Subscriber access provided by University of Birmingham
Article
Can producing oil store carbon? Greenhouse Gas footprint of CO2EOR, offshore North Sea R Jamie Stewart, and R Stuart Haszeldine Environ. Sci. Technol., Just Accepted Manuscript • Publication Date (Web): 19 Mar 2015 Downloaded from http://pubs.acs.org on March 20, 2015
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 26
Environmental Science & Technology
1
Can producing oil store carbon? Greenhouse Gas
2
footprint of CO2EOR, offshore North Sea
3
4
R Jamie Stewart*, R Stuart Haszeldine
5
Scottish Carbon Capture & Storage,
6
School of GeoSciences, University of Edinburgh,
7
EH9 3JW, Scotland, UK
8
9
*Corresponding author – email: [email protected], phone: +44 (0)131 650 5936,
10
fax: +44 (0)131 650 7340
11
12
1 ACS Paragon Plus Environment
Environmental Science & Technology
13
Abstract
14
Carbon dioxide enhanced oil recovery (CO2EOR) is a proven and available technology used
15
to produce incremental oil from depleted fields whilst permanently storing large tonnages
16
of injected CO2. Although this technology has been used successfully onshore in North
17
America and Europe, there are currently no CO2EOR projects in the United Kingdom. Here,
18
we examine whether offshore CO2EOR can store more CO2 than onshore projects
19
traditionally have, and whether CO2 storage can offset additional emissions produced
20
through offshore operations and incremental oil production. Using a high-level Life Cycle
21
system approach, we find that the largest contribution to offshore emissions is from flaring
22
or venting of reproduced CH4 and CO2. These can already be greatly reduced by regulation.
23
If CO2 injection is continued after oil production has been optimised, then offshore CO2EOR
24
has the potential to be carbon negative - even when emissions from refining, transport and
25
combustion of produced crude oil are included. The carbon intensity of oil produced can be
26
just 0.056-0.062 tCO2e/bbl if flaring/venting is reduced by regulation. This compares against
27
conventional Saudi oil 0.040tCO2e/bbl, or mined shale oil >0.300tCO2e/bbl.
28 Abstract Art 29 2 ACS Paragon Plus Environment
Page 2 of 26
Page 3 of 26
Environmental Science & Technology
30
Introduction
31
Carbon Dioxide Enhanced Oil Recovery (CO2EOR) may be an option for reducing the cost of early-
32
stage carbon capture and storage (CCS) as more and more projects around the European Union (EU)
33
utilising saline aquifer storage are deemed too expensive and fail to go beyond Front End
34
Engineering Design (FEED).
35
CO2EOR is a technique that has been applied in the United States (US) since the mid 80’s to enhance
36
oil recovery from mature fields.1 In the United States where large volumes of CO2 are readily
37
available from natural sources, CO2 has been injected as the primary recovery agent in these EOR
38
operations. In Europe however the lack in availability of low cost CO2 has meant that EOR operations
39
have conventionally used water injection or methane gas injection, with currently no CO2EOR
40
projects in an offshore environment.2-5 With the prospect of Carbon Capture technology developing
41
across the EU however, the required volumes of CO2 may become available through anthropogenic
42
sources.6 CO2EOR is regarded as an option for storing large volumes of this captured CO2 with the
43
added benefit of improving the recovery rate from depleted oil fields. This oil production may hold
44
potential to improve CCS project economics and get projects beyond FEED.
45
However it is known from currently operating CO2EOR projects onshore that the operations and
46
processes involved in CO2EOR are energy intensive and could result in significant atmospheric
47
emissions.7,8 This paper presents the conclusions from a medium to high level life cycle assessment
48
of CO2EOR operations, in which emissions and CO2 stored are modelled for a theoretical offshore
49
North Sea project. Consideration of the uncertainties involved in operating a CO2EOR project in an
50
offshore environment means a number of scenarios and uncertainty ranges are presented. Particular
51
attention is paid to the flaring and venting of reproduced gases which is found to have a large
52
control on the emissions associated with the production of oil through CO2EOR.
3 ACS Paragon Plus Environment
Environmental Science & Technology
53
Several other studies have attempted to quantify how the emissions from EOR operations, and from
54
the final combustion of the petroleum products, relate to the total volume of CO2 stored in EOR
55
projects.9-12 These studies focus on historical data from onshore developments and vary widely in
56
their outcomes.
57
The primary reason for the varying conclusions relates to inclusion or exclusion of the large
58
contribution to emissions made by the combustion of produced petroleum products, and the
59
principal of additionality. Jaramillo et al., (2009)10 who looked at 5 onshore North American CO2 EOR
60
projects found that when emissions from the full system boundary, from coal mining to final product
61
combustion are included, then onshore CO2 EOR projects have historically been net emitters of CO2.
62
They assumed that oil produced through CO2EOR is additional to the global system and therefore
63
emissions from the combustion of the final petroleum products should be included in the study,
64
which resulted in overall net emissions.10,11 Other studies such as Faltison and Gunter., (2011)12 who
65
have analysed emissions and CO2 stored at 8 onshore US CO2EOR fields, argue that oil produced
66
through CO2EOR will displace oil produced through other sources and emissions from final product
67
combustion should therefore not be included.12
68
In this work, the net carbon balance of CO2EOR is examined with results from both including and
69
excluding emissions from end product combustion. To date only one study by Hertwich et al.,
70
(2008)13 has attempted to model offshore EOR in the North Sea to quantify emissions from the EOR
71
process. Although they model a theoretical development in a similar fashion to this study only one
72
CO2 storage scenario was considered within the EOR modelling. Hertwich et al., (2008)13 also did not
73
consider emissions from flaring/venting of recycled CO2 and CH4 which we believe may be of
74
significant importance to the carbon balance of a CO2EOR project.
