Carbon Capture Corrosion Current Practice 2023

Paper No.
19539 MARCH 19-23, 2023 | DENVER, COLORADO
Downloaded from http://onepetro.org/amppcorr/proceedings-pdf/AMPP23/All-AMPP23/3188479/ampp-2023-19539.pdf by University of Calgary user on 27 September 2023

Current Industry Practice in Carbon Capture Corrosion, and the Knowledge Gap it Produces
Mark McLeod Ivan Gutierrez

OGC Energy OGC Energy
66 Eldon Street 66 Eldon Street
Sheffield, S1 4GT Sheffield, S1 4GT
United Kingdom United Kingdom
Brad Healey Tundzhay Kurtulan

OGC Energy OGC Energy
66 Eldon Street 66 Eldon Street
Sheffield, S1 4GT Sheffield, S1 4GT
United Kingdom United Kingdom
ABSTRACT
Leveraging their background in industry and working on large carbon capture projects in the UK,
Netherlands, Australia, and Canada, the authors examine the methodologies currently used to mitigate
corrosion risks in supercritical CO2 streams saturated with impurities, highlighting areas of conservatism,
and produce the next steps necessary to address the knowledge gap.
The authors will also propose work to address this gap, through an early-stage concept for a mesoscale
experimental project which seeks to categorise the effect impurities have on supercritical CO 2 transport.
The project aims to reduce conservatism and allow greater uptake of carbon capture across industries.
Key words: Supercritical, Dense, Liquid, CO2, corrosion, dehydration, impurities, thermodynamics
INTRODUCTION
Vital work is being carried out across engineering to ensure the net-zero commitments as laid out in the
Paris agreement are met. Due to increased government investment, carbon capture, utilisation, and
storage (CCUS) has become key to achieving these commitments, with some industries only able to
decarbonise through CCUS, such as concrete or fertiliser production. Carbon capture has also moved
away from vertically integrated systems, with single emitters having dedicated downstream transport and
storage sites, to larger systems gathering CO2 with a shared transport and storage infrastructure. This
allows smaller CO2 emitters the potential to tie into these capture networks and make use of the scale of
the project. Although this gives greater savings by utilising economies of scale, it also introduces a
considerable challenge to the network operator.
© 2023 Association for Materials Protection and Performance (AMPP). All rights reserved. No part of this publication may be reproduced, stored in a retrieval system,
or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording, or otherwise) without the prior written permission of AMPP.
Positions and opinions advanced in this work are those of the author(s) and not necessarily those of AMPP. Responsibility for the content of the work lies solely with
the author(s).
1
As highlighted in previous work1,2, it has been proven that dense phase CO 2 when saturated with
impurities poses a much higher corrosion risk than pure CO 2. This clean CO2 has been extensively used
in EOR in the US and the middle east, with little incident3. The formation of acids, and potentially liquids
or solids, has led the industry to take strong stances on the composition of fluid allowed in transport
pipelines. The impurities present will vary from project to project, but impurities such as NO x, SOx, and
H2S, are likely to present as by-products of combustion, and could potentially lead to acid formation in
the fluid stream, likewise, the presence of O 2 and H2O will serve to create the conditions that allow CO2
corrosion or provide the conditions for reactants to form. These common impurities have so far been
present in each fluid composition.

These composition guidelines take the form of impurity limits for major impurities expected to be found
in the CO2 stream. It is currently best practice to place the responsibility on emitters to ensure their CO2
is within these limits, either through scrubbing or by changing separation technology to costly solvents
which leave fewer impurities. These added costs and strict limitations reduce the likelihood of smaller
asset owners from implementing carbon capture and joining the sequestration network.
This paper will present the approach and methodology taken by engineering teams when attempting to
perform corrosion assessments and material selection philosophies for the qualification of pipelines for
transporting CO2 from industrial, multi-hub CCUS. The variability, project to project, of the CO 2 stream,
due to the range of industries, capture methods, and other process controls, challenges any existing
methodology. A novel approach is required and the steps presented here have been common to projects
in the UK, Netherlands, Australia, and Canada.
Experience within these projects has been used to comment on the state of the industry, and specifically
how a lack of understanding of the specific interactions between carbon dioxide and the impurities
introduced from industrial CCUS, as early identified in academia, have translated over to industry. From
interactions with network operators, engineers, academics, and stakeholders, this paper presents a better
understanding of transport networks’ challenges and how it is necessary to foster communication and
collaboration between researchers and engineers as they tackle these roadblocks.
Another challenge in CCUS is the drive to reuse infrastructure and reduce the need to use new pipelines
for projects. This, combined with the restricted economics of CCUS, demands the use of carbon steel as
a material of construction to enable CCUS projects globally. Pipelines in the North Sea and elsewhere
are for the most part made of carbon steel connecting offshore oil fields to onshore facilities. Reusing
these pipelines for CO2 transport could lead to a large cost reduction in establishing a carbon capture
network. On the other hand, when a new pipeline is required for CCUS, the economics of CCUS proscribe
the use of CRAs.
Work is underway to determine the suitability of existing oil and gas wells for CO2 injection4 as required
by the regulator. This paper will look at the corrosion risks of new and existing carbon steel pipelines
transporting CO2 laden with impurities typical of industrial hub CCUS networks, how these challenges
have been approached in industry, and the need for further study to mitigate these challenges.
RISK MANAGEMENT AND MITIGATION
The engineering process for carbon capture projects undergoes a similar gate process to that of other
energy projects, with field planning and concept development leading to FEED, detailed design followed
by EPC and eventually injection of CO2. During the concept development and pre-FEED stages it is
typical that in-depth work is performed to detail the risk caused by predictable factors such as corrosion,
pipeline fracture, and changes to production. These factors are then accounted for and mitigated where
possible and avoided in design otherwise.
Impurities
the author(s).
2
Effect of impurities on the damage mechanisms
An approach during early stages of a CCUS project to mitigate corrosion is attempting to quantify the risk
of certain impurities damaging the pipeline. The limited data for CO2 transport with impurities makes this
first step a challenging one. This first step requires the identification of credible damage mechanisms for
pipelines, according to the impurities present in the fluid stream. The first point of call for this exercise is
to look at the existing experience in other energy industries. Damage mechanisms that are common in
hydrocarbon production have been studied and well-documented, and their findings are relevant for
CCUS, especially data from acid gas disposal and enhanced oil recovery.

