Energy Strategy Reviews 22 (2018) 61–81

Contents lists available at ScienceDirect

Energy Strategy Reviews

journal homepage: www.elsevier.com/locate/esr

An assessment of CCS costs, barriers and potential T

a,b,∗ c b,d b,c a,b
Sara Budinis , Samuel Krevor , Niall Mac Dowell , Nigel Brandon , Adam Hawkes
a
Department of Chemical Engineering, Imperial College London, London, SW7 2AZ, UK
b
Centre for Process Systems Engineering, Imperial College London, London, SW7 2AZ, UK
c
Department of Earth Science and Engineering, Imperial College London, London, SW7 2AZ, UK
d
Centre for Environmental Policy, Imperial College London, London, SW7 2AZ, UK

A R T I C LE I N FO A B S T R A C T

Keywords: Global decarbonisation scenarios include Carbon Capture and Storage (CCS) as a key technology to reduce
Energy carbon dioxide (CO2) emissions from the power and industrial sectors. However, few large scale CCS plants are
CO2 operating worldwide. This mismatch between expectations and reality is caused by a series of barriers which are
CCS preventing this technology from being adopted more widely. The goal of this paper is to identify and review the
Cost
barriers to CCS development, with a focus on recent cost estimates, and to assess the potential of CCS to enable
Integrated assessment model
access to fossil fuels without causing dangerous levels of climate change.
The result of the review shows that no CCS barriers are exclusively technical, with CCS cost being the most
significant hurdle in the short to medium term. In the long term, CCS is found to be very cost effective when
compared with other mitigation options. Cost estimates exhibit a high range, which depends on process type,
separation technology, CO2 transport technique and storage site.
CCS potential has been quantified by comparing the amount of fossil fuels that could be used globally with
and without CCS. In modelled energy system transition pathways that limit global warming to less than 2 °C,
scenarios without CCS result in 26% of fossil fuel reserves being consumed by 2050, against 37% being con-
sumed when CCS is available. However, by 2100, the scenarios without CCS have only consumed slightly more
fossil fuel reserves (33%), whereas scenarios with CCS available end up consuming 65% of reserves. It was also
shown that the residual emissions from CCS facilities is the key factor limiting long term uptake, rather than cost.
Overall, the results show that worldwide CCS adoption will be critical if fossil fuel reserves are to continue to be
substantively accessed whilst still meeting climate targets.

1. Introduction system transition to a low carbon world, providing estimates of the

future use of fossil fuels, CCS, and other energy resources that are
Carbon Capture and Storage (CCS) is a technology aiming at se- consistent with climate change mitigation targets. IAMs use a range of
parating, transporting and permanently storing carbon dioxide (CO2) methodological approaches that determine which technologies are se-
underground in order to avoid its emission into the atmosphere. CCS is lected, along with a range of input data assumptions like costs and
often argued to be a key technology for the decarbonisation of the performance, which all have a strong bearing on outcomes. For ex-
global energy system and can be applied to both power generation and ample, the IEA [3] has employed an integrated assessment model to
industrial production. For the industrial sector in particular, when support a roadmap assisting governments and industry in integrating
material or process replacement is not technically or economically CCS in their emissions reduction strategies. Adoption of this roadmap
feasible, CCS is currently the only technology able to drastically reduce would enable storage of a total cumulative mass of approximately 120
carbon emissions. GtCO2 between 2015 and 2050 (according to the Carbon Tracker In-
The pivotal role of CCS as a transitional technology towards a low or itiative, this value would be higher and equivalent to 125 GtCO2 [4]).
zero emission future has been highlighted by a number of recent pub- In this context, the importance of CCS is evident. This technology
lications [1,2]. A key tool in the assessment of possible pathways for the could enable countries to continue to include fossil fuels in their energy
decarbonisation of the energy system are global Integrated Assessment mix [5] without further exacerbating climate change and therefore
Models (IAMs). These models are used to produce scenarios of energy could unlock assets that would otherwise be stranded [6,7]. Moreover,

∗
Corresponding author. Department of Chemical Engineering, Imperial College London, London, SW7 2AZ, UK.
E-mail address: [email protected] (S. Budinis).

https://doi.org/10.1016/j.esr.2018.08.003
Received 9 January 2018; Received in revised form 15 May 2018; Accepted 1 August 2018
Available online 20 August 2018
2211-467X/ © 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/BY/4.0/).
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

Abbreviations GHG Greenhouse Gas

HadSCCCM1 Hadley Centre Simple Climate-Carbon-Cycle Model
AR Assessment Report IAM Integrated Assessment Model
BECCS Bio Energy with Carbon Capture and Storage (CCS) IEA International Energy Agency
CCGT Combined Cycle Gas Turbine IGCC Integrated Gasification Combined Cycle
CCS Carbon Capture and Storage IPCC Intergovernmental Panel on Climate Change
CDR Carbon Dioxide Removal LCOE Levelized Cost OF Electricity
CO2e Equivalent carbon dioxide LHV Low Heating Value
COE Cost Of Electricity NET Negative Emission Technology
COP Conference Of the Parties NGCC Natural Gas Combined Cycle
CPS Current Policies Scenario NGO Non-Governmental Organization
ECBM Enhanced Coal Bed Methane NOAK Nth Of A Kind
ECMR Enhanced Coal Bed Methane Recovery NPS New Policies Scenario
EGR Enhanced Gas Recovery NPV Net Present Value
EMF Energy Modelling Forum PC Pulverised Coal
EOR Enhanced Oil Recovery SiMCaP Simple Model for Climate Policy assessment
EWS Efficient World Scenario TRL Technology Readiness Level
FOAK First Of A Kind UKERC UK Energy Research Centre

meeting climate targets without adopting CCS would mean up to 138% understood, at least to the extent that these reserves, if converted to
increase in total discounted mitigation costs [8]. CO2 and released into the atmosphere, are demonstrably larger than the
Despite the positive estimates on the potential role of CCS reported allowable carbon budget for a 2 °C world. Less compelling evidence
in the literature, the current number of operating CCS plants is limited. exists on likely outcomes with respect to fossil fuel utilisation, where
According to the Global CCS Institute, there are currently 39 large-scale the use of abatement technology such as CCS might unlock fossil fuel
CCS projects worldwide in either ‘early development’, ‘advanced de- reserves whilst meeting carbon emission targets.
velopment’, ‘in construction’ or ‘operating’ phase [9]. Among the pro- This article identifies and reviews potential CCS barriers, with a
jects currently in operation (17), nine are based in the United States, focus on CCS costs, and reviews the evidence for the potential role of
followed by Canada (3 projects), Norway (2 projects) and Brazil, Saudi CCS technology in unlocking fossil fuel assets that would otherwise be
Arabia and United Arab Emirates (1 project each). The Boundary Dam stranded in a world where CO2 emissions are severely constrained.
Carbon Capture and Storage Project [10] and the Petra Nova Carbon In section 2, the paper covers the evidence on this broad issue in-
Capture Project [11] are the only two examples of CCS applied to power cluding the climate science, global data on fossil fuel reserves and re-
generation, while the remaining 15 operating projects are on industrial sources and quantification of unabated burnable carbon. Section 3
production (ethanol, fertilizers, hydrogen, iron and steel, synthetic summarises the potential barriers to the full development of CCS, in-
natural gas) and natural gas processing [12]. cluding costs (which are covered in details in section 4), geo-storage
The total number of large scale CCS projects has fallen in the past capacity, source-sink matching, supply chain and building rate, policy
five years, from 75 (2012) to 39 currently (2017). At the same time the regulation and market, and public acceptance. Section 4 summarises
number of projects in the ‘operating’ phase has increased from 8 (2012) cost metrics and estimates for CCS energy and efficiency penalty; CO2
to 17 (2017). These trends reflect both the technical feasibility of CCS capture, transport and storage; capital and operating costs. Section 5
and its struggle to emerge as a game-changing technology against cli- includes a review of a multi-model IAM comparison study that con-
mate change. sidered CCS in relation to the unburnable carbon concept, and quan-
The cost of CCS has been previously identified as a major barrier to tifies the potential of CCS to give access to fossil fuels in the long term
its adoption, however there are other potential barriers which are while meeting stringent climate targets. Section 6 provides an analysis
preventing its wider implementation. One of the goals of this article is on the influence of residual CO2 emissions on the adoption of CCS in the
to identify the barriers to the global adoption of CCS, with a focus on its energy scenarios. This leads to recommendations on the treatment of
costs. The second goal of the paper is to quantify CCS potential, in this aspect of CCS in unburnable carbon assessments in future (section
particularly with reference to the concept of ‘unburnable carbon’. This 7) and conclusions (section 8).
concept points out that known fossil fuel reserves cannot all be con-
verted to CO2 emitted to the atmosphere (i.e. burned or otherwise) if
2. Background
the world is to avoid dangerous climate change. A number of reports
have been published recently on the topic, though it is by no means a
2.1. The global greenhouse gas budget
new issue, with analysis available from as early as the 1990s. These
studies present a range of insights, from commentary on how the un-
2.1.1. The need for emissions mitigation
burnable issue may or may not imply the existence of a ‘carbon bubble’
It is unequivocal that climate change is influencing the planet, with
in terms of impact on fossil fuel company value, through to analysis
a range of effects already observable [13]. It is also extremely likely
identifying specific fossil fuel related projects that may not be needed
that this is caused by emissions of GHG ensuing from human activities,
given the perception of an impending reduction in fossil fuel demand,
directly (e.g. fossil fuel combustion, cement production) or indirectly
combined with their potentially high cost relative to other projects.
(e.g. deforestation). Given the observed impacts to date, the extreme
Analyses on unburnable carbon exists in the grey literature, pro-
nature of potential future effects on natural and human systems [14],
duced by banks, consultancies, insurers, think tanks and NGOs (Non-
and the rapidly increasing emissions [8], it is pressing that decision
Governmental Organisations). Academic research behind the insights is
makers consider options to mitigate climate change by reducing emis-
also available in specific areas, but few studies exist that span the topic.
sions, and plan adaptation to deal with climate change that does occur.
In particular, a substantial body of research exists in the climate science
On the mitigation side, this has led to the concept that the world has
domain on the extent of the global carbon budget and the impacts of
a constrained greenhouse gas emissions budget; a cumulative emissions
climatic change. Also, the extent of fossil fuel reserves is fairly well
limit which if breached is likely to lead to a global mean surface

62
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

temperature rise of more than 2 °C [15]. Peak warming given by cu- and economic (i.e. reserves) and discovered sub-economic (resources)
mulative emissions has been adopted by the scientific community as a [26]. Since 1972, various nomenclatures have been proposed and
reliable measure of climate change [16], while the 2 °C limit was chosen adopted [27–33].
because the best evidence on projected impacts and damage indicate Broadly speaking, “reserves” refers to the quantity of fossil fuels that
that effects are more limited and more certain below this level [17]. is likely to be extracted under economic conditions (i.e. a given set of
However, even 2 °C cannot be considered completely safe, with “well fossil fuel prices versus project costs) that make a specific project fa-
below” 2 °C emerging from the Paris Agreement [18], and in any case vourable. The commercial criteria depend on what can be defined as
adaptation will still be required. “commercial”. In most definitions, commercial is used as being sy-
nonymous to “economic”, which means that “the project income will
2.1.2. Carbon budget estimations cover the cost of development and operations (at zero discount rate)”
Carbon budgets that lead to warming of greater than 2 °C have also [35]. Simplistically, fossil fuel price is in turn determined by the mar-
been produced using climate models such as MAGICC 6.0 [19] [15], ginal cost of production, which is the cost of the most expensive fossil
HadSCCCM1 [16] [20,21] and SiMCap EQW [22,23]. These models fuels at that point in time. Therefore the extent of aggregate global
have been employed by different research groups and institutions in reserves is a function of the prevailing fossil fuel price, which itself has
order to estimate the carbon budget, which represents the maximum proven to be a very volatile quantity. This makes any estimate of re-
amount of CO2 that can be released to the atmosphere in order to limit serves open to debate, and indeed the supply curve for each fossil fuel is
the temperature rise below a certain target. In a range of studies at- dynamic in nature.
tempting to quantify this budget, the authors almost universally ac- The extent of reserves is contentious with respect to its link to the
knowledge the uncertainties associated with the estimations, in that the ‘carbon bubble’ concept, which is driven by the fact that if some re-
chain of causes and effects from emission through to temperature rise is serves are unburnable the companies that own those reserves might be
very complex. Key sources of uncertainty are budget type definition, the overvalued in the stock market [36]. However Mayer and Brinker [37]
underlying data and modelling, the scenario selection, temperature have argued that the perception of carbon risk has been inflated by the
response timescales and accompanying pathway of CO2 and non-CO2 choice of definition for the reserves, in particular that reserves esti-
emissions [24]. Climate science is a rich and active area of research and mated using the SEC (Security and Exchange Commission) method are
as such estimates of the global carbon budget are likely to be refined not as high as some other methods, and also that these reserves are
over time. likely to be monetised quickly. Others argue that regardless of a par-
Table 1 summarises the carbon budgets as estimated by reported ticular company's exposure in terms of ownership of fossil fuel reserves,
sources. According to Meinshausen et al. [15], the probability of ex- the impact of the unburnable issue on fossil fuel prices is likely to have
ceeding 2 °C can be limited to below 25% (50%) by keeping 2000–2049 an influence on the degree that companies value their assets i.e. an
cumulative CO2 emissions from fossil sources and land use change to indirect ‘carbon bubble’ effect [38].
below 1000 (1440) GtCO2. They also estimate that non-CO2 GHGs
(including methane, nitrous oxide, hydrofluorocarbons, per- 2.2.2. Reserves and resources estimations
fluorocarbons and sulphur hexafluoride) may constitute 33% of overall In order to evaluate the amount of unburnable fossil fuel reserves in
emissions. Allen et al. [16] estimate that if total emissions between a low carbon scenario, the next step is to evaluate the overall potential
1750 and 2500 are 3670 GtCO2, then the most likely peak warming will carbon emissions within these reserves, and compare this with the
be 2 °C. However, half of the emissions have already been released to global carbon budget. The extent of reserves has been reviewed by
the atmosphere since 1750. Therefore, this would mean a carbon Meinshausen et al. [15], who state that the mid-estimate from the lit-
budget of about 1835 GtCO2 in 2009, when the paper was published. erature could produce 2800 gigatonnes (Gt) of CO2 emissions in a
Finally, it is clear that the global carbon budget is being rapidly scenario of unabated combustion, with an 80%-uncertainty range of
exhausted. Over the period 2002 to 2011, the global fossil fuel, cement 2541 to 3089 GtCO2. Reserve estimates have also been reported by
and land use change CO2 emissions were approximately 34 GtCO2 per McCollum et al. [39], which summarised conventional and unconven-
year [25]. Therefore the global carbon budget for temperature rise to tional fuel estimates. This reported a lower estimate of 3683 GtCO2,
remain below 2 °C is likely to be exhausted before 2050 unless action is which corresponds reasonably to that reported by McGlade and Ekins
taken quickly. [40] (3613 GtCO2). McCollum also presented an upper estimate of 7118
GtCO2. Clearly there is great uncertainty regarding estimates of global
2.2. Classification and estimation of fossil fuel reserves and resources fossil fuel reserves, particularly where as-yet undiscovered reserves are
included.
2.2.1. Classification Reserves and resources estimates by fossil fuel type have been
The methodology for determining fossil fuel reserves is a contested summarised in Table 2. Three different units have been reported to
subject. One of the first attempts to classify them is represented by the represent the amount of reserves and resources: gigatonnes (Gt), their
McKelvey box, which classifies resources as undiscovered, discovered energetic content (EJ) and the amount of CO2 they would release to the

