An Interoperable CO2 Transport Network Towards Specifications for the Transport of Impure CO2.pdf

Ref. Ares(2023)6340073 - 19/09/2023
ANNEX 1:  Interoperable  CO2  Transport  Network  –  Towards  Specifications  for  the  Transport  of
Impure CO2’ CCUS Forum Expert Group on CO2 Specifications
Report of the CCUS Forum Expert Group on CO2 Specifications

‘An Interoperable CO2 Transport Network –
Towards Specifications for the Transport of Impure CO2’

September 2023

We would like to express our gratitude to all members and contributors of the CCUS Forum expert
group. We would particularly like to thank the three co-chairs Roland Span (Ruhr-Universität
Bochum), Andy Brown (Progressive Energy), and Harald Tlatlik (Wintershall Dea) for their
commitment to the work.

1

Executive Summary
The report of the CCUS Forum expert group on CO2 specifications, supported by the Zero Emissions
Platform (ZEP), complements the report on CO2 infrastructure and provides a common understanding
and clear recommendations regarding CO2 transport specifications.

The expert group makes the following recommendations to the European Commission:
•  Develop as rapidly as possible a network code and standards for a multimodal CO2 transport
network  in  the  EU/EEA.  Determining  these  standards  and  a  network  code  will  require  the
development of scenarios and an analysis of specific considerations for different transport modes,
based on fundamental assumptions on the future European CO2 transport network.

•  Develop  a  strategy  and  clear  targets  for  a  common  European  CO2  transport  network.  This
strategy should comprise key components, including a fit-for-purpose regulatory framework for
CO2 transport with non-discriminatory, open-access, multimodal CO2 transport infrastructure (see
report on CO2 infrastructure). Targets should cover the future size of the network by development
stage, the role of each transport mode, and its precise geographical span.

•  Support  and  prioritise  research  in  the  identified  fields.  In  addition  to  providing  advice  on
additional  policy  actions,  the  report  raises  several  questions  for  which  further  research  is
requested. This includes, in a non-exhaustive way:
o  potential interactions between several different compounds in the CO2 stream (for example
the stabilisation of aqueous phases by polar molecules);
o  prediction of phase transitions and the accompanying partitioning of the impurities into the
new  phases  and  non-equilibrium  phase  transition  conditions,  in  the  context  of  multi-
component systems;
o  potential chemical reactions of impurities within the CO2 stream;
o  fundamental research on chemical reactions paths and kinetics;
o  prediction of the mass flow and the composition in two-phase flows;
o  dynamics of a pipeline network for CO2 containing impurities for dense phase transport;
o  the impact of extended two-phase regions on operational procedures for pipeline transport
in the dense phase;
o  challenges  resulting  from  solid  phase  formation  and  operational  procedures  for  pipeline
pressures above about 2 MPa for transport in gaseous states;
o  energy-efficient technologies for heating up CO2 streams;
o  practical limits regarding the flexibility of mass flows into storage reservoirs;
o  risks linked to chemical reactions that could result from the mixing of different impurities from
different CO2 streams; and
o  the design of facilities to process the landed CO2 that could be captured from the exhaust gas
of ships.

Safe transport  of impure CO2  streams is possible today.  Wherever fundamental understanding of
processes  is  incomplete,  the  design  of  transport  networks  follows  common  engineering  practice:
safety factors are applied to make sure that any design is on the safe side. The questions raised in this
report aim at an improved design of next generation transport networks, which can be designed more
economically if a better understanding of fundamental effects allows for less conservative margins.
Improved theoretical understanding alone does not result in better transport networks – it has to
go together with experience from practical implementation, which has to start now!
3

1.  Introduction
a)  Background
As  part  of  the  CCUS  Forum,  the  European  Commission  has  set  up  a  working  group  on  CO₂
infrastructure  to  provide  clear  recommendations  on  how  to  develop  and  deploy  sustainably  a
European  CO₂  transport  and  storage  infrastructure  grid  to  reach  climate  neutrality  by  2050.  The
European  Commission  has  invited  the  Zero  Emissions  Platform  (ZEP)  to  support  and  coordinate  a
working  group  of  experts  to  complement  this  work  and  provide  clear  recommendations  on
specifications for CO₂ stream transport1. The medium to long-term goal of the European Commission
is to establish a Europe-wide network for CO₂ stream transport, which utilises all transport modes that
are  required  for  industry  decarbonisation,  and  which  helps  establish  a  European  CO₂  market.  It  is
intended for emitters of different sizes to have undiscriminated access to this network. While point-
to-point solutions directly connecting large emitters with sinks can define their own specifications for
the transported CO₂  stream (with regards to pressure, temperature,  impurities…) in accordance to
existing national and international standards for CO2 stream transportation, a multimodal European
transport network requires an agreed international network code, which specifies characteristics for
CO2 inserted into different transport modes. Storage requirements have to be considered in case they
exceed the quality requirements of the transportation infrastructure, e.g., with regard to allowable
impurities. Finally, the goal is to deliver, as closely as possible, economically optimal solutions that
consider the whole chain, from capture to storage.

b)  Objectives
The objectives of this report and of the expert group supported by ZEP is not to establish standards
competing  with  standards  set  up  by  standardisation  bodies  like  the  International  Organization  for
Standardization (ISO) or the Comité Européen de Normalisation (CEN-CENELEC). The goal is to provide
recommendations  on  the  lines  along  which  characteristics  of  CO2  streams  can  be  regulated  for
multimodal  transport  and  to  summarise  a knowledge  base  on  which  stakeholders  can  negotiate a
network code. It aims at providing high-level guidance for some research activities required to close
identified  knowledge  gaps.  The  objective  is  to  arrive  at  specifications  that  are  based  on  verified
knowledge.  Wherever  science  cannot  yet  provide  traceable  and  sufficiently  accurate  answers,
specifications will have to be stricter than might be necessary, so that safety is always ensured. Costs
to comply with these specifications will, as a consequence, be higher than perhaps would otherwise
be necessary.
As guiding principles, the expert group has agreed on three statements:
•  two-phase flow should be avoided as much as is practicable (If not everywhere and under all
operational conditions, then at least at measuring points because, using current technology,
it is not possible to predict the mass flow and in particular the composition in two-phase flows
with confidence. More research work is recommended);
•  the formation of corrosive phases must be avoided;
•  the concentration of all impurities in a CO₂ stream should be specified to be such that their
health and safety impact is always less than that of the carbon dioxide itself.

1 ZEP is the technical adviser to the European Commission on the deployment of CCS and CCU – a European
Technology and Innovation Platform (ETIP) under the Commission’s Strategic Energy Technologies Plan (SET-
Plan).
4

c)  Content of this report
Chapter 2 of this report will summarise the relevant assumptions made for the development of the
European  CO2  grid.  Interfaces  between  the  three  principal  process  steps  (capture,  transport  and
storage) will be discussed. Chapter 3 introduces a new distinction that resulted from discussions within
the  working  group:  the  distinction  between  “negotiable  impurities”  (for  which  allowable
concentrations can be optimised in a way that considers the cost minimisation for the whole chain;
primarily  these  are  “non-condensable  gases“  such  as  nitrogen,  argon,  hydrogen  or  methane)  and
“non-negotiable  impurities”  (for  which  excessive  concentrations  of  some  impurities  would  create
health  or  safety  issues  or  could  potentially  damage  some  of  the  hardware  associated  with  high
investment costs). Chapter 4 summarises specific considerations for the different transport modes,
including  buffer  storage,  the  infrastructure  required  in  ports  to  process  CO2  captured  on  board of
ships, and constraints resulting from injection and storage requirements. Chapter 5 summarises the
resulting  conclusions  and  recommendations  on  specifications  for  CO₂  stream  transport  in  the
European  Economic  Area  (EEA).  The  report  is  complemented  by  an  executive  summary  for
policymakers.

5

2.  Assumptions and nomenclature
a)  Assumptions on the development of the European CO₂ grid
In the European Economic Area (EEA) the need to transport large amounts of CO2 is currently being
driven  by  individual  projects  in  some  member  states  linking  industrial  clusters  to  storage  sites  via
pipeline or ship transport (e.g. Porthos, Aramis, Northern Lights, Greensand…). Consequently, these
projects  have  published  or  announced  their  own  minimum  CO2  stream  quality  specifications,
reflecting the requirements of the individual storage site or transport medium. These specifications
are  mostly  labelled  as  ‘work  in  progress’  and  are  partly  conservative  with  respect  to  pipeline  and
reservoir integrity, reflecting knowledge gaps. On the other hand, in terms of CO2 stream purity the
requirements are usually specific, reflecting the internationally established minimum of 95 mole-%,
with individual constraints on impurities.  Lower requirements would presumably be aiming to unlock
more sources of CO2 by avoiding purification except where necessary for a specific technical basis.

If  the  EU  wants  to  reach  its  climate  goal  of  net-zero  emissions  by  2050  and  become  net-negative
afterwards,  a  rapid  increase  in  storage  and  transport  volumes  is  going  to  be  necessary.  The  first
objective is to reach at least 50 million tonnes of annual injection capacity in 2030 under the European
Commission  proposal  for  a  regulation  ‘Net-Zero  Industry  Act’.  While  it  is  beyond  the  scope  of  this
document  to  make  recommendations  on  infrastructure  planning,  some  conditions  related  to  CO2
stream quality can be identified to ensure that transport capacity is not the bottleneck in the progress
towards permanent CO₂ storage:
1.  Access  to  infrastructure  should  be  transparent  and  non-discriminatory.  This  is  already
stipulated under Directive 2009/31/EC.
2.  Intermodality:  Emission  sources  with  no  pipeline  access  will  have  to  apply  non-pipeline
transport (NPT) solutions such as ships, barges, rail cars or trucks. These modes transport CO₂
streams  discontinuously  in  liquid  phase,  usually  at  temperatures  significantly  lower  than
ambient.  Considering  the  possible  pipeline  transportation  modes  “gas  phase”  and  “dense
phase”, it might lead to two different pipeline sectors (emission collection and trunk lines)
with two different CO2 stream specifications. Hybrid solutions combining pipelines and NPT
might also be considered.
3.  Extension: Future interconnections between early pipeline systems connected to individual
storage sites or shipping terminals are likely to improve economic efficiency as well as access.
4.  Treatment: Overall costs along the value chain are expected to decrease when the optimal
locations for CO₂ stream treatment are found.
5.  Interoperability: Transport systems should be technically interoperable across borders.
6.  Economic  efficiency  will  be  severely  compromised  if  CO2  streams  from  different  sources
cannot be blended as required without violating pipeline integrity criteria, for instance, due
to inter-impurity chemical reactions.
7.  Pipeline Operations: While designing pipeline infrastructure, consideration should be given to
bidirectionality and diverse routes. Shut-ins may be necessary during e. g. maintenance and
will temporarily impact transport capacity.
8.  Climate Change:  Rising ambient temperatures might affect all transport modes, i.e. pipeline
and NPT. For example, maximum ground temperatures may impact transport of CO₂ streams
in onshore pipelines, river barge transport capacity could significantly diminish during summer
due  to low water  levels or the maximum travelling time of rail cars may be limited due to
increased ambient temperatures raising the temperature of the CO₂ cargo.
6

9.  Distances: Infrastructure is likely to be built first in Member States bordering the North Sea.
CO2  streams  from  Southern  European  and  Central  and  Eastern  Europe  (CEE)  countries will
probably be stored separately or connected to existing infrastructure only at a later stage. Any
future legal framework must accommodate transit of CO2 through Member States.
10. Predictability:  Lastly,  the  creation  of  CCUS  value  chains  constitutes  a  “chicken  and  egg”
situation. Investments in capture, transport and storage facilities will only be triggered if the
other parts of the chain can be reasonably assumed by investors to be in place in time.

