Reply Brief — Midwest Ozone Group, Applicant v. Environmental Protection Agency, et al.
Supreme Court briefAug 23, 2024
Ask Donna
What actually matters in this document.
Text
No. 24A105
In the Supreme Court of the United States
M IDWEST O ZONE G ROUP ,
Applicant,
v.
E NVIRONMENTAL P ROTECTION A GENCY AND M ICHAEL S. R EGAN ,
A DMINISTRATOR , ET AL.
Respondents.
APPLICANT’S REPLY IN SUPPORT OF EMERGENCY
APPLICATION FOR IMMEDIATE STAY OF FINAL AGENCY
ACTION PENDING DISPOSITION OF PETITION FOR REVIEW
To the Honorable John G. Roberts, Jr.,
Chief Justice of the Supreme Court of the United States and Circuit Justice
for the District of Columbia Circuit
Ancil C. Ramey (Lead Counsel)
David M. Flannery
Kathy G. Beckett
Keeleigh S. Huffman
STEPTOE & JOHNSON PLLC
Post Office Box 1588
Charleston, WV 25326
(304) 353-8000
Ancil.Ramey@Steptoe-Johnson.com
Edward L. Kropp
STEPTOE & JOHNSON PLLC
Post Office Box 36425
Indianapolis, IN 46236
Counsel for Midwest Ozone Group
TABLE OF CONTENTS
Page
TABLE OF CONTENTS ......................................................................................................... i
TABLE OF AUTHORITIES .................................................................................................... ii
INTRODUCTION .......................................................................................................... 1
ARGUMENT .................................................................................................................. 2
I.
This Case Warrants the Court’s Discretionary Review and
Applicants Are Likely to Succeed on the Merits in this Case ....................... 2
A.
Failure to demonstrate feasibility with respect to CCS and cofiring.......................................................................................................... 2
II. Grid Operators, Power Companies and Members of Congress All
Find That The Rule Presents A Significant Near-Term Threat To
The Reliability Of The Electric Power Grid ................................................... 6
III. The Balance of the Equities and the Public Interest Favor an
Immediate Stay ............................................................................................. 11
CONCLUSION............................................................................................................. 12
i
TABLE OF AUTHORITIES
Page(s)
Cases
Hollingsworth v. Perry, 558 U.S. 183, 190 (2010) ........................................................ 2
West Virginia v. EPA, 597 U.S. 697 (2022) ......................................................... 2, 3, 10
Statutes, Rules and Regulations
18 CFR 292.204(b)(1) ..................................................................................................... 5
42 U.S.C. §7411(a)(1) ............................................................................... 1, 2, 3, 5, 6, 10
Clean Air Act Section 111(b)(2) ..................................................................................... 5
89 Fed. Reg. 39,798, 40,005 – 40,006 (May 9, 2024) .................................................... 1
89 Fed. Reg. 39,803 .................................................................................................... 6, 9
89 Fed. Reg. 39,927 ........................................................................................................ 3
89 Fed. Reg. 40,011-20................................................................................................... 8
Other Authorities
Application For Immediate Stay Of Final Agency Action During Pendency Of
Petitions For Review, National Mining Association and America's Power
v. EPA (U.S. 24A97) at 16 (filed July 24, 2024) ...................................................... 5
Motion For Stay State of West Virginia, State of Indiana, et al. v. EPA (241120), at 20-21 (D.C. Cir. filed May 13, 2024) (denied July 19, 2024) ................... 1
SPP Statement On The Recent EPA Greenhouse Gas Emissions Rule (May
20, 2024) (access at https://tinyurl.com/w9bja7ku) ................................................. 7
ii
INTRODUCTION
EPA is statutorily obligated under Section 111 of the Clean Air Act to take
“energy requirements” into account and to select technology that has been
“adequately demonstrated.” This Rule fails these criteria although EPA and the
Respondent-Intervenors attempt to articulate otherwise.
An immediate stay is necessary to preserve the status quo pending litigation
of the merits of the Rule which would otherwise impose significant and irreparable
harm on sources affected by the Rule including the membership of the Midwest Ozone
Group. Contrary to the assertions of Respondents, there will be little to no harm to
EPA, Intervenors, or the public from an immediate stay pending judicial review since
current state and federal programs regulate these sources and related emissions.
No one disputes that the Rule targeting greenhouse gases from the affected
sources will cost billions of dollars that cannot be recovered. 89 Fed. Reg. 39,798,
40,005 – 40,006 (May 9, 2024). Those costs begin now. Additionally, while the Rule
does not require emission reductions for at least three years and compliance deadlines
are years out, a stay of the Rule is necessary to avoid the irreparable harm that would
be caused by the need to initiate immediate planning and related acquisitions, and
which do not support the claim of harm to the public interest. See Motion For Stay
State of West Virginia, State of Indiana, et al. v. EPA (24-1120), at 20-21 (D.C. Cir.
filed May 13, 2024) (denied July 19, 2024).
1
It is essential to grant an immediate stay to protect the regulated community
from harm while the litigation uncovers EPA’s contortions orchestrated to disguise
its slight of the law found in this Rule.
Midwest Ozone Group supports and incorporates herein the replies filed by
Applicants who support an immediate stay of the entire Rule.
ARGUMENT
I. This Case Warrants the Court’s Discretionary Review and Applicants
Are Likely to Succeed on the Merits in this Case.
There is a reasonable probability that four Justices will consider the issues
raised in this case sufficiently important to grant certiorari. See Hollingsworth v.
Perry, 558 U.S. 183, 190 (2010). EPA and Respondent-Intervenors try to portray this
case as a routine matter unworthy of this Court’s attention. Yet even the Respondent
Parties claim that the challenged rulemaking is one of immense importance. EPA
Resp. at 39; State and Municipal Resp. at 39. This Rule is part of a series of rules
that are intended to remove fossil fueled power generation from the grid a goal so
important to EPA that it ignores the law and negative energy impacts (i.e., grid
reliability).
A. Failure to demonstrate feasibility with respect to CCS and co-firing.
The Clean Air Act does not create authority for EPA to develop aspirational
goals nor agency derived missions to effect energy policy or grid reliability, but
instead looks for a “best system” that “has been adequately demonstrated.” 42 U.S.C.
§7411(a)(1) (“Section 111”). EPA has suggested it has “returned to the traditional
source-based approach that the West Virginia Court has used as its benchmark.”
2
EPA Resp. at 14 (referring to West Virginia v. EPA, 597 U.S. 697 (2022)). A
“traditional” Section 111 rule would not require implementation of CCS technology
and the related pipelines when there are no “demonstrated” full-scale CCS projects
that support 90% capture this Rule requires. See SaskPower Proposed Rule
Comments, Doc. ID. EPA-HQ-OAR-2023-0072-0687, App. at 0001a; Energy &
Environmental Research Center Final Report at 6, Doc. ID. EPA-HQ-OAR-20230072-0710, Attachment D, App. at 0011a.
Neither Bellingham nor Mongstad, both cited by EPA as examples of CCS with
90% capture, are what EPA claims they are. Bellingham was a slip stream off of a
much larger plant that did not attempt to operate its CCS facility at full scale and
Mongstad is a Norwegian plant that has operated CCS on a 12 MW flue gas stream
from a natural gas combined cycle cogeneration plant at Mongstad power station.
Neither are exemplars of full scale CCS facilities. See Cichanowicz & Hein Report at
4, Doc. ID. EPA-HQ-OAR-2023-0710, Attachment C, App. at 0032a; see also, 89 Fed.
Reg 39927 at n.768. Further, adequate pipeline infrastructure is not yet available to
support CCS. Power Generators Air Coalition Proposed Rule Comments at 35-37,
Doc. ID EPA-HQ-OAR-2023-0072-0710, App. at 0114-16a.
Joe Goffman has declared that the rule has no near-term harm on source
operations with the exception of planning. Goffman Decl., ¶105, App. at 0248-49a.
EPA asserts there are no plans that need to be made at this time. EPA Resp. 54. The
regulated industry holds an entirely different view of the immediacy of the
operational and financial complexities of planning. See McLennan ¶55 (“Complying
3
with the Final Rule’s 20232 deadline will require Minnkota to immediately begin
design and development and negotiate project contracts by the end of the current
calendar year (at the very latest)”), App. at 0326a, Grooms ¶26 (explaining that
imminent retirement is the only option for Cardinal because “the Final Rule’s other
purported compliance pathways are not demonstrated and are not commercially
viable….”), App. at 0359a, Hasten ¶31 (providing that there will be a $3 billion
investment in additional generation because of this Rule, which will be felt by
ratepayers), App. at 0403-04a, Soderberg ¶29 (stating that a race dynamic will be
created by the Rule, where EGUS will be required to purchase equipment as early as
possible to attempt to avoid processing and delivery delays by inundated equipment
manufacturers), App. at 0430a, Tudor ¶37 (explaining that federal Rural Utilities
Service is a very lengthy process and there is “serious doubt” whether this can be
navigated given the compliance timelines in the Rule.), App. at 0457a, McCollam ¶64
(“Working backwards from the Final Rule’s compliance dates, the engineering
should have already begun.”), App. at 0499a. The Agency also asserts that EPA
alone is empowered by the Clean Air Act with the task of making technical judgments
about achievability and cost. EPA Resp. at 54. The regulated community and this
Court have heard this song before. The only “traditional” aspect of this Rule is EPA’s
repeated eagerness to again claim it is owed deference. However, EPA is not entitled
to such deference for an improperly developed rule that is not reasonable.
With respect to its contention that 40% co-firing with natural gas is
adequately demonstrated for coal plants, EPA notes only that many plants already
4
use “some amount” of natural gas. EPA Resp. at 29. EPA fails to respond, however,
to the point made by Applicants that very few plants use natural gas for anything
more than startup and that co-firing actual generation would involve costly
modifications with respect to gas pipeline capacity. Application For Immediate Stay
Of Final Agency Action During Pendency Of Petitions For Review, National Mining
Association and America's Power v. EPA (U.S. 24A97) at 16 (filed July 24, 2024).
Neither does the agency’s posturing address the simple truth that coal refuse
fueled plants operate under a fundamentally different set of federal rules that govern
fuel utilization. This point is demonstrated by EPA’s repeated statement that its 40%
natural gas co-firing requirement requires only minor changes to a coal plant’s boilers
and is otherwise cost-effective. EPA Resp. at 19, 29, 43. EPA ignores the fact that, in
order to be a qualified facility under the Public Utility Regulatory Policy Act, 75% of
heat input must be provided by the use of coal refuse. See Gibbons ¶2, App. at 0507a.
Accordingly, qualified coal refuse fueled units are prohibited by law from using more
than 25% natural co-firing. See 18 CFR 292.204(b)(1). There is nothing reasonable
about EPA’s rule relative to coal refuse.
In addition, the Rule entirely fails to address efforts by the coal refuse fuel
fired EGUs to educate the agency about its unique and confounding factors.
Appalachian Region Independent Power Producers Association (“ARIPPA”),
Schuylkill Energy Resources, and Ri-Corp each recommended EPA consider a
separate subcategory as provided for in Clean Air Act Section 111(b)(2) for coal refuse
facilities to recognize, among other things, their environmental benefits, net-neutral
5
GHG emissions, unique boiler characteristics and applicable legal requirements. See
ARIPPA Proposed Rule Comments, EPA-HQ-OAR-2-23-0708, App. at 0524a;
ARIPPA Supplemental Notice of Proposed Rule Comments, EPA-HQ-OAR-20238215, App. at 0624a; Schuylkill Energy Resources Proposed Rule Comments, EPAHQ-OAR-2023-00720-0560, App. at 637a; Ri-Corp Proposed Rule Comments, EPAHQ-OAR-2023-0072-0559, App. at 0645a. EPA failed to acknowledge these comments
in the Response to Comments, in the Final Rule, or in its response in opposition to
applications for stay. These facts alone demonstrate the unlawful, arbitrary and
capricious nature of this Rule and therefore supports an immediate stay pending
judicial review.
II. Grid Operators, Power Companies and Members of Congress All Find
That The Rule Presents A Significant Near-Term Threat To The
Reliability Of The Electric Power Grid
EPA’s position that the text of the Rule demonstrates a thoughtful rulemaking
relative to grid reliability
is unsupported when assessed relative to comments
received. See 89 Fed. Reg. 39,803; Goffman Decl., ¶113, App. at 0255-56a. The Rule is
instead a piecemeal experiment with unproven technologies countered by valid
concerns from grid experts including the balancing authorities, power companies, and
Congress. EPA continues to ignore these points to downplay this Rule as an unlawful
and illogical outgrowth of its authority under Section 111.
Electric
Reliability
Council
of
Texas,
Inc
(“ERCOT”),
Midcontinent
Independent System Operator, Inc. (“MISO”), PJM Interconnection, L.L.C. (“PJM”),
and Southwest Power Pool, Inc. (“SPP”), urged that EPA address grid reliability by
6
making four requests: allowing compliance deadline "tapering" in regions with
projected resource shortfalls over the following six years, providing a process to
confirm the shortfall determinations to its satisfaction, annually reviewing or
amending its guidance on enforcement discretion based on revised projected shortfall
conclusions, and implementing the "tapering" on a fact-specific basis depending on
how far away the region is in its forward analysis from meeting its target reserve
margins. ERCOT, MISO, PJM, and SPP Proposed Rule Comments at 11, Doc. ID.
EPA-HA-OAR-2023-0072-8207, App. at 664a. SPP specifically noted that its “study and
projected increase in demand did not consider the additional at-risk generation that
may retire and not be adequately replaced in a relatively short time frame resulting
from the compliance time frames contained in the Final Rule.” SPP Statement On
The Recent EPA Greenhouse Gas Emissions Rule (May 20, 2024) (access at
https://tinyurl.com/w9bja7ku), App. at 0653a. EPA failed to respond to any of these
issues raised.
Power companies also have expressed concerns about grid reliability,
specifically the ability to provide reliable electric service to the public as a direct
result of the Rule. See Brown Decl. ¶24, App. at 0677-78a; Beam Decl. ¶98, App. at 0729a;
Komaromy Decl. ¶43, App. at 0752a. These companies have operations across many
different states, and their united stance on this issue indicates this is a matter of
national significance. EPA maintains that the Rule will allow for the electric grid to
reliably operate, yet the vast majority of power companies have told the Agency this
is simply not true.
7
Even as this litigation develops, EPA has continued to marginalize these
concerns and has avoided saying much more than it met with the balancing agencies
and the Department of Energy on these issues. In EPA’s sixty page brief, only one
single page mentioned grid reliability. See EPA Response Br. at 44. Power companies
shared robust modeling, reports, and comments with EPA supporting exactly why
this Rule will negatively impact the grid and reliability of electricity. EPA counters
this by stating that they “devoted multiple pages of analysis” to the issue. Id. That
said, the nine pages of the Federal Register publication EPA alludes to do not contain
a meaningful contemplation or assessment of the raised concerns, but instead, are a
regurgitation of the same flowery language that is intended to assuage power
generation anxieties. See 89 Fed. Reg. at 40,011-20.
State and Municipal Respondents rely heavily on the modeling performed by
EPA for the proposition that the Rule “will have almost no effect on coal-fired
generation when the Rule is fully implemented in 2040, and that the Rule will not
undermine grid reliability.” Municipal Response at 26-32.
These comments fail to recognize, however, that the record demonstrates
numerous errors that are contained in EPA’s modeling that render any analysis
arbitrary and capricious. NRECA Proposed Rule Comments at 32, Doc. ID EPA-HQOAR-2023-0072-0770, App. at 0784a. As was demonstrated in these comments, EPA’s
modeling incorrectly projected the retirement of 66 coal units representing 40% of the
retired coal capacity in 2030. Marchetti IPM Final Report at 17-22, Doc. ID EPAHQ-OAR-2023-0072-0770, App. at 0809-14a. This modeling conclusion was reached in
8
the absence of any public statements or filings related to those units indicating that
they intend to retire. Id. This error can be attributed to EPA’s factually and legally
unreasonable assumptions regarding implementation of the Inflation Reduction Act.
Id.
EPA refuses to engage with the regulated community in a meaningful way, to
ensure that the suite of regulations promulgated do not impact generators’ ability to
meet demand and consistent with capacity for cost and available technology. The
Goffman Declaration purports to make these assurances but falls short of actually
identifying actual regulatory provisions that would alleviate the stressors on the
regulated community. EPA, by and through the Goffman Declaration, attempts to
resolve all stated concerns citing market changes, tax incentives, renewable
electricity production, etc. Goffman Decl., ¶33-38, App. at 0204-07a. The Rule advances
its coal and gas-fired EGU regulatory strategy and Goffman offers enforcement
discretion for after-the-fact relief if compliance or grid reliability becomes a problem.
See 89 Fed. Reg. at 39,803. Electricity demand is increasing at a rate that requires
immediate action to manage the harmful impact of this Rule. Gibbons Decl., ¶27, App.
at 0520-21a. Further, EPA’s dismissive response that retirements of power plants are
a business decision, and no plant is required by EPA to limit operations falls short of
a well-reasoned defense of the Rule. Goffman Decl. ¶103, App. at 0247-48a. This
Goffman claim is demonstrably false as the Rule requires units to choose between
two difficult compliance pathways: (1) set a retirement date prior to 2039 or (2) add
carbon capture controls that will reduce the net generation of the unit by 40% or
9
more. Both options undeniably operations. EPA ignores the commitments and, in
indeed, regulatory obligations of these companies to operate in a manner that is
lawful and sound as defined by economics and environmental stewardship.
If the intention of this Rule was to ensure Section 111 determinations were
well-reasoned, EPA might have taken the opportunity to engage with grid operators
and power companies to learn what is being done now and what additional costeffective and technologically feasible changes can be made moving forward. Those
discussions did not occur and because of that this Rule does not achieve the balance
required by Section 111’s best system adequately demonstrated criteria.