Table 1
Global emissions budgets from selected sources.
Budget (Gt) Gases Scope Timeframe Probability statement Model Reference

886 CO2 fossil sources, land use change 2000–2049 20% chance of exceeding 2 °C MAGICC 6.0 [15]
1000 CO2 fossil sources, land use change 2000–2049 25% chance of exceeding 2 °C MAGICC 6.0 [15]
1437 CO2 fossil sources, land use change 2000–2049 50% chance of exceeding 2 °C MAGICC 6.0 [15]
1356 Kyoto gases fossil sources, land use change 2000–2049 20% chance of exceeding 2 °C MAGICC 6.0 [15]
1500 Kyoto gases fossil sources, land use change 2000–2049 26% chance of exceeding 2 °C MAGICC 6.0 [15]
1678 Kyoto gases fossil sources, land use change 2000–2049 33% chance of exceeding 2 °C MAGICC 6.0 [15]
2000 Kyoto gases fossil sources, land use change 2000–2049 50% chance of exceeding 2 °C MAGICC 6.0 [15]
3670 CO2 fossil sources, land use change 1750–2500 50% chance of exceeding 2 °C (according to [34]) HadSCCCM1 [16]
1635–1752a Kyoto gases fossil sources, land use change 2000–2050 50% chance of exceeding 2 °C (low aerosol scenario) SiMCaP EQW and MAGICC [23]
1631–1897 Kyoto gases fossil sources, land use change 2000–2050 50% chance of exceeding 2 °C (high aerosol scenario) SiMCaP EQW and MAGICC [23]

a
Note that these budgets required global emissions peak between 2014 and 2016, which is now accepted to be impossible.

63
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

Table 2 having a more relaxed temperature rise target increases the carbon
Estimation of reserves and resources of oil, gas and coal [41–47]. budget by 2–16% in 2020 and by 38–100% in 2035.
Fossil fuel Gigatonnes (Gt) Exajoules (EJ) Carbon (GtCO2) Cumulative emission budgets have been reported in Table 5 for
type temperature rise targets between 1.5 °C and 4 °C. The most stringent
target (1.5 °C) would reduce the carbon budget by about 300 GtCO2 in
Reserves Oil 219–240 9264–10145 679–744
2050 and by 330–370 GtCO2 in 2100, while the more relaxed targets
Resources Oil 334–847 14128–35845 1036–2627
Reserves Gas 125–155 6016–7461 338–453
would increase the carbon budget by 1610–1790 GtCO2 (3 °C) to
Resources Gas 427–540 20518–25921 1151–1454 2660–3440 GtCO2 (4 °C) in 2100. Carbon budgets having different
Reserves Coal 892–1004 25141–28313 2378–2678 likelihoods to meet their temperature targets should not be compared
Resources Coal 21208–22090 598066–622924 56577–58929 directly. Therefore these budget extensions represent only an indication
Reserves Total 1236–1399 40421–45919 3395–3876
of the sensitivity of the budget to various temperature targets.
Resources Total 21969–23477 632712–684690 58764–63010

2.3.2. Circumstances that could make unburnable carbon a reality

atmosphere if burned unabated (GtCO2). According to the values re- In order for fossil fuel reserves to become uneconomic or otherwise
ported in the table, the overall amount of reserves (including oil, gas inaccessible, some important developments would be needed in the
and coal) is equivalent to 3395–3876 GtCO2. next decade. The three key developments would include a potent global
agreement to mitigate climate change; the implementation of an ef-
2.3. Unburnable carbon fective policy, regulatory and market mechanism; the adoption of
technology approaches avoiding or limiting emissions. The agreement
Considering the range of carbon budgets and the extent of fossil fuel reached during COP21 in 2015 [18] indicates that it is possible to
reserves discussed above, it is apparent that not all of the reserves can achieve a potent global agreement on climate change, however further
be converted to CO2 and released to the atmosphere, if the world is to more ambitious binding commitments will be required. Implementing
avoid temperature rise greater than 2 °C. In this context, the term effective policy, regulatory and market mechanisms at national and
‘stranded assets’ or ‘unburnable carbon’ has been used to indicate any international levels in order to meet the agreed commitments would be
reserves surplus greater than a given carbon budget. Therefore, this likely to include carbon trading or taxation mechanisms similar to those
term refers to the amount of fossil fuel that cannot be burnt in a miti- already included in the ETS [49]. Finally, technological approaches that
gated climate change scenario. avoid or limit the emissions associated with the use of fossil fuels (e.g.
Unburnable carbon has been recently investigated by a number of CCS) would not be commercialised, or proven uneconomic or otherwise
sources, some of which have been reported in Table 3, summarising unacceptable relative to other means to reduce global emissions, such
overall reserves and unburnable and burnable carbon for different as renewable energy technologies.
timeframes. In all the reported references, unburnable carbon is be- Some sources have confirmed the possibility of unburnable carbon
tween 49% and 80% of overall reserves. A prominent example is the becoming a reality [41,50,51], while others have denied the ‘carbon
World Energy Outlook 2012 [41], which estimates overall reserves to bubble’ as a real problem, such as Mayer and Brinker [37]. While cli-
be equal to 2860 GtCO2. Without CCS, less than a third (i.e. less than mate change is generally acknowledged by oil and gas companies [52],
953 GtCO2) can be burnt in the 2DS. This finding is based on the IEA their position on the ‘carbon bubble’ is cautious and highlights how the
assessment of global carbon reserves, measured as the potential CO2 outcome depends on many factors and is therefore difficult to predict.
emissions from proven fossil fuel reserves. Almost two-thirds of these This paper has distinguished between the concepts of ‘carbon bubble’
carbon reserves are related to coal, 22% to oil and 15% to gas. Although being a financial issue, and unburnable carbon being a technological
IEA considers CCS a key option to mitigate CO2 emissions, it also issue. However the two issues are clearly linked and in the grey lit-
highlights the uncertainty regarding its pace of deployment. erature there are controversial opinions regarding the likelihood of
unburnable carbon becoming a major issue in the global energy system.

2.3.1. Sensitivity to temperature rise targets

The amount of burnable carbon (i.e. the carbon budget) reported in 3. Barriers to CCS development
Table 3 shows a wide variability which depends on factors such as the
reference, the modelling methodology and assumptions, the timeframe Many abatement technologies affect the use of fossil fuels. The
under analysis and the temperature rise target. range of options is large, and includes both direct approaches, affecting
The 2 °C target has received a lot of attention since it was introduced the use of fossil fuels or their emissions into the atmosphere (e.g. en-
as an EU climate target back in 1996 [48]. However, other targets have hanced energy efficiency and conservation; replacement of coal by
been taken into account as well, which were more or less stringent with natural gas; greater use of nuclear power; carbon capture, utilisation
respect to the 2 °C target. For example, at COP21 (Conference Of the and storage) [53,54]; and indirect approaches, increasing the re-
Parties), the conference “invites the Intergovernmental Panel on Cli- maining carbon budget (such as the development of mass market re-
mate Change to provide a special report in 2018 on the impacts of newable energy technologies, afforestation and reforestation). In
global warming of 1.5 °C above pre-industrial levels and related global
greenhouse gas emission pathways” [18] (this report is currently under Table 3
review). While the most stringent target of 1.5 °C was already requested Estimates of unburnable and burnable carbon from selected sources.
in 2008 by the Alliance of Small Island States and the Least Developed Unburnable Burnable Overall Timeframe Reference
Country group [15], other less stringent targets include temperature carbon (GtCO2) carbon remaining
rises up to 5.3 °C. The analysis of scenarios that would bring about si- (GtCO2) reserves
(GtCO2)
milar targets are motivated by the desire to show what would happen
without an emission reduction framework in place. 1360 1440 2800 2000–2050 [15]
Table 4 shows four different scenarios proposed by IEA [41], which "more than less than 2860 until 2050 [41]
correspond to four different temperature rise targets (with a probability 2/3" > 1907 1/3 < 953
2230 565 2795 2010–2050 [36]
of 50% to meet the target). CO2 emissions have been reported for the
1960 900 2860 2013–2049 [4]
years 2020 and 2035, showing how a reduction in global emissions 2565 1049 3613 until 2050 [40]
could still bring the temperature rise to 3 °C. According to the IEA [41],

64
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

Table 4 barriers, including the lack of market mechanisms and incentives, fewer
IEA scenarios and corresponding CO2 emissions for different temperature rise effective mechanisms to penalise major CO2 emitting sources, in-
targets (modified from Ref. [41]). adequate legal framework allowing transport and storage (both inland
Scenario Temperature rise CO2 emissions CO2 emissions and offshore) and public awareness and perception [53,63,64].
target (⁰C) (Gt, 2020) (Gt, 2035) At the current CO2 capture rate (which represents the percentage of
CO2 emitted by the process that will be captured and ultimately se-
450 Scenario (or 2 31.4 22.1
questered), no major purely technological barriers exist for capture,
2DS)
Efficient World 3 32 30.5 transport and storage of CO2. Indeed, CO2 separation and reinjection is
Scenario (EWS) a common feature of regular oil and gas industry operations. At the
New Policies 3.6 34.6 37 same time, the cost of capturing CO2 in a non-regulated market is
Scenario (NPS)
preventing progress.
Current Policy 5.3 36.3 44.1
Scenario (CPS)
Regarding location and capacity of storage sites, the main factors to
take into account include cumulative capacity of carbon storage, rates
of release and uptake, connection from source to store and climate
Table 5 impact of storage timescale [65].
Fossil fuel carbon budget for different maximum temperature rises [4] IPCC [8].
Timeframes: 2011–2050; 2011–2100. 3.1. Cost of CCS
Fossil fuel carbon budget (GtCO2)
Cost of CCS has been identified as the major challenge preventing
Temperature target Until 2050 Until 2100 Probability (%) the widespread adoption of this technology, and has been investigated
(°C)a (GtCO2) (GtCO2) more in detail in section 4. Estimating actual CCS cost and expressing it
in a clear way is challenging. This is mainly due to lack of empirical
1.5 550–1300 630–1180 14–51
2 860–1600 960–1550 39–68 data (currently, in the power sector there are only two full scale CCS
3 1310–1750 2570–3340 57–74 plant in operation [11,66]), difficulty in choosing the baseline when
4 1570–1940 3620–4990 61–86 comparing different CCS plants, a variety of currencies and currency
a
base years in the reported literature, cost differences due to unavail-
Relative to 1850–1900.
ability of transport and storage infrastructure and a variety of processes,
operating conditions and capture processes. Section 4 reports on how
extremely emissions constrained scenarios (e.g. net zero emissions in
the cost of CCS can be expressed and which values have been estimated
the second half of this century as put forward in COP21), indirect
in the literature. Costs have been reported in $2015 by converting
measures will be ineffective simply because there will be no carbon
single currencies into US dollars and then taking inflation into account.
budget left to open up.
The results of the analysis show a great variability among sources, with
Carbon capture and storage refers to a process that separates CO2
a lack of data for specific processes or capture technologies, and iden-
from a gas stream and stores it underground. CCS can be applied to
tify the capture step as the most expensive step of the CCS chain.
power generation and industrial facilities and includes three main steps
which are the separation of CO2 from the gas stream, its compression
3.2. Geo-storage capacity
and transportation (via pipeline or shipping) and its storage in a sui-
table geological site (e.g. saline aquifers, depleted oil and gas re-
Global demand for CO2 storage in climate scenarios maintaining
servoirs). CCS is categorised according to the class of capture process
CO2 concentrations at 400–500 ppm are estimated at an accumulated
(post-combustion, pre-combustion, and oxy-combustion) and type of
store of 600–2000 GtCO2 by 2100 [67–69]. A number of studies have
separation technology (absorption, adsorption, membranes, cryogenic
compiled regional estimates of CO2 storage resources to suggest that
distillation, gas hydrates, and chemical looping) [54–56]. While post-
cumulative resources to be in the range of 10,000–30,000 GtCO2 in-
combustion and pre-combustion capture technologies are widely used
cluding 1000 Gt in depleted oil and gas reservoirs [67]. This suggests an
(between TRL 1 and 5), oxy-combustion capture is still under devel-
abundance of storage capacity relative to demand over the century.
opment and not yet commercial, while pre-combustion is likely to be
The regional studies vary widely in their approach to estimation
“decades away from commercial reality” [57].
[70,71]. More recently, efforts by the Society of Petroleum Engineers
Carbon capture and storage can be integrated in processes which
and the United Nations Economic Commission for Europe have devel-
therefore can be classified as carbon-positive, near carbon-neutral or
oped storage resource classification systems that will allow for com-
carbon-negative. Carbon-positive processes still emit CO2 to the atmo-
parable regional estimates to be made, with classifications indicative of
sphere, while near-carbon neutral do not and carbon-negative process
the certainty around commercial viability [72,73]. The global storage
reduce the amount of CO2 which is already in the atmosphere [58]. An
availability estimates are classified as prospective under these schemes,
example of carbon-negative processes is Bio-Energy with CCS (BECCS)
with little to no physical reservoir characterisation performed for the
processes, which usually include either electricity [59] or biofuel pro-
vast majority of potential locations. This leaves a high degree of un-
duction [60]. BECCS are part of a class of technologies known as Ne-
certainty as to their ultimate potential. The development of scores of
gative Emission Technologies (NETs), which also include reforestation
projects will be required before this uncertainty can be significantly
and afforestation, various forms of geo-engineering, Carbon Dioxide
reduced.
Removal (CDR) such as CO2 capture from the air and ocean fertilization
In saline aquifers (i.e. fields without hydrocarbon production) re-
[61]. BECCS and reforestation would arguably be the most attractive
servoir pressurisation will limit the accessible CO2 geo-storage capacity
options to create negative emissions [61]. According to McLaren [62],
in the absence of pressure management strategies [68]. Recent work
NETs cannot be expected to offer an economically viable alternative to
using reservoir simulation has found that only 0.01–1% of the pore
mitigation in the coming decades. At the same time, their limited de-
volume of saline aquifers will be available for storage over timescales of
ployment (10–20 GtCO2/yr) can help reducing the overall CO2 emis-
around 50 years of injection, in the absence of brine production from
sions by 2030–2050.
the reservoir. This is due to the requirement that pressures in the re-
The review presented in this section has identified as main chal-
servoir remain below that which would induce fractures or reactivate
lenges to the uptake up of CCS its cost and energy penalty, followed by
faults in a sealing caprock. However, the first generation of storage
location and capacity of storage sites. There are several non-technical
capacity is underpinned by the potential use of existing oil and gas

