Technical Support Document for
Process Emissions from Cement: Proposed Rule for
Mandatory Reporting of Greenhouse Gases
Office of Air and Radiation
U.S. Environmental Protection Agency
January 28, 2009
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
Contents
1. Industry Description 3
2. Total Emissions 3
2.1 Process Emissions 3
2.1.1 Calcination Process 4
2.1.2 Raw Materials 4
2.2 Combustion Emissions 5
3. Review of Existing Programs and Methodologies 5
3.1 2006 IPCC Guidelines 6
3.2 U.S. National Inventory Report 2008 Method 8
3.3 WRI / WBCSD Protocol (Cement Sustainability Initiative) 9
3.4 California AB32 9
3.5 California Climate Action Registry (CCAR) 9
3.6 Department of Energy's 1605(b) Voluntary Reporting Program 9
3.7 EPA Climate Leaders 10
3.8 EUETS 10
3.9 New Mexico Mandatory GHG Reporting Program 11
3.10 The Climate Registry 11
4. Options for Reporting Threshold 12
4.1 Options Considered 12
4.1.1 Emissions-based Threshold 12
4.1.2 Clinker Production Capacity-based Threshold 12
4.1.3 All Clinker Production Facilities 13
4.2 Analysis of Emissions and Facilities Covered Per Option 13
4.2.1 Emissions-based Threshold 13
4.2.2 Clinker Production Capacity-based Threshold 14
4.2.3 All Clinker Production Facilities 16
5. Options for Monitoring Methods 16
5.1 Option 1: Direct Measurement (Annual Reporting) 17
5.2 Option 2: Hybrid Method (Annual Reporting) 17
5.3 Option 3: Simplified Emission Calculation Method 19
6. Options for Estimating Missing Data 19
6.1 Direct Measurement 19
6.1.1 Continuous Emissions Monitoring System (CEMS) 19
6.1.2 Stack Testing 20
6.2 Facility-specific Emission Calculation 21
7. QA/QC Requirements 21
7.1 Direct Measurement 21
7.2 Facility-specific Emission Calculation 22
8. Types of Emissions Information to be Reported 23
8.1 Additional Data to be Retained Onsite 24
9. References 25
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
1. Industry Description
In 2006, there were 113 cement plants in 37 U.S. States and 2 cement plants in Puerto Rico.
Total cement production capacity was approximately 115 million metric tons. About 94 million
metric tons of portland cement and 6 million metric tons of masonry cement were produced
(USGS 2007).
EPA developed a Draft Cement Database (EPA Draft Cement Database; EPA 2007), based on
Portland Cement Association's (PCA's) 2004 plant level summary data that contained 115
cement plants, excluding Puerto Rico (PCA 2006). It includes 107 integrated cement plants that
house both kilns for producing clinker and mills for grinding cement from clinker, and 8 fine
grinding-only plants that do not produce clinker and house only grinding mills for producing
cement from purchased clinker. Grinding-only facilities use purchased clinker and other
additives, are generally operated using purchased electricity, and have very limited on-site
combustion emissions. Plant level threshold and emissions coverage analysis reported
subsequently in this document were developed from this database.1 Since there are no onsite
process-related GHG emissions from the grinding-only facilities and their onsite combustion
emissions are very limited, grinding only facilities will not be discussed further in this
document.
U.S. clinker production (including Puerto Rico) totaled 88,453 thousand metric tons (EPA
2008a) from an industry capacity of almost 95,000 thousand metric tons (based on EPA Draft
Cement Database) in 2006. Imports of clinker were also higher due to higher demand. The 6
leading cement producing states (i.e., Texas, California, Pennsylvania, Florida, Michigan, and
Alabama) accounted for 48% of U.S. production in 2006 (USGS 2007).
The process-related emissions of CC>2 from 2006 cement production were estimated to be 45.7
MMTCO2 (EPA 2008a). This is equivalent to 0.52 metric tons of process CC>2 per metric ton of
clinker and accounts for more than half of total GHG emissions (comprising both combustion
and process-related) from cement industry.
2. Total Emissions
Total combustion and process-related GHG emissions from 2006 cement production, including
CH4 and N2O emissions from fossil fuel combustion based on plant-specific characteristics in the
EPA Draft Cement Database, were estimated to be 86.8 MMTCO2e (EPA 2007). This is
equivalent to 0.98 metric tons of CC^e per metric ton of clinker, of which 0.46 metric tons are
attributable to fuel combustion.
2.1 Process Emissions
Process-related CC>2 emissions from cement production are the second largest source of
industrial CC>2 emissions in the United States (EPA 2008a). Cement production process
comprises the following two steps: (i) clinker production and (ii) finish grinding. Process-related
1 Although this database was used for performing threshold analysis, the names and locations of individual cement
plants and their production capacities included in this database have been withheld from reporting in this document
due to non-disclosure copyright information contained in PCA (2006).
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
GHG emissions from cement production are mainly CC>2 emissions that arise during the clinker
production process. There are no CC>2 emissions from the finish grinding process, during which
clinker is ground finely with gypsum and other materials to produce cement.
Predominant sources of process-related CO2 emissions arise from calcination of carbonates that
formed clinker and from calcination of carbonates that formed clinker kiln dust (CKD).
Additional process-related CO2 emissions may arise from non-carbonate, total organic carbon
contained in the raw materials consumed for clinker production. The CC>2 generation process
during cement production is described below.
2.1.1 Calcination Process
During the cement production process, first, calcium carbonate or calcite (CaCOs) (common
sources of which include limestone and chalk) is heated in a cement kiln at a temperature of
o o
about 1,450 C (2,400 F) to form lime (i.e., calcium oxide or CaO) and CC>2 in a process known
as calcination or calcining (EPA 2008a). The lime then reacts with silica-containing materials
(such as sand and shale) and iron oxide and alumina in the raw materials to produce clinker,
which is an intermediate product. Very small amounts of carbonates other than CaCOs, such as
magnesite (or magnesium carbonates), and non-carbonate organic carbon may also be present in
the raw materials fed to the kiln (these raw materials are discussed in the next section). The
other carbonates and non-carbonate organic carbon also contribute to generation of additional
CC>2 during the calcination process (IPCC 2006). CC>2 generated during the calcination process as
a by-product is released to the atmosphere.
Clinker typically contains a large fraction of CaO and may contain a very small fraction of MgO
(magnesium oxide), which is formed during the calcination process from magnesite in the raw
materials. The CaO and MgO contents of clinker are controlled to tight specifications (IPCC
2006). The clinker is then cooled and is mixed with a small amount of gypsum, and other
materials, such as slag. The mixture, whose composition is varied depending on the type of
Portland cement that needs to be produced, is then ground together in a fine-grinding mill to
make portland cement. Portland cement is used both as an end-use product and also as an
intermediate product in producing different types of blended cement (FHWA 2008).
During clinker production, some of the clinker precursor materials instead of forming clinker,
form partially or fully calcinated cement kiln dust (CKD) (EPA 2008a). CKD is entrained in the
hot flue gases and is carried out the feed end of the kiln and eventually removed by fabric
filtration or electrostatic precipitation. Depending on the chemical makeup of the CKD,
especially its alkalinity, CKD may be recycled back to the kiln, added to the cement at the
finishing end, or disposed of on-site or off-site. Because CKD represents product, the emphasis
is on using it rather than wasting it. Additional process-related CO2 emissions are generated in
the kiln with the formation of partially or fully-calcined CKD.