Under these assumptions the development of transport infrastructure will be impacted by CO2 stream
composition in several ways:
1.  Predictability: The need to provide a given CO2 stream quality can significantly impact both
capital expenditure (CAPEX) and operational expenditure (OPEX) for CO2 capture. Treatment
is  characterised  by  diminishing  returns.  Quantitative  optimisation of  the  total  costs  of  CCS
would require finding a CO2 stream purity that is not too high. Setting a purity requirement
for transport infrastructure  will ease  risk management  for today’s emitters when weighing
mitigation options. This still holds if the purity is not optimal, e.g. because the data used was
necessarily incomplete.
2.  Access:  Among  other  things,  non-discriminatory  access  also  means  that  appropriate
specifications should be set to be independent of the CO2 stream source.
3.  Intermodality:  Batch  transport  of  liquid  CO2  at  low  temperature  has  more  stringent
requirements in terms of CO2 purity and water content than those for pipelines. Even then
some trace components may still compromise reservoir integrity.
4.  Extension:  Bidirectional  interconnections  without  treatment  will  only  be  possible  if  CO2
stream quality is harmonised.
5.  Treatment: Higher CO₂ stream purity is always beneficial for transport because it could allow
a  larger  operational  envelope,  thus  reducing  risks  associated  with  pipeline  operational
integrity.  Higher CO₂ stream purity can also enable lower pressures for batch transport (e.g.
by  ship).  However,  purities  that  might  be  preferred  for  pipelines  and  storage  may  impose
prohibitive costs on the emitters, since purity comes with additional energy requirements and
higher costs. Purification technology possibly implies additional CO2 emissions, which need to
be considered over the whole value chain. There is a possibility that inter-impurity chemical
reactions  in  pipelines  may  create  unwanted  intermediate  products,  e.  g.  water.    The  CO₂
stream  specification must  account  for  this.   Even  when  this  is  the case,  the  possibility  still
exists when the CO₂ streams originate from diverse processes, that inlet specification limits
are exceeded slightly at the exit point due to inter-impurity chemical reactions. Treatments
like further drying of the CO₂ stream might be considered as an alternative to the introduction
of separate limits for entries and exits but maintaining the non-corrosive nature of the CO₂
stream under all normal, transient, and upset scenarios is key.
6.  Interoperability: Harmonised national specifications that are suitable for liquid, dense and gas
phase  would  enable  transport  across  borders  without  technical  facilities  except  for  fiscal
metering of the CO2 mass flow. It would also eliminate the need for treatment facilities  at
intersections involving phase changes (the need for compression/cooling or heating/throttling
notwithstanding).
7.  Economic Efficiency: Will in part be facilitated by source-independent specifications.
7

o  On the  flipside  of the accommodation of  arbitrary blending,  pipeline  integrity  must  not
depend on certain trace components being diluted through blending.
o  This  report  considers  linear  CCS  value  chains  for  the  purpose  of  permanent  geological
storage. The expected gradual phase-in of CCU will be enabled by e. g. non-discriminatory
third-party access. Beyond that no specific requirements resulting from the use of the CO2
as a feedstock will be considered here. It is however likely that CCU applications will require
high purity of CO2 as well as the absence of catalyst poisons.  Higher quality of CO₂ streams
will therefore benefit usage by decreasing the need for treatment after transport, but it
must be understood that this will come at significant cost to the emitting agency. Demands
on CO2 purity for catalyst-based utilisation processes are very specific; it is considered very
likely  that  a  further  CO2  purification  optimised  for  the  specific  utilisation  process  is
unavoidable.
8.  Pipeline Operations: When designing pipeline infrastructure, consideration should be given to
bidirectionality. Flow assurance is an issue and needs to be considered (see ISO/TR 27925),
especially during shut-in operations, which may be necessary, e.g. during maintenance, and
will temporarily impact transport capacity because the temperature and pressure of the CO2
stream can assume any ground or water temperature. The composition of the CO₂ stream will
determine the maximum saturation pressure within that temperature interval. The saturation
pressure,  in  turn,  will  determine  the  pipeline  wall  thickness  required  to  rule  out  Running
Ductile Fractures (RDF) in dense phase pipelines. Higher CO2 stream purity will result in lower
pipeline  wall  thicknesses  and,  potentially,  significantly  reduced  CapEx,  especially  if  the
maximum temperature is low.
9.  Climate Change: Rising ambient temperatures might affect all transport modes, i.e. pipeline
and NPT, (see above item 9). Elevated ground temperatures will result in elevated saturation
pressures, requiring thicker pipeline walls, in particular when considering impurities and the
effect on saturation pressures.
10. Distances:  For  long  distances  dense  phase  pipelines  will  usually  be  economically  more
advantageous than gas phase transport. The minimum pressure in dense phase is lower if CO2
stream purity is higher, and allowable pressure drop is therefore higher.

b)  Interfaces between capture, transport and storage
It is generally accepted that the purity requirements for the initial transport mode are most effectively
met at the capture plant. Additional purity requirements resulting from transport modes further down
the transport chain (e.g. ships) may either be met at the capture plant or by post-processing in local
hubs before the transport modes are changed.

Most operational CCS systems and mature projects are isolated value chains in which specifications
are tailored to the planned transportation mode and stores. An EU-wide connected system for CCS
could introduce several transport modes in series between capture and store and introduce interfaces
in  between  those.  The  full  combination  of  these  transfers  is  summarised  in  the  table  below2  and
discussed  per  receiving  transport  mode  hereafter  (noting  that  water  removal  is  one  of  the
“Purification” steps):

2 Definitions of the terms used in this table can be found under ‘Nomenclature’.
8

To
Gas
phase  Dense  phase  MP shipping
LP shipping
Rail and truck
From
pipeline
pipeline
(14-17.5 bara)
(6.5-8 bara)
Gas phase

Fully
Purification
Purification
Not likely
compatible3
Dense phase
Exceptional

Purification
Purification
Not likely
MP shipping
Not likely
Fully

Unexplored
Fully
compatible
compatible
LP shipping
Not likely
Fully
Unexplored

Fully
compatible
compatible
HP shipping
Unexplored
Unexplored
Unexplored
Unexplored
Unexplored

Compared  to  dense-phase  transport,  CO₂  stream  transport  in  the  gas  phase  is  relatively  energy
inefficient for large-scale transport. It is an option for a collection network within an industrial cluster
but not expected to be the final mode of transport towards the injection site unless in the early stages
of injection into a depleted gas reservoir, or the pipelines need to pass through or close to, dense
population centres. As such the transfer into a gas phase system from the other transportation modes
is classified as exceptional but might be required for integrating of portions of repurposed pipelines.

A pipeline for a dense phase CO₂ stream transportation system that must be depressurised below the
critical pressure for maintenance activities (e.g. valve inspection) will pass through the gas phase state
as  the  CO2  is  boiled  off  inside  the  pipe.  This  rare  upset  condition  is  likely  to  drive  CO2  stream
specifications to be similar to those for the gas phase, as the formation of a separate corrosive liquid
phase during depressurisation is also an undesirable event.

CO₂ stream processing units, which feed into a dense phase pipeline, are typically based either on
compression and cooling, when starting from gas phase, or based on pumping and heating of the CO₂
stream when starting from liquid phase.  If no further processing is intended at the interfaces between
the different transport modes (gas or dense), this implies that the CO₂ stream specification for the gas
and shipping compositions must be the same as for the dense phase pipeline. This includes CO₂ stream
requirements to ensure that crack growth by RDF is avoided, and also to avoid the formation of strong
acids in dense phase CO₂ streams at low temperatures. The minimum pressure in dense phase can be
lower if CO₂ stream purity is higher, and the allowable pressure drop is therefore higher. Compression
from  gas  phase  conditions  to  dense  phase  conditions  becomes  less  energy  intensive  for  low
concentrations of non-condensable gases. In this case, the gaseous CO₂ stream can be compressed to
an intermediate pressure and can be liquefied by heat removal at this pressure level, before an energy
effective pump is used to increase the pressure to the final pipeline pressure. High concentrations of
non-condensable gases increase the dew point pressure and do not allow for the application of pumps

3 In the case where gas-phase transport allows for high concentrations of non-condensables, removal of some
non-condensables might be necessary. To include emissions from small emitters, it may be advantageous to
define demands for gas-phase pipelines with largely unprocessed CO₂, which then has to be processed at
processing hubs downstream.
9

at  pressures  substantially  below  pipeline  pressure  (see  Section  3.b).  However,  since  the  cost  for
additional purification may be very high, the overall economic case should be considered4.

Refrigeration of CO2 streams for shipping requires significantly higher purity levels compared to those
necessary  for  pipeline  operation.  Without  purification5,  impurity  levels  typical  for  various  capture
processes  drive  liquefication  conditions  to  a  combination  of  high  pressure  (25  bar)  and  lower
temperature (< -30 °C)6.
Ships  need  to  receive  liquified  CO2  (LCO2)  at  conditions  around  their  lowest  allowable  operating
pressure to manage pressure build-up during voyage as required by the IGC Code7:
•  Venting of the cargo to maintain cargo tank pressure and temperature shall not be acceptable.
•  The containment system (cargo storage tank on the ship) insulation and design pressure shall
be adequate to provide a reasonable margin for the operating time and temperatures.

To meet the above requirements, a ship or barge suitable for a trade typically of 15 to 21 days is used
for international shipping considering the voyage and the associated weather delays. The present ship
designs do not include any conditioning equipment for managing boil-off gases at medium-pressure
(operating range of roughly 14 bara to 17.5 bara); instead they rely on pressure retention. The same
approach is also adopted in the development of liquid petroleum shipping8.

In addition to agreeing the maximum impurity levels for specific components to avoid corrosion, the
level  of  inerts  (non-condensable  gases)  acceptable  for  transfer  into  a  ship  may  be  specified  on  a
combination of operating pressures  and temperatures  (the  latter being the boiling temperature at
that pressure) and accept the resulting impurity-levels of the inerts within the CO₂ stream. The transfer
of refrigerated CO2 into ships results in the reverse transfer of vapour returns. Ship pressure control
during loading results in vapour return with a higher volume than that of liquid transferred, thereby
creating a demand on the export terminal for a boil-off gas (BOG) system.  The impurity levels of low
boiling components such as H2, CO, NO and H2S will be high in the vapour return gas, but predicting
the  actual composition in practice  is difficult. Ship vapour return can introduce  impurities  into the
liquefaction  unit  that  handles  the  boil-off  gas,  which  are  unexpected  and  there  is  a  risk  that  the
accumulated levels might drive the LCO2 off spec. Specifications for liquid phase transport need to
consider this possible enrichment effect; further research and some practical experience will help to
avoid overly strict limitations.

4 Conversion from liquid phase at low temperature to dense phase at ambient temperature requires a
substantial amount of heat. For the environmental assessment of the process, it is important that this heat is
supplied in an efficient way.
5 Enrichment of low boiling impurities, such as water and SOx, in the remaining dense phase has to be
considered. The corresponding effects are not completely understood yet
6 Engel and Kather, 2018. Example provided by Figure 2 in Improvements on the liquefaction of a pipeline CO2
stream for ship transport, 2018, https://doi.org/10.1016/j.ijggc.2018.03.010.
7 International Code for the Construction and Equipment of Ships Carrying Liquefied Gases in Bulk (IGC Code),
International Maritime Organisation.
8 Notaro, Gabriele and Belgaroui, Jed and Maråk, Knut and Tverrå, Roe and Burthom, Steve and Sørhaug, Erik
Mathias, 2022. Ceto: Technology Qualification of Low-Pressure CO2 Ship Transport, Proceedings of the 16th
Greenhouse Gas Control Technologies Conference (GHGT-16), http://dx.doi.org/10.2139/ssrn.4272083.
10

The combined liquefaction and purification will increase costs but will also create a side stream and
lower  the  CO2  recovery  in  the  liquid  product.  CO2  recoveries  of  96-98%  have  been  calculated  in
simulation  studies9.  Moreover,  the  side  stream  has  a  significant  CO2  content  and  will  most  likely
contain low-boiling components such as CO, CH₄, and NO that cannot be simply vented. Integration of
the purification unit with the capture unit and the manufacturing complex can provide an outlet for
this  stream.  The  presence  of  an  outlet  for  this  stream  at  the  harbour  is  key,  and  concepts  for  its
effective management need to be developed.

Under the EU ETS directive each step in the CCS value chain must be monitored using a form of mass
balance  approach,  where  some  of  the  CO2  stream  entering  or  leaving  the  installation  (i.e.  at  the
custody  transfer  points)  is  monitored  using  continuous  measurement  systems10.  The  loss  of  a  few
percent of CO2 is a significant penalty. Current means to measure the mass flow (with an uncertainty
of 1% to 2%) are not sufficiently accurate to determine small losses by differences between two large
flows. The accuracy of mass flow measurements should be improved further. Whereas LCO2 terminals
under  development  aim  to  identify  and  include  ‘no  regret’  items  to  allow  handling  of  LCO2  at  LP
conditions  in  the  future,  the  transfer  of  LCO2  between  two  pressure  levels  is  unexplored.  A  CO2
specification for LP shipping is expected to be more stringent for water content, but data is lacking for
other components.

A  particular  interface  between  transport  and  storage  is  for  solid  impurities  (particulate  matter)  as
these can directly impact injectivity. Stringent specifications on particulate content (total suspended
solids) are very well known in well interventions and produced water re-injection schemes from oil
and gas operations. The requirement can be linked to the target reservoir by the bridge theory11. The
removal of solids >5 micron is required even for the best reservoirs under matrix injection and more
stringent requirements are typical for most reservoirs under study in the North Sea. A specification
for particulate matter across the value chain is key to enforce discipline in pipeline commissioning and
avoid frequent filter change-out directly upstream the offshore well.

c)  Nomenclature
Relevant states of carbon dioxide (CO2) as they are named in this report (see also Figure 2):
Gas phase
denotes states below the vapour pressure of pure CO2, specifically those below the
pressure  on  the  saturated  vapour  line  of  CO2-rich  mixtures.    In  the  context  of  this
document, the term gas phase refers to fluid CO2 with a density normally below 100
kg/m3. At these states, compressors have to be used to increase the pressure further.
Dense phase  denotes states above the vapour pressure of pure CO2, specifically those above the
pressure  on  the  saturated  liquid  line  of  CO2-rich  mixtures  at  close  to  ambient
temperature.  In the context of this document, the term dense phase refers to fluid
CO2  with  a  density  above  500  kg/m3,  whereby  densities  in  dense  phase  pipeline
transport will be above 750 kg/m3 in most cases. At these states, pumps can be used
to increase the pressure further.