EPA is aware that this Rule, particularly when combined with the other
recently promulgated rules targeting fossil fuel fired EGUs, is not a traditional or
“regular” rulemaking. EPA insists that this Rule was promulgated in normal course
and any retirements that may occur will be “incidental” and should not “cast doubt
on its validity.” EPA Response Br. at 18 (citing West Virginia v. EPA, 577 U.S. 1126
(2016)). It is true that West Virginia suggests that “a rule that may end up causing
an incidental loss of coal’s market share” would not necessarily impact its validity.
597 U.S. at 731 n.4 (emphasis added). But the circumstances in the present case are
not that EPA is simply vaguely aware that fossil fuel EGUs may find themselves
contemplating retirement as a result of the Rule. In fact, these EGUs, and other
sector stakeholders, have directly communicated to EPA that many facilities will be
forced to retire early as a result of the Rule because CCS is not a viable option and
the suite of rules promulgated require too many other investments in the shortened
10
life of the generators. A serious likelihood of multiple fossil fuel EGU retirements
directly resulting from the Rule in addition to the other rules promulgated at the
same time is a consideration that would (and has) properly cast doubt on EPA’s
purpose for this Rule.
III. The Balance of the Equities and the Public Interest Favor an
Immediate Stay.
EPA has failed to explain how a stay would harm the public interest. EPA
Response Br. at 56-58. The Rule was designed by EPA without legal authority, period.
Its impact is to substitute the agency’s opinions over the Clean Air Act, Congress,
grid operators, power generators, and the public. It appears that only the comments
of a few were woven into the Rule. Only “benefits” the agency chooses to observe were
included. For example, the Rule and the EPA Response Brief each fail to acknowledge
that coal-refuse facilities actually reduce greenhouse gas emissions by burning coal
refuse as a fuel. See ARIPPA Proposed Rule Comments at 19, App. at 0542-43a. This
is particularly concerning since EPA asserts the challenged Rule has the stated
purpose of reducing greenhouse gas emissions.
EPA states that fossil-fuel power plants are the largest emitters of greenhouse
gas in the nation, yet it offers no support for concluding that the reductions of carbon
dioxide emissions required by the Rule would affect the consequences of climate
change in any manner. It is unlawful to regulate units that are necessary for the
continued reliability of the electric grid, without any evidence demonstrating that
the regulation will result in substantial decrease in emissions levels. EPA cannot
show how this Rule will improve emissions.
11
EPA has misused its authority under the Clean Air Act to advance its agenda
against the continued operation of fossil-fuel power plants. EPA does not demonstrate
how climate change is resolved or otherwise significantly addressed as a result of the
Rule. As such, the balance of the equities and the public interest favor an immediate
stay of this rule, pending review of the merits.
CONCLUSION
For the foregoing reasons, Applicant respectfully requests an immediate stay
of the Rule to prevent irreparable harm to its membership and the domestic
electricity grid.
/s/ Ancil G. Ramey
Ancil G. Ramey
Counsel of Record
David M. Flannery
Kathy G. Beckett
Keeleigh S. Huffman
STEPTOE & JOHNSON PLLC
707 Virginia Street, East
Post Office Box 1588
Charleston, WV 25326
(304) 353-8000
Ancil.Ramey@steptoe-johnson.com
Edward L. Kropp
STEPTOE & JOHNSON PLLC
PO Box 36425
Indianapolis, Indiana 46236
Counsel for Midwest Ozone Group
12
APPENDIX
INDEX OF APPENDICES
Exhibit A
SaskPower Proposed Rule Comments
0001a
Exhibit B
Energy & Environmental Research Center Final
Report Final Report
0003a
Exhibit C
Cichanowicz & Hein Report
Exhibit D
Power Generators Air Coalition Proposed Rule
Comments
Exhibit E
Goffman Declaration
Exhibit F
McLennan Declaration
Exhibit G
Grooms Declaration
Exhibit H
Hasten Declaration
Exhibit I
Soderberg Declaration
Exhibit J
Tudor Declaration
Exhibit K
McCollam Declaration
Exhibit L
Gibbons Declaration
Exhibit M
ARIPPA Proposed Rule Comments
Exhibit N
ARIPPA Supplemental Notice of Proposed Comments
Exhibit O
Schuylkill Proposed Rule Comments
Exhibit P
Ri-Corp Proposed Rule Comments
Exhibit Q
ERCOT Proposed Rule Comments
Exhibit R
SPP Press Release
Exhibit S
Brown Declaration
0022a
0076a
0189a
0296a
0344a
0387a
0413a
0435a
0461a
0506a
0524a
0624a
0637a
0645a
0652a
0653a
0673a
Exhibit T
Beam Declaration
Exhibit U
Komaromy Declaration
Exhibit V
NRECA Proposed Rule Comments
Exhibit W
Marchetti IPM Final Report
Exhibit X
Final Rule
0734a
0737a
0753a
0791a
0815a
Perez, JuanB
From:
Sent:
To:
Cc:
Subject:
Jeff Jickling <jjickling@saskpower.com>
Friday, August 4, 2023 11:49 AM
A-AND-R-DOCKET
Darren Foster; Joel Cherry; Cole Goertz
Docket ID No. EPA–HQ–OAR–2023–0072: SaskPower Correction of Reference to
Boundary Dam Unit 3 Emissions Performance in Proposed Rule
Follow Up Flag:
Flag Status:
Follow up
Flagged
In the Proposed Rule for the New Source Performance Standards for Greenhouse Gas Emissions From New, Modified,
and Reconstructed Fossil Fuel-Fired Electric Generating Units; Emission Guidelines for Greenhouse Gas Emissions From
Existing Fossil Fuel-Fired Electric Generating Units; and Repeal of the Affordable Clean Energy Rule, there is a reference
to SaskPower’s Boundary Dam Unit 3 CCS Facility ‘successfully demonstrating the commercial-scale feasibility of 90
percent capture rates’. As the owner and operator of this facility, we are providing the following correction to the
emissions performance of the Boundary Dam Unit 3 CCS Facility.
SaskPower’s CCS facility was the first of its kind, and we have acknowledged the technical issues encountered at
the facility, such as amine degradation resulting from fly ash ingress. We have consistently made modifications
during the past eight years to stabilize operations, improve reliability and maximize capacity.
SaskPower’s CCS facility is not capturing 90 per cent of emissions from Boundary Dam Unit 3, though that is its
nameplate capacity. Our CCS facility has only operated at full nameplate capacity for a few days shortly after it
was commissioned.
To maintain long-term reliable operation, only a portion of the total flue gas from BD3 can be processed by the
CCS facility. The portion that cannot be processed through the CCS facility is released to the atmosphere.
Recent performance has shown that the CCS facility can capture at least 90% of the CO 2 from the partial flue gas
stream it processes.
To ensure a higher level of overall equipment reliability and process efficiency, SaskPower has optimized the CO 2
capture rate at a target of 65 to 70 per cent of total Boundary Dam Unit 3 emissions on an ongoing basis.
If you have any questions, or would like further information, please contact Jeff Jickling at (306)566-2374 or
jjickling@saskpower.com.
Jeff Jickling, P.Eng., PMP
SaskPower | Sr. Business Advisor, Corporate Strategy & Planning | Strategy, Technology & Finance
p. 306-566-2374 | c. 306-551-9557 | saskpower.com | Twitter | Facebook
10W – 2025 Victoria Avenue
Regina, SK S4P 0S1
1
0001a
This email including attachments is confidential and proprietary. If you are not the intended recipient, any redistribution
or copying of this message is prohibited. If you have received this email in error, please notify us by return email, and
delete this email.
2
0002a
EXAMINATION OF EPA’S PROPOSED EMISSION
GUIDELINES UNDER 40 CFR PART 60
Final Report
Prepared for:
Daniel Walsh
National Rural Electric Cooperative Association
4301 Wilson Boulevard
Arlington, VA 22203
Prepared by:
John P. Kay
Wesley D. Peck
John A. Brunner
Tyler K. Newman
Michael P. Warmack
Amanda J. Livers-Douglas
2023-EERC-08-04
August 2023
0003a
TABLE OF CONTENTS
LIST OF FIGURES ......................................................................................................................... i
EXECUTIVE SUMMARY............................................................................................................. ii
INTRODUCTION .......................................................................................................................... 1
SUBCATEGORIZATION OF ELECTRIC GENERATION UNITS ............................................. 1
Thermal Degradation............................................................................................................. 4
Oxidative Degradation .......................................................................................................... 4
Degradation by Reaction with Impurities ............................................................................. 4
CARBON CAPTURE EXAMPLES............................................................................................... 5
AES Corporation’s Warrior Run Generating Station ............................................................ 5
AES Corporation’s Shady Point Generating Station ............................................................. 5
Searles Valley Minerals Soda Ash Plant................................................................................ 5
Quest CO2 Capture Facility ................................................................................................... 5
Saskpower’s Boundary Dam Unit 3 ...................................................................................... 6
NRG Energy’s Petra Nova Facility ....................................................................................... 6
AVAILABILITY FACTOR ............................................................................................................. 6
NATURAL GAS CARBON CAPTURE ........................................................................................ 7
INTEGRATED GASIFICATION COMBINED CYCLE............................................................... 8
TRANSPORTATION AND GEOLOGIC STORAGE ................................................................... 8
TIMING ........................................................................................................................................ 10
Carbon Capture Timing Issues ............................................................................................ 11
Transport and Storage Timing Issues .................................................................................. 11
COSTS .......................................................................................................................................... 12
SUMMARY .................................................................................................................................. 13
REFERENCES ............................................................................................................................. 14
LIST OF FIGURES
1
Availability of Boundary Dam Unit 3 capture facility and CO2 compressor relative to
the Unit 3 power plant ........................................................................................................... 7
2
Timeline to storage as presented by EPA in the TSD .......................................................... 10
3
More realistic development timeline for the capture portion of a CCS project .................. 12
i
0004a
EXAMINATION OF EPA’S PROPOSED EMISSION GUIDELINES UNDER 40 CFR
PART 60
EXECUTIVE SUMMARY
The U.S. Environmental Protection Agency (EPA) is proposing new emission guidelines for
CO2 at existing fossil fuel-fired electric generating units. The guidelines propose that the best
system for emission reduction for coal-fired electric generating units is carbon capture and storage
(CCS). Carbon capture rates must meet a minimum of 90%. EPA believes that CCS is a mature
technology that can be implemented to meet a 2030 deadline.
Examples are given in the guidelines to show the maturity of CCS; however, these examples
spotlight facilities that are small in size, and all but two examples, Saskpower’s Boundary Dam
Unit 3 and the Petra Nova project, perform no subsurface injection at all. The examples are of
slipstream systems and production facilities. No example is given of a facility larger than Petra
Nova’s 240-MW facility capturing CO2 and injecting it into the subsurface because one does not
exist.
With respect to the transport and storage of CO2, sufficient demonstration of CCS with all
the appropriate regulatory frameworks in place has not occurred. Documentation is not present to
support EPA’s geographic analysis, and the information the Agency does possess is out of date.
The timeline for implementation of CCS is expected to take much longer than anticipated
by EPA. One example is EPA review of UIC (underground injection control) Class VI permits. In
the last year, the number of permits under review has risen from 9 to 98, and historically, it appears
the process can take more than 6 years per permit. Overall, from evaluation to commercial
operation, an optimistic timeline indicates it can take at least 7 years to complete. Any disruptions
to permitting, design, or construction can extend this for many additional years, which will be
incompatible with meeting compliance in 2030.
Costs related to CCS can vary widely depending on conditions at the location and the
permitting required. Based on experience, the costs to design and construct the carbon capture
facility can exceed $1 billion. The pipeline for transportation of the CO2 to the injection site can
cost $600,000–$2,500,000 per mile or more, with development of the injection site costing
$30 million or more, depending upon the number of injection and monitoring wells required. As a
result of these substantial costs, final investment decisions on the construction and commissioning
of carbon capture and transportation systems are often contingent upon an associated geologic
storage facility permit being available and approved. This further extends the time to implement
new carbon capture and storage well beyond the proposed compliance date.
Although CCS technology is progressing, it is too early to label it as commercially mature
technology, and more projects need to be completed to substantiate the performance levels
suggested by EPA. Based on the supported conclusions and the current status of carbon capture
technology, EGUs cannot meet the CO2 capture rates or the timeline that EPA proposes.
ii
0005a
EXAMINATION OF EPA’S PROPOSED EMISSION GUIDELINES UNDER 40 CFR
PART 60
INTRODUCTION
The proposed change to the U.S. Environmental Protection Agency’s (EPA’s) rules within
40 CFR Part 60 details proposed carbon emission standards for both fossil fuel-fired steam
generating units and fossil fuel-fired stationary combustion turbines. In this document, EPA
outlines climate change and its impacts; recent developments in emissions; proposed requirements
for both new and reconstructed stationary combustion turbine electric generation units (EGUs);
requirements for new, modified, and reconstructed fossil fuel-fired steam generating units; the
proposed regulatory approach for existing fossil fuel-fired steam generating units; the proposed
regulatory approach for emission guidelines for existing fossil fuel-fired stationary combustion
turbines; and impacts of the proposed actions. Within the document, EPA discusses the best system
of emission reduction (BSER) for various subcategories of fossil fuel-fired steam generating units
and subcategories of fossil fuel-fired stationary combustion turbines. The primary focus of this
review is to examine the application of carbon capture and storage (CCS) technologies to these
EGUs through their present state of readiness and adequacy of demonstration. CCS includes the
carbon capture process itself, transportation of the CO2, and storage or sequestration.
SUBCATEGORIZATION OF ELECTRIC GENERATION UNITS
The EPA categorizes EGU’s into two primary groups: fossil fuel-fired steam generating units
and fossil fuel-fired stationary combustion turbines. Various subcategories exist under the umbrella
of these two categories, which are further discussed below.
Of the eleven subcategories for fossil fuel-fired steam generating units, EPA is proposing the
application of CCS to one: long-term existing coal-fired steam generating units. These units are
coal-fired steam generating units that have not elected to commit to permanently cease operations
by January 1 of 2040. This CCS system is required to have a CO2 capture rate of 90%, with the
associated degree of emission limitation a CO2 reduction of 88.4% lb CO2/MWh-gross (proposed
rule pages 33359 and 33360).
EPA is proposing to regulate existing fossil fuel-fired stationary combustion turbines in two
segments, with only the first outlined in this proposed EPA regulation, the second to be released in
a separate regulation document later. In this first segment, EPA proposes regulation for baseload
turbines over 300 MW. EPA defines baseload as having a capacity factor greater than 50%
(proposed rule page 33362).
EPA believes that two technologies are possible BSERs for fossil fuel-fired stationary
combustion turbines over 300 MW operating at a capacity factor of greater than 50% coupled with
heat-rate improvements: i) cofiring with low greenhouse gas (GHG) hydrogen and ii) CCS. EPA
believes that the 300-MW threshold for applicability is appropriate because it focuses on the units
with the highest emissions where CCS is likely to be the most cost-effective.
1
0006a
ADEQUACY AND APPLICABILITY OF CARBON CAPTURE
Several technologies are included under the umbrella of carbon capture: postcombustion,
precombustion, oxyfuel combustion, and direct air capture. Direct air capture does not capture CO2
directly from a GHG point source prior to its emission but after. Therefore, direct air capture is not
further discussed, as it is not applicable to being integrated into fossil fuel-fired steam generating
EGUs or fossil fuel-fired turbines.
Oxyfuel combustion involves combining a fuel, such as coal or natural gas, with pure
oxygen. Since the oxidant stream is pure oxygen instead of air, such as in a conventional
combustor, the combustion reaction does not create other combustion by-products such as NOx.
CO2 and H2O steam are produced from the reaction, which is then used to power a turbine. Since
this combustion reaction creates a stream of pure CO2, the CO2 can be captured without the need
for additional systems that are required in pre- or postcombustion CO2 capture. However, these
systems do require a constant and sizable supply of pure oxygen, often necessitating an air
separator to be included in the process (1). Additional challenges of oxyfuel combustion systems
are the high capital costs, energy consumption, and operational challenges of oxygen separation
(2). Research of oxyfuel combustion is still ongoing, with projects focusing on lab-, bench-, and
pilot-scale testing to understand the combustion mechanics of oxyfuel combustion at high
temperatures and pressures, verify system design and operation concepts, and improve the
performance of ancillary system components (3). Some demonstrations of oxyfuel combustion
systems have been conducted, the largest being a retrofitted 100-MWth PC boiler in Central
Queensland, Australia, which operated from December 2012 to March 2015. In that time, the unit
achieved 10,000 hours of oxyfuel combustion and 5500 hours of carbon capture (4).
Precombustion CO2 capture constitutes the removal of CO2 from a fuel source prior to its
combustion. This is commonly achieved through fuel gasification, in which the feedstock, such as
coal, is partially oxidized with steam and oxygen-rich air under high temperature and pressure to
form syngas, which is a mixture of hydrogen, carbon monoxide, CO2, and smaller elements of
other gases, such as methane. The syngas can then undergo the water-gas shift reaction, which
converts the carbon monoxide and water in the gas to hydrogen and CO2. The CO2 can then be
captured, and the H2-rich fuel combusted. Since the precombustion fuel stream is rich in CO2 and
at a higher pressure, extraction of the CO2 from the stream is easier than in postcombustion
systems. However, the cost of a gasification system is often greater than a traditional coal-fired
power plant (5). Therefore, precombustion CO2 capture is not considered a leading technology for
CO2 emissions reduction in the electrical generation industry. But it has been shown to be effective
in the chemical processing industry, with Great Plains Synfuels Plant in Beulah, North Dakota,
having been in operation for the past 25 years and remaining the only coal-to-synthetic natural gas
facility in the United States. Great Plains Synfuels Plant produces synthetic natural gas from lignite
coal and captures its CO2 for utilization in enhanced oil recovery (EOR) in Canada. The plant is
capable of capturing up to 3 million tons of CO2 per year. Since 2000, CO2 emissions at the
Synfuels Plant have been reduced by 45% (6).
