United States    s  Office of Air Quality       EPA-450/3-85-003a
           Environmental Protection  Planning and Standards     May 1985
           Agency        Research Triangle Park NC 27711

           Air
&ERA     Portland Cement
           Plants-
           Background
           Information for
           Proposed Revisions
           To Standards

-------
                                  EPA-450/3-85-003a
             Portland Cement-
Background Information for Proposed
          Revisions to Standards
             Emission Standards and Engineering Division
            U.S. ENVIRONMENTAL PROTECTION AGENCY
                  Office of Air and Radiation
             Office of Air Quality Planning and Standards
                Research Triangle Park, NC 27711
                      May 1985     ^ $ Environmental Protection Agency
                                Region 5, Library (PL-12J)
                                77 West Jackson Boulevard, 12th Floor
                                Chicago, IL 60604-3590

-------
This report has been reviewed by the Strategies and Air Standards Division of the Office of Air Quality
Planning and Standards,  EPA,  and approved for publication.  Mention of trade names or commercial
products is not intended to constitute endorsement or recommendation for use. Copies of this report are
available through the Library Services Office (MD-35), U.S. Environmental Protection Agency, Research
Triangle Park, N.C. 27711,  or from National Technical Information Services, 5285 Port Royal Road,
Springfield, Virginia 22161.

-------
                             TABLE  OF  CONTENTS
                                                                       Page

 LIST  OF  FIGURES	     iv

 LIST  OF  TABLES	     v

 CHAPTER  1.   EXECUTIVE  SUMMARY   	     1-1
      1.1  Regulatory History of Current Standards   	     1-1
      1.2  Industry Trends	     1-2
      1.3  Control Technology 	     1-2
      1.4  Compliance Test Data	     1-3
      1.5  Cost Considerations Affecting the NSPS	     1-4
      1.6  Enforcement  Aspects   	     1-4

 CHAPTER  2.   INDUSTRY DESCRIPTION  	     2-1
      2.1  Introduction	     2-1
      2.2  Process Description   	               2-1
           2.2.1  Raw Material Handling 	     2-1
           2.2.2  Clinker Production  	     2-2
           2.2.3  Cement Manufacture and Shipment 	     2-5
      2.3  Industry Characterization  	     2-6
           2.3.1  Geographic Distribution 	     2-6
           2.3.2  Production	     2-6
           2.3.3  Growth Trends	     2-7
           2.3.4  Process Developments  	     2-7
      2.4   Emissions From Portland Cement Plants  	     2-9
           2.4.1  Particulate Emissions 	     2-9
           2.4.2  Sulfur Oxide Emissions  	     2-9
           2.4.3  Nitrogen Oxide Emissions  	     2-10
      2.5   References for Chapter 2	     2-31

CHAPTER  3.  CURRENT STANDARDS FOR PORTLAND CEMENT PLANTS 	    3-1
      3.1   New Source Performance Standards 	    3-1
           3.1.1  Summary of New Source Performance Standards ...    3-1
           3.1.2  Testing and Monitoring Requirements 	    3-2
           3.1.3  Recordkeeping and Reporting Requirements  ....    3-2
      3.2   State Regulations  	    3-3
     3.3   References for Chapter 3	    3-4

CHAPTER 4.  CONTROL TECHNOLOGY AND COMPLIANCE TEST RESULTS ....    4-1
     4.1  Available Particulate Control  Technology 	    4-1
          4.1.1  Kiln	    4-1
          4.1.2  Clinker Cooler	    4-8
          4.1.3  Other Facilities   	    4-9
     4.2  Summary of Particulate Compliance Test Results 	    4-9
          4.2.1  Kiln	    4-9
          4.2.2  Clinker Cooler	                 4-11
          4.2.3  Other Facilities   	    4-13
                                    m

-------
                      TABLE OF CONTENTS (continued)
     4.3  Available Gaseous Pollutant Technology 	    4-13
          4.3.1  Sulfur Dioxide	    4-13
          4.3.2  Nitrogen Oxides	    4-22
     4.4  References for Chapter 4	    4-23

CHAPTER 5.  COST ANALYSIS	    5-1
     5.1  Approach	    5-1
     5.2  Estimated Capital and Annualized Costs of
          Emission Control 	            5-1
          5.2.1  Kiln	    5-2
          5.2.2  Clinker Cooler	    5-2
          5.2.3  Other Facilities  	    5-2
     5.3  Comparison of Estimated and Reported Capital Cost
          Data	    5-3
     5.4  Cost Effectiveness	    5-3
5.5  References for Chapter 5	         5-12

CHAPTER 6.  ENFORCEMENT ASPECTS  	    6-1
     6.1  Varied Exhaust Gas Ducting Configurations  	    6-1
     6.2  Bypass of Electrostatic Precipitators  	    6-2
          6.2.1  CO Trips	    6-2
          6.2.2  Kiln Startup and Shutdown	    6-5
     6.3  Continuous Opacity Monitors  	    6-6
     6.4  -Recordkeeping and Reporting Requirements 	    6-6
     6.5  References for Chapter 6	    6-7

APPENDIX A.   SUMMARY OF PORTLAND CEMENT FACILITIES SUBJECT
     TO NSPS	                 A-l
APPENDIX B.   SUMMARY OF STATE REGULATIONS FOR PORTLAND CEMENT
     PLANT FACILITIES	               B-l
APPENDIX C.   PARTICULATE EMISSIONS AND OPACITY DATA FOR
     FACILITIES SUBJECT TO THE NSPS SINCE THE 1979 REVIEW  ....    C-l
                                   IV

-------
                             LIST OF FIGURES

                                                                      Page
Figure 2-1   Typical Wet Process Material Handling 	    2-11
Figure 2-2   Typical Dry Process Material Handling 	    2-12
Figure 2-3   Typical Clinker Production Process  	    2-13
Figure 2-4   Four-Stage Suspension Preheater With a Precalciner  .    2-14
Figure 2-5   Traveling Grate Preheater System  	    2-15
Figure 2-6   Finish Mill Grinding and Shipping 	    2-16
Figure 2-7a  Portland Cement Plant Locations—Western U. S.  ...    2-17
Figure 2-7b  Portland Cement Plant Locations—Eastern U. S.  ...    2-18
Figure 2-8   Fuel Consumption Per Ton of Clinker Produced by Fuel
             Type and Clinker Production Process 	    2-19
Figure 2-9   Number of Plants Using Wet or Dry Clinker
             Production Process  	    2-20
Figure 2-10  Kiln Construction by Year	•	    2-21
Figure 2-11  Detail of Roller Mill  That Combines Crushing,
             Grinding, Drying,  and Classifying in One  Vertical
             Unit	     2-22
Figure 2-12  Particle Size Distribution of Cement Dust 	     2-23
Figure 4-1   Particulate Mass Emissions From Kilns  That Have
             Become Subject to the  NSPS Since 1979	     4-10
Figure 4-2   Particulate Mass Emissions From Clinker Coolers That
             Have Become Subject to the NSPS Since  1979	     4-12
Figure 4-3   S02 Emissions  Versus Sulfur in  the Coal	     4-20

-------
                             LIST OF TABLES
Table 2-1  U.S. Clinker Production, Kiln Capacity, and Capacity
           Utilization—1972-1982  	    2-24

Table 2-2  Facilities Subject to the NSPS Since 1979 Review  . .  .    2-25

Table 2-3  S02 Emission Test Results for Portland Cement
           Facilities That Have Become Subject to the NSPS
           Since 1979	    2-28

Table 2-4  NO  Emission Test Results for Portland Cement
           Facilities That Have Become Subject to the NSPS
           Since 1979	    2-29

Table 4-1  Potential Sources of Particulate Emissions and
           Typical Control Practices 	    4-2

Table 4-2  Particulate Control Technology Practices at Plants
           With Facilities That Have Become Subject to the NSPS
           Since the 1979 Review	    4-3       ~

Table 4-3  Summary of Carbon Monoxide Trip Data for Electrostatic
           Precipitators on Kilns That Have Become Subject to the
           NSPS Since 1979	    4-7

Table 4-4  S02 Emissions From Lone Star Industries, Incorporated .    4-16

Table 5-1  Summary of Model Kiln Facility Parameters 	    5-4

Table 5-2  Estimated Capital and Annualized Costs of
           Particulate Emission Control Equipment for Model Kiln
           Facilities	    5-5

Table 5-3  Summary of Model Clinker Cooler Facility Parameters .  .    5-6

Table 5-4  Estimated Capital and Annualized Costs of Particulate
           Emission Control Equipment for Model Clinker Cooler
           Facilities	    5-7

Table 5-5  Summary of Parameters for Model  Other Facilities  ....  5-8

Table 5-6  Estimated Capital and Annualized Costs of
           Particulate Emission Control Equipment for Model Other
           Facilities	    5-9

Table 5-7  Comparison of Estimated Capital  Costs of Emission
           Control with Reported Capital Data Costs (From
           Industry)	    5-10
                                   VI

-------
                       LIST OF TABLES (continued)
Table 5-8  Cost Effectiveness of Participate Emission Reduction by
           Model Facilities	     5-11

Table A-l  Summary of Portland Cement Facilities Subject
           to NSPS	     A-l

Table B-l  Summary of State Regulations for Portland Cement Plant
           Facilities	     B-l

Table C-l  Particulate Emissions and Opacity Data for Facilities
           Subject to the NSPS Since the 1979 Review	     C-l
                                   vn

-------
                          1.  EXECUTIVE SUMMARY

     The Clean Air Act Amendments of 1977 require that the U. S.
Environmental Protection Agency (EPA) review and, if appropriate, revise
new source performance standards (NSPS) every 4 years.  This report
presents information on developments that have occurred in the port!and
cement industry since the last review of the standards in 1979.

1.1  REGULATORY HISTORY OF CURRENT STANDARDS

     The NSPS for the portland cement industry were proposed on August 17,
1971, promulgated by EPA on December 23, 1971, and revised in response
to a court remand on November 12, 1974 (40 CFR 60, Chapter I, Subpart F).
The standards apply to kilns, clinker coolers, raw mill systems, finish
mill systems, raw mill dryers, raw material storage areas, clinker
storage areas, finished product storage areas, conveyor transfer points,
bagging, and bulk loading and unloading systems that had begun
construction or modification on or after August 17, 1971.

     The standards prohibit the discharge into the atmosphere from any
kiln, exhaust gases which:

     1.   Contain particulate matter in excess of 0.15 kilograms (kg) per
megagram (Mg) of feed (dry basis) to the kiln or 0.30 pounds (lb) per
ton of feed to the kiln, or
     2.   Exhibit greater than 20 percent opacity.

The standards prohibit the discharge into the atmosphere from any clinker
cooler,  exhaust gases which:

     1.   Contain particulate matter in excess of 0.05 kg/Mg of feed (dry
basis) to the kiln (0.10 lb/ton), or
     2.   Exhibit 10 percent opacity or greater.

Finally, the standards prohibit the discharge into the atmosphere from
any affected facility other than the kiln or clinker cooler, exhaust
gases which exhibit 10 percent opacity or greater.

     The first review of the standard, published in 1979,  recommended
that no  changes be made to the particulate mass  or the visible emission
limits.   A recommendation to require opacity monitoring was made.   In
addition, it was recommended that a monitoring program be  initiated to
determine nitrogen oxide (NO ) and sulfur dioxide (S02) emission rates
                                    1-1

-------
 for  kilns that  have become subject to the  NSPS and  that  research  and
 development be  funded to determine means of  reducing  NO   emissions  from
 kilns.                                                 x

     The following sections  summarize the  results and conclusions of  the
 second  review of the NSPS for portland cement plants.

 1.2  INDUSTRY TRENDS

     Since the  1979 review,  37 cement plants have added,  reconstructed,
 or modified facilities so as to bring them under the  NSPS  for portland
 cement  plants.  Fourteen plants have installed all  new facilities (i.e.,
 kilns,  clinker  coolers, and  other associated equipment such as mills  and
 storage and transfer facilities), and the  remainder have  added new  kiln
 capacity and/or other equipment.

     Ninety-two percent of the kilns built since the  1979  review use  the
 dry  process of  cement production instead of the wet process because the
 dry  process is  more fuel efficient.  The fuel efficiency  of the dry
 production process can be increased further by adding a  preheater, which
 uses the kiln exhaust gases  to preheat the raw feed,  or  by combining  a
 preheater with  a precalciner to preheat and partially precalcine the  raw
 feed prior to the kiln.  Of  dry process kilns built since 1979, 17 percent
 use  a preheater system, and  79 percent use a preheater/precalciner
 system.

     Fuel efficiency can be  improved also  by directing all or a portion
 of the exhaust gases from the kiln, the preheater (if one exists), or
 the  clinker cooler through the raw mill  prior to a control device for
 further heat exchange between the gases and the raw feed material.
 Twenty percent  less energy was needed to produce I Mg (1.1 ton) of clinker
 in 1982 than was needed in 1972.

 1.3  CONTROL TECHNOLOGY

     Fabric filters or electrostatic precipitators are used to control
 emissions from portland cement kilns.   Compliance with the particulate
mass and visible emission standards has been demonstrated using either
control device.

     At 28 plants with one or more kilns that have become subject to the
NSPS since the 1979 review,  17 kilns are controlled by fabric filters,
and 13 kilns are controlled by electrostatic precipitators (3 kilns at
one plant are controlled by one electrostatic precipitator).

     Fabric filters most commonly control   emissions from clinker coolers.
Of 23 clinker coolers subject to the NSPS  since the 1979 review,  17 are
controlled by fabric filters, 2 are controlled by electrostatic
precipitators, and 4 are controlled by gravel bed filters.

     Other affected facilities at cement plants are typically controlled
by fabric filters;  however,  two finish mills are controlled by
electrostatic precipitators.


                                    1-2

-------
      Air pollution control agency personnel expressed concern that
 excess particulate emissions from kilns controlled by electrostatic
 precipitators were occurring during periods of carbon monoxide (CO)
 trips.  Because a spark source is present, electrostatic precipitators
 used to control kiln particulate mass emissions are equipped with
 combustibles or CO monitors that de-energize the electrostatic
 precipitator if preset levels are reached that may present an explosion
 hazard.   Carbon monoxide trips last from less than a minute to more than
 20 minutes and may occur from a few times per year to more than 600 times
 per year.   Annually, particulate emissions resulting from such trips can
 be significant.

      Emission test data from 19 cement kilns show that kilns can be
 major sources of S02 emissions.   These emissions result from both sulfur
 in the fuel  (coal) and sulfur in the raw feed material.   Both components
 can vary significantly from plant to plant.   Data and mass balance
 calculations indicate that S02 emissions are reduced by 35 to 75 percent
 in the production process;  the sulfur can be absorbed into the clinker,
 the raw feed, or the control  device dust or can be emitted as a gas.
 The EPA and  the portland cement industry have examined the use of fabric
 filters  in controlling S02  emissions as well  as the use of flue gas
 desulfurization systems as  potential  add-on control.   Data on the-amount
 of S02 emission reduction achieved by control  devices on cement kilns
 are inconclusive because many unpredictable factors affect emissions,
 such as  the  sulfur content  of the feed,  the point in the process  at
 which S02  removal  occurs,  and the relative importance of process  variables,

      Since the  1979  review  of the NSPS,  research has  been conducted on
 the emission reduction of NO  .   Although there  are  several  process
 modifications that appear to  affect NO   emissions,  additional  research
                                      /\
 is  required  to  demonstrate  control  technology  for NO   emissions.
                                                     /\
 1.4  COMPLIANCE TEST  DATA

      Thirty  kilns  have become  subject to  the NSPS  since  the  1979  review;
 however, three  of  these  are under construction,  and compliance  test data
 are  not  available.  All  of  the 27 operational  kilns that  have become
 subject  to the  NSPS since the  1979  review are  in  compliance  with  the
 NSPS  particulate mass  and visible emission  limits.  Twenty-three  clinker
 coolers  have  become subject to the  NSPS  since 1979; two are  completing
 construction, and  compliance test data are not  available.  Nineteen of
 the  twenty-one  operational clinker  coolers that  have  become  subject to
 the  NSPS since  the 1979  review are  in compliance with the particulate
 mass  limit; two of the clinker coolers, which were tested under conditions
 not  representative of  those during  normal operation, were found to
 exceed the particulate mass limit and will be retested during normal
 operation.   One clinker cooler that is in compliance with the particulate
mass  limit exceeds the visible emission limit; plant modifications are
 underway to bring the visible emissions below 10 percent opacity.  All
of the other affected facilities  (mills and storage and transfer
 facilities) have been reported to be in compliance with the 10 percent
visible emission limit.
                                    1-3

-------
1.5  COST CONSIDERATIONS AFFECTING THE NSPS

     To estimate the cost effects of the NSPS, model facility descriptions
were developed based on information from the industry.   The capital and
annualized costs for the control system for each model  plant were estimated
using guidelines in the CARD Manual and information supplied by industry.
Costs were updated to July 1983 dollars using the Chemical Engineering
Journal plant cost index.

     The cost effectiveness of controlling particulate emissions from
kilns was estimated to range from $34 to $49 per Mg ($31 to $45 per
ton). The cost effectiveness of controlling particulate emissions from
clinker coolers was estimated to range from $27 to $44 per Mg ($25 to
$40 per ton).  The cost effectiveness of controlling particulate emissions
from other affected facilities was estimated to range from $30 to $167 per
Mg ($27 to $151 per ton).

1.6  ENFORCEMENT ASPECTS

     Chapter 6 discusses Federal, State, and local air pollution control
agency personnel concerns about (1) interpretation of the mass emission
limits for various duct configurations of affected facilities, (2) the
need to bypass an electrostatic precipitator during periods of CO trips,
startups, and shutdowns, (3) monitoring requirements, and (4) recordkeeping
and reporting requirements.
                                    1-4

-------
                         2.   INDUSTRY DESCRIPTION

 2.1   INTRODUCTION

      Manufacturing of hydraulic cement  is covered by the Standard
 Industrial Classification  (SIC) code 3241, which includes plants that
 manufacture portland, natural, masonry, and pozzolan cements.  Over
 95 percent of the hydraulic cement manufactured in the United States is
 Portland cement, which consists mainly  of tricalcium silicate and dicalcium
 silicate.1,2  The portland cement production process involves three
 basic steps.  First, raw materials are  crushed and mixed.  Second, the
 mixture is heated to high temperatures  in a kiln where chemical reactions
 take  place and a rock-like substance called clinker is formed.  The
 clinker is then cooled in a clinker cooler.   Third, the cooled clinker
 is crushed, and ground gypsum or other  materials are added to'obtain the
 properties desired in the finished cement.  In the following sections of
 this  chapter, the portland cement production process is described, the
 industry is characterized, and uncontrolled emissions are discussed.

 2.2   PROCESS DESCRIPTION

 2.2.1  Raw Material Handling

      Portland cement is composed of combinations of calcium, silica,
 alumina, iron, and gypsum.   Limestone is the most common source of
 calcium, although oyster shells, chalk,  coral  rock, or aragonite are
 used  in some parts of the country.3  Limestone can also have naturally
 high amounts of clay or shale, which contain aluminum silicates or free
 silica.   For example, the mineral  components of "cement rock" limestone
 from the Lehigh Valley of Pennsylvania are so correctly proportioned
 that no additional  raw materials are required to make clinker.3  More
 commonly,  raw materials such as clay,  shale, or iron ore must be added
 to adjust the chemical  composition of the clinker.   Processing of these
 raw materials into kiln feed involves  a  quarrying and crushing phase  and
 a mixing and grinding phase.

     Limestone is usually obtained from  an open quarry located on or
 near the plant site.   An explosive such  as ammonium nitrate  and fuel  oil
 (ANFO) is  often used to quarry the limestone,  although,  in  some instances,
the materials may be quarried mechanically.   Raw materials  not quarried
at the site are typically brought  to the plant by truck or  rail  and
stored in  stockpiles near the crushing machinery.
                                  2-1

-------
     The raw materials are crushed in a primary crusher to a maximum
size of approximately 15.2 centimeters (cm) (6 inches [in.]) in diameter.4
Primary crushers may be of the gyratory, jaw, roll, or hammer type.
Secondary crushers, often hammermills, crush the rock to smaller than
2.5 cm (1 in.) in diameter.5  Crushed raw materials are stored in silos
or stockpiles.

     During the mixing and grinding phase of raw material handling, the
crushed materials are proportioned, ground so that 70 to 90 percent will
pass through a 200 mesh sieve, and then blended prior to being fed into
the kiln.3,6  Sometimes both proportioning and blending occur after the
grinding phase.  Mixing and grinding of raw materials can be done using
either a wet or a dry process.

     In the wet grinding process, ball mills or compartment mills (a
ball mill combined with a tube mill) are used, and water is added to the
mill with the crushed raw materials (see Figure 2-1).7,8  The propor-
tioned and ground raw feed is discharged from the mill as a slurry
containing from 30 to 40 percent water.9  Slurry composition is adjusted
in correcting tanks if necessary, and the slurry is then stored in a
slurry basin.  This slurry may be fed directly to the kiln or may
first be dewatered to form a cake containing about 20 percent moisture
or dried in a dryer heated by exhaust gases from the kiln or the clinker
cooler.9,10

     In the dry grinding process, ball mills, roller mills, or compart-
ment mills are also used, but the materials are ground without water
(see Figure 2-2).  Crushed raw materials are dried in the mill  itself or
in a direct-contact rotary dryer until the free moisture content is less
than 1 percent.5  Heat for the mill or dryer can be supplied by direct
firing, although it is usually supplied by recirculation of hot exhaust
gases from the kiln or clinker cooler.  If a roller mill is used, all
kiln exhaust gases can be directed through the mill for drying and
preheating; if a ball mill is used, only a portion of the exhaust gases
can be directed to the mill.11  The feedstock is typically blended using
compressed air in homogenizing silos and then stored until  the material
is fed into the kiln.9

2.2.2  Clinker Production

     Figure 2-3 presents a schematic of the basic process of clinker
production.  Raw feed (wet slurry or dry feed) is fed into the upper end
of an inclined rotary kiln and conveyed slowly toward the lower end of
the kiln by gravity and rotation of the kiln cylinder.  Kilns are fired
from the lower end so that the hot gases pass countercurrent to the
descending raw feed material.  The temperature of the feed material
increases to a maximum of about 1500°C (2700°F) during passage through
the kiln.12  The temperature increase is accompanied by a series of
physical and chemical changes:  (1) evaporation of the free water,
(2) evaporation of the combined water in the clay, (3) calcination of
the magnesium carbonate (MgC03 •* MgO + C02), (4) calcination of the
calcium carbonate (CaC03 -> CaO + C02), and (5) combination of the lime
                                  2-2

-------
 and  clay  oxides  at  the  firing  end  of  the  kiln  to  form  the  rock-like
 substance called clinker.13,14  Clinker is  comprised of  four  major
 compounds:   tricalcium  silicate  [(CaO)3 • Si02],  dicalcium silicate
 [(CaO)2  •  Si02],  tricalcium  aluminate [(CaO)3  • A1203],  and tetracalcium
 alumino-ferrite  [(CaO)4-Al203-Fe203].14

      2.2.2.1 Wet process  of clinker  production.   In the wet  process  of
 clinker production,  feed material  enters  the kiln in a wet slurry  form.
 For  a slurry containing 40 percent moisture, 2.6  megagrams (Mg)  (2.8  tons)
 of slurry feedstock  will yield 0.9 Mg (1  ton)  of  clinker.15,lg   The
 balance of. the feedstock,  about  1.7 Mg (1.8 tons),  is  lost during  clinker
 production  a-s water  vapor, carbon  dioxide,  and other volatile compounds.3
 Wet  process  kilns average  160  meters  (m)  (525  feet) in length, and
 evaporation  of moisture from tbe feed occurs in the first  20  to  25 percent
 of the kiln's length.15 Metal-chains are often hung inside the  kiln  to
 aid  in heat  transfer to the  wet  slurry and  to  help break up clumps of
 raw  materials.1-7

      2.2.2.2 Dry process  of clinker  production.  The  only difference in
 the  calcination  process between a  wet process  and a dry  process  kiln  is
 that less moisture needs to  be evaporated from dry process  feed  material.
 Because dry  kiln  feed typically contains  less  than 1 percent  moisture,
 approximately 1.6 Mg (1.8  tons) of raw feed are needed to  produce  0.9 Mg
 (1 ton) of clinker.5,15,16   Again, the remainder  of the  kiln  feed,
 0.7  Mg (0.8  ton), is lost  during clinker production as water  vapor,
 carbon dioxide,  and  other  volatile compounds.3  Dry process kilns  can be
 20 to 25  percent  shorter than wet  process kilns because  little or  no
 kiln  residence time  is  needed  to evaporate water  from  dry  feed.15  The
 water vapor  produced in a wet  kiln increases the  heat  loss  from  the
 kiln.  Therefore, dry process  kilns require less  fuel   per  kilogram of
 clinker produced  than wet process  kilns.18  In 1982, average  consumption
 of kiln fossil fuel  per kilogram of clinker produced by  the wet  process
 was  6.5 megajoules (MJ) (5.6 million  British thermal units  [Btu's] per
 ton) compared to 4.6 MJ (4.0 million  Btu's per ton) per  kilogram of
 clinker produced by  the dry  process.19

     Dry process  kilns  that  have become subject to the new  source
 performance  standards (NSPS) since 1979 commonly  employ  a preheater or
 preheater/precalciner system.13  Both the  preheater and  the preheater/
 precalciner  systems allow the sensible heat in kiln exhaust gases to
 preheat and partially calcine the  raw feed before the   feed enters the
 kiln.

     Addition of a preheater to a dry process  kiln permits use of a kiln
 one-half to two-thirds shorter than a dry  kiln without a preheater
 because heat transfer to the dry feed (whether ground  or pelletized)  is
more efficient in a preheater than in the  preheating zone of the kiln.14
Also, because of the increased heat transfer efficiency,  a preheater
 kiln system requires less  energy than a wet kiln or a  dry kiln without a
preheater to achieve the same amount of calcination.   Wet raw feed
 (containing 20 to 40 percent moisture) requires a longer residence time
 for preheating,  which is best provided in  the kiln itself.   Therefore,
                                  2-3

-------
wet process plants do not use preheater systems.20  Compared to a wet
process kiln, a dry process kiln with a preheater system can use 50 percent
less fuel.21  There are two kinds of preheater systems:   the suspension
system and the traveling-grate system.

     The suspension preheater is the most commonly used preheater system
and usually consists of a vertical tower containing a multistage cyclone-
suspension process interconnected with pipes (see Figure 2-4).   Dry
ground feed typically containing less than 1 percent moisture enters at
the top of the tower and exits at the bottom into the feed end of the
kiln.5  Hot kiln exhaust gases exit at the feed end of the kiln and
travel upward through the preheater system countercurrent to the flow of
the descending feed.  The dry feed particles can be entrained by and
uniformly dispersed within the ascending hot gas stream.13  Thus, the
feed is separated and preheated in each stage, and, in the lower stages
of the preheater where the off-gases are the hottest, up to 40 percent
of the calcining may occur.22

     In the traveling-grate preheater system, the blended raw feed is
moistened to form small pellets that measure about 2.5 cm (1 in.) in
diameter and that contain 10 to 12 percent water.12  These pellets are
spread upon a grate that travels slowly toward the feed end of the kiln.
Hot exhaust gases leaving the kiln pass through the pellet bed, drying,
heating, and partially calcining the pellets.12  A traveling-grate
preheater is shown in Figure 2-5.

     Addition of a precalciner system to a preheater system allows about
95 percent of the calcining of the raw material to be accomplished
before the raw material enters the kiln.17,21  Figure 2-4 depicts a
suspension preheater/precalciner kiln system.  In this system,  a vessel
called a flash precalciner is located between the preheater and the kiln
and is fueled by a separate burner.   The calciner may use air from the
kiln (air-through system) or from the clinker cooler (air-around system)
and, depending on the specific system,  will burn 40 to 60 percent of the
total kiln fuel.23  Rapid calcination occurs in the precalcining vessel.
By monitoring the precalciner temperature, adjustments to the calcination
rate can be quickly made.  This helps to yield uniform calcination of
the kiln feed material.24  Gases from the precalciner continue up through
the preheater.17

     The direct contact that occurs in preheater and preheater/precalciner
systems between hot kiln exhaust gases and the raw feed can allow conden-
sation of sulfur and alkalies on the feed, which can result in a high
concentration of these substances in the clinker.   Excessive sulfur in
the cement can delay some of the hydration reactions until after the
final setting of the concrete.  The delayed hydration reactions can
cause expansion of the concrete and cracking of the final structure.25
Therefore, the American Society of Testing and Materials (ASTM) limits
the total sulfur trioxide (S03) content of finished cement to 2.3 to
4..5 percent, depending on the type of cement and the content of tri-
calcium aluminate [(CaO)3 • A1203].9  Alkalies in cement can react with
certain aggregates to cause swelling and weakening of the concrete.26
                                  2-4

-------
 Also,  excessive  alkalies can  lead to  ring  formations  inside  the  kiln  or
 preheater  vessels, which can  adversely affect clinker  formation.
 Therefore, ASTM  has placed an optional limit on the total  alkali  content
 in  portland cement of 0.6 percent.9

     Alkali metals (sodium and potassium oxides) and sulfates are
 volatilized in the calcining  area of  the kiln; and, if the kiln  exhaust
 gases  travel  through a preheater, raw mill, or dryer,  these  alkali
 metals and sulfates condense  on the raw feed that is entering the kiln.27
 This condensation can set up  a recirculation of volatile compounds that
 could  increase the alkali metal and sulfur content of  the  clinker.27  To
 avoid  excessive  buildup of alkali and sulfur on the raw feed, some
 preheater  kiln systems have an alkali bypass exhaust gas system  added
 between the kiln and the preheater.28  Some of the kiln exhaust  gases
 are ducted to the alkali bypass prior to the preheater, thus reducing
 the alkali fraction passing through the feed.29  Particulate emissions
 from the bypass  are controlled by a separate pollution control device.
 Dust collected in the alkali  bypass control device is  usually disposed
 of, although  it  can be recycled to the kiln after leaching to remove the
 alkali content.

