United States Solid Waste and
Environmental Protection Emergency Response EPA530-R-94-001
Agency (5305) DecembeM993
4>EPA Report to Congress on
Cement Kiln Dust
Methods and Findings
(Graphics not included on Diskettes)
Primed on /?t>.".,-.Vj P^per
-------
15
§3001(b)(3)(C), reach a final Regulatory Determination on the management status of CKD
within six months of submission of this Report. The Regulatory Determination requires the
Agency only to determine to promulgate regulations under Subtitle C, or determine that Subtitle
C is unwarranted. Thus, if RCRA §3004(x) or Option 5 is chosen, EPA would have time beyond
six months to promulgate a Final Rule.
-------
2-7
Exhibit 2-3
Geographic Distribution of Active Portland Cement Production Facilities in 1991 in the U.S.
o
Smitic I'lullaml C'cutenl Association, I992b. op. tit.
-------
2-11
Exhibit 2-5
Steps in the Manufacture of Purl land Cement
Step 2 - Kaw Materials are Ground to Powder and Blended
Dry
Process
Step!
Kaw materials arc quarried,
crushed to 125mm si/.c,
then to 20mm, and stored.
Air Separator
L
Raw —K-' Grinding
Materials F~ Mill
Dust Collector
Dry Mixing
and Blending
Silos and
Storage
I-inc Materials
Step 2 - Raw Materials are Ground. Mixed with Water to
Form a Slurry, and Blended
Wet
Process
yVater added
here
'Vibrating Screen
Raw
Grinding l—
Materials f 1 Mill |~
Slurry
To
Kiln
To
Kiln
Slurry is Mixed and Blended, then Stored '
Sunn i- Ail.»|>tctl lioin I'oitl.iiKl ('email Association. IW2 [i\i\ /\/M/V.V/V <
-------
2-12
Kxhihit 2-5 (continued)
Steps in the Manufacture of Portland Cement
Slop 3 - Burning Chemically Changes Raw Mix into Cement Clinker
Exhaust
Slack
Oust Materials added
Collector lo kiln
Raw Mix is Burned lo Partial
Fusion al I4SO-I650"C
puc| added
Ll^--
Clinker and
Gypsum (Stored
Separately) Conveyed
To Grinding Mill
Siq> 4 - Clinker with Gypsum Ground into Portland Cement
Air Separator
Grinding
Mill |—-i
Dust Collector
Transportation of Hulk
Material by Truck
oV Train
Cciiicnl Pump
.11, r A,l.i|iu-.l lioiii I'uiil.uiil I ciiu-iu Awui.iliim, l'*1^' IAnAnnl\\n n/.SV/n /<•/ lm\c Mri,il.\ in Ci'iinnninil Kiln Dual >
-------
3-5
Exhibit 3-3
Schematic Diagrams of Common Types of Air Pollution Control Devices
Collected Dust
Cyclone Collector
Source: Adapted from Kirk-Othmer, 1979
Collected Dust
Counter-current Gravity Separator
Source: Adapted from Kohlhaas et al. 1983
Collected Dust
Inertial Separator
Source: Adapted from Kohlhaas et al, 1983
<^ i Qean gas
^^•1 Dust-laden gas
a Coarse particles
b Fine particles
-•'"' Streamlines of the fluid
-------
3-6
Exhibit 3-3 (continued)
Schematic Diagrams of Common Types of Air Pollution Control Devices
Perforated
Metal
Screen •
Granular
Activated
Carbon
Air or Gas for Pulse-Cleaning
Granular Bed Filter
Source: Adapted from Kirk-Othmer. 1979
Dusty .Air
Side
Screen Filter
Source: Adapted from Kirk-Othmer, 1979
High-Voltage Discharge
Electrode (-)
Charging Field
Charged (-) Particles
Particle Path
Collecting Baffle
Grounded (+) Collecting
Surface
Discharge Electrode
Tension Weight
Electrostatic Precipitator
^ i Clean gas
^I^B Dust-laden gas
- ; ; " " Streamlines of the fluid
Source: Adapted from Kirk-Othmer, 1979
-------
6-3
Exhibit 6-1
Overview of Risk Assessment Methodology
Evaluate Chemical Concentrations
in GKD
Compare sampling data from
20 cement plants to
risk-screening criteria
CKD constituents, exposure pathways, and
waste streams that warrant more detailed study
Evaluate Risks of CKD
When Managed On Site
1. Evaluate risk potential ol 15 initial case-
study facilities
2. Model risk at selected facilities with highest
risk potential
3. Model risks in potentially high-risk scenarios not
captured by site-specific modeling
Evaluate Risks of Off-site Beneficial
Uses of CKD
1. Evaluate risk potential at 5 sample off-site
locations
2. Identify scenarios that, represent relatively high-risk
management practices
3. Model risks for potentially high-risk scenarios
-------
Intrinsic c
Hazard of ;
CKD ; ^:
'"• • • - •','• '"'•• •'-
Contamination
Potential; v
-transport
^Potential
Exposure
Potential
6-IK
Exhibit 6-5
Site Specific Factors Used to Evaluate Risk Potential of On-Site CK1) Management
Ground Water
1. Concentration of mobile and persistent
chemicals relative to drinking water
criteria
2. Number of constituents exceeding
toxterty criteria
1. Ground-water monitoring data
2. Area of CKD unit at base
3. Availability ol water to seep through
CKD unit
a. CKD under water, or
b. Net recharge
4. Containment
5 CKD unt type
6. Primary unaaturated zone materials
underlying management unit
7. Depth to ground water
1. Lateral containment (slurry walls)
2. Potential tot.ground water to discharge
directly to surface water
3. Karst or fault area
4, Distance from unit to downgradient
property boundary
5. Distance from unit to nearest existing
downgradient residence
1 Ciound tvutot presently used
2 f'opulolHH] williin one mile downyiadiont
Surface Water
1. Concentration of persistent chemicals
relative to surface water criteria.
2. Number of constituents exceeding
toxlclty criteria
1. Surface water monitoring data
2. Potential for contamination via
atormwater runoff
a. Containment
b. Location relative to 100-yr ftoodplaln
c. Distance to receiving surface water
d. Intensity of rainfall
e. Area of CKD unit at base
3. Potential for contamination via
ground-water discharge
a. Ground- water contamination potential
b. Potential tor ground water to
discharge to surface water
c. Distance to receiving surface water
4. Potential for contamination via airborne
release
a. Air contamination potential
b. Width of receiving surface water
c. Distance to receiving surface water
1. Surface water dilution potential
2. Distance from potential point of
release to surface water to potential
point of use by people
I Ecological sensitivity
2 P/H.senf human uses oi surface
1. Concentration of persistent chemicals
relative to air release oH-srte
exposure criteria
2. Number of constituents exceeding
toxicity criteria
1. Ambient air monitoring data
2. CKD transported in open-air trucks
3. Containment
4. Exposed surface area of CKD unit '
5. Mean annual wind speed
6. Ratio of erosion threshold wind speed
to fastest speed wind
7. Precipitation-evaporation index
8. Number of days per year with
2 0.01 inch of precipitation
1. Distance from CKD unit to nearest
property boundary
2. Distance from CKD unit to nearest
existing residence
1. Number ot existing residences within
one mile ol property boundary
2. Surrounding land uses
-------
6-31
Exhibit 6-10
(Graphical Illustration of On-sitv Risk Modeling Scenarios
SeniWvHyAnalyib of Higher Risk On-SHe Management Scenarios
95th Locolton U»
^ Honest Rarcentlle Adjocentto Adjoc
sk Potential Analysis of
>le of 1 5 Facilities
Relative Ri
Samp
Measuea MBloJs /"ortaitural Surt
aoxlrs Concentrations Field Vtt
J*" under\*ater - Subsbtenoe
>6fu lO
. Management Food
ace •
In Quarry Consumption
3ter
• Best Estimate • Best Estimate • Best Estimate • Best Estimate • Best Estimate • Best Estimate
• upper Bound •upper Bound •upper Bound •upper Bound •upper Bound •upper Bound
* baseline Modeling
*" /Analysis of
* OrvSlteCKD
b Monagement
Fl w • Cenlrd Tendency Rbte
Selected •HgnEndRlsks
FaclllHes
-------
K-IO
!• vl.il.il 8-2
Process Flow Diagram of Recovery Scrubber
Kiln Dust Input
Scrubbed Exhaust Reaction
Tank
->• CKD Movement
-^ Water Movement
^ Gas Movement
Acid Movement
Potassium
Sulfaie
Output
'in. «• Ail.i|Kcil limn l',iss.iin.i()ii(H!(ly Technology
-------
8-15
Exhibit 8-3
Process Flow Diagram of Fluid Bed Dust Recovery Process
(With Heat Recovery System)
Vent gases to
Secondary dust collection
Suspension
Preheater
Cyclone
Fluddized
Bed
Reactor
Start-up Air
Heater
Reactor Fluidizing
Air Blower
Vent gases to
Secondary dust collection
Cooler
Cvclone
Cooler Fluidizing
Air Blo*er
Material Movement
Gas Movement
Source: Adapted from Fuller Company
-------
CONTENTS
REPORT TO CONGRESS ON CEMENT KILN DUST
Volume II: Methods and Findings Page
LIST OF ACRONYMS
CHAPTER ONE: INTRODUCTION 1-1
1.0 PURPOSE AND SCOPE 1-1
1.1 GENERAL METHODS AND INFORMATION SOURCES 1-2
1.1.1 General Analytical Methods 1-3
1.1.2 EPA Data Collection Activities 1-3
A. 1991 Portland Cement Association (PCA) Survey of
Cement Kiln Facilities 1-3
B. 1992 and 1993 CKD Sampling and Analysis 1-3
C. Damage Case Collection 1-4
D. EPA Site Visits 1-4
E. RCRA §3007 Waste and Site Characteristics Data Request .... 1-5
1.2 EPA'S DECISION MAKING RATIONALE 1-5
U CONTENTS AND ORGANIZATION OF REPORT 1-7
CHAPTER TWO: CEMENT INDUSTRY OVERVIEW 2-1
2.0 INTRODUCTION AND DESCRIPTION OF CKD 2-1
2.1 CEMENT INDUSTRY STRUCTURE AND CHARACTERISTICS 2-1
2.1.1 What Is Cement? 2-1
2.1.2 The Cement Industry 2-5
2.2 CEMENT MANUFACTURING 2-9
2.2.1 The Basic Production Process 2-10
Mining 2-10
Crushing 2-10
Grinding and Blending 2-13
Pre-Drying 2-13
Drying and Preheating 2-13
Calcining 2-15
Burning 2-15
Cooling 2-15
-------
CONTENTS (continued)
Volume II: Methods and Findings Page
Finish Milling/Loading 2-16
2.23. Kiln Design 2-16
Wet Process Kiln 2-18
Dry Process Kiln 2-19
Preheater Kiln 2-21
Precalciner Kiln 2-23
Semidry Process Kiln 2-23
Trends in Kiln Technology and Use 2-25
2.2.3 Process Inputs 2-26
Raw (Feed) Materials 2-26
Fuels 2-33
Fossil Fuels . 2-37
Non-hazardous Waste Fuels 2-39
Hazardous Waste Fuels 2-40
. Overview of the Hazardous Waste Fuels Industry 2-40
Extent of Hazardous Waste Consumption by Cement Kilns
and Types of Waste Burned 2-41
Hazardous Waste Fuel Technologies 2-42
Relevant Environmental Regulations 2-42
2.3 SUMMARY AND RELEVANCE TO SUBSEQUENT CHAPTERS 2-43
CHAPTER THREE: CKD GENERATION AND CHARACTERISTICS 3-1
3.0 INTRODUCTION 3-1
3.1 CKD GENERATION 3-1
3.1.1 Dust Collection Devices 3-2
3.1.2 Plant-Level CKD Generation Rates 3-2
3.1.3 Quantities and Fate of CKD Generated 3-16
Differences in CKD Generation Rates Across Process Types 3-16
Differences in CKD Generation Rates Across Process Types
and Fuel Usage 3-18
Differences in Gross CKD Generation Rates 3-18
Differences in Net CKD Generation Rates 3-19
CKD Recycling 3-19
Recycling Differences 3-21
Differences in Fate of CKD Across Process Types 3-21
-------
Ill
CONTENTS (continued)
Volume II: Methods and Findings Page
Differences in Fate of CKD Across Process Types and
Fuel Usage 3-22
3.2 CKD GROSS CHARACTERISTICS 3-25
3.2.1 Physical Characteristics 3-25
3.2.2 Bulk Chemical Characteristics 3-29
3.3 CKD TRACE CHARACTERISTICS 3-30
3.3.1 EPA Sampling Program 3-30
332 Total Concentrations 3-32
Metals 3-33
Dioxins and Furans 3-37
General Chemistry 3-41
Volatile Organics 3-41
Semi-Volatile Organics 3-44
Pesticides 3-44
PCBs 3-45
Radionuclides 3-45
33.3 Leachable Concentrations 3-47
Metals 3-47
Dioxins and Furans 3-50
General Chemistry 3-50
Volatile Organics 3-50
Semi-Volatile Organics 3-50
Pesticides 3-52
PCBs 3-52
Radionuclides 3-52
3.4 STATISTICAL ANALYSES OF CKD CHARACTERIZATION RESULTS 3-53
3.4.1 Metals 3-53
3.4.2 Dioxins and Furans 3-61
3.5 CLINKER CHARACTERISTICS 3-62
CHAPTER FOUR: CURRENT MANAGEMENT PRACTICES FOR CKD 4-1
4.0 INTRODUCTION 4-1
-------
IV
CONTENTS (continued)
Volume II: Methods and Findings Page
4.1 ON-SITE LAND DISPOSAL 4-1
4.1.1 LandQlls 4-2
4.1.2 Piles 4-2
4.1 J Ponds 4-2
4.1.4 Dimensions 4-3
4.1.5 Codisposal 4-3
4.1.6 Remaining Useful Life 4-3
4.2 POTENTIAL EXPOSURE PATHWAYS 4-4
4.2.1 Ground Water 4-5
4.2.2 Surface Water 4-6
4.23 Air 4-6
4.3 ENVIRONMENTAL PROTECTION PRACTICES 4-7
4.4 BENEFICIAL USE OF CKD 4-9
4.4.1 Waste Stabilization 4-10
4.4.2 Soil Amendment (Fertilizer) 4-12
4.4.3 Liming Agent 4-12
4.4.4 Materials Additive 4-12
4.4.5 Road Base 4-13
4.4.6 Other Uses 4-13
CHAPTER FIVE: DOCUMENTED DAMAGES FROM MANAGEMENT OF CKD 5-1
5.0 INTRODUCTION AND METHODOLOGY 5-1
"Tests of Proof 5-1
Identification of Prospective Damage Cases 5-2
Information Collection 5-2
Damage Case Preparation and Review 5-3
-------
CONTENTS (continued)
Volume II: Methods and Findings Page
Limitations of the Damage Cases 5-3
5.1 OVERVIEW OF FINDINGS, TRENDS, AND CONCLUSIONS 5-4
5.1.1 Findings 5-4
5.1.2 Overall Trends and Conclusions 5-5
5.2 DOCUMENTED GROUND AND SURFACE WATER DAMAGE CASE
SUMMARIES 5-9
Cases of Documented Damage 5-9
5.2.1 Holnam Incorporated, Mason City, Iowa 5-9
5.2.2 Lehigh Portland Cement Company, Leeds, Alabama 5-14
5.2J Lehigh Portland Cement Company, Mason City, Iowa 5-17
5.2.4 Portland Cement Company, Salt Lake City, Utah 5-22
5.2.5 Southwestern Portland Cement (Southdown, Inc.), Fairborn, Ohio 5-26
5.2.6 National Gypsum Co./Lafarge Corp., Alpena, Michigan 5-32
5.2.7 Ash Grove Cement West, Montana City, Montana 5-34
5J CASES OF POTENTIAL DAMAGE TO GROUND AND SURFACE WATER . . . 5-37
5.3.1 Texas Industries, Inc., Midlothian, Texas 5-37
5.3.2 Holnam, Inc., Artesia, Mississippi 5-40
5.3.3 Markey Machinery Property, Seattle, Washington 5-41
5.4 DOCUMENTED AIR DAMAGES 5-45
5.5 CKD MANAGEMENT SCENARIOS OF CONCERN 5-49
CHAPTER SIX: POTENTIAL DANGER TO HUMAN HEALTH
AND THE ENVIRONMENT 6-1
6.0 INTRODUCTION 6-1
Purpose and Scope 6-1
-------
VI
CONTENTS (continued)
Volume II: Methods and Findings Page
Overview of Approach 6-2
Major Results and Conclusions 6-4
6.1 INITIAL RISK SCREENING 6-6
6.1.1 Approach and Methods 6-6
CKD Composition Data 6-6
Risk-Screening Criteria 6-7
Other Constituent-Speciflc Factors 6-10
6.1.2 Risk-Screening Results 6-11
6.2 EVALUATION OF RISKS WHEN CKD IS MANAGED ON SITE 6-14
6.2.1 Risk Potential Ranking of Initial Case Studies 6-15
Approach and Methods 6-15
Results of Risk Potential Ranking 6-19
Risk Potential Ranking for the Ground-water Pathway 6-20
Risk Potential Ranking for the Surface Water Pathway 6-23
Risk Potential Ranking for the Air Pathway 6-26
6.2.2 Risk Modeling of On-site CKD Management 6-29
Analytical Methodology 6-29
Release. Fate, and Transport Modeling Methodology 6-32
Characterization of Exposed Populations 6-34
Exposure Assessment and Risk Characterization 6-35
Sensitivity Analysis of Higher Risk Potential Scenarios 6-38
Results of On-site Risk Modeling 6-40
Baseline On-site CKD Management 6-40
Sensitivity Analysis of Potentially Higher Risk Scenarios 6-45
6.23 Summary of Risks from On-site CKD Management 6-51
Ground-Water Risks 6-51
Surface Water Risks to Human Health 6-52
Aquatic Ecological Risks 6-52
Air Pathway Risks from Windblown Dust 6-52
6.3 EVALUATION OF RISKS FROM OFF-SITE BENEFICIAL USES OF CKD . . . 6-53
6.3.1 Approach and Methods 6-54
-------
V1J
CONTENTS (continued)
Volume II: Methods and Findings Page
Hazardous Waste Stabilization and Disposal ...................... 6-54
633 Sewage Sludge Treatment and Use ............... ............... 6-56
6.3.4 Building Materials Addition ................................... 6-57
6.3.5 Road Construction .......................................... 6-58
Analysis of Risk Factors ...................................... 6-58
Risk Modeling Results for Unpaved Traffic Surfaces ................ 6-59
6.3.6 Agricultural Liming ......................................... 6-60
Analysis of Risk Factors ...................................... 6-60
Risk Modeling Results for Liming .............................. 6-62
6.3.7 Summary of Risks from Off-site Beneficial Uses ................... 6-63
CHAPTER SEVEN: EXISTING REGULATORY CONTROLS ON CKD MANAGEMENT . 7-1
7.0 INTRODUCTION AND METHODS .................................. 7-1
7.0.1 Objectives ................................................. 7-1
7.0.2 Methodology ............................................... 7-1
7.0.3 Summary of Findings ........................................ 7-2
7.0.4 Limitations of the Analysis ................................... 7-4
7.1 AIR [[[ . 7-5
7.1.1 Federal Controls ........................................... 7-5
Clean Air Act .............................................. 7-5
Implementing Regulations .................................... 7-5
State Implementation Plans ............................. 7-5
New Source Performance Standards ...................... 7-6
Prevention of Significant Deterioration .................... 7-6
Nonattainment Review ................................ 7-7
Hazardous Air Pollutants .............................. 7-7
Boiler and Industrial Furnace Regulations ....................... 7-8
7.1.2 State Controls ............................................. 7-9
-------
vm
CONTENTS (continued)
Volume II: Methods and Findings Page
Ambient Air Quality Standards 7-11
Authority To Construct 7-11
Permit To Operate 7-12
Prevention of Significant Deterioration and New Source
Review 7-12
Michigan 7-13
Ambient Air Quality and Permits 7-13
Paniculate Matter and Visible Emissions Limitations 7-14
Pennsylvania 7-14
Nonattainment Review 7-15
Particulate Matter and Visible Emissions Limitations 7-15
Texas 7-16
Other Requirements Applicable to Cement Plants 7-16
Special Report on Texas Cement Plants' Excess Emissions 7-18
7.2 WATER 7-18
7.2.1 Federal Controls 7-18
The Clean Water Act 7-18
Safe Drinking Water Act 7-20
7.2.2 State Controls 7-21
California 7-22
Process Wastewater Requirements 7-23
Water Quality Standards 7-23
Stormwater Management Requirements 7-24
Michigan 7-24
Water Quality Standards 7-25
Ground-Water Protection 7-25
Pennsylvania 7-25
Water Quality Standards 7-26
Ground-Water Protection 7-26
Stormwater Management Requirements 7-27
Texas 7-27
Water Quality Standards 7-28
Ground-Water Protection 7-28
Stormwater Management Requirements 7-29
73 SOLID WASTE MANAGEMENT 7-29
73.1 Federal Controls 7-30
The Resource Conservation and Recovery Act 7-30
Boiler and Industrial Furnace Rule 7-31
-------
IX
CONTENTS (continued)
Volume II: Methods and Findings Page
Comprehensive Environmental Response, Compensation, and
Liability Act 7-32
7.3.2 State Controls 7-32
California 7-33
Michigan 7-36
Pennsylvania 7-37
Texas 7-40
CHAPTER EIGHT: ALTERNATIVE CKD MANAGEMENT PRACTICES AND
POTENTIAL UTILIZATION 8-1
8.0 OVERVIEW 8-1
8.1 MINIMIZATION OF CKD REMOVAL FROM THE KILN SYSTEM 8-2
8.1.1 Control of CKD Generation Rates 8-2
8.1.2 Direct Return of CKD to the Kiln 8-3
Return to Flame End 8-5
Return with Raw Feed 8-5
8.1 J Treatment and Return of CKD to the Kiln 8-6
Pelletizing 8-6
Leaching with Water 8-6
Leaching with a Potassium Chloride Solution 8-8
Alkali Volatilization 8-8
Recovery Scrubbing 8-9
Fluid Bed Dust Recovery 8-13
8.2 BENEFICIAL USE OF REMOVED CKD 8-16
8.2.1 Stabilization of Sludges, Wastes, and Contaminated Soils 8-17
Sewage Sludge 8-17
Oil Sludge 8-21
Acid Waste 8-22
Miscellaneous Wastes and Contaminated Soils 8-23
8.2.2 Soil Stabilization 8-24
8.2.3 Land Reclamation 8-24
-------
CONTENTS (continued)
Volume II: Methods and Findings Page
8.2.4 Agricultural Applications 8-25
Fertilizer 8-25
Liming Agent 8-28
8.2.5 Livestock Feed Ingredient 8-28
8.2.6 Lime-Alum Coagulation in Water Treatment , 8-31
8.2.7 Construction Applications 8-31
Blending with Portland Cement 8-32
CKD as the Only Blending Agent 8-32
CKD as a Co-Blending Agent 8-33
Use as a Road Base Material 8-34
8.2.8 Sanitary Landfill Daily Cover 8-35
8.2.9 Mineral Filler 8-36
8.2.10 Lightweight Aggregate 8-36
8.2.11 Glass Making 8-36
8.3 ON-SITE LAND DISPOSAL 8-36
8.4 SUMMARY AND FINDINGS 8-38
8.4.1 Technical Feasibility 8-43
8.4.2 Human Health/Environmental Considerations 8-44
8.4.3 Economic Feasibility 8-44
8.4.4 Current Extent of Use and Trends 8-45
CHAPTER NINE: COST AND ECONOMIC IMPACTS OF ALTERNATIVES TO
CURRENT CKD DISPOSAL PRACTICES 9-1
9.0 INTRODUCTION 9-1
9.1 APPROACH AND METHODS 9-1
9.1.1 Data Sources 9-2
-------
XI
CONTENTS (continued)
Volume II: Methods and Findings Page
9.1.2 Approach to Estimating Costs and Impacts of CKD Management
Alternatives 9-2
Case Study Plants 9-2
Methods for Estimating Facility Costs 9-3
Methods for Extrapolating from the Case Study Sample to the Industry . . 9-4
9.13 Cost Accounting Assumptions 9-5
9.1.4 Limitations of the Analysis 9-5
9.2 DESCRIPTIONS AND COSTS OF BASELINE AND ALTERNATIVE CKD
MANAGEMENT METHODS : . . 9-6
9.2.1 Current Practices 9-7
Direct Recycling of Collected Dust 9-8
Off-Site Beneficial Use 9-9
Current Land Disposal Practices 9-9
9.2.2 Alternative Land Disposal Practices 9-10
Conventional Subtitle C Technology and Administrative Standards 9-10
Subtitle C Costs (Exclusive of Corrective Action Costs) 9-11
Potential Subtitle C Corrective Action Costs 9-11
Alternative Subtitle C Costs Under RCRA §3004(x) 9-14
Tailored Contaminant Release Controls 9-15
9.23 Alternative On-Site CKD Recycling and Recovery Techniques 9-19
Increasing Direct CKD Recycling 9-19
Innovative CKD Recovery Technologies 9-20
Alkali Leaching 9-22
Fluid Bed Dust Recovery 9-26
The Passamaquoddy Technology Flue Gas Desulfurization
Process : 9-28
9.2.4 Other Operating Practices 9-31
Revised Standards for Cement Products 9-31
Curtailing Use of Hazardous Waste Fuels 9-32
9.2.5 Summary of the Costs of Alternative CKD Management Methods 9-34
9.3 POTENTIAL IMPACTS OF ALTERNATIVE MANAGEMENT PRACTICES . . . 9-37
-------
XII
CONTENTS (continued)
Volume II: Methods and Findings Page
9.3.1 Individual Plant-Level Impacts 9-37
932 Nationwide Cement Industry Impacts 9-40
9.3.3 Conclusions and Relationships to Regulatory Requirements 9-43
CHAPTER TEN: STUDY FINDINGS AND REGULATORY OPTIONS 10-1
10.1 Study Findings 10-1
10.1.1 Sources and Volumes of Waste (Study Factor 1) 10-1
10.12 Waste Management Practices (Study Factors 2 and 8) 10-2
10.1.3 Waste Characteristics and Potential Risks to Human Health and the
Environment (Study Factor 3) 10-2
10.1.4 Documented Evidence of Damage (Study Factor 4) 10-4
10.1.5 Potential Costs and Impacts of Subtitle C Regulation (Study Factors 5,
6, and 7) 10-4
10.2 Environmental Justice 10-5
10.3 Recommendations 10-5
10J.1 Decision Rationale and Options 10-5
10.3.2 Regulatory Options 10-7
10.3.3 Next Steps 10-11
GLOSSARY
BIBLIOGRAPHY
-------
LIST OF EXHIBITS
REPORT TO CONGRESS ON CEMENT KILN DUST
Volume II: Methods and Findings
Page
Exhibit 2-1 General Cement Types and Uses 2-2
Exhibit 2-2 1990 Clinker Production by State 2-6
Exhibit 2-3 Geographic Distribution of Active Portland Cement Production
Facilities in 1991 in the U.S 2-7
Exhibit 2-4 U.S. Clinker Capacity for the Period 1973-1991 . . 2-9
Exhibit 2-5 Steps in the Manufacture of Portland Cement 2-11
Exhibit 2-6 Material Temperature Ranges in the Kiln Portion of the Cement
Manufacturing Process 2-14
Exhibit 2-7 Typical Energy Requirements for Each Kiln Type 2-18
Exhibit 2-8 Typical Process Flow Diagram for a Wet Process Cement Kiln 2-20
Exhibit 2-9 Typical Process Flow Diagram for a Suspension Flash Preheater
Cement Kiln . ; 2-22
Exhibit 2-10 Typical Process Flow Diagram for a Precalciner Coal and Waste
Fired Cement Kiln 2-24
Exhibit 2-11 Age and Capacity of Existing Kilns 2-25
Exhibit 2-12 Feed Mixtures in 1990 2-28
Exhibit 2-13 Feed Mixtures in 1985 2-29
Exhibit 2-14 Raw Material Consumption in 1990 and 1985 2-30
Exhibit 2-15 Typical Composition of Raw Materials 2-31
Exhibit 2-16 Typical Feed Mixtures in 1990 2-32
Exhibit 2-17 Cement Kiln Fuel Consumption in 1990 and 1985 2-34
Exhibit 2-18 Cement Kiln Fuel Mixtures in 1990 2-35
Exhibit 2-19 Cement Kiln Fuel Mixtures in 1985 2-36
Exhibit 2-20 Average Coal Composition by Rank 2-38
Exhibit 2-21 Coal Producing Regions of the United States and Coal Types
Produced 2-38
-------
XIV
LIST OF EXHIBITS (continued)
Volume II: Methods and Findings
Page
Exhibit 2-22 Nominal Fossil Fuel Prices in 1985 and 1990 2-39
Exhibit 2-23 Breakdown of Solid Waste Fuel Consumption 2-40
Exhibit 2-24 Breakdown of Hazardous Waste Fuel Consumption 2-41
Exhibit 3-1 Flow Chart of Gross CKD Management Pathways 3-2
Exhibit 3-2 Air Pollution Control Devices Used at Cement Kilns 3-4
Exhibit 3-3
Exhibit 3-4
Exhibit 3-5
Exhibit 3-6
Exhibit 3-7
Exhibit 3-8
Exhibit 3-9
Exhibit 3-10
Exhibit 3-11
Exhibit 3-12
Exhibit 3-13
Exhibit 3-14
Exhibit 3-15
Exhibit 3-16
Exhibit 3-17
Exhibit 3-18
Schematic Diagrams of Common Types of Air Pollution Control
Devices
3-5
1990 Gross CKD Collection by Different Types of Air Pollution
Control Devices 3-7
Relationship Between Net and Gross CKD Generated in 1990 3-8
Gross and Net CKD Generated (1990) 3-10
Share of Net CKD Generated and Clinker Production Capacity (1990) .. 3-11
Facilities With High Net CKD Generation Relative to Clinker
Capacity 3-12
Percentages of Gross CKD Recycled, Sold, and Wasted (1990) 3-13
Recycling Rates Among Facilities That Operate Dry Kilns 3-14
Recycling Rates Among Facilities That Operate Wet Kilns 3-15
Average CKD Generation Rates Per Ton of Product (1990) 3-17
Fate of CKD as a Percent of Gross CKD (1990) 3-23
Particle Size Distribution of CKD by Process Type 3-26
Particle Size Distribution of CKD Midwest Portland Cement
Company, Zanesville, Ohio 3-27
Hydraulic Conductivity of Freshly Generated and Managed CKD 3-29
Typical CKD Bulk Constituents 3-31
Trace Metal Concentrations in As Generated CKD 3-34
-------
XV
LIST OF EXHIBITS (continued)
Volume II: Methods and Findings
Page
Exhibit 3-19 Trace Metal Concentrations in As Managed CKD 3-36
Exhibit 3-20 Trace Elements Commonly Found in Native Soils 3-37
Exhibit 3-21 Total Concentrations of Dioxins and Dibenzofurans in As Generated
CKD 3-39
Exhibit 3-22 Total Concentrations of Dioxins and Dibenzofurans in As Managed
CKD 3-40
Exhibit 3-23 Summary of Dioxin and Dibenzofuran Concentrations in CKD 3-42
Exhibit 3-24 Summary of Combined 1992-1993 Dioxin/Furan Sampling Results CKD
2,3,7,8, TCDD Toxicity Equivalence 3-43
Exhibit 3-25 Comparison of Maximum and Average Metals Concentrations in As
Generated Dust with TC Standards 3-48
Exhibit 3-26 Comparison of Maximum and Average Metals Concentrations in As
Managed Dust with TC Standards 3-49
Exhibit 3-27 TCLP Concentrations of Dioxins and Dibenzofurans in As Generated
CKD 3-51
Exhibit 3-28 T-test Comparison of Fuel Burning Effects on Metals Concentrations .. 3-56
Exhibit 3-29 T-test Comparison of Kiln Type on Metals Concentrations 3-58
Exhibit 3-30 Analytical Results of Clinker Analyses for Inorganics By Fuel Type 3-64
Exhibit 4-1 1991 CKD Waste Management Unit Dimensions 4-3
Exhibit 4-2 Remaining Life of Waste Management Units 4-4
Exhibit 4-3 Typical Cement Plant Layout 4-5
Exhibit 4-4 Environmental Protection Practices at CKD Waste Management Units
Active in 1990, by Kiln Fuel Use Type 4-8
Exhibit 4-5 Estimated Off-Site Uses for CKD Sold/Given Away 4-11
Exhibit 5-1 Summary of Cases of Documented and Potential Damage to Human
Health and/or Environmental 5-4
Exhibit 5-2
Summary of Documented Water Damages 5-6
-------
XVI
LIST OF EXHIBITS (continued)
Volume II: Methods and Findings
Page
Exhibit 5-3 Cases of Potential Damage 5-8
Exhibit 5-4 Site Diagram - Holnam Incorporated, Mason City, Iowa 5-10
Exhibit 5-5 Site Diagram - Lehigh Portland Cement Company, Leeds, Alabama .... 5-15
Exhibit 5-6 Site Diagram - Lehigh Portland Cement Company, Mason City, Iowa ... 5-18
Exhibit 5-7 Site Diagram - Lime Creek Nature Center (Lehigh Portland Cement),
Mason City, Iowa 5-19
Exhibit 5-8 Site Diagram - Portland Cement Company, Salt Lake City, Utah 5-24
Exhibit 5-9 Site Diagram - Southwestern Portland Cement, Fairborn, Ohio 5-28
Exhibit 5-10 Summary of Exceedances of State Metals Limits Southwestern Portland
Cement - Fairborn, Ohio Landfill #6 5-29
Exhibit 5-11 Site Diagram - National Gypsum Co./LaFarge Corp., Alpena, Michigan . 5-33
Exhibit 5-12 Summary of Exceedances of State Metals Limits National Gypsum
Co./Lafarge Corp., Alpena, Michigan 5-34
Exhibit 5-13 Site Diagram - Ash Grove Cement West, Montana City, Montana 5-36
Exhibit 5-14 Site Diagram - Texas Industries, Inc., Midlothian, Texas 5-38
Exhibit 5-15 Site Diagram - Holnam, Inc., Artesia, Mississippi 5-42
Exhibit 5-16 Site Diagram - Markey Machinery Property, Seattle, Washington 5-43
Exhibit 5-17 Summary of Air Damage Case Findings 5-46
Exhibit 5-18 Example of CKD Disposal Adjacent to an Agricultural Field 5-50
Exhibit 6-1 Overview of Risk Assessment Methodology 6-3
Exhibit 6-2 Basis for Risk-Screening Criteria 6-9
Exhibit 6-3 CKD Constituents That Exceeded Risk-Screening Criteria at EPA Sample
Facilities 6-12
Exhibit 6-4 Comparison of 15 Sample Facilities to Other Facilities 6-16
Exhibit 6-5 Site Specific Factors Used to Evaluate Risk Potential of On-Site CKD
Management 6-18
-------
XV11
LIST OF EXHIBITS (continued)
Volume II: Methods and Findings
Page
Exhibit 6-6 Risk Potential Rankings for the Ground-water Pathway 6-21
Exhibit 6-7 Risk Potential Rankings for the Surface Water Pathway 6-24
Exhibit 6-8 Particle Size Distribution of CKD by Kiln Type 6-27
Exhibit 6-9 Risk Potential Rankings for the Air Pathway 6-28
Exhibit 6-10 Graphical Illustration of On-site Risk Modeling Scenarios 6-31
Exhibit 6-11 Aquatic Ecological Benchmark Levels 6-38
Exhibit 6-12 Baseline On-Site Management Cancer Risks for Direct Exposure Pathways
for 15 Case Study Facilities 6-41
Exhibit 6-13 Baseline On-Site Management Cancer Risks for Foodchain Exposure
Pathways for 15 Case Study Facilities 6-42
Exhibit 6-14 Constituents Contributing to Adverse Health Effects In On-site CKD Risk
Modeling Analysis 6-43
Exhibit 6-15 Results of Central Tendency and High End Ecological Effects Analysis .. 6-45
Exhibit 6-16 Sensitivity Analysis of Maximum CDD/CDF Cancer Risks for Direct
Exposure Pathways 6-46
Exhibit 6-17 Sensitivity Analysis of Maximum CDD/CDF Cancer Risks for Foodchain
Exposure Pathways 6-47
Exhibit 6-18 Sensitivity Analysis of Location Adjacent to Agricultural Field for Foodchain
Exposure Pathways : 6-48
Exhibit 6-19 Sensitivity Analysis of Location Adjacent to Surface Water for Direct and
Foodchain Exposure Pathways 6-49
Exhibit 6-20 Sensitivity Analysis of Highly Exposed Individuals for Foodchain Exposure
Pathways 6-50
Exhibit 6-21 Off-site Beneficial Uses Examined in the Risk Assessment 6-54
Exhibit 7-1 New Source Performance Standards for Portland Cement Plants 7-7
Exhibit 7-2 Summary of State Air Pollution Controls 7-10
Exhibit 7-3 Summary of State Water Pollution Controls 7-21
-------
XVI11
LIST OF EXHIBITS (continued)
Volume II: Methods and Findings
Page
Exhibit 7-4 Summary of State Solid Waste Management Controls 7-33
Exhibit 8-1 Flow Chart of Gross CKD Management Pathways 8-1
Exhibit 8-2 Process Flow Diagram of Recovery Scrubber 8-10
Exhibit 8-3 Process Flow Diagram of Fluid Bed Dust Recovery Process 8-15
Exhibit 8-4 Summary of Alternatives for Minimization of CKD Removal from the
Kiln System 8-39
Exhibit 8-5 Summary of Alternatives for Beneficial Utilization of CKD Removed
from the Kiln System 8-41
Exhibit 9-1 Facilities Included in the Cost Analysis 9-3
Exhibit 9-2 Subtitle C Disposal Costs 9-12
Exhibit 9-3 Subtitle C-Minus Disposal Costs 9-15
Exhibit 9-4 Tailored Contaminant Release Controls Costs: Continued Sales of CKD for
Off-Site Use 9-17
Exhibit 9-5 Tailored Contaminant Release Controls Costs: Curtailed Sales of CKD for
Off-Site Use 9-18
Exhibit 9-6 Key Design Conditions for the Nine Case Study Cement Plants 9-23
Exhibit 9-7 Estimated Incremental Net Costs for the Alkali Leaching System 9-25
Exhibit 9-8 Estimated Incremental Net Costs for the Fuller Fluidized Bed System . . 9-27
Exhibit 9-9 Estimated Incremental Net Costs for the Passamaquoddy Technology
Recovery Scrubbing Process 9-30
Exhibit 9-10 Economic Benefits from Burning Hazardous Waste Fuels 9-33
Exhibit 9-11 Total Incremental Annualized Costs of CKD Management Alternatives for
EPA Case Study Cement Plants 9-35
Exhibit 9-12 Annualized Incremental Costs of CKD Management Alternatives per Metric
Ton of CKD for EPA Case Study Cement Plants 9-36
Exhibit 9-13 Annualized Incremental Costs of CKD Management Alternatives per Metric
Ton of Cement for EPA Case Study Cement Plants 9-38
-------
XIX
LIST OF EXHIBITS (continued)
Volume II: Methods and Findings Page
Exhibit 9-14 Capital Investment Requirements for Implementing CKD Management
Alternatives for EPA Case Study Cement Plants 9-39
Exhibit 9-15 Incremental Costs of Alternative CKD Management Practices Relative to
Value of Cement Sales (Incremental cost per ton of cement/revenue per ton
of cement — cents per dollar of sales) 9-41
Exhibit 9-16 Industry Wide Costs of Alternative Management Practices (Cost in
$ Million) 9-42
-------
LIST OF ACRONYMS
ADEM
APCD
ARARs
ARE
ASTM
AWQC
BACT
BDAT
BIFs
BMPs
BOM
CDDs
CDFs
CELDS
CERCLA
CKD
CWA
DAF
DOE
DOT
DRE
ESPs
FGD
FHA
FIT
FML
GAC
GEMS
HBLs
HOPE
HSWA
1C
LCNC
LCS
LDR
LOAELs
LPCC
MCLs
MDNR
MEI
MERA
NAAQS
NESHAPs
NOV
NPDES
NPL
NSPS
OEPA
PCA
Alabama Department of Environmental Management
air pollution control device(s)
applicable and relevant or appropriate requirements
Air Resources Board
American Society for Testing and Materials
ambient water quality criteria
Best Available Control Technology
Best Demonstrated Available Technology
boilers or industrial furnaces
best management practices
U.S. Bureau of Mines
Chlorinated Dibenzo-p-dioxins
Chlorinated Dibenzofurans
Computer-Aided Environmental Legislative Data System
Comprehensive Environmental Response, Compensation, and Liability Act
cement kiln dust
Clean Water Act
dilution and attenuation factor
U.S. Department of Energy
Department of Transportation
destruction and removal efficiency
electrostatic precipitators
flue gas desulfurization
Federal Highway Authority
Field Investigation Team
flexible membrane liner
granular activated carbon
Graphical Exposure Modeling System
health-based levels
high density polyethylene
Hazardous and Solid Waste Amendments of 1984
Information Circular
Lime Creek Nature Center
leachate collection systems
Land Disposal Restrictions
lowest observed adverse effect levels
Lehigh Portland Cement Company
Maximum Contaminant Levels
Michigan Department of Natural Resources
maximum exposed individual
Michigan Environmental Response Act
National Ambient Air Quality Standards
National Emission Standards for Hazardous Air Pollutants
Notice of Violation
National Pollutant Discharge Elimination System
National Priorities List
New Source Performance Standards
Ohio Environmental Protection Agency
Portland Cement Association
-------
LIST OF ACRONYMS (continued)
PCBs - polychlorinated biphenyls
PCS - EPA's Permit Compliance System
PCU - Portland Cement Company of Utah
PICs - products of incomplete combustion
POTWs - publicly owned treatment works
ppm - parts per million
PSD - Prevention of Significant Deterioration
PVC - polyvinyl chloride
RCRA - Resource Conservation and Recovery Act
SCS - Soil Conservation Service
SDWA - Safe Drinking Water Act
SIP - state implementation plan
SPLP - Synthetic Precipitation Leaching Procedure
SWMUs - solid waste management units
TC - Toxicity Characteristic
TCLP - Toxicity Characteristic Leaching Procedure
TDS - total dissolved solids
TEF - toxicity equivalent factor
TSDF - treatment, storage, or disposal facilities
TSP - total suspended paniculate
TSS - total suspended solids
TWC - Texas Water Commission
UBK - uptake/biokinetic
UDEQ - Utah Department of Environmental Quality
USGS - U.S. Geological Survey
VOCs - volatile organic compounds
WMUs - waste management units
WWTPs - wastewater treatment plants
-------
CHAPTER ONE
INTRODUCTION
Section 3001(b)(3)(A)(iii) of the Resource Conservation and Recovery Act (RCRA)
excludes cement kiln dust waste from regulation under Subtitle C of RCRA, pending completion
of a Report to Congress required by §8002(o) and a determination by the EPA Administrator
either to promulgate regulations under Subtitle C or that such regulations are unwarranted (as
required by §3001(b)(3)(C)). This report has been prepared by EPA to meet the requirements
of §3001 (b)(3) and §8002(o) that the Agency study cement kiln dust (CKD) waste generated in
the production of cement clinker and prepare a Report to Congress on the findings of the study.
This introductory chapter is organized into four sections. The first section discusses the
purpose and scope of this report, while section two presents EPA's general study methods and
the major sources of information used in preparing this document. Section three describes the
Agency's decision making rationale that will be used in making the final regulatory
determination. Finally, section four provides an overview of the content and organization of this
report.
1.0 PURPOSE AND SCOPE
On October 21,1976, Congress enacted RCRA (Pub. L. 94-580). Section 3001 of RCRA
mandated that the EPA Administrator "promulgate regulations identifying characteristics of
hazardous waste, and listing particular hazardous wastes which shall be subject to the provisions
of this subtitle." Section 3004 required the Administrator to promulgate standards applicable to
owners and operators of hazardous waste treatment, storage, and disposal facilities.
In response to these requirements, EPA proposed regulations for managing hazardous
wastes under Subtitle C of RCRA on December 18,1978 (43 FR 58946). In this regulatory
proposal, EPA proposed to defer most of the RCRA Subtitle C requirements for six categories
of wastes, which it termed "special wastes," until information could be gathered and assessed and
the most appropriate regulatory approach determined. Special wastes are typically generated in
large volumes, are thought to pose less risk to human health and the environment than wastes
(to be) regulated as hazardous wastes, and may be inappropriately regulated under the proposed
technical requirements implementing Subtitle C. EPA identified CKD waste as one of these
"special wastes".1
In 1979, Congress began work on reauthorization of RCRA. During the reauthorization
process, Rep." Thomas Bevill (Alabama) offered an amendment (now frequently referred to as
the Bevill Amendment) which, among other things, modified §3001 to temporarily exempt
"cement kiln dust waste" (along with two other categories of waste) from Subtitle C regulation,
pending completion of certain studies. On October 12, 1980, Congress enacted the Solid Waste
Disposal Act Amendments of 1980 (Pub. L. 96-482), which added §3001(b)(3)(A)(i-iii) (the
Bevill Amendment) to RCRA. These amendments also added §8002(o), which required the
1 The other five proposed "special wastes" specifically identified in the 1978 proposed rule were mining waste; utility
waste; phosphate rock mining, beneficiation, and processing waste; uranium mining waste; and oil and gas drilling muds
and oil production brines.
-------
1-2
Administrator to study the adverse effects on human health and the environment, if any, from
the disposal of "cement kiln dust waste," and submit a Report to Congress on its findings. The
1980 amendments also added §3001(b)(3)(C), which requires the Administrator to make a
regulatory determination, within six months of the completion of the §8002(o) study, whether or
not to regulate CKD waste under Subtitle C of RCRA.
In response to the 1980 RCRA amendments, on November 19,1980, EPA published an
interim final amendment to its hazardous waste regulations to reflect the provisions of the Bevill
Amendment (45 FR 76618), which is codified at 40 CFR 261.4(b)(8). Consequently, since that
time, CKD has been exempt from Subtitle C of RCRA -- that is, this material has never been
regulated as a hazardous waste under federal law.2
The purpose of this report is to comply with the Congressional edict and to establish the
factual basis for EPA decision making regarding the appropriate regulatory status, under RCRA,
of CKD waste. In keeping with the statutory requirements, this report addresses the following
eight study factors, as articulated at §8002(o) of RCRA:
(1) The source and volumes of [CKD] generated per year;
(2) Present disposal practices;
(3) Potential danger, if any, to human health and the environment
from the disposal of [CKD];
(4) Documented cases in which danger to human health or the
environment has been proved;
(5) Alternatives to current disposal methods;
(6) The costs of such alternatives;
(7) The impact of those alternatives on the use of natural resources;
and
(8) The current and potential utilization of [CKD].
In addition, the report includes a review of applicable state and federal regulations, so
regulatory decisions that derive from the report will avoid duplication of existing requirements.
1.1 GENERAL METHODS AND INFORMATION SOURCES
In preparing this report, EPA has developed industry-wide, and in some cases, facility-
specific data and analytical methods that reflect the complexity of the issues that are addressed in
this report. The facilities that generate CKD waste vary considerably in size, location,
operational aspects, and waste management techniques. Moreover, to examine in detail the
2 It should be noted here that under the RCRA Subtitle C Boilers and Industrial Furnaces (BIF) Rule, CKD generated
by kilns that bum hazardous waste as fuel may be ineligible for Bevill Exclusion under certain conditions (see 40 CFR
266.112).
-------
1-3
broad array of study factors mandated by RCRA §8002(o), EPA had to develop approaches and
methods that were sufficiently sophisticated to take into account the special nature of CKD.
This section outlines the general methods and the data sources that the Agency employed to
respond to the statutory study factors, beginning with a discussion of the major data collection
initiatives that EPA's Office of Solid Waste (OSW) has conducted during the past three years.
1.1.1 General Analytical Methods
To address the RCRA §8002(o) study factors in a thorough and systematic fashion, EPA
has organized this report and conducted its supporting analyses along functional lines. The
Agency has combined its examination of certain study factors into groups and has presented its
analysis of the others in a logical sequence. For that reason, certain key concepts are addressed
in several chapters of the Report, albeit with different emphases. Examples of such cross-cutting
issues include CKD composition, cement kiln design and operation, and kiln fuel type.
The specific methods that EPA used to address each of the study factors are described in
detail in Chapters 3 through 9. Additional information on the methods used and supporting data
are contained in the Background Documents to this Report.
1.1.2 EPA Data Collection Activities
To develop an adequate information base to address the eight study factors, OSW
conducted a number of data collection activities. The focus of most of these efforts was site-
specific. As a result, EPA was able to compile reasonably detailed industry-wide information,
which was used extensively to prepare this report. The major information-gathering initiatives
are identified and discussed in the following paragraphs.
A. 1991 Portland Cement Association (PCA) Survey of Cement Kiln Facilities
In December 1991, the Portland Cement Association (PCA) prepared and distributed a
written questionnaire to the operators of the 115 cement kiln facilities in the U.S. that EPA
believed generated CKD waste. These facilities were identified from information in existing
Agency files, information provided by PCA, and from data supplied by the U.S. Bureau of Mines
(BOM). The questionnaire was designed to elicit information both on CKD waste generation
and management at clinker-producing facilities, and on the operational characteristics of the
facilities. The majority of the questions addressed waste management and were ordered so as to
"track" CKD from the point of generation through ultimate disposition.
Approximately 80 percent of the facilities that currently generate CKD responded to the
questionnaire. PCA has made these survey responses available to the Agency, and hence, to the
public record. Responses were entered into a computerized data base, which EPA used in
conducting the analyses described below. A description of the survey is contained in the
Background Documents to this Report. Copies of the survey instrument, as well as all available
non-confidential company responses, may be found in the supporting docket for this report.
B. 1992 and 1993 CKD Sampling and Analysis
Because CKD has not been studied by OSW previously, and because existing facility-
specific data on this waste were sparse, EPA conducted a CKD and cement clinker sampling and
-------
1-4
analysis program in early 1992. The Agency's field sampling teams visited 15 cement facilities,
recorded observations of operational practices, photographed waste management activities, and
collected samples. In most cases, EPA sampled both "as generated" and "as managed" CKD.
To clarify certain analytical issues raised by the results of the 1992 sampling and analysis
effort (referred to throughout this Report as "Phase I"), EPA conducted a second, more focused
CKD sampling and analysis program in May 1993 (referred to as Phase II). The Agency visited
and took CKD (and in some cases, clinker) samples from six cement plants (one of which had
been visited during Phase I), and performed various chemical analyses. One important
distinction between the two sampling and analysis programs is that the analytical methods
employed for measuring dioxin and dibenzofuran concentrations in Phase II analysis were far
more sensitive than those used in Phase I.
The data developed in EPA's two-phase CKD sampling program is summarized in the
supporting docket for this report. A description of EPA's waste sampling study is presented in a
Technical Background Document, which may also be found in the docket.
C. Damage Case Collection
To respond to the need to describe "documented cases in which danger to human health
or the environment has been proved," (referred to in this report as "damage cases") as directed
by RCRA, EPA conducted an exhaustive examination of the extent to which CKD has been
implicated in human health or environmental contamination incidents. This effort began by
contacting appropriate staff in all EPA regions and states in which one or more cement kiln
facilities is located. When available, the information was obtained through the mail or through
visits to state/local officials having regulatory jurisdiction over CKD management.
The Agency's damage case analysis is based primarily on documented evidence, rather
than on visits to the sites being evaluated. However, the 15 Phase I waste sampling visits
included an effort to collect information on the existence of potential environmental pathways
through which CKD and its constituents might migrate and cause adverse impacts. The result of
this effort is a compilation of information regarding the past and present management practices
that have been applied to CKD, and the environmental or human health consequences of these
practices.
Damage case findings are presented in Chapter 5 of this report; the individual sites that
have been evaluated are listed in a supporting Technical Background Document, which also
provides more extensive discussions and supporting evidence.
D. EPA Site Visits
In addition to the waste sampling and damage case collection efforts described above,
EPA visited two cement facilities during the summer of 1991 to enhance the Agency's general
understanding of CKD generation and management practices. The knowledge and insights
gained during these and subsequent sampling visits have enabled the Agency to understand and
evaluate current waste management practices, and to draw conclusions and make
recommendations regarding the appropriate regulatory status of CKD.
-------
1-5
E. RCRA §3007 Waste and Site Characteristics Data Request
To augment existing EPA waste and site characterization data and to allow affected
facilities to have meaningful input into the Agency's evaluation of the physical and chemical
characteristics of CKD, EPA issued a formal written request, under authority of RCRA §3007, to
cement plant operators. The purpose of the request was to obtain any available information on
the characteristics of the dust that they generate. In particular, the Agency sought new
information on the presence and concentrations of organic constituents in CKD, and patterns of
off-site CKD use in productive applications, as well as site-specific environmental characteristics
of those plants for which PCA surveys had not been received. The request did not specify the
quantity of data required by EPA or a data format, so as to make compliance by the facility
operators as simple as possible.
EPA reviewed all data submittals and collected and used the data that are most relevant
to the analyses presented in this study.
1.2 EPA'S DECISION MAKING RATIONALE
Based on the analysis of the study factors found at §8002(o), EPA has arrived at
preliminary findings that are relevant to the appropriate regulatory status of CKD under RCRA.
These findings suggest two general EPA options that were developed through the systematic
evaluation process described below. In this process, the Agency considered the study factors in a
step-wise fashion. This methodology is consistent with previous Bevill Amendment decisions,
such as those made for mineral processing. In applying this framework, EPA used a number of
assumptions, which are described in the following paragraph.
The first assumption that the Agency made is that decision criteria were needed so that
reasonable decisions regarding the need for additional regulatory controls can be achieved. The
second major assumption guiding EPA's decision-making process was that the study factors that
are most important in establishing the regulatory status of CKD are 1) the risks posed and
documented damages caused by the dust, and 2) the costs and impacts that would be associated
with more stringent regulatory controls, if such additional controls were warranted. The reason
for this is that in the absence of potential risk and/or documented damages, there is no need for
hazardous waste regulation under RCRA Subtitle C (the key issue in question); if greater
regulatory controls are needed because of significant potential or documented danger, the costs
and impacts of regulatory controls are the critical factor in determining whether a given
alternative would lead to the desired outcome (adequate protection of human health and the
environment, and continued operation of the affected facilities). EPA also believes that it has
developed and analyzed regulatory compliance scenarios that are realistic from an operational
and engineering standpoint, and that are likely to be adequately protective of human health and
the environment (i.e., could be implemented by facility operators and would result in societal
benefits).
-------
1-6
Evaluation Criteria
Step 1. Does management of CKD pose human health/environmental problems? Might current
practices cause problems in the future?
Critical to the Agency's decision-making process is whether CKD either has caused or
could cause human health or environmental damage. To resolve this issue, EPA posed the
following key questions:
(1) Has CKD, as currently managed, caused documented human health impacts or
environmental damage?
(2) Does EPA's analysis indicate that CKD could pose significant risk to human
health or the environment at any of the sites that generate it (or in off-site use),
under either current management practices or plausible mismanagement
scenarios?
(3) Does CKD exhibit any of the characteristics of hazardous waste?
If the answer to any of these three questions was yes, then EPA would conclude that further
evaluation was necessary. If the answer to all of these questions was no, then the Agency would
conclude that regulation of CKD under RCRA Subtitle C is unwarranted.
Step 2. Is more stringent regulation necessary and desirable?
If CKD has caused or may cause human health or environmental impacts, then EPA
would conclude that an examination of alternative regulatory controls was appropriate. Given
the context and purpose of the present study, the Agency focused on an evaluation of the
likelihood that such impacts might continue or arise in the absence of Subtitle C regulation, by
posing the following two questions:
(1) Are current practices adequate to limit contaminant release and associated risk?
(2) Are current federal and state regulatory controls adequate to address the
management of CKD?
If current practices and existing regulatory controls are adequate, and if the potential for actual
future impacts is low (e.g., facilities in remote locations), then the Agency would tentatively
conclude that regulation of CKD under Subtitle C is unwarranted. Otherwise,, further
examination of regulatory alternatives was necessary.
Step 3. What would the operational and economic consequences be of a decision to regulate
CKD under Subtitle C?
If, based upon the previous two steps, EPA believed that regulation of CKD under
Subtitle C might be appropriate, then the Agency would evaluate the costs and impacts of two
potential regulatory options. The focus of this inquiry was whether the magnitude and
distribution of regulatory compliance costs might jeopardize the continued economic viability of
-------
1-7
one or more generators if the waste were to be regulated under the Subtitle C scenario. The key
questions in the Agency's decision-making process were as follows:
(1) Are predicted economic impacts associated with the Subtitle C scenario significant
for any of the affected facilities?
(2) To what extent could these compliance costs be avoided through the
implementation of alternative CKD management practices?
(3) In the event that significant impacts are predicted, might a substantial proportion
of domestic capacity or product consumption be affected?
(4) What effects would hazardous waste regulation have upon the viability of the
beneficial use or recycling of CKD?
In EPA's judgment, absence of significant impacts would suggest that Subtitle C regulation might
be appropriate for CKD if findings indicate that it poses significant risk. If even less stringent
Subtitle C standards impose widespread and significant impacts on facilities, and/or deter the safe
and beneficial use of the dust, EPA would conclude that regulation under some form of Subtitle
D program might be more appropriate.
1.3 CONTENTS AND ORGANIZATION OF REPORT
This Report to Congress consists of two volumes, as follows:
Volume I: Executive Summary
u This volume provides an overview of the methods and data sources used to
conduct the study, the technical findings of the study, a description of the
regulatory options considered, EPA's conclusions and preliminary
recommendations, and a discussion of the next steps the Agency plans to
undertake.
Volume II: Methods and Findings
ii Chapter 1. Introduction, summarizes the purpose and scope of the report, general
methods and information sources used, and EPA's decision-making rationale.
u Chapter 2. Cement Industry Overview, provides a description of CKD
waste, cement industry structure and characteristics, the cement
manufacturing process, the types of production processes used, and
significant process inputs.
u Chapter 3. CKD Generation and Characteristics, describes the generation
of CKD, dust collection devices and recycling of dust back to the kiln, and
the physical and chemical characteristics of CKD.
-------
1-8
U Chapter 4. Current CKD Management Practices, outlines the range of
CKD management practices currently employed at domestic cement
plants.
u Chapter 5. Documented Damages from Management of CKD, provides a
discussion of case studies in which it has been shown that currently used CKD
management practices have led to documented environmental impairment.
U Chapter 6. Potential Danger to Human Health and the Environment,
presents a discussion of EPA's risk assessment in which the Agency
examined inherent hazards posed by CKD, evaluated site-specific risk
factors, and performed quantitative transport, fate, and exposure modeling.
il Chapter 7, Existing Regulatory Controls on CKD Management, reviews
applicable federal and state laws and regulations controlling CKD
management with respect to the various environmental media.
u Chapter 8. Alternative Management Practices and CKD Utilization,
investigates a variety of alternative management practices for CKD and
beneficial uses of CKD removed from the production system, and
examines the technical feasibility, human health/environmental
considerations, and economic feasibility of each option.
u Chapter 9. Cost and Economic Impacts of Alternatives to Current CKD
Disposal Practices, discusses methods and data sources used, cost
modeling results, and analysis of the economic impacts under each of
several regulatory and operational scenarios.
u Chapter 10. Study Findings and Regulatory Options.
Additional documentation regarding the methods, data sources, and assumptions used in
preparing this report and the analyses contained herein may be found in the RCRA docket
(docket number F-93-RCKA-FFFFF).
-------
1-9
CHAPTER ONE
INTRODUCTION
1.0 PURPOSE AND SCOPE 1
1.1 GENERAL METHODS AND INFORMATION SOURCES 2
1.1.1 General Analytical Methods 3
1.1.2 EPA Data Collection Activities 3
A. 1991 Portland Cement Association (PCA) Survey of Cement Kiln
Facilities 3
B. 1992 and 1993 CKD Sampling and Analysis 3
C. Damage Case Collection ; 4
D. EPA Site Visits 4
E. RCRA §3007 Waste and Site Characteristics Data Request 5
1.2 EPA'S DECISION MAKING RATIONALE 5
1.3 CONTENTS AND ORGANIZATION OF REPORT 7
-------
CHAPTER TWO
CEMENT INDUSTRY OVERVIEW
2.0 INTRODUCTION AND DESCRIPTION OF CKD
CKD is a fine-grained solid material generated as the primary by-product of the
production of cement. Cement production occurs in very large rotary kilns at high temperatures;
finely ground raw material enters and rolls downward from the "cool end" of the kiln, while fuels
and combustion air are introduced and drawn upward from the "hot end." This air that is drawn
into and through the kiln carries with it some of the finely ground solid raw materials, condensed
fuel components, and partially reacted feed. As this air exits the cool end, the entrained solid
matter is collected before the air is vented to the atmosphere, through the large gas emission
"smokestacks" that are found at all cement production facilities. CKD generation results directly
from this control of particulate matter that would otherwise be discharged. In contrast to many
other residues of industrial production, CKD is essentially an "off-specification" product; it much
more closely resembles the raw materials entering and product leaving the operation than many
other industrial solid wastes. The effective control of stack emissions at cement plants has
occurred only during the past 30 years or so. Therefore, the generation of CKD as a "solid
waste" in significant quantities is a relatively recent phenomenon. Nonetheless, existing
stockpiles of this material are quite large at some facilities, and substantial quantities of
additional CKD are generated (though not necessarily accumulated) on a continuous basis at all
cement plants.
All cement kilns generate CKD, and the quantities and characteristics of the CKD
generated depend upon a number of operational factors and characteristics of the inputs to the
manufacturing process. A critical examination of CKD management and its impacts cannot be
conducted without an understanding of the industry and the basic process of manufacturing
cement. Accordingly, this section presents an overview of the cement industry and describes
cement types, the basic structure of the cement industry, manufacturing processes, and variations
in kiln design. A section on process inputs provides discussion of the characteristics of the raw
feed and the fuels used in the manufacturing process. This overview should assist the reader in
understanding some of the issues surrounding the chemical characteristics of CKD that are
discussed in considerable detail in subsequent chapters of this report. This chapter concludes
with a brief summary and "road map" to the remainder of this document.
2.1 CEMENT INDUSTRY STRUCTURE AND CHARACTERISTICS
2.1.1 What Is Cement?
For purposes of this Report to Congress, cement refers to the commodities that are
produced by burning mixtures of limestone and other minerals or additives at high temperature
in a rotary kiln, followed by cooling, finish mixing, and grinding. This is the manner in which the
vast majority of commercially-important cementitious materials are produced in the United
States. Cements are used to chemically bind different materials together. The most commonly
produced cement type is "Portland" cement, though other standard cement types are also
produced on a limited basis. The major cement types and their applications are presented in
-------
2-2
Exhibit 2-1. Portland cement is a hydraulic cement, meaning that it sets and hardens by chemical
interaction with water and is capable of doing so under
-------
2-3
Exhibit 2-1
General Cement Types and Uses
Cement Name
Normal Portland Cement
Modified Portland
Cement
High -Early-Strength
Portland Cement
Low-Heat Portland
Cement
Sulfate-Resisting
Portland Cement
Air-Entraining Portland
Cements
Portland-Blast Furnace
Slag Cements
White Portland Cement
Portl and -Pozzol an
Cement
•type
I
n
m
IV
V
1A
IIA
IIIA
IS
IS-A
MH
MS
N/A
IP
IP-A
Characteristics
Non-specialty hydraulic cement
Generates less heat from its
hydration and is more resistant to
sulfate attacks than type I.
Allows earlier removal of forms and
shorter periods of curing.
Generates less heat during
hydration than type II. Gains
strength more slowly than type I.
The tricalcium aluminum content is
limited to seven percent.
High sulfate resistance cement that
gains strength more slowly than
type I. The tricalcium aluminate
content is limited to a maximum of
five percent.
Air-entraining agents, interground
with the cement clinker, purposely
cause air, in minute, closely spaced
bubbles, to occur in concrete.
Made by grinding granulated high-
quality slag with Portland-cement
clinker. Type IS cements gain
strength more slowly in initial
stages, but ultimately have about
the same 28 -day strength as type I
cements.
Desirable aesthetic qualities. High
in alumina and contains less than
0.5 percent iron.
A blended cement made by
intergrinding Portland cement and
pozzolanic materials (slaked lime
and granulated blast-furnace slag or
other material similar to natural
lava), without burning.
Uses
Most structures, pavements, and
reservoirs.
Structures having large cross
sections, such as large abutments
and heavy retaining walls. Also in
drainage where a moderate sulfate
concentration exists.
When high strengths are required
within a few days.
Mass concrete constructions such
as large dams where high
temperature rises would create
special problems.
Special cement, not readily
available, to be used when
concrete is exposed to severe
sulfate attack.
Entrained air makes the concrete
more resistant to the effects of
repeated freezing and thawing and
of the deicing agents used on
pavements.
Air entrainment type is IS-A;
Moderate heat-of-hydration type is
MH; Moderate sulfate resistance
type is MS.
Architectural and ornamental
work.
Used under certain conditions for
concrete not exposed to the air.
Air entrainment type is IP-A.
Covered in ASTM specification
C3-40.
-------
Masonry Cement
C91
(ASTM)
2-4
Hydraulic cement made by
combining either natural or
Portland cements with fattening
materials such as hydrated lime
and, less frequently, with air-
entraining admixtures.
Used in place of job cement-lime
mixtures to reduce the number of
materials handled and to improve
the uniformity of the mortar.
Exhibit 2-1 (continued)
General Cement Types and Uses
Cement Name
Waterproofed Cements
Natural Cement
Type
N/A
N/A
Characteristics
Produced by grinding with certain
soaps and oils.
Manufactured from limestone
containing clay, with chemical
constituents similar to those of
Portland cement.
Uses
Sometimes used where a
waterproof or water-repellant
concrete or mortar is desirable.
The effectiveness is limited to
three or four feet of water
pressure.
Sometimes used as common
mortar for brick or stonework.
Note: N/A = not applicable
Source: Adapted from Baumeister and Marks, 1968. Standard Handbook for Mechanical Engineers.
water.1 When combined with sand, gravel, water, and other materials, Portland cement forms
concrete, the most widely used building material in the world. Portland cement comprises about
90 percent of all hydraulic cement produced, with masonry, lime, and natural cements
constituting the remainder. Cement produced and sold in the U.S. must meet specifications
established by the American Society for Testing and Materials (ASTM). Five types of Portland
cements are covered by ASTM specifications (number C150). Each type requires specific
additives or changes in the proportions of the raw material mix to make products for specific
applications.
Together, Portland and masonry cement consumption exceeded 69 million metric tons (76
million tons) in 1991,2 far surpassing the use of all other cement types combined. Seventy-three
percent of all U.S. cement is used by the ready mix concrete industry, while 12 percent is used by
concrete product producers (pipe, pre-cast, and prestressed), and 5 percent is used by highway
contractors. Other users include building materials dealers and other contractors. Most of the
Portland cement sold in the U.S. is shipped in bulk form, with less than 5 percent of the total
being shipped in bags.3
1 American Concrete Institute, 1990. Cement and Concrete Terminology, American Concrete Institute Committee
116, publication SP-19 (90), Detroit, Michigan.
2 Portland Cement Association, 1992a. U.S. Cement Industry Fact Sheet, 10th Edition, Market and Economic
Research, Skokie, Illinois. April 1991
1 U.S. Department of Commerce, 1991. Cement Industry, U.S. Industrial Outlook 1991, Washington, D.C.
-------
2-5
Portland cement concrete is utilized in a wide variety of construction applications,
including houses, offices, roads, sidewalks, playgrounds, and water supply systems. Use in
buildings accounts for approximately 60 percent of all cement consumption. Half of this is for
residential buildings, approximately 20 percent for commercial buildings, and the remainder for
public and farm buildings.4 Public infrastructure development and maintenance accounts for the
remaining 40 percent of cement demand.5 The majority of public infrastructure applications are
for streets and highways, but other important applications include water and waste systems, burial
vaults, and oil wells.6 As discussed in more detail below, cement producing facilities in the
United States generally serve local and regional markets.
Portland cement consists of a mixture of synthetic materials or compounds. The four
principal compounds are:
u Tricalcium silicate - (CaO)3SiO2;
u Dicalcium silicate - (CaO)2SiO2;
u Tricalcium aluminate - (CaO)3Al203); and
u Tetracalcium-aluminoferrite - (CaO)4(Al2O3)(Fe2O3).
Portland cement can be defined as "the product obtained by finely grinding clinker produced by
calcining to incipient fusion (i.e., sintering) an intimate and properly proportioned mixture of
argillaceous and calcareous materials."7 Roughly translated, cement is made by sintering a
mixture of materials containing lime, silica, alumina, and iron oxide. Typically, these materials
include limestone, sand, clay, iron ore, and/or other minerals and mineral processing residues.
Cement production involves heating these raw materials, usually limestone and clay, to
approximately 1,482°C (2,700°?) to form a material called "clinker," which is granular and highly
variable in size. Clinker is then cooled, and ground with a smaller amount (approximately five
percent) of gypsum to make cement. The heating process is performed in a rotary cement kiln,
which is a brick-lined cylinder, inclined slightly from the horizontal, that rotates on its
longitudinal axis at a slow and constant speed (generally a couple of revolutions per minute).
Raw materials are introduced at the higher end, while a fixed combustion source is operated at
the lower end. Thus, the raw materials and the heated air (which induces the chemical reactions
in the raw feed) travel countercurrent to one another. There are a number of different cement
kiln configurations that employ different technologies and equipment, and thus, there is
considerable variation in the size and operating characteristics of cement kilns. Nonetheless,
cement kilns are all quite large in comparison with most other types of industrial equipment;
4 ibid.
* Sanbom, S., 1992. Cement & Aggregates, Value Line Publishing. July 24, 1992.
6 U.S. Department of Commerce, 1991, op. at.
7 "Portland Cement", Kackman, A.H. and Brown, R.W., Section Editors; SME Mineral Processing Handbook,
Society of Mining Engineers of the American Institute of Mining, Metallurgical, and Petroleum Engineers, Inc., New
York, NY, 1985, Vol. 2, Section 26, pp. 26-1 to 26-52.
-------
2-6
typical cement kilns are approximately 152.4 meters (m) long (500 feet) and 3.7 m in diameter,8
and some are considerably larger. There are two primary types of kiln design: wet process kilns
that accept feed materials in slurry form, and dry process kilns that accept feeds in dry, ground
form. A more detailed description of the cement production process and its variations is
presented in Section 3.2, below.
2.1.2 The Cement Industry
Because cement is used in almost all construction activities, the cement industry is an
important part of the nation's economic and industrial base. Figures characterizing the cement
industry, such as production, capacity, and consumption statistics, change as a result of
fluctuations in domestic construction activities, plant modernizations, economic conditions, and
the level of cement imports. In 1991, reported consumption of Portland and masonry cement in
the U.S. was 69.3 million metric tons (76.3 million tons), representing the lowest consumption
since the early 1980s.9 The companies that comprise the U.S. cement industry, and the number
and location of the plants that they operate are presented in Appendix B-l.
The most recent figures on clinker production are from 1990.10 In 1990, the domestic
cement industry consisted of 43 companies operating 115 clinker-producing plants, consisting of
218 kilns, in 37 states and Puerto Rico. Total domestic clinker production in 1990 was nearly
65.4 million metric tons, comprising an 85.7 percent utilization of the total available capacity of
close to 76.2 million metric tons. Dry kilns accounted for approximately 45 million metric tons
of this production, representing over 65 percent of the total. The slight dominance in terms of
production by dry kilns reflects the recent trend toward greater reliance on this more energy-
efficient process.
California was the largest clinker producing state in 1990, followed by Texas,
Pennsylvania, Missouri, and Michigan. Exhibit 2-2 presents 1990 clinker production by state.
This geographic distribution reflects the regional nature of the industry. Because of the low
value-to-weight ratio of cement and the resulting high relative cost of transportation, most
cement plants are located within 321.9 kilometers (200 miles) of their principal markets.11
Therefore, the states with the largest populations and economies, which have enjoyed dynamic
construction trends, tend to be the largest cement markets and also the largest cement producers.
There has been little change over the past 30 years in terms of the largest cement-consuming
states; California, Texas, and Florida have been the leaders throughout this period.
8 Environmental Toxicology International, 1992. All Fired Up: Burning Hazardous Waste in Cement Kins, Seattle,
Washington.
9 Portland Cement Association, 1992b. U.S. and Canadian Portland Cement Industry: Plant Information Summary
1991, Market and Economic Research, Skokie, Illinois. August, 1992.
10 Production statistics for this report were derived from capacity information presented in the 1991 PCA facility
survey and capacity utilization data reported by the U.S. Bureau of Mines. For more detailed information on the
derivation of these data, refer to Appendix B-2.
11 The exception to this distance limitation comprises those production facilities that have ready access to water
transportation, enabling them to extend their marketing areas through the use of distribution terminals. Although
ultimate delivery of cement to the consumer is almost always performed by truck, 90 percent of shipments from
production plants to distribution terminals are done by rail and waterway (Environmental Toxicology International,
1992, op. tit.)
-------
2-7
Exhibit 2-3 presents the geographic distribution of cement plants across the U.S. The
location of plants also corresponds roughly to the location of the larger cement markets.
California, Texas, and Pennsylvania each had 11 cement facilities in 1990. These larger states
contain several distinct markets, each with multiple cement producers supporting the construction
requirements of individual metropolitan areas. On the other hand, there are a number of one-
plant companies that serve the needs of a particular regional market and a number of states that
exist as individual cement markets. In addition, 13 states and the District of Columbia currently
have no cement-producing plants and rely upon cement supplies from other (usually adjacent)
states. The 13 states are:
-------
2-8
Exhibit 2-2
1990 Clinker Production by State
State
CALIFORNIA
TEXAS
PENNSYLVANIA
MISSOURI
MICHIGAN
ALABAMA
INDIANA
NEW YORK
IOWA
SOUTH CAROLINA
ILLINOIS
FLORIDA
PUERTO RICO
MARYLAND
KANSAS
OKLAHOMA
ARIZONA
OHIO
VIRGINIA
GEORGIA
ARKANSAS
UTAH
COLORADO
TENNESSEE
14 OTHER STATES
Totals
Number
of Plants
11
11
11
5
4
5
5
4
4
3
4
4
2
3
4
3
2
3
1
2
2
2
3
2
15
115
Number of Kilns
Wet
1
9
8
2
3
0
3
4
2
6
0
5
4
2
4
2
0
4
0
0
5
2
1
2
12
81
Dry
19
11
15
5
6
6
6
1
4
1
8
3
2
5
7
5
7
1
5
3
0
1
2
1
13
137
Total
20
20
23
7
9
6
9
5
6
7
8
8
6
7
11
7
7
5
5
3
5
3
3
3
25
218
Reported
Capacity
(tons per year)1
10,389,000
7,853,940
5,226,000
4,158,000
3,866,390
4,096,000
2,794,900
2,941,630
2,498,000
2,493,570
2,265,000
2,816,700
2,188,007
1,750,000
1,737,144
1,791,000
1,640,000
1,507,000
1,100,000
1,082,730
1,239,000
853,000
1,250,000
940,000
7,506,400
75,983,411
Reported
Capacity
Utilization
85.0%
89.0%
81.8%
99.8%
98.6%
74.9%
97.4%
86.1%
99.8%
95.5%
98.4%
71.4%
78.8%
93.7%
80.4%
74.0%
71.6%
76.0%
87.9%
87.6%
69.1%
99.7%
66.8%
87.6%
80.3%
85.7%
Calculated
Production
(tons clinker)
8,830,650
6,990,007
4,274,868
4,149,684
3,812,261
3,067,904
2,722,233
2,532,743
2,493,004
2,381,359
2,228,760
2,011,124
1,724,150
1,639,750
1,396,664
1,325,340
1,174,240
1,145,320
966,900
948,471
856,149
850,441
835,000
823,440
6,027,639
65,117,783
• These numbers have been converted from short tons per year as reported in the source.
Source: Capacity data derived from facility responses to the 1991 PCA Survey, 1992 North American Cement
Directory, and the Bureau of Mines 1990 Annual Report on Cement. Utilization data taken from the
Bureau of Mines 1990 Annual Report on Cement.
-------
2-9
Exhibit 2-3
Geographic Distribution of Active Portland Cement Production Facilities in 199U
Ui
Source: Portland Cement Association, 1992b, op. at.
-------
2-10
Alaska Minnesota North Dakota
Connecticut New Hampshire Rhode Island
Delaware New Jersey Vermont
Louisiana North Carolina Wisconsin
Massachusetts
In 1990 the five largest U.S. cement companies in terms of clinker production were as
follows: Holnam, Incorporated; Lafarge Corporation; Lehigh Portland Cement Company; Lone
Star Industries; and Southdown, Incorporated. These companies also ranked identically in terms
of capacity, with the top five accounting for approximately 39 percent of industry clinker capacity.
Clearly, the cement industry is not one with highly concentrated ownership; the five largest
producers in many other types of manufacturing industries account for 80 percent or more of the
total industry capacity.12
Nevertheless, the structure of the domestic cement industry has changed radically over
the last 20 years. There are now fewer companies with one- and two-plant operations, and many
established cement companies are out of business or have sold off their holdings. In general, the
industry has been shrinking in terms of capacity over the last decade. Exhibit 2-4 presents U.S.
clinker capacity for the period 1973-1991. Also, foreign ownership of U.S. clinker capacity has
increased from 5 percent in the late 1970s to over 70 percent in 1992.13 Both Holnam Inc. and
Lafarge Corporation are foreign owned.14
Seasonal cycles are typical in the cement industry. Demand during peak summer months
may be three times that of a winter month. Typically, a cement producer will build up
inventories during the winter months in anticipation of peak demands during the summer
construction period. Trade literature reports that about two thirds of U.S. cement consumption
generally occurs during the six month period from May to October.
Historical data document the fact that trends in cement consumption generally follow
trends in construction activity. In the past, the cement industry has experienced prolonged cycles
in activity over 15 to 20 years, as well as the shorter traditional business cycles.
Imports of cement usually account for a small percentage of total U.S. cement
consumption. Historically, cement and clinker importation has been highly cyclical, but still has
only accounted for between three and 11 percent of consumption. The 11 percent figure
occurred in 1979, and the lower figures generally appear during the later stages of recessionary
periods. Higher rates of importation usually occur when plants are operating at full capacity and
are still unable to meet consumer requirements. Imports of cement tend to affect domestic
coastal markets, like Florida and California, to a greater degree, because in these markets, costly
ground transportation of cement product to consumers is not required. Historically, it has been
Canada, Japan, Spain, and Mexico that have been
12 Portland Cement Association, 1984. U.S. Cement Industry - an Economic Report, Skokie, Illinois. January 1984.
13 U.S. Department of Commerce, Date Unknown. Construction Materials Database, Building Materials and
Construction Division, Office of Forest Products and Domestic Construction, International Trade Administration,
Washington, D.C.
14 Portland Cement Association, 1992b, op. at.
-------
2-11
Exhibit 2-4
U.S. Clinker Capacity for the Period 1973-1991
Source: Portland Cement Association, 1992b, op. at.
key exporters of cement to the U.S. Exports of U.S. cement are reported to have seldom
exceeded 1 percent of total production. Not surprisingly, exports tend to occur when domestic
market demand is low relative to foreign markets, and during domestic recessions.
2.2 CEMENT MANUFACTURING
This section provides a more detailed presentation of cement manufacturing, focusing on
those aspects that influence the generation and characteristics of CKD. This section begins with
a general description of the basic steps common to all cement production processes, followed by
a comparison of the different kiln types that are in operation in the U.S., focusing on noteworthy
operational differences and industry trends. The section concludes with a discussion of the
inputs into the process of manufacturing cement (i.e., raw materials and fuels). The growing use
of hazardous waste-derived fuels is an issue that is of growing importance not only to the
regulatory status of CKD but to the larger issue of hazardous waste management in the United
States.
-------
2-12
2.2.1 The Basic Production Process
Although a variety of cement types are produced in the U.S., cement production
generally follows a standard series of steps. The focus of this section is on the manufacture of
Portland cement, but the basic production steps are common to most other types as well.
Portland cement is derived from a combination of calcium (usually in the form of limestone),
silica, alumina, iron oxide, and small amounts of other materials. These raw materials are
quarried, crushed, ground together, and then burned in rotary kilns at temperatures near
1,482 C (2,700 F). The resulting material (a mixture of marble- to fist-sized pellets and sand-
sized particles) is called clinker. The clinker is finely ground into a gray powder and mixed with
gypsum to slow down the "setting" (i.e., hardening) of the cement when it is used in concrete.15
Manufacturers use clinker and specific additives in various proportions to create cements having
different properties for specific construction applications. The general manufacturing steps are
discussed below in greater detail and are presented schematically in Exhibit 2-5.
Typical Portland cement consists largely of limestone, clay and/or shale, and a small
amount of gypsum and other minerals, such as iron ore, sand, or bauxite. Most of these mineral
inputs into the production of cement are quarried on site, ground and blended, then fed into a
kiln system. The type of Portland cement being produced determines the specific proportions of
these raw materials. High levels of impurities in the raw feed may cause operational problems in
the kiln, and/or produce inferior cement. A more detailed discussion of the differences in raw
material types and composition and their effects is presented in Section 2.2.3.
Mining
Material mined for cement production consists primarily of limestone. Mining usually
occurs in open quarries, although underground mining has been used in the past. During a
typical surface mining operation, shovels or bulldozers remove overburden, rock is blasted, and
front-end loaders or power shovels load the blasted rock into trucks or railroad cars.16 The size
of the mined rock is up to one meter in diameter. The rock is transported to the crushing plant
located in the quarry or at the cement plant. Overburden, or waste rock, is often disposed on
site in or adjacent to the quarry.17 Overburden has also been used for riprap and fill material
or sold as aggregate for concrete or road base use.18
Crushing
Once removed, the raw material (e.g., limestone) is crushed to a smaller size. The type
of crusher used is dependent on the nature of the rock (e.g., hardness, lamination, and quarry
u Portland Cement Association, 1984, op. at.
16 Not all quarries mine by blasting hard rock. In the Southeastern U.S., draglines are used to break up and load
semi-consolidated material.
17 Muelberg, P.E., et al., 1977. Industrial Process Profiles for Environmental Use: Chapter 21, The Cement Industry,
U.S. Environmental Protection Agency, Office of Research and Development, Industrial Environmental Research
Laboratory, Cincinnati, Ohio. February 1977.
18 Johnson, W., 1992. Cement, Annual Report 1990, U.S. Bureau of Mines, Washington, D.C. April 1992.
-------
2-13
product size). Common crusher types include: gyratory, jaw, and roll crushers and hammer and
impact mills.
-------
2-14
Exhibit 2-5
Steps in the Manufacture of Portland Cement
-------
2-15
Exhibit 2-5 (continued)
Steps in the Manufacture of Portland Cement
-------
2-16
Typically, a primary crusher reduces the rock from power shovel size to 0.1 to 0.25 meter (0.328
to 0.82 feet) in diameter and a secondary crusher reduces the product to 0.01 to 0.05 meters in
diameter. This material is then conveyed with other raw materials to the grinding and blending
step. Partial drying of rock is sometimes accomplished during the crushing process by passing
furnace-heated air, clinker cooler exhaust air, or kiln exhaust gases through the crusher.19-20
Grinding and Blending
In general, during the grinding process crushed raw material is fed into the grinding mill,
ground to a fine size range, and blended to obtain the correct composition for kiln feed. This
material is commonly referred to as raw mix or raw meal. In the dry process, ground and
blended materials are usually conveyed to a pre-drying step before being fed to the kiln. In the
wet process, the raw materials are mixed with about 30 to 40 percent water during grinding or
blending to form a well-homogenized slurry. Low concentrations of slurry thinners may be
added, such as sodium carbonates, silicates, and phosphates, as well as lignosulfates and modified
petrochemicals.21-22
Pre-Drying
In the dry process, raw materials must be dried before they are fed into the kiln. The
moisture content of the ground raw materials is usually three to eight percent, but may reach 20
percent. Drying reduces moisture content to less than one percent. Ground stone is usually
dried with furnace-heated air, clinker cooler exhaust air, or kiln exhaust gases in a cylindrical
rotary dryer.23
Drying and Preheating
Within the kiln system, different chemical reactions and phase formations occur that are
defined by specific temperature ranges of the feed material. Exhibit 2-6 shows the various
reaction zones and the raw material temperatures in a typical rotary kiln.
The drying and preheating zone is where the first true thermal treatment occurs within
the kiln system. In some kiln configurations (e.g., in wet kilns) this zone is located in the cool
end of the kiln, while in others it resides in the preheater or precalciner. In any case, the
following three sequential processes occur in this zone: evaporating all free water (material
19 Sapp, J.E., 1981. Energy and Materials Flows in the Cement Industry, prepared for U.S. Department of Energy,
June 1981.
20 Muelberg, et ai, 1977, op. at.
21 Sapp, 1981, op. cit.
22 Kirk-Othmer, 1979. Cement, Kirk-Othmer Encyclopedia of Chemical Technology, Volume 5, John Wiley and
Sons, New York, New York.
23 Muelberg, et a!., 1977, op. cit.
-------
2-17
temperature lOO'C); dehydrating clay minerals (material temperature beginning at 549 C); and
increasing the feed temperature to the calcining temperature (material temperature 804 C).24
24 Peray, K.E., 1986. The Rotary Cement Kiln, Second Edition, Chemical Publishing Co., Inc., New York, New
York.
-------
. 2-18
Exhibit 2-6
Material Temperature Ranges in the Kiln Portion of the Cement Manufacturing
-------
2-19
Calcining
Calcining, or calcination, is the process through which carbonates or other compounds
are decomposed by application of heat. Carbon dioxide is driven off from limestone (CaCO3)
and magnesium carbonate (MgCO3) contained in the feed, leaving free lime (CaO) and magnesia
(MgO). This process occurs at a material temperature range between 804°C and 1,200°C.25
Complete calcination of the kiln feed before it enters the burning zone is essential to
proper burning and clinker formation.26 The successful calcining process, which produces a
grayish-green clinker, requires the appropriate temperature and an oxidizing atmosphere to
completely decompose the carbonates in the feed materials. Insufficient oxidizing conditions
yield a brown clinker that produces an inferior cement.27
Burning
Though often used synonymously with the term sintering, the burning of the calcined kiln
feed is actually a three-stage process occurring in the hot end of the kiln (refer to Exhibit 2-6).
The area in which burning occurs can be divided into three sections: the upper transition zone,
the sintering zone, and the cooling or lower transition zone. This production stage is also
commonly referred to as clinkering, because passing through these zones results in the kiln feed
becoming clinker. In the upper transition zone, interim-phase formations occur while some
calcination is still being completed. The upper transition zone (material temperature range
1,200-1,400 C) is identified in a temperature profile by a rapid rise in material temperature just
at the end of the calcining zone. The final stages of clinker compound formation take place in
the sintering zone. This process involves exothermic reactions and is the zone of highest material
temperature (1,400-1,510 C). The last 3.05 to 6.1 m of the kiln's discharge (hot) end constitute
the cooling zone (material temperature range from a high of 1,510 C to a low of 1,290 C).28
Cooling
After leaving the kiln, the clinker is further cooled in rotary, planetary, or grate-type
coolers by air pulled into the unit by dedicated cooler fans, and then transferred by conveyor to
the finish mill.29 The cooling process conditions can significantly influence the quality of the
clinker. Generally, faster cooling rates result in a higher quality clinker.30
25 ibid.
*ibid.
27 Kohlhaas, B., et al., 1983. Cement Engineers' Handbook, Bauverlag GmbH, Wiesbaden and Berlin.
28 Muelberg, et al., 1977, op. at. These units have been converted to C from °F as reported.
"Ibid.
30 Peray, 1986, op. at.
-------
2-20
Finish Milling/Loading
At the finish milling stage, Portland cement is produced by grinding clinker together with
about five percent gypsum to a fine powder (94 to 98 percent particles with diameters less than
0.044 millimeters). It is then loaded into bulk carriers or packaged into bags. The gypsum is
added to retard the setting time of the cement, thereby making it more suitable for common
construction applications. It is at this stage that various additives along with gypsum are
introduced to create specialty Portland cements. For example, masonry cement production
includes blending crushed limestone with the clinker and gypsum. Other typical finish milling
additives include: blast furnace slag, fly ash, and natural pozzolans (such as volcanic rocks,
diatomaceous earth, and burned oil shale residue).31
During the finish milling process, the proportioned materials are drawn up by belt
conveyors to a two-compartment ball mill. A separator device recycles oversized products and
sends correctly sized product to storage. To prevent dehydration of the gypsum, either air or
water cooling is used during grinding.32
2.2.2 Kiln Design
Rotary cement kilns are horizontal, inclined rotating cylinders that are refractory lined,
internally fired, and designed to produce clinker through the intense heating of raw materials.
Raw materials are fed into the upper, cool end while fuels are normally fed into the lower, hot
end. As a function of the inclined surface, combustion gases and raw materials move
counterflow in kilns. Thus, the raw materials get progressively hotter as they travel down the
length of the kiln to become clinker at the low, hot end. U.S. cement kilns range in length from
35.4 to 231.6 meters (m) (116 to 760 feet), and 2.4 to 7.3 m in diameter.33
Clinker is manufactured in five kiln types:
u Wet process;
u Dry process;
u Preheater;
u Precalciner; and
u Semidry process.
These kiln types represent variations on two primary cement kiln designs: wet process
and dry process (preheater, precalciner, and semidry process kilns are variations on the standard
dry process). Raw materials are generally the same for both wet and dry processes, but the
sequences and operations for raw material crushing, grinding, and blending are different.34 In
both process types, the kiln generally slopes at an angle of about 3 degrees. Most kilns, both wet
31 KoWhaas, el al., 1983, op. at.
32 Muelberg, el al., 1977, op. at.
33 Johnson, W., 1992, op. tit.
34 Beers, A., 1987. Hazardous Waste Incineration: The Cement Kiln Option, New York State Legislative
Commission on Toxic Substances and Hazardous Wastes, Albany, New York. December 1987.
-------
2-21
and dry, are equipped with a 18.3 to 30.5 m "chain" section in the upper, or cool end of the kiln,
to increase the heat transfer rate from the discharge gas to the raw feed and to help keep the
process material moving down the kiln. Chain sections also provide a filtering action to reduce
dust emissions.35-36 Though the chains are expensive and require a high level of maintenance,
the improved energy efficiency of the kiln justifies their use.37-38
.• One major difference between wet and dry kilns is the kiln length.39 As described above
and shown in Exhibit 2-6, the raw material travels through several reaction zones before
becoming clinker. The number of these zones required and their relative length varies by kiln
type. Wet process kilns require additional length for preheating the slurry feed. To obtain the
necessary heat transfer for water evaporation, wet kilns must be long, typically ranging from 137
to over 183 m in length.40'41 Dry process kilns do not require this slurry preheat zone, and
require about 10 percent less length for evaporation.42 Therefore, standard dry process kilns
may be somewhat shorter than otherwise comparable wet kilns. The preheater, precalciner, and
semidry process kilns have only calcining and burning zones, because the material has been dried
before it enters the kiln. Accordingly, these kiln types may in some cases be very short (under
61 m).43-44 Therefore, in general, dry kiln types tend to be shorter than wet kilns of the same
production capacity (i.e., their length to capacity ratio is smaller). As a result, the use of dry kiln
technology has allowed the construction of higher capacity kilns. For example, precalciner kilns
in operation in Japan reportedly produce over 9,072 metric tons of clinker per day,45 while the
largest U.S. clinker capacity for a preheater/precalciner kiln is 4,700 metric tons per day.46 By
comparison, available data indicate that the largest wet kilns in the U.S. have a capacity of
approximately 3,630 metric tons per day.
35 Muelberg, el ai, 1977, op. at.
36 Engineering-Science, 1987. Background Information Document for the Development of Regulations to Control the
Burning of Hazardous Wastes in Boilers and Industrial Furnaces, Volume II: Industrial Furnaces, submitted to U.S.
Environmental Protection Agency, Waste Treatment Branch, Washington, B.C. January 1987.
3' Peray, 1986, op. at.
M Muelberg, el ai, 1977, op. at.
39 Engineering-Science, 1987, op. at.
40 Engineering Science, 1987, op. at.
41 Johnson, 1992, op. at.
43 Peray, 1986, op. at.
43 Engineering Science, 1987, op. at.
44 Johnson, 1992, op. tit.
43 Kirk-Othmer, 1979, op. at. '
46 Portland Cement Association, 1992b, op. at.
-------
2-22
Another major difference between wet and dry processes is evident in their heat
requirements.47 At any given location in a kiln, the gas temperature always exceeds the material
temperature. However, the closer that these two temperatures approach one another, the more
efficient the heat exchange between the gases and the material. Each of the five kiln types has a
characteristic temperature profile. In dry kilns, gas and material temperatures tend to be much
closer over the entire length of this profile than in wet kilns.48 Therefore, dry kilns tend to be
more efficient than wet kilns, requiring less energy input per ton of clinker produced. Exhibit 2-
7 presents reported typical energy requirements for various kiln types.
Exhibit 2-7
Typical Energy Requirements for Each Kiln Type*
Kiln Type
Wet Process Kiln
Dry Process Kiln
Semidry (Lepol) Kiln
Preheater Kiln
Precalciner Kiln
Energy Required
(Kcal/kg output)
1,529 - 1,668
1,251 - 1,390
945.2
750.6 - 889.6
Unknown"
' Energy requirements provided in this exhibit are those reported in the literature. Average energy
requirements calculated from facility responses to the 1991 PCA survey are comparable. The sample of 46 wet kilns for
which complete information was provided averaged 1,520.7 Kcal/kg (5.47 million Btus). The sample of 83 dry kilns,
including semidry, preheater, and precalciner kilns, averaged 1,175.9 Kcal/kg.
b Specific data were not available for precalciner kilns. However, while precalciners burn 30-50 percent of their
total energy input at the rear of the kiln, reducing the heat load on the burning zone, preheater and precalciner kilns
consume approximately the same amounts of fuel. Therefore, while precalciners reduce the heat requirements of the kiln
itself, it is believed that the total heat required for the complete process (calcination through clinker production) is
unchanged and similar to that required for a typical preheater kiln.
Sources: Engineering Science, 1987, op. at.; Perry's Chemical Engineers' Handbook, 1984. Sixth Edition, McGraw-Hill,
New York, New York.
The specific characteristics of the five kiln types are discussed below in more detail.
Wet Process Kiln
In the wet process, the limestone and other raw mix components are ground wet and
slurried at a moisture content of 30 to 40 percent. This slurry is fed into the upper, or cool, end
of the kiln and flows down slope through the kiln to the hot discharge end. A typical process
47 Engineering Science, 1987, op. at.
48 Peray, 1986, op. at.
-------
2-23
flow diagram for a wet process kiln is presented in Exhibit 2-8. Wet process kilns are longer
than dry process kilns because a substantial portion of the kiln length (20 to 25 percent) must be
used for evaporation of the slurry water.49-50
When compared to dry process kilns, reported advantages of wet process kilns include
more uniform feed blending, generally lower emissions of kiln dust, and compatibility with moist
climates where complete drying of raw feed is difficult to achieve.51-52 The primary
disadvantage associated with wet process kilns, however, is that they require significantly more
energy, because large quantities of water must be evaporated from the raw feed, resulting in
higher operational costs. Typical energy requirements for wet process kilns range from 1,529 to
1,668 Kcal/kg of clinker produced.53-54
Dry Process Kiln
In the dry process, dry raw mix is pneumatically pumped to the upper end of the kiln.
The meal flows down through the sloped kiln as it is thermally treated. Dry process kilns have
diameters similar to wet process kilns, but are shorter because there is no evaporation zone
required. Dry process kilns operate with a high exit gas temperature of approximately 450 C
(840°F), and typically, employ water sprays to cool the gas before it enters the dust control
equipment.55
Kiln gases exiting dry process kilns do not pass through a wet raw mix that would
significantly decrease exit gas temperatures. The high exit gas temperatures can therefore be
used for cogeneration of electrical power. This fact may be significant to existing dry process
plants because cogeneration could be energy-conserving. In some locations it may even be more
economical to add a power plant to an existing dry kiln than to retrofit the kiln with a
preheater.56 The number of facilities currently coupled with power plant operations is not
presented in the literature. In the absence of cogeneration equipment, the hot exit gas is used as
supplemental combustion air for the kiln fuel.57
* Muelberg, et al., 1977, op. at.
50 Engineering Science, 1987, op. at.
51 Peray, 1986, op. at.
" Beers, 1987, op. at.
53 Engineering Science, 1987, op. at.
54 Perry's Chemical Engineers' Handbook, 1984, op. at.
" Peray, 1986, op. at.
"Ibid.
57 Muelberg, et al., 1977, op. at.
-------
2-24
Exhibit 2-8
Typical Process Flow Diagram for a Wet Process Cement Kiln
-------
2-25
Increased energy efficiency is a major advantage of dry process kilns in comparison to wet
process kilns. Available data indicate that dry kilns are approximately 10-25 percent more
thermally efficient than wet kilns, requiring 1,251 to 1,390 Kcal/kg of clinker produced.58-59
Prebeater Kiln
Preheater kilns, more accurately referred to as suspension preheater kilns, preheat and
partially calcine raw meal by passing it through a system of heat exchange cyclones before it
enters the kiln.60-61 A typical, four-stage suspension flash preheater kiln is illustrated in Exhibit
2-9. As the raw material passes through each of the four stages, it gets hotter and becomes more
processed before entering the kiln, resulting in more uniformly processed material. Suspension
preheaters reduce by one half to two thirds the required kiln length of the rotary dry process
kilns they precede. Because of the increasing fuel costs experienced by kiln operators during the
1970's and early 1980's, suspension preheaters came into common use.62
Suspension preheater kilns are the most energy-efficient types of kilns available,
producing clinker at energy consumption rates ranging from 750.6 to 889.6 Real/kg.63'64 The
addition of a preheater with or without a precalciner improves process efficiency, through fuel
savings and scale economies associated with larger production units.65 The feed in precalciner
and preheater kilns also tends to be much more uniformly calcined than in dry kilns and even in
longer wet kilns. Operating conditions in the precalciner and preheater kilns are also easier to
control.66
One disadvantage of the preheater kiln is that plug-up problems can occur at the lower
cyclone stage and kiln inlet due to high concentrations of volatile constituents such as alkalies,
sulfur, and chlorides in the kiln exit gases. To reduce the severity of this problem, alkali and
sulfur bypass systems allow evacuation of some of the kiln exit gases before they reach the
preheater cyclones.67
58 Engineering-Science, 1987, op. dt.
59 Peny's Chemical Engineer's Handbook, 1984, op. dt.
40 Peray, 1986, op. dt.
61 Muelberg, el al, 1977, op. dt.
62 Peray, 1986, op. dt.
° Engineering Science, 1987, op. dt.
64 Perry's Chemical Engineers' Handbook, 1984, op. dt. The units were converted to Kcal/kg from million
BtusAon as reported.
65 Engineering-Science, 1987, op. dt.
66 Peray, 1986, op. dt.
67 Ibid.
-------
2-26
Exhibit 2-9
Typical Process Flow Diagram for a Suspension Flash Preheater Cement Ki
-------
2-27
Precalciner Kiln
Precalciner kilns, the most recent advance in cement manufacturing technology, are
essentially suspension preheater kilns that are equipped with a secondary firing system (flash
furnace) attached to the lower stage of the preheater tower. A pyroclone precalciner kiln is
illustrated in Exhibit 2-10. Precalciner kilns fall into two categories - kilns with and kilns
without tertiary air ducts. Kilns with tertiary air ducts are supplied with air from the exhaust
gases from the clinker cooler. Precalciner kilns without tertiary air ducts receive air from the
kiln itself. The kilns with tertiary air ducts are generally more difficult to control.68 Both
categories process the raw feed similarly.
Precalciner and preheater kilns have several advantages over conventional dry and wet
long kilns. As mentioned above, the feed in precalciner and preheater kilns is much more
uniformly calcined than in conventional dry and wet long kilns. Operating conditions in the
precalciner and preheater kilns are also easier to control. Moreover, precalciner kilns greatly
reduce the residence time needed for raw material to become clinker. A conventional dry kiln,
for example, requires approximately 45.7 m to achieve 90 percent calcination, a distance that
takes the feed about one hour to traverse. The precalciner kiln, in contrast, completes the same
amount of thermal work in less than a minute. Precalciners do not improve the energy efficiency
of the calcining process, but they do reduce the required heat load in the rotary kiln, thereby
extending refractory service life, and reducing operational costs. Additionally, less expensive,
lower grade fuels, such as subbituminous coal, lignite, and oil shale, as well as tires and waste oil,
can be burned in the auxiliary firing unit, reducing the fuel cost per unit of clinker. Precalciner
kilns also exhibit output rates that were previously considered unattainable.69
The addition of a preheater or precalciner significantly improves the overall energy
efficiency of the cement production process. Depending on whether a preheater and/or
precalciner is used, dry process plants can have a long, medium, or short kiln.70
Semidry Process Kiln
The semidry process kiln, also known as the "grate process" or the "Lepol" kiln, is a type
of preheater kiln that begins with raw feed nodules containing 10 to 15 percent moisture. In the
semidry process, raw material nodules travel on a grate through a preheater in which they are
partially calcined. The partly calcined material then falls down a chute into the rotary kiln where
the final clinkering takes place. The partial precalcination allows the rotary kiln to be only about
one-third the usual length. This type of kiln usually requires additional labor to monitor the
thickness of the feed bed and to oversee production of the nodules, which is not required in
operating conventional rotary kilns. Some plants use filter press cakes rather than nodules for
kiln feed. In these cases, the wet-feed slurry is passed through large presses to remove free
water, and more importantly, to remove alkalies before the cakes are fed to the kiln.71
"Ibid.
"ibid.
TO Engineering Science, 1987, op. at.
71 Peray, 1986, op. cit.
-------
2-28
-------
2-29
Exhibit 2-10
Typical Process Flow Diagram for a Precalciner Coal and Waste Fired Cemefl
-------
2-30
Although nearly as energy efficient as preheater and precalciner kilns, the semidry
process kiln has a limited output capacity. The semidry process is practical, however, in
geographic areas where raw material moisture is so high that it cannot be economically removed
by waste heat from the kiln. An additional advantage is that the semidry process kiln operates
with much lower dust content in the waste gases than preheater and precalciner kilns.72 In
1990, however, only four of 115 domestic cement plants reported using a semi-dry process kiln
system. The small number of these kilns operating in the U.S. is believed to be related to low
output rates and the labor intensive nature of this process.
Trends in Kiln Technology and Use
Exhibit 2-11 shows the age distribution of kilns both in terms of numbers and capacity.
Clearly, there has been a trend over the years toward larger capacity kilns of both basic types.
While the earliest domestic facilities produced Portland cement by the wet process, the dry
process also has been utilized for some time. From the 1950's through the 1970's, wet process
kilns remained competitive with the more energy-efficient dry kilns because they were less labor-
intensive than dry
Exhibit 2-11
Age and Capacity of Existing Kilns
Date of Kiln
Installation
After 1980 '
1976-1980
1971-1975
1966-1970
1961-1965
1956-1960
1951-1955
1946-1950
1941-1945
1936-1940
1931-1935
Before 1931
Wet Kilns
Number of
Kilns
0
0
11
13
20
19
5
6
1
1
0
4
Average
Capacity
(Ktons)
0.0
0.0
459.6
424.9
355.4
260.5
193.6
158.0
150.0
153.0
0.0
159.3
Dry Kilns
Number of
Kilns
26
18
32
19
38
46
19
9
0
2
0
0
Average
Capacity
(Ktons)
690.3
653.1
476.8
425.0
336.1
199.6
176.6
164.0
0.0
138.0
0.0
0.0
Overall
Number of
Kilns
26
18
43
32
58
65
24
15
1
3
0
4
Average
Capacity
(Ktons)
690.3
653.1
470.9
426.9
346.3
226.7
181.1
160.0
150.0
143.0
0.0
159.3
Source: Adapted from Portland Cement Association, 1992b, op. at.
• Ibid.
-------
2-31
kilns.73 However, no new wet kilns have been constructed since 1975. This change to more
energy-efficient dry technologies may be attributed to rising energy prices. Because dry kilns can
be constructed with larger capacities than wet kilns, the trend toward larger capacities also may
have made this technology more attractive because of positive economies of scale.
Kilns normally operate 24 hours/day and 7 days/week, but temporary shutdowns for
refractory relining and other maintenance activities reduce the effective annual operating period
substantially. The most recent figures report the average annual operating time at 313 days/year
as of 1990. Although the kilns operate nearly year round, they do not necessarily operate at
maximum capacity at all times. The portion of capacity utilized depends upon market factors.
As indicated in the preceding discussion of industry structure, annual clinker capacity utilization
is approximately 86 percent.74
2.23 Process Inputs
Many of the issues surrounding CKD focus on the health and environmental impacts
associated with exposure to CKD or its constituents,75 and on the development of technologies
to improve recovery rates, minimize generation, improve operational efficiencies in existing kiln
systems, and seek alternative beneficial applications for existing, stockpiled CKD. An
understanding of these issues requires knowledge of both the raw materials and the fuels used in
cement kiln systems, because it is these inputs, coupled with the manufacturing process, that to a
large extent, determine the characteristics and quantities of CKD generated.
The following discussion provides insight into how the inputs into the process of
manufacturing Portland cement can affect process operations and the ultimate composition and
fate of CKD. The focus is on the major (bulk) constituents of raw mix and fuel inputs. A
detailed discussion of trace constituents and their sources, fate, and impacts is presented in
Chapters 3, 5, and 7.
The data used in this section of the report were taken from facility responses to the 1991
PCA survey; however, not all domestic facilities responded to this survey. For example, the
totals for raw material and fuel consumption reported here do not reflect total consumption for
the industry. Rather, they reflect consumption for the sample of facilities that responded to the
pertinent questions in the survey. EPA has no reason to believe that these facilities are not
representative of the population at large, and has conducted the analyses presented in this report
accordingly. In addition, the sample of facilities providing information for 1990 in the survey was
not necessarily the same as the sample providing information for 1985. Therefore, the reader
should not compare or make any inferences regarding total consumption for these two years.
Raw (Feed) Materials
Portland cement is commonly made from a mixture of calcareous (calcium containing)
materials, typically limestone, and smaller amounts of materials that contain silica, alumina, and
71 Kirk-Othmer, 1979, op. at.
74 Johnson, W, 1992, op. at.
15 Exposure to CKD can be direct or through the contamination of environmental media such as ground water.
-------
2-32
iron. Kiln feed is generally comprised of about 80 percent carbonate of lime and about 20
percent silica with much lower quantities of alumina and iron. The number of raw materials
required to achieve this blend depends on the composition of the materials and their
availability.76 Exhibits 2-12, 2-13, and 2-14 show the types of raw materials consumed and the
types of feed mixtures used in 1990 and 1985 at facilities that responded to the 1991 PCA
survey.77 There has been very little change in the types of materials used for cement
manufacture over the last five years. Limestone is the primary source of calcium for nearly all
cement plants, though one facility reported using cement rock (a type of limestone that is nearly
perfectly balanced for cement manufacture) in 1990, and one facility reported using marl (an
earthy material containing significant amounts of calcium carbonate). In the past, plants in
Alabama and Arkansas have used chalk, and plants along the Gulf of Mexico and the coast of
California have used crushed coquinoid limestone (often referred to as shell hash) as sources of
calcium.78 However, no facilities reported using these materials in the 1991 PCA survey.
Shale, clay, and sand are the primary materials fed as sources of silica and alumina.
Some facilities also utilize ash (typically fly ash or bottom ash) as a primary or secondary feed for
its silica and alumina content. A majority of facilities use either iron ore or steel mill scale to
supplement the iron content of their mix. Other materials used in significant amounts in cement
kilns include the following: gypsum, for its calcium content; bauxite, an aluminum ore;
diatomaceous earth, which is high in silica; and slag, for its silica and calcium content. Exhibit
2-15 presents chemical composition data for several types of limestone and many of these other
feed materials.
In addition to the chemical composition of the desired product, the proportion of each
type of raw material used in a given cement kiln will depend on the composition of the specific
materials available to the operator. Exhibit 2-16 shows the quantity ranges and typical mixtures
of materials fed to kilns in 1990. The specific blend of materials used at a given kiln at a given
time is often calculated through an iterative process using computer programs. The
proportioning process takes into account the ratios of calcium, silica, alumina, and iron needed
to produce good quality clinker, as well as the "burnability" of the raw mix (i.e., the requirements
in terms of time, temperature, and fuel to process the material).79 In addition, kiln operators
pay close attention to the presence of "impurities" in the mixture, including magnesia, sulfur,
chlorides, and oxides of potassium and sodium (referred to as "alkalies"). Magnesia (MgO) levels
are carefully monitored because they can lead to the production of clinker that is unsound if not
cooled rapidly (i.e., such clinker used to make concrete can cause destructive expansion of
hardened concrete through slow reaction with water).80 Magnesia can, however, be desirable to
some extent because it acts as a flux at sintering temperatures, facilitating the burning process.
Alkalies can react in the cool end of the kiln with sulfur dioxide, chlorides, and carbon dioxide
76 Peray, 1986, op. tit.
77 For more detailed information on the derivation of the raw material consumption statistics used in this report,
refer to the Technical Background Document.
78 Boynton, R.S., 1980. Chemistry and Technology of Lime and Limestone, Second Edition, John Wiley and Sons,
New York, New York.
79 Kirk-Othmer, 1979, op. at.
K Taylor, H.F.W., 1990. Cement Chemistry, Academic Press Inc., San Diego, California.
-------
2-33
contained in the kiln gas and can lead to operational problems. Therefore, high levels of
alkalies, sulfur, and chlorides in the feed mix are undesirable.81
81 Peray, 1986, op. at.
-------
2-34
Exhibit 2-12
Feed Mixtures in 1990
Primary
Feed(s)1
Limestone
Limestone
and Gay
Limestone
and Shale
Limestone
and Sand
Limestone
and Ashe
Limestone,
Sand, and
Shale
Limestone,
Sand, and
Clay
Other*
Total
Number
of
Hants'1
19
16
14
10
6
5
4
5
79
Secondary Feedsc (Number of Plants Utilizing)
Sand
12
7
9
--
2
—
--
2
32
Shale
5
1
-
2
0
—
0
2
10
Clay
7
--
0
1
2
1
-
1
12
Iron
Ore
15
7
2
6
3
4
2
5
44
Ashd
3
3
2
1
-
0
1
0
10
Mill
Scale
3
3
0
2
1
1
1
0
11
Gypsum
3
2
2
2
1
0
0
0
10
Other Secondary Feeds
Electroplating Sludge,
Filter Cake, Slag,
Bauxite
Foundry Wastes, Spent
Catalysts, Carbon Black
Bauxite, Iron Flue Dust,
Mag Rock
Staurolite
Slag, Aluminum Silicate
1 For purposes of this analysis, primary feeds are defined as those materials that make up five percent or
more of the total feed.
available.
b This exhibit presents information only for the 79 facilities for which 1990 raw feed information was
c For purposes of this analysis, secondary feeds are those materials that make up less than five percent of the
total feed.
d Includes bottom ash, fly ash, and alumina ash (calcined alumina).
e Includes facilities for which the primary feeds were limestone and ash; limestone, ash, and sand; limestone,
ash, and clay; and limestone, ash, and slag.
' Other primary feed mixtures in 1990 included: limestone, shale, and clay, limestone and gypsum; limestone,
shale, and diatomaceous earth; cement rock and limestone; and marl and limestone.
Source: Facility responses to the 1991 PCA Survey.
-------
2-35
Exhibit 243
Feed Mixtures in 1985
Primary
Feed(s)1
Limestone
Limestone
and Gay
Limestone
and Shale
Limestone
and Sand
Limestone
and Ash*
Limestone,
Sand, and
Shale
Limestone,
Sand, and
Clay
Other'
Total
Number
of
Plants11
26
15
10
5
5
5
5
5
76
Secondary Feeds0 (Number of Plants Utilizing)
Sand
18
3
5
„
4
..
„
3
33
Shale
7
0
„
1
0
„
0
3
11
day
7
-
0
2
1
0
„
2
12
Iron
Ore
19
6
6
2
2
3
4
4
46
Ashd
2
3
0
1
..
1
2
0
9
Mill
Scale
4
3
0
0
2
1
1
0
11
Gypsum
4
2
1
0
0
0
0
0
7
Other Secondary Feeds
Slag, Bauxite, Dolomite
Slag, Fluorspar, Calcium
Chloride, Carbon Black
Iron Flue Dust
Staurolite
1 For purposes of this analysis, primary feeds are defined as those materials that make up five percent or
more of the total feed.
available.
b This exhibit presents information only for the 76 facilities for which 1985 raw feed information was
c For purposes of this analysis, secondary feeds are those materials that make up less than five percent of the
total feed.
d Includes bottom ash, fly ash, and alumina ash.
* Includes facilities for which the primary feeds were limestone and ash; limestone, ash, and clay, and
limestone, ash, and slag.
' Other primary feed mixtures in 1985 included: limestone, shale, and clay; limestone, shale, and
diatomaceous earth; marl and limestone; limestone, sand, shale, and clay; and limestone, sand, and mag rock
(limestone with high magnesium content).
Source: Facility responses to the 1991 PCA Survey.
-------
2-36
Exhibit 2-14
Raw Material Consumption in 1990 and 1985d
Feed Material
Limestone
Shale
Clay
Sand
Marl
Ashc
Iron Ore
Gypsum
Bauxite
Mill Scale
Diatomaceous
Earth
Slag
Mag Rock
Other
Total
1990 Consumption"
Metric Tons
67,852,083
3,685,150
2,989,094
2,076,220
1,526,864
1,158,200
512,659
321,122
89,280
76,623
74,386
57,580
9,980
80,875
80,510,116
Percent of Total
84.28
4.58
3.71
2.58
1.90
1.44
0.64
0.40
0.11
0.10
0.09
0.07
0.01
0.10
100.00
1985 Consumption1"
Metric Tons
60,359,409
3,215,690
3,098,777
1,475,413
644,430
743,550
451,471
149,674
36,320
87,442
81,013
134,286
56,065
20,241
70,553,781
Percent of Total •
85.55
4.56
4.39
2.09
0.91
1.05
0.64
0.21
0.05
0.12
0.11
0.19
0.08
0.03
100.00
* Reflects 1990 raw material consumption for the 79 facilities for which data were available.
b Reflects 1985 raw material consumption for the 76 facilities for which data were available.
c Includes bottom ash, fly ash, and alumina ash.
d These numbers have been converted from short tons per year as reported in the source.
Source: Facility responses to the 1991 PCA survey.
-------
2-37
Exhibit 2-15
Typical Composition of Raw Materials
Feed Material
Constituent (weight percent)
CaO
Si02
A1A
FeA
MgO
Ignition
Loss
(weight
percent)
CALCIUM SOURCES
Indiana high calcium limestone
Virginia high calcium limestone
Kansas cretaceous high calcium limestone (chalk)
Illinois Niagaran dolomitic limestone
Northwestern Ohio Niagaran dolomitic limestone
New York magnesian limestone
Lehigh Valley, Pennsylvania limestone (cement
rock)
Pennsylvania limestone (cement rock)
Marl
54.5
55.3
52.5
31.2
29.5
45.7
38.9
41.8
49.1
0.7
0.4
2.4
0.1
0.1
2.6
19.8
13.4
6.0
0.7
0.1
1.6
0.3
0.0
0.2
5.4
4.6
0.6
0.1
0.1
0.6
0.2
0.1
0.2
1.6
0.6
2.3
0.6
0.5
0.6
20.5
21.1
7.1
2.7
1.9
0.4
40-43
40-43
40-43
44
44
44
32-37
32-37
40.4
SILICA AND ALUMINA SOURCES
Shale
Clay
Sand
Blast Furnace Slag
Bauxite
3.2
0.5-
0.9
0.8
35.5
-
53.8
61.0-
67.8
70.0
33.1
10.6
18.9
14.3-
16.9
15.0
9.1
57.5
7.7
4.5-
12.4
5.0
0.9
2.6
2.2
0.4-
1.2
0.2
16.4
—
8.2-
13.1
8
8.6
2.1
28.4
IRON SOURCES
Iron Ore
Steel Mill Scale
-
--
6.7
Z5
1.4
1.1
89.7
89.9
0.4
-
0.2
4.0
Sources: Boynton, 1980, op. cit.; Peray, 1986, op. cit.; Kirk-Othmer, 1979, op. at.
-------
2-38
Exhibit 2-16
Typical Feed Mixtures in 1990
Primary Feed(s)
Limestone
Limestone
and Gay
Limestone
and Shale
Limestone
and Sand
Limestone
and Ashb
Limestone,
Sand, and
Shale
Limestone,
Sand, and
Clay
Range
Average
Range •
Average
Range
Average
Range
Average
Range
Average
Range
Average
Range
Average
Range of Above
All
Facilities'
Range
Average
Feed Composition (weight percent)
Limestone
89.67 -
99.14
95.05
74.54 -
91.10
83.08
69.27 -
91.98
82.48
80.52 -
93.89
88.85
77.58 -
91.4
83.69
78.47 -
83.58
81.18
75.76 -
82.93
80.27
69.27.
99.14
19.85-
99.14
84.98
Sand
0.00-
3.76
1.29
0.00-
4.91
1.08
0.00-
4.60
1.37
5.10-
17.68
8.09
0.00-
10.32
2.87
5.68-
9.36
7.40
5.77-
9.71
8.28
0.00-
17.68
0.00-
17.68
2.90
Shale
0.00-
4.40
0.63
0.00-
3.16
0.19
5.01-
30.73
15.18
0.00-
4.29
0.76
0.00-
0.00
0.00
7.86-
11.61
9.94
0.00-
0.00
0.00
0.00-
30.73
0.00-
30.73
3.89
Clay
0.00-
4.41
0.76
6.95-
22.67
14.54
0.00-
0.00
0.00
0.00-
2.63
0.26
0.00-
13.04
2.88
0.00-
1.53
0.31
5.37-
14.86
8.80
0.00-
22.67
0.00-
22.67
3.93
Iron
Ore
0.00-
2.77
1.11
0.00-
0.78
0.16
0.00-
1.44
0.38
0.00-
2.24
0.64
0.00-
1.24
0.50
0.00-
2.36
1.16
0.00-
1.99
0.89
0.00-
2.77
0.00-
2.77
0.63
Ash'
0.00-
4.80
0.29
0.00-
4.29
0.43
0.00-
185
0.22
0.00-
1.61
0.16
8.21-
13.27
10.01
0.00-
0.00
0.00
0.00-
1.20
0.30
0.00-
13J7
0.00-
13J7
0.96
Mai
Scale
0.00-
0.72
0.08
0.00-
1.66
0.18
0.00-
0.00
0.00
0.00-
0.72
0.14
0.00-
0.25
0.05
0.00-
0.07
0.01
0.00-
2.25
0.56
0.00-
2.25
0.00-
2.25
0.11
Gypsum
0.00-
3.58
0.36
0.00-
2.28
0.15
0.00-
3.00
0.35
0.00-
2.56
0.27
0.00-
0.00
0.00
0.00-
0.00
0.00
0.00-
0.00
0.00
0.00-
3.58
0.00-
10.42
OJ5
Other
0.00-
3.27
0.44
0.00-
1.95
0.19
0.00-
0.39
0.03
0.00-
0.00
0.00
0.00-
0.00
0.00
0.00-
0.00
0.00
0.00-
3.62
0.91
0.00-
3.62
0.00-
3.62
0.19
1 Includes bottom ash, fly ash, and alumina ash.
b Includes facilities whose primary feeds were limestone and ash; limestone, ash, and sand; limestone, ash, and
clay, and limestone, ash, and slag.
c Reflects feed mixtures for the 79 facilities for which data were available and includes facilities that do not
fall into one of the primary feed categories listed above (i.e., facilities from the "other" category shown in Exhibit 2-14).
Source: Facility responses to the 1991 PCA survey.
-------
2-39
Available data indicate that the cement industry consumes almost 1.6 metric tons of raw
material per ton of clinker produced. The portion of raw material that does not become clinker
is either lost on ignition or becomes CKD. Exhibit 2-15 shows typical ignition losses for each
material type. Many of the feed materials, including limestone, can contain significant quantities
of trace metals. Consequently, trace metal levels from feed materials in CKD can vary from
plant to plant based on differences in feed mixture components. As stated above, the occurrence
and impacts of trace metals are discussed further in Chapters 3, 5, and 7 of this report.
Fuels
The process of producing cement requires tremendous amounts of energy. The cost of
this energy can constitute as much as 40 percent of total production costs for a cement facility.82
The most energy-intensive part of the process is the maintenance of adequate temperatures
inside the cement kiln.83 As indicated in the discussion of the manufacturing process, material
temperatures inside the kiln reach as high as 1,510°C (2,750°F). The heat source (combustion
gas) must therefore be hotter than these materials, or hotter than the temperatures needed to
form cement. In 1990, the average energy required for a kiln to produce one kg of clinker was
1,245 Reals.84
The U.S. Bureau of Mines estimates that in 1990, the domestic cement industry
consumed 71 trillion Kcals.85 Industry responses to the 1991 PCA survey, however, suggest that
this figure may be an underestimate. For example, 71 facilities that provided complete fuel
information in the survey consumed an estimated 52 trillion Kcals. Also, the energy required to
manufacture the nearly 70 million metric tons of clinker produced in 1990 at the average of 1,245
Kcals/kg (as estimated from the PCA surveys) would be over 81 trillion Kcals.
Clearly, the elevated combustion temperatures involved in cement production require
fuels with a high heat content. Furthermore, the large volume of fuel consumed by the industry
necessitates fuels that are available in large quantities at reasonable cost. Historically, these
circumstances have dictated an almost exclusive reliance on fossil fuels of one type or another.
Exhibits 2-17, 2-18, and 2-19 characterize fuel consumption by the cement industry in 1985 and
1990. These exhibits show that coal and, to a lesser extent, other fossil fuels have been and
continue to be the primary fuels burned in most cement kilns. However, there has been a trend
toward exploiting other, lower cost fuel alternatives, particularly waste fuels. In particular, a
number of cement kilns across the country now fire non-hazardous solid wastes and/or hazardous
waste liquids and solids as fuel. The following sections discuss the use of fossil fuels, solid waste
fuels, and hazardous waste fuels in more detail.
82 Johnson, W., 1992, op. at.
0 Smith, J.D., 1990. Cement Kilns 1990, El Digest, Environmental Information, Ltd. June 1990.
84 Fuel consumption statistics for this report were derived from facility responses to the 1991 PCA survey. For
more detailed information on the derivation of these statistics, refer to the Technical Background Document. These
units were converted to Kcals from Btus as reported.
85 Johnson, W., 1992, op. tit.
-------
2-40
Exhibit 2-17
Cement Kiln Fuel Consumption in 1990 and 1985
Fuel Input
Coal
Coke
Hazardous
waste
Natural Gas
Oil
Solid waste
Other
Total
1990 Consumption1
Total Consumption'
5,548,250 metric
tons
866,732 metric tons
NA
384 km'
43,551 kiloliters
NA
NA
—
Energy
Equivalen
t (million
Kcals)
36,969,297
6,820,688
3,535,117
3,455,517
424,025
491,003
54,959
51,750,606
Percent
of
Total
71.4%
13.2%
6.8%
6.7%
0.8%
0.9%
0.1%
100.0%
1985 Consumption1"
Total Consumption
5,838,240 metric
tons
518,274 metric tons
NA
121km5
12,452 kiloliters
NA
NA
—
Energy
Equivalent
(million
Kcals)
38,901,568
4,088,875
563,693
1,087,839
124,253
2,666
35,626
44,804,520
Percen
t
of
Total
86.9%
9.1%
1.3%
2.4%
0.3%
0.0%
0.1%
100.0%
* Reflects 1990 fuel consumption for the 71 facilities for which data were available.
b Reflects 1985 fuel consumption for the 65 facilities for which data were available.
c The units in this table have been converted to metric from standard as reported.
NA: Data not available.
Source: Facility responses to the 1991 PCA survey.
-------
2-41
Exhibit 2-18
Cement Kiln Fuel Mixtures in 1990
Primary Fuel(s)'
Coal
Coal, Coke
Coal, Coke, Hazardous Waste
Coal, Hazardous Waste
Coal, Natural Gas
Natural Gas, Coke
Natural Gas
Coal, Natural Gas, Coke
Coke, Hazardous Waste
Coal, Solid Waste
Coal, Oil
Solid Waste
Natural Gas, Oil, Coke
Natural Gas, Solid Waste
Total
Number
of
Hants11
39
11
6
6
5
3
2
2
2
1
1
1
1
1
81
Supplemental Fuel' (Number of Plants Utilizing)
Coal
--
—
—
—
—
2
0
—
0
—
—
0
0
0
2
Natural
Gas
19
5
3
2
»
--
—
—
2
1
0
1
..
--
33
Oil
17
1
2
3
1
0
1
0
0
0
-
0
—
0
25
Coked
4
—
-
1
2
—
0
--
0
0
0
0
-
0
7
Solid
Waste
8
3
2
1
1
0
1
1
0
«
0
—
0
--
17
Hazardous
Waste
2
1
—
—
1
0
0
0
—
0
0
0
0
0
4
Other*
3
0
2
0
0
0
0
0
0
0
0
0
0
0
5
1 For purposes of this analysis, primary fuels are defined as those inputs that make up ten percent or more
of the total heat value input.
b This exhibit presents information only for the 81 facilities for which 1990 fuel input information was
available.
c For purposes of this analysis, supplemental fuels are those inputs that make up less than ten percent of the
total heat value fed.
d Includes petroleum coke.
e Other fuels utilized in 1990 included: coke dust, carbon black, re-refined oil, carbon dust, and propane.
Source: Facility responses to the 1991 PCA survey.
-------
2-42
Exhibit 2-19
Cement Kiln Fuel Mixtures in 1985
Primary Fuel(s)*
CoaJ
Coal, Coke
Coal, Natural Gas
Coal, Natural Gas, Coke
Natural Gas, Coke
Natural Gas
Coal, Hazardous Waste
Coal, Coke, Hazardous Waste
Natural Gas, Oil, Coke
Natural Gas, Coke, Hazardous
Waste
Total
Number
of Plants'1
49
13
4
3
1
1
1
1
1
1
75
Supplemental Fuel' (Number of Plants Utilizing)
Coal
—
-
—
--
0
0
-
--
0
1
1
Natural
Gas
21
7
-
—
-
-
0
1
--
--
29
Oil
17
3
2
0
1
0
0
0
-
0
23
Coked
6
-
1
«
--
0
0
--
-
--
7
Solid
Waste
1
2
0
0
0
0
0
0
0
0
3
Hazardous
Waste
3
1
0
0
0
0
"
--
0
"
4
Other1
4
0
0
0
0
0
0
0
0
0
4
* For purposes of this analysis, primary fuels are defined as those inputs that make up ten percent or more
of the total heat value input.
b This exhibit presents information only for the 75 facilities for which 1985 fuel input information was
available.
c For purposes of this analysis, supplemental fuels are those inputs that make up less than ten percent of the
total heat value fed.
d Includes petroleum coke.
e Other fuels utilized in 1985 included: coke dust, sublime pitch, re-refined oil, carbon dust, and propane.
Source: Facility responses to the 1991 PCA survey.
-------
2-43
Fossil Fuels
Although the industry is now utilizing increasing amounts of alternative fuels, most
cement kilns continue to combust traditional fossil fuels to produce the enormous process heat
required for cement production. Fossil fuels still accounted for more than 90 percent of the total
energy consumed in cement kilns in 1990. In contrast to the trend in the 1970's and early 1980's,
when rising oil prices resulted in coal displacing oil as kiln fuel, there has been a recent trend
away from coal toward other types of fossil fuels (coke, natural gas, and oil) and toward waste
fuels. Exhibits 2-17, 2-18, and 2-19 show that coal's share of cement kiln fuel consumption
declined from 1985 to 1990, both in terms of total energy provided and in the percentage of
facilities using it. Coal, however, still remained the most commonly used fuel.
Most cement plants firing coal as a primary fuel obtain the coal from a local source.
Data from the 1991 Keystone Coal Industry Manual suggest that most kilns usually obtain coal
from within the same state or from the closest state with a sufficient coal supply. Coal is
primarily organic matter consisting of carbon, hydrogen, oxygen, nitrogen, and sulfur. The
composition of coal varies from place to place throughout the U.S. Coal is ranked according to
composition into classes ranging from anthracitic to bituminous to subbituminous to lignitic.
Exhibit 2-20 shows average coal composition by rank, and Exhibit 2-21 displays the geographic
distribution of each rank.
Coal can contain significant quantities of sulfur, trace metals, and halogens, and their
concentrations are dependent on the area in which the coal was mined. Consequently,
contaminant levels attributable to coal in CKD can vary from plant to plant based on regional
differences in coal composition. The occurrence of trace metals is discussed further in Chapter 3
of this report. Sulfur (in the form of SO3) will vaporize in the kiln to form sulfur dioxide (SO2),
and condense in the form of sulfates. Within the kiln, these sulfates combine with calcium and
potassium, causing operational problems in the cool end of the kiln.86 Halogens are of concern
because chlorides can cause operational problems similar to those caused by sulfur. Chlorine
concentrations in coal can range from 100 to 2,800 parts per million.87
The other fossil fuels utilized in cement kilns include coke, natural gas, and oil. Coke is
the solid, cellular, infusible material remaining after the carbonization of coal, pitch petroleum
residues, and certain other carbonaceous materials.88 The coke used by cement kilns is typically
petroleum coke. Natural gas is a naturally occurring mixture of hydrocarbon and
nonhydrocarbon gases found in the porous geologic formations beneath the earth's surface.
Processed natural gas is principally methane, with small amounts of ethane, propane, butane,
pentane, carbon dioxide, and nitrogen.89 Because of its high heat transfer rate, natural gas is
used to perform initial firing of kilns at many cement plants. When the kiln reaches operating
temperature, the primary fuel is then brought on-line. Coke, natural gas, and oil are considered
"cleaner" than coal because they contain less sulfur per Kcal provided.
K Peray, 1986, op. at.
87 Environmental Toxicology International, 1992, op. at.
K Perry's Chemical Engineers' Handbook, 1984, op. at.
89 Kirk-Othmer, 1979, op. at.
-------
2-44
-------
2-45
Exhibit 2-20
Average Coal Composition by Rank
Real (per kg.)
Moisture
Volatile matter
Fixed carbon
Ash
Hydrogen
Carbon
Nitrogen
Oxygen
Sulfur
Sulfate sulfur
Pyritic sulfur
Organic sulfur
Average
6,210
10.0
29.9
48.8
11.3
5.1
64.1
1.1
16.4
2.0
0.12
1.19
0.70
Anthracite
7,100
1.4
6.5
79.5
12.6
2.4
80.1
0.8
3.2
0.8
0.02
0.35
0.48
Bituminous
6,820
4.8
32.3
51.2
11.7
5.0
69.1
1.3
10.3
2.7
0.16
1.70
0.88
Subbituminous
5,230
18.4
33.8
39.0
8.8
5.9
54.3
1.0
29.3
0.7
0.04
0.35
0.32
Lignite
2,780
41.5
23.0
20.9
14.6
6.8
29.9
0.5
46.5
1.7
0.24
0.68
0.75
(All values except Kcals in weight percent)
* These units have been converted to Kcal/kg from Btus/lb as reported.
Sources: Swanson, V.E., et ah, 1976. Collection, Chemical Analysis, and Evaluation of Coal Samples in 1975, U.S.
Geological Survey, Open File Report 76-468.
Exhibit 2-21
Coal Producing Regions of the United States and Coal Types Produced
Region
Eastern Province (Alabama, Kentucky, Maryland, Ohio,
Pennsylvania, Tennessee, Virginia, West Virginia)
Interior Province (Arkansas, Blinois, Indiana, Iowa, Kansas,
Kentucky, Michigan, Missouri, Nebraska, Oklahoma)
Gulf Province (Alabama, Arkansas, Mississippi, Texas)
Northern Great Plains Province (Montana, North Dakota,
Wyoming)
Rocky Mountain Province (Arizona, Colorado, New Mexico,
Utah, Wyoming)
Pacific Coast Province (Washington)
Coal Ranks Produced
Primarily anthracite, with large deposits
bituminous
of
Medium- to high-volatile bituminous, with smaller
volumes of low-volatile bituminous and anthracite
Lignite
Lignite, subbituminous
Bituminous, subbituminous
Bituminous, subbituminous, lignite
Source: Magee, E.M., et ah, 1973. Potential Pollutants in Fossil Fuels, prepared for U.S. Environmental Protection
Agency, Esso Research and Engineering Co., Linden, New Jersey. June 1973.
-------
2-46
Exhibit 2-22 compares relative costs per unit of energy of each of the fossil fuels. Coal
(or at least bituminous, subbituminous, and lignite coal) is inexpensive compared to the other
fuel types, which certainly contributes to its dominance as the primary energy source for cement
kilns. However, natural gas prices have fallen much more rapidly than coal prices over the last
five years. These reduced prices, combined with the lower sulfur content of natural gas, may
help explain its increasing share of the kiln fuel market. Moreover, domestic supplies of both
coal and natural gas are abundant, providing a more secure long-term fuel source, as compared
to oil.
Exhibit 2-22
Nominal Fossil Fuel Prices in 1985 and 1990
Fuel
Coal (Anthracite)
Coal (Bituminous,
Subbituminous, and Lignite)
Natural Gas
Oil
Coke
1990 Price (cents
per million Kcals)
692.7
395.0
613.7
1,370.8
Not Available
1985 Price (cents
per million Kcals)
810.6
455.7
896.0
1,648.7
Not Available
Percent Change
-14.5
-13.3
-31.5
-16.9
-
Source: Energy Information Administration, 1992. Annual Energy Review 1991. June 1992.
Non-hazardous Waste Fuels
In an effort aimed largely at reducing production costs, the cement industry has been
actively investigating alternative fuel sources. As a result, there has been a significant increase in
the use of wastes, both hazardous and non-hazardous, as fuels by cement kilns. Three facilities
reported using small amounts of non-hazardous solid waste90 as supplemental fuels in 1985. By
comparison, 20 facilities reported using solid waste as either primary or supplemental fuel in
1990. One of these facilities even obtained the majority (over 70 percent) of the energy it
consumed from solid waste (waste oil). The primary types of non-hazardous solid waste used as
fuel in 1990 were used tires and waste oil. One facility reported using waste wood chips as fuel.
Exhibit 2-23 further details the consumption of solid waste fuel by type. Reportedly, the industry
is also evaluating other types of non-hazardous solid waste, including almond shells and
municipal solid waste, for use as kiln fuel.91 Although available waste fuel price data are
extremely limited, it is EPA's understanding that the burning of such alternative fuels can be
done at zero net cost or even at a small profit to cement plant operators.
90 The term, "solid waste", is used throughout this analysis in its regulatory sense. The term does not have a direct
relationship to the physical form of the material and includes sludges and liquid wastes (such as waste oil) as well as
solids.
91 Portland Cement Association, 1992a, op. at.
-------
2-47
Exhibit 2-23
Breakdown of Solid Waste Fuel Consumption0
Solid Waste
Tires
Waste OU
Wood Chips
Total
1990 Consumption*
Total
Consumption
69,228 metric tons
19,148 kilolitere
943 metric tons
-
Energy
Equivalent
(million
Kcals)
278,391
207,841
4,792
491,024
Percent
of
Total
56.7%
42.3%
1.0%
100.0%
1985 Consumption1"
Total
Consumption
0
3 14 kilolitere
0
-
Energy
Equivalent
(million
Kcals)
0
2,666
0
2,666
Percent of
Total
0.0%
100.0%
0.0%
100.0%
1 Reflects 1990 fuel consumption for the 71 facilities for which data were available.
k Reflects 1985 fuel consumption for the 65 facilities for which data were available.
c These units have been converted to metric units from the standard units reported.
Source: Facility responses to the 1991 PCA survey.
Hazardous Waste Fuels
In addition to increasing consumption of solid waste fuels, cement kiln operators have
substantially increased the consumption of hazardous wastes as fuel, which has come to account
for a significant portion of fuel consumption by cement kilns. Available data indicate that
hazardous waste fuels now supply as much of the total energy consumed as natural gas (see
Exhibit 2-17). The remainder of this section discusses the hazardous waste fuels industry in
general, the extent of hazardous waste consumption by kilns, the types of hazardous waste
burned, the technologies used to feed the waste to kilns, and environmental regulations
applicable to this practice. The section concludes by highlighting some of the subjective
arguments made by supporters and opponents of burning hazardous wastes in cement kilns.
Overview of the Hazardous Waste Fuels Industry
As part of EPA's RCRA Subtitle C Land Disposal Restrictions (LDR) Program (see 40
CFR Part 268), EPA has established treatment standards that must be met before a hazardous
waste may be land-disposed. These standards are based on the performance of the Best
Demonstrated Available Technology (BOAT). For most non-aqueous wastes contaminated with
organic constituents (e.g., solvents, petroleum refining wastes), EPA has determined that
incineration is the BDAT. Use as fuel in boilers or industrial furnaces (BIFs) (e.g., cement kilns)
is an acceptable means of complying with LDRs for many of these wastes. Because of the
significant demand for treatment of hazardous wastes, the limited availability of and high price of
using commercial hazardous waste incinerators, and the sizable fuel requirements of cement
-------
2-48
kilns, cement kilns can successfully compete with commercial incinerators in the waste treatment
market, and have become a major component of the commercial hazardous waste management industry.
Because the business of managing hazardous wastes is quite different from the business of
manufacturing cement, and because cement plants are often located far from waste generators,
many cement companies rely on intermediate fuel processors to supply them with waste fuels.
These fuel processors may be located at the cement plant site, but more often they are not. The
fuel processor typically accepts or collects wastes from many individual waste generators and
stores and blends wastes to meet cement kiln specifications. In some cases, these fuel processors
also manage or participate in hazardous waste fuel management activities (e.g., sampling and
recordkeeping) at the cement facility. Waste fuels may be stored in tanks or in container storage
areas prior to being fed to the kiln or, in some cases, may be fed directly from tanker trucks.
Extent of Hazardous Waste Consumption by Cement Kilns and Types of Waste Burned
Environmental Information Digest lists 23 cement plants that were routinely burning
hazardous waste in 1990.92 Eighteen of the 81 facilities that provided fuels information in
response to the 1991 PCA survey reported burning hazardous waste as either primary or
supplemental fuel. As noted above and shown in Exhibits 2-17, 2-18, and 2-19, the practice of
burning hazardous waste in cement kilns has increased significantly during the last five years,
both in terms of the number of facilities utilizing hazardous waste and in terms of percent of
total energy consumed.
