Chemical and Mechanical Methods of Pipeline Integrity [1 ed.] 9781613996171, 9781613994962

Citation preview

Chemical and Mechanical Methods for Pipeline Integrity

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 1

03/11/17 6:50 PM

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 2

03/11/17 6:50 PM

Chemical and Mechanical Methods for Pipeline Integrity

Wayne W. Frenier, FNACE Frenier Chemistry Consultants Written in Collaboration with T. D. Williamson, Incorporated Foreword by Richard B. Williamson Chairman (Ret.), T. D. Williamson, Incorporated

Society of Petroleum Engineers

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 3

03/11/17 6:50 PM

© Copyright 2018 Society of Petroleum Engineers All rights reserved. No portion of this book may be reproduced in any form or by any means, including electronic storage and retrieval systems, except by explicit, prior written permission of the publisher except for brief passages excerpted for review and critical purposes. Printed in the United States of America.

Disclaimer This book was prepared by members of the Society of Petroleum Engineers and their wellqualified colleagues from material published in the recognized technical literature and from their own individual experience and expertise. While the material presented is believed to be based on sound technical knowledge, neither the Society of Petroleum Engineers nor any of the authors or editors herein provide a warranty either expressed or implied in its application. Correspondingly, the discussion of materials, methods, or techniques that may be covered by patents implies no freedom to use such materials, methods, or techniques without permission through appropriate licensing. Nothing described within this book should be construed to lessen the need to apply sound engineering judgment nor to carefully apply accepted engineering practices in the design, implementation, or application of the techniques described herein.

ISBN 978-1-61399-496-2

First Printing 2018

Society of Petroleum Engineers 222 Palisades Creek Drive Richardson, TX 75080-2040 USA

http://www.spe.org/store [email protected] 1.972.952.9393

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 4

03/11/17 6:50 PM

Dedication I dedicate this book to my family. My wife, Dolores, our children Andrew Frenier and Kathleen Turner and our grandchildren are the inspiration for the work that I have accomplished. – Wayne Frenier

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 5

03/11/17 6:50 PM

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 6

03/11/17 6:50 PM

Foreword Critical Needs Pipelines are the safest, most reliable, and most cost-effective means of transporting hazardous fluids. In comparison to other forms of transportation, their construction and operations have substantially less initial and ongoing impact on the environment. With approximately 2.3 million miles of hazardous substance pipelines operating in as many as 120 countries, these lines are some of the most important infrastructures in the world. The number will continue to grow with population growth, changing patterns of fuel usages for power generation, and industrial, commercial, and residential heating and transportation. In just the last 50 years, these pipeline systems have grown from approximately 0.5 million to 2.3 million miles of pipelines built and in operation for crude oil, refined liquids, and natural gas and gas liquids in regions across the world. These more recently constructed pipelines reflect the growing industrial and post-war generations’ requirements of the industrializing and industrial economies. They also reflect the critical role that pipelines continue to play in scaling the ability of economies to expand their energy consumption in a safe and cost-effective manner. However, these newer pipelines—built with new and improved material and construction techniques—as well as those constructed pre-1970 all require continuing efforts to assure that they remain safe from external forces and that they are resistant to corrosion, for example, and to ensure that the pipelines remain free of materials and conditions that would impede the effective transport of fluids. A Short History of Petroleum Pipelines The history of the pipeline industry is largely the story of industrialized societies’ growing use of and reliance on hydrocarbon liquids and gas for energy and essential chemicals. While it is not completely clear when the first iron-based pipelines were used, the development of the first drilled oil/gas wells in Pennsylvania circa 1865 was a major driver. There is also a claim that iron-based lines were used in Canada circa 1860 to transport crude oil. Gas pipelines made of wood are claimed to have been used as early as 1830. A Canadian source also says that a 25-km iron gas pipeline was constructed in 1856 (NRC 2016). At this early point in time, the market for distilled crude oil (kerosene) and gas was for lighting. However, in 1885 Robert Bunsen’s invention of what is now known as the Bunsen burner opened vast new opportunities to use natural gas. Once effective and numerous pipelines began to be built in the 20th century, the use of natural gas expanded to home heating and cooking, manufacturing and processing plants, and utility plants to generate electricity. The development of the internal combustion engine and motor vehicles in the early 20th century dramatically changed the markets for distilled oil-based products. An interesting note is that during the 19th century, there were significant competitions between transportation of petroleum products in pipelines and transportation using horse-drawn wagons (oil in whiskey barrels, for example) and then in rail cars. This competition greatly affected the price of the delivered products. Similar types of competitions between pipelines and rail delivery of crude oil continue today. Technical, Regulatory, and Societal Challenges Pipeline maintenance and integrity management will become more necessary in hostile climates and remote locations such as the Arctic and deep water. Additional challenges will include producing additional appropriate devices for “unpiggable” lines. Chemical issues will include formulating, vii

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 7

03/11/17 6:50 PM

viii Foreword

sourcing, and injecting “greener” corrosion inhibitors and other treatment chemicals into longer line sections. Waste disposal of water and treatment of chemicals, as well as ecological requirements and problems with heavy and waxy crude oils and high-water-cut wells, will require new solutions. We are living in an age of almost instant communication of news about transportation failures. Societies around the world will require that hazardous liquids and gases be transported safely and reliably. The result will be more laws and regulations and the development of best practices to reduce these failures. The growing experience in the pipeline industry as well as the current and future infrastructure improvements can be used to continue to reduce spills and failures. The Importance of Training The various parts of the pipeline systems are physically connected from the wells to the consumers. They are also connected by a relatively small set of mechanical and chemical principles. Once these principles are understood and taught to pipeline engineers and operators, consistent and improved practices can be developed. To this end, this book—inspired, authored, compiled, and edited by Wayne Frenier, one of the true experts in our industry—is a guide to the practicing pipeline engineer and others who serve and support the pipeline industry. Richard B. Williamson, Chairman (Ret.), T. D. Williamson, Incorporated

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 8

03/11/17 6:50 PM

Preface Pipelines are a major part of the circulatory and transportation systems of the oil/gas and petrochemical industries. Fluids are produced from the wells and flow through gathering lines or/and flowlines to central points where some degree of processing and separation may take place (in onshore and offshore facilities). The liquids and gases are then placed into pipelines for conveyance to refineries, to chemical plants, or to power producers. The refined products also may pass through pipelines to the ultimate users. At various points wastewater streams (including vast amounts of produced and hydraulic fracture water) are generated, and these very diverse fluids may pass through pipelines before disposal. This frequently means reinjection of a fraction of the fluids into the earth through the injection-well system of a field or into licensed wastewater wells. The author and technical consultants of this book (see “Acknowledgments”) understand that some of the transportation of well fluids as well as the finished products is accomplished using tanker trucks, barges, trains, and ships of many designs. At the date of the compilation of this book, there was a shortage of pipelines to transport liquids or gas. The cost of rail transport was USD 10–15 more for a barrel of crude oil than by pipeline (Batheja 2014). The book, however concentrates on the piping-conveyed segments of the transportation system. Although the analogy of a circulatory system breaks down to some extent because all the fluids are not returned to the producing formations, very large volumes of liquids and gases may be reinjected as part of pressure maintenance, enhanced oil recovery, stimulation activities (see Frenier and Ziauddin 2014)., and waste disposal. The “heart” of the processes includes the natural pressure of the producing formations as well as innumerable pumps and gas compressors of all descriptions that maintain the flowing pressure of the systems. The pipelines are vital and integral parts of the oil/gas systems of the world. Because of recent advances in production of gas and oil from shale formations (Boyer et al. 2011; PI 2012), the future growth in pipeline infrastructure will be significant. INGAA (2011) has predicted the growth for North America through 2035. The values in the following predictions are from this reference. • Natural gas transmission infrastructure  43 Bcf/D of new natural gas transmission capability  400,000 miles of gathering lines (at 16,000 miles/yr)  1,400 miles/yr of new gas transmission mainline  600 miles/yr of new laterals to and from natural-gas-fired power plants, processing facilities, and storage fields  24 Bcf/yr of new working gas capacity in storage  197,000 hp/yr for pipeline compression • Natural gas liquids and oil infrastructure  USD 0.6 billion/year or a total of USD 14.5 billion over the study period for natural gas liquids pipeline expenditures  USD 1.3 billion/year or USD 31.4 billion over the study period of capital expenditures for oil pipelines Fig. P-1 is a photograph of a pipeline under construction showing some of the welded segments. Fig. P-1a shows sections in the construction ditch with welded and unwelded sections. Fig. P-1b demonstrates the large equipment needed for laying pipeline segments.

ix

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 9

03/11/17 6:50 PM

x Preface

(a) Pipeline showing sections

(b) Equipment laying sections Fig. P-1—Pipelines under construction: (a) individual sections, (b) handling equipment.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 10

03/11/17 6:50 PM

Preface xi

An important driver for current construction of additional pipelines is the new light oil as well as the loss of associated natural gas produced in several of the shale oil fields (Bakken in North Dakota, USA, and the Eagle Ford in Texas, USA). Because of the lack of pipelines and gas-treating facilities, as much as 30% of the gas is flared (Fig. P-2) or is used for directly powering hydraulically operated equipment that then vents the gas into the atmosphere (Rahim 2013). A report by Aleklett (2013) includes satellite photos claimed to have been taken by the National Aeronautics and Space Administration of multiple oilfield gas flares in the Bakken and Eagle Ford plays. The illumination from the flares seems to compare in intensity with the illumination of major cities in the regions near the flares. The state of North Dakota as well as a major producer (described in Rahim 2013) are claimed to be committed to dramatically reducing the waste of these hydrocarbon streams by constructing adequate transportation and other ways to use the products in the near future. This new oil and gas also affects the location of any new pipelines. Speakers (Solomon et al. 2015; Banerjee 2015) at the 2015 Pipeline + Energy Conference and Exposition in Tulsa, Oklahoma, USA, noted that oil quality (light vs. heavy) also plays a role in the geographic direction of new lines as well as the mode of transportation. They noted that the refineries best suited for heavy oil are on the US coast on the Gulf of Mexico and that those designed for light oil are mostly on the US East Coast. Thus, in some cases, rail transportation is currently needed, even if it is costlier than in-place pipelines. In addition, methane is a potent “greenhouse gas.” USEPA (2013a) noted that the lifetime of methane (CH4) in the atmosphere is much shorter than that of carbon dioxide (CO2), but CH4 is more efficient at trapping radiation than is CO2. The report claims that pound for pound, the comparative impact of CH4 on climate change is over 20 times greater than that of CO2 over a 100-year period. Johnson (2014) has published a short review of the role that oil/gas production plays in contributing to the release of methane to the atmosphere and the actual role it has in global climate change. He cites information that up to 29% of the annual methane loss to the atmosphere comes from oil/gas

Fig. P-2—Orvis State natural gas flare, Arnegard North Dakota (Wikimedia 2013).

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 11

03/11/17 6:50 PM

xii Preface

production and transportation. He claims that because of the trapping-of-heat factor of methane vs. CO2, reducing methane emissions in the short term (i.e., in the next 10 years) should be a global priority. The needed construction of new gas-handling facilities is in turn driven by the economic value of these resources as well as US Environmental Protection Agency (EPA) regulations (EPA 2012), 40 CFR 98 Subpart W (USEPA 2012), and 40 CFR 60 Subpart OOOO (USEPA 2013b). Details regarding the difficulties of maintaining these current types of oil/gas facilities and new lines are described in Sections 2.2.3 and 2.3.1. The overall result for some sectors is “Midstream Mania.” Walton (2013) claims that companies that construct, own, or maintain pipelines, tankage, and midstream processing plants are working overtime to meet demand. One section of the article is titled “The Golden Age of Pipelines.” However, the author and consultants for this book note that significant problems have already arisen that must be addressed and thus are the major subject of this book. Note that economic conditions as well as market forces also will affect demand for new pipelines. Goal and Organization of This Book This current publication provides an overview of the science and technology of the use of a wide range of pipeline chemicals and mechanical equipment for a general technical audience, with emphasis placed on the basic chemical/physical principles by which the chemicals and devices can enhance or maintain product delivery. Some knowledge of chemistry (equivalent to an introductory college general chemistry course) is assumed. More-advanced concepts are introduced in Chapter 1. The introductory chapter describes the varied pipeline environments, problems that require chemical or mechanical intervention, and thus the need for the thousands of different chemicals and devices that are in use. This chapter also reviews the important chemical/physical principles that are common to most if not all the enhancement treatments. The applications of interventions described are primarily in the upstream and midstream oil/gas business, but many of the methods can also be used in refineries or product pipelines. This book is limited to internal pipeline flow assurance and reliability issues and does not specifically address external corrosion, internal coatings, or cathodic protection. See Goldschmidt and Streitberger (2003) for references to pipeline coatings and Lazzari and Pedeferri (2006) for cathodic protection information, as well as NACE (2011) for a general review of pipeline corrosion control. The following is an outline for analysis of potential internal problems in pipelines and facilities: 1. 2. 3. 4.

Is there a problem that requires an intervention? If there is a problem, how bad will it be? Can the problem be managed through engineering and/or chemical means? Evaluate the results of an intervention or control strategy.

Each major section and most subsections will include reviews of current literature as well as summaries of the consensus understandings from the literature cited. Chapter 1, “Introduction to the Technology of Flow and Integrity Management,” provides an overview of the reasons pipelines require intervention to enhance or maintain product delivery and introduces the various types of chemical and mechanical interventions in use. This chapter also reviews basic chemical and engineering processes that occur in pipeline operations. It also emphasizes the commonality shared among many of the chemical and engineering processes across the pipeline systems as well as the well production processes. Chapter 2, “From the Well to the Consumer,” describes how the aqueous fluids, hydrocarbon liquids, and gases change from the wellhead, through the gathering lines and surface or subsurface facilities, then through transmission (trunk) pipelines to a consumer. It outlines the chemical/

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 12

03/11/17 6:50 PM

Preface xiii

physical forces that ultimately affect the delivery of the products as well as the plans to anticipate and alleviate problems. Chapter 3, “Corrosion Processes in Pipelines and Facilities,” provides a review of internal corrosion and corrosion mechanisms that affect pipeline/facility operations. Here also are reviewed pertinent texts. In addition, the chapter provides a short introduction to integrity management processes. Chapter 4, “Chemistry of Product Flow Impairment in Pipelines and Facilities,” describes processes that impair the flow of oil, gas, and water in pipelines and facilities. These include inorganic solids, organic solids, mixed deposits, and emulsions. Chapter 5, “Mechanical Methods of Enhancement and Assessment of Pipeline Operations,” reviews the many pigs (scrapers), moles, coiled tubing, and jetting equipment used to maintain the lines and to place chemicals in them. The use of hydrostatic testing as well as in-line detection and inspection devices is also reviewed. Chapter 6, “Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations,” shows how chemicals and mechanical devices perform to prevent and inhibit the formation of deposits, corrosion, and emulsions. This chapter also reviews gas dehydration methods, acid gas removal, and flow enhancement chemicals. Chapter 7, “Cleaning of Pipelines and Facilities,” describes the use of pigging and various chemical cleaning solvents to clear fouled pipelines and units in the facilities. Methods reviewed include chemicals, testing, evaluation, and application of solvents. Chapter 8, “Pipeline/Facility Maintenance Health, Safety, and Environmental Issues,” reviews issues of health, safety, and the environment related to the maintenance of pipelines and facilities. Chapters 1 through 8 each concludes with a “Summary and Lessons Learned” section that summarizes the major findings revealed by the review of the technologies discussed and how this knowledge can be applied to pipeline management projects. Chapters 3, 4, 5, 6, and 7 also have a section titled “Best Practices and Case Studies for Chemical/Mechanical Management of Pipelines.” Here, the science and engineering principles described in the earlier sections are illustrated through practical demonstrations of chemical/mechanical intervention and remediation. Thus, Chapters 1 through 4 describe the root problems and Chapters 5 through 8 provide a large range of mechanical and chemical solutions that can be accomplished in safe and environmentally acceptable ways. Definition of the Pipeline Treatment Environment and Summary The scope of this book is limited primarily to chemical/mechanical intervention and enhancement in the production (upstream) and transfer (midstream) oilfield environment. This includes flowlines and gathering lines, associated-gas/liquid-treating facilities, and US Department of Transportation– regulated crude-oil trunk and gas transmission pipelines. The discussion will not specifically include problems in refineries or finished product pipelines. However, many of the needed techniques and technologies, especially those for treating separation/gas-treating units, are similar to those described in this book and could be applied with appropriate modifications. See Frenier (2001) for a review of the cleaning of industrial equipment, including downstream oil/gas equipment. The scope is limited primarily to internal flow assurance and integrity issues. Note that external corrosion problems are usually addressed using cathodic protection and coating methodologies that are beyond the scope of this book. However, some in-line inspections will detect and assess exterior corrosion as well as the effectiveness of cathodic protection. See Lazzari and Pedeferri (2006) for an explanation of cathodic protection and Munger and Vincent (1999) for discussions of protective coatings. Also see the comprehensive books on pipeline activities by Revie (2015), Peabody (2001), and Hevle (2012) and the Petrowiki (2014) web page.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 13

03/11/17 6:50 PM

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 14

03/11/17 6:50 PM

Acknowledgments The author acknowledges the extensive help and advice of T. D. Williamson, Incorporated. The engineers at T. D. Williamson participated in the planning, evolution, and technical reviews of this book. I wish especially to thank Richard B. Williamson, Chairman (Ret.) of T. D. Williamson, Incorporated, for his continuing support and help with the project. I also recognize the extensive advice of Lee Shouse, Chuck Harris, Woody Smith, Eric Freeman, and Gordon Blair of T. D. Williamson. In addition, I acknowledge the help and guidance of Rick Underwood of Enable Midstream Partners (Enable 2013) (Ret.), Michael Volk Jr. of the University of Tulsa, Mohsen Achour of ConocoPhillips, and Murtaza Ziauddin of Schlumberger. I also acknowledge the help and encouragement of David Wint, Jeff Foote, and Ian Lisko as well as the members of the Tulsa Section of NACE, International.

xv

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 15

03/11/17 6:50 PM

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 16

03/11/17 6:50 PM

Table of Contents Foreword������������������������������������������������������������������������������������������������������������������������� vii Preface������������������������������������������������������������������������������������������������������������������������������ ix Acknowledgments���������������������������������������������������������������������������������������������������������� xv 1  Introduction to the Technology of Flow and Integrity Management������������������������ 1 1.1  Description of Pipelines and Operating Environments�������������������������������������������� 2 1.2 Need for Chemical and Mechanical Enhancements to Pipelines and Facilities������������������������������������������������������������������������������������������������������������ 4 1.3  Economic and Market-Related Forces Affecting Pipeline Maintenance������������������ 5 1.3.1  General Economic Issues for Oilfield Treatments����������������������������������������� 5 1.3.2  Pipeline-Specific Economic Considerations������������������������������������������������� 6 1.4  Review of the Physics and Chemistry of Pipeline Interventions������������������������������ 7 1.4.1  Pipeline Materials for Construction and Pressure Requirements����������������� 7 1.4.2 Fluid Mechanics and the Effects of Fluids and Phases on Pipeline Operations���������������������������������������������������������������������������������������������������� 9 1.4.3  Viscosity and Rheology of Fluids��������������������������������������������������������������� 17 1.4.4  Thermodynamics and Kinetics of Pipeline-Fouling Reactions������������������� 21 1.4.5  Surface Chemistry�������������������������������������������������������������������������������������� 24 1.4.6  Testing of Pipeline Fluids���������������������������������������������������������������������������� 28 1.5  Summary and Lessons Learned��������������������������������������������������������������������������� 29 2  From the Well to the Consumer�������������������������������������������������������������������������������� 31 2.1  Description of Well Production Fluids Entering the Pipeline System��������������������� 31 2.1.1  Aqueous Phases���������������������������������������������������������������������������������������� 31 2.1.2  Hydrocarbon Liquids���������������������������������������������������������������������������������� 33 2.1.3  Gaseous Phases���������������������������������������������������������������������������������������� 33 2.1.4 Solids��������������������������������������������������������������������������������������������������������� 34 2.1.5  Emulsions, Foams, and Solid Dispersions������������������������������������������������� 34 2.2  Effects of the Life Cycle and Reservoir Type on Pipeline Maintenance����������������� 34 2.2.1  Life Cycle of a Hydrocarbon-Producing Reservoir������������������������������������� 34 2.2.2  Conventional Reservoirs of Oil and Gas���������������������������������������������������� 38 2.2.3  Unconventional Reservoirs of Oil and Gas������������������������������������������������ 39 2.3  Problems Anticipated in Different Areas of Pipeline-Connected Systems������������� 44 2.3.1  Gathering Lines and Wastewater Lines����������������������������������������������������� 45 2.3.2  Surface and Subsurface Facilities�������������������������������������������������������������� 46 2.3.3  Water-System-Related Issues�������������������������������������������������������������������� 60 2.3.4  Summary of Problems in Surface/Subsurface Treatment Plants���������������� 62 2.3.5  Crude Oil Trunk and Gas Transmission Pipelines�������������������������������������� 63 2.4  Planning for Pipeline/Facility Reliability and Flow Assurance�������������������������������� 64 2.5  Summary and Lessons Learned��������������������������������������������������������������������������� 67 3  Corrosion Processes in Pipelines and Facilities����������������������������������������������������� 69 3.1  Fundamentals of Corrosion Chemistry������������������������������������������������������������������ 70 3.1.1  Effects of Pipeline Metallurgy on Corrosion Processes����������������������������� 72 3.1.2  Effect of Dissolved Salts and the Hydrocarbon Phases����������������������������� 76 3.1.3  Carbon Dioxide Corrosion�������������������������������������������������������������������������� 79 xvii

xviii  Table of Contents

3.1.4  Hydrogen Sulfide Corrosion and Various Cracking Conditions������������������ 81 3.1.5  Oxygen Corrosion�������������������������������������������������������������������������������������� 84 3.1.6  Organic Acid Corrosion������������������������������������������������������������������������������ 85 3.1.7  Microbiologically Influenced Corrosion������������������������������������������������������� 85 3.1.8  Corrosion From Cleaning/Stimulation Fluids���������������������������������������������� 89 3.2  Consequences of Corrosion: Manifestations in Pipelines�������������������������������������� 91 3.2.1  Top-of-Line Corrosion��������������������������������������������������������������������������������� 91 3.2.2  Localized Corrosion����������������������������������������������������������������������������������� 95 3.2.3  Erosion/Corrosion and Impingement Damage������������������������������������������� 99 3.2.4 Summary of Pipeline Corrosion Locations and Manifestation Types��������������������������������������������������������������������������� 102 3.3  Corrosion Processes in Facilities������������������������������������������������������������������������ 102 3.4  Corrosion and Inhibitor Testing���������������������������������������������������������������������������� 104 3.4.1  Common Methods for Corrosion Rate Determination������������������������������ 104 3.4.2 Specific Laboratory Test Methods for Studying Corrosion and Inhibition�������������������������������������������������������������������������������������������� 112 3.4.3  Physical and Chemical Examination of Surfaces������������������������������������� 124 3.4.4  Comparison of Laboratory Tests�������������������������������������������������������������� 126 3.4.5  Field Monitoring of Corrosion Processes������������������������������������������������� 127 3.4.6 Placement of Probes and Coupons for Maximum Effectiveness��������������������������������������������������������������������������� 135 3.5  Introduction to Integrity Management������������������������������������������������������������������ 136 3.5.1  Short History of Integrity Management Processes���������������������������������� 136 3.5.2  Preview and Nomenclature of Direct Assessment����������������������������������� 139 3.5.3  US State Regulations������������������������������������������������������������������������������� 139 3.5.4  US Federal Inspection Regulations���������������������������������������������������������� 142 3.5.5  Worldwide Pipeline Safety Regulations and Practices����������������������������� 144 3.5.6 Summary of Standards for Corrosion Assessments and Safe Operation���������������������������������������������������������������������������������� 146 3.6  Corrosion Prediction and Assessment Processes���������������������������������������������� 147 3.6.1  Methods of Corrosion Prediction�������������������������������������������������������������� 147 3.6.2 Internal Corrosion Direct Assessment and Risk Assessment Methods������������������������������������������������������������������������������� 152 3.7  Summary and Lessons Learned������������������������������������������������������������������������� 158 3.8  Best Practices and Case Histories for Corrosion Control������������������������������������ 159 3.8.1 Monitoring and Controlling Corrosion in an Aging Sour-Gas-Gathering System: A Nine-Year Case History (Nelson et al. 2007)���������������������������������������������������������������������������������� 159 3.8.2 Top-of-Line Corrosion in Multiphase Gas Lines: A Case History (Gunaltun et al. 1999)������������������������������������������������������ 159 3.8.3 Special Issues Related to the Application of Direct Assessment (Klechka 2009)�������������������������������������������������������������������� 159 3.8.4 Internal Corrosion Direct Assessment of Buried Steel Pipeline in Chinese Oil Industry (Qimin and Guibai 2008)����������������������� 160 4  Chemistry of Product Flow Impairment in Pipelines and Facilities��������������������� 161 4.1  Why and Where Inorganic Scale and Organic Deposits Form���������������������������� 162 4.2  Inorganic Scale Formation���������������������������������������������������������������������������������� 163 4.2.1  Mineral Precipitation Scales��������������������������������������������������������������������� 163 4.2.2  Corrosion Product Scales������������������������������������������������������������������������ 172

Table of Contents  xix

4.3  Organic Deposits������������������������������������������������������������������������������������������������� 173 4.3.1  Waxes (Paraffins)������������������������������������������������������������������������������������� 173 4.3.2 Asphaltenes��������������������������������������������������������������������������������������������� 174 4.3.3 Naphthenates������������������������������������������������������������������������������������������� 175 4.3.4  Clathrate Gas Hydrates���������������������������������������������������������������������������� 177 4.4 Mechanism of Formation of Mixed Deposits in Pipelines and Facilities�������������������������������������������������������������������������������������������������������� 179 4.4.1  Overlapping Precipitation Regimes���������������������������������������������������������� 179 4.4.2  Pressure-Induced Deposits���������������������������������������������������������������������� 180 4.4.3  Alkaline Earth Scale and Organic Solids������������������������������������������������� 181 4.4.4  Corrosion-Triggered Mixed Deposits Including “Black Powder”���������������� 181 4.5  Deposits Found in Facilities��������������������������������������������������������������������������������� 187 4.6  Emulsions, Foams, and High-Viscosity Causes of Flow Problems��������������������� 188 4.6.1 Emulsions������������������������������������������������������������������������������������������������� 188 4.6.2 Foams������������������������������������������������������������������������������������������������������ 191 4.6.3  High Viscosity and Turbulent Flow������������������������������������������������������������ 191 4.7  Summary and Lessons Learned������������������������������������������������������������������������� 191 4.8 Best Practices and Case Studies of Flow Assurance Control in Pipelines���������������������������������������������������������������������������������������������������������� 192 4.8.1 Review of Issues Associated with Inhibition of Scale-Deposit-Covered Pipelines (Turgoose et al. 2006)������������������������� 192 4.8.2 A Case History of Heavy-Oil Separation in Northern Alberta: A Singular Challenge of Demulsifier Optimization and Application (Wylde et al. 2010)���������������������������������������������������������� 192 4.8.3 Best Practice for Organic Deposit Removal (Montgomery et al. 1996)������������������������������������������������������������������������� 193 4.8.4 Multichemical Application Case History (Shepherd et al. 2012)������������������������������������������������������������������������������ 194 5 Mechanical Methods of Assessment and Enhancement of Pipeline Operations�������������������������������������������������������������������������������������������������� 195 5.1  Introduction to Pigs, Scrapers, and In-Line Inspection Devices�������������������������� 195 5.2  Types and Uses of In-Line Pipeline Devices�������������������������������������������������������� 199 5.2.1  Utility Pigs������������������������������������������������������������������������������������������������ 199 5.2.2  Pig Application Problems and Solutions�������������������������������������������������� 202 5.2.3  In-Line Operations Chemical Issues and Solutions��������������������������������� 207 5.2.4  Selection and Application of Mechanical Pigs������������������������������������������ 210 5.3  Hydrostatic Pressure Testing������������������������������������������������������������������������������� 214 5.3.1 General Procedures for Performing Hydrostatic Pressure Tests������������������������������������������������������������������������������������������ 215 5.3.2  Water Quality and Wet Layup������������������������������������������������������������������� 216 5.3.3  Drying the Line and Dry Layup���������������������������������������������������������������� 217 5.4  In-Line Inspection Tools��������������������������������������������������������������������������������������� 218 5.4.1  Introduction to In-Line Inspection Technologies��������������������������������������� 218 5.4.2  Geometry and Mapping Devices�������������������������������������������������������������� 222 5.4.3  Magnetic Flux Leakage Technologies������������������������������������������������������ 224 5.4.4  Ultrasonic Technique Method������������������������������������������������������������������� 227 5.4.5  Eddy Current In-Line Inspection Methods������������������������������������������������ 230 5.4.6  Miniaturized In-Line Inspection Sensors�������������������������������������������������� 231 5.4.7  Cathodic Protection (CP) In-Line Tools���������������������������������������������������� 233

xx  Table of Contents

5.4.8 Comparison/Selection of In-Line Inspection Methods and Data Analysis������������������������������������������������������������������������������������ 235 5.4.9  Analyses of In-Line Inspection Data and Tool Performance��������������������� 238 5.4.10  Summary of In-Line Inspection Developments�������������������������������������� 243 5.5  In-Line Handling Equipment and Other Devices������������������������������������������������� 245 5.5.1  Traps (Receivers) and Launchers������������������������������������������������������������ 245 5.5.2  Pig Indicators and Tracking Devices��������������������������������������������������������� 247 5.5.3  Coiled Tubing for Conveying Tools Into Pipelines������������������������������������� 249 5.6  Summary and Lessons Learned������������������������������������������������������������������������� 251 5.7 Best Practices and Case Studies for Mechanical Management of Pipelines������������������������������������������������������������������������������������ 251 5.7.1 Unpiggable Pipelines: What a Challenge for In-Line Inspection! (Schmidt 2004)���������������������������������������������������������� 251 5.7.2 Case Study: Bayu Undan Pipeline and Darwin Liquefied Natural Gas Project (Weatherford 2009)���������������������������������� 251 5.7.3 Single-Trip Pigging of Gas Lines During Late Field Life (Mandke et al. 2002)�������������������������������������������������������������������������������� 251 5.7.4 In-Line Inspection on an Unprecedented Scale (Brockhaus et al. 2015)���������������������������������������������������������������������������� 252 6 Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations�������������������������������������������������������������������������������������������������� 253 6.1  Corrosion Inhibitors and Inhibition Mechanisms�������������������������������������������������� 253 6.1.1 Review of Production and Pipeline Corrosion Inhibitor Mechanisms������������������������������������������������������������������������������� 257 6.1.2 “Sweet” Corrosion Inhibitors��������������������������������������������������������������������� 258 6.1.3 Sour Brine Corrosion Inhibitors and Inhibition of Microbiologically Influenced Corrosion����������������������������������������������������� 262 6.1.4 Inhibitors for Gas-Containing Lines and Multiphase Lines����������������������� 263 6.1.5  Selection of Corrosion Inhibitors for Use in Pipelines������������������������������ 269 6.1.6  Chemistries for Inhibition of Cleaning Fluid Acid Attacks������������������������� 271 6.2  Inorganic Scale Inhibitors and Inhibition Mechanisms���������������������������������������� 271 6.3  Organic Deposit Inhibition����������������������������������������������������������������������������������� 275 6.3.1  Paraffin Deposition and Pour Point Inhibitors������������������������������������������� 275 6.3.2  Asphaltene Inhibitors�������������������������������������������������������������������������������� 275 6.3.3  Gas Hydrate Inhibitors����������������������������������������������������������������������������� 277 6.3.4  Calcium/Sodium Naphthenate Inhibition�������������������������������������������������� 280 6.4  Flow Enhancement, Biocides, and Oxygen Scavengers������������������������������������� 281 6.4.1  Demulsifiers, Foaming Agents, and Defoamers��������������������������������������� 281 6.4.2  Flow Enhancers: Drag-Reducing Agents������������������������������������������������� 284 6.4.3 Biocides���������������������������������������������������������������������������������������������������� 288 6.4.4  Oxygen Scavengers��������������������������������������������������������������������������������� 290 6.5  Acid Gas Scavenger Chemistry�������������������������������������������������������������������������� 291 6.5.1  Oxidizing Agents�������������������������������������������������������������������������������������� 292 6.5.2 Aldehydes������������������������������������������������������������������������������������������������ 292 6.5.3  Nitrogen-Based Scavengers�������������������������������������������������������������������� 292 6.6  Chemistry for Producing Pipeline Gels���������������������������������������������������������������� 294 6.7  Fluid Additive Injection Methods and Equipment������������������������������������������������� 297 6.7.1  Continuous Treatments With Pipeline Chemicals������������������������������������� 299 6.7.2 New Formulations for Application of Production Corrosion and Scale Inhibitors���������������������������������������������������������������������������������� 306

Table of Contents  xxi

6.7.3 Applications of Treatment Chemicals in the Batch Mode and Using Pigs��������������������������������������������������������������������� 307 6.8  Summary and Lessons Learned������������������������������������������������������������������������� 315 6.9 Best Practices and Case Histories for Maintenance Treatments of Pipelines��������������������������������������������������������������������������������������� 316 6.9.1 Monitoring and Controlling Corrosion in an Aging Sour-Gas-Gathering System: A Nine-Year Case History������������������������� 316 6.9.2 Black Powder in Pipeline: Cleaning Program (Sirnes and Gundlach 2012)�������������������������������������������������������������������� 316 6.9.3 Beneficial Effects of Chemical Treatment and Maintenance Pigging Programs in Returning an Offshore Pipeline to Pre–Hurricane Ike Conditions Following a Breach and the Ingress of Seawater and Sand, and the Effects of Bacteria-Generated H2S (Powell et al. 2010; Powell et al. 2012)������������������������������������������������������������������������������������ 316 6.9.4 Integrated Production Chemistry Management of the Schoonebeek Heavy-Oil Redevelopment in the Netherlands: From Project to Startup and Steady-State Production (Shepherd et al. 2012)������������������������������������������������������������������������������ 316 7  Cleaning of Pipelines and Facilities������������������������������������������������������������������������ 319 7.1  Maintenance Pigging������������������������������������������������������������������������������������������� 319 7.2 Automated Methods and Mechanical Equipment for Maintenance Cleaning����������������������������������������������������������������������������������������� 322 7.2.1  Horizontal Multiple-Pig Launchers����������������������������������������������������������� 323 7.2.2  Vertical Multiple-Pig Launchers���������������������������������������������������������������� 323 7.2.3  Automatic Sphere Launchers������������������������������������������������������������������� 325 7.3 Chemical Solvents and Mechanisms for Cleaning Pipelines and Facilities�������������������������������������������������������������������������������������������������������� 326 7.3.1  Introduction to Solvent Chemistry������������������������������������������������������������ 327 7.3.2  Chemicals for Removing “Acid-Soluble” Inorganic Solids������������������������ 331 7.3.3  Solvents for Alkaline Earth Sulfates��������������������������������������������������������� 340 7.3.4  Corrosion Inhibitors for Inorganic Scale Removal Solvents��������������������� 343 7.3.5  Solvents for Organic Solids���������������������������������������������������������������������� 345 7.3.6  Solvents for Mixed Deposits��������������������������������������������������������������������� 356 7.4  Testing of Deposits To Develop Solvents������������������������������������������������������������� 357 7.4.1 Work Flow for Evaluation and Treatment of Inorganic, Organic, or Mixed Deposit������������������������������������������������������������������������ 358 7.4.2  Chemical Identification of a Deposit��������������������������������������������������������� 359 7.4.3  Evaluation of Solvents for Removing Inorganic/Organic Deposits����������� 360 7.5  Application of Chemical Cleaning in Pipelines���������������������������������������������������� 365 7.5.1  Liquid-Phase Solvents����������������������������������������������������������������������������� 365 7.5.2  Examples of Treatment of Pipelines To Remove Mixed Deposits������������� 366 7.5.3  Uses of Gelly Pigs in Pipeline Cleaning��������������������������������������������������� 370 7.5.4  Use of Foamed Fluids in Pipeline Cleaning and Treatment��������������������� 373 7.6  Cleaning of Topside Facilities������������������������������������������������������������������������������ 377 7.6.1  Typical Topside Equipment Needing Cleaning����������������������������������������� 377 7.6.2  Methods for Cleaning Topside Facilities and Heat Exchangers��������������� 378 7.6.3  On-Site Cleaning Issues�������������������������������������������������������������������������� 381 7.6.4  Examples of Cleaning of Oil and Gas Facilities��������������������������������������� 382

xxii  Table of Contents

7.7 A Company’s Best Practice for Maintaining Gas-Gathering and Transmission Lines���������������������������������������������������������������������������������������������� 383 7.8  Summary and Lessons Learned������������������������������������������������������������������������� 383 7.9 Best Practices and Case Histories for Renovation/Remediation of Pipelines and Facilities������������������������������������������������������������������������������������ 384 7.9.1 Pigging of Pipelines With High Wax Content (Tordal 2006)��������������������� 384 7.9.2 Recommissioning of Mothballed Pipelines Offshore California: A Success Story of Cleaning, Pigging, Monitoring, and Integrity Management (Wylde 2009)���������������������������������������������������������������������� 384 7.9.3 Available Methodologies Concerning the Treatment and Removal of Sand From Pipelines, With Associated Case Studies (Mackay 2013)������������������������������������������������������������������� 384 7.9.4 Innovative Methodology for Cleaning Pipes: Key to Environmental Protection (Buzelin and Lima 2008)��������������������������������� 385 7.9.5  Cleaning the Valhall Offshore Oil Pipeline (Marshall 1990)���������������������� 385 7.9.6 Conclusions���������������������������������������������������������������������������������������������� 386 8 Pipeline/Facility Maintenance Health, Safety, and Environmental Issues����������������������������������������������������������������������������������������������� 387 8.1 Introduction���������������������������������������������������������������������������������������������������������� 387 8.2 Health and Safety Considerations During Pipeline/Facility Maintenance Operations������������������������������������������������������������������������������������� 387 8.2.1  General Considerations for All Oil/Gas Operations���������������������������������� 387 8.2.2  Specific Pipeline/Facility Considerations�������������������������������������������������� 393 8.3  Health, Safety, and Environmental Management������������������������������������������������ 395 8.3.1 Chemical Selection To Enhance Health, Safety, and Environmental Management Compliance����������������������������������������� 395 8.3.2  Chemical Development Processes���������������������������������������������������������� 396 8.3.3 Chemical Handling Processes To Promote Health, Safety, and Environmental Management Improvements�������������������������������������� 401 8.4 Handling, Reuse, and Disposal of Pipeline/Facility Treating Fluids and Solids������������������������������������������������������������������������������������������������� 403 8.4.1 Planned Waste Disposal Options for Pipeline/Facility Cleaning Fluids���������������������������������������������������������������������������������������� 403 8.4.2 Control and Remediation of Spills in Water Bodies and on Land��������������������������������������������������������������������������������������������� 407 8.4.3 Use of Remediation Chemicals in Bodies of Water and Proposed Mechanism of Action�������������������������������������������������������� 411 8.5  Summary and Lessons Learned (Chapter and Book)����������������������������������������� 415 8.5.1  Chapter 8 Lessons Learned��������������������������������������������������������������������� 415 8.5.2  General Lessons From This Book������������������������������������������������������������ 415 Appendix A: Glossary�������������������������������������������������������������������������������������������������� 417 Appendix B: Nomenclature����������������������������������������������������������������������������������������� 421 References�������������������������������������������������������������������������������������������������������������������� 423 Index������������������������������������������������������������������������������������������������������������������������������ 475

Chapter 1

Introduction to the Technology of Flow and Integrity Management This chapter introduces the broad scope of this book. Here, you will find the concepts you need to know to deal with issues you may encounter. These concepts will help you determine the specific actions needed to reduce negative impacts on the delivery of products to a consumer. Important subjects introduced here include the need for interventions and the economics of interventions. In addition, this chaper provides a review of physical and chemical principles that underlie the processes described in this book. Appendix A of this book contains a glossary of terms that are especially relevant to pipelinerelated usages, given that various names are used for the same cleaning and assessment tools and other devices in different parts of an overall supply system, as well in different locations within the petroleum industry worldwide. In addition, various hydrocarbon products are commonly known by a variety of names. Some of the definitions come from the Schlumberger Glossary (Schlumberger 2012), PAPA (2013), and the Pipeline and Hazardous Materials Safety Administration (PHMSA) (PHMSA 2016). These references (websites) can be accessed for definitions of important additional pipeline or oil/gas terms. A list of symbols used in this document is in Appendix B. Acronyms are defined the first time they occur in each chapter. All the subjects introduced in this chapter are discussed in more detail in later chapters, of course. Maintaining flow [also known as flow assurance (FA)] and ensuring the physical integrity of the pipelines and facilities themselves are core necessities for providing useful products to the ultimate consumers, as well as encouraging a healthy worldwide economy. The fluids are extremely variable. They frequently consist of multiple phases and are subject to unpredictable changes in pressure, temperature, and composition. Therefore, the formation of solids and emulsions is a constant threat. Solids that result from equilibrium changes can decrease the effective diameter of the pipe, completely block the pipe, or change the viscosity of the fluid. “Wet”-gas lines are frequently impeded because of the amount of condensed low-molecular-weight hydrocarbons present as the temperatures are decreased. Additional equilibrium changes can also cause corrosion or degradation of the pipe wall, leading to a leak, loss of product, and possibly major environmental damage and loss of life. Many chemical and mechanical processes are available and in current use to maintain the constant flow of vital products and to assure the integrity of this important infrastructure. This book has been written to provide an improved understanding of the role of chemical reactions and mechanical devices for enhancing and maintaining the conveyance of oil, gas, and related products. Among the many books that describe the thousands of chemicals used in the pipeline industry are those by Fink (2003, 2012) and Kelland (2009, 2013). In addition, there are the works of Frenier and Ziauddin (2014), Frenier and Ziauddin (2008), and Frenier et al. (2010), which describe many of the chemical

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 1

03/11/17 6:50 PM

2  Chemical and Mechanical Methods for Pipeline Integrity

reactions that occur in the pipeline environment. In this book, the focus is on the application of these chemistries to the maintenance of a productive pipeline environment. This study will demonstrate the synergism between these chemical and mechanical processes and will make the case for the necessity of using mechanical devices in these applications. Books that describe additional details of pipeline maintenance in general include Mohitpour et al. (2010), which is part of the pipeline series from the American Society of Mechanical Engineers (New York). A book by Mokhatab et al. (2006) covers many discussions of gas transmission technologies, including details of gas treatment chemistries and plants. The extensive “handbook” by Revie (2015) contains 52 peer-reviewed chapters that cover various aspects of integrity and safety of both production and transmission pipelines. Kennedy (1993) also describes aspects of pipelines, including types of pipelines, pipe manufacture and coating, and fundamentals of pipeline design, as well as types of pumps and compressors in use when his book was published. 1.1  Description of Pipelines and Operating Environments API (2007) described several categories of petroleum/gas-related pipelines. Crude oil gathering lines (approximately 2- to 8-in. outer diameter (OD) collect product from various wells, transmit it to collection/treatment facilities, and then move it onward into “trunk” (transmission) lines. These can be approximately 8- to 25-in. OD, and a few (such as the Alaskan pipeline) may be as much as 48-in. OD. The larger lines may cross state, territorial, and international borders. Most pipelines on land are buried from several up to tens of feet in the ground. In specific areas such as the Alaskan pipeline or for some river crossings, pipelines may also be placed above ground. Subsea gathering lines and trunk lines are usually on the sea floor, but they may also be buried. Risers then convey the well products to surface (or subsea) treatment facilities. There are more than 40,000 miles of crude oil gathering lines and 55,000 miles of trunk lines just in the US. Once at the treatment facilities, the crude oil is refined into hundreds of different products. Natural gas [and natural gas liquids (NGLs)] is collected in more than 200,000 miles of gathering lines (PHMSA 2012) in the US and transmitted in more than 270,000 miles of larger (20- to 42-in.OD) transmission lines. For home and industrial use, the transmission lines connect to distribution systems of local/regional gas companies (API 2007a). Note that gas is usually treated near the wellhead and oil fields to remove water, NGLs, and acid gases (Chapter 2 explains this). Worldwide as of 2010, according to Mohitpour et al. (2010), there were more than 500,000 miles of gas pipelines alone. PHMSA (2012) provided a map (Fig. 1.1) that shows the general location of gas (red) and hazardous liquid (blue) pipelines in the US. The US Central Intelligence Agency (CIA 2008) reported that as of 2008, there were approximately 62,000 miles of lines in Canada, 160,000 in Russia, and 40,000 in China. As a result of the increases in production from shale oil/gas wells, the need for new pipelines of all types is expected to increase. INGAA (2011) predicted the need for new natural gas lines through 2035 and estimated new needed infrastructure (see Fig. 1.2). This increase in production from shale formations is increasing the flow of liquid hydrocarbons that also will stimulate the construction of thousands of miles of new lines to support that part of the oil/gas industry (Boyer et al. 2011; PI 2012). Note that this includes 16,000 miles/yr of new gathering lines that are especially susceptible to blockages and corrosion (see Sections 2.3.1 and 3.2). In addition, Tubb (2013) estimated that as of 2013, more than 116,000 miles of pipelines were planned or under construction worldwide. Unless liquefied, compressed (packed), or stored in underground cavities or old wellfields, the gas is used as produced and is more difficult to store than liquid hydrocarbons. In some mixed hydrocarbon wellfields, the gas is reinjected to maintain reservoir pressure and has been flared on-site if storage or transmission lines are not available (Fig. P-2). As noted in this book’s preface as well as in Jacobson (2014), pipelines are one of the most economical and energy-efficient methods for transporting any bulk product (see Fig. 1.3).

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 2

03/11/17 6:50 PM

Introduction to the Technology of Flow and Integrity Management  3

Pipelines Hazardous liquid Gas transmission

Fig. 1.1—Gas and hazardous pipelines in the US (PHMSA 2012).

Added Gas infrastructure Inter-regional pipeline capacity (Bcfd)

2011-2020 29

2011-2035

Average annual

43

1.7

Miles of transmission mainlines (1,000s)

16.4

35.6

1.4

Miles of laterals to/from power plants, storage fields and processing plants (1,000s)

6.6

13.9

0.6

Miles of gathering lines(1,000s)

491

1,043

42

30

29

30

142

304

12

22

22

20

592

1,518

61

4

4

4

3,039

4,946

197

Average pipe size (in.) Inch-miles of laterals to/from power plants, storage fields, and processing plants (1,000s) Average pipe size (in.) Inch-miles of gathering lines (1,000s) Average pipe size (in.) Compression of pipelines (1,000 HP) Gas storage (Bcf)

NA

589

24

Processing capacity (Bcfd)

18.1

32.5

1.3

Fig. 1.2—Added natural gas infrastructure to 2035 (INGAA 2011).

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 3

03/11/17 6:50 PM

4  Chemical and Mechanical Methods for Pipeline Integrity

Fig. 1.3—Costs and energy of transportation of bulk products (Jacobson 2014).

Pipelines are highly regulated by US states, the US federal government, and international organizations (see Section 3.5), with the safety and integrity of the lines a constant goal of all stakeholders. General information on pipelines and approximately 30 references to pipeline systems are found in Petrowiki (2014). 1.2  Need for Chemical and Mechanical Enhancements to Pipelines and Facilities The drivers for needing chemical and mechanical interventions and enhancements are the overlapping concepts of flow assurance (FA) and integrity management (IM). Brown (2002) described FA as the “production operation that generates a reliable, manageable, and profitable flow of fluids from the reservoir to the sales point.” FA is especially critical for deepwater assets. Because of limited access to the seafloor infrastructure in deepwater areas, blockages and corrosion in tubulars and lines from deposit formation or from other causes may lead to expensive workovers (Brown 2002). IM is specifically associated with the risks of a pipeline failure. Stephens and Playdon (1998) noted that pipeline characteristics (or attributes) must first be evaluated to produce a line-specific estimate of the failure probability for each segment within the system as a function of failure cause, which might be • • • •

Metal-loss corrosion Mechanical damage Ground movement Crack-like defects

Then, an estimate of the potential consequences of segment failure must be made in terms of three distinct consequence components: lifetime line safety, environmental damage, and economic impact. Cause-specific failure probability estimates are then multiplied by a global measure of the loss potential associated with the different consequence components to produce a single measure of operating risk for all failure causes associated with each segment. Segments can then be ranked by failure cause and according to the estimated level of risk. The author of this book contends that IM

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 4

03/11/17 6:50 PM

Introduction to the Technology of Flow and Integrity Management  5

is inherently a part of FA because, for example, a failed, leaking pipeline segment cannot effectively and efficiently move products to the consumers. In many countries, including the US and Canada, various inspections are required by governmental regulations (PHMSA 2012). Idem (2015) claimed that this was a response to a number of serious pipeline ruptures that had devastating environmental effects on land and water. Thus, the US Department of Transportation PHMSA announced proposed regulations to require that all hazardous liquid pipelines have a system for detecting leaks and establish a timeline for inspections of affected pipelines following an extreme weather event, natural disaster, or operator negligence. President Barak Obama’s transportation secretary, Anthony Foxx, said, “Hazardous liquid pipelines crisscross the country and pipeline failures can have profound impacts on local communities and the environment.” He declared, “This proposed rule is an important step forward to enhance safety, and protect people and the environment.” The mechanical and chemical enhancements and interventions described in this book are an integral part of both FA and IM by locating, anticipating, and assessing possible damages and interruptions of flow conditions and then providing proactive solutions to prevent or remedy the situations. Details of IM systems are provided in Section 3.5. 1.3  Economic and Market-Related Forces Affecting Pipeline Maintenance This section includes a short review of the economic issues that affect most oil/gas treatment projects as well as concerns specific to the pipeline industry. 1.3.1  General Economic Issues for Oilfield Treatments. Chemical and mechanical treatments are performed for many different reasons, and the general market for all oilfield-related chemicals may exceed USD 15 billion (Freedoniagroup 2008). FA methods such as the use of corrosion, scale, and organic deposit inhibitors, as well as the application of demulsifieres, are used to prevent problems from occurring. Many chemicals are also used in stimulation, in enhanced-oil-recovery (EOR) activities (Kelland 2009; Frenier and Ziauddin 2014), and in pipeline infrastructure maintenance. In the case of tight formations, hydraulic-fracturing methods are used initially to cause enough conductivity to allow the well to produce economical amounts of hydrocarbons. Fracturing activities, along with directional drilling in shale oil/gas formations may produce enough hydrocarbons to change the entire energy calculus of countries and regions of the world. Jaffe (2010) said that because of these methods, there may be enough gas potentially available to change some geopolitical balances of power. Also, discoveries of retrievable liquid hydrocarbons associated with various shale plays (RRCT 2011) may likewise greatly affect hydrocarbon supplies and economics and pipeline activities (see Fig. 1.2). For all the processes used, there is an economic element that also includes environmental and political questions (Hess 2010). These economic issues must be taken into account in business calculations before a decision is made to pump chemicals or to perform pipeline maintenance activities. Initially, a calculation must determine whether to consider any enhancements. Then one must decide which method should be used. Deciding which type of treatment to use may involve more than economic issues, of course. Examples of other issues include the formation type, location, equipment availability, time constraints, local availability of chemicals, and urgency of treating a problem. If enough information is available to add up all the potential costs as well as the loss/gain of products, then the economics equations described next can be used. Economides and Boney (2001) discussed calculation methods and decisions that should be made to decide on a project. They examined several indicators, shown here. Payout Time. Payout time includes the total costs associated with the project, but not the value of the money or the profit. This is a measure of liquidity for the project. Net Present Value. Net present value (NPV), shown in Eq. 1.1, is the definition for cumulative discounted cash flow (CDCF). The NPV is the maximum of this CDCF. NPV gives a dollar value added to the property at present time. If it is positive, the investment is attractive; if it is negative,

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 5

03/11/17 6:50 PM

6  Chemical and Mechanical Methods for Pipeline Integrity

it means an undesirable investment. According to Economides and Boney (2001), NPV is the most widely used indicator showing a dollar amount of net return. Here ∆USDn is the incremental revenue (minus the incremental expenses and taxes that are a result of operations), n is the time period increments (e.g., years), and i is the interest rate. ∆USD n − cost .���������������������������������������������������������������������������������������������������������� (1.1) n n =1 (1 + i ) n

NPV = ∑

Rate of Return. The rate of return (ROR) is a comparison with other investments and is determined by setting i = 0. Return on Investment. The payback is the total amount of money earned from the investment in the project. This is the return on investment (ROI). Investment relates to the amount of resources put into generating the given payback. ROI =

( Payback − Investment ) • 100 .�������������������������������������������������������������������������������������� (1.2) Investment

Corporate Goals and Risk. Corporate goals and risks are less tangible, but are important considerations for judging projects. These also include matching or exceeding the capabilities of the competition in some markets. Risks may include environmental or ecotox issues. 1.3.2  Pipeline-Specific Economic Considerations. Pipeline maintenance includes the use of chemicals and mechanical equipment for FA as well as for IM. Many projects will require consideration of the general guidelines described in Section 1.3.1. However, calculations of conditions that lead to a failure, especially calculations of corrosion susceptibilities, are mandated by governmental regulations (Section 3.5) as well as by good business practices. Several specific examples are noted here. NACE (1994; 2011) described two additional economic calculations. These include the discounted payback period (DPBP) and benefit/cost ratios, explained next. Discounted Payback Period. The DPBP method requires a calculation of discounted cash inflow (DCI). DCI =

(Actual Cash Inflow) ,���������������������������������������������������������������������������������������������������� (1.3) (1 + i)n

where I is the discount rate and n is the time period in which the cash inflow happens. Then, DPBP = A +

B . ������������������������������������������������������������������������������������������������������������������������ (1.4) C

A is the last period with negative discounted cumulative cash flow, B the absolute value of discounted cumulative cash flow at the end of period A, and C the discounted cash flow during the period after A. Benefit/Cost Ratios. The benefit/cost ratio is the ratio of the benefits of a project or proposal, expressed in monetary terms, relative to its costs, also expressed in monetary terms. All benefits and costs should be expressed in discounted present values (see Eq. 1.3 and 1.4; NACE 2011). Also note that Appendix A and Appendix B in NACE (2011) show examples of these calculations made on the basis of pipeline examples. This book (NACE 2011) has more than 100 references that may be of value for these discussions. Wei et al. (2009) described an integrated approach, made on the basis of corrosion modeling and laboratory testing, to optimize the use of carbon steel in corrosive service for applications such as

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 6

03/11/17 6:50 PM

Introduction to the Technology of Flow and Integrity Management  7

downhole tubulars, pipelines, and facilities. This approach presents economic advantages, such as reducing the use of expensive corrosion-resistant alloys while ensuring the operational integrity of equipment and facilities. A key part of this integrated approach is to apply reliable corrosion models underpinned with laboratory data (see Section 3.4). These authors claimed that to be most effective, the models should account for the relevant chemistry and physics of the corrosion process, including the effects of detailed water chemistry, liquid hydrocarbons, and the degree of protection from iron carbonate or iron sulfide scales (these pipeline conditions and chemistries are described in Section 3.1 of this book). Thus, in their report Wei et al. (2009) claimed that ideally, models should account for variations in conditions and flow characteristics along the length of a wellbore or pipeline. Additional information on modeling is in Section 3.6. Dawotola et al. (2011) proposed a data-driven approach to find the optimal inspection interval (see Section 3.5) for a petroleum pipeline that is subject to long-term corrosion failure. This approach is claimed to account for the determination of both the failure frequency and consequences of failure. Three forms of corrosion are studied for use in the model: uniform corrosion (Section 3.1), pitting corrosion (Section 3.2.2), and stress corrosion and other forms of cracking (Section 3.2.2). Failure frequency is estimated by fitting historical pipeline failure data into either a homogeneous Poisson process (Wikipedia 2014b)—a stochastic program that counts the number of events and the time points at which these events occur in a given time—or a power-law process. The consequences of corrosion attack are calculated in terms of economic loss, environmental damage, and human safety and are determined for small leaks, large leaks, and rupture of pipeline. Failure frequency and consequences are both used to estimate total loss resulting from pipeline operation. A risk-based IM optimization of the pipeline is obtained by minimizing the economic loss of pipeline, taking human risk and maintenance budget as constraints. 1.4  Review of the Physics and Chemistry of Pipeline Interventions The next sections (1.4.1 through 1.4.6) provide an introduction to the mechanical and chemical principles that control and inform all the maintenance processes described in the subsequent chapters of this book. For those readers skilled in the art and sciences of pipeline operations, this section (or parts of it) is optional and may be skipped. The technologies reviewed here include pipeline construction materials, simplified fluid dynamics, viscosity, rheology, simplified thermodynamic principles, surface chemistry, and testing of well/pipeline fluids. These principles are applied in the remainder of this book to explain and connect the various processes discussed. 1.4.1  Pipeline Materials for Construction and Pressure Requirements. Most oilfield pipe goods are constructed using various carbon steels and must comply with ANSI/API SPEC 5L (2011) and/ or ANSI/NACE MR0175-2009 (2009) standards, depending on the expected product, temperatures, and flow rates. High-strength carbon steel pipes (up to 80 ksi) are in use onshore and subsea. For some subsea pipelines or facilities in sour or high carbon dioxide (CO2) service, 13% Cr martensitic steel may be specified (see Section 3.1 for more details on metallurgy’s role in corrosion). NACE SPO106-2006 (2006) provides design considerations that apply to pipelines made of steel used to transport natural and manufactured gas, crude oil, and refined products for the control of internal corrosion. A corrosion specialist should be consulted during pipeline design and construction. Two major recommendations are to completely determine the compositions of the gases, liquids, and other fluids and to design for flow velocities that mitigate corrosion and pipeline cracks. Chapter 4 of Heidersbach (2011) reviews the materials and metallurgy of oilfield equipment. He noted that pipeline metals (mostly various steels) are specified in API SPEC 5L (2011). As with most oilfield materials, specifications are made on the basis of performance standards, not chemical specifications such as SAE (2015) standards (e.g., for 1010 carbon steel) that have maximum or minimum elemental limits. He also noted that most line piping is composed of “low-carbon steel” (less than 0.3% C). Other chemical elements may be added (Mn and small amounts of Cr, Ni, for example)

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 7

03/11/17 6:50 PM

8  Chemical and Mechanical Methods for Pipeline Integrity

to produce a “high-strength/low-alloy steel” if field specifications are to be met. Chapters 3, 4, and 5 in Revie (2015) discuss various aspects of pipeline design to address stress and fracture control. Here, Section 3.1.1 provides additional details of the metallurgies of pipeline steels. Also, Petrowiki (2014) provides a comprehensive list of pipeline metal grades, along with references. Most pipeline segments are connected using various welding methods—these include electrical as well as laser methods (see Komizo 2008)—and the welds require extensive inspections and testing. Fig. 1.4 shows a drawing of the weld bead, the heat-affected zone (HAZ), and possible inclusions (or voids) and cracks that could affect reliability. The weld bead and the HAZ are subjected to temperatures that will change the metal’s microstructure (and the weld bead is essentially a casting), so these differences can affect the corrosion and mechanical properties compared to the base metal. Various heat treatments, including annealing, quenching, tempering, and normalizing, can be used to reduce some effects of the welding and forming processes (see Heidersbach 2011). In the US, many requirements for the allowed pressure ratings of pipelines are driven partially by Class Location F Department of Transportation regulations (see Sec1 10 or less occupied buildings 0.72 tion 3.5). Wint (2011) describes the Title 49 CFR 2 Greater than 10, but less than 46 0.6 §192.105 (USDOT/OPS 2004) requirements for 3 More than 46 0.5 design of steel pipe in service in high-consequence areas (HCAs) such as urban environments. The most 4 Where buildings of more than 4 0.4 stories are prevalent restrictive requirements (Table 1.1) have the lowest value for the maximum-allowed design operating Table 1.1—HCA location class. pressure (MAOP). This value is calculated with  2St  MAOP =   ( E • F • T ) . �������������������������������������������������������������������������������������������������������� (1.5)  D  The values in Eq. 1.5 are MAOP = maximum-allowed design operating pressure, S = yield strength [specified minimum yield strength (SMYS)], D = outside diameter, t = wall thickness, F = design factor HCA Class, §192.111, E = longitudinal joint factor, §192.113, T = temperature derating factor, §192.115. Weld Bead

Heat-affected zone Pipe wall

Defects Cracks Fig. 1.4—Pipeline weld and defects.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 8

03/11/17 6:50 PM

Introduction to the Technology of Flow and Integrity Management  9

There are various factors that can affect the thickness of the pipe (such as corrosion damage; see Chapter 3) and may cause the pipe to be derated (also see DNV-RP-F101 2010; Leewis 2003). Ashby (2013) noted that the classification of the HCA location classes may change in the future, so more miles of piping will fall into the most restrictive class. Note that the complexities of the regulations are beyond the scope of this book; furthermore, current regulations should always be reviewed carefully when making these calculations. These restrictions and classification are covered in Chapters 3 and 5. Polymers (plastic-fiberglass) as well as polymer-coated steel pipe segments are being placed in many sections of operating and transmission environments. The different general types of polymers and coatings include thermoplastic resins that soften when heated and thermosetting resins that undergo a chemical reaction during the formation process. Thermoplastic plastics include high-density polyethylene, polyvinyl chloride (PVC), and fluoro polymers. Thermosetting plastics include epoxies, polysilicones, and polyurethanes. Elastomers are rubbery thermosetting plastics that may have less crosslinker and possibly other additives that allow deformation for use in seals, packers, or other such materials. Also, mixtures of resins with fiberglass or carbon fibers are present in some uses. The benefits of plastic piping may include lower weight and better corrosion resistance, as well as easier connections than otherwise possible. However, many FA issues such as scaling and emulsions may affect flow in polymer/resin-coated lines. In addition, pressure limitations and cost considerations may dictate the choice between steel and resins. Plastic piping, including fiberglass, is used in some downhole applications and for transportation of oilfiled water and flowback fluids. Pipeline systems also involve a number of ancillary devices that include pumping/compression equipment, valves, taps, pig-launching and -retrieval equipment, and devices for introduction of different treating chemicals. Fig. 1.5 is a diagram of a generic system of wells, gathering lines, and flowlines that terminate in a separation/treating facility or a refinery (Idachaba 2016). This diagram, while based on a design for a land-based oil/gas system, also applies to subsea systems. These systems are all designed for gas, crude oil, and mixed well fluids. Pigging problems associated with valves and the like are described in Section 5.2.2. If the pipeline is a subsea system (Albert et al. 2011), “risers” (or pipes) may convey the various fluids to the surface. Some fluids may also go through subsea piping to injection wells. When all the on-site processing is complete, the crude oil and gas enter pipeline systems (or other methods of transport) that send the products to a “consumer.” The long-distance transition systems will also have periodic pumping/compression stations, innumerable valves, and laterals and other devices for measurement and maintenance. Some of these devices are described in subsequent chapters. 1.4.2  Fluid Mechanics and the Effects of Fluids and Phases on Pipeline Operations. The mechanics of fluid flow in the multiple pipeline segments affects and is affected by the changing nature of the fluid phases (details of the chemistries of the various phases, which include aqueous liquid and hydrocarbon liquids, gases, and solids, are in Section 2.1). The pipelines themselves are also affected by the production of scale, which is inorganic deposits (as noted in Chapter 4). The fluid mechanics at any point in the pipeline system and the presence of deposits will affect the causes of corrosion damage (see Chapter 3), and pressure and flow changes affect deposition of scale. Thus, there is a feedback loop in which mechanical and chemical conditions influence each other. All these conditions must consequently be understood and anticipated for successful FA and IM. The following subsections provide an introduction to single-phase and multiphase flow in pipes. More details are in Brill (1987), Asante (2002), Shoham (2006), and Falcimaigne and Decarre (2008). Single-Phase Flow. Single-phase flow equations and concepts provide a starting point for the much-more-complex multiphase flow conditions that actually exist in many pipeline segments.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 9

03/11/17 6:50 PM

10  Chemical and Mechanical Methods for Pipeline Integrity

a) Typical pipeline units (and key) Well location

Flow station

1

2

Export terminal/ refinery

Main manifold

3

4 Pipelines

Delivery lines

Flowlines

Row = Pipeline right of way

Small manifold

5 Trunk lines

b) Details of a complex well field and pipeline organization 1

Well locations

2

5

2

1

Row 3

Row

Row

Oil field

Row

Row

Small manifold

4

Flow station Row 2

4 Main manifold

1

Row

Access roads

Fig. 1.5—Generic layout of a pipeline system (Idachaba 2016).

