TECHNICAL MEMORANDUM
TO:
DATE:
SUBJECT:
Docket for Rulemaking, "Revised Cross-State Air Pollution Rule (CSAPR) Update for the 2008
Ozone NAAQS" (EPA-HQ-OAR-2020-0272)
September 1, 2020
Assessing Non-EGU Emission Reduction Potential
Introduction
Because there are many types of non-EGU emissions sources or units that emit NOx and many control
technologies or combinations of control technologies for these units, there are many approaches to assessing
emission reduction potential from non-EGU emissions sources. The EPA completed an assessment of emission
reduction potential from these sources on a compressed schedule, and this memorandum presents one approach.
The remainder of this memorandum summarizes this approach to assessing non-EGU emission reduction
potential and the related air quality impacts associated with the estimated reductions. The memorandum includes
the following sections:
Model and Methodology Used to Assess Non-EGU Emission Reduction Potential
Background for Determining Source Size/Threshold for Non-EGU Emissions Sources
Air Quality Impacts from Potential Non-EGU Emissions Reductions
Further Verifying and Refining Estimated Non-EGU NOx Emissions Reduction Potential
Detailed Verification and Review of Controls on Non-EGU Sources in Four States
Conclusions of Verification and Review of Controls on Non-EGU Sources in Four States and Potential
Emissions Reductions
Caveats and Limitations of the Cost Analysis
Control Installation Timing
Request for Comment and Additional Information
Model and Methodology Used to Assess Non-EGU Emission Reduction Potential
For this assessment the EPA used the Control Strategy Tool (CoST) with the maximum emission reduction
algorithm1-2-3, the Control Measures Database (CMDb)4, and the 2023 emissions projections based off of the 2016
NEIvl.5 We used the maximum emission reduction algorithm to estimate the largest quantity of potential
emissions reductions from each emissions source or unit that might impact downwind receptors. CoST also
includes a least cost algorithm that works to identify the set of controls that achieves a given percent reduction or
target emissions reduction at the least cost. If that target emission reduction can't be achieved, then the resulting
strategy will be, by definition, the maximum emissions reduction strategy. That is, the primary objective of the
strategy will be focused on getting emissions reductions and not on lowering costs.
1 Further information on CoST, including a peer review of the tool, can be found at the following link:
https://www.epa.gov/economic-and-cost-analvsis-air-pollution-regulations/cost-analvsis-modelstools-air-pollution.
2 We made a few minor changes to the CoST tool that are not reflected in this assessment. These changes could result in less
than 30 additional tpy of potential emissions reductions and ~$2 million less in total costs.
3 The maximum emission reduction algorithm assigns to each source the single measure (if a measure is available for the
source) that provides the maximum reduction to the target pollutant, regardless of cost. For more information, see the CoST
User's Guide available at the following link:
https://www.cmascenter.org/cost/documentation/3.5/CoST%20User's%20Guide/.
4 The CMDb is available at the following link: https://www.epa.gov/economic-and-cost-analvsis-air-pollution-
regulations/cost-analvsis-modelstools-air-pollution.
5We used the 2023 inventory files with/7? in filename.
1
September 1, 2020
-------
For 2023, we summarized emissions reductions and average annual cost per ton for the 12 states identified from
the 2023 air quality modeling and linked to downwind receptors.6The cost per ton values are annual costs and the
estimated reductions are annual emissions reductions. In addition, in the assessment CoST applied controls to
emissions units with a 150 tons per year (tpy) or more pre-control NOx emissions threshold (see section below on
Background for Determining Source Size for Non-EGU Emissions Sources for options on NOx emissions thresholds).
The results of the CoST run are summarized in an Excel workbook titled CoST Control Strategy - Max Reduction
$10k 150 tpy cutoff 12 States Updated Modeling - No Replace - 07-23-2020.
The 12 states in this assessment are the 12 states EPA proposes to find linked to a downwind receptor in 2021 in
this proposed action: IL, IN, KY, LA, MD, Ml, NY, NJ, OH, PA, VA, and WV.
States across the U.S. reported NOx emissions from approximately 81,000 non-EGU facilities with point sources.
Of these, states reported control information for facilities with one or more controls for approximately 17,000
non-EGU facilities, or 21 percent of these facilities.7 As such, this assessment of emission reduction potential from
non-EGU emissions sources reflects a large degree of uncertainty because information about existing controls on
emissions sources is missing for some states and incomplete for some sources.8 As an example, Table 1 below
includes emissions totals, uncontrolled emissions, and percent of uncontrolled emissions using information from
the 2017 NEI.
Table 1. For Facilities w/>150 tpy of Emissions in the 2017 NEI - By State,
Total NOx Emissions and Uncontrolled NOx Emissions (ANNUAL tpy)
Total
Uncontrolled
Percent of
State
Emissions
NOx Emissions
Emissions
(ANNUAL)
(ANNUAL)
Uncontrolled
IL
17,655
16,773
95%
IN
32,926
31,567
96%
KY
19,121
16,445
86%
LA
91,952
87,295
95%
MD
6,354
2,339
37%
Ml
35,399
34,459
97%
NJ
3,753
2,261
60%
NY
12,418
11,065
89%
6 In projecting emissions from 2016 to 2023, a percent emissions reduction can be applied to certain emissions units or
sources without knowledge of specific controls for those units or sources - these reductions are labeled as being from
unknown measures. Some of the units or sources included in this assessment had reductions estimated from unknown
measures. In some cases, CoST removed those unknown measures and applied controls to some of those units or sources,
resulting in approximately 20 thousand tons of emissions reductions estimated from the CoST-applied controls. Because CoST
didn't know what the unknown measures were, CoST might be applying controls that aren't appropriate. In addition, in some
cases CoST didn't have a control, so it didn't remove those unknown measures and apply controls.
7 This summary was based on a query of the 2017 NEI.
8 As noted, control information in the NEI is not consistently provided, but there are two columns that contain control
information. The Control IDs has a number associated with a control device, and the % Reduction has the control efficiency.
Either of these columns may be populated, or both, or none. In cases where only the Control IDs column is populated and we
don't know what the control efficiency is, CoST treats the source as uncontrolled and applies a replacement control. We are
likely overestimating potential emissions reductions in these cases. For the 12 states in this assessment there are 488
possible emissions sources to control with pre-control emissions >150 tpy. Of these, 130 have something in the Control IDs
column (which means CoST may be inappropriately applying a control), 129 have something in the % Reduction column, and
only 28 have both.
2
September 1, 2020
-------
Total
Uncontrolled
Percent of
State
Emissions
NOx Emissions
Emissions
(ANNUAL)
(ANNUAL)
Uncontrolled
OH
35,186
33,891
96%
PA
31,680
30,437
96%
VA
19,394
14,317
74%
WV
11,507
11,255
98%
From the CoST run, Table 2 below summarizes potential emissions reductions by industry sector and the range of
annual cost per ton estimates across units to which CoST applied controls in each industry sector. This summary
can be found in a worksheet titled Control Summary-by NAICS (2) in an Excel workbook titled Control Summary -
Max Reduction $10k 150 tpy cutoff 12 States Updated Modeling - No Replace - 05-18-2020.
Table 2. Annual NOx Emission Reduction Potential and Cost Per Ton Ranges by Industry Sector in 2023 for
Twelve States
NOx Emission Reduction Annual Cost/Ton Range
NAICS Title
Potential (ANNUAL tpy) (2016$)
Chemical Manufacturing
11,577
$914- $9,703
Nonmetallic Mineral Product Manufacturing
19,092
$64 - $4,204
Petroleum and Coal Products Manufacturing
4,363
$2,028 - $8,911
Pipeline Transportation
32,593
$721 - $8,333
Paper Manufacturing
2,058
$3,796- $8,911
Other
1,846
$212 - $7,963
Utilities
392
$1,279 - $6,046
Primary Metal Manufacturing
6,392
$1,395 - $9,495
Oil and Gas Extraction
2,4S4
$634 - S5.6S3
The EPA categorized the CoST results for the control technologies that comprise approximately 92 percent of the
total estimated potential emissions reductions from the non-EGU sources in the 2023 projected inventory with
150 tpy or more of NOx emissions in the 12 linked states; the technologies and related emissions sources include:
a. Layered combustion (lean burn IC engines - natural gas),
b. NSCR or layered combustion (industrial natural gas IC engines, SCCs with technology not
specified),
c. SCR (glass manufacturing - container, flat & pressed, ICI boilers, IC engines (oil-fired and natural
gas)),
d. SNCR (cement manufacturing - dry and wet kilns, municipal waste combustors), and
e. Ultra-low NOx burner and SCR (ICI boilers).
The EPA incrementally included additional details in these summaries, including:
a. Emissions source group (ICI boilers, IC engines, cement kilns, glass furnaces),
b. State, and
c. Industry sector (cement/glass manufacturing, paper manufacturing, pipeline transportation).
In addition, we calculated a weighted average cost per ton for each technology, plotted the weighted average
costs, and observed a clear breakpoint in the curve at $2,000 per ton.9 This identified two tranches, or buckets, of
9 By technology, the Agency calculated a weighted average cost per ton so that some of the outlier cost per ton values did not
disproportionately impact the "average" value used to plot the curve.
3
September 1, 2020
-------
potential emissions reductions (see Figure 1 below).10 The summaries discussed above and the figure below are
also available in the Excel workbook titled Control Summary - Max Reduction $10k 150 tpy cutoff 12 States
Updated Modeling - No Replace - 05-18-2020.