Postcombustion CO2 capture involves the removal of CO2 from the flue gas of an EGU.
After the fuel has been combusted, the exhaust gases are processed to filter out potential
contaminants such as ash and SO2, then the exhaust gases go to the postcombustion CO2 capture
2
0007a
system, which captures the CO2 from the gas stream through a reaction with a chemical solvent
(amine). This solvent captures the CO2 gas, and the solvent and gas are later separated in the
stripper column, where heat and pressure are used to regenerate the solvent and create a stream of
pure CO2, that can then be compressed, transported, and sequestered. The use of chemical solvents
for carbon scrubbing is the most commonly acknowledged process for capturing CO2 from gas
mixtures (7) and has been used in the natural gas industry to separate CO2 from other gases since
the 1930s (8). Current federally funded work in solvent-based postcombustion capture is seeking
to address key challenges to deployment, which include scale-up, parasitic load, process
integration, water use, and capital costs (8). Additionally, solvent degradation can be a significant
issue.
The large-scale carbon capture facilities that are in operation throughout the world are mostly
focused on natural gas processing (9). Only two facilities are operating at coal-fired power plants:
Saskpower’s Boundary Dam Unit 3 (110 MW) and the Petra Nova Project (240 MW equivalent),
which will be discussed later. Postcombustion CO2 capture has yet to be demonstrated at a baseload
facility larger than Boundary Dam Unit 3. Parasitic load requirements for the operation of the
carbon capture system decrease the net power output of the EGU by roughly 20% (10).
With respect to carbon capture technology, the proposed rule states that:
“The EPA is proposing that the CO2 capture component of CCS has been adequately
demonstrated and is technically feasible based on the demonstration of the technology
at existing coal-fired steam generating units…” [page 33291]
The design and integration of CO2 capture facilities can vary based on the configuration of
the EGU and fuel source. Variations, such as the CO2 purity of the emission stream, facility design,
local energy costs, emission volumes, flue gas temperature and pressure, the presence of
contaminants, transition from cold or warm (standby) condition to operation condition, and
ramping due to load changes, all affect the applicability and cost of implementing CCS at fossil
fuel-fired EGUs (11). For example, in Wyoming, most of the existing power generation fleet is not
equipped with environmental control systems that remove enough NOx, SOx, and other air
pollutants to prevent the accelerated degradation of the amine solvent inside of the CO2 capture
system. 87% of EGUs have flue gas desulfurization systems, and 56% of EGUs have
postcombustion NOx control systems, whereas nearly all Wyoming EGUs only have particulate
and mercury control devices installed. Before a CCS system could be constructed and retrofitted,
these facilities would need to be upgraded to meet these requirements (12). These upgrade
requirements are not considered by EPA in the proposed capture requirements.
Typical solvents utilized in carbon capture systems are amine-based. The name amine refers
to a chemical function group that includes compounds with a nitrogen atom and a lone pair. A
common amine in CCS systems is monoethanolamine (MEA), colloquially referred to in industry
as amine. Amines are susceptible to degradation, and solvent management can be a significant
challenge. Amine degradation can reduce solvent efficiency or cause an unintentional release into
the atmosphere. This degradation can happen because of several factors: thermal or oxidative
degradation or reaction with impurities in the flue gas stream. Advanced amines are being
3
0008a
developed, which reduce thermal or oxidative degradation of the amine and improve its capture
efficiency; however, impurities still have a significant impact on the life of amine.
Thermal Degradation
The heat involved in the regeneration process, where CO2 is stripped from the amines, can
cause these molecules to break down, leading to loss of capture efficiency and the need for frequent
solvent replacement. This aspect is poorly addressed in literature due to the sensitive nature of
sharing specific information from vendors. This quickly increases operating costs and results in
large quantities of liquid waste. The EPA does not recognize or address this issue.
Oxidative Degradation
The solubility of oxygen in amine solutions is a key issue in dealing with problems like
degradation and corrosion (13). An increased level of oxygen in the amine solvent changes the
solvent chemistry, increasing its tendency to cause oxidation and corrosion. The Technology
Centre Mongstad (TCM) studied amine degradation in a combined heat and power plant (CHP)
and noted significant corrosion. Significant material thickness reduction and leakage on the CHP
reboiler heat exchanger plates were observed. The CHP stripper packing surface was coated by a
layer of corrosion products. This layer was “leaching” out in the solvent upon restart of the CHP
stripper, resulting in rapidly increasing iron content in the fresh solvent (14). The application of an
oxygen scavenger, a chemical additive to the amine solvent, could be used as a preventive measure
to keep solvent degradation low. This form of degradation of the amine solvent is correlated to the
composition of the EGU’s flue gas and not the capture rate of the CCS system; therefore, the effects
of the flue gas on the solvent chemistry must be individually investigated at each facility.
Degradation by Reaction with Impurities
When TCM tested amines on a residue fluidized catalytic cracker (RFCC), they had not been
able to operate the amine plant with RFCC flue gas because of very high amine emissions
(>20 ppm) caused by sulfuric acid aerosol and dust particles present in the flue gas (15). With
installation of a Brownian diffusion (BD) filter upstream from the absorber, more than 95% of the
aerosols were removed, and together with optimization of plant process parameters and
configuration, the amine emissions were reduced. It is known that both SO2 and NOx will give
unwanted reactions with MEA (16).
Although the degradation mechanisms for MEA have been extensively studied in the
literature (16–19), testing, understanding, and mitigating amine degradation on a plant-by-plant
basis are crucial for the sustainable and efficient application of CCS technology. Research is
ongoing to develop advanced amines and to improve the process design to minimize amine losses,
such as optimizing operating conditions, implementing solvent purification processes, and better
managing impurity variability in the flue gas.
The above discussion illustrates that additional investigation is required at the specific
facility being considered for installation of a carbon capture system that may include significant
construction and redesign to accommodate CCS implementation.
4
0009a
CARBON CAPTURE EXAMPLES
EPA cites several examples of successful plant operation within its proposed guidelines, and
a few of them will be briefly discussed. These examples do show that CCS is possible and
promising and present a promising solution for the future; however, they do not reflect the needs
as set forth by EPA as they are examples of slipstream systems, are smaller capacity units, do not
employ the full CCS process, and are capturing CO2 at levels below 90%.
AES Corporation’s Warrior Run Generating Station
The Warrior Run Station is a 180-MW bituminous coal-fired power plant located in
Maryland. The installed CO2 capture system captures a small slipstream of the facility’s flue gas
to produce 330 t CO2/day of food-grade CO2 for use in food processing. The process used is an
ABB Lummus unit with MEA as its solvent (20). The installed CO2 capture system captures
anywhere between 4% to 6% of the CO2 emissions of the plant (21). The important highlights are
that the system is a slipstream of a small power plant which produces a product and does not inject
CO2 into the subsurface. Therefore, the small capacity, slipstream system employed here
demonstrates a small portion of the required CO2 capture rate and has little correlation to the levels
EPA would mandate under their proposed guidelines.
AES Corporation’s Shady Point Generating Station
Shady Point Power Plant is a 320-MW circulating fluidized-bed subbituminous coal-fired
power plant located in Oklahoma. A slipstream of the power plant’s flue gas is captured to produce
200 t CO2/day of food-grade CO2 for use in food processing. With the plant emitting 1.24 million
t CO2/year, the yearly capture rate approximates to 6%. This process uses an ABB Lummus scrubber
system with MEA as its solvent (20). Like the Warrior Run Generating Station, this is a small
slipstream which produces a product and does not inject CO2 into the subsurface. This example
has little correlation to the levels EPA would mandate under their proposed guidelines.
Searles Valley Minerals Soda Ash Plant
The Searles Valley Minerals Soda Ash Plant, located in California, captures approximately
800 t CO2/day from the flue gas of the 62.5-MW Argus Cogeneration Plant, a subbituminous coalfired power plant that generates electricity and steam. The CO2 is captured with an ABB Lummus
MEA capture unit, and the captured CO2 is used for the carbonation of brine in the production of
soda ash (20). With the plant emitting 1.63 million t CO2/year, the capture rate approximates to
18%. Like the previously discussed facilities, the small capacity system employed here
demonstrates a small portion of the required CO2 capture rate and has little correlation to the levels
EPA would mandate under their proposed guidelines. The correlation between this facility and
what is expected under the proposed guidelines is minimal.
Quest CO2 Capture Facility
The Quest Carbon Capture and Storage Project is a CCS facility in Alberta, Canada, that
began operation in 2015. Quest removes CO2 from the process gas streams of three hydrogen
5
0010a
manufacturing units (HMUs), equating to 1 million t CO2/year, within the Scotford upgrader
facility, which emits 3 million t CO2/year. Between the years of 2015 and 2021, Quest has been
able to capture between 77.4% and 83% of CO2 emissions from the HMUs, with the average CO2
capture rate of 79.4% (22). Although the facility has demonstrated the ability to store CO2, the
overall capture rate of the facility falls short of EPA’s proposed 90% minimum capture rate.
Saskpower’s Boundary Dam Unit 3
Saskpower’s Boundary Dam Unit 3 (BD3) is a 110-MW lignite-fired unit in Saskatchewan,
Canada. Development of the CCS facility began in 2007, with the decision to move forward with
construction in 2010, and provides CO2 for both EOR and sequestration. Saskpower selected the
CANSOLV process, an amine solvent system, for its CO2 capture process. During its first year of
operation, BD3 achieved a CO2 capture rate of 50% of its designed volume. This capture rate has
been improved through design and operations optimizations, although the capture rate is still below
its designed CO2 production levels. Significant issues included operational difficulties from
construction and design deficiencies, issues with fly ash carryover, a lack of redundancy and
isolation capabilities, and amine degradation and foaming (23). Lessons learned from this CCS
facility are slow to be released, and there may be other operational challenges that industry and
EPA do not know about. This facility exemplifies the site-specific challenges that are to be
expected with CCS implementation and is only one-third the scale of plants that would be
addressed in EPA’s proposed emission guidelines.
NRG Energy’s Petra Nova Facility
The Petra Nova facility, located in Texas, is a 240-MW equivalent slipstream of flue gas
from the W.A. Parish coal-fired facility. This postcombustion capture facility started operation in
2017 and fulfilled its objectives of demonstrating carbon capture at this scale coupled with
compression and transportation of CO2 to an oil field for EOR only. The facility was shut down
because of low oil prices in May 2020 due to no alternate method of sequestration. Market-driven
EOR alone does not adequately demonstrate CCS that will meet EPA’s proposed continuous
emission reduction. During its 3-year operation, it suffered frequent outages and missed its carbon
capture targets by ∼17% (24).
AVAILABILITY FACTOR
The availability factor is a measure of the amount of time a system is in operation and not
undergoing maintenance, repair, and unexpected down time and is given as a percentage. The most
relevant example given by EPA for an EGU utilizing a fully integrated carbon capture system is
Boundary Dam Unit 3, as mentioned above. Since its start-up in 2014, the unit has experienced
operational issues that have led to more frequent capture facility outages than originally
anticipated. The primary issues experienced have been with fly ash and fly ash component buildup
in the CCS facility. Heat-transfer surfaces, such as the reboilers, fouled over time. The packing in
the absorbers and the strippers also experienced fly ash buildup and the development of organic
deposits. These issues affected the capture capacity of the facility as the heat-transfer efficiency
decreased and the gas flow rate became limited because of deposits. The implementation of
6
0011a
advanced demister wash systems has extended the facility’s operational time between maintenance
outages, and wash systems for the booster fan and redundant heat exchangers, with isolations, have
enabled the facility to conduct online maintenance. The capture facility has also experienced
compressor failures that increased its unplanned outage time for the years of 2021 and 2022.
Figure 1 shows the yearly percent availability of the Boundary Dam Unit 3 capture facility’s
availability, planned maintenance outages, and unplanned outages for the years of 2014 through
2022 relative to the operation of the power plant (25).
Figure 1. Availability of Boundary Dam Unit 3 capture facility and CO2 compressor relative
to the Unit 3 power plant (25).
The figure shows that annually the capture facility was not able to operate the full time the
power plant was in operation and for only 2 years was the facility operating above 90% of the
available time. Based on this information, even with the capability to capture greater than 90% of
the CO2 emissions, it is premature to expect that a capture facility will be able to operate with an
availability factor sufficient to comply with the annual emission requirements of the proposed rule.
NATURAL GAS CARBON CAPTURE
Many of the demonstrations studied by EPA are coal-fired power plants with small slip
stream CO2 capture systems. EPA proposes CCS as the BSER for stationary combustion turbines
for greater than 300 MW and over 50% capacity factor; however, CCS has been studied less at
natural gas EGUs than coal-fired EGUs. Among coal-fired EGUs, each facility has different CCS
retrofitting and integration needs, due to the operational parameters, facility differences, and the
7
0012a
composition of flue gas. One of the primary challenges with natural gas is its lower carbon load.
Natural gas-fired generation produces less CO2 per MWh than coal-fired facilities, meaning that
the capture plant design has to be adjusted for a lower concentration of CO2 in the flue gas (26).
Additionally, gas turbine EGUs are more likely to be throttled for load-following applications than
coal-fired EGUs, adding significant demands and stresses on the attached CCS system to ramp
with the power plant (27).
INTEGRATED GASIFICATION COMBINED CYCLE (IGCC)
Currently, no commercial-scale IGCC plants are in operation or under development. The
Kemper County, Mississippi, IGCC project struggled, with major problems stemming from overly
complex technology, complex supply chain, and equipment reliability issues (28). After significant
cost overruns, the original plan for a gasification plant was abandoned, and the plant was converted
to natural gas operation (29).
TRANSPORTATION AND GEOLOGIC STORAGE
Several fundamental assertions are made by EPA in its proposed guidelines in relation to
CO2 transportation and geologic storage. In the following sections, those assertions will be directly
addressed.
With respect to the geologic storage of CO2, the proposed rule states that:
“The EPA proposes that CCS at a capture rate of 90 percent is the BSER for longterm coal-fired steam generating units because CCS is adequately demonstrated, as
indicated by the facts that it has been operated at scale and is widely applicable to
sources and that there are vast sequestration opportunities across the continental
U.S.” [page 33346]
The issue is with the assertion that “CCS is adequately demonstrated” and “has been
operated at scale.” EPA describes in the May 23, 2023, technical support document (TSD) titled
GHG Mitigation Measures for Steam Generating Units TSD on page 22 that there are only two
large-scale CCS facilities in North America on existing coal steam EGUs. One of which was Petra
Nova which only operated from 2017 to May 2020 and involved CO2 EOR. The other is Boundary
Dam in Canada, which is not subject to EPA’s underground injection control Class VI rules for the
storage of CO2. To date, no commercially operated CCS project capturing CO2 from a coal steam
EGU in the United States has operated under EPA Class VI regulations. The only CCS projects
that are in operation in the United States under EPA Class VI regulations are Archer Daniels
Midland processing plant (capturing approximately1 million tonnes/year) in Decatur, Illinois, and
the Red Trail Energy ethanol facility (capturing approximately 180,000 tonnes/year) near
Richardton, North Dakota. For comparison, a 300-MW coal-fired facility would capture
approximately 2.5 million tonnes/year. These two projects are not enough to demonstrate that the
appropriate regulatory frameworks are in place for the operational phase of projects that will
require flexibility and likely regular updates to permitted operational parameters.
8
0013a
Another issue is with the assertion of “vast sequestration opportunities.” This assertation is
seemingly founded on a geographic analysis performed by EPA:
“The EPA performed a geographic availability analysis in which the Agency examined
areas of the country with sequestration potential in deep saline formations,
unmineable coal seams, and oil and gas reservoirs; information on existing and
probable, planned or under study CO2 pipelines; and areas within a 100-kilometer
(km) (62-mile) area of locations with sequestration potential.” [page 33298]
However, no documentation of this geographic analysis is provided. The May 23, 2023, TSD
titled GHG Mitigation Measures for Steam Generating Units TSD also referenced this geographic
analysis. Figure 1 of the TSD showing geologic storage potential from the NATCARB website
includes an antiquated map layer for unminable coal seams. The U.S. Geological Survey (USGS)
has a more accurate map of unminable coal seams that could be used for CCS (30). The USGS
map accounts for EPA Class VI regulations, 40 CFR 144.3, and 40 CFR 144.6 (31), which prohibit
CO2 storage in formations with salinity lower than 10,000 mg/L (30, 32). Figure 1 of the TSD also
shows an erroneous map layer for deep saline formations. Using the correct deep saline formation
map layer (showing the proper extent of assessed formations based on minimum depth
requirements), the USGS coal layer, and the pertinent stationary CO2 sources will show that the
spatial relationship of CO2 capture to geologic storage is not as opportune as suggested by EPA.
The result is that more and longer pipelines will be needed to transport captured CO2 to feasible
storage locations. This implication cascades into additional time (and money) needed to construct
a fully integrated CO2 capture, transport, and storage project. Other aspects of EPA’s geographic
analysis that contribute to the overstatement of “vast and nearby” geologic storage opportunities
are:
• Proximity does not factor into the feasibility/suitability of geologic storage.
• The EPA geographic availability analysis is based on a generation unit being within
100 km of a state with geologic storage potential, rather than from the storage location
itself, which erroneously oversells the spatial relationship between CO2 source and
geologic sink.
• The analysis does not integrate evolving local (state, county, parish) CO2 transportation
and storage laws, some of which are looking to ban the geologic storage of CO2.
In the TSD, EPA states:
“DOE’s assessment focuses on the potential physical constraints for sequestering
CO2; it does not include economic or other constraints.”
And
“While the NETL and USGS characterize potential storage, site-specific technical,
regulatory, and economic considerations will ultimately factor into the attractiveness
of a given storage resource for a particular project.”