The pressure drop over a distance, L, of a single-phase incompressible fluid can be obtained from the mechanical-energy-balance equation (Economides et al. (1994) as ∆pL = ∆pPE + ∆pF + ∆pKE.���������������������������������������������������������������������������������������������������������� (1.6) In an expanded form, it becomes 2 f f ρν 2 l  g ρ ∆pL =   ρl sin θ + + ∆ν 2.�������������������������������������������������������������������������������� (1.7) gC d 2 gC  gC  Note that ∆pL is the difference between the upstream and the downstream pressure and ∆pPE is the pressure drop because of potential-energy change. It is the hydrostatic head of the fluid and accounts for the pressure change caused by the fluid column weight. Also note that r is the fluid density and q is the pipe deviation from horizontal. Thus, the hydrostatic head for a horizontal pipe (q = 0°) is zero. For a vertical tubing, q = 90° for upward flow and q = –90° for downward flow. For fresh water, the potential-energy pressure drop per foot of vertical distance is 0.433 psi/ft. ∆pF is the pressure drop resulting from pipe friction and is obtained from the Fanning equation (John Thomas Fanning, 1837–1911): ∆pF =

2 f f ρν 2 gc d

,�������������������������������������������������������������������������������������������������������������������������� (1.8)

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 10

03/11/17 6:50 PM

Introduction to the Technology of Flow and Integrity Management  11

where v is the fluid velocity and d is the pipe diameter. For a pipe with a constant cross section, the fluid velocity can be expressed in terms of flow rate, q, as

ν=

4q .������������������������������������������������������������������������������������������������������������������������������������ (1.9) π d2

The Fanning friction factor, ff, for laminar flow is ff =

16 4 π dµ = ,������������������������������������������������������������������������������������������������������������������ (1.10) qρ N Re

where the Reynolds number of the flow is N Re =

4 qρ qd = . ������������������������������������������������������������������������������������������������������������������ (1.11) π dµ ν A

Substituting v and ff in the expression for ∆pF yields the Hagen-Poiseuille law for pressure drop in a pipe (Hagen 1839; Poiseuille 1840): ∆p f =

128µ ql .������������������������������������������������������������������������������������������������������������������������ (1.12) π gc d 4

From the foregoing definitions and equations, we can see that for a given flow rate, the pressure drop is proportional to 1/d4 for laminar flow, and for turbulent flow in a smooth pipe, the pressure drop is proportional to 1/d4.75. Hence, any reduction in effective pipe diameter from deposit buildup can lead to a dramatic increase in pressure drop and hence to a similar decrease in the production from the reservoir. For turbulent flow in a smooth pipe, ff = 0.079/NRe0.25. In Eq. 1.6, ∆pKE is the pressure drop caused by change in kinetic energy between various positions in the pipe. Generally, it is much smaller compared to ∆pPE and ∆pF. It is equal to zero if there is no change in fluid velocity between the two points of measurement. For example, given an incompressible fluid f flowing through a pipe of uniform cross-sectional area, the velocity does not change and ∆pKE is equal to zero. In cases of high gas volumes or high gas/oil ratios, a rapid change in velocity may occur, but even then ∆pKE generally accounts for less than 10% of the pressure loss. Elements in Eqs. 1.6 through 1.12 have the following values: g = gravitational acceleration, m/s2, gc = gravitational acceleration at the center of mass, m/s2, ff = Fanning friction factor, dimensionless, q = flow rate, m3/s, l = length of pipe, m, A = area of pipe, m2, d = pipe inside diameter (for round pipe), m, v = flow velocity, m/s, r = density, kg/m2, m = dynamic viscosity, kg/(m·s), n = kinematic viscosity (m/r), m2/s. Note that in the various segments of the pipeline industry, expressions used in pipeline models differ; however, NRe is universally important because the transitions from laminar to turbulent flow is generally described by this number with laminar flow conditions 4,000,

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 11

03/11/17 6:50 PM

12  Chemical and Mechanical Methods for Pipeline Integrity

Laminar flow Clean pipe Slightly turbulent flow Dirty pipe

Turbulent flow Chocked-down pipe

and a transition range between these values. The Darcy friction factor (fD), also called the DarcyWeisbach friction factor, is used in some applications and equals 4(ff) or 64/NRe. In addition to velocity, pipe diameter, and viscosity, other factors can affect the flow regime. Shannon (2010) claims that the presence of debris or scale in a pipe can change the flow patterns. The preceding equations assume a smooth pipe, while Fig. 1.6 shows three possible conditions for the same velocity and viscosities. Thus, in this set of conditions, there is laminar flow for a smooth pipe, slightly turbulent flow for a “dirty” pipe, and fully turbulent flow when the diameter has constricted enough to cause this flow condition. The friction factor equation (Eq. 1.10) will also change for a dirty, rough pipe. An equation to calculate the change is

Fig. 1.6—Effects of deposits on flow.

 ε 1 2.51  = −2 log10  +  .�������������������������������������������������������������������������������������������� (1.13) f  3.7d N Re f  In Eq. 1.13, e is the height of the roughness (mm). Although this can be measured for laboratory measurements, it is usually calculated from the observed pressure drop in a line (see Eq. 1.14). Multiphase Flow. Understanding the multiphase flow conditions in the lines is an important aspect in predicting corrosion and solids accumulations, as well as distribution of inhibitors in pipelines. Shoham (2006) stated that the hydrodynamics of single-phase flow is well-understood and that pressure drop vs. flow rate and heat transfer can be calculated straightforwardly (see flow rate equations in the previous subsection, Eqs. 1.6 through 1.13). However, adding a second (or third phase) greatly complicates the analyses. This short section provides an introduction to the different flow regimes found in piping and modeling methods to predict hydrodynamic problems. Shoham (2006) and Asante (2002) are recommended for many more details of the methods for calculating important variables such as the friction factor and the pressure drop during multiphase flow conditions. Different multiphase flow combinations in piping include • • • •

Hydrocarbon liquid/water Hydrocarbon gases/hydrocarbon liquids Hydrocarbon liquid/water/gases Combinations of phases

The gas phases may also include acid gases such as acetic acid (HAc), hydrogen sulfide (H2S), and CO2, and temporally dispersed phases of water and liquid hydrocarbons will form. Because of the density and viscosity differences between the different phases, they will have different mass and volume velocities, and these differences cause the complex regimes to form. Fig. 1.7, adapted from Shoham (2006) and Kee et al. (2015), depicts several conditions that may occur given the rate of flow in pipelines experiencing horizontal multiphase conditions. This figure shows water and oil/gas phases. As the flow rates increase, various different regimes form. Some possible regimes from the papers by Shoham (2006) and Kee et al. (2015) are described below with letters referencing Fig. 1.7. Note that the Fig. 1.7 diagram shows the continuous phases as well as bubbles and droplets of liquids temporally dispersed. These conditions can affect corrosion

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 12

03/11/17 6:50 PM

Introduction to the Technology of Flow and Integrity Management  13

as well as the application of inhibitors. Two-phase flow follows a similar sequence, but without the second liquid phase. Flow direction Key Gas

a) Stratified (ST)

Water Oil Gas bubble

b) Elongated bubble (EB)

Water droplet Oil droplet

Increasing flow velocities

c) Slug (SL)

d) Wavy annular (WA)

e) Annular mist (AM) Fig. 1.7—Horizontal multiphase (three-phase) flows in gas/liquid pipelines (after Shohan 2006 and Kee et al. 2015).

Stratified Flow (Fig. 1.7a). Stratified flow is characterized by the concurrent flow of liquid streams at the bottom and a gas stream at the top of the pipe. The two liquid phases are often separated or slightly dispersed at the oil/water interface. The gas/liquid interface may be smooth or show some waviness caused by the drag of the gas passing over the liquid. Elongated Bubble Flow (Fig. 1.7b). Elongated bubble flow is also called plug flow and is a form of intermittent flow that occurs at low gas velocity. Slug Flow (Fig. 1.7c). Slug flow occurs when the liquid bridges the entire pipe cross section, forming a liquid slug, while the gas flows as a large bubble between the trains of liquid slugs. The large gas bubble moves on top of a slower-moving stratified liquid layer characterized as the gas-bubble/ liquid-film zone. Smaller bubbles of gas as well as droplets of oil and water are dispersed in the various phases. Wavy Annular Flow (Fig. 1.7d). Wavy annular flow occurs at the transition between slug and annular flow. The flow lacks the characteristic pressure fluctuation found in slug flow. The upper wall is occasionally wetted by an unstable liquid film that keeps falling diagonally downward. Annular Mist Flow (Fig. 1.7e). Annular mist flow occurs at very high gas velocity when gas flows at the pipe core and liquid moves as an annular film enveloping the pipe wall. The turbulent gas contributes to rough gas/liquid interfaces containing interfacial waves of varying amplitudes. When liquid-containing lines, such as the current Alaskan pipeline (Abrams 2011), are not operating at capacity, the lines will not be full and three-phase flow may exist. Some of these lines could contain solid particles as a result of changing conditions or the introduction of contaminants. The flow regimes described in Fig. 1.7 may be present when lines are horizontal. Kesana (2013), Thome (2012), and Brill (1987) describe these flow regimes. These authors describe plug flow as liquid

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 13

03/11/17 6:50 PM

14  Chemical and Mechanical Methods for Pipeline Integrity

plugs separated by elongated gas bubbles that are smaller than the tube diameter. Gas Gas Gas Because the bubbles are more buoyant than the liquid, there is a continuous film of Gas Gas Condensate Condensate liquid in the tube bottom. As H 2O H2O a line is inclined, changes may occur as a result of flow rate changes. Fig. 1.8a depicts a flow regime of liquids and a b) Types of flow patterns in vertical pipes gas and shows the effects of terrain. Here, the hills and valleys may cause liquids to remain in the line bottom, especially at low flow rates. This may lead to corrosion in the low points or to additional backpressure on the line caused by holdup of the liquids. Fig 1.8b shows flow regimes in vertical pipes (see Brill 1987; Thome 2012). In these vertical orientations the regimes can include bubble flow, slug flow, churn flow (a form of intermittent flow in which periBubble Slug Churn Annular odical flow reversal of flow flow flow flow liquid film is observed), and annular mist, dependFig. 1.8—Effects of terrain (a) and vertical flow regimes (b) (Brill 1987 ing on the flow velocities. and Thome 2012). Kesana (2013) claimed that slug flow, especially if it is carrying solids, can result in severe metal loss by erosion (erosion and erosion/corrosion of metals are described in Section 3.2.3). Wang et al. (2013) provided additional information on the variety of regimes that occur in multiphase flow in the piping. This group studied a range of flow conditions and pressure drops in a loop described in this report. They also used a lower °API value (28.5 °API) and higher-viscosity (dead-oil viscosity is 1.1 Pa·s at 15.6°C) oil phase than is usual (see Pan et al. 1995). In addition, these investigators used natural gas (instead of a more inert gas) at a range of elevated pressures as well as changing oil flow, gas flow, and water flow velocities. They also used a sapphire window to observe the phases. A series of flow regimes were identified and are depicted in Fig. 1.9. Types of flow in which the oil is dispersed in the water are depicted in Figs. 1.9a and 1.9b. Conditions of water dispersed in oil are shown in Fig. 1.9c. In all cases, the tests had horizontal flows under intermittent and slug flow conditions. Modeling Multiphase Flow Conditions. Griffith (1984) reviewed qualitative aspects of predicting the pressure drop in pipelines with multiphase flow. He claimed that the most important a) Effects of terrain

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 14

03/11/17 6:50 PM

Introduction to the Technology of Flow and Integrity Management  15

factor is the friction factor (see Eq. 1.10), but he recommended using Moody curves (Moody 1944), assuming that the fluid is flowing at the velocity of the mixture. A Moody diagram showing the Darcy friction factor (fD) plotted against Reynolds number for various roughness factors is shown as Fig. 1.10 (from Wikipedia 2013b).  d 2  fD =    2  ∆p .���������������������������������������������������������������������������������������������������������������� (1.14)  L   ρq  Asante (2002) described some of the models used to calculate the consequences of multiphase flow in gas pipelines. They parallel the regimes seen in in Fig. 1.7. Computations using single-phase and multiphase models are in use to estimate important flow properties. Shoham (2006) explained that the modeling methods can be divided into “empirical-physical models,” computational models (including computational fluid dynamics), and mixed models. The methods describe next have used experiments as well as some level of computational methods to describe the flow regimes occurring at different points in a pipeline. Single-Phase Approaches for Multiphase Flow Models. These models use calculation of observed friction factors ( f in the Moody diagram shown in Fig. 1.10) and pressure drop (∆P) for a tested fluid mixture and then use the single-phase equations (Eqs. 1.6. through 1.11) for predictions. (a)

Gas phase with thick oil film on the wall

Oil and gas dispersed in water

Oil layer

INT(O/W-S&SOW-F) Horizontal intermittent flow with O in W dispersion slug and stratified oil and water film

Water layer

Slug body region

Liquid film region

(b)

Gas phase with thick oil film on the wall

Oil and gas dispersed in water

INT(O/W-S&O/W-F) Horizontal intermittent flow with O in W dispersion slug and oil and water dispersion film

Oil dispersed in water layer

Slug body region

Liquid film region

(c)

Gas phase with thick oil film on the wall

Water and gas dispersed in oil

Water dispersed in oil layer

Slug body region

INT(W/O-S&W/O-F) Horizontal intermittent flow with W in O dispersion slug and water and oil dispersion film

Liquid film region

Fig. 1.9—Various flow patterns in three-phase flow with heavy oil and natural gas (Wang et al. 2013).

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 15

03/11/17 6:50 PM

16  Chemical and Mechanical Methods for Pipeline Integrity

Moody Diagram 0.1 0.09 0.08 0.07

Transition region

0.06 0.05

0.02 0.015 0.01

Friction Factor, f

0.04

0.005

0.03

0.02 0.015

0.01

Laminar flow 64 Re Material Concrete, course Concrete, smooth Drawn tubing Glass, plastic Iron, cast Steel, mortar lined Steel, rusted Steel, forged Sewers, old Water mains, old

103

0.002 0.001 ε(mm) 0.25 0.025 0.0025 0.0025 0.15 0.1 0.5 0.025 3.0 1.0

104

5×10–4 2×10–4 10–4 5×10–5

Complete turbulence

Smooth pipe

105

106

107

Relative Pipe Roughness, ε/D

0.05 0.04 0.03

10–5 5×10–6 10–6 108

Reynolds Number, Re Fig. 1.10—Moody diagram.

Shoham (2006) considered this to be part of the “experimental” approach but noted that it is applicable only under the conditions that are close to those of the experiment. Multiphase Approaches for Multiphase Flow Models. Asante (2002) claimed that the simplest models, termed “homogeneous models,” would apply to the mist flow in Fig. 1.7 and use modifications of the single-phase equations (Eqs. 1.6 through 1.11) and adjust the friction factor ( f ), the density (r), and the viscosity (m). An example of this model is in Fancher and Brown (1963), who discussed homogeneous calculations using the correlation of Poettmann and Carpenter (1952):    dp 1 q2 M 2 = ∆p = − ρ + f  .���������������������������������������������������������������������� (1.15) 10 5  dl 144   7.413 × 10 ρ d   In this expression, p = pressure, psi; L = length, ft; r = flowing density, lbm/ft3; f = Fanning friction factor; d = pipe diameter, ft; q = oil flow rate, B/D; and M = total mass of gas and liquid, lbm. Fancher and Brown (1963) noted that as the gas/liquid ratio increases and the liquid rate decreases, and pressure gradients calculated from Eq. 1.15 are influenced more by the second term of this equation than by the first. Stratified flow and annular mist flow conditions (Figs. 1.7, 1.8, and 1.9) are frequently encountered and are relevant to many FA and IM conditions. They are highly complex, and so the reader is directed to Asante (2002) and Shoham (2006). Wang et al. (2013) describe multiphase flow tests and correlations with the multiphase models of Zhang and Sarica (2006), and the authors found that the water flows show reasonable predictions, whereas the pressure drops and liquid holdups were underpredicted. A three-phase flow model based on experimental as well as computational data was described by Karami et al. (2016). In the nomenclature of Shoham (2006), this would be a mixed model.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 16

03/11/17 6:50 PM

Introduction to the Technology of Flow and Integrity Management  17

To simulate oil, water, and gas flowing conditions, an experimental setup—a 6-in.-ID pipe flow loop (see Fig. 1 in Karami et al. 2016)—used mineral oil, tap water, and air to simulate the gas phase. The experiments were conducted under low-liquid-loading condition, which is commonly observed in wet-gas pipelines. The analyzed flow characteristics included wave pattern, liquid holdup, water holdup, pressure gradient, and wetted-wall fraction. Videos of the wave patterns of the liquid phases were also recorded for analyses. The observed wave patterns included stratified smooth and stratified wavy with 2D waves, 3D waves, roll waves, and atomization flow. The transitions between the flow patterns vary as a function of water cut. The trends of pressure gradient, liquid holdup, and water holdup with respect to gas velocity (vSg) and liquid velocity (vSL), as well as water cut, were observed. Good correlation with several mathematical models was demonstrated. Details of additional flow models that are in use are described in NACE SP0208-2008 (2008), Appendices A and B. Because corrosion, scale, and use of control chemicals depend on the phase that coats the surfaces, the complex models are critical both to prediction and to pipeline treatments. Hilgefort (2014) as well as NACESP 0208-2008 (2008) provided information on multiphase flow models and how the various conditions affect corrosion in pipeline segments (many more details of corrosion and its manifestations are provided in Chapter 3). Hilgefort (2014) stated that flow models help predict important conditions that may cause corrosion. Thus, a first step is determining the potential for accumulation of water and debris if stratified flow exists, given that internal corrosion occurs where water (or hygroscopic solids) comes into contact with the pipe wall. Therefore, if it is known which pipelines are likely to experience water or solids accumulation, the operator knows which are susceptible to internal corrosion. If the operators know where, along the pipeline, the water and debris accumulation will occur, they can perform inspections to check for internal corrosion. The most important factors in the models are the (1) critical water velocities and inclination angles for water and/or solids and (2) critical inclination angles (q ), which are  ∆(elevation)  θ = arcsin  .������������������������������������������������������������������������������������������������������ (1.16)  ∆(distance)  These critical factors are then compared to the elevation and inclination profile of the pipeline (like those in Fig. 1.7d) to find where water or solids may accumulate. Solids moving in a pipeline, Hilgefort (2014) noted, are subject to gravitational forces, which deposit the particles, and also to turbulent forces, which keep the particles in suspension. At lower flow rates, particles tend to settle out and can sit at the pipeline bottom. 1.4.3  Viscosity and Rheology of Fluids. The viscosity of a fluid is an essential property for characterization of flow behavior. This property is a measure of a fluid’s resistance to being deformed by either shear stress or tensile stress. Viscosity describes a fluid’s internal resistance to flow and may be thought of as a measure of fluid friction. Wikipedia (2009) claims that James Clerk Maxwell (1831–1879) called viscosity “fugitive elasticity” because of the analogy that elastic deformation opposes shear stress in solids, whereas in viscous fluids, shear stress is opposed by rate of deformation. Viscosity is a primary determinent of the energy needed to move fluids in pipes. The Hagen- Poiseuille equation (Eq. 1.12) notes that pressure drop is directly proportional to viscosity, and therefore more pumps or larger ones will be required to move a more viscous fluid compared with a less viscous material. Thus, knowledge of the viscosity as a function of the piping and temperatures is critical information for all pipeline operations. Viscosity is a function of the fluid’s chemical composition and physical state (usually temperature). Two different definitions of viscosity are in common use: absolute viscosity and kinematic

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 17

03/11/17 6:50 PM

18  Chemical and Mechanical Methods for Pipeline Integrity

viscosity. Absolute (intrinsic) viscosity (h) is a measure of the resistance to flow that a fluid offers when it is subjected to shear stress, shown as Eq. 1.17. This equation relates the shear stress (s) exerted on a fluid to the resultant strain rate (g ). Kinematic viscosity (υ) is defined as the ratio of absolute viscosity to the density of the fluid at the same temperature. The apparent (measured) viscosity (µa) is the value usually reported in oilfield application and may depend on the shear rate.

σ = ηγ.������������������������������������������������������������������������������������������������������������������������������������ (1.17) The viscosity of the crude oil and the viscosities of various treating fluids described in this book are critical values. In general, the viscosity of crude oil increases with its density—that is, the lower the °API value of a crude oil (denser oil), the higher the viscosity. Typically, the viscosity of dead oil is experimentally determined as a function of temperature, and the viscosity of live oil is determined as a function of pressure at reservoir temperature (for reservoir engineering purposes) and at a lower temperature (for facility design purposes). In addition, the viscosity of various petroleum fractions is important in downstream applications. The viscosity of petroleum fractions also increases with a decrease in the °API value; for residues and heavy oils with °API value of less than 10 (specific gravity above unity), the viscosity varies from several thousands to several million poises. Viscosity is a bulk property that can be measured for all types of petroleum fractions in liquid form. Kinematic viscosity is a useful characterization parameter for heavy fractions in which boiling point data are not available because of thermal decomposition during distillation. Viscosity is not only an important physical property but also a parameter that can be used to estimate other physical properties as well as composition and quality of undefined petroleum fractions (Riazi 2005). Generally, the kinematic viscosities of petroleum fractions are measured at standard temperatures 37.8°C (100°F) and 98.9°C (210°F). As Section 4.6.1 will demonstrate, when the fluid being transported contains several phases, emulsions or dispersions may form. Kalra et al. (2012) showed that emulsions will generally increase fluid viscosity, and this will require the resizing of pumps and other facilities to accommodate the needed flow rates. The viscosity of various treating fluids is also of great importance because it affects the chemical/ physical properties of fluids. The measurement of viscosity is accomplished by several techniques. In general, either the fluid remains stationary and an object moves through it, or the object is stationary and the fluid moves past it. The drag caused by the relative motion of the fluid and a surface is a measure of the viscosity. The flow conditions must also have a sufficiently small value of the Reynolds (Re) number for there to be laminar flow (see the discussions of flow equations in Section 1.5). For Newtonian liquids (i.e., viscosity that is independent of shear rate), viscosity can be measured by capillary U-tube viscometers (Fig. 1.11). This devices measures the time it takes for the test liquid to flow through a capillary of a known diameter of a certain factor between two marked points. By multiplying the time taken for the fluid to flow times the factor of the viscometer, one obtains the kinematic viscosity. Viscometers are usually placed in a constant-temperature water bath for this measurement, and kinematic viscosity is measured at temperatures from 15 to 100°C (≈60–210°F). The test method is described in more detail in ASTM D445-04e1 (2004). In this method, repeatability and reproducibility are 0.35 and 0.7%, respectively (Denis and Briant (1997). A large variety of “complex fluids” are encountered in oilfield treatments, especially in reactive stimulation, hydraulic fracturing, and EOR. These are fluids that behave neither like a liquid nor like a solid under flow, but show a mixed behavior. The study of these types of fluids is “rheology.” The simplest definition of rheology is the study of the flow of matter. For simple fluids such as water or many organic liquids, including some very viscous fluids, the relationship of the shear stress ( s ) applied to the resultant shear strain rate ( g ) is linear and the ratio of these quantities is the liquid viscosity ( u or h ) as defined

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 18

03/11/17 6:50 PM

Introduction to the Technology of Flow and Integrity Management  19

U-tube (Ostwald) Viscometer

Rotational (Couette) Viscometer

Fig. 1.11—Viscometer types (Frenier et al. 2010).

Shearing Stress, t

in Eq. 1.17. These fluids are termed “Newtonian fluids.” Shear thinning Fig. 1.12 shows types of Bingham plastic stress/strain responses for Newtonian several different types of fluids that include Newtonian and non-Newtonian materials. Note that a Bingham plastic has a yield value and does not respond until that point; however, it then acts as a µ Newtonian fluid. For many types of complex fluids such 1 as emulsions, suspensions, slurries, gels, foams, polymer solutions, and other complex Shear thickening mixtures of substances, the shear-stress/shear-rate relationship cannot be characterized by a single value of viscosity (u) (i.e., at a fixed Rate of Shearing Strain, dµ/dy temperature). Instead, the viscosity is a function of the Fig. 1.12—Types of viscous responses. operating conditions, such as the shear rate. One task of rheology is to establish the relationships between viscosity and stresses by adequate measurements and modeling. Two examples of the non-Newtonian behavior are shear thinning and shear thickening fluids. Thixotropic fluids are even more complex because a scan of stress vs. strain does not have the same values when the strain is reversed. Latex paints provide an example of a thixotropic fluid.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 19

03/11/17 6:50 PM

20  Chemical and Mechanical Methods for Pipeline Integrity

For polymer melts (and wax/oil mixtures) and suspensions, generally, the viscosity decreases as the shear rate increases. This type of behavior, shear thinning, is of considerable industrial significance. –1 10 For example, fracturing fluids are Newtonian shear thinning. Some fluids exhibit region different types of behavior depending on the shear field. CarbohyPower law region drate-based fracturing fluids [e.g., 10–2 hydroxypropyl guar (HPG)] exhibit 10–2 10–1 100 101 102 103 104 such complex behavior when in Shear rate (s–1) water. Fig. 1.13 shows plots of 0.48% HPG in water as a function Fig. 1.13—Hydroxypropyl guar (HPG) rheology as a function of of shear rate and temperature (Guiltemperature (Constein et al. 2001). lot and Dunand 1985). Two models will be required to fit this type of data. Eq. 1.17 is useful in the linear (Newtonian) low-shear region, but a power-law model gives an apparent viscosity (ma): Viscosity (Pa·s)

100

60°F 80°F 100°F 125°F 150°F

µa = K / γ (1− n ) .������������������������������������������������������������������������������������������������������������������������ (1.18) Eq. 1.18 is required for the high-shear region. Here, K is the consistency index in lbf-sn/ft2 or kPa·sn, and n is the flow behavior index (dimensionless). These relationships hold for most fracturing fluids over the range of shear rates in which the fluid displays non-Newtonian behavior. A loglog plot of s vs. g usually yields a straight line over a portion of the shear range. The slope of the straight-line portion is equal to the behavior index n, and the value of t at g = 1.0 s–1 is equal to the consistency index K. A log-log plot of ma vs. g has a straight-line slope of n – 1 when the power-law model is applicable. The slope is zero for Newtonian behavior (Constein et al. 2001). To better predict the full range of fracturing fluid viscosity, a rheology model must use not only n and K but also a zero-shear viscosity term. The Ellis model (Matsuhisa and Bird 1965) added zero-shear viscosity at γo to the power-law model to improve viscosity prediction: 1 1 1 = + .�������������������������������������������������������������������������������������������������������������������� (1.19) µ a µ o Kγ n −1 Here, n and K are defined from the high-shear data. Shear thickening is a less frequently observed phenomenon whereby the material exhibits an increasing viscosity with increasing shear rate. A property of some complex fluids (e.g., some types of clay in water and latex paint, as noted earlier), thixotropy refers to time-dependent viscosity. A thixotropic fluid displays a decreasing viscosity with time at a constant shear rate. Materials that exhibit the opposite behavior— that is, increasing viscosity with time at a constant shear rate—are rheopectic. A complex fluid may also exhibit a yield stress. Below a certain value of applied stress, the yield stress, the fluid does not flow; but above this stress, the fluid flows. It may be noted that waxy crude oils, for example, are usually both shear thinning and thixotropic, and they may exhibit a yield stress at lower temperatures as well. A rheometer is an instrument that is used to measure the rheology of fluids. In addition to applying a constant shear stress or a constant shear rate, rheometers can also apply oscillatory motion. This is called “dynamic oscillatory rheometry.” The response to the oscillatory motion can be used to determine the solid-like behavior and the liquid-like behavior of the fluid. Further, the stress in the direction

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 20

03/11/17 6:50 PM

Introduction to the Technology of Flow and Integrity Management  21

normal to the flow can Elastic solid also be measured. Thus, Strain the stress tensor can be Stress related to the strain. The δ=0 Measured stress response measured response can Viscous fluid be factored into two comStrain ponents: the in-phase and Stress the out-of-phase. The inphase response defines Viscoelastic fluid δ = π/2 the storage modulus, G′, Strain which gives informaApplied strain deformation Stress tion on the elasticity; and the 90° out-of-phase 0 < δ < π/2 response, G″, is the loss modulus that gives inforFig. 1.14—Dynamic oscillation rheometry. mation about the viscous properties of the fluid. To perform the experiment, a sinusoidal stress is applied at a frequency, v, and the strain (in phase and out of phase) is plotted vs. the frequency (see Fig. 1.14). Rheometers can be programmed to run temperature sweeps, shear sweeps, and creep tests to determine various rheological properties of fluids. A study of this behavior can provide mechanistic information about the arrangement and rate of change of the polymer molecules. Another type of viscometer is a rotary viscometer, which is used for a wide range of shear rates, especially for low shear rate and viscous fluids such as lubricants and heavy-petroleum fractions and fracturing fluids. In these viscometers, fluid is placed between two surfaces: One is fixed, and the other is rotating. Rotational viscometers use the idea that the torque required to turn an object in a fluid is a function of the viscosity of that fluid. They measure the torque required to rotate a disc or bob in a fluid at a known speed. “Cup” and “bob” viscometers work by defining the exact volume of a sample that is to be sheared within a test cell; the torque required to achieve a certain rotational speed is measured and plotted. There are two classical geometries in cup and bob viscometers, known as either the “Couette” or the “Searle” system, distinguished by whether the cup or the bob rotates. The rotating cup is preferred in some cases because it reduces the onset of Taylor vortices (i.e., axisymmetric toroidal vortices), but it is more difficult to measure accurately. See Fig. 1.11 for diagrams of the two types of viscometers. Also see Wikipedia (2009) for examples of other types of viscometers. The above methods are used to measure viscosity of liquids at atmospheric pressure. To measure the viscosity of the live fluid, the viscometers must operate under pressure. Typically, three types of viscometers are used for these measurements: a rolling ball viscometer, a capillary flow viscometer, and an electromagnetic viscometer. The reservoir fluid sample is transferred to the viscometer at an elevated pressure to ensure monophasic transfer. In the rolling ball viscometer, a ball is allowed to fall through the fluid, and the time required for the ball to travel through the fluid is correlated to the fluid viscosity. In the capillary flow viscometer, the live fluid is allowed to flow under pressure through a long capillary tube. The pressure drop across the tube is measured, and the viscosity of the fluid is calculated on the basis of the HagenPoiseuille equation for laminar flow (Eq. 1.12). Another device is an electromagnetic viscometer (CVI 2011). In this method, a piston moves back and forth through the pressurized fluid, and the drag on the piston is measured, thus inferring the viscosity. 1.4.4  Thermodynamics and Kinetics of Pipeline-Fouling Reactions. The concepts of aqueous solution thermodynamics are crucial for understanding many processes in oilfield chemistry. The topic is covered in great depth in textbooks on aqueous chemistry. For example, see Zemaitis et al.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 21

03/11/17 6:50 PM

22  Chemical and Mechanical Methods for Pipeline Integrity

Rate of Reaction

(1986) and Langmuir (1997). Therefore, only a simplified description of important principles is presented here. The reader should consult the above-mentioned texts Forward reaction for a more in-depth discussion. r1 = r2 A system at chemical equilibrium represents a dynamic state in which two or more opposing reactions are taking place Reverse reaction at the same time and at the same rate. Fig. 1.15 shows the approach of a chemical system to equilibrium. At early time, the Time rate of forward reaction (r1)—that is, the formation of products from reactants—is Fig. 1.15—A chemical system’s approach to equilibrium. large. As the system progresses in time, the rate of forward reaction decreases and the rate of reverse reaction (r2) increase. When the rates are equal, the system is in equilibrium and the net rate (i.e., rate of formation of products minus rate of formation of reactants) is equal to zero. Once a closed system reaches chemical equilibrium, the chemical composition of the system is, by implication from that point on, independent of time and previous history. Equilibrium constants are used to relate the amounts of reactants and products at equilibrium. Consider, for example, a simple ideal chemical system consisting of reactants A and B, and products C and D, for which the reaction stoichiometry is given by aA + bB

⇔ cC + dD .�������������������������������������������������������������������������������������������(1.20a)

This would imply that the forward and reverse reactions are aA + bB

→ cC + dD (forward reaction)������������������������������������������������������������ (1.20b)

and cC + dD

→ aA + bB (reverse reaction) .�������������������������������������������������������������� (1.20c)

If the system is ideal, then the equilibrium constant for the system can be expressed in terms of concentrations as K eq =

[C ]c [ D]d ,���������������������������������������������������������������������������������������������������������������������� (1.21) [ A]a [ B]b

where Keq is the equilibrium constant and [ ] denotes concentration of the species. The equilibrium constant can also be expressed in terms of the free-energy change for the reaction (DG°) as  −∆G   K eq = exp  .���������������������������������������������������������������������������������������������������������������� (1.22)  RT  Here, R is the gas constant and T is the temperature. The Keq is a function of temperature and pressure only and does not depend on composition of the system. Therefore, if the composition of the system changes as a result of subsequent reactions or addition of new reactants or products to

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 22

03/11/17 6:50 PM

Introduction to the Technology of Flow and Integrity Management  23

the system, the same value of the equilibrium constant(s) can be used to calculate the equilibrium distribution in the new system provided that the temperature and pressure remain constant and the equilibrium assumption is valid. Equilibrium constants are not constants in the true sense because they do depend on temperature and pressure. An increase in temperature may affect the forward and reverse reactions differently. The reaction that absorbs the most heat will increase that rate to a larger extent than the other reaction. A new equilibrium constant will now represent the new situation. The simple equation by van’t Hoff (J. H. van’t Hoff, 1852–1911) can be used to compute the change in equilibrium constant caused by a change in temperature. The equation can be expressed as d ln K eq dT

=

∆H ° ,���������������������������������������������������������������������������������������������������������������������� (1.23) RT 2

where DH° is the standard enthalpy change of reaction, T is the temperature, and R is the gas constant. If the reaction is exothermic (i.e., if DH° for the reaction is negative), then the equilibrium constant decreases as temperature increases. Conversely, Keq increases with temperature for an endothermic reaction. If the standard enthalpy change of reaction is assumed independent of temperature, then integrating Eq. 1.23 gives an even simpler result: ln

K eq K eq,ref

  = − ∆H °  1 − 1  .�������������������������������������������������������������������������������������������������� (1.24) R  T Tref 

Here, Keq,ref is the value of the equilibrium constant at the reference temperature Teq,ref. This approximate equation implies that a plot of ln Keq vs. the reciprocal temperature gives a straight line. This equation is helpful in interpolating and extrapolating equilibrium constant data with reasonable accuracy. The effect of pressure on equilibrium constants is typically smaller than the effect of temperature. However, for deep wells it can be significant and needs to be considered along with the change in temperature. The pressure dependence of the equilibrium constant can be calculated from  ∂ln K eq  ∆V °  ∂ P  = RT , ���������������������������������������������������������������������������������������������������������������� (1.25) T where ∆V° is the molar volume change of the reaction with all reactants and products in their standard states (Langmuir 1997). If the molar volume change of the reaction is independent of pressure, then the integration of Eq. 1.25 yields ln

K eq K eq,ref

=−

∆V °( P − Pref ) ,������������������������������������������������������������������������������������������������������ (1.26) RT

where Keq is the equilibrium constant at the desired pressure P. Keq,ref is the equilibrium constant at Pref, which is typically 1 bar. In very dilute aqueous solutions, the anions and cations behave in an “ideal” manner, in which each ion will act as if it is independent of all other ions in the solution. In real solutions, especially those in which there are high concentrations of other ions (such as in a produced brine), the ions are affected by the other ions in solution, and so the fluid may not behave as if it has exactly the same number of ions as described by the concentration. The equilibrium constants in such nonideal systems are then expressed in terms of species activity. If, for example, the chemical system

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 23

03/11/17 6:50 PM

24  Chemical and Mechanical Methods for Pipeline Integrity

considered previously of components A, B, C, and D is nonideal, then the equilibrium constant is given by K eq =

 −∆G   {C }c { D}d = exp  , ������������������������������������������������������������������������������������������ (1.27) a b { A} { B}  RT 

where { } denotes the activity of the species. The activity of the species can be thought of as an effective concentration of the species in solution. Programs to calculate or estimate the activity coefficients are beyond the scope of this discussion, but there are several general purpose geochemical models available in the public domain that can be used to predict formation of oilfield scale. Most of them are available at no or minimal charge through the Internet and have been extensively reviewed in texts on aqueous chemistry. They include Mangold and Tsang (1991); Glynn et al. (1992); Wolery (1992); van der Heijde and Elnawawy (1993); Langmuir (1997); Butler and Cogley (1998). These concepts can be applied directly to inorganic scales. Organic deposits such as wax (paraffin), asphaltenes, gas hydrates, and naphthenates also become supersaturated before they can start to deposit. However, because of system complexity, equilibrium constants cannot usually be calculated. See Section 4.2.1 for discussions of the rates of scale dissolution or deposition of solids and how they are influenced by the thermodynamic properties described in this section. 1.4.5  Surface Chemistry. The chemistry and physics of surfaces affect a great number of processes and services performed in oil/gas transportation and apply to solids as well as liquids. The surface of any solid or liquid is an interface between that medium and some other that could be a solid, liquid, or a gas. The chemistry of the interface is affected by both surfaces. In one example (Fig. 1.16), a drop of oil is placed on a solid surface in a jar of water. The forces at that surface—that is, interfacial surface tension (IST, or g )—are caused by the cohesion of the molecules in that surface and depend on the molecules in both surfaces. Therefore, IST is not a property of the liquid alone but of the liquid’s interface with another material. In Fig. 1.16, the surface (on the left) is covered with a water film (it is hydrophilic), and the oil drop is repelled and the contact angle (q ) is close to 0°. In the middle case, the surface is partially oil-wetting, and in the right figure the surface is completely oil-wet and the oil droplet spreads out on the surface. If a liquid is in a container, such as a pipe or tank, then there will be a liquid/air interface at its top surface and also an interface between the liquid and the container walls. The IST between the liquid and the air is usually different from its IST with the container walls. Where two surfaces meet, the consequence must also be such that all forces are in balance. As noted, where the two surfaces meet (and there are three phases), they form a contact angle, q, which is the angle that the tangent line of the liquid surface makes with the solid surface. Fig. 1.17 shows an example of solid, liquid, and gas interfaces. Here, tension forces ( f ) are shown for the liquid/air interface, liquid/solid interface, and solid/air interface.

θ

θ ~ 0°

γow γso

γso = γsw + γow Cos θ

θ

γsw

θ ~ 180°

Fig. 1.16—Wetting angles of water, oil, and mineral surface (Abdallah et al. 2007).

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 24

03/11/17 6:50 PM

Introduction to the Technology of Flow and Integrity Management  25

The vertical and horizontal forces must both cancel exactly at the contact point. The horizontal component of fla is canceled by the adhesive force, fA.

fsa

θ

f1a

fA f1s

fA = fla sin θ .�������������������������� (1.28) The other balance of forces is in the vertical direction. The vertical component of fla must exactly cancel the force, fls. fls = fsa + fla cos θ .���������������� (1.29) Because the forces are in direct proportion to their respective surface tensions, we also have

γ ls = γ sa + γ la cos θ , ���������������� (1.30)

Fig. 1.17—Forces at surfaces (Wikipedia 2010d). Liquid

Solid

Contact Angle

Water Ethanol Diethyl ether Carbon tetrachloride Glycerol

Soda-lime glass Lead glass Fused quartz

180°

Paraffin wax

73°

Silver

90°

Acetic acid Water

where gls is the liquid/solid IST, gsa is the solid/air IST, and gla is the liquid/air IST. Table 1.2—Contact angle values for solid/liquid converted to Eq. 1.30 is known as Young’s equation. oilfield convention (Sears and Zemanski 1955). Table 1.2 lists contact angle data for common fluids, and Table 1.3 shows IST values for a number of liquids with air. The values of the contact angles and the IST control many of the properties of processes described in subsequent chapters. Note in Table 1.3 that the IST of pure water is much higher than that of solutions of ethanol and water. Many surface-active agents (i.e., surfactants) also lower the IST of water and thus affect the contact of water with surfaces. The convention used in the oil/gas industry is that a near-180° contact angle indicates that the liquid will spread on (“wet”) that surface. The reader should note which convention is being used when comparing data, given that some conventions call wetting to have a 0° contact angle. The contact angle and IST can be measured in several ways. The sessile-drop method is illustrated in Fig. 1.18a. Here, a drop of liquid is seen after it has been placed on a surface in air, and the three IST vectors are seen. The contact angle (q) can be measured visually by taking a photograph of the surface, usually using a microscopic device; see Howard et al. (2010) for a specific example. Figs. 1.18a and 1.18c illustrate the importance of the solid surface itself. In these examples, a water drop has been placed on two surfaces in an oil medium. When bronze is the solid (Fig. 1.18b), the water does not displace the oil, whereas on glass (Fig. 1.18c), the water does displace the oil and wets the surface. Fig. 1.18 also illustrates a possible caution when interpreting literature information, because either the acute or the obtuse angle could be reported. If water is used to test the “wetting” of a surface, low numbers refer to a hydrophilic surface and high values refer to a hydrophobic surface. Also see Tadmor (2004). The IST of a liquid vs. the IST of air can be measured using several methods, among them the Du Nouy ring, the pendant-drop test, and the Wilhemly plate. These are explained next. Du Nouy Ring. A manual method apparatus, the Du Nouy ring is depicted in Fig. 1.19. This is the traditional method used to measure surface or interfacial tension. Wetting properties of the surface or interface have little influence on this measuring technique. Maximum pull exerted on the ring

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 25

03/11/17 6:50 PM

26  Chemical and Mechanical Methods for Pipeline Integrity

Temperature (°C)

IST, g (dyne/cm)

Acetic acid

20

27.6

Acetic acid (40.1%) + water

30

40.68

Acetic acid (10.0%) + water

30

54.56

Acetone

20

23.7

Diethyl ether

20

17.0

Ethanol

20

22.27

Ethanol (40%) + water

25

29.63

Ethanol (11.1%) + water

25

46.03

Glycerol

20

63

n-Hexane

20

18.4

Hydrochloric acid 17.7M aqueous solution

20

65.95

2-proponal

20

21.7

Methanol

20

22.6

n-Octane

20

21.8

Sodium chloride 6.0M aqueous solution

20

82.55

Sucrose (55%) + water

20

76.45

Water

 0

75.64

Water

25

71.97

Water

50

97.91

Liquid

Table 1.3—IST of various liquids measured against air (Dean 1961).

γla

γsl

θ γsa

(a) Illustration of sessile drop method with liquid droplet partially wetting a solid substrate

(b) Water droplet immersed in oil resting on a brass surface

(c) Water droplet immersed in oil resting on a glass surface

Fig. 1.18—Sessile-drop method.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 26

03/11/17 6:50 PM

Introduction to the Technology of Flow and Integrity Management  27

Fig. 1.19—Du Nouy ring (Wikimedia 2016).

by the surface is measured (PHYWE 2004). The most important experimental issues are using very clean equipment and fire cleaning the platinum ring used in the test. Pendant-Drop Test. Used with oils, the pendant-drop test depends on surface tension to suspend a drop of liquid from the end of a tube, as shown in Fig. 1.20. The force from the surface tension is proportional to the length of the liquid/tube boundary, with the proportionality constant usually denoted as Given that the length of this boundary is the circumference of the tube, the force attributable to surface tension is given by

d

Fg

α

m

Fγ = π dγ .������������������������������������������������������������������������ (1.31) Note that Here d is the tube diameter. The mass m of the drop hanging from the end of the tube can be found by equating the force caused by gravity,

Fig. 1.20—Surface tension ensured by pendant-drop method.

Fg = mg.���������������������������������������������������������������������������������������������������������������������������������� (1.32) The component of the surface tension in the vertical direction is Fγ = sin α . Thus, mg = π dγ sin α . ���������������������������������������������������������������������������������������������������������������������� (1.33) Here, a is the angle of contact with the tube, and g is the acceleration caused by gravity. The limit of this formula, as a goes to 90°, gives the maximum weight of a pendant drop for a liquid with a given surface tension, g. Thus,

γ=

mg .������������������������������������������������������������������������������������������������������������������������������������ (1.34) πd

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 27

03/11/17 6:50 PM

28  Chemical and Mechanical Methods for Pipeline Integrity

Wilhemly Plate Method. Another method especially suited to check surface tension over long time intervals is the Wilhemly plate method, described Fig. 1.21. A vertical plate of known perimeter is attached to a balance, and the force resulting from wetting is measured. This method uses the interaction of a test plate with the liquid being tested. Biolinscientific (2009) described one instrument that uses this method. The calculations for l air this technique are based on the geometry of a fully wetted plate in contact with, but not submerged in, the liquid liquid phase. In this method, the position of the probe relative to the surface is significant. As the surface is brought θ into contact with the probe, the instrument will notice this event through the change in forces it experiences. It will register the height at which this occurs as the “zero Fig. 1.21—Wilhemly plate. depth of immersion.” The plate will then be wetted to a set depth to ensure that there is indeed complete wetting of the plate (i.e., zero contact angle). When the plate is later returned to the zero depth of immersion, the force it registers can be used to calculate surface tension (see Eq. 1.35). F

γ la =

F ; l = 2w + 2d . ������������������������������������������������������������������������������������������������������ (1.35) 2lcosθ

A number of important properties and uses of pipeline chemicals are related to wetting and capillary forces during pumping in the formation. Howard et al. (2010) explained that the Laplace and Washburn equations can both be of importance (Grattoni et al. 1995). The Laplace equation relates capillary pressure—the difference between the phase pressures of a nonwetting (P2) and a wetting phase (P1) such as water and oil, Pc—to surface tension, g, and contact angle, q. Here, r is the radius of a capillary tube, L is the height of the fluid rise in the tube, r is the density of the fluid, and g is the gravitational acceleration (980 cm/sec2). ( P2 − P1 ) = Pc = L ρ g = 2γ cos θ / r �������������������������������������������������������������������������������������������� (1.36) To evaluate wetting of two fluids in the same geometry, the ratio of two Laplace equations can be used. For example, using water and a solution with unknown properties leads to Lw ρw g = 2γ w cos θ w / r ������������������������������������������������������������������������������������������������������������ (1.37) and Lu ρ u g = 2γ u cos θ u / r . ������������������������������������������������������������������������������������������������������������ (1.38) Dividing one by the other and removing constants gives Lu γ u cos θ u = . ���������������������������������������������������������������������������������������������������������������������� (1.39) Lw γ w cos θ w 1.4.6  Testing of Pipeline Fluids. Testing of the well/pipeline fluids provides the base data to predict conditions for FA and IM planning. These short subsections give an outline. Details are in Frenier and Ziauddin (2008) and Frenier et al. (2010).

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 28

03/11/17 6:50 PM

Introduction to the Technology of Flow and Integrity Management  29

Compositional Characterization of Crude Oil/Gas. Important ways for characterizing crude oil include • • • • •

Boiling points from distillation or gas liquid chromatography Density distribution from the assay cuts SARA (saturates, aromatics, resins, and asphaltenes) PNA (paraffin/naphthene/aromatic) Water content

Physical Characterization. Common characterization processes include • • • • •

Pressure/volume/temperature relationships Density Viscosity Rheology Refractive index

Composition of the Aqueous Phases. The composition of the aqueous phases may greatly affect scale potential, corrosivity, emulsion tendencies, and hydrate formation. Analysis of water chemistry can give important information on the scaling tendencies of the formation water or of water that may become mixed with the formation water. There are a number of standard methods for analyses of water, including those found in the volume from the American Public Health Association (Clesceri et al. 1999). Most of the cations that are important for determination of scaling tendencies can be determined using atomic absorption spectrophotometry or inductively coupled plasma optical emission spectrophotometry. There are also portable kits (Hach 2005) for analysis of water components that require addition of specific reagents and examine the development of a color change that can be read using a simple optical spectrophotometer. Analysis of surface water samples will give some indication of the presence of scaling ions, but unless a sample can be collected at the formation face and maintained at the bottomhole temperature and pressure, important scaling ions will not be present. The concentrations of acid gases and the resultant pH values are determined by methods described in Davies and Scott (2006). A major complication with this analysis is that the water content of a fluid and the chemical composition change in time and place in the pipeline train as a result of physical changes and introduction or removal of components in the separation and treating stages. Stimulation and EOR can likewise change the chemical composition of the fluids entering the system at any point. Additional testing methods, including microbiological methods, which are used especially for field samples, are discussed in Section 3.4.3. 1.5  Summary and Lessons Learned This chapter has provided an introduction to the physical and chemical technologies required to understand processes for maintaining and protecting the vital pipeline systems. • The pipeline systems of the oil/gas industry extend from the wellhead to the final consumer of the products, and the same types of chemical/physical processes will apply throughout the system. There are various physical and chemical reactions that are common to all systems, and these were reviewed in this chapter and referenced for further study. The most universal principles are the varied reactions at surfaces (liquid/liquid, liquid/gas, liquid/solid, and gas/ solid) and the realities of multiphase flow conditions. • Multiphase flow of some form is the norm for all line segments, though degree and conditions may change dramatically in different portions of the system. The different flowing forms may greatly affect corrosion, scale formation, and other FA issues.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 29

03/11/17 6:50 PM

30  Chemical and Mechanical Methods for Pipeline Integrity

• Pipe materials of construction are critical issues for maintaining integrity and long life and must be addressed when planning a line segment and when maintenance is needed. • Although an almost infinite variety of chemicals may transit a system, the division into aqueous liquids, hydrocarbon liquids, gases, and solids will help make sense of the mixtures. • As much chemical information about each chemical class and phase as is possible is necessary for making predictions about FA and IM issues.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 30

03/11/17 6:50 PM

Chapter 2

From the Well to the Consumer Hydrocarbon-producing wells deliver a vast number of chemicals and chemical mixtures into the flowlines, into subsurface and surface treating facilities, and then into the pipelines that deliver a product to a customer. This chapter describes the continuum of maintenance and integrity concerns that affect the connected pipeline and surface facilities and thus the steady flow of products from the wellhead to the final user of hydrocarbon chemicals. The consumer facilities may use a product, such as natural gas, to directly generate power and heat, or the hydrocarbons may be subjected to processing, such as at a refinery or chemical plant. In this book, the chemicals produced from the wells that then transit the pipeline systems are grouped by phase as • • • •

Aqueous liquids Hydrocarbon liquids Gases Solids

This chapter includes short descriptions of the fluids and some solid phases. In addition, here are described the influence of the reservoir history and type and the effects various possible fluid phases have on the different categories of pipeline problems found in the upstream and midstream petroleum environments. 2.1  Description of Well Production Fluids Entering the Pipeline System Sections 2.1.1 through 2.1.4 provide details of the composition of the fluids and importance to flow assurance (FA) and integrity management (IM) of the different fluid phases. 2.1.1  Aqueous Phases. Most of the hydrocarbon streams coproduce vast quantities of salt solutions. The coproduced water includes connate water associated with the native hydrocarbons, as well as improved oil/gas recovery (IOR) water (including water for pressure maintenance). Additional water sources include those produced by stimulation treatments. In addition, condensed water from different operations, such as facilities, forms as a consequence of changes in temperature and pressure. Most important, all the aqueous phases may differ in composition as a result of times and locations. Included in the stimulation category are aqueous streams from fracturing and acidizing treatments (see Frenier and Ziauddin 2014), as well as any other aqueous fluids used during separation activities. The chemical compositions of these water streams vary significantly from almost potable to very concentrated brines. Fig. 2.1 (USGS 2002) shows a diagram of salinities of produced water from various parts of the US. Enhanced oil recovery (EOR) and stimulation, especially hydraulic fracturing (HF), may significantly change the water quantity and quality that is produced at the end of all the well treatment operations. EOR and pressure maintenance using injected water are major

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 31

03/11/17 6:50 PM

32  Chemical and Mechanical Methods for Pipeline Integrity

Fig. 2.1—Total dissolved solids in produced water, US land (USGS 2002).

reasons that the water cut usually increases as the well field matures. The aqueous phases present in a specific section of a pipeline system may also change as a function of the well’s life cycle (POSC 2006; see also Section 2.2.1) and various production activities. Dissolved alkaline and alkaline earth salts and various acids constitute the majority of the “inorganic”- type materials present. However, almost any soluble element may be present in small quantities, and HF may introduce additional chemical species (see Shen et al. 2012). The amount and composition of the aqueous phase may affect or cause corrosion, inorganic scale, gas hydrates, and emulsions as well as affecting the use of the final hydrocarbon product. Reports by Ruegamer et al. (2013) and Wilson (2014) claim that the amount of water used in HF treatments is changing as a result of new knowledge and the changing nature of the wells. Jacobs (2016) has reviewed the problems of production and injection of water from the Mississippian Lime formation in western Oklahoma, USA. The play is a carbonate formation, and unconventional techniques, especially directional drilling, are used to recover oil and gas from it. Last year, production in the Mississippian Lime was estimated to be approximately 100,000 B/D of oil, which accounted for a quarter of the state’s overall oil production. The problem is that the water cuts were not just higher than normal, they were often extraordinarily high. Newly completed oil wells have been known to pump out as much as 98% water at a rate of thousands of barrels per day. According to Jacobs (2016), the injection of so much water into a porous formation (the Arbuckle) has probably contributed to induced seismic activity in Oklahoma. Therefore, the continued production from this important resource is in doubt. Definitions and examples of stimulation treatments are presented in Frenier and Ziauddin (2014). Any of these treatments and the changes in them may affect water quality and volume.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 32

03/11/17 6:51 PM

From the Well to the Consumer  33

The failure of a pipeline carrying waste water can be as significant as the failure of a line carrying oil or gas in some situations. The Associated Press (AP 2014) has reported that a leaking underground pipeline near Mandaree, Montana, USA, spilled approximately 1 million gal of saltwater (from fracturing activities in the Bakken plays) near Bear Den Bay, a tributary of the Missouri River, which provides potable water to some communities. SPE recognizes that the availability and production of water is one of the most important issues facing the hydrocarbon production and transport industries (SPE 2011). Water vapor will also be entrained with natural gas as well as the liquid hydrocarbons, and its removal, termed “dehydration,” is described in Section 2.3.2. 2.1.2  Hydrocarbon Liquids. Hydrocarbon liquids well as the gaseous chemicals described in this section are usually the primary goal of petroleum well production activities. The in-situ crude oil being produced constitutes a continuum of soluble chemicals, from C1 to large molecules with molecular weight in the 750-dalton range. Fig. 2.2 shows some basic chemical structures that constitute crude oil. The chemicals range from nonpolar chemicals (i.e., saturated hydrocarbons) to highly polar ones (i.e., aromatic asphaltenes and nitrogen-substituted chemicals). The lower hydrocarbons are considered in the gases Paraffins described in the next section. However, it should be understood that at undisturbed n-octane reservoir conditions there exists a single hydrocarbon phase, and the gaseous hydrocarbon phases do not emerge until the bubn-pentadecane blepoint (Bp) pressure has been reached. Napthenes When and where the phase changes take place may greatly affect the viscosity and stability of the flowing crude oil and the production of wax, asphaltenes, and gas hydrates (these subjects are reviewed in isooctylperhydrophenanthren Chapter 4). Depending on the source of the oil, it may consist of heavy fraction as Aromatic well as lightweight fractions. Some shale N formations produce a very light (and flammable) product—termed “natural gas liquid” (NGL) (see Appendix A)—that may xylene cause flow and gathering line problems quinoline (these are described in Section 2.3.1). Resins and Asphaltenes 2.1.3 Gaseous Phases. Included in the category of gaseous phases are hydrocarbons from C1 to approximately C5 as well as CO2, H2S, N2 (Alvarado et al. 1998), water vapor, low-molecular-weight organic acids, and possibly oxygen (O2). All these gases (except O2) may be dissolved in the hydrocarbon or aqueous phases at reservoir equilibrium and form a separate gaseous phase at different times and places in the production and pipeline train. Both chemical reactions and

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 33

HO An asphaltene

6-dodecylnaphthalene Fig. 2.2—Chemical component families in crude oil.

03/11/17 6:51 PM

34  Chemical and Mechanical Methods for Pipeline Integrity

mechanical processes (temperature, pressure, flow rates) affect the presences and quantities of the gaseous phases. The dissolved gases also partially control the aqueous fluid pH and may thus determine the corrosivity of the fluids. Organic acids such as acetic acid can become volatile under some conditions and may be one cause of top-of-line corrosion (TLC). O2 is always considered to be an introduced contaminant, given that the undisturbed reservoir conditions are usually reducing and any oxygen would quickly be reacted. Production of natural gas (mostly methane, with various amounts of C2–C5 hydrocarbons) is the goal of many oilfield operations, and this mixture will usually be separated or removed from the production stream before conveyance to a customer. In some wells, the production of CO2 or rare gases such as helium (He) may also be a primary or secondary goal (see Daly 2005). CO2 has become an important commodity because this gas is a significant EOR chemical. 2.1.4 Solids. Solids are complex materials usually considered to be unwanted contaminants. Sand and other formation fines can result from uncontrolled production methods or migration of HF proppants. Corrosion products, organic solids (especially paraffins, asphaltenes, gas hydrates, and naphthenates), and scale particles also could enter a flow system. The impingement of flowing solids onto a surface can cause mechanical and corrosion/mechanical damage to many well system components (see Section 3.2.3). A sufficient amount of solids can change the flow regime or actually block the pipe. 2.1.5  Emulsions, Foams, and Solid Dispersions. A wide range of mixed phases can form in pipelines as well as in facilities. Emulsions are a dispersion of one or more liquids in another liquid (Kokal 2006). Foams are a dispersion of a gas in a liquid. Solids also may be dispersed in liquids and gases. Important characteristics causing all dispersions in the oil/gas industry include these: • • • • •

More than one phase is present. Shear forces are needed to mix the phases. Surface-active molecules are present to stabilize dispersions. Mixtures are thermodynamically unstable. Mixtures have a finite lifetime before individual phases separate.

Because of the heterogeneous nature of pipeline-associated fluids, all these factors may present at any time. More details of problems and applications of mixed phases are provided in Sections 4.6 and 7.5.4 and in Frenier and Ziauddin (2014). 2.2  Effects of the Life Cycle and Reservoir Type on Pipeline Maintenance The requirements to combat FA and IM problems in pipelines as well as the solutions to them are largely dictated by the reservoir fluids that are produced, as described briefly in Section 2.1. Conventional and unconventional plays may produce different fluids and fluid volumes that can affect scale types and locations, corrosion types and locations, and formation of organic solids. Sections 2.2.2 and 2.2.3 provide short descriptions of several conventional and unconventional reservoirs and how production and completion methods may affect pipeline problems and solutions. The actual life cycle of the well, as well as the various stages in the production phase, can also greatly affect the need for corrosion, scale, and organic solids control in pipelines. 2.2.1  Life Cycle of a Hydrocarbon-Producing Reservoir. The phases of the life cycle of a hydrocarbon-producing reservoir have been identified in the industry as exploration (discover), appraisal (define), development (develop), production (deplete), and abandonment (dispose) (POSC 2006). Except for the earliest phases of exploration, in which geologic and seismic methods are used to find

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 34

03/11/17 6:51 PM

From the Well to the Consumer  35

promising areas where hydrocarbons may be located, large volumes of chemicals are used to aid production. They are applied during the drilling, completion, production, and abandonment phases. However, even when additional chemicals are not used, chemistry is important in all the phases (see Fig. 2.3). Consider, for example, the following applications. • Exploration/appraisal: Elucidate formation geochemistry and chemistry of the fluids. • Development: Apply drilling and cementing chemicals to form and complete the well. • Production: Apply chemicals for stimulation, flow assurance, EOR in the mature field, and characterization (i.e., tracers). • Abandonment: Apply cementing chemicals to seal the well and perform monitoring. Exploration/Appraisal. In the exploration/appraisal phase, geochemical analyses are performed on the basis of seismic and well probe data and analyses of outcrops or cores. McCarthy et al. (2011) described the tests that geochemists have used to determine the hydrocarbon-producing potential of a formation from the rock samples collected. These include total organic carbon analysis to find the maximum amount of carbon in the rock, as well as a pyrolosis process in which the rock is heated to increasing temperatures and the effluents are analyzed by several methods described in the papers. Fluids may be captured in test wells for evaluation, and chemical probes may be placed using wireline or coiled tubing. Short summaries of methods used to analyze the liquid samples are described in Frenier et al. (2010, Chap.3). Development. In the development phase, various water-based and oil-based fluids are used in drilling of most wells. Complex oilfield cements are then used to stabilize the production tubing and to isolate various zones from communication with the surface and from nonproducing formations. In addition, completion fluids may be used to maintain and control the well’s pressure balance. Residual effects of drilling and completion chemicals may necessitate the use of acids or HF to remove damage (see Frenier and Ziauddin 2014). Production. Various chemicals are applied during all the phases of production to maintain, control, and, frequently, to enhance the flow of the oil, gas, and aqueous phases. Chemistry in the Well Life Phases Exploration/ Appraisal

Development

Production

Abandonment

Geochemical analyses

Drilling and fluids cementing services

Stimulation flow assurance in mature field: EOR characterization: tracers

Cementing Monitoring

Fig. 2.3—Chemistry in the phases of a well’s life.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 35

03/11/17 6:51 PM

36  Chemical and Mechanical Methods for Pipeline Integrity

Various types of pipelines are used during this phase, and the chemicals affect them in ways that constitute major themes of this book. Abandonment. When the well is abandoned, cements and other chemicals are employed to make sure that the hydrocarbons or other fluids will not reach the surface or pollute aquifers or damage property. As well, pipelines and gathering lines directly associated with production must be decommissioned and safely removed. Production Subphases. The production activity involves subphases that may affect pipeline activities. See the review by Lindley (2001) on the production phases; an abstract follows. Primary Phase. In this phase, the reservoir fluids flow mostly as a result of the initial and internal pressure of the reservoir. Note that the production is controlled by the pressure differential between the formation and the bottomhole well pressure. While the pressures may be sufficient for initial production, stimulation using fracturing and/or reactive chemical treatments may be applied to some wells to remove formation damage or to improve returns from tight formations (Frenier and Ziauddin 2014). Other production chemicals can also be used to maintain flow, including inhibitors and surfactants (these are described in Chapter 6; see also Frenier et al. 2010). For the most part, during the primary phase, chemicals, including any injected water or gas, are not added to the reservoir except near the wellbore, so the reservoir is not changed significantly from a chemical standpoint. However, just by flowing the wells, important equilibrium conditions may be changed. At some point (either early or very late in the production phase) pressure maintenance will be required. This may be defined as the secondary phase. Some types of pipelines described in Section 2.3 are in use during this phase. Secondary Phase. The pressure to move the fluids through the formations (see Eqs. 1.6 and 1.7) can be maintained or enhanced by adding a downhole pump to reduce the flowing pressure or by injecting fluids into the formation. This latter action presents a radical change to the reservoir, given that a large number of injection wells may be required. This will require a large increase in the number of pipelines and other piping to serve the expanded production area. A typical arrangement is a “five-spot” pattern, in which four input or injection wells are located at the corners of a square—the exact shape of the flood and number of wells depends on the reservoir dimensions (Singh and Kiel 1982; Chang 2010)—and the production well is placed in the center of the square. The injection fluid, which is normally water (brine), steam, or gas, is pumped or injected simultaneously through the four injection wells to displace the oil toward the central production well (Schlumberger 2010). Otott (2007) presented one description of this layout, and Fig. 2.4 shows a number of different plans developed on the basis of the reservoir characteristics (Singh and Kiel 1982). The injected fluids act to drive the hydrocarbons to the production wells as well as to maintain the flowing pressure. However, these activities may introduce multiple problems, including an increase in water cut, a change in water saturation and possibly a change in the wettability of the formation, and introduction of scale-causing ions such as new cations and sulfate. In addition, the water should be treated to remove dissolved oxygen and a biocide should be added. Not adequately treating the water is a major cause of corrosion and the “souring” of the reservoir through introduction of sulfate-reducing bacteria. Increasing the water cut as well as possibly changing ions in the water can change the corrosion rates and locations of damage, as well as scaling (more details are in Chapter 3 and Section 4.2). Because there may now be four injection wells for each production well, the number of connecting pipes as well as treating facilities may increase and/or change. Although steam injection may be considered an integral part of an EOR process (or of primary production for very heavy oils), it can cause multiple problems, including the formation of mixed deposits (see Frenier et al. 2010; also see Chapter 4). Many chemical treatments can be used in this phase to maintain production, including inhibitor injections as well as reactive chemical and propped-fracture treatments. These are described in more detail in Chapters 3, 4, and 5 of Frenier and Ziauddin (2014).

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 36

03/11/17 6:51 PM

From the Well to the Consumer  37

Corner well

Side well

Injection well (a) Direct line drive

(d) Nine-spot

Four-spot

Seven-spot

(b) Staggered line drive

Smallest area of flow symmetry

(c) Five-spot, special case of (b) where d/a = 1/2

(e) Seven and four spots

Fig. 2.4—Injection/producer patterns (Singh and Kiel 1982).

At some point in the life of many reservoirs, the removal of additional hydrocarbons is not possible or is economically unproductive because the oil is trapped in the pore spaces and so strongly adsorbed onto the rock surfaces that injection of water, natural gas, or steam cannot remove economic amounts. Because of uneven coverage of the reservoir resulting from permeability differences, oil may also have been bypassed by the sweep fluids. At this point, the massive injection of external chemicals may be planned and the well may be considered to be in the tertiary production phase. Tertiary Phase. Frenier and Ziauddin (2014) define this production phase as the tertiary use of EOR chemicals that may be part of an overall IOR process. As much as 2 × 1012 bbl of conventional oil and as much as 5 × 1012 bbl of heavy oil will remain in the world’s reservoirs after the primary and secondary production phases have reached their economic limits. This incremental production is difficult and expensive but will remain as one of the methods for prolonging production from mature fields. Note that steam injection may be used at earlier phases in some heavy-oil fields (Thomas 2008).Thus, the injection of large amounts of chemicals to remove some of this oil usually defines the tertiary phase. Much of the current interest is driven by the high price of oil. EOR activities, especially CO2 flooding (see Frenier and Ziauddin 2014, Chap. 5), will have a dramatic effect on pipeline FA and IM activities and concerns. The amount of produced water will increase, and the low pH can greatly affect corrosion potential as well as scaling patterns. Changes in inhibitors will probably be required.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 37

03/11/17 6:51 PM

38  Chemical and Mechanical Methods for Pipeline Integrity

Different types of reservoirs may produce different chemicals and may involve changing ratios of hydrocarbons to water, as discussed next in Sections 2.2.2 and 2.2.3. 2.2.2  Conventional Reservoirs of Oil and Gas. According to Frenier and Ziauddin (2014), conventional reservoirs include those drilled, completed, and stimulated using the tools that have been used and developed in the past 100 years. Many of the most important techniques reviewed in the book were developed for use in these types of plays, which exist in carbonate and sandstone formations. The stimulation methods include acidizing as well as fracturing using proppants. By exception, they are not “unconventional” (described in Section 2.2.3). However conventional reservoirs may also include consolidated and unconsolidated formations. Consolidated Reservoirs. Usually a sandstone formation is cemented together to form a mass that has substantial compressive strength (see Fig. 2.5a, which shows the constituents, and Fig. 2.5b, presenting a Berea micrograph). Although carbonate formations—CaCO3 or CaMg(CO3)2—are much more homogeneous than sandstones, they look much like a sandstone, even under a microscope, except that most of the matrix is soluble in hydrochloric acid (HCl). As a consequence, consolidated limestone and sandstone will have high compressive strengths and can be used as primary building materials. These reservoirs can be stimulated by matrix acidizing and acid fracturing (carbonates only), as well as by proppant fracturing methods (using both carbonates and sandstones). Filling a reservoir with a conductive proppant will provide enhanced permeability and may also improve connections within the reservoir. Lack of connections either areally or vertically is a major reason for bypassed hydrocarbons. At some point in the well’s life cycle (POSC 2006), the formation will probably require waterflooding, CO2 flooding, or another

(a) Constituents of Sandstones

(b) Micrograph of Berea Sandstone

Fig. 2.5—Sandstone constituents (a) and micrograph of Berea sandstone (b).