9 Deng, Roussanaly and Skaugen, 2019. Appendix B in Techno-economic analyses of CO2 liquefaction: Impact of
product pressure and impurities, 2019, performance of state-of-art designs is scarce,
https://doi.org/10.1016/j.ijrefrig.2019.04.011.
10 Section 8.3 in EU monitoring and reporting regulation Guidance Document No.1, Version, 10 February 2022.
11 The effect of solid impurities on field operations, FTC white paper.
11

Liquid phase
denotes states on the saturated liquid line of CO2 or of CO2-rich mixtures in tanks of
ships or train cars (LCO2). CO2 or CO2-rich mixtures in tanks are commonly transported
as  saturated  liquid.  Saturated  liquid  implies that  there  is  a  saturated  vapour  phase
above the liquid, which fills the headspace of the tank. However, in general the mass
of the saturated vapour is very small compared to the mass of the saturated liquid in
the tank. The density of CO2 in the liquid phase typically exceeds 1000 kg/m3.
Vapour phase  denotes states on the saturated vapour line of CO2 or of CO2-rich mixtures. The density
of CO2 in the vapour phase is typically below 60 kg/m3.
Solid phase
denotes  solid  CO2,  typically  at  temperatures  below  −54  °C.  Solid  CO2  is  frequently
referred to as dry ice, even though dry ice technically is a compressed form of solid
CO2. Without compression, solid CO2 has a more snow-like appearance.

The following abbreviations and terms are used throughout the report:
Auto refrigeration of CO2 refers to the effect that the temperature of a CO2 stream decreases as the
pressure is reduced without (sufficiently large) heat input from the outside. Compared to other
fluids, this effect is particularly strong in CO2. Expansion of CO2 can result in temperatures below
-50 °C and even in the formation of solid CO2.
Carbon Capture and Storage (CCS) refers to the process that consists of CO2 capture at the emitter’s
site, processing of the captured CO2, transport to a storage site and injection for permanent
storage in a safe geologic structure.
Carbon  Capture  and  Utilisation  (CCU)  refers  to  the  process  that  consists  of  CO2  capture  on  the
emitters site, processing of the captured CO2, transport to a utilisation site and utilisation of the
CO2 as part of a different value chain.
CO2  stream  refers  to  a  stream  of  captured  CO2,  which  necessarily  contains  impurities.  The
overwhelming majority of the stream is CO2 (at least > 95% on a molar basis).
Corrosion  resistant  alloy  (CRA)  refers  to  highly  alloyed  stainless  steel  alloys  in  the  context  of  this
document.
Direct air capture (DAC) refers to technologies that separate CO2 that is contained in ambient air at
low concentrations.
European  Economic  Area  (EEA)  denotes  the  members  of  the  European  Union  plus  Iceland,
Lichtenstein and Norway, which form a single market. Switzerland is not part of the EEA but
part of the single market. The UK is neither part of the EEA nor of the single market but remains
an important partner in European CCS activities.
Impurities  are  components  in  a  CO2  stream  other  than  CO2,  which  are  typically  present  at  low
concentration.
Intermodality refers to transport schemes, in which different transport modes (e.g. pipeline transport
and ship transport) interact in the sense that CO2 is transferred from one mode to the other on
its way from source to sink.
Interoperability refers to the ability of transport systems developed in different countries to enable
transport of CO2 over boarders without technical or legal restrictions.
Joule-Thomson  effect  /  coefficient  describes  the  temperature  drop  that  results  from  pressure
reduction without heat input from the outside.
Linear CCS value chain refers to a value chain, where CO2 is separated and transported to storage,
without using a part of the CO2 as, e.g., chemical feedstock, which is part of a different value
chain.
12

Low  pressure  (LP)  shipping  refers  to  ship  transport of  CO2  in  large  vessels, which  are  expected  to
operate  at tank pressures of about 7 bar and at temperatures around  -50 °C. Yet no vessels
operating at LP conditions have been build.
Liquefaction  refers  to  a  change  to  the  liquid  phase  in  the  context  of  this  document;  the  process
involves cooling to temperatures typically below -20 °C.
Medium  pressure  (MP)  shipping  refers  to  ship  transport  of  CO2  in  relatively  small  vessels,  which
typically  operate  at  tank  pressures  between  13  and  18  bar  and  at  temperatures  between
-35 °C and -22 °C without liquefaction system for boil-off gas, thus allowing for pressure build
up during transport.
Network code is a document that defines the physical parameters of a CO2 stream to be injected into
a  CO2  transport  network.  It  is  the  basis,  upon  which  commercial  arrangements  can  be
negotiated.
Non-condensable gases are components in a CO2 stream that, as pure fluids, cannot become liquid at
the temperatures that are characteristic for processes with CO2. Nevertheless, a liquid or dense
phase  CO2  stream  can  contain  small  amounts  of  such  non-condensable  gases.  Typical  non-
condensable gases are nitrogen, argon, oxygen, hydrogen, methane, and hydrogen, which may
be found as impurities in CO2 streams depending on the origin of the CO2.
Running Ductile Fraction (RDF) describes a failure mode, where a puncture of a pipeline develops into
an open crack over a significant length along the pipeline. Due to the phase behaviour of CO2,
pipelines transporting CO2 in the dense phase are particularly sensitive to RDF and need to be
designed in a way that avoids this failure mode.

13

3.  Impurities
When  discussing  CO2  stream  quality  regarding  the  interoperability  of  infrastructure  and  investors
predictability, a distinction can be made between two categories of impurities:

1.  Hazardous  impurities  are  those  with  an  impact  on  health  and  safety  or  the  integrity  of
pipelines or geological reservoirs. Established limits for each of these components are in the
ppm or ppb range. Due to their possible severe impact on integrity or health and safety, the
concentration of the impurities in the CO₂ stream is not subject to economic considerations
and is therefore non-negotiable.
2.  Non-condensable  impurities  whose  partial  removal  becomes  necessary  to  enable  flow
assurance and simply to leave more room for CO2 in reservoirs. In summary, several percent
of these components are usually considered acceptable, with a CO2 stream minimum purity
of 95 % being an internationally recognised standard. Since the effect of these impurities can
be  controlled  by  technical  means  and  since  low  levels  of  non-condensable  impurities  in
general  reduce  OPEX,  CAPEX  and  energy  demand  for  transport, while  they  increase  OPEX,
CAPEX and energy demand for purification, they might be subject to economical optimisation
and are therefore negotiable. Examples include H₂, N₂, Ar, and CH₄.

a)  Hazardous impurities
Health, Safety, and Environment (HSE) related limits cannot be subjected to economic optimisation.
Many, but not all, corrosion and other phenomena related to hazardous impurities will not arise if
corrosion-resistant alloys (CRA) are chosen (and others which are not found with carbon steel might
be identified), but their higher cost compared to carbon steel means that this option is not normally
chosen. Good Process Engineering methods should be applied at all times to produce a safe system.

Toxic components such as carbon monoxide (CO), hydrogen sulphide (H2S), nitrogen oxides (NOx) and
sulphur oxides (SOx) must be limited to satisfy the requirements of occupational and external safety.
In case of release of a CO2 stream, risks should be limited to that of exposure to CO₂ alone. Since most
toxic components have  a higher  volatility than CO2, they will accumulate  in the  gas phase if a CO2
stream is released to the environment as a multiphase flow consisting of gas and solid particles. The
resulting concentrations can be calculated using appropriate models. Alternatively, all toxic molecules
can conservatively be assumed to be present in the gas phase.

Corrosion of carbon steel is possible if an aqueous phase forms within the CO2 stream. Compared to
natural  gas  pipelines,  for  instance,  the  resulting  corrosion  rates  are  much  higher  because  CO2  and
water will form carbonic acid in this phase. Oxygen (O2) will also promote corrosion when dissolved.
Furthermore, some combustion products like NOx and SOx can form the much more potent nitric and
sulphuric acids. Some impurities can enable the formation of an aqueous phase, even if the water
content is sufficiently low to be normally fully dissolved in CO2. The most prominent examples are
glycols, whose use should be ruled out entirely until reliable results on applicable concentrations are
available  if  transfer  into  the  pipeline  is  possible.  To  a  lesser  extent  this  effect  is  also  known  from
methanol  and,  crucially,  amines.  The  maximum  allowable  impurity  levels  need  to  be  determined
covering the operating pressure and temperature window for the entire multimodal transport system
(gas/dense phase/NPT).

14

CO/CO₂  stress  corrosion  cracking12  is  another  possible  phenomenon  that  should  be  avoided,  and
research work to define this, and examine at what level of CO this ceases to be a problem needs to be
carried out.

The integrity of geological reservoirs may be compromised, for instance through caprock degradation
or precipitation of iron oxides. This can be caused by the presence of O2, NOx or SOx. Since water will
be  present  underground,  the  use  of  corrosion  resistant  alloys  (CRA)  will  often  be  necessary,  for
instance, in permanently wetted components. With sufficient amounts of NOx/SOx, especially in shut-
in  conditions,  the  droplets  of  water  may  show  very  low  pH,  leading  to  conditions  outside  of  the
application window of super duplex and similar steels (the classic CRAs). At the present time there are
no standardised tests to assess this behaviour. Limits on impurities may therefore be  governed by
either HSE or integrity criteria.

b)  Non-condensable impurities
These  impurities  include  hydrogen  (H2),  nitrogen  (N2),  argon  (Ar)  and  hydrocarbons  (mainly  CH4).
While for any pure fluid the phase change between gas and liquid can be charted as a line in a p-T
diagram, these components (on their own or in combination) lead to an envelope where both phases
are  present.  While  the  thermodynamic  states  of  saturated  vapour  (dashed  lines)  are  only  slightly
shifted towards higher pressures even at temperatures close to the critical temperature, the pressures
of saturated liquid (solid lines) can be raised substantially dependent on the concentration, especially
in  the  case  of  hydrogen  (Figure  1).  Where  possible,  states  within  the  two-phase  region  (between
saturated vapour and saturated liquid line) should be avoided for CO2 stream transport, at least in
pipelines.

12M. Gonuguntla et al., Wet CO-CO2 stress corrosion cracking in CO2 transport pipelines, Corrosion 2023, paper
number 19052, Houston TX, AMPP 2023.
15

Figure  1:  Influence  of  hydrogen  and  nitrogen  impurities  on  the  phase  envelope  of  CO2  streams,
according to TREND and RefProp.

For dense phase pipeline transport, a large fraction of non-condensable results in a higher minimum
pressure and thus lower allowable pressure drop and capacity. Costs for the transport of impure CO2
may  rise  because  it  is  necessary  to  increase  the  pipeline  wall  thickness  to  cater  for  the  increased
Maximum  Operating  Pressure,  operational  events  caused  by  fluid  hammer  and  to  avoid  RDF.  This
16

impacts the CapEx depending on the saturated liquid pressure of the transport fluid. Current basis for
design, accounting for RDF (DNV RP F10413), effectively limits the pipeline diameter, and the one large-
diameter pipeline may need to be replaced by parallel pipelines of smaller diameter. Standards and
codes may need subsequent adaption to allow for larger diameters, and by this reduce the number of
pipe strands for large clusters. For pipeline transport of gaseous CO2 streams and geological storage
non-condensable gases will simply result in lower capacity.

In the case of NPT at low temperature liquid conditions (LCO2), limits imposed on non-condensable
impurities are significantly lower than for pipeline transport or storage. This is because the Maximum
Operating Pressure of tanks is directly related to the saturated liquid pressure – in the tank there is
always  a  two-phase  system,  in  which  the  state  of  the  liquid  corresponds  to  saturated  liquid  with
roughly the medium composition, while the state of the gas phase above the liquid corresponds to
saturated  vapour  at  the  same  temperature  and  pressure,  but  with  deviating  composition.  Wall
thickness is proportional to Maximum Operating Pressure and related to CapEx. It is also limited to
around 50 millimetres due to welding constraints.

Limits on non-condensable impurities for pipeline  transport should consider the  costs  incurred for
their removal during capture and/or before Non-Pipeline Transport. It should be considered that the
purity requirements for shipping transport are more onerous than those for pipelines. If an intermodal
spine  transportation  system  is  to  be  realised,  then  the  purity  of  the  CO₂  stream  is  a  fundamental
decision.  If the CO₂ grid is designed around shipping transport then all emitters will need to comply
and additional costs will be borne by them compared to a design around pipeline transport, but further
processing downstream can be avoided.  A CO₂ grid built around a pipeline standard would require
further  processing  of  the  CO₂  stream  to  bring  it  to  shipping  standard,  but  the  rest  of  it  could  be
exported by pipeline. Addressing this matter as a design basis decision requires a clear vision for CO₂
stream transport and will be influenced by the route by which it is envisaged that the majority of the
CO₂ will be exported and the relative costs thereof. Differences in the availability of electrical power
and heat resulting from the considered processes and from local constraints may have a significant
impact  on  the  results.  Energy  efficiency  and  energy  integration  are  important  criteria  for  the
optimisation of all elements of the CCUS chain.