9
0014a
These are nontrivial comments that have a strong impact on project timelines and budgets
when a geologic storage option is pursued for captured CO2. A major consideration from a
regulatory perspective is access to federally owned pore space, a topic that has yet to be fully
addressed by any federal agency.
TIMING
In the May 23, 2023, TSD titled GHG Mitigation Measures for Steam Generating Units
TSD, EPA denotes that deployment of CCS is economically reasonable and can be done within
5 years:
“Deployment of CCS technology at EGUs involves a project schedule that can be
completed in roughly five years. For affected sources who choose to implement CCS,
the project will involve several phases, many of which can occur concurrently and
simultaneously.” [page 35 TSD]
“There are many site-specific considerations to individual sources that influence the
project timeline and schedule. Nonetheless, EPA believes that a five-year project
timeline for deploying CCS, and related infrastructure and equipment, is reasonable.”
[page 36TSD].
This 5-year period, as depicted in the example timeline shown in Figure 2, which was also
presented in the TSD, is not realistic and is completely unachievable. The timeline has several
issues related to the sequencing of events, as discussed below.
Figure 2. Timeline to storage as presented by EPA in the TSD.
10
0015a
Carbon Capture Timing Issues
This timeline assumes that procurement of capture equipment will be started prior to the
storage site being permitted. Most facilities looking to develop CCS will be reliant on third-party
financing or loans. These financing avenues have historically required permits for the storage site
to be in place to increase regulatory certainty and reduce investment risks. Based on that
experience, even on EPA’s inaccurately short and overlapping time frames for each step, the carbon
capture timelines would be delayed by at least an additional 1.5 years.
The timing of the site work and construction at the carbon capture site are shown to
commence approximately 6 months after the start of engineering and procurement. Without
detailed information concerning the constructability of the site such as soil analysis, location
parameters (temperature changes, wind loading, etc.), the footprint of the capture facility, etc., that
requires measurement and testing to provide detailed information for the construction of the
capture facility, the present timeline is likely compressed and would be expected to require
additional time (3–6 months) to complete the requisite evaluations prior to initiation of work.
Transport and Storage Timing Issues
The timeline shown in Figure 2 depicts 2.5 years for storage feasibility, site characterization,
and permitting. This is an extremely optimistic and aggressive timeline. For example, the U.S.
Department of Energy’s CarbonSAFE Program assumes a 5-year timeline to address feasibility,
characterization, and permitting. Even for states with Class VI primacy such as North Dakota,
storage feasibility, site characterization, and permitting could take up to 4.5 years (33). One of the
only ways to accelerate this timeline would be if there were existing site-specific data that were
sufficient to address UIC Class VI requirements. For states without primacy, storage feasibility,
site characterization, and permitting could take up to 6.5 years based on historical EPA permitting
timelines from the two approved EPA UIC Class VI permits (34). In June of 2022, nine projects
were waiting for Class VI permit approvals (35). EPA now has 98 UIC Class VI permits (in 35
projects) to review (36).
Another issue with the proposed 5-year timeline is that it does not adequately factor in the time
needed to lease pore space. Much of the prime geologic storage opportunity lies beneath federally
owned lands. As such, any storage operation that will emplace captured CO2 in pore space
managed by the federal government will need to work through federal permitting and NEPA
(National Environmental Policy Act) review. This process alone can add years to a project’s
development timeline. In addition to the federal land issue, many states have yet to address pore
space ownership. Challenges to amalgamation authority on nonfederal land, achieving 100%
consent of private pore space owners where amalgamation rules do not exist, and states lacking
established pore space rules result in significant uncertainty regarding how much of the nation’s
geologic CO2 storage resources can be developed and permitted, particularly within the time frame
of the proposed rules.
Pipeline feasibility, design, and permitting stage is listed at 2.5 years. Depending on the route
of the pipeline, permits for water body crossings, federal lands, and the Army Corps of Engineers
can take a year or more to acquire, if the permit is allowed at all. In addition, agreements with
11
0016a
landowners for rights of way (ROWs) for the pipeline can take a year or longer, depending on the
length of the pipeline. In all, the listed time of 2.5 years for pipeline feasibility, design, and
permitting appears to be overly optimistic. In addition, with the current supply chain issues, the
ability to secure the required piping and equipment may take up to 1 year to acquire. With multiple
major projects planned to be undertaken, supply chain issues would be expected to worsen. Finally,
as the number of projects grow, the demand for labor will increase, adding to the expected cost.
A subsidiary supporting attachment provided by EPA to augment the GHG Mitigation
Measures for Steam Generating Units TSD entitled “CCS Schedule Sargent and Lundy” contains
the development timeline in Figure 3. As stated in the supporting document, “This schedule is for
the on-site CCS system only and does not include the scope associated with the development of
the CO2 off-take/storage (including transportation, sequestration, EOR, utilization, and/or
utilization).” Although the 7-year schedule shown in Figure 3 is quite aggressive, it is more realistic
than the EPA’s schedule shown in the upper part of Figure 2. There is no explanation as to why
EPA chose to arbitrarily dismiss this timeline in favor of one that seemingly fits its regulatory goals
better.
Figure 3. More realistic development timeline for the capture portion of a CCS project.
COSTS
It is difficult to give precise costs for a CCS project because of the factors outlined above
and the specific needs of a specific facility to achieve a minimum of 90% CO2 capture. It has been
our experience that the general range for the capture facility at EGUs alone is $0.8–$1.3 billion.
In determining the needs for CO2 transportation, a “rule-of-thumb” (ROT) estimate for the
installed cost of a pipeline can be calculated with the following expression:
Installed Cost = Pipe O.D. (inches) × Pipeline Length (miles) × $100,000
This ROT estimate is a based on the FECM/NETL CO2 Transport Cost Model (2022)
(Model), where the installed cost reflects 2019 dollars and is based on the Parker equation within
the central U.S. region, referenced within the Model. Pipelines in other areas as well as any pipeline
specific needs (environmental impacts, water body crossings, large elevation changes, etc.) would
need to be addressed in addition to the estimated costs provided by the Model. The Model reflects
costs for new pipeline installations and reflects the steel pricing used within the model. If the
12
0017a
pricing of steel for the pipeline under consideration is different than the pricing included in the
Model, the Model would need to be adjusted to reflect the current pricing or additional cost
included to reflect the anticipated pricing.
EPA references $280K per inch-mile for pipeline installations and states that is the cost “to
construct new natural gas pipelines (“laterals”) and is an average for lateral development within
the contiguous U.S.” (page 33353). The term “lateral” typically refers to a line that delivers product
from a main line to a customer. As such, the installation of a lateral will normally include costs
such as hot tapping of the main line to install the lateral offtake, potentially shutting down the main
line to install the lateral offtake, etc. These are high-cost projects and are not typically included in
new pipeline construction.
From our experience in CCS field development, using an example of one injection well with
one monitoring well, assuming the CO2 is at pressure and not requiring additional pressurization
at the injection pad, and injecting 1 million metric tons of CO2 per year, the cost of injection field
development surpasses $30 million. Additional injection and monitoring will cause this value to
quickly increase. Also, the need for premium casing such as corrosion-resistant alloys (CRAs) can
add substantial cost to the drilling costs associated with the injection and monitoring wells
necessary for the project. In addition, the lead time for the CRA material can be 1 year or longer.
If material testing is required to determine which CRA would best serve the system, the time to
design, perform, and evaluate the material tests can require 6 to 12 months (depending on the
number and types of materials for testing) before the CRA material can be purchased. Given the
wide range of CO2 streams from the EGU and other facilities, different targeted injection horizons,
and very little information available on CRA testing in saline environments with CO2 streams with
multiple impurities, it is anticipated that material testing would be required to determine which
CRA material would be required. The effects that the material testing needs and the availability of
CRAs would have on a project are not evident in EPA’s consideration.
SUMMARY
Although CCS technology is progressing, it is too early to label it as commercially mature
technology, and more projects need to be completed to substantiate the performance levels
suggested by EPA. No large-scale (greater than 240-MW) CCS systems on EGUs are in operation
in the United States by which to determine the feasibility of CCS as a BSER option. Each facility’s
design considerations are unique and can vary widely due to variables such as the CO2 purity of
the emission stream, facility design, local energy costs, emission volumes, flue gas temperature
and pressure, the presence of contaminants, transition from cold or warm (standby) condition to
operation condition, ramping due to load changes, and the required purity of the CO2 emission
stream. When examining the case of Boundary Dam Unit 3 capture facility’s availability factor,
since the start of operation, the expectation of a capture facility to operate long enough through
the year to meet EPA’s proposed annual emission requirements has fallen short, and there is no
expectation that facilities in the United States will not see similar issues. The EPA storage
assumptions are not adequately documented, and the complexity of the permitting required will
greatly affect timelines for facilities to implement CCS. Based on the supported conclusions and
13
0018a
the current status of carbon capture technology, EGUs cannot meet the CO2 capture rates or the
timeline that EPA proposes.
REFERENCES
1.
Southwest
Research
Institute.
www.swri.org/industry/advanced-power-systemsconventional-power-generation/oxy-fuel-combustion (accessed July 2023).
2.
Kanniche, M.; Gros-Bonnivard, R.; Jaud, P.; Valle-Marcos, J.; Amann, J.-M.; Bouallou, C.
Pre-Combustion, Post-Combustion and Oxy-Combustion in Thermal Power Plant for CO2
Capture. Applied Thermal Engineering 2010, 30, 53–62.
3.
https://netl.doe.gov/node/7477 (accessed July 2023).
4.
www.csenergy.com.au/ArticleDocuments/222/Callide_Oxyfuel_Project_-_Completion__LEGACY_PUBLICATION_-_FINAL_eBook_artwork_-_smaller_size-1.pdf.aspx
(accessed 2023).
5.
Office of Fossil Energy and Carbon Management. www.energy.gov/fecm/pre-combustioncarbon-capture-research (accessed July 2023).
6.
www.netl.doe.gov/research/Coal/energy-systems/gasification/gasifipedia/great-plains
(accessed July 2023).
7.
Aghel, B.; Janati, S.; Wongwises, S.; Shadloo, M.S. Review on CO2 Capture by Blended
Amine Solutions. Int. J. of Greenhouse Gas Control 2022, 119.
8.
National Energy Technology Laboratory. DOE/NETL Carbon Capture Program – Carbon
Dioxide Capture Handbook; Aug 2015.
9.
Yamada, H. Amine-Based Capture of CO2 for Utilization and Storage. Polymer Journal
2020, 53, 93–102.
10.
Congressional Research Service. Carbon Capture and Sequestration (CCS) in the United
States, R44902, Oct 5, 2022.
11.
Dixon, R. Pathways to Commercial Liftoff: Carbon Management. Carbon Management.
12.
Abdallah, D.; Coddington, K.; Garofalo, J.F.; Wu, W.; Zhai, H. Policy-Driven Potential for
Deploying Carbon Capture and Sequestration in a Fossil-Rich Power Sector. Environmental
Science & Technology 2022, 56 (14), 9872–81.
13.
Rooney, P.C.; Daniels, D.D. Oxygen Solubility in Various Alkanolamine/Water Mixtures.
Petroleum Technology 1998, Q2, 97–101.
14.
Morken, A.K.; Pedersen, S.; Kleppe, E.R.; Wisthaler, A.; Vernstad, K.; Ullestad, O.; Flo,
N.E.; Faramarzi, L.; Hamborg, E.S. Degradation and Emission Results of Amine Plant
Operations from MEA Testing at the CO2 Technology Centre Mongstad. Energy Procedia
2017, 114, 1245–1262.
14
0019a
15.
Lombardo, G.; Shah, M.I.; Fostas, B.; Hvidsten, O.A.; Faramarzi, L.; de Cazenove, T.;
Lapaumier, H.; Rogiers, P. Results from Testing of a Brownian Diffusion Filter for Reducing
the Aerosol Concentration in a Residual Fluidized Catalytic Cracker Flue Gas at the
Technology Centre Mongstad. Presented at the 14th Greenhouse Gas Control Technologies
Conference
(GHGT-14)
Melbourne,
Australia,
Oct
21–26,
2018,
http://dx.doi.org/10.2139/ssrn.3366032.
16.
Fostas, B.; Gangstad, A.; Nenseter, B.; Pederson, S.; Sjovoll, M.; Sorensen, A.L. Effects of
NOx in the flue gas degradation of MEA. Energy Procedia 2011, 4, 1566–1573.
17.
Da Silva, E.F.; Hoff, K.A.; Booth, A. Emissions from CO2 Capture Plants: An Overview.
Energy Procedia 2013, 37, 784–790.
18.
Lepaumier, H.; Picq, D.; Carrette, P. New amines for CO2 Capture. I. Mechanisms of Amine
Degradation in the Presence of CO2. Ind. Eng. Chem. Res. 2009, 48 (20), 9061–9067.
19.
Davis, J.; Rochelle, G. Thermal Degradation of Monoethanolamine at Stripper Conditions.
Energy Procedia 2009, 1, 327–333.
20.
Dooley, J.J.; Davidson, C.L.; Dahowski, R.T. An Assessment of the Commercial Availability
of Carbon Dioxide Capture and Storage Technologies as of June 2009. Pacific Northwest
National Laboratory, June 2009.
21.
Capital News Service. Cnsmaryland.org/2013/12/03/in-maryland-a-potential-future-forcleaner-coal/ (accessed July 2023).
22.
Alberta Department of Energy. Quest Carbon Capture and Storage Project Annual Summary
Report; Revised Aug 2021.
23.
Giannaris, S.; Janowczyk, D.; Ruffini, J.; Hill, K.; Jacobs, B.; Bruce, C.; Feng, Y.; Srisang,
W. SaskPower’s Boundary Dam Unit 3 Carbon Capture Facility – The Journey to Achieving
Reliability. Presented at 15th International Conference on Greenhouse Gas Control
Technologies, Mar 15–18 2021.
24.
Groom N. Problems Plagued U.S. CO2 Capture Project Before Shutdown. Reuters, 2020,
www.reuters.com/article/us-usa-energy-carboncapture/problems-plagued-u-s-co2-captureproject-beforeshutdown-document-idUSKCN2523K8 (accessed 2023).
25.
Pradoo, P.; Jacobs, B.; Hill, K.; Wilkes, S.; Ruffini, J.; Srisang, W.; Feng, Y.; Daverne, D.;
Nelson, C. Improving the Operating Availability of the Boundary Dam Unit 3 Carbon
Capture Facility, Presented at the 16th International Conference on Greenhouse Gas Control
Technologies (GHGT-16), October 23–27, 2022, Lyon, France.
26.
www.eia.gov/todayinenergy/detail.php?id=48296 (accessed July 2023).
27.
www.energy.gov/fecm/articles/carbon-capture-opportunities-natural-gas-fired-powersystems (accessed July 2023).
15
0020a
28.
Wagman, D.C. The Three Factors That Doomed Kemper County IGCC. IEEE Spectrum,
2017,
https://spectrum.ieee.org/energywise/energy/fossilfuels/the-three-factors-thatdoomed-kemper-county-igcc (accessed 2023).
29.
U.S. Department of Energy Office of Fossil Energy. Southern Company – Kemper County,
Mississippi. https://www.energy.gov/fecm/southern-company-kemper-county-mississippi
(accessed 2023).
30.
Jones, K.B.; Barnhart, L.E.; Warwick, P.D.; Corum, M.D. Preliminary GIS Representation
of Deep Coal Areas for Carbon Dioxide Storage in the Contiguous United States and Alaska.
U.S.
Geological
Survey
Open-File
Report
2018–1178,
2019,
21
p.
https://doi.org/10.3133/ofr20181178.
31.
U.S. Environmental Protection Agency. Federal Register, Part III, 40 CFR Parts 124, 144,
145 et al. Dec 10, 2010. EPA preamble page 77242, www.govinfo.gov/content/pkg/FR2010-12-10/pdf/2010-29954.pdf (accessed Sept 2022).
32.
Goodman, A.; Hakala, A.; Bromhal, G.; Deel, D.; Rodosta, T.; Frailey, S.; Small, M.; Allen,
D.; Romanov, V.; Fazio, J.; Huerta, N.; McIntyre, D.; Kutchko, B.; Guthrie, G. U.S. DOE
Methodology for the Development of Geologic Storage Potential for Carbon Dioxide at the
National and Regional Scale. International Journal of Greenhouse Gas Control 2011, 5 (4),
952–965.
33.
Livers-Douglas, A.J.; Christianson, C.C.; Richards, T.L.; Dalkhaa, C.; Massman, N.;
Crocker, C.R.; Connors, K.C.; Vettleson, H.M. Project Development and Permitting
Strategies from the First Wave of Geologic CO2 Storage Projects in North Dakota; Task 4
Deliverable D8b for U.S. Department of Energy National Energy Technology Laboratory
Cooperative Agreement No. DE-FE0031838; Energy & Environmental Research Center:
Grand Forks, ND, 2022.
34.
Bachtel, L.A.; Lau, P.K.; Pogue, E.R.; Smith, D.D.; Klugman, N.C. Carbon Capture,
Utilization, and Storage: Class VI Wells and U.S. State Primacy. 2022, Mayer Brown,
https://www.mayerbrown.com/en/perspectives-events/publications/2022/06/carboncaptureutilization-and-storage-class-vi-wells-and-us-state-primacy (accessed Sept 2022).
35.
U.S. Environmental Protection Agency. EPA Report to Congress: Class VI Permitting; Oct
2022; www.epa.gov/system/files/documents/2022-1/EPA%20Class%20VI%20Permitting%
20Report%20to%20Congress.pdf (accessed 2023).
36.
U.S. Environmental Protection Agency. Class VI Wells Permitted by EPA; 2023,
www.epa.gov/uic/class-vi-wells-permitted-epa (accessed 2023).