Figure 1. Cumulative NOx Emission Reduction Potential (annual tons) by Weighted Average Cost Per Ton
(annual cost per ton) for Control Technologies in 2023
ฃ
O
7,000
6,000
5,000
$6,631
Layered
Combustion
o
" vv 4,000
gp id
< o
-a 3,000
op
'aj
2,000
1,000
0
$1,917
$5,675
Ultra-low NOx
Burner and SCR
SNCR
NSCRor
Layered
Combustion
10,000 20,000
SCR
30,000 40,000 50,000
Cumulative NOx Reductions
(tons)
60,000 70,000 80,000
Dotted vertical line separates the two tranches.
For the technologies above, we then:
a. Within each technology, further organized by source group, and
b. Looked closer at cost per ton within these technology/source group "bins".
These summaries are available in the Excel workbook with the CoST run results titled CoST Control Strategy - Max
Reduction $10k 150 tpy cutoff 12 States Updated Modeling - No Replace - 07-23-2020.
The first tranche of potential emissions reductions had a weighted average cost of approximately $2,000 per ton
and a cost range from ~$64 per ton - ~$5,700 per ton and included the following technology/source groupings11:
a. SCR - glass manufacturing - container, flat & pressed, IC engines, oil-fired and natural gas (in
pipeline transportation and oil & gas extraction industry sectors), and
b. SNCR - cement manufacturing - dry and wet, municipal waste combustor.
See Table 3 for details - note that the potential emissions reductions are annual tons not ozone season tons.
Additional details on this first tranche, including the potential emissions reductions and number of emissions units
by state are shown in Table 4. To analyze potential emissions reductions in step 3 of the 4-step framework, we
determined that the potential emissions reductions in tranche 1 are potentially relatively cost-effective because
10 This assessment assumes annual cost per ton values. To consider whether the tranches would change using ozone season
cost per ton values, we divided total annual cost by ozone season tons. The technology/source groupings stay the same, and
the ozone season cost per ton values are higher.
11 For the emissions unit estimated to generate emissions reductions at $64 per ton, the emissions and cost estimates were
incorrect. The 2023 projected emissions for the unit were significantly overestimated as a result of a growth factor the EPA
received for these emissions from a multi-jurisdictional partner organization. Further, the equation used to estimate the cost
was mis-specified in CoST, and the true cost is likely on the order of $800 per ton. Changes to these underlying factors will
likely guide an updated assessment for a final rulemaking.
4
September 1, 2020
-------
the $2,000 cost per ton for non-EGU emissions reductions is similar to the control stringency for EGUs
represented by $1,600 per ton (see section below on Further Verifying and Refining Estimated Non-EGU NOx
Emissions Reduction Potential for additional discussion).
Table 3. Annual NOx Emission Reduction Potential and Annual Cost Per Ton Range by Technology and
Source Group for Tranche 1 in 2023
Technology
SCR
SCR
SCR
SCR
SNCR
SNCR
Source Group
Ammonia - NG-Fired Reformers
Glass Manufacturing - Container,
cJat & Pressed*
!C Engines - Natural Gas, Oi!
iron & Steel - In-Process
Combustion - Bituminous Coal
Cement Manufacturing - Dry and
Wet
Municipal Waste Combustors
Total
NOx Emission Reduction
Potential (ANNUAL tpy)
Annual Cost/Ton
Range (2016$/ton)
2,113
15,570
8,843
154
3,711
145
30,537
S3,300
$64 - $4,200
$1,200 - $5,700
S4,200
SI,300' - $2,000
$1,900
""Installing controls on glass furnaces typ
-------
Table 4. Annual NOx Emission Reduction Potential by Technology, State, and Source Group for Tranche
1 in 2023
NOx Emission Reduction
State Number of Units Potential (ANNUAL tpy)
Technology Source Group
Ammonia - NG-Fired Reformers
SCR
Louisiana
5
Glass Manufacturing - Container, Flat & Pressed
2.113
SCR
Illinois
4
1,113
SCR
Indiana
3
652
SCR
Louisiana
3
535
SCR
New York
3
1,162
SCR
Ohio
3
973
SCR
Pennsylvania
6
10,804
SCR
Virginia
2
331
IC Engines - Natural Gas, Oil
SCR
Illinois
1
274
SCR
Indiana
5
2,024
SCR
Louisiana
7
3,860
SCR
Michigan
5
1,514
SCR
Ohio
1
131
SCR
Virginia
2
1,040
Iran & Steel - in-Process Combustion - Bituminous Coal
SCR
Indiana
1
Cement Manufacturing - Dry and Wet
154
SNCR
Indiana
8
2,236
SNCR
Maryland
1
149
SNCR
Michigan
13
1,326
Municipal Waste Combustors
SNCR Maryland I 145';
The second tranche of potential emissions reductions had a weighted average cost range from approximately
$5,000 per ton to $6,600 per ton and a cost range from ~$1,400 per ton - ~$9,700+ per ton and primarily included
the following technology/source groupings:
a. Layered Combustion - lean burn IC engines - natural gas (in pipeline transportation and oil & gas
extraction industry sectors),
b. NSCR or Layered Combustion - industrial natural gas IC engines, SCCs with technology not
specified (in pipeline transportation and oil & gas extraction industry sectors), and
c. Ultra-low NOx burner and SCR - ICI boilers (in paper manufacturing, petroleum and coal products
manufacturing, chemical manufacturing, and primary metal manufacturing industry sectors).
See Table 5 for details - note that the potential emissions reductions are annual tons, not ozone season tons.
Additional details on this second tranche, including the potential emissions reductions and number of emissions
units by state are shown in Table 6. To analyze potential emissions reductions in step 3 of the 4-step framework,
we made no determination as to whether the potential emissions reductions in tranche 2 are cost-effective, and
6
September 1, 2020
-------
we did not look further in tranche 2 because we assumed the $1,600 per ton cost threshold for reductions from
EGU sources was an equivalent cost threshold for comparison. The underlying details and summary Tables 3
through 6 are available in the Excel workbook titled CoST Control Strategy - Max Reduction $10k 150 tpy cutoff 12
States Updated Modeling - No Replace - 07-23-2020.
Table 5. Annual NOx Emission Reduction Potential and Annual Cost Per Ton Range by Technology and
Source Group for Tranche 2 in 2023
Technology
Layered Combustion
Source Group
Lean Burn ICE - NG
Industrial NG ICE, SCCs
with technology not
specified
NOx Emission
Reduction Potential Annual Cost/Ton
(ANNUALtpy) Range (2016$/ton)
10.963 $5,500' - $6,600'
[NiSCR of Layered
Combustion
Ultra-low \Ox Burner
and SCR
ICI Boilers
13,176
17,341 $1,400 - $9,700*
$6,400 - $8,300
Total
*Weighted average cost/ton is '"$5,100.
7
September 1, 2020
-------
Table 6. Annual NOx Emission Reduction Potential by Technology, State, and Source Group for Tranche
2 in 2023
State
Numberฎ# Units
NOx Emission Reduction
Potential (ANNUALtpy)
Technology
Source Group
Lean Bum ICE - NG
Layered Combustion
Illinois
7
1,444
Layered Combustion
Indiana
10
2,376
Layered Combustion
Kentucky
4
752
Layered Combustion
Louisiana
3
944
Layered Combustion
Michigan
2
307
Layered Combustion
New York
1
324
Layered Combustion
Ohio-
8
1,521
Layered Combustion
West Virginia
14
3,294
Industrial NG ICE, SCCs with technology not specified
NSCR or Layered Combustion
Illinois
10
1,708
NSCR or Layered Combustion
Kentucky
1
417
NSCR or Layered Combustion
NSCR or Layered Combustion
Louisiana
Ohio
31
23
6,731
4,319
JO Boilers
Ultra-tow NOx Burner and SCR
Illinois
2
387
Ultra-tow NOx Burner and SCR
Ultra-low NOx Burner and SCR
Indiana
Kentucky
9
3
3,810
905
Ultra-tow NOx Burner and SCR
Louisiana
25
7,119
Ultra-low NOx Burner and SCR
New York
4
1,343
Ultra-tow NOx Burner and SCR
Ohio
7
2,059
Ultra-low NOx Burner and SCR
Pennsylvania
3
748
Ultra-low NOx Burner and SCR
West Virginia
3
969
Background for Determining Source Size/Threshold for Non-EGU Emissions Sources
In assessments of non-EGU emission reduction potential for previous interstate transport rulemakings, we
assessed units with pre-control NOx emissions > 100 tpy, which is the major source threshold for moderate ozone
nonattainment areas. For this assessment, the EPA included units with pre-control NOx emissions > 150 tpy,
which is an emissions threshold comparable to 25 MW for EGUs used in prior interstate transport rulemakings. To
derive this emissions threshold, we used emissions expected from an average 25 MW EGU unit operating at a
median heat rate, emission rate, and capacity factor for a coal-fired unit. A description of this derivation is below.
The CSAPR trading program is currently restricted to EGU sources greater than 25 MW electric generating capacity
in the regulation. Since non-EGU sources are not all rated in electric generating capacity, we estimated an
equivalent threshold for these sources on an annual NOx emissions basis. We estimated that 150 tons of NOx
emissions per year is a reasonable approximation for a typical 25 MW EGU.
