U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
PB80-110976
Fugitive Emissions from
Iron Foundries
Midwest Research Inst, Kansas City, MO
Prepared for
Industrial Environmental Research Lab, Research Triangle Park,
Aug 79
-------
&EPA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
EPA-600/7-79-195
August 1979
Fugitive Emissions
from Iron Foundrie
nteragency
Energy/Environment
R&D Program Report
-------
NOTICE
THIS DOCUMENT HAS BEEN REPRODUCED'
FROM THE BEST COPY FURNISHED US BY
THE SPONSORING AGENCY. ALTHOUGH IT
IS RECOGNIZED THAT CERTAIN PORTIONS
ARE ILLEGIBLE, IT IS BEING RELEASED
IN THE INTEREST OF MAKING AVAILABLE
AS. MUCH. INFORMATION AS .POSSIBLE.
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research ana Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide-range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service. Springfield, Virginia 22161.
-------
TECHNICAL REPORT DATA
(Please read Inunctions on the reverse tit fore comrtletingi
1. REPORT NO.
EPA-600/7-79-1S5
2.
4. TITLE AND SUBTITLE
Fugitive Emissions from Iron Foundries
6. PERFORMING ORGANIZATION CODE
3. pcriPiFNT-s ACCESSION?NO.
r& 16-I/O
6. REPO'RT DATE
August 1979
7. AUTHOR(S)
Dennis Wallace and Chatten Cowherd Jr.
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
10. PROGRAM ELEMENT NO.
1AB015
11. CONTRACT/GRANT NO.
68-02-2120
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PI
Final; 6/75 - 2/78
,0 PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES JERL-RTP project officer is Robert V. Hendriks, MD-62, 919/
541-2733.
is. ABSTRACT
report describes the assessment of fugitive emissions of air pollutants
discharged from process operations in iron foundries, and the need for the develop-
ment of control technology for the most critical sources. Data indicates that the most
significant fugitive emissions control problem in foundries is the pouring of hot metal
into sand molds and the subsequent cooling of castings of these molds. Other signifi-
cant fugitive emissions sources which have control problems are the electric arc
furnace, preparation of molds and cores using organic binders, and casting shakeout.
These conclusions are tempered by the fact that, for most fugitive emission sources
in iron foundries, data is insufficient to determine accurate emission factors.
Research and development programs are recommended: to better quantify the most
significant sources; to evaluate currently available control technology for electric arc
furnaces and shakeout; and to develop new control technology for pouring and cooling
and for core and mold preparation using organic binders.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Pollution Emission
Iron and Steel Industry
Processing
Leakage
Foundries
Electric Arc Furnaces
Pollution Control
Stationary Sources
Fugitive Emissions
13B
11F
13H
14B
131
13A
8. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
2'
PAGES
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------
EPA-600/7-79-195
August 1979
Fugitive Emissions from Iron Foundries
by
Dennis Wallace and Chatten Cowherd Jr.
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
Contract No. 68-02-2120
Program Element No.1AB015
EPA Project Officer: Robert V. Hendriks
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, IMC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
PREFACE
This report was prepared for the Environmental Protection Agency to pre-
sent the results of work performed under Contract No. 68-02-2120. Mr. Robert V.
Hendriks served as EPA Project Manager.
The program was conducted in the Environmental and Materials Sciences Di-
vision of Midwest Research Institute. Dr. Chatten Cowherd, Head, Air Quality
Assessment Section, served as Program Manager. Mr. Dennis Wallace was the prin-
cipal investigator on the iron foundry portion of the study, which is the sub-
ject of this report.
iii
-------
CONTENTS
Preface ill
Figures vi
Tables vii
Summary ............................. 1
Conclusions ........ ........ 2
1.0 Introduction 3
2.0 Fugitive Emission Source Identification .. . 7
2.1 General process description 7
2.2 Fugitive emissions sources . 9
2.3 Materials flow 29
3.0 Fugitive Emissions Quantification . . 32
,3.1. Emission factor quantification methods-. . , . . . 32
3.2 Emission factors for the foundry industry .... 37
3.3 Inventory of iron foundry emissions 47
4.0 Fugitive Emissions Control ................. ,^9
4.1 Raw materials input .. 53
4.2 Melting and casting 56
4.3 Product finishing ....... 73
4.4 Mold and core preparation ............ 79
4.5 Waste handling 84
5.0 Research and Development Recommendations ..... 87
5.1 Determination of critical sources 87
5.2 Analysis of control availability 91
5.3 Current research 93
5.4 Suggested research 95
6.0 References 98
7.0 Glossary 102
8.0 English to Metric Unit Conversion Table 104
Preceding page blank
-------
FIGURES
Number Page
2-1 Composite flow diagram for the gray iron foundry industry . 10
2-2 Raw material storage and handling 11
2-3 Mold and core making 12
2-4 Melting and casting 13
2-5 Cleaning and finishing 14
2-6 Sand handling 15
2-7 Methods of iron inoculation 23
2-8 Iron foundry industry weighted material flows 30
3-1 Predictive emission factor equation for storage pile for-
mations by means of translating conveyor stacker .... 38
3-2 Predictive emission factor equation for transfer of aggre-
gate from front-end loader to truck 39
3-3 . Predictive emission factor equation .for wind erosion from
storage piles 40
'3-4. Quality assurance. (QA) -rating scheme for emission factors . 41
3-5 Line drawing of Canton Malleable's sand system showing
plowoff points and resultant sand temperatures 44
4-1 Hooding system for a vibrating screen 55
4-2 Hooding system for revolving screen 57
4-3 Fixed hood for cupola tapping 59
4-4 Movable hood for cupola tapping 60
4-5 Close capture hooding system for electric arc furnaces . . 62
4-6 Close capture hooding system for electric induction fur-
nace 64
4-7 Enclosure for ladle inoculation 66
4-8 Hooded pouring station 70
4-9 Mold tunnel 71
4-10 Movable pouring hood 72
4-11 Side draft hooding for vibrating shakeout 74
4-12 Rotary shakeout 76
4-13 Swing frame grinder booth with transfer car 77
4-14 Downdraft grinding control 78
4-15 Mixer and muller ventilation 80
4-16 Control system for shell molding 82
5-1 Flow diagram to determine the need for R&D . 88
vi
-------
TABLES
Number Page
1-1 Sources of Fugitive Emissions from Iron Foundries 5
2-1 Chemical Specifications for Iron Castings 8
2-2 Fugitive Emission Sources in Iron Foundries 17
2-3 Major Sources of Fugitive Emissions 19
2-4 Principal Organic Core Binders in Use in the United
States 25
2-5 Products of Thermal Decomposition of Sand Binders 27
2-6 Functional Groups Observed in Infrared Absorption Spectra
of Condensed Liquid Phases 28
3-1 Analysis of Fugitive Emission Measurement Methods 35
3-2 Fugitive Emissions Inventory 48
4-1 F.ugitive Emissions Control Technology Summary 51
-4-2 :'. Comparison of' Emissions from GreenvSand and Permanent .Mold :"
•: Processes for Producing a 13-Lb Uncored Casting under.