16
0021a
Technical Comments on the
Carbon Capture Utilization and Sequestration Aspects of the Proposed
New Source Performance Standards for GHG Emissions from New and
Reconstructed EGUs; Emission Guidelines for GHG Emissions from Existing
EGUs; and Repeal of the Affordable Clean Energy Rule
Prepared by
J. Edward Cichanowicz
Consultant
Saratoga, CA
Michael C. Hein
Hein Analytics, LLC
Whitefish, MT
Prepared for the
American Public Power Association
National Rural Electric Cooperative Association
August 7, 2023
a
0022a
Table of Contents
1
Summary ......................................................................................................................... i
2
INTRODUCTION .............................................................................................................. 1
3
4
5
6
7
CCUS EXPERIENCE RELEVANT TO BSER ............................................................................ 2
3.1
Criteria for “Adequately Demonstrated” .............................................................................. 2
3.1.1 Industrial Applications ............................................................................................................ 2
3.1.2 Engineering FEED Studies ....................................................................................................... 4
3.2
Stages of Emerging Technology ............................................................................................ 6
3.2.1 First, Nth-of-a-Kind ................................................................................................................. 6
3.2.2 Commercially Availability ....................................................................................................... 7
3.3
North American Utility Scale Processes ................................................................................ 7
3.3.1 Sask Power Boundary Dam 3 .................................................................................................. 8
3.3.2 Petra Nova ............................................................................................................................ 10
REVIEW OF EPA’s PROJECTION OF CCUS COST ................................................................13
4.1
Overview ........................................................................................................................... 13
4.2
Inadequate Experience for Cost Basis ................................................................................. 13
4.3
FEED Study Capital Cost ..................................................................................................... 14
4.3.1 Coal-Fired Applications ......................................................................................................... 15
4.3.2 NGCC Applications ................................................................................................................ 18
4.4
Inadequate Basis for Levelized $/Tonne Calculation ........................................................... 19
4.4.1 Coal-fired Application ........................................................................................................... 19
4.4.2 NGCC Application ................................................................................................................. 21
CO2 Pipeline Permitting Issues .......................................................................................23
5.1
Background ........................................................................................................................ 23
5.1.1 Pipeline Length ..................................................................................................................... 23
5.1.2 Pipeline Projects: Select Description .................................................................................... 25
5.2
Permitting Experience ........................................................................................................ 26
5.2.1 Iowa ...................................................................................................................................... 26
5.2.2 Nebraska ............................................................................................................................... 27
5.2.3 Illinois.................................................................................................................................... 28
5.3
Timeline Summary ............................................................................................................. 28
CRITIQUE OF CCUS SCHEDULE ........................................................................................30
6.1
S&L Proposed Schedule ...................................................................................................... 30
6.2
Global CCS Institute Schedule............................................................................................. 31
6.3
EPA’s Compressed Schedule ............................................................................................... 32
6.4
Real World CCUS Project Schedules.................................................................................... 33
6.4.1 NGCC Schedule ..................................................................................................................... 34
6.4.2 Coal-Fired CCUS Applications ............................................................................................... 35
EPA-PROJECTED CCUS INSTALLATIONS ...........................................................................40
Appendix A. Flawed Cost Extrapolations for NGCC Application ..............................................44
Appendix B. Example CCUS Project Schedules .......................................................................46
b
0023a
List of Figures
Figure 3-1. Reliability of Boundary Dam Unit 3, CCS Process: Q1 2021 to Q1 2023....................... 9
Figure 4-1. Evolution of Wet FGD Technology: The First Decade................................................. 14
Figure 4-2. CCUS Capital Cost as Reported for Coal-Fired Demonstrations, FEED Studies .......... 16
Figure 4-3. CCUS Capital Cost as Reported for NGCC FEED Studies ............................................. 18
Figure 5-1. Candidate CO2 Pipeline Routing, Length: Plants Daniel and Miller ............................ 24
Figure 6-1. S&L CCS Deployment Schedule .................................................................................. 30
Figure 6-2. Global CCS Institute Deployment Schedule............................................................... 32
Figure 7-1. Geographic Location of Coal-Fired Generating Units EPA Projects to Retrofit CCUS:
100 km Proximity ................................................................................................................. 42
Figure 7-2. Geographic Location of Coal-Fired Generating Units EPA Projects to Retrofit CCUS:
200 km Proximity ................................................................................................................. 43
Figure B-1. Elk Hills Project Schedule: Post-FEED Study Activities .............................................. 46
Figure B-2. Minnkota Power Milton R Young Station: Complete Schedule ................................. 46
Figure B-3. Prairie State Final Engineering, Procurement, Construction Schedule ..................... 47
List of Tables
Table 4-1. Sensitivity Results: Role of Capital Cost, CCUS Reliability of Projected CO2 $/tonne . 20
Table 6-1. Summary Schedule Information: NGCC CCUS Projects ............................................... 34
Table 6-2. Summary Schedule Information: Coal-fired CCUS Projects ......................................... 35
Table 7-1. Units Projected by EPA IPM to Adopt CCUS by 2030 .................................................. 40
c
0024a
Summary
1 Summary
The U.S. Environmental Protection Agency (EPA) on May 23, 2023 proposed five separate
actions under Section 111 of the Clean Air Act addressing greenhouse gas emissions (GHG)
from fossil fuel power plants generating electrical power. EPA bases the proposed GHG rule on
many unverified assumptions, but the most egregious is that carbon capture utilization and
storage (CCUS) is a demonstrated technology and qualifies as best system of emission reduction
(BSER). EPA (improperly) designates CCUS as BSER, then extrapolates CCUS cost metrics to
a wide variety of generating units. That EPA uses questionable means to generalize CCUS cost
is of concern, but such concern is secondary to the unsubstantiated claim – and flaw in EPA’s
proposal - that CCUS is BSER. Consequently, all CCUS-related cost and performance
predictions fail.
This critical observation, supplemented with several others, is further described as follows:
The CCUS Utility experience base is inadequate.
There is a single CCUS process operating in North America relevant to utility power
generation—Sask Power Boundary Dam Unit 3. This unit has operated since 2014, and over
eight years of refinement exhibits increased reliability– which although improved can still be
compromised by failure of specialty, hard-to-acquire components that cannot be readily “spared”
on-site.
A second CCUS operating unit relevant to utility power application – the Petra Nova
“slipstream” project at the W.A. Parish station - operated for three years before termination in
March of 2020. As further discussed in Section 3, both demonstrations were significantly cofunded by federal (and for Sask Power Boundary Dam) the local (provincial) governments.
This collective large-scale CCUS experience – comprised of two units with one operating for an
abbreviated period – does not reflect the variety of conditions for CCUS application to the U.S.
generating fleet. Of particular note is that small-scale pilot plant tests for two proposed
demonstrations – conducted in 2015 (Minnkota Power Milton R. Young) and presently ongoing
(Basin Electric Dry Fork) and are necessary to address remaining risk. The lean CCUS
experience is in sharp contrast to real-world lessons accumulated in the early- and mid-70s with
first-generation flue gas desulfurization (FGD) technology, in which 20 generating units were
equipped with FGD and operated (some for five years) prior to a federal mandate to limit sulfur
dioxide (SO2) emissions.
Industrial CCUS applications are inadequate to reflect utility power generation.
EPA cites numerous industrial applications that due to scale, effluent gas treated, atypical CO2
content and process conditions, limited removal of CO2, or intermittent operation, are of
i
0025a
Summary
peripheral import to coal-fired utility application. Consequently, experience with industrial
applications has no impact on qualifying CCUS as utility-scale BSER.
Engineering “FEED” studies – regardless of the detail – do not deliver real-world operating
experience, and are not a substitute for “lessons learned” from authentic operation.
EPA, lacking relevant CCUS experience, cites up to 15 engineering (Front-End Engineering
Design, or FEED) studies as a basis for BSER. EPA’s premise is invalid for two reasons. First,
FEED studies do not address “final” design – the latter exercise a separate step, prior to
equipment procurement. Second, and more important, FEED studies are exclusively paper and
digital exercises that do not include the critical follow-through of building, operating, and
documenting experience that almost without exception leads to revised design.
This view is shared by two contractors that supported EPA in this rulemaking. Sargent & Lundy
Engineers (S&L) and Bechtel National Corporation state CCUS FEED studies leave risks that
are not addressed. Specifically, EPA sponsored S&L to develop model CCUS cost calculations
referenced in the Technical Support Documents, which state CCUS is an evolving technology.
Bechtel, prime contractor for the FEED study addressing CCUS retrofit to the Panda/Sherman
natural gas/combined cycle (NGCC) generating unit, state the present level of CCUS experience
is inadequate; they recommend – prior to full-sale application at Panda/Sherman – a large
capacity pilot plant test be conducted.
The CCUS cost basis – both capital requirement and the levelized cost per ton ($/ton) to avoid
CO2 - is highly uncertain, and will remain so without additional large-scale demonstrations.
EPA attempts to compensate for the lack of experience by featuring paper and digital
calculations, derived from unverified FEED studies, to determine the cost to avoid CO2
($/tonne).1 The shortcomings for coal and NGCC applications differ, and are described
separately.
Coal Applications. First, EPA – although citing FEED studies as a basis for BSER – ignore them
as a source of capital cost for actual sites. Alternatively, EPA uses capital costs for a hypothetical
unit, as determined by the National Energy Technology Laboratory (NETL) of the Department of
Energy (DOE). A more authentic cost would be derived from the “average” of the six FEED
studies – that even with uncertainty is “grounded” by actual site specifics. The difference in cost
is not small - EPA’s selected hypothetical unit capital cost is approximately 30% less than the
average of the six FEED studies.
Conversely, EPA, when seeking estimates of cost to avoid CO2 ($/tonne basis), changes course
and features the FEED studies ignored for capital cost. EPA highlights FEED study results along with several from international studies – to showcase that cost to avoid a tonne of CO2
($/tonne) cluster near the research and development (R&D) target of $40/tonne. As previously
described, FEED results are paper and digital exercises, describing facilities never built or tested.
1
All references to avoided cost are cited in terms of cost per metric tons ($/tonne).
ii
0026a
Summary
Further, key factors that drive the levelized cost result – capacity factor and remaining unit
lifetime – are not presented. EPA’s reporting of these costs is not transparent.
NGCC Applications. Similar to the case for coal duty, EPA ignores FEED studies as a source for
CCUS capital. EPA again defers to an NETL study of a hypothetical unit for capital cost – but
not really, as EPA “discounts” the capital inferred. Specifically, EPA determines CCUS capital
cost per net power output – the conventional metric - by normalizing the CCUS cost by net
power generated prior to CCUS. This unusual combination – normalizing CCUS cost by net
power prior to retrofit – is unprecedented, and ignores the loss of 33 MW consumed by CCUS.
No explanation is offered for what is effectively a discount.
In summary – for both coal-fired and NGCC application – CCUS costs remains highly uncertain.
EPA’s projected schedule for CCUS deployment – from concept evaluation to injection of CO2
for sequestration or enhanced oil recovery – is unrealistic and compressed even compared to
optimistic projects.
EPA ignores schedules to retrofit CCUS issued by two sources: the contractor S&L whom they
engaged for this purpose, and the Global CCS Institute. S&L developed for EPA a CCUS retrofit
schedule describing 6.25-7 years as necessary, and concede this applies to a partial scope of
duties by ignoring CO2 transportation (e.g. pipeline construction and permitting) and terrestrial
sequestration (e.g. site development and permitting). The Global CCS Institute cites almost nine
years as necessary, but “pass” on realistic permitting challenges – by noting their schedule
assumes “…. there is no significant community opposition” to the project. Experience in the
U.S. particularly the Midwest – belies this assumption.
EPA assumes the responsibility of completing the schedule. EPA adds activities to S&L’s scope
but compress the schedule by about two years. The resulting five-year schedule – slightly more
than half of the 8.25 years advised by the Global CCS Institute - allocates one half-year to for
CO2 “transport and storage” feasibility and two years for CO2 sequestration “site characterization
and permitting.” These estimates are contrary to plentiful evidence such timeframes are not
credible. Section 5 describes how acquiring a CO2 pipeline permit – such as the proposed
Navigator project in Iowa - appears to require 3.5 years and only if no other roadblocks emerge
prior to end-of-year 2024. Section 6 summarizes detailed schedules developed for the FEED
studies and show under ideal conditions – a “head-start” for sequestration site development and
no barriers to CO2 pipelines – eight years are required. Some projects will require possibly 12
years.
These studies suggest not only that the five-year time frame is unrealistic, with 10 years or more
required for many projects.
CCUS does not qualify as BSER.
EPA is to select BSER after considering if a technology is “adequately demonstrated”,
“commercially available,” and can be deployed for a cost that is “reasonable”, all while
iii
0027a
Summary
representing the best balance of economic, environmental, and energy considerations. Two utility
demonstrations – both with significant government cofunding – do not comprise an adequate
demonstration. Process equipment for CCUS can be purchased – but without meaningful
guarantees from process supplier, the technology is not fully commercially available. Costs,
projected mostly from paper and digital FEED studies, are highly uncertain.
CCUS is distinguished from all precedent environmental controls in that a significant fraction of
power produced that would be directed to the grid – 20-30% for coal- and 10% for NGCCapplication – is consumed by the process. This collection of conditions does not qualify CCUS as
BSER in the present state of development.
iv
0028a
Introduction
2 INTRODUCTION
The U.S. Environmental Protection Agency (EPA) on May 23, 2023 proposed five separate
actions under Section 111 of the Clean Air Act addressing greenhouse gas emissions (GHG)
from fossil fuel power plants generating electrical power. New Source Performance Standards
(NSPS) for stationary combustion turbines and coal-fired generating units to limit emissions of
CO2 are proposed, as well such limits for existing fossil fuel generating units fired by coal, or gas
turbines operating in simple or combined cycle duty.
Of the elements of EPA’s proposed regulation, there is one critical premise – the role EPA
assigns to carbon capture utilization and storage (CCUS). EPA submits that CCUS – in the
present state-of-art technology –is commercially proven and feasible for utility application to
both coal-fired and natural gas combined cycle (NGCC) generating units. EPA projects via its
Integrated Planning Model (IPM) that 39 coal-fired power plants – totaling almost 14 gigawatts
(GW) of capacity – will adopt CCUS by 2030.2 The premise of EPA’s modeling results in
arbitrarily determining that CCUS is the best system of emissions reduction (BSER).
This report addresses the technology status of CCUS in terms of designation as BSER. The
operating experience to underpin future applications of CCUS technology is reviewed,
considering commercial-scale duty, laboratory tests, and the paper or digital design studies
funded by the National Energy Technology Laboratory (NETL) and others.
This report is comprised of seven sections and two appendices. Section 3 addresses the
shortcomings with industrial experience and Front-End Engineering and Design (FEED) studies,
the features of emerging technology, and the limited experience with two units equipped with
CCUS. Section 4 reviews EPA’s evaluation of CCUS cost, addressing capital required and the
levelized cost to avoid CO2 on a dollar per metric tonne basis ($/tonne), including the impact of
tax benefits accrued through the Inflation Reduction Act (IRA). Section 5 highlights one aspect
of CCUS EPA does not address in detail – the task of securing CO2 pipelines for delivery to sites
for sequestration or use for enhanced oil recovery (EOR). Section 6 addresses EPA’s assumption
that a five-year deployment schedule is realistic. Section 7 projects on a continental map of
North America the locations of EPA projected CCUS applications, showing the relationship to
existing and proposed CO2 pipeline routing and potential geological sequestration or EOR sites.
Select backup material is presented in Appendices A and B.
2
U.S. EPA, Integrated Proposal Modeling and Updated Baseline Analysis, Memo to the Docket
(EPA_HQ_OAR_2023_0072), July 7, 2023. Hereafter EPA 2023 Integrated Baseline Analysis.
1
0029a
Critique of EPA’s
Designation of CCUS as BSER
3 CCUS EXPERIENCE RELEVANT TO BSER
The EPA has designated CCUS as BSER based on the following rationale:
The technology has been studied, examined, and tested for decades and it has reached a point in
its development where it is adequately demonstrated and commercially available.3
The additional economic incentives are important for establishing that the cost of CCS is
reasonable, and an appropriate BSER.4
Section 3 reviews the technical basis of CCUS, focusing on relevant utility power generation
experience, considering the definition of technology as adequately demonstrated and
commercially available, and the incurred cost.
It should be noted EPA does not propose criteria by which to gauge CCUS in terms of the
metrics “adequately demonstrated”, “commercially available”, and a cost that is “reasonable”,
and “appropriate.” Nor does EPA address the decision to select a technology with the “best”
balance of economic, environmental, and energy considerations.
3.1 Criteria for “Adequately Demonstrated”
A technology is considered “demonstrated” when there is (a) adequate experience that reflects
projected operating duty, (b) confidence that operation is reliable over extended periods of time,
and (c) the technology suppliers can offer meaningful guarantees, more than equipment and
engineering services for sale. EPA in several instances distorts the meaning of the term
“demonstrated”. Most notable are (a) application at industrial or small-scale processes, and (b)
the significance of engineering studies, the latter without corroborating results. These are
described as follows:
3.1.1 Industrial Applications
EPA submit that industrial application of CCUS – particularly for cases that “report” 90% CO2
capture – contribute to demonstrating CCUS for utility applications.
Industrial applications significantly differ from utility-scale power generation. Utility
applications are distinguished by continual 24 x 7 duty, operation at high reliability, and
processing flue gas with CO2 content that differs from utility power generation – the latter
typically 3-4% CO2 for NGCC application and 11-13% CO2 content for coal-fired application.
Almost all non-utility applications treat product gases with higher CO2 concentrations – such as
3
Greenhouse Gas Mitigation Measures for Steam Generating Units – Technical Support Document.
Docket EPA-HQ-OAR-2023-0072. Page 35. Hereafter Steam EGU TSD.
4
Ibid.