2.1.2 Raw Materials
Cement production requires a major source of calcium and smaller sources of silicon, aluminum
and iron (PC A 2008). In the United States, limestone is the predominant source of calcium in
cement production. Other sources of calcium used in cement production include marl and chalk.
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
Silicon sources include sand, shale, clay, and fly ash. The iron and aluminum are added in the
form of iron ore and bauxite, or recycled metals, if not contained in the silicon source added.
Several other raw materials are added and/or supplemented to provide specific properties. A
wide range of raw materials are used in cement kilns for clinker production. In some cases, raw
materials serve both as raw materials and an energy source, such as in the case of hazardous
wastes that have CaO (van Oss 2004). Examples of the raw materials used in the U.S. cement
kilns are listed in Table 2.1 (PCA 2006).
Table 2.1 Example Raw Materials for U.S. Cement Kilns
Limestone
Alumina Corrective
Alumina Sludge
Aluminum Oxide
Bauxite
Biosolids
Blast Furnace Slag
Bottom Ash
Clay
Copper Slag
Filter Cakes
Fly ash
Foundry Sand
Foundry Sludge
Iron Dross
Iron Fines
Iron Ore
Iron Sludge
Iron Waste
Marl
Mill Scale
Petroleum
Contaminated Soil
Sand
Sandblast Grit
Shale
Silica Gel
Spent Catalyst
Waste Aluminum
2.2 Combustion Emissions
Combustion emissions include CO2, N2O and CFL; emissions that result from the combustion of
carbon-based fuels in the cement kiln and other onsite combustion equipment. The cement kiln
is the most significant of these combustion units and typically is fueled with coal. Other fossil
fuels are generally too expensive to be used for kiln fuel, however carbon-based wastes are
commonly combusted in the kilns. These wastes include solvents, oils and tires. The three
primary types of kilns used in the U.S. are wet kilns, simple dry kilns, and dry kilns equipped
with a preheater and precalciner; with the latter kiln being the most efficient. The other sources
of combustion equipment at cement plants consist of transportation equipment used in the
mining and transport of raw and finished materials
3. Review of Existing Programs and Methodologies
This section first summarizes the Tier 1, Tier 2 and Tier 3 methods recommended by the 2006
IPCC Guidelines for National GHG Inventories, followed by the method adopted in the
Inventory of U.S. GHG Emissions and Sinks 1990-2006 (EPA 2008a) for calculating process-
related emissions from cement production. Other emissions calculation methods under existing
reporting programs and protocols are described further in this section. Most of these programs
and protocols essentially apply IPCC Tier 2 and Tier 3 approaches with some degree of
flexibility (EPA 2008b). These programs include:
World Resources Institute/World Business Council for Sustainable Development
(WRI/WBCSD) Cement Sustainability Initiative,
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
California's Global Warming Solutions Act under Assembly Bill No. 32 (also simply
referred to as California AB32\
California Climate Action Registry,
DOE/EIA's Voluntary Reporting of Greenhouse Gases Program under Section 1605(b) of
the Energy Policy Act of 1992,
EPA Climate Leaders,
European Union Emission Trading System (EU ETS),
New Mexico's Mandatory Reporting Program, and
The Climate Registry.
In addition, three alternative approaches are discussed based on EPA's review of these existing
programs and current practices within the cement industry.
3.1 2006 IPCC Guidelines
The IPCC Guidelines recommend three different Tiers for calculating cement emissions.
Tier 1 Method
Under this method, process-related CC>2 emissions are calculated based on clinker production
estimates, derived from cement production data, after adjusting for imports and exports of
clinker, and a default CC>2 emission factor (of 0.52 metric tons of CC>2 per metric ton of clinker).
It is adjusted for CC>2 from CKD. The default emission factor assumes that the clinker contains
65 percent CaO, all of which is assumed to be from carbonate sources; it also assumes that all
carbonate sources are calcined 100 percent in the kiln. It is generally accepted that this is not a
preferred approach for facility level emissions estimation due to large uncertainty associated with
the CC>2 emission estimates associated with this approach.
CO 2 Emissions = fZt (Mci * Ccu) -Im + ExJ* EF
cic
Where:
CC>2 Emissions = emissions of CC>2 from cement production (metric tons)
Mc; = weight (mass) of cement produced of type i (metric tons)
Ccii = clinker fraction of cement type i (fraction)
Im = imports for consumption of clinker (metric tons)
Ex = exports of clinker (metric tons)
EFcic = emission factor for clinker in the particular cement, (metric tons of CC>2 per metric ton of
clinker). The default clinker emission factor (EFcic) is corrected for CKD.
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
Tier 2 Method
Under this method, process-related emissions are calculated directly from clinker production data
and a CO2 emission factor, determined by the CaO content of clinker. Usually, the CaO
content of clinker falls within the range of 60 to 67 percent and the CaO content of clinker within
a facility remain relatively stable (i.e., with a minimal 1 to 2 percent variation). For example, not
accounting for CKD, the CC>2 emission factor is 0.51 metric tons of CC>2 per metric ton of
clinker, when the CaO content is 65 percent; the emission factor falls within the range of 0.47
metric tons of CO2 per metric ton of clinker (for 60 percent CaO content of clinker) and 0.53
metric tons of CO2 per metric ton of clinker (for 67 percent CaO content of clinker).
However, if a facility derives significant amount of CaO from non-carbonate source such as steel
slag or fly ash, this fraction of CaO should be subtracted from the CaO content of clinker and the
CO2 emission factor based on clinker production must be adjusted accordingly. For example, if 5
percent of CaO in a 65 percent CaO clinker is from slag, the CaO from carbonate is 60 percent
and accordingly, the adjusted CO2 emission factor is 0.47 metric tons of CO2 per metric ton of
clinker, instead of 0.51 metric tons of CO2 per metric ton of clinker that has 65 percent CaO, all
from carbonate sources. This method does not adjust for CO2 emissions attributable to
calcination of carbonates in forming MgO content of clinker.
Process-related CO2 emissions attributable to calcined CKD not recycled to the kiln are added to
the clinker production based CO2 estimates by using a CKD adjustment factor of 1.02 (or 2% of
CO2 emissions attributable to clinker production). Under this method, CO2 emissions can be
calculated as follows.
Emissions = Mci EFci
Where,
CO2 Emissions = process-related emissions of CO2 from cement production (metric tons)
Mci = weight (mass) of clinker produced (metric tons)
EFci = emission factor for clinker (metric ton of CO2 per metric ton of clinker)
This clinker emission factor (EFci) is not corrected for CKD.