Ventilated Conditions '•.. ... .... . '. . ... ..:.... . . .-.. - 68
4-3 Pollutant Removal Systems for Shelieore and Mold Machines . 83
4-4 Example Surface Crusting Agents for Storage Piles and Ex-
posed Areas 86
5-1 Ranking of Particulate Emission Sources .... 90
5-2 Current Research 94
vii
-------
SUMMARY
The study reported herein was directed to the assessment of fugitive emis-
sions of air pollutants discharged from process operations in iron foundries,
and the need for the development of control technology for the most critical
sources. The major study tasks included (a) identification and quantification
of fugitive emissions based on available data; (b) prioritization of sources
according to the need for emissions control; (c) analysis of current control
practices and deficiencies in control technology; and (d) recommendation of
research and development programs to provide the required control technology.
It was found that the most significant fugitive emissions control problem
in foundries is the pouring of hot metal into sand molds and the subsequent
cooling of castings of these molds. Other significant fugitive emissions
sources which have control problems are the electric arc furnace, preparation
of molds and cores using organic binders, and casting shakeout. Research and
development ^programs are recommended ,(a) to better quantify the-mp's.t. signifi.--
cant sources;, (b) to evaluate,currently available control .technology for elec-
tric arc furnaces and,shakeout; and (c) to. develop new control technology for
pouring'and coolingrand for core and mold preparation using organic binders.
-------
CONCLUSIONS
The major conclusions derived from this investigation are as follows;
1. For most fugitive emissions sources in iron foundries, data are in-
sufficient to determine accurate emission factors with a high level of con-
fidence.
2. Accurate emissions testing methods for fugitive emissions sources in
foundries have not been adequately developed.
3. Many fugitive emissions sources, especially those involving sand han-
dling and preparation, are well controlled in most foundries for worker health
reasons.
4. By far, the most significant fugitive emissions control problem in
foundries is the pouring of hot metal into sand molds and the subsequent cool-
ing of castings in these molds.
5. Other significant fugitive emissions sources which have control prob-
lems are the electric arc furnace, preparation of molds and cores using or-
ganic binders, and casting shakeout.
6. Little research on iron foundry fugitive emissions problems is cur-
rently being conducted. The major exceptions are a NIOSH study on control
methods employed at the best controlled foundries and a multimedia study by
the Environmental Protection Agency. The results of these studies should be
carefully reviewed.
7. Research and development programs are recommended (a) to better quan-
tify the most significant sources; (b) to evaluate currently available control
technology for electric arc furnace and shakeout and (c) to develop new con-
trol technology for pouring and cooling and for core and mold preparation us-
ing organic binders.
-------
SECTION 1.0
INTRODUCTION
Until recently the national effort to control industrial sources of air
pollution has focused on emissions discharged from stacks, ducts or flues and
carried to the point of discharge in confined flow streams. Control strategies
have been based on the assumption that the primary air quality impact of in-
dustrial operations resulted from the discharge of air pollution from conven-
tional ducted sources.
However, failure to achieve the air quality improvements anticipated from
the control of ducted emissions has spurred a detailed reexamination of the
industrial air pollution problem. Evidence is mounting which indicates that
fugitive (non-ducted) emissions contribute substantially to the air quality
impact of industrial operations and, in certain industries, may greatly exceed
the effects, of-stack emissions. . ; ,'/•:•,.. •• •'' . '."',".'"' • :"?•••' .
Iron foundry processes,-which are'characteristically batch or semicohtin-
uous Derations, entail the generation of substantial quantities of fugitive
emissions at numerous points in the process cycle. Frequent materials handling
steps occur in the storage and preparation of raw materials and in the disposal
of process wastes. Additionally, fugitive emissions escape from reactor vessels
during charging, process heating, and tapping.
Fugitive emissions occurring in iron foundries constitute a difficult air
pollution control problem. Emissions are discharged with a highly fluctuating
velocity into large volumes of carrier gases having poorly defined boundaries.
Emissions from reactor vessels contain large quantities of fine particulate
with smaller amounts of vaporous metals and organics in hot, corrosive gas
streams. Enclosures and hooding of fugitive sources, with ducting to conven-
tional control devices, have met with limited success in controlling emissions.
This report presents the results of an engineering investigation of fugi-
tive emissions in the iron foundry industry. This study was directed to the
accomplishment of the following objectives:
1. Identification of fugitive emission sources within integrated iron
foundries.
-------
2. Prioritization of identified emissions sources based on relative en-
vironmental impact»
3. Recommendations of future research, development and/or demonstration
to aid in the reduction of fugitive emissions from the sources determined to
be the most critical.
Fugitive emissions in the iron foundry industry generally come from one
of five process areas: (a) raw materials storage and handling; (b) melting
and casting; (c) finishing; (d) core and mold preparation; and (e) waste mate-
rials handlings A complete listing of the sources is given in Table 1-1.
The technical approach used to conduct the subject investigation con-
sisted of the performance of the following seven program tasks:
Task 1 - Identify Fugitive Emission Sources; A comprehensive information
collection and data compilation effort was carried out to identify all poten-
tially significant sources of fugitive emissions occurring within iron found-
ries.
Task 2 - Quantify Fugitive Emissions; Available emissions data based on
source tests and estimating techniques were used to characterize the types and
quantities of fugitive emissions from sources identified in Task 1.