2
0030a
Critique of EPA’s
Designation of CCUS as BSER
chemical and ethanol production, and processing of hydrogen and ammonia, by up to a factor of
10. These high concentrations of CO2 elevate the “driving force” for mass transfer and
adsorption, which combined with a smaller scale and shorter physical distance over which to
effect mixing and CO2 absorption present different challenges than for power generation.
EPA’s industrial “reference applications” are not relevant to utility duty. Specifically, EPA
claims CCUS viability is “…. further corroborated by CO2 capture projects assisted by grants,
loan guarantees, and Federal tax credits for “clean coal technology” authorized by the
EPAct05. 80 FR 64541–42 (October 23, 2015).”5 EPA cite a compilation of 72 CCUS projects –
demonstration tests, pilot plant test, CO2 storage, and transport activities – as relevant supporting
their assessment, per Excel file “Attachment_1”,6 of which only two treat the entirety of gas flow
generated. These two facilities – the Searles Valley Minerals caustic soda plant and the Quest
methane reformer – do not represent large-scale utility duty, nor is there evidence that CO2
removal matched that proposed by EPA for 24x7 duty. Other sites referenced by EPA are the
“slip stream” category of process testing for which CCUS reliability does not limit that of the
host unit.7 Two “slip-steam” tests cited in the “Attachment 1” reference file are discussed – the
Bellingham Energy Center for NGCC duty, and the Petra Nova demonstration (discussed in
Section 3.3).
The sites reported to process the entirety of product gas – Searles Valley Mineral and Quest – are
further described as follows:
Searles Valley Minerals. Public information suggests CO2 capture is either intermittent or
derives CO2 removal well below 90%. The Searles site is comprised of three coal-fired units –
two generating 27.5 MW and a third at 7.5 MW.8 The CO2 removal capability is cited as 800
tons per day9 which suggests relaxed duty. Specifically, if the CO2 removal process treats flue
gas from the smallest (7.5 MW) capacity unit, operation at 80% capacity factor will generate
2,375 tons of CO2 per day – and daily CO2 removal of 800 tons implies either a 33% removal for
a complete 24-hour day, or 90% CO2 removal for 35% operating time (perhaps one “daytime”
shift). These performance metrics are not adequate to qualify CCS as demonstrated technology.
Quest. The effluent from this methane reforming process does not reflect combustion products,
as CO2 content is elevated compared to utility application. Experience with CO2 removal at
highly elevated content – although contributing to general CCUS knowledge – is not a basis to
designate CCUS as BSER for utility application.
5
Steam EGU TSD. Page 22.
EPA-HQ-OAR-2023-0072-0061_attachment_1.
7
Three additional facilities are listed as operating CO2 capture, but as a “slipstream’. (AES Warrior Run,
AES Shady Point, and Bellingham Energy Center). The slipstream process arrangement – a useful means
for research and development - does not link the reliability of the host process to the CO2 capture
technology – and thus cannot represent conditions for 24x7 utility power generation demonstration.
8
Energy Information Agency 860 Data, File 3_1_Generator_Y2021. Operable tab, Rows 9148-9150.
9
Elmoudir, W. et. al., HTC Solvent Reclaimer system at Searles Valley Minerals Facility in Trona, CA,
Energy Procedia 63 (2): 6156-6165, December 2014.
6
3
0031a
Critique of EPA’s
Designation of CCUS as BSER
Bellingham Energy Center. This NGCC unit is host to a 40 MW slip-stream employing a firstgeneration amine-based process (that evolved as the Flour Econoamine process). There is no data
available to describe these results – a DOE “fact sheet” reports the unit operated from 1991
through 2005, with CO2 removal of “85-95%”.10 It is not known if operation was continual
versus intermittent, pending market demand for commercial grade CO2. If periods of 85-95%
CO2 removal are interspersed with lower targets, this experience does not support BSER for
utility application.
In summary, experience with industrial CCUS applications, although contributing to CCUS
technology evolution, does not qualify CCUS as demonstrated for utility duty.
3.1.2 Engineering FEED Studies
EPA claims studies of CCS feasibility for utility duty – “Front End Engineered Design” or FEED
studies – contribute to designating the technology as “demonstrated”.
Three phases of analysis are typically employed to develop a CO2 capture design. The first step
defines the overall features of the design, using general site information, and “budgetary” cost
quotations. This “pre-FEED” study presents a feasibility “yes/no” test.
The second step – the FEED study – is intended to (a) develop in more detail process flowsheets
and/or equipment arrangement drawings, and (b) solicit budgetary quotations from suppliers to
establish cost and availability. Some FEED studies include a construction plan, addressing the
fabrication and delivery of the largest components to the site. At present, there are 13 such
complete FEED studies (listed in Section 5) addressing coal-fired and NGCC generators.
The third phase is detailed engineering which specifies equipment physical attributes, layout, and
an operating plan in detail to develop a request for proposal and solicit a supplier “firm” designs
and cost. This detailed engineering step has been completed only for the Sask Power Boundary
Dam 3 and the Petra Nova projects. For developed technology, this third phase should solicit
performance and/or reliability guarantee from equipment suppliers.
EPA cite four FEED studies for coal and three for NGCC,11 with seven more planned described
in Attachment _1.12 EPA rightfully identifies these FEED studies as “…projects in the early
stages of assessing the merits of retrofitting coal steam EGUs with CCS technology”, with
potential for “…the application of CCS to existing gas facilities”.13
10
U.S. Department of Energy (DOE). Carbon Capture Opportunities for Natural Gas Fired Power
Systems. Available at https://www.energy.gov/fecm/articles/carbon-capture-opportunities-natural-gasfired-power-systems.
11
Steam EGU TSD. P. 23.
12
EPA-HQ-OAR-2023-0072-0061_attachment_1.
13
Steam EGU TSD. P. 23.
4
0032a
Critique of EPA’s
Designation of CCUS as BSER
As will be shown for several projects, there remain significant “post-FEED” details in design and
specifications for procurement. Most importantly, FEED studies as paper and digital exercises
are absent the critically important “learning by doing” – the frequently quoted guidance from the
Global CCS Institute as necessary to evolve CCUS.14
Four FEED studies are cited in the Steam EGU TSD for coal-fired duty: Basin Electric Dry Fork,
Prairie State Generating Station, the Milton R. Young Station of Minnkota Power, and Nebraska
Public Power District’s Gerald Gentleman Station. Each of these studies is complete and project
CCUS capital cost, and with assumptions of unit lifetime and capacity factor project an implied
cost to avoid CO2 ($/tonne). Capital cost results from these projects – in addition to analogous
studies addressing Enchant Energy San Juan and Sask Power’s Shand station – are addressed in
Section 4.
Four newly launched studies have not progressed to delivering cost estimates. These are Cleco
Brame Energy Center Madison Unit 3 (pet coke/bit coal) (Lena, LA); Duke Energy’s
Edwardsport integrated gasification combined cycle (IGCC) facility (Edwardsport, IN); Four
Corners Station (located on the Navajo Nation in AZ); and CWL&P Dallman Unit 4
(Springfield, IL).
FEED studies are important - but on their own – are inadequate to qualify a technology as
commercial. In at least two instances, FEED study authors advised additional pilot plant testing.
Basin Electric Dry Fork Coal-Fired. A 2020 FEED study by S&L evaluated MTR’s membrane
CO2 capture technology for application to the Basin Electric Dry Fork station, and had advised
the next phase of activities a 10 MW “large” pilot plant test,15 evolving to a “slip stream”
configuration for “partial capture conditions” at 400 MW capacity. This advisement offered in
2020 is testament to the evolving nature of CCUS technology.
NGCC Combined Cycle. A FEED study conducted by Bechtel National examined retrofit of a
generic monoethanolamine (MEA) process to the 758 MW Panda Sherman Power Project. The
principal investigators noted: “At the time of this FEED study, no full-scale NGCC power plants
with PCC was built anywhere in the world; even pilot studies using NGCC flue gas conditions
were limited. This leads to a lack of data for process simulation model validation under
conditions of interest for commercial NGCC+PCC plants….”.16
14
Technology Readiness and Cost for CCS, Global CCS Institute, March 2021. Available at
https://www.globalccsinstitute.com/resources/publications-reports-research/technology-readiness-and-costs-of-ccs/.
15
Freeman, B. et. al., Commercial-Scale FEED Study for MTR’s Membrane CO2 Capture Process,
presentation to the Carbon Capture Front End Engineering Design Studies and CarbonSafe 2020
Integrated Review Webinar, August 17-19, 2020. P. 23.
16
Elliot, W.R. et. al., Front-End Engineering Design (FEED) Study for a Carbon Capture Plant Retrofit
to a Natural Gas-Fired Gas Turbine Combined Cycle Power Plant (2x2x1 Duct-Fired 758-MWe Facility
with F Class Turbines), Final Scientific/Technical Report, DE-FE0031848, March, 2022. P. 2. Hereafter
Panda Sherman 2022 Final Report.
5
0033a
Critique of EPA’s
Designation of CCUS as BSER
The principal investigator then concludes: “A pilot testing program is therefore proposed to
resolve most of these design uncertainties, generally duplicating all process elements of the fullscale PCC unit apart from CO2 product compression.”17
This is S&L’s second advisement that CCUS is emerging technology – in addition to
recommending a pilot plant test at Dry Fork prior to commercial demonstration, S&L describe
the technology as “emerging” in an explanatory note issued with the proposed CCUS schedule.18
FEED Studies are critical to project development for CCS as this technology is an emerging
technology with very limited full-scale / commercial installations.
In summary, FEED studies develop the arrangement of process equipment and preliminary cost
for CCUS. These conceptual exercises are inadequate to qualify CCUS as BSER.
3.2 Stages of Emerging Technology
Commercially available technologies are characterized by operating experience that enables
process suppliers to provide meaningful performance guarantees.
As noted by S&L, CCS is considered an “emerging technology”19 which typically evolve in
several stages. Early projects are based on limited experience and the role of process suppliers
evolved during this period. It must be emphasized there is stark contrast between a supplier
offering “for sale” an engineered design and fabricated hardware, in contrast to providing
meaningful process guarantees. This subsection further addresses these topics.
3.2.1 First, Nth-of-a-Kind
Any new process – or application of an evolving process to conditions outside present-day
experience – is considered the “first” of a “kind” (FOAK). Such FOAK designs are characterized
by uncertainty in terms of equipment arrangement, process conditions (reaction chemistry, flow
field, temperature), and operating duty, and the risk to achieve environmental control
performance and reliability.
FOAK designs can address risk and uncertainty but only by large scale testing and operation for
extended periods. Projects subsequent to FOAK are described as the “Nth-of-a-Kind” (NOAK),
in which additional (the nth) application addresses evolving conditions. There is no clear
delineation between the number of FOAK applications necessary to evolve to NOAK.
Power industry technologies are not considered “demonstrated” until adequate “NOAK”
applications operate for sufficient time, defining and resolving uncertainties. There is no broadly
recognized threshold for the number of acceptable NOAK projects to be completed prior to
17
Ibid.
S&L_CCS_Schedule_EPA-HQ-OAR-2023-0072-0061_attachment_16.pdf.
19
Ibid.
18
6
0034a
Critique of EPA’s
Designation of CCUS as BSER
commercial maturity. The DOE acknowledges this uncertainty with regard to CCUS, in noting
NOAK designs can include equipment that “…. are not fully mature (e.g. plants with IGCC and
any plant with CO2 capture…”, and will incur costs higher than reflected within their most recent
analysis.20
The fact that CCUS is a FOAK or NOAK is evidenced by the demonstrations at the Basin
Electric Dry Fork and Minnkota Power Milton R. Young station. As described in Section 6, the
site-specific process design for these sites relies heavily on pilot plant tests – either completed (in
2015) or presently underway – at the site. The uncertainties which remain are best addressed at
pilot scale which is proof CCUS technology is not mature.
The uncertainty of FOAK designs is also recognized in the Princeton “Net-Zero” study.21 The
analysis suggests five FOAK designs must be built and operated for – in their opinion –
sufficient time for costs to “settle”; but with broader implications for mitigating risk.
3.2.2 Commercially Availability
EPA implies CCUS processes are commercially available when suppliers offer to sell the
necessary process equipment and engineering services. However, a supplier offering to design,
procure and install such hardware does not constitute commercial availability. The missing
requirement is meaningful guarantees of process performance, backed with remedial action if
goals for emissions removal or reliability are not attained.
Neither Sask Power or Petra Nova process hardware were reported as awarded performance
guarantees. That absence of commercial guarantees is the reason both projects were significantly
co-funded by federal and local governmental entities, with additional funds defraying risk
inherent to a FOAK concept.
3.3 North American Utility Scale Processes
At present, there is one operating CCUS unit in North America from which to assess commercial
feasibility – Sask Power Boundary Dam Unit 3. A second CCUS-equipped unit – Petra Nova –
operated for 3 years (terminating in March 2020). Both of these demonstrations provide
significant experience – but on their own does not establish CCUS as demonstrated and
commercially available.
A summary of these two projects is presented in this subsection.
20
Cost and Performance Baseline for Fossil Energy Plants Volume 1: Bituminous Coal and Natural Gas
to Electricity, DOE/NETL – 2023/4320, October 14, 2022. Hereafter 2020 Baseline CCUS Costs. P.50
21
The Princeton Net-Zero Project - Potential Pathways, Infrastructure, and Impacts. Available at
https://netzeroamerica.princeton.edu/?explorer=year&state=national&table=2020&limit=200
7
0035a
Critique of EPA’s
Designation of CCUS as BSER
3.3.1 Sask Power Boundary Dam 3
Overview. Sask Power has operated CCUS at Boundary Dam Unit 3 since 2014, employing an
early generation Cansolv CO2 process. Inherent to the Cansolv process is a SO2 removal step –
controlling SO2 to less than 10 parts per million (ppm) – that combined with improved
particulate matter control protects the amine sorbent from degradation.
This activity was significantly co-funded by the Canadian and Saskatchewan provincial
governments. The capital budget is approximately $1.2 B (USD), of which $240 M is provided
by the Canadian and provincial government. The retrofit of CCUS was contemporaneous with
refurbishing the steam turbine and the electric power generator to support 30-year operation.
CO2 Disposition. CO2 is compressed to 2,500 pounds per square inch gauge (psig) and
transported 70 kilometers (km) by pipeline to the Weyburn oilfield for EOR, where it is injected
1.7 km underground. CO2 not employed for EOR is transported 2 km for sequestration in the
Deadwood saline aquifer (referred to as Aquistore).
As the Steam EGU TSD notes, a key issue is protecting the amine sorbent from decay with
exposure to trace metals and SO2. Several issues not unique to CCUS process equipment have
compromised reliability. EPA note CCUS reliability was compromised in 2Q 2021 due to a
failed CO2 compressor but dismiss this as not inherent to CCUS reliability. However, Sask
Power cites these large, special purpose components as rare, and due to limited inventory are not
immediately accessible. The cost to maintain “spares” on site is prohibitive. To assure high
reliability, additional capital cost should be allocated to provide access to spare equipment;
alternatively, enhanced operation and maintenance (O&M) should be planned and include
downtime for “preventive” maintenance.
Observations are offered for Sask Power Boundary Dam 3 in three categories: reliability, cost of
CO2 capture ($/tonne), and implementation schedule.
Reliability. The availability of the Boundary Dam 3 CCUS facility is publicly reported in the
Sask Power’s CCUS Blog.22 This latter source reports the reliability separately of the host boiler
and CCUS process since Q1 2021. Figure 3-1 presents two quarterly reports that describe
reliability continuously from Q1 2020 through Q1 2023 (available as of July 24, 2023). The top
portion of each chart reports Boundary Dam Unit 3 availability (white background) and the
lower portion of each chart reports CCS facility availability (gray background).
Considering CCS facility alone, Figure 3-1 shows the average of availability from Q2 2021
through Q1 2023 is 64.5% over this period. The loss of the compressor is a major contributor to
this shortfall and a factor to be encountered in commercial duty.
22
https://www.saskpower.com/about-us/our-company/blog/2023/bd3-status-update-q1-2023.
8
0036a
Critique of EPA’s
Designation of CCUS as BSER
Figure 3-1. Reliability of Boundary Dam Unit 3, CCS Process: Q1 2021 to Q1 2023
Cost. As a FOAK retrofit, Boundary Dam 3 cost although not representative is informative. As
previously described, capital cost (including plant refurbishment was of $1.2 B (U.S, 2014-dollar
basis),23 with the Canadian government contributing $240 M.24
Sask Power report 50 percent of the cost is attributable to the CO2 capture and regeneration
process, 30 percent for power plant refurbishment, and 20 percent for other emissions control
and other efficiency upgrades.25 Consequently, $600 M of capital is accounted for CCUS,
equivalent to $5,405/kW (net, w/CCUS).
The levelized cost to avoid one tonne of CO2, as reported by the CCS Knowledge Center, is
$105. This cost estimate is based on a capacity factor of 85 percent, lifetime of 30 years, and a
credit for CO2 as EOR.26 It should be noted CCUS availability since 1Q 2021 has prevented this
cost of $105/tonne from being achieved.
23
https://financialpost.com/commodities/energy/jim-prentice-to-wind-down-carbon-capture-fund-inalberta-new-projects-on-hold?. Canadian dollar values at 0.86 USD in 2014.
24
See: https://www.powermag.com/saskpowers-boundary-dam-carbon-capture-project-wins-powershighest-award/.
25
Giannaris et. al. 2021.
26
The Shand CCS Feasibility Study Public Report, November 2018, CCS Knowledge Center. Available
at https://ccsknowledge.com/initiatives/2nd-generation-ccs---Shand-study. Hereafter Shand 2018
Feasibility Report.