CO2 can cause carbonic acid corrosion, once dissolved in water. Carbonic acid lowers the pH and locally
corrodes mild steels5, leading to uniform loss of metal6. Steels in environments containing CO2 can form
protective scales, most of which are FeCO 3, but they often break down. At high pH values, carbonate
films form on metal surfaces, contrastingly, at low pH values, the bare metal is exposed to liquid acids.
The solubility of these films and scales differ with pH and temperature7. While the scale kinetics are not
fully understood, CO2 corrosion is understood well enough for large hydrocarbon production applications.
Furthermore, CO2 transport in gas, liquid and dense phases has been successfully managed for many
years8. Carbonic acid corrosion in CO2 transport has been controlled with adequate dehydration. Several
studies have been dedicated to determining the water content required to eliminate this damage
mechanism9,10. These studies, combined with industrial practice, demonstrate that CO2 can be safely
transported at various pressures with a water content as high as 630 ppmv11. Current dehydration
technologies can achieve dehydration levels down to 30 ppmv12 eliminating the risk of carbonic acid
corrosion during normal operation.
H2S can cause several forms of damage. Corrosion occurs as H 2S acidifies water, reducing the in-situ
pH and catalysing the penetration of atomic hydrogen into the steel, which may increase the risk of
hydrogen embrittlement. Wet H2S mainly causes three types of corrosion, metal loss corrosion, sulphide
stress cracking and stepwise cracking. These H2S related damage mechanisms can result in gradual or
catastrophic degradation, all requiring mitigation 13. Sulphide stress cracking is a form of hydrogen
embrittlement which requires a combination of a hydrogen environment, where atomic hydrogen is able
to enter metal, a susceptible steel, and adequate stress13. Cracking is caused by the dissolution and
diffusion of hydrogen into the steel when subjected to tensile stress and is difficult to monitor and detect
in practice, particularly in a timely manner. Stress-oriented hydrogen-induced cracking produces small
cracks, which are formed approximately perpendicular to the principal stress, residual or applied, in steel.
These form “ladder-like” cracks linking small pre-existing cracks. Typically, these relatively long cracks
will run in the through-thickness direction of a structure. Hydrogen-induced cracking can also be caused
by wet H2S which results in metal blistering as a result of high hydrogen concentration. Cracks tend to
form parallel to the surface and the hoop stress direction. Hydrogen-induced cracking can also occur as
a result of hydrogen entering the steel and collecting at impurities or inclusions within the steel at relatively
low temperatures. Hydrogen is generated as a H 2S corrosion product7.
H2 gas as an impurity can potentially cause hydrogen embrittlement in susceptible steels. This damage
mechanism needs to be considered for metals exposed to an environment with hydrogen gas, where
molecular gas can be adsorbed on an oxide free surface (e.g. a fresh crack with no oxygen in the system)
and dissociate into atomic hydrogen that diffuses through the metal lattice. The exact mechanism of
hydrogen embrittlement in steels is not yet fully understood7. Research has shown that the adsorbed
atomic hydrogen in these scenarios may be too low to be considered a risk 14, however there is limited
experimental evidence to confirm this. CCUS projects need to evaluate scenarios where high hydrogen
gas concentration can occur, for example, migration of volatile gases like hydrogen into the first CO2 gas
bubbles at the bubble point when depressurising from CO2 liquid.
Oxygen can cause corrosion as oxygen is introduced into the solution through the interface or air to the
metal surface during oxygen transportation. Oxygen migration through the main layer of the corrosive
the author(s).
3
medium to the diffusion layer ends on the metal surface, where it takes part in a negative reduction
reaction15. A basic understanding of the oxygen corrosion mechanism shows that the rate is affected by
dissolved O2 concentration and flow rate13. Oxygen is also known to exacerbate other damage
mechanisms in CO2 transport as discussed further in the NOx and SOx section of this paper. Interestingly,
oxygen can interact with other damage mechanisms by reducing its likelihood. As previously mentioned,
for molecular hydrogen to be adsorbed, an oxide free surface is required, this is unlikely to occur in an
environment with oxygen as an impurity.
Certain impurities like amines and carbamates may react and form solids. While the quantities may be
too low for under-deposit corrosion to be a credible threat, nevertheless, these solids need to also be
considered with regards to reservoir injectivity.