75
Alongside quantifying the emissions specifically related to operating a CO2 EOR project offshore this
76
paper aims to assess the carbon budget performance of a realistic offshore CO2EOR operation with
77
regard to both incremental oil produced and CO2 stored for two EOR scenarios. A common way to 4 ACS Paragon Plus Environment
Page 4 of 26
Page 5 of 26
Environmental Science & Technology
78
assess the storage performance of an operation is to assess how many tonnes of CO2 are stored per
79
barrel of incremental oil produced. Established onshore projects have stored 0.160-
80
0.300tCO2/bbl8,10,11 (3.0-5.7 Mscf/STB). However a number of studies have estimated that CO2EOR
81
operations may store between 0.300-0.600tCO2/bbl (5.7-11.3Mscf/STB) if optimised to store
82
CO2.9,13,14 This study highlights that gross storage rates (not incorporating emissions) at North Sea
83
CO2EOR developments may be significantly higher than storage rates seen at traditional onshore
84
CO2EOR developments.
85
injection beyond the time period needed to optimise oil recovery may be needed to give a North Sea
86
CO2EOR project a negative carbon balance.
87
Although it is technically useful to estimate the storage factor (CO2 stored/bbl of oil produced) and
88
net carbon balance of a project, it must be noted that the assessment of CO2 stored in the carbon
89
balance of a CO2EOR development is based on a number of important assumptions. The most
90
fundamental of these assumptions is the counterfactual that CO2 which is being stored in the EOR
91
operation would otherwise have entered the atmosphere. This assumption allows the CO2 storage
92
element of an operation to be seen to offset the operational emissions and end use emissions of the
93
produced crude (not including emissions from CO2 capture plant and resource mining etc.). If
94
however the CO2 that is stored within the EOR development would have otherwise been captured
95
and stored using pure CO2 storage (no EOR) then the carbon balance calculation should not
96
incorporate the CO2 stored in the EOR development. In this study the former of the two scenarios is
97
used to estimate the net carbon balance. This issue will be examined further in the discussion
98
section of this paper.
99
It is also proposed by a number of studies10,15 that an effective way of assessing the climate benefits
100
of CO2EOR is to estimate the emissions associated with the production of every barrel of oil
101
produced (tCO2e/bbl). This can then be compared to production of oil from other sources. This paper
Interestingly however the study also highlights that CO2 import and
5 ACS Paragon Plus Environment
Environmental Science & Technology
102
presents estimations of the carbon intensity of oil produced through CO2EOR in the UK North Sea for
103
a number of different EOR scenarios.
104
Designing the system boundary
105
To complete a life cycle assessment of CO2EOR operations, a system boundary must be drawn which
106
identifies the processes, activities and materials used within the operation. Once these activities
107
have been identified the emissions associated with each process flow can be modelled.
108
Although CO2EOR relies on many associated activities such as resource mining for construction of a
109
power plant, carbon capture at a power plant, CO2 transport, crude oil refining and crude oil
110
consumption, these processes are all common to either CCS with aquifer storage or conventional
111
crude oil production. The production operations which relate to processes and activities that take
112
place at the production platform are however unique to CO2EOR projects. For this reason a number
113
of studies have attempted to address this phase of the CO2EOR chain.7,8,13 Advanced Resources
114
International (2010)14 found that in traditional onshore projects, upstream CO2EOR operations are
115
dominated by three energy demanding processes. CO2 compression has the largest contribution to
116
energy demand and therefore associated emissions. Although variable from project to project, gas
117
separation and artificial lifting also significantly contribute to the energy demands of CO2EOR
118
operations.14 Hertwich et al., (2008)13 found that the emissions associated with these energy
119
intensive processes were largely controlled by whether equipment was powered by gas/diesel
120
turbines or connected to a larger electricity grid. This paper also presents the results from a ‘gate to
121
gate’ LCA where modelling focused on the offshore production operations, but also incorporates
122
flaring/venting emissions and fugitive emissions. The results from this work may be integrated with
123
a broader ‘cradle to grave’ life cycle assessment (LCA) for the full chain as diplayed below in figure 1.
124
This study includes the emissions associated with aspects of the construction process but focuses
125
on the operational phase. Emissions associated with site evaluation and characterisation,
6 ACS Paragon Plus Environment
Page 6 of 26
Page 7 of 26
Environmental Science & Technology
126
construction, closure and post closure monitoring have been found by Dilmore8 to contribute less
127
than 1% of green house gases emitted in association with CO2EOR activity. The remaining 99% can
128
be attributed to the operational phase of the project.
129
The CO2EOR operation modelled is based on a theoretical oil field development that has experienced
130
secondary production (water flood). Therefore much of the infrastructure that would be required to
131
develop a green field development is already in place. Only processes and infrastructure associated
132
with incremental tertiary production are recognised within this study. For the same reason oil
133
production included in this study relates only to incremental oil produced through CO2EOR activities.
134
It is however recognised that base oil (non EOR oil) production may continue throughout the lifetime
135
of the tertiary EOR operation. Although LCA’s are often compiled to cover many environmental
136
impacts along a production chain, such as global warming potential, human toxicity, land use,
137
pollutants and acidification, this study only accounts for the large volume greenhouse gases (see
138
figure 1).
139 140 Figure 1 – Above- The CCS full production chain. This study will focus on the offshore production operations which are unique to a CO2EOR operation.
141 7
Below- Simplified overview of the system boundary included in the study. CO2(r) and CH4 (r) relate to recycled gases. Input CO2 relates to CO2 imported from a carbon capture plant and input CH4 relates to natural gas that is imported to power production equipment. Only gases that are within the red box are included in this study.ACS Paragon Plus Environment
Environmental Science & Technology
142
143
144
Overview of the EOR Process
145
The inputs to the CO2EOR system included within the system boundary are CO2 and CH4. Pure dense
146
phase CO2 from an anthropogenic source is delivered to the platform through a dedicated CO2
147
pipeline. A separate pipeline delivers CH4 to be utilised as a fuel gas to power offshore operations.