     Dry process kilns with a preheater or preheater/precalciner  have
 higher production capacities  than simple dry process kilns of the same
 diameter.22   Preheaters can increase  the capacity of a dry process kiln
 by 20 to 30 percent, and a flash calciner can add another 25 percent
 clinker production capacity.21  Kiln  capacity increases because  the
 preheater and precalcining vessels accomplish some of the feed calcination
 much more quickly than can occur in the kiln.   Also, because some drying
 and calcining of the feed has already been accomplished by the preheater
 or preheater/precalciner systems, the kiln itself can be shorter  and,
 therefore, can be rotated more quickly while maintaining proper  feed
 residence time and bed depth.21,30

     2.2.2.3  Clinker cooling.  Clinker is discharged from the kiln to a
 clinker cooler.  Ambient air  is passed through a moving bed of hot
 clinker, cooling the clinker  from about 1500°C (2700°F) to about 65°C
 (150°F).13,31  Clinker coolers can be the planetary, grate, or vibrating
 type.   Cooled clinker can be stored in silos,  storage halls,  or outdoor
 stockpiles.  Clinker cooler exhaust gases can be ducted to emission
 control equipment or can be recycled to the kiln,  the preheater  (or
 precalciner),  the raw mill,  or a raw feed dryer.

 2.2.3  Cement Manufacture and Shipment

     Figure 2-6 presents a schematic of finished cement grinding and
 shipping.   Cooled clinker is mixed with about 5 percent gypsum and
 ground to a size such that 90 to 100 percent of it passes a minus-325 mesh
 sieve.16,32  Gypsum is  added to regulate  the setting time of the finished
cement.33  Depending on the  type of cement being made,  other additives
may be mixed in at this time.   These other additives could include
dispersal,  water proofing,  or air-entraining agents.8   The finish mill
can be an open circuit,  where the material  passes  through the mill
                                  2-5

-------
regardless of particle size, or a closed circuit, where air classifiers
send over-sized clinker back through the mill for further grinding.34
The finished cement is packaged in bags or bulk loaded and delivered by
rail,  truck, or ship.

2.3  INDUSTRY CHARACTERIZATION

     As of December 1983, there were 143 portland cement manufacturing
plants in 40 States and Puerto Rico.  Eight of these plants do not
produce clinker but grind purchased clinker into finished cement.20  The
143 plants are operated by 45 different companies.20  By comparison, in
1979,  53 companies operated 166 cement plants, and,  in 1974, 51 companies
operated 179 cement plants.35,36  As of December 1983, 56 percent of the
industry clinker capacity was owned by 10 companies.  The five companies
that owned about 36 percent of industry clinker capacity at that time
were:   Lone Star Industries, Inc.  (11.9 percent), General Portland, Inc.
(7.0 percent), Ideal Basic Industries, Inc. (6.2 percent), Gifford-Hill
& Company, Inc. (5.3 percent), and Lehigh Portland Cement Company
(5.1 percent).37

2.3.1  Geographic Distribution

     Geographic distribution of domestic portland cement plants as of
December 1983 is shown in Figures 2-7a and 2-7b.  Portland cement plants
tend to be located near adequate supplies of suitable raw materials,
sufficient fuel of a consistent quality, electrical  power, and a source
of labor.38  Because portland cement is expensive to transport, proximity
and economical transportation to regional markets is also necessary.
About 95 percent of portland cement is shipped less  than 483 kilometers
(300 miles).39

     Regional concentration of cement plants has shifted in recent
years.  Previously, clinker capacity was concentrated in the Eastern and
the Great Lakes-Midwestern regions of the U.S. where construction activity
was high.  Clinker capacity has increased in the West and the South
Central regions of the country because of the changing construction
market, the availability of limestone, and, in the case of the South
Central region, the availability of inexpensive fuel.36

     California plants have the capacity to produce  the largest quantity
of domestic cement, followed by plants in Texas and  Pennsylvania.39
Texas, however, accounts for the largest consumption of portland cement,
followed by California and Florida.39

2.3.2  Production

     Growth of the portland cement industry is closely tied with growth
of the construction industry.  As shown in Table 2-1, clinker production
reached a peak of 70.9 xlO6 Mg (78.2 xlO6 tons) of clinker in 1973.40
Clinker production reached a 10-year low of 54.7 xlO6 Mg (60.2 xlO6 tons)
in 1982.39  Cement consumption was 57.2 xlO6 Mg (63.1 xlO6 tons) in 1982
and increased to 63.0 xlO6 Mg (69.4 xlO6 tons) in 1983.41,42  The Portland
                                  2-6

-------
 Cement Association predicts  68.9 xlO6  Mg (76 xlO6  tons)  of cement
 consumption in  1984,  74.8 x  106  Mg (82.5 xlO6 tons)  in 1985,  and
 77.8 xlO6  Mg (85.8 xlO6  tons)  in 1986,  yielding an estimated  average
 annual  increase in cement consumption  of 7  percent.42

 2.3.3  Growth Trends

      As of December 1983,  64 of  the  143 port]and cement  plants  are
 subject to the  NSPS for  the  portland cement industry.  One plant is  a
 grinding-only facility.   Of  the  63 conventional  plants,  24 plants have
 all-new facilities such  that the entire plant is subject to the NSPS;
 31  plants  have  at  least  one  kiln subject to the NSPS;  and 8 plants have
 nonkiln facilities only,  such  as a finish mill  or  transfer facilities,
 subject to the  NSPS.  Appendix A lists  information for plants with
 facilities subject to the  NSPS.

      Table 2-2  lists 37  cement plants  that  have facilities that have
 become  subject  to  the standards  since  the 1979  review  and identifies the
 affected facilities and  the  control  equipment used at  these facilities.
 Fourteen of the plants have  installed  all new facilities  (i.e.,  kilns,
 clinker coolers, and other associated  equipment such as  mills,  transfer
 facilities,  and storage  facilities)  since 1979,  and 23 plants have added
 new kiln capacity  and/or other equipment.

      Growth  of  the portland  cement industry  after  1971 was  projected to
 be  about six kilns and six clinker coolers  per year.43   As  shown  in
 Appendix A,  63  kilns and 54  clinker  coolers  have become  subject  to the
 NSPS  in the  12  years since 1971.   This  growth rate is equivalent  to
 about five  kilns and more  than four  clinker  coolers per year.

      Construction  of several entirely new cement plants  is  planned in
 the U.S.   Four  new cement  production plants  have received permits for
 construction, and  two more plants  have  submitted permit applications.
 Four  additional  plants have  in the past  had  active, approved construction
 permits, but  the permits have  expired and would have to be  reapproved
 before  construction could  commence.44   In addition, several expansions
 or modifications of existing facilities  have been planned.  Three plants
 not currently subject to the NSPS  have modification/reconstruction plans
 that would bring them under the standards.44  These plans include adding
 new kilns and converting from  the wet process to the dry process.  Four
 existing plants  with facilities already  subject to NSPS each plan to add
 an additional kiln.  One of these will  be a wet process kiln,  and the
 other three will be dry process preheater/precalciner systems.  Another
 plant plans to add a preheater/precalciner system to an existing dry-
process  kiln currently subject to the NSPS.44

 2.3.4   Process Developments

     Three developments  have occurred in the manufacture of portland
cement  in the last decade.
                                  2-7

-------
     First, many plants have converted their kilns to coal firing because
of the high cost of oil and gas fuels.  In 1983, 98 percent of all
cement kilns were fired by coal; in 1973, 31 percent of the kilns were
coal fired.45  Many plants continue to have the capability to use oil or
gas as a backup fuel.22  Figure 2-8 illustrates consumption of coal,
oil, and natural gas by the cement industry from 1970 to 1980.

     Waste fuels are sometimes used as alternative kiln fuels because
they are less expensive than oil and gas.  Waste fuels used in some
Portland cement plants include solvents (eight plants), waste oil (three
plants), and wood chips (one plant).46  No waste fuel was burned in
cement kilns in 1972; in 1982, 525 xlO6 MJ (498 xlO9 Btu's) of energy
were generated in cement kilns from waste fuels.19

     The second development in the port!and cement industry has been a
trend from the wet process of clinker production to the dry process,
usually including a preheater/precalciner system.   Figure 2-9 illustrates
the change over time in the number of plants using the wet or dry produc-
tion process, and Figure 2-10 shows the construction of wet and dry
clinker production capacity in the U.S. between 1930 and 1982.  Until
recently, the wet process was more common than the dry process because
wet raw materials blend more easily and more consistently, producing a
higher quality clinker.47  Dry raw materials are,  however, easier to
handle, and dry blending and material handling techniques have improved
significantly.47

     Overall, about 62 percent of cement plants use the dry cement
production process.  Eighty percent of the post-1971 kilns use the dry
process compared to 46 percent of the pre-1971 kilns.48  Additionally,
67 percent of the newer kilns have preheater or precalciner systems;
whereas only 6 percent of the pre-1971 kilns have preheater systems, and
none have precalciner systems.48  The trend in the portland cement
industry is toward the construction of dry process kilns as a means of
conserving energy, increasing production capacity, and reducing material
handling problems.

     The third development has been a trend toward the use of the roller-
type raw mill systems.  These mills combine drying and classifying of
the raw feed with crushing operations.  Drying is accomplished by the
use of hot exhaust gases recovered from the kiln,  preheater, or clinker
cooler.  The use of this type of raw mill system improves productivity
and energy efficiency.49  Figure 2-11 depicts a roller mill.

     These three developments have resulted in an increase in energy
efficiency and average kiln capacity.  Fuel efficiency in cement produc-
tion has increased because of the increased use of the dry process of
clinker production and associated preheater and preheater/precalciner
systems and because of increased use of kiln or clinker cooler gases to
preheat raw materials in the raw mill.  Twenty percent less energy was
needed to produce 1 Mg (1.1 ton) of clinker in 1982 than was needed in
1972.50
                                  2-8

-------
      Average plant capacity has  increased  because  production  costs  per
 ton of product are less  for the  larger dry process plants.51   For this
 reason,  the recent economic downturn  caused the  closing  of  many  smaller
 wet process facilities,  which, while  decreasing  the total number of
 operational  cement plants,  increased  average kiln  capacity.52  Twenty-
 eight percent of  the  274 kilns in  operation in the U.S.  by  the end  of
 1983 have  been built  since  1971, and  these kilns represent  47 percent  of
 the domestic clinker  capacity.53   Clinker  capacity from  these  kilns
 averages 496,000  Megagrams  per year (Mg/yr) (547,000 tons/yr) per kiln,
 which is more than twice the  clinker  production  potential of their
 pre-1971 counterparts.53

 2.4  EMISSIONS FROM PORTLAND  CEMENT PLANTS

 2.4.1  Particulate Emissions

      Portland cement  plants were selected  for NSPS  development because
 cement clinker production facilities  can be significant  sources  of  par-
 ticulate matter.   The most  significant  sources of  particulate emissions
 at  a cement  plant are the kiln and clinker cooler.    Kilns controlled by
 a cyclone  dust collector for  product  recovery purposes can  emit  as  much
 as  22.5  kilograms  of  particulate matter per megagram  (kg/Mg) of  raw
 material (45  pounds per  ton [lb/ton]), and clinker  coolers  controlled  by
 a cyclone  dust collector can  emit as  much  as 15 kg/Mg (30 lb/ton) of raw
 material.54   Thus,  a  plant with facilities  controlled only  by cyclones
 and  producing 544,000 Mg/yr of clinker (600,000 tons/yr) would emit
 about 21,900  Mg/yr (24,200 tons/yr) of particulate matter from the  kiln
 and  about  14,600  Mg/yr (16,100 tons/yr) from the c-1 inker cooler.

      Figure 2-12  presents particle size distribution ranges for
 uncontrolled  particulate emissions from a  dry process kiln,  a wet process
 kiln,  and  a clinker cooler.55  Approximately 50 percent of  the particles
 in exhaust gases  from a  dry process kiln with a preheater are smaller
 than  1.5 to 3.5 micrometers (urn)  in diameter (i.e., the mass median
 diameter [MMD]  is  1.5 to 3.5 urn), and 85 to 99 percent of the particles
 are  smaller than 10 urn.   Similarly, for wet process kiln exhaust gases,
 the MMD  is 7  to 40 urn, and 20 to  60 percent of the  entrained particulate
 matter is  smaller  than 10 urn in diameter.   However, the clinker cooler
 exhaust gas particles are larger; the MMD  is 30 to  over 100  urn, and less
 than  20 percent of clinker cooler dust is smaller than 10 urn in diameter.

 2.4.2  Sulfur Oxide Emissions

      Emissions of  sulfur oxides from portland cement kilns are caused by
 fuel combustion and clinker formation.  Sulfur oxide emissions  are
almost solely in the form of sulfur dioxide (S02),  although  small
quantities  of sulfuric acid (H2S04) and S03 may exist in kiln  exhaust
gases.

     Actual S02 emission test results  for facilities that have  become
subject to  the NSPS since 1979 range from 0.2 to  265 parts per  million
(ppm) by volume and from 0.09 to  277 kg/h  (0.2 to 611 Ib/h).  Table  2-3
                                  2-9

-------
 presents  the  S02  emission test results by plant.  Assuming  7,200  hours
 per year  (h/yr) of  operation, approximately  half of the  plants would  be
 considered  significant  sources of S02 emissions (i.e., greater than
 91 Mg/yr  [100 tons/yr]  of S02 emissions).

      The  S02  emissions  result from both  sulfur in the fuel  and sulfur in
 the raw materials.  Direct correlation of these factors  with S02  emissions
 is difficult  because of the complex chemistry of sulfur  in  the kiln.
 Sulfur can  be absorbed  into the clinker, raw feed, or dust  collected  in
 a control device  or emitted as a gas.  In addition, the  amount of sulfur
 found in  the  fuel and the feed can vary  significantly from  plant  to
 plant.  The sulfur content of the fuel ranges from 0.5 to 3 percent.  As
 shown in  Table 2-3, wet process kilns tend to emit larger quantities  of
 S02 than  dry process kilns because they  burn more coal per  Mg of  clinker
 produced  than do  dry process plants.   The sulfur content of the raw feed
 material  is known to vary considerably.   One source reported an average
 sulfur content of 0.05  percent by weight in the feed of  nine California
 cement plants.56  One of these California plants reported 0.2 percent
 sulfur in the feed.57   A plant in Oregon reported 0.02 percent sulfur in
 the feed.58  A plant in Colorado, which  uses shale containing kerogen as
 a raw material, reported 0.6 percent sulfur in the feed.59

      Emissions of S02 from the kiln are  reduced significantly by the
 production process because the S02 is absorbed into the clinker.   About
 75percent of the S02 formed in the kiln reportedly is absorbed into the
 clinker.60  One mass balance calculation measured approximately 38 percent
 removal  of S02 into the clinker.61  Industry personnel state that removal
 efficiencies within the production process can exceed 90 percent.62
 Data  on reduction of S02 emissions in the production process vary widely
 because of differences  in process parameters and in sulfur content of
 raw feed material  and fuel.

 2.4.3  Nitrogen Oxide Emissions

      Parameters that affect emissions of nitrogen oxides (NO )  from
                                                            s\
cement kilns include the nitrogen content in the coal  and raw materials.
Nitrogen oxides can form in portland cement kilns at temperatures of
1400° to 1650°C (2600° to 3000°F).   Because clinkering occurs at  about
1500°C (2732°F),  temperatures favorable  for NO  formation are reached in
routine  kiln operation.63                     x

     As  shown in  Table 2-4,  actual  kiln  NOV emissions  range from  116 to
                                          s\
609 ppm  by volume  and from 14 to  294  kg/h (31 to 649 Ib/h).   Assuming
7,200 h/yr of operation, all  but  one  kiln would be  considered a major
source of NOX emissions (i.e.,  greater than 91 Mg/yr [100 tons/yr] of
NO  emissions).
                                  2-10

-------
                            CRUSHED RAW
                             MATERIALS
                          FROM STOCKPILE
ro
i
                                                                                     SLURRY
                                                                                      BASIN
RAW MATERIALS —
ARE PROPORTIONED
                                                               RAW MATERIALS
                                                                   MILL
                                                                                                    SLURRY
                                                                                                    FEEDER
                                Figure 2-1.  Typical  wet process  material handling.
                                         64 65

-------
      CRUSHED RAM
     MATERIALS  FROM
       STOCKPILE
INJ

Ul

t-
Ul
UJ
Z
JLlL 	
f'gH
o

oc
_-J > —
-.»...

Q
<
IA
"Tl
>i


11
;



«
X¥¥Y i
AW MATERIALS -» \l
RE PROPORTIONED
                                             RAW MATERIALS
                                                 MILL
-n ~t
': •- :•


'.-

>

,' .•
h ., ,iu— n-i
TO
PREHEATER
4
DRV MIXING AND
BLENDING SILOS
GROUND RAW
MATERIAL STORAGE
                                     Figure 2-2.  Typical dry  process material handling.
                                                                                             64

-------
                    EXHAUST STACK
ro
i
Co
                                             GAS FLOW
                                       MATERIAL FLOW
                                                             PRIMARY AIR
                                                               AND FUEL
                       RAW
                      FEED
                    MATERIAL
                                                  SECONDARY
                                                      AIR
COOLER
               U
            CLINKER
            OUTLET
                              Figure 2-3.  Typical clinker production process.28

-------
                    CXHRUST -«-EEEE
ro
i
                                                 FEED  (FROM 3RD  STflCEJ




                                                        FUEL  (COflL. CflS.  OIL. ETC. J



                                                        OXYGEN CCOOLER.  KILN, flTMOSPHERE)




                                                 PRODUCT  CTO 4TH STflGEl
                                                FLASH PRECALCINER
                                                   KILN
                                                                  BURNER
                                                                    COOLER
                                                                           	»- VENT
                                                                          CLINKER
                           Figure  2-4.   Four-stage suspension preheater with a  precalciner.28

-------
I
t—•
(Jl
PELLETIZER
   PAN
TO COLLECTION
DEVICE
         Figure 2-5.  Traveling  grate preheater system.66

-------
ro
i
                          5«/. GYPSUM |
                            ADDED
                                                                   BULK STORAGE
 BULK  BULK BOX  PACKAGING TRUCK
TRUCK CAR  CAR   MACHINE
                                       Figure 2-6.   Finish mill  grinding and shipping.
                                                                                          65

-------
ro
i
                                                                                                  ANNUAL GRAY CEMENT GRINDING CAPACITY
                                                                                                       • Under SOO.OOO Shorl Tons
                                                                                                              454.000 Metric Tons

                                                                                                       • SOO.OOO lo 900.000 Shod Tons
                                                                                                         454.000 lo 816.000 MeUlc Tons

                                                                                                       • O«e(800.000 Shorl Tons
                                                                                                             816.000 Metric Tons

                                                                                                       • Grinding Only
                                    Figure 2-7a.   Portland  cement  plant  locations—Western U.S.67

-------
r\j
i
oo
                                                                                      ANNUAL GRAV CEMENT GRINDING CAPACITY

                                                                                           •  Under 500,000 Short Tons
                                                                                                  4S4.000 Metric Tons

                                                                                           •  500.000 to 900.000 Short Tons
                                                                                             454.000 to 816.000 Metric Tons

                                                                                           • Over 900.000 Short Tons
                                                                                                816.000 Metric Tons

                                                                                           •  Grinding Only

                                                                                             Puerto Rico (1  plant)
                                   Figure  2-7b.   Portland  cement plant  locations—Eastern  U.S.67

-------
           Mq  (tons)

      0.2  (0.2) -
      0.1 (0.1)
                                  cubic meters
                                  (barrels)
                                0.03
                                (0.2) "
                                0.03
                                (0.1)
                 I   I    I   I   I   I   I   i   I   T
             70    72      74     76     78     80
                                           year

                        (a)  Coal
                  000    (000
                  liters  cubic feet)
                  93 (3.0)
                  62 (2.0) -
                  31 (1.0)
                                     70     72     74     76


                                                (b)  011
                                                 Wet process

                                                 Dry process
1   I   l
78     80
   year
                         70    72     74    76    78    80


                                (c)  Natural  Gas
                                                             year
Figure 2-8.   Fuel
uel  consumption per megagram (ton) of clinker  produced  by
fuel  type and  clinker  production  process.68
                                        2-19

-------
=  WET PROCESS
=  DRY PROCESS
=  WET AND DRY PROCESS
                200
            oo
                                2%
                                         4%
                 20
                      1966  1973  1980  1982

                               YEAR
        Figure 2-9.   Number of plants using wet or dry
              clinker production process.53,68,69
                               2-20

-------
                    TOTAL WET AND
                     DRY PROCESS
                            1	1	1	1	1	f	!
1930  1935  1940  1945  1950  1955  1960  1965  1970  1975  1980  1985
                   YEAR OF CONSTRUCTION
  Figure 2-10.   Kiln  construction by year.
53
                       2-21

-------
RAW MATERIAL
  FEED SPOUT

GRINDING ROLLER
                                     PRODUCT DISCHARGE PORT
                                     CLASSIFIER BLADE
GAS INTAKE PORT
                                    HOT GAS FROM KILN,
                                   PREHEATER OR COOLER
 Figure 2-11.   Detail of roller mill that combines crushing,
  grinding, drying, and classifying in one vertical unit.70
                             2-22

-------
                                                   Leas Than Indicated  Particle Diameter,  %
ro
i
rv>
         to
         c.
         rt>
Oi
r+
n
(D
in
tM
ID
Q.
en
         CT
         c
         n-
         O
         -t>
         O
         (V
         fD
         ri-
         Q.
          01
          tn

-------
            TABLE 2-1.   U.S.  CLINKER PRODUCTION,  KILN CAPACITY,
                    AND CAPACITY UTILIZATION—1972-198239
Year
Clinker production,
  106 Mg (106 ton)
 Kiln capacity,
106 Mg (106 ton)
Utilization
   rate,
  percent
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
     70.2 (77.4)
     70.9 (78.2)
     70.8 (78.1)
     58.5 (64.5)
     62.2 (68.6)
     65.3 (72.0)
     68.5 (75.5)
     69.0 (76.1)
     63.2 (69.7)
     61.4 (67.7)
     54.7 (60.3)
    77.5 (85.4)
    78.8 (86.9)
    82.5 (90.9)
    83.7 (92.3)
    77.5 (85.4)
    80.0 (88.2)
    80.8 (89.1)
    81.4 (89.7)
    83.5 (92.1)
    82.8 (91.3)
    80.6 (88.9)
  90.6
  90.0
  85.8
  70.0
  80.2
  81.6
  84.7
  84.9
  75.8
  74.1
  67.9
                                   2-24

-------
TABLE 2-2.   FACILITIES SUBJECT  TO THE NSPS SINCE  1979 REVIEW
















1
fXJ
en



















Company/
plant location
Alamo Cement Co.
San Antonio, Tex.
Alaska Basic Ind.c
Anchorage, Alaska
Ash Grove Cement Co.d
Louisville, Nebr.
California Portland
Cement Co.
Mojave, Calif.
Capitol Aggregates,
Inc.T
San Antonio, Tex.
Centex Corp.
Buda, Tex.
Columbia Cement Co.
Zanesville, Ohio
Davenport Ind.
Buffalo, Iowa
Dixie Cement Co.
Knoxville, Tenn.
Genera) Portland, Inc.
New Braunfels, Tex.
Genstar, Ltd.
Redding, Calif.
San Andreas, Calif.9


Gulf Coast Portland
Cement Co.
Houston, Tex.
Ideal Basic Ind. , Inc.
Theodore, Ala.

La Porte, Colo.9

Tijeras, N. Hex.


Date- type
1981-D.PC

--

1982-0, PC
1
1981-D, PC


1983-D, PC


1983-D, PC

--

1981-D, PC

1979-D.PC

1980-D.PC

1981-D, PC

1945-W
1952-W
1956-W
--


1981-D, PC


1981-D.PH




Clinker
capacity,
105 Mg/yr
(103 tons/yr)
474
(523)
.-

506
(558)
907
(1,000)

453
(500)

425
(468)
--

734
(809)
464
(512)
794
(875)
518
(571)
174 (192)
174 (192)
174 (192)
—


1,284
(1,415)

399
(440)


Kiln
Fuel,
sulfur
content, %
Coal, 1.5/
Coke, 3.9
„

Coal, 0.9

Coal, 0.53


Coal/coke,
3.35

Coal

__

Coal

Coal, 1.5

Coal

Coal /wood,
2.0
Coal,
0.6

--


Coal, 1.5


Coal, <1.0


AFFECTED FACILITIES


Clinker cooler
Emission
control
ESP (w/cooler
and raw mi 11)
__

ESP

FF(-) (w/raw
mill)

FF


FF (w/raw mill)

__

FF (w/raw mill)

FF(-)

2 ESP's (w/raw
mill)
FF(-) (w/cooler
and raw mill)
ESP


--


FF(-) (w/cooler
and raw mill
dryers)
FF(+)

"
Trans-
mi sso-
meter
Yes

—

Yes

NA (e)


Pro-
posed

No

—

NA

Yes

Yes

Yes

Yes


--


Yes


Yes

"
Date
1981

__

1982

1981


1983


__

—

1981

1979

1980

1981

--


--


1981


1981


Emission
control
(w/kiln and raw
mill)
__

FF( + )

FF(-)


FF


—

—

FF

FF(-)

GB

FF(-) (w/raw mill
and kiln)
..-


--


FF(-) (w/kiln and
raw mill dryers)

FF(-) (w/raw mill)

"
Date
1981

1982

-.

1981


1983


1983

1978

1981

--

1980

1981

--


1973
& 1978
1978
1981


1981

>1979

Other
Facil ity-control
Entire plant except
finish mill-FF
Finish mill , stor-
age, transfer FF(-)
--

Raw mill FF(-)
(w/kiln)

Entire plant-FF


Raw mill (w/kiln)

Finish mill-FF

Entire plant-FF

--

Entire plant-FF

Raw mill-FF(-) (w/
kiln and cooler)
--


Finish mill

Storage-FF
Entire plant-FF(-)


Entire plant, except
finish mill-FF
F inish mi 1 1-fF
                                                                               (continued}

-------
TABLE  2-2.   (continued)
              AFFECTED FACILITIES
Company/
plant location
Kaiser Cement Corp.
lucerne Valley, Calif
Permanente, Calif.

San Antonio, Tex.


Lehigh Portland
Cement Co.
Mason City, Iowa
Lone Star Ind. , Inc.
Davenport, Calif.
Ewa Beach, Hawai i
Cape Girardeau, Mo.
ro
I Pryor, Okla.
ro
Maryneal , Tex.
Salt Lake City, Utahd

Martin Marietta Corp.
Lyons, Colo.
Leamington, Utah

Monolith Portland
Cement Co.
Laramie, Wyo.
Moore McCormack
Cement, Inc.
Brooksville, Fla.
Oregon Portland Cement
Durkee, Oregon
River Cement Co.
Festus, Mo.

Date-type3
1982-D, PC

1981-D.PC

1975-D.2PC
(second PC
in 1979)
1979-D.PC


1981-D.PC

-
1981-D.PC

1979-0

--
1979-W

1979-0, PC

1982-D, PC

1981 -W


1982-0, PH

1979-D.PH

--


Clinker
capacity,
105 Mg/yr
(103 tons/yr)
1,379
(1,520)
1.J79
(1,520)
703
(775)

493
(543)

675
(744)
-
900
(992)
242
(267)
--
136
(150)
367
(405)
547
(603)
272
(300)

508
(560)

454
(500)
--

Kiln
Fuel,
sulfur
content, %
Coal

Coal, <0.5

Coal, 1.0


Coal


Coal

--
Coal, 3

Coal, 3-4

--
Coal, oil
gas, 0.4-0.6
Coal, 0.52

Coal,
0.4-0.6
Coal,
0.5-0.9

Coal, 1.5

Coal, <1.0

--

Clinker cooler
Emission
control
FF (w/raw mill )

FF(-)

3 ESP's
FF (on
alkali bypass)
ESP (w/raw mill)


ESP (w/raw mill)

--
ESP

FF

--
FF(-)

FF(-)

FF (w/raw mill)

ESP


FF(-)

ESP (w/cooler)

--

Trans-
misso-
meter
NA

No

No


No


Yes

—
Yes

No

--
NA

Yes

NA

No


NA

Yes

--

Date
1982

1981

1975


--


1981

-
1981

1979

—
1979

--

1982

1981


1982

1979

—

Emission
control
FF (with alkali
bypass)
FF(-)

FF


--


G8

--
2 FF's (w/raw mill)
mill
GB

--
FF(-)

--

FF

FF(-)


FF(-)

ESP (w/kiln)

--

Date
1982

1981

1977


1980


1981

>1979
1981

--

1979
--

1979

1982

1981


--

1979

>1979

Other
Faci lity-control
Entire plant-FF

Entire plant except
finish mill-FF(-)
Finish mill-FF


Mil 1 , separators-FF


Entire plant-FF

Mill, storage-FF
Entire plant-FF

—

Coal transfer-FF
--

Limestone dryer-
-FF(-)
Entire plant-FF

finish mill,
Cement cooler-Fr(-)

--

Entire plant-FF(-)
Finish mill-ESP
Raw mill-FF


-------
                                                      TABLE  2-2.    (continued)
















ro
t
ro



Company/
plant location
Southwestern Portland
Cement Co.
Victorville, Calif.

Bush land, Tex.

Odessa, Tex.

Texas Industries, Inc.
Hunter, Tex.

Midlothian, Tex.





Clinker
capacity,
, 105 Mg/yr
Date-type (103 tons/yr)


1984-D

--

1978-D


1980- D

--
?Kiln types: W = wet process; D
Emission control types: ESP =
cfabric filter; and GB = gravel


,PC 726
(800)
—

,PH 253
(279)

,PC 602
(664)
--
= dry process; D,PH = dry
electrostatic precipitator
bed filter.

Kiln

Fuel,
sulfur
content, %


Coal

--

Coal, 0.5


Coal, 1.2

--
process with
; FF = fabric
AFFECTED


Emission
control


FF

--

FF


ESP

--
preheater; and D, PC
filter (baghouse); F
FACILITIES
Clinker cooler

Trans-
misso- Emission
meter Date control Date


NA 1984 GB 1984

1981

No — -- 1982


Yes 1980 FF 1980

1979
= dry process with preheater/precalciner
F( + ) = positive-pressure fabric filter; FF(-)

Other

Facility-control


Entire plant-FF

Coal storage,
Coal transfer-FF
Coal transfer,
Coal storage-FF

Entire plant-FF

Finish mill-ESP
= negative-pressure
grinding only is performed.
mPlant has more than one kiln; other kilns subject to NSPS installed prior to 1979.
,NA = not available.
 facilities under construction.
uPlant is closed.