Erosion and erosion-corrosion are both possible damage mechanisms in a carbon capture network. The
removal of surface films, usually passive material or mineral scale, leaves the underlying metal exposed
to corrosion7. Where aggressive chemical environments are present, this can lead to exceptionally fast
rates of corrosion noted by its intermittent nature 13. Flow accelerated corrosion may also be present in
piping made of carbon steel. This is caused by chemical dissolution and mass transfer, rather than a
mechanical process16.
There is limited data surrounding the effect of NO x and SOx impurities on dense-phase CO2 pipelines,
experiments have shown corrosion occurring at water concentrations far below the water-solubility of a
water-CO2 system. Particularly high corrosion rates in excess of 5mm/y were reported in the presence of
NO2. Impurities were shown to be consumed in reactions, slowing the corrosion rate 2.
According to early engineering work, other impurities can be expected in the CO2 mixture depending on
the industrial source. HCN and other acids need to be considered as possible impurities. Their effect in
gas and liquid CO2 has not been studied yet. The CCUS industry has a significant challenge validating
the presence of these and other impurities and quantifying the corrosion risks related to them.
Effect of impurities on the thermodynamic behaviour of the fluid
The impurities in the CO2 fluid not only impact the corrosion reactions, but also increase the risk of liquid
dropout within the pipeline that could potentially initiate corrosion. Phase behaviour can be predicted
through the use of an adequate equation of state (EOS), many such models have been developed over
the years, to address certain mixtures of CO2 and impurities17. Recently the Energy Institute published a
guideline for EOS in CCUS18.
It is important to characterise the liquid-vapour phase behaviour of the CCUS stream for several reasons,
particularly when water is present in the CO 2. Even at concentrations below which water causes
corrosion, CO2 can form hydrates19 with other impurities. Equations of state are useful as they can provide
mathematical interrelation between the temperature, pressure, volume and composition of a fluid.
Overall, there is a reasonable degree of accuracy when compared to current experimental data 20,21,
matching the pressure-composition data of many binary and ternary CO 2-gas mixtures, where data for
these is available. This is important to note, as the prediction of these binary and ternary gas mixtures is
relied upon heavily in the prediction of phase behaviour for CCUS streams. Depending on the impurities
in the system, other liquid phase may also be present. TEG and amines have the thermodynamic
potential to form liquid phases that may be rich in water and CO 2 albeit their phase fraction being too low
to provide a wetting effect.
Data validating EOS to measure the solubility of CO2 in TEG and Amine solutions give mixed results 22,23
for the predictive ability of EOS at low TEG concentrations, and further work is likely necessary to address
the uncertainty brought about by this. Utilising the interaction parameters present in literature various
EOS have shown good results with multiple mixtures of CO 224,25. It is important to recognise however,
the author(s).
4
that there is not a single EOS which will map exactly to the gas expected in the CCUS stream. Work
continues to quantify the ideal model, or models, used for the prediction of liquid drop-out within CCS,
and extensive work has been performed utilising the Cubic-Plus-Association equation implemented within
Multiflash17, particularly to determine the suitability of these EOS for CO 2 with a large set of impurities.
Interaction parameters are not available for all potential impurities, though an exhaustive list would not
be possible, as even within CCUS literature the entirety of potential impurities and their effect is still an
unknown.
Further work is still required to validate the current EOS models, experimental validation is required to
fully understand the interaction between more than two impurities.

Despite the known limitations of these predictive tools, these models have enhanced the corrosion
protection methodology employed by several industrial projects. Mapping operating modes, including
transient conditions such as depressurisation, against vapour-liquid diagrams allows for risk management
decisions to be semi-quantitative rather than entirely qualitative, though several areas of risk, including
the possibility of acid formation, will need further work. In discussion with the transport hub operator, a
full operating portfolio is determined, highlighting areas of risk. These are then used in a full failure mode,
effect, and criticality analysis which is fully explored later in this paper.
Process
Alongside controls on impurities, there are certain process controls which can assist in controlling
corrosion risk in CCUS transport. However, these controls can only effectively take place at the point of
capture, leading to variability in the fluid stream during transport. The emitter stream compositions are
typically fixed before the project reaches the materials selection and corrosion management phase of
design. Any changes to the specification for impurity limits poses the risk of reducing emitter uptake due
to modifications in capture technology to remove impurities being potentially economically unfeasible. As
seen in the Netherlands26, an emitter leaving the project can have a huge impact on the viability of the
project overall.
Dehydration
It is necessary to dehydrate CO2 to mitigate the presence of corrosion and the formation of hydrates
within the pipeline. The CO2 capture process used, as well as the composition of the flue gas, can have
a significant effect on the level of water in the CO2 stream. There exist a variety of methods for
dehydration, giving rise to a variability in the water content of each emitter’s CO2 stream.
The most cost-effective measure for dehydrating CO2 from industry CCUS is to use a molecular sieve,
or TEG12, this comes from a study focusing on four main elements, an evaluation of the process used,
an evaluation of the monitoring and management of the process, an evaluation of the process to meet
the various requirements for CO2, and an analysis of future systems. A variety of moisture levels were
considered, ranging from fairly high (550 ppmv) to very low (<10 ppmv), alongside two typical gas flow
rates. While molecular sieves give better performance than TEG, degradation and blockage can occur,
particularly in the presence of O2 and acids.
Although the use of glycols such as TEG is recommended to achieve consistent dehydration of the CCUS
stream, it was determined during a project that further research to quantify risks of liquid dropout in the
pipeline caused by the carryover of glycols into the system interacting with water. The composition of the
potential liquid phase formed is unknown, and could be a highly corrosive water-based aqueous phase,
or a more benign amine, methanol or glycol rich aqueous phase. The addition of TEG lead to the dew
point of the CCUS stream to be increased significantly, increasing the range where an aqueous phase
could occur. It should be noted, however, that a CCUS project in Canada has been operating for several
years with TEG carryover and no detrimental impact has been reported27.
the author(s).
5
As part of system wide monitoring, including the CCUS stream composition, the monitoring of dehydration
systems will play a large role in maintaining the integrity of carbon capture networks. The use of alarms
and on-line analysis should be considered in the concept development stage of a CCUS transport
network.
Operating Modes
During normal operation, the most efficient method of transporting CO 2 is in the dense or supercritical
phase 28. As a result, most pipelines will aim to operate well above 100 bar(a) to maintain this phase
within the fluid. However, gas transport is likely in gathering networks, due to safety concerns surrounding
leaks of CO2 around inhabited areas, and initially CCUS projects are likely to be operated in the gas