65
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

fields, and high quality saline aquifer reservoirs. As a result, reservoir 3.3. Source-sink matching
pressurisation is unlikely to present major barriers for the first gen-
eration of commercial CO2 storage. Some studies have evaluated the impact of a potential limit on
Only a few studies evaluate the impact of a potential limit on sto- storage capacity on the deployment of CCS in integrated assessment
rage capacity on the deployment of CCS in integrated assessment models [75,77]. Koelbl et al. [77] addresses the issue of regional dis-
models [74–77]. In Koelbl et al. [76] the varying levels of deployment tribution of storage. In their study, storage supply was found to be
of CCS in 12 integrated assessment models were assessed against sev- limiting in China, Japan, and South Korea. Storage capacity in Japan
eral assumptions, including the existence of global and regional capa- and South Korea is highly uncertain with some estimating significant
city constraints, which ranged from 3500 to 20,000 Gt. The maximum resources offshore, particularly in Japan that would provide sufficient
cumulative storage demand was 3000 GtCO2 by 2100. Because the supply for at least decades [84]. In the case of China, the model appears
limiting capacity was not reached, the varying levels of deployment in to use values for storage capacity a factor of three less than those re-
the models were not correlated to the total CO2 storage supply. A ported in the source data of Dahowski et al. [79] and CCS is being
sensitivity study of one model in Koelbl et al. [77] also showed that the actively pursued as a large scale mitigation technology [85].
deployment of CO2 storage until 2050 was not sensitive to a regional In general, where regional storage supply estimates are most de-
storage capacity estimates ranging from 4500 to 10,000 GtCO2. The veloped (e.g. North America, Europe including Scandinavia, Brazil),
primary reason was again because the capacity in most regions was not source-sink matching shows that CCS will not be constrained by local
approached by 2050. On the other hand, the study found that storage availability of storage resources. Outside of these areas, storage avail-
could be limited beyond 100 years of full scale deployment should there ability is highly uncertain, although the global distribution of sedi-
be significant uptake of CCS. mentary basins is such that it is possible there will be few locations
Keppo and van der Zwaan [75] analysed the impact of more severe where local storage availability will be a limiting factor.
constraints on CO2 storage capacity until 2100, comparing a baseline
scenario with a pessimistic scenario where capacity is limited to half 3.4. Supply chain and building rate
that available in depleted oil and gas fields alone. This corresponds to a
reduction of global capacity from approximately 10,000 to 500 GtCO2 In the literature, the rate of technology deployment and cost re-
(when comparing hydrocarbon and non-hydrocarbon storage re- duction of CCS has been compared to development timescales in the oil
sources). By 2100 CCS deployment is very limited due to the capacity and gas industry (e.g. 3–5 years for the build-up of a giant gas field,
constraints. However, the early deployment of CCS until 2050, prior to according to Söderbergh et al. [86]), and also to the more recent ex-
the approach of capacity constraints, are mostly unaffected. Implicit in perience of implementation of post-combustion capture of sulphur
this is that volumetric estimates of global storage capacity are only an oxides and nitrogen oxides at coal-fired power plants based in the US
order of magnitude from levels where the deployment over the next [57].
century would be affected. In 2012, IEAGHG commissioned a study on potential supply and
An estimated 1000 Gt of storage capacity is available in oil and gas capacity constraints associated with equipment for CCS plants [87]. The
reservoirs alone. The analysis of integrated assessment models in Koelbl study focused on the global scale and included the full CCS chain
et al. [76] showed that from 2010 to 2050 between 100 and 500 Gt of (capture, transport and storage) but excluded the power or industry
storage demand would be consistent with a 2 °C pathway. This suggests equipment. Part of the purpose of this study was to understand if the
that there will be few storage capacity limits to the first generation of CCS roadmap proposed by IEA [2] could meet major barriers due to
commercial CCS deployment, even under scenarios of high demand for supply or capacity constraints. The results of the study have identified
CCS, as all of the demand can be met with very low cost storage options, as major potential supply chain constraints hydrogen turbines for the
such as oil and gas fields. capture step, pipelines for the transport step, availability of geo-en-
Integrated assessment models incorporate potential storage cost gineers and drilling rigs for the storage step and finally availability of
limitations through a set of rules that generally ignore the issues of petroleum engineers across the full CCS chain. The conclusion of the
pressurisation and pressure management. The most flexible storage cost study did not identify any insurmountable obstacles to the deployment
supply curves have been developed by Dooley and Friedman [78] for of CCS as suggested by IEA [2]. However, the construction rate for CCS
North America, and by Dahowski et al. [79] for China. A commonly applied to the power industry would be lower than historical power
used regionally distributed supply cost curve for the rest of the globe plant construction rates. In addition, the suggested deployment of CCS
was developed by Hendriks and Graus [80]. Notably, these datasets in the industrial sectors (capture of 65% of current emissions by 2050)
were developed prior to the work done by Birkholzer and Zhou [81], has been considered optimistic. Overall, the most significant risk is
demonstrating the first order impacts of regional pressure build-up on represented by the competition between CCS and the oil and gas sector
storage capacity. Key capacity constraints built into the supply curves for experienced staff and drilling equipment necessary for exploration
include total capacity, and the requirement that supply must be avail- activities. Similar issues have been identified in a study focussing on the
able for a particular source for a minimum of 10 years. Pressurisation is UK market [88].
partially taken into account by limiting the amount of CO2 that can be Similar challenges have been discussed in an interview with CCS
injected into a single well – a proxy for the risk of near wellbore frac- developers during the course of this project, where the following issues
turing. The impact of this limit, however, is the construction of a new were cited to be important when considering barriers to CCS: geological
well in the storage basin when costs are justified. While local injectivity appraisal and power station build; availability of skilled labour; and
may be dealt with in this way, it is clear that regional pressurisation of regulatory shortfalls. The geological appraisal of a store takes 3–4
the storage resource may not [82]. Thus, an additional constraint years, while a power station build takes 3–4 years for gas turbines and
should be built into the models in which regional pressurisation may 5–6 years for solid-fuelled systems (these timelines are representative
trigger the deployment of pressure management strategies. Pressure for a plant based in EU or USA but could vary for developing countries
management and the handling of waste brine are longstanding practices such as China). Therefore, if appraisal and power station build are si-
in the oil and gas industry. As such, costs estimates suitable for use in multaneous, the CCS aspect may be on the critical path. But if the power
integrated assessment models should be readily available from existing station build is dependent on the suitability of the store, appraisal may
literature [83], or by interviews with relevant oilfield operators. need to proceed prior to power station build. However, if national CO2
transportation infrastructure were already present, any dependency
would be largely eliminated. The availability of sufficient skilled labour
could represent a bottleneck in the long term. However in the short

66
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

term, there may be a higher workforce availability due to the recently has reviewed the analyses on public perception. According to Seigo
depressed oil and gas prices resulting in a number of job losses [89]. As et al. [95], the public knows that climate change exists, but is unsure
an example, the White Rose project in the UK was estimated to need on- about what causes it and the various mitigation options. In particular,
average 4000–5000 people over approximately five years, with a peak estimates of the emissions reduction needed are underestimated while
of 9000 people. Finally, at present the regulatory environment for CCS the role of renewable energies is overestimated. The risk perception
infrastructure is not well developed, leading to uncertainty regarding focuses on sustainability of CCS, leakages and overpressurisation of the
development timeframes and price models. storage sites.
The process of the 3–4 year appraisal period for a CCS site is not Furthermore, a further concern is that public investments in CCS
new, and is already regularly undertaken by the oil and gas industry. would reduce the budget for renewable alternatives [95–97]. A survey
Overall, construction-related barriers to CCS development appear to be with 60 participants from Pittsburgh (Pennsylvania, US) on preferences
a minor issue, meaning that the risk is largely non-technical in nature, for emission reductions reported that the most preferred portfolio in-
which could mean that financial environments and/or regulation will cluded energy efficiency, followed by nuclear power, integrated gasi-
change significantly over the construction period. fication combined-cycle coal with CCS and wind [98]. Therefore, these
studies report a moderate public acceptance of CCS as long as it is part
3.5. Policy, regulations and market of a wider portfolio of carbon emission reduction options.
The importance of informing the public in an adequate and neutral
As previously mentioned, the cost of CCS has been identified as a way is highlighted by Seigo et al. [95] and other publications. For ex-
major barrier to its wider adoption. At the moment there is no market ample van Alphen et al. [96] explored the effect of the media (in par-
for CCS and this is mainly because a plant with CCS will always be more ticular the Dutch press) on public perception of CCS and reported that
expensive (in terms of capital and operating costs) than the same plant media and stakeholders (government, industry, NGOs) share the same
without CCS. Enhanced oil and gas recovery options represent the only concerns towards CCS, including the previously cited sustainability,
exception and in fact have been employed for many decades. Without leakages from the storage site and reduced investments in renewable
effective mechanisms to underpin uptake, the deployment of CCS to a energy. At the same time a lack of knowledge seems in some cases to be
level that would be adequate to meet the climate change targets will responsible for decreased support, and in others, for increased risk and
remain implausible. reduced benefit perception [97]. This suggests a need for a closer col-
Possible policy options include carbon trading, such as the EU laboration between experts from engineering and communications in
Emissions Trading System (EU ETS) mechanism, or carbon taxation; order to inform the public.
targeted investment support, especially needed for the initial capital
costs; feed-in schemes, which guarantee a fixed fee in order to com- 4. CCS cost estimates
pensate for the higher costs of the project when compared to conven-
tional alternatives; a carbon floor price; low-carbon portfolio standard 4.1. Cost metrics for CCS plants
with tradable certificates; minimum standards, such as a CCS obligation
for new installations after 2020 [90]. Various metrics have been suggested to estimate or measure the cost
Some low-carbon initiatives that could encourage CCS include the of carbon capture and storage and they depend on the system under
Clean Energy Future Package in Australia, the Regional Greenhouse Gas analysis and on the purpose of the analysis itself.
Initiative in US, the Western Climate Initiative (British Columbia, The cost of CCS is often expressed as an energy or efficiency penalty,
Manitoba, Ontario, Quebec and California), the Framework Act on Low where the performance of a plant without CCS is compared with the
Carbon and Green Growth in Korea, the General Law on Climate performance of the same plant with CCS. Energy penalty applies to the
Change in Mexico, the National Policy on Climate Change in Brazil, just power generation sector while efficiency penalty can be used for both
to cite a few [91]. According to Lohwasser and Madlener [92], the ef- power and industrial sectors. Energy penalty and efficiency penalty
fectiveness of policies promoting ’learning-by-doing’ (i.e. cumulative have been expressed by means of the following equations:
deployment) or ’learning-by-searching’ (i.e. cumulative R&D efforts)
depends on their spending levels. At lower policy costs (up to US$ Energy penalty
587 M), both methods are about equally effective, while at higher Power output without CCS − Power output with CCS ⎞
spending levels policies promoting cumulative deployment are more = 100 ⎜⎛ ⎟

⎝ Power output without CCS ⎠ (1)

effective than those promoting R&D efforts.
In May 2015, some of the major oil and gas companies wrote to the Efficiency penalty = Efficiency without CCS (%) − Efficiency with CCS (%)
UNFCCC secretariat asking for “clear, stable, long-term, ambitious
(2)
policy frameworks”, stating that a price on carbon “should be a key
elements of these frameworks” [93]. This would encourage reduction of While Equation (1) gives “the proportional loss in power output
CO2 emissions by means of increased efficiency, fossil fuel switch (from capacity with reference to a base case without capture”, Equation (2)
coal to gas) and investment in CCS, renewable energy, smart buildings shows “the decrease in plant efficiency percentage points due to cap-
and grids and new mobility business models. Moreover, a price on ture” [99].
carbon in such a framework would avoid “uncertainty about investment For the power sector, the Levelized Cost Of Electricity (LCOE) is
and disparities in the impact of policy on business”. often used ($/MWh). LCOE is often labelled as increased Cost Of
Electricity (COE), expressed as [100]:
3.6. Public acceptance (TCC )(FCF ) + (FOM )
COE = + VOM + (HR)(FC )
(CF )(8766)(MW ) (3)
Public acceptance has a key role in the deployment of carbon cap-
ture and storage, locally and globally. There is no general model able to where COE=Cost Of Electricity generation ($/MWh), TCC = Total
explain public acceptance of new technology, however a framework has Capital Costs ($), FCF=Fixed Charge Factor (fraction/year), FOM=-
been proposed that includes a range of different factors affecting ac- Fixed Operating and Maintenance costs ($/year), VOM=Variable non-
ceptance. These factors include acceptance and attitude, knowledge, fuel Operating and Maintenance costs ($/MWh), HR = net power plant
experience, trust, fairness, affect, perceived costs, risk and benefits, Heat Rate (MJ/MWh), FC = unit Fuel Cost ($/MJ), CF = plant Capacity
outcome efficacy and problem perception [94]. Factor (fraction), 8766 = total hours in an average year and MW = net
This framework has been adapted to CCS by Seigo et al. [95], who plant capacity (MW).