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 38

03/11/17 6:51 PM

From the Well to the Consumer  39

type of IOR if it is an oil-producing formation. These operations will change the hydrocarbon/ water ratio as well as the chemical constituents, which will alter the impact on the connecting pipelines. Unconsolidated Reservoirs. Soft, unconsolidated sand formations exist and frequently have high permeabilities, but still may be improved through stimulation processes including acidizing or hydraulic fracturing. Because they may produce sand, frac-pack designs may apply (Morales et al. 2003) to control the sand production. Sand-producing formations can cause significant damage to piping through erosion/corrosion processes and impingement of the solids (see Section 3.2.3). As the reservoirs mature, changes in oil, gas, and water ratios and amounts will change pipeline operating conditions. 2.2.3  Unconventional Reservoirs of Oil and Gas. According to Holditch et al. (2007), the term “unconventional” usually applies to a low-permeability reservoir (< 0.1 md) that produces mainly dry natural gas. Many of the low-permeability reservoirs developed in the past are sandstone, but significant quantities of gas are also produced from low-permeability carbonates as well as shales and coalbed deposits. Some shale formations have also yielded liquid hydrocarbons (RRCT 2011). Fig. 2.6 describes the geographic distribution of unconventional original gas in place (Dong et al. 2011) for various sections of the Earth. Note that at the time of the assessment by Dong et al. (2011), tight gas sands were the largest known source of future gas. However, development of shale gas may change these estimates significantly. Cramer (2008) reviewed the stimulation of the “unconventional reservoir,” and he noted that this term has different meanings to different people. Certain reservoirs that are termed “unconventional” have a rock matrix consisting of interparticle pore networks with very small pore connections imparting very poor fluid flow characteristics. The author claims that abundant volumes of oil or gas can be stored in these rocks, and often the rock is high in organic content and is the source of the hydrocarbon. However, because of marginal rock matrix quality, these reservoirs generally require both natural and induced fracture networks to enable economic recovery of the hydrocarbon. Rock types in this class include shale and coalbeds. Cramer (2008) noted that the term “shale” is a catch-all for any rock consisting of extremely small framework particles with minute pores charged with hydrocarbon and includes carbonate- and quartz-rich rocks. He claimed that another type of unconventional reservoir is the stacked pay unit, which exhibits somewhat better pore characteristics than in the case just outlined but with the individual units tending to be lenticular in shape and having an extremely small size or volume. These two classes of unconventional reservoirs are amenable to well stimulation.

Region

Coalbed Methane (P50)

Tight Sands Gas (P50)

Shale Gas (P50)

Total (TCF) (P50)

Austral-Asia

1,348

6,253

2,690

10,291

North America

1,629

10,784

5,905

18,318

859

28,604

15,880

45,343

13

3,366

3,742

7,122

9

15,447

15,416

30,872

176

3,525

2,194

5,895

Commonwealth of Independent States Latin America Middle East Europe Africa

18

4,000

3,882

7,901

World

4,052

71,981

49,709

125,742

Fig. 2.6—Geographic distribution of unconventional original gas in place, in trillion cubic feet (TCF) (Dong et al. 2011).

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 39

03/11/17 6:51 PM

40  Chemical and Mechanical Methods for Pipeline Integrity

When the above rock types become commercially exploited, they are known as resource plays. Once a low priority, the depletion of conventional reservoirs and improving price for oil and gas has driven unconventional reservoirs to an important place in the oil/gas industry. In some regions (i.e., Rocky Mountain Province in the US), unconventional reservoirs represent the primary target of current activity and remaining hydrocarbon development. Cramer (2008) claimed that given their unique petrophysical properties, each type of unconventional reservoir requires a unique approach to well stimulation, with often differing objectives than exist with conventional reservoir types. The paper reviews the characteristics of the basic unconventional reservoir types, lessons learned, and successful stimulation practices developed in completing these reservoirs; it also suggests areas for improvement in treatment and reservoir characterization and in treatment design. Hydraulic fracturing has contributed greatly to the economic producibility of these reservoirs. Natural gas production from shallow, fractured shale formations in the Appalachian and Michigan basins in the US has been under way for decades. What changed the game, as it were, was the recognition that one could “create a permeable reservoir” and high rates of gas production by intensely stimulating horizontal wells through multistage fracturing. Chapters 3 and 4 of Frenier and Ziauddin (2014) describe some of these advances. Details of several important unconventional reservoirs are presented next. Shale Gas and Oil Reservoirs. A very important type of predominately silicate formation that contains shale beds is now being frequently treated using proppant fracturing methods. Shale is the most common sedimentary formation on Earth (Boyer et al. 2011), but it is quite different from sandstones that contain sand grains mixed with clays and other minerals. Shale has been defined as a fine-grained clastic sedimentary rock formed from a clay mud that was compacted over time (Wikipedia 2010c). As such, the major minerals represented are largely low-permeability reservoir kaolinite, montmorillonite, and illite. This report noted that clay minerals of Late Paleozoic mudstones contain expandable clays, whereas in older rocks, especially in Middle to Early Paleozoic shales, illite clays predominate. The shales may be very dark, almost black in color, because of the presence of unoxidized carbon compounds as well as iron oxides (Fig. 2.7). These are also called “organic-rich shales.” Some shale formations, such as the Haynesville (Buller 2010), may also contain calcite and dolomite and may have HCl solubility up to 15%. Akrad et al. (2011) called these formations “prospective shales” because the clay content is usually less than 50%. They noted that the high carbonate content of some of these rocks makes them

(b) Shale sheets (a) Organic-rich shale Fig. 2.7—Shale rocks (USGOV.jpg) and shale sheets (Alexander et al. 2011; Boyer et al. 2011).

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 40

03/11/17 6:51 PM

From the Well to the Consumer  41

“soft,” and thus working these formations may require different fracturing fluids and proppants for stimulation. Shale has long been considered a “source” rock (Cramer 2008; McCarthy et al. 2011) as well as a barrier trap for migrating hydrocarbons. Fracturing of shale now makes it a viable producing formation, and these formations are now called “resource rock plays” (see Cramer 2008). The trapped carbon is the source of the methane/liquids and is the object of the fracture stimulation treatments. Shales can also be called slates. The major physical attributes are extremely thin lamella or parallel bands less than 1 cm thick (see Fig. 2.7); these are known collectively as fissility. The shale beds have very low permeabilities. Soeder (1988) examined a Marcellus shale that was free of a mobile liquid phase and had a measured gas porosity of approximately 10% under stress with a fairly strong “adsorption” component. Permeability to gas (k) was highly stress dependent, ranging from approximately 20 microdarcies (md) at a net stress of 3,000 psi down to approximately 5 md at a net stress of 6,000 psi (note that most oil/gas-producing sandstones have permeability values in the 1- to 1,000-md range). These properties make extraction of gas from shales extremely difficult unless they have been fracture treated. Kaufman et al. (2008) report that shale gas formations also have microfractures and cleats that provide access to the gas. In this characteristic, they share similarities with coalbeds that can also be fractured to produce methane. Very important fields of gas-producing shale include the Barnett Shale in central Texas and the Marcellus Group that underlies parts of New York, Pennsylvania, Ohio, and West Virginia. An important formation is the Eagle Ford Shale Group of Texas, which runs from the Mexican border almost to Dallas in the US. According to the RRCT (2011), the lower section of the Eagle Ford consists of organic-rich deposits and fossiliferous marine shales. Also, a small area of the Eagle Ford consists of a thin unit between the shales, and this area is especially amenable to hydraulic fracturing. The wells in the deeper part of the play deliver a dry gas, but moving northeastward (and with an updip), the wells produce more liquids (see the discussion in Boyer et al. 2011; McCarthy et al. 2011). One of the fields is actually an oil field (Eagleville, Eagle Ford). Even though the conditions are severe, these are the fields’ reservoir characteristics: • • • • •

6 to 10% porosity 200 to 600 md 7,000- to 10,000-psi bottomhole pressure 2.0 to 4.5 million psi Young’s modulus Bottom hole static temperature (BST) 270 to 300°F

More than 5 million bbl of oil have been produced in 2 years (RRCT 2011) from this reservoir. The field ends in the vicinity of Dallas, where an outcrop of Austin Chalk over the shale can be seen (Fig. 2.8). Thus, it runs for almost 400 miles (width approximately 50 miles) and with a maximum thickness of approximately 250 ft (RRCT 2011). These dimensions are unusual, because the kerogen in most oil shale formations must be removed using heat. Fig. 2.8 also shows the layers of the Marcellus Shale, which is a very important gas play that covers several eastern US states. Fig. 2.9 shows the location of major shale oil and gas plays. Note that other shale formations such as the Bakken in North Dakota, USA, are also producing significant volumes of liquid hydrocarbons. A US Energy Information Administration report (EIA 2014) claimed that as of 2014, the Bakken region is producing 1 million BOPD of liquids and 1 billion ft3/D of gas. The reader will understand that this is an ever-changing value. Table 2.1 describes the properties of the major shale basins in the US. Because these are active drilling/exploration areas, the map is continually changing. Boyer et al. (2011) and Alexander et al. (2011) also claimed that there are significant shale gasand oil-containing formations in Europe, Africa, Russia, China, and South Asia, but current production is not as developed as in North America.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 41

03/11/17 6:51 PM

42  Chemical and Mechanical Methods for Pipeline Integrity

Austin chalk

Eagle Ford shale Marcellus shale Fig. 2.8—Outcrops of Austin chalk, Eagle Ford shale (Wikipedia 2011d), and Marcellus shale (Arthur et al. 2009).

Fig. 2.9—Shale gas and shale oil plays (Boyer et al. 2011).

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 42

03/11/17 6:51 PM

From the Well to the Consumer  43

Gas Shale Basin

Barnett

Fayetteville

Haynesville

Marcellus

Woodford

Atrium

New Albany

Estimated area, square miles

5,000

9,000

9,000

95,000

11,000

12,000

43,500

Depth, ft

6,500– 8,500

1,000– 7000

10,500– 13,500

4,000– 8,500

6,000– 11,000

600–2,200

500–2,000

Net thickness, ft

100–600

20–200

200–300

50–200

120–220

70–120

50–100

Depth to base of treatable water, ft

~1200

~500

~400

~850

~400

~300

~400

Rock column thickness between top of pay and bottom of treatable water, ft

5,300– 7,300

500–6,500

10,100– 13,000

2,1257,650

5,600– 10,600

300–1,900

100–1,600

Total organic carbon, %

4.5

4.0–9.8

0.5–4.0

3–12

1–14

1–20

1–25

Total porosity, %

4–5

2–8

8–9

10

3–9

9

10–14

3–350

60–220

100–300

60–100

200–300

40–100

40–80

N/A

N/A

N/A

N/A

N/A

5–500

5–5,000

Well spacing, acres

60–160

80–160

40–560

40–160

640

40–160

80

Original gas in place, tcf

327

52

717

1,500

23

76

160

Technically recoverable resources, tcf

44

41.6

215

262

11.4

20

19.2

Gas content, scf/ton Water production, bbl/day

References to various values are in the cited paper. Table 2.1—Properties of key shale gas basins in the US (Kell 2009).

Some of the properties of the oil and gas produced from the shale plays, especially the Eagle Ford and ­Bakken, may cause significant difficulties for pipeline operations. ­ Table 2.2 shows gas compositions for dry-gas lines (Wocken et al. 2013), in addition to a Bakken gas, which is “wet” with higher hydrocarbons. As the fluids are cooled, NGLs will form and will make transmission difficult, which may necessitate interventions described in Chapters 5 and 6.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 43

Dry Pipeline Gas Composition, wt%

Bakken Gas Composition, wt%

Methane, CH4

92.2

55

Ethane, C2H6

5.5

22

Propane, C3H8

0.3

13

Chemical

Butane, C4H10

5

Pentane, C5H12

1

Hexane, C6H14

0.25

Heptane, C7H16

0.1

Nitrogen, N2

1.6

3

Carbon dioxide, CO2

0.4

0.5

1041

1495

Higher heating value, Btu.scf

Table 2.2—Gas composition comparisons.

03/11/17 6:51 PM

44  Chemical and Mechanical Methods for Pipeline Integrity

Directional drilling and hydraulic fracturing are vital technologies that have allowed these resources to be developed (see Arthur et al. 2009; King 2010). A diagram showing the placement of multiple fractures is in Fig. 2.10 (PackersPlus 2011). Additional physical descriptions of fractures are noted in a report by Rassenfoss (2015). Details of fluids and treatment methods specifically for shale are in in Frenier and Ziauddin (2014). These formations may produce dry gas, wet gas, and other liquid hydrocarbons, as well Multiple fractures as large amounts of nonconnate water that may affect and complicate pipeline operations. Fig. 2.10—Stimulation of a shale bed with multiple Tight Gas. Deep (hot), tight gas has been fractures (PackersPlus 2011). a major focus of fracturing and has led to the development of many of the chemistries described in Frenier and Ziauddin (2014, Chap. 4). Some tight gas carbonates can also be stimulated using acid fracturing, as explained in Frenier (2014, Chap. 3) and Bustos et al. (2007). Key to proppant fracturing design are the length of the fracture and temperature. Length is important to provide contact with the formation. Also, because temperatures are usually hot (>200°), fluid selection to accommodate the reactivity of the fluids is critical. Note that the conductivity of the proppant is not important, given that it will always be much higher than that of the formation. Because the hydrocarbon phase is dry gas (except for the fracturing water), the impact on the piping and pipelines may be less severe compared to a continual wet-gas flow that is encountered in shale gas completions. Coalbed Methane. Coal also is a low-permeability solid (Wikipedia 2010a). Almost all the permeability of a coalbed is usually considered attributable to fractures, which in coal are in the form of cleats. The permeability of the coal matrix is negligible by comparison. Coal cleats are of two types: butt cleats and face cleats, which occur at nearly right angles. The face cleats are continuous and provide paths of higher permeability, whereas butt cleats are noncontinuous and end at face cleats. Wikipedia (2010a) claims that on a small scale, fluid flow through coalbed methane reservoirs usually follows rectangular paths. The ratio of permeabilities in the face cleat direction over the butt cleat direction may range from 1:1 to 17:1. Because of this anisotropic permeability, drainage areas around coalbed methane wells are often elliptical in shape. Coalbeds are usually very wet, and therefore water must be removed to release the gas. Fracturing treatments and the use of chemical surfactants to drain the water are required. In addition, extremely large volumes of water are associated with retrieving coalbed methane. The high water volumes must be accommodated by appropriate scale and corrosion protection systems for the piping and lines. 2.3  Problems Anticipated in Different Areas of Pipeline-Connected Systems Different types of maintenance problems can be anticipated in gathering lines, surface facilities, and transmission/trunk pipelines. Many of these issues are, however, controlled by equilibrium changes in temperature, pressure, flow rate, and product composition and thus have similar chemical/ physical origins. This section describes known FA and integrity problems in three segments in the hydrocarbon supply train. This separation is somewhat arbitrary, given that piping and pipelines are integral to every system, including the refineries and chemical plants. In addition, what happens (including introduction of new chemicals) may affect the stability of subsequent sections because there is some carryover even if separation of the various chemicals is accomplished. Note that from a US regulatory standpoint (USDOT 2012), a hazardous pipeline includes “the pipeline or pipeline system, which means all parts of a pipeline facility through which a hazardous

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 44

03/11/17 6:51 PM

From the Well to the Consumer  45

liquid or carbon dioxide moves in transportation, including, but not limited to, line pipe, valves, and other appurtenances connected to line pipe, pumping units, fabricated assemblies associated with pumping units, metering and delivery stations and fabricated assemblies therein, and breakout tanks.” The primary federal regulatory agency in the US is the Pipeline and Hazardous Materials Safety Administration (see PHMSA 2012). Refineries and chemical plant facilities are not within the scope of this book because many of the chemical reactions needed to refine the crude oil are more complex (and at much higher temperatures) than those in the upstream/midstream sectors. However, the refineries and facilities are affected by the chemicals applied in the upstream parts of the systems. In addition, the unit operations and types of equipment in some refinery operations are similar to those in upstream treating facilities (see Section 2.3.2). For information on refinery cleaning processes, see Frenier (2001) and Frenier (2017b). Various types of additional products not directly associated with oil/gas production are conveyed using pipelines. Examples of the range of other materials include, but are not limited to, • • • • •

Coal and other solid slurries Alcohols and other liquid or gaseous chemicals Anhydrous ammonia Finished petroleum product (e.g., gasoline and diesel fuels) Potable water

All these lines may have FA and IM issues similar to those faced by oil/gas pipelines, but these lines are excluded from the current book because of space and time limitations. The reader is encouraged to examine Mohitpour et al. (2010), a book that mentions finished products but is also primarily associated with oil/gas transportation. In the normal parlance of the oil/gas industry, gathering lines, transfer lines, and surface facilities are considered part of the “upstream environment”; the transmission lines are described as “midstream”; and refineries and chemical plants are “downstream.” Note, however, that as long as the conditions are the same, chemistry and physics work the same regardless of the sector. So the lines between the various oil/gas sectors are frequently blurred, and operators therefore need knowledge of the entire supply train. 2.3.1  Gathering Lines and Wastewater Lines. Gathering lines are the piping systems that connect individual wellheads to larger collection and treatment facilities. These systems include onshore as well as offshore well production systems. See Fig. 1.5 for one example of a well gathering line layout. In this category are included lines that move partially treated fluids to further treatment before the fluids enter a transmission system. Also included is piping that feeds water for flooding, EOR, and disposal activities. The water is generally recycled water from other production wells. The wastewater lines can be composed of polymers, which are not easily corroded, as well as large steel lines (see Harris et al. 2010). The latter are subject to corrosion, scale, and microbiologically influenced corrosion (MIC). The waste water from oil/gas operations is becoming a valuable commodity. A report by Boschee (2015) describes the very complex system of water recovery and treatments conducted by Chevron in California. In this case, the recovered water is used for agricultural treatments. The article mentions an 8-mile internally coated pipeline constructed as part of this system. In the shale oil/gas fields, very large volumes of water are needed for fracturing treatments; there, the highly contaminated returned fluids are being reclaimed, and much of the water transits newly constructed pipelines (see Locke and Kimball 2013; Rao 2015). Methods for cleaning recycled water are described in Section 8.4.1. Fig. 2.11 shows a small section of a complex onshore system with numerous connections, valves, filters, and control devices that can become fouled and/or corroded. Ruptures of wastewater pipelines have caused damage in various oil/gas fields, including several in the Bakken play (Scheyder 2015; AP 2014). Details of wastewater treatment methods and spill cleanup are in Sections 8.4.1 and 8.4.2.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 45

03/11/17 6:51 PM

46  Chemical and Mechanical Methods for Pipeline Integrity

Gathering lines may initially experience wellhead temperatures and pressures as well as multiphase and multichemical mixtures. The temperature and pressure changes that occur as the fluids transit the gathering lines may set up conditions for various types of corrosion, including MIC; formation of inorganic scales; and organic deposits, including gas hydrates and wax. Fig. 2.11—Fluid injection system. Producers with subsea wells usually explicitly plan to deal with wax and hydrate deposition if conditions favorable to their deposition exist at any time in the well life cycle (Frenier et al. 2010; Kelland 2013; Borden 2014). Pressure changes and mixing of incompatible waters provide the conditions for forming calcite and sulfate scales. Asphaltene deposition is less likely in this segment, but it is not unknown. Risers in subsea producing systems connect the seafloor facilities (i.e., gathering systems and subsea separators) to the surface. These piping systems may convey untreated well fluids, like gathering lines, or they may convey partially treated gases or liquids. Under the categories of gathering lines are the various connections to/from the wellhead for various activities, including injecting fluids into the well/reservoir systems. The mixing of various well fluids at manifolds may also change the chemical compositions, initiating scale, organic solids, and conditions for corrosion. Gathering lines are constantly subject to chemical/ physical upsets as wellfield conditions change. Thus, many kinds of problems may occur in a gathering line. Wint (2013) claimed that there is a range of issues for gathering lines serving shale plays that produce “rich gas” (i.e., wet gas), which contains significant levels of liquefiable hydrocarbons (such as ethane or propane and higher alkanes), along with methane gas. Liquids can accumulate at low-elevation points along gathering systems where the high liquid concentrations in the gas streams cause significant issues with slugging, high differential pressures (liquids loading), and corrosion (see Preface as well as Section 2.2.3). These conditions may require almost constant pigging (see Sections 5.5.1 and 5.2.1). In addition, crude oil containing high levels of paraffin and other flow-reducing contaminants (e.g., fracturing sand, chlorides, and spent chemicals) presents flow restriction issues in these upstream pipeline systems. Many factors contribute to the overall performance and flow efficiency of pipeline systems that may include the elevation profile, flow volumes, product quality, and temperature; all pipelines must be evaluated on an individual basis. The gathering lines and piping connections are subject to a wide range of possible corrosive attacks (see Chapter 3). Because of the wide variety of chemicals and mixtures in the lines, corrosive attack is difficult to eliminate, especially as the lines age (see discussion in Berger 2007). 2.3.2  Surface and Subsurface Facilities. Surface and subsurface facilities include primary and secondary separation facilities where the aqueous, hydrocarbon liquid, and gaseous phases are treated to produce pipeline-transmission-quality products. Mokhatab et al. (2006) noted that processing is designed to separate natural gas, condensate, incondensable liquids (e.g., water and hydrocarbons), and acid gases. These processes will allow the further processing and sale or reuse of the valuable products. Several current engineering/chemical operations are in use to achieve these goals. Fig. 2.12 shows a generic flow chart of the processes. In the diagram, the fluids are usually subjected to a physical separation sequence to produce aqueous, liquid hydrocarbon, and gaseous phases. The different phases may then be given additional treatment, depending on the original composition and needs (details are in the 2.3.2 subsections).

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 46

03/11/17 6:51 PM

From the Well to the Consumer  47

Acid gas removal

Well fluids Physical separation

Solids

Oil

Gas

Compression

Water Pipeline

Refinery

Injection/recycle

Fig. 2.12—Well fluids treatment processes.

Treatment facilities range in complexity from a single tank or tower at a well production site to very large integrated plants. Walton (2013) showed a photo (Fig. 2.13) of a gas/liquids processing plant (ONEOK Partners) in the US state of North Dakota. This type of substantial plant is necessary in field locations to prepare large volumes of well products for further transit. Such plants may serve a part of a well field, with the individual well connected to units with pipelines. Fig. 2.14 shows a photo­­ graph of a land-based gas treatment plant that is slightly different from the plant shown in Fig. 2.13. Note the amount of piping that connects to underground lines. The complexity will be determined by the fluid chemical quality, volumes, and other processing re­­ quired downstream to achieve transport-quality products.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 47

Flare stack Separation or treatment towers

Pipelines

Fig. 2.13—Gas treatment plant, ONEOK Partners Stateline Processing plant in North Dakota (Walton 2013).

Piping

Reactors/separators

Fig. 2.14—Land-based gas treatment plant.

03/11/17 6:51 PM

48  Chemical and Mechanical Methods for Pipeline Integrity

Shell and Tube Heat Exchanger

Outlet tube side and valve

Inlet shell side and valve Inlet tube side and valve

Tube sheet

Head (open)

Inlet shell side and valve

Fig. 2.15—Drawing of a shell and tube heat exchanger.

Reboiler

Hot process vapor and liquid outletq Steam inlet

Tubes

Hot condensate outlet

Warm process liquid inlet

Fig. 2.16—Reboiler schematic.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 48

An example of a complex integrated treatment plant was described by Jarragh et al. (2013). They noted that the Kuwait Oil Company has facilities that consist of 22 crudeprocessing plants (or gathering centers) as well as four gas processing plants (or booster stations). Two effluent water disposal plants, as well as a seawater treatment plant and injection plants, are included in a vast network of pipelines carrying different products, as well as early production facilities. They also provided several detailed diagrams that show the large number of individual treatment units, pumps, and connecting piping that must be protected from corrosion and flow problems. Note that these diagrams are not reproduced here because of their complexity. Jarragh et al. (2013) described several important types of processing units. These include heat exchangers (Fig. 2.15), reboilers (Fig. 2.16), and various towers

03/11/17 6:51 PM

From the Well to the Consumer  49

used for hydrocarbon separation as well as for contacting a liquid with a gas to cause a chemical or physical reaction. This will include dehydration or removal of an acidic gas. One example of a bubble-cap tower is shown, though many other designs are possible. See Fig. 2.17 for details of the gas contact in the countercurrent flow reactor (the cleaning of these units is covered in Section 7.6). Numerous types of physical separation devices are shown in Figs. 2.18 and 2.19. A wide range of organic, inorganic, and mixed deposits can foul these units. These solids come from the inlet gases and liquids as well as from the reactions that take place in the units. See Engel and Sheilan (2014) as well as examples in Section 4.5.

Liquid flow

Gas flow

Acid gas removal tower

Overflow weir Downcommer

Liquid flow

Gas flow

Details of flow Fig. 2.17—Bubble-cap gas removal tower. Gas to compression, dehydration

Typical crude oil treatment system (separation, heating, dehydration, stabilization, storage, metering, pumping)

Multiphase meter (Test separator)

Gas to compression, processing

MP separator

Oil dehydration Gas to compression, processing

Electrostatic treater, heater treater degasser, desalter

Water-to-water treatment

LP separator Crude heater

From wells

Manifold

Water-to-water treatment

Crude oil storage tank

Water-to-water treatment

Crude/crude exchanger

Crude oil storage tank

Crude oil metering skid Crude oil booster pumps

To crude oil export pipeline Crude oil export pumps

Fig. 2.18—Separator layout (Schlumberger 2010).

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 49

03/11/17 6:51 PM

50  Chemical and Mechanical Methods for Pipeline Integrity

Coalescing plates

Safety valve Inlet breaker

Deflector plate

Foam breaker

Mist extractor

Gas line

Effluent inlet

Water-level controller and float

Water line Weir plate

Oil line

Oil-level controller and float Vortex breaker

Fig. 2.19—Gravity settler (Sims 2010).

The production of oil and gas from the shale formations of North America has developed so rapidly that the industry has challenges keeping up with the demand. Boschee (2014d) reported that the shorter time frame for construction of various facilities in the shale environment may reduce the planning time. This means that the safety of the designs, especially pressure containment (and thus the integrity of the piping), is a major consideration. Physical Separation. The surface facilities are similar to some refinery operations in that chemical and mechanical methods are used, including heating and cooling of fluids. Most initial separations depend on gravity and density. This may take place in tanks as well as in more-complicated equipment. The aqueous, liquid hydrocarbons and the gases may also undergo additional physical and chemical reactions to dehydrate the hydrocarbons and remove CO2 and hydrogen sulfide (H2S). Fig. 2.18 shows a detailed layout for initial surface separation of the different phases when oil, water, and gases are present. Kokal (2006) and Kelland (2009) note that application of heat promotes oil/water separation and accelerates the treating process. An increase in temperature has three significant effects: It reduces the viscosity of the oil, increases the mobility of the water droplets, and elevates the settling rate of the water droplets. Fig. 2.19 describes a gravity separator in more detail. Some subsea separation equipment can provide a level of dehydration and liquid/gas separation (for more information, see Estefen et al. 2005). Mokhatab et al. (2006, Chap. 5) examined phase separation theory and designs. Because of the great variety of chemical processes that are used, separation facilities are frequently fouled with mixtures of organic solids, inorganic solids, and corrosion products. Bedwell et al. (2015) reviewed problems found in aging facilities and how both engineering and chemistry can lengthen the life of these vital units. A drawing by these authors indicates some of the problems, which include foam, emulsions, and scale. The planning “circle” (Fig. 2.20) indicates process steps for reducing the impact of these issues, which are addressed more fully in Chapter 4. In many cases, the separation of solids, as noted by Rawlins (2013), is a highly important process step that must preceed some of the other treating steps. This author reviewed the literature on solids prevention at the well using sand control methods described in the paper and by King et al. (2003). Also reviewed was physical separation in the surface facilities. Rawlins further noted that the solids can be natural substances from the reservoir (including sand, clay fines, and scale) and introduced solids (including fracturing proppant). This latter category is increasingly present as oil/gas production from shale plays includes more hydrocarbon production.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 50

03/11/17 6:51 PM

From the Well to the Consumer  51

Mist elim.

Foarm brkr.

Inlet

Foarm brkr.

Gas outlet Foarm brkr.

Mist elim.

Foarm brkr.

Gas outlet

1. Theoretical modeling Oil outlet Water outlet

Water outlet 5. Optional process dianostics

4. Chemistry and flow assurance

2. Fluids and solids sampling and analysis

3. System mapping

1,000 ppm oil-in-water

5% Basic sediment and water

Fig. 2.20—Flow assurance (FA) problems and planning for maintaining separation facilities (Bedwell et al. 2015).

Rawlins (2013) claims that desanders can be incorporated into a surface facility (Fig. 2.21a), or they can be standalone if sand/solids production is of a higher volume than can be handled by the sand filter (Fig. 2.21b). Dehydration. After the primary separation, a natural gas stream may require further dehydration and removal of the acid gases (H2S, CO2, and organic acids such as acetic acid). These operations are required to reduce corrosion to pipelines and to provide a method for controlling scale and gas hydrate formations. Dehydration may follow acid gas removal or may be a standalone operation if the gas does not need acid removal treatment (see the following subsection for details of acid gas treatments). Mokhatab et al. (2006) claimed that there are two fundamentally different processes for dehydration: refrigeration as well as removal of the water using liquid or solid desiccants. Chapter 7 of their book has detailed information on these processes. Several low-temperature processes, among them refrigeration, are able to effect some degree of dehydration and have been used for many years (Records and Seely 1951). Records and Seely (1951) described the technology for low-temperature dehydration of natural gas using the JouleThomson (J-T) effect. The change of temperature with respect to a change of pressure in a J-T process is the J-T coefficient (Kittel and Kroemer 1980):

( )

µ JT = ∂T ∂P

H

. ���������������������������������������������������������������������������������������������������������������������������� (2.1)

The value of mJT is typically expressed in °C/bar (SI units: K/Pa) and depends on the specific gas as well as the temperature and pressure of the gas before expansion. The process works by reducing the temperature of a water-saturated gaseous mixture by expansion through a restrictive orifice (i.e., J-T effect) or by other methods described in subsequent paragraphs. It also reduces the equilibrium water content of the gaseous mixture as a direct function of temperature. Then, gravity difference forms and separates hydrate particles from the gaseous mixture.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 51

03/11/17 6:51 PM

52  Chemical and Mechanical Methods for Pipeline Integrity

Facilities Desander (a) Wellhead

Produced water desander

Choke

Well-stream desander with integral accumulator Well-stream desander with external accumulator

Solids transport and disposal system

Wellhead Desander (b)

Fig. 2.21—Desander applications (Rawlins 2013).

Note that chilling a gas stream by J-T expansion can also lead to hydrate formation if the pressure is high enough. Dehydration by a mechanical refrigeration using this or other methods (e.g., external gas expansion chillers using gas J-T processes) with glycol injection to prevent hydrate formation is in current use (Huffmaster 2004). Mokhatab et al. (2006) noted that propane is a commonly used refrigerant in such units. These types of chillers can also be used to produce NGL. Hubbard (1991) claimed that the removal of water from natural gas can also be accomplished in several additional ways. Commercial methods currently used when this report was written included absorption using glycol dehydration and adsorption using a dry desiccant. These authors noted that the two methods use mass transfer of the water molecule into a liquid solvent (glycol solution) or a crystalline structure (dry desiccant). The third method uses cooling to condense the water molecule to the liquid phase, with the subsequent injection of inhibitor (glycol or methanol) to prevent hydrate formation. Glycol absorption and dehydration involve the use of a liquid desiccant to remove water vapor from the gas. In this process, the water forms a stable solution with the complex alcohols. Although many liquids possess the ability to absorb water from gas, the liquid that is most desirable to use for commercial dehydration purposes should possess the following properties: • • • • • •

High absorption efficiency Easy and economic regeneration Noncorrosive and nontoxic properties No operational problems when used in high concentrations No interaction with the hydrocarbon portion of the gas No contamination by acid gases

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 52

03/11/17 6:51 PM

From the Well to the Consumer  53

HO The glycols, particularly monoethylene glyOH col (MEG), diethylene glycol (DEG), triethMonoethylene glycol ylene glycol (TEG), and tetraethylene glycol (TTEG) (see Fig. 2.22), can satisfy most of O these criteria to varying degrees. HO OH Water and the glycols show complete mutual Diethylene glycol solubility in the liquid phase because of hydroO OH gen-oxygen bonds, and their water vapor presO HO sures are very low. One frequently used glycol for Triethylene glycol dehydration is TEG. The wet gas is dehydrated in the absorber, and the heat in the stripping column O O regenerates the water-free glycol. A drawing of a HO O OH generic recycle system is presented in Fig. 2.23. Tetraethylene glycol Additional details of reboilers, exchangers, and towers are in Figs. 2.16 and 2.17. Fig. 2.22—Glycols used in dehydration plants. Glycol absorption and dehydration as described is equivalent to any distillation process with an alcohol/water mixture. Note that the gas enters at the bottom of the contactor and moves countercurrent to the glycol, which is regenerated by heating. Devold (2013) claimed that the reboilers operate at 260 to 350°F and the distillation process may take several hours to strip out all the water from the “rich” glycol. These high-temperature processes can produce complex deposition products that may require mechanical or chemical removal (Sections 5.2 and 7.3). Another description of the operation of a glycol dehydrator is in Moore et al. (2008). The glycol stream should be recharged constantly because some the absorbent may react and form heaver molecules, which should be removed by the filter. Note that some of these same glycols (along with methanol) are used as inhibitors of hydrate formation. Mokhatab et al. (2006) described the use of solid desiccants, which adsorb water onto a veryhigh-surface-area solid without causing a specific chemical reaction other than formation of hydrogen bonds. The authors list three types of solid desiccants: silica gel (noncrystalline SiO2); activated alumina (Al2O3); and molecular sieves, which are crystalline aluminosilicates that include interconnecting cavities. These may be similar to some natural tectosilicates described in Frenier and Ziauddin (2014). An example is in Fig. 2.24. Gas out

Scrubber Lean glycol

Vaporized liquid

Reboiler A

Heat Reboiler B

Heat Gas in

Holding tank Rich glycol

Fig. 2.23—Glycol recycle (Devold 2013).

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 53

03/11/17 6:51 PM

54  Chemical and Mechanical Methods for Pipeline Integrity

Fig. 2.24—Tectosilicates structure.

The dry desiccants may produce more-effective dehydration than the glycols, but they are bulky. They also must be regenerated using a heated gas at 400 to 500°F. The choice of process depends on the specification of the gas stream and the physical location and cost requirements (for details, see Mokhatab et al. 2006). The piping in the dehydration systems and transfer pipelines are subject to corrosive attack and fouling by polymeric residues of the glycol solvents produced at the high temperatures. Acid Gas Removal. Dehydration of gas is a mostly physical process whereby the water is absorbed or dissolved into a second phase (with TEG). Most of the chemistry involved may, especially for the glycols, be harmful to the systems because of the breakdown of the desiccants as a result of the heat of the regeneration steps. The treatment of the acid gases involves more chemistry for the removal, as well as for the regeneration steps. Although there are many processes for removing H2S and/or CO2 from gas streams (Kidnay and Parrish 2006), the most widely used are the amine systems (Mokhatab et al. 2006). These systems depend on the reactivity of H2S and CO2 with an amino nitrogen. The most currently used processes are monoethanolamine (MEA), monodiethanolamine (MDEA), diglycolamine (DGA), and ­diisopropylamine (DIPA). Diglycolamine is a trademark of the Jefferson Chemical Company; the other designations are chemically accurate in referring to the structure of the compound (Fig. 2.25). Acid gas removal involves acid/ base chemical reactions. Fig. 2.26. HO H shows a schematic of an acid gas NH2 N removal and dehydration plant HO OH Monoethanol amine Diethanolamine DEA (Al-Qahtani and Garland 2013); this figure also shows the layout of various towers and heat exchanges described in Figs. 2.15, 2.16, and NH2 OH 2.17. This unit ­(Fig. 2.26.) has a NH2 N two-stage acid gas removal proHO cess, followed by dehydration of O HO N-methyldiethanolamine MDEA the gas by contact with TEG. Note Diglycolamine DGA that if the gas is not acidic, only the TEG process may be needed. Fig. 2.25—Acid gas absorption chemicals.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 54

03/11/17 6:51 PM

From the Well to the Consumer  55

Lean amine: no acid gas Rich amine: loaded with acid gases Sales gas Propane chillers Lowtemp/highpressure acid gas contactor (1) Slug catcher

Lean amine chiller

TEG contactor

Lean amine cooler

Lean/rich amine exchanger

High-temp/lowpressure acid gas regenerator (1)

To thermal oxidizer

Acid gas Contactor (2) Acid gas feed chiller

Lean amine chiller

Lean amine cooler

Lean/rich amine exchanger

Acid gas regenerator (2)

To sulfur-recovery unit feed header

Fig. 2.26—Schematic of acid gas removal and dehydration plant (Carpenter 2014; Al-Qahtani and Garland 2013).

Different reactions are needed for removal of the acids and regeneration of the amine. These are acid/base reactions that form salts that can be regenerated by heating. Note that the first step is the absorption of the acid gas by the liquid phase, which is an amine/water mixture. Typical concentrations are as follows (Kohl and Nielson 1997): • • • •

MEA: approximately 20% for removing H2S and CO2 and 32% for removing only CO2 Diethanolamine: approximately 20 to 25% for removing H2S and CO2 Methyldiethanolamine: approximately 30 to 55% for removing H2S and CO2 DGA: approximately 50% for removing H2S and CO2

The rate and mechanisms of this process must be understood for each gas and liquid pair. Zare Aliabad and Mirzaei (2009) noted that the chemistry is different for primary and secondary amines (MEA and DEA; see Fig. 2.26.) as compared with tertiary amines such as MDEA. In the following set of equations (Eqs. 2.2 and 2.3), A and B represent primary amines (or a secondary amine with a minor change), and C–E represent tertiary amines.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 55

03/11/17 6:51 PM

56  Chemical and Mechanical Methods for Pipeline Integrity

With H2S. (A) 2RNH 2 + H 2 S ⇔ (RNH 3 )2 S (B) (RNH 3 )2 S + H 2 S ⇔ 2(RNH 3 HS) (C) H 2 S ⇔ HS− + H + +

(D) RR ′ R ″ N + H ⇔ RR ′ R ″ NH

. ���������������������������������������������������������������������� (2.2) +

(E) RR ′ R ″ NH + + HS− ⇔ (RR ′ R ″ NH + )(HS− ) With CO2. (A) 2RNH 2 + CO 2 + H 2 O ⇔ (RNH 3 )2 CO 2 (B) (RNH 3 )2 CO 2 + H 2 O + CO 2 ⇔ 2(RNH 3 HCO 3 ) (C) CO 2 + H 2O ⇔ H + + HCO3_ +

(D) RR ′ R ″ N + H ⇔ RR ′ R ″ NH

.���������������������������������������������������������������� (2.3) +

(E) RR ′ R ″ NH + + HCO _3 ⇔ (RR ′ R ″ NH + )(HCO _3 ) MEA has been the most widely used of these compounds (Frenier 2001). It is a clear, colorless liquid that boils at 338°F (170°C) at atmospheric pressures. Of all the amines used, it is the strongest base and so reacts most readily with acid gases. However, Kelland (2009) claimed that at the time of publication of his book, MDEA was frequently used for H2S removal. As noted in Eqs. 2.2 and 2.3, because MDEA is a tertiary amine and has no mobile proton, it cannot react directly with CO2, but will form a salt with the dissolved bicarbonate (HCO3–). Kelland also claims that the CO2 can then be removed with another amine or by carbonate/bicarbonate absorption. MDEA also forms a salt with HS–, but the reaction is faster than with CO2 (see the discussion of the chemistry by Zare Aliabad and Mirzaei 2009). This MDEA material also has a lower vapor pressure than MEA as well as a lower heat of reaction with H2S and CO2 and is thus more economical to regenerate. An important understanding is that these reactions proceed to the right at low temperature and high H2S partial pressures. For this reason, the absorption is normally conducted at pressure of 300 psi or higher and at or near ambient temperature. Regeneration is normally performed at or near atmospheric pressure and at the solution boiling point. The sour gas containing H2S and/or CO2 enters the plant through a scrubber, which removes any free liquids and/or entrained solids. The sour gas then enters the bottom of the absorber (similar to Fig. 2.23) and flows upward in countercurrent contact with the descending aqueous amine solution (Fig. 2.25). Note that the equilibrium reactions (Eqs. 2.2 and 2.3) are reversed by the higher temperatures and lower pressures in the regeneration units; however, complete regeneration cannot be accomplished. Temperature and flow rate are several of the controllable variables (Zare Aliabad and Mirzaei 2009). Sweetened gas leaves the top of the absorber and flows to a dehydration unit (Fig. 2.26), where saturation water from the aqueous amine solution is removed. The fluids are potentially extremely corrosive, and therefore choice of the proper metallurgies is critical. There is also the potential for breakdown of the reactants and production of fouling solids. As noted in Fig. 2.25, the H2S removed from the gas stream is then converted to elemental sulfur (S0) using a chemical process. A major method is the Claus process (USEPA 1995), which uses multiple steps that are conducted at high temperatures (>1800°F in a furnace, then 600°F in a catalytic reactor). The overall reaction is 2H 2 S + O 2 → 2S + 2H 2 O .���������������������������������������������������������������������������������������������������������� (2.4)

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 56

03/11/17 6:51 PM

From the Well to the Consumer  57

A report by Boschee (2014c) claims that handling sour crude oil and gas is becoming an important feature of oil/gas production, especially in the MEA sector. The author also described alternative processes in addition to the Claus reactions noted in Eq. 2.4. See this review for details. Also see Kidnay and Parrish (2006) and Mokhatab et al. (2006) for more-comprehensive information on gas treatment technologies. The acid gas and dehydration equipment are subject to fouling from the breakdown of the glycols and the amines producing polymers as well as various corrosive attacks and corrosion products. Any oxygen content in the gas stream or overheating in the reboiler will cause amine degradation and the formation of heat stable salts, which can cause erosion/corrosion. FeS deposits will be a common solid, and it is difficult to remove this scale (see Sections 3.1.4, 4.2.2, 6.5.1, and 7.3.2). The hightemperature foulants produced in Claus plants may require mechanical cleaning (see Section 7.6.2 and Frenier 2017b). Offshore Processing. Processing offshore includes topside facilities on platforms of all descriptions, from floating production storage and offloading (FPSO) platforms meant to handle as much as 100,000 BOPD (Gaidhani 2014) to the large compliant tower and tension leg structures in the North Sea and Gulf of Mexico. These units may also handle very large volumes (100,000 BOPD of oil equivalent) produced as multiphase fluids (Shell 2014). In these facilities, Gaidhani (2014) noted, the major separation/treatment is dehydration as well as oil/gas separation. A setup similar to Fig. 2.18 would be used. The author of this report (Shell 2014) noted that corrosion of steel (solved with addition of a corrosion inhibitor—see Section 6.1) is one issue. Shell (2014) also claimed that changes in viscosity because of water entrainment is an FA concern and may affect throughput and heat transfer in the exchangers and separators (see Fig. 2.19). Also see Section 2.1.5. Subsea Technologies. Subsea units that process hydrocarbon/water streams are considered to be a major emerging technology area and are still in the prelarge-scale commercial stage at the time of this report (Rassenfoss 2013). This subsection reviews several technologies that may make major contributions to subsea well developments. Rigzone (2011) claimed that subsea processing can encompass a number of different processes to help reduce the cost and complexity of developing an offshore field. The main types of subsea processing could include • • • • • •

Subsea water removal Reinjection or disposal Single-phase and multiphase boosting of well fluids Sand and solid separation Gas/liquid separation and boosting Gas treatment and compression

Possible technologies have been proposed or implemented in small scale. A patent by Anderson and Allen (2004), for example, describes a subsea separator that can be a free-water knockout type, which could be a vertical vessel standing upright or a horizontal vessel lying on its side. Optionally, a subsea separator can include a three-phase separator to separate water, liquid hydrocarbons, and gaseous hydrocarbons from well fluid conveyed from a subsea wellhead. The water that is separated in the subsea separator typically still has gaseous and possibly liquid residual hydrocarbons. The water with gaseous and possibly liquid residual hydrocarbons is contaminated and is not acceptable to be dumped into the sea without further treatment. Another patent, this by Parksa and Aminb (2012), describes a method that uses the concepts of gas cooling through expansion and the controlled formation and management of gas hydrates to reduce the water content of a saturated gas stream to levels suitable for gas transport in subsea pipelines. A pilot plant was implemented to test the solution design at inlet pressures up to 10 MPa using natural gas from a domestic gas main at flow rates up to 35 std m3/h. Parksa and Aminb (2012) claimed that

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 57

03/11/17 6:51 PM

58  Chemical and Mechanical Methods for Pipeline Integrity

the pilot plant successfully demonstrated dehydration of the natural gas stream to a water content suitable for transportation, without condensation of liquid water, at subsea pipeline temperatures of 4°C. Although the concept has been demonstrated, the implementation of the concept in a fieldviable manner is pending. Estefen et al. (2005) described the design of a subsea gas-gathering system including subsea dehydration and compression. The various possible layouts are considered, but the details of the separations are not listed. Bass (2006) described a subsea gas dewpoint modifying/dehydration process that could be used for several applications. One is to reduce FA costs by eliminating or minimizing the need for continuous hydrate inhibition to reduce pipeline costs by removing water. This will reduce corrosion and may allow use of carbon steel rather than a corrosion-resistant alloy due to the reduced water content. However, hydrates will form on these alloys if the conditions are correct (see Section 4.3.4). A second application is to process the gas to sales quality; doing this also addresses FA needs. A different technology has been proposed for subsea. These applications would use a supersonic nozzle (Laval nozzle) to condense vapors from a gas feed and inlet guide vanes that impart a spin to the fluid to separate the condensed liquids. At present several supersonic separator units are in commercial operation. One of these units has been in operation on an offshore platform in Malaysia since 2004, and a conceptual design has been completed for a subsea unit. The technology is driven by free pressure drop comparable to a turbo-expander system, which is significantly less than a Joule-Thompson cooling process. Consequently, it is more likely to be applicable to a high highpressure gas fields where it has proven significant added value. Twister (2012) describes this process: The separator has thermodynamic principles similar to those of a turbo-expander, combining the following process steps in a compact, tubular device • Expansion • Cyclonic gas/liquid separation • Recompression Whereas a turbo-expander transforms pressure to shaft power, supersonic separators achieves a similar temperature drop by transforming pressure to kinetic energy (i.e., supersonic velocity). It is stated that multiple inlet guide vanes generate high-vortices concentric swirl (up to 500,000 G, with G signifying the gravity force). A Laval nozzle (Fig. 2.27) is used to expand the saturated feed gas to supersonic velocity, which results in a low temperature and pressure. This results in the formation of a mist of water and hydrocarbon condensation droplets. The results claimed are as follows

Saturated Feed Gas

Cyclone Separator (500,000g)

Static Guide Vanes

Vortex Generator

Diffuser Dry gas

Tapered inner body Leval Nozzle

Liquids + Slip-gas

Fig. 2.27—Sonic separator (Twister 2012).

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 58

03/11/17 6:51 PM

From the Well to the Consumer  59

• The high-vortices swirl centrifuges the droplets to the wall. • The liquids are split from the gas using a cyclonic separator. • The separated streams are slowed down in separate diffusers, typically recovering 70 to 75% of the initial pressure. • The liquid stream contains slip-gas, which will be removed in a compact liquid degassing vessel and recombined with the dry-gas stream. Wen et al. (2013) described developments of a subsea system based on the Laval system. They stated that the supersonic separator mainly consists of a Laval nozzle, a swirl device, and a diffuser. The Laval nozzle is used to expand the saturated gas to supersonic velocity, resulting in low temperature and pressure. In these conditions, nucleation of the water vapor and hydrocarbon may occur, followed by droplet growth. The swirl device generates a high-vortices swirl to centrifuge the droplets to the wall, and liquids are separated from the gas mixtures. Then the diffuser will recover 60 to 80% of the initial pressure of the dry gas, which will stay in the center area of the tube. There are many advantages of this novel technique. First, it is an environmentally friendly apparatus because it requires no chemicals and has no emissions. Second, it is a static device, given that it does not use rotating equipment. Also, it is small in size and lightweight and enables unmanned operation for personnel safety. Therefore, it is suitable for platforms and subsea gas processing. Subsea Installations. Several subsea installations have been made or are planned. A project in the Norwegian area of the North Sea is claimed to be the first full-scale commercial subsea separation, boosting, and injection system (FMC 2013). FMC Technologies (FMC 2013) noted that the well fluid from the Tordis field is first routed into the subsea separator vessel. The inlet cyclone in the vessel completes a first separation, in which the majority of the gas is separated out and routed through a separate gas bypass. The remaining water, oil, and gas are separated through the gravity principle inside the vessel. The water is then pumped by means of a water injection pump directly back into the Utsira reservoir through the tree. Any deposit of sand inside the separation tank is handled by the sand removal system. Moraes et al. (2012) considered the problem of sand to be the main challenge for the development of an extensive whole subsea water separation and reinjection pilot system for Marlim field offshore Brazil, a project known as the SSAO Marlim Project. Extensive onshore pilot-level tests have developed the use of a gravity-based sand separator, followed by a two-stage hydrocyclonic water/oil separation system, which is claimed to provide water-quality values that are acceptable for subsurface reinjection of the water. Albuquerque et al. (2013) provided more details of a proposed subsea processing system. Fig. 2.28 shows a block diagram of the process steps in the proposed system. Some level of gas/liquid separation—called a vertical annular separation system—was claimed to be in use (or tested) in the North Sea and in offshore Brazil projects; however, the other operations are in the test phase. Grave and Olson (2014) described developmental activities for a compact subsea processing facility that could be used in deep water. Types of units evaluated include • • • • •

Hydrocyclone-based separators In-line desanders Electrostatic coalescers Oil/water pipe separators Gas deoilers

Subsea Technology Issues. In relation to subsea separation technologies, FA issues include scaling and fouling by gas hydrates and waxes because the product of separation may be subjected to low temperatures during various transfer operations. Duarte et al. (2012) described the concerns in multiphase lines, water lines, and water injection systems.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 59

03/11/17 6:51 PM

60  Chemical and Mechanical Methods for Pipeline Integrity

Floating production unit Gas

Oil

Pump Gas Solids Gas Gas/liquid separation Oil water

Liquid (Gas< 3%)

Solid/liquid separation

Gas solids

Well production

Liquid (Gas< 3%)

Water/oil separation

Liquid (BSW< 50%)

Water/oil electrostatic separation

Oil

Water

Water Hydrocyclones

Pump Water for reinjection Injection well

Fig. 2.28—Diagram of proposed subsea multiphase separation system (Albuquerque et al. 2013).

Hydrate prevention is extremely challenging because of several open connections between the multiphase lines and the water lines. Hence, such usual means as MEG inhibition and thermal insulation have not been sufficient to ensure the hydrate prevention strategy, and new strategies have been developed. It has been necessary to challenge the strategies in every part of the system. 2.3.3  Water-System-Related Issues. The volume of water coproduced with oil and gas or the flowback water that is present after stimulation treatments frequently exceeds the volume of oil and gas produced. As reservoirs age and innovative processes such as hydraulic fracturing of shale beds increases (Frenier and Ziauddin 2014), use, reuse, and disposal of water increasingly affects the oil, gas, and pipeline industries. SPE (2011) considered water availability and produced-water handling to be one of the “Grand Challenges” of the future. The reuse of produced/flowback water is common within the oil/gas production sector, given that these fluids are valuable commodities for pressure maintenance, EOR, or hydraulic fracturing. After preliminary separation of hydrocarbons, some level of treatment may be required to meet various water-quality specifications. A large variety of piping systems connect the water treatments to the wells and to possible injection and reuse facilities. Several levels of treatment may be required. Fig. 2.29 shows a diagram of a simple system that separates the phases and has a tank to allow gravity settling of entrained solids. The fluids may also flow into a fracturing flowback pit (Tinto and Solomon 2010), shown as Fig. 2.30, and this may be a stage in the collection/treatment process. Fig. 2.31 shows a diagram of a primary water treatment process for recycle (Boschee 2012). Boschee (2012) notes that the economics of using recycle of flowback water for large well fields include the benefits of lower transportation and disposal costs as well as environmental benefits.

BK-SPE-CHEMICAL_AND_MECHANICAL-170274.indb 60

03/11/17 6:51 PM

From the Well to the Consumer  61

Each step may remove a different potential undesirable entity. The anaerobic basin tries to remove sulfate-reducing bacteria, and aeration may remove iron. Note that this process does not alter the brine chemistry by changing the ionic makeup. Some operators are using large fixed-site treatment operations. Boschee (2012) described a fixedsite system in use in the Pinedale Anticline in Wisconsin, USA, a Bakken play that shows a coagulant and flocculation tanks with a clarifier (see Fig. 2.32) next to a collection lagoon with aerobic (aeration) treatments. Portable operations are also in use. Halliburton (2011) introduced at portable water treatment/recycle system called CleanWaveTM. Specifics of the system are that it • Coagulates suspended matter using electrical charge (electrcoagulation precipitation) • Is a containerized unit with 100-kVA generator • Is scalable, with ability to handle total flow in real time • Reduces total suspended solids 99% • Coagulates particles 0.50 in. if pipe diameter 2% (or >0.50 in. if diameter 6% (or >0.5 in. if pipe diameter 6% (or >0.50 in. if pipe diameter % (or > 25 in. if pipe diameter 1.0. As discussed later in this section, SR is a measure of the degree of supersaturation and can affect the rate of scaling. Much of the art and science of prediction of scaling involves calculation of the real SR in complex brines. The log(SR) is SI, the saturation index. This is a more convenient designation because it uses

164  Chemical and Mechanical Methods for Pipeline Integrity

Mineral Name

Common Name

Chemical Formula

Acmite

Sodium iron silicate

NaFe(SiO3)2

Analcite

Sodium aluminum silicate

Anhydrite

Calcium sulfate

CaSO4

Aragonite

Calcium carbonate-rhombic

CaCO3

Barite

Barium sulfate

BaSO4

Brucite

Magnesium hydroxide

Calcite

Calcium carbonate-hexagonal

NaAlSi2O6 · H2O

Mg(OH)2 CaCO3

Chalcocite

Copper sulfide

Chalcopyrite

Copper iron sulfide

CuFeS2

Cu2S

Chromite

Iron chromium spinels

CrFe2O4

Copper

Copper

Covellite

Copper sulfide

Cuprite

Copper oxide

Gibbsite

Aluminum hydroxide

Cu CuS Cu2O Al(OH)3

Goethite

Hydrous ferric oxide

Gypsum

Calcium sulfate dehydrate

FeOOH

Halite

Salt

NaCl

Hematite

Ferric oxide

Fe2O3

CaSO4 · 2H2O

Hydromagnesite

Magnesium carbonate and hydroxide

Hydroxyapatite

Calcium phosphate

3MgCO3 · Mg(OH)2 · 3H2O

Mackinawite

Iron sulfide

Magnesia

Magnesium oxide

MgO

Magnetite

Ferric-ferrous oxide

Fe3O4

Magnesite

Magnesium carbonate

Montmorillonite

Aluminum silicate (one type of a clay)

Noselite

Sodium aluminum silicate

Pyrite

Iron sulfide

FeS2

Pyrolusite

Manganese dioxide

MnO2

Pyrrhotite

Iron sulfide

FeS2

Serpentine

Magnesium silicate

Siderite

Iron carbonate

Silica

Quartz

Silica, amorphous

Hydrated silica

Sodalite

Sodium aluminum silicate

Wüstite

Ferrous oxide

Ca10(OH)2(PO4)6 FexSy

MgCO3 Al2O3 · 4SiO2 · 4H2O Na8Si2O7O24 · H2O

Mg3Si2O7 · 2H2O FeCO3 SiO2 SiO2 · 2H2O Na8Al6Si6O24 · 2Cl FeO

Table 4.1—Summary of common inorganic-scale-forming compounds.

small numbers. Various correlations are available in the literature for estimating SI in terms of easily measurable quantities such as pH, alkalinity (Alk), and concentration of dissolved ions and ionic strength of the fluid. For example, the saturation index from Stiff and Davis (1952) is calculated as SI = pH − k − pCa 2+ − pAlk ������������������������������������������������������������������������������������������������������ (4.2)

Chemistry of Product Flow Impairment in Pipelines and Facilities  165

In Eq. 4.2, which applies only to the supersaturation of calcium carbonate, Ca2+ is the concentration of the dissolved calcium and Alk is the measured alkalinity (a titration), which is a general measure of the bicarbonate (HCO3–) concentration. Many other empirical saturation indices have been used in various industries to predict scaling conditions. See, for example, Davies and Scott (2006) and Frenier and Ziauddin (2008). Under the best of conditions, scaling and corrosion can be controlled with minimal use of chemicals if pH [Ca2+] and p[CO2] are tightly controlled. This approach is very difficult to achieve in practice, so various scale and corrosion inhibitors are also used (see Chapter 6). Carbonate scales form in the near-wellbore region; in fractures, sandpacks, screens, and downhole equipment; in tubulars and pipelines; and on surface equipment. The mixing of incompatible waters, such as seawater and formation water, when the solubility product of a salt is exceeded in a local environment can causes sulfate scales. These are found in most of the same environments as carbonates, though usually not at the same time, but are not seen frequently in surface equipment. This type of scale can form in the matrix, perforations, screens, tubing, connection lines, pipelines, and facilities. Another example is fouling of injection wells from placement of seawater into an incompatible formation. Barium sulfate scale, formed from mixing incompatible water, is a particularly important scale because it is extremely difficult to redissolve (Frenier and Ziauddin 2008). If these scales are present in pipelines, clogging will occur. In addition, slightly radioactive deposits, called “NORM,” can form with the sulfates and contaminate equipment, including pigs. Crabtree et al. (1999) proposed a mechanism of formation of the most common types of inorganic scale, including calcium carbonate and alkaline earth sulfate deposits. These are categorized as “salt formers.” However, the corrosion product scales (i.e., siderite hydrated iron oxides and the iron sulfides) are subject to exactly the same types of chemical equilibrium conditions. The only difference is that the cations (Fe2+ and Fe3+) can originate from the construction metals, not directly from the water phases (soluble Fe can also be found naturally in some formations). Although the driving force for scale formation may be a temperature or pressure change, out-gasing, a pH shift, or contact with incompatible water, many oilfield waters that have become oversaturated and scale prone do not always produce scale on surfaces. For a scale to form, it must grow in solution. The first development within a saturated fluid is a formation of unstable clusters of atoms, a process called “homogeneous nucleation.” Fig. 4.4a, from Crabtree et al. (1999), depicts this process from supersaturation, to formation of ion pairs, to the final deposition of a crystal. The final step of the process is deposition of scale crystals on a surface. This can occur coincidental to nucleation and crystal growth or subsequent to the first two steps. Fig. 4.4b (Jacoby 2014) and Nielsen et al. (2014) illustrate a micrographic view of a flowing solution of supersaturated solution of CaCO3 that catches the transition from the amorphous state (red line) to the first incipient crystals (purple line). These authors used state-of-the-art technologies, including a transmission electron microscope (TEM) to describe the details of crystal formation. Fig. 4.5 shows microscopic views of some incipient BaSO4 scale crystals. Deposition onto a surface is frequently the most damaging process for flow assurance, but mobile solids can also plug the formation as well as the filters and may affect other processes, such as the stability of emulsions (see Section 4.6.). The rate at which scales form is very important and is usually described in terms of induction time (tind) after a supersaturated condition exists but before the appearance of scale. It is assumed that no scale will form in the system if the induction time is less than the residence time of the fluid in the supersaturated condition. Although induction time models do not predict when the scale buildup starts to affect operation of the system, they are critical starting points for that type of estimation. Assuming that the nucleation time is much greater than the time required for crystal nuclei to grow to a detectable size, the induction period can be expressed in terms of the primary nucleation rate, J, as 1 t ind = .�������������������������������������������������������������������������������������������������������������������������������������� (4.3) J

166  Chemical and Mechanical Methods for Pipeline Integrity

Homogeneous nucleation Supersaturation

(a) Ion pairs

Clusters/nuclei Transient stability Imperfect crystalites

Ba+2

Further growth at sites of crystal imperfections

SO4–2 Heterogeneous nuleation

2 µm

(b) TEM of Amorphous CaCO3 Forming Aragonite

Supersaturation Ion pairs

Surface imperfections

Fluid flow

Pipe well

Fig. 4.4—Formation mechanisms of inorganic scale (Crabtree et al. 1999; B: Jacoby 2014).

Fig. 4.5—Barium sulfate crystals (Sorbie and Laing 2003).

Chemistry of Product Flow Impairment in Pipelines and Facilities  167

The primary nucleation rate is given by   γ 3υ 2 J = A exp  −φβ  3 2 .���������������������������������������������������������������������������������������������� (4.4)  ( kBT ) ( ln SR )  Here, A is the frequency factor, f is the energy barrier factor (f = 1 for homogeneous nucleation and f < 1 for heterogeneous nucleation), b is a shape factor, g is a crystal surface energy, u is the molecular volume of the crystalline phase, kB is the Boltzman constant, T is the absolute temperature, and SR is the supersaturation ratio. See Tantayakom et al. (2005). Combining Eq. 4.3 with Eq. 4.4 gives ln t ind = φβ

γ 3υ 2

( kBT ) ( ln SR ) 3

2

1 + .������������������������������������������������������������������������������������������������ (4.5) A

Eqs. 4.3 and 4.5 and Fig. 4.4 can be applied to elucidate the nature of the nucleation mechanism. As can be seen in Fig. 4.6, a change in slope of the ln(tind) as a function of (ln SR)–2 reveals a transition from homogeneous to heterogeneous nucleation. Moreover, the surface energy can be determined by extracting slope values. Models of nucleation kinetics are also described by Tomson et al. (2004) and are generally grounded on the theory of nucleation by Mullin (1997). They are characterized by the logarithm of the induction time being proportional to the nuclei surface tension divided by the saturation index squared, to various exponents. For example, they have found that the measured induction time for barite, in the absence of inhibitor (t0ind.), over approximately five orders of magnitude can be described by the following function: 0 Log(t ind. ,sec) =

2.2 + 1, 087 − .3T (o K )  T 2 SI

2

−

0.12 + 1, 087 − .3T (o K )  T 3 SI 2

3

.���������������������������������� (4.6)

ln(tind)

The term in brackets in Eq. 4.6 is related to the surface tension function from classical nucleation theory, and the calculated interfacial surface tension is close to that reported for barite, and correspondingly for calcite. Several other researchers have attempted to predict the effect of inhibitors on mineral scale formation. They note that one of the most successful theories of precipitation kinetics is the Burton-CaberraraFrank (BCF) spiral-growth Heterogeneous mechanism. In this theory, kink nucleation sites grow continuously in a spiral, thereby overcoming the Homogeneous expected activation energy for nucleation nucleation of a new site on a flat γ 3υ 2 slope = φβ particle surface. A corollary of 3 kBT this is that inhibition is treated as an adsorbed inhibitor blocking the advance of a spiral. Only the active growth sites need be poisoned to prevent the spiral from continuing to grow. Specific (lnSR)–2 examples are in subsections of Section 4.2. Fig. 4.6—Nucleation model (Tantayakom et al. 2005).

168  Chemical and Mechanical Methods for Pipeline Integrity

Clear liquid Surface roughness

Flow direction

Solid

(a) Stage 1

Surface roughness

Flow direction

Liquid with scale (CaCO2) molecules

Solid

(b) Stage 2 Scale crystals Flow direction Solid

(c) Stage 3

Flow direction

Solid

(d) Stage 4 Flow direction

Solid

(e) Stage 5 Fig. 4.7—Scale deposition in a flowing pipe (Hamid et al. 2013).

Actual scale adhesion and accumulation in a pipeline can be affected by flow dynamics. A study by Hamid et al. (2013) claims that a five-step process controls the amount of deposition in a flowing pipe. In Fig. 4.7a the flow in a clear but unsaturated solution over a rough surface is shown. In Fig. 4.7b, the solution is saturated with a salt and microscale crystals start to form (see also Fig. 4.4). In the remaining stages, scale will form at the boundary area if the flow is above a minimum level and will continue to accumulate until the flow is above this minimum. This is because of erosion of the scale particles. Specific important inorganic scales that foul pipelines and facilities are described in the next subsections. Alkaline Carbonate Scales. Alkaline and alkaline earth carbonate scale formation is controlled by the most complex chemistry of all the minerals commonly encountered in the oilfield. To predict carbonate precipitation, the chemistry of three reactions must be understood. The chemical sequence is threefold: 1. Carbon dioxide dissolves in water to form carbonic acid. 2. Carbonic acid dissociates to carbonate and bicarbonate, which lowers the pH. 3. Carbonate ions interact with alkaline/alkaline earth ions (Na+, Ca2+, Mg2+).