148
Within this study the source and transport distance of the anthropogenic CO2 do not affect the CO2
149
operations and therefore these parameters are excluded from this analysis. When transporting CO2
150
by pipeline the CO2 is normally compressed to a pressure (+9.6 MPa) where it will be in a
151
supercritical phase. This allows larger volumes of CO2 to be transported in a smaller diameter pipe.
152
When delivered at the platform CO2 may have to be recompressed by additional CO2 pumps on the
153
platform to the required injection pressure.16
154
In the modelled scenarios, CO2 is injected continuously into the reservoir. Although established
155
projects have injected water at specific intervals to sweep the reservoir, this study presumes no
156
water injection is undertaken. For a time CO2 injection commences and all injected CO2 is stored
157
without instantaneous enhanced oil production. After a period incremental oil production will occur
158
due to the increase of pressure in the reservoir as a result of CO2 injection. However oil production
159
does not solely increase due to the increase in pressure. At pressures above what is termed the
160
minimum miscibility pressure (MMP) CO2 reduces oil viscosity, increasing oil mobility.17 This
161
production, which is known as miscible CO2 displacement, results in dissolved miscible CO2 being
162
produced at the production well, alongside some free phase CO2, after a period of time. The volume
163
of CO2 reproduced at the production wells will increase over time until the gas recycle capacity of
164
the facility is reached. It must be made clear that although the retention of injected CO2 will not be
165
100% as CO2 recycling will inevitably occur, with historically around 50% of injected CO2 being
8 ACS Paragon Plus Environment
Page 8 of 26
Page 9 of 26
Environmental Science & Technology
166
retained in the reservoir,18 almost all of the imported CO2 (minus CO2 lost through fugitive emissions
167
and flaring/venting) will be stored in reservoir. After CO2 breakthrough a three phase mixture of oil
168
(with associated CH4), CO2 and water is produced at the platform. Using a number of processes the
169
crude oil and water is separated from the CH4 and CO2 which is now in a gaseous phase. The crude
170
oil can then be exported, and the water treated and disposed overboard. The operator can then
171
choose to either re-inject the CH4 / CO2 mixture alongside the fresh CO2 delivered to the platform, or
172
separate the CH4 from the CO2 to be used as a fuel gas or for export. A fourth phase of solid
173
asphaltene precipitate may also be present.19
174
In this process of CO2 injection and recycling, emissions to atmosphere come from a number of
175
sources. Emissions of CO2 are released from production process equipment, which are assumed to
176
be powered by gas turbines. Fugitive emissions of CO2 and CH4 relate to unintentional leaks from
177
valves and seals. Emissions of CO2 and CH4 occur when produced gas is flared or vented during upset
178
operating conditions or for maintenance. In some cases fugitive emissions can be used also to
179
describe emissions from flaring or venting. Within this work they are regarded as separate entities.
180
Methods
181
Model design
182
The study models an ‘anchor field’ development that would be the first CO2EOR project in an area,
183
with a field big enough to accommodate the CO2 supply from a commercial size carbon capture
184
plant.20 The theoretical field modelled here is designed to accept CO2 from a full scale carbon
185
capture project with a gross power output of around 1GW and CO2 output of around 5Mt per
186
annum. The CO2 supply of around 5Mt/yr is here considered to be required to allow EOR project
187
economics to be justifiable. Two scenarios have been developed which represent CO2EOR operations
188
with varying goals. EOR case 1 represents a scenario where the CO2 supply (5Mt/yr) is diverted to
189
another field after 10 years and only the injection of recycled CO2 continues for a further 10 years.
9 ACS Paragon Plus Environment
Environmental Science & Technology
190
For economic reasons, this may be the most likely scenario for an operator focused on oil
191
production, as continuation of CO2 import for a further period is here not modelled to increase the
192
oil production profile (100MMbbl in both cases). EOR case 2 represents a scenario where CO2 is
193
continuously supplied to the field for 20 years and is injected alongside recycled CO2. This scenario
194
represents an injection strategy optimised for CO2 storage. The design assumptions such as recycle
195
rate, percentage of reproduced gas flared/vented and percentage of CH4 in the reproduced gas
196
stream remain constant in both scenarios.
197
To allow estimates of both injected gas volumes and produced gas and fluids for each EOR scenario
198
to be made, simplified reservoir numerical models were constructed in Microsoft Excel (See figures
199
S1 and S2 in the Supporting Information). These models allow predictions of inputs and outputs
200
within the system boundary on an annual basis. The primary input to the model is based on
201
importing and injecting 5Mt (minus fugitive emissions) of CO2 annually with 100MMbbl of
202
incremental oil being produced over the twenty year period. EOR case 1 is based on an oil
203
production rate of 2bbl of incremental oil produced per tonne of CO2 imported. As discussed later in
204
the paper other studies have found utilisation rates of between 1.7-6.5 bbl of incremental /tonne of
205
CO2 imported.12,21,22 Here a rate of 2bbl incremental / tonne of CO2 imported is used as a
206
conservative estimate. Although an oil production rate is estimated over the 20 year period it is here
207
modelled to have no control on operational emissions as they are based on gas recycling rates and
208
cumulative oil production. As described above EOR case 2 is based on importing and injecting 5Mt of
209
CO2 for a further 10 years. However it is modelled here that because incremental oil production is
210
already at its maximum, supplementary CO2 import and injection of CO2 will not lead to an increase
211
in oil production. Details of the equations and assumptions used in the model for the two EOR cases
212
can be found in the Supporting Information (Section S1).