67
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

The cost of CCS can also be expressed as a cost of carbon ($/tCO2), chemical looping combustion. The reported costs present a wide
which may refer to the CO2 avoided, captured or abated, as reported in variety, ranging from 20 $2015/tCO2 (refineries and natural gas pro-
equations (4)–(6) [5,100]. The equations refer to the power generation cessing) to 100 $2015/tCO2 (cement production). Post-combustion
sector, where COE is the cost of electricity generation ($/MWh) while with amine separation presents the highest maximum cost (110 $2015/
NPV is the Net Present Value of the specified scenarios. The subscripts tCO2) while storage via Enhanced Oil Recovery/Enhanced Gas Recover
“CCS”, “ref” and “cc” refer respectively to plants with CCS, plant (EOR/EGS) has a smaller range characterised by a higher minimum cost
without CCS and to the capture step only. but a lower maximum cost when compared to CCS storage.
The cost of avoided CO2 is inclusive of capture, transport and sto- Table 8 reports the cost of CO2 transport depending on the pipeline
rage steps, and therefore represents the full CCS chain. At the same capacity (MtCO2/yr) and location (onshore or offshore). As expected,
time, it heavily depends on the baseline (“ref”) that is used for the the lowest cost of transport refers to the onshore pipelines having
comparison, which may or may not be the same type of plant as “CCS”. higher capacity (1.3–2.2 $2015/tCO2/250 km with capacity 30 MtCO2/
The cost of captured CO2 refers only to the capture step, without taking yr).
into account transport or storage. Finally, the cost of abated (or re- Table 9 reports the cost of storage depending on the storage site
duced) CO2 refers to multiple CO2 emission sources and therefore has (depleted oil and gas fields or saline formations), the location (onshore
been suggested as more appropriate for Integrated Assessment Models or offshore) and the possibility to reuse already existing oil and gas
as it enables comparison of different energy systems. The subscripts wells. The cheaper storage solution corresponds to the onshore depleted
“ref” and “low-C” refer to values respectively before and after a speci- oil and gas fields, with a small positive margin given by reusing already
fied carbon reduction scenario [100]. existing wells.
Table 10 reports the cost of avoided CO2 depending respectively on
(COE )CCS − (COE )ref
Cost of avoided CO2 = process plant, capture technology and storage solution. This cost in-
(tCO2/ MWh)ref − (tCO2/ MWh)CCS (4) cludes capture, transport and storage steps and depends heavily on the
(COE )CC − (COE )ref selection of the reference plant. Therefore, a wide variability in the cost
Cost of captured CO2 = is observed depending on the type of process plant.
(tCO2/ MWh)captured (5)

(NPV )low − C − (NPV )ref 4.4. Cost of electricity

Cost of abated CO2 =
(tCO2)ref − (tCO2)low − C (6)
Table 11 reports the cost of electricity ($2015/MWh) depending on
Finally the cost of CCS can be reported in the literature in a more process plant, capture technology and storage solution, respectively.
traditional way which includes estimation of capital and operating The lowest cost corresponds to gas-fired power generation (54 $2015/
costs. MWh) while the highest cost corresponds to Integrated Gasification
Combined Cycle (278 $2015/MWh). The cost of electricity does not
4.2. Energy and efficiency penalty vary with different capture technologies (111–265 $2015/MWh) but
may be much higher when CCS is adopted instead of EOR or EGR for
According to Clark and Herzog [6], the major barrier to CCS in the the storage of CO2.
power industry is the high capital cost and energy penalty compared to
traditional fossil fuel fired generators. As an example, the net efficiency 4.5. Capital and operating costs
penalty (LHV) of CCS for coal-fired power generation is about 10%
[101]. This penalty does not depend on the type of power plant but Capital and operating costs are reported in Table 12 for coal-fired
rather on the capture process, which contributes to about two thirds of and gas-fired power generation. Capital costs are expressed in $2015/
the overall energy penalty. kWel.net and represent capital expenditure (CAPEX) costs or overnight
In a study by Hammond et al. [102], the energy penalty of a pul- capital costs while operating costs are either fixed ($2015/kW-yr) or
verised–coal (PC) power plant is about 16% and it is higher than the variable ($2015/MWh). The operating fixed costs appear to be much
energy penalty associated with integrated gasification combined cycle higher when CCS is applied to coal-fired power generation (69–84
(about 9%) and Natural Gas Combined Cycle (NGCC) plants (about 7%) $2015/kW-yr) rather than gas fired power generation (around 8
when combined with carbon capture and storage. $2015/kW-yr).
According to Page et al. [99], energy penalty values for PC plants
with capture range from 15% to 28% while efficiency penalties range 4.6. Discussion
from 8% to 15.4%. For Natural Gas Combined Cycle (NGCC) plants, the
energy penalty is around 15–16% while the efficiency penalty varies The cost of CCS reported shows a great variability among sources,
between 6% and 11.3%. Finally, the energy penalty for IGCC plants with a lack of data for specific processes or capture technologies. The
varies from 4.9% to 20% while the efficiency penalty ranges from 5% to capture step is definitely the most expensive step of the CCS chain, with
10.3% (Table 6). a cost of carbon equivalent to $2015/tCO2 20–110. Transport cost
ranges between 1.3 and 15.1 $2015/tCO2/250 km, depending on lo-
4.3. Capture, transport and storage costs cation and length of the pipeline. Storage cost depends on the type of
storage site and the possible reuse of existing facilities and is between
The cost of captured CO2 refers only to the capture step and does not
include transport or storage costs. However, various sources report Table 6
different capture costs depending on the type of storage site. This is Energy and efficiency penalty for Pulverised Coal (PC), Natural Gas Combined
because the cost of captured CO2 depends on the operating conditions Cycle (NGCC) and Integrated Gasification Combined Cycle (IGCC) power plants
of the capture step, which in turn depend on how carbon dioxide is [99,101,102].
transported to the storage site, together with its location and type. Pulverised Coal Natural Gas Integrated Gasification
Table 7 reports the cost of captured CO2 depending respectively on Combined Cycle Combined Cycle
process plant, capture technology and storage solution. The main re-
Energy penalty 15–28% 15–16% 4.9–20%
ported industries include power generation, refineries, iron and steel
Efficiency 8–15.4% 6–11.3% 5–10.3%
and cement production. The main reported capture technologies in- penalty
clude post-combustion separation with amine, oxy-combustion and

68
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

Table 7 Table 10
Cost of captured CO2 for different process plants, capture technologies and Cost of avoided CO2 for different process plants, capture technologies and
storage solutions. storage solutions.
Cost ($2015/tCO2) References Cost ($2015/tCO2) References

Min Max Min Max

Process plant Process plant

Coal-fired power 41 62 [103–106] Coal-fired power 24 110 [103–106,109–112]
Gas-fired power 52 100 [103,107] Gas-fired power 67 115 [103,107,110–112]
Iron and steel 57 69 [106,108] Iron and steel 52 120 [103,106,113]
Refineries and natural gas processing 20 79 [106,108] Refineries 6 160 [103,106,113]
Cement production 35 110 [106,108] Pulp and paper 47 93 [113]
Natural gas combined cycle 75 95 [104–106] Cement production 27 146 [103,106,113]
Oxyfuel combustion 45 50 [106] NGCC 10 146 [104–106,109]
Capture technology Oxyfuel combustion 48 99 [106,109]
Post-combustion (amine) 50 110 [108] IGCC 3 140 [106,109]
Chemical looping 35 52 [108] Chemicals + bio or synfuel 20 111 [103,113]
Oxy-combustion 45 66 [106,108] Capture technology
Storage Post-combustion (amine) 63 87 [110]
CCS 20 110 [103–108] Pre-combustion 47 60 [110]
EOR/EGR 52 62 [107] Storage
CCS 20 113 [106]
EOR/EGR 71 84 [107]
Table 8
Cost of CO2 transport for onshore and offshore pipelines with different capa-
cities (modified from Ref. [109]). Table 11
Cost of electricity for different process plants, capture technologies and storage
Methods Capacity (MtCO2/yr) Transport cost ($2015/tCO2/250 km) solutions.
Min Max Cost ($2015/ References
MWh)
Onshore pipelines 3 4.4 11.1
10 2.2 3.8 Min Max
30 1.3 2.2
Offshore pipelines 3 7.3 15.1 Process plant
10 3.5 4.9 Coal-fired power 59 167 [57,104,105,110–112,114–116]
30 1.9 2.4 Gas-fired power 54 231 [104,107,110–112,114,115]
NGCC 61 102 [105,115]
Oxyfuel combustion 111 271 [110,114,117]
IGCC 111 278 [57,114,117]
Table 9
Capture technology
Cost of CO2 storage for various storage sites (modified from Ref. [109]). Post-combustion 111 266 [114] [110],
Properties Storage cost ($2015/tCO2) (amine)
Oxy-combustion 111 265 [114] [110],
Min Max Storage
CCS 59 271 [57,104,105,107,110–112,114–116]
Depleted oil and gas field – reusing wells onshore 1.6 11 EOR/EGR 68 87 [107]
Depleted oil and gas field – no reusing wells onshore 1.6 15.7
Saline formations onshore 3.1 18.8
Depleted oil and gas field – reusing wells offshore 3.1 14.1 Table 12
Depleted oil and gas field – no reusing wells 4.7 22
Capital and operating costs for coal and gas fired power plants (modified from
offshore
Refs. [104,111,112,118]).
Saline formations offshore 9.4 31.4
Process plant Capital costs Operating fixed Operating variable
($2015/kWel.net) costs ($2015/kW- costs ($2015/MWh)
1.6 and 31.4 $2015/tCO2. The overall cost of carbon for CCS can be yr)
estimated by summing the cost of carbon for capture, transport and
Coal-fired 3552–6816 69–84 9–10
storage steps. For example, for a pipeline length of 250 km, this cost power
would range between 22.9 and 156.5 $2015/tCO2. These numbers are Gas-fired 2313–5088 14–33 11–16
comparable to those reported in Table 10, which report the cost of power
carbon for the avoided CO2.
It is important to remember that the cost of CCS reported in the
academic and grey literature is based on estimations which are com- NOAK varies between 3.4% and 8.1% for the power generation sector,
paring CCS technology to other similar technologies that have been while is around 9.3% for the industrial sector (US$ per tonne of CO2
recently developed. For instance at the moment there are only two avoided). Other aspects that are important to consider when estimating
examples of full scale CCS applied to power generation [11,66] and the cost of CCS include cost uncertainties, cost baseline and limited
therefore they represent First Of A Kind (FOAK) projects. Costs related operating experience.
to FOAK projects are not representative of the costs of the technology
when it will be fully developed into Nth Of A Kind (NOAK) solutions. 5. CCS potential in the context of unburnable carbon
The transition from FOAK to NOAK takes place along a technology
learning curve, which depends on many factors including policy, reg- 5.1. Selected industrial and academic outlooks
ulations and market, as discussed in the next section. According to the
Global CCS Institute [106], the cost reduction of CCS from FOAK to The role of CCS in future energy scenarios has been analysed by
industrial and academic sources. For instance, while BP highlights the