This estimate represents a generic 25 MW EGU and relied on assumptions of three factors: heat rate, capacity
factor, and NOx emissions rate. To develop an estimate for each of these factors, we evaluated EGUs ranging from
25 MW - 30 MW, which represent the smallest EGUs currently included in the CSAPR trading program. This
8
September 1, 2020
-------
sample included nine units from the following six plants (ORIS codes): 50931, 2790, 50611, 50835, 57046,
2935. We excluded one outlier unit with a NOx rate that was nearly three times higher than the next highest NOx
rate. We calculated the median and average heat rate and NOx rate based on the assumptions included in NEEDS
v6 rev: 3-26-2020. We calculated the median and average annual capacity factor based on Air Markets Program
data reported to EPA in 2019. These values are summarized below.
Median
Average
Heat Rate (Btu/kWh)
12,140
12,291
NOx Rate (Ibs/MMBtu)
0.18
0.23
Capacity Factor (%)
61%
61%
The estimated annual emissions from a typical 25 MW unit based on the assumptions above ranges from about
141 annual tons (median values) to 188 annual tons (average values). Given the small sample sizes, we believe the
median values are more representative than average values. Therefore, we estimated that 150 tons per year is a
reasonable approximation of the annual NOx emissions at a typical 25 MW EGU. Since non-EGUs sources are not
universally rated in MW electric generating capacity, we believe that NOx emissions of 150 tons per year is an
equivalent threshold for use in this assessment.
Air Quality Impacts from Potential Non-EGU Emissions Reductions
Tables 7 and 8 below provide estimates of the air quality impacts at the Westport, CT receptor of the potential
non-EGU emissions reductions in linked upwind states. We chose the Westport site for this assessment because it
is likely the only site to remain a receptor during the time period when non-EGU controls could be implemented,
assuming those controls take longer than 18 months to install. The results for Westport, CT are representative of
the impacts for other coastal Connecticut receptors. In Tables 7 and 8 below, the air quality data are provided for
individual upwind states and by industry sector, source category, and technology for all linked upwind states
combined. Tables 7 and 8 (and the tables that follow) include potential emissions reductions in units of ozone
season tons for appropriate comparison to potential EGU emissions reductions.
The estimated air quality impacts of the potential non-EGU emissions reductions are based on multiplying the
estimated emissions reductions by the parts per billion (ppb) per ton values for each linkage.12 The ppb per ton
values were derived from the state-by-state contribution modeling. Since the contribution modeling included
emissions from all anthropogenic sources in each state, rather than just non-EGUs, the ppb per ton values used
for this analysis introduce some degree of uncertainty in the results.
In addition, because the precursor emissions in the New York City portion of New York state are a large portion of
the total state emissions and given the proximity of the coastal CT receptors to New York City, the contributions
from the state of New York in the modeling largely reflect the contribution from emissions within New York City
and adjacent areas of southern New York. As such, the ppb per ton values for New York based on the modeling
are likely to overstate, by a large amount, the ppb per ton values from sources outside of New York City. In this
assessment, the estimated impacts at Westport and other coastal CT receptors of the potential non-EGU
emissions reductions in New York state are likely overstated because the ppb per ton values used in the
calculations are dominated by the contributions from New York City, whereas the potential non-EGU emissions
reductions are from emissions units in the western part of the state. Also note that there were no potential NOx
12 We applied the calibration factor for this receptor that is used in the Air Quality Assessment Tool (AQAT) for calculating the
ozone impacts of EGU emissions reductions. The AQAT is discussed and documented in the Ozone Transport Policy Analysis
TSD for this proposal. Calibration factors are intended to account for the non-linear response of ozone to NOx emissions
reductions.
9
September 1, 2020
-------
emissions reductions from New Jersey because the projected 2023 emissions inventory did not include non-EGU
point sources in New Jersey with pre-control NOx emissions greater than 150 tpy for which CoST had applicable
control measures.
Table 7. Non-EGU Emissions in 2023 and PPB Reductions at Westport, Connecticut
for Individual Linked Upwind States
Linked States
OS NO*
Reductions
PPB
Reduction
Pennsylvania
4,813
0.144
New York
1,179
0.107
Ohio
3,751
0.043
Indiana
4,931
0.037
West Virginia
2,117
0.029
Michgan
1,311
0.013
Illinois
2,053
0.003
Kentucky
364
0.007
Virginia
571
0.006
Maryland
123
0.003
Total
21,764
0.401
Table 8. Non-EGU Emissions in 2023 and PPB Reductions at Westport, Connecticut by
Industry Sector or Source Category, Technology, and Weighted Average Cost Per Ton
from Linked Upwind States13
Secto n Tech nology
05 NOx
Reductions
PPB
Reduction
Approx.
$2,000/ton
Glass Manufacturing: SCR
5,780
0.189
Lea n Bum IC Engs: SCR
2,710
0.025
lrcin& Steel: 5CR
64
0.000
Cement: 5NCR
1,546
0.014
MWC: 5NCR
61
0.001
SubTotal
10,161
0.230
Approx.
$5,000 tD
$6,600/tDn
Lean Burn IC Eng: L Comb
4,040
0.052
Industrial Natural Gas: N5CR/L Comb
2,685
0.027
ICI Boilers: Ultra-low NOx Burner/SCR
3,700
0.092
SubTotal
10,425
0.172
Total
20,586
0.401
Further Verifying and Refining Estimated Non-EGU NOx Emissions Reduction Potential
Because information for existing controls on non-EGU emissions sources is missing in the 2016 base year
inventory for some states and incomplete for some sources, the EPA went through a process to further verify
13 After the initial assessment of non-EGU reduction potential, we further reviewed information related to applying SCR to IC
engines (discussed below). CoST estimated the control cost inappropriately, as it applied a cost equation to a source much
larger than the predictive range of the equation. In such cases, CoST should apply a cost per ton value, which would be above
the $2,000 per ton threshold for cost-effectiveness. As a result, we removed these units from further consideration.
10
September 1, 2020
-------
existing control information and refine the NOx emission reduction potential estimated by CoST, the CMDb, and
the 2023 projected inventory in Tables 3 through 6 above. The steps the EPA took, discussed in more detail below,
include:
Considered the air quality impacts by state and focused on upwind states with the largest estimated
potential air quality impacts from potential non-EGU emission reductions;
Assumed that the potential reductions in tranche 1 were potentially cost-effective because tranche l's
weighted average cost of $2,000 per ton is similar to the control stringency for EGUs represented by
$1,600 per ton;
Looked at potential emissions reductions in tranche 1 that were estimated to cost less than $2,000 per
ton; and
For those potential reductions in tranche 1 that were estimated to cost less than $2,000 per ton, reviewed
online facility permits and industrial trade literature to verify and determine if the estimated emissions
reductions may be actual, achievable emissions reductions.
First, we considered the potential ppb impacts by state in Table 7 and prioritized the verification and refinement
of the NOx emission reduction potential for a subset of the states with the largest estimated potential air quality
impacts. We reviewed potential controls and estimated emissions reductions in Pennsylvania, New York, Ohio,
Indiana, and West Virginia. The EPA identified these states using an estimate of 0.02 ppb as a threshold for air
quality improvement that may be obtained from reductions from non-EGUs in each state. The Agency is not
applying a 0.02 ppb impact threshold as a step in the Step 3 multi-factor test. Rather, this threshold value allowed
the Agency to better target its efforts toward the potentially effective states for non-EGU NOx emissions
reductions.
Next, to continue analyzing potential emissions reductions in step 3 of the 4-step framework, we determined that
the potential reductions in tranche 1 (Table 3 above) were potentially relatively cost-effective because the $2,000
cost per ton cost for reductions from non-EGU sources is similar to the control stringency for EGUs represented by
$1,600 per ton. While we made no determination as to whether the potential emissions reductions in tranche 2
were cost-effective, we did not look further in tranche 2 (Table 5 above) because we assumed the $1,600 per ton
control stringency for proposed reductions from EGU sources was an equivalent cost threshold for comparison.14
Note that the emissions reductions from tranche l15 are in the section of Table 8 with the weighted average cost
of $2,000 per ton (2016$), and the emissions reductions from tranche 2 are in the section with the weighted
average cost range of $5,000 to $6,600 per ton. Tranche 1 includes:
1. SCR:
a.glass manufacturing - container, flat & pressed,
b.lC engines - natural gas, oil (in pipeline transportation and oil & gas extraction industry
sectors), and
2. SNCR:
a.cement manufacturing - dry and wet kilns,
b.municipal waste combustors.
The total estimated potential emissions reductions from non-EGU sources in Pennsylvania, New York, Ohio, and
Indiana in tranche 1 were 7,556 ozone season tons. Note that West Virginia dropped out because as indicated
below CoST estimated control costs for two IC engines inappropriately, and CoST did not apply cost-effective
14 Details on these tranches can be found in the Summary SCR and SNCR and Summary Other Technologies worksheets in the
CoST Control Strategy - Max Reduction $10k 150 tpy cutoff 12 States Updated Modeling - No Replace - 07-23-2020 Excel
workbook.
15 In tranche 1 the cost per ton ranges from ~$64 per ton - ~$5,700 per ton.
11
September 1, 2020
-------
controls to any other emissions units in the state. Below we note exceptions where in tranche 1 CoST applied cost-
effective controls that were not included in the results.
CoST applied controls to two IC engines in West Virginia for additional potential emissions reductions of
341 ozone season tons (in tranche 1 CoST did not apply controls to any other emissions units in the state).
