The Fourth International Forum on
Transportation of CO2 by Pipeline

CO2 Pipelines Short Course 17-18 June
Conference 19-20 June
Newcastle, UK

Wednesday 19 June

by Dr Julia Race, School of Marine Science and Technology, University of Newcastle, Newcastle upon Tyne, UK

Standards and guidance for CO2 transportation

by Dr Achim Hilgenstock, Dr-Ing. Sven Anders, and Rabea Visse, E-ON New Build & Technology, Gelsenkirchen, Germany

Effective climate protection to avoid a global rise in temperature is only possible if CO2 emissions are reduced significantly. Considering that worldwide energy demand is, at the same time, rising steadily, E.ON is making an important contribution to reducing CO2 output with its technological improvement of conventional power plants. The Group’s work in this respect includes innovative methods of CO2 capture in coal-fired power plants and subsequent storage of CO2 in geological formations (carbon capture and storage or CCS).

Challenging technology is used for capturing CO2 from the flue gas and injecting it underground. But while it is a key element in the CCS process chain, transportation of CO2 from where it is captured to where it is stored is often left out of the public debate. This paper deals with line-based transportation of CO2 and presents the current status in national and international standardization.

by Hamish Holt, DNV, Hovik, Norway

DNV has released new industry guidance on managing major accident risks associated with handling large quantities of CO2. The guidance, developed within the CO2RISKMAN joint-industry project (JIP), provides a comprehensive reference source aimed at helping to raise understanding of the issues, challenges, and major accident potential of handling large quantities of captured CO2.

The new guidance was developed over a 15-month period with support from 16 industry and regulator organizations including Shell, Chevron, Maersk Oil, E.ON, National Grid, Gassco, AMEC, and the HSE. It provides valuable knowledge and information for projects, operations, and regulators involved in CCS. The guidance describes how the properties and characteristics of the impure CO2 stream can lead to loss of containment due to, for example, low-temperature embrittlement, damage to elastomers, lubricant breakdown, and rapid internal corrosion. It gives details on escalation potential as well as possible consequences to people and to the local environment. To increase value, the guidance also offers a CCS project a targeted list of measures that should be considered within a project’s risk management process to reduce the likelihood and/or consequences of CO2 leaks.

The 400-page guidance which is subdivided into four documents can be downloaded at no charge from Since its launch on 31st January it has been downloaded by nearly 200 people in over 25 countries.

The presentation will introduce the new guidance, highlighting its content, layout and value within and across CCS projects.

Research project updates

by Russell Cooper, National Grid, Warwick, UK

National Grid’s COOLTRANS (CO2Liquid pipeline TRANSportation) research programme is due to be completed in December, 2013. The aim of this £8-million programme is to address and resolve the key issues relating to the safe routeing, design, and construction of onshore pipelines for the transportation of anthropogenic, high-pressure, dense-phase CO2 from power stations and other industrial emitters to offshore locations for underground storage. The research has provided key learning relating to the differences between dense-phase CO2 with impurities and other fluids, and the challenges which must be addressed in the design of dense-phase CO2 pipelines in terms of the differences and the gaps in existing codes and standards.

As presented at this Forum in 2011 and 2012, the COOLTRANS research work is being undertaken by a number of UK academic and industry experts. The findings of the research carried out by different researchers must be integrated and variances reconciled at chosen interfaces, and the effect of uncertainties assessed in order to develop practical guidance for real projects. The strategy developed to achieve this has involved the development of a number of case studies in which theoretical predictions are compared with experimental data. In the final stage of the COOLTRANS programme, a major case study of direct relevance to the demonstration of safe design and operation of a dense-phase CO2 pipeline is being carried out. This case study will use the results of state-of-the-art research to verify and inform the assumptions applied in the study of pipeline failures and releases which will underpin the safety justification for the operation of the pipeline.