Task 3•- Review Existing Control Technology; Information was collected
and analyzed .to .evaluate the effectiveness of available systems and techniques
applicable to the control of fugitive emissions.
Tasks 4 and 5 - Develop Emissions Classification System and Classify
Emissions; A generic classification system was developed and applied to iden-
tify the similarities and differences in fugitive emission sources, thereby
defining generalized control problems which might most effectively be treated
in an integral manner.
Task 6 - Determine Critical Control Needs; Using background information
developed in previous tasks, the identified fugitive sources were ranked ac-
cording to their relative environmental benefit of (or need for) emissions
control requiring, if necessary, the development and demonstration of effec-
tive control techniques.
Task 7 - Recommend Research and Development Programs; Having identified
and prioritized control needs in Task 6, priority R and D program areas were
recommended to address these needs, taking into account deficiencies in avail-
able control technology and the expected results of research programs already
underway.
-------
TABLE 1-1. SOURCES OF FUGITIVE EMISSIONS
FROM IRON FOUNDRIES
1.0 Raw Materials Storage and Handling
1.1 Storage
• Coke
• Sand
• Scrap
1.2 Handling and Transfer
• New sand handling
• Spent sand handling
• Coke handling
2.0 Melting and Casting
2.1 Cupola
• Tapping
2.2 EAF
• Charging
• Tapping
• Leakage
2.3 Induction Furnace
. :2.4 Inoculation .. . • . '•;•! •
'2.5 Pouring and Cooling
3.0 Finishing
3.1 Shakeout
3.2 Grinding
4.0 Core and Mold Preparation
4.1 Mulling
4.2 Shell or hot box
• Heating
• Holding pallet
4.3 Cold set box
4.4 Core wash
5.0 Waste Handling
5.1 Slag quench
5.2 Waste sand transfer
5.3 Sand and slag storage
-------
This report is organized by subject area as follows:
. Section 2 identifies 'fugitive emission sources within iron foundries.
. Section 3 presents data on the quantities of fugitive emissions.
. Section 4 summarizes control technology applicable to fugitive emis-
sions sources.
. Section 5 presents a ranking of critical control needs and defines
priority R&D program areas directed to the development of control
technology for fugitive emissions.
. Section 6 lists the references cited in this report.
. Section 7 presents a Glossary of Terms which defines special terminol-
ogy used in this report to describe and characterize fugitive emission
sources.
. Section 8 is an English to metric conversion table.
-------
SECTION 2.0
FUGITIVE EMISSION SOURCE IDENTIFICATION
The iron foundry industry utilizes iron and steel scrap to manufacture
cast iron products ranging in size from less than an ounce to several tons per
casting. Many of the processes involved in the production of castings have the
potential for release of gaseous and/or particulate fugitive emissions to the
foundry environment and subsequently to the external atmosphere. The first two
subsections below describe the basic foundry processes and identify those
sources which have fugitive emissions potential. Industry-wide materials flow
is developed in the final subsection.
2.1 GENERAL PROCESS DESCRIPTION
The typical iron foundry processes various grades of iron and steel scrap
'to produce iron castings. For- the remainder of 'the report, .any foundry which
produces gray, ductile or malleable iron will be considered an iron foundry.
This classification is reasonable in .that most processes used to produce the
three types of iron are identical. ALso, the chemical specifications for gray
iron shown in Table 2-1 nearly encompass those for ductile and malleable iron
given in the same table.
The four basic operations present in all foundries are: raw materials
handling and storage, melting, pouring of metal into molds and removal of
castings from the molds. Other operations present in many but not all found-
ries include: (a) preparation and assembly of molds and cores; (b) mold
cooling; (c) shakeout; (d) casting, cleaning, and finishing; (e) sand han-
dling and preparation; and (f) hot metal inoculation.
For purposes of this report the iron foundry has been divided into five
areas of operation:
1. Raw materials storage and handling.
2. Melting and casting.
3. Cleaning and finishing.
-------
TABLE 2-1. CHEMICAL SPECIFICATIONS FOR IRON CASTINGS-'
I/
Element
Carbon
Silicon
Manganese
Sulfur
Phosphorus
Gray iron
(%)
2.5-4.0
1.0-3.0
0.25-1.0
0.02-0.25
0.05-1.0
Malleable iron
(cast white )
(%)
2.00-2.60
1.10-1.60
0.20-1.00
0.04-0.18
0. 18 maximum
Ductile iron
(7.)
2.4-4.0
1.8-2.8
0.10-1.00
0, 03 maximum
0. 10 maximum
-------
4. Mold and core preparation
5. Waste handling
A general flow diagram of foundry operations is presented in Figure 2-1. Block
diagrams of each of the first four basic areas are presented in Figures 2-2
to 2-5. A separate figure (Figure 2-6) is presented for sand handling, which
may be involved in all areas of operation. It should be noted that while most
iron foundries have operations falling into each of the broad categories
listed above, the foundry industry is so diverse that specific operations
will vary greatly from plant to plant. Described below are the operations
most commonly utilized in iron foundries.
As can be seen in Figure 2-1, raw materials enter the foundry in one of
two areas, :the melt shop or the core and mold-making area. At the melt shop
the primary raw materials are iron scrap, borings and turnings, limited quan-
tities of pig iron and foundry returns used for metallic content, coke for
cupolas and fluxing material such as limestone, dolomite, fluorspar, and cal-
cium carbonate. The metallics are generally melted in one of three furnace
types: cupola, electric arc or electric induction furnace. Reverberatory or
air furnaces are currently in limited use. After the iron is melted, required
ladle additions are made, either in the furnace or the ladle, and the iron
is transferred by ladle to the pouring area for casting in molds.' •
Upon reaching the casting area, the hot metal is poured into a mold to
produce an iron casting. The four types of molding processes which have re-
ceived most attention are green sand molds, shell sand molds, cold set molds
and permanent molds or centrifugal casting. Of these, green sand molding is
by far the most prevalent. Reference 2 discusses dry sand molding, the full
mold process and the Rheinstahl process; however, they were not examined as
a part of this study. If a sand mold is used, the mold and casting are then
transferred to a shakeout area where the casting is removed from the sand.
The spent sand is then recycled and the casting is taken to the finishing
shop for cleaning, grinding and finishing.