213
The basic outputs of the model which allow emission estimations to be made are annual volume of
214
CH4 produced and the annual volume of CO2 recycled. The volume of CO2 recycled is based on a
10 ACS Paragon Plus Environment
Page 10 of 26
Page 11 of 26
Environmental Science & Technology
215
maximum recycle rate of 600mmscfd which is reached over a period of 5 years and then remains
216
constant for the remainder of the 20 year time frame. The ratio of CH4 to CO2 in the reproduced gas
217
was modelled to increase from 100% CH4 to 10% CH4 over a period of 5 years, with CO2 making up
218
the remaining 90%. This assumption is based on work by Goodyear et al., (2011)16 who suggest that
219
CO2 may make up 90mol% of the gas in the produced gas stream. An important assumption made
220
within the models is that recycled volumes are constant in both EOR cases even though volumes of
221
injected CO2 over the 20 year project life are different. This is based on the assumption that the
222
maximum recycle rate will be reached and sustained in both EOR cases, and is not a direct function
223
of injected CO2 volumes. Although these models were constructed through time, the results
224
presented in this paper focus predominantly on the cumulative CO2 and CH4 emissions and CO2
225
storage at the end of the 20 year project life. These outputs of volumes of recycled CO2 and CH4 are
226
then be used to calculate emissions for each EOR case using formulas described in the Supporting
227
Information (Section S2).
228
Input Data
229
Specicfic design related data was acquired after communication with a CO2EOR developer. For the
230
modelled anchor project a dedicated gas supply pipeline is used to supply fuel for gas turbines. A gas
231
turbine emission factor of 610g/kwhr was used.23 Since there are no current CO2EOR operations in
232
the United Kingdom Continental Shelf (UKCS), assumptions have been made about the processes
233
required for the modelled anchor project, and their associated energy requirements. Details of
234
assumptions can be found in the Supplementary Information (Table S1). This study estimates that
235
the power demand for additional CO2 infrastructure lies between 18-22kwhr/tonne CO2 recyled. This
236
is slightly lower but broadly similar to the energy use figures stated by the Pembina Institute (2013)22
237
who suggest a range of 35-120 kwhr/tonne CO2 recycled.
238
A large percentage of the modelled operations energy demand comes from the recycle process
239
(66%) (Table S1). Here the produced CO2 + CH4 mix is separated from the produced crude and brine
11 ACS Paragon Plus Environment
Environmental Science & Technology
240
and recompressed for injection. Additional CO2 pumps may also be required to increase the pressure
241
of delivered CO2 to the required injection pressure (additional compression – Table S1). Fuel gas
242
separation, although included in Table S1 is not included in the modelled scenarios. Here it is
243
assumed that the CAPEX and module weight of gas treatment units, with current technology, is too
244
high to be justified for an offshore environment, and therefore produced hydrocarbon gases will be
245
re-injected.16
246
required as after CO2 breakthrough the wells will ‘auto-lift’.16 This is the same assumption used by
247
Hertwich et al. (2008)13.
248
(Table S1) from steel required to build a new bridge linked platform are also incorporated and are
249
displayed in the results section of this paper.
250
Flaring / venting & fugitive emissions
251
In this study an average figure for the venting/flaring of reproduced gas (CO2 + CH4) was calculated
252
using flaring rates from UK North Sea oil fields in 2011 where on average 3.5% of produced gas was
253
flared or vented, due to maintenance and safety reasons.24 Emissions associated with flaring and
254
venting were calculated by applying this percentage to the volume of annually reproduced gas
255
predicted in the Excel models described previously. This is a parameter that a number of other
256
studies have disregarded when modelling CO2EOR.10-14 Due to the uncertainty in this value a range of
257
flaring/venting rates of 1-5% is here used (See Supporting Information section S3 for further details
258
on this parameter). The control of this parameter is also discussed in the Discussion section of this
259
paper.
260
Little data is currently available relating to fugitive emissions from CO2EOR projects. In Dilmore8 a
261
range of 0-1% loss of purchased CO2 is assumed. Personal communication with US operators also
262
revealed that an estimated 1-2% of purchased CO2 is lost to fugitive emissions. Here, total fugitive
263
CO2 emissions are modelled to be 1% of CO2 imported to the platform. Given the large uncertainty of
264
this parameter a range of 0.5-5% is used to give lower and upper estimates.
For simplification, all the modelled scenarios assume that ‘artificial lift’ is not
Emissions associated with drilling new wells and embedded emissions
12 ACS Paragon Plus Environment
Page 12 of 26
Page 13 of 26
Environmental Science & Technology
265
266
Crude transport, refining and combustion
267
To quantify the emissions associated with the incremental oil production, a number of emission
268
factors were extracted from the literature. Downstream emissions from crude oil production can be
269
broadly grouped into three catagories; transport, refining and combustion. An emission factor of
270
0.004 (0.0011-0.0119) tCO2/bbl was used for crude oil transport,14,25 0.03(0.027-0.05) tCO2/bbl for
271
crude oil refining10,14,21,26,27 and 0.431 (0.428-0.433) tCO2/bbl for final product combustion.27 For
272
further details and justifcation of the ranges used for each emission factor please see the Supporting
273
Information (section S4).
274
Results
275
Cumulative emissions
276
Total cumulative emissions for the 20 year project life time for both EOR cases are displayed in table
277
1. Total emissions from EOR case 1 accumulate to 13.5 (5.2-20.5) MtCO2e over the 20 year life of the
278
project. Total cumulative emissions from EOR case 2 although larger at 13.7 (5.5-23.1) MtCO2e, are
279
relatively similar considering case 2 represents an additional injection of around 5MtCO2 per year for
280
ten years. This similarity is due to the largest contribution of emissions arising from the recycle
281
process which remains constant in both cases.
282
As can be seen in both EOR scenarios venting of CO2 and CH4 has the largest contribution (~80%) to
283
greenhouse gas emissions over the 20 year life time of the project. This is due to the high levels of
284
CO2 in the produced gas stream preventing ignition as discussed in the Supporting Information.