This paper presents an overview of the knowledge obtained during the COOLTRANS research programme to date, presents further results of integrated case studies, and explains how the research work is being used and applied in a practical pipeline safety justification study.

by Prof. Valerie Linton, C Lu, G Michal, A Godbole, X Liu, N Birbilis, J Hayes, Peter Tuft, Phil Venton, and P Balfe, Energy Pipelines CRC, Faculty of Engineering, University of Wollongong, Wollongong, Australia

In Australia, gas and oil pipelines are designed, constructed, and operated to the Standard AS2885: Pipelines – gas and liquid petroleum. Should pipelines be built in Australia to carry CO2 from capture sites to storage sites, AS2885 could be used to cover the design and operation of supercritical CO2 pipelines. A gap analysis was conducted on AS2885 to identify parts of the Standard that required revision to be applicable to CO2, and this work produced a draft informative appendix that has been incorporated into AS2885.1, published in 2012.

The Energy Pipelines CRC is well advanced on a programme of work to fill in gaps in the current knowledge of CO2 behaviour so that this information can be offered for incorporation incorporated into later versions of the Standard. Additionally, the Energy Pipelines CRC is carrying out work on the public safety, community consultation, and organizational requirements for CO2 pipelines. Finally, a cost-benefit exercise is being conducted on the application of the results of the work. This paper provides an overview of this work and the benefits for CO2-carrying pipelines in Australia.

by Dr Stefan Jäger, Christoph Bosch, Salzgitter Mannesmann Forschung GmbH, Duisburg, Germany,  Michel Meyer, GDF Suez, France, and Antonio Lucci, Centro Sviluppo Materiali, Italy, and presented by Dr-Ing. Marion Erdelen-Peppler, Salzgitter Mannesmann Forschung GmbH

The concept of carbon capture and storage (CCS) might become a huge contribution for reducing the amount of anthropogenic CO2 released to the atmosphere and thus can help in containing the greenhouse effect. Since the implementation of this concept requires the application of new techniques for capture, transport, and storage of liquefied CO2, a multitude of research projects were started on these topics; first, mainly on capture and storage, but later on also on transport of CO2. One of these projects, SARCO2-A, dealing with several aspects of the pipeline transportation issue for CCS, was carried out by a consortium of companies encompassing an electricity and gas transporter (National Grid), two energy providers (GdF Suez and Open Grid Europe), an oil and gas company and energy provider (ENI), three pipe manufacturers (Europipe, Salzgitter Mannesmann Line Pipe, and Vallourec & Mannesmann Tubes), and two research centres (Centro Sviluppo Materiali and Salzgitter Mannesmann Forschung), all contributing to other aspects of transporting CO2.

This work provides an update on the detailed research approach of the SARCO2-A project and presents selected results found out in the course of investigations. Findings of an in-depth study on likely composition and constitution of CO2 mixtures coming out of the different capture processes are presented. This fundamental study was used as basis for the experimental work in the project. The experimental part encompasses investigation of different corrosion phenomena like general and localized corrosion and corrosion forms related to certain impurities in anthropogenic CO2. Corrosion-protection concepts in shape of inhibitor systems or coatings are studied in small and full-scale tests for classifying their applicability in CO2 pipelines. For conducting safety and reliability analyses of such pipelines, the leakage behaviour of CO2 upon release is assessed, with experimental data facilitating the simulation of CO2 release phenomena.

by Prof. Haroun Mahgerefteh1, Sergey Martynov1, Solomon Brown1, Michael Fairweather2, Robert Woolley2, Sam Fall2, Vagesh Narasimhamurthy3, Ole Jacob Taraldset3, Trygve Skjold3, Ioannis Economou4, Dimitrios Tsangaris4, Laurence Cusco5, Mike Wardman5, Simon Gant5, Jill Wilday5, Yong Chun Zhang6, Shaoyun Chen6, and Christophe Prous7

1 University College London, UK
2 University of Leeds, UK
3 GEXCON AS, Norway
4 National Research Centre for Physical Sciences, Greece
5 Health and Safety Laboratory, UK
6 Dalian University of Technology, People's Republic of China
7 Institut National de l’Environnement et des Risques, France

Without a clear understanding of the hazards associated with the failure of CO2 pipelines, carbon capture and storage (CCS) cannot be considered as a viable proposition for tackling the effects of global warming. The development of reliable validated pipeline outflow and dispersion models are central to addressing this challenge. This information is pivotal to quantifying all the hazard consequences associated with failure of CO2 transportation pipelines forming the basis for emergency response planning and determining minimum safe distances to populated areas.