69
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

role of gas as a cleaner fossil fuel for power generation in future pro- are part of the IPCC Fifth Assessment Report Database, decarbonisation
jections, encouraging research and development toward higher energy happens first in electricity generation, followed by industry, buildings,
efficiency routes [42], Shell and ExxonMobil explicitly mention CCS as and transport [121]. In this context, the importance of CCS is evident,
a technology able to reduce carbon emissions. While ExxonMobil says being applicable to power generation and industrial production.
that the development of CCS could be significantly limited by “eco- Moreover, BECCS and other NETs would be able to extend the 2050
nomic and practical hurdles” [119], Shell propose energy scenarios in carbon budget by 11–13% (for a 50–80% probability to remain below
which CCS plays a key role, helping to decarbonise electricity by 2060 2 °C temperature increase [122]).
and to reduce world CO2 emission to zero by 2100 [120].
Although many academic publications cover a range of aspects re- 5.3. Review of the model comparison exercise EMF27
lated to CCS, only few have explicitly investigated this technology in
the context of unburnable carbon in their projections. This is the case 5.3.1. Description of the project and models involved
for the UCL Institute for Sustainable Resources, who has released two The IPCC Fifth Assessment Report Database includes 31 models and
publications on the topic. The first publication [34] focused on the 1184 scenarios [123]. The majority of the scenarios were provided via
volumes of oil that cannot be used up to 2035, and the results estimate model inter-comparison exercises, so the outcome of various models for
that 500–600 billion barrels (Gb) of current reserves should not be the same scenarios can be compared. The Energy Modelling Forum
burnt. The lower estimate (500 Gb) excludes CCS from the energy centred at Stanford University since 1976 is one of the first major model
scenario while the higher estimate (600 Gb) assumed a widespread comparison efforts. EMF27 builds on previous model inter-comparison
adoption of this technology. When CCS is not available, the cost of exercises such as EMF19, EMF21 and EMF22 and compares 18 in-
decarbonisation increases and therefore affects the cost of CO2 emis- tegrated assessment models [124], which have also been analysed in
sions. The consequence is that oil consumption is affected as well, not the projects AMPERE2 [125] and AMPERE 3 [126].
because CCS would otherwise be applied to oil consumption but rather One of the main purposes of EMF27 is to analyse the role of tech-
because it would generate a larger carbon budget for oil consumption nology for achieving climate policy objectives. According to Kriegler
when applied to gas and coal. According to McGlade and Ekins [34], et al. [124], CCS is deployed at a substantial scale in almost all the
40–55% (with CCS-without CCS) of yet to be found deepwater re- EMF27 mitigation scenarios with full technology availability. The im-
sources should not be developed. In both technological scenarios, arctic portance of CCS is mainly due to its flexibility, which includes the
oil and most light tight oil resources remain undeveloped while un- capability of sequestering carbon dioxide from the atmosphere when
conventional oil production is generally incompatible with low CO2 applied with bioenergy [127].
energy system.
The second UCL publication on the topic of unburnable carbon [40] 5.3.2. Investigated climate and technology scenarios
considers all fuels and their geographical location. The proposed sce- The analysis presented in this paper includes all the models that
narios include three mitigation scenarios (2, 3 and 5 °C increase of were part of EMF27 that have been employed for generating the sce-
temperature) and two technology scenarios (with and without CCS). narios included in the AR5 database and that were able to run up to the
The results for the 2 °C scenario are summarised in Table 13, which year 2100 (therefore the models AIM-Enduse, DNE21+, ENV-Linkages
presents the overall reserves, divided by fossil fuel type, and the un- and Phoenix have been excluded from the analysis). It is important to
burnable/burnable carbon in the two technology scenarios. CCS enables highlight that not all the models were able to give an output for specific
use of 1% more oil, 3% more gas and 7% more coal by 2050. According scenarios. This behaviour has been taken into account as an indication
to McGlade and Ekins [40], CCS has the largest effect of any technology that the specific target was technically or economically infeasible, fol-
on cumulative fossil fuel production levels. However its effect before lowing the approach by Kriegler et al. [124]. In particular, both Kriegler
2050 is modest because of its cost, late introduction and maximum rate et al. [124] and Krey et al. [127] reported that most of the models were
of construction. not able to meet the most stringent climate target without CCS. While in
In essence, both publications from McGlade and Ekins suggest that a specific case (referring to the IMAGE model), it was reported that the
CCS makes little difference to the extent of unburnable carbon. scenario was not feasible due to the lack of sufficient alternative miti-
However, these scenarios are not the only resource that can be used to gation potential [128], for the remaining models the availability or
assess the impact of CCS on fossil fuel use. As part of the Fifth otherwise of CCS has the strongest impact on carbon prices [129] and
Assessment Report, the International Panel on Climate Change (IPCC) on the variation of mitigation costs [124,125].
made an open call to collect energy projections coming from various The scenarios selected for the analysis reported in this paper are
integrated assessment models. characterised by climate mitigation target and technological avail-
ability. The climate mitigation scenarios include the scenarios
5.2. Integrated assessment models “450 ppm” (reported in the main body of the paper) and “550 ppm”
(reported in Appendix), aiming to reach atmospheric GHG
Integrated assessment models are models that can depict scenarios
of global change related to climate change. They are inherently multi- Table 13
disciplinary, incorporating climate science, engineering and economics Unburnable reserves before 2050 for the 2 °C scenarios with and without CCS
as a minimum. They are global in geographical scope, incorporate the (modified from Ref. [40]).
century-long time horizons relevant to climate change, and cover all
Fossil fuel Overall With CCS Without CCS
sectors of the economy and land use. This very broad scope is required reserves
to adequately assess potential responses to the threat of climate change, Unburnable Burnable Unburnable Burnable
allowing modellers to capture the key interrelationship in complex
Oil (Gb) 1306 431 (33%) 875 (67%) 449 (34%) 857 (66%)
systems of energy production, climate, and economics. IAMs are natu-
Gas (Tm3) 194 95 (49%) 99 (51%) 100 (52%) 94 (48%)
rally predisposed to analyses on unburnable carbon, given their cov- Coal (Gt) 999 819 (82%) 180 (18%) 887 (89%) 112 (11%)
erage of technology options, economics and climate. Oil (GtCO2) 531 175 (33%) 356 (67%) 183 (34%) 349 (66%)
As the energy sector is the primary source of CO2 emissions, several Gas (GtCO2) 418 205 (49%) 213 (51%) 215 (51%) 202 (48%)
studies have used IAMs to estimate how the current energy system may Coal 2664 2185 (82%) 480 (18%) 2366 (89%) 299 (11%)
(GtCO2)
evolve in order to be compatible with climate change objectives. Most
Overall 3613 2565 (71%) 1049 (29%) 2764 (77%) 850 (24%)
of them suggest that CCS will be crucial to meet the 2 °C limit cost- (GtCO2)
effectively [117]. In most of the integrated modelling scenarios which

70
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

concentration at levels of respectively 450 ppm CO2e and 550 ppm atmospheric concentration until 2100. As expected, all of these sce-
CO2e by 2100 [126]. The selected technology scenarios include the full narios have approximately the same cumulative emissions of CO2, as
technology scenario (“Fulltech”), the conventional solutions scenario they all reach the same atmospheric concentration over the time period.
(“Conv”) and the scenario without CCS (“noCCS”). The full technology The shapes of the profiles are slightly different, reflecting the impact of
scenario has a full portfolio of technologies which may scaled up in the technology options and constraints on the abatement pathway chosen
future in order to meet the climate targets; in the conventional solution by the models.
scenario solar, wind and biomass potentials are limited and therefore Fig. 2 reports the projections for the captured CO2 over the time-
energy demand is met by means of conventional technologies based on frame 2005–2100. As expected, the ‘noCCS’ scenario does not capture
fossil fuel deployment in combination with CCS and/or nuclear; finally any CO2 emissions in any scenario. Both ‘Conv’ and ‘Fulltech’ reach
in the scenario without CCS carbon capture and storage never becomes very significant levels of capture and storage by both 2050 and 2100,
available [124,125,127]. and in virtually all scenarios the rate of capture is still increasing at the
end of the time horizon in 2100. Fig. 2 shows that the total level of
5.3.3. CCS modelling assumptions capture and storage achieved in the 450 ppm (i.e. more climate-con-
As part of the EMF27 project, Koelbl, et al. [76] looked at the way strained) scenario is lower than that of the 550 ppm scenario (reported
CCS was characterised in each model, focussing on assumptions such as in Appendix). This could be explained by the fact that the residual
fuel prices, baseline emissions, type of model, modelling technology emissions from CCS in the 450 ppm scenario are far more important
change and CCS modelling approach. However, none of the model as- than in the 550 ppm scenario due to the more onerous overall con-
sumptions could clearly be associated with the amount of CO2 captured. straint on emissions. This limits the usefulness of CCS in the 450 ppm
Therefore, the authors suggested that further research is needed in scenario. Moreover, comparing ‘Conv’ and ‘Fulltech’ scenarios in the
order to investigate the impact of CCS modelling parameters on the 450 ppm case, ‘Conv’ utilises CCS less than ‘Fulltech’. On first con-
simulation outcomes. According to what reported by Koelbl et al. [76], sideration, this may seem unexpected because ‘Conv’ has more con-
most of the models are not including any limitation to either storage strained access to the alternatives to CCS for decarbonisation. However,
rate or capacity, while only one model includes all the types of storage ‘Fulltech’ has greater access to biomass than ‘Conv’, meaning that
sites (on and offshore EOR, depleted gas, undepleted gas, depleted oil, ‘Fulltech’ can use BECCS to a greater extent, and therefore displays
as well as Enhanced Coal Bed Methane (ECBM) onshore, and two types greater overall use of CCS. Overall, many factors contribute to these
of aquifers). The assumptions on the availability of CCS cover today (4 outcomes including the availability of biomass/BECCS, relative costs of
models), 2020 (7 models) and 2030 (1 model). fossil fuels and biomass, technology performance and lifetimes, to name
Among the 64 references listed in the AR5 database webpage, only a few, which are important topics for further research. The residual
one reference [130] reports the marginal abatement cost of CCS. Annex emissions challenge is discussed further in section 6.
III of the IPCC report “Climate Change 2014: Mitigation of Climate
Change” [1] reports, for CCS combined with power generation, the 5.4.2. Fossil fuel consumption with and without CCS
overnight capital expenditure to be between 2000 and 4000 $2010/kW, Fig. 3 reports the total fossil fuel use for the three technology sce-
construction time around 4–5 years, fixed annual operation and main- narios for the 450 ppm scenario. There is a large variation of model
tenance cost between 13 and 58 $2010/kW and variable operation and results (reported in Appendix), and this variation increases for the
maintenance cost between 8.3 and 15 $2010/kW. timeframe 2005–2100, highlighting the increased uncertainty that
Investment costs and efficiencies for power generation combined characterises the model outputs after 2050. With regard to the top chart
with CCS have been estimated by Koelbl et al. [77] and the results have in Fig. 3, it is clear that fossil fuel use drops in all scenarios, revealing
been reported in Table 14 and Table 15 for capture and transport of the challenges faced by these energy forms over coming decades and
CO2, respectively. Data presented in Table 14 are representative of only the competition from renewable sources of energy under climate
a small subsection of potential CCS technologies, and exclude coal with change mitigation scenarios. This is in contrast with what has been
either post-combustion or oxy-combustion capture technologies. When reported by IEA [43] and also by BHP Billiton [131], who still forecast a
comparing the costs reported by Koelbl et al. [77] with the costs re- growing fossil fuel demand in the future. However, the range of model
ported in the literature (section 4), the following considerations apply. outcomes for consumption of fossil fuels is large, with some models
CCS investments costs reported in Table 14 for 2020 are lower than the indicating a stabilisation or increase of fossil use in the ‘Conv’ and
capital costs reported in the literature (Table 12) for both coal-fired ‘Fulltech’ scenarios. The range of outcomes from the models for the
($2015/kWe 1181–4942 vs. 3552–6816) and gas-fired ($2015/kWe ‘noCCS’ case are much tighter towards the end of the time horizon, and
856–2394 vs. 2313–5088) power generation. The efficiency penalties
reported in Table 14 are similar to the penalties reported in Table 6 Table 14
(6–11% for CCGT, 5–11% for IGCC). Ranges of investment costs, efficiency and efficiency loss (p.p. percentage
The transport costs reported in the literature (Table 8) varies be- points of capture efficiency loss) for power plants and capture unit (modified
tween 1.3 (onshore pipelines, capacity 30 MtCO2/yr) and 15.1 (offshore from Ref. [77]). Investment costs are expressed in $2015 per kWe.
pipelines, capacity 3 MtCO2/yr) $2015/tCO2/250 km against figures Investment costs Efficiency
reported in Table 15, where the transport costs varies between 0.5 and
42.5 $2015/tCO2/250 km. This shows that a larger range of transport 2020 2050 2020 2050
costs has been adopted in the EMF27 models. The same consideration
no CCS no CCS (%)
applies to the cost of storing CO2 in depleted oil and gas fields
(Table 16). According to Rubin et al. [109], it varies between 1.6 and IGCC Coal 914–3464 643–3300 38–52 40–58
22 $2015/tCO2 (as reported in Table 9), while according to Koelbl et al. IGCC Biomass 1416–3966 997–3780 32–50 35–54
[77] it varies between 1 and 35.1 $2015/tCO2. Negative storage costs CCGT 532–1158 432–1055 48–64 50–67

represent an income, coming from fossil fuel recovery. 2020 2050 2020 2050

5.4. CCS and unburnable carbon up to 2100 Capture Capture (p.p.)

IGCC Coal 267–1479 107–1353 4–11 3–9

5.4.1. Emissions and capture of carbon dioxide
IGCC Biomass 669–1100 333–1007 5–11 3–7
Fig. 1 reports the emissions of the three selected technology sce- CCGT 325–1236 156–1058 6–11 5–9
narios (Fulltech, Conv, noCCS) for 450 ppm CO2 equivalent

71
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

Table 15
Ranges of CO2 transport costs per distance category (modified from Ref. [77]).
Distance in < 50 50–200 200–500 500–2000 2000–∞
km