CFckd = emissions correction factor for CKD
Tier 3 Method
This method requires full accounting of carbonates in the raw materials. Under this method,
process-related CO2 emissions are calculated using data on the type, composition, quantity and
the emission factors of the carbonates consumed for clinker production. In addition, any organic
carbon and/or carbon residues in the raw materials (such as limestone, shale and fly ash) will
also need to be accounted for in estimating the CO2 emissions from clinker production
accurately. IPCC (2006) recommends accounting for CO2 emissions from organic carbon (or
kerogen) content in raw materials, if it accounts for more than 5 percent of the total heat used in
the kiln. IPCC (2006) recommends facilities to perform chemical analyses with sufficient
frequency to estimate the carbonate contents of raw materials used in clinker production at the
facility level and derive a facility-specific emission factor to estimate CO2 emissions based on
clinker production. The Tier 3 method includes an adjustment factor for subtracting the carbon in
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
the uncalcined carbonate within CKD not reused in the kiln. The following formula facilitates
calculating facility-specific CC>2 emissions from clinker production (IPCC 2006):
CO2 Emissions = ^(EF, *M, * F,)-[Md * Cd *(l-Fd)* EFd] + Zk(Mk * Xk * EFk )
Where,
CO2 Emissions = process-related emissions of CC>2 from cement production (metric tons)
EF; = emission factor for the particular carbonate /' (metric ton of CC>2 per metric ton of
carbonate)
M; = weight or mass of carbonate / consumed in the kiln, (metric tons)
F; = fraction calcination achieved for carbonate /'
Md = weight or mass of CKD not recycled to the kiln (metric tons)
Cd = weight fraction of original carbonate in the CKD not recycled to the kiln
Fd = fraction calcination achieved for CKD not recycled to kiln
EFd = emission factor for the uncalcined carbonate in CKD not recycled to the kiln,
(metric ton of CC>2 per metric ton of carbonate)
Mk = weight or mass of organic or other carbon-bearing non-fuel raw material k (metric
tons)
Xk = fraction of total organic or other carbon in specific non-fuel raw material k
EFk = emission factor for other carbon-bearing non-fuel raw material k (metric ton of
CC>2 per metric ton of carbonate)
3.2 U.S. National Inventory Report 2008 Method
In the Inventory, the process-related CC>2 emissions from cement manufacturing process for 2006
were calculated by multiplying a CC>2 emission factor (in tons of CC>2 released per ton of clinker
produced) and the total amount of clinker produced. The emission factor used in this analysis
was derived by assuming an average 65 percent lime fraction in clinker and a constant reflecting
the mass of CC>2 released per unit of lime. The CC>2 emission factor for clinker production is
calculated as follows:
EF Clinker = 0.65 CaO x [44.01 g/moleCO2 / 56.08 g/moleCaOJ
= 0.51 metric tons CO 2 per metric ton of clinker
The process-related CC>2 emissions calculated using the above specified clinker production
emission factor do not include the incremental CC>2 emissions attributable to the calcinated
portion of the CKD.
Therefore, consistent with the 2006 IPCC Guidelines, the additional process-related CC>2
emissions attributable to CKD are estimated as two percent of the total process-related CC>2
emissions calculated from clinker production (IPCC 2006). Total process-related CC>2 emissions
from cement manufacturing thus include the CC>2 emissions arising from both clinker production
and CKD generation.
This method is similar to the 2006 IPCC Tier 2 method in the sense that process-related CC>2
emissions are calculated based on clinker production and CC>2 emissions factor associated with
clinker production and CKD. However, because the estimates are developed at the national level,
this method uses default emission factors (as the representative emission factors) for calculating
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
process-related CO2 emissions from clinker production and from CKD. These emission factors
are consistent with the IPCC Tier 1 emission factors.
3.3 WRI / WBCSD Protocol (Cement Sustainability Initiative)
Under this protocol, corporate- or facility-level emissions can be calculated using either of the
following two approachesbased on corporate or facility-level clinker production data or based
on corporate or facility-level cement production data and facility-specific or recommended
default emission factors for CO2 from clinker production. This protocol takes into account both
CaO and MgO content of clinker and associated CO2 emissions in the clinker production data-
based approach. The default emission factor for clinker is 0.525 metric tons per metric ton of
clinker. This emission factor is consistent with the IPCC Tier 1 emission factor for clinker
production, which also includes emissions attributable to CKD. The protocol also includes a
more detailed methodology for calculating facility-specific clinker and CKD emission factors.
3.4 California AB32
Under the California Mandatory GHG reporting program, facility level process-related CC>2
emissions from clinker production must be calculated based on the volume and composition of
clinker produced and the amount of CKD discarded during the manufacturing process. Each
facility is required to calculate facility-specific clinker and CKD emission factors. This method
requires taking into account the CaO and MgO content of clinker and the fraction of this CaO
and MgO that is non-carbonate based, in calculating the facility level emissions. In addition, this
program also requires calculating CO2 emissions associated with organic carbon in raw materials
based on the weight of raw material consumed and the assumption that two tenth of 1 percent
(=0.2%) total organic carbon is contained in raw materials.
3.5 California Climate Action Registry (CCAR)
CCAR recommends corporate entities and preferably, facilities and sources, to calculate and
report their process-related CO2 emissions from cement manufacturing based on the amount of
clinker produced, CaO and MgO content of clinker, non-carbonate CaO and MgO, and the total
amount of CKD not recycled to the kiln. The clinker and CKD emission factors can be derived
based on the company-(or facility-/ source-) specific data collected on these factors. In the
absence of company-specific data to calculate emission factors, use of a default emission factor
is permitted. Process-related CO2 emissions from total organic carbon in raw materials is
calculated by assuming the raw materials contain 0.2% of total organic carbon.
3.6 Department of Energy's 1605(b) Voluntary Reporting Program
Three methods for calculating CO2 emissions are prescribed under this program. In the case of
the A-rated method (i.e., the best method), process-related CO2 emissions are calculated by
multiplying clinker production by the respective CaO and MgO emission factors, based on
measured CaO and MgO contents of clinker. In this method, facility-specific emission factors are
calculated. The B-rated method uses default emission factors or IPCC-recommended default
CaO and MgO contents of clinker and clinker production to calculate CO2 emissions. The C-
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
rated method is based on calculating CO2 emissions by applying a default CO2 emission factor to
total cement production. For calculating CC>2 emissions from CKD, the program's guidelines
prescribe A, B and C-rated methods, in the descending order of their stringency and preference
ranking.
3.7 EPA Climate Leaders
The Climate Leaders Program recommends 3 approaches for calculating and reporting corporate
level process-related CO2 emissions from cement manufacturing. The first approach involves
calculating and reporting CC>2 emissions based on clinker production, bypass dust produced, and
CKD produced, and carbon containing non-fuel raw materials consumed. This method, similar to
the WRI/WBSCD Cement Sustainability Initiative, also takes into account both facility-specific
CaO and MgO contents of clinker and CC>2 emissions associated with calcination of carbonates
that generated both these chemical compounds.
The second approach is the cement production-based approach, which takes into account the
amount of cement produced by type, adjusts for imports and exports and uses data on the clinker
and carbonate fractions, if data are available, in developing the process-related CC>2 emissions.
This is not a preferred approach due to large uncertainty associated with the CC>2 emission
estimates associated with this approach.
The third approach calculates process-related CC>2 emissions based on carbonate inputs, non-
carbonate carbon containing non-fuel raw materials, and CKD, and corresponding default
emission factors. This method also takes into account both CaO and MgO content of clinker and
CO2 emissions associated with calcination of carbonates attributed to these.