An example for the calcite formation reaction is Ca 2+ + 2 HCO3− = CaCO3

+ H 2 O + CO 2 .���������������������������������������������������������������� (4.7)

All the chemical reactions are linked such that modifying any one factor (e.g., pH or partial pressure of CO2) has an effect on the others. An important contribution to scale formation is that reducing partial pressure of CO2 causes precipitation of calcium carbonate. This is called “flashing” or “auto scaling.”

Chemistry of Product Flow Impairment in Pipelines and Facilities  169

Fig. 4.8 shows a diagram Distance in line that describes auto scaling p (also “flash scaling”), which Precipitation produces dro re u is caused by a reduction of further pressure drop s es r P leading to further acidity as the acid gas CO2 precipitation. leaves the solution. The reduction of pH causes more Ca2+ + CO23− = CaCO3(S) bicarbonate to form and calCO2 + H2O = H2CO3 cium carbonate to precipitate. H2CO3 = H+ + HCO−3 pKa = 3.6 Note, as listed in Table 4.2, HCO−3 = H+ + CO32− pKa = 10.3 that the general solubilities of Ca(HCO ) = CaCO + H2O + CO2 3 2 3(S) the carbonates is Mg2+ > Ca2+, but the solubility of a given Pressure drop causes CO2 to break out of solution, salt is greatly dependent on causing precipitation of calcium carbonate. p[CO2] and pH. Also, these units are moles per 1000 g of Fig. 4.8—Auto (flash) scaling of calcium carbonate. water. Unfortunately, there is no standard way of recording solubilities, and so mixed units and differing experimental conditions are found in many compilations. Flash scaling also appears in natural environments. Fig. 4.9 is a photograph of a part of Mammoth Hot Springs in Yellowstone National Park, Wyoming, USA. The geothermic processes in the park dissolve large amounts of calcium in the presence of CO2. As the brine reaches the surface, the pressure is released and a calcite-containing mineral (travertine) precipitates in vast quantities. Carbonate scaling can occur by incompatible mixing of waters as well as by flashing formation, or both simultaneously. Almost absolute accuracy in water analysis is required for prediction of scaling because of the critical dependence of this reaction on pH. Calculation errors in pH, which is a function of all solution species, and the pressure of CO2, can give orders-of-magnitude errors in saturation factors for carbonates. Physical samples of scale are often required to confirm scale predictions from water analyses. Sulfate Scales. Because the basic chemical reactions for the precipitation of alkaline earth sulfates is much less complex than carbonate formation, the literature on formation of these salts has concentrated on the morphology, adherence to surfaces, and rate at which crystallization is initiated. Three different calcium sulfates occur, each with different amounts of hydration water: • Anhydrite (CaSO4) • Hemihydrate (CaSO4 · 1/2H2O) • Gypsum (CaSO4 · 2H2O). The other sulfates, barite (BaSO4) and celestine (SrSO4), are not hydrated. Magnesium forms complex crystals with sulfate, but these have high solubility and do not usually present problems. M2+

Radius (10–8 cm)

Fluorides

Hydroxides

Mg

0.31

V.S

1.2 × 10

1.2 × 10

Ca

0.65

2 × 10–4

2.1 × 10–2

1.5 × 10–4

1.5 × 10–2

Sr

1.13

1 × 10–3

6.5 × 10–2

7.0 × 10–4

5.0 × 10–3

Ba

1.35

1 × 10–2

2.8 × 10–1

1.0 × 10–4

1.0 × 10–5

–11

Moles per 1000 g of solid in water at 1 atmosphere and saturated with CO2 for carbonates

Table 4.2—Solubility of alkaline earth salts (Becker 1998b).

Carbonates –3

Sulfates 2.4

170  Chemical and Mechanical Methods for Pipeline Integrity

Fig. 4.9—Calcium carbonate deposits at Mammoth Hot Springs, Yellowstone National Park.

Strontium Sulfate solubility, mol/L

5 × 10–3

25°C

4 × 10–3

150°C

3 × 10–3 2 × 10–3

250°C

1 × 10–3 0

1

2

3 4 NaCI mol/L

Seawater

5

6

Fig. 4.10—Mineral solubility (molality) vs. salinity. T (°C) 0 –2

25

50

75

100

125

150

Melanterite (FeSO4·7H2O)

log10Ksp

–4 –6 –8

Gypsum (CaSO4·2H2O)

Anhydrite (CaSO4)

Celestite (SrSO4) Barite (BaSO4)

–10

RaSO4

–12 Fig. 4.11—Solubility of sulfate minerals as a function of temperature.

Radium is also an alkaline earth metal, exists naturally, and can substitute for barium in barite and cause the deposit to become radioactive. Additional details on various crystallographic structures are found in the Mineralogy Database (http://www.webmineral.com). The composition, hydration, and morphology can become critical when remediation methods must be implemented because these factors can greatly affect the rate of dissolution or physical removal. Calcium, barium, and strontium sulfate scale formation is usually the result of mixing incompatible waters (e.g., seawater with high sulfate content and formation water with high barium content). As this water continues to be produced up the wellbore or in a pipeline, the pressure and temperature fall and additional scale will form. Solubilities of the various salts as a function of temperature, pressure, and salt content are a highly complex subject (see Table 4.2, which shows the solubilities of various salts). Kan et al. (2005) discussed various models. Note the changes with respect to salinity and temperature of SrSO4 in that it is more soluble in some higher concentrations of brines (see Fig. 4.10). Fig. 4.11 shows a plot of sulfate mineral solubilities (ksp values from Langmuir 1997) for barite, gypsum, and anhydrite as a function of temperature. There are very large differences already noted for the sulfates. We see that gypsum is unstable at

Chemistry of Product Flow Impairment in Pipelines and Facilities  171

temperatures >50°C and that anhydrite is unstable at temperatures below that value. Because gypsum is hydrated, it is more easily dissolved compared with anhydrite. The solubilities of theses salts in water are a major determination of scaling. Becker (1998b) related the solubilities of a range of alkaline earth metals to crystal structure. He proposed that the coordination numbers of the central metal are not altered by the change of the anion from carbonate to sulfate. However, the metallic salt resulting from the neutralization of charge by the larger and stronger sulfate anion causes a greater distortion of the coordinated complex. Thus, the crystal forms of the multiple combinations of complexed salts present different morphologies when the individual unit cells are deformed. His theory is that the deformation of the complex also reduces the ability to accept water at the open ligand sites. This inability to accept coordination water changes the solubility characteristics of the salt complex. A general trend of increasing water solubility is observed with increasing coordination by water at the available ligand sites of the metallic salts. Therefore, decreases in the number of available ligand sites result in lower water solubility. Coordination numbers (i.e., ligand sites) assumed by the central metal depend on the availability of energy levels possessed by the metal. They are determined by the physical conditions to which the metal cation is exposed. Becker (1998b) noted that the highest coordination number is determined by the highest quantum state reached by the available metallic orbitals. Electronic effects exerted by ionic ligand groups also affect changes in the nature of the complex and the coordination number assumed by the metallic cation. Comparison of the alkaline earth metal salt complexes found in some crude oil systems indicates that salts of large ions follow trends expected if changes in hydration energy dominate. Thus, the solubility trends show the following order: BaSO4, 1% d/D

EMAT

Transverse MFL

High resolution axial MFL

Specific threat

Wrinkle bends

Dents, wrinkles, buckels, & ripples

Scratch/ gouges

Pipe Deformation Excavation damage

Earth movement (strain)

Mechanical Methods of Assessment and Enhancement of Pipeline Operations   237

Waker et al. (2004) summarized general selection criteria for ILI tools. They noted that technological advances are constantly occurring in the pipeline inspection industry and that it is important to keep up to date with these advancements. Their suggestions are that before selecting an ILI tool, it is important to decide • • • •

Objectives for performing pipeline inspections Types of defect to be detected Needed level of accuracy and reliability of the data set Technology that is available at the time of selection

A general selection flow chart from Waker et al. (2004) is shown as Fig. 5.41. ILI tools can be combined into an almost limitless number of “trains” of tools to achieve multiple functions (see the discussion by Willems et al. 2010). A report from TDW (2010) shows an example of the use of mapping, spiral MFL, axial MFL, and deformation/geometry measurements along with the output signals while scanning the same area of a pipeline. See Fig. 5.42, that also shows the types of output files that result from each tool. It should be clear from this figure that the analysis of a large number of files as the tools transit the pipeline is a significant task. Pirtle (2013) claimed that running a number of different tools provides an overlapping data set that allows detection of defects that are extremely difficult to quantify with a single tool. Examples are • Dents with volumetric metal loss • Metal loss in seamless piping • Metal loss crossing girth welds Pipeline Data

Pipeline History

Operating Conditions

Trap dimensions Outer diameter WT Bends Valves Coating type/ quality

Product Flow MOP

Primarily inspecting for Metal loss?

Key to Acronyms

Previous ILI/Type Defects: severity, type, number CP Known Corrosion Leaks Planning ILI

N

WT-Wyes and Tees MOP-Maximum Operating Pressure CP-Cathodic Protection ILI-Inline Inspection MFL-Magnetic Flux Leakage

Are pipeline geometry Measurement needed?

Is feature orientation required?

Y Is this a high-risk line?

Y N

N Is this a high-priority line?

N N

Outer/inner diameter differentiation?

Medium -resolution MFL tool

Y Y

High-resolution metal-loss tool

Proceed with ILI run

Y

Y

N

High-resolution crack detection tool

N

Can tool be run in the current operating conditions and pipeline configurations?

Y

Caliper multichannel

Can the facilities or line be modified? Performance Modifications

Y

Fig. 5.41—Selection flow chart for ILI (Waker et al. 2004).

N

Caliper single channel

Consult advisor

238  Chemical and Mechanical Methods for Pipeline Integrity

Residual & battery

Axial MFL

Def & IDOD

Spiral MFL

ODO & Drive

Key to Signals RES - residual magnetic field DEF - deformation and cross section SMFL - spiral MFL MFL - axial MFL IDOD - interior and outside diameter Fig. 5.42—A train of ILI tools and output signals (TDW 2010).

5.4.9 Analyses of In-Line Inspection Data and Tool Performance. This subsection describes the processes used to evaluate ILI data as well as actual tool performance, compared with a physical examination of a pipe defect. This process provides the data for making decisions about pipeline integrity. The tools described in Sections 5.4.2 through 5.4.7 make innumerable measurements that may, according to Taberner et al. (2002), detect hundreds, possibly thousands, of anomalies that must be assessed to determine their danger to the pipeline. Depending on the tool type, the devices characterize the anomalies by their measured depth, length, and circumferential extent with varying degrees of accuracy. The data are processed to provide several different views of the anomalies, including real-time scans with the pulse echo (A-Scan), side view (B-Scan), and top view (C-Scan). These readings are recorded along with the measurement uncertainty of the inspection tool itself. Significant anomalies may be verified by actually examining the pipe surfaces. This frequently requires that the pipe be dug up, which is a costly process. The goal of ILI examinations is to evaluate defects using the probe data and to reduce the number of actual surface examinations. Willems et al. (2010) performed side-by-side tests of MFL, EC, and EMAT tools. These authors produced artificial defects (internal, external, and of different types) on a pipeline section and then scanned them with a single tool that had the three components in a train (similar to Fig. 5.42). Fig. 5.43a, from Willems et al. (2010), shows a photograph of the pipe with some of the external drilled defects; internal and welded defects not shown. Presenting the pattern and length from a top view, Fig. 5.43b shows a comparison of C-Scans of the internal and external anomalies detected. Note that the different scans gave different resolutions of some of the same defect types. The authors claimed that depth can be measured directly from the EMAT scan, which shows depth (not shown in the figure), yielding the correct value of 4 mm. They also claimed that as expected for external metal

Mechanical Methods of Assessment and Enhancement of Pipeline Operations   239

(a) Test pipe and external defects

External defects

(b) Comparison of C-scan presentations Internal

EMAT

External Internal

MFL

External

Internal

EC

External Fig. 5.43—Test pipe and comparisons of scans of defects, using three tools (Willems et al. 2010).

loss, almost no EC signal is produced. The MFL signal shows the radial component of the strayflux field. It exhibits a very sharp gradient at the edges of the defect, thus allowing for an extremely precise length measurement. The conclusion is that all three tools could be used to give the most accurate view of the line and its defects. Note that different proprietary software packages currently in use present and manipulate the data to allow interpretation. Taberner et al. (2002) discussed how data obtained from various ILI tools are compared with acceptable defect criteria to determine the “criticality” of the defect. There are various methods available, and defects that fail an assessment against the criteria may need to be excavated for verification and possible repair. Described in their paper are methods of analysis that may be used with data from a tool: • ASME B31G-1991 (ASME 1991). A modification is known as “Modified B31G” (Kiefner and Vieth 1989). • RSTRENG® refers to “remaining strength” (Kiefner and Vieth 1989); this is also known as “effective area method” (see Kiefner 2006). • DNV-RP-F101 (DNV 2010). • API 579-1/ASME FFS-1 (2007). Once acceptability criteria (i.e., depth and length) of a defect are set, these methods provide statistical analytical tools for determining which anomalies require direct inspection. Cost constraints as well

240  Chemical and Mechanical Methods for Pipeline Integrity

as hazard classes (see Table 1.1) may also affect the analyses depending on the local regulations (see Section 3.5). t d A report by Kiefner (2006) reviewed and compared Exact area ASME B31G-1991, Modified 31G, and RSTRENG methL ods for assessing the remaining strength of a pipeline that t d had requested a waiver from the Pipeline and Hazardous Parabolic area Materials Safety Administration (PHMSA) to increase (a) A=2/3 Ld (B31G) pressure regulatory limits. The author called these processes the “stranded methods” for assessing pipeline strength after L corrosion damage has been detected. The report notes that d the B31G method consults tables of acceptable length dimensions of corroded areas depending on the maximum B31G- A=2/3Ld measured depth and the dimensions of the pipe, along with (b) Mod B31G- A=0.85Ld optional calculations for “safe operating pressure” considFig. 5.44—Area of defect strength ering the measured length and depth of the corroded area. methods (Kiefner 2006). A diagram from this author illustrates the differences in methods (see Fig. 5.44). On the basis of their analyses, they consider Modified B31G to be less conservative than B31G and more conservative than RSTRENG. Anderson (2011) noted that another of these methods (API 579-1/ASME FFS-1 2007) describes three levels of evaluation of ILI data: L

A

• Level 1. This is a basic assessment that can be performed by properly trained inspectors or plant engineers. A Level 1 assessment may involve simple hand calculations. • Level 2. This is more complex than Level 1 and should be performed only by engineers trained in the API 579-1/ASME FFS-1 standard. The author claimed that most Level 2 calculations can be performed with a spreadsheet containing proper equations from this API standard. • Level 3. This is the most advanced assessment level, which should be performed only by engineers with a high level of expertise and experience. The author asserted that Level 3 assessments may include computer simulation, such as finite element analysis (see Wikipedia 2014c) or computational fluid dynamics (see Wikipedia 2014a). Anderson (2011) and Kiefner (2006) claimed that the three assessment levels represent a tradeoff between simplicity and accuracy. They asserted that simplified assessment procedures are necessarily more conservative than more-sophisticated engineering analyses. To aid in the more-complex assessments, Anderson (2011) then claimed to have developed software that can rapidly process large quantities of ILI wall-loss data and evaluate the MAOP at discrete locations. They contended that the ranking of these MAOP values serves as a rational and rapid means for prioritizing the severity of corrosion throughout the line. In this paper, the data come from UT measurements. They claimed also that their software—which is not described in detail—will assess metal loss; the remaining strength factor is computed, which can be used to compute a MAOP. The summary of remaining strength methods were described by Kiefner (2006) in a chart that shows the complexity of the methods vs. the conservative nature of the prediction (see Fig. 5.45). Note that the explanations are in Anderson (2011) and Kiefner (2006). Thus, the operator must balance the difficulty with the ability to reduce the necessity of digging more piping for direct inspections. A standard for evaluation of the performance of a tool used in ILI inspections is provided in API STD 1163 (2005). Several discussions of this standard are noted because the performance of a tool will affect the process for deciding if an anomaly must be physically examined. Desjardins (2005) stated that using this API standard, ILI tool performance verification is performed on the basis of four principle measurements vs. actual (or simulated) defects.

Mechanical Methods of Assessment and Enhancement of Pipeline Operations   241

Simple

Complexity

Difficult

He noted that one assessment is the FEA API 579 L3 POD, which is the measure of the tool’s ability to detect an anomaly of a given size and type. In addition, the POI is a measure of the tool’s ability to properly identify the anomaly type: corrosion, inclusion, or crack. Effective Area RSTRENG Also, the sizing accuracy is the meaAPI 579 L2 sure of the tool’s ability to correctly report the size or severity of a given anomaly. The tool must also be able Modified B31G to report the location and position of ASME API 579 L1 anomalies so that corrective action B31G can be taken if required. The author assessed these percentage values and Less conservative More conservative More exact Less exact variation range (+/−) by comparing Conservatism ILI data to field measurements when a pipe is exposed and the anomalies Fig. 5.45—Spectrum of assessment complexity vs. are measured. He claimed that a new conservatism (Kiefner 2006). tool can be evaluated by comparing the results with a previous ILI run that has been verified. Table 5.7, adapted from API STD 1163 (2005), gives an example of the data about a tool’s performance that the tool provider should provide to the pipeline operator. Examples of other detection criteria are also in API STD1163 (2005). McCann et al. (2007) provided a statistical analytical method for evaluating a tool’s performance on the basis of API STD 1163 (2005). The general acceptance or rejection criteria include levels of • Tolerance (e.g., within ±10% of wall thickness) • Certainty (i.e., probability), such as 80% of time a reported anomaly depth is within a specified tolerance • Confidence, indicating the confidence with which the tolerance and certainty levels are satisfied (e.g., 95%) The authors proposed using a binomial distribution function for determining if the tool’s detection properties are acceptable or are rejected. A summary of the accuracy/precision of assessing the defects includes • Detection of an anomaly (% POD) at x% confidence • Identification of type of anomaly (% POI) at x% confidence • Accuracy of the depth/size/ shape (%) at x% confidence Depth Detection • Location in the line of the defect Threshold POD (%) at x% confidence Tomar and Fingerhut (2006) directly addressed the techniques used for comparing field-measured defects with ILI data. These authors claimed that manually acquired field data using hand tools have two

Qualifiers and Limitations

10%t

90%

Extended metal loss length and width >3t

15%t

90%

Pits, t1mpy?

Have liquids been recovered from the line?

Are there offsets, dead legs or low spots?

Sample and test pipeline liquids

yes

no Does Water exist?

no

yes

Threat of IC exists

yes

no

no

no Remove dead legs not needed.

yes

Has any NDT shown IC?

Will pipeline transport high levels of CO2 or H2S?

Is the pipeline piggable?

yes

no

no

Periodically analyze results to update strategy for minimizing IC.

yes

Has ILI shown IC?

Monitor pipeline liquid removal volume.

yes

Is the pipeline piggable? no

yes Periodically analyze results to update strategy for minimizing IC.

Are there offsets, dead legs or low spots? no Monitor gas quality.

Fig. 5.48—Internal corrosion decision tree (Abayarathna et al. 2015).

Fig. 5.49—Improvements in line piggability by ILI (Steinvoorte 2013).

Remove dead legs not needed. Calculate flow velocities, critical angles and sweep velocities..

Mechanical Methods of Assessment and Enhancement of Pipeline Operations   245

Barbian and Beller (2012) reviewed current technologies and projected future trends: • There will be a trend to combine inspection tools.  So-called combo tools will make use of different technologies to perform different tasks at the same time. An example is combining geometry and metal-loss inspection. See Fig. 5.41 for a representation of multiple ILI tools and output signals (TDW 2010).  Multitechnology tools will use different technologies to enhance and optimize one specific inspection task. See the discussion in Willems et al. (2010). • As electronics become more powerful, combined tools and multitechnology tools may grow together, thereby forming a new generation of tools that will offer combined inspections with optimized tool performance and defect specification. 5.5  In-Line Handling Equipment and Other Devices Additional mechanical equipment, including traps, launchers, and jetting processes that use coiled tubing are described in this section. The items discussed are critical to the success of mechanical pigging and the use of ILI tools. Chemical injection equipment is reviewed in Section 6.7. 5.5.1  Traps (Receivers) and Launchers. Warriner (2008) noted that pig “traps” (i.e., receivers) provide a means of loading/unloading pipeline tools, including intelligent pigs, into pipelines. The traps take the form of a vessel that can use diverted flow for either launching or receiving a unit. Thus, the vessel can be isolated from the pipeline to facilitate the loading/unloading of pipeline tools. Because when in use the launcher is essentially part of the pipeline, it is important that it be capable of withstanding pipeline conditions such as pressure, temperature, and effects of the service medium (i.e., corrosion). These devices are thus pressure vessels that allow isolation from the pipeline for injection and retrieval of the pig. Warriner (2008), TDW (2005b), and Morrow (2011) claimed that various types of equipment are needed to handle the pig operation in addition to launchers and receivers. Some items are seen in Fig. 5.50, which illustrate retrievers with sludge (Fig. 5.50a), a badly fouled pig (Fig. 5.50b), and a handling tray (Fig. 5.50c). These types of ancillary equipment will be sized for the equipment being launched or retrieved. Large-diameter and long-length ILI tools may require heavy-lifting devices such as a crane (Fig. 5.51).

(a) Pig retrieval with sludge in a tray

(b) Badly fouled pig

(c) Pig-handling tray

Fig. 5.50—Pig-handling equipment (a, c) and fouled pig (b) (TDW 2005b).

246  Chemical and Mechanical Methods for Pipeline Integrity

(a) Launcher system

(b) Receiver system

Fig. 5.51—Typical launching and receiving setup (TDW 2011).

Typical setups of valves and pressure equipment for launching/ retrieving devices are found in Fig. 5.51, which shows the launch setup (Fig.5.51a) and the receiving setup (Fig. 5.51b). Nonspecific instructions for safe launching /retrieving of pigs are in TDW (2011), but each piece of equipment will have different controls for safe operations. Note well the general safety discussion in Section 8.2 because dealing with high pressures as well as flammable chemicals requires operators to be well-trained and adhere to equipment procedures. Details of a retriever are shown in Fig. 5.52, with several ports for relieving internal pressure before opening the door to remove the pig and the deposits removed from the line. Drawings of bypass devices as well as automated and multipledevice-launching equipment are found in Sections 7.1 and 7.2. Additional considerations noted by Morrow (2011) and Warriner (2008) are these:

Pig signal

Mainline trap valve

Vent Valve

Vent Valve

Pressure gauge

Closure

Drain Mainline bypass valve

Equalization valve

Return line valve

Fig. 5.52—Details of a pig retriever device (TDW 2005b).

Mechanical Methods of Assessment and Enhancement of Pipeline Operations   247

• Mechanical design. The units must be mechanically designed to meet the pipeline section design specifications.  Design pressure, design code.  Design factor, design temperature.  Appropriate materials. • Size. The devices must be dimensionally suited for the type of pigging that is expected on the pipeline section. • Location. The units should be situated away from ignition sources, and there should be sufficient space for loading/unloading. • Layout. Layout should allow accessibility and orientation for safe operations. • Environment. Drip trays and other setups are required to prevent environmental contaminations. • Logistics. Lifting and other facilities must be available for forces to operate valves and the like. Fig. 5.50 illustrates an important problem, which is the sludge (or pig trash) that comes out with the pig, a very dirty device, and is also an object of treatment. Many units will be reused, and so the sludge requires disposal (Section 8.4). For reuse, the pig must be thoroughly cleaned (see image of a fouled pig in Fig. 5.50b). The ILI devices and especially the UT devices can be especially sensitive to debris on the sensors. TDW (2005a) suggests cleaning pigs or spheres as soon as possible after removal from the pipeline. The document notes that this is especially important when the pig or sphere has been used in a line containing a high percentage of hydrogen sulfide. The document also describe methods of removing various pipeline residues from pigs. A pig washer (see Fig. 5 in TDW 2005a) is claimed to be able to clean pigs automatically without solvents. Warriner (2008) described different categories of automated handling equipment. Included are • • • •

Bidirectional devices that can launch/receive a pig Vertical/horizontal launchers (Sections 7.1 and 7.2) Vertical/horizontal retrievers Multiple-pig launchers (horizontal as well as vertical), including ball launchers (more details in Section 7.2)

5.5.2  Pig Indicators and Tracking Devices. Pig passage indicators are devices that are mechanically or electronically tripped when a pig passes a particular point in the line. They can produce a “flag” or make an electronic indication that can be transmitted to a central station. Common indicators have a trigger extending into the pipeline through a tapped welded fitting. When a pig passes, the trigger is tripped, activating the signaling mechanism. A drawing from Ver-Nooy (1963), Fig. 5.53, shows a mechanical tripper in the received position (Fig. 5.53a) and in the pipeline mode (Fig. 5.53b). Pig trackers are devices that are used to track the passage and pinpoint locate a pig that is equipped with an appropriate transmitter. These devices enable operating and maintenance companies to minimize the costs involved with excavating and removing stuck pigs. TDW (2005b) described a magnetic device, or transmitter, that can be towed behind a pig or incorporated into it. The signal of the movement is detected. Receivers with an antenna are placed on/in the ground above the pipeline and detect the signal of passage and store data for analyses (see Fig. 5.54). Some receivers can send a message to the operator that the pig has passed a specific location. An aboveground marker detects the magnetic field of an MFL device and other signals to mark and record the passage of a device. Other types of trackers detect pig movement by sound amplification of the noises created by the pigs moving through a pipeline. These noises occur when the pigs traverse circumferential weld penetration at weld seams (TDW 2005b), along with the normal scraping of the cups and cleaning elements against the pipe wall.

1

2

3

4

Key 1 Tapping machine 2 Spring-loaded balls 3 Tapping valve 4 Plug assembly 5 Trigger element 6 Housing 7 Pipeline 8 Pipe opening 9 Pivoting flanges

1

3

2

5

4 6

6

7

7

8

8 (a) Tripper recessed

9

5

(b) Tripper extended

Fig. 5.53—Pig tripper element (Ver-Nooy 1963).

Receiver

Transmitter

Antenna

Aboveground marker Fig. 5.54—Magnetic pig tracker (TDW 2005b).

Mechanical Methods of Assessment and Enhancement of Pipeline Operations   249

Very sensitive geophones are placed in the ground or on a pipe component to hear the noises of movement. Photographs of different types of location transmitters, various tool location and tracking systems, and an electronic tool detector are shown in Rosen (2014).

Injector 5.5.3 Coiled Tubing for Conveying head Tools Into Pipelines. On the surface, hand jetting or automated lances can be used to affect deposit removal when a unit is facing completely blocked lines, but downhole or in some pipelines, the tool will require some type of conveyance to the trouble location. Details of the use of high-pressure water to clean Lubricator heat exchangers are provided in Section 7.6.2. This current short section describes the use of coiled tubing to place different tools into pipelines. The tubing is used frequently to place various jetting tools Shear seal/ into the wellbore or some sections of a Combi BOP pipeline to clean a section of the line. The coiled tubing is a continuous length of pipe wound on a spool. The Quad BOP pipe is straightened before being pushed into a unit and then rewound to coil the pipe back onto the transport and storage spool. Fig. 5.55 shows a tool that places the tubing into the well. Depending on Wellhead the pipe diameter (1 to 4.5 in.) and the spool size (Fig. 5.56 for a drawing of a truck-mounted coiled tubing rig), the tubing can range from 2,000 to 15,000 ft. (i.e., 610 to 4570 m) or greater in Fig. 5.55—Coil tubing injector head. Courtesy of length. See Kumar et al. (2008) and Schlumberger (2004). Afghoul et al. (2004) for more-complete discussions of coil tubing interventions; they reported that methods such as chemical dissolvers, slickline tools, and downhole motors have been used successfully for removing various types of organic/inorganic scale. Many types of tools may be attached to the length of coiled tubing. Included are blaster-type services that use high-pressure water, solvents, or abrasives to remove deposits without damaging the tubing; more details of particulate blaster services are described by Frenier and Ziauddin (2008). An example is the use of a roto-wash tool used with coil tubing rigs that have been used for paraffin removal. Such a tool, seen in Fig. 5.57, was designed to jet clean the inner walls of the tubing, casing, pipelines, and the like, on a coiled tubing string (Schlumberger 2007). This rotary wash tool is designed with four ports on the circumference, which cause the wash shoe to rotate, and with one bottom port for jetting in that direction. High-pressure water, possibly containing detergents, can be sprayed from the tool at various pressures needed for the cleaning.

250  Chemical and Mechanical Methods for Pipeline Integrity

The rotation of the wash tool is driven by the thrust generated through the offset nozzle ports. As pump rate increases, the rotation of the wash shoe increases. The side ports will jet the pipe walls clean, and the port facing down will clean sand bridges and other loose fill. A centralizer should be run directly above the tool to provide stability and a more uniform cleaning action to the entire pipe surface. In addition, a wire scratcher can be run above the tool to improve cleaning action. High-pressure water as well as water/solids can be used to remove the deposit. Coil tubing tools are also used extensively to place and circulate the various cleaning chemicals (to be described in Section 7.3). Baker-Hughes (2003), covering various uses of coil tubing in scale removal activiFig. 5.56—Coil tubing rig (from Afghoul et al. 2004). ties, also shows various tools and mechanCourtesy of Schlumberger. ical cleaning devices. This report states that wellbore-cleaning solutions can include a variety of high-pressure jet washing tools. These tools are used for sand, paraffin, asphaltene, or scale cleanouts. High-pressure jet washing tools are also highly effective for cleaning completion accessories such as nipples, gas lift mandrels, and screens for which mechanical means of cleaning are limited. Halliburton (2005) provided a jetting service that combines fluid jet technology with the performance of coiled tubing; the service provided and claimed an effective solution for many wellbore cleanout problems. Also see Ali et al. (2002) for a case study on use of coiled tubing techniques in scale removal operations. Torres et al. (2005) described the use of coiled tubing for removal of asphaltene sludge. Crude produced in the North of Monagas Estate, East Venezuela, has a high asphaltene content, which comes out of solution in both the wellbore tubulars and the pipeline. This can lead to complete plugging of the pipeline and may increase the cost of maintaining production because of the need to periodically remove these organic deposits. In a specific case, noted by Torres et al. (2005), 9 km of 8-5/8-in.-OD production pipeline was successfully cleaned out using 2-in.-OD coiled tubing to regain pipeline production. The operator and the service company established a joint team to perform the feasibility study and the engineering. Some of the key points were the design of a frame to lay down the injector head, the placement of entry points along the pipeline, the selection of the bottomhole assembly, and the choice of fluids to use.

Fig. 5.57—A roto washing tool head. Courtesy of Schlumberger.

Mechanical Methods of Assessment and Enhancement of Pipeline Operations   251

Another issue was the measurement of stresses (push/pull) on the coiled tubing so that it could be run as far as possible in the pipeline without damaging either it or the pipeline. The pipeline was successfully cleaned, the coiled tubing being run seven times from five different entry points in the pipeline. The fluid chosen was a mixture of diesel with xylene, and gelled water was also used to help remove the debris. The line was cleaned during 22 days with eight entries at designated points. This resulted in substantial savings for the operator and significantly reduced the time to recover normal production in the pipeline. Buali et al. (2014) describe the use of coiled tubing with a jetting tool to remove FeS from well tubing. This job also used a linear hydroxypropyl guar fluid instead of water in the jetting process. The viscous fluid reduced the horsepower needed as well as being able to suspend the removed deposits. A description of these gels is presented in Section 6.6 and in Frenier and Ziauddin (2014, Chap. 4). 5.6  Summary and Lessons Learned • A large variety of mechanical devices, including pigs, scrapers, ILI tools, injectors, and other intervention tools, are in daily use to maintain and renovate pipeline infrastructure worldwide. • There is no single universal device, but each set of tools must be planned for use on the basis of the wide variety of pipeline conditions present. • ILI developments are being driven by IM requirements and are able to identify a large range and number of possible pipeline defects. A vital part of the process is classifying and providing further analyses of the defects found. • Several programs were reviewed for classifying anomalies found by ILIs and for determining which ones require on-site inspections. • Ancillary equipment such as pig launchers/receivers and trackers, as well as coil tubing equipment, is in use in pipeline operations. Additional devices are discussed in Sections 6.7 and 7.2. 5.7  Best Practices and Case Studies for Mechanical Management of Pipelines 5.7.1  Unpiggable Pipelines: What a Challenge for In-Line Inspection! (Schmidt 2004). A piggable pipeline is a pipeline that is designed to allow a standard inspection tool to negotiate it, which basically requires a more or less constant bore and sufficiently long radius bends and traps to launch/ receive the pigs. This way, an unpiggable pipeline can be defined as not designed like this. The paper describes options for using ILI devices in difficult-to-pig lines. Also see Wint (2011). 5.7.2  Case Study: Bayu Undan Pipeline and Darwin Liquefied Natural Gas Project (Weatherford 2009). This study involved precommissioning/commissioning services for a 26- to 28-in. dualdiameter pipeline that is 312.3 miles long. The report provides details of a very large dewatering operation that consisted of propelling six multidiameter long-run pigs with high-pressure, superdry compressed air at a velocity averaging 0.5 m/s. The pig train was designed to desalinate the pipeline during the dewatering process and leave a minimal layer of water on the internal pipe to achieve efficient air drying. 5.7.3  Single-Trip Pigging of Gas Lines During Late Field Life (Mandke et al. 2002). Recently, the industry has developed subsea pig launchers that are specifically designed for deepwater and ultradeepwater applications. However, there are significant operational challenges associated with single-trip pigging that need to be resolved. This paper draws attention to potential difficulties in the pigging of multiphase gas lines during late field life condition when the line is likely to hold a large amount of liquid. A pig traveling through such a flowline is likely to stall at the base of a deepwater riser as a result of the buildup of a high static head of liquid column in front of it and insufficient drive energy provided by the reservoir fluid.

252  Chemical and Mechanical Methods for Pipeline Integrity

There are some options, such as using gas lift at the riser base, which are available to deal with this problem. Pigging with bypass is another cost-effective method to mitigate this problem. Using a hypothetical field, pigging simulation results are presented to demonstrate how inclusion of 10 to 15% bypass in the pig benefits the pigging operation. Typical pipe-cleaning pigs are limited to 3 to 5% bypass (see examples of bypass pigs in Fig. 5.18). 5.7.4  In-Line Inspection on an Unprecedented Scale (Brockhaus et al. 2015). This report documents the inspection of what was the world’s longest subsea gas pipeline at the time of the report. This line is 1173 km in length and connects Norway to the UK. The authors claimed that the line had not been inspected until the ILI run described in the report. Some of the combined technologies described in Section 5.4 were used, and the problems with line diameter changes had to be addressed. They claimed the line was successfully assessed in 2009.

Chapter 6

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations Chemical and mechanical processes for maintaining various pipeline activities are required in support of the dual goals of flow assurance (FA) and pipeline integrity management (IM). Essentially all the chemical enhancements (corrosion, scale, and organic-foulant inhibitors, and flow enhancers) require at least one mechanical step for application. Thus, both types of processes are needed for operational successes. The extensive methods used for cleaning and renovating fouled pipelines, including progressive pigging and pigs staged with chemical solvents, are described in Chapter 7. Note that Chapter 7 overlaps somewhat with the current chapter as well as with Chapter 5. The next sections of this chapter, Sections 6.1 through 6.4, provide short reviews of the various chemicals used in pipeline maintenance. These include corrosion inhibitors, inorganic scale and organic deposit inhibitors, and friction reducers, gastreating chemicals, and demulsifiers. Many more details of these chemicals and applications are found in Frenier and Ziauddin (2008), Frenier et al. (2010), Frenier and Ziauddin (2014), Kelland (2009; 2013), Ali et al. (2016), and Hill and Cassidy (2016). Also note the references in the following sections. Section 6.5 describes chemical methods for maintaining oil/gas treatment facilities. Section 6.6 describes the use and mechanism of chemical “gel” pigs that can be applied as a substitute for, or in tandem with, mechanical pigs for maintenance tasks. Section 6.7 demonstrates the use of various mechanical devices for applying the chemicals to the lines and process equipment. In the next chapter, Sections 7.1 and 7.2 discuss the use of mechanical pigs for removing solids/liquids from the lines as part of ongoing pipeline maintenance. Details of the many cleaning chemicals in use are provided in Sections 7.3 through 7.6. 6.1  Corrosion Inhibitors and Inhibition Mechanisms The use of corrosion inhibitors for control of downhole, internal corrosion in pipelines and in facilities is a highly developed industry. Some aspects that pertain specifically to control of pipeline corrosion are described in this section. For more complete discussions of the wide variety of inhibitor molecules see Frenier and Ziauddin (2014), Frenier and Ziauddin (2008), Kelland (2009, 2013), Fink (2003), Raman and Labine (1993), and EFC (1994). This section describes “surface-active” corrosion inhibitors that block oxidizing agents from attacking a metal surface. Corrosion can also be prevented by removing the oxidizing agents from the fluids. These materials, often called “scavengers,” are described in Sections 6.4.4 and 6.5.

254  Chemical and Mechanical Methods for Pipeline Integrity

Commercial corrosion inhibitor formulations for use in well fluids, pipelines, and cleaning fluids will have a number of components, including • • • •

The “active ingredient” that provides most of the inhibitor protection A surfactant package to help disperse the inhibitor Solvents to make a one-component mixture and enhance storage lifetimes Possibly inhibitor “aids” to improve performance in special conditions (e.g high-temperature or high-alloy steel)

These components affect the use range (e.g., brine/oil-phase compositions, acid gases, acid strength for cleaning chemicals, and temperature) as well as the overall impact of the inhibitor on the environment. Thus, the whole package must be tested, not just the active chemical as reported in many of the published papers cited in this book. In addition to corrosion testing, the formulation must not cause emulsions or foams to form. To control corrosion from O2 in water systems of oil/gas pipelines or facilities, the industry uses chemicals that are very different from those that control various acid-type attacks. Controlling O2 corrosion usually entails either removing the oxygen from the system or promoting a passive film on the steel. Nitrite salts (X-NO2) are in use as passivating inhibitors (also see Section 6.4.4). Water treatment is integral to the oil/gas industry but is generally beyond the scope of this book; see the discussions in Davies and Scott (2006). The materials describe here are designed to reduce the attack of corrosive gases and liquids (CO2, H2S, HAc, and cleaning formulations) on pipeline steels, as described in Sections 3.1 and 3.2. Recall that the laboratory methods for determining corrosion rates were described in Section 3.4.2. Controlling the internal corrosion also partially controls iron carbonate and iron sulfide scale (see Section 4.2.2). The goal of inhibition, according to NACE (2013c), is to reduce the corrosion rate, from the uninhibited value, CRu, to a lower inhibited value, CRi. On the basis of an inhibited rate, the inhibitor should, according to this report, provide inhibition protection (values of >95% reduction. This protection value (% Inhibition) is calculated as  CR − CRi  % Inhibition (Inh) =  u × 100 ���������������������������������������������������������������������������������� (6.1)  CRu  Even if the % Inhibition values are very high, the absolute rates suggested by NACE (2013a) may be at issue (see Table 6.1).These depend on temperature as well as the overall plans/models described in Section 3.6. Note that Table 6.1 does not apply to inhibition of acidic cleaning fluids that are described in Section 7.3.4. The data from Eq. 6.2 can be used, according to NACE (2013a), to provide life estimates of a pipeline service, assuming uniform corrosion. The report describes a corrosion allowance (CA) “availability” model,

Temperature Range, °C

Inhibited Corrosion Rate, mm/y

Up to 120°C

0.1

>120°C and 10 to adhere to iron and thus would be effective only as a cooling water inhibitor, not in acidic brines. However, Craddock et al. (2006) claimed that some polyaspartate chemicals and some polyglucosidse molecules (Fig. 6.4), and especially when mixed together, provide corrosion inhibition in brine/oil/CO2 fluids that is as effective as “traditional” chemistries. A patent in the names of Guzmann et al. (2012) discloses an inhibitor formulation that specifies a two-part corrosion inhibitor: a. The first part is a glycoside component comprising at least one glycol side of the formula R(OG)x, wherein R is an aliphatic hydrocarbon radical having 1 to 25 carbon atoms. The G is residue of a saccharide moiety selected from the group consisting of fructose, glucose, mannose, galactose, talose, gulose, allose, altrose, idose, arabinose, xylose, lyxose, and ribose andalkoxylated derivatives thereof, and x is 1 to 30. b. The second part comprises at least one polymerization product of aspartic acid, optionally in form of a copolymeriate with fatty acids, polybasic carboxylic acids, anhydrides of polybasic carboxylic acids, polybasic hydroxycarboxylic acids, monobasic polyhydroxycarboxylic acids, alkoxylated alcohols, alkoxylated amines, amino sugars, carbohydrates, sugar carboxylic acids, and polymers. Bain et al. (2003) claimed to have tested low-molecular-weight thermal polyaspartates as scale and corrosion inhibitors in an open, recalculating cooling water system running at pH 8.5. This particular application agrees with the predictions of Kalota and Silverman (1994), who noted that high pH is required for polyasparates to be effective. The ethoxy moiety gives the compound increased water solubility. If sulfide is not present, the formulation may also include a phosphate ester. Increasing water solubility of the inhibitor can be a major achievement because marine toxicity frequently is related to solubility of the material in the cell’s lipid bilayer (Clewlow 1995). Furthermore, water-soluble inhibitors will not need toxic hydrocarbon solvents. Fischer and Parker (1997) increased water solubility of the classic tall oil fatty acid condensate by making the anhydride of the fatty acid then partially neutralizing it to an ammonium or potassium salt. The authors claimed good film persistency and corrosion inhibition properties. Harris et al. (2010) noted that film-forming corrosion inhibitors are often selected to control CO2 corrosion and that their effectiveness vs. microbiologically influenced corrosion (MIC) is desirable in systems that suffer from both forms of corrosion. Traditional corrosion inhibitor tests (e.g., bubble tests) have unfavorable conditions for microbial Fig. 6.4—Glucoside molecule structures.

262  Chemical and Mechanical Methods for Pipeline Integrity

activity and are inadequate for evaluating MIC control. Biocide-screening test methods have been used to evaluate microbial kill with toxic chemicals added batchwise, providing very little direct information about controlling corrosion. Once-through flow cells containing corrosion coupons were inoculated with a field consortium enriched in synthetic produced water to simulate MIC field activity. Maximum pitting rate on the coupons was the key performance indicator for screening inhibitors. Results indicated that many of the corrosion inhibitors tested increased the maximum MIC pitting rates when compared to untreated controls. In at least one case, a less toxic inhibitor provided better MIC control than a more toxic inhibitor. Data suggest that the field microbial consortia used in the testing developed resistance to an incumbent inhibitor that has been used for many years. The results indicate that inhibitor selection made on the basis of MIC control is not simply a function of ability to control bacterial growth and activity. 6.1.3  Sour Brine Corrosion Inhibitors and Inhibition of Microbiologically Influenced Corrosion. Most of the studies described above have been directed at inhibitors for “sweet” systems in which the corrosion reactions involve CO2. Sour (H2S) well/pipeline fluids must also be inhibited. The inhibitor formulations are similar to those used in CO2 conditions but must be formulated and tested with H2S present to ensure adequate performance. The inhibitors used for fluids in contact with CO2 when elemental sulfur and H2S are also in the systems include cocodimethyl benzyl ammonium chloride as well as aminoethyl fatty imidazolines and alkyl pyridinium ammonium chloride. These are similar in structure to some CO2 inhibitors. See Fig. 6.5 for typical sour fluid inhibitor molecules; these must usually be specially formulated to be dispersible in the more acidic H2S environments and in the presence of an FeS-containing scale. Moore and Liu (2009) addressed some principal properties of corrosion inhibitors and how these properties affect the ultimate fate of an inhibitor in a production system. Specifically, they presented affinities of various corrosion inhibitors for sand, iron sulfide, barium sulfate, iron carbonate, and emulsion drop surfaces. Their paper reported the behaviors and performance of selected corrosion inhibitors (above) in the presence of elemental sulfur. This study looked at systems where corrosion was taking place under a sulfur deposit, and a cell for such studies is described in the report. They found that the quaternary inhibitors will adsorb on sand and sulfur, and thus the effectiveness may be reduced. Ramachandran et al. (2002) used a force field program that was developed to understand mackinawite, an important iron sulfide scale formed during H2S corrosion. They claim that the dissolution of this scale to be a rate determining step during the corrosion of iron in H2S environments. They proposed that inhibitors retard the dissolution of scale by binding to it and preventing acidic attack of the scale. This work is claimed to form an important step in using molecular modeling techniques to study the corrosion of iron in H2S environments and its inhibition. The + N corrosion inhibitor used in the study was Cl1-benzyl-2,6 dimethylpyridinium chloride. The calculations showed that the 1-benzyl-2,6 dimethylpyridinium chloride lowest energy structure arrived at through annealed dynamics is that the pyridine ring of 1-benzyl-2,6 dimethylpyridinium chloride does not follow a flat orientation N+ with the surface but adopts a 70° angle Clwith the surface. The most favorable interactions are those in which the hydron-dodecyl, n,n-dimethyl benzyl ammonium chloride gen atoms of the pyridine ring bind with Fig. 6.5—Sour (H2S) corrosion inhibitors. the surface sulfur atoms of mackinawite.

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   263

Several investigators have presented papers/patents that claim improved corrosion inhibitors for mixed CO2/H2S brine systems. A presentation of their findings as well as citations of the most recent patent applications follow next. Stewart, Jovancicevic, Menendez et al. (2010) described the development of a new corrosion inhibitor for application in sour gas/oil production. During the development process, controlling pitting corrosion in laboratory testing was important for quantifying inhibitor performance. This criterion was set because pitting or localized corrosion is the most significant internal corrosion threat in sour gas production. Yang and Jovancicevic (2009) disclosed imidazoline dimer-type compounds that are prepared by the reaction of dimer fatty acid and a dialkylene triamine, such as diethylenetriamine (DETA), which are claimed to be useful for corrosion inhibition in water-containing fluids contacting metal, particularly fluids containing CO2 and/or H2S. Because MIC is a cause of some of the H2S present (described in Section 3.1.7), MIC may be tcontrolled by • Preventing/destroying the sulfate-reducing bacteria (SRB) that lead to some H2S formation • Inhibiting the MIC with corrosion inhibitors described in this section that may also be effective as biocides. However, this may affect the biodegradation of the formulation A report by Greene et al. (2006) cited by Harris et al. (2010) claimed that corrosion inhibitors, (especially nitrogen quats), may be used in concert with SRB biocides (see Section 6.4.3) for a more effective control of SRB biofilms and thus MIC. The filming corrosion inhibitors described by these authors include imidazolines, primary amines, diamines, amino-amines, oxyalkylated amines, fatty acids, and dimer and trimer acids (see Figs. 6.1 and 6.5 for some structures). However, they also claimed that though some of these chemical moieties are toxic to microbes, others are susceptible to microbial degradation. Harris et al. (2010) used a flow cell (Fig. 6.6) and reportedly found that results indicate that some corrosion inhibitors may enhance MIC pitting by acting as a nutrient source, whereas other corrosion inhibitors inhibit MIC as a result of unknown mechanisms. In pipelines where multiple corrosion mechanisms are at work (e.g., MIC and CO2), it would be advantageous to know if the corrosion inhibitor selected is acting as a promoter of MIC or as an inhibitor; therefore testing as described by the authors is suggested. The authors also found bioscale and pitting in wastewater lines despite treatments with different corrosion inhibitors. 6.1.4  Inhibitors for Gas-Containing Lines and Multiphase Lines. This section describes the need as well as the application of inhibitors in gas-containing lines, including gathering lines, any pipeline with multiphase flow, and sales gas lines. Note that the inhibitor chemicals have been described in the previous sections, but the formulations may have different requirements. Inhibitors for Top-of-Line Corrosion Conditions. There are several chemical factors as well as approaches for providing protection for pipelines periodically exposed to condensing corrosive water, or top-of-line corrosion (TLC) (see Section 3.2.1). The conventional method is to formulate/ synthesize the inhibitor formulations to persist during the time between periodic applications (see Section 6.7.4 for several methods of application). During the nonapplication period, the surfaces will still be bathed in the corrosive fluids, so methods such as those described in Section 3.4.2 must be used to determine the protection time and thus the intervals for retreatment. Much of this section has been influenced by Frenier and Wint (2014). Inhibitors used for sweet corrosion must also protect the steel from attack by acetic-acid-containing fluids that may be present in the top of pipelines. Martin (2009) claimed that proper blends of imidazoline salts and light amines can inhibit vapor space corrosion far better than either component alone and that the ratio of blending is important. He also noted that these blends were subsequently optimized for both weight-loss corrosion and hydrogen entry. For any TLC inhibitor, the persistence of the material (see Section 3.2.1) must be optimized to give protection for as long as possible.

264  Chemical and Mechanical Methods for Pipeline Integrity

Gas trap

Coupon sites

Temperature probe

Chemical injection point

Effluent

Syringe pump

Nutrient reservoir

Working reservoir

Fig. 6.6—Schematic of once-through flow cell indicating input from the working reservoir and oncethrough flow cell where coupons are inserted at multiple sites. A syringe pump provides continuous dosing of inhibitors (Harris et al. 2010).

Section 6.1.2 has described the basic chemistries of inhibitors used to reduce the impact of sweet as well as sour corrosion. These types of molecules are termed “active ingredients” (AIs). The actual liquid inhibitor formulations that that are pumped or otherwise placed into a pipeline segment require several other components to be effective. The choice of the AIs is predicated on the corrosion process, fluid phase that requires inhibition, and target location and mechanism for replenishment of the inhibitor film. Continuously placed inhibitors usually go into the aqueous phase and thus require a solvent as well as surfactants to help dispersion into the aqueous phase. Miksic et al. (2004) proposed an inhibitor formulation for use in continuous injection pipeline operations. The formulation is a solvent used as a “carrier.” In their patent of AIs, the authors gave examples that include a fatty acid anhydride and a 21-carbon dibasic acid with an amine or imidazoline to form a corrosion inhibitor consisting essentially of a fatty acid derivative. The inhibitor is subsequently dissolved in a fatty acid oil or ester selected from the group consisting of soybean oil or methyl soya ester. To make this inhibitor water dispersible, one can then add sulfonates and a long-chain ethoxylated alcohol, adjusting the viscosity with an alcohol comprising isopropanol. Note that this patent gives a good example of making a useful corrosion inhibitor formulation. Batch/pig-placed inhibitors (see Section 6.7.3) are frequently targeted at TLC and may need to be soluble/dispersible in a nonaqueous phase. Thus, in addition to the AIs, an oil-soluble solvent (such as diesel oil) and oil-soluble surfactants are required. Because the inhibitor film is not continuously replenished, the film must persist during contact with gases/liquids until the next treatment. Agents that increase the viscosity and the film persistence may also be in the formulation. DeMarco et al. (2001) and DeMarco et al. (2002) studied a number of pure corrosion inhibitors to understand the persistence at a surface molecular level (Table 6.2 and Fig. 6.7). The authors used film persistency tests as well as several surface-analytical methods to study adsorption. The film persistence measurements have been conducted extensively using various

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   265

Inhibitor Compound

Adsorption Mechanism

Hexadecenylsuccinic anhydride

Chemisorption

Hexadecylsuccinig anhydride

Chemisorption

Octenylsuccinic anhydride

Chemisorption

Benzyldimethylhexadecylammonium chloride

Physisorption and/or chemisorption

Trimethylhexadecylammoniym chloride

Physisorption and/or chemisorption

Lauric acid

Physisorption and/or chemisorption

Sebacic acid

Physisorption and/or chemisorption

Cetylpyridinium chloride

Physisorption and/or chemisorption

Cetylamine

Physisorption and/or chemisorption

Stearylamine

Physisorption and/or chemisorption

Dimethyldodecylamine

Physisorption and/or chemisorption

Nalco/Exxon product

Physisorption and/or chemisorption

Cetylcyanide

Physisorption and/or chemisorption

Cetylmercaptan

Physisorption and/or chemisorption

Dimethyldipalmitylammonium chloride

Possibly physisorption

Folic acid

Chemisorption

Table 6.2—CO2 inhibitors studied and probable adsorption mechanisms (DeMarco et al. 2001).

techniques (see Section 3.4), as well as electrochemical impedance spectroscopy, polarization resistance, electrochemical noise analysis, and impinging jet electrodes. They claimed that some work has been carried out to extend these studies to investigations of inhibitor films in the presence of both high fluid shear stresses and multiple phases. DeMarco et al. (2001, 2002) claimed to have developed a tentative quantitative structure/activity relationship for 16 corrosion inhibitors, shown in Table 6.2; the results demonstrated that some compounds undergo chemisorption and display good persistence, whereas other inhibitors are not persistent and may experience physisorption. Their work is also claimed to agree with surface-analysis studies referenced in DeMarco et al. (2001). These included surface-enhanced Raman spectroscopy, surface reflection Fourier-transform infrared spectroscopy, and X-ray diffraction and showed that inhibitors based on a succinic anhydride functionality are chemisorbed at the surface of mild steel through the formation of an Fe(II)/ (III)-dicarboxylate complex. They also claimed that even if the inhibitor is strongly adsorbed, to provide good persistency it still has to adsorb (and remain) in the presence of an oil. The reactions with the surface are claimed to include the chemisorption of the dicarbocylic acid onto the surface: 2+ 2+ Fe (ads) + R(COOH)2(ads) → R(COO− )2 Fe (ads) + 2H+(aq) ������������������������������������������������������������������ (6.4)

On the basis of their experimental work, DeMarco et al. (2001, 2002) also noted that the most persistent inhibitors/formulations also adsorb a film of hydrocarbon that increases the hydrophobicity of the surface product. This extends the time for removal and then the increase of the corrosion rate to the uninhibited values. Note that this is similar to the mechanism proposed in Fig. 6.2. This mechanism is also supported by the work of Li et al. (2013), who tested model inhibitors (a quaternary ammonium chloride and a fatty amine) in an oil/water mixture under laboratory flowing conditions. They found that by increasing the oil-wetting of steel, the inhibitor molecules helped form an antiwater film on the surface, thus improving corrosion inhibition.

O

O

O

HO

O

O

O

Hexadecylsuccinic anhydride

Hexadecenylsuccinic anhydride

N+

O

Cl-

Cl-

Sebacic acid

Lauric acid

O

O

OH

OH

+

N

N

N

N

OH

Cl-

N

HN O

HN

Dimethyldipalmitylammonium chloride HO

H3C(H2C)13

N+

Cetylmercaptan H3C(H2C)13

Cetylcyanide

Dimethyldodecylamine

Stearylamine

Cetylamine

Cetylpyridinium chloride

N

O

O HO (2S)-2-[(4-{[(2-amino-4-hydroxypteridin-6-yl)methyl]amino}phenyl)formamido]pentanedioic acid (Folic acid)

H 2N

HS

N

H2N

H2N

Cl-

N+

Fig. 6.7—Structures of tested CO2 corrosion inhibitors.

Trimethylhexadecylammonium chloride

Benzyldimethylhexadecylammonium chloride

Octenylsuccinic anhydride

O

O

O

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   267

Cain and Rosenthal (1993) claimed to have developed a polyamine-based material that is said to be a film-persistent, water-dispersible corrosion inhibitor. It is a quaternary ammonium salt that was optimized for brine dispersability and corrosion inhibition. The authors noted that the inhibitors were derivatives of ethylated tertiary amines made from a process using methyl chloride, methyl sulfide, and benzyl chloride and methyl phosphoric acid. The inhibitor was tested at several concentrations under many accepted industry methods, including kettle tests and continuous and film persistence wheel tests. They claimed that the inhibitor performance in the tests was equivalent or superior to the performance of common quaternary intermediates at only one-quarter of their concentrations. Most important, these authors claimed that the inhibitor displays excellent film persistence. This can be a major advantage, given that most existing water-soluble corrosion inhibitors have very poor film persistence and must be administered continuously rather than in batches. They also do not strongly adhere to metal surfaces and are ineffective unless continually present in the surrounding fluid. Narasaiah et al. (2013) described work (chemistries not revealed) to develop inhibitor formulations that will protect the pipe bottom as well as the top of the line by using a single formulation that is added continuously to the pipeline flow, not batched. The author of the current book concludes that such a formulation may contain both nonvolatile and volatile filming molecules. Because complete distribution of chemicals to the top of the line is difficult to achieve or predict under the varying pipeline conditions, several different chemical additives are claimed to improve TLC corrosion inhibition, even under unfavorable flow conditions. Because most inhibitor formulations are designed to stay in the liquid phase, the issue of volatility is being addressed by several researchers. An alternative method for protecting against TLC may use inhibitors that vaporize from the flowing aqueous phase at the bottom of the line (note that vapor phase corrosion inhibitors, vapor phase corrosion inhibitors also may be part of a more conventional inhibitor formulation). The vapor space of pipelines may be protected by inhibitors, usually called vapor-phase inhibitors), designed to be volatile under flowing conditions. Martin (2009) alluded to these types of chemicals, and specific citations by Oehler et al. (2012) described the testing of primary, secondary, and tertiary volatile amines as specific TLC inhibitors. The exact chemicals are not described, but patents disclose the use of several volatile amines seen in Fig. 6.8 (see Anbarasi et al. 2013). Miksic et al. (2013) also described the testing of volatile TLC inhibitors using devices described in Section 3.4. Their results show that some azoles and acetylene alcohols provided top-of-line protection, whereas other tested acetylene alcohols showed pitting. They also tested sulfur-containing compounds that showed pitting. Blended products were claimed to be more effective in an aceticacid-containing fluid than any single chemical. See Jenkins (2011) and Frenier (2003a) for examples of organic acid/chelate solvent inhibitors having an AI blended with sulfur. Schmitt et al. (2001) claimed “film-spreading agents” can be added to the inhibitor formulation to promote the chemicals to spread and cover all the surfaces. The central claim of the report is that careful tuning of the spreading system, composed NH2 of single- or two-phase bottom-of-line liqH2N uids, spreading agents, and the corrosion HN inhibitor, yields optimal corrosion protecNH2 tion on the entire pipeline surface even at Cyclohexylamine Guanidine high water condensation rates. The authors noted that perfluorosurfactant tetraethyl CH3 OH ammonium perfluorooctane sulfonate (TEPOS), may be combined with nonionic N N HO surfactants. Examples of nonionic surfacH tants are tallow alcohol ethoxylate with an N,N-Dicyclohexylamine Methyldiethanolamine average content of five ethoxy (EO) groups Fig. 6.8—Volatile amine corrosion inhibitors. per alcohol moiety. The authors showed

268  Chemical and Mechanical Methods for Pipeline Integrity

that this combination reduced the top-of-line spreading time of the inhibitors compared with controls. On the basis of the laboratory tests described in the paper, the preferred spreading agents are surfactants that reduce the surface tension of the aqueous part of the bottom-of-line liquid already at low concentrations. The corrosion inhibitor must be chosen according to the type of the corrosive gas (CO2, H2S, or mixtures thereof ) and the corrosion severity expected. Two-Part Coating Inhibitors for Pipelines. An alternative inhibition technology claims to produce a more-adherent and -long-lasting corrosion protection film than can be provided using the batch methods to be described in Section 6.7.3. These methods use some of the inhibitor molecules described previously in this section and are illustrated by Figs. 6.1, 6.5, and 6.7. The binary technologies are developed on the basis of the work described in the patents of Zaid (1999), Zaid and Wolf (2001), Zaid et al. (2008), and Zaid et al. (2013).The concept, described in detail in the paper by Zaid and Sanders (2005), involves the application of a two-part epoxy amine–containing coating that can be applied to well tubing and to pipelines and will supply longer-term corrosion control compared with conventional filming inhibitors. The documents describe the reaction of a low-molecular-weight epoxy resin with a “hardener” that will incorporate a corrosion-inhibiting amine functionality into a crosslinked polymer coating. A proposed mechanism is shown in Fig. 6.9. In this drawing, epichlorohydrin reacts with bisphenol-A in the presence of NaOH to produce a low-molecular-weight epoxy resin (Fig.6.9a). This resin is then “hardened” (i.e., cured) in the presence of a dibasic amine (Fig. 6.9b). This process, which may be exothermic, incorporates the corrosion-protecting amine into the film and also provides crosslinks; see other amines in Zaid (1999). See Pascault and Williams (2010) for more details of the very useful epoxy polymer technologies. A mechanistic section in the appendix of Zaid and Sanders (2005) shows several models for the functioning of the amine hardeners in producing a 3D-crosslinked matrix. OH

Cl

O

+

NaOH OH

Epichlorohydrin

a)

Bis-phenol-A

+ Epoxy resin

b)

H2N

N H

N-dodecyl-1,3-diaminopropane

Corrosion-inhibiting crosslinked polymer film Fig. 6.9—Two-part epoxy inhibitor coating.

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   269

Payne et al. (2012) and Freeman and Williamson (2006) described the use of a two-part inhibitor, developed on the basis of this technology, that is claimed to exhibit longer film life (and so longer corrosion protection) compared with one-part formulations. These authors did not specifically disclose the chemicals. However, they did also claim that the binary inhibitor can be premixed and applied as one part with a “pot life” of approximately 8 hours or in two parts using a spray pig (see Section 6.7.3) for the first half of the process and then a conventional batch-pig placement for the second half of the binary formulation. The authors presented results from laboratory and pipeline tests that showed improved corrosion protection compared with untreated and conventional batch inhibitors. Payne et al. (2012) and Freeman and Williamson (2006) also claimed that the pipe surface should be clean (or new) for maximum film persistence. Methods for chemical and mechanical cleaning of heavily fouled pipe surfaces are described in Chapter 7. Corrosion Inhibition for Sales Gas Lines. Pipe lines that deliver treated natural gas to a consumer (e.g., a power plant) or to the “city gate” for ultimate delivery to local homes and businesses, are referred to as “sale gas lines.” The conditions in the “dry” gas lines should preclude the need for inhibitors to protect the integrity of the lines; however, the formation of black powder (Section 4.4.4) in long-distance gas pipelines is considered as a condition requiring chemical inhibitors (Sherik and Jabran 2011; Zhang et al. 2012). Although the corrosion rates may be quite low (1), the conditions exist for scale formation on surfaces. From the standpoint of the production and pipeline operator, the packet of supersaturated water must leave the critical area before scale can start to affect production or cause some other undesirable problem, such as contribute to corrosion. Iron oxides will not form unless dissolved oxygen is in the water. Also see the discussion of equilibrium equations in Section 1.7.

272  Chemical and Mechanical Methods for Pipeline Integrity

O

O

HO P O HO

OH P

HO OH HO

O P

HO

OH

1-Hydroxyethane-1,1-diphosphonic acid

O

P

O N

P

N

O

OH

HO

OH HO

P

HO

OH

P

O

P

HO O

HO HO

OH

OH

P

P OH

O HO

Ethylenediaminetetra(methylenephosphonic acid) O H

CH COOH

CH2

m

P OH

CH2

CH H n COOH

Phosphonated poly acrylate, MW ~ 3600

H

CH2

H

CH n SO3H

Polyvinylsulfonic acid MW~15,000

Fig. 6.11—Chemical structure of generic scale inhibitors.

A scale inhibitor is defined as any chemical agent that reduces the rate of formation of a fouling scale on a surface of interest. Note that the scale may form somewhere else as long as the water remains in a supersaturated state, but the water may become undersaturated when it reaches the exterior or may no longer be able to contact critical surfaces. Inhibitors are different from chelating agents that chemically tie up all scaling ions, because inhibitors act on scaling surfaces and incipient scale nuclei, not on the dissolved ions. Inhibitors can be extremely efficient and thus are frequently referred to as ”threshold” materials. The most basic mechanistic idea is that the inhibitors adsorb onto critical sites on the forming scale crystals and thus block formation of larger crystals before they can actually cause fouling. Dispersant-type chemicals can also prevent any crystals from agglomerating into large clusters and then settling onto surfaces. Corrosion-related scale inhibitors may also act to protect metal surfaces from corroding and thus releasing iron ions (many more details will be provided). In this discussion, the use of methods to reduce the concentration of sulfates in injection water and chemicals that act on the surface of a growing scale crystal is considered to be part of the “inhibition” technology set. The problem for the operator and the inhibitor provider is to develop chemicals that will effectively delay formation of the type of scale expected in the particular pipeline environment and to provide ways for the inhibitor to be in the water when it becomes supersaturated. This frequently means that the inhibitor must be placed into the water-producing zone or at the bottom of the well or injected into a pipeline system. This also usually means that all the supersaturated water must be treated. Next is a general discussion of scale inhibition chemistry, followed by a description of the chemicals for treating various types of scale. This section includes descriptions of inhibitors developed for reducing the rate of formation of precipitated scale (Ca/Mg carbonates, Ca/Ba/Sr sulfates, silicates) and corrosion-related scale (Fe carbonates, oxides, and Fe sulfides). The types of inhibitors that are able to prevent scaling by alkaline earth salts such as calcite, gypsum, or barite are usually much different from chemicals used to prevent iron-based scale because the mechanisms of inhibition are quite different. Section 6.7 will describe various application methods. Many very effective molecules exist for actually inhibiting scale formation, and these must be tailored through testing for specific scaling conditions. The key issue is to provide a minimum concentration of the inhibitor molecules in the well fluid necessary to

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   273

suppress the formation of the target scale. This is termed “minimum inhibitor concentration.” The application technique used to place the inhibitor as well as salts that may accompany the inhibitor can greatly influences the maximum life of the inhibitor treatments. The overall goal of the inhibitor treatment is to provide the longest protection time possible before another intervention is required. Hundreds of chemicals are available for inhibition of scale formation processes, and because this is a domain of active research, more are being developed. These chemicals are placed into the production tubing during either a matrix squeeze treatment or a stimulation treatment or are continuously pumped in through a spaghetti pipe. Although the mechanisms of action of some phosphonates are being studied and a generalized picture of mode of action is emerging, this is a largely empirical domain. As a consequence, there is no guarantee that the ranking of effectiveness of various chemicals made in the laboratory holds true under downhole conditions. Another fact that needs to be kept in mind is the possible antagonistic effect of the various inhibitors used downhole: For instance, corrosion inhibitors could be enhancers for wax deposition because of the hydrophobic film they form (San Miguel and Rodger 2000). Several investigators have proposed generalized mechanisms for the functioning of salt-forming scale inhibitors that emphasize different aspects of a rather complex physical/chemical process. The general categories of salt-type scale inhibitors currently in commercial use include phosphonates, polyacrylates, and surfactants (dispersants). Tomson et al. (2004) developed a comprehensive model for this type of scale inhibitor. They proposed that the primary driving force for adsorption is related to simple hydrophobic repulsion from solution of a macro neutral molecule. On the basis of the nucleation study described above, the investigators observed that the inhibitor concentration needed to completely inhibit barite formation is approximately equal to 16% surface coverage. An equation to predict the minimum inhibitor need was proposed on the basis of this model and compared with field observations. The range of predicted inhibitor concentrations is quite similar to what is observed in the field as a minimum effective dose, even though it was arrived at by a completely independent method of calculation. These investigators modeled the influence of inhibitors on the nucleation time, assuming a separation of the effect of the inhibitor from that of the uninhibited mineral; for example, for barite,  b L inh 0 log10 (t ind ,sec) = log10 (t ind ,sec) +  inh  × Cinh ( mg / L ).���������������������������������������������������������� (6.6)  mg  Here, Cinh is the concentration of inhibitor added and binh is an inhibitor effectiveness term. This empirical equation (Eq. 6.6) works quite well for a large range of inhibitors and minerals, but is not mechanistic in origin. Furthermore, the empirical binh term can become complicated over the whole range of pH, T, P, and compositions encountered in oil/gas wells. The advantage of Eq. 6.6 for modeling the effect of inhibitors is that the effect of the inhibitor is separated from the basic nucleation of the scale-forming mineral, but the disadvantage is that there is no fundamental understanding of the binh term to aid in estimating its value for a new inhibitor or for additional conditions. Given Eq. 6.6, the effect of inhibitors can be viewed as increasing the interfacial surface tension between the growing nuclei and the solution. Continuing research has focused along five parallel lines: • • • • •

Adsorption isotherm of inhibitors on different solids at various solution conditions Solubility of numerous metal-inhibitor solids Minimum effective inhibitor concentration Induction time Effect of trace heavy metals on presence of inhibitor concentrations

Fig. 6.12 depicts a general mechanism (Crabtree et al. 1999; Frenier and Ziauddin 2008). Here, inhibition may be affected by • Crystal modification of an individual seed crystal • Threshold inhibition of growing crystal groups • Dispersion of formed crystal groups to prevent adhesion to a surface

274  Chemical and Mechanical Methods for Pipeline Integrity

Primary and Secondary Inhibition Mechanisms Microcrystals with adsorbed antiscale molecules

Crystal modification

Dispersion No crystal agglomeration due to charge repulsion

Threshold inhibition

Crystal morphology distorted, preventing formation of regular crystalline lattice and buildup of adherent scale

No further crystal growth: active sites blocked

Fig. 6.12—General scale inhibitor mechanisms (Chen et al. 2004).