However, CoST estimated the control cost inappropriately, as it applied a cost equation to a source much
larger than the predictive range of the equation. In such cases, CoST should apply a cost per ton value,
which in this instance would be above the $2,000 per ton threshold for cost-effectiveness. As a result, it
was determined that there are no actual controls available at the selected level of cost-effectiveness for
these units. We reviewed the permits for these units - the permit indicates that the units currently do not
have control devices installed but do require periodic tune-ups and performance tests.
CoST applied a control to an IC engine in Indiana for additional potential emissions reductions of 292
ozone season tons. Like the West Virginia controls, the cost of this control was underestimated for a
source of this size, and the cost per ton for this source is above the threshold for cost-effectiveness. We
reviewed the permit for this unit - the permit indicates that the unit currently does not have a control
device installed but does require performance tests and a preventive maintenance plan.
Next, we looked at the potential emissions reductions in tranche 1 that were estimated to cost <$2,000 per ton,
which were 6,346 ozone season tons, or 84 percent of the estimated reductions in tranche 1 in these states; the
remaining 16 percent of estimated reductions, or 1,210 ozone season tons, was above the $2,000 per ton
threshold.
The steps taken to verify and refine the NOx emission reduction potential information were based first on
technology application and related costs (as detailed above in the section on Model and Methodology for
Assessing Non-EGU Emission Reduction Potential), then on a representative sample of states, and then on likely
cost-effective reductions (i.e., reductions < $2,000 per ton) in those states, which led to key industry sectors; we
did not directly select key industry sectors to review for applicability. In the review of the potential controls in
tranche 1 for Pennsylvania, New York, Ohio, and Indiana, we concluded that the likely cost-effective emissions
reductions were from SCR applied to glass furnaces and SNCR applied to cement kilns.16 Please see the additional
discussion on these estimated emissions reductions in the section below titled Conclusions on Verification and
Review of Controls on Non-EGU Sources in Four States and Potential Emissions Reductions.
We did not review the potential controls for emissions sources in tranche 1 for the remaining five states in Table
7. Based on the additional verification and review we conducted for potential controls on emissions sources in
Pennsylvania, New York, Ohio, Indiana, and West Virginia (summarized in the section below titled Detailed
Verification and Review of Controls on Non-EGU Sources in Four States), however, we believe it is reasonable to
conclude that a similarly small quantity of additional NOx emissions reductions could be identified. For the
remainder of the analysis, to be conservative, we assume that the tranche 1 reductions identified by CoST at or
less than $2,000 per ton in these five states are real emissions reductions.17
Detailed Verification and Review of Controls on Non-EGU Sources in Four States
After determining it was appropriate to verify the potential emissions reductions that were estimated to cost
<$2,000 per ton, we took the additional step of verifying and refining the information on potential controls for
16 Note that for non-EGUs not all industry sectors are present in each of the 12 states.
17 As discussed in more detail in the Conclusions of Verification and Review of Controls on Non-EGU Sources in Four States and
Potential Emissions Reductions section, the possible emissions reductions from Pennsylvania, New York, Ohio, Indiana, and
West Virginia were approximately 903 ozone season tons. The assumed emissions reductions from the remaining five states
would be approximately 664 ozone season tons.
12
September 1, 2020
-------
emissions sources in tranche 1 for Pennsylvania, New York, Ohio, and Indiana. Note that West Virginia dropped
out because CoST estimated control costs for two IC engines inappropriately, and CoST did not apply likely cost-
effective controls to any other emissions units in the state. To verify and refine the information, we reviewed
facilities' online Title V permits for likely cost-effective emissions reductions associated with SCR applied to glass
furnaces and SNCR applied to cement kilns, and also reviewed industrial trade literature for these facilities and
their parent companies. These permit and industrial trade literature reviews were completed as of July 31, 2020.
By state, in Tables 9 through 12 below, we include information on 20 emissions units at glass manufacturing and
cement manufacturing facilities including the facility name, NEI Unit ID, type of emissions unit, existing NOx
control, NOx monitoring device, type of fuel used, and related notes. Of the 20 emissions units, 10 units either (i)
have controls and monitors (primarily CEMS) already, (ii) are installing controls and CEMS or consolidating
operations in the next few years as a result of recent consent decrees issued as part of the EPA's New Source
Review Air Enforcement Initiative, (iii) have shut down, or (iv) are planning to shut down by 2023. Based on
information collected through the permit review, we believe the units in categories (i) and (iii) don't present an
opportunity to generate emissions reductions as part of this analysis and should be removed from further
consideration. With respect to categories (ii) and (iv), for purposes of a focused analysis of potential cost-effective
non-EGU emissions reductions, we excluded these units from further consideration.
Reviewing online facility permits does not always resolve outstanding questions. Permits can be 100 pages or
more in length with detailed information about a facility and the units at the facility, and the accuracy and extent
of information can vary by state. Matching NEI information to information in the permit is not always straight
forward. For example, the NEI Unit IDs don't always match the unit ID information in the permit and even more
research or refinement is needed.
13
September 1, 2020
-------
Table 9. Pennsylvania Glass Manufacturing Facilities
Facility Name/NEI
Unit ID18
Ultimate
Parent
Company
Type of
Emissions
Unit
County
2023
Projected
NOx
Emissions
Estimated
Reductions
(OS Tons)
from SCR
Existing
NOx
Control
NOx
Monitoring
Device/
Technique
Fuel Used
by Furnace
Other
Notes
Status of
Estimated
Reductions19
Ardagh Glass Inc/Port
Allegany Pit
(NEI Unit ID
19110913)
Ardagh Group
S.A.
Container
Glass: Melting
Furnace
McKean
152
47.43
LNB + OEAS
CEMS
Natural Gas
194.73
tons annual
emissions
limit
Remove from
consideration
(Already
controlled)
Vitro Flat Glass
LLC/Carlisle
(NEI Unit ID
18725313)20
Vitro, Inc.
Flat Glass:
Melting
Furnace
Cumberland
10,514
3,285.65
No Control
CEMS
Natural
Gas/Oil #2
(permitted
for both
fuels; natural
gas is the
typical fuel
used)
Emissions
limit of
26.75
lb/ton glass
produced
Uncertain -
see notes
below
(NEI
discrepancy)
Vitro Flat Glass
LLC/Carlisle
(NEI Unit ID
18725413)
Vitro, Inc.
Flat Glass:
Melting
Furnace
Cumberland
1,236
386.27
SCR
CEMS
Natural
Gas/Oil #2
(permitted
for both
fuels; natural
gas is the
typical fuel
used)
Emissions
limit of 7.0
lb/ton glass
produced
Remove from
consideration
(Already
controlled)
18 Pennsylvania's online permits are available at the following link: http://www.depreportingservices.state.pa.us/ReportServer/Pages/ReportViewer.aspx7/Air Qualitv/AQ Permit Docs.
19 The category indicated in italics and parentheses is associated with the categories in Table 13 below.
20 The cost per ton and potential emissions reductions for this emissions unit reflect a high degree of uncertainty. The uncertainty comes from the following two sources: (i) discrepancies
between the underlying information for this unit in the 2023 projected inventory and other emissions data, and (ii) the equation in CoST that is used to estimate the emissions reductions
and cost per ton value. In the 2023 projected inventory, the reported pre-control emissions are much larger than what appears in the PA Air Emissions Report
(http://www.depgreenport.state.pa.us/powerbiproxv/powerbi/Public/DEP/AQ/PBI/Air Emissions Report) for this facility and significantly larger than any other glass furnace in this
analysis, and the projected inventory does not show a control on any unit at this facility, even though a review of the permit indicates that one unit does have a control. Lastly, the
equation used to estimate the costs is misspecified and yields artificially low cost per ton estimates for a source this large. The default cost per ton value from CoST for this source is
roughly $800/ton (2016$) and is still within the range of costs in tranche 1.
14
September 1, 2020
-------
Facility Name/NEI
Unit ID18
Ultimate
Parent
Company
Type of
Emissions
Unit
County
2023
Projected
NOx
Emissions
Estimated
Reductions
(OS Tons)
from SCR
Existing
NOx
Control
NOx
Monitoring
Device/
Technique
Fuel Used
by Furnace
Other
Notes
Status of
Estimated
Reductions19
Pittsburgh Glass
Works/Meadville
Works 821,22
(NEI Unit ID
19025613)
Vitro, Inc.
Flat Glass:
Melting
Furnace
Crawford
1,739
543.49
No Control
CEMS
Natural Gas
766.5 tons
annual
emissions
limit
Uncertain -
closed a line
on June 10,
2020
(Shutdown)
Guardian Ind
Corporation/Jefferson
Hills
(NEI Facility ID
2989611)
Allegheny
512
159.93
Facility
closed at
end of
2015.
Remove from
consideration
(Shutdown)
Subtotal
4,422.77
21 Vitro acquired this facility in 2017 - https://www.post-gazette.com/business/career-workplace/2020/04/13/Meadville-Vitro-glass-COVID-19-lavoffs-pennsvlvania/stories/202004130Q94.