phase as injection rates are ramped up. It is unlikely, given the information gathered from thermodynamic
modelling, for impurities to form aqueous phases at this pressure. It is then of the utmost priority to control
transient operating modes where the pressure/temperature profile of the fluid stream is likely to lead to
the formation of a corrosive aqueous phase.
Control of transient operation such as decompression is already necessary to protect against brittle
fracture of pipelines, as pipeline cooling due to the decompression of the stream has the potential to
reduce the material below the ductile-brittle transition temperature, so it is not unexpected for similar
controls to be implemented as necessary for corrosion control.
A variety of operating modes can be seen in Figure 1 below which shows the operating envelopes for
four industry projects, and a typical vapour-liquid equilibrium phase diagram for CO2 with one impurity,
H2O at 250 ppmv. Pressure and temperature scenarios that intersect the two phase region indicates
where a water dominated liquid phase will occur alongside the CO2 dominated gas phase. It should be
noted that the VLE line will alter drastically depending on the impurities present within the CO2 stream,
and their quantities. Project 1 shows the normal operating modes of a CCUS operator, with no transient
modes. Projects 2 and 4 show all expected operating modes, with project 2 avoiding the VLE line by
controlling operating modes. Project 3 represents a simple injection profile, where only a supercritical
stream is expected. Each project will have a unique makeup of operating modes, depending on the
transport properties of the stream.
Each of the lines in the figure below represents the range of operating modes that a project expects to
encounter through normal operation. To avoid the presence of liquid formation within the pipeline it is
important to control the operating modes and minimise the potential for water dominated aqueous
phases. Without more robust equations of state, and further laboratory testing to validate them, any
thermodynamic modelling of CO2 with impurities will not necessarily be an accurate prediction of CO2
stream behaviour, assessment should be taken with caution.
The liquids formed as a result of phase changes could be any component within the stream, depending
on the composition, reacting with other impurities to form acids or solids which can impact the long-term
integrity of the pipeline.
For one operator in the UK, a high hydrogen specification from emitters required a careful approach to
the control of operating modes. It was determined that during decompression there was an appreciable
risk that small amounts of hydrogen-dominated gas would be present at the bubble point, which could
lead to an embrittlement scenario. Extensive work was performed to determine the controls to avoid
transient conditions where high concentrations of H 2 were present, and that pressures where those
concentrations of H2 were particularly high (40%+) be avoided.
Each project will require careful consideration of the stream composition, which influences the potential
that liquid drop out can occur, and how to manage transient conditions to avoid these areas.
the author(s).
6
Figure 1: Operating modes of four industry projects, operating in both liquid/supercritical and
gas phase transportation.
Mitigation Strategies
The energy industry has developed the knowledge and the skills to tackle the challenges of mitigating
the risks present in carbon capture transport. Decades of work in oil and gas has developed engineers’
understanding of the metallurgy, the dangers, and the solutions which keep assets safe. Due to the
emerging nature of multi-emitter CCUS systems, there are no best practices currently available. Work is
being carried out by organisations like AMPP with the initiative 21532 and ISO within WG8. However,
there is still significant work ahead before a CCUS materials selection standard is available.
In the meantime, stakeholders in CCUS projects are cautious of changes to well established
methodologies - the energy industry is used to mitigate corrosion with corrosion allowances measured in
mm (or fractions of an inch). This strategy is not suitable for CCUS projects where corrosion occurs at
much higher corrosion rates in highly localised areas and a different approach is needed. This paper
proposes a method to systematically evaluate the damage mechanisms in a CCUS pipeline.
Failure modes and effects analysis
To carry out a corrosion and stress cracking assessment, it is proposed to use a methodology where
corrosion shall be considered as a damage mechanism where there is general loss of metal due to
chemical reaction and stress cracking is considered a damage mechanism where the combination of
environment and stress result in metal cracking. The assessment is carried out based on EN IEC
60812:201829, failure modes and effects analysis.
the author(s).
7
In a CCUS system it is important to identify areas which may lead to increased risk. As mentioned
previously, the pressure drop associated with transient conditions can lead to rapid cooling of the pipeline
and very low temperatures. These low temperatures can lead to an increased risk of brittle failure. Also
a concern in CCUS systems, though not specifically related to corrosion and with pipeline material
typically fixed in projects, is the potential for running ductile fracture of carbon steel30,31. Some guidance
is present in ISO 27913 for mitigating this risk 32, however it has been shown that the Battelle two curve
method (BTCM), even with correction factors, is “grossly non-conservative”33. A more accurate
representation could be obtained from combination fluid-static finite element simulation, and a fitness for
service assessment performed for the pipeline selected for use.
The stream composition has a great effect on the failure modes assessed in the pipeline, and worst case-