-------
ro
rv>
oo
                             TABLE  2-3.   S03  EMISSION TEST  RESULTS  FOR  PORTLAND  CEMENT  FACILITIES  THAT
                                                HAVE  BECOME  SUBJECT  TO  THE  NSPS  SINCE  1979
Process type
Wet


Dry
No PH, PC
Preheater



Preheater/
precalciner



















Facility
1.
2.


3.
4.
5.
6.
7.
8.



9.

10.
11.
12.
13.
14.
15.



16.
17.
18.
19.

20.
Kiln
Kiln,
Kiln1

Kiln
Kiln
Kiln
Kiln
Kiln (with cooler)
Kiln

Bypass

Kiln + raw mill
Bypass
Kiln.
KilnK
Kiln + raw mill
Kiln
Kiln
Kiln + raw mill
Bypass
Kiln + raw mill
Bypass
Kiln
Kiln
Kiln
Kiln,+ raw mill
Kiln*
Kiln
Control
type
ESP
ESP
ESP

FF
FF(-)fl
FF
FF(+)9
ESP
FF(-)

FF

FF
FF
FF(-)
FF(-)
FF
FF(-)
FF
ESP
ESP
ESP
ESP
ESP
ESP
ESP
ESP
ESP
ESP
Date
tested
5/82
10/79
11/79

3/80
9/82
2/83
4/82
5/80
10/80

10/80

11/83
11/83
5/83
5/81
5/83
5/82
9/82
1/83
1/83
10/83
4/84
5/82
7/81
3/82
12/83
12/83
4/81
SO, emissions3
Fuel,
Coal,
Coal,
Coal,

Coal,
Coal,
Coal,
Coal,
Coal,
Coal,



Coal,
Coal
Coal,
Coal,
Coal
Coal,
Coal
Coal,
Coal,
Coal,
Coal,
Coal
Coal,
Coal,
Coal
Coal
Coal
% S
coke, 0.76b
°'6d
0.6d
A
3.5d
1.0db
0.5d.
<1.0d
0.55b
0.52b


H
0.5d
ft
0.53d
2.0fl
A
<0.50d

coke
coke
coke
coke
ft
l'2d
3.0d



ppm, vol.
--
240e
237e

17.8
—
0.2
71
6.5

-------
                 TABLE  2-4.  NO   EMISSION TEST  RESULTS  FOR PORTLAND CEMENT FACILITIES THAT HAVE BECOME
                                            SUBJECT  TO  THE NSPS SINCE 1979
ro
<£>
NO., emissions
Process type
Wet


Dry
No PH, PC
Preheater
Preheater/
precalciner










Facility
1.
2.


--
3.
4.
5.



6.

7.
8.
9.
10.
11.
Kiln
K11nd
Kilna


Kiln
Kiln
Kiln

Alkali

Kiln +
Alkali
Kiln
Kiln
Kiln +
Kiln
Kiln +









bypass

raw mill
bypass


raw mi 1 1

raw mill
Control
type
ESP
ESP
ESP


FF
FF
FF(-)e

FF

FF
FF
FF(-)
FF(-)
FF
FF(-)
ESP
Date
tested
6/81
10/79
11/79


2/83
9/82
10/80

10/80

11/83
11/83
5/83
5/81
5/83
5/82
1/83
ppm,
vol.
— _
258C
332C


384
—
259f
116?
320T
809
--
--
279, 462C
55
103, 219C
145
631
	 y\ 	
kg/h (Ib/h)
46.
54.
102


108
98.
--
—
—
—
112
7.0
181
14.
93.
108
227
9 (103)
4 (120)
(225)


(238)
4 (217)




(248)
(15.4)
(399)
1 (31.0)
9 (207)
(237)
(502)
kg/Mg (Ib/ton)
__
0.
0.


1.
	
--
—
--
--
1.
0.
0.
0.
0.
0.
I.

39
73


64





03
06
9
16
38
43
63
+ cooler







12.
Alkali
Kiln +
+
Kiln +
bypass
raw mi 1 1
cooler
cooler
ESP
ESP

ESP
1/83
10/83

5/82
76
--

609
3.8
141

294
(8.3)
(311)

(649)
0.
--

1.
025


64

(0.78)
(1.47)


(3.28)





(2.06)
(0.13)
(1.8)
(0.31)
(0.76)
(0.86)
(3.26)

(0.05)


(3.28)
                                                                                                     (continued)

-------
                                                TABLE 2-4.  (continued)
ro
i
CO
o
NO,, emissions

Process
Precalci
(cont1


type
ner
d)



Facil
13.

14.
Ki
Ki
Ki

ity
ln,+ raw mill
lnh
In
Control
type
ESP
ESP
ESP
Date
tested
12/83
12/83
4/81
ppm,
vol .
220
184g


kg/h
112
78.0
205.

(Ib/h)
(247)
(172)n
9 (454)

kg/Mg
0.72
0.50
1.44

(Ib/ton)
(1.45).
(1.01)n
(2.87)
 Emission test results received from State and local air pollution control agencies, EPA regional
.offices, and industry contacts.
 Average of 1.8 kilns in operation.
 . ppm normalized to 3 percent 02
 Average of 3 kilns in operation.
,(-) = negative-pressure fabric filter.
 Type I clinker production
uType II clinker production.
 Kiln in raw mill bypass mode; i.e., raw mill  is off.

-------
 2.5  REFERENCES FOR CHAPTER 2

  1.   U.  S.  Environmental Protection Agency.  Multimedia Assessment and
      Environmental  Research Needs of the Cement Industry.  Publication
      No.  EPA-600/2-79-111.   May 1979.   p.  2.

  2.   Kirk,  R.  and D.  Qthmer.   Cement.   In:   Encyclopedia of Chemical
      Technology,  3rd Edition, Vol.  5.   New York.  John Wiley and Sons,
      Inc.  1979.   p.  163.

  3.   U.  S.  Environmental Protection Agency.  Environmental Considerations
      of  Selected  Energy Conserving  Manufacturing Process Options:
      Vol.  X—Cement Industry Report.   Publication No.  EPA-600/7-76-034J.
      December  1976.   p.  84.

  4.   Portland  Cement Association.   Cement  and Concrete Reference Book.
      Chicago,  Illinois.   1964.   p.  17.

  5.   Kreichelt, T.,  D.  Kemnitz,  and S.  Cuffe.   Atmospheric Emissions
      From  the  Manufacture of Portland  Cement.   U.  S.  Department of
      Health, Education,  and Welfare.   Cincinnati,  Ohio.   Publication
      No. AP-17.   1967.   p.  10.

  6.   Letter  from  Greer,  W.,  Lone Star  Industries,  Inc.,  to Cuffe, S. ,
      EPA/ISB.   August 28,  1984.   Response  to request  for comments on the
      draft  review document,   p.  3.

  7.   U. S.  Environmental  Protection Agency.   Industrial  Process Profiles
      for Environmental  Use:   Chapter 21~The Cement Industry.   Publication
      No. EPA-600/2-77-023u.   1977.   p.  17.

  8.   Reference  1, p.  37.

  9.   Reference 2, p.  183.

10.   Reference 5, p.  11.

11.   Telecon.  Maxwell,  C., MRI, with Kreisberg, A., Fuller Company.
      April 12, 1984.  Discussion of  kiln design  parameters.

12.   U. S.  Environmental Protection Agency.   Inspection  Manual  for
      Enforcement of New  Source Performance  Standards:   Portland Cement
      Plants.  Publication No. EPA-340/1-75-001.  September 1975.  p. 3-5.

13.   Reference 2,  p.  165.

14.   Reference 3,  p.  18, 19.

15.   Reference 7,  p.  20.
                                   2-31

-------
16.  Reference 7, p. 24.

17.  Reference 2, p. 184.

18.  Energy Conservation in the Cement Industry.   Pit  and  Quarry.   July
     1982.  p. 61.

19.  Portland Cement Association.  Energy Report:  U.  S. Portland  Cement
     Industry.  Skokie, Illinois.  October, 1983.

20.  Portland Cement Association.  U. S. and Canadian  Portland  Cement
     Industry:  Plant Information Summary.  Skokie,  Illinois.   December 31,
     1983.  pp. 55-80.

21.  Reference 3, p. 24.

22.  Reference 1, p. 38.

23.  Letter from Gebhardt, R., Lehigh Portland Cement  Company,  to  Cuffe, S.,
     EPA/ISB.   June 6,  1984.   Response to request  for  comments  on  draft
     review document,   p. 2.

24.  Reference 3, p. 37.

25.  Reference 3, p. 63.

26.  Letter and attachments from Venturini, P., California Air  Resources
     Board, to Cuffe,  S., EPA/ISB.  January 17, 1984.  Data  for California
     plants,  p.  27.

27.  Reference 3, p. 25.

28.  KVB, Inc.  Emissions Reductions by Advanced Combustion  Modification
     Techniques for Industrial Combustion Equipment.   Prepared  for U.  S.
     Environmental Protection Agency.  Industrial  Advisory Panel Meeting.
     June 8, 1983.

29.  Reference 3, p. 22.

30.  Reference 23, p.  3.

31.  Friedman, D.  Coal-Fired Preheater/Flash Furnace  Boosts Production
     and Fuel  Efficiency.  Pit and Quarry.  July 1981.  p. 109.

32.  Reference 5, p. 13.

33.  Reference 3, p. 62.

34.  Reference 2, p. 185.

35.  Barrett,  K.   A Review of Standards of Performance for New  Stationary
     Sources—Portland Cement Industry.  U. S. Environmental  Protection
     Agency.  Publication No. EPA-450/3-79-012.  March 1979.  p. 4-1.

                                   2-32

-------
 36.   Reference 20,  p.  1.

 37.   Reference 20,  p.  13.

 38.   Reference 5,  p.  3.

 39.   Portland  Cement  Association.   United States Cement Industry Fact
      Sheet.  Second edition.   Skokie,  Illinois.   September 1983.

 40.   Portland  Cement  Association.   The U.  S.  Cement Industry—An Economic
      Report.   Third edition.   PCA  Market and Economic Research.   Skokie,
      Illinois.   January  1984.   p.  17.

 41.   Portland  Cement  Association.   Portland Cement Consumption.   Vol.  5,
      No.  12.   PCA Market and  Economic  Research.   Skokie,  Illinois.
      February  22, 1984.  p. 3.

 42.   Portland  Cement  Association.   U.  S.  Cement  Consumption Forecast.
      PCA  Market  and Economic  Research.   Skokie,  Illinois.   October,
      1983.

 43.   U. S. Environmental Protection Agency.   Background Information for
      Proposed  New Source Performance Standards:   Steam Generators,
      Incinerators,  Portland Cement Plants,  Nitric Acid Plants,  and
      Sulfuric  Acid  Plants.  Publication  No.  APTD-0711.   August  1971
      p. 32.

 44.   Memorandum  from  Clark, C., MRI, to  Project  File.   January  20,  1984.
      Compilation of data from  telephone  contacts  with  State and  local
      air  pollution  control agencies regarding construction and  modifica-
      tions of  cement  plants.

 45.   Reference 20,  p.  I.

 46.   Portland  Cement Association.   U.S.  and  Canadian  Portland Cement
      Industry:    Plant Information  Summary.   Skokie, Illinois.   December 31
      1982.  p.  37.

 47.   Reference 1, p. 33.

 48.   Reference 20,  p.  9.

 49.   Reference 40,  p.  6.

 50.   Reference 40,  p.  10.

 51.   Reference 1, p. 21.

52.  Reference 20, p.  31-34.

53.  Portland Cement Association.    Design and Control  of Concrete Mixtures.
     Skokie,  Illinois.  1979.   pp.   32-40.
                                   2-33

-------
54.  Reference 43, pp. 28-29.

55.  Reference 35, p. 4-16.

56.  Reference 26, pp. 112-121.

57.  Reference 26, p. 117.

58.  Information from Bosserman, P., Oregon Department  of  Environmental
     Quality, to Clark, C., MRI.  January 9, 1984.   Summary  of  source
     test results for Oregon Portland Cement Company, Lake Oswego,
     Oregon,  p. 15.

59.  Information from Clouse, J., Colorado Air Pollution Control  Division,
     to Clark, C., MRI.  November 10, 1983.  Notice  of  intent to  construct
     and operate Martin Marietta Cement, Lyons, Colorado,  p. 15.

60.  Ketels, P., J.  Nesbitt, and R.  Oberle (Institute for  Gas Technology)
     Survey of Emissions Control and Combustion Equipment  Data  in Industrial
     Process Heating.  Prepared for U. S. Environmental Protection
     Agency.  Publication No. EPA-600/7-76-022.  October 1976.  p.  72.

61.  Reference 26, p. 238.

62.  Reference 23, p. 4.

63.  Reference 3, p. 102.

64.  Reference 53.   pp. 18-19.

65.  Reference 1, p. 35.

66.  Reference 12, p. 3-8.

67.  Reference 20, p. 53-54.

68.  Reference 18, p. 62.

69.  Reference 5, pp. 33-47.

70.  Reference 40, p. 8.
                                   2-34

-------
             3.  CURRENT STANDARDS  FOR PORTLAND CEMENT  PLANTS

3.1  NEW SOURCE PERFORMANCE STANDARDS

     On August 17, 1971, the Environmental Protection Agency proposed
standards for port!and cement facilities under Section  111 of the Clean
Air Act to control particulate matter and visible emissions.  The standards
were promulgated on December 23, 1971, and revised in response to a
court remand on November 12, 1974.x-3

3.1.1  Summary of New Source Performance Standards

     The affected facilities under  the new source performance standards
(NSPS) for Portland cement plants are the:  kiln, clinker cooler, raw
mill system, finish mill system, raw mill dryer, raw material storage,
clinker storage, finished product storage, conveyor transfer points, and
bagging and bulk loading and unloading systems.4

     The standards prohibit the discharge into the atmosphere from any
kiln, exhaust gases which:

     1.  Contain particulate matter in excess of 0.15 kilograms per
megagram (kg/Mg) (0.30 pounds per ton [lb/ton]) of feed (dry basis) to
the kiln, or
     2.  Exhibit greater than 20 percent opacity.

     The standards prohibit the discharge into the atmosphere from any
clinker cooler, exhaust gases which:

     1.  Contain particulate matter in excess of 0.05 kg/Mg (0.10 Ib/ton)
of feed (dry basis) to the kiln, or
     2.  Exhibit 10 percent opacity or greater.

     Finally, the standards prohibit the discharge into the atmosphere
from any affected facility other than the kiln or clinker cooler, exhaust
gases which exhibit 10 percent opacity or greater.1

     The standards apply to any facilities that have been built,  modified,
or reconstructed after August 17,  1971.   The term "modified facility"
applies to facilities to which physical  or operational  changes have been
made that caused an increase in the emission rate of particulate  matter
or visible emissions (i.e., the pollutants to which this standard
applies).5  The term "reconstructed facility" applies when the
                                    3-1

-------
replacement cost of components exceeds 50 percent of the cost of building
a comparable new facility.6

3.1.2  Testing and Monitoring Requirements

     3.1.2.1  Particulate Matter.   Test methods used to determine
compliance with the standards covering particulate matter emissions are:

     1.   Method 5 for the concentration of particulate matter and the
associated moisture content of the exhaust gases,
     2.   Method 1 for sample and velocity traverses,
     3.   Method 2 for stack gas velocity and volumetric flow rate deter-
minations, and
     4.   Method 3 for analysis of exhaust gases for carbon dioxide
(C02), excess air, and dry molecular weight.

     The sampling time for Method 5 must be at least 60 minutes for
emission testing of the kiln or the clinker cooler.   The sample volume
collected using Method 5 must be at least 0.85 dry standard cubic meters
(dscm) (30 dry standard cubic feet [dscf]) for testing of the exhaust
gases from the kiln and 1.15 dscm (40.6 dscf) for testing of the clinker
cooler.4  Particulate mass emission rate in grams per hour (g/h) can be
calculated by multiplying the volumetric flow rate of the gases (in
dscm/h)  as determined by Method 2 times the particulate concentration
(in g/dscm) as determined by Method 5.

     Total kiln feed rate (excluding fuel) must be determined during
each testing period by suitable methods in order to calculate particulate
mass emissions per unit of kiln feed.   Total  kiln feed rate is expressed
in units of Mg (or tons) per hour of dry feed to the kiln and is to be
confirmed by a material balance over the production system.

     At all times, the air pollution control  equipment associated with
the affected facility (or facilities) should be maintained and operated
to minimize particulate emissions.  Monitoring of operation or maintenance
procedures may include opacity observations,  review of procedures, and
facility inspections.  The owner or operator of a portland cement plant
with one or more facilities subject to the NSPS is required to monitor
and record daily production rates and kiln feed rates.4

     3.1.2.2  Opacity.  Methods for determining compliance with opacity
standards are defined in Section 60.11 of the Code of Federal Regula-
tions.7  Method 9 is used for measuring visible emissions from stationary
sources.  Continuous monitoring of opacity is not required.

3.1.3  Recordkeeping and Reporting Requirements

     Notification of construction, reconstruction, or modification as
well as  initial startup is to be provided to the Administrator of the
EPA.8
                                    3-2

-------
      Within 60 days after achieving the maximum production (or throughput)
 rate of an affected facility but no later than 180 days after initial
 startup of the facility,  the owner or operator of the plant is required
 to conduct a performance  test and furnish to the Administrator a report
 of the test results.   Emissions measured during periods of startup,
 shutdown,  and malfunction are not considered representative for the
 purpose of demonstrating  compliance.   Under Section 114 of the Clean Air
 Act, performance tests may be required by the Administrator at other
 times.9

      Records are to be maintained by  the plant owner or operator of  the
 occurrence and duration of startup,  shutdown,  and malfunctions in the
 process and of malfunctions of air pollution control  equipment.8

      A file of all  performance tests  and other reports  and records
 required is to be kept for a period of at least 2 years.8

 3.2  STATE REGULATIONS

      Portland cement  manufacturing plants are  currently operating in
 40 States  and Puerto  Rico.   Appendix  B presents a summary  of  particulate
 and visible emissions  regulations  for these  States  as well  as  regulations
 for S02  and NOX  that  are  applicable to Portland cement  production processes
 in the absence of NSPS.10   Enforcement authority  for  the NSPS  for the
 Portland cement  industry  has  been  delegated  to most States.

     •Of  the 40 States,  24  States  have particulate matter regulations for
 all  or part of the  State  that  are  defined by one  of two sets  of  process
 weight rate equations.  For a  kiln feed  rate of 136 Mg/h (150  tons/h),
 the  allowable (State)  emissions are 17.5  and 25.1  kg/h  (38.6  and  55.4 Ib/h),
 for  the  two sets  of process weight rate  equations.  These  emissions
 convert  to  0.128  and 0.185  kg/Mg  (0.257  and  0.369 Ib/ton)  of  kiln feed
 (units of  the NSPS  for  particulate mass  emissions).

     Limitations  on particulate matter emissions  for existing  sources in
 the  16 other  States range from 0.15 to 0.75  kg/Mg (0.30 to 1.5 Ib/ton)
 of  kiln  feed.  However, variations in  exhaust  gas flow  rates from particular
 facilities  or variations in emission  testing methods could result in
 some States requiring more  stringent  emission  control levels than the
 NSPS.

     Most of  the States limit visible emissions to 20 percent opacity or
 less for new  facilities and to 40 percent opacity or less for existing
 facilities.   Some State regulations are more stringent than the NSPS
 (i.e., requiring visible emissions of 10 percent opacity or less).

     Sulfur dioxide regulations are specified in one of five categories:

     1.  Parts per million (ppm), by volume,
     2.  Kilograms per megajoule (kg/MJ) (pounds per million British
thermal units [Btu's]) of  heat input,
                                    3-3

-------
     3.  Ambient air quality levels similar to or the same as the national
ambient air quality standard (NAAQS) for S02)
     4.  Kilograms per megagram (kg/Mg) (pounds per ton [lb/ton]) of
material processed, or
     5.  Requirements on the sulfur content of the fuel.

     Sulfur dioxide regulations for 34 of the 40 States fall into
categories (1), (2), and (3) above.  The most stringent regulations in
each of the first two categories stipulate (a) less than 500 ppm of S02,
by volume, and (b) less than 0.003 kilograms of S02 per megajoule
(0.7 pounds of S02 per million Btu's) of heat input.

     Only 6 of the 40 States have regulations specific to NO  that may
                                                            X
be applicable to portland cement plants.   California requires the lowest
achievable emission rate (LAER), and some districts within California
have specific NO  regulations.   The remaining five States have regulations
                /\
specific to fuel-burning equipment (expressed in units of pounds per
million Btu's).  According to telephone contacts with the State air
pollution control agencies, only two of the five States (Indiana and
Oklahoma) enforce the fuel-burning standards at portland cement
facilities.

     In addition to State regulations or NSPS, some portland cement
plants may be required to achieve more stringent emission levels under
regulations for the Prevention of Significant Deterioration (PSD).11
Also, if a new plant is located in a nonattainment area for National
Ambient Air Quality Standards (NAAQS), the LAER would be required for
the nonattaining pollutant.12  Facilities that are subject to PSD as
well as NSPS (since the 1979 review) are listed in Table 2-2.

3.3  REFERENCES FOR CHAPTER 3

 1.  Federal Register.   Standards of Performance for New Stationary
     Sources.Proposed Standards for Five Categories.   Washington, D.C.
     Office of the Federal Register.   August 17, 1971.   36 FR 15704-15722.

 2.  Federal Register.   [Promulgation of Standards for Five Categories
     Proposed August 17, 1971.]  Washington, D.C.   Office of the Federal
     Register.   December 23, 1971.   36 FR 24876-24899.

 3.  Federal Register.   [Revision to opacity standard for portland
     cement plants.]  Washington, D.C.   Office of the Federal Register.
     November 12, 1974.   36 FR 39872-39877.

 4.  U. S. Environmental Protection Agency.   Code of Federal Regulations.
     Title 40,  Chapter I, Part 60.   Sections 60.60 through 60.64.
     Washington, D.  C.   Office of the Federal Register.   July 1, 1982.

 5.  Reference 4, Section 60.14.
                                    3-4

-------
             4.   CONTROL TECHNOLOGY AND  COMPLIANCE TEST RESULTS

      This  chapter presents  participate  and gaseous emission control
 technology used  on portland cement facilities  that have become subject
 to  the  NSPS since 1979.   Emission  test  results for those facilities  are
 also  presented.   Appendix C presents  compliance test  results by plant.
 This  information was  obtained  from State  and local  air pollution control
 agencies,  EPA  regional  offices,  and individual  portland cement plants.

 4.1  AVAILABLE PARTICIPATE  CONTROL TECHNOLOGY

      Fabric filter control  of  wet  and dry  process  kilns,  clinker coolers,
 and other  facilities  as  well as  electrostatic  precipitator  control of
 kilns provide the basis  for the  particulate  matter and visible emissions
 standards  that were proposed and promulgated in 1971  and revised (visible
 emissions  only)  in 1974.1

     Typical methods  used for  control of particulate  emissions from
 potential  sources at  portland  cement manufacturing  facilities  are  listed
 in  Table 4-1.  The kiln  and clinker cooler are  the  first  and second
 largest sources,  respectively, of  particulate  emissions  at  a cement
 plant.  Particulate emissions  also  occur during material  handling,
 transfer,  and storage.

     Table  4-2 summarizes the  particulate control  technology currently
 in  use at  facilities  that have become subject  to the  NSPS since  the  1979
 review.   Particulate  emissions from kilns are  controlled by  either
 fabric filters or electrostatic  precipitators.   Particulate  emissions
 from clinker coolers  and other facilities (mills,  storage facilities,
 and transfer facilities) are typically controlled by  fabric  filters.
 Two plants  have  finish mills controlled by electrostatic precipitators.

 4.1.1  Kiln

     At the 28 plants with  one or more kilns that have become  subject to
 the NSPS since the 1979 review,  17  kilns are controlled by fabric filters
 and 13 kilns are  controlled by electrostatic precipitators (3  kilns at
 one plant are controlled by one  electrostatic precipitator).

     4.1.1.1  Fabric Filters.  Most of the fabric filters used for
 control  of  kiln emissions are the negative-pressure (suction) type.
Only one positive-pressure fabric filter system is used for control of
emissions from a  kiln that has become  subject to the NSPS since 1979.
                                    4-1

-------
        TABLE 4-1.  POTENTIAL SOURCES OF PARTICULATE EMISSIONS AND
                       TYPICAL CONTROL TECHNOLOGIES
Source
Particulate control technology
Raw material system (including
  crushing and grinding)

Raw material dryer

Raw material storage (except
  coal piles)

Kiln (including preheater/
  precalciner systems and
  alkali bypass systems)

Clinker cooler

Clinker storage

Finish mill system (excluding
  fugitive emissions)

Finished product storage

Conveyor transfer points (e.g.,
  to primary crusher, secondary
  crusher, elevators, material
  storage, grinding mill)

Packaging (i.e. , bagging)
Low flow fabric filter systems


Fabric filters

Low flow fabric filter systems
Fabric filters, electrostatic
  precipitators
Fabric filters, gravel bed filters

Low flow fabric filter systems

Fabric filters, electrostatic pre-
  cipitators (on large mills)

Low flow fabric filter systems

Low flow fabric filter systems
Low flow fabric filter systems
                                    4-2

-------
        TABLE 4-2.  PARTICULATE CONTROL TECHNOLOGY CURRENTLY IN USE
         AT PLANTS WITH FACILITIES THAT HAVE BECOME SUBJECT TO THE
                        NSPS SINCE THE 1979 REVIEW

Affected
facil ity
Kiln:
Wet
Dry, without
Parti cul ate
FF
1
1
control
ESP
o'
technology3
GB
0
0

Total
3
1
    preheater or
    precalciner

  Dry, with preheater           3             lc
Dry, with preheater/
precalciner
Cl inker cooler

Other facilities
5,§d,
2e
13 2d
2I'
31
' 4,3d,
le
lc,le

2f
0

4

0
20

23

31
 ESP = electrostatic precipitator; FF = fabric filter; GB = gravel bed
 filter; NA = data not available.   Note that some of these facilities
.may have cyclone precollection devices.
 Exhaust gases from 3 kilns (at one plant) ducted to one control system.
 .Kiln and clinker cooler exhaust gases combined.
 Exhaust gases combined with raw mill.
 Kiln, clinker cooler, and raw mill or raw mill  dryer exhaust gases
..combined.
 Finish mill facilities at these two plants are controlled by electro-
 static precipitators; the remaining other facilities at these two plants
 are controlled by fabric filters.
                                    4-3

-------
     Fabric filter systems (often called baghouses) consist of a structure
containing tubular bags made of woven fabric through which the exhaust
gas stream is passed.  Particles are collected on the upstream side of
the fabric.  Dust on the bags is periodically removed and collected in a
hopper.

     The efficiency of a fabric filter is directly proportional to the
fabric area.  Design efficiencies of greater than 99.9 percent are
typical.  The air-to-cloth ratio of fabric filters ranges from about
1.3:1 to 2:1 for kilns and alkali bypass systems.  The bags are typically
made of fiberglass and cleaned by reverse air.

     Kiln exhaust gases must be cooled to about 200° to 315°C (400° to
600°F) before entering the fabric filter to preclude damage to the
filter fabric.2  Cooling of exhaust gases from dry process kilns may be
accomplished by water sprays and/or bleed-in air.s  Bleed-in air (i.e.,
colder air), which is the most commonly used coolant, condenses alkali
material onto the particulate.   Control of the alkali content of the
clinker is effectively accomplished by cooling a portion of the kiln
exhaust gases (i.e., alkali bypass) and then directing them to a separate
fabric filter.   At plants using dry process kilns with preheater or
preheater/precalciner systems,  kiln exhaust gases may be ducted to a raw
mill (or raw material dryer) to dry the raw feed material; this procedure
increases the moisture content and reduces the temperature of exhaust
gases entering the fabric filter.4  For wet process kilns, the high
moisture content of exhaust gases requires adequately insulated fabric
filter systems to prevent corrosion of the ducts and blinding of the
filter bags because of a wet filter cake.3

     The temperature of gas entering the fabric filter must be maintained
above the dew point of the gas  to prevent blinding of the filter bags.5
One plant has experienced blinding of the bags as a result of kerogen (a
bituminous material  in the raw feed) in the dust.   Process modifications
made to burn off the kerogen prior to feed entering the kiln are expected
to correct the problem.6

     Bag life is affected by the abrasiveness of the particulate matter
in the exhaust gases, temperature of the gases,  and maintenance
practices.7,8  Abrasion of filter bags (and peripheral  equipment) can
also be a problem because of high flow rates.5  Heat recovery techiques

and new bag materials (e.g.,  Nomex® and Gore-tex®) are expected to
increase bag life.7   Improved methods of detecting leaks in bags and
fastening bags  are also available.7

     Disadvantages of fabric filters include the need for a high pressure
drop (necessitating  high energy consumption), a low resistance to
temperatures above 3158C (60CrF),  and the potential  for blinding of the
bags at temperatures below the  dew point.9

     Advantages  of fabric filters  include high efficiencies,  simplicity
in operation,  reliability,  and  compartments that can be isolated for
repairs.9
                                    4-4

-------
     4.1.1.2  Electrostatic Precipitators.  Cleaning of exhaust gases
using electrostatic precipitators  involves three steps:   (a) passing the
suspended particles through a direct-current corona to charge them
electrically, (b) collecting the charged particles on a grounded plate,
and  (c) removing the collected particulate from the plate by a mechanical
process (i.e., rapping).

     At 11 of the 28 plants with kilns subject to the NSPS since the
1979 review, emissions from the cement kiln are controlled by electro-
static precipitators.  Design efficiencies of greater than 99.9 percent
are  typical.  The specific collection area (SCA) is a parameter used to
ensure design efficiency of an electrostatic precipitator.  The SCA is
defined as the ratio of the total  plate area to the gas flow rate.  As
the  SCA of an electrostatic precipitator increases, collection efficiency
improves.  Information from industry contacts indicates that the SCA's
for  electrostatic precipitators controlling kilns that have become
subject to the NSPS since 1979 range from 1.0 to 1.9 square meters per
cubic meter per minute (m2 per 1,000 m3/min) (310 to 570 square feet per
1,000 actual cubic feet per minute [ft2/!,000 acfm]).