Exhibit 2-24 provides information on the physical form of hazardous wastes burned in
cement kilns. Although liquid wastes comprise most of the hazardous wastes consumed, cement
plant operators also have expanded their consumption of hazardous wastes in solid form in
recent years. In
Exhibit 2-24
Breakdown of Hazardous Waste Fuel Consumption
User name:
Hazardous Waste
Physical Form
IS3ASSO (43)
1990 Consumption'
Energy Equivalent
(million Kcals)
Queue: R6HASSB.SSS6
Percent of Total
13100582D *3.5%
1985 Consumption1"
Energy Equivalent
(million Kcals)
Percent of Total
0
Directory:
Description: LPT1 Catch
January 13,94
8:57am
***********************
******************************
RRRR CCC A SSS SSS OOO
R RC C AA S SS SO O
R RC A AS S O O
RRRR C A A SSS SSS O O
RR C AAAAA S SO O
R R C CA AS SS SO O
R R CCC A A SSS SSS OOO
***********************************************************
L SSS T1T1T
L S S T
L S T ::
-------
L SSI ::
LLLLL SSS T ::
********************************************************************************
-------
0.0%
Liquid
Sludge
Total
3,411,450
0
3,535,272
96.5%
0.0%
100.0%
563,718
0
563,718
100.0%
0.0%
100.0%
92 Smith, J.D., 1991. Cement Kilns 1991, El Digest, Environmental Information, Ltd. August 1991.
-------
2-49
1 Reflects 1990 fuel consumption for the 71 facilities for which data were available.
b Reflects 1985 fuel consumption for the 65 facilities for which data were available.
c These units have been converted to Kcals from Btus as reported.
Source: Facility responses to the 1991 PCA survey.
-------
2-50
addition, though several facility operators have reported that they have the capacity to burn
hazardous waste sludges,93 none (in the 1991 PCA survey) reported burning waste in this form.
Hazardous Waste Fuel Technologies
Several different technologies are employed to feed hazardous waste fuels into cement
kilns. These different methods are important in that they dictate the types of wastes that a kiln
can burn (i.e., liquid, sludge, or solid) and the location of the waste entry point relative to the
flame in the kiln. In general, there are four important types of feed systems:
u Liquid feed systems, whereby liquid waste fuels are fed either through primary
(fossil fuel) fuel ports or through similar ports in the hot end of the kiln;
u Sludge feed systems, which are similar to liquid feed systems except that they have
higher solids, particle size, and viscosity tolerances;
u Dry solids systems, in which dry, finely divided waste particles (like coal dust) are
blown into the flame area of the kiln; and
u Container feed systems, in which small buckets or bags are fed to the mid-section
or calcining zone of the kiln. In wet process kilns, containers are fed at the mid-
point of the kiln, while in preheater/precalciner kilns, containers are fed at the
cool end, between the preheater and the rotating kiln body.
Relevant Environmental Regulations
Until August 1991, the burning of hazardous waste as fuel in cement kilns was exempt
from RCRA permitting requirements under the premise that this practice constituted recycling.
As a guideline for identifying recycling, EPA established a minimum waste heating value limit of
2,780 Kcal/kg. Hazardous waste fuel burners meeting this limit were required to notify EPA of
their activities, obtain an EPA identification number, and maintain certain records (40 CFR
266.36).
On February 21,1991 (56 FR 7134), EPA promulgated new rules to control the burning
of hazardous waste in boilers and industrial furnaces. The so-called Boiler and Industrial
Furnace (BIF) Rule controls the emissions of toxic organic compounds, toxic metals, hydrogen
chloride (HC1), chlorine gas (C12), and paniculate matter from boilers and industrial furnaces
that burn hazardous wastes. In particular, the BIF Rule establishes, at 40 CFR Part 266, Subpart
H, the following regulatory compliance criteria:
u Demonstrate 99.9999 percent destruction or removal of dioxin-listed wastes and
99.99 percent destruction or removal of all other hazardous organic constituents;
u Adhere to limits on carbon monoxide and/or hydrocarbon flue gas concentrations;
93 U.S. Environmental Protection Agency, 1990. Commercial Combustion Capacity for Hazardous Waste Sludges
and Solids, Office of Solid Waste, Washington, B.C. August 1990.
-------
2-51
u Meet risk-based (i.e., health-based) emission limits for four carcinogenic metals
(arsenic, beryllium, cadmium, and chromium), six noncarcinogenic metals
(antimony, barium, lead, mercury, silver, and thallium), HC1, and C12. Under
EPA's tiered approach to metal standards, higher emissions rates and feed rates
are permitted as more detailed, site-specific emissions testing and dispersion
modeling are conducted;
u Limit paniculate matter emissions to 0.08 gr/dscf, corrected to 7 percent oxygen
(the same standard required of hazardous waste incinerators); and
u Comply with general facility standards specified under 40 CFR 264 for hazardous
waste treatment, storage, and disposal facilities.
Affected facilities were given until August 21,1991 to notify EPA of their waste burning
practices and their plans for certifying precompliance with the rule. With few exceptions, BIF
operators were required to certify compliance with interim status standards by August 21,1992.
The rule did not place a deadline on final permit decisions. [Note to reader: EPA recently
granted a stay to the two-part test.]
23 SUMMARY AND RELEVANCE TO SUBSEQUENT CHAPTERS
This chapter presented an overview of the cement industry and the basic cement
manufacturing process. The discussion of the industry's structure and description of kiln
technology presented herein will assist the reader in understanding the issues raised in
subsequent chapters of this report. Specifically, the overview of the cement industry serves as an
introduction that will result in a clearer understanding of the economic impacts of potential CKD
regulation (see Chapter 9). The discussion of production technologies provides the reader with
the industry-specific knowledge needed to understand the current CKD collection, recycling, and
management practices discussed in Chapters 3 and 4 and management alternatives presented in
Chapter 8. Finally, the last section of this chapter on process inputs will help the reader
understand the connection between raw material and fuel inputs to cement making and the
chemical characteristics of CKD, a topic that is discussed more fully in Chapter 3.
-------
2-52
CHAPTER TWO
CEMENT INDUSTRY OVERVIEW
2.0 INTRODUCTION AND DESCRIPTION OF CKD 1
2.1 CEMENT INDUSTRY STRUCTURE AND CHARACTERISTICS 1
2.1.1 What Is Cement? 1
2.1.2 The Cement Industry 5
2.2 CEMENT MANUFACTURING 9
2.2.1 The Basic Production Process 10
Mining 10
Crushing 10
Grinding and Blending 13
Pre-Drying 13
Drying and Preheating 13
Calcining 15
Burning 15
Cooling 15
Finish Milling/Loading 16
2.2.2 Kiln Design 16
Wet Process Kiln 18
Dry Process Kiln 19
Preheater Kiln 21
Precalciner Kiln 23
Semidry Process Kiln 23
Trends in Kiln Technology and Use 25
-------
2-53
2.2.3 Process Inputs 26
Raw (Feed) Materials 26
Fuels 33
Fossil Fuels 37
Non-hazardous Waste Fuels 39
Hazardous Waste Fuels 40
Overview of the Hazardous Waste Fuels Industry 40
Extent of Hazardous Waste Consumption by Cement Kilns and Types of
Waste Burned 41
Hazardous Waste Fuel Technologies 42
Relevant Environmental Regulations 42
2J SUMMARY AND RELEVANCE TO SUBSEQUENT CHAPTERS 43
-------
2-54
Exhibit 2-1
Exhibit 2-2
Exhibit 2-3
Exhibit 2-4
Exhibit 2-5
Exhibit 2-6
Exhibit 2-7
Exhibit 2-8
Exhibit 2-9
Exhibit 2-10
Exhibit 2-11
Exhibit 2-12
Exhibit 2-13
Exhibit 2-14
Exhibit 2-15
Exhibit 2-16
Exhibit 2-17
Exhibit 2-18
Exhibit 2-19
Exhibit 2-20
Exhibit 2-21
LIST OF EXHIBITS
General Cement Types and Uses 2
1990 Clinker Production by State 6
Geographic Distribution of Active Portland Cement Production Facilities
in 1991 in the U.S 7
U.S. Clinker Capacity for the Period 1973-1991 9
Steps in the Manufacture of Portland Cement 11
Material Temperature Ranges in the Kiln Portion of the Cement
Manufacturing Process 14
Typical Energy Requirements for Each Kiln Type 18
Typical Process Flow Diagram for a Wet Process Cement Kiln 20
Typical Process Flow Diagram for a Suspension Flash Preheater Cement
Kiln
22
Typical Process Flow Diagram for a Precalciner Coal and Waste Fired
Cement Kiln 24
Age and Capacity of Existing Kilns 25
Feed Mixtures in 1990 28
Feed Mixtures in 1985 29
Raw Material Consumption in 1990 and 1985 30
Typical Composition of Raw Materials 31
Typical Feed Mixtures in 1990 32
Cement Kiln Fuel Consumption in 1990 and 1985 34
Cement Kiln Fuel Mixtures in 1990 35
Cement Kiln Fuel Mixtures in 1985 36
Average Coal Composition by Rank 38
Coal Producing Regions of the United States and Coal Types Produced .... 38
-------
2-55
Exhibit 2-22 Nominal Fossil Fuel Prices in 1985 and 1990 39
Exhibit 2-23 Breakdown of Solid Waste Fuel Consumption 40
Exhibit 2-24 Breakdown of Hazardous Waste Fuel Consumption 41
-------
CHAPTER THREE
CKD GENERATION AND CHARACTERISTICS
3.0 INTRODUCTION
Cement production processes in current use in the U.S. generate CKD as an intrinsic
process residue. During cement production, kiln combustion gases flow countercurrent to the
raw feed and exit the kiln under the influence of induced draft fans. The rapid gas flow and
continuous raw feed agitation are turbulent processes that result in large quantities of paniculate
matter being entrained in the combustion gases. The entrained paniculate matter (as well as
various precipitates) is subsequently removed from the kiln exhaust gases by air pollution control
equipment; this paniculate matter constitutes CKD.
For purposes of this report, as discussed in greater detail later in this chapter, gross CKD
constitutes the dust collected at the air pollution control device(s) (APCDs) associated with a
kiln system. Gross CKD is generated as an inherent process residue at all cement plants, though
the fate of this material varies by facility. After collection, gross CKD is either recycled back to
the kiln system or removed from the kiln system as net CKD. Net CKD is either treated and
returned to the kiln system, disposed in an on-site waste management unit, or sold or given away
for beneficial use. Exhibit 3-1 illustrates the potential management pathways for gross CKD.
Although a number of plants directly recycle all gross CKD back to the kiln system, most plants
remove a significant quantity of CKD from the system, for subsequent treatment and recycle
back into the kiln, disposal, or for beneficial use.
Although CKD generation is unavoidable, the amount and characteristics of the dust that
is generated (and the degree to which it is recycled or reused) can be influenced by several
factors. These factors include kiln type, cement production rate, raw feed material types and
proportions, fuel type(s), and the types and numbers of APCDs employed. Through variations in
these factors, many facilities recycle some portion of their generated dust back to the kiln. This
chapter presents information on CKD generation rates and characteristics (including current
recycling practices and limitations on CKD that is returned to the kiln system) and describes the
physical and chemical characteristics of CKD. Finally, for the interest of the reader, this chapter
also presents information on the characteristics of clinker materials.
3.1 CKD GENERATION
As mentioned above, gross CKD generation is defined in this report as the collection of
dust via APCDs from cement kiln exhaust gases. This definition excludes that portion of
generated CKD that passes the APCDs and exits the kiln system with the exhaust gases. Based
on typical APCD efficiency standards, generally between 98 and 100 percent of all paniculate
matter is captured before exiting the kiln system.1
1 Engineering-Science, 1987. Background Information Document For The Development of Regulations To
Control The Burning of Hazardous Wastes In Boilers and Industrial Furnaces. Vol. II: Industrial Furnaces. January,
1987. p. 3-47.
-------
3-2
Exhibit 3-1
Flow Chart of Gross CKD Management Pathways
Based on an extrapolation of the data supplied to EPA by respondents to the 1991 PCA
Survey, which represents data from 64 percent of active U.S. cement kilns,2 the U.S. cement
industry generated an estimated 12.9 million metric tons (14.2 million tons) of gross CKD and
4.6 million metric tons of net CKD in 1990. Operators of U.S. kilns recycled about 8.3 million
metric tons, or 64 percent, of the gross CKD. This section discusses how CKD is collected,
provides information on plant level CKD generation rates, and finally, addresses factors that may
affect gross CKD generation, recycling rates, and net CKD generation.
3.1.1 Dust Collection Devices
APCDs are used to limit dust emissions from the kiln system to the atmosphere. The
combustion gases that exit the kiln consist primarily of carbon dioxide, water, fly ash (i.e., fine
solid particles of ashes, dust, and soot from burning of fuels), sulfur, and nitrogen oxides. The
components of these gases are derived from the combustion of fuels, contaminants (organic and
2 Operators of 144 kilns provide usable data in response to the PCA Survey, there are approximately 225 kilns in
the U.S. industry.
-------
3-3
inorganic) in the kiln solids, small particles of feed and clinker material, and (for wet kilns) slurry
water. After passing through the air pollution control system, the remaining combustion gases,
which are discharged through a stack, consist primarily of carbon dioxide and water.3
Undesirable contaminants (in terms of clinker quality) may volatilize in the burning zone of the
kiln and precipitate as alkalies, sulfates, and chlorine compounds to become part of the CKD.4
CKD as collected is a fine-grained, solid, highly alkaline material that is generated at a
temperature near l,482eC (2,700CF). These characteristics tend to limit the types of dust
collection devices that can be used to control air pollutant emissions from cement kilns. For
example, because its fine-grained nature (diameter ranging from near zero micrometers or
microns [/im] to greater than 50 /xm) allows CKD to be easily entrained in exhaust gases, settling
chambers that rely on gravity to separate particulate matter from a gas stream can only be used
as a primary dust collection device to remove coarse dust particles and, in general, must be
combined with more complex devices such as fabric filters (i.e., baghouses) or electrostatic
precipitators. Wet scrubbers, commonly used in many mineral processing industries, cannot be
used in the cement industry because adding water to the captured CKD causes it to harden ("set
up") due to its cementitious properties.
The predominant APCDs in use at cement plants are electrostatic precipitators (ESPs)
and fabric filters arrayed in baghouses. Both are often preceded by one or more cyclones.
Additional APCDs include gravity/inertial separators and granular bed filters. Dust collection
systems at cement plants may involve a combination of the above units. These systems typically
remove dust at an efficiency ranging from 98 to nearly 100 percent.5 Each of these technologies
is described below in Exhibit 3-2 and is illustrated in Exhibit 3-3.
Dust collection systems are sensitive to the temperature of the inlet gases because very
low or high temperatures may damage APCD components; the moisture and sulfur content of
the gases require that the temperature be controlled within a set range. For example, moisture
can condense in a baghouse or in an ESP when the temperature falls below the dewpoint of the
gases. Such condensation can cause plugging problems and result in corrosion of the dust-
collection equipment. A conservative minimum dust collector inlet temperature should be 176°C
to allow for the additional temperature drop that may occur within a baghouse. At the other
end of the range, the temperature of gases passing through most baghouses cannot exceed 299°C
before damage to the filters occurs.6
Exhibit 3-4 summarizes the quantities of CKD collected in 1990 by APCD type among
respondents to the 1991 PCA Survey. APCD types are characterized according to the four
choices provided in the survey: baghouse, multiclone, ESP, or other APCD. Some respondents
reported quantities for systems with two different APCD types. This exhibit shows that ESPs
and baghouses in isolation are the predominant APCDs used by facilities. These two types of
devices collected approximately 65 percent of the CKD generated by the 123 kilns for which
survey responses are available. In contrast, 23 percent of the survey respondents collected CKD
J Engineering-Science, 1987, op. cit., p. 3-4.
4 Kohlhaas, B., el at., 1983. Cement Engineer's Handbook. Bauverlag GMBH, Wiesbaden and Berlin, p. 624.
5 Engineering-Science, 1987, op. cit.t pp. 3-47 and 3-49.
6 Peray, Kurt E., 1986. The Rotary Cement Kin. Chemical Publishing Co., Inc. New York, NY. p. 172.
-------
3-4
with a multiclone in combination with a baghouse in 1990. Other combinations accounted for
the remaining 12 percent of the CKD collected. Very few survey respondents indicated "other"
to describe their APC system.
-------
3-5
Exhibit 3-2
Air Pollution Control Devices Used at Cement Kilns*-b
APCD
Description
Electrostatic Precipitators
Fabric Filters
Cyclones
Gravity llnertial Separators
Granular Bed Filters
One or more high intensity electrical fields are generated and cause
particles to acquire an electrical charge. These charged particles migrate
to a collecting surface that has the opposite electrical charge. The
collecting surface may be wet or dry. Facility operators then retrieve the
captured CKD. One advantage of this technology is that flow is not
restricted during collection. Collection efficiencies can be as high as
99.75 percent.
Filters remove particulate matter from gas streams by retaining the
particles in a porous structure, and are typically used in series to form a
baghouse. The porous structure is generally a woven or felt fabric with a
retention efficiency that improves as the interstices fill with captured
dust, but with the negative effect of increased flow resistance. Thus,
regular filter cleaning is required to maintain efficiency. Baghouse filters
can also be constructed of siliconized glass fibers (i.e., fiberglass). Fabric
filters can remove submicron-sized particles at collection efficiencies as
high as 99.95 percent.
A vortex within a collector propels particles to deposition areas for
removal. Cyclones may be operated either wet or dry. They deposit the
collected particulate matter into a hopper for eventual collection.
Cyclones have collection efficiencies that range from 58 to 97 percent.
Multiple cyclones used as part of one unit are referred to as multiclones.
Multiclones have collection efficiencies that range from 85 to 94 percent
for dust particles with diameters of 15 to 20 microns.
These devices collect particulate matter by gravity or centrifugal force,
but do not depend upon a vortex as do cyclones. Examples include
settling chambers, baffled chambers, louvered chambers, and devices in
which the gas and particulate mixture passes through a fan. In general,
collectors of this type are of relatively low collection efficiency and are
frequently followed by other types of collectors. Gravity settling
chambers remove coarse dust particles at collection efficiencies ranging
from 30 to 70 percent.
Dust is captured and bound on a porous medium through the principle
of adsorption. The most commonly used medium is granular activated
carbon (GAC). Collection efficiencies have been reported to be as high
as 99.9 percent.
* Kohlhaas, B., et a/., 1983. Cement Engineer's Handbook. Bauverlag GMBH, Wiesbaden and Berlin, p. 635.
b Duda, W.H., 1976. Cement-Data-Book: International Process Engineering in the Cement Industry. Bauverlag
GMBH Wiesbaden and Berlin, pp. 403-417.
-------
3-6
Exhibit 3-3
Schematic Diagrams of Common Types of Air Pollution Control Devices
-------
3-7
Exhibit 3-3 (continued)
Schematic Diagrams of Common Types of Air Pollution Control Devices
-------
3-8
Exhibit 3-4
1990 Gross CKD Collection by Different Types of Air Pollution Control Devices*
APCD
ESP
Baghouse
Multiclone with
Baghouse
Baghouse with ESP
Multiclone with ESP
Baghouse and Other
Total
No. of
Facilities
with
APCD
Type
37
19
7
5
4
1
73
Percent
of
Facilities
50.7
26.0
9.6
6.8
5.5
1.4
100.0
Gross CKD Generation1' (Metric Tons)
CKD
Collected
3,578,934
1,662,784
2,128,021
678,642
552,804
112,945
8,714,130
Cumulative
CKD
Collected
3,578,934
5,241,718
7369,739
8,048381
8,601,185
8,714,130
8,714,130
Percent of
Total CKD
Collected'
41.1
19.1
24.4
7.8
6.3
1.3
100.0
Cumulative
Percent
Collected
41.1
60.2
84.6
92.4
98.7
100.0
100.0
• Based on the usable responses from 73 facilities reported in the PCA Surveys.
b This exhibit presents only APCD collection quantities, while the remainder of this chapter considers only
quantities of CKD reported as gross or generated CKD. A number of facilities reported gross CKD generation rates
that were different than the associated CKD collection rates reported for the APCDs.
c (Collected CKD by given APCD) * (Total CKD Collected) x 100
Responses in this category are either variations on one of the previously discussed APCDs (e.g.,
cyclones as opposed to multiclones) or consist of uncommon APCDs (e.g., fallout chambers).
3.1.2 Plant-Level CKD Generation Rates
To better understand the nature of contemporary CKD practices, EPA has performed an
extensive evaluation of plant-level gross and net CKD generation rates. Any significant patterns
with respect to CKD generation and in-line recycling could have important implications with
respect to EPA's analysis of the adequacy of current management practices and the feasibility of
CKD management alternatives.
As stated previously, CKD generation rates vary widely among facilities on both a gross
and net basis. These rates do not, however, necessarily vary in direct proportion with one
another. A scatter plot of gross versus net CKD generation for plants responding to the 199]
PCA Survey is presented in Exhibit 3-5. As this exhibit demonstrates, there is no clear, apparent
relationship between the amount of gross CKD generated and the amount of net CKD
generated, even within a given process type. Facilities that generate large quantities of gross
CKD do not necessarily generate large amounts of net CKD. Conversely, a facility may generate
moderate quantities of CKD on a gross basis, but may be one of the larger net CKD generators
by virtue of the fact that it recycles none of its
-------
3-9
Exhibit 3-5
Relationship Between Net and Gross CKD Generated in 1990
Source: 1991 PCA Survey Responses.
Note: For clarity of presentation, available data on the LaFarge Corporation, Alpena facility was excluded
from the above exhibit because it was an outlier. In 1990, this facility generated about 650 thousand metric tons of
gross CKD and about 430 thousand metric tons of net CKD.
-------
3-10
CKD. Therefore, gross CKD generation rate is not an accurate indicator of the magnitude of
waste management issues at individual cement facilities.
In comparison with the other kiln types, operators of preheater/precalciner kilns recycle
higher percentages of the gross CKD that they generate; this difference is especially apparent at
the higher gross CKD generation rates (i.e., more than 200,000 metric tons per year). Moreover,
while there is a considerable amount of scatter in the data, it appears that at lower gross CKD
generation rates (less than 200,000 metric tons per year), the distribution of points corresponding
to both the wet and the preheater/precalciner plants follows one of two patterns: 1) the points
fall along the X-axis (net CKD is zero or close to zero); or 2) the points fall along a diagonal
line (net CKD equals or is close to gross CKD), meaning that recycled CKD is or is close to
zero. Plants operating long dry kilns, in contrast, are more difficult to categorize with respect to
CKD generation trends.
To further examine industry-wide CKD generation rates and trends, EPA analyzed CKD
generation and management data comprising useable results from PCA Survey responses
covering 79 active U.S. cement plants; the data obtained have been tabulated and, for purposes
of presentation, split into ten groups of eight facilities.7 Exhibit 3-6 shows gross and net CKD
generation for these groups arrayed in descending order, by net CKD generation. The top ten
percent of facilities generated about 2.25 million metric tons of gross CKD in 1990, which was
two to three times more than the gross CKD generated by any other group (except for the
seventh decile), and was about 10 times higher than the gross CKD generated by the sixth decile.
The gross CKD generation rates of the plants in the remaining groups do not differ markedly,
with most generating between 0.5 and a little more than one million metric tons per year, with an
average of about 700,000 metric tons per year.
It is apparent from the relative heights of the bars on the right half of the diagram that
approximately one-half of all 79 plants in the sample directly recycle all, or almost all, of the
gross CKD that they generate. In general, it also would appear, based upon an examination of
the heights of the gross CKD bars of deciles 1-5 with those of deciles 6-10, that the gross CKD
generation rates of the groups with very high aggregate recycling rates are comparable to those
of the groups generating significant quantities of net CKD. Finally, it is clear that some facilities
in all of these groups recycle significant quantities of CKD, due to the substantial differences in
gross and net CKD generation rates in each group represented.
Plant-by-plant net CKD generation rates vary dramatically among facilities in the U.S.
cement industry. The top ten facilities together accounted for over 50 percent of the total net
CKD generated in 1990. In fact, the top three facilities alone accounted for close to 30 percent
of the net CKD generated during this period. To determine whether the share of net CKD
generated was simply a function of facility size or throughput, the Agency compared net CKD
generation to clinker capacity. Exhibit 3-7 shows the share of total net CKD and clinker
production capacity accounted for by each group of eight facilities. The top ten percent
accounted for almost 47 percent of net CKD generated in 1990. Although as a group these
facilities are also the top ten percent in terms of production capacity, they represent only about
15 percent of total industry capacity. In the remaining groups, no pattern with respect to net
CKD generation and production capacity emerges. For example, the seventh decile represents
close to 12 percent of capacity but only two percent of net CKD generation, whereas the second
7 The last group contains only seven facilities.
-------
3-11
decile represents less capacity (9.4 percent) but accounts for a substantially larger share of net
CKD generation (19.7 percent).
Exhibit 3-6
Gross and Net CKD Generated (1990)
Source: 1991 PCA Survey Responses.
Note 1: Data on gross CKD generated was not available for one data point in the fifth decile and one data
point in the eighth decile. Therefore, the gross CKD indicated for these two deciles is the total generated at only
seven of the eight facilities in each decile.
Note 2: The last decile contains only seven facilities.
-------
3-12
Exhibit 3-7
Share of Net CKD Generated and Clinker Production Capacity (1990)
Source: 1991 PCA Survey Responses.
Note 1: Data on clinker production capacity was not available for one data point in the third decile and one
data point in the last decile.
Note 2: The last decile contains only seven facilities.
-------
3-13
Exhibit 3-8 lists seven facilities whose share of net CKD generated in 1990 was three to
four times higher than their share of clinker production capacity. All seven facilities, not
surprisingly, fall in the top two deciles of net CKD generators. The facilities that display the
sharpest difference between shares of clinker production capacity and net CKD generation are
the LaFarge, Alpena, facility, which is also the largest generator of net and gross CKD, and the
Holnam, Ada, facility. Five of the seven facilities in the exhibit have wet process kilns. With
respect to fuel usage, there appears to be a fairly even distribution between hazardous and non-
hazardous waste burners. Five of the seven facilities displayed in Exhibit 3-8 are Holnam
facilities and the remaining two are owned by LaFarge Corporation.
Exhibit 3-8
Facilities With High Net CKD Generation Relative to Clinker Capacity
Facility
LaFarge, Alpena
Holnam, Holly Hill
Holnam, Clarksville
Holnam, Ada
Holnam, Florence
Holnam, Fort
Collins
LaFarge, Fredonia
Share of Clinker
Capacity (1)
3.28
2.15
2.22
1.06
0.82
0.77
0.65
Share of Net CKD
Generated (2)
13.50
8.11
7.11
4.50
3.05
2.67
2.11
Ratio
(2)/(l)
4.12
3.77
3.20
4.25
3.72
3.47
3.25
Fuel
H
H
H
N
N
N
H
Process
Dry Long
Wet
Wet
Wet
Wet
DryPH/PC
Wet
Conversely, there are several facilities that generate little or no net CKD; however, these
facilities account for a fairly large percentage (around two percent or more) of clinker capacity.
The Kaiser Cement, Cupertino, facility accounted for 2.73 percent of total clinker capacity, yet
generated almost zero percent of total net CKD in 1990. Most of these facilities operate dry
kilns (usually preheater/precalciner kilns) and do not burn hazardous waste fuels.
Exhibit 3-9 provides information on the percentages of gross CKD that were recycled,
sold, and wasted in 1990 for the ten groups of eight facilities, again arranged in descending order
by net CKD generation rate. The percentage of gross CKD recycled in 1990 ranges from about
34 percent for the top net CKD generators all the way up to 100 percent. The percentage sold
does not follow any discernible pattern and generally varies from zero to approximately 10
percent, with the marked exception of the sixth decile. The facilities in the sixth decile sold as
much as 26.7 percent of the gross CKD they generated in 1990. The last three deciles, which
consist of facilities that recycle large portions of their gross CKD, not surprisingly sell negligible
quantities of CKD. Finally, the percentage of CKD wasted increased marginally from 53.7
percent in the top decile to 56.6 percent in the second decile and then decreased substantially in
each decile, with the exception of the sixth, reaching zero in the tenth and final group.
-------
3-14
Exhibit 3-9
Percentages of Gross CKD Recycled, Sold, and Wasted (1990)
Source: 1991 PCA Survey Responses.
Note 1: Data on CKD recycled was not available for one data point in the fifth decile and one data point in
the eighth decile.
Note 2: The last decile contains only seven facilities.
-------
3-15
Finally, the Agency looked at percentages of CKD recycled at the plant level based on
the individual kiln type and fuel usage. Of the 79 facilities for which PCA data are available, 48
operate dry kilns and the remaining 31 operate wet kilns. Of the 48 dry kiln facilities, 14 are
equipped with dry long kilns and 34 with preheaters/precalciners (Ph/Pc).
Exhibit 3-10 summarizes recycling rates among the 48 facilities in EPA's data base that
operate dry kilns. Hazardous waste burners in general do not appear to be recycling as much as
non-hazardous waste burners, and operators of dry long kilns seem to recycle less than those
operating Ph/Pc kilns. Low recycling rates, however, do not always imply that a large percentage
of CKD is wasted. For example, the Heartland Cement Company's Independence facility, which
operates a dry long kiln and burns hazardous waste, wastes only about 8.5 percent of the CKD it
generates even though it recycles only around 37 percent. The facility sold close to 55 percent of
the gross CKD it generated in 1990. Similarly, Southdown's Dixie facility, a hazardous waste
burner operating a Ph/Pc kiln, sold 44 percent of its CKD in 1990 and wasted only 16 percent.
Exhibit 3-10
Recycling Rates Among Facilities That Operate Dry Kilns
Fuel Type
Number of
facilities
Percent of CKD Recycled (Number of
Facilities)
>50%
>90%
100%
Facilities That Operate Dry Long Kilns
Hazardous Waste
Non-Hazardous Waste
All Fuels
5
9
14
2
6
8
1
3
4
0
1
1
Facilities That Operate Dry Ph/Pc Kilns
Hazardous Waste
Non-Hazardous Waste
All Fuels
5
29
34
3
21
24
1
18
19
1
12
13
Facilities That Operate Dry Kilns (Long or Ph/Pc)
Hazardous Waste
Non-Hazardous Waste
All Fuels
10
38
48
5
27
33
.2
21
23
1
13
14
Note: Data are not available on one facility that operates dry long kilns and uses non-hazardous fuels and on one facility
that operates dry Ph/Pc kilns and uses non-hazardous waste fuels.
-------
3-16
Among facilities that burn non-hazardous waste fuels and operate Ph/Pc kilns, an
interesting pattern can be observed. These facilities fall at one of two extremes -- they either
recycle a large percentage or none of their gross CKD. In 1990, as shown in Exhibit 3-10 above,
18 of the 29 Ph/Pc plants for which data are available recycled over 90 percent of the gross CKD
they generated, and three others recycled over 50 percent. In contrast, six of the remaining eight
facilities recycled zero percent of the gross CKD they generated (though the operators of two
facilities, LaFarge, Davenport and Monarch, Humboldt, each sold over 40 percent of their
CKD).
Exhibit 3-11 summarizes recycling rates among the 31 facilities in EPA's data base that
operate wet kilns. As shown in this exhibit, operators of facilities with wet kilns do not appear to
recycle as much as dry kiln operators. The most glaring difference is the fact that no facilities
that operate wet kilns recycle 100 percent of the CKD they generate. Among hazardous waste
burners, only two of the eight facilities recycled over 50 percent of their CKD in 1990, and none
recycled over 90 percent. The percentage of CKD wasted is high, with five of the eight facilities
wasting over 50 percent of their CKD. Among the non-hazardous waste burners, 10 out of 23
recycled over 50 percent and three recycled over 90 percent. A total of six facilities sold over 40
percent of their CKD in 1990. The operator of the Holnam facility in Seattle recycled only
about 51 percent of the CKD it generated, but sold the rest.
Exhibit 3-11
Recycling Rates Among Facilities That Operate Wet Kilns
Fuel Type
Hazardous Waste
Non-Hazardous Waste
All Fuels
Number of
facilities
8
23
31
Percent of CKD Recycled (Number of
Facilities)
>50%
2
10
12
>90%
0
3
3
100%
0
0
0
Finally, because of the observed wide variability in gross and net CKD generation rates,
as well as the demographic characteristics (i.e., age distribution) of domestic cement kilns, EPA
examined whether the age of individual cement kilns appears to influence the generation of
CKD, on either a gross or net basis. The first step was to perform a simple linear correlation
analysis of CKD generation and kiln age for the kilns within the sample of 79 cement plants
providing useable data. The Agency conducted this test using all plants, then repeated the
procedure separately for plants both burning and not burning hazardous waste fuels, and for wet
and dry process kilns (i.e., the Agency conducted a set of five correlation analyses). Results of
this exercise showed that for gross CKD generation, the correlation coefficients were negative
and statistically significant at the 95 percent confidence level for all kilns, non-hazardous waste-
burning kilns, wet process kilns, and dry process kilns; correlation coefficients for these groups
ranged from -0.26 to -0.38. That is, the older kilns generate less gross CKD than the newer
kilns, all else being equal. However, on a net CKD basis, no such relationship is apparent; no
statistically significant correlation coefficients (at the 95 percent confidence level) were found
within any of the five groups with respect to kiln age.
-------
3-17
Because, as documented in Chapter 2, kiln capacity (and therefore, potential gross CKD
generation) has increased over time, EPA proceeded to repeat the above analysis using CKD
generation data that were normalized for clinker capacity. That is, we divided the per-kiln CKD
generation rates by reported kiln capacity to eliminate the potential effects of the trend toward
larger cement kilns over the past few decades.8 Results of this exercise show that for both gross
and net CKD, the normalized CKD generation rates are not related to kiln age for any of the
five groups (i.e., none of the correlation coefficients are significantly different than zero). EPA
therefore concludes that CKD generation is not related to kiln age, even if adjusted for fuel type
or processing technology.
3.1.3 Quantities and Fate of CKD Generated
In an effort to further characterize the highly variable gross and net CKD generation
rates described above, EPA conducted an analysis of potentially significant cement kiln design
and operating variables. The two primary factors that are identified and examined in this section
are kiln process type (wet or dry), and fuel type (i.e., whether the kiln is or is not fired with
hazardous waste). As discussed earlier in this chapter, U.S. cement plants generated a total of
about 12.9 million metric tons of gross CKD in 1990, and 4.6 million metric tons of net CKD,
that is, material removed from the kiln system. As discussed further in Chapter 8, few practical
process modifications can alter gross CKD generation rates. Nonetheless, it is important to look
at differences between kiln types and operating practices to identify the process factors that may
influence the gross quantity of dust generated.
Exhibit 3-12 presents tabulated data that summarize CKD generation rates per ton of
clinker produced as a function of fuel usage (i.e., burning or not burning hazardous waste) and
process type (i.e., wet, dry long, dry with preheater/precalciner). CKD generation data per ton of
product eliminate differences in generation rates that are a function of differences in kiln size;
this allowed the Agency to examine whether there are differences in generation rates that appear
to be directly related to process types and/or fuel usage.
Differences in CKD Generation Rates Across Process Types
Exhibit 3-12 reveals the following relationships with respect to CKD generation across
different process types:
• Wet kilns, which comprise 36 percent of all kilns in Exhibit 3-12, on average
generate less gross CKD per ton of product than dry kilns. The data indicate that
wet kilns generate about 24 percent less CKD per ton of product than dry long
kilns and about eight percent less than Ph/Pc kilns.
• Operators of wet kilns, however, recycle a lower percentage of CKD than
operators of dry kilns; on average, they generate more net CKD per ton of
product (9.5 percent more than dry long kilns and 167 percent more than Ph/Pc
kilns).
• With respect to dry kilns, Ph/Pc kilns generate about 17 percent less gross CKD
and about 60 percent less net CKD per ton of product than dry long kilns.
1 Ideally, the data would have normalized using kiln-specific clinker production data. Because, as discussed above,
such data are unavailable, EPA used clinker capacity data as the best available proxy.
-------
3-18
-------
3-19
Exhibit 3-12
Average CKD Generation Rates Per Ton of Product (1990)'
Kiln Type
No. of Kilns
CKD Per Ton of Product Ratio'
Gross
CKD
Net CKD
CKD
Recycled'
Wet Kilns
Burning Hazardous Waste
No Hazardous Waste
All Fuels
13
28
41
0.219
0.179
0.192
0.163
0.093
0.115
0.056
0.086
0.077
Dry Long Kilns
Burning Hazardous Waste
No Hazardous Waste
All Fuels
13
19
32
0.236
0.264
0.253C
0.131
0.087
0.105C
0.105
0.177
0.148
Dry Preheater/Precakiner Kilns
Burning Hazardous Waste
No Hazardous Waste
All Fuels
6
34
40
0.175
0.215
0.209*
0.071
0.038
0.043'
0.104
0.177
0.166
All Dry Kilns
Burning Hazardous Waste
No Hazardous Waste
All Fuels
19
53
72
0.217
0.233
0.228
0.112
0.055
0.070
0.105
0.178
0.158
All Kilns
Burning Hazardous Waste
No Hazardous Waste
All Fuels
32
81
113
0.218'
0.214C
0.215
0.13.3'
0.068°
0.087
0.085
0.146
0.128
* Source: Bureau of Mines.
b In general, Gross CKD = Net CKD + CKD Recycled.
e Computed from Bureau of Mines data.
EPA has evaluated the significance of these data further by performing pair-wise t-test
comparisons of the average waste-to-product ratio value (means) provided in Exhibit 3-12. The
results of this exercise demonstrate that there are no statistically significant differences (at a 95
percent confidence level) in the normalized gross CKD generation rates between any of the
groups identified in the exhibit. That is, despite the apparent differences in average gross CKD
generation rates per unit of product between, for example, hazardous waste-burning wet kilns
and non-hazardous waste burning wet kilns, these differences are not sufficient, on a statistical
basis, to indicate that these two groups are fundamentally different with respect to this variable.
In marked contrast, a number of statistically significant differences are apparent between
various groups with respect to net CKD generation relative to production. Looking at the
sample as a whole, the 32 kilns burning hazardous waste generate substantially more net CKD
per unit of product, on average, than the 81 kilns that are not fired with this alternative fuel.
-------
3-20
The difference (average of 0.113 versus 0.068 tons of net CKD per ton of product, respectively)
is significant at the 99 percent confidence level. A similar pattern is observed within kiln type
groups: both wet and dry kilns burning hazardous waste fuels have significantly higher (at a 95
percent confidence level) average net CKD generation rates than kilns of the same technology
type that do not burn hazardous waste fuels. Interestingly, the differences within the dry process
kiln type category diminish when considering dry long and dry Ph/Pc kilns individually; although
average net CKD generation rates are higher for the hazardous waste burners within each of
these technology type sub-groups, the differences between these rates and those of the non-
hazardous waste burning kilns within their respective sub-groups are not statistically significant at
a 95 percent confidence level.
Average normalized net CKD generation rates also appear to vary significantly by kiln
technology type alone. Wet process kilns have average net dust generation rates that are
significantly higher than those of dry process kilns, and within the dry process kiln type, dry long
kilns generate significantly more net CKD per unit of product, on average, than Ph/Pc kilns. The
significance levels for EPA's comparisons between these groups approach or exceed 99 percent.
Differences in CKD Generation Rates Across Process Types and Fuel Usage
Further examination of the data in Exhibit 3-12 reveals the following findings:
• Wet kilns that burn hazardous waste fuels generate about 22 percent more gross
CKD per ton of product than those that do not burn hazardous wastes. In the
case of dry kilns, the data suggest the opposite - dry kilns that burn hazardous
waste generate about 7 percent less gross CKD per ton of product.
• The inverse relationship between hazardous waste burning and gross CKD
generation in dry kilns becomes more marked with increasing technological
sophistication. Dry long kilns burning hazardous waste generate close to 11
percent less gross CKD per ton of product than dry long kilns not burning
hazardous waste; Ph/Pc kilns that burn hazardous waste generate almost 20
percent less gross CKD per ton of product.
• Across all process types, operators of kilns that bum hazardous waste
(representing almost 30 percent of all kilns), recycle significantly less CKD per ton
of product than non-hazardous waste burners. The result is that hazardous waste
burners generate almost twice as much net CKD per unit of product as non-
hazardous waste burners, though they generate only about two percent more gross
CKD.
• In the case of dry kilns, lower recycling rates (per ton of product) can be partly
attributed to the fact that kilns burning hazardous waste generate lower quantities
of gross CKD per ton of product than kilns that do not burn hazardous waste.
Operators of dry kilns that burn hazardous waste, however, recycle a lower
percentage of the gross CKD they generate than kilns that do not burn hazardous
waste -- 48 percent compared with 70 percent, respectively.
Differences in Gross CKD Generation Rates
-------
3-21
The observations highlighted above reveal that for the wet and dry kilns overall (i.e.,
regardless of which kilns burn hazardous waste), gross CKD generation rates per ton of product
appear to be slightly lower in wet kilns (19.2 percent) than in dry kilns (22.8 percent). This
finding supports at least one source that states wet kilns generate less gross CKD than dry kilns.9
Gross CKD generation rates per unit of product, when all kilns are considered together,
are slightly higher for hazardous waste burners than for kilns not burning hazardous waste.
When different process types are considered, this relationship holds true only for wet kilns.
Therefore, there is no conclusive evidence indicating that burning hazardous waste results in
increased gross CKD generation, though such a finding was indicated in an early EPA study on
this topic.10 One argument against this conclusion is that kilns that burn hazardous waste
should generate less ash per unit of energy consumed than kilns that burn coal. Burning
hazardous waste fuels, however, may allow the facility operator to burn a lower grade of coal
(i.e., with a higher ash content) than it could otherwise, thus maintaining a relatively high overall
ash content.
Differences in Net CKD Generation Rates
Based on observations from Exhibit 3-12, net CKD generation rates are higher in kilns
burning hazardous waste. Although EPA has not found definitive evidence that burning
hazardous waste causes increased net CKD generation rates, limited documentation suggests a
link between the two variables. In one early study conducted by EPA in 1981, trial burns were
conducted at three dry process cement kilns and two wet process kilns to compare results when
coal was burned alone and when coal was co-fired with hazardous waste at unspecified rates.
(None of these kilns was identified in the study.) In one of the dry process kilns, normal coal-
fired operations generated approximately 91 metric tons per month of net CKD. When
hazardous waste was co-fired, this figure increased to 1,800 metric tons per month, reportedly to
keep system chloride levels within prescribed limits. Information on changes in the amount of
dust generated at the other four facilities was not reported.11
More recent data are not adequate to support conclusions regarding any cause-and-effect
relationship(s) between combustion of hazardous waste fuels and net CKD generation rates,
because the available data were not collected over time and do not include observations obtained
during both hazardous waste fuel burning and the absence of this practice at the same plants.
Nonetheless, the data analyzed and presented in this report reflect the actual operating
experience of the majority of active cement plants in the U.S. and do allow the Agency to make
some interesting comparisons. These data show that within each kiln type group, on average,
kilns fired with hazardous waste fuels have net CKD generation rates that are substantially
higher than those of kilns not fired with hazardous wastes. In each of the three basic kiln type
groups, net CKD generation, normalized for actual production rates, was from 50 to 87 percent
higher in kilns burning hazardous wastes than in kilns not burning these alternative fuels. In
addition, net CKD generation rates are substantially higher for wet kilns than for
preheater/precalciner kilns, and are somewhat higher than for dry long kilns. This pattern is
apparent in both fuel type groups.
9 Engineering-Science, 1987, op. cit., p. 3-12.
10 Ibid.
11 Engineering-Science, 1987, op. cit., p. 4-18.
-------
3-22
CKD Recycling
Although net CKD may be viewed as a waste, its nature as essentially an "off-spec
clinker," or intermediate product, makes direct return to the kiln, or recycling, a desirable option
for cement plant operators. If more CKD could be returned to the kiln system via recycling, less
net CKD would be generated. Decreasing net CKD quantities reduces the quantity of dust that
must be managed in some other manner. Moreover, reduced net CKD generation saves energy
and raw materials through minimizing raw feed demand and heat lost to materials that are
removed from the system and not productively used.
A variety of methods are currently used to directly return CKD to the kiln system. First,
dust is injected, or insufflated, through or near the flame at the hot end of the kiln. Second,
CKD is conveyed to a shroud, or sleeve, that embraces the middle of the kiln near the material
inlet to the calcining zone. In both of these cases, CKD is mechanically conveyed by a screw
conveyor from the APCD to the point of re-introduction to the kiln. Third, operators introduce
CKD at the front end along with the raw feed. Alternatively, CKD can be returned to the kiln
after first treating it for removal of undesirable contaminants (e.g., through leaching,
volatilization, or recovery scrubbing). These approaches to returning CKD to the kiln are the
subject of further elaboration in Chapter 8.
Conceptually, the ideal strategy for any cement plant operator would be to return all of
the gross CKD to the process, which would eliminate any need to dispose of or find alternative
uses for waste CKD. Returning CKD to the kiln system, however, involves balancing savings in
resources, energy, and waste management costs with the costs of increased concentrations of
certain CKD constituents and the capital and operating costs of the necessary equipment. While
cement plant operators do typically recycle some portion of the gross CKD, the gradual
accumulation of alkalies in the dust usually necessitates that some CKD be removed from the
system as a net waste. Through recycling, chlorine and alkalies tend to accumulate in the gross
CKD that is generated. These constituents can continue to build up in the kiln system as alkalies
and alkali salts, which may impair the cement production process in three primary ways:
- • Increased paniculate matter emissions. As discussed in Sections 3.2.1 and 3.2.2,
below, alkalies and chlorides may decrease the efficiency of ESPs, resulting in
increased paniculate matter emissions.12
• Kiln damage and/or preheater plugging. Alkali chlorides, which can damage kiln
linings, condense more readily in the kiln than oxides, which help to protect kiln
linings. Alkali chloride condensation can also lead to preheater plugging and
ultimately to increased alkali recirculation.
• Inferior quality cement product. Alkali levels also affect the quality of the cement
product. The American Society for Testing and Materials (ASTM) sets specific
limits on alkali levels in portland cement. ASTM C 150 mandates that cement
contain no more than 0.6 percent alkali.13 This standard was created to
12 Beers, A., 1987. New York State Legislature, Legislative Commission on Toxic Substances and Hazardous
Wastes. Hazardous Waste Incineration: The Cement Kiln Option. December, p. 11.
" American Society for Testing and Materials, 1987. ASTM C150-86, Vol. 04.02. Concrete and Aggregates, p. 91.
-------
3-23
minimize a detrimental phenomenon in concrete called alkali-aggregate
reactivity.14 Through this phenomenon, chemical reactions between the cement
gel and the aggregates (i.e., gravel that is mixed with the cement to form
concrete) cause the concrete to expand and may result in cracking or other
structural defects. To limit alkali concentrations in the cement product, some
CKD must typically be removed from the system.
' Recycling Differences
To help identify some of the factors influencing recycling rates, it is useful to compare a
facility that recycles all of its gross CKD to a facility that recycles none of its gross CKD. The
operator of the active kiln at the Kaiser Cement Company in Cupertino, California, for example,
recycles almost all (99.97%) of the gross CKD back to the raw feed. The unit is a dry process
kiln with a preheater and precalciner, and is fired primarily with low-sulfur coal, and to a lesser
extent, petroleum coke. The facility operators attribute the high recycling rate to the inherently
low alkali, chloride, and sulfate levels in the raw material and fuel inputs.15
In contrast to the Kaiser Cupertino plant, the operator of the four active wet process kiln
systems at Texas Industries, Inc., in Midlothian, Texas, recycles none of the gross CKD. Each
kiln is fired with a mixture of coal, natural gas, petroleum coke, and liquid hazardous waste.
According to the facility operator, the raw feed is high in alkalies, and because Texas Industries
produces low-alkali cement, the operator believes that as generated CKD cannot be recycled
back to the kiln."
The idea that recycling rates are solely dependent on raw feed yields a simple
generalization. Generally, higher alkalies should result in reduced recycling for a given cement
grade. However, process and fuel differences can also significantly influence recycling rates. The
remainder of this section discusses these differences in greater depth by examining the PCA
Survey data for recycling rate differences across process type and fuel type. The influence of raw
feed cannot be assessed in this analysis because appropriate raw feed characterization data are
not readily available.
Exhibit 3-13 presents summary data that express CKD going to the various management
pathways as a weight percent of gross CKD, as a function of whether they burn or do not bum
hazardous waste, and process type (i.e., wet, dry long, and dry with preheater/precalciner). The
data in Exhibit 3-13 demonstrate a number of interesting relationships that are outlined below.
This discussion enhances the discussion relating to Exhibit 3-12, particularly with respect to the
percentage of gross CKD that is disposed and sold.
Differences in Fate of CKD Across Process Types
Exhibit 3-13 reveals the following relationships with respect to CKD fate across different
process types:
14 Kosmatka, S. and Panarese, W., 1990. Design and Control of Concrete Mixtures. Portland Cement Association.
Skokie, Illinois, pp. 42-43.
u U.S. EPA, 1992. Sampling Trip Reports.
16 Ibid
-------
3-24
• Wet kilns generate, on average, 59 thousand metric tons of gross CKD, compared
with about 69 thousand metric tons generated in dry long kilns and 60 thousand
metric tons generated in Ph/Pc kilns. As previously noted, however, operators of
wet kilns recycle a lower percentage of their gross CKD -- 45 percent in contrast
with 65 percent in the case of dry long kilns, and 82 percent in the case of Ph/Pc
kilns.
• The percentage of gross CKD that is sold is about five times higher for wet kilns
than it is for dry kilns. Operators of wet kilns sell 13 percent of all the gross
CKD they generate, compared with less than three percent for either category of
dry kiln (though operators of hazardous waste-burning Ph/Pc kilns sold about 9.5
percent of their gross CKD in 1990).
• Despite the relatively large percentage of gross CKD that is sold by wet kilns, the
lower percentage of CKD recycled by the operators of such kilns results in them
wasting the highest percentage of gross CKD. In 1990, operators of wet kilns
disposed of over 40 percent of all the CKD they generated. In the same year, dry
long kiln operators disposed of about one-third of the CKD they generated, while
the operators of Ph/Pc kilns disposed of only about 15 percent.
Differences in Fate of CKD Across Process Types and Fuel Usage
Examination of the fuel usage data in Exhibit 3-13 reveals the following relationships:
• Across all process types, hazardous waste-burning kilns generate, on average,
more gross CKD than those not burning hazardous wastes. The difference is the
greatest in the case of wet and Ph/Pc kilns, where hazardous waste burners
generate almost twice as much gross CKD as non-hazardous waste burners. Part
of this phenomenon is explained by the fact that kilns burning hazardous waste
fuels tend to be larger than those not burning such fuels. The difference in size is
apparent when Exhibit 3-13 is compared with Exhibit 3-12, where CKD generation
rates have been normalized by production rates. For example, gross CKD per ton
of product for wet kilns is only 1.2 times higher for hazardous waste burners than
it is for non-hazardous waste burners (see Exhibit 3-12), though the former
generate almost twice as much gross CKD as the latter in terms of absolute
quantity (see Exhibit 3-13). In the case of dry kilns, gross CKD generated per ton
of product is less for hazardous waste burners than it is for non-hazardous waste
burners (see Exhibit 3-12), though the former generate larger average quantities
of gross CKD (see Exhibit 3-13).
• Across all process types, operators of hazardous waste-burning kilns recycle a
lower percentage of their gross CKD than those operating non-hazardous waste-
burning kilns. Operators of wet kilns, as well as dry kilns that burn hazardous
waste fuels, recycle almost 35 percent less of their gross CKD than those who do
not burn hazardous waste fuels (34 percent compared with 52 percent in the case
of wet kilns, and 55 percent compared with 83 percent in the case of dry kilns).
Overall, kilns burning hazardous wastes generate about twice as much net CKD as
a weight percentage of gross CKD than do kilns not burning hazardous wastes.
-------
3-25
This difference in recycling rates based on fuel usage is the most marked in the
case of dry long kilns, where the percentage of gross CKD recycled by hazardous
waste burners is only about half that of non-hazardous waste burners. Hazardous
waste burners in this process type generate almost three times as much net CKD
as a percentage of gross CKD - 54 percent, compared with 20 percent in the case
of non-hazardous waste burners.
-------
3-26
Exhibit 3-13
Fate of CKD as a Percent of Gross CKD (1990)1
Fate of CKD as a Percent of Gross CKD (Averages)
Kiln Type
No. of
Plants
No. of
Kilns
Average
Gross
CKD
per kiln
(metric
tons)
Weight Percent of Gross CKD
Net
Recycled
Disposed
Sold"
Wet
Burning Hazardous Waste
No Hazardous Waste
All Fuels
8
23
31
14
39
53
84,724
49,944
59,131
65.96
47.90
54.74
34.04
52.19
45.32
54.32
33.34
41.28
11.64
14.56
13.45
Dry Long
Burning Hazardous Waste
No Hazardous Waste
All Fuels
5
8
13
15
19
34
69,081
68,707
68,872
5435
19.65
35.01
45.65
80.39
65.01
53.05
17.28
33.11
1.31
2.37
1.90
Dry PH/PC
Burning Hazardous Waste
No Hazardous Waste
All Fuels
5
28
33
6
51
57
94,253
56,389
60,374
29.02
17.06
19.03
70.98
83.73
81.63
19.53
14.97
15.72
9.49
2.09
3.31
All Dry Kilns
Burning Hazardous Waste
No Hazardous Waste
All Fuels
10
36
46
21
70
91
76,273
59,732
63,549
45.41
17.87
25.50
54.59
82.69
74.90
41.21
15.69
22.76
4.20
2.18
2.74
All Kilns
Burning Hazardous Waste
No Hazardous Waste
All fuels
18
59
77
35
109
144
79,653
56,230
61,923
54.15
27.41
35.77
45.85
72.99
64.51
46.79
21.30
29.27
7.36
6.11
6.50
' Based on usable responses from 1991 PCA Survey.
k "CKD Sold" quantities may also include CKD that was given away.
-------
3-27
• No patterns emerge from the data in Exhibit 3-13 with respect to the percentage
of gross CKD sold by operators of hazardous versus non-hazardous waste-burning
cement kilns. In the case of wet kilns, hazardous waste burners sell a lower
percentage of their CKD than non-hazardous waste burners, though the difference
is not striking. In the case of dry kilns, however, hazardous waste burners sold a
higher percentage of the CKD they generated in 1990 than non-hazardous waste
burners. Operators of Ph/Pc kilns that burned hazardous waste fuels sold over
four times more, as a percentage of gross CKD, than did the operators of the
Ph/Pc kilns that do not burn hazardous waste fuels.
• Hazardous waste burners across all process types dispose of a larger percentage of
the gross CKD they generated as compared with non-hazardous waste burners.
This result is consistent with the earlier observation with respect to recycling rates.
Although no pattern emerged in the case of CKD sold, the percentages of CKD
sold are generally low relative to CKD recycled and it is not surprising that the
percentages of CKD sold did not significantly influence the outcome with respect
to CKD wasted.
• Across the entire sample of 144 kilns and 77 plants, close to 65 percent of the
gross CKD generated was internally recycled in 1990, and of the more than 35
percent comprising net CKD, approximately 82 percent was disposed and 18
percent was sold or given away.
The data contained in Exhibits 3-12 and 3-13 show that recycling rates do differ between
process type and between fuel type. Although some CKD must be removed from the system to
reduce alkali levels, it is possible that some facility operators recycle none of their CKD to
reduce the complexities of clinker quality control. Any explanation for the greatly reduced
recycling rates observed among wet kilns compared to dry kilns is probably based on economics,
because full recycling (in some cases with treatment of CKD) could probably be achieved for any
process if cost were not an issue. Possible recycling technologies are discussed in greater depth
in Chapters 8 and 9. Nonetheless, three potential explanations are presented here for the finding
that wet kilns recycle less CKD than dry kilns.
One reason that wet kilns recycle less CKD than dry kilns is apparently because, relative
to the number of kilns in each process type, a larger proportion of wet kilns burn hazardous
waste than dry kilns. As discussed above, decreased CKD recycling rates are associated with
hazardous waste burning. According to the 1991 PCA Survey, 13 of 43 wet kilns (30 percent)
burn hazardous waste while only 19 of 82 dry kilns (23 percent) bum hazardous waste. This
practice is probably driven by economics. Because of the higher water content of the raw feed,
which must be dried, wet process kilns are inherently less energy efficient than dry kilns.17 To
supplement the large energy demands required by wet process kilns in a cost-competitive
manner, wet kiln operators have presumably looked to hazardous waste as an inexpensive source
of fuel with high heat content.18
17 Beers, A., 1987, op. at.
" Gossman, D., 1992. The Reuse of Petroleum and Petrochemical Waste in Cement Kilns. Environmental
Progress (Vol. 11, No. 1). February, p. 5.
-------
3-28
Another possible explanation is the fact that wet kilns generally represent older systems
than dry kilns, and that perhaps these kilns also represent a sector of the cement industry where
any system modifications are not projected to be cost-effective. Hence, operators of wet kilns
may be wasting a greater proportion of dust to simplify their process operations. Rather than
installing recycling process equipment and constantly monitoring CKD constituents to determine
appropriate recycling rates, such operators may find it easier and more economical to waste the
dust.
As mentioned previously, observed recycling differences between kiln systems may owe to
plant-specific differences in raw feed inputs, where the raw feed inputs may typically be higher in
alkalies for wet kilns than for kilns not burning hazardous waste, or for dry process kilns. Due to
a general paucity of raw feed data, the only relevant comparison that can readily be made to
explore this possibility is to compare the geographic distribution of kilns based on recycling rate,
which can indicate regional differences in the geology and chemistry of raw materials. If regional
trends were noted, EPA could infer an influence from raw feed inputs. In reviewing the
geographic distribution of recycling rates (through mapping), however, EPA noted no recycling
patterns by region for process type or for fuel type.
The most plausible explanation for decreased recycling rates among kilns burning
hazardous waste is that chloride, alkali, and/or sulfate levels in some hazardous wastes may
significantly increase the loading rates of these contaminants in the dust. To maintain acceptable
levels of chloride, alkali, and sulfate in the system, more CKD may need to be removed from the
system than if the kiln had not been burning hazardous waste as fuel. Some facility operators
report that the burning of hazardous waste with high chlorine levels can induce the precipitation
of alkali chlorides in the kiln. Bleeding of CKD then removes these alkali chlorides from the
system.1' This idea is supported in the co-firing study cited above, where net CKD in one kiln
was increased from 91 metric tons per month to 1,800 metric tons per month to control chloride
levels.20
3.2 CKD GROSS CHARACTERISTICS
CKD is comprised of thermally unchanged raw materials, dehydrated clay, decarbonated
(calcined) limestone, ash from fuel, and newly formed minerals corresponding to all stages of
processing up through the formation of the clinker.21-22 An unusual feature of CKD is that,
unlike typical process wastes that are substantially different than the product, CKD is essentially
cement clinker that does not quite meet commercial specifications.
3.2.1 Physical Characteristics
Although the relative constituent concentrations in CKD can vary significantly, CKD has
certain physical characteristics that are relatively consistent. When stored fresh, CKD is a fine,
dry, alkaline dust that readily absorbs water. When managed on site in a waste pile, CKD can
"ibid.
20 Engineering-Science, 1987, op. cit., p. 4-18.
21 Kohlhaas, B., et al., 1983, op. cit., pp. 624.
22 Engineering-Science, 1987, op. cit., pp. 3-4, 3-12, 4-18.
-------
3-29
retain these characteristics within the pile while developing an externally weathered crust, due to
absorption of moisture and subsequent cementation of dust particles on the surface of the pile.
Exhibit 3-14 provides particle size distributions for CKD generated by various process
types. It demonstrates that the size distribution of CKD can vary significantly, with diameters
ranging from near zero to greater than 50 /xm. (The lack of statistical information from the
sources of these figures necessitates that only qualitative conclusions about particle size
distributions be drawn.) The data show that from 15 to 90 percent of CKD has a diameter
below 10 /xm, within the respirable range for humans. Moreover, Exhibit 3-14 suggests that at
least 55 percent of CKD measures less than 30 /xm in diameter, while a nearly uniform 82
percent falls below 50 /im. Although the data contained in Exhibit 3-14 are limited in scope and
therefore inconclusive, it appears that dry precalciner kilns generate larger CKD particles than
wet kilns or dry long kilns, while dry long kilns may produce the smallest CKD particles. Dry
long kilns appear to generate nearly all (90 percent) of their CKD in the respirable range, while
only 17 percent of CKD from dry kilns with precalciners is in this size range. Median sizes also
suggest that dry, long kilns may have the smallest CKD particles, at 3 /xm, followed by wet kilns
at 9.3 jim, and finally dry kilns with precalciners at 22.2 /xm.
Exhibit 3-14
Particle Size Distribution of CKD by Process Type
Particle
Size (jan)
0-5
5-10
10-20
20-30
3040
40-50
>50
Median
Particle
Size
SOURCE 1"
Unspecified Process Type
(weight percent)
5
10
30
17
13
7
18
No Data
SOURCE 2k
Wet Kilns
(weight percent)
26
19
20
9
8
1
17
93
Long Dry Kilns
(weight percent)
45
45
5
1
1
0
3
3.0
Dry Kilns with
Precalciner
(weight percent)
6
11
15
23
18
9
18
22.2
' Kohlhaas, et al., 1983, op. cit., p. 640. The number of samples used to develop data was not specified.
b Todres, H., A. Mishulovich, and J. Ahmed, 1992. CKD Management: Permeability. Research and
Development Bulletin RD103T, Portland Cement Association, Skokie, Dlinois, p. 2. It appears that one sample per
process type was analyzed to develop the data presented above.
-------
3-30
Exhibit 3-15 presents particle size data submitted by Midwest Portland Cement Company
in Zanesville, Ohio, in response to a request for data by EPA under Section 3007 of RCRA.
This facility operates two wet process kilns. These data show that "peaks" in the particle size
distribution occur at diameters of 22 fim (approximately 13 percent of total dust volume) and 3.9
pm (about 11 percent of the total dust volume). Thirty percent or more of the CKD examined
in this analysis had an aerodynamic diameter of less than 10 /im, which is the respirable range for
humans.
Coplay (ESSROC Materials) Cement Company of Frederick, Maryland, in response to
the same RCRA Section 3007 request, submitted particle-size screen analysis data on "typical
stack dust."23 These data were similar to those submitted by Midwest Portland, showing 99.9
percent of dust passing a screen size of approximately 185 /un, 99.4 percent passing 130 /zm, 88.8
passing 110 /zm, and 72.4 percent passing 72 /*m. (These screen sizes were converted from
screen sizes of #20 mesh, #50 mesh, #100 mesh, and #200 mesh, respectively.)
Exhibit 3-15
Particle Size Distribution of CKD
Midwest Portland Cement Company, Zanesville, Ohio*
Particle Size (pm)
176.0
125.0
88.0
62.0
44.0
31.0
22.0
16.0
11.0
7.8
5.5
3.9
2.8
1.9
Percent Volume Passing
1.6
7.0
93
9.1
8.0
8.5
12.8
5.7
9.6
6.7
7.4
11.4
2.5
0.4
Cumulative Percent
Volume Passing
100.0
98.4
91.4
82.1
73.0
64.9
56.5
43.7
38.0
28.4
21.7
143
2.9
0.4
23 ESSROC Materials, Inc., 1992. Particle size distribution of "typical stack dust," February 14, 1992. Submitted
on November 3, 1992 in response to RCRA Section 3007 request for information by U.S. EPA, August 18, 1992.
-------
3-31
1 L4
0.0
0.0 1
' Midwest Portland Cement Company, 1992. Particle size distribution of kiln dust laboratory sample,
February 14, 1992. Submitted on September 2, 1992 in response to RCRA Section 3007 request for information by
U.S. EPA, August 18, 1992.
The fine-grained nature of CKD makes it easily transportable in air, a factor that
necessitates the use of effective air pollution control devices to remove this material from kiln
exhaust gases. The smallest particles may not be fully captured by air pollution control devices,
and may instead be released into the atmosphere. Particles smaller than 75 ^m can be
suspended in the air and tend to follow air currents. At 30 /tm or less, these particles can travel
long distances before settling.
The ability of CKD to absorb water stems from its chemically dehydrated nature, which
results from the thermal treatment it receives in the kiln system. The action of absorbing water
(rehydrating) releases a significant amount of heat from non-weathered dust, a phenomenon that
can be exploited in beneficially using CKD. For example, CKD can be used to dewater
municipal sewage sludge, while the heat of hydration can be used to sterilize the blended
material. Such uses are discussed further at the end of this chapter and in more detail in
Chapter 8.
Hydraulic conductivity represents a physical characteristic of particular interest for CKD
managed in piles or beneficially used in applications such as bulk fill. If, for example, CKD were
to conduct water fairly well, it could be used as a bulk fill without concern about ponding and
structurally unstable saturated material. Disposal of CKD in a waste management unit that
readily conducted water might, however, require controls to prevent release of leachate to the
environment.
EPA's data on CKD hydraulic conductivity are limited to two sources. Source 1 is a
report on CKD pile characteristics by General Portland (now National Cement), in Los Robles,
California.24 The report presents test results comparing fresh CKD to CKD that had been
placed in a waste pile. In a laboratory experiment, the two dust types were compacted to varying
degrees under an empirically determined optimum compacting moisture content, and their
hydraulic conductivities were measured. Source 2 is a conductivity study conducted by
researchers for PCA.25 In this study, investigators compared the conductivity of CKD from
three process types: wet, dry long, and dry with precalciner. Exhibit 3-16 summarizes the results
of these studies.
As shown in Exhibit 3-16, the hydraulic conductivity of CKD is inherently low, at least
compared to typical soil types. Compacted CKD conductivities are as low as IxlO'10 cm/sec, an
extremely low value compared to the typical conductivity of a compacted clay landfill liner, which
is about 1 x 10"7 cm/sec. The highest conductivity was 3 x 10"3 cm/sec, which indicates moderate
permeability. Making solely qualitative comparisons because additional data were not provided,
24 Chadbourne, J. and E Bouse, General Portland, 1985. Los Robles Cement Plant CKD Waste Classification
Report. August, p. 5.
B Todres, H., A. Mishulovich, and J. Ahmed, 1992. CKD Management: Permeability. Research and Development
Bulletin RD103T, Portland Cement Association, Skokie, Illinois, p.7.
-------
3-32
no clear pattern was observed between process types from Source 2. For example, although dust
from the wet process exhibited lower permeabilities at medium and high compactions, this same
dust had the highest conductivity under light compaction. For Source 1, the waste pile dust
appeared to exhibit a lower conductivity than the fresh dust. The authors of Source 1 attributed
the lower conductivity of the waste pile dust to its "setting" during prolonged weathering.
Although all or most of the cemented aggregations that might have formed during weathering
would most likely be broken down during compaction in the test, the disaggregated particles in
the managed dust appear to more readily bind and reduce conductive pore space than in the
generated dust. The data from this test suggest that, although CKD hydraulic conductivity
appears to be inherently low, this property appears to decrease even further with time, especially
when CKD is exposed to atmospheric influences such as humidity and rain.
-------
3-33
Exhibit 3-16
Hydraulic Conductivity of Freshly Generated and Managed CKD
SOURCE r
Percent of Maximum
Compacted Density
85
90
95
Hydraulic Conductivity
(cm/sec)'
Waste Pile
6.5x10*
5.5xlO-<
1.9x10-"
Fresh
3.5xl(T5
1.3xl05
4.7x10*
SOURCE 2"
Degree of
Compaction'
Light
Medium
Heavy
Hydraulic Conductivity
(cm/sec)
Wet
Process
1.5x10'
7.6x10*
IxlO'10
Dry
Process
3.0x10'
7.0x10"*
OxlO4
Dry Process with
Precalciner
S.lxlO"
2.1X10'5
1.6x10-'
' Chadbourne, J., and E. Bouse, 1985, op. cit., p. 5.
6 Todres, H., et al, 1992, op. cit., p. 7.
c Relative hydraulic conductivities between soil/rock types (Todres el al., 1992, p. 5):
Relative Permeability Hydraulic Conductivity Typical Soil/Rock Type
High
Medium
Low
Very Low
Near Impervious
>ia' (cm/sec)
104 to Iff' (cm/sec)
10J to 10J (cm/sec)
Iff3 to 10-' (cm/sec)
< Iff' (cm/sec)
Coarse gravel
Sand, find sand
Silty sand, dirty sand
Silt, fine sandstone
Clay
* Degrees of compaction were designed to simulate the following treatments: Light Compaction = "as dumped", with little
or no compact'rve effort; Medium Compaction (Standard Proctor) = Compaction in the Held at appropriate moisture content, using
moderate equipment on thin lifts; Heavy Compaction (Modified Proctor) = Compaction in the Held at appropriate moisture content,
using heavy equipment on controlled lift thicknesses.
3.2.2 Bulk Chemical Characteristics
An analysis of the chemical composition of CKD can be conducted on two levels: 1) the
major, or "bulk" constituents of the material, and 2) minor, or "trace" constituents that may
comprise very small percentages of total CKD mass but nonetheless be important from an
operational and/or environmental standpoint. Bulk constituents are defined herein as those that
exceed 0.05 percent by weight in the CKD. Bulk constituents are primarily those found in
clinker, though they also may be present at levels in CKD that are unacceptable in the cement
product. Although the types of bulk constituents found in CKD do not vary significantly among
samples from different plants and over time, the relative proportions of these constituents vary
widely. Trace CKD characteristics are presented in Section 3.3 and clinker characteristics are
presented in Section 3.5.
Exhibit 3-17 summarizes some of the data on bulk constituents available for comparison,
which suggest that significant CKD variability can occur even between kilns with relatively minor
process differences. The first two columns of data show ranges for data provided by respondents
to the 1991 PCA Survey, divided between all wet kilns and all dry kilns for which there are data.
-------
3-34
The third and fourth columns contain analytical data representing two types of dry preheater
kilns: dry with cyclone preheater and dry with grate preheater.
Although wide concentration ranges exist for most constituents, Exhibit 3-17 generally
shows that the primary bulk constituents in CKD are silicates, calcium oxide, carbonates
(expressed as loss of CO2 and H2O on ignition), potassium oxide, sulfates, chlorides, various
metal oxides, and sodium oxide. The information presented in Exhibit 3-17 suggests that few
inferences can be drawn from these data regarding process influences on CKD chemical
composition. That is, variability in CKD composition appears to be without trend among the
process types, based upon this limited sample. The chloride content of the wet kiln CKD may
exceed that of the dry kiln CKD by a factor of about two. Exhibit 3-17 does serve to
demonstrate that factors other than process type (e.g., fuels, feed, product specifications) may be
influencing CKD chemical characteristics, because of the high degree of constituent
concentration variability within each process type.
As an additional measure of chemical characteristics, Exhibit 3-17 shows that CKD is
inherently alkaline. This characteristic is a clear function of the large quantity of CaO and other
alkaline compounds, such as KjO, NaOH, Na2CO,, and Na2SO4, that comprise CKD. Again,
however, conclusions based on process differences are tenuous using the available data. In
general, the pH of CKD leachates (using standard EPA leachate procedures) falls between 11
and 13.26 The significance of these leachate levels with respect to environmental risk is
discussed in detail in Chapter 6.
3 J CKD TRACE CHARACTERISTICS
Trace constituents are generally found in concentrations of less than 0.05 percent by
weight and are typically expressed as milligrams per kilogram (mg/Kg), or parts per million.
These constituents include certain organic chemicals, metals such as cadmium, lead, and
selenium, and radionuclides. Trace constituents are important to an analysis of the chemical
characteristics of CKD because some of these elements and compounds are toxic or otherwise
harmful at low concentrations, and as discussed below in Chapter 5, CKD has been managed in a
way that may release these trace constituents to the environment. Furthermore, the use of
hazardous waste and other wastes (e.g., slag) and raw materials as fuel and raw material inputs in
cement kilns has raised concerns regarding the concentrations of certain heavy metals in CKD
generated by plants that use these alternative materials.
3.3.1 EPA Sampling Program
With the exception of metals and general chemistry data,27 the Agency found that
existing chemical characterization data on CKD was insufficient for the purpose of determining
what organic and inorganic constituents may be present in CKD. Therefore, the Agency decided
to undertake a sampling program in order to characterize CKD more fully for this Report to
26 Although some leachate pH values from dry kiln-derived CKD have been measured below 9.0, the Agency does
not believe that kiln type exerts a significant influence on the alkalinity of CKD, and believes that the validity of some
reported data is questionable. For example, results from EPA's 1992 field sampling and analysis yielded several as
generated and as managed CKD samples with reported laboratory leachate values below pH 9.0. The Agency believes
these pH levels may not be representative of typical CKD leachate characteristics.
27 General chemistry analytes are also referred to as "major ions" and Vet chemistry" in this document.
This class of analytes includes chlorides, sulfates, sulfides, fluorides, cyanide, and total organic carbon.
-------
3-35
Congress. During the spring of 1992 and spring of 1993, EPA visited a total of 20 cement
manufacturing facilities for the
-------
3-36
Exhibit 3-17
Topical CKD Bulk Constituents
Constituent
SiO2
A1A
TiO2
FejO3
Mn2O3
CaO
MgO
SO,
K2O
cr
Na2O
LOI (COj+HjO)
PH
Concentration Ranges (weight percent)
Long Dry Kilns, Dry
Kilns with Preheaters,
and/or Calciners*
43-10.1
1.0-33
0.07-0.2
0.7-23
0.01-0.2
11.0-45.0
0.4-2.0
0.1-7.7
0.2-9.7
0.08-2.7
0.07-1.2
Not Available
6.11 - 12.83f
(S.U.)
Wet Kilns*
4.1-7.7
13-33
0.08-0.2
0.8-2.0
0.02-0.04
15.9-38.0
0.4-1.9
0.1-6.0
0.2-12.1
4.2-6.3
0.1-4.1
22-25'
11.64-12.98*
(S.U.)
Rotary Kiln
with Cyclone
Preheater'
7-11
3-6*
1-3'
41-51
0.5-2
0.5-4
0.5-4
0-0.3
0-0.5
29-38
No Data
Rotary Kiln
with Grate
Preheater'
2-19
0.5-8*
0.5-4d
6-26
0-2
7-41
14-40
0.9-4.5
0.5-3
4-24
No Data
• Based on 28 tests from 12 facilities responding to 1991 PCA Survey.
b Based on 19 tests from 9 facilities responding to 1991 PCA Survey.
' Kohlhaas, B, et al., 1983, op. cit., p. 623. No information was provided on the size of the population
samples or operational characteristics.
d The responses for the corresponding constituents are aggregated.
' Range based on (1) a Dragon Products Company memorandum (December 6, 1991) from Steve Wallace to
John Bangeman regarding typical analyses of several Dragon Products materials; and (2) a typical analysis of Stable
Sorb at Keystone Cement Company (February 18, 1991).
' Based on EPA sampling data for TCLP and SPLP leachate tests on as generated CKD from seven facilities.
These leachate samples are obtained using an acid solution, so that actual CKD pH values may be higher than
indicated here.
« Based on EPA sampling data for TCLP and SPLP tests on as generated CKD from eight facilities. These
leachate samples are obtained using an acid solution, so that actual CKD pH values may be higher than indicated
here.
-------
3-37
primary purpose of obtaining samples of cement kiln dust to determine its composition.28
Fifteen of the facilities were selected at random from the population of U.S. facilities. The other
five were selected for individual characteristics such as kiln type (e.g., wet or dry), use of
hazardous waste as kiln fuel, and geographic location, so that these factors would be represented
in the final data set. Then, the Agency collected samples of "as generated" and "as managed"
CKD from each of the visited facilities and subjected them to chemical analysis.29
. The Agency selected the following classes of analytes to characterize cement kiln dust:
volatile organic compounds, semi-volatile organic compounds, dioxins, furans, pesticides, poly-
chlorinated biphenyls (PCB), metals, radionuclides, and general chemistry. The individual target
compounds that were determined are identified in the respective sampling project plans. The
project plans are available in the EPA docket for this Report to Congress.30 The organic
analytes were selected primarily from EPA's list of RCRA Hazardous Constituents, which is
presented in Appendix VIII of 40 CFR Part 261.
Because CKD is typically managed in open piles outdoors, the Agency believed it
necessary to also examine the potential for constituents to leach from CKD into the surrounding
environment. Therefore, the Agency also prepared TCLP and SPLP leachates from subsamples
of selected CKD samples and subjected them to analysis for certain of the analytes identified
above, including metals, dioxins, furans, pesticides, radionuclides, and general chemistry.31
TCLP is a laboratory method that simulates the generation and release of leachate from an
improperly disposed solid waste (i.e., a mis-management scenario). In certain cases, EPA uses
an analogous method, SPLP, to simulate land disposal of inorganic wastes in monofills, a
situation that commonly occurs at domestic cement plants.32 Under both leaching procedures,
the analyte concentrations that are measured in the leached extract are compared with a set of
EPA regulatory standards, which are based on 100 times the respective EPA primary drinking
water standards (i.e., toxicity characteristic).
3.3.2 Total Concentrations
The total concentrations (i.e., mass of a particular constituent per mass of CKD) of trace
constituents found in CKD are presented below by the following classes: metals, dioxins and
24 The EPA sampling efforts conducted in 1992 and 1993 are also referred to as the Phase I and Phase II sampling
efforts, respectively. Also, one of the 20 facilities was re-visited and sampled a second time for analysis verification
purposes. The re-visit is not included in the facility counts presented in this document.
29 As generated refers to newly generated CKD that was obtained from the air pollution control device at the kiln.
As managed refers to CKD that was obtained from the facilities' on-site CKD storage or disposal piles. The as-
managed CKD samples were obtained from storage or disposal pile areas containing dust that had typically been in
storage or disposal status for up to six months.
30 The sampling project plans are entitled Cement Kin Industry Sampling and Analysis and Quality Assurance Project
Plan, dated March 1992, and Cement Kin Industry Sampling and Analysis and Quality Assurance Project Plan • Phase 2,
dated May 1993.
31 TCLP stands for Toxicity Characteristic Leaching Procedure. SPLP stands for Synthetic Precipitation Leaching
Procedure. The protocols for these procedures are found in SW-846 analytical methods numbered 1311 and 1312,
respectively.
32 Only the TCLP test and its results have regulatory significance; though the SPLP is an official EPA method, it is
not used for identifying hazardous wastes under 40 CFR Part 261.
-------
3-38
furans, general chemistiy, volatile organic compounds, semi-volatile organic compounds,
pesticides, polychlorinated biphenyls (PCB), and radionuclides.
Metals
EPA collected as generated CKD samples from the 15 facilities that were sampled by
EPA in 1992 and as managed CKD samples from 13 of these facilities.33 EPA analyzed these
samples for the following metals: antimony, arsenic, barium, beryllium, cadmium, chromium,
lead, mercury, nickel, selenium, silver, thallium, and vanadium. With a few exceptions, all of the
metals were detected in all of the as generated and as managed CKD samples. As discussed
below, these data were used to supplement the metals data made available to the Agency by the
cement manufacturing industry and in literature published by the U.S. Bureau of Mines.
Upon conducting several assessments of the types and concentrations of trace metal
constituents in as generated and as managed CKD, the Agency has concluded that a number of
trace metal constituents occur in CKD at highly variable concentrations. Exhibits 3-18 and 3-19
present basic univariate statistics (number of samples, number of non-detected values, and mean,
minimum, maximum, and median concentrations) describing the occurrence of several trace
metal constituents in as generated and as managed CKD, respectively. These data were
generated by separate studies conducted by the Portland Cement Association, the U.S. Bureau of
Mines, and EPA in its 1992 sampling and analysis effort.
Data on the total constituent concentrations of trace metals found in both the as
generated and as managed CKD show that the eight Toxicity Characteristic (TC) metals listed in
40 CFR §261.24 and nine other metals are consistently present (although at variable
concentrations) in CKD. In general, the predominant trace metals include antimony, barium,
lead, manganese, strontium, thallium, and zinc, and the minor trace metals include beryllium,
copper, hexavalent chromium, mercury, nickel, silver, and thallium. A comparison of the data
characterizing the as generated and as managed CKD suggests that the total constituent
concentrations of trace metals found in the as generated CKD are greater by as much as an
order of magnitude; however, the apparent differences may be attributable to changes in
composition of materials charged to the kiln over time.
Intuitively, one would expect a mineral production waste such as CKD to contain the
same types of constituents naturally present in the parent material. The concentration of the
constituents found in such a waste, however, is likely dependent on whether the operator utilized
a concentrating or extracting processing procedure. To help assess whether CKD contains
elevated levels of any specific trace metal, the Agency compared the highest of the average
concentrations of each trace metal observed in the five studies presented in Exhibit 3-18 for the
as generated CKD and the two studies presented in Exhibit 3-19 for the as managed CKD to the
range of trace metals commonly found in native soils. As shown in Exhibit 3-20, the levels of
several of the trace metals found in CKD are within the range commonly found in native soils.
Interestingly, these data suggest that CKD contains seven trace metals (antimony, cadmium, lead,
M One additional set of metals data was generated too late for consideration during the development of this
report.
This data set includes EPA's analysis for the 14 metals in as managed CKD from the six facilities that were sampled
by EPA in 1993 (Phase 2). These data have been included in the EPA RCRA docket for this report for access by
interested parties, and they will be considered by the Agency during its formulation of the final regulatory
determination for CKD.
-------
3-39
mercury, selenium, silver, and zinc) at levels outside the range commonly found in native soils.
These data also show that CKD may have arsenic and strontium at levels that are within the
range of naturally occurring soils but that exceed the average
-------
3-40
Exhibit 3-18
Trace Metal Concentrations in As Generated CKD
(parts per million)
Data Source
EPA SAMPLING1
PCA SURVEY1
PCA REPORT I'
Analyte
Antimony
Aneaic
Bui urn
Beryllium
Cadmium
Chromium
Lead
Mercury
Nickel
Selenium
SUver
Thallium
VtnacSum
Ancuncoy
Anemc
Barium
Beryllium
Cadmium
Chromium
Chromium (VI)
Copper
Lead
MangineM
Mercury
Nickel
Selenium
SUver
Thallium
Vanadum
Zinc
Antimony
Anenic
Barium
Beryllium
Cadmium
Chromium
Lead
Mercury
Silver
Thallium
Number of
Sampler
17
17
17
17
17
17
17
17
17
17
17
17
17
1
3
1
3
3
3
2
1
3
5
3
3
1
1
1
3
3
6
6
6
6
6
6
6
6
6
6
Number of
Non-Detected
Value.
1
0
0
1
0
0
0
3
0
0
0
0
0
1
0
0
3
0
0
0
0
0
0
3
0
0
1
1
0
0
0
2
0
0
0
0
0
3
0
0
Mean
7.7
6.9
172.1
0.71
13.2
26.6
388.4
1.0
19.0
17.3
6.9
17.1
41.6
0.53
34J
150.0
0.517
8.05
39.0
7.82
28.4
210-3
211.2
0.104
1SJ
6-5
0.504
4.616
33.5
104.3
nit,
20.4
IKS
3.88
18.6
35.9
283.7
0.062
9.17
88.0
Minimum
1.77
2.1
no
0.158
0.89
11.5
5.1
0.005
6.9
2.5
1.1
0.99
6.6
0.53
3.7
150.0
0-509
3.0
317
7.05
28J
151.0
200.0
0.100
10.0
6J
OJ04
4.616
23.0
86.0
37.8
3.726
101 X)
2.86
4.73
18.1
53.2
0.003
5.71
68.6
Maximum
27.2
20J
779 JO
1.6
80.7
81.7
1,490.0
14.4
39.0
109.0
22.6
108.0
204.0
0.53
53.0
150.0
0.523
111
49.0
8.59
28.7
270.0
222.0
0.107
23^
6J
OJ04
4.616
39J
116.0
161.0
80.7
323.0
4.64
44.0
58-5
819.0
OJ05
117
146.0
Median
6-2
4.9
103.0
0.59
4.6
'18.1
287.0
0.11
15.9
11J
3.7
3J
25.9
OJ3
46.2
150.0
0.521
9.05
35J
7.82
28.4
210.0
212.0
0.106
21.1
6.5
OJ04
4.616
38J
111.0
142.5
tM
141 J
3.81
13J
361
171.0
0.015
933
79.0
-------
3-41
Exhibit 3-18 (continued)
Trace Metal Concentrations in As Generated CKD
(parts per million)
Data Source
PCA REPORT 2*
BOM 1C 8885-
Analytc
Antony
Anew
Barhm
Beryllium
CKfanhm
Oromh.,
Lad
Lithium
Mercury
Nictel
Selenium
Silver
TblllJiMn
Amjnmy
Aneaic
C^mixm
Chromium
Copper
Lead
Lithium
Maotuete
Nickrl
Silver
Strootium
Zinc
Number of
Samples
95
95
95
95
95
95
95
0
95
95
94
95
95
113
113
113
113
113
113
113
113
113
113
113
113
Number of
Non-Detected
Values
86
45
0
1
14
0
0
0
30
27
40
3
6
71
0
17
0
0
5
15
0
51
54
0
0
Mean
0.395
13.0
185.8
0.645
8.83
40.8
434.5
0
17.3
0.49
18.3
10.3
40.6
3.3
23.8
20.0
41.6
30.1
252.9
18.0
385.6
19.3
5.1
669.0
462.0
Minimum
0.083
IJ23
35.0
0.032
0.008
8.25
33.5
0
1.0
0.001
0.227
3.549
0.109
0.701
1.3
0.687
11.0
7.0
11.335
1.754
63.0
5.421
1.291
100.0
32.0
Maximum
3.43
159.0
1,402.0
3.54
59.6
293.0
7,390.0
0
60.0
25.5
307.0
40.7
776.0
70.0
518.0
352.0
172.0
206.0
1,750.0
76.0
2,410.0
91.0
17.0
8,800.0
8,660.0
Median
0.21
9.07
133.0
0.539
3.27
29.1
188.0
0
14.0
0.045
7.^
9.28
8.96
0.83
10.0
7.6
35.0
24.0
148.0
16.0
284.0
16.0
4.7
400.0
167.0
' Data from EPA's 1992 sampling effort.
b Data from the 1991 PCA Survey of U.S. cement plants.
c Portland Cement Association, 1992. An Analysis of Selected Trace Metals in Cement and Kiln Dust (Draft).
PCA Report SP109T, Skokie, IL.
d Portland Cement Association, 1992. An Analysis of Selected Trace Metals in Cement and Kiln Dust. PCA
Report SP109T, Skokie, IL, 56 pages.
* Haynes, B., and G. Kramer, 1982. Characterization of U.S. CKD. Bureau of Mines Information Circular
(1C) 8885, U.S. Department of Interior. Bureau of Mines. Office of Assistant Director. Minerals and Materials
Research, Washington, D.C
-------
3-42
Exhibit 3-19
Trace Metal Concentrations in As Managed CKD
(parts per million)
Data Soiree
EPA SAMPLING*
PCA SURVEY*
Analjte
Antimony
Anenic
Barium
Beryllium
f*»/4fMiiim
Chromium
Lad
Mcrauy
Nickel
Selenium
Silver
Thallium
Vanadum
Antimony
Anenic
Barium
BeryUium
Cadmium
Chromium
Chromium (VI)
Copper
Lead
MangtneM
Mercury
Nickel
Seleaium
Silver
Strontium
Thallium
Vanadium
Zinc
Number of
Samples
14
U
U
14
14
14
14
14
14
14
14
14
14
37
44
42
34
44
44
7
1
44
2
42
11
34
41
1
36
1
2
Number of
Non-Detected
Valuei
2
0
0
2
0
0
0
3
0
0
2
0
0
13
11
0
5
0
0
1
0
0
0
24
0
9
2
0
18
0
0
Mean
6.5
7.7
144.5
0.68
114
35.0
359.1
0.121
19.4
10.7
4.2
4.1
33.3
• 27.7
16.0
235.2
1.1
24.3
40.1
0.11
7.15
857.9
165.9
1.0
214
15.0
7.4
4224
9.7
30.0
128.6
Minimum
1.581
2.1
39.8
0.175
0.62
9.6
40.6
0.009
6J
1.4
0.348
1.1
7.6
0.099
0.514
10
0.141
0.41
3.3
0.02
7.15
3.12
IB O
0.002
3.6
0.518
0.187
4224
2.0
30.0
37.2
Maximum
10.9
194
560.0
1J
27.4
110.0
863.0
0430
54.7
43.9
17 1
14.6
120.0
360.0
811
635.0
6.7
8S.7
132.0
0.23
7.15
4.230.0
2084
4.7
46J
100.0
57.9
4224
68.6
30.0
220.0
Median
6.6
6.4
136.5
.52
10.1
21.4
380.5
0.075
14.9
7.7
1.95
2J
19.6
7.2
9.7
207.0
0.69
17 J
34.8
0.13
7.15
441.0
16S.9
1.1
16.0
7J
3.0
4224
5.7
30.0
128.6
• Data from EPA's 1992 sampling effort.
k Data from the 1991 PCA Survey of U.S. cement plants.
-------
3-43
Exhibit 3-20
Trace Elements Commonly Found in Native Soils (mg/Kg)
Trace Elements
Antimony
Arsenic
Barium
Beryllium
Cadmium
Chromium
Copper
Lead
Manganese
Mercury
Nickel
Selenium
Silver
Thallium
Vanadium
Zinc
Strontium
Common Ranges
2- 10
1-50
100 - 3,000
0.1 -40
0.01 - 0.7
1 - 1,000
2- 100
2-200
20 - 3,000
0.01 - 0.3
5-500
0.1 -2
0.01 - 5
NA
20-500
10-300
50- 1,000
Average
Concentrations
NA
5
430
6
0.06
100
30
10
600
0.03
40
0.3
0.05
NA
100
50
200
As Generated
(Highest Average)'
112.8
34.3
185.8
3.8
20
41.6
30.1
434.5
385.6
17.3
19.3
18.3
10.3
40.6
41.6
462
669
As Managed
(Highest Average)'
27.7
16
235.2
1.1
24.3
40.1
7.1
857.8
165.9
1
22.4
15
7.4
9.7
33.3
128.6
422.8
Source: Hazardous Waste Land Treatment. Table 6.46 - Trace Elements of Soils, U.S. Environmental Protection
Agency, Cincinnati, Ohio, PB89-179014, April 1983, page 273.
* EPA used the highest of the average concentrations of each trace metal observed in the five studies
presented in Exhibit 3-18 for the as generated CKD and the two studies presented in Exhibit 3-19 for the as managed CKD.
native soil concentration by a factor of two or more. CKD, therefore, could be a potential
contributor of these metals at higher than natural levels to the environment. For some metals
(e.g., arsenic), the high end of the naturally occurring range of concentrations in native soils may
present risk to human health and the environment. The potential risks to human health and the
environment posed by these metals are identified and discussed in Chapters 5 and 6.
Dioxins and Furans
EPA analyzed dioxins and furans in as generated and as managed CKD samples from 11
facilities. A number of the dioxin and furan target compounds were detected in both as
generated and as managed CKD. Because these compounds were detected rather consistently in
CKD, the Agency continued with analysis of the analytical data. Analyses were performed to
-------
3-45
non-hazardous waste burning facilities. These data show that dioxins and dibenzofurans
(HpCDD, HpCDF, HxCDD, HxCDF, OCDD, OCDF, PeCDD, PeCDF, TCDD (including
2,3,7,8-TCDD) and TCDF also
-------
3-44
determine whether the data from the 1992 EPA sampling could be pooled with the data from the
1992 sampling, and to determine whether there are relationships between dioxin and furan levels
in CKD and two important operating factors: (1) use of hazardous waste as a fuel; and (2) use
of wet process kilns versus dry process kilns.
One facility that was sampled during the Phase I sampling effort, River Cement
Company, was re-sampled as part of the Phase II effort. This was done in part to conduct
confirmatory analyses on the levels of dioxins that were reported from the Phase I effort at this
facility. The CKD from this facility exhibited, by far, the highest levels of dioxins observed
during the EPA sampling program. The results of the Phase II sampling did confirm the
presence of dioxins in CKD at this facility, although the Phase II results showed dioxin levels
generally three to four times lower than those measured during the Phase I sampling effort. As
such, the Agency considers this facility to be non-typical of the industry in this respect. The
Agency believes that the production of dioxins and furans in processes such as these may be
heavily influenced by the incinerator or kiln exhaust gas temperatures, specifically at the inlet to
the air pollution control devices (APCD) (e.g., baghouses, electrostatic precipitators). The levels
of these organic constituents of CKD may be controllable through relatively minor process
modifications to reduce exhaust gas or APCD inlet temperatures.
Exhibits 3-21 through 3-24 provide dioxins and dibenzofurans data resulting from
sampling by EPA of as generated and as managed dust at six cement plants in 1993. It is worthy
of note that dioxins and dibenzofurans were found in CKD samples collected from both facilities
burning hazardous waste fuels and those not burning hazardous waste fuels.
Exhibit 3-21 presents total constituent concentration data for dioxins and dibenzofurans
obtained from analyses of as generated CKD produced by both hazardous waste burning and
non-hazardous waste burning facilities.
These data indicate that dioxins and ______ ______ _____ _______
dibenzofurans (HpCDD, HpCDF,
HxCDD, HxCDF, OCDD, OCDF,
PeCDD, PeCDF, TCDD (including
2,3,7,8-TCDD) and TCDF are present
at very low concentrations in CKD
generated by both hazardous and non-
hazardous waste fuel burning facilities.
Most of the homologs, however, were
detected at concentrations below 100
ppt, while several samples had
homolog concentrations approaching
one ppb. Only one homolog was
detected at a concentration exceeding
one ppb (total HxCDD at 1.5 ppb).
These results correspond with the
results obtained from EPA's Phase I -_________I^___0^ ^ „,
analyses of dioxins and dibenzofurans,
with the exception of one of seven samples where total HpCDD, HxCDD, HxCDF, PeCDD,
PeCDF, TCDD, and TCDF were all detected at concentrations exceeding one ppb.
DIOXINS AND DIBENZOFURANS
* Tetrachlorinated dibenzo-p-dioxins (TCDD)
* Tetrachlorinated dibenzofurans (TCDF)
* Pentachlorinated dibenzo-p-dioxins (PeCDD)
* Pentachlorinated dibenzofurans (PeCDF)
* Hexachlorinated dibenzo-p-dioxins (HxCDD)
* Hexachlorinated dibenzofurans (HxCDF)
* Heptachlorinated dibenzo-p-dioxins
(HpCDD)
* Heptachlorinated dibenzofurans (HpCDF)
* Octachlorinated dibenzo-p-dioxins (OCDD)
* Octachlorinated dibenzofurans (OCDF)
Exhibit 3-22 presents total constituent concentration data for dioxins and dibenzofurans
obtained from analyses of as managed CKD generated by both hazardous waste burning and
-------
3-46
Exhibit 3-21
Total Concentrations of Dioxins and Dibenzoftirans in As Generated CKD (/ig/Kg)
ANALYTE
1,23,4,6,7,8-HpCDD
Total HpCDD
1,23,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
Total HpCDF
1,23,4,7,8-HxCDD
1,23,6,73-HxCDD
1,23,7,8,9-HxCDD
Total HxCDD
1,2,3,4,7,8-HxCDF
U3.6,73-HxCDF
1,2,3.7.8,9-HxCDF
23.4,6,7,8-HxCDF
Total HxCDF
OCDD
OCDF
1,2,3.7,8-PeCDD
Total PeCDD
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
Total PcCDF
2,3.7,8-TCDD
Total TCDD
23,7,8-TCDF
Total TCDF
HW-1
Sample 1
0.0019
0.0037
< 0.00056
< 0.00052
0.0043
< 0.00079
< 0.00064
< 0.00095
0.012
< 0.00073
< 0.00044
< 0.00044
J 0.0004
J 0.0004
0.0036
< 0.001
< 0.0005
0.0021
< 0.00046
< 0.00053
0.00039
< 0.00031
0.0054
0.0005
0.0028
HW-1
Sample 2
0.0018
0.0037
< 0.001
< 0.001
N.A.
< 0.0014
< 0.0021
< 0.0019
0.0076
< 0.0023
< 0.0016
< 0.0017
< 0.0015
N.A.
0.0036
< 0.0019
< 0.0012
N.A.
< 0.00089
< 0.00072
N.A.
< 0.0011
0.0035
< 0.00065
N.A.
HW-2
B 0.25
B 0.55
0.037
0.0074
B 0.067
0.025
0.049
0.041
1.5
0.024
0.025
0.014
< 0.037
0.23
B 0.1
0.01
0.03
0.85
0.033
0.064
0.53
0.0056
B 0.44
0.038
0.96
HW-3
B 0.0048
B 0.011
0.0024
< 0.0017
B 0.0024
< 0.0014
< 0.0017
< 0.00093
0.00059
0.0028
0.0028
0.00096
0.0023
0.024
B 0.034
0.0017
< 0.0012
N.A.
0.0061
0.0038
0.063
< 0.00088
N.A.
0.0044
0.076
NH-1
B 0.0051
B 0.0098
< 0.0013
< 0.0013
0.00047
< 0.0012
< 0.0018
< 0.0014
N.A
< 0.0011
< 0.00076
< 0.00073
< 0.00076
N.A.
B 0.018
< 0.0032
< 0.0015
N.A.
0.00052
< 0.00046
0.00052
< 0.0013
N.A.
0.00039
0.00039
NH-2
J 0.0011
0.0026
J 0.00028
< 0.00069
B 0.0013
< 0.00083
< 0.00066
< 0.001
N.A.
< 0.0007
< 0.00049
< 0.00085
J 0.00057
0.00089
B 0.0046
< 0.001
< 0.00057
N.A.
< 0.00051
< 0.00033
0.00071
< 0.00037
N.A.
< 0.00087
0.014
NH-3
< 0.003
0.0079
< 0.0076
< 0.00079
N.A.
< 0.0012
< 0.00096
< 0.0015
0.012
< 0.00096
< 0.00099
< 0.0012
< 0.00074
0.0019
B 0.0079
< 0.0014
< 0.0021
0.01
< 0.00058
< 0.00057
N.A.
< 0.0016
0.0091
< 0.00099
N.A.
MAXIMUM
0.25
0.55
0.037
0.0074
0.067
0.025
0.049
0.041
1.5
0.024
0.025
0.014
0.037
0.23
0.1
0.01
0.03
0.85
0.033
0.064
0.53
0.0056
0.44
0.038
0.96
AVERAGE
0.03824
0.08410
0.00716
0.00191
0.01509
0.00455
0.00812
0.00695
0.30644
0.00466
0.00458
0.00284
0.00618
0.05144
0.02453
0.00289
0.00530
0.28737
0.00601
0.01006
0.11892
0.00159
0.11450
0.00654
0.21064
AVERAGE
DETECTED
0.04412
0.08410
0.01323
0.00740
0.01509
0.02500
0.04900
0.04100
0.30644
0.01340
0.01390
0.00748
0.00109
0.05144
0.02453
0.00585
0.03000
0.28737
0.01321
0.03390
0.11892
0.00560
0.11450
0.01082
0.21064
N.A.
"B'
"J"
Average
Delected
Not Detected, the Associated Value is the Detection Limit.
Detection limits are not available for total concentrations.
The Constituent was Detected in an Associated Blank.
KS
The Concentration is an Estimate. The Constituent Was Positively Identified at a Trace Value
or is a Nontarget Constituent for which no Calibration was Performed.
= The average of the samples, excluding those that were not delected.
HW-1 -- Keystone Cement Co., Bath, PA
HW-2 -- River Cement Co., Festus, MO
HW-3 -- Heartland Cement Co., Independence,
NH-1 -- Ash Grove West, Inc., Inkom, ID
NH-2 -- Calaveras Cement Co., Tehachapi, CA
NH-3 -- Holnam, Inc., Artesia, MS
-------
3-47
Exhibit 3-22
Total Concentrations of Dioxins and Dibenzofurans in As Managed CKD (/tg/Kg)
ANALYTE
1,2,3,4,6,7,8-HpCDD
Total HpCDD
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
Total HpCDF
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
Total HxCDD
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
Total HxCDF
OCDD
OCDF
1,2,3,7,8-PeCDD
Total PeCDD
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
Tola! PeCDF
2,3,7,8-TCDD
Total TCDD
2,3,7,8-TCDF
Total TCDF
HW-1
Sample 1
0.012
0.023
< 0.0012
< 0.00063
N.A.
< 0.0011
0.0018
< 0.0016
0.057
< 0.00089
< 0.00049
< 0.00035
< 0.00065
N.A.
0.009
< 0.0024
< 0.00076
0.032
< 0.0012
< 0.0006
N.A.
< 0.001
0.036
< 0.00033
N.A.
HW-1
Sample 2
0.012
0.023
< 0.00082
< 0.00062
N.A.
< 0.0027
0.0023
< 0.0015
0.06
< 0.0016
< 0.00077
< 0.00083
J 0.00045
J 0.0045
0.008
< 0.0023
< 0.0019
0.029
< 0.0015
< 0.0011
N.A.
< 0.0011
0.035
< 0.0015
0.0006
HW-2
B 0.24
B 0.54
0.1
0.012
B 0.15
0.022
0.038
0.026
0.86
0.045
0.042
0.016
0.067
0.46
B 0.19
0.022
0.021
0.55
0.038
0.085
0.97
0.0034
B 0.24
0.029
1.1
HW-3
B 0.014
B 0.027
0.0014
< 0.00077
B 0.0025
< 0.00054
< 0.00088
< 0.00057
0.0081
0.00048
0.00046
< 0.00033
0.00074
0.0027
B 0.046
0.0016
< 0.00057
N.A.
0.00048
0.00044
0.0054
< 0.00054
B 0.002
0.00068
0.0091
NH-1
J 0.00067
J 0.00067
< 0.00047
< 0.00078
N.A.
< 0.00052
< 0.00085
< 0.00062
N.A.
< 0.00035
< 0.0003
< 0.00065
J 0.00043
J 0.00043
B 0.0035
< 0.0011
< 0.00045
N.A.
< 0.00029
< 0.00031
N.A.
< 0.00059
N.A.
< 0.00037
N.A.
NH-2
< 0.0021
N.A.
< 0.00039
< 0.00068
N.A.
< 0.00095
< 0.00068
< 0.00081
N.A.
< 0.00033
< 0.00027
< 0.00063
J 0.00042
J 0.00042
B 0.0027
< 0.0029
< 0.00043
N.A.
< 0.00047
< 0.00041
N.A.
< 0.001 1
N.A.
< 0.00073
N.A.
NH-3
0.0039
0.014
< 0.00087
< 0.0014
N.A.
< 0.0016
< 0.0016
< 0.0018
0.02
< 0.00063
< 0.00021
< 0.0011
< 0.0012
N.A.
B 0.011
< 0.0032
< 0.0017
0.0071
< 0.00078
< 0.00096
0.00084
< 0.0013
0.0094
< 0.0012
N.A.
MAXIMUM
0.24
0.54
0.1
0.012
0.15
0.022
0.038
0.026
0.86
0.045
0.042
0.016
0.067
0.46
0.19
0.022
0.021
0.55
0.038
0.085
0.97
0.0034
0.24
0.029
1.1
AVERAGE
0.040667
0.104612
0.015021
0.002411
0.076250
0.004201
0.006587
0.004700
0.201020
0.007040
0.006357
0.002841
0.010127
0.093610
0.038600
0.005071
0.003830
0.154525
0.006103
0.012689
0.325413
0.001290
0.064480
0.004830
0.369900
AVERAGE
DETECTED
0.047095
0.104612
0.050700
0.012000
0.076250
0.022000
0.014033
0.026000
0.201020
0.022740
0.021230
0.016000
0.013808
0.093610
0.038600
0.011800
0.021000
0.154525
0.019240
0.042720
0.325413
0.003400
0.064480
0.014840
0.369900
N.A.
•B'
T
Average
Detected
Not Detected, the Associated Value is the Detection Limit.
Detection limits are not available for total concentrations.
The Constituent was Detected in an Associated Blank.
The Concentration is an Estimate. The Constituent Was Positively Identified at a Trace Value
or is a Nontarget Constituent for which no Calibration was Performed.
The average of the samples, excluding those that were not detected.
HW-1 - Key stone'Cement Co., Bath, PA
HW-2 - River Cement Co., Festus, MO
HW-3 - Heartland Cement Co., Independence, KS
NH-1 - Ash Grove West, Inc., Inkom, ID
NH-2 - Calaveras Cement Co., Tehachapi, CA
NH-3 - Holnam, Inc., Artesia, MS
-------
3-48
are present at very low concentrations in the as managed CKD generated by both hazardous and non-
hazardous waste fuel burning facilities. As was the case in the analyses of the as generated CKD, the
majority of the dioxins and dibenzofurans were detected at concentrations below 100 ppt, while
several samples had homolog concentrations approaching one ppb. Only one homolog was detected at
a concentration exceeding one ppb (this time, total TCDF was detected at 1.1 ppb). These results
also correspond with the results obtained from EPA's Phase I analyses of dioxins and dibenzofurans,
with the exception of one of seven samples where total HpCDD, HxCDD, HxCDF, PeCDF, and
TCDF were detected at concentrations exceeding one ppb. As summarized in Exhibit 3-23, the levels
of dioxins and dibenzofurans detected in CKD appear to be slightly higher than those levels detected
in samples of the as managed CKD, with the exception of total HxCDD, PeCDD, and TCDD
homologs. The significance of this difference is not known; however, it is likely explained by both
sample and analytical variation.
Finally, Exhibit 3-24 presents a summary of EPA's dioxin and furan analytical data collected
in 1992 and 1993 that have been normalized to 2,3,7,8-TCDD equivalence.
Because dioxins and furans were detected in CKD from all 11 sampled facilities, the Agency
believes it appropriate to carry consideration of these compounds through the risk assessment and
decision rationale components of this report, which means that the presence of these compounds in
CKD will influence the Agency's decisionmaking on the RCRA regulatory status of CKD.
General Chemistry
The following general chemistry target analytes were analyzed in all of the samples of as
generated and as managed CKD obtained at 15 of the facilities sampled by EPA: chloride, fluoride,
sulfate, sulfide, total organic carbon, total cyanide, and moisture content (or percent solids). Except
for sulfide and sulfate, the same target analytes were also analyzed in all of the CKD samples
obtained at the other five facilities sampled by EPA. Except for cyanide, the general chemistry target
compounds were analyzed for general information, such as comparison with similar basic composition
data supplied by the industry.
With one exception, total cyanide was not detected in any of the CKD samples. At one
facility, cyanide was reported as detected in the as generated and as managed CKD samples.
However, the reported levels were less that the method detection limit.
The Agency believes that no further consideration should be given to total cyanide for the
purposes of this report because it does not appear to be present in CKD on an industry-wide basis.
All of the analytical data from this effort are available in the docket for this Report to Congress.
Volatile Organics
Because of the nature of as generated CKD (i.e., temperature of 300° F or more, very dry
matrix), the Agency considered it unlikely that volatile organic compounds would be present in this
material. To confirm this, volatile organic compounds were analyzed in as generated CKD samples
from 11 facilities.
The chemical analysis of the as generated CKD samples revealed a number of instances in
which volatile organics were detected. The following discussion identifies those instances and
presents the Agency's conclusions regarding their validity and implications!
-------
3-49
Exhibit 3-23
Summary of Dioxin and Dibenzofuran Concentrations in CKD
ANALYTE
1,2,3,4,6,7,8-HpCDD
Toul HpCDD
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
Total HpCDF
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDD
1,2,3,7,8,9-HxCDD
Toul HxCDD
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
Toul HxCDF
OCDD
OCDF
1,2,3,7,8-PeCDD
Toul PeCDD
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
Toul PeCDF
2,3,7,8-TCDD
Toul TCDD
2,3,7,8-TCDF
Toul TCDF
AS GENERATED - TOTAL (ug/Kg)
MAXIMUM
0.25
0.55
0.037
0.0074
0.067
0.025
0.049
0.041
1.5
0.024
0.025
0.014
0.037
0.23
0.1
0.01
0.03
0.85
0.033
0.064
0.53
0.0056
0.44
0.038
0.96
AVERAGE
0.03824
0.08410
0.00716
0.00191
0.01509
0.00455
0.00812
0.00695
0.30644
0.00466
0.00458
0.00284
0.00618
0.05144
0.02453
0.00289
0.00530
0.28737
0.00601
0.01006
0.11892
0.00159
0.11450
0.00654
0.21064
AVERAGE
DETECTED
0.04412
0.08410
0.01323
0.00740
0.01509
0.02500
0.04900
0.04100
0.30644
0.01340
0.01390
0.00748
0.00109
0.05144
0.02453
0.00585
0.03000
0.28737
0.01321
0.03390
0.11892
0.00560
0.11450
0.01082
0.21064
AS GENERATED - TCLP (ug/L)
MAXIMUM
0.00003
0.000032
< 0.000017
< 0.000019
0.000007
< 0.000018
< 0.000021
< 0.000019
N.A.
< 0.00001
< 0.00001
< 0.000024
0.000021
0.000006
0.00017
0.000055
< 0.000024
N.A.
< 0.000017
< 0.000008
N.A.
< 0.000022
0.000005
< 0.00001
N.A.
AVERAGE
0.000021
0.000018
< 0.000011
< 0.000014
0.000007
< 0.000013
< 0.000012
< 0.000012
N.A.
< 0.000007
< 0.000006
< 0.000013
0.000011
0.000005
0.000073
0.000028
< 0.000012
N.A.
< 0.000009
< 0.000005
N.A.
< 0.000013
0.000005
< 0.000007
N.A.
AVERAGE
DETECTED
0.000019
0.000024
0.000007
0.000005
0.000005
0.000080
0.000011
0.000005
AS MANAGED - TOTAL (ug/Kg)
MAXIMUM
0.24
0.54
0.1
0.012
0.15
0.022
0.038
0.026
0.86
0.045
0.042
0.016
0.067
0.46
0.19
0.022
0.021
0.55
0.038
0.085
0.97
0.0034
0.24
0.029
1.1
AVERAGE
0.040667
0.104612
0.015021
0.002411
0.076250
0.004201
0.006587
0.004700
0.201020
0.007040
0.006357
0.002841
0.010127
0.093610
0.038600
0.005071
0.003830
0.154525
0.006103
0.012689
0.325413
0.001290
0.064480
0.004830
0.369900
AVERAGE
DETECTED
0.047095
0.104612
0.050700
0.012000
0.076250
0.022000
0.014033
0.026000
0.201020
0.022740
0.021230
0.016000
0.013808
0.093610
0.038600
0.011800
0.021000
0.154525
0.019240
0.042720
0.325413
0.003400
0.064480
0.014840
0.369900
"<•
N.A.
Average
Detected =
Not Delected, the Associated Value is the Detection Limit.
Detection limits ire not available for toUl concentrations.
The Constituent was Detected in an Associated Blank.
The Concentration is an Estimate. The Constituent Was Positively Identified at a Trace Value
or is a NonUrget Constituent for which no Calibration was Performed.
The average of the samples, excluding those that were not detected.
-------
3-50
Exhibit 3-24
Summary of Combined 1992-1993 Dioxin/Furan Sampling Results
CKD 2,3,7,8, TCDD Toxicity Equivalence (ppm)
Plant
River-Festus, MO*
Hoi nam-Tij eras, NM
Heartland-Independence, KS"
LaFarge-Fredonia, KS"
Giant-Harleyville, SC"
Ash Grove-Inkom, ID
Independent-Catskill, NY
Calaveras-Tehachapi, CA
Keystone-Bath, PA*
Holnam-Artesia, MS
Ash Grove-Chanute, KS'
As Generated
2.475 x 104 b
3.2 x 10'5 k
3.6 x 10^
1.5 x 10-6
8.2 x lO'7
1.3 x lO'7
4.0 x lO'8
7.5 x 10'8
6.7 x lO'8 b
8.0 x 10-'
ND
As Managed
1.955x 10"4 b
ND
6.8 x ia7
9.0 x 10-6 b
4.3 x KT6
5.3 x ia8
ND
4.5 x lO"8
3.6 x ia7
5.0 x ia8
ND
Sample Year
1992 and 1993
1992
1993
1992
1992
1993
1992
1993
1993
1993
1992
' - Hazardous Waste Burner
b - Denotes average of two samples
ND denotes a non-detect
In several instances the volatile organic compound methylene chloride was detected in as
generated CKD samples. In each instance, however, this compound was also detected in one or more
of the corresponding quality assurance blanks, including method, trip, field, and equipment blanks.
Therefore, its presence in the CKD sample is attributed to contamination of the sample. Methylene
chloride is a common laboratory contaminant, the presence of which at low concentrations is not
unexpected and is usually attributed to contamination of the ambient atmosphere in the laboratory.
In several instances, volatile organic compounds were detected in CKD at only one or two
facilities, usually near the detection limit. The compounds acetone, carbon disulfide, chlorobenzene,
chloroform, ethyl benzene, tetrachloroethene, trichloroethene, and xylene are in this category.
Acetone was detected at only one facility. Carbon disulfide, chlorobenzene, chloroform, ethyl
benzene, and tetrachloroethene were measured just above the detection limit at only one facility each.
Trichloroethene was measured just above the detection limit at only two facilities, and xylene was also
detected at only two facilities. The Agency believes that no further consideration should be given to
these compounds for purposes of this report because their measured levels are near the analytical
detection limit and they do not appear to be present in CKD on an industry-wide basis.
-------
3-51
Two volatile organic compounds were detected several times, but only in the samples
analyzed by one of the two laboratories that conducted the volatile organics analyses. The Agency
believes that their detection is due solely to laboratory contamination or artifacts of the process of
analyzing the samples that are unique to the laboratory that reported the analytes as detected. The
compounds acetonitrile and trichlorofluoromethane are in this category. Acetonitrile was detected in
all of the CKD samples analyzed at one laboratory but in none of the CKD samples analyzed at the
other laboratory. Trichlorofluoromethane was detected in five of the CKD samples analyzed at one
laboratory but in none of the CKD samples analyzed at the other laboratory. The Agency believes
that no further consideration should be given to these compounds for purposes of this Report to
Congress because their presence is due solely to laboratory contamination or artifacts of the analytical
procedures used.
Four other volatile organic compounds were reported as detected in several CKD samples and
are believed to be present due to sample contamination during the process of collecting and analyzing
the samples or from artifacts of the analytical procedures. These compounds are benzene, 2-
butanone, isobutyl alcohol and toluene. Additionally, the as generated CKD sample from one
facility34 had considerably higher levels of these compounds than did samples from the other
facilities. The Agency believes that the integrity of the sample is suspect and therefore should not be
considered. The Agency believes that no further consideration should be given to these compounds
for purposes of this Report to Congress because their presence is believed to be due to laboratory
contamination or artifacts of the analytical procedures.
No as managed CKD samples were subjected to analysis for volatile organics because the
Agency believed that any such compounds, even if present in the as generated CKD, would have
separated from the CKD due to prolonged exposure of the CKD to the elements (i.e., up to six
months).
Semi-Volatile Organics
As was the case with volatile organic constituents, the Agency also considered it unlikely that
volatile organic compounds would be present in this material (i.e., temperature of 300° F or more,
very dry matrix). To confirm this, semi-volatile organic compounds were analyzed in as generated
and as managed CKD samples from six facilities. None of the semi-volatile compounds were detected
in either the as generated or as managed CKD samples.
The Agency believes that the semi-volatile organic compounds should not be considered
further for purposes of this Report to Congress because they do not appear to be present in CKD, and
accordingly, has not included them in the analysis that follows later in this report.
Pesticides
Thirteen target pesticide compounds were analyzed in as generated and as managed CKD
samples from 11 facilities. Three of the target pesticide compounds were detected at a total of two
facilities. Endrin and heptachlor epoxide were both detected in as generated CKD at one facility.
Endosulfan was detected in both as generated and as managed CKD at another facility.
Calaveras Cement Company, Tehachapi, CA facility. The results of the split sample analysis were similar.
-------
3-52
Because only three pesticide compounds were detected in CKD samples at only one facility
each, the Agency believes that the pesticides do not warrant further consideration for this Report to
Congress because they do not appear to be present in CKD on an industry-wide basis. Accordingly,
the pesticide compounds are not included in the analysis that follows later in this report.
PCBs
Seven PCB compounds were analyzed in as generated and as managed CKD samples from 11
facilities. None of the target PCB compounds were detected in either the as generated or as managed
CKD samples obtained by EPA.
The Agency believes that PCBs should not be considered further for purposes of this Report
to Congress because they do not appear to be present in CKD, and accordingly, has not included
them in the analysis that follows later in this report.
Radionuclides
Raw materials are the major source of common, natural radionuclides that may be found in
cement kiln dust because such materials are part of the earth's crust. Therefore, for example, if the
limestone used in the manufacture of cement was slightly enhanced with background levels of the
radioisotopes of uranium or thorium, slightly enhanced levels of these radionuclides would be
expected to be present in the cement kiln dust. In order to properly evaluate any potential risks
associated with management of CKD, the Agency conducted radiochemical analyses on samples of
CKD for a number of the natural elements. The Agency also decided to analyze the samples for
man-made elements, which could be present in raw materials due to their prior release in the
environment, for example, from fallout from above-ground nuclear weapons testing and from the
explosion of a satellite containing plutonium in the earth's atmosphere.
The Agency conducted gross alpha, gross beta, and gamma analyses as well as isotopic
analysis involving chemical separations for the following specific analytes: radium-226, plutonium-
238, plutonium-239, uranium-234, uranium-235, uranium-238, thorium-227, thorium-228, thorium-
230 and thorium-232.35iM Gamma analyses and plutonium isotopic analyses were performed on the
as generated CKD samples from all 20 facilities and also on the as managed CKD samples from six
of these facilities. Gross alpha and beta analyses and isotopic analyses for the other specific analytes
listed above were determined for all of the CKD samples from six of the facilities.
Several of the naturally occurring radionuclides were detected in the CKD samples collected
by EPA, including isotopes of lead, radium, uranium, thorium, and potassium. With the possible
exception of uranium and potassium, the radiological activities determined for the naturally occurring
radionuclides are considered to be within the range of activities normally found in environmental
M The radionuclide analyses were performed by EPA's National Air and Radiation Environmental Laboratory.
The analytical methods that were used for the samples associated with this study are presented in the EPA document
entitled Eastern Environmental Radiation Facility Radiochemistry Procedures Manual (doc. # EPA 520/5-84-006).
16 The following radionuclides can be detected by the gamma spectroscopy method used in this sampling and
analysis program: Be-7, Na-22, K-40, Cr-51, Mn-54, Co-56, Co-57, Co-58, Fe-59, Co-60, Zn-65, Sr-85, Y-88, Zr-95,
Nb-95, Ru-103, Ru-106, Cd-109, Ag-110, Sn-113, Sb-124, Sb-125,1-131, Ba-133, Cs-134, Cs-136, Cs-137, Ba-140.
La-140, Ce-141, Ce-144, Hg-203, Bi-206, Bi-207, T1-2Q8, Pb-212, Bi-214, Pb-214, Ra-226, Ra-228, U-235, U-238.
-------
3-53
samples of comparable composition. That is, the activity levels observed for these radionuclides
measured in CKD are expected to be no different than, for example, those found in samples of soil
and rock that are randomly selected and sampled.
The activity levels for the uranium isotopes may be considered to be slightly higher than
average values for these isotopes found naturally in soils and rocks. However, based on the
equilibrium state of the isotopes, the levels are consistent with the expected values for environmental
samples containing natural uranium which has not undergone any isotopic separation, enhancement, or
depletion process. This demonstrates that the observed uranium activity levels are due to unaltered
natural uranium. In addition, there is great variability in the natural concentrations of the isotopes of
uranium in soils and rocks. Also, the incineration process in a cement kiln could reasonably be
expected to slightly increase the concentration of the isotopes of uranium due to the substantial
reduction in volume of the fuels burned and materials processed in the kiln. Therefore, the activity
levels of uranium isotopes and decay products, and thorium isotopes as well, are consistent with what
would be expected in the residual material resulting from the processing of materials containing
naturally-occurring radionuclides.
For the man-made elements, the Agency subjected the samples from the 1992 EPA sampling
program to gamma scan analysis. Certain samples from four of the facilities were also subjected to
gross alpha and gross beta analyses, and isotopic analysis involving chemical separations for isotopes
of plutonium, uranium and thorium. Based on the results, the Agency proceeded to analyze all of the
samples for the plutonium isotopes.37 Also, all of the CKD samples from the six facilities sampled
by EPA in 1993 were analyzed for the man-made elements. Two of the man-made elements were
detected3* in CKD samples as follows. Three samples of as managed CKD had detectible levels of
plutonium-239. Cesium-137 was detected in CKD samples as follows: in as generated and as
managed CKD at four facilities; in as generated CKD two facilities; and in the TCLP extract of the
as generated CKD from one facility. This is consistent with prior findings that, due to past above-
ground weapons testing, very small amounts of Pu-239 and Cs-137 are routinely detected in soils and
comparable media.
For the man-made radionuclides, the radiological activities determined for the EPA samples
are considered to be within the range of activities normally found in environmental samples of
comparable composition.
In summary, the Agency considers that the radiological activities determined for the whole
CKD samples collected by EPA to be within the range of activities found in environmental samples of
comparable composition. That is, the activity levels observed for the radionuclides measured in CKD
are expected to be no different than, for example, those found in samples of soil and rock that are
randomly selected and sampled. Nevertheless, because the Agency's sampling and analytical program
did reveal detectable amounts of certain radionuclide species in CKD samples, it has decided to
" This included re-analysis of the samples from the original four facilities. The Agency considers the original
analytical results to be valid analytical data. The re-analyses were conducted for comparison purposes.
M The analytical detection limit for the EPA radionuclide analyses is considered to be the minimum detectable
activity (MDA) value. MDA is the smallest activity that must be present in a sample in order to yield a count rate
that will be detected with 97Vi % probability given detection criteria that give a 21/: % probability of falsely detecting
activity in a blank sample. (The confidence levels cited here are those used by EPA for its analysis of the EPA
samples.)
-------
3-54
include certain of them in the risk analysis on the basis of their presence at levels exceeding defined
risk criteria. Chapter 6 of this report identifies the radionuclide species that were included in the risk
analysis along with the basis for their inclusion.
3.3.3 Leachable Concentrations
• EPA has established four sets of tests, or characteristics, that are used to determine whether a
given waste stream should be managed as a hazardous waste. Of these four characteristics that define
RCRA-designated hazardous wastes (ignitability, corrosivity, reactivity, and toxicity), only toxicity is
relevant to CKD. This is because CKD is a solid, inorganic, non-flammable substance that is not
ignitable or reactive. Although CKD is highly alkaline, it is not considered corrosive under EPA's
definitions, because the characteristic does not apply to solid materials. Examining the characteristic
of toxicity in CKD is important in that the test is designed specifically to evaluate the potential for
toxic trace metals to leach and migrate from solid wastes.
Metals
To assess the potential of CKD to exhibit the toxicity characteristic, EPA performed TCLP
and SPLP leachate analyses of subsamples of the CKD samples. EPA compared the maximum and
average concentrations (i.e., mass of a particular constituent per unit volume of extract) in TCLP and
SPLP leachates from as generated and as managed CKD, as collected from all available sources, with
TCLP standards for arsenic, barium, cadmium, chromium, lead, mercury, selenium, and silver. The
results of these comparisons are presented in Exhibits 3-25 and 3-26. These comparisons show that,
in general, concentrations of trace metals are well below their corresponding TCLP standards. In
fact, for some constituents, the maximum observed leachate concentration is more than an order of
magnitude below the corresponding regulatory standard. Among these data, however, four samples
(two TCLP, two SPLP extracts) of a total group of 244 samples of as generated dust analyzed for
lead yielded concentrations greater than the TCLP standard of 5 parts per million (ppm) (see Exhibit
3-25); the maximum concentration for these four samples was 16.5 ppm. Also, two samples of a
group of 129 samples of as generated dust analyzed for selenium yielded concentrations greater than
the TCLP standard of 1.0 ppm; the maximum concentration of these two samples was 1.711 ppm.
As shown in Exhibit 3-26, one of 88 samples of as managed dust analyzed for barium yielded a
concentration higher than the TCLP standard (102.000 ppm versus 100.0 ppm), and one of 88
samples of as managed dust analyzed for cadmium yielded a concentration above the TCLP standard
(2.55 ppm versus 1.0 ppm).
The reader should note that the outcome of the above analysis may change if EPA revises
several existing TC levels and promulgates TC levels for new chemicals based on updated national
primary drinking water standards (NPDWS). Specifically, EPA recently revised the NPDWS for
barium, cadmium, chromium, lead, and selenium to 2 mg/L, 0.005 mg/L, 0.1 mg/L, 0.015 mg/L,
and 0.05 mg/L, respectively. EPA also established a NPDWS for nickel at 0.1 mg/L.
Because most of the target metals analytes were detected in all of the EPA CKD samples, the
Agency believes it appropriate to carry consideration of these elements through the decision rationale
and risk assessment process of this report, which means that the presence of these elements in CKD
will influence the Agency's decision-making on the RCRA regulatory status of CKD.
-------
3-55
Exhibit 3-25
Comparison of Maximum and Average Metals Concentrations in As Generated Dust with TC Standards
As Generated CKD-TCLP Test Results (parts per million)
Analyte
Arsenic
Barium
Cadmium
Chromium
Uad
Mercury
Selenium
Silver
TC
Standard
5.0
100
1.0
5.0
5.0
0.2
1.0
5.0
Number of
Samples
227
227
227
227
227
227
112
227
Maximum
Concentration
0.636
9.190
0.508
1.290
9.718
0.022
1.711
0.166
Ratio of Maximum
Concentration to
Standard
0.13
0.09
0.51
0.26
1.94
0.11
1.71
0.03
Maximum
Concentration
Minus Standard
-4.364
-90.810
-0.492
-3.710
4.718
-0.178
0.711
-4.834
Average
Concentration
0.02035
0.59762
0.01480
0.04571
0.21396
0.00083
0.07302
0.04147
Ratio of
Average
Concentration
to Standard
0.004
0.006
0.015
0.009
0.043
0.004
0.073
0.008
Average
Concentration
Minus
Standard
-4.979
-99.402
-0.985
-4.954
-4.786
-0.199
-0.926
-4.958
Number of
values >
Standard
0
0
0
0
2
0
2
0
As Generated CKOSPLP Test Results (parts per million)
Arsenic
Barium
Cadmium
Chromium
food
Mercury
Selenium
SUver
5.0
100
1.0
5.0
5.0
0.2
1.0
5.0
17
17
17
17
17
17
17
17
0.014
1.860
0.004
0.128
16.500
0.0001
0.276
0.030
0.003
0.019
0.004
0.026
3.3
0.0005
0.276
0.006
-4.987
•98.140
-0.996
-4.872
11.500
-0.1999
-0.724
-4.970
0.00606
0.50462
0.00382
0.0243
2.13729
0.0001
0.04578
0.00697
0.001
0.005
0.004
0.005
0.427
0.0005
0.046
0.001
-4.994
-99.495
-0.996
-4.976
-2.863
-0.1999
-0.954
•4.993
0
0
0
0
2
0
0
0
-------
3-56
Exhibit 3-26
Comparison of Maximum and Average Metals Concentrations in As Managed Dust with TC Standards
As Managed CKD-TCLP Test Results (parts per million)
Analyte
Arsenic
Barium
Cadmium
Chromium
Lead
Mercury
Selenium
Silver
TC
Standard
5.0
100
1.0
5.0
5.0
0.2
1.0
5.0
Number of
Samples
74
74
74
74
74
70
66
73
Maximum
Concentration
0.867
102.000
2.550
1.290
4.570
0.100
0.303
0.500
Ratio of Maximum
Concentration to
Standard
0.173
1.020
2.550
0.258
0.914
0.500
0.303
0.100
Maximum
Concentration
Minus
Standard
•4.133
2.000
1.550
-3.710
-0.430
-0.100
-0.697
-4.500
Average
Concentration
0.05958
2.15876
0.08654
0.13824
0.33766
0.00385
0.05055
0.04772
Ratio of Average
Concentration to
Standard
0.0119
0.0216
0.0865
0.0277
0.0675
0.0192
0.0506
0.0095
Average
Concentration
Minus
Standard
-4.940
-97.841
•0.913
-4.862
-4.662
•0.196
•0.949
-4.952
Number of
values >
standard
0
I
I
0
0
0
0
0
As Managed CKD-SPLP Test Results (parts per million)
Arsenic
Barium
Cadmium
Chromium
Lead
Mercury
Selenium
Silver
5.0
100
1.0
5.0
5.0
0.2
1.0
5.0
14
14
14
14
14
14
14
14
0.013
0.869
0.004
0.373
1.790
0
0.086
0.026
0.003
0.009
0.004
0.075
0.358
0.002
0.086
0.005
-4.987
-99.131
-0.996
-4.627
-3.210
-0.200
-0.914
-4.974
0.00416
0.39564
0.00336
0.09348
0.50310
0.00012
0.02348
0.00706
0.0008
0.0040
0.0034
0.0187
0.1006
0.0006
0.0235
0.0014
-4.996
-99.604
-0.997
-4.907
•4.497
•O.200
-0.977
-4.993
0
0
0
0
0
0
0
0
-------
3-57
Dioxins and Furans
As expected due to the extremely insoluble nature of dioxins and dibenzoftirans, Exhibit 3-27
shows that no homologs, except OCDD, were detected in TCLP extracts using detection limits
ranging from 0.000003 /zg/L to 0.000037 /xg/L. Total OCDD was detected in two TCLP extracts of
as generated CKD samples obtained from non-hazardous waste fuel burning facilities; the measured
levels are 0.00017 /*g/L and 0.00011 /zg/L.
Based on these results, the Agency does not believe that leachable dioxins and furans should
be considered further for purposes of this Report to Congress, and accordingly, has not included them
in the analysis that follows later in this report.
General Chemistry
EPA did not conduct TCLP analyses of any as generated or as managed CKD for the general
chemistry parameters. In addition, as discussed .earlier in this chapter, EPA did not expect total
cyanide to be present in CKD; therefore, EPA also did not conduct TCLP analyses for cyanide (i.e.,
as specified in 40 CFR §261, Appendix II, "If a total analysis of the waste demonstrates that
individual analytes are not present in the waste, or that they are present but at such low concentrations
that the appropriate regulatory levels could not possibly be exceeded, the TCLP need not be run").
Based on the preceding discussion, the Agency believes that cyanide should not be considered
further for purposes of this Report to Congress, and accordingly, has not included it in the analysis
that follows later in this report.
Volatile Organics
The Agency did not conduct TCLP analyses of the CKD for any of the volatile organic
compounds due to EPA's expectations that low or non-detectable total concentrations of the volatile
organic constituents would be found in CKD materials (i.e., as specified in 40 CFR §261, Appendix
II, "If a total analysis of the waste demonstrates that individual analytes are not present in the waste,
or that they are present but at such low concentrations that the appropriate regulatory levels could not
possibly be exceeded, the TCLP need not be run"). The Agency believes that the futility of
performing such analyses was demonstrated by the fact that no volatile organic constituents were
confirmed present in CKD.
Based on the preceding discussion, the Agency believes that the volatile organic compounds
should not be considered further for purposes of this Report to Congress, and accordingly, has not
included them in the analysis that follows later in this report.
Semi-Volatile Organics
The Agency did not subject CKD leachates to analysis for semi-volatile organic compounds
because EPA did not expect that any semi-volatile organic constituents would be found in CKD
materials (i.e., as specified in 40 CFR §261, Appendix II, "If a total analysis of the waste
demonstrates that individual analytes are not present in the waste, or that they are present but at such
low concentrations that the appropriate regulatory levels could not possibly be exceeded, the TCLP
need not be run"). The Agency believes that the futility of performing such analyses was
demonstrated by the fact that no semi-volatile organic constituents were confirmed present in CKD.
-------
3-58
Exhibit 3-27
TCLP Concentrations of Dioxins and Dibenzofurans in As Generated CKD (pg/L)
ANALYTE
1,2,3,4,6,7,8-HpCDD
Total HpCDD
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8.9-HpCDF
Total HpCDF
1,2,3,4,7,8-HxCDF
1,2,3,6,7,8-HxCDF
1,2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
Toul HxCDF
OCDD
OCDF
2,3,7,8-TCDD
Total TCDD
HW-1
Sample 1
< 0.000019
N.A.
< 0.000017
< 0.000015
N.A.
< 0.00001
< 0.000008
< 0.00002
< 0.000021
N.A.
< 0.000035
< 0.000026
< 0.000022
N.A.
HW-1
Sample 2
< 0.000023
N.A.
< 0.000012
< 0.000018
N.A.
< 0.00001
< 0.00001
< 0.000024
< 0.000007
N.A.
B 0.000037
< 0.000033
< 0.000013
N.A.
HW-2
< 0.000009
N.A.
< 0.000003
< 0.000005
J 0.000007
< 0.000004
< 0.000003
< 0.000006
J 0.000006
J 0.000006
B 0.000027
< 0.000009
< 0.000005
N.A.
HW-3
J 0.000016
J 0.000032
< 0.000008
< 0.00001
N.A.
< 0.000005
< 0.000003
< 0.000004
J 0.000004
J 0.000004
J 0.000077
I 0.000011
< 0.000008
J 0.000005
NH-1
J 0.000021
J 0.000021
< 0.000017
< 0.000019
N.A.
< 0.000006
< 0.000006
< 0.000012
< 0.000016
N.A.
0.00017
< 0.000022
< 0.000009
N.A.
NH-2
< 0.000028
J 0.000019
< 0.000014
< 0.000019
N.A.
< 0.00001
< 0.000009
< 0.00001
< 0.000015
N.A.
0.00011
< 0.000055
< 0.000017
N.A.
NH-3
< 0.00003
N.A.
< 0.000009
< 0.000014
N.A.
< 0.000007
< 0.000006
< 0.000012
< 0.000008
N.A.
J 0.000057
< 0.000041
< 0.000014
N.A.
MAXIMUM
0.00003
0.000032
<0.000017
<0.000019
0.000007
< 0.00001
<0.00001
< 0.000024
0.000021
0.000006
0.00017
0.000055
<0.000022
0.000005
AVERAGE
0.000021
0.000018
< 0.0000 11
<0.000014
0.000007
< 0.000007
< 0.000006
<0.000013
0.000011
0.000005
0.000073
0.000028
<0.000013
0.000005
AVERAGE
DETECTED
0.000019
0.000024
0.000007
0.000005
0.000005
0.000080
0.000011
0.000005
N.A.
•B"
Average
Detected =
Not Detected, the Associated Value is the Detection Limit
Detection limits are not available for total concentrations.
The Constituent was Detected in an Associated Blank.
The Concentration is an Estimate. The Constituent Was Positively Identified at a Trace Value
or is a Nontarget Constituent for which no Calibration was Performed.
The average of the samples, excluding those that were not detected.
HW-1 - Keystone Cement Co., Bath, PA
HW-2 - River Cement Co., Festui, MO
HW-3 - Heartland Cement Co., Independence, KS
NH-1 - Ash Grove West, Inc., Inkom, ID
NH-2 - Calaveras Cement Co., Tehachapi, CA
NH-3 - Holnam, Inc., Artesia, MS
-------
3-59
The Agency believes that the semi-volatile organic compounds should not be considered
farther for purposes of this Report to Congress because they do not appear to be present in CKD, and
accordingly, has not included them in the analysis that follows later in this report.
Pesticides
EPA conducted TCLP leachate analyses for pesticides using as generated CKD from six
facilities. There were no pesticide compounds detected in the TCLP leachates.
The Agency believes that the pesticides do not warrant further consideration for this Report to
Congress because they do not appear to be present in CKD on an industry-wide basis. Accordingly,
the pesticide compounds are not included in the analysis that follows later in this report.
PCBs
The Agency conducted TCLP leachate analyses for PCBs on samples of the as generated CKD
collected from six facilities. EPA did not detect any PCB compounds in the TCLP leachates.
Because there were no PCB compounds detected in any of the CKD samples obtained by
EPA, the Agency believes that they do not warrant further consideration for the purpose of this
Report to Congress. Accordingly, the PCB compounds are not included in the analysis that follows
later in this report.
Radionuclides
The Agency conducted gross alpha, gross beta, and gamma analyses as well as isotopic
analysis involving chemical separations for the following specific analytes on TCLP leachates of
CKD: radium-226, plutonium-238, plutonium-239, uranium-234, uranium-235, uranium-238,
thorium-227, thorium-228, thorium-230 and thorium-232. Gamma analyses and plutonium isotopic
analyses were performed on TCLP leachates of as generated CKD from six of the facilities. Gross
alpha and beta analyses and isotopic analyses involving chemical separations for the other specific
analytes listed above were determined for all of the CKD samples from six of the facilities.
Several of the TCLP leachates of the as-managed CKD samples had elevated levels of
potassium-40 compared to the other aqueous samples. Potassium-40 is a naturally occurring
radionuclide of relatively high natural abundance. It is reasonable to expect that the leaching process
would concentrate the potassium-40, thereby producing the elevated activity levels. Since potassium-
40 is a beta emitter, this would also explain the somewhat elevated gross beta activities of the TCLP
leachate samples. In addition, Cesium-137 was detected in the TCLP extract of one sample of the as
generated CKD sample.
For the man-made radionuclides, the radiological activities determined for the EPA samples
are considered to be within the range of activities normally found in environmental samples of
comparable composition.
As discussed earlier in this chapter, the Agency considers that the radiological activities
determined for the whole CKD samples collected by EPA to be within the range of activities found in
environmental samples of comparable composition. That is, the activity levels observed for the
radionuclides measured in CKD are expected to be no different than, for example, those found in
samples of soil and rock that are randomly selected and sampled. Nonetheless, because EPA detected
-------
3-60
several radionuclide species in CKD samples, it has decided to include certain of them in the risk
analysis on the basis of their presence at levels exceeding defined risk criteria. Chapter 6 of this
report identifies the radionuclide species that were included in the risk analysis along with the basis
for their inclusion.
3.4 STATISTICAL ANALYSES OF CKD CHARACTERIZATION RESULTS
• As discussed in the previous sections of this chapter, the Agency detected potentially
significant concentrations of certain trace metals and dioxins and furans in both the as generated and
as managed forms of CKD. In an attempt to better understand the significance of these findings, EPA
conducted a number of additional analyses.
The Agency notes at the outset that a number of caveats apply to this analysis and the results
obtained thereby. These caveats limit the extent to which the results can reasonably be used to draw
inferences or conclusions concerning the influence of cement kiln design and operating factors on
CKD constituent concentrations. First, most of the CKD composition data in EPA's possession were
obtained from one-time "grab" sampling at operating cement plants. The Agency believes that this
was the general method employed to generate the PCA Survey and PCA Report data, as well as
EPA's 1992 and 1993 field sampling data. Consequently, the data ascribed to a particular facility
were collected at a particular point in time, and may or may not accurately reflect the typical
composition of CKD or clinker over an extended period of time. In this regard, EPA notes that
examination of some of the time series (process control) data submitted by some facility operators in
response to the Agency's RCRA §3007 request indicate significant variation, on a day-to-day basis, in
the concentrations of major CKD constituents; it seems reasonable that trace constituents may also
vary in this manner. Second, the Agency's understanding of the CKD data provided by respondents
to the PCA Survey is quite limited; information on collection methods and conditions is largely
absent. Thus, for example, CKD constituent concentrations that are attributed to the burning of
hazardous waste may actually have been collected when the cement kiln in question was not burning
hazardous waste at all or under normal operating conditions.