This paper presents an overview of the main findings of the recently completed CO2PipeHaz project ( focused on the hazard assessment of CO2 pipelines to be employed as an integral part of CCS. Funded by the European Commission’s FP7 Energy programme, the project’s objective is to address this fundamentally important issue.

The development of the state-of-the-art multi-phase heterogeneous discharge and dispersion models for predicting the correct fluid phase during the discharge process are given special consideration given the very different hazard profiles of CO2 in the gas and solid states. Model validations are based on both small-scale controlled laboratory conditions as well as large-scale field trials using a unique CCS facility in China, the world’s largest emitter. The large-scale tests involve the full-bore rupture and puncture of a specially constructed fully instrumented 250-m long, 230-mm internal diameter, pipeline containing dense-phase CO2. The heterogeneous flow patterns in the ruptured pipe and the near-field dispersion region are fully investigated and modelled in order to obtain a complete understanding of the discharge phenomena and gas plume behaviour following a large-scale CO2 release.

The understanding gained is used for evaluating the adequacy of control measures in CO2 pipelines, with best practice guidelines being developed.

The research leading to this work has received funding from the European Union 7th Framework Programme FP7-ENERGY-2009-1 under grant agreement number 241346.

by Dan Allason and Keith Armstrong, GL Noble Denton, Spadeadam Test Site, UK, and Julian Barnett, National Grid Carbon, Warwick, UK

The COOLTRANS research programme has included a major programme of experimental testing carried out by GL Noble Denton for National Grid. The test work has included above-ground shock-tube test and venting test releases, and below-ground puncture and rupture releases and full-scale fracture-propagation tests.

The GLND Spadeadam test site is a large-scale experimental test facility which is continuously manned with full supporting infrastructure, including on-site workshops equipment, offices, and conference facilities. GLND has wide-ranging experience of, and capability for, handling and testing dense-phase CO2, and has the capability for accurate preparation of large-scale mixtures using a circulating loop to achieve homogeneous conditions with a heat exchanger allowing accurate temperature control. The Spadeadam test facility is located with an MOD RAF site, which allows the application of large, controlled, exclusion zones. GLND has extensive and recognized experience and expertise carrying out full-scale crack-propagation and dispersion tests for natural- and rich-gas inventories. The expertise, experience, and facilities at Spadeadam have enabled National Grid to develop an efficient programme of experimental tests, in which – where possible – the test scope is extended so that the same test rigs are used for more than one purpose. This included, for example, the use of the full-scale test rig to investigate under-pressure welding requirements under-flowing conditions, and to carry out venting trials including measurement of fluid and material temperatures. A key feature of the programme is the reuse of equipment such as pumps, flow meters, and instrumentation.

This paper presents the experimental testing undertaken in support of the National Grid COOLTRANS programme at Spadeadam, and describes the test set-up and demonstrates how the experience and expertise available in GLND has added value to the research programme.

Fracture propagation

by Dr Kamal K. Botros, NOVA Research and Technology Centre, Calgary, Alberta, Canada, and Eduardo Hippert Jr and Pedro Craidy, Petrobras/ CENPES/ PDP/ TMEC, Rio de Janeiro, Brazil