13 DNV-RP-F104 “Design and operation of carbon dioxide pipelines, Recommended practice”, Edition 2021-02 -
Amended 2021-09.
17

4.  Specific considerations
a)  Dense-phase pipeline transport
For  the  purposes  of  this  document  dense-phase  pipeline  transport  is  defined  as  transport  of  CO2
streams  with  a  density  above  500  kg/m3.  This  includes  liquid  states  and  an  adjacent  part  of  the
supercritical domain, see Figure 2. This handy value is defined from a technical perspective based on
the  experience  that  for  500 kg/m³  and  above,  pumps  can  normally  be  used  for  maintaining  the
operational pressure.

typical
dense phase
gas phase

Figure 2: Phases of pure and impure CO2

Within this range factors, like low compressibility and rotor dynamics, allow for pressurisation using
pumps rather than compressors. Along with the higher pipeline capacities, the resulting comparatively
low energy demand of these pumps results in lower transport costs at least over distances sufficient
to make up for any increased energy demand of initial compression and cooling. Allowable pressure
drops are limited by the saturated liquid pressure for the given composition and temperature.

For larger CO2 streams high-density transport may be the only option permitted by route planning
considerations due to the smaller pipeline diameters required. High-density CO2 is a powerful solvent,
and no inner pipeline coating system is currently known to maintain its integrity in this environment.
18

That means that surface roughness and thus pressure drop and OPEX cannot be decreased further
through internal coating.
When released in an uncontrolled manner, CO2 streams can reach temperatures of -78.5 °C and form
a solid phase (below 5.2 bar), which will sublimate over time. The dynamics of the pipeline system
needs  to  be  fully  considered,  but  the  required  wall  thickness  will  usually  be  governed  by  the
requirement to arrest a Running Ductile Fracture. The resulting wall thickness will largely depend on
the CO2 stream purity and, in particular, the hydrogen content. Water content for the CO2 stream is
best measured using mass spectrographs at the moment. Entries exceeding their limits can be shut
off.

While  high  density  pipeline  transport  is  an  established  technology,  some  design  or  operation
parameters  are  usually  chosen  conservatively  based  on  experience  without  detailed  quantitative
knowledge of some of the underlying mechanisms. In particular further research in the following areas
could lower future costs of the technology:
•  The methodology for design against RDF laid out in DNV RP F104 is based on limited
experimental data with modern X60 and X65 materials. Hence, the method is strictly
applicable only to those materials which could be a major limiting factor. Other pipelines
using other material types, for example X70 and X52, should not be designed or re-purposed
by this method. Additionally, the standard de facto limits the pipeline diameter which can be
used for dense-phase CO2 transport. The most obvious option to overcome this situation is
to enhance the scientific and technical understanding by conducting more mid-scale or full-
scale tests.
•  Impact of impurities on the formation of an aqueous phase, either through absorption or
creation of water through chemical reactions.
•  Concepts for tracking or monitoring of all relevant impurities.
•  Develop and validate models to predict the formation of solids from water, hydrogen and
other hydrate-building species.
•  Investigate whether an aqueous phase is a necessary condition for corrosion or if it can also
be facilitated by, for example, adsorption effects.
•  Dynamic simulation of the pipeline grid to identify challenges from dynamic reactions and to
trace impurities.

b)  Gas-phase pipeline transport
Gas phase transport carries the potential for a second phase to be formed by condensation, and the
impurity levels will need to be set such that this is an unlikely event. Formation of a liquid phase is
undesirable  because  it  can  cause  severe  operational  problems.  For  pure  CO2,  the  temperature-
dependent vapour pressure establishes the upper pressure limit for gas phase transport. For CO2-rich
mixtures, the corresponding pressure on the saturated vapour line (the dew point) depends not only
on temperature, but also on the concentration of less volatile impurities in the CO2. The most common
of these impurities is water, which can lead to the formation of a corrosive liquid phase. Thus, for the
layout of gas phase pipelines it is important to limit impurities in the CO2 stream to a level at which no
condensation occurs for maximum operating pressure and minimum operating temperature. Phase
equilibria  for  CO2/water  systems  are  well  described,  limits  for  allowable  water  contents  can  in
principle be calculated as a function of pressure and temperature in the pipeline. For temperatures
19

operating below  approximately 10 °C, hydrate and ice  formation have  to be  considered when two
phase formation is investigated.

However, there is a strong interaction between water and other impurities such as SO2, SO3, and NOx,
which can result in the formation of an acidic liquid phase at pressures below the dew point pressure
of CO2 containing only water. This influence, which is known as “acid dew point” in typical combustion
gases,  is  less  well  described  for  CO2-rich  mixtures.  Initial  specifications  for  allowable  contents  of
combined  impurities  have  to  be  conservative,  potentially  resulting  in  additional  cost  for  CO2
processing. Further research is recommended to derive traceable limits.
Other impurities that may lead to the formation of a second phase are traces of capture agents like
amines, amine mixes or ammonia. To specify allowable concentrations for capture agents, the possible
formation of a solid phase has to be considered; models allowing for an accurate description of the
corresponding  phase-equilibria  in  CO2-rich  mixtures  are  not  yet  available,  as  a  result  of  which
specifications have to be conservative.
In experiments, corrosion (at relatively low rates) has been observed at states at which existing models
do not predict the formation of a free corrosive phase. Whether the possible formation of a liquid-like
layer  by  adsorption  on  metal  surfaces  is  the  reason  for  these  observations  should  be  investigated
further.

Non-condensable gases (such as nitrogen, argon, oxygen, methane, hydrogen) increase the pressure
on the saturated vapour line and have therefore no adverse effect on allowable pipeline pressures in
gas-phase transport. Still, their maximum concentration should be specified because they increase the
required compression work and the volume flow in the pipeline. Except for oxygen, this limitation can
be guided by purely commercial aspects; low limits may go along with increased energy consumption
and  CO2  slip  during  processing.  The  allowable  oxygen  content  can  be  limited to  a  low  value  as  an
additional safety measure against corrosion but needs to be low to avoid adverse effects in the storage
media.

As explained above, the maximum operating pressure of a gas-phase pipeline is given by the pressure
on the dew line, which depends on temperature. Thus, the allowable operating pressure will largely
depend on the expected operating temperature – of course the effect of impurities described above
must be also considered. High pressure levels reduce the relative pressure loss along the pipeline. A
network-code for gas-phase CO2 stream transport should define standard pressure-levels to allow for
the development of standardised equipment.

Operating  procedures,  e.g.  for  venting  of  pipelines  or  shut-in  in  depressurised  pipelines,  and  case
studies for pressure-loss scenarios must consider the particularly strong Joule-Thomson effect in CO2.
Upon quasi adiabatic expansion from the gas phase, CO2 at a pressure above about 2 MPa can form a
liquid or even a solid phase. States that may result in the formation of a liquid phase are indicated in
yellow in Figure 3; states that may also result in the formation of a solid phase upon expansion to less
than 5.2 bar are indicated in orange. This drives the harmonisation of specified impurity levels in both
gas-phase  and  dense-phase  pipeline  transport,  since  similar  effects  arise  for  depressurisation  of
dense-phase  pipelines  as  well.  However,  the  influence  impurities  have  at  this  point  is  not  well
investigated yet. Formation of hydrates or water ice is likely and needs to be checked, based on the
allowed water content, considering the effect of potentially hydrate-forming impurities as well.
20

Figure 3: States relevant for gas-phase transport, at which adiabatic expansion of gaseous CO2 leads
to liquid (yellow) or for expansion pressures below 0.516 MPa even to solid (orange) formation.

In the long run, gas-phase pipes and pipelines carrying largely unprocessed CO2 streams from small
emitters to processing hubs, where the CO2 is dried and purified for further transport, may become
commercially attractive. Gas phase transport of largely unprocessed CO2 streams results in a number
of challenges that have not been properly addressed yet. This includes suitable choices for (corrosion
resistant) materials, safety regulations and operational procedures avoiding the formation of a solid
phase.  An example of this would be where different CO₂ sources in an industrial area employ a central
capture/processing facility which may be operated by an independent entity.

c)  Buffer storage
Buffer storage of CO₂ streams is needed to balance intermittent with continuous flow. An example
would  include  delivery  by  ship  into  a  port  and  subsequent  introduction  into  a  transportation
infrastructure or geological storage facility, both of which operate best under steady-state conditions.
Thus, the capacity of buffer tanks has to be adapted to the size and frequency of delivered charges.

To combine high capacity with low weight and costs of buffer tanks, storage of liquid CO2 (LCO2) at
low temperatures is technically and economically advantageous. For pressure levels and temperatures
in buffer tanks the same considerations apply as for ship transport (see Section 4.d). The same is true
for allowable impurities. If CO2 is delivered to the buffer as gas phase or dense-phase stream, the low
temperature liquefaction process is energy intensive, but it offers the potential for relatively simple
removal  of  non-condensable  impurities  like  air  components.  For  the  storage  facilities  themselves,
experience from industrial gas companies already exists for the storage of pure CO₂ (e.g. Linde, Air
Liquide).  Design  codes  and  regulations  (e.g.  from  the  European  Industrial  Gases  Association)  are
available as well. For impure CO₂, the Northern Lights project can act as an example.
21

No technological issues that would inhibit buffer storage of CO₂ streams are known. However, storage
of  CO2  containing  impurities  requires  consideration  of  some  additional  engineering  aspects,  in
particular when the composition of the delivered CO2 stream is not constant over time. In this case,
e.g.,  tanks  may  need  to  be  designed  in  a  way  that  avoids  “roll  over”-like  effects  due  to  different
densities and boiling points of different CO2-rich mixtures. In case the CO2 stream leaving the buffer-
tank needs to be heated up to pipeline or injection conditions, the energy efficiency of the process
requires special attention.

d)  Ship transport
As an alternative to pipeline transport, liquid CO2 (LCO2, liquid CO2 at low temperature and medium to
low pressure levels below 20 bar and above 5.2 bar) transport by ship is a link in the CCS chain. Ship
transport requires considerable CAPEX investment for liquefaction, storage and vessels, coupled with
higher liquefaction power requirements compared to the initial pipeline compression. Emitters that
do  not  have  access  to  local  sequestration  may  need  to  access  remote  storage  sites  that  are  only
accessible through marine routes. These could benefit most from LCO2  transport. Ship transport of
LCO₂ is also important where land or near-shore storage capacity is insufficient for delivery by pipeline,
or where CO2 stream has to be delivered to remote locations rich in renewable power as part of CCU
concepts. Hence, LCO2  transport is a key element in the CCS chain to provide flexibility and options
that can minimise the total cost of carbon abatement.

A generic depiction of a CCS chain that involves LCO2  transport is shown in Figure 4. In addition to the
LCO2  shipping  vessel,  the  terminal  requires  a  liquefaction  system  and  potentially  some  additional
treatment upstream  of  liquefaction. It  also requires LCO2  storage that holds the  CO₂ stream cargo
between loadings.  Similar storage facilities may be required at the reception location to allow a more-
or-less constant flow into the store.

Figure 4: Generic CCS chain with LCO2 link

Given the focus of this report on CO₂ stream specification, the different LCO2 shipping and storage
conditions  that  could  influence  CO2  stream  specification  requirements  need  emphasising.  Three
potential shipping conditions have been discussed within the industry:
•  High  Pressure  (HP):  Operating  at  near  ambient  temperature  and  correspondingly  high
pressure (higher pressure than for MP & LP transport, which operate at lower temperatures).
22

HP shipping can be considered a transport mode in development but is not currently the main
focus of the shipping industry.
•  Medium Pressure (MP): ≈13 bara, with pressure rising during the voyage up to approximately
18  bara.  MP  shipping  is  commercially  available  for  the  current  LCO2  trade.  For  pure  CO2  the
operating temperature after loading saturated liquid at 13 bara is -32°C.
•  Low Pressure (LP): ≈7 bara, with on-board refrigeration for boil-off gas (BOG) management. LP
shipping is not currently commercially available. Pure CO₂ saturated at a pressure of 7 bara
would operate  at  -49°C. On-board refrigeration would maintain this condition with limited
pressure rise during the voyage.

The transport of CO2 as boiling liquid at low temperature results in high demands on purity. Especially
the concentrations of water and volatile compounds need to be maintained at very low levels. The
presence of water might lead to freezing and the stabilisation of aqueous phases and hydrates, while
higher  levels  of  volatile  compounds  would  decrease  the  bubble  point  temperature,  thus  the
temperature in the tank at which the CO2 stream has to be transported. The motivation for LP shipping
is  to  achieve  economies  of  scale  by  transporting  larger  volumes  over  longer  distances. Minimising
shipping costs might turn out to be critical to optimise the value chain and can possibly be achieved
with LP LCO2 shipping due to the lower pressure requirements. But all of this comes at the cost of even
higher purity requirements of the CO2 stream.