3.4.1 Metals
As indicated above, for potentially important trace metal constituents, data were available
from the following sources and were used for these analyses:
• 1992 EPA Sampling Data
• Individual 1991 Portland Cement Association (PCA) Surveys
• PCA CKD Metals Analysis Report (Draft and Final - January, 1992)
• Bureau of Mines (BOM) Information Circular 8885 (1982)
The Agency conducted the additional analyses of the metals data in three primary steps:
• Step 1 - Examine the concentration data from each source to determine if the
measurements are random samples from a normal distribution. If appropriate,
calculate a random concentration, the value of which lies between zero and the
detection limit for the analytical measurement method, for constituent concentrations
that were reported by the laboratory as "undetected."
-------
3-61
• Step 2 - Compare the metal constituent concentrations from each of the data sources
to determine if there are significant differences between the mean concentrations.39
If significant differences are not found, then the data from these sources may be
assumed to be drawn from the same population and can be combined for subsequent
analysis.
• Step 3 - Examine the data for correlations or trends that may suggest how metals
concentrations may be influenced by the design and operating characteristics of
individual cement kilns.
A detailed description of these analyses and the results obtained thereby are presented in a Technical
Background Document for this Report to Congress which is available in the RCRA docket.
For Step 1, EPA observed substantial improvement in the test for normality (the Shapiro-Wilk
statistic) in most cases by including the calculated concentrations for the "undetected" constituents.
Moreover, substituting these randomly calculated non-zero concentrations for the "undetected"
constituent concentrations results in concentrations that are in all cases normally distributed; that is,
the hypothesis that the data are normally distributed could not be rejected at the 95 percent confidence
level for any material type, analysis type, or constituent. As mentioned above, because standard
statistical analyses presuppose normally distributed data, all of the Agency's subsequent analyses are
based upon the inclusion of these calculated data.
In Step 2, EPA compared the calculated means from each of the other data sources to the
EPA sampling data means using the student t-test. The EPA data served as the basis of comparison
because the Agency has the highest level of confidence in its own data set. These comparisons of the
means resulted in the following observations:
• Most of the means from the various sources are not significantly different from the
EPA sampling data means (at the 95 and 99 percent confidence level).
• For those means that are significantly different from the EPA sampling data at the 95
percent confidence level, most are higher than the EPA sampling data means. By
combining these data, the effect would be to increase the calculated mean constituent
concentrations from the original EPA measurements. All subsequent analyses would
be more conservative as a result.
• For those means that are significantly lower than the EPA sampling data, all but three
have substantial overlap between the minimum and maximum concentrations reported
for each data source. This suggests that the difference may be an artifact of the
sampling technique (i.e., the sampling was not random) and that therefore one cannot
reject the hypothesis that the samples are from the same population. Further,
differences in means did not involve "critical" constituents (i.e., those flagged as
hazardous constituents) except for antimony (discussed below).
• Only three mean concentrations were found to be significantly lower than the
corresponding EPA sampling data value and to not have overlap in the range of
39 The Agency used mean values in the parametric statistical tests described in this section after establishing that
the data are normally distributed. Parametric statistics require use of the mean rather than some other measure of
central tendency, such as the median.
-------
3-62
observed concentrations. All three of these are mean concentrations for antimony and
were calculated from PCA Report 2 and Bureau of Mines sources for "as generated"
dust using "TCLP" and "Total" analytical methods. The reason for this anomaly is
not known. It is interesting to note, however, that the mean antimony values reported
in PCA Report 2 were two to three orders of magnitude lower than those in PCA
Report 1, for reasons that are not fully explained in PCA Report 2 (the final report).
Because of this significant, unexplained swing in reported values, the Agency is
inclined to discount these data. In addition, the mean antimony concentrations in the
EPA and Bureau of Mines data sets differ by only a factor of about two.
Based upon this analysis, EPA concluded that the majority of the calculated mean
concentrations for the EPA sampling data are not significantly different than the means from the other
sources. The few concentration means that are significantly different do not adversely affect the
overall analysis, for the reasons discussed above. Consequently, the Agency believed that it was
reasonable to combine, or pool, the data from all of the available sources prior to conducting
subsequent (Step 3) analyses.
Nonetheless, because in the next phase of the analysis the constituent data were examined in
light of plant-specific design and operating factors, the available data set does not include observations
from all active portland cement plants; it does, however, contain data from a substantial percentage of
them (47 of the 115 active facilities). That is, because the analysis presented here focuses on cement
kiln design and operating factors, only those composition data that can be attributed to specific plants
or design and operating factors can be used. Accordingly, the data from the two PCA reports and the
1982 Bureau of Mines cement kiln dust study have not been used in this analysis, because these
documents present no information on the design and operating factors of interest, nor do they identify
the specific facilities that gave rise to the data presented therein.
EPA's next step was to attempt to determine whether CKD trace metal constituent
concentrations might be affected by cement kiln design and operating factors. Given the disparities
noted earlier in net CKD generation rates between kiln types and especially across fuel types, EPA
focused its examination on these two variables. Accordingly, the Agency conducted t-test
comparisons of the mean concentrations of trace metals in CKD found within these respective groups.
Results of these analyses are presented below.
T-test to Examine Hazardous Waste Burning Effects. For this analysis, EPA pooled the
data from both available sources, i.e., those containing material composition data and an indication of
whether the corresponding facility does or does not burn hazardous wastes as fuel. The Agency
calculated the mean of the metal concentration data for each Sample Type and Analysis Type
subgroup. EPA then compared the mean concentration for each metal within each sub-group for
those facilities burning hazardous wastes with those that do not. EPA used the t-test to determine
whether the null hypothesis can be rejected for two means representing the same population at a given
confidence level.
The majority of the means were not significantly different at the 95 percent confidence level
for these two sub-groups; that is, there is no statistically significant difference in the mean
concentration of most metals in most material and sample types in CKD generated by kilns burning
hazardous waste versus those not burning hazardous waste. Exhibit 3-28 lists those means that may
be considered different at a 95 percent confidence level.
-------
3-63
Exhibit 3-28
T-test Comparison of Fuel Burning Effects on Metals Concentrations
fmcludes only EPA* and PCA Survey Data)
Constituent
Cadmium
Chromium
Lead
Thallium'
Arsenic
Barium
Cadmium
Nickel'
Selenium
Thallium'
Mercury1
Nickel
Thallium'
Aluminum'
Arsenic
Barium'
Cadmium'
Chromium'
Mercury*
Lead'
Selenium'
Silver"
Cadmium
Mercury
Selenium'
Chromium
Silver'
Cadmium'
Lead
Aluminum'
Sample Type
As Generated
As Generated
As Generated
As Generated
As Generated
As Generated
As Generated
As Generated
As Generated
As Generated
As Generated
As Generated
As Generated
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
Analysis Type
Total
Total
Total
Total
TCLP
TCLP
TCLP
TCLP
TCLP
TCLP
SPLP
SPLP
SPLP
Total
Total
Total
Total
Total
Total
Total
Total
TCLP
TCLP
TCLP
TCLP
SPLP
EP
EP
EP
RAI
Mean Concentration (ppm)
Kilns Burning HW
1.4 x 10'
3.9 x 10'
53 x 101
1.8 x 10°
7.2 x 10 J
9.8 x 10 •'
4.1 x 10°
1.4 x 10 "2
3.4 x 10 *
2.0 x 10 '2
1.0 x 10 •*
1.4 x 10 •'
2.3 x 10 '2
7.3 x 10s
1.8 x 10'
2.3 x 10 2
2.7 x 10'
4.5 x 10'
9.7 x 10 •'
1.1 x 10s
1.4 x 10'
3.0 x 10 '2
1.8 x 10 •'
6.0 x 10 •'
8.2 x 10 "2
1.8 x 10"'
3.3 x 10 •'
2.2 x 10 •'
2.9 x 10 "2
2.3 x 10 '
Kilns Not Burning HW
5.2 x 10°
1.6 X 10'
2.0 x 10 2
2.8 x 10'
1.2 x 10 "2
3.9 x 10 •'
1.7 x 10 "2
5.3 x 10 "2
9.3 x 10 '2
5.4 x 10 •'
9.8 x 10 "'
1.3 x 10 •'
2.0 x 10"'
1.5 x 10 4
9.9 x 10°
1.4 x 10 2
6.0 x 10°
2.1 x 10'
3.4 x 10 •'
1.2 x 10 2
7.5 x 10°
6.2 x 10 "2
3.3 x 10 "2
1.2 x 10 •'
3.4 x 10 "2
2.7 x 10 "2
5.0 x 10 "2
1.4 x 10 •'
1.4 x 10°
1.6 x 10 4
Ratio (HW + non-HW)
2.7
2.4
2.6
.066
0.58
2.5
0.25
0.26
0.37
.037
1.06
1.07
0.12
0.48
1.8
1.6
4.5
2.1
2.8
8.8
1.9
0.48
5J5
5.2
2.4
6.8
.065
.016
.020
1.4
* 1993 sampling and analysis data not included.
b Confidence level of 99 percent.
-------
3-64
Cadmium, chromium, and lead are found at mean concentrations in as generated CKD
that are from 2.4 to 2.7 times higher in dust from facilities burning hazardous wastes than in
CKD from facilities not burning hazardous waste fuels. On the other hand, thallium
concentrations are decidedly lower in the dust generated by the hazardous waste-burning kilns
(and at the 99 percent confidence level); this pattern holds not only for total concentrations but
also for results of both leaching procedures (TCLP and SPLP). A number of other constituents
are found at significantly lower concentrations in TCLP leachate in CKD from hazardous waste-
burning kilns relative to non-hazardous waste burning kilns, including arsenic, cadmium, nickel,
and selenium. However, barium is found at higher concentrations in the leachates from
hazardous waste-burning kilns.
For as managed CKD, a number of heavy metals are found at significantly higher
concentrations in dust from kilns burning hazardous wastes; these include arsenic, barium,
cadmium, chromium, mercury, lead, and selenium. Only aluminum is found at a lower mean
concentration. Leach test results are somewhat inconsistent, with some constituents (e.g.,
cadmium) exhibiting higher concentrations in dust from hazardous waste-burning kilns using one
leach test (TCLP), and the opposite result using a different though similar leach test (EP).
Overall, certain metals appear to be present at a consistently higher mean concentration
in CKD generated by kilns burning hazardous waste than in CKD generated by kilns not using
this type of alternative fuel. Lead, cadmium, and chromium are the most prominent examples.
T-lest to Examine Influence of Kiln Type (Dry vs. Wet). For this analysis, EPA pooled
the data from both available sources, i.e., those containing material composition data and an
indication of whether the corresponding facility has dry or wet kilns (only four plants nationwide
have both). The Agency calculated the mean of the metal concentration data for each Sample
Type and Analysis Type subgroup. EPA then compared the mean concentration for each metal
within each sub-group for those facilities with wet kilns to those that have dry kilns. This was
performed with the t-test to determine whether the null hypothesis can be rejected, i.e., the two
means represent the same population at a given confidence level.
The majority of the means were not significantly different at the 95 percent confidence
level for these two sub-groups; those means that are significantly different at this confidence level
are presented in Exhibit 3-29.
No statistically significant differences between the wet and dry process are apparent in
the total metals concentrations of as generated CKD. For four metals, however, TCLP test
results are higher for dust generated by the dry process kilns; mean concentrations of aluminum,
cadmium, nickel, and selenium in TCLP leachate from dry kiln as generated dust ranged from
2.4 to 4.4 times those from wet kiln as generated dust. Antimony levels as determined by the
SPLP test appear to be almost twice as high for dust from the wet process as from the dry
process.
Results from the total metals concentrations tests (acid digestion and RAI40) on as
managed CKD are striking. Significantly higher concentrations of aluminum, arsenic, barium,
cadmium, chromium, copper, mercury, lead, titanium, and zinc are found in the CKD from the
40 The Agency used x-ray diffraction data for metal oxides reported in the PCA surveys to estimate total
constituent concentrations of specific metals. These estimated total constituent concentrations have been designated
"RAI" in the Report.
-------
3-65
wet process kilns; mean concentration differences range from a factor of about two to almost ten,
and most of the differences in mean concentration are significant at the 99 percent confidence
level. TCLP results for aluminum and chromium are similar. Additional test results from use of
the EP method show higher
Exhibit 3-29
T-test Comparison of Kiln Type on Metals Concentrations
(includes only EPA* and PCA Survey Data)
Constituent
Aluminum
Cadmium
Nickel*
Selenium*
Antimony
Arsenic
Barium*
Cadmium*
Chromium*
Mercury*
Lead*
Aluminum
Chromium
Arsenic*
Selenium*
Aluminum*
Copper
Titanium*
Zinc*
Sam pie Type
As Generated
As Generated
As Generated
As Generated
As Generated
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
As Managed
Analysis Type
TCLP
TCLP
TCLP
TCLP
SPLP
TotaJ
Total
Total
Total
Total
Total
TCLP
TCLP
EP
EP
RAI
RAI
RAI
RAI
Mean Concentration (ppm)
Dry Kilns
1.1 X 10 •'
1.8 X 10"
5.6 X 10 •"
1.1 x 10 '
2.7 x 10 •»
1.1 X 10'
8.9 x 10'
1.3 x 10'
2.4 x 10'
2.8 X 10 •'
1.5 X 10'
6.1 X 10"
9.6 x 10 •*
5.2 x 10 •»
3.6 x 10 -1
1.7 x 10*
1.2 x 10J
1.0 X 10'
3.6 X 101
Wet Kilns
4.3 x 10 2
4.0 x 10 •'
1.4 x 10 J
3.2 x 10 •»
4.6 X 10 J
1.9 x 10'
2.9 x 10 J
2.8 x 10'
4.9 x 10'
1.2 x 10°
1.3 x 101
2.1 x 10 •'
2.0 x 10 '
1.4 x 10 -1
3.7 X 10 '
2.0 x 10 4
3.0 x 105
2.2 x 10'
2.2 x 10'
Ratio (Dry + Wet)
2.4
4.4
4.0
3.5
0.58
0.57
0.30
0.46
0.50
0.24
0.11
0.29
0.48
3.8
0.10
0.82
0.40
0.45
0.17
• 1993 sampling and analysis data not included.
* Confidence level of 99 percent.
concentrations of selenium in as managed CKD from the wet process, though for most
constituents, any differences in mean concentration are not statistically significant.
EPA can discern few overall trends from these results. Lead concentrations seem to be
lower in CKD when using the dry process rather than the wet process, and total metals
concentrations seem to be generally higher in as managed dust from the wet process. Otherwise,
there do not appear to be consistent trends in metals content of these materials with respect to
kiln technology type.
-------
3-66
To better understand the determinants of trace metal concentrations in CKD, EPA
wanted to determine whether there might be design and/or operating factors that influence
contaminant concentrations in a direct manner. Therefore, as a next step, EPA examined the
correlation between chemical concentration and four individual variables (kiln age, kiln length,
recycling rate, and percentage of fuel consumption supplied by coal), to examine the validity of
the hypothesis that each in isolation is directly and linearly related to the concentration of
individual metallic constituents.
Linear correlation analysis is a quantitative technique that is used to estimate the degree
to which two variables are related. Strong positive linear correlation means that as the value of
one variable increases, the value of the other increases in direct proportion. Conversely, negative
linear correlation means that as the value of one variable increases, the other decreases in direct
proportion. Correlation analysis can therefore be used to identify variables that may be useful in
explaining or even predicting the value of a variable or phenomenon of interest, and is
particularly useful as a preliminary step leading to application of more sophisticated quantitative
techniques, such as multiple regression analysis.
The Pearson Correlation Coefficient measures the strength of the linear relationship
between two variables; the coefficient value ranges from -1 to +1. When two variables are highly
and positively correlated, the coefficient approaches +1. Alternatively, if the variables are
negatively correlated (value between -1 and 0), the variables are related inversely. A correlation
of zero means that each variable has no linear predictive value with respect to the other. For
purposes of its analysis, EPA assumed that Pearson Correlation Coefficient values greater than
+.75 or less than -.75 indicate that a given variable has a possibly significant effect on constituent
concentration. EPA has made this assumption because in all likelihood, constituent
concentrations in CKD and other materials are a function of several variables, rather than a
simple linear function of just one. The Agency's purpose was to identify potentially significant
variables for possible further analysis, and to ensure that it did not overlook any clear or
dominant explanatory variables. EPA's threshold value of +/- .75 is an arbitrary level to identify
those variables that appear to signify or be used to "predict" trace constituent levels.
EPA tested four factors for correlation with constituent concentration values:
• Kiln age;
• Kiln length;
• CKD recycling rate; and
• Percent of energy consumption supplied by coal.
EPA chose these factors because it appears that they may have some impact on the distribution
of metallic and non-metallic constituents in the kiln system and CKD.
The Agency examined kiln age because the predominant kiln type being used has
changed over time; in recent years many older wet process kilns have been replaced with more
energy efficient dry process kilns. Dry and wet process kilns have different material residence
times, temperature profiles, and other operating characteristics that may influence constituent
concentrations. Moreover, the physical age and condition of a unit may affect its operating
performance and, thereby, the characteristics of both product (clinker and cement) and waste
(CKD). The kiln length influences the amount of time that dust and raw materials remain in the
kiln, possibly altering their chemical composition. Because metallic (and other inorganic)
constituents are not destroyed in the kiln system, their build up in and removal from the kiln
-------
3-67
system can be key factors in influencing the composition of CKD. CKD recycling rates may be a
good indicator of the importance of these phenomena. Finally, EPA analyzed the possibility that
the percentage of the energy consumed in making cement clinker that is supplied by coal may
correlate with the concentrations of certain constituents in the CKD and other materials. A
significant negative correlation may suggest that the use of alternative fuels (e.g., hazardous
wastes) exerts an important influence on, for example, the composition of CKD (e.g., with
respect to heavy metal concentrations).
In the majority of the cases in which EPA observed extreme (very high or low) Pearson
correlation coefficient values there are a small number of observations (eight or less). Significant
coefficient values (as defined within this context) are few for the analyses that involved larger
data sets. Thus, it is difficult to draw any broad conclusions regarding the possible importance of
the operating factors that EPA has examined here with respect to constituent concentrations.
Nonetheless, there are some interesting findings from this analysis, which are discussed below.
Correlation Analysis of Kiln Age. No significant correlations (i.e., > .75 or < -.75) are
apparent in the as generated CKD data for kiln age. For as managed dust, EP test results
suggest both positive (antimony, molybdenum, selenium, and silver) and negative (arsenic,
mercury, and zinc) correlations with kiln age. There are no instances of extreme coefficients for
any constituent for more than one analysis type using the combined data. For cement, chromium
and nickel concentrations are highly and positively correlated with kiln age, as reflected in both
total and TCLP test results, while selenium is negatively correlated using these two analysis types.
The impact of this finding is limited by the small number of data points (four to seven). TCLP
leachate for mercury is positively correlated while total mercury concentration is somewhat
negatively correlated.
EPA found no constituents with extreme correlation coefficients using the EPA sampling
data. The results using only the PCA data parallel those using the combined data set for cement
and as managed CKD because the PCA data set comprises almost all of the data points for
cement and as managed CKD.
Correlation Analysis of Kiln Length. In the combined data set, there are no extreme
correlation coefficient values from analysis of the as generated CKD data. For the EPA data for
as managed dust, the analysis suggests both positive (antimony, molybdenum, and selenium) and
negative (arsenic, mercury, and zinc) correlations with kiln length. It is noteworthy that this
pattern with the same metals also appeared in the kiln age analysis presented above. Again,
sample sizes for this analysis were quite limited (two to eight). A different analysis type ("RAT)
indicates a strong positive correlation between kiln length and zinc concentration in as generated
CKD. No instances of extreme coefficients resulted for any constituent for more than one
analysis type for kiln length.
There are no constituents with extreme correlation coefficients using the EPA sampling
data, and no constituents have extreme coefficients for more than one analysis type using PCA
Survey Data. The EP and RAI results are, of course, identical to those from the combined data
set, because these analysis types are not represented in the EPA data. In the very limited data
(three observations from two facilities) for as generated CKD provided in the PCA Survey
responses, total concentrations of arsenic, beryllium, cadmium, nickel, vanadium, and zinc have a
strong negative correlation with kiln length, while total chromium and lead positively correlate
with kiln length.
-------
3-68
Correlation Analysis of Recycling Rate. For as generated CKD, only total zinc
concentrations display a significant (negative) correlation with recycling rate, and this finding is
based on only three observations. For as managed CKD, only a few constituents and analysis
types show a strong linear relationship to recycling rates, and in most of these cases, the
correlation results are based on only a few data points. Total concentrations of manganese and
zinc, and EP concentrations of molybdenum and zinc are negatively correlated, while EP
concentrations of antimony and TCLP concentrations of zinc are positively correlated. With the
exception of the TCLP zinc concentrations, all of these correlations are influenced by very small
sample sizes (two to four observations).
There are no instances of extreme coefficients for a constituent for more than one
analysis type for the recycling rate variable.
Within the EPA data set, only silver and thallium concentrations in as managed dust
(positive) show any significant correlation with recycling rate. In the PCA Survey data, results
for the cement material type are identical to those in the combined data set. For as generated
CKD, there are only three observations. These indicate significant negative correlations for total
arsenic, beryllium, cadmium, nickel, vanadium, and zinc as well as significant positive correlations
for chromium and lead. In as managed CKD, negative correlations are apparent for total
manganese and zinc, and for EP concentrations of molybdenum and zinc; positive correlations
are observed in EP results for antimony and TCLP results for vanadium and zinc. All but the
last of these cases are drawn from very small data sets (two to four data points per analysis).
Correlation Analysis of Percent of Energy Consumption Supplied by Coal. For the as
generated CKD data, only total zinc concentrations are strongly (negatively) correlated to coal
use; this finding is based upon only three data points. For the as managed CKD data, total
concentrations of manganese and zinc, EP concentrations of antimony, and zinc concentrations
measured using the "RAT method are negatively correlated with coal use. In contrast, EP
concentrations of cadmium, molybdenum, silver, and zinc are positively correlated with the
percentage of energy value derived from coal. Numbers of observations available in these cases
range from two to twelve, and there are no instances of extreme coefficients for a constituent for
more than one analysis type.
For the EPA data set, no extreme correlation coefficients appear in any material or
analysis type, with the exception of total concentrations of lead, which are negatively correlated
with the extent of coal use in both as generated and as managed CKD. In as generated CKD,
total arsenic, beryllium, cadmium, nickel, vanadium, and zinc are negatively correlated, and total
chromium and lead are positively correlated, with the extent of coal use. In as managed CKD,
results are identical to those obtained using the combined data set, with the exception that TCLP
concentrations of vanadium are positively correlated with the percentage of energy value derived
from coal, in addition to the EP results described above.
For the PCA data, as before, cement analysis results are the same as in the combined
data set.
3.4.2 Dioxins and Furans
In parallel with its examination of trace metal constituents, EPA evaluated the dioxins
and dibenzofurans data for significant relationships and trends. First, the Agency attempted to
determine whether or not the data obtained from Phase I and Phase II sampling are comparable
-------
3-69
and can be pooled for further analysis. This analysis was necessary because the laboratory
methods used during Phase II analysis were selected specifically because they offered far greater
sensitivity than those employed for Phase I analysis. T-test comparisons of individual analytes by
sample type yielded the following results. No statistically significant differences at the 99 percent
confidence level were observed in as generated or as managed CKD, though there were a few
instances in which differences were apparent at the 95 percent level (1,2,3,4,7,8,9-heptachloro-
dibenzofuran in both as generated and as managed dust, and total tetrachlorodibenzodioxin in as
managed dust). The means for all analytes in CKD are, as expected, considerably higher in the
Phase I data than in the Phase II data, due to the more sensitive laboratory methods employed
to generate the Phase II data. Phase II mean values are generally about ten to 30 percent of
Phase I means in as generated CKD, while in as managed CKD, these ratios range between 15
and 50 percent.
Based upon these results, EPA concluded that, despite the absence of statistically
significant differences between Phase I and Phase II results for CKD, it was probably not
reasonable to combine the data for further analysis. To do so would imply that the data are all
derived from the same population, an assumption that is known to be false, due to the very
different ways in which the two data sets were generated.
Consequently, as a next step, the Agency ran some t-test comparisons of the dioxin and
dibenzofuran constituents in the Phase II data set, i.e., the data with more detected values, to
determine whether any statistically significant differences exist in CKD generated by cement kilns
falling into various groups. Because of the limited number of available data points, EPA was
able to conduct only two simple pair-wise comparisons:
(1) Wet process kilns and dry process kilns; and
(2) RCRA hazardous waste-burning kilns and non-hazardous waste burning kilns.
The results of these t-test comparisons yield, with only two exceptions, no statistically
significant differences between the respective groups for any group, material type, analysis type,
or analyte.
Comparisons of wet and dry process kilns produce an interesting result. As generated
and as managed CKD organic constituent concentrations are generally more than one order of
magnitude higher in CKD generated by dry process kilns than in CKD generated by wet process
kilns.
Mean concentrations in as managed dust generated by kilns burning hazardous waste
fuels (as indicated by both total and TCLP analyses) appear to be higher, often by more than
one order of magnitude, than in CKD generated by kilns not burning hazardous wastes. The
same pattern is observed in the totals analyses of as generated CKD, though in TCLP results,
concentrations are marginally higher in the samples collected from kilns not burning hazardous
waste fuels. In this latter case, however, nearly all observations are estimated values, because the
corresponding measured TCLP concentrations were below detection limits. Nonetheless, these
observations are constrained by the relatively small sample population (six facilities - three
RCRA hazardous waste burners and three non-hazardous waste burners) and thus the difficulty
in establishing statistically significant differences between these groups. Additionally, the Agency
recognizes that other types of fuel (e.g., coal) that may be either burned exclusively or else co-
-------
3-70
fired with RCRA hazardous wastes in the kilns could be significant contributors of organics and
other constituents that are measured in the CKD.
3.5 CLINKER CHARACTERISTICS
During most of the facility sampling visits conducted by EPA for this study, samples of
newly generated clinker were obtained and subjected to chemical analyses similar to those for
cement kiln dust. There are two main reasons why EPA collected clinker samples during this
study. First, the Agency wanted to be able to compare the actual and relative amounts of certain
analytes in clinker with those in CKD as part of the RTC development effort. Second, the
analytical data derived from analysis of the clinker samples will be used by EPA in another study.
This study is entitled Use of Hazardous Waste in Cement Production. EPA's goal in this study is
to examine how federal regulations and policies can and do affect the use of hazardous waste in
cement production, and, should it be necessary, to determine the level of control necessary to
protect human health and the environment.41
The samples of clinker that were obtained during this study are from a total of 18
facilities.42 Eleven of the 18 were using hazardous waste for some portion of their fuel during
the EPA sampling. The other seven facilities were not using hazardous waste for fuel during the
EPA sampling.
All clinker samples were analyzed for metals, radionuclides, and major ions.43 Nine of
those samples were also analyzed for dioxins, furans, semi-volatile organics, pesticides, and
PCB's. Four of these nine samples were also analyzed for volatile organic compounds. Several
sets of leachates were also prepared from the whole samples of clinker obtained during the EPA
sampling visits. These TCLP and SPLP leachates were analyzed as follows: 18 sets of leachates
for metals, and four sets of leachates for dioxins, furans, pesticides, PCB's, radionuclides, and
major ions. Volatile organics and semi-volatile organics were not analyzed in clinker leachates
because it is believed that these compounds, if present in the original whole samples, would have
been driven from the samples during the leachate preparation process and thus could not be
quantitated. Exhibit 3-30 presents the results of the clinker characterization for inorganics by
fuel type (i.e., non-hazardous and hazardous waste fuels).
Dioxins and furans were not detected in either the whole clinker samples or in the TCLP
and SPLP leachates. Thus, it was not feasible to compare dioxin and furan levels in clinker with
general fuel type or other operating factors. Accordingly, dioxin and furan levels in clinker do
not appear to correlate with levels found in CKD produced at the same facility, i.e., although
dioxins were detected in CKD at several facilities, they were not detected in clinker produced at
the same facilities.
No pesticides or PCB's were detected in either whole clinker or TCLP and SPLP
leachates prepared from the whole clinker samples.
41 For more information, see the report titled RCRA Implementation Study (RJS) Update: The Definition of Solid
Waste (EPA 530-R-92-021. July, 1992).
42 One of the 18 facilities was re-visited and sampled a second time for analysis verification purposes. The re-visit
is not counted in the statistics or facility counts presented in this section.
43 At three of the 18 facilities, more than one clinker sample was obtained for sampling and analytical quality
assurance purposes.
-------
3-71
Only one volatile organic compound was detected in whole clinker. This compound is
toluene. It was detected only once and its quantitated level is close to the detection limit. For
reasons stated earlier, no TCLP or SPLP leachates were subjected to analysis for volatile organic
compounds.
There were no semi-volatile organic compounds detected in whole clinker. There were
two instances of quantitation estimates for di-n-butyl-phthalate at levels near 200 ppb. This is,
however, below the established analytical method detection limit. For reasons stated earlier, no
TCLP or SPLP leachates were subjected to analysis for semi-volatile organic compounds.
Several of the naturally occurring radionuclides were detected in the clinker samples
collected by EPA, including isotopes of lead, radium, uranium, thorium and potassium. For the
man-made elements, pIutonium-238 was detected in clinker from one facility, and plutonium-239
was detected in clinker from another facility.
The Agency has drawn no conclusions at this time regarding the significance of any of the
clinker data. The analytical results from clinker characterization for this study are available in
the EPA docket for this Report to Congress. The Agency invites comments on all aspects of this
clinker characterization data, including the above findings from the Agency's preliminary analysis
of the clinker data.
-------
3-72
Exhibit 3-30
Analytical Results of Clinker Analyses for Inorganics By Fuel Type
(includes only EPA* Data)
Constituents
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Lead
HWFueb
Analysis:
Units
Number of samples
Number of detects
Minimum detected value
Maximum detected value
Average of detected values
Number of samples
Number of detects
Minimum detected value
Maximum detected value
Average of detected values
Number of samples
Number of detects
Minimum detected value
Maximum detected value
Average of detected values
Number of samples
Number of detects
Minimum detected value
Maximum detected value
Average of detected values
Number of samples
Number of detects
Minimum detected value
Maximum detected value
Average of detected values
Number of samples
Number of detects
Minimum detected value
Maximum detected value
Average of detected values
No
Total
"S/k*
7
5
9.5
16.5
12.2
7
7
3.8
25.4
9.4
7
5
0.75
2.4
1.5
7
0
ND
_i
..
7
7
26.1
138
60.9
7
7
0.77
21
4.7
Yes
Total
«>S/1<8
9
9
5.3
27.9
13.4
9
9
1.4
14.7
7.3
9
9
0.86
2.7
1.4
9
0
ND
..
..
9
9
44.3
150
83.5
9
9
0.68
33.1
9.3
No
TCLP
mg/L
7
6
0.05
0.09
0.06
7
7
0.002
0.006
0.003
7
0
ND
._
—
7
0
ND
..
-
7
7
0.02
0.33
0.15
7
6
0.002
0.016
0.007
Yes
TCLP
•ng/L
9
7
0.05
0.08
0.06
9
9
0.002
0.004
0.003
9
0
ND
_.
-
9
0
ND
..
..
9
6
0.02
0.95
0.4
9
8
0.002
1.9
0.25
-------
3-73
Exhibit 3-30 (continued)
Analytical Results of Clinker Analyses for Inorganics By Fuel Type
(includes only EPA* Data)
Constituents
Mercury
Nickel
Selenium
Silver
Thallium
HWFuet
Analysis:
Units:
Number of samples
Number of delects
Minimum detected value
Maximum detected value
Average of detected values
Number of samples
Number of detects
Minimum detected value
Maximum detected value
Average of detected values
Number of samples
Number of detects
Minimum detected value
Maximum detected value
Average of detected values
Number of samples
Number of detects
Minimum detected value
Maximum detected value
Average of detected values
Number of samples
Number of detects
Minimum detected value
Maximum delected value
Average of detected values
No
Total
«ngft«
7
0
ND
„
..
7
7
13.1
82
33.9
7
1
0.52
0.52
0.52
7
6
1.5
8.2
3.5
7
1
0.19
0.19
0.19
Yes No
Total TCLP
ingflsg || mgfl.
9
1
0.07
0.07
0.07
9
9
20.1
91.1
36.5
9
2
0.92
1.6
1.3
9
5
1.1
9.9
3
9
2
0.18
0.54
0.36
7
1
0.0002
0.0002
0.0002
7
0
ND
_.
..
7
0
ND
__
—
7
3
0.01
0.06
0.02
7
1
0.001
0.001
0.001
Ves
TCLP
mg/L
9
0
ND
_.
„
9
1
0.03
0.03
0.03
9
3
0.001
0.014
0.009
9
3
0.01
0.06
0.03
9
0
ND
„.
»
* 1993 sampling and analysis data not included.
-------
3-74
CHAPTER THREE
CKD GENERATION AND CHARACTERISTICS
3.0 INTRODUCTION 1
3.1 CKD GENERATION 1
3.1.1 Dust Collection Devices 2
3.1.3 Quantities and Fate of CKD Generated 16
Differences in CKD Generation Rates Across Process Types 16
Differences in CKD Generation Rates Across Process Types and Fuel
Usage 18
Differences in Gross CKD Generation Rates 18
Differences in Net CKD Generation Rates 19
CKD Recycling 19
Recycling Differences 21
Differences in Fate of CKD Across Process Types 21
Differences in Fate of CKD Across Process Types and Fuel Usage 22
3.2 CKD GROSS CHARACTERISTICS 25
3.2.1 Physical Characteristics 25
3.2.2 Bulk Chemical Characteristics . . 29
3.3 CKD TRACE CHARACTERISTICS 30
3.3.1 EPA Sampling Program 30
3.3.2 Total Concentrations 32
Metals 33
Dioxins and Furans 37
General Chemistry 41
Volatile Organics 41
Semi-Volatile Organics 44
Pesticides 44
PCBs 45
Radionuclides 45
3.3.3 Leachable Concentrations 47
Metals 47
Dioxins and Furans 50
General Chemistry 50
Volatile Organics 50
Semi-Volatile Organics 50
Pesticides 52
PCBs 52
Radionuclides 52
3.4 STATISTICAL ANALYSES OF CKD CHARACTERIZATION RESULTS 53
3.4.1 Metals 53
3.4.2 Dioxins and Furans 61
3.5 CLINKER CHARACTERISTICS 62
-------
3-75
-------
3-76
LIST OF EXHIBITS
Exhibit 3-1
Flow Chart of Gross CKD Management Pathways 2
Exhibit 3-2
Air Pollution Control Devices Used at Cement Kilns 4
Exhibit 3-3
Schematic Diagrams of Common Types of Air Pollution Control Devices 5
Exhibit 3-4
1990 Gross CKD Collection by Different Types of Air Pollution Control Devices 7
Exhibit 3-5
Relationship Between Net and Gross CKD Generated in 1990 8
Exhibit 3-6
Gross and Net CKD Generated (1990) 10
Exhibit 3^7
Share of Net CKD Generated and Clinker Production Capacity (1990) 11
Exhibit 3-8
Facilities With High Net CKD Generation Relative to Clinker Capacity 12
Exhibit 3-9
Percentages of Gross CKD Recycled, Sold, and Wasted (1990) 13
Exhibit 3-10
Recycling Rates Among Facilities That Operate Dry Kilns 14
Exhibit 3-11
Recycling Rates Among Facilities That Operate Wet Kilns 15
Exhibit 3-12
Average CKD Generation Rates Per Ton of Product (1990) 17
Exhibit 3-13
Fate of CKD as a Percent of Gross CKD (1990) 23
Exhibit 3-14
Particle Size Distribution of CKD by Process Type 26
Exhibit 3-15
Particle Size Distribution of CKD
Midwest Portland Cement Company, Zanesville, Ohio 27
Exhibit 3-16
Hydraulic Conductivity of Freshly Generated and Managed CKD 29
Exhibit 3-17
Typical CKD Bulk Constituents 31
Exhibit 3-18
Trace Metal Concentrations in As Generated CKD 34
Exhibit 3-19
Trace Metal Concentrations in As Managed CKD 36
Exhibit 3-20
Trace Elements Commonly Found in Native Soils 37
Exhibit 3-21
Total Concentrations of Dioxins and Dibenzofurans in As Generated CKD .... 39
Exhibit 3-22
Total Concentrations of Dioxins and Dibenzofurans in As Managed CKD 40
-------
3-77
Exhibit 3-23
Summary of Dioxin and Dibenzofuran Concentrations in CKD 42
Exhibit 3-24
Summary of Combined 1992-1993 Dioxin/Furan Sampling Results
CKD 2,3,7,8, TCDD Toxicity Equivalence 43
Exhibit 3-25
Comparison of Maximum and Average Metals Concentrations in As Generated Dust with TC
Standards 48
Exhibit 3-26
Comparison of Maximum and Average Metals Concentrations in As Managed Dust with TC
Standards 49
Exhibit 3-27
TCLP Concentrations of Dioxins and Dibenzofurans in As Generated CKD .... 51
Exhibit 3-28
T-test Comparison of Fuel Burning Effects on Metals Concentrations
56
Exhibit 3-29
T-test Comparison of Kiln Type on Metals Concentrations
58
Exhibit 3-30
Analytical Results of Clinker Analyses for Inorganics By Fuel Type 64
-------
CHAPTER FOUR
CURRENT MANAGEMENT PRACTICES FOR CKD
4.0 INTRODUCTION
• Four general approaches are used in managing CKD at cement plants: direct recycling,
treatment and return to the kiln system, landfilling/stockpiling, and/or off-site use. As discussed
in Chapter 3, direct recycling of CKD to the raw feed is preferable when practical; however,
excessive alkali content, as well as other operational factors, may limit this practice.1 Dust that
is removed from the system may be disposed in waste management units (WMUs) or sold or
given away for beneficial use off site.
This chapter describes current management practices for CKD that is not recycled to the
kiln -- that is, it is removed from the kiln system. The first section discusses on-site land disposal
of CKD, including the three major types of WMUs. The potential environmental impacts of on-
site disposal of CKD, including potential exposure pathways and environmental protection
practices at WMUs, are examined in the second and third sections. Finally, in the last section,
the off-site beneficial uses of CKD are described.
4.1 ON-SITE LAND DISPOSAL
Waste CKD is most commonly land-disposed in on-site WMUs. Respondents to the 1991
PCA Survey (representing usable data from 79 plants and 145 kilns) reported that they land
disposed an average of about 33,000 metric tons of CKD per plant in 1990. Of this aggregated
average, wet process facilities disposed of 41,735 metric tons per plant and dry process facilities
disposed of 27,419 metric tons per plant. F^rtrapolating these averages for wet and dry kilns to
the entire industry, an estimated 4.2 million metric tons of CKD were land disposed nationwide
in 1990.
Of 81 facilities responding to the 1991 PCA Survey, 62 (77 percent) indicated that they
manage CKD on site (no off-site CKD WMUs have been reported). Only two of the
respondents indicated that they had more than one active WMU. All but one of the facilities in
the sample having a wet kiln disposes of CKD on site, and the net CKD from about two thirds of
the wet kilns is sold for off-site use. CKD is disposed at somewhat more than half of the dry
kilns with preheaters/precalciners. CKD from all but four of the 32 kilns burning hazardous
waste is disposed on site, while the CKD from 66 of 93 kilns not burning hazardous waste is so
managed. Some facility operators view CKD placement in waste management units as temporary
stockpiling rather than disposal, with the expectation that the CKD will ultimately be removed
for beneficial utilization. The 23 percent of facilities that do not dispose CKD in WMUs are
those that recycle all of their CKD, or sell all of their net CKD.
The 1991 PCA Survey defined land disposal units for CKD as being comprised of three
general types: landfills, piles, and ponds. Ninety-seven percent of the respondents with WMUs
1 In this case, CKD can be treated using various methods to remove alkalies and can then be recycled to the kiln
system. These methods are described in Chapter 8. When in-plant closed loop recycling has reached its practical
limit, CKD must be removed from the system.
-------
4-2
classified their units as one of these three types. Three percent of the responses, or two
facilities, indicated that they managed CKD using a slight variation of these types.
4.1.1 Landfills
At cement production facilities, landfills are reportedly the most common on-site
management method. Landfills accounted for 52 percent of the WMUs in the survey responses.
Landfills are generally defined as WMUs in which material is disposed below the topographic
grade and is sometimes buried between layers of earth. Usually landfills are artificial structures
equipped with an engineered liner and a leachate/run-off collection system. For CKD disposal,
however, landfills are generally not engineered structures; that is, they generally were not
constructed in the manner that current solid waste landfills are constructed, with liners or run-off
collection systems. CKD is typically dumped into a retired portion of the existing limestone
quarry or in a nearby ravine. The CKD is either transported by truck to the quarry, pumped as a
slurry, or insufflated through pipelines. In a typical operation, CKD is transported by truck to
the quarry where it is dumped at the edge. The dust typically remains where dumped for a
period of weeks or months to "weather," after which it is bulldozed over the edge into the quarry.
As an example, the River Cement plant in Festus, Missouri transports pelletized waste
dust to its on-site CKD monofill. The monofill is located in a ravine that the facility has
reportedly closed off with an earthen berm at its base. Once at the monofill, the CKD is
bulldozed into a desired location.
4.1.2 Piles
With a slightly lower count than landfills, 43 percent of the WMUs in the survey
responses were reported as piles. Like landfills for CKD, piles also are not engineered structures
but are instead accumulations of CKD in designated areas. Such piles may or may not be above
grade, and they may or may not be contained within the quarry. EPA believes that there was
probably little differentiation between piles and landfills from the perspective of the respondents
to the survey.
For example, the Ash Grove Cement plant in Inkom, Idaho, prior to installing a CKD
dust leaching system approximately 20 years ago, disposed waste CKD in several large piles at
the edge of the facility's limestone quarry. The piles have since been covered with shale, and
during a May 1993 sampling visit attended by EPA personnel, vegetation was observed growing
on many of the pile surfaces. During another sampling visit conducted in May 1993 at the
Keystone Cement plant in Bath, Pennsylvania, EPA personnel observed several large CKD piles
that had accumulated on open ground adjacent to active cropland.
4.1.3 Ponds
Disposal ponds at cement plants are different from landfills and piles in that CKD is
stored underwater. This is an unusual WMU type and only one of the survey respondents
indicated that it managed CKD in this manner. However, the Holnam Incorporated cement
plant in Artesia, Mississippi has constructed an active CKD pile along the edge of an inactive
limestone quarry, which has filled with water to form a lake. During a May 1993 sampling visit
to the facility attended by EPA personnel, the active CKD pile was observed extending into the
lake. The active pile will be extended further into the lake as more CKD is added.
-------
4-3
Use of ponds creates a permanent hydraulic head on the dust, which imposes a
continuous downward pressure and creates the potential for downward migration of
contaminants into ground water. Additional discussion of this type of WMU and its implications
for environmental and human health risk may be found in Chapter 6.
4.1.4 Dimensions
The size of CKD waste management units can vary considerably, depending upon such
factors as unit type, age, and the quantity of dust discarded. The 1991 PCA Survey responses
generally did not provide clear data on the volume of CKD contained in the waste management
units. However, unit thickness and basal area measurements were reported, and these are
presented in Exhibit 4-1. The information available from the survey is insufficient to calculate
total volumes, because unfounded assumptions of uniform unit geometry (e.g., a cylindrical or
rectangular shape) would provide inaccurate results.
As shown in Exhibit 4-1, piles tend to be the largest units, attaining a maximum height or
thickness of 56.4 meters (m) (185 feet), according to usable responses to the survey, and
averaging 15 m in thickness or height. Landfills average 14 m in thickness, with a maximum
reported thickness of 34.6 m. The landfill units averaged twice the basal area of the piles, at
approximately 7.9 hectares (19.4 acres), compared to about 3.6 hectares for the piles. These
units can occupy significant land areas, covering up to 54.2 hectares. The pond and the "other"
units are very small in comparison in basal area.
Exhibit 4-1
1991 CKD Waste Management Unit Dimensions
Pile
Landfill
Pond
Other
# WMUs
With Usable
Responses*
18 of 28
15 of 34
lof 1
Iof2
Thickness
(meters)
Min.
3.05
43
3.7
24.7
Max.
56.4
34.6
3.7
24.7
Avg.
15
14
3.7
24.7
Basal Surface Area
(hectares)
Min.
.04
.19
.36
.03
Max.
8.4
54.2
.36
.03
Avg.
3.8
7.9
.36
.03
' Based on usable responses to the 1991 PCA Survey.
4.1.5 Codisposal
Facility operators also use land disposal units for small quantities of materials other than
CKD. Information provided in the PCA Survey responses indicated that, the 66 CKD WMUs,
23 percent contained non-CKD waste materials in addition to CKD. These materials, totalling
22,333 metric tons, include furnace brick, concrete debris, and tires, and constitute less than one
percent of the material reporting to disposal in these units in 1990. The quantity of quarry
overburden co-disposed with CKD in 1990 nearly equalled CKD disposal quantities in 1990.
-------
4-4
This material, because of its earth-like nature, was not considered a "waste material" when
performing this analysis.
4.1.6 Remaining Useful Life
As facility operators continue to land-dispose CKD, the available capacity of existing
waste management units will decrease. The 1991 PCA Survey responses yielded data regarding
the remaining useful life of WMUs. Exhibit 4-2 provides a breakdown of these data in 10-year
intervals. Of the 53 respondents with usable data, most (55 percent) of the CKD WMUs will be
full to capacity within the next 20 years.
Exhibit 4-2
Remaining Life of Waste Management Units'
Remaining
Useful Life
(Range in Years)
0-9
10-19
20-29
30-39
4049
50-59
60-69
70-79
80-89
90-99
100-109
200-209
Total
Number of
CKD Waste
Management
Units
13
16
9
1
5
3
1
0
0
1
3
1
53
Percent of
CKD Waste
Management
Units"
24.5
30.2
17.0
1.9
9.4
5.7
1.9
0.0
0.0
1.9
5.7
1.9
100
1 Based on usable responses from 1991 PCA Survey
b (# CKD WMU's in a given range / Total # CKD WMU's )
x 100
4.2 POTENTIAL EXPOSURE PATHWAYS
CKD management practices may affect human health and the environment through three
primary exposure pathways: ground water, surface water, and air. The potential for release to
these media varies according to the CKD management practices and the control measures
-------
4-5
employed at a given facility. This section introduces the mechanisms by which CKD constituents
are released to each medium, while Chapter 6 evaluates the human health and environmental
risks associated with the different release and exposure pathways. Exhibit 4-3 presents a layout
of a typical cement plant that illustrates potential exposure pathways. At a typical facility, a
surface water body runs past the facility, while CKD is managed in a waste management unit in a
retired quarry that is near ground-water level.
Exhibit 4-3
Typical Cement Plant Layout
4.2.1 Ground Water
Precipitation may percolate through CKD and leach constituents into the liquid phase.
Based on the constituents' tendency to remain bound up in the matrix of the waste (i.e., the
mobility of the constituents and their solubility in water), they may then migrate through the
vadose zone (i.e., unsaturated zone) and enter an underlying aquifer. After release to ground
water, the constituents will move with the general flow of the ground water, although at a
-------
4-6
velocity that is slower than the ground water itself depending on their individual tendencies to
bind to soil. Exposure to ground-water contaminants can occur through the domestic use (e.g.,
drinking water source) of untreated ground water. Potential migration to ground water is a
particular concern when CKD is managed underwater, such as in surface impoundments or
flooded quarries. The standing water column in these types of waste management units exerts a
downward pressure (hydraulic head) that forces water through the vadose zone to the ground
water.
For CKD management, ground water may be a potential exposure pathway because
several CKD constituents (e.g., arsenic) are particularly mobile in ground water under high-pH, a
condition associated with CKD leachate. In addition, many of the facilities are underlain by
shallow aquifers, and only a few facilities have control measures in place to prevent or detect the
migration of CKD leachate. Such control measures include the installation of synthetic or
natural (e.g., compacted clay) liners, installation of leachate detection/collection systems, capping
CKD management units to prevent leaching by precipitation, and installation of slurry walls to
prevent lateral ground-water migration.
4.2.2 Surface Water
Stormwater run-off from a waste management unit is an important release mechanism, as
precipitation may carry constituents in either a dissolved or suspended form through natural flow
patterns to nearby surface waters or farm fields. Flooding or overflow of submerged WMUs may
also result in CKD constituents being released to streams and rivers. In addition, constituents of
concern may be released to surface water by migrating through ground water that discharges to a
surface water body.
Human and ecological receptors may be exposed to surface water contamination through
various means, including drinking water intake, ingestion of contaminated fish or shellfish, and
direct contact with contaminated water. Common practices for controlling releases2 to surface
water include the leachate controls as described above for ground water; stormwater run-on/run-
off controls that divert water from piles on landfill areas and/or collect run-off from the WMU
for treatment prior to release to surface waters; and capping or covering.
4.2.3 Air
CKD constituents can be released to the air in the form of a gas or a particle. The only
constituents that can be released as a gas are volatile or semivolatile organic chemicals (e.g.,
benzene and toluene), which tend to be present in relatively low concentrations in CKD, if
present at all. Most CKD constituents (e.g., metals) are not volatile but could be released to air
through fugitive dust emissions. Dust particles may be suspended in the air by either wind
erosion or mechanical disturbances. The extent to which dust is blown into the air by wind
erosion depends on a number of site-specific characteristics, including the texture (particle size
distribution) and moisture of CKD on the surface of piles, the presence of nonerodible elements
such as clumps of grass or stones on the pile, the existence of a surface crust, and wind speeds.
Mechanical disturbances that can serve to suspend CKD constituents in the air include vehicular
traffic on and around CKD piles, CKD dumping and loading operations, and transportation of
CKD around a plant site in uncovered trucks. Cement plants may use a variety of control
2 Stormwater controls and other regulatory requirements addressing releases to surface waters are described in
Chapter 7.
-------
4-7
measures to limit the release of CKD to the air.3 For example, CKD may be "nodulized" in a
pug mill, compacted, wetted, covered, and/or mixed with large chunks that are not susceptible to
wind erosion.
CKD constituents that are released to the air are transported and dispersed by the winds
and eventually deposited onto land or water, either by settling in a dry form or by being
entrained in precipitation. Humans and other organisms can be exposed to the constituents in a
number of ways. For example, airborne particles that are equal to or smaller than 10
micrometers (jtm) in size are respirable and may be inhaled directly. Contaminants that have
settled onto soil may be incidentally taken into the mouth and ingested, and contaminants that
have been deposited on vegetation may be ingested via the food chain. In the specific case of
radionuclides, people may be exposed to direct radiation emanating from radionuclides in the air
or deposited onto the ground.
4.3 ENVIRONMENTAL PROTECTION PRACTICES
As noted above, CKD WMUs can represent permanent placement of a large volume of
material that extends over a significant area. Because CKD disposal units are generally
uncovered, they are subject to all of the climatic conditions of the geographic region in which
they are located. Precipitation events are notable because they can transport particles and
solubilized constituents beyond the boundaries of the WMU. Ousting winds are another
potential transport mechanism.
To reduce the potential for off-site migration of CKD and CKD constituents, facilities
managing CKD in WMUs can employ various environmental protection practices. These include
run-off control/collection, run-off collection/treatment, leachate control/collection, leachate
collection/treatment, slurry walls, liner systems, dust suppression, dust compaction, ground-water
monitoring, and the preparation and implementation of closure plans. Exhibit 4-4 displays these
practices as reported for 66 WMUs at the 62 facilities responding to the PCA Survey for which
usable data were available. To relate these data to analyses presented in previous sections, EPA
has classified WMUs by those receiving CKD from kilns burning hazardous waste and those
receiving CKD from kilns not burning hazardous waste. Although statistical conclusions are
tenuous given the small number of observations in some cases, Exhibit 4-4 reports the
frequencies of each practice as both number and percentage of WMUs in the respective fuel use
category. It should be noted that a single WMU may employ several of the listed environmental
protection practices.
Run-off control/collection involves diverting precipitation away from the WMU to a
discharge area (e.g., a stream) or into a collection unit (e.g., a treatment impoundment) or
directly to a receiving stream. As presented in Exhibit 4-4, about 50 percent of both kilns
burning hazardous waste and those not burning hazardous waste perform some type of run-off
control. The diverted run-off is typically either discharged directly to a surface water stream, or
discharged after some form of treatment. Existing treatment methods are unknown, though over
20 percent of the WMUs possess systems that reportedly treat collected run-off.
Leachate controls are any devices or approaches (e.g., underdrains) to prevent aqueous
liquid that has entered managed CKD from exiting the WMU in an uncontrolled manner,
5 Plant-specific air pollution control permits often explicitly address fugitive dust emissions from CKD piles and
other sources. A more complete discussion of this topic is presented in Chapter 7.
-------
4-8
particularly to the ground water. Overall, about 18 percent of the WMUs have leachate control,
while about half that number also treat the leachate in some manner. This practice is more
prevalent among the facilities not burning hazardous waste; 10 of 50 WMUs in this category have
leachate controls, while only two of 16 WMUs containing CKD from hazardous waste-burning
kilns are so equipped.
Slurry walls are very low-permeability walls cast-in-place in trenches of varying depth and
width around a waste management unit. This technology is one of the more costly environmental
control measures, and is reported at only nine of the 66 WMUs (approximately 15 percent)
overall. Seven of these nine are WMUs receiving CKD from hazardous waste burners, while
only two (representing four percent) of the WMUs from kilns not burning hazardous waste have
such devices.
-------
4-9
Exhibit 4-4
Environmental Protection Practices
at CKD Waste Management Units Active in 1990, by Kiln Fuel Use Type
Environmental
Protection Practices*
Run-off Control/Collection
Run-off Collection/Treatment
Leachate Control/Collection
Leachate Collection/Treatment
Slurry Walls
Modified Natural Liner
Dust Suppression
Dust Compaction
Other
Ground-Water Monitoring
Closure Plan (Approved)
Closure Plan (Not Approved)
None
Total in Response Groupb
All Fuel Types
No. of
WMUs
34
15
12
5
9
7
29
22
14
11
10
2
26
66
Percent of
WMUse
51
22
18
8
14
11
44
33
21
17
15
3
39
Hazardous Waste
Burners
No. of
WMUs
8
4
2
1
7
1
9
6
3
4
3
1
5
16
Percent of
WMUs
50
25
13
6
44
6
56
38
19
25
19
6
31
Not Hazardous Waste
Burners
No. of
WMUs
26
11
10
4
2
6
20
16
11
7
7
1
21
50
Percent of
WMUs
52
22
20
8
4
12
40
32
22
14
14
2
42
* A WMU and/or a facility may have more than environmental protection practice.
b Based on 81 usable facility responses to 1991 PCA Survey, 62 facilities of which had active WMUs in 1990.
e Calculated as number of WMUs in a fuel-type column for the relevant environmental protection practice
divided by the total WMUs for that fuel type. For example: 8 WMUs from hazardous waste burners practice run-off
control/collection, divided by 16 total WMUs from hazardous waste burners, equals 50 percent.
-------
4-10
Liners may be used in a WMU to restrict leachate from entering permeable soil layers
and underlying ground-water aquifers. Liners can be natural or synthetic. Examples of natural
liners are the indigenous bedrock or in-situ clay/shale. Natural liners can be modified through
compaction to reduce downward migration channels. Synthetic liners may include compacted
clay/shale, asphalt, concrete, or a manufactured woven fabric. Twenty-two percent of the
respondents to the 1991 PCA Survey who have WMUs indicated that liners are not used. Of the
78 percent of respondents who indicated that they did use liners in their WMUs, none reported
use of synthetic liners. This information is in accordance with EPA's observations during site
visits, which revealed that all WMUs visited had only natural liners, typically the bedrock within
a retired portion of a limestone quarry. As shown in Exhibit 4-4, seven (about 11 percent) of the
respondents across both fuel types stated that they used modified natural liners.
Dust suppression/control is defined in the 1991 PCA Survey as any means of reducing the
level of ambient breathable dust. Controls under this practice can include wetting, compacting,
or covering CKD. Almost half of the WMUs across both fuel types reportedly have some form
of dust control system. Dust compaction involves the densification of waste material to increase
available disposal space and ameliorate dust migration. About 30-40 percent of the WMUs of
both fuel types reportedly undergo some type of dust compaction.
Respondents to the "other" category indicated methods used to cover or contain a CKD
WMU (soil cap, clay cap, berm, rip-rap cap, tree planting, etc.). Such activities were reported
for about 20 percent of units, both within and across fuel types.
Overall, approximately 17 percent of WMUs have some type of ground-water monitoring
system. In absolute terms, more non-hazardous waste burners (seven) monitor ground-water
quality than hazardous waste burners (four), though in percentage terms, nearly twice as many
hazardous waste burners in the sample of 66 monitor ground water as non-hazardous waste
burners (25 percent vs. 14 percent).
In general, closure plans do not appear to be significantly more common for WMUs at
hazardous waste-burning facilities than for WMUs at facilities that do not burn hazardous waste.
Ten of the twelve units addressed by a closure plan have been approved by the pertinent
regulatory agency.
Finally, 39 percent of the WMUs have none of the environmental controls listed in
Exhibit 4-4; this finding applies to 31 percent of WMUs from hazardous waste-burning facilities,
and to 42 percent of the facilities not burning hazardous waste. Since, however, the quality and
effectiveness of reported systems is unclear, it is difficult to assess what, if any, increased
exposure risks might exist at these WMUs compared to-WMUs that do utilize environmental
protection practices. As mentioned above, the implications of the use or lack of use of various
environmental protection practices is discussed in greater detail in Chapter 7.
4.4 BENEFICIAL USE OF CKD
When CKD is not put back into the kiln or disposed on site, a facility may sell it for off-
site beneficial use. EPA's data regarding beneficial uses of CKD came from two sources, the
1991 PCA Survey and cement facility responses to EPA's 1992 request for information under
RCRA section 3007. Besides information on the beneficial uses of CKD, the PCA Survey data
included information on gross and net CKD generation rates, kiln type, and fuel type. The
§3007 responses contained information on beneficial uses of CKD, but did not include that other
information. Some facilities submitted both the 1991 PCA Survey and a response to the §3007
-------
4-11
request for information, however, there were other facilities that only sent a response to the
section 3007 request. Therefore, only 1991 PCA Survey data were used to relate off-site use to
generation rates, kiln type, and fuel type. The aggregated data were used to calculate
percentages of CKD sold (or given away) for off-site beneficial uses.
Of the approximately 9.8 million metric tons (9.4 million tons) of CKD generated from
145 kilns at 79 plants providing usable responses in the 1991 PCA Survey, approximately 6.5
percent of gross CKD was sold off site or given away for beneficial use by 44 facilities.
Responses to EPA's 1992 RCRA §3007 request indicate that at least 15 additional plants sold or
gave away CKD in 1990. Respondents to the 1991 PCA Survey sold (or gave away) an average
of 3,920 metric tons of CKD per kiln in 1990, representing 5.8 percent of the gross CKD. Of
this aggregated average, wet kilns sold 7,833 metric tons per kiln and dry kilns sold 1,866 metric
tons per kiln, accounting for 13.3 percent and 2.6 percent of gross CKD. Of the dry kilns, dry
long kilns sold or gave away 1,993 metric tons per kiln (2.8 percent of gross CKD) and dry kilns
with preheaters/precalciners sold 1,721 metric tons per kiln (2.3 percent of gross CKD). Wet
kilns sell or give away a higher average percentage of their gross CKD than dry kilns. When
considering fuel type, there is no apparent link between fuel type and percent of gross CKD sold
or given away. Dry kilns that burn hazardous waste sell or give away a higher percentage of
gross CKD than dry kilns that do not burn hazardous waste. On the other hand, wet kilns that
do not burn hazardous waste sell or give away a higher percentage of gross CKD than wet kilns
that burn hazardous waste.
The primary end-use applications for CKD sold off site as categorized in the 1991 PCA
Survey were waste stabilization, soil amendment (both as a soil stabilizer and as a fertilizer),
liming agent, materials addition, road base, and "other." Exhibit 4-5 below provides survey and
§3007 data regarding end-use applications for CKD sold off site. As shown in the exhibit, 71
percent of the approximately 0.94 million metric tons of CKD sold off site in 1990 was used for
waste stabilization. Soil amendment accounts for the second largest use, approximately 12
percent. The category "other" includes uses such as wet scrubbing or general undefined
agricultural use. These categories are briefly discussed below. More detailed discussion of both
these current and potential beneficial utilizations of CKD may be found in Chapter 8. In
addition, researchers have investigated using CKD in other applications, including as an
ingredient in livestock feed, as a lime-alum coagulant, as a mineral filler, as an ingredient in the
manufacture of lightweight aggregate, and as a replacement for soda ash in the manufacture of
green glass.
4.4.1 Waste Stabilization
Waste stabilization was, by far, the most common beneficial use of CKD, accounting for
just under 71 percent of the total in 1990 (see Exhibit 4-5). CKD can absorb excess liquids and
provide an alkaline environment to neutralize acids. Through its absorption capacity, CKD can
dewater contaminated materials to increase weight-bearing capacity and to reduce the presence
of free leachate. One of the primary forms of waste stabilization for which CKD is used is for
municipal sewage treatment sludge. It is an economical and effective means of dewatering and
stabilizing raw or digested sewage treatment sludges, thereby rendering such sludges more
conducive to handling. The treated sludges can then be used as landfill cover, structural fill
material, dike construction material,
-------
4-12
Exhibit 4-5
Estimated Off-Site Uses for CKD Sold/Given Away*
Off-Site Uses of CKD
Total CKD Sold/Given Away
1990
Used for Waste Stabilization
Used as Soil Amendment
Used as Liming Agent
Used for Materials Addition
Used as Road Base
Other
Quantity
(Metric
Tons)
940,000
668,274
110,676
52,480
25,365
10,832
75,840
% of Total
Off Site"
100.0
70.8
11.7
5.6
2.7
1.2
8.0
# Facilities
Burning HW
16
15
1
1
0
0
0
# Facilities
Not Burning HW
47
35
13
5
3
4
13
1 The information in this table was estimated from the 1991 PCA Plant Information Summary, the 1991 PCA
Surveys returned by 88 facilities and the responses of 85 facilities to the EPA's 1992 request for information under
RCRA section 3007. The data obtained thereby address 109 facilities. The data in this table were extrapolated to the
industry as a whole, i.e., from 109 to 115 facilities.
b (tons used off site for given use/total tons used off site) x 100
and in agricultural applications.*15'6-7 Waste stabilization with CKD is found at wastewater
treatment plants (WWTPs) and chemical production facilities.
In addition to municipal sludge stabilization, the use of CKD to solidify oil sludge also
has evoked a fair amount of interest and research.8-9-10 According to one source, CKD has
4 Keystone Cement Company, date unknown. StableSorb: A Coproduct of Cement Manufacturing With a Variety of
Uses. Product Brochure.
5 Bumham, J.C., 1988. CKD/Lime Treatment or Municipal Sludge Cake, Alternative Methods For Microbial and
Odor Control. Paper from Proceedings of National Conference on Municipal Sewage Treatment Plant Sludge
Management. June 27-29. Palm Beach, Florida.
6 Personal communication with J. Patrick Nicholson, N-Viro Soil, December 7, 1992.
7 Kelley, W.D., D.C Martens, R.B. Reneau, Jr., and T.W. Simpson, 1984. Agricultural Use of Sewage Sludge: A
Literature Review. Bulletin 143. Virginia Water Resources Research Center, Virginia Polytechnic Institute and State
University, Blacksburg, Virginia. December, p. 38.
* Morgan, David S., et al., 1984. Oil Sludge Solidification Using CKD. Journal of Environmental Engineering.
October.
9 Thorsen, J.W., et al., 1983. In Situ Stabilization and Closure of an Oily Sludge Lagoon. 3rd Ohio Environmental
Conference. March. Columbus, Ohio.
-------
4-13
proven to be one of the most efficacious and economical means of solidifying non-recoverable
waste oil sludge, producing a stable and compactible fill material with good compressive strength.
Solidification of oily sludge in landfills makes it possible to use a reclaimed landfill site for
industrial construction.11
CKD has also been used to neutralize and stabilize some additional wastes such as acid
waste, synthetic wastes, contaminated dredged materials, and non-degradable liquid hazardous
wastes. Finally, CKD has been used both alone and in conjunction with other soil stabilizing
agents to temporarily or permanently increase the stability of soils at locations such as
construction sites. CKD has also been utilized on a limited basis to reclaim settling ponds,
lagoons, and abandoned mines.
4.4.2 Soil Amendment (Fertilizer)
As shown in Exhibit 4-5, about 12 percent of the CKD used beneficially in 1990 was used
as a soil amendment, mostly as fertilizer. Like agricultural lime, CKD is alkaline and contains a
number of essential plant nutrients. Because of these parallel characteristics, CKD has been
used as an agricultural soil amendment for a number of years. CKD possesses significant
fertilizer potential, particularly because of its high potassium content. It has been used to this
end at the state and local levels in Ohio, Illinois, and Pennsylvania, because it provides savings
over substitute products.12 Agricultural studies relating to the use of CKD as a fertilizer have
been undertaken in several countries around the world, including Russia, Poland, Netherlands,
Czechoslovakia, and India.
Although there has been a considerable amount of research conducted on CKD use as a
fertilizer, existing applications of CKD for this purpose have been mostly anecdotal, and there is
only limited evidence that commercial CKD use as a fertilizer is growing significantly.
4.4.3 Liming Agent
Nearly six percent of the CKD used beneficially in 1990 was used as a liming agent.
CKD has significant potential as a liming agent because of its high alkalinity. Substances that
can and have been neutralized with CKD include industrial acidic wastes such as spent pickle
liquor, leather tanning wastes, and cotton seed delinting chemicals. CKD also has been used as
an agricultural liming agent to treat acidic soils. In the mid-eighties, it was used as an
agricultural lime on a regional basis in New York.13-14
10 Zarlinski, SJ. and J.C. Evans, 1990. Durability Testing of a Stabilized Petroleum Sludge. Paper from Hazardous
and Industrial Wastes, Proceedings of 22nd Mid-Atlantic Industrial Waste Conference. July 24-27. Pennsylvania.
11 Morgan, David S., et al., 1984. op. cit.
12 Personal Communication with Marc Saffley, Soil Conservation Service (SCS), November 18, 1992.
u Naylor, L.M., J.C. Dagneau, and I.J. Kugelraan, 1985. CKD - A Resource Too Valuabk to Waste? Proceedings
of the Seventeenth Mid-Atlantic Industrial Waste Conference on Industrial and Hazardous Wastes. June 23. pp. 353-
366.
14 Naylor, L.M., E.A. Seme, and TJ. Gallagher, 1986. Using Industrial Wastes in Agriculture. BioCycle. February.
pp. 28-30.
-------
4-14
4.4.4 Materials Additive
Approximately 2.7 percent of CKD used beneficially was used in 1990 for materials
additive applications, where CKD is blended with cement either alone or with other additive
materials and aggregates to make concrete. CKD also has been used as a mineral filler for
bituminous paving materials and asphaltic roofing materials. In addition, glassmakers have used
CKD in glass that does not have stringent color restrictions or requirements for chemical
stability.
4.4.5 Road Base
Approximately 1.2 percent of the CKD used beneficially in 1990 was used for road base
construction. CKD provides an economically attractive substitute for road base products such as
fill materials and lime. Use of CKD for this purpose has, however, been limited thus far and the
subject does not appear to have attracted much continuing attention.
4.4.6 Other Uses
About eight percent of the CKD used beneficially in 1990 was used in other uses (e.g.,
wet scrubbing and general undefined agricultural use).
-------
4-15
CHAPTER FOUR
CURRENT MANAGEMENT PRACTICES FOR CKD
4.0 INTRODUCTION 1
4.1 ON-SITE LAND DISPOSAL 1
4.1.1 Landfills 2
4.12 Piles 2
4.13 Ponds 2
4.1.4 Dimensions 3
4.1.5 Codisposal 3
4.1.6 Remaining Useful Life 3
4.2 POTENTIAL EXPOSURE PATHWAYS 4
4.2.1 Ground Water 5
4.2.2 Surface Water 6
4.2.3 Air 6
43 ENVIRONMENTAL PROTECTION PRACTICES 7
4.4 BENEFICIAL USE OF CKD 9
4.4.1 Waste Stabilization ; 10
4.4.2 Soil Amendment (Fertilizer) 12
4.43 Liming Agent 12
4.4.4 Materials Additive 12
4.4.5 Road Base 13
4.4.6 Other Uses 13
-------
4-16
LIST OF EXHIBITS
Exhibit 4-1 1991 CKD Waste Management Unit Dimensions 3
Exhibit 4-2 Remaining Life of Waste Management Units 4
Exhibit 4-3 Typical Cement Plant Layout 5
Exhibit 4-4 Environmental Protection Practices at CKD Waste Management Units
Active in 1990, by Kiln Fuel Use Type 8
Exhibit 4-5 Estimated Off-Site Uses for CKD Sold/Given Away 11
-------
CHAPTER FIVE
DOCUMENTED AND POTENTIAL DAMAGES FROM MANAGEMENT OF CKD
5.0 INTRODUCTION AND METHODOLOGY
Section 8002(o)(4) of RCRA requires that EPA's study of CKD waste examine
"documented cases in which danger to human health or the environment has been proved." In
order to address this requirement, EPA defined danger to human health or the environment in
the following manner. First, danger to human health includes both acute and chronic effects
(e.g., directly observed health effects such as elevated blood lead levels or loss of life) associated
with management of CKD waste. Second, danger to the environment includes the following
types of impacts:
(1) Significant impairment of natural resources (e.g., contamination of any current or
potential source of drinking water, with contaminant concentrations exceeding
drinking water and/or aquatic ecologic standards);
(2) Ecological effects resulting in degradation of the structure or function of natural
ecosystems and habitats; and
(3) Effects on wildlife resulting in damage to terrestrial or aquatic fauna (e.g.,
reduction in species' diversity or density, or interference with reproduction).
This approach parallels that used in the previous RCRA §8002 studies prepared by the Agency.1
This section describes the approach the Agency has employed to address the §8002(o)(4)
requirement, including the "tests of proof and the methods used to identify potential cases,
information on actual damage cases, and verification of the accuracy and completeness of the
resulting case studies. In addition, this section provides a discussion of the limitations associated
with interpreting the results obtained. Throughout the discussion, cases where damage to the
environment has been proved are referred to as damage cases.
"Tests of Proof
The statutory requirement is that EPA examine proven cases of danger to human health
or the environment. Accordingly, EPA developed "tests of proof to determine if
documentation available on a case provides evidence that danger/damage has occurred. (These
are the same criteria used in the Report to Congress on Special Wastes from Mineral
Processing.) These "tests of proof consist of three separate tests; a case that satisfies one or
more of these tests is considered "prove." The tests are as follows:
• Scientific investigation. Damages are found to exist as part of the findings of a
scientific study. Such studies should include both formal investigations supporting
litigation or a state enforcement action, and the results of technical tests (such as
1 See, for example, U.S. EPA, 1990. Report to Congress on Special Wastes from Mineral Processing. Office of
Solid Waste. July.
-------
5-2
monitoring of wells). Scientific studies must demonstrate that damages are
significant in terms of impacts on human health or the environment. For
example, information on contamination of a drinking water aquifer must indicate
that contamination levels exceed drinking water standards.2
• Administrative ruling. Damages are found to exist through a formal
administrative ruling, such as the conclusions of a site report by a field inspector,
or through existence of an enforcement action that cited specific health or
environmental damages.
• Court decision. Damages are found to exist through the ruling of a court or
through an out-of-court settlement.
Identification of Prospective Damage Cases
EPA identified damage case sites by compiling a list of (1) currently operating cement
manufacturing facilities and currently inactive or closed facilities that were active during the last
10 to 20 years based on industry and government sources (e.g., the Portland Cement Association
and the U.S. Bureau of Mines); and (2) cement manufacturing facilities investigated under the
Comprehensive Environmental Response, Compensation and Liability Act (CERCLA) and listed
in EPA's CERCLIS data base. Additional facilities were identified during the information
collection process described below when state or federal contacts indicated that these facilities
should be considered. The initial search resulted in the identification of 127 active and inactive
sites as a basis for searching records for documented damages. Some cases, though they did not
meet the "tests of proof were identified as "potential" damage cases because they showed
evidence of on-site contamination, but lacked any information regarding whether or not
contaminated media migrated off site.
Information Collection
In addition to gathering information from its regional offices, EPA contacted state, other
federal, and local agencies to collect information. Telephone contacts were made with agencies
in all states in which cement is currently produced. These agencies included state environmental
regulatory agencies; state, regional, or local departments of health; and other agencies potentially
knowledgeable about damages related to the management of CKD waste. EPA also contacted
professional and trade associations, and public interest and citizens groups, seeking additional
information and perspective on prospective damage cases.
The Agency then visited four states identified in the initial telephone screening to collect
information about specific sites from state and local agency files. These four states (California,
Missouri, South Carolina, and New York) account for 23 of the 127 sites investigated for
potential documented damages. EPA selected the states to be visited based on (1) the type and
extent of site-specific information available in the files (based on contacts with state and local
personnel); and (2) the ability of the Agency to combine data collection activities with scheduled
CKD sampling visits. Where feasible, information also was collected by mail from state and local
agency personnel. EPA did not conduct file searches in all states in which CKD sampling visits
2 We recognize that comparison of drinking water standards and constituent levels in groundwater is not routine.
But because of the lack of benchmark standards for constituents in leachate, we believe it is a useful comparison.
-------
5-3
occurred because, based on contacts with state and local government personnel, EPA determined
that no relevant information was available in the files of some agencies.
During visits to the regulatory agencies in the four states described above EPA reviewed
documentation on sites on the list of potential damage cases, and collected documentation on
those cases that appeared to meet one or more of the "tests of proof." Follow-up contacts were
also made with relevant agencies, groups, and individuals based upon initial information review.
Damage Case Preparation and Review
Following completion of the data collection efforts, EPA prepared detailed damage case
study notes of the information obtained for documented damage case sites. These notes provide
the basis for the discussions of damage case findings for CKD waste management that are
covered in this report. The detailed damage case notes are available in the RCRA docket.
Limitations of the Damage Cases
The damage case findings that resulted from the process described above must be
interpreted with care, for several reasons. First, CKD waste disposal sites are often co-located
with limestone mining operations (e.g., active and exhausted quarries) that may also be used for
storage of other cement manufacturing feedstocks (e.g., petroleum coke). Similarly, and more
importantly, CKD waste is or has been co-managed with other wastes such as refractory brick at
many sites. In such cases, it is often difficult to determine if the documented damages were
caused by management of CKD waste, or if the stored raw material or co-managed waste may
have caused or contributed to the observed damage. The sites included in this report are those
for which available data indicate that the documented damages are attributable, in whole or in
large part, to the management of CKD waste.
Second, the extent to which the findings can be used to draw conclusions concerning the
relative performance of waste management practices among states is limited by variations in
requirements and recordkeeping. For example, recordkeeping varies significantly among states.
Some states have up-to-date central enforcement or monitoring records on cement
manufacturing facilities within the state. Where states have such records, information on
damages may be readily available.
More often, enforcement and monitoring records are incomplete and/or distributed
throughout regional offices within the state. Data collection efforts generally were focused on
the central office of the appropriate state agencies. In some instances, information may have
been available at a state regional office that was not available in the central office.
The third reason for caution is that, because CKD waste is not regulated under Subtitle C
of RCRA, many states do not specifically regulate the management of CKD at cement
manufacturing facilities. As a result, monitoring and, thus, detection of problems at cement
manufacturing facilities has occurred on a very limited basis, if at all, in some states. Therefore,
while damages may have occurred in states that do not have an environmental monitoring or
regulatory program specifically for CKD wastes, these damages could not be identified in this
study.
Finally, because environmental contamination resulting from waste disposal practices
often takes years to become evident, documented examples of danger that have resulted from
-------
5-4
particular waste disposal practices may reflect conditions that no longer exist. Specifically,
processing operations, waste characteristics, and/or waste management practices may have
changed. As a result, damage cases associated with CKD waste do not necessarily demonstrate
that current CKD waste management practices or regulations affecting CKD waste generation
and management are in need of change. Conversely, failure of a site to exhibit documented
damages at present does not necessarily suggest that past or current waste management practices
have not or will not cause damage. The Agency believes, however, that information on dangers
posed by past waste management practices is useful in understanding the potential for
environmental and human health impacts when releases to the environment occur.
5.1 OVERVIEW OF FINDINGS, TRENDS, AND CONCLUSIONS
5.1.1. Findings
Using the methodology described above, EPA collected information regarding damages to
human health and the environment at 115 cement plants that were active in the United States in
1990. EPA also investigated the possibility of damages at 12 additional sites, including
abandoned (inactive) cement plants and inactive, off-site disposal areas, at which CKD has been
disposed within the past 20 years.
Based on its investigation, the Agency compiled the following information concerning the
recorded documentation alleging human health and/or environmental damages at these 127 sites:
Exhibit 5-1
Summary of Cases of Documented and Potential Damage to
Human Health and/or Environmental
Number of Sites
90
15
19
3
Documented and Potential Damages
No allegations of damages.
Alleged damages: documentation insufficient to
support a test of proof.
Information available to support at least one test
of proof for damages. These are cases of
documented damage to surface water and/or
groundwater and/or air.
Information available to indicate that on-site
surface water has been impacted, but there is no
data to indicate that damaged media has impacted
ground water or has migrated off site. These are
cases of potential damage.
From its investigation of compliance with environmental regulations and CKD
management and disposal practices at these sites, EPA was able to document damages to human
health and/or the environment at 19 cement plants in the United States using the tests of proof
described above. Three additional sites are classified as cases of potential damage, because there
-------
5-5
is no substantial evidence the damaged media has migrated off site. Exhibit 5-2 presents a
summary of EPA's findings at seven of the sites where there has been documented damage to
surface water and/or ground water, including waste management practices at the time the
damage occurred, the environmental media impaired, and the chemical constituents of concern in
the affected media. They are among the 19 cases of documented damage identified above in
Exhibit 5-1. Exhibit 5-3 presents the same information for the two cases of potential damage.
The eight documented damages are described in more detail under Documented Ground and
Surface Water Damage Case Summaries in Section 5.2 of this chapter. The two documented
cases of potential damage are described after description of the eight cases of documented water
damage in Section 5.3. Documented damages to air were found at 12 sites and are listed in
Exhibit 5-17. Air damages are summarized in this chapter in Section 5.4.
5.1.2 Overall Trends and Conclusions
Damages the Agency has documented are in the form of exceedances of established
constituent limits; no direct impacts on human health have been demonstrated during the
conduct of this analysis. In cases where damages to surface and ground water from the
management of CKD have been documented, there are exceedances of a Federal or State
minimum concentration limits (MCLs) for constituents of drinking water, and/or exceedances of
aquatic/ecologic MCLs for constituents of surface water. In the air damage cases, damages are
exceedances of opacity limits adopted by States in compliance with the Clean Air Act.3 In all
damage cases the available data included no evaluation of or information on potential for actual
human exposure to waste constituents. Waste management practices included disposal in unlined
units: waste piles, abandoned quarries, or landfills; two of the 19 damage case facilities disposed
of CKD in off-site units. These waste management practices are common at many sites across
the country.
At five of the seven sites where documented water damages have occurred, both surface
water and ground water have been affected as a direct result of past waste management
practices. Typical concerns at these facilities include elevated pH, total dissolved solids, and
sulfate above secondary MCLs in ground water and surface water, as well as elevated levels of
toxic metals such as arsenic, cadmium, and lead above primary drinking water MCLs.
At the three sites where there are potential water damages, on-site surface waters have
been impacted by the disposal of CKD, but there is no significant evidence that these waters
have migrated off site. At one of these sites CKD is managed underwater in an inactive quarry.
In addition to the documented damages to both surface water and ground water, EPA
identified 21 incidents of air damage at 12 facilities. Notices of Violation (NOVs) were issued
for these incidents, with three cases eventually settled through a judicial settlement. Six of these
facilities have received more than one NOV. With the exception of two cases associated with the
accumulation of fugitive dust, all of the cases were associated with visible emission violations
(opacity) related to equipment and process malfunctions associated with the dust management
system.
5 Opacity is an indirect measurement of the concentration of PM10> the MCL of which is protective of human
health. Althoug
-------
5-6
There are several sites with waste management practices similar to those in the
documented damage cases that the Agency has investigated under CERCLA. At these sites, the
Agency either found no cause for further action under CERCLA, or recommended further
action that has not yet occurred. However, further action under CERCLA is based on a ranking
system which is weighted towards proximity to human population centers. Therefore, failure to
investigate further may overlook the existence of ecologic damage and/or risk to small human
populations.
-------
5-7
Exhibit 5-2
Summary of Documented Water Damages
Site
Waste Management Practice
Damaged Media
Constituents of
Concern
(concentrations
exceed MCLs)
Other Constituents of
Concern
(concentrations exceed
background levels)
Hnlnamlncorporated,
MasonCity,Jowa
LehighPortlandCement,
Leeds, Alabama
LehighPortlandCeraeat,
CKD former! jdisposedin abandoned^inlined
quarrjpartialljfitled with precipitatiomnd ground
water (water ta hie risesinto quarry). Currently
(I990)recyclesl00% of CKD.
CKD formerljdisposedin unlinedwastepilesat
edge of clay pit/quarry.Currenth/1990) recycles
100% of CKD.
CKDcnrrentlj(1990)disposedin on'site,unlioed
wastepilesat perinteiemf ahandonedquany^
CKD alsoformerljdisposedin off-site,unlined[
wastepilesat site currentljusedas Lime Creek
NatuwCenter.
GroundWater
SurfaceWater
SurfaceWater
GroundWater
SurfaceWater
elevatedpH, elevated Cl,K>]Sfa,
TDS,CI,Mg, phenols
SO4,Cr
eJevatedpH, ekvatedK,Na - ;
804'
elevatedpH,
TDS
SO4*
TDS, As
PortlandCementCompany,
Salt Lake City, Utah
CKD formerljdisposednearbybutin off-site, SurfaceWater
unlinedwaste pileswith drainageditchthroughsite
andsurpluxanaladjacentto site. Kilncurrently
inactive. GroundWater
Soils
elevated pH, As, elevated Mo
hexavalenCr, Pb
elevatedpH, elevated K, Mo
TDS, As, Cd,F,
Pb,S04
elevated Mo, Pb
-------
5-8
Exhibit 5-2 (continued)
Summary of Water Damages
Site
Waste Management Practice
Damaged Media
Constituents of
Concern
(concentrations
exceed MCLs)
Other Constituents of
Concern
(concentrations exceed
background levels)
Soutlidownjnc.,
Fairborn£)hio
CKD formerlvdisposedin on-sUe,unlinedlandfills. SurfaceWater
CKDcurrentljstoredinon-sitesilospriorto
recycling. ExcessCKD soldand shippedofF site for
an unknownpurpose.
GroundWater
efevatedpH,As, elevatedCu, Zn
Cd, Cr, F«s, Pb,
Se , ,
elevated pH,A$,
Cd>Cr,
Pb,Se
NationaiGypsum
LafargeCorporation
Alpena, Michigan
31+ hectaresjnactivedisposafcite. National
Gypsumformerlydisposedof CKD on the shoresof
Lake Huron. CKD piled 18 metersabove the lake
level. On shore there are 9-meterhighbanksthat
are activelybeingundercutby wave action. Pile
containsdrutns^juckets, airpollutioncontrolbags,
and otherdebriswhichare all erodinginto the lake.
SurfaceWater
Soils
elevatedAs, Pb
elevatedAs, Pb,
Se, Zn
Ash Grove CementWest, Inc.
MontanaCity, Montana
Currentlj(1990) disposesof CKD by landfillin^
djmwon the east side of the active quarry. Surface
run-ofrduringstormsflowsintoholdingpondsfor
sedimehtremovatbefoire dischaJtgdinto PricklyPear
Creek. Two catastrophircleasesof CKD^bearing
j$!udge$from holdingpond into creek.
SorfaceWater elevatedTDS e1evatedPb,tl
AI=aluminumAs=ai^enicCd=cadmiumCI=chlorideCr=chromiunlCu=copper^'=fluorideFe=iron4>b=:|eadJvIn =
Ni=nickelK=potassiuni$e=seleniuniNa=sodiumSO4=:sulfate|1=thalliumTDS=totaUissolvedsolids^u=zinc
-------
5-9
Site
Exhibit 5-3
Cases of Potential Damage
Waste Management Practice
Damaged Media
Constituents of Concern
(concentrations exceed MCLs)
Texas Industries
Midlothian, Texas
Holnam, Inc.
Artesia, Mississippi
Markey Machinery
Property
Seattle, Washington
60% to 80% of CKD collected in HSP
currently (1990) disposed in ori'-site landfill in
a depleted quarry area. 20% to 40% used
beneficially off site. CKD formerly disposed in
unlined waste pile.
Currently (1990) disposes of non-waste derived
CKD in water-filled quarry. Waste derived
CKD disposed of in open pile with bermed
boundaries.
Approximately 38,000 cubic meters of CKD
wa$ disposed off site OB a parcel of la»d withia
the city limits of Seattle. Site is an old truck
park located vdthia 1,200 meters of a State
fishery (DuWamish River) and has substantial
nearby population:
Surface Water
Surface Water
(quarry water,
process water
discharge into
quarry)
Surface Water
Ground Water
Soils
elevated pH, As, Cr, Pt>
\
\
elevated pH
elevated pH, Pb "'<
elevated Pb
elevated A*, Pb
As=arsenic, Cr=chromium, Pb=lead
-------
5-10
S3. DOCUMENTED GROUND AND SURFACE WATER DAMAGE CASE SUMMARIES
As described above, EPA contacted officials at local and state regulatory agencies and at
EPA Regional ofGces in all states in which cement is produced to gather information
documenting the environmental performance of waste management practices for CKD. In
addition to interviewing these officials, EPA reviewed files obtained either through the mail or
during visits to regulatory agencies. Through the above-described case studies, EPA found
documented environmental damages of either ground or surface water associated with CKD
management at the following seven facilities:
• Holnam Incorporated, Mason City, Iowa;
• Lehigh Portland Cement Company, Leeds, Alabama;
• Lehigh Portland Cement Company, Mason City, Iowa;
• Portland Cement Company, Salt Lake City, Utah; and
• Southwestern Portland Cement (Southdown, Inc.), Fairborn, Ohio.
• National Gypsum Co./Lafarge Corp., Alpena, Michigan
• Ash Grove Cement West, Inc., Montana City, Montana
EPA has also found cases of potential environmental damage at the following three
facilities:
• Texas Industries, Inc., Midlothian, Texas
• Holnam, Inc., Artesia, Mississippi
• Markey Machinery Property, Seattle,Washington
Documented damages at these seven facilities are summarized below, followed by a summary of
the three cases of potential damage in Section 53.
Cases of Documented Damage
5.2.1 Holnam Incorporated, Mason City, Iowa
The Holnam, Inc. facility (formerly Northwestern States Portland Cement Company)
occupies 97 hectares (240 acres), and is located in Cerro Gordo County, Iowa adjacent to the
northern boundary of a residential development in Mason City. The plant operates one long dry
process kiln and manufactures Types I, II, and III Portland cements and masonry cement.
The site is bordered to the west by a railroad right-of-way, to the north by the property
line of the Lehigh Portland Cement Company, to the east by Highway 65 and to the south by
streets bordering the residential areas of Mason City. Cahnus Creek crosses the northwest
portion of the property on its way to Winnebago Creek nearly 1 kilometer (0.62 miles) away. To
the east and west are rural agricultural areas.
From 1969 to 1985, plant operators landfilled CKD waste into a large inactive quarry
located on the western portion of the facility property. Known as West Quarry, the disposal site
4PortlandCementAssociation.l991. PCA CKD Survey: ResponsefromHolnam,Inc.,MasonCity,Iowa.
-------
5-11
was originally 61 hectares in area and 12 meters (m) (39 feet) deep. When disposal activities
ceased in 1985, 73 percent of available quarry volume was filled with approximately 1.8 million
metric tons (2.0 million tons) of kiln dust, and the open volume of the quarry (now known as
West Quarry Pond) was reduced to approximately 16 hectares and was filled with approximately
1.59 million kiloliters (420 million gallons) of water. An indeterminate amount of dust also
was disposed in East Quarry, located east of Highway 65. No record is available regarding
CKD disposal prior to 1969. Exhibit 5-4 provides a diagram of the Holnam site.
Exhibit 5-4
Site Diagram - Holnam Incorporated, Mason City, Iowa
^NorthwestenfitatesPortlandCementCompany, 1985. Hydrogeohgic Investigation - West Quarry Site,
Northwestern States Portland Cement Company, Mason City, Iowa. Prepared by IT Corporation. July, 1985.
Iowa Department of Natural Resources, 1990. Record of Decision for Northwestern States Portland Cement
Company Site, Mason City, Iowa. June, 1990.
"''Portland Cement Association, 1991. PCA CKD Survey: Response from Holnam, Inc., Mason City, Iowa.
-------
5-12
Holnam no longer wastes CKD at its Mason City, Iowa facility. The raw materials in the
manufacturing process have been changed so that the kiln dust can be placed back into the
product; the dust is 100 percent recycled.
There are two aquifers in the vicinity of the Holnam facility, both of which supply potable
water to people living nearby. Mason City municipal wells and the high capacity wells of both
Holnam and Lehigh Cement tap sandstones comprising the Jordan aquifer at depths greater than
370 meters. Wells also tap a shallower limestone and dolomite aquifer located within 90 meters
of the surface, which supplies the drinking and industrial needs of both facilities. The shallower
limestone and dolomite aquifer also supplies the drinking water needs of about 300 residents in a
subdivision (Winnebago Heights) located two kilometers north of the site.
Five municipal and five industrial water wells are located within 1.6 kilometers of West
Quarry Pond. The municipal wells, located southeast of West Quarry, help supply drinking water
to the Mason City public supply system, which serves over 30,000 people. Most residences in the
vicinity of Holnam draw water from the municipal water wells. Some of these residences also
have shallow private wells used for gardening and other outdoor activities. Of the five
industrial wells, two are owned by Holnam and are located on site; two are owned by Lehigh
Portland Cement Company, and are located on Lehigh property north of the Holnam facility;
and one is owned by the American Crystal Sugar Company, and located to the north within one
mile of West Quarry Pond.
In April, 1974, a change in color in the quarry water prompted Northwestern States
personnel to initiate a pH monitoring program in the West Quarry. From April 1974 to January
1976, the pH level in the water increased from 8.0 to 8.7. By April, 1976, the pH level had
increased sharply to 11.8, and reached 12.8 in late 1980. The Iowa Department of Natural
Resources attributed the increase in pH to a collapse of the natural buffering system that was
sustaining the quarry water at a near-neutral pH. A quarry dewatering program, initiated in
1987, which reduced the water level in the West Quarry Pond from 12 to 4.6 meters, succeeded
in lowering the pH level to 10.6 by 1990.9
Also, a report on the water quality of Calmus Creek prepared by the University of
Iowa in 1984, describes a blowout, or seep on the northeast side of the West Quarry. Water
from this seep, before merging with Calmus Creek, was observed to have a high pH (11.3) and
elevated levels of several constituents, including sulfate (1,700 mg/L), sodium (1,280 mg/L),
potassium (2,400 mg/L), and phenol (230 /xg/L) relative to Calmus Creek (pH: 7.7-8.0, sulfate:
32-44 mg/L, sodium: 4.7-6.6 mg/L, potassium: 2.8-5.0 mg/L, phenol: 2-4 pgfL).10 Benthic
populations of aquatic animals were reported to be non-existent downstream, with very little
spawning activity within the affected reach of Calmus Creek. Immediately downstream of the
blowout, water in Calmus Creek showed an increase in turbidity (from 20 Natural Turbidity
Units (NTUs) to 50 NTUs downstream) and elevated levels of sulfate (65 mg/L) and potassium
o
0 U.S. Department of Health and Human Services, 1991. Health Assessment for Northwestern States Portland
Cement Company, Mason City, Cerro Gordo County, Iowa. December, 1991.
9Ibid,
10Universlty of Iowa, 1984. Calmus Creek Water Quality Study. Report 85-1.
-------
5-13
(47 mg/L) relative to sampling sites upstream of the blowout. In April 1985, the State ordered
the facility to cease discharges from the seep area to Calmus Creek. At the same time, the
facility was ordered to stop disposal of CKD in the quarry and to conduct a hydrogeologic
investigation.
In 1985, a hydrologic investigation of the West Quarry site prepared for Northwestern
States Portland Cement Company showed waste kiln dust to be the original source of
contamination at the site. Reported analyses of waste kiln dust show high levels of magnesium
(4,000-5,000 mg/kg), potassium (4,400-13,000 rag/kg), and sulfur (4,100 mg/kg). A 10 percent
slurry mixture of water and kiln dust produced a solution with a high pH (11.8-12.4).
The report concluded that water from the West Quarry was also a source of
contamination. Samples were characterized by a high pH (>12.0), as well as high concentrations
of Total Dissolved Solids (1,800-13,000 mg/L), potassium (430-2,300 mg/L), sodium (48-250
mg/L), chloride (36-130 mg/L), and sulfate 320-3,030 mg/L). Concentrations of chromium (0.06-
033 mg/L) also exceeded the Federal primary drinking water standard (0.015
The investigation noted that water in the West Quarry is hydrogeologically connected
with the surrounding ground water and, as a result, there is potential for migration of the
contaminants in the ground water. Water sampled in wells placed between the West Quarry and
Calmus Creek showed elevated pH levels (103-13.1) that decreased with depth. Also, levels of
TDS (6,700-30,000 mg/L), aluminum (1.5-48 mg/L), potassium (1,100-3,900 mg/L), sodium (170-
620 mg/L), chloride (71470 mg/L), and sulfate (160-2,500 mg/L) were generally similar to levels
observed in water in the West Quarry. Levels of these constituents in water sampled from
background wells were considerably lower (pH: 6.8-7.4, TDS: 900-1,800 mg/L, aluminum: 1.5-4.5
mg/L, potassium: 2.4-3.0 mg/L, sodium: 21-22 mg/L, chloride: 26-65 mg/L, sulfate: 76-380
mg/L)/5
The facility installed an acid-neutralization system in June 1987, adjacent to Calmus
Creek in the northwestern portion of the filled West Quarry. In addition to treating the seep
water, the system was used to dewater of the West Quarry Pond. The treated water is
discharged to Calmus Creek in accordance with a NPDES permit issued by the Iowa Department
of Natural Resources. These actions taken by the facility have eliminated untreated discharges
from the West Quarry to Calmus Creek. However, discharge of water from the acid-
neutralization facility still poses potential water quality problems in Calmus Creek due to
elevated levels of total dissolved solids and phenols/6
"ibid.
12Ibid.
Northwestern Stales Portland Cement Company, 1985. Hydrologic Investigation of the West Quarry Site,
Northwestern States Portland Cement Co., Mason City, Iowa. Prepared by IT Corporation. Jufy.
"ibid.
"ibid.
Northwestern Slates Portland Cement Company, 1989. Remedial Investigation/Feasibility Study on the West
Quarry, Mason City, Iowa. Prepared by Layne Geosciences, Inc. Project No. 61.1099.
-------
5-14
In May 1985, the facility installed two ground-water extraction wells in the vicinity of the
seeps to control the discharge to Calmus Creek. The water that was collected by the wells was
circulated back into West Quarry Pond.
' The site currently has a series of 16 monitoring wells. Analytical results of ground water
discharging to Calmus Creek from sampling conducted in 1988, as part of EPA's Field
Investigation Team (FIT) investigation, revealed a pH of 13.1 and sulfate and phenols
concentrations of 1,500 mg/L, and 0.16 mg/L, respectively. Both pH and sulfate levels exceed .
national secondary drinking water standards.
On August 30, 1990, the Holnam site was listed on the National Priorities List. In its
June 1990 Superfund Record of Decision, the Iowa Department of Natural Resources
summarized the major concerns at the site as contaminated surface water and ground water.
The primary problems have been sharp increases in pH and mineral deposition in on-site ground
water and nearby surface water as a result of contact with waste CKD in the West Quarry.
In the June 1990 Record of Decision for this site, the Iowa Department of Natural
Resources determined that the selected remedy for the site would include the following actions
to control and remediate existing ground-water contamination and to reduce the potential for
future contamination of ground water and surface water:
• Dewatering of the West Quarry (completed in September 1989);
• Construction of a permanent drain system in the dewatered West Quarry to
collect precipitation run-off and ground-water inflow to the quarry;
• Placement of an engineered clay cap over the area of the West Quarry filled with
CKD to minimize infiltration through the kiln dust;
• Installation of bedrock extraction wells to collect contaminated ground water
beneath the West Quarry, prevent migration of contaminated ground water from
the site, and maintain ground-water levels below the CKD;
• Installation of kiln dust dewatering wells, if necessary, and
• Treatment of contaminated waters to meet Iowa NPDES discharge permit limits
for discharge to Calmus Creek.
The initial remedial actions taken at this facility, dewatering of the West Quarry Pond,
and neutralization of pond water, have proved to have some positive impact. However,
according to the Superfund Record of Decision, additional remedial actions are still necessary to
reduce the potential risk of future contamination. These include construction of a permanent
drain, placement of a clay cap over the quarry, and installation of bedrock extraction wells.
17lbuL
JQ
Iowa Department of Natural Resources, 1990. Record of Decision for the Northwestern States Portland
Cement Company Site, Mason City, Iowa. June, 1990.
-------
5-15
The disposal of CKD in unlined, abandoned quarries is a common waste management
practice utilized at cement plants. Damages at this site resulting from this disposal practice
consist of impairment of Calmus Creek from the overland flow of high pH water from West
Quarry Pond, and ground-water discharges to the creek. These discharges have elevated the pH
of the stream above the State's water quality standard. This damage has been documented in
several studies, the most recent being conducted in 1989. On-site ground-water contamination
also has been identified at this site. The contaminants of concern include pH, total dissolved
solids, potassium, sulfate, and phenols. These constituents have been observed at levels that
exceed primary and secondary drinking water standards.
5.2.2 Lehigh Portland Cement Company, Leeds, Alabama
Lehigh Portland Cement Company's Leeds plant is located in Jefferson County,
approximately 24 kilometers (km) (15 miles) east of Birmingham, Alabama. The plant has
operated a single dry-process kiln at the site since 1976 and manufactures Types I, II, and III
Portland cement and masonry cement. In 1990, the facility utilized coal for 96 percent of its fuel
needs, and natural gas for the remaining four percent. The plant currently recycles all of its
CKD; there is no land disposal of CKD either on or off site.
The entire plant encompasses 270 hectares (668 acres) and is located within a 100-year
floodplain with karst topography and faulted bedrock. The population within a 2 km radius of
the plant was 7,000 in 1990, and the nearest residence is 91 meters (m) (300 feet) to the
northwest of the plant's boundary. No public or private drinking water wells exist within two
kilometers of the plant.20
Prior to 1978, the previous owners of the facility, the Atlas Cement Company and U.S.
Steel, disposed of an undetermined portion of its waste CKD in two on-site piles. These piles
lie within 150 meters of the plant's limestone quarry, which is located to the south of the plant's
kiln. Neither the State of Alabama nor Lehigh personnel know the total amount of CKD
disposed in the piles, or if any material is co-disposed with the dust. One of these piles is
currently seeded with grass.
Both waste CKD piles drain into a sedimentation pond, the water from which is pumped
uphill and dispersed as a spray in a grove of pine trees. Run-off from the spray flows downslope
away from Moores Creek, the natural drainage channel located south of Lehigh's limestone/clay
quarry. Moores Creek receives stormwater run-off from the plant property through five NPDES
outfalls.22 The site layout is shown in Exhibit 5-5.
During the 1980s, the Alabama Department of Environmental Management's (ADEM)
Water Division observed two incidents of elevated pH in Moores Creek caused by storm-water
^Portland Cement Association, 1991. PCA CKD Survey: Response from Lehigh Portland Cement Company,
Leeds, Alabama.
20Ibid.
Lehigh Portland Cement Company, 1993. Personal communication with Charlie KJotz, Safety Training, and
Environmental Manager, Leeds, Alabama facility.
igh Portland Cement Company, Leeds, Alabama, 1993. Personal communication with Charlie Klotz, op.cit.
-------
5-16
run-off from both dust piles and the plant proper. In April 1984, ADEM issued a NOV to
Lehigh Portland Cement for violations of the Water Division's regulations. In February 1987,
ADEM issued a Notice of
Alabama Department of Environmental Management, 1984. Letter from H.H. Beiro, Pollution Control
Specialist, to M.F. McCarthy, Lehigh Portland Cement Co., Leeds, Alabama. April 19, 1984.
-------
5-17
Exhibit 5-5
Site Diagram - Lehigh Portland Cement Company, Leeds, Alabama
Noncompliance for exceedances of limits for pH and total suspended solids (TSS) specified in
the facility's NPDES permit during the fourth quarter of 1986.2*
In the April 1984 NOV to Lehigh, ADEM noted that surface run-off from the facility's
waste CKD stockpiles had elevated the pH of Moore's Creek, the receiving stream, from a level
of 6.9 upstream of the plant to a level of 9.5 downstream of the plant, constituting a violation of
the State's Water Quality Standard for pH in the stream of 8.5. In response, Lehigh Portland
collected seven samples in May 1984, at the stream's "low flow" from various points on Moore's
Creek (both upstream and downstream of the plant), and at the base of one of the dust piles.
Samples collected above and below Outfall #003 (located at the southern end of the
limestone/clay quarry) yielded pH levels of 9.12 and 8.84, respectively. Lehigh Portland stated in
Alabama Department of Environmental Management, 1987. Notice of Noncompliance from S.Jenkins to M.
McCarthy, -Lehigh Portland Cement Company. February 3, 1987.
25Lehigh Portland Cement Company, 1984. Letter fiom R. Gebhardt to H. Beiro, ADEM. May 30, 1984.
-------
5-18
a letter to ADEM accompanying the sampling results, that "areas of the old, consolidated kiln
dust piles that look like water courses are not; no water runs down them even during very heavy
rains"; however, the facility offered no explanation for the elevated pH levels in the stream.
Lehigh Portland initiated several ADEM-approved pollution abatement measures^ in an
attempt to control run-off into Moores Creek. In 1986, Lehigh Portland installed diversion
ditches and an unlined sedimentation pond to the south of the clay pit and the dust piles, to
allow settling of CKD in run-off from the piles prior to its discharge to Moores Creek through
Outfall #006. Lehigh Portland also seeded the dust piles with grass in the summer and fall
of 1986 in an effort to control run-off.
Lehigh Portland's Leeds facility was again cited by ADEM in February, 1987, for
violations of its NPDES permit.29 The violations consisted of three exceedances of the pH limit
of 9.0 for Outfall 006 during the fourth quarter of 1986 (measured pH: 9.2-10.0), and an
exceedance of the daily average total suspended solids limit for the same outfall (25 mg/L)
measuring 112.5 mg/L. During the first quarter of 1987, the daily average TSS for Outfall 006
exceeded the permit limit of 25 mg/L for each month (January: 88 mg/L, February: 58 mg/L,
March: 51.5 mgfL).30
By May 1987, after determining that vegetation alone would not sufficiently control the
run-off, Lehigh Portland sealed the discharge pipe from the sedimentation pond to Outfall 006 to
prevent further discharge to Moore's Creek. The plant also installed a pump and spray system
to recirculate the water from the sedimentation pond away from Moores Creek. An emergency
spillway to Moore's Creek was retained in the event of emergency overflow. According to the
ADEM's Water Division, no additional violations or noncompliance with permit conditions have
been observed, as determined through the Division's review of the plant's Discharge Monitoring
Reports (the Division does not regularly inspect or monitor discharges at this facility). In
addition, no ground-water contamination below the sedimentation pond has been observed. '
Alabama Department of Environmental Management, 1985. Letter from Kirk S. Kreamer, ADEM, to A.P.
Mahatekar, AmTech Services, Inc., August 21, 1985.
2°Amtech Services, Inc., 1986. Letter from M. Holder, Professional Engineer, AmTech Services, Inc. to P. Prysey,
ADEM. Jufy 24, 1986.
Alabama Department of Environmental Management, 1987. Notice of Noncompliance from S.Jenkins to M.F.
McCarthy, Lehigh Portland Cement Co., Leeds, Alabama. February 3, 1987.
^Alabama Department of Environmental Management, 1987. DMR Violation Report, 1st Qtr. 1987, for Lehigh
Portland Cement Co., Leeds, Alabama.
Lehigh Portland Cement Company, 1987. Letter from L. Copple to S. Jenkins, Alabama Department of
Environmental Management. February 20, 1987.
Alabama Department of Environmental Management, 1992. Personal communication with S. Jenkins. January,
1992.
Alabama Department of Environmental Management, 1992. Personal communication with C. McRoy. October,
1992.
-------
5-19
Damage at this site consists of the impairment of the water quality of Moore's Creek
through the discharge of run-off from inactive CKD disposal piles. The discharge elevated the
pH of the stream to levels exceeding the State's designated water quality standard for the stream;
the discharge also exceeded the discharge limit for pH specified in the facility's NPDES permit.
This was documented on two separate occasions during the 1980s, a period in which the CKD
waste piles were inactive. The initial remedial action taken by Lehigh Portland, vegetating the
piles and installing a sedimentation pond to extract CKD from the discharge, proved ineffective
as demonstrated by the noncompliance with the NPDES-permitted discharge limits. The final
remedial action, eliminating discharge through the outfall, has been effective to date. "
5.23 Lehigh Portland Cement Company, Mason City, Iowa
The Lehigh Portland Cement Company (LPCQ site, in operation since 1911, is located
at 700 25th Street Northwest, on the north side of Mason City, Cerro Gordo County, Iowa. The
facility operates one kiln, and manufactures Types I and III Portland cement.
The site covers approximately 61 hectares (150 acres) and is bordered on the south by
Calmus Creek (a tributary of the Winnebago River), and on the east by U.S. Highway 65. The
facility is located in an urban area and a small residential neighborhood is located approximately
2.4 kilometers (km) (1.5 miles) to the north. The Lime Creek Nature Center (LCNC) is
approximately 1.6 km northeast of the site. The plant is located within the 100-year floodplain.
The Northwestern States Portland Cement Company site (now owned by Holnam, Inc.) is
immediately south of the Lehigh site. Calmus Creek flows between these two sites to the
Winnebago River, which is located approximately 450 meters (m) (1,476 feet) north and east of
the two facilities:36
The LCNC, although separate from the plant area, has been the site of past disposal of
CKD by the Lehigh Portland Cement Company. The LCNC covers 247 hectares and is owned
by the County of Cerro Gordo and operated as an outdoor recreation area. It was opened to
the public in May 1984. Portions of the current LCNC were formerly owned by Lehigh Portland
Cement Company. The property was transferred to Cerro Gordo County in 1979.
In 1990, Lehigh Portland, utilizing normal fossil fuels (85 percent coal, 8 percent natural
gas, 7 percent coke), generated approximately 171,984 metric tons (189,577 tons) of CKD, of
which 162,789 metric tons (95 percent) was recycled and used as raw material in the kiln. An
estimated 8,620 metric tons of wasted CKD were landfilled in a clay quarry. This landfill first
began receiving CKD waste in 1986.38
EPA has also promulgated rules that will require control of surface run-off; thus, if implemented, these
regulations should prevent similar types of violations in the future.
U.S. Environmental Protection Agency, 1991. Record of Decision: Lehigh Portland Cement, Mason City, Iowa.
Prepared by the Office of Emergency and Remedial Response. June, 1991.
36Ibid.
37Ibid.
oo
Portland Cement Association, 1991. PCA CKD Survey: Response from Lehigh Portland Cement Company,
Mason City, Iowa.
-------
5-20
Prior to disposal in the current landfill, Lehigh Portland deposited CKD in locations
throughout facility property, including an exhausted quarry north of the plant (now known as the
CKD Reclamation Area), as well as other on-site inactive quarry areas (now partially re-filled
with water) located northeast of the plant proper, including Area "C" Pond, Arch Pond, Blue
Waters Pond, and West Quarry Pond. Prior to 1979, when Lehigh Portland owned the LCNC
property, plant operators also disposed of waste CKD in an abandoned quarry on the west side
of the property (now water-GUed and known as Quarry Pond), and in a 16 hectare site located
along the west bank of the Winnebago River, known as the "Badlands". The actual amount of
CKD disposed of on site may exceed 900,000 metric tons. >4° CKD disposal areas and plant
operations are shown in Exhibits 5-6 (CPCC site) and 5-7 (LCNC site).
Exhibit 5-6
Site Diagram - Lehigh Portland Cement Company, Mason City, Iowa
•* U.S. Environmental Protection Agency, 1988. Final Report Site Investigation: Lehigh Portland Cement Company.
Mason City, Iowa. Prepared by Ecology and Environment Field Investigation Team for Region Vll. MarcJi, 1988.
40U.S EPA, 1991. Record of Decision: Lehigh Portland Cement, LA. op.cit.
-------
5-21
Exhibit 5-7
Site Diagram - Lime Creek Nature Center (Lehigh Portland Cement), Mason City, Iowa
The LPCC site was placed on the National Priorities List (NPL) on August 30, 1990. In
litigation, Lehigh identified a number of concerns regarding the hazard ranking score. After
reviewing the issues regarding the calculation of the score on the hazard ranking system, the
Agency decided not to contest Lehigh's challenge to the listing decision. The listing was vacated
by mutual consent in October 1992. Removal of that site from the NPL does not affect clean-up
at the site.
There are two aquifers in the vicinity of the Lehigh Portland facility, both of which supply
potable water to people living nearby. Wells serving the population of Mason City tap a
sandstone aquifer greater than 370 meters in depth. Lehigh Portland, as well as the adjacent
Holnam facility, utilize a shallower limestone and dolomite aquifer located within 90 meters of
the surface. This aquifer supplies the drinking and industrial needs of both facilities. In
addition, it supplies the drinking water needs of about 300 residents in a subdivision located
north of the site. In 1987, EPA Field Investigation Team (FIT) personnel at Lehigh Portland
observed shallow (1-3 meter depth) static water levels in pre-existing on-site ground-water
-------
5-22
monitoring wells.** Quarry floors are below this depth, hence any CKD waste disposed in them
was likely deposited directly into the shallow (1-90 meter) aquifer.
Problems with the site were first identified in 1981 during a routine hydrochemical test of
Blue Waters Pond, one of four water-Glled abandoned quarries on Lehigh property. The results
of the test indicated that the pond water was alkaline (pH: 10.6) and exceeded the State standard
(pH: 9.0) for discharge into warm water streams. At this time, Lehigh had installed an
overflow control structure at the southeastern corner of Blue Waters Pond. The flow control
structure allowed water from the pond to be discharged directly to Calmus Creek to eliminate
possible back-flooding of critical equipment.**
Lehigh hired an independent consulting firm to determine the source of the high pH
waters. Twenty-eight surface water samples from various locations were collected and analyzed.
The results of the report identified three potential sources, of which Arch Pond contributed the
most significant quantities of high pH water to Blue Waters Pond. As a result, the facility
transferred the water from Blue Waters Pond to the Area "C" Pond and retained the water
behind two earthen dikes. These dikes have since failed due to high rainfall.
In August 1984, the State of Iowa conducted a Comprehensive Work/Quality Assurance
project on Calmus Creek, which is located approximately 300 meters south and downgradient of
Blue Waters Pond. This investigation found that surface water contamination was directly
related to the Lehigh facility as a result of discharges from the pond into the creek via a tile
drain outlet southeast of the plant. The discharged water had a pH of 11.4, and total dissolved
solids of 4,700 mg/L, including 2,000 mg/L potassium and 829 mg/L sulfates. The investigation
also determined that the Arch Pond immediately west of the Blue Waters Pond could also
contribute an unknown quantity of run-off from the western half of the plant to Calmus
Creek.45
The study concluded that the biological quality of Calmus Creek had deteriorated as a
result of effluent discharges from the Lehigh plant and the Holnam facility site located to the
south. The study stated that because of the deterioration of the chemical balance in Calmus
Creek and the quarry ponds, the number and variety of fish and benthic organisms were found to
be substantially reduced downstream of the tile drain. As a result of this study, Lehigh was
required to eliminate the discharge into Calmus Creek.46
U.S. Environmental Protection Agency, 1988. Final Report, Site Investigation, Lehigh Portland Cement, Mason
City, Iowa, op.cit.
^ Portland Cement, 1989. Site Investigation Protocol for the Lehigh Portland Cement Company Plant, Mason
City, Iowa. Prepared for Lehigh Portland by Layne GeoSciences, Inc. October, 1989.
"ibid. .
Iowa Department of Water, Air and Waste Management, 1984. Calmus Creek Water Quality Study, May-August,
1984.
46U.S. EPA, 1991. Record of Decision: Lehigh Portland Cement, IA. op.cit.
-------
5-23
Subsequently, at some unknown time, dikes were constructed to separate Arch Pond, the
Area "C" Pond, and Blue Waters Pond, and an aboveground piping system was installed to pump
water from Blue Waters Pond into the Area "C" Pond. Lehigh also constructed a lined ditch to
channel the surface water run-off collected by the drain system from the adjacent highway back
into the tile drain located southeast of Blue Waters Pond. The long-term goal of this effort was
to eliminate Blue Waters Pond by backfilling and regrading the area.
An EPA site investigation conducted in April 1987 confirmed that the on-site quarry
ponds and shallow ground-water table are contaminated locally and that contaminants have the
potential to migrate off site to Calmus Creek and the Winnebago River. Seepage has occurred
from the quarry ponds and is contaminating the ground water. The FIT investigation concluded
that contamination could occur during high intensity rainfall events, leading to ground-water
infiltration and flooding, and that the potential exists for human and biological exposures to the
hazards present at the site.4
Surface water samples taken from each of the on-site quarry ponds (Blue Waters, Arch
and Area "C" ponds), which contain disposed CKD, showed elevated levels of pH, metals,
potassium, sodium, and sulfate relative to samples taken from Calmus Creek and the Winnebago
River. Levels of total aluminum (0.82-1.8 mg/L), total sodium (28.0-180.0 mg/L), and total
potassium (120.0-290.0 mg/L) in samples from the ponds were nearly ten times greater than
levels of the same compounds found in samples taken from Calmus Creek (aluminum: 0.15 mg/L,
sodium: 7.2 mg/L, potassium: undetected) and the Winnebago River (aluminum: 0.12 mg/L,
sodium: 0.92 mg/L, potassium: 0.77 mg/L). Sulfate concentrations in the same ponds ranged from
270 mg/L to 1,160 mg/L and were as much as 34 times background levels in the creek and river
(34.0-47.0 mg/L). Except for West Quarry Pond (pH: 8.52), which showed a pH close to levels
found in Calmus Creek (pH: 7.84), and the Winnebago River (pH: 8.49), values of pH in pond
waters were uniformly high (pH: 11.19-11.23). Arsenic was detected in waters from Arch Pond
(0.051 mg/L) at about the same level as the Federal drinking water standard (0.05 mg/L), while
lead was detected in the duplicate sample from Blue Waters Pond (0.038 mg/L) at a level 2.5
times the Federal drinking water standard (0.015 mg/L).48
The sample from the tile drain outlet into Calmus Creek, which drains Blue Waters
Pond, had a pH value close to background (7.90), and had no detectable levels of arsenic or lead.
Levels of potassium (19.0 mg/L), sodium (11.0 mg/L), and sulfate (63.5 mg/L), however, were
elevated above background levels in Calmus Creek and the Winnebago River.49
Three pre-existing water wells, which are used to monitor ground-water flow and
chemistry, were sampled during the EPA site investigation. These wells are located between the
Area "C" Pond and Blue Waters Pond (MW #2), between the Arch Pond and Blue Waters Pond
(MW #3), and hydrologically downgradient from Blue Waters Pond, between the pond and the
Winnebago River at the eastern facility boundary (MW-#1). All three wells are less than 20
meters deep (MW #1: 19.1 meters, MW #2: 12.8 meters, MW #3: 9.1 meters) and penetrate the
shallow ground-water table (static water levels: 1.2-2.7 meters below ground level). Samples
47U.S. EPA, 198S. Final Report, Site Investigation, Lehigh Portland Cement, Iowa, op.cit.
48Ibid
49Ibid.
-------
5-24
collected from wells MW #2 and MW #3 had elevated levels of pH (11.06, 12.04 respectively),
above the national secondary drinking water standard (9.5). In addition, arsenic was present in
well MW #3 at a concentration (dissolved 0.072 mg/L) 1.4 times the Federal drinking water
standard (0.05 mg/L). Zinc was found in MW #3 at levels five to six times background
concentration, but below Federal drinking water standards. MW #1 had a pH close to
background levels (pH: 7.9), however levels of calcium (130 mg/L) and potassium (1.9 mg/L)
were elevated relative to upgradient wells MW #2 and MW #3. One deep on-site Lehigh
drinking water well was also sampled. This well did not exhibit concentrations of constituents
above primary or secondary MCLs.SO
In 1989, Lehigh hired Layne GeoSciences to perform the Remedial
Investigation/Feasibility Study for the site. Nine monitoring wells were installed on the site, one
being a nested well. The first of four sampling rounds was conducted in June 1990. Elevated
pH values, total dissolved solids, and similar contaminants as prior studies were found in the
ground water and surface water. The pH levels ranged from background to as high as 11.43 in
one well. Total dissolved solids in this well were also as high as 7,000 mg/L. The pH levels in
the on-site ponds were higher than previously detected (13.0 in Arch Pond), with TDS levels at
11,000 mg/L.51
In the fall of 1990, it was also determined by the Iowa Department of Natural Resources
that the LCNC needed to be investigated for the same contaminants as the Lehigh site. As with
the Lehigh site, the primary concerns in the LCNC area include elevated pH and TDS levels.
The CKD samples that were collected showed high values for extractable and final pH (11 -
12.7). Elevated pH levels were detected in the Quarry Pond (9.5) and one monitoring well (#14,
pH: 10.4). This high pH was not found in the LCNC water well, which is assumed to be
downgradient of the CKD deposits.
There are two specific contamination concerns at the LCNC site:
• Elevated ground-water pH beneath the Badlands area; and
• Elevated ground-water pH and TDS levels in the Quarry Pond.
Local ground water and surface water have been affected at this site by high pH levels,
an increase in total dissolved solids content, and elevated concentrations of potassium, sulfate,
and sodium. These constituents have been monitored at levels that exceed national drinking
water standards. In addition, ground-water contamination is evident beneath the Lime Creek
Nature Center, a past off-site disposal area for CKD. These damages have been documented in
several studies, and the situation has not changed significantly since 1989.
5.2.4 Portland Cement Company, Salt Lake City, Utah
From 1965 to 1983, the Portland Cement Company of Utah (PCU) disposed of CKD at
five sites in and around Salt Lake City, Utah. The largest of these sites, designated as Portland
Cement Co. site numbers two and three (Kiln Dust #2 & #3), is estimated to be 29 hectares (71
S°Ibid.
Lehigh Portland Cement Company, 1991. Remedial Investigation/Feasibility Study for the Lehigh Portland Cement
Company Plant, Mason City, Iowa. Prepared by Layne GeoSciences, Inc. April, 1991.
-------
5-25
acres) in area and is listed on the NPL. Lone Star Industries purchased PCU in 1979, and has
been identified by the EPA and the Utah Department of Environmental Quality (UDEQ) as one
of several Potentially Responsible Parties.^
The Kiln Dust #2 & #3 site is located in Salt Lake City approximately 2.5 kilometers
(km) (1.6 miles) southeast of the Utah International Airport, and 1.6 kilometers south of
Interstate 80. The site consists of three adjacent CKD disposal areas, site #2, site #3, and the
West Site. The property is bounded on the north and east by city streets, and on the south and
west by the Jordan River Surplus Canal."54
Land use in the vicinity of the site is characterized by mixed residential and commercial
development. The immediate area surrounding the site is zoned for commercial and light
industrial use. East of the site are residential areas. Vacant areas or agricultural lands are
common in the surrounding area. The EPA, in the 1990 Record of Decision, estimates between
6,000 and 12,000 people live within one mile of the site. Exhibit 5-8 provides a diagram of the
site.
Between 1965 and December 1983, approximately 378,700 cubic meters (m3) (495,718
cubic yards) of CKD was disposed in site #2, site #3, and the West Site, by PCU and/or Lone
Star Industries. The waste CKD was disposed of as a slurry on site 2, while on site 3 it was
disposed of in dry form. Within the boundaries of site #2, site #3 and the West Site, CKD is
present in thicknesses ranging from one meter to more than two meters (m) (3.3-6.6 feet).-56
Co-disposed with the CKD is 327 metric tons (360 tons) of chromium brick. At the West Site,
CKD is mixed in discontinuous layers with an indeterminate amount of industrial debris,
including rubble, soils, scrap iron, concrete slabs, asphalt, common bricks, alumina kiln bricks,
and common trash.
Two drainage features pass through or are adjacent to the site. A drainage ditch, known
as the City Drain, flows through the site, carrying urban storm run-off in a northwesterly
direction. The Jordan River Surplus Canal carries water from the Jordan River northwestward
to the Great Salt Lake. The City Drain is part of an urban storm sewer system and the water it
U.S. Environmental Protection Agency, Region V1I1, and Utah Department of Health and Emironmental
Quality, 1990. Declaration for the Record of Decision, Portland Cement Company (Kiln Dust #2 & #3), Operable Unit
1, Salt Lake City, Utah. July, 1990.
53United Stales Bankruptcy Court, Southern District of New York, 1991. Proof of Claim of the Stale of Utali
Department of Environmental Quality, Case No. 90-B-21277, in reference to New York Trap Rock Corporation, Lone
Star Industries, Inc. et.al., Debtors.
54U.S. EPA and UDEQ, 1990. Declaration for the Record of Decision, Portland Cement Company, Salt Lake
City Utah, op.cit.
57ibid.
-------
5-26
carries is protected by the State of Utah. The water in the Surplus Canal is protected by the
State of Utah for nongame fish, water-oriented wildlife, and agricultural uses.
"ibid.
-------
5-27
Exhibit 5-8
Site Diagram - Portland Cement Company, Salt Lake City, Utah
Near-surface ground water underneath the site is characterized by a shallow, unconGned
ground-water body, and local, perched water bodies. Both of these water bodies are above a
deeper confined aquifer, which is the principal source of ground water in Salt Lake Valley.
Although the local gradient of the shallow aquifer is generally to the northeast, toward the
Jordan River, it is strongly influenced by the Jordan River Surplus Canal, the City Drain, and an
underground sewer drain along the west side of CKD disposal areas #2 and #3.
The 1990 EPA Record of Decision summarizes the results of several studies of the Kiln
Dust #2 and #3 CKD disposal site. Disposed kiln dust contains the elements arsenic (3.0-27
mg/kg), cadmium (2.1-5.5 mg/kg), chromium (8.7-28 mg/kg), lead (90-1,274 mg/kg), and
molybdenum (8.7-51.7). Of these, concentrations of molybdenum and lead are generally above
those found in typical soils of the western United States (molybdenum: 3-7 mg/kg, lead: 10-700
59U.S. EPA and UDEQ, 1990. Declaration for the Record of Decision, Portland Cement Company, Salt Lake
City, Utah, op.cit.
-------
5-28
me/kg). Concentrations of metals showed little variation among site #2, site#3, and the West
Site.®
Analytical results of ponded water at the site are also summarized in the 1990 EPA
Record of Decision. Ponded water was observed in several pools along the edges of disposal
sites #2 and #3, and within the boundaries of site #3. Based on samples collected during
regular observations at the site since 1984, the reported results show elevated levels of arsenic
(2.53 mg/L maximum), chromium (3.00 mg/L maximum), and lead (0.37 mg/L maximum) above
Federal drinking water limits, reaching as high 50 times the standard for arsenic (0.05 mg/L), 30
times the standard for hexavalent chromium (0.1 mg/L), and nearly 25 times the standard for
lead (0.015 mg/L). According to the Record of Decision, state officials observed ponded
water migrating off site through a ditch that flows west into the City Drain.
The Record of Decision for the site concluded that the soil, ground water, and surface
water are contaminated with CKD constituents both on and off site. A contaminant plume is
present in the shallow ground water (approximately 2 meters below the surface) beneath the site
and off site. The plume is highly alkaline (pH: 12.6 maximum) and contains elevated
concentrations of arsenic (11.4 mg/L maximum), lead (0.45 mg/L maximum), chromium, and
other constituents including cadmium (6.04 mg/L maximum), fluoride (123 mg/L maximum),
sulfate (15,500 mg/L maximum), and total dissolved solids (90,000 mg/L maximum). The plume
has been detected immediately north of the site near a sewer alignment, and flows north across
the site. Ground-water sampling results from the Remedial Investigation indicate exceedances
of the primary drinking water standards for pH, arsenic (2.3 X MCL), cadmium (4 X State
MCL), chromium, and lead (30 X MCL). Remediation of the ground water cannot begin until
the sources of contamination are controlled or removed.
Surface water samples collected from the City Drain, which flows through the site,
indicate exceedances of the primary drinking water standards for pH and arsenic.
Fugitive dust emissions also have been observed by state officials during high wind events,
but apparently no NOVs have been issued. Modeling results of fugitive air emissions show
airborne particulatc concentrations in excess of the EPA 24-hour Significant Impact Limit of
5/ig/m , for an area extending 3.5 kilometers north of disposal site #3.
In a 1990 Declaration for the Record of Decision, the State of Utah indicated, and EPA
concurred, that excavation and off-site disposal of the CKD was their preferred alternative for
remediation of the site. Remediation will begin approximately 18 months after completion of the
62U.S. EPA and UDEQ, 1990. Declaration for the Record of Decision, op.cit.
<%;<£
<*lKd.
65Ibid.
-------
5-29
remedial design. The CKD will be removed to an off-site, state-approved, noncommercial,
double-lined, industrial landfill in Salt Lake City, yet to be constructed. The bricks and soil will
be treated on site. EPA and the State of Utah have yet to determine the method of treatment.
Further ground-water monitoring will be conducted during the Remedial Action to determine
whether the contamination is lessening or if the ground water has to be treated
53.5 Southwestern Portland Cement (Southdown, Inc.), Fairborn, Ohio
The Southdown, Inc. facility (formerly Southwestern States Portland Cement Company)
site is located on approximately 1,620 hectares (4,000 acres), 2 kilometers (km) (1.2 miles) east
of the City of Fairborn in Bath Township, Greene County, Ohio. The facility has operated one
dry-process kiln since at least 1930. Prior to Southwestern Portland's purchase of the property in
1924, the site was owned by Universal Atlas Cement Company. During 1990, the principal
commercial products manufactured at the facility included Types I, LA, II, and III Portland
cement, masonry cements, and expansive cements. A limestone quarry currently operated by
the facility is located north of the site.
In 1990, the facility fueled its kiln with pulverized coal, waste tires, liquid hazardous
waste-derived fuel, and fuel oil (for start-up only). Along with fossil fuels (coal: 68,600 metric
tons (75,618 tons), oil: 504 kiloliters (133,089 gallons)), an estimated 4,170 metric tons of tires
and 10,230 kiloliters of liquid hazardous waste were burned by Southwestern Portland in its
cement manufacturing process.
Facility property boundaries are adjacent to Mad River and the Beaver Creek watershed.
A portion of the site (acreage unknown) is located in the 100-year floodplain. Wetlands that
drain into Beaver Creek have been identified adjacent to the western property boundary. South
and west of facility property are glacial deposits of unconsolidated gravel, sand, and clay that
contain the aquifers that supply drinking water to the city of Fairborn. As of December 1991,
approximately 35 people resided within the facility boundary, and an additional 30,000 residents
lived within one mile of the plant. Both public and private drinking water wells are located
within one mile of the facility boundary.
An estimated 707,800 metric tons of CKD waste were landfilled in quarries owned by
Southwestern Portland from 1924 through 1978.71 Two tenths of one percent of all disposed
material is chromic oxide brick, which was co-disposed along with CKD by Southwestern
Portland from 1965 to 1978. CKD disposal occurred at 10 landfills dispersed .within the facility
67U.S. EPA and UDEQ, 1990. Declaration for the Record of Decision, op.cil.
j\ft
Portland Cement Association, 1991. PCA CKD Survey: Response from the Southwestern Portland Cement
Company, Fairbom, Ohio.
7°Ibid,
'Southdown, Inc., 1991. Site Assessment of Southwestern Portland Cement Properties Near Fairbom, Ohio: Phase 1,
Preliminary Investigations, Part 1. Prepared by Panterra Corporation. March 11, 1991.
-------
5-30
property boundary. Universal Atlas Cement also may have disposed of CKD prior to 1924 into
Landfill #1. The landfills are unlined and do not have leachate collection systems/Plant
operations and CKD disposal areas are shown in Exhibit 5-9.
Since the facility ceased its landfilling operations in 1980, CKD has been managed by
temporarily storing the waste in five cement storage silos. A significant portion of the CKD at
this facility is also recycled and used as raw material in the kiln.
The Ohio Environmental Protection Agency (OEPA) is concerned about the potential of
contaminant releases at Southwestern Portland from Landfill #1 and Landfill #6. Landfill #1
covers 73 hectares and contains an estimated 11 million cubic meters (14.4 million cubic yards)
of CKD-bearing fill. It is adjacent to Mud Run, a tributary to the Mad River which is
classified as a state resource water (recreational fishery). Landfill #6, an 11 hectare site,
contains an estimated 920,000 cubic meters of CKD, co-disposed with kiln brick, plant and
domestic trash, clean fill, and cover soil. It is adjacent to 21 hectares of wetlands and overlies
buried sand and gravel deposits that contain aquifers tapped by public water supply wells.
Landfill #1 is the closest (1.6 kilometers) of all ten disposal sites to Fairborn's North Well Field.
Landfill #6 is within 2.8 kilometers of four public water supply wells serving the needs of the
38,000 residents of the City of Fairborn.77
Contaminant releases have been observed in surface and ground waters associated with
Landfill #6. Exhibit 5-10 summarizes the results of several sampling efforts that have been
completed for this site. Surface water samples collected by OEPA (unpublished data) from seeps
and streams around the toe of the landfill during March 1993 had elevated levels of arsenic (1 to
3 times OEPA standard), iron (8 to 31 times OEPA standard), and selenium (1 to 3 times above
OEPA standard) above OEPA limits for drinking water. Levels of lead were at, or slightly below
the Federal drinking water standard (0.015 mg/L). Ground-water samples collected at the same
time near the seeps had elevated levels of arsenic (24 times OEPA standard), iron (31 times
OEPA standard), and selenium (1.8 times OEPA standard) above OEPA drinking water limits.
The surface water samples had very alkaline pH levels, reaching as high as 13.6.
Ohio Etnironmeiital Protection Agency, 1992. Personal communication with M. Leliar. January, 1992.
73Portland Cement Association, 1991. PCA CKD Survey, op.cit.
Ohio Environmental Protection Agency, 1993. Personal communication with Thomas ScJineider, Site
Coordinator. April, 1993.
75Ibid
'°Ohio Environmental Protection Agency, 1993. Fact Sheet: Southwestern Portland Cement Company — Landfill
No. 6, Fairbom, Ohio. March, 1993.
Ohio Environmental Protection Agency, 1986. Preliminary Assessment, Southwestern Portland Cement, Fairbom,
Ohio (Landfill #6), Part 3 • Description of Hazardous Conditions and Incidents. July 18, 1986.
-------
5-31
The EPA has summarized reported analyses of surface waters and ground waters
associated with Landfill #6.78 As shown in Exhibit 5-10, surface water samples collected from
December,
Ohio Environmental Protection Agency, 1992. Director's Final Findings and Order in the Matter of Southdown,
Inc., 506 East Xenia Drive, Fairbom, Ohio.
-------
5-32
Exhibit 5-9
Site Diagram - Southwestern Portland Cement, Fairborn, Ohio
-------
5-33
Exhibit 5-10
Summary of Exceedances* of State Metals Limits
Southwestern Portland Cement • Fa inborn, Ohio
Landfill #6
Date/Media Sampled*1
3/93: Ground water*-"
As: 1.2
Fe: 9.4
Hg: 0.0016
Ni: 0.23
Se: 0.018
Zn: 0.05
pH: 13.38
3/93: Surface water*
Fe: 2.39
pH: 12.8
3/93: Surface water4"
As: 0.09
Fe: 9.24
Se: 0.026
pH: 12.89
3/93: Composite
Ground Water*-0
As: 0.927
Cd: 0.024
Cr: 0.105
Pb: 0.108
Ni: 0.283
Se: 0.022
pH: 12.08
3/93: Surface water"-"
As: 0.14
pH: 12.88
12/90-3/91: Surface Water1-0
As: 0.83
Cd: 0.02
Cr: 0.100
Pb: 0.037
Ni: 0.283
pH: 12.9
3/93: Surface water*-"
As: 0.157
Fe: 5.14
Se: 0.028
pH: 13.6
10/90:
Surface Water1-0
As: 0.388
Pb: 0.070
Se: 0.07
-------
5-34
Exhibit 5-10 (continued)
Summary of Exceedances* of State Metals Limits
Southwestern Portland Cement - Fairborn, Ohio
Landfill #6
Date/Media Sampled11
11/90: pH readings^
pH6: 13.44
pH8: 9.76
pH9: 13.63
pHIO: 13.70
pHll: 1330
Drinking Water1
As: 0.05 P
Ba: 1.0 P
Cd: 0.01 P
Cr: 0.05 P
Cu: 1.0 S
Fe: 0.3 S
Pb: 0.05 P
Mn: 0.5 S
Hg: 0.002 P
Se: 0.01 P
Ag: 0.05 P
Zn: 5.0 S
pH: 7 - 10.5 S
Agricultural Water1
As: 0.1
Be: 0.1
Cr: 0.1
Cu: 0.5
Fe: 5.0
Pb: 0.1
Hg: 0.01
Ni: 0.2
Se: 0.05
Zn: 25.0
Background"1
As: <0.05
Be: < 0.004
Cd: < 0.001
Cr: <0.01
Cu: 0.02
Fe: 0.12
Pb: < 0.003
Mn: <0.005
Hg: <0.20
Ni: <0.01
Se: < 0.005
Ag: < 0.001
Zn: 0.02
pH: 7.10
* Constituent concentrations higher than Stale standards are marked in bold. No violation of water standards is
implied.
b All concentrations in mg/L except pi! in standard units.
c Ground water from seep (MW-3) located at toe of landfill.
d Surface water sample (WN-2) from drainage from toe of landfill.
* Surface water sample (SW-16) from seep at toe of landfill.
f Surface water sample (SW-17) from seep at toe of landfill.
8 Surface water sample (SW-15) collected from stream at west toe of landfill.
h Composite of ground-water samples from on-site monitoring wells. Reported constituent levels are the
highest concentrations observed during the sampling period.
1 Surface water and Icachate samples from landfill. Listed constituent levels are the highest concentrations
observed during the sampling period.
' Surface water sample of ponded Icachate collected south of landfill.
k Readings from surface streams around the southern and western edge of landfill.
1 Water Quality Standards, State of Ohio.
m Ground-water well located upgradient from Landfill #6.
" Ohio EPA, 1993, Unpublished surface water and ground-water monitoring data from Landfill #6,
Southwestern Portland Cement Co, Fairborn, Ohio.
0 Ohio EPA, 1992, Administrative Order against Southwestern Portland Cement Co., Greene Co., Ohio.
p EPA, 1991, Table 4-3: Field Investigation Team (FlT)-collected pH readings.
-------
5-35
1990 to March, 1991 around Landfill #6 showed elevated levels of arsenic 17 times OEPA
standard), cadmium (2 times OEPA standard), and chromium (2 times OEPA standard) above
Ohio EPA drinking water limits. Levels of nickel were 1.4 times the State limit for agricultural
waters. Highly alkaline ground waters (pH > 12) sampled during January and February, 1991,
had similar degrees of exceedance for arsenic, cadmium, chromium, and nickel. In addition,
levels of lead were reported to be as high as 7 times the Federal drinking water limit (0.015
mg/L). The pH readings from surface streams collected in November, 1990 were reported as
high as 13.7.
In July 1992, the OEPA issued an administrative enforcement order against the facility
for past disposal activities at Landfill #6. In the order's findings of fact, OEPA determined that
the wastes disposed of in Landfill #6, including CKD, contained arsenic, lead, mercury, nickel,
selenium, zinc, cadmium, chromium, copper, and phenolics, and, therefore, are industrial wastes.
OEPA also determined that the leachate, because of its high pH (up to 13.7), is a hazardous
waste, and when released from Landfill #6, constitutes disposal of hazardous waste. According
to State law, the deposit (i.e., disposal) of industrial waste and hazardous waste in surface and
ground waters constitutes pollution (i.e., damage) of State waters. The order requires that a
CERCLA Remedial Investigation and Feasibility Study be conducted for this area.79 To date,
no remedial actions have been undertaken at this site.
Surface water and ground-water samples collected from streams around Landfill #1 are
characterized by high pH, but only arsenic, iron, and selenium are elevated above State water
quality standards. In a reconnaissance of the site in June, 1991, the Ohio EPA reported levels of
arsenic (0.06 mg/L) 1.2 times the OEPA drinking water limit of 0.05 mg/L, iron (0.51 mg/L)
three times the OEPA secondary drinking water limit (0.3 mg/L), and selenium (0.021 mg/L) 2.1
times the OEPA drinking water limit of 0.01 in surface water from a seep at the point of
emergence along the north toe of the landfill. The pH of the water was highly alkaline (11.58)
and exceeded the State drinking water standard of 10.5.80 Elevated levels of arsenic (0.12 mg/L,
2.4 times OEPA drinking water limit) and iron (4.1 mg/L, 13.6 times Ohio EPA drinking water
limit) in ground water associated with a seepage along the northwest slope of Landfill #1 also
were reported in a site assessment of the landfill prepared for Southdown.81
The OEPA has also reported elevated levels of copper, lead, zinc, and selenium in excess
of standards for warmwater wildlife habitats, in surface water samples collected along the margin
of Landfill #1.82 Although the concentrations of these elements are below the general State
drinking water standards (copper: < 10 - 45 ppb, lead: 10 ppb, zinc: 16 - 60 ppb), these elements
are considered elevated due to the very low water hardness of these samples (12-41 ppm CaCO3)
relative to normal water hardness (200-400 mg/L CaCO3). The low water hardness increases the
sensitivity of aquatic organisms to these constituents. The State limits for lead in waters with low
19 ibid.
K Ohio Environmental Protection Agency, 1991. Memo from Louise T. Snyder, DWQPA, SDWO on the
Southwestern Portland Cement facility, Landfill #1. September 9, 1991.