Graham et al. (2004) summarized several mechanistic ideas to explain inhibition of barium sulfate scale by phosphonate chemicals as well as other chemicals. These authors noted that different inhibitor types may operate by different mechanism and that commercial products may have more than one active ingredient. Fig. 6.13 describes a proposed theory by Fig. 6.13—Generalized inhibition mechanism (Graham et al. Graham et al. (2002) and Graham 2004.). et al. (2004) such that the inhibition in one case arises from adsorption of a Ca-phosphonate complex from solution onto to specific sites on the developing scale crystal. The consensus is that 10 to 25% coverage of the critical sites will cause nearly 100% inhibition of scale formation. Note that this theory agrees with the amount of surface coverage estimated by Tomson et al. (2004). The inhibitors also affect deposition onto a steel surface. An inhibitor concentration greater than the minimum inhibitor concentration causes the Ba2+ ions to stay in solution, thereby greatly reducing the amount of surface scale. Concentrations below the minimum actually increase adherence of the scale and change the crystal structure in a deleterious manner. Shen et al. (2012) claimed that the high level of dissolved iron commonly present in the flowback water from some gas shale plays, especially the Marcellus, in the Appalachian Basin, in the US, adversely affects the ability of the scale inhibitor to inhibit calcium carbonate scale. In this study, the inhibition performance of two new chemicals and some commercial products was evaluated under static as well as dynamic test conditions using synthetic Marcellus waters at varying iron concentrations. It was shown that both new chemicals (claimed to be modified phosphonates) were able to control calcium carbonate scale effectively in the presence of dissolved iron up to 200 ppm, whereas the performance of polycarboxylic acid, amino tri(methylene phosphonic) acid, and carboxymethyl inulin dropped sharply even in the presence of small amounts of Fe2+ (5 ppm).

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   275

The inclusion of iron-sequestering agents with these chemicals and the effect of iron on calcium sulfate inhibition are also discussed by these authors. This document also has an excellent description of mineral scale inhibitor test methods. Iron sulfide can also form in pipelines or pipeline units and may be controlled by adding H2S scavengers (Section 6.5.2 and Frenier and Ziauddin 2014). Fowles et al. (2014) noted that for ferrous sulfide scale, there is a need for a quick and reliable test method. They claimed that the preparation of ferrous sulfide under sour conditions is not straightforward because of both the difficulty in achieving sufficiently anaerobic conditions and the associated hazards of managing H2S. The authors describe an improved test method. Several types of inhibitors were tested with some initial success. Included are phosphonates, organic acids, and tetrakis (hydroxymethyl) phosphonium sulfate. Section 6.5.2 also describes H2S scavengers. 6.3  Organic Deposit Inhibition Inhibition of organic deposition (i.e., deposition inhibitors) may be a method used in some pipeline operations. The author of this book contends that some situations may require introducing inhibitors into a flow stream to prevent damage. Note that the fouling compounds were described in Section 4.3. Short presentations on paraffin (wax), asphaltene, gas hydrate, and calcium naphthenate inhibition are provided in the following sections. For more details, see Frenier et al. (2010), Kelland (2009, 2013), and Fink (2011). 6.3.1  Paraffin Deposition and Pour Point Inhibitors. According to several authors (Tung et al. 2001; Jennings and Newberry 2008), there are two major reasons for employing wax inhibitors: 1. Reducing or preventing the accumulation of wax and wax gel deposits in flowlines 2. For high-pour-point waxy crude oils, paraffin inhibitors (in this context, often called “pour point depressants” or “dispersants”) are added to improve the oil flow characteristics. Note that a number of different chemicals are in use to achieve these purposes. Pedersen and Ronningsen (2003) claimed that functionally, these chemical inhibitors can act as wax crystal modifiers, wax dispersants, and detergents (surfactants). The detergent (surfactant) inhibitors require water to be effective (Tung et al. 2001). The literature discloses a number of polymers proposed for inhibiting wax deposition as well as reducing the pour point and modifying rheological properties of the fluids. Wax crystal modifiers include ethylene-vinyl acetate (EVA) polymers, alpha-olefin-maleic anhydride (OMAC) polymers, ethylene-butene-PEB-n copolymers, and ester-type paraffin inhibitors such as polyacrylate and polymethacrylates, including poly (octadecyl acrylate). Modifications include copolymers with N-vinylpyrrolidone and alkyl acrylates and/or methacrylates and polymeric phosphate esters. Surfactants also have been used to reduce wax (and asphaltene) deposition and may also function as pour point modifiers if water is present (Bernadiner 1993). These include alkylaryl sulfonates mixed with polyethyleneglycol ether di-terbutylphenoles, OH Triton-X (ethoxylated phenols), and Tween-type surfactants (Ahn OH et al. 2005); the latter are sorbitan based (see Fig. 6.14 for the base structure). An example of proposed mechanisms is presented in Fig. 6.15, which shows the surfactants interfering with the strucOH ture of the wax, thus delaying precipitation. 6.3.2  Asphaltene Inhibitors. Asphaltenes, discussed in Section 4.3.2 and Frenier et al. (2010), are the only type of organic deposit that have no crystalline compounds present. Crystallization is not

OH

O

sorbitan Fig. 6.14—Sorbitan.

276  Chemical and Mechanical Methods for Pipeline Integrity

observed because there are thousands of different asphaltene molecules in crude oil, and they precipitate as an indeterminate mass. Therefore, all asphaltene “inhibitors” have a dispersion component. Two classes of inhibitors have been defined by Kelland (2009) and are presented in the following subsections. These have designations by this author as having dispersant properties as compared with inhibitory properties. See the comparisons with scale inhibitors in Section 6.2. Asphaltene Dispersant Type (Monomeric) Asphaltene Inhibitors. Various monomeric inhibitors are described by Kelland (2009), who provided an extensive list of known or proposed asphaltene dispersant (AD) materials that include nonpolymeric amphiphiles and sulfonicFig. 6.15—Wax inhibitor mechanism (Venkatesan based nonpolymeric surfactants. In addition, he describes nonpolymers with acid head groups, 2004). amide and imide surfactants, alkylphenols, and ion-pair surfactants. These materials are claimed to influence the stability of the crude oil; however the author of this book has not found conclusive proof of this theory. Active Ingredient (Polymeric) Inhibitors. The AI inhibitors are polymeric dispersants that help prevent the deposition of the asphaltenes on surfaces. Kelland (2009) described a large number of materials that are thought to have inhibitory properties (AI materials) in that they may affect the particle size of the asphaltenes. The author names this class of materials “oligomeric (resinous)” and “polymeric” AIs. The chemical structures include polyesters and polyamide/imides, alkylphenolaldehyde resins, lignosulfonates, and several other categories of oil-soluble polymers. According to Smith et al. (2008), the asphaltene inhibitor specificity can be explained by acid-base-type interactions between the inhibitor and polar species in the crude oils or asphaltene fractions. This study, according to the authors, provides the first evidence for inhibitor effectiveness related to heteroatom content derived from detailed polar chemical composition. The authors of this book note that all AI/AD formulations must be tested with the on-site crude oil for inhibition effectiveness and must be in the oil before/as the precipitates form (see Frenier et al. 2010). Because of the heterogeneous nature of the asphaltenes, there are no materials that provide crystallization-level inhibition. Most documents and patents that show chemical compositions claim materials that act as oil-soluble surfactants or dispersants and the specific chemistry of the oil, asphaltenes, and inhibitors must be matched. Several authors claimed that the heteroatoms (N, P, and metals) affect the effectiveness of the inhibitors. Organic solvents are included in many of the formulations. As evident by the number of patents, researchers are still trying to develop improved asphaltene inhibitors. There is much less information on the mechanisms of action of asphaltene inhibitors compared with wax inhibitors. The general conclusion of the papers abstracted is that the dispersant chemicals act as “resins,” peptizing the asphaltenes and keeping them dispersed as very small particles. The major chemical structures are those of an oil-soluble dispersing agent. It is difficult to judge the effectiveness of asphaltene inhibitor use in oil fields. Stankiewicz et al. (2002) claimed that use of continuous injection of asphaltene inhibitors (type not disclosed) resulted in operational expense savings of USD 2,000–500,000/yr in some deepwater wells. It is also difficult to judge the inhibitor effectiveness vs. the solvency effectiveness of AD/AI materials because many packages will contain an asphaltene solvent as well as an “inhibitor.” Usually, production companies follow the production loss of a well that is thought to be fouled by asphaltenes and then

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   277

introduce an inhibitor package, which usually also contains a solvent; they then retreat when production reaches a critical stage. In the scope of this book, which includes gathering lines, surface facilities, and production pipelines, asphaltenes may cause many problems, especially in surface separation equipment. This situation is caused by the asphaltenes being emulsion-causing agents. See the discussions on emulsion stabilization in Sections 4.3.2 and 4.6.1. 6.3.3  Gas Hydrate Inhibitors. As noted in Section 4.3.4, clathrate gas hydrates can and do form in many pipeline segments. They are especially prevalent in subsea gathering and transit lines containing gas, water, and hydrocarbons. Because the hydrates are a crystalline class of materials (though there are a number of different structures and stoichiometries), some overlap on inhibition mechanisms with inorganic scales can be used. There are three different types of hydrate inhibitors in current use. The following subsections provide very short reviews. See Frenier et al. (2010) and the rather extensive discussions in Kelland (2013) for more details. Thermodynamic Inhibitor Agents. These chemicals function on the basis of the collegative properties of solutions. Chemicals that shift the thermodynamic equilibrium conditions that exist between hydrate, water, ice, and liquid structures are thus frequently called “thermodynamic inhibitors” (THIs). They act in the way “antifreeze” materials used in automotive and other industries do. Despite its being the predominant chemical class in commercial use, the large problem is the high volume needed. For severe hydrate conditions, more than 5% volume of the chemical, based on the water phase, may be needed. Commonly used THIs include • • • •

Methanol Ethanol Glycols Aqueous-phase salt

Note that the glycols include MEG, diethylene glycol (DEG), and triethylene glycol (TEG) and that the aqueous-phase salts are NaCl and CaCl2. Structures of the organic chemicals are shown in Table 6.4. These materials function by changing the chemical potential of the solution phase. If x is the mole concentration of the solute, the change in the in temperature of melting (T) by adding the solutes is −

dx ∆H = , ���������������������������������������������������������������������������������������������������������������������������� (6.7) dT RT 2

where DH is the melting heat of the pure solvent (Fink 2003). Hammerschmidt (1934, 1939) proposed the following formula to compute the shift in hydrate formation temperature caused by adding a thermodynamic inhibitor (see Table 6.4): ∆T =

Ks ,������������������������������������������������������������������������������������������������������ (6.8) ( M i / M w )(100 − Ci )

where DT = temperature drop with addition of a solute, °C, Ks = constant dependent on the nature of the inhibitor, Ci = inhibitor concentration in aqueous phase, wt%, Mi = molar mass of inhibitor, and Mw = molar mass of water.

278  Chemical and Mechanical Methods for Pipeline Integrity

Inhibitor

Structure

K

Methanol

2335

Ethanol

2335

Monoethylene glycol (MEG)

2700

Diethylene glycol (DEG)

4000

Triethylene glycol (TEG)

5400

Table 6.4—Commonly used thermodynamic hydrate inhibitors.

For associative inhibitors such as inorganic salts, a correction is made in Eq.6.9, where Mo is substituted for Mi (Mohammadi and Tohidi 2005). Here a is the degree of ionization and n is the number of ions from each salt. Mo =

M i ���������������������������������������������������������������������������������������������������������������������� (6.9) α (n − 1) + 1

The addition of a thermodynamic inhibitor such as salt, methanol, or ethylene glycol to the aqueous phase will shift the curve to lower temperatures (or higher pressures). This shifts the hydrate to the left (see Fig. 6.16). Note that the presence of THI materials such as methanol or a glycol in a pipeline fluid may affect the action of a corrosion inhibitor, and therefore its presence must be considered during product testing. As noted in this section, while very effective, high doses of the THI materials are required for protection from hydrate formation, more-efficient chemicals, the “low-dose hydrate inhibitors” (LDHIs), have been developed and are in use in some well/pipeline systems. The two types of LDHIs in current use are kinetic inhibitors (KHIs) and antiagglomerate (AA) materials. The two inhibitors types are described in the next two subsections. Low-Dose Kinetic Inhibitors. KHIs work at relatively low concentrations (compared with methanol or glycol) by delaying the nucleation and growth of the hydrate crystals. As the driving force becomes greater—that is, as the distance into the hydrate region increases—the time before the start of hydrate formation becomes less. Because the thermodynamic driving force is difficult to calculate, the amount of penetration into the hydrate zone is referred to in the literature as the amount of “subcooling” and is frequently designated as ∆T (not to be confused with the change in temperature, ∆T, in the hydrate formation temperature with the use of thermodynamic inhibitors).

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   279

Hydrates form No hydrates form

Aqueous phase Deionized water (DIW) DIW with 10% methanol

Live-gas condensate

DIW with 10% ethylene glycol

Fig. 6.16—Hydrate inhibition curves for deionized water (DIW) and methane (Notz et al. 1996).

To delay hydrate crystal nucleation and disrupt crystal growth, the inhibitors may contain functional groups designed to be similar to gas molecules. It was postulated by Talley and Oelfke (1999) that these kinetic inhibitors prevent hydrate crystal growth by becoming incorporated into the growing hydrate crystals, thereby disrupting further hydrate crystal growth. The growing hydrate crystals complete a cage by combining with the partial hydrate-like cages around the kinetic hydrate inhibitor moieties containing gas-like groups. These inhibitors are effective with or without the presence of a liquid hydrocarbon phase, but they are typically less effective in preventing the hydrate formation as the production pressure increases and the penetration into the hydrate envelope increases. Additional theories are considered later in this section. Examples of KHIs include poly(N-methylacrylamide), poly(N,N-dimethylacrylamide), poly(Nethylacrylamide), poly(N,N-diethylacrylamide), poly(N-methyl-N-vinylacetamide), poly(2-ethyloxazoline), poly(N-vinylpyrrolidone), and poly(N-vinylcaprolactam). These, also considered the “first generation of KHIs,” are all water-soluble polymers (Sloan 1990; Kelland et al. 1995) and can be used with high water cuts (up to 60%). The first generation of KHls was able to deliver approximately 15°F of subcooling. This limit increased to approximately 20°F for the second generation (Klomp et al. 1997). The latter limit appears to be close to the theoretical limit for most KHls, according to Mehta et al. (2003). Fig. 6.17 shows structures of two KHI monomers. Readers of the literature are cautioned to be careful of the nomenclature and whether °F or °C ∆T values are quoted. The largest issue to be considered is that although only low doses (>1,000 ppm) may be required, these are true inhibitors in that they only delay but do not prevent hydrate formation. Indeed, when the time delay that may extend to hours has elapsed, hydrate formation may be very rapid. Low-Dose Antiagglomeration Chemicals. The other major classes of hydrate inhibitors are the AAs. Mehta et al. (2003) reviewed the development of the AA-type materials. The concept of hydrate dispersants such as sulfosuccinates had been proposed by Sugier et al. (1990) as another means of achieving hydrate control. A class of chemicals that could disperse submillimeter hydrate crystals into an oil or condensate (liquid hydrocarbon) phase, was developed by Klomp et al. (1995). Examples of such dispersants are tributyldecylammonium, tripentyldecylammonium, tributyloleylammonium, and tributylhexadecylammonium cations. A generalized quaternary structure is presented in Fig. 6.18. Klomp et al. (1995) contended that instead of attempting to completely eliminate the nucleation of hydrates, the objective has shifted to preventing the formation and accumulation of large hydrate crystals into a composite hydrate blockage. The action of AA LDHI types depends on having a

280  Chemical and Mechanical Methods for Pipeline Integrity

O

N

vinylpyrrolidone

X-

R1

O N N-vinylcaprolactam

Fig. 6.17—Monomers of some polymeric KHIs.

+ R4

M

R2 R3

Diagram of quaternary compound M = N or P in most cases. R groups vary

Fig. 6.18—Quaternary compound general structure.

“hydratephilic” head incorporated within the hydrate crystals and a “hydratephobic” (or “oleophilic”) tail that disperses the hydrates into a liquid hydrocarbon phase (Mehta et al. 2003). The most powerful chemical, according to Mehta et al. (2003), are certain organic quaternary ammonium and phosphonium salts with one or two long oleophilic tails. Some of the AA-type LDHls are also known to provide some kinetic hydrate inhibition and are sometimes referred to as “hydrate growth inhibitors” (HGls). In other instances, however, the AAs can actually promote hydrate formation because their inhibition mechanism does not depend on stopping hydrate formation per se but rather preventing accumulation into a plug. Mehta et al. (2003) also show photographs of dispersed hydrates in a test cell. Mehta et al. (2003) also describe other characteristics of the AA inhibitors. They contend that the AAs work by emulsifying water in the hydrocarbon liquid. When hydrates form, they are carried as a nonagglomerated slurry in the hydrocarbon phase, with no increase in viscosity up to 50% water cuts. The water-cut limit to AAs is approximately 60%. In the field case study, the AAs effectively inhibited hydrate formation, with no significant downstream problems. Flowlines are continuously treated with LDHIs before shut-in, eliminating the need for instantaneous corrective actions on shutin (e.g., flowline depressurization). The volume reduction of inhibitor to be injected can be reduced by a factor of 25, allowing less topside storage space, easier transportation, and smaller umbilicals. AAs eliminate methanol discharge in overboard water and oil/gas export lines. Thus, most of the AAs developed to date act as hydrate dispersants that will allow hydrate crystals to disperse into a liquid hydrocarbon phase, instead of allowing the formation and accumulation of hydrate crystals into a blockage. The antiagglomerate properties of LDHI are caused by a “hydratephilic” head that is incorporated within the hydrate crystals and a “hydratephobic” tail that disperses the hydrates into a liquid hydrocarbon phase (Mehta et al. 2002). The first field test of AA LDHI for a subsea oil well in the Gulf of Mexico was reported by Frostman and Downs (2000). Kelland (2013) reviewed the use of AA-type hydrate inhibitors in pipelines and also describe chemicals such as those that form emulsions (sodium di-2-ethylhexylsulfosuccinate or polymeric surfactants). This author also provided detailed descriptions of the quaternary-type “hydratephilic” chemicals. The chemical structures of several large quats proposed as pipeline AA inhibitors are noted in this book. A reference (Delgado-Linares et al. 2013) provides a model for these types of hydrate control systems. 6.3.4 Calcium/Sodium Naphthenate Inhibition. The formation of calcium naphthenate solids appears to be a growing problem in some areas of the petroleum production environment, and so methods to stop or “inhibit” this type of solid is the subject of current research and development. The sodium salts of naphthenates can also cause emulsions (see Section 4.3.3). Because naphthenate soaps or calcium naphthenate solids will not form unless the pH of the fluid is greater than approximately 6, the addition of various acids to the brine has been a major treatment method. Maintaining pressure to keep CO2 in solution also is an effective prevention method, but doing this may be difficult to achieve in practice. Inhibition of calcium or sodium soap scales with naphthenates has also entailed the addition of demulsifiers. Because this is specific to the particular crude oil and there are a very large number of demulsifiers in use, readers are directed to the discussions of the

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   281

many types of chemicals in Fink (2003) and Kelland (2009) as well as the general discussion of emulsions in Becker (1997). The use of acidic materials for pH control and thus reduction of calcium naphthenate scales was reviewed by Turner and Smith (2005). The addition of glacial acetic acid (99.9%) to decrease the pH of the water to the value at which the carboxylate groups were protonated has been shown to reduce the mass of solids formed. Phosphoric acid, as well as other organic acids such as sulfonic acids and muriatic acid (a C-14 acid), has also been used with some effectiveness. Additional acids were mentioned at a workshop on naphthenate scales that include glycolic acid as well as formic acid (Techbits 2008). These materials are stronger acids (lower pKa values) than glacial acetic acid and may be more effective. However, all these acids are volatile and may cause corrosion in vapor spaces or corrosion of other ferrous equipment. Vindstad et al. (2003) reported that addition of HCl at a concentration of 200 to 400 mg/L had been found to completely eliminate naphthenate fouling in the Heidrun field, off the coast of Norway, but had the unexpected consequence of increasing the water content of the produced oil, so these producers have also had to use dispersants and inhibitors (compositions not described). These authors did not specifically mention corrosion of ferrous equipment, but this will be an issue unless corrosion inhibitors are added to the acids or the pH control is precise. Recently, additional materials for control of naphthenate deposits have been proposed. Hurtevent and Ubbels (2006) claimed that low-dose naphthenate inhibitors can be used in place of acetic acid, and they provided laboratory and field data. These inhibitors are based on demulsifier chemistry, but details were not provided in this report. However, a US patent application may provide some of the chemistry. In this application, Ubbels (2005) claimed the use of surfactants selected from phosphates, sulfates, sulfonates, sulfosuccinates, polysulfosuccinates, phenols, betaines, thiocarbamates, xanthates, and combinations as applicable as naphthenate formation inhibitors. The inhibitors are added at 250 to 500 ppm, and the inhibitor composition may include injecting the inhibitor composition downhole, dosing the inhibitor composition to an oil/water separator, or providing the inhibitor composition at another desirable point. As another option, the inhibitor composition may be added as the oil/water mixture is being produced from a formation. Preferably, the inhibitor composition is added before a choke, manifold, turret, or a combination thereof. The method may include reducing the pressure after adding the inhibitor composition; this will release carbon dioxide gas from the oil. The method may also include reducing the pressure to release carbon dioxide gas from the mixture before adding the inhibitor composition. 6.4  Flow Enhancement, Biocides, and Oxygen Scavengers Demulsifiers, foaming agents, and antifoam chemicals and flow enhancers are surface-active chemicals that orient themselves at fluid, gas, or solid surfaces. Their function is to change the surface energies and effect a desired difference between two phases. Biocides are added to pipelines principally to reduce MIC (see Section 3.1.7), and oxygen scavengers are used to reduce corrosion caused by this introduced gas. Short reviews of these chemical types follow. See Kelland (2009, 2013) and Frenier and Ziauddin (2014) for more details. 6.4.1  Demulsifiers, Foaming Agents, and Defoamers. Highly surface-active chemicals (Section 1.4.5) such as demulsifiers, foaming agents, and defoamers are used in pipelines as well as in facilities to help maintain flow and assist in cleaning operations (Section 7.5.4) and placement of inhibitors (Section 6.7.2). Demulsifiers. An extensive number of demulsifiers are being used in the production phase of the petroleum industry to help resolve/prevent emulsions. Emulsions must be broken to allow efficient separation of the oil, gas, and aqueous phases before further purification (see Section 2.1.5 and 2.3.2). Kokal (2006) claimed that the only clear generalization regarding demulsifiers is that they mostly have a high molecular weight (approximately the same as natural surfactants) and, when used as

282  Chemical and Mechanical Methods for Pipeline Integrity

demulsifying agents, they tend to establish an emulsion opposite in type to that stabilized by natural Film surfactants. This author showed Oil drainage a drawing (Fig. 6.19) in which the demulsifiers displace the Water natural stabilizers present in the Interfacial film interfacial film around the water droplets. This displacement is Demulsifier brought about by the adsorption of the demulsifier at the interface Natural surfactant and influences the coalescence of water droplets through enhanced Fig. 6.19—Demulsifier mechanism (Kokal 2006). film drainage. The structure of the chemical must be such that it can compete with the emulsifying chemicals, but it must also destabilize the film so that gravity will start the draining process. This requires a specific chemical characteristic that depends on the oil chemistry as well as the water chemistry. Thus, there is the difficulty of finding universal demulsifiers. Fink (2003, 2011) and Kelland (2009, 2013) reviewed in main sections in their books the large number of demulsifier chemicals proposed or in use. Principal divisions of the types are small anionic surfactants such as dodecylbenzyl sulfonic acid (DDBSA) as well as polymers made from polyethylene or polypropylene oxide. Fink (2003) also provided tables that list materials recommended for water-in-oil (W/O) emulsions and those that may work with oil-in-water (O/W) emulsions. Materials listed by these authors include Water

• • • • • • • • • • •

Polyalkoxylate block copolymers and ester derivatives Alkylphenol-aldehyde resin alkoxylates Polyalkoxylate polyamines Vinyl polymers Polysilicones as boosters (which are also antifoaming agents) Cationic amide-ether esters Polyamines Polyamides Alkoxylated fatty oils Biopolymers Nonionic and anionic surfactants

The structures of several small-molecule demulsifiers are shown in Fig. 6.20. Testing methods for determining the most effective material are in Kokal (2006) and Kelland (2013). Note that the choice of a chemical is specific to the oil/water qualities; thus, on-site samples are required. Foaming Agents and Defoamers. Foams are dispersions of gases in a liquid (see Figs. 1.16 and 1.17 for a description of forces at a surface). Foams are similar to the emulsions described in this section, but the inner phase is a gas. See Fig. 4.30 for a microscopic view of foam as it settles. Various natural chemicals in a pipeline/facility can stabilize a foam, or one can add an artificial surfactant to cause a stable foam to form. Foaming agents stabilize the surface. Examples of materials that can produce a high-temperature foam are given in Table 6.5. Useful foams, applied for cleaning pipelines, are described in Section 7.5.4.

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   283

O HO

p-Dodecylbenzylsulfonic acid

S O

Na+

O OO S Na+

-

O

O -

O

O

Na+

Sodium bis(2-ethylhexyl)sulfosuccinate

HO

(CH2 CH2-O)9

Nonylphenol 9EO Fig. 6.20—Small-molecule demulsifiers.

Unwanted foams require the use of defoamers that may be injected at locations where foam may occur. Foams can also form in facilities where there are high-turbulence areas, which are particularly susceptible to foam (Section 2.3.2). Defoamers such as silicones and polyglycols (Kelland 2009, 2013) help the foam to collapse. Oilfieldwiki (2013) claimed that the following types of defoamers are in use:

Surfactant

Type

A

Amphoteric alkyl amine

B

Blend of ethoxylated alcohols

C

Blend of ammonium alcohol, ethoxysulfate, and ethoxylated alcohols

D

Formulated amphoteric alkyl amine

Table 6.5—Surfactants for hydrostatic foam tests (Kam et al. 2007).

• Oil-based defoamers that have an oil carrier. The oil might be mineral oil, vegetable oil, white oil, or any other oil that is insoluble in the foaming medium except silicone oil.

284  Chemical and Mechanical Methods for Pipeline Integrity

• •

•

•

•

An oil-based defoamer may also contain a wax and/or hydrophobic silica to boost the performance. Powder defoamers are oil-based defoamers on a particulate carrier such as silica. Water-based defoamers use different types of oils and waxes dispersed in a water base as well as water-soluble surfactants that impede foam formation. The surfactant defoamers may be long-chain fatty alcohol or fatty acid soaps or esters. Methanol has been used in some formulations, but it is toxic and flammable. Silicone-based defoamers have a silicone compound as the active component. These might be delivered as an oil- or a water-based emulsion. The silicone compound consists of a hydrophobic silica dispersed in a silicone oil. Emulsifiers are added to ensure that the silicone spreads fast and well in the foaming medium. The silicone compound might also contain silicone glycols and other modified silicone fluids. Ethylene oxide/propylene oxide–based defoamers contain polyethylene glycol and polypropylene glycol copolymers. They are delivered as oils, water solutions, or water-based emulsions. These copolymers normally have good dispersing properties and are often well suited when deposit problems are an issue. Alkyl polyacrylates–based defoamers.

See Al-Qahtani and Garland (2013) for an example of use of a defoamer in a gas treatment plant. 6.4.2  Flow Enhancers: Drag-Reducing Agents. A large pressure drop in a flowing pipeline can result in increased expenditures to lift/pump the oil. This may necessitate upgraded pump equipment or reduced throughput. Thus, the need to reduce the drag—that is, the pressure drop, ∆P, in a segment of piping (see Eqs. 1.8 through 1.10)—has been investigated. Chemicals as well as engineering methods have been applied extensively in the pipeline industry to reduce drag. Langsholt (2012) and Toms (1948) reported on the use of drag-reducing agents (DRAs). This section reviews chemicals used in reducing drag during pipeline activities; however, see Kelland (2013, Chap.17) for a much more comprehensive treatment of this subject. Chemicals that affect the viscosity or the drag of flowing hydrocarbon or emulsions are in use to reduce the pressure drop and thus increase flow rate of various fluids. One approach for reducing drag is to reduce oil viscosity. The transportation of crude oil from surface tar sands is facilitated by mixing the product with diesel fuel (Larsen 2012). Fan and Shafie (2013) proposed using derivatives of turpentine as “greener” viscosity reducers for heavy crude oils and bitumen. This disclosure provides a method of reducing drag, reducing friction, reducing viscosity, and/or improving flow of viscous hydrocarbons. The method can include the steps of introducing an effective amount of a drag-reducing composition containing a blend of turpentine liquids. In one embodiment, the blend of turpentine liquids includes a-terpineol, b-terpineol, b-pinene, and p-cymene. These materials are similar to the solvents shown in Section 7.3.5. Also, heavy oil can be heated to lower viscosities, but this method is not practical for all situations. Drag-Reducing Chemicals. In addition to changing oil viscosity, specific chemicals may modify the internal drag generated by turbulent flow in crude oil pipelines. However, this application may be limited to systems operating under turbulent as opposed to laminar conditions. Thus, constraints on flow type also prevent drag reducers from being used with most viscous heavy oils, which tend to experience laminar flow. Despite these limitations, DRAs are in wide use because they can frequently be used at low levels (10 to100 ppm) and can be economical alternatives to mechanical solutions (i.e., more pumps or larger pipe diameters). Lee and Brown (2013) claimed that DRAs have the following uses, depending on each application: • Pipeline flow (capacity) increase • Throughput assurance during maintenance periods, for example

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   285

• • • •

Capital avoidance (i.e., smaller diameter, fewer pump stations) Increased oil production Reduced tanker turnaround time Power cost saving

Frictional Pressure Drop, DP

This section reviews known chemistries and mechanisms of operation of DRA chemicals. As noted, Kelland (2013) devoted an entire chapter to DRAs that are used in various oilfield applications. Fink (2011) also included a chapter that contains many structures and references. Examples described by these authors include oil-soluble polymers such as polyalkylenes and polyolefins that are used to reduce the drag in various pipelines. In addition, 2-hydroxy phosphonoacetic acid (HPAA) materials and water-soluble drag-reducing polymers include polyethyleneoxides, polyvinyl alcohols, and various polysaccharides such as the guar. In addition, these authors proposed using critical micelle concentration and hydroxyethyl cellulose. Surfactants such as some of the sulfonates (Rassoul and Hadi 2007), as well as betaines, can also act as DRAs. DRAs target specific components of the crude oil, or they target specific flow characteristics of the system to improve flow conditions in upstream pipelines. A representation of the effects of DRAs and turbulence on the frictional pressure loss (∆P) is shown in Fig. 6.21. Langsholt (2012) claims that much of the current understanding of the mechanisms of DRAs depends on the work of Virk (1975), who introduced the concept of an elastic sublayer between the viscous sublayer and the outer Newtonian region. According to Langsholt (2012) as well as Mohitpour et al. (2010), this concept still seems to be generally accepted, and it is claimed that Laminar Turbulent flow the polymeric DRA reduces flow Without DRA the shear stresses and wallnormal velocity fluctuations. The interpretation of this is With DRA that turbulent bursts interact with the coiled polymer molecules in such a way that they Projection of are stretched, and so absorb or laminar dissipate the energy, reducflow ing the turbulent momentum transfer. See a depiction Flow Rate of the effect in Fig. 6.22 (Kelland 2013). Fig. 6.21—Effects of turbulence and DRAs on ∆r. DRA Injection

Laminar sublayer Buffer region

Turbulent core

Fig. 6.22—Drug-reducing mechanism (adapted from Kelland 2009).

0.3 mm

1.2 mm

150 mm

286  Chemical and Mechanical Methods for Pipeline Integrity

A simple equation from Rassoul and Hadi (2007) has been used to describe the DRA (DR) effect: % DR =

∆Pb − ∆Pa ⋅ 100.���������������������������������������������������������������������������������������������������������� (6.10) ∆Pb

Another analysis has been used by Langsholt (2012). In this approach, a plot of the friction factor vs.Reynolds number (Re) obtained in single-phase pipe flow is presented in so-called PrandtlKarman coordinates and takes the form f −0.5 = C1 log(Re s f 0.5 ) + C0 .���������������������������������������������������������������������������������������������������� (6.11) For turbulent flow of a Newtonian fluid in a smooth pipe, Langsholt (2012) claimed that C1 ≈ 2.0 and C0 ≈ −0.8, whereas for a DRA-fluid, that follows the Virk (1975) maximum drag reduction (MDR) asymptote, where the coefficients are C1 = 9.5 and C0 = −19.0. In a plot of a partially degraded solution of a polymer, the coefficient values normally fall between the Newtonian and the MDR values. A Prandtl-Karman plot for a 10-ppm solution degraded in the 60-mm pipe at three different (successively higher) flow rates is shown in Fig. 6.23. The figure shows that the gradually higher shear rates, applied during the partial degradation process, bring the behavior closer to the non-DRA values (pure solvent). More details of degradation are described later in this section. The constraints on flow type also prevent drag reducers from being used with most viscous heavy oils, which tend to experience laminar flow. Paraffin wax inhibitors modify wax crystal structure to lower the pour point or wax appearance temperature (wax crystal modifiers) and/or interfere with the flocculation of wax crystals (wax dispersants). See Section 6.4 of Frenier et al. (2010). Similarly, asphaltene inhibitors prevent asphaltene deposition by interacting with the asphaltene particles to delay flocculation, a process that will slow and potentially prevent precipitation and deposition. Examples of DRAs containing large polyalkyl polymers, HPAAs, and surfactants follow. Johnston and Fry (1994) and Labude et al. (2002) described DRAs that are high-molecular-weightpolyolefin ultrahigh-molecular-weight polymer. An example is a linear poly(a-olefin) composed of monomers with a carbon chain length of between 4 and 20 carbons or 10 mixtures of two or more such linear poly(a-olefins). Specific examples of these linear poly(a-olefins) include poly(l-octene), poly(l-nonene) and poly(l-decene). Polyisobutylene is also mentioned. The ultrahigh-molecular-weight polymer may also be a copolymer, such as a polymer composed of two or more of 15 types of monomers, as long as all monomers used have a carbon chain length of between 4 and 20 carbons. 12.000

Pure solvent 10 ppm_1.0 m/s 10 ppm_1.8 m/s 10 ppm_2.5 m/s MDR asymptote

MDR : y = 9.5x - 19

11.000

f -0.5

10.000 9.000 8.000

10 ppm_2.5 m/s : y = 3.4x - 6.3 10 ppm_1.8 m/s : y = 4.5x - 9.5 10 ppm_1.0 m/s : y = 6.4x - 15

7.000 Solvent : y=1.94x - 0.74

6.000 5.000 3.000

3.200

3.400

3.600

3.800

4.000

4.200

4.400

log(Res f 0.5) Fig. 6.23—Friction factors for 10-ppm polymeric DRA at three levels of degradation (Langsholt 2012).

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   287

The chief problem with these * CH2 CH * polymers is that they are difficult to dissolve in crude oil. Thus, the technologies in these two patents O C include freeze drying or grindn ing the polymers to small size and then dissolving in a solvent such as NH2 diesel oil or an emulsion that will disperse in crude oil. Polyacrylamide repeat unit. For HPAA, some groups are hydrolyzed to carboxylate (-COOH) groups. A DRA process that uses 200 ppm of an HPAA material, as shown in Fig. 6.24—repeat unit for HPAA. Fig. 6.24, was claimed by Kang and Jepson (2000) to be effective in reducing the average pressure drop significantly for superficial liquid/ gas velocities in both slug and annular flow. The test medium was oil with a viscosity of 2.5 cp, and the gas phase was CO2. Average pressure-drop reduction of up to 82% for slug flow and 47% for annular flow was achieved. For slug flow, the effectiveness generally decreased with increase in superficial liquid velocity. The maximum pressure drop across the slug also reduced with addition of DRA. The effectiveness for maximum pressure drop was more than 28% at a superficial liquid velocity of 1.5 m/s for all gas velocities. Kang and Jepson (2000) also claimed that at superficial liquid/gas velocities of 0.5 and 2 m/s, the flow pattern changed from slug flow to stratified flow by decreasing the slug frequency to 0 slug/min when 50-ppm DRA was added. At higher liquid/gas velocities, the slug frequency also decreased significantly. The effective height of the liquid film decreased with addition of DRA. Hénaut et al. (2012) also studied the use and degradation of HPAA DRA materials. Drag reduction and degradation studies by Jouenne et al. (2015) found that the level of drag reduction (Eq. 6.10) increases with fluid velocity and decreases with polymer concentration of an HPAA solution. The concentration effect vanishes at high velocities. A general trend is obtained when the drag reduction is plotted as a function of the generalized Reynolds number. Jouenne et al. (2015) also found that the critical velocity at which mechanical degradation occurs in turbulent flow increases with the pipe diameter. They also concluded that design guidelines for water transport will suffice for polymer. A water injection network can be converted to a polymer injection network for tertiary enhanced-oil-recovery (EOR) polymer projects. Some surfactants also can be applied as DRA chemicals. Rassoul and Hadi (2007) claimed that O sodium p-dodecylbenzyl sulfonic S acid (SDSA) is useful (Fig. 6.25). -O Sodium p-dodecylbenzenesulfonic acid Using an Iraqi crude oil, the Na+ O authors found that 250-ppm SDSA Fig. 6.25—A surfactant DRA chemical. produced a 54% reduction on drag using pipe flow tests. Additional Chemical Drag-Reducing Agents. Demulsifier chemicals, which disrupt interfacial oil/water films facilitating droplet coalescence, are generally used to break W/O emulsions, but if applied properly they may increase the droplet size of the water internal phase, reducing the apparent viscosity of the produced emulsion. Faust and Weathers (2011) reported that when an O/W emulsion is formed, the viscosity of the fluid is reduced and the flow is enhanced. They found a pair of two polymeric surfactants that made this type of emulsion using Canadian heavy oil. They also reported that the droplet size of the oil phases is approximately 50 µm and that the emulsion is stable only under high-shear flowing conditions and will resolve when the fluid remains still.

288  Chemical and Mechanical Methods for Pipeline Integrity

Faust and Weathers (2011) described biphasic viscosity reducer chemicals that target the bulk fluid properties of crude oil, regardless of the source of viscosity, by dispersing oil into free water, creating a highly flowable water-external emulsion of low apparent viscosity. Screening tests described by these authors confirmed the capacity of certain polymers to emulsify heavy oils, with gravities well below 20 °API, as well as waxy crudes from different locations around the world into 20 to 25% water solutions, creating stable, water-external emulsions. They also claimed that in all cases the emulsion exhibited significant levels of apparent-viscosity reduction, generating improved flowability in a bench-top flow loop, as well as emulsion resolution under standard field separation conditions, including heat and traditional emulsion-breaking chemicals. In addition, Faust and Weathers (2011) claimed that solution degradation is a process that depends on several parameters: the maximum shear that the fluid is exposed to, the duration of the exposure, and the type and concentration of the DRA. In their study, the fluid was circulated in a closed loop, the concentration was known, and the flow rate, or shear, could be controlled. By increasing the flow rate—for example, by exposing the fluid to higher shear—the degradation continued, before a new stable situation emerges. This systematic, shear-dependent degradation is illustrated schematically in Fig. 6.26. Note that during the study, the solution was degraded in one step, representing a shear rate higher than the maximum shear the fluid will be exposed to during the experimental campaign. This section has described DRA chemicals. Another flow-related condition that especially affects multiphase pipeline conditions is slug flow (see a description in Section 1.6.2 and Fig. 1.8). This type of flow condition may cause corrosion damage if solids are being transported (Palacios and Quintero 2003). An invention by Ramachandran et al. (2010) described an injection process (see Section 6.7.1) and the use of surfactants that are designed to reduce the surface tension of the fluid phase; the agent then converts a two-phase flow to a foam that behaves as a more viscous single-phase flow. Section 6.1.5 described the interaction of various chemicals used to treat different pipeline problems; it was noted that scale and corrosion inhibitors can interact in a synergistic or antagonistic manner (Silvestri et al. 2010). Kelland (2013) described articles that note that DRAs can also affect corrosion inhibition in various ways in different pipeline-operating flow regions.

dP/dx (pressure gradient) V (velocity)

6.4.3 Biocides. Biocides are added to well/pipeline fluids to protect important chemicals such as polymers from degradation and also to kill SRB-type microorganisms (see Section  3.1.7). Such materials can be considered to be corrosion control agents by retarding MIC-forming organisms. Two different types of biocides are described by Kelland (2013) and McIlwaine (2005) The first class of chemicals includes oxidizing chemicals such as chlorine, chlorine dioxide, bromine chloride, and various peroxides. These materials are extremely effective Water-soluble acrylate polymer in oil emulsion for killing microorganisms and may be used to sterilize injection V water or other water sources. The nonoxidizing biocides include aldehydes (glutaraldehyde), chlorophosonates, and quaternary amines and phosphoniums. dP/dx Oxidizing Agents. Chlorine and chlorine dioxide are in wide Time use to disinfect many water Adding DRA sources, among them swimming pools and possibly some oilfield Fig. 6.26—Polyacrylic acid (PAA) degradation vs. velocity (Faust waters. Because these chemicals and Weathers 2011).

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   289

themselves may be corrosive and dangerous to handle, they may not be suitable for direct pipeline activities. Carpenter and Nalepa (2005) reviewed information on the family of bromine-based biocides and discussed the basics of their chemistries and properties. They focused on a new stabilized bromine chloride– based biocide that has been demonstrated in a slickwater fracturing fluid field trial (see Eq. 6.12). BrCl + H 2 O ⇔ HOBr + HCl . �������������������������������������������������������������������������������������������� (6.12) Here, excellent microbiological control was claimed to be maintained cost-effectively with relatively low biocide residuals. These oxidizing materials kill the bacteria by disrupting their cell membranes. Some of these materials may also act as surfactants. Note that comparability with the well fluids as well as with environmental laws must be considered. Abraham (2014) and Rovison et al. (2010) suggested the use of peracetic acid (PAA) as a microbiological treatment agent and for clarifying water. The reaction for the production of PAA is O +

HO

HO

OH

O Acetic acid

Hydrogen peroxide

HO

O

Peracetic acid

+ H

O

H

. �������������������������������� (6.13)

Water

As this reaction shows, the final products will contain different amounts of all four chemicals in the mixture, depending on the reaction kinetics. Note that various peroxide-type chemicals have long usage in well stimulation operations. See Section 4.4 of Frenier and Ziauddin (2014). Nonoxidizing Biocides. Also in current use are a large variety of materials that affect microbes in different ways when compared with oxidizers (McIlwaine 2005). Kaufman et al. (2008) described tetrakis-hydroxylmethylphosphonium sulfate (THPS) as tetrahydro- 3,5-dimethyl-1,3,5-thiadiazinane2-thione. This biocide, “thione,” has been shown not to interfere with friction reducer and is extremely effective in killing acid-producing bacteria as well as SRB. It is claimed to be a broad spectrum biocide. Powell et al. (2010) noted that THPS has been used successfully to help renovate a damaged subsea pipeline. Walsh (2014) reviewed treatment methods of water sources (specifically from unconventional plays) to reduce the impact of various microbiological organisms. This author described many of the chemicals listed by Kelland (2009) and Kaufman et al. (2008), as well as 2,2-dibromo-3nitrilopropionamide (DBNPA) and 2-bromo-2-nitropropane-1,3-diol (bronopol). Walsh (2014) also produced useful tables for comparing the various types of materials. Fig. 6.27a shows a table of specific characteristics of different families of chemicals, all assumed to kill the microbes; Fig.6.27b compares cost, efficiencies, and ease of operations of these. The higher numbers are more favorable. The author of this book notes that the oxidizing biocides (air, chlorine, hypochlorite, chlorine dioxide. and bromine chloride) can be used only to pretreat water before entering an iron-based pipeline, where all oxidizing chemicals must be absent to reduce corrosion. Walsh (2014) lists additional references to water treatments that may apply to pipelines. Nitrates have also been used as a biocide for SRB because they accelerate the production of nitratereducing bacteria, which are not linked directly with MIC. Thorstenson et al. (2002) described tests comparing glutaraldehyde with nitrate injection to prevent formation souring (and thus production of H2S). These authors concluded that nitrate was more effective than glutaraldehyde and thus that this was an effective method in the reservoirs being treated.They found that • Nitrate injection led to reduction in SRB numbers and activity and a concomitant enrichment of nitrate-reducing bacteria.

290  Chemical and Mechanical Methods for Pipeline Integrity

a) Comparison of some characteristics of biocides Characteristic

Acrolein

Quaternary Ammonium

Glutaraldehyde

Chlorine

Hypochlorite

Chlorine dioxide

Reacts with H2S

Yes

No

No

No

Yes

Yes

Penetrates biofilm

No

No

No

No

No

Yes

Converts ferrous iron to hydrogen ferric oxide

No

No

No

Yes

Yes

Yes

Human toxicity concerns

High

High

Medium

High

Medium

Low

Environmental concerns

High

High

Medium

High

Medium

Low

b) Comparison of some selection factors for biocides Type

Cost

Efficiency

Operations

Aeration

3

1

5

Hypochlorite

5

4

4

Chlorine dioxide

1

5

3

QACs

2

1

2

Aldehydes

2

4

1

Bronopol

2

4

2

DBNPA

2

3

1

Fig. 6.27—Comparison of biocides (Walsh 2014).

• The effect of nitrate injection was evident 4 months after start of injection. • Corrosion measurements on metal coupons showed a decrease in weight loss from 0.7 mm/yr before nitrate addition to 0.2 mm/yr with nitrate treatment. Note that nitrate has not been tested as a direct-injection chemical in pipelines. See further discussion of MIC in Section 3.1.7. Jenneman et al. (2010) and Greene et al. (2006) claimed that SRB can be inhibited more efficiently, thus leading to less MIC, if nonoxidizing biocides and a metabolic inhibitor that impedes secondary regrowth of the microorganisms are used in combination. The authors list types of biocides as aldehydes (e.g., formaldehyde, glutaraldehyde, and acrolein), amine-type compounds (e.g., quaternary amine compounds and cocodiamine), halogenated compounds (e.g., bromopol and DBNPA), sulfur compounds (e.g., isothiazolone and carbamates), and quaternary phosphonium salts (e. g., THPS). Examples of metabolic inhibitors listed include nitrite, molybdate, tungstate, selenate, and anthraquinone compounds. The authors claimed that the preferred combination is glutaraldehyde and small amounts of a nitrite compound. Note that Voordouw et al. (2004) also described both nitrates and nitrites as acting to reduce reservoir souring. The uses in pipeline operations have not been documented in the public literature. Because all these materials are designed to kill or disrupt various microorganisms, it may be difficult to prove that they can pass the various tests to allow use in various locations that have strict testing requirements (see Section 8.3 for more information). Sections of Frenier (2017a) give details of various test methods. 6.4.4  Oxygen Scavengers. Oxygen scavengers are materials used to protect pipelines and other equipment in facilities from corrosion attack by O2. Various reducing agents, especially sulfites, bisulfites, and metabisulfites, are in current use to control the amount of O2 in a fluid. Here, the reaction of a bisulfite is 2HSO−3 + O 2 → 2HSO−4.���������������������������������������������������������������������������������������������������������� (6.14)

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   291

OH Note that this reaction is O Na+ NH2 H2 N slow at low temperatures N -O S (15,000 ppm) reduced the breakthrough time significantly. A likely mechanism is the reduction in fluid pH, which would shift the equilibrium, shown in Fig. 6.35 to the left. As noted in the first paragraph of this section, many of the reaction products are insoluble in the fluids and may foul the line or other equipment. Horton et al. (2010) state that when using conventional triaziane scavenging treatments, care must be taken that a sufficient amount of the scavenger is injected into a well/pipeline. If there is an insufficient amount of the triaziane chemical injected for the amount of H2S, these authors claimed, the dithiazine (Fig. 6.35) may react with more H2S to form trithiane (Fig. 6.29), which is a sour-smelling white solid. The author of this book has observed this reaction with formaldehyde and H2S. To prevent this solids reaction, Horton et al. (2010) claimed, there is a process that provides a preflush into the well with ammonia or an amine before the thiazine. This is performed to remove a portion of excess H2S to prevent the solids-forming reaction. As noted above, some of the simple aldehydes produce a “more” soluble product, but these are also more expensive. 6.6  Chemistry for Producing Pipeline Gels When there are many intrusions into a line or the pipeline segment to be maintained is very long or mechanical pigs may not be practical, another option is a viscous fluid called a gel or “gelly pig.” These materials consist of a length of gelled oil or water that fills a specific volume of a pipeline. These terms are confused in the nomenclature, but this book designates a “gel” as an uncrosslinked but thickened fluid, whereas a gelled (or gelly) pig is a much thicker crosslinked fluid. They can be used to apply treatment chemicals (Section 6.7) and for cleaning pipelines (Section 7.5.3).

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   295

The uses of chemical gels/gelly include • • • • • •

Batching or fluid separation Debris pickup and cleaning (see Section 7.5.3) Application of chemicals such as inhibitors and biocides (Sections 6.7.3 and 7.5.3) Dehydration of lines Condensate removal Heavy chemical cleaning (Section 7.5)

This section provides information on the chemistry of formation of gel/gelly pigs. Because they are liquid, though highly viscous, gels can be pumped through any line that will accept liquids, although they may have high K values (see Eq. 5.1). Gel/gelly pigs can be used alone in liquid lines, in place of batching pigs, or in conjunction with various conventional pigs. When used with conventional pigs, gelled pigs can improve overall performance while almost eliminating the risk of sticking a pig. Gel pigs do not wear out in service as do conventional pigs. Note, however, that the fluids can be susceptible to dilution and gas cutting or degradation resulting from shear forces on the polymer backbone. Care must be taken, therefore, when designing a pig train that incorporates gel pigs to minimize fluid bypass of the pigs; a conventional pig should be placed at the back of the train when displacing with gas, according to PetroMin Hydrocarbon Asia (PM 2013). This source also claimed that gel pigs have been used to temporally seal valves during hydrotesting. Fig. 6.36 shows a cartoon of a gelled (polymer fluid) pig removing debris by using the technologies of Purinton (1984, 1985) and Purinton and Mitchell (1987). Note in this figure that a mechanical (brush) pig is also used. The polymer (gelled) pigs are similar to a thick fracturing fluid that may include gelled water, or gelled oil such as diesel fluid. Using the nomenclature described above, a polymer is mixed with water or oil to produce an uncrosslinked gel that has increased viscosity. In a process that is similar to the formation of fracturing fluids—see Frenier (2014, Chap. 4) for details of many formulations—these gels, called “pickup” or “debris transport” gels, may be formed on the fly. This is accomplished by mixing a polymer solution (such as a guar, hydroxyethyl cellulose, or xanthan). If a much thicker gelled fluid is needed, the polymers may be mixed with a crosslinking agent. For some polymers this is a borate compound at high pH. By varying the amount of polymer and crosslinker, the viscosity and debris-suspending characteristics can be controlled. See chemical structures of guar and xanthan in Figs. 6.37 and 6.37b. The crosslink reactions are ligand exchange processes. One such generic reaction is 2[R car − (OH)2 ] + ML 4 = M(O−2 − R car )2 + 4H+. ���������������������������������������������������������������������� (6.15) Here, R car − (OH)2 represents the carbohydrate polymer and L is a ligand associated with the ­metallic crosslinker. B R U S H

FLOW

P I G

“Turbulent” pick-up region (shear thinning effect)

Laminar concentrate region suspends debris

Plug flow transport region moves debris forward, away from trailing pig

Fig. 6.36—Gelly pig for debris removal (Frenier 2001).

296  Chemical and Mechanical Methods for Pipeline Integrity

a) Guar repeat unit

c) Polyacrylamide repeat unit b) Xanthan repeat unit

Fig. 6.37—Gel chemicals: (a) guar, (b) xanthan, and (c) polyacrylamide.

An equilibrium coefficient, Keq, can describe the extent of the reaction, and a kinetic coefficient can describe the rate of crosslinking. Many factors can affect both of these properties. These include the relative equilibrium coefficient for each polymer and metal. In addition, the rate at which the reaction occurs depends on the individual reactions, concentrations of reactants, and temperatures. Usually, a high pH drives the reaction to the right as the protons are reacted to form water (Eq. 6.15). However, low-pH crosslinking in the presence of CO2 is possible if some Ti or Zr compounds are used. Gels can also be made using various partially hydrolyzed polyacrylamide chemicals (Fig. 6.37c). These polymers can be crosslinked using metals such as Fe3+ or Cr3+ ions. Initially, the alcohol groups are protonated and must lose the proton to form the bond with the boron ion. Because boron forms relatively weak coordinate bonds with the sugar, the pH must be raised before the metal can react. Titanium and zirconium form 6-coordinate octahedral complexes with some ligands in solution, and boron forms 4-coordinate tetrahedral complexes. During the flow of a gelled pig, there is a high shear in the turbulent section near the walls (Fig. 6.36); there is also an area of laminar flow in the center that helps to suspend solids removed from the pipe walls. These fluids are complex and will follow non-Newtonian behaviors that make them very useful is some pipeline applications (more details of the rheology of these materials are found in Section 7.5). At the end of a job, the gelled fluid must be “broken” so that the viscosity is reduced to enable the fluid to be processed for disposal. For the guar-based fluids that are gelled with sodium borate, addition of an acid to reduce the pH may achieve the viscosity reduction. Breakers that are used in hydraulic fracturing gels are described in Frenier and Ziauddin (2014) and Gulbis and Hodge (2001). A diagram/photo from Mackay (2013) of various possible gel structures is shown in Fig. 6.38. Fig. 6.38a shows a linear gel structure; Fig. 6.38b, different gel mixes; and Fig. 6.38c, a crosslinked gelled structure.

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   297

a) Linear gel structure

b) Gels showing high to low viscosities

c) Crosslinked gelled structure

Fig. 6.38—Various gel structures (Mackay 2013).

Gelled-oil pigs have been used and are formed using a specific phosphate ester with petroleum oil (see Gulbis and Hodge 2001; Frenier and Ziauddin 2014) .The Pipeline and Pigging Service Association (PPSA 2008a) noted that because of the flexibility of the gels, these materials may also be used to dislodge stuck pigs. Abney and Browne (2006) show the use of solid pigs with gelled pigs for debris removal in a process by which the solid pig removes the deposits and the gel acts to suspend the particles. Akhiyarov (2007) described gel processes that have also other applications in pipeline system. These include • • • • •

Isolation (sealing) of pipeline section under pressure Carryover of gas/water accumulations Extraction of stuck pigging devices Series fluids separation Chemical treatment of wall surface

More details of the use of gel/gelled pigs in transporting corrosion inhibitors and other chemicals are presented in Section 6.6, and renovation of highly fouled pipelines using gels is described in Section 7.5. Note that many of these applications are also used in conjunction with solid (or polymer) pigs in complicated pig trains. 6.7  Fluid Additive Injection Methods and Equipment This section describes methods for injecting chemicals into pipeline systems as well as strategies for selecting injection points and methods of injection. Treating chemicals such as corrosion inhibitors and scale inhibitors, described in Sections 6.1 through 6.4, are injected into a pipeline or well tubing using several individual pieces of equipment that are part of a system. These can include • • • • •

A storage device for the chemical A pump or other injection device A connection and injector into the pipeline or well bottom Valves and controllers Sensors near the equipment to be protected

There are a number of different methods for applying scale inhibitors, corrosion inhibitors, and other chemicals into the downhole/pipeline environments. Most of these processes also include the mechanical units described in subsequent sections.

298  Chemical and Mechanical Methods for Pipeline Integrity

The following subsections describe continuous/batch treatments, squeeze applications, and use of various gels, foams and pigging applications. Gilson (2014) notes that the application method may depend on the flowing conditions (for multiphase flow, see Section 1.6.2) as well as chemical compatibilities. A diagram shown in Fig. 6.39, developed on the basis of this report, describes general selections of the application methods, discussed in Sections 6.7.1 through 6.7.3. Depending on the flow velocities, either slug or other methods, such as pigs, gels, foams, and spray pigs (see Sections 6.7.2 and 6.7.3), are preferred in situations in which complete inhibitor coverage cannot be achieved because of low flow conditions. Determining the location of the various types of chemical injection equipment (described in Sections 6.7.1 and 6.7.3) will be a critical task for preventing various deleterious conditions (e.g., corrosion, scale formation, and emulsions) that are treated with the chemicals described in the current chapter. While there is no formula for determining the location of injectors, some suggestions include considering the following conditions. 1. Determine the conditions to be treated. a.  Scale (organic and inorganic) b.  Corrosion, including MIC and TLC c.  Emulsions or foam prevention d.  DRAs, biocides, and other chemicals described in this chapter 2. Analyze the flowing conditions (i.e., perform flow modeling) to determine where water or deposits will accumulate and if the chemicals will have enough residence time to effect the change. 3. Analyze temperatures, pressures, multiphase flows, and terrain that will affect these processes. 4. Consider whether there will be interactions with the other chemicals or attack on the equipment. 5. Determine the location of in-place motive forces (i.e., pumps, wellheads, compressors, and merging flowing fluids). Frequently the injection point is downstream of these units. 6. Assess the relationship of injection to monitoring stations (corrosion, or possible sampling points). A diagram suggested by Abayarathna (2014) and seen as Fig. 6.40 shows monitoring points both above and below chemical injection stations; these will provide a baseline as well as aid in the success of the chemicals. 7. Determine if it is physically possible to introduce chemicals at a selected location. Flow Velocity Stratified Flow

Wet-gas lines: 0–7.5 FPS Oil lines: 0–3.5 FPS Water accumulates and is very stagnant

Annular Flow

Wet-gas lines: 7.5 – 15 FPS Wet-gas lines: Oil lines: 3.5–7 FPS 15–25 FPS All liquids still on bottom of line but less pooling

Wet-gas lines: >25 FPS Oil lines: > 7 FPS

Annular flow

Preferred Corrosion Inhibitor Application Method Slug or continuous

Continuous Fig. 6.39—Inhibitor application for given flow conditions.

Mist-annular flow

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   299

Production well

Production well

Flow Injection Monitor

Monitor

Monitor

Chemical Injection

Chemical Injection

To pipeline Monitor

Gathering and treatment Monitor Chemical Injection

Layout of chemical injection and monitoring system Production well Fig. 6.40—Injection and monitoring layout for a multiwell system.

6.7.1  Continuous Treatments With Pipeline Chemicals. Continuous treatments are designed to provide effective amounts of the inhibitor or other treatment molecules as the fluids come from the formation or from other upstream sources. Various chemicals can be injected into a pipeline system for reasons described in Sections 6.1, 6.3, 6.4, and 6.5. Injection Into a Well. Inhibitors and other chemicals that are designed to treat the tubulars may be placed in the well bottom using a treatment Macaroni string string of small-diameter tubing, called a “macaroni” string. Some of these chemicals may then Chemical squeeze enter the pipeline systems, especially the gathering lines. Note that these chemicals may have a useful or unuseful effect on the pipelines downstream, but they must be considered as integral to the pipeline chemistry. Fig. 6.41 presents a drawing of injection into a well. This figure also shows chemicals being “squeezed” into the producing formation for longer-term functioning. See Frenier and Ziauddin (2008, Chap. 5) as well as the next subsection of this book for information on scale inhibitor squeeze treatments. Inhibitors also have been injected in the gas lift ­mandrel (Fleming et al. (2003) and through the well Fig. 6.41— Inhibitor placement methods.

300  Chemical and Mechanical Methods for Pipeline Integrity

liner (Poggesi et al. 2002). See the best practice [Graham et al. (2002)] in Section 5.3.2 of Frenier and Ziauddin (2008). For shallow wells, the inhibitor batch may be dumped directly into the well tubing. Byars (1999) listed details of a number of additional direct methods for treating oil/gas wells, including tubing displacement, use of a dump bailer, and batch flushing down the annulus (see Appendix 8A of his book). An issue with the application of liquid scale inhibitors through steel tubing is the possible corrosion of the application tubing. Many scale inhibitors are mixtures of phosphonic acids, and the formulation can have a low pH value. Many of these materials will chelate iron as well as calcium and can thus possibly accelerate corrosion even if the pH is near neutral. Daminov et al. (2006) reported that corrosion rates in excess of 20-mm/yr penetration of the steel lines in West Siberian fields. As many as 1,200 wells could be affected. The authors suggested adding corrosion inhibitors to the scale inhibitor formulations. See Fig. 6.42 for one type of setup for injection into a well. Once the various mixtures of liquids and gases enter the gathering lines, treatment facilities, and transmission lines, additional inhibition may be necessary because the chemicals that were injected into the wells have been depleted or removed or may be harmful to a downstream process. On shore and at a readily accessible locations, injection of inhibitors or other treatment ­chemicals may be as simple as locating a tank of the chemical at the wellhead or at a convenient place along a pipeline, Flush line Flowline

Inhibitor(s)

Check valve Pump

Inhibitor Oil

60 ft capillary tube

Pipe wrap, six joints

Pump

Res. 1 Entrance

Intake Motor

Res. 2 Entrance Res. 3 Entrance

Fig. 6.42—Example of injection setup for well bottom placement.