Further descriptions of the specific foundry operations are included in
the following section on sources of iron foundry fugitive emissions.
2.2 FUGITIVE EMISSIONS SOURCES
Iron foundries contain a variety of process sources with the potential
for emitting gaseous or particulate air pollutants to the plant environment
and on to the atmosphere. As indicated in Section 2.1, specific operations dif-
fer greatly in different foundries. Hence the specific operations which present
an emissions problem in one foundry may not be a problem in another foundry.
-------
LIMESTONE
BINDER
SAND
00 00" 'OO
SAND
STORAGE
BIN
CORE & MOLD MAKING AREA
(See Figure 2-3)
CORE SAND
SAND STORAGE
CORE BINDER (fj) (Th
BINDER STORAGE
EMISSIONS
CAS1INGS
SHIPPING
Figure 2-1. Composite flow diagram for the gray iron foundry industry.
(Numbers refer to source listings in Table 2-2.)
-------
From Foundry
From Machin
Shop or
Scrap Dealer
T I
Carbonates
Fluorides
Carbides
2/
Figure 2-2. Raw material storage and handling.—'
11
-------
Sand and
Binders or
Preodred
Sand
Sand and
Binders
Chemical
Catalyst
Figure 2-3. Mold and core making.
(Numbers refer to
Table 2-2.)
12
-------
Electric
Induction Furnace
3/
Figure 2-4. Melting and casting.-
13
-------
4/
Figure 2-5. Cleaning and finishing.—'
14
-------
I/I
Sand Reclamation
System
L ~L
Used
Sond Bin
Reclaimed
Sand Bin
New
Sand
Cereal
Binder Bin
Clay
Bin
Other
Additives
Resin
Bin
Acli\
Refuse
Figure 2-6. Sand handling.—
5/
-------
Based on a limited number of contacts with industry personnel, probable sources
of fugitive emissions were identified and are presented in Table 2-2. Obviously,
all of these are not major sources of fugitive emissions in the foundry industry.
Those fugitive sources which may be major sources at a specific plant are
presented in Table 2-3. For each of these sources, available data on composition
of the emission stream are also given. Each of these sources is discussed in the
following subsections.
2.2.1 Raw Materials Storage and Handling
Raw materials are used in two areas of the foundry. Metallies and possibly
coke and some type of fluxing material are needed for the production of molten
iron in the melt shop. Sand and binders or a prepared mixture of sand with
binders are needed for mold and core preparation. Depending upon the method
used, both the storage and handling of these materials may become a fugitive
emissions problem. However, appropriate processing and storage methods should
minimize emissions.
2.2.1.1 Storage—
The primary materials entering storage ares (a) the sand which is used
for core and mold production, (b) various grades of scrap to be charged as
metallics to the furnace, and (c) coke for firing of the cupola. Minor quan-
tities of fluxing materials, refractories and binders are also received by
foundries. However, these are of little significance as emission sources.
Generally foundries purchase mold and possibly core sand which is washed,
dried, and screened before shipping. Upon receipt the sand is transferred by
covered conveyor or pneumatic tube to an enclosed storage area to preserve
sand quality. Hence new sand storage should present minimal emissions problems.
Coke and scrap arrive at the foundry by railcar or truck. These have
traditionally been stored in open scrapyards. However, many foundries now have
covered storage areas to prevent degradation from weathering of both coke and
scrap. Minimal emissions from storage of coke and dirty scrap may be generated
during load-in and load-out from both open and covered storage. Wind erosion
from open yards and those covered areas not having protection against the wind
may generate some emissions-,
2.2.1.2 Handling and Transfer—
The sand handling system shown in Figure 2-6 can be the largest source of
fugitive emissions in an iron foundrya However, many of these emissions are
larger particles (> 50 jim in diameter) and so will settle out before leaving
the foundry. As indicated above, sand is often transferred to storage areas
pneumatically. However, if mechanical means are used, conveyor dump and trans-
fer points will be sources of fugitive emissions.
16
-------
TABLE 2-2. FUGITIVE EMISSION SOURCES IN IRON FOUNDRIES
Source-
No.
Source identification
Pollutant
Partic- Sulfur Carbon Metal
ulate oxides monoxide fumes
Hydro-
carbons
Raw materials storage & handling
1
2
3
4
5
6
7
8
9
-10
11
12
13
14
15
Melting
16
17
18
19
20
21
22
23
Limestone handling
Unloading
Transfer to storage
Storage
Trans f er to furnace
Coke handling
Unloading
Transfer to storage
Storage pile
Transfer to furnace
Metallic charge handling
Unloading
Storage pile '
Transfer to furnace
Binder unloading
Binder storage
Sand unloading
Sand storage
& casting
Cupola furnace
Tapping
Charging
Electric arc furnace
Charging
Leakage
Tapping
Induction furnace
Charging
Melting
Tapping
X
X
X .
X
X
X
X
X
X
•'. " ; '.X:;':'"W •" "":"'•'' ' '-''"." - ' '.
. X
X' • '"•' •.••••••••;
X
X
X
X X X X
X X
X
X XX
X X
X X
X X
X X
(continued)
X
X
X
X
X
17
-------
TABLE 2-2. (continued)
a/
Source-
No.
24
25
26
Cleaning
27
28
29
30
Pollutant
Partic- Sulfur Carbon Metal
Source identification ulate oxides monoxide fumes
Iron inoculation
Pouring
Cooling
& finishing
Shakeout
Return sand system
Cooling & cleaning
Grinding
X X
X X
X X
X X
X
X
X
Hyd ro -
carbons
X
X
X
Mold & core preparation
31
32
33
34
35
36
37
38
39
40
41
Sand charge to mixer/muller
Dry sand mixing or mulling
Holder
Cold set mold
Oven bake core box
Core oven leakage . '.'.-..
Shell or hot box heat
Cold box core or mold
No bake core box
Core cooling
Core wash
X
X
X
X
X
X
X
X
X
X
X
X
X
Waste handling
42
43
44
45
46
Slag quench
Waste sand transfer
Waste material storage
Transfer to landfill
Baghouse catch
X X
X
X
X
X
a/ Sources are identified by number in Figures 2-1 and 2-3
18
-------
TABLE 2-3. MAJOR SOURCES OF FUGITIVE EMISSIONS
Emissions source
Pollutant
a/
Concentrations.'