81 Southdown, Inc., 1992. Subarea 1 Site Evaluation, Southwestern Portland Cement Company, Fairbom, Ohio.
Prepared by Ground Water Associates, Inc, April, 1992.
82 Ohio Environmental Protection Agency, 1991. Memo from Louise Snyder, op.cit.
-------
5-36
hardness (12-41 ppm CaCO3) range from 9.1 ppb to 42 ppb. The ranges for copper and zinc are
2.1-7.2 ppb and 20-55 ppb, respectively.83
Damages at this site include contamination of on-site surface water and ground water.
Damages have been documented in several studies. The contaminants of most concern to human
health at both CKD landfills include pH and arsenic. The metals arsenic, selenium, chromium,
lead, and pH have all been observed at levels exceeding either primary or secondary drinking
water standards. These damages have resulted from the disposal of CKD in unlined landfills.
No remedial actions have been initiated for this site. However, a 1992 Administrative Order
issued by the Ohio Department of Natural Resources requires the company to undertake a
remedial investigation and feasibility study for Landfill #6.
5.2.6 National Gypsum CoTLafarge Corp., Alpena, Michigan
National Gypsum Company owned and operated a cement manufacturing facility
northeast of Alpena, Michigan on the shore of Lake Huron's Thunder Bay. In 1986, Lafarge
Corporation purchased the facility from National Gypsum and is the current owner and operator.
Cement has been manufactured at this site since at least the 1890s.84
During the 1980s, National Gypsum disposed of its CKD in a waste pile located northeast
of the facility along the edge of Lake Huron. The site covers more than 30 hectares (77 acres)
and is approximately 300 meters (984 feet) x 600 meters, with CKD piled as high as 18 meters
above the level of the lake.85 The site has been inactive since 1986, when Lafarge took over
operations. All CKD in this pile was generated prior to Lafarge's decision to burn hazardous
waste fuels.86 A site layout is provided in Exhibit 5-11.
Evidence of environmental release of CKD originating from the pile has been
documented by the Michigan Department of Natural Resources (MDNR). During a site visit in
March, 1993, MDNR inspectors reported CKD washing into a large erosion ditch (1 meter wide
x 3 meters deep) leading to Lake Huron, along with other debris, including airbags, drums, kiln
brick, and other miscellaneous debris co-managed with the dust. In addition, waves from the
lake were reported to be actively eroding the pile along 6- to 9-meter high banks on the south
end of the shoreline.87 MDNR has provided the Agency with photographs and videotapes
M Michigan Department of Natural Resources, 1993. Personal communication with JoAnn Merrick, Acting Chief,
Compliance and Enforcement Section, Waste Management Division, Michigan Department of Natural Resources.
July, 1993.
85 Michigan Department of Natural Resources, 1993. Letter from John W. Vick, Environmental Response
Division to Rebecca Beasly, Assistant General Council, National Gypsum Company. May, 1993.
" Michigan Department of Natural Resources, 1993, Interoffice Communication from Jim Sygo, Chief, Waste
Management Division, to Russell Harding, Deputy Director. April, 1993.
87 Michigan Department of Natural Resources, 1993. John W. Vick letter, op. cit.
-------
5-37
showing CKD washing into Lake Huron by means of flow down erosion channels on the pile and
wave action along the shore.88
M Michigan Department of Natural Resources, 1993. Photos of National Gypsum CKD pile taken during MDNR
site visits on March 30, 1993 and April 22, 1993. Videotape of National Gypsum CKD pile taken by John W. Vick on
April 30, 1993.
-------
5-38
Exhibit 5-11
Site Diagram - National Gypsum CoTLaFarge Corp., Alpena, Michigan
Evidence of contamination was found in soil and surface water samples obtained from the
pile near the shore of Lake Huron. As shown in Exhibit 5-12, surface water samples from the
erosion ditch and nearby Lake Huron show levels of arsenic and lead in excess of standards
specified under the Michigan Environmental Response Act (MERA, 1982 PA 307, as amended).
Grab samples of soil from the beach and upslope from the shore on the CKD pile had elevated
levels of arsenic, selenium, lead, and zinc, all above default values for soil cleanup.89
MDNR considers the presence of heavy metals in CKD and nearby surface waters to be a
"release of hazardous substances under MERA," which "represents a threat to public health and
the environment." MDNR has advised both National Gypsum Co. and Lafarge Corp. that they
89 Michigan Department of Natural Resources, 1993. John W. Vick letter, op.cit.
-------
5-39
are in violation of the Michigan Water Resources Commission Act (MWRC, PA 1929, as
amended).90
"ibid.
-------
5-40
Exhibit 5-12
Summary of Exceedances of State Metals Limits
National Gypsum CoTLafarge Corp., Alpena, Michigan
Date: Media Sampled
3/93: Surface Water; Shore of
Lake Huron adjacent to
CKD pile.
3/93: Surface Water: Erosion
ditch on CKD pile 20
feet from shore.
3/93: Soil Sample: Surface
grab sample taken from
CKD pile.
3/93: Soil Sample: Surface
grab sample taken from
beach northeast of
erosion ditch.
3/93: Soil Sample: Surface
grab sample taken from
sediment at mouth of
erosion ditch.
Constituent/Observed
Concentration
As: 30 ppb
Pb: 32 ppb
As: 6.52 ppm
Se: 0.546 ppm
Zn: 53 ppm
As: 27.1 ppm
Pb: 36 ppm
Zn: 115 ppm
As: 23.2 ppm
Pb: 51 ppm
Se: 3.15 ppm
Zn: 134 ppm
State
Standard*
As: 0.02 ppb
Pb: 8 ppb
As: 5.8 ppm$
Se: 0.41 ppm$
Zn: 47 ppmt
As: 5.8 ppm$
Pb: 21 ppm$
Zn: 47 ppm$
As: 5.8 ppm$
Pb: 21 ppm$
Se: 0.41 ppm$
Zn: 47 ppm$
* Standards specified under the Michigan Environmental Response Act (MERA) (1982 PA 307, as
amended).
$ MERA Type A soil cleanup criteria
Currently, MDNR is negotiating with both companies to initiate interim response actions to
prevent further erosion and deposition of contaminants into Lake Huron.91
5.2.7 Ash Grove Cement West, Montana City, Montana
Ash Grove Cement West's Montana City facility is located on a 197 hectare (486 acre)
site less than 10 kilometers (km) (6.2 miles) south of the city of Helena, Montana. The plant
utilizes a wet process to manufacture cement in one kiln, which has an annual capacity of
91 Michigan Department of Natural Resources, 1993. Personal communication with John W. Vick, Environmental
Quality Analyst, Environmental Response Division.
-------
5-41
269,510 metric tons (297,079 tons).92 Facility boundaries are adjacent to the unincorporated
town of Montana City, with an estimated 300 residents living within 0.8 km of the facility's
boundary. No known sensitive areas (e.g., wetlands or endangered species habitats) are located
nearby. However, there are private drinking water wells within 0.8 km of the facility's
boundary.93
In 1990, the facility utilized predominantly natural gas, coal, and coke for its fuel needs.
These fuels were supplemented by 8,270 kiloliters (2.18 million gallons) of waste pitch.94 In
1991, Ash Grove West applied for precompliance certification to burn hazardous waste under the
Boiler and Industrial Furnace rule, but was denied status by Region 8.
Waste CKD is landfilled in a draw on the east side of the quarry. In 1990, Ash Grove
West in Montana City generated an estimated 29,000 metric tons of CKD, of which 19,000
metric tons were landfilled (the remainder being returned to the kiln). Prior to 1989, CKD was
co-managed with shale overburden mined from quarry operations. Since the fall of 1989, CKD
has been monofilled over the co-managed pile. At the end of 1991, the landfill was estimated to
hold 77,000 metric tons of cumulative material.95
Stormwater run-off flows into one of two holding ponds, each of which discharges south
of the plant proper via permitted outfalls into Prickly Pear Creek. Run-off from the active CKD
landfill flows into a lower holding pond where it percolates through a gravel dam and discharges
into Prickly Pear Creek (outfall is currently valved shut).96-97 Run-off from a second upper
pond discharges into Prickly Pear Creek, 245 meters (800 feet) further upstream from the
discharge outfall from the lower pond.98 CKD disposal areas are shown in Exhibit 5-13.
In December 1990, the State of Montana Department of Health and Environmental
Sciences filed a Letter of Complaint and Application for Injunction against Ash Grove West, Inc.
for violations of discharge permit limits at the Montana City facility.99 In its claim, the
Department describes two catastrophic releases from the plant's wastewater ponds into Prickly
Pear Creek. Both releases involved quantities of CKD which flowed into the creek.
92 Portland Cement Association, 1992. PCA CKD Survey: Response from Ash Grove West, Inc., Montana City,
Montana.
93 Ibid.
"Ibid.
95 Portland Cement Association, 1992. PCA CKD Survey: Response from Ash Grove West, Inc., Montana City,
Montana.
96 Ibid.
97 Ash Grove Cement West, Montana City, Montana facility, 1993. Personal communication with plant personnel.
July, 1993.
* Ibid.
99 Department of Health and Environmental Sciences, State of Montana, 1990. Slate of Montana ex. rel v. Ash
Grove Cement West, Inc. Complaint and Application for Injunction, Cause No. 8442, Montana 5th Judicial District
Court, Jefferson County. December 11, 1990.
-------
5-42
-------
5-43
Exhibit 5-13
Site Diagram - Ash Grove Cement West, Montana City, Montana
As described in the Complaint, the first violation, which occurred on June 28,1990,
involved the release into Prickly Pear Creek of substantial quantities of sludge which had been
previously excavated that morning from the bottom of the lower wastewater pond. A late
morning/early afternoon storm washed substantial quantities of the excavated materials into
Prickly Pear Creek. Subsequent measurements of creek waters downstream of the discharge
point showed a total suspended solids level of 586.8 mg/L, compared to 10.1 mg/L upstream of
the discharge point.100 This is a violation of the plant's State effluent limit of 50 mg/L for total
dissolved solids.101
100 ibid.
101 Department of Health and Environmental Sciences, State of Montana, 1990. State of Montana ex. reL v. Ash
Grove Cement West, Inc., Consent Decree, Stipulation and Order, Cause No. 8442, Montana 5th Judicial Court,
Jefferson County.
-------
5-44
In the second violation, on August 16,1990, the lower holding pond failed after Ash
Grove pumped dense liquid sludge from the slurry tanks into the holding pond. Catastrophic
failure of the holding pond resulted in discharges into Prickly Pear Creek that raised the
concentration of total dissolved solids from 5.2 mg/L upstream to 37,368 mg/L near the discharge
point, and 4,453 mg/L 150 meters downstream from the discharge point.102 This is also a
violation of the plant's State limit of 50 mg/L for total dissolved solids.
Ash Grove acknowledged in a Consent Decree that both events allowed materials to
pollute Prickly Pear Creek in violation of State law. In addition to exceeding the State permit
limit for total dissolved solids, the discharges increased the turbidity above naturally occurring
conditions, and "created a nuisance, harmed aquatic life, and formed objectionable emulsions and
deposits .. ."ltD
5.3 CASES OF POTENTIAL DAMAGE TO GROUND AND SURFACE WATER
The Agency has identified cases of potential damage at three sites: (1) Texas Industries
facility in Midlothian, Texas, (2) Holnam facility in Artesia, Mississippi and, (3) Markey
Machinery Property in Seattle, Washington. In these cases there is information available to
indicate that surface water located on site has been contaminated above Maximum Contaminant
Levels (MCLs), but there is no data to indicate whether or not such levels have interacted with
either nearby ground water or other surface waters off site. For example, at the Texas Industries
facility, exceedances of metal standards were found in small, isolated puddles that were in no
obvious communication with any other body of surface water. Furthermore, there is no known
ground-water contamination at the Holnam facility.
5.3.1 Texas Industries, Inc., Midlothian, Texas
The Texas Industries facility is located in Ellis County on a 643 hectare (1,587 acre) tract
of land 3.5 kilometers (km) (2.17 miles) southwest of Midlothian, Texas.104 The plant
manufactures approximately 1,088,900 metric tons (1,200,286 tons) of Portland cement per year
in four wet process rotary kilns, and is authorized to burn hazardous waste for energy
recovery.105 Land use in the vicinity is predominantly agricultural, with low-density rural
residential areas located adjacent to facility property boundaries to the east, south, and
northwest.106
102 Department of Health and Environmental Sciences, State of Montana, 1990. Complaint and Application for
Injunction, op.clt.
105 Department of Health and Environmental Sciences, State of Montana, 1990. State of Montana ex. rel v. Ash
Grove Cement West, Inc., Consent Decree, Stipulation, and Order, Cause No 8442, Montana 5th Judicial District
Court, Jefferson County. December 19,1990.
1M Texas Industries, Inc., 1992. Part B Permit Application to EPA and tlie Texas Water Commission. Prepared by
Entellect Environmental Services for Texas Industries, Inc. p. 1-9.
105 U.S. EPA Region 6, 1992. Complaint, Compliance Order, and Notice of Opportunity for Hearing in the Matter of
Texas Industries, Inc., Midlothian, Texas. Docket No. RCRA VI-203-H, Hazardous Waste Management Division.
September, 1992.
106 Texas Industries, Inc., 1992. op.clt.
-------
5-45
Each kiln produces 40 to 45 metric tons of CKD per day, all of which is wasted from the
system. Sixty to 80 percent of the CKD is pelletized in a pug mill and landfilled on site, while
the remaining 20 to 40 percent is sold either as roadbed filler or as a stabilizer.107 The facility
has two on-site CKD landfills, an active landfill in a depleted quarry area, and an inactive capped
landfill located in the quarry to the southwest of the active disposal area.108 A diagram of the
site is provided in Exhibit 5-14.
Exhibit 5-14
Site Diagram - Texas Industries, Inc., Midlothian, Texas
Temperate climactic conditions
-------
5-35
1990 to March, 1991 around Landfill #6 showed elevated levels of arsenic 17 times OEPA
standard), cadmium (2 times OEPA standard), and chromium (2 times OEPA standard) above
Ohio EPA drinking water limits. Levels of nickel were 1.4 times the State limit for agricultural
waters. Highly alkaline ground waters (pH > 12) sampled during January and February, 1991,
had similar degrees of exceedance for arsenic, cadmium, chromium, and nickel. In addition,
levels of lead were reported to be as high as 7 times the Federal drinking water limit (0.015
mg/L). The pH readings from surface streams collected in November, 1990 were reported as
high as 13.7.
In July 1992, the OEPA issued an administrative enforcement order against the facility
for past disposal activities at Landfill #6. In the order's findings of fact, OEPA determined that
the wastes disposed of in Landfill #6, including CKD, contained arsenic, lead, mercury, nickel,
selenium, zinc, cadmium, chromium, copper, and phenolics, and, therefore, are industrial wastes.
OEPA also determined that the leachate, because of its high pH (up to 13.7), is a hazardous
waste, and when released from Landfill #6, constitutes disposal of hazardous waste. According
to State law, the deposit (i.e., disposal) of industrial waste and hazardous waste in surface and
ground waters constitutes pollution (i.e., damage) of State waters. The order requires that a
CERCLA Remedial Investigation and Feasibility Study be conducted for this area.79 To date,
no remedial actions have been undertaken at this site.
Surface water and ground-water samples collected from streams around Landfill #1 are
characterized by high pH, but only arsenic, iron, and selenium are elevated above State water
quality standards. In a reconnaissance of the site in June, 1991, the Ohio EPA reported levels of
arsenic (0.06 mg/L) 1.2 times the OEPA drinking water limit of 0.05 mg/L, iron (0.51 mg/L)
three times the OEPA secondary drinking water limit (0.3 mg/L), and selenium (0.021 mg/L) 2.1
times the OEPA drinking water limit of 0.01 in surface water from a seep at the point of
emergence along the north toe of the landfill. The pH of the water was highly alkaline (11.58)
and exceeded the State drinking water standard of 10.5.80 Elevated levels of arsenic (0.12 mg/L,
2.4 times OEPA drinking water limit) and iron (4.1 mg/L, 13.6 times Ohio EPA drinking water
limit) in ground water associated with a seepage along the northwest slope of Landfill #1 also
were reported in a site assessment of the landfill prepared for Southdown.81
The OEPA has also reported elevated levels of copper, lead, zinc, and selenium in excess
of standards for warmwater wildlife habitats, in surface water samples collected along the margin
of Landfill #1.82 Although the concentrations of these elements are below the general State
drinking water standards (copper: < 10 - 45 ppb, lead: 10 ppb, zinc: 16 - 60 ppb), these elements
are considered elevated due to the very low water hardness of these samples (12-41 ppm CaCO3)
relative to normal water hardness (200-400 mg/L CaCO3). The low water hardness increases the
sensitivity of aquatic organisms to these constituents. The State limits for lead in waters with low
"Ibid.
K Ohio Environmental Protection Agency, 1991. Memo from Louise T. Snyder, DWQPA, SDWO on the
Southwestern Portland Cement facility, Landfill #1. September 9, 1991.
81 Southdown, Inc., 1992. Subarea 1 Site Evaluation, Southwestern Portland Cement Company, Fairbom, Ohio.
Prepared by Ground Water Associates, Inc, April, 1992.
K Ohio Environmental Protection Agency, 1991. Memo from Louise Snyder, op.dt.
-------
5-36
hardness (12-41 ppm CaCO3) range from 9.1 ppb to 42 ppb. The ranges for copper and zinc are
2.1-7.2 ppb and 20-55 ppb, respectively.83
Damages at this site include contamination of on-site surface water and ground water.
Damages have been documented in several studies. The contaminants of most concern to human
health at both CKD landfills include pH and arsenic. The metals arsenic, selenium, chromium,
lead, and pH have all been observed at levels exceeding either primary or secondary drinking
water standards. These damages have resulted from the disposal of CKD in unlined landfills.
No remedial actions have been initiated for this site. However, a 1992 Administrative Order
issued by the Ohio Department of Natural Resources requires the company to undertake a
remedial investigation and feasibility study for Landfill #6.
5.2.6 National Gypsum CoTLafarge Corp., Alpena, Michigan
National Gypsum Company owned and operated a cement manufacturing facility
northeast of Alpena, Michigan on the shore of Lake Huron's Thunder Bay. In 1986, Lafarge
Corporation purchased the facility from National Gypsum and is the current owner and operator.
Cement has been manufactured at this site since at least the 1890s.84
During the 1980s, National Gypsum disposed of its CKD in a waste pile located northeast
of the facility along the edge of Lake Huron. The site covers more than 30 hectares (77 acres)
and is approximately 300 meters (984 feet) x 600 meters, with CKD piled as high as 18 meters
above the level of the lake.85 The site has been inactive since 1986, when Lafarge took over
operations. All CKD in this pile was generated prior to Lafarge's decision to burn hazardous
waste fuels.86 A site layout is provided in Exhibit 5-11.
Evidence of environmental release of CKD originating from the pile has been
documented by the Michigan Department of Natural Resources (MDNR). During a site visit in
March, 1993, MDNR inspectors reported CKD washing into a large erosion ditch (1 meter wide
x 3 meters deep) leading to Lake Huron, along with other debris, including airbags, drums, kiln
brick, and other miscellaneous debris co-managed with the dust. In addition, waves from the
lake were reported to be actively eroding the pile along 6- to 9-meter high banks on the south
end of the shoreline.87 MDNR has provided the Agency with photographs and videotapes
0 ibid.
84 Michigan Department of Natural Resources, 1993. Personal communication with JoAnn Merrick, Acting Chief,
Compliance and Enforcement Section, Waste Management Division, Michigan Department of Natural Resources.
July, 1993.
85 Michigan Department of Natural Resources, 1993. Letter from John W. Vick, Environmental Response
Division to Rebecca Beasly, Assistant General Council, National Gypsum Company. May, 1993.
K Michigan Department of Natural Resources, 1993, Interoffice Communication from Jim Sygo, Chief, Waste
Management Division, to Russell Harding, Deputy Director. April, 1993.
17 Michigan Department of Natural Resources, 1993. John W. Vick letter, op. cit.
-------
5-37
showing CKD washing into Lake Huron by means of flow down erosion channels on the pile and
wave action along the shore.88
M Michigan Department of Natural Resources, 1993. Photos of National Gypsum CKD pile taken during MDNR
site visits on March 30, 1993 and April 22, 1993. Videotape of National Gypsum CKD pile taken by John W. Vick on
April 30, 1993.
-------
5-38
Exhibit 5-11
Site Diagram - National Gypsum Co./LaFarge Corp., Alpena, Michigan
Evidence of contamination was found in soil and surface water samples obtained from the
pile near the shore of Lake Huron. As shown in Exhibit 5-12, surface water samples from the
erosion ditch and nearby Lake Huron show levels of arsenic and lead in excess of standards
specified under the Michigan Environmental Response Act (MERA, 1982 PA 307, as amended).
Grab samples of soil from the beach and upslope from the shore on the CKD pile had elevated
levels of arsenic, selenium, lead, and zinc, all above default values for soil cleanup.89
MDNR considers the presence of heavy metals in CKD and nearby surface waters to be a
"release of hazardous substances under MERA," which "represents a threat to public health and
the environment." MDNR has advised both National Gypsum Co. and Lafarge Corp. that they
89 Michigan Department of Natural Resources, 1993. John W. Vick letter, op.cit.
-------
5-39
are in violation of the Michigan Water Resources Commission Act (MWRC, PA 1929, as
amended).90
'Ibid.
-------
5-40
Exhibit 542
Summary of Exceedances of State Metals Limits
National Gypsum CoVLafarge Corp., Alpena, Michigan
Date: Media Sampled
3/93: Surface Water; Shore of
Lake Huron adjacent to
CKD pile.
3/93: Surface Water: Erosion
ditch on CKD pile 20
feet from shore.
3/93: Soil Sample: Surface
grab sample taken from
CKD pile.
3/93: Soil Sample: Surface
grab sample taken from
beach northeast of
erosion ditch.
3/93: Soil Sample: Surface
grab sample taken from
sediment at mouth of
erosion ditch.
Constituent/Observed
Concentration
As: 30 ppb
Pb: 32 ppb
As: 6.52 ppm
Se: 0.546 ppm
Zn: 53 ppm
As: 27.1 ppm
Pb: 36 ppm
Zn: 115 ppm
As: 23.2 ppm
Pb: 51 ppm
Se: 3.15 ppm
Zn: 134 ppm
State
Standard*
As: 0.02 ppb
Pb: 8 ppb
As: 5.8 ppmt
Se: 0.41 ppmt
Zn: 47 pprnj
As: 5.8 ppm$
Pb: 21 ppm$
Zn: 47 ppmt
As: 5.8 ppm$
Pb: 21 ppm$
Se: 0.41 ppm$
Zn: 47 ppm$
• Standards specified under the Michigan Environmental Response Act (MERA) (1982 PA 307, as
amended).
$ MERA Type A soil cleanup criteria
Currently, MDNR is negotiating with both companies to initiate interim response actions to
prevent further erosion and deposition of contaminants into Lake Huron."
5.2.7 Ash Grove Cement West, Montana City, Montana
Ash Grove Cement West's Montana City facility is located on a 197 hectare (486 acre)
site less than 10 kilometers (km) (6.2 miles) south of the city of Helena, Montana. The plant
utilizes a wet process to manufacture cement in one kiln, which has an annual capacity of
91 Michigan Department of Natural Resources, 1993. Personal communication with John W. Vick, Environmental
Quality Analyst, Environmental Response Division.
-------
5-41
269,510 metric tons (297,079 tons).92 Facility boundaries are adjacent to the unincorporated
town of Montana City, with an estimated 300 residents living within 0.8 km of the facility's
boundary. No known sensitive areas (e.g., wetlands or endangered species habitats) are located
nearby. However, there are private drinking water wells within 0.8 km of the facility's
boundary.93
In 1990, the facility utilized predominantly natural gas, coal, and coke for its fuel needs.
These fuels were supplemented by 8,270 kiloliters (2.18 million gallons) of waste pitch.94 In
1991, Ash Grove West applied for precompliance certification to bum hazardous waste under the
Boiler and Industrial Furnace rule, but was denied status by Region 8.
Waste CKD is landfilled in a draw on the east side of the quarry. In 1990, Ash Grove
West in Montana City generated an estimated 29,000 metric tons of CKD, of which 19,000
metric tons were landfilled (the remainder being returned to the kiln). Prior to 1989, CKD was
co-managed with shale overburden mined from quarry operations. Since the fall of 1989, CKD
has been monofilled over the co-managed pile. At the end of 1991, the landfill was estimated to
hold 77,000 metric tons of cumulative material.95
Stormwater run-off flows into one of two holding ponds, each of which discharges south
of the plant proper via permitted outfalls into Prickly Pear Creek. Run-off from the active CKD
landfill flows into a lower holding pond where it percolates through a gravel dam and discharges
into Prickly Pear Creek (outfall is currently valved shut).96-97 Run-off from a second upper
pond discharges into Prickly Pear Creek, 245 meters (800 feet) further upstream from the
discharge outfall from the lower pond.98 CKD disposal areas are shown in Exhibit 5-13.
In December 1990, the State of Montana Department of Health and Environmental
Sciences filed a Letter of Complaint and Application for Injunction against Ash Grove West, Inc.
for violations of discharge permit limits at the Montana City facility.99 In its claim, the
Department describes two catastrophic releases from the plant's wastewater ponds into Prickly
Pear Creek. Both releases involved quantities of CKD which flowed into the creek.
92 Portland Cement Association, 1992. PCA CKD Suivey: Response from Ash Grove West, Inc., Montana City,
Montana.
91 Ibid.
94 Ibid.
95 Portland Cement Association, 1992. PCA CKD Suivey: Response from Ash Grove West, Inc., Montana City,
Montana.
* Ibid.
97 Ash Grove Cement West, Montana City, Montana facility, 1993. Personal communication with plant personnel.
July, 1993.
*Ibid.
99 Department of Health and Environmental Sciences, State of Montana, 1990. State of Montana ex. rel v. Ash
Grove Cement West, Inc. Complaint and Application for Injunction, Cause No. 8442, Montana 5th Judicial District
Court, Jefferson County. December 11, 1990.
-------
5-42
-------
5-43
Exhibit 5-13
Site Diagram - Ash Grove Cement West, Montana City, Montana
As described in the Complaint, the first violation, which occurred on June 28, 1990,
involved the release into Prickly Pear Creek of substantial quantities of sludge which had been
previously excavated that morning from the bottom of the lower wastewater pond. A late
morning/early afternoon storm washed substantial quantities of the excavated materials into
Prickly Pear Creek. Subsequent measurements of creek waters downstream of the discharge
point showed a total suspended solids level of 586.8 mg/L, compared to 10.1 mg/L upstream of
the discharge point.100 This is a violation of the plant's State effluent limit of 50 mg/L for total
dissolved solids.101
100 ibid.
101 Department of Health and Environmental Sciences, State of Montana, 1990. State of Montana ex. reL v. Ash
Grove Cement West, Inc., Consent Decree, Stipulation and Order, Cause No. 8442, Montana 5th Judicial Court,
Jefferson County.
-------
5-44
In the second violation, on August 16,1990, the lower holding pond failed after Ash
Grove pumped dense liquid sludge from the slurry tanks into the holding pond. Catastrophic
failure of the holding pond resulted in discharges into Prickly Pear Creek that raised the
concentration of total dissolved solids from 5.2 mg/L upstream to 37,368 mg/L near the discharge
point, and 4,453 mg/L 150 meters downstream from the discharge point.102 This is also a
violation of the plant's State limit of 50 mg/L for total dissolved solids.
Ash Grove acknowledged in a Consent Decree that both events allowed materials to
pollute Prickly Pear Creek in violation of State law. In addition to exceeding the State permit
limit for total dissolved solids, the discharges increased the turbidity above naturally occurring
conditions, and "created a nuisance, harmed aquatic life, and formed objectionable emulsions and
deposits .. "m
5.3 CASES OF POTENTIAL DAMAGE TO GROUND AND SURFACE WATER
The Agency has identified cases of potential damage at three sites: (1) Texas Industries
facility in Midlothian, Texas, (2) Holnam facility in Artesia, Mississippi and, (3) Markey
Machinery Property in Seattle, Washington. In these cases there is information available to
indicate that surface water located on site has been contaminated above Maximum Contaminant
Levels (MCLs), but there is no data to indicate whether or not such levels have interacted with
either nearby ground water or other surface waters off site. For example, at the Texas Industries
facility, exceedances of metal standards were found in small, isolated puddles that were in no
obvious communication with any other body of surface water. Furthermore, there is no known
ground-water contamination at the Holnam facility.
53.1 Texas Industries, Inc., Midlothian, Texas
The Texas Industries facility is located in Ellis County on a 643 hectare (1,587 acre) tract
of land 3.5 kilometers (km) (2.17 miles) southwest of Midlothian, Texas.104 The plant
manufactures approximately 1,088,900 metric tons (1,200,286 tons) of Portland cement per year
in four wet process rotary kilns, and is authorized to bum hazardous waste for energy
recovery.105 Land use in the vicinity is predominantly agricultural, with low-density rural
residential areas located adjacent to facility property boundaries to the east, south, and
northwest.106
102 Department of Health and Environmental Sciences, State of Montana, 1990. Complaint and Application for
Injunction, op.cit.
103 Department of Health and Environmental Sciences, State of Montana, 1990. State of Montana ex. rel v. Ash
Grove Cement West, Inc., Consent Decree, Stipulation, and Order, Cause No 8442, Montana 5th Judicial District
Court, Jefferson County. December 19, 1990.
104 Texas Industries, Inc., 1992. Part B Permit Application to EPA and the Texas Water Commission. Prepared by
Entellect Environmental Services for Texas Industries, Inc. p. 1-9.
105 U.S. EPA Region 6, 1992. Complaint, Compliance Order, and Notice of Opportunity for Hearing in the Matter of
Texas Industries, Inc., Midlothian, Texas. Docket No. RCRA VI-203-H, Hazardous Waste Management Division.
September, 1992.
106 Texas Industries, Inc., 1992. op.cit.
-------
5-45
Each kiln produces 40 to 45 metric tons of CKD per day, all of which is wasted from the
system. Sixty to 80 percent of the CKD is pelletized in a pug mill and landfilled on site, while
the remaining 20 to 40 percent is sold either as roadbed filler or as a stabilizer.107 The facility
has two on-site CKD landfills, an active landfill in a depleted quarry area, and an inactive capped
landfill located in the quarry to the southwest of the active disposal area.108 A diagram of the
site is provided in Exhibit 5-14.
Exhibit 5-14
Site Diagram - Texas Industries, Inc., Midlothian, Texas
Temperate climactic conditions in the region feed intermittent streams that flow over
impermeable clayey soils. Surface run-off from the plant proper discharges into the eastern
branch of Cottonwood Creek, and 5 kilometers further downstream into Joe Poole Lake, a public
drinking water reservoir. Surface run-off from the inactive pile flows into the East Branch of
107 Texas Industries, Inc., 1992. Persona] communication with plant operators during EPA CKD sampling visit.
March, 1992.
1M
Ibid.
-------
5-46
Cottonwood Creek,109 while run-off from the active CKD disposal area spills into the quarry
and is confined to the facility property.110 The quarry floor (18 meter (60 feet depth) is
fractured, and there is a large body of ponded water near the active disposal pile. Perched water
tables are within 11 meters of the surface and are above the quarry floor.111 Beneath the
facility the uppermost aquifer is located 73 meters below grade.112
CKD from the Texas Industries facility is known to contain leachable chromium.
Samples of CKD collected by the Texas Water Commission (TWC) in 1991, directly from
beneath one kiln (Kiln #4) had chromium levels in leachate of 0.44 mg/L. Kiln dust from the
active landfill had a level of chromium below, but close to 0.08 mg/L. One sample of CKD from
a "fugitive dust landfill" had a total chromium content of 881 mg/kg.113
During January 1992, inspectors from the TWC noticed pools of reddish-brown liquid
seeping from the inactive pile during a RCRA compliance inspection of Texas Industries'
hazardous waste treatment, storage, and disposal facilities.114 This seepage, believed to be
storm run-off,115 was noted as an "Area of Concern" in a Notice of Violation letter to the
facility describing violations of solid waste rules.116 Analysis of a sample of this liquid showed
levels of arsenic of 0.2 mg/L and lead of 0.03 mg/L.117
The seepage was again observed during a subsequent inspection of the facility in March
1992. Sample analysis showed the liquid to be extremely alkaline (pH: 13), with levels of arsenic
of 0.46 mg/L and chromium at 1.07 mg/L.118 As a result of the March 1992, inspection, EPA
109
Texas Water Commission, 1993. Personal communication with Sara Barrett, Field Investigator. July, 1993.
""Texas Industries, Inc., 1992. Persona] communication with facility personnel during EPA CKD sampling visit.
March, 1992.
111 Texas Water Commission, 1990. RCRA Facility Assessment Facility Checklist for Texas Industries,
Incorporated, Peter F. Lodde, reviewer.
112
Ibid.
113 Texas Water Commission, 1991. Letter from Allen Hayes, Environmental Quality Specialist, to files regarding a
review of laboratory analyses of samples collected on October 21 and 22, 1991, from North Texas Cement Co., Texas
Industries, Inc., and Box Crow Cement Co. November 18, 1991.
114 Texas Water Commission, 1992. Interoffice communication from Sam Barrett, Field Investigator, to files
regarding compliance inspection at Texas Industries, Inc., Midlothian, Texas. February, 1992.
115 Texas Water Commission, 1992. Personal communication with Sara Barrett, Field Investigator. June, 1992.
116 Texas Water Commission, 1992. Letter from Mary B. Adrian, Section Leader, Enforcement Section, TWC, to
E.L. Faciane, Staff Vice-President, Environmental Affairs, Texas Industries, p.3. April 15, 1992.
117 Texas Water Commission, 1992. Interoffice memorandum from Sara Barrett, Field Investigator to files
regarding record review of analytical results of samples collected from Texas Industries, Inc. during inspections on
January 27, 1992, March 10, 1992, and April 10, 1992. October 21, 1992.
118 Texas Water Commission, 1992. Letter from Sam Barrett to files regarding samples taken at Texas Industries,
op. cit.
-------
5-47
Region 6 filed a Letter of Complaint with Texas Industries, Inc. for violations of RCRA Subtitle
C regulations.119
Although seepage from the old landfill was observed only as localized pools, there exists a
potential for contaminants to migrate beyond plant boundaries. First, the cap on the old disposal
area can become eroded120 and allow stormwater access to disposed CKD. Secondly, the old
disposal pile is in close proximity (90 meters) to the East Branch of Cottonwood Creek.
Uncontrolled run-off from the old disposal pile would flow into both Cottonwood Creek and
adjacent Newton Creek.121 Furthermore, the probability of an uncontrolled release of CKD
into either creek would be highest during a storm event. The characteristic low permeability of
soils (10's cm/sec)122 within plant boundaries and in the immediate vicinity reduces the effect of
rainfall infiltration into the ground, decreasing the volume of surface run-off during a storm
event.
As a result of an inspection of both CKD disposal areas in March 1990, the TWC
concluded a potential exists for contaminant release from the landfills.123 TWC based its
finding on the presence of the shallow (11 meter depth) water table. CKD in the active area is
disposed on the quarry floor at a depth of 18 meters, which is below the level of the perched
water table (11 meters).124 In addition, the volume of disposed dust is high (estimated to be
28,350 cubic meters (37,059 cubic yards and nearby ponded water.125 The shallow ground-
water table combined with the high volume of waste in the active disposal area, the lack of a
landfill liner, and the proximity of the active landfill to ponded water combine to create an
"unknown potential" for release.126
5.3.2 Holnam, Inc., Artesia, Mississippi
The Holnam facility is located in Lowndes County, about 5 kilometers (km) (3.1 miles)
south of Artesia, Mississippi along Route 45. Facility property encompasses an estimated 120
hectares (300 acres) and is partly located in the 100-year flood plain. The surrounding land use
is predominantly rural and agricultural. In 1990, an estimated 60 residents lived within two
kilometers of the facility property boundary, with the nearest residence located 900 meters (2,953
119 U.S. Environmental Protection Agency, Region 6, 1992. Complaint, Compliance Order, and Notice of
Opportunity for Hearing in the Matter of Texas Industries, Inc. Docket Number RCRA VI-203-H. September 30, 1992.
120 Texas Water Commission, 1992. Personal communication with Sam Barrett, Field Investigator.
November, 1992.
121 Texas Water Commission, 1993. Personal communication with Sam Barrett, Field Investigator. July, 1993.
122 Texas Industries, Inc., 1992. Part B Permit Application, op.cit.
123 Texas Water Commission, 1990. RCRA Facility Assessment Facility Checklist, op.cit.
124 Ibid.
-------
5-48
feet) to the northwest. At least one private drinking water well is located within the facility's
boundary.127
Holnam's Artesia facility utilizes one wet process kiln with a 454,000 metric ton (500,440
ton) capacity to produce clinker. In 1992, the plant produced 426,670 metric tons of clinker
while burning coal (95,750 metric tons) almost exclusively for its energy needs. The facility
started burning hazardous waste as a fuel supplement in June, 1993.m
Non-waste derived CKD at Holnam is disposed in an abandoned, water-filled quarry
located northeast of the kiln. An estimated 253,000 metric tons of CKD is landfilled in the
quarry along the eastern edge. Two other waste CKD disposal areas also exist within facility
boundaries at Holnam. A large, older CKD disposal area, with an indeterminate amount of
CKD, is located 300 meters east of the active disposal area. In 1993, Holnam created a new
disposal area 9 to 12 meters south of the quarry disposal area, to manage hazardous waste-
derived CKD.
The quarry lake is filled to a depth of 3.2 meters with 632,000 kiloliters (167 million
gallons) of water, comprised of rain water and industrial process water. Water from the quarry
lake is pumped to make raw-material slurry and process water for the wet scrubbers. Industrial
process water, originating from the clinker cooler scrubber, flows into the quarry from a
discharge point located on the southwest side of the quarry lake.129 Stormwater run-off from
the quarry lake discharges via an NPDES permitted outfall into a tributary of the South
Branch.130 Exhibit 5-15 shows the Artesia site.
In May 1993, while collecting samples of CKD and clinker, the Agency measured
elevated levels of pH, in surface waters and discharge points within the property boundaries of
Holnam's Artesia facility. The pH of water in the quarry lake (described in Agency field notes
as a settling pond) was measured at 11.0 at a point along the northeast corner of the abandoned
quarry where grading permitted access to the edge of the water. In an open culvert near the
discharge point into the quarry lake, clinker cooler water had a measured pH of 11.6. Water in
a retention basin at the site of the old CKD waste pile had a measured pH of 11.2.131 The
Agency has no data regarding the potential for release at this site.
127 Portland Cement Association, 1991. PCA CKD Survey: Response from Holnam, Inc., Artesia, Mississippi.
124 United States Environmental Protection Agency, 1993. Unpublished field notes from Phase II CKD sampling
trip. May 25, 1993.
129 U.S. Environmental Protection Agency, 1993. Unpublished field notes collected during visit to Holnam, Inc.,
Artesia, Mississippi, May 25, 1993.
130 State of Mississippi, 1992. Permit for storrnwater run-off for Holnam, Inc., Artesia, Mississippi; NPDES Permit
No. MSR320017.
131 U.S. EPA, 1993. Unpublished field notes collected during CKD sampling visit to Holnam, Artesia, Mississippi.
op.cit.
-------
5-49
5.3.3 Markey Machinery Property, Seattle, Washington
The Markey Machinery Property site is a rectangular, 1.8 hectare (4.4 acre) CKD landfill
on industrial property within the city limits of Seattle, Washington."2 Between 1977 and 1978
an estimated 38,250 cubic meters (m3) (50,000 cubic yards [yd3]) of CKD was disposed on the
property
132 GeoEngineers, Inc., 1989. Environmental Site Assessment, CKD Landfill, Markey Machinery Property, Seattle,
Washington. Prepared for Helsell, Fetterman, Martin, Todd, & Hokanson. August, 1989.
-------
5-50
Exhibit 5-15
Site Diagram - Holnatn, Inc., Artesia, Mississippi
as fill, allegedly by Ideal Cement.133 The site, an old truck park, is located within 1,220 meters
(4,003 feet) of the Duwamish River, which is classified as a fishery by the State.134 Although
properties immediately adjacent to the site are industrial,1" there is a nearby population of
over 2,600 residents within 0.8 kilometers (0.5 miles) of the site.136
The site is immediately adjacent to surface drainage. Along the north boundary is the
eastward flowing Ham Creek, which intersects the Duwamish River further downstream. The
U4 Department of Ecology, State of Washington, 1992. Site Hazard Assessment, Markey Property, Parcel 4, south
96th Street/lOth Avenue Soutti, Seattle, Washington. September, 1992.
135 GeoEngineers, Inc., 1989. op.cit.
"'Department of Ecology, State of Washington, 1992. op.cit.
-------
5-51
east boundary of the site is marked by a ditch with intermittent flow which drains into Ham
Creek.137 Total annual rainfall averages nearly 86 cm (34 inches).138 The vertical depth to
ground water at the site is less than eight meters.139 The site is shown in Exhibit 5-16.
Exhibit 5-16
Site Diagram - Markey Machinery Property, Seattle, Washington
Analyses of four samples of CKD collected from test pits at the site in 1989, showed
elevated concentrations of heavy metals, including arsenic, cadmium, copper, lead, and zinc that
were higher than for uncontaminated soils.140 Two of these samples were collected at locations
along the southern margin of the landfill at the furthest distance away from the overlying waste
U7 GeoEngineers, Inc., 1989. op.cit.
1JS Department of Ecology, State of Washington, 1992. op.cit.
' GeoEngineers, Inc., 1989, op.cit.
-------
5-52
debris. Levels of lead (960-1,730 ppm) and arsenic (150-210 ppm) in samples of CKD from the
Markey site exceed State soil cleanup standards specified in the Model Toxics Control Act (lead:
250 ppm, arsenic: 20 ppm).141 Levels of cadmium (1.6-3.8 ppm) exceed the State soil cleanup
standard (2.0 ppm)142 in three out of four samples.143 The average pH of laboratory leachate
from the four samples is highly alkaline (12.4) and just below the State Dangerous Waste
criterion of 12.5.144
Analyses of ground-water samples collected in 1989 showed the concentrations of
dissolved metals to be below established drinking water limits. At the same time, the level of
lead in the wells ranged from less than 5 ppb to 8 ppb,145 and slightly exceeded the State
cleanup level for ground water (5.0 ppb)146 in three out of four wells. Analysis of water level
measurement in four ground-water monitoring wells at the site suggests the predominant flow of
shallow ground water is northeast toward Ham Creek.147
Analysis of surface water has shown the impact of the presence of CKD at Markey
Property. A surface water sample collected in 1989 from the ditch along the eastern boundary
had an elevated pH of 10.2 and a concentration of lead (0.36 ppm) 24 times the Federal limit for
drinking water (0.015 ppm; conversion assumes the density of water to be 1.0 g/cm3).148 A
sample of standing water along the southern boundary of the site had an alkaline pH of 9.4, and
a concentration of lead (0.025 ppm).
The State of Washington Department of Ecology has ranked the Markey Property CKD
landfill site a "3" on a scale of one to five, with one representing the highest level of concern and
five the lowest.149 The ranking is a measurement of potential risk to human health and the
environment relative to other contamination sites in the State.150
141
173-340
Department of Ecology, State of Washington, 1991. The Model Toxics Control Act Cleanup Regulation, Chapter
'0 WAC. As Amended, February, 1991. p. 108.
wlbid.
143 GeoEngineers, Inc., 1989. op.cit.
144
Ibid.
145 GeoEngineers, Inc., 1989. op.cit.
146 Department of Ecology, State of Washington, 1992. The Model Toxics Control Act Cleanup Regulation, Chapter
173-340 WAC. As amended. February, 1992. p. 93.
148 GeoEngineers, Inc., 1989. op.cit.
149 Department of Ecology, State of Washington, 1992. Site Hazard Assessment, Markey Property, Parcel 4, South
96th Street/lOlh Avenue South, Seattle, Washington. September, 1992.
130 Department of Ecology, State of Washington, 1992. Washington Ranking Method, Scoring Manual. As
amended. April, 1992.
-------
5-53
Several site characteristics contribute to release potential at the Markey Property site,
including: 1) the quantity of CKD used as fill (38,250 m3); 2) the lack of run-on or run-off
controls, cover, liner, or leachate containment system at the site; 3) the site's close proximity to
populated areas; and, 4) the close proximity of the site to environmentally sensitive surface
waters such as the Duwamish River.151 Enhancing the potential risk that this site poses to
human health and the environment are confirmed releases of lead to surface and ground waters
around the site.152
5.4 DOCUMENTED AIR DAMAGES
In addition to examining documented cases of damage to surface- and ground-water,
EPA reviewed available information for evidence of damage to the air media. In most cases, the
standard of proof of air damage was an administrative ruling in the form of an NOV of a State
or Federal regulation, issued by a State or Federal inspector153. For many cases, however,
although the Agency was provided with anecdotal information from an interview with a State
official, the Agency was unable to locate sufficient documentation to qualify them as damage
cases. Additionally, air damage information was gleaned from Hazardous Waste Site Preliminary
Assessment forms. The cases that met the standard of proof and the other cases that are less
well-documented suggest that cement kilns can be a significant cause of localized air quality
problems.
In conducting this study, EPA identified 21 incidents at 12 facilities that met one of the
tests of proof. NOVs were issued for these incidents, with three cases eventually settled through
a judicial settlement. Six of these facilities have received more than one NOV. With the
exception of two cases associated with the accumulation of fugitive dust, all of the cases were
associated with visible emission violations (opacity) related to equipment and process
malfunctions associated with the dust management system. This usually involved the baghouse,
clinker cooler, or dust screw conveyors. The 21 incidents that meet the test of proof are outlined
in Exhibit 5-17, Summary of Air Damages.
In general terms, if a visual inspection performed according to Method 9154 shows
opacity to be in excess of 20 percent, the facility is found to be in violation. Most states have
adopted the standard of 20 percent, with some states promulgating more stringent standards,
such as 10 percent.
Opacity limits are independently enforceable standards set out in the Clean Air Act (see
40 CFR, Part 60, New Source Performance Standards). Opacity is defined as the power of the
plume to obscure a background. Opacity is also an indirect measure of paniculate matter. EPA
uses opacity as an indicator of a problem with the combustion process or an air control device.
Since high opacity correlates with high paniculate matter, it may signify a health hazard. If
"' Department of Ecology, State of Washington, 1992. Site Assessment, Markey Property, op.cit.
U2Ibid.
m In many cases it was difficult to discern whether CKD was the source of the violation, since some notices
merely listed the air control rule that was violated. Where there is no description linking CKD to the violation, the
cases are not considered documented damage cases.
U4 40 CFR 60.60.
-------
5-54
opacity is high, EPA will ask for a compliance test to see if the facility meets the PM10
standard155.
155
This stack test measures the size of participate matter.
-------
5-55
Exhibit 5-17
Summary of Air Damage Case Findings
SITE
DESCRIPTION OF VIOLATION
TEST OF PROOF
REGULATING
AUTHORITY
Atlantic Cement
Company,
Ravena, NY
Opacity of kiln stack emissions
exceeded 20 percent for a total of
293 minutes during an observation
time of 60 minutes. The highest
opacity was 81 percent.
N.O.V.
October 21,1983
USEPA Region 2 issued
N.O.V.
•NY State Department of
Environmental
Conservation took
enforcement lead
Hercules Cement
Company
Stockertown, PA
State determined that emissions
from the baghouse dust disposal
area exceeded the limits of the
State's Air Pollution Control Rules
and Regulations'
N.O.V./Consent
Order
April 21, 1978
Pennsylvania Department
of Environmental
Resources
Keystone
Portland Cement
Bath, PA
Between May 22, 1979 and
February 1,1980, the State
observed and noti&ed the company
of various paniculate emissions,
fugitive paniculate emission, and
visible emission violations caused
by point and area air
contamination sources:
clinker discharge, rock dump, kiln
No. 2 waste dust tank, finish mills,
No. 2 cement kiln seal, dust dump,
raw material storage. No. 1 kiln
waste dust system and plant
roadway.b
N.O.WConsent
Order
August 27,1980
Pennsylvania Department
of Environmental
Resources '• .:.",:;..">•
Blue Circle,
Atlanta, GA
Excessive opacity from Kiln #1
expansion joint; 20 percent opacity.
"Probable" violation of emissions
standard at kiln baghouse exhausts.
Inspection resulted from citizen
complaints of paniculate matter
collecting on cars, swimming pool,
lawn chairs and other items outside
homes, originating at Blue Circle.
N.O.V.
August 10, 1990
N.O.V.
September 25,
1987
Georgia Department of
Environmental Resources
. Holnam, Inc.,
. Holly Hill, SC
Visible emissions exiting from
cement kiln #1 and #2 were
observed exceeding'the maximum
allowable State and Federal limit
of 20% opacity.
The clinker dust and/or emissions
from the baghouse were observed
exceeding the maximum allowable
10% opacity.
N.O.V.
July 16,1991
N.O.V.
July 11,1991
:South Carolina , :
Department of Health
:and Environmental
Control
-------
5-56
Exhibit 5-17 (continued)
Summary of Air Damage Case Findings
SITE
DESCRIPTION OF VIOLATION
TEST OF PROOF
REGULATING
AUTHORITY
Santee Cement
Company
(now Holnam,
Inc.)
Holly Hill, SC
Opacity excess emissions totalling
approximately 12% for the quarter
(14,011 minutes) due to
equipment/process malfunctions
associated with kiln #2.
Opacity excess emissions 13.2% of
the quarter (17,605 minutes) due to
equipment/process malfunctions
associated with kiln #2.
Opacity emissions in excess of 20%
were observed being emitted from
kiln #1 for more than six minutes in
a one hour period.
N.O.V.
May 15, 1990
N.O.V.
February 26, 1990
N.O.V.
August 10, 1989
South Carolina
Department of Health and
Environmental Control
Giant Cement
Harleyville, SC
In excess of opacity limits from the
stack serving kiln #4 and #5 and
from the clinker handling and
storage area.
N.O.V./Consent
Order
February 20, 1991
South Carolina
Department of Health and
Environmental Control
Lafarge
Corporation
Alpena, MI
Excessive visible emissions from
pugmill/pelletizer used to mix CKD
and water. This process was
observed in operation and visible
emissions readings were conducted
of the CKD pellets dropping off the
conveyor and onto the disposal pile.
76.67% opacity.
N.O.V.
August 5, 1991
Michigan Department of
Natural Resources
Lone Star
Industries
Cape Girardeau,
MO
Opacity was found to be in excess of
15% from clinker cooler and in
excess of 30% from the main kiln
stack. This was in violation of the
court settlement described below
between DNR and Lone Star.
Existing air pollution control
equipment was not of sufficient size
to handle periods of high dust
loading. Lone Star was violating
State's opacity regulation as well as
New Source Performance Standards.'
N.O.V.
February 4, 1991
Court Settlement
September 24, 1990
Missouri Attorney General
(Missouri Department of
Natural Resources)
Missouri Department of
Natural Resources
-------
5-57
Exhibit 5-17 (continued)
Summary of Air Damage Case Findings
SITE
Lone Star
Industries
Pryor, OK
Holnam, Inc.
Ada, OK
National Cement
Lebec, CA
Kaiser Cement
(now Mitsubishi
Cement)
Lucerne Valley,
CA
Calveras Cement
Co.
Monolith, CA
DESCRIPTION OF VIOLATION
A sizable accumulation of baghouse
waste dust was present on the
property outside of the building. In
violation of Oklahoma air pollution
control regulations governing
fugitive dust.
Excessive particulate emissions from
kiln dust storage area blowing off of
plant property.
Excessive emissions from kiln
baghouse. Emissions ranged from
40 to 100% opacity.
Particulate emissions from baghouse
controlling kiln emissions, above
allowable limits.
Excessive dust from ductwork
carrying gases from kiln to
baghouse; grey plume 20% -40%
opacity.
Excessive emissions from chute to
kiln baghouse; 60%-100% opacity.
Excessive emissions from dust
collection bin west of rotary kiln;
35% -50% opacity.
TEST OF PROOF
N.O.V.
October 3, 1990
N.O.V.
July 23, 1991
N.O.V.
October 7, 1992
N.O.V.
September 23, 1987
N.O.V.
August 5, 1992
N.O.V.
February 7, 1992
N.O.V.
August 22, 1991
REGULATING
AUTHORITY
Oklahoma State
Department of Health
Oklahoma State
Department of Health
Kern County, California
Air Pollution Control
District
San Bernadino, California .
Air Pollution Control :
District
San Joaquin Unified
Air Pollution Control
District
(Kern Co., CA)
* Hercules agreed to install air pollution control equipment to eliminate dust emissions from the baghouse area.
b Keystone Portland agreed to take corrective measures to keep the above-described emissions to a minimum.
* Settlement between Missouri Department of Natural Resources, the State Attorney General, and Lone Star. In
order to satisfy the regulations and ensure they could meet the regulation, Lone Star undertook an agreement to reduce
emissions by 100 tons per year through the installation of new air pollution control equipment. The agreement provided for
a 30 percent opacity limit until new air pollution control equipment was installed.
EPA also identified 50 citizen complaint forms from the files of three states aimed at
seven different cement kiln plants. In the case of the Blue Circle Cement plant in Atlanta,
Georgia, such complaints resulted in an NOV (Exhibit 5-17). In this case, a number of citizens
in the vicinity of the cement plant complained of particulate matter originating at the plant,
collecting on their cars, lawn chairs, window sills and other items located outside of their homes.
-------
5-58
Although at the time of the inspection, the opacity of the plume did not appear excessive,
considering the large exhaust area of the baghouse monitors to the atmosphere, State officials
concluded that mass emissions probably exceeded Georgia Air Quality Rules. Generally, the
other citizen complaints were similar in nature to those received for Blue Circle Cement.
In addition, nine citizens complained of respiratory problems believed to be associated
with emissions originating from the cement kiln plant. The health complaints were
unsubstantiated, however.
5.5 CKD MANAGEMENT SCENARIOS OF CONCERN
There are a few CKD management scenarios which may pose a high calculated risk under
specific reasonable worst case conditions. These situations are highlighted below. The risks
associated with these scenarios are described in more detail in Chapter 6. While they are
believed to be relatively infrequent, they are, nevertheless, plausible given the range of observed
concentrations of constituents in CKD.
In particular, disposal of CKD in exposed, unlined piles that are adjacent to actively tilled
agricultural fields may present higher risks (Exhibit 5-18). Analysis shows there is a greater
potential for risk through the foodchain from the ingestion of vegetables, meat, milk, and soil
contaminated by arsenic and dioxins through atmospheric deposition of CKD from nearby piles.
The close proximity of an active agricultural field to the exposed CKD pile has been observed
twice in the course of EPA site visits.
The Agency is also concerned about the practice of management of CKD underwater,
and in quarries, in particular. CKD disposal in a quarry that later filled with water is a
prominent factor in two cases of documented damage, one of which is a National Priorities List
Superfund site. Investigations at these sites noted that CKD-contaminated waters were likely
sources of contamination of surrounding surface waters and groundwaters.
Although the Agency's calculated risk associated with the management of CKD under
water is low, the Agency did not assume karst topography (an irregular topography with sinks,
underground streams, and caverns) when it modeled CKD management underwater in quarries.
This risk could be higher in scenarios where CKD is managed in areas with limestone bedrock
and karst topography. Cavernous limestones are highly jointed and fractured and can conduct
large volumes of groundwater rapidly for significant distances. Water-CKD mixtures migrating
through cavernous limestones can enter shallow groundwater bodies with little or no attenuation,
exposing to risk all nearby population that may drink the water and degrading the environmental
quality of nearby groundwaters and surface waters.
-------
5-59
Exhibit 5-18
Example of CKD Disposal Adjacent to an Agricultural Field
-------
5-60
CHAPTER FIVE
DOCUMENTED DAMAGES FROM MANAGEMENT OF CKD
5.0 INTRODUCTION AND METHODOLOGY 1
"Tests of Proof 1
Identification of Prospective Damage Cases 2
Information Collection 2
Damage Case Preparation and Review 3
Limitations of the Damage Cases 3
5.1 OVERVIEW OF FINDINGS, TRENDS, AND CONCLUSIONS 4
5.1.1 Findings 4
5.1.2 Overall Trends and Conclusions 5
5.2 DOCUMENTED GROUND AND SURFACE WATER DAMAGE CASE
SUMMARIES 9
Cases of Documented Damage 9
5.2.1 Holnam Incorporated, Mason City, Iowa 9
5.2.2 Lehigh Portland Cement Company, Leeds, Alabama 14
5.2.3 Lehigh Portland Cement Company, Mason City, Iowa 17
5.2.4 Portland Cement Company, Salt Lake City, Utah 22
5.2.5 Southwestern Portland Cement (Southdown, Inc.), Fairborn, Ohio 26
5.2.6 National Gypsum CoVLafarge Corp., Alpena, Michigan 32
5.2.7 Ash Grove Cement West, Montana City, Montana 34
5.3 CASES OF POTENTIAL DAMAGE TO GROUND AND SURFACE
WATER 37
5.3.1 Texas Industries, Inc., Midlothian, Texas 37
5.3.2 Holnam, Inc., Artesia, Mississippi 40
5.3.3 Markey Machinery Property, Seattle, Washington 41
5.4 DOCUMENTED AIR DAMAGES 45
5.5 CKD MANAGEMENT SCENARIOS OF CONCERN 49
-------
5-61
LIST OF EXHIBITS
Exhibit 5-1
Summary of Cases of Documented and Potential Damage to
Human Health and/or Environmental 4
Exhibit 5-2
Summary of Documented Water Damages 6
Exhibit 5-3
Cases of Potential Damage 8
Exhibit 5-4
Site Diagram - Holnam Incorporated, Mason City, Iowa 10
Exhibit 5-5
Site Diagram - Lehigh Portland Cement Company, Leeds, Alabama 15
Exhibit 5-6
Site Diagram - Lehigh Portland Cement Company, Mason City, Iowa 18
Exhibit 5-7
Site Diagram - Lime Creek Nature Center (Lehigh Portland Cement), Mason City, Iowa 19
Exhibit 5-8
Site Diagram - Portland Cement Company, Salt Lake City, Utah 24
Exhibit 5-9
Site Diagram - Southwestern Portland Cement, Fairborn, Ohio 28
Exhibit 5-10
Summary of Exceedances' of State Metals Limits
Southwestern Portland Cement - Fairborn, Ohio
Landfill #6 29
Exhibit 5-11
Site Diagram - National Gypsum Co./LaFarge Corp., Alpena, Michigan 33
Exhibit 5-12
Summary of Exceedances of State Metals Limits
National Gypsum Co./Lafarge Corp., Alpena, Michigan 34
Exhibit 5-13
Site Diagram - Ash Grove Cement West, Montana City, Montana 36
Exhibit 5-14
Site Diagram - Texas Industries, Inc., Midlothian, Texas 38
Exhibit 5-15
Site Diagram - Holnam, Inc., Artesia, Mississippi 42
Exhibit 5-16
Site Diagram - Markey Machinery Property, Seattle, Washington 43
Exhibit 5-17
Summary of Air Damage Case Findings . 46
Exhibit 5-18
Example of CKD Disposal Adjacent to an Agricultural Field 50
-------
CHAPTER SIX
POTENTIAL DANGER TO HUMAN HEALTH AND THE ENVIRONMENT
6.0 INTRODUCTION 6-1
Purpose and Scope 6-1
Overview of Approach 6-2
Major Results and Conclusions 6-4
6.1 INITIAL RISK SCREENING 6-6
6.1.1 Approach and Methods 6-6
CKD Composition Data 6-6
Risk-Screening Criteria 6-7
Other Constituent-SpeciGc Factors 6-10
6.12 Risk-Screening Results 6-11
6.2 EVALUATION OF RISKS WHEN CKD IS MANAGED ON SITE 6-14
6.2.1 Risk Potential Ranking of Initial Case Studies 6-15
Approach and Methods 6-15
Results of Risk Potential Ranking 6-19
Risk Potential Ranking for the Ground-water Pathway 6-20
Risk Potential Ranking for the Surface Water Pathway 6-23
Risk Potential Ranking for the Air Pathway 6-26
6.2.2 Risk Modeling of On-site CKD Management 6-29
Analytical Methodology 6-29
Release. Fate, and Transport Modeling Methodology 6-32
Characterization of Exposed Populations 6-34
-------
Exposure Assessment and Risk Characterization 6-35
Sensitivity Analysis of Higher Risk Potential Scenarios 6-38
Results of On-site Risk Modeling 6-40
Baseline On-site CKD Management 6-40
Sensitivity Analysis of Potentially Higher Risk Scenarios 6-45
6.2.3 Summary of Risks from On-site CKD Management 6-51
Ground-Water Risks 6-51
Surface Water Risks to Human Health 6-52
Aquatic Ecological Risks 6-52
Air Pathway Risks from Windblown Dust 6-52
6J EVALUATION OF RISKS FROM OFF-SITE BENEFICIAL USES OF CKD 6-53
6.3.1 Approach and Methods 6-54
6.3.2 Hazardous Waste Stabilization and Disposal 6-54
6JJ Sewage Sludge Treatment and Use 6-56
6.3.4 Building Materials Addition 6-57
6.3.5 Road Construction 6-58
Analysis of Risk Factors 6-58
Risk Modeling Results for Unpaved Traffic Surfaces 6-59
6.3.6 Agricultural Liming 6-60
Analysis of Risk Factors 6-60
Risk Modeling Results for Liming 6-62
6.3.7 Summary of Risks from Off-site Beneficial Uses 6-63
-------
LIST OF EXHIBITS
Exhibit 6-1
Exhibit 6-2
Exhibit 6-3
Exhibit 6-4
Exhibit 6-5
Exhibit 6-6
Exhibit 6-7
Exhibit 6-8
Exhibit 6-9
Exhibit 6-10
Exhibit 6-11
Exhibit 6-12
Exhibit 6-13
Exhibit 6-14
Exhibit 6-15
Exhibit 6-16
Exhibit 6-17
Exhibit 6-18
Overview of Risk Assessment Methodology
Basis for Risk-Screening Criteria
CKD Constituents That Exceeded Risk-Screening Criteria at EPA Sample
Facilities
Comparison of 15 Sample Facilities to Other Facilities
Site Specific Factors Used to Evaluate Risk Potential of On-Site CKD
Management
Risk Potential Rankings for the Ground-water Pathway . .
Risk Potential Rankings for the Surface Water Pathway . .
Particle Size Distribution of CKD by Kiln Type
Risk Potential Rankings for the Air Pathway
Graphical Illustration of On-site Risk Modeling Scenarios
Aquatic Ecological Benchmark Levels
Baseline On-Site Management Cancer Risks for Direct Exposure
Pathways for 15 Case Study Facilities
Baseline On-Site Management Cancer Risks for Foodchain Exposure
Pathways for 15 Case Study Facilities
Constituents Contributing to Adverse Health Effects In On-site CKD Risk
Modeling Analysis
Results of Central Tendency and High End Ecological Effects Analysis . . .
Sensitivity Analysis of Maximum CDD/CDF Cancer Risks for Direct
Exposure Pathways
Sensitivity Analysis of Maximum CDD/CDF Cancer Risks for Foodchain
Exposure Pathways
Sensitivity Analysis of Location Adjacent to Agricultural Field for
Foodchain Exposure Pathways
. 6-3
6-9
6-12
6-16
6-18
6-21
6-24
6-27
6-28
6-31
6-38
6-41
6-42
6-43
6-45
6-46
6-47
6-48
-------
LIST OF EXHIBITS (cent.)
Exhibit 6-19 Sensitivity Analysis of Location Adjacent to Surface Water for Direct and
Foodchain Exposure Pathways 6-49
Exhibit 6-20 Sensitivity Analysis of Highly Exposed Individuals for Foodchain Exposure
Pathways 6-50
Exhibit 6-21 Off-site Beneficial Uses Examined in the Risk Assessment 6-54
-------
CHAPTER SIX
POTENTIAL DANGER TO HUMAN HEALTH AND THE ENVIRONMENT
6.0 INTRODUCTION
Section 8002(o)(3) of RCRA requires that EPA's study of CKD waste analyze potential
danger to human health and the environment from disposal. In response to this requirement,
EPA assessed the risks of potential releases of CKD contaminants to the environment, both
during the routine management of the dust at cement plants and when the dust is beneficially
used at other locations. This assessment relies heavily on the information developed on the
amounts and characteristics of CKD generated (discussed in Chapter 3), CKD management
practices (discussed in Chapter 4), and alternative CKD management practices and uses
(summarized in Chapter 8). In addition, the risk assessment is intended to complement the
damage case study presented in Chapter 5. The damage cases provide actual instances of
environmental contamination, sometimes attributable to management practices and facility
settings not considered in the risk assessment. The risk assessment covers the potential for
certain more subtle or long-term risks that might not be evidenced in the damage case files.
This chapter summarizes the methods and results of EPA's risk assessment of CKD
disposal and use. Additional details on various aspects of the study are provided in Technical
Background Document, Human Health and Environmental Risk Assessment in Support of the
Report to Congress on CKD Waste (referred to as the "Risk Assessment Technical Background
Document" in the rest of this chapter). Before presenting the specific elements of the study, this
section provides background on the purpose and scope of the risk assessment, as well as an
overview of the study approach. This introduction also summarizes the major results and
conclusions that are developed in greater detail in the remainder of the chapter.
Purpose and Scope
One of the primary objectives of the risk assessment was to investigate, as realistically as
possible, the baseline risks of CKD management practices at actual sites. This was accomplished
by focusing on a sample of case-study cement plants and off-site beneficial use scenarios that
appeared to reasonably represent the universe of sites where CKD is disposed and used. For
each sample site, EPA evaluated the potential for CKD contaminants to be released into the
environment, migrate to possible human and ecological receptors, and result in exposures and
adverse effects. This evaluation included a combination of qualitative analyses designed to
document and describe major factors contributing to (or limiting) risks, and quantitative
modeling designed to estimate the magnitude of risks. The study focused on the potential for
releases and exposures through all media and pathways (ground water, surface water, air, and the
food chain), and examined risks both to maximally exposed individuals and total populations
around each case-study site.
Recognizing that potentially higher risk conditions may exist at other sites not included in
the case-study sample, EPA designed the study to evaluate potential adverse effects under a
variety of hypothetical scenarios. These scenarios were constructed by modifying the conditions
evaluated at the case-study sites to reflect a reasonable worst-case set of waste characteristics,
environmental settings, or CKD management practices.
-------
6-2
Overall, the study examines the range of conditions that exist across the industry, while
also focusing on those scenarios that have the greatest potential for adverse effects. The case
studies are believed to fairly represent the range of risks that exist at "typical" sites. At the same
time, to characterize the upper end of the risk distribution, priority was given to identifying and
evaluating those management scenarios that pose the greatest threat.
Overview of Approach
The risk assessment approach consisted of three primary steps, as shown in Exhibit 6-1.
First, the Agency conducted an "initial risk screening" of the chemical concentrations in CKD.
Using EPA's sampling data for 20 cement plants, as well as data provided by industry, this
screening compared chemical concentrations to a set of criteria. Concentrations that fell below
these screening criteria were judged to pose a low or negligible risk that did not need further
study. Conversely, concentrations above the criteria indicated that more detailed study was
needed to determine the risks associated with certain CKD constituents, exposure pathways, and
facility-specific waste streams under more realistic management conditions. This initial risk
screening is summarized in Section 6.1.
Second, those constituents, exposure pathways, and CKD waste streams that could not be
ruled out based on the initial risk screening were evaluated at a sample of actual cement plants.
For each of the 15 plants visited during the 1992 sampling study, EPA collected site-specific data
on a number of management practices and environmental factors that influence the potential for
damage through releases to ground water, surface water, and air when the dust is managed on
site at cement plants. Based on an analysis of these factors, the facilities were grouped into risk
potential categories (negligible, low, moderate, and high) for each pathway. The Agency then
performed quantitative modeling to estimate the human health and environmental risks at five of
these 15 plants in order to estimate both central tendency and high end risks. In addition, the
sensitivity of these modeled risk results to selected key parameters was examined in order to
identify potentially higher risk management scenarios and environmental settings not captured by
the 15 sample sites. Section 6.2 summarizes this evaluation of risks when CKD is managed on
site at cement plants.
Third, those constituents, exposure pathways, and CKD waste streams that the initial risk
screening could not exclude from further study were evaluated in the context of off-site beneficial
uses. The Agency reviewed data on the nature, extent, and location of off-site CKD uses to
identify five case studies for further risk analysis. These cases represented five major categories
of off-site use: 1) hazardous waste stabilization and disposal, 2) sewage sludge stabilization and
use, 3) building materials addition, 4) road construction, and 5) agricultural liming. EPA
collected data on major risk factors for each case study to determine the potential for adverse
effects and to prioritize the beneficial use categories for quantitative modeling. Hypothetical
scenarios designed to represent the two categories that appeared to pose the highest risk were
then developed and modeled for the purpose of risk estimation. This analysis of off-site
beneficial uses is presented in Section 6.3.
-------
6-3
Exhibit 6-1
Overview of Risk Assessment Methodology
-------
6-4
Major Results and Conclusions
Major results and conclusions from the evaluation of potential danger to human health
and the environment from the management of CKD are presented below.
• The pH of CKD leachate measured in laboratory tests typically ranged from 11 to
13. High pH levels in ground water and surface water may result in a variety of
adverse effects, including the mobilization of certain metals and other constituents
that could pose lexicological problems, human tissue burns (at pH levels above
12.5 or more), corrosion in pipes, and objectionable taste in drinking water. In
addition, high pH levels could cause a wide variety of adverse ecological effects.
• Seventeen radionuclides were found in detectable concentrations in CKD,
including members of the naturally occurring uranium-238 and thorium-232 decay
chains and anthropogenic radionuclides that have been dispersed throughout the
environment along with fallout from nuclear weapons tests. The concentrations of
these radionuclides in CKD, however, are not elevated compared to the range of
natural background levels, and modeling results for those nuclides with the highest
potential for adverse health effects showed negligible risk.
• Based on a detailed qualitative review of site-specific risk factors at 15
representative cement plants, on-site CKD handling and disposal does not appear
to have a high potential for adverse human health and environmental risks.
However, selected risk factors, observed or reported at these or other cement
plants, required more detailed qualitative evaluations.
• Quantitative risk modeling of case-study plants yielded central tendency risk
estimates for cancer and noncancer health effects that were below levels of
concern. Of the seven potential exposure pathways examined in this baseline
analysis, including direct contact and indirect foodchain pathways, estimated
increased individual cancer risks never exceeded a level of 1x10"* (most pathway
risks never exceeded IxlO"8). The noncancer hazard estimates were always less
than one order of magnitude of the noncancer effects threshold.
• Modeling estimates of high end risks from on-site management indicated a greater
potential for human health effects. High end facility cancer risks due to
recreational exposures to surface water reached an upper bound value of 2xlO"s;
the ingestion of vegetables grown in agricultural fields contaminated by CKD
reached an upper bound cancer risk of 3x10"*, and consumption of recreationally-
caught fish reached an upper bound risk of 4xlO'5. The other high end direct and
indirect exposure pathway estimates were all less than 1x10"*.
• Although the central tendency results for the baseline risk modeling analysis
showed no exceedances of ambient water quality criteria or other aquatic
ecological benchmarks, the high end results indicated a potential for aquatic
ecological damages. The high end ecological risks reflect contributions of CKD
from overland run-off, atmospheric deposition, and ground-water discharge all
entering the receiving water body. While most of the high end results indicated
that aquatic ecological benchmarks would be exceeded by small amounts for most
constituents, two constituents (cadmium and chromium) exceeded benchmarks by
-------
6-5
more than two orders of magnitude and two others (arsenic and lead) exceeded
by a factor of ten or more.
The sensitivity analysis of hypothetical but plausible (based on conditions
infrequently observed) higher risk scenarios indicated a potential for more
significant human health threats in a number of scenarios. These analyses
indicated that the proximity to potential exposure points (such as agricultural
fields and surface water bodies), high end concentrations of individual toxic
constituents (such as dioxins, arsenic, or heavy metals), or the possible presence of
extreme exposure situations (such as subsistence food consumption), would be
major factors that could increase the potential for damages from CKD plants.
Dioxins/furans did not contribute substantially to cancer risks for either the
central tendency orTiigh end plants in the baseline case studies. Sensitivity
analysis, based on high end measured dioxin concentrations, also suggested
negligible or low risks in the direct exposure pathways. However, for indirect
foodchain pathways, high dioxin concentrations applied to base case plant settings
increased central tendency cancer risks to levels as high as IxlO"4 and high end
plant risks to as high as IxlO"3.
Sensitivity analyses indicated that, other factors being equal, CKD units located
adjacent to crop fields and pastures or surface water bodies (both settings having
been observed in field site visits) would increase general health and/or aquatic
resource damages by an order of magnitude or more over the base-case estimates.
Although subsistence level food consumption exposure patterns were not observed
in the field or otherwise reported to the Agency, sensitivity analyses incorporating
these extreme indirect foodchain exposure situations yielded the highest estimated
risks in the EPA studies. Although these subsistence consumption risks did not
exceed levels of concern for the central tendency base case plants, when combined
with any other high end risk factor, cancer risks typically exceeded IxlO"4 for
subsistence fish consumption and IxlO'5 for subsistence farming.
Off-site beneficial byproduct use of CKD as a stabilizing agent for hazardous
waste, sewage sludge stabilizer, road sub-base, asphalt additive, and additive for
building materials (e.g., concrete and masonry block) does not appear to pose
significant risks to human health or the environment. Although there is some
potential for releases of CKD contaminants and subsequent exposures when the
dust is used in the construction of unpaved roads and parking lots, modeling of a
parking lot scenario indicates that this risk should be small (predicted cancer risks
of IxlO"7 or lower and noncancer risks of at least two orders of magnitude below
effects thresholds for all potential exposure pathways).
Utilization of CKD as an agricultural liming agent appears to pose more of a risk
than other byproduct beneficial uses. The Agency's analysis indicated that cancer
risks and noncancer effects could exceed relevant levels of concern in the
foodchain pathway in several scenarios for those CKD sources with very high
concentrations of arsenic and dioxins. While best estimate risks indicated a
maximum exposure to a subsistence farmer of about 7x10"* due to arsenic, the
-------
6-6
upper bound risks in this exposure scenario reached a maximum of ZxlO"4 as a
result of dioxin exposures.
6.1 INITIAL RISK SCREENING
EPA started its risk assessment by comparing the concentrations of chemicals measured
in CKD to a set of benchmarks, or "risk-screening criteria." These criteria were developed using
accepted toxicity values and chemical release, transport, and exposure assumptions that represent
reasonable mismanagement scenarios when CKD is managed on site at cement plants. EPA first
compared chemical concentrations to the screening criteria to identify CKD constituents that
need further study to determine if there is a potential to pose a human health or environmental
risk when the dust is managed on site. The Agency then evaluated other chemical and physical
properties (i.e., mobility and persistence in the environment, and normal background
concentrations) that may tend to mitigate, intensify, or otherwise qualify the risks associated with
those CKD constituents found at levels above the screening criteria.
The purpose of this initial risk screening was threefold:
• To identify individual CKD constituents that may have the potential to pose risks,
and, if so, how pervasively across cement plants;
• To identify exposure pathways that are most likely to convey risks (ground water.
surface water, air, and direct contact); and
• To identify CKD waste and product streams on a facility-specific basis that may
have the potential to pose risks under reasonable mismanagement scenarios.
Those CKD constituents, exposure pathways, and CKD streams believed to pose a low or
negligible risk based on the results of the risk screening could be excluded from further analysis.
Conversely, those constituents, pathways, and CKD streams that could not be ruled out based on
this initial screening would warrant a closer, site-specific assessment. The Agency then
proceeded to analyze these constituents, pathways, and cement plants in more detail in
subsequent steps of the risk assessment.
The remainder of this section summarizes the methods and results of this initial risk
screening. More detail is provided in the Risk Assessment Technical Background Document.
Section 6.1.1 provides a brief overview of the risk screening approach and methods. Section 6.1.2
presents the risk-screening results for different exposure pathways, and discusses their
implications for subsequent steps in the risk assessment.
6.1.1 Approach and Methods
This section describes the CKD composition data, risk-screening criteria, and other
constituent-specific factors used in the initial risk screening.
CKD Composition Data
-------
6-7
For the purpose of the initial risk screen, EPA examined the concentrations of 25 dioxins
and furans, 14 metals, 17 radionuclides, fluoride, and pH.1 The screening focused primarily on
, concentrations measured during the Agency's 1992 and 1993 sampling study, introduced in
Chapter 1. EPA believes that it is appropriate to focus this risk screen on its own sampling data
(as opposed to data from the PCA Survey, PCA Reports, and Bureau of Mines) for three main
reasons:
• • EPA's data set is the only source of data on dioxins, furans, and radionuclides
(the other sources do not provide any data on these constituents);
• The Agency data can be related in all instances to specific waste management
practices and environmental settings for subsequent case-study purposes; and
• As discussed in Section 3.2.2 of this report, a statistical analysis indicates that the
vast majority of calculated mean concentrations for metals in the EPA sampling
data are not significantly different than the means from the other data sources.
Nevertheless, the Agency recognizes that the other data sources report higher concentrations of
some metals than observed in the EPA sampling, and that limiting this initial risk screening to
only the EPA sampling data might ignore some metal concentrations that would yield higher risk
conclusions. Therefore, the risk screen also considered the full range of metal concentrations
reported in the other data sources.
EPA's data set of constituent concentrations consists of a total of 45 CKD samples from
20 different cement plants, including ten facilities that burn hazardous waste as fuel and ten
facilities that do not burn hazardous waste. Not all samples were analyzed for every constituent,
however. Metals data (both totals and leach extract) are available for 15 facilities, and dioxins
data are available for 11 facilities (although only six facilities have leachate data). The number
of facilities for which radionuclide data are available ranges from seven to 20, depending on the
particular radionuclide and test type. For this analysis, EPA did not differentiate between the
"as generated" and "as managed" dust samples, but rather combined the sampling results (there
were 24 "as generated" samples and 21 "as managed" samples). Similarly, the results from TCLP
and SPLP extract analyses, discussed in Chapter 3, were not differentiated for the initial risk
screen.
Leachate extract analyses were conducted for dioxins, furans, and radionuclides at the six
cement plants examined in the 1993 sampling, but not at the 15 plants examined in 1992. The
Agency filled this data gap by estimating leachate concentrations of these constituents for the
1992 sampling results. In particular, EPA determined the median ratio of total concentrations to
leachate extract concentrations observed for each dioxin, furan, and radionuclide examined in the
1993 sampling, and then multiplied these ratios by the corresponding total concentrations
observed in 1992. These estimated leachate concentrations were then pooled with the measured
concentrations from 1993 for comparison to the risk-screening criteria.
Risk-Screening Criteria
1 EPA also measured the concentrations of chloride, total organic carbon, total cyanide, sulfate, and sulfide in
CKD totals analyses during the 1992 sampling. The Agency did not examine these chemicals in the risk screening,
however, as there are no accepted toxicity values on which to base screening criteria.
-------
6-8
Because this evaluation was intended to identify constituents, pathways, and CKD streams
that warrant further analysis and rule out those that present negligible risk, EPA designed the
screening criteria to be reasonably "conservative" to avoid false negative conclusions. That is, the
criteria are based on release, transport, and exposure assumptions that are more likely to indicate
risk than actual CKD management practices at cement plants.
Separate criteria were developed for four release and exposure pathways: ground water,
surface water, air, and on-site direct contact. For the ground-water pathway, the Agency used
two criteria to evaluate the potential for adverse health effects through drinking water exposures:
one based on the drinking water primary Maximum Contaminant Levels (MCLs) and the other
based on health-based levels (HBLs). Four criteria were used for surface water: two for
evaluating the potential for human exposure through drinking water (based on the same MCLs
and HBLs used for the ground-water criteria), one for evaluating the potential for aquatic
ecological effects (based on the Ambient Water Quality Criteria), and one for evaluating the
potential for human exposure through fish ingestion. One criterion was developed for the air
pathway and the on-site direct contact pathway. The basis for each of the risk-screening criteria
is summarized in Exhibit 6-2. The Risk Assessment Technical Background Document provides
more detail on the derivation of these criteria, as well as the numerical values used for the
different criteria.
An individual lifetime cancer risk of IxlO"5 was used as the basis for the screening criteria
for carcinogens, indicating that the chance of an individual contracting cancer over a 70-year
lifetime,2 as a result of the exposure being assessed, is approximately 1 in 100,000. This risk
level is consistent with EPA policy of selecting risk management targets between IxlO"4 and 1x10"*
(55 FR 8716; March 9, 1990). An individual cancer risk of IxlO"5 is appropriate for developing
screening criteria in this context because the total population exposed to CKD is relatively small,
and because using a lower target risk in conjunction with the conservative exposure assumptions
underlying the screening criteria would unnecessarily compound the conservatism of the criteria.
For example, assuming a 70-year exposure duration introduces substantial conservatism
compared to the 9-year average exposure duration assumed in most current generic risk
assessments (assuming a 9-year exposure would raise the screening criteria for carcinogens by a
factor of almost eight). Using a higher target risk would be inappropriate because the screening
analysis was designed to be reasonably conservative and to minimize false negatives.
To develop the ground-water and surface water pathway criteria, EPA used a dilution
and attenuation factor (DAF) to account for the decrease in concentration that occurs as
contaminants are released from a waste management unit, mix in the flow of ground water or
surface water, and migrate to a location where a person, plant, or animal might be exposed. A
DAF of 10 was used for the ground-water pathway and a DAF of 100 was used for the surface
water pathway (i.e., adverse effect levels were multiplied by 10 for the ground-water criteria and
by 100 for the surface water criteria). These are the same DAFs that EPA used in conducting a
similar risk-screening analysis in the Report to Congress on Special Wastes from Mineral
2 EPA assumed a 70-year exposure duration in developing the risk-screening criteria as one means of ensuring that
the criteria are conservative (i.e., to help avoid false negative conclusions in this step of the analysis). In the risk
modeling step of the analysis, EPA assumed an exposure duration of 9 years, which is the 50th percentile (median)
duration of occupancy at one residence (Exposure Factors Handbook, U.S. EPA Office of Health and Environmental
Assessment, EPA/600/8-89/043, July 1989). The Agency used 9 years in the risk modeling to develop a risk estimate
that is more realistic than the conservative risk potential conclusions from the initial risk screen.
-------
6-9
Processing.3 The Agency believes that these factors account for a minimal amount of dilution
and attenuation in ground water and surface water under reasonable CKD mismanagement
scenarios.
5 Report to Congress on Special Wastes from Mineral Processing, Volume II, Methods and Analyses, U.S. EPA
Office of Solid Waste, July 1990.
-------
6-10
Exhibit 6-2
Basis for Risk-Screening Criteria3
Screening Criterion
Major Underlying Assumptions and Parameters
Ground-
water
Pathway
lOx Primary MCL
The Primary Maximum Contaminant Levels (MCLs) established for drinking water supplies are designed to be protective of human health. Ten
times the primary MCL represents the constituent concentrations in CKD leachate that could result in an exceedance of the primary MCL (and
the risk of associated adverse human health effects) if the leachate is released and migrates in ground water to a downgradient drinking water
well with less than a 10-fold dilution. In the case of pH, the Agency used one standard unit above the upper bound of the secondary MCL
(equivalent to a factor of 10) because there is no primary MCL. The secondary MCL for pH is intended to limit corrosrvity and taste effects, not
necessarily adverse health effects.
lOx Health-Based
Level
The Agency developed health-based levels (HBLs) using chemical-specific lexicological values along with equations for calculating preliminary
remediation goals for ground water at Supcrfund sites. These levels assume that an adult directly ingests contaminated ground water and inhales
volatile contaminants from whole-house water use (such as from the shower or faucet). The HBLs are based on an individual lifetime cancer
risk of IxlO"5 for carcinogens and noncancer effect thresholds for noncarcinogens. The Agency multiplied these HBLs by 10 to develop criteria
that represent concentrations in CKD leachate that may pose health risks if leachate is released and migrates in ground water to a nearby
drinking water well with less than a 10-fold dilution.
lOOx Primary
MCL
These are the same MCLs used for the ground-water criteria, simply multiplied by 100 rather than 10 to account for greater dilution expected in
surface water.
lOOx Health-
Based Level
These are the same HBLs used in deriving the ground-water criteria, but multiplied by 100 instead of 10.
Surface
Water
Pathway
lOOx AWQC
When available, the Agency used chronic ambient water quality criteria (AWQC) for freshwater organisms. When AWQC were not available,
the Agency derived "AWQC-like" values by extrapolating lowest observed adverse effect levels for chronic exposures of freshwater organisms.
These criteria are designed to be protective of aquatic organisms (not humans), accounting for the potential for constituents to bioconcentrate
and cause adverse effects through food chain exposures.
Human Fish
Ingestion Health
Factor
The Agency developed human health screening criteria for contaminated fish ingestion using chemical-specific lexicological values and
bioconoenlration factors, along with equations for calculating exposure from the ingestion of contaminated fish at Superfund sites. The levels are
based on an individual lifetime cancer risk of 1x10'' for carcinogens and noncancer effect thresholds for noncarcinogens. The Agency multiplied
these levels by 100 to develop criteria that represent concentrations in CKD leachate that may pose human health risks if constituents are
released, migrate to a surface water with only a 100-fold dilution, and bioconcentrate in fish that are consumed by humans.
Air Release-Off-site Exposure
Pathway
These criteria represent concentrations that, if CKD is suspended in air and transported to a downwind receptor location, could lead to an
individual lifetime cancer risk of 1x10'' or an exceedance of a noncancer effect threshold. The underlying assumptions are that particulars from
a CKD pile are blown into the air by the wind, dispersed to a hypothetical "backyard gardener's' property located 230 meters (750 feet) away,
and deposited onto soil and vegetables at that point. The receptor is then assumed to be exposed to CKD contaminants via four routes: (1)
inhalation of particulates; (2) incidental ingestion of soil contaminated by airborne deposition of participates (i.e., inadvertent ingeslion of soils as
a result of normal mouthing of objects or hands); (3) ingestion of leafy vegetables contaminated by deposited particulates; and (4) for
radionuclides, exposure to direct radiation from the contaminated ground surface without any shielding.
On-siic Direct Contact
Pathway
These criteria are based on a highly conservative, hypothetical scenario in which an individual is assumed to live directly on uncovered CKD, and
over a lifetime, incidentally ingests the dust, inhales parliculates suspended into the air, inhales constituents that have volatilized from the dust,
and is exposed to direct radiation with no shielding. No dilution is taken into account; the exposed individual is assumed to live directly on
CKD, not CKD mixed with soil or any other material. The criteria are based on an individual lifetime cancer risk of 1x10' for carcinogens and
noncancer effect thresholds for noncarcinogens. The Agency calculated these levels using equations and parameters developed for calculating
preliminary remediation goals for soil at Superfund sites.
Sec the Risk Assessment Technical Background Document for the numerical values used for each criterion and more detail on their derivation.
-------
6-11
To develop appropriate screening concentrations for dioxins and furans, EPA followed
the methodology presented in Interim Procedures for Estimating Risks Associated with Exposures to
Mixtures of Chlorinated Dibenzo-p-dioxins and -Dibenzofurans (CDDs and CDFs), 1989 Update.
According to this methodology, concentrations of 2,3,7,8-substituted CDDs and CDFs (i.e.,
CDDs and CDFs with a chlorine substituted on the 2, 3, 7, and 8 carbon atoms) are converted to
equivalent concentrations of 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD), the most potent
carcinogen that has been evaluated by EPA. Equivalent concentrations of 2,3,7,8-TCDD for
each 2,3,7,8-substituted congener4 are calculated by multiplying the concentration of each
2,3,7,8-substituted congener by its respective toxicity equivalent factor (TEF). CDDs and CDFs
that do not have chlorine substitutions at the 2, 3, 7, and 8 carbons are assigned a TEF of zero.
After each congener is multiplied by its TEF, the concentrations for all the congeners are
summed to determine the 2,3,7,8-TCDD equivalent for the mixture.
Other Constituent-SpeciQc Factors
For those constituents found to exceed one of the risk-screening criteria, the Agency
evaluated three other constituent-specific factors that may affect the potential for human health
and environmental risks. These other factors were used to qualify the results of the criteria
comparisons, not as a basis for excluding constituents of potential concern from the analysis.
The values used in evaluating each of these factors are outlined in the Risk Assessment
Technical Background Document.
First, the Agency evaluated each constituent's mobility in ground water5 by examining its
soil-water partition coefficient (K,,), which reflects the tendency of a chemical to attach to soil.6
EPA evaluated this factor because, even though a constituent may exist in CKD leachate in
relatively high concentrations, it may pose little or no risk to off-site receptors if it migrates very
slowly in ground water.
Second, each constituent's persistence in the environment was evaluated. A constituent
that degrades rapidly may not pose a substantial risk, even if it exists in relatively high
concentrations. Many constituents present in CKD are elements that do not degrade in the
environment. However, EPA evaluated the half-life of dioxins in ground water as reported in the
U.S. Department of Energy's (DOE's) MEPAS database. The persistence of dioxins in air or
surface water was not evaluated, because the travel time in these media to a possible exposure
point is nearly instantaneous. For radionuclides, EPA used radioactive half-lives documented in
the Radiological Health Handbook (1970) published by the U.S. Public Health Service.
4 The term "congener" refers to any one member of the same chemical family. There are 75 congeners of
chlorinated dibenzo-p-dioxins; seven of these have chlorine substituted at the 2, 3, 7, and 8 carbons. Likewise, there
are 135 congeners of chlorinated dibenzofurans; ten of these have chlorine substituted at the 2, 3, 7, and 8 carbons.
f EPA assumed that all constituents would be mobile in surface water or air if released to these media.
* This partition coefficient, or Kj, represents the equilibrium ratio of a chemical adhering to soil that is present in
ground water. The Agency reviewed each constituent's K,, as developed by EPA's Office of Research and
Development (ORD) (documented in EPA's Corrective Action chemical database). If a value was not developed by
ORD, Kj values were selected from the Department of Energy's Chemical Data Bases for the Multimedia
Environmental Pollutant Assessment System (MEPAS). Both of these sources provide Revalues for different pH
categories, and EPA selected values from the highest pH category to best represent conditions that are most likely to
exist in CKD leachate.
-------
6-12
Third, EPA evaluated the normal background concentrations of radionuclides in the
environment. Most of the radionuclides detected in CKD are naturally occurring (such as
members of the uranium-238 and thorium-232 decay chains), while others are anthropogenic but
have become ubiquitous in the environment (such as cesium-137 and plutonium-238/239, which
exist essentially everywhere due to fallout from nuclear weapons tests). The Agency reviewed
background concentration data available in the literature and provided by DOE. If a
radionuclide was found to exist in CKD in concentrations within the normal range found in the
environment, it may not pose a risk that warrants special attention.
6.1.2 Risk-Screening Results
Although substantial variability was found in the concentrations of individual
contaminants at the 20 facilities sampled, all 20 facilities had one or more constituents that
exceeded the risk-screening criteria for every pathway. The constituents that exceeded screening
criteria at each facility are presented in Exhibit 6-3.7 (For additional detail, including the
magnitude of exceedances at each facility, see the Risk Assessment Technical Background
Document.) As shown, every facility had at least four constituents that exceeded the ground-
water pathway criteria, at least one constituent that exceeded the surface water pathway criteria,
and at least five constituents that exceeded the very conservative on-site direct contact criteria.
In addition, every facility tested for metals had CKD that exceeded the air release off-site
exposure criteria for at least one constituent.
Those facilities that burn hazardous waste as fuel are identified in Exhibit 6-3 with an
asterisk. For the most part, the facilities that bum hazardous waste as fuel had the same
constituents exceeding screening criteria by the same order of magnitude as the facilities that do
not burn hazardous waste. However, dioxin, lead, chromium, pH, and Tl-208 levels at hazardous
waste burners tended to exceed certain criteria by a slightly wider margin than at other facilities.
Conversely, thallium, Bi-214, Pb-214, and Ra-226 concentrations tended to exceed the criteria by
a slightly wider margin at facilities that do not burn hazardous waste.
In terms of the results for individual constituents, the initial risk screening suggests the
following:
• Ground Water. The constituents needing further study for ground water are
antimony, arsenic, thallium, and pH. Dioxins (2,3,7,8-TCDD equivalents), lead,
beryllium, and cadmium also exceeded risk screening criteria, but these
constituents are relatively immobile under the high pH conditions expected for
CKD leachate (they would be expected to migrate readily only at sites where
fractures or solution cavities exist in the subsurface). In addition, K-40, Ra-228,
and U-238 exceeded the screening criteria, but these radionuclides appear to be
present in CKD in concentrations that are within the range of background levels
found in normal rock and soil.
7 The absence of a chemical for a given facility in Exhibit 6-3 may be the result of a lack of data for that facility,
rather than the result of low chemical concentrations that fall below the screening criteria. Specifically, dioxins were
not analyzed at Facilities B, C, G, I, J, L, N, Q, and S. Metals were not analyzed at Facilities K, M, P, R, and T.
Radionuclide data also are not available for every facility.
-------
6-13
Exhibit 6-3
CKD Constituents That Exceeded Risk-Screening Criteria
at EPA Sample Facilities'
Facility
Facility A*
Facility B
" Facility C*
Facility D
Facility E
Facility F«
Facility G
Facility H*
Facility I*
Facility J
Facility K
Facility L
Facility M
Ground-water
Pathway1
Sb, As, Pb, K-40, pH
As, Pb, Tl, K-40, pH
As, Pb, K-40, pH
Sb, As, Tl, K-40, U-
238, 23,7,8-TCDD
equiv., pH
Sb, As, Tl, K-40, pH
Sb, As, Pb, K-40, U-
238, 23,7,8-TCDD
equiv., pH
Sb, As, Pb, Tl, K-40,
pH
As, Pb, K-40, U-238,
23,7,8-TCDD,
equiv., pH
Sb, As, Pb, Tl, Ra-
228, K-40, pH
Sb, As, Pb, Tl, Ra-
228, K-40, U-238,
PH
K-40, U-238, 2,3,7,8-
TCDD equiv.
Sb, As, Pb, Tl, K-40,
PH
K-40, U-238, 23,7,8-
TCDD equiv.
Surface Water Pathway
lOOx MCL or
100xHBLk
pH, K-40
K-40, pH
Pb, K-40, pH
Tl, U-238,
TCDD
equiv., pH
K-40, pH
Pb, K-40, pH
Pb, Tl, K-40,
pH
Pb, K-40, U-
238, pH,
23,7,8-
TCDD equiv.
K-40, pH
K-40, U-238,
pH
K-40, U-238
Tl, pH
U-238
lOOx AWQC*
Pb.pH
PH
Pb.pH
pH
PH
Pb, pH
Pb, pH
Pb, pH,
23,7,8,-
TCDD equiv.
Pb.pH
PH
pH
Fish
Ingestion*
Tl
Tl
Tl, 23,7,8-
TCDD equiv.
Tl
Tl, 23,7,8-
TCDD equiv.
Tl
Tl, 23,7,8-
TCDD equiv.
Tl
Tl
23,7,8-TCDD
equiv.
Tl
23,7,8-TCDD
equiv.
Air Release -
OfT-site
Exposure
Pathway*
As, Cr
As, Cr
As, Cr
As, Cr
As, Cr
As, Cr
As, Cr
As, Cr, 23,7,8-
TCDD equiv.
As, Cr
As, Be, Cd, Cr
As, Cr, Tl
Oo-site Direct
Contact Pathway
As, Pb, Bi-214, K-
40, Pb-214, Ra-226,
Ra-228, Tl-208
As, Be, Bi-214, Pb-
214, K-40, Ra-226,
Ra-228, Tl-208
Pb-214, K-40, Ra-
226, Ra-228, TI-20S
As, Tl, Bi-214, Pb-
214, K-40, Ra-226,
Ra-228, Tl-208,
23,7,8-TCDD
equiv.,
TCDD+TCDF
As, Pb-214, K-40,
Ra-226, Ra-228, Tl-
208
As, Pb, Pb-214, K-
40, Ra-226, Tl-208
As, Bi-214, Cs-137,
Pb-214, K-40, Ra-
226, Ra-228, Tl-208
As, Pb, Pb-214, K-
40, Ra-226, Ra-228,
Tl-208, 23,7,8-
TCDD equiv.,
TCDD+TCDF
As, Bi-214, Pb-214,
K-40, Ra-226, Ra-
228, Tl-208
As, Be, Bi-214, K-
40, Pb-214, Ra-226,
Ra-228, Ti-208
Bi-214, K-40, Pb-
214, Ra-226, Ra-
228, Tl-208
Tl, Bi-214, K-40,
Pb-214, Ra-226,
Ra-228, Tl-208
Bi-214, K-40, Pb-
214, Ra-226, Ra-
228, Tl-208
-------
6-14
Exhibit 6-3 (continued)
CKD Constituents That Exceeded Risk-Screening Criteria
at EPA Sample Facilities*
FacUity
Facility N*
Facility O*
Facility P«
Facility Q
Facility R"
Facility S*
Facility T
Ground-water
Pathway*
Sb, As, Pb, K-40, pH
Sb, As, Pb, Tl, Ra-
228, K-40, U-238,
23,7,8-TCDD equiv.,
PH
Ra-228, K-40, U-238,
23,7,8-TCDD equiv.,
PH
As, TI, K-40, pH
K-40, U-238, 23,7,8-
TCDD equiv., pH
Sb, As, Pb, Tl. K-40,
PH
Ra-228, K-40, U-238,
23,7,8-TCDD equiv.,
PH
Surface Water Pathway
lOOx MCL or
lOOxHBL'
Pb, K-40, pH
K-40, U-238,
PH
U-238, pH
Tl
K-40, U-238,
pH
pH, K:40
K-40, U-238,
PH
lOOx AWQC*
Pb,pH
Pb.pH
pH
Tl
pH
pH
Fish
Ingestion*
Tl
Tl, 23,7,8-
TCDD equiv.
23,7,8-TCDD
equiv.
Tl
23,7,8-TCDD
equiv.
Tl
23,7,8-TCDD
equiv.
Air Release -
Off-site
Exposure
Pathway*
As, Cr
As, Be, Cr
Th-228
As, Cr, Tl
As, Cr
Go-site Direct
Contact Pathway
As, Be, Pb, Bi-214,
Pb-214, K-40, Ra-
226, Ra-228, TI-208
As, Be, Pb, Bi-214,
K-40, Pb-212, Pb-
214. Ra-226, Ra-
228, TI-208
Bi-214, K-40, Pb-
214, Ra-226, Ra-
228, TI-208
As, Tl, Bi-214, Pb-
214, K-40, Ra-226,
Ra-228,.TI-208
Bi-214, K-40, Pb-
214, Pb-212, Ra-
226, Ra-228, TI-208
As, Bi-214, Pb-214.
K-40, Ra-226, Ra-
228, TI-208
Bi-214, K-40, Pb-
214, Ra-226, Ra-
228, TI-208
• Burns hazardous waste as fuel.
• Dioxins were not analyzed at Facilities B, C, G, I, J, L, N, Q, and S. Metals were not analyzed at Facilities K, M, P, R, and T.
Radionuclide data also are not available for every facility.
k Metals data reported by industry (not developed by EPA) indicate that, in addition to the above exceedances, beryllium and
cadmium occasionally exceed ground-water screening criteria, arsenic occasionally exceeds the HBL-based surface water criterion, and
mercury occasionally exceeds the AWQC-based and fish ingestion criteria. Because the identity of the facilities exceeding the criteria
for these constituents is not known, they could not be displayed in this exhibit.
Surface Water. Dioxins and furans (2,3,7,8-TCDD equivalents), lead, thallium,
arsenic, K-40, U-238, and pH need further study to determine their potential
drinking water threats. Dioxin, lead, thallium, mercury, and pH levels exceeded
the AWQC-based criteria and require further study to determine their potential
for aquatic ecological risk. Considering the potential for these constituents to
bioconcentrate in fish tissue, dioxins, thallium, and mercury could pose an added
threat of human exposures through the fish ingestion pathway. Of these
constituents, dioxins, lead, thallium, and mercury are relatively immobile in
ground water (if fractures or solution cavities that facilitate flow do not exist) and
thus would tend to migrate to surface water primarily by stormwater run-off or
-------
6-15
atmospheric deposition, rather than via ground-water discharge. In addition, the
surface water risks associated with K-40 and U-238 do not appear to be greater
than the risks associated with natural background radioactivity.
• Air. The constituents needing further study to determine airborne releases and
exposures include dioxins (2,3,7,8-TCDD equivalents), arsenic, beryllium,
cadmium, thallium, and chromium (conservatively assuming all of the chromium
in CKD is present in its more toxic hexavalent form). Th-228 also could pose a
risk via the air pathway, but no more than the risk associated with natural
background concentrations of this radionuclide.
• On-site Direct Contact. Dioxins, arsenic, beryllium, lead, thallium, and eight
radionuclides may be present at some facilities in concentrations that may be
harmful under the highly conservative scenario in which an individual lives directly
on uncovered CKD. Although the radionuclides may pose a risk under this
exposure scenario, this radiation threat should not be any greater than that
associated with natural background radioactivity.
Based on these screening results, EPA concluded that more detailed study was needed to
determine the risks of several CKD constituents, exposure pathways, and facility-specific waste
streams. The Agency proceeded to evaluate these risks more closely by examining existing
conditions at a sample of actual cement plants and off-site locations where CKD is beneficially
used.
6.2 EVALUATION OF RISKS WHEN CKD IS MANAGED ON SITE
In the second step of the risk assessment, EPA conducted a closer examination of the
cement plants and CKD constituents that were found to have the potential for risks in the initial
risk-screening. The results of the preceding analysis of constituent concentrations in CKD were
combined with a site-specific evaluation of CKD management practices and environmental
settings at a sample of actual cement plants.
This more detailed evaluation of risks was conducted in two phases. First, EPA
evaluated the "risk potential11 at initial case-study facilities by analyzing a number of site-specific
factors relating to the potential for on-site CKD management to pose risks via ground-water,
surface water, and air pathways. The purpose of this evaluation was to document and describe
the major factors contributing to or limiting risk at each case-study facility, and to prioritize the
facilities for further analysis through quantitative modeling. This evaluation of risk potential is
presented in Section 6.2.1.
Second, the Agency performed quantitative modeling to estimate the magnitude of risks
associated with on-site CKD management at cement plants. In particular, site-specific modeling
was performed to estimate the risks at case-study cement plants that could pose higher risks
based on the preceding evaluation of risk potential. The Agency also modeled potentially higher
risk scenarios not captured by the sample of cement plants considered in the evaluation of risk
potential. This risk modeling of on-site CKD management is presented in Section 6.2.2.
-------
6-16
6.2.1 Risk Potential Ranking of Initial Case Studies
This section summarizes the methods and results of the risk potential ranking conducted
by EPA to determine factors that strongly influence the risks of on-site CKD management and to
prioritize cement plants for risk modeling. The Risk Assessment Technical Background
Document provides more detail on this evaluation.
Approach and Methods
EPA focused this ranking on a subset of the constituent concentration data and 20
sample facilities analyzed in the initial risk screening. Only some of the constituents and
facilities were examined to develop an initial sample of case-study facilities that could be
evaluated on a "level playing field." In particular:
• Dioxin concentrations were not considered because only 11 of the 20 sample
facilities were analyzed for dioxins. Considering dioxins, therefore, would have
resulted in artificially high risk potential rankings for some facilities that are based
more on data availability than on true differences that exist across sites.
• The five cement plants sampled by EPA in 1993 were not considered. These
facilities were excluded from the risk potential ranking because their CKD was
not tested for metals, which could result in a bias in the ranking.
It is important to clarify that EPA excluded dioxins and the five facilities sampled in 1993 only
from this risk potential ranking and not from the rest of the risk assessment. As discussed in
Section 6.2.2, the Agency modeled the risks of dioxins under several actual and hypothetical
management scenarios, as well as potential higher-risk conditions found at some of the five
facilities sampled in 1993, but not observed in the sample of 15 facilities sampled in 1992.
EPA believes that it is reasonable to focus on the 15 cement plants sampled in 1992 as
initial case-study facilities because they appear to provide a representative sample of other
cement plants. Specifically, the sample is large and diverse, representing approximately 10
percent of the universe of existing U.S. cement plants as well as a diversity of fuel types, process
types, and geographic locations (e.g., eight of the facilities burn hazardous waste as fuel and
seven do not). Moreover, the sample of 15 cement plants compares favorably with the complete
set of 83 plants for which data are available, as shown in Exhibit 6-4. Specifically, the two sets of
facilities are quite similar in terms of a number of factors that influence risk, including CKD
management unit types, the size of CKD management units, the proximity to "sensitive"
environmental features (karst terrain, geological faults, 100-year floodplains, and endangered
species habitats), the number of residents presently within one mile, and the distance to the
nearest existing residence. The 15 sample facilities, however, generate relatively large volumes of
net CKD compared to the broader set of 83 plants, and do not represent the management of
CKD underwater (which is practiced at three of the 83 facilities). Finally, a statistical analysis
indicates that the concentrations of metals in CKD at these 15 facilities are similar to the
concentrations observed at other cement plants. As discussed in Section 3.2.2, most of the
calculated mean concentrations for metals at the 15 sample facilities are not significantly
different than the means from other data sources that cover a larger sample of facilities
(including PCA Report 2, which provides data on the concentration of metals in CKD from 79
cement plants).
-------
6-17
Exhibit 6-4
Comparison of 15 Sample Facilities to Other Facilities
Parameter
Total net CKD generated
CKD management unit
type
CKD managed
underwater?
Basal area of CKD
management unit(s)
Facility in karst area?
Facility in fault area?
Facility in 100-year
floodplain?
Facility in endangered
species habitat?
Number of residents
presently within one mile
of property boundary
Distance from property
boundary to nearest
existing off-site residence
Range of Values for 15 Sample
Facilities
25% >. 63,500 MT (70,000 tons)
50% >. 40,500 MT (45,000 tons)
75% >. 16,500 MT (18,200 tons)
60% landfill CKD in an on-site quarry
27% manage CKD in an above-grade
pile
13% (2 facilities) have no active CKD
unit
100% no
25% > 63,500 m2 (683,000 ft2)
75% >. 6,700 m2 (72,000 ft2)
80% no
20% yes
67% no
33% yes
47% no
53% yes
100% no
25% >. 1,020 people
75% >. 25 people
25% > 850 m (2,800 ft)
75% >. 15 m (50 ft)
Range of Values for All 83 Facilities for
Which Data are Available*
25% > 53,500 MT (59,000 tons)
50% >. 21,800 MT (24,000 tons)
75% > 1,100 MT (1,200 tons)
43% landfill CKD in an on-site quarry
40% manage CKD in an above-grade
pile
11% landfill CKD in other units (mines,
slopes)
1% (1 facility) manage CKD in a pond
4% use other management units
97% no
3% (3 facilities) yes
25% > 58,600 m2 (630,000 ftz)
75% >. 3,700 m2 (39,800 ft2)
85% no
15% yes
86% no
14% yes
60% no
40% yes
98% no
2% yes
25% >. 2,020 people
75% :> 33 people
25% >. 790 m (2,600 ft)
75% >. 30 ra (100 ft)
' A total of 79 cement plants, including 11 of 15 sample facilities, returned completed PCA mail survey
questionnaires. Comparable data for the other four sample facilities were developed during the sampling visits.
-------
6-18
For the sample of 15 cement plants, the Agency analyzed site-specific information on a
number of factors that determine the degree to which CKD constituents are likely to be released
into the environment and transported to locations where humans or ecological receptors could be
exposed. The particular factors considered are listed in Exhibit 6-5. As shown, EPA conducted
separate analyses of factors that relate to the potential for CKD management to pose risks via
the ground-water, surface water, and air pathways (including risks from the ingestion of food
contaminated through these different pathways). For each pathway, four sets of factors were
systematically considered at every site:
• Factors related to the intrinsic hazard of CKD. These factors included the
frequency and magnitude with which chemical concentrations and pH levels
exceeded the risk-screening criteria discussed in Section 6.1. Again, dioxins were
not considered in this step to avoid biasing the ranking toward the subset of
facilities whose CKD was analyzed for dioxins. In addition, EPA did not consider
immobile constituents in the ground-water pathway ranking, or radionuclides for
any pathway because they were all measured in CKD at levels that fall within the
range of typical background levels.8
• Factors related to ground-water, surface water, and air contamination potential.
These factors included CKD management practices (size of pile, presence of liners
and run-off controls, dust suppression practices, etc.) and environmental features
(e.g., depth to ground water, distance to surface water, and wind speeds) that have
a bearing on the potential for contaminants to migrate from waste management
units and contaminate environmental media.
• Factors related to transport potential. These factors included the presence of
natural and man-made barriers to contaminant migration in environmental media,
such as slurry walls or surface water bodies that might impede the migration of
ground-water contaminants, and karst terrain or fractures that may facilitate
contaminant migration in ground water. The distance to closest potential
receptors also was considered as a transport potential factor, giving the risk
potential ranking an element of a maximum exposed individual (MEI) risk
assessment.
• Factors related to exposure potential. These factors included the present human
uses of nearby ground water, surface water, and air, as well as the size of
potentially exposed populations. By considering the size of potentially exposed
populations, the ranking also included elements of a population risk assessment
Depending on the size of the population, this factor had the effect of moderating
or intensifying the risk potential ranking based on MEI distances alone.
* EPA believes that leaving immobile ground-water contaminants and radionuclides out of this ranking provides a
more realistic evaluation of risk potential. However, once a facility was selected for risk modeling based on this
ranking, all constituents that exceeded one of the risk-screening criteria were modeled.
-------
6-19
Exhibit 6-5
Site Specific Factors Used to Evaluate Risk Potential of Oil-Site CKD Management
-------
6-20
The Agency assembled site-specific values for each of these factors using, when available,
information collected during the site visits. When data were not available from the site visits, a
variety of sources were used to fill in data gaps, including the PCA mail survey, local offices of
State governments and the U.S. Geological Survey (USGS), USGS topographic maps, the
Graphical Exposure Modeling System (GEMS), and environmental data collected by EPA for
nearby facilities as part of other risk assessment projects.
For each pathway (ground water, surface water, and air), the various factors were
combined to develop rankings (negligible, low, moderate, and high) regarding intrinsic hazard,
contamination potential, transport potential, and exposure potential at each site. The Agency
then combined these four rankings to develop an overall ranking of the ground-water, surface
water, and air risk potential at each plant. In developing this overall ranking for the different
media, the lowest ranking was selected from among the scores assigned to intrinsic hazard,
contamination potential, transport potential, and exposure potential. For example, if the ground-
water pathway at a facility was assigned a low intrinsic hazard, a high contamination potential, a
moderate transport potential, and a moderate exposure potential, the facility was assigned an
overall low ground-water risk potential. In this way, the Agency evaluated the individual risk
factors to determine if there were any factors that would limit the potential for significant risk at
a given site. If a risk-limiting factor was identified (e.g., intrinsic hazard was low, as in the above
example), the overall risk for that pathway could not be high. Chapter 7 of the Risk Assessment
Technical Background Document describes this methodology in more detail, presents the
individual factors and criteria used to develop risk potential rankings, and documents the results
for each of the 15 case-study facilities.
In performing this ranking, EPA considered only the current conditions that exist at each
cement plant, such as the current CKD pile sizes and containment features, the current land and
water use practices in surrounding areas, and the current population distributions in off-site
areas. Insufficient data were available to support a meaningful analysis and prediction of
possible future conditions. However, significant changes in the current conditions at these 15
plants could result in some facilities being assigned higher or lower risk potential rankings.
Results of Risk Potential Ranking
The case-study site rankings represent best professional judgments on the potential for
current CKD management practices at the 15 sample plants to pose risks to human health and
the environment, based on the analysis of factors outlined above. The results provide a means of
evaluating the risk potential at each of the 15 sites relative to each other, not a definitive
assessment of the absolute risk at each site (e.g., a site ranking cannot be translated into a
numeric cancer or non-cancer risk estimate). Considering the rigor of the methodology, this
ranking provides a credible basis for prioritizing the sites and selecting plants, that warrant risk
modeling. At the same time, the results indicate the general level of risk expected to exist at
each site, based on the Agency's understanding of risk-influencing parameters and the results of
previous risk analyses and modeling projects. This is especially the case for sites that are
assigned a negligible risk potential, where one or more site factors allow the Agency to conclude,
with some certainty, that risks for a given release and exposure pathway are indeed sufficiently
low to be ignored. As previously discussed, available information indicates that the site
conditions and distribution of risk potential rankings across the sample of 15 plants reasonably
represents the larger universe of active cement plants, but may not reflect particularly high-risk
conditions or factors that have been discovered at the damage case sites or observed during site
visits.
-------
6-21
Risk Potential Ranking for the Ground-water Pathway
Exhibit 6-6 summarizes the risk potential rankings for the ground-water pathway at the 15
sample cement plants. These rankings address only the potential for human health risks through
drinking water ingestion, not the potential for health or ecological risks associated with the
discharge of contaminated ground water to a surface water body (which are considered in the
next section on surface water risk potential). As shown, the Agency developed separate hazard
potential rankings for each plant based on the intrinsic hazard of chemical concentrations and
pH levels in CKD leachate. The plants are ordered in the exhibit from highest to lowest ground-
water risk potential based on the concentrations of chemicals in CKD leachate. The risk
potential ranking of the plants considering pH levels is slightly different, as indicated by the
number in parentheses in the far right column.
Based on the results in Exhibit 6-6, none of the 15 facilities are expected to pose an
overall high ground-water risk. Although the Agency's methodology ranked certain factors in
isolation as having a high risk potential, the scores for these individual factors were moderated
when combined with the other factors that determined overall site risk. For example, even
though the potential for ground-water contamination was ranked high at Facility G, the overall
risk potential for the facility was ranked moderate considering the other factors (intrinsic hazard,
transport potential, and exposure potential) that influence risks at the site.
The Agency ranked four facilities as having an overall moderate risk potential for the
ground-water pathway, considering the chemical concentrations in CKD leachate. In order of
descending risk potential, these are Facilities G, A, C, and J. These same facilities also were
ranked among the top considering pH levels of the CKD leachate. Facilities A and C burn
hazardous waste as fuel, while Facilities G and J do not use hazardous waste as an alternative
fuel. The primary factors that contributed to these facilities being ranked relatively high
included:
• At Facility G, the potential for ground-water contamination appears high because,
among other factors, the water table is shallow (0.3 to 1 meter [1 to 3 feet]
beneath the CKD pile), the underlying soils are a permeable sand, and net
recharge is high (38 cm/year, or 15 in/year). However, the potential for ground-
water contamination to migrate to off-site drinking water wells and result in
significant exposures is only moderate because the nearest downgradient residence
is approximately 1,600 meters (one mile) from the CKD pile. Furthermore, local
water suppliers have stated that residences in the area derive their drinking water
from community water systems (although ground water is used for domestic
purposes in the area and the possibility of a private well at nearby residences
cannot be ruled out). The size of the population that may be exposed to any
ground-water contamination within a mile downgradient of the facility's CKD pile
is about 20 people.
• At Facility A, the contamination potential is not as high as at Facility G because
the material underlying the CKD pile is a less permeable limestone and siltstone
and because the net recharge is smaller (15 cm/year). As at Facility G, ground
water is used for domestic purposes in the area, but according to local water
suppliers, residences around Facility A derive their drinking water from a nearby
river. If any nearby residences do have private wells, the nearest downgradient
-------
6-22
residence that may be exposed to ground-water contamination is about 490 meters
-------
6-23
Exhibit 6-6
Risk Potential Rankings for the Ground-water Pathway
Facility
Facility G
Facility A*
Facility C*
Facility J
Facility D
Facility I*
Facility F*
Facility B
Facility S*
Facility O*
Facility H«
Facility N*
Facility E
Facility Q
Facility L
Intrinsic Hazard
Potential
Chemical
Moderate
Moderate
Moderate
Moderate
High
Low
Low
Low
Low
Moderate
Moderate
Moderate
Low
Moderate
Moderate
pH
High
Moderat
e
Moderat
e
High
High
High
Moderat
e
Moderat
e
High
High
Moderat
e
Moderat
e
Moderat
e
Moderat
e
High
Ground-water
Contain mation
Potential
High
Moderate
High
Moderate
Low
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Negligible
Negligible
Transport
Potential
Moderate
Moderate
Moderate
Moderate
High
Moderate
Moderate
High
Moderate
Low
Low
Low
Low
Negligible
Negligible
Current
Exposure
Potential*
Moderate
High
Moderate
Moderate
High
High
High
Moderate
Moderate
Negligible
Negligible
Negligible
Negligible
Negligible
Negligible
Overall Ground-water Risk
Potential (Rank)
Chemical
Moderate (1)
Moderate (2)
Moderate (3)
Moderate (4)
Low (5)
Low (6)
Low (7)
Low (8)
Low (9)
Negligible (10)
Negligible (11)
Negligible (12)
Negligible (13)
Negligible (14)
Negligible (15)
PH
Moderate (2)
Moderate (3)
Moderate (5)
Moderate (4)
Low (9)
Moderate (1)
Moderate (6)
Moderate (8)
Moderate (7)
Negligible (10)
Negligible (11)
Negligible (13)
Negligible (12)
Negligible (15)
Negligible (14)
• Burns hazardous waste as fuel.
• Future development of ground-water uses around these facilities could increase the exposure potential rankings and, depending on
the risk rankings for the other site factors (intrinsic hazard, ground-water contamination potential, and transport potential), could
result in higher overall ground-water risk potential rankings.
(1,600 feet) from the CKD pile and the total population within a mile
downgradient is 450 people, larger than the potentially exposed population at
Facility G.
At Facility C, there appears to be a high potential to contaminate ground water
because the water exists just three meters below the CKD pile, the unsaturated
zone is moderately permeable (a clayey sand), and net recharge is high (33
cm/year). Although ground water is used as a drinking water source in the area,
-------
6-24
the nearest downgradient residence that may be affected is farther away from the
CKD pile than at Facilities G and A (1,100 meters). Additionally, the only
residence that might be affected by any ground-water contamination is located on
site, in between the CKD pile and a large river, which borders the site. All other
residences in the direction of ground-water flow are on the other side of the river
and are unlikely to be exposed to any ground-water contamination originating
from Facility C.
• At Facility J, the ground-water contamination potential appears moderate because
the water table is moderately deep (9 meters), the net recharge is moderate (20
cm/year), and the permeability of the shale underlying the site's CKD pile is low.
Ground water is presently used in the area for domestic purposes, and the nearest
downgradient residence that may have a private well is roughly 550 meters from
the CKD pile. There are approximately 40 people within a mile downgradient
that may be exposed to any ground-water contamination originating from the pile.
Five facilities were ranked as having an overall low ground-water risk potential. All of
these facilities were ranked as low because one or more critical factors that determine overall site
risk potential were scored low according to the Agency's ranking methodology. For example, the
intrinsic hazard of the chemical concentrations in CKD leachate at Facilities I, F, B, and S is low,
making the overall ground-water risk potential low at those sites regardless of the ground-water
contamination, transport, and exposure potential. Similarly, even though the intrinsic hazard of
the dust at Facility D is ranked high, the overall ground-water risk potential at the site appears
low because of the low potential for ground-water contamination at the site (the water table is
about 30 meters deep, net recharge is very low, and the underlying clay and shale is very
impermeable).
Six facilities were ranked as having an overall negligible ground-water risk potential. Two
of these facilities, Q and L, were assigned a negligible hazard because they presently recycle all
of their CKD and do not have an on-site CKD management unit. The other facilities were
assigned a negligible hazard because there is a negligible potential for exposure to any ground-
water contamination that might originate from on-site CKD management. In particular:
• All ground water at Facility O discharges directly into the site's quarry (ground
water is pumped at the site to dewater the quarry). Even after mining operations
cease and ground-water contours are allowed to return to normal, any ground-
water contamination originating from the plant's CKD pile would migrate just 150
meters to the northern property boundary where it would discharge directly into a
surface water body without being withdrawn for human use.
• If ground water beneath the CKD pile at Facility H were to become
contaminated, it would likely discharge directly into a river with a large dilution
potential located 1,200 meters downgradient. All of the property between the pile
and the river is owned by Facility H and presently uninhabited.
• There presently are no residences within a 1,600 meters downgradient from the
CKD pile at Facility N. Also, the nearest downgradient property boundary where
off-site exposures could occur is relatively far (1,400 meters) from the CKD pile.
-------
6-25
• Any ground-water contamination originating from the CKD pile at Facility E is
expected to discharge directly to a river with a large dilution potential 370 meters
downgradient without being withdrawn for human use (all of the property
between the pile and the river is owned by Facility E and presently uninhabited).
Even if all the contamination did not discharge into the river, the closest
downgradient residence that could be exposed to the contamination is
approximately 2,300 meters away.
Risk Potential Ranking for the Surface Water Pathway
Exhibit 6-7 summarizes the risk potential rankings of the 15 case-study cement plants for
the surface water pathway. These rankings address the potential for human health risk via
drinking water, fish ingestion, and other surface water uses, as well as the potential for risk to
aquatic organisms. As for the ground-water pathway, the Agency developed two separate
rankings, one considering only the concentrations of chemicals (not pH levels) in CKD leachate
and the other considering both the concentrations of chemicals and pH levels. Contamination
potential scores were developed by considering three contaminant migration pathways:
stormwater run-off to surface water, ground water to surface water migration, and air deposition
to surface water. The highest score from among these three scores at a given facility was
selected as that facility's surface water contamination potential. The overall surface water risk
ranking at a site was determined by selecting the lowest score for any of the critical factors at
that site (i.e., intrinsic hazard, contamination potential, transport potential, and current exposure
potential). The plants are ordered in Exhibit 6-7 from highest to lowest overall risk potential
considering the concentrations of chemicals in CKD leachate. The alternate ranking considering
pH levels is indicated in the far right column of the exhibit.
Based on these results, none of the 15 facilities are expected to present a high risk to
human health and aquatic organisms via the surface water pathway. As discussed for the
ground-water pathway, the Agency's methodology ranked some facilities high for one or more
aspects, but at each facility, at least one critical factor lowered the overall risk potential. For
example, Facility F scored high for transport potential, but received an overall moderate risk
ranking when the other factors were considered.
As shown in Exhibit 6-7, EPA ranked seven facilities as having a moderate surface water
risk potential. Five of these seven facilities (O, F, A, I, and N) burn hazardous waste as fuel.
The main factors that contributed to these rankings include:
• At Facility O, CKD could blow into the air and deposit in a water body with a
large surface area just 150 meters to the north. Additionally, after current
ground-water pumping to dewater the quarry ceases, any ground-water
contamination originating from the on-site CKD pile would be expected to
migrate 150 meters to the north and discharge into the same water body. Such
contamination, including possible increases in pH levels in affected areas, has the
potential to cause ecological damage, but would not be expected to pose a human
drinking water threat because the water is not used for drinking. The potential
for surface water contamination via stormwater run-off appears low, given surface
drainage patterns and ditches that divert run-off from the CKD pile into the
quarry, through a series of settling ponds, and eventually out to the surface water
body through an NPDES-permitted outfall.
-------
6-26
At Facility J, there is a potential for stormwater run-off carrying contaminants
from the on-site CKD pile to migrate approximately 2,100 meters through a
drainage ditch and discharge into a reservoir. Given the pile's containment
features and the site's hydrogeology and meteorology, there also is a potential for
CKD contaminants to migrate via ground-water discharge and airborne deposition
-------
6-27
Exhibit 6-7
Risk Potential Rankings for the Surface Water Pathway
Facility
Facility O*
Facility J
Facility F*
Facility D
Facility A*
Facility P
Facility N"
Facility G
Facility S«
Facility B
Facility E
Facility C«
Facility H«
Facility Q
Facility L
Intrinsic Hazard
Potential
Chemical
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
pH
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Surface Water Contamination
Potential by Different
Migration Pathways
Storm
Water
Low
Low
Low
Mod.
Mod.
Mod.
Low
Neg.
Low
Low
Low
Low
Low
Neg.
Neg.
Ground
Water
Mod.
Low
Mod.
Low
Neg.
Mod.
Neg.
Low
Neg.
Low
Mod.
Low
Low
Neg.
Neg.
Air
Mod.
Mod.
Low
Low
Low
Mod.
Mod.
Low
Neg.
Low
Mod.
Mod.
Mod.
Neg.
Neg.
Transport
Potential
High
High
High
High
High
Mod.
Mod.
High
High
High
Low
Low
Neg.
Neg.
Neg.
Current
Exposure
Potential*
High
High
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
Mod.
High
Mod.
Neg.
Neg.
Overall Surface Water
Risk Potential (Rank)
Chemical
Mod. (1)
Mod. (2)
Mod. (3)
Mod. (4)
Mod. (5)
Mod. (6)
Mod. (7)
Low (8)
Low (10)
Low (9)
Low (11)
Low (12)
Neg. (13)
Neg. (14)
Neg. (15)
PH
Mod. (1)
Mod. (2)
Mod. (5)
Mod. (4)
Mod. (3)
Mod. (6)
Mod. (7)
Low (8)
Low (10)
Low (9)
Low (11)
Low (12)
Neg. (13)
Neg. (15)
Neg. (14)
* Burns hazardous waste as fuel.
* Future development of surface water uses around these facilities could increase the exposure potential rankings and, depending on
the risk rankings for the other site factors (intrinsic hazard, surface water contamination potential, and transport potential), could
result in higher overall surface water risk potential rankings.
to this same reservoir, located 1,000 meters directly downgradient and downwind
from the on-site pile. This reservoir has minimal flow, so any contamination
reaching the water is unlikely to be transported downstream and diluted
significantly. In addition, there is a high potential for human exposures through
the fish ingestion pathway because the reservoir is actively fished.
At Facility F, there is a potential for contaminants to migrate through ground
water from the on-site CKD pile to a creek located 600 meters downgradient.
There also is a potential for windblown dust from the pile to deposit in the same
creek, given the limited controls on dusting and the on-site meteorological
conditions. This creek has a low flow and dilution capacity, and currently is used
in the vicinity of the cement plant for agricultural purposes, creating the potential
for human exposures through the food chain. Lead and pH levels measured in
-------
6-28
extract analyses of this facility's CKD also exceed AWQCs, indicating a potential
for aquatic ecological damage if the creek is contaminated.
• At Facility D, a moderate potential for surface water contamination through
stormwater run-off exists because run-off is only partly controlled and the nearest
surface water body that may receive run-off is 300 meters away. In addition,
there is a potential for this same creek to be contaminated by airborne deposition
of CKD, because windblown dusting from the pile is not prohibited entirely (e.g.,
although the pile is occasionally wetted, it is not covered or compacted over its
entire surface). The potentially receiving water body has a very low flow (annual
average of 0.06 m3/sec, or 2 cfs), and thus has a very limited dilution capacity.
The low flow makes it unlikely that the water body is used as a human drinking
water supply. However, thallium and dioxin concentrations measured in leachate
extracts of the facility's CKD indicate a potential for human health risks through
the fish ingestion pathway. The high pH levels of CKD leachate at the facility
also create the potential for aquatic ecological damage.
• At Facility A, there is a contamination potential via the stormwater run-off and
air migration pathways because of the close proximity of the nearest water body
(15 meters), a moderate potential for windblown dusting from the site's CKD pile,
and limited stormwater run-off controls. The potentially receiving surface water
has a low dilution capacity (annual average flow of 2 m3/sec), and people could
come in direct contact with the receiving water at the point of contamination (i.e.,
the water body is off site and access to it is unrestricted). Given the water's low
flow, any surface water contamination is probably not a human drinking water
threat, although it could pose a health threat via the fish ingestion pathway
(thallium appears to the primary constituent of potential concern for this
pathway). Also, elevated lead and pH levels measured in leachate extracts of the
plant's CKD indicate a potential for adverse aquatic ecological effects.
• At Facility I, the nearest water body to the CKD pile is a river located only 90
meters away. There is a potential for contaminants to migrate into the river
through ground-water discharges because there is a moderate potential for
ground-water contamination at the site (given limited engineering controls and the
site's hydrogeology), the river is located in a downgradient direction, and the river
is likely to receive ground-water inputs. A potential also exists for contaminants
to migrate to the river via stormwater run-off and windblown dusting, given site
meteorology and limited controls on the pile (e.g., stormwater is not diverted in
drainage ditches or subject to NPDES permitting prior to discharge). However,
the water's relatively large flow (annual average of 132 mVsec) is expected to
significantly dilute any contamination that enters the river.
• At Facility N, CKD containment features and site environmental conditions
combine to create a moderate potential for contaminants to blow into the air and
deposit in a river about 1,400 meters from the on-site CKD pile. There also is a
low potential for contaminants to migrate to this same creek along with
stormwater run-off, given the pile's run-off controls and distance from surface
water. The potentially receiving river has a moderate flow (annual average of
almost 100 m3/sec), can be accessed by people in the area where CKD
contaminants would enter the water, and is presently used for recreation, fishing,
-------
6-29
and irrigation. Elevated lead and pH levels measured in leachate extract analyses
of the facility's CKD also suggest the potential for aquatic ecological damage.
The Agency ranked five facilities as having an overall low risk potential for the surface
water pathway. All of these facilities were assigned a relatively low surface water risk because
one or more critical factors (e.g., low contaminant concentrations, low transport potential) were
found to pose a low risk according to the Agency's ranking methodology.
Three facilities were ranked as having an overall negligible potential for surface water
risk. As discussed for the ground-water rankings, two of these facilities, Facilities Q and L, were
assigned a negligible risk potential because they presently recycle 100 percent of their CKD. The
other facility, Facility H, was assigned a negligible surface water risk potential because the
nearest surface water is located 1,200 meters from the on-site CKD pile. This relatively long
distance makes it unlikely that the river will receive large CKD loads via any migration pathway
(ground water, stormwater run-off, or air). Even if CKD migrated to the river, it would be
quickly diluted because of the river's high flow in the vicinity of Facility H (over 5,000 m3/sec on
average).
When fully implemented, the Agency's recently promulgated stormwater runoff control
regulations (described in Section 7.2.1 of Chapter 7) could substantially mitigate or eliminate
human health risks and aquatic ecological damages to surface waters attributable to stormwater
runoff of CKD contaminants. These regulations would not, however, control delivery of CKD
contaminants to surface waters via ground-water or air pathways.
Risk Potential Ranking for the Air Pathway
The air pathway is of concern for CKD because the dust is a fine paniculate matter that
is readily suspendable, transportable, and respirable in air. In general, particles that are <.100
micrometers (/on) may be suspended in the wind and transported. Within this range, particles
that are <30 /xm can be transported for considerable distances downwind. However, only
particles <.10 jtm are respirable by humans. The significance of paniculate size for CKD is
illustrated in Exhibit 6-8, which displays the particle size distribution for dust samples by kiln
type. Virtually all of the dust generated at the 15 case-study sites may be suspended and
transported in the wind (i.e., the vast majority of particles are <.100 /im), and over two-thirds of
all dust particles generated may be transported over long distances. Additionally, a significant
percentage of the total dust generated (from 22 to 95 percent, depending on kiln type) is
comprised of respirable particles that are <.10 /an.
In an effort to keep the dust down, many facilities add water to CKD prior to disposal to
form larger clumps or nodules. In addition, as CKD sits in a pile exposed to the elements,
occasional wetting by rainfall results in the formation of a thin surface crust in inactive areas of
the pile. However, based on field observations during the site sampling trips, neither the
formation of nodules nor the natural surface crusting eliminates the potential for CKD to blow
into the air. Nodulizing the dust prior to disposal provides incomplete and temporary control
because the entire dust volume is not nodulized and because the dust eventually dries and
returns to a fine paniculate that is available for suspension and transport. Likewise, a surface
crust may develop, but (1) the crust breaks when vehicles or people move on the pile, and (2)
fresh dust is regularly added to the pile providing a continual, exposed reservoir of fine particles.
-------
6-30
-------
6-31
Exhibit 6-8
Particle Size Distribution of CKD by Kiln Type"
Kiln Type
Long, wet rotary
Long, dry rotary
Dry, with
precalciner
Number of Kilns
in Case Study*
20
2
6
Percentage of
Particles £100 pm
95
100
98
Percentage of
Particles <30 pm
77
99
66
Percentage of
Particles <. 10 /zm
53
95
22
* Data for particle size distribution from: Todres, H.A. et al. 1992. CKD Management
Permeability, Research and Development Bulletin RD103T, Portland Cement Association, Skokie,
IL.
b The number of kilns represents the total number of kilns at the 15 facilities sampled in 1992,
such that if one facility had three kilns each of the kilns was counted.
Although these intrinsic properties of CKD make the dust conducive to airborne
suspension and transport, other site-specific factors must be considered when evaluating the
overall risk potential for the air pathway. For this risk potential ranking, EPA has focused
primarily on the potential for CKD releases as the dust is transported across a site and disposed
in piles. The Agency recognizes that an unknown quantity of CKD also may be released from
fugitive emissions during loading and unloading of vehicles transporting CKD, during CKD
removal from the dust collection systems (e.g., electrostatic precipitators), and from other points
in the process (e.g., process leaks or stack emissions). However, insufficient information was
available to evaluate these potential release sources in a meaningful way in this risk ranking.
Exhibit 6-9 summarizes the air risk potential rankings for the 15 case-study cement plants.
The Agency developed overall risk potential rankings based on the intrinsic hazard of the dust
(based on total concentrations measured in dust), the air contamination potential, transport
potential, and current exposure potential. The plants are listed in the exhibit from highest to
lowest risk potential.
None of the facilities were ranked as posing a high risk potential for the air pathway
considering the many site-specific factors that influence risk. Several facilities were ranked high
for at least one critical factor, but this was moderated when combined with other factors. For
example, the exposure potential was ranked high at Facility B, but other factors such as the
intrinsic hazard of the facility's dust, the moderate exposed surface area of the pile (51,200 m2 or
550,000 ft2), the high precipitation-evaporation index (indicating a relatively moist environment),
and the distance to the nearest residence (460 meters) suggest that overall risk potential is
moderate rather than high.
The Agency ranked 11 facilities, including seven hazardous waste burners and four
facilities that do not burn hazardous waste, as posing a moderate risk potential for the air
pathway. The similarity in scores across the range of facilities is related to the similarities in
intrinsic hazard scores (all 15 plants scored moderate for intrinsic hazard) and similarity in
-------
6-32
Exhibit 6-9
Risk Potential Rankings for the Air Pathway
Facility
Facility A*
Facility J
Facility D
Facility B
Facility G
Facility F*
Facility O*
Facility I»
Facility N*
Facility H*
Facility S*
Facility C*
Facility E
Facility L
Facility Q
Intrinsic Hazard
Potential
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
- Moderate
Moderate
Moderate
Moderate
Air
Contamination
Potential
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Negligible
Negligible
Transport
Potential
High
High
High
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
Low
Low
Negligible
Negligible
Current
Exposure
Potential*
High
Moderate
High
High
High
Moderate
Moderate
Moderate
Moderate
Moderate
Moderate
High
Moderate
Negligible
Negligible
Overall Air Risk
Potential (Rank)*
Moderate (1)
Moderate (2)
Moderate (3)
Moderate (4)
Moderate (5)
Moderate (6)
Moderate (7)
Moderate (8)
Moderate (9)
Moderate (10)
Moderate (11)
Low (12)
Low (13)
Negligible (14)
Negligible (15)
* Burns hazardous waste as fuel.
' Future development of land uses around these facilities could increase the exposure potential rankings and,
depending on the risk rankings for the other site factors (intrinsic hazard, air contamination potential, and
transport potential), could result in higher overall air risk rankings.
b The distinction between pH and chemicals is not applicable for the air pathway. The intrinsic hazard ranking
for the air pathway is based only on results of totals analyses, which do not include pH (pH is only relevant for
liquids).
management practices (13 of the 15 facilities scored moderate for contamination potential).
However, to prioritize plants for risk modeling, EPA identified the individual plants posing the
greatest potential risk for the air pathway. The three plants ranked as having the greatest risk
potential were Facilities A, J, and D. Primary factors that contributed to their ranking included:
• At Facility A, a large exposed surface area of the dust pile (206,000 m2), limited
dust suppression measures (e.g., the pile is not wetted and is only partially
compacted), moderate wind speeds, and a relatively moist setting (relatively
frequent rainfall and limited evaporation) contributed to an overall moderate
ranking for air contamination potential. The close proximity of the CKD pile to
the site boundary (30 meters) and moderate distance to the nearest residence (490
meters) suggest a high potential for CKD to be transported to receptors if it is
-------
6-33
released in the air. Finally, Facility A has a relatively large population within one
mile (3,000 people) and much of the land surrounding the plant is used for
agriculture (the facility leases some of its own property to nearby farmers),
suggesting that both inhalation and food chain exposures could occur if CKD is
released to air.
• At Facility J, limited dust suppression practices (e.g., the on-site CKD pile is
uncovered, not wetted, and only partially compacted) and moderate rainfall and
wind speeds result in a moderate potential for wind erosion from the on-site CKD
pile. The proximity to property boundaries (140 meters) and a residence (300
meters) also suggests that CKD could be transported to receptors if released to
air. Surrounding agricultural land and pastures provide a pathway for food chain
exposure in addition to direct inhalation and incidental ingestion exposures.
• At Facility D, an active dust suppression program (e.g., wetting the pile)
moderates the potential releases from a pile with a large exposed surface area
(102,000 m2) in a dry climate. However, the close proximity of the CKD pile to
property boundaries (150 meters) and the large nearby population (1,400 people
within one mile) suggest that transport of dust and exposures to nearby
populations may occur.
Only Facilities C and E were ranked as having a low risk potential for the air pathway.
Both of these facilities were ranked low because the potential for transport to exposed
individuals for each facility was low. Specifically, the nearest residence at Facility C is 1,100
meters from the pile, and the nearest residence at Facility E is over 1,600 meters from the pile.
Therefore, it is unlikely that significant inhalation exposures would occur. However, the air
pathway risks do not appear to be negligible because land around each facility is used, in part,
for agricultural purposes, creating the potential for human exposures through the ingestion of
food contaminated by atmospheric deposition.
Only two facilities were ranked as having a negligible risk potential for the air pathway.
These facilities, Facilities L and Q, were assigned negligible risk because all generated CKD is
currently recycled.
6.2.2 Risk Modeling of On-site CKD Management
This section presents the methodology and results of the Agency's quantitative fate and
transport modeling analysis of on-site CKD management. The first part presents the analytical
methodology and the second part presents the results of the on-site risk modeling.
Analytical Methodology
The Agency conducted a quantitative fate and transport modeling analysis to estimate the
potential human health and environmental effects associated with current on-site CKD
management practices. This modeling analysis extended the results of the risk potential ranking
analysis (presented in Section 6.2.1) by quantifying risks at five of the 15 facilities evaluated in
that ranking analysis. For each of the three primary direct exposure pathways scored (i.e., air,
surface water, and ground water), the two highest ranking facilities in each exposure pathway
were selected for the modeling analysis to provide a basis for quantifying the upper end of the
risk distribution for the 15 case-study plants. Because some facilities were the first or second
-------
6-34
highest scoring facility in more than one pathway, this approach resulted in the selection of a
total of five facilities for modeling.