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   301

such as at a ­pumping station. Then, the treatment chemicals are injected using an appropriately sized pump or using an educator to draw the chemicals from the tank. As long as there is enough turbulence to disperse the chemicals, 360o coverage may be accomplished. However, many types of complications exist, especially because in the distance between injection stations, which are frequently downstream of pumps, chemical and flow conditions may differ. Some of the many physical and chemical conditions that may affect the coverage are described in the next sections of this book. Many types of pumps are Fig. 6.43—Additive injection on surface into an inlet pipe. available, but their descriptions are not within the scope of this book. One example of injection at a line or wellhead is shown in Fig. 6.43. Squeeze Treatments. Scale, organic solids, and corrosion inhibitors often are squeezed into a formation to achieve a slow release of the chemical that has been adsorbed onto the formation rocks. Fig. 6.40 shows one diagram of a squeeze treatment in which the chemicals are seen to go into one zone of the formation. During production, the adsorbed (precipitated) chemicals will be released to protect the metal surfaces from scale or corrosion damage. Note that this method requires injection at pressures that do not fracture the formation. See Frenier and Ziauddin (2014). Details of this complex method are described in Frenier and Ziauddin (2008). Stalker et al. (2014) claimed that in a number of carbonate reservoirs producing from different regions, corrosion inhibitor squeeze treatments have been shown to provide corrosion protection for downhole production tubulars when no alternative downhole chemical deployment technique is available. They also claimed that corrosion inhibitor squeeze treatments are much less commonplace than scale inhibitor squeeze treatments. The author noted differences between squeeze chemicals used in these two treatment approaches: • Scale inhibitor adsorption and release mechanisms have been studied in detail (see Tomson et al. 2004). • Many corrosion inhibitors are put into an oil base for effective distribution, so adsorption on a water-wet formation may not occur (these are partition coefficient issues). • There may be difficulty detecting and tracking corrosion inhibition to determine if the solution has an effective concentration of the inhibitor. • Formation damage is possible. Extensive laboratory testing by Stalker et al. (2014) found that some chemical families may be used as possible corrosion inhibitor squeeze candidates, but the partitioning into the correct phase

302  Chemical and Mechanical Methods for Pipeline Integrity

for contact with the corroding surface will be the major issue. The author of this book notes that the paper does not consider the use of squeeze treatments for pipeline situations; however, small reaches may be treatable. Longer reaches of pipelines may require additional inhibitor applications, which are controlled by corrosion probes (Section 3.4.3) or by chemical analyses for scale inhibitors. Nelson et al. (2007) described the corrosion inhibitor applications by continious injection as well as periodic batch injections to protect a large gathering line system in North Dakota. Equipment such as shown in Fig.6.43 was used in the injection systems. The retention of a scale inhibitors in the formation is the determining factor on the squeeze life (lifetime to minimum inhibitor concentration). As noted in Frenier and Ziauddin (2008), a precipitation squeeze, in which process the anionic inhibitor (usually a phosphonate or a polyacrylate with phosphorous as an analytical tag) is precipitated as a Ca2+ salt, usually lasts longer than an adsorption squeeze, in which the inhibitor molecules adhere directly to the limestone, sandstone, and clays. A number of other ways to enhance the lifetime of a squeeze also have been proposed. Kelland (2009) provided a short discussion of a number of “squeeze enhancement” methods that include Ca2+, Zn2+, and the quaternaries to be mentioned later, as well as kaolin clays. As noted, the use of Ca2+ has been extensively researched (see Frenier and Ziauddin 2008). Shuler (1991, 1993) provided good “teachings” on squeeze technologies and inhibitors that can be reviewed by the reader. He describes a method that involves injecting an aqueous solution of polyquaternary amines into the formation. This solution can be injected before, simultaneously with, or after injection of the scale inhibitors. The scale inhibitors are preferably a blend of nonpolymeric/polymeric scale inhibitors. Preferably, the polyquaternary amine is a poly-(dimethylamineco-epichlorohydrin) or a poly-(diallyldimethylammonium chloride). Shuler said that materials such as the clay stabilizer polyepeichlorhydrin are effective. In a typical process, an amine polymer solution is injected into a well, followed by injection of a scale inhibitor and an overflush. The well is then shut in for 20 O to 24 hours before production recommences. Shuler (1991, 1993) teaches that to ensure facile injection of the polyquaternary amines into the formation during the process, their N HCl n molecular weight should be below 50,000. Selle et al. (2003), Chen et al. (2006), and Montgomerie Poly(dimethylamine-co-epichlorohydrin) et al. (2010) claimed that squeeze life can be extended if a cationic monomer or polymer is placed into the formation before the inhibitor is squeezed. Montgomerie et al. (2010) Cl N+ used a homopolymer formed from diallyl dimethyl ammonium chloride (DADMAC). Possible structures of bridging Diallyldimethylammonium chloride agents are in shown in Fig. 6.44. DADMAC These compounds are then followed by the scale inhibitors of the types mentioned by Frenier and Ziauddin (2008) and Kelland (2009). Montgomerie et al. (2010) proposed a mecha* nism by which the use of charged polymers such as those described in the aforementioned applications enhances reten+ Cl tion of scale inhibitors in subterranean formations. This does N so by a mechanism through which the adsorption of the posiH 3C CH3 n tively charged compounds to the formation reduces its negative charge. As a result, scale inhibitors, which are often negatively PolyDADMAC charged, are more readily retained on the formation. This mechanism may also involve a combination of effects, Fig. 6.44—Possible polymer repeat including modification of the formation surface to enhance unit structures for squeeze enhancement chemicals. adsorption and precipitation. It is believed by Montgomerie

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   303

et al. (2010) that when polymers formed from a diallyl ammonium salt are combined with scale inhibitors, and in particular scale inhibitors comprising a carboxylate group, a solid such as gel, for example, forms. This gel acts as a precipitate in that it is easily retained in a hydrocarbon well. However, the reaction to form the solid is believed to be reversible. The method of the invention is claimed by Montgomerie et al. (2010) to be different from those earlier techniques using relatively low-molecular-weight poly salts—poliquaternary amines (Shuler 1991)—because the earlier methods rely solely on adsorption of the amine to the rock to improve retention of scale inhibitor. It is also claimed to be different from the earlier technique of using Ca2+ because the polymer formed from a diallyl ammonium salt can itself be adsorbed onto a rock surface, preferably in a preconditioning step. Thus, a further advantage of the method of the invention is that the polymer formed from a diallyl ammonium salt can be preinjected followed by an injection of a scale inhibitor. This is claimed to reduce or prevent formation damage (i.e., loss in permeability). Squeeze treatments and other scale inhibitor treatments may be required in very hot wells. Wang et al. (2012) tested several phosphonate and polymeric scale inhibitors at temperatures up to 200°C to determine the thermal stability and thus the effectiveness of the materials for delaying precipitation of BaSO4. Using scale delay tests and chemical analyses, the authors found that all the inhibitors tested (three phosphonates and three polymers) were quite stable up to approximately 150°C, but the polymers were more stable at 200°C. However, adsorption on sandstone core materials materially improved the stability of diethylenetriamine pentamethylene phosphonic acid, even at 200°C. The implication is that squeezing the inhibitor applications will stabilize the inhibitor in very hot wells. Direct Injection Into a Pipeline. Ramachandran et al. (2011) described a method for injecting treatment chemicals into a pipeline using the setup of Fig. 6.45. This diagram shows sensors, automated valves, and programs for injecting the correct volume of chemical for a number of different treatments. In this case, a surfactant was injected to promote single-phase flow in the line. Chen et al. (2003) claimed to have developed a mechanistic chemical dispersion model to aid in optimizing corrosion inhibitor injection into a pipeline. The mechanism of corrosion inhibition is attributed to the adsorption of inhibitor to the pipe wall to create a hydrophobic layer (see Fig. 6.2). To accomplish optimization of injection, a mechanistic corrosion inhibitor dispersion model, which is linked to a commercial flow simulator, was developed. It is claimed to account for the mechanisms governing the distribution of corrosion inhibitor, including the gross distribution of oil/water/gas in the pipeline, turbulent dispersion in the free-water phase, partitioning of inhibitor between oil/water phases, and inhibitor adsorption/desorption on the pipe wall. This model is also claimed to be able to calculate the inhibitor concentration in the liquid phases as well as on the wall in the flowline. The basis of the model, according to Chen et al. (2003), Control unit Control fluid (F)

Sensor

Programs Models Data

F

Pump Valve

Fig. 6.45—Chemical injection system (Ramachandran et al. 2011).

304  Chemical and Mechanical Methods for Pipeline Integrity

is a description of the solute (corrosion inhibitor) distribution, which entails an accurate calculation of velocity and phase distribution in a pipe. The 3D flow phenomenon is modeled by a 1D Tanks approximation. This approach has allowed the authors to solve the distribution of solute in terms of the cross-section-averaged variables. Thus, the authors contended, this mechanistic model should be a useful tool in the design of optimal corrosion Different inhibitor injection for corrosion control—that is, chemicals for determining the minimum inhibitor concentration required to provide the maximum inhibition protection. Spill Ismail (2014) contended that inefficient chemicontainment cal management can lead to later problems with FA and infrastructure integrity—to corrosion and possibly even failures, particularly from H2S corFig. 6.46—Chemical injection tanks (Ismail 2014). rosion. This author also claimed that recognizing this, operators have recently become more ­mindful of implementing metric-driven chemical management programs. Fig. 6.46 shows a photograph of chemicals being stored on the surface for injection into facilities, into pipelines, or downhole for control of many problems described in this chapter. The processes of chemical management described in the Chen et al. (2003) report include • Identification of key performance indicators of successful treatments • Data collection • Data standardization The report describes the optimization of the use of H2S scavengers (Section 6.4), as well as paraffin buildup inhibitors to minimize chemical usages (Section 6.3.1). Wang et al. (2002) claimed that pipeline conditions that result in slug flow (see Thome 2012), in which bubbles may form in the line, may result in an incomplete formation of an inhibitor film, even when there is enough turbulence to transport the inhibitors to the top of the line. The authors of this paper cited Kaul (1996), who had shown that inhibitors behave differently in full pipe and slug flows. The claim is that the inhibitors are more effective in full pipe flow than in slug flow, the effectiveness in full pipe flow being close to 100%, whereas it is less than 50% in many cases of slug flow. The location of both the chemical injection and the corrosion measurement is important for controlling the unwanted conditions, such as corrosion. Abayarathna (2014) makes several recommendations/suggestions for location of injection and monitoring (see Section 3.4.3). • Track baseline corrosion rates by monitoring just upstream of any corrosion inhibitor injection point. • Locate monitors at/near the pipeline end rather than upstream. This may be more critical because it provides a measure at the end of line to show the effectiveness of any remediation measures (e.g., corrosion inhibition). • Obtain velocities on the basis of production rates and internal diameters of pipelines to plan injections and monitoring. • Check elevation changes to determine if critical angles could be exceeded and liquids could accumulate.

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   305

• Check flow regimes (e.g., stratified or turbulent) because they affect inhibitor distributions.  The key is to place sensing element within the most corrosive phase—the produced water.  Also check if there are solids within the pipeline that could cover the coupon or probesensing element. Additional complications can arise during treating fluid injection as a result of unwanted chemical interactions. Gonzalez et al. (2013) described the use of both corrosion and scale inhibitors in a production train. They noted that the injection of multiple chemicals, such as antiscale and anticorrosion additives, must often take place sequentially or simultaneously in very close injection points. Thus, it is necessary not only to evaluate the efficiency of chemicals individually but also to screen for incompatibilities and/or interference between chemicals. These interferences can affect either the antiscale efficiency or the anticorrosion efficiency, or even both efficiencies at the same time. This kind of mutual reduction of efficiency may have been the reason for past pipeline failures, in which scaling and corrosion were both observed despite the addition of large quantities of both additives. On the basis of their experiments, these authors found that both materials are effective in only a small concentration range of a scale inhibitor and a corrosion inhibitor. See Gonzalez et al. (2013, Fig. 3) as well as Poggesi et al. (2002) for examples. Gonzalez et al. (2013) claimed that many candidate additives for oilfield applications have been evaluated and that “interaction maps” have been constructed. These maps indicate regions of concentrations at which one or both additives are not effective. Further, these maps are useful for the selection of products that should be effective without risks of interactions. To finalize the validation, we can perform an “ultimate” blocking test in which both additives, water, and oil are in contact at one point of the experiment, to also take into account the partitioning of additive between oil and water. Freeman and Williamson (2006) described a process that includes the use of an injection nozzle inserted into a line at or near the wellhead or just downstream of a compressor/pumping station. This nozzle injects/pumps corrosion inhibitor into the line under high pressure, vaporizing it where it mixes with the gas and/or liquids in the line. Corrosion inhibitors used in these applications are typically water soluble. The chemical goes into solution and travels with liquid water on the pipeline bottom. Direct contact between the pipe wall and the water is the primary mechanism for inhibitor transfer. Gravity effects, low flow rates, and stratified flow make transfer of the corrosion inhibitor to the top of the pipe improbable at many locations. As water vapor and other gases travel along the pipeline, they are cooled. The cooling allows water vapor to condense on the upper portions of the internal pipe wall. The condensed water often contains no or insignificant quantities of corrosion inhibitor and does not provide any corrosion protection. In the presence of certain acid gases, corrosion rates can be extremely high in the upper portions of the line. The other extreme may be the injection of chemicals into a subsea well system. Ludlow et al. (2010) described a high-pressure pump for use in the injection of liquid chemicals into subsea oil/ gas wells. The pump was intended to be positioned in the subsea environment adjacent to the wellhead. The system comprises a piezoelectric actuator for reciprocating a plunger that acts to compress and expand the effective volume of a pumping chamber having a valved inlet connected to a source of the liquid and a valved outlet to lead the liquid to the well. The device is claimed to have a minimum of moving parts and in particular avoids the need for any rotating parts and as well as high-performance bearings and seals. Joosten et al. (1999) described the testing and design of corrosion-monitoring electrical resistance (ER) probes (Section 3.4.2) and a generic subsea setup for injection of the inhibitor (Fig. 6.47), as well as corrosion monitoring similar to that described in Fig. 6.48. The authors noted that the subsea ER probes are flush mounted at the 6 o’clock position in the outlet of the stainless-steel test and production headers. The ER probes are positioned within the stainless-steel header piping but as close as possible to the stainless steel to material transition where the flowlines attach to the headers. This allows continuous treatment of the fluids entering the pipeline.

306  Chemical and Mechanical Methods for Pipeline Integrity

Inhibitor injection Duplex SS Production Manifold

Connector for future tie-in

10 meters Individual well tie-ins

Carbon steel pipeline

Corrosion monitoring probe

Fig. 6.47—Subsea inhibition injection manifold with a monitoring probe (Joosten et al. 1999).

In many cases, a corrosion probe or other sensor can be set to automatically start inhibitor/chemical introduction when a set point is reached and to cease introduction when the desired effect is achieved. The ER probes are at the heart of automated monitoring and corrosion control systems. Szabo et al. (2009) asserted that many of the costs directly related to corrosion may be mitigated and managed with continuously monitored corrosion transmitters as part of a comprehensive plantwide control strategy. Process parameter effects related to electrochemical corrosion may be minimized by means of direct, continuous corrosion feedback for active control and optimization of neutralizing agents (e.g., inhibitors). A system described in that source has an LPR probe (depicted in Fig. 48a) with readings of general corrosion, localized corrosion, and fluid conductance. The signals are sent to a data analyzer that then interprets the input signals and sends the data to an alarm/control unit. An example of a corrosion rate output signal is in Fig. 6.48b. The signals can alert the operator to activate corrosion inhibitor flows, or the system can be programmed to start the injection systems automatically.

Fig. 6.48—Automated corrosion monitoring and alarm/control system

6.7.2  New Formulations for Application of Production Corrosion and Scale Inhibitors. The use of microemulsions to place and disperse an inhibitor is reviewed. Collins and Hewartson (2002) claimed that as the industry moves toward increasing deepwater production, the very high cost of well intervention places large constraints on the performance of scale control options. The paper discusses using EOR-type surfactant systems (see Frenier 2014, Chap. 5) in conjunction with scale inhibitors to extend the squeeze inhibitor treatment lifetime. Collins and Hewartson (2002) contended that the treatment lifetime is extended by the miscible displacement of organic material from formation surfaces, thereby

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   307

increasing the surface area available for the scale inhibitor to adsorb onto the surfaces that are placed in microemulsions (Section 4.6.1). This extensive paper provides many details on the preparation of the microemulsions and the incorporation various oilfield chemicals in them. Several patents provide useful details about using microemulsions for placing scale and corrosion inhibitors. For example, Collins and Vervoort (2003) describe a microemulsion comprising (i) an oil phase, (ii) an aqueous phase comprising an aqueous solution of a water-soluble oil/gas field production chemical or an aqueous dispersion of a water-dispersible oil/gas field production chemical, and (iii) at least one surfactant, wherein the aqueous phase is distributed in the oil phase in the form of droplets having a diameter in the range 1 to 1000 nm. The patent claims that the droplets are in the form of microdominans having at least one dimension of length, breadth, or thickness in the range 1 to 1000 nm. Halim et al. (2011) described the synthesis and testing of a wax inhibitor to protect a pipeline carrying waxy crude oil. To assess the use of the wax inhibitor for squeeze applications, a coreflood study was conducted to determine its capability to adsorb onto formation. A formation damage study was also conducted to ensure that there is no formation damage coupled with the injection of wax inhibitor. It was found that the wax inhibitor can be retained in the formation up to 88.5% without significant formation damage. As a next step, it is planned to run the coreflood and wax inhibitor release tests to refine the design of squeeze treatment. A patent by Zaid and Wolf (2000) proposes a method for making solid pellets of treating chemicals that can dissolve slowly for a continuous release of chemicals. Specifics of the products are that the solid composites are formed by mixing a quantity of melted nonylphenol ethoxylate. The desired active ingredients (corrosion inhibitors, scale inhibitors, bactericides, scale converters, foaming agents, and mixtures thereof) are then added to the melted mixture followed by stirring for approximately 15 minutes. The hot material is then poured into a belt mold or ball mold and alloyed to cool, forming pellets of the desired size and shape. 6.7.3  Applications of Treatment Chemicals in the Batch Mode and Using Pigs. Corrosion inhibitors as well as organic solids inhibitors and biocides can be placed in pipelines using continuous direct injection. For these, see Section 6.7.1 and Figs. 6.42 and 6.46; also see the review by Frenier and Wint (2014). Batch treatments, however, can involve a short period of treatment chemical injection, followed by periods when the material is not injected. Direct Batch Treatments. In many pipeline environments, it may be economically or physically impossible to treat some stretches of line using continuous injections, so batching may be the preferred method. Online monitoring by methods described in Section 3.4.2, as well as some of the methods described in this section, may be required to determine batching intervals for corrosion inhibitors. Further, testing of TLC and inhibition requires specialized equipment, also described in Section 3.4.2. In addition to periodic injections, operators can use various types of pigs to help coat a pipe interior. Pruett (2005) noted that one method of applying a treating liquid to the interior of a pipeline is accomplished where the treating liquid is captured between two pipeline pigs, called “batch pigs,” that move in tandem through a pipeline pig with the treating liquid there between. Although this method is widely accepted and used, it does not necessarily ensure that the upper quadrant of the pipeline interior is adequately coated with or exposed to the treating liquid. This is similar to the pig train with slugs of fluids described in Figs. 5.10 and 5.11. One way to help ensure coverage requires that enough inhibitor or other treating chemical must be applied and the total amount estimated from the line surface area. The pig train must move at a velocity so that all surfaces are contacted. The chemicals should be applied so that the slug of inhibitor has at least 10 to 15 seconds of contact with the pipe surfaces so as to coat the top of the pipe. Gelder et al. (1988) suggested that a 5% solution of the inhibitor in oil should contact the surfaces for at least 10 seconds and that the presences of a hydrocarbon liquid in the line may reduce the inhibitor effectiveness.

308  Chemical and Mechanical Methods for Pipeline Integrity

Bojes et al. (2001) noted that batch corrosion inhibitors are widely used for the corrosion control of production wells and pipelines in the oil/gas industry. The authors claimed that rules of thumb that incorporate film thickness, contact time, and surface area are still commonly used to calculate the volume of batch inhibitor required for pipeline applications. However, actually measuring the thickness of the inhibitor film on the metal and the impact of different variables on the film (e.g., inhibitor type, contact time, diluents type, dilution ratio, and shear stress) offers the potential to provide a better understanding for optimizing the application procedure and required batch frequency. The inhibitors tested had viscosities from 4 to18 cp, and the investigators claimed that percent coverage is more important than thickness as a predictor of longevity of corrosion protection under shear (rotating cage) test conditions. Various inhibitors were described in Sections 6.1, 6.3, and 6.4.3. However, the formulations for batch applications can vary. Yang et al. (2007) claimed that properly formulated water-dispersible inhibitors can be applied using the batch methods as well as oil-soluble inhibitors. The film persistency was found to be excellent, and 90% coverage of the metal surface was achieved. Wylde et al. (2013) claimed that that batch corrosion inhibitors frequently are oil soluble. But even if they are water soluble/dispersible, they are designed to provide a protective chemical barrier against corrosion. Oil-soluble corrosion inhibitors are generally acknowledged to have improved film-forming capabilities over that of their water-soluble counterparts. The drawback, however, is that many mature fields that require increased corrosion protection also produce large volumes of water, which makes dispersability of oil-soluble products a challenge. Furthermore, oil-soluble corrosion inhibitors tend to have a poorer environmental impact profile when compared to water-soluble products. The investigators claimed that diluents such as field condensate/crude, diesel, or other aromatic solvents can be used to dilute the inhibitor. This results in a larger batch volume, thereby yielding improved contact time with the metal surface to be protected. These authors tested • A blend of water-soluble alkoxylated amide amine, quaternary ammonium compound, and imidazoline with sulfur synergist • An oil-soluble blend of imidazoline, fatty acid alcohol, and dimer acids • A water-dispersible blend of aliphatic amines, ethoxylated alcohol derivatives, and a sulfur synergist In their laboratory tests, the oil-soluble blend was the most effective, and the water-dispersible product was determined to be preferable to the water-soluble blend. The next section covers application using gels and foams. Gels and Foam as Application Media. The methods described in this section can also be viewed as batch treatments, but the phases of the fluids have been modified for efficiency of application. Gelled Pigs. See descriptions of the gelling chemicals in Section 6.6. Various formulations of chemically gelled water/oil can be used to place inhibitor chemicals onto pipeline surfaces. The use of gels or “gelly pigs” in pipeline operations, including for inhibitor placement, has a long history. Because it is liquid, although highly viscous, the gel can be pumped through any line that will accept liquids. Gel pigs can be used alone in liquid lines, in place of batching pigs, or in conjunction with various types of conventional pigs. When used with conventional pigs, gelled pigs can improve overall performance while almost eliminating the risk of sticking a pig. Gel pigs do not wear out in service as do conventional pigs, though they can degrade from shear. Note, however, that the fluids can be susceptible to dilution and gas cutting. Therefore, when designing a pig train that incorporates gel pigs, care must be taken to minimize fluid bypass of the pigs and to place a conventional pig at the back of the train when displacing with gas (PM 2013). This source also claimed that gel pigs have been used to temporally to seal valves during hydrostatic testing. As in the applications in fracturing, the fluids can be a liner gel— called, simpy, a “gel”—which has increased viscosity compared to the normal liquids.

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   309

Crosslinked gels—in this case, called a gel or “gelly” pig—can be very thick and almost like a polymer pig (see TDW 2011). Inhibitors or possibly other chemicals can be added to either the gel or the gel pig. The most important idea is that the inhibitors must be tested to be compatible with the gel and allow transport through the line. Gel applications require a full line segment and the pressure available to move the “train.” Fig. 6.49 illustrates three possible ways to use various pigs. In Fig. 6.49a is a foam (mechanical) pig, an uncrosslinked gel pig that can contain a chemical, and a crosslinked (very viscous) gel pig; Fig. 6.49b is without solid pigs for inhibitor applications, and Fig. 6.49c shows the use of solid pigs capable of some cleaning action. Many combinations of these configurations are possible. This figure shows the use of aqueous fluids, but oils can also be gelled (Gulbis and Hodge 2001) and possibly used with inhibitors. The inhibitors could be placed into any of the fluids, and the “gel” section (light blue) could be an ungelled aqueous fluid. These fluids can be used to deposit various inhibitors (Kennard and McNulty 1992), as well as biocides (PM 2013). The chemistry of these materials was described in Section 6.6; the use of these types of chemicals for heavy cleaning of pipelines is discussed in more detail in Section 7.5.3. Because the gels are extremely viscous, they will support and aid in the placement of treating chemicals over the entire surface of the line. Note that the gels can be made from aqueous brines or can be oil based to be compatible with the treatment being performed. PM (2013) claimed that many inhibitors are dispersed into gelled oil formulations. A reason for using gels (and foams) may apply if the line is very difficult to pig. Quarini and Shire (2007) reviewed the use of various types of pigs, including gel pigs, and they determined that several attributes of these fluids will aid in inhibitor placement: • They are unlikely to become stuck. • Because they conform to pipeline irregularities, all surfaces are contacted. • The gels enhance the sealing capacity of any mechanical pig in the train. Direction of pig train

Filtered Seawater

Foam Pig (Mechanical)

120 m3 Gel

(Linear polymer with treating chemical)

Gel Pig (crosslinked polymer)

a) Direction of pig train

Filtered Seawater

Gel Pig (crosslinked polymer)

120 m3 Gel

(Linear polymer with treating chemical)

Gel Pig (crosslinked polymer)

b) Direction of pig train Filtered Seawater

70 m3 Gel Foam brush pig

(Linear polymer with treating chemical)

c)

Fig. 6.49—Pig train options for inhibitors (Mackay 2013).

Foam brush pig

310  Chemical and Mechanical Methods for Pipeline Integrity

Uzu et al. (2000) described gel pig technologies and noted that gel pigs can be used for many purposes, including for cleaning (described in Sections 6.6 and 7.5.3) as well as for placing chemicals such as corrosion inhibitors, biocides, dehydrating fluids, and paraffin solvents or inhibitors. These authors claimed that as much as 20 vol% of the pig can be an inhibitor formulation and that the pig can then effectively coat long sections of a line. They also claimed that commercial applications of inhibitors in a gel were performed in 1982 to a 279-mile segment of a 36-in. North Sea subsea wet-gas line. Keys (2000) described the 1997 conversion of an oil pipeline to gas transport and the use of gel pigs with various mechanical devices to clean and inhibit the surfaces. The gel stages helped to remove more than 1 million lbm of debris for a 151-mile line as well as provide an initial coat of corrosion inhibition. Akhiyarov (2007) confirms that the turbulent flow in a gel (Fig. 6.36) is useful for distributing an entrained/dissolved inhibitor efficiently over the pipe surfaces. Schreurs et al. (1994) discussed the development of gelled pigs for use in the drying/dewatering of pipelines. This will often entail the gelling of alcohol solutions or using alcohol between gel/ solid pigs to dry the line after maintenance or before commissioning (see Section 5.3). This operation provides a very low dewpoint and greatly reduces the corrosion potential. The authors claimed that water-based and methanol-based gels have been developed that provide excellent performance. It has been demonstrated that it is possible to design and operate dewatering trains even for long pipelines with virtually no gas bypass. The contributions of this paper are claimed to include the functional specifications and the methodology for the development of appropriate gels for pipeline dewatering/drying applications. The laboratory-based simulations of the dilution and shearing of gels during the train of fluids and gels was also claimed to be useful. Foams. See Section 7.5.4. Foams can be formed in pipelines, and various treating chemicals can be put into the foaming fluid and used to coat all the line surfaces. Note that these are “foamed” liquids, as compared to a polyfoam pig. See Section 7.5.4 for the details of the formation and movement of foams in pipelines and how they can be used for dewatering lines. Foams have some of the rheological properties of gels, such as shear thinning and non-Newtonian properties; however, the greatest benefit is that foams greatly reduce the volume of liquids required to fill the volume of a line segment. Usually, the volume of gas/liquid will exceed 8.5:1. Several other authors have reported on the use of foam to place corrosion inhibitors on top of pipelines. Note that spray devices (to be descried in ­Section 6.7.3) are also claimed to be useful for placing inhibitors on all sides of a pipeline. Achour et al. (2010, 2011) described the use of foams formed in pipelines to control TLC. They claimed that the idea is to inject the corrosion inhibitor within a foam matrix. The foam slug is formed and carried along the pipeline by the produced gas, thereby contacting the whole circumference of the pipe along a given distance. They also claimed that the foam can be broken before separation using a defoamer if it does not naturally break. This process ensures homogeneous delivery of the inhibitor to the pipe wall along pipe sections suffering from TLC. This technique does not interrupt or slow down the production and is more cost-effective than the classical methods of treating TLC using pigging or batch inhibition. Achour et al. (2010, 2011) tested both foaming agents and corrosion inhibitors. See Table 6.6 for a description of foaming Foaming Agent Corrosion Inhibitor agents and corrosion inhibiN-decyl-N-dimethylamine oxide 3-methoxyproplamine tors that were tested by these Dodecylaminodipropionate 3-Methoxyproplamine bromide authors. Achour et al. (2010, 2011) Sodium C14-16 olefin sulfonate Didicyldium ethyl ammonium bromide described a foaming test based Dodecylbenzene sulfonic acid Decylamine caprylate on ASTM D892-98 (1998). Dentritic polymer (DP foamer) Dodecylbenzene octylamine caprylate They then tested mixtures of Glucopon 215 UP Cyclohexylamine caprylate foamers with various corroTable 6.6—Foaming agents and corrosion inhibitors. sion inhibitors to produce an

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   311

optimal blend for foam stability and corrosion inhibition. Their conclusion is that foam tendency and stability results show that the choice of corrosion inhibitor and foaming agent is a key factor in applying the new method. The corrosion-testing results confirmed the proof of the concept of delivering the corrosion inhibitor to the top of the line by means of the foam. A diagram of the injection setup is shown as Fig. 6.50. Applications of Inhibitors With Specialty Pigs. If a line is piggable, a specialty pig can be used to overcome problems with contacting all line surfaces. Pruett (2005) and Freeman and Williamson (2006) described an alternative method. It applies the Bernoulli (or Venturi) effect to a specially constructed spray-producing pig that uses the energy of bypass flow to do the work needed to redeploy and redistribute residual inhibitor chemicals throughout the pipe run. Reusing or effectively transferring corrosion chemicals in pooled accumulations along the pipeline bottom has been viewed as an alternative to “chemical batching” and an effective step in solving the problem of TLC. Fig. 6.51 shows a depiction of this device. The authors noted that one method of using the spray pig can be applied when continuous injection is the primary means of introducing corrosion inhibitor fluid into the pipeline. This method is effective on relatively level lines and pipelines with a continuous upgrade, such as those associated with offshore wells. The spray pig has proved to be a very effective dewatering pig while it distributes inhibitor-containing fluids to the top of the pipe. In this manner, a dense vapor cloud is created in front of the pig as it splashes through and jets the liquids. Freeman (2009) reported on the use of a spray pig system (similar to Fig. 6.51) for treating TLC that has been semiautomated to collect performance data. To provide more-immediate feedback on the performance of a spray pig in any given application, engineers have incorporated some existing in-line inspection technologies into the body of the pig itself as an on-board data logger. It is, essentially, a “semismart” spray pig. The author claimed that the unit is capable of operating under normal spray pig conditions and that the pig is equipped to collect a variety of information during every run. This includes data on differential pressure (∆P), which is of key importance to achieving proper jetting action. The semismart spray pig gauges pressure at both the front and the back of the pig. To be effective, a spray pig must deliver inhibitor to the top of the pipe. To verify that this is indeed happening, the semismart design also monitors rotation/orientation of the pig. Check valve

Control valve Surfactant and water

Gas space

Liquid

Pump

Check valve

Mixer

Gas Inhibitor or other chemical

Pump

InjectFoam or

Check valve

Pipe

Flow

Fig. 6.50—Foaming setup for gas, liquid, and chemical injection (Achour et al. 2010).

312  Chemical and Mechanical Methods for Pipeline Integrity

Sealing cups and disks provide the tightest seal available

Middle jets spray inhibitor fluid to Saturate the top quadrant of the pipe

Counterweights ensure that the spray pig remains properly oriented in the pipeline The spray pig will act as its own reservoir by capturing the fluid between the front and rear sealing elements. This provides the volume of fluid to supply the nozzles and ensure complete treatment of all pipe surfaces.

Using the typical differential pressure associated with pigging the spray pig siphons the residual inhibitor laying in front of the pig through the nozzles and redistributes it directly to the unprotected top inner wall of the pipe.

Fig. 6.51—Inhibitor spray pig (T. D. Williamson 2015).

The data acquired by the device offer a complete 3D profile of how the pig moved throughout the run, including tool rotation, spray nozzle orientation, and port orientation. Should an improper orientation occur, the pig’s sensors can tell whether this was temporary or permanent and also where it took place in the pipeline (e.g., relative to known corrosion areas). Freeman and Williamson (2006) also noted that the systems is most effective if used in a “train” of a separator, a spray pig with inhibitor, and a foam pig to complete the application (see Fig. 6.52). Lim (2013) provides additional uses of the spray pig concept described in Figs. 6.51 and Fig. 6.52. For most-effective placement of an inhibitor or even another chemical such as a biocide, the pipe surfaces should be extremely clean. The author recommended the use of a pit-cleaning pig (left side of Fig. 5.1c), if possible, with long steel bristles that can dig scale from a pit. For this type of tool, he also made specific recommendations: • Rotational orientation. To be effective, the spray pig must deliver inhibitor to the top of the pipe. The data acquired by the logger offered a complete profile of the pipeline throughout the run, including tool orientation (spray nozzle and port orientation). Separator/scraper pig or spray pig

Spray pig

Liquid inhibitor Foam pig Fig. 6.52—Spray pig with train of additional devices.

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   313

The pig sensors can indicate whether the entire top of the pipe had been properly sprayed or if and, from odometer data, where an improper orientation took place in the pipeline relative to expected or known corrosion areas. • Differential pressure. The spray pig gauges pressure at both the front and the back of the pig, allowing for the calculation of the differential pressure across the pig body. Pressure differential is of key importance to achieving proper jetting action because bypass flow and differential pressure are the dynamics that create the Venturi effect, which drives the spraying action. • Pig speed. The spray pig should stay within speed limits to allow for proper contact time between the inhibitor and the pipe wall. The pig’s odometer generates the speed profile on the run as well as the accurate location of events. Inhibitor contact time for each specific pipe segments of interest can be calculated, and this can help establish future frequency of spray pig usage. • Acceleration. Speed excursions are typical when pigging in gas lines, but these are not conducive to consistent inhibitor application. Matching the data with the location obtained from the odometer provides a good overview of the overall jetting performance in a run. • Temperature. Gauging line temperature throughout a run can highlight areas that may be most susceptible to corrosion, in particular TLC, which is believed to occur in specific temperature ranges. Use of Pigs to Apply Epoxy-Based Inhibitor Films. Shouse et al. (2012) claimed a method for performing a proper application of inhibitors to all parts of a pipeline using a two-part corrosion inhibitor. The binary corrosion inhibitor is applied by a modified batch method (Fig. 6.53) using the spray pig technology that ensures 360o coverage of the internal pipe wall. Note that this is an application of the patent from Zaid and Sanders (2005) and Zaid et al. (2008). It proposes an epoxy-based inhibitor process described in Sections 6.1.4 and 6.1.5. See Fig. 6.9 for the chemistry. As shown in Fig. 6.53, the binary corrosion inhibitor is typically applied as a two-part system where the first part (the epoxy resin dispersed in a solvent, Fig. 6.53a) bonds molecularly with the metallic pipeline wall and then the second part (the amine hardener, Fig. 6.53b) reacts and bonds with the first, allowing it to be applied to pipelines with limited cleaning and surface preparation requirements. Proper application of the binary corrosion inhibitor delivers superior internal corrosion protection for the pipeline operator/owner at a better economic cost/benefit performance over traditional corrosion inhibitor treatment methods. Baker (2015) shows examples of the components being sprayed onto the interior of a pipe section. Field performance and film persistency tests have, according to Shouse et al. (2012), indicated that the active lifespan far exceeds that of existing conventional pipeline corrosion inhibitors. Results from a 1,000-ft loop test using corrosive water showed a reduction in corrosion rates from 169 mpy to 0.01 mpy after a 6-hr soak of the binary inhibitor. Field performance and film persistency tests (i.e., soaking treated pipe sections in pH 2 acid solution) have indicated that the active lifespan exceeds that of existing conventional pipeline corrosion inhibitors. Tests by these authors also showed that this binary corrosion inhibitor has been proved to significantly reduce wax deposition problems in crude oil lines, which results in lower maintenance and improved flow characteristics. Shouse (2014a) proposed a mechanism of action of the binary inhibitor whereby the materials penetrate a rust layer (Fig. 6.54a) and produce a hydrophobic, long-lasting protective film on the pipe surfaces (Fig. 6.54b). Comparison of Inhibitor Application Methods. This section provides this book’s author’s and advisors’ evaluations of the various application methods. These evaluations are made on the basis of their experience in this field and the literature citations in Frenier and Wint (2014). Table 6.7 shows a comparison of benefits as well as known disadvantages of the application methods described in Frenier and Wint (2014). Different line configurations—piggable or nonpiggable (see TDW 2011)—may dictate different methods. Shear conditions and amounts of liquids/gases in a line may also affect the choice. Several different developmental techniques are being investigated and were described.

314  Chemical and Mechanical Methods for Pipeline Integrity

Batching pig

At least 8 hours later run a wiper pig.

Spray pig Wall coating spray

Uncoated

Foam wiper

Mixed slug or liquid Liquid in front spray pig

Coated pipeline

a) Pig train for gas line: premix binary parts A and B

Sealing Pig/wiper

Spray pig Wall coating spray Coated pipeline

Slug or liquid pill Part A

Uncoated

Liquid in front spray pig

b) Binary inhibitor part A (first) pill configuration for gas line

Batching pig

At least 8 hours later run a wiper pig.

Spray pig Wall coating spray

Part A coated pipeline

Slug or liquid pill Part B

Foam wiper

Liquid in front spray pig

Coated pipeline

c) Binary inhibitor part B (second) pill and foam w/wiper pig for gas line Fig. 6.53—Application methods of binary inhibitor for gas lines (Shouse et al. 2012).

Binary Filming Pipe Inhibitors Mechanisms a) b)

Without binary corrosion inhibitor

With binary corrosion inhibitor

These agents penetrate into the microscopic surface of the metal and forms a bond with the oxide film of the metal, becoming part of the metal matrix and thereby resulting in a tenacious barrier that prevents corrosion attacks.

Unlike conventional batch corrosion inhibitors, these filming compounds preferentially adhere to the surface being treated and will subsequently and spontaneously cure under water to a tough, adherent, corrosion-resistant film.

Fig. 6.54—Binary inhibitor mechanism.

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   315

Method of Application

Benefits

Disadvantages

Commercial Status

References

Injection of contemporary inhibitors

Uses current equipment and systems, used on line

Depends on line turbulence for distribution and top-of-line contact

Commercial

Chen et al. (2003)

Injection of advanced inhibitors

Uses current equipment and systems, on line and pigs not needed

Complete surface coverage and film life not proved

Research phase

Schmitt et al. (2001)

Batch treatments with slugs and pigs

Many years of experience and histories

Depends on line turbulence for distribution and top-ofline contact, may require excess inhibitor. Line must be piggable and off line.

Commercial

Bojes et al. (2001)

Foams

Uses less liquid, fills all parts of line, may be used on line

Depends on line turbulence for distribution and TOL contact, film life not proved

Development phase

Achour et al. (2010)

Gels

Excellent surface contact, Off line; with pigs, has can be used with pigs for downside of any pig cleaning treatment

Commercial

Uzu et al. (2000)

Spray pigs

Excellent surface coverage, uses less inhibitor for coverage

Off line, line must be piggable

Commercial

Freeman and Williamson (2006)

Spray pig with two-step Inhibitor

Excellent surface coverage, long film life

Off line, line must be piggable

Development phase

Shouse et al. (2012)

Table 6.7—Comparison of pipeline TLC inhibitors.

The importance of proper surface preparation cannot be overemphasized for effective application and film life of an inhibitor. The inhibitor film will not adhere properly to a heavily scaled surface, so application onto a pipeline surface requires thorough mechanical and possibly chemical cleaning before an application, especially for batch treatments. This is a complex situation; indeed, using laboratory tests of the interactions, Kowata and Takahashi (1996) demonstrated the effects of scale on corrosion inhibitors. The more advanced methods of application (foam, gels, sprays) described above will promote adhesion, but prior cleaning of the surface is nonetheless recommended. See Frenier (2001) and Frenier and Ziauddin (2008) for various chemical solvents. Various cleaning pigs are described by TDW (2011) and PPSA (2008b). Industry documents also suggest cleaning before inhibition (TDW 2011). The various types of cleaning pigs are shown in Figs. 5.8 and 5.9. 6.8  Summary and Lessons Learned Maintenance of pipelines can be enhanced using mechanical devices, through chemistry, and with these methods used together. • There is a large range of corrosion and scale inhibitor molecules that must be tailored to the pipeline flow and chemistry conditions. • Chemicals for inhibition and scale inhibition/deposition are described in detail with mechanisms that have been proposed. • Additional needed chemicals that are described include foamers/defoamers, de-emulsifiers, oxygen scavengers, and H2S treatment chemicals. • Various application methods are selected for the pipeline conditions and include  Gelled chemical pigs and foamed inhibitors.  Injection of chemicals with various mechanical devices used to introduce the chemical and to move it along the line.  Various specialized pigs and pig trains.

316  Chemical and Mechanical Methods for Pipeline Integrity

6.9  Best Practices and Case Histories for Maintenance Treatments of Pipelines Several case histories on the use of maintenance pigging chemicals for treatment of pipelines and surface facilities are described here. 6.9.1 Monitoring and Controlling Corrosion in an Aging Sour-Gas-Gathering System: A Nine-Year Case History. See Section 3.8.1. 6.9.2  Black Powder in Pipeline: Cleaning Program (Sirnes and Gundlach 2012). The authors described the mechanical cleaning of a rich-gas piping system and facilities in the Norwegian sector of the North Sea. They claimed that the business driver for cleaning the pipeline is to be able to perform an in-line inspection and avoid underdeposit corrosion in the pipeline. First operational cleaning runs after pipeline commissioning gave slurry as the pigging product. Some years later, the slurry changed to black powder and caused operational challenges. Because of the low radioactivity and the composition of the dust, some restrictions for the receiving facility were identified. Modifications at the receiving facilities have been performed to be able to clean the pipeline in a safe manner. The main business driver for the modifications at the receiving facilities is to ensure an environmentally friendly and safe operation. The dust has not given any problems in daily operation. After the first incident with black powder, the gas-processing plant installed filter modules to handle the dust in daily operation. Before the start of the pipeline-cleaning activities, additional filter capacity has been installed to ensure a proper protection of the processing plant. The authors claimed that the cleaning program aims mainly to increase the aggressiveness of the cleaning pig after each pig run (Fig. 6.55). In addition, the condition of the pig after arrival, the flow and pressure conditions during the pig run, and the amount of transported dust need to be evaluated. An evaluation group agrees if the next level of aggressive pig can be launched or if a pig of the same aggressiveness level has to be resubmitted. Also, a step back in aggressiveness can be considered and is expected at the later stage of the program. 6.9.3  Beneficial Effects of Chemical Treatment and Maintenance Pigging Programs in Returning an Offshore Pipeline to Pre–Hurricane Ike Conditions Following a Breach and the Ingress of Seawater and Sand, and the Effects of Bacteria-Generated H2S (Powell et al. 2010; Powell et al. 2012). The authors described an overview of pipeline repairs, initiatives to dewater the pipelines, biocide (Section 6.4.3), and H2S (Section 6.5) scavenger treatments, as well as the pigging program that was implemented to re-establish control over SRB and the H2S they generated, thereby minimizing that threat to pipeline integrity. 6.9.4  Integrated Production Chemistry Management of the Schoonebeek Heavy-Oil Redevelopment in the Netherlands: From Project to Startup and Steady-State Production (Shepherd et al. 2012). This paper describes the production chemistry management process undertaken during the design, commissioning, and startup phases of the Schoonebeek redevelopment.

Least aggressive

More contact

Disks and scrapers

Fig. 6.55—Progressive scraper pigs (Sirnes and Gundlach 2012).

Chemical and Mechanical Treatments To Enhance/Maintain Pipeline Operations   317

Wellsite

Demulsifier

Biocide/oxygen scavenger

Antifoam

Scale inhibitor

Corrosion inhibitor

H2S scavenger

From vent gas compressor

Oxygen scavenger To vent gas compressor

To export gas metering

CVR separator

To vent gas compressor

To drain tank

Separator To vent gas compressor

Boiler Wash water

Drains tank

Various streams

Knock out drum

To skim tank

Slop tank To vent gas compressor

Various streams Oil export

To skim tank

Oil export Dehydration tank

Well site

Oil/water/gas CVR fluids

Off gas Wash water

Oil Produced water

Water/condensate/ oil

Wash tank

To vent gas compressor

To water export Skim tank

Fig. 6.56—Overview of chemical usages at a central fluid treatment facility (Shepherd et al. 2012).

Challenging separation issues and saline water, together with a multitude of other process conditions, resulted in a complex but robust application portfolio. This was established during the design stages of the project. Early involvement of the production chemistry discipline aided this process. Chemical selection was conducted in adherence to health, safety, security, and environmental directives and focused on unique produced-fluid properties. Since startup, the success of chemical performance has been the result of the availability of chemical treatment programs and surveillance/ sampling plans. No contingency chemicals have so far been needed at the facilities since startup. Export oil and water key performance indicators have been met most of the time. Further optimization of chemical applications is an ongoing process that will follow the life of the field. Fig. 6.56 provides an overview of the types and locations of chemical treatments required to maintain flow and reliability in the plant. These chemical types are described in detail in Sections 6.1, 6.2, and 6.4.

Chapter 7

Cleaning of Pipelines and Facilities If the use of the various scale and corrosion inhibition treatments (Sections 6.1, 6.2, and 6.3) and the engineering processes are not adequate to maintain flow and integrity, then more-aggressive application of chemicals and mechanical equipment may be required. This chapter describes the use of various pigging procedures that include progressive maintenance pigging and automated equipment for periodic line maintenance. In addition, chemical solvents may be applied using mechanical devices to prepare, renovate and remediate fouled pipelines as well as topside fluid treatment facilities (Sections 2.3.2 and 7.6.2). The chapter’s text references the more detailed descriptions of chemicals and cleaning methods found in Frenier (2001), Frenier and Ziauddin (2008), and Frenier et al. (2010). The short report by Roberts (2015) in the comprehensive handbook by Revie (2015) also describes mechanical devices (i.e., pigs) and some chemical methods. While the chemicals used in cleaning of well tubing and industrial equipment (Frenier 2001) may be identical to those used in pipeline segments, the temperatures that exist in pipelines are frequently much lower than those experienced in refineries (and possibly some parts of topside facilities) or well tubing applications. Thus, solvent selection and testing will be a critical step in the process to develop the most effective and safest fluids. The author of this book notes that mechanical devices such as pigs, pumps, and hydraulic jetting (Section 7.6.2) as well as many of the chemicals, especially surface-active molecules, have been described in Chapters 1 through 6 and that this current discussion is another application of the same mechanical and chemical principles highlighted throughout this book. 7.1  Maintenance Pigging The methods described in this section are used in the normal maintenance program for pipelines without necessarily shutting down product flows for long time periods. In normal or scheduled maintenance, some lines must be pigged on a daily or possibly more regular basis, especially if there is a mixture of gas and condensate. Progressive pigging described later in this section could require several days if the line is heavily fouled. Various types of pigs, shown early in Figs. 5.1, 5.7, 5.8, and 5.9, are also shown in Fig. 7.1, and various units may be used depending on the conditions. A review of the mechanics of pigging in Section 5.1 also is suggested as a prerequisite to the current sections (7.1 and 7.2.). The reasons for performing various cleaning operations include • Maintenance (this section) • Preparation for hydrostatic testing (Section 5.3) or in-line inspection (ILI) operations (Section 5.4) • Preparation for inhibitor or other chemical treatments (Section 6.7)

320  Chemical and Mechanical Methods for Pipeline Integrity

• Major changes such as preparation for a product change • Post-cleaning following a catastrophic event The cleaning process may have to remove any of or all the types of deposits described in Chapter 4 and corrosion products (Chapter 3). SolCleaning pigs ids removed will include rust, mill scale, black powder, and any other solids such as welding rods, wire, dirt, sand, and wax. In many cases, water and liquid hydrocarbons also will be removed. If the line has heavy fouling, it is necessary to assess the thickness and hardness of the deposit as well as the ability for safely conveying a pig through the blockage. It is also Ultrasonic in-line inspection tool important to determine if access at the beginning and end of the line Fig. 7.1—Devices used at Red Hill Hawaii (Regin et al. 2008). segment is possible, the diameter of the line is acceptable, and the lines are not blocked by intrusions such as valves and probes. Leontaritis (2007) discussed some of the issues with pigging wax deposits. Pigging of flowlines or export pipelines is being accomplished routinely for a multitude of purposes in the oil industry, including wax removal. Leontaritis (2007) contended that pigging wax “slush” is a much more risky operation than pigging liquid or even wax deposits in liquid-full lines. In addition to the wax slush, or slurry, there can also be water present in the pipeline, which leads to the formation of a wax emulsion. This may lead to problems with pigging as well. It is the opinion of the author of this book that many total blockages of subsea pipelines are caused by stuck pigs; thus care should be exercised in the design of a pigging job. Wasden (2003) noted that many operators prefer a very conservative approach to wax control, with routine pigging operations (which in subsea systems require a looped flowline arrangement or subsea pig-launching capability) and highly specialized surveillance programs to warn of incipient deposition. To better understand the technology of pigging for wax removal, experiments on the mechanics of wax removal using pigs were conducted by Wang et al. (2001). Different types of pigs were pulled through a cast wax pipe. The wax thickness, oil content, pig type, and pigging efficiency were investigated. They found that the pigging force was divided into three parts: • Baseline force • Breaking force • Wax plug transportation force The baseline force is the force required to move a pig in a clean pipe, which varies with pig types and pig size. The breaking force, similar to the wax shear strength, is the force that causes plastic deformation of the wax layer, which depends on wax thickness, wax properties, and pig type. The wax plug transportation force is the force required to move the wax plug that has been cut off from the pipe wall out of the pipe. The author of this book notes that these types of forces are also present when restarting flow in pipelines, especially if waxy oil is present.

Cleaning of Pipelines and Facilities  321

Tordal (2006) described the planning needed for successful pigging of a pipeline containing highwax-content crude oil: 1. Before a pipeline with high wax deposit is being pigged, a pigging program has to identify the type and numbers of pigs. The use of a nonsuitable pig can, in the worst case, lead to full pipeline blockage. 2. As the pig enters the pipeline, it removes the wax deposit of the pipe wall and pushes the wax in front of the pig. As the pigging length increase, the volume of wax also increases in front of the pig. 3. The wax deposit can, depending on pipeline size, length, and volume of wax, build up to a plug several hundred meters long. 4. During pigging, the wax deposit can build up in front of the pig until the differential pressure required for driving the pig equals the pigging pump pressure, causing a full pipeline blockage to occur. To reduce the chance of a stuck pig, Tordal (2006) recommended pigs that contain holes to allow oil to bypass the pig. To illustrate this idea, a diagram and photograph of a bypass pig designed for wax removal is shown in Fig. 5.18. Lindner (2006) also recommended the use of a bypass pig to remove mixed deposits that contain black powder (Section 4.4.4). One method of reducing the risk of a stuck pig is by practicing what is called “progressive pigging.” In this method, initially a soft foam pig is sent through the line, followed by more-aggressive cleaning pigs. The returns from the soft pig are used for determining the possibility of using moreaggressive pigs. Shouse (2010) suggested that a progression of mechanical devices can be run through a line to determine the ability of the line to take a cleaning process and then to progressively remove deposits for achieving the goal of a treatment. A typical sequence of progressive pigging, according to Shouse (2010), may include • • • •

Foam wipers to determine the initial clearance in the line A geometry tool to determine passage of further pigs (Section 5.4.2) Foam scrapers with bristles (with detergent solutions) Mandrel tool (Figs. 7.1 and 5.1) with  Scrapers and cups (Section 7.3.5)  Aggressive brushes and detergents • Gauging tool (Section 5.4.2) • Tool with magnets (Fig. 5.8) • Geometry tools and/or ILI tools, depending on the goal Also see the discussion by Fretwell (2007) on the steps in a progressive pigging plan. This author suggests several pig selection criteria for progressive pigging that include • • • • • • • • •

Pipe size Length of the pipeline section Minimum bend radius used in the construction of the line Product being transported by the pipeline Number of changes in pipeline inner diameter (ID) because of wall thickness changes Valve types Pipeline location (e.g., subsea or cross country) Pig trap design Type of debris to be removed

322  Chemical and Mechanical Methods for Pipeline Integrity

In addition, see pig selection discussions in Section 5.2.3. Several examples of maintenance pigging are noted next. Plaziat (2013) described the use of several pigs and ILI devices to clean and inspect a gas line in Europe. The 48-in. line ran for 1200 km, and the passage of the tools pushed by the gas flow took 3 to 4 days. The author claimed that during the inspection process, three different tools were used: • A gauge tool, that is, a tool with gauge plates (Fig. 5.7), for finding protruding obstructions before use of the other devices • A cleaning tool • An ILI tool, which maps potential corrosion and metal loss as well as the exact curvature of the lines by means of an inertial navigation system The author then reported that a cleaning tool that had brushes as well as magnets was used. A gas bypass feature allowed for speed control (see Section 5.2.2). Material pushed out was then analyzed to determine possible corrosion mechanisms. The final ILI run had a magnetic flux leakage (MFL) unit, and distance recording, as well as caliper and shallow internal corrosion sensors. He claimed that the integrity of the line was confirmed using these tools. Regin et al. (2008) discussed the project management and technical challenges for the metal-loss inspection of the 813-mm (32-in.) diesel pipeline at Red Hill, Pearl Harbor, Hawaii. The pipeline has several “unpiggable” features such as 28° mitered bends, reduced valve sections, and no launcher/ receiver. Even though the pipeline is aboveground, it rests directly on over 300 supports and is encased in more than 20 bulkheads, each 4 ft wide. Various types of cleaner pigs (Figs. 5.8 and 7.1) and ultrasonic devices (Section 5.4) were then used to assess the corrosion in these lines. Kinnari et al. (2007) claimed a “thruster pig”-based method (see Fig. 5.19 and Section 5.2.2) for removing hydrates that includes inserting a thruster pig connected to a return flowline into the pipeline. The pig is advanced forward into the pipeline by pumping a thrusting fluid into an annulus between the pipeline and the return flowline while removing deposits continuously or intermittently and returning flow as appropriate from ahead of the pig through the return flowline. The return flowline is then disconnected from the pig after sufficient deposit removal and retrieved from the return flowline. Next, the pig may then be transported through the pipeline by the pipeline fluid to a pig-receiving location once the pipeline flow is resumed. The author of this book does not know if this method is in wide use. Additional literature on maintenance pigging has been provided by TDW (2011), Pirtle (2007), and Cowan and Weintritt (2004). Also see a University of Tulsa report (TU 2004) that gives a review of pigging technology. Pig “trains” are another option; these combine mechanical and chemical cleaning. See Frenier (2001), who discussed various chemical pigs and pig trains in which stages of chemical solvents provide removal of inorganic scales (Fig. 5.7). Pig trains have also been used for removing organic deposits. More details of chemical pig trains are presented in Section 7.5. 7.2  Automated Methods and Mechanical Equipment for Maintenance Cleaning The launch/recovery devices described in Section 5.5 (see Figs. 5.50 and 5.51) provide the capability of launching/receiving a single maintenance pig or ILI. However, a number of processes have been engineered to automatically inject into a pipeline different types of devices, including spheres and mandrel devices (see Fig. 5.1). These systems can also inject several tools sequentially from various types of storage units. Section 5.5.1 describes the basic functions of traps and device launchers, and the automated devices are modifications of these types of equipment. Multipig and automated processes can be much more efficient for injecting some tools into a pipeline compared with manual processes. An advantage is that a multipig automated process may be safer than manually injecting each device separately because launcher system components can remain pressurized, even when the launch chamber is depressurized to receive the next device.

Cleaning of Pipelines and Facilities  323

In addition, spheres or foam pigs and other devices can be injected on the basis of a chosen time schedule or controlled from a remote location. An analogy is the comparison of a breach-loading single-shot cannon with an automated, multiround system. Several automated systems are described in Sections 7.2.1 through 7.2.3. However, all of them must have processes to move a pig into the injection position, propel it into the line, and then move the next device into the injection position. The remaining pigs must also be separated or restrained in some manner. 7.2.1  Horizontal Multiple-Pig Launchers. Warriner (2008) described several types of automated systems. Some of the devices and techniques have been engineered specifically for wet gas and other pipeline systems that require routine pigging, especially to prevent liquid holdup that limits flow and promotes wax buildup. Automated systems may also be needed in areas such as subsea, where physically opening/closing valves and other devices may be limited to specific conditions of the line and environment (Kongsberg 2015). Some of these devices use various sets of valves, or other methods, that are actuated for each pig in the launcher (Warriner 2008). They can be operated manually (or by a remotely operated vehicle) or by a remote system. Kongsberg (2015) described a valve-type subsea automated pig launcher system that is designed for individual launching of pigs from subsea, remotely controlled or from a platform. This system uses a cassette of several pigs. Because these systems are horizontal, they must use gas or fluid pressure differentials to launch the pig, whereas vertical systems may also use gravity to inject the pig. Tarlton (2002) discloses an automated subsea launcher system that has a rotating magazine for storage of several pigs. A valving/launcher system includes a dual-bore collet connector for coupling to the launcher manifold. During launching, a kicker valve is opened to bypass the throttle valve and direct the higher upstream pressure through the secondary bore of the dual-bore collet connector and into the rotary magazine of pigs. Davis (1992) claimed a horizontal multiple-pig-launching system that incorporates a tubular magazine for containing several pigs. A source of hydraulic fluid medium is provided and that fluid is injected into the magazine behind the free piston for movement of the free piston and the pigs for pig launching. A gas-energized liquid pump is operated by gas pressure from the gas transmission pipeline under the control of a timer-operated gas supply valve for selective introduction of hydraulic fluid into the magazine. 7.2.2  Vertical Multiple-Pig Launchers. A vertical system allows for loading multiple pigs that are separated by hydraulically operated pins or a valving system. In the systems described by Warriner (2008), each pig is launched by retracting the pin after the first pig is launched. A manually operated or automated hydraulic system is needed to launch each device. Another example of a pin system is described by Rajabali and West (1999). In this disclosure, the pigs are stacked vertically and are controlled by a pair of axially spaced actuators for release one at a time into the pipeline. The operation of the actuators is effected by a control system using high-pressure nitrogen gas; this system can be remotely operated to effect launching of a pig into the pipeline at any desired time without the need for on-site personnel. Bath and Yemington (2000) describe a vertical multipig-launching system that uses a “pig gate” that is actuated by changes in hydraulic pressure to isolate one pig in a vertically stacked system. At launch time, the gate opens and allows a single pig to be launched into the line. Disher and Chernuka (2002) disclosed a device that has a circular rotating magazine that allows pigs to be dropped through a valve opening (Fig. 7.2). The authors claimed that the supply magazine is a generally circular disc, with several holding chambers spaced in an organized way around the disc so that the holding chambers can each in turn be positioned over the pig-launching chute such that any pigs contained therein can drop by gravity into the chute.

324  Chemical and Mechanical Methods for Pipeline Integrity

Pig inlet Rotating shaft Mechanism

Supply magazine

(a) Pig chamber Holding chamber

Holding chamber

Rotating shaft

Supply magazine

Floor

Holding chamber Floor opening Piglaunching chute

Holding chamber Pig (b) Rotating mechanism inlet Floor Rotating shaft Opening Pig-launching chute

Fig. 7.2—Rotating pig magazine (Disher and Chernuka 2002).

They contend that the supply magazine functions both in pig storage and in pig delivery to the pipeline. In this case, many pigs can be loaded into the supply magazine at once. In addition, because the disc is oriented on a generally horizontal plane, the apparatus is vertically compact, which avoids the height problems that are encountered in vertical storage mechanisms. Wilkinson (2011) described an automatic multiplepig-launching (AMPL) system that had been developed and has undergone a commercial use. The system individually launches multiple pigs from a preloaded “cassette,” shown in Fig. 7.3). This figure shows a set of mand­ rel pigs ready for launch.

Fig. 7.3—Automated multiple-pig-launching system cassette of pigs (Wilkinson 2011).

Cleaning of Pipelines and Facilities  325

The author noted that each AMPL pig has its own hydraulic launch Same-size spheres control mechanism that is designed 24 in such a way that the next pig to be 24 24 launched is armed only when the 24 launcher has been depressurized Sphere detector after the previous launch, and so the next pig cannot be launched 16 18 accidentally. This launcher has 20 Sphere detector an additional safety system in place so that if the control 16 18 20 24 mechanism fails, no pigs can or Graduated-size spheres will launch. The system is claimed to require no modification to Fig. 7.4—Automate sphere launcher with graduated spheres the existing pipeline launcher (Vinson 1968). because the pig-launching process is controlled by a hydraulic system incorporated with the pigs rather than by using complex pipeline on the launcher. McNabney and Marbach (2014) disclose a system that has a rotating magazine that contains one or more pigs. The automatic barrel pig launcher comprises a plate bore, bottom plate bore, upper case bore, and through bore of a ball valve that when aligned provide loading of the pig into the through bore of the ball valve. H

H

H

H

H

H

H

H

H

H

H

7.2.3  Automatic Sphere Launchers. Several types of spheres (balls) are used to remove liquids from multiphase gathering lines (see Fig. 5.1 and Section 5.2.3). Various multiball launchers have been proposed. The need for this form of automatic pigging was described by Vinson (1968) as a method to control liquid slugs in wet-gas gathering lines. The author noted that each sag in the line (see Fig. 1.8) contributes an additive pressure-drop equivalent to the head of liquid held in its downstream uphill section. The liquid head in each sag varies inversely with the gas velocity. The author suggested that automatic on-stream spherical pigging (sphere example in Fig. 5.1) offers a practical solution to the pressure-drop problem. The solution consisted of an automated sphere launcher. A schematic of the launcher is presented in Fig. 7.4. The launcher is automated to open/close valves to discharge a sphere at set times. In this process spherical pigs matching the pipeline ID pushes a train of smaller-diameter spheres ahead of it. Payne and Wint (2012) claimed that automated vertical launchers can release a spherical pig using the hydraulic-actuated pin system; see also Rajabali and West (1999). TDW (2011) and Wint (2013) explained that an automated combination pigging system delivers more options, more control, and more throughput compared with single-pig launchers. Improvements claimed include that such a system • Has the ability to launch a pigging program automatically • Requires less labor to launch a pig • Releases spherical pigs one at a time at predetermined intervals, as shown in Payne and Wint (2012), with a launch (Fig. 7.5a), recovery (Fig. 7.5b), and pin release mechanism • May also launch a combination of standard cleaning pigs, batching pigs, or ILI tools • Allows multiple spherical pigs to be loaded at one time • Eliminates need to open/close valves to conduct a launch with a flow-through barrel design • Increases valve life and reduces maintenance cost • Reduces gas releases to the atmosphere and explosive environments • Through reduction in opening/closing launchers/receivers, reduces the risk of injuries to personnel

326  Chemical and Mechanical Methods for Pipeline Integrity

Connection of valve and pipe must be smooth and inclined so spheres will roll freely Launching valve

Oversized barrel Pressure equalization valve

Reducer

Pressure equalization line

Blowdown valve

Guide bars to prevent spheres from being sucked Into blowdown opening

Pig signal Blowdown valve Receiver valve Pig signal

End closure

Actuators for pins

Eccentric reducer

Oversized barrel

Guide bars to prevent spheres from being sucked Into blowdown opening

(a) Sphere launcher

Guide bars designed so that there will be no diametrical squeeze on the sphere and so that the sphere will “stand-off” from branch opening and rolling freely into valve

Suction valve

Drain and pressure equalization valve

Drain and pressure equalization line Dual connections spaced so that it is impossible for one or more spheres to block off both connections

(b) Sphere receiver

c) Photograph of sphere launch mechanism

Spheres

Retention pins

Fig. 7.5—Automated ball launcher (a), receiver (b), and photograph of mechanism (c) (Payne and Wint 2012).

Single-pig launchers, as well as the multiple, automated systems described in this section, enable operators to maintain flow in their pipeline systems. Using equipment such as seen in Fig. 7.6. Fig. 7.6a shows an automated system with a bypass loop, and Fig. 7.6b is a photograph of this type of rig on a site. The insert shows several types of cleaning units that may be used in this type of equipment. Zellou (2015) claimed that the benefits are • • • • • • •

Increased operational efficiency Reduced manpower Reduced downtime Reduced energy consumption Decreased gas emission Regulatory compliance Enhanced safety

When these processes are not sufficient for providing clean surfaces, a very wide array of chemicals can be used to remove unwanted solids and liquids. These are described in Sections 7.3 through 7.6. The author of this book has provided the figures for a description of one type of automated cleaning unit launcher, but other options of different solutions are possible. 7.3  Chemical Solvents and Mechanisms for Cleaning Pipelines and Facilities This section describes various chemical solvents that may or may not be staged with pigs for various cleaning tasks in pipelines and in units in facilities. Solvents for inorganic solids and organic foulants do not have the same chemistries, given that the chemical structures of the solids are markedly different. Section 7.3.2 describes solvents that are formulated for inorganic salt foulants that include iron compounds (oxides, sulfides, and carbonates), as well as calcium carbonate and calcium and barium and sulfate scales (Section 4.2).

Cleaning of Pipelines and Facilities  327

(a) Diagram of ball launcher

Ball launcher

Valve

Bypass loop

Launcher controller (b) Photograph of multipig launcher and pigs

Mandrel pigs Spheres Fig. 7.6—Automated ball launcher with bypass (a) and photograph of launcher and pigs (b) (Wint 2013).

Separate sections then describe solvent formulations for organic solids (see Sections 4.3 and 7.3.5) and mixed deposits (see Sections 4.4.4 and 7.3.6). Then, testing methods to develop appropriate solvents (Section 7.4) for the total fouling matrix are described. 7.3.1  Introduction to Solvent Chemistry. For total chemical dissolution of any deposit, the solvent needs to have enough capacity to bring all the molecules (or ions) into a true thermodynamic solution. Depending on the deposit to be removed and the solvents available, chemical cleaning may produce the cleanest surface with the least damage to the metal of construction. However, because many of the solvents for inorganic solids will react with (corrode) the surface as well as with the deposit, corrosion control (i.e., use of special inhibitors) may complicate the process. Solvents for removing only organic solids, such as wax or asphaltenes, may not be corrosive; however, they are often used in stages with or in emulsions of aqueous corrosive liquids. In addition to the complete dissolution of an organic solid, aqueous (or oil) mixtures of detergents (surfactants) are also frequently used to disperse the solid. Much of this section has been abstracted from Frenier and Ziauddin (2008) and Frenier and Ziauddin (2014), and aspects apply to most inorganic solids as well as to crystalline organic materials such as paraffin wax and gas hydrates. Listed in Fig. 7.7 are general types of solvent chemicals currently in use. The figure identifies solvent types for mineral scales as well as for organic and mixed deposits. Note that various formulations and temperature ranges are required for different scales and operating conditions. A preliminary example is shown in Fig. 7.8. It describes the dissolution of iron oxides from a pipeline surface at a low temperature and demonstrates the importance of the solvent choice.