(uncontrolled)
Particle size
distribution—
Raw material storage and
handling
Storage piles
Sand
Coke
Scrap
Handling and transfer
Sand transfer to
storage
Sand transfer from
storage to mold area
Coke transfer
Return sand conveyors
Particulate
Particulate
Particulate
Particulate
Particulate
Particulate
Particulate
3-5 gr/cu ft
3-5 gr/cu ft|' .
3-5 gr/cu ft-7 30-1,000 un£7
3-5 gr/cu ft-7"
3-5 gr/cu fti/
3-5 gr/cu fti7"
507, 2-15
30-1,000
507, 2-15 var
507, 2-15
6/
6/
1/2-2 gr/cu ft°/ 507, 7-15 um-
(907, > 50 ]im)-/
Mold and core preparation
Mulling (charging and
mixing) •
Shellcore and mold heat
Cold set core
Box exhaust
Holding pallet
Particulate
CO
Formaldehyde
Amines
Phenol
Amines
Phenol
Formaldehyde
Amines
Phenol
Formaldehyde
3-5 gr/cu ft~
700
10 ppml/
250 pproZ/
20
400-4,000
3-20 Ppm|/
5-15 ppm2'
350-1,400 ppm§/
3-8
507, 2-15 urni7"
3-5 ppmi/
Melting and casting
Cupola tapping
EAF charging
Tapping
Leakage
Induction furnace
(charging, tapping,
and leakage)
Particulate
Particulate
Hydrocarbon mist
Particulate
Particulate
Hydrocarbon
Particulate
(continued)
< 0.7
807, < 5 umi2/
0.1-1 urn!/
807, < 5 .
807, < 5 um!P_/
19
-------
TABLE 2-3. (continued)
Emissions source
Iron inoculation
Metal pouring
Cooling
Cleaning and finishing
Shakeout
Grinding
Waste handling
Slag quench
Waste sand transfer
Sand and slag storage
Pollutant
Particulate
Particulate
CO
Hydrocarbon
Particulate
Particulate
CO
Hydrocarbon
Particulate
Particulate
so2
Particulate
Particulate
Concentration—
(uncontrolled)
Heavyi/
0.00291
gr/scfli/
1,500 pproli'
250 ppm —
0.00291
gr/scfH/
0.1654
gr/scfii/
670 ppmll/
215 pproll/
0.5-5 gr/scf£/
-
, ™
. -
Particle size
distribution-/
< 0.7 W
98% < 15
(977= <
98% < 15
(97% <
98% < 15
(46% <
-
—
,
>/
6 um)
Um12/
6 urn)
6 urn)
a/ Superscripts refer to references in Section 6.
20
-------
As cereal and clay binders are added to the sand this will increase the amount
of fines and hence the potential for emissions. As sand and binders are mixed
in the muller, they are generally wetted, and emissions are not generated for
the remainder of the preparation process. After shakeout, the spent sand is
again dry and has the potential for emissions. All conveyor drop points and
screening and reclamation operations will be sources of fugitive emissions.
Methods for handling scrap and, in foundries using cupolas, coke vary
greatly. Smaller, older foundries may transfer materials manually using fork-
lifts and buckets and hand charging while larger foundries may have a completely
mechanized materials handling system. In either case the amount of emissions is
dependent upon the quality of the scrap or coke. If the scrap or coke contain
significant amounts of fines, vibrating conveyors, conveyor drop points, and
manual materials dumps will all be sources of fugitive emissions. However, han-
dling of coke and scrap is generally a minor source of emissions.
2.2.2 Melting and Casting
The operations that occur in an iron foundry from the time scrap is charged
into a furnace for melting until the time the casting is to be removed from the
mold constitute the greatest fugitive emission sources for which generally ap-
plicable control measures have not been identified. Most iron castings are pro-
duced from scrap which has been melted in either a cupola, an electric arc fur-
nace (EAF), or an electric induction furnace. The primary fugitive emission
sources from melting are (a) cupola tapping; (b) the total EAF cycle; and (c)
induction furnace charging and melting. Other major emission sources in this
area include (a) inoculation of ductile iron, (b) pouring hot metal into molds,
and (c) cooling the filled molds before shakeout.
2.2.2.1 Cupola—
The cupola furnace is an upright brick-lined cylindrically shaped vessel
which uses the heat from the charged coke to melt iron. The cupola operation
is continuous, with metallics, coke and fluxing agents being charged in layers
at the top and the molten iron tapped from the bottom. The operation and pri-
mary emissions problems of the cupola are described in detail in Reference 13
and are not discussed here.
Because the cupola is kept under negative pressure for emission control
purposes, charging is generally not a fugitive emissions problem. The only
source of fugitive emissions is the tapping of the molten metal from the
furnace. The metal is tapped in one of two ways. In the first case, the
metal is tapped to a forehearth where the slag is skimmed and then the iron
is transferred into a ladle for pouring. In this case, the slag skimming
and transfer into the ladle are minor sources of fine particulate emissions.
In the other case, the metal is tapped directly to a ladle and the slag is
skimmed from the ladle. This is also a minor source of fine particulate
emissions.
21
-------
2.2.2.2 Electric Arc Furnace—
The electric arc furnace (EAF) is a refractory-lined cup-shaped vessel
with a refractory-lined roof. Three graphite electrodes are placed through
holes in the roof to provide the electrical energy for melting iron0 The EAF
can be charged through the side, or the roof can be removed and the furnace
charged from the top. Most newer and larger furnaces are of the top charge
variety.
Primary emissions control for the EAF during melting is generally accom-
plished through some form of direct shell evacuation (DSE) or by the use of
a canopy hood. The problem with the DSE system is that it is not operational
when the roof is rembved for charging or tapping, both of which are considered
to be major sources of fugitive particulate emissions. Also, if oily scrap is
charged to the EAF, charging can be a source of hydrocarbon emissions. It is
also possible to have a small amount of leakage, around the electrodes during
melting, but this is a relatively minor source of emissions if the primary
system is operating properly. However, malfunction or inferior design of the
primary DSE system can lead to a major fugitive emissions problem.,
In the case where a canopy hood is used as the primary emissions control
system, emissions from all phases of the operation are captured to some degree.