3.8 EU ETS
In the EU ETS protocol the recommended CO2 calculation methodology for cement differs by
the two reporting periods. All facilities that have the production capacity of greater than 500
metric tons of cement per day are required to report their facility level process-related CO2
emissions using one of the recommended approaches.
First Reporting Period
Under Method A (also referred to as the carbonate consumption based approach), the process-
related CO2 emissions are calculated by measuring the amount of carbonates in process inputs
and multiplying those with the default emission factors for carbonates consumed. These
emissions are adjusted to account for unreleased CO2 in the uncalcined carbonates. This method
accounts for CO2 emissions from calcination of calcium and magnesium carbonates. The EU
ETS protocol recommends two tiers for measuring carbonates in inputs using metering. While
Tier 1 method allows for a maximum of-5% to + 5% uncertainty, Tier 2 method allows for a
maximum uncertainty range of-2.5% to +2.5% in the metered amount of pure carbonates
contained in the process inputs. Default emission factors can be used under both tiers.
10
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
Under Method B (also referred to as clinker production based approach), process-related CC>2
emissions from clinker production and CKD production are first calculated and the CKD
emissions are added to the clinker-production based CC>2 emissions. This method includes three
tiers of activity data measurement stringency. In the case of clinker production, the maximum
permissible uncertainty surrounding the amount of clinker in process output metered is -5% to
+5% under Tier 1, -2.5% to +2.5% under Tier 2, and is -1.5% and +1.5% under Tier 3. In the
case of CKD production, the maximum permissible uncertainty surrounding the amount of CKD
in process output metered is -10% to +10% under Tier 1 and -5% to +5% under Tier 2. Default
emission factor can be used under Tier 1 and emission factors are calculated using WRI/WBCSD
formula.
Second Reporting Period
Under Method A (also referred to as Kiln Input approach), the process-related CC>2 emissions are
calculated by measuring the process input and multiplying it with the CC>2 emission factor for
carbonates and conversion factors, which are factors that account for the fraction of carbon in the
raw materials that are not converted to CC>2. Three tiers exist for estimating inputs to kiln, each
with varying degrees of maximum permissible uncertainty range. The maximum permissible
uncertainty range associated with measurement of inputs to kiln is -7.5% to +7.5% under Tier 1,
-5% to +5% under Tier 2 and -2.5% to +2.5% under Tier 3. It recommends calculating an
emission factor for each relevant kiln input and provides two tiers of conversion factors.
Under Method B (also referred to as Clinker method), the process-related CC>2 emissions are
calculated by adding CC>2 emissions associated with calcination of CKD and bypass dust, CC>2
emissions attributable to non-carbonate raw materials and CC>2 emissions from clinker
production. The two tiers of uncertainty associated with measurement of clinker production are: -
5% to +5% under Tier 1 and -2.5% and +2.5% under Tier 2. Non-carbonate raw materials are
measured with maximum possible uncertainty of-15% to +15% under Tier 1 and -7.5% and
+7.5% under Tier 2. Stringency of calculating emissions and conversion factors also differ by the
different tiers: it ranges from using default emission factors to adopting best industry practice for
calculation to using country-specific and facility-specific emission and conversion factors.
3.9 New Mexico Mandatory GHG Reporting Program
Under this reporting program, two approaches are recommended for measuring facility-level
process-related CC>2 emissions from cement production. The first approach recommends
measuring CC>2 emissions using CEMS. The second method is the production based approach,
under which the production process related CC>2 emissions are calculated as the sum of clinker
production based emissions, CKD production based emissions and total organic carbon content
in raw materials.
3.10 The Climate Registry
Under the reporting guidelines of the Registry, facility level process-related CC>2 emissions from
cement production can be calculated using either clinker method or carbonate method. The
clinker method involves adding CC>2 emissions from calcination of carbonates attributable to
11
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
clinker and CKD production and from non-carbonate, total organic carbon content in the raw
materials consumed. Facility-specific or default emission factors for clinker production and CKD
production can be used. Default value of total organic carbon content of 0.2% of raw material by
weight can be used in the calculations. The carbonate method is similar to the Tier 3 IPCC
method, described above, but incorporates Registry-specific default values for emission factors.
For additional information on all of the programs listed above, please refer to the Review of
Existing Programs memorandum (EPA-HQ-OAR-2008-0508-053).
4. Options for Reporting Threshold
Several alternative emission and capacity threshold options for reporting facility-level GHG
emissions from the cement industry were analyzed. This section describes the reporting options
considered and associated emissions and the coverage of cement production facilities under each
option.
4.1 Options Considered
Three different threshold options were evaluated for purposes of this analysis: an emission-based
threshold, a clinker-based threshold, or no threshold. Although the main focus of this document
is to analyze the process-related CO2 emissions from cement production, (i) because combustion
accounts for nearly half of the total GHG emissions from the cement production facilities, (ii)
because the affected facilities will be required to report all their total emissions, and (iii) because
the emission threshold will be calculated based on the total emissions rather than process-related
emissions only, the emissions threshold and coverage analyses for the cement industry included
in this section correspond to total GHG emissions (comprising both process and combustion
related emissions), unless explicitly specified otherwise. Note also that a cement production
plant may contain one cement kiln or multiple cement kilns. For the purposes of this threshold
analysis, the entire cement production plant (including all of the collocated cement kilns) is
considered a single facility.
4.1.1 Emissions-based Threshold
An option is to require all facilities that exceed a total, annual on site emission-based threshold to
report. The threshold would be based on total emissions, including both production process and
combustion-related emissions. For this analysis the standard thresholds examined for all source
categories under the GHG rulemaking were analyzed: 1,000 metric tons of CO2e, 10,000 metric
tons of CO2e, 25,000 metric tons of CO2e and 50,000 metric tons of CO2e and 100,000 metric
tons of CO2e.
4.1.2 Clinker Production Capacity-based Threshold
Another option is to require facilities that exceed a given clinker production capacity threshold to
report. Two types of clinker production capacity threshold can be established: (a) Daily
production capacity-based threshold; and (b) Annual production capacity-based threshold.
12
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
Four annual clinker production capacity thresholds were analyzed for this study, based on the
clinker production rates that correspond to the standard CC>2 emission thresholds discussed
above: 10,000 metric tons per year, 25,000 metric tons per year, 50,000 metric tons per year or
100,000 metric tons per year. Similarly, four daily clinker production capacity thresholds that
generally correspond to the annual thresholds were analyzed for this study: 250 metric tons per
day, 500 metric tons per day, 750 metric tons per day and 1,000 metric tons per day.
4.1.3 All Clinker Production Facilities
Another option might be to require all integrated cement facilities to report their plant level
emissions. The EPA Draft Cement database contains 107 integrated cement facilities.
4.2 Analysis of Emissions and Facilities Covered Per Option
This section reports the results of the cement plant industry analysis for each of the following
reporting threshold option specified:
Emissions-based threshold
Clinker production capacity-based threshold
All clinker production facilities
Tables 2.1 through 2.4 illustrate the results of the data analyses in support of these alternative
options. In these tables, a facility corresponds to a cement manufacturing plant; the number of
emissions sources (i.e., kilns) located within the cement manufacturing plants has not been taken
into account.