The following recommendations can be made:
1)  It  needs  to  be  decided,  whether  a  common  value  for  water  content  within  the  CO₂  stream  is
specified  for  both  LP  and  MP  shipping,  as  it  may  have  only  minimal  or  no  impact  on  cost  or
technology selection for dehydration and it would simplify the LCO2  specification requirements.
2)  Separate recommendations for LP and MP LCO2 based on temperature rather than composition
are made and justified as follows. Since the solubility of volatile compounds is low at LP conditions,
it  will  be  difficult  to  establish  a  composition-based  allowance  of  volatiles  that  can  be  reliably
measured  and  designed  for.  Instead,  it  is  recommended  that  a  minimum  temperature  for  the
bubble point at given pressure is specified. In case of LP shipping, this temperature should provide
a reasonable margin to the triple point or phase boundary of solid CO2. A preliminary temperature
of -52°C at 7 bar could be applied; this results in a purity requirement of approximately 99.9%.
3)  The  same  rationale  for  a  temperature-based  specification  should  be  applied  to  MP  shipping.
Taking additional engineering considerations into account, a recommendation with a minimum
temperature of -40 °C at 13 bar could be made. This results in a LCO2 purity of approximately 99.5%
and allows for sufficient margin for pressure build-up during transport (MP tanks usually allow for
a  maximum  pressure  of  18  to  20  bar).  However,  currently  only  tank  materials  for  a  minimum
temperature of -35 °C are qualified in shipping codes. With a security margin, loading at  -30 °C
with 99.5% purity (with N2 as main impurity) would result in an initial pressure of about 17.6 bar,
leaving too little room for pressure build-up. Tank materials qualified for lower temperatures are
urgently required.

However,  the  given  temperature  specifications  are  considered  preliminary  suggestions  and  are
subject to further discussion. To convert measured compositions or boiling temperatures at deviating
pressure  into  equivalent  boiling  temperatures  at  7  bar  or  13  bar,  suitable  algorithms  need  to  be
agreed.
23

e)  Rail & truck transport
Today CO₂ rail transport is carried out in the liquid phase at low temperature (LCO2). Conditions in the
tank are very similar to conditions for MP shipping (see Section 4.d). Most rail tank-cars are approved
for  temperatures  down  to  -40°C,  corresponding  to  a  pressure  of  about  13  bar  at  0.5%  volatile
impurities, see Section 4.d. The loading temperature for pure CO2 is normally not lower than -34°C,
which corresponds to a pressure of about 12 bar. The test pressure of the tanks is at approximately
26 bar, pressure relief valves are normally set 1.3 times lower (20 bar) than the tank test-pressure.
Thus, for pure CO2 there is sufficient room for pressure build-up during transport due to heat intake.
However, for CO2 with 0.5% volatile impurities -34° C corresponds to a tank pressure of approximately
16 bar; the reserve for pressure build-up is significantly reduced. Existing tank designs are suitable for
the transport of LCO2 with impurities as discussed in Section 4.d, but the loading temperature needs
to  be  close  to  the  lowest  approved  temperature  to  compensate  for  the  pressure  effect  of  the
impurities.

As far as can be ascertained, there are currently no discussions with rail tank-car operators to allow
transport of CO₂ streams in the dense phase at ambient temperature. The operating pressure would
need  to  be  much  too  high  (more  than  80  bar)  for  an  economically  feasible  tank  car  construction.
Transport  temperatures  of  -50°C  or  lower,  which  correspond  to  the  specifications  for  LP  ship
transport, are an option but are not requested by the market today. Transport via (ISO) containers,
mostly with vacuum isolation, is a viable option for smaller volumes or intermodal solutions. As long
as pressure levels in the container are the same as in conventional tanks, the design of the tank has
no effect on the required CO2 stream specifications.

Whilst operating pressures, temperatures and the resulting constraints on purity are similar to those
for  MP  ship  transport,  operating  procedures  may  need  to  be  different.  Operating  procedures
established for the rail transport of pure CO2 are likely to be applicable for CO2 streams containing the
specified levels of impurities as well, but they need to be reviewed for possible implications arising
from the higher levels of impurities.

Collection of CO₂  streams from small emitters, in which the CO2  contains higher concentrations of
impurities,  and  then  subjecting  them  to  further  purification  at  the  collection  hub  may  be  an
economically attractive proposition as a result of the economies of scale possible for the purification
plant. However, in this case the level of impurities within the CO₂ stream whilst in the intermediate
transportation stage will be a relevant issue for the construction of new rail tank-cars. Higher water
and sulphur contents would result in challenges with possible corrosion of the tanks and the potential
for the formation of a solid phase in the associated pipes, valves and tanks. Tank cars and appropriate
operating procedures for this approach are not yet available.

Future  CCS  and  CCU  rail  transport  will  differ  from  today’s  market  mainly  by  increased  transport
volumes. This means logistics concepts will need to focus on large block trains, fast (un)loading and
quick round trips. CO₂ stream transport times in the rail tank-cars will be less on average in the CCS
and  CCU  market,  but  holding  times  will  still  be  relevant.  Transport  logistics  will  require  priority
transport  corridors  to  the  CO₂  terminals,  integrated  infrastructure  planning  and  support  of  the
industry for rail-network access and efficient (un)loading stations.
24

Truck transport of CO2 will be most likely to play only a minor role but may be relevant during the
start-up phase and also in the long run, when it becomes necessary to collect CO2 streams from small
emitters  as  well.  When  CO2  is  transported  as  LCO2  on  trucks,  the  same  considerations  as  for  rail
transport will apply. Where trucks are serving small emitters, the CO2 stream may be transported as
dense phase at ambient temperature as well. In this case bundles of high-pressure tanks with limited
diameter  could  be  used.  The  resulting  requirements  on  CO2  purity  are  in  general  similar  to  those
discussed for dense-phase pipeline transport in Section 4.a. Tanks and operating procedures allowing
for high-pressure transport of CO2 with higher concentrations of impurities have not been developed
yet.

f)  Geological storage – injection
Many aspects of the injection well design and operation are driven by the properties of bulk CO₂ rather
than its impurities14. Auto refrigeration of CO₂ streams in the well tubing might be an issue in low-
pressure reservoirs. When injection is interrupted quickly at the top of the well, the bottom of the
liquid CO₂ column will keep moving, whilst at the top of this column, some of the CO₂ will evaporate,
and  a  gas/liquid  interface  will  be  established  with  relatively  cold  temperatures.  This  feature  can
theoretically be avoided by deploying a smooth closing procedure when the reservoir pressure is high
(as in some aquifers). In highly depleted gas fields, it is anticipated that upon an interruption of the
flow, the entire wellbore will reach an equilibrium between the vapour phase CO₂ and the reservoir
pressure. If phase transitions take place within the tubing during shut-in, the accompanying thermal
cycling will be an additional potential failure mechanism that needs further understanding to avoid.
The presence of impurities (e.g., H2 and/or N2) results in a two-phase envelope instead of a saturation
line for pure CO2, which increases the likelihood of two-phase flow conditions being present in the
wellbore; this situation should, where possible be avoided in view of the possible associated hazards
and  computational  challenges  that  need  to  be  overcome  to  inform  the  operator  about  the
thermodynamic  behaviour  of  the  CO₂  stream.  Non-condensable  impurities  decrease  the  stream
density,  causing  the  mass  flux  to  decrease  over  the  same  pressure  drop,  and  thus  reduce  the
injectivity.

In the highly unlikely event of an uncontrolled CO₂ release from the well, the (sub)surface safety valve
(SSV)  or  other  barriers  fitted  in  the  well  must  withstand  the  low  temperature  associated  with
sublimation  of  CO2  until  near  atmospheric  pressure  while  containing  the  high-pressure  liquid
underneath.

During injection, at the well/reservoir interface, water will dissolve in the CO₂ stream resulting in the
formation of carbonic acid. Continuous injection may push the formation water away from the near
wellbore. However, cross flow/high permeability streaks may lead to water prevailing. This typically
drives the material selection in the well to corrosion resistant alloys (CRA). The elevated CO₂ stream
pressure  can  significantly  lower  the  pH  of  condensing  water  to  create  a  condition  where  de-
passivation of  CRA can occur,  also  possibly  driven  by  the  further  significant  decrease  in  pH  by  the
presence of SOx and NOx. De-passivation can cause localised corrosion and even increase the risk of

14 A more elaborate description of these effects can be found in Sections 3 and 4 of Acevedo and Chopra, 2017,
Influence of Phase Behaviour in the Well Design of CO2 Injectors, Energy Procedia.
25

cracking in these parts. Additionally, there is no general analysis of the potential for damage to the
minerals in the near-wellbore zone. As indicated above, the CO2 stream may contain water over a long
time, dissolving some of the impurities (like SO3 and SO2). The resulting strong acids may attack the
minerals in the  near-wellbore  zone,  leading to changing environments near to the injection points
which are, however,  usually far away from the  cap rock  relevant  for the  integrity of the reservoir.
Additionally, since  the  flow  pattern in the reservoir may  not  be known to the appropriate level of
granularity, the aggressive mixture of reservoir fluid and CO2 stream may come into contact with the
tubulars from the reservoir side and corrode from the outside. Due to this potential damage options,
an as-simple-as-possible, damage-tolerant completion is advisable.

For  CO₂  streams  with  typically  low  O2  levels  (~10  ppm-molar)  this  drives  the  selection  of  25Cr  to
mitigate the risk of de-passivation. This constraint is particularly relevant for CO₂ storage in depleted
gas  fields  as  these  would benefit  from the  conversion of existing  production wells  into  injectors15.
Existing wells are typically constructed from low-grade CRA such as 13Cr and some parts of the well
may not be replaceable. In addition to well integrity concerns this may lead to injectivity impairment
through corrosion by-products plugging the formation.

Backflow of CO₂ and formation water from the reservoir back into the well tubing cannot be ruled out.
This does not only influence the lower part of the well directly in the potential water column but also
shallower  sections  of  the  well  which  may  also  be  in  contact  with  this  corrosive  medium  through
convection and condensation effects. This is important for injection wells which are in intermittent
service (as opposed to continuous injection).  In particular those projects using direct injection from a
ship will mean a situation whereby short injection periods are followed by long waiting times for the
next ship to arrive. Corrosion takes place during the shut-in times, possibly all along the tubing, by
small bufferless condensed water droplets. Some impurities (for example the sulphur oxides) partition
preferentially from the CO2 phase into the aqueous phase, leading to extremely low pH values. Oxygen
and NO2 do not partition strongly into the aqueous phase but they maintain an oxidative potential
within the droplets which, in principle, render the mixture as very aggressive to stainless steels. Since
the  volume  of  those  droplets  is  small, the macroscopic  corrosive  attack  is  hard  to  forecast,  and  is
possibly very dependent on the location in the tubing string. An integrated computer model taking
into account natural convection, solution of water from the reservoir-CO2 interface, droplet-droplet
interaction and precipitation along the tubing incorporating the chemical effects would be necessary
to model this situation. Since no long-term field experiences with impure CO2 containing considerable
amounts of SO3, O2, and/or NO2 are available, it cannot automatically be assumed that the usual oil
and gas material solutions will work. Furthermore, sufficiently standardised tests for these situations
are not  available, and need to be established. Intermittent service might  lead to  frequent thermal
cycling which, due  to  the possibly  phase  transition in the CO2  case,  is normally not  experienced in
traditional injection wells and therefore needs to be looked at in greater detail.

For well cement, a suitably well-designed Portland cement, with low permeability, will have sufficient
resistance to diffusion-dominated carbonation along the axial length of the well such that it does not
pose a risk for loss of isolation. Carbonic acid can degrade cement through dissolution and subsequent

15 The re-use of wells can be to limit development costs but also due to the challenges associated with drilling
of new wells into highly depleted formations (underbalance drilling).
26

leaching of minerals from the cement, provided the cement is exposed to a considerable volume of
(flowing) water, which may result in an increase in porosity and reduction in strength. Flowing water
may not be found at caprock levels, especially if the seepage flow consists of humid CO2 (the more
buoyant phase) rather than of CO2-rich brine. Subsequent reprecipitation of carbonate minerals may
occur, slowing the rate of cement degradation and in some cases, imparting self-healing properties16.
It is anticipated that the extent of Portland cement degradation via bulk diffusion will be in the order
of 1 - 10 m over 10,000 years17. It should also be noted that extensive mineralisation of the CO2 will
take place within this timeframe.

The risk of H2S formation in the reservoir due to supply of nutrients for sulphur reducing bacteria or
hydrogen in the feedstock that can reduce sulphur containing minerals must be considered for failure
scenarios. This is not a generic requirement. For example, Rotligand reservoirs in the Southern North
Sea contain minerals that acts as H2S scavengers and will therefore limit H2S accumulation. A CRA such
as 25Cr is resistant to backflow of H2S.

g)  Geological storage – reservoir
Regarding the impact of impurities on the reservoir, three different regimes can be described: near
wellbore phenomena, their impact on capacity, and their impact on the containment.