$2015/t CO2 0.06–3.9 0.13–21.96 0.83–59.78 1.95–244 7.32–263.52

$2015/t 0.002–0.16 0.001–0.18 0.002–0.17 0.001–0.2 0.002–0.09
CO2/km

fossil fuel use drops rapidly to very low levels late in the century. From
the analyses, it is possible to conclude that CCS is extremely important
for the continued use of fossil fuels in the medium to long term, with the
technology having significant impact on usage from 2030 onwards.
Gas and coal are the fuels where increased consumption is observed
through the availability of CCS. While coal has the most significant
difference between Conv and noCCS scenarios, gas is the only fossil fuel
that increases its utilisation between 2005 and 2050, and also almost
maintains its contribution in absolute terms between 2050 and 2100.
Results referring to use of single fuels have been reported in Appendix.
The results referring to the 550 ppm scenario are similar to those Fig. 1. Average global emissions of CO2 (GtCO2/yr) for the 450 ppm scenario
reported in Fig. 3 in the way they report the projections of fossil fuel across EMF27 models.
usage for the three technology scenarios. As expected, the presence of
CCS in these scenarios unlocks more fossil fuel reserves than the
450 ppm scenario, though at the expense of the climate, manifesting as
a higher probability of exceeding 2 °C peak warming. When considering
the impact of CCS on a fuel-by-fuel basis, again coal sees the greatest
gains from addition of CCS to the technology mix, and in fact becomes
the dominant fossil fuel in energy terms by 2100, almost doubling
consumption on 2005 levels. Gas also sees significant gains due to CCS,
and increases aggregate utilisation in the energy mix.
Numerical average values for fossil fuel usage in 2050 and in 2100
across EMF27 models are reported in Table 17. Values from individual
models and for each fossil fuel are presented in Appendix and show the
range of outcomes observed.
In summary, Table 18 shows the average cumulative consumption
of fossil fuels over two timeframes (2005–2050 and 2005–2100) ob-
served across the models. This consumption has been express in giga-
tonnes of equivalent carbon dioxide (GtCO2), exajoules (EJ) and per-
centages of reserves as reported by McCollum et al. [39](low estimate
value). Clearly CCS has a very significant impact on fossil fuel con-
sumption post 2050, enabling 65% of reserves to be used instead of
33% in the scenario without CCS, whilst still delivering climate targets.
Fig. 2. Average capture of CO2 (GtCO2/yr) for the 450 ppm scenario across
EMF27 models.
5.4.3. Discussion
A primary point of interest when considering the potential of CCS as
seen by integrated assessment models is an understanding of what Scenario ($140/tCO2 in most OECD countries in 2040). Therefore the
factors in the models are limiting its uptake. While the presented results cost of CCS in future is highly competitive with the other available
clearly point to the importance of CCS in underpinning the role of fossil abatement options, if climate change is to be limited to less than 2 °C. It
fuels in future low carbon energy systems, they still leave a significant is worth noting that the costs reported here are not an assumption of the
question unanswered: why is CCS not adopted in greater quantities. EMF models, but rather an output of the models (i.e. they are the
Akimoto et al. [130] suggests that the marginal cost of CCS across marginal abatement cost, which can be interpreted as the cost of the
the entire possible range of fossil fuel reserves (i.e. up to ∼4000 GtCO2) most expensive measure that was used to meet the climate target in
is less than US$100/tCO2. However, as shown in Fig. 4, the marginal each year, or alternatively as the global carbon price required to
cost of abatement produced in the 450 ppm ‘Conv’ scenario is well achieve the climate target).
above this value, indicating that the model would adopt the technology One possible explanation for this phenomenon is that the rate of
at the maximum possible rate if it were able to do so. uptake of CCS-equipped facilities is limited in the models. From what
The cost of carbon reported in Fig. 4 for the 450 ppm scenario is reported by Koelbl et al. [76], we can conclude that CCS uptake is not
well above the cost of carbon assumed by the IEA [132] for the 450 limited by storage capacity or growth thereof. Therefore, another

Table 16
CO2 storage cost ranges per storage type (modified from Ref. [77]).
$2015/tCO2 stored Enhanced Oil Recovery (EOR) Remaining gas Depleted oil Depleted gas Enhanced Coal Bed Methane Recovery (ECMR) Aquifer

Onshore −128.3–64.1 1–16.9 1–16.9 1–16.9 −36.7–210.5 0.5–12.1

Offshore −128.3–125.8 1.9–35.1 1.9–35.1 1.9–35.1 1–42.4

72
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

Fig. 3. Average total primary energy from fossil fuel use (EJ/yr) for the Fig. 4. Average cost of carbon (CO2) for the 450 ppm scenario across EMF27
450 ppm scenario across EMF27 models. models.

Table 17 evaluation difficult. Testing the hypothesis on residual emissions is

Average total primary energy from fossil fuel use (EJ/yr) in 2050 and 2100 for outside the scope of this paper. However, it will be the subject of further
the 450 ppm scenario across EMF27 models. investigation in future research.
Climate mitigation scenario 450 ppm
6. Analysis of residual emissions
Technology scenario Conv Fulltech noCCS

Primary Energy (fossil, EJ) - 2050 326 364 140 Most CCS studies assume a capture rate for CO2 emissions. The re-
Primary Energy (fossil, EJ) - 2100 256 215 36 sidual emissions of the process are the remaining percentage of emis-
sions that are emitted to the atmosphere. The capture rate assumed in
most studies is in the range of 85–90% for power generation. However,
option is a limit on the rate that CCS-enabled facilities can be built (e.g. there is no evidence that these capture rates represent the maximum
maximum capacity or activity growth rates, maximum new capacity technically achievable, and indeed the basis for this assumption is
installation by region, etc.), or how quickly infrastructure related to rarely discussed. Though additional cost would be incurred, higher
CCS can be built. However, the detailed review produced on CCS as- capture rates, even greater than 95% may be technically achievable.
sumptions in the relevant models [76] did not cite any limits on uptake This study does not seek to quantify the technical limits or economic
of these technologies, and further personal communications with the impact of higher capture rates, but instead seeks to provide an initial
relevant modellers confirmed that any such limits were likely to be non- investigation into the implications for unburnable carbon if such a
binding, particularly in later model years. technology were available. To support this analysis a global integrated
This paper hypothesises that the constraint on CCS is therefore not assessment model, TIAM-Grantham [135], has been applied to examine
cost related or supply chain related (i.e. build rate limited), particularly the impact of a range of capture rates on the use of fossil fuels in the
in later years, but rather that the residual emissions from CCS make it global energy system. The range of capture rates investigated was
an unfavourable option in climate change mitigation scenarios; even 66%–96% in 2% increments, with results for global gas primary energy
these low levels of emissions are sufficiently high to conflict with ex- supply as shown in Fig. 5. While neither coal nor oil experience sig-
tremely constrained global carbon budgets. This hypothesis is sup- nificant changes across the capture rate sensitivity examined, gas sees
ported by previous work produced by the UK Energy Research Centre the strongest impact in this initial study. Earlier in the time horizon the
(UKERC) [133] and IEAGHG [134], who both report a capture rate of capture rate does not have a large impact due to the relatively low price
90% for coal based power generation with CCS. IEAGHG [134] de- of other abatement opportunities across the global economy at that
monstrated that increasing the capture rate from 90% to 98% would not time. However, beyond 2050 the capture rate becomes very important,
increase but rather reduce (−3%) the cost per tonne of CO2 avoided for and by 2100 a high capture rate of 96% leads to an almost doubling of
oxy-combustion and IGCC applications. Capture technology developers the primary gas supply relative to an 85% capture rate. This additional
have so far focussed on 85–90% capture rates however this could not be 100 EJ per year of primary gas supply has a wholesale value of ap-
sufficient with tighter global emission limits. However, the lack of data proximately £500 billion per year at UK gas prices at the time of
regarding state-of-the-art capture rates of CCS plants makes the writing. Further research is required, including multi-model

Table 18
Cumulative fossil fuel consumption in the timeframes 2005–2050 and 2005–2100. Reserves ‘low’ estimate from McCollum et al. [39]. “without CCS” scenario
corresponds to the noCCS scenario while “with CCS” scenario corresponds to the Fulltech scenario.
GtCO2 Exajoules (EJ) % of reserves

without CCS with CCS without CCS with CCS without CCS with CCS

Up until 2050 953 1347 13,166 18,356 26 37

Up until 2100 1208 2380 16,823 32,376 33 65

73
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

the future prevalence and need for reservoir pressure management and
management of the produced brine. Moreover, there is a need for fur-
ther Research, Development & Demonstration (R,D&D) in order to
improve storage efficiency.
It should also be a high priority to update CCS components in in-
tegrated assessment models with costs associated with the need for
brine production to relieve pressure with increased rates of CO2 injec-
tion.
The agreement reached during COP21 invites the
Intergovernmental Panel on Climate Change to “provide a special re-
port in 2018 on the impacts of global warming of 1.5 °C above pre-
industrial levels and related global greenhouse gas emission pathways”
[18]. Therefore, the assessment provided in this Paper should be re-
visited in the future in order to take into account the outcomes of the
IPCC special report.

8. Conclusions

This paper has reviewed potential barriers to the worldwide adop-

tion of CCS and also considered whether this technology has the po-
Fig. 5. Sensitivity of primary energy supply of natural gas in 2050, 2080 and
2100 to CCS capture rate, produced by means of the TIAM-Grantham model. tential to enable access to more fossil fuel reserves in the future, where
these reserves would otherwise be ‘unburnable’. The authors have cri-
tically reviewed the studies that have considered CCS in the context of
comparison studies, in order to fully explore the capture rate issue. Such
unburnable carbon, analysed the status and costs of CCS, studied its
a study is out of the scope of this paper, but will be pursued in future
impact on fossil fuel consumption across a selection of global climate
research.
change mitigation models used in the IPCC 5th assessment report, and
The result produced here is dependent on a range of assumptions in
examined the extent of global CO2 geo-storage capacity. Finally, a new
the model, including the relative fossil fuel extraction costs, bioenergy
analysis has been performed demonstrating the impact of the capture
cost, CCS technology cost, performance and availability, and regional or
rate of CCS technology on unburnable carbon outcomes.
global constraints on CO2 geo-sequestration and total technology up-
There have been a number of recent studies reviewing the un-
take. Therefore, it is recommended that further IAM modelling studies
burnable carbon topic. These have broadly reached the same conclu-
are undertaken to understand the drivers of the shift of impact of re-
sion; that some portion of fossil fuel reserves is unburnable in scenarios
duced residual emissions between gas and coal (and, to an extent, oil),
where climate change induced warming is limited to a reasonable
with the expectation that some sensitivity in the results produced
chance of temperature rise less than 2 °C. Only a few of these studies
herein is likely to be observed.
have explicitly considered the impact of the availability of CCS tech-
nology. Those studies that did consider this issue explicitly indicated
7. Recommendations for further research that CCS has a limited impact on the amount of reserves that are
burnable. However, none of these studies focused on the potential of
The reported results have highlighted the need for further research CCS, or questioned why results indicated a less prominent role for the
relating to the potential impact of technology on the extent of un- technology than might otherwise be expected.
burnable fossil fuels. Key areas for future research topics include sen- In order to fill this gap, an analysis specifically on CCS and un-
sitivity analysis of design and operational parameters on CCS capture burnable carbon has been undertaken. Insights are drawn from the
rate, regional dynamic assessment of CO2 storage resources and CCS EMF27 multi-model comparison, which produced a set of scenarios of
modelling in integrated assessment models. energy system change to mitigate climate change. EMF27 included
This paper has performed an initial investigation into the impact of scenarios with and without CCS, and therefore provides a robust and
the CCS capture rate on the use of fossil fuels in the long term, with the consistent basis for investigation of the impact of CCS on fossil fuel
key finding being that it is indeed an important factor. Further mod- reserve utilisation. Analysis of results confirm that CCS availability has
elling studies are required to clarify the nuances of this point, and a large bearing on the extent of fossil fuel consumption in climate-
specifically what the impact of fossil fuel prices, CCS technology cost, constrained scenarios; approximately 200 EJ per year more fossil fuel is
performance and availability, and competition with other low carbon utilised per year in scenarios with CCS, as opposed to a scenario without
technologies has on outcomes. These studies should not only perform the technology. A key difference between this study and previous ef-
sensitivity analysis on these parameters, but should also examine results forts is that the dynamics of CCS uptake were considered, with the
across multiple models to determine robustness of results to modelling observation that CCS adoption is still ramping up at 2050 (previous
approach. studies limited the time horizon of consideration to 2050).
The analysis of the literature has highlighted a lack of data on the The extent to which EMF27 modelling assumptions limit CCS up-
state-of-the-art capture rate for CCS plants. Most references indicate a take has also been reviewed. Based on the evidence available from
capture rate of 90%; however this value may not be enough. Previous EMF27 models, there are few limiting assumptions made on the avail-
research [134] has already shown that increasing the percentage of ability of CCS. Almost all models reviewed reported no capacity or
capture to 98% would not increase the cost per tonne of CO2 abated for uptake-rate limits for the transport and storage phases of CCS. While
oxy-combustion and pre-combustion applications. Therefore, further less evidence was available for the capture phase, it is unlikely that such
research may be needed in order to increase the capture rate of CCS constraints are preventing uptake substantially, particularly later in the
plants closer to 100% and to understand the technology and cost im- time horizon (i.e. 2040 onwards).
plications of this. Also, the cost of CCS technology assumed in the models does not
It should be a high priority for all countries considering large scale appear to be a significant barrier. The key observation is that the capital
deployment of CO2 storage to perform regional dynamic assessments of and operating costs of CCS technology are generally much lower than
the CO2 storage resource. This will provide important information on the marginal abatement costs observed in the models (these are from

74
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

hundreds to thousands of US$ per tonne across the models, which is this issue would be able to explore the issue more fully.
substantially higher than the cost of CCS). Therefore, if CCS is available
(and not unfavourable for other reasons) further adoption should be Acknowledgements
observed in the models. The only plausible explanation that such
adoption is not observed is that there is another factor in the models Part of this paper has been developed in collaboration with the IEA
preventing uptake. Greenhouse Gas R&D Programme (IEAGHG), who are gratefully ac-
This paper investigated the possibility that residual emissions from knowledged for their funding and intellectual contribution. The authors
CCS installations, usually modelled as approximately 10–15% of emis- would like to especially thank Tim Dixon and Dr Jasmin Kemper for
sions from the source in question, is the reason further uptake is not their advice and inputs.
observed. The TIAM-Grantham model was applied to consider this The authors would also like to acknowledge the funding from the
question, running a scenario constrained to 2 °C warming across a range Sustainable Gas Institute, which was founded by Imperial College
of capture rates from 66% to 96%. The result of this study was that London and BG Group (now part of Royal Dutch Shell). Funding for the
capture rate is indeed very important for the role of natural gas, in Sustainable Gas Institute is gratefully received from Royal Dutch Shell,
particular, in future energy systems. From 2050 onwards very high Enagas SA and the Newton/NERC/FAPESP Sustainable Gas Futures
capture rates lead to natural gas retaining market share while the other project NE/N018656/1. Note that funding bodies were not involved in
fossil fuels consistently decline. A further multi-model comparison on the design, implementation or reporting of this study.