4.2.1 Emissions-based Threshold
Table 4.1 summarizes the number of cement facilities that will be affected under alternative
facility-level total (production process- and combustion-related) annual emission-based threshold
and the corresponding industry emission coverage, based on 2006 facility-level emission
estimates. Covered emissions include both combustion- and process-related emissions. The
results of the analysis indicate that all 107 integrated cement facilities will be covered under the
program if the emission threshold is 50,000 metric tons of CC^e or below, accounting for all of
the total GHG emissions from cement manufacturing.
By increasing the annual emission threshold to 100,000 metric tons of CC>2, only one integrated
cement plant, whose 2006-emissions accounted for over 90,000 metric tons of CC^e and
approximately one-tenth of one percent of the cement industry emissions in 2006, will not be
covered under this program; the remaining 106 integrated cement plants, accounting for 99.9
percent of the emission will be covered by this threshold.
13
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
Table 4.1 Summary of the Impact of Various Total Annual Facility Level Emissions Thresholds on
the Effectiveness of the Reporting Rule.*
Number of cement facilities exceeding the
threshold
Total Cement Industry Emissions covered
(million metric tons of CO2e)
% of Total CO2e Emissions Covered by the
facilities exceeding the threshold
Total Clinker Production Capacity (1,000
metric tons)
% of Total Clinker Production Capacity
Threshold - Total Annual Facility Level CO2e Emissions
1,000
metric
tons of
CO2e
107
86.83
100.0%
94,385
100.0%
10,000
metric
tons of
CO2e
107
86.83
100.0%
94,385
100.0%
25,000
metric
tons of
CO2e
107
86.83
100.0%
94,385
100.0%
50,000
metric
tons of
CO2e
107
86.83
100.0%
94,385
100.0%
100,000
metric tons
of CO2e
106
86.74
99.9%
94,283
99.9%
Note: * Includes both combustion and process-related GHG emissions. Includes CH4 and N2O emissions from
stationary fossil-fuel combustion in the cement plants. Clinker production capacity was based on 2004 plant level
data published by Portland Cement Association. Emissions coverage corresponds to 2006 emissions calculated by
scaling 2005 clinker production and emissions developed in the EPA Draft Cement Database.
Source: Summarized from EPA Draft Cement Database (2007).
4.2.2 Clinker Production Capacity-based Threshold
Table 4.2 illustrates the number of affected facilities under alternative daily clinker production
capacity threshold, based on the EPA's 2005 Draft Cement Plant Database (2007) and associated
sector-wide CC>2 emissions coverage in 2006. The results indicate that at the production capacity
of 750 metric tons of clinker per day or below, over 99 percent of the 2006-total cement industry
GHG emissions and 96 percent of the total integrated cement facilities will be covered at this
threshold. If the threshold is increased to 1,000 metric tons of clinker production per day, over 97
percent of the 2006-emissions and about 91 percent of the total integrated cement facilities will
be covered. To put this in context, EU ETS recommends a production capacity threshold of 500
tons/day.
14
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
Table 4.2 Summary of the Impact of Various Daily Clinker Production Capacity Thresholds on the
Effectiveness of the Reporting Rule
Number of cement facilities exceeding the threshold
Total Daily Clinker Production Capacity (metric tons/day)
of facilities exceeding the threshold
Equivalent Total Annual Clinker Production Capacity
(1,000 metric tons/year) of facilities exceeding the threshold
% of Clinker Production Capacity covered by facilities
exceeding the threshold
% of Total CO2e Emissions covered by the facilities
exceeding the threshold
Threshold clinker production capacity
250 metric
tons / day
107
287,537
94,385
100.0%
100.0%
500 metric
tons / day
104
286,610
94,074
99.7%
99.6%
750 metric
tons / day
103
285,862
93,814
99.4%
99.3%
1,000 metric
tons / day
97
280,332
92,002
97.5%
97.2%
Note: Clinker production capacity was based on 2004 plant level data published by Portland Cement Association.
Source: Summarized from EPA Draft Cement Database (2007).
Table 4.3 below illustrates the number of facilities exceeding the annual clinker production
capacity threshold levels of 10,000 metric tons, 25,000 metric tons, 50,000 metric tons and
100,000 metric tons and associated clinker production and emissions coverage in 2006. The
results indicate that even at 100,000 metric tons of clinker production capacity per year
threshold, nearly all of the integrated facilities will be reporting and that nearly all of industry
emissions will be covered.
Table 4.3 Summary of the Impact of Various Annual Clinker Production Capacity Thresholds on
the Effectiveness of the Reporting Rule
Number of cement facilities exceeding the threshold
Total Clinker Production Capacity (1,000 metric tons /
year) by facilities exceeding the threshold
% of Clinker Production Covered by facilities
exceeding the threshold
% of Total CO2e Emissions Covered by the facilities
exceeding the threshold
Threshold annual clinker production capacity
10,000
metric tons
/year
107
94,385
100.0%
100.0%
25,000
metric tons
/year
107
94,385
100.0%
100.0%
50,000
metric tons
/year
107
94,385
100.0%
100.0%
100,000
metric tons
/year
106
94,288
99.9%
99.9%
Note: Clinker production capacity was based on 2004 facility level data published by Portland Cement Association.
Emissions coverage corresponds to 2006 emissions calculated by scaling 2005 clinker production and emissions
developed in the EPA's Draft Cement Database. They include both combustion and process related emissions and
CH4 and N2O emissions.
Source: Summarized from EPA Draft Cement Database (2007).
15
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
4.2.3 All Clinker Production Facilities
Table 4.4 illustrates the number of integrated cement facilities and the total emissions and clinker
production capacity attributable to these facilities in 2006. All of these facilities and their
associated emissions would be included under a rule requiring all clinker production facilities to
report emissions.
Table 4.4 Summary of the Impact of Including All Clinker Production Facilities on
the Effectiveness of the Reporting Rule
Facility Type
Number of integrated cement facilities
Total Annual Clinker Production Capacity (1,000 metric tons / year)
Process-related CO2e Emissions covered by the Integrated Facilities (million metric tons
ofCO2e)
Total CO2e Emissions covered by the Integrated Facilities (million metric tons of CO2e)*
% of Total Annual Clinker Production Capacity covered by these facilities
% of process related CO2 Emissions covered by these facilities
% of Total CO2e Emissions covered by the Integrated Facilities*
Integrated Cement
Facilities
107
94,385
45.7
86.8
100%
100%
100%
Note:* Includes both combustion and process-related GHG emissions. Includes CH4 and N2O emissions from
stationary fossil fuel combustion in the cement facilities. Clinker production capacity was based on 2004 facility
level data published by Portland Cement Association. Total emissions and emissions coverage correspond to 2006
emissions calculated by scaling 2005 production and emissions developed in the EPA's Draft Cement Database. The
process related emission estimate was obtained from the U.S. GHG Inventory, referenced below.
Source: (1) Based on EPA Draft Cement Database; (2) Inventory of U.S. Greenhouse Gas Emissions and Sinks:
1990-2006, Final. U.S. Environmental Protection Agency, Washington, DC. April 15, 2008.