     The high resistivity of particles in cement kiln exhaust gases
requires that the gases be conditioned prior to entering the electrostatic
precipitator.  Resistivity is about a factor of 10 lower for wet process
kilns than for dry process kilns because of the moisture in the gases;
however, the resistivity of exhaust gases from the dry process kiln can
be lowered by spray cooling.10  Exhaust gases from dry process kilns
with preheaters have higher resistivity than those from dry process
kilns without preheaters.11  Electrostatic precipitators can operate at
high temperatures and at temperatures below the dew point.

     Startup of the kiln requires  a period of several hours (for a
downtime of only a few hours) to more than 24 hours (for a cold start).
During startup, there are more combustible materials in the kiln than
are present during normal operation.   Because of this hazard, some kiln
operators reportedly deenergize the electrostatic precipitator during
the startup period because sparks  in the electrostatic precipitator
could ignite the combustibles.   As a result,  particulate emissions could
be uncontrolled for a period of several  minutes to more than a day.
Similarly, the electrostatic precipitator could be deenergized during
gradual  cool down of the kiln because of the potential  for ignition of
combustibles.  However, electrostatic precipitator vendors and plant
operators state that, because of improved process control, it is now
normal  practice for new electrostatic precipitators to start up and shut
down concurrent with the kiln induced draft fan.12-14

     Due to the presence of a spark source, shut-offs of the electrostatic
precipitator can also occur if carbon monoxide (CO) or excess air concen-
trations reach a preset critical  level  at which an explosion could occur
in the electrostatic precipitator.   This automatic shut-off is called a
CO trip.   The fundamental cause of potential  explosive conditions is
incomplete combustion of the fuel  in the presence of a spark source.15
These conditions result if there are irregularities in the feed,
                                    4-5

-------
 disturbances in the coal  conveying and feeding,  insufficient fuel
 preparation (i.e.,  drying,  grinding),  insufficient combustion chamber
 temperature, or disturbances in the air and gas  flow system (i.e   mill
 bypass,  preheater draft).16

      Table 4-3 summarizes CO trip data from electrostatic precipitators
 on kilns subject to the NSPS since 1979.   The composition of the kiln
 exhaust  gases may be monitored at a given plant  by CO,  02,  or
 combustibles monitors.   A trip level and, in some  cases,  an alarm  level
 are set  by each plant.   Table 4-3 lists the gases  that  are  monitored to
 set off  the alarm or the  CO trip at each  plant and the  location  of the
 monitor.   The annual  frequency of CO trips and the duration of each trip
 are also presented.   Levels set for shut-offs (CO  trips)  range from 0 2
 to 6 percent CO for kilns subject to the  NSPS since 1979, and duration
 of CO trips per occurrence  ranges from less than 1 minute to about
 20 minutes.17  Two  types  of trips are  common:  a  spike CO  event of  short
 duration (i.e.,  a few seconds to 5 minutes) and  long term instability
 (10 minutes to 4 hours).  The frequency of trips ranges from a few trips
 per year to over 600  trips  per year.17  State air  pollution control
 agency enforcement  personnel  indicated  that CO trips of electrostatic
 precipitators are typically treated as  malfunctions  of  the  control
 device;  emissions during  malfunctions  are not considered  representative
 for the  purpose  of  demonstrating compliance (see Chapter 3).

      Electrostatic  precipitator vendors and plant  personnel  state  that
 if a kiln  is  properly  designed and  operated,  CO trips of the  precipitator
 should be  infrequent.18-22   Several equipment vendors noted  that one  or
 two CO trips  per month  is an  average frequency for  a properly  operated
 kiln.21,22   Each trip would  average about 3  minutes.22  Chapter 6  dis-
 cusses some  design  and  operation  parameters  that can be used  to minimize
 the occurrence of CO trips.

      Cleaning  of the electrostatic  precipitator plates is sometimes
 hampered by  the  fineness  of  the  dust.   In  addition,  air leaks, high
 moisture content, low gas temperature,  and  the alkali,  sulfur and  chloride
 content  of  the exhaust  gases  may  produce  conditions  that promote corrosion
 within the precipitator that  may  cause reduced efficiency, short circuits,
 and  downtime.  Good operation  and maintenance  practices should prevent
 these  problems.

     Advances  in  the design of electrostatic precipitators such as the
 use  of wide duct  spacing, prechargers,  or pulse energization are available
 and  could improve dust collection and reduce costs.25  Electrostatic
 precipitation is  used almost exclusively  for control of kiln emissions
 in  Europe and Japan where such design advances have been applied.24

     4.1.1.3  Cyclones.  Cyclone collection systems consist of one or
more conically shaped vessels  in which  the gas stream follows a circular
motion prior to outlet (typically at the bottom of  the  cone).25  Collec-
tion efficiency  is a function of (a) size of particles  in  the gas stream,
(b) particle density, (c) inlet gas velocity, (d) dimensions of the
cyclone,  and (e) smoothness of the cyclone wall.26
                                    4-6

-------
-p»
I
                   TABLE 4-3.   SUMMARY OF CARBON MONOXIDE  TRIP DATA FOR ELECTROSTATIC  PRECIPITATORS ON

                                KILNS THAT HAVE BECOME  SUBJECT TO THE NSPS SINCE 1979.17
Process typea
1980- D, PH/PC
(K)
1975-D, PH/PC
(K)
1980-D, PH/PC
(K)
1975-D, PH (K);
1982-0, PH/PC (K)
1979-D, PH/PC
(K)
\n /
1981-D PH/PC
(K + RM)
1979-D, PH
(K + CC)
1983-D, PH/PC
(K + CC)
1981-W
(K)
\nJ
1981-0, PH/PC
(K)
processdasC-PF -'
KE = kiln exit; P
ESPO = outlet to
Coal
firing method Location
Direct (K); ESPI
Indirect (PC)
Pneumatic blowers FSO, ESPI
to riser duct
Direct (K); KE, PE, ESPI
Indirect (PC)
tec
I\C
Injected KE, ISO, FSO

SC-PF KE, TSO, FSO
Fluidizing KE, PE
pump
OC, PE,
KE, ESPI
--

--

Ic^Sllfw-VJssufi'fan^ = ^ ""^
E = precalciner exit- TSO = third-sta
electrostatic Drecinitstnr ^
Monitor
Measured gas
Combustible
gas
Combustible
gas, 02
CO, 02


Combustible
gas, 02, C03
Combustible
CO, 02
Combustible
gas
CO, 02



CC = clinker cooler; 1
tl t- FSO - f '
i e , bu - Tirst-sta(


Alarm level Trip level
>0.6%
2% 5%
0.7% 1.5%


None 0.8%-TSO
0. 6%-FSO
1% CO 1% CO
0.4% 0.6%
0.2%/ 2.0% NGC
0.8% 4.0% coal
>2%-CO
<1.5%-02


) = dry process; PH = preheater;
_
je outlet; ESPI = inlet to electi
CO trip
Frequency Duration per
per year event, minutes
15 12.8
7.7 4
690.7 11.3

177 19.7
177 4.4

3.0 4
15 <1
122.2 3.3
Seldom <3



PC = precalciner; W = wet

rostatic precipitator;
      NG = natural gas.

-------
      In the cement  industry, cyclone-type collection systems are used
for product recovery.  Cyclones are typically used as precollection
systems in combination with fabric filters or electrostatic
precipitators.27

4.1.2  Clinker Cooler

     Of 23 clinker  coolers subject to the NSPS since the 1979 review, 17
are controlled by fabric filters, 2 are controlled by electrostatic
precipitators, and  4 are controlled by gravel bed filters.

     4.1.2.1  Fabric Filters.   Most of the fabric filters used for
control of clinker  cooler emissions are the negative-pressure type.
Only one positive-pressure fabric filter system is used for control of
emissions from a clinker cooler that has become subject to the NSPS
since 1979.

     The bags in fabric filters controlling clinker coolers are typically
cleaned by a pulse  jet cleaning mechanism and have air-to-cloth ratios
ranging from about  4:1 to 9:1.   One fabric filter controlling clinker
cooler exhaust gases is cleaned by reverse air flow and has an air-to-cloth

ratio of 2:1.   The  bags may be made of fiberglass, Nomex®, or Gore-Tex®.
Clinker cooler exhaust gas temperatures range from about 93° to 230°C
(200° to 450°F).

     4.1.2.2  Electrostatic Precipitators.   No plants that have become
subject to the standard since 1979 use electrostatic precipitators for
control of exhaust  gases from the clinker cooler.   In two cases where
the kilns are controlled by an electrostatic precipitator and where
planetary type clinker coolers are used, the gases from the cooler are
ducted to the kiln  as preheated combustion air.28,29  One precipitator
vendor states that  in other countries electrostatic precipitators are
successfully used for control  of grate-type clinker coolers in the
cement industry.29

     4.1.2.3  Gravel Bed Filters.   Gravel  bed filters consist of a bed
of granules for particle collection.   Particles are collected by inertial
impaction, flow interception,  diffusional  collection, and gravity
settling.30  The first such system was installed in 1973, and three are
currently in use on clinker coolers that have become subject to the
standard since 1979.  One clinker cooler under construction will be
controlled by a gravel  bed filter.   Gravel  bed filters  generally have
reverse air cleaning, separate compartments,  and no electric field
augmentation.30,31

     Gravel bed filters have been applied  to  control  emissions from
five clinker coolers subject to the NSPS (four clinker  coolers since
1979).  The principal advantages of gravel  bed filters  for control  of
clinker cooler exhaust gases are the ability  to withstand temperatures
exceeding about 480°C (900°F)  and to provide  continuous control  of
emissions at wide temperature  fluctuations.31
                                    4-8

-------
 4.1.3  Other Facilities

      Affected facilities other than the kiln and clinker cooler are:
 raw mill  system,  finish mill  system,  raw mill dryer, raw material storage,
 clinker storage,  finished product storage,  conveyor transfer points,
 bagging,  and bulk loading and unloading systems.

      Raw materials,  clinker,  and cement handling are typically controlled
 by enclosures (total  or partial) and/or hooding of transfer points with
 exhaust gases directed to fabric filters.   Thirty-one plants have
 facilities other  than the kiln and clinker  cooler that have become
 subject to the NSPS  since the 1979 review.   All  of these other affected
 facilities are controlled by  fabric filters except for two finish mills
 that are  controlled  by electrostatic  precipitators.

      The  air-to-cloth ratios  of fabric  filters  controlling other
 facilities range  from 4:1 to  8:1.   The  bags are  less heat resistant than
 those used to control  kilns or clinker  coolers  (i.e.,  used at temperatures
 from ambient to 107°C [225°F]) and may  be made  of polyester felt, Dacron®
 felt,  or  polypropylene.   The  fabric filter  bags  are  cleaned by pulse  jet
 cleaning  mechanisms  at most facilities  subject  to the  standard since
 1979.

 4.2   SUMMARY OF PARTICIPATE COMPLIANCE  TEST RESULTS

 4.2.1   Kiln

      Since the  1979  review, 30 kilns  (at 28 plants)  have become  subject
 to the  NSPS.  Of  the  30  kilns,  emission test data are  available  for
 27 kilns.  Three  of the  30 kilns are  completing construction.  The
 27 kilns  produce  from  136,000  to 1,380,000  megagrams of  clinker  per year
 (Mg/yr) (150,000  to 1,520,000  tons  per year)  with an average  production
 of 570,000 Mg/yr  (630,000 tons/yr).

     Figure  4-1 shows  particulate mass emission data for  kilns subject
 to the  NSPS  since the  1979 review.  In some  cases, exhaust  gases  from
 the  kiln  are  ducted individually;  in  other  cases  the exhaust  gases from
 the  kiln  and  one  or more additional affected  facilities  (e.g., clinker
 cooler  and/or raw mill) are vented  together.  These varied  ducting
 configurations are discussed in Section 4.2.4.  Exhaust gases  from kilns
 are  controlled by fabric filters or electrostatic precipitators.  All  of
 the  27  operational kilns comply with  the NSPS of  0.15  kilograms per
 megagram  (kg/Mg)  (0.30 pounds per ton [lb/ton]) of feed  (dry  basis) to
 the  kiln.

     Visible emissions from the kiln  are limited  by the NSPS  to less
 than or equal to 20 percent opacity.  Opacity data for 11 of  the 30 kilns
 are presented in Appendix C;  these data represent observations (by plant
 or State/local personnel) of visible emissions using EPA Reference
Method 9.   All 11 of these kilns are  in compliance with the visible
emission regulation.   The visible emissions  range from 0 percent opacity
to 10 to 15 percent opacity.   State and local agency personnel have
                                    4-9

-------
             PARTICULATE
             MASS EMISSIONS,
             KG/HG (LB/TON)
               0.15 (0.30)"
KILN NSPS  LIMIT
                0.1 (0.20) - -
-pi
I
               0.05  (0.10)

(3 KILNS)

O

D
D

O
D •
a o

D
D
P




O



_.
U 0


a
a D


a
KILN +
RAW KILL











O



KILN +
CLINKER
COOLER
O



a








D

KILN +
CLINKER COOLtR
+ RAW MILL
                                                                                                                   LEGEND
                                                                                                                          WET
                                                                                                                        PROCESS
                                                                                                          ELECTROSTATIC
                                                                                                            PRECIPITATOR

                                                                                                          FABRIC FILTER
                                                                  DRY
                                                                PROCESS

                                                                  O

                                                                  D
                                                              FACILITY
           Figure 4-1.   Participate mass  emissions from  kilns  that  have become  subject  to the NSPS  since  1979.

-------
 indicated that none of the 30 operational  kilns had problems complying
 with the visible emission limit.   However,  some plants have had detached
 plumes.   At least 13 of the 28 plants have  transmissometers for monitoring
 opacity of kiln exhaust gases.

      One kiln that has become subject to the NSPS since 1979 has a
 detached plume.32  The dry process coal-fired kiln controlled by an
 electrostatic precipitator is in  compliance with the mass standard, and
 opacity is monitored by a transmissometer.   The cause of the detached
 plume is unknown.32

      Two plants have corrected detached  plume problems caused by kerogen
 (a  bituminous material) in the limestone feed material.   Both plants  use
 fabric filters for control.   One  plant added a precalciner and shortened
 the kiln.   The precalciner is operated at a temperature high enough to
 combust kerogens from the kiln feed.33  The other plant uses a uniquely
 designed preheater system.6

      Another  plant had a  detached plume  on  wet process kilns controlled
 by  electrostatic precipitators  and on  wet process kilns  controlled  by
 fabric filters when the kilns were oil-fired.   The plant now operates
 one dry  process kiln that is  coal-fired  and controlled by a  fabric
 filter,  and the plant has had no  further problems.34

      Although detached plumes have been  studied  extensively  at  several
 facilities, no one cause  appears  to be responsible for their occurrence.
 The raw  materials,  the fuel,  the  blasting explosive  used  in  mining, and
 the ambient temperature are potential  contributing causes.35-37

 4.2.2  Clinker Cooler

      Since  1979,  23  clinker coolers have become  subject  to the  NSPS.
 Emission  test  data  are  available  for 21  of  the 23  facilities.   Two
 plants are  completing  construction of  their clinker coolers.

      Figure 4-2  presents  the  particulate mass emission data  for 21  clinker
 coolers.   In  some  cases,  exhaust gases from the  clinker cooler  are
 ducted to  individual control  devices and stacks,  and,  in  other  cases,
 exhaust gases  from the  clinker cooler  are vented to one or more additional
 affected  facilities  prior to  the control  device  (see Section 4.2.4).

     Two of the  21 facilities exceed the NSPS mass emission  limit of
 0.05 kg/Mg  (0.10  lb/ ton) for the clinker cooler.  At  one facility, a
 portion of the exhaust gases  from the clinker cooler is recycled to the
 kiln and a portion is exhausted through the roller mill.  The clinker
 cooler and roller mill combined emissions are 0.095 kg/Mg (0.19 Ib/ton),
which exceeds the particulate mass emission limit for the clinker cooler.38
This plant is  uniquely designed and, at the time of testing, process
conditions were not representative of normal operating conditions.6  The
clinker cooler will be retested.  One other facility with a similar
configuration (combined clinker cooler and raw mill emissions ducted to
a fabric filter) is able to meet the 0.05 kg/Mg (0.10 Ib/ton) standard.
                                    4-11

-------
PARTICULATE
MASS EMISSIONS,
KG/MG (LB/TON)
   0.15 (0.30H-
                                            KILN NSPS LIMIT
 0.1 (0.20)
0.05 (0.1)	


D

nn A
~~| i J ^~^ ^~^
n
D
i — i n n n
— U 	 U— -LJ-LJ- 	
CLINKER COOLER
FACIL

q

CLINKER COO


D '
CLINKER COOLER
+ RAM MHL
TY

•

_ER NSPS
O


CLINKER
COOLER +
KILN
O
D


LIMIT

D

CLINKER COOLER
+ KILN
+ RAW HILL
                                                                                           LEGEND

                                                                                    ELECTROSTATIC  PRECIPITATOR  Q
                                                                                    FABRIC FILTER


                                                                                    GRAVEL BED FILTER
                                                                                                              D

                                                                                                              A
               Figure  4-2.   Particulate mass  emissions from  clinker  coolers
                     that have  become  subject to the  NSPS since 1979.

-------
 A second facility controlled by a fabric filter was recently tested and
 found to exceed the participate mass emission limit.   Data from these
 tests were not representative of normal  operating conditions because the
 clinker cooler was tested during startup.   During the test,  the air flow
 was 125 percent of design,  and the production rate was only  68 percent
 of design.39  The clinker cooler will  be retested.

      The NSPS  limits visible emissions from the clinker cooler to less
 than 10 percent opacity.   State and local  air pollution control  agency
 contacts indicated that 22  of the 23 clinker coolers  are in  compliance
 with the visible emissions  limit.

      Opacity data for five  clinker coolers  that have  become  subject to
 the NSPS since 1979 are presented in Appendix C.   Data show  that visible
 emissons are 0 percent at four plants  and  5 to 10 percent at one plant.

      The one clinker cooler that exceeds the visible  emission limit is
 controlled by  a gravel  bed  filter and  is in compliance with  the  mass
 emission standard.   The plant expects  to correct  the  visible emission
 problem by venting the exhaust gases from  the clinker cooler to  a closed
 loop heat exchanger system  and returning the exhaust  gases to the
 cooler.40

 4.2.3  Other Facilities

      Fourteen  plants  have installed  all  new facilities  (i.e.,  kilns,
 clinker coolers,  and  other  associated  equipment such  as  mills, transfer
 facilities,  and  storage facilities)  since 1979.   Seventeen additional
 plants  have  added equipment  other  than kilns  or clinker  coolers  since
 1979.   State agency  personnel  indicated  that  none of  these facilities
 had problems meeting  the  visible  emission limit of  less  than  10  percent
 opacity.   (There  is  no  mass  emission limit  for these  facilities).

 4.3   AVAILABLE GASEOUS  POLLUTANT TECHNOLOGY

 4.3.1   Sulfur Dioxide

      Sulfur dioxide  (S02) emissions  from kilns can be controlled  in  the
 process  itself by  (a)  reduction of the sulfur content of the  fuel and
 the  raw  feed material,  (b) absorption of S02 by calcium carbonate (CaC03)
 in  the  raw feed material (in the preheater and in the raw mill),
 (c)  maintenance of excess oxygen in the  kiln at an optimal level  (about
 1 to 2 percent), and  (d) combination of the sulfur with alkali oxides
 (in  the  firing end of the kiln) to form alkali sulfates within the
 clinker.41,42  The degree to which each of these methods affect S02
 reduction can vary considerably depending on process parameters.

     Data were obtained to determine those control devices and process
modifications that might reduce S02 emissions from portland cement
plants.   Total  potential S02 emissions from a kiln are equal  to sulfur
from the coal combustion plus sulfur from the raw feed calcination.
                                    4-13

-------
Both components can vary significantly from region to region and, within
a region, from plant to plant.

     Between 4.6 and 6.5 xlO9 Joules (4 and 5.6 xlO6 Btu's) are needed
to produce 1 Mg (1 ton) of clinker.43  Assuming an average of 26.7 xlO6
Joules per kilogram (11.5 xlO3 Btu's per pound) of coal, 158 to 221 kg
(348 to 487 Ib) of coal are needed to produce 1 Mg (1 ton) of clinker.
A typical plant produces approximately 544,000 Mg (600,000 tons) of
clinker in a year.   With the use of coal that is 1 percent sulfur by
weight, potential S02 emissions from fuel combustion would range between
1,894 and 2,650 Mg (2,088 and 2,922 tons) per year.   The potential for
S02 emissions is much higher when the sulfur content of the raw materials
used to produce Portland cement is considered.  As an example, sulfur
content in the raw feed was reported to be 0.02 percent by weight at one
plant and 0.6 percent at another plant.44,45  At a typical plant
(544,000 Mg [600,000 tons] of clinker a year), use of raw feeds containing
these percentages of sulfur would add between 388 and 11,657 Mg (428 and
12,852 tons) per year to the potential S02 emissions mentioned above
that are attributable to the coal.  Thus, the potential S02 emissions
from both the coal  and raw feed would range between 2,282 and 14,307 Mg
(2,516 and 15,774 tons) per year.

     The actual S02 emissions from portland cement plants (although, in
some cases, greater than 91 Mg/yr [100 ton/yr]) are significantly less
than potential S02  emissions because sulfur is retained in the product
during production.   It was reported in the 1979 portland cement NSPS
review that 75 percent of the S02 emission -potential is absorbed in the
clinker as potassium or sodium sulfates.46  Assuming this 75 percent
reduction does occur, S02 emission potential from coal  combustion in the
kilns would decrease to between 473 and 663 Mg (522 and 731 tons) per
year.  For raw feed calcination, potential S02 emissions would decrease
to between 97 and 2,914 Mg (107 and 3,213 tons) a year.  The total
potential S02 emissions would decrease to between 570 and 3,577 Mg (629
and 3,944 tons) per year.

     Sulfur dioxide emission data were obtained from source test reports.
Potential S02 emissions based on the sulfur in the coal burned by the
kiln were calculated.  Actual S02 emissions were subtracted from this
amount.  The result was divided by the calculated potential S02 emissions
from the coal to determine potential S02 reduction efficiency from the
production process.  The reduction efficiency does not include potential
S02 emissions from the raw feed.  Based on potential S02 emissions from
fuel alone, reduction efficiencies higher than 75 percent can be achieved.
These percent reduction levels would be higher if sulfur in raw feed was
accounted for.  (Only three plants reported the sulfur content in the
raw feed.)

     4.3.1.1  Flue Gas Desulfurization Systems.  Three types of flue-gas
desulfurization (FGD) systems exist that could provide control of S02
emissions from portland cement kilns:  the lime spray-drying system, the
wet limestone desulfurization system, and the dry lime injection system.
                                    4-14

-------
      Lime spray drying is being successfully introduced into utility and
 industrial  boiler systems to reduce S02 emissions.   Sulfur dioxide
 reduction efficiencies of 60 to 87 percent are guaranteed by equipment
 vendors for those plants that will be using this control  process.47  In
 lime spray  drying, atomized lime slurry reacts with S02-laden flue gas
 in a spray  dryer.   Either an electrostatic precipitator or a fabric
 filter then collects the dried particulate matter exiting the spray
 dryer.48 The byproduct from the scrubbing of the spray dryer could be
 used for fertilizer, boiler S02 control,  soil stabilization, aggregate
 road bases, or as an acid neutralizing agent.49,50

      There  is no known application of a full-scale  lime spray-drying S02
 control  system in the portland cement industry.   However,  to meet
 California  regulations,  one portland cement plant,  Lone Star Industries,
 Inc.,  is experimenting with a pilot-scale lime spray-drying system to
 reduce S02  emission levels.51  Lone Star  has had mixed  results  with this
 system;  nevertheless,  the company  is planning to install  a full-scale
 system.

      To  install  the pilot-scale system, the main conditioning tower,
 which  is upstream of the electrostatic precipitator, was  retrofitted  to
 be used  as  a  type  of lime spray-dryer ("dry scrubber").   In this  spray-
 dryer  tower,  slurry containing 90  percent available  lime  is atomized  and
 mixed  with  the kiln exhaust gases.   Slurry droplets  react  with  the S02
 and are  then  dried by  the hot exhaust gases as  shown by the following
 simplified  reaction:

     Ca(OH)2  + S02   -»•   CaS03  + H20.

 There  is  a  gas  retention  time of 4 seconds  in the tower.   The resulting
 dry particulate  matter  is  usually  exhausted from the tower to the  raw
 mill.  When the  raw mill  is  not operating,  particulate  matter passes
 directly  to the  main electrostatic  precipitator.

     Lone Star  Industries  has  found  that  the  efficiency of the  spray-dryer
 system in reducing  S02 emissions is  affected  primarily  by  two factors:
 the  temperature  of  the exhaust  gases  and  the  use of the raw mill.   Lower
 gas  temperatures in  the tower  allow  better  sulfur absorption by the
 lime.  However, when the  raw  mill  is  operating, the temperature of  gases
 leaving the tower must remain  high  so  that  materials can be dried  in  the
 mill;  When the  raw mill  is not operating,  lower gas temperatures  are
 possible, and  gases are ducted directly from  the tower to  the electrostatic
 precipitator.   Therefore, when the mill is  on, the spray dryer achieves
 a  significantly  lower percent S02  reduction than when the  mill is  off.
 However, when the raw mill is on, a  significant amount of  reduction in
 S02  occurs in the mill itself by reactions  of S02 with the  CaC03 in the
 raw materials.  Thus, temperature of exhaust gases and use  of a raw mill
 counterbalance each other to bring about S02  reductions presented  in
Table 4-4.

     The operating permit for the plant allows emissions of no more than
37 kg (82 Ib)  of S02 per hour (about 54 ppm).  The lime spray-dryer has
                                    4-15

-------
    TABLE 4-4.  S02 EMISSIONS FROM LONE STAR INDUSTRIES, INCORPORATED
                               kg/h

Raw mill off (gas temperature
177°C [350°F])
Raw mill on (gas temperature
260° to 288SC [500° to 550°F])
Lime spray-
dryer off
193 (4-2ST
102 (-2250"
Lime spray-
dryer on
81 ei?9")
78 (iH)
Based on S02 emission tests of a pilot-scale lime spray-dryer system.
                                  4-16

-------
 not enabled the plant to meet this limit.   Lone Star believes that this
 is because the calculations on which the permit was based assumed that
 all the S02 was generated from fuel  combustion.   Lone Star Industries
 has found, however,  that sulfur from the fuel  tends to be absorbed into
 the clinker and that only sulfur from the raw materials tends to be
 emitted.   The company has calculated a 95 percent correlation between
 S02 emissions and sulfur in the raw  feed material.   Generation of S02 in
 the preheater from pyrites in the raw feed was not forecast by the
 company or the FGD vendor.

      Although lime spray drying is demonstrated in other industries,
 there are differences in exhaust characteristics that lessen the perfor-
 mance of  this technology in the portland cement industry as demonstrated
 in this pilot study.

      The  wet limestone  desulfurization system  involves mixing the kiln
 exhaust gas stream with an alkali  slurry in a  wet scrubber located down-
 stream of the particulate matter control  device.   The S02  in the gas
 stream is reacted with  the alkali  slurry.   This  technology has  been
 demonstrated on sources such  as utility  and industrial  boilers.   Sulfur
 dioxide removal  efficiencies  of greater  than 90  percent are possible
 with  this control  device.52

      The  Lone Star Industries  portland cement  plant  in  California that
 installed the pilot-scale lime spray-drying system  considered an alkali
 slurry scrubbing  control  system but  decided against  use of such  a system
 for several  reasons.  First,  a wet scrubber would  require  380 to 473  liters
 (100  to 125 gallons)  of water  per  minute,  and  the plant might not always
 have  that much  water  available.  Second, the steam plume that occurs  as
 a  result  of the  use of  a wet  scrubber  might not  be acceptable to neighbors
 or  the  local  air  pollution control agency.   Third, the  kiln  electrostatic
 precipitator  is a  component of the production  process.   All  process
 materials  from  the raw  mill (about 180 Mg  [200 tons]  per hour) go  into
 the electrostatic  precipitator,  and  raw  feed that is  collected in  the
 electrostatic precipitator is  conveyed to  the  preheater  tower.   Gases
 are exhausted to  the  atmosphere.  Therefore, if  the electrostatic
 precipitator  were  to  be  shut down for  any  reason, the wet  scrubber,
 which would be downstream of the electrostatic precipitator,  would  also
 need  to be  shut down.   Fourth,  the sludge  from the wet  scrubber  would
 require disposal.  The  sludge  could be recycled, or it  could  be  converted
 to  gypsum  and, if the quality  of gypsum were satisfactory, combined with
 the ground  clinker.  However,  there would  be costs associated with
 drying  and  converting the sludge.  Because a dry lime spray-dryer would
 not require such a large water  supply, would produce no steam plume,
 could be placed upstream of the electrostatic precipitator, and disposal
 of the  dust would be relatively simple, Lone Star Industries decided
 that the  lime spray-dryer was  the better S02 control system for  its
Portland cement plant.M  (No wet limestone desulfurization systems have
been installed at portland cement plants, however.)

     In the dry lime injection system, lime or limestone is injected
 into the exhaust gas  where the S02 is absorbed into the lime.  The
                                    4-17

-------
resulting dry powder is collected in the particulate control device.
One kiln that has become subject to the NSPS since 1979 has a modified
precalciner with a limestone dust injection system to remove sulfur.53
The raw feed material at this plant includes kerogen-bearing shale,
which has a high sulfur content.

     4.3.1.2  Fabric Filters^and Electrostatic Precipitators.  EPA and
industry personnel have examined the possibility that fabric filters
used for particulate control can provide some control of S02 in industries
such as portland cement manufacturing where the particulate fabric
filter cake is alkaline in nature.   Studies of the industrial boiler
industry have shown that fabric filters located downstream of lime
spray-dryers effect from 5 to 30 percent of the overall S02 removal,
depending on the ratio of lime to S02, the approach temperature, and the
fabric filter pressure drop.54  These same studies report that, in
contrast, electrostatic precipitators located downstream of a lime
spray-dryer remove as much as 6 percent of the overal1 S02 removal.54

     Typical raw kiln feed contains about 75 percent calcium carbonate.
Typical kiln dust contains from 40 to 65 percent free and combined
calcium oxide depending on the process, degree of calcination, and
degree of reaction.55  Studies indicate that a fabric filter that controls
a kiln, or kiln and raw mill, and that collects the compounds mentioned
above may have some potential for S02 reduction through reaction of S02
to calcium sulfate.