The  near  wellbore  phenomena  strongly  depend  on  the  phase  behaviour  of  the  fluid  entering  the
reservoir.  Although  the  conditions  for  the  formation  and  dissociation  of  CO2  hydrates  in  brines  of
varying salinity are well understood, there are few studies, and thus little expertise, on the effects of
impurities such as CH
18
4,  N2,  CO  and  H2
. The same is true for salt precipitation, which is one of the
major  challenges  reducing  injectivity  in  geological  storage  systems.  The  effects  of  impurities  are
considered negligible in most studies. However, a few experimental studies suggest that SO2 and NO2
content  can  significantly  increase  permeability  by  further  lowering  the  pH  of  reservoir  brine  with
associated rock mineral dissolution, with the caveat that most studies use a large water:rock ratio that
is  not  representative  for  subsurface  conditions  and  that  overestimates  dissolution  effects.  The
dissolution may counteract the effect of salt precipitation near the wellbore, especially if the reservoir

16 Wolterbeek, T.K.T., Peach, C.J., Raoof, A., Spiers, C.J. (2016), Reactive transport of CO2-rich fluids in
simulated wellbore interfaces: Flow-through experiments on the 1–6 m length scale, International Journal of
Greenhouse Gas Control, 54(1), 96-116, https://doi.org/10.1016/j.ijggc.2016.08.034.
17  OEUK Well Decommissioning for CO2 Storage Guidelines (Issue 1, Nov 2022) - section 3.2 & Appendix B.3 &
B.4.
18 Horvat, K., Kerkar, P., Jones, K., Mahajan, D., 2012. Kinetics of the Formation and Dissociation of Gas
Hydrates from CO2-CH4 Mixtures. Energies 5(12), 2248–2262, doi:10.3390/en5072248.
Eslamimanesh, A., Babaee, S., Gharagheizi, F., Javanmardi, J., Mohammadi, A. H., Richon, D., 2013. Assessment
of clathrate hydrate phase equilibrium data for CO2+CH4/N2+water system. Fluid Phase Equilibria 349, 71–82,
https://doi.org/10.1016/j.fluid.2013.03.015.
Liu, J., Yan, Y., Li, S., Xu, J., Chen, g., Zhang, J., 2016. Structure and Stability of Binary CH4-CO2 Clathrate
Hydrates. Computational & Theoretical Chemistry (Accepted Manuscript),
http://dx.doi.org/10.1016/j.comptc.2016.04.010.
Sadeq, D., Al-Fatlawi, O., Iglauer, S., Lebedev, M., Smith, C., Barifcani, A., 2020. Hydrate Equilibrium Model for
Gas Mixtures Containing Methane, Nitrogen and Carbon Dioxide. OTC-30586-MS,
https://doi.org/10.4043/30586-MS.
27

rock contains carbonates.19 To be able to interpret case studies on reactive transport effects properly
it  is  important  to  consider  all  influencing  factors  and  interactions;  this  is  both  a  technical  and  a
scientific challenge, which is not easily addressed using current physical and numerical modelling tools
and expertise.

Impurities in the injected CO2 stream and/or gases already present in the storage formation can affect
storage capacity, although they are generally not considered in the preliminary capacity estimation.
Non-condensable  impurities  may  potentially  cause  a  reduction  in  the  storage  capacity  via  two
different ways:
1)  by taking up pore space that could have been used for the CO2, and/or
2)  by reducing the density of the CO2 weighted gas mixtures by decreasing the compressibility.

In an IEAGHG study20 the effect of the impurities was investigated for three scenarios of gas streams
in which the CO2 purity changes from low (85%-90% mol) to medium (95%-97% mol) and high purity
(>  99%  mol).  The  impurities  assumed  were  O2,  N2,  Ar,  H2O,  NO2,  SO2,  CO,  H2  and  H2S  at  various
concentrations in the CO2 stream. The storage capacity for a given storage geometry was calculated
using  Peng-Robinson  Equation  of  State  to  estimate  the  phase  behaviour.  It  is  found  that  the
normalised  capacity  (i.e.,  the  actual  storage  capacity  over  the  nominal  capacity  for  pure  CO2)
decreases with increasing concentration of impurities, and that at a certain pressure and temperature
there would be a maximum decrease in storage capacity of 60% of the nominal capacity, emphasising
the relative importance of CO2 density and compressibility of the mixture. More recent studies support
this  conclusion.  Due  to  the  relevance  of  pressure  and  temperature  on  mixture  density  and
compressibility,  the  importance  of  impurities  for  storage  capacity  decreases  with  increasing  depth
(higher pressure and temperature)21.

The effects of the impurities on the dissolution of CO2 in the brine and thus on the storage capacity
have not been widely studied in the literature to date. As most high-concentration impurities under
consideration  in  geological  storage  have  lower  solubility  in  water  than  CO2,  the  presence  of  these
impurities would reduce the partial pressure of CO2 and, therefore, reduce the dissolution of CO2 in
formation  water.  Acid  impurities,  such  as  SOx  and  NOx,  would  decrease  the  solubility  of  CO2  by
decreasing the pH of the formation water. However, dissolved rock minerals (carbonates) can serve
as a pH buffer and weaken the effect of acid impurities on CO2 dissolution. In one of the rare modelling
studies being published the total solubility of SO2 + CO2 in water varies exponentially with respect to
SO2 concentrations,  i.e.,  at  low  concentrations  of  SO2  (up  to  5%  as  a  maximum  case  in  geological
storage – which is already well above limits discussed for transport) total changes in solubility of the
CO2 in water are estimated to be negligible22.


19 Aminu, M. D., Nabavi, S. A., & Manovic, V. (2018). CO2-brine-rock interactions: The effect of impurities on
grain size distribution and reservoir permeability. International Journal of Greenhouse Gas Control, 78, 168–
176, doi:10.1016/j.ijggc.2018.08.008 .
20 Effects of Impurities on Geological Storage of CO2, IEAGHG, 2011.
21 Neele F, Koornneef J, Poplsteinova J, Brunsvold A, Eickhof C (2017) Toolbox of effects of CO2 impurities on
CO2 transport and storage systems. Energy Procedia 114:6536–6542.
22 Miri, R., Aagaard, P., & Hellevang, H. (2014). Examination of CO2–SO2 solubility in water by Saft 1.
implications for CO2 transport and storage. The Journal of Physical Chemistry B, 118(34), 10214–10223,
https://pubs.acs.org/doi/10.1021/jp505562j.
28

An important but poorly understood topic is the effect of impurities on the formation and growth of
gravitational instabilities in saline aquifers. It is well known that the formation of CO2-saturated brine
fingers and their movement in aquifers from top to bottom is a long-term but important contributor
to  storage  capacity  by  accelerating  dissolution  through  convective  transport.  For  a  given  Rayleigh
number, dissolution of N2 and H2S impurities makes the system stable, while dissolved SO2 accelerates
the onset of instability.23

The  effect  of  impurities  in  the  CO2  stream  on  containment  is  related  to  their  long-term  chemical
impact.  In  saline  aquifers,  during  and  after  the  storage  injection,  the  CO2  plume  (including  the
impurities)  tends  to  migrate  towards  the  storage  reservoir-caprock  boundary  potentially  inducing
geochemical reactions that may result in mineral dissolution and/or precipitation even in the cap rock
formation. A low pH brine in the vicinity of the CO2 plume may come into contact with the caprock
and may cause dissolution of the caprock minerals, especially if carbonates are a part of it. If SO2 and
NOx  are  present  in  the  CO2  stream  they  would  form  H2SO4  and  HNO3  in  brine  and  promote  the
dissolution of rocks. H2SO4 may also cause precipitation but HNO3 will not.
However, concentrations of these impurities will likely be limited to the ppm range due to concerns
regarding transport. And the way from the injection point to the caprock is long. Possible chemical
reactions will likely affect the vicinity of the injection site rather than the caprock – these effects have
been discussed in the previous sections. As further important natural safeguard, any reactions with
competent caprock (i.e., in absence of a pre-existing leak path like a conductive fault zone) are limited
to the lowermost portion of the caprock (centimetre scale). This is demonstrated by modelling studies
as well as by direct observations on sealing formations that have been exposed to CO2 for over 100,000
years24.  The  main  reason  for  this  is  the  slow  (diffusive)  transport  mechanism  in  caprock:  CO2  and
impurities can only enter the caprock through dissolution and diffusion in the formation water.

To date, no case has been reported where impurities in the CO2 stream affected the integrity of the
caprock.  However,  further  research  is  recommended  to  understand  the  processes  induced  and/or
accelerated by the impurities in the CO2 stream on the quality of the confinement – not to guarantee
the safety of confinements currently considered, which can be taken for granted based on today’s
knowledge, but to improve our understanding of fundamental effects, which may become relevant
for the assessment of confinements that have to be explored in the future to allow for wide spread
use  of  CCS  technologies.  In  addition  to  high  computational  capabilities,  reactive  flow  modelling
requires a thermodynamic database completed for the reaction parameters of the impurities in brines
with minerals representing the reservoir and caprock. Their coupling with geomechanical models is
important for evaluating reactive effects on the quality of the confinement further.

h)  Relevance of capture technologies for transport
The CO2 capture process has a significant impact on the type and level of impurities in the CO2 stream
for storage. These impurities broadly depend on 3 variables:

23 Kim, M. C., & Song, K. H. (2017). Effect of impurities on the onset and growth of gravitational instabilities in a
geological CO2 storage process: Linear and nonlinear analyses. Chemical Engineering Science, 174, 426–444.
24 Kampman, N., Busch, A., Bertier, P., Snippe, J., Hangx, S., Pipich, V., Di, Z., Rother, G., Harrington, J.F., Evans,
J.P. and Maskell, A., 2016. Observational evidence confirms modelling of the long-term integrity of CO2-
reservoir caprocks. Nature Communications, 7(1), p.12268.
29

1)  The chemical composition of the original carbon-containing material (natural gas, coal, coke,
liquid hydrocarbons, limestone, biomass, etc.). Any element in this original material (sulphur,
nitrogen,  chlorine,  heavy  metals…)  may  potentially  end  up  in  the  CO2  stream  for  storage.
Waste incineration can result in high and fluctuating levels of impurities due to the inevitable,
yet unpredictable fluctuations in the composition of the waste.

2)  The process through which the original material has undergone to produce the CO₂, resulting
in  the  diluted  stream  that  is  the  feed  for  the  capture  process.  Several  categories  can  be
identified:
-
Combustion,  when  air  is  used  as  oxidising  agent.  The  CO2  stream  produced  is  mainly
diluted with nitrogen, and smaller quantities of excess oxygen, water, and other impurities
produced in the combustion process, such as NOx, COS, SO2, and SO3. Capture strategies
targeting  the  flue  gas  streams  from  combustion  are  described  as  post-combustion
capture.
-
Oxy-fuel combustion, which uses essentially pure oxygen as the oxidising agent for the
combustion. In this case, the raw flue gas stream typically contains above 70%mol CO2.
The main impurities here are excess oxygen, inert gases not completely separated from
the oxygen at the Air Separation Unit (e.g., nitrogen, and argon), and water. Depending
on the fuel composition, impurities produced in the combustion process (SO2, etc.) may
be present, at higher concentrations than those in the case of combustion with air.
-
Partial oxidation processes, such as steam reforming or gasification, normally found in the
context of hydrogen production (e.g. to be used in hydrogenation / reduction processes,
ammonia  plants,  ‘blue’  hydrogen  production,  or  as  fuel  in  pre-combustion  capture
concepts).  In  any  of  these  cases,  the CO2-rich  stream  may  contain  hydrogen, nitrogen,
argon,  and  smaller  amounts  of  CO,  methane  and  other  light  hydrocarbons,  as  well  as
sulphur  compounds  (e.g.,  H2S)  were  this  element  to  have  been  present  in  the  feed
material.
-
Limestone (calcium carbonate) calcination produces CO2 and calcium oxide. This process
takes  place  during  clinker  manufacturing  in  cement  plants  and  requires  heat,  which  is
nearly always provided by combustion. Thus, the CO2 produced from calcination is likely
to be mixed with flue gas from combustion, which could either be based on air or on oxy-
fuel  combustion.  The  main  composition  difference  between  a  CO2  stream  from
combustion  and  that  from  a  cement/lime  plant  is  the  higher  CO₂  concentration;  other
impurities are essentially the same.
-
Another combined situation occurs in steelmaking facilities using the Blast Furnace-Basic
Oxygen  Furnace  (BF-BOF) route,  where  the  CO2  is  generated  at  different  points  and  in
different type of processes across the plant. Impurity levels are typically a combination of
those found in hydrogen production and in combustion processes with air, in different
proportions depending on the particular streams targeted for capture.
-
If CO2 is captured as part of natural gas / biogas treatment processes (gas sweetening),
the CO2 is expected to be diluted with light hydrocarbons, as well as with small quantities
of sulphur compounds.
-
Finally, CO2 can be captured from the atmosphere (DAC, Direct Air Capture), in which case
typical impurities are air components and traces of capture agents.