Offshore oil and gas production operations produce off gas (residue gas) that is primarily a binary mixture of CO2 and methane (CH4) in varying compositions. This gas is then transported back to the reservoir location from the processing facilities via dedicated high-pressure pipelines and then re-injected into the reservoir. While pure CO2 pipeline technology is well-established, the mixture composition of this residue gas involving CH4 content of up to 45% (by mole) is unique in so far as its implication on the fracture controls aspects of the design of such a pipeline. Supercritical CO2 (and CO2+CH4 for that matter) is a particularly challenging fluid from this point of view, because its thermodynamic characteristics are such that a very high driving force for fracture can be sustained for a long time. Whilst the properties of supercritical CO2 are well understood, the impacts of additional component, such as CH4 in the CO2, are not. Even though CO2 is not flammable, it is an asphyxiating gas that is denser than air, and can collect in low-lying areas. From a safety point of view, it is as important to control fracture in a CO2 pipeline as in one transporting a flammable fluid. The determination of the required toughness for the arrest of ductile fracture requires knowledge of the decompression behaviour of the contained fluid.

This paper presents experimental results of the decompression wave speed obtained from shock-tube tests conducted on mixtures of CO2+CH4 from initial pressure of approximately 30 MPa and initial temperature of approximately 40oC. Comparison between measured vs. predicted decompression wave speeds by models based on GERG-2008 and Peng-Robinson (EOS) are made. The results show that for the mixtures tested, GERG-2008 shows much better performance than PR, particularly in the PρT and wave speed. The part of the decompression wave speed below the plateau (saturation pressure) is very small. This was determined to be due to the fact that the initial pressure is quite higher than the bubble point pressures: hence the isentrope does not intersect the two phase region until the pressure significantly drops to a value close to the choke pressure. The implication of the plateau pressure on the required material toughness will be discussed.

by Hakon O Nordhagen1, E. Aursand2, M. Hammer2, S. T. Munkejord2, and Dr Cato Dørum1

1 SINTEF Materials and Chemistry, Trondheim, Norway
2 SINTEF Energy Research, Trondheim, Norway

A common assumption made in current empirically based ductile-crack-arrest methodologies is that long-running ductile fractures take place in a steady-state condition. That is, a constant crack speed is necessary (but not sufficient) for a long running fracture to occur. However, as witnessed in the literature on results from full-scale pipeline crack-arrest experiments, the ductile-fracture velocity (DFV) measurements often show a strong and seemingly irregular behaviour. Based on timing-wire measurements of the DFV, a variation of more than 100 m/s within meters of crack propagation is not uncommon – also for pipe segments with insufficient fracture resistance for self-arrest. Although this behaviour is most often seen in backfilled conditions, it might also occur in pipelines with no backfill. No common explanation for this phenomenon can be found in the literature.

Possible mechanisms that may explain this scatter in DFV data include (1) the ability of timing wires to break consistently – even in un-backfilled tests, (2) variations in depth, density or strength of the backfilled soil, or (3) local variation in pipeline material properties such as dynamic fracture resistance and elastic-plastic properties. Because this sensitivity of the DFV to slight variations in the system parameters, simple (uncoupled) models, such as those used in, cannot be expected to predict DFVs which are always close to the measured values.

Through a computationally efficient, physically based, and fully coupled crack-propagation model we have identified a fourth possible source for variations that can be seen in the DFV. For gases such as CO2 and rich natural gas, where a vapour-to-liquid phase transition takes place upon depressurization, a combination of a high saturation pressure and a low acoustic velocity of the gas (and a low fracture resistance) can cause a ductile propagating fracture to outrun the depressurization wave. As we will present in this paper, the position of the crack tip will fluctuate around the transition point between the liquid and the liquid-vapour region in the pipeline. This will lead to a systematically fluctuating DFV with amplitude of more than 100 m/s, caused mainly by the sudden change of both pressure and acoustic velocity (decompression speed) across the phase boundary.

Though this simulated systematic variation of DFV has not been compared to full-scale crack-arrest experiments with CO2 (not available yet), it might add to the discussion on the analysis and interpretation of scatter in already existing crack-arrest data sets where a two-phase boundary might exist.

by Hiroaki Nakai and Prof. Shuji Aihara, Dept. Systems Innovation, Graduate School of Engineering, The University of Tokyo, Japan

Prevention of unstable crack propagation is one of the most crucial subjects in maintaining the safety of CO2 pipelines. There are some existing models for evaluating the crack propagation and arrest in natural gas pipeline, i.e. the Battelle Two-Curve model, HLP model, and others. We have to be careful in applying these methods to CO2 pipelines because they are more or less dependent on experimental data of full-scale burst tests of natural gas pipelines.