However, inefficient capture can result from cross drafts and cause significant
quantities of emission to escape. This constitutes a major fugitive emissions
problem.
2.2.2.3 Electric Induction Furnace-- . .
The two types of induction furnaces used in foundries are the channel in-
duction and coreless induction furnaces. The coreless induction furnace is'most
often used for iron melting and this type presents the more significant fugitive
emission problem. The coreless induction furnace is a cup-shaped vessel which
uses electrical energy to induce eddy currents in the metallic charge to produce
molten iron. Since very clean or preheated scrap must be charged to the induc-
tion furnace, emissions are generally less than the cupola or the EAF. Hence,
these furnaces are often uncontrolled. In that case, the total furnace operation
becomes a fugitive emission problem.
2.2.2.4 Iron Inoculation--
Approximately 1570 of the iron castings produced in the United States util-
ize ductile iron. Generally^ ductile iron is produced by the addition of mag-
nesium or similar alloying compound to the molten iron after it has been tapped
to the ladle. Several methods are used to introduce the magnesium to the molten
metal. Several of these are illustrated in Figure 2-7. These methods are de-
scribed in detail in Reference 14. Since the primary emissions from iron inoc-
ulation are magnesium oxide particulate, the severity of the emissions problem
is related to the level of magnesium recovery of the particular process. Modi-
indicates that three processes which have yielded good results and are widely
applied are (a) pourover method, (b) sandwich methods, and (c) plunging method.
22
-------
"SANDWICH
'TRIGGER'
POUR-OVER
THROW-IN
PLUNGING
14/
Figure 2-7. Methods of iron inoculacion.—
23
-------
Of these three, sandwich methods appear to have the best magnesium recovery
and hence least emissions.
2.2.2.5 Iron Pouring and Cooling—
Two of the most significant sources of fugitive emissions in the iron
foundry are the pouring of hot metal into sand molds and subsequent cooling
of the castings. These processes vary significantly in different foundries.
In nonmechanized foundries, the molds are generally placed in a large open
area. The hot metal ladle is then moved by a overhead pulley system to the
mold and the casting is poured and cooled in place. In more mechanized found-
ries, the mold is placed on a conveyor and moved to the pouring station and
then moved onto a cooling area. Emissions problems are comparable for both
processes. The emissions are contained in a relatively high temperature, buoy-
ant, moist stream. The constituents of the stream are fine metallics from the
hot metal and organics produced by thermal decomposition of the binders, as
discussed in Section 2.2.4. The damp buoyant stream and the organic emissions
make control of these sources difficult.
2.2.3 Cleaning and Finishing
The only major sources of fugitive emissions in the finishing area are
casting shakeout and grinding. Shakeout is the method by which the iron cast-
ing is removed from the sand mold. Shakeout varies more from plant to plant
than any other foundry operation. Observations during a limited number of
foundry visits revealed shakeout being accomplished manually by forklift or
hand shovel, mechanically on-a .grate shakeout, and by elevating the flask and
pneumatically shaking the sand and casting out. In any case the emissions con-
sist of dust from the dried sand, organic residue from binders, and water va-
por. Ease of control appears to be dependent upon the type of shakeout used.
Grinding may also be a source of fugitive emissions in an iron foundry.
Four basic types of grinders are used in foundries: bench, floor stand, por-
table, and swing. Each of these is a source of particulate emissions. Little
information was obtained on the partizle size or total amount of emissions from
grinding. However, some plant operators indicated that the finishing room was a
significant industrial hygiene problem.
2.2.4 Mold and Core Preparation
The primary fugitive emissions problem in the core and mold preparation
area is the release of organic vapors from the binders used in heated core box,
no-bake, and cold box cores and molds. The major organic binders in use in the
United States are shown in Table 2-4. In addition, dust emissions may be pro-
duced in the dry mixing of sand and binders.
Shell core-making or shell-molding is a process whereby cores or molds
having a thickness of 1/8 to 3/8 in. are produced. These are used for the most
24
-------
TABLE 2-4. PRINCIPAL ORGANIC CORE BINDERS IN USE IN THE UNITED STATES—
177
Binders
Approximate annual
current consumption
Organic
Oven bake
90 x 106
1. Oleoresinous
2. Urea-formaldehyde resins
3. Phenol-formaldehyde resins
4. Cereal binders
Heated core box
Shell
Hot box
Gassed core
No-bake
85 x 106
45 x 106
3 x 106
20 x 106
1. Phenol-formaldehyde novolaks
1. Furan resins (UFFA)
2. Phenol resins (UPF)
3. Phenol-modified resins
1. Cold box (isocyanate)
1. Air set (oil-oxygen)
2. Furan no-bake
3. Oil no-bake
4. Urethane (phenolic-isocyanate)
25
-------
part in applications requiring a great amount of precision. Sand and approxi-
mately 57o thermosetting resin (usually having a phenol formaldehyde base) may
be dry mixed in a muller..^? 16/ xhe sands may also be prepared by cold, warm,
or hot coating. This mix is then blown into a metal box housing the pattern
plate, which has been heated to a temperature of 350 to 700°?.-^=' The binder
within 1/8 to 3/8 in. of the pattern is melted and the material turned into
a dough-like substance. Excess sand is dumped off and the shell is then
hardened. The primary emissions from the process are CO, formaldehydes,
amines, ammonia, and phenols^
Hot box binders are those resins that rapidly polymerize in the presence
of acidic chemicals and heat to form a mold or core3 The original hot box
resins were developed by modifying urea-formaldehyde resins with the addition
of 20 to 457» of furfuryl alcohol. This type of hot box resin is commonly re-
ferred to as a furan resin. The furan resins were then modified with the addi-
tion of phenol to produce urea-phenol-formaldehyde hot box resins, which are
referred to as phenolic resins or UPF resins. The UPF resins have a pungent
odor and adequate ventilation at the core-making machines is required. More
recently, urea-free phenol-formaldehyde-furfuryl alcohol binders have been
developed. These have a much lower volatile content and odor compared with
other hot box resins as a consequence of eliminating urea from the formula-
tion.—
A two-part polyurethane cold box binder system was developed about 1967
that required gassing rather than b'aking or heating to achieve a cure. Part I
of the system is a phenolic resin, and Part II is a polyisocyanate, both dis-
solved in solvents. In the presence of a catalyst, triethylamine (TEA) or di-
methyl ethylamine (DMEA), the hydroxy groups of the liquid phenolic resin com-
bine with the isocyanate groups of the liquid polyisocyanate to form a solid
urethane resin which serves as the sand binder. Following introduction of the
catalyst into the cold box, air is used to sweep any remaining vapors through
the core, after which the core is removed from the core box. The amine cata-
lysts are volatile, flammable, organic liquids and excessive vapors present
safety hazards*=•?•'
The so-called no-bake binders represent modifications of the oleo-
resious, urea-formaldehyde, phenol-formaldehyde and polyurethane binder sys-
tems previously described, in which various chemicals are incorporated to pro-
duce polymerization in an unheated core box.—
Decomposition products of the various binders are presented in Tables 2-5
and 2-6. It should be noted that these values were obtained by direct venting
of prepared cores and are not representative of in-plant ambient levels.