As discussed in Section 6.2.1, the risk ranking evaluated a sample of 15 CKD facilities
shown to be reasonably representative of the universe of 115 CKD facilities in the U.S. By
evaluating risks at two facilities believed to represent the highest risk potential in each of the
three direct pathways, EPA selected plants that would be most likely to include those
combinations of CKD constituent characteristics, management practices, and exposure settings
that might pose the greatest risk to human health from the larger 15-facility sample.
While the methodology focused on evaluating the potential high end of the risk
distribution, it also provided an estimate of the central tendency portion of the national risk
distribution, because three of the five modeled facilities represented midrange scores in each of
the pathways. For example, while Facility G was selected as the highest ranking facility in the
ground-water pathway, it represented the eighth ranking facility out of 15 for the surface water
pathway. Thus, the modeled surface water risk estimates corresponding to this facility and two
others could be used to represent the central portion of the national risk distribution. In this
manner, the Agency was able to characterize the central tendency portion of the national risk
distribution.
In focusing on the 15 case-study cement plants, it is possible that certain less frequent but
potentially high risk CKD management scenarios might not be represented, potentially
understating the true high end nationwide risks from CKD disposal. Consequently, the case-
study baseline analysis was supplemented with a number of potentially higher risk scenarios to
more fully characterize the upper tail of the distribution of national risks. Exhibit 6-10 illustrates
these six sensitivity analysis scenarios and their relationship to the baseline central tendency and
high end scenarios evaluated in the initial case study analysis.
The baseline on-site CKD management scenarios simulated, as closely as feasible, the
actual waste management practices and environmental conditions at the five modeled facilities in
order to estimate order-of-magnitude risks at relevant exposure points. Risks were estimated
using a standard Agency screening-level model (MMSOILS), a mix of site-specific and regional
geographical data, and standard Agency exposure assessment and risk characterization methods.
Both individual cancer risks and noncancer human health effects were estimated via air, ground-
water, surface water, soils, and the foodchain pathways. Aquatic ecological effects also were
estimated for potentially affected surface waters.
The sensitivity analyses of higher risk modeling scenarios were conducted to quantify
effects from potentially higher risk waste characteristics, environmental settings, or CKD
management practices that have been observed nationally but that were not found at the five
baseline facilities. Thus, they are hypothetical yet plausible. Each of these scenarios was based
primarily on the baseline case-study facility characteristics; only key risk factors were modified to
simulate a potentially higher risk condition. For example, hypothetical upper bound dioxin risks
were estimated by simulating dioxin/furan concentrations at the highest levels measured by EPA
at each of the five modeled facilities. Thus, this sensitivity analysis scenario combined the basic
transport and exposure characteristics of the five original baseline facilities with one selected high
risk potential factor (increased dioxin concentrations) to provide an upper sensitivity estimate of
the potential contribution of dioxins/furans to CKD risks.
-------
6-35
Exhibit 6-10
Graphical Illustration of On-site Risk Modeling Scenarios
-------
6-36
The six higher risk scenarios examined, in turn, the following waste characteristics,
environmental settings, management practices, or exposure scenarios:
• Disposal of CKD with the highest levels of 2,3,7,8-substituted chlorinated dibenzo-
p-dioxins (CDDs) and dibenzo-furans (CDFs) measured by EPA;9
• Disposal of CKD with the 95th percentile highest measured metals concentrations
based on combined EPA and industry samples from nearly 100 CKD facilities;
• Simulation of a CKD pile located directly adjacent to an agricultural field with
uncontrolled erosion of CKD impacting the crops;
• Simulation of a CKD pile located directly adjacent to a surface water body (a lake
and a river) with uncontrolled CKD eroding directly to the water;
• Simulation of CKD management in the bottom of a quarry that is covered with
water resulting from ground-water seepage; and
• Simulation of potential risks to highly exposed individuals relying on locally-grown
produce, beef, and milk, and locally-caught fish for subsistence purposes.
The primary components of the risk modeling methodology used in the baseline on-site
scenarios, the higher risk scenarios, and the off-site use scenarios, are summarized below. A
more-detailed presentation of the modeling methodology is presented in Chapter 8 of the Risk
Assessment Technical Background Document.
Release. Fate, and Transport Modeling Methodology
The CKD risk modeling analysis used the MMSOILS model, a screening-level multimedia
contaminant release, fate, and transport model, to estimate ambient concentrations of
constituents of concern in ground water, air, surface water, soils, and the foodchain. MMSOILS
was developed by EPA's Office of Research and Development to simulate the release of
hazardous constituents from a wide variety of waste management scenarios and their subsequent
multimedia transport through key environmental pathways.10 MMSOILS also simulates
numerous cross-media transfers of contaminants (e.g., atmospheric deposition to soil and ground
water discharge to streams). As a screening-level model, MMSOILS was designed to provide
rough order-of-magnitude exposure estimates in relatively simple environmental settings (e.g.,
granular porous aquifers and relatively flat terrain). Greater uncertainty is associated with the
model's application to more complex and heterogeneous environmental settings. See Chapter 8
of the Technical Background Document for a more detailed description of MMSOILS and its
use in this risk analysis.
The Agency adopted a screening-level methodology for this analysis both in the selection
of MMSOILS and in the nature of the data used in the simulations. The Agency used site-
9 The Agency is currently conducting a scientific reassessment of the cancer potency of CDDs/CDFs. Because this
reassessment has not yet been completed, the CDD/CDF risk estimates are subject to revision.
10 U.S. Environmental Protection Agency, Office of Research and Development, MMSOILS: Multimedia
Contaminant Fate, Transport, and Exposure Model. Documentation and User's Manual. September 1992 (updated in
April 1993).
-------
6-37
specific, regional, and national level data to characterize the five actual cement kiln facilities.
Data for the baseline on-site facilities was obtained primarily from three sources:
• Site-specific data that were collected by the Agency from actual CKD facilities;
• PCA mail survey;" and
• • Previously collected data on facilities located in similar geographical regions as the
case study CKD facilities.12
These data represent the best readily available sources for simulating waste characteristics, CKD
management practices, environmental settings, and receptor locations at the five baseline on-site
facilities. Because many of the environmental setting data characterize the regional setting of a
facility rather than its site-specific features, the modeling results represent a rough screening-level
indication of contaminant fate and transport in the various environmental media.
The Agency estimated ambient concentrations of CKD constituents of concern in the
following exposure pathways/routes:
• Direct inhalation of air;
• Ingestion of contaminated ground water;
• Recreational exposures to contaminated surface water13
• Incidental ingestion of contaminated soil; and
• Foodchain ingestion of contaminated vegetables, beef, and milk.
The MMSOILS documentation describes the mathematical approaches used in estimating
ambient concentrations in each of these pathways, along with key assumptions and limitations.
There are many sources of analytical uncertainty in any exposure or risk assessment. To '
better characterize this uncertainty, the Agency's guidance on risk characterization recommends
developing both "central tendency" and "high end" risk estimates when conducting risk
assessments.14 The central tendency estimate represents the best estimate of risk, while the high
end estimate represents a plausible estimate of the individual risk for those persons at the upper
end of the risk distribution. This study adopted the Agency's recommended approach by
developing both central tendency and high end risk estimates for CKD facilities.
In addition, EPA guidance recommends accounting for analytical uncertainty wherever
possible in risk assessments. In developing this CKD risk assessment methodology, the Agency
identified the most significant sources of uncertainty that could result in understating individual
risks at the baseline facilities. Because no analytical data were available at the five facilities .
11 The 1991 Portland Cement Association mail survey.
12 These data were collected from EPA Regional offices and states for the Corrective Action Regulatory Impact
Analysis currently being conducted by EPA's Office of Solid Waste.
a Because none of the five baseline facilities had drinking water supply intakes in any of the rivers downstream of
the CKD facilities, exposures from ingestion of surface water as a drinking water source were not estimated in this
analysis.
14 U.S. EPA, 1992. Guidance on Risk Characterization for Risk Managers and Risk Assessors.
-------
6-38
quantifying environmental concentrations of CKD constituents at exposure points, it was not
possible to calibrate the fate and transport modeling methodology with actual site data.
Consequently, it was judged that the modeled exposure concentrations represented the most
significant source of analytical uncertainty. Accordingly, the Agency generated "best estimate"
and "upper bound" constituent concentrations in each exposure pathway at each facility based on
best estimate and upper end characterizations of the key environmental transport parameters
contributing most to uncertainties in the ambient concentration estimates.
Characterization of Exposed Populations
Data were collected on the locations of individuals that could be exposed to ambient
concentrations of CKD constituents in each of the exposure pathways analyzed with MMSOILS.
The methodology focused on estimating plausible exposure that could reasonably be expected
based on actual nearby residential exposure locations. The risk modeling did not estimate risks
corresponding to exposures directly on the CKD pile, because the Agency did not identify any
residences on abandoned CKD piles; this hypothetical exposure scenario was not addressed
further in the study. In addition, because the methodology was based on a risk screening
approach, it was not possible to characterize the distribution of risks received by exposed
populations surrounding each CKD facility. The approach used in characterizing the exposure
points evaluated in the modeling analysis is briefly summarized below.
Direct Inhalation
For estimating individual exposure from direct inhalation of windblown CKD
contaminants, USGS quadrangle maps and site visits were used to identify the nearest residence
to the CKD pile in any compass direction. For estimating the total exposed population at the
site, the total number of residences surrounding the facility were identified out to a distance of
10 kilometers from the CKD pile. (In addition to estimating direct inhalation of airborne
contaminants, indirect exposure resulting from wind erosion of CKD particulates were estimated
in several of the other exposure pathways described below.)
Surface Water
Sources of drinking water in the vicinity of the five baseline facilities were identified
through contacts with the water utilities serving the vicinity of each facility, and it was
determined that none of the five areas withdrew surface water for public water supplies
downstream from the CKD facilities. Consequently, exposures were not estimated for ingestion
of surface water as a source of drinking water. Surface water exposures through recreational
swimming were estimated at the point in the nearest surface water body closest to the CKD pile.
(Exposures through ingestion of locally-caught fish in the nearest surface water body were also
estimated as part of the foodchain analysis.)
Ground Water
The extent of ground-water usage as a local drinking water source was determined
through contacts with the water utilities serving communities around each facility. The Agency
determined that one of the five facilities had significant private ground-water usage downgradient
of the site, while three facilities primarily served by public water supplies were likely to have only
limited private well usage; one of the facilities had no ground-water usage within one mile of the
facility. Accordingly, individual ground-water exposures were estimated at the nearest residence
-------
6-39
downgradient of the four facilities with potential ground water usage, while ground-water
exposures were not estimated at the fifth facility. Exposure to the potentially affected population
at the one site with significant ground-water usage were based on all residences located
downgradient of the facility within a distance of two miles; at the other three facilities population
risks were not estimated (only individual risks).
Incidental Soil Ingestion
Exposure due to the incidental ingestion of soil were estimated at the residence nearest
to each facility that could potentially receive atmospheric deposition and/or erosion from the
CKD facility. This location generally represented the closest residence identified for estimating
direct inhalation exposures.
Foodchain Pathway
The foodchain pathway analysis generated constituent concentrations in vegetables, beef,
milk, and fish at different exposure points in the vicinity of the facility. For vegetables, beef, and
milk, foodchain concentrations were estimated at the agricultural field or pasture nearest to the
facility. The locations of these fields were identified during the site visits (at one facility), or
estimated based on the percentage of agricultural land in the county (at the other four facilities).
While these fields and grazing lands were intended to be located on family farms for purposes of
the exposure assessment, the actual crops grown and use of these fields was not known (and thus
may significantly overstate actual foodchain exposures).15 Constituent concentrations in fish
were estimated at the nearest point in the surface water body closest to the facility. It was not
known what edible species of fish were present in these streams or whether the streams are
actually used for recreational fishing. Consequently, this scenario may also overstate actual
foodchain exposures through fish ingestion.16
Exposure Assessment and Risk Characterization
The Agency followed standard guidance, methods, and practice in estimating risks due to
exposures at the five baseline facilities.17'18119 Best estimate and upper end individual lifetime
cancer and noncancer effects were calculated at each exposure point using the best estimate and
upper end exposure concentrations from the MMSOILS modeling results. (The Technical
u The foodchain exposure pathway analysis was based on the assumption that vegetables grown for human
consumption originate in a field located adjacent to the pasture used for grazing the beef .and dairy cattle.
16 The Agency is currently revising its "Methodology for Assessing Health Risks Associated with Indirect Exposure
to Combustor Emissions," Interim Final, EPA/600/6-90-003, January 1990, and consequently this foodchain risk
methodology is subject to revision.
17 U.S. EPA, 1989. Risk Assessment Guidance for Superfund Volume I: Human Health Evaluation Manual (Part
A). Office of Emergency and Remedial Response. EPA/540/1-89/002.
'* U.S. EPA, 1991. Human Health Evaluation Manual. Supplemental Guidance: Standard Default Exposure
Factors. Office of Emergency and Remedial Response. OSWER Directive: 9285.6-03.
19 U.S. EPA, 1989. Exposure Factors Handbook. Office of Health and Environmental Assessment. EPA/600/8-
89/043.
-------
6-40
Background Document for this human health and environmental risk assessment provides .
significantly greater detail on the exposure and risk assessment methodologies.)
In this analysis, the Agency estimated individual excess cancer risk for each pathway.
which represents the increase above background in the probability of developing cancer over an
individual's lifetime in response to contaminant exposures. To estimate excess cancer risks, the
Agency multiplied the daily intake of each carcinogen by the cancer slope factor published in
EPA's Integrated Risk Information System (IRIS).20 The highest individual risk for each
exposure pathway is the sum of cancer risks calculated for each carcinogenic constituent resulting
from exposures at the nearest location to the facility at which an exposure through that pathway
could occur (e.g., for the ground-water pathway, the nearest point of ground-water use
downgradient of the facility). Total pathway cancer risks represent the constituent-specific risks
aggregated across chemicals within each pathway (following Agency guidance, cancer risks were
not aggregated across exposure pathways).
The Agency evaluated noncancer effects by determining the ratio of the estimated dose
of a particular contaminant to a standard Agency reference dose (RfD). These ratios are
referred to as "hazard quotients." Hazard quotients greater than one for individual chemicals
represent an exceedance of an Agency threshold of concern and the possibility of an adverse
health effect. Total individual noncancer effects were evaluated by adding the chemical-specific
hazard quotients within each pathway, referred to as the "hazard index."
Direct Ingestion Pathways
Exposures through direct inhalation, drinking water ingestion, incidental soil ingestion,
and recreational ingestion of surface water were estimated using national average exposure rates,
frequencies, and durations reported in standard Agency guidance documents.
Foodchain Pathways
Exposures for the vegetable, beef, and milk foodchain pathways were based on the
assumption that the exposed individuals live on a family farm at which they raise a portion of
their annual consumption of these food products (or live in a farming community where a
significant portion of their food could originate from one local source). Moreover, it was
assumed that the home-grown vegetables they consume all originate from the identified
agricultural field receiving CKD from the facility, and that their beef and dairy cattle are
provided feed from pasture land in the same location. While the extent of consumption of
home-grown vegetables, beef, and dairy products will vary significantly on a site-specific basis
depending on the types of crops grown, the type of farm, and individual behavior, the
considerable variation in exposures on a site-specific basis could not be accounted for in this
analysis. In general, it is believed that these exposure estimates may significantly overstate actual
consumption patterns. Exposures through ingestion of recreationally caught fish were estimated
using behavior patterns for the average individual in the general population as reported in
standard Agency guidance.
20 U.S. Environmental Protection Agency, Integrated Risk Information System Database.
-------
6-41
Sensitive Subpopulations (Childhood Exposures to Lead)
Exposures to lead were calculated for one sensitive subpopulation -- children up through
the age of seven years located at the residence nearest to each baseline facility. Because EPA
has not published a reference dose for this systemic toxicant, the lead uptake/biokinetic (UBK)
model was used to estimate the increased blood lead levels from exposure to lead in CKD.21
The lead uptake/biokinetic (UBK) model provides a method to predict blood lead levels in target
populations (i.e., children ages 0 to 7) exposed to lead in air, diet, drinking water, indoor dust,
soil, and paint. Based on user-supplied lead concentrations in each of these potential sources of
exposure, the UBK model estimates the relative contributions of each exposure source and the
total lead uptake from all sources.
The model presents several different indicators of potential health effect from lead. First,
it generates a distribution of blood lead levels for each year of the exposure period (ages 0 to 7)
based on the total lead uptake. Second, the UBK model estimates the geometric mean blood
level in the exposed population. Finally, the model estimates the percentage of the exposed
population that is expected to be at or above a specified blood-lead threshold level (a blood lead
level greater than 10 /xg/dL was assumed to be the threshold of interest, based on exposure and
effect relationships that have been established in infants and children at blood lead
concentrations as low as 10 /ig/dL).22
The UBK model accepts inputs for several sources of lead exposure not estimated in this
CKD risk analysis: paint ingestion, indoor dust, and drinking water exposures resulting from lead
solder pipes. For these sources, which were assumed to be unaffected by the CKD facilities,
average background lead concentration levels presented as default values in the UBK model
were employed. For those exposure routes used in the UBK model that were estimated by
MMSOILS in this analysis, which included dietary intakes (through vegetables, beef and milk,
and fish), soil intakes, and atmospheric exposures, the estimated lead concentrations from
MMSOILS were added to the national average background values presented in the UBK model.
Thus, the blood lead levels estimated in this analysis represent an increment above the national
average background childhood blood lead levels estimated by the UBK model resulting from
exposures to CKD.
Aquatic Ecological Effects
The Agency estimated potential aquatic ecological effects from CKD releases by relating
ambient surface water constituent concentrations to benchmarks for the protection of aquatic
life. These benchmarks were either published EPA chronic ambient water quality criteria
(AWQC) for the protection of aquatic life, or, where these were not available, lowest observed
adverse effect levels (LOAELs) divided by a factor of 5 to account for variations in species
sensitivity. Exhibit 6-11 shows the eight constituents for which aquatic ecological benchmarks
were available based on AWQC documents and their values. AWQCs are intended to protect
21 U.S. EPA, 1991. "A PC Software Application of the Uptake/Biokinetic Model, Version 0.5," Office of Health
and Environmental Assessment (ECAO-CIN-2178A).
22 A threshold for the noncancer effects of lead is believed to lie within or below the 10 - 15 ug/dL range. Note,
however, that this range is regarded as a "level of concern" warranting attention from a medical viewpoint and not a
dose level or threshold below which no adverse health effects would be expected to occur. (From U.S. Environmental
Protection Agency, "Technical Support Document on Lead," Office of Research and Development, ECAO-CIN-757
(January 1991).
-------
6-42
aquatic communities against adverse effects on structure or function by protecting 95 percent of
the species against adverse population-level effects. These adverse effects are species-dependent
and could include reduced reproduction, growth, or survival. Effects of contaminated sediments
on benthic communities are not considered.
Exhibit 6-11
Aquatic Ecological Benchmark Levels
Constituents
Antimony
Arsenic (III)
Beryllium
Cadmium*
Chromium (VI)
Lead*
Thallium
2,3,7,8-TCDD
Aquatic Ecological
Benchmark (mg/L)
3.2 x 10'1
1.9 x lO'1
1.1 x lO'3
1.1 x 10'3
1.1 x lO'2
3.2 x 10'3
8.0 xlO'3
2.0 x lO'9
Source
LOAEL
AWQC
LOAEL
AWQC
AWQC
AWQC
LOAEL
LOAEL
Assumes a water hardness of 100 rag/L CaCOj
Sensitivity Analysis of Higher Risk Potential Scenarios
The Agency conducted a sensitivity analysis of various factors that could indicate the
potential for higher risks from CKD management than exhibited in the baseline risk modeling
analysis. A selected number of low probability but potentially higher risk waste characteristics,
environmental settings, and exposure assumptions were identified based on site visits, reports
from CKD facilities (other than those modeled in the baseline analysis), and in some cases,
hypothetical scenarios that could potentially occur but were not specifically observed. In this
sensitivity analysis, the Agency examined the extent to which these selected higher risk potential
factors, when combined with the baseline central tendency and high end modeling scenarios,
could indicate a potential for more significant risks resulting from CKD management.
Two of the sensitivity analysis scenarios examined the sensitivity of the baseline results to
changes in waste characteristics:
• The highest measured dioxins scenario estimated the risks associated with the
disposal of CKD containing the highest levels of 2,3,7,8-substituted chlorinated
dibenzo-p-dioxins (CDDs) and dibenzo-furans (CDFs) measured by EPA during
its CKD facility sampling and analysis program. This scenario examined the
-------
6-43
extent to which the baseline central tendency and high end risk estimates would
change if this high CDD/CDF wastestream were present at each of the facilities.
• The 95th percentile inorganic constituent concentrations scenario evaluated the
change in risk associated with the disposal of CKD exhibiting the 95th percentile
highest constituent concentrations at all five baseline facilities. These 95th
percentile concentrations reflect data taken from the EPA sampling and analysis
effort and the PCA survey.23 Because of the low probability that a wastestream
containing each of the inorganic CKD constituents at their respective 95th
percentile concentration could be found at any single facility, this scenario does
not examine the total incremental risk associated with this wastestream
characterization, but rather examines only the potential for individual constituents
to exceed health effects levels of concern.
Three sensitivity analysis scenarios examined the sensitivity of the baseline risk modeling
results to higher risk potential environmental transport scenarios or CKD management practices:
• An EPA damage case study identified a CKD pile located directly adjacent to an
agricultural field with uncontrolled erosion of CKD impacting the field. To
simulate this scenario, the transport characteristics at two of the facilities modeled
in the baseline analysis were modified to simulate exposures associated with the
location of an agricultural field or pasture directly next to a CKD pile lacking
erosion controls. This scenario focused on the potential effects of this setting on
the terrestrial foodchain pathway alone.
• Several EPA damage case studies identified CKD piles that were located directly
adjacent to surface water bodies with uncontrolled erosion of CKD entering the
water bodies. To simulate this scenario, the environmental transport
characteristics at two of the facilities modeled in the baseline analysis were
modified to simulate the location of a surface water body (one facility had a river
and the other a lake) next to the CKD pile. This sensitivity scenario examined
the incremental risks to the recreational swimming and fish ingestion exposure
pathways.
• EPA identified several facilities practicing CKD management underwater in a
quarry, at which CKD was disposed in a quarry that had been excavated during
cement production and that subsequently was filled with water entering the quarry
through ground-water seepage. This scenario focused on examining the potential
for increased risks through the ground-water transport pathway, although it also
examined potential reductions in risk potential through the atmospheric and soil
erosion pathways.
The sixth scenario examined in the sensitivity analysis examined the incremental change
in foodchain risks associated with an assumption that an individual could rely on vegetables, beef
and milk, and fish originating in locations affected by the CKD pile as major components of their
diet:
25 Portland Cement Association, 1991. op. cit.
-------
6-44
• Potential high exposure due to subsistence food consumption was addressed by
estimating exposures of two categories of individuals: subsistence farming and
subsistence fishing. The subsistence farming scenario simulates the exposures that
could be received by an individual ingesting a high percentage of homegrown
produce, beef, and dairy products. For this hypothetical scenario, seventy-five
percent of the subsistence farmer's beef, milk, and vegetables are assumed to
originate in the CKD-contaminated agricultural field or pasture. The subsistence
fishing scenario simulates the exposures to an individual that ingests a high
proportion of fish caught locally in a CKD-contaminated surface water body. For
this scenario, 75 percent of the fish consumed by the subsistence fisherman is
assumed to be caught in the contaminated water body nearest to the facility.
These exposure scenarios represent relatively infrequent behavior patterns that
have not actually been observed or reported at any of the facilities examined by
the Agency.
Results of On-site Risk Modeling
The results from the on-site risk modeling analysis are presented in this section, first for
the baseline on-site facilities and then for the sensitivity analysis of higher risk scenarios. Unless
otherwise indicated, all cancer risks are reported in terms of excess individual lifetime risk of
cancer. Noncancer effects are reported using the previously described hazard index.
Baseline On-Site CKD Management
The cancer and noncancer baseline modeling results are presented below for the direct
exposure pathways (i.e., air, ground water, surface water, and soil ingestion) and the foodchain
pathways (i.e., vegetables, beef and milk, and fish).
Baseline Direct Exposure Pathway Risks
The Agency calculated a range of high end cancer risks corresponding to both a "best
estimate" of facility transport conditions and an "upper bound" characterization of facility
transport parameters. This range of high end values presented for each pathway, as shown in
Exhibit 6-12, reflects the facility with the highest estimated risks in each respective pathway from
among the five modeled facilities. As anticipated in the previous qualitative ranking of risk
potential, different facilities were responsible for the highest risk estimated in each of the
different pathways.
The central tendency results for the distribution reflect the best estimate of risks from the
three facilities with the lowest risk estimates. The central tendency results are presented as a
range "less than" the highest value estimated for these three facilities. Thus, the Agency believes
that best estimate of the central tendency will generally be less than the reported value (see
Exhibit 6-12).
The central tendency baseline modeling results generally indicate a low potential for
adverse health effects from current CKD management via the direct exposure pathways. Of the
five pathways presented in Exhibit 6-12, the surface water pathway exhibited the highest central
tendency risks, estimated at less than an individual cancer risk level of IxlO"8. The other direct
exposure pathway risks were also negligible. The central tendency results for noncancer health
effects were all more than four orders of magnitude below the health effects threshold (i.e., the
-------
6-45
hazard quotients were less than IxlO"4 in all five direct exposure pathways). These results suggest
that most CKD management facilities will not present significant hazards through direct exposure
pathways.
The high end risks generally indicated a low threat through the direct exposure pathways,
with the exception of the surface water pathway. High end risks resulting from exposures during
recreational swimming ranged from 4X10"6 to 2xlO'5, attributable to arsenic concentrations in the
surface water body. The upper bound of the high end risks in the other four pathways never
exceeded 1x10"*, while the best estimate risks at the high end facilities were all less than IxlO'9.
The only noncancer effect within one order of magnitude of the reference dose were for
recreational swimming exposures, resulting from the combined systemic effects of arsenic and
cadmium. However, the noncancer estimates did not exceed a hazard quotient of 1.0, indicating
a low potential threat in the surface water pathway. The high end noncancer estimates in the
other four pathways were negligible.
These surface water risks reflect potential exposures in the lake located near Facility J.
Because MMSOILS assumes that the lake acts as a sink both for CKD eroding from the pile and
subsequently traveling overland to the lake during its operating period, and for windblown CKD
that reaches the lake from the pile, it indicates a potential for accumulation of CKD constituents
in the lake bed sediments. The potential surface water effects are based on the partitioning of
the CKD constituents from the lake bed sediments into the water column. Because of the
relative simplicity of the lake simulation component of MMSOILS, the Agency believes these
results could overstate the actual high end risks at this facility. Additionally, when fully
implemented, the Agency's recently promulgated stormwater runoff control regulations
(described in Section 7.2.1 of Chapter 7) could substantially mitigate or eliminate human health
risks from surface waters contaminated by stormwater runoff from CKD piles. These regulations
would not, however, control delivery of CKD contaminants to surface waters via ground-water or
air pathways.
Exhibit 6-12
Baseline On-Site Management Cancer Risks for
Direct Exposure Pathways for 15 Case Study Facilities
Exposure
Pathway
Ground water
Surface water
Direct inhalation
Soil ingestion - adult
Soil ingestion - child
Excess Individual Lifetime Cancer Risk
High End
Best Estimate
2xlO'9
4x10*
2xlO'12
5xlO'12
8xlO'12
Upper
Bound
6x1 0-8
2xlO's
3x1 0'12
lxlO'7
2xlO'7
Central Tendency
Less than IxlO'18
Less than IxlO"8
Less than IxlO'14
Less than IxlO'13
Less than IxlO'12
-------
6-46
Baseline Foodchain Pathway Risks
As shown below in Exhibit 6-13, while the baseline foodchain pathway results indicated a
somewhat higher potential for health effects than in the direct exposure pathways, the central
tendency individual cancer risks were still below IxlO"6, with the beef and milk and fish exposure
routes showing negligible central tendency risks at levels less than IxlO"8. The central tendency
noncancer effects were also more than two orders of magnitude below the threshold effects level
(i.e., the hazard quotients were all less than IxlO'2), indicating a negligible likelihood of
.noncancer impact at those CKD facilities represented by the central tendency estimate.
The high end foodchain estimates varied by pathway, with the ingestion of fish resulting
in the highest risks (ranging from 4X10"6 to 4xlO'5) due to the combined cancer effects of arsenic,
beryllium, and potassium-40. High end risks from the ingestion of vegetables ranged from 2x10"*
to SxlO"6, resulting from arsenic uptake into the vegetables. The beef and milk exposure pathway
results were all less than IxlO'6 and ranged from 2xlO'7 in the best estimate to 4xlO'7 in the upper
bound. The high end noncancer foodchain effects exceeded a hazard quotient of 1.0 in one
pathway: ingestion of fish at Facility J resulted in an estimated high end hazard quotient ranging
from 4.1 to 16 due to exposures to cadmium. In addition to cadmium, chromium also
contributed to this high end noncancer effect with hazard quotients ranging from 0.17 to 0.66.
The high end noncancer effects were negligible in the other two foodchain pathways.
Exhibit 6-13
Baseline On-Site Management Cancer Risks for
Foodchain Exposure Pathways for 15 Case Study Facilities
Exposure
Pathway
Vegetable
Beef & Milk
Fish
Excess Individual Lifetime Cancer Risk
High End
Best Estimate
ZxlO-6
2xlO'7
4x10-*
Upper Bound
SxlO-6
4xlO'7
4xlO'5
Central Tendency
Less than IxlO-6
Less than IxlO"8
Less than IxlO"8
In estimating the terrestrial foodchain effects (i.e., vegetables, beef and milk), the
assumptions concerning the amount of erosion transported from the CKD pile to the agricultural
field may result in an overestimate of the impacts in the high end analysis. While the Agency
observed effective erosion controls at the five baseline facilities that were believed to effectively
restrict the off-site movement of CKD by the erosion pathway, it was believed that these erosion
controls could potentially fail in extreme storm events or due to failure of engineered controls.
Consequently, the high end analysis adopted a worst case assumption at three of the facilities
that these erosion controls would completely fail. Because the high end risks in the terrestrial
foodchain pathway were associated with two of these facilities, these high end results may
overstate the likely upper bound risks to the foodchain pathway at these facilities.
-------
6-47
In total, the baseline modeling analysis simulated the release, fate, and transport of 14
constituents (or in the case of CDDs and CDFs, groups of constituents) that have been detected
in CKD and have known cancer or noncancer health effects that could be modeled using current
Agency guidance and available data. Exhibit 6-14 shows all of these constituents and the
exposure pathways where they exceeded a cancer risk of 1x10"* or a hazard quotient of 0.1.
-------
6-48
Exhibit 6-14
Constituents Contributing to Adverse Health Effects In On-site CKD Risk Modeling Analysis
Constituents of
Concern
Antimony
Arsenic
Barium
Beryllium
Cadmium
Chromium
Thallium
2378 TCDD
Equivalents
Potassium 40
Radium 226/228
Uranium 234
Uranium 238
Thorium 230
Exposure Pathways
,0
"•5
TT rt
o> "a
n ,c
CT HE
Recre at ion ai::mvm
iwunmings : :.;s;s::; : : s H
CT HE
v/
v/
5
C9
^ a
1-3
-j irt
2 8,
n ,<=
CT HE
f tijnpv) noijsaSu
|:;i;|;:fips:;iBi(vU3ptS3-i
CT HE
Residentiai ''jSbfl'f||':
ngestioh (ChiJd||
CT HE
8
1
S
00
CT HE
v7
1
u
CT HE
J3
—
CT HE
v7
«/
*/
\l
\i
J
Health Effects
Decreased blood cell synthesis
Skin cancer and skin damage
Increased blood pressure
Gross tumors
Kidney damage
Central nervous system effects
Liver damage
Multiple cancers
Cancer
Cancer
Cancer
Cancer
Cancer
-------
6-49
Exhibit 6-14 illustrates on a constituent-specific basis that none of the central tendency
estimates exceeded a cancer risk of IxlO"6 or hazard quotient of 0.1. The exhibit shows that the
high end risks were the result of exposures to six CKD constituents: arsenic, beryllium,
cadmium, chromium, potassium-40, and thallium. Exposures to CDDs/CDFs were not of
concern in the baseline risk analysis.
Arsenic, which can cause both systemic and carcinogenic effects, contributed to cancer
risks and/or noncancer effects in three high end exposure pathways: exposures during swimming,
ingestion of vegetables, and ingestion of recreationally-caught fish. In the swimming and fish
ingestion exposure pathways, arsenic exceeded both cancer and noncancer levels, while in the
vegetable ingestion pathway, it only indicated a potential threat for its cancer effect (it reached a
level just exceeding IxlO"6). Cadmium resulted in exposures exceeding a noncancer hazard
quotient of 0.1 in two high end pathways at one facility: exposures during swimming (ranging
from 0.056 to 0.22) and ingestion of fish (ranging from 3.8 to 15). The remaining three
constituents contributed to health effects of potential concern only in the high end fish ingestion
scenario. Of these constituents, beryllium and potassium-40 indicated potential cancer effects,
while thallium exceeded a noncancer hazard quotient of 0.1 for fish ingestion at one facility.
As mentioned previously, full implementation of the Agency's recently promulgated
stormwater runoff control regulations could substantially limit human health risks from the
ingestion of fish from surface waters contaminated by stormwater runoff from CKD piles. These
regulations would not, however, limit the migration of CKD contaminants to surface waters via
ground-water or air pathways.
Baseline Estimated Increased Blood Lead Levels
The Agency's methodology for characterizing potential adverse health effects resulting
from exposures to lead generated an estimate of the increased blood lead levels above national
background levels for children. Using the default assumptions in the UBK model for national
average background lead concentrations in the various exposure routes through which children
could be exposed to lead, it estimated a national average mean blood lead level of 3.14 /zg/dL in
children ages one through seven. Thus, in those cases where releases from the CKD facility did
not increase exposures to lead above assumed national background levels, the UBK model would
estimate a mean blood lead level of 3.14 /zg/dL. Where releases from the CKD facility increased
exposures to lead, the resulting estimate of the mean blood lead level would represent an
increment above this national background estimate.
The estimated mean blood lead levels exceeded the baseline value of 3.14 /zg/dL at three
of the five baseline facilities, while the two remaining facilities were estimated to result in no
increase above national background levels. The estimates exceeded the blood lead effect level of
concern of 10 izg/dL at two of these facilities. The highest exceedance took place at Facility J,
where the best estimate mean blood lead level was approximately 14 izg/dL, while the upper
bound estimate was approximately 48 /zg/dL. These increased exposures above background
primarily reflect the simulated ingestion of lead in fish caught in the lake at this facility; exposure
to lead through the other exposure routes were generally below the national average background
levels. Facility A also exceeded the national average background estimates with a best estimate
mean blood lead level of about 5 ^g/dL and a upper bound estimate of about 13 izg/dL. Finally,
the central tendency estimates at Facility F did not exceed background, while the upper bound
estimate was approximately 8 jzg/dL (which is below the health effect level of concern for blood
lead).
-------
6-50
Most of the blood lead level estimates indicating an increase above the national
background were attributable to ingestion of fish caught in the nearest surface water body to the
respective facilities. Because these blood lead estimates are based on the conservative
assumption that 20 percent of the child's fish originates in the contaminated surface water body,
the Agency believes that they most likely overstate that actual lead exposures associated with
most CKD facilities.
Baseline Aquatic Ecological Effects
The examination of aquatic ecological effects focused on eight constituents for which
aquatic ecological benchmarks were available (see Exhibit 6-11): antimony, arsenic, beryllium,
cadmium, chromium (VI), lead, thallium, and 2,3,7,8-TCDD. In the high end analysis, five of
these eight constituents exceeded their aquatic ecological benchmarks (arsenic, beryllium,
cadmium, chromium [VI], and lead). The highest values occurred in the nearby lake at Facility
J, and ranged from about two times the benchmark value for arsenic to about 300 times the
benchmark for cadmium (see Exhibit 6-15 below). Given the relative simplicity of the
MMSOILS lake exposure model, it is likely that these values could significantly overstate the
actual constituent concentrations in this lake. As Exhibit 6-15 shows, none of the constituents
exceeded their respective aquatic ecological health effects benchmarks in the central tendency
analysis.
Exhibit 6-15
Results of Central Tendency and High End Ecological Effects Analysis
Modeling
Scenario
High End
Central Tendency
Ratio of Surface Water Concentration to Ecological Effects Criteria
Arsenic
14-50
Below AWQC
Beryllium
0.5 -2
Below LOAEL
Cadmium
80 - 320
Below AWQC
Chromium
37 - 150
Below AWQC
Lead
14-54
Below AWQC
Again, as mentioned previously, full implementation of the Agency's recently
promulgated stormwater runoff control regulations could substantially mitigate or eliminate
aquatic ecological damages to surface waters attributable to stormwater runoff of CKD
contaminants. These regulations would not, however, limit the migration of CKD contaminants
to surface waters via ground-water or air pathways.
Sensitivity Analysis of Hypothetical Higher Risk Scenarios
The sensitivity analysis of potentially higher risk scenarios quantified the change in the
baseline risks associated with the superimposition of selected high risk potential facility and
environmental setting characteristics on the baseline facility characterization. The Agency
examined six high risk potential scenarios, which selectively modified the baseline facility
estimates as described earlier. The results for each of these six scenarios are presented below in
the following order: maximum measured dioxin concentrations; 95th percentile metal
concentrations; location directly adjacent to an agricultural field; location directly adjacent to a
-------
6-51
receiving surface water body; management underwater in a quarry; and risks to possibly highly
exposed farmers and fisherman.
Maximum Measured Dioadn Concentrations
This sensitivity analysis examined the change in risks that would occur at the five baseline
facilities, based on the hypothetical management of CKD containing the highest measured
CDD/CDF concentrations found in EPA's sampling at 11 cement plants (see the Docket for the
report on the sampling and analysis results). In order to estimate the sensitivity of the original
case-study plant risk estimates to CDD/CDF concentrations, the highest CDD/CDF measured
concentrations were substituted into each of the five original facility settings. This scenario is
presented to provide an upper tail estimate of potential CDD/CDF risks nationwide. (Because
the Agency has only published cancer slope factors for CDD/CDF congeners and has not
published reference doses, this sensitivity analysis only examined incremental individual cancer
risks and did not address noncancer effects.)
Exhibit 6-16 presents the high end and central tendency results for this sensitivity analysis.
In the three primary direct inhalation and ingestion pathways (i.e., ground water, surface, and
air), the results were found to be identical to the original baseline risks (presented previously in
Exhibit 6-12). The lack of incremental increase in risks above the baseline estimates reflects the
fact that CDDs/CDFs did not contribute to these baseline exposure pathway risks due to their
lack of mobility in subsurface systems, low solubility in water, and relatively low concentrations in
air. The sensitivity analysis did indicate a potential increase in the soil ingestion pathways. The
central tendency risks increase by about three orders of magnitude, although they remain
negligible (below IxlO'9). The high risks increased to a similar degree, resulting in risks to adults
ranging from 3xlO"10 (best estimate) to TxlO"6 (upper bound). The risks to children ingesting soil
increased to IxlO"9 in the best estimate to 2xlO'5 in the upper bound.
Exhibit 6-16
Sensitivity Analysis of Maximum CDD/CDF Cancer Risks for Direct Exposure Pathways
Exposure
Pathway
Ground water
Surface water
Direct inhalation
Soil ingestion: adult
Soil ingestion: child
Excess Individual Lifetime Cancer Risk
High End
Best Estimate
Identical to Baseline
Identical to Baseline
Identical to Baseline
SxlO'10
IxlO'9
Upper Bound
Identical to Baseline
Identical to Baseline
Identical to Baseline
7x10*
2x1 0'5
Central
Tendency
Identical to Baseline
Identical to Baseline
Identical to Baseline
Less than IxlO'10
Less than IxlO'9
The sensitivity analysis indicated a similar increase in risks in the foodchain exposure
pathways (Exhibit 6-17). Because the baseline estimates had been higher in the foodchain
pathways, the foodchain risks in the sensitivity analysis were correspondingly higher. The risks to
-------
6-52
the central tendency facilities were about two orders of magnitude greater than in the baseline
analysis, and were less than IxlO'5 in the beef and milk and fish pathways, and less than IxlO"4 for
the ingestion of vegetables. In the high end analysis, the highest risks were found in the fish
ingestion pathway and reached an upper bound value of 2xlO'3. Ingestion of vegetables and beef
and milk resulted in risks ranging from 2x10"* to 6x10"*.
In both the terrestrial foodchain scenarios and the soil ingestion scenarios, the upper
bound risks reflect an assumption concerning the failure of erosion controls at the facilities. As
was the case in the baseline analysis, the results presented in this maximum CDD/CDF
concentration sensitivity analysis are likely to overstate the risks associated with CKD
management.
Exhibit 6-17
Sensitivity Analysis of Maximum CDD/CDF
Cancer Risks for Foodchain Exposure Pathways
Exposure
Pathway
Vegetable
Beef & Milk
Fish
Excess Individual Lifetime Cancer Risk
High End
Best Estimate
2X10-4
ZxlCT1
SxlO-4
Upper Bound
SxlO"4
6X10"4
2xlO'3
Central Tendency
Less than 1x10"*
Less than 1x1 Q's
Less than IxlO'5
95th Percentile Metals Concentrations
The 95th percentile metals concentration sensitivity analysis examined on a constituent-
specific basis the potential for additional CKD constituents to exceed health effects levels of
concern. This scenario was evaluated by scaling the baseline risk estimates for each constituent
based on the ratio of the metals concentrations in the baseline facility sample and the 95th
percentile metals concentrations (see the Docket for the results of EPA's CKD sampling and
analysis program). This simplified approach assumes that the risk results in each exposure
pathway will be linear with respect to constituent concentration. While this approach may be as
accurate as evaluating all of these scenarios directly with MMSOILS, the Agency believes it
represents a reasonable estimation of the risks associated with these higher constituent
concentrations.
The primary incremental change over the baseline results in this sensitivity analysis was
the increased noncancer effects associated with thallium. While it was only within one order of
magnitude of the reference dose for the high end fish ingestion in the baseline analysis, it was
within one order of magnitude of the reference dose in four additional high end exposure
pathways using the 95th percentile concentrations: residential soil ingestion by adults and
children, vegetables, and beef and milk. In one of these pathways, vegetable ingestion, thallium
was within one order of magnitude of the threshold concentration in the central tendency
analysis.
-------
6-53
Only two other constituents were within one order of magnitude of the reference dose in
a single pathway in addition to those found in the baseline. Antimony had a hazard quotient of
0.12 in the high end surface water pathway. Chromium had a high end hazard quotient of 0.49
in the surface water pathway.
Location Adjacent to an Agricultural Field
This sensitivity scenario focused on the potential for increased risks when an agricultural
field or pasture was located directly adjacent to the CKD pile without erosion controls.
Accordingly, this scenario only compares the baseline and sensitivity analysis results for the
terrestrial foodchain pathways (i.e., ingestion of vegetables and ingestion of beef and milk). This
sensitivity analysis only examined two of the five baseline facilities: Facility F (representing the
high end estimate) and Facility J (representing the central tendency estimate).
The sensitivity results indicated that risks could increase in the baseline vegetable and
beef and milk exposure pathways by between one and two orders of magnitude if the facilities
were located directly next to an agricultural field (Exhibit 6-18). Because the scenario assumes
no erosion loss during transport between the CKD pile and the field, this high risk scenario
would be expected to result in significantly higher risks than at the actual baseline facilities.
Both the beef and milk and vegetable exposure routes had similar high end risks, approximately
4xlO"s. The central tendency risks were somewhat lower, with the vegetable risks about one
order of magnitude higher than the beef and milk risks.
Noncancer effects in this sensitivity analysis exceeded the reference dose, unlike in the
baseline analysis. The high end hazard quotient for the vegetable pathway was about 6, while the
high end hazard quotient for the beef and milk pathway was about 3. The central tendency
vegetable pathway hazard quotient was about 5, while the central tendency beef and milk hazard
quotient, with a value of about 0.8, did not exceed the reference dose.
Exhibit 6-18
Sensitivity Analysis of Location Adjacent to Agricultural Field
for Foodchain Exposure Pathways
Exposure
Pathway
Vegetable
Beef & Milk
Excess Individual Lifetime Cancer Risk
High End
Best Estimate
3.8xlO's
4.0xlO'5
Upper Bound
4.2xlO'5
4.3xlO'5
Central Tendency
Less than 3xlO'5
Less than 4x10"*
-------
6-54
Location Adjacent to a Surface Water Body
This sensitivity scenario focused on the potential for increased risks when a surface water
body was located directly adjacent to the CKD pile. Because this scenario only affects the
exposure pathways associated with the ambient concentrations in surface water, this scenario only
examined the recreational swimming and fish ingestion pathways. This sensitivity analysis
examined two of the five baseline facilities, both of which are representative of the high end
risks: Facility F (representing the high end estimate for a facility bordering a river) and Facility J
(representing the high end estimate for a facility bordering a lake). Central tendency risks were
not estimated in this sensitivity analysis, which only examined potential changes to the high end
estimates.
The high end results for the adjacent surface water scenario showed increased health
effects in the recreational swimming and fish ingestion pathways (Exhibit 6-19). In the
recreational swimming scenario, the risks were about one order of magnitude higher than the
baseline risks, while the noncancer effects reached a maximum hazard quotient of 1.0 in one
case. In the fish ingestion scenario, the sensitivity analysis risks were between five and seven
times higher than in the baseline analysis. The most significant change in the sensitivity analysis
was associated with the increased noncancer effects in the fish ingestion scenario, which reached
a maximum hazard quotient of 35 due to uptake of cadmium.
Exhibit 6-19
Sensitivity Analysis of Location Adjacent to Surface Water
for Direct and Foodchain Exposure Pathways
Exposure
Pathway
Excess Individual Lifetime Cancer Risk
High End
Best Estimate
Recreational Swimming 2.4xlO'5
Fish Ingestion || 2.3xlO$
Upper Bound
3.3xlO'5
3.2x10^
Central
; Tendency
Not evaluated
Not evaluated
Management Underwater in a Quarry
This scenario simulated the increased potential for ground-water contamination resulting
from disposal of CKD in a quarry that subsequently fills with water due to ground-water seepage.
This sensitivity analysis scenario generated the highest ground-water estimates among all the
baseline and hypothetical scenarios. The best estimate ground-water effects, however, remained
below a cancer risk of IxlO"7, and more than four orders of magnitude below the noncancer
effects level. The high end individual cancer risks reached an upper bound value of about 7xlO"7,
while the noncancer hazard quotient was within one order of magnitude of a potential effect.
While this sensitivity analysis scenario was not designed to examine risks in the other
pathways, the results indicated that this scenario would have the lowest air, surface water, and
foodchain effects. Because the CKD is managed underwater and below grade, there is minimal
-------
6-55
potential for air emissions and erosion run-off, both of which were primary driving forces in the
soil and foodchain pathway effects.
Subsistence Level Food Consumption Risks
The Agency evaluated potential risks to individuals highly exposed through two
subsistence food consumption scenarios: subsistence farming and subsistence fishing. These
hypothetical scenarios evaluated, respectively, potential exposures to an individual that receives
75 percent of his/her vegetables, beef, and milk from sources contaminated by CKD, and an
individual that receives 75 percent of his/her diet of fish from a local stream contaminated by
CKD. These increased exposure assumptions were superimposed, in turn, on the baseline
analysis, the maximum dioxin sensitivity analysis, the adjacent agricultural field scenario (for the
subsistence farmer), and in the adjacent surface water body scenario (for subsistence fishing).
Thus, this sensitivity analysis examined the combined effects of high foodchain exposures with
several individual central tendency and high end risk settings. Consequently, at least in the
higher risk scenarios, this analysis tends to compound certain of these highly conservative
assumptions related to both the surface water and soil erosion pathways. Accordingly, these
results reflect worst case assumptions with a low probability of occurring, and should be
evaluated as an indication of the sensitivity of the baseline results to combinations of high risk
assumptions.
As would be expected, this scenario produced the highest estimates of risks from the on-
site management of CKD. Exhibit 6-20 shows the high end and central tendency cancer risks for
subsistence fishing and farming in the baseline analysis, the maximum dioxins analysis, and in the
respective adjacent locations sensitivity analyses.
Exhibit 6-20
Sensitivity Analysis of Subsistence Level Food Consumption Risks
Exposure
Pathway
Excess Individual Lifetime Cancer Risk
High End
Best Estimate
Upper Bound
Central
Tendency
Subsistence Fishing
Baseline Analysis
Maximum Dioxins
Adjacent Surface Water
2.0x10^
1.3xlO'2
3.5xlO'3
1.7X10'3
6.7X10'2 .
1.4X10'2
Less than 6xlO'7
Not estimated
Not estimated
Subsistence Farming
Baseline Analysis
Maximum Dioxins
Adjacent Agricultural Field
1.3xlO'5
4.7xlO'3
6-lxlO-4
2.0X10'5
7.2xlO-3
6.7X10-4
Less than 7x10"*
Not estimated
Not estimated
-------
6-56
The baseline analysis of highly exposed individuals estimated maximum risks at the
central tendency facilities of less than TxlO"6 for subsistence farming and less than 6xlO'7 for
subsistence fishing. The central tendency noncancer effects were generally within one order of
magnitude of the health effects threshold, but did not exceed a hazard quotient of 1. In reality,
the Agency does not believe that the average facility represented by the central tendency
estimate is likely to have subsistence-level exposures, as this is believed to be a relatively
uncommon practice. But these central tendency results suggest that were such individuals
located near CKD facilities, most would receive risks ranging below these values.
The high end baseline and sensitivity estimates indicate the greatest risk potential in these
two subsistence exposure scenarios. The subsistence fishing scenario results ranged from 2x10"*
to 7xlO~2, with the highest risks in the upper bound estimate associated with the maximum dioxin
concentration analysis. The subsistence farming results were somewhat lower, ranging from 1x10"
5 to 7xlO"3, with the highest risks again occurring in the maximum dioxin concentration scenario.
Generally, the high end subsistence level cancer risks were driven by dioxins, arsenic, and in
some cases beryllium, while the noncancer effects were driven by arsenic, cadmium, chromium,
and thallium.
6.2.3 Summary of Risks from On-site CKD Management
Based on a limited comparison, the sample of cement plants examined in this analysis
appears to be generally representative of typical cement plants across the nation in terms of
several factors that influence risks. By prioritizing the plants according to risk potential and
focusing the modeling on the five facilities that appear to pose the highest risks, EPA attempted
to quantify the upper range of the distribution of risks likely to be associated with the 15 case-
study plants. In addition, the analysis was designed to quantify the middle range of this risk
distribution as characterized by the "central tendency" estimates. The Agency recognizes that the
high end results do not necessarily capture the upper bound of the risks that exist across the full
universe of 115 active cement plants, as site-specific factors at some plants may contribute to
higher risks than estimated for the 15 sample facilities. Therefore, the Agency also conducted a
sensitivity analysis of several hypothetical scenarios representing combinations of potentially
higher risk scenarios that may exist at other facilities. The findings pertaining to each primary
exposure pathway are presented below.
Ground-water Risks
On-site CKD management practices and hydrogeologic conditions create a moderate
potential for ground-water contamination at most of the 15 case-study plants. For example, none
of the on-site CKD piles examined in the sample are equipped with a synthetic liner or other
engineered control to prevent the migration of contaminants to the subsurface, and most sites
exist in locations where the net recharge, depth to ground water, subsurface permeability, and
other factors could permit shallow ground-water contamination. However, the potential for any
such contamination to pose significant risks is diminished greatly by other factors at the majority
of sites, including relatively low concentrations of contaminants in CKD leachate, the tendency
for several CKD contaminants to sorb to soil and migrate very slowly in ground water, the
distance to potential downgradient receptors, and existing ground-water use patterns.
Considering all of these factors on a site-specific basis, the central tendency estimate of
individual risks for the ground-water pathway were low at each of the facilities modeled
(generally, significantly less than an increased individual cancer risk of IxlO'10 and noncancer
effects several orders of magnitude below the relevant effects thresholds). Even in the high end
-------
6-57
and sensitivity analyses, increased individual risks through ingestion of ground water never
exceeded IxlO"6. Additionally, no cancer cases or noncancer effects were predicted for the
populations surrounding the model facilities.
-------
6-58
Surface Water Risks to Human Health
The potential for significant human health risks from direct exposures to surface water
also appears low at present at most of the case-study plants, due to the lack of surface water
usage for drinking purposes downgradient of the facilities. Because the surface water was not
used for drinking water purposes, the risk modeling analysis examined exposures resulting from
recreational swimming. In the central tendency analysis, the human health effects were below an
individual cancer risk of 1x10"* and several orders of magnitude below relevant noncancer effect
thresholds, based oh a recreational swimming scenario assuming exposures from dermal
absorption and incidental ingestion of surface waters. In the high end analysis, the risk potential
was shown to be greater, with individual risks ranging from 4x10"* to 2xlO's. Important factors
contributing to the low central tendency risk estimates include the frequent practice of
intercepting and diverting stormwater run-off from CKD piles through on-site ditches prior to
discharge to surface water bodies, as well as the distance to and dilution capacity (high flow rate)
of receiving creeks and rivers. However, the high end and sensitivity analysis modeling results
indicate that higher risks from direct exposure to surface water may exist if stormwater run-off is
not adequately controlled and receiving waters have a negligible dilution capacity.
Potential human health risk estimates are higher for ingestion of fish from contaminated
waters. While central tendency estimate of effects from consumption of fish caught
recreationally were found to be less than 1x10"* for cancer and well below the noncancer effect
threshold, the high end results reached an increased individual cancer risk of about 4xlO"s and a
noncancer hazard quotient for cadmium at a level about ten times higher than its corresponding
threshold.
In cases where CKD facilities are located directly adjacent to a surface water body, both
the best estimate recreational swimming and fishing scenarios showed increased cancer risks and
noncancer effects roughly similar to the baseline high end estimates. In cases where facilities
manage CKD containing the highest concentrations of dioxins measured by EPA, however, the
estimated upper bound risks could exceed a cancer risk level of one in one thousand.
In cases where an exposed individual receives 75 percent of their fish from the
contaminated surface water body (a subsistence fisherman), the risk analysis predicted significant
cancer and noncancer effects. While this may be a relatively rare scenario at actual facilities, the
modeling analysis showed this practice to be of relatively significant concern were it to occur.
Aquatic Ecological Risks
The risk modeling analysis evaluated the potential for CKD constituents to exceed
aquatic ecological benchmark values in receiving surface waters near the plant. The central
tendency results showed no values exceeding chronic ambient water quality criteria (AWQC) for
the protection of aquatic life. In the high end analysis, five of the fourteen modeled constituents
were shown to have a potential for exceeding ecological levels of concern.
Air Pathway Risks from Windblown Dust
The air pathway is of concern for on-site CKD management because the dust is a fine
paniculate matter that is readily suspendable, transportable, and respirable in air. Many of the
sample facilities add water to CKD prior to disposal to form larger clumps or nodules in an
effort to keep the dust down, and some dust suppression is achieved naturally as thin surface
-------
6-59
crusts form on inactive portions of CKD piles as they are exposed to the elements. Nevertheless,
these appear to be temporary and incomplete measures of fugitive dust control at most facilities.
Quantitative modeling of air pathway risks to people living near case-study facilities
indicated that wind erosion and mechanical disturbances of on-site CKD piles do not result in
significant risks at nearby residences via direct inhalation (e.g., central tendency and high end
risks estimates were all less than 1x10"" increased individual cancer risk at all five facilities
modeled). However, fugitive dust from on-site CKD piles was estimated to be one of two
contributors in some cases to higher risk estimates for indirect exposure pathways (which were
primarily a result of direct surface run-off from the CKD pile reaching an agricultural field).
Central tendency foodchain cancer risk and noncancer effects for ingestion of vegetables,
beef, and milk, were below individual risks levels of IxlO"6 at all five facilities. In the high end
baseline facility scenarios, foodchain risks for ingestion of vegetables reached a maximum of
about 3x10"*. In the sensitivity analysis scenarios, however, these risks reached a maximum of
about 2x10"* due to uptake of maximum measured CKD dioxin concentrations in vegetables. The
estimated risks and hazards for the highly exposed subsistence farmer were significantly higher,
reaching a maximum cancer risk exceeding IxlO'2 in the upper bound sensitivity analysis scenario
that simulated the worst case dioxin concentrations. While the frequency of these less common
exposure scenarios is likely to be relatively low on a national basis, these risk estimates indicate a
potentially significant threat were they to occur.
6.3 EVALUATION OF RISKS FROM OFF-SITE BENEFICIAL USES OF CKD
As discussed in Chapter 8, approximately 943,000 metric tons (1,040,000 tons) of CKD
was sold or given away in 1990 for off-site beneficial uses. Most commonly, the dust is used to
stabilize hazardous and non-hazardous waste for disposal purposes. About 70 percent of off-site
CKD use in 1990 was for this purpose, which is approximately six times more than for any other
single use. The next most common off-site use is as a soil amendment, in which CKD mixed
with sewage sludge is used as a fertilizer, soil conditioner, or landfill cover. The third most
common single use is as a liming agent, in which raw CKD is land-applied directly to agricultural
fields. Together, the amount of CKD used as a soil amendment and liming agent accounts for
roughly 17 percent (160,000 metric tons) of the total quantity of CKD sold or given away in
1990. A number of other uses also exist, but they are much less common, both in terms of the
number of cement plants and quantity of CKD involved. For example, three cement plants sold
or gave away about 25,000 metric tons (3 percent of the total) to be used as an additive to
concrete and other building materials, and four plants sold or gave away approximately 11,000
metric tons (1 percent of the total) for use in the construction of roads.
This section evaluates the human health and environmental risks associated with these
various beneficial uses of CKD. It starts with an overview of the risk assessment approach and
methods. The section then evaluates the risks of the following major categories of beneficial
uses in turn: hazardous waste stabilization and disposal, sewage sludge stabilization and use,
building materials addition, road construction, and agricultural liming. Included in the discussion
of sewage treatment and use is an evaluation of the use of stabilized sewage as a landfill cover
(one example of soil amendment). Other uses of CKD as a soil amendment (e.g., soil stabilizer)
are not addressed because they are expected to pose similar, if not smaller, risks than the direct
application of CKD as a liming agent to food crops and pastures. Furthermore, additional uses
of CKD, such as an ingredient in livestock feed, a lime-alum coagulant, a mineral filler, an
ingredient in lightweight aggregate manufacture, and in glass making, are not evaluated in this
-------
6-60
chapter because of their limited use. These potential beneficial uses are described in detail in
Chapter 8.
6J.I Approach and Methods
As a basis for evaluating risks of off-site uses, EPA collected information on how and
where the dust is used. This information was obtained primarily through telephone interviews
with personnel at a sample of five principal independent companies that receive, process, and/or
market CKD at off-site locations. These companies are listed in Exhibit 6-21.
These five companies were selected for three reasons. First, the Agency selected a
number of off-site recipients that is roughly proportional to the relative frequency of each
category of off-site use: four recipients that mix CKD with either hazardous waste or sewage
sludge, and one recipient each for liming agent, road construction, and building materials
addition. Second, each company receives and handles a relatively large amount of CKD. With
one exception, the sites received more than 900 metric tons (1,000 tons) of CKD from more than
one cement manufacturing plant in 1990. Third, the sample of off-site locations represents a
diversity of geographical areas.
Exhibit 6-21
Off-site Beneficial Uses Examined in the Risk Assessment
RECEIVING
LOCATION
Farmland Ind.,
Coffeyville, KS
VFL Technology,
Matverne, PA
NewLime, Ravena, NY
National N-Viro
Energy Systems, Sioux
City, LA
U.S. Ash Inc.,
Roanoke, VA
BENEFICIAL USE
Hazardous Waste Stabilization
(petroleum refining sludge)
Landfill Cover
(sewage sludge mixture)
Waste Stabilization
Road Construction
Liming Agent
Soil Amendment
(sewage sludge mixture)
Materials Addition
(concrete admixture)
QUANTITY OF CKD
RECEIVED IN 1990
Metric Tons (Short
Tons)
123,000 (136,000)
19,000 (20,900)
53,000 (58,300)
8,000 (8,800)
23,000 (25,300)
6,000 (6,600)
10,400 (11,440)
Information was developed on productive processes at recipient companies and on basic
environmental features at locations where the dust is ultimately used. The Agency then analyzed
the factors influencing CKD release, transport, and exposure potential for each category of use,
considering the conditions that exist at the sample off-site locations. The purpose of this analysis
was to document and describe the major factors that could influence risks from each beneficial
use, and to prioritize the uses for further risk analysis through quantitative modeling.
6.3.2 Hazardous Waste Stabilization and Disposal
-------
6-61
Farmland Industries is a petroleum refinery that uses CKD to stabilize petroleum sludges
prior to land disposal. In 1990, Farmland received approximately 123,000 metric tons (136,000
tons) of CKD, accounting for roughly 17 percent of all the CKD used off-site for waste
stabilization that year.
Farmland has used CKD to stabilize a variety of petroleum refining wastewater treatment
sludges. The largest quantity of CKD was used as part of a project to close and renovate the
refinery's oily sludge ponds in early 1990. These unlined ponds held various wastewaters and
sludges, including API Separator Sludge (K051).24 Investigations conducted at the facility in the
early 1980s revealed that ground-water wells downgradient from the oily sludge ponds contained
elevated levels of lead, phenols, and hexavalent chromium,25 as well as a thick layer of oil on
top of the water.26 Therefore, before the land disposal restrictions for K051 became effective in
November 1990, Farmland closed the ponds by excavating all the sludge, mixing it with CKD to
stabilize it, and disposing of the mixture in a specially created landfill on top of the excavated
sludge ponds. The landfill was lined and capped with a local clay.
Since this large closure project in 1990, Farmland has continued to use smaller quantities
of CKD to stabilize other listed oil/water/solids separation sludges (i.e., F037 and F038).
Farmland presently sends these sludges mixed with CKD off site for disposal in a Subtitle C
landfill. Farmland indicates that once the land disposal restrictions for these other hazardous
sludges become effective in June 1994, it will begin using a reconfigured wastewater treatment
system that will eliminate the need to use CKD as a stabilizing agent.
The potential for CKD to cause or contribute to significant ground-water contamination
as it is used by Farmland appears remote. The liner and cap at the landfill containing stabilized
wastes from the old oily sludge ponds limit the extent to which water seeps through the wastes
and percolates into the subsurface. Monitoring wells have been installed around the landfill and,
according to Farmland personnel, have shown no sign of ground-water contamination thus far.
Additionally, the off-site landfill where Farmland presently sends its stabilized F037 and F038 is •
equipped with appropriate controls required under Subtitle C to minimize the risk of ground-
water contamination.
The containment provided at the on-site (oily sludge) landfill and off-site Subtitle C
landfill also serves to limit the potential for CKD to significantly contaminate surface water. For
example, the liner and cover used at the on-site landfill should significantly reduce the extent to
which landfill contaminants can migrate to the nearby Verdigris River, either via overland run-off
along with stormwater or via ground-water discharge. In addition, Farmland has constructed
dikes and berms to control flooding and limit the direct flow of stormwater run-off from the site
into the Verdigris River.
24 Environmental Priorities Initiative, Preliminary Assessment, Farmland Industries Site, Coffeyville, KS, Ecology and
Environment, Inc. prepared for U.S. EPA Hazardous Site Evaluation Division, October 15, 1990.
B Evaluation of the Potential for Migration of Hazardous Waste Constituents from the Disposal Site to Water Supply
Sources, Farmland Industries, Coffey\nlle, Kansas, Engineering Enterprises, Inc., 1991.
26 Inspection of Ground-water Monitoring, Farmland Industries, Coffey\ille, Kansas, Draft Report, Harding Lawson
Associates, 1984.
-------
6-62
Similarly, once mixed with hazardous waste, the dust exists in an oily mixture that is not
susceptible to wind erosion. This mixture is ultimately disposed in a covered landfill that
effectively prohibits the potential for significant airborne emissions.
Prior to mixing CKD with hazardous waste, Farmland will accumulate a maximum of 9
metric tons (10 short tons) of the dust in an unlined, uncovered pile at the site. Although there
is a potential for contaminants to migrate from this CKD pile into the environment, this
potential threat appears small compared to that posed by the much larger piles kept on-site at
some cement plants. Furthermore, there appears to be nothing unique about the environmental
setting at the site that leads EPA to believe that the threat of releases from this small pile at
Farmland is any greater than those evaluated for cement plants themselves.
Based on this case-study example, EPA believes that the use of CKD for hazardous waste
stabilization does not pose a significant threat to human health or the environment. When mixed
with hazardous waste, CKD is subject to full Subtitle C regulation. In fact, solidification with
CKD and other similar agents has been designated as the Best Demonstrated Available
Technology for the disposal of several metal-bearing wastes that exhibit a hazardous waste
characteristic (55 FR 22520; June 1, 1990). Small quantities of CKD are handled and possibly
released into the environment at off-site use locations before the dust is mixed with hazardous
waste, but the risks associated with these releases are expected to be minimal. For these reasons,
the Agency did not perform quantitative risk modeling for hazardous waste stabilization.
6.3.3 Sewage Sludge Treatment and Use
CKD is commonly used in the treatment of sewage sludge that is then used as landfill
cover, fertilizer, or soil conditioner. One treatment approach, the N-Viro process, accounts for a
large amount of all of the CKD used in this manner. As discussed in more detail in Chapter 8,
the N-Viro process combines CKD with sewage sludge through a patented reaction to produce a
"soil-like product."
To evaluate potential risks associated with the use of N-Viro soil, the Agency contacted
two vendors that have licensed the process: National N-Viro Energy Systems in Sioux City, IA,
and VFL Technology in Malverne, PA. National N-Viro produces and sells N-Viro soil for many
uses (e.g., landfill cover, soil fertilizer). VFL, in contrast, operates a production facility at the
Middlesex County Municipal Landfill in Middlesex, NJ for the exclusive purpose of producing
landfill cover. The VFL plant has an N-Viro soil production capacity of 120 dry tons per day, 7
days per week. In evaluating risk potential, the Agency focused on the use of N-Viro soil as a
landfill cover. Use as a soil fertilizer or conditioner is expected to pose similar, if not smaller,
risks than the direct application of CKD as a liming agent (evaluated in Section 6.3.6).
The potential for contamination and adverse effects through the ground-water, surface
water, and air pathways appears minor when CKD is combined with municipal sludge and used
as a landfill cover. At the Middlesex landfill, for example, the landfill itself must meet basic
design and operating standards for Subtitle D municipal landfills, including standards designed to
limit the seepage of constituents through the landfill base.27 Ground-water monitoring also is
v In October 1991, EPA promulgated expanded criteria in 40 CFR Parts 257 and 258 for solid waste disposal
facilities regulated under Subtitle D of RCRA, including the co-disposal of sewage sludge with household wastes in
municipal solid waste landfills (56 FR 50978, October 9, 1991). This rule set forth minimum federal criteria for
municipal solid waste landfills like the Middlesex County Landfill, including location restrictions, facility design and
-------
6-63
conducted, and according to VFL personnel, no ground-water contamination has been detected
since the plant began operation in 1991. The Middlesex landfill also is equipped with berms and
dikes to limit stormwater run-on/run-off and subsequent contamination of surface water. The
greatest potential for air releases exists during transport of the raw dust to the N-Viro facility.
VFL, however, reportedly transports the dust in covered trucks and transfers it directly into the
plant via enclosed pipelines. Once CKD is combined with wet sludge, the dust particles are
bound to the mixture and are prevented from being suspended in the air. When dried, the
potential for airborne releases from the N-Viro product is limited because the fines are bound in
large soil-like clumps.
Although there is some potential for the highly alkaline nature of CKD leachate to
mobilize certain trace metals that exist in sewage sludge, this threat appears substantially limited
by physical processes and existing regulatory and administrative controls. When added to sewage
sludge, CKD raises the pH and chemically binds most heavy metals in the sludge. For example,
barium, beryllium, cadmium, copper, mercury, nickel, lead, thallium, and zinc tend to be more
immobile in ground water under high pH conditions than under low or neutral pH conditions;
the reverse tends to be true only for arsenic, hexavalent chromium, antimony, molybdenum, and
selenium.28 In addition, EPA recently promulgated technical and permitting regulations that
apply to sewage sludge beneficial use and disposal practices (40 CFR Part 503). Thus, fertilizers
and soil amendments derived from CKD-sewage sludge mixtures pose minimal risk because these
final products are required to be tested to assure they comply with all provisions of 40 CFR 503,
which are fully protective of human health and the environment. N-Viro routinely analyzes their
sewage sludge to assure compliance with concentration limits established in this "clean sludge"
rule for arsenic, cadmium, chromium, copper, lead, molybdenum, mercury, nickel, selenium, and
zinc. According to N-Viro personnel, this testing has not detected any exceedances of the clean
sludge levels since the facility opened in 1991.
Based on this review, mixing CKD with sewage sludge for use as a municipal landfill
cover does not appear to pose a threat to human health or the environment, and the Agency did
not undertake more detailed risk analysis through modeling.
6-3.4 Building Materials Addition
To evaluate potential hazards of adding CKD to building materials, EPA contacted U.S.
Ash, Inc. in Roanoke, VA, which purchased approximately 43 percent of all of the dust used for
this purpose in 1990. U.S. Ash uses CKD to replace cement in general use concrete. Thirty
percent of the cement is replaced with CKD and fly ash in equal proportions (i.e., 15 percent of
the cementitious product from U.S. Ash is CKD). U.S. Ash does not purchase dust from kilns
that burn hazardous waste. The dust is added in dry form to the cement, which is sold to many
different customers and used in many different applications.
The possible scenarios for using CKD-containing cement are as numerous and diverse as
those that exist for normal cement, and can include its use as water distribution pipelines and
structural members of buildings and bridges. Therefore, it is difficult to generalize about
potential exposure scenarios associated with this category of use.
operating criteria, ground-water monitoring requirements, corrective action requirements, financial assurance
requirements, and closure and post-closure care requirements.
M Based on soil-water partition coefficients (K^'s) in EPA's Corrective Action chemical database.
-------
6-64
One generalization that may be possible, however, is that dust or leachate from CKD-
containing cement is unlikely to be significantly different in composition than that from normal
cement. This is based partly on the fact that CKD is mixed with cement in only small
proportions. Side-by-side leach test data for trace metals published by PCA also suggest that the
composition of leachate from cement and CKD are similar.2' Although these PCA data
indicate that the concentrations of relatively volatile metals (mercury, selenium, thallium,
cadmium, and lead) may be 13 to 40 times higher in CKD leachate than cement leachate, TCLP
tests of both materials yielded metal concentrations that were non-detectable and/or below TC
regulatory levels in virtually all cases. Perhaps more relevant results are provided by an
independent study30 that found that metals concentrations in leachate from concrete products
were below detectable limits for all metals tested with the exception of chromium, which was
measured at 72 ppb, well below the chromium MCL and TC regulatory level. These results
indicate that metals, once bound into the cementitious matrix, are unlikely to leach from cement
in appreciable quantities and probably do not pose a risk via waterborne pathways.
Similarly, once CKD is locked into concrete, the potential for airborne releases appears
low. A potential for air releases does exist during materials handling prior to forming of the
concrete, analogous to those observed at the cement production facilities. A potential for
fugitive dusting from concrete also exists during use and when the concrete is cut apart or
broken up, either in construction or demolition projects. Such releases, however, would be
temporary and the amount of dust emitted to the air is likely to be small compared to that
emitted from the large, uncovered CKD piles at cement plants.
For these reasons, EPA believes that the use of CKD as an additive in building materials
is not likely to result in significant incremental releases of contaminants to the environment.
Additional modeling to quantify risks from this type of use was not conducted.
6.3.5 Road Construction
General evaluation of risk factors suggests that use of CKD in road construction could
present a potential threat greater than the other uses discussed above. To evaluate this threat in
greater detail, the Agency performed quantitative modeling of a road construction scenario.
Analysis of Risk Factors
NewLime in upstate New York distributed almost 8,200 metric tons (9,000 short tons) of
the CKD used in road construction, or approximately 76 percent of all CKD used for this
purpose in 1990. Based on telephone interviews with NewLime personnel, CKD can be used in
three different ways for road construction: as a road sub-base, mixed with asphalt that is used
for the road surface, and in the construction of unpaved roads and parking lots. The potential
for releases into the environment varies with these different types of uses.
The potential for releases appears small when the dust is used as a road sub-base. In
these situations, the dust is usually mixed with gravel and fly ash. Because CKD and fly ash are
29 Portland Cement Association, 1992. An Analysis of Selected Trace Metals in Cement and Kin Dust. Skokie, 1L.
30 Kriech, Anthony J., Leachability of Asphalt and Concrete Pavements, Heritage Research Group,
March 1992.
-------
6-65
pozzolanic,31 the sub-base sets up to form a solid layer that binds the CKD constituents in
place. Leaching and migration from the sub-base also is expected to be limited by little to no
direct contact with water, as the sub-base is overlain by relatively impermeable asphalt or
concrete. The primary occasions when water may flow under the road and leach CKD
contaminants are likely to be associated with freeze/thaw conditions. In addition, there appears
to be little potential for windblown dusting, except during the actual application of CKD as a
sub-base and the brief period that it is uncovered.
When used as an additive to asphalt, dust in the asphalt could be submerged during
rainfall. In principle, CKD constituents could leach from the asphalt mixture and migrate to
ground water or flow overland to surface waters. However, a study by the Heritage Research
Group32 shows that metals normally present in asphalt do not tend to leach in appreciable
concentrations during TCLP tests. For almost every metal tested, concentrations in asphalt
leachate were below detectable limits. The only exception was chromium, which was detected in
asphalt leachate at a level of 0.10 ppm, 50 times below its TC regulatory level. It is not known
how CKD affects the leachability of asphalt, if at all. However, the generally low concentrations
of chemicals observed in CKD leach tests and the relatively small proportion (five percent or
less) of CKD that is mixed with asphalt suggest that asphalt mixed with CKD would produce
leachates very similar to asphalt by itself. The potential for airborne releases when CKD is used
in asphalt also appears low because the dust is locked into a matrix through a pozzolanic
(hardening) reaction. The presence of small proportions of CKD is not expected to significantly
affect the quantity and quality of particulates that are suspended from the asphalt during road
use.
A greater potential for CKD contaminants to migrate into the environment appears to
exist when the dust is mixed with clayey soils to form unpaved roads or parking lots. In these
cases, the dust may be applied in an indiscriminate manner that is not designed to optimize a
pozzolanic reaction. Moreover, the dust is not covered by a hardened road surface like asphalt
or concrete, and engineered controls are not used to prevent CKD contaminants exposed to the
elements from leaching into the subsurface or migrating to any nearby fields or surface waters
along with storm water run-off. CKD also could be blown into the air by the wind, and vehicular
traffic both during and after construction could periodically and temporarily suspend paniculate
matter into the air. The primary factors that would influence the amount of CKD suspended in
the air include the particle size of the material on the road surface, traffic volumes, the speeds
and other characteristics of vehicles (e.g., number of wheels and weights), and rainfall patterns.
Based on this evaluation, there does not appear to be a significant human health or
environmental risk associated with the use of CKD as either a road sub-base or an additive to
asphalt. However, since there appears to be a greater potential for releases of CKD
contaminants and subsequent exposures when the dust is used in the construction of unpaved
roads or parking lots, these risks were studied in greater detail through quantitative modeling.
Jl That is, finely divided siliceous or siliceous and aluminous material that reacts chemically with slaked lime at
ordinaiy temperature and in the presence of moisture to form a strong, slow-hardening cement.
52 Kriech, Anthony J., Evaluation of Not Mix Asphalt for Leachability, Heritage Research Group,
October 1992.
-------
6-66
Risk Modeling Results for Unpaved Traffic Surfaces
The Agency employed the same basic modeling methodology for quantifying the risks
from off-site use of CKD for unpaved roads and parking lots as was used in the on-site analysis
(Section 6.2.2). The primary difference relates to the design of the road paving scenario.
Because such uses may take place in virtually any location in the U.S., this risk scenario is largely
hypothetical and was developed using best professional judgment. Thus, the results from this
analysis should be considered rough indications of the kinds of risks that might correspond to
this CKD management approach.
The Agency simulated direct addition of CKD to the other materials (clayey soils and
aggregate) used in the construction of an hypothetical off-site parking lot. Because MMSOILS
requires a square source term, it cannot effectively simulate releases from a long thin source such
as would be required to simulate a road. Accordingly, the analysis was limited to the use of
CKD in unpaved parking lots. The release modeling for unpaved parking lots considered three
of the four pathways evaluated in the on-site modeling: ground water, air, and surface water.
Only one exposure pathway in the foodchain pathway was evaluated: ingestion of fish in a
nearby stream.
Based on this modeling, the estimated risks associated with use of CKD as a surface for
unpaved parking lots were generally quite low. None of the pathways examined were found to
have cancer risks exceeding the 10"* risk range or noncancer effects exceeding the threshold dose.
The highest risks were in the foodchain pathway (for ingestion of fish in the nearby
stream receiving run-off from the parking lot). The only foodchain pathway effect evaluated for
unpaved parking lots corresponds to the ingestion of fish caught recreationally in the nearby
stream. The increased individual cancer risk associated with recreational fishing was estimated to
be about IxlO'7, due to exposures to 2,3,7,8-TCDD. Noncancer effects were about two orders of
magnitude below the effects threshold, with thallium representing the highest intake to reference
dose ratio.
The maximum ground-water risks from the unpaved parking lot were estimated at 5.3x10"
9, which were driven by potassium-40. The noncancer effect was nearly seven orders of
magnitude below the effect threshold. The low ground-water effects resulted from the low
permeability of the unpaved surface (resulting in minimal leachate generation) combined with the
relatively low concentrations of the CKD constituents in the leachate generated by the parking
lot material.
The increased individual cancer risks through exposure to air emissions were estimated to
be 1.4x10'" to the individual living closest to the parking lot. Noncancer effects were negligible
and could not be quantified. These low air risks reflect the small size of the unpaved parking lot,
which is unlikely to serve as a large enough source to result in elevated ambient concentrations
of CKD constituents in the air.
The estimated maximum risk resulting from dermal absorption and incidental ingestion of
water while swimming was 2.4 x 10"', and was due primarily to arsenic. Noncancer effects were
estimated to be about five orders of magnitude below the health effects threshold.
-------
6-67
6.3.6 Agricultural Liming
Because of the potential for bioaccumulation and the direct ingestion of contaminated
food products, CKD used as a liming agent appears, on first evaluation, to pose more of a
potential risk than any other CKD use. To further explore this risk potential, EPA performed
quantitative modeling of an agricultural liming scenario.
Analysis of Risk Factors
Like agricultural lime, CKD is alkaline and contains a number of essential plant nutrients.
According to the 1991 PCA Survey and the RCRA §3007 responses, approximately 53,000 metric
tons (58,000 tons) of dust were sold or given away in 1990 for liming from five cement plants
(one plant each in New York, Pennsylvania, and Kansas, and two in Idaho). To evaluate this
application, the Agency contacted the NewLime Company in upstate New York. NewLime
distributes dust to over 1,600 farmers and accounted for approximately 46 percent of the total
dust used for liming in 1990.
NewLime is the exclusive CKD agent for Blue Circle Cement in Ravena, NY. The dust
is transported from Blue Circle to storage silos at NewLime via enclosed trucks. CKD in storage
at NewLime is subsequently transported from the silos to specific points of application via bulk
tanker trucks. The dust is not modified in any way prior to application. The typical point of
application is a 41-hectare (100-acre) farm that grows alfalfa, corn, and soybeans for livestock
feed. Alfalfa and corn account for about 90 percent of the crop output. Approximately half of
the CKD from NewLime is applied in New York.
Liming may occur during any season of the year with the majority occurring in the fall.
Once the dust arrives at a farm, it is placed in spreader boxes of spreader trucks. These boxes
commonly hold up to 11 metric tons of CKD and measure approximately 10 meters (33 feet) in
width. Two and a half centimeter (1-inch) diameter holes on the bottom of the spreader boxes
are spaced every 10 centimeters, which enables CKD to be applied evenly to the fields. CKD is
applied in four steps. The farmer first disks the soil and harrows the ground; CKD is then
spread; the soil is disked again; and the farmer plows a final time. CKD is usually tilled to a
depth of 15 to 20 centimeters. Typically, 4.5 metric tons of CKD are spread per hectare with
CKD application occurring once every three to five years. This is the same as the application
rate for regular lime.
A paucity of available data on the composition of agricultural lime prevents a complete
comparison of CKD and lime in terms of trace contaminant concentrations. However, a
preliminary analysis suggests that, compared to CKD, agricultural lime can contain higher totals
concentrations of some constituents (such as barium), about equal concentrations of some
constituents (such as chromium), and lower concentrations of other constituents (including lead,
nickel, silver, vanadium, and copper).33 Agricultural lime would not be expected to contain
dioxins because it is simply crushed limestone, and not, like CKD, manufactured in a combustion
process along with chlorine precursors that might yield dioxins.
The potential for ground-water contamination from liming is a function of the amount of
CKD applied, dust leachability, and the particular environmental conditions that exist at a farm
51 Boynton, Robert S., Chemistry and Technology of Lime and Limestone, Second Edition, John Wiley & Sons, Inc.,
New York, Chichester, Brisbane, and Toronto.
-------
6-68
(e.g., rainfall and recharge rates, soil chemistry and permeability, and depth to ground water).
As noted above, the dust is applied infrequently in small amounts, just like regular lime. No
data are available, however, to compare contaminant concentrations in leachate from regular
lime to those measured in CKD leachate extracts. The environmental conditions where the CKD
is applied may vary widely because CKD liming takes place not only throughout upstate New
York, but also in a few other locations in the United States. At some sites, these conditions may
be conducive to ground-water contamination, such as when net recharge rates are high, soils are
permeable, and ground water is shallow. Furthermore, because ground water in rural areas
around farms is often used for drinking and other purposes, any ground-water contamination
associated with the use of CKD as a liming agent may have the potential to result in human
exposures.
Similarly, there is a potential for this use of CKD to result in surface water
contamination. The only measures that may exist to prevent CKD contaminants from migrating
into surface waters are likely to be occasional irrigation ditches and agricultural management
techniques designed to preserve topsoil, such as terracing. Vegetation may slow run-off to
surface waters during the growing season, but when CKD is applied initially, little or no
vegetation exists. Even during the growing season, much bare soil is exposed to the elements in
fields with row crops. Factors such as CKD application rates, CKD properties (e.g., chemical
composition and leachability), annual rainfall, the slope of the land, the nature of on-site soils,
the extent of crop cover, and the distance to surface waters will all contribute to the potential for
surface water contamination.
In general, when properly handled, the potential for release to air when CKD is used as a
liming agent appears smaller than the potential for release to ground water and surface water.
In the specific example of NewLime, the dust is covered and contained during all phases of
storage and transport prior to the time it is applied to a field. In particular, CKD is transported
in enclosed trucks to the NewLime storage facility, where it is then stored in enclosed silos. The
dust is then transported from NewLime to individual farms in enclosed tanker trucks where it is
placed in enclosed spreader boxes. The dust is dropped only centimeters above the ground and
quickly tilled into the soil; it is not broadcast in the air and then allowed to settle onto the
ground. In the Agency's telephone interview, NewLime personnel indicated that little dust
becomes airborne even on windy days.
The greatest potential for contaminant exposures resulting from the use of CKD as a
liming agent is through the foodchain. Crops cultivated in fields limed with CKD by NewLime
are used as feed for livestock. CKD constituents, therefore, may be ingested directly by animals
and concentrated in food products (milk, meat) that are ingested by humans.
Risk Modeling Results for Liming
The Agency conducted a quantitative analysis to estimate the potential magnitude of risks
resulting from the agricultural use of CKD. As in the unpaved road analysis, the modeling
methodology for the agricultural applications of CKD was based largely on the approach used in
analyzing on-site risks. The primary differences concern the focus on the two foodchain
exposure pathways relevant to agricultural applications: vegetables, and beef and milk. Because
the CKD is assumed to be tilled directly into the soil, this analysis did not quantify potential
impacts to surface water, air, or fish ingestion, as it is assumed that these results would be
significantly lower than risks from the ingestion of agricultural products. Another difference
from the on-site modeling analysis included the simulation of three basic risk scenarios: best
-------
6-69
estimate, high end, and upper bound. The best estimate analysis assumed the best estimate CKD
application rate and CKD concentrations representing CDD/CDF values from a facility that has
been documented as selling CKD for use as a liming agent, and 50th percentile values for the
metals and radionuclides from the Agency's database on CKD concentrations. The high end
analysis used a high end CKD application rate and CKD concentrations corresponding to the
highest risk potential wastestream from the five baseline facilities. Finally, the upper bound
value used the high end application rate, and CKD concentrations from the facility with the
highest measured CDD/CDF concentrations from the Agency's sampling and analysis program.
The Agency simulated direct incorporation of CKD on a hypothetical agricultural field
assumed to grow corn or alfalfa for use as cattle feed. The field used in the simulation
represented a typical 41-hectare (100-acre) field. The analysis employed an assumed CKD
loading rate of 4.5 metric tons per hectare (2 tons per acre) every four years in the best estimate
and the same loading every two years in the high end and upper bound analyses.
The best estimate results for the liming agent analysis showed the following for the three
foodchain exposure scenarios examined. The highest best estimate risks were in the subsistence
farmer scenario, reaching a maximum cancer risk of VxlO"6 for arsenic, with the next highest risk
resulting from beryllium exposures (at 4.7xlO"7). Ingestion of vegetables (non-subsistence)
resulted in a risk of 8.4xlO'7 for arsenic, while the ingestion of beef and milk resulted in a risk of
l.lxlO-7.
The high end results (more frequent every two year application of CKD with higher
constituent concentrations) exceeded those in the best estimate by one to two orders of
magnitude. The subsistence farming scenario had the highest total cancer risks of 2.5xlO'5,
resulting equally from exposures to arsenic and 2,3,7,8-TCDD equivalents. The risks for
ingestion of vegetables were 1.7x10"* (due to arsenic), while the risks from beef and milk
ingestion were IxlO"6 due to 2,3,7,8-TCDD equivalents and arsenic.
The upper bound scenario simulated the tilling of CKD with EPA's highest measured
CDD/CDF concentrations in the field. This scenario produced the highest risk estimates, with a
maximum risk of 2.1x10"* for the subsistence farming scenario (due primarily to 2,3,7,8-TCDD
equivalents), a cancer risk of 1.7xlO"s for beef and milk ingestion (dominated by 2,3,7,8-TCDD),
and a cancer risk of l.lxlO'5 for the ingestion of vegetables (resulting from 2,3,7,8-TCDD,
arsenic, and beryllium).
None of the noncarcinogens exceeded the effects threshold in the liming agent analysis,
although several constituents resulted in hazards within one order of magnitude of the threshold
(i.e., a hazard ratio between 0.1 and 1.0): antimony (high end subsistence farming only),
cadmium (all subsistence farming scenarios and high end vegetables), and thallium (subsistence
farming and beef and milk ingestion).
6.3.7 Summary of Risks from Off-site Beneficial Uses
By far, the most common off-site use of CKD is for waste stabilization, both for
hazardous and non-hazardous waste prior to disposal and non-hazardous waste (municipal
sewage sludge) prior to beneficial use. Based on an evaluation of the conditions that exist at
sample off-site locations where CKD is used, EPA believes that these uses do not pose a
significant threat to human health or the environment. Hazardous waste stabilization presents a
low risk because CKD mixed with hazardous waste is subject to full Subtitle C regulation,
-------
6-70
including requirements for disposal in lined units to prevent ground-water contamination,
appropriate run-on and run-off controls to prevent surface water contamination, and capping of
landfills upon closure to prevent air releases. Releases to various media are further minimized
because CKD is generally mixed with sludges to form a stabilized solid that is less susceptible to
dispersion (e.g., via wind erosion) than CKD by itself. For non-hazardous waste stabilization, the
risks are also expected to be small because at least half of the CKD used in this manner in 1990
was used in the N-Viro process, which combines CKD with sewage sludge. Similar to Subtitle C
regulations for hazardous wastes, sewage sludge disposal is controlled by recently promulgated
permitting regulations (40 CFR Part 503). These regulations set forth concentration limits for
metals in sludge before disposal. Compliance monitoring of stabilized sludge at a sample off-site
use location indicates that no exceedances of clean sludge levels have occurred.
Three other off-site uses — road sub-base, additive to asphalt, and materials addition —
also do not pose significant risks. When used for these purposes, CKD is mixed with other
materials, such as asphalt or cement, to form a solid matrix. In this form, it is unlikely that the
CKD will contaminate environmental media because: (1) the CKD makes up only a small
fraction of the total solid matrix (e.g., less than five percent in the case of asphalt mixtures); and
(2) the solid matrix is generally not susceptible to significant releases to ground water, surface
water, or air.
Preliminary evaluation identified two types of uses that could have a greater potential to
pose risk to human health and the environment: agricultural liming and construction of unpaved
roads and parking lots. The primary risk conclusions for these off-site uses are as follows:
• For agricultural liming, releases to ground water and surface water are possible
due to leaching and surface run-off. Air releases are not expected to be
significant because the dust is covered and contained at all times during transport
and delivery, dropped only centimeters from the ground during application, and is
quickly tilled below the surface. EPA's modeling predicted potential risks via the
foodchain pathway for this practice for ingestion of vegetables from the field, beef
and milk raised on feed from the field, and most significantly, for a farmer
subsisting on both vegetables, beef, and milk raised from the field. The best
estimate cancer risks reached a maximum of 7x10"*, while the maximum high end
risks were 2.5xlO"5. In the bounding analysis, the subsistence farming scenario
showed the greatest risk potential with a risk estimate of 2.1X10"4.
• For use in unpaved roads and parking lots, releases to ground water, surface
water, and air could occur because the CKD is not fixed in a solid matrix, but is
slightly compacted, exposed to the elements, and disturbed by vehicular traffic.
However, the Agency's modeling predicted very low risks (less than 5x10'') for the
ground-water, air, and surface water pathways, and only IxlO"7 for the worst-case
scenario of fish ingestion in the adjacent surface water body. Noncancer risks
were found to be below the combined effects threshold for all pathways evaluated.
-------
CHAPTER SEVEN
EXISTING REGULATORY CONTROLS ON CKD MANAGEMENT
7.0 INTRODUCTION AND METHODS
7.0.1 Objectives
EPA's objective in this analysis was to identify and evaluate the existing regulatory
controls that pertain to management of CKD. The Agency carried out this analysis in
accordance with the spirit of RCRA Section 8002(o), which suggests that EPA "review studies
and other actions of other Federal and State agencies... with a view toward avoiding
duplication of effort." The regulatory analysis also served more generally to help characterize
current waste management practices. This knowledge will guide the development of the Agency's
position on any additional CKD management regulations that EPA may deem appropriate.
7.0.2 Methodology
EPA has addressed federal, state, and local regulations, based on the environmental
media that they were established to protect. Therefore, the Agency examined those regulations
that protect air quality and surface and ground water. In examining existing management
controls on CKD, the Agency evaluated both the strengths and areas needing improvement for
the present regulatory framework. EPA also evaluated cross media impacts created by
regulation; one example of such an impact is the effect that air pollution control devices,
installed to remove dust from cement plant exhaust gases, had on the prevalence of collecting,
landfilling, and storing CKD.
In the initial phase of the analysis, EPA examined the relevant statutes and regulations
pertaining to air quality, water quality, and solid waste as they might apply to the management of
CKD. To develop a baseline of information about current federal and state regulations, EPA
conducted an on-line search of the Computer-Aided Environmental Legislative Data System
(CELDS), a data base containing abstracts of federal and state environmental regulations. By
querying CELDS with various combinations of key words, such as "cement plants," "dust," and
"fugitive emissions," the Agency obtained abstracts of federal and state environmental regulations
that might affect on-the-ground management of CKD.
EPA identified and evaluated the existing federal regulatory controls on CKD, focusing
on programs-and requirements established by EPA. This characterization is necessary for two
reasons. First, some states do not have EPA-approved programs for regulating air pollution
emissions to the atmosphere or discharging contaminants to surface waters. In those states,
federal EPA regulations take precedence. Second, the federal government has not delegated
authority to the states for implementing some environmental protection statutes and regulations
and is thus responsible for their implementation. EPA contacted EPA Regional staff in those
states that do not have federally approved programs for implementing the major environmental
statutes (e.g., RCRA, The Clean Water Act [CWA]), and performed a detailed regulatory
analysis of the implementation of existing federal statutes and regulations that pertain to the
management of CKD.
-------
7-2
The next step of EPA's analysis was to perform a more detailed review of statutes and
regulations in four selected states. Based on time and resources, EPA limited this review to four
of the states with the largest clinker production and finished grinding capacities. Together,
California, Michigan, Pennsylvania, and Texas represent over 35 percent of 1990 clinker
production capacity. EPA assumed that these states would have the most extensive experience in
controlling the management of CKD and would have the greatest interest in regulating CKD.
Based on the state statutory and regulatory language, EPA discovered that the scope of
state programs in these four states was not always clear. Therefore, EPA contacted state and
local officials involved with implementing CKD management requirements to learn how those
statutes and regulations are interpreted in practice, and to obtain facility-specific implementation
information. The information compiled from these contacts was combined with the existing
information on statutory and regulatory requirements to produce the final implementation
analysis.
7.0.3 Summary of Findings
The control of stack emissions at cement plants during the last 30 years may have
prompted an increase in landfilling and stockpiling of CKD as air pollution control devices
removed non-useable dust from kiln exhaust gases.1 Prior to regulation of air pollutant
emissions, CKD was released into the atmosphere from stack and fugitive emissions with little or
no control. More recently, air pollution control has been expanded to regulate fugitive dust
emissions from non-stack sources (e.g., storage piles and transportation equipment).
As cement plant stack and fugitive emissions have been increasingly regulated, the
generation and long-term management and disposal of CKD have become solid waste
management issues. States have become concerned about contamination of surface and ground
water from improper management of solid wastes. This growing concern has manifested itself in
closer solid waste regulatory control of CKD at the state level. More stringent solid waste
requirements for landfilling and stockpiling may in turn be a factor in the rise of CKD recycling
and beneficial use.
On the federal level, air quality has been improved through implementation of controls
on releases of CKD through kiln stacks and via fugitive dust emissions. Under the New Source
Performance Standards (NSPS) for cement plants, a facility must comply with specific emission
limitations for paniculate matter. Prevention of Significant Deterioration (PSD) review also is
required for a cement plant located in a geographic area that is classified as an attainment area.
In addition, cement plants are subject to Nonattainment Review if they are located in an air
quality control area that is not in compliance with the National Ambient Air Quality Standards
(NAAQS) for a given pollutant (e.g., paniculate matter or sulfur dioxide). Cement kilns that
burn hazardous waste fuels also are being controlled under new regulations for the Burning of
Hazardous Waste in Boilers and Industrial Furnaces (BIF), which imposed new controls on those
facilities.
At the state level, air quality requirements for CKD management incorporate the federal
standards as a baseline but many states have established additional, more stringent requirements.
Individual states subject cement plants to visible emission or opacity limitations that are more
1 Personal communication with Gary Linns, Pennsylvania Department of Environmental Resources, Bureau of Air
Quality, October 9, 1992.
-------
7-3
stringent than the federal standards; for example, Pennsylvania requires opacity to be measured
over a three-minute period rather than the six-minute period in the federal standard. Texas also
requires notification of "excessive emissions" and establishes paniculate matter ground level
concentration limits. Michigan and Pennsylvania have established fugitive dust control programs;
Pennsylvania has actually required two cement plants to store CKD and clinker in a warehouse
or silo to control fugitive emissions. California, Michigan, Pennsylvania, and Texas all require
permits to construct and operate cement plants. In all states, the permit review process is
designed to ensure that emissions from such sources will not interfere with the attainment and
maintenance of ambient air quality standards.
Similarly, treatment of wastewater and stormwater run-off from cement plants has been
beneficial in maintaining and improving water quality. National Pollutant Discharge Elimination
System (NPDES) permits establish effluent limits on suspended solids, pH, toxic pollutants such
as heavy metals, and material run-off from CKD storage piles. Monitoring and reporting
requirements assure compliance with the applicable effluent limitations, water quality standards,
and pretreatment standards. In addition to controls on process wastewater from cement plants,
stormwater run-off is also regulated. Under new federal stormwater regulations, pollution
prevention plans and Best Management Practices (BMPs) should continue to reduce the amount
of uncontrolled CKD contained in stormwater discharges.
In states where responsibility for implementation of the NPDES program to regulate
discharges to surface water has been delegated, such as California, Michigan, and Pennsylvania,
only one state water permit regulating the discharge of pollutants is required. In Texas, where
delegation of authority has not occurred, both federal and state permits must be obtained.
Cement plants in any of the states generally are subject to State Pollutant Discharge Elimination
System and pretreatment regulations that are either identical or similar to the federal
requirements. Each of the four states has adopted both descriptive (e.g., all waters should
support agricultural use) and numeric surface water quality standards.
In addition, state regulation of discharges has been expanded beyond the scope of the
NPDES to include discharges to ground water. Ground-water protection requirements include
facility siting restrictions, design standards, ground-water monitoring, and the designation of
wellhead protection areas. Michigan has established ground-water quality regulations. Similarly,
Pennsylvania's Pollutant Discharge Elimination System program applies both to streams and
ground water. California's ground-water protection policy includes closely regulating a number
of potential sources of ground-water degradation, such as waste management facilities.
In implementing the federal stormwater requirements, each state has established different
permit issuance policies. California has decided not to issue permits specific to industrial
categories, but has established a general permit that applies to all stormwater dischargers.
Cement plants in Pennsylvania and Texas, where delegation has not occurred, will be subject to a
general permit program that is industry-specific. In Michigan, on the other hand, individual
cement plants or groups of plants must apply for stormwater permits.
While CKD is temporarily excluded from regulation under Subtitle C of RCRA, CKD is
still subject to regulation as a non-hazardous solid waste under RCRA Subtitle D. In the area of
solid waste management, CKD is landfilled or stored in piles on site at many cement plants.
Under Subtitle D, flexibility exists for states to implement requirements for industrial non-
hazardous waste and this flexibility results in a diverse collection of state Subtitle D programs.
Enforcement of Subtitle D is primarily a state's responsibility. The federal government, however,
-------
7-4
has the authority and resources under the Comprehensive Environmental Response,
Compensation, and Liability Act (CERCLA, or Superfund) to respond to situations in which
CKD is or may be released into the environment such that it poses an imminent and substantial
danger to human health and the environment. Two cement kiln facilities are currently on the
National Priorities List (NPL) in response to CKD-related environmental damage.
Because the Bevill Amendment does not preclude more stringent (e.g., hazardous waste)
requirements for CKD at the state level, states (such as California) are free to characterize CKD
as a hazardous waste; indeed, several differences in CKD management requirements exist
between the states. However, California does not enforce the management of CKD as a
hazardous waste because of recent legislation imposing a one year moratorium on enforcement
of these requirements pending a study of CKD (See Section 7.3, subsection 7.3.2 for further
discussion). Pennsylvania, on the other hand, classifies CKD as a residual waste, regulating CKD
less stringently than if it was considered a hazardous waste, but still requiring comprehensive
waste management practices. Michigan and Texas characterize CKD as an industrial, non-
hazardous solid waste and, therefore, subject CKD to fewer management requirements than
either California or Pennsylvania. Pennsylvania and Michigan also require beneficial use
approvals for CKD. This approval is required if a cement facility plans to reuse CKD for a
particular beneficial purpose, such as to make fertilizer or to use CKD as a soil stabilizer.
State solid waste management requirements for CKD control appear to be in transition
because of a trend toward creation of new state industrial solid waste programs and expansion of
existing programs. Concerns about ground-water protection and the desire to examine the
effects of burning hazardous waste as a fuel supplement also support this trend. Pennsylvania is
an example of a state in which new regulations on residual waste management apply more
stringent requirements to all industrial solid waste, including CKD. Texas has new waste
classification requirements and will be proposing new boiler and industrial furnace regulations in
the near future. California has imposed a moratorium on enforcement of regulations that affect
CKD pending further evaluation. In contrast, Michigan appears to be reviewing CKD
management practices as well as examining with interest other states' programs for examples of
effective innovations.
7.0.4 Limitations of the Analysis
This regulatory analysis must be interpreted with care, as the scope of the state regulatory
review was limited. Time and resource constraints precluded a detailed analysis of all of the
states that contain cement plants. In addition, EPA found that the scope of state programs was
not always clear from the state statutory and regulatory language reviewed. As a result, EPA
contacted state and local officials to interpret legal requirements. State and local regulatory
officials, while helpful, sometimes had differing interpretations of requirements.
The ability to draw conclusions concerning the relative performance of waste
management controls among states is limited by variations in requirements and recordkeeping
among the states. Recordkeeping varies significantly among states; where states have pertinent
records, information on implementation may be readily available. Also, as this study was limited
to four states, the analysis may not be completely representative of CKD management controls
throughout the country.
Most often, because CKD waste is not regulated under Subtitle C of RCRA, states do
not specifically regulate the management of this waste at cement manufacturing facilities.
-------
7-5
Similarly, enforcement and monitoring records are incomplete and/or distributed throughout
regional offices within a state, making the effectiveness of existing CKD management controls
more difficult to evaluate.
7.1 AIR
7.1.1 Federal Controls
Clean Air Act
The federal Clean Air Act requires EPA to establish the maximum ground level
concentrations of pollutants in the ambient air that protect public health and the environment.2
Currently, national ambient air quality standards (NAAQS) exist for sulfur dioxide (SO2),
paniculate matter smaller than 10 microns in size (PM10), carbon monoxide (CO), ozone (O3),
nitrogen oxides (NOJ; and lead (Pb).3 The NAAQS for paniculate matter is the standard
having the most effect in controlling the release of CKD to the atmosphere through kiln stack or
fugitive dust emissions. The standard for paniculate matter was changed in 1987 from one
measuring total suspended paniculate (TSP) to one measuring paniculate matter ten microns in
diameter or smaller (PM,0).4 PM10 emission limitations apply not only to kiln dust, but also to
particulates from grinding and milling processes, coal dust, and quarry dust.
The NAAQS establish ceilings for individual pollutant concentrations and require the
development and implementation of emission limitations pursuant to other sections of the Clean
Air Act. Therefore, NAAQS determine the degree of control that will be imposed on existing
sources and the restrictions on location of new sources, depending on whether air quality is
better or worse than the NAAQS in the particular area where a source is or will be located.
Regulatory agencies enforce the emission limitations to comply with the NAAQS. Various
implementing regulations and air pollution control programs are described below.
Implementing Regulations
State Implementation Plans
The state implementation plan (SIP) under Section 110 of the Clean Air Act5 is the
primary regulatory mechanism by which emission controls are imposed by the states on stationary
sources in order to meet NAAQS. EPA's approval of a state plan makes its provisions
enforceable by the federal government, the state, and by citizen suit. All the states have SIPs,
but the 1990 Amendments to the Clean Air Act require many changes in current SIPs, as
delineated below.
2 42 U.S.C. §§ 7401-7671q.
3 40 CFR Part 50.
4 Fine particles pose a greater hazard to human health as they can pass through the body's natural defenses and
penetrate deep into the lungs.
5 42 U.S.C. § 7410.
-------
7-6
In particular, Section 110(a)(2) of the Clean Air Act has been amended to require that
an acceptable SIP contain detailed provisions addressing the following topics:
• Emission limitations and control measures;
• Monitoring requirements;
• Review of new and modified sources for compliance with new source performance
standards (NSPS), prevention of significant deterioration (PSD), and
nonattainment;
• A demonstration of adequate legal authority to operate and enforce the program;
• Emergency authority similar to that granted EPA under Section 303 of the Clean
Air Act; and
• A permit program.
New Source Performance Standards '
EPA established NSPS for portland cement plants in 40 CFR 60 Subpart F. These
performance standards apply to plants that were constructed or modified after August 17,1971.6
Components of cement plants (referred to as "facilities") specifically affected are kilns, clinker
coolers, raw mill systems, finish mill systems, raw mill dryers, raw material storage facilities,
clinker storage facilities, finished product storage facilities, conveyor transfer points, and bagging
and bulk-loading and unloading systems. For these plants, EPA establishes performance
standards that reflect the emission limitations achievable through application of the best available
pollution control technology. The performance standards consider other environmental (e.g.,
increased water pollution in exchange for reduced air pollution) and energy impacts.7
In accordance with the NSPS, no portland cement plant owner or operator may exceed
the paniculate matter emission limits. Owners or operators must monitor each stack using a
continuous opacity monitoring system or a certified visible emissions observer. In all cases, each
owner or operator must submit semi-annual reports of excess emissions (defined as all 6-minute
periods during which the average opacity exceeds the standard) and of equipment malfunctions.
The emission standards for these facilities are shown in Exhibit 7-1. In addition, owners or
operators must record daily production rates and kiln feed rates and conduct monitoring
activities.
Prevention of Significant Deterioration
The goal of the PSD program is to avoid deterioration of air quality in attainment ("clean
air") areas by maintaining pollutant emissions levels such that ambient air quality remains below
the NAAQS. For example, Oakland County, Michigan, is an attainment area for paniculate
6 A "modification" is any physical or operational change of an existing facility that increases the emission of any air
pollutant.
742U.S.C. §7411.
-------
7-7
matter, SO2, NO,, and lead; thus, the PSD program seeks to limit emission levels of these
pollutants so that ambient concentrations remain within their respective NAAQS.
-------
7-8
Exhibit 7-1
New Source Performance Standards for Portland Cement Plants
Affected Facility
Kiln
Clinker Cooler
All Other Affected Facilities
New Source Performance Standards
Particular Matter
(kg/mt)'
0.15
0.05
-
Opacity
(percent)
20
10
10
* Paniculate matter is measured in terms of kilograms of paniculate matter per metric ton of feed (dry basis) to the kiln.
Section 165 of the Clean Air Act requires a PSD permit prior to construction or
modification of a source in an attainment area. To obtain a PSD permit, a source must
demonstrate that it will use Best Available Control Technology (BACT), to reduce emissions for
each pollutant subject to regulation under the Clean Air Act. Section 169(3) defines BACT as
air pollution controls that achieve the
maximum degree of [emission] reduction ... which the permitting authority, on a case-
by-case basis, taking into account energy, environmental, and economic impacts and other
costs, determines is achievable for such facility ....
BACT limitations must be at least as stringent as those limitations required by applicable NSPS.
The BACT provision gives EPA or the authorized state the ability to tighten emission control
technology requirements by incorporating state-of-the-art control technology developments.
Nonattainment Review
Whereas PSD review applies where a new or modified emission source is to be located in
an attainment area, Nonattainment Review applies where an air quality control area is in
nonattainment of the NAAQS. State implementation plans must require that permits be
obtained for the construction or modification of major stationary sources in a nonattainment
area.
Nonattainment status typically means that more stringent emission limitations will be
necessary for the source than if it were being built in an attainment area. In addition, the
stationary source must obtain an emission offset, which is a reduction in emissions of the
nonattainment pollutant by an existing source (or sources) in the same area. Regulations
establish offsets on the basis of total emission discharges. The offsets must be somewhat greater
than the potential emissions of the new or modified source to produce a net air quality benefit or
"reasonable further progress" toward compliance with the NAAQS.
Hazardous Air Pollutants
The 1990 Clean Air Act Amendments completely revised Section 112 of the Clean Air
Act that had provided for national emission standards for hazardous air pollutants (NESHAPs).
The revised Section 112(b)(l) establishes a program to regulate emissions of 189 toxic air
-------
7-9
pollutants through technology-based standards.8 EPA will not establish control requirements
directly on a substance-by-substance basis. Instead, EPA identified categories of industrial
facilities, including Portland Cement plants, that emit substantial quantities of each air toxic.9
NESHAPs apply to facilities that emit or have the potential to emit 10 tons or more of
any single hazardous air pollutant in a year. Alternatively, a facility that emits or has the
potential to emit more than 25 tons per year of any combination of hazardous air pollutants
would also be subject to NESHAPs. In setting standards, EPA is allowed to distinguish between
new and existing facilities, and to set less stringent technology-based standards for existing
facilities. The standards can compel a wide range of control measures, including not only the
installation of control equipment but also process changes or the substitution of materials.
Within eight years of establishing a NESHAPs for a source category, EPA must provide for a
second phase of regulatory controls aimed at protecting public health with "an ample margin of
safety."10 If necessary, additional health-based standards will be required.
EPA is currently developing NESHAPs for Portland Cement plants that will address stack
emissions and fugitive emissions. Hazardous air pollutant emissions from CKD storage piles will
be considered for regulation. It is uncertain how, or if, CKD storage piles will be regulated. The
NESHAPs for Portland Cement plants are scheduled to be promulgated no later than November
1997.
Boiler and Industrial Furnace Regulations
On February 21,1991, EPA finalized regulations that expanded controls on the burning
of hazardous waste in boilers and industrial furnaces.11 The boiler and industrial furnace (BIF)
regulations require owners and operators of boilers and industrial furnaces burning hazardous
waste to limit the emissions of toxic metals, carbon monoxide, hydrogen chloride, chlorine gas,
and particulate matter.12 Cement kilns that burn hazardous waste are subject to the regulation
because they are defined as industrial furnaces." Prior to the BIF rule, cement kilns burning
hazardous waste for energy recovery in urban areas could do so only if they complied with the
emission standards applicable to hazardous waste incinerators.14 These urban cement kilns,
along with all other cement kilns that burn hazardous waste, are now subject only to the BIF rule
requirements.
* 42 U.S.C. § 7412(b)(l).
9 42 U.S.C. § 7412(d).
10 42 U.S.C. §7412(0(2).
11 56 Fed. Reg. 7134.
12 While the BIF rules are promulgated under the authority of the Resource Conservation and Recovery Act, 42
U.S.C. §§ 6901 to 6992K, for the purpose of this report, the discussion concerning the BIF rule emission limits is
included under the section on air pollution controls. For further discussion of the impacts of the BIF rule on solid
waste management, see Section 7.4.1.
13 40 CFR § 260.10.
14 42 U.S.C. § 6924(q)(2)(C). See also 40 CFR § 266.3 l(c).
-------
7-10
Burning hazardous waste that contains toxic organic compounds under poor combustion
conditions can result in emissions of organic compounds. EPA regulates the emissions of
organics as follows:
• A 99.99 percent destruction and removal efficiency (DRE) standard for organic
hazardous constituents in waste feeds and a 99.9999 percent DRE for dioxin-listed
hazardous waste;
• Limits on flue gas concentrations of carbon monoxide (CO) and hydrocarbons to
control products of incomplete combustion; and
• Special controls for chlorinated dibenzodioxins and dibenzofurans (CDD/CDF)
for BIFs burning hazardous wastes under specific circumstances.
The rules also establish emission limits for toxic metals based on site-specific testing and
analyses (e.g., emissions testing, dispersion modeling).15 Emissions of hydrogen chloride and
free chlorine are regulated under the same general approach. EPA established a paniculate
matter emission limit to control emissions of toxic metals; metals and organic compounds may
adsorb onto particulate matter in the flue gas. Under the terms of an October 22, 1993
settlement, EPA will initiate a new rulemaking on whether to revise provisions of existing BIF
regulations that establish standards for cement kilns by September 20, 1995.16
In an effort to protect the public from health risks associated with burning hazardous
wastes, EPA announced on September 28, 1993 enforcement actions against 11 cement kilns for
violations of the BIF rules. These enforcement actions are seeking over $13 million in penalties
from owners and operators for violations that range from failure to comply with emission
standards and inadequate monitoring of hazardous waste fuel feeds to failure to maintain proper
records.
The BIF rule directly affects the regulatory status of CKD generated by cement kilns
burning hazardous waste as fuel. These effects are discussed in detail in the Solid Waste
Management Federal Controls section (7.4.1).
7.1.2 State Controls
At the state level, California, Michigan, Pennsylvania, and Texas have established air
quality requirements for CKD management that incorporate the federal standards and also
subject cement plants to more stringent visible emission or opacity limitations. A summary of
the four states' air pollution controls can be found in Exhibit 7-2 and a discussion of the
individual state's air quality requirements follows.
15 Owners and operators must analyze the hazardous waste to be burned and comply with the standards for each
of the 10 metals (antimony, arsenic, barium, beryllium, cadmium, hexavalent chromium, lead, mercury, nickel,
selenium, silver,'and thallium) that could reasonably be expected to be in the waste.
16 Horsehead Resource Development Co. Inc. v."EPA, No. 91-1221 (D.C. Cir. Oct. 22, 1993). EPA also agreed to
describe the option of adopting technology-based emission standards for cement kilns. Final rulemaking would be
required by December 15, 1996.
-------
7-11
Exhibit 7-2
Summary of State Air Pollution Controls
States
Permits
N
General
Requirements
Paniculate Matter
Emission Controls
Fugitive Dust
Emission Controls
California
Permit to construct
and to operate; some
air pollution control
districts require
permitting of CKD
storage facilities
Compliance with
Federal NSPS, and
PSD and
Nonatlainment Review
Paniculate matter
emission limits; opacity
limits
Air pollution control
district may require
fugitive dust control
program
Michigan
Permit to construct
and to operate
Compliance with
Federal NSPS, and
PSD and
Nonattainment Review
Paniculate matter
emission limits; opacity
limits
State may require
fugitive dust control
program
Pennsylvania
Permit to construct
and to operate; permit
for fugitive emissions
Compliance with
Federal NSPS, and
PSD and
Nonattainment Review
Paniculate matter
emission limits; opacity
limits
Fugitive emission
controls as permit
condition
Texas
Permit to construct
and to operate
Compliance with
Federal NSPS, and
PSD and
Nonattainment Review
Paniculate matter
emission limits; opacity
limits
Opacity limits
California
The California Air Resources Board (ARE), local or regional air pollution control
districts, and air quality management districts ("air districts") are the state agencies primarily
responsible for controlling air pollution.17 The ARE has responsibility to set air standards,
measure local compliance, assist air districts in the preparation of plans to attain the standards,
and review those plans and their implementation.
Air districts have the primary responsibility for enforcement of state and air district
regulations. Enforcement options available to the air districts include notices of violation,
abatement orders, administrative penalties, civil and criminal penalties, and permit
revocations.18 The ARE reviews district enforcement practices and is authorized to exercise
district enforcement authority if it finds that a district's actions are inadequate.19
The air districts closely monitor compliance with fugitive dust emission limits for cement
plants. The North Central Coast Air Basin inspectors review cement plant operations and
records monthly.20 Other air districts usually inspect cement plants annually unless there has
17 Cal. Health and Safety Code §§ 39000^4384.
w Cal. Health and Safety Code §§ 42400-42402.
19 Cal. Health and Safety Code §§ 41502-41507.
20 Personal communication with Greg Chee, Air Quality Engineer, North Central Coast Air Basin, March 3, 1993.
-------
7-12
been a complaint or reason to believe a cement plant is in violation.21 Inspectors review the
measures that facilities use to control fugitive dust emissions from CKD such as storage tanks or
storage buildings.
Ambient Air Quality Standards
The state ambient air quality standards adopted by the ARE are in addition to the
NAAQS adopted by EPA under the Clean Air Act. In adopting state ambient air quality
standards, the ARE is required to consider "the public health, safety, and welfare, including, but
not limited to health, illness, irritation to the senses, aesthetic value, interference with visibility
and effects on the economy." According to a December 1988 ARE report, 23 of California's 41
air districts were in federally-designated nonattainment areas.22 Therefore, most regulatory
attention focuses on developing rules necessary to attain the federal standards.
Currently, California state ambient air quality standards exist for the following pollutants:
ozone (O3);
carbon monoxide (CO);
nitrogen dioxide (NO2);
sulfur dioxide (SO2);
paniculate matter (PM,0);
sulfates (SO4);
paniculate lead (Pb);
hydrogen sulfide (H2S); and
visibility-reducing particles.23
Authority To Construct
An owner or operator of a cement
plant proposing to construct or modify a
stationary source in California that may emit
pollutants into the atmosphere must first
obtain an Authority to Construct from the
county or regional air pollution control
district or air quality management district
where the source is or will be located. The
air districts must ensure that emissions from
such sources will not interfere with the
attainment and maintenance of state ambient
air quality standards.
Generally, an air district requests the following
information before granting an Authority to
Construct:
• Description of the business, including the
materials used and the particle sizes of all
bulk solids involved;
• Type of air pollution control equipment and
its anticipated degree of efficiency;
• Types of fuel; and
• Operating schedule for emission sources.
21 Personal communication with Tom Krinke, Air Quality Specialist, Compliance Section, San Bernadino Air
Quality Control District, July 28, 1992.
22 Air Resources Board, Stationary Source Division, Technical Review Group Emissions Credit Systems and New
Source Review Programs: A Report to the Legislature (Dec. 8, 1988).
17 CCR § 70200.
-------
7-13
Permit To Operate
Cement plants must also obtain a Permit to Operate from the air district for the area in
which the facility is located. In general, the air district asks the applicant to verify that
construction or modification of a facility was completed in accordance with the Authority to
Construct and that the facility will meet the district's regulations.
- Prevention of Significant Deterioration and New Source Review
Unless the local air district has been delegated the authority by EPA to issue these
permits, a new or modified project may also require a PSD permit from EPA. A PSD permit is
only required if the facility is located in an attainment area for a pollutant.
In nonattainment areas, air districts have adopted New Source Review Rules, which
regulate all new or modified sources with emissions exceeding a specified limit for any pollutant
for which a state or national ambient air quality standard exists. For example, standards exist for
sulfur dioxide and paniculate matter. The average "trigger" level is 250 pounds per day. If this
level is exceeded by the projected operation of the new source or modification of an existing
facility, New Source Review requirements become applicable.
Approval for the new source or modification will be granted only if the owner or
operator provides "offsets" for all net emission increases.24 The owner or operator must reduce
emissions within the source at a 1 to 1 ratio, or reduce other sources in the nonattainment area
at a ratio of at least 1.2 to 1. For example, an applicant proposing a new or modified source
producing 1,000 pounds of pollutants per day must eliminate a minimum of 1,000 pounds of
pollutants per day from his/her existing source or 1,200 pounds of pollutants per day from other
existing sources. Offsets must be located upwind in the same or adjoining counties, or within 15
miles downwind of the proposed new or modified source.
In addition to complying with applicable New Source Review and/or PSD requirements,
owners and operators of cement plants must comply with the federal NSPS.
Cement plants are also subject to visible emission limitations in accordance with Section
41701 of the California Health and Safety Code. This provision prohibits any air contaminant
discharge to the atmosphere that continues for an aggregate period of more than three minutes
in any one hour in which such emission is either:
• As dark or darker than "No. 2 on the Ringelmann Chart";25 or
• Of such opacity as to obscure an observer's view in the same degree as smoke
equalling Ringelmann No. 2.26
14 Emission Offset Interpretive Ruling, 40 CFR Part 51, Appendix S.
25 A Ringelmann Chart grades the shade or opacity of visible air contaminant emissions. The chart is a color
coded strip that an observer holds up and compares to the shade of color of the air emissions discharging from a
smokestack. Darker colors and higher numbers correspond with greater emissions. Ringelmann's Scale for Grading
the Density of Smoke as published in United States Bureau of Mines Information Circular 8333.
M Cal. Health and Safety Code §§ 41701, 41704.
-------
7-14
Some Air Districts, such as the South Coast Air Quality Management District, set more stringent
opacity limits (emissions no darker than No. 1 on Ringelmann Chart) and these have been
applied to cement plants.27
These visible emission limitations also regulate fugitive dust emissions from CKD storage.
In addition, some air pollution control districts require permitting of CKD storage facilities and
fugitive dust control programs. For example, air pollution control districts have required cement
plants to control fugitive emissions by managing CKD in storage tanks or in warehouses before
the CKD is insufflated (recycled to the feed stream) or landfilled.
Michigan
Ambient Air Quality and Permits
The Air Pollution Control Commission within the Michigan Department of Natural
Resources (DNR) establishes state ambient air quality standards and requires permits for
construction and operation of a cement plant.28 Permits, for the most part, are issued by the
regional DNR offices.2' Permit requirements subject cement plants to the same general
requirements as in California (e.g., compliance with the U.S. EPA's New Source Performance
Standards for cement plants, and PSD and Nonattainment Review).30
As in California, inspections and enforcement occur primarily at the district (Michigan
DNR) level. The Michigan DNR Air Quality Division has approximately 35-40 inspectors,
including district supervisors, to review all stationary sources of air pollution.31 For cement
plants, inspections are performed annually. The district offices may inspect more frequently in
response to complaints or if a facility has a history of compliance problems. The Wayne County
Health Department, Air Pollution Control Division,, monitors the cement facility located in
Wayne County.32
On the basis of submitted monitoring reports or plant inspections, the DNR district office
will send a notice of violation for any noncompliance with federal or state requirements or
permit conditions. This notice of violation requires initiation of corrective actions and may
establish a deadline for compliance. If a voluntary agreement is not entered into with DNR,
DNR may initiate further enforcement actions such as administrative orders, injunctions, and
civil and criminal penalties.
27 SCAQMD Rule 401.
M Mich. Comp. Laws § 336.13.
29 The Air Pollution Control Commission issues a permit only when a facility's emissions significantly affect a PSD
or nonattainment area.
30 Mich. Admin. Code §§ 336.1203 and 336.1208.
51 Personal communication with Barb Rosenbaum, Michigan Department of Natural Resources, Air Quality
Division, September 10, 1993.
32 Personal communication with Bob Zabick, Wayne County Health Department, Air Pollution Control Division,
Enforcement, October 13, 1992. Currently, the cement plant located in Wayne County only grinds clinker delivered
from Canadian cement plants. The kiln which was permitted to burn hazardous waste for fuel was shut down for
economic reasons.
-------
7-15
Particulate Matter and Visible Emissions Limitations
Michigan has established paniculate matter emission limits for cement manufacturers that
are equivalent to the federal new source performance standards, but the state may impose more
stringent emission limits. The Air Pollution Control Commission usually applies the opacity limit
in the federal new source performance standards for cement plants (that range from 20 percent
for kilns to 10 percent for clinker coolers). The state regulations require a cement plant not to
exceed 27 percent opacity for more than one six-minute period per hour." The Michigan Air
Pollution Control Commission is also authorized to require a more stringent standard on a case-
by-case basis.
In addition, the Air Pollution Control Commission may request that a cement plant's
operator submit a fugitive dust control program.34 This requirement applies to any facility that
"processes" bulk materials, including raw materials for cement manufacture or clinker on its way
to a grinding unit. These requirements are triggered only by a notification from the Commission.
The Commission has required fugitive dust control programs that include the pneumatic
conveyance of CKD, the use of paniculate matter collection devices during transfer operations,
the application of dust suppression liquids to haul roads (twice each month), and weekly
sweeping of paved haul roads. In addition, at one cement facility, the Commission required that
CKD be pneumatically pumped to the floor of a quarry rather than simply dumped from the top
of the quarry. More recently, the Commission required a facility to mix its CKD with water to
form pellets as a means of reducing fugitive emissions.
Pennsylvania
Under the Pennsylvania Clean Air Act, the Department of Environmental Resources
(DER) establishes ambient air quality standards for the Commonwealth.35
In a manner similar to that used in California and Michigan, permitting and enforcement
of air programs generally are handled by the six regional DER offices. However, Allegheny and
Philadelphia Counties have autonomous air pollution control programs that have been approved
by DER. As in the other states, permits to construct and operate are required.
DER conducts annual inspections of cement plants and reviews monitoring reports
submitted by facilities. Generally, DER enforcement procedures parallel the enforcement
procedures employed in California and Michigan. DER is concerned with fugitive dust emissions
from CKD storage and other plant operations.36 Inspectors are placing some emphasis on
reviewing fugitive dust control programs and discouraging the use of open storage of CKD.
Currently, all the operating cement plants recycle some of the CKD back into the kiln. DER is
encouraging this trend to reuse CKD.
53 Mich. Admin. Code § 336.1301 (l)(a).
34 Mich. Admin. Code §§ 336.1371-72.
55 35 Pa. Stat. § 4004.
36 Personal communication with Gary Linns, Pennsylvania Department of Environmental Resources, Air Quality
Control Division, October 9, 1992.
-------
7-16
Nonattainment Review
In Pennsylvania, special permit requirements exist for a cement plant locating in or
significantly affecting nonattainment areas." Pennsylvania's nonattainment review is slightly
different than the reviews required by California and Michigan. These special permit
requirements only apply to cement plants constructed or modified after June 30,1979. In
addition, a cement plant must discharge greater than 50 tons per year of emissions, 1,000 pounds
per day, or 100 pounds per hour, whichever is more restrictive. Finally, facilities must also be
significantly affecting a nonattainment area. To be considered significantly affecting a
nonattainment area, a facility's discharge must exceed established emission limits. For example,
the significance levels for ambient total suspended paniculate and sulfur dioxide are one
microgram per cubic meter annually, or five micrograms per cubic meter in a 24-hour period. In
determining whether a source exceeds the emission rates or significance levels, all emissions
resulting from the operation must be considered, including flue (e.g., stack) and fugitive (e.g.,
material transfer, storage piles, and roads on the plant property) emissions. To be permitted, a
facility that exceeds the significance levels must offset its emissions by reducing emissions from
its own facility or another facility located in the nonattainment area or from a facility affecting
the nonattainment area. To improve air quality in the nonattainment area, the ratio of
paniculate matter or sulfur dioxide emission reductions required for any new emissions must be
equal to or greater than 1.3 to 1 for flue emissions and 5 to 1 for fugitive emissions.
Paniculate Matter and Visible Emissions Limitations
Pennsylvania has additional criteria for paniculate matter that apply independently of
PSD and nonattainment review. No source may cause the emission of visible air contaminants
that exceed either of two opacity limits:
• 20 percent for a three-minute period in any hour; or
• 60 percent at any time.38
Visible emissions may be measured by either (1) any device approved by DER to provide
accurate opacity measurements or (2) by a trained observer.
The Pennsylvania Air Pollution Control regulations also establish paniculate matter
emission limits for cement plants.39 Cement plants may not emit paniculate matter, at any
time, in excess of either a rate calculated by a formula (variables exist for clinker production and
clinker cooling) or 0.02 grains per dry standard cubic foot in the effluent gas.
As in Michigan, the emission into the atmosphere of fugitive air contaminants from a
source is prohibited unless permitted by the state.40 DER may require an owner or operator of
a source to provide a description of proposed control measures, characteristics of emissions,
quantity of emissions, and ambient air quality analysis showing the impact of the source on
37 25 Pa. Code §§ 127.61-127.73.
38 25 Pa. Code § 123.41.
"25 Pa. Code § 123.13.
40 25 Pa. Code § 123.1.
-------
7-17
Permits may only be issued if the fugitive emissions,
after being treated by an appropriate air pollution
control, meet the following requirements:
• Emissions are of minor significance with
respect to causing air pollution; and
• Emissions are not preventing or interfering
with the attainment or maintenance of any
ambient air quality standard.
ambient air quality. At least two cement
plants in Pennsylvania have been required to
store CKD in a warehouse or silo to control
fugitive emissions.
Texas
The Texas Clean Air Act authorizes
the Texas Air Control Board to set standards
and emission limits for air pollution.41 As in
the other states, the 11 cement plants in
Texas must have permits for construction and
operation and must comply with applicable air quality standards.
On September 1,1993, the Texas Air Control Board was abolished and all powers and
duties were transferred to the Texas Water Commission. This transfer of authority completed
the process of consolidating all environmental protection programs into one agency. At this
time, the agency became the Texas Natural Resources Conservation Commission.
As in the other three states, primary responsibility for compliance monitoring occurs in
the regional offices. Personnel from 12 regional offices conduct annual inspections of cement
plants.42 The regional offices also respond to citizen's complaints, review upset reports (i.e., if a
facility violates its emission limitations, the facility must report this violation to the Commission),
perform investigations, and if necessary, recommend enforcement actions. The primary
enforcement mechanism is the notice of violation. The notice of violation provides a facility
operator the opportunity to correct any problems within 30 days. Facilities usually come into
compliance within this time period. The Commission will take no further action if the violation
is not continuing or not a repeating problem. If a facility fails to conduct remedial activities,
however, additional enforcement activities may be initiated.
Other Requirements Applicable to Cement Plants
Because the Air Quality Program of the Texas Natural Resources Conservation
Commission requires the control of air pollution from visible emissions and paniculate matter,
cement plants are subject to visible emission requirements that vary depending upon age and
exhaust gas flow rate.43 Currently, the opacity limits for existing cement plants in Texas range
from 10 percent (the facility was subject to more stringent PSD requirements) to 30 percent.
Visible emissions must not exceed the following opacities:
(1) 30 percent average over a six-minute period for any source on which construction
was begun on or before January 31,1972;
(2) 20 percent average over a six-minute period for any source on which construction
was begun after January 31, 1972; or
41 Tex. Health and Safety Code § 1.05.
42 Persona] communication with Richard Lee, Natural Resources Conservation Commission, Air Quality Program,
Compliance Division, August 25, 1993.
1 Tex. Admin. Code tit. 31, § 111.111.
-------
7-18
(3) 15 percent average over a six-minute period for any source having a flow rate
greater than or equal to 100,000 cubic feet per minute unless a continuous opacity
monitoring system is installed.
Categories (1) and (2) apply to facilities that utilize a continuous opacity monitoring system but
otherwise would be subject to the third category. Fugitive dust emissions from CKD piles or
roads on which CKD is transported must not exceed an opacity of 30 percent over a six-minute
period.
Continuous opacity monitors and annual inspections assist in keeping a facility in
compliance. Currently, all 11 portland cement manufacturing plants in Texas have continuous
opacity monitors. Facilities submit quarterly monitoring reports and notify the Air Quality
Program of any exceedance of the permit requirements. Monitoring and report records must be
maintained on site for two years. Inspectors review these on-site records as part of the annual
facility inspection.
In addition, ground level paniculate matter concentration limits exist for all sources,
including cement plants.44 Emissions of paniculate matter from a source may not exceed either
of the following net ground level concentrations:
• 200 micrograms per cubic meter of air sampled, averaged over any three
consecutive hours; or
• 400 micrograms per cubic meter of air sampled, averaged over any one hour
period.
The owner or operator of a cement plant also must notify the Air Quality Program of any
major upset condition that causes or may cause an excessive emission.45 A "major upset" is
defined as "[a]n unscheduled occurrence or excursion of a process or operation that results in an
emission of air contaminants that contravenes the Texas Clean Air Act and is beyond immediate
control .. .Il4
-------
7-19
Special Report on Texas Cement Plants' Excess Emissions
From November 1990 to May 1991, the Air Quality Program conducted a special study of
all 11 portland cement manufacturing plants in Texas to gain a better understanding of
emissions.47 The study analyzed whether the kilns and clinker coolers were in continuous
compliance with paniculate matter and stack opacity limits, such as those found in the federal
NSPS (40 CFR Part 60, Subpart F). The Program conducted the cement plant study to gain a
better-understanding of emissions, especially as more facilities are using hazardous waste as a
fuel.
The Air Quality Program concluded that quantities and causes of excess emissions varied
greatly between companies. The Program reviewed the causes of excess emissions, their levels
and duration, and what actions were taken to prevent a recurrence in the future. In response to
the study, the Air Quality Program developed guidance for companies to use in reporting excess
emissions. The Air Quality Program also provided resources for the Compliance Division to hire
an additional staff person with the responsibility to monitor maintenance and upsets at
facilities.48
7.2 WATER
7.2.1 Federal Controls
The Clean Water Act
The basic framework for federal water pollution control is the Federal Water Pollution
Control Act of 1972, also known as the Clean Water Act (CWA).49 The CWA establishes
national goals to eliminate the discharge of pollutants into navigable waters. The principle
means to achieve these goals is to impose effluent limitations on, or otherwise to prevent,
discharges of pollutants into any waters of the United States.
Under the Clean Water Act, the discharge of a pollutant from a point source into any
waters of the United States, except as authorized by a permit, is illegal. Accordingly, any cement
production facility seeking to discharge wastewater effluent and/or a point source discharge of
stormwater to surface waters must apply for and obtain an NPDES permit. A cement facility has
typically two types of discharges: process wastewater and stormwater run-off.
To impose limitations on pollutant discharges, the CWA established a nationwide NPDES
permit program (see 40 CFR 122). An NPDES permit establishes specific "effluent limitations"
and conditions regarding any discharges to surface waters. For the cement industry, these
effluent limitations include requirements for total suspended solids, temperature, and pH. A
permit writer may impose additional limits on toxics if an adequate basis exists using a Best
Professional Judgment (BPJ) determination. Monitoring and reporting requirements assure
47 Cement Plant Resources Group, draft Final Report on Cement Plant Excess Emissions (July 31, 1991).
** Personal communication with Richard Lee, Natural Resources Conservation Commission, Air Quality Program,
Compliance Division, August 25, 1993. This person's responsibilities include not only monitoring maintenance and
upsets at cement plants but also at facilities that emit vinyl chloride.
4'33U.S.C. §§ 1251-1387.
-------
7-20
For regulatory purposes, process wastewater has
been divided into two categories. The first category
of process wastewater includes the following
discharges:
• water that comes into contact with kiln dust
as an integral part of the manufacturing of
cement; or
« water that is used in wet scrubbers to
control kiln stack emissions.
The second category of process wastewater includes:
• water that does not come into contact with
kiln dust as an integral part of the
manufacturing process; and
• water that is not used in wet scrubbers to
control kiln stack emissions.
The other discharge of concern is storrnwater run-off
from materials storage piles and exposed surfaces at
a facility.
compliance with the applicable effluent
guideline limitations (40 CFR 411), water
quality standards, and pretreatment standards.
Under Section 402(b) of the CWA,
responsibility for administration of the
NPDES program can be approved for
individual states. To obtain program
approval, a state must have a statutory
program for regulating discharges to surface
waters. EPA has approved state programs for
implementing the NPDES permit system for
three of the states analyzed in this report:
California, Michigan, and Pennsylvania. In
these states, only one state water permit
regulating the discharge of pollutants is
required. In Texas, which has not received
authority, both federal and state permits must
be obtained.
For those cement facilities that
discharge to publicly owned treatment works
(POTWs) and not directly to surface waters, different but comparable treatment standards exist.
These indirect discharges are regulated by pretreatment standards. Pretreatment standards
protect the operation of POTWs (e.g., prohibit the introduction of pollutants that create fire or
explosion hazards) and prevent the discharge of pollutants that might pass through POTWs
without receiving adequate treatment. Cement facilities are subject to the general pretreatment
standards in 40 CFR Part 403 as modified by or in addition to the effluent guideline limitations
in 40 CFR Part 411 discussed above. Pretreatment requirements are directly enforceable by
EPA and states with NPDES permitting authority.
The CWA also requires that states establish water quality standards for all surface waters.
The standards are subject to EPA approval. States are allowed to set more stringent water
quality standards than those derived by EPA in water quality criteria documents. In establishing
NPDES permit requirements, the effluent limitations discussed above (that established a
technology-based minimum treatment standard) may be superseded by more stringent effluent
limitations necessary to maintain water quality in specific water bodies. Therefore, the stringency
of particular water quality standards, established to protect designated uses for sections of a
water body, can significantly affect the final effluent guideline limitations specified in a facility's
NPDES permit.
In addition to controls on process wastewater from cement production facilities,
storrnwater run-off is also regulated. Generally, storrnwater discharges from cement facilities
contain CKD from materials storage piles and surficial areas at a facility. On November 16,
1990, EPA adopted a rule setting forth NPDES permit application requirements for storrnwater
discharges associated with industrial activity.50 Storrnwater run-off from cement facilities is
50 40 CFR § 122.26.
-------
7-21
considered a discharge associated with industrial activity.51 Facilities located in Michigan may
apply for coverage under a stormwater permit through an individual application. In addition to
an individual application for a permit, facilities in Texas, California, and Pennsylvania may also
obtain coverage under a general permit.52
Under the stormwater permit application regulations, pollution prevention plans and best
management practices will be required to reduce pollutants in stormwater discharges. Prior to
1990, cement plants already had to meet the effluent limitations for run-off from materials
storage piles. Numeric limitations existed for discharges of total suspended solids (TSS) and pH
levels. Facilities meeting these limitations are deemed to be in compliance with the new
stormwater requirements for the remainder of an existing NPDES permit.53 If an existing
permit, which covers discharges of stormwater, expires, the facility is required to obtain separate
permits for both their stormwater discharge and any process wastewater discharge.
Safe Drinking Water Act
The Safe Drinking Water Act (SDWA) has several provisions that can be significant to
the operation of cement plants, including requirements for setting drinking water regulations and
maximum contaminant levels (MCLs) for toxic water contaminants, as well as wellhead
protection area programs. MCLs are "the maximum permissible level of a contaminant in water
which is delivered to any user of a public water system."54 EPA is responsible for establishing
MCLs for pollutants as part of the primary drinking water regulations.
The importance of MCLs can be found in determining the level of cleanup required at
Superfund sites containing CKD (for further discussion of Superfund and CKD, see section
7.4.1). Two cement production facilities described in the Damage Case Study (see Chapter 6)
are currently listed on the National Priorities List. These facilities are:
Holnam Incorporated, Mason City, Iowa; and
Portland Cement Company, Salt Lake City, Utah.
For these facilities, MCLs constitute one of the primary classes of applicable and relevant or
appropriate requirements (ARARs) when any hazardous substance, pollutant, or contaminant
will remain on a Superfund site. MCLs have been established for many of the compounds found
in CKD, including arsenic (As), barium (Ba), cadmium (Cd), chromium (Cr), mercury (Hg),
selenium (Se), and silver (Ag).
The SDWA also requires EPA to establish national secondary drinking water regulations,
standards that reflect welfare concerns such as odor, taste, and color. While these less stringent
51 40 CFR § 122.26(b)(14)(ii).
52 EPA issued NPDES general permits for stormwater discharges associated with industrial activity on September
6, 1992. 57 Fed. Reg. 41236. Facilities in Texas are eligible for general permits because EPA maintains NPDES
authority for Texas while California and Pennsylvania have been approved to issue general permits as part of their
NPDES programs and have chosen to issue general permits. Michigan does not yet issue general permits.
51 40 CFR § 122.26(e)(6).
34 42 U.S.C. §§ 300f-30Qj.
-------
7-22
standards protect public welfare other than human health, their violation can be used to justify
the abandonment of a water source or treatment to remedy the problem. The national
secondary drinking water standards most applicable to CKD are those for pH and total dissolved
solids.
In addition, the wellhead protection area program encourages states to develop systematic
and comprehensive programs within their jurisdictions to protect public water supply wells and
wellfields. A wellhead protection area is defined as "the surface and subsurface area surrounding
a water well or wellfield, supplying a public water system, through which contaminants are
reasonably likely to move toward and reach such well or wellfield."55 Based on the 1991 PCA
Survey, 25 of 91 operating cement kiln facility respondents indicated that they are located within
one mile of a public drinking water well.