285
Emissions associated with drilling new wells, working over old wells and for the manufacturing of a
286
new bridge link platform, which are assumed to all occur prior to CO2 injection, have a very small
287
contribution to overall emissions at
Article
Can producing oil store carbon? Greenhouse Gas footprint of CO2EOR, offshore North Sea R Jamie Stewart, and R Stuart Haszeldine Environ. Sci. Technol., Just Accepted Manuscript • Publication Date (Web): 19 Mar 2015 Downloaded from http://pubs.acs.org on March 20, 2015
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 26
Environmental Science & Technology
1
Can producing oil store carbon? Greenhouse Gas
2
footprint of CO2EOR, offshore North Sea
3
4
R Jamie Stewart*, R Stuart Haszeldine
5
Scottish Carbon Capture & Storage,
6
School of GeoSciences, University of Edinburgh,
7
EH9 3JW, Scotland, UK
8
9
*Corresponding author – email: [email protected], phone: +44 (0)131 650 5936,
10
fax: +44 (0)131 650 7340
11
12
1 ACS Paragon Plus Environment
Environmental Science & Technology
13
Abstract
14
Carbon dioxide enhanced oil recovery (CO2EOR) is a proven and available technology used
15
to produce incremental oil from depleted fields whilst permanently storing large tonnages
16
of injected CO2. Although this technology has been used successfully onshore in North
17
America and Europe, there are currently no CO2EOR projects in the United Kingdom. Here,
18
we examine whether offshore CO2EOR can store more CO2 than onshore projects
19
traditionally have, and whether CO2 storage can offset additional emissions produced
20
through offshore operations and incremental oil production. Using a high-level Life Cycle
21
system approach, we find that the largest contribution to offshore emissions is from flaring
22
or venting of reproduced CH4 and CO2. These can already be greatly reduced by regulation.
23
If CO2 injection is continued after oil production has been optimised, then offshore CO2EOR
24
has the potential to be carbon negative - even when emissions from refining, transport and
25
combustion of produced crude oil are included. The carbon intensity of oil produced can be
26
just 0.056-0.062 tCO2e/bbl if flaring/venting is reduced by regulation. This compares against
27
conventional Saudi oil 0.040tCO2e/bbl, or mined shale oil >0.300tCO2e/bbl.
28 Abstract Art 29 2 ACS Paragon Plus Environment
Page 2 of 26
Page 3 of 26
Environmental Science & Technology
30
Introduction
31
Carbon Dioxide Enhanced Oil Recovery (CO2EOR) may be an option for reducing the cost of early-
32
stage carbon capture and storage (CCS) as more and more projects around the European Union (EU)
33
utilising saline aquifer storage are deemed too expensive and fail to go beyond Front End
34
Engineering Design (FEED).
35
CO2EOR is a technique that has been applied in the United States (US) since the mid 80’s to enhance
36
oil recovery from mature fields.1 In the United States where large volumes of CO2 are readily
37
available from natural sources, CO2 has been injected as the primary recovery agent in these EOR
38
operations. In Europe however the lack in availability of low cost CO2 has meant that EOR operations
39
have conventionally used water injection or methane gas injection, with currently no CO2EOR
40
projects in an offshore environment.2-5 With the prospect of Carbon Capture technology developing
41
across the EU however, the required volumes of CO2 may become available through anthropogenic
42
sources.6 CO2EOR is regarded as an option for storing large volumes of this captured CO2 with the
43
added benefit of improving the recovery rate from depleted oil fields. This oil production may hold
44
potential to improve CCS project economics and get projects beyond FEED.
45
However it is known from currently operating CO2EOR projects onshore that the operations and
46
processes involved in CO2EOR are energy intensive and could result in significant atmospheric
47
emissions.7,8 This paper presents the conclusions from a medium to high level life cycle assessment
48
of CO2EOR operations, in which emissions and CO2 stored are modelled for a theoretical offshore
49
North Sea project. Consideration of the uncertainties involved in operating a CO2EOR project in an
50
offshore environment means a number of scenarios and uncertainty ranges are presented. Particular
51
attention is paid to the flaring and venting of reproduced gases which is found to have a large
52
control on the emissions associated with the production of oil through CO2EOR.
3 ACS Paragon Plus Environment
Environmental Science & Technology
53
Several other studies have attempted to quantify how the emissions from EOR operations, and from
54
the final combustion of the petroleum products, relate to the total volume of CO2 stored in EOR
55
projects.9-12 These studies focus on historical data from onshore developments and vary widely in
56
their outcomes.
57
The primary reason for the varying conclusions relates to inclusion or exclusion of the large
58
contribution to emissions made by the combustion of produced petroleum products, and the
59
principal of additionality. Jaramillo et al., (2009)10 who looked at 5 onshore North American CO2 EOR
60
projects found that when emissions from the full system boundary, from coal mining to final product
61
combustion are included, then onshore CO2 EOR projects have historically been net emitters of CO2.
62
They assumed that oil produced through CO2EOR is additional to the global system and therefore
63
emissions from the combustion of the final petroleum products should be included in the study,
64
which resulted in overall net emissions.10,11 Other studies such as Faltison and Gunter., (2011)12 who
65
have analysed emissions and CO2 stored at 8 onshore US CO2EOR fields, argue that oil produced
66
through CO2EOR will displace oil produced through other sources and emissions from final product
67
combustion should therefore not be included.12
68
In this work, the net carbon balance of CO2EOR is examined with results from both including and
69
excluding emissions from end product combustion. To date only one study by Hertwich et al.,
70
(2008)13 has attempted to model offshore EOR in the North Sea to quantify emissions from the EOR
71
process. Although they model a theoretical development in a similar fashion to this study only one
72
CO2 storage scenario was considered within the EOR modelling. Hertwich et al., (2008)13 also did not
73
consider emissions from flaring/venting of recycled CO2 and CH4 which we believe may be of
74
significant importance to the carbon balance of a CO2EOR project.