     Industry personnel state, however, that, depending on the chemistry
of the filter cake, no significant S02 reduction may occur in the fabric
filter.  A fabric filter may have insufficient moisture present to allow
formation of calcium sulfate on the filter cake.56  If kiln exhaust
gases do not pass through a raw material mill prior to the fabric filter,
the filter would probably contain substantial amounts of calcium oxide,
which might absorb significant quantities of S02.57  If, however, kiln
exhaust gases do pass through a raw material mill prior to the fabric
filter, the filter cake may be primarily calcium carbonate, which may
not react appreciably with S02 at the high temperature and low humidity
found in a fabric filter.57

     In the raw mill itself, however, for raw feed with a high surface
area that is exposed to both the S02-laden gas stream and to 15 to 20
percent moisture levels up to 50 percent of the S02 is reported to be
absorbed into the raw materials as calcium sulfate.57,58

     Industry personnel also note that dry process plants have long gas
paths between the point of formation of S02 and the control device,
which allow potential absorption of S02 prior to the control device.18

     Recent studies are inconclusive regarding significant reduction in
S02 emissions through a fabric filter.5^,60  One study states that,
although S02 molecules would have substantially more contact with the
dust surface in a fabric filter than in an electrostatic precipitator,
particle reactivity seems to have a greater influence on S02 reduction
                                    4-18

-------
 than does particle/gas contact.61  Particle reactivity would not be
 significantly affected by the control  device used.   The study concluded
 that a fabric filter probably would not absorb significant amounts of
 S02  because the system studied currently provides poor S02 emissions
 absorption.62  However,  another study,  performed by Price Waterhouse for
 the  Portland Cement Association (PCA),  used sulfur material  balance data
 to show an average reduction  in potential  S02  emissions of 66 percent
 from plants using electrostatic precipitators  and 70 percent from plants
 using fabric filters.63   These reductions  include sulfur absorbed by the
 product and sulfur retained in the  dust collected by the control  device.

      No continuous monitoring or Method 6  inlet and outlet data  are
 available for facilities  subject to the NSPS,  however,  to assess  S02 ^
 removal  through a fabric  filter or  an electrostatic precipitator  on a
 cement kiln.   Figure 4-3  presents S02 outlet emission data for 19 of the
 30 cement kilns that have become subject to the NSPS since the 1979
 review.   The S02  emissions in Ib (kg) of S02 per ton (Mg) of raw  feed
 are  plotted versus the percent sulfur in the fuel  for both fabric filter
 and  electrostatic precipitator control.  These  data do  not show
 significantly lower S02 emissions from  kilns controlled by fabric filters
 than from kilns controlled by electrostatic precipitators.

      Information  on the amount by which particulate control  devices
 reduce S02  emissions  from cement kilns  is  inconclusive.   This  is  because
 many unpredictable factors affect emissions  such  as  the sulfur content
 of the feed,  the  point in the process at which  S02  removal occurs  (e.g.,
 clinker,  control  device,  exhaust gases), and the  relative importance  of
 process variables  (e.g.,  temperature, moisture,  feed  chemical  composi-
 tion).

      4.3.1.3   Process  Modifications.  Process modifications  that  can
 affect S02  emission  levels include  using the dry  rather than the wet
 process,  increasing  the oxygen  level in  the  kiln, reducing the percent
 sulfur in the  coal,  switching  to  raw feed materials  that  are low  in
 sulfur, and  use of  coal slurry.

      Approximately  6.5 xlO9 Joules  (5.6  xlO6 Btu's) are  required to
 manufacture  1  Mg  (1 ton)  of clinker in  a wet process  portland  cement
 plant.  In a dry process  plant,  only 4.6 xlO9 Joules  (4 xlO6 Btu's) are
 needed.  The added  coal required  in wet  process kilns increases the S02
 emission potential.  A nationwide 50-plant  survey sponsored by the PCA
 reported that  dry process kilns  emitted  half as much  S02  as wet process
 kilns.64  The  typical  S02 emission  level from the dry process electro-
 static precipitator-controlled plants is 23  kg/h (50  Ib/h) less than
that  from the wet process electrostatic precipitator-controlled plants.
The  increased  energy efficiency of the dry process results in substan-
tially decreased production costs.  Because of this energy cost savings,
the dry production process has become the predominant process in  the
Portland cement industry for  new plants.  Of the 30 kilns subject to the
NSPS since 1979, only 5 have  used the wet production process, and 3 of
the 5 are older kilns that were converted to coal-firing.
                                    4-19

-------
 Ib/ton
 (kg Mg)
 6.0
(3.or-
 4.0
(2.0!
 2.0.
(1.0)
a
        a
        a
     a
      B
                   a
        1    2-3
     PERCENT SULFUR
    FABRIC FILTER
                    1b/ton
                    (kg/Mg)
                   6.0
                   (3.0TT
 4.0
(2.0;
                   2.04.
                   (1.0)
                                  a
                  LEGEND
                D DRY PROCESS
                • WET PROCESS
                                           a
                           1    2    3
                        PERCENT SULFUR
        Figure 4-3. S02 emissions versus sulfur in coal
                          4-20

-------
      Unfortunately, the PCA survey did not take into account the control
 device the plant used or the level of sulfur in the fuel  or raw materials.
 Further study is needed to determine exact relationship between S02
 emissions and the type of production process.

      Several  California studies have shown that increased oxygen levels
 in the kiln will reduce S02 emissions.65,66  It is theorized that the
 S02 reacts with the oxygen to form S03,  which  reacts better with the
 alkali dust from the raw materials and is  absorbed by the clinker or the
 dust cake on  a fabric filter.65

      The oxygen level  in the kiln is easily controlled and would not
 involve any changes or additions  in equipment.   However,  operators  might
 strongly resist a requirement to  maintain  a specific oxygen level.   The
 oxygen level  presently used is based on  the individual  operator's
 experience with the level  that results in  a consistent product.   Varia-
 tion of oxygen level  at one plant and among plants is  unknown.   Also,
 increasing oxygen levels may increase nitrogen  oxide emissions.

      Because  coal  is  usually the  main source of sulfur in the portland
 cement process,  burning coal  with a low  sulfur  content is a simple  way
 to reduce potential  S02  emissions.   Most plants in this industry use
 coal  with less than 2  percent sulfur by weight.  Much  of  the growth in
 the portland  cement industry occurs  at sites in close  proximity  to
 sources of low sulfur  coal.67   One disadvantage to this method  is that
 lower sulfur  coal  is more  expensive  on the  average than high sulfur
 coal:

      Switching from raw  feed  materials with  a high sulfur content to
 those with a  low sulfur  contents  can  also  reduce potential  S02 emissions.
 For instance,  oil  shale  that  is occasionally used  as part of the  raw
 feed  materials  can  contain  high levels of  organic  materials rich  in
 sulfur.   Oil  shale  has  been  identified as a  major  source  of S02  emissions
 at  two  Colorado  cement plants.6   Oil  shale can  be  replaced by shale with
 lower sulfur  content or  replaced  by other silicon-rich materials such as
 clay,  sand, or  sandstone.

      Limestone can  contain  sulfur-rich pyrite (FeS2).  Pyrite has been
 identified  as  the cause  of  high S02 emissions at one California plant.17
 Limestone with pyrite could be replaced either  by  pyrite-free limestone
 or  other calcium-rich products, or the pyrite could be "washed" from the
 limestone  by allowing it to settle out when  the limestone  is pulverized
 and mixed with water.  The amount of pyrite  in  limestones around the
 country varies depending on mining techniques and  the purity of the
 limestone  formation.

     An estimated 50 percent of all industrial  waste products could be
 used  in the production of portland cement,  and these products are lower
 in sulfur than most natural fuels.  Because most portland cement plants
are near population and industrial centers,  industrial waste products as
raw feed could be readily available and reduce the  reliance of the plant
on natural materials.
                                    4-21

-------
     Although many natural and man-made substitutes are available for
 limestone and shale, the costs of replacement materials or "cleaning" of
 the  raw materials could be higher than costs for raw materials presently
 being used.  Because of the high cost of transporting cement, port!and
 cement plants must be located near areas of product demand and must make
 do with the raw materials locally available in order to keep costs
 reasonable.  Also, some raw materials that increase potential S02
 emissions are attractive to plant operators for other reasons.  For
 example, the use of oil shale reduces energy costs because oil shale
 adds combustible products that help fire the kiln.  Also, the iron in
 pyrite-rich limestone is a necessary ingredient for portland cement
 production.  Although the relationship between the chemical composition
 of different raw materials and the amount of S02 emissions has been
 shown in some parts of the country, further study is needed to determine
 if this relationship exists nationwide.

     Use of coal slurry has been estimated by one vendor to reduce S02
 emissions up to 90 percent compared to emissions from other fuel
 sources.68  Pulverized coal is mixed with water to allow any pyrite in
 the coal to settle out.   A commercial coal slurry process has been
 demonstrated on asphalt concrete plants in Illinois and is being promoted
 for portland cement plants.  The vendor of the commercial process stated
 that coal slurry produced in this process is two to three times less
 expensive than fuel oil.68  If low sulfur coal is unavailable or
 prohibitively expensive, this process may be a cost-effective way of
 reducing the sulfur content of other coals.

     The cost of a new portland cement plant would increase by the
 amount required to build the coal slurry process plant unless a local
 commercial coal  slurry process were available.   Costs would be incurred
 for replacing or modifying the burners in plants to burn coal slurry.
 Exact costs for the use of a coal slurry in a portland cement plant are
 unknown because no coal  slurry process plant large enough to supply fuel
 for a portland cement plant has been built.   However, the costs would be
 higher for processing and firing coal slurry than for firing coal  dry.
 Disposal costs might increase because of the increased solid waste from
 the coal slurry production process.

     Because excessive sulfur in the clinker can cause cracking of the
 final cement product, ASTM has set recommended standards for the amount
 of sulfur allowed in the clinker.  Plant operators meet clinker specifica-
 tions (and, as a side benefit, reduce potential  S02 air emissions) by
 limiting the sulfur content of raw materials and fuels.

4.3.2  Nitrogen  Oxides

     At least six kilns  that have become subject to the NSPS since 1979
 have nitrogen oxides (NO ) monitors.

     Since the 1979 review of the NSPS,  research has been conducted on
N0x emission reduction by combustion modification techniques on a  pilot-
 scale system,  a  small  scale kiln, and a full  scale kiln.   This study has


                                    4-22

-------
shown that N0x emissions may be reduced through recirculation  of  flue
gases into the primary combustion air of the kiln.69   Recirculation  of
flue gases reduces the local oxygen content in the kiln  flame, which, in
turn, lowers the flame temperature.  The lower flame temperature  reduces
N0x emissions that are caused by coal burning.

     Although the process modifications tested did reduce NO   emissions,
the effects of flue-gas recirculation could not be separatedxfrom
variations in process parameters.   Therefore, additional research is
needed to demonstrate the effectiveness of flue gas recirculation or
other NO  emission reduction methods.
        /\

4.4  REFERENCES FOR CHAPTER 4


 1.   U.  S.  Environmental  Protection Agency.   Response to Remand Ordered
     by U.  S.  Court of Appeals for the District of Columbia in Portland
     Cement Association v.  Ruckelshaus.   Publication EPA-450/2-74r023
     November 1974.   p.  7-14.

 2.   U.  S.  Environmental  Protection Agency.   Control  Techniques for
     Particulate  Emissions  from Stationary Sources.   Volume II.  Publi-
     cation No.  EPA-450/3-81-005b.   September 1982.   p. 9.7-78.

 3.   Phillips,  N.  and W.  Brumagin.   Fabric Filters in the Cement Industry.
     Pit and  Quarry.   Volume 73.   Number 1.   July  1980.  p.  96.

 4.   U.  S.  Environmental  Protection  Agency.   Inspection Manual  for
     Enforcement  of  New Source  Performance  Standards.   Portland Cement
     Plants.   Publication No.  EPA-340/1-75-001.  September  1975.   p.  3-13.

 5.   Reynolds,  J.  and L Theodore.   Analysis  of an APCA Baghouse  Operation
     and Maintenance  Survey.  Journal  of the  Air Pollution  Control
     Association.  November  1980.  pp. 1255-1257.

 6.   Telecon.   Clark,  C., MRI,  to Clouse, J. , Colorado  Air  Pollution
     Control Division.  October 13,  1983.  Discussion of  cement plants
     in  Colorado.

7.   Murray, J. and C. Rayner.  The  State-of-the-Art of Dust Collectors
     on  Preheater Kilns.  (Presented at  the 22nd IEEE Cement Industry
     Technical Conference.   Toronto, Ontario, Canada.   May 19-22,  1980 )
     pp.  15-16.

8.   Bundy, R.  Operation and Maintenance of Fabric Filters.  Journal of
    the Air Pollution Control Association.  July  1980.   p. 757.

9.  Barrett, K.  A Review of Standards of Performance  for New  Stationary
    Sources—Portland Cement Industry.  U. S. Environmental Protection
    Agency.  Publication No. EPA-450/3-79-012.   March  1979.  pp  4-21
    4-22.
                                   4-23

-------
10.  Reference 7, p. 5.

11.  Reference 9, p. 4-20.

12.  Comments and attachments presented by Brown, R., Environmental
     Elements Corp., to the National Air Pollution Control Techniques
     Advisory Committee.  August 30, 1984.   Response to the NSPS  for
     Portland cement plants,  p. 5, 7-9.

13.  Letter from Lotz, W., Lehigh Portland Cement Company, to Farmer, J.,
     EPA/OAQPS.   September 6, 1984.  Transmitting summary of comments
     presented to the National Air Pollution Control Techniques Advisory
     Committee on August 30, 1984.   p.  2, 3.

14.  Letter from Greer, W., Lone Star Industries, Inc., to Cuffe, S. ,
     EPA/ISB.  June 22, 1984.  Response to request for comments on  the
     draft review document,  p.  5.

15.  PEDCo Environmental,  Inc.  Technical Assistance to the State of
     Iowa--Excess Emissions at Lehigh Cement, Mason City, Iowa.   EPA
     Contract No. 68-01-6310.  February 1984.  p. 59.

16.  Reference 15, p. 61.

17.  Reference 15, pp. 93-94.

18.  Comments presented by Orem, S., Industrial Gas Cleaning Institute,
     to the National Air Pollution Control  Techniques Advisory Committee.
     August 30,  1984.  Response to the NSPS for portland cement plants.
     p. 1.

19.  Reference 12, pp. 4,  5.

20.  Comments presented by von Seebach, M. , Polysius Corp., to the
     National Air Pollution Control Techniques Advisory Committee.
     August 30,  1984.  Response to the NSPS for portland cement plants.
     p. 3.

21.  Reference 13, p. 3.

22.  Letter from Riley, J., Lurgi Corp., to Farmer, J., EPA/OAQPS.
     August 31,  1984.  Transmitting summary of comments to have been
     presented to the National Air Pollution Control Techniques Advisory
     Committee on August 30, 1984.   p.  1.

23.  Petersen, H.  New Trends in Electrostatic Precipitation:  Wide Duct
     Spacing, Precharging, Pulse Energization.  (Presented at the 22nd
     IEEE Cement Industry Technical Conference.  Toronto, Ontario,
     Canada.   May 19-22, 1980.)  pp.  4-14.

24.  Reference 23, p. 14.
                                    4-24

-------
    25.  U. S. Environmental Protection Agency.   Control  Techniques  for
         Participate Emissions from  Stationary  Sources.   Volume  I.
         Publication No. EPA-450/3-81-005a.   September  1982.   p.  4.2-6.

    26.  Reference 16, pp. 4.2-19 -  4.2-30.

    27.  Reference 9, pp. 4-17 - 4-19.

s  28.  U. S. Environmental Protection Agency.   Portland Cement  Plant
A       Inspection Guide.  Publication No. EPA-340/1-82-007.  June  1982.
         p. 27.

    29.  Comments presented by Prior, 0., F.  L.,  Smidth and Company,  to  the
         National Air Pollution Control Techniques Advisory Committee.
         August 30, 1984.  Response  to the  review of NSPS for  portland
         cement plants,   p.  3.

    30.  Reference 25, p. 8-32.

    31.  Reference 9, pp. 4-22 - 4-25.

    32.  Telecon.  Clark, C.-, MRI, with Frances,  J. , Nebraska  Department of
         Environmental Control.  November 4, 1983.  Discussion about  Ash
         Grove Cement Company,  Louisville, Nebraska.

    33.  Herod, S.  Martin Marietta Clears the Air Over Colorado  Plant.  Pit
         and Quarry.   July 1981.   p.  85.

    34.  Telecon.  Clark, C., MRI, with Johnson,  H.,  Bay  Area Air Quality
         Management District.   October 27, 1983.  Discussion about Kaiser
         Cement Corp., Permanente, California.

    35.  Armstrong, J.,  P.  Russell,  and M. Plooster.   Balloon-Borne Particulate
         Sampling of the Glens  Falls  Portland Cement Plant Plume.  EPA Grant
         No.  R805926-01.   July  16, 1979.

    36.  Information from Fulton,  R., Jefferson County Bureau of Environmental
         Health,  to Clark,  C.,  MRI.   January 16, 1984.   Formation of  a
         Detached Plume  in the  Exhaust Gas of a Portland Cement Kiln.  EPA
         Contract No.  68-01-4146.   September 1981.

    37.  Winberry, W., Jr.,  Engineering-Science, Inc.   Survey of Detached
         Plumes With Potential  Formation  of Inhalable Particulates.    Durham,
         North Carolina.   June  1983.

    38.  Information  from Clouse,  J., Colorado Air Pollution Control   Division,
         to Clark, C., MRI.   November 10,  1983.   Results for February 25 and
         26,  1982, roller mill/clinker cooler tests  at Ideal  Basic Industries,
         Inc., Fort Collins,  Colorado.
                                        4-25

-------
39.   Information  from  Phelps, D.,  Iowa  Department  of  Water,  Air,  and
     Waste Management, to Clark, C., MRI.   November 4,  1983.   Emission
     test summary for  Davenport  Cement.

40.  Telecon.  Eddinger, J., EPA/ISB, with  Thoits, F.,  Monterey  Bay
     Unified Air  Pollution  Control  District.  August  27,  1984.   Discussion
     about visible emissions.

41.  Letter and attachment  from  Miller,  E. , Oregon Portland  Cement
     Company, to  Farmer, J., EPA/OAQPS.   November  22, 1983.   Response  to
     Section 114  information request,   p. 21.

42.  Telecon.  Clark,  C., MRI, with  Lewis M., and  J.  Macias,  Texas  Air
     Control Board, Region  8.  December 7,  1983.   Discussion  about
     cement plants in  the Region.

43.  Portland Cement Association.   Energy Report:  U.S. Portland  Cement
     Industry.  Skokie,  Illinois.  October  1983.

44.  Information from  J. Clouse, Colorado Air Pollution Control  Division,
     to C. Clark, MRI.   November 10, 1983.  Notice of intent  to  construct
     and operate Martin  Marietta Cement,  Lyons, Colorado,  p.  15.

45.  Information from  P. Bosserman,  Department of  Environmental Quality,
     Oregon, to C. Clark, MRI.   January  9,  1984.   Summary of  source  test
     results for an Oregon  portland  cement  plant.

46.  Ketels, P., 0.  Nesbitt, and R.  Oberle, (Institute-for Gas Technology.)
     Survey of Emissions Control and Combustion Equipment Data in Industrial
     Process Heating.   Preapred for  U.   S. Environmental Protection
     Agency.  Research Triangle Park, North Carolina.    Publication
     No. EPA-600/7-76-022.   October  1976.   p. 72.

47.  Kelly, M. and S.   Shareef (Radian Corporation).  Third Survey of Dry
     S02 Control Systems.   Prepared  for  U.  S. Environmental Protection
     Agency.  Research Triangle Park, North Carolina.    Publication
     No. EPA-600/S7-81-097.   August  1981.   p. 4.

48.  Apple, C. and M.   Kelly.  Mechanisms of Dry S02 Control Processes.
     U. S. Environmental Protection  Agency.  Publication
     No. EPA-600/S7-82-026.   June 1982.  p. 2.

49.  Reference 9, p.  4-12.

50.  U. S. Environmental Protection Agency.  Multimedia Assessment and
     Environmental Research Needs of the Cement Industry.   Publication
     No. EPA-600/2-79-111.   Cincinnati, Ohio.  May 1979.  p.   53.

51.  Memorandum from Clark,  C.,  MRI, to Tabler, S., EPA/SDB.   January  30,
     1985.  Minutes of October 2, 1984, discussion about the  Lone Star
     flue-gas desulfurization system.
                                    4-26

-------
 52.   Telecon.   M. Maul, MRI, to D. Seaward, Wheelabrator-Frye,  Inc.
      March 30, 1984.   Alkaline slurry injection system costs for portland
      cement plants.

 53.   Memorandum from Maxwell, C., MRI, to Glowers, M., EPA/ISB.  November  18,
      1983.   Trip report:   Ideal Basic Industries, Inc., Fort Collins,
      Colorado, on November 8, 1983.

 54.   Sedman, C.   Performance of Spray Drying FGD Systems.  Draft.  U. S.
      Environmental  Protection Agency.  Research Triangle Park, North
      Carolina.  August 8,  1984.  p.  13.

 55.   Letter from Gebhardt, R.,  Lehigh Portland Cement Company, to Cuffe, S.,
      EPA/ISB.   June  6, 1984.   Response to request for comments on draft
      review document,   p.  4.

 56.   Reference 12, p.  4.

 57.   Reference 14, p.  6.

 58.   Reference 20, p.  2.

 59.   Reference 12, pp.  2-4.

 60.   Energy and  Resource Consultants, Inc.   Background Document on SO
      and  NO Emissions From  Five  Industrial  Process  Categories.       x
      September 20, 1982.   Prepared for the  Office of Technology Assess-
      ment.   Washington, D.C.  p.  2-12.

 61.   KVB,  Inc.   An Evaluation of  S02  Removal  Across  a Fabric  Filter.
      July  1983.   Prepared  for Lone Star  Industries,  Inc.   p.  21, 30.

 62.   Reference 61, p.  23.

 63.   Comments  presented by Gebhardt,  R.,  Lehigh  Portland  Cement Company,
      to the  National Air Pollution Control  Techniques Advisory  Committee.
      August  30,  1984.  Response to the review of NSPS for  portland
      cement  plants,  p. 4.

 64.   Letter  from  Schneeberger,  C., Portland Cement Association  (PCA),  to
      Maxwell,  C., MRI.  November 17,  1983.  Transmittal of  PCA  report:
      Kiln Gaseous Emissions Survey.   August 18,  1983.   p.  2 of  6.

 65.   Reference 47, pp. 20, 24,  25.

 66.   KVB,  Inc.   Emissions  Reduction by Advanced  Combustion  Modification
     Techniques for Industrial  Combustion Equipment.   Prepared  for U.  S.
     Environmental Protection Agency  Industrial  Advisory Panel Meeting.
     June 8, 1983.  p. 58.

67.  Wark, K. and C.  Warner.   Air  Pollution:   Its Origin and Control.
     2nd Edition.  New York,  Harper and Row.  1981.   p. 351.
                                    4-27

-------
68.   Duncan, C.  and V. Swanson.  Organic-Rich Shale of the U.S. and
     World Land Areas.  U.S. Geological Survey.   Circular 523.119.4/2:53.
     U.S.  Department of Interior.  Washington, D.C.  1965.  pp. 1-27.

69.   KVB,  Inc.   Application of Combustion Modification Technology for
     NO  Control to Cement Kilns.  Prepared for U. S.  Environmental
     Protection Agency.
                                    4-28

-------
                              5.  COST ANALYSIS

     This chapter presents the costs of complying with the new source
performance standards  (NSPS) for affected portland cement facilities.

5.1  APPROACH

     Model facility parameters were established to represent the range
of facilities that have become subject to the NSPS since the 1979 review.1
Capital and annualized costs of emission control equipment for the model
facilities were estimated using the CARD manual (in December 1977
dollars).2  These costs were updated to July 1983 dollars using the
Chemical Engineering fabricated equipment cost index and statistics from
the Bureau of Labor and the Bureau of Industrial Economics.3-5

     The capital cost of a control system includes the cost of design
and installation of the major control device and of auxiliaries such as
fans and instrumentation; the cost of foundations, piping, electrical
wiring, and erection; and the cost of engineering construction overhead
and contingencies.6,7

     The annualized cost of a control system represents the yearly cost
to the company of owning and operating the system.   This cost includes
direct operating costs such as labor, utilities, and maintenance and
capital related charges such as depreciation, interest, administrative
overhead, property taxes, and insurance.   Annualized costs presented in
this chapter do not include credits for product recovery.6/

     The estimated capital  and annualized costs of emission control
equipment are presented in Section 5.2.   A comparison of estimated and
reported capital costs is presented in Section 5.3.   Cost effectiveness
of emission control  is presented in Section 5.4.

5.2  ESTIMATED CAPITAL AND ANNUALIZED COSTS OF EMISSION CONTROL

     Estimated capital  and annualized costs for each of 17 model  facilities
(7 kilns,  3 clinker coolers,  and 7 other facilities, [i.e.,  raw mill,
blending silos,  clinker storage,  2 finish mills,  cement storage and
transfer facilities]) are presented in the following subsections.
                                     5-1

-------
5.2.1  Kiln

     Model kiln facilities were developed to represent emission control
by fabric filters and by electrostatic precipitators on kilns installed
since 1979.   Parameters describing the facilities and emission control
equipment are presented in Table 5-1.   The exhaust gas flow rate is the
critical parameter for costing both types of emission control equipment.
Exhaust gas flow rates for each model  kiln were developed using reported
air flow data from industry and State and local agencies in combination
with design flow data from a control equipment manufacturer.8

     Table 5-2 presents the estimated capital and annualized costs in
July 1983 dollars of control equipment for each of the seven model kiln
facilities.   Kilns A, B, and C represent small, medium, and large kilns,
respectively, with exhaust gases controlled by fabric filters.   Kiln D
represents a medium kiln with a main fabric filter emission control
system and an alkali-bypass fabric filter control system (which handles
about 25 percent of the exhaust gases).   Kilns E, F, and G represent
small, medium, and large kilns, respectively, with exhaust gases
controlled by electrostatic precipitators.

5.2.2  Clinker Cooler

     Most of the 23 clinker cooler facilities subject to the NSPS since
the 1979 review are controlled by fabric filters.

     Fabric filter control of a clinker cooler was evaluated for three
sizes of model facilities:  small, medium,  and large.  Parameters that
describe the fabric filters used to control clinker cooler exhaust gases
are shown in Table 5-3.  The data for exhaust gas flow rates and for
other parameters were derived from data for similar facilities that have
become subject to the NSPS since 1979.

     Table 5-4 presents the estimated capital and annualized costs in
July 1983 dollars of control equipment for each of the model clinker
cooler facilities.

5.2.3  Other Facilities

     Other affected facilities (mills,  storage, and transfer facilities)
at portland cement plants are subject only to a visible emissions limit
under the NSPS of less than 10 percent opacity.  Fabric filters are used
to control emissions from most of these facilities that are subject to
NSPS since the 1979 review.  Two plants have finish mills controlled by
electrostatic precipitators.

     Capital and annualized costs were estimated for six model  facilities
(raw mill, blending silo, clinker storage,  finish mill, cement storage,
and transfer) controlled by fabric filters and for one model facility
(finish mill) controlled by an electrostatic precipitator.   Parameters
describing the emission control equipment for each model facility are
presented in Table 5-5.  These parameters were based on reported data
                                     5-2

-------
 from medium-size  facilities  that  have  become  subject  to  the  NSPS  since
 1979 and  are  representative  of  facilities  at  a  medium size plant
 (544,000  Mg/yr  [600,000 tons/year]  per kiln).

     Table 5-6  presents the  capital and annualized  costs  of  particulate
 emission  control  equipment for  these model  facilities.   Because a portland
 cement plant  will  have more  than  one of several of  the facilities  shown
 in Table  5-6, total plant capital costs for control of other facilities
 can be substantially more than  the  sum of  the individual  costs in the
 table.  Two plants with facilities  that have  become subject  to the NSPS
 since 1979 reported more than 50  fabric filter  control devices (at each
 plant) controlling exhaust gases  from  these other facilities.  Based on
 the capital costs  for fabric filter control of  the  individual facilities
 shown on  Table  5-6, the total plant capital cost of such  fabric filter
 systems is estimated to be $5,000,000  per  plant.

 5.3  COMPARISON OF ESTIMATED AND  REPORTED  CAPITAL COST DATA

     Reported capital cost data were obtained from  individual plants.
 These data were converted to 1983 dollars  for comparison  with the
 estimated capital cost data presented  in Section 5.2.9-12  Table  5-7
 presents  the estimated and reported capital costs by  facility size  and
 type of control equipment.

 5.4  COST EFFECTIVENESS

     Table 5-8.presents the cost effectiveness of the 17  model facilities.
 Cost effectiveness is the annualized cost  of emission control divided by
 the annual emission reduction.   Uncontrolled and NSPS (allowable)
 particulate emissions are calculated for each model  facility by
 multiplying the raw material  feed rate by  an emission factor.1^,14
 Uncontrolled emissions are assumed to be those exiting a  product
 recovery cyclone.  The annual emission reduction is calculated as
 uncontrolled emissions minus  NSPS (allowable) emissions.