7.2.2 State Controls
California, Michigan, and Pennsylvania have been delegated responsibility for
implementation of the NPDES program to regulate discharges to surface water; in Texas,
however, where delegation has not occurred, Texas has its own water program but EPA
continues to manage the NPDES program. In California, Michigan, and Pennsylvania, only one
state water permit regulating the discharge of pollutants is required. In Texas, where delegation
has not occurred, both federal and state permits must be obtained. A summary of the four
states' water pollution controls can be found in Exhibit 7-3, below, and a discussion of individual
state's water quality requirements follows.
Exhibit 7-3
Summary of State Water Pollution Controls
States
EPA-Approved
Permit Program
Water Quality
Standards
California
Yes; adopted all
federal NPDES
requirements
Existing and
anticipated beneficial
uses; regional water
boards adopt numeric
water quality standards
for specific water body
segments
Michigan
Yes; adopted all
federal NPDES
requirements
All waters protected
for agricultural use,
public water supply,
and recreation;
numeric water quality
standards for sped Tic
water body segments
Pennsylvania
Yes; adopted all
federal NPDES
requirements
Designated water uses;
numeric water quality
standards for specific
water body segments
Texas
No; separate State
permit and a
NPDES permit from
EPA Region VI
Three classifications:
recreation, domestic
water supply, and
aquatic life; numeric
water quality
standards for
sped fie water body
segments
55 42 U.S.C. § 300h-7(e).
-------
7-23
Exhibit 7-3 (continued)
Summary of State Water Pollution Controls
Slates
Stormwater
Requirements
Ground-Water
Protection Policy
California
General Stormwater
permit applicable to
all Stormwater
dischargers (not by
industrial category)
Non-degradation;
classification system
based on beneficial
uses
Michigan
Individual Stormwater
permits (no general
permitting authority)
Non-degradation; no
degradation above
local background
levels in all current
and potential
drinking water
sources
Pennsylvania
Individual permit for
run-off from material
storage piles; general
permit for all other
Stormwater discharges
Non-degradation; use
of best demonstrated
control technology and
best management
practices to protect
ground-water
resources
Texas
EPA administers the
Stormwater program
but must incorporate
state hazardous
metal effluent
limitations to comply
with Texas water
quality standards
Non-degradation;
four classes of
ground water; EPA-
approved wellhead
protection program
California
The California Porter-Cologne Water Quality Act establishes the Water Resources
Control Board and nine regional water resources control boards within the California
Environmental Protection Agency. The state board and the regional boards are authorized to
perform the following activities:
• Adopt water quality plans;
• Regulate discharges to surface and ground water; and
• Require cleanup of discharges of hazardous materials and other pollutants.
Responsibility for water quality planning is shared by the state board and the nine regional
boards.
Holders of state-issued NPDES permits must provide monthly discharge monitoring
reports. The monitoring data are input into a computerized data base (EPA's Permit
Compliance System (PCS)).56 The monitoring data are compared to the effluent limitations
included in the permit, and cement plants that exceed permit limits may be subject to possible
enforcement action.
The state and regional boards may inspect a facility as necessary to ensure compliance
with water quality requirements. The state board does not consider cement plants major
dischargers (as compared to some POTWs that may discharge over a million gallons of water per
day) and therefore, these facilities are subject to less frequent inspections. However, the state
board must inspect each facility at least once per year. The regional boards will consider more
56 PCS is a computerized management information system which contains data on the NPDES permit-holding
facilities. PCS tracks permit status, permit limits, discharge monitoring reports, violations, and enforcement activities.
-------
7-24
frequent inspections if a cement plant has received a complaint, exceeded permit limits, or if the
facility has a past record of permit violations.
The state water resources control board has broad enforcement authority. The state
board may designate and authorize regional water quality control boards to exercise enforcement
authority. The state board and designated regional boards have administrative and civil penalty
authority.57 A regional board can issue a cease and desist order to a discharger who is violating
a discharge requirement or prohibition if a likelihood exists that the violation will continue in the
future. The board may direct the discharger to comply immediately or in accordance with a time
schedule set by the board to remedy or prevent future violations. A regional board may, through
the Office of the Attorney General, seek a Superior Court injunction to prohibit an actual or
threatened waste discharge if it constitutes an emergency. This applies if the discharge or
threatened discharge causes or will cause a condition of pollution or nuisance constituting an
emergency that requires immediate action to protect public health, safety, or welfare.58
The Porter-Cologne Act contains a number of civil and criminal penalty provisions.59
Civil penalties may be imposed on owners or operators that negligently or intentionally violate a
cease and desist order. Persons who, even unintentionally, cause or permit the discharge of a
hazardous substance (under California law, CKD is considered a hazardous substance, see
Section 7.4.2 for further discussion) that causes pollution may be strictly liable for civil penalties
up to $25,000 per day. Criminal penalties ranging up to one year's imprisonment may be
•imposed for an owner's or operator's failure to report an unintentional discharge of hazardous
substances or for the falsification of required reports.
Process Wastewatcr Requirements
California has a federally-approved NPDES permit pretreatment management program
and the authority to issue general permits. The regional water quality boards implement the
NPDES permit program, subject to EPA review. California's water quality regulations adopt by
reference all applicable federal NPDES and pretreatment regulations and, therefore, cement kiln
facilities are subject to these requirements.
Water Quality Standards
The California Water Resources Control Board has adopted state-wide water quality
principles and guidance that the regional boards may make more stringent. In addition, each
regional board has adopted water quality standards for specific water body segments, and these
water quality standards are included in the region's water quality control plan. State water
quality policy requires long-range resource planning, including ground-water and surface water
management programs, and control and use of reclaimed water. Wastewater discharges must be
treated to protect existing and anticipated beneficial uses of such water. California's ground-
water protection policy includes closely regulating a number of potential sources of ground-water
degradation, such as waste management facilities.
57 Cal. Water Code §§ 13300-13306 and §§ 13261, 13385, 13387.
" Cal. Water Code § 13340.
59 Cal. Water Code §§ 13350-13371.
-------
7-25
Stormwater Management Requirements
In November 1991, the California Water Resources Control Board adopted a General
Industrial Stormwater Permit to comply with the federal requirements for stormwater discharges.
In so doing, California decided against issuing general permits to specific industrial categories,
and instead established a general permit that applies to all industrial stormwater dischargers.
Cement kiln facilities are subject to both the stormwater requirements of this general permit and
to the-federal effluent limitations for materials storage pile run-off. Cement kiln facilities will be
required to develop pollution prevention plans, implement best management practices to control
stormwater discharges, and establish monitoring programs.
Michigan
The Michigan Water Resources Commission Act authorizes the Water Resources
Commission to issue permits that regulate the discharge of all pollutants to the waters of the
state.60 As in California, these permits are to assure compliance with the CWA and the
NPDES program. Michigan has been delegated authority to administer the NPDES permit
program for industrial facilities and to conduct its own pretreatment program. The state office
issues NPDES permits that apply the federal effluent limitations, including the numeric
limitations for run-off from materials storage piles (see 40 CFR 411).61 These permits contain
pH and total suspended solids limitations for treated process waters, treated quarry ground water
and quarry stormwater. The state, however, has not been delegated by EPA the authority to
issue general permits. Therefore, cement plants must submit individual applications to comply
with the new federal requirements for stormwater discharges associated with industrial activity.
Regional DNR offices conduct inspections and enforce permits.62 Some cement plants
are considered major dischargers and are inspected annually. Like California, Michigan uses
PCS to meet the information, inspection, and enforcement needs of its water quality program.
The violations discovered most often for all industrial facilities, including cement plants, are
effluent limitation violations and permit compliance schedule violations. Monitoring and
recordkeeping violations occur less frequently. About 80 percent of all industrial and municipal
dischargers are found to be "significantly in compliance" with their permit requirements. An
additional 10 percent are found to be out of compliance, with less serious violations. The
remaining 10 percent are out of compliance, with more serious violations. Overall, industrial
facilities are in better compliance today than in the past due to increased environmental
awareness. The cement manufacturing industry is not generally considered to have major
compliance problems.63
The type of enforcement action taken depends on the severity of the violation. When a
facility is found to be out of compliance, DNR first issues a verbal or written notice of
noncompliance. If no remedial action occurs, a formal notice of violation is issued. Negotiated
60 Mich. Comp. Laws §§ 323.1-323.13.
61 Personal communication with Pete Ostlund, Chief of Industrial Permits Section, Surface Water Quality Division,
Michigan Department of Natural Resources, September 15, 1992.
62 Personal communication with Tom Rohrer, Chief of Enforcement Unit, Surface Water Quality Division, May 24,
1993.
63 Ibid.
-------
7-26
settlements bring most facilities into compliance. These settlements include penalties and
schedules to return a facility into compliance. For more serious violations, emergency situations,
and excessive recalcitrance on the part of the facility, DNR is authorized to issue abatement
orders, revoke permits, and file civil and criminal actions.
Water Quality Standards
The state's water quality standards provide, as a minimum, that all waters be protected
for agricultural use, navigation, industrial water supply, public water supply at the point of water
intake, warm-water fish, and wading. Areas that must be suitable for swimming include all of the
Great Lakes, their connecting waterways, and all inland lakes. The swimming rule does not
apply in mixing zones, areas where a point source discharge is mixed with a receiving water. The
overall aim is that all waters outside mixing zones must be suitable for swimming during the
summer months. No degradation of waters may occur without a demonstration that such an
activity would not be unreasonable and would promote the public interest. Dissolved solids must
not exceed concentrations that are or may become injurious to any of the above designated uses.
Ground-Water Protection
Michigan also has established ground-water quality regulations to protect the public and
to maintain the quality of ground waters in all usable aquifers for individual, public, industrial,
•and agricultural water supplies. These regulations establish the following:
• the goal of non-degradation of ground-water quality in useable aquifers;
• the requirements for hydrogeological study before permitting a discharge into
ground waters;
• water quality parameters (e.g., metals, organic compounds, and toxic materials);
and
• a ground-water monitoring system based on the hydrogeological study, local
conditions, and the type of discharge for new and existing waste management
facilities.
Pennsylvania
Pennsylvania's Clean Streams Law authorizes the Department of Environmental
Resources (DER) to establish a program to prevent water pollution and to improve the purity of
Pennsylvania's waters.64 Pennsylvania has full authority to administer the federal NPDES
permit program and to issue general permits. Unlike Michigan, DER's six regional offices issue
individual permits. DER adopts the federal effluent limitations for cement plants and is
implementing the federal requirements for stormwater discharges. An important feature of the
state's NPDES program is that it applies both to streams and to ground water.
64 35 Pa. Stat. § 691.
-------
7-27
The six regional offices conduct monitoring and enforcement activities, and have
approximately 55 inspectors to regulate about 4,500 sewage and industrial dischargers.65
Regional staff conduct annual inspections and respond to emergencies, pollution events, and
complaints. Inspections include observations of treatment unit operation, effluent sampling,
identifying problems, recommending solutions, and citing violations. During the past few years,
Pennsylvania has made a considerable effort to enhance its monitoring program through
automating the effluent limits data and the discharge monitoring using PCS. Monitoring data for
all NPDES dischargers are reviewed on a routine basis.
Compliance and enforcement actions rely on both on-site monitoring and inspection data.
Violations may result in notices, orders issued on site, penalty assessments, and civil and criminal
actions. If environmental damage or willfulness was not involved, an attempt is made to obtain
voluntary compliance. Usually a notice of violation requests correction by a specific date or the
submittal of a compliance schedule. In more serious situations, higher level administrative, civil,
or criminal actions may be the first step.
Water Quality Standards
Pennsylvania has established designated water uses and water quality criteria.66 These
designatd water uses include: support of warm-water fish; potable water supply, after treatment;
industrial, livestock, and wildlife water supply; and irrigation, boating, fishing, water-contact
sports, and aesthetics. In addition, specific water quality criteria for pH and total dissolved solids
apply state-wide unless other numeric criteria are established for specific water body segments.
In general, the pH limit is a range between 6.0 and 9.0, and the limit for total dissolved solids is
500 milligrams per liter on a monthly average with a maximum limit at any time of 750
milligrams per liter.67
Pennsylvania participates with the neighboring states of New Jersey, New York, and
Delaware, as well as with the federal government, in the Delaware River Basin Compact. This
agreement requires cooperative efforts to preserve the water's recreational and fish-producing
value. The Delaware River Basin Commission, composed of the governors of the signatory states
(or their designees) and the Secretary of the Interior, establishes comprehensive water quality
standards. These standards impose limitations for pH and total dissolved solids that are more
stringent than those imposed by Pennsylvania's water quality standards. Five cement facilities are
located within the Delaware River Basin and are therefore subject to the Delaware River Basin
Commission's water quality standards.
Ground-Water Protection
Pennsylvania has developed a Ground Water Protection Strategy to protect ground-water
resources from contamination through the application of best demonstrated control technologies,
the use of BMPs, monitoring of permit compliance, detection of ground-water contamination,
and assessment and remediation. The Commonwealth's ultimate goal is non-degradation of
65 A total of 5,614 inspections were conducted in FY 91 of industrial and municipal dischargers. See Pennsylvania
Department of Environmental Resources 1992 Water Quality Assessment.
66 25 Pa. Code § 16.
67 25 Pa, Code § 93.6.
-------
7-28
ground-water quality. While the state does not have ground-water quality standards, ground-
water protection is considered when establishing permit requirements for wastewater and
stormwater discharges or solid and residual waste management facility design standards.
Stormwater Management Requirements
Unlike Michigan, Pennsylvania has been delegated NPDES general permit authority by
EPA. Cement kiln facilities can apply for coverage under Pennsylvania's general permit for
discharges of stormwater associated with industrial activities. Cement plant operators are
required to apply for individual stormwater discharge permits from DER for run-off from
materials storage piles because they are subject to the Cement Manufacturing Point Source
Category effluent limitations. The general permit, however, authorizes other stormwater
discharges at a cement facility. The general permit requires that the owner of a facility develop
and implement a Preparedness, Prevention, and Contingency Plan; comply with effluent
limitations (e.g., pH between 6 and 9); and conduct annual monitoring.
Texas
In Texas, cement facilities subject to the federal NPDES permit program must receive an
NPDES permit from EPA Region VI as well as an industrial discharge permit from the Texas
Natural Resources Conservation Commission (Commission). Texas has not been delegated
authority by EPA to administer the NPDES permit program for industrial facilities. Though
cement facilities need dual endorsement, the requirements are similar. As Texas does not have
any categorical effluent limitations for cement plants, both EPA Region VI and the Commission
utilize the effluent guideline limitations for cement plants (40 CFR 411). These limitations are
used as a minimum baseline, subject to more stringent limits, if necessary, to meet state water
quality standards.
Unlike the other states, dual enforcement authority exists because both EPA Region VI
and the Commission are responsible for compliance monitoring. EPA Region VI is responsible
for monitoring compliance with the federal effluent limitations and the federal stormwater
requirements. Per an agreement with EPA, the Commission will be primarily responsible for
enforcement of the federal standards.68 In addition, the Commission must identify alleged
Texas Water Code violations and bring violators into compliance with the statutes.
The Commission has 14 field offices that conduct inspections (a 15th office will be added
in 1994). Together these offices have approximately 50-70 inspectors." The Field Operations
Division, located at the Commission's headquarters in Austin, coordinates inspection activities
among these field offices. The frequency of inspections is established based on the type of
facility. The field offices inspect all major facilities annually. Cement plants, however, are not
considered major facilities because they do not discharge large quantities of effluent.70 After
inspecting all major facilities, time and budget permitting, the offices will inspect other, minor
dischargers such as cement plants. The offices will consider minor dischargers a high priority for
68 Personal communication with Everett Spenser, EPA Region VI, Enforcement Branch, September 9, 1993.
69 Personal communication with Rick Ruddell, Watershed Management Division, and Earnest Heyer, Field
Operations Division, May 16, 1993.
-------
7-29
inspection if they have received a complaint or if the facility has a past record of violations.
Inspectors review a facility's compliance with the terms of its permit, observe treatment facility
operations, and sample effluent discharges.
The Commission takes enforcement actions based on results of inspections by the field
offices, monitoring data included in the monthly effluent report (Texas also uses PCS to manage
discharger information), and public complaints. In a manner similar to that used in other states,
minor violations are handled by the field office. These offices send out notices of violations and
require the permittee to return to compliance within 30-60 days. The field offices refer to
headquarters all severe and continuing violations. The Commission coordinates enforcement
activities with EPA Region VI for violations at facilities with NPDES permits. In addition, the
Commission meets with EPA Region VI staff on a quarterly basis to discuss enforcement
activities and noncompliance reports. Commission and EPA Region VI personnel do not
consider cement plants to have had major compliance problems.71
Water Quality Standards
As in the other states, Texas establishes both narrative and numeric water quality
standards. The state has three major categories of water quality designations: (1) contact (e.g.,
swimming) and noncontact (e.g., boating) recreation; (2) domestic water supply; and (3) five
subcategories of aquatic life (i.e., limited quality, intermediate quality, high quality, exceptional
quality, and oyster waters).72 The domestic water supply and aquatic life designations are of
special concern to cement facility operators, as both impose limits on concentrations of heavy
metals, as well as toxic and chemical materials. In addition, the State has numeric criteria for
each classified segment of a given water body.
Ground-Water Protection
A state ground-water protection policy was adopted in 1989 that sets nondegradation of
ground-water resources as its goal. The policy recognizes the variability of Texas' aquifers, the
importance of maintaining water quality for existing and potential uses, the protection of the
environment, and the maintenance and enhancement of the long-term economic health of the
state. Discharges of pollutants, disposal of wastes, and other regulated activities must be in a
manner that will maintain present uses and not impair potential uses of ground water or pose a
public health hazard. The State legislature is currently debating proposals that would require
ground-water monitoring for disposal facilities located in the Edwards Aquifer region.
Texas is the only state (of the four states included in this report) to have its Wellhead
Protection Program approved by EPA. The Wellhead Protection Program identifies the roles of
state and local agencies and attempts to coordinate the state's existing ground water and well
water protection legislation. A wellhead protection program attempts to protect ground water
surrounding a well or wellfield that supplies a public drinking water system. A wellhead
protection program promotes the use of best demonstrated technology to prevent contamination.
11 Everett Spenser, EPA Region VI, Enforcement Branch, September 9, 1993, and Rick Ruddell, Watershed
Management Division, and Earnest Heyer, Field Operations Division, Texas Natural Resources Conservation
Commission, May 16, 1993.
72 Tex. Admin. Code tit. 31, § 307.7.
-------
7-30
Stormwater Management Requirements
No requirement exists under Texas regulations for the permitting of stormwater
discharges. However, cement facilities are subject to the numeric limitations for run-off from
materials storage piles found in the federal effluent limitations. These numeric limitations are
incorporated as baseline requirements in the Texas Industrial Discharge permit and the federal
NPDES permit.
As Texas does not have NPDES authority, EPA will administer the new stormwater
requirements, including the issuance of general permits. However, EPA must incorporate the
Texas hazardous metal effluent limitations (e.g., for cadmium, lead) as part of any stormwater
permit, to comply with Texas water quality standards.73
7J SOLID WASTE MANAGEMENT
Three general approaches are used in managing CKD at cement plants: recycling,
beneficial use, and land management. While the least wasteful method for managing CKD
would be to recycle it to the raw feed, excessive alkali content in the dust limits the amount of
CKD that can be directly recycled without upsetting the proper functioning of the kiln (see
Chapter 3, Section 3.1.1). When recycling can not be used to manage all CKD, the CKD may be
disposed or stockpiled on site in waste management units or sold off site for beneficial use.
Beneficial uses of CKD include applications as soil amendments, material additives, liming
agents, road sub-bases, or waste stabilization agents. Various solid waste management
requirements exist for disposal and stockpiling of CKD. In addition, some states require a
permit before allowing off-site beneficial uses of CKD.
Under federal law, the Bevill Amendment (Section 3001 of RCRA) temporarily excludes
CKD from regulation as a hazardous waste under Subtitle C of RCRA, pending study. While
temporarily excluding CKD from regulation as a hazardous waste, the Bevill Amendment did not
preclude CKD regulation under other provisions of federal or state law. Currently, CKD is
subject to federal criteria as a non-hazardous solid waste under Subtitle D of RCRA. In
addition, CERCLA provides the federal government with the authority and resources to respond
to situations in which CKD wastes are or may be released into the environment such that they
pose an imminent and substantial danger.
Because nothing prevents states from imposing more stringent hazardous waste
requirements, states (such as California) may, and in some cases do, characterize CKD as a
hazardous waste. Pennsylvania, on the other hand, classifies CKD as a residual waste.
Pennsylvania regulates CKD less stringently than if CKD was considered a hazardous waste but
still requires comprehensive waste management practices. Michigan and Texas characterize CKD
as an industrial, non-hazardous solid waste and subject CKD to management requirements that
vary from those of the other states.
73 57 Fed. Reg. 41236.
-------
7-31
7-3.1 Federal Controls
The Resource Conservation and Recovery Act
The Resource Conservation and Recovery Act of 1976 (RCRA), as amended, is the
primary statute governing the management of solid and hazardous waste.74 The principle
objectives of RCRA are to:
• Promote the protection of human health and the environment from potential
adverse effects of improper solid and hazardous waste management;
• Conserve material and energy resources through source reduction and waste
recycling;
• Reduce or eliminate the generation of hazardous waste as expeditiously as
possible; and
• Improve solid waste management practices.
Special requirements for hazardous wastes are found in Subtitle C of RCRA. Subtitle C
provides a statutory framework for tracking all hazardous and toxic wastes from "cradle to grave,"
that is, from their generation to their final disposal, destruction, or recycling.
Under Section 3006 of RCRA, EPA may authorize states to administer and enforce a
state hazardous waste program in lieu of the federal Subtitle C program. In order to receive
authorization, a state's program must contain hazardous waste management regulations at least
as stringent as federal Subtitle C standards. A state's enforcement provisions must also provide
at least equivalent penalties to those required in RCRA and enforcement activities performed by
EPA. EPA has approved the state-level programs for implementing the Subtitle C hazardous
waste management system of all four of the states analyzed in this report.
Pursuant to regulations issued by EPA (40 CFR Part 261), solid wastes that meet EPA
hazardous waste criteria with respect to "toxicity, persistence, degradability in nature, potential
for accumulation in tissue, and other related factors such as flammability, corrosiveness. .." are
subject to RCRA's Subtitle C requirements. Generators of these wastes are generally required
to comply with labeling, storage, transportation, and disposal requirements.
In 1980, however, Congress enacted the Bevill Amendment, which temporarily exempted
certain categories of high volume solid wastes (including CKD) from regulation as a hazardous
waste under Subtitle C of RCRA, pending study and a Regulatory Determination. While
temporarily excluding CKD from regulation as a hazardous waste, the Bevill amendment did not
exempt CKD from regulation under other provisions of federal or state law. Currently, CKD is
subject to regulation as a non-hazardous solid waste (hereafter referred to as solid waste) under
Subtitle D of RCRA.75 Subtitle D established a cooperative framework for federal, state, and
local governments to control the management of solid waste. The actual planning and
implementation of solid waste programs are state and local functions.
74 42 U.S.C. §§ 6901 to 6992K.
75 42 U.S.C. §§ 6942 - 6949a.
-------
7-32
The broad definition of solid waste in the federal regulations includes industrial waste
such as CKD. "Solid waste" is defined as "any garbage, refuse, sludge from a waste treatment
plant, water supply treatment plant, or air pollution control facility and other discarded material,
including solid, liquid, semisolid, or contained gaseous material resulting from industrial,
commercial, mining, and agricultural operations ... "7
-------
7-33
to perform adequate testing of CKD to determine if it should be characterized as hazardous
waste. EPA alleges that the other facility stored hazardous CKD without a permit.
Comprehensive Environmental Response, Compensation, and Liability Act
Run-off, leachate, and other air and water emissions from CKD can be subject to the
regulatory and liability provisions of the Comprehensive Environmental Response,
Compensation, and Liability Act (CERCLA, or Superfund). Superfund provides the federal
government with the authority and resources to respond to situations in which pollutants or
contaminants are or may be released into the environment such that they pose an "imminent and
substantial danger to the public health or welfare...."81
CERCLA authorizes EPA to respond to immediate threats to the environment or human
health in situations in which a responsible party cannot act or cannot be readily identified. In
such situations, EPA can proceed with necessary containment or removal actions. Where
conditions allow, the Agency can also undertake more detailed remedial actions. Section 106
provides authority for administrative orders necessary to protect public health and the
environment.
In those situations in which responsible parties that can respond "properly and promptly"
can be identified, EPA is authorized to establish what remedial actions are required and to
oversee the responsible parties' cleanup efforts. In all cases, the owners and/or other responsible
parties are liable for the costs of cleaning up the hazardous waste problem, and for correcting
damages to natural resources. Two cement kiln facilities are currently listed on the NPL and are
discussed in detail in the Damage Case Evaluation (Chapter 5), above. These two NPL sites are
as follows:
• Holnam Incorporated, Mason City, Iowa; and
• Portland Cement Company, Salt Lake City, Utah.82
7.3.2 State Controls
In the area of solid waste, significant differences in CKD management exist between the
states, specifically with respect to how states characterize CKD waste. California characterizes
CKD as a hazardous waste. Pennsylvania, on the other hand, classifies CKD as a residual waste,
regulating CKD less stringently than if it was considered a hazardous waste, but still requiring
comprehensive waste management practices. Michigan and Texas characterize CKD as an
industrial, non-hazardous solid waste and therefore, subject CKD to fewer management
requirements than either California or Pennsylvania. A summary of the four states' solid waste
management controls can be found in Exhibit 7-4, and a discussion of individual state's solid
waste management requirements follows.
81 42 U.S.C. § 9604.
82 A third site, Lehigh Portland Cement Company's Mason City, Iowa, plant was placed on the NPL on August
30, 1990. In litigation, Lehigh identified a number of concerns regarding the hazard ranking score. After reviewing
the issues regarding the calculation of the score on the hazard ranking system, the Agency decided not to contest
Lehigh's challenge to the listing decision. The listing was vacated by mutual consent in October, 1992. Removal of
Lehigh's Mason City site from the NPL does not affect clean-up at the site. For further discussion of this site, see
Section 5: Documented and Potential Damages from Management of CKD.
-------
7-34
Exhibit 7-4
Summary of State Solid Waste Management Controls
States
CKD
Classification
Permit
required
Ground-Water
Monitoring
Reporting
Requirements
Beneficial Use
Approvals
Required
California
Hazardous waste;
moratorium on enforcement
Yes; design standards, siting
restrictions, and operating
requirements
Monitoring system designed
to provide best assurance of
earliest possible detection of
a release from facility
Permit will include frequency
of reporting of ground-water
sampling results; annual
hazardous waste report
No
Michigan
Industrial solid waste
Yes; design standards,
siting restrictions, and
operating requirements
Monitoring system to
evaluate ground water at
the solid waste facility
boundary, number and
location of wells will vary
depending on
hydrogeological study
Quarterly ground-water
sampling results
Sometimes; for uses that
could be considered
disposal and do not
involve a licensed waste
management facility (e.g.,
use as fill material)
Pennsylvania
Residual waste
Yes; design standards, siting
restrictions, and operating
requirements
Sufficient number of
monitoring wells to be
representative of water
quality (at least one
upgradient and three
downgradient wells)
Quarterly and annual
reporting of sampling results;
biennial residual waste report
and source reduction plan
Yes; general permits may be
issued on a regional or slate-
wide basis for a particular
use
Texas
Industrial
solid waste
No';
notification to
State of waste
management
activities
Technical
guidelines on
ground-water
monitoring
systems, but
monitoring
not required
Notification
to State of
waste
classification
and waste
management
activities
No
' On-site management of industrial non-hazardous waste does not require a permit. CKD transported off site must be
managed in a permitted facility.
California
In addition to assigning responsibilities to the California Water Resources Control Board
and the regional water resource control boards to regulate water discharges, the Porter-Cologne
Act gives the boards specific authority to regulate discharges of waste to land.83 The boards
also have the authority to manage landfills and waste piles. This aspect of the California water
pollution control law complements the state's Hazardous Waste Control Law. Under the
Hazardous Waste Control Law, the Department of Toxic Substances Control (Department)
manages California's hazardous waste program. The California Water Resources Control Board
and the California Environmental Protection Agency's Department of Toxic Substances Control
are presently revising their regulations to make them more consistent.
Currently, CKD is considered a non-RCRA hazardous waste under California's
Hazardous Waste Control Act but may be reclassified as a "special waste." California defines a
83 Cal. Water Code §§ 13172, 13226-13227.
-------
7-35
"special waste" as "a waste which is a hazardous waste only because it contains an inorganic
substance or substances which cause it to pose a chronic toxicity hazard to human health or the
environment ...n84 The waste must meet all of the criteria and requirements of a special waste
as specified in Sections 66261.122 and 66261.124 of the state's hazardous waste regulations; for
example, the waste must not exhibit the characteristics of corrosivity, ignitability, reactivity, or
toxicity (presumably, "acute" toxicity), as defined by the state.85 The hazardous waste
management regulations specifically list CKD as a waste that may be classified as a special
waste.86
For CKD to be classified as a special waste, the owner or operator of a cement plant
must submit an application to the Department. The application must contain the following
information:
• Address where the waste is generated;
• Description of the waste that includes its source, physical state, quantity, and rate
of generation; and
• Chemical analysis data.87
A representative CKD sample can be used in the chemical analysis because the plant
continuously uses the same kinds of raw materials with respect to their origin, composition, and
properties.
Upon written approval by the Department, a cement plant operator may manage CKD as
a special waste, allowing it to be disposed in a landfill that is not permitted or operated under
the more stringent hazardous waste requirements. As a special waste, the landfill does not have
to comply with the hazardous waste facility design, closure and post-closure care, and financial
assurance requirements.88 However, the waste management facility must comply with any waste
discharge requirements issued by the regional water quality control board.89 The owner or
operator of a facility also must have been granted a variance that allows for the disposal of
special wastes. Unless specifically waived by a variance, the owner or operator of a waste
management facility that accepts special waste is subject to the hazardous waste enforcement,
manifesting, and reporting requirements.90
84 Cal. Admin Code tit. 22, § 66260.10 (also referred to as the California Code of Regulations).
85 A solid waste considered hazardous under California's more stringent corrosivity or toxicity characteristics
definition might not be considered a hazardous waste under the federal RCRA Subtitle C hazardous waste
characteristics definition.
** Cal. Admin. Code tit. 22, § 66261.120.
87 Cal. Admin. Code tit. 22, § 66261.124.
** Cal. Admin. Code tit. 22, § 66261.126(a).
19 As stated above, in addition to regulating water discharges, regional water quality control boards are authorized
to regulate discharges of waste to land. Cal. Water Code §§ 13172, 13226-13227.
90 Cal. Admin. Code tit. 22, § 66261.126(c).
-------
7-36
Even if a cement plant's CKD was to be classified as a special waste, the owner or
operator of the plant is subject to all of the requirements that apply to a generator of hazardous
waste.91 As a generator of hazardous waste, the owner or operator must analyze the CKD and
obtain an identification number. Manifesting and other reporting requirements also apply, as do
requirements and limitations for storing hazardous waste. Owners or operators may, however,
obtain a variance from any or all of these requirements. In practice, because the state has never
enforced the management of CKD as a hazardous waste, the majority of cement plants in
California have not applied to reclassify their CKD as a special waste, nor do they manage it as a
hazardous waste.92 Only one cement plant began the process of having CKD reclassified as a
special waste, but it never provided the Department with all of the required studies.93
Therefore, the Department has never made a determination on whether CKD could be
reclassified as a special waste.
In response to potential CKD enforcement concerns, the Governor of California signed a
bill in October 1992 that places a one year moratorium on the enforcement of hazardous waste
requirements for CKD that fails the California hazardous waste corrosivity characterization
test.94 This bill, effective January 1, 1993, and extending through January 1, 1994, authorizes a
study on the health-based effects of CKD with funding by the California Cement Manufacturers
Association. This study is subject to review by a committee consisting of California EPA, the
Department of Toxic Substances Control, the State Water Resources Control Board, the
California Cement Manufacturers Association, and an environmental organization. The study
will analyze whether the hazardous waste corrosivity criteria, including testing protocols, should
be applied to CKD. Due to procurement problems, this study remains at the bid stage and the
California Cement Manufacturers Association will be asking the legislature for a one-year
extension on the moratorium so the study can be completed.95
The moratorium only provides a temporary exemption from the enforcement of
hazardous waste management requirements for CKD that fails the California hazardous waste
corrosivity characterization test. One cement plant also failed California's hazardous waste
characterization test because of the presence of lead in the CKD.96 The Department, however,
Jssued a letter in 1985 stating that the cement plant operator could treat the CKD as a non-
hazardous waste because the lead had little potential to leach out of the CKD. Under this
variance, the plant operator must wet the CKD and allow it to solidify prior to managing it in
waste piles. The waste piles do not have liners or concrete pads. Ground-water monitoring wells
91 Cal. Admin. Code tit. 22, § 66261.126(d).
92 Persona] communication with Chris Marxen, California Environmental Protection Agency, Department of Toxic
Substances, Waste Evaluation Unit, and Fred Fontus, California Environmental Protection Agency, Department of
Toxic Substances, Surveillance and Enforcement, September 10, 1993.
93 Letter from Stanford Lau, Toxic Substances Control Division, to Ralph Mitchell, Lone Star Industries, Inc., July
20, 1988.
94 Assembly Bill 3789, Cal. Health and Safety Code § 25141.1.
* Personal communication with Chris Marxen, California Environmental Protection Agency, Department of Toxic
Substances, Waste Evaluation Unit, September 10, 1993.
96 Persona] communication with Fred Fontus, California Environmental Protection Agency, Department of Toxic
Substances, September, 10, 1993.
-------
7-37
near the waste piles consistently demonstrate no ground-water contamination has occurred from
lead leaching from the CKD.
Recently, a weekly sample from newly-generated CKD at this cement plant demonstrated
elevated lead levels. The plant immediately notified the Department, The plant segregated this
CKD and will manage this batch of CKD as a hazardous waste.
• The moratorium does not affect inspections or enforcement of other environmental
requirements such as ground-water protection. Cement plants continue to be inspected annually.
Michigan
The Michigan Solid Waste Management Act authorizes the Michigan Department of
Natural Resources (DNR) to manage industrial solid waste and hazardous waste disposal.97
"Solid waste" is defined broadly to include "garbage, rubbish, ashes, incinerator ash, incinerator
residue, street cleanings, municipal and industrial sludges, and solid commercial and solid
industrial waste .. ."'8 CKD is considered a solid industrial waste under this definition, and is
not characterized as a hazardous waste under Michigan regulations. CKD is considered a Type
II solid waste." DNR has become concerned about the potential environmental impact of past
disposal of CKD in quarries and is studying CKD disposal practices at the three currently-
operating cement plants.100 DNR has discovered some ground-water contamination
downgradient from CKD piles, but has no proof that the contamination originates from the CKD
and is not the result of natural leachate from bedrock fractures. This situation continues to be
monitored.
If the owner or operator wants to
manage CKD on site, a construction permit
and operating permit for landfills are
required. The permitting process requires a
separate license to operate a facility. The
license application must be accompanied by
an engineer's certification that construction
was completed in accordance with the
previously approved plans. A bond to cover i^——«^—^——^^—^—^——^—
the costs^of closure and post-closure
monitoring is also required. With regard to post-closure land use, the Department is in the
The construction permit application is to be
accompanied by the following:
• A hydrogeological report and monitoring
program;
• Engineering plans; and
• An environmental assessment.
97 Mich. Comp. Laws §§ 401-436.
98 Mich. Comp. Laws § 299.407.
99 Wastes listed as hazardous or that demonstrate hazardous characteristics are classified as Type I wastes. Type
III wastes are inert and essentially insoluble (e.g., demolition debris, rock, or dirt). Type II wastes are all the wastes
that cannot be considered Type I or Type III and include garbage and rubbish. See Letter from Mindy Koch, Acting
Chief, Waste Management Division, Department of Natural Resources, to Myron Black, LaFarge Corporation, dated
September 27, 1991.
100 Personal communication with Brad Venman, Michigan Department of Natural Resources, Waste Management
Division, October 9, 1992. Currently, the cement plant located in Wayne County only grinds clinker delivered from
Canadian cement plants. The kiln which was permitted to burn hazardous waste for fuel was shut down for economic
reasons.
-------
7-38
process of reviewing a cement facility's closure plan to turn a limestone quarry containing CKD
into an inland lake or marina.101
Another option is to have CKD receive a "designation of inertness" from DNR.102 This
designation would exempt CKD from most solid waste management requirements (e.g.,
construction or operating permits). To be classified as inert material, the CKD must meet
Toxicity Characteristic Leaching Procedure (TCLP) concentration limits. If the designation of .
inertness is approved for CKD, the owner or operator of the cement plant would be required to
characterize the CKD a minimum of once per year. One facility failed this test because the
leachable concentration of lead was too high.103
At another cement plant, DNR outlined three options for future CKD disposal: 1)
transport the CKD to an off-site facility licensed to manage Type II wastes; 2) develop an on-site
facility that is permitted and licensed; or 3) obtain a "designation of inertness" from DNR for the
continued disposal of CKD in the on-site quarry.104 If CKD is to be transported off site for
disposal, the disposal must be consistent with county solid waste management plans.
DNR only requires beneficial use permits for certain uses. For example, no beneficial
use approval is necessary when CKD is used to solidify liquid hazardous wastes because controls
and reporting requirements exist and are the responsibility of the liquid hazardous waste
management facility. Similarly, CKD used to solidify drilling muds and cuttings from oil and gas
exploration activities or tank bottoms from oil and gas production facilities would not need to be
permitted because the solidified wastes would be taken to a licensed landfill. DNR is more
concerned about how CKD is ultimately disposed. Therefore, approval and reporting
requirements would apply if the beneficial use of CKD involved activities such as river bank
stabilization or restoration projects that would use CKD as fill. Historically, CKD has been used
as fill along the Detroit River. Today, this type of use would receive close scrutiny.
Pennsylvania
Unlike California and Michigan, CKD is classified by Pennsylvania as a residual waste.
The Pennsylvania Department of Environmental Resources (DER) defines "residual waste" as
"[g]arbage, refuse, other discarded material or other waste including solid, liquid, semisolid, or
contained gaseous materials resulting from industrial, mining and agricultural operations ... if it
is not hazardous."105 Pennsylvania's definition of residual waste is similar to Michigan's
definition of industrial waste.
101 Personal communication with Brad Venman, Department of Natural Resources, Waste Management Division,
October 9, 1992.
102
Mich. Comp. Laws § 299.408(3), Mich. Admin. Code §§ 299.4102(h) (viii) and 299.4301(3).
103 Letter from Mindy Koch, Acting Chief, Waste Management Division, Department of Natural Resources, to
Myron Black, LaFarge Corporation, dated September 27, 1991.
104 Ibid.
105 25 Pa. Code § 287.1.
-------
7-39
The six regional DER offices conduct inspections; inspections based on the new residual
waste regulations are pending.106 Due to limited resources, inspections of hazardous waste
management facilities and municipal waste management facilities are a higher priority than
residual waste management inspections. The regulations explicitly state that DER does not have
a duty to conduct a minimum number of inspections per year at a facility.107 While no
minimum number of inspections is required, the regulations recommend that DER conduct 12
inspections per year for each residual waste landfill. DER may conduct as many inspections as
necessary for public health or safety reasons.
Currently, DER residual waste management enforcement is concentrating its efforts on
obtaining compliance with notification requirements. These requirements mandate that residual
waste generators provide to DER by March 1,1993, basic information about their waste and its
management. Approximately 67 percent of residual waste generators submitted the required
report.108
In a manner similar to that of the other states, DER has administrative, civil, and
criminal enforcement authority, including the authority to levy penalties.10' The residual waste
management regulations establish specific violations for which a civil penalty should be assessed
(e.g., acceptance of waste not approved under a permit), but do not restrict DER from assessing
penalties for violations not explicitly set forth.110 These regulations also establish factors for
determining penalty amounts (e.g., willfulness of violation). For most cases of noncompliance,
DER plans to send a notice of violation and arrange for a meeting to discuss the violations,111
at which a date will be established by which the violations must be abated. If the violations are
not corrected, DER may seek consent orders, civil penalties, and in extreme situations, criminal
penalties.
As a residual waste, CKD is regulated less stringently than it would be if classified as a
hazardous waste, but is still subject to comprehensive waste management controls. Cement
facilities are subject to the residual waste management requirements in Pennsylvania's Solid
Waste Management Act and the new Residual Waste Management Regulations.112 These new
regulations, adopted in July 1992, reflect the increasing complexity of industrial waste
management and afford greater protection of the State's ground water. These new regulations
expand residual waste management requirements and replace older residual waste standards.
106 Personal communication with Reno Vacheski, Pennsylvania Department of Environmental Resources, Region
U, Waste Management Bureau, September 10, 1993.
107
25 Pa. Code § 287.421.
I0* Persona] communication with Sam Sloan, Pennsylvania Department of Environmental Resources, Bureau of
Waste Management, Division of Municipal and Residual Waste, September 10, 1993.
109 Penn. Stat. §§ 6018.602-6018.606.
110 25 Pa. Code § 287.411.
111 Personal communication with Reno Vacheski, Pennsylvania Department of Environmental Resources, Region
II, Waste Management Bureau, September 10, 1993.
112 35 Penn. Stat. Ann. § 6018.102 and 25 Pa. Code § 287.
-------
7-40
Cement plants that manage CKD are required to obtain a Residual Waste Processing
and/or Disposal Permit.113 This permit attempts to balance state-of-the-art environmental
protection methods and the risks presented by particular wastes at a facility. This permitting
process requires the following elements:
• Environmental assessment;
• Analysis of the waste(s);
• Source reduction strategy; and
• Plan for ongoing analysis of the waste(s).114
In establishing disposal requirements for a specific facility, a waste classification system is used to
determine landfill design standards. A leaching analysis based on EPA's methods compares the
amount of contaminants in the waste's leachate to ground-water parameters; this analysis
determines the appropriate landfill design class.
A ground water parameter is one of the foDowing:
• the maximum contaminant level goal
(MCLG) under the Federal Safe Drinking
Water Act;
• the primary or secondary maximum
contaminant level (MCL);or
• for other contaminants, concentrations
derived from the U.S. EPA's Integrated
Risk Information System (IRIS).
Depending on the results of the leach
tests, CKD may be placed in one of three
different types of landfills with various liner
and other requirements (e.g., Class I landfills
must comply with more stringent liner, siting,
and operating standards than either Class II
or Class III landfills). Under the industrial
waste regulations, a waste that leaches more
than 50 times the ground-water parameter for
any of the contaminants it contains would be
required to be disposed of in a Class I
landfill. If the contaminants are greater than
25 times the ground-water parameters for
metals and other cations, or more than 10
times the ground-water parameters for other contaminants, the CKD would be required to be
disposed of in a Class II landfill. Other wastes, in general, could be disposed of in a Class III
landfill.
The residual waste regulations require permits that include provisions for liners, leachate
collection systems, monitoring wells, and disposal of leachate. Though similar to the federal
Subtitle D criteria with regard to prohibitions on where facilities may be located (e.g., within the
100-year floodplain) the residual waste regulations are more stringent. Residual waste permits
will be issued for up to 10 years, but DER must review these permits every five years. Facilities
without permits must document planned closure procedures within a certain time frame.
Under the residual waste regulations, a cement facility must develop a source reduction
strategy and update that plan every five years. The strategy must contain the following elements:
• the methods and procedures that will be used to achieve a reduction in the weight
or toxicity of waste generated on the premises;
113 25 Pa. Code § 287.101.
114 25 Pa. Code §§ 287.121-287.134.
-------
7-41
• the magnitude of the projected reduction; and
• a timetable for when the reductions will occur.
Every two years, a cement facility must file a report describing the types of waste
generated, where the wastes are processed or disposed, and the facility's efforts to implement its
source reduction plan. The report is intended to provide baseline data about industrial waste
generation and source reduction in Pennsylvania.
Cement facilities may also be required to obtain a Residual Waste Beneficial Use
Approval.115 This approval is required if a cement plant plans to reuse constituents of residual
waste for a particular beneficial purpose. This requirement is currently in effect. Past uses of
CKD to make fertilizer and for use in soil stabilization, land reclamation, waste remediation, and
sewage sludge dewatering would now be required to obtain DER beneficial use approval. On its
own initiative, or at the request of other parties, DER may issue general permits for beneficial
use on a regional or Commonwealth-wide basis. The general permit may establish concentration
limits for contamination or place restrictions on the use. As of this date, DER has not received
and approved any general permit applications or issued on its own initiative a general permit for
the beneficial uses of CKD. Those cement plants using CKD for a beneficial purpose need to
apply for a permit.
Texas
The Texas Solid Waste Act grants the Texas Natural Resources Conservation
Commission jurisdiction over hazardous and industrial waste management.116 Under new waste
classification regulations finalized in November 1992, owners or operators of a facility must
submit a registration form to the Commission that makes a determination of their waste
classification and assigns a waste code to their own wastes.
The Commission conducts inspections of all permitted non-hazardous waste management
facilities annually.117 Cement plants that burn hazardous waste for fuel are inspected annually
and are permitted RCRA facilities. The majority of cement plants, however, do not burn
hazardous waste and manage their CKD on site. These cement plants are exempt from
permitting requirements (this permit exemption for facilities managing their CKD on site is
discussed in greater detail below) and annual inspections. The Commission will inspect cement
plants that are exempt from permitting requirements if there are complaints.
Texas has administrative and civil enforcement authority, including the authority to levy
monetary penalties of up to $10,000 per day of violation. Generally, the Commission attempts to
issue agreed upon Corrective Action Directives to encourage voluntary compliance and
implementation of corrective action in an expedited manner. Administrative orders with
penalties for violations are also issued. In situations where the Commission believes that an
imminent and substantial endangerment to human health or the environment has occurred, an
115 25 Pa. Code § 287.611.
"'Tex. Health and Safety Code § 361.
117 Persona] communication with Earnest Hire, Texas Natural Resources Conservation Commission, Industrial and
Hazardous Waste Enforcement Section, September 10, 1993.
-------
. 7-42 .
order may be issued without the facility's consent to facilitate immediate corrective action. No
major enforcement or compliance problems are known to exist at cement plants.118
The Texas Solid Waste Management Regulations establish standards for all aspects of the
management and control of industrial solid waste; CKD is considered a non-hazardous industrial
solid waste. "Industrial solid waste" is defined as any solid waste resulting from or incidental to
any process of industry or manufacturing.119 Non-hazardous industrial solid waste is classified
as follows:
• Class 1 wastes that may pose a substantial danger to human health and the
environment because of hazardous characteristics;
• Class 2 wastes that are all wastes that cannot be considered Class 1 or Class 3;
and
• Class 3 wastes that are inert and essentially insoluble (e.g., rock, bricks, or dirt).
CKD would be considered a Class 1 corrosive waste if it is a semi-solid or solid which, when
mixed at a 1:1 ratio with water, produces a solution with a pH less than or equal to 2 or greater
than or equal to 12.5. Based on EPA studies, some CKD could be classified as a Class 1
industrial waste under this standard. The cement facility owner or operator would then be
required to manage the CKD only in permitted hazardous waste management facilities and
would be subject to manifesting and reporting requirements.
Wastes are classified in the most protective manner unless knowledge and/or data
demonstrate that a less conservative classification (Class 2 or 3) is applicable. If an owner or
operator considers its CKD to be a Class 2 or 3 waste, testing results must verify this
position.120 To be classified as a Class 3 waste, the cement kiln owner or operator must
demonstrate that the CKD does not exceed maximum leachable concentrations (based on EPA's
Toxicity Characteristic Leaching Procedure (TCLP)), or exceed Primary Drinking Water
Standards or the Total Dissolved Solids limit of the secondary standards. To be considered a
Class 2 waste, CKD must not fail the corrosivity test described above. Basically, a Class 2 waste
is not as totally innocuous and inert as is required for a Class 3 waste, but also does not present
the potential threat of a Class 1 waste.
The Industrial Waste Management Regulations exempt non-hazardous industrial waste
(Class 2 or 3) disposal facilities from the requirement to obtain a solid waste facility permit if the
waste is: 1) disposed on site; 2) the disposal site is located within 50 miles of the plant or
operation; and 3) the waste is not "commingled" with waste from another source.121 Off-site
waste management units must be permitted.
119 Tex. Admin. Code tit. 31, § 335.1.
120 Owners or operators must use an approved testing method such as those described in "Test Methods for the
Evaluation of Solid Waste, Physical/Chemical Analysis." In addition, the owner or operator must maintain
documentation of the sampling procedures. The Commission may review a waste characterization at any time to
determine if the waste has been appropriately classified.
121
Tex. Admin. Code tit. 31, § 335.2(d).
-------
7-43
Owners and operators of facilities that are exempt from the permitting requirement must
still comply with the following industrial waste management requirements:
• Owners and operators must notify the Commission 90 days prior to the onset of
disposal activities;
• Records must be kept of the description of the waste, quantities stored or
disposed, and quantities shipped off site;
• The storage or disposal of waste must not cause a nuisance or endanger the public
health or welfare; and
• Owners or operators must file a notice in the county deed records of the disposal.
In addition, owners or operators may be required to submit information on waste
management methods, facility engineering plans, and the geology of the facility's location. An
owner or operator is required to submit details of closure activities only if requested by the Texas
Natural Resources Conservation Commission.
The Commission provides technical guidelines to advise owners/operators of on-site waste
piles on appropriate liner materials and thickness, closure and post-closure care activities, and
site selection criteria, but these guidelines are not requirements that can be enforced. For
example, owners or operators are not expressly required to place liners under waste piles or to
monitor ground water. In addition, no closure and post-closure care requirements exist for on-
site non-hazardous industrial solid waste piles.
-------
7-44
CHAPTER SEVEN
EXISTING REGULATORY CONTROLS ON CKD MANAGEMENT
7.0 INTRODUCTION AND METHODS 1
7.0.1 Objectives 1
7.0.2 Methodology 1
7.0.3 Summary of Findings 2
7.0.4 Limitations of the Analysis 4
7.1 AIR ." -. 5
7.1.1 Federal Controls 5
Clean Air Act . 5
Implementing Regulations 5
State Implementation Plans 5
New Source Performance Standards 6
Prevention of Significant Deterioration 6
Nonattainment Review 7
Hazardous Air Pollutants 7
Boiler and Industrial Furnace Regulations 8
7.1.2 State Controls 9
California 10
Ambient Air Quality Standards • 11
Authority To Construct 11
Permit To Operate 12
Prevention of Significant Deterioration and New Source Review .... 12
-------
7-45
Michigan 13
Ambient Air Quality and Permits 13
Paniculate Matter and Visible Emissions Limitations 14
Pennsylvania 14
Nonattainment Review 15
Particulate Matter and Visible Emissions Limitations 15
Texas 16
Other Requirements Applicable to Cement Plants 16
Special Report on Texas Cement Plants' Excess Emissions 18
7.2 WATER 18
7.2.1 Federal Controls 18
The Clean Water Act 18
Safe Drinking Water Act 20
7.2.2 State Controls 21
California 22
Process Wastewater Requirements 23
Water Quality Standards 23
Stormwater Management Requirements 24
Michigan 24
Water Quality Standards 25
Ground-Water Protection 25
Pennsylvania 25
Water Quality Standards 26
Ground-Water Protection 26
Stormwater Management Requirements 27
-------
7-46
Texas 27
Water Quality Standards 28
Ground-Water Protection 28
Stormwater Management Requirements 29
7J SOLID WASTE MANAGEMENT 29
7.3.1 Federal Controls 30
The Resource Conservation and Recovery Act 30
Boiler and Industrial Furnace Rule 31
Comprehensive Environmental Response, Compensation, and Liability Act . . 32
7.3.2 State Controls 32
California 33
Michigan 36
Pennsylvania 38
Texas 40
-------
LIST OF EXHIBITS
Exhibit 7-1 New Source Performance Standards for Portland Cement Plants 7
Exhibit 7-2 Summary of State Air Pollution Controls 10
Exhibit 7-3 Summary of State Water Pollution Controls 21
Exhibit 7-4 Summary of State Solid Waste Management Controls 33
-------
CHAPTER EIGHT
ALTERNATIVE CKD MANAGEMENT PRACTICES AND POTENTIAL UTILIZATION
8.0 OVERVIEW
As discussed in Chapter 3, gross CKD is the dust collected at the air pollution control
device(s) associated with a kiln system. Gross CKD is generated as an inherent process residue
at all cement plants, though the ultimate fate of this material varies by facility. Exhibit 8-1
illustrates the potential management pathways for gross CKD. After collection, gross CKD is
either recycled back to the kiln system or removed from the kiln system as net CKD. Although a
number of plants recycle all gross CKD back to the kiln system, most plants remove a significant
quantity of CKD from the system. On average, 0.20 tons of gross CKD are generated per ton of
clinker produced, and 0.07 tons of net CKD are generated for the same amount of clinker (i.e.,
about 65 percent of gross CKD is recycled and about 35 percent is removed from the kiln).
Exhibit 8-1
Flow Chart of Gross CKD Management Pathways
-------
8-2
When CKD is removed from the kiln system, it can be treated for return to the kiln
system, beneficially utilized, or disposed. Net CKD represents a loss of resources when it is
removed from the manufacturing process and discarded because CKD is essentially derived from
the raw feed (which has been quarried, ground, and blended), and, to a lesser extent, the kiln
fuel(s). Additionally, the fuel value of the CKD is lost when the dust at elevated temperatures is
removed.1 Lastly, removing CKD from the system imposes handling, transportation, and
disposal costs. Hence, the first efficiency goal of the kiln operator should be to remove less
CKD from the kiln system when possible. Second, when CKD must be removed from the kiln
system, losses to disposal can potentially be minimized by using this material as a resource.
This chapter discusses the technologies that are available and under development to
minimize the quantity of CKD that must be removed from the kiln system and alternative
management practices and potential uses for the net CKD that is generated.
8.1 MINIMIZATION OF CKD REMOVAL FROM THE KILN SYSTEM
Conceptually, three general approaches can be used to minimize the removal of CKD
from the kiln system:
• Control of CKD generation rates;
• Direct return of CKD to the kiln; and
• Treatment and return of CKD to the kiln.
The first approach involves process controls to minimize dust generation. The other two
approaches address methods for returning dust to the kiln system once it is generated and
collected. These approaches are discussed below in more detail.
8.1.1 Control of CKD Generation Rates
One approach to minimizing the quantity of net CKD is to generate less gross CKD.
Based on the limited information available on this topic, however, there appear to be few
practical modifications that can significantly decrease the amount of CKD generated by a given
kiln. Nonetheless, there are three primary factors that can influence the gross CKD generation
rate within a kiln system. First, dust generation can be minimized by reducing gas turbulence in
the kiln and avoiding excessive flow velocities. This practice is, to EPA's knowledge, being
implemented to the extent possible as a basic process efficiency parameter. Second, the use of
chains near the cool end of the kiln helps to trap CKD before it is entrained into the kiln
exhaust.2 Most kilns are equipped with such cool-end chain sections. Third, the ash content of
fuels can vary, yielding differing amounts of particles that become a part of the CKD. For
example, liquid hazardous wastes will tend to have a lower ash content than coal.3 This issue is
not likely to drive fuel usage decisions at a kiln.
1 Based on EPA's sampling study, CKD was removed from the kiln system at a typical temperature exceeding 93
'C (200 *F).
2 Peray, K.E., 1986. The Rotary Cement Kin. Chemical Publishing Co., Inc. New York, New York. p. 108.
J Gossman, D., 1992. The Reuse of Petroleum and Petroleum Waste in Cement Kins. Environmental Progress.
Volume 11, Number 1. February, pp. 1-6.
-------
8-3
The amount or percentage by which gross CKD generation can be reduced by any of
these factors is not reported in the available literature. Based on the information reviewed, a
process-oriented approach to minimizing CKD generation rates appears to have a limited
potential impact in comparison to other approaches, such as increased recycling.
8.1.2 Direct Return of CKD to the Kiln
When gross CKD is generated, minimizing its removal from the kiln system involves
recycling gross CKD from its collection point(s) to some part of the manufacturing process.
According to the 1991 PCA Survey responses, kiln operators recycled 52 percent of the gross
CKD generated in 1990. For each kiln, however, the amount of dust that can be returned to the
kiln depends upon (1) the content of minor elements (alkalies4, sulfur, and chlorine) in the dust;
(2) the technology used to recycle the dust; and (3) the type of cement being produced. In
addition, the type of kiln system (wet, dry, or preheater/precalciner) will influence the type of
return system that can be employed.
Direct return of CKD to the kiln is the simplest recycling practice. Some operators may
opt for removal and disposal of CKD rather than installing return systems or monitoring quality.
Product specifications and local market demands dictate clinker alkali levels, and thus can
influence the quality of CKD that can be returned to the kiln system. Because clinker quality
can be reduced by the presence of alkalies and other constituents, in some cement markets, only
CKD that is within specified limits for these components can be directly returned to the kiln
system in significant amounts. In electrostatic precipitators (ESPs), CKD of acceptable quality is
generally only obtained from the initial ESP stages, while the CKD from the later ESP stages is
not of sufficient quality for direct return to the kiln because of higher alkali metal content.
As mentioned previously, the major factor limiting the direct recycling of dust to the
manufacturing process is its alkali level. The American Society for Testing and Materials
(ASTM) specifies a limit of 0.6 percent alkali in portland cement;5'6 cement with higher alkali
content is considered an inferior product and is not suitable for all uses because the alkalies can
react with some concrete aggregates and cause the concrete to crack.7 Similarly, chlorine can
react with alkalies to form alkali chlorides, which can also result in structurally-defective
concrete. Sulfur, in the form of sulfate, can reduce the structural quality of concrete as well.8-9
Quality specifications for a given product, which depend upon cement type and local market
conditions, will dictate these constituent concentrations in the clinker.
4 Alkalis refer to the alkali metals in group IA of the periodic table of the elements, including lithium, sodium,
potassium, rubidium, cesium, and francium. These are light, highly reactive metals. Of primary concern to cement
producers are sodium and potassium.
5 Wilson, R.D. and W.E. Anable, 1986. Removal of Alkalies From Portland CKD. U.S. Department of Interior,
Bureau of Mines Report of Investigations Number 9032. p. 2.
6 Kosmatka and Panarese, 1990. Design and Control of Concrete Mixtures. 13th Ed. Portland Cement Association,
SkokJe, Illinois, pp. 15-16.
7 Davis, T.A., et al, 1975. Disposal and Utilization of Waste Kin Dust From Cement Industry. Southern Research
Institute. May. p. 17.
8 Kosmatka and Panarese, 1990, op. cit.
9 Mehta, P.K., 1986. Concrete: Structure, Properties, and Materials, Prentice Hall, Englewood Cliffs, NJ.
-------
8-4
As discussed in Chapter 3, the input materials used to produce cement (raw feed and
fuel) influence the chemical composition of the CKD generated. At facilities where a significant
quantity of CKD is removed from the system, the use of alternative input materials might
improve CKD characteristics so that a larger portion of CKD could be directly returned to the
kiln. However, given the weights of these alternative input materials it may be impractical to
substitute materials because of high transportation costs.
Fuel inputs can significantly influence CKD chemical characteristics. For example,
burning low-sulfur coal instead of less expensive coal that is high in sulfur can yield a reduction
in CKD sulfur levels. Similarly, using hazardous waste as fuel can affect CKD alkali levels.10
Chapter 3 presents data showing that kilns burning hazardous waste recycle less CKD than kilns
not burning hazardous waste. This practice is also suggested in the literature. In one study,
burning hazardous wastes containing chlorine reportedly resulted in reduced recycling rates. In
this study, normal coal-fired operations at a dry kiln yielded 90 metric tons (100 tons) of net
CKD per month for disposal. However, when hazardous waste was co-fired, this figure increased
to 1,800 metric tons (2,000 tons) per month to limit chloride levels in the system.11
Raw feeds also influence CKD quality and recycling rates. The greatest raw feed
limitation for the industry appears to be excessive alkali levels in the limestone feedstock. Unlike
fuels, which are generally the only major inputs that are transported to a kiln from off-site,
limestone feed materials, which account for about 85 percent of the raw material consumption,
are almost always quarried on site. Because transportation costs can be prohibitive, the viable
options for alternative limestone raw feed inputs are limited for most facilities. Some facilities
may, however, find it possible to substitute raw materials, such as sand, shale, or clay. For
example, the Calaveras Cement Company facility in Tehachapi, California substitutes the locally-
available sand with low-alkali sand (sweet sand), which they purchase from an off-site source.
This low-alkali sand balances the high-alkali content of the limestone the facility quarries,
thereby enabling the facility to recycle 100 percent of the generated CKD.
Finally, process type appears to affect CKD recycling rates. Chapter 3 demonstrated
how, relative to units of clinker product, wet kilns recycle less CKD than dry long kilns, and dry
long kilns recycle less CKD than preheater or preheater/precalciner kilns. (The reasons for these
differences are not fully understood and analyses are continuing.)
After collection in air pollution control devices and removal of unacceptable CKD, the
acceptable portion of the CKD is conveyed back to the kiln system. CKD conveyance
mechanisms vary between facilities, but generally consist of augers, belts, positive pressure air
conveyors, and negative pressure air conveyors. CKD can be returned to the kiln system at three
general locations: CKD can be introduced at the flame (hot) end of the kiln, at the middle of
the kiln, or at the raw input (cool) end of the kiln (including blending and storage with the raw
mix before reaching the kiln). The equipment needed to return CKD to the kiln system can be
expensive, but these costs can be outweighed by the resulting savings on avoided resource losses
and CKD management costs.12 Return system installation costs can also fall within reasonable
10 Gossman, D., 1992, op. cit.
11 Engineering Science, 1987. Background Information Document For The Development of Regulations To Control
The Burning of Hazardous Wastes In Boilers and Industrial Furnaces. Vol. II: Industrial Furnaces. January, pp. 4-18.
12 Persona] communication with Hans Steuch, Director of Engineering, Ash Grove West, December 9, 1992.
-------
8-5
limits. For example, an Ash Grove Cement Company facility invested $100,000 to install a
system to return to the hot end of the kiln the 50 to 90 metric tons of CKD per day that had
previously been wasted.13
Return to Flame End
Insufflation involves the introduction of unaggregated CKD into the hot end of the kiln.
Dust is injected through or near the burner pipe into the kiln, where the CKD very rapidly
reaches reaction temperature. In general, the amount of dust returned through insufflation
represents about 15 percent of kiln feed. Although this return method is common, two primary
limitations exist. First, the amount of dust at a given kiln that can be returned through
insufflation is limited by the reduced flame temperature that it causes in the burning zone.14 A
second disadvantage of insufflation is the continuous resuspension of dust that it causes. This
results in a recirculating dust load that requires additional energy for CKD collection and
reheating.15
Return to Mid-Kiln
CKD can also be conveyed to a shroud in the middle of the kiln near the material inlet
to the calcining zone. Mid-kiln return of the dust is expensive, however, because the kiln must
be cut to accommodate the shroud.16 Scoops mounted on the kiln at this location pick up the
dust and drop it into the kiln through openings in the shell. Scoops tend to allow fugitive
emissions of dust because of sealing problems between the shroud and the kiln. Use of mid-kiln
scoops is not a common dust return method.17
Return with Raw Feed
CKD can be returned to the kiln system at the input end (also referred to as the exhaust,
or cool end) by combining it with the raw feed. The return process differs between dry and wet
kilns. In dry process kilns, the dust is conveyed to kiln feed silos or returned directly to the kiln
feed system where it is blended with the raw feed to obtain a uniform mix. In wet process kilns,
CKD must be returned in a different manner to avoid hardening and thickening of the feed
slurry. There are several solutions to this problem:
• Dry CKD can be added to the feed slurry where the slurry enters the kiln;
• A separate CKD slurry can be formed and pumped directly into the kiln; or
15 Ibid.
14 Steuch, Hans E., 1992. Review of Dust Return Systems. Paper presented at the Portland Cement Association
Seminar on Emerging Technologies for Kiln Dust Management. March 4. Chicago, Illinois, p. 2.
15 Davis, T.A., et al, 1975, op. cit.
16 Davis, T.A., et a!., 1975, op. cit.
17 Steuch, H.E., 1992, op. cit.
-------
8-6
• Chemical additives, such as molasses or lignosulfates, can be added to retard the
setting of CKD when it is hydrated, and thereby improving its flow characteristics.18
The addition of water and/or chemicals, however, limits the amount of CKD that can be
returned," presumably because kiln efficiency would be reduced. In the case of water addition,
too much water would require excessive energy for dehydration. In the case of chemicals, the
introduction of too many additives could yield recirculating loads of unusable CKD that negate
the benefits of treating the material in this manner. Lastly, this alternative can be costly.
8.13 Treatment and Return of CKD to the Kiln
CKD that contains alkalies or possesses other undesirable characteristics may be treated
so that it can be returned to the kiln system. Although few treatment processes have been
commercially adopted on a wide scale, research into CKD treatment and recycling has yielded a
number of promising technologies. These include pelletizing, leaching with water, leaching with
potassium chloride solution, alkali volatilization, recovery scrubbing, and fluid bed dust recovery.
Pelletizing
Pelletizing is generally a physical transformation of CKD that makes it more manageable
in certain situations. In addition to pelletizing for return to the kiln system, much CKD is also
pelletized prior to disposal in waste management units. This technology has been in existence for
over 15 years, and can be used to return CKD to the flame or the feed ends of the kiln.20 In
contrast to insufflation of unaggregated CKD, the pelletizing process gives the CKD the
necessary strength to withstand the forces of being fired into the flame without resuspending
large quantities of paniculate matter. No binder is necessary, and no deterioration is found
when the pellets are introduced through the chain section.21 Pelletizing also avoids the need for
any major modification of flame characteristics. The pellets form clinker, which is chemically
indistinguishable from normal clinker.22
Pelletizing may also involve adding a binder and/or raw feed to the CKD. As discussed
below, pelletized CKD can be used with other treatment technologies to improve its handling
characteristics. Technologies discussed below in which pelletized CKD has been used include
alkali volatilization and fluid bed dust recovery.
Leaching with Water
By leaching alkali salts out of CKD using water, the amount of dust that can be recycled
to the kiln can be increased. In the leaching process, dust is mixed with water in a tank or pug
mill to produce a slurry of about 10 to 20 percent solids. The slurry is thickened in a clarifier
w Rates of addition are not provided in the reference source.
19 Steuch, H.E., 1992, op. cit.
20 Sell, Nancy J. and Fritz A. Fischbach. 1978. Pelletinng Waste CKD for More Efficient Recycling. Industrial and
Engineering Chemistry, Process Design and Development. Volume 17, Number 4. October, pp. 468-473.
21 Ibid.
nlbld.
-------
8-7
where solids settle to the bottom and excess water overflows the top. The underflow from the
clarifier contains about 50 percent solids and is returned to the kiln to produce clinker. In wet
process kilns, the underflow slurry is either mixed with the feed slurry or pumped into the kiln
through a pipe parallel to the kiln feed.23 In dry process kilns, the underflow must be filtered
and dried before it can be blended with the raw mix or its components. Leaching with hot water
has been found to remove more alkali than leaching at ambient conditions.24-25
The alkaline wastewater from the leaching process must be treated before it can be
discharged because it has a high pH and high concentrations of dissolved and suspended solids.
To solve this problem, electrodialysis has been investigated for use after leaching to remove
alkali salts from the leachate and to recover salts that may be marketable as fertilizers by
evaporation and fractional crystallization.26 An electrodialysis stack creates an internal electric
potential that forces ions from the leachate through semipermeable membranes into a
concentrated brine. Water that enters the brine by osmosis carries with it the concentrated salts.
The remaining partially desalted water is reused to leach alkali from CKD; no wastewater is
discharged.27 The concentrated brine from the electrodialysis stack contains about 20 percent
dissolved solids (mainly potassium, sodium, carbonate, and sulfate). Depending on the ratio of
potassium to sodium, the brine may be suitable as a liquid fertilizer, or potassium can be further
purified and concentrated by fractional crystallization and evaporation.28 In some cases, high
calcium concentrations can interfere with electrodialysis. To reduce excessive calcium
concentrations, wastewater can be run through two carbonators in which the calcium will
combine with carbon dioxide and precipitate out as CaCO3 before the wastewater is sent through
the electrodialysis stack.29 The extent to which these technologies are used commercially, or
their associated costs, were not found in the available literature.
Generation of wastewater is not a precondition to the use of alkali leaching, as
demonstrated by the Ash Grove Cement, Inkom facility. In a modified version of alkali leaching,
the operators of this facility in Idaho concentrate the alkalies in the leaching water to produce a
potassium sulfate solution. Ash Grove has been leaching alkalies from CKD and selling
potassium sulfate to farmers since the 1950s. In this process, CKD is moved from the air
pollution control units to a concrete holding tank using a dust elevator. Water is added to this
tank, where the alkalies are leached from the CKD and subsequently concentrated through solar
evaporation in two holding ponds; the bottom sludges from the holding tank are pumped to the
raw mix slurry tank. The facility operator performs analysis on the sludge returned to the kiln
feed to account for input variations. From the evaporation ponds, the facility sells the potassium
sulfate for approximately $2.80 per metric ton to a local broker who purchases about 9,000
metric tons (10,000 tons) per year of the product. Local potato farmers spray-apply this solution
23 Davis, T.A., et al, 1975, op. oil.
24 Ibid.
25 Persona] Communication with Henry Voldbaek, Ash Grove, Inkom, Idaho, December 9, 1992.
26 Davis, T.A., et al., 1975, op. cit.
"Ibid.
Davis, T.A., et al., 1975, op. cit.
-------
8-8
to their fields. According to Ash Grove Cement, this leaching and return process results in 100
percent recycling of CKD.30
In addition to the Ash Grove, Inkom facility, only one other facility is known to treat
CKD with water leaching -- the Holnam Plant in Dundee, Michigan. The operators of this
facility use a water circulation system to leach alkalies from CKD. In this process, CKD is blown
down to a thickening tank for clarification (settling). The slurry contained in the underflow is
recycled back to the wet process kiln. The clarified portion of the tank contents is pH-
neutralized using waste acid, and directed to a holding tank. From this tank, most of the water is
recycled for process water, while a smaller portion is discharged through a NPDES-permitted
outfall. This technology produces no saleable byproducts. According to Holnam, the start-up
costs for implementing this technology are extremely low.31
The economic costs and benefits of CKD leaching with water are described more fully in
Chapter 9.
Leaching with a Potassium Chloride Solution
Another procedure to increase the quality, and hence, quantity, of CKD returned to the
kiln system involves leaching alkali out of the dust with a potassium chloride solution. In this
process, CKD is mixed with a hot KC1 solution that produces a slurry high in pH, dissolved
solids, and suspended solids. The slurry is then treated with an oily hydrocarbon and a long-
chain fatty acid to flocculate the slurry solids for separation.32 After separation of the solids,
the remaining aqueous phase is cooled to induce KC1 crystallization. Optimal leaching
conditions were found at 70 to 80°C (158 to 176°F).33 No recent discussion of this technology
has been found, suggesting that it has not been applied commercially in the U.S.
Alkali Volatilization
Alkali volatilization represents another method to recover alkali from the surface of CKD
particles. This technique generally involves subjecting the CKD to a high temperature flame,
then condensing the resulting alkali vapors from the hot gases onto a cooler surface. (Although
volatilization occurs during insufflation under normal kiln operating conditions, separation of the
alkali is not accomplished because the alkali recondenses onto the CKD in the kiln rather than
being removed.)34
Sintering35 is the primary method used to achieve alkali volatilization from CKD. In
one pilot study, pelletized samples were sintered for one to two hours at a temperature of 1,100
30 Personal communication with Henry Voldbaek, Ash Grove, Inkom, Idaho, December 9, 1992.
" Personal communication with Harry Hackett, Holnam, Dundee, Michigan, January 28, 1993.
52 McCord, A.T., 1977. CKD Treatment - By Leaching with Hot Potassium Chloride Solution to Recover Alkali
Values. U.S. Patent No. 4031184.
53 Ibid.
M Davis, T.A., et al, 1975, op. cit.
35 Sintering is the process of heating a material such that it becomes a coherent mass without melting.
-------
8-9
to 1,300°C. The study also examined melting pelletized samples at temperatures of 1,600 to
1,700°C for 30 minutes. Researchers noted a first order decomposition of the alkalies in heated
CKD through this process, and concluded that the most likely mechanism for the removal of
alkalies is the formation of volatile potassium and/or sodium. After thermal treatment, both
sintered and melted samples met ASTM standard cement criteria levels for A12O3, CaO, Fe2O3,
KjO, and MgO.36 X-ray diffraction analysis and compression tests also indicated that cement
made from sintered or molten CKD would be within ASTM alkali standards. In this study, the
alkali content found in sintered or melted CKD was less than 0.094 percent, almost an order of
.magnitude below the ASTM standard of 0.6 percent. Researchers concluded that temperatures
above 1,300°C, and in the presence of carbon (i.e., a reducing atmosphere) are sufficient for
effective alkali volatilization from CKD.37 The current status of this technology has not been
determined.
In a similar process, CKD was mixed with fly ash containing alumina, and sintered for the
purpose of recovering the alumina.38 This process involves combining fly ash with CKD and
soda ash, pelletizing the mixture, and sintering at temperatures between 1,200 and 1,300°C. The
sintered pellets are leached with a dilute solution of soda ash to recover the alumina. The waste
from this recovery process is dicalcium silicate, which shows promise as a raw material in the
manufacture of low-alumina and conventional portland cement.39 The current status of this
technology has also not been determined.
Recovery Scrubbing
Another CKD treatment technology is the flue gas desulfurization (FGD) process, or
recovery scrubber. This process enables all CKD to be recycled as kiln feed by removing
alkalies, chlorides, and sulfates from the dust. The recovery scrubber creates a recycling system
that produces potassium fertilizer as well as reusable feed, and reportedly discharges only clean
air and distilled water.40-41-42
Exhibit 8-2 illustrates the various stages of the recovery scrubber process. The first step
in the process requires saturating a stream of water with carbon dioxide by introducing it to kiln
exhaust gases from the heat exchange process (located at right of the diagram). The CKD input
36 Wilson, R.D. and W.E. Anable, 1986, op. oil.
"Ibid.
M Burnet, G., 1987. Alumina Recovery from Fly Ash by the Lime-Soda Sinter Process. Proceedings of the
International Symposium on Ash. February 2. Pretoria, South Africa.
" Ibid.
40 Morrison, G.L., 1990. CKD and Flue Gas Scrubbing: Tlie Demonstration at Dragon Products Company.
July. p. 7.
41 Morrison, G.L., 1991. Flue Gas Scrubbing and Waste Elimination - An Application of the Recovery Scrubber.
Paper presented at The World Coal Institute Conference and Exhibition on Coal in the Environment. April 3.
London, United Kingdom.
n Anonymous, 1991. SO, Technology Pays for Itself and Then Some. Coal and Synfuels Technology. January 14.
p. 1.
-------
8-10
is then m.ixed with the carbon dioxide-saturated water in the mix tank (on the upper left) to form
a slurry. The use
-------
8-11
Exhibit 8-2
Process Flow Diagram of Recovery Scrubber
-------
8-12
of carbon dioxide-saturated water helps to prevent the slurry from setting or solidifying. The
slurry is then introduced to the reaction tank of the scrubber. In the reaction tank, the calcium
in the slurry reacts with CO2 from the heat exchange process to form insoluble CaCO3. At the
same time, alkali metal (e.g., sodium and potassium) hydroxides react to form the corresponding
sulfates.43
From the reaction tank, dissolved solids and suspended solids in the reacted slurry are
separated into two process streams via a series of settling tanks (bottom left and right) combined
with a dilution tank (bottom center) where the slurry is mixed with an equal quantity of water to
reduce the alkali content.44 The dissolved solids are crystallized through evaporation, using
waste heat from the kiln exhaust. The suspended solids (minus soluble alkalies, chlorides, and
sulfates), are returned to the cement kiln raw material feed system for reuse in the production of
cement; the CaCO3 that forms from the calcium and CO2 yields "new" limestone. The sulfur
from the sulfur dioxide and calcium-sulfur compounds in the exhaust gas combines with the
potassium in the CKD to make potassium sulfate, which can be used as a fertilizer.45
The recovery scrubber reportedly removes 90 to 98 percent of the sulfur dioxide in the
flue gas and improves the kiln's paniculate CKD capture efficiency.46-47-48 In addition to
increased cement production, the process yields two marketable by-products: potassium sulfate
(a fertilizer) and distilled water. According to all available sources, no process waste flows (i.e.,
waste material in either liquid or solid form) result from the process.49-50-51 Although water is
added to this process, there is no liquid effluent because all the added water is evaporated into.
the flue gas.52
According to Passamaquoddy Technology, which is marketing the system, the process can
be widely used with modifications in both wet and dry process cement kilns. Site-specific factors
43 Anonymous, 1991. Chloride-Free Potash Fertilizer from Waste SO2 and CKD. Phosphorous and Potassium. July-
August, p. 48.
44 Morrison, G.L., 1990. Exhaust Gas Scrubbing and Waste Elimination, An Application of the Recovery Scrubber to
a Cement Kin. Paper presented at the 1990 SO, Control Symposium. May 8. New Orleans, Louisiana.
45 Anonymous, 1991. "Home-Grown" Scrubber Protects the Environment of an Indian Reservation. Sulphur. July-
August, p. 24.
44 Anonymous, 1991. SO, Technology Pays for Itself and Then Some. Coal and Synfuels Technology. January 14.
p. 1.
47 Anonymous, 1991. Recovery Scrubber Meets Design Goals. Coal and Synfuels Technology. April 22. p. 5.
48 Morrison, G.L., 1991, op, cit.
49 Morrison, G.L., 1992. CKD Management Using a Recovery Scrubber: Operation and Economics. Paper
presented at the Portland Cement Association Seminar on Emerging Technologies for Kiln Dust Management. March
4. Chicago, Elinois.
50 Anonymous, 1991. Recovery Scrubber Meets Design Goals. Coal and Synfuels Technology. April 22. p. 5.
51 Personal communication with Garrett Morrison, Passamaquoddy Technology, November 24, 1992.
52 Anonymous, 1991. Chloride-Free Potash Fertilizer from Waste SO2 and CKD. Phosphorous and Potassium. July-
August, p. 48.
-------
8-13
requiring design modifications would include kiln characteristics, raw feed characteristics, cement
product specifications, and fuel type.53-54 For example, if these factors yield CKD high in
sodium, a double potassium crystallizer would be required to make the potassium sulfate by-
product marketable.55
Currently, the Dragon Products Company in Thomaston, Maine, owns the only
commercial-scale recovery scrubber system in operation. This $18 million56 trial plant was co-
funded by the U.S. Department of Energy's Clean Coal Technology program and began
operation in 1990.57 Although there are no other installations, Passamaquoddy Technology has
conducted about 35 plant-specific, engineering evaluations to determine if application of the
technology at specific plants is justified and if so, in what configuration.58 Although it has
evoked considerable interest in the United States and abroad, the extent to which this technology
will be adopted across the industry is unclear.
With respect to start-up and operating costs, the recovery scrubber system may be more
capital-intensive than a traditional system, but the savings in avoided resource losses and CKD
disposal costs reportedly yield a net savings. Moreover, the system requires no net increase in
personnel. Modifications can, however, significantly increase the cost. For example, a double
crystallizer, if required, would add about $1 million to the start-up costs. A continuous metals
extractor for the CKD can also be installed for an additional $200,000 to $300,000.59 The
recovery scrubber system also requires higher energy inputs than traditional systems.
The recovery scrubber reportedly confers cost savings due to avoided resource losses and
CKD disposal costs. Moreover, the process allows the facility increased flexibility in using high
sulfur fuel (which tends to be less costly than low sulfur fuel) in the kiln. Dragon Products
expects payback from installation of the recovery scrubber within two to three years. After the
payback period, the recovery scrubber at the Dragon facility is expected to generate process
savings, avoided CKD disposal costs, and additional income from the sale of by-
products.60'61-62 According to Dragon Products, the potassium sulfate (potash) production
55 Morrison, G.L., 1990, op. tit.
54 A discussion of specific design and operating limitations of this process has not been located in any of the
available literature.
55 Personal communication with Garrett Morrison, Passamaquoddy Technology, November 24, 1992.
56 Persona] communication with Garrett Morrison, Passamaquoddy Technology, July 2, 1993.
57 Anonymous, 1991. SO2 Tecfinology Pays for Itself and Then Some. Coal and Synfuels Technology. January 14.
p.l.
" Personal communications with Garrett Morrison, Passamaquoddy Technology, November 24, 1992 and July 2,
1993.
"Ibid.
60 Morrison, G.L., 1992, op. tit.
61 Anonymous, 1991. SO, Technology Pays for Itself and Then Some. Coal and Synfuels Technology. January 14.
p.l.
-------
8-14
rate is expected to range from 7 to 9.1 metric tons per day; this material may be worth as much
as $220 per metric ton.63 According to independent trade and government sources, if the
product is of sufficient quality, it may be worth $198 to $355 per metric ton.64'65
Dragon Products recycles not only all the CKD it generates, but also consumes CKD
from its stockpile of dust generated during previous years. For possibly as many as 100 years,
Dragon Products had been disposing its CKD in an on-site pile at a rate, that until recently,
averaged 230 metric tons per day. Material from this pile is now being mined and reintroduced
to the kiln at a rate of approximately 90 to 270 metric tons per day, so that Dragon Products
expects to eliminate its existing CKD waste pile in the near future.66'67
In addition to recycling their own dust, it may be possible that cement plants with
recovery scrubbers could consume wastes generated by other industries, such as paper mills and
biomass-burning power plants, and charge a tipping fee for the service. Dragon Products will
reportedly be accepting about 45,000 metric tons per year of wood and coal ashes, at a tipping
fee of $33 per metric ton.68 Other wastes that reportedly can be accommodated include acids
or bases and ash from other burners.69-70
The economic costs and benefits of the recovery scrubber technology are examined in
further detail in Chapter 9.
Fluid Bed Dust Recovery
The fluid bed dust recovery process, or Fuller process, thermally treats the CKD (on
either a gross or net basis). Although this process does not return CKD to the kiln system, it is
functionally similar to such return technologies because it yields a usable cement clinker product
rather than treated CKD. The fluid bed process is designed to accept all CKD generated from a
kiln, pelletize it, and calcine it into clinker on a fluid bed instead of in a typical rotary kiln.71
0 Ibid.
63 Personal communication with Garrett Morrison, Passamaquoddy Technology, July 2, 1993.
64 Chemical Prices Weekly, 1992. Potassium Sulfate. Ending December 25. p. 28.
65 Bureau of Mines, 1989. Minerals Yearbook, Vol. 1, Metals and Minerals, Potash (James P. Searls auth.).
Department of the Interior, pp. 801-805.
66 Anonymous, 1991. "Home-Grown" Scrubber Protects the Environment of an Indian Reservation. Sulphur. July-
August, p. 24.
*' Morrison, G.L., 1990, op. cit.
68 Personal communication with Garrett Morrison, Passamaquoddy Technology, July 2, 1993.
69 Anonymous, 1991. "Home-Grown" Scrubber Protects the Environment of an Indian Reservation. Sulphur. July-
August, p. 24.
10 Morrison, G.L., 1990, op. cit.
71 Cohen, S.M., 1992. Fluid Bed Dust Recovery. Paper presented at the Portland Cement Association Seminar on
Emerging Technologies for Kiln Dust Management March 4. Chicago, Illinois.
-------
8-15
The unusable constituents are concentrated into a waste dust that represents 10 percent of the
original input dust volume.
During pelletization, CKD enters a pug mill, where the free lime in the CKD reacts with
introduced moisture, and the CKD is pelletized. Generally CKD requires only water for
pelletization. However, in some cases, a binder (e.g., five percent portland cement) may be
required. As illustrated in Exhibit 8-3, the Fuller process operates on a gravity system, in which
dried CKD pellets are gravity fed into the reaction zone of the fluid bed reactor for a retention
time of 1-2 hours. The pelletized CKD reacts to form clinker and moves along the fluid bed
and drops into a pipeline for cooling and subsequent grinding.72
In the fluid bed, the combustion and calcining gases flow upward and are directed away
from the raw feed. The gases released from the clinkering process exit the fluid bed without
contacting the incoming raw material. In contrast, the exhaust gases in conventional cement
kilns contact incoming raw material and reprecipitate alkalies onto the raw material particles.
The fluid bed process reportedly removes as much as 90 percent of the K2O, 70 percent of the
Na2O, 90 percent of the SO3, and almost 100 percent of the chlorides contained in the original
feed dust. These compounds are concentrated in the unusable dust fraction (i.e., 10 percent of
the original CKD input) that is removed from the fluid bed system by directing exhaust through
a baghouse.
The fluid bed process yields clinker output at 60 percent of the CKD input (i.e., 0.6 tons
of clinker is generated per ton of CKD input). Waste dust is generated at approximately 10
percent of the CKD input. The fate of the remaining 30 percent of the original CKD input has
not been addressed. Presumably, it is either returned to the system or lost as gaseous emissions
or both. These generation rates are comparable to those of kilns using virgin raw materials.
Because the unusable dust contains approximately 40 percent potassium, it may be of value as a
fertilizer feedstock. Further evaluation is needed, however, to assess the effects of other
constituents that may also be present in the dust. The presence of heavy metals presents one
potential concern, although initial tests indicate that the unusable dust from the Fuller process is
no higher in heavy metals than dust from conventional rotary kiln cement processes.73
A wide range of solid fuel sources (e.g., coal, petroleum coke) can be added to CKD
during pelletization to provide up to 90 percent of the process energy requirement. A clinker
bed moves above a grid plate through which air is driven. This clinker bed is held at 1,300°C by
injection of oil or gas directly into the moving bed of material. The added fuel supplies the
remaining 10 to 15 percent of the heat requirement after the fuel in the pellets is burned.
Projected fuel consumption for a commercial level plant is 1.15 million kilocalories (Kcals) per
metric ton (4.14 million Btus per ton) of clinker. According to Fuller, this figure is-competitive
with both wet and long dry commercial systems but not with preheater or flash calcining systems,
if compared to normal clinker production. This claim is verified by the data in Exhibit 2-7 of this
report. As shown in Exhibit 2-7, energy consumption (Kcal/Kg of output) is 1,529 to 1,668 for
wet process kilns, 1,251 to 1,390 for dry process kilns, 945.2 for semi-dry kilns, and 750.6 to 889.6
for preheater kilns. Specific information is not available for precalciner kilns, but the energy
consumption of such kilns is believed to be similar to that of preheater kilns. The Fuller process
72 Ibid.
73 Personal communication with Sidney Cohen, Fuller Company, July 20, 1993.
-------
8-16
minimizes energy loss by reclaiming the heat contained in exhaust gases with a heat recovery
system.74-75
74 ibid.
75 Personal communication with Sidney Cohen, Fuller Company, November 13, 1992.
-------
8-17
Exhibit 8-3
Process Flow Diagram of Fluid Bed Dust Recovery Process
-------
8-18
This technology has only been used to date on a pilot scale. Although reportedly ready
for market, this system is still undergoing extensive research. For example, a uniform set of
operational parameters has not yet been developed. Tests indicate that the operational
parameters are facility-specific, and that system designs would have to be individually tailored.
Many aspects of the reaction dynamics are also not fully understood. The end product and by-
products of the reactor are being analyzed to confirm cement (clinker) quality and to determine
the dynamics of the recovery process as they relate specifically to metallic and other components.
Material and heat balances are being developed to establish the overall energy consumption for
the process, as well as to provide data for economic evaluations.76
In pilot scale tests, this system has successfully produced a clinker product and a highly
concentrated alkali dust product using CKD from 10 different cement plants. The fluid bed
recovery process can provide the benefit of generating a usable product, and decrease by up to
90 percent the amount of CKD that is regularly disposed. The remaining 10 percent of the
received CKD becomes a potentially marketable by-product that is high in potassium.
Alternatively, this material can be leached to remove alkalies, producing solids that can be
returned to the process and a highly concentrated solution that can be utilized for chemical
production.77 As mentioned above, the economic viability of the process depends on each
facility, and is a function of the quantity of dust generated, kiln capacity, and dust management
costs. Therefore, a unit cost for the process is not realistic.78 According to the manufacturer,
estimated capital investment for a complete 270 metric ton per day (300 TPD)79 installed system
(as CKD input, to produce up to 160 metric tons clinker) would be $9 to $10 million,80 with an
estimated payback period of six years.81
8.2 BENEFICIAL USE OF REMOVED CKD
It is likely that even with advances in recycling technologies, some CKD will need to be
removed from kiln systems. Because resources are lost when CKD is permanently disposed, and
because disposal practices can be burdensome, finding alternative uses for waste CKD can help
facilities avoid disposal costs and even generate additional revenue. CKD has been used
beneficially for at least 15 years, and interest in uses for CKD as a valuable resource appears to
be growing. According to responses from the 1991 PCA Survey and §3007 requests, 779,916
metric tons (859,709 tons) of CKD were used beneficially in 1990, or 5.4 percent of the gross
CKD generated in 1990, and about 16 percent of the net CKD for that year. Of this total, about
71 percent (670,000 metric tons) was used for waste stabilization, 12 percent (111,000 metric
tons) for soil amendment, 5.6 percent (53,000 metric tons) as liming agent, nearly three percent
(25,000 metric tons) as materials additive, about one percent (11,000 metric tons) as road base,
and eight percent (76,000 metric tons) for other uses.
76 Cohen, S.M., 1992, op. cit.
77 Persona] communication with Sidney Cohen, Fuller Company, July 20, 1993.
71 Personal communication with Sidney Cohen, Fuller Company, November 13, 1992.
79 An installation of this size would be adequate to treat total net dust generated at most plants in the U.S.
w Cohen, S.M., 1992, op. cit.
81 Personal communication with Sidney Cohen, Fuller Company, November 13, 1992.
-------
8-19
The ASTM suggests that CKD may be useful in a variety of applications, including
construction, stabilization, waste treatment, and agriculture. Due to the variability in dust
composition, however, ASTM advises that use of CKD should be undertaken only after the
material's characteristics have been properly evaluated with respect to the intended application.
ASTM also recommends frequent performance testing until the degree of variability has been
established.82
Currently, CKD is used beneficially for sludge, waste, and soil stabilization, land
reclamation, waste remediation, acid neutralization, agricultural applications, such as fertilizer
and lime substitution, and construction applications. These CKD uses, however, appear to
represent only a portion of potential beneficial applications for this material. The manner and
extent of CKD adaptation for beneficial applications is in constant flux as research and
development of CKD use continue to grow.
8.2.1 Stabilization of Sludges, Wastes, and Contaminated Soils
CKD has been used as an agent to solidify and stabilize waste materials and
contaminated soils since at least 1982. According to respondents to the 1991 PCA Survey and
§3007 requests, 70.8 percent of the CKD that was sold or given away was used for waste
stabilization. CKD has been used with reported success on sewage sludge, waste oil sludge, and
miscellaneous other wastes. The elevated pH of CKD helps to neutralize the acid conditions and
decrease the mobility of heavy metals in these materials. CKD can also help to dewater
contaminated materials and thereby increase weight-bearing capacity and possibly reduce the
threat of leachate migration.
Sewage Sludge
Economical and effective treatment technologies for municipal sewage sludge have been
sought for many years. Without treatment, sludge may contain unwanted microbial pathogens
and it may not be in a solid form conducive to handling. Treating the sludge with CKD may
improve its physical, chemical, and biological characteristics. For example, trace metals in the
sludge are immobilized by precipitation and coprecipitation as carbonates, oxides, hydroxides,
phosphates, and sulfates.83 A considerable amount of research has been conducted on the use
of CKD as
a medium for dewatering and stabilizing raw or digested sewage treatment sludges.84-85'86'87'88
42 ASTM, 1991. Standard Guide for Commercial Use of Lime Kin Dusts and Portland CKDs. 1990 Annual Book of
American Society for Testing and Materials Standards. Volume 11.04. Method Number D5050-90. pp. 172-174.
a Based on notes developed by ICF Incorporated during the 5th Annual International Conference on Alkaline
Pasteurization and Stabilization, Somerset, New Jersey, May 18-19, 1992.
84 Anonymous, 1985. Mobile Hazardous Waste Treatment Services • What's Available? Hazardous Waste
Consultant. September-October.
45 Burnham, J.C., el al, 1990. CKD Stabilization of Municipal Wastewater Sludge. Annual International Solid
Waste Exposition "Vancouver 90." August. Canada.
86 Kovacik, T.L., 1988. Sludge, Kin Dust Make Fertilizer. Water Engineering Management. December.
87 Metry, A.A., et al., 1985. A Cost-Effective Approach For Stabilization and Closure of an Organic Waste Sitperfund
Site. 31 Annual Meeting of the Institute of Environmental Sciences. April 29. Las Vegas, Nevada.
-------
8-20
Presently, CKD is being used commercially to stabilize municipal sewage sludge by at
least two companies: (1) Keystone Cement Co., which markets CKD under the name StableSorb
as a sewage sludge dewatering agent; and (2) National N-Viro Energy Systems (N-Viro), which
markets a CKD-stabilized sewage sludge as N-Viro Soil.
The treated sludges have been used as landfill cover, structural fill material, dike
construction material, and for agricultural purposes.89-90-91 Agricultural application of sewage
sludge has also been a common method of sludge management. Sludge contains a number of
nutrients beneficial to plants, including nitrogen, phosphorus, sulfur, calcium, potassium,
magnesium, and a host of nutrients needed in small quantities.92 Because, however, these
nutrients can occur within a wide range of concentrations, sludge used for agricultural purposes
may require nutrient supplements.
In 1992, Keystone sold over 38,000 metric tons (42,000 tons) of StableSorb, a
considerably lower figure than its record sales of 74,000 metric tons in 1990, but greater than its
1991 total of 25,000 metric tons. The anomalously high figure for 1990 was apparently the result
of an unusually large project that used StableSorb during that year.93 The low price of CKD
compared to its substitutes makes it highly competitive in the market. Keystone sells the dust for
$10 per metric ton ($9 per ton) F.O.B. (freight on board)94. Transportation costs, however, can
add up to $22 per metric ton ($20 per ton) to the sale price of the dust. Keystone supplied CKD
for two years to a utility in New Jersey for use in waste stabilization. Prior to the use of CKD,
the utility used fly ash for this purpose. The utility combined the dust and waste in a slurry tank,
in which the waste was thereby stabilized; the stabilized waste was used as a landfill cover. They
also blended CKD with dry wastes to form a landfill filler. Some buyers have been concerned
about potential dust quality impacts when Keystone burns hazardous waste as a kiln fuel.
According to Keystone, these concerns have been allayed through the use of constituent test
results, though Keystone is reportedly not selling StableSorb pending EPA's regulatory
determination for CKD. For specialized requests, Keystone will modify the CKD by blending it
with cement to give the product more strength. For such a treatment, the price increases by the
" Zier, R.E. and E. Wood, 1991. Sludge: Sludge Solutions. Waste Information Digests. May.
K Keystone Cement Company, date unknown. Stablesorb: A Coproduct of Cement Manufacturing With a Variety of
Uses. Product Brochure.
90 Bumham, J.C., 1988. CKD/Lime Treatment or Municipal Sludge Cake, Alternative Methods For Microbial and
Odor Control. Paper from Proceedings of National Conference on Municipal Sewage Treatment Plant Sludge
Management. June 27-29. Palm Beach, Florida.
91 Persona] communication with J. Patrick Nicholson, N-Viro Soil, December 7, 1992.
92 Kelley, W.D., D.C Martens, R.B. Reneau, Jr., and T.W. Simpson, 1984. Agricultural Use of Sewage Sludge: A
Literature Review. Bulletin 143. Virginia Water Resources Research Center, Virginia Polytechnic Institute and State
University, Blacksburg, Virginia. December, p. 38.
93 Personal communication with Doug Glasford and Bill Fischer, Keystone Cement, November 24, 1992.
94 The F.O.B. price of a product is the price that would be charged if the product were to be picked up from the
shipping dock; it excludes the cost of loading goods aboard a carrier, transportation costs, and all other costs beyond
the port of export.
-------
8-21
percentage of added cement multiplied by $55 per metric ton (e.g., a 10 percent blending adds
$5.50 per metric ton to the price).95
N-Viro Soil is used with or instead of lime to disinfect and deodorize municipal sewage
sludge, and according to the company, it provides a safe and socially-acceptable solution for
treating the sludge. N-Viro Soil is produced by combining CKD and municipal sludge through a
patented process called "Advanced Alkaline Stabilization with Subsequent Accelerated Drying."
This N-Viro technology uses a combination of microbiological stresses to kill pathogens and
stabilize the sludge to produce what is described as "a soil-like product." CKD's high alkali and
exothermic properties produce a pH of about 12 and generate temperatures between 52 and
62°C when mixed with the moisture contained in sewage sludge.96
N-Viro Soil contains between 35 and 75 percent CKD by weight. The alkali and CaO
content of CKD reportedly contribute to the characteristics of the product and make it a suitable
agricultural lime substitute. The large surface area and low moisture content of fine-grained
CKD particles provide odor control and accelerate drying. When combined with sludge, CKD
reportedly dilutes trace metal concentrations and reduces the solubility of trace metals. Further,
the heat produced from the hydrolysis reaction between CKD and the sludge moisture, combined
with elevated pH levels, apparently kills all pathogens in the sludge.97 In addition, this mixture
produces an artificial soil that can be used as a cover material for landfills and as an agricultural
lime substitute for soils.98 CKD also contributes most of the base elements (i.e., Ca, Mg, K,
Na) in N-Viro Soil that make the product a useful soil amendment.99 The product reportedly
can be stored for long periods of time without deterioration.100'101 One of the primary
drawbacks of N-Viro Soil is that a relatively large quantity of kiln dust is required to treat a
given unit of sludge, meaning that significant quantities of CKD must be transported and
handled to treat a given quantity of sludge.102
Regulations that govern the quantity of metals that can be applied to land in sludge may
limit the application rate of CKD-stabilized sludge (since sludge itself has high metals
concentrations);103 N-Viro Soil ostensibly meets these requirements.104'105 In addition, 40
95 Ibid.
96 N-Viro Energy Systems, 1991. Promotional Bulletin - N-Viro Soil.
97 Personal communication with Robert Bastion, Office of Wastewater Enforcement and Compliance, EPA,
November 1992.
* Ibid.
99 N-Viro Energy Systems, 1991, op. cit.
100 In tests, N-Viro Soil has apparently been stored for over 500 days. (PR Newswire, September 21, 1988).
101 Kovacik, T.L., 1987. Successful Recycling for Sludge and Solid Waste. BioCycle Southeast Conference.
November 4. Orlando, Florida.
102 Persona] communication with Robert Bastian, Office of Wastewater Enforcement and Compliance, EPA,
November 1992.
103 Ibid.
-------
8-22
CFR Part 503, promul-gated on November 27,1992, has established permitting regulations with
respect to sludge use and disposal practices.106 According to N-Viro, their purchases of CKD
must meet specifications for metals levels set by EPA's clean sludge rule as outlined below:
Constituent
Arsenic
Cadmium
Chromium
Copper
Lead
Molybdenum
Mercury
Nickel
Selenium
Zinc
Concentration Limit
(ppm)
<41
<39
<1200
<1500
<300
<18
<17
<490
<36
<2800
The greatest barrier to CKD use is likely to be the lack of specifications for metals contents and
products of incomplete combustion (PICs). N-Viro reportedly does not purchase CKD from
plants that burn hazardous waste. However, this policy is followed more because of public
perception than for technical reasons.107
The N-Viro process appears to be less costly than existing methods of sludge treatment.
In 1988, the city of Toledo, Ohio, signed a five-year contract with National N-Viro Energy
Systems to build and operate a $3 million facility to convert sewage sludge into fertilizer.
Toledo's Director of Public Utilities stated that a comparable composting plant would have cost
$25 to $30 million.108 The cost of operating the plant was estimated at about $39 per wet
metric ton ($35 per wet ton), after dewatering, compared to the $52 per wet metric ton that the
city was spending at the time to haul its sludge to a reclamation project. The city expects to
realize a profit from fertilizer sales resulting from the project within five years.109
The use of N-Viro Soil has increased rapidly as indicated by sales of the product that
have doubled each year for the past four years. In 1992, more than 900,000 metric tons of N-
Viro Soil were sold.110-111 Some CKD obtained by N-Viro goes to uses other than sewage
104 Anonymous, 1988. Toledo City Council Approves Five-Year Contract With National N-Viro Technology, Inc. For
Sewage-Agricultural Facility. PR Newswire. September 21.
105 Bumham, J.C., 1988, op. cit.
10* Personal communication with Robert Bastian, Office of Wastewater Enforcement and Compliance, EPA,
November 1992.
107 Personal communication with J. Patrick Nicholson, N-Viro Soil, December 7, 1992.
108 Anonymous, 1988. Toledo City Council Approves Five-Year Contract With National N-Viro Technology, Inc. For
Sewage-Agricultural Facility. PR Newswire. September 21.
109 Anonymous, 1988. Toledo Tries a Sludge First. Engineering News-Record. September 29.
110 CKD comprises only about 35 percent of this N-Viro soil.
111 Personal communication with J. Patrick Nicholson, N-Viro Soil, December 7, 1992.
-------
8-23
sludge stabilization. Although 80 percent of the CKD the company sells is used to stabilize
sewage by producing N-Viro Soil, the remaining 20 percent is sold as dust for direct application
in other beneficial uses.112 The company claims that use of N-Viro Soil has resulted in
substantial savings (more than $42 per acre) to farmers by reducing chemical fertilizer costs and
by increasing yields on crops such as soybeans, corn, and alfalfa.113 N-Viro Soil mixed with fly
ash has also been used as an aggregate diking material in Wilmington, Delaware.114 The
product is, however, expensive to transport. N-Viro sells 136,000 metric tons a year to a
company in New Jersey at $28 to $39 per metric ton, of which transportation costs represent 50
percent."5
Although N-Viro's technology has grown rapidly as a promising management alternative
for CKD that would otherwise be disposed, several factors may impede its market growth
potential. Such factors include regulatory uncertainties.116 Alternative additives may also
reduce the importance of CKD to sewage sludge stabilization. For example, N-Viro also uses
sulfur scrubbing residue and fluidized bed residue as a sludge additive, because these materials
have more activated carbon and impart better odor control than CKD.117 Although any trend
is unclear, these substances may replace CKD in the future.
Sewage sludge stabilization has been implemented on a more modest commercial scale as
well. In Cayce, South Carolina, government regulations required treatment of activated sewage
sludge before landfill disposal. After evaluating several processes, the city opted for a screw
press dewatering system, supplemented by the application of CKD. The sludge is dewatered to a
cake before it is mixed with the dust. The dust raises the cake's pH level from 9.0 to 11.2 to
destroy bacteria and other pathogens and chemically binds any heavy metals in the sludge. The
new system has reportedly saved the city money and keeps odors to a minimum.118
Oil Sludge
In addition to municipal sludge stabilization, the use of CKD to solidify oil sludge has also
elicited a fair amount of interest and research.119|120>121 According to one source, CKD has
112 Ibid.
113 Anonymous, 1992. N-Viro Achieves Record Year. PR Newswire. February 19.
114 Personal communication with Robert Bastion, Office of Wastewater Enforcement and Compliance, EPA,
November 1992.
115 Persona] communication with J. Patrick Nicholson, N-Viro Soil, December 7, 1992.
116 Ibid.
wlbid.
118 Billings, C.H., 1992. Screw Press/Kiln Dust Combination Doubles Drying Bed Production. Public Works. May.
119 Morgan, D.S., et a/., 1984. Oil Sludge Solidification Using CKD. Journal of Environmental Engineering.
October.
120 Thorsen, J.W., et al., 1983. In Situ Stabilization and Closure of an Oily Sludge Lagoon. 3rd Ohio Environmental
Conference. March. Columbus, Ohio.
-------
8-24
proven to be one of the most efficacious and economical means of solidifying non-recoverable
waste oil sludge, producing a stable and compactible fill material with good compressive strength.
Solidification of oily sludge in landfills makes it possible to use a reclaimed landfill site for
industrial construction.122
In 1983, CKD was used by the city of Wichita, Kansas, to solidify highly acidic oil sludge.
Oily sludge had accumulated during the 1950s and 1960s in the John's Sludge Pond from the oil
recycling and reclamation operations of Super Refined Oil Co. The use of sulfuric acid to refine
waste oil for recycling created an acidic layer on top of the sludge pond that frequently
overflowed into nearby surface waters. In 1983, under orders from EPA, the city excavated and
solidified the sludge using CKD with redeposition of the treated sludge into a compacted clay-
lined cell followed by capping, using a compacted clay cap. No contaminant levels requiring
further action were detected in surface and ground water following this action.123
In another example, CKD was used to solidify oil sludge during a site remediation project
near Dallas, Texas, in 1982. Operators determined CKD to be the best of several alternatives in
terms of chemical properties and cost effectiveness. The reaction of CKD with water to form
calcium hydroxide removed a significant portion of the free water from the sludge. This reaction
also provided a solid matrix of sufficient density and weight-bearing capacity that it could be used
as a fill. In addition, the binding effect of the kiln dust reportedly prevented the oil from
leaching out of compacted layers in the landfill to which it was transferred.124 The project
required an estimated 68,000 metric tons of kiln dust, which was blown into the sludge pit.
During the project, operators found that stockpiled kiln dust required a greater dust-to-oil mixing
ratio than recently generated dust,125 presumably because the newer dust contained more
unreacted lime.
Acid Waste
The alkaline nature of CKD makes it an effective neutralizing agent for treating acidic
materials. Substances that have been neutralized with CKD include industrial acidic wastes, such
as spent pickle liquor, and wastes from leather tanning and cotton seed delinting processes.126
Liquid hazardous wastes have been neutralized, oxidized, or reduced, and then solidified by the
addition of CKD and fly ash to form material that has the consistency of coarse gravel.127
121 Zarlinski, SJ. and J.C. Evans, 1990. Durability Testing of a Stabilized Petroleum Sludge. Paper from Hazardous
and Industrial Wastes, Proceedings of 22nd Mid-Atlantic Industrial Waste Conference. July 24-27. Pennsylvania.
122
Morgan, D.S., et al, 1984, op. cit.
123 Environmental Protection Agency, Office of Emergency and Remedial Response, 1989. Superfund Record of
Decision (EPA Region 7): John's Sludge Pond, Wichita, KS, (First Remedial Action). Washington, DC., September 22.
IM Anonymous, 1982. A Method For Oil Sludge Solidification and Disposal Using CKD Has Reclaimed 133 Acres
Near Dallas, Texas. Waste Age. April.
125 N-Viro Energy Systems, 1991, op. cit. pp. 100-102.
124 Davis, T.A., et al., 1975, op. cit.
127 Razzell, W.E., 1990. Chemical Fixation, Solidification of Hazardous Waste. Waste Management Resources.
April.
-------
8-25
As discussed below in more detail, CKD can also be used in land reclamation activities to
neutralize acid mine drainage,128 and has reportedly been used to neutralize acidic
wastewater.12' Other neutralization possibilities include CKD use to treat acidic mine waste
piles and leachate from hazardous waste and sanitary landfills.
Miscellaneous Wastes and Contaminated Soils
The use of CKD as a solidifying and stabilizing medium for a variety of wastes and
contaminated soils, in addition to municipal sludge, oil sludge, and acid wastes, has been studied
on several occasions. For example, laboratory studies have been conducted on the use of CKD
in conjunction with portland cement and rice husk ash to immobilize synthetic wastes containing
cadmium, lead, aldrin, chlordane, and electroplating wastes.130 Researchers have also
investigated the use of CKD to solidify and stabilize contaminated dredged materials.131 Some
uses have been implemented in the field as well. Some examples of miscellaneous CKD
stabilization uses are summarized below:
• In 1988, EPA initiated an ash solidification project to evaluate the performance of
several techniques, including mixing incinerator ash with CKD. CKD was used to
stabilize the ash and harden the mixture into a monolithic block, improving the
physical and handling characteristics of the ash.132
• In Brisbane, Australia, non-degradable liquid hazardous wastes are chemically
treated and then solidified by the addition of fly ash and CKD before permanent
burial in clay cells. The wastes include pesticides, paints, organic solvents, and
oily wastes. The researchers concluded that leaching test results have been "below
10 times EPA drinking water guidelines."133
• The use of CKD to treat sludges can make it possible to use on-site management
techniques instead of more expensive off-site disposal alternatives. CKD was used
to stabilize an organic sludge at a Superfund site. The stabilized material
reportedly provided a sound base for the final cover system.134
• Experiments have demonstrated the potential use of CKD to stabilize PCB-
contaminated sites. Last year, however, EPA's Risk Reduction Engineering
1M Davis, T.A., el al., 1975, op. at.
129 Personal communication with Ben Haynes, Engineering, Technology, and Research Division, Bureau of Mines,
U.S. Department of the Interior, November 1992.
130 Ahn, K.H., et al, 1988. Solidification of Hazardous Wastes: An Approach Using Cementitioits Binders. Paper
From Proceedings of 1988 Pacific Basin Conference on Hazardous Waste. February 1-6. Honolulu, Hawaii.
131 Betteker, J.M., et al., 1986. Solidification/Stabilization of Contaminated Dredged Material. Proceedings of Mid-
Atlantic Industrial Waste Conference. June 29-July 1. Lancaster, Pennsylvania.
152 Anonymous, 1988. EPA, Cities, Industry Press Congress For Incinerator Ash Legislation. Cogeneration Report.
April 22.
133 Razzell, W.R, 1990, op. cit.
134 Metry, A.A., et al., 1985, op. cit.
-------
8-26
Laboratory found that the heat produced from the quicklime reaction with PCBs
causes these carcinogens to volatilize. The Agency stated that further experiments
were planned to examine PCB decomposition and volatilization under simulated
field conditions.135
8.2.2 Soil Stabilization
As a soil stabilizer, CKD can decrease shifting, subsidence, fugitive dust emissions, and
erosion, and thereby provide temporary or permanent stability of soil at locations such as
construction sites. It mixes easily with existing soils and maximizes compacted density.
Respondents to the 1991 PCA Survey indicated that 12 percent of the CKD used beneficially was
used as a soil amendment, some of which probably included soil stabilization. According to
Keystone Cement, use of CKD as a soil stabilizer reduces construction costs when it is used to
speed up construction schedules (e.g., due to reduced necessary paving thickness, reduced
dewatering time).136
Blending CKD with other soil stabilizing agents can also be effective. For example, CKD
has been found to enhance the ability of waste sulfate to stabilize and strengthen soils. The
addition of fly ash to such a mixture can further improve soil strength.137 CKD can be injected
like lime into the ground using rig-mounted tubes that can be driven to desired depths, or it can
be mixed with soil using earth moving equipment. Keystone Cement and N-Viro both market
CKD as a soil stabilization product.
8.2.3 Land Reclamation138
In a manner similar to that used for soil stabilization, CKD has been utilized to reclaim
settling ponds, lagoons, or other lands. The added CKD stabilizes and dewaters these lands and
can render them useful for industrial, commercial, or residential purposes. Unlike other
reclamation procedures, processes, and materials, CKD can reportedly accommodate many
treatment procedures without the use of additional materials.139 The use of CKD to reclaim
lands that have been mined has also been studied.140 CKD is marketed by Keystone Cement
and N-Viro for use in land reclamation projects.
In addition to the high lime content in CKD, the easy flowing nature of CKD makes .it an
attractive neutralizing agent to pump into abandoned mines to treat acid mine drainage. In
1975, one cement plant reportedly disposed CKD in strip mines where it neutralized acid mine
135 Anonymous, 1991. Quicklime Volatilizes PCBs, EPA Finds. Superfund. June 28.
136 Keystone Cement Company, op. cit.
137 Nebgen, J.W., el al, 1976. Use of Waste Sulfate For Remedial Treatment of Soils, Volume I, Discussion of Results.
U.S. Department of Transportation, Federal Highway Administration. August.
"' Land reclamation efforts generally attempt to restore lands adversely affected by human activity to their original
state.
139
Keystone Cement Company, op. cit.
140 Libicki, J., 1984. Reclamation in Mountains, Foothitts, and Plains: Doing it Right. 9th Annual Meeting of the
Canadian Land Reclamation Association. August 21. Canada.
-------
8-27
drainage and precipitated iron from the run-off water.141 This treatment method may also help
reduce seepage of water from the mine.142 Specific quantities of CKD and water treated were
not provided in the literature, and EPA has not found more recent accounts of CKD use in mine
reclamation.
8.2.4 Agricultural Applications
CKD, like agricultural lime, is alkaline and contains a number of essential plant nutrients.
Because of these parallel characteristics, CKD has been used as an agricultural soil amendment
for a number of years. For example, in the mid-seventies, many U.S. cement manufacturers
reported that local farmers would occasionally visit their plants and haul away truckJoads of kiln
dust to spread on their fields.143 To better understand the advantages and limitations of CKD
as an agricultural amendment, numerous studies (cited below) have been conducted. These
studies, many of which took place outside of the United States, have sought to determine such
factors as the fertilizer equivalence and the lime equivalence of CKD, so that optimal CKD
application rates could be determined. The use of CKD as fertilizer and as a liming agent is
discussed below.
Fertilizer
Respondents to the 1991 PCA Survey and §3007 requests indicated that 11.7 percent of
the CKD used beneficially was as a soil amendment, some or all of which probably included use
as a fertilizer. CKD possesses significant fertilizer potential, particularly because of its high
potassium content. It has been used to this end at state and local levels in Ohio, Illinois, and
Pennsylvania because it provides savings over substitute products.144 Researchers have
suggested that a 0.9 metric-ton-per-acre (one-ton-per-acre) application of CKD would meet the
initial potassium requirement for corn on many soils.145 Soil scientists have also suggested that
other key plant nutrients contained in CKD, such as calcium, phosphorus, and zinc, might be
beneficial in some fertilizer applications.146-147
Numerous agricultural studies have been conducted to address specific applications of
CKD as a fertilizer. In Russia and Poland, several studies found CKD to be an acceptable and
inexpensive fertilizer for potatoes. Unlike most inexpensive potassium fertilizers, which contain
high amounts of undesirable chloride, the CKD used in this study had essentially no chloride.
141 Davis, T.A., et al, 1975, op. tit.
"2Ibid.
MIbid.
144 Persona] communication with Marc Saffley, Soil Conservation Service (SCS), November 18, 1992.
145Anonymous, 1981. CKD Use as Lime-Potash Fertilizer. Farm Chemicals. April.
146 Ibid.
147 Mettauer, H. and A.P. Conesa, 1981. Agronomic Value of Residua! Cement Dust. Comptes Rendus des Seances
de 1'Academie d'Agriculture de France. Volume 67, Number 9. pp. 772-781.
-------
8-28
Also, the sulfate in the dust led to a higher starch content in potatoes.148 Dutch researchers
found that comparable yields of oats were achieved when CKD versus limestone and K2SO4 were
used as fertilizer. Mixed peas-and-oats crops fertilized with kiln dust contained slightly more
protein than crops grown with KC1 fertilizer. In comparison to KCl-fertilized crops, the dust was
also found to yield fodder containing more starch, and sugar beets containing more sugar.149 In
a Czechoslovak an study, pot experiments using cereals and sunflower as test crops showed that
CKD was similar in effect to a potassium fertilizer.150
Before using CKD as fertilizer, however, it may be useful to treat it in some manner. For
example, dry CKD is easily wind blown, and some form of binding (e.g., pelletizing) may be
desirable. A Russian patent describes the preparation of granules by rolling the dust in water.
A rotary unloader can be used for this purpose, after which the granules are treated with CO2 to
make them non-hygroscopic151 and mechanically strong. Other treatment methods may be used
to modify the chemistry of CKD-based fertilizer to meet specific soil and crop needs. For
example, a Russian group used chlorination roasting to raise the K2O content of kiln dust to over
20 percent.152 Indian researchers developed a method to recover about 16.5 kilograms of
potassium sulfate per metric ton (15 kilograms per ton) of coarse CKD. The recovered potash
salts reportedly were pure enough to be used as fertilizer for crops such as potatoes and
tobacco.153 Similarly, the potassium sulfate recovered by the Dragon Products recovery
scrubber (described in Section 8.1.3) is also reportedly pure enough for use as fertilizer.154
Dragon Products currently has an agreement with a wholesaler to purchase the by-product.
In addition to pretreating CKD, it may also be worthwhile to blend CKD with other
fertilizer ingredients. For example, the magnesium content in CKD must be supplemented from
another source to achieve the required magnesium to calcium ratio for plant growth.155-156
Other examples are highlighted below.
• A Russian patent describes a process in which kiln dust is mixed with nitric acid-
phosphate extract to yield an N-P-K fertilizer.157
Ia Davis, T.A., et al, 1975, op. cit.
149 Ibid.
uo Kulich, J. and A. Ragas, 1973. Furnace Dusts from Cement Works as a Source of Available Nutrients.
Pol'nohospodarstvo. Volume 19, Number 2. pp. 113-121.
m Hygroscopic materials readily take up and retain moisture.
m Davis, T.A., et al, 1975, op. cit.
155 Chari, N.R. and D.K. Sahu, 1980. Studies on the Feasibility of Recovery of Potassium Salts from CKD. Fertilizer
Technology. Volume 17. pp. 69-70.
Vi Anonymous, 1991. Chloride-free Potash Fertilizer from Waste SO2 and CKD. Phosphorus and Potassium. July-
August. p. 48.
135 CKD is high in calcium, but contains relatively little magnesium.
156 Personal communication with Hillary Inyang, Professor, University of Wisconsin, November 16, 1992.
157 Davis, T.A., et al, 1975, op. cit.
-------
8-29
French researchers have suggested that CKD mixed with distillery sludge may
yield a material having a beneficial effect on crop yield and plant composition.
158
• A Canadian investigation found that CKD could enrich a slurry of swine manure
by increasing the levels of extractable calcium and potassium. The CKD mixture
also reduced odor levels.159
• • Researchers at Penn State University have suggested that CKD could be used to
produce a lime-potash fertilizer containing 35 percent calcium oxide, six percent
magnesium oxide, five percent potash, and four percent sulfur. With adequate
quality control, the product was projected to be worth $33 to $39 per metric ton
($30 to $35 per ton) as fertilizer.160
Use of CKD as a fertilizer may be of benefit to the physical characteristics of the soil as
well. A French study analyzed the effects of CKD on soil structure and infiltration and its value
as an amendment and fertilizer for rye grass. Based on laboratory, pot, and field experiments,
the study concluded that CKD could be a useful fertilizer and soil amendment. The study
documented no heavy metal toxicity to plants at normal application rates and suggested the
possible use of CKD as a replacement for gypsum in the treatment of saline soils.161
Although there has been a considerable amount of research conducted on CKD use as a
fertilizer, existing applications of CKD for this purpose have been mostly anecdotal, and there is
only limited evidence that commercial CKD use as a fertilizer is growing significantly. In
addition, the Soil Conservation Service (SCS), an authority on agricultural soils, is not conducting
any research on CKD.162 Nonetheless, N-Viro Energy Systems claims that N-Viro Soil has
resulted in substantial savings (more than $42 per acre) to farmers by reducing chemical fertilizer
costs and by increasing yields on crops such as soybeans, corn, and alfalfa.163 In 1990, an Iowa
farmer reported successful use of sewage sludge and CKD as fertilizer as an alternative to
conventional agricultural chemicals.164 As discussed previously, Dragon Products expects to
market one of its CKD recovery scrubber by-products as a potassium sulfate (potash)
u* Mettauer, H. and A.P. Conesa, 1981, op. cit.
159 Harrington, S.F. and A.F. MacKenzie, 1989. Enrichment of Swine Manures through CKD Incorporation.
Biological Wastes. Volume 29, Number 1. pp. 1-10.
160 Anonymous, 1981. CKD Use as Lime-Potash Fertilizer. Farm Chemicals. April.
161 Mettauer, H. and A.P. Conesa, 1981, op. cit.
162 Personal communication with Marc Saffley, Soil Conservation Service (SCS), November 18, 1992.
163 Anonymous, 1992. N-Viro Achieves Record Year. PR Newswire. February 19.
1M Looker, D., 1990. Boone Couple Adds Fuel to Sustainable Ag Debate. Des Moines Register. January 10.
Volume 141, number 171. p. 1A+.
-------
8-30
fertilizer.165 However, the amount of potash recovered from CKD for fertilizer applications is
currently insignificant in comparison to production from traditional sources.166
Liming Agent
As mentioned briefly above, CKD has significant potential as a liming agent. However,
the effectiveness of CKD relative to agricultural lime is a subject of some dispute. According to
one sdurce, CKD performed as well as lime in raising pH on a weight basis in tests on certain
soil types. However, in other tests, it took one and a half times as much CKD as lime to achieve
equivalent results.167 Research at the USDA station in Beltsville, Maryland, found that CKD
had about 80 percent of the soil neutralizing capacity of lime and about the same liming qualities
as pulverized limestone.168 Studies in Latvia showed that kiln dust could fully replace lime to
treat acidic soils to grow sugar beets or corn, and the dust could partially replace lime for
growing potatoes and rye.169 An Australian study found that CKD was effective in neutralizing
acid soils, and suggested that if given a choice of limestone alternatives, the selection should be
based on the relative costs of the purchase, transport, and application of the various
materials.170
Based on responses to the 1991 PCA Survey and §3007 requests, use of CKD as a liming
agent accounted for 5.6 percent of the CKD that went off site for beneficial use, and less than
0.5 percent of the gross CKD generated in 1990. Other documentation reports that CKD was
being marketed and used as an agricultural lime on a regional basis in New York in the mid-
eighties.171-172 It is not clear what limitations or benefits were encountered from this activity.
8.2.5 Livestock Feed Ingredient
The alkaline properties of CKD have given rise to strong interest in the past in using
CKD as a feed ingredient in livestock diets. In performing research on this application, however,
the presence and effects of various trace metals, such as arsenic, barium, cadmium, chromium,
165 Anonymous, 1991. Chloride-free Potash Fertilizer from Waste S02 and CKD. Phosphorus and Potassium. July-
August, p. 48.
166 Anonymous, 1991. Potash Mining in Alsace, France. Phosphorus and Potassium. September-October, p. 21.
167 Anonymous, 1980. Supplementing Ruminant Feeds with CKD Improves Livestock Performance, According to
Agriculture Canada. Feedstuff's. June 2. p. 38.
'" Davis, T.A., el al, 1975, op. cit.
169 Ibid.
170 Dann, P.R., B.S. Dear, and R.B. Cunningham, 1989. Comparison of Sewage Ash, Crushed Limestone, and CKD
as Ameliorants for Acid Soils. Australian Journal of Experimental Agriculture. Volume 29. pp. 541-549.
171 Naylor, L.M., J.C. Dagneau, and I.J. Kugelman, 1985. CKD - A Resource Too Valuable to Waste? Proceedings
of the Seventeenth Mid-Atlantic Industrial Waste Conference on Industrial and Hazardous Wastes. June 23. pp. 353-
366.
172 Naylor, L.M., E.A. Seme, and TJ. Gallagher, 1986. Using Industrial Wastes in Agriculture. BioCycle. February.
pp. 28-30.
-------
8-31
lead, mercury, and selenium, have been regarded with concern.173 Although international
interest in the use of CKD for livestock feed has been relatively high, as of the mid-eighties, U.S.
regulatory agencies forbade the use of CKD in the diets of animals destined for human
consumption.174-175 Under interstate commerce regulations, the FDA has not approved the
use of CKD as a livestock feed ingredient. If the meat from an animal fed with CKD does not
leave the state the animal lived in, however, this would not violate FDA requirements.
Therefore, it is possible for some animals to be fed CKD, but the FDA is not aware of this
occurring.176 According to an FDA representative, no approval requests are currently being
processed for CKD use as an animal feed, though occasional inquiries are received.177
Experiments conducted on steers and lambs in the late seventies revealed that diets
containing 3.5 percent CKD, with and without supplemental protein, provided better growth
results than diets without CKD. Further, carcasses from steers fed kiln dust were superior to
those of other steers (e.g., they had more fat over the rib, a higher marbling score, and "graded"
higher).178 An analysis of the complete diets, in correlation with kidney and liver tissues,
showed that there was no undesirable accumulation of elements such as arsenic, cadmium, lead,
or selenium.179-180 The results of the experiment were attributed to the neutralizing effect of
CKD on rumen acids in the animals' gastrointestinal tracts, the presence of macro and trace
mineral elements, and the possible increased mineral availability afforded by CKD.181 Similar
research has not produced findings of significantly elevated levels of heavy metals, or any cases of
toxicity, in the animals studied. Further, one study concluded that the long-term feeding of CKD
to steers did not elevate metal levels sufficiently to cause any real concern. According to one
source, based on World Health Organization standards, meat from these steers would be of little
concern in a well-balanced diet.182 A Russian study found that CKD fed to cattle increased
173 U.S. Department of Agriculture, 1979. Letter from William E. Wheeler, Research Animal Scientist, Nutrition,
to John P. Lehman, Hazardous Waste Management Division, Office of Solid Waste, U.S. EPA. February 23.
174 Wheeler, W.E., 1981. Variability in Response by Beef Steers to CKD in High Concentrate Diets. Journal of
Animal Science. March. Volume 52. pp. 618-627.
175 Bush, R.S. and W.G. Nicholson, 1985. The Effect of CKD on Tissue Accumulation of Trace Minerals in Steers.
Canadian Journal of Animal Science. June. Volume 65. pp. 429-435.
176 Persona] communication with Dr. Donna Waltz, Center for Veterinary Medicine, Animal Feeds Division, FDA,
November 1992.
177
Ibid.
"' Wheeler, W.E. and R.R. Oltjen, 1979. CKD in Complete Diets for Finishing Steers and Growing Lambs. Journal
of Animal Science. March. Volume 48. pp. 658-665.
179 Wheeler, W.E., 1978. CKD: A Potential Feed Ingredient for Livestock. Cereal Foods World. Volume 23. pp.
296-297, 299, 311
IM Wheeler, W.E. and R.R. Oltjen, 1979, op. cit.
181 Ibid.
112 Bush, R.S. and W.G. Nicholson, 1985, op. cit.
-------
8-32
body weight gain and reduced the percentage of premature culling of stock.183 Similar
experiments with rats and swine during this same period showed a positive growth effect like that
found in cattle.184-185 Canadian experiments showed weight gain rates of 22 percent in sheep
and nine percent in cattle fed with CKD. The best results were with finishing lambs and young
Holstein heifers.186
Researchers have also found that feed supplemented with CKD did not consistently
stimulate growth of rats, mice, hamsters, and lambs.187-188'189 Italian researchers found that
while CKD had no adverse effect on the health of lambs, it also had no significant effect on rate
of weight gain, feed intake, feed conversion efficiency, or carcass weight.190 Similarly,
additional studies of steers, yearling beef heifers, dairy cows, and lambs also indicated an
inconsistent or nonexistent growth response to CKD.19U92il93'194'195>19fi This variability
in response appeared to be the result of variability in the composition of CKD between different
183 Karadzhyan, A.M., A.G. Chirkinyan, L.V. Efreraova, and A.A. Evoyan, 1982. Effect of CKD on Growth and
Development of Young Cattle. Trudy Erevanskogo Zootekhnichesko-veterinarnogo Instituta. Number 53. pp. 10-14.
1M Roginski, E.E. and W.E. Wheeler, 1978. A Growth Effect of Georgia CKD in Rats. Federation Proceedings.
Volume 37. p. 404.
185 Newton, G.L. and O.M. Hale, 1979. CKD and Carboxylin as Feed Additives for Swine. Journal of Animal
Science. October. Volume 49. pp. 908-914.
116 Anonymous, 1980. Supplementing Ruminant Feeds with CKD Improves Livestock Performance, According to
Agriculture Canada. Feedstuffs. June 2. p. 38.
'" Galvano, G., A. Lanza, L. Chiofalo, and M. Mal'an, 1982. CKD as a Mineral Source in Feeding Ruminants.
World Review of Animal Production. Volume 18, Number 4. pp. 63-71.
1M Roginski, E.E. and W.E. Wheeler, 1979. The Response of Monogastric Species to CKD in the Diet. Federation
Proceedings. Volume 38. p. 614.
189 Zinn, R.A., R.A. Lovell, D.R. Gill, F.N. Owens, and K.B. Poling, 1979. Influence of CKD on Animal
Performance and Nutrient Availability. Journal of Animal Science. Volume 49. p. 422.
190 Galvano, G., A. Lanza, L. Chiofalo, and M. Mal'an, 1982, op. cit.
191 Anonymous, 1980. Supplementing Ruminant Feeds with CKD Improves Livestock Performance, According to
Agriculture Canada. Feedstuffs. June 2. p. 38.
192 Ward, G.M., C.A. Olds, D.D. Caveny, and G.A. Greathouse, 1979. CKD in Finishing Lamb Diets. Journal of
Animal Science. September. Volume 49. p. 637.
193 Noller, C.H., J.L. White, and W.E. Wheeler, 1980. Characterization of CKDs (Fed to Animals) and Animal
Response. Journal of Dairy Science. November. Volume 63. pp. 1947-1952.
194 Zinn, R.A., R.A. Lovell, D.R. Gill, F.N. Owens, and K.B. Poling, 1979. op. cit.
195 Wheeler, W.E., 1981, op. cit.
196 Felix, A., D.R. Rao, C.B. Chawan, and P.I. Deem, 1980. Effect of CKD on Nutrient Digestibility in Sheep.
Annual Research Report: School of Agriculture, Alabama A&M. pp. 103-109.
-------
8-33
sources and even within the same source.197 Other researchers agreed that the variable
composition of CKD makes it somewhat unreliable as a feed additive."8 The feed mixture fed
to animals could also be a source of these varying results. For example, diets containing alfalfa
hay appeared to be unaffected by CKD supplementation because they already had sufficient
buffering capacity.199
Along with inconsistent results, some research has indicated potentially adverse impacts
from using CKD as a livestock feed amendment. German researchers found that in instances
where CKD was fed to cattle, weight increases were up to 14 percent lower than would be
expected. In one case, an extremely low weight gain required a reduction of the percentage of
CKD fed to the animal.200 A study of CKD fed to rainbow trout showed no significant
difference in growth rate or feed conversion. It also showed an increase in selenium
concentrations in the fishes' livers, though no fish died and there were no pathological signs.201
CKD fed to swine depressed body weight gain and apparently interfered with normal bone
metabolism to the extent of causing bone lesions on the humerus.202 CKD fed to broiler chicks
caused no significant improvement in growth rate or feed utilization when fed at low levels, and
caused severe rickets when fed at high levels (five to nine percent).203
The use of CKD as a feedstock additive in animals used for human consumption does not
appear to be viable in the United States in the immediate future, despite the fact that most
research has reported positive or neutral effects.
8.2.6 Lime-Alum Coagulation in Water Treatment
In water treatment, CKD can be substituted for lime in coagulation processes.204 CKD
was reportedly used in 1975 in Oregon as a partial and total replacement for lime in the
preparation of alum floe to remove turbidity from water. Use of kiln dust was successful in
197 Noller, C.H., J.L. White, and W.E. Wheeler, 1980, op. cit.
1M Hogue, D.E., PJ. Van Soest, J.R. Stouffer, G.H. Earl, W.H. Gutenmann, and D.J. Lisk, 1981. CKD as a
Selenium Source in Sheep Rations. The Cornell Veterinarian. January. Volume 71. pp. 69-75.
199 Noller, C.H., J.L. White, and W.E Wheeler, 1980, op. cit.
200 Flachowsky, G., HJ. Lohnert, G. Stubedorff, E. Flachowsky, G. Staupendahl, and A. Hennig, 1982. The Use of
Portland CKD in the Feeding of Fattening Bulls. Archiv fur Tierernahrung. Volume 32. pp. 93-98.
201 Rumsey, G.L., W.H. Gutenmann, and D.J. Lisk, 1981. CKD as an Additive in the Diets of Rainbow Trout.
Progressive Fish-Culturist. Volume 43. pp. 88-90.
202 Pond, W.G., D.A. Hill, C.L. Ferrell, and L. Krook, 1982. Bone Lesions in Growing Swine Fed 3 Percent CKD as
a Source of Calcium. Journal of Animal Science. January. Volume 54. pp. 82-88.
203 Veltmann, J.R. and L.S. Jensen, 1979. Effect of Georgia CKD on Broiler Chick Performance. Poultry Science.
Volume 58. p. 1027.
204 Eger, V.G. and O.N. Mandryka, 1984. The Possibility of Using Bentonite Clay for Purification of Wastewaters
From Apatite Processing. Journal of Applied Chemistry of the USSR. Volume 57. pp. 2420-2422.
-------
8-34
neutralizing the water and in improving flocculation.205 More recent examples citing the use of
CKD for this purpose have not been found.
8.2.7 Construction Applications
Although CKD is not typically blended into the finished cement product at the facility,
CKD has been found in some construction applications to perform well when blended with
cement and aggregates to make concrete. In these instances, CKD is often blended together
with other additives, such as fly ash and/or lime. The concrete made from CKD blending can be
used for purposes such as road base construction. There have been a number of studies on the
suitability of concrete made with CKD for this and other construction applications.
Blending with Portland Cement
A large number of studies have demonstrated that CKD can successfully replace a
portion of portland cement in making concrete. Various blends have been researched, using only
CKD as an additive, and using CKD with other additives. According to responses to the 1991
PCA Survey and §3007 requests, about 2.7 percent of the CKD that was sold or given away in
1990 was used as a materials additive.
CKD as the Only Blending Agent
Some research shows that CKD can replace over 50 percent of prescribed portland
cement for certain concrete applications. A study of concrete made with five percent CKD
found that the properties of CKD concrete were almost the same as those of normal concrete
mixes.206 Additional research demonstrated that a five percent replacement of portland cement
with CKD did not appreciably affect the freeze-thaw durability of cement.207 Another study of
concrete made with various proportions of CKD concluded that, while replacing cement with
CKD generally increased water demand and decreased concrete strength, CKD could replace
cement by up to 15 percent without causing significant strength loss.208 Martin Marietta
Corporation has submitted a patent application for concrete blocks containing 10 to 60 percent
CKD. The blocks reportedly show improved compressive strength compared to blocks without
CKD.209
Coupled with the positive results outlined above, the use of CKD as a replacement for
portland cement has also been shown to have limitations. According to some studies, excessive
quantities of CKD in cement (amounts vary depending on CKD composition) will decrease the
strength and
205 Davis, T.A., et al., 1975, op. cit.
206 Ramakrishnan, V., 1986. Evaluation of Kiln Dust in Concrete. American Concrete Institute, pp. 821-839.
207 Ramakrishnan, V. and P. Balaguru, 1987. Durability of Concrete Containing CKD. American Concrete
Institute, pp. 305-321.
208 Ravindrajah, R.S., 1982. Usage of CKD in Concrete. International Journal of Cement Composites and
Lightweight Concrete. May. Volume 4, Number 2. pp. 95-102.
209 Martin Marietta Corporation, date unknown. Poured, Moulded, or Pressed Concrete Blocks Contain Aggregate,
Cement, and CKD.
-------
8-35
workability of the cement product.210-211 CKD also tends to retard setting time.212-213-214
The high levels of sulfate and alkali contained in certain dusts may also have an undesirable
effect on concrete durability.215 One study showed that alkali-aggregate reactivity in cement
made with CKD caused greater expansion at six months than ordinary cement.216 In a CKD
treatment (alkali removal) and use investigation, however, the U.S. Bureau of Mines concluded
that concrete made from either sintered or melted CKD exhibited strength equal to or greater
than ASTM standards.217
CKD as a Co-Blending Agent
Some of the limitations of CKD as a concrete ingredient may be overcome by
incorporating additional materials, such as fly ash or slag, along with the dust. A 1980 study
found that pozzolanic218 concrete containing CKD and fly ash had the property of
autogenous219 healing and concluded that such concrete was potentially useful as road base and
merited further development.220 Subsequent studies showed that the addition of either slag or
fly ash to cement-CKD blends resulted in better or similar characteristics (strength, setting time,
and workability) in comparison to ordinary concrete.221 Fly ash also reportedly acts to inhibit
the expansion resulting from alkali-aggregate reactivity.222 This effect might yield a higher
alkali content in portland cement, allowing increased CKD recycling at applicable kilns if the
ASTM standard were changed. Little information has been found on this topic.
210 Bhatty, M.S.Y., 1985. Use of CKD in Blended Cements: Alkali-Aggregate Reaction Expansion. World Cement
December. Volume 16, number 10. pp. 386, 388-390, 392.
211 Ravindrajah, R.S., 1982, op. cit.
212 Bhatty, M.S.Y., 1984. Use of CKD in Blended Cements. World Cement. May. Volume 15, Number 4. pp.
126-134.
213 Ramakrishnan, V., 1986, op. cit.
214 Ravindrajah, R.S., 1982, op. cit.
215 Valley Forge Laboratories, Inc., 1982. Kin Dust-Fly Ash Systems for Highway Bases and Subbases. U.S.
Department of Transportation and U.S. Department of Energy. September.
216 Bhatty, M.S.Y., 1984, op. cit.
217 Wilson, R.D. and W.E Anable, 1986, op. cit.
214 A pozzolan is a material rich in silica or silica and aluminum that is chemically inert and possesses little or no
value as a cementing agent, but, when in a finely divided form and in the presence of water, will react with calcium
hydroxide to form compounds possessing cement-like properties. The most commonly available pozzolan in use in the
United States is fly ash (Valley Forge Laboratories, 1982, p. 7).
219 Originating or derived from a source within the same subject.
220 Miller, C.T., D.G. Bensch, and D.C Colony, 1980. Use of CKD and Fly Ash in Pozzolanic Concrete Base
Courses. Transportation Research Record, pp. 36-41.
221 Bhatty, M.S.Y., 1984, op. cit.
222 Bhatty, M.S.Y., 1985, op. cit.
-------
8-36
Some research has moved into more commercial stages of development. For example,
the U.S. Patent Office has received applications and, in some cases, issued patents for the use of
various blends of CKD in concrete. Examples of patent applications include those for the
following products:
• A high iron hydraulic cement manufactured from red mud and CKD by calcining
them with gypsum, lime, and alumina. This composition reportedly sets rapidly, is
stronger in compression than portland cement, is more tolerant to the presence of
alkali oxides, and has enhanced bonding to steel reinforcement. It is also low in
cost due to the use of waste materials and the lower fusion temperature used in manufacturing;
• Cement containing blast furnace slag, CKD, and/or calcium carbonate, and
optionally, gypsum. This composition is reportedly suitable as a partial
replacement for portland cement, is less expensive, and forms concretes with
better setting times and compressive strengths;224
• A mixture of fly ash or pozzolan, CKD, aggregate, and water that, when
compacted and reacted at ambient temperature, can be used as stabilized base
material to underlie road surfacing. This composition reportedly minimizes the
use of energy-intensive materials such as lime and asphalt;225 and
• A blend of portland cement, CKD, and phosphogypsum for use in producing air-
filled concrete panels.226
The use of CKD as a blending ingredient for concrete is apparently being actively
researched and marketed. Ongoing studies of applications for CKD-blended concrete may
provide new alternative uses in the future.