As has been recognized by some studies, the decompression behaviour of CO2 fluid associated with pipe fracture is very different from that of natural gas. Nowadays, some experimental data on full-scale tests of CO2 pipeline have become available. In the present paper, the existing models with the one developed by the authors are subjected to evaluation using these data and the results are compared and discussed. The UT model developed by the authors is not based on full-scale tests data of natural gas pipelines but on physical assumptions regarding deformation of pipe wall, crack propagation, gas decompression, and coupling of these features. It is expected that the UT model can be applied to CO2 pipelines without modification by, for instance, parameter adjustment. A comparison of the models for evaluating decompression behaviour of CO2 fluid is also made, and the result is discussed.

by Julian Barnett, National Grid, Warwick, UK

The design principles for pipelines to be used for transporting high-pressure gas or liquid with a high vapour pressure require that the pipe toughness of the pipeline is sufficiently high to arrest a propagating crack. Dense-phase CO2 containing impurities is a high-vapour-pressure fluid, so pipelines transporting this fluid are susceptible to long running fractures. In addition, the vapour pressure of the fluid is sensitive to the level of impurities, so a maximum allowable vapour pressure for the pipeline must be established.

Fracture propagation is a complex phenomenon which is not fully understood. The methods for estimating the toughness required to arrest a running ductile fracture are semi-empirical. Consequently, when new grades of line pipe steel are developed or when pipelines are to be used to transport new types of fluids (for example, rich gas or CO2) or to transport fluids at higher pressures, or at higher design factors, full-scale fracture-propagation tests are required to extend the range of validity of the existing semi-empirical methods. These tests are expensive, but necessary.

National Grid has conducted two full-scale fracture-propagation tests using a dense-phase CO2-rich mixture. The two tests were successfully conducted at GL Noble Denton’s Spadeadam test site in Cumbria, UK, on behalf of National Grid, in April and October, 2012. These are the first such tests of their type using modern, high-toughness, linepipe steel. The main aim of the two tests was to determine the level of impurities that could be transported in a 914-mm diameter, 25.4-mm wall thickness pipe. The results, which are confirmed to be valid, indicate that the existing semi-empirical models are non-conservative and do not apply to dense-phase CO2 pipelines. This key learning has been obtained and is being addressed through the COOLTRANS research programme. The results and the implications of the full-scale fracture-propagation tests are covered in this presentation.


Thursday 20 June

Materials degradation and cracking

by Arne Dugstad, Malgorzata Halseid, and Bjørn Morland, Institute for Energy Technology, Kjeller, Norway

CO2 has been transported and used for EOR for more than 30 years with a good track record. Corrosion seems to be insignificant as long as the water content in the CO2 is below the threshold where an aqueous phase can form.

Regardless of the safe water level, accidental ingress of water in a complex network of pipelines might happen. Water and other impurities might also precipitate if the system needs to be depressurized. The acceptable response time after water contamination will be system-specific and depend on the corrosion rate and corrosion allowance. Presently, the corrosion rate in a pipeline exposed to an aqueous phase cannot be estimated accurately due to lack of corrosion data.

The present paper discusses upset conditions and gives a review of the very few publications actually reporting relevant corrosion data. The paper also discusses results obtained in recent experiments at IFE and the experimental challenges that are encountered when impurities like SOx and NOx are present.

by Daniel Sandana and Mike Dale, Macaw Engineering, Newcastle, UK, and Dr E A Charles, and Dr Julia Race, Newcastle University, Newcastle, UK

Transporting anthropogenic CO2 in pipelines, either in dense phase or gaseous phase, is an essential component in the practical realisation of carbon capture and storage (CCS). Whichever phase is considered, the likelihood and severity of internal degradation mechanisms arising from CO2 transportation under normal operating conditions and under process upsets needs to be assessed.