22 This facility shut down one of its two production lines effective June 10, 2020. The company stated that it will be too expensive to rebuild the production line.
https://www.glassmagazine.com/news/vitro-shut-down-float-line-automotive-glass
https://www.post-gazette.com/business/career-workplace/2020/04/13/Meadville-Vitro-glass-COVID-19-lavoffs-pennsvlvania/stories/202004130Q94
23 https://www.post-gazette.com/business/pittsburgh-companv-news/2015/06/24/Guardian-lndustries-to-close-Jefferson-Hills-plant-more-than-100-face-lavoffs-
pittsburgh/stories/201506240183
https://www.wtae.com/article/guardian-industries-closing-iefferson-hills-plant-idling-114/7472247
15
September 1, 2020
-------
Table 10. New York Glass Manufacturing Facilities
Facility Name/NEI
Ultimate
Type of
County
2023
Estimated
Existing
NOx
Fuel Used
Other
Status of
Unit ID24
Parent
Emissions
Projected
Reductions
NOx
Monitoring
by Furnace
Notes
Estimated
Company
Unit
NOx
Emissions
(OS Tons)
from SCR
Control
Device/
Technique
Reductions25
Furnace #1
Possible
has an
(Possible
annual
emissions
emissions
reductions)
limit of 1.2
Two
lb NOx/ton
Anchor Glass Container
Corp
(NEI Unit ID 2854113)
Anchor Glass
Container
Corp
Container
Glass: Melting
Furnace
Chemung
450
140.63
furnaces -
Furnace #1 -
SCR;
Furnace #2 -
no control
CEMS
Natural Gas
of glass
produced.
Furnace #2
has an
annual
emissions
limit of 4.5
lb NOx/ton
of glass
produced.
Furnace A
Possible
has an
(Possible
annual
emissions
emissions
reductions)
Two
furnaces -
No controls
indicated
limit of 4.0
Owens Brockway Glass
Container Inc
0-1 Glass, Inc.
Container
Glass: Melting
Cayuga
309
96.69
CEMS
Natural Gas
lb NOx/ton
of glass
(NEI Unit ID 2863113)
Furnace
produced.
Furnace B
has an
annual
emissions
limit of ?
24 New York's online permits are available at the following link: http://www.dec.nv.gov/dardata/boss/afs/issued atv.html.
25 The category indicated in italics and parentheses is associated with the categories in Table 13 below.
16
September 1, 2020
-------
Guardian Geneva Float
Glass Facility
(NEI Unit ID 18725413)
Koch
Industries,
Inc.
Flat Glass:
Melting
Furnace
Ontario
790
246.83
SCR
CEMS
Natural Gas
Annual
emissions
limit of 770
tons
Remove from
consideration
(Already
controlled)
Subtotal
484.15
Table 11. Ohio Glass Manufacturing Facility
Facility Name/NEI
Ultimate
Type of
County
2023
Estimated
Existing
NOx
Fuel Used
Other Notes
Status of
Unit ID27
Parent
Emissions
Projected
Reductions
NOx
Monitoring
by Furnace
Estimated
Company
Unit
NOx
Emissions
(OS Tons)
from SCR
Control
Device/
Technique
Reductions28
Furnace #1
Uncertain -
has annual
recent stack
emissions
test shows
limit of 364.7
emissions well
Two
tons - a
below permit
furnaces -
recent stack
limit
None
Natural Gas
or Oil
(permitted
for both;
natural gas
the typical
fuel used)
test show
(Already
indicated for
emissions at
controlled)
Pilkington North
America Inc.
(NEI Unit ID 55204113)
Flat Glass:
Melting
Furnace
Wood
755
236
furnace #1;
3R
technology29
for furnace
#2
(technology
is
proprietary)
CEMS
the furnace
are 41.64
tons NOx.
Furnace #2
has an annual
emissions
limit of 945
tons -- CEMS
data show
recent
emissions of
792.98 tons.
Subtotal
236
26 This facility is subject to a consent decree with the U.S. requiring that SCR be installed on its furnace to be shutdown with compliance actions to be taken between December 31, 2017
and December 31, 2024. A NOx CEMS is already in place. Consent decree is at https://elr.info/sites/default/files/doi-consent-decrees/united states v. guardian industries corp.pdf.
27 Ohio's online permits are available at the following link: https://www.epa.ohio.gov/dapc/permits/permits.
28 The category indicated in italics and parentheses is associated with the categories in Table 13 below.
29 3R is a NOx control technology that involves combustion modification using excess natural gas to create reducing conditions within a glass furnace in order to reduce NOx emissions. A
BACT permit by Cardinal Glass in Portage, Wl, filed with the Wisconsin Department of Natural Resources (WIDNR) (December 8, 2017) indicates that this technology may lead to long-term
furnace and refractory damage based on their experience with the use of 3R at 3 other plants of theirs in the U.S. The CMDb does not include the 3R process as a NOx control technology.
17
September 1, 2020
-------
Table 12. Indiana Glass Manufacturing and Cement Manufacturing Facilities
Facility Name/NEI
Ultimate
Type of
County
2023
Estimated
Existing
NOx
Fuel Used
Other
Status of
Unit ID30
Parent
Emissions
Projected
Reductions
NOx
Monitoring
by Furnace
Notes
Estimated
Company
Unit
NOx
Emissions
(OS Tons)
from SCR
and SNCR
Control
Device/
Technique
Reductions31
Furnace #2
Possible
has an
(Possible
annual
emissions
emissions
reductions)
limit of
Ardagh Glass Inc.
(NEI Unit ID 65375713)
Ardagh
Group S.A.
Container
Glass: Melting
Furnace
Randolph
312
97.63
No Control
No monitors
Natural Gas
506.9 tons.
This
furnace
may be
emitting
under its
permit
limit.
Furnace #1
Possible
has an
(Possible
annual
emissions
emissions
reductions)
limit of
Ardagh Glass Inc.
(NEI Unit ID 65375813)
Ardagh
Group S.A.
Container
Glass: Melting
Furnace
Randolph
280
87.65
No Control
No monitors
Natural Gas
389.24
tons. This
furnace
may be
emitting
under its
permit
limit.
30 Indiana's online permits are available at the following link: https://www.in.gov/apps/idem/caats/.
31 The category indicated in italics and parentheses is associated with the categories in Table 13 below.
18
September 1, 2020
-------
Facility Name/NEI
Ultimate
Type of
County
2023
Estimated
Existing
NOx
Fuel Used
Other
Status of
Unit ID30
Parent
Emissions
Projected
Reductions
NOx
Monitoring
by Furnace
Notes
Estimated
Company
Unit
NOx
Emissions
(OS Tons)
from SCR
and SNCR
Control
Device/
Technique
Reductions31
Facility-
Possible
wide
(Possible
annual
emissions
emissions
reductions)
Anchor Glass Container
Anchor Glass
Container
limit of 396
Corporation
Container
Glass: Melting
Dearborn
276
86.28
No control
No monitors
Natural Gas
tons, which
(NEI Unit ID 28314513)
Corp.
Furnace
likely
includes
additional
emissions
sources.
NOx
Remove from
control &
consideration
monitoring
(Shutdown)
required
Lehigh Cement
Company
(NEI Unit ID 5813813)
Heidelberg
Cement
Long Kiln
Clark
187
38.89
SNCR
CEMS
Natural Gas
(coal or oil
as backup
fuels)
under
consent
decree
(Essroc).
Plant to
cease
operations
during
2022.
NOx
Remove from
control &
consideration
monitoring
(Shutdown)
Lehigh Cement
Company
(NEI Unit ID 5813313)
Heidelberg
Cement
Preheater Kiln
Clark
394
82.08
SNCR
CEMS
Natural Gas
(coal or oil
as backup
fuels)
required
under
consent
decree
(Essroc).
Plant to
cease
operations
19
September 1, 2020
-------
Facility Name/NEI
Unit ID30
Ultimate
Parent
Company
Type of
Emissions
Unit
County
2023
Projected
NOx
Emissions
Estimated
Reductions
(OS Tons)
from SCR
and SNCR
Existing
NOx
Control
NOx
Monitoring
Device/
Technique
Fuel Used
by Furnace
Other
Notes
Status of
Estimated
Reductions31
during
2022.
Lehigh Cement
Company
(NEI Unit ID 4232613)
Heidelberg
Cement
Preheater Kiln
Lawrence
552
115.09
No control
No monitoring
Natural gas
Plant is
subject to
NOx
requireme
nts in
consent
decree; a
single kiln
will replace
all 3
preheater
kilns in
2023.
Permit
indicates
MKFor
LNBas
controls for
ozone
season, but
no
evidence of
installation.
Remove from
consideration
(Lehigh
Cement - kiln
replacements)
Lehigh Cement
Company
(NEI Unit ID 4232813)
Heidelberg
Cement
Preheater Kiln
Lawrence
495
103.21
No control
No monitoring
Natural gas
Plant is
subject to
NOx
requireme
nts in
consent
decree; a
single kiln
will replace
all 3
Remove from
consideration
(Lehigh
Cement - kiln
replacements)
20
September 1, 2020
-------
Facility Name/NEI
Unit ID30
Ultimate
Parent
Company
Type of
Emissions
Unit
County
2023
Projected
NOx
Emissions
Estimated
Reductions
(OS Tons)
from SCR
and SNCR
Existing
NOx
Control
NOx
Monitoring
Device/
Technique
Fuel Used
by Furnace
Other
Notes
Status of
Estimated
Reductions31
preheater
kilns in
2023.
Permit
indicates
MKFor
LNBas
controls for
ozone
season, but
no
evidence of
installation.
Lehigh Cement
Company
(NEI Unit ID 4233913)
Heidelberg
Cement
Preheater Kiln
Lawrence
711
148.10
No control
No monitoring
Natural gas
Plant is
subject to
NOx
requireme
nts in
consent
decree; a
single kiln
will replace
all 3 kilns
by 2023.