5. Options for Monitoring Methods
This section includes review of existing CC>2 emission estimation and monitoring methodologies,
recommended by relevant reporting programs for monitoring / measuring CC>2 emissions from
cement production processes.
Three common CC>2 emissions estimation methods emerge from the review of various
monitoring and measurement protocols for calculating and reporting process-related CC>2
emissions from cement production. They are: (i) cement-production based method, (ii) clinker-
production based method and (iii) non-fuel raw material-consumption based method. The cement
production based method tends to introduce a large amount of uncertainty in the CC>2 emission
estimates calculated. This method is not consistent with good practice (IPCC 2006). The other
two methods that are consistent with good practice (IPCC 2006) and have relatively lower level
of uncertainty are:
Calculation of process-related CO 2 emissions based on clinker-production, after
adjusting for CO 2 emissions attributable to CKD. Some protocols extend this IPCC
(2006) Tier 2 method by adjusting for the CO2 emissions attributable to bypass dust and
total organic carbon in raw materials.
16
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
Calculation of process-related CO 2 emissions based on the quantity, type and
composition of raw material consumed (including their carbonates content) and the
extent ofuncalcined carbonates leaving the kiln. This is similar to IPCC (2006) Tier 3
method.
The stringency of CC>2 monitoring methods varies across the existing monitoring programs
reviewed, depending on the method adopted for calculating emission factors and the stringency
of activity data measurement.
At the facility-level, process-related CC>2 emissions from cement production can be monitored
directly using emission measurement equipment or using one of the calculation methods
specified above. For purposes of this reporting program, the following three alternative emission
monitoring / measurement options were analyzed.
5.1 Option 1: Direct Measurement (Annual Reporting)
In cement facilities, where process emissions and/or combustion GHG emissions are contained
within a stack or vent, direct measurement constitutes either measurements of individual GHG
concentration in the stack gas and the flow rate of the stack gas using a Continuous Emissions
Monitoring System (CEMS), or periodic measurement of the individual GHG concentration in
the stack gas and the flow rate of the stack gas using periodic stack testing. Under either a
CEMS approach or a stack testing approach, the emissions measurement data would be reported
annually.
Elements of a CEMS include a platform and sample probe within the stack to withdraw a sample
of the stack gas, an analyzer to measure the concentration of each GHG pollutant (e.g., CO2) in
the stack gas, and a flow meter within the stack to measure the flow rate of the stack gas. The
emissions are calculated from the concentration of the specific GHG pollutant in the stack gas
that is monitored using the concentration monitor for that pollutant and the flow rate of the stack
gas. The CEMS continuously withdraws and analyzes a sample of the stack gas and
continuously measures each of the individual GHG pollutant concentration and flow rate of the
stack gas.
For direct measurement using stack testing, sampling equipment would be periodically brought
to the site and installed temporarily in the stack to withdraw a sample of the stack gas and
measure the flow rate of the stack gas. Similar to CEMS, for stack testing the emissions are
calculated from the concentration of GHGs in the stack gas and the flow rate of the stack gas.
The difference between stack testing and continuous monitoring is that the CEMS data provide a
continuous measurement of the emissions while a stack test provides a periodic measurement of
the emissions.
5.2 Option 2: Hybrid Method (Annual Reporting)
Under a hybrid method, facilities that already have CEMS installed for other purposes could be
required to configure their CEMS to monitor CC>2 from their stacks. Facilities that have CEMS
17
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
installed for measuring NOX and/or SCh, could also be required to install a CC>2 monitor and test,
quality assure and operate the system as required under Part 75, Continuous Emissions
Monitoring rule, in Volume 40 of the Code of Federal Regulations (CFR). Facilities that already
have CC>2 CEMS could be required to test, quality assure and operate the system as required
under Part 75.
Under a hybrid approach, facilities that do not have CEMS could calculate their process-related
emissions using a facility-specific non-fuel raw material input consumption-based approach.
This mass-balance approach is conceptually similar to the Tier 3 method of the 2006 IPCC
Guidelines and Method A (Kiln Input Approach) recommended under the EU ETS for the
second reporting period. Under this method, cement facilities would calculate their raw material
specific carbonate contents, total organic carbon (TOC) content and emission factors. Cement
facilities again readily assess carbonate contents of raw material inputs with every batch.
Under this approach, facilities would establish facility-specific weighted average annual
emission factors, carbonates content, and non-carbonate carbon content, based on the weighted
average of sample chemical analysis conducted specified number of times (such as on a quarterly
basis) during the year. Cement facilities may, however, analyze these factors more frequently
either on a daily basis or every time the batch of raw material mix changes.
The CC>2 emissions attributable to uncalcined carbonate within CKD not reused in the kiln would
be subtracted from the CC>2 attributable to carbonates contained in the raw materials consumed.
This would require developing facility-specific CKD emission adjustment factor or using a
default factor of 1, which assumes that 100% of all carbonates in CKD is calcined. To establish a
CKD adjustment factor, facilities could conduct chemical analysis with sufficient frequency to
estimate the facility-specific fraction of uncalcined carbonate in CKD that is not recycled to the
kiln.
Techniques for measuring inputs and clinker
In the laboratories of cement facilities, chemical analysis is regularly performed as a quality
control tool to ensure the clinker, and in turn, resulting cement, meets the required standards
(Cement Americas 2001). In order to maintain the desired composition of clinker output, the
carbonate contents and non-carbonate total organic carbon content in the raw materials are
monitored and controlled within appropriate ranges. Clinker that does not meet the specified
standard cannot be used for cement production, but must be recycled back to the kiln or
discarded. X-ray Fluorescence (XRF) analysis is the most widely used technique for determining
the chemical composition of raw materials and products (such as clinker). Further, for other
environmental considerations, facilities may also use thermal analysis to screen raw materials
and kiln feed to evaluate their potential to contribute to volatile organic compounds and
particulate matter emissions (Cement Americas 2001).
18
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
Method used to measure CKD
CKD production depends on kiln configuration, raw materials and fuel(s) used in clinker
production, and process characteristics (Gebhardt 1999). CKD primarily consists of raw
materials contaminated with volatile organic compounds, such as sulfur, alkalis and chlorides
(Kessler 1995). Depending on its composition, some CKD is recycled back to the kiln, some
CKD is recycled into cement, and some CKD must be discarded. However to the extent possible,
the CKD is recycled. As a quality control measure, cement facilities measure the composition of
their CKD on a regular basis. However, for purposes of calculating CO2 emissions using the non-
CEMS method, facility-specific uncalcined carbonate content in CKD that is not recycled back
to the kiln can be estimated through chemical analysis.
Method used to measure TOC
The CO2 emissions from the non-carbonate total organic carbon (TOC) in the raw materials
entering the kiln can be determined using chemical analysis. Alternatively similar to other
protocols (California AB32), a default factor can be used in the place of chemical analysis to
determine the non-carbonate TOC in the raw materials. An example default factor might be two-
tenth of one percent of the total raw material weight.
5.3 Option 3: Simplified Emission Calculation Method
This method requires measuring the clinker production and CKD production, something cement
facilities do as a part of regular business operations.