75
Alongside quantifying the emissions specifically related to operating a CO2 EOR project offshore this
76
paper aims to assess the carbon budget performance of a realistic offshore CO2EOR operation with
77
regard to both incremental oil produced and CO2 stored for two EOR scenarios. A common way to 4 ACS Paragon Plus Environment
Page 4 of 26
Page 5 of 26
Environmental Science & Technology
78
assess the storage performance of an operation is to assess how many tonnes of CO2 are stored per
79
barrel of incremental oil produced. Established onshore projects have stored 0.160-
80
0.300tCO2/bbl8,10,11 (3.0-5.7 Mscf/STB). However a number of studies have estimated that CO2EOR
81
operations may store between 0.300-0.600tCO2/bbl (5.7-11.3Mscf/STB) if optimised to store
82
CO2.9,13,14 This study highlights that gross storage rates (not incorporating emissions) at North Sea
83
CO2EOR developments may be significantly higher than storage rates seen at traditional onshore
84
CO2EOR developments.
85
injection beyond the time period needed to optimise oil recovery may be needed to give a North Sea
86
CO2EOR project a negative carbon balance.
87
Although it is technically useful to estimate the storage factor (CO2 stored/bbl of oil produced) and
88
net carbon balance of a project, it must be noted that the assessment of CO2 stored in the carbon
89
balance of a CO2EOR development is based on a number of important assumptions. The most
90
fundamental of these assumptions is the counterfactual that CO2 which is being stored in the EOR
91
operation would otherwise have entered the atmosphere. This assumption allows the CO2 storage
92
element of an operation to be seen to offset the operational emissions and end use emissions of the
93
produced crude (not including emissions from CO2 capture plant and resource mining etc.). If
94
however the CO2 that is stored within the EOR development would have otherwise been captured
95
and stored using pure CO2 storage (no EOR) then the carbon balance calculation should not
96
incorporate the CO2 stored in the EOR development. In this study the former of the two scenarios is
97
used to estimate the net carbon balance. This issue will be examined further in the discussion
98
section of this paper.
99
It is also proposed by a number of studies10,15 that an effective way of assessing the climate benefits
100
of CO2EOR is to estimate the emissions associated with the production of every barrel of oil
101
produced (tCO2e/bbl). This can then be compared to production of oil from other sources. This paper
Interestingly however the study also highlights that CO2 import and
5 ACS Paragon Plus Environment
Environmental Science & Technology
102
presents estimations of the carbon intensity of oil produced through CO2EOR in the UK North Sea for
103
a number of different EOR scenarios.
104
Designing the system boundary
105
To complete a life cycle assessment of CO2EOR operations, a system boundary must be drawn which
106
identifies the processes, activities and materials used within the operation. Once these activities
107
have been identified the emissions associated with each process flow can be modelled.
108
Although CO2EOR relies on many associated activities such as resource mining for construction of a
109
power plant, carbon capture at a power plant, CO2 transport, crude oil refining and crude oil
110
consumption, these processes are all common to either CCS with aquifer storage or conventional
111
crude oil production. The production operations which relate to processes and activities that take
112
place at the production platform are however unique to CO2EOR projects. For this reason a number
113
of studies have attempted to address this phase of the CO2EOR chain.7,8,13 Advanced Resources
114
International (2010)14 found that in traditional onshore projects, upstream CO2EOR operations are
115
dominated by three energy demanding processes. CO2 compression has the largest contribution to
116
energy demand and therefore associated emissions. Although variable from project to project, gas
117
separation and artificial lifting also significantly contribute to the energy demands of CO2EOR
118
operations.14 Hertwich et al., (2008)13 found that the emissions associated with these energy
119
intensive processes were largely controlled by whether equipment was powered by gas/diesel
120
turbines or connected to a larger electricity grid. This paper also presents the results from a ‘gate to
121
gate’ LCA where modelling focused on the offshore production operations, but also incorporates
122
flaring/venting emissions and fugitive emissions. The results from this work may be integrated with
123
a broader ‘cradle to grave’ life cycle assessment (LCA) for the full chain as diplayed below in figure 1.
124
This study includes the emissions associated with aspects of the construction process but focuses
125
on the operational phase. Emissions associated with site evaluation and characterisation,
6 ACS Paragon Plus Environment
Page 6 of 26
Page 7 of 26
Environmental Science & Technology
126
construction, closure and post closure monitoring have been found by Dilmore8 to contribute less
127
than 1% of green house gases emitted in association with CO2EOR activity. The remaining 99% can
128
be attributed to the operational phase of the project.
129
The CO2EOR operation modelled is based on a theoretical oil field development that has experienced
130
secondary production (water flood). Therefore much of the infrastructure that would be required to
131
develop a green field development is already in place. Only processes and infrastructure associated
132
with incremental tertiary production are recognised within this study. For the same reason oil
133
production included in this study relates only to incremental oil produced through CO2EOR activities.
134
It is however recognised that base oil (non EOR oil) production may continue throughout the lifetime
135
of the tertiary EOR operation. Although LCA’s are often compiled to cover many environmental
136
impacts along a production chain, such as global warming potential, human toxicity, land use,
137
pollutants and acidification, this study only accounts for the large volume greenhouse gases (see
138
figure 1).
139 140 Figure 1 – Above- The CCS full production chain. This study will focus on the offshore production operations which are unique to a CO2EOR operation.
141 7
Below- Simplified overview of the system boundary included in the study. CO2(r) and CH4 (r) relate to recycled gases. Input CO2 relates to CO2 imported from a carbon capture plant and input CH4 relates to natural gas that is imported to power production equipment. Only gases that are within the red box are included in this study.ACS Paragon Plus Environment
Environmental Science & Technology
142
143
144
Overview of the EOR Process
145
The inputs to the CO2EOR system included within the system boundary are CO2 and CH4. Pure dense
146
phase CO2 from an anthropogenic source is delivered to the platform through a dedicated CO2
147
pipeline. A separate pipeline delivers CH4 to be utilised as a fuel gas to power offshore operations.