Three other possible sources of emissions in the mold and core area are
mulling, molding, and core washing. After castings are removed at the shakeout,
the spent sand is returned to the muller where it is mixed with water and
26
-------
TABLE 2-5. PRODUCTS OF THERMAL DECOMPOSITION OF SAND BINDERS!!/
N>
Concentration in effluent (ppm by volume )£'
Product
Carbon monoxide
Hydrogen cyanide
Methane
Ethylene
Acetylene
Carbon dioxide
Ammonia
Aldehydes (as
formaldehyde )
Phenol^/
Threshold limit
value (ppm)£/
50
10
-
-
-
5,000
25
2
5
Polyurethane
40,000
16
2,000
1,500
1,500
7,000
> 1,500
200
17.5 mgd/
Oil base
40, 000
400
40, 000
7,000
1,500
11,000
500
> 400
0.6 mgl/
Urea-
formaldehyde
. 40, 000
320
2,000
1,500
1,500
7,000
1,500
400
1.5 mgl/
Phenolic
40, 000
60
2,000
1,500
1,500
1,000
-
> 400
0.4 mgl/
a/ All products except phenol were determined in the gas phase. The approximate volumes of the gas phases
collected from each binder material were as follows: polyurethane, 200 ml; oil base, 300 ml; urea-
formaldehyde, 1,000 ml; phenolic, 200 ml.
b/ Threshold limit values (TLV) established by the American Conference of Governmental and Industrial
Hygienists.
c/ Phenol was determined in the condensed liquid phase.
d_/ The values given are the total weights of phenol found in the condensed liquid phase.
-------
TABLE 2-6. FUNCTIONAL GROUPS OBSERVED IN INFRARED ABSORPTION
SPECTRA OF CONDENSED LIQUID PHASES*iIZ/
Binder material
Urea-
Functional group Polyurethane Oil base formaldehyde Phenolic
Aliphatic CH + 4- 4- +
Aromatic CH 4- 4- 4-
Ester C=O +4- 4-
COOH 4- +
Aldehyde C=0 + 4-
Amide 4- +
Secondary amide 4-
Acidic OH . 4-
Phenyl 4-4- 4-
Substituted phenyl 4-4- 4-
a/ The total weights of the -condensed liquid phase collected.from each
binder material were as follows: polyurethane, 120 mg; oil base,
500 mg; urea-formaldehyde, 200 mg;•phenolic, 80 mg.
28
-------
makeup sand and binders. After the sand enters the muller, it is moistened
and should present minimal emissions problems. However, transfer of the spent
sand to the muller can be a source of particulate emissions.
Sand molding at a foundry is generally accomplished by manually or mechan-
ically packing the sand in the risers. Transfer of material or vibrating of the
mold may result in limited quantities of particulate.
After cores have been cured, most are coated with water, alcohol, or .
naptha-based washes. Common materials included in the wash composition are:—
Graphite Asbestos
Silica Mica
Talc Coal
Magnesite Coke
Alumina Coal tar
Zircon Pitch
These washes can be applied through either spraying or dipping. After the
wash has been applied, moisture is evaporated by torching in the case of water-
based washes and by ignition in the case of naptha or alcohol-based washes.
Core washing may be a source of both particulate and organic emissions.
2.2.5 Waste Handling
The primary waste materials produced at a foundry are (a) the slag from
the melting operations and (b) spent sand from molds and cores. With each type
of melting furnace, the molten metal is transferred to a holding furnace or
ladle. The slag rises to the top of the metal and is skimmed off and solidi-
fied by air cooling or water quenching. Either cooling method can produce
small amounts of SC^ and ^S with water quenching producing higher levels.
In foundries using green sand molds, about 2 to 3% of total foundry sand
is replaced daily to insure proper sand quality. This sand is generally stored
temporarily in either outdoor piles or hoppers. It is then transferred (along
with slag) to a landfill for disposal. Potential for dust emissions exists dur-
ing handling and transfer of the materials to the storage area. If outdoor
piles are used, emissions can be generated from wind erosion. Finally, if cov-
ered transport vehicles are not used, emissions are generated during the trans-
fer to a landfill for final disposal.
2.3 MATERIALS FLOW
As an aid to determining the relative importance of sources of fugitive
emissions in the foundry, an industry-wide materials flow diagram was devel-
oped, as shown in Figure 2-8. The materials flows were estimated using 1974
as a base year.
29
-------
u>
o
Slog and Wasle
11
Coke
| 27
(0.77)
Scrap
20 3
(1.23)
1
87
21.7 ^
/ i 121
\ ' • J' 1
4'9 ^
(0 30)
'•9 ^
(0.12)
°-5 .-
(0.03)
1(
f 1
Slag and
!.
Electric
f(
n
Electric
Induction
O.O/)
A). 6
( 1 2"il
Wosle
0.3
0.02)
4.6
(0 28)
Waste
01
Of\\\
1.8
(0.11)
0.5
(0.03)
.-
3.7
(n ??^
27.5 .;
(1-67)
5.5
3.3
16.5 Finished
(I 00)*" Costings
Mold/Core
System
Return Sand
149.33
(9.1)
154
(0.53)
Waste Scrap
2.3
(O.H)
Wosle S<
5.67
(0.34)
ind
(9.3)
•Numbers indicate annual
flows in 10° slioil Ions.