30

3)  The last variable affecting the  type and level of impurities is the capture technology itself,
which concentrates the diluted CO2 stream in order to improve the efficiency and cost of CO2
transport and storage. In the case of chemical (e.g., amines) or physical solvent technologies,
traces of the solvent may be found in the captured CO2 stream. If solid adsorbents (e.g., PSA,
VSA, or TSA processes) are used, solid particles could be expected. Cryogenic processes will
typically produce very pure CO2 streams due to the low-temperature process requirements,
which  are  normally  stricter  than those  for  pipeline  transport  or  geological  storage.  On  the
other hand, CO2 membranes, which may be used for biogas or natural gas treatment, are not
very  selective,  so  significant  amounts  (up  to  10%  mol,  or  even  more)  of  other  gases  (for
example, methane or nitrogen) have the potential to remain in the CO2 stream.

Based on the type and level of impurities in the captured stream, the following considerations can be
made on the CO2 treatment strategy to fulfil transport and storage specifications:
•  Dehydration – water will always need to be removed to ppm levels, since condensation may
lead  to  severe  corrosion  of  the  carbon  steel  materials.  Two  main  types  of  dehydration
processes are normally considered: glycol systems, which can reach water contents around
50  ppmv,  and  solid  adsorbents  (e.g., molecular  sieves),  which  can  reach  levels as  low  as 1
ppmv. At typical transport pressure and temperature in pipelines, both dehydration concepts
would  be  applicable.  However,  traces  of  glycols  in  the  captured  CO₂  might  lead  to  the
formation of a corrosive phase at transport conditions. Thus, glycol concentrations have to be
limited  to  very  low  levels  or  eliminated  altogether.  This  is  particularly  relevant  for  TEG
(triethylene  glycol),  which  is  frequently  used  in  dehydration  processes  for  natural  gas.
Accurate information on phase equilibria in CO₂ containing traces of water and glycols is not
yet  available.  Thus,  limits  for  glycol  concentrations  in  the  CO₂  need  to  be  chosen
conservatively to date. If the CO₂ stream is to be liquefied, water levels below 10 ppmv are
required, leading to the preferential use of solid adsorbents.
•  Particulate removal can be readily achieved with conventional filters down to micron levels.
•  Sulphur components removal – there are multiple sulphur removal technologies; selection is
normally based on the content and type of sulphur species in the CO₂ stream, together with
the  treated  CO₂  stream  specification.  Depending  on  the  specification  level  defined  for  the
transport,  the  selection  of  the  capture  technology  may  be  affected,  potentially  leading  to
higher capture costs. This is probably one of the cases where an optimum trade-off between
cost and impurity levels can be achieved.
•  Non-condensable gases (nitrogen, oxygen, argon, hydrogen, carbon monoxide, methane, etc.)
at the low concentrations typically found after the capture process cannot be easily removed
from  the  captured  CO₂  stream.  The  carbon  monoxide  content  should  be  reduced  to  the
specified level at the capture plant due to its toxicity. Hydrogen and methane can be oxidised
with catalysts, if necessary, under the addition of oxygen. On the other hand, oxygen can be
removed by catalytic reaction with (added) hydrogen. An energy intensive low temperature
process is required to remove nitrogen and argon. To minimise the whole-chain cost, it may
transpire that it is advantageous to reduce the concentrations of nitrogen and argon to the
low levels required for low temperature liquid-phase transport (ship or train / truck, buffer
storage) at central hubs as part of the liquefaction process. As in the case of sulphur, very
strict transport & storage specifications for the total amount of non-condensable gases or for
31

individual species may rule out  some in principle feasible capture  technologies, potentially
leading to higher capture costs.
•  Finally, specification levels for minor impurities such as acids (e.g., HCl, HF, HCN), nitrogen
compounds (e.g., NH₃, NOx), or organic compounds (e.g., amines, glycols, alcohols, aldehydes,
lubes, etc.) need to be established. In general, these impurities are relevant for the safety in
transport; concentration limits  need to be met  at the  capture plant. Stricter limits may  be
required for low temperature liquid-phase transport; to reduce concentrations further, post
processing at hubs before liquefaction may be an option.
•  Special consideration needs to be given for the case of mixtures of CO2 streams from different
capture processes. Mixing of varying streams can lead to compositions largely fluctuating over
time.  These  situations  present  specific  challenges  due  to  potential  interactions  between
different  impurities,  leading  to  unforeseen  variations  in  the  CO2  stream  composition.
Reactions may produce additional water that could raise the CO2 stream dew point, increasing
the  risk  of  acidic  water  drop-out  leading  to  pipeline  corrosion.  Solids  can  be  formed,  for
example  by  reaction  of  ammonia  with  CO2.  Both  the  dynamics  and  the  chemistry  of  these
processes  and  the  impact  of  changed  composition  on  the  (acid)  dew  point  need  to  be
investigated  further  to  come  to  optimised  specifications.  At  this  point  in  time,  transport
specifications for mixtures of CO2 streams need to be more stringent (i.e., with lower levels
for  individual  impurities)  than  if  each  CO2  stream was  to  be  transported  alone,  where  the
composition and the phase behaviour is more easily predictable.
•  Purification processes generally produce purge streams, which  contain some CO2 together
with  the  potentially  hazardous  impurities  that  have  been  removed.  The  handling  of  these
purge  streams  needs  to  be  controlled  carefully,  for  instance,  in  accordance  with  local
regulations.  Concepts  need  to  be  developed  to  assign  impurities  separated  at  hubs
downstream  of  the  capture  process  (together  with  the  associated  costs)  to  emitters.  In
particular,  processes  for  the  removal  of  non-condensable  components  can  be  energy
intensive.  The  energy  efficiency  of  the  whole  capture  and  transport  system  needs  to  be
considered.

i)  Port infrastructure for CO2 captured on board of ships
To decarbonise the shipping sector and to implement carbon capture onboard not only procedures
and technologies for CO2 capture and storage on board must be developed and implemented, but also
infrastructure needs to be provided at the receiving ports. The amount of CO2 that could potentially
be handled is substantial: for example, the fuel traded in Rotterdam, which is the largest bunkering
port in Europe, corresponds to annual CO2 emissions of more than 30 million tonnes when it is burnt.
If  CCS  on  ship  becomes  a  viable  and  widely  used  solution  for  the  decarbonisation  of  maritime
transport,  ports  will  need  to  act  as  CO2  hubs  in  the  future.  This  will  include  CO2  post-processing
onshore, since CO2 qualities delivered by ships may not be in line with demands for further transport.
Global standards need to be implemented to ensure the offloading of captured CO2, potentially with
fuel, propulsion, capture-type and ship-type dependent specifications. These standards would have
to cover not only pressure and temperature levels, the required purity and tolerable impurities but
also the offloading procedures.  These would have to be suitable for different vessels, regardless of
propulsion  type,  fuel,  size  or  manufacturer.  Technologies  and  procedures  required  to  monitor  the
quality of delivered CO2 would need to be implemented. Regular off-loading of CO2 in each port is
crucial because of the increasing mass of CO2 that would have to be moved, taking up space that would
32

otherwise be used for cargo. Storage capacities (possibly both for pre-processed CO2 coming from the
ships and for post-processed CO2 waiting for further transport) have to be provided to compensate
for the strongly fluctuating delivery of CO2. Ownership claims and responsibilities for the delivered
CO2 and for the impurities in the CO2 stream need to be clarified.

Thus, the following aspects must be covered by international standards and codes to ensure a viable
CCS chain:
•  Minimum purity of CO2 captured on board, likely to come from in different ship classes.
•  Maximum concentrations of certain components, likely to be from different classes.
•  Temperature and pressure levels for on-board storage and unloading.
•  Definition of responsibilities for purification, ownership of the delivered CO2 streams.

The  following  infrastructure  would  need  to  be  provided  in  ports,  and  appropriate  procedures  for
permitting and concepts for ownership of the infrastructure need to be established:
•  Technologies for CO2 offloading, connections for transmission: these are likely to be based on
bunker vessels.
•  Facilities for post-processing of CO2 and temporary storage capacity.
•  Options for further transport of the captured CO2, either by ship or by pipeline.
•  Procedures and equipment for monitoring quantity and quality of incoming CO2.
•  Heating and cooling as required by the following transport mode.

The ongoing development of technologies for CO2 capture, purification and storage onboard of ships
needs to be observed closely to deliver optimised solutions for the whole chain.

33

5.  Conclusions and recommendations
At  this  stage  of  CCUS  development,  setting  limits  for  individual  impurities  (or  an  effective  sum  of
several  impurities,  which  may  cause  similar  or  interdependent  effects)  may  well  be  the  most
prominent single problem. This is mainly because the level of impurities in a CO2 stream significantly
influences  each  part  of  the  value  chain,  and  different  stages  of  the  process  show  conflicting
requirements and different stages of technical maturity. For example, it seems reasonable that on the
capture side  a wide specification would lower costs and reduce  effort, while  on the transport  and
storage side an as-pure-as-possible stream would reduce many technical challenges and associated
costs. It is worth considering that specifications for pipeline transport can be less stringent than for
ship transport. For an optimisation of the whole value chain, more knowledge on every single step is
required  but  –  moreover  –  a  clear  political  framework  is  essential, which  defines  an  unambiguous
vision and clear targets of a common European CO2 transport network. However, since this industry
is an emerging one and the costs and need for purification in specific environments is not fully set and
understood, rules need to be set in a flexible way, where possible.

A complete picture of the quantitative description of the physical and chemical behaviour of impure
CO2 is still not available. Qualitatively, the phase behaviour of impure CO2 mixtures is well understood
for most technical impurities. There is a lack of understanding concerning the potential interactions
between several different compounds in the CO2  stream (for example  the stabilisation of aqueous
phases by polar molecules). For the prediction of phase transitions and the accompanying partitioning
of the impurities into the new phases and non-equilibrium phase transition conditions, there is little
reliable literature available for multi-component systems. This may have a direct impact on CapEx (for
example for the wall thickness requirements for avoiding Running Ductile Fracture conditions) and on
blowout  or  leakage  scenarios  and  their  safety  and  health  considerations.  More  public  research  is
required for all points listed above in order to reduce the degree of technical conservatism, which is
being applied at the moment in order that project development can continue.
Chemical reactions of impurities within the CO2 stream are hardly described and quantified in publicly
available literature. This is especially important considering that chemical reactions may take place at
every single step of the CCUS value chain, but of course the processes which contain a reactive CO2
stream mixture for prolonged times may face the highest consequences. From the research already
available it is obvious that the time scale for possible reactions is expected to be within the residence
time  for  transport  on  ships  or  in  pipelines  –  so  the  nature  and  quantity  of  reaction  products  and
possibly  additional  phases  needs  to  be  fully  understood  to  ensure  safe,  economic  and  reliable
transport without extra security margins. Since chemical reactions might  lead to the situation that
certain  impurity  levels  are  exceeded  during  storage  or  transportation  starting  from  an  in-spec
composition  (for  example  by  production  of  water  during  pipeline  transport),  a  common  political
concept  to  handle  this  situation  needs  to  be  elaborated.  Accordingly,  as  a  final  recommendation,
fundamental research on chemical reactions paths and kinetics seem to be of prime importance. On
that basis, safe and economic (because less conservatism has to be applied compared to thresholds
defined today) impurity threshold concentrations could be worked out.

The goal to establish standards and a network code that allow for the development of a multimodal
European CO2 transport network optimised to minimise the whole chain costs will result in the need
for scenario development and in specific considerations for different transport modes. Tools that are
able to combine the development of European CO2 stream transport scenarios with such a detailed
34

technical analysis are not yet available and need to be developed to come to cost-optimal solutions.
However, the development of transport scenarios is outside of the scope of this report. This report
focusses on characteristics of CO2 streams that are relevant for different elements of the transport
infrastructure.

Pipeline transport in the dense phase will most likely be the backbone of a European CO2 transport
network. Beside the overarching limitations for toxic impurities, the content of “non-condensable”
gases  (typically  the  air  components  nitrogen  and  argon,  but  also  methane  and  hydrogen  –  for  the
oxygen concentration low limits apply due to concerns regarding corrosion) plays an important role.
High  concentrations  of  these  gases  (up  to  5%  in  total)  can  be  accommodated  but  can  result  in
substantially more expensive pipelines and a higher energy demand than for gas phase transport. On
the other hand, low impurity limits increase the cost and energy demand of purification. An optimum
balance still has to be determined. The water content, the content of acid forming components, and
the  content of amines  and glycols have to be  limited strictly to avoid the formation of a corrosive
second phase, but allowable limits are higher than for other transport modes. A network code needs
to specify pressure levels for transport and operational procedures. Optimised operational procedures
need  further  thought  and  research,  mainly  because  throttling  of  CO2  in  the  dense  phase  leads  to
extreme  temperature  drops  and  may  result  in  the  formation  of  a solid  phase  or of  a  gas  phase  in
equilibrium  with  a  corrosive  liquid  phase.  The  dynamics  of  a  pipeline  network  for  CO2  containing
impurities  needs  to  be  understood,  as  well  as  the  impact  of  extended  two-phase  regions  on
operational procedures, including in particular the operation of pumps and compressors at transient
states of the pipeline. Standards for the design and permitting of pipelines that do not rely on tests
for every single project need to be established. For pipeline transport in general, the need to provide
accurate online monitoring of impurities still is a point of concern.