9
0037a
Critique of EPA’s
Designation of CCUS as BSER
Schedule. Sask Power does not report schedule details from concept inception to delivering CO2
for EOR, but reports the project took six years from “…commitment to completion”.27 Given
the proximity to both an existing oil field (Weyburn) and saline reservoir (~10 mile) the actions
to acquire permits – not reported by Sask Power – are likely atypical for most of the U.S.
domestic fleet.
Sask Power’s schedule may be relevant only for units situated in oil producing regions.
Considering the cost subsidy, the reliability issues, and the incurred cost of CO2 control,
Boundary Dam 3 experience does not represent CCUS as “adequately demonstrated” or
“commercially available.”
3.3.2 Petra Nova
Overview. NRG, owners of the W. A. Parish Generating Station, operated the Petra Nova CCUS
process at Unit 3 from March 2017 through March 2020. This process employed the secondgeneration KM-CDR solvent developed by MHI and Kansai Electric Power Company,
previously tested at 25 MW scale at Alabama Power Company’s Barry Station.
The Petra Nova demonstration, significantly co-funded by the U.S. DOE, required capital of
approximately $1 B. The CCUS process is not applied to the entirety of Unit 3 flue gas, but
rather a 240 MW-equivalent slipstream, thus not affecting host unit reliability. Petra Nova’s
CCUS process hardware is unique – a 78 MW gas turbine (GE 7FA) was installed with a heat
recovery steam generator (HRSG), the latter the source for CCUS auxiliary steam. The power
generated by the gas turbine not consumed by the CCUS process (reported as 35 MW) is sold to
the energy grid.28
CO2 Disposition. CO2 upon regeneration is compressed to 1,900 psig and transported 81 miles by
pipeline for EOR at the West Ranch site, requiring injection between 5,000 feet to 6,000 feet
underground. Unlike Boundary Dam Unit 3, there is no alternative means of CO2 disposition.
Similar to Boundary Dam Unit 3, numerous operating issues were encountered with ancillary
components. Heat exchangers processing reagent denoted as cool lean (without CO2) and hot
rich (with CO2) were prone to leaks, while the gas quencher accumulated deposits that restricted
performance. Some issues are attributed to penetration of SO2 entering the capture process.
These components are necessary for CCUS, and their failure should not be dismissed as
incidental. In the third operating year, additional factors such as tube corrosion in the solvent
reclaimer were encountered that – similar to Sask Power – can compromise CO2 compressor
performance.
27
SaskPower’s Boundary Dam Carbon Capture Project Wins Powers Highest Award, Power,
https://www.powermag.com/saskpowers-boundary-dam-carbon-capture-project-wins-powers-highestaward/.
28
W.A. Parish Post-Combustion CO2 Capture and Sequestration: Demonstration Project DOE Award
Number DE-FE0003311 Final Scientific/Technical Report, Report DOE-PNPH-03311, March 31, 2020.
Hereafter Petra Nova 2020 Final Report.
10
0038a
Critique of EPA’s
Designation of CCUS as BSER
Observations are offered for the Petra Nova project in three categories: Reliability, cost of CO2
capture ($/), and implementation schedule.
Reliability. CCS reliability increased each year. Considering both the CO2 capture system and
the source of auxiliary steam, in the last operating year (2019) 49 days were fully or partially
lost. Although an improvement from the 108 observed in 2017, the CCUS process was still not
available for 13.4% of operating time in the third and best year.
Cost. Petra Nova reports a $1B capital cost with approximately 60% expended for the CO2
capture equipment, gas turbine, and the HRSG – the latter to provide auxiliary steam. The
remaining approximately 40% of the cost was dedicated to administrative matters, the share of
the CO2 pipeline, and improvements to the oil field to enable higher CO2 injection for EOR.
Funding sources were a DOE grant of $190 M, financing of $250 M, and equity offered by the
sponsors. One trade journal noted Petra Nova financing conditions were unique: “Like other
early CCS demonstration projects, Petra Nova’s financial viability relied on a rare alignment of
incentives, including a DOE grant, cheap credit from Japan, and part-ownership of an oilfield,
which probably has limited relevance for future CCS plans under the new fiscal policy.”29
The project sponsors are not forthcoming with actual incurred cost per tonne ($/tonne). The final
report to DOE30 does not address this cost metric. The EPA in the Steam EGU TSD cite a cost of
$65/tonne, as referenced to the Global CCS Institute,31 whom in turn cite a Petra Nova Technical
Report from a period (July 2014 through December 2016) prior to unit operation.32
Consequently, the $65/tonne is a pre-operational estimate, no different than a FEED evaluation,
for which basic parameters of unit lifetime and capacity factor are not shared. Also, project
economics should account for the incremental revenue derived from the 35 MW delivered by the
gas turbine (acquired under the CCUS budget) to the grid. (This revenue could lower CCUS
levelized cost, but no details are provided.
Schedule. Petra Nova required a 6-year schedule for their activities, with work initiating in early
2011 to enable an air permit to be filed in September 2011,33 although details are absent in the
public schedule.34 Petra Nova is unique as the Texas Gulf Coast provides an ideal location for
CCUS given existing pipeline corridors and proximity of oilfields that can readily accept
significant CO2 injection.
29
https://www.nsenergybusiness.com/features/petra-nova-carbon-capture-project/#.
Petra Nova 2020 Final Report.
31
Technology Readiness and Costs of CCS, March 2021, the Global CCS Institute. See page 35.
32
W.A. Parish Post-Combustion CO2 Capture and Sequestration Project, Topical Report/Final Public
Design Report, Award No. DE-FE0003311, for July 01, 2014 to December 31, 2016. See page 30.
33
Ibid. P. 13.
34
Petra Nova Carbon Capture, presented to the Carbon Capture, Utilization and Storage, and Oil and
Gas Technologies Integrated Annual Review Meeting, August, 2019. Graphic 3. Available at:
https://netl.doe.gov/sites/default/files/netl-file/Anthony-Petra-Nova-Pittsburgh-Final.pdf.
30
11
0039a
Critique of EPA’s
Designation of CCUS as BSER
The Petra Nova project schedule may be relevant only for units situated in oil producing locales.
Considering the cost subsidy required, and complicated by reluctance to release the final costs,
the Petra Nova project – although contributing to CCUS technology development - does not
qualify CCUS as BSER.
12
0040a
Review of EPA’s Projection
of CCUS cost
4 REVIEW OF EPA’s PROJECTION OF CCUS COST
4.1 Overview
Section 4 critiques EPA’s cost evaluation of CCUS. As noted in Section 3, there are only two
verified capital cost reports for CCUS –Sask Power Boundary Dam Unit 3 and Petra Nova.
EPA’s proposed trajectory of CCUS evolution more optimistic compared to that observed for
flue gas desulfurization (FGD) technology, in which multiple demonstration tests (many <100
MW) operated for up to 5 years prior to federal legislation mandating FGD deployment. Further,
EPA is inconsistent in their selection of references – after lauding FEED studies that EPA
submits demonstrate the technology as commercial – EPA ignores these results when seeking
capital cost. Finally, EPA does not consider the risk to reliability presented by CCUS, that
compromises CO2 removed and tax benefits accrued through the IRA.
These are further described as follows.
4.2 Inadequate Experience for Cost Basis
There is little verified experience with CCUS to base EPA’s estimate of cost. In contrast, FGD
evolved through approximately 20 commercial-scale processes that provided significant
experience at utility conditions, prior to federal legislation mandating their use.
Figure 4-1 presents for FGD technology the installation date and flue gas equivalent generating
capacity treated for installations through mid-1978. It should be noted that 20 FGD installations
were installed and operating prior to the 1977 Clean Air Act Amendments - with at least 10
operating for up to five years.35 This experience served as the basis to mandate the use of
FGD.36
Figure 4-1 shows that – prior to 1977 and drafting of the Clean Air Act Amendments in that year
– FGD technology evolved in a logical manner. The first three years (through 1975) saw 10
installations, of which all but three were of 150 MW of capacity or less. Notably, three
installations that exceeded 400 MW in capacity were an early design variant – the “combined
particulate/SO2” process – which incurred either reliability or SO2 removal challenges. These
combined particulate/SO2 processes – almost without exception – required refurbishment or
replacement with “conventional” limestone FGD technology.
35
Shattuck, D. et. al., A History of Flue Gas Desulfurization (FGD) – The Early Years. Available at
https://www.science.gov/topicpages/g/gas+desulphurization+fgd.
36
Aldy, J. E. et. al., Looking Back at Fifty Years of the Clean Air Act, Resources for the Future Report
20-01 October 2020, Revised December 2020. Available at: https://media.rff.org/documents/WP_2001_rev._Looking_Back_at_Fifty_Years_of_the_Clean_Air_Act_hmvW55y.pdf.
13
0041a
Review of EPA’s Projection
of CCUS cost
1000
900
Nameplate Capacity (MW)
800
700
600
Early Generation
Particulate/SO2
Venturi Concept
500
400
300
200
100
0
Jan-71
Jan-72
Jan-73
Jan-74
Jan-75
Jan-76
Jan-77
Jan-78
Jan-79
Generating Unit In-Service Date
Figure 4-1. Evolution of Wet FGD Technology: The First Decade
In summary, compared to the status of FGD technology at the time of federal legislation
mandating use, CCUS at present is characterized by inadequate experience, affecting cost and
reliability. Consequently, CCUS experience is inadequate to base federal regulation for CO2
removal at the scope and timescale as proposed.
4.3 FEED Study Capital Cost
EPA, after lauding FEED studies to justify CCUS as BSER, ignores FEED results when seeking
a realistic capital cost for use in their analysis of avoided CO2 cost ($/tonne). FEED studies
provide a better estimate of CCUS capital cost then EPA’s use of a hypothetical “model” plant.
As described in Section 3, FEED studies are the second step of a three-phase process to develop
engineering details for a CCUS design. Even with six FEED results “in-hand”, EPA uses an
S&L “model” to generate CCUS capital cost for a “hypothetical” unit, reporting results in Table
7 of the Steam EGU TSD. Of note are three S&L’s disclaimers in the source document
describing the limits in the use of the model to generate costs.37 These address scope, site
factors, and the lack of a cost “benchmark” – as described as follows:
37
IPM Model – Updates to Cost and Performance for APC Technologies: CO2 Reduction Retrofit Cost
Development Methodology, Final Report, Project 13527-002, March, 2023. Hereafter S&L 2023 CO2
IPM.
14
0042a
Review of EPA’s Projection
of CCUS cost
Scope:
Transportation, storage, and monitoring (TS&M) of the captured CO2 are not included in
the base cost estimates and instead costs can be included as a user input on a $/ton
basis.
Site Factors:
The IPM cost equations do not account for site-specific factors that can significantly
affect costs, such as flue gas volume and temperature, and do not address regional labor
productivity, local workforce characteristics, local unemployment and labor availability,
project complexity, local climate, and working conditions.
Cost “Benchmark” or Validation:
Due to the limited availability of actual as-spent costs for CO2 capture projects, the cost
estimation tool could not be benchmarked against recently executed projects to confirm
how accurately it reflects current market conditions.38
These disclaimers are clear – scope is not complete and terminates with CO2 at the fence line;
site factors are ignored; and results are not validated with experience. Consequently, cost
estimates for CCUS capital and the levelized cost to avoid CO2 ($/tonne) are at-risk. An
alternative approach is to use FEED site specific results and adopt the average capital cost.
4.3.1 Coal-Fired Applications
Figure 4-2 presents CCUS capital cost per net generating capacity after CCUS for the two
demonstrations and the six FEED studies for coal-fired generating units. Capital cost is reported
for Sask Power Boundary Dam 3,39 Sask Power Shand,40 Petra Nova,41 Basin Electric Dry
Fork,42 Minnkota Milton R. Young,43 Enchant Energy San Juan,44 Nebraska Public Power
38
S&L 2023 CO2 IPM at p. 1.
Coryn, Bruce, CCS Business Cases, International CCS Knowledge Center, Aug 16, 2019, Pittsburgh, PA.
40
Giannaris, S. et. al., Implementing a second-generation CCS facility on a coal fired power station –
results of a feasibility study to retrofit SaskPower’s Shand power station with CCS, available at:
https://ccsknowledge.com/pub/Publications/2020May_Implementing_2ndGenCCS_Feasibility_Study_Re
sults_Retrofit_SaskPower_ShandPowerStation_CCS.pdf.
41
Final Scientific/Technical Report, W.A. Parish Post-Combustion CO2 Capture and
Sequestration Demonstration Project, DOE Award Number DE-FE0003311, Petra Nova Parish Holdings
LLC, March 31, 2020, Report DOE-PNPH-03311. Hereafter Petra Nova 2020 Final Report.
42
Commercial-Scale Front-End Engineering Design Study for MTR’s Membrane CO2 Capture Process,
Final Technical Report, November 10, 2022. Hereafter 2022 MTR FEED Report.
43
Project Tundra: Postcombustion Carbon Capture on the Milton R. Young Station in North Dakota,
NRECA Update, October 2022.
44
Crane, C., Large-Scale Commercial Carbon Capture Retrofit of the San Juan Generating Station,
Overall Feed Package Report for DOE Cooperative Agreement DE-FE0031843, September 30, 2022.
39
15
0043a
Review of EPA’s Projection
of CCUS cost
District Gerald Gentleman,45 and Prairie State.46 Figure 4-2 also reports capital cost for one of
the hypothetical unit evaluated by NETL: 640 MW (net) with a 10,000 Btu/kwh gross heat rate.47
CCUS Capital Required, $/kw, net w/CCUS (Variable Yr Basis)
6000
5000
Average Excluding
Boundary Dam 3, NETL
Reference: $3,198/kW)
4000
S&L IPM Model
"Intermediate"
Cost: $2,222/kW)
3000
2000
1000
0
111
217
240
324
450
601
640
700
Boundary
Shand
Petra Nova
Dry Fork
Minnkota
San Juan
NETL
NPPD
735
Prairie State
Dam
Figure 4-2. CCUS Capital Cost as Reported for Coal-Fired Demonstrations, FEED Studies
Figure 4-2 displays the capital cost from one of EPA’s “reference” units (Table 4 of the Steam
EGU TSD) used to calculate levelized cost to avoid CO2 ($/tonne). This calculation, using the
S&L IPM model, is conducted for a 400 MW plant with a 10,000 Btu/kWh heat rate,
approximating the average conditions of generating capacity and heat rate of units in Figure 4-2.
The CCUS capital cost of $2,222/kW (net, with CCUS) for this reference unit is superimposed on the
figure as a reference point for Figure 4-2 results.
45
Carbon Capture Design and Costing: Phase 2 (C3DC2), Final Project Report, Final
Scientific/Technical Report, DOE-FE0031840, March 2023.
46
Full-Scale FEED Study for Retrofitting the Prairie State Generating Station with an 816-MWe Capture
Plant Using Mitsubishi Heavy Industries America Post-Combustion CO2 Capture Technology, August 2,
2022. Hereafter 2022 Prairie State FEED Report.
47
Cost and Performance Baseline for Fossil Energy Plants Volume 1: Bituminous Coal and Natural Gas
to Electricity, DOE/NETL Report 2023-4320, October 14, 2022. Hereafter 2022 Bituminous/NGCC
CCUS Retrofit.
16
0044a
Review of EPA’s Projection
of CCUS cost
Data in Figure 4-2 vary widely by site. Capital cost per net generating capacity after CCUS
determined by the FEED studies for all but two units exceeds the $2,222/kW (net, with CCUS) derived
using the S&L IPM procedure for the reference 400 MW unit. The average capital cost from
these FEED studies and demonstration tests – excluding the highest and lowest values – provides
a more authentic estimate of CCUS capital cost.
Excluding both the highest (Boundary Dam) and lowest (NPPD) costs reported in Figure 4-2, the
average capital cost of units in Figure 4-2 is $3,198/kW (net, with CCUS); a 44% increase to S&L’s
reference unit. These FEED study results, even though not “benchmarked” to actual data, are
transparent and can be reviewed – unlike costs generated by the S&L IPM model, which include
“proprietary data”.48
It is important to recognize capital cost data in Figure 4-2 reflects only CO2 capture,
compression, and preparation for transport from the fence line – but not for transport to the
sequestration or EOR site, injection, and plume monitoring.
Sites requiring minimal pipeline length still incur significant costs for the sequestration step.
Two example sites for which information is available are the Minnkota Power and Petra Nova
projects.
Minnkota Power’s Milton R. Young Station. This site requires only 0.5 mile of pipeline for CO2
transport to the sequestration site. However, additional facilities are required for substations for
CO2 metering and pumps, monitoring for seismic activity, and plume migration. The injection of
CO2 requires four wells drilled – three for injection and one for subsurface monitoring – to as
deep as 10,000 feet. Environmental monitoring instrumentation as required for Underground
Injection Control (UIC) Class VI wells is included to assure successful sequestration, as well as
financial assurance in accordance with the regulatory requirements of UIC Class VI wells. These
ancillary support facilities and provisions are estimated to require an additional $100M – or,
$289/kW (net, after CCUS).
Petra Nova. Section 3.3.2 reports of the $1B for all activities, $600 M was devoted to CO2
capture at the plant site with the remaining $400 million dedicated to, among other needs, the
CO2 transport and upgrade of the West Ranch site. This includes the cost for the 81-mile CO2
pipeline and for upgrading the oilfield wells to accept more CO2 for EOR. As a transparent
accounting of projects costs has not been released, it is not known how much of the $400 M is
dedicated to these activities.
48
S&L 2023 CO2 IPM, page 3. “Cost algorithms developed for the IPM model are based primarily on a
statistical evaluation of cost data available from various industry publications as well as Sargent &
Lundy’s proprietary database and do not take into consideration site-specific cost issues. By necessity, the
cost algorithms were designed to require minimal site-specific information and were based only on a
limited number of inputs such as unit size, gross heat rate, baseline emissions, removal efficiency, fuel
type, and a subjective retrofit factor.”