6. Options for Estimating Missing Data
Procedures for estimating missing data provide reporting facilities methods to use substitute data
that are close to the actual values. These procedures differ by the type of monitoring method
adopted. Appropriate missing data procedures are described below.
6.1 Direct Measurement
Alternative direct emissions measurement methods include measuring CO2 emissions using
CEMS and through stack testing. The missing data procedures for these two measurement
options are described below.
6.1.1 Continuous Emissions Monitoring System (CEMS)
For direct measurement of CO2 flow rates or direct measurement of CO2 emissions using CEMS,
Part 75 establishes procedures for substituting missing data. In general, missing data from
operation of the CEMS may be replaced with substitute data to determine the CO2 flow rates or
CO2 emissions during the period in which CEMS. Two different missing data procedures
"initial" and "standard" routineare described in Part 75. For both the initial and missing data
procedures, the appropriate substitute values are calculated and applied automatically by the
19
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
Data Acquisition and Handling System (DAHS), which is a component of the CEMS (EPA
2005a).
The initial missing data procedures are used until a certain number of quality assured CEMS data
are obtained. For CC>2, initial missing data procedures are applicable during the initial 720 hours
of operation of the CEMS unit after the initial certification at the site. The initial procedures are
simple procedures, such as average of the measured CEMS data before and after the period
during which the data were missing, or arithmetic average of historic data for similar operating
conditions.
Under Part 75, standard missing data procedures must be applied (i) after the initial 720 hours of
quality assured CEMS data have been obtained (prior to the lapse of three year period) after
obtaining initial certification or (ii) after the lapse of three year period after obtaining initial
certification, even if 720 hours of quality assured CEMS data have not been obtained. The
standard missing data procedures are nearly identical to initial missing data procedures even after
720 hours of operation of the CEMS, if the percent monitor data availability (PMA), which the
ratio of total number of hours of quality assured CEMS data to the total number of hours the
monitor operated, is 95 or more (i.e., >95%) and the period during which the data are missing
does not exceed 24 hours. These substitution data procedures are as close to the actual data as
possible (without any punitive component). PMA estimates are calculated hourly by the DAHS.
As a part of 2002 revisions to Part 75, EPA added new provisions that allow sources to
implement the standard missing data procedures for purpose of obtaining more representative
substitute data values. For example, affected sources burning different types of fuels are given
the option to separate their historical CEM data according to fuel type and to apply the standard
missing data procedures on a fuel-specific basis. Part 75 also accommodates using substitute data
value for each missing data hour, based on the appropriate operational characteristics of that
time.
In the case of a mandatory GHG reporting program, however, it would be necessary to ensure
that the missing data procedures are as accurate as a possible to the actual missing data and that
they are neither conservative nor liberal. The missing data procedures from Part 75 described
above are consistent with the objective of the mandatory reporting program, which seeks to
develop accurate inventory of GHG emissions from each affected facility.
Some of the missing data procedures established in Part 75 are, however, not consistent with the
objective of the mandatory GHG reporting program. For example, for PMAs less than 95%, the
standard missing data procedures under Part 75 become increasingly conservative to ensure that
emissions are not under-reported in a cap-and-trade program. These conservative emission
reporting procedures result in over estimates of emissions and, therefore, are not consistent with
a non-cap-and-trade, mandatory GHG reporting program.
6.1.2 Stack Testing
For measuring CC>2 emissions using stack testing, missing data procedures are not generally
anticipated. EPA's Clean Air Act National Stack Testing Guidance (EPA 2005b) document
20
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
contains the quality assurance guidance for collecting appropriate data based on pre-established
site-specific test plan. For more details, refer to EPA 2005b.
6.2 Facility-specific Emission Calculation
Procedures for estimating missing data do not apply to CC>2 process emissions calculated based
on clinker production and plant specific emission factors. If the carbonate content analysis
required to calculate the plant specific emission factor are missing or invalid, facilities could
undertake a new chemical analysis. Similarly, missing or invalid CKD analysis or organic
carbon analysis could be replaced by new analysis.
There is no reason for a plant to be missing information on clinker production. Clinker
production is measured with redundant systems at multiple locations within the plant; for
example at the clinker cooler exit, at the inlet to the clinker storage silos, and /or between the
storage silos and the finish mill. The blending of clinker and other feed materials to manufacture
cement and the dispatch of finished cement is also tightly monitored and can be used to
accurately measure the production of clinker.
7. QA/QC Requirements
Facilities could perform quality assurance and quality control (QA/QC) procedures and self data
verification procedures of the measurement of actual emissions and/or of all the input and output
data used in the calculation of emissions.
As part of the data verification requirements, the owner or operator could submit a detailed
explanation of how company records of measurements are used to quantify all sources of carbon
input and output within receipt of a written request from EPA or from the applicable State or
local air pollution control agency.
Monitoring method-specific QA/QC and verification procedures are described below.
7.1 Direct Measurement
In the case of CEMS being used for monitoring emissions, the applicable QA/QC procedures
prescribed under Part 75 that are related to ensuring the accuracy of the CC>2 concentration
monitor, the flow meters that measure the flue gas flow rate, and the DAHS that record the
CEMS measurements and compute the emissions, heat rate and other relevant information could
be adopted. In the case of stack testing, the QA/QC procedures described in the EPA's Clean Air
Act National Stack Testing Guidance document (EPA 2005b) could be followed.
Part 75 established specific tests and testing frequency for the individual components of the
CEMS systems to quality assure the data obtained by CEMS. Examples of these checks include:
Daily calibration error test and daily interference check for flow monitors
Quarterly leak check for differential pressure-type flow monitors
21
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
Annual or Semi-annual Relative Accuracy Test Audit (RATA) and Bias test for flow
monitors and for pollution concentration monitors
Primary elemental visual inspection once every three years for Orifice and nozzle
See Part 75 for details on test specification, test acceptance criteria, and exceptions for the
various types of QA checks recommended.
Some of the key good practice QA/QC and data verification procedures for maintenance of direct
emissions measurement equipment, based on Acid Rain and other relevant programs, are
summarized below.
Conduct regular maintenance of monitoring equipment, such as CC>2 concentration
monitors, flow meters and sampling probes
Keep a written record of procedures needed to maintain the monitoring system in proper
operating condition and of a schedule for implementing those maintenance procedures. In
other words, develop a facility and equipment specific maintenance procedure manual(s)
and regular maintenance implementation schedule(s).
Keep a maintenance log to record of all testing, maintenance, or repair activities
performed on any monitoring system or component in a location and format suitable for
inspection. The maintenance log must include: date, time, and description of any testing,
adjustment, repair, replacement, or preventive maintenance action performed on any
monitoring system and records of any corrective actions associated with a monitor's
outage period. Further, details of any resultant changes to the system's ability to record
and report emissions data (such as changes to the flow monitor, changing of temperature
and pressure coefficients and dilution ratio settings) must be recorded and maintained at
the facility, long with written explanations of the procedures used to make the
adjustment(s).
7.2 Facility-specific Emission Calculation
The Option 1 above describes the QA/QC and data verification procedures for CEMS.
For non-CEMS methods involving facility-specific calculations using kiln input-based mass-
balance approach, the following procedures, which were adopted from the existing available
reporting programs (such as Acid Rain, EU ETS and Climate Leaders programs), could be used.