     The cost effectiveness of controlling emissions from kilns ranges
 from $34 to $50 per Mg of emissions ($31 to $45 per ton).  The cost
 effectiveness of controlling  emissions from clinker coolers ranges  from
 $27 to $44 per Mg ($25 to $40 per ton).  The cost effectiveness of
 controlling emissions from other facilities was estimated to range  from
$30 to $167 per Mg ($27 to $151 per ton).
                                     5-3

-------
                                 TABLE  5-1.  SUMMARY OF MODEL KILN FACILITY PARAMETERS
en
i
Model
Size
facility

Produc- Exhaust gas
tion rate. rate, mVmin
Mg (tons) (acfm)
Temp, inlet/ Pressure
outlet, drop, .
°C (°F) in. WG Other parameters
I. Fabric filter control
A Small:

B Medium:

C Large:

D Medium
alkali




272,000
(300,000)
544,000
(600,000)
1,089,000
(1,200,000)
with
bypass:
544,000
(600,000)


4,250
(150,000)
8,500
(300,000)
17,000
(600,000)
Main:
7,650
(270,000)
Bypass:
2,270
(80,000)
246 (475)7
149 (300)
246 (475)7
149 (300)
246 (475)7
149 (300)
232 (450)7
149 (300)

260 (500)7
149 (300)

6 For all fabric filter con-
trolled facilities:
6 Air-to-cloth ratio: 1.3:1 to
1.5:1; 7,200 h/yr operation;
6 fiberglass bags; reverse air
cleaning.
5


5


II. Electrostatic precipitator control
E Small:

F Medium:


G Large:



272,000
(300,000)
544,000
(600,000)

1,089,000
(1,200,000)


3,540
(125,000)
7,080
(250,000)

14,160
(500,000)


177 (350)7
177 (350)
177 (350)7
177 (350)

177 (350)7
177 (350)


5 For all electrostatic precipi
tator controlled facilities:
5 precipitator efficiency =
99.95%; precipitation rate
parameter =5.5 m/min
5 (18 fpm); SCA = 1.4 m2 per
mVmin (420 ft2 per
1,000 acfm); 7,200 h/yr
operation.
      .Megagrams (tons) of clinker produced per year per kiln.

       m3/min = Actual cubic meters per minute; acfm = actual cubic feet per minute.
       Temperature estimated at inlet to and outlet of control device.

-------
TABLE 5-2.  ESTIMATED CAPITAL AND ANNUALIZED COSTS OF PARTICULATE EMISSION
                CONTROL EQUIPMENT FOR MODEL KILN FACILITIES
Model kiln
I. Fabric filter
A
B
C
D
II. Electrostatic
E
F
G
Model
facility type
control
Small
Medium
Large
Medium, with
alkali bypass
precipitator control
Small
Medium
Large
Capital
cost, $a
1,925,000
3,572,000
10,344,000
3,904,000
2,212,000
3,542,000
8,748,000
Annual ized
cost, $
548,000
924,000
2,030,000
1,099,000
480,000
765,000
1,615,000
aJuly 1983 dollars.
                                     5-5

-------
                           TABLE  5-3.  SUMMARY OF MODEL CLINKER COOLER FACILITY PARAMETERS
cr>
cl
A
B
C
Model
inker cooler
Small:
Medium:
Large:
Model
facility size,
Mg (tons)3
272,000
(300,000)
544,000
(600,000)
1,089,000
(1,200,000)
Exhaust
gas flow
rate, mVmin
(acfm)D
2,830
(100,000)
5,660
(200,000)
9,910
(350,000)
Temp.
inlet/
outlet,
°C (°F)
177 (350)/
121 (250)
204 (400)/
121 (250)
204 (400)/
121 (250)
Pressure
drop,
in. W.G.
4
6
6
Air-to-
cloth
ratio
5:1
5:1
5:1
Other
parameters
For all model
clinker cooler
facilities:
7,200 h/yr of
operation;
Nomex bags;
pulse jet
cleaning.
      .Megagrams  (tons)  of  clinker produced per year per kiln.
       mVmin  = actual cubic  meters per minute; acfm = actual cubic feet per minute.
       Temperature  estimated  at  inlet to and outlet of control device.

-------
TABLE 5-4.  ESTIMATED CAPITAL AND ANNUALIZED COSTS OF PARTICULATE EMISSION
          CONTROL EQUIPMENT FOR MODEL CLINKER COOLER FACILITIES


                                     Model         Capital,      Annualized
Model clinker cooler             facility size     cost, $       cost, $


          A                           Small         931,000        321,000

          B                          Medium      1,731,000        523,000

          C                           Large      2,959,000        800,000

.Exhaust gases controlled by fabric filter.
DJuly 1983 dollars.
                                     5-7

-------
                            TABLE  5-5.  SUMMARY OF PARAMETERS FOR MODEL OTHER FACILITIES
en
i
oo
Exhaust
gas flow rate,
Model other facility mVmin (acfm)
I. Fabric filter control
Raw mill 1,130
(40,000)
Blending silo 280
(10,000)
Clinker storage 170
(6,000)
Finish mill 710
(25,000)
Cement storage 420
(15,000)
Transfer facility 70
(2,500)
II. Electrostatic precipitator control
Finish mill 710
(25,000)







Temp.
inlet/outlet
°C (°F)C

88 (190)/
38 (100)
38 (100)/
38 (100)
38 (100)/
38 (100)
121 (250)/
52 (125)
38 (100)/
38 (100)
38 (100)/
38 (100)

121 (250)/
93 (200)







Pressure Air-to-
drop, cloth .
in. W.G. ratio Other parameters

4 7:1 For fabric filter-controlled
facilities:
4 • 5:1 5,000 h/yr operation (except
7,200 h/yr for finish mill);
4 8:1 polyester or Dacron bags;
pulse jet cleaning.
4 6:1

4 7:1

4 6:1


5 -- For electrostatic precipi-
tator controlled facilities:
precipitator efficiency
= 99.95%; precipitation rate
parameter =5.5 m/min
(18 fpm); SCA = 1.4 m2 per
mVmin (420 ft2 per
1,000 acfm); 7,200 h/yr
operation.
      All  facilities are  representative of facilities at a medium-size plant (i.e., 544,000 Mg of clinker
     .produced per year per  kiln [600,000 tons/yr]).
      rnVmin = actual cubic  meters per minute; acfm = actual cubic feet per minute.
      Temperature estimated  at inlet to and outlet of control device.

-------
TABLE 5-6.  ESTIMATED CAPITAL AND ANNUALIZED COSTS OF PARTICULATE EMISSION
               CONTROL EQUIPMENT FOR MODEL OTHER FACILITIES
Model
I.






II.

other facility3
Fabric filter control
Raw mill
Blending silo
Clinker storage
Finish mill
Cement storage
Transfer facility
Electrostatic precipitator control
Finish mill
Capital.
cost, $

344,000
154,000
84,000
254,000
146,000
68,000

914,000
Annual i zed
cost, $

102,000
54,000
38,000
81,000
54,000
34,000

189,000
a.,, , •-[•+• 4. 4. • *• -i-4.-
 All facilities are representative of facilities at a medium-size plant
 (i.e., 544,000 megagrams of clinker produced per year per kiln
,[600,000 tons/yr]).
 July 1983 dollars.
                                    5-9

-------
       TABLE 5-7.  COMPARISON OF ESTIMATED CAPITAL COSTS OF EMISSION
          CONTROL WITH REPORTED CAPITAL DATA COSTS (FROM INDUSTRY)

                                   	Capital cost, 1983 $
                                    Estimated                  Reported
Model facility

Kiln               .
     A.   Small (FF)D                1,925,000                 1,500,000
     B.   Medium (FF)                3,572,000
     C.   Large (FF)                10,344,000                14,000,000^
     D.   Medium, with               3,904,000                 l,000,000a
           alkali bypass (FF)
     E.   Small (ESP)D               2,212,000
     F.   Medium (ESP)               3,542,000                 3,200,000,
                                                              3,500,000
     G.   Large (ESP)                8,748,000                 4,700,000

Clinker cooler
     A.   Small (FF)                   931,000
     B.   Medium (FF)                1,731,000                 1,000,000
     C.   Large (FF)                 2,959,000

Other facilities
     (FF)                              68,000                    42,000
                                   to 344,000                 to 73,000
                                                                 28,000,
                                                             and 73,000
     (ESP, finish mill)               914,000                   850,000

 Reported capital costs from industry (References 9-12).  Reported costs
 are installed costs of control systems (assumed to include the cost of
.the control device and all auxiliaries).
 FF = Fabric filter; ESP = electrostatic precipitator.
 .Control system for large kiln and clinker cooler.
 Cost of alkali bypass system only.
/Two-year-old electrostatic precipitator purchased from another company.
 The $28,000 capital cost is the average for 14 dust collectors.
                                   5-10

-------
               TABLE  5-8.    COST  EFFECTIVENESS OF  PARTICIPATE EMISSION
                                  REDUCTION  BY  MODEL  FACILITIES
Model facility3
I. Kiln
A. Small (FF)

B. Medium (FF)

C. Large (FF)

D. Medium, with
alkali bypass (FF)
E. Small (ESP)

F. Medium (ESP)

G. Large (ESP)

II. Clinker cooler
A. Small (FF)

B. Medium (FF)

C. Large (FF)

III. Other facilities1
Raw mill (FF)

Blending silo (FF)

Clinker storage (FF)

Finish mill (FF)

Cement storage (FF)

Transfer facility (FF)

Finish mill (ESP)

Facility
size,
Mg/yr .
(tons/yr)D

272,000
(300,000)
544,000
(600,000)
1,089,000
(1,200,000)
544,000
(600,000)
272,000
(300,000)
544,000
(600,000)
1,089,000
(1,200,000)

272,000
(300,000)
544,000
(600,000)
1,089,000
(1,200,000)















Annual-
ized
cost,
$/yrc

548,000

924,000

2,030,000

1,099,000

480,000

765,000

1,615,000


321,000

523,000

800,000


102,000

54,000

38,000

81,000

54,000

34,000

189,000

Particulate d
emissions, Mq/yr (tons/yr)
Un-
controlled

11,200
(12,300)e
22,300
(24,600)e
44,600
(I9,200)e
22,300 .
(24,600)
11,200 .
(12,300)e
22,300
(24,600)e
44,600 a
(49,200)

7,300
(8.040)9
14,600
(16.100)9
29,200
(32.200)3

1,420 .
(1,570)J
1,420 .
(1,570)J
1,420 .
(1,570)J
1,420 .
(1,570)J
1,420 .
(1.570)3
1,420 .
(1,570)J
1,420 .
(1,570)J
NSPS

73 f
(80)r
146 f
(161)
292 f
(322)T
146 f
(161)
73 f
(SO/
146 f
(161)
292 f
(322)r

24 h
(27)h
49 h
(53)n
97
(107)"

286 .
(315)k
286
(315)*
286 ..
(315)K
286 .,
(315)K
286
(315r
286 .,
(315)"
286
(315)K
Emission
reduction

11,100
(12,200)
22,100
(24,400)
44,300
(48,900)
22,100
(24,400)
11,100
(12,200)
22,100
(24,400)
44,300
(48,900)

7,250
(8,000)
14,500
(16,000)
29 , 100
(32,100)

1,130
(1,250)
1,130
(1,250)
1,130
(1,250)
1,130
(1,250)
1,130
(1,250)
1,130
(1,250)
1,130
(1,250)
Cost
effectiveness
STRg ($/ton)

49

42

46

50

43

34

36


44

36

27


90

48

34

72

48

30

167


(45)

(38)

(42)

(45)

(39)

(31)

(33)


(40)

(33)

(25)


(82)

(43)

(30)

(65)

(43)

(27)

(151)

&FF = Fabric filter;  ESP  = electrostatic  precipitator.
 Megagrams  (tons) of  clinker produced per year per kiln.
iJuly 1983  dollars.
 1.7 Mg kiln feed per Mg  cement; 1.05 Mg  cement per Mg  clinker; Reference 14.
fKiln:   23  kg of particulate emissions per Mg of kiln feed (45 lb/ton);  Reference 13
'Kiln:   NSPS limit of 0.15 kg/Mg kiln feed (0.30 lb/ton).
^Clinker cooler:  15  kg of particulate emissions per Mg of kiln feed (30 lb/ton); Reference
10.
•Clinker  cooler:  NSPS  limit of 0.05 Kg/Mg kiln feed (0.10 lb/ton).
 All  other facilities are  representative of facilities  at medium-size  plants that have  become subject  to the
.NSPS since 1979.
J0ther facilities:   estimated 10 kg of  particulate emissions per Mg  of cement (20 lb/ton) (Reference 14).  Assumed
k?5 percent control  efficiency if controlled by cyclone dust collector.
 Other facilities:   estimated 10 kg/Mg  of cement or 20  lb/ton (Reference 14), and assumed 95 percent control
 efficiency (minimum) if controlled by  fabric filter, i.e. a conservative estimate used to calculate maximum cost
 effectiveness.
                                                   5-11

-------
5.5  REFERENCES FOR CHAPTER 5

 1.  Memorandum from Maxwell, C.,  MRI, to Eddinger, J., EPA/ISB.  August 6,
     1984.   Model facility sizes and exhaust gas characteristics.

 2.  Neveril, R. (Gard, Inc.) Capital and Operating Costs of Selected
     Air Pollution Control Systems.  Prepared for the U.S. Environmental
     Protection Agency.  Publication No.  EPA-450/5-80-002.  December
     1978.

 3.  Economic Indicators.   Chemical Engineering.  November 28, 1983.
     p.  7.

 4.  Producer Prices and Price Indexes for Commodity Groupings and
     Individual Items.   U.S.  Department of Commerce.  Bureau of  Labor
     Statistics.  November 1983.  pp. 90, 112, 114, 115.

 5.  1983 U.S.  Industrial  Outlook for 250 Industries With Projections
     for 1987.   U.S. Department of Commerce.   Bureau of Industrial
     Economics.  January 1983.  p. 2-8.

 6.  Memorandum from Maxwell, C.,  MRI, to Eddinger, J. , EPA/ISB.  August 6,
     1984.   Capital and annualized costs of air pollution control equipment.

 7.  Memorandum from Maxwell, C.,  MRI, to Eddinger, J. , EPA/ISB.  August 6,
     1984.   Revised cost analysis.

 8.  Telecon.  Maxwell, C., MRI, to Kreisberg, A., Fuller Company.
     April  12,  1984.  Discussion of design parameters for cement  kiln
     control equipment.

 9.  Letter and attachments from Miller,  E.,  Oregon Portland Cement
     Company, to Farmer, J.,  EPA/OAQPS.   November 22, 1983.  Response to
     Section 114 information request,  p. 35.

10.  Letter and attachments from Powledge, H., Ideal Basic Industries,
     Inc.,  to Farmer, J.,  EPA/OAQPS.  December 30, 1983.  Response to
     Section 114 information request,  p. 97.

11.  Letter and attachments from Gebhardt, R., Lehigh Portland Cement
     Company, to Farmer, J.,  EPA/OAQPS.   January 6, 1984.  Response to
     Section 114 information request,  pp. 10, 14, and 15.

12.  Letter and attachments from Greer,  W.,  Lone Star Industries, Inc.,
     to Farmer, J., EPA/OAQPS.  March 30, 1984.  Response to Section 114
     information request,   pp. 12, 13 and 51.

13.  U.  S.  Environmental Protection Agency.   Background Information for
     Proposed New Source Performance Standards; Steam Generators,
     Incinerators, Portland Cement Plants, Nitric Acid Plants, Sulfuric
     Acid Plants.  Publication No. APTD-0711.  August 1971.  pp.  28-29.
                                     5-12

-------
14.   U.  S.  Environmental Protection Agency.  Industrial Process Profiles
     for Environmental Use:  Chapter 21--The Cement Industry.  Publi-
     cation No.  EPA-600/2-77-023u.  February 1977.  pp. 18, 20, 22, 24.
                                     5-13

-------
                          6.  ENFORCEMENT ASPECTS

     This chapter presents the concerns of Federal, State, and  local  air
pollution control agencies based on their experience  in enforcing the
Portland cement NSPS.  These enforcement concerns may be grouped into
four areas:  (1) interpretation of the particulate mass emission limits
for various duct configurations of affected facilities; (b) the emissions
generated during bypass of an electrostatic precipitator during periods
of carbon monoxide  (CO) trips, startups, and shutdowns; (c) monitoring
requirements; and (d) recordkeeping and reporting requirements.

6.1  VARIED EXHAUST GAS DUCTING CONFIGURATIONS

     At some plants with facilities subject to the NSPS since 1979,
exhaust gas streams are ducted from one facility through another facility
prior to a control device.  Exhaust gases can also be split among several
control devices.  Enforcement personnel expressed concern that either of
these configurations can result in redistribution of particulate matter
emissions.

     In 13 cases, plants are ducting the kiln or clinker cooler exhaust
gases through additional facilities prior to the emission control equip-
ment and stack.   Hot exhaust gases from the kiln, the preheater, or the
clinker cooler may be ducted through the raw mill.   These gases allow
the raw mill to dry the raw materials in addition to crushing and
classifying them.  This preheating is a recent development to improve
productivity and fuel efficiency.1

     Seven plants currently duct kiln exhaust gases through the raw
mill, and one of the plants under construction also plans to duct kiln
exhaust gases through the raw mill.   In all  seven cases, particulate
mass emissions are below the kiln NSPS limit of 0.15 kg/Mg (0.30 Ib/ton).
One plant with combined kiln and clinker cooler emissions is in compliance
with the kiln NSPS and also with the more stringent clinker cooler NSPS
of 0.05 kg/Mg (0.10 Ib/ton).   Three plants combine kiln, clinker cooler,
and raw mill or raw mill  dryer emissions;  all  three of these plants are
in compliance with the kiln NSPS.   At one of two plants with combined
clinker cooler and raw mill  exhaust gases,  emissions are below the
clinker cooler NSPS limit.   The second plant exceeds the NSPS limit.
This is a uniquely designed plant,  and,  at the time of testing, process
conditions were not representative  of normal  operating conditions.2  The
clinker cooler will  be retested under normal  operating conditions.
                                  6-1

-------
     In other cases, plants are splitting exhaust gases from the kiln or
the clinker cooler.  For example, exhaust gases from a dry process kiln
are often split to allow part of the gases to travel through a preheater
and/or a raw mill and part to bypass these facilities.  Such a bypass
system reduces buildup of alkalies and sulfates from the exhaust gases
onto the raw feed.  Emissions from the bypass may be recombined with the
preheater or raw mill gases prior to a control device or may be controlled
by a separate control device.  In the latter case, the particulate mass
emissions from the main stack and bypass stack should be added to obtain
total kiln emissions.

6.2  BYPASS OF ELECTROSTATIC PRECIPITATORS

     Air pollution control agency personnel from several States expressed
concern that excess particulate emissions from kilns controlled by
electrostatic precipitators were occurring during bypass of the control
device because of a CO trip (see 4.1.1.2).  These CO trips of electro-
static precipitators typically are treated as malfunctions of the control
device.

     After a kiln system has achieved smooth, normal operation, some CO
trips may still occur, although infrequently.  These CO trips may be
caused by one or more of the following reasons:   malfunction of equipment
ahead of the ESP, poor maintenance, or operator error.  Malfunctioning
equipment could include the kiln, fans,  preheater (plugups), coal mill,
and gas analysis equipment.   Poor maintenance will cause all system
conditions to deteriorate with time.   There may also be improper actions
or reactions by the operator that contribute to CO trips.

     Electrostatic precipitators may also be bypassed during startup and
shutdown of the kiln due to increased combustibles in the flue gases.
Emissions during startup, shutdown, and malfunctions are not considered
representative for the purpose of demonstrating compliance with the
NSPS, and emissions in excess of the applicable emission limit during
startup, shutdown, and malfunctions are not considered a violation
unless otherwise specified in the applicable standard.  However,
Section 60.11(d) of the General  Provisions requires that "[a]t all
times, including periods of startup,  shutdown, and malfunction, owners
and operators shall, to the extent practicable,  maintain and operate any
affected facility including associated air pollution control equipment
in a manner consistent with good air pollution engineering control
practice for minimizing emissions."

6.2.1  CO Trips

     Information gathered during this review from plant personnel and
equipment vendors concerning operation and maintenance of electrostatic
precipitators controlling cement kilns is presented below.

     As shown in Table 4-3,  the frequency of CO trips of electrostatic
precipitators can range from a few times a year to over 600 times a
year.   Assuming 7,200 hours per year of operation and 23 kg/Mg
                                  6-2

-------
 (45  Ib/ton) of uncontrolled emissions (i.e., cyclone control only),
 annual particulate emissions during CO trips can vary from 0.21 Mg/yr
 (0.23 ton/yr) for one CO trip of 4-minutes1 duration to 390 Mg/yr
 (430 tons/yr) for 600 trips of average 11-minutes1 duration.

     Electrostatic precipitator vendors and plant operators indicate
 that, if process, control, and monitoring equipment are properly designed,
 operated, and maintained, CO trips of the precipitator should be infre-
 quent.3-7  Short CO increases, or CO spikes, can be disregarded or
 eliminated, and proper attention to complete fuel combustion can minimize
 the  number of extended CO increases that necessitate de-energization of
 the  electrostatic precipitator to ensure safety of the control equipment.
 Several equipment vendors noted that one or two CO trips per month is an
 average frequency for a properly operated kiln.5,7  Each CO trip would
 average about 3 minutes.7  Another source stated that a electrostatic
 precipitator could be operated to reduce CO trips to three or four
 occurrences per year.8

     As described in the following sections, the use of continuous com-
 bustibles monitors with time delay trips and careful attention to the
 coal metering system and general process operation should minimize
 unpreventable CO trips.

     6.2.1.1  Continuous Monitor

     6.2.1.1.1  Type, number,  and location of monitor.   Continuous
 monitors can be installed to measure oxygen, carbon monoxide,  or com-
 bustibles in the gas stream entering the electrostatic precipitator.9
 Combustibles monitors are especially advantageous because they monitor
 CO as well  as methane gas.8,10  At new plants in the cement industry,
 continuous monitors are typically installed at three locations:   one at
 the kiln exit, one at the outlet to the third-stage preheater cyclone,
 and one at the outlet to the first-stage preheater cyclone.   Gases
 extracted from these locations are cooled and cleaned prior to analysis.9
This conditioning causes a delay of 30 seconds to 2 minutes from the
 time of sample extraction until  the data are available to the kiln
 operator.11  Because the most  common increase in kiln combustibles is a
CO spike, which lasts less than 30 seconds, the monitor delay usually
allows the potentially explosive combustibles to exit the stack before
the recorder registers the event.12  De-energization of the electrostatic
precipitator in these cases is too late to be effective and,  thus,
serves no purpose.4

     In situ continuous  monitors require no time to condition  the gases
and, therefore,  instantaneously register CO or combustibles levels.8,11
Such monitors cannot be  used where high temperatures or dust loadings
are present,  and would be most useful,  therefore,  for monitoring exhaust
gases at the outlet of the electrostatic precipitator.10,13  Monitors
that require conditioning of the gases  would probably be used  in
conjunction with in situ monitors.
                                  6-3

-------
     Monitors  located at the first-stage preheater outlet, the electro-
 static precipitator inlet, or the precipitator outlet may receive a
 lower level of combustibles than those at earlier process locations.
 Conditions prior to the ESP are most likely to cause CO trips.13  Monitors
 at the earlier locations can reduce the chance that the electrostatic
 precipitator will be de-energized because of a CO spike.  The monitor at
 the  kiln exit  is most commonly used only for information purposes because
 of maintenance problems with the extractive probe.13  One vendor indicated
 that, in some cases, plants de-energize the electrostatic precipitator
 only when two or three monitors indicate that de-energization is
 appropriate.10

     6.2.1.1.2  Trip and alarm levels.   As reported in Chapter 4
 (Table 4-3), the uu trip level for electrostatic precipitators varies
 from 0.2 to 5 percent combustibles or CO.   Alarms, which warn the kiln
 operator that the level of combustibles is approaching the trip level,
 are often set at levels that range from 0.2 to 2 percent combustibles or
 CO.  Three electrostatic precipitator vendors believe that an alarm is
 appropriate and should be set to go off from 0.2 to 0.5 percent
 combustibles or CO.10,14,15  They state that the CO trip level should be
 set from 0.7 to 2 percent combustibles  or CO.10,14,15  Further, time
 delays can be incorporated into the monitor so that instantaneous CO
 spikes are disregarded and only an extended CO increase can de-energize
 the electrostatic precipitator.15  For  all  types of monitors, including
 in situ monitors, such time delay trips could be designed.   For example,
 a monitor that records the combustible  level  once per minute could be
 programmed to de-energize the electrostatic precipitator only after
 receiving 2 to 5 consecutive readings above the trip level  (thereby
 allowing a 1 to 4 minute trip delay).   The electrostatic precipitator
 would, thus, only be de-energized for these extended increases in com-
 bustibles, i.e.,  those that risk the safe  operation of the precipitator.
 One electrostatic precipitator vendor stated that even a 4-minute delay
 would not endanger the electrostatic precipitator.16

     Some plants incorporate time delays of about 2 minutes before
 re-energizing the electrostatic precipitator after corrective action has
 been taken to ensure that the problem has  been solved.   Other plants
 re-energize the electrostatic precipitator immediately (within about
 5 seconds).17

     6.2.1.2  Coal  Metering System.   The method by which coal is  fired
 both in the kiln and in the precalciner can affect CO trips.   Coal  can
 be fired by the direct or the indirect  method.   In a direct-fired system,
 coal  is conveyed directly from the coal  mill  to the kiln;  this is the
 system generally used in the cement industry.18  In an indirect-fired
 system,  coal is stored in bins between  the  mill and the kiln.  Because
 the coal  is not stored in a direct-fired system,  no combustible gases
 can accumulate in the system.   Coal  irregularities (particle size,
moisture) or disturbances in coal  feeding  and conveying (surges), how-
 ever, are translated immediately into the  kiln, potentially resulting in
 incomplete combustion.18  Although indirect coal  firing generally allows
more consistent fuel  conveying,  precipitator vendors state  that proper
                                  6-4

-------
 coal  metering  can  be  accomplished with both  direct  and  indirect  coal-fired
 systems.10,15

      6.2.1.3   General  Process Operation.  The  ultimate  causes  of CO
 trips are process  upsets  upstream of the electrostatic  precipitator.
 For this reason, experienced personnel that  are well trained in  standard
 procedures  for operation  and maintenance are essential  to provide  kiln
 operation that eliminates  unnecessary CO increases  and  to ensure prudent
 use of monitor information  in deciding which CO increases necessitate
 de-energization.   Plant personnel can make process  modifications as
 necessary to minimize  these trips and should maintain the required
 equipment (e.g., parts for  the combustibles  analyzer).10

      Increases in  combustibles that necessitate de-energization  of the
 electrostatic  precipitator  are the result of hours  of improper kiln
 operation,  not of  an  instantaneous process event.    One  vendor states
 that  every  electrostatic precipitator fire he  has examined has been the
 result of 24 to 48 hours of kiln maloperation  with  no operational
 combustibles analyzer  present.10  No electrostatic  precipitator  fires
 observed by the vendor have been caused by an  instantaneous CO spike.10

      Another vendor noted that environmental regulations in Europe
 require that the raw feed to the kiln and the  raw material drying and
 grinding systems be interlocked with the high  voltage supply of  the
 electrostatic precipitator.19  This ensures shutdown of air flow and
 feed  to the raw mill and kiln during a CO trip.19   Although the  kiln may
 continue to rotate, emissions would be significantly less than during
 full  operation.  Such an interlock system could provide strong incentive
 for plant personnel to minimize the number of  CO trips.19

 6.2.2  Kiln Startup and Shutdown

      Electrostatic precipitator vendors and plant operators state that,
 because of improved process control  and advancements in microprocessor
 control, it is now normal  practice for new electrostatic precipitators
 to start up and shut down concurrent with the  kiln  induced draft fan.20-23
 Startup of a large "cold"  precalciner kiln can take from 20 to 36 hours
 depending on the type of refractory brick installed in the kiln.19  In
 the first few hours, a low flame is used to cure and dry the refractory,
 no induced draft fan is used (natural  draft is sufficient),  no raw
material  is fed to the kiln, and the kiln is occasionally turned for
 about one-third of its circumference to ensure consistent warming.21
When the refractory is heated,  the temperature can be raised;  therefore,
more fuel  is added, and the induced draft fan  is turned on to  provide
more oxygen.21   At this point,  the electrostatic precipitator  is often
energized because the draft will  stir dust in the  kiln,  causing emissions.
Feed to the kiln is begun  at some later point when the necessary
temperature has been reached.21

     Most modern kilns start up  on oil  or gas because a stable  flame is
easier to maintain with these  fuels  and because the coal is  often dried
                                  6-5

-------
by kiln or clinker cooler exhaust gases.21  Therefore, the coal mill is
turned on only after the kiln produces sufficient heat to dry the coal.21

     Plant personnel also note that if kiln oxygen levels are kept above
4 percent until the kiln is near full production and at normal temperature
levels, complete combustion is better assured.21  High oxygen levels are
not economical for general  kiln operation, however.21

6.3  CONTINUOUS OPACITY MONITORS

     The current NSPS do not require continuous monitoring of visible
emissions.  EPA Reference Method 9 is used to assess compliance with the
visible emissions limit.

     Continuous opacity monitors can automatically alert facility personnel
to a control device problem, thereby facilitating prompt repair of the
device.  Continuous opacity monitors are effective in all weather condi-
tions and at night.   At least 13 of the 28 plants that have become
subject to the NSPS since the 1979 review have installed continuous
opacity monitors because of State requirements.

     Continuous opacity monitors work well on all types of control
devices where flue gases are exhausted to the atmosphere through a
single stack.  A single continuous opacity monitor may not, however,
measure accurately the opacity of visible emissions from the multiple
stacks or monovents associated with some positive-pressure fabric filters
or the multiple stacks associated with some negative-pressure fabric
filters.24  One kiln and one clinker cooler that have become subject to
the standard since 1979 are controlled by positive-pressure fabric
filters.   In both cases, however, the fabric filters are vented to a
single stack.  One company in this industry is known to use negative-
pressure fabric filters with multiple stacks at two of its plants that
have become subject to the NSPS since 1979.

6.4  RECORDKEEPING AND REPORTING REQUIREMENTS

     Performance test results for affected facilities must be reported
to the EPA within 60 days after achieving maximum production but no
later than 180 days after startup of the facility.   The Portland Cement
Association, plant personnel, and personnel at one State pollution
control agency indicated that this time allowance is too short.25-28
Personnel at one plant noted that they could not correct all the
mechanical, electrical, and physical problems involved with startup of
their entire new plant within 180 days.26  Personnel at another plant
stated that optimal  operating settings for a fabric filter cannot be
developed until all  process systems are stabilized and that testing
before this time gives results that are not indicative of the normal
operation of the control equipment.27
                                  6-6

-------
 6.5  REFERENCES FOR CHAPTER 6

'  1.  Portland Cement Association.  The U. S. Cement  Industry:   An  Economic
      Report.  Skokie, Illinois.  January 1984.  pp.  6,  8.

  2.  Telecon.  Clark, C., MRI, to Clouse, J. , Colorado  Air  Pollution
      Control Division.   October 13, 1983.  Discussion of  cement plants
      in Colorado.