Appendix A. Supplementary data

Supplementary data related to this article can be found at https://doi.org/10.1016/j.esr.2018.08.003.

Appendices. A. Projections for the climate change scenario: 550 ppm

Fig. A.1. Average global emissions of CO2 (GtCO2/yr) for the 550 ppm scenario across EMF27 models.

CAPTURED CO 2
CLIMATE CHANGE SCENARIO: 550 ppm
Fulltech Conv noCCS

35

30

25
Captured CO (GtCO /yr)

20

15

10

0
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Years

Fig. A.2. Average capture of CO2 (GtCO2/yr) for the 550 ppm scenario across EMF27 models.

75
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

FOSSIL FUEL USAGE

CLIMATE CHANGE SCENARIO: 550 ppm
Fulltech Conv noCCS

600

500

Fossil fuel usage (EJ/yr)

400

300

200

100

0
2005 2015 2025 2035 2045 2055 2065 2075 2085 2095
Years

Fig. A.3. Average total primary energy from fossil fuel use (EJ/yr) for the 550 ppm scenario across EMF27 models.

Table A.1
Average total primary energy from fossil fuel use (EJ/yr) in 2050 and 2100 for the 550 ppm scenario across EMF27 models.

Climate mitigation scenario 550 ppm

Technology scenario Conv Fulltech noCCS

Primary Energy (fossil, EJ) - 2050 474 457 299
Primary Energy (fossil, EJ) - 2100 478 437 143

Fig. A.4. Cost of carbon (CO2) for the 550 ppm scenario. Timeframe 2005–2100, B. Projections for single EMF27 models.

76
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

B. Projections for single EMF27 models

Fig. B.1. Emissions from fossil fuel usage in the timeframe 2005–2050 according to the EMF27 models. Usage = 0 means scenario infeasibility.

Fig. B.2. Emissions from fossil fuel usage in the timeframe 2005–2100 according to the EMF27 models. Usage = 0 means scenario infeasibility, C. Projections for
single fuels use.

C. Projections for single fuels use

Fig. C.1. Total primary energy from oil use (EJ/yr) for the 450 ppm scenario across EMF27 models.

77
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

Fig. C.2. Total primary energy from gas use (EJ/yr) for the 450 ppm scenario across EMF27 models.

Fig. C.3. Total primary energy from coal use (EJ/yr) for the 450 ppm scenario across EMF27 models.

Fig. C.4. Average distribution of total primary energy from fossil fuel use (EJ/yr) by type of fuel for the 450 ppm scenario across EMF27 models for the Fulltech
scenario.

78
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

Fig. C.5. Average distribution of total primary energy from fossil fuel use (EJ/yr) by type of fuel for the 450 ppm scenario across EMF27 models for the Conv
scenario.

Fig. C.6. Average distribution of total primary energy from fossil fuel use (EJ/yr) by type of fuel for the 450 ppm scenario across EMF27 models for the noCCS
scenario.

References and Vulnerability. Contribution of Working Group II to the Fifth Assessment

Report of the Intergovernmental Panel on Climate Change, Cambridge University
Press, Cambridge, United Kingdom and New York, NY, USA, 2013.
[1] IPCC, Climate Change 2014: Mitigation of Climate Change. Contribution of [15] M. Meinshausen, N. Meinshausen, W. Hare, S.C.B. Raper, K. Frieler, R. Knutti,
Working Group III to the Fifth Assessment Report of the Intergovernmental Panel D.J. Frame, M.R. Allen, Greenhouse-gas emission targets for limiting global
on Climate Change, Cambridge University Press, Cambridge, United Kingdom and warming to 20C, Nature 458 (2009) 1158–1162.
New York, NY, USA, 2014. [16] M.R. Allen, D.J. Frame, C. Huntingford, C.D. Jones, J.A. Lowe, M. Meinshausen,
[2] IEA, Technology roadmap: carbon capture and storage 2011, in: IEA (Ed.), N. Meinshausen, Warming caused by cumulative carbon emissions towards the
Technology RoadmapOnline, 2011. trillionth tonne, Nature 458 (2009) 1163–1166.
[3] IEA, Technology roadmap: carbon capture and storage 2013, in: IEA (Ed.), [17] IPCC, Climate Change 2007: Synthesis Report. Contribution of Working Groups I,
Technology RoadmapOnline, 2013. II and III to the Fourth Assessment Report of the Intergovernmental Panel on
[4] Carbon Tracker Initiative, Unburnable carbon 2013: wasted capital and stranded Climate Change, IPCC, Geneva, Switzerland, 2007.
assets, C.T. Initiative, 2013. [18] COP21, Adoption of the Paris agreement, COP21, 2015.
[5] IPCC, IPCC Special Report: Carbon Capture and Storage, (2005). [19] M. Meinshausen, S.C.B. Raper, T.M.L. Wigley, Emulating IPCC AR4 atmosphere-
[6] V.R. Clark, H.J. Herzog, Can “stranded” fossil fuel reserves drive CCS deployment? ocean and carbon cycle models for projecting global-mean, hemispheric and land/
Energy Procedia 63 (2014) 7261–7271. ocean temperatures: MAGICC 6.0, Atmos. Chem. Phys. Discuss. 8 (2008)
[7] J.J. Gale, T. Dixon, CCS as a Critical Part of the Carbon Budget, (2014). 6153–6272.
[8] IPCC, Climate change 2014: summary for policy makers. Mitigation of climate [20] P. Friedlingstein, P. Cox, R. Betts, L. Bopp, W. Von Bloh, V. Brovkin, P. Cadule,
change, Contribution of Working Group III to the Fifth Assessment Report of the S. Doney, M. Eby, I. Fung, G. Bala, J. John, C. Jones, F. Joos, T. Kato,
Intergovernmental Panel on Climate Change, Cambridge University Press, M. Kawamiya, W. Knorr, K. Lindsay, H.D. Matthews, T. Raddatz, P. Rayner,
Cambridge, United Kingdom and New York, NY, USA, 2014. C. Reick, E. Roeckner, K.G. Schnitzler, R. Schnur, K. Strassmann, A.J. Weaver,
[9] Global CCS Institute, The Global Status of CCS: 2016 summary report, Global CCS C. Yoshikawa, N. Zeng, Climate-carbon cycle feedback analysis: results from the
Institute, 2016. (CMIP)-M-4 model intercomparison, J. Clim. 19 (2006) 3337–3353.
[10] SaskPower, Boundary Dam Carbon Capture Project, (2017). [21] C. Huntingford, J.A. Lowe, B.B.B. Booth, C.D. Jones, G.R. Harris, L.K. Gohar,
[11] NRG, Petra Nova, (2017). P. Meir, Contributions of carbon cycle uncertainty to future climate projection
[12] Global CCS Institute, Large Scale CCS Projects, (2017). spread, Tellus B 61 (2009) 355–360.
[13] IPCC, Climate change 2013: summary for policy makers. The physical science [22] M. Meinshausen, B. Hare, T.M.L. Wigley, D. Van Vuuren, M.G.J. Den Elzen,
basis, Contribution of Working Group I to the Fifth Assessment Report of the R. Swart, Multi-gas emissions pathways to meet climate targets, Climatic Change
Intergovernmental Panel on Climate Change, Cambridge University Press, 75 (2006) 151–194.
Cambridge, United Kingdom and New York, NY, USA, 2013. [23] A. Bowen, N. Ranger, Mitigating Climate Change through Reductions in
[14] IPCC, Climate Change 2014: Summary for Policy Makers. Impacts, Adaptation, Greenhouse Gas Emissions: the Science and Economics of Future Paths for Global