Pipeline transport at gas states will become relevant to connect single medium to large emitters to
hubs  feeding  the  CO2  into  the  backbone  established  by  high-density  pipelines  or  to  make  use  of
repurposed pipelines or in populous areas. For large amounts of CO2, pipeline transport at gas states
is usually not attractive. High concentrations of non-condensable gases can be tolerated, but limits on
water  content,  acid  forming  gases,  amines  and  glycols  need  to  be  even  stricter  than  for  pipeline
transport in the dense phase; a corrosive dense phase might be formed by condensation. The effect
of acid forming gases on condensation in a CO2 atmosphere is quantitatively not well described yet.
Challenges  resulting from solid phase  formation and operational procedures for pipeline pressures
above about 2 MPa (solid or liquid phase formation during expansion) need further consideration and,
if necessary, research.

Buffer storage  is an important element of all concepts  including non-continuous transport modes.
Storage in the dense phase under high pressure is possible and results in conditions similar to pipeline
transport in the dense phase. However, for large tanks storage as liquid phase at low temperature is
more  suitable  to  limit  the  required  wall  thickness,  even  though  the  energy  consumption  for
liquefaction at low temperature is much higher. Conditions are similar to those for ship transport. The
industry has long-lasting experience with low-temperature storage of pure CO2. However, impurities
in  captured  CO2  streams  may  result  in  effects  that  have  previously  not  been  experienced.  These
challenges  can  be  handled  but  need  to  be  anticipated  –  this  requires  further  consideration  and,  if
necessary, research.
35

Inland ship transport is relevant to collect CO2 streams along rivers and to transport the CO2 streams
to hubs, where it can be injected into the pipeline backbone or transferred to bigger sea going ships.
Sea going ships can be used to transport CO2 to storage sites. For remote storage sites this will likely
remain the preferred solution, whilst an increasing number of more favourably located storage sites
will be connected to pipelines over time. Ship transport will likely be particularly relevant for a fast
ramp up of CCS solutions. On board of ships, CO2 is generally transported in the liquid phase at low
temperature.  Small  to medium  ships will  likely transport  CO2  as  boiling  liquid  at medium  pressure
(about 13 to 18 bar at temperatures between -35 °C to -22 °C). Industrial experience with small ships
carrying pure CO2 at these conditions is available. For larger ships travelling further, transport at low
pressure (7 bar at  -49 °C) seems favourable. In general, liquid-phase  transport at low temperature
results in stricter requirements on purity than transport at ambient temperature. An effective removal
of non-condensable gases during liquefaction is possible, but requires additional equipment to limit
the loss of CO2 during the process. Corresponding regulations are pending. Strict limits are required
for components that might form solid phases at low temperatures (water, but also amines and glycols
– allowable concentrations for amines and glycols still need further research). Overly strict limits for
glycols  (in  particular  TEG)  may  make  dehydration  processes  more  expensive;  with  regard  to  TEG
further  research  is  urgent.  Shipping  at  low  pressure  (LP,  7  bar  at  -49  °C)  is  not  yet  established
technology. The vicinity of LP shipping conditions to the triple-point pressure of CO2 (5.2 bar, below
this pressure solid formation would unavoidably occur) requires further thoughts and maybe research
with  regard  to  operating  procedures.  Ship  transport  necessarily  includes  the  need  to  establish
infrastructure in ports (CO2 liquefaction and possibly post-processing, buffer storage, loading). Either
for injection into a reservoir or further transport in pipelines, the liquid CO2 transported by ship at low
temperature  needs  to  be  heated  up  again.  While  the  liquefaction  process  necessarily  is  energy
intensive, efficient technologies for heating up the CO2 stream need to be promoted.

Rail and possibly truck transport will be important transport modes during the start-up phase of a
CO2-transport network. In the long run they will be likely to be restricted to small to medium remote
emitters. Rail cars and trucks usually transport pure CO2 as a boiling liquid at low temperature and
medium pressure; industrial experience for pure CO2 is available. In general, constraints regarding CO2
purity and operating procedures are similar to medium pressure ship transport. However, additional
thoughts on operating procedures are required, e.g., to limit the amount of CO2 that escapes as gas
phase while filling the tank. Vacuum isolated containers may turn out to be an attractive alternative
to classical tank cars.

Additional constraints on CO2  characteristics may  result  from the injection into geological storage
sites. Most issues related to fluctuating injection largely depend on the characteristics of CO2 itself,
rather  than  on  impurities  in  the  CO2  stream.  Materials  for  the  injection  pipe  and  flow-wetted
components in the well need to be chosen in a way that they can withstand a CO2/brine mixture in
case of a backflow. Acid forming impurities in the CO2 increase the risk of corrosion and the threat to
the cement plug closing the borehole. However, acceptable limits are engineered to be higher than
for transport. In order to pose a significant threat to cement integrity, the presence of acid forming
impurities  in the  CO2  would have  to fundamentally alter the reactive  transport processes  involved
compared to CO2. None of the studies to date provide evidence or indicate that this will be the case.
Thermal cycling due to fluctuating CO2 flows and its effect on the injection facilities and the reservoir
36

closure to the injection point are considered critical operational issues. Further thoughts and research
are recommended to determine practical limits with regard to the flexibility of mass flows into the
reservoir.

On the reservoir side, impurities in the CO2 stream may affect the near well-bore area, the capacity of
the reservoir, and theoretically, its integrity. In the near well bore area, chemical reactions induced or
intensified by impurities can be a challenge for the injectivity. More frequent drilling of new wells may
be an economically severe management solution. The capacity of the reservoir can be significantly
reduced  by  non-condensable  gases,  whereby  the  corresponding  processes  can  be  complex  and
adverse effects could result from other impurities – this needs to be considered when the capacity of
reservoirs is estimated. Theoretically, chemically reacting impurities (in particular from acid-forming
components like SOx and NOx) can be a threat to the integrity of the caprock. However, concentrations
of  these  impurities  are  already  restricted  to  the  ppm  level  to  address  concerns  with  upstream
infrastructure, and the distance from the well bore to the caprock is usually long. Possible chemical
reactions are more likely to take place close to the well bore rather than at the caprock and, simply
due to stoichiometry, the amount of rock that can be affected is small. Even if the caprock is affected,
the key natural safeguard against any reactions with an intact caprock is that they are limited to the
lowermost portion of the caprock (centimetre scale) due to the slow (diffusive) transport mechanism
in caprock. To date, no cases have been identified where the caprock was weakened due to impurities
on the levels anticipated for CO2 stream transport.

The  chosen  capture  technology  is  relevant  for  the  impurities  contained  in  captured  CO2,  as  is  the
source  of  the  CO2  (e.g.,  coal,  gas,  waste,  biomass,  cement)  and  the  conversion  process  (e.g.,
combustion,  gasification,  fermentation,  calcination).  Different  combinations  of  these  three  factors
result in very different requirements for the processing of captured CO2, but they cannot be reflected
in  different  access  conditions to  a European  CO2  transport  network.  Risks  can arise  from chemical
reactions that could result from the mixing of different impurities from different CO2 streams and may
include solid formation and the formation of an unpredicted corrosive phase. These risks are partly
understood and can be circumvented by defining conservative limits for impurity concentrations, but
further thoughts and research are required to deliver economically optimal limits in the long run.

CO2 capture from the exhaust gas of ships is a means to reduce the CO2 emissions of marine transport.
Globally,  several  research,  development  and  demonstration  projects  address  this  technology.
Although the technology by itself is not relevant for a European CO2 transport network, it needs to be
understood that it may turn ports into important sources of CO2. Regarding the characteristics of the
landed CO2 (pressure and temperature levels, impurities), global standards need to be established to
allow ships to unload their CO2 in arbitrary ports. These standards will probably not be identical to
standards defined for the European transport network. Thus, ports need to be equipped not only with
unloading facilities, buffer storage and connections to the transport network, but also with facilities
to process  further the  landed CO2. This aspect  needs further thoughts,  research, and international
cooperation to come to optimal solutions.
37

6.  Sources

Acevedo and  Chopra.  Sections 3  and 4 of  Influence  of Phase  Behaviour  in  the Well Design of  CO2
Injectors, 2017.
Aminu, Nabavi, & Manovic, 2018. CO2-brine-rock interactions: The effect of impurities on grain size
distribution and  reservoir permeability.  International  Journal  of  Greenhouse  Gas  Control, 78, 168–
176, doi:10.1016/j.ijggc.2018.08.008.
DNV-RP-F104  “Design  and  operation  of  carbon  dioxide  pipelines,  Recommended  practice”,  Edition
2021-02, Amended 2021-09.
Effects of Impurities on Geological Storage of CO2, IEAGHG, 2011.
Engel  and  Kather,  2018.  Example  provided  by  Figure  2  in  Improvements  on  the  liquefaction  of  a
pipeline CO2 stream for ship transport, 2018, https://doi.org/10.1016/j.ijggc.2018.03.010.
Eslamimanesh, Babaee, Gharagheizi, Javanmardi, Mohammadi, Richon, 2013. Assessment of clathrate
hydrate  phase  equilibrium  data  for  CO2+CH4/N2+water  system.  Fluid Phase  Equilibria  349, 71–82,
https://doi.org/10.1016/j.fluid.2013.03.015.
Gonuguntla et al., Wet CO-CO2 stress corrosion cracking in CO2 transport pipelines, Corrosion 2023,
paper number 19052, Houston TX, AMPP 2023.
Horvat, Kerkar, Jones, Mahajan, 2012. Kinetics of the Formation and Dissociation of Gas Hydrates from
CO2-CH4 Mixtures. Energies 5(12), 2248–2262, doi:10.3390/en5072248.
International Code for the Construction and Equipment of Ships Carrying Liquefied Gases in Bulk (IGC
Code), International Maritime Organization.
Kampman,  Busch,  Bertier,  Snippe,  Hangx,  Pipich,  Di,  Rother,  Harrington,  Evans  and  Maskell,
Observational  evidence  confirms  modelling  of  the  long-term  integrity  of  CO2-reservoir  caprocks.
2016. Nature Communications, 7(1), p. 12268.
Kim  and  Song, 2017. Effect  of  impurities on  the  onset  and  growth of  gravitational  instabilities  in  a
geological CO2  storage  process: Linear and nonlinear analyses. Chemical Engineering Science, 174,
426–444.
Liu,  Yan,  Li,  Xu,  Chen,  Zhang,  2016.  Structure  and  Stability  of  Binary  CH4-CO2  Clathrate  Hydrates.
Computational
&
Theoretical
Chemistry
(Accepted
Manuscript),
http://dx.doi.org/10.1016/j.comptc.2016.04.010.
Miri, Aagaard, & Hellevang (2014). Examination of CO2–SO2 solubility in water by Saft 1. implications
for  CO2  transport  and  storage.  The  Journal  of  Physical  Chemistry  B,  118(34),  10214–10223,
https://pubs.acs.org/doi/10.1021/jp505562j.
Neele, Koornneef, Poplsteinova, Brunsvold, Eickhof (2017) Toolbox  of effects of CO2  impurities  on
CO2 transport and storage systems. Energy Procedia 114:6536–6542.
Notaro,  Belgaroui,  and  Maråk,  Tverrå,  Burthom,  Sørhaug,  2022.  Ceto:  Technology  Qualification  of
Low-Pressure  CO2  Ship  Transport,  Proceedings  of  the  16th  Greenhouse  Gas  Control  Technologies
Conference (GHGT-16), http://dx.doi.org/10.2139/ssrn.4272083.
38

Offshore Energies UK, Well Decommissioning for CO2 Storage Guidelines (Issue 1, Nov 2022) - section
3.2 & Appendix B.3 & B.4.
Sadeq,  Al-Fatlawi,  Iglauer,  Lebedev,  Smith,  Barifcani,  2020.  Hydrate  Equilibrium  Model  for  Gas
Mixtures
Containing
Methane,
Nitrogen
and
Carbon
Dioxide.
OTC-30586-MS,
https://doi.org/10.4043/30586-MS.
Wolterbeek,  Peach,  Raoof,  Spiers,  Reactive  transport  of  CO2-rich  fluids  in  simulated  wellbore
interfaces:  Flow-through  experiments  on  the  1–6  m  length  scale,  2016,  International  Journal  of
Greenhouse Gas Control, 54(1), 96-116, https://doi.org/10.1016/j.ijggc.2016.08.034.

39