Use as a Road Base Material
According to the 1991 PCA Survey responses and §3007 requests, approximately 1.2
percent of the CKD used beneficially in 1990 was used for road base construction. This
application of CKD has been researched since the 1970s. A study in the late seventies by the
Transport and Road Research Laboratory in the United Kingdom concluded that freshly-
produced CKD has little application in road making, but that well-weathered CKD could be
useful as bulk fill.227 More recently, however, research studies have demonstrated broader
I
223 Regents of the University of California, 1986. High Iron Hydraulic Cement Manufactured from Red Mud and
CKD. PCT Patent Application Number WO--86-05773. October 9.
224 Standard Concrete Material, Inc., 1983. Cement Composition as Substitute for Portland Cement Containing Blast
Furnace Slag, CKD, and/or Calcium Carbonate, and Optionally Gypsum. PCT Patent Application Number WO--83-
01443. April 28.
225 Nicholson Realty, date unknown. Cementitious-hardening Paving Base Composition Using Waste Materials.
226 Bryansk Technical Institute, date unknown. Solution for Building Applications Containing Portland Cement,
CKD, and Phosphogypsum.
227 Sherwood, P.T., L.W. Tubey, and P.G. Roe, 1977. The Use of Waste and Low-Grade Materials in Road
Construction. Transport and Road Research Laboratory. Crowthorn, England.
-------
8-37
applications for CKD in road construction. Saudi Arabian researchers, for example, found that
sand stabilized with CKD could be utilized for base materials in highway construction.228
CKD blending has also been investigated for use in road construction. For example, the
U.S. Department of Transportation and the U.S. Department of Energy tested the effectiveness
of substituting CKD for hydrated lime in lime-fly ash-aggregate road base systems. CKD was
found to perform well in pozzolanic road base compositions involving some form of lime-fly ash
stabilization. CKD generally yielded mixes with high resistance to freezing and thawing and
some mixes developed early strength, possibly extending the normal cut-off dates for late season
construction. The study found that, with few exceptions, fresh CKD worked with nearly any fly
ash to produce strengths as high or higher than those observed with commercial hydrated lime
and fly ash, although larger CKD quantities were required compared to normal hydrated lime to
achieve the same strength. The study also found that aged CKD from stockpiles had a lower
free lime content and resultant poor reactivity. Additionally, CKD from dry process plants
tended to produce the highest strength concrete. Total dusts containing both fine and coarse
CKD were better than separated CKD. The study concluded that, owing to its calcium oxide or
unreacted lime content, CKD may be used in place of hydrated lime or portland cement as a
pozzolanic road base material.229
Research has also demonstrated that some CKD blends do not perform satisfactorily.
For example, a study by the Florida Institute of Phosphate Research concluded that a mixture of
phosphogypsum and CKD was not useful for road construction.230 CKD has an inherent
propensity to degrade or react in place. This instability can lead to differential settlement
problems. Further, CKD has no shear strength due to its fine-grained particle size. The shear
strength due to cohesion alone is minute unless the dust is modified for use. The fine-grained
character of CKD introduces the additional problem of erodability and sediment transport.231
The use of CKD as a road base material appears to have developed to a commercial
stage and to be of continued interest, though quantities used for this purpose are currently small.
Depending on the project, a state can specifically require the use of CKD or a contractor may
request it. One explanation for the limited use of CKD as a road sub-base may be that new
construction calls for flexible pavements with drainable bases. CKD, in contrast, is rigid and has
a low permeability. In addition, less new sub-base construction is currently taking place, giving
way to increased sub-base rehabilitation instead. Nonetheless, CKD provides an economically
viable alternative to substitute products, such as fill materials and lime, because it is more
economically competitive. Transportation costs can, however, add substantially to the price and
ultimately drive the market for the dust.232
^Baghdad!, Z.A. and M.A. Rahman, 1990. Potential of CKD for the Stabilization of Dune Sand in Highway
Construction. Building and Environment. Volume 25, Number 4. pp. 285-289.
2!9
Valley Forge Laboratories, Inc., 1982, op. cit.
230 May, A., J.W. Sweeney, and J.R. Cobble, 1983. Use of Florida Phosphogypsum in Synthetic Construction
Aggregate. Florida Institute of Phosphate Research Publication Number 01-008-026.
231 Personal communication with Hillary Inyang, Professor, University of Wisconsin, November 16, 1992.
232 Personal communication with Mike Rafalowski, DOT-FHA, November 16, 1992.
-------
8-38
Despite the apparent economic advantages of using CKD as a road base material, the
subject does not appear to have attracted much continuing attention. Neither the Department of
Transportation (DOT) nor the Federal Highway Authority is funding research on the subject, nor
does DOT maintain data records on the use of CKD in this manner.233
8.2.8 Sanitary Landfill Daily Cover
Because of the fine nature of CKD particles, the use of CKD as a landfill cover can
probably only be achievable when it is blended with some other material. N-Viro Soil, for
example, is used as a daily cover for a number of municipal landfills with contracts set for up to
15 years beginning in 1990. N-Viro Soil is being used for this application at the rate of 9 to 104
dry metric tons (10 to 115 dry tons) per day per landfill, depending on the landfill.234-235
8.2.9 Mineral Filler
According to an early source, CKD has been used as a mineral filler for bituminous
paving materials and asphaltic roofing materials. It has also been suggested as a filler for plastics
and for asphaltic products such as insulating board, concrete expansion strips, and sound
deadening material.236 EPA's research has not yielded more recent discussions of such
applications.
8.2.10 Lightweight Aggregate
In the mid-seventies there was at least one process under development to use CKD in the
manufacture of lightweight aggregate.237 However, EPA's research has not yielded more recent
discussions of this application.
8.2.11 Glass Making
Researchers have reported success in the use of CKD to make glass for which color and
high chemical stability are not essential considerations. According to this finding, CKD can serve
as a partial replacement for soda ash in the manufacture of green glass because it increases .the
rate of sulfate decomposition, the main cause of foaming in glass baths.238 The developmental
stage of this technology is uncertain, and EPA's research has not yielded more recent discussions
of the use of CKD in glass making.
8.3 ON-SITE LAND DISPOSAL
234 N-Viro Energy Systems, 1991, pp. 9-10, op. cit.
35 Persona] communication with Robert Bastion, Office of Water-Wastewater Enforcement, EPA, November 1992.
236 Davis, T.A., et al, 1975, op. cit.
mlbid.
-------
8-39
Net CKD that is removed from the kiln system and not beneficially utilized is disposed,
generally in landfills, piles, or ponds. In a landfill, CKD is generally disposed below grade (e.g.,
in mines, stopes, or quarries) and is sometimes buried between layers of earth. Piles are typically
above-grade deposits of CKD. Submerged management of CKD in ponds accounts for only a
small portion of on-site CKD management. As discussed in Chapter 4, responses to the 1991
PCA Survey and EPA observations during 1992 sampling activities demonstrate that the
predominant CKD waste management practice at cement plants is disposal in a retired portion of
the limestone quarry.
An alternative to these land disposal practices is to dispose CKD in an engineered
landfill. Engineered landfills are typically constructed with environmental controls that are
designed to contain wastes within the disposal unit. Monitoring of ground water and other
environmental media in the vicinity of the landfill is often performed to ensure that the
environmental controls are functioning properly. Daily operations are performed according to
procedures that limit exposure of nearby populations to windblown dust and other potential
hazards.
The remainder of this section discusses, in general terms, the design and operating
practices frequently used at engineered landfills. The specifics of landfill design actually vary
widely from site to site depending on numerous factors such as the intrinsic hazard of the waste;
the requirements imposed by federal, state, and local regulations; the climate and hydrogeology
of the site; the resource value of the underlying ground water; the proximity of nearby
populations and endangered species; and the location of the site relative to sensitive
environments such as floodplains, seismic impact zones, and wetlands.
Engineered landfills are designed with run-on control systems. Run-on from adjacent
property can increase the amount of water percolating into the landfill and contribute to leachate
formation; leachate is liquid that has percolated through the wastes and extracted dissolved or
suspended materials. Run-on can be controlled through construction of diversion ditches, trench
drains, and other devices. Typically, run-on control systems are designed to prevent flow onto
the active portion of the landfill during the peak discharge from a 25-year storm. Well-designed
landfills also have run-off control systems to prevent surface run-off from the site from entering
nearby areas and streams. Run-off control systems are typically designed to collect and control
the water volume resulting from a 24-hour, 25-year storm.
Engineered landfills are equipped with components that contain and remove leachate.
Liner systems are frequently installed prior to placement of wastes in landfills to prevent leachate
from entering ground water. Liner systems are constructed with low-permeability soils and/or
synthetic materials that are sloped to divert the leachate to underdrain pipes, which collect the
leachate for treatment; these are known as leachate collection systems (LCS). Liner
configurations frequently used include a single layer of compacted clay; a flexible membrane liner
(FML) made of high density polyethylene (HDPE) or other material underlying a LCS; a
"composite" liner system consisting of a LCS and FML overlying a two- to three-foot layer of
compacted clay; and a "triple" liner system consisting of two FMLs with LCSs above and between
them, overlying a layer of compacted clay. Well-designed LCSs maintain less than a 30-cm depth
of leachate over the liner. All components of the liner and LCS must be constructed of
materials that have appropriate chemical properties and sufficient strength and thickness to
prevent failure due to pressure gradients, physical contact with the waste and leachate, and other
stresses.
-------
8-40
Landfills equipped with leachate collection systems must also have mechanisms in place
for leachate treatment and disposal. Frequently, collected leachate is recirculated back into the
landfill. Alternatively, leachate may be treated on site and then discharged to a surface water
body. A third alternative is to discharge the leachate to a municipal wastewater treatment plant
with or without prior treatment, depending on the characteristics of the leachate and the
requirements of the sewage treatment plant. Many different types of biological and
physical/chemical treatment technologies are available for treating leachate prior to discharge to
surface water or a wastewater treatment plant. Discharges of collected leachate and run-off to
surface water must be performed in accordance with National Pollutant Discharge Elimination
System (NPDES) requirements established pursuant to Section 402 of the Clean Water Act.
Ground-water monitoring is frequently conducted to detect leachate releases from
landfills and evaluate the degree and significance of resultant ground-water contamination.
Effective monitoring well systems must comprise a sufficient number of appropriately located
wells able to yield ground-water samples that represent the quality of background ground water
and the quality of ground water downgradient of the fill area. The number, spacing, and depths
of monitoring wells are based on site-specific characteristics. Samples are collected periodically
and analyzed for hazardous constituents or for parameters that indicate that a release has
occurred. Statistical analysis of the samples is performed to help determine whether a release
has occurred and the nature and extent of the contamination. Operators of some landfills also
conduct monitoring of surface water, soils, and air.
If contamination is significant, corrective action is taken to clean up the environment to
the extent feasible. Many different types of ground-water corrective action technologies are
available, including source controls to minimize further releases (e.g., excavation of the waste,
placing a low-permeability cap over the fill area); ground-water recovery wells that remove
ground water from the subsurface and treat it to reduce contaminant levels; and slurry walls,
which restrict ground-water flow and thereby minimize further spread of the contamination. The
technical feasibility, costs, and effectiveness of these technologies vary widely from site to site.
Closure and post-closure care are important components of environmentally protective
landfill management. When a landfill (or a portion of the landfill) is filled to capacity, a final
cover is installed to minimize infiltration and erosion. The cover may consist simply of vegetated
top soil. More sophisticated covers also contain a liner made of polyvinyl chloride (PVC) or
other synthetic material underlying a drainage collection system; "composite" cover systems also
include a two-foot clay layer. The cover should be designed with a permeability less than or
equal to the permeability of the bottom liner system or natural subsoils to prevent ponding at the
bottom of the landfill. After the landfill is closed, post-closure care is conducted for many years.
Post-closure care activities typically include maintenance of the integrity of the landfill cover,
operation of the LCS, and monitoring of ground water.
A wide variety of additional design and operating features are practiced at landfills,
including access controls (e.g., installation of fences to prevent public exposure to hazards), the
use of daily cover (covering each day's fill with soil to prevent dust from blowing), and others.
The degree of latitude that a landfill owner or operator may exercise in selecting among the
environmental controls discussed above depends largely on the regulatory status of the landfill.
Without federal or state requirements, the need for these controls depends on the characteristics
of the waste and the environmental and exposure characteristics of the waste disposal site.
-------
8-41
8.4 SUMMARY AND FINDINGS
This chapter has presented an overview of CKD management alternatives. The length of
discussion allotted for the various technologies does not necessarily reflect EPA's views on their
relative merits, but instead is a function of disparities in the availability of information.
Moreover, the conclusions derived from these discussions are tentative and subject to
reevaluation upon receipt of new information.
Although investigators have explored numerous alternative CKD management practices,
the scope of practices being utilized commercially remains limited. The perceived economics of
CKD management and familiarity with existing techniques may be limiting more widespread
adoption of alternatives to CKD disposal. For example, though available technologies could
allow nearly 100 percent of gross CKD to be recycled to the kiln, the capital cost of adopting this
practice may seem excessive for many operators, especially given the absence of strong incentives
to reduce the quantities of net CKD generated. EPA has examined in detail the economic
feasibility of some of these alternative management practices. This analysis is presented in
Chapter 9 of this report.
Exhibits 8-4 and 8-5 (on pages 8-43 to 8-46) summarize the alternatives discussed in this
Chapter for minimizing CKD removal from the kiln system and for beneficially using CKD that
is removed from the system. The exhibits indicate general characteristics of each alternative in
terms of technical feasibility, environmental considerations, economic considerations, and trends
in use. Overall findings with respect to these factors are discussed below.
-------
8-42
Exhibit 8-4
Summary of Alternatives for Minimization of CKD Removal from the Kiln System
ALTERNATIVE
Recovery Scrubbing
Leaching with Water and
Return to Kiln
Fluid Bed Dust Recovery
Leaching with Kd Solution
and Return to Kiln
Alkali Volatilization
Control of Gross CKD
Generation Rates
Control of Raw Feed and
Fuel Inputs to the Kiln
Direct Return to Flame
End
TECHNICAL FEASIBILITY;
DEVELOPMENTAL1
STAGE"
Pilot/
Limited Commercial
Limited Commercial
Pilot
Theoretical/Bench
Theoretical/Bench
Full
Commercial
Full
Commercial
Full
Commercial
OPERATIONAL
IMPACTS'
-Increased system
efficiency
-Versatile fuel choices
-Sludge drying system
required for dry kilns
None Identified
None Identified
None Identified
-Can decrease process
stability
-Can decrease product
quality
Can decrease product
quality
-Can reduce flame
temperature
-Resuspends Dust
OPERATIONAL
COMPLEXITY
Moderate to High
- However, no
net increase of
personnel is
required
Low
High
Unknown
Unknown
Low
Low to Moderate
Low
ECONOMICCONSIDERATIONS: |
START-
UP
COSTS
High
Low
High
Unknown
Unknown
Low
Low
Low
OPERATING;
COST
SAVINGS'
High-
Investment
paybadc
expected in <5
yrs.
Moderate
High
Low to High
Low to High
Low
Low to
Moderate
Low to High
NEW
PRODUCTS
-Potassium sulfate potentially
marketable as fertilizer
-Distilled water
Leaching solution suitable as
liquid potassium fertilizer
Unusable dust may be of value
as a fertilizer
Unknown
Can yield cement product or
cement additive
No
No
No
ENVIRONMENTAL
.CONSIDERATIONS;
-Reportedly discharges only dean air
and distilled water
-Pilot plant is beginning to consume- its
backlog of dust previously generated
-Reportedly removes 90 to 98 percent
of the SO, in the flue gas
-Increased materials and energy
efficiency
-Less CKD to Disposal
-Increased materials and energy
efficiency
-Less CKD to Disposal
-Other aspects unknown
-Increased materials and energy
efficiency
-Less CKD to Disposal
-Other aspects unknown
Unknown
-Increased materials and energy
efficiency
-Less CKD to Disposal
-Increased materials and energy
efficiency
-Less CKD to Disposal
-Increased materials and energy
efficiency
-Less CKD to Disposal
NOTES
Reportedly improves kiln's
paniculate capture efficiency of
CKD and unburned organic*
-Leaching with hot water removes
more alkali than leaching at ambient
conditions
In some cases, a binder may be
required for pelletization of CKD
Optimal leaching conditions were
found at 70 to 80*C
Sintering is the primary method
-Primary control is to minimize
turbulent conditions in kiln
-Major process changes to reduce
gross CKD would rarely be
economically justifiable
-Alternative raw feed sources limited
-Use of alternate fuel can affect
temperature profile of kiln and must
be monitored
-Hazardous waste fuels containing
higher levels of chlorine can increase
net CKD generation
-Limited by reduced Qame
temperature that dust causes in
burning zone
-Causes continuous resuspension of
dust
-------
8-43
Direct Return to Mid-Kin
Limited
Commercial
None Identified
Low
Low
Low to High
No
-Increased materials and energy
efficiency
-Less CKD to Disposal
-Can release fugitive dust
Difficulties can arise in mounting the
sleeve
Exhibit 8-4 (continued)
Summary of Alternatives for Minimization of CKD Removal from the Kiln System
ALTERNATIVE:
Direct Return with Raw
Feed
Pelletizing and Return to
Kiln
TECHNICAL FEASIBILITY :
DEVELOPMENTAL?
STAGE:
Full
Commercial
Full
Oonun era al
OPERATIONAL
IMPACTS*
Some obstacles for wet
shiny
Reduced resuspension of
dust
OPERATIONAL
COMPLEXITY
Low
Unknown
START-
UP.
COSTS
Low
Low
ECONOMIC CONSIDERATIONS
OPERATING
COST
SAVINGS'
Low to High
Low to High
NEW
PRODUCTS ;
No
No
ENVIRONMENTAL •;
CONSIDERATIONS:
-Increased materials and energy
efficiency
-Less CKD to Disposal
-Increased materials and energy
efficiency
-Less CKD to Disposal
NOTES
Can return CKD either at kiln input
or at blending stage
-Gives CKD necessary strength to
withstand forces of being 6red into
flame
-Avoids need for flame characteristic
modification
* "Developmental Stage" describes the usage level at which the technology is being implemented, classified as Theoretical, Bench Scale, Pilot Scale, or in Limited Commercial or
Full Commercial development.
b "Operational Impacts" describes effects of the technology on the kiln system or the clinker product.
c "Operating Cost Savings" provides a qualitative assessment of the operating costs savings likely to be realized through using a given technology, and is considered
from start-up costs. Classes are low, moderate, and high, with many technologies ranging from low to high because savings depend upon the amount of net CKD that is re
-------
8-44
Exhibit 8-5
Summary of Alternatives for Beneficial Utilization of CKD Removed from the Kiln System
ALTERNATIVE
Stabilization of Sewage
Sludge
Sanitary Landfill Daily
Cover
Stabilization of Oil Sludge
Stabilization of Add
Wastes
Stabilization of
Miscellaneous Wastes
Soil Stabilization
Road Base
Land Reclamation
FertDizer
Liming Agent
Blending Agent with
Portland Cement
Lime-Alum Coagulation in
Water Treatment
Blending Agent with Other
Materials
DEVELOPMENT At
STACE-
Full
Conun era u
Full
Commercial
Full
Commercial
Full
Commercial
Full
Commercial
Full
Commercial
Full
Commercial
Full
Commercial
Full
Commercial
Full
Commercial
Bench
Unknown
Bench
VALUE OF
PRODUCT*
High
High
Moderate
Moderate
Moderate
Moderate
Low
Low
Moderate
Moderate
Low - Moderate
Unknown
Low - Moderate
MARKET STATUS*
POTENTIAL
DEMAND
High
High
High
High
High
High
Moderate to
High
High
High
High
Moderate
Unknown
Moderate?
CURRENT
STATUS'
Common
limited- Common
Common
Common
Common
Common
Limited-Common
Limited-Common
Limited-Common
Common
Limited-Common
Unknown
Limited-Common
APPARENT
TRENDS
Growing
Growing
Growing
Growing
Growing
Unknown
Growing
Unknown
Stable-Growing
Stable-Growing
Growing
Unknown
Growing
ENV1RONMENTALCONSIDERATIONS
CKD may be dispersed in environment with undear
impacts
CKD is managed under controlled conditions
CKD is dispersed in environment with undear impact •
CKD is dispersed in environment with undear impacts
CKD is dispersed in environment with undear impacts
-CKD is dispersed in environment with undear impacts
-Helps reduce fugitive dust and erosion from soil
CKD is dispersed in environment with undear impacts
CKD is dispersed in environment with undear impacts
-CKD is dispersed in environment with undear impacts
-Potential limitations for food chain crops
-CKD is dispersed in environment with undear impacts
-Potential limitations for food chain crop*
CKD is dispersed in environment with undear impacts
Unknown impact on treated water
CKD is dispersed in environment with undear impacts
NOTES
-Successfully marketed as a treatment
•gent under the names Stablesorb and
N-Viro Sofl
-Municipality can save $13 per wet
metric ton ($12 per wet too) on
sewage disposal
Most applications use CKDfeewage
sludge blend
Decrease* shifting and subsidence
Variable composition of CKD hinders
reliability
-Large percentages wilt decrease
strength and workability, and retard
setting time
-Variable composition of CKD
hinders reliability
-Variable composition of CKD
hinders reliability
-Adds autogenous healing to
pozzolanic concrete
-------
8-45
Exhibit 8-5 (continued)
Summary of Alternatives for Beneficial Utilization of CKD Removed from the Kiln System
ALTERNATIVE
Mineral Filler
Lightweight Aggregate
Glass Making
Livestock Feed Ingredient
DEVELOPMENTAL
STAGE-
Limited
Commercial
Unknown
Theoretical/Bench
Theoretical/Non-
oonuneraal Pilot
VALUE OF
PRODUCT*
Unknown
Unknown
Unknown
Unknown
; MARKET STATUS*
POTENTIAL
• DEMAND
Unknown
Unknown
Unknown
Unknown
CURRENT
STATUS
Unknown
Unknown
Unknown
Not beEeved to
be used in VS.
APPARENT
TRENDS
Unknown
Unknown
Unknown
No Growth
ENVIRONMENTAL CONSIDERATIONS
Unknown
CKD is dispersed in environment with undear impacts
. Unknown
CKD in diets of animab destined for human
consumption is not permitted in U.S.
NOTES
One process under development in
the mid-seventies
Succeosful when color and high
chemical stability are not essential
Variable composition of CKD hinders
reUabiEty
1 "Developmental Stage" describes the usage level at which the technology is being implemented, classified as Theoretical, Bench Scale, Pilot Scale, or in Limited Commercial or
Full Commercial development.
b Value of product is qualitatively ranked relative to anticipated value of other beneficial uses (highly speculative).
c "Market Status" is comprised of three categories: Potential Demand, based on a qualitative estimate of the user market; Current Status, based on available information about
current CKD demand for a given use; and Apparent Trends, based on a qualitative assessment of current demand trends.
-------
8-46
8.4.1 Technical Feasibility
Most of the management alternatives discussed above are technically feasible to at least
some degree. The differences lie in how practical these alternatives are in terms of investment
requirements, expected benefits, and performance standards. Process controls that minimize
gross dust generation rates and alkali levels are commonly used throughout the industry.
Significant reductions in current CKD generation probably cannot be achieved through these
means without compromising product quality. Although process differences can influence the
amount of CKD recycled (e.g., fuel type, process type, feed inputs, etc.), the incremental benefit
of initially removing less CKD from the kiln system is unlikely to induce significant and
potentially costly process changes.
CKD treatment and return systems show the greatest promise for increasing the amount
of CKD returned to the kiln system. These treatment systems minimize net CKD generation by
removing alkalies and other contaminants and returning treated dust to the system without
compromising product quality. The promising technologies in this area include recovery
scrubbing, alkali leaching, and fluid bed dust recovery. Water-based alkali leaching shows the
greatest promise for effective CKD treatment with minimal technological requirements.
Recovery scrubbing and fluid bed dust recovery, in contrast, continue to undergo development,
and require significant expertise for design, installation, and optimization.
Beneficial uses of CKD are at a further stage of development than recycling technologies.
In contrast to CKD treatment and recycling technologies, all of the beneficial utilization
technologies appear to be readily feasible technically (at least at the pilot scale), and may involve
activities as simple as blending CKD with other materials. Nonetheless, research is required to
more fully commercialize such uses, since even blending requires knowledge of appropriate
mixing ratios. Incorporation of fly ash with cement reportedly reduces the negative influence of
alkalies on concrete strength. Additional research should be conducted to determine whether
blending fly ash with cement might allow greater alkalies in clinker, and therefore greater CKD
recycling rates in the kiln system. Aside from blending, the more technically challenging
beneficial uses include glass making and coagulation in water treatment.
Although many viable beneficial uses of CKD have been identified, the inherent
variability of CKD, as established in Chapter 3, poses a significant limitation on its widespread
use. Cement kiln operations are designed to optimize the chemical and engineering
characteristics of clinker, while CKD is a byproduct for which specifications cannot be developed.
Hence, depending upon its ultimate use, the dust may require testing and possibly further
processing. This situation may limit the option of using CKD as a material in scenarios such as
when a project planner knows little or nothing about the available CKD nearest to the project
location and may be willing to use a more expensive, but more standardized material.
Strictly managed land disposal practices are technically feasible, as has been demonstrated
with landfills containing many other materials. Numerous engineering firms have the capability
to design all necessary environmental protection features, including leachate collection and
treatment systems, liners, ground-water monitoring systems, and run-on and run-off controls.
These operations tend to be extremely costly, however, a reality that may provide an incentive
toward developing other management alternatives.
-------
8-47
8.4.2 Human Health/Environmental Considerations
Direct recycling or treatment and recycling practices provide an inherent environmental
benefit by minimizing the amount of net CKD that will be disposed of. These practices conserve
energy and resources originally used to prepare and heat the raw feed that would otherwise
become wasted CKD. Furthermore, total CKD return to the kiln system eliminates the
environmental liabilities associated with land disposal.
In addition to influencing CKD generation rates, the impact of alternative practices on
other process waste streams should be considered. For example, some alternatives, like recovery
scrubbing, can improve kiln emissions quality. This technology also reportedly generates only
distilled water and potash. In contrast, any wastewater generated from alkali leaching may
warrant some concern. Water containing alkalies may be released to surface or ground waters if
this material is handled improperly. Based on EPA's information, however, this system can be
and has been installed in a way that is environmentally protective. Further investigation of alkali
leaching wastewater disposal may be warranted before the impacts of this technology can be fully
assessed.
The environmental implications for beneficially using CKD are uncertain. The nature of
most alternatives for beneficial utilization of CKD is to disperse it in some manner into the
environment. CKD managed in this manner will generally be exposed to climatic influences such
as precipitation and wind. Some alternatives, however, occur in controlled conditions, such as
those in which CKD is blended with sewage sludge and used as a municipal landfill daily cover.
Alternatives that involve contact with human food chain products, such as fertilizer production
and livestock feed, may require more careful consideration. When appropriate, however, the use
of CKD may reduce the demand for traditional materials (e.g., limestone) that would have to be
mined at some environmental cost. This holds true for uses such as road base construction, soil
stabilization, and waste stabilization.
CKD management in land disposal units represents wasted quantities of mined and milled
raw materials. As documented in Chapter 6, CKD disposal in quarries may not adequately
protect human health and the environment. One option is disposal in strictly managed landfills,
which would significantly reduce threats to human health and the environment.
8.4.3 Economic Feasibility
Although the start-up costs of many of the CKD recycling technologies discussed in this
chapter are low to moderate, inducing significant changes in the CKD management practices at
cement facilities may require time and successful demonstration projects. The least capital-
intensive CKD treatment technology appears to be water-based alkali leaching. Leaching systems
can be installed with minor start-up and maintenance costs, especially as applied to wet process
kilns. This technology may also generate marketable by-product(s). Recovery scrubbing and
fluid bed dust recovery also show market potential. In particular, recovery scrubbing appears to
show the greatest potential for adaptability and effectiveness. Notwithstanding the high start-up
costs associated with these technologies, net benefits can be realized within a few years. Each of
these technologies provides the economic incentives of increased product yield, and reduced
resource losses to disposed CKD.
The economic incentives to sell CKD for beneficial utilization appear to be moderate but
growing. In particular, uses such as stabilization of municipal sewage treatment sludge could
-------
8-48
prove to be in high demand. Municipalities will reportedly pay over $11 per metric ton ($10 per
ton) for CKD as a new treatment alternative or as an alternative to lime. Not only does selling
CKD for beneficial use allow operators to use the material rather than simply disposing of it, but
kiln operators can realize significant revenues from such activities. For example, one facility
(Ash Grove Cement) reportedly sells its dust at $11 to $22 per metric ton, a figure that may be
considered full profit since the material would otherwise be disposed. Cement product, in
contrast, is typically sold with a profit of about $5.50 per metric ton.239 Regardless of the price
at which CKD might be sold for such uses, the fact that operators can receive income makes it
highly feasible economically. Additionally, CKD is significantly less expensive to users than most
or all alternative materials. However, this savings can be lost in transport costs if the distance
between the user and the cement plant is too great.
The increased costs of more strict land disposal practices would represent a significant
liability to all plants disposing CKD. Implementation of Subtitle C requirements would result in
costs that might make a number of CKD treatment technologies economically viable. Even
eventual Subtitle D requirements, though less onerous than Subtitle C requirements, could
impose a significant economic burden on operations that currently dispose their CKD.
8.4.4 Current Extent of Use and Trends
Currently, CKD management incentives appear to be diverging in two directions, with
both recycling and beneficial use offering attractive prospects. On one hand, operators want to
minimize CKD removal from the kiln to conserve resources and energy lost to a waste material,
and to minimize operating costs for CKD waste management units. Considering the necessary
investments in capital and labor, however, the prospect of avoided costs associated with most of
the technologies to minimize CKD removal from the kiln appears to hold little appeal at present.
In contrast, the beneficial use of CKD as a marketable product appears to be growing,
such that an operator might find positive incentives to remove CKD from the kiln system and sell
it, thereby avoiding a costly recycling system. If EPA finds that CKD does not warrant
hazardous waste regulation, many tentative markets for beneficial utilization of CKD may
develop more fully to create a significant increased demand for and price of CKD. Nonetheless,
several factors could increase incentives to minimize removal of CKD from the kiln. These
include stricter CKD management regulations, increased fuel and feed costs, and the
development of more economical recycling technologies. To minimize resource losses through
disposal, and to minimize the use of CKD in beneficial areas of unknown environmental impact,
EPA believes that the first CKD management objective should be to economically recycle as
much CKD to the kiln as possible without compromising the quality of the clinker product.
In general, some of the most promising technologies for minimization of CKD removal
from the kiln system appear to be recovery scrubbing, alkali leaching, and fluidized bed recovery.
These technologies utilize all or nearly all of the CKD generated, may produce one or more
marketable products, and provide a return on the invested capital.
Among the beneficial uses, stabilization of municipal sewage sludge appears to have a
great deal of potential as a means of beneficially utilizing CKD, particularly as a daily municipal
landfill cover, where the location of CKD use and final disposition is carefully controlled. EPA
does not believe that widespread agricultural use of CKD for producing crops that are consumed
239 Personal communication with Hans Steuch, Ash Grove West, December 9, 1992.
-------
8-49
by humans should be practiced without full characterization of the CKD from each source.
Caution should also be exercised in using CKD as a livestock feed supplement. Other potentially
useful applications of CKD include waste stabilization, soil stabilization, soil amendment (liming
agent), road base construction, and blending with portland cement to make miscellaneous
cement-based construction products.
If EPA determines that Subtitle C regulation of CKD is warranted, the significant
potential cost of complying with Subtitle C land disposal requirements may increase the
marketability of many recycling technologies. Subtitle D requirements, if implemented and
enforced at the state level, could also increase the economic viability of CKD recycling
technologies, which would result in less land disposal of CKD and less wastage of raw material.
Requirements under these regulatory alternatives are described in detail in Chapter 9 of this
report.
-------
8-50
CHAPTER EIGHT
ALTERNATIVE CKD MANAGEMENT PRACTICES AND POTENTIAL UTILIZATION
8.0 OVERVIEW 1
8.1 MINIMIZATION OF CKD REMOVAL FROM THE KILN SYSTEM 2
8.1.1 Control of CKD Generation Rates . . . . 2
8.1.2 Direct Return of CKD to the Kiln 3
Return to Flame End 5
Return with Raw Feed 5
8.13 Treatment and Return of CKD to the Kiln 6
Pelletizing 6
Leaching with Water 6
Leaching with a Potassium Chloride Solution 8
Alkali Volatilization 8
Recovery Scrubbing 9
Fluid Bed Dust Recovery 13
8.2 BENEFICIAL USE OF REMOVED CKD 16
8.2.1 Stabilization of Sludges, Wastes, and Contaminated Soils 17
Sewage Sludge 17
Oil Sludge 21
Acid Waste 22
Miscellaneous Wastes and Contaminated Soils 23
8.2.2 Soil Stabilization 24
8.2.3 Land Reclamation 24
-------
8-51
8.2.4 Agricultural Applications 25
Fertilizer 25
Liming Agent . 28
8.2.5 Livestock Feed Ingredient 28
8.2.6 Lime-Alum Coagulation in Water Treatment 31
8.2.7 Construction Applications 31
Blending with Portland Cement 32
CKD as the Only Blending Agent 32
CKD as a Co-Blending Agent 33
Use as a Road Base Material 34
8.2.8 Sanitary Landfill Daily Cover 35
8.2.9 Mineral Filler 36
8.2.10 Lightweight Aggregate 36
8.2.11 Glass Making 36
8.3 ON-SITE LAND DISPOSAL 36
8.4 SUMMARY AND FINDINGS 38
8.4.1 Technical Feasibility 43
8.4.2 Human Health/Environmental Considerations 44
8.4.3 Economic Feasibility 44
8.4.4 Current Extent of Use and Trends 45
-------
8-52
LIST OF EXHIBITS
Exhibit 8-1 Flow Chart of Gross CKD Management Pathways 1
Exhibit 8-2 Process Flow Diagram of Recovery Scrubber 10
Exhibit 8-3 Process Flow Diagram of Fluid Bed Dust Recovery Process 15
Exhibit 8-4 Summary of Alternatives for Minimization of CKD Removal from the
Kiln System 39
Exhibit 8-5 Summary of Alternatives for Beneficial Utilization of CKD Removed from
the Kiln System 41
-------
CHAPTER NINE
COST AND ECONOMIC IMPACTS OF ALTERNATIVES
TO CURRENT CKD DISPOSAL PRACTICES
9.0 INTRODUCTION
Section 8002(o)(5) of RCRA requires EPA to analyze "alternatives to current disposal
methods" for cement kiln dust (CKD) waste, while Section 8002(o)(6) requires the Agency to
analyze "the costs of such alternatives" and Section 8002(o)(7) directs EPA to address "the
impact of those alternatives on the use of natural resources." This chapter presents EPA's
analysis of the cost and potential economic impacts of adopting a wide variety of alternative
practices for managing CKD, including the use of emerging technologies. This analysis draws on
the information presented in preceding chapters addressing current management practices
(Chapter 4) and potential management alternatives (Chapter 8). The results of this analysis
contributed to the formulation of a range of regulatory status options, which are presented for
public review and comment in Chapter 10.
This chapter consists of three sections. The first describes the approach and methods
used to develop the cost and impact estimates. The second presents and discusses the costs of
managing CKD under a variety of different management practices. The final section explores
potential impacts on the cement industry and its markets and relates these impacts to a general
discussion of regulatory management options.
9.1 APPROACH AND METHODS
This section describes how EPA conducted its cost analysis. A short section on the
conceptual framework used for the analysis is followed by a description of the methodology used
to estimate facility costs and a discussion of data sources and limitations. Details on the
Agency's approach, methods, and results are provided in a Cost Background Document1
prepared in support of this report, which may be found in the RCRA docket (No. F-93-RCKA-
FFFFF).
RCRA requires EPA to analyze alternatives to current CKD disposal methods and their
costs. "Alternatives" can be thought of in two distinct ways: alternative regulatory frameworks
that EPA might select, and alternative management practices that individual cement kiln
operators might adopt in response to regulatory changes. To avoid confusing these two ideas, this
report will refer to the choices to be made by EPA as regulatory scenarios, and the choices made
by cement kiln operators in response to changing regulations as CKD management alternatives
or responses. This report first provides estimates of the costs of CKD management alternatives
for the case study plants, and then uses these cost estimates to address the broader question of
the potential industry-wide costs and impacts associated with these practices, as described in
Section 9.3.
1ICF Incorporated, 1993. Technical Background Document: Cost and Economic Impacts of Alternatives to Current
CKD Disposal Practices.
-------
9-2
In EPA's basic analytical framework, the costs imposed by an alternative management
practice are measured as the difference in cost between the current management practice
(referred to hereafter as the "baseline") and the (generally different) alternative practice.
9.1.1 Data Sources
Detailed site-specific data on cement plant operation and CKD generation and waste
management practices form the basis for the analyses presented in this chapter. The more
important data were drawn primarily from the 1991 Portland Cement Association (PCA) Survey
responses, supplemented by EPA observations made during the 1992 field sampling visits to the
10 cement plants addressed in this analysis and referenced in earlier chapters. Data on emerging
CKD management technologies were obtained from both published and primary sources,
including detailed discussion and correspondence with the developers of these technologies.
Most of the industry and market data upon which EPA has based its assessment of the economic
conditions facing the cement industry were obtained from documents published by PCA and the
U.S. Bureau of Mines (BOM). PCA's U.S. and Canadian Portland Cement Industry. Plant
Information Summary. December 31.1991. and BOM's Cement (1990) were particularly useful,
as were documents published by the U.S. International Trade Administration, International
Trade Commission, and Bureau of the Census.
9.1.2 Approach to Estimating Costs and Impacts of CKD Management Alternatives
EPA's basic approach to analyzing the costs of CKD management alternatives in this
report is to estimate the financial costs of each alternative as they would be experienced by a
sample of 10 selected cement manufacturing plants. These estimates were made by applying
cost-estimating functions to the specific conditions found at the facilities in the sample. This
approach may be contrasted with an exhaustive analysis of the costs at every facility in the
country, for which sufficient data were unavailable, and a model facility approach, which may
have lacked realism for specific facilities. A disadvantage of this approach is that in order to
extrapolate the results of the cost analysis at specific facilities to estimates of nationwide costs,
one must assume that the sample is representative of the industry as a whole. To address this
issue, EPA has taken care to select plants that encompass much of the range of conditions found
across the industry.
Case Study Plants
EPA selected a sample of 10 plants for detailed analysis and discussion. The 10 plants
examined were drawn from 15 plants at which EPA conducted CKD sampling during February
and March, 1992. Of these 15 facilities, the 10 selected as the sample for the costing analysis are
those plants for which EPA's knowledge of existing operational and waste management practices
is most complete. These 10 plants are identified in Exhibit 9-1, which also presents some of the
key data on plant operations, CKD generation, and CKD management that are used in the
analyses. As discussed in Chapter 1, the Agency has assembled this sample so as to reflect, to
the extent possible, the full range of cement kiln technology types, geographic regions of the
U.S., and types of fuel used. Accordingly, EPA believes that the 10 facilities examined in this
chapter are adequately representative of the population of active CKD-generating plants in the
U.S. to support general conclusions regarding the cost and economic impacts of adopting
alternative CKD management practices.
-------
9-3
To verify that the sample is representative of the population, EPA conducted t-test
comparisons of the average gross and net CKD generation rates of the sample of 10 and the
remaining 69 cement plants for which data are available. These variables were selected because
they have a strong bearing on the costs of adopting various CKD management methods. The
mean gross and net CKD generation rates of these two groups cannot be distinguished at a 95
percent confidence level, suggesting that they are drawn from the same overall population and
-------
9-4
Exhibit 9-1
Facilities Included in the Cost Analysis
Company
Ash Grove
Cement
Dixie (Southdown)
Essroc Materials
Hoi n am
Holnam
Independent
Cement
Kaiser Cement
LaFarge
Rinker Portland
Cement
River Cement
Location
Chanute,
Kansas
Knoxville,
Tennessee
Speed,
Indiana
Clarksville,
Missouri
Tijeras, New
Mexico
Catskill,
New York
Cupertino,
California
Fred on i a,
Kansas
Miami,
Florida
Festus,
Missouri
Kiln
Type
Wet
Dry
Ph/Pc
Dry
Ph/Pc
Wet
Dry
Ph/Pc
Wet
Dry
Ph/Pc
Wet
Wet
Dry
Long
Hazardous
Waste
Burner
(Yes/No)
Yes
Yes
No
Yes
No
No
No
Yes
No
Yes
Gross
CKD
Generation
Rate
(Metric
Tons/Yr.)
80,244
27,431
172,249
344,700
58,659
57,299
162,388
71,372
95,246
69,054
Net CKD
Generation
Rate
(Metric
Tons/Yr.)
77372
16406
28,274
227,000
25,755
57,293
545'
67,446
900
53,784
CKD
Disposed
(Metric
Tons/Yr.)
59,963
4386
25,008
214,753
25,755
36,025
454
67,446
900
53,784
CKD
Sold
(Metric
Tons/Yr.
)
17,409
12,120
3,266
12,247
0
21,268
91
0
0
0
Source: Responses to 1991 Portland Cement Association (PCA) Survey.
1 Although the operator of this plant reported a net CKD generation rate of 545 metric tons/yr. for 1990, EPA
determined during its 1992 site sampling visit that the plant currently recycles 100 percent of the gross CKD
generated. Accordingly, in the remainder of this chapter, EPA assumes that net CKD generation by this plant is
zero.
therefore, that the sample adequately represents the larger group of cement plants for which
EPA has data, and by inference, the industry as a whole.
Methods for Estimating Facility Costs
To calculate the costs of managing CKD in various ways for the 10 plants, EPA
developed and applied cost-estimating functions, based on an engineering analysis of each
alternative and its component operations and activities. These functions were developed to
-------
9-5
express CKD management costs as a function of waste generation rate and other plant-specific
operating variables.
In EPA's cost estimating analysis, the first step was to estimate the costs and benefits2 of
waste management activities and their distribution over time. The second step was to discount
all future costs to the present and then calculate the equivalent annualized compliance cost or
benefit. The annualized compliance cost or benefit is the average annual cost or benefit (i.e.,
annuity), over the assumed operating life of the facility, that has the same total present value as
the sum of the actual expenses incurred and revenues received at their actual times. This
method offers the distinct advantage of allowing comparisons among alternative technologies
whose costs and benefits may be incurred at different times.
Cost estimating functions were developed from an engineering analysis of each
technology, and were generally based on empirical data regarding each cost element for a given
technology. The sum of the costs of these elements equals the total facility cost for a particular
CKD management strategy. Similarly, the benefits accruing to the facility operator of adopting a
particular CKD management alternative are expressed on an annualized basis; these benefits are
in the form of operating savings and additional revenues. In all cases, EPA's cost estimating
procedures consider both initial capital investment costs, and annual operating and maintenance
(O&M) costs (e.g., materials, labor, and utilities). Results are expressed as annualized total
costs, total and annualized capital costs, and unit costs (e.g., cost per unit of waste or product).
For certain CKD management alternatives, e.g., disposal of CKD in a secured landfill
under RCRA Subtitle C, two additional categories of costs may be incurred. In one category are
the capital costs for disposal facility closure and annual costs of post-closure care and
maintenance, which are simply capital and O&M costs that are incurred beginning at facility
closure. In the other category are the costs associated with potential corrective actions for solid
waste management units that release hazardous constituents to the environment; these costs
would apply only to cement plants that are newly regulated under RCRA Subtitle C. EPA has
not explicitly included the costs of corrective action in the final impact analysis due to the wide
range of uncertainty associated with these cost estimates. Nonetheless, some estimates of
possible plant-level corrective action costs for typical facilities are presented in Section 9.2 for
illustrative purposes.
Methods for Extrapolating from the Case Study Sample to the Industry
Having estimated the costs of various CKD management methods for a representative
sample of plants in the cement industry, the Agency performed an analysis of the economic
impacts of regulatory scenarios. This analysis is presented in Section 9.3.
As part of this exercise, EPA extrapolated the estimates obtained for the sample of 10
plants to the industry as a whole, to gauge the potential nationwide impacts of the scenarios
considered. EPA estimated nationwide impacts of general regulatory management options in two
steps.
2 In this analysis, EPA has considered both the operating savings and the income generated through the sale of
new byproducts and services associated with certain CKD management methods. These savings and income streams
are referred to throughout this chapter as "benefits."
-------
9-6
Cost impacts were first scaled up from the 10 case study plants to the larger 79 facility
sample for which the Agency has CKD generation rate data from the PCA Survey. Next, cost
impacts were scaled up from these 79 plants to all plants in the domestic industry. To scale the
cost from the 10 sample plants to the 79 plants, EPA multiplied total costs estimated for the 10
plants by the ratio of (1) total net CKD for the 79 plants to (2) total net CKD for the 10 plants.
To estimate total costs for the additional 36 plants (for which CKD generation data were not
available), the Agency extrapolated costs from the 79 survey plants based on relative cement
production rather than CKD. To estimate nationwide costs, EPA thus scaled up the total cost
for the 79 survey respondents by the ratio of (1) total U.S. cement production capacity to (2)
estimated cement production capacity of the 79 plants.
For waste management alternatives that might affect only the 35 plants currently burning
or projected to burn hazardous waste, the analysis was conducted in the same manner except that
costs for the five hazardous waste burning plants in the sample were first scaled up to the 17
hazardous waste burning plants represented among the 79, and then to the 35 hazardous waste
burning plants nationwide.
9.1.3 Cost Accounting Assumptions
Costs of regulations can be viewed in two contexts, economic and financial. The two
perspectives consider regulatory costs in two very different ways for different purposes. The
economic context considers impacts on resource allocation for the economy as a whole, while the
financial context evaluates private sector effects on facilities, firms, and other discrete entities.
For this report, EPA has focused on the financial context (i.e., impacts on facilities and the
industry), in keeping with the statutory directives articulated at RCRA §8002(o), by evaluating
the costs of alternative management practices and their effects on the industry.
Consequently, in conducting this analysis, EPA has employed data and cost accounting
assumptions that reflect the viewpoint of cement producers. For example, the Agency has
employed a discount rate (9.49 percent) that approximates the likely cost of obtaining financing
for regulatory compliance-related expenditures, rather than a "social" discount rate, or cost to
society. This discount rate is based on an estimate of the weighted average cost of capital to
U.S. industrial firms.3 Similarly, costs and benefits have been calculated on an after-tax basis, to
better reflect the actual financial impacts of prospective regulatory requirements.
In estimating the costs of applying specific waste management technologies, the Agency
made a number of additional assumptions, as described in the Cost Background Document.
9.1.4 Limitations of the Analysis
The analytical results presented below are based upon the application of simple cost
engineering models to a sample of 10 cement plants that EPA has assumed are representative of
the industry as a whole. To the extent that this assumption is not valid (i.e., there are important
operational practices or technologies being used to generate and manage CKD that are not
known to the Agency), the results of this analysis may yield biased conclusions. Given, however,
the scope and depth of EPA's information collection process (e.g., site visits to nearly 20 percent
3ICF Incorporated, 1990. Regulatory Impact Analysis for tlie Proposed Rulemaking on Corrective Action for Solid
Waste Management Units (Draft). Prepared for Economic Analysis Staff, Office of Solid Waste, U.S. EPA. June 25,
1990.
-------
9-7
and detailed survey results from nearly 70 percent of the plants in the industry), the Agency
believes it unlikely that any important CKD management technologies or site-specific CKD
management practices have been overlooked.
There are important limitations in EPA's understanding of the technical aspects and costs
of direct CKD recycling. While virtually all cement plants directly recycle some portion of the
CKD collected, the Agency has only a limited understanding of the extent to which facility
operators currently attempt to maximize the quantity recycled (or control the trace metal or
dioxin concentrations), the engineering and operational constraints on this practice, and the
economic trade-offs between the incremental costs of increasing recycling rates and the benefits
of recovering the resource value contained in the CKD. Because increasing direct CKD recycling
is perhaps the simplest and most effective means of eliminating CKD disposal and its associated
impacts, the Agency views this as a key information gap.
Finally, EPA's knowledge and understanding of certain market and technical issues limits
the Agency's ability to offer definitive conclusions regarding the feasibility of certain CKD
management options. For example, while it is clear that alkali content of the cement product
can be a controlling factor in the extent of CKD recycling in some parts of the country, EPA
does not have sufficient information on the regional variability in cement product markets (which
are often driven by state transportation department specifications) and raw material composition
to determine which plants are or may be constrained by the 0.6 percent alkali limit established by
ASTM and which are not.4 Similarly, the feasibility of raw material substitution as a means of
increasing CKD recycling rates broadly across the industry cannot be determined based upon
current information. The Agency does, however, view this option as a promising, low-cost
alternative to land disposal of CKD, at least for some plants.
9.2 DESCRIPTIONS AND COSTS OF BASELINE AND ALTERNATIVE CKD
MANAGEMENT METHODS
The waste management practices discussed in this report reflect the range of practices
that are currently employed to manage CKD, as well as alternative management techniques that
the Agency believes could be employed by facility operators in response to new regulatory
requirements. These practices fall into four basic categories: (1) current practices; (2)
alternative land disposal practices; (3) alternative recycling and recovery; and (4) other operating
practices.
At present, at least some operators of U.S. cement kilns are using most of these practices,
and many use combinations of several. As shown in Chapter 3 of this report and in Exhibit 9-1,
cement plant operators most often directly recycle some fraction of their gross CKD, many sell
some portion of their net CKD for off-site use, and most dispose the remainder in on-site waste
management units. In only a few cases is CKD treatment and recovery practiced, and in no
instance, to EPA's knowledge, is CKD currently managed as a RCRA Subtitle C hazardous
waste.
As discussed in Chapters 4 and 8, based upon extensive research and evaluation, EPA
believes that certain trends in CKD management are apparent, and that at least some of these
trends may continue regardless of the Agency's ultimate decision concerning the regulatory status
of CKD. The most important trend observed is the move away from disposal and toward waste
4 The reader is referred to Section 8.1.2 of this report for background information on this topic.
-------
9-8
reduction, recovery, or productive use. For example, some operators have been successful in
increasing direct CKD recycling, either by reconfiguring dust handling systems or modifying their
raw material mix; in at least a few cases, net CKD generation rates have been cut to zero in the
process. At the same time, sales of CKD for off-site use have been increasing,5 as cement
companies have more aggressively promoted the sale and use of this material for stabilizing
wastes, amending agricultural soils, and other applications. Finally, the historical interest in
recovering and reusing CKD to produce cement clinker (described in Chapter 8) is resulting in
the limited application of several CKD recycling technologies. Based upon a preliminary
evaluation (described more fully below), EPA believes that several of these technologies may find
more widespread application in the cement industry during the next few years. The level of
interest in this type of CKD management approach is evidenced by the large number of
additional site-specific engineering evaluations that have been requested of and conducted by the
developers of these technologies.6
As described in Section 9.1, EPA has developed cost estimating equations to calculate
baseline costs reflecting the current waste management practices employed by nine of the 10
facilities in the sample (the tenth generates no net waste), as well as the costs under various
alternative management practices. EPA's current and alternative land disposal cost estimates
reflect the assumption that disposal costs are a function of several variables:
• Quantities of CKD generated, recycled, and disposed;
• Physical and chemical characteristics of CKD;
• Depth to ground water;
• Current ground-water and surface water monitoring practices;
• Location characteristics of the facility;
• Characteristics of CKD waste management units; and
• Remaining useful life of existing CKD waste management units.
Differences in these variables across facilities explain, in large part, differences in results among
facilities that may have comparable CKD generation rates. Similarly, the Agency's costing
functions for CKD treatment and recovery technologies are based upon waste generation rates,
the chemical composition of the CKD entering the recovery process, and in some cases, cement
kiln technology type and local climatic conditions.
9.2.1 Current Practices
The major current practices that are applied to CKD fall into three basic categories,
which are discussed below: 1) direct recycling; 2) off-site beneficial use; and 3) on-site land
disposal. Each of these three approaches to dust management confers economic benefit to or
imposes costs on the facility operators that employ it. EPA's evaluation of alternative CKD
management methods and their costs builds upon an understanding of the current, or baseline,
practices for CKD management that are described in this section.
5 Recent information suggests, however, that the operators of a number of hazardous waste-burning cement plants
have suspended'sales of CKD, pending the outcome of EPA's decision-making process regarding the RCRA status of
this material.
6 For example, a principal of Passamaquoddy Technology has reported conducting 35 such evaluations for his
technology. Personal communication with Garrett Morrison, Passamaquoddy Technology, Inc., July 2, 1993.
-------
9-9
Direct Recycling of Collected Dust
Direct recycling, as described in Chapters 3 and 8, is practiced at a majority of the
operating cement plants in the U.S., at least to some degree. In brief, direct recycling involves
returning CKD to the cement kiln (or raw material storage) directly as an input, without any
treatment or reclamation. Approximately 18 percent of cement plants, or 20 plants nationwide,
recycle all of their CKD.7
As a general matter, as stated in Chapter 8, it is in the facility operator's interest to
remove as little CKD from the kiln system as possible. Nonetheless, there are wide disparities
across the industry in both the quantities and the percentages of gross CKD generated that are
directly returned to the kiln (or raw feed) system. In general, operators of dry kilns tend to
recycle a greater percentage of their gross CKD than operators of wet process systems, and
operators of hazardous waste-burning kilns recycle a lower percentage of their CKD than
operators of kilns not burning hazardous wastes. CKD that is recycled is typically pneumatically
conveyed, or "insufflated" to the flame end of the kiln, where it is reintroduced through or
adjacent to the burner pipe. Alternatively, the collected CKD may be conveyed to raw material
storage (silos or tanks, for dry and wet process kilns, respectively).
One approach to decreasing the amount of CKD removed from the system (and
therefore, destined for disposal) is to reduce the total amount of dust leaving the kiln (i.e.,
decrease the gross CKD generation rate). As discussed in Chapter 8, however, facility operators
are already motivated by process efficiency and cost considerations to limit the quantities of dust
that exit the kiln; most kilns are equipped with chain sections and most operators limit air flow
velocities to reduce turbulence, in order to control excessive dust production. Accordingly,
opportunities for reductions in gross CKD, or total collected dust, in cement kiln systems appear
to be quite limited.
Options for returning the material collected, however, are more numerous. CKD has
significant value as a raw material in cement making, particularly because it has already been
quarried, crushed, ground, blended, and partially calcined. One industry source has indicated
that this material (net CKD) has a value to the cement producer of $4-12 per short ton;8 this
range is consistent with estimates obtained from other industry sources.
In performing this analysis of the impacts of CKD management alternatives, EPA has not
explicitly calculated the baseline cost savings at plants that already recycle their CKD or the
industry-wide benefits of increasing recycling rates, because of data limitations. As stated above,
however, the Agency believes that the average reductions in variable operating costs of increased
recycling are on the order of $9 per metric ton (range of about $4.50 to $13.50 per metric ton) of
CKD recycled, less handling costs. These estimates do not consider the capital and operating
costs associated with installing direct CKD recycling equipment (e.g., for insufflation), nor do
they reflect any off-setting credits associated with the avoided costs of CKD disposal.
7 Based on 18 percent of the sample of 79 plants in the Portland Cement Association survey and 115 plants total
in the U.S.
1 Morrison, G.L., 1993. Passamaquoddy Technology Recovery Scrubber Operations Update and Forecast. July. pg.
7.
-------
9-10
Off-Site Beneficial Use
As discussed above in Chapters 4 and 8, CKD may be used for a number of off-site
beneficial purposes, including stabilizing wastes, fertilizing farmland, and neutralizing waste acids.
In 1990, off-site utilization accounted for approximately 943,000 metric tons of CKD, which was
about 6 percent of the gross CKD and 20 percent of the net (non-recycled) CKD generated in
the U.S., according to PCA Survey data. Approximately 70 percent of the total quantity of CKD
going to off-site beneficial use was used for waste stabilization, including dewatering and
stabilizing municipal sewage sludge and oil sludge. Twelve percent of the CKD used off site was
applied as a combination fertilizer and agricultural liming agent, due to its potassium content and
its alkalinity (which is beneficial where acidic soils are prominent).
EPA has extremely limited data on the prices obtained for CKD destined for off-site
use.9 Only one actual price quote is currently available: Keystone Cement in Bath,
Pennsylvania reportedly sells its CKD for about $10.00 per metric ton plus transportation
costs.10 Based upon observations made during field visits, EPA believes that the operators of
other cement plants sell CKD for a few dollars per metric ton, or give it away. In this chapter
and supporting analyses, the Agency has assumed that cement plants receive a nominal price for
their CKD ($5 per metric ton), because of the availability of low-cost competing materials in
many areas. The net revenues received from CKD sales are assumed to be equal to the sale
price (f.o.b.) because of the minimal handling or processing required for typical off-site uses.
Current Land Disposal Practices
Most CKD that is removed from the kiln system (i.e., is not directly recycled to the kiln)
and is not used off site in a beneficial application is disposed of on land. EPA believes that, in
the absence of new regulatory controls, this would continue to be an important waste
management practice across the industry. Of the 81 facilities responding to the 1991 PCA
survey, 77 percent dispose of some CKD on site. The remaining 23 percent recycle all of their
CKD, or sell all of their non-recycled, or net, CKD. No off-site disposal of CKD has been
reported. Extrapolating from data provided by the PCA Survey respondents to the entire
industry, an estimated 3.8 million metric tons of CKD were land-disposed nationwide in 1990. A
full description of current land disposal practices is provided above in Chapter 4.
Facilities relying on on-site disposal typically dump CKD into an unlined, retired portion
of the limestone quarry associated with the cement plant. Alternatively, they may dump CKD in
large unlined piles at other on-site locations. Only one respondent to the PCA Survey reported
use of a pond, in which the CKD disposal area collects and retains water that covers the CKD.
About one-fourth of the plants reportedly co-dispose CKD with other waste materials, such as
furnace brick, concrete debris, and tires; typically, the co-disposed wastes amount to less than
one percent of the quantity of CKD disposed. In addition, quarry overburden (the earth and
rock removed to reach unmined deposits of limestone and other raw materials) is co-disposed
with CKD at some plants. Across all plants represented in the PCA Survey data, total quantities
of overburden co-disposed with CKD nearly equalled the amount of CKD disposed in 1990.
9 Although EPA has received information on off-site use in response to its 1992 RCRA §3007 request indicating
that the operators of at least 35 plants sold CKD in 1990, and that an additional 31 plant operators either sold or gave
away CKD during that year, none of these responses provided CKD pricing data.
10 Personal communication with Doug Glasford, Keystone Cement, November 24, 1992.
-------
9-11
Increasingly, on-site CKD management practices are being affected by non-RCRA federal
environmental control regulations and standards developed and applied at the state level. As
discussed in Chapter 7, national controls on stormwater run-off have been developed under the
Clean Water Act, and plants in many states are subject to limitations on fugitive dust emissions
from operating and waste management units, including CKD piles. In some instances, state
government agencies have required special controls on CKD management units, to limit
contaminant releases to the environment.
EPA's cost estimates for current baseline CKD land disposal practices for the case study
plants include costs for land, land clearing, heavy equipment, operator labor, utilities, and, as
appropriate, environmental control measures such as dust suppression and run-on/run-off
controls. The bulk of these costs are associated with equipment to convey CKD from the cement
plant to the disposal site and place it in the desired location. For consistency, the Agency has
assumed throughout that CKD disposal would be performed with dedicated equipment, and that
certain more or less fixed costs would be incurred by the operator, irrespective of the CKD
quantity disposed.
Of the nine facilities in the sample used in EPA's cost analysis that rely on on-site CKD
disposal, estimated costs for land disposal ranged from about $83,000 to just under $400,000 per
year; the median cost is about $3.50 per metric ton of CKD. In part because of EPA's
simplifying assumptions, facilities disposing the least CKD have the highest estimated unit
disposal costs.
9.2.2 Alternative Land Disposal Practices
In the event of a change in the RCRA regulatory status of CKD, it is likely that changes
in existing management practices would be required at most plants for regulatory compliance.
These modifications would likely be driven by specific regulatory requirements, which cannot be
precisely defined at this time. EPA has prepared an analysis of the costs of several different
approaches to more stringent regulation of the land disposal of CKD, which are described in this
section. The regulatory framework for these approaches is Subtitle C of RCRA, which provides
for a comprehensive system for the management of hazardous wastes. This section presents
descriptions of and costs associated with three different approaches to land disposal of CKD
within the context of Subtitle C: (1) a conventional Subtitle C scenario in which all existing
program elements are applied; (2) a modified Subtitle C scenario that incorporates the flexibility
in establishing site-specific requirements provided by §3004(x) of RCRA; and (3) a much more
limited approach that might be implemented to control CKD contaminant releases to
environmental media.
Conventional Subtitle C Technology and Administrative Standards
EPA regulations promulgated pursuant to RCRA Subtitle C define stringent "cradle to
grave" management practices that must be applied to hazardous wastes generated and managed
in the United States. Under these regulations, only carefully defined approaches to hazardous
waste management are permissible, and all of these approaches are adapted to the conditions
found at individual hazardous waste management sites through permits. As an inorganic solid
material, only a very few options are available for the permanent disposal of CKD as a hazardous
waste. CKD could be managed for short periods of time in a waste pile, but long term disposal
would require the use of a landfill meeting EPA-specified minimum technology standards.
Accordingly, EPA has identified and categorized all requirements under Subtitle C that might
-------
9-12
have cost implications for the management of CKD in a hazardous waste landfill, including
requirements related to notification, permitting, technical standards for land disposal, monitoring,
closure, post-closure care, financial responsibility, and corrective action for continuing releases
due to past practices; possible Land Disposal Restrictions program requirements have not been
included, due to uncertainties regarding appropriate pre-disposal treatment for this material.
More detailed requirements are described in the Background Document.
• Subtitle C Costs (Exclusive of Corrective Action Costs)
Assuming Subtitle C landfill disposal, annualized costs for eight of the nine facilities with
non-zero net CKD generation range from about $2.4 million to more than $14 million over and
above baseline waste management costs, and average about $6.3 million per facility. The ninth
plant, facility G, generates a relatively small amount of net CKD; its annualized incremental costs
are much lower - about $140,000 per year. For the eight facilities with significant net waste
generation rates, capital costs comprise 50 to 77 percent of the annualized costs. Total capital
costs for each facility range from about $7.8 million to $74 million, except for facility G, which
has an estimated capital cost of only $20,000 because its operator is assumed to send the CKD
off site for disposal, and requires only a temporary storage area. Overall, average capital costs
are $25.8 million for the case study plants. The relative importance of capital costs is a reflection
of the major capital expenditures that would be required to construct on-site Subtitle C secured
landfills for managing CKD.11 Estimated Subtitle C disposal costs for each facility are shown in
Exhibit 9-2.
The highest cost per metric ton of net CKD for Subtitle C disposal is more than $153 per
metric ton for Facility G, which is assumed to send its waste off site because that would be less
costly than constructing an on-site unit, given this plant's low net waste generation rate and scale
economies. Most of the remaining facilities also have costs of more than $100 per metric ton of
CKD. Even at these relatively high unit costs, however, construction of on-site disposal units is
the most cost-effective response for most of the operators of the plants in the sample, because of
scale economies.
Potential Subtitle C Corrective Action Costs
One potentially important and very costly component of regulating CKD under RCRA
Subtitle C is corrective action requirements. Section 3004(u) of the Hazardous and Solid Waste
Amendments of 1984 (HSWA) to RCRA requires permitted Subtitle C facilities to undertake
corrective action for toxic releases to all media, from all solid waste management units (SWMUs)
located on their premises. These requirements would affect all newly-permitted cement plants
under the Subtitle C requirements if (1) they manage (store or dispose) newly generated CKD
on site and (2) they have prior releases from solid waste management units (SWMUs) requiring
cleanup. The following classes of facilities would not be affected by this Subtitle C requirement:
(1) Facilities that presently burn hazardous waste fuels and are already subject to
Subtitle C permit requirements, because cement plants burning hazardous waste
fuels are already subject to facility-wide corrective action requirements (if they
release hazardous constituents to the environment);
11 Major line item costs include those of procuring and installing one clay, two sand, and two synthetic liners,
leachate collection systems, and run-on/run-off controls, as well as site preparation and excavation costs.
-------
9-13
(2) Facilities that send all newly generated waste off site for disposal; and
-------
9-14
Exhibit 9-2
Subtitle C Disposal Costs
Facility
A
B
C
D
E
F
G
H
I
SAMPLE
TOTAL
MINIMUM
MAXIMUM
AVERAGE
Estimated
Current Waste
Management
Cost
(SOOO/YR)
214
104
396
148
128
140
83
118
80
1,411
80
396
157
Subtitle C Costs Incremental to Current Practices
Loss of
Revenue
From CKD
Sales
(SOOO/YR)
87
18
61
0
106
0
0
0
61
333
0
106
37
Annualized
Cost
(SOOO/YR)
10,613
3,182
14382
2,996
8,569
4,095
138
3,958
2379
50312
138
14382
5^90
Total
Capital
Cost
($000)
52495
11,637
74383
10,863
40,749
17471
21
16,736
7,809
232,364
20
74382
25,818
Annualized
Capital Cost
(SOOO/YR)
7,848
1,736
11,099
1,621
6,080
2,622
3
2,497
1,165
34,671
3
11,098
3,852
Annualized
Cost per
MTCKD
(S/MT)
137.2
112.5
63.4
116.3
149.6
60.7
153.3
73.6
144.1
--
60.7
153.3
112.3
Annualized
Cost per
MT
Cement
(S/MT)
27.9
3.0
11.6
9.3
17.8
14.0
0.4
3.7
4.7
--
0.4
27.9
10.3
Note: Current waste management cost is calculated from the quantity of CKD currently being wasted. Regulatory
cost increments to current waste management costs are calculated from the net waste generation rate, which includes
both the quantity of CKD currently sold and the quantity wasted.
(3) Facilities that generate and manage CKD on site but do not have toxic releases to
ground water, soil, air, or surface waters warranting mandated cleanup or control.
Note that facilities in category (2) would not be affected even if they have SWMUs on site that
release hazardous constituents, because such facilities could avoid being brought into the Subtitle
C regulatory system for hazardous waste treatment, storage, or disposal facilities (TSDFs) in the
first instance. For this analysis, EPA has calculated and presented potential corrective action
costs for all nine CKD-generating facilities in the sample for illustrative purposes; as stated
above, the five hazardous waste-burning plants in the sample are already subject to facility-wide
corrective action requirements. In addition, the Agency has assumed that all historically disposed
wastes at each site would require a corrective action response; this worst-case assumption
obviously produces higher estimated costs than likely actual costs if CKD were to be newly
-------
9-15
regulated as a hazardous waste. The purview of this analysis includes both active and inactive
SWMUs at active CKD-generating facilities.12
Based upon data on corrective action strategies and costs for remediating contaminated
media at cement plants and analogous industrial facilities, EPA developed two basic conceptual
approaches to prospective corrective actions at CKD-generating facilities:
(1) Excavation, treatment, and secured disposal of wastes in a waste management unit
(on- or off-site), referred to as Remedial Strategy 1 in this analysis; and
(2) Capping, cap maintenance, and future use restrictions, referred to as Remedial
Strategy 2.
Because of the nature of the waste and contaminants in question, and the philosophy of EPA's
corrective action program, the emphasis in developing these two strategies is on contaminant
source control. Depending upon the severity and areal extent of contamination at individual
facilities, additional steps (e.g., ground-water pumping and treatment) might be required at some
facilities. EPA has not, however, included the costs of such actions in this analysis.
Estimated annualized upper bound corrective action costs at the nine sample plants range
from $108,000 to almost $15 million for Strategy 1, and from $450,000 to $775,000 for Strategy 2.
The wide disparity in estimated costs under Strategy 1 reflects the great differences among the
sample plants with respect to existing CKD (and other waste) quantities (12,900 to 1.8 million
cubic meters). If Remedial Strategy 1 is required at sites in the sample of nine, the annualized
costs of regulatory compliance under Subtitle C could increase by as much as 470 percent over
and above general facility and waste disposal costs; costs at all but one facility could increase by
at least 30 percent. In contrast, if Strategy 2 (involving capping) were to be adopted, upper
bound Subtitle C costs would increase by only 5.2 to 15.2 percent. The Technical Background
Document provides a detailed description of the methods used in this analysis, and presents the
site-specific potential corrective action costs developed under Strategies 1 and 2.
The results of the analysis suggest that corrective action requirements could, at some
facilities, add significantly to the costs of Subtitle C disposal and, at certain plants, exceed all
other costs related to Subtitle C land disposal of future CKD generation. EPA notes that plants
having low net CKD generation rates, and correspondingly low on-site CKD disposal costs, could
have high corrective action (and total compliance) costs, and vice versa, because of the variability
in 1) the quantities of CKD and other wastes historically accumulated at each site, and 2) the
environmental conditions that drive corrective action costs.
With respect to corrective action, it should be noted that the 35 or so cement plants
already permitted (or in the process of being permitted) as hazardous waste burners will already
be subject to facility-wide corrective action, if needed, under Subtitle C of RCRA. Thus,
additional corrective action responsibilities could accrue only to the 85 plants not permitted as
hazardous waste burners. How many of these plants, if any, might require corrective action
would remain to be determined by site-specific studies.
11 Releases from SWMUs at inactive cement plants would be controlled under Superfund.
-------
9-16
Alternative Subtitle C Costs Under RCRA §3004(x)
The Agency has also examined a less costly disposal option that would represent
somewhat less stringent disposal practice requirements under a modified form of Subtitle C.
RCRA Section 3004(x) allows for flexible Subtitle C regulation for hazardous CKD waste, as well
as several other special waste categories, under certain conditions. Under this provision, many
significant RCRA requirements13 may be modified at the Administrator's discretion
"...to take into account the special characteristics of such wastes,
the practical difficulties associated with implementation of such
requirements, and site-specific characteristics ... so long as such
^ modified requirements assure protection of human health and the
environment."
Accordingly, EPA has estimated costs for a "Subtitle C-Minus" alternative, assuming that
on-site CKD disposal would need to meet less stringent technology requirements than under full
Subtitle C, due to site-specific variability in potential risk to ground water. Plants located in
areas with deep ground water and relatively impermeable soils (low risk sites) would be allowed
to continue using their current landfills, while plants located in areas with more vulnerable
ground water would be required to construct new landfills with liners (clay for moderate risk
sites, composite for high risk sites) and leachate collection systems. These less stringent liner and
leachate collection system requirements result in capital cost savings of several million dollars at
most of the case study plants. All plants would, however, still be required to conduct ground-
water monitoring and many other activities mandated by existing standards.
Based upon the case study risk assessments presented in Chapter 6, two of the nine
sample plants considered in this analysis are classified as "high" risk facilities. Annualized
compliance costs for one of these two facilities are about 18 percent less than under full Subtitle
C. The second high risk facility (Facility G), has a very low waste generation rate, and its
operator would face the same land disposal costs under Subtitle C-Minus as it would under
Subtitle C, because it relies on off-site disposal.
Six of the remaining facilities have moderate risk levels, and their costs are about 50
percent (range of 37 to 60 percent) lower than under full Subtitle C.
The difference is even more dramatic for the only low risk plant (facility D), which has
compliance costs that are 78 percent lower. Overall, EPA estimates that disposal costs could
average about 42 percent lower for Subtitle C-Minus than for full Subtitle C disposal. Estimated
costs for the nine case study facilities under Subtitle C-Minus are shown in Exhibit 9-3.
Additional detail regarding the manner in which EPA has computed the costs for the
land disposal alternatives is provided in the Technical Background Document.
13 Specifically, RCRA sections 3004(c) through (g) (land disposal restrictions), (o) (minimum technology standards),
(u) corrective action for continuing releases), and 3005(j) (permitting of interim status treatment, storage, and disposal
surface impoundments) are covered by this provision.
-------
9-17
Exhibit 9-3
Subtitle C-Minus Disposal Costs
Facility
A
B
C
D
E
F
G
H
I
SAMPLE
TOTAL
MINIMUM
MAXIMU
M
AVERAGE
Risk
Level-
Moderate
Moderate
High
Low
Moderate
Moderate
High
Moderate
Moderate
Estimated
Current
Waste
Management
Cost
(SOOO/YR)
214
104
396
148
128
140
83
118
80
1,411
80
396
157
Subtitle C-Minus Costs Incremental to Current Practices
Loss of
Revenue
From CKD
Sales
(SOOO/YR)
87
18
61
0
106
0
0
0
61
333
0
106
37
Annualized
Cost
($000/YR)
4,763
1,993
11,817
671
4,096
2,185
138
2,251
1,536
29,450
138
11,817
3,272
Total
Capital
Cost
($000)
19,438
4,988
59,834
662
15,438
6,822
21
7,142
3,117
117,462
21
59,834
13,051
Annualized
Capital Cost
(SOOO/YR)
2,900
744
8,928
99
2308
1,018
3
1,065
465
17430
3
8,928
1,947
Annualized
Cost Per
MTCKD
($/MT)
61.6
70.5
52.1
26.1
71.5
32.4
152.5
41.9
93.1
--
26.1
152.5
66.9
Annualize
d Cost Per
MT
Cement
($/MT)
12.5
1.9
9.6
2.1
8.5
7.5
0.4
2.1
3.1
--
0.4
12.5
5.3
Note: Current waste management cost is calculated from the quantity of CKD currently being wasted. Regulatory
cost increments to current waste management costs are calculated from the net waste generation rate, which includes
both the quantity of CKD currently sold and the quantity wasted. ,
• As documented in the site-specific hazard potential analyses presented in Chapter 6.
Tailored Contaminant Release Controls
A less stringent alternative to the very complex and costly technical and administrative
requirements associated with even a flexible Subtitle C approach could consist of tailored
upgrades to existing land disposal units. Under this approach, the objective would be to employ
site-specific contaminant release controls to ensure that CKD and its constituents were not
released to adjacent environmental media, and hence, would not migrate to potential
environmental and human receptors.
Based upon the results of the risk analysis presented in Chapter 6 of this report, the
primary potential risk pathways of concern for most plants are fugitive dust that might result in
-------
9-18
CKD deposition on crop and grazing land, and human health and ecological risk from
stormwater run-off releases to fields and surface waters from disposal piles. Though perhaps of
less frequent concern, there is also the possibility of ground-water contamination associated with
the disposal of CKD in sub-grade units in areas of shallow ground water or under fractured-flow
conditions.
EPA thus estimated costs for upgrading existing active land management units at the case
study facilities examined in this chapter. Many possible types and degrees of upgrading could be
considered under various future regulatory scenarios. For purposes of providing an illustrative
example, however, the Agency estimated costs for just one set of typical upgraded practices,
consisting of the following set of contaminant release control elements:
• Fugitive dust emission controls;
• Run-on/run-off controls;
• Ground-water monitoring;
• Waste pile capping at unit closure;
• Post-closure care; and
• Costs related to engineering studies and permitting.
Fugitive dust emission controls consist of water lines that are installed around the
perimeter of a waste management unit; these lines are equipped with spray nozzles that are used
to wet the material inside the unit on a periodic basis. Run-on/run-off controls are comprised of
drainage ditches and culverts that are installed around the perimeter of the unit, and pipes and
pumping that are employed to convey stormwater away from the unit. Ground-water monitoring
systems involve single wells, placed at 200 foot intervals, around one-half (i.e., a down-gradient
edge) of the perimeter of the unit. The screen depth is the midpoint of the aquifer. Ground-
water sampling and analysis are conducted quarterly from all wells in the system. Waste pile
capping involves regrading the deposited material, as necesjsary, then installing a two foot thick
soil cap and planting grass to stabilize the cover material.14 Post-closure care under this
alternative consists of continued ground-water monitoring, maintenance of run-on/run-off
controls and the integrity of the soil cap (through mowing and fertilizing the grass), and site
security. Finally, costs related to engineering studies and permitting are assumed by EPA to be
$250,000, all of which is incurred in Year 1.
These controls are scaled to the predicted size of a facility's waste management unit, as
determined through EPA's estimates of current waste management costs. Facilities that currently
employ one or more of these practices would bear no additional cost for the corresponding
program element. For example, plants that currently monitor ground-water quality would
experience no additional costs for this requirement. The resulting cost estimates for this type of
tailored approach are presented in Exhibits 9-4 and 9-5 for the nine relevant case study cement
plants.
14 Waste management unit capping at facilities in arid areas (e.g., New Mexico) involves placement of a rock cap
ler than a soil cap.
-------
9-19
Exhibit 9-4
Tailored Contaminant Release Controls Costs:
Continued Sales of CKD for Off-Site Use
Facility
A
B
C
D
E
F
G
H
I
SAMPLE
TOTAL
MINIMUM
MAXIMU
M
AVERAGE
Estimated
Current
Waste
Management
Cost
($000/YR)
214
104
396
148
128
140
83
118
80
1,411
80
396
157
Loss of
Revenue
From CKD
Sales
(SOOO/YR)
0
0
0
0
0
0
0
0
0
0
0
0
0
Costs Incremental to Current Practices
Total
Capital
Cost
($000)
301
319
370
271
287
314
263
2,032
427
4,584
263
2,032
509
Annualized
Capital Cost
(SOOO/YR)
45
48
55
40
43
47
39
303
64
684
39
303
76
Annualize
d Capital
and O&M
Costs
(SOOO/YR)
97
85
157
59
83
100
51
364
83
1,079
51
364
120
Annualize
d Cost Per
MTCKD
($/MT)
1.6
3.4
0.7
2.3
2.3
1.5
56.7
6.8
18.9
--
0.7
56.7
10.5
Annualized
Cost Per
MT
Cement
($/MT)
0.26
0.08
0.13
0.18
0.17
0.34
0.13
0.34
0.17
--
0.08
0.34
0.20
-------
9-20
Exhibit 9-5
Tailored Contaminant Release Controls Costs:
Curtailed Sales of CKD for Off-Site Use
Facility
A
B
C
D
E
F
G
H
I
SAMPLE
TOTAL
MINIMUM
MAXIMU
M
AVERAGE
Estimated
Current
Waste
Management
Cost
(SOOO/YR)
214
104
396
148
128
140
83
118
80
1,411
80
396
157
Loss of
Revenue
From CKD
Sales
(SOOO/YR)
87
18
61
0
106
0
0
0
61
333
0
106
37
Costs Incremental to Current Practices
Total
Capital
Cost
($000)
308
322
373
271
296
314
263
2,032
529
4,707
263
2,032
523
Annualized
Capital
Cost
($000/YR)
46
48
56
40
44
47
39
303
79
702
39
303
78,
Annualized
Capital
and O&M
Costs
(SOOO/YR)
104
87
160
59
94
100
51
364
110
1,125
51
364
125
Annualized
Cost Per MT
CKD
(S/MT)
1.34
3.07
0.70
2.28
1.65
1.48
53.38
6.78
6.70
--
0.70
53.38
8.60
Annualized
Cost Per
MT
Cement
($/MT)
0.27
0.08
0.13
0.18
0.20
0.34
0.13
0.34
0.22
-
0.08
0.34
0.21
Under this set of improvements to current land disposal practices and assuming
continued sale of CKD for off-site use at current levels, annualized CKD management costs
increase by 40 to 300 percent, ranging from $51,000 to more than $360,000 per year. For the
majority of plants, annualized CKD land management costs would increase by 40 to 80 percent.
Capital costs for installing these release controls would generally fall in the $200,000 to $400,000
range, though one facility (H) would face new capital requirements of over $2.0 million because
of its location in a flood plain. On a unit basis, incremental costs average about $10 per metric
ton of CKD (within a range of about $1.00 to $53) and about $0.20 per metric ton of cement,
ranging from $0.08 to $0.34 per ton of cement.
Using an alternative (extreme) assumption that all sales of CKD for off-site use would be
curtailed (e.g., due to regulatory changes), costs for these controls increase somewhat for the five
cement plants in the sample that reportedly sell CKD (Exhibit 9-5). Cost increases are due both
to the loss of revenue from CKD sales and to the need to dispose of larger CKD quantities. The
-------
9-21
range of annualized costs under this variant remains the same as in the previous case, but the
average impact increases from $120,000 to just under $125,000 per year. Effects on capital
requirements are relatively modest, and unit impacts, expressed as the annualized cost per metric
ton of CKD, are negligible for most of the plants studied.
In the event that the continued use of the existing CKD management unit(s) resulted in
release of contaminants to ground water (as determined by the quarterly ground-water sampling
and analysis required under this alternative), corrective action would be necessary. As shown in
the Subtitle C corrective action cost analysis presented above, the most cost-effective means of
controlling releases to ground water is generally waste management unit capping with an
impermeable cover to control further leachate formation within the unit. These costs would be
incurred during a single year, and represent the expense of engineering and installing a
composite liner and top soil layer on the entire CKD waste pile at its predicted maximum size
(i.e., assuming 15 years of waste accumulation).15
Based on this approach, the estimated corrective action cost for any of the case study
plants that might require corrective action range from just over $100,000 to about $2.2 million.
The average per-facility cost is about $1.36 million assuming continued sales of CKD for off-site
use at 1990 levels, or about $1.45 million if CKD sales were curtailed. Six of the nine case study
plants fall within the range of $1.35 million to $1.8 million.
The Agency assumes that only a relatively small number of plants would face new to
corrective action responsibilities under this scenario (and, as previously noted, these
responsibilities would apply only to currently active land placement units). Only the 50 or so
plants currenly generating net waste and managing it on site (and that are also not hazardous
waste burners already subject to corrective action) could have new corrective action
responsibilities under this costing scenario. In addition, of these 50 plants, only a fraction would
be likely to have ground water contaminant releases requiring correction (based on results of the
ground water risk pathway analyses reported in Chapter 6). At $1.4 million per plant, if 10 to 20
plants out of the 50, for example, were to require such corrective action, total capital costs would
be on the order of $14 to $28 million for the industry as a whole.
Overall, even with corrective action for ground-water releases, this tailored upgrading of
existing units is far less costly than either the full Subtitle C or the Subtitle C-Minus land
management standards.
9.23 Alternative On-Site CKD Recycling and Recovery Techniques
There are several available alternatives to the on-site disposal of CKD. Some build upon
practices that are already in widespread use, while others rely upon unconventional methods to
chemically treat CKD so that it may be converted into useful products. The more prominent of
these alternative approaches are discussed below.
Increasing Direct CKD Recycling
u Unlike typical Subtitle C corrective action provisions, EPA has assumed here that corrective action would affect
only the units being employed to manage currently generated and disposed CKD. Other SWMUs at a cement plant
would be unaffected.
-------
9-22
Despite the clear economic incentive to recycle as much of the collected (gross) CKD as
possible, there are several factors that can limit the ability of the kiln operator to directly reuse
this material. The primary limitation appears to be buildup of alkalis (sodium and potassium)
and sulfur in the recirculating dust load and in the clinker. As discussed at length in Chapters 3
and 8, a large percentage of plants in the U.S. must meet the ASTM standard of 0.6 percent or
less alkali in most or all of their product.16 The conventional method for complying with this
limitation has been to periodically remove ("bleed") CKD from the system as a waste or by-
product. An emerging alternative is to selectively reformulate raw materials input combinations
to yield total alkali concentrations within acceptable product limits, despite continuous recycling
of CKD. EPA's research indicates that at least some facilities have been able to recycle all of
their CKD on a continuous basis in this manner through the selective use of high purity raw
materials, often purchased from off-site sources.17 At present, however, there is insufficient
information either to assess the extent to which the alkali limit actually influences CKD
generation across the industry, or to evaluate the national potential of raw material substitution
as a feasible low-cost means of reducing or eliminating net CKD generation or its constituent
levels.
Other in-plant factors that may limit direct CKD recycling can include reliance by the
facility operator on CKD return methods (e.g., insufflation) that cannot accommodate the entire
CKD stream generated by the facility, or on mechanical systems that are incapable of
accommodating fluctuations in CKD generation rates, i.e., have no surge capacity. For example,
cement plants with relatively high total dust collection rates may not be able to recycle all of this
CKD through insufflation without reducing the kiln flame temperature beyond limits that would
adversely affect product (clinker) quality. The operator of one of the facilities visited during
EPA's 1992 CKD sampling program reportedly disposes of about 50 percent of its total collected
CKD solely for this reason.18 EPA has not been able to determine the extent to which such
considerations currently limit CKD recycling across the domestic cement industry, and therefore
is not in a position to predict trends or the magnitude of waste reduction opportunities
associated with overcoming these types of engineering problems.
Innovative CKD Recovery Technologies
As an alternative to the predominant CKD management practices currently in use,
several technologies have been developed for recovering the values contained within this
material. In general, these techniques both recover the lime, silica, and other components that
are used to produce cement clinker, and produce a residue containing relatively high
concentrations of alkali salts that may have value as a fertilizer. The Agency has identified at
least three innovative technologies for treating and recovering CKD that would otherwise be
wasted. These technologies are examined here in detail because they appear to be promising
" Types I and II Portland cement (which must comply with the 0.6 percent alkali limit) comprise the vast majority
of Portland cement, and Portland cement comprises the vast majority of hydraulic cement, produced and used in the
U.S.
17 For example, Calaveras Cement Company is able to continuously recycle all of the CKD generated at its
Tehachapi, CA plant through the use of a low-alkali ("sweetener") sand purchased from an off-site, local source; this
low alkali sand counterbalances the relatively high alkali content in the native limestone. (Source: Personal
communication with Lars Oberg, Calaveras Cement Company, Tehachapi, CA, May 20, 1993.)
18 Persona] communication with Brian Graf, ESSROC Materials, Inc. (Logansport, IN), March 17, 1992.
-------
9-23
from the standpoint of both technical and economic feasibility and pollution prevention potential.
The three technologies are as follows:
1. Alkali leaching;
2. Fluid bed recovery (Fuller process); and
3. Recovery scrubbing (Passamaquoddy Technology process).
Each of these processes is based upon the premise that by removing some or most of the alkali
salts contained in CKD, the treated CKD can either be returned to the clinker production
process or manufactured into clinker directly. The alkali salts (primarily potassium sulfate), in
turn, can then be sold (with or without further purification) for their fertilizer value. Each
technology employs a different approach for separating the alkalis from the CKD, and each
produces a somewhat different primary treatment residue (by-product). One of the technologies,
the Passamaquoddy Technology recovery scrubber, also reportedly confers a number of other
process cost savings and new revenue streams.
Using information collected from the published literature, site visits, and extensive
interviews with the principals involved in developing these technologies, the Agency has
developed costing equations covering the major capital equipment and O&M cost items, as well
as the operating savings and by-product revenues, associated with implementing each of the three
.technologies. These equations and the assumptions upon which they are based are presented in
the Technical Background Document.19
Because these technologies are not in widespread use and because of the variability in
potential input and by-product market conditions across the country, EPA has constructed both
"high value" and "low value" cases. In the high value case, all potential savings and by-product
revenues are received by the facility operator, while in the low value case, only the most certain
benefits of installing the technology are realized; other, more market-driven benefits are assumed
to be zero or negative (i.e., impose costs). As a result, the Agency's predicted overall costs
•(benefits) of installing and operating these innovative technologies should be interpreted as
ranges rather than point estimates.
In addition, to assess the cost-effectiveness of larger versus smaller CKD feed rates the
Agency estimated costs twice for each technology, once with units sized for the reported gross
CKD generation rate, and once using the units sized for net CKD generation rate. Despite
reported scale economies associated with the technologies, EPA's analysis suggests that in
general, these recovery processes are most economically applied using the net CKD generation
rate as the input.20 Another assumption made in this analysis is that the facility operators
would find it more cost-effective to feed CKD that is currently being sold for off-site use to these
alternative recovery technologies. That is, EPA assumed that the benefits of recovering the raw
19ICF Incorporated, op cit.
20 One key reason for this outcome is that in EPA's analysis, CKD that is currently recycled confers no
incremental raw material value, because this value has already been captured by the facility operator. Consequently,
the facility operator choosing to treat the gross CKD stream receives new raw material credits only for the net portion
of this stream, even though the equipment and operating expenses are scaled up to process the entire gross CKD
quantity. Using this set of assumptions, the operator would choose to treat the gross stream only if the incremental
by-product credits were sufficient to offset the additional capital and O&M costs or if the net CKD stream was of an
insufficient quantity to support a particular technology.
-------
9-24
mix values (at $9 per metric ton) coupled with the possible additional benefits from by-product
revenues would outweigh the lost revenues associated with CKD sales (at $5 per metric ton) and
the additional costs of scaling up the CKD processing equipment to handle the larger material
volume.
EPA made a number of additional assumptions. The more important ones are as follows:
1) the three technologies are readily available, that is, they would be freely licensed and/or
installed by their developers, i.e., would be available to all domestic cement plant operators;21 2)
CKD that is treated and returned to the kiln system has a value of $9/metric ton in all cases; 3)
water used, saved, or recovered has a value of $1/1000 gallons; and 4) CKD recovery processes
are operated 24 hours per day, 330 days per year.
Other important costing assumptions apply differently to the high value and low value
cases. For the high value case, the following additional assumptions have been applied: 1)
potassium sulfate has a value of approximately $220/metric ton; 2) by-products containing
significant concentrations of potassium sulfate have a value that is directly proportional to their
potassium sulfate concentration; and 3) by-products are marketable throughout the region
surrounding each cement plant, i.e., the entire quantities produced can be sold at the estimated
price. For the low value cases, EPA has assumed that between two to 10 percent of the
incoming CKD that is removed must be disposed and is sent to secure disposal at an off-site
commercial landfill (either for non-hazardous solid wastes or for RCRA Subtitle C hazardous
wastes), at a cost of about $50 and $277 per metric ton (including transportation), respectively.22
Moreover, other sources of revenue related to additional by-products or services that apply to
particular processes are assumed to be unavailable to the plant operator.
Finally, it is worthy of note that two of the technologies, the alkali leaching system and
the Passamaquoddy Technology recovery scrubber, have to date been applied only to wet process
kiln systems. In this analysis, EPA has assumed that with additional expenditures (e.g., for a
rotary dryer and ancillary equipment), operators of dry process kilns would be able to adapt
these two technologies to their own operations without significant technical difficulties.
Accordingly, the Agency has calculated and presented data from the application of each
technology to each plant in the analysis, irrespective of kiln type, and has included, where
appropriate, the incremental capital and O&M costs associated with the necessary additional
equipment.
Other basic design conditions for the application of the three technologies to our sample
of nine net CKD-generating cement plants are summarized in Exhibit 9-6.
Alkali Leaching
The alkali leaching process is the simplest of the three technologies considered in this
section; the process involves combining CKD with water at a ratio of about 1:5, agitating the
mixture, allowing the leached CKD solids to settle, then recycling this slurry (muds) to the
21 It should also be noted that EPA has no information on the likely licensing and/or royalty arrangements and
fees that would be required to install these technologies. Accordingly, costs associated with these arrangements have
not been included in this analysis, even though they might be non-trivial.
n For purposes of discussion in this chapter, only the results for Subtitle C disposal of the recovery residues are
presented. The intermediate case results may be found in the Technical Background Document.
-------
9-25
process and removing the liquid fraction for concentration and eventual sale as a liquid fertilizer.
When installed at dry process kilns or plants, muds from the leaching process must be dried and
stored in a raw feed silo, for return to the kiln to produce clinker. In wet process kilns the
underflow
-------
9-26
Exhibit 9-6
Key Design Conditions for the Nine Case Study Cement Plants
Sample
Plant
A
B
C
D
E
F
G
H
I
Process
•type
Wet
Dry
Wet
Dry
Wet
Wet
Wet
Dry
Dry
Net Annual
Evaporation
Rate
(Inches/Yr.)
15
<0
<0
50
<0
15
<0
<0
<0
Estimated
Clinker
Capacity
(Metric Tons)
449,922
1,026,384
1,179,230
430,873
533,375
346,868
511,604
1,023,209
544,260
Clinker
Capacity
Utilization
(Percent)
80.4
97.4
99.8
71.6
86.1
80.4
71.4
99.8
87.6
Estimated Annual
Clinker Production
(Metric Tons/Yr.)
361,737
999,698
1,176,872
308,505
459,236
278,882
365,286
1,021,162
476,772
Percent
K2Oin
CKD1
3.2
4.7
4.7
13
4.7
4.7
4.7
2.9
4.7
* Facility-specific average calculated from daia provided in response to EPA's 1992 RCRA §3007 request for plants A, D, and H; for
the other plants, industry average calculated from all available responses.
slurry is either mixed with the feed slurry or pumped into the kiln through a pipe, parallel to the
kiln feed. In the high value case, the liquid (potash) fertilizer solution is sold at a price of $15
per metric ton, while in the low value case, this material is not saleable and must be evaporated
to the point at which it can be handled as a dry sludge, then disposed in a landfill (both Subtitle
C and non-hazardous waste landfills are considered). In any event, about 90 percent of the
original CKD is returned to the process as kiln feed, and the remaining 10 percent is either sold
or disposed.
The leaching process technology has been in use for at least 20 years; two cement plants
(at Inkom, ID and Dundee, MI) currently use the technology, which has enabled the operators of
these two facilities to eliminate on-site CKD disposal. As discussed in Chapter 8, the operator of
the Inkom facility (Ash Grove Cement) has been operating its leaching process for many years
and has been selling its potash solution to a local agri-chemicals dealer throughout this period.23
The cost estimation for the alkali leaching process considers (1) the annual quantity of
CKD fed to the process, (2) the facility's baseline waste management costs, (3) the K2O
concentration of the CKD, (4) whether the plant uses the wet or a dry process, and (5) the
annual evaporation rate where the plant is located. The capital expenditures required for alkali
leaching include the costs of procuring and installing the following equipment:
23 Personal communication with Craig Southworth, Ash Grove Cement Company, May 20, 1993.
-------
9-27
• A dust elevator, a leaching tank with a slow-moving agitator, pumps, and piping
(all facilities);
• Either a mechanical evaporator or an evaporation pond (if the annual site-specific
evaporation rate is high enough); and
• For dry process kilns only, a leached slurry (muds) dewatering system to dry the
recovered dust prior to reentry to the kiln.
• Under the low value case, where the potash solution cannot be sold as a fertilizer,
an additional evaporator and dewatering system are required to concentrate the
potash solution to a sludge for disposal.
Capital costs are annualized and added to annual operating and maintenance expenses
for an estimate of total costs. Annual savings are estimated for (1) the amount of raw feed for
the kiln that is replaced by recycled CKD, and (2) the elimination of disposal costs for CKD that
is no longer land disposed. In the high value case, annual income also is estimated for fertilizer
(potash solution) sales. Estimated annual costs, savings, and income (if any) are combined to
obtain an estimate of the annual net cost or net benefits from utilizing the process. For each
facility, estimates based on processing both gross CKD and net CKD are compared; and the
facility is assumed to select the volume with the lower net costs (or, the higher net income). The
results of this exercise are presented in Exhibit 9-7, which displays EPA's costing results using the
high value and low value cases. Detailed results of these cases, as well as the intermediate case,
are presented in the Background Document.
Under the high value case where the operator can find markets for the potash solutions,
the alkali leaching system yields benefits to five of the nine plants. The estimated annualized
benefit for these five plants ranges from about $68,000 to more than $1.6 million, and averages
$563,000 per year. The remaining four plants are predicted to experience net costs of $150,000
to just over $200,000 annually. For most plants, the better point of application of this technology
appears to be to the net CKD stream, as estimated net benefits are higher for net CKD than for
gross CKD at all but one of the sample facilities, and the difference for this plant is small (less
than four percent), given the level of resolution of this analysis. For the five plants experiencing
economic gain, these results suggest an average net benefit of almost $20 per ton of CKD and
close to $1.00 per ton of cement product. The impacts on the four plants predicted to
experience increased costs range from about $3.75 to $10.50 per ton of CKD processed, or from
$0.15 to $0.46 per ton of finished cement.
Using more pessimistic assumptions about the marketability and regulatory status of the
by-product potassium sulfate solution from alkali leaching, the predicted economic benefits are
reduced substantially. If the by-product solution is not marketable and instead must be
concentrated to a sludge, dewatered, and disposed in an off-site Subtitle C landfill (e.g., if the
material exhibited characteristics of hazardous waste or if the Subtitle C derived-from rule were
to apply), the technology is profitable for only one of the nine plants in the sample. For the
other eight, estimated cost increases exceed $3.5 million annually at one facility, $1 million
annually at four, and $500,000 at two others. Costs per ton of CKD at these eight plants range
from about $16 per metric ton to more than $38 per metric ton, while costs per metric ton of
cement range from $0.85 to $4.19.
-------
9-28
Exhibit 9-7
Estimated Incremental Net Costs for the Alkali Leaching System
Case
High
Value
Low
Value
(Subtitle C
Disposal
of
Residual
Alkalies)
Sample
Facility
A
B
C
D
E
F
G
H
I
Wtd.
Avg.,
Haz.
Waste
Burners
Wtd.
Avg.,
All
Facilities
A
B
C
D
E
F
G
H
I
Wtd.
Avg.,
Haz.
Waste
Burners
Wtd.
Avg.,
All
Facilities
Burns
Hazardous
Waste
Yes
No
Yes
No
No
Yes
No
Yes
Yes
Yes
No
Yes
No
No
Yes
No
Yes
Yes
Total
Capital
Costs
($000)
769
1,642
2384
1458
639
732
51
2353
1,215
1,491
1,260
1,664
2,161
3,495
2,052
1399
1^37
132
3,089
1,604
2,278
1,904
Total
Annual
O&M and
Annualized
Capital
Costs
($000/Yr.)
406
492
1,109
455
359
332
23
697
390
587
474
2,036
1,142
5371
1,052
U94
1,756
62
1,862
794
2364
1,741
Total
Annual
Savings
and
Income
($000/Yr.)
822
337
2,754
304
633
747
91
496
216
1,007
711
661
251
1,706
282
459
530
88
398
166
692
505
Annual
Net Cost
($000/Yr.)
-416
155
-1,645
150
-274
-414
-68
201
175
-420
-237
1375
891
3,665
770
1,135
1,226
-27
1,465
628
1,672
1,237
Net
Cost/
Metric
Ton of
CKD
($/MT)
-5.38
5:49
-4.77
5.83
-4.78
-5.81
-75.08
3.74
10.57
-3.72
-3.16
17.77
31.51
16.16
29.90
19.81
18.18
-29.38
27.23
38.03
18.92
20.08
Net
Cost/
Metric
Ton of
Cement
($/MT)
-1.10
0.15
-1.33
0.46
-0.57
-1.41
-0.18
0.19
0.35
-0.60
-0.37
3.62
0.85
2.97
2.38
2.35
4.19
-0.07
1.37
1.25
2.40
1.95
-------
9-29
Because the alkali leaching technology is the least complex of the three alternatives
examined, it has the lowest capital costs for equipment and installation. Total capital investment
costs range from less than $1 million to about $2.4 million in the high value case, even at dry
process plants, which require substantially more equipment than wet process plants to adopt this
technology. The average annualized capital cost for the sample of nine plants is just over
$181,000 in the high value case. Under the low value cases, where additional equipment is
required to concentrate the potash solution and dry the sludge adequately for land disposal, total
and annualized capital investment requirements more than double at several plants and on
average increase by about 50 percent.
For the five plants at which this technology yields a positive return under the high value
case, annual return on total investment ranges from 54 percent to 133 percent. The average net
annual return across the nine case study plants is 18.8 percent under the high value case.
Fluid Bed Dust Recovery
The Fuller Company's fluid bed pocess does not remove alkalis from CKD to prepare it
for ^introduction to the kiln, but instead thermally treats the dust to produce clinker directly
from the CKD, while concentrating the alkalis into a by-product stream for sale as a fertilizer or
disposal as a processing residue. The fluid bed process can be designed to thermally treat either
gross or net CKD. In the fluid bed process, CKD is pelletized and calcined into clinker on a
fluid bed instead of in a typical rotary kiln; clinker yields are on the order of 60 percent of the
CKD treated. The fluid bed process also produces a by-product material representing about 10
percent of the original input CKD volume. The remaining 30 percent of the incoming CKD is
removed in the form of exhaust gases (ignition loss).24 The fluid bed process has been
demonstrated only on a pilot scale, though several evaluations are underway worldwide. A
detailed description of the fluid bed process may be found in Chapter 8.
In addition to the initial purchase and installation costs of the fluid bed reactor itself, the
required capital expenditures include the costs of procuring and installing feed tanks, a pug mill,
pelletizers, a rotary dryer, a roll crusher, a screen, dust collectors, a surge bin, a heat exchanger,
a fluid bed cooler, a spray tower, fans, piping, and pumps. Because this technology produces
cement clinker directly rather than a treated CKD slurry, there is no difference in required
equipment or cost when applied to dry versus wet process kilns. Annual savings are estimated
for the elimination of disposal costs for CKD that is no longer land-disposed, and annual income
is estimated for the sale of clinker that is produced by the process. Under the high value case,
the operator receives additional revenue from the sale of the by-product (fertilizer) dust. Under
the low value case, this material is disposed in an off-site landfill (as a Subtitle C-reguIated
waste). Costing results for the high value and low value cases are displayed in turn in Exhibit 9-
8.
Under the high value case, application of the fluid bed CKD recovery system yields
substantial economic benefits for one of the nine facilities. The most cost-effective level of
application is the net CKD stream, though two plants produce too little net CKD to use net
CKD as the input, based on minimum fluid-bed technology sizing requirements. Because
application of the technology to the gross CKD stream would impart net costs rather than
24 Ignition loss represents the thermal decomposition of hydrates and carbonates in untreated CKD to form solid
oxides (e.g., CaO, MgO) that comprise the clinker and CKD, and H2O and CO2that are emitted as stack exhaust
gases.
-------
9-30
benefits and because other available technologies could impart net benefits at these plants, EPA
has assumed that the operators of plants G and I would choose not to install the fluid bed
reactor system. For the
-------
9-31
Exhibit 9-8
Estimated Incremental Net Costs for the Fuller Fluidized Bed System
Case
High Value
Low Value
(Subtitle C
Disposal of
Residual
Alkalies)
Sample
Facility*
A
B
C
D
E
F
H
Wtd.
Avg.,
Haz.
Waste
Burners
Wtd.
Avg.,
All
Facilities
A
B
C
D
E
F
H
Wtd.
Avg.,
Haz.
Waste
Burners
Wtd.
Avg.,
All
Facilities
Burns
Hazardous
Waste
Yes
No
Yes
No
No
Yes
Yes
Yes
No
Yes
No
No
Yes
Yes
Operating
Days Per
Year
330
253
330
231
330
330
330
330
305
330
253
330
231
330
330
330
330
305
Total
Capital
Costs
($000)
9,641
5,090
24307
5,090
7,446
8^67
7,052
12^92
9^99
9,641
5,090
24307
5,090
7,446
8,567
7,052
12392
9,599
Total
Annual
O&Mand
Annualized
Capita]
Costs
(SOOO/Yr.)
2,630
1,163
6,529
1,094
2,058
2,287
1,883
3332
2,520
4,039
1,569
10,661
1,430
3,100
3,515
2,861
5,269
3,882
Total
Annual
Savings
and
Income
(SOOO/Yr.)
2,454
748
6,962
685
1,787
2,093
1,676
3,293
2344
1,806
569
5,063
535
1307
1,528
1,225
2,406
1,719
Annual
Net Cost
($000/Yr.)
176
415
-434
409
270
194
207
36
177
2,234
1,000
5,597
895
1,793
1,987
1,635
2,863
2,163
Net
Cost/
Metric
Ton of
CKD
($/MT)
2.27
14.67
-1.91
15.90
4.72
2.87
3.85
0.34
2.31
28.87
35.37
24.68
34.75
31.29
29.46
30.41
26.93
28.21
Net
Cost/
Metric
Ton of
Cementb
($/MT)
0.46
0.40
-0.35
1.26
0.56
0.66
0.19
0.05
0.26
5.88
0.95
4.53
2.76
3.72
6.79
1.53
3.84
3.13
' Facilities G and I do not generate sufficient quantities of net CKD to meet the minimum practical scale for a commercial scale fluid
bed system.
b Includes incremental cement clinker production arising from operation of the fluidized bed system.
plant at which application of the fluid bed recovery technology confers estimated benefits, cost
savings amount to about $434,000 annually, or 35 cents per metric ton of cement. Net costs for
the remaining six plants are in the $175,000 to $410,000 range, suggesting a unit cost of $2.25 to
almost $16 per metric ton of CKD processed, and $0.19 to $0.66 per metric ton of cement.
-------
9-32
Under the low value case, the process would not produce net savings for any plant in the
sample. Because of the relatively large quantity of by-product generated by this process (10
percent of the incoming CKD quantity), Subtitle C regulation of the process residue would
increase net costs dramatically. Typical and average plant costs would increase by about 10 times
(to more than $2.1 million annually), and unit costs would approach or exceed $4 per metric ton
of cement at half of the plants in the sample.
Estimated total capital costs average almost $9.6 million. Under the high value case,
total and annualized capital costs are on average almost eight times higher for the fluid bed
reactor system than for the alkali leaching system. Under the low value case, this gap narrows to
a factor of five, because in contrast to the alkali leaching system, no additional equipment would
be required for the fluid bed system if the by-product had to be disposed. At the one plant for
which the estimated net benefit of installing this technology is positive, the net annualized return
on total invested capital is 1.8 percent.
EPA's research suggests that the Fuller Company process is unique among the CKD
treatment technologies in that it could be constructed as a stand-alone facility to receive CKD
from other sources and produce cement clinker and by-product fertilizer. Such an arrangement
would not require capital investment on the part of the cement company and it would not
directly affect the production process at the CKD-generating cement plant. EPA has not
formally analyzed this possibility because no examples are available to provide the necessary data.
Nonetheless, the developer of this technology believes that this concept could be economically
viable if applied in a suitable location, and has conducted preliminary evaluations of this idea.25
The Passamaquoddy Technology Rue Gas Desulfurization Process
The flue gas desulfurization process, or recovery scrubber, developed by Passamaquoddy
Technology, Inc. with support from the U.S. Department of Energy's Clean Coal Technology
Program, reportedly enables all CKD to be recycled as kiln feed by removing alkalies, chlorides,
and sulfates from the dust. The recovery scrubber produces potassium sulfate fertilizer crystals
(at a rate of about two percent of CKD processed, by weight) as well as reusable cement kiln
feed (the remaining 98 percent), and reportedly discharges only scrubbed exhaust gases (and
internally consumed distilled water). The process also may produce several additional income
streams. First, the process can accommodate alkaline ashes of various types as feedstock
materials; the operator of a recovery scrubber may be able to earn tipping fees from ash
generators. Second, because reported flue gas scrubbing efficiency is on the order of 90-95
percent, the facility may be able to burn higher sulfur coal (at lower cost) than might be
permitted otherwise, under SO2 emissions limits. Moreover, if EPA should expand its SO2
emission allowance trading program to include industrial facilities as well as public utilities,26
cement plants equipped with the recovery scrubber could conceivably sell emission allowances on
the open market. Because some of these prospective benefits are related to the presence of
significant amounts of sulfur in the kiln combustion gases, this technology might not be suitable
for application to the small percentage of U.S. cement plants that do not rely upon coal for at
25 Personal communication with Sidney Cohen, The Fuller Company, July 20, 1993.
M Keynote Address by Carol Browner, EPA Administrator, at the Clean Air Marketplace 1993. September 9,
1993.
-------
9-33
least part of their energy needs.27 A detailed description of the recovery scrubbing process may
be found in Chapter 8.
This system has been installed and is operating at the Dragon Products, Inc. cement plant
in Thomaston, ME. According to a representative of Passamaquoddy Technology, this plant is
recycling all of its newly generated CKD, is producing high-purity potassium sulfate crystals, and
has recently signed contracts for receipt of alkaline ash generated off site, which will be fed to
the recovery scrubber system in the same manner as the CKD.28 Dragon Products anticipates
receiving a $30 per ton tipping fee for accepting this material.
The Passamaquoddy Technology recovery scrubber is by far the most complex of the
recovery technologies examined in this chapter. The capital equipment and associated costs of
the technology can be grouped into major functional categories, as follows:
• Gas handling equipment (duct work and fan);
• CKD processing equipment (mixing tank, reaction tank, pumps, and piping);
• Fertilizer production equipment (heat exchanger, circulating pump, evaporation
tank, condenser, and centrifuge);
• Equipment controls (instrumentation, electrical distribution, and miscellaneous
construction - also includes engineering, design, and project management); and
• A dewatering system (for dry process kilns only - pressure filter press, steel filtrate
tank, filter cake storage bin, rotary drum dryer, conveyor, pumps, piping,
electrical, and instrumentation).
The cost estimation for the recovery scrubber process includes the following elements:
(1) the annual quantity of CKD; (2) the process type (wet or dry); (3) the percentage of K2O in
the dust feed; and (4) the facility's baseline waste management costs. For the high value case,
EPA also assumes: (1) the scrubber is designed with excess capacity and processes not only
currently generated CKD but also CKD from stockpiles and/or alkali ash from off-site sources
(at a ratio of 5:2); (2) the facility can sell excess SO2 allowances for $300 per ton SO2; (3) the
plant switches from low sulfur coal to high sulfur coal, at savings of $2 per ton of coal; (4) the
potassium sulfate output is sold as fertilizer at $175 per metric ton; and (5) off-site sources pay
the facility $33 per metric ton to receive their alkali ash, which is used as an input to the process.
In the low value cases, the recovery scrubber is sized only to accommodate current on-site CKD
generation, and none of the additional sources of savings or revenue are assumed to be available
to the facility operator. In addition, off-site land disposal of the by-product crystals is required,
in either a commercial Subtitle D or Subtitle C landfill. Costing results for the high value and
low value (Subtitle C Disposal) cases are displayed in Exhibit 9-9.
27 Only 10 of the 81 respondents to the 1991 PCA Survey providing useable data indicated that coal was not used
as a primary kiln fuel at their plants.
28 Personal communication with Garrett Morrison, Passamaquoddy Technology, Inc., July 2, 1993.
-------
9-34
Exhibit 9-9
Estimated Incremental Net Costs for the Passamaquoddy Technology Recovery Scrubbing Process
Case
High Value
Case
Low Value
(Subtitle C
Disposal of
Residua]
Alkalies)
Sample
Facility
A
B
C
D
E
F
G
H
I
Wtd.
Avg.,
Haz.
Waste
Burners
Wtd.
Avg.,
All
Facilities
A
B
C
D
E
F
G
H
I
Wtd.
Avg.,
Haz.
Waste
Burners
Wtd.
Avg.,
All
Facilities
Burns
Hazardous
Waste
Yes
No
Yes
No
No
Yes
No
Yes
Yes
Yes
No
Yes
No
No
Yes
No
Yes
Yes
Total
Capital
Costs
($)
10,552
24,742
22,738
5,488
9,142
10,568
12,790
9,751
5,021
11,649
12,267
6,525
4,895
14,402
3,664
5,958
6,613
430
6,553
3,516
7,522
5,840
Total
Annual
O&M and
Annualized
Capital
Costs
($000/Yr.)
1,996
5,057
4,247
1,113
1,716
1,911
2326
1,946
1,006
2^21
2369
I486
1,094
3,641
874
1,385
1,479
76
1,562
796
1,812
1,388
Total
Annual
Savings
and
Income
($000/Yr.)
2,709
5,112
8,621
780
2,206
2,695
2,736
1,705
669
3,280
3,026
683
263
1,772
293
475
549
88
421
173
720
524
Annual
Net Cost
($000/Yr.)
-712
-55
-4,375
333
-490
-784
-410
241
336
-1,059
-657
903
831
1,870
580
909
930
-12
1,141
623
1093
864
Net
Cost/
Metric
Ton of
CKD
($/MT)
-8.87
-0.32
-19.32
12.94
-8.55
-10.99
-4.31
4.48
20.38
-11.78
-7.40
11.67
29.39
8.25
22.53
15.87
13.79
-12.85
21.21
37.73
12.37
14.03
Net
Cost/
Metric
Ton of
Cement
($/MT)
-1.87
-0.05
-3.54
1.03
-1.02
-2.68
-1.07
0.22
0.67
-1.52
-1.03
2.38
0.79
1.51
1.79
1.89
3.18
-0.03
1.06
1.24
1.57
1.36
-------
9-35
In the high value case, installation of the recovery scrubber yields net benefits for six of
the nine facilities with positive net CKD generation rates. The other three have relatively low
net CKD generation rates and operate dry process plants, both of which would make the
economics of this alternative less favorable than they would be otherwise; the projected cost
impacts are on the order of only $240,000 to $340,000 per year. The optimal application point
for six of the nine plants is to the net CKD stream rather than the gross CKD quantity. For two
of the remaining plants (A and B) the differences are marginal. For the six plants in the sample
showing a positive return on investment, net benefits average more than $1.1 million per year,
and range from $55,000 to almost $4.4 million annually. On a unit basis, benefits for the plants
under this case average about $7.40 per metric ton of CKD and $1.03 per metric ton of cement.
Under the low value case, the recovery scrubber process produces net benefits at only
one plant in the sample, and is uniformly most cost-effective when applied to the net CKD
generation rate. Under the assumption of hazardous waste disposal of the recovery scrubber
residue, the process is predicted to generate a (small) net benefit at only one plant, which has a
very low waste generation rate and a relatively high estimated current disposal cost. Unit costs
average about $1.35 per ton of cement, and range from about zero to $3.20.
Capital installation costs associated with this technology are comparable to those for the
fluid bed technology, and are considerably higher than those for the alkali leaching system. Total
capital costs under the high value case average more than $12.2 million, and range up to $24.7
million. Under the low value case, capital costs are actually lower, because in the absence of
expected revenues from receipt of alkaline ash from off-site and SO2 emission allowance sales,
the technology would be designed and installed at a significantly (about 50 percent) smaller scale.
In these cases, capital installation costs average about $5.8 million and run from $3.5 to about
$6.5 million for most of the sample facilities. As with the other technologies, capital costs would
be unaffected by whether the process residue is managed in non-hazardous or hazardous waste
management units, because all such disposal is assumed to occur at off-site locations, due to
scale economies.
Under the high value case, the annualized return on total invested capital for the six
plants with estimated net benefits from installing the recovery scrubber ranges from 0.2 to about
19 percent, with four of the facilities falling in the range from 3.2 to 7.4 percent per year. Across
the nine case study plants, the average annualized return on capital is 5.4 percent. These rates of
return are substantially lower than those associated with the alkali leaching system, even though
total net benefits are in all cases higher.
9.2.4 Other Operating Practices
In addition to modifying the chemical characteristics of and/or management practices
applied to CKD after it is removed from the kiln system, it may be possible to effect reductions
in disposal rates by relaxing some of the constraints that appear to limit CKD recycling or by
attempting to modify some of the inputs that may be resulting in increased generation and
removal of dust. Two possible approaches are presented and briefly discussed in this section.
Revised Standards for Cement Products
One possible means for overcoming recycling limitations imposed by-product quality
concerns would be to modify the ASTM alkali limit, at least for certain applications. The
purpose of the ASTM standard is to prevent reactions between the alkalis in cement and the
-------
9-36
lime and silica in the aggregates used in making concrete. Such reactions are to be avoided
because they expand, crack, and weaken the concrete. If the reactions could be reliably
prevented by some means other than limiting the alkali content of cement, specifications for the
alkali content of cement could be relaxed, and more CKD could be recycled. At least some
members of the cement industry favor such an approach, and the Mid-Atlantic Regional
Concrete Technical Committee is considering a recommendation favoring relaxed cement alkali
standards in combination with concrete aggregate standards (the committee consists of
representatives from the cement, aggregate, and concrete industries, state and federal highway
agencies, and the U.S. Army Corps of Engineers).29 At present, however, it is unclear whether
any such modifications in product standards will be undertaken. EPA has not been able to
evaluate the feasibility of implementing this type of approach to reducing CKD removal and
disposal rates.
Curtailing Use of Hazardous Waste Fuels
A second possible means of increasing recycling rates could be to reduce the use of
hazardous wastes as fuel. Only a subset of all cement plants and kilns burn hazardous waste.
Kilns at twenty-five cement plants are known to bum hazardous waste, while about 10 additional
plants have received and/or have applied for approval to do so. These thirty-five plants represent
less than a third of all cement plants in the country, and constitute about 25 percent of total
industry clinker capacity. PCA Survey data show, however, that the hazardous waste burning
plants tend to generate disproportionate quantities of CKD: the Agency projects that almost half
of the net CKD generated and land-disposed is associated with hazardous waste-burning cement
plants. Because (as stated above) hazardous waste provides about seven percent of the industry's
energy inputs and hazardous waste-burning plants constituted 25 percent of the industry (in
terms of clinker capacity) it can be estimated that hazardous waste provides about 28 percent of
the energy for the subset of kilns that use it.30
If the Bevill Exclusion for CKD were to be removed, one potential response of affected
cement plant operators could be to suspend hazardous waste burning in their kilns. The costs of
this response, per metric ton of cement, are calculated as follows:
(1) the amount of energy derived from hazardous waste fuel per metric ton of
cement, multiplied by the sum of
(2a) the revenues received for accepting hazardous waste per unit of energy, plus
(2b) the cost, per unit of energy, of replacing the energy value of the hazardous waste
with fossil fuel.
As shown in Exhibit 9-10, EPA estimates that for the five hazardous waste-burning plants
in the sample of nine, the average gross benefit from burning hazardous waste amounted to
29 Reardon, Patrick W., Jr., "Low Alkali Cement Requirements for Northeast U.S. Markets," Cement Technology,
November 1991, pp. 61-63.
30 Assuming that energy use per unit of clinker production does not differ on average between the facilities that
burn hazardous waste and those that do not, and if hazardous waste burners constitute 25 percent of clinker capacity,
then the hazardous waste burners also use 25 percent of all of the fuel used in the industry. If hazardous waste
provides seven percent of all energy, it must provide 28 percent (i.e., 7 percent divided by 25 percent) of the energy,
on average, at the subset of facilities that burn hazardous waste.
-------
9-37
Exhibit 9-10
Economic Benefits from Burning Hazardous Waste Fuels
Sample
Plant
A
C
F
H
I
Average
Quantities of Hazardous Waste
Fuels Received
Solid
(Metric Tons)
11,014
0
86
0
548
2,330
Liquid
(Metric Tons)
36,638
68,438
29,644
44,474
131
35,865
Gross Benefits from Hazardous Waste Fuel Burning
Revenues
from Receipt
of Wastes
($000)
12,325
12,821
5,596
8,332
296
7,874
Fuel
Cost
Savings
($000)
1,094
1,759
772
1,176
13
963
Total
Benefit
($000/yr.)
13,419
14,581
6,369
9,508
310
8,837
Benefit per
Metric Ton
of Cement
(S/yr.)
35.33
11.80
21.75
8.87
0.62
15.67
approximately $15.70 per metric ton of cement,31 which includes both revenues from receiving
hazardous waste from generators and alternative fuel cost savings.32 These do not reflect the
permitting, engineering, administrative, or operating costs associated with installing a hazardous
waste fuel burning operation at a cement plant; they are presented for illustrative purposes only.
The average value taken across the five hazardous waste burners in the sample obscures a
high degree of variability in the benefit of this practice among the individual plants. One of the
plants in the sample reported burning less than 1,000 metric tons of hazardous waste, which was
less than one unit of hazardous waste for every 500 units of cement produced. At the other
extreme, another plant operator reported consuming almost 50,000 tons of hazardous waste fuel,
or more than one unit of hazardous waste for every 10 units of cement. Consequently, estimated
revenues from hazardous waste burning range from less than $0.70 to more than $35 per metric
ton of cement.
51 Removal of the one facility with a very low hazardous waste fuel consumption rate raises the average benefit to
the remaining four plants to more than $19.40 per metric ton of cement.
52 According to a recent (May 1993) draft report published by EPA's Office of Waste Programs Enforcement,
Estimating Costs for the Economic Benefits of RCRA Non-Compliance. the median value received by cement plants
for burning bulk, non-halogenated solvents and organic liquids (the most prevalent hazardous waste fuel) was about
$170 per short ton, with a heating value of about 10,000 Btu per pound, which is about $34 per million Kcal. Prices
are even higher for halogenated solvents and organic liquids, for wastes in drums, or for solid wastes. EPA estimates
that bulk solid hazardous wastes bring a typical price of $450 per short ton. To this value must be added the cost of
replacing the hazardous waste with fossil fuel, which EPA estimates to be approximately $4.00 per million Kcal (see
Exhibit 2-22 of this report) assuming that bituminous, subbituminous, or lignite coal would be substituted for the
hazardous waste.
-------
9-38
This wide range of potential financial impacts resulting from curtailment of hazardous
waste fuel burning suggests that this alternative might be a cost-effective response to a change in
the regulatory status of CKD for some operators under certain conditions." This high degree
of variability also suggests, however, that the costs of suspending hazardous waste fuel burning at
some cement plants would exceed the costs of on-site Subtitle C disposal of CKD.
9.2.5 Summary of the Costs of Alternative CKD Management Methods
From the foregoing, it is clear that the operators of U.S. cement plants have a number of
options for managing the CKD that they generate. In most cases, current practices involve on-
site land disposal of CKD in unlined, non-engineered piles or portions of quarries. Some
operators, however, have been successful in either directly recycling a major portion of the gross
CKD that they generate or selling a substantial fraction of the dust that is not recycled, or both.
At a few facilities, novel approaches to recovering CKD that would otherwise be disposed have
been implemented, though the technical and economic feasibility of some of these technologies
have not been demonstrated at a full commercial scale. CKD also could be managed under
more stringent controls in accordance with existing or modified RCRA Subtitle C standards.
This would involve constructing on-site (because of scale economies) hazardous waste landfills,
and disposing waste CKD in these new units. Alternatively, regulations could establish tailored
contaminant release controls that would be installed and operated within the facility's existing
waste management system. Finally, in the event of a change in the RCRA regulatory status of
CKD, there might be incentives for kiln operators that burn hazardous waste fuels to cease this
practice in order to avoid regulatory compliance costs.
To facilitate further examination of and comparisons among these disparate approaches
to CKD management, the comparative costs of adopting these alternative practices are
summarized in Exhibits 9-11 through 9-13. Exhibit 9-11 presents summary statistics on the total
estimated cost impacts of adopting these strategies for the nine sample plants and for the subset
of five facilities that burn hazardous waste fuels. Exhibits 9-12 and 9-13 display these results in
normalized form; Exhibit 9-12 provides estimated facility-level cost impacts per metric ton of
CKD, while Exhibit 9-13 provides these results on a per ton of cement product basis. Each
exhibit provides impacts incremental to the estimated costs of current practices, and, where
applicable, displays both high value and low value results.
Exhibit 9-11 shows that, for the median and average plant results, two of the CKD
recovery technologies indicate net benefits (revenues) to the facility operator under the high
value case, for both the nine case study plants and the subset of five hazardous waste-burning
facilities. Tailored contaminant release controls and the third CKD recovery technology under
the high value case have the lowest costs. These alternative approaches would approximately
double current management costs, assuming the median and average values in the exhibit. From
there, the cost of alternative practices jumps substantially, to typical costs of more than $1
million annually per facility. In this category would fall the three CKD recovery technologies
under the low value case/Subtitle C and Subtitle C-Minus land disposal, and cessation of
hazardous waste burning. These results are generally consistent between the group of nine and
M In addition, if social costs were analyzed, they would include the lost surplus to suppliers of hazardous wastes
(i.e., the price each supplier would be willing to pay to be rid of its hazardous waste, minus the price each supplier
actually paid). This lost surplus could be a significant loss to society, though it would not be a factor in the decisions
of cement kiln operators except to the extent that the price for accepting hazardous waste at kilns that remain in the
business of burning hazardous wastes could rise substantially.
-------
9-39
the sub-set of five, with the obvious exception of the alternative of cessation of hazardous waste
burning.
-------
9-40
Exhibit 9-11
Total Incremental Annualized Costs of CKD Management Alternatives for EPA Case Study Cement Plants
CKD Management
Alternative
Recovery Scrubber/
High Value
Alkali Leaching/High
Value
Tailored
Contaminant Release
Controls/Sale of Dust
Fluidized Bed/High
Value
Tailored
Contaminant Release
Controls/No Sale of
Dust
Recovery
Scrubber/Low Value
Alkali Leaching/Low
Value
Fluidized Bed/Low
Value
Subtitle C- Landfill
Stop Burning of
Hazardous Waste
Subtitle C Landfill
All Facilities in Sample (9 Plants)
Minimum
($000/yr)
-4,375
-1,645
51
-434
51
-12
-27
895
138
0
138
Median
($000/yr)
-410
-68
83
207
100
903
1,135
1,793
2,185
310
3,958
Maximum
($000/yr)
336
201
364
415
364
1,870
3,665
5,597
11,817
14,581
14,382
Weighted
Average
($000/yr)
-657
-237
144
177
148
864
1,237
2,163
6490
4,910
9,511
Hazardous Waste Burners in Sample (5 Plants)
Minimum
($000/yr)
-4,375
-1,645
83
-434
100
623
628
1,635
1,536
310
2^79
Median
($000/yr)
-712
-414
100
176
110
930
1,375
1,987
2,251
9,508
4,095
Maximum
($000/yr)
336
201
364
207
364
1,870
3365
5,597
11,817
14,581
14382
Weighted
Average
($000/yr)
-1,059
-420
160
36
164
1,093
1,672
2,863
7,566
8,837
10,437
-------
9-41
Per plant averages.
Exhibit 9-12
Annualized Incremental Costs of CKD Management Alternatives per Metric Ton of CKD for EPA Case Study Cement Plants
CKD Management
Alternative
Recovery Scrubber/
High Value
Alkali Leaching/High
Value
Fluidized Bed/High
Value
Tailored
Contaminant Release
Controls/No Sale of
Dust
Tailored
Contaminant Release
Controls/Sale of Dust
Recovery Scrubber/
Low Value
Alkali Leaching/Low
Value
Fluidized Bed/Low
Value
Subtitle C- Landfill
Stop Burning
Hazardous Waste
All Facilities in Sample (9 Plants)
Minimum
($/MT)
-19.32
-75.08
-1.91
0.70
0.70
-12.85
-29.38
24.68
26.10
0.00
Median
($/MT)
-4.31
-4.78
3.85
2.28
2.30
15.87
27.23
30.41
61.60
18.75
Maximum
($/MT)
20.38
10.57
15.90
53.38
56.70
37.73
38.03
35.37
15148
176.78
Weighted
Average
($/MT)
-7.40
-0.37
2.31
2.03
2.53
14.03
20.08
28.21
53.20
79.74
Hazardous Waste Burners in Sample (5 Plants)
Minimum
($/MT)
-19.32
-5.81
-1.91
0.70
0.70
8.25
16.16
24.68
32.39
18.75
Median
($/MT)
-8.87
-4.77
2.27
1.48
1.60
13.79
18.18
28.87
52.10
94.43
Maximum
($/MT)
20.38
10.57
3.85
6.78
18.90
37.73
38.03
30.41
93.10
176.78
Weighted
Average
($/MT)
-11.78
-0.60
0.34
1.89
2.40
12.37
18.92
26.93
51.05
100.00
-------
9-42
Subtitle C Landfill
60.71
116.33
153.30
90.80
60.71
73.59
144.10
80.16 1
-------
9-43
On a cost per unit of waste basis, the results are quite similar, with respect to both the
direction and the relative magnitude of cost impacts. As shown in Exhibit 9-12, the recovery
scrubber and alkali leaching systems again show net benefits in the central tendency (median and
weighted average), high value case, and adoption of the fluidized bed recovery system or tailored
contaminant release controls imposes impacts of about the same magnitude as current practices.
CKD recovery under the low value case and the Subtitle C landfill alternatives would impose
impacts ranging between $14 and $91 per ton CKD, though these values are far lower than the
unit cost of commercial off-site Subtitle C disposal, and all but the Subtitle C landfill and
cessation of hazardous waste burning alternatives are less costly than typical off-site non-
hazardous waste disposal.
Incremental costs per ton of cement product follow much the same pattern. Exhibit 9-13
shows that the alternative practices fall in the same rank on a cost per unit product basis, and
that the central tendency measures (median and weighted average) suggest impacts of less than
$0.30 per metric ton of cement for five of the alternatives to current practice. CKD recovery
under the low value case imposes estimated cost impacts of between about $1.35 and $3.15 per
metric ton of cement, or 2.4 to 5.7 percent of the value of sales (about $55 per metric ton). The
Subtitle C land disposal and cessation of hazardous waste burning alternatives suggest typical
impacts exceeding $5 per metric ton, or more than the typical net margin received by cement
producers. In a few extreme cases, estimated impacts approach the typical sales price of cement.
Finally, capital investment requirements for installing these various CKD management
alternatives vary widely, as displayed in Exhibit 9-14. The capital costs of cessation of burning
hazardous waste are assumed to be negligible given EPA's costing assumptions regarding sunk
capital. Installation of contaminant release controls on existing CKD management units would
require highly variable investments of capital on the part of the facility operator, due to the
variability of existing management controls at CKD-generating plants and site-specific
environmental conditions (and associated risk potential). In most cases, however, capital
investment requirements for this alternative do not exceed $350,000 per plant. Installation of
CKD recovery technologies or Subtitle C landfill disposal alternatives, on the other hand, would
require much greater capital resources. Weighted average capital investment costs range from
about $1.2 million to almost $50 million. As stated above, the alkali leaching process is the least
complex and capital-intensive of the CKD recovery technologies, and occupies the lower end of
this cost range. Typical values for the capital costs of the other alternatives are in the $6 million
to $12 million range. Not surprisingly, the full Subtitle C land disposal alternative imposes the
greatest capital costs, due to the complexity and expense of installing multiple liner and leachate
collection systems and other aspects of regulatory compliance.
9.3 POTENTIAL IMPACTS OF ALTERNATIVE MANAGEMENT PRACTICES
Based on the cost estimates described in the preceding section, this section provides the
Agency's perspectives on potential impacts from implementing these alternative management
practices, first at the level of our typical case study cement plants and then for the industry as a
whole.
9.3.1 Individual Plant-Level Impacts
-------
9-44
In considering impacts at the individual plant level, the added cost of an alternative dust
management practice relative to the market value of cement produced (added cost per dollar of
sales) provides a direct measure of relative importance and a first general measure of impact.
-------
9-45
Exhibit 9-13
Annualized Incremental Costs of CKD Management Alternatives per Metric Ton of Cement for EPA Case Study Cement Plants
CKD Management
Alternative
Recovery Scrubber/
High Value
Alkali Leaching/High
Value
Fluidized Bed/High
Value
Tailored
Contaminant Release
Controls/Sale of Dust
Tailored
Contaminant Release
Controls/No Sale of
Dust
Recovery Scrubber/
Low Value
Alkali Leaching/Low
Value
Fluidized Bed/Low
Value
Subtitle C- Landfill
Stop Burning
Hazardous Waste
Subtitle C Landfill
All Facilities in Sample (9 Plants)
Minimum
($/MT)
-3.54
-1.41
-0.35
0.08
0.08
-0.03
-0.07
0.95
0.40
0.00
0.40
Median
($/MT)
-1.02
-0.18
0.46
0.17
0.20
1.51
2.35
3.72
3.10
0.62
9.30
Maximum
($/MT)
1.03
0.46
1.26
0.34
0.34
3.18
4.19
6.79
12.50
35.33
27.90
Weighted
Average
($/MT)
-1.03
-0.37
0.26
0.20
0.21
1.36
1.95
3.13
7.96
8.71
13.27
Hazardous Waste Burners in Sample (5 Plants)
Minimum
($/MT)
-3.54
-1.41
-0.35
0.13
0.13
1.06
1.25
1.53
2.10
0.62
3.70
Median
($/MT)
-1.87
-1.10
0.19
0.26
0.27
1.51
2.97
4.53
7.50
11.80 •
11.60
Maximum
($/MT)
0.67
0.35
0.66
0.34
0.34
3.18
4.19
6.79
12.50
35.33
27.90
Weighted
Average
($/MT)
-1.52
-0.60
0.05
0.21
0.22
1.57
2.40
3.84
8.63
15.67
13.60
-------
9-46
Exhibit 9-14
Capital Investment Requirements for Implementing CKD Management Alternatives for EPA Case Study Cement Plants
CKD Management
Alternative
Stop Burning
Hazardous Waste
Tailored
Contaminant Release
Controls/Sale of Dust
Tailored
Contaminant Release
Controls/No Sale of
Dust
Alkali Leaching/High
Value
Alkali Leaching/Low
Value
Recovery Scrubber/
Low Value
Fluidized Bed/High
Value
Fluidized Bed/Low
Value
Recovery Scrubber/
High Value
Subtitle C- Landfill
Subtitle C Landfill
All Facilities in Sample (9 Plants)
Minimum
($000)
0
263
27
51
132
430
5,090
5,090
5,021
21
21
Median
($000)
0
314
167
1,216
1,664
5,958
7,446
7,446
10,552
6,823
16,736
Maximum
($000)
0
2,032
1,887
2384
3,495
14,402
24307
24307
24,742
59,834
74383
Weighted
Average
($000)
0
501
507
1,260
1,904
5,840
9,599
9,599
12,267
30,712
47,105
Hazardous Waste Burners in Sample (5 Plants)
Minimum
($000)
0
301
308
732
1,537
3,516
7,052
7,052
5,021
3,117
7,809
Median
($000)
0
370
373
1,216
1,664
6453
8,567
8,567
10,552
7,142
17471
Maximum
($000)
0
2,032
2,032
2384
3,495
14,402
24307
24307
22,738
59,834
74383
Weighted
Average
($000)
0
554
560
1,491
2,278
7422
12392
12392
11,649
36,149
52,404
-------
9-47
Exhibit 9-15 summarizes the cost-relative-to-sales impacts of the alternative practices for the nine
typical cement plants studied in this Report. For this purpose, the cost per ton of cement
estimates (from Exhibit 9-13, above) have been compared to a nominal cement price of $55 per
metric ton, an approximate national average for recent years. As has been done previously,
results are presented both for the entire set of case study plants and, separately, for the five
hazardous waste fuel-burning plants. The alternative practices are ranked from highest cost to
lowest cost (or net revenue, in the case of high-value dust recovery).
The results presented in Exhibit 9-15 indicate a very wide variation in cost per dollar of
sales, both among plants for particular management practices and across the different alternative
practices. For Subtitle C land disposal, as an example, the difference between the low cost and
the high cost plant ranges from less than one cent to over 50 cents per dollar of sales. As noted
in earlier sections of this chapter, this extreme difference across plants is a reflection of the wide
variations in waste dust generation rates and other observed plant-specific variables affecting
total costs of waste management. Hazardous waste burners would generally face higher
management costs for any of the land disposal practices because of their higher waste generation
rates. Wide variation in relative costs implies great differences in competitive disadvantage,
should plants competing for the same regional market become subject to one of these alternative
practices.
The high absolute cost impact of the Subtitle C and C-Minus land management practices
is due to the combined effects of a high waste-to-product ratio (about 0.06 tons of net CKD per
ton of cement for the median kiln), the high incremental cost of Subtitle C practices and the
relatively low value of product (at about $55 per metric ton). Both median and high end cost-to-
sales ratios for Subtitle C and C-Minus land management are extremely high by any industry
pollution control standard, and generally exceed traditional industry profit rates by a wide margin
or multiple.
In contrast, the low cost alternative practices - both the tailored contaminant release
control land management and the various dust recovery technologies - suggest relatively
affordable approaches at the individual plant level, ranging from less than 1/2 cent per dollar of
sales (or possible net profit in some instances) up to a few cents per sales dollar, depending on
the particular alternative practice scenario.
9.3.2 Nationwide Cement Industry Impacts
As a first step in projecting nationwide impacts of alternative management practices for
waste dust, Exhibit 9-16 presents a hypothetical extrapolation of average plant-level costs for
each of the individual alternatives to the maximum relevant universe of affected cement plants.
For example, if Subtitle C land management standards were to be employed in the future in
place of current land management practices for waste dust, about 85 U.S. cement plants would
be affected, given EPA's estimate that about 25 percent of today's kilns recycle all or virtually all
of their collected dust. If all 85 plants were to manage all their current net CKD under these
Subtitle C standard practices, the required new capital investment cost would amount to about
$2.4 billion, and total annualized future dust management costs for the industry as a whole would
increase by more than $500 million per year.
By contrast, with respect to the Tailored Contaminant Release Control land management
scenario, the Agency estimates that perhaps 15 to 20 percent of the 85 facilities currently land
-------
9-48
Exhibit 9-15
Incremental Costs of Alternative CKD Management Practices Relative to Value of Cement Sales
(Incremental cost per ton of cement/revenue per ton of cement — cents per dollar of sales)
CKD Management
Alternative
Stop Burning
Hazardous Waste
Subtitle C Landfill
Subtitle C- Landfill
Recovery Low Value C
Tailored Contaminant
Release Controls/No
Sale of Dust
Tailored Contaminant
Release Controls/Sale
of Dust
Recovery High Value
All Facilities in Sample (9 Plants)
Minimum
0
0.7
0.7
-0.1
0.2
0.2
-6.4
Median
1.1
16.9
5.6
2.8
0.4
0.3
-1.9
Maximum
64.2
50.7
22.7
5.8
0.6
0.6
0.8
Weighted
Average
15.8
24.1
14.5
2.5
0.4
0.4
-1.9
Hazardous Waste Burners in Sample (5 Plants)
Minimum
1.1
6.7
3.8
1.9
0.2
0.2
-6.4
Median
21.5
21.1
13.6
2.8
0.5
0.5
-3.4
Maximum
64.2
50.7
22.7
5.8
0.6
0.6
1.2
Weighted
Average
28.5
24.7
15.7
2.9
0.4
0.4
-2.8
-------
9-49
Exhibit 9-16
Industry Wide Costs of Alternative Management Practices
(Cost in $ Million)
Subtitle C Standard Landfill
Subtitle C-Minus Landfill
Cease Haz. Waste Burning
Tailored Contaminant Release
Controls - No Sale
Tailored Contaminant Release
Controls - Sale
Dust Recovery - Low Value
Dust Recovery - High Value
Industry Wide
No. of
Affected
Plants
85'
85
35
70b
70
85
85
Total
Capital
Cost
1,919
970
0
39
38
912
434
Total
Annual
Cost
416
243
280
9
9
64
-49
35 Hazardous Waste
Burners
Total
Capital
Cost
1,070
610
0
23
22
238
369
Total
Annual
Cost
224
143
280
5
5
35
-34
' - Assuming 30 plants with zero or negligible net waste.
b - Assuming 15-20% of the 85 plants with net waste are already essentially in compliance.
disposing waste dust may already be employing these or equivalent practices.34 Hence, under
this scenario, only about 70 of the 115 plants, or 60 percent of the industry nationwide might
initially incur costs for this particular set of alternative practices. Similarly, aggregate costs or
other impacts affecting hazardous waste burning cement plants, viewed as a subset, would relate
only to the 35 or so plants expected to be in that category.