79
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

Annual Emissions, (2009). [65] V. Scott, R.S. Haszeldine, S.F.B. Tett, A. Oschlies, Fossil fuels in a trillion tonne
[24] J. Rogelj, M. Schaeffer, P. Friedlingstein, N.P. Gillett, D.P. van Vuuren, K. Riahi, world, Nat. Clim. Change 5 (2015) 419–423.
M. Allen, R. Knutti, Differences between carbon budget estimates unravelled, Nat. [66] C.C.S. SaskPower, Boundary Dam carbon capture project, SaskPower, 2016
Clim. Change 6 (2016) 245–252. Online.
[25] IPCC, Climate change 2013: technical summary, The Physical Science Basis. [67] IEAGHG, Can Carbon Capture and Storage Unlock ‘unburnable Carbon’? (2016).
Contribution of Working Group I to the Fifth Assessment Report of the [68] M.L. Szulczewski, C.W. MacMinn, H.J. Herzog, R. Juanes, Lifetime of carbon
Intergovernmental Panel on Climate Change, Cambridge University Press, capture and sorage as a climate-change mitigation technology, Proc. Natl. Acad.
Cambridge, United Kingdom and New York, NY, USA, 2013. Sci. Unit. States Am. (2012) 5185–5189.
[26] V.E. McKelvey, Mineral Resource Estimates and Public Policy: better methods for [69] J.J. Dooley, Estimating the supply and demand for deep geologic CO2 storage
estimating the magnitude of potential mineral resources are needed to provide the capacity over the course of the 21st Century: a meta-analysis of the literature,
knowledge that should guide the design of many key public policies, Am. Sci. 60 Ghgt-11 37 (2013) 5141–5150.
(1972) 32–40. [70] C.-T. Liu, B.-Z. Hsieh, C.-C. Tseng, Z.-S. Lin, Modified classification system for
[27] Society of Petroleum Engineers (SPE), Petroleum Reserves & Resources estimating the CO2 storage capacity of saline formations, Int. J..Greenh.gas. Contr.
Definitions, (2018). 22 (2014) 244–255.
[28] SPE, Petroleum resources management system, S.S.o.P. Engineers, 2008 Online. [71] S.M. Frailey, O. Tucker, G.J. Koperna, The genesis of the CO2 storage resources
[29] USA Government, U.S. Secturities and Exchange Commssion, (2018). management system (SRMS), Energy Procedia 114 (2017) 4262–4269.
[30] USGS, World Petroleum Assessment, (2018). [72] Society of Petroleum Engineers (SPE), CO2 Storage Resources Management
[31] Norwegian Petroleum Directorate, Resource Accounts & Analysis, (2018). System, (2018).
[32] The Russian Government, Ministry of Natural Resources and Environment of the [73] Task force on application of UNFC - 2009 to injection projects specifications for
Russian Federation, (2018). the application of the united Nations classification for fossil energy and mineral
[33] Oil and Gas Reserves Committee, Comparison of selected reserves and resource reserves and resources 2009 (UNFC-2009) to injection projects for the purpose of
classifications and associated definitions, SPE, 2005. geological storage, UNECE, 2016.
[34] C. McGlade, P. Ekins, Un-burnable oil: an examination of oil resource utilisation in [74] N. Bauer, Carbon Capturing and Sequestration: an Option to Buy Time? Doctoral
a decarbonised energy system, Energy Pol. 64 (2014) 102–112. Thesis University Potsdam Faculty of Economics and Social Sciences, 2005.
[35] Society of Petroleum Engineers, Comparison of Selected Reserves and Resource [75] I. Keppo, B. van der Zwaan, The impact of uncertainty in climate targets and CO2
Classifications and Associated Definitions, (2005). storage availability on long-term emissions abatement, Environ. Model. Assess. 17
[36] Carbon Tracker Initiative, Unburnable Carbon – are the world's financial markets (2012) 177–191.
carrying a carbon bubble? in: C. Tracker (Ed.), Carbon Tracker Initiative, 2011. [76] B.S. Koelbl, M.A. van den Broek, A.P.C. Faaij, D.P. van Vuuren, Uncertainty in
[37] N. Mayer, L. Brinker, Deflating the "carbon bubble, I. Energy, 2014. Carbon Capture and Storage (CCS) deployment projections: a cross-model com-
[38] P. Spedding, K. Mehta, N. Robins, Oil & carbon revisited, H.G. Research, 2013. parison exercise, Climatic Change 123 (2014) 461–476.
[39] D. McCollum, N. Bauer, K. Calvin, A. Kitous, K. Riahi, Fossil resource and energy [77] B.S. Koelbl, M.A. van den Broek, B.J. van Ruijven, A.P.C. Faaij, D.P. van Vuuren,
security dynamics in conventional and carbon-constrained worlds, Climatic Uncertainty in the deployment of Carbon Capture and Storage (CCS): a sensitivity
Change 123 (2014) 413–426. analysis to techno-economic parameter uncertainty, Int. J..Greenh.gas. Contr. 27
[40] C. McGlade, P. Ekins, The geographical distribution of fossil fuels unused when (2014) 81–102.
limiting global warming to 20C, Nature 517 (2015) 187–190. [78] J.J. Dooley, S.J. Friedman, A Global but Regionally Disaggregated Accounting of
[41] IEA, World energy outlook 2012, IEA, 2012 Online. CO2 Storage Capacity: Data and Assumptions for Compiling Regional CO2 Storage
[42] BP, BP statistical review of world energy, BP, 2015. Capacity Supply Curves for Incorporation within ObjECTS - > MiniCAM,
[43] IEA, World energy outlook 2014 - executive summary, IEA, 2014 Online. Lawrence Livermore National Laboratory, 2005 UCRL-SR-209663.
[44] World Energy Council, World Energy Resources: 2013 Survey, (2013). [79] R.T. Dahowski, X. Li, C.L. Davidson, N. Wei, J.J. Dooley, Regional opportunities
[45] BGR, Reserves, Resources and Availability of Energy Resources, (2014). for carbon dioxide capture and storage in China, P.N.N.L. PNNL-19091, 2009.
[46] Oil & Gas Journal, Worldwide look at reserves and production, Oil Gas J. 112 (1) [80] C. Hendriks, W. Graus, Global carbon dioxide storage potential and costs, E. EEP-
(2014) [104] is an online report: http://www.zeroemissionsplatform.eu/library/ 02001, 2004.
publication/165-zep-cost-report-summary.html. [81] J.J. Birkholzer, Q. Zhou, Basin-scale hydrogeologic impacts of CO2 storage: ca-
[47] L. Herrmann, E. Dunphy, J. Copus, Oil and Gas for Beginners, Deutsche Bank, pacity and regulatory implications, Int. J..Greenh.gas. Contr. 3 (2009) 745–756.
2010. [82] W.G. Allinson, Y. Cinar, P.R. Neal, J. Kaldi, L. Paterson, CO2 storage capacity –
[48] European Council, Communication on community strategy on climate change: combining geology, engineering and economics, SPE Econ. Manag. 6 (2014)
council conclusions, European Council, 1996 Brussels. 15–27.
[49] European Commission Climate Action, The EU Emissions Trading System, EU ETS, [83] IEAGHG, Extraction of Formation Water from CO2 Storage, (2012).
2016. [84] T. Ogawa, S. Nakanishi, T. Shidahara, T. Okumura, E. Hayashi, Saline-aquifer CO2
[50] D. Hone, The carbon bubble reality check, Shell, 2013. sequestration in Japan – methodology of storage capacity assessment, Int.
[51] BP, Climate change, BP, 2014. J..Greenh.gas. Contr. 5 (2011) 318–326.
[52] CarbonBrief, What the Fossil Fuel Industry Thinks of the ‘carbon Bubble’, (2014). [85] Asian Development Bank, Roadmap for Carbon Capture and Storage
[53] D.Y.C. Leung, G. Caramanna, M.M. Maroto-Valer, An overview of current status of Demonstration and Deployment in the People's Republic of China, (2015).
carbon dioxide capture and storage technologies, Renew. Sustain. Energy Rev. 39 [86] B. Söderbergh, K. Jakobsson, K. Aleklett, European energy security: an analysis of
(2014) 426–443. future Russian natural gas production and exports, Energy Pol. 38 (2010)
[54] S.D. Kenarsari, D.L. Yang, G.D. Jiang, S.J. Zhang, J.J. Wang, A.G. Russell, Q. Wei, 7827–7843.
M.H. Fan, Review of recent advances in carbon dioxide separation and capture, [87] IEAGHG, Barriers to Implementation of CCS: Capacity Constraints, (2012).
RSC Adv. 3 (2013) 22739–22773. [88] N. Odeh, Assessing the domestic supply chain barriers to the commercial de-
[55] M.E. Boot-Handford, J.C. Abanades, E.J. Anthony, M.J. Blunt, S. Brandani, N. Mac ployment of carbon capture and storage within the power sector, AEA, 2012.
Dowell, J.R. Fernandez, M.C. Ferrari, R. Gross, J.P. Hallett, R.S. Haszeldine, [89] J. Miller, Life after oil: can Aberdeen rise again? B. News, 2016.
P. Heptonstall, A. Lyngfelt, Z. Makuch, E. Mangano, R.T.J. Porter, [90] H. Groenenberg, H. de Coninck, Effective EU and member state policies for sti-
M. Pourkashanian, G.T. Rochelle, N. Shah, J.G. Yao, P.S. Fennell, Carbon capture mulating CCS, Int. J..Greenh.gas. Contr. 2 (2008) 653–664.
and storage update, Energy Environ. Sci. 7 (2014) 130–189. [91] A. Kossoy, P. Guigon, State and trends of the carbon market 2012, W. Bank, 2012
[56] J.C.M. Pires, F.G. Martins, M.C.M. Alvim-Ferraz, M. Simões, Recent developments Washington DC.
on carbon capture and storage: an overview, Chem. Eng. Res. Des. 89 (2011) [92] R. Lohwasser, R. Madlener, Relating R&D and investment policies to CCS market
1446–1460. diffusion through two-factor learning, Energy Pol. 52 (2013) 439–452.
[57] E.S. Rubin, H. Mantripragada, A. Marks, P. Versteeg, J. Kitchin, The outlook for [93] BG Group plc, BP plc, ENI SpA, Royal Dutch Shell Plc, Statoil ASA, Total SA,
improved carbon capture technology, Prog. Energy Combust. Sci. 38 (2012) Paying for Carbon Letter, (2015).
630–671. [94] N.M.A. Huijts, E.J.E. Molin, L. Steg, Psychological factors influencing sustainable
[58] J. Gibbins, H. Chalmers, Is all CCS equal? Classifying CCS applications by their energy technology acceptance: a review-based comprehensive framework, Renew.
potential climate benefit, Energy Procedia 4 (2011) 5715–5720. Sustain. Energy Rev. 16 (2012) 525–531.
[59] M. Fajardy, N. Mac Dowell, Can BECCS deliver sustainable and resource efficient [95] S.L. Seigo, S. Dohle, M. Siegrist, Public perception of carbon capture and storage
negative emissions? Energy Environ. Sci. 10 (2017) 1389–1426. (CCS): a review, Renew. Sustain. Energy Rev. 38 (2014) 848–863.
[60] J. Koornneef, P.v. Breevoort, C. Hamelinck, C. Hendriks, M. Hoogwijk, K. Koop, [96] K. van Alphen, Q.V.T. Voorst, M.P. Hekkert, R. Smits, Societal acceptance of
M. Koper, T. Dixon, A. Camps, Global potential for biomass and carbon dioxide carbon capture and storage technologies, Energy Pol. 35 (2007) 4368–4380.
capture, transport and storage up to 2050, Int. J..Greenh.gas. Contr. 11 (2012) [97] L. Wallquist, V.H.M. Visschers, M. Siegrist, Impact of knowledge and misconcep-
117–132. tions on benefit and risk perception of CCS, Environ. Sci. Technol. 44 (2010)
[61] D. van Vuuren, S. Deetman, J. van Vliet, M. van den Berg, B. van Ruijven, 6557–6562.
B. Koelbl, The role of negative CO2 emissions for reaching 2 °C—insights from [98] L.A. Fleishman, W.B. de Bruin, M.G. Morgan, Informed public preferences for
integrated assessment modelling, Climatic Change 118 (2013) 15–27. electricity portfolios with CCS and other low-carbon technologies, Risk Anal. 30
[62] D. McLaren, A comparative global assessment of potential negative emissions (2010) 1399–1410.
technologies, Process Saf. Environ. Protect. 90 (2012) 489–500. [99] S.C. Page, A.G. Williamson, I.G. Mason, Carbon capture and storage: fundamental
[63] IEAGHG, Barriers to Overcome in Implementation of CO2 Capture and Storage (2): thermodynamics and current technology, Energy Pol. 37 (2009) 3314–3324.
Rules and Standards for the Transmission and Storage of CO2, (2003). [100] E.S. Rubin, Understanding the pitfalls of CCS cost estimates, Int. J..Greenh.gas.
[64] J. Gibbins, H. Chalmers, Carbon capture and storage, Energy Pol. 36 (2008) Contr. 10 (2012) 181–190.
4317–4322. [101] K. Goto, K. Yogo, T. Higashii, A review of efficiency penalty in a coal-fired power

80
S. Budinis et al. Energy Strategy Reviews 22 (2018) 61–81

plant with post-combustion CO2 capture, Appl. Energy 111 (2013) 710–720. [122] B. Caldecott, G. Lomax, M. Workman, Stranded carbon assets and negative
[102] G.P. Hammond, S.S.O. Akwe, S. Williams, Techno-economic appraisal of fossil- emissions technologies, S.A.P.a.t.U.o.O.s.S.S.o.E.a.t. Environment, 2015 Working
fuelled power generation systems with carbon dioxide capture and storage, Energy Paper SeriesOnline.
36 (2011) 975–984. [123] IPCC, Integrated assessment modelling consortium (IAMC) AR5 scenario database
[103] G. Simbolotti, CO2 capture and storage, IEA ETSAP, 2010 Technology Brief. (working group III), I.I.f.A.S.A. 2014, 2014 Online.
[104] European technology platform for zero emission fossil fuel power plants (ZEP), the [124] E. Kriegler, J. Weyant, G. Blanford, V. Krey, L. Clarke, J. Edmonds, A. Fawcett,
costs of CO2 capture, Transport and Storage (2011). G. Luderer, K. Riahi, R. Richels, S. Rose, M. Tavoni, D. van Vuuren, The role of
[105] T. Fout, A. Zoelle, D. Keairns, M. Turner, M. Woods, N. Kuehn, V. Shah, V. Chou, technology for achieving climate policy objectives: overview of the EMF 27 study
L. Pinkerton, Cost and Performance Baseline for Fossil Energy Plants - Volume 1a: on global technology and climate policy strategies, Climatic Change 123 (2014)
Bituminous Coal (PC) and Natural Gas to Electricity - Revision 3, (2015). 353–367.
[106] Global CCS Institute, Economic assessment of carbon capture and storage tech- [125] K. Riahi, E. Kriegler, N. Johnson, C. Bertram, M. den Elzen, J. Eom, M. Schaeffer,
nologies - 2011 update, WorleyParsons, Schlumberger, GCCS, 2011. J. Edmonds, M. Isaac, V. Krey, T. Longden, G. Luderer, A. Méjean, D.L. McCollum,
[107] N. Kuehn, K. Mukherjee, P. Phiambolis, L.L. Pinkerton, E. Varghese, M. Woods, S. Mima, H. Turton, D.P. van Vuuren, K. Wada, V. Bosetti, P. Capros, P. Criqui,
Current and Future Technologies for Natural Gas Combined Cycle (NGCC) Power M. Hamdi-Cherif, M. Kainuma, O. Edenhofer, Locked into Copenhagen pledges —
Plants, (2013). implications of short-term emission targets for the cost and feasibility of long-term
[108] D. Leeson, J. Fairclough, C. Petit, P. Fennell, A Systematic Review of Current climate goals, Technol. Forecast. Soc. Change 90 (Part A) (2014) 8–23.
Technology and Cost for Industrial Carbon Capture - Grantham Report 7, (2015). [126] E. Kriegler, K. Riahi, N. Bauer, V.J. Schwanitz, N. Petermann, V. Bosetti,
[109] E.S. Rubin, J.E. Davison, H.J. Herzog, The cost of CO2 capture and storage, Int. A. Marcucci, S. Otto, L. Paroussos, S. Rao, T. Arroyo Currás, S. Ashina, J. Bollen,
J..Greenh.gas. Contr. 40 (2015) 378–400. J. Eom, M. Hamdi-Cherif, T. Longden, A. Kitous, A. Méjean, F. Sano, M. Schaeffer,
[110] M. Finkenrath, Cost and performance of carbon dioxide capture from power K. Wada, P. Capros, D.P. van Vuuren, O. Edenhofer, Making or breaking climate
generation, International Energy Agency, 2011. targets: the AMPERE study on staged accession scenarios for climate policy,
[111] Global CCS Institute, The Costs of CCS and Other Low-carbon Technologies, Technol. Forecast. Soc. Change 90 (Part A) (2014) 24–44.
(2011). [127] V. Krey, G. Luderer, L. Clarke, E. Kriegler, Getting from here to there – energy
[112] L. Irlam, The costs of CCS and other low-carbon technologies in the United States - technology transformation pathways in the EMF27 scenarios, Climatic Change 123
2015 update, GCCS, 2015. (2014) 369–382.
[113] J. Davison, CCS costs and economics, in: IEAGHG (Ed.), IEAGHG-OPEC Workshop [128] J. van Vliet, A. Hof, A. Mendoza Beltran, M. van den Berg, S. Deetman, M.J. den
on CCS CDM,Vienna, 2013. Elzen, P. Lucas, D. van Vuuren, The impact of technology availability on the
[114] Department of Energy & Climate Change, The Potential for Reducing the Costs of timing and costs of emission reductions for achieving long-term climate targets,
CCS in the UK, (2013). Climatic Change 123 (2014) 559–569.
[115] EIA, Levelized Cost and Levelized Avoided Cost of New Generation Resources in [129] V. Krey, K. Riahi, Implications of delayed participation and technology failure for
the Annual Energy Outlook 2015, (2015). the feasibility, costs, and likelihood of staying below temperature targets—-
[116] Congressional budget office, federal efforts to reduce the cost of capturing and greenhouse gas mitigation scenarios for the 21st century, Energy Econ. 31
storing carbon dioxide, Congress Of the United States, 2012. (Supplement 2) (2009) S94–S106.
[117] S. Bassi, R. Boyd, S. Buckle, Fennell Paul, N. Mac Dowell, Z. Makuch, I. Staffell, [130] K. Akimoto, F. Sano, T. Homma, K. Wada, M. Nagashima, J. Oda, Comparison of
Bridging the gap: Improving the Economic and Policy Framework for Carbon marginal abatement cost curves for 2020 and 2030: longer perspectives for ef-
Capture and Storage in the European Union, (2015). fective global GHG emission reductions, Sustain Sci 7 (2012) 157–168.
[118] EIA, Updated Capital Cost Estimates for Utility Scale Electricity Generating Plants, [131] B.H.P. Billiton, Climate change: portfolio analysis, B. Billiton, 2015.
(2013). [132] IEA, World Energy Model Documentation - 2014 Version, (2014).
[119] ExxonMobil, The outlook for energy: a view to 2040, ExxonMobil, 2015. [133] G. Anandarajah, N. Strachan, P. Ekins, K. Ramachandran, N. Hughes, Pathways to
[120] Shell, New lens scenarios, Shell, 2013. a low carbon economy: energy systems modelling - working paper, UKERC, 2008.
[121] IPCC, climate change 2014: Synthesis report, Contribution of Working Group I, II [134] IEAGHG, CO2 capture at coal based power and hydrogen plants, IEAGHG, 2014.
and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate [135] T. Napp, A. Gambhir, R. Thomas, A. Hawkes, D. Bernie, J. Lowe, Exploring the
Change, Cambridge University Press, Cambridge, United Kingdom and New York, Feasibility of Low-carbon Scenarios Using Historical Energy Transitions Analysis
NY, USA, 2014. (C3), (2015).

81