Plant is
subject to
NOx
control by
MKFor
LNBin
ozone
season, but
no
evidence of
installation.
Remove from
consideration
(Lehigh
Cement - kiln
replacements)
21
September 1, 2020
-------
Facility Name/NEI
Unit ID30
Ultimate
Parent
Company
Type of
Emissions
Unit
County
2023
Projected
NOx
Emissions
Estimated
Reductions
(OS Tons)
from SCR
and SNCR
Existing
NOx
Control
NOx
Monitoring
Device/
Technique
Fuel Used
by Furnace
Other
Notes
Status of
Estimated
Reductions31
Lehigh Cement
Company
(NEI Unit ID 65392513)
Heidelberg
Cement
Wet Kiln
Cass
314
65.49
Wl (water
injection)
CEMS
Coal or oil
Kiln #2 has
an
emissions
limit of
4.75 lb
N Ox/ton
clinker
produced
Possible
(Possible
emissions
reductions)
Lehigh Cement
Company
(NEI Unit ID 65392613)
Heidelberg
Cement
Wet Kiln
Cass
242
50.33
SNCR +Wl
CEMS
Coal or oil
Kiln #1
does not
have an
emissions
limit
indicated in
the permit
Remove from
Consideration
(Already
controlled)
Lone Star Industries Inc.
(NEI Unit ID 9180513)
Buzzi Unicem
Semi-Dry Kiln
Putnam
1,578
328.66
Low NOx
calciner +
good
combustion
practice
(GCP)
CEMS
Coal or Oil
Kiln has an
emissions
limit of
5.5141b
NOx/ton of
clinker
produced
Possible
(Possible
emissions
reductions)
Subtotal
1,203.41
22
September 1, 2020
-------
Conclusions of Verification and Review of Controls on Non-EGU Sources in Four States and Potential Emissions
Reductions
CoST identified cost-effective (i.e., $2,000 per ton or less) control technologies for 20 emissions units at glass
manufacturing and cement manufacturing facilities in Pennsylvania, New York, Ohio, and Indiana. None of these
units are owned by small businesses as defined by the Small Business Administration's (SBA) small business size
standards for these two industry sectors.32 The total potential emissions reductions in Pennsylvania, New York,
Ohio, and Indiana in tranche 1 were 7,556 ozone season tons. We looked at potential emissions reductions in
tranche 1 that were estimated to cost <$2,000 per ton (likely cost-effective), which were 6,346 ozone season tons.
We reviewed online permits for these 20 units and as indicated in Tables 9 through 12, 10 of these units either (i)
have controls and monitors (primarily CEMS) already, (ii) are installing controls and CEMS or consolidating
operations in the next few years as a result of recent consent decrees issued as part of the EPA's New Source
Review Air Enforcement Initiative, (iii) have shutdown, or (iv) are planning to shut down by 2023.33 Table 13 below
summarizes the status of the potential emissions reductions.
Table 13. Status of Estimated Emissions Red
uctions
# of
OS Tons
(% of Total)
Emissions
Units
Shutdowns
4
824
13
Lehigh Cement - Kiln Replacements
3
366
6
NEI Discrepancy/Uncertain
1
3,286
51
Already Controlled/Uncertain
5
967
15
Possible Emissions Reductions
7
903
14
TOTAL
20
6,346
Based on the 2023 projected inventory, the emissions reductions from the plant shutdowns and consolidated
operations (between 2015 and 2023) are estimated to be approximately 824 tons, or 13 percent of the potentially
cost-effective ozone season emissions reductions in tranche 1. These emissions reductions are not currently
reflected in the estimated air quality impacts shown above in Tables 7 and 8. In addition, for the Lehigh Cement
manufacturing facility in Lawrence County, Indiana (emissions reductions estimated to be 366.40 tons, or 6
percent of the potentially cost-effective ozone season emissions reductions in tranche 1) that is subject to a
consent decree, the 2023 projected inventory emissions are 1,758 tons and we currently do not know what the
expected emissions reductions may be. We have found that the three older kilns currently in operation will shut
down by 2023 and be replaced with a single new kiln whose production capacity will be almost 3 times as large
(2.8 million tons of clinker compared to 1 million tons of clinker currently) and whose NOx emissions are
unknown.
Ten facilities, summarized again below in Table 14, were estimated to have the potential to generate some
emissions reductions. However, results from the review of online permit review and industrial trade literature
suggest that some of those potential reductions may not be true potential reductions. The status of the potential
reductions at the ten facilities is summarized below, along with an assessment of the likelihood that
32 U.S. Small Business Administration (SBA). Table of Small Business Size Standards as of August 19, 2019. Available at
https://www.sba.gov/sites/default/files/2019-
08/SBA%20Table%20of%20Size%20Standards_Effective%20Aug%2019%2C%202019_Rev.pdf.
33 The status of three of these 10 facilities reflects some uncertainty. Those facilities include Vitro Flat Glass LLC/Carlisle, PA,
Pittsburgh Glass Works/Meadville Works 8, PA, and Pilkington North America Inc., OH. The uncertainty associated with the
potential emissions reductions from these three facilities is discussed in this section.
23
September 1, 2020
-------
recommended controls could generate any emissions reductions. The assessment for each facility concludes with
either "uncertain" or "possible" depending on the likelihood of potential emissions reductions.
Vitro Flat Glass LLC/Carlisle. PA (NEI Unit ID 18725313). 3.285.65 OS tons - The cost per ton and estimated
emissions reductions for this emissions unit reflect a high degree of uncertainty. The uncertainty comes
from the following two sources: (i) discrepancies between the underlying information for this unit in the
2023 projected inventory and other more recent emissions data, and (ii) the equation in CoST that is used
to estimate the emissions reductions and cost per ton value. In the projected inventory, the pre-control
2023 emissions for one of the emissions units (10,514 tons) are much larger than what appears in the
2018 PA Air Emissions Report for the entire facility (1,770 tons) and six times larger than any other glass
furnace in this analysis.34 The discrepancies in emissions (roughly 8,700 tons) and estimated emissions
reductions (3,285.65 ozone season tons, or 51 percent of the likely cost-effective ozone season emissions
reductions in tranche 1) are not currently reflected in the estimated air quality impacts shown above in
Tables 7and 8. In addition, the projected inventory does not show a control on any unit at this facility,
even though a review of the permit indicates that one unit does have a control.
Emission Reduction Potential: Uncertain, NEI discrepancy
Pittsburgh Glass Works/Meadville Works 8. PA (NEI Unit ID 19025613). 543.49 OS tons -- This facility shut
down one of its two production lines effective June 10, 2020. The company stated that it is too expensive
to rebuild the production line.
Emission Reduction Potential: Uncertain, potentially shutdown
Anchor Glass Container Corp. NY (NEI Unit ID 2854113). 140.63 OS tons -- Furnace #2 has an annual
emissions limit of 4.5 lb NOx/ton of glass produced and no current control.
Emission Reduction Potential: Possible
Owens Brockway Glass Container Inc., NY (NEI Unit ID 2863113), 96.69 OS tons-The permit shows two
furnaces with no controls.
Emission Reduction Potential: Possible
Pilkington North America Inc., OH (NEI Unit ID 55204113), 236 OS tons - In the permit, Furnace #1 was
listed with an annual emissions limit of 364.7 tons, but a recent stack test indicates emissions at the
furnace are 41.64 tons of NOx.
Emission Reduction Potential: Uncertain, potentially already controlled
Ardagh Glass Inc.. IN (NEI Unit ID 65375713). 97.63 OS tons - In the permit, Furnace #2 has an annual
emissions limit of 506.9 tons and the 2023 projected emissions are 312 tons. It is possible the source is
currently operating under its permit limit.
Emission Reduction Potential: Possible
Ardagh Glass Inc.. IN (NEI Unit ID 65375813). 87.65 OS tons - In the permit, Furnace #1 has an annual
emissions limit of 389.24 tons and the 2023 projected emissions are 280 tons. It is possible the source is
currently operating under its permit limit.
Emission Reduction Potential: Possible
Anchor Glass Container Corporation. IN (NEI Unit ID 28314513). 86.28 OS tons - Facility-wide annual
emissions limit of 396 tons, which likely includes additional emissions sources.
34 In the 2017 NEI, the NOx emissions for the larger furnace are approximately 2,076 tons.
24
September 1, 2020
-------
Emission Reduction Potential: Possible
Lehigh Cement Company. IN (NEI Unit ID 65392513). 65.49 OS tons - Currently uses water injection as
control technology.
Emission Reduction Potential: Possible
Lone Star Industries Inc., IN (NEI Unit ID 9180513), 328.66 OS tons - Currently uses low NOx calciner +
good combustion practice (GCP) as control technology.