Check the accuracy of the emissions calculations by step to ensure that computational
errors are avoided. Some emission calculation checks that can avoid errors include: (a)
checking the units of measurement for different variables and parameters used in the
calculations for correctness and for consistency with the emissions calculated; (b)
checking the conversion factors used in the calculations; and (c) checking the aggregation
of data by emission sources (e.g., kilns), product type, and/or production batch number.
22
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
Assess the representativeness of carbon and carbonate contents of raw materials
determined through laboratory analysis, by comparing those values with IPCC or other
available default values
Check for consistency in production data, carbon content data, and emission estimate
across time periods. Outliers must be capable of being explained by changes in the
facility's operational conditions. If the differences across the annual data cannot be
attributed to changes in (a) production levels, (b) type and amounts of fuels and raw
materials used, and/or (c) emission generation process (for example, due to energy
efficiency improvements), the plausibility of measurement error must be assumed,
investigated and corrected for.
Compare the emission estimate with prior years' emissions estimates and the national
emissions estimate(s) for the industry and evaluate if the differences in the underlying
input data factors (such as carbonate and carbon contents, quantity and types of raw
materials used and quantity of CKD generated and recycled) are consistent with the
differences in the emissions estimates. If the differences in emissions estimates could not
be explained reasonably, plausibility of monitoring error should be assumed and further
efforts must be undertaken to verify the calculations and rectify any error in the source
data.
Maintain comprehensive documentation of the data used and the methodology adopted,
including data received through personal communication. The chemical analysis used for
developing carbonate contents and non-carbonate TOC content of raw materials must be
well documented and the data accurately recorded. The methodology used to derive the
average quarterly emission factors and other parameters must be well documented and be
made available for audit when needed.
8. Types of Emissions Information to be Reported
To ensure completeness, cement production facility owners or operators would report annual
GHG emissions from cement production including both combustion-related CC>2, CH4 and N2O
emissions and process-related CC>2 emissions.
Additional data for verification could include process raw material and product feed rates and
carbon contents. Such data listed below would illustrate the process operating conditions at
which the emissions monitoring data were obtained. EPA could use such data, for example to
check the reported emissions against activity-data-based emission factors for the process.
quantity of clinker produced
quantity of raw material consumed, by type of raw material
carbonate contents of raw materials by type of raw materials obtained through laboratory
analysis of sample data
carbon content of non-fuel raw materials, by type of raw material used
quantity of CKD produced
quantity of CKD recycled and disposed of
23
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
carbonate content of the uncalcined CKD obtained through laboratory analysis of
sampled data
CC>2 estimation methodology adopted
Initial data reporting could include submission of facility-specific information (including the
name and location of the facility, type of facility, number of emissions sources, production and
other unit characteristics, primary fuels and raw materials typically used, name of and contact
information for the designated person authorized to represent the facility and to certify and
submit its emissions reports and respond to questions), a monitoring plan, results of the
monitoring system certification tests, stack testing plan, and facility-specific emission
calculation methodology, depending on the type of the monitoring system adopted.
8.1 Additional Data to be Retained Onsite
Facilities could be required to retain data concerning monitoring of GHG emissions onsite for a
period of at least three years from the reporting year. For CEMS, these data could include
CEMS monitoring system data, including continuous-monitored GHG concentrations and stack
gas flow rates, calibration and quality assurance records. For stack testing these data could
include stack test reports and associated sampling and chemical analytical data for the stack test.
Process data including process raw material and product feed rates and carbon contents could
also be required to be retained on site for a period of at least three years from the reporting year.
EPA could use such data to conduct trend analyses and potentially to develop process or activity-
specific emission factors for the process.
24
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
9. References
Cement Americas. 2001. "A Cement Plant Toolbox to Simplify Your Life,"
by Linda M. Hills, Ella Shkolnik, F. MacGregor Miller and Vagn Johansen, Cement
Americas. Jul 1, 2001 12:00
http ://cementamericas. com/mag/cement_cement_plant_toolbox/
EPA 2005a. "Plain English Guide to Part 75 Rule." Clean Air Market Division, U.S.
Environmental Protection Agency, Washington, DC.
http://www.epa.gov/airniarkt/emissions/docs/plain_english_guide jart75_rule.pdf
EPA 2005b. "Clean Air Act National Stack Testing Guidance." Office of Enforcement and
Compliance Assurance, U.S. Environmental Protection Agency. September 30.
http://www.epa.gov/conipliance/resources/policies/monitoring/caa/stacktesting.pdf
EPA 2007. "Plant-level Cement GHG Database: Revised Draft," prepared by ICF for the
Program Integration Branch, Climate Change Division, U.S. Environmental protection
Agency, May 2007.
EPA2008a. Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2006, Final. U.S.
Environmental Protection Agency, Washington, DC. April 15.
http://www.epa.gov/climatechange/emissions/iisinventonfreport.htmltf
EPA 2008b. Cement process-related CC>2 emissions calculation methodology: A review of
existing reporting programs," Cement_4 4 08.xls, Program Integration Branch, Climate
Change Division, U.S. Environmental Protection Agency. April.
Federal Highway Administration (FHWA 2008). "Portland Cement." Department of
Transportation, 2008 http://www.fhwa.dot.gov/infrastructure/materialsgrp/cement.html
Gebhardt, Ronald Frank 2001. "Method of treating cement kiln dust for recovery and recycle,"
U.S. Patent No. 6331207, Patent Issued on December 18, 2001.
http ://www. patentstorm.us/inventors-patents/Ronald_Frank_Gebhardt/1712631/1. html
Holcim 2008a. "Emission Monitoring and Reporting Standard," Holcim U.S.
http://www.holcim.us/USA/EN/id/56785/mod/home/page/editorial.html:
Holcim 2008b. "Locations in the United States," Holcim U.S.
http://www.holcim.us/US A/EN/id/536892814/mod/gnmO/page/location_map.html.
IPCC 2006. 2006IPCC Guidelines for National Greenhouse Gas Inventories. Volume 3:
Industrial Processes and Product Use. Intergovernmental Panel on Climate Change.
Kessler, Grant R. 1995. "Cement Kiln Dust (CKD) Methods for Reduction and Control, " IEEE
Transactions on Industry Applications, Vol.31, No.2, March/April. Abstract:
http://ieeexplore.ieee.org/xpl/freeabs all.isp?tp=&arnumber=370292&isnumber=8490
25
-------
Technical Support Document for Cement Manufacturing: Proposed Rule for Mandatory Reporting of Greenhouse Gases
PCA 2006. U.S. and Canadian Portland Cement Industry: Plant Information Summary (as of)
December 31, 2004. Portland Cement Association.
PCA 2008."How Portland Cement is Made" Internet site of Portland Cement
Association, http ://www. cement. org/basics/howmade. asp
USGS 2007. "Mineral Commodities Summary: Cement."
http://minerals.usgs.gov/minerals/pubs/commodity/cement/cemenmcs07.pdf
van Oss, Hendrick 2004. ICF staff personal conversation with Hendrick van Oss of the United
States Geological Survey, October 19, 2004.
26
------- |