Whilst internal corrosion has been a focus of research in this area, the risk of stress-corrosion cracking (SCC) has not been extensively investigated. This paper explores the level of risk posed by SCC in CO2 pipelines, and gaps in current knowledge, together with a presentation of preliminary test results that investigate the presence of SCC in simulated CO2 environments in the presence of impurities.

by Dr Shiladitya Paul and Dr Bernadette Craster, TWI, Granta Park, Cambridge, UK

In the papers presented at previous editions of this Forum, TWI discussed the advances made and technology gaps that remain in understanding materials behaviour in high-pressure CO2. It became evident from the technology gap review that although there is considerable experience of testing materials in lower-pressure gaseous CO2, the experimental data in dense-phase CO2, especially in the presence of a second phase such as H2S, is sparse. As the CO2, either captured from power plants or obtained in produced fluids from oil and gas exploration, is likely to contain multiple components, test methods need to be developed to assess the behaviour of materials such as metals, polymers, and metal-polymer systems in contact with such environments. The test protocols should include, but not be limited to, the assessment of: (a) corrosion damage of pipeline metallic materials and welds, (b) permeation, ageing and decompression behaviour of polymeric seals and liners, and (c) structural integrity and fitness for service in high-pressure CO2. The experimental data generated in (a) and (b) when incorporated in (c) give key information regarding a pipeline’s suitability and its life in CO2 service.

In this paper we describe a selection of initial results from on-going research at TWI to address some of the critical issues involved with transport of CO2 for carbon capture and storage (CCS). This will improve understanding of materials to allow prudent materials selection for confident new build design or rerating of existing infrastructure to proceed with increased assurance. Development of novel test methods for monitoring corrosion and materials’ degradation, their limitations, and applicability of phase equations and safety issues, are also discussed.

Design and operations

by  Julian Barnett, National Grid Carbon, Warwick, UK, and Dick Wilkinson and Keith Armstrong, GL Noble Denton, Loughborough, UK (abstract to follow)

National Grid has extensive experience in the management and execution of under-pressure operations on the UK natural gas pipeline system. Under-pressure operations are of major importance, as they allow the pipeline to be modified and/or repaired without interrupting operation or requiring decommissioning. Under-pressure operations include welding, hot-tap and stopple, and the installation of sleeve repair systems.

For gaseous-phase CO2 pipelines, the experience and procedures National Grid applies to the natural gas system are applicable. However, the higher pressures and different phase behaviour of dense-phase CO2 require careful consideration. Under-pressure operations on dense-phase CO2 pipelines are likely to be needed in order to allow required activities to be carried out without affecting the emitter’s operation and access to the pipeline, and avoiding the phase changes and associated temperature changes that will be incurred due to pressure reduction if decommissioning is required. Under-pressure operations are hazardous, so there is a need to develop procedures and set requirements for dense-phase CO2 pipelines in a safe and controlled way. Initial work has been carried out by National Grid during the COOLTRANS experimental work. Advantage was taken of the flow loop set-up for the full-scale fracture-propagation test to carry out under-pressure welding trails under full-scale operating conditions.

This paper presents the results of the under-pressure welding trails, discusses the issues identified, and considers the additional developmental work needed to ensure a comprehensive suite of under-pressure operations and procedures are available for the pipeline operator.

by Justin Alexander, FEESA Ltd, Farnborough, UK

Proposed schemes for CCS are becoming increasingly complex in terms of pipeline network architecture. Schemes are being proposed that use a pipeline network to route CO2 gathered from clusters of CO2 producers to a central CO2 compression/dehydration facility, after which the CO2 is routed to the injection wells, which potentially involves another pipeline network. Such schemes have many permutations of network architecture, in turn leading to a potentially vast number of system-design permutations given the available selections of pipeline diameters, well tubing diameters, pipeline insulation options, booster-heater options, etc.