  3.  Comments presented by Orem, S., Industrial Gas  Cleaning  Institute,
      to the National Air Pollution Control Techniques Advisory  Committee.
      August 30, 1984.  Response to the review of NSPS for portland
      cement plants,  p.  1.

  4.  Comments and attachments presented by Brown, R., Environmental
      Elements Corp., to the National Air Pollution Control  Techniques
      Advisory Committee.  August 30, 1984.   Response to the NSPS for
      Portland cement plants,   p.  4, 5.

  5.  Comments presented by von Seebach, M., Polysius Corp., to  the
      National Air Pollution Control Techniques Advisory Committee.
      August 30, 1984.  Response to the NSPS for portland  cement plants.
      p.  3.

  6.  Letter from Lotz,  W.,  Lehigh Portland Cement Company,  to Farmer, J.,
      EPA/OAQPS.   September 6,  1984.   Transmitting summary of comments
      presented to  the National Air Pollution Control Techniques Advisory
      Committee on  August 30,  1984.   p.  3.

  7.  Letter from Riley,  J.,  Lurgi Corp.,  to Farmer, J.,  EPA/OAQPS.
      August 31,  1984.   Transmitting summary of comments  to  have been
      presented to  the National Air Pollution Control Techniques Advisory
      Committee on  August 30,  1984.   p.  1.

  8.  Telecon.   Clark, C., MRI, with Hawks,  R.,  PEDCo Environmental, Inc.
      August 22,  1984.   Discussion of CO trips,  startup,  and shutdown of
      an  electrostatic precipitator.

  9.  PEDCo  Environmental, Inc.   Technical  Assistance to  the State of
      Iowa--Excess  Emissions  at Lehigh  Cement,  Mason City,  Iowa.   Prepared
      for U.  S.  Environmental  Protection Agency.   February 1984.   p.  27.

 10.  Telecon.   Clark,  C. , MRI,  with Brown,  R.,  Environmental Elements
      Corp.   September 14, 1984.   Discussion of  techniques  to minimize CO
      trips  of electrostatic  precipitators.

 11.   Reference 9,  p.  31,  61.

 12.   Reference 9,  p.  95.

 13.   Reference 9,  p.  31.
                                   6-7

-------
14.  Telecon.  Clark, C., MRI, with von Seebach, M.,  Polysius  Corp.
     September 14, 1984.  Discussion of techniques  to minimize CO  trips
     of electrostatic precipitators.

15.  Telecon.  Clark, C., MRI, with Prior, 0., F. L  Smith  and Company.
     October 2, 1984.  Discussion of CO trip  levels.

16.  Telecon.  Clark, C., MRI, with Brown, R., Environmental Elements
     Corp.  October 2, 1984.  Discussion of time delays  for CO trips.

17.  Reference 9, p.  63.

18.  Reference 9, p.  20-23, 61.

19.  Reference 5, p.  3, 4.

20.  Reference 4, p.  5, 7-9.

21.  Reference 6, p.  2.

22.  Letter from Greer, W., Lone Star Industries, Inc.,  to  Cuffe,  S. ,
     EPA/ISB.  June 22, 1984.   Response to request  for comments on the
     draft review document,   p. 5.

23.  Letter from Schneeberger, C.,  Portland Cement  Association, to
     Cuffe, S., EPA/ISB.  August 29, 1984.  Comments on  draft  review
     document and Advanced Notice of Proposed Rulemaking for the review
     of standards for portland cement plants,  p. 3.

24.  U. S. Environmental Protection Agency.   Electric Arc Furnaces and
     Argon-Oxygen Decarburization Vessels  in  Steel  Industry—Background
     Information for Proposed Revisions to Standards.  Draft EIS.
     Publication No.  EPA-450/3-82-020a.   July 1983.   pp. D-18  to D-20.

25.  Reference 23, p. 10.

26.  Letter and attachments from Gebhardt, R., Lehigh Portland Cement
     Company, to Farmer, J. , EPA/OAQPS.   January 6, 1984.   Response to
     Section 114 information request,   p.  13.

27.  Letters and attachments from Powledge, H., Ideal Basic  Industries,
     Inc., to Farmer, J., EPA/OAQPS.  December 30,  1983.  Response to
     Section 114 information request,   p.  5.

28.  Telecon.  Clark, C., MRI, with Gore,  R., Alabama Department of
     Environmental Management.  September 29, 1983.   Discussion of
     cement plants in the State of Alabama.
                                  6-8

-------
                     APPENDIX A



SUMMARY OF PORTLAND CEMENT FACILITIES SUBJECT TO NSPS
                        A-l

-------
TABLE A-l.  SUMMARY OF  PORTLAND CEMENT FACILITIES SUBJECT TO NSPS
PLANT DATA
Name/ local ion
tPA Region 11
Moore McCormack
Cement, Inc.
Glens Falls Portland Cement
313 Warren St.
Glens Falls, N.Y. 12801
San Juan Cement
GPO Box 2888
San Juan, Puerlo Rico

LPA Region III
Coplay Cemenl Manufacturing Co.6
Nazareth, Pa. 18064



General Portland, Inc.
Whitehall Cement
Whitehall, Pa. 18052


Lone Star Industries, Inc.6
P.O. Box 27
Cloverddle, Va. 24077
(Roanoke, Va. )

LPA Region IV
General Portland, Inc.6
(formerly Citadel Cement Corp.)
Arcola Rd.
KILN DATA
lota) cement
capacity
Kiln year
fuel/%S wet or dry

450
Coal




Oil (coal by
end of 1983)


1,095
Coal



800
Coal
(No. 1, No. 2 kiln)
Oil
(No. 3 kiln)
1,200
Coal
1.28



750
Coal


1973-D




1967-W
1967-W
1975-W


1978-D




1956-D
1965-0
1975-D ,


1951-D
1951-D
1953-D
1956-D
1976-D

1977-D


Clinker^ Preheater/
capacity precalciner

543









972




174
333
257


140
140
140
140
540

708



ft Neither
M Preheater
u Precalciner


M Neither
o Preheater
n Precalciner


u Neither
M Preheater
n Precalciner


u Neither
M Preheater
n Precalciner
(All kilns)

u Neither
M Preheater
u Precalciner
(1976 kiln only)


n Neither
M Preheater
n Precalciner
DATA ON FACILITIES SUBJECT TO NSPS
Facilities
subject
to NSPS

Kiln
Cooler



Kiln
Cooler
Mills
Storage

Kiln
Cooler
Mills
Storage
Transfer
Kiln




Kiln
Cooler




Kiln
Cooler

Reason-
N, M, RV
date

N-1973
N-1973



N-1975
N-1975
N-1975
M-Conveyors

N-1978
N- 1978
N-1978
N-1978
N-1978
N-1975




N-1976
N-1976




N-1977
N-1977

Control .
equipment

ESP
FF



FF
FF
FF
FF

FF
FF
FF
FF
FF
FF(O




ESP





ESP
FF

                                                                               (continued)

-------
TABLE A-l.   (continued)
PLANT DATA

Name/location
EPA Region IV (continued)
Ideal Basic Industries, Inc.a
Theodore Industrial Park
Theodore, Ala. 36582





Lehigh Portland Cement Co.6'9
(formerly Universal Atlas Cement)
800 Second Ave.
Leeds, Ala. 35094


j. National Cement Co.6
i Highway 144
00 Ragland, Ala. 35131
Moore McCorroack Cement, Inc.e
Florida Mining and Materials
605 W. Broad St.
Brooksville, Fla. 33512



Lone Star Industries, Inc.6
(formerly Maule Industries, Inc )
Hialeah, Fla. 33012
Medusa Cement Company6
Clinchfield, Ga. 31013



Total cement
capacity3
fuel/%S

2,365
Coal
1.5





600
Coal


800
Coal
1,200
Coal
1.04



1,200
Coal , gas
790
Coal
Low sulfur

KILN

Ki In year
wet or dry

1981-D





1976-D


1976-D
1975-0
1982-D ,



1970-W
1970-W
1975-W
1961-W
1974-D


DATA

Clinker-
capacity

1,415





558


804
560
560



231
231
752
193
546


DAIA ON FACII ITIFS SIIR.IFf

Preheater/
precalciner

Q Neither
M Preheater
W Precalciner





n Neither
M Preheater
o Precalciner


u Neither
M Preheater
M Precalciner
n Neither
on Preheater
Q Precalciner
(both kilns)



M Neither
ci Preheater
a Precalciner
Q Neither
os Preheater
n Precalciner
(1974 kiln)
Faci 1 ities
subject
to NSPS

Kiln
Cooler
2 raw mi 1 1
dryers
Raw mill
Finish mill
Storage
Transfer
Kiln
Cooler
Mills
Storage
Transfer
Kiln
Cooler
Storage
Kiln
Kiln
Cooler
Cooler
Mills
Storage
Transfer
Kiln
Cooler
Kiln
Cooler


Reason:
N, M, R /
date

N-1981)
N-1981 >
N-1981)

N-1981
N-1981
N-1981
N-1981
N-1976
N-1976
N-1976
N-1976
N-1976
N-1976
N-1976
N-1975
N-1982
N-1975
N-1982
N-1975
N-1975
N-1975
N-1975
N-1975
N-1974
N-1974


'T TO NSPS

Control
equipment

FF(-)

FF(-)
FF(-)
FF(-)
FF(-)
ESP
Gravel bed


ESP
Gravel bed
FF(-)
FF(-)
FF(-)
FF(-)
FF(-)
FF(-)
FF(-)
ESP
FF
FF(79)


                                                          (continued)

-------
TABLE A-l.   (continued)
PLANT DATA
Name/location
EPA Region IV (continued)
Moore McCormack Cement, Inc.c
(formerly Flintkote Co.)
Kosmos Cement Co.
Kosmobdale, Ky. 40272


Texas Industries, Inc.6
United Cement
Artesia, Miss 39/36



Giant Portland Cement Co.e
P.O. Box 218
Hdrleyville, S.C. 29448

Giffot'd-Hill & Company, Inc.e
P.O. Box 326
Harleyville, S.C. 29448



Dundee Cement Co.e
Santee Portland Cement
Hwy. 453 South
Holly Hill, S.C. 29059
Moore McCormack Cement, Inc.
Dixie Cement Co. (formerly Ideal
Basic Industries, Inc.)^
Knoxvil le, Tenn. 37914
KILN DATA
Total cement
capacity
fuelTts

660
Coal
0.6-0.8


525
Coal
1.0



855
Coal

650
Coal



1,700
Coal
750
Coal
1.5
Kiln year
wet or dry

19/4-0


1974-W



1952-W
1957-W
1960-W
1974-W
1974-D



1966-W
1972-W
19/9-D
Clinker^ Preheater/
capacity precalciner

651 11 Neither
M Preheater
o Precalciner


456 Hi Neither
o Preheater
o Precalciner



200 M Neither
185 ci Preheater
185 o Precalciner
200
551 a Neither
M Preheater
a Precalciner



363 M Neither
693 n Preheater
o Precalciner
512 u Neither
M Preheater
B Precalciner
DATA ON FACILITIES SlIR.lFrT TO NSPS
Facilities
subject
to NSPS

Kiln
Cooler
Finish mill
Raw blending
silo
Transfer
Kiln
Cooler
Mills
Storage
Transfer
Kiln
Cooler

Kiln
Dryer
Cooler
Mills
Storage
Transfer
Kiln
Cooler
Kiln
Coo lei-
Raw m i 1 1
Finish mi 1 1
Reason-
N, M, RC/
date

N-1974
N-1974
N-1974
N-1974

N-1974
N-1974
N-1974
N-1974
N-1974
N-1974
N-1974
N-1974

N-1974)
N-1974)
N-1974
N-1974
N-1974
N-1974
N-1972
N-1972
N-1979
N-1979
N-1974
N-1975
Control .
equipment

ESP
FF
FF
FF

FF
ESP
Wet scrub-
ber
FF( + )
FF( + )
FF( + )
FF
FF

FF
Ff



ESP
Fl
FF(-)
FF(-)
FF
FF
                                                           (continued)

-------
TABLE A-l.   (continued)
PLANT DATA
Name/loration
EPA Region V
Centex Corp.
Illinois Cement Co.
P.O. Box 442
LaSalle, 111. 61301

Missouri Portland Cement Co.e
Joppa, 111. 62953



Lehigh Portland Cement Co.e
Mitchell, Ind. 47446


Louisville Cement Co.e
Speed, Ind. 47172



National Gypsum Co.6
Huron Cement
Alpena, Mich. 49707


Columbia Cement Co. (Ashland Oil)
P.O. Box 1531
Zanesville, Ohio 43701
Southwestern Portland Cement Co.e
Fairborn, Ohio 45324

KILN DATA
Total cement
capacity
fuel/%S

380
Coal
2.0


1,314
Coal
2-2.5


725
Coal


1,094
Coal



2,450
Coal
3


700
Coal

730
Coal

Kiln year
wet or dry

1974-D




1963-D
1975-D



1960-D
1960-D
1976-D

1973-D
1977-D



1962-D
1965-D
1965-0
1975-D
1975-D
1955-W
1963-W

1955-W
1974-0

Clinker.
capacity

428




544
672



248
248
264

331
602



318
318
318
508
508
241
360

124
569

Preheater/
precalciner

n Neither
M Preheater
D Precalciner


a Neither
a Preheater
n Precalciner
(1975 kiln)

a Neither
a Preheater
D Precalciner
(1976 kiln)
a Neither
a Preheater
D Precalciner
(1977 kiln)

a Neither
n Preheater
o Precalciner


a Neither
n Preheater
D Precalciner
n Neither
a Preheater
D Precalciner
DATA ON FACILITIES SUBJECT TO NSPS
Facilities
subject
to NSPS

Kiln
Cooler
Mills
Storage
Transfer
Kiln
Cooler
Mills
Storage
Transfer
Kiln
Cooler


Kiln
Cooler
Kiln
Cooler
Mortar kiln
Kiln
Kiln
Cooler
Cooler

Finish mill


Kiln
Cooler

Reason:
N, M, RV
date

N-1974
N-1974
N-1974
N-1974
N-1974
N-1975
N-1975
N-1975
N-1975
N-1975
N-1976
N-1976


N-1973
N-1973
N-1977
N-1977
N->1971
N-1975
N-1975
N
N

N-1978


N-1974
N-1974

Control .
equipment

FF(-).
FF(-)h
FF(-)
FF(-)
FF(-)
ESP .
FF(-)1
FF(-)
FF(-)
FF(-)
ESP
FF(-)


ESP
FF(-)
ESP
FF(-)
ESP
FF

FF
FF

FF


ESP
FF(+)

                                                            (continued)

-------
                                                 TABLE  A-l.   (continued)
cr>
PLANT DATA
Name/location
EPA Region VI
Lone Star Industries, Inc.
(formerly OKC Cement)
Louisiana Cement Division
New Orleans, La. 70129
Ideal Basic Industries, Inc.
Tijeras, N.M. 87059


Lone Star Industries, Inc.®
Oklahoma Cement
9250 Amberton Pkwy.
Pryor, Okla. 74361
Alamo Cement Co."
5675 FM 1604 NE
San Antonio, Tex. 78233


Capitol Aggregates, Inc.
Capitol Cement
Nacigdiches at Bulve
San Antonio, Tex. 78233



Centex Corp.e
Texas Cement Co.
Buda, Tex. 78610





KILN DATA
Total cement
capacity
fuel/%S

750
Coal
<0.7

660
Coal


725
Coal
3-4

600
Coal /coke
1.5-coal
3.9-coke

800
Coal /coke
3.35




si ,300
Coal






Kiln year
wet or dry

1964-W
1974-W


1959-D
1960-D


1961-D
1962-D
1979-D

1981-0




1965-W
1983-D





1978-D
1983-D






Clinker.
capacity

347
347


237
237


205
205
267

523




338
500





468
=468






Preheater/
precalciner

a Neither
n Preheater
D Precalciner

a Neither
S> Preheater
D Precalciner
(Both kilns)
a Neither
n Preheater
a Precalciner

a Neither
a Preheater
a Precalciner


B Neither (wet)
a Preheater
a Precalciner
(PH, PC- 1983
kiln)


a Neither
B Preheater
a Precalciner
PH-1978 kiln
PH.PC-1983 kiln



DATA ON FACILITIES SUBJECT TO NSPS
Facilities
subject
to NSPS

Kiln
Cooler
Transfer

Finish mill



Kiln
Cooler


Kiln
Cooler
Raw mi 1 1
Storage
Transfer
Kiln
Cooler
Raw mill
Coal/coke
transfer
Storage
Transfer
Kiln
Raw mill
Cooler
Mill
Storage
Transfer
Kiln
Raw mill
Reasoni
N, M, RV
date

N->1974
N->1974
N->1974

N->1979



N-1979
N-1979


N-1981
N-1981
N-3981
N-1981
N-1981
N-1983
N-1983
N-1983
N-1983

N-1983
N-1983
N-1978
N-1978
N-1978
N-1978
N-1978
N-1978
N-1983
N-1983
Control .
equipment

ESP



FF



FF
Gravel bed


ESP


FF
FF
FF
FF
FF
FF

FF
FF
FF

FF
FF
FF
FF
FF
— _ 	 _,, 	 _-

-------
TABLE  A-l.  (continued)
PLANT DATA
Name/location
LPA Region VI (continued)
General Portland, Inc.
Wald Rd. & Solms Rd.
New Braunfels, Tex. 78130
(Balcones, Tex. )


Gulf Coast Portland Cement Co.
6203 Industrial Way
Houston, Tex. 77011
e f
Kaiser Cement Corp. '
P. 0. Box 34210
San Antonio, Tex. 78265

lone Star Industries, Inc.
Jet. FM 608 FM 1170
Maryneal, Tex. 79556

Southwestern Portland Cement Co.
Bushland, Tex. (Amarillo, Tex.)

Southwestern Portland Cement Co.
Odessa, Tex. 79760


lexas Industries, Inc.a
8100 Carpenter Frwy
Hunter, Tex. 78130


Texas Industries, Inc.
Midlothian, Tex. 76065

-
KILN DATA
Total cement
capacity
fuel/%S
925
Coal




940
Coal

490
Coal
1.0

545
Coal


242
Coal

550
Coal
0.5

840
Coal
1.2


1,400
Coal


Kiln year
wet or dry
1980-D





1961-W


1975-D



1951-D
1951-D
1953-D

1963-W

1958-D
1978-D


1980-0




1960-W
1963-W
1967-W
1972-W
Clinker.
capacity
875





333


775



151
151
243

230

248
279


664




304
304
304
304
Preheater/
precalciner
u Neither
M Preheater
S) Precalciner



M Neither
n Preheater
o Precalciner
D Neither
M Preheater
M Precalciner
(2nd PC added
LI Neither
M Preheater
Q Precalciner
(All kilns)
M Neither
n Preheater
n Precalciner
H Neither
M Preheater
n Precalciner
(1978 kiln)
n Neither
M Preheater
a Precalciner


W Neither
u Preheater
N Precalciner

DATA ON FACILITIES SUBJEC1
Faci lities
subject
to NSPS
Kiln
Raw mill
Cooler
Finish mil 1
Storage
Transfer
Finish mill
Finish mill
Storage
Kiln

Cooler
1979) Finish mill
Coal transfer



Coal storage
Coal transfer

Kiln
Coal storage
Coal transfer

Kiln
Cooler
Mills
Storage
Transfer
Finish mi 1 1
Transfer


Reason:
N, M, RV
date
N-1980)
N-1980/
N-1980
N-1980
N-1980
N-1980
N-1973
N-1978
N-1978
N-1975

N-1975, 1979
N-1979
N-1979



N-1981
N-1981

N-1978
N-1982
N-1982

N-1980
N-1980
N-1980
N-1980
N-1980
N-1979
N-1979


TO NSPS
Control £|
equipment
2 LSP's

Gravel bed
FF
FF
FF
FF
FF
FF
3 ESP's
FF (AB)
FF
FF
FF



FF
FF

FF
FF
K

ESP
FF
FF
FF
FF
ESP
FF


                                                           TcontiiuiedJ

-------
                                               TABLE A-l.   (continued)
oo
PLANT DATA
Name/location
EPA Region VII
Davenport Industries
Davenport Cement
Buffalo, Iowa 52808
(Scott City, Iowa)


Lehigh Portland Cement Co.
Mason City, Iowa 50401


Northwestern States Portland Cement Co.e
N. Federal 17th St.
Mason City, Iowa 50401
Monarch Cement Co.
Humboldt, Kans. 66748



Lone Star Industries, Inc.^
Marquette Cement Co.
Box 520
Cape Girardeau, Mo. 63701


River Cement Co.
Festus, Mo. 63028

Ash Grove Cement Co.e
Louisville, Nebr. 68037



KILN DATA
Total cemegt
capacity
fuel/%S

850
Coal




750
Coal


1,150
Coal

600
Gas



1,200
Coal
3.0



1,060
Coal

790
Coal
0.9


Kiln year
wet or dry

1981-D





1958-D
1979-D


1960-D
1966-D
1976-0
1972-D
1974-D
1975-0


1981-D





1965-D
1969-D

1975-D
1982-D



Clinker.
capacity

809





233
543


163
450
252
116
248
248


992





558
558

400
558



Preheater/
precalciner

n Neither
a Preheater
a Precalciner



D Neither
a Preheater
8 Precalciner
(1979 kiln only)
a Neither
a Preheater
D Precalciner
n Neither
N Preheater
n Precalciner
(1974, 1975
kilns)
a Neither
H Preheater
a Precalciner



a Neither
a Preheater
D Precalciner
a Neither
a Preheater
a Precalciner
(PH-1975 kiln)
(PC-1982 kiln)
DATA ON FACILITIES SUBJECT TO NSPS
Facilities
subject
to NSPS

Kiln
Raw mill
Cooler
Finish mill
Storage
Transfer
Kiln
Mill
Separators

Kiln
Cooler

Kiln
Cooler



Kiln
Cooler
Raw mill
Finish mill
Storage
Transfer
Raw mill


Kiln
Cooler
Kiln
Cooler

Reason-
N, M, RV
date

N-1981*
N-1981/
N-1981
N-1981
N-1981
N-1981
N-1979)
N-1980/
N-1980

N-1976
N-1976

N-1975
N-1975



N-1981
N- 19811
N-1981 /
N-1981
N-1981
N-1981
N>1979


N-1975
N-1975
N-1982
N-1982

Control d
equipment

FF

FF
FF
FF
FF
ESP



FF
FF
(1979)
FF
FF



ESP
2 FF's

FF
FF
FF
FF


ESP
FF(+)
ESP
FF(-)


-------
TABLE  A-l.   (continued)
PLANT DAI A
Name/ local ion
EPA Region VIII
Ideal Basic Industries, Inc.
Boettcher Plant
P.O. Box 2227
Ft. Collins, Colo. 80522
(La Porte, Colo.)

Ideal Basic Industries, Inc.6
Portland, Colo. 81226


Martin Marielta Corp. ^
P.O. Box 529
Lyons, Colo. 80540
South Dakota Cement Plant6
(State of South Dakota)
> P.O. Box 351
ua Rapid City, S.D. 57709
Lone Star Industries, Inc.6
Utah Portland Cement Co.
615 W. 8th South
Salt Lake City, Utah 84110
Martin Marietta Corp.9
P 0 Box 40
. Leamington, Utah 84648



Monolith Portland Cement Co f
P.O. Box 40
Ld ramie, Wyo. 82070

KILN DATA
Total cement
capacity
fuel AS

768
Coal


1,070
Coal
<1.0

405
Coal
0.52
1,100
Coal
Low sulfur
420
Coal , oil ,
gas
0.4-0.6
650
Coal
0.4-0.6



700
Coal
0.5-0.9

Kiln year
wet or dry

1981-D


1948-W
1948-W
1974-W

1979-0
1950-W
1956-W
1958-W
1978-0
1960-W .
1975-W
19/9-W
1982-D




1961-W
1981-W

Clinker.
capacity

440


184
184
480

405
151
151
151
503
120
150
150
603




200
300

Preheater/
precalciner

U Neither
M Preheater
o Precalciner


» Neither
a Preheater
fi Precalciner

Q Neither
8 Preheater
M Precalciner
n Neither
8 Preheater
Q Precalciner
(1978 kiln)
« Neither
o Preheater
a Precalciner
a Neither
M Preheater
M Precalciner



M Neither
n Preheater
n Precalciner

DATA ON FACIIITJF"; UlR.lFfT rn N<;P<;
Facilities
subject
to NSPS

Kiln
Cooler
Raw mill
Storage
Transfer
Kiln
Cooler
Mill
Storage
Kiln
Limestone
dryer
Kiln
Cooler
Kiln
Cooler
Kiln
Raw ra i 1 1
Cooler
Finish mill
Storage
Transfer
Kiln
Cooler
Finish mi) 1
Cement cooler
Reason-
N, M, Rc/
date

N-1981
N- 19811
N-1981/
N-1981
N-1981
N- 1974 1
N-1974/


N-1980
N-1979
N-1978
N-1978
N-1979
N-1979
N- 19821
N-1982/
N-1982
N-1982
N-1982
N-1982
N-1981
N-1981
N-1981
N-1981
Control .
equipment1

FF(0
FF(-)
FF
FF
LSP
FF
FF
FF(-)
FF(-)
FF(-)
FF(-)
FF(-)
FF(-)
FF
H
FF
FF
FF
tSP
FF(-)
FF(-)
FF(-)
                                                           (continuelT)

-------
TABLE A-l.   (continued)
PLANT DATA

Name/location
EPA Region IX
California Portland Cement Co.a
SOC 24 TUN R1AW
Mojave, Calif. 93501


Genstat', Ltd.
Genstar Cement and Lime Co.
Redding, Calif. 96001

Genstar, Ltd.
Genstar Cement and Lime Co.
San Andreas, Cal if. 95249
Kaiser Cement Corp.9
CuslienLuiry Plant
Star Route Box 400
Lucerne Valley, Calif. 92356



Kaiber Cement Corp.9
Perraanente Plant
Permanente, Calif. 95014


Lone Star Industries, Inc.9
Davenport, Calif. 95017
(Santa Cruz, Calif. )



KILN DATA
lotal cement
capacity11
fuel/%S

1,700
Coal
0.53


600
Coal /wood
chips
2.0
770
Coal
0.6
1,600
Coal



1,600
Coal
<0.5


775
Coal




Ki In year
wet or dry

1955-D
1955-D
1981-D


1981-D

1945-w
1952-W
1956-W
1982-D

,

1981-D


J981-D




Clinker^ Preheater/
capacity precalciner

218 11 Neither
218 W Preheater
1,000 M Precalciner
(1981 kiln only)

571 cj Neither
M Preheater
M Precalciner

192 M Neither
192 a Preheater
192 u Precalciner
1,520 n Neither
M Preheater
a Precalciner



1,520 u Neither
w Preheater
W Precalciner


744 o Neither
M Preheater
M Precalciner



DATA ON FACILITIES SUBJECT TO NSPS
Facilities
subject
to NSPS

Kiln
Raw mill
Cooler
Storage
Transfer-
Kiln
Cooler
Raw mi 1 1

Kiln
Kiln
Kiln
Kiln
Raw mill
Cooler
Alkali bypass
Finish mill
Storage
Transfer
Kiln
Cooler
Mills
Storage
Transfer
Kiln
Raw mi 1 1
Cooler
Finish Mill
Raw feed silo
Transfer
Reason'
N, M, R /
date

N- 19811
N-1981/
N-1981
N-1981
N-1981
N-1981)
N-1981)
N-1981)

M-1975)
M-1975 >
M-1975)
N-1982)
N- 1982 (
N-1982)
N-1982
M-1982
N-1982
N-1981
N-1981
N-1981
N-1981
N-1981
N- 19811
N-1981)
N-1981
N-1981
N-1981
N-1981

Control .
equipment

FF(-)
FF(-)
FK-)
FF(-)
FE(-)

Multi-
cyclone/ESP
IF
FF
FF
FF
FF
FF(-)
FF(-)
FF(-)
FF(-)
FK-)
ESP
Gravel hed
FF(-)
FF(-)
FF(-)
                                                          "Tcontinued]

-------
                                                         TABLE  A-l.    (continued)
PLANT DATA
Name/location
EPA Region IX (continued)
Monolith Portland Cement Co.e
Monolith, Calif. 93548
(Kern, Calif. )



Southwestern Portland Cement
Victorville, Calif. 92392





Lone Star Industries, Inc.
Lone Star Hawaii Cement
Ewa Beach, Hi. 96706
EPA Region X
Alaska Basic Ind.
Anchorage, Alaska 99501

Oregon Portland Cement^
Durkee, Oreg. 97905




KILN DATA
Total cemegt
capacity
ftiel/%5

1,000
Coal
2.0



1,400
Coal





270
Coal
<1. 1

260


630
Coal
0.55



Ki In year
wet or dry

1974-w





1949-W
1953-W
1954-W
1955-W
1956-W
1965-0
1984-D
1972-0



GRINDING


1979-D





Clinkerb
capacity

518





• 78
124
124
124
124
574
800
257



ONLY


500





Preheater/
precalciner

a Neither
n Preheater
o Precalciner



D Neither
8 Preheater
M Precalciner
(1984 kiln only)



n Neither
H Preheater
D Precalciner

H Neither
n Preheater
a Precalciner
u Neither
ta Preheater
a Precalciner



DATA ON FACILITIES SUBJECT TO NSPS
Facil ities
subject
to NSPS

Kiln
Cooler
Finish mill
Storage
Loading
Ore conveyor
Kiln
Cooler
Mills
Storage
Transfer


Mill
Storage


Finish mil 1
Storage
Transfer
Kiln
Cooler
Finish mill
Raw mill
Storage
Transfer
Reason:
N, M, RV
date

N-1974
N-1974
N-1972
N-1974
N-1974
N-1974
N-1984
N-1984
N-1984
N-1984
N-1984


N->1979
N->1979


N-1982
N-1982
N-1982
N-1979)
N-1979)
N-1979
N-1979
N-1979
N-1979
Control .
equipment

FF
FF
FF
FF
FF
FF
FF
Gravel bed
FF
FF
FF


FF
FF


FF(-)
FF(-)
FF(-)
ESP

ESP
FF(-)
FF(-)
FF(-)
.1,000 short tons  cement per year.
 1,000 short tons  clinker per year.
C,N = new facility,  R  =  reconstructed facility,  M = modified  facility.
 ESP = electrostatic  precipitator; FF = fabric  filter;  (») =  positive pressure; (-) = negative pressure;
 gravel bed = gravel  bed filter; AB = alkali  bypass.
^Listed in the 1979 review of the NSPS for the  portland cement  industry.
 Plants visited in 1983.
?Plants sent Section  114 information reguests in 1983
•Baghouse with heat exchanger.
'Baghouse with gravel bed filter.