17
0045a
Review of EPA’s Projection
of CCUS cost
4.3.2 NGCC Applications
Figure 4-3 presents capital cost estimated by FEED studies of NGCC assets that have been
reported in the public domain. These FEED studies address the Panda Sherman,49 Golden
Spread Mustang,50 Daniel 4,51 and Elk Hills52 generating units.
CCUS Capital Required, $/kw, net w/CCUS (Variable
Yr Basis)
2000
1800
Average of Four
FEED Studies:
$1,496/kW
Capital Cost ($/kW)
1600
1400
1200
1000
800
600
400
200
0
Panda/Sherman
Golden Spread/Mustang
Daniel 4
Elk Hills
NETL H-Frame
Figure 4-3. CCUS Capital Cost as Reported for NGCC FEED Studies
49
Panda Sherman 2022 Final Report.
Rochelle, G., Piperazine Advanced Stripper (PZAS™) Front End Engineering Design (FEED) Study,
DE-FE0031844, 2022 Carbon Management Research Project Review, August 17, 2022.
51
Lunsford, L., et. al., Front End Engineering Design of Linde-BASF Advanced Post-Combustion CO2
Capture Technology at a Southern Company Natural Gas-Fired Power Plant, Final Scientific/Technical
Report, per DE FE0031847, September 30, 2022. Hereafter 2022 Daniel FEED Report.
52
Front-End Engineering Design Study for Retrofit Post-Combustion Carbon Capture on a Natural Gas
Combined Cycle Power Plant, Agreement DE-FE0031842, for US DOE/NETL, January 2022. Hereafter 2022
Elk Hills FEED Report.
50
18
0046a
Review of EPA’s Projection
of CCUS cost
Figure 4-3 also includes CCUS capital for retrofit to hypothetical NGCC units, as evaluated by
NETL.53 The NETL study estimates capital cost F-Frame and H-Frame gas turbine designations.
The H-Frame result is shown in Figure 4-3 for CCUS capital cost, reported as $/kW (net, with CCUS).
Capital costs reported in Figure 4-3 vary widely by site, driven by, among other factors, the
steam source for CCUS. For example, CCUS capital cost projected for Panda Sherman
($1,135/$/kW (net, with CCUS) is the lowest as the existing HRSG provides steam CCUS duty – but at
the cost of a capacity penalty. Conversely, the highest capital cost (~$1,700/$/kW (net, with CCUS)) is
estimated for two units (Mustang, Daniel 4) as project scope includes auxiliary boilers to provide
steam, preserving generating capacity.
The average of the four FEED studies – albeit representing different concepts to provide CCUS
steam –is $1,496/$/kW (net, with CCUS). This value represents a 20% premium to the cost developed
by NETL.
4.4 Inadequate Basis for Levelized $/Tonne Calculation
EPA employs different methodologies to calculate the levelized cost to avoid CO2 ($/tonne),
including the impact of the IRA, for coal-fired and NGCC generating units. For coal, EPA’s
calculations are recorded in the docket54 but NGCC calculations are inadequately explained or
referenced.
EPA’s calculation methodology is reviewed in this section to document shortcomings. However,
as stated previously, CCUS is not BSER and cost are not confidently defined; thus, EPA’s
calculations are speculative and do not reflect present state-of-art in the proposed rulemaking
docket.
4.4.1 Coal-fired Application
EPA calculations presented in Table 8 of the Steam EGU TSD, which defined levelized cost per
(short) ton including the benefits of the IRA, are invalid for numerous reasons. First, as noted in
Section 4.3.1., the capital cost used by EPA for this calculation is derived from the S&L IPM
model, for “hypothetical” sites. As noted in Section 4.3.1, this source does not provide capital
cost “referenced” to a specific site, nor based on fully transparent data. The example 400 MW
unit with a 10,000 Btu/kWh heat rate is assigned a cost of $2,222/$/kW (net, with CCUS) 31% less
than capital from FEED studies ($3,198/$/kW (net, with CCUS)).
Second, calculations are based on the optimistic premise that the CCUS process will operate at
100% availability, thus always be available to accrue tax benefits and defray operating cost. As
the bulk of CCUS costs are capital, incurred whether the unit is operating or not, periods of
53
Cost and Performance of Retrofitting NGCC Units for Carbon Capture – Revision 3, DOE/NETL2023/3848, May 31, 2023. Hereafter 2023 NGCC CCUS Retrofit.
54
EPA-HQ-OAR-2023_0072-0061_attachment_3.
19
0047a
Review of EPA’s Projection
of CCUS cost
restricted duty will limit CO2 delivered and tax benefits. A compromise in availability directly
affects the calculated cost to avoid CO2.
Table 4-1 compares the levelized cost per tonne ($/tonne) for EPA’s optimistic case, and two
sensitivity cases that explore the role of CCUS capital cost and process availability.55 Table 4-1
presents EPA’s results as calculated using Tables 8 and 9 Steam EGU assumptions, the
“intermediate” capital cost ($2,222/kW (net, with CCUS), and perfect availability (100%). The costs
are presented for 50% and 70% capacity factor, and include the benefit of the IRA.56 Also shown
are results to sensitivity analysis.
Table 4-1. Sensitivity Results: Role of Capital Cost, CCUS Reliability of Projected CO2 $/tonne
Capacity
Factor (%)
50
50
70
70
EPA Assumption
Capital Cost
CCS
($/kW)
Reliability
2,222
100%
2,222
90
2,222
100
2,222
90
$/Tonne
15
23
-9
-2
FEED Study Average
Capital Cost
CCS
$/Tonne
($/kW)
Reliability
3,198
100
49
3,198
90
53
3,198
100
15
3,198
90
23
The sensitivity of the levelized cost (including IRA benefits) to avoided CO2 ($/tonne) to
changes in CCUS capital and reliability are described as follows:
EPA Capital, Compromised CCUS Availability. This case retains EPA’s optimistic capital cost
of $2,222/kW (net, with CCUS), but recognizes that – as witnessed at Sask Power and Petra Nova CCS availability is typically less than 100%. Results for the two capacity factors are as follows:
•
•
Perfect (100%) Availability. Estimated $/tonne cost is reported as $15 at 50% and -$9 at
70% capacity factor.
Compromised (90%) Availability. Estimated $/tonne costs elevates to $23 at 50% and $2 at 70% capacity factor.
FEED Study Capital, Compromised CCUS Availability. Applying the average of FEED study
capital of $3,198/kW (net, with CCUS) for 100% and 90% CCUS reliability derives the following:
•
Perfect (100%) Availability. Estimated $/tonne costs elevates to $49 for at 50% and $15
at 70% capacity factor.
55
It should be noted the author could not corroborate why Table 8 of the Steam EGU TSD specifies the
variable O&M cost used in the calculation is $5/MWh, compared to $23/MWh reported by the S&L IPM
source document for what appears to be comparable conditions. For the purpose of this report,
calculations adopt EPA’s $5/MWh to assure a valid comparison. However, the difference is noted and
should be further explored.
56
The “negative” costs presented in Table 4-1 for two cases reflect EPA’s projection that CO2 removal
and sequestration will comprise a profitable venture.
20
0048a
Review of EPA’s Projection
of CCUS cost
•
Compromised (90%) Availability. Estimated $/tonne costs elevates to $53 for at 50% and
$23 at 70% capacity factor.
It should be noted that –without the IRA subsidy – the cost to avoid CO2 per tonne for some
cases is a factor of 10 higher compared to 100% CCUS reliability. For capital cost of $3,198/kW
(net, with CCUS), the levelized cost to avoid CO2 at 50% capacity factor is $127 and at 70% capacity
factor is $93.
4.4.2 NGCC Application
As noted for coal-fired duty, CCUS for NGCC duty is not BSER. Both S&L and Bechtel have
opined there is negligible experience with CCUS on NGCC conditions. EPA project CCUS
capital cost for NGCC using an unconventional metric that biases costs low and extrapolate costs
to a wide range of applications using both NGCC and coal-derived basis. These results are
flawed, as described as follows.
Capital Cost. EPA projects CCUS capital cost using an incorrect metric. Table 7 of the
Combustion Turbine TSD reports capital, fixed O&M, and variable O&M costs for hypothetical
NGCC units employing the F-Frame and H-Frame technologies, as derived by NETL for
“greenfield” application. Table 7 presents capital cost per net generating capacity (a) replicated
from the NETL study57 and (b) inferred by EPA.
The implied capital for CCUS depends on whether NETL’s “conventional” method is chosen, or
EPA’s inexplicable variant. NETL’s conventional method – taking the difference in capital cost
with and without CCUS – implies a capital cost of $1,199/kW (net, with CCUS) for F-Frame and of
$1,055/kW (net, with CCUS) for the H-Frame applications
EPA inexplicably changes the capital cost metric. The capital cost EPA attributes to CCUS in
Table 7 –$949/kW for the F-Frame and $823/kW for the H-Frame – is lower than inferred from
NETL’s methodology, as EPA normalizes the inferred CCUS cost by net generating capacity
prior to CCUS retrofit.58 This approach is flawed as it does not account for 33 MW of net power
consumed due to the CCUS process.
Extrapolation to Different Applications. EPA’s Combustion Turbine GHG TSD employs a series
of extrapolations to infer CCUS capital, fixed operating, and variable operating cost for a variety
of combustion turbine applications.
EPA (a) misuses the power law relationships describing the change in equipment cost with
generating capacity, and (b) fails to recognize the difference in CCUS process conditions
57
2022 Bituminous/NGCC CCUS Retrofit. Exhibit 9-5 at 710.
Personal Communication, Lisa Thompson to Liz Williamson, July 25, 2023. The $949/kW cost in
Table 7 is calculated by dividing the absolute difference in the costs of the combined cycle EGU with CCS
and without CCS divided by the net output of the combined cycle EGU without CCS. In this case, 688
million divided by 727,000 kW (rounded).
58
21
0049a
Review of EPA’s Projection
of CCUS cost
between coal-fired vs NGCC duty. As a consequence, EPA projects CCUS cost for NGCC duty
(3-4% CO2) based on coal-fired duty (with 12% CO2). Notably, a 2013 NETL report59 cautions
extrapolations such as these, which EPA follows to produce Figures 1-5 in the Combustion
Turbine TSD.
In perspective, these EPA cost results are not of consequence as CCUS is not a demonstrated
technology on NGCC (or coal-fired application), and a basis for cost extrapolations does not
exist. The shortcomings in EPA’s methodology are further discussed in Appendix A for
reference.
In Summary:
•
EPA estimates of CCUS capital cost for coal applications in Tables 6 and 7 of the Steam
EGU TSD are low. The real-world source is the average capital cost derived from the two
industrial demonstrations and FEED studies, eliminating the high (Boundary Dam 3) and
lowest (NETL) cost units. These real-world projects define a cost of $3,198, a 43%
premium to that generated by the IPM model. Revised estimates of $/tonne incurred –
using FEED-study capital cost and accounting for a 10% compromise in CCS reliability increases cost calculated for 50% capacity factor from $23 to $53/tonne with the IRA
credit, and for 70% capacity factor from $2 to $23/tonne if CCUS works as planned for at
least 12 years.
•
EPA estimates of CCUS capital cost for NGCC application presented in the Combustion
Turbine GHG Mitigation TSD are not transparent. EPA infers CCUS capital from a
NGCC CCUS retrofit study issued May 28, 2023, in lieu of the more real-world approach
of averaging cost from the four FEED studies. This latter approach derives capital cost
exceeding that of the NETL-derived hypothetical site by 20%. Most notably, there are no
applications of CCUS on NGCC units – thus no sources to verify the design from which
cost is derived. Two EPA contractors agree. Specifically, both (a) S&L in reporting the
projected CCUS schedule and IPM model and (b) Bechtel in the FEED study for
Panda/Sherman both state limited experience with CCUS on NGCC brings uncertainties,
which compromise the authenticity of any cost estimate.
59
Quality Guidelines for Energy System Studies: Capital Cost Scaling Methodology, DOE/NETL341/013113, January 2013. Hereafter 2013 Scaling Quality Guidelines.
22
0050a
CO2 Pipeline Permitting Issues
5 CO2 Pipeline Permitting Issues
Broad CCUS deployment will require a significant increase in CO2 pipeline capacity. Securing
new pipelines requires design, permitting, and construction tasks – all within a time frame that
will not delay the entire project. Section 5 presents examples of ongoing permitting conflicts,
demonstrating how delays can be incurred. The takeaway from this discussion is used in the
critique of the CCUS implementation schedule presented in Section 6.
5.1 Background
Deploying CCUS to numerous generating units – such as the 39 units EPA estimates to deploy
per the 2023 Integrated Baseline Analysis - requires expanding CO2 pipelines capability. One
limiting step to CCUS deployment is acquiring the necessary right-of-way for pipelines to
transport the CO2. EPA in their projected CCUS schedule estimate 130 weeks to be required for
permitting a pipeline. The Global CCS Institute assumes that in acquiring pipeline access during
their proposed almost 9-year schedule “…. there is no significant community opposition.” 60
A key factor determinate in the schedule is the pipeline length to access either EOR or terrestrial
sequestration. Each additional mile of pipeline requires additional owners’ land to access and
acquire right-of-way. Pipeline permitting issues are addressed following a brief discussion of
pipeline length.
5.1.1 Pipeline Length
The length of the pipeline to transport CO2 from candidate CCUS sites can vary by an order of
magnitude. This range is evidenced by several units that have completed CCUS FEED studies.
The CO2 pipeline length for projects located adjacent to the generating site – such as for Project
Tundra at the coal-fired Dry Fork station, and the Elk Hills NGCC application – are less than a
few miles. Conversely, and as shown in Figure 5-1, the pipeline length necessary to transport
CO2 to the ECO2S Regional Storage Complex from Mississippi Power’s Daniel Unit 4 is 180
miles and from Plant Miller 150 miles.61 Although it appears desirable to rely on CCUS
installations on units located at or adjacent to a disposition site, such a strategy is unrealistic as
host units may not have favorable characteristics (generating capacity, capacity factor, remaining
lifetime).
60
CCS Institute report 20-22; p. 48.
Riestenberg, D. et. al., Establishing an Early Carbon Dioxide Storage Complex in Kemper County, MI:
Project EICO2S, 2020 DOE/NETL Integrated Review Webinar, August 17-19, 2020. Hereafter 2020 Kemper
County Storage Complex.
61
23
0051a
CO2 Pipeline Permitting Issues
Figure 5-1. Candidate CO2 Pipeline Routing, Length: Plants Daniel and Miller
Both the DOE and EPA adopt a typical pipeline length to be 100 km – 62 miles – for which there
is no technical basis; EPA concedes this assumption as a means for “standardization”.62 The
DOE applies this “default” 100 km pipeline length in their cost evaluation for “hypothetical”
plant. EPA states “…. there are 43 States containing areas within 100 km from currently
assessed onshore or offshore storage resources in deep saline formations, unmineable coal seams,
and depleted oil and gas reservoirs”;63 this observation is inadequate to justify the 100-km length
as a default.
Pipeline length will be driven by finding adequate storage volume to accept the CO2 quantity
from a large generating unit; it is unlikely the required storage will be located at the nearest
boundary of any terrestrial basin. The NETL Atlas64 - developed to provide “high-level”
assessment and not a detailed assay of disposition sites - reveals significant heterogeneity of
features that affect CO2 injection rate and storage. The quantity of CO2 to be stored for a coal62
88 Fed. Reg at 33,297, n 333.
Ibid; 33,298.
64
NETL Carbon Storage Atlas; Fifth Edition, DOE Office of Fossil Energy, August 2015. Hereafter 2015
DOE/NETL Storage Atlas.
63
24
0052a
CO2 Pipeline Permitting Issues
fired or NGCC unit of generating capacity large enough for CCUS to be feasible (i.e., 400 MW
or more) is far greater than demonstrated at all but a few sequestration sites permitted to date.
The Global CCS Institute reports 22 projects either in operation or construction for 2024 or 2025
duty with only two sequestering 5 or more million tonnes of CO2 per year (Mt/a).65
In summary, EPA’s assumption of a 100-km average pipeline length to access an acceptable
reservoir for power generation units is not substantiated.
Section 7 presents a graphic depicting arrangement of the 39 units projected by EPA to adopt
CCUS, showing the “footprint” required for pipelines of 100 and 200 km.
5.1.2 Pipeline Projects: Select Description
The Midwest is the nexus for CO2 pipeline permitting. Several entities are well into the process
of developing pipelines to acquire CO2 from ethanol facilities. The major actors are
Summit/Midwest Carbon Solutions, Navigator, and Wolf Carbon. Key features of each project
are summarized as follows:
•
Navigator66 proposes 900-mile pipeline bisecting Iowa from northwest to southeast and
transporting CO2 to Illinois. (~$3.2B). A total of 1,300 miles via South Dakota,
Nebraska, Minnesota, in addition to Iowa, is proposed. The permit application was filed
in July 2022.
•
Wolf Carbon67 propose 280 miles of pipeline to transport CO2 from ADM ethanol
producing facilities in eastern Iowa to Decatur, IL for terrestrial sequestration.
•
Summit Carbon68 will build 700 miles of pipeline in western and northern Iowa to
transport CO2 to North Dakota, for existing EOR application. In Iowa alone, the
proposed pipeline will cross 30 counties.69
These entities are pursuing pipeline permits in several states: Iowa, Minnesota, North Dakota,
Nebraska, and South Dakota. The permitting requirements vary significantly by state– Iowa
presents perhaps the most structured “steps”, and Nebraska the least. The lack of structured steps
currently in Nebraska does not imply permitting requirements are less strict than Iowa; but that
Nebraska’s process for permitting CO2 pipelines is evolving.
65
Global Status of CCS 2022, issued by the Global CCS Institute. Section 6.2. Available at
https://www.globalccsinstitute.com/resources/global-status-of-ccs-2022/.
66
https://heartlandgre
This text is long and has been trimmed here. Open the source document for the complete record.
This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.