328  Chemical and Mechanical Methods for Pipeline Integrity

Major Scale Removal Chemicals Cleaning Agent

Application Temperature, °F

HCl 5–15%

Application, Dissolution

6 in.), there will be more volume/capacity of solvent to dissolve more scale. This is a critical calculation needed during job planning. Examples of capacity tests are found in Section 7.4.2.

Cleaning of Pipelines and Facilities  331

The morphology of a “real” scale may differ substantially from case to case depending on hardness, permeability, and the presence of any foreign agents. This may greatly affect the practical dissolution rate. Thus, a mixture of calcite and gypsum will have a different overall rate of removal compared with the pure phases. The heterogeneous nature of the scale may help or hinder final removal. If one component is very soluble, it may be possible to dissolve that component and flush the rest out of the system being cleaned. On the other hand, the presence of an impervious outer layer of barite or even an organic deposit may greatly hinder the overall removal. Campbell (2000b) claimed that the penetration of a pipeline deposit by a solvent or dispersant may greatly affect the fluid’s ability to remove the deposit. This is also a function of the morphology of the deposit noted in the preceding paragraph. The author tested water-soluble as well as oilsoluble cleaners on one (unidentified) pipeline sludge and found a marked difference in the time the solvent penetrated the deposit. This would affect cleaning time. He also noted that detergents must also suspend the deposit to remove it from the unit. One of the most important engineering requirements involves obtaining a representative sample of the scale to allow analysis and then solvent testing. The author of this book realizes that this may be an extremely difficult problem to solve in the pipeline environment (unless a pig has been run through the line), but every practical avenue must be pursued to obtain a sample of the fouling scale. Solvents for inorganic solids (described in Section 7.3.2) will usually be formulated with a corrosion inhibitor (Section 7.3.4) to protect the steel. Wetting agents—these may be various surfactants (Section 7.3.5)—will also be part of the formulation. These materials affect the solids wetting conditions, may promote penetration, and also may affect the overall rate of dissolution. Shank and McCartney (2013) tested various surfactants as well as different inorganic/organic acids as solvents for CaCO3 and observed that the acid strength (see Table 7.1) as well as the surfactant affected the rate. The authors found that low interfacial surface tension (IST) fluids spread and attacked the carbonate, but they also caused a foam to form that may impede dissolution. The flow patterns of the fluid also affect these dissolution properties. The next subsections (7.3.2 and 7.3.3) provide more details of solvents for two different types of inorganic scales: acid soluble and acid insoluble. A wide range of solvents for inorganic oilfield deposits have been developed (Frenier and Ziauddin 2008), and some of these are reviewed here as appropriate for selected pipeline or facilities situations. This next section starts with a short review of inorganic solids dissolution chemistry and reviews dissolving oxides, sulfides, and carbonates (i.e., acid-soluble) and acid-insoluble sulfates. 7.3.2  Chemicals for Removing “Acid-Soluble” Inorganic Solids. This section describes the solvents available for the acid-soluble scales that include calcite, siderite, various iron “oxides,” rusts, and iron sulfides. However, the capacities of the solvents for each scale as well as the reaction rates and the response of each reaction rate to temperature vary greatly. A large number of solvent molecules can possibly be used in some situations. Chemical characteristics are found in Table 7.1, and structures of solvent molecules are presented in Fig. 7.10. All these chemicals have acid functionalities; however, only some of them are available commercially in a liquid acid formulation that is useful in many low-temperature pipeline conditions. Mineral Acids. HCl is a widely produced chemical that can be used at concentrations from approximately 5 to 15% for dissolution of carbonates as well as various iron salts. The reaction products include FeCl2, CO2, H2, H2S, CaCl2, and MgCl2, and these are all quite water soluble (the chlorides are approximately 50% soluble in water). Therefore, the formation of a precipitate or a separate CO2-rich phase is generally not a problem. This acid can be used on many types of oilfield-related and process equipment. It can be inhibited at temperatures up to approximately 250°F (122°C). However, HCl solutions are not used to clean Series 300 stainless steel, free machining alloys, magnesium, zinc, aluminum, cadmium, or galvanized steel.

10

0.1

0.5

40

HEDTA—hydroxyethyl ethylenediamine-triacetic acid

EDTA—ethylenediaminetetraacetic acid

DTPA—diethylenetriaminepentaacetic acid

GLDA-L-glutamic acid, N,N-diacetic acid

263

393

292

278

192

60

46

97

98

20

36.5

Acid

351

503

380

344

258

Na salt

MW

9.36

10.5

10.2

9.8

5.7

4.8

3.75

1.0

–5

3.2

–7

pK1

5.0

8.5

6.1

5.4

4.4

2

pK2

3.5

4.3

2.7

2.6

2.9

pK3

pKa Values

2.6

2.6

2.0

pK4

1.8

pK5

Table 7.1—Characteristics of reactive chemicals (data from Martell and Smith 1978; Jones and Williams 2001).

v.

v.

Formic acid

50

v

H2N-SO3H Sulfamic acid

Acetic acid

v.

H2SO4

Citric acid

v.

v.

HCl

HF

Compound

Sol. Water, pH 2 or less (wt%)

11.7 (est.)

28.0

25.0

19.8

11.5

3.4

3.1

Fe (III)

5.9

10.9

10.7

8.4

3.5

1.18

1.43

Ca (II)

8.9

7.8

6.2

2.6

1.07

1.38

Ba (II)

LogK Equilibrium Constant

18.7

16.5

14.4

8.0

1.5

1.36

Al (III)

Cleaning of Pipelines and Facilities  333

O H

H

Cl

F

HO

HF

HCl

S

OH

HO

HO O Formic acid

O Sulfuric acid

O Acetic acid

O

OH

O HO

S

HO

O

NH2

Sulfamic acid

OH

HO

HO

O

Glycolic acid

OH

O O Citric acid- CA

HO

OH

O

O

HO O

HO

N N

N N

O

OH

O

O OH N-(hydroxyethyl)-ethylenediaminetriacetic acidHEDTA

HO OH Ethylenediaminetetraacetic acid- EDTA

OH

O

O

HO OH

O N

HO

HO O

O O

L-glutamic acid, N, N-diacetic acid- GLDA

N

N

OH

HO

O

O

N

OH

O Diethylenetriaminepentaacetic acid- DTPA

OH O

Fig. 7.10—Structures of solvent molecules for inorganic solids.

Dangers of general or localized attack on the construction metals limit the use of HCl for cleaning the previously mentioned metals. Most other construction metals can be adequately protected during cleaning using corrosion inhibitors (see Section 7.3.4). HCl will dissolve carbonates, phosphates, most sulfate scales, ferrous sulfide, iron oxides, and copper oxides. Also, using additives (including HF) , one can remove copper, clays, and silica from surfaces with inhibited HCl. In some cases in which Series 300 stainless steel is present, dilute solutions of sulfuric acid (H2SO4) can be used to dissolve iron oxides and iron sulfides, but not calcium-based deposits. Organic acids and chelating agents (discussed in the next subsection) can be used on many sensitive metals to remove inorganic scales.

334  Chemical and Mechanical Methods for Pipeline Integrity

Although the reaction rate depends on the temperature (Eq. 7.1), HCl can be an effective solvent, even at the low temperatures found in some pipeline sections or facilities for dissolving carbonate and some iron salts. However, the very acidic solvents must be flushed from the units and the surfaces then flushed with alkaline solutions such as sodium carbonate and small amounts of NaOH. These solutions may contain small amounts of NaNO2 to help passivate the surfaces. All the solutions must then be neutralized and sent to a disposal site. See information in Frenier (2001) as well as Section 8.4.1. The carbonates, calcite, and siderite are the most easily soluble, even at low temperatures. The general reaction for calcite is 2H+ + CaCO3 = Ca 2+ + CO 2 + H 2 O ������������������������������������������������������������������������������������������ (7.4) The reaction for HCl is 2HCl + CaCO3 = CaCl 2 + CO 2 + H 2 O �������������������������������������������������������������������������������������� (7.5) The reaction for HCl with siderite is 2HCl + FeCO3 = FeCl 2 + CO 2 + H 2 O���������������������������������������������������������������������������������������� (7.6) Note from Eq. 7.1, and Fig. 7.9 that HCl will dissolve iron oxides, even at low temperatures. Iron sulfides (FeSx) can be found in many parts of the production, connection, facilities, and transport systems. At the higher temperatures that may be found in H2S removal equipment and during long periods of contact, iron sulfide deposits (generally FeS, but sometimes FeS2) will form. See Section 3.1.4. The low-sulfur-ratio sulfides (close to stoichiometric ratio of Fe and S) can easily be dissolved using mineral acids: FeS + 2H+ = Fe 2+ + H 2 S ������������������������������������������������������������������������������������������������������������ (7.7) The reaction rates depend on the exact mineral present as well as the temperature; however, if solvent temperatures cannot be maintained at 150°F, chemical dissolution may not be an option. HCl alone and HCl with additives are commonly used to clean parts of surface-based gas removal facilities (Frenier 2001). Mackinawite and pyrrhotite react rapidly; other sulfides such as pyrite are essentially insoluble in acid. While the acid reaction is a highly effective procedure, it produces large amounts of toxic hydrogen sulfide. This gas produces severe safety and operational problems. Lawson et al. (1980) reviewed the major procedures for safely removing iron sulfide deposits: • Mineral acid with an acid/gas scrubber • Mineral acids with H2S suppression chemicals • Multiple stages of oxidizing agents with acids • Alkaline cleaners If none of the suppression/control technologies is employed, a surface vessel being cleaned must be vented through an NaOH scrubber with enough capacity to absorb all the H2S during the shortest possible reaction time. As an alternative, the vessel can be vented to a flare. In both cases, it is advisable to vent through a knockout tank to remove any entrained liquid. During cleaning of well tubulars, the H2S evolved makes the acid fluid extremely corrosive to the steel and presents a hazard when the well is flowed back. Fatalities to service crews have occurred.

Cleaning of Pipelines and Facilities  335

Also, pigging operations can be affected by toxic gases, and especially H2S and the toxicity of “pig trash” (Figs. 5.22 and 5.50). Thus, the venting of gas before opening a pig trap must be accomplished carefully. The gas can be checked with an H2S meter, and personnel must have sufficient protective equipment (see Section 8.2.2). Several different suppression technologies are available as additives to mineral acids, such as HCl or H2SO4. Buske (1981), Frenier (1982), and Ball and Frenier (1984) developed suppression agents that contain aldehydes. The most efficient agent is formaldehyde, which reacts stoichiometrically with hydrogen sulfide to produce trithiane, a very insoluble material. Sometimes, a backup scrubber system is used to ensure complete removal of sulfide gas. O R1

C

S

H H

+

H2S

3 CH 2 O + 3 H2 S

R1

C S

H

+

H S

H2 O

..... (7.8)

S Formaldehyde

Hydrogen Sulfide

Trithiane

Tests conducted in the laboratory and backed up with treatments demonstrated that formaldehyde used with sulfuric acid can dissolve FeS with essentially 100% suppression of H2S evolution. There are several problems with the use of formaldehyde. This chemical is listed on several compilations of suspected carcinogens. In addition, formaldehyde can react with hydrochloric acid to form chloromethyl ethers (Frankel et al. 1974); these are known human carcinogens. Further, sulfuric acid cannot usually be used in downhole operations because of concerns with formation of calcium sulfate deposits. Because of concerns about formaldehyde toxicity, this chemical was replaced by glyoxal. The chemical, glyoxylic acid (Buske 1981), can be used with hydrochloric and sulfuric acids. Formaldehyde can also be generated in situ by adding hexamethylenetetraamine to strong acid. N N

N

N

+ 4 H

+ 6 H2 O

6 CH 2 O

+

4

NH 4

.������������������������ (7.9)

Hexamethylenetetraamine

Additional less-toxic suppression agents that include mixtures of glyoxylic acid, cinnamaldehyde, and glyoxal (Fig. 6.30) have been used in some situations in the petroleum industry. Frenier and Hill (2002) developed a scavenger system that does not significantly inhibit FeS dissolution that contains glyoxylic acid and an unsaturated aldehyde. This system uses glyoxal or preferably glyoxylic acid and cinnamaldehyde to dissolve FeS, capture the H2S, and lower the corrosion rate to acceptable values. All these materials have been formulated in HCl. Miller (2004) claimed an FeS solvent cleaning composition that consists of formaldehyde or glutaraldehyde, methanol, 2-butoxyethanol, isopropyl alcohol, hydrochloric acid, surfactant intermediate, surfactant/corrosion inhibitors, glacial acetic acid, and water. This formulation is a combination of previous technologies but adds an organic acid (acetic acid). Nasr-El-Din et al. (2000a) claimed that a drawback of aldehyde-based scavengers in acidizing treatments is the formation of oily material, which may adversely affect the performance of water injectors.

336  Chemical and Mechanical Methods for Pipeline Integrity

This paper discusses a scavenger that is uses hydroxyalkyl triaziane (Fig. 6.31). Laboratory work by these authors included compatibility of the old and new hydrogen sulfide scavenger with live acids. The impact of the new scavenger on corrosion of low-carbon steel was also addressed. They claimed that on the basis of the encouraging results from the laboratory, field trials were conducted on a number of wells. Post-acidizing flowback samples were collected and analyzed to confirm laboratory results. Finally, injectivity results from the field trial were compared with offset-well performance to validate the technical benefits of the new hydrogen sulfide scavenger. Examples of field use of these methods in the pipelines and facilities are included in the following reference. Dilute aqueous solutions of sulfuric acid (H2SO4 at approximately 10 to 15 wt%) were used in conjunction with the aldehydes (Frenier 1982). Frenier et al. (1980) described the use of this solvent with aldehydes described above to safely remove FeS-containing deposits from refinery and oilfield gas-treating facilities. These solutions can be used on metals (such as Series 300 stainless steels) that may be attacked by HCl-based solutions. In an example described by Ball and Frenier (1984), a heat exchanger in a refinery, fouled by FeS deposits, was cleaned using a solution of 10% H2SO4 and an aldehyde, at approximately 125°F. The metal analyses from the solutions showed that more than 4,000 lbm of FeS was removed without H2S being detected. An example of a field treatment of piping fouled with FeS is presented in Fig. 7.11, which shows the fouled pipe section (Fig. 7.11a) and the Fe/time dissolution curve for this treatment (Fig. 7.11b). From these data, the system was cleaned in approximately 7 hours of circulation (see more information on application methods in Sections 7.5 and 7.6). Note that heat exchangers as well as piping fouled with FeS exist in gas-treating plants (Section 2.3.2) as well as in many other parts of the pipeline systems. Note, also, that the setup described by Ball and Frenier (1984) included a gas scrubber containing a dilute NaOH solution as a backup (see layout and circulation chemical details in Section 7.6.2). Similar jobs (known to the author of this book, but not published) have been accomplished in oilfield gas plants. It is extremely important to know that the sulfuric acid descaling solutions require a different corrosion inhibitor family compared with an HCl inhibitor. Chemistries similar to those used for organic acids and chelating agents (Section 7.3.4) must be used in H2SO4 solvents. Note that flushing, passivation, and disposal processes described for HCl also must be conducted for this acidic solvent. However, a more recent publication on the use of THPS as a solvent suggests that a polymeric additive with THPS may also be effective. A patent application by Talbot and Jones (2015) claims (a) Pipe fouled with FeS Pipe wall (b) Dissolution of FeS in aldehyde/ 10% H2SO4 0.7 0.6 [Fe], wt%

0.5 0.4 0.3

Fe

0.2 FeS Scale

0.1 0

0

1

2

3

4

5

6

7

8

Time, Hours Fig. 7.11—Field cleaning of FeS-fouled piping at 125°F (data from Ball and Frenier 1984).

Cleaning of Pipelines and Facilities  337

that adding an acidic polymer such as a polyacrylate terminated with vinylidene-1,1-diphosphonic acid (VDPA ) to THPS will dissolve FeS. Fig. 7.12 describes the structure of THPS (a) and part (b) of this figure shows the reactivity of THPS formulations with different additives. The data from the applications indicates that the more acidic additive dissolved more FeS than THPS alone. Mattox (2005) also claimed that the solutions of THPS/ammonium chloride react with and complex soluble and insoluble iron(II) sulfide compounds through a series of chemical reactions (not described) at ambient temperature, resulting in the solubilization of the iron(II) compounds. This author further claimed that the formulation can be pumped through pipelines as slugs between pigs as a pig train (see Section 7.5.1 and Fig. 6.45). Additional experimental FeS dissolution formulations are claimed to be based on chelating agents similar to those proposed for calcite and iron oxide dissolution. See the subsection on organic acids and chelating agents as well as Yap et al. (2010) and Wang et al. (2013). Although these alternative high-pH formulations appear to be promising in laboratory tests, the author of this book cannot document any field tests or commercializations (or relative costs) of these alternative formulations on FeS removal in pipeline/facility environments.

(a)

Tetrakishydroxymethylphosphoniumsulfate (THPS) 3.3

90 80

3.25

(b)

3.2

60 50

3.15

40

3.1

pH

% FeS Dissolved

70

30

3.05 FeS pH

20

3

10 0 THPS

VDPA

THPS + 5% VDPA

295 THPS + 20% VDPA

Dissolution of FeS and Solvent pH, for 20 Hrsat 50 oC Fig. 7.12—Structure and reactivity of THPS with FeS (Talbot et al. 2002).

338  Chemical and Mechanical Methods for Pipeline Integrity

Other Inorganic Acids in Cleaning Operations. Sulfuric acid (H2SO4) and sulfamic acid (Fig. 7.10) can be used to clean chloride-sensitive metals or as part of a chelating-agent-based formulation (see the next section). See also Frenier (2001). Organic Acids and Chelating Agents. Formulations with organic acids and chelating agents may be considered for very sensitive situations in which concentrated HCl or other mineral acids may not be acceptable. These situations could include removal of scale from units that contain Series 300 stainless steel that is susceptible to chloride stress corrosion cracking (SCC) or high-temperature applications (as in some gas treatment plants). Acetic acid as well as formic acid formulations are in use for cleaning surfaces; however, low reaction rates at low temperatures may limit uses when the surface temperatures exceed approximately 120°F. Fig. 7.10 and Table 7.1 show the structure and chemical characteristics. These formulations are capable of dissolving iron oxides, carbonates, and possibly FeS (Yap et al. 2010). Chelating agent solvents that are in use in the oil field and facilities include salts of citric acid (Na+, K+, and NH4+), EDTA, HEDTA, and GLDA. Low-pH formulations of HEDTA and GLDA are in use or are proposed for dissolution of acid-soluble deposits, including iron oxides and calcite. See Frenier et al. (2000) and DeWolf et al. (2010). The dissolution of alkaline earth carbonates by acids is one of the most investigated surface reactions, especially as it is applied to acidizing and permeability stimulation by wormhole formation. The details are beyond the scope of this book. See the reviews by Fredd and Fogler (1998), Fredd (2000), and Hill and Schechter (2001). Additional studies that are applicable for calcite scale removal and iron oxide and iron carbonate are described below. If the solvent has available protons, the reaction with the carbonate to produce CO2 also acts to help break up the scale lattice because the structure of the crystal is destroyed. Because the surface reaction rate is not a major factor in the overall process, especially at elevated temperatures, the key consideration is the capacity of the solvent, and this is controlled by the solubility of the calcium salt formed by this reaction, as well as the molecular weight of the anions. Calcium chloride formed from HCl is very soluble. Com3 mon organic acids such as formic acid and CaCO dissolved Dissolver (lbm/1000 gal) acetic acid will also readily dissolve calcium 20% Trisodium HEDTA (pH 2.5) 738@ carbonate, but they will form insoluble precipitates at high enough concentrations of calcium 20% Tetrasodium EDTA (pH 12) 520@ to limit the capacity of organic acid solvents. 21% Diammonium EDTA (pH 4.5) 735@ The molecular capacity of the solvents is like20% Disodium HEIDA (pH 2.5) 250@ wise important. HCl has the lowest ratio of 20% Tetrasodium GLDA (pH 1.5) 1600# molecular weight to calcite molecular weight 15% HCl 1833@ of any practical solvent, so the theoretical 10% Formic acid 845@ capacity is the highest for any solvent. Table 7.2 below shows laboratory data for dissolu5% Citric acid 5@ tion tests performed at atmospheric pressure 20% Trisodium NTA (pH 4) 5@ that compare capacities of a number of possible 10% Acetic acid 422@ solvents. They reflect realistic solvent concenKey: @ = Frenier (2001), # = LePage et al. (2009) trations. These tests, described in Frenier et al. Table 7.2—Calcite dissolution capacity of various (2000) and LePage et al. (2009), were run for dissolvers (excess chelant and calcite). 24 hours at 150°F in a loosely closed reactor that contained an excess of calcite. The reaction equation for acetic acid with calcite is CaCO3 + 2(CH 3 − COOH) = Ca(CH 3 − COO− ) + CO 2 + H 2 O.�������������������������������������������� (7.10) Citric acid was one of the first organic acids to be used in solvents to remove iron oxides from process equipment. While inhibited citric acid was used to clean various power-generating units in

Cleaning of Pipelines and Facilities  339

the 1950s, formation of a gray precipitate was noted. Citric acid is an inexpensive, environmentally friendly chemical; however, its use in oilfield applications has been limited because of the low solubility of calcium citrate that could form if the fluid was spent on a high calcium-containing deposit. Citric acid is a chelating agent as well as an organic acid. Chelating agents include the amino carboxylic agents such as EDTA or DTPA (see Table 7.1 and Fig. 7.10). These chemicals can be used to dissolve calcite and iron oxide, usually at acidic pH ranges, and constitute the key ingredients in all other known sulfate-removing formulations, which are used at high pH and usually high temperatures. Chelating solvents work on the equilibrium concentrations of the metal ions in solutions with the solid. Even though the equilibrium concentration of the metal may be rather low, a powerful chelating agent such as EDTA or DTPA may be complex enough to enable the metal to shift the equilibrium to the right, thus affecting dissolution of the salt. In addition, chelating agents can be used to remove carbonate scales. Even though these materials are much more expensive than mineral and organic acids, they may be required at high temperatures or in the presence of sensitive metal alloys. Selection is especially important if corrosion or toxicity problems are a consideration. A chelating molecule can surround a metal ion in solution and greatly increase the solubility. A structure of a Ca(II)-HEDTA chelate ring is shown as an example of the chelation complex (Fig. 7.13). The formation of rings by the multidentate ligand (i.e., ligand having many reactive sites) gives added stability to the complex. The equilibrium for the reaction, M + nL = MLn.������������������������������������������������������������������������������������������������������������������������ (7.11) does not go as far to the right for monodentate ligands as it does when a chelating agent is present. This is known as the “chelate effect.” The equilibrium constants for these reactions, K = [ MLn ] / [ M ][ L ] ,������������������������������������������������������������������������������������������������������������ (7.12) n

defines the “strength” of the chelate complex. For example, the equilibrium constant for Cu2+ reacting with four molecules of ammonia (NH3) is 1013 (log K = 13), whereas for the complex of Cu2+ with the chelating agent EDTA, log K is 19. See Table 7.1, which is a compilation of stability constants of ligands and metals of interest for scale removal studies. Highly acidic chelating agent formulations, such as mixtures of HEDTA and a mineral acid (see Fig. 7.10) will dissolve calcite and iron oxides/carbonates at temperatures as low as 40oF (Hoy 1987). If calcium salts are not present, mixtures of HEDTA with other mineral acids (e.g., sulfuric acid) have also been used in some pipeline cleaning operations (Frenier and Hoy 1986). Fig. 7.8 is an example of the dissolution of an iron oxide from a water line at 100oF comparing HEDTA/pH 1.7 with a fluid containing Diammonium EDTA at pH 5. Note that the lower-pH fluid is much more effective at this low temperature. Frenier and Hoy (1986) reported that the lowpH fluids are effective for scale removal at temperatures as low as 40°F. An example of the reaction of acidic HEDTA with calcite is shown in Fig. 7.14. Sarathy et al. (2000) reported that a modified ammonium citrate solution was used Fig. 7.13—Chelate ring formation.

340  Chemical and Mechanical Methods for Pipeline Integrity

O– O

Ca++ O–

HO

+

–O

N N

Na+ O

OH

O O Calcium carbonate

OH

Monosodium N-hydroxyethylethylenediaminetriacetic acid O–

Na+

O HO

N N

O

+ CO2 + H2O

O– O

Ca++

O–

Monosodium calcium N-hydroxyethylethylenediaminetriacetic acid Fig. 7.14—Reaction of calcite with acidic HEDTA.

for the preoperational cleaning of a diglycolamine gas treatment system (Sections 2.3.2 and 7.6) to remove iron oxide, and possibly Cu, and to passivate the surfaces. 1. For iron oxide removal, they used 3 wt% citric acid pH adjusted to 3 to 3.5 with ammonia hydroxide. This was circulated and maintained at a minimum temperature of 75°C up to 95°C. 2. In the neutralization/passivation stage, they ensured that the minimum available free citric acid was >1% and raised the pH to 9.5 by adding sodium carbonate. Once the pH was a uniform 9.5 pH, 0.5 wt% sodium nitrite was added and circulated for passivation. See layout and circulation details in Section 7.6.2. Formulations of several chelating agents with HCl or H2SO4 to provide acidity have been used widely in the oil field. Frenier et al. (2001) described the use of HCl/HEDTA solutions for dissolution of calcite. Acidic aqueous fluids containing GLDA and various acids (pH > 3) have also been used commercially to dissolve calcite in oilfield applications (Boonstra et al. 2009; DeWolf et al. 2010). An example of chelating agents dissolving iron oxide from a pipe is shown in Fig. 7.8. Flushing, neutralization, and possibly passivation and disposal of the wastes are also steps in the process with organic acids and chelating agents. However, because the initial pH values are much higher than with mineral acids, less neutralization will be required. See Sarathy et al. (2000) in the above example. 7.3.3  Solvents for Alkaline Earth Sulfates. The three main sulfate minerals that could foul pipelines, wells, and surface equipment are gypsum, anhydrate, and barite. Unless an X-ray diffraction (XRD) test has been performed, the deposit will be described only as barium or a calcium sulfate.

Cleaning of Pipelines and Facilities  341

The state of hydration may dramatically affect the dissolution rate of calcium sulfate, but this will be considered to be a “morphology” issue unless the exact crystalline form is known. It is also possible for calcium sulfate and barium sulfate crystals to be mixed into a composite scale. In addition, calcite can mix with the sulfates (see Section 4.4.3). Some fields also have scales that contain strontium sulfate, and some sulfate deposits can contain radium or radon. These mildly radioactive deposits are naturally occurring radioactive materials, commonly designated “NORM.” These different solids may also contaminate pigs that have cleaned lines associated with oil/gas fluids and may require cleaning of the pigs before reuse or disposal. Sulfate deposits are difficult to dissolve because they are not soluble in acids. This happens because there is no way to effectively remove the sulfate anion from the system. In addition, oxides or carbonate scales become “opened up” physically during the dissolution process because the formation of the CO2 and H2O reaction products can produce microcracks. This process can improve penetration of the scale by the fresh solvent. Therefore, the effective rate of dissolution of a carbonate or oxide increases dramatically during the process as the available surface area increases. The dissolution of a sulfate scale takes place on the surface of the crystal until natural inhomogeneities cause the mass to fragment. This fragmentation is important to the overall dissolution rate but is not a predictable phenomenon. In addition, there are large differences in the solubility of calcium sulfate and barium sulfate (Table 4.2). As described below, the barium salt is especially difficult to remove. Because both sulfates are products of the mixing of production water streams, analyses of these water sources will give an indication of either barium or calcium sulfates. However, only a chemical examination of the deposit found in a line (Section 7.4) can distinguish between these two deposit types and thus enable the specific selection of a solvent. Before the introduction of chelating agents to dissolve sulfate deposits, the only method for removal of these scales (usually gypsum deposits in surface equipment) consisted of stages of alkaline agents such as sodium carbonate (soda ash), which were soaked into the deposit to “convert” the surface to calcium carbonate (references in Rybacki 1973). The “converted” deposit is then soluble in HCl. Multiple stages of soda ash followed by HCl were often needed to remove the deposit. This was an extremely slow process and required large volumes of chemicals. Chelating agents can remove these deposits in one treatment if there is enough solvent capacity. The reactions of a chelating agent, HEDTA, with calcite are described in Fig. 7.14. The reactions of chelating agents with other salts are as shown here: CaCO3 + Na 4 EDTA = CaEDTA 2− + 4Na + + CO32−. ���������������������������������������������������������������� (7.13) CaSO 4 + Na 3 HEDTA = CaHEDTA− + 3Na + + SO 42−.�������������������������������������������������������������� (7.14) BaSO 4 + K 4 EDTA = BaEDTA 2− + 4K + + SO 24− ���������������������������������������������������������������������� (7.15) Note that for a chelating solvent to be effective, the ratio of the solubility product to the chelate equilibrium constant (Ksp/Keq) must be greater than unity (see discussion in Section 7.3.2 as well as Eqs. 7.11 and 7.12). The author of this book notes that EDTA and DTPA could theoretically dissolve CaSO4 because the ratio is >>1, whereas none of the chelating agents has a ratio >1 for BaSO4. This makes dissolution of barite much more difficult (i.e., slow) compared with gypsum dissolution. These solvents are effective only at high pH (12) and high temperatures (>200°F), so they have limited use in the pipeline environment unless a section can be isolated and circulated or several days of soaking can be tolerated. See Figs. 7.15 and 7.16 for examples of rates of reaction. Note that several chelating agents at high pH (12) will readily dissolve gypsum; barite (Ba) is extremely difficult to dissolve and after 50 hours (Fig. 7.16) less is in solution than in the case ofgypsum (Ca) after only 5 hours. In addition only K4EDTA with an X additive was effective at the lower level. Note that many “additive” kinetic accelerators for dissolving barite are reviewed in Frenier and Ziauddin (2008, Chap. 5).

342  Chemical and Mechanical Methods for Pipeline Integrity

Fig. 7.15—Dissolution of gypsum.

Fig. 7.16—Dissolution of barite.

HO HO

P

O

Use of these chemicals will require temperatures above approximately 150°F or very long contact time (a day or more), so there may be limited uses in pipeline environments. A patent by Zaid and Wolf (2002) lists a solvent for alkaline earth sulfates that contains a combination of ingredients that include EDTA, a bicarbonate, ammonia, and a phosphonate chelating agent such nitrilotri(methylphosphonic acid); see Fig. 7.17. The authors stated that 25 mL of a solution of these chemicals was placed in a beaker and heated to approximately 160°F in a hot-water bath. .

O HO

N P

OH

O P

OH

OH Nitrilotri(methylphosphonic acid) Fig. 7.17—Component of barium sulfate solvent.

Cleaning of Pipelines and Facilities  343

Then, a piece of barium sulfate was added to the heated solution. The barium sulfate was completely dissolved within 15 minutes. 7.3.4  Corrosion Inhibitors for Inorganic Scale Removal Solvents. Essentially all the solvents used for dissolving inorganic deposits, as well as some of the solvents for organic deposits, are corrosive to some construction metals. Most pipeline equipment requiring cleaning is constructed of iron-based alloys such as carbon and alloy steels. These metals will be attacked by the aggressive solvents (mineral acids, organic acids, and low/neutral-pH chelating agents), unless the cleaners contain corrosion inhibitors. An uninhibited solvent would damage the equipment being cleaned, and the corrosion would consume the expensive solvent. This is true even for neutral-pH solvents. Thus, without the use of corrosion inhibitors, most chemical descaling operations should not be performed. Note that this application requires different molecules compared to the materials that inhibit sweet/ sour corrosion (Section 6.1). For more-complete information see Frenier and Ziauddin (2008). Corrosion Reaction. In most pipeline operations, the corrosion reaction that the inhibitor must minimize is an acid attack of some type. The electrochemical equation is 2H+ + Fe 0 = H 2 + Fe 2+ .������������������������������������������������������������������������������������������������������������ (7.16) This type of acid attack takes place in hydrochloric acid, organic acids, and chelating agents, including formulations that have pH values as high as ≈9, especially if they contain ammonium (NH4+) ions. The rate of corrosions can vary by several orders of magnitude depending on the acid strength and temperature. In general terms, the amount of damage potential increases with acidity and temperature, but most chemical scale removers are corrosive and require use of corrosion inhibitors. The review references cited below give much more information about corrosion in general. Other types of corrosion, such as pitting and SCC, may also occur, but they are usually initiated by acid attack. The technology of acid corrosion inhibition was reviewed by Frenier and Growcock (1989) and Frenier and Ziauddin (2014). Also refer to the excellent reviews in Corrosion Inhibitors, the European Federation of Corrosion’s Publication No.11 (EFC 1994). For more-general information on chemical cleaning science and technology and the requirements for corrosion inhibitors, the reader is referred to Frenier and Nesbitt (1995), the Electric Power Research Institute manual (Wackenhuth 1984), and the Powell (1984) chemical cleaning manual. Additional information on inhibitors is included in the compilations of Oakes (1972), Riggs (1973), Robinson (1979), Rozenfeld (1981), and Frenier (2001). Cleaning Solvents for Pipelines. A short review for the use of corrosion inhibitors in pipelinecleaning solvents is presented in the following paragraphs. An inhibitor formulation that is actually used in a pipeline treatment has a number of components, including • The “active ingredient” that provides most of the inhibitor protection • A surfactant to help disperse the inhibitor • Solvents to make a one-component mixture • Inhibitor “aids” to improve performance in special conditions (e.g., high temperature or highalloy steel) These components affect the use range (e.g., acid strength and temperature) as well as the overall impact of the inhibitor on the environment. Different types of corrosion inhibitors are required for these two broad types of fluids. HCl-Based Solvents. Solvent fluids that are HCl based usually contain an amine and/or an unsaturated oxygen inhibitor compound. The mechanisms by which many corrosion inhibitors function

344  Chemical and Mechanical Methods for Pipeline Integrity

Fig. 7.18—Formation of polymer film in HCl from acetylenic alcohol and quaternary.

were elucidated over the past 50 years. The reader is referred to Damaskin et al. (1971) for an excellent introduction to this subject. The diagram in Fig. 7.18 shows a generic mechanism in which two different inhibitors are protecting steel in the solution containing HCl. Shown are an acetylenic alcohol and a pyridinium quaternary. See Fig. 7.19 for typical chemical structures for inhibiting HCl-containing cleaning solvents. Inhibitors for Sulfuric Acid, Organic Acids, and Chelating Agent Solvents. The most successful inhibitor formulations for sulfuric and organic acid inhibitors and chelating fluids usually contain amines and a reduced sulfur compound, or combinations of a nitrogen compound (amine, quaternaries, or polyfunctional compound) and a sulfur compound. Typical sulfur chemicals include mercapto acetic acid, thiourea, or compounds that produce a small amount of a soluble sulfide. The only major exceptions are the sulfonium compounds. Jofa (1965) investigated the mechanism of mixtures of quaternary ammonium or alkyl ammonium compounds used with sulfur compounds, such as thiourea. He concluded that the sulfur compounds form HS-ions in solution. These adsorb onto the iron surface, thus attracting the cationic amine inhibitor. As compared with HCl solvents, the metal surfaces do not have a negative (Cl–) covered surface to attract the amine or quaternary. R − SH = HS− . ������������������������������������������������������������������������������������������������������������������������ (7.17) −

Fe + HS− = Fe ( HS) . �������������������������������������������������������������������������������������������������������������� (7.18) In+ + Fe(HS)− = Fe(InSH).������������������������������������������������������������������������������������������������������ (7.19) The author of this book believes that the nitrogen/sulfur inhibitors function in a similar manner in the chemical cleaning solvents (Fig. 7.20 shows a diagram of this type of mechanism). The sulfur-containing inhibitors must be used with some of the special alloys found in nuclear steam generators. Currently, there is no complete explanation for the specificity of these metals for the nitrogen/sulfur combination. Virtually all commercial scale-cleaning inhibitor formulations contain these active ingredients (see Kennedy 1987; Frenier 1997, 2003a, 2003b). Table 7.3 shows a short summary of corrosion inhibition requirements and the use of various inhibitors for different metals, solvents, and temperatures. Inhibitor “aids” are noted for completeness, but they will probably not be required in pipeline-related cleaning situations. See Frenier and Ziauddin (2008) for more details and references. Corrosion inhibitors for use in descaling fluids must be rigorously tested using high-pressure equipment and the actual metals that may be found in the treatment path. Unfortunately, there is no

Cleaning of Pipelines and Facilities  345

N Compounds Dicyandimide

H2N H 2N

H C

N

C

N

Piperizine

N

N H

Pyrrole Rosin amine Mannich product (R1 = abietyl)

N H

Indole

R1

O N

Nitrobenzene

N

CH2

CH2

CH2

R2

C

R1

NO2

H S/N Compounds O Benzyl thiocyanate

CH2

Sulfoximine

SCN

R1 S

R2

NH

Product of a Thiol with a nitrile R1

S

CH2

CH

C

O Compounds

N N O

O Crotonaldehyde

CH CH

CH3

C

O

C

CH

C

H

H

O

Furfural

H

CH

Cinnamaldehyde

Phenyvinyl ketone

H

C

C

CH2

O

β-Hydroxypropiophenone

C

CH2

CH2

OH

O 3-Phenyl-2-propyn-1-ol

C

C CH2

H OH p-Anisaldehyde H3C

O

C

O

OH Hexynol

CH3CH2CH2

CH OH

C

CH

Ethyloctynol CH3(CH2)4

CH

C

C

CH

CH2CH3

Fig. 7.19—HCl corrosion inhibitor structures.

standard method for evaluation of inhibitors and fluids. Each vendor uses slightly different procedures. Hill et al. (2003) described some procedures required for development of inhibitors. 7.3.5  Solvents for Organic Solids. Waxes and asphaltenes constitute the main categories of organic deposits that may foul pipelines. These deposits may be mixed with residual oil and in most cases with inorganic solids (see the discussion in Section 7.3.6). Gas pipelines operating at high pressures and low temperatures can also be fouled with gas hydrates. The formation mechanisms are described in Section 4.3 and in more detail in Frenier et al. (2010).

346  Chemical and Mechanical Methods for Pipeline Integrity

Fig. 7.20—Mechanism for organic acids and chelating solvents. Temp. Max., °F

Metallurgy

HCl/HF

200

All

Ammonium quaternary and unsaturated oxygen compound

None

HCl/HF

350

Carbon steel

Ammonium quaternary and unsaturated oxygen compound

Formic acid

HCl/HF

350

Cr steels

Ammonium quaternary and unsaturated oxygen compound

Formic acid Antimony chloride

Organic acids

200

All

Ammonium compound and reduced sulfur compound

None

Organic acids

400

All

Ammonium compound and reduced sulfur compound

None or KI of CuI

Chelating agents

400

All

Ammonium compound and reduced sulfur compound

None

Fluid

Generic Inhibitor Type

UInhibitor Aid

Table 7.3—Inhibitor selection for pipeline/facility scale removal fluids.

This section considers the many chemical solvents that are in current use or have been proposed in the patent literature for removing organic fouling solids. The dissolution of organic solids shares many of the aspects of inorganic scale dissolution. Similarities are that they require • Identification of the deposit composition, amount, and location • Testing of the deposit to determine solubility in various fluids, assuming adequate samples can be obtained • Determination of the best application method • Circulation/placement of an adequate solvent volume to achieve the treatment goal However, a major difference between solvents for inorganic and organic scales is the degree of dissolution that is possible and needed. Using the appropriate amount and type of solvent, most inorganic scales can be removed completely—see the various discussions in books on inorganic scale (e.g., Cowan and Weintritt 2004; Frenier and Ziauddin 2008). Because of the heterogeneous nature of field organic foulants, complete solvency and thus removal may not be possible using a single solvent or even with several stages. In addition, usually a fraction of the deposit is inorganic in character.

Cleaning of Pipelines and Facilities  347

(

Therefore, use of emulsions, dispersions, and cosolvents is commonly required for removing deposits containing organic materials. Also, the thermodynamic requirement of total dissolution cannot usually be proved. Thus, the stability-constant/solubility-product nomenclature used in the inorganic scale literature is usually not applied to describe the dissolution process (Frenier and Ziauddin 2008). Nonetheless, quasithermodynamic functions exist for some systems and will be described later in this section. Ultimately, testing and development of custom formulations at the field level are a significant requirement for success. This section discusses the chemistry and types of commercial solvents and other fluids in current use to remove pipeline area organic deposits. Techniques developed to clean refinery and petrochemical equipment may provide a guide for developing solvents for the removal of production-related organic deposits because there are chemical similarities. Note that deposits in some topside facilities will be similar to refinery deposits. Listed in Table 7.4 are generalized methods for dissolving various organic scales found in these plants (Frenier 2001). These deposits are found on refinery front-end equipment as well as in operational deposits. Solids found in the transport portion of the hydrocarbon supply chain will usually be composed of the first four categories of organic deposits described in Table 7.4. The waxes frequently found in flowlines and production tubing consist of higher-carbon-number alkanes compared with the refinery hydrocarbons, but they are soluble or dispersible using the same types A cationic surfactant of solvents as used in refineries. Thus, deterNH3+ Cl– gents or organic solvents should be able to 1-dodecyl-, ammonium chloride remove most pipeline-related organic deposits. Onshore processing plants or any off_ N + O Na An anionic surfactant shore units where heat is applied to affect O S O a separation can experience fouling by the higher-molecular-weight deposits described as the F5–F8 categories. These types of foulants will require high-temperature solvents 1-dodecyl sulfonate, Na salt or mechanical methods for removal to be A nonionic surfactant accomplished. Detergents. There is a vast variety of surfactants (i.e., detergents) used in removing organic solids and dispersing inorganic/mixed solids. The key categories are cationic, anionic, )9 O and nonionic. Simple molecules that are repreOH p-n-nonylphenol, 9 EO sentative of these types are shown in Fig. 7.21. Fig. 7.21—Surfactant types. Code

Deposit Type

General Cleaning Solvents

F1

Light hydrocarbons, C1-C5

Degassing and light oil removal by detergents

F2

Gasoline, diesel and fuel oils, C4-C5

Degassing and light oil removal by detergents

F3

Motor oils, greases, bunker C, crude oil

Remove with alkaline detergents and light solvents

F4

Tar and asphaltic deposits

Aromatic solvents followed by alkaline detergents

F5

Black, hard solids with a polymer matrix

Alkaline detergents, organic solvents with accelerators

F6

Black, hard solids with a coke matrix

Mechanical methods

F7

Linear polyolefins

High-temperature oils or solvents followed by detergents

F8

Crosslinked polymers

Mechanical methods

Table 7.4—Generic organic deposits and solvent types.

348  Chemical and Mechanical Methods for Pipeline Integrity

There are also many amphoteric surfactants that can be anionic or cationic depending on the pH of the aqueous phase (see Frenier and Ziauddin 2014) . A very wide range on nonionic surfactants are available and are often a first choice for at least a part of a cleaning formulation because they are compatible with other components. Materials with various hydrophilic/lipophilic balance values are shown in Fig. 7.22. This system allows for an estimation of the balance between the water-soluble parts—mostly ethylene oxide (EO)—and the hydrocarbon parts. This enables the formulator to try to match the systems to be cleaned. See Griffin (1954), Kelland (2009), and (Frenier and Ziauddin 2014) for more details of hydrophilic/lipophilic balance calculations. For many applications, especially in cleaning pipe sections that are not in a controlled environment (e.g., a company facilities plant), it may be desirable to use environmentally benign surfactants and solvents such as those that are natural-product-based. An example from Sasol (2012) is shown in Fig. 7.23. Fig. 7.23a shows the base structure of a branched-chain sulfonate (anionic detergent), and Fig. 7.23b shows a number of natural product alcohols that can be used to form an array of materials with varying solvency ranges for different types of deposits. TDW (2012) also claimed the use of environmentally friendly surfactants for cleaning pipelines. Many oilfield and pipeline chemicals, including detergent cleaning mixtures, corrosion inhibitors, and defoamers, contain a mixture of different chemicals. These incorporate the active ingredients, solvents (water, alcohols), and possibly cosolvents (see a later section). Werff (2006) reported on a proprietary solvent that is claimed to contain concentrated liquid surfactants, polymers, and dispersants. The formulation must be soluble/dispersible in the flowing stream and must be stable on-site and in unheated/cooled storage facilities. Many inhibitors and surfactant formulations have used methanol as a universal solvent (and possibly as an antifoaming agent). However, this alcohol is toxic and highly flammable and can be substituted for by materials such as glycol ethers. See the discussions on less toxic inhibitor formulations by Frenier (2003b) and Sitz et al. (2012). At critical micelle concentration, the molecules of the surfactant in water associate to form a micelle. Fig. 7.24 show a diagram of a spherical micelle with the hydrophilic “head” at the water interface and the hydrophobic “tails” associated in the center of the micelle. In the presence of a hydrocarbon or fat molecule, these materials can be emulsified and thus removed from a surface.

Fig. 7.22—Common nonionic surfactants.

Cleaning of Pipelines and Facilities  349

(a)

Varying the hydrophobe length and degree of branching as well as the PO and EO number

(b) Hydrophobe Families

Fig. 7.23—Surfactants from natural products (Sasol 2012).

The cleaning action of detergents involves changing the IST between the metal and the water (see Figs. 1.16 and 1.17) such that the surface becomes water-wet, not oil-wet. This action allows the organic compounds to be displaced. The remaining surfactant then forms an emulsion with the oil, preventing redeposition. Flow in the pipe keeps the organic soils suspended. Frequently, organic solvents or alkaline agents (sodium carbonate or trisodium phosphate) are added to help convert any fatty acids in the oil to a natural surfactant and improve detergency. General Organic Deposit Solvents. If detergent solutions cannot effectively remove an organic deposit in a pipeline, then organically based solvents may be required. Table 7.4 notes that common aromatic solvents such as toluene and xylene (see Figs. 7.25a and 7.25b) and aromatic naphtha, a petroleumderived solvent, have long been used to dissolve hydrocarbon-based solids. However, these materials are toxic to humans and to other organisms and are not allowed or are restricted in some markets. As a result, formulations containing terpene solvents such as d-limonene and turpentinebased solvents have been proposed (see Figs. 7.25e, 7.25g, and 7.25h). An example includes a patent by Matta (1985) to remove hydrocarbon deposits. These compositions include a terpene solvent, propyleneglycol ethers, and surfactants in a water base. They are claimed as hardsurface cleaners that will dissolve oils, asphaltenes, and heavy oils. Mehta and Krajieck (1995) and Krajieck et al. (1995) described processes that use terpenes and surfactants for cleaning and decontaminating refinery vessels and for removing benzene to allow maintenance to take place. An alternative technology that contains enzymes and N-oxide type nonionic surAqueous solution factants was claimed by Mestetsky (1995) Hydrophilic head as a solvent that easily releases the oil for enhanced separation and disposal of the waste water. The cited documents demonstrate that more ecologically acceptable and safer methods (compared to Hydrophilic tail using aromatic solvents and chlorinated hydrocarbons) are being developed to clean refinery equipment. The structures of some lower toxicity chemicals (along with toluene and xylene) are shown in Fig. 7.24—Surfactant micelle. Figs. 7.25c through 7.25h.

350  Chemical and Mechanical Methods for Pipeline Integrity

(a)

(b)

Toluene

Xylene O H 3C

(c)

Dicyclopentadiene

(e)

N

(d)

n-Methyl-2-pyrrolidinone

(f)

d-Limonene 7-Methyl-3-methylene-1,6-octadiene Myrcene

OH (g)

Pinene

(h)

2-(4-Methyl-1-cyclohex-3-enyl)propan-2-ol Terpineol Fig. 7.25—Standard (a, b) and environmentally improved (c–h) solvent molecules.

Additional solvents and formulations are reviewed here. Penney (1986) patented a method for removing and preventing deposition of organic deposits by contacting the deposit with a penetrating solvent and a cationic perfluoro compound. These materials are said to prevent or reduce the wetting of the surface by the hydrocarbons. Many solubilizing formulations have been proposed for removing waxes and asphaltenes. McClaflin and Yang (1987) stated that a water-soluble ethoxylated alkyl phenol and a short-chain alcohol could be used to clean the wellbore. Another common formulation is an oil-soluble surfactant and a light or aromatic hydrocarbon. Kruka (1987) also proposed such a formulation. Jennings et al. (2011) claimed that an exothermic reaction can occur between selected solvents and an acid, and the heat evolved may help to remove organic solid deposits. The acids may include organic acid compounds such as sulfonic acids, sulfuric acid, and nitric acid. The solvents may include terpene and terpene-derivative-containing solvents, including, but not necessarily limited to, limonene, pinene, dipentene, turpentines, and compounds having at least one double bond such as methyl furan, dienes, styrene, and vinyl acetate. Although the reactions between some of these chemicals (some drawn in Fig. 7.25) were shown to generate substantial heat, it is not clear what the reactions do to the solvency of the organic molecules or whether the reactions can occur at pipeline temperatures. Emulsions and microemulsions of various organic solvents can be produced and are in use to reduce the amount of organic solvent needed for deposit removal. Microemulsions are a special category of emulsions wherein the droplet size is reduced so that the emulsion appears to be thermodynamically stable. To form microemulsions, a specific ratio of surfactant/oil/water, as well as (usually) a cosolvent such as a butanol, must be present. Fig. 7.26 shows that only a very small area of the phase diagram will yield the microemulsion, but these fluids can be powerful solvents with reduced environmental impacts in some situations.

Cleaning of Pipelines and Facilities  351

Model oil-in-water microdroplet Continuous water phase

Polar hydrophilic heads

100% buOH:CTAB (1:2) Microemulsion region-larger area

Variety of phases

Hydrophobic chains Cosurfactant

Surfactant

(a) Model of oil-in-water microdroplet with effect of the cosurfactant

100% 20% DeS060

100% Hexadecane

(b) Phase diagram of butanol/CTAB, water solution of a SI (DeS060), and a hydrocarbon (hexadecane)

Fig. 7.26—Microemulsion-phase diagram showing (a) the general formation of a microdroplet in water and (b) a specific example of forming an emulsion with a scale inhibitor.

This figure shows the phases for emulsifying a scale inhibitor (e.g., DeS060). Collins and Vervoort (2001) described a microemulsion comprising (1) an oil phase; (2) an aqueous phase, which is an aqueous solution of a water-soluble oil/gas production chemical or an aqueous dispersion of a water-dispersible oil/gas field production chemical; and (3) at least one surfactant. The aqueous phase is distributed in the oil phase in the form of droplets having a diameter in the range 1 to 1000 nm or in the form of microdomains having at least one dimension of length, breadth, or thickness in the range 1 to 1000 nm. This microemulsion is claimed to be effective for removing various organic deposits in well environments. Emulsions using acids and chelating agents can also be used to remove mixed organic/inorganic deposits. These are described in more detail in Section 7.3.6. Solvents and techniques for specific types of pipeline deposits are reviewed in the next sections. Note that these are listed separately, frequently from the patent literature; however, often the products are effective for a wide variety of organic pipeline solids. Solvents for Wax. The solvents for wax and asphaltenes described were developed for downhole use, but they could be models for surface pipeline use. Bernadiner (1993) made an extensive evaluation of several surfactants and mixtures for removing wax and preventing redeposition. The surfactant classes studied included alkyl sulfonates, alkylaryl sulfonates, and polyethyleneglycol ether di-t-butylphenol materials. The author of this detailed study used rotating discs coated with wax to examine the effect of chemical type as well as flow. By plotting the square root of the angular velocity (w) of the disc vs. the dissolution rate, the investigator was able to indicate that there was a significant effect of the detergent flow rate on the dissolution process in accordance with the Levich (1962) relationships. Using mixtures of these surfactants at 0.4 to 1.25%, the author claimed to have cleaned and prevented wax deposition in wells located in Kazakhstan. The papers of Bernadiner (1993) and Fan and Llave (1996) are unusual in that they use surface (mineral) dissolution kinetic approaches to look at organic dissolution mechanisms. They imply that the rates are diffusion limited at least under the test conditions. Zhang et al. (2003) described a dewaxing fluid comprising at least one each of these: • Paraffin-solubilizing organic solvent selected from the group consisting of aromatic hydrocarbons, terpenes, and isoparaffinic hydrocarbons, said paraffin-solubilizing organic solvent comprising from approximately 25 to 75 vol% of said composition

352  Chemical and Mechanical Methods for Pipeline Integrity

• Water-soluble polar organic solvent, said water-soluble polar organic solvent comprising from approximately 25 to 75 vol% of said composition • Water-soluble surfactant, said water-soluble surfactant comprising from approximately 0.5 to 20 wt% to volume of said composition The organic solvents in this patent include limonene, isoparaffinic hydrocarbon, and an alkylbenzene or dialkylbenzene. Examples of the polar organic include methanol, ethanol, isopropanol, butanol, acetone, ethylene glycol, and propylene glycol. A wide variety of water soluble surfactants include cationic, nonionic, or anionic materials. Note that the formulator would have to develop the final formulation incorporating all the ingredients chosen to provide a stable emulsion. This patent is a good example of typically used oilfield formulations for wax removal. Dyer (2007) summarized commonly used acids, detergents, and emulsions for removing wax. Anionic surfactants used in detergents comprise fatty acid soaps; alpha olefin sulfonate; sulfonates; amine ethoxylates; amine salts or linear alkyl benzene sulfonic acid; aromatic sulfonates comprising cumene, xylene, and toluene sulfonate; earth metal salts of olefin sulfonate and alcohol sulfates and sulfonates; and blends of such anionic surfactants. Nonionic surfactants suitable for use comprise ethoxylated nonionic surfactants selected from the group consisting of condensation products of ethylene oxide with aliphatic alcohols having from 8 to 22 carbon atoms in either straight- or branched-chain configuration as well as ethoxylated nonionic surfactants selected from the group consisting of condensation products of ethylene oxide with nonyl phenol, phenol, butyl phenol, di-nonyl phenol, octyl phenol, or other phenols, as well as blends of such nonionic surfactants. The acid constitutes from 0.1 to approximately 15 wt% based on the total weight of the emulsion, and preferably from approximately 0.1 to 5 wt%. Suitable acids include hydrochloric, phosphoric, sulfuric, hydrofluoric, nitric, citric, oxalic, maleic, acetic, fumaric, malic, glutaric, or glutamic acids, as well as mixtures of such acids. The preferred acid is hydrochloric acid. The hydrocarbon solvent constitutes from approximately 0.1 to 20.0 wt% based on the total weight of the emulsion, and preferably from approximately 0.1 to 10 wt%. The hydrocarbon solvents suitable for use in accordance with the invention discussed include kerosene, gasoline, diesel, jet fuel, xylene, and mixtures thereof. The preferred solvent is kerosene. Dyer (2008) described a method for cleaning oil wells to increase the flow of oil with a unique aqueous cleaning emulsion comprising water, hydrocarbon solvent, and detergent. This one-step method provides for the simultaneously cleaning/removal of asphaltene and/or paraffin and scale. This method can be used alone or with the assistance of a wash tool that is a combination pressure/surge wash tool having a nipple assembly. A bypass port is coupled to the nipple assembly, and a diverter cup is coupled to the bypass port. Several pressure wash cups are positioned on the tool. A pressure wash port is located between the pressure wash cups, and a pump shoe assembly is coupled to a bottom pressure wash cup. The first claim is that this formulation that contains water, a surfactant, and a hydrocarbon to form a stable emulsion can be used to remove wax and asphaltenes as well as “scale.” Mention is made in the specifications of also adding a mineral acid to the emulsion. Two authors claimed that biodegradable fluids could be used as solvents to remove wax (and possibly to inhibit wax formation). Tukenov (2014) described a solvent and a process claimed to be grounded in “nanochemistry.” This author specifically claimed that the chemical works when mixed and activated with crude oil by a proprietary nanochemical mechanism and attacks wax differently than do other chemistries. Also claimed is that flushing wax with activated crude oil removes the wax in layers and creates a waxophobic (i.e., wax-repelling) condition for an extended period. The chemistry is also environmentally friendly, being nontoxic and nonvolatile, and the activator is safe to handle. Spills can be washed down, and no special handling requirements are needed. Neither the chemistry nor the mechanisms are revealed.

Cleaning of Pipelines and Facilities  353

A patent by Andrecola (2014) also claimed that nanoemulsions of specific chemistries could efficiently remove wax deposits; however, the author did not claim residual inhibition of wax formation. Specific embodiments of this system are these: • The fluid system comprises a blend of micellar solution of fatty acids, a vegetable oil solvent, and one or more surfactants. The preferred fatty acids are fatty acids from coconut oil or tall oil, or they are a blend of C8–C16 fatty acids such asisoleic or linoleic acids. In a preferred embodiment, tall oil is the preferred fatty acid. • In particular, the fluid system used in the invention is in the form of a nanoemulsion, defined as a multiphase system consisting of water, fatty acid and vegetable oil blends, emulsifier(s), and alcohol. This is a transparent and thermodynamically stable liquid solution. • In a preferred embodiment, the nanoemulsion of droplet dispersions of oil-in-water (O/W) has an average particulate size range in the order of approximately 5 to 50 nm in drop radius. Nanoemulsions of the current disclosure are prepared by providing an external energy input to the oil/water/surfactant system using high shear stress or inertial disruption to overcome the effect of interfacial tension to fragment large microscale droplets into the nanoscale. Unlike microemulsions, where the average drop size grows continuously with time so that phase separation ultimately occurs, the nanoemulsions of the current disclosure are thermodynamically and kinetically stable. The drops of microemulsion are generally large (>0.1 microns) and often exhibit a milky or cloudy, rather than a translucent, appearance as seen in nanoemulsions. Note that Andrecola (2014) also contains short reviews of methods for removing wax such as hot oiling. Asphaltene Removal Methods. Because asphaltenes contain more aromatic and unsaturated groups than paraffins and frequently contain a broad mixture of different chemical species, the solvents tend to be more aromatic and frequently more complex. In the experience of the author of this book, xylene is commonly used to remove asphaltenes, but other methods are described in the patent literature. These are presented next. Thomas et al. (1995) studied the deposition of asphaltenes in oil-producing formations. The authors claimed that the selection of chemical control agents in the past has been limited to bulk dissolution studies on samples retrieved from production systems. Until recently, the accepted way to treat these problems has been through the use of xylene, toluene, or other aromatic solvents. This method requires the use of large amounts of these solvents, as well as a high frequency of treatment. Preliminary dispersant and solvency tests were conducted by Thomas et al. (1995) using an asphaltene dispersant test in hexane. Chemicals that provided promising results in dissolving and dispersing asphaltenes in the nonsolvent medium of hexane were selected as candidates for field application. This consisted of making asphaltene pellets and noting how fast the brown color dispersed into hexanes in the presence of a dispersant, or for additional testing in a core flow deposition removal test. The core flow test apparatus provided a method to introduce asphaltene fouling into a core to enable study of its removal by the use of chemical agents. Using core samples and asphaltenes from the production resource under consideration allows the selection of the best removal chemical. The test measured the return permeability after cleaning with various solvents. Several proprietary solvents showed improvements compared with a xylene control and were field tested in several oil fields. An important conclusion is the need to continue an “inhibitor” treatment after the formation has been cleaned. Del Bianco and Stroppa (1995) described a composition that includes saturated species, as well as alkylbenzenes and polyaromatics that are useful for dissolving asphaltenic residues present in oil wells. The authors developed a graphical method for making the blends. The compositions are selected from those listed on a ternary chart, and they contained various polyaromatics, alkylbenzenes, and other hydrocarbons.

354  Chemical and Mechanical Methods for Pipeline Integrity

In some cases, solvent treatment of the oil is considered to be beneficial, according to Mansoori (2001, 2008), because it dilutes the crude oil and reduces the tendency of the heavy organics to precipitate. Solvent treatments may not be very successful largely because the solvents that can be used are limited to aromatic solvents. Xylene is generally the most common solvent selected for well stimulations, workovers, and heavy organics inhibition and cleaning. In some cases, xylene injection through the nonproducing string may actually help to minimize the heavy organic deposition problem. The author noted that in oil fields with frequent need for aromatic wash, it may be necessary to design an aromatic solvent with stronger wash power and better economy for the particular deposit in mind (Garcia-Hernandez 1989). Laboratory tests may be necessary to blend the most appropriate aromatic solvent and/or dispersant for a given oil field from the points of view of effectiveness, economy, and environmental friendliness. Del Bianco and Stroppa (1997) proposed a process for the dissolution of asphaltene sediments in oil wells that consists in introducing a solubilizing composition for the above asphaltenes into the oil wells. This process is characterized by a solubilizing composition that comprises the following elements: • Basically a hydrocarbon fraction consisting of at least 70 wt%, preferably at least 80 wt%, of aromatic and alkyl aromatic hydrocarbons, the alkyl group being from C1 to C4 • A fraction consisting mainly of quinoline and isoquinoline as such or alkyl substituted, preferably as such, the alkyl group being from C1 to C4 Note that the weight ratio between the two fractions ranges from 97.5:2.5 to 75:25, preferably from 97:3 to 90/10. For example the fraction called “wash oil” deriving from the distillation of coal tar, which in itself contains a certain content of quinoline (usually from 5 to 10 wt%), has proved to be particularly effective. The wash oil product has an initial distillation point (ASTM D2887) of between 198 and 210°C and a final distillation point of between 294 and 310°C, with 50% that distills at a temperature lower than 230 to 250°C (1CAS No. 309-985-4; EINEX No. 101896-27-9). The wash oil can be used as such, or another quinoline or isoquinoline can be added as required. Thorssen and Loree (1998) proposed a wax and asphaltene solvation fluid for use in oil/gas wells derived as a residual fluid from a feedstock that includes a greater mass percentage of trimethylbenzene than decane and is preferably sour. Mass percentage of both aromatics and asphaltenes in the residual fluid is in the 30 to 70% range, and a complex mixture of both are described (see also Loree 2000). Several authors have proposed using various oilfield fluids for removing asphaltenes. Thus, Jamaluddin and Nazarko (1996) developed a method that comprises the injection of deasphalted oil into the near-wellbore formation followed by a soaking period and a production period. Their method has the great advantage of readily dissolving precipitated asphaltene in a well environment without other costly treatments such as the use of solvents such as xylene or toluene. Furthermore, their method does not require the incorporation of any additives in the deasphalted oil. In an aspect of the invention, deasphalted oil is injected at a temperature ranging from 20 to 100°C to increase the solubilizing power of asphaltene particles. The deasphalted oil can be any oil from which the natural asphaltene or asphalt components have been removed. In yet another aspect of this method, the deasphalted oil is obtained directly from the well to be treated, and the deasphalting of the crude oil is realized by precipitating asphaltene molecules with n-pentane. Deasphalted oil can also be obtained from oil-refining or heavy-oil upgrading processes. Cimino et al. (1995) noted that monocyclic aromatic solvents such a toluene or xylene gave less than 50% dissolution of many asphaltenes. Thus, they recommended inexpensive distilled refinery blends. A preferred mixture for testing could contain 1-methylnaphthalene, n-hexadecane, and toluene. Lightford et al. (2008) describes the laboratory development and field application of a water/ aromatic solvent emulsion system that has been successfully used to clean/dissolve asphaltene and leave the carbonate fractured formation in a water-wet state to delay the production decline.

Cleaning of Pipelines and Facilities  355

Other advantages when using this type of emulsion are cost reduction and improved effectiveness in removing asphaltene deposits when compared to alternative solvents that have been used. This is of particular significance to these wells in which large volumes of a washing phase have to be pumped downhole. Hazards have also been reduced by using relatively high-flashpoint aromatics. Continuous mixing of the emulsion when pumping reduces waste improves the logistics involved in pumping the large volumes needed to treat long, openhole sections and/or to treat the fractures deeper in the near-wellbore region. Although this paper does not specifically describe the composition of the formulation, it does describe dissolution tests using a number of aromatic solvents, xylene + ethylbenzene, a mixture of terpenes, terpenoids with a terpene-limenoid fraction, and a heavy aromatic naphtha. In addition, solvency was improved by adding a cosolvent such as a glycol ethers (Fig. 7.27). All these solvents dissolved essentially 100% of the asphaltene deposit. Buzelin and Lima (2008) described the use of a chemical product composed of diesel, isopropane, benzene, and naphthalene to clear subsea pipes clogged with mixtures of wax and asphaltenes. The pipes need to be treated with these chemicals for a minimum of 12 hours, and better results could be obtained by increasing the treatment period. A special filtration procedure is required to separate the liquids and solids from the material that is discarded from the pipes, and the liquids are recycled through the pipe again until the pipes are fully treated. In this process, which was accomplished by pumping from the surface, the cleaning mixture was introduced and then flushed through the pipeline. Clathrate Gas Hydrate Solvents and Applications. Because alcohols are well-known inhibitors used for preventing hydrate formation (Section 6.3.3 and Frenier et al. 2010), alcohols, especial methanol, can be used to dissolve hydrates in lines. Precautions must be taken when removing hydrates because the sudden evolution of a gas may cause an unwanted pressure spike. Therefore, the use of any solvent, including water, requires careful application methods. Lee et al. (2009) developed a method for placing methanol (an inhibitor and a solvent) for gas hydrates at the site of a plug in a subsea line. The example is for a subsea line, but the method could be used in some surface lines with modifications. The technique involves pumping methanol into the well to flush any fluids trapped at the wellhead down into the well. HO

OH Ethyleneglycol

N

N,N-dimethyl-9-decenamide

OH HO Propyleneglycol HO O Ethyleneglycol monobutyl ether

HO

O O Butyl-3-hydroxybutyrate

O O

O O

O

O

OH Propyleneglycolmethyl ether

O

HO

O

O

O

O

Dipropyleneglycolphenyl ether

Ethyl 3-(2,4-dimethyl-1,3-dioxolan-2-yl) propanoate Ethyl levulinate glycerol ketal

–O OH Dipropylene glycol monomethyl ether

HO

OH

1,3-Propanediol Methyl-9-dodecenoate

Fig. 7.27—Alcohol-type solvents and cosolvents.