The paper describes the application of an established thermal-hydraulic pipeline network modelling tool to a CCS scheme proposed for NW England, which allowed conceptual designers to rigorously and efficiently evaluate many different permutations of the pipeline network design (6000 permutations were evaluated over a two-week period). The study results enabled the project team to then focus on a selected network architecture concept which was both technically feasible and more economically attractive than other concepts. The paper also discusses how the pipeline network model facilitated the evaluation of flow-assurance issues such as avoidance of hydrates, and maintaining the CO2 stream in dense-phase flow throughout the injection system.

Novel aspects included the ability to model the system using a ‘life-of-field’ approach that allowed the conceptual designers to evaluate how increasing reservoir pressures through time affect the system design (in particular, the injection compression system), and also ensure that the flow-assurance issues were evaluated comprehensively throughout field life, rather than for an unmethodical selection of snapshots in time. Additional functionality of the modelling tool is discussed, such as the ability to incorporate user-configurable logic, benefits of which include allowing the user to control bringing on and abandoning of wells through the ‘life-of-field’ simulation, and also enabling generation of results in terms of economic parameters.

by Dr Hannah Chalmers, Roger Watson, and Prof. Jon Gibbins, School of Engineering, University of Edinburgh, Edinburgh, UK

by Thomas A Demetriades and Dr Trevor C Drage, Efficient Fossil Energy Technologies Centre, University of Nottingham, UK and Richard S Graham, School of Mathematics, University of Nottingham, UK

Abstract to follow

Leak and dispersion testing and modelling

by Dr Andrew Cosham1, David G Jones2, Keith Armstrong3, Dan Allason3, and Julian Barnett4

1 Atkins, Newcastle upon Tyne, UK
2 Pipeline Integrity Engineers, Newcastle upon Tyne, UK
3 GL Noble Denton, Spadeadam Test Site, UK
4 National Grid Carbon, Warwick, UK

A gas or liquid expanding through a leak in a pipeline will cool: this is effect is commonly referred to as Joule-Thomson cooling.

The low temperatures that occur in the expanding fluid when natural gas or CO2 escapes through a leak in a pipeline has led to the suggestion that localized cooling of the material surrounding the leak will occur and that this cooling is sufficient to significantly reduce the toughness of the steel, resulting in a brittle fracture. The implication of this postulated mechanism is that a leak (an otherwise stable through-wall defect) in a pipeline might become unstable, resulting in a rupture.

There is no operational evidence to support this postulated failure mechanism, despite considerable experience of leaks from operational natural gas pipelines around the world. Nevertheless, it remains a contentious issue. The perception is that the cooling that will occur in a leak in a pipeline transporting dense-phase CO2 will be greater than that in a leak in a typical natural gas pipeline. Therefore the issue has gained renewed interest in the context of the transportation of CO2 in pipelines.

There are two questions that need to be answered in order to determine whether or not the postulated failure mechanism is credible: (1) what is the temperature drop in the pipe wall surrounding the leak, and (2) what are the implications of this temperature drop?

A large research and development programme has been initiated by National Grid Carbon to determine the feasibility of transporting dense-phase CO2 in pipelines in the UK. As part of this programme, National Grid Carbon has undertaken several studies to answer these questions.

Four full-scale tests have been conducted to measure the temperature of the pipe wall around a 3-mm diameter hole, a 6-mm diameter hole, a 0.5 x 47-mm machined slit, and an (approximately) 150-mm long fatigue crack during a steady-state release of dense-phase CO2 from a 914-mm diameter, 25.4-mm wall thickness pipe buried in the ground. The pipe was instrumented with an array of thermocouples on the internal and external surfaces of the pipe wall. The surrounding soil was also instrumented with thermocouples.

The experimental set-up is described, and the results of the four tests are presented and discussed.