scenarios (liquid dropout, corrosion, cracking, etc), from analysis of damage mechanisms, are mapped
onto phase diagrams to determine the specific areas of risk during operation. It is then prudent to divide
the system into sub-systems and identify operating modes (e.g. standard operation, venting,
maintenance, temporary or long shut down). These subsystems and operating modes are determined
collaboratively with the client, at the beginning of the project. These are mapped against the phase
diagrams produced from the worst-case scenarios to identify operating modes which may interact with
liquid forming phases.
Synergistic interactions between two or more chemical components can be evaluated by identifying
damage mechanisms that can be exacerbated through the presence of two or more impurities. An
example of this is stress cracking, where oxidizing agents can cause accelerated breakdown of the iron
scale on the surface of steel, this is not typically seen in oil and gas where oxygen is unlikely to exist as
an impurity. Also of concern is the possibility of reactants forming in the CCUS stream, including strong
acids or elemental sulphur34.
Even with well understood stream composition and operating modes, upsets, either at the emitter or
within the fluid processing, can lead to unexpected high-risk operating conditions. Some consideration
must be given to the possibility of out of specification fluid entering the system, with failsafe modes
preventing the emitter from adding to the bulk stream present which can ensure pipeline integrity.
A risk matrix can be then completed, with each damage mechanism individually evaluated, first to
determine if a quantitative method of available to determine risk (such as corrosion rate calculation, stress
corrosion thresholds based on partial pressures, or the temperature of the fluid). Where such a
quantitative method does exist, the risk can be calculated accordingly. Where no quantitative method is
reliable, a qualitative assessment is carried out, at the agreement of the client and following whichever
standards/regulations are available. While it can be difficult to qualitatively assess damage mechanisms
with little data, utilising what literature is available to provide a well-reasoned case is necessary to ensure
some concern is given to these degradation modes.
The produced risk matrix is evaluated to identify areas of higher risk, and mitigating strategies are
proposed. Examples of potential mitigating measures to reduce the risk of a specific damage mechanism
are altering the operating modes, adjusting CO2 specifications, or altering the fluid processing to produce
a drier stream. A similar methodology is undertaken for determining the suitability of wells, valves, and
other hardware along the pipeline.
Gaps present in methodology
The work already undertaken in the study of carbon capture has highlighted several areas where further
study is necessary. From understanding the interactions of two or more impurities on the corrosive
behaviour of the fluid1, to the experimental validation of thermodynamic modelling to verify the prediction
of water or liquid phases in the pipeline 17. Further work and cooperation are necessary to bridge this
knowledge gap. In order to account for this lack of information in the methodology, typical
recommendations and mitigations can be over conservative or costly.
the author(s).
8
The specifications in industry are set on a variety of different factors. Polar impurities with low molecular
weight can have a significant effect on the two-phase region of the CCUS stream, necessitating the
limiting of H2, N2, CO and Ar, controlling these impurities also mitigates the risk of ductile running fracture.
Other impurities are limited due to concerns with reservoir integrity such as O 2 and SOx. Except for limits
set as a result of toxicity relative to CO2, the concentration of the impurities within the CCUS stream could
see relaxation following further work into the behaviour of these impurities.
Reservoir testing could allow the limit of O2 and SOx to be increased, as understanding the impact these
impurities have on the propagation of sulphate reducing bacteria, and the production of H2S leading to
injection blockage. For NOx and SOx limits could be revisited following analysis which includes the volume