356  Chemical and Mechanical Methods for Pipeline Integrity

The subsea safety valve should be opened to allow any fluids to fall to the well bottom, and then the valve is closed. The gas column trapped in the upper section of the tubing can then be bled back to the platform through an available umbilical line. It is recommended to dose methanol into the wellhead while bleeding the pressure back to the platform to minimize the risk of forming hydrates in the umbilical line. The gas pressure in the well should be lowered to a pressure that will not cause valve damage from the high differential pressure across the valve while opening. This author advises that, after the flowline and well pressure have equalized, the valves be closed and the pressure in the flowline on the wellhead side of the plug (after choke pressure) be monitored for any changes. An increase in pressure can be attributed to the dissociation of hydrates from the plug and can help to reveal that the composition of the plug does contain hydrates. 7.3.6  Solvents for Mixed Deposits. Most if not all fouling deposits in pipelines and facilities are mixtures of several or even many different solids (see the discussions in Section 4.4). Frequently, enough solids can be removed by the pigging methods described in Sections 6.9 and 7.1 to clear the line for a needed maintenance process. However, if there are several organic solids as well as inorganic deposits such as black powder in the pipeline, chemical methods may be needed to remove enough of the materials to allow corrosion assessments or other maintenance to be performed. Wylde and Slayer (2009) made particularly useful observations about the selection of solvents to remove deposits from pipelines. They noted that to remove the very mixed organic portion of a pipeline deposit, a number of chemical factors should be considered. These include • • • • •

Wetting of the solids by the solvents Solubilization (actual dissolution) Emulsification Dispersion Detergency

The point made by Wylde and Slayer (2009) is that a number of different chemicals—five to seven, in fact—may be required, and these must be custom blended on the basis of the composition of the solids to be removed. More details regarding solvent testing and selection are found in Section 7.4, and practical examples are presented in Section 7.5. Removal of mixed organic/inorganic deposits and acid sludges poses unique problems because even more different chemistries must be addressed simultaneously. There are three broad strategies for removing mixed deposits: applying solvents in stages, using emulsion with aqueous fluids, and adding surfactants and cosolvents to aqueous fluids. These approaches are described next. Stages of Solvents. A common strategy is to pump/spray an organic solvent to remove an organic solid and then follow it with an acid/chelant/detergent stage to dissolve or disperse the inorganic solids. Frequently, several cycles may be need for complete removal of the deposit. Montgomery et al. (1996) have described the use of various solvent stages for deposit removals. Stages of solvents can be used with pig trains, as described in Section 7.5. Emulsions or Dispersions of Organic Fluids With Aqueous Fluids. Oil outer-phase and oil innerphase emulsions can be generated depending on the need. Many combinations of organic solvents and acids and/or chelating agents can be formed as an emulsion (Coffey et al. 1974; Lyons and Plisga 2005). Many microemulsions (these appear to be one phase) have also been produced (Collins and Vervoort 2001). The benefit of a single stage is that from the standpoint of placement, it is easier to control than multiple stages. In all these fluids, a major issue is the use of a properly tested corrosion inhibitor (Frenier and Ziauddin 2008; also Section 7.3.4), which is necessary to protect the base metal of the tubing and flowlines from attack by the aggressive fluids needed to remove the mineral scale deposits.

Cleaning of Pipelines and Facilities  357

This is a difficult problem because the organic portion of the fluid may dissolve the inhibitor film if the formulation is not properly tested and formulated. Aqueous Fluids Containing Surfactants and Cosolvents. Mixtures of aqueous fluids can be blended with relatively large amounts of methanol, ethanol, or isopropyl alcohol as well as ether alcohols such as ethyleneglycolmonobutyl ether or dipropyleneglycol methyl ether (see Fig. 7.27). These types of fluids can dissolve some mixtures of organic/inorganic foulants but are not as aggressive as emulsions if there are very large concentrations of a heavy organic solid such as an asphaltene. Frenier and Brady (2008) proposed a mixture of several types of glycols with chelating agents and other solvents and surfactants that were effective for removing inorganic/organic deposits from the near-wellbore region of the formation. The fluids were a single phase and could be used with various corrosion inhibitors at high temperatures and in high-salt-content fluids. Organic acids such as acetic acid (HAc) or stronger acids (HCl or H2SO4) have also been used with surfactants and mutual solvents to remove mixed foulants. McCoy (2015) describes the recent development of alcohol-based solvents—as well as cosolvents (see the right side of Fig. 7.27)—that are now in use in many solid-surface cleaners and detergents. The materials are claimed to be derived from natural products and are less toxic than ethylene glycol monobutyl ether (left side of Fig. 7.27). When used with surfactants, these materials may be used to clean pipeline surfaces. The choice of an application method and the choice of the individual solvent components depend on the total makeup of the fouling solids. The methods described can be used to choose and optimize the final formulation. Various specific examples of solvents for acid sludge and mixed deposits are abstracted. McLaughlin and Richardson (1978) proposed an acidizing fluid for asphaltic formations that can be cleaned with an aqueous acid without causing a deposition of permeability-impairing iron/ asphaltene compounds. The improvement comprises adding to the first portion of the aqueous hydrochloric acid solution or homogeneous dispersion of at least enough salicylic acid to chelate with and prevent the formation of iron/asphaltene solids. It does so, substantially, as all the ferric ions within the acid that enters the earth formation become dissolved. Many other iron-chelating agents could likewise be applied for ferric control (Frenier et al. 2001), but whether the salicylic acid also provides additional solvency for the organic materials remains unclear. Mixtures of various organic solvents (Section 7.3.5 and Fig. 7.25), surfactants (Fig. 7.21), and cosolvents (Fig. 7.27) can be formulated to wet, emulsify, and disperse mixtures of organic/inorganic solids from pipelines. Specific formulations are claimed from several vendors, but details are trade secrets. An example from Miska et al. (2015) is given in Section 7.5.1. 7.4  Testing of Deposits To Develop Solvents Cowan and Weintritt (2004, Chap. 9) provided a very useful section on scale removal that describes important tasks in scale sampling, analysis, and testing. The following steps are recommended by Cowan and Weintritt (2004) and the author of this book as a general plan for removing deposits from pipelines and facilities. 1. Obtain a representative sample. Many deposits are heterogeneous, so the sampling processes must take this into account, and samples from different locations may be required for an accurate representation of the problem. In the pipeline sections of this industry, samples of deposits are frequently obtained after a pigging operation and may be called “pig trash.” Sample several parts of the material removed; if several batches are produced from one line, sample all of them. The sample should be preserved in an air-tight container if possible. Note that the sludge may be toxic or flammable and therefore care is recommended. Note also that all the possible deposit classes discussed in Chapter 4 may be present. The author of this book appreciates that this may be the most difficult step in the process.

358  Chemical and Mechanical Methods for Pipeline Integrity

2. Estimate the amount of the foulants. Do this because it is impossible to know the quantity of chemicals required for removal if the amount of deposit in the line is unknown. Thus, the volume/mass of scale to be removed must be estimated. Some of the ILI devices described Section 5.4.2 may be used to assess scale volumes as determined by changes in pipeline diameters. A thesis by Ebenezer (2006) reviewed several additional known methods for estimating the deposit load. Refer to the references in this document for details. • Pressure pulse method (activating a valve in a multiphase pipeline to measure deposits) • Pressure drop (related to change in diameter of line; Eq. 1.12) • Pigging and ILI (Section 5.4) • Thermal methods (thermal flux); not recommended by Ebenezer (2006) for multiphase lines • Spool piece removal where an actual small segment can be observed, the scale removed, and the mass calculated • Tracers (chemicals not described) 3. Identify the deposit (after collection) using chemical/physical means, including Fouriertransform infrared spectroscopy (FT-IR), GC/LC mass spectroscopy, X-ray fluorescence (XRF), and XRD. Specific examples are in a later subsection. 4. Perform chemical testing; see Fig. 9.5 in Cowan and Weintritt (2004), which shows a scheme of chemical analysis from ASTM (1969). Some chemical tests are described here in Sections 7.4.1 and 7.4.2. 5. Proceed with physical testing of the deposits and evaluating solvents. Some additional details of the sample identification of the deposit and solvent evaluation process are described in this section. Several different methods have been proposed for testing solvents and deposits in Sections 7.4.1 and 7.4.2. The author of this book notes that the testing protocol must determine the • Correct chemical formulation for removing the deposit, including the concentration of the active ingredients required (either oil based or water based). • Volume of solvent required on the basis of the mass of the solids requiring removal. The test protocol should also suggest the application methods appropriate for the sites, amount of deposit, and physical environments (see Section 7.5). A subject frequently neglected is determining the utilities available to support temporary equipment. The next Sections (7.4.1 and 7.4.3) provide detailed methods that have been used in various fields. 7.4.1 Work Flow for Evaluation and Treatment of Inorganic, Organic, or Mixed Deposit. Fig. 7.28 shows a flow diagram of a process (arrows) to determine the composition of scale and organic deposits on the basis of chemical tests (Frenier and Ziauddin 2008). The author of this book notes that this test protocol has been used for many years to analyze deposits from a range of industries, including pipeline power production, pipelines, refineries, facilities, and petrochemical plants. 1. A weighed sample of the deposit is ashed in a high-temperature furnace to drive off any hydrocarbon-containing chemicals. The weight loss is the loss on ignition (LOI). 2. The remaining solids are extracted with high concentration of HCl (frequently 28%) at 200°F to determine the maximum solubility in HCl. An analysis of the dissolved metals in the HCl supernatant gives the total amount of metals in the sample. Any remaining solids are extracted with HCl/HF to give the silica content (SiO2). Analysis of the metals in solution from the acid dissolutions can help determine the scale composition. Inorganic analyses can also be performed using a variety of chemical methods, including atomic absorption spectrophotometry, inductively coupled plasma optical emission spectrophotometry, or wet chemical tests.

Cleaning of Pipelines and Facilities  359

Scale Loading Procedure

Perform loss on ignition test (LOI) 10% LOI extraction tests needed

SiO2 by HCI/HF

Metals and Anions

Gas evolution CO2, H2S HCN, SO2

MPS + LOI + SiO2 = 100% Fig. 7.28—Deposit analysis process.

3. If the LOI is 10% or less, acidification of the solids will give a qualitative assessment of carbonates or sulfides on the basis of the formation of gas. In addition, X-ray analysis of the scale can give a quantitative assessment of the elemental composition (XRF) or a semiquantitative assessment of crystalline compounds present (XRD). This type of analysis may also identify specific crystalline minerals in the scale. 4. If the LOI is more than 10%, then extraction with organic solvents will be required to determine a process needed to dissolve the deposit. If there is a very high mass of organic deposit, separate hot extractions using alternate volumes of toluene and hexane can give a qualitative estimate of the chemical composition of the organic deposit. Also, subjecting the solution to FT-IR or GC mass spectroscopy may provide qualitative information that may facilitate a solvent choice (see the next subsection). If the maximum solubility in HCl is more than 20%, multiple stages of an organic/inorganic solvent may be required or an acid emulsion or dispersion with an organic phase will be needed for complete removal. 7.4.2  Chemical Identification of a Deposit. Chemical means for determining the deposit composition may be useful for picking the appropriate solvent. This section provides information on the application of an analytical technique, FT-IR, to aid in identifying organic deposits. This information may be used in the development of chemical dissolution formulations and methods for the subsequent removal of such a deposit. Curtis and Weaver (1998) described the uses of this highly promising technique. The remainder of this section was abstracted from their CORROSION 98 paper (Curtis and Weaver 1998) and is used with permission. This method was originally developed to identify deposits in refinery equipment and other surface facilities, but it is as useful for determining the composition of pipeline/facility deposits. Identification of Classes of Organic Compounds by Fourier-Transform Infrared Spectroscopy. The first step in this fingerprinting process is the separation of component materials from the composite sample. Once separated, the functional groups or class of the component materials may be identified by FT-IR. This separation may be influenced by either physical or chemical means. Separation by Physical Methods. One of the more common examples of physically separating components in a refinery or petrochemical soil is based on the material’s volatility relative to temperature. By increasing the temperature using an infrared lamp, heat gun, or other heat source, lower-boiling materials will volatilize from the sample. At temperatures near or below the freezepoint of water, often higher-boiling materials will “gel” or “solidify.”

360  Chemical and Mechanical Methods for Pipeline Integrity

Once solidified, these materials can be removed by filtration or other suitable means. By analyzing the materials obtained from these physical separations, a better understanding of the nature of the organic materials within the sample is achieved. By electronically subtracting the FT-IR spectra of each component from those of the composite, spectral insights for the entire sample are brought about. Separation by Chemical Methods. A common method of chemically separating refinery or petrochemical soil components uses their different solubilities. Some components will preferentially partition into a polar solvent, whereas others will be soluble in nonpolar solvents. Solvents are classified as polar and nonpolar by their molecular structure. Water is the most common polar solvent. Additional polar solvents are linear alcohols such as methanol or ethanol. Traditional benchtop methods to evaluate organic soils can be time consuming and occasionally ineffective. Study of the materials obtained from these physical/chemical separations using FT-IR not only reduced analysis time but also provided insights that might otherwise be inaccessible. A sample of a foulant appeared to be black, sooty metallic shavings. Following solvent extractions of this material with xylene and Freon™, the solvents were evaporated, and the residues were analyzed by FT-IR. Because chlorinated solvents attack the epoxy that secures the zinc selenide crystal to the horizontal attenuated total reflectance boat, Freon was air-dried from the residue. The residue was resolvated in acetone for mounting purposes. A spectral library search, restricted to the fingerprint region, strongly suggested that the residue from each extraction was a defoamer. The class of defoamer was narrowed to a polyalkyl methacrylate copolymer in hydrocarbon oil. There were subtle differences between the residues from an acetone extraction and those of Freon and xylene. The appearance of peaks at 1852, 1774, and 1104 cm–1 strongly indicate a high-molecular-weight carboxylic anhydride as a component in the acetoneextracted material. A spectral library search confirmed this by identifying the compound class as di- and tetra-substituted carboxylic anhydrides. These materials commonly occur in emulsifiers and lubricants. Follow-up provided insight to the producer of the information afforded by FT-IR analysis of the deposit. The FT-IR analysis established that there were organics present in the form of surfactants and defoamer additives. These additives were believed to be the origin of the problem. Summary. On the basis of the scale analysis from the previous sections, a short list of candidate solvents is developed. The previous step is necessary because it is known that not all solvents will dissolve a certain scale type. A set of possible solvents should be selected on the basis of the scale type, temperature of the facility or pipeline, and metallurgy and on the basis of logistical and economic considerations. Then, a laboratory test to determine solvent capacity is required (Section 7.4.2). For example, using “pig trash,” a sample from a pigging operation in a West Texas pipeline, Wylde and Slayer (2009) developed an identification and solvent evaluation that included an energy-dispersive X-ray spectroscopy analysis of the solids to identify the inorganic solids after the sample was ashed (i.e., through LOI). Because the sample showed more than 50% organics, the authors performed visual and gravimetric solvency tests using a wide range of formulations to develop an optimal cleaning fluid that could be used at the pipeline temperature. Lowtemperature and some high-temperature procedures for testing specific solvents are outlined in Section 7.4.3. 7.4.3  Evaluation of Solvents for Removing Inorganic/Organic Deposits. Low-temperature as well as high-temperature (and -pressure) evaluation methods are in use and described in this section. These can be used for evaluation of solvents for pipelines (usually low temperatures) and facilities (frequently higher temperatures). Several very specific methods also are noted. Low-Temperature Evaluation of Solvents. Two specific tests are described, the first by Frenier et al. (2010), and the second by Casey (2006).

Cleaning of Pipelines and Facilities  361

Frenier et al. (2010) provided the following procedure that can be used to measure the effectiveness of different organic/inorganic solvents. A.  Equipment Needed. • Four-ounce bottles • Balance • Hot-water bath • Wrist shaker B. Procedure. 1. Place approximately 10 g of sample into a clean, preweighed, 4-oz. glass bottle. Prepare one bottle for each solution to be tested. 2. Place the bottles containing the 10-g samples in hot-to-boiling water. Completely melt any organics if possible. 3. Carefully remove the bottles from the hot-water bath and roll the bottles horizontally as the organics begin to solidify. 4. Repeat the heating, cooling, and rolling procedure until a uniform, solid coating of organics is deposited on the inside of each test bottle. 5. Allow the paraffin bottles to cool to room temperature. Weigh each bottle. 6. Place 100 mL of the appropriate organic solvent, inorganic solvent,or emulsion in each bottle. Cap and seal each bottle. 7. Place the bottles in a wrist-shaker and shake for the duration of the test (up to 24 hours). 8. Note the visual results. 9. Carefully decant the solvent from the bottles. Dispose of the decanted solvent in an approved manner. Gently rinse the bottles with water and acetone and allow them to dry. 10. Reweigh the bottles. 11. Calculate the percent solid removed. The percent deposit removed equals ([P + B ]2 − B1) − ([P + B ]3 − B1) × 100 , �������������������������������������������������������������� (7.20) ([P + B ]2 − B1) where [P + B]2 = original weight of organics and bottle from Step 5, B1 = original weight of bottle, and [P + B]3 = final weight of paraffin and bottle from Step 10. 12. Record the relative efficiency of each solvent system tested. 13. The supernatant as well as any sludge can be further analyzed by metal analyses (AA, ICP) or by FT-IR liquid chromatography with a mass spectrophotometer or any other appropriate method that is available. Note that the Frenier et al. et al. (2010) test (above) can be modified for testing water-based solvent on inorganic minerals. The modification involves deleting Steps 2 through 5. Casey (2006) described the following low-temperature test procedure for organic solids (i.e., grease). It uses a gravimetric method to test solvents for removal of organic deposits such as greases on surfaces. The device uses the bob of a rheometer as the test item, and this can be rotated to improve the rate of solids removal. 1. Preweigh a viscometer rotating bob to the nearest 0.01 g and record as “Weight Before.” (This is approximately 150 g.) 2. Lightly smear a uniform layer of axle grease on the exterior surface of the lower two-thirds of the rotor.

362  Chemical and Mechanical Methods for Pipeline Integrity

3. Weigh and record as “Before + Solids.” After this step, care is taken not to disturb the layer of grease (Fig. 7.29). 4. Carefully attach the grease-smeared sleeve on the viscometer apparatus. 5. Prepare 500 mL of the test solvent in a glass beaker. The normal solvent concentration is 10 vol%. 6. Place the beaker of test solution on the viscometer fluid shelf. Adjust the fluid depth on the sleeve until the fluid meniscus and the etched ring on the rotor coincide. 7. Start the viscometer rotor rotating at 300 rev/min. Initialize/reset the stopwatch to zero. 8. Continue the rotating sleeve smeared with grease in the test solution for 30 minutes. 9. Stop the viscometer rotor rotation. Lower the liquid shelf and properly discard the used solvent. 10. Carefully remove the rotor from the viscometer to allow for air-drying. 11. Allow the rotor to air-dry overnight. Weigh the rotor and record as “Weight After.” 12. Formula to determine the percent weight of grease removed by the solvent: wt% Removed =

(Weight After) − (Weight Before) × 100 ���������������������������������� (7.21) ([Before + Grease] − Weight Before)

Additional methods for low-temperature dissolution of deposits are cited but not abstracted. See Spurrell and Bibbs (2000) for a method for evaluating detergent formulations for removing organic deposits from pipelines and facilities. Campbell (2000b) also lists simple methods for evaluating pipeline deposit cleaners.

Beaker Organic deposit Rotator bob

Fig. 7.29—Photograph of rotator with organic deposit (Casey 2006).

Cleaning of Pipelines and Facilities  363

The next subsections describe tests in which the temperatures of the units being cleaned are >200°F. This is an unusual condition for pipelines but could exist in some facility units that may be heated to higher temperatures to speed the cleaning operations. High-Temperature and -Pressure Procedures. A generic test procedure was taken from Frenier and Ziauddin (2008), but it has been modified for testing predominantly mixed scales. Tests should be performed in cells pressurized to approximately 600 psig. This equipment can be used for hightemperature and -pressure simulations. This test is most appropriate for testing inorganic or possibly mixed deposits. The procedure to be used for laboratory test is as follows: 1. Prepare the scale dissolver (solvent). 2. Pour 100 mL of the scale dissolver into a test cell. 3. Weigh approximately 12 g of a scale sample and place it in a scale basket (record exact mass). 4. Lower the scale basket into the test cell. 5. Pressurize the test cell to 500 psig. 6. Place the test cell in a temperature bath for the desired time (2 to 24 hours). 7. If the cell can be sampled under pressure, a dissolution/time curve can be produced. 8. Remove the test cell and release the pressure. 9. Filter the contents onto a preweighed filter paper. 10. Rinse the solid particles with water, then with acetone to remove organics; place the filter containing solids in a drying oven. 11. Weigh the dry solids and filter paper. 12. Test the dry solids for HCl-soluble material, and develop a solvent for that portion. The amount of scale removed by the organic dissolver is Steps 3 through 10. Required parameters include these: • Piping interval length to be treated, h (ft) • Internal radius of tubing, Tubing (in.) • Internal radius of scale, rscale (in.) • Scale density, rs (lbm/ft3) • Dissolution capacity, G (lbm/gal) The calculations for determining the needed volume of solvents are presented in Fig. 7.30.

1) Volume of scale

2  2 R    Rtubing  Vs (ft 3 ) = 3.1416 * h *   −  scale    12    12   

2) Volume of chemical to remove all scale

V * ρs Vc (gal ) = s G

3) Initial pipe volume

r  Vi (ft 3 ) = h * 3.1416 *  scale   12 

4) Estimated number of soaks

=

2

Vc * ρs  Rtubing r   + scale  12 12  h * G * 176.715 *    2    

Fig. 7.30—Calculations for solvent volumes.

2

364  Chemical and Mechanical Methods for Pipeline Integrity

Additional Test Methods for Various Deposits. Several test methods for specific types deposits are listed in this section to illustrate alternative methodologies. Del Bianco and Stroppa (1995) used the following method to test and custom-blend asphaltene solvents. To correctly perform the measurements, two steps were necessary: first to prepare the solution appropriately and then to determine the solubility curves. The following operations are necessary to complete these steps. Preparation. The solution containing a known content of asphaltenic material is dissolved in order to obtain calibration straight lines as necessary to correlate absorbance and concentration at the three wave lengths selected in order to perform the measurements (400, 600, and 800 nm). Such solutions were prepared by starting from a mother solution obtained by filtering a mixture of approximately 100 mg of deposit in 100 mL of solvent. Inasmuch as the asphaltenic material submitted to the measurements is a compositional continuum, and therefore the spectrum at ultraviolet/visible wave lengths can be a function of the amount and of the quality of the dissolved material, the evaluation of the data relevant to the concentration of the dissolved product during the determination of the solubility curves was obtained by calculating the average values from the absorption values measured at said three wave lengths. Note that the wave-length range within which the measurements were carried out is the widest possible one considering the instrument limits and what the solvent could potentially absorb under 400 nm. In most cases taken into consideration, the calibration straight lines show a very good linearity within the examined concentration range. By means of linear regression calculations, it is possible to compute the extinction coefficients relevant to each individual wave length, which will subsequently be used to compute the concentrations once solubility curves are determined. Determination of the Solubility Curves. The solubility curves for the evaluation of the solvent capacities are determined by measuring the amount of asphaltenic material dissolved by mixtures at different values of the deposit; this is the solvent ratio. By means of experiments, a set of mixtures was prepared that contained a known amount of deposit and increasing solvent volumes. Such mixtures were sonicated during 20 minutes and then were kept overnight with mechanical stirring. The resulting suspension was filtered under pressure by using syringes equipped with Teflon™ filters of 0.5 microns, and on the filtrate the absorbance measurements were carried out, from which the concentrations of dissolved organic material were obtained. In these examples from Del Bianco and Stroppa (1995), the solubility kinetics reported were determined by using ultraviolet/visible absorption (400, 600, and 800 nm) to measure, at room temperature, the concentration of asphaltenic material dissolved from specimen pellets of deposit soaked in the solvent being tested; this was performed as a function of soaking time. The pellets are prepared by pressing an exactly weighed amount of 100 mg of sample with a Perkin-Elmer press, with a pressure of 10 000 kg/cm2, to obtain small discs of 13-mm diameter having a thickness of 0.7 mm. The aforementioned pellets are then charged to the interior of a sample carrier composed of two wire networks supported by a tripod and soaked in 1 L of solvent to be studied. The deposit/ solvent ratio is such that after an infinite time, the maximal solubility level allowed for each solvent should be reached (grams of deposit/liter). During the test, the solution is kept weakly stirred by means of a magnetic anchor to secure the homogeneousness of the solution, with the fragmentation of the pellet being simultaneously avoided. Ivanova and Shitz (2008) developed a method to evaluate the efficiency of the action of solvents with additives for removing mixed organic deposits in a Russian oil/gas field. The deposit sample was heated to the softening temperature, thoroughly mixed, and shaped in the form of a cylinder 12 × 20 mm. It was then cooled and placed into the previously weighed basket from the brass (steel) gauze, with the size of a cell being 1.5 × 1.5 mm. The dimensions of the basket are 70 × 15 × 15 mm. The basket with the samples was weighed and placed in a glass hermetic cell into which 100 mm of the solvent under study is poured. The experiment temperature is 10°C.

Cleaning of Pipelines and Facilities  365

In 4 hours, the basket with the residual nondestructed component of the deposit is taken out and dried to a constant weight. The destructed but undissolved component of the deposit, which had fallen from the basket in the cell, is filtrated, dried to a constant weight, and weighed. 7.5  Application of Chemical Cleaning in Pipelines The chemicals to be used for cleaning fouled pipelines or facilities (Section 7.6) can be applied as singlephase liquid solvent or emulsions or as gels or foams. This section described these practical methods. 7.5.1 Liquid-Phase Solvents. Additional specific reasons and cautions for performing a cleaning include • Preoperational cleaning to remove construction debris and to lower the dewpoint before introduction of products (also see hydrostatic pressure testing in Section 5.3). Methanol would be a preferred final solvent fill to help produce a low dewpoint. • Operational cleaning to remove debris before inspection, testing (including hydrostatic testing), and maintenance or to clear blockages and reductions in effective line diameters. • Pipeline conversion from one product to a different product (i.e., gasoline to diesel fuel). There are also practical issues to take into account. These include • Any chemical cleaning of pipelines, including conventional use of “pigs,” introduces special problems. • Most pipelines are at “ambient” temperature, but they are frequently in the temperature range of 40 to 80°F (4 to 27°C). Thus, conventional chemical cleaning technologies must function at the lower end of their operational range. • Pipelines are often buried underground or from several feet to thousands of feet under the ocean; they also run under rivers and cities or are in isolated geographic areas. • Usually there are very limited access points, and circulation of “solvents” is frequently impossible. • Mechanical pigs that have scrapers and brushes are used to remove debris from pipelines. These were described in Sections 7.1 and 6.9.3. For extremely dirty sections, or for other reasons, so-called pig trains of chemical solvents (Section 7.3) or of gelly pigs are used to clean pipeline sections (see Fig. 7.31); these trains use stages of degreasing chemicals followed by acid cleaners and a neutralizing medium. These stages may be many feet long and are separated by mechanical or gel pigs. The selection of the solvent and tests were described in Section 7.4. Note that more than just the efficiency of the solvent may control the chemistry used or whether chemicals are used at all. A highly respected participant in chemical cleaning activities for many years, Greg Casey reported1 that the location of the fouled line may play a significant role in solvent/method choice. He noted that “within the fence” of a facility or plant, aggressive chemicals such as acids, chelating agents, or organic solvents may be applicable. In these locations, spills can be contained and waste disposal (see Section 8.4) of a relatively small volume can be managed. However, for long reaches of lines with few pig launchers and traps, less-aggressive solvents such as detergent solutions or very benign solvents (e.g., terpenes, alcohols) are less problematical for spills, leaks, and disposal. The aggressive use of brush pigs (Section 7.1 and Fig. 5.9) may be the practical method for deposit removal. It may be feasible to pump a solvent train back and forth to increase solvent contact with the deposit. Personal communication, Greg Casey to Wayne Frenier, 2006.

1

366  Chemical and Mechanical Methods for Pipeline Integrity

Chemical Pig Trains

Separator gel

Drive fluid

Inhibitor gel

Cup pig

Drive fluid

Debris Separator gel gel

Inhibitor gel

Line fill

Brush pigs

Neutralizing solution

Cup pig

Acid solution

Degreasing solution

Line fill

Cup pig

Fig. 7.31—Chemical and mechanical pig trains.

Methods for Following the Progress of a Pipeline Chemical Cleaning Treatment. How to follow and assess the progress of a pipeline cleaning operation depends on the deposit to be removed, the solvent in use, and the access points for retrieving samples. If short segments of line are being cleaned to remove inorganic deposits (e.g., iron oxides, iron sulfides, or calcium salts) and an aqueous solvent is being used, then the metals in the solvent can be determined as a function of time. On-site portable equipment and mobile laboratories (see Frenier 2001, 2015) can be used for the metals analyses. The endpoint is the leveling of the increase in metal ions over approximately 1 hour. If an acid such as HCl is the solvent, the change in the acid content—a simple acid/base titration (Hach 2005)—will show how much free solvent is present. If stages of solvent or dispersing chemicals are applied to remove solids without dissolving them, then several on-site tests can be performed. Brush pigs (see Fig. 5.1, 5.7, and 5.8) can be run between solvent batches to determine if debris is continually being removed. The endpoint is when an agreed-on cleanliness value is achieved. Miska et al. (2015) described a pipeline-cleaning process whereby solvent/dispersants mixed with a hydrocarbon were staged with pigs over approximately 1 week to clean a line. The job was followed by sampling the liquids/solids that had been flushed from the trap to a fracturing tank. The authors also recorded the volumes of solvents injected vs. the solvent retrieved. Initially, the solvents were consumed by the solids and only subsequent batches, laden with solids, showed positive amounts of free solvent. The volume of solids found after each batch was measured by drying and weighing samples. Brush pigs were likewise examined after each cleaning segment. 7.5.2  Examples of Treatment of Pipelines To Remove Mixed Deposits. Cleaning Chemicals Used With Solid Pigs or Other Equipment. Bell et al. (2008) described the cleaning of a 17.3-mile subsea pipeline installed in 1980 from Platform Elly to Long Beach Harbor, California, USA. The line was fouled with paraffin, asphaltenes, and inorganic deposits. After several attempts, the operators used aggressive pigs, staged with solvents (not described), as well as brush pigs and removed a 30-ft deposit described as a wax “candle.” They concluded that chemicals were needed because mechanical methods were insufficient.

Cleaning of Pipelines and Facilities  367

Abney and Browne (2006) showed examples of the use of several different pig trains for the removal of wax, black powder (see Section 4.4.4), and emulsion-covered scale. The conclusions from this report are that different pigs as well as a variety of solvents and detergents may be required for different types of deposits, especially if they are mixed deposits. See Fig. 4.25 for a photo of the solids removed from this line. Wakeham and Fleming (2011) described the chemical and mechanical cleaning of a largediameter (30-in.) sweet natural gas pipeline before running ILI tools. The cleaning included two liquid slugs of cleaning agents. First was a 190 000-L slug of hydrocarbon-based gel and distillate; it was then followed by a 37 000-L slug of breaker to return residual gel to a liquid state for easier removal from the pipeline. Allan (2011) discussed development of a cleaning chemical for hydrocarbon transport pipelines. Details of the formulation research and laboratory testing are summarized, along with case histories of applications in the western US and the US Gulf Coast. As noted in Section 4.3, hydrocarbon transportation can result in precipitation of paraffins, asphaltenes, and naphthenates, which adsorb onto the walls of the pipeline, thereby becoming associated with iron sulfides forming from corrosion. Pigging operations alone do not always remove all these deposits because of compaction and a strong adherence to the pipeline walls. It therefore becomes necessary to add surfactant-based chemicals to assist in the breakup and removal of these deposits. Poole et al. (2008) presented a case history of inspecting, cleaning, and re-establishing communication in a deepwater pipeline that had plugged because of paraffin deposition over an extended period. Early mitigation attempts did not remove the solid mass causing the blockage but, rather, compounded the problem. A chemical capable of dispersing the wax mass was applied to the plugged pipeline through a snubbing unit. Access to the pipeline and the wax plug was gained by disconnecting the line at the subsea site, lifting the pipeline to a work rig, installing the snubbing unit, and applying the chemical solution in seawater at a prescribed dilution ratio. One critical step in dispersing and removing the solid wax mass was selecting the dispersant package. In addition, after the correct chemistry was identified (see Section 7.4), its application required a complex operation. A snubbing unit inside the pipeline jetted the chemical dispersant and drilled through the wax mass. The dispersed fluid was then returned to the working rig where the paraffins were disposed of. The operation was claimed to be successful, and the pipeline was placed back into production. However, it was complicated by the ocean floor topography: The deepwater environment adds a degree of difficulty that is not experienced in shelf waters or on land. The authors of the foregoing study stated that the exercise taught several lessons, the most significant being that a slow and consistent effort is ideal. Once the mass is dispersed, it is best to continuously remove this fluid. Stop/start operations allow the solubilized mass to resolidify as the solvent drains away from the mass under static conditions. Wylde (2011) described the use of an organic deposit solvent (see Section 7.3.5 for chemical options) with a mechanical pig to dislodge a stuck pig from a pipeline. The author claimed that a total of 168 gal (4 bblm) of specialty formulation was pumped neat, followed by 1,680 gal (40 bblm) of clean water. The product was pushed with a pig toward the lodged pig. The lodged pig was moved to the south, and then the flow was reversed and both pigs arrived at the northern junction and were removed from the pipeline. The water from the line contained large amounts of iron and hydrocarbons. When the two pigs arrived at the trap, a large amount of solids was recovered from the pipeline. Wylde (2011) and Poole et al. (2008) claimed that the mechanism of cleaning of mixed deposits where inorganic solids are not dissolved includes • Wetting: the action of a surfactant to reduce surface tension of a medium. This reduction is achieved by molecular attraction toward a dissimilar surface. For pipeline cleaning, wetting agents help to remove hydrocarbon deposits from oil-wet scale, thereby allowing access to inorganic materials. See the discussion of surface tension in Section 1.6.4 that applies to the other parts of a cleaning mechanism.

368  Chemical and Mechanical Methods for Pipeline Integrity

• Detergency: the ability of a surfactant to remove particles from a surface. This also applies to oils that become emulsified. • Emulsification: the formation of a stable emulsion of two or more immiscible liquids, a process enabled by surfactants. This is similar to micellular solubilization, just with larger solubilized particles. • Solubilizers: surfactants that can affect otherwise insoluble materials. When a surfactant concentration is high enough, micelle structures can form to incorporate the insoluble materials and bring them into an apparent solution. Cosolvents such as glycol ethers may be added to improve the ability of the emulsion to remove oils. • Dispersion: surfactants that retain insoluble particles in suspension by preventing aggregation of particles with one another. Ideally, particles are small and this will lead to a more stable dispersion. Note that similar to emulsification, the dispersion property of surfactants prevents redeposition of solid particles by maintaining them in suspension. For more information on emulsions and dispersions, see Kokal (2006). Wylde (2011) as well as the author of this book note that pipeline chemical solvents may include a number of different components to achieve a particular chemical cleaning goal. Several different surfactants may be mixed with cosolvents and either water or a hydrocarbon to affect the line cleaning. Deposit and solvent testing (Section 7.4) in addition to physical application and on-site testing will also be required for a successful cleaning operation. Additional Example Formulations of Chemicals for Pipeline Cleaning. Various publications and patents are abstracted in this section. Riad et al. (2012) cleaned a pipeline (with a 68,000-gal capacity) in Egypt to remove gypsum and organic deposits (see Section 4.4.3 and Fig. 4.22). The treating method used a stage of a 1:1 mixture of diesel oil and xylene to soak and to remove the organic components of the deposits. A second stage of an unidentified gypsum solvent—see Section 7.3.3 and Frenier and Ziauddin (2008)—was used to dissolve/soften the deposit. This fluid was flushed with water, and the solids, as shown in Fig. 4.22, were removed from the line. The authors reported that 500 drums of solids were removed and that the line was returned to satisfactory operations. Thompson (1974) patented an emulsion that is claimed to be successful for removing calcium sulfate deposits that are coated and/or commingled with an organic deposit. The emulsion comprises a continuous aqueous phase of at least one of sodium citrate, a water soluble salt of ethylenediaminetetra acetic acid, or potassium glycolate, having an organic solvent dispersed in it and with a surfactant present as an emulsifier, an acid phosphate ester. Examples of the emulsifier include at least one phosphate ester with an ethoxylated straight-chain alcohol (8 to 10 carbon atoms) containing 4 moles of ethylene oxide, or a phosphate ester of ethoxylated tridecyl alcohol containing 6 moles of ethyleneoxide. Sherik et al. (2008) claimed that there are several different types of chemical cleaning agents used in the removal of black powder from gas pipelines. These include gel formulations that suspend the black powder (see Section 7.5.3). The authors noted that gels show excellent capability to carry large amounts of solids; but in situations where cleaning has to be performed online, dealing with large gel batches becomes problematic. Also, removal of gel residues from the pipeline must be accomplished. Various surfactant cleaning agents were claimed by these authors to remove black powder. These chemicals (not described) can be dissolved in diesel or organic solvents (dissolution in water should be avoided to ensure that the pipeline is not exposed to water). Surfactants will have the ability to penetrate contaminants and lower the surface tension properties of the pipeline, leading to removal of large amounts of black powder. Unless the surfactants are implemented in acidic fluids, the cleaning action will be to form a loose suspension of the high-surface-area particles and sweep them along in the fluid flow.

Cleaning of Pipelines and Facilities  369

Ford and Hollenbeak (1987) described a dicyclopentadiene and a mixture of naturally occurring terpenes from pine trees. This mixture is dispersed into the treating acids by emulsifying them with an aliphatic ethoxy alcohol and used to dissolve sludge, organic deposits, and inorganic scale. Crowe (1993) developed an acid system for combating sludge with iron control. Antisludge agents containing nonionic and anionic surfactants are also used for sludge and emulsion control during acidizing. This acid uses a reducing agent (erythorbic acid) and ammonium mercapto acetate. Wong et al. (1997) studied a number of crude oils in the Permian Basin of Texas/New Mexico, in the US; these oils were known to cause formation of acid sludge. The authors compared nine different antisludging formulations. They found that ferric iron control was necessary and recommended a “catalytic reducing agent.” Using FT-IR analyses, they found that the acid sludge was stabilized by asphaltenes containing oxygen-containing functional groups. They also stressed the need for eliminating air contact with the crude oils before the test. Curtis and Weaver (1998) demonstrated the power of preidentification of an organic deposit by FT-IR before trying to pick a solvent, whereas Paulis and Sharma (1997) tested a number of nonylphenol formaldehyde resins (from Witco) as demulsifiers for crude oil in contact with brine or HCl/Fe3+ ions. These materials were claimed to be highly effective, but no data in the presence of iron in HCl were provided. Lysandrou and Dulaney (1987) detailed a process and composition for acidizing wells that contain paraffin or asphaltenes, comprising methanol and ethanol, a proponal, or tertiary-butanol, or any combination thereof and a mixture of C6 through C10 alcohols. These fluids are mixed and dispersed into the acids (HCl, HF) to dissolve the acid-soluble deposits as well as the organic foulants. The authors claimed that these mixtures are miscible in 25% HCl. Wylde and Slayer (2010) described details of the development of a cleaning chemical for heavyoil/gas pipelines. Information is offered regarding the reason for development, the research involved in formulation of the new product, and the laboratory testing. The authors contended that the key properties required of these chemistries were wettability alteration; solubilization efficacy of organic materials; and emulsification of phases, dispersion, detergency, and defoaming. These authors used a static dissolution test (see Section 7.4 for details of other test methods) using balls of pig trash, which contained organic as well as inorganic deposits from a pipeline. The method was used to rate a number of different organic solvents as well as mixtures with water-based solvents (see Fig. 7.32). One case history (Wylde and Slayer 2010) details how a pipeline operator unsuccessfully tried to clean a 12-in., 9-mile section of pipeline with a pig. The pig was launched and became stuck along the length of the pipeline. Application of the newly developed product was able to free the stuck pig and removed significant debris. The conclusion was that for this line, the use of chemical solvents along with pigs was more effective than pigs alone. This reference has an excellent list of references on specific pipeline solvents.

Static test using debris from pig scrapings

Fig. 7.32—Static solvent test (Wylde and Slayer 2010).

370  Chemical and Mechanical Methods for Pipeline Integrity

Removal of so-called black powder, an inherently mixed deposit (Section 4.4.4), is described by Sirnes and Gundlach (2012) and Sirnes and Klungland (2015). The papers detail the use of a progression of pig types of increasing aggressiveness. They report that the program started with a standard foam pig, followed by foam pigs with brushes. The next step involved ordinary mechanical cleaning pigs, also without and with brushes, followed by a dummy MFL pig, and in the end an aggressive active-cleaning tool with magnets and brushes and flushing over the discs. (Note that all these devices are shown in Figs. 5.1, 5.7, and 5.9.) The total amount of black powder taken out of the pipeline system described by these authors was approximately 10 tons. Powder was taken out from the collector unit, filter unit, and both pig receivers. The powder is hazardous waste and stored in water barrels. An analysis showed both FeS (mackinawite) and Fe3O4 that could be pyrophoric if allowed to dry. Smith et al. (2011) claimed production of a single acidizing fluid for dissolving organic/inorganic components fouling wells. It contains a miscibility solvent that substantially prevents any phase separation between the constituents and lack of dispersion of the additives in the fluid. One aspect of the invention is the production of a single fluid-acidizing formulation that comprises a relatively high concentration of mineral or organic acid, a miscibility solvent, a surfactant, and at least an anticorrosion additive, which when premixed is stable for extended periods and substantially for periods such as exceeding 1 year and possibly longer at ambient and wellbore temperatures (when under wellbore pressure). The major claim is that the formulation comprises • Aqueous 15% hydrochloric acid solution of approximately 13.8 wt% of the formulation • Miscibility solvent of approximately 59 vol% of the formulation for forming a substantially stable single fluid, the miscibility solvent further comprising approximately  15 wt% of an aromatic solvent  63 wt% of a blend of alcohols, the blend of alcohols being approximately 5 to 15 wt% longchain alcohols and approximately 40 to 60 wt% short-chain alcohols • Approximately 19 to 24 wt% of a surfactant • Corrosion inhibitor of approximately 1.5 wt% of the formulation • Demulsifier of approximately 0.04 wt% of the formulation • Antisludge additive of approximately 0.4 wt% of the formulation • Iron control additive of approximately 1.5 wt% of the formulation • Water as the balance 7.5.3  Uses of Gelly Pigs in Pipeline Cleaning. In the last 30 years, there were developments on the basis of the concept of a gelled pig (see also Section 6.6 for a description of the chemistry of forming these materials and Section 6.7.3 on the use of gels for placing corrosion inhibitors). Several patents and articles described the use and development of this technology. Jaggard and Scale (1977), for example, taught the use of a hydrocarbon fluid gelled by an orthophosphate ester. The gel may be used as an interfacial control for different fluids and to remove residual fluids and solids. Scott (1983) patented a gelled fluid formed by crosslinking xanthan gum using a multivalent metal such as ferric sulfate, aluminum sulfate, or chromium chloride. These “Bingham plastic” gels were used to remove debris from pipelines. Purinton (1984) described the idea of using a gelled fluid as a substitute or as an adjunct to mechanical pigs for cleaning purposes. The underlying concept is that the gelled fluid would be able to flow around obstructions in the pipe (e.g., probes, changes in diameter) without getting stuck. In addition, the gelled fluid may be able to suspend debris. One of the reasons that mechanical pigs become stuck is that piles of debris build up on the leading edge of the pig as it goes down the pipeline. A pig train that is made up of mechanical pigs as well as segments of gelled fluids is shown in Fig. 7.31. This diagram demonstrates how the gel picks up and distributes the solid materials in the gel volume.

Cleaning of Pipelines and Facilities  371

Several different types of rheological properties may be required for different purposes. Gels that are required to go through geometrically different environments and to suspend debris must exhibit pseudoplastic and viscoelastic characteristics and must be shear thinning. Fluids containing crosslinked polysaccharides are a type of fluid that exhibits these behaviors. These are known as powerlaw fluids in that their rheology is characterized by

µ = k′γ n′−1,�������������������������������������������������������������������������������������������������������������������������������� (7.22) where m is the viscosity, g is the shear rate, k′ is the consistency index, and n′ is the behavior index of the fluid. Fluids with this type of rheology will become less viscous when sheared (so the fluids will flow around objects) but will suspend debris when the shear rate is reduced. Many specialized gelled pig formulations were developed. The Purinton (1984) patent just described used a hydrocarbon solvent that was gelled using an alkyl oleyl phosphate and an alkali aluminate salt. These gels could be used for dissolving hydrocarbon-like fouling deposits or for separating various hydrocarbon products. Purinton (1981) also was able to produce a gel of an alcohol, such as methanol, by using hydroxypropyl cellulose (HPC). A pig train consisting of water gelled using hydroxypropyl guar (HPG) and crosslinked with sodium borate was followed by the HPCthickened methanol, followed by a slug of methanol. This sequence was used to remove residual water and thus to dry the pipeline. The HPG gel (Purinton 1985) could also be used to remove debris or could be staged with acidic formulations (Frenier et al. 1980) to remove iron sulfide deposits. The HPG gels can also be used to lay down a biocide coating. One of the largest projects that used gelled pigs was the cleaning of the FLAGS (Far North Liquids and Associated Gas System) gas line in the North Sea. Scott and Zijlstra (1981) described this very large job. The FLAGS gas pipeline runs from offshore St. Fergus, Scotland, for 380 miles to the Brent gas field in the North Sea. A gelled plug system using a Bingham plastic fluid and plastic scraper pigs was used to remove more than 350 tons of mill scale, and other debris from this line. A specific use for gelled fluid pigs was described by Keys and Evans (1991). A total of 119,000 gal of gel and degreaser was used to clean approximately 65 miles of an abandoned crude oil pipeline that was to be converted to natural gas service. A single, complex pig train was able to clean the line. The train consisted of a number of stages of mechanical and gelled pigs as well as degreaser solutions. Nesbitt (1991) described the use of gelled pigs to address stuck conventional mechanical pigs. The gelled solution, made from hydroxyethyl cellulose (HEC) polymer, was able to bypass several stuck mechanical pigs, remove debris from mechanical pigs, and restore the seal. This allowed the drive fluid to push the mechanical pigs to the takeout point. Misra et al. (2013) presented the results of tests conducted to design a fast-breaking gelled fluid for a debris-removal operation in a long, large-diameter subsea pipeline in the Asia Pacific region. In addition to exhibiting good stability and debris-carrying capacity under laboratory test and field conditions, the gel had to be compatible and demonstrate stability when mixed with other additives in the transported hydrocarbon. A highly viscous, aqueous-based linear polysaccharide gel (xanthan based) was designed to satisfy these requirements. Details of the preparation of this gel are in the report and included in Table 7.5. Misra et al. (2013) reported that the xanthan (Fig. 6.36) gel will support up to 2 lbm/gal of debris (See Fig. 7.33) and is compatible with oilfield treatment chemicals such as biocides and paraffininhibiting polymers (Section 6.3.1). A key environmental and operational consideration was the easy and fast disposal of the gel after completion of the cleaning operation. With oxidizing breaker, gels prepared in potable water resulted in viscosity of 5010 mg/kg (Dermal LD50; Rabbit) = 700 mg/kg (Oral LD60; Rat)

IARC Group 1 or 2

skin, eyes, respiratory system

Target Organ Effects

Component

Hydrochloric acid

Component

LD60 / LC50

None known. Not known to cause heritable genetic damage. Not known to cause birth defects or have a deleterious effect on a developing fetus. Not known to adversely affect reproductive functions and organs. See COMPONENT TOXICOLOGICAL INFORMATION below.

COMPONENT TOXICOLOGICAL INFORMATION

Reproductive toxicity: Target organ effects:

Carcinogenic effects: Mutagenic effects: Teratogenic effects:

Chronic Health Hazard

Corrosive. Rapidly causes pain, bums, comeal injury. May cause permanent damage and blindness. Corrosive. Rapidly causes pain, bums, redness, swelling and damage to Skin contact: tissue. Corrosive. Causes pain and severe bums to mouth, throat and stomach. Ingestion: Inhalation: Corrosive. Short exposure can injure lungs, throat, and mucous membranes. Causes pain, bums, choking, and coughing. Not known to cause allergic reaction. Sensitization - Iung: Sensitization - skin: Not known to cause allergic reaction. Toxicologically synergistic None known. products: Other Information: Prolonged exposure at low concentration may cause erosion of the teeth.

Eye contact:

Acute Health Hazard

PRODUCT TOXICOLOGICAL INFORMATION

11. TOXICOLOGICAL INFORMATION

Other Information: Gives off hydrogen by reaction with metals.

Shelf life - (Years):

Hazardous polymerization: Hazardous polymerization does not occur.

Fig. 8.2—Material safety data sheet for 36% HCl.

36

Weight%-Range

Odor: Pungent

Corrosive to metals.

DANGER

EMERGENCY OVERVIEW

Main physical hazards:

2. HAZARDS IDENTIFICATION

Emergency Telephone Number:

Company Identification:

Your Chemical Company

Hydrochloric Acid 36% Unihibited H36

Product Name:

Use of the Substance/Preparation:

H036

Product Code:

1. IDENTIFICATION OF THE SUBSTANCE/PREPARATION AND THE COMPANY/UNDERTAKING

Version:

(USA) (Compiles with USA OSHA 29 CFR 1910.1200 and ANSI Z400.1)

SAFETY DATA SHEET

Product Code: H036

Pipeline/Facility Maintenance Health, Safety, and Environmental Issues   391

392  Chemical and Mechanical Methods for Pipeline Integrity

• As many of the potential safety problems or concerns as possible should be brought to the attention of everyone. Maximum pressure limits should be set at this time, and every highpressure pump operator must be aware of these limits. Instructions for pressure testing the lines and connections must also be covered. The high-pressure treating lines should be tested, and a properly tested line includes tests of each pump in addition to the main treating line. The pressure rating of the pipeline/facility units should be checked to make sure it exceeds the treating pressure. The pretreatment safety meeting is the principal communications to all personnel. A well-organized safety meeting helps ensure that the treatment is an operational success without being a threat to human safety. The safety meeting should be a part of the preoperational planning document and also should include operator personnel so that all ­communications are in place. • Pressure control. Ensure that the pipeline equipment pressure control is always maintained. Many types of valves are in use in pipeline operations and in process plants and have to be maintained as well as being operated by trained personnel using approved procedures. • Precautions for flammable fluids. Note the MSDS as well as the company manual’s handling of the pipeline fluids. Oil-based fluids should be tested for volatility before they are accepted as a fracturing fluid. An oil is generally considered safe to pump if it has a Reid vapor pressure less than unity, °API value less than 50, and open-cup flashpoint of 10°F (–12°C). However, even if the fluid is considered safe to pump, several additional safety rules should be followed when pumping oil. Storage tanks for flammable fluids should be diked and spotted at least 150 ft from the operating area. Many other treating chemicals may have low flashpoints and may be flammable. Any fluid that contains any alcohol may be flammable. Spotting the fluids in this manner helps minimize exposing the work area to fire if problems occur during pumping. Also, all low-pressure hoses should be enclosed in a hose cover to prevent oil from spraying on hot engine components of the trucks, should a hose leak. Care must be taken to ensure that there is no smoking on location. It is a good idea to have all personnel check matches and lighters when they arrive on location to prevent them from unintentionally lighting up. Finally, fire-fighting equipment should be on location and ready to be operated. In this way, a small fire may be contained before it has a chance to spread and become a major disaster. To further reduce the risk for an on location fire, equipment should be both bonded and grounded. Bonding forms an electrical conduit between vessels containing flammable fluids and the equipment that transfers this fluid and pumps it. It ensures that the electrical potential from electrostatic buildup is controlled between all the units handling the flammable fluid. Grounding controls the electrostatic buildup between the equipment and the ground. When bonding and grounding are combined, it ensures that all points between the wellhead and each piece of equipment are at the same potential and that any static electricity generated during the operation is dissipated without the possibility of a static discharge. • Environmental considerations. Treatment operations should be conducted using sound environmental practices to minimize the potential for contamination of air, water, and soil. All operations should comply with all applicable environmental laws and regulations. Hazardous material spills should be cleaned up quickly in accordance with a spill plan. All waste and unused materials should be handled and disposed of in accordance with locally approved regulations. Note that more details regarding ecotox management and waste disposal are in Sections 8.3, 8.4, and 8.4.2. An American Petroleum Institute (API) recommended practice, API RP 54 (1999), covers all the items described in the preceding list, as well as others such requirements for job site setup, in ­significant detail. The main topics in the API RP 54 table of contents are listed here.

Pipeline/Facility Maintenance Health, Safety, and Environmental Issues   393

• • • • • • • • • • • • • • • • • • •

General References Definitions Injuries and first aid Personal protective equipment Operations Fire prevention and protection Flammable liquids Drilling and well-servicing rig equipment Drilling and well-servicing rig electrical systems Pumping units Special services Stripping and snubbing Drillstem testing Acidizing, fracturing, and hot oil operations Cementing operations Gas, air, or mist-drilling operations Hot tapping and freezing operations Hot work, welding, and flame-cutting operations

A generalized oil/gas industry health and safety process that has been developed for use in sites for unconventional resources such as the shale plays is described by Devlin et al. (2014). This report claims that an API (2016) Exploration and Production Health Issues Group, has recently been formed to provide research, scientific analysis, and guidance on health concerns regarding unconventional resource production activities. This group has a particular focus on potential community health impacts. The principal objectives of the group are to • Identify, analyze, and respond to health-related allegations • Develop and manage research projects that provide sound data for industry and government risk management decisions • Communicate all scientific findings through peer-reviewed publications • Through documents and various public forums, advance public understanding of the benefits and risks from production of unconventional resources Specific interest areas include chemical disclosure, public health impacts, industry activity to minize risk, and community outreach. Note that the items described above are to be used not as standards but as common-sense considerations applied on the basis of one’s experiences. The legal and regulatory aspects of safety and ecotox management vary by state, country, and area and are continually evolving. These concerns are beyond the scope of this book. 8.2.2  Specific Pipeline/Facility Considerations. Most of the items noted in the previous paragraphs apply to many if not all oil/gas industry operations. A report by Smart (2003) lists safety concerns and suggestions that are unique to pipeline operations. This author’s recommendations include • Do not stand in the projected path of a pig, including the bends and traps (see Figs. 5.52 and 7.5). These tools can move with high velocities and with high forces and are known to knock the door off of a trap or to penetrate the wall of a bend. • Never work alone. Always have another employee with you.

394  Chemical and Mechanical Methods for Pipeline Integrity

• Restrict access to authorized persons only. • Take special cautions for loading/unloading pigs. Regarding the latter, launchers and receivers must be pressurized at various times, and releasing pressure to provide access must be performed with the knowledge that the pigs may store pressure as a result of a tight fit in the barrel. Furthermore, paraffin and hydrates may block communication within the pressurized sections and make safe release of pressure difficult. The author of this book contends that the handling of the devices after they have transited the line and have entered the receiver is particularly hazardous. All systems must allow for the receipt of pigs at the launcher, because blockages in the pipeline may require the pigs to be pushed back to the launcher. There are inherent risks in opening the barrel to atmosphere, and so care must be taken to ensure that the barrel is depressured before opening. If the barrel is not completely depressured, the pig can be ejected from the barrel. In fact, operators have been severely injured when standing in front of an open receiver barrel door (see Fig. 5.3). When the pipeline product is sour, the barrel pressure should be evacuated to a flare or another system where the sour gas is burned or subjected to a scrubber. Operators should have a self-containing breathing apparatus available when working on sour systems. The pig trash (Fig. 4.22) pushed out of the line may contain toxic or environmentally hazardous material and must be disposed of in compliance with federal and state laws. In some cases, the sludge may be contaminated by naturally occurring radioactive materials (NORM) (Fisher 1995), and then the pig itself must be decontaminated before reuse or disposal (see TDW 2005a, 2011). All the fluids/solids leaving the line after a pigging operation must be contained, separated, and treated if necessary. Note that some pig wastes from gathering lines may be considered to be exempt as a legal hazardous waste (RRCT 1998), but still are regulated as far as disposal is concerned. See Section 8.4 for information of handling of oilfield waste products. The reader should review the additional details provided by Smart (2003). Appendices in Smart’s report itemize specific safety checks that should be accomplished while working around active pipelines. Also note several additional articles on pipeline safety. McIlroy (2009), for example, describes a program that calculates the direction and concentration of toxic gases such as H2S that may be evolved from activities at a wellsite or pipeline worksite. This program could be used in the case of a gas release during operations to warn local communities. Tayab (2012) claimed that preventing the “next accident” is a key issue that poses a challenge to high-risk businesses such as the oil/gas industry. Over the last 5 years, Tayab’s company has exercised significant efforts and resources to achieve a high safety standard at its worksites, but trends of reoccurring incidents have been observed. Concerns have been raised whether any lessons have been learned from previous accidents. The author also claimed that extra efforts were put in place in incident investigation processes by introducing extended root-cause analysis training and use of HEARTS (the HSE electronic action, reporting, and tracking system), to accomplish trend analysis. The data indicated that most root causes (72%) can be linked to human activity, resulting in unsafe behavior and/or creating unsafe conditions. Consequences of not addressing behavioral issues and the recommendations for influencing the positive behaviors are discussed here at length. Very specific pipeline/facility issues include injuries sustained as a consequence of the operation of the many valves that must be opened and closed during maintenance and operations. Tys and Attwood (2013) noted that one constant in every process plant is the high number and diversity of valves that control the flow of feedstock, products, and service liquids and gases. The ergonomic issues they associated with the design, operation, and maintenance of valves include • Physical stress to open and close the valves • Potential for injuries and subsequent related costs

Pipeline/Facility Maintenance Health, Safety, and Environmental Issues   395

• Lack of access to process-critical valves • Difficulty to remove and replace the valves • Potential process upsets when valves cannot be operated in required time These authors also describe a study by a refinery to evaluate and change valve placement and operations to reduce muscular and skeletal injuries. ANSI/API RP1173 (2015), which covers pipeline safety management, provides pipeline operators with safety management system requirements that when applied provide a framework to reveal and manage risk, promote a learning environment, and continuously improve pipeline safety and integrity (see Section 3.5). At the center of a safety management system is the operator’s existing pipeline safety system, including the operator’s pipeline safety processes and procedures. ANSI/ API RP 1173 is claimed to provide a comprehensive framework that defines the elements needed to identify and address safety for a pipeline’s life cycle. These safety requirements identify what is to be done but then leave to individual pipeline operators the details associated with implementation and maintenance of the requirements. The recommended organization framework involves a wideranging approach of four key steps: 1. 2. 3. 4.

Plan Do Check Act

This approach is named the American National Standard for pipeline safety management systems. 8.3  Health, Safety, and Environmental Management Developing “green,” low-toxicity formulations or less hazardous chemicals has become the goal of many formulators in HSE management; however, there is not total agreement on what so-called green chemistry is (Gupta 1998), even though many proposed green formulations are shown in the following sections. Darling and Rakshpal (1998) of the US Environmental Protection Agency defines green products as “pollution prevention at the molecular level, designed to reduce or eliminate the use or generation of chemicals that are hazardous to human health or the environment.” Note that Darling and Rakshpal’s definition is used throughout this book wherever possible, but also note that many of the references cited here define “green” in relationship to more-limited criteria, such as a particular governmental regulation or a specific list of regulations. Some of these are listed in Section 8.3.1, as well as in Kelland (2009, Chap. 1). 8.3.1  Chemical Selection To Enhance Health, Safety, and Environmental Management Compliance. Developing, achieving, and managing an effective HSE program must start at the beginning of any chemical-related project. A US Department of Energy report from an industry and government committee (Deutch et al. 2011) agrees with this assessment and notes that the major recommendations for safer shale gas production include replacing diesel oil from all fracturing formulations (Frenier and Ziauddin 2014) and disclosing most components of the mixtures. Intense interest in green chemistry or other HSE issues is a fairly recent phenomenon. At the 1985 European Symposium on Corrosion Inhibitors (6SEIC), there were no papers with “toxicity” or “green” in the title, though some papers described lower-toxicity materials. By 1995, the first NACE International symposium was considering green inhibitors. Since then, additional NACE symposia were held in 1998 and 1999. Likewise, many of the papers at 7SEIC in 1990 and 8SEIC in 1995 were concerned with toxicity or the environment. The patent literature has also started to describe formulations in which toxicity values or environmental improvement are the subject of the major claims of the inventors.

396  Chemical and Mechanical Methods for Pipeline Integrity

Today, most articles or presentations involving corrosion inhibition and other oilfield chemicals have some mention of an HSE type of assessment or concern. Clearly, HSE or green concerns are now politicized issues. Many HSE-related conferences have been held within the oil/gas production chemicals community. In particular, papers have been presented in the Formation Damage and Oilfield Chemical symposia, and SPE frequently holds conferences such as the SPE International Conference on Health, Safety, and Environment in Oil and Gas Exploration and Production held 12–14 April 2010 in Rio de Janeiro, Brazil. This conference covered a wider range of HSE issues than ever before. Several specific programs for developing improved and safer chemicals were described. 8.3.2  Chemical Development Processes. A major oil/gas service company described a quantitative assessment of HSE impacts of prospective new well production chemicals that includes four areas of concern (Purinton and Manning 1996): global restricted use, physical and chemical properties, environmental screening, and exposure hazards. The program considers specific criteria in each category, with scores identified. The total scores are added, and the worst products (i.e., those with the highest scores) are then slated for elimination or modification. The four areas of concern are listed as follows and under them are the specific items that are evaluated or tested, with scores noted in parentheses. Note that this program was developed in the mid-1990s, and some of the individual control items may have changed; however, the methodology can be used with appropriate updates. 1. Global “Restricted Use” Consideration. This consideration notes if the product and its ­components are listed on any of the following: European Union (EU) controlled substances (4) EU land-banned material (4) North Sea banned list (3) Organohalogens (nonpolymeric) (4) CERCLA/CWA Hazardous substance table (3) 302.4 with RQ 100 lbm or less Environmental Protection Agency (3) extremely hazardous materials Carcinogenic, mutagenic, or toxic to reproduction (CMR) substances (3) Marine pollutants (2) Resource Conservation and Recovery Act waste code U or P (4) California Prop 65 (2) Priority Pollutants Clean Water Act (2010) (4) Process safety management chemicals list (4) Arms Control and Disarmament Agency (ACDA) schedule (arms control substances) (2) US Department of Health Annual Carcinogen Report (3) Toxic metals list (4) 2. Physical/Chemical Properties. An assessment then gives scores for physical/chemical ­properties as per the following values. Flammability based on flashpoint: >141°F (60.5°C) (0) 73 to 141°F (23 to 60.5°C) (1) 10 to 73°F (–12 to 23°C) (2) 3 and MW < 600 (2) Log Po/w < 3 and/or MW > 600 (0) Toxicity: EC50 1 mg/L (4) LC50 1–10 mg/L (3) 10–100 mg/L (2) >100 mg/L (0) 4. Exposure Hazards. Assessed are exposure hazards using the following data sources and data provided on vendor MSDSs. The testing procedures and nomenclature were described by Frenier (1996): • Registry of toxic effects of chemical substances • MDL Information Systems, Inc. Hayward, CA, MSDS database • Canadian Center for Occupational Health and Safety Information (CCINFO) MSDS database • Other resources as applicable Score assignments are as follows: Inhalation: Highly toxic: LC50 < 0.5 mg/L (4) Toxic: 0.5 < LC50 < 2 mg/L (3) Low-toxic: LC550 > 2 mg/L (0) Ingestion: Extremely toxic: LD50 < 5 mg/kg (4) Highly toxic: 5 < LD50 < 50 mg/kg (3) Toxic: 50 < LD50 < 500 mg/kg (2) Low-toxic: LD50 > 500 mg/kg (0) Skin absorption: Highly toxic: LD50 < 40 mg/kg (4) Toxic: 40 < LD50 < 200 mg/kg (3) Low-toxic: LD50 > 200 mg/kg (0) Carcinogen: If the product or its components are classified, the following International Agency for Research on Cancer group rating (or equivalent rating from other agencies) is used: 1 = 4, 2A = 3, 2B = 2, 3 or none = 0. If the product or its components is a mutagen: yes = 2, no = 0. The assessment process will “score” the potential product and will attempt to produce a product with the lowest overall score. A score of zero is the goal. This initial assessment will

398  Chemical and Mechanical Methods for Pipeline Integrity

probably not be the final consideration of toxicity. If the material is to be used in North Sea countries, or if the material is not registered in various potential use countries, additional testing may be required. Jenkins (2011) provided a review of the current UK offshore regulations and the protocols for screenings. See Fig. 8.3 for a diagram of the Harmonized Mandatory Control Scheme (HMCS). This chart takes the investigator through the steps to determine if additional data are required before a new chemical can be introduced. The acronyms are • • • • •

OSPAR, convention for protection of environment (Europe) PLONAR, pose little or no risk to environment HOCNF, harmonized offshore chemical notification format OEDC, various methods of biodegradation determination CHARM, chemical hazard assessment, and risk management model

The EU’s Parliament has passed new regulations for the safe use of chemicals in the EU countries. These regulations are called “REACH” (EC 2006). Notably, the regulations stipulate that all substances manufactured or imported into the EU at 1 ton/ yr or more per manufacturer/importer must be tested and registered. These materials include • Substances supplied in their pure form or in preparations • Substances supplied in products intended to be released from the products • Certain substances that are present in the products and have some adverse effect on health or the environment

Is substance on the PLONOR list

Yes

No Is substance on Annex 2 of the OSPAR Strategy with regard to Hazardous Substances or considered by authorities to be at special concern for the marine enviroment?

Expert judgement positive

Yes

D. Refusal of Permission

Yes

No

No FULL HOCNF NEEDED

Yes

No

Is substitute available?

Yes Is LC50 or EC50 Yes Is the substance inorganic?