The minimum internal surface temperature in the four tests was in the range 0 to -30°C (approximately), and the external surface temperature in the four tests was in the range -10 to -80°C (approximately). The cooling of the pipe wall would appear to be primarily caused by direct contact with the bulb of frozen ground that develops around the leak, and not by the fluid expanding through the leak.

by Dr Janice A Lake and Dr Karon Smith, The University of Nottingham, Nottingham, UK

The linked projects of COOLTRANS (Dense Phase CO2 PipeLine TRANSportation) and RISCS (Research into Impacts and Safety in CO2 Storage) aim to provide the CCS community with data on CO2 distribution within the soil and plant-canopy environments with simultaneous measurement of crop performance, health, and ultimately crop yield resulting from an hypothetical pipeline leakage. Field studies attained a mean soil-gas concentration of ~40% CO2 in gassed plots, contrasting with control plots at ~1.1% over a full growing season period. A significant impact on plant growth was detected under elevated levels of soil CO2 with a reduction in plant biomass and yield by ~30% in most crops in areas of soil CO2 concentration above 20%.

Additional complimentary short-term experiments to investigate these effects in different soil types (organic, loam, sandy, and limed) were carried out in controlled-environment growth facilities to standardize all other environmental variables. As CO2 competes by volume with O2 (a necessary component for healthy plant performance), these experiments allowed for evaluation of CO2-specific effects by the incorporation of N2-gassed O2-depleted controls. CO2 levels attained the same levels as field trials (~40%) with similar plant responses, validating the laboratory-based experiments as comparable to field conditions. Biomass was reduced by >40% in CO2-gassed and between 10 and 20% in N2-gassed crops. Different soil types elicited different magnitudes of response in each crop.

Here we present a joint report on vegetation effects of high CO2 concentrations in the soil environment and the underlying mechanisms involved.

by Philip Cleaver, Ann Halford and Karen Warhurst, GL Noble Denton, Loughborough, UK, and Julian Barnett, National Grid Carbon, Warwick, UK

National Grid is funding research, referred to as the COOLTRANS programme, to address gaps in knowledge on the safe design and operation of onshore pipelines for transporting anthropogenic, high-pressure, dense-phase CO2 from large industrial emitters to storage. UK safety legislation requires that the risks associated with high-pressure pipelines are ‘as low as reasonably practicable’ (ALARP), and demonstration of this generally requires compliance with recognized pipeline codes. However, current pipeline codes do not apply to dense-phase CO2. The aim of the COOLTRANS research programme is to provide data and information as technical justification for the operation of dense-phase CO2 pipelines, and the development, scrutiny, and publication of code requirements, which National Grid considers essential to ensure the safe design and operation of onshore dense-phase CO2 pipelines in the UK.

Part of the COOLTRANS research work has involved experimental studies of the behaviour produced when a below-ground release of CO2 takes place from pressurized pipework. The experiments have been carried out to simulate the behaviour that would be produced if a buried pipeline were punctured or ruptured. Measurements have been made during the experiments of the conditions within the pipework system from which the release was taking place and of the concentration in the resulting dispersing CO2 cloud. In general, the releases took place from a section of pipework that was buried in soil to an appropriate depth to represent a typical cross-country pipeline. Measurements were taken of the size and shape of any below-ground crater that was produced by these releases. In addition, a smaller number of experiments took place in a pre-formed crater of the same dimensions that had been observed in an earlier experiment. As there was no violent expulsion of soil in these experiments without backfill, measurements of the temperature of the resulting flow could be made in a horizontal plane immediately above the crater.

The purpose of this paper is to explain how the results from these experiments have been used to produce correlations to predict the size of the crater that is formed by this type of release. How this information could be used in deciding on the separation distance required for parallel pipelines is considered. In addition, the typical conditions that were observed in the flow emerging from the crater are noted. Finally, how the behaviour of the resulting CO2 plume is influenced by the wind is discussed and the possibility that the flow may stall over the crater is considered.

Abstract to follow


Organized by:

Clarion Technical Conferences

Tiratsoo Technical

Supported by: Newcastle University    
  PIPE Pipelines International The Journal of Pipeline Engineering