148
Within this study the source and transport distance of the anthropogenic CO2 do not affect the CO2
149
operations and therefore these parameters are excluded from this analysis. When transporting CO2
150
by pipeline the CO2 is normally compressed to a pressure (+9.6 MPa) where it will be in a
151
supercritical phase. This allows larger volumes of CO2 to be transported in a smaller diameter pipe.
152
When delivered at the platform CO2 may have to be recompressed by additional CO2 pumps on the
153
platform to the required injection pressure.16
154
In the modelled scenarios, CO2 is injected continuously into the reservoir. Although established
155
projects have injected water at specific intervals to sweep the reservoir, this study presumes no
156
water injection is undertaken. For a time CO2 injection commences and all injected CO2 is stored
157
without instantaneous enhanced oil production. After a period incremental oil production will occur
158
due to the increase of pressure in the reservoir as a result of CO2 injection. However oil production
159
does not solely increase due to the increase in pressure. At pressures above what is termed the
160
minimum miscibility pressure (MMP) CO2 reduces oil viscosity, increasing oil mobility.17 This
161
production, which is known as miscible CO2 displacement, results in dissolved miscible CO2 being
162
produced at the production well, alongside some free phase CO2, after a period of time. The volume
163
of CO2 reproduced at the production wells will increase over time until the gas recycle capacity of
164
the facility is reached. It must be made clear that although the retention of injected CO2 will not be
165
100% as CO2 recycling will inevitably occur, with historically around 50% of injected CO2 being
8 ACS Paragon Plus Environment
Page 8 of 26
Page 9 of 26
Environmental Science & Technology
166
retained in the reservoir,18 almost all of the imported CO2 (minus CO2 lost through fugitive emissions
167
and flaring/venting) will be stored in reservoir. After CO2 breakthrough a three phase mixture of oil
168
(with associated CH4), CO2 and water is produced at the platform. Using a number of processes the
169
crude oil and water is separated from the CH4 and CO2 which is now in a gaseous phase. The crude
170
oil can then be exported, and the water treated and disposed overboard. The operator can then
171
choose to either re-inject the CH4 / CO2 mixture alongside the fresh CO2 delivered to the platform, or
172
separate the CH4 from the CO2 to be used as a fuel gas or for export. A fourth phase of solid
173
asphaltene precipitate may also be present.19
174
In this process of CO2 injection and recycling, emissions to atmosphere come from a number of
175
sources. Emissions of CO2 are released from production process equipment, which are assumed to
176
be powered by gas turbines. Fugitive emissions of CO2 and CH4 relate to unintentional leaks from
177
valves and seals. Emissions of CO2 and CH4 occur when produced gas is flared or vented during upset
178
operating conditions or for maintenance. In some cases fugitive emissions can be used also to
179
describe emissions from flaring or venting. Within this work they are regarded as separate entities.
180
Methods
181
Model design
182
The study models an ‘anchor field’ development that would be the first CO2EOR project in an area,
183
with a field big enough to accommodate the CO2 supply from a commercial size carbon capture
184
plant.20 The theoretical field modelled here is designed to accept CO2 from a full scale carbon
185
capture project with a gross power output of around 1GW and CO2 output of around 5Mt per
186
annum. The CO2 supply of around 5Mt/yr is here considered to be required to allow EOR project
187
economics to be justifiable. Two scenarios have been developed which represent CO2EOR operations
188
with varying goals. EOR case 1 represents a scenario where the CO2 supply (5Mt/yr) is diverted to
189
another field after 10 years and only the injection of recycled CO2 continues for a further 10 years.
9 ACS Paragon Plus Environment
Environmental Science & Technology
190
For economic reasons, this may be the most likely scenario for an operator focused on oil
191
production, as continuation of CO2 import for a further period is here not modelled to increase the
192
oil production profile (100MMbbl in both cases). EOR case 2 represents a scenario where CO2 is
193
continuously supplied to the field for 20 years and is injected alongside recycled CO2. This scenario
194
represents an injection strategy optimised for CO2 storage. The design assumptions such as recycle
195
rate, percentage of reproduced gas flared/vented and percentage of CH4 in the reproduced gas
196
stream remain constant in both scenarios.
197
To allow estimates of both injected gas volumes and produced gas and fluids for each EOR scenario
198
to be made, simplified reservoir numerical models were constructed in Microsoft Excel (See figures
199
S1 and S2 in the Supporting Information). These models allow predictions of inputs and outputs
200
within the system boundary on an annual basis. The primary input to the model is based on
201
importing and injecting 5Mt (minus fugitive emissions) of CO2 annually with 100MMbbl of
202
incremental oil being produced over the twenty year period. EOR case 1 is based on an oil
203
production rate of 2bbl of incremental oil produced per tonne of CO2 imported. As discussed later in
204
the paper other studies have found utilisation rates of between 1.7-6.5 bbl of incremental /tonne of
205
CO2 imported.12,21,22 Here a rate of 2bbl incremental / tonne of CO2 imported is used as a
206
conservative estimate. Although an oil production rate is estimated over the 20 year period it is here
207
modelled to have no control on operational emissions as they are based on gas recycling rates and
208
cumulative oil production. As described above EOR case 2 is based on importing and injecting 5Mt of
209
CO2 for a further 10 years. However it is modelled here that because incremental oil production is
210
already at its maximum, supplementary CO2 import and injection of CO2 will not lead to an increase
211
in oil production. Details of the equations and assumptions used in the model for the two EOR cases
212
can be found in the Supporting Information (Section S1).