of reactant products produced, and kinetic of reactions, to quantify the impact each has on the formation
of strong acids which can corrode carbon steel. Ammonia (NH3) may produce solids which can impact
injectivity, and studies which quantify this could relax limits.
Specifications seen in the industry at present are exceptionally limiting, controlling the presence of
impurities such as NOx to the ppm level, and the difficulty at measuring these impurities has been
discussed elsewhere34. Whilst these limits are set based on the most accurate knowledge to date, the
possibility of upsets, the limiting nature of expensive CAPEX to remove impurities from industrial origin
CO2, and the uncertainty that corrosion controls are effective are high hurdles for new projects to clear.
Further work to identify the most damaging impurity interactions, the conditions for the production of
reactants, the impact of impurities to cause liquid dropout, and the makeup of these aqueous phases
would bolster confidence from stakeholders and engineers.
SOLVING THE KNOWLEDGE GAP
DECIDE – DEfine CO2 Impurities : Reduce Emissions
As a solution to this knowledge gap, collaborative work with academia and industry has led to a concept
for a novel experimental scheme under the acronym DECIDE, which seeks to quantify the impurity
interactions with the CO2 stream and determine their impact on the integrity of CCUS infrastructure.
Current work relies on static experimentation, which will not accurately capture the erosion-corrosion
expected due to elevated flow rates. Passivating scale and a lack of regeneration of impurities could
allow for poor interpretation of results. The proposed work would act as an in-situ experiment, matching
the conditions seen in transport pipelines as close as reasonably practicable. Work was completed
alongside two universities as part of the Sheffield Innovation Program (SIP) which has served as the
foundation for the work presented here.
The flow apparatus is formed of two distinct areas. A static bottle will serve as a tool for validating and
recreating previous work, and a flow pipe with corrosion coupons and a sacrificial pipe for testing. Gases
will be individually injected into the flow apparatus and allowed to freely mix. Gas will be compressed to
dense phase and either circulated across the sacrificial pipe or piped into the static autoclave. Present
plans are to inject CO2 with five impurities, identified earlier as common within CCUS projects, with
flexibility of design allowing for adding or removing impurity sources.
Of particular interest, and a key part of the DECIDE study, will be to test our hypothesis that during normal
operating conditions in CCUS transport that the risk of corrosion with dehydrated CO2 is minimal, and
that the transient modes, such as depressurisation, are what cause acid drop out and heavy corrosion
as seen in previous work1. To this end, the system has been designed such that depressurisation and
other transient conditions can be produced, with the ability to purge the line with an inert gas such as N2
prior to depressurisation. A full schematic of the flow loop will follow in later work.
the author(s).
9
Flow experiments will mirror field conditions as close as reasonably practicable, with pipeline size, flow
rate, and process conditions determined by those in use for dense CO 2 transport. Coupon selection,
installation, and monitoring will follow AMPP standard practice SP0775.
The next steps for the design of the flow loop are to define the monitoring apparatus and recycling
strategy for the CO2 gas. Work is currently underway to develop strategies which will allow for the effective
monitoring of corrosion, gas composition, and change of impurity makeup. Further work seeks to
determine a feasible footprint and cost of construction for the flow apparatus.
Experiments have been devised such that the parameters which have the greatest effect on the corrosive
nature of the CO2 fluid can be determined. To effectively examine the areas the literature has shown are

necessary to relax conservatism, impurities which can have the greatest effect on the integrity of the
asset will form the first round of testing. H 2O, H2S, NOx, SOx, and O2 will be mixed with bulk CO2 in
differing amounts, as described by a fractional factorial approach. It is expected that the removal of certain
impurities, NOx in particular, will allow the relaxation of several impurities.
It is expected that the results of this experiment will answer some of the questions raised by previous
work, further examining the role impurity interaction plays on the overall risk of corrosion within CCUS.
Working with thermodynamics experts modelling the stream will allow the validation of the predictions
given by the equations of state or highlight areas of non-conformity so that further thermodynamics work
can be completed.
It is the intention of the proposed work that this flow apparatus continue to be in use beyond the life of
the initial DECIDE experiments. To that end, care has been taken to ensure the suitability of the design
to accept and examine exotic impurities not yet studied which may arise as a result of the proliferation of
carbon capture, and quantify the effects, if any. It is hoped that the equipment can be used to help develop
standards for use in the CCUS industry.
Next Steps
As the carbon capture market grows, and new industries seek to reduce their CO 2 footprint through point
capture carbon capture, it will be necessary to better understand the effect of impurities on the stream.
Though early in development, it is the intent of DECIDE that this increase in understanding will lead to
less conservativism within the industry, allowing smaller emitters to link up to carbon transport hubs, while
simultaneously enabling the feasibility of widescale carbon capture by increasing the window of impurities
allowed within the gas.
Further work into CCUS standards will hopefully continue, with the authors supporting the AMPP and
ISO initiatives as these seeks to identify the necessity of new comprehensive standards.
CONCLUSIONS
The work required to design and construct a new CCUS pipeline, or to qualify an existing pipeline as fit
for service in CCUS is both multi-disciplinary and extensive, requiring a knowledge of not only the
effects of CO2 on a pipeline but also the damage mechanisms of several cocktails of gas impurities.
These impurities can have a multitude of effects on the gas, such that a one size fits all solution is not
possible. Previous work35,36 has shown that the development of a single acceptable limit for all
impurities is impractical and likely to exclude several swathes of industry.
Due to the lack of standards in the field it is necessary to approach each carbon capture project with a
wide methodology. As process, thermodynamics and metallurgy are intrinsically linked and vary with
the impurities which are present in each CO2 pipeline, a robust methodology is necessary to look
holistically at the entire CCUS market chain identifying areas of improvement and further study. The
the author(s).
10
authors have proposed an experimental study into these interactions in order to increase the viability of
carbon capture, as well as make infrastructure more accessible to small and medium emitters.
Open cooperation between operators will be necessary to continue the growth of carbon capture. For
several industries, carbon capture is the only path to de-carbonising available, and although some
standards exist as initial guidance, including ISO 2791332 for pipeline transport and ISO 2791437 for
geological storage, an overall lack of best practice for managing corrosion in CO2 transport or “Ready
for CCUS” will only serve to delay their adoption. The development of the AMPP guide 21532 should
expand these recommended practices, it is also understood that ISO workgroup 8 is currently
identifying the need for a material selection standard for CCUS.