-------
                           APPENDIX B



SUMMARY OF STATE REGULATIONS FOR PORTLAND CEMENT PLANT FACILITIES
                              B-l

-------
                             TABLE  B-l.    SUMMARY  OF  STATE  REGULATIONS  FOR  PORTLAND  CEMENT  PLANT FACILITIES
03
I
n'A
Rpgion
1






11



111


State
Conn.
Me

Mass.
N.H.
R.I.
Vt.
N.J
N.Y
P.R.

Dela.
Md.
No. of
plants3
0/0
1/0

0/0
0/0
0/0
0/0
0/0
5/1
2/1

0/0
3/0
State regulation .
Nitrogen oxides Sulfur dioxide Particulate ' '

Regulations based Equation Set 1
on fuel type and
sulfur content of
fuel.





NAAQS' E = 0.024P0-665;
P < 100, 000 Ib/h
E = 0.05 gr/dscf;
P > 100, 000 Ib/h
After 12/31/80; NSPS
41,000 ppm E = 3.59P,0-62;
P = <30 tons/h
E = 4. IP0-67,
P = 30 tons/h
E = 55P°-"-40;
P >30 tons/h

<0 7 lb/106 NAAQS E = linear inter-
Btu polation from table;

Opacity0

4 20% except for 5 min
in any 60-min period;
or NSPS.






Existing plants 420%;
New plants 410% for 3
or more min during 60 min
period.
420% for 3 min or more
during 60 min period.
460% any time.


40% or 420%, depending
on area except for 6 min
Ai r pol 1 ution
regulation reference

Chapters 101; 105, 106;
December 22, 1982.






Part 220; March 16, 1973.
Sections 403, 40/, 412,
June 27, 1980.


litle 10; June 24, 1983.
                Penn.
                Va.
                W.  Va.
                              11/2
                              1/1
                              1/0
4500 ppm
42,000 ppm
42000 ppm
SO.05  gr/dscf or
E = 55po.11-40 for
  P >30 ton/h.

K:£ =  6.23P0-12;
  P =  tons/h dry  feed
CC:E = 3.93p°-4*;
  P =  tons/h product.

Equation set 2
0-99  )b/h for wet
cement processes.
0-21.2 Ib/h for dry
cement processes.
                                                                                                    period in 60 min 440%;
                                                                                                    or NSPS.
     for 3 min or more
during 60 min period.
460% any time.

420% except for one
6-min period in any
hour not to exceed
60%; or NSPS.
420% except for one
2-min period in any
60-min period 440%.
0% for storage
structures.
                                                                                                                             Paragraphs 123 13, 123.21,
                                                                                                                             123.41, May 13,  1983
                                                                                                                             Rules  EX 2, 4,  5; March 1.
                                                                                                                             1983
                                                                                                                             Regulations VII  and X;
                                                                                                                             April 8, 1982.
                                                                                                                                                       ecll

-------
                                                     TABLE  B-l.   (continued)
CO
I
CO
EPA No. of
Region State plants3
IV Ala 5/4



Fla. 6/2
Ga 2/1

Ky. 1/1


Miss. 1/1





N.C. 0/0



S.C.f 3/3












Tenn. 2/1



V 111. 4/2





State regulation A
Nitrogen oxides Sulfur dioxide
Class 1 county:
a. 8 lb/106 Btu.
Class 11 county:
S4.0 lb/106 Btu
-
Ib/h limit based
on stack height.
<0.2-0.8 lb/106 Btu


<4.8 lb/106 Btu
or
Existing process:
<2,000 ppm
New process:
<500 ppm
'2.3 lb/106 Btu



NAAQS; cement
processes not
considered fuel
burning sources.









<4 lb/106 Btu;
<2,000 or <500 ppm
depending on
county.
<2,000 ppm.





Participate ' •
Class I county:
Equation Set 1.
Class II county:
Equation Set 2.
Equation Set lh
Equation Set 2

0.8-3.0 lb/106 Btu


Equation Set 2,
except equation 2B
for P = 30 tons/h.



Emissions SO. 437 lb/
barrel cement. 99 7%
efficiency of control
system.
For production rate
(R), in tons/h (each
kiln):
R = 10 E < 14
R = 15 E < 18
R = 20 E < 22
R = 25 E < 25
R = 30 E < 29
R = 50 E < 40
R = 60 E < 42
R - 80 E c 45
R = 100 E < 47
R = 120 E < 48
Equation Set 2



Equation Set 2, except
equation 2B for
P = 30 tons/h




Opacity0
S20% except for one
6-min period in any
60-min period not to
exceed 40%; or NSPS.
S20%, or NSPS.
S40% or NSPS.

Equation Set 1
(exception for plants
with heat exchangers).
S40%; or NSPS.





S40% except for 5 min
period in one hour.


$40% for existing
sources or S20% for
new sources.










NSPS for new
facilities.


Existing plants: S30%;
no period >60% for
8 min period during
1 hour and less than
3 times per day
New plants: S10%
Air pol lution
regulation reference
Chapters 4 and 5; March 23,
1982.


Part VI; July 1, 1983.
Chapter 391-3-1; August 27,
1982.
S20%; or NSPS


APC-S-1: Sections J, 4, 6;
December 8, 1982.




Title 15; Subchapter 20,
March 1, 1983.


Standard I, Standard III, and
Standard IV-Section 111.
June 24, 1983.










Chapter 1200-3-7, 1200-3-14,
March 2, 1983


Rules 203(b), 203(d), 204;
April 8, 1983.




                                                                                                                   (conlinueTn

-------
                                                                         TABLE  B-l.    (continued)
 tPA
Region   State
                  Ind.
                  Mich
No  of
plants
	   State legislation	c_,j_e_
Nftrogen oxides    Sulfur dToxide      Participate  '  '
                                 4/2
                                           '0 7 lb/10r'Btu   '6  lb/101' Btu.
                                 6/1
                                          NAAQS;  cement
                                          processes  not
                                          considered fuel-
                                          burning sources.
                                                                      Prior  to 12/6/68:
                                                                      E - 8.6 P»-fi7
                                                                        P <  30 tons/h
                                                                      [ -- 15.0 P°-s"
                                                                        P >  30 tons/h
                                                                      After  12/6/83:   Plant
                                                                      specific or NSPS.

                                                                      K:  <0.25  lb/1,000 Ib
                                                                      gas
                                                                      CC.  <0 30 lb/1,000  Ib
                                                                      gas.   ESP1s must have
                                                                      automatic  controller.
                                                                                                               Opacity
                                                                                           Attainment areas:
                                                                                           540% in 6 mill average;
                                                                                           460% in 15 min average
                                                                                           Nonattainment areas:
                                                                                           J30% in 6 min average;
                                                                                           $60% in 15 min average.
                                                                              $20% except:   (a) $40%
                                                                              for S3 min in 60 min
                                                                              period no more than
                                                                              3 times per 24 h,
                                                                              (b) water vapor, or
                                                                              (c) technologically and
                                                                              economically not feasible.
                                                                                                                             An pel lution
                                                                                                                          requl.it inn reference
                                                                                                          Article b. Rule 3,  Section 2,
                                                                                                          Article 5, Article  /,  June 15,
                                                                                                          1983
                                                                                              General  rules,  Parts 3 & 4,
                                                                                              August 21,  1981
Minn

Ohio
00
 I
-P"
                        0/0

                        6/2
                                                            <7  lb/10'1 Btu.
                                                                               f = 0 !>51,
                                                                                 P ' 0.05 ton/h.
                                                                               Equation Set 2;
                                                                                 P f 0.05 tons/h
                                                                              S20% except for one
                                                                              6-min average in any
                                                                              60-min period not to
                                                                              exceed 60% and except
                                                                              for water vapor and
                                                                              startup/shutdown/
                                                                              incidents.
                                                                                              litle J/45-1/-11   Aiiqust 1,
                                                                                              1982
         VI
                  Wis

                  Ark.
               2/0

               2/0
                                                            Ambient S02
                                                            levels: < 0.2 ppm
                                                                       Specific  for K and CC's
                                                                       at existing plants.
                                                                              Existing plants:   S40%
                                                                              except for 5 min period
                                                                              in 60 min.
                                                                                              Sections 4,  7,  8;  July 30,
                                                                                              1973
                  La.
                  N.M
                  Okla.
                  Tex.
                1/1


                1/1


                3/1
                                           <0 7 Ib/Btu.
                                 20/11
                                                            52,000 ppm.
                                                            NAAQS.
                                          Ambient  concentra-
                                          tion outside  plant
                                          property <0.52  ppm
                                          for 5 min period,
                                          <0.48 ppm for 1-h
                                          period;  <0.05 ppm
                                          for 24 h period.

                                          Ambient  concentra-
                                          tion <0.4 ppm for
                                          any 30 min period.
                                                              Equation  Set  2.
                                                                      $230% mg/m:s  (adopted
                                                                      January  23,  1970).

                                                                      Equation- Set 2.
                                                                       E =  0.048 Qn-b2
                                                                       Q =  stack flow rate,
                                                                       in acfm.
                                                                                                    $20% except for 4 min
                                                                                                    period in 60 min.

                                                                                                    $20%.
                                                                              $20% except for 5 mm
                                                                              period in 60 min.
                                                                              Existing plants: $30%
                                                                              over 5 min period.   New
                                                                              plants after 1/31/72:
                                                                              $20% over 5 min period.
                                                                                              Sections 19, 24; January 27,
                                                                                              1983.

                                                                                              Regulation 401,  'M?,
                                                                                              November 24, 1980

                                                                                              Regulations 3 I, 32,  J 4,
                                                                                              3.5; Apiil 9, 1982
                                                                                              Regulation I, February 1, 1982;
                                                                                              Regulation II, March 4, 1981;

-------
                                                     TABLE  B-l.   (continued)
CO
en
tl'A Nn nf

Region jUU plants Nitrogen oxides Sulfur dioxide
VU Iowa 1/3 - <500ppm

Kans 9 5/1 - NAAQS
Mo 5/2 -- Ambient air.
<0 25 ppm for 1 h
except for once
in 4 days; and
<0.07 ppm for 24 h
except for once in
90 days; Existing
sources: ^2,000
ppm New sources:
<500 ppm
Nebr 2/1 - NAAQS

VUI Col°- 3/3 - / )b/lon of
material (in-
cluding fuel)
processed.
Mont- 2/° — Requirements on
sulfur content of
fuel .



N D. 0/0
s D- 1/1 " <3.0 lb/106 Btu.






Utah 3/2 -- NAAQS

Wyo 1/1 <0. 7 lb/106 Btu NAAQS.

	 ... _. . 	 _ . . ..
b
regulation ,
Particulate0'"78"
K. <0.1 gr/dscf and
$0.3 % of inlet mass
loading or NSPS.
Equation Set 2
Equation Set 2





Equation Set 2.

Equation Set 1
Equation Set 2.




Equation Set 2






Existing and new
facil ities same as
NSPS.
Equation Set 2.


— _ — =_:— ^-. — --._-=-_- r _ — __;

Opacityc
$40% or NSPS

$40% or NSPS.
120% for existing
sources or $10% for new
sources except for
6 min period in any
60 min not to exceed
60% or NSPS




$20% or NSPS.

520% or NSPS.
Facilities prior to
12/23/68: $40% for
6-tnin period.
Facilities after
12/23/68: S20% for
6-min period; or NSPS

Existing sources: $20%
except for one 3-min
period in 60-min period
not to exceed 60%;
New sources: 410% from
kiln and all other
affected facilities.
$20% or NSPS.

Existing sources: «40%
New sources: $20% or
NSPS.
_=_ ,_ _ . 	
Air pol lut ion
regulation leference
Chapter 4 November 17, 1982.

Title 28, Part 3, May 1, 1983
Title 10: 10CSR 10-2 060;
10 CSR 10-2.160, 10 CSR
10-3 050, 10 C',R 10-3.080;
May 12, 1983





Rules 4, 5, 9, and 13;
August 6, 1982.
Part 1, Common Provision
Regulations 11, 111, VI,
July 1, 1983.
Rules 16 8. 1403, 16.8:1404,
16.8.1411, September 30, 1982.




Chapter 74.26:06-
Chapter 74 26-07,
Chapter 74:26:12
March 18, 1982



Parts 111 and IV, July 29,
1982

Sections 4, 10, 14; August 26
1981.

                                                                                                                  (conTTnuedy

-------
                                                                      TABLE B-l.    (continued)
03
I
Cr>
[PA
Region
IX





















State
Ariz







Calif.
(Monterey)
(Mountain)
(San B. )

(Shasta)
(SCAQMD)
(Bay A )
(Kern)
Hawa i i


Nev. f

No of
plants'*
2/0







13/8
(1/1)
(1/1)
(3/2)

(1/1)
(3/0)
(1/1)
(3/2)
2/1


1/0

State r
Ultrogen oxides Sulfur dioxide
For K.  193 tons/h of feed
CC:
E = 0.1 P
  P s 733 tons/h of feed
E = 55 P"-"-40
  P >733 tons/h of feed.

Installation prior to
11/1/82.   NAAQS.
Installation after
11/1/82:  NSPS.

E = 0.45(PW)°-BO;
PW < 17,000 Ib/h for
existing sources or
PW <9,250 Ib/h for new
sources.   E = 1.12(PW)°-27,
PW S17.000 Ib/h for existing
sources   E -- l.O(PW)0-25;
PW i9,250 Ib/h for new
sources.
                                                                                                          Other S10%
$20% for 3 min period
in 60 min.
                                                                                                                                     Sections 8 and 13; May  13,
                                                                                                                                     Section ]6 3 1;  July  1981
                                                                                                                                     Article 1,  198'i
Sections  1-1201, 1-1329,  and
1-1330    September 5, 1980
                                                                                                                                                         TcontTnueiTy

-------
                                                                        TABLE  8-1.    (continued)
00
LPA
Region State
X Oreg.




Wash



No of
plants8
2/1




4/0



State
Nitrogen oxides Sulfur dioxide
NAAQS




< 1,000 ppm
(corrected to 7%
02) for 60 min
period
regulation H
Particulatec'a'°
Equation Set 2.




•-0.1 gr/dscf.




Opacity0
Facilities prior to
6/1/70; «40% for 3 min
in 1 h Facilities since
6/1/70; «20% for 3 rain in
1 h or NSPS.
 30 tons/h of feed (IB)
 Equation Set 2:
  E = 4.1 P0-67 for P s 30 tons/h of  feed (IB); E - 55P"-"-40 for P >  30 tons/h of feed (2B)
'NSPS not formally adopted in SIP.
jjState has not accepted NSPS delegation.
|Exception for General Portland,  Inc., plant in Tampa,  Florida.
 •NAAQS = National ambient  air quality standard.
JNA = not available.
Source.   State Air taws.  Environment Reporter.  Bureau of  National Affairs, Inc ,   Washington,  0 C   1983.
                                                                                                                                             Air pol Iulion
                                                                                                                                          regulation  reference
                                                                                                                                       Division 21,  January 22, 1982
                                                                                                                                       Title  173, WAC  1/1-400-040,
                                                                                                                                       WAC 173-400-060,
                                                                                                                                       WAC 173-400-115; Apri I 15,

-------
                     APPENDIX C

PARTICULATE EMISSIONS AND OPACITY DATA FOR FACILITIES
      SUBJECT TO THE NSPS SINCE THE 1979 REVIEW
                        C-l

-------
                            TABLE  C-l.   PARTICULATE EMISSIONS AND OPACITY DATA FOR  FACILITIES

                                         SUBJECT  TO THE NSPS  SINCE THE  1979 REVIEW
o
I
ro
EPA
region
IV


V
VI


Company/locat ion
Ideal Basic Ind., Inc.
Theodore, Ala.
Moore McCormack
Cement, Inc.
(Fla. Mining 4 Materials)
Brooksvi 1 le, F la.
Moore-McCormack
(Dixie Cement Co.)
(formerly Ideal Basic)
Knoxvi 1 le, Tenn.
Columbia Cement Co.
Zanesvl 1 le, Ohio
Ideal Basic Ind., Inc.
lijeras, N. Hex. '
lone Star Ind. Inc.
(Oklahoma Cement )
Pryor, Ok la.
Alamo Cement Co.
San Antonio, lex.

Clinker
capa-
city.
Date- tons/
typea yr
1981- 1,415
D,PC
1982- 560
O.Ptl
1979- 512
0,HC
—
--
1979-D 267
1981- 523
D,PC

Emission
control
equipment
(FF (-)
(w/cooler
and raw
mi 1 1
dryers )
FF(-)
FF(-)
-
--
FF
ESP
(w/cooler
and raw
mi 1 1 )
KILN

Participate Opa-
^ emissions city,
" Ib/h
52.18
(9/83)
6.5
(9/82)
4.63
(12/79)
-
—
6.5
(3/80)
22
Main stack
15.8
Bypass stack
Ib/ton J
0.22 12.06
(9/83) (Avg.)
0.058 0
(9/82) (9/82)
0.039
(12/79)
—
~-
0.112
(3/80)
0.19
( 1 /83 )
0.01
(1/83)
CLINKER COOLER
Emission Particulate
control h emissions
Date equipment Ib/h Ib/ton
1981 FF(-) (w/
kiln and
raw mitt
dryers)
1982 FF(-) 4.98 0.044
(9/82) (9/82)
1979 FF(-) 0.82 0.008
(12/79) (12/79)
—
—
1979 GB 13.8 0.10
(3/80) (3/80)
1981 ESP (w/kiln
and raw mi M )
OTHER FACILITIES
Opa- Emission
city, control .
t Date equipment
1981 Entire
plant-FF(-)
0
(9/82)
.. _ _
19/8 Finish
mill-FF
>1979 Finish
mlll-FF,
CYC
—
1981 Entire
plant-FF
(except
finish mill )
Opa-
city,
I


~
1C
1C
—
"
                                                   4.0    0.024

                                                Bypass stack  (4/84)
                                                                                                                  1715531

-------
                                                     TABLE C-l.   (continued)
o
I
OJ


tPA
region Company/local ion
VI Capital Aggregates, Inc.
San Antonio, Tex.


Centex Corp.
(Texas Cement )
Buda, lex.
General Portland, Inc.
New Uraunfefs, Tex.
Gul 1 Coast Portland
Cement Co,
(Uiv. ol McOonough)
Houston, lex.

Kaiser Cement Corp.
(Longhorn plant)
San Antonio, Tex.

Lone Star Ind. Inc.
Maryneal , Tex.

Southwestern Portland
Cement Co.
Bush land (Amarillo), Tex.


Southwestern Portland
Cement
Odessa, Tex.


KILN
	 (' 1 ! nL*ir- 	 ~ 	 ~~~ — " ~~ ~~ ~— 	 ^— — — — — — 	 .
L 1 1 nker
capa-
city, tmission Particulate Opa-
Date- tons/ control . emissions city.
tyPe yr equipment Ib/h Ib/ton <
1983- 500 FF NAc'd NAC'd NAc>d
O.PC


1983- 468 FF (w/raw NAC NAC NAC
D,PC mill)

1980- 875 2-F.SP's ?5.6 0.129 0-5
".PC (-/raw (5/8?) (5/82)
mi II )





1975- 775 3-ESP's; FF 13.89 0.088 -10
D, 2PC on alkal i
(2nd PC bypass
in 1979)
—


__




1978- 279 FF 10.73 0.148 3.7
°.PH (2/83) (2/83) (2/83)


CLINKER COOLER OTHER FACILITItS

Fmisslon Particulate Opa-
control ^ emissions city.
Date equipment Ib/h Ib/ton t Date
1983 FF NAC'd NAc'd NAC'd 1983


1983

1980 GB 19.3 0.100 5-10 1980
(7/82) (7/82)

1973
and
1978
1978
1975 FF 1.889 <0.05 — 1977



1979


1981



1982



Emission
control
equipment
Mil Is,
storage.
transfer -FF
Raw mi 1 1 -II
(w/ki In)
Entire
plant-FF

Finish
mil Is-FF

storage-f F
Finish
mi II -FF


Coal
transfer -FF

Coal
storage.
coal
transfer-FF
Coal
storage.
coal
transfer, FF

Opa-
city,
%



__

-

1C



..



	


	



"


                                                                                                                    (conf TnuetTT

-------
TABLE C-l.   (continued)
EPA
region Company/location
Texas Industries
Hunter, Tex.
Texas Industries
Midlothian, lex.

VII Davenport Industries
BuUalo, lo»a
Lehigh Port land
Cement Co.
Mason City, Iowa
lone Star tnd. Inc.
(Marquette Cement )
Cape Girardeau, Mo.
River Cement Co.
Festus, Mo
<"> Ash Grove Cement Co.
^ Louisvi 1 le, Nebr.
VIII 1 dea 1 Bas i c 1 ad . , 1 nc .
(Boettcher plant)
la Porte (Fort
Col lins), Colo.
Martin Marietta Corp.
Lyons, Colo.




Date-
type3
1980-
D,PC
„


1981-
D.PC
1979-
D,PC

1981-
D,PC

—

1982-
D,PC
1981-
D,PII


1979-
D,PC



KILN

Clinker
capa-
clty, Emission Particulate Opa-
tons/ control K emissions city.
yr equ i pmen t 1 b/h
664 ESP 31.87
(7/81)
	


809 FF (w/raw 20.37
mill) (8/83)
543 ESP (w/raw 35.4
mill) (6/83)

992 ESP 29.17
(3/82)

„

558 tSP 7.87
(7/83)
440 FF(t) 13.1
(4/82)


405 FF(-) 10.29®
9.98
(10/80)
FF(-) 3.30
(Alkali (10/80)
bypass >
Ib/ton %
0.229
(7/81 )
	


0.1358
(8/83)
0.266
(6/83)

0.12
(3/82)

—

0.065
(7/83)
0.14
(4/82)


0.095?
0.094
(10/80)
0.03
(10/80)

CLINKER COOLER OTHER FACILITIES
Emission Particulate Opa- Emission Opa-
control h emissions city, control < city,
Date equipment Ib/h Ib/ton ( Date equipment I
1980 FF 3.25 0.0233 — 1980 Entire
(12/81) (12/81) plant-FF
1979 Finish
ml II-ESP,
transfer-FF
1981 FF 20.75 0.138 -- 1981 Entire
(8/83) (8/83) plant-FF
1980 Mill,
separators-
FF
1981 2 FF "s 13.7 0.06 — 1981 Entire
(N/raw mill) plant-FF

>1979 Rawmill-FF

1982 FF(*) 5.83 0.048
(7/83) (7/83)
1981 FF(-)(w/raw 16.4 0.191 — 1981 Entire
(mill) (2/82) (2/82) plant
(exept finish
mill)-FF
1979 Limestone
dryer -FF(-)




-------
                                               TABLE C-l.   (continued)
o
I

EPA
region Company/location
VIII lone Star Ind., Inc.
(Utah Portland Cement)
Salt lake City, Utah
Martin Marietta Corp.
Leamington, Utah


Monolith Portland
Cement Co.
Laranne, Wyo.
IX Cal ifornia Portland
Cement Co.
Mojave, Cal i 1.

Genstar, Ltd.
(Div. ol Flintkote)
Read ing, Ca 1 i I .

Genstar, Ltd.
(Div. of Fl intkote)
San Andreas, Cal if.
Kaiser Cement Corp.
Cushenbury Plant
Lucerne Valley, Calif.
Kaiser Cement Corp.
(Cupert ino)
Permanent e, Cal i f .

Date-
type3
1979-W
1982-
D.PC


198 I-W
1981-
D,PC

1981-
D,PC

1945-H
1952-W
1956-W
1982-
D,PC
1981-
D,PC

capa-
clty,
tons/
yr
150
603


300
1,000

571

192
192
192
1,520
1,520
KILN
Emission
control
equipment
FH-)
FF <«/raw
mi II )
FF
(Alkali
bypass )
ESP
FF(-)
(M/raw
mill)

FF(-)
(w/rax
mi 1 1 and

Multi-CYC,
ESP
FF (H/rav
mil 1 )
FF(-)
Part icu late Opa-
emissions city,
Ib/h
8.74
(9/80 )
9.949
(6/83J
5.8?"
(10/82)

5.5
(No. 2
kiln
5/82)
15.34
(5/83)

2.87?
5.061
(5/81)

45J
(11/79)
18
(5/83)
8.0
(9/83,
10/83)
Ib/ton I
0.294 0
(9/80) (9/80)
0.089
(6/83J
0.0411
(10/82)

O.lll 5-15
(No. 2 (No. 2
kiln kiln
5/82) 6/82)
0.07
(5/83)

0.027? 0
0.05)'
(5/81 )

0.29 4
(11/79)
0.066
(5/83)
0.030
(9/83,
10/83)
CLINKER COOLER
Emission
control
Date equipment
1979 FF(-)
1982 FF


1981 FF(-)
1981 FF(-)

1981 FF(-)
(•/raw mi 1 1
and cooler)



1982 FF («/
alkali
bypass)
1981 FF(-)
OFHER FACILITIES
Part icu late Opa-
emissions city.
Ib/h
1.58
(10/80)
3.96
(!2/83>


0.17
(5/82)
4.98
(5/83)

—



1.9"
(5/83)
2.3
(10/83)
1 b/ton I Date
0.049 0
(10/80) (10/80)
0.034 0 1982
(12/83) (12/83)


0.004 0 1981
(5/82) (5/82)
0.04 — 1981
(5/83)

1981



0.006" — 1982
(5/83)
0.0082 — 1981
(10/83)
Emission Opa-
control . cHy,
equipment %
-
Entire 1C
plant-FF


F inish mi 1 1 ,
cement
cooler-FF(-)
Raw mill-
IP!-)
(•/kiln)
Raw mi 1 1-
FF(-) (*/
kiln and
cooler)


Entire
plant-FF
Entire
plant-FF (- )

-------
                                                                           TABLE  C-l.    (continued)
o
cr>

KILN
Ci;n|(er 	 — 	 	 	 	
capa-
fp. city. Emission Particulate Opa-
. Date- tons/ control emissions r.tw
region Company/location Ivoe vr .n,,i™L t° — ll/i -ILJI CI'Y.

IX tone Star Ind., Inc. 1981- 744 ESP (./ran 3.75? 0 0229 3
Davenport (Santa Cruz), D,PC m, ,, ) 6.451 O.OK1 (12/83)
'-<""• (12/83) (12/83)



Southwestern Portland 1984- 800 NAc'd NAc>d u»c'd u»c.d
Cement Co. 0 PC Nfl NA
Viclorvi lie, Cal if .
lone Star Ind., Inc.
Ewa Beach, Hawai >
X Alaska Basic Inc.
Anchorage, Alaska



Oregon Portland Cement 1979- 500 ESP (w/ 6 35 0 055 I
Durkee, Oreg. 0,PH COOier) (10/83) (10/83)




Cl INKER COOLER OTI€R FACILITIES

Emission Particulate Opa- Emission Opa-
control . emissions city rontmi ^i*u
Date equipment" -fb>h Ib/ton /' Date equfpmenl" ' \'
1981 M "-54 0.056 - 1981 Finish mill
(12/83) (12/83) ra« feed
si lo.
transfer-
FF(-)



>1979 Mill,
storage-FF
1982 Finish mill.
storage.
transfer-
FF(-)
1979 ESP  - - - )979 Ent|re
plant-FF(-),
finish mlll-
ESP

        cfabrlc filter; GB = gravel  bed  filter.
        dNA - Data not available; test data not completed.
        fiFacililies under construction.
        flype \ clinker production.
         Fype 2 clinker production.
        ^Rao mi I I  on-IIne.
         Alkali  b/pass.
         Haw mill  bypassed.
         Ihree  ktlns  in opetatioo.
                                                                 ocess with  preheater, D, PC  = Dry process «tth precalclner/preheater
                                                                 preclpltator; fF - fabric filter (baghouse); FF(O = positive-pressure fabr
Ic filter; FF(-) =  negative-pressure

-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing/
1. REPORT NO.
EPA-450/3-85-003a
4. TITLE AND SUBTITLE
Portland Cement Plants--Bac
Proposed Revisions to Stand
7. AUTHOR(S)
9. PERFORMING ORGANIZATION NAME AIN
Office of Air Quality Planni
U. S. Environmental Protecti
Research Triangle Park, Nort
12. SPONSORING AGENCY NAME AND AOC
Director for Air Quality Pla
Office of Air and Radiation
U. S. Environmental Protecti
Research Triangle Park, Nort
2. 3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
kground Information for May 1985
ar(js 6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
IO ADDRESS 10. PROGRAM ELEMENT NO.
ng and Standards
on Agency 11. CONTRACT/GRANT NO.
h Carolina 27211 68-02-3817
RESS 13. TYPE OF REPORT AND PERIOD COVERED
nning and Standards Final
14. SPONSORING AGENCY CODE
on Agency FPA/200/04
h Carolina 27711 hPA/^uu/U4
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Revisions to the standards
cement plants (40 CFR Part
Section 111 of the Clean Ai
information gathered during
17.
a. DESCRIPTORS
Air pollution
Pollution control
Standards of performance
Portland cement plants
Kilns
Clinker coolers
Particulates
18. DISTRIBUTION STATEMENT
Unlimited
of performance for the control of emissions from port! and
60.60) are being proposed under the authority of
r Act. This document contains a summary of the
the review of this new source performance standard.
KEY WORDS AND DOCUMENT ANALYSIS
b.lDeNTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Air Pollution Control 13 B
19. SECURITY CLASS (This Report) 21. NO. OF PAGES
Unclassified 123
20. SECURITY CLASS (This page) 22. PRICE
Unclassified
EPA Form 2220-1 (Rev. 4-77)
                              PREVIOUS EDI TION IS OBSOLETE

-------
.4

-------
  UnHed Statps
  Environmental Protection
  Agency
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park NC 27711
Official Business
PuM.illy lor Private Use
'..100
                                                                          If your address is incorrect, please change on the above label.
                                                                          tear off, and return to the above address
                                                                          If you do not desire to continue receiving this technical report
                                                                          series, CHECK HERE D , tear off label, and return it to the
                                                                          above address

-------