213
The basic outputs of the model which allow emission estimations to be made are annual volume of
214
CH4 produced and the annual volume of CO2 recycled. The volume of CO2 recycled is based on a
10 ACS Paragon Plus Environment
Page 10 of 26
Page 11 of 26
Environmental Science & Technology
215
maximum recycle rate of 600mmscfd which is reached over a period of 5 years and then remains
216
constant for the remainder of the 20 year time frame. The ratio of CH4 to CO2 in the reproduced gas
217
was modelled to increase from 100% CH4 to 10% CH4 over a period of 5 years, with CO2 making up
218
the remaining 90%. This assumption is based on work by Goodyear et al., (2011)16 who suggest that
219
CO2 may make up 90mol% of the gas in the produced gas stream. An important assumption made
220
within the models is that recycled volumes are constant in both EOR cases even though volumes of
221
injected CO2 over the 20 year project life are different. This is based on the assumption that the
222
maximum recycle rate will be reached and sustained in both EOR cases, and is not a direct function
223
of injected CO2 volumes. Although these models were constructed through time, the results
224
presented in this paper focus predominantly on the cumulative CO2 and CH4 emissions and CO2
225
storage at the end of the 20 year project life. These outputs of volumes of recycled CO2 and CH4 are
226
then be used to calculate emissions for each EOR case using formulas described in the Supporting
227
Information (Section S2).
228
Input Data
229
Specicfic design related data was acquired after communication with a CO2EOR developer. For the
230
modelled anchor project a dedicated gas supply pipeline is used to supply fuel for gas turbines. A gas
231
turbine emission factor of 610g/kwhr was used.23 Since there are no current CO2EOR operations in
232
the United Kingdom Continental Shelf (UKCS), assumptions have been made about the processes
233
required for the modelled anchor project, and their associated energy requirements. Details of
234
assumptions can be found in the Supplementary Information (Table S1). This study estimates that
235
the power demand for additional CO2 infrastructure lies between 18-22kwhr/tonne CO2 recyled. This
236
is slightly lower but broadly similar to the energy use figures stated by the Pembina Institute (2013)22
237
who suggest a range of 35-120 kwhr/tonne CO2 recycled.
238
A large percentage of the modelled operations energy demand comes from the recycle process
239
(66%) (Table S1). Here the produced CO2 + CH4 mix is separated from the produced crude and brine
11 ACS Paragon Plus Environment
Environmental Science & Technology
240
and recompressed for injection. Additional CO2 pumps may also be required to increase the pressure
241
of delivered CO2 to the required injection pressure (additional compression – Table S1). Fuel gas
242
separation, although included in Table S1 is not included in the modelled scenarios. Here it is
243
assumed that the CAPEX and module weight of gas treatment units, with current technology, is too
244
high to be justified for an offshore environment, and therefore produced hydrocarbon gases will be
245
re-injected.16
246
required as after CO2 breakthrough the wells will ‘auto-lift’.16 This is the same assumption used by
247
Hertwich et al. (2008)13.
248
(Table S1) from steel required to build a new bridge linked platform are also incorporated and are
249
displayed in the results section of this paper.
250
Flaring / venting & fugitive emissions
251
In this study an average figure for the venting/flaring of reproduced gas (CO2 + CH4) was calculated
252
using flaring rates from UK North Sea oil fields in 2011 where on average 3.5% of produced gas was
253
flared or vented, due to maintenance and safety reasons.24 Emissions associated with flaring and
254
venting were calculated by applying this percentage to the volume of annually reproduced gas
255
predicted in the Excel models described previously. This is a parameter that a number of other
256
studies have disregarded when modelling CO2EOR.10-14 Due to the uncertainty in this value a range of
257
flaring/venting rates of 1-5% is here used (See Supporting Information section S3 for further details
258
on this parameter). The control of this parameter is also discussed in the Discussion section of this
259
paper.
260
Little data is currently available relating to fugitive emissions from CO2EOR projects. In Dilmore8 a
261
range of 0-1% loss of purchased CO2 is assumed. Personal communication with US operators also
262
revealed that an estimated 1-2% of purchased CO2 is lost to fugitive emissions. Here, total fugitive
263
CO2 emissions are modelled to be 1% of CO2 imported to the platform. Given the large uncertainty of
264
this parameter a range of 0.5-5% is used to give lower and upper estimates.
For simplification, all the modelled scenarios assume that ‘artificial lift’ is not
Emissions associated with drilling new wells and embedded emissions
12 ACS Paragon Plus Environment
Page 12 of 26
Page 13 of 26
Environmental Science & Technology
265
266
Crude transport, refining and combustion
267
To quantify the emissions associated with the incremental oil production, a number of emission
268
factors were extracted from the literature. Downstream emissions from crude oil production can be
269
broadly grouped into three catagories; transport, refining and combustion. An emission factor of
270
0.004 (0.0011-0.0119) tCO2/bbl was used for crude oil transport,14,25 0.03(0.027-0.05) tCO2/bbl for
271
crude oil refining10,14,21,26,27 and 0.431 (0.428-0.433) tCO2/bbl for final product combustion.27 For
272
further details and justifcation of the ranges used for each emission factor please see the Supporting
273
Information (section S4).
274
Results
275
Cumulative emissions
276
Total cumulative emissions for the 20 year project life time for both EOR cases are displayed in table
277
1. Total emissions from EOR case 1 accumulate to 13.5 (5.2-20.5) MtCO2e over the 20 year life of the
278
project. Total cumulative emissions from EOR case 2 although larger at 13.7 (5.5-23.1) MtCO2e, are
279
relatively similar considering case 2 represents an additional injection of around 5MtCO2 per year for
280
ten years. This similarity is due to the largest contribution of emissions arising from the recycle
281
process which remains constant in both cases.
282
As can be seen in both EOR scenarios venting of CO2 and CH4 has the largest contribution (~80%) to
283
greenhouse gas emissions over the 20 year life time of the project. This is due to the high levels of
284
CO2 in the produced gas stream preventing ignition as discussed in the Supporting Information.
285
Emissions associated with drilling new wells, working over old wells and for the manufacturing of a
286
new bridge link platform, which are assumed to all occur prior to CO2 injection, have a very small
287
contribution to overall emissions at