ACKNOWLEDGEMENTS
REFERENCES
1. Morland, B. H., Dugstad, A. & Svenningsen, G. Experimental based CO2 transport specification
ensuring material integrity. Int. J. Greenh. Gas Control 103697 (2022)
doi:10.1016/J.IJGGC.2022.103697.
2. Dugstad, A., Halseid, M. & Morland, B. Effect of SO2 and NO2 on corrosion and solid formation
in dense phase CO2 pipelines. Energy Procedia 37, 2877–2887 (2013).
3. Gale, J. & Davison, J. Transmission of CO2-safety and economic considerations. Energy 29,
1319–1328 (2004).
4. Koning, M. et al. Deliverable D1.7 Final Report Project name ERA-NET ACT: Re-using EXisting
Wells for CO2 Storage Operations Work package WP1-Project Management and coordination
Dissemination level Public. https://rex-co2.eu (2022).
5. Lotz, U., van Bodegom, L. & Ouwehand, C. The effect of type of oil or gas condensate on
carbonic acid corrosion. Corrosion 47, 635–645 (1991).
6. Kotwica, D. & Minevski, L. Causes and Prevention of Corrosion in Carbon Steel Natural Gas
Coolers. (1998).
7. Heidersbach, R. Internal Metal Loss Corrosion Threats. in Corrosion and Materials in
Hydrocarbon Production 57–71 (2019).
8. National Energy Technology Laboratory. Carbon Dioxide Enhanced Oil Recovery Untapped
Domestic Energy Supply and Long Term Carbon Storage Solution.
9. Bondor, P. L. Applications of carbon dioxide in enhanced oil recovery. Energy Convers. Manag.
33, 579–586 (1992).
10. Kwak, D. H., Yun, D., Binns, M., Yeo, Y. K. & Kim, J. K. Conceptual process design of CO2
recovery plants for enhanced oil recovery applications. Ind. Eng. Chem. Res. 53, 14385–14396
(2014).
11. Brownsort, P. A. Briefing on carbon dioxide specifications for transport 1st Report of the
Thematic Working Group on: CO 2 transport, storage and networks Release Status: FINAL.
(2019).
12. Kemper, J., Sutherland, L., Watt, J. & Santos, S. Evaluation and Analysis of the Performance of
Dehydration Units for CO2 Capture. Energy Procedia 63, 7568–7584 (2014).
13. Heidersbach, R. Corrosive Environments. in Metallurgy and Corrosion Control in Oil and Gas
Production, Second Edition 13–35 (wiley, 2011). doi:10.1002/9781119252351.
14. Comité Européen de Normalisation. CEN/TR 17797 - Gas infrastructure - Consequences of
hydrogen in the gas infrastructure and identification of related standardisation need in the scope
of CEN/TC 234. (2022).
15. Yan, X., Wang, Y., Du, Q., Jiang, W. & Li, R. Research progress on factors affecting oxygen
corrosion and countermeasures in oilfield development. doi:10.1051/e3sconf/201913101031.
the author(s).
11
16. Kain, V., Roychowdhury, S., Ahmedabadi, P. & Barua, D. K. Flow accelerated corrosion:
Experience from examination of components from nuclear power plants. Eng. Fail. Anal. 18,
2028–2041 (2011).
17. Luna Ortiz, E. CO2cktails in a pipeline’: the phase behaviour of CO2 with>20 impurities. in The
11th Trondheim Conference on Carbon Capture, Transport and Storage 10–18 (SINTEF
Academic Press, 2021).
18. Energy Institute. Review of Equations of State and Available Experiment Data for CCS Fluids.
(2022).
19. de Koeijer, G. et al. CO2 transport–Depressurization, heat transfer and impurities. Energy
Procedia 4, 3008–3015 (2011).
20. Diamantonis, N. I., Boulougouris, G. C., Mansoor, E., Tsangaris, D. M. & Economou, I. G.