It is clear from Exhibit 9-16 that total national costs for the alternative practices,
considered independently from one another, vary substantially across the wide variety of
management methods evaluated. Within the subset of land management practices alone
(including Subtitle C Standards, C-Minus, and Tailored Contaminant Release Control), new
capital investment requirements could range from $2 billion down to $38 million, and annualized
costs for the industry could range from $416 to $9 million. Under the two most favorable dust
recovery technology options, in the two cases considered, capital investment costs for the 85
affected plants would also be quite substantial (on the order of $500 million to over $1 billion for
85 cement plants). But if these emerging technologies should prove universally adaptable and
cost-effective, within the range tentatively estimated by the Agency, then total industry-wide costs
34 This is a very rough estimate based on land management practices reported in the PCA Survey for 1990. See
Chapter 4, Section 4.3, for summary information on practices.
-------
9-50
could be in a quite moderate range between plus or minus 6 percent of sales and with a median
close to break-even.
The potential impacts on the industry and its markets could also be expected to vary to
an extreme degree across the several technical dust management alternatives. For example, the
high-cost Subtitle C practices, with both high initial capital requirements and costs per ton of
cement ranging over 30 percent of the value of product, would place an extreme competitive
strain on a large segment of the industry. Here, one would expect, with 25 to 30 percent of the
domestic industry as well as potential foreign competition unaffected, that an initial impact would
entail a substantial decrease in the number of financially viable domestic cement plants. Since
the major portion of surviving plants would be operating under substantially higher long run total
costs, economic theory would also project a substantial increase in regional and national average
prices of cement necessary to cover the increased costs of waste management. The natural
market corollaries of this impact scenario also suggest a decrease in overall domestic demand for
cement in relation to other substitute building and construction materials, and a larger relative
and absolute market share for imported cements.35
The Tailored Contaminant Release Control scenario for continued land management of
waste dust represents an alternative with much less potential for adverse industry impacts.
Indeed, with median incremental costs at only about one-half of one percent of sales and a high
end cost at one percent, there would appear to be little concern regarding industry-wide impacts.
For the industry subgroup of hazardous waste fuel burners, the alternative practice option
of ceasing to burn hazardous waste-derived fuels would also, as noted previously, imply rather
extreme competitive disruption. With average and median plant-level financial impacts over 20
percent of the average value of cement, many and perhaps most of these plants would not
consider this a financially viable alternative under current or future market conditions.
The emerging recovery technologies represent the most interesting as well as the most
uncertain set of CKD management alternatives from a potential impacts standpoint.
9.3.3 Conclusions and Relationships to Regulatory Requirements
The preceding discussion has focused primarily on costs and implications of individual
CKD management practices as technological alternatives, viewed independently in isolation from
one another, and in the absence of any particular regulatory context. In reality, cement
companies are not and would not in the future be restricted to just one industry-wide
alternative. As demonstrated in this and preceding chapters, plants generally have a choice of
several existing and emerging dust control and management options. Subject to existing
regulatory constraints, managers can be expected to chose the alternative or combination of
alternatives that tend to contribute most to company profitability, by minimizing disposal costs
and/or by exploiting CKD by-product use potentials.
In the absence of further state or federal CKD regulation, the principal competitive
choices foreseeable in the baseline would appear to include: (1) traditional land disposal; (2) a
continued substantial role for various off-site by-product uses (the most profitable option for
35 As recently as 1988, about one-fifth of U.S. cement demand was supplied by imports. This has decreased to
about 10 percent in 1991, and is expected to decrease further to seven percent by 1992. U.S. Bureau of Mines,
Mineral Commodity Summaries, 1993, p. 42.
-------
9-51
many plants); (3) further development of the emergent dust recovery technologies; and
(4) possible in-plant dust generation reduction and recycling via process adjustments, improved
system controls, and raw material or fuel substitutions.
Given the wide variety and the plant-specific nature of the options, it is virtually
impossible to predict future baseline CKD management trends with any degree of accuracy.
High technology land management practices, of the type required for hazardous waste land
placement under Subtitle C of RCRA, would not be considered an economically competitive (or
perhaps even an economically feasible) CKD management option by private management in the
baseline regulatory context. Most plants might continue to view currently uncontrolled land
placement as the optimal practice from the plant's profit-and-loss standpoint. Nonetheless,
evidence presented in Chapter 8 and earlier in this chapter suggest that a substantial segment of
the industry is looking towards CKD reduction, recovery, and off-site use options as economically
superior. EPA's cost analysis suggests that one or more of the currently available or emerging
recovery technologies could in fact be economically attractive, under a reasonable set of
assumptions, particularly for plants with high CKD generation. To the extent that these
approaches do prove successful from a private profit-and-loss standpoint, the industry's future
baseline trend would be towards increased natural resource conservation and decreased land
disposal.
New state or federal regulations directed at controlling traditionally unrestricted on-site
land placement could assume many forms and degrees of restriction. Several levels of such
controls were simulated by the Agency's engineering cost studies, including two variations of
incremental contaminant release controls and two increasingly more severe versions of RCRA
Subtitle C technology standards. Essentially, these regulatory approaches would remove low cost
land practice options and shift cement plant decisions toward a choice between the (increasingly)
higher cost land disposal options and the other CKD reduction and recovery options.
Two main conclusions were drawn regarding the relative costs of these shifting choices.
The first is that incremental land management practices, of the type simulated in the tailored
contaminant release control scenarios, could be implemented at relatively modest additional cost
at most of the 70 or so cement plants not now already in compliance. Although CKD
management costs would about double, on average, relative to baseline, the absolute cost
increment relative to the market value of cement would average about one-half of one percent of
sales. The maximum added cost-to-sales estimate for the EPA case study sample was about one
percent.
This cost increment could probably be absorbed without severe industry impact or a
substantial number of plant dislocations. There would be some fractional percentage increases in
regional cement prices to cover the added dust management costs, but this would not appear
sufficient to materially influence international trade flows. At the same time, the alternative
non-disposal CKD reduction and recovery options would become incrementally less costly (or
more profitable) relative to land disposal. As a result, one would also expect at least a modest
shift away from land disposal towards the other options.
The second major conclusion from the cost analysis is that the incremental land disposal
costs implied by the RCRA Subtitle C minimum technology and administrative standards would
indeed be much higher in both relative and absolute terms than either current typical land
practices or the incremental contaminant release scenario described above. The Subtitle C
regulatory scenarios projected incremental cost-to-sales ratios for the median affected facility of
-------
9-52
5,5 and 17 percent, respectively, for the C-Minus and full Subtitle C versions. (For hazardous
waste burners, median and average costs would be even higher than for the industry as a whole
due to generally higher net waste generation rates.) High end ratios for the EPA case study
sample were estimated at 21 percent and 52 percent.
For the Subtitle C scenarios, any impact assessment becomes inherently uncertain.
Under the industry-wide waste listing version, a substantial majority of the Nation's cement
plants (about 85 currently managing new CKD in land-based units out of 115 U.S. plants) would
be faced with a rather extreme set of dust management options. (Under an alternative
regulatory option, where removing the Bevill regulatory exemption would affect only hazardous
waste burners, about 35 of the 115 operating cement plants would face the same choice among
extreme options.) On the one hand, affected plants would face a very high cost but well known
and relatively simple-to-implement land disposal technology. The alternative waste reduction,
recycle, and recovery options, though potentially much lower in cost, are likely to be much more
technically complex to adopt and operate, and much more uncertain in outcome. Some may not
work as promised; some may not be as cost effective as the original prototypes or EPA's
preliminary cost estimates might suggest.
The possible outcomes of this potential decision process, viewed in the context of a
generally new and developing set of dust reduction and recovery technologies, cover a broad
spectrum. Although the Agency's studies of alternative practices and preliminary engineering
cost estimates suggest that currently or soon to be available innovative technologies could well
play a critical future role in the industry, at this early stage in their development there can be no
real guarantees. Under a Subtitle C regulatory scenario, the economic risk is that a significant
portion of the industry might be unable to develop or adapt to what would amount to a
technological revolution in cement kiln dust generation and management practices. Under this
circumstance, a substantial portion of the industry could be forced out of business due to the
high costs of land disposal, as described above. On the other side of the equation are the
substantial economic benefits, the savings in energy and other natural resources, and the
improvements in environmental quality that could potentially accrue from the new recovery
technologies, should they prove successful.
-------
9-53
CHAPTER NINE
COST AND ECONOMIC IMPACTS OF ALTERNATIVES
TO CURRENT CKD DISPOSAL PRACTICES
9.0 INTRODUCTION 9-1
9.1 APPROACH AND METHODS 9-1
9.1.1 Data Sources 2
9.1.2 Approach to Estimating Costs and Impacts of CKD Management
Alternatives 2
Case Study Plants 2
Methods for Estimating Facility Costs 3
Methods for Extrapolating from the Case Study Sample to the Industry .... 4
9.1.3 Cost Accounting Assumptions 5
9.1.4 Limitations of the Analysis 5
9.2 DESCRIPTIONS AND COSTS OF BASELINE AND ALTERNATIVE CKD
MANAGEMENT METHODS 6
9.2.1 Current Practices 7
Direct Recycling of Collected Dust 8
Off-Site Beneficial Use 9
Current Land Disposal Practices 9
9.2.2 Alternative Land Disposal Practices 10
Conventional Subtitle C Technology and Administrative Standards 10
Subtitle C Costs (Exclusive of Corrective Action Costs) 11
Potential Subtitle C Corrective Action Costs 11
Alternative Subtitle C Costs Under RCRA §3004(x) 14
Tailored Contaminant Release Controls 15
9.2.3 Alternative On-Site CKD Recycling and Recovery Techniques 19
-------
9-54
Increasing Direct CKD Recycling 19
Innovative CKD Recovery Technologies 20
Alkali Leaching 22
Fluid Bed Dust Recovery 26
The Passamaquoddy Technology Flue Gas Desulfurization Process .. 28
9.2.4 Other Operating Practices 31
Revised Standards for Cement Products 31
Curtailing Use of Hazardous Waste Fuels 32
9.2.5 Summary of the Costs of Alternative CKD Management Methods 34
9.3 POTENTIAL IMPACTS OF ALTERNATIVE MANAGEMENT PRACTICES 37
9.3.1 Individual Plant-Level Impacts 37
9.3.2 Nationwide Cement Industry Impacts 40
9.3.3 Conclusions and Relationships to Regulatory Requirements 43
-------
9-55
LIST OF EXHIBITS
Exhibit 9-1 Facilities Included in the Cost Analysis 3
Exhibit 9-2 Subtitle C Disposal Costs 12
Exhibit 9-3 Subtitle C-Minus Disposal Costs 15
Exhibit 9-4 Tailored Contaminant Release Controls Costs: Continued Sales of CKD
for Off-Site Use 17
Exhibit 9-5 Tailored Contaminant Release Controls Costs: Curtailed Sales of CKD
for Off-Site Use 18
Exhibit 9-6 Key Design Conditions for the Nine Case Study Cement Plants 23
Exhibit 9-7 Estimated Incremental Net Costs for the Alkali Leaching System 25
Exhibit 9-8 Estimated Incremental Net Costs for the Fuller Fluidized Bed System 27
Exhibit 9-9 Estimated Incremental Net Costs for the Passamaquoddy Technology
Recovery Scrubbing Process 30
Exhibit 9-10 Economic Benefits from Burning Hazardous Waste Fuels 33
Exhibit 9-11 Total Incremental Annualized Costs of CKD Management Alternatives for
EPA Case Study Cement Plants 35
Exhibit 9-12 Annualized Incremental Costs of CKD Management Alternatives per
Metric Ton of CKD for EPA Case Study Cement Plants 36
Exhibit 9-13 Annualized Incremental Costs of CKD Management Alternatives per
Metric Ton of Cement for EPA Case Study Cement Plants . 38
Exhibit 9-14 Capital Investment Requirements for Implementing CKD Management
Alternatives for EPA Case Study Cement Plants 39
Exhibit 9-15 Incremental Costs of Alternative CKD Management Practices Relative to
Value of Cement Sales (Incremental cost per ton of cement/revenue per
ton of cement — cents per dollar of sales) 41
Exhibit 9-16 Industry Wide Costs of Alternative Management Practices (Cost in $
Million) 42
-------
10-13
CHAPTER TEN
STUDY FINDINGS AND REGULATORY OPTIONS
10.1 Study Findings 10-1
10.1.1 Sources and Volumes of Waste (Study Factor 1) 10-1
10.1.2 Waste Management Practices (Study Factors 2 and 8) 10-2
10.1.3 Waste Characteristics and Potential Risks to Human Health and the
Environment (Study Factor 3) 10-2
10.1.4 Documented Evidence of Damage (Study Factor 4) 10-4
10.1.5 Potential Costs and Impacts of Subtitle C Regulation (Study Factors 5, 6,
and 7) 10-4
10.2 Environmental Justice 10-5
10.3 Recommendations 10-5
10.3.1 Decision Rationale and Options 10-5
10J.2 Regulatory Options 10-7
10.33 Next Steps , 10-11
-------
CHAPTER TEN
STUDY FINDINGS AND REGULATORY OPTIONS
10.1 Study Findings
Based on the information collected for this Report to Congress, this chapter presents a
summary of the Agency's findings regarding various aspects of the generation and management
of cement kiln dust (CKD) waste, as well as our initial regulatory options for this waste. Results
of EPA's analysis of the eight Congressionally-mandated study factors (see Chapter 1) are
presented as follows: sources and volumes of waste (Study Factor 1) in Section 10.1.1; current
and potential uses of CKD (Study Factor 8), and present disposal practices (Study Factor 2) in
Section 10.1.2; potential danger to human health and the environment (Study Factor 3) in
Sections 10.1.3; documented cases of damage to human health and the environment (Study
Factor 4) in Section 10.1.4; and costs and impacts of alternative CKD management scenarios
(Study Factors 5, 6, and 7) in Section 10.1.5.
10.1.1 Sources and Volumes of Waste (Study Factor 1)
In 1990, the cement manufacturing industry in the United States consisted of 43
companies operating 115 clinker-producing plants (218 kilns) in 37 states and Puerto Rico.
California was the largest clinker producing state in 1990, followed by Texas, Pennsylvania,
Missouri, and Michigan. Although all cement is manufactured in inclined rotary kilns using
similar raw materials (primarily limestone, clay, and sand), variations in the manufacturing
process and kiln design affect energy requirements and production capacity at each facility. The
cement industry burns large amounts of high Btu fuels during the manufacturing process,
primarily coal and other fossil fuels. In 1990, however, 23 facilities also burned hazardous waste
as fuels.
Based on an analysis of existing data, including industry data collected by the Portland
Cement Association and EPA survey data collected under RCRA §3007 authority from the
operators of cement manufacturing facilities, the Agency has documented that cement plants
generate large quantities of cement kiln dust waste. In 1990, the generation of gross CKD (i.e.,
CKD that is collected by air-pollution control devices) was 12.7 million metric tons; there are,
however, wide variations among kilns in total gross CKD generated and gross CKD generated
per ton of clinker.
In addition, there are wide variations among kilns in the amount of net CKD that is
generated (i.e., CKD that is either disposed or used beneficially off-site). For example, twenty-
five percent of facilities produce essentially no net CKD (CKD that is either disposed or sold),
while 10 percent of the largest net generators produce almost 50 percent of all net CKD.
Finally, the Agency also found that the burning of hazardous waste appears to affect the
volume of dust that is actually disposed of. Kilns that burn hazardous waste remove from the
kiln system an average of 75 to 104 percent more dust per ton of clinker than kilns that do not
burn hazardous waste. The Agency is interested in receiving additional information regarding
how the burning of RCRA hazardous wastes, non-hazardous wastes (such as tires and non-
-------
10-2
hazardous used oils), and fossil fuels affect the quantity and chemistry of generated CKD, as well
as the partitioning of toxic metals, chlorides, and alkalis between stack gases, CKD, and clinker.
10.1.2 Waste Management Practices (Study Factors 2 and 8)
For that portion of CKD that is disposed of, industry practice is to manage it in piles,
quarries, and landfills, most of which are unlined and uncovered. (Most of the gross CKD -- 8.2
million metric tons or 64% -- is currently recycled directly back into the kiln or raw feed system.)
Some active piles are also managed underwater or adjacent to surface water and/or actively tilled
agricultural lands. Although most CKD waste is disposed on-site, some is sold for off-site use.
For example, in 1990, 7% of gross CKD generated (897,000 metric tons) was sold for off-site
use, most of it as a waste stabilizer, liming agent, or materials additive.
Opportunities do exist, however, to further reduce the amount of net CKD that is
disposed of or sold off-site for use by recycling it back into the kiln. The Agency has identified a
number of pollution prevention opportunities, including flue gas desulfurization, fluid-bed dust
recovery, and leaching with water, that may, in some instances, represent low cost and potentially
profitable alternatives to CKD disposal practices.
Federal statutes that potentially affect CKD management include the Clean Air Act
(CAA), Clean Water Act (CWA), the Resource Conservation and Recovery Act (RCRA), and
the Comprehensive Environmental Response, Compensation and Liability Act (CERCLA).
Provisions of regulations developed under authority of the CAA and CWA impose regulatory
controls on releases of CKD to the air (via stack or fugitive dust emissions) and water (from
stormwater run-off and point source effluent discharges), respectively. Under both RCRA and
CERCLA, the federal government can respond to situations where the release of CKD or its
constituents presents an imminent and substantial danger to human health and the environment.
CKD that is not directly recycled is also subject to regulation under Subtitle D of RCRA. In
addition, CKD generated in kilns that burn RCRA hazardous waste is subject to the RCRA
Boiler and Industrial Furnace rule (40 CFR 266.112).
Based on an analysis of state regulations, the Agency has found that cement kiln dust
waste is regulated under state and local laws, but the requirements vary significantly from state to
state. For example, California regulates CKD as a non-RCRA hazardous waste, but has
suspended enforcement of the management requirements for CKD that fails the State's
hazardous waste corrosivity test, pending the results of further study of CKD. Pennsylvania
regulates CKD as a residual waste, requiring facilities to comply with site-specific disposal
requirements and waste reduction strategies, which are both periodically updated by the State.
In contrast, Michigan and Texas both consider CKD an industrial non-hazardous waste.
Michigan requires permits, ground-water monitoring, and regular reports of ground-water
sampling results; whereas Texas issues non-enforceable guidance.
10.1.3 Waste Characteristics and Potential Risks to Human Health and the Environment
(Study Factor 3)
EPA's analysis of cement kiln dust chemistry shows that CKD does contain toxic
constituents, including metals and organic by-products. Constituents identified in dust solids and
leachate include arsenic, thallium, antimony, lead, chromium, total-2,3,7,8-substituted dioxins,
-------
10-3
and total hexachloro-dibenzodioxin. In addition, water-CKD mixtures are often RCRA corrosive
(see 40 CFR 261.22), with pH levels commonly in excess of 12.5 standard units.
In addition, on the basis of our analysis of leachate test results, EPA has found that no
significant distinction can be made between CKD generated from kilns that burn hazardous
waste from those that do not burn hazardous waste. (This rinding was corroborated for metals
content in CKD by leachate test results submitted to the Agency by the cement industry.) For
example, laboratory analysis of CKD using the Toxicity Characteristic Leaching Procedure
(TCLP) shows that trace metal concentrations rarely exceed RCRA toxicity limits whether or not
the CKD is generated at kilns that burn hazardous waste. At the same time, certain metals, such
as lead, cadmium, and chromium are present in the CKD at a consistently higher mean
concentration from those kilns that burn RCRA hazardous waste than those that do not
(alternatively, thallium is higher in CKD generated from kilns that burn predominantly non-
hazardous fuels).
While it is not possible to establish statistically significant differences between these
groups due to small sample sizes, detectable, but low, concentrations of dioxins and(
dibenzofurans were detected in CKD, (ranging in concentration from a few parts per trillion to
7.7 parts per billion), at both hazardous waste burning facilities and non-hazardous waste burning
facilities. However, the highest concentrations that were measured in CKD came from kilns that
burn hazardous waste. [Note: The levels of dioxins in dust observed at the River Cement facility
in Festus, Missouri, a facility that burns hazardous waste, are at least 15 times higher than those
found at any other facility for which EPA has data. We believe that River Cement is likely
atypical of the industry as a whole.] Volatile and semivolatile compounds were generally not
found in CKD.
With respect to exposure scenarios associated with on-site CKD management, EPA
modeled both direct and indirect exposure pathways, including contaminated surface water and
ground water used as a drinking water source, direct inhalation and ingestion of windblown
CKD, and the ingestion of foodstuffs (beef, milk, fish, and\or vegetables) originating from
agricultural fields or streams that are receiving releases of CKD from nearby piles. The sample
of cement plants examined in this analysis appears to be generally representative of typical
cement plants in terms of several factors that influence risk. Based on this analysis, cancer risks
for individuals living around cement plants under average conditions of transport and exposure
(defined as central tendency estimates) were low (below 1 x 10"4). In addition, noncancer effects
were below the threshold effects level, indicating a negligible likelihood of noncancer impact.
This analysis also quantified the high end of the distribution of risks around these same cement
plants. While the risks were somewhat higher, they are generally considered within an acceptable
risk range.
The Agency recognizes that the high end results obtained above may not necessarily
capture the upper bound of the risks that exist across the whole universe of cement plants, as
site-specific factors at some plants may contribute to higher risks. Therefore, in addition to a
central tendency analysis, the Agency also conducted a sensitivity analysis of several hypothetical
scenarios representing a combination of potentially higher risk transport and exposure situations.
This analysis estimated that the potential cancer risks for individuals living around cement plants
assuming plausible worst-case conditions (i.e., modeled utilizing the highest measured
concentrations of dioxins and arsenic found in CKD and leachate derived from CKD) were in
the risk range of 10"5 to 10'2 (for purposes of this analysis, these individuals are hypothetical
-------
10-4
individuals highly exposed to CKD intake who were created for purposes of the Agency's risk
characterization). The hypothetical scenarios are: (1) subsistence fish consumers ingesting fish
caught in nearby waters; or (2) subsistence farmers ingesting elevated amounts of vegetables
grown in, or beef and milk derived from animals who ingested grasses originating from
agricultural fields receiving releases from nearby CKD piles through air deposition.
The Agency does not have sufficient information to determine whether these plausible
worst-case conditions, of high transport and high exposure potential exist around cement
manufacturing facilities, and if yes, their prevalence. Therefore, the Agency is interested in
receiving additional information regarding the extent to which activities such as farming
(including recreational gardening), fishing, and swimming occur around these facilities.
10.1.4 Documented Evidence of Damage (Study Factor 4)
Migration of potentially hazardous constituents, including metals, has occurred from
cement kiln dust waste sites. EPA has documented seven cases of damage to surface water and
ground water, and 21 cases of documented damage to air from cement kiln dust waste. By
damage, the Agency means that toxic constituents have contaminated ground water and/or
surface water, and/or air above Maximum Concentration Limits or some other standard.
Constituents of concern being released to ground and surface waters include arsenic, chromium,
and lead, among others. When ground-water and surface water exceedances do occur, the
magnitude of the exceedance is generally small, although in certain instances it was as high as
two orders of magnitude above the Maximum Concentration Limit for drinking water.
Environmental damage generally affects the area in the immediate vicinity of the waste
disposal site. However, in some cases, nearby wetlands and streams that are off-site were also
impacted. For example, releases from two facilities in Mason City, Iowa caused severe
degradation of the aquatic habitat in nearby Calmus Creek. Observed releases are commonly
chronic at sites at which exceedances have been noted.
It should be noted that information on environmental quality, on which this evidence is
based, is limited by available data from each of the 127 sites evaluated. For those sites that had
data, available files contained information on releases, but little human exposure data. Because
there is little evidence of direct human exposure to environmental releases from CKD, it appears
that the observed damages are not widespread.
Waste disposal practices at sites where water damages have been documented include
management in waste piles, abandoned quarries, or landfills, all of which were unlined. Air
damages are primarily due to mechanical failure of dust handling equipment. There is no
evidence that any damage has directly affected human health. In particular, drinking water wells
are located far enough away, and/or tap aquifers are isolated enough to be very unlikely to access
contaminated ground water.
10.1.5 Potential Costs and Impacts of Subtitle C Regulation (Study Factors 5, 6, and 7)
If CKD were required to be managed as a RCRA hazardous waste under the existing
regulatory scheme, there would likely be significant compliance costs for these facilities. These
costs may potentially be reduced if they could recycle their dust. For these facilities costs would
-------
10-5
be upwards of 20 percent or more of product sales. In addition, domestic and international
competition limits the ability for those facilities to pass costs through to customers.
The costs of managing CKD as a hazardous waste could be reduced, if RCRA Section
3004(x) authority is used to modify certain Subtitle C requirements (e.g., prohibitions on land
disposal, minimum technological requirements for managing CKD). Costs would also be
imposed under the Option 5 management standards, although those costs are likely to be much
less than under Options 3 and 4. Removing the exemption, but not specifically listing CKD
would have less cost impact, as most CKD is not RCRA characteristic hazardous waste. While
those cement kilns that burn listed RCRA hazardous waste would be required to handle their
CKD as hazardous, they will likely be able to at least partly absorb the costs of Subtitle C
compliance with revenue from accepting and burning these wastes. In addition, these facilities
are already subject to a number of the more costly RCRA requirements (e.g., requirement to
obtain a permit, corrective action).
Pollution prevention opportunities, including, flue gas desulfurization, fluid-bed dust
recovery, and alkali leaching show promise as low cost, and potentially profitable alternatives to
disposal in piles. Flue gas desulfurization creates new lime for use as raw material, scrubs stack
gases of sulfur, and creates pelletized alkali sulfates that may be sold as fertilizer. The alkali
leaching process dissolves alkalis from CKD, enabling more CKD to be returned to the kiln.
The process creates an alkali solution that may be sold as a fertilizer. The fluid-bed dust
recovery process takes CKD and converts it directly into clinker. All threp technologies can be
used to process old CKD piles. The Agency is interested in receiving additional information
regarding how these processes affect the quantity and chemistry of air emissions, as well as the
partitioning of toxic metals, chlorides, and alkalis between CKD and clinker.
10.2 Environmental Justice
In addition to the eight study factors specifically identified in the statute, the Agency is
interested in determining whether there are environmental justice issues associated with the
management of CKD. The Agency's risk modeling results indicate that subsistence farmers and
subsistence fish consumers would be most susceptible to the risks posed by the management of
cement kiln dust.1 It is not known, however, how prevalent these activities are around existing
cement manufacturing facilities. Although the Agency acknowledges that its concern is
speculative, the prospect that subsistence farmers and subsistence fish consumers may be of low
income or minority status suggests that there might also be environmental justice issues
associated with cement manufacturing facilities. The Agency is interested in receiving additional
information regarding the extent to which activities such as farming (including the recreational
gardening of vegetables) and fishing occur around these facilities, and in particular, whether
subsistence farming and subsistence fishing exists. The Agency is also interested in learning of
concerns related to environmental justice (i.e., the fair treatment of people of all cultures,
incomes, and educational levels with respect to protection from environmental hazards)
associated with the management of cement kiln dust.
10.3 Recommendations
1 For purposes of this report, subsistence farmers and subsistence fish consumers are hypothetical individuals highly
exposed to CKD intake who were created for purposes of the Agency's risk characterization.
-------
10-6
10J.1 Decision Rationale and Options
Based upon the analysis of the eight study factors in RCRA §8002(o), EPA has reached
some preliminary findings. Utilizing the three step procedure described in Chapter 1 of this
volume (Section 1.2), EPA has arrived at tentative answers to the questions posed in its decision
rationale, which are described below. The decision rationale contributed to development of the
five proposed options for managing CKD waste (listed in Section 10.3.2), although the Agency
has not yet made a final decision. EPA is soliciting comment on how the decision rationale can
be used in the Agency's decision-making process.
Step 1: Does management of CKD pose human health and environmental problems?
Might current practices cause problems in the future?
After reviewing evidence of damage to human health and the environment, performing a
risk assessment, and reviewing the results of laboratory analyses of waste samples, EPA has
concluded that risks associated with CKD management are generally low. There is, however, a
potential under certain circumstances for CKD to pose a danger to human health and
environment, and it may do so in the future.
Data collected from state files and EPA site visits identify common CKD waste
management practices, including management in exposed, unlined piles, abandoned quarries, and
landfills, that have caused, and may continue to cause, contamination of air and nearby surface
water and ground water. Management practices such as disposal in a water-filled quarry and
management in piles adjacent to grazing and agricultural fields or surface water bodies also pose
a potential danger to human health and the environment. In addition, risk modeling results
support the conclusion that CKD can potentially pose risks to human health and the
environment under certain hypothetical, yet plausible scenarios.
Step 2: Is more stringent regulation necessary or desirable?
EPA has reached no conclusions with respect to the need for more stringent regulation.
EPA's preliminary analysis of the effectiveness of state and federal regulations and controls
suggests that additional controls should be evaluated; for example, controls for CKD
management scenarios which potentially present high risks, if those scenarios exist. While CKD
is regulated under state and local laws, the specific requirements for CKD vary from state to
state. In many instances, minimal controls are applied to these wastes. Also, recycling
technologies could be used as a means to improve waste management practices.
Step 3: What would be the operational and economic consequences of a decision to
regulate CKD under Subtitle C?
Operational costs of CKD regulation are largely dependent on the management
alternative selected. If CKD is managed as a hazardous waste under RCRA Subtitle C, facilities
that manage their CKD through on-site land disposal are estimated to incur significant
compliance costs. However, the financial burden of compliance, even for waste dust generated in
kilns that burn RCRA hazardous waste, may be reduced or potentially turned into net income, if
facilities are able to adopt pollution prevention technologies that recycle CKD.
-------
10-7
The possible economic outcomes of a decision to regulate CKD under RCRA Subtitle
C cover a broad spectrum. An economic analysis of innovative pollution prevention technologies
(including alkali leaching, flue gas desulfurization, and fluid bed dust recovery), suggests that the
potentially high compliance costs of CKD land disposal may drive the industry toward more
recycling of their CKD. However, at this early stage of their development, it is uncertain that
these recycling technologies can be widely adopted by the industry. Moreover, even if CKD is
recycled, some facilities may incur substantial disposal costs.
10J.2 Regulatory Options
This section presents a series of options the'Agency is considering concerning the
management of cement kiln dust waste based on the findings of this Report. In accordance with
RCRA §3001(b)(3)(C), EPA will make a regulatory determination for cement kiln dust waste
after submitting this Report to Congress, holding a public hearing, and accepting and reviewing
public comments.
As stated previously, cement kiln dust waste generally presents a low inherent toxicity, is
only rarely characteristically hazardous, and, in most cases based on risk modeling, does not
present a risk to human health and the environment. However, cement kiln dust waste may pose
a potential threat to human health and the environment considering plausible worst-case
conditions under certain hypothetical management scenarios (see Chapters 5 and 6). Major
factors increasing the potential for human health and environmental damages include proximity
to potential exposure points such as agricultural fields and surface water bodies, as well as the
concentrations of key constituents of concern.
Based on the findings, and an initial evaluation of regulatory options, the Agency has not
decided whether to retain or remove the CKD exemption. The Agency considered a number of
options which represent a wide range of scenarios that would subject CKD to different
management requirements and enforcement oversight. From these, the Agency has chosen to
highlight five, including three in which CKD would be managed under Subtitle C, with the intent
to focus public comment from environmental groups, industry, and other interested parties
regarding the most appropriate approach to manage CKD.
EPA notes that regulations for the management of CKD waste under Subtitle C may not
be warranted or appropriate if other Agency-administered programs are better suited to address
the concerns identified in this report. Among the statutes that may have authority to address the
indirect foodchain risks associated with CKD are the Clean Water Act (stormwater management
regulations), the Clean Air Act (the program defining the National Emissions Standards for
Hazardous Air Pollutants), and the Toxic Substances Control Act (which gives the Agency
authority to issue appropriate regulations to address the risks from hazardous chemical
substances or mixtures). In particular, when fully implemented the Agency's recently
implemented stormwater control regulations could substantially mitigate damages related to the
surface water pathway. These alternative authorities are being explored and a decision to pursue
regulation of CKD under one or more of these statutes may form the basis for a decision that
Subtitle C regulation of CKD may be limited or even unwarranted.
Whether or not the Agency lifts the exemption, dust suppression and stormwater
management at facilities that burn hazardous waste, as well as on-site CKD management
practices at all other facilities would be subject to current and potential future regulation under
-------
10-8
the Federal Clean Air and Clean Water Acts, and where such provisions exist, all applicable state
laws and regulations. Damages at existing CKD disposal sites also could be addressed by RCRA
§7003 and CERCLA §104 and §106, if the site posed an imminent and substantial danger to
human health and the environment.
Option 1: Retain the CKD Exemption.
Since CKD exhibits low inherent toxicity and poses minimal risk when evaluating the
various exposure pathways using average or best-case conditions, it may be appropriate to retain
the exemption for cement kiln dust waste, that is, maintain the status quo. Under this option,
CKD management would continue to be regulated by the states, if at all.
Option 2: Retain the CKD Exemption, but enter into discussions with the industry,
in which they voluntarily implement dust recycling technologies, reduce
waste, and monitor and control certain off-site uses.
Since certain management scenarios may present risks when assuming plausible worst-
case conditions and pollution prevention alternatives may be promising in certain instances, the
Agency could enter into discussions with the cement manufacturing industry to urge it to
implement selected waste minimization/pollution prevention technologies or implement, more
environmentally protective management practices, including controlling certain off-site uses.
For example, some of the potential higher risk situations that have been identified in the
hypothetical scenarios relate to on-site CKD management and derive from CKD releases from
waste piles or other points via wind-blown dust or stormwater run-off or a combination of the
two. These contaminant release situations may be controllable (and at some facilities are
currently being controlled) at relatively low cost by careful location of the waste pile and active
use of conventional dust suppression and stormwater management practices. The Agency would
hold discussions with the industry to encourage them to voluntarily agree to implement these
practices.
An exception to the above conclusion would appear to be the 15 percent or so of cement
plants where CKD waste is managed in areas of karst topography or other areas characterized by
flow in fractured or cavernous bedrock, where leachate may directly percolate to ground water
with little or no attenuation. For some of these facilities, the ground-water pathway may become
of increased concern, depending on other site-specific considerations. Again, EPA would discuss
with the industry opportunities to either use appropriate liners or relocate the CKD management
unit.
About 20 percent of current net CKD generation is used off-site for a wide variety of
purposes, most of which according to the Agency's risk assessment do not pose human health or
other risks. However, the use of raw CKD containing higher measured levels of certain metals
and/or dioxins as a direct substitute for lime on grazing fields, agricultural fields, and gardens can
concentrate toxic constituents in crops and animal products at levels of concern for human
health. This use of CKD, though not widely practiced at present, is otherwise not currently
controlled, and may warrant further consideration by the Agency.
The Agency, under this option, could also develop guidance for states regarding site
management, off-site uses, and pollution prevention and waste minimization technologies. This
-------
10-9
guidance would assist states in reducing the potential risks posed by mismanagement of CKD and
recommend implementation of technologies that would promote recycling of CKD.
Under this option, CKD management would not be controlled by the provisions of
RCRA Subtitle C. However, since the exemption for CKD remains in place, CKD generated in
kilns that burn hazardous waste would still be subject to the two-part test for residuals under 40
CFR 266.112. If CKD does not pass the two-part test, it would be treated to standards for land
disposal (40 CFR 268.43) and disposed in a Subtitle C facility. Damages at existing CKD
disposal sites would still be addressed by RCRA §7003 and CERCLA §104 and §106, if the site
posed an imminent and substantial danger to human health and the environment.
Option 3: Remove the CKD Exemption but delay implementation for some period of
time (e.g., two years), that would allow industry time to employ pollution
prevention options.
While CKD may not present risks when evaluating the various exposure pathways using
average or best-case conditions, CKD may pose a potential danger to human health and the
environment if managed in certain ways under a limited set of exposure pathways assuming
plausible worst-case conditions. Also, damages to the environment resulting from poor CKD
management practices have been recorded and are continuing to occur at some facilities. For
these reasons, removing the Bevill exemption (codified at 40 CFR 261.4(b)(8)) may be
appropriate. Accordingly, provisions of the Boiler and Industrial Furnace rule (40 CFR 266.112)
would no longer apply to hazardous waste-derived CKD.
Under this option, on-site CKD management practices at those facilities with dust that
exhibited any of the RCRA hazardous waste characteristics, or CKD derived from the burning of
listed hazardous wastes (see 40 CFR 261.3(c)(2)(i)) would be affected by the provisions of
RCRA Subtitle C. CKD disposal piles which are inactive on or before the effective date of the
Final Rule would be unaffected by the provisions of Subtitle C, unless subsequently managed.
By delaying lifting the exemption for some period of time (e.g., two years after the
Regulatory Determination), industry would be provided an opportunity to implement pollution
prevention alternatives and thus, manage the hazardous waste management costs they would
incur. During this interim period between submittal of the Report to Congress and the effective
date of the Final Rule, the CKD exemption would still be in effect. The Agency believes that
many of the affected facilities would utilize the time to adopt pollution prevention technologies
which would reduce, if not eliminate the amount of hazardous CKD they generate, or stop
burning hazardous waste.
Once the exemption is removed, CKD generated from cement manufacturing facilities
that burn RCRA hazardous wastes would be RCRA hazardous waste under the derived-from
rule (40 CFR 261.3(c)(2)(i)). The goal of avoiding Subtitle C compliance costs would provide an
incentive for each facility to look for pollution prevention alternatives to recycle their CKD and
reduce the amount of hazardous waste generated. The Agency is requesting additional
information on the viability of the CKD recycling options discussed in the RTC and any other
available pollution prevention or recycling option not considered in the Report.
Those facilities that do not burn hazardous waste would not generally be affected by
removing the exemption unless they generated characteristic RCRA hazardous waste. The
-------
10-10
Agency expects the number of non-hazardous waste burning facilities affected by this option
would be small, since CKD rarely exhibits a characteristic of hazardous waste. These facilities
would have an incentive to control their cement manufacturing process to avoid generating
characteristic CKD.
Option 4: Remove the CKD Exemption, and rely on existing hazardous waste rules
to control cement kiln dust
This option is similar to Option 3, except the exemption would be removed in accordance
with RCRA §3010(b). (Under Subtitle C of RCRA, wastes brought under regulatory control
have up to six months from the Regulatory Determination before they become subject to
hazardous waste control.) Thus, CKD that is hazardous waste-derived or exhibits a RCRA
hazardous characteristic would be made subject to the provisions of RCRA Subtitle C.
Otherwise, this option is the same as Option 3.
Option 5: Promulgate Regulatory Standards for the Management of CKD Waste.
As previously stated, the Agency's analysis of the risks associated with cement kiln dust
suggest that by merely lifting the exemption at 40 CFR 264.1(b)(8), certain pathways of potential
concern under the hypothetical scenarios may not be adequately addressed under Options 3 and
4, should EPA decide that Subtitle C regulation is warranted for CKD in the first instance.
Specifically, EPA's risk assessment indicates indirect foodchain risks are of potential concern
from releases of CKD from disposal piles to nearby surface waters and crop lands and from the
direct application of CKD to croplands as a soil amendment assuming reasonable worst-case
conditions. The Agency acknowledges, as discussed in detail in Chapter 6, that these modelled
risks, while plausible, are of probably minimal incidence.
As described above, the likely regulatory result under Options 3 and 4 would be to make
CKD generated by a kiln that burns listed hazardous wastes itself a hazardous waste under the
derived-from rule (40 CFR 261.3(c)(2)(i)). The indirect foodchain risks potentially identified in
this Report, however, are not associated only with CKD generated by hazardous waste burning
kilns. As a result, EPA is also considering regulatory mechanisms that would specifically address
these risks, including promulgating regulatory standards under Subtitle C for the management of
CKD waste that would provide adequate protection against these risks.
RCRA §3001(b)(3)(C) provides that EPA shall within six months of the RTC "determine
to promulgate regulations under this subchapter... or determine that such regulations are
unwarranted." The statute does not describe the type of regulation that EPA should consider
promulgating (if any), other than that such regulation be under Subtitle C of RCRA. For
example, RCRA does not expressly direct EPA to determine whether to list CKD as hazardous,
as required for other wastes under the mandates in RCRA §3001 (c). Furthermore, RCRA
§2002(a) gives the Administrator the broad authority to "prescribe ... such regulations as are
necessary to carry out his functions under this chapter." The Agency believes it has the authority
where appropriate to promulgate federally-enforceable regulatory standards under Subtitle C for
the management of CKD. EPA could explore mechanisms for imposing regulatory standards for
CKD, e.g., under grant of rulemaking authority under 3001(b)(3)(C). Alternatively, EPA could
-------
10-11
consider conditioning the CKD exemption from the definition of hazardous waste (40 CFR
261.4(b)(8)) on compliance with appropriate management standards.
EPA could promulgate minimally burdensome management standards for cement kiln
dust that would adequately control the indirect foodchain risks, such as: (1) requiring that dust
piles be kept covered to control fugitive emissions and institute surface water run-off and erosion
controls; (2) maintaining ground-water protection, perhaps by requiring that CKD piles be
maintained on a non-earthen base or by requiring a liner; and (3) establishing risk-based
concentration thresholds for all constituents of concern (including 2,3,7,8-TCDD, arsenic,
cadmium, and lead) for CKD used as a direct soil amendment. Additional or alternative
standards may be appropriate, and EPA welcomes comments and suggestions on this aspect of its
options.
Of the five options being considered by the Agency, Options 3, 4, and 5 would provide
more control through implementation of the provisions of Subtitle C. The principal difference
between Options 3 and 4 is the timing of the implementation of the regulatory controls. Option
3 provides industry additional time to implement waste minimization/pollution prevention options
and more protective CKD management standards. Option 4 would bring CKD under Subtitle C
regulatory control more quickly. Removing the exemption also would impose regulatory equity
between CKD generated from kilns that burn RCRA hazardous waste and residues from other
incinerators that burn RCRA hazardous waste that do not have such an exemption. Option 5
would provide management standards to control all CKD, and would be targeted to specifically
address only those risks of potential concern.
The Agency did not evaluate the risk from the land application of agricultural lime, so it
cannot determine whether there is an increase in incremental risk when CKD is substituted. In
any event, CKD-sewage sludge derived fertilizers and soil amendments are considered safe for
such uses as fertilizer and pose minimal risk because these final products are required to be
tested to assure they comply with all provisions of 40 CFR 503, which are fully protective of
human health and the environment. It should be noted that if the exemption is removed,
fertilizer that is derived from CKD generated from a kiln that burns listed hazardous waste is
itself a hazardous waste under the derived-from rule (40 CFR 261.3(c)(2)(i)); the extent of
regulation, however, is limited (see 40 CFR 266.20(b)).
In addition, it should also be noted that under current rules, if CKD is recycled, the
resulting clinker is not automatically subject to the provisions of Subtitle C. By removing the
exemption, however, clinker may be affected by the derived-from rule (40 CFR 261.3(c)(2)(i)) if
the kiln burns listed hazardous waste, thereby becoming a hazardous waste. The Agency has not
yet fully analyzed available data on trace constituents in clinker. Based on our understanding of
current data, however, the Agency does not believe that clinker produced from kilns that burn
listed hazardous waste generally poses a hazard to human health and the environment. The
Agency is, therefore, considering crafting appropriate regulatory language for clinker. The
Agency, however, is interested in receiving comment on this issue.
10.3.3 Next Steps
After an evaluation of public comments on this RTC, the Agency will, in accordance with
RCRA §3001(b)(3)(C), reach a final Regulatory Determination on the management status of
CKD within six months of submission of this Report. The Regulatory Determination requires
-------
10-12
the Agency only to determine to promulgate regulations under Subtitle C, or determine that
Subtitle C is unwarranted. Thus, if RCRA §3004(x) or Option 5 is chosen, EPA would have
time beyond six months to promulgate a Final Rule.
-------
GLOSSARY
acid mine drainage - Water draining from closed or abandoned mines that is highly acidic, often
due to high concentrations of acidic sulfates.
acid neutralization - Reaction of an acid with a base to lower the pH of the base, i.e., make it
more neutral.
acidic soils - Soils exhibiting a pH of less than 7.
aggregate - A mixture of mineral substances (e.g., sand, gravel, crushed rock, slag, etc.) which,
when cemented, forms concrete. Uncemented aggregate can be used as a component of road
pavement and in manufacturing processes.
agricultural lime substitute - Any substance used as a substitute for lime to add calcium oxide to
soils, usually to control pH.
air pollution control devices - Devices used to limit dust emissions from the kiln system to the
atmosphere. Dust collection systems at cement plants generally involve a combination of
electrostatic precipitators, fabric filters arrayed in baghouses, cyclones, gravity/inertial separators,
and granular bed filters.
air districts - California air quality management districts that share the responsibility with the
California Air Resources Board, and local or regional air pollution control districts, to set air
pollution control standards.
alkali volatilization - A technique to recover alkali from the surface of CKD particles. This
technique generally involves subjecting the CKD to a high temperature, then condensing the
resulting alkali vapors from the hot gases onto a cooler surface.
alkali-aggregate reactivity - Susceptibility of aggregate to the alkali-aggregate reaction, a
chemical reaction in either mortar or concrete between alkalies (sodium and potassium) from
Portland Cement or other sources and certain constituents of some aggregates. The reaction
may cause deleterious expansion of the concrete or mortar.
alkalies - "Impurities" in the mixture of raw material (and to some extent, fuels) used in a
cement kiln, including the univalent, mostly basic metals of group I of the periodic table, and
their oxides.
alkaline materials - Water-soluble materials that yield a high concentration of hydroxyl (OH")
ions in solution, or otherwise produce a solution with a pH > 7 when dissolved.
alum - Aluminum sulfate, commonly used as a coagulant in wastewater treatment processes.
applicable and relevant or appropriate requirements (ARARs) - Standards used to establish
cleanup levels at Superfund sites.
argillaceous materials - Term applied to rocks that contain silt- to clay-sized sediments, which
often contain a high percentage of organic materials and a high clay mineral content.
-------
A-2
as generated CKD - Term used to describe a sample of CKD collected after it has exited the
kiln, but before it has been returned to the kiln system or placed in a CKD management unit.
as managed CKD - Term used to describe a sample of CKD collected from a CKD management
unit (e.g., a waste pile).
aquifer - A subsurface formation containing water in quantities sufficient to be withdrawn. A
useable aquifer is one that may be used for agricultural and industrial purposes as well as human
consumption.
background concentrations - Ambient concentrations of naturally occurring or anthropogenic
chemicals present in the environment not due to CKD management. These concentrations are
used as baseline levels to compare with chemical concentrations measured in CKD.
baghouse filter - Large fabric bag, usually made of glass fibers, used to eliminate intermediate
and large (greater than 20 microns in diameter) particles. This device operates in a way similar
to the bag of an electric vacuum cleaner, passing the air and smaller paniculate matter, while
entrapping the larger particles. The porous structure of baghouses is generally a woven or felt
fabric with a retention efficiency that improves as the interstices fill with captured dust, but with
the negative effect of increased flow resistance. Thus, regular filter cleaning is required to
maintain overall efficiency.
baghouse waste dust - Particles removed from the porous structure of a baghouse.
ball mill - A type of size reduction equipment used in the cement industry for the grinding of
raw materials and clinker. A ball mill consists of a rotating cylinder containing steel or ceramic
balls that are used to break up materials placed into the mill.
best management practices - Structures (such as storage ponds and infiltration trenches)
designed to receive and contain stormwater, and/or procedures, methods, and devices, that
improve or maintain the quality of environmental media.
Best Available Control Technology (BACT) - Air pollution controls that achieve the "maximum
degree of [emission] reduction ... which the permitting authority, on a case-by-case basis taking
into account energy, environmental, and economic impacts and other costs, determines is
achievable for such facility..." To obtain a Prevention of Significant Deterioration (PSD) permit,
a source must demonstrate that it will use BACT to reduce emissions for each pollutant subject
to regulation under the Clean Air Act.
Bevill Amendment - Section 3001 of RCRA, which temporarily excludes CKD (and other specific
waste categories) from regulation as a hazardous waste under Subtitle C of RCRA, pending
study. While temporarily excluding CKD from regulation as a hazardous waste, the Bevill
Amendment does not preclude CKD regulation under other provisions of federal or state law.
bioaccumulation - The net uptake of a chemical in the environment into biological tissues via all
exposure pathways. It includes the accumulation that may occur by direct exposure to
contaminated media (e.g., dermal absorption, ingestion) as well as exposure from food. This
phenomenon can result in higher concentrations of substances in biological tissue than in
surrounding environmental media.
-------
A-3
Boiler and industrial furnace (BIF) regulations - Regulations that require owners and operators
of hazardous waste-burning boilers and industrial furnaces to limit the emissions of toxic metals,
carbon monoxide, hydrogen chloride, chlorine gas, and paniculate matter. Cement kilns are an
example of boilers and industrial furnaces.
brine - A concentrated solution of inorganic salts, formed by the partial evaporation of saline
waters.
British thermal units (Btus) - A unit of measurement of heat; one Btu will raise the temperature
of one pound of water one degree fahrenheit.
bulk constituents - For purposes of this report, constituents that exceed 0.05 percent by weight in
CKD. Bulk constituents are primarily those found in clinker, though they also may be present at
levels in CKD that are unacceptable in the cement.
burnability - The requirements in terms of time, temperature, and fuel to process the raw
material used in a cement kiln.
calcareous materials - General term used for rocks containing calcium carbonate, such as
limestone and dolomite, and for materials that are regarded as basic or alkaline.
calcination - Heating an ore, mineral product, or intermediate product in a furnace or kiln to
decompose carbonates or intermediate compounds to CO2 and associated oxides. In a cement
kiln, this process occurs at a material temperature range between 1,480 and 2,190°F.
cancer/noncancer risks - The increased probability resulting from exposure to a hazardous
substance of an individual or population experiencing cancer or adverse, noncancer effects.
carbonator - Wastewater treatment used to remove excessive calcium concentrations.
Wastewater is run through carbonators in which the calcium will combine with carbon dioxide
and precipitate out as CaCO3.
carcinogenic - A substance that produces or incites cancer.
cement clinker - The material that is formed when heating the raw materials used in the
production of cement, usually limestone and clay, to approximately 2,700°F. Clinker is granular
and variable in size, and is usually cooled and ground with a smaller amount of gypsum to form
cement.
cement compression strength - The resistance of cement to rupture under compression,
expressed as force per unit area.
cement kiln fuel - Fuel used in the process of producing cement. The elevated combustion
temperatures involved in cement production require fuels with a high heat content (e.g., fossil
fuels).
cement plants - Facilities that produce commodities by burning mixtures of limestone and other
minerals or additives at high temperature in a rotary kiln, followed by cooling, and finish mixing
and grinding.
-------
A-4
cement workability - The capacity of cement for being shaped.
central tendency risk estimate - The best estimate of risk. This represents the risk to which most
of the population may be exposed.
CERCLA - Comprehensive Environmental Response, Compensation, and Liability Act, also
known as Superfund.
chlorination roasting - Addition of chlorine compounds (e.g., sodium or calcium chloride) to the
raw feed, followed by a reaction in the kiln with other raw materials to produce potassium
chloride (KC1). The KC1 is sublimed and collected with the CKD, raising the K2O content.
chlorinated dibenzofurans - A class of organic compounds that are listed hazardous wastes under
the following RCRA waste codes: F020, F021, F022, F023, F026, F027, F028, and F032 (see 40
CFR Part 261). These compounds are also important combustion products from the burning of
hazardous waste fuels.
CKD - A fine grayish material created in the process of calcining feed to produce cement clinker.
The primary byproduct of the production of cement.
CKD cohesion - Molecular attraction by which the particles of CKD are united throughout the
mass.
CKD management practices - Methods used to manage gross CKD generated by a cement plant.
These generally include direct recycling, treatment and return to the kiln system,
landfilling/stockpiling, and/or beneficial use.
CKD shear strength - Measure of the internal force in CKD tangential to the section on which it
acts.
CKD beneficial use - The use of CKD in a variety of productive applications including soil
stabilization, land reclamation, waste remediation, sewage sludge stabilization, agricultural
applications, lime substitution, and construction.
clarification - The removal of paniculate matter, chemical floe, and precipitates from suspension
in a fluid through gravity settling.
clarifier - A tank in which solids are settled to the bottom and are subsequently separated from a
fluid as a sludge.
clay - Sedimentary particles with diameters less than 1/256 mm. When in a formation, clay layers
are usually less permeable than other sedimentary materials such as sandstone or limestone.
clinker cooler - After leaving the kiln, the clinker is cooled in a rotary, planetary, or grate-type
clinker cooler using air pulled into the unit by dedicated cooler fans, and then transferred by
conveyor to the finish mill.
closure and post-closure care plan - A written plan that identifies and describes the steps that
will be carried out to close, dismantle, decommission, and/or reclaim a waste management unit at
a cement plant.
-------
A-5
co-combustion residues - Residues derived from the burning of hazardous waste fuels in
conjunction with fossil fuels.
coagulation processes - Processes by which a colloid is made to come out of solution by
aggregation. In water treatment, CKD can be substituted for lime in coagulation processes.
congener - A term used to refer to any one member of the same chemical family. For example,
there are 75 congeners of chlorinated dibenzo-p-dioxins, seven of these congeners have chlorine
substituted at the 2, 3, 7, and 8 carbon atoms.
coke-petroleum coke - The solid, cellular, infusible material remaining after the carbonization of
coal, pitch petroleum residues, and certain other carbonaceous materials. The coke used by
cement kilns is typically petroleum coke.
combustion air - Air that is introduced into the hot end of the kiln, reacted with the kiln fuel,
and drawn upward into the cool or "feed" of the kiln.
compression test - A method used to measure the amount of force that can be applied to an
object of known area before failure.
concrete durability - The ability of a material (concrete) to function properly and resist
destruction over a long period of time, including resistance to: failure under load, weathering,
freezing and thawing, corrosion, rotting, abrasion, and changes in properties depending on the
environment in which the material is being used.
confined aquifer - An aquifer that is overlain by a confining bed, which has significantly lower
hydraulic conductivity than the aquifer.
constituent mobility - The tendency of a substance to move through the environment.
Commonly used in this volume to refer to the mobility of a substance in ground-water systems
based on the substance's soil-water partition coefficient (Kj), with lower K,,s related to more
mobile substances.
constituent persistence - The tendency of a substance to remain in the environment. Generally
based on a substance's half-life in water, air, and soil. Substances with longer half-lives (i.e.,
more persistent) may present a greater hazard.
contaminant plume - A body of contaminated ground water spreading from a surface or
subsurface source of contamination.
corrosivity - One of the four characteristics of hazardous waste as defined by EPA, based upon
pH values of less than 2.0 or greater than 12.5 (see 40 CFR §261.22).
cross media impacts - Results of industrial or waste management activities that affect at least
two of the three primary environmental media - air, ground water, and surface water.
crystallizer - Device used to separate the components of a solution, mixture, or slurry into a solid
and a liquid phase by the application of cooling, evaporation, or other means.
-------
A-6
cyclone - An air pollution control device in which a vortex within a collector propels particles to
deposition areas for removal. Cyclones generally deposit the collected paniculate matter into a
hopper for eventual collection.
damage cases - Situations in which damage to human health and/or the environment has been
proved.
dehydration - The removal of bound water or hydrogen and oxygen from a chemical compound
in the proportion in which they form water. CKD has a chemically dehydrated nature, which
results from the thermal treatment it receives in the kiln system.
densification - The process of making CKD more dense. For example, dust compaction involves
the densification of waste material to increase available disposal space and ameliorate dust
migration.
destruction and removal efficiency - Percentage of organic hazardous constituents in a waste fuel
that are destroyed when the fuel is burned, e.g., in a boiler or industrial furnace.
diatomaceous materials - Natural deposits from a past lake or deep sea environment, composed
of silica, which are used as a mild abrasive, as a filtering medium, and for insulation of boilers
and blast furnaces.
dilution and attenuation factor (DAF) - A factor used to account for the decrease in
concentration of a substance after it is released from a waste management unit, mixes in
environmental media (e.g., ground water or surface water), and migrates to a location where a
person, plant, or animal might be exposed. As used in this report, mathematically equal to the
concentration of a substance in CKD divided by the estimated concentration of the substance at
a point of possible exposure.
dioxin - Any of a family of compounds known chemically as dibenzo-p-dioxins. Tests on
laboratory animals indicate dioxins are among the more toxic man-made chemicals known. Some
dioxins are also potent carcinogens, mutagens, and/or teratogens. Concern about dioxins arises
from their potential contamination of commercial products.
dispersion modeling - Mathematical simulation of the transport of a pollutant away from its
source in one or more environmental media - air, soils, ground water, or surface water.
disposal ponds - On-site CKD management method in which CKD is stored underwater. Use of
ponds creates a permanent hydraulic head on the dust.
distillery sludge - Residues from the removal of alcohol (by distillation) from fermented grain
mash.
dolomite - Carbonate rock (dolostone) composed of calcium/magnesium bicarbonate
(CaMg(COj)2), which is frequently used as a building stone and in the manufacture of bricks for
furnaces.
dry process kilns - Cement kilns in which raw materials are ground, conveyed, blended, and
stored in a dry condition. The dry raw mix is pneumatically pumped to the upper end of the
-------
A-7
kiln. As in all kilns, the raw materials flow down through the sloped kiln as they are thermally
treated. Dry process kilns are shorter and more thermally efficient than wet kilns.
dust compaction - A method of dust suppression by compressing CKD to reduce the level of
ambient suspendable dust.
dust screw conveyors - Mechanical devices used to transport gross CKD from air pollution
control equipment to a storage tank or other components of a cement facility.
dust suppression/control - Any means of reducing the level of ambient breathable dust.
Commonly used controls include wetting, compacting, or covering CKD.
effluent - Waste materials discharged to the environment, often in liquid form, such as treated
wastewater from a treatment plant.
electric arc furnace dust - Emission control dust or sludge from the primary production of steel
in electric arc furnaces. Electric arc furnace dust is a listed hazardous waste under the RCRA
waste code K061 (see 40 CFR Part 261).
electrodialysis - A process that uses electrical current and an arrangement of semi-permeable
membranes to separate soluble minerals from water.
electrostatic precipitator - An air pollution control device that generates one or more high
intensity electrical fields that cause particles to acquire an electrical charge. These charged
particles migrate to a collecting surface that has the opposite electrical charge.
emission offset - A reduction in emissions of a nonattainment pollutant from an existing source
(or sources) in the same area as a prospective new source of the same pollutant. Emission
offsets are required in all nonattainment areas in the U.S.
emissions testing - Sampling and analysis of air emissions from an industrial facility to measure
pollutant concentrations.
environmental media - One or more of the following: air, soils, ground water, or surface water.
EP method - Laboratory test used to determine whether a material exhibits the hazardous waste
characteristic of toxicity. Materials that are shown to leach one of 14 hazardous constituents at
concentrations exceeding 100 times primary drinking water standards are considered EP toxic.
The 14 hazardous constituents include arsenic, barium, cadmium, chromium, lead, mercury,
selenium, silver, endrin, lindane, methoxychlor, toxaphene, 2,4-D, and 2,4,5-TP (Silvex). This
method has been superseded by the Toxicity Characteristic Leaching Procedure (TCLP).
exothermic reaction - Chemical reaction in which heat is evolved.
exposure pathways - The course a substance takes from the point where it is released into the
environment (i.e., the source) to an exposed organism. Each exposure pathway includes a source
or release from a source, an exposure point (a location of potential contact between an organism
and a substance), and an exposure route (the way a substance comes in contact with an
organism, such as ingestion, inhalation, and dermal contact).
-------
A-8
exposure potential - The likelihood of individuals, resources, or populations being exposed to
CKD contaminants.
fabric filters-baghouses - Fabric filters are an air pollution control device in which filters remove
paniculate matter from gas streams by retaining the particles in a porous structure. Fabric filters
are typically used in series to form a baghouse.
fate and transport modeling - Modeling used to predict what will happen to chemicals released
to the environment. These models incorporate chemical, physical, and biological
transformations, bioaccumulation, and ease of transport based on various physical and chemical
properties to predict the movement and fate of each chemical in the environment.
Federal drinking water standards - Contaminant limits for water destined for human
consumption established by EPA pursuant to the Safe Drinking Water Act (SDWA).
filter press cake - The semi-solid residue left after a slurry is passed between the plates of a filter
press.
finish mill system - Equipment used to convert clinker into finished cement. At the finish
milling stage, Portland Cement is produced by grinding clinker together with about five percent
gypsum to a fine powder, and then loading it into bulk carriers or packaging it into bags. It is at
this stage that various additives along with gypsum are introduced to create specialty Portland
Cements.
floe - A clump of solids formed in wastewater treatment by biological or chemical action.
flocculation - Process in which aggregates of solids are formed and settled during wastewater
treatment by biological or chemical action.
floodplain - The land areas adjacent to a stream or river consisting of unconsolidated sediments
that are occasionally covered by water during and after storm events.
flue gas desulfurization (recovery scrubber) - CKD treatment technology that enables all CKD to
be recycled as kiln feed by removing alkalies, chlorides, and sulfates from the dust.
fluid bed recovery process (Fuller process) - Process that thermally treats CKD for recovery.
The process is designed to accept CKD generated from a kiln or from stockpiles, pelletize it, and
calcine it into clinker on a fluid bed instead of in a typical rotary kiln. This process yields a
usable cement clinker product rather than treated CKD.
fluorspar - Material composed of fluorite that is often used as a flux in the smelting of iron or in
the chemical industry, and is used as a secondary feed material for cement plants.
fly ash - Non-combustible residual particles from the combustion process, carried by flue gas.
fractional crystallization - The removal of early-formed crystals from an originally homogeneous
mixture so that these crystals are unable to react further with the parent mixture (e.g., removal of
salt from alkaline CKD wastewater for beneficial use).
-------
A-9
freeze-thaw durability - The ability of a material to resist destruction from repeated cycles of
freezing and thawing.
fugitive dust - CKD constituent particles suspended in the air by either wind erosion or
mechanical disturbances.
fugitive air emissions - Pollutant emissions into the air not caught by a control system.
grab sample - A single sample of a material (e.g., soil, CKD) that is collected for laboratory
analysis.
granular bed Filters - Air pollution control device in which dust is captured and bound on a
porous medium through the principle of adsorption. The most commonly used medium is
granular activated carbon.
gravity/inertial separator - Air pollution control device that collects paniculate matter by gravity
or centrifugal force, but does not depend on a vortex as do cyclones.
gross CKD - The dust collected at the air pollution control device(s) associated with a kiln
system. Gross CKD is generated as an inherent process residue at all cement plants.
ground water - The water contained within the pore spaces of subsurface formations below the
water table and within the zone of saturation.
ground-water monitoring - Recording of water levels in wells or water quality of samples taken
from wells to study either a site specific area (e.g., adjacent to a landfill) or a regional area
(county to several county level).
gypsum - Insoluble, evaporite mineral composed of calcium sulfate.
halogen - Any of a group of five chemically-related nonmetallic elements that includes bromine,
fluorine, chlorine, iodine, and astatine.
hazardous waste - According to Federal Law (40 CFR 261) a solid waste, or combination of
solid wastes, which, because of its quantity, concentration, or physical, chemical, or infectious
characteristics, may (1) cause, or significantly contribute to, an increase in mortality or an
increase in serious irreversible, or incapacitating reversible, illness; or (2) pose a substantial
present or potential hazard to human health or the environment when improperly treated, stored,
transported, disposed of, or otherwise managed.
hazardous waste stabilization - The use of a substance (e.g., CKD) as a dewatering and
solidifying agent prior to the land disposal of sludges containing hazardous wastes.
health-based levels (HBLs) - As used in this volume, risk-screening criteria for ground water and
surface water developed by EPA using chemical-specific lexicological values and equations for
calculating preliminary remediation goals for ground and surface water at Superfund sites.
heavy metals - Metallic elements with high atomic weights, e.g., mercury, chromium, cadmium,
arsenic, and lead. They can damage living things at low concentrations and tend to accumulate
in the food chain.
-------
A-10
high sulfur fuel - Kiln fuel containing high levels of sulfur compounds. During combustion of
high sulfur fuel, sulfur (in the form of SO3) will vaporize in the kiln to form sulfur dioxide (SO2),
and condense in the form of sulfates.
hydra ted lime - A dry powder obtained by hydrating quicklime (calcium oxide) with enough
water to form a hydroxide.
bydration - The formation of a compound by the combination of water with another substance.
hydraulic conductivity - A physical characteristic measuring the ability of water to flow through a
solid material, e.g., CKD.
hydrologically downgradient - The direction towards which ground water and surface water flow.
hydrolysis - The reaction of an ion with water to produce either hydronium ion (H3O+) or
hydroxide ion (OH").
ignitability - One of the four characteristics of a hazardous waste as defined by EPA (see 40
CFR Part 261). A solid waste is ignitable if it has the ability to combust at or near 140°F, can
cause fire through friction, is an ignitable compressed gas, or is an oxidizer.
induced draft fans - Large fans used to draw air into cement kilns causing kiln combustion gases
to flow countercurrent to the raw feed and exit the kiln.
infiltration - The flow of water downward from the land surface (e.g., as managed CKD) through
the upper soil layers, which may eventually lead to ground water resources.
inorganic constituents - Chemical substances derived from mineral sources that do not usually
contain carbon.
insufflation - The pneumatic introduction of unaggregated CKD into the hot end of the kiln.
ion exchange - Reversible substitution of ions in a crystal with other ions in solution, without
disturbance of the crystal lattice or its electrical neutrality. This occurs by diffusion, particularly
in crystals where weakly bonded ions form one- or two- dimensional channelways. Artificial ion
exchange resins with three-dimensional hydrocarbon networks are commonly used (e.g., in water
softeners; for separating isotopes; in desalination; and in the chemical extraction of elements
from ores).
iron flue dust - Secondary kiln feed used as an iron additive to achieve the desired consistency of
the general kiln feed, depending on the composition and availability of primary feed materials.
karst topography - A type of geographic terrain underlain by carbonate rocks (e.g., limestone)
where significant dissolution of the rock has occurred due to flowing ground water.
Characteristic features include sinkholes, caves, and streamless valleys.
kiln exhaust - Exhaust gases exiting the upper end of the kiln. Kiln exhaust is sometimes used to
pre-dry feed materials prior to conveying them to the kiln.
-------
A-ll
land reclamation - Process of changing the landscape back to conditions similar to those present
at a given location prior to a land altering activity (e.g., excavation of a landfill or quarry).
landfill cover - Material used to cap a landfill at closure. Low-permeability materials (e.g., clay,
plastic) are used to prevent rainwater or run-off from entering the landfill and increasing the
likelihood of movement of the contaminants from the landfill (e.g., leachate).
landfill liners - Low-permeability materials used in sealing a landfill to prevent leachate from
escaping beneath or laterally from a landfill. Liner materials range from bedrock and in-situ clay
to synthetic plastics.
landfill - A waste management unit in which material is disposed below topographic grade and is
sometimes buried between layers of earth.
leachate - Water that contains a high amount of dissolved solids and is created by liquid
percolating through layers of a landfill and collected by the landfill liner or seeps from a landfill
to surrounding soil and ground water.
leachate collection system - A system installed in conjunction with a liner to capture the leachate
that may be generated from a landfill so that it may be removed and treated.
leather tanning wastes - Chemical wastes generated from the treatment and preservation of hides
during leather production.
Lepol kiln - This preheater kiln, also known as a "grate kiln," begins with raw feed nodules
containing 10 to 15 percent moisture. In the semidry process, these raw material nodules travel
on a grate through a preheater in which they are partially calcined. This partly calcined material
then falls through a chute into the rotary kiln where final clinkering takes place. As a result of
the partial precalcination the rotary kiln can be one-third the usual length.
lignosulfate - Chemical additive that is used to retard the setting of CKD when it is hydrated and
thereby improve its flow characteristics.
limestone quarry - An open mine for extracting limestone from the earth for beneficial use.
limestone - A bedded sedimentary rock composed mainly of calcium carbonate, or a rock type
composed of, in general, at least 80 percent of carbonates of calcium and magnesium, which
yields lime when burned.
liming agent - A substance used to raise the pH of acidic soils.
low-alkali sand - Also known as "sweet sand," this material is used to balance the high alkali
content of the limestone quarried at some facilities and enable the plant to recycle a greater
percentage of its generated CKD.
low-sulfur coal - Coal that contains relatively low concentrations of sulfur and emits lower
quantities of sulfur oxides when burned.
-------
A-12
lowest observed adverse effects levels (LOAELs) - From dose-response relationships, LOAELs
are the lowest exposure level at which there are statistically or biologically significant increases in
frequency or severity of adverse effects between the control group and the test population.
mag rock - Limestone with a high magnesium content.
masonry cement - A hydraulic cement for use in mortars for masonry construction, containing
one or more of the following materials: Portland Cement, Portland blast furnace-slag cement,
Portland-pozzolan cement, natural cement, slag cement, or hydraulic lime; and usually containing
one or more additional materials (e.g., hydrated lime, limestone, chalk, calcareous shell, talc,
slag, or clay) as prepared for this purpose.
Maximum Contaminant Levels (MCLs) - The maximum permissible level promulgated under the
Safe Drinking Water Act for a contaminant in water that is delivered to any user of a public
water system. Primary MCLs are established in 40 CFR 141 to be protective of human health;
secondary MCLs are established in 40 CFR 143 to protect the aesthetic quality of drinking water
(e.g., taste, odor, color, and appearance).
maximum exposed individual (MEI) - The actual or hypothetical individual, who based on
location, sensitivity, and exposure pattern, is subject to the highest concentration of a substance,
and therefore has the highest reasonable risk. The MEI may vary by exposure pathway.
metric ton - A unit of weight measure equivalent to one million grams, or approximately 1.102
short tons.
microbiological stresses - Techniques used in the N-Viro soil technology to kill pathogens and
stabilize sludge to produce a "soil-like product."
mineral processing - Steps used to concentrate and refine mineral ores (raw or beneficiated) into
more useful forms.
minimum waste heating value limit - A minimum standard established by EPA for the burning
of hazardous wastes in cement kilns prior to the promulgation of the BIF rule. Operators of
cement kilns burning hazardous wastes with a heating value of at least 5,000 Btu/pound were
exempt from RCRA permitting requirements.
MMSOILS - A multimedia model used to estimate the concentrations of contaminants in the
environment, as well as human exposure and health risk, associated with the release of chemical
substances from waste sites. The model predicts the transport and fate of a chemical in ground
water, surface water, soil, the atmosphere, and the food chain.
monofill - A landfill that contains one type of waste, such as CKD.
N-Viro soil - The product of a patented waste treatment process that was developed in 1987, and
is used with lime or instead of lime to disinfect and deodorize municipal sewage sludge. N-Viro
soil contains between 35 and 75 percent CKD by weight.
National Priorities List - EPA's list of the top-priority hazardous waste sites in the country that
are subject to the Superfund Program, as established by the Comprehensive Environmental
Response, Compensation, and Liability Act (CERCLA).
-------
A-13
net CKD - CKD removed from the kiln system after collection.
New Source Performance Standards (NSPSs) - Federal air pollution control standards that must
be achieved by newly operational or significantly modified industrial facilities. These standards
include limits on paniculate matter emissions from cement plants.
nodulized - Refers to CKD that has been converted into small clumps or "nodules" by treating
the dust with water. A pug mill is often used for this purpose.
Non-hygroscopic - Not having the property of readily absorbing moisture from the atmosphere.
Nonattainment areas or status - Air quality control areas not in compliance with the National
Ambient Air Quality Standards for a given pollutant.
NPDES permits - EPA permits to discharge wastewaters from a point source into surface
waterways, issued under the National Pollutant Discharge Elimination System (NPDES).
oil sludge - The highly viscous fraction left over from petroleum refining that cannot be further
refined.
opacity - Refers to the percent decrease in the transmission of light measured in air emissions
from the stack of an industrial facility. Cement plants are often required to operate under
specific opacity limitations.
organic constituents - In a mixture of elements and/or compounds, those compounds that are
carbon-based.
osmosis - Movement of a solvent through a semipermeable membrane into a solution of higher
solute concentration that tends to equalize the concentrations of solute on the two sides of the
membrane.
overburden - Any unconsolidated material that overlies bedrock.
particle size distributions - The comparative amounts of particles of different diameters within a
defined volume.
participate CKD capture efficiency - A measure of the effectiveness of a CKD recovery
technology to remove CKD particulate matter from kiln exhaust gases.
Pearson Correlation Coefficient - A measure of the closeness to a linear relationship between
two variables. Also referred to as the true product-moment correlation.
pelletizing - A method in which finely divided material is rolled in a drum or on an inclined disk,
so that the particles cling together and form small spherical pellets. This technique is sometimes
used to treat CKD prior to disposal or recovery.
perched aquifer - Unconfined ground water separated from an underlying main body of ground
water by an impermeable unsaturated zone (e.g., clay layer).
-------
A-14
permeability - The capacity of subsurface strata to transmit a fluid, expressed as the rate at which
a fluid of standard viscosity (e.g., water) can move a specified distance. Permeability is
dependent on the size and shape of pores in the stratum or strata, the size and shape of
interconnections between pores, and the extent of these interconnections.
pH - The negative logarithm of the effective hydrogen ion concentration or hydrogen ion activity
in gram equivalents per liter used in expressing both acidity and alkalinity on a scale whose
values run from 0 to 14, with 7 representing neutrality, numbers less than 7 increasing acidity,
and numbers greater than 7 increasing alkalinity.
phosphogypsum - Phosphorus-containing gypsum produced by the controlled reaction of
phosphate rock with sulfuric acid during phosphoric acid production.
planetary cooler - Cylinder or series of cylinders attached radially to a rotary kiln that promotes
heat exchange between hot clinker and cold air. Planetary coolers have no moving parts and all
air passing through them enters the kiln and is used for combustion.
pneumatic conveyance - Transport of material (e.g., CKD) via forced air.
pollution prevention plan - Document that describes the implementation of practices to reduce
pollutants in stormwater discharges associated with industrial facilities.
Portland Cement Association - An industry organization representing most North American
cement manufacturers.
Portland Cement - A hydraulic cement produced by pulverizing Portland Cement clinker and
usually containing a small quantity of calcium sulfate.
pozzolan - A material rich in silica or silica and aluminum that is chemically inert and possesses
little or no value as a cementing agent, but, when in a finely divided form and in the presence of
water, will react with calcium hydroxide to form compounds possessing cement-like properties.
The most commonly available pozzolan in use in the U.S. is fly ash.
precalciner kiln - Suspension preheater kilns that are equipped with a secondary firing system
(flash furnace) attached to the lower stage of the preheating tower. These kilns are the most
recent advance in cement manufacturing technology.
preheater kiln-suspension - A type of kiln in which raw meal is preheated and partially calcined
by passing it through a system of heat exchange cyclones before it enters the kiln. This is the
most energy efficient type of kiln available.
pretreatment standards - Wastewater contaminant limits to be achieved by on-site treatment
before sending wastewaters to publicly-owned treatment works.
Prevention of Significant Deterioration - A key element of the Clean Air Act that establishes
non-degradation of airsheds with acceptable air pollutant levels as the first priority of state-level
air quality programs.
-------
A-15
process wastewater - Waters used or generated in one or more mineral processing units that
have accumulated contaminants to such an extent that they must be removed from the processing
unit(s).
products of incomplete combustion (PICs) • Compounds resulting from the incomplete thermal
breakdown and oxidation of organic chemicals.
pug mill - A mixer having a stationary cylindrical mixing compartment, with the axis of the
cylinder horizontal, and one or more rotating horizontal shafts to which mixing blades or paddles
are attached.
pulverized coal - Coal that has been crushed into small pieces to increase the surface area of the
coal and thus allow the coal to bum more rapidly.
quicklime - Calcium oxide (CaO).
radionuclides - Elements that emit alpha, beta, and/or gamma rays by the spontaneous
disintegration of atomic nuclei.
raw feed - Ground, proportioned, and blended raw materials that are conveyed into the upper
end of a cement kiln. These materials are generally comprised of about 80 percent carbonate of
lime and 20 percent silica, with much lower quantities of aluminum and iron.
raw mill systems - Components of the cement production process in which crushed raw materials
are fed into grinding mills, ground to a fine size range, and blended to obtain the correct
composition for kiln feed.
raw mill dryers - Equipment used in the dry process of cement manufacturing to reduce the
moisture content of ground, blended raw materials to less than one percent prior to feeding them
into the kiln.
RCRA - Resource Conservation and Recovery Act of 1976. The Federal statute that provides
EPA with the authority to regulate the treatment, accumulation, storage, disposal, and
reclamation of solid and hazardous wastes.
ready mix concrete - Concrete manufactured for delivery to a purchaser in a plastic and
unhardened state.
recovery scrubbing - CKD treatment technology also known as the flue gas desulfurization
(FGD) process. This technology enables all CKD to be recycled as kiln feed by removing
alkalies, chlorides, and sulfates from the dust.
refractory relining - The process of periodically replacing the kiln refractory, a material that is
used to protect the steel shell of the kiln from high temperatures generated during clinker
production. The kiln refractory usually consists of brick.
rehydration - Reincorporation of water that has been removed from a substance.
residual waste - Unused materials or byproducts of a process that have no immediate use.
-------
A-16
reverse osmosis - A flow of fresh water through a semipermeable membrane when pressure is
applied to a solution (e.g., seawater) on one side of it.
Ringelmann Chart - A series of shaded illustrations used to measure the opacity of air pollution
emissions. The chart ranges from light grey to black and is used to set and enforce emissions
standards.
riprap - A foundation, sustaining wall, or layer of stones, chunks of concrete, or similar durable
material placed on an embankment or slope to prevent erosion.
risk assessment - A formalized methodology for analyzing the adverse effects resulting from
releases of hazardous substances. Risk assessment generally includes the collection of data and
background information, an exposure assessment, a hazard assessment, a dose-response
evaluation, and risk characterization.
risk potential - As used in this volume, the potential for CKD management at cement plants to
contribute to adverse effects via the ground water, surface water, and air pathways, based on a
qualitative (i.e., non-modeling) analysis of factors that influence risk.
risk screening criteria - As used in this report, a set of chemical-specific benchmarks used to
compare to concentrations measured in CKD for the purpose of determining the intrinsic hazard
of the dust. Concentrations in CKD that fell below these criteria were judged to pose a low or
negligible risk that did not need further study. Concentrations above the criteria indicated that
more detailed study was needed to determine the risks associated with certain CKD constituents
and exposure pathways.
road base - Aggregate beneath the macadam of a road that acts as a support or substrate.
rotary coolers - A sloped, revolving cylinder that receives clinker from the kiln. Cooling is
achieved by drawing air through the cooler and into the kiln opposite to the flow of clinker.
Rotary coolers are frequently located underneath the rotary kiln and are often connected to the
kiln by a vertical shaft.
rotary kilns - Horizontal, inclined rotating cylinders, refractory lined and internally fired,
designed to produce clinker through the intense heating of raw materials.
saline soils - Soils that contain enough soluble salt to reduce their fertility. The lower limit is
usually defined as 0.4 Siemens per meter.
seeps/seepage - Springs/leakage to underlying aquifers through stream beds or the emergence of
ground water into a stream channel, but may also relate to flow between different aquifer units.
semi-volatile organic compounds - A class of organic compounds that have a moderate tendency
to vaporize.
semidry process kiln - see Lepol kiln
sewage sludge stabilization - The use of a substance (e.g., CKD) as a dewatering or solidifying
agent for sewage sludges prior to disposal or beneficial use, such as for fertilizer or soil
amendment.
-------
A-17
shale - Fine-grained, fissile, sedimentary rock composed of clay-sized and silt-sized particles of
unspecified mineral composition.
Shapiro-Wilk statistic - A value that is computed to test the null hypothesis that a group of data
values are a random sample drawn from a normal distribution.
sintering - Causing to become a coherent mass by heating without melting.
slag - The residue from the melting of metals or the reduction of ores.
slurry - A mixture of water and any finely divided insoluble material, such as Portland Cement,
CKD, or clay in suspension. Also, a watery mixture of insoluble material that results from
certain pollution control techniques.
slurry walls - A type of containment system that prevents leachate from migrating through
ground water systems. Typically, slurry walls are formed in place by excavating a trench outside
the edge of a waste management unit or ground water contaminant plume, mixing the removed
materials with grout (e.g., bentonite clay, asphalt, etc.) and immediately redepositing the slurried
mixture into the trench.
soil-water partition coefficient (KJ - The equilibrium ratio of a chemical adhering to soil to that
present in ground water. Reflects a substance's mobility in ground water, with substances with
low KjS expected to migrate faster in ground water than substances with high KjS.
soil amendment - Material added to soils to change their chemical characteristics to improve
crop production. For example, CKD is used to improve the quality of soil, including pH
adjustment, for agricultural purposes.
soil stabilizer - A material used to prevent soil from shifting, subsiding, drifting away as fugitive
dust, or eroding.
special wastes - Four categories of wastes, including CKD, for which EPA is required to defer
most RCRA Subtitle C requirements until comprehensive studies are presented to the U.S.
Congress, and the most appropriate regulatory approach is determined (RCRA Section
3001(b)(3)).
spent pickle liquor - A liquid waste generated by steel finishing operations of facilities within the
iron and steel industry. Spent pickle liquor is a listed hazardous waste under the RCRA waste
code K062 (see 40 CFR Part 261).
StableSorb - CKD marketed by Keystone Cement Co. as a sewage sludge dewatering agent.
stormwater run-on/run-off collection system - A system for preventing water from infiltrating
land-based waste management units (e.g., landfills, waste piles) during storm events. Such
systems typically include drainage ditches, land grading, impervious substances, and other
measures.
Synthetic Precipitation Leaching Procedure (SPLP) - A laboratory analytical method (No. 1312,
SW-846) that simulates land disposal of inorganic wastes in monofills, a situation that often
occurs at cement plants.
-------
A-18
t-test - A statistical technique used for hypothesis testing, e.g., to determine whether samples
have been drawn from the same population.
tertiary air duct - Conveyance used in certain precalciner kilns that runs parallel to the kiln and
supplies waste gases from the clinker cooler to the combustion zone in the preheater tower.
tests of proof - Criteria used in this report for evaluating prospective cases in which CKD
management has resulted in documented damages to human health and the environment.
total dissolved solids (TDS) - A measure of the dissolved solids in wastewater, effluent, or water
bodies. Dissolved solids are disintegrated organic and inorganic material contained in water.
Excessive amounts make water unfit to drink or use in industrial processes.
total suspended particulate-ambient - A measure of the concentration of solid particles present
in a specific place in the absence of new emissions.
total suspended solids (TSS) - A measure of the suspended solids in wastewater, effluent, or
water bodies, determined by using tests for "total suspended non-filterable solids." Suspended
solids are small, undissolved particles of solid pollutants.
toxicity - The degree of danger posed by a substance to animals or plant life.
Toxicity Characteristic Leaching Potential (TCLP) - A laboratory method (No. 1311, SW-846)
that simulates the generation and release of leachate from an improperly disposed solid waste.
This procedure is applied to solid wastes to determine whether they exhibit the hazardous waste
characteristic of toxicity.
unconfined aquifer - An aquifer characterized by the absence of an aquitard above it, so that the
water table forms the upper boundary of the aquifer and is free to move with atmospheric
influences such as atmospheric pressure. Also referred to as a water table aquifer.
uptake/biokinetic (UBK) model - An EPA-developed model used to estimate blood-lead levels in
children. The model considers intake from diet, direct inhalation, ingestion of dust, soil, paint,
and drinking water.
volatile organic compounds - A class of organic compounds that have a high tendency to
vaporize.
volatilization - The process of passing into vapor from a liquid state.
waste stabilization - Treatment with the following reagents (or waste reagents) or combination of
reagents to reduce the leachability of hazardous metals or inorganics: (1) Portland Cement; or
(2) lime/pozzolans (e.g., fly ash and CKD). This does not preclude the addition of reagents (e.g.,
iron salts, silicates, and clays) designed to enhance the set/cure time and/or compressive strength
(see 40 CFR 268.42).
waste management units - Locations at which wastes are treated, stored, accumulated, recovered
for reuse, and/or disposed. Waste management units include wastewater treatment plants,
surface impoundments, waste piles, landfills, and quarries.
-------
A-19
waste stream - Material containing potentially toxic constituents. In this report, the waste stream
of interest is CKD, with different cement plants generating their own streams of CKD with
unique chemical compositions.
water table - The level below which the soil or rock is saturated with water. It is also the upper
boundary of the saturated zone. At this level, the hydraulic pressure is equal to atmospheric
pressure. Also used to refer to an aquifer that exhibits unconfined conditions (i.e., a water table
aquifer).
wellhead protection area - An area delineated around and upgradient of a drinking water well in
which activities and substances that may result in contamination of a well are regulated.
wet process kilns - A clinker manufacturing process used in rotary kilns in which the feed enters
the kiln in the form of a slurry with a moisture content of 30-40 percent. In comparison with dry
process kilns of the same diameter, wet process kilns require an additional section (dehydration
zone) to drive off the water from the kiln feed. As a result, wet process kilns must be
considerably longer to achieve the same production rate.
wet scrubbers - Air pollution control devices that employ water sprays to remove sulfur oxides,
paniculate matter, and other air pollutants from exhaust (usually combustion) gases.
X-ray diffraction - Reflection of X-rays at definite and characteristic angles from crystal
structures, yielding data for identification of a given mineral species.
-------
BIBLIOGRAPHY
Abrons, M.D., et. al. 1985. Respiratory Effects of Portland Cement Dust. U.S. Department of
Commerce, National Technology Information Service.
Ahn, K.H., et al. 1988. Solidification of Hazardous Wastes: An Approach Using Cementitious
Binders. Paper From Proceedings of 1988 Pacific Basin Conference on Hazardous Waste.
February 1-6. Honolulu, Hawaii.
Air Resources Board, Stationary Source Division, Technical Review Group Emissions Credit
Systems and New Source Review Programs: A Report to the Legislature (Dec. 8,1988).
Alabama Department of Environmental Management, 1984. Letter from H.H. Beiro, Pollution
Control Specialist, to M.F. McCarthy, Lehigh Portland Cement Co., Leeds, Alabama.
April 19,1984.
Alabama Department of Environmental Management, 1985. Letter from Kirk S. Kreamer,
ADEM, to A.P. Mahatekar, AmTech Services, Inc., August 21,1985.
Alabama Department of Environmental Management, 1987a. Notice of Noncompliance from S.
Jenkins to M. McCarthy, Lehigh Portland Cement Company. February 3,1987.
Alabama Department of Environmental Management, 1987b. DMR Violation Report, 1st Qtr.
1987, for Lehigh Portland Cement Co., Leeds, Alabama.
American Society for Testing and Materials, 1987. ASTM C150-86, Vol. 04.02. Concrete and
Aggregates, p. 91.
American Concrete Institute, 1990. Cement and Concrete Terminology, American Concrete
Institute Committee 116, publication SP-19 (90), Detroit, Michigan.
Amtech Services, Inc., 1986. Letter from M. Holder, Professional Engineer, AmTech Services,
Inc. to P. Prysey, ADEM. July 24,1986.
Anonymous. 1980. Supplementing Ruminant Feeds with Cement Kiln Dust Improves Livestock
Performance, According to Agriculture Canada. Feedstuffs. June 2. p. 38.
Anonymous. 1981. Cement Kiln Dust Use as Lime-Potash Fertilizer. Farm Chemicals. April.
Anonymous. 1982. A Method For Oil Sludge Solidification and Disposal Using Cement Kiln Dust
Has Reclaimed 133 Acres Near Dallas, Texas. Waste Age. April.
Anonymous. 1984. Controlling NOx Emissions Can Improve Cement Quality in Kilns. Chemical
Engineering. April 30. pp. 18-19.
Anonymous. 1985. Mobile Hazardous Waste Treatment Services - What's Available? Hazardous
Waste Consultant. September-October.
Anonymous. 1985. VA-PASEC Precaldnation System Implemented for First Time. World
Cement. December.
-------
B-2
Anonymous. 1987. U.S. Briefs: Cogeneration: Santa Maria Aggregate Corp. Oil & Gas Journal.
August 10. p. 24.
Anonymous. 1988. EPA, Cities, Industry Press Congress For Incinerator Ash Legislation.
Cogeneration Report. April 22.
Anonymous. 1988. River Cement to Use Hazardous Waste as Fuel. Pit & Quarry. August, p.
150.
Anonymous. 1988. Toledo City Council Approves Five-Year Contract With National N-Viro
Technology, Inc. For Sewage-Agricultural Facility. PR Newswire. September 21.
Anonymous. 1988. Toledo Tries a Sludge First. Engineering News-Record. September 29.
Anonymous. 1988. Lone Star Introduces World's First Four-Hour Cement. Concrete Products.
December, p. 16.
Anonymous. 1989. Puerto Rican Cement Company. Caribbean Update. April, p. 15.
Anonymous. 1989. Waste Fuel System Installed at Ash Grove. Pit & Quarry. May. p. 18.
Anonymous. 1989. Portland Cement Reference Materials Available. Ceramic Bulletin.
September 22. p. 1105.
Anonymous. 1989. Titan Cement Company S.A. Eurobusiness. October.
Anonymous. 1990. Garbage In, Cement Out. Forbes. January 8. p. 189.
Anonymous. 1990. Eye on Recycling. PIMA. March, p. 16.
Anonymous. 1990. Industry Barometer. Pit & Quarry. May. p. 16.
Anonymous. 1990. Market Set for Decade of Change and Restructuring. Lloyd's List
International. May 31. p. 4.
Anonymous. 1990. Portland Cement Consumption by Type of Construction. Concrete Products.
June. p. 11.
Anonymous. 1990. New 2200 tpd Production Line for Ash Grove. World Cement. October 10.
pp. 464-465.
Anonymous. 1990. Southdown Buys Waste Processing Firms. Houston Business Journal.
October 8. Volume 20, p. 24.
Anonymous. 1990. Southdown Buys Hazwaste Units. Chemical Week. October 17. p. 6.
Anonymous. 1990. Cement Imports Hit Stone Wall Journal of Commerce. October 29. p. 13c.
Anonymous. 1991. 1990 Cement Totals. Pit & Quarry. March, p. 10.
-------
B-3
Anonymous. 1991. A Monthly Analysis of Trends in the Construction & Cement Industries (Issues
March-Nov. 1991). The Monitor.
Anonymous. 1991. Cement Producer to Buy 12% of Lincoln. Globe and Mail (Toronto).
February 15. p. B7.
Anonymous. 1991. Near Thomaston, ME, Commercial-Scale Tests Are Underway. Enironmental
Science and Technology. March, p. 358.
Anonymous. 1991. High Temperature Cement Kiln Offers a Balanced ''Industrial Ecosystems for
the Safe Recycling of Hazardous Waste. News Release. April 2.
Anonymous. 1991. Quicklime Volatilizes PCBs, EPA Finds. Superfund. June 28.
Anonymous. 1991. Chloride-Free Potash Fertilizer from Waste SO2 and Cement Kiln Dust.
Phosphorous and Potassium. July-August, p. 48.
Anonymous. 1991. "Home-Grown" Scrubber Protects the Environment of an Indian Reservation.
Sulphur. July-August, p. 24.
Anonymous. 1991. SO3 Technology Pays for Itself and Then Some. Coal and Synfuels
Technology. January 14. p. 1.
Anonymous. 1991. Potash Mining in Alsace, France. Phosphorus and Potassium. September-
October, p. 21.
Anonymous. 1991. Recovery Scrubber Meets Design Goals. Coal and Synfuels Technology. April
22. p. 5.
Antosh, N. 1988. Mergers, Foreign Buyouts Stir Up Formerly Dormant Cement Stocks. Houston
Chronicle. April 10. p.l.
ASTM. 1991. Standard Guide for Commercial Use of Lime Kiln Dusts and Portland Cement Kiln
Dusts. 1990 Annual Book of American Society for Testing and Materials Standards.
Volume 11.04. Method Number D5050-90. pp. 172-174.
Barrington, S.F. and A.F. MacKenzie. 1989. Enrichment of Swine Manures through Cement Kiln
Dust Incorporation. Biological Wastes. Volume 29, Number 1. pp. 1-10.
Beers, A., 1987. Hazardous Waste Incineration: The Cement Kiln Option, New York State
Legislative Commission on Toxic Substances and Hazardous Wastes, Albany, New York.
December 1987.
Betteker, James M., et al. 1986. Solidification I Stabilization of Contaminated Dredged Material.
Proceedings of Mid-Atlantic Industrial Waste Conference. June 29-July 1. Lancaster,
Pennsylvania.
Bhatty, M.S.Y. 1984. Use of Cement Kiln Dust in Blended Cements. World Cement. May.
Volume 15, Number 4. pp. 126-134.
-------
B-4
Bhatty, M.S.Y. 1985. Use of Cement Kiln Dust in Blended Cements: Alkali-Aggregate Reaction
Expansion. World Cement. December. Volume 16, number 10. pp. 386, 388-390, 392.
Billings, Clayton H. 1992. Screw Press/Kiln Dust Combination Doubles Drying Bed Production.
Public Works. May.
Boynton, R.S., 1980. Chemistry and Technology of Lime and Limestone, Second Edition, John
Wiley and Sons, New York, New York.
Bryansk Technical Institute. Date Unknown. Solution for Building Applications Containing
Portland Cement, Cement Kiln Dust, and Phosphogypsum.
Bureau of Mines. 1989. Minerals Yearbook, Vol 1, Metals and Minerals, Potash (James P. Searls
auth.). Department of the Interior, pp. 801-805.
Burnet, G. 1987. Alumina Recovery from Fly Ash by the Lime-Soda Sinter Process. Proceedings of
the International Symposium on Ash. February 2. Pretoria, South Africa.
Burnham, J.C. 1988. Cement Kiln Dust/Lime Treatment or Municipal Sludge Cake, Alternative
Methods For Microbial and Odor Control. Paper from Proceedings of National
Conference on Municipal Sewage Treatment Plant Sludge Management. June 27-29.
Palm Beach, Florida.
Burnham, J.C., et al. 1990. Cement Kiln Dust Stabilization of Municipal Wastewater Sludge.
Annual International Solid Waste Exposition "Vancouver 90." August. Canada.
Bush, R.S. and W.G. Nicholson. 1985. The Effect of Cement Kiln Dust on Tissue Accumulation
of Trace Minerals in Steers. Canadian Journal of Animal Science. June. Volume 65. pp.
429-435.
Cal. Admin Code tit. 22, § 66260.10 (also referred to as the California Code of Regulations).
Cal. Admin. Code tit. 22, § 66261.120.
Cal. Admin. Code tit. 22, § 66261.124.
Cal. Admin. Code tit. 22, § 66261.126(a).
Cal. Admin. Code tit. 22, § 66261.126(c).
Cal. Admin. Code tit. 22, § 66261.126(d).
Cal. Health and Safety Code § 25141.1.
Cal. Health and Safety Code §§ 39000-44384.
Cal. Health and Safety Code §§ 41502-41507.
Cal. Health and Safety Code §§ 41701, 41704.
-------
B-5
Cal. Health and Safety Code §§ 42400-42402.
Cal. Water Code §§ 13172,13226-13227.
Cal. Water Code §§ 13261.
Cal. Water Code §§ 13300-13306.
Cal. Water Code § 13340.
Cal. Water Code §§ 13350-13371.
Cal. Water Code §§ 13385,13387.
Cantu, T. 1990. Dust Removal at Lincoln Heights to Cost Alamo Cement $1.5 Million. San
Antonio Business Journal. November 9. Volume 4, p. 3.
Cement Plant Resources Group, draft Final Report on Cement Plant Excess Emissions (July 31,
1991).
Chadbourne, J. and E. Bouse, General Portland, 1985. Los Robles Cement Plant Cement Kiln
Dust Waste Classification Report. August, p. 5.
Chari, N.R. and D.K. Sahu. 1980. Studies on the Feasibility of Recovery of Potassium Salts from
Cement Kiln Dust. Fertilizer Technology. Volume 17. pp. 69-70.
Chemical Prices Weekly. 1992. Potassium Sulfate. Ending December 25. p. 28.
Cohen, S.M. 1992. Fluid Bed Dust Recovery. Paper presented at the Portland Cement
Association Seminar on Emerging Technologies for Kiln Dust Management. March 4.
Chicago, Illinois.
Constantino, D. 1989. High-Tech Cement. Pit & Quarry. March, pp. 31-32.
Dann, P.R., B.S. Dear, and R.B. Cunningham. 1989. Comparison of Sewage Ash, Crushed
Limestone, and Cement Kiln Dust as Ameliorants for Acid Soils. Australian Journal of
Experimental Agriculture. Volume 29. pp. 541-549.
Davis, Thomas A., et al. 1975. Disposal and Utilization of Waste Kiln Dust From Cement
Industry. Southern Research Institute. May. p. 17.
Department of Ecology, State of Washington, 1991. The Model Toxics Control Act Cleanup
Regulation, Chapter 173-340 WAC. As amended, February, 1991. p. 108.
Department of Ecology, State of Washington, 1992. Site Hazard Assessment, Markey Property,
Parcel 4, South 96th Street/lOth Avenue South, Seattle, Washington. September, 1992.
Department of Ecology, State of Washington, 1992. The Model Toxics Control Act Cleanup
Regulation, Chapter 173-340 WAC. As amended, February, 1992. p. 93.
-------
B-6
Department of Ecology, State of Washington, 1992. Washington Ranking Method, Scoring
Manual. As amended, April, 1992.
Department of Health and Environmental Sciences, State of Montana, 1990. State of Montana
ex. rel v. Ash Grove Cement West, Inc. Complaint and Application for Injunction, Cause
No. 8442, Montana 5th Judicial District Court, Jefferson County. December 11,1990.
Department of Health and Environmental Sciences, State of Montana, 1990. State of Montana
ex. rel v. Ash Grove Cement West, Inc., Consent Decree, Stipulation, and Order, Cause
No. 8442, Montana 5th Judicial District Court, Jefferson County. December 19, 1990.
Dodds, J. Date Unknown. Waste Tires as Auxiliary Fuel for Cement Kilns. U.S. Department of
Energy, Office of Industrial Programs.
Dodds, J. 1984. Scrap Tire Fuel for Cement Kilns, Exchange Meetings Summary. U.S.
Department of Energy. October 2.
Duda, W.H., 1976. Cement-Data-Book: International Process Engineering in the Cement Industry.
Bauverlag GMBH Wiesbaden and Berlin, pp. 403-417.
Ecology and Environment, Inc. 1990. Environmental Priorities Initiative, Preliminary Assessment,
Farmland Industries Site, Coffeyville, KS. Prepared for U.S. Environmental Protection
Agency, Hazardous Site Evaluation Division, October 15.
Eger, V.G. and O.N. Mandryka. 1984. The Possibility of Using Bentonite Clay for Purification of
Wastewaters From Apatite Processing. Journal of Applied Chemistry of the USSR.
Volume 57. pp. 2420-2422.
Emission Offset Interpretive Ruling, 40 CFR Part 51, Appendix S.
Energy Systems. 1991. Promotional Bulletin - N-Viro Soil.
Engineering Enterprises, Inc. 1991. Evaluation of the Potential for Migration of Hazardous Waste
Constituents from the Disposal Site to Water Supply Sources, Farmland Industries,
Coffeyville, Kansas.
Engineering-Science, 1987. Background Information Document for the Development of Regulations
to Control the Burning of Hazardous Wastes in Boilers and Industrial Furnaces, Volume II:
Industrial Furnaces, submitted to U.S. Environmental Protection Agency, Waste
Treatment Branch, Washington, D.C. January 1987.
Environmental Toxicology International, 1992. All Fired Up: Burning Hazardous Waste in
Cement Kilns, Seattle, Washington.
ESSROC Materials, Inc., 1992. Particle size distribution of "typical stack dust," February 14,
1992. Submitted on November 3, 1992 in response to RCRA Section 3007 request for
information by U.S. EPA, August 18, 1992.
-------
B-7
Felix, A., D.R. Rao, C.B. Chawan, and P.I. Ikem. 1980. Effect of Cement Kiln Dust on Nutrient
Digestibility in Sheep. Annual Research Report: School of Agriculture, Alabama A&M.
pp. 103-109.
Flachowsky, G., HJ. Lohnert, G. Stubedorff, E. Flachowsky, G. Staupendahl, and A. Hennig.
1982. The Use of Portland Cement Kiln Dust in the Feeding of Fattening Bulls. Archiv fur
Tierernahrung. Volume 32. pp. 93-98.
Galvano, G., A. Lanza, L. Chiofalo, and M. Mal'an. 1982. Cement Kiln Dust as a Mineral Source
in Feeding Ruminants. World Review of Animal Production. Volume 18, Number 4. pp.
63-71.
Garrett-Price, B.A. 1985. Potential for Energy Conservation in the Cement Industry. U.S.
Department of Energy. February.
GeoEngineers, Inc., 1989. Environmental Site Assessment, Cement Kiln Dust Landfill, Markey
Machinery Property, Seattle, Washington. Prepared for Helsell, Fetterman, Martin, Todd,
& Hokanson. August, 1989.
GeoEngineers, Inc., 1990. Environmental Sampling Report, Upper Fill, Proposed Parcel 4, South
96th Street & 10th Avenue South, Seattle, Washington. Prepared for Helsell, Fetterman,
Martin, Todd & Hokanson. June, 1990.
Godbey, T. 1991. Re-evaluating Dust Control in Cement Operations. Pit & Quarry. April, pp.
27-34.
Gossman, D. 1992. The Reuse of Petroleum and Petroleum Waste in Cement Kilns.
Environmental Progress. Volume 11, Number 1. February, pp. 1-6.
Harding Lawson Associates, 1984. Inspection of Ground-water Monitoring, Farmland Industries,
Coffeyville, Kansas, Draft Report.
Hartline, C. 1990. Dealing with Dust Control Pit & Quarry. December, pp. 22-25.
Haynes, B., and G. Kramer, 1982. Characterization of U.S. Cement Kiln Dust. Bureau of Mines
Information Circular (1C) 8885, U.S. Department of Interior. Bureau of Mines. Office
of Assistant Director. Minerals and Materials Research, Washington, D.C.
Hogue, D.E., PJ. Van Soest, J.R. Stouffer, G.H. Earl, W.H. Gutenmann, and D.J. Lisk. 1981.
Cement Kiln Dust as a Selenium Source in Sheep Rations. The Cornell Veterinarian.
January. Volume 71. pp. 69-75.
ICF Incorporated, 1990. Regulatory Impact Analysis for the Proposed Rulemaking on Corrective
Action for Solid Waste Management Units (Draft). Prepared for Economic Analysis Staff,
Office of Solid Waste, U.S. EPA. June 25.
ICF Incorporated, 1993. Technical Background Document: Cost and Economic Impacts of
Alternatives to Current CKD Disposal Practices (in progress).
-------
B-8
Iowa Department of Water, Air and Waste Management, 1984. Calmus Creek Water Quality
Study, May-August, 1984.
Iowa Department of Natural Resources, 1990. Record of Decision for Northwestern States
Portland Cement Company Site, Mason City, Iowa. June, 1990.
International Trade Association, 1993. U.S. Industrial Outlook, 1993. Washington, D.C., pp. 7-5
to 7-8.
Johnson, F. 1983. Recover Fugitive Cement Dust at Dockside. Pit & Quarry. September, p. 68.
Johnson, W. 1988. Cement. Cement Minerals Yearbook, pp. 201-228.
Johnson, W., 1992. Cement, Annual Report 1990, U.S. Bureau of Mines, Washington, D.C. April
1992.
Johnson, W and H. Sullivan. 1990. Cement in 1989. U.S. Department of Interior, Bureau of
Mines. July 25.
Johnson, W and H. Sullivan. 1991. Cement in February 1991. U.S. Department of Interior,
Bureau of Mines. April 12.
Karadzhyan, A.M., A.G. Chirkinyan, L.V. Efremova, and A.A. Evoyan. 1982. Effect of Cement
Kiln Dust on Growth and Development of Young Cattle. Trudy Erevanskogo
Zootekhnichesko-veterinarnogo Instituta. Number 53. pp. 10-14.
Kelley, W.D., D.C. Martens, R.B. Reneau, Jr., and T.W. Simpson, 1984. Agricultural Use of
Sewage Sludge: A Literature Review. Bulletin 143. Virginia Water Resources Research
Center, Virginia Polytechnic Institute and State University, Blacksburg, Virginia.
December, p. 38.
Keynote Address by Carol Browner, EPA Administrator, at the Clean Air Marketplace 1993.
September 9, 1993.
Keystone Cement Company. Date Unknown. Stablesorb: A Coproduct of Cement Manufacturing
With a Variety of Uses. Product Brochure.
Kirk-Othmer, 1979. Cement, Kirk-Othmer Encyclopedia of Chemical Technology, Volume 5, John
Wiley and Sons, New York, New York.
Kohlhaas, B., et al, 1983. Cement Engineer's Handbook. Bauverlag GMBH, Wiesbaden and
Berlin, p. 624.
Kosmatka, S. and Panarese, W., 1990. Design and Control of Concrete Mixtures. 13th Ed.
Portland Cement Association. Skokie, Illinois, pp. 15-16, 42-43.
Kovacik, T.L. 1987. Successful Recycling for Sludge and Solid Waste. BioCycle Southeast
Conference. November 4. Orlando, Florida.
-------
B-9'
Kovacik, T.L. 1988. Sludge, Kiln Dust Make Fertilizer. Water Engineering Management.
December.
Kriech, Anthony J., 1992. Evaluation of Hot Mix Asphalt for Leachability, Heritage Research
Group, October.
Kriech, Anthony J., 1992. Leachability of Asphalt and Concrete Pavements, Heritage Research
Group, March.
Kulich, J. and A. Ragas. 1973. Furnace Dusts from Cement Works as a Source of Available
Nutrients. Pol'nohospodarstvo. Volume 19, Number 2. pp. 113-121.
Lachica, E.D. 1990. Agency Rules Cement Imports Hurt U.S. Firms. Wall Street Journal.
August 14. p. 9.
Lebel, R. 1990. Cement Producers Say They Can Bum Old Tire in Kilns. Montreal Gazette.
May 19. p. C-3.
Lehigh Portland Cement Company, 1984. Letter from R. Gebhardt to H. Beiro, ADEM. May
30,1984.
Lehigh Portland Cement Company, 1987. Letter from L. Copple to S. Jenkins, Alabama
Department of Environmental Management. February 20,1987.
Lehigh Portland Cement Company, 1989. Site Investigation Protocol for the Lehigh Portland
Cement Company Plant, Mason City, Iowa. Prepared for Lehigh Portland by Layne
GeoSciences, Inc. October, 1989.
Lehigh Portland Cement Company, 1991. Remedial Investigation/Feasibility Study for the Lehigh
Portland Cement Company Plant, Mason City, Iowa. Prepared by Layne GeoSciences, Inc.
April, 1991.
Letter from Stanford Lau, Toxic Substances Control Division, to Ralph Mitchell, Lone Star
Industries, Inc., July 20, 1988.
Letter from Mindy Koch, Acting Chief, Waste Management Division, Department of Natural
Resources, to Myron Black, LaFarge Corporation, dated September 27,1991.
Libicki, J. 1984. Reclamation in Mountains, Foothills, and Plains: Doing it Right. 9th Annual
Meeting of the Canadian Land Reclamation Association. August 21. Canada.
Longenbach, J, et. al. 1991. EPA/CKD Meeting Agenda, Keystone Cement Company. Keystone
Cement Company. July 30.
Looker, Dan. 1990. Boone Couple Adds Fuel to Sustainable Ag Debate. Des Moines Register.
January 10. Volume 141, number 171. p. 1A+.
Lynch, M. 1991. Steel Cement Plant Set to Convert Slag. Australia Financial Review. March 20.
-------
B-10
Martin Marietta Corporation. Date Unknown. Poured, Moulded, or Pressed Concrete Blocks
Contain Aggregate, Cement, and Cement Kiln Dust.
May, A., J.W. Sweeney, and J.R. Cobble. 1983. Use of Florida Phosphogypsum in Synthetic
Construction Aggregate. Florida Institute of Phosphate Research Publication Number 01-
008-026.
McCord, A.T. 1977. Cement Kiln Dust Treatment - By Leaching with Hot Potassium Chloride
Solution to Recover Alkali Values. U.S. Patent No. 4031184.
McNamara, V. 1990. Southdown Buys Waste Processors. Houston Business Journal. June 11.
Volume 20, p. 10.
Mehta, P.K. 1986. Concrete: Structure, Properties, and Materials, Prentice Hall, Englewood Cliffs,
NJ.
Metry, A.A., et al. 1985. A Cost-Effective Approach For Stabilization and Closure of an Organic
Waste Superfund Site. 31 Annual Meeting of the Institute of Environmental Sciences.
April 29. Las Vegas, Nevada.
Mettauer, H. and A.P. Conesa. 1981. Agronomic Value of Residual Cement Dust. Comptes
Rendus des Seances de 1'Academie d'Agriculture de France. Volume 67, Number 9. pp.
772-781.
Mich. Admin. Code §§ 299.4102(h) (viii) and 299.4301(3).
Mich. Admin. Code §§ 336.1203 and 336.1208.
Mich. Admin. Code § 336.1301 (l)(a).
Mich. Admin. Code §§ 336.1371-72.
Mich. Comp. Laws § 299.407.
Mich. Comp. Laws § 299.408(3).
Mich. Comp. Laws §§ 323.1-323.13.
Mich. Comp. Laws § 336.13.
Mich. Comp. Laws §§ 401-436.
Michigan Department of Natural Resources, 1993. Letter from John W. Vick, Environmental
Response Division to Rebecca Beasly, Assistant General Council, National Gypsum
Company. May, 1993.
Michigan Department of Natural Resources, 1993. Interoffice Communication from Jim Sygo,
Chief, Waste Management Division, to Russell Harding, Deputy Director. April, 1993.
-------
B-ll
Michigan Department of Natural Resources, 1993. Photos of National Gypsum CKD pile taken
during MDNR site visits on March 30,1993 and April 22,1993. Videotape of National
Gypsum CKD pile taken by John W. Vick on April 30,1993.
Miller, C.T., D.G. Bensch, and D.C. Colony. 1980. Use of Cement Kiln Dust and Fly Ash in
Pozzolanic Concrete Base Courses. Transportation Research Record, pp. 36-41.
Morgan, David S., et al, 1984. Oil Sludge Solidification Using Cement Kiln Dust. Journal of
Environmental Engineering. October.
Morrison, G.L. 1990. CKD and Flue Gas Scrubbing: The Demonstration at Dragon Products
Company. July. p. 7.
Morrison, G.L. 1990. Exhaust Gas Scrubbing and Waste Elimination, An Application of the
Recovery Scrubber to a Cement Kiln. Paper presented at the 1990 SO2 Control
Symposium. May 8. New Orleans, Louisiana.
Morrison, G.L. 1991. Flue Gas Scrubbing and Waste Elimination - An Application of the Recovery
Scrubber. Paper presented at The World Coal Institute Conference and Exhibition on
Coal in the Environment. April 3. London, United Kingdom.
Morrison, G.L. 1992. CKD Management Using a Recovery Scrubber: Operation and Economics.
Paper presented at the Portland Cement Association Seminar on Emerging Technologies
for Kiln Dust Management. March 4. Chicago, Illinois.
Morrison, G.L., 1993. Passamaquoddy Technology Recovery Scrubber Operations Update and
Forecast. July. pg. 7.
Mournighan, R.E. and M. Branscome. 1987. Hazardous Waste Combustion in Industrial
Processes: Cement and Lime Kilns. U.S. Environmental Protection Agency, Office of
Research and Development. November.
Muelberg, P.E., et al., 1977. Industrial Process Profiles for Environmental Use: Chapter 21, The
Cement Industry, U.S. Environmental Protection Agency, Office of Research and
Development, Industrial Environmental Research Laboratory, Cincinnati, Ohio. February
1977.
Naylor, L.M., J.C. Dagneau, and IJ. Kugelman. 1985. Cement Kiln Dust - A Resource Too
Valuable to Waste? Proceedings of the Seventeenth Mid-Atlantic Industrial Waste
Conference on Industrial and Hazardous Wastes. June 23. pp. 353-366.
Naylor, L.M., E.A. Seme, and TJ. Gallagher, 1986. Using Industrial Wastes in Agriculture.
BioCycle. February, pp. 28-30.
Nebgen, J.W., et al. 1976. Use of Waste Sulfate For Remedial Treatment of Soils, Volume I,
Discussion of Results. U.S. Department of Transportation, Federal Highway
Administration. August.
Newton, G.L. and O.M. Hale. 1979. Cement Kiln Dust and Carboxylin as Feed Additives for
Swine. Journal of Animal Science. October. Volume 49. pp. 908-914.
-------
B-12
Nicholson Realty. Date Unknown. Cementitious-hardening Paving Base Composition Using Waste
Materials.
Noller, C.H., J.L. White, and W.E. Wheeler. 1980. Characterization of Cement Kiln Dusts (Fed
to Animals) and Animal Response. Journal of Dairy Science. November. Volume 63.
pp. 1947-1952.
Norman, J.R. 1987. Battling for Survival in the Concrete Jungle. BusinessWeek. September 14.
pp. 66-74.
Northwestern States Portland Cement Company, 1985. Hydrogeologic Investigation - West Quarry
Site, Northwestern States Portland Cement Company, Mason City, Iowa. Prepared by IT
Corporation. July, 1985.
Northwestern States Portland Cement Company, 1989. Remedial Investigation/Feasibility Study on
the West Quarry, Mason City, Iowa. Prepared by Layne Geosciences, Inc. Project No.
61.1099.
N-Viro Achieves Record Year. PR Newswire. February 19, 1992.
Ohio Environmental Protection Agency, 1986. Preliminary Assessment, Southwestern Portland
Cement, Fairbom, Ohio (Landfill #6), Part 3 - Description of Hazardous Conditions and
Incidents. July 18, 1986.
Ohio Environmental Protection Agency, 1991. Memo from Louise T. Snyder, DWQPA, SDWO
on the Southwestern Portland Cement facility, Landfill #1. September 9, 1991.
Ohio Environmental Protection Agency, 1992. Director's Final Findings and Order in the Matter
of Southdown, Inc., 506 East Xenia Drive, Fairbom, Ohio.
Ohio Environmental Protection Agency, 1993. Fact Sheet: Southwestern Portland Cement
Company -- Landfill No. 6, Fairbom, Ohio. March, 1993.
Pennsylvania Department of Environmental Resources 1992 Water Quality Assessment.
Peray, K.E., 1986. The Rotary Cement Kiln, Second Edition, Chemical Publishing Co., Inc., New
York, New York.
Personal communication with Barb Rosenbaum, Michigan Department of Natural Resources, Air
Quality Division, September 10,1993.
Personal communication with Ben Haynes, Engineering, Technology, and Research Division,
Bureau of Mines, U.S. Department of Interior, November 1992.
Personal communication with Brian Graf, ESSROC Materials, Inc. (Logansport, IN), March 17,
1992.
Personal communication with Bob Zabick, Wayne County Health Department, Air Pollution
Control Division, Enforcement, October 13, 1992.
-------
B-13
Personal communication with Brad Venman, Department of Natural Resources, Waste
Management Division, October 9,1992.
Personal communication "with C. McRoy, Alabama Department of Environmental Management.
October, 1992.
Personal communication with Charlie Klotz, Safety Training, and Environmental Manager,
Lehigh Portland Cement Company, Leeds, Alabama facility. 1993
Personal communication with Chris Marxen, California Environmental Protection Agency,
Department of Toxic Substances, Waste Evaluation Unit, September 10,1993.
Personal communication with Clifton Calderwood, Manager Compliance Division, San Joaquin
Valley Air Basin, March 31,1993.
Personal communication with Craig Southworth, Ash Grove Cement Company, May 20,1993.
Personal communication with Dan McGee, Michigan Department of Natural Resources, January
22,1993.
Personal communication with Doug Glasford and Bill Fischer, Keystone Cement, November 24,
1992.
Personal communication with Dr. Donna Waltz, Center for Veterinary Medicine, Animal Feeds
Division, FDA, November 1992.
Personal communication with Earnest Hire, Texas Natural Resources Conservation Commission,
Industrial and Hazardous Waste Enforcement Section, September 10,1993.
Personal communication with Ed Kendig, Compliance Director, Monterey Bay Air Pollution
Control District, July 28,1992.
Personal communication with Everett Spenser, EPA Region VI, Enforcement Branch, September
9,1993.
Personal communication with facility personnel during EPA-CKD sampling visit, Texas
Industries, Inc., March 1992.
Personal communication with Fred Fontus, California Department of Environmental Protection,
Department of Toxic Substances, September, 10,1993.
Personal communication with Garrett Morrison, Passamaquoddy Technology, November 24,1992
and July 2,1993.
Personal communication with Gary Linns, Pennsylvania Department of Environmental
Resources, Bureau of Air Quality, Air Quality Control Division, October 9,1992.
Personal communication with Greg Chee, Air Quality Engineer, North Central Coast Air Basin,
March 3,1993.
-------
B-14
Personal communication with Hans Steuch, Director of Engineering, Ash Grove West, December
9,1992.
Personal communication with Harry Hackett, Holnam, Dundee, Michigan, January 28,1993.
Personal Communication with Henry Voldbaek, Ash Grove, Inkom, Idaho, December 9, 1992.
Personal communication with Hillary Inyang, Professor, University of Wisconsin, November 16,
1992.
Personal communication with J. Patrick Nicholson, N-Viro Soil, December 7, 1992.
Personal communication with JoAnn Merrick, Acting Chief, Compliance and Enforcement
Section, Waste Management Division, Michigan Department of Natural Resources. July,
1993.
Personal communication with John McDowell, DOE, December 7,1992.
Personal communication with John W. Vick, Environmental Quality Analyst, Environmental
Response Division, Michigan Department of Natural Resources, 1993.
Personal communication with M. Lehar, Ohio Environmental Protection Agency. January, 1992.
Personal communication with Marc Saffley, Soil Conservation Service (SCS), November 18,1992.
Personal communication with Maune Currie, Michigan Department of Natural Resources,
Regional Office, October 10,1992.
Personal communication with Michael Kussow, Air Pollution Control Officer, North Sacramento
Valley Air Quality Monitoring District, August 11, 1993.
Personal communication with Mike Rafalowski, DOT-FHA, November 16, 1992.
Personal communication with Pete Ostlund, Chief of Industrial Permits Section, Surface Water
Quality Division, Michigan Department of Natural Resources, September 15,1992.
Personal communication with plant personnel, Ash Grove Cement West, Montana City, Montana
facility. July, 1993.
Personal communication with Reno Vacheski, Pennsylvania Department of Environmental
Resources, Region II, Waste Management Bureau, September 10, 1993.
Personal communication with Richard Lee, Natural Resources Conservation Commission, Air
Quality Program, Compliance Division, August 25, 1993.
Personal communication with Rick Ruddell, Watershed Management Division, and Earnest
Heyer, Field Operations Division, May 16, 1993.
Personal communication with Robert Bastion, Office of Water-Wastewater Enforcement, EPA,
November 1992.
-------
B-15
Personal communication with S. Jenkins, Alabama Department of Environmental Management.
January, 1992.
Personal communication with Sam Barrett, Field Investigator, Texas Water Commission, June
1992, November 1992, and July 1993.
Personal communication with Sam Sloan, Pennsylvania Department of Environmental Resources,
Bureau of Waste Management, Division of Municipal and Residual Waste, September 10,
1993.
Personal communication with Sidney Cohen, Fuller Company, November 13,1992 and July 20,
1993.
Personal communication with Thomas Schneider, Site Coordinator, Ohio Environmental
Protection Agency. April, 1993.
Personal communication with Tom Krinke, Air Quality Specialist, Compliance Section, San
Bernadino Air Quality Control District, July 28,1992.
Personal communication with Tom Rohrer, Chief of Enforcement Unit, Surface Water Quality
Division, May 24, 1993.
Portland Cement, Kackman, A.H. and Brown, R.W., Section Editors; SME Mineral Processing
Handbook, Society of Mining Engineers of the American Institute of Mining,
Metallurgical, and Petroleum Engineers, Inc., New York, NY, 1985, Vol. 2, Section 26,
pp. 26-1 to 26-52.
Pond, W.G., D.A. Hill, C.L. Ferrell, and L. Krook. 1982. Bone Lesions in Growing Swine Fed 3
Percent Cement Kiln Dust as a Source of Calcium. Journal of Animal Science. January.
Volume 54. pp. 82-88.
Portland Cement Association, 1984. U.S. Cement Industry - an Economic Report, Skokie, Illinois.
January 1984.
Portland Cement Association, 1991a. PCA Cement Kiln Dust Survey: Response from Lehigh
Portland Cement Company, Leeds, Alabama.
Portland Cement Association, 1991b. PCA Cement Kiln Dust Survey: Response from the
Southwestern Portland Cement Company, Fairborn, Ohio.
Portland Cement Association, 1991c. PCA Cement Kiln Dust Survey: Response from Holnam,
Inc., Artesia, Mississippi.
Portland Cement Association, 1991d. PCA Cement Kiln Dust Survey: Response from Holnam,
Inc., Mason City, Iowa.
Portland Cement Association, 1992a. U.S. Cement Industry Fact Sheet, 10th Edition, Market and
Economic Research, Skokie, Illinois. April 1992.
-------
B-16
Portland Cement Association, 1992b. U.S. and Canadian Portland Cement Industry: Plant
Information Summary 1991, Market and Economic Research, Skokie, Illinois. August,
1992.
Portland Cement Association, 1992c. An Analysis of Selected Trace Metals in Cement and Kiln
Dust (Draft). PCA Report SP109T, Skokie, IL.
Portland Cement Association, 1992d. An Analysis of Selected Trace Metals in Cement and Kiln
Dust. PCA Report SP109T, Skokie, IL, 56 pages.
Portland Cement Association, 1992e. PCA Cement Kiln Dust Survey: Response from Ash Grove
West, Inc., Montana City, Montana.
Ramakrishnan, V. 1986. Evaluation of Kiln Dust in Concrete. American Concrete Institute, pp.
821-839.
Ramakrishnan, V. and P. Balaguru. 1987. Durability of Concrete Containing Cement Kiln Dust.
American Concrete Institute, pp. 305-321.
Ravindrajah, R.S. 1982. Usage of Cement Kiln Dust in Concrete. International Journal of
Cement Composites and Lightweight Concrete. May. Volume 4, Number 2. pp. 95-102.
Razzell, W.E. 1990. Chemical Fixation, Solidification of Hazardous Waste. Waste Management
Resources. April.
Reardon, Patrick W., Jr., 1991. "Low alkali cement requirements for Northeast U.S. Markets,"
Cement Technology. November, pp. 61-63.
Reed, J.E., et. al. 1983. Co-firing of Refuse-Derived Fuel (RDF) with Natural Gas in a Cement
Kiln, Summary of Test Results. Prepared for the U.S. Department of Energy, Office of
Industrial Programs. March.
Regents of the University of California. 1986. High Iron Hydraulic Cement Manufactured from
Red Mud and Cement Kiln Dust. PCT Patent Application Number WO-86-05773.
October 9.
Ringelmann's Scale for Grading the Density of Smoke as published in United States Bureau of
Mines Information Circular 8333.
Roberts, A. 1990. French Firm to Challenge Brinish Cement Imports. Virginian-Pilot. June 28.
pp. C8-C9.
Roginski, E.E. and W.E. Wheeler. 1978. A Growth Effect of Georgia Cement Kiln Dust in Rats.
Federation Proceedings. Volume 37. p. 404.
Roginski, E.E. and W.E. Wheeler. 1979. The Response of Monogastric Species to Cement Kiln
Dust in the Diet. Federation Proceedings. Volume 38. p. 614.
Rumsey, G.L., W.H. Gutenmann, and D.J. Lisk. 1981. Cement Kiln Dust as an Additive in the
Diets of Rainbow Trout. Progressive Fish-Culturist. Volume 43. pp. 88-90.
-------
B-17
Saito, I, et. al. 1987. Effective Use of Waste Tyres by Gasification in Cement Plant. World
Cement. September, pp. 264-269.
Sanborn, S., 1992. Cement & Aggregates, Value Line Publishing. July 24,1992.
Sansbury, T. 1988. Cement Makers Provide Market for Hazardous Waste, as Fuel. Journal of
Commerce. December 8. p. 4b.
Sapp, I.E., 1981. Energy and Materials Flows in the Cement Industry, prepared for U.S.
Department of Energy, June 1981.
Schiavi, W. 1990. Solid Improvements on an Old Idea. Pit & Quarry. September, pp. 30-37.
Seebacher, N. 1991. Cementing a Spot on the Fortune 500. The Detroit News. April 14. pp.
Fl-4.
Sell, Nancy J. and Fritz A. Fischbach. 1978. Pelletizing Waste Cement Kiln Dust for More Efficient
Recycling. Industrial and Engineering Chemistry, Process Design and Development.
Volume 17, Number 4. October, pp. 468-473.
Sherwood, P.T., L.W. Tubey, and P.O. Roe. 1977. The Use of Waste and Low-Grade Materials in
Road Construction. Transport and Road Research Laboratory. Crowthorn, England.
Smith, J.D., 1990. Cement Kilns 1990, El Digest, Environmental Information, Ltd. June 1990.
Smith, J.D., 1991. Cement Kilns 1991, El Digest, Environmental Information, Ltd. August 1991.
pp. 20-32.
Southdown, Inc., 1991. Site Assessment of Southwestern Portland Cement Properties Near Fairborn,
Ohio: Phase 1, Preliminary Investigations, Pan 1. Prepared by Panterra Corporation.
March 11,1991.
Southdown, Inc., 1992. Subarea 1 Site Evaluation, Southwestern Portland Cement Company,
Fairbom, Ohio. Prepared by Ground Water Associates, Inc, April, 1992.
Standard Concrete Material, Inc. 1983. Cement Composition as Substitute for Portland Cement
Containing Blast Furnace Slag, Cement Kiln Dust, and/or Calcium Carbonate, and
Optionally Gypsum. PCT Patent Application Number WO--83-01443. April 28.
Stanley, D. 1988. Florida Mining Proposes Waste-Burning Cement Plan. Tampa Tribune.
September 15, p. Dl.
State of Mississippi, 1992. Permit for stormwater runoff for Holnam, Inc., Artesia, Mississippi;
NPDES Permit No. MSR320017.
Steuch, Hans E. 1992. Review of Dust Return Systems. Paper presented at the Portland Cement
Association Seminar on Emerging Technologies for Kiln Dust Management. March 4.
Chicago, Illinois, p. 2.
-------
B-18
Supplementing Ruminant Feeds with Cement Kiln Dust Improves Livestock Performance, According
to Agriculture Canada. Feedstuffs. June 2, 1980, p. 38.
Taylor, H.F.W., 1990. Cement Chemistry, Academic Press Inc., San Diego, California.
'Test Methods for the Evaluation of Solid Waste, Physical/Chemical Analysis."
Tex. Admin. Code tit. 31, § 101.1.
Tex. Admin. Code tit. 31, § 101.6.
Tex. Admin. Code tit. 31, § 111.111.
Tex. Admin. Code tit. 31, § 111.155.
Tex. Admin. Code tit. 31, § 307.7.
Tex. Admin. Code tit. 31, § 335.1.
Tex. Admin. Code tit. 31, § 335.2(d).
Tex. Health and Safety Code § 1.05.
Tex. Health and Safety Code § 361.
Texas Industries, Inc., 1992. Part B Permit Application to EPA and the Texas Water Commission.
Prepared by Entellect Environmental Services for Texas Industries, Inc. p. 1-9.
Texas Water Commission, 1990. RCRA Facility Assessment Facility Checklist for Texas
Industries, Incorporated, Peter F. Lodde, reviewer.
Texas Water Commission, 1991. Letter from Allen Hayes, Environmental Quality Specialist, to
files regarding a review of laboratory analyses of samples collected on October 21 and 22,
1991, from North Texas Cement Co., Texas Industries, Inc., and Box Crow Cement Co.
November 18,1991.
Texas Water Commission, 1992. Interoffice memorandum from Sam Barrett, Field Investigator
to files regarding record review of analytical results of samples collected from Texas
Industries, Inc. during inspections on January 27, 1992, March 10, 1992, and April 10,
1992. October 21, 1992.
Texas Water Commission, 1992. Interoffice communication from Sam Barrett, Field Investigator,
to files regarding compliance inspection at Texas Industries, Inc., Midlothian, Texas.
February, 1992.
Texas Water Commission, 1992. Letter from Mary B. Adrian, Section Leader, Enforcement
Section, TWC, to E.L. Faciane, Staff Vice-President, Environmental Affairs, Texas
Industries, p.3. April 15, 1992.
-------
B-19
Thorsen, J.W., et al., 1983. In Situ Stabilization and Closure of an Oily Sludge Lagoon. 3rd Ohio
Environmental Conference. March. Columbus, Ohio.
Tidona, R.J., et. al. 1983. Project Summary: Evaluation of Combustion Variable Effects on NOx
Emissions from Mineral Kilns. U.S. Environmental Protection Agency. November.
Todres, H., A. Mishulovich, and J. Ahmed, 1992. Cement Kiln Dust Management: Permeability.
Research and Development Bulletin RD103T, Portland Cement Association, Skokie,
Illinois, p.7.
U.S. Bankruptcy Court, Southern District of New York, 1991. Proof of Claim of the State of Utah
Department of Environmental Quality, Case No. 90-B-21277, in reference to New York
Trap Rock Corporation, Lone Star Industries, Inc. et.aL, Debtors.
U.S. Department of Agriculture. 1979. Letter from William E. Wheeler, Research Animal
Scientist, Nutrition, to John P. Lehman, Hazardous Waste Management Division, Office
of Solid Waste, U.S. EPA. February 23.
U.S. Department of Commerce, Date Unknown. Construction Materials Database, Building
Materials and Construction Division, Office of Forest Products and Domestic
Construction, International Trade Administration, Washington, D.C.
U.S. Department of Commerce, 1991. Cement Industry,'U.S. Industrial Outlook 1991,
Washington, D.C.
U.S. Department of Commerce, International Trade Administration. 1987. A Competitive
Assessment of the U.S. Cement Industry. July 1.
U.S. Department of Energy. 1983. Use of Solid Waste Material as a Raw Material in a Cement
Process, Final Report. December 31.
U.S. Department of Energy. 1989. Health Assessment for Portland Cement Company, Waste
Cement Kiln Dust Sites 2 & 3, Salt Lake City, Utah. February 28.
U.S. Department of Energy, Energy Efficiency Office. 1988. Energy Efficient Demonstration
Scheme, Refuse Burning in Cement Kilns, Demonstration at Blue Circle Westbury Works.
February.
U.S. Department of Health and Human Services, 1991. Health Assessment for Northwestern States
Portland Cement Company, Mason City, Cerro Gordo County, Iowa. December, 1991.
U.S. Department of the Interior, Bureau of Mines. 1985. Cement, A Chapter from Minerals
Facts and Problems 1985 Edition.
U.S. Department of the Interior, Bureau of Mines. 1987. Cement.
U.S. Department of Interior, Bureau of Mines. 1990. Minerals Yearbook.
U.S. Department of Transportation and U.S. Department of Energy. 1982. Kiln Dust-Fly Ash
Systems for Highway Bases and Subbases. October 1.
-------
B-20
U.S. Environmental Protection Agency, 1988. Final Report Site Investigation: Lehigh Portland
Cement Company, Mason City, Iowa. Prepared by Ecology and Environment Field
Investigation Team for Region VII. March, 1988.
U.S. Environmental Protection Agency, 1989. Exposure Factors Handbook. Office of Health
and Environmental Assessment. EPA/600/8-89/043.
U.S. Environmental Protection Agency, 1989. Risk Assessment Guidance for Superfund Volume
I: Human Health Evaluation Manual (Part A). Office of Emergency and Remedial
Response. EPA/540/1-89/002.
U.S. Environmental Protection Agency, 1990. Commercial Combustion Capacity for Hazardous
Waste Sludges and Solids, Office of Solid Waste, Washington, D.C. August 1990.
U.S. Environmental Protection Agency, 1990. Report to Congress on Special Wastes from Mineral
Processing, Volume II, Methods and Analyses, Office of Solid Waste, July.
U.S. Environmental Protection Agency, 1991. "A PC Software Application of the
Uptake/Biokinetic Model, Version 0.5," Office of Health and Environmental Assessment
(ECAO-CIN-2178A).
U.S. Environmental Protetion Agency, 1991. Human Health Evaluation Manual. Supplemental
Guidance: Standard Default Exposure Factors. Office of Emergency and Remedial
Response. OSWER Directive: 9285.6-03.
U.S. Environmental Protection Agency, 1991. Record of Decision: Lehigh Portland Cement,
Mason City, Iowa. Prepared by the Office of Emergency and Remedial Response. June,
1991.
U.S. Environmental Protection Agency, 1992. Guidance on Risk Characterization for Risk
Managers and Risk Assessors.
U.S. Environmental Protection Agency, 1992. Sampling Trip Reports.
U.S. Environmental Protection Agency, 1993. Unpublished field notes collected during visit to
Holnam, Inc., Artesia, Mississippi, May 25,1993.
U.S. Environmental Protection Agency, 1993. Water Systems Byproducts Treatment and Disposal
Cost Document (Final Draft). April.
U.S. Environmental Protection Agency, 1989. Superfund Record of Decision (EPA Region 7):
John's Sludge Pond, Wichita, KS, (First Remedial Action). Office of Emergency and
Remedial Response. Washington, DC. September 22.
U.S. Environmental Protection Agency, 1992. MMSOILS: Multimedia Contaminant Fate.
Transport, and Exposure Model. Documentation and User's Manual. Office of Research
and Development. September 1992. (updated in April 1993).
-------
B-21
U.S. Environmental Protection Agency, 1990. Summary of Industrial Processes, Products, Wastes
Generated and Regulation of the Cement Kiln Industry. Office of Solid Waste, December
14.
U.S. Environmental Protection Agency, Region VI, 1992. Complaint, Compliance Order, and
Notice of Opportunity for Hearing in the Matter of Texas Industries, Inc., Midlothian, Texas.
Dockett No. RCRA VI-203-H, Hazardous Waste Management Division. September,
1992.
U.S. Environmental Protection Agency, Region VIII, and Utah Department of Health and
Environmental Quality, 1990. Declaration for the Record of Decision, Portland Cement
Company (Kiln Dust #2 & #3), Operable Unit 1, Salt Lake City, Utah. July, 1990.
U.S. International Trade Commission. 1990. Gray Portland Cement and Clinker from Mexico.
August.
U.S. International Trade Commission. 1991. Gray Portland Cement and Clinker from Japan.
April.
University of Iowa, 1984. Calmus Creek Water Quality Study. Report 85-1.
Valley Forge Laboratories, Inc. 1982. Kiln Dust-Fly Ash Systems for Highway Bases and
Subbases. U.S. Department of Transportation and U.S. Department of Energy.
September.
Veltmann, J.R. and L.S. Jensen. 1979. Effect of Georgia Cement Kiln Dust on Broiler Chick
Performance. Poultry Science. Volume 58. p. 1027.
Ward, G.M., C.A. Olds, D.D. Caveny, and G.A. Greathouse. 1979. Cement Kiln Dust in
Finishing Lamb Diets. Journal of Animal Science. September. Volume 49. p. 637.
Weitzenkamp. 1988. Utilization of Waste Heat from a Cement Production Rotary Kiln for Heating
and Domestic Water Supply of Buildings. Interatom. May 26.
Wheeler, W.E. 1978. Cement Kiln Dust: A Potential Feed Ingredient for Livestock. Cereal Foods
World. Volume 23. pp. 296-297, 299, 312.
Wheeler, W.E. 1981. Variability in Response by Beef Steers to Cement Kiln Dust in High
Concentrate Diets. Journal of Animal Science. March. Volume 52. pp. 618-627.
Wheeler, W.E. and R.R. Oltjen. 1979. Cement Kiln Dust in Complete Diets for Finishing Steers
and Growing Lambs. Journal of Animal Science. March. Volume 48. pp. 658-665.
Willey, C.R. and C.S. Weinberger. 1985. Demonstration Test of Refuse Derived Fuel as a
Supplement Fuel in Cement Kilns. Ohio Environmental Protection Agency. April.
Wilson, R.D. and W.E. Anable. 1986. Removal of Alkalies From Portland Cement Kiln Dust.
U.S. Department of Interior, Bureau of Mines Report of Investigations Number 9032. p.
2.
-------
B-22
Z.A. and M.A. Rahman. 1990. Potential of Cement Kiln Dust for the Stabilization of Dune Sand
in Highway Construction. Building and Environment. Volume 25, Number 4. pp. 285-
289.
Zarlinski, S.J. and J.C. Evans, 1990. Durability Testing of a Stabilized Petroleum Sludge. Paper
from Hazardous and Industrial Wastes, Proceedings of 22nd Mid-Atlantic Industrial
Waste Conference. July 24-27. Pennsylvania.
Zier, Robert E. and Elizabeth Wood. 1991. Sludge: Sludge Solutions. Waste Information
Digests. May.
Zinn, R.A., R.A. Lovell, D.R. Gill, F.N. Owens, and K.B. Poling. 1979. Influence of Cement
Kiln Dust on Animal Performance and Nutrient Availability. Journal of Animal Science.
Volume 49. p. 422.
17 CCR § 70200.
25 Pa. Code § 16.
25 Pa. Code § 93.6.
25 Pa. Code § 123.1.
25 Pa. Code § 123.13.
25 Pa. Code § 123.41.
25 Pa. Code §§ 127.61-127.73.
25 Pa. Code § 287.1.
25 Pa. Code § 287.101.
25 Pa. Code §§ 287.121-287.134.
25 Pa. Code § 287.411.
25 Pa. Code § 287.421.
25 Pa. Code § 287.611.
33 U.S.C. §§ 1251-1387.
35 Penn. Stat. Ann. § 6018.102 and 25 Pa. Code § 287.
35 Penn. Stat. §§ 6018.602-6018.606.
35 Pa. Stat. § 4004.
35 Pa. Stat. § 691.
-------
B-23
40 CFR § 60.60.
40 CFR § 122.26.
40 CFR § 257.2.
40 CFR § 260.10.
40 CFH § "66.31(c).
\
40 CFR J /1 50.
40 CFR Part 257.
42 U.S.C. § 300h-7(e).
42 U.S.C. § 6924(q)(2)(C).
42 U.S.C. § 6944.
42 U.S.C. § 6944(a).
42 U.S.C. § 6973.
42 U.S.C. § 7410.
42 U.S.C. § 7411.
42 U.S.C. § 9604.
42 U.S.C. §§ 300f-300j.
42 U.S.C. §§ 6901 to 6992K.
42 U.S.C. §§ 6942 - 6949a.
42 U.S.C. §§ 7401-7671q.
56 Fed. Reg. 7134.
57 Fed. Reg. 41236.
------- |