Emission Reduction Potential: Possible
25
September 1, 2020
-------
Table 14. Potential Emissions Reductions from Glass Manufacturing Facilities in Pennsylvania, New York, Ohio, and Indiana
Facility Name/NEI Unit ID
Type of
Emissions Unit
County (State)
2023 Projected
NOx Emissions
Estimated
Reductions
(OS Tons)
Existing NOx
Control
NOx
Monitoring
Device/
Technique
Status of
Estimated
Reductions
Vitro Flat Glass LLC/Carlisle
(NEI Unit ID 18725313)
Flat Glass:
Melting Furnace
Cumberland (PA)
10,514
3,285.65
No Control
CEMS
Uncertain,
NEI
discrepancy
Pittsburgh Glass
Works/Meadville Works 8
(NEI Unit ID 19025613)
Flat Glass:
Melting Furnace
Crawford (PA)
1,739
543.49
No Control
CEMS
Uncertain,
potentially
shutdown
Anchor Glass Container Corp
(NEI Unit ID 2854113)
Container Glass:
Melting Furnace
Chemung (NY)
450
140.63
Two furnaces -
Furnace #1-SCR;
Furnace #2 - no
control
CEMS
Possible
Owens Brockway Glass
Container Inc
(NEI Unit ID 2863113)
Container Glass:
Melting Furnace
Cayuga (NY)
309
96.69
Two furnaces - No
controls indicated
CEMS
Possible
Pilkington North America
Inc.
(NEI Unit ID 55204113)
Flat Glass:
Melting Furnace
Wood (OH)
755
236
Two furnaces -
None indicated for
furnace #1; 3R
technology for
furnace #2
(technology is
proprietary)
CEMS
Uncertain,
potentially
already
controlled
Ardagh Glass Inc.
(NEI Unit ID 65375713)
Container Glass:
Melting Furnace
Randolph (IN)
312
97.63
No Control
No monitors
Possible
Ardagh Glass Inc.
(NEI Unit ID 65375813)
Container Glass:
Melting Furnace
Randolph (IN)
280
87.65
No Control
No monitors
Possible
Anchor Glass Container
Corporation
(NEI Unit ID 28314513)
Container Glass:
Melting Furnace
Dearborn (IN)
276
86.28
No control
No monitors
Possible
Lehigh Cement Company
(NEI Unit ID 65392513)
Wet Kiln
Cass (IN)
314
65.49
Wl (water
injection)
CEMS
Possible
Lone Star Industries Inc.
(NEI Unit ID 9180513)
Semi-Dry Kiln
Putnam (IN)
1,578
328.66
Low NOx calciner +
good combustion
practice (GCP)
CEMS
Possible
26
September 1, 2020
-------
In summary, the total potential emissions reductions in Pennsylvania, New York, Ohio, and Indiana in tranche 1
were 7,556 ozone season tons. We looked at potential emissions reductions in tranche 1 that were estimated to
cost <$2,000 per ton, which were 6,346 ozone season tons. Between unit shutdowns and potentially incorrect
emissions data in the 2023 projected inventory (and a resulting incorrect estimate of potential emissions
reductions), of the 6,346 tons approximately 4,110 tons, or 64 percent, of the likely cost-effective emissions
reductions are not or may not be true emissions reductions. The potential emissions reductions associated with
applying CoST-recommended controls that are considered possible are 903 ozone season tons, or 14 percent of
the likely cost-effective emissions reductions.
Caveats and Limitations of the Cost Analysis
The EPA acknowledges several important caveats and limitations of the non-EGU cost assessment included in this
memorandum, which include the following:
Boundary of the cost analysis: In this cost analysis we include only the impacts to the sectors and facilities that are
the focus of this analysis. We include the costs for purchase, installation, operation, and maintenance of control
equipment over the lifetime of the equipment. Recordkeeping, reporting, testing and monitoring costs are not
included.35 Additional revenue may be generated by vendors that would build, install, and test new control
technologies for use at sources in the directly affected sectors, especially for control equipment manufacturers,
distributors, or service providers. These revenue and employment impacts are not included in this cost analysis.
Cost and effectiveness of control technologies: The application of control technologies reflect average retrofit
factors nationally and equipment life. We do not account for regional or local variation in capital and annual cost
items such as energy, labor, and materials. The estimates of control technology costs may over- or under-estimate
the costs depending on how the difficulty of actual retrofitting and equipment life compares with the control and
cost assumptions. In addition, the estimates of control efficiencies for control technologies included in the
assessment assume that the control devices are properly installed and maintained.
Interest rate: We apply an interest rate of 7 percent to annualize capital costs in the analysis. In addition, while
this interest rate is consistent with guidance as found in the EPA Air Pollution Control Cost Manual,36 (hereafter
called the "Control Cost Manual") the actual interest rate may vary for control cost estimation at each facility
included in this analysis.
Accuracy of control costs: We estimate that there is an accuracy range of +/- 30 percent for non-EGU point source
control cost estimates. This level of accuracy is described in the Control Cost Manual, which is a basis for the
estimation of non-EGU control cost estimates included in this memorandum. This level of accuracy is consistent
with either the budget or bid/tender level of cost estimation (or Class 4) as defined by the American Association
for Cost Engineering (AACE) International and explained in Section 1, Chapter 2 of the Control Cost Manual. In
addition, the accuracy of costs is also influenced by the availability and extent of data underlying the cost
estimates for individual control technologies.
35 Many of the sources included in this cost analysis already have NOx monitors (primarily CEMS) installed, as shown in Tables
9-14, which partially offsets this limitation.
36 U.S. EPA, Office of Air Quality Planning and Standards. EPA Air Pollution Control Cost Manual. Section 1, Chapter 2, pp. 15-
17. Available on the Internet at https://www.epa.gov/sites/production/files/2017-
12/documents/epaccmcostestimationmethodchapter 7thedition 2017.pdf.
27
September 1, 2020
-------
Control Installation Timing
We previously examined the time necessary to install the controls listed above for different industries. The 2016
Final Technical Support Document (TSD)for the Final Cross-State Air Pollution Rule for the 2008 Ozone NAAQS,
Assessment of Non-EGU NOx Emission Controls, Cost of Controls, and Time for Compliance Final TSD (CSAPR
Update non-EGU TSD) provided preliminary estimates of installation times for a variety of NOx control
technologies applied to a large number of sources in non-EGU industry sectors.37 For virtually all NOx controls
applied to cement manufacturing and glass manufacturing information on installation times was not available to
provide an estimate, and we concluded that the installation time for these controls was "uncertain." There was an
exception for SNCR applied to cement kilns, and the installation time estimate of 42-51 weeks listed in the CSAPR
Update non-EGU TSD does not account for implementation across multiple sources, the need to have NOx
monitors installed, and other steps in the permitting and construction processes.
To improve upon information from the CSAPR Update Non-EGU TSD on installation times for SCR on glass
furnaces and SNCR on cement kilns, EPA reviewed information from permitting actions and a consent decree. For
two glass manufacturing facilities that installed SCR on glass furnaces, from the time of permit application to the
time of SCR operation was approximately 19 months for one facility and is currently at least 20 months for
another facility.38 These installation times do not reflect time needed for pre-construction design and engineering,
financing, and factors associated with scaling up construction services for multiple installations at several
emissions units. With respect to cement kilns, an April 2013 consent decree between EPA and CEMEX, Inc.
required installation of SNCR at a kiln within 450 days, or approximately 15 months, of the effective date of the
consent decree.39 Similarly, this installation time does not reflect time associated with scaling up construction
services for multiple control installations at several emissions units.
Request for Comment and Additional Information
To develop a more complete record the EPA requests comment on several questions related to specific control
strategies the Agency evaluated, and in particular seeks feedback and data from stakeholders with relevant
expertise or knowledge. Should such additional information and analyses show that emissions reductions from
non-EGU sources in the linked upwind states would be more cost-effective than what is included in the EPA's
current assessment, available for installation earlier than the EPA estimates, or more impactful on downwind air
quality than the EPA's current information suggests, then the Agency remains open to the possibility of finalizing a
rule requiring such controls as may be justified under the Step 3 multi-factor test.
As indicated above, information about existing controls on non-EGU emissions sources in the inventory was
missing for some states and incomplete for some sources. The approach the EPA used in this proposal was to
assess emission reduction potential using CoST and the projected 2023 inventory to identify emissions units that
were uncontrolled. Given that the EPA's assessment of any other NOx control strategies would also rely on CoST,
the CMDb, and the inventory to identify emissions units that were uncontrolled and to assess emission reduction
potential from non-EGU sources, the Agency believes such an assessment would likely lead to a similar conclusion
that estimated emission reduction potential is uncertain.
37 The CSAPR Update non-EGU TSD is available on the EPA's website at the following link:
https://www.epa.gov/airmarkets/assessment-non-egu-nox-emission-controls-cost-controls-and-time-compliance-final-tsd.
38 Cardinal FG Company submitted a permit application to the Wisconsin Department of Natural Resources (WIDNR) to
construct an SCR in December 2017 at a facility in Portage, Wisconsin. The SCR was expected to be ready for testing in mid-
July 2019. In addition, Cardinal FG Company submitted a permit application to the WIDNR to construct an SCR in January
2019 at a facility in Menomonie, Wisconsin. The SCR is currently not operational.
39 The consent decree can be located at the following link: https://www.epa.gov/sites/production/files/documents/cemex-
lvons-cd.pdf.
28
September 1, 2020
-------
As such, for this and future regulatory efforts, to improve the underlying data used in an assessment of emission
reduction potential from non-EGU sources, we request comments on: (i) the existing assessment of emission
reduction potential from glass furnaces and cement kilns; (ii) emission reduction potential from other control
strategies or measures on a variety of emissions sources in several industry sectors; and (iii) the feasibility of
further controlling NOx from IC engines and large ICI boilers, including optimizing combustion and installing ultra-
low NOx burners. The three sections below introduce the areas for comment and describe workbooks generated
by CoST, the CMDb, and the 2023 projected inventory with the underlying data to review for comment.