Evaluation of Cubic, SAFT, and PC-SAFT Equations of State for the Vapor–Liquid Equilibrium
Modeling of CO2 Mixtures with Other Gases. Ind. Eng. Chem. Res. 52, 3933–3942 (2013).
21. Rowland, D. et al. Reliable prediction of aqueous dew points in CO2 pipelines and new
approaches for control during shut-in. Int. J. Greenh. Gas Control 70, 97–104 (2018).
22. Wise, M. & Chapoy, A. Carbon Dioxide Solubility in Triethylene Glycol and Aqueous Solutions.
Fluid Phase Equilib. (2016) doi:10.1016/j.fluid.2016.03.007.
23. Skylogianni, E., Perinu, C., Cervantes Gameros, B. Y. & Knuutila, H. K. Carbon dioxide solubility
in mixtures of methyldiethanolamine with monoethylene glycol, monoethylene glycol–water,
water and triethylene glycol. J. Chem. Thermodyn. 151, 106176 (2020).
24. Afsharpour, A. & Haghtalab, A. Implementation of electrolyte CPA EoS to model solubility of
CO2 and CO2 + H2S mixtures in aqueous MDEA solutions. Chinese J. Chem. Eng. 27, 1912–
1920 (2019).
25. Dawass, N. et al. Solubility of Carbon Dioxide, Hydrogen Sulfide, Methane, and Nitrogen in
Monoethylene Glycol; Experiments and Molecular Simulation. J. Chem. Eng. Data 66, 524–534
(2021).
26. Gasunie. ⁣Athos-project ends after Tata Steel decision. https://www.gasunie.nl/en/news/athos-
project-ends-after-tata-steel-decision (2021).
27. Quest CCS. Quest CO2 dehydration performance 2021. (2021).
28. Wang, H. et al. A Review of Pipeline Transportation Technology of Carbon Dioxide. IOP Conf.
Ser. Earth Environ. Sci 310, 32033 (2019).
29. BSI. BSI Standards Publication Failure modes and effects analysis (FMEA and FMECA). (2018).
30. Martynov, S. B., Talemi, R. H., Brown, S. & Mahgerefteh, H. Assessment of Fracture
Propagation in Pipelines Transporting Impure CO2 Streams. Energy Procedia 114, 6685–6697
(2017).
31. Biagio, D., Di Biagio, M., Lucci, A., Mecozzi Centro Sviluppo Materiali SpA, E. & Carlo Spinelli
Eni SpA, I. M. FRACTURE PROPAGATION PREVENTION ON CO2 PIPELINES: FULL SCALE
EXPERIMENTAL TESTING AND VERIFICATION APPROACH. Pipeline Technol. Conf. (2017).
32. International Organization for Standardization. ISO 27913 - Pipeline Transportation Systems.
(2016).
33. Barnett, J. & Cooper, R. An Operator’s Perspective on Fracture Control in Dense Phase CO2
Pipelines. in (American Society of Mechanical Engineers Digital Collection, 2016).
doi:10.1115/ipc2016-64466.
34. Morland, B. H., Svenningsen, G. & Dugstad, A. The Challenge of Monitoring Impurity Content of
CO2 Streams. Process. 2021, Vol. 9, Page 570 9, 570 (2021).
35. Luna-Ortiz, E., Yao, C., Barnes, J., Winter, M. & Healey, M. Development of A CO2 Specification
for Industrial CCS Transport Networks: Methodology, Limitations and Opportunities. (2022)
doi:10.4043/32103-ms.
36. Harkin, T., Filby, I., Sick, H., Manderson, D. & Ashton, R. Development of a CO2 Specification
for a CCS Hub Network. Energy Procedia 114, 6708–6720 (2017).
37. International Organization for Standardization. ISO 27914 - Carbon dioxide capture,
transportation and geological storage — Geological storage. (2017).
the author(s).
12

Carbon Capture Corrosion Current Practice 2023

Uploaded by

Copyright:

Available Formats

Carbon Capture Corrosion Current Practice 2023

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Carbon Capture Corrosion Current Practice 2023

Uploaded by

Copyright:

Available Formats

Paper No.

19539 MARCH 19-23, 2023 | DENVER, COLORADO

Downloaded from http://onepetro.org/amppcorr/proceedings-pdf/AMPP23/All-AMPP23/3188479/ampp-2023-19539.pdf by University of Calgary user on 27 September 2023

Mark McLeod Ivan Gutierrez

Brad Healey Tundzhay Kurtulan

Downloaded from http://onepetro.org/amppcorr/proceedings-pdf/AMPP23/All-AMPP23/3188479/ampp-2023-19539.pdf by University of Calgary user on 27 September 2023

RISK MANAGEMENT AND MITIGATION

Downloaded from http://onepetro.org/amppcorr/proceedings-pdf/AMPP23/All-AMPP23/3188479/ampp-2023-19539.pdf by University of Calgary user on 27 September 2023

Downloaded from http://onepetro.org/amppcorr/proceedings-pdf/AMPP23/All-AMPP23/3188479/ampp-2023-19539.pdf by University of Calgary user on 27 September 2023

Effect of impurities on the thermodynamic behaviour of the fluid

Downloaded from http://onepetro.org/amppcorr/proceedings-pdf/AMPP23/All-AMPP23/3188479/ampp-2023-19539.pdf by University of Calgary user on 27 September 2023

Downloaded from http://onepetro.org/amppcorr/proceedings-pdf/AMPP23/All-AMPP23/3188479/ampp-2023-19539.pdf by University of Calgary user on 27 September 2023

Failure modes and effects analysis

Downloaded from http://onepetro.org/amppcorr/proceedings-pdf/AMPP23/All-AMPP23/3188479/ampp-2023-19539.pdf by University of Calgary user on 27 September 2023

Gaps present in methodology

Downloaded from http://onepetro.org/amppcorr/proceedings-pdf/AMPP23/All-AMPP23/3188479/ampp-2023-19539.pdf by University of Calgary user on 27 September 2023

SOLVING THE KNOWLEDGE GAP

DECIDE – DEfine CO2 Impurities : Reduce Emissions

Downloaded from http://onepetro.org/amppcorr/proceedings-pdf/AMPP23/All-AMPP23/3188479/ampp-2023-19539.pdf by University of Calgary user on 27 September 2023

Downloaded from http://onepetro.org/amppcorr/proceedings-pdf/AMPP23/All-AMPP23/3188479/ampp-2023-19539.pdf by University of Calgary user on 27 September 2023

Downloaded from http://onepetro.org/amppcorr/proceedings-pdf/AMPP23/All-AMPP23/3188479/ampp-2023-19539.pdf by University of Calgary user on 27 September 2023

You might also like