First, the EPA requests comment on the aspects of the assessment presented above of emission reduction
potential from the glass and cement manufacturing sectors. To help inform review and comments, please see the
following Excel workbooks available in the docket: (i) for a summary of the CoST run results CoST Control Strategy
- Max Reduction $10k 150 tpy cutoff 12 States Updated Modeling - No Replace - 07-23-2020, and (ii) for
summaries of emissions reductions by control technologies, Control Summary - Max Reduction $10k 150 tpy
cutoff 12 States Updated Modeling - No Replace - 05-18-2020. Note that the CoST Control Strategy - Max
Reduction $10k 150 tpy cutoff 12 States Updated Modeling - No Replace - 07-23-2020 Excel workbook includes a
READ ME worksheet that provides details on the parameters used for the CoST run.
Specifically, the EPA is soliciting comment on the following:
Are applying SCR to uncontrolled or under-controlled glass furnaces and SNCR to uncontrolled or under-
controlled cement kilns in the linked states feasible approaches to achieve cost-effective emissions
reductions? If not, what types of cost-effective controls can be applied to these sources?
Does the EPA have the right and most up to date information on emissions and existing control
technologies for the units included in this assessment? If not, what is the correct and more up to date
information?
After looking at the underlying CoST run results, are the cost estimates accurate and reasonable? If not,
what are more accurate cost estimates?
What is the earliest possible installation time for SCR on glass furnaces?
What is the earliest possible installation time for SNCR on cement kilns?
For the non-EGU facilities without any emissions monitors, what would CEMS cost to install and operate?
How long would CEMS take to program and install?
In addition to the assessment of emission reduction potential from the glass and cement manufacturing sectors,
for the 12 linked states the EPA attempted to summarize all potential control measures for emissions units with
150 tpy or more pre-control NOx emissions in 2023 in several industry sectors. This information illustrates that
there are many potential approaches to assessing emissions reductions from non-EGU emissions sources or units.
We used the Least Cost Control Measure worksheet from a CoST run.40 By state for the 12 linked states and then
by facility, this information is summarized in the Excel workbook titled CoST Control Possibilities $10k 150 tpy
cutoff 12 States Updated Modeling - 06-30-2020, also available in the docket.
Second, specifically the EPA requests comment on the following:
Other than glass and cement manufacturing, are there other sectors or sources that could achieve
potentially cost-effective emissions reductions? What are those sectors or sources? What control
technologies achieve the reductions? What are cost estimates and installation times for those control
technologies?
Are there other sectors where cost effective emission reductions could be obtained by, in lieu of installing
40 The Least Cost Control Measure worksheet is a table of all possible emissions source-control measure pairings (for sources
and measures that meet the respective criteria specified for a control strategy), each of which contains information about the
cost and emissions reductions achieved if the control measure were to be applied to the emissions source.
29
September 1, 2020
-------
controls, replacing older, higher emitting equipment with newer equipment?
Are there sectors or sources where cost effective emission reductions could be obtained by switching
from coal-fired units to natural gas-fired units?
For non-EGU sources without emissions monitors, what would CEMS cost to install and operate? How
long would CEMS take to program and install? Are monitoring techniques other than CEMS, such as
predictive emissions monitoring systems (PEMS), sufficient for certain non-EGU facilities that would not
be brought into a trading program? If so, for what types of non-EGU facilities, and under what
circumstances, would PEMS be sufficient? What would be the cost to install and operate monitoring
techniques other than CEMS?
Third, in the workbook titled CoST Control Possibilities $10k 150 tpy cutoff 12 States Updated Modeling - 06-30-
2020 the EPA included two worksheets with information on controls for ICI boilers and IC engines: (i) Boilers -
ULNB and (ii) IC Engines - LEC. For the 12 linked states, the EPA summarized CoST's application of ultra-low NOx
burners (ULNB) on ICI boilers and low emission combustion (LEC) on IC engines. Assuming that the estimated
emissions reductions from CoST's application of these controls are real and cost-effective, there could be
approximately 5,000 ozone season tons of emissions reductions from 52 ICI boilers and 8,000 ozone season tons
of emissions reductions from 69 IC engines. This information is summarized in Table 15 below.
Table 15. Summary of Potential Emissions Reductions from ULNB on ICI Boilers and LEC on IC Engines
ICI Boilers
IC Engines
Number of Emissions Units in the 12 Linked States
(>150 tpy NOx emissions)
52
69
2023 Projected Total NOx Emissions in the 12 Linked
States (ozone season tons, reflects any existing
control before ULNB or LEC were applied)
6,779
9,260
2023 Projected Total NOx Emissions in the 12 Linked
States after Applying ULNB to Boilers (ozone season
tons)
1,695
-
2023 Projected Total NOx Emissions in the 12 Linked
States after Applying LEC to IC Engines (ozone season
tons)
-
1,231
Number of Units with No Known Existing Control
51
57
The EPA is requesting comments on the feasibility of further controlling NOx from IC engines and large ICI boilers,
including optimizing combustion and installing low NOx burners. The Agency understands that it is generally
possible to install low NOx burners on EGU boilers fairly quickly and that these burners can significantly reduce
NOx emissions. We note that in the original interstate transport rule, the NOx SIP call, the Agency concluded that
controls on large, non-EGU boilers and turbines were cost effective and allowed states to include those emissions
sources in their budgets as a means of providing additional opportunities to reduce state-wide NOx emissions in a
cost-effective manner.41 Therefore, we solicit comment on whether the EPA should require that large non-EGU
boilers and turbines - as defined in the NOx SIP call as boilers and turbines with heat inputs greater than 250
mmBTU per hour or with NOx emissions greater than 1 ton per ozone season day42 - within the 12 states employ
controls that achieve emissions reductions greater than or equal to what can be achieved through the installation
of low NOx burners.
41 See 63 FR 57402.
42 Note that the 250 mmBTU/hr for ICI boilers and turbines is equivalent to 25 MW heat input for an EGU. Also, the tonnage
per source was 1 ton per ozone season day. Because controls on non-EGUs operate year-round, the emissions would be 365
tons per year.
30
September 1, 2020
-------
Also, five of the 12 states that are subject to this rulemaking are also within the Ozone Transport Region (OTR) -
Maryland, New Jersey, New York, Pennsylvania, and Virginia. As member states of the OTR, these five states are
required to implement reasonably available control technology (RACT) state-wide on major sources of
emissions.43 It is likely that NOx controls, such as low NOx burners, are already in wide-spread use within states
these five states. However, such controls may not be as widely used in states outside of the OTR. Therefore, we
also solicit comment on the (i) magnitude of the emissions reductions that could be achieved by requiring that
large non-EGU boilers and turbines install controls that achieve emissions reductions greater than or equal to
what could be achieved through the installation of low NOx burners, (ii) prevalence of these or better NOx
controls already in place on this equipment in these 12 states, and (iii) time it typically takes to install such
controls.
In addition to the above, the EPA is requesting comments on the following:
How effective are ultra-low NOx burners or low NOx burners in controlling NOx emissions from ICI
boilers?
Are they generally considered part of the process or add-on controls? If they are part of a process, how
could the EPA estimate the cost associated with changing the process to accommodate ultra-low NOx
burners and low NOx burners?
What are the costs (capital and annual) for these as add-on control technologies on ICI boilers?
What are the earliest possible installation times for these control technologies on ICI boilers? The EPA
believes it is generally possible to install low NOx burners on EGU boilers relatively quickly and that low
NOx burners can significantly reduce NOx emissions. The EPA solicits comment on whether this is also
true for large non-EGU ICI boilers.
Do some of the emissions units included in the summary already have either add-on controls or controls
that are part of a process? If so, what control is on the unit and what is the control device (or removal)
efficiency?
Natural gas compressor stations are the largest NOx-emitting non-EGU sector44 affecting the 12 states
that are the subject of this proposal, and many of these facilities are powered by decades-old,
uncontrolled IC engines. Should emissions reductions be sought from the IC engines at these stations,
either through installing controls, upgrading equipment, or other means?
How effective is low emission combustion in controlling NOx from IC engines?
What is the cost (capital and annual) for low emission combustion on IC engines?
What is the earliest possible installation time for low emission combustion on IC engines? In lieu of
installing controls, is replacing older, higher emitting equipment with newer equipment a cost-effective
way to reduce emissions from IC engines?
Do some of the emissions units included in the summary already have either add-on controls or controls
that are part of a process? If so, what control is on the unit and what is the control device (or removal)
efficiency?
Attachments
1. CoST Control Strategy - Max Reduction $10k 150 tpy cutoff 12 States Updated Modeling - No Replace - 07-
23-2020.xlsx
2. Control Summary - Max Reduction $10k 150 tpy cutoff 12 States Updated Modeling - No Replace - 05-18-
2020.xlsx
43 One exception to the requirement of state-wide RACT within the OTR is for Virginia. Only the Northeast portion of the state
is included within the OTR and only facilities within that portion of the state are subject to RACT.
44 Based on data from the 2017 National Emissions Inventory (NEI) database. For additional details on the 2017 NEI data
summaries, please see the Excel workbook titled 2017 NEI Data_Twelve States_Merged_Greater than 100 Tons in the docket.
31
September 1, 2020
-------
CoST Control Possibilities $10k 150 tpy cutoff 12 States Updated Modeling - 06-30-2020.xlsx
2017 NEI Data_Twelve States_Merged_Greater than 100 Tons.xlsx
32
September 1, 2020
------- |