• Numbers in parentlieses
are ratios of material flows
to total iron produced.
Figure 2-8. Iron foundry industry weighted material flows.
-------
Scrap consumption in the iron foundry industry was calculated by di'ffer-
ence. In 1974, total scrap consumption for cupolas and electric arc furnaces
was 53.1 million tonsri2' and scrap consumption in the iron and steel industry
was 30.2 million tons,-£2' leaving foundry consumption at 22.9 million tons.
Assuming iron castings account for 88.8% of foundry production,-L2' iron foundry
consumption was 20.3 million tons per year.
Reference 19 reported total shipments of 16.5 million tons of gray, duc-
tile and malleable iron in 1974. Information obtained during foundry visits
suggests an average yield of 60% good castings. Thus, a total hot metal pro-
duction of 27.5 million tons was required. Assuming 1.05 tons of scrap are
required to produce 1 ton of hot metal, approximately 29.0 million tons of
scrap including internal foundry returns were charged to furnaces in iron
foundries.
Hot metal production was apportioned among furnaces as follows:
Cupola 74.8%
Electric arc furnace 16.8%
Induction furnace 6.7%
Other 1.7%
19/
Based on ductile iron production of 13.3% in 1974,— it was assumed in Figure
2-8 that 13.3% of the hot metal produced was treated.
Both sand and coke inputs were calculated using 1972 Census of Manufac-
turers data.for foundry consumption normalized to 1974 production levels. Total
sand throughput was calculated assuming a sand-to-metal ratio of 7:1. It should
be noted that the sand throughput value obtained is consistent with the 2 to 57,
removal values indicated by industry personnel during plant visits. Finally,
the division between sand molds and nonsand molds was assumed to be 4:1, as in-
dicated in Reference 9.
31
-------
SECTION 3.0
FUGITIVE EMISSIONS QUANTIFICATION
In order to adequately assess the need for control of fugitive emissions
in iron foundries, it was essential that estimates of fugitive emissions be
developed. In general, compilation of emissions inventories requires emission
factors, throughput data, and information on the level of control for each
process inventoried. The following subsections of this report (a) discuss the
methods by which emission factors are determined and the inaccuracies involved
in each of these methods; (b) present and analyze all emission factors developed
for the iron foundry industry, and (c) develop an uncontrolled total particulate
and fine particulate emissions inventory for major fugitive sources in the
foundry industry.
3.1 EMISSION FACTOR QUANTIFICATION METHODS
Usually the first step in conducting any emissions inventory is the deter-
mination of emissions factors in units of weight of emissions per-weight of
process throughput. The units used in this study will be pounds of emissions
per short ton of throughput (Ib/ton). These values can be converted to kilograms
per ton using the conversion factor in Section 8«0. Obviously the best way
to accurately determine emission factors is by several repetitions of a highly
reliable emissions testing method. However, fugitive emissions testing methods
are still in the developmental stages, and only a limited number of fugitive
emissions tests have been performed. Hence, it is necessary to estimate emission
factors based on stack emissions data and fugitive emissions data from similar
processes.
The following two subsections discuss the various fugitive emissions test"
ing methods and those estimating techniques which can be used if test data.
are not available.
3.1.1 Fugitive Emissions Testing Methods
Three basic strategies for sampling fugitive emissions have been proposed
under EPA Contract No. 68-02-1815-li/ These are (a) quasi-stack testing, (b)
roof monitor testing, and (c) upwind-downwind testing. Two additional methods
which have been examined under the current study are (a) exposure profiling
and (b) dilution profiling. Presented below is an analysis of the limitations
32
-------
of each method with respect to the quantification of iron foundry emissions
sources.
Quasi-stack testing requires the hooding or enclosure of the fugitive
emissions source in such a manner that complete capture of the fugitive emis-
sion stream is achieved. Negative pressure then draws the emission stream
through a ducting system where standard stationary source test methods are
used to determine emissions. Another version of this method utilizes the cap-
ture system on a well-controlled fugitive emissions source. Standard source
testing methods are used to determine the emissions by sampling in the ductwork
prior to removal of pollutants from the stream by a control device.
The quasi-stack sampling technique has limited application for the foundry
industry. The major problems encountered when using the first version of this
method are the costs involved in constructing an efficient capture system for
anything but small sources and the difficulty of designing a complete capture
system which does not interfere with normal foundry operations. The difficulty
involved in using installed capture devices to collect samples is that most
systems on foundry sources appear to be significantly less than 100% efficient.
It may be difficult to locate foundries with capture systems adequate for
sampling purposes.
Even if the problems stated above are overcome, two aspects of fugitive'
-emissions may limi.t the .accuracy .of the emissions measurements. First of all,
many of the fugitive emissions in iron foundries result from the fine particu-
late nature .of materials involved. It is possible that the negative pressure
applied at some sources may erode additional fine particulate from the process
stream. This will produce an artificially high emission rate for these sources.
Also, the accuracy of standard stack sampling methods is partially dependent
on a relatively constant emission stream. However, many fugitive sources have
a.high degree of variation with time. This may lead to inaccurate emission
factors if standard methods are used to measure the emission stream.
The roof monitor technique utilizes the natural or forced flow from roof
monitors or similar major building exhaust areas to determine total fugitive
emissions from a building. This technique uses some pollutant measuring devices
such as a hi-vol or gas sampling train and a scanning method (by moving one
sampler or by using multiple samplers) to determine an "average" pollutant
concentration escaping through the opening. The velocity of the exit gas stream
is measured and emissions are calculated.
Roof monitor sampling is best suited for those situations in which a lim-
ited number of easily distinguishable operations are conducted in the same
building. Foundry operations are such that this is not usually the case. Hence,
it is difficult to isolate specific sources for testing. The other major prob-
lem in roof monitor sampling is the difficulty in ensuring that the measured
33
-------
concentrations are in fact the average concentration escaping from the roof
monitor and that the measured velocity is representative of the velocity through-
out the stream.
Upwind-downwind emissions testing is generally applied to outdoor sources
such as roads or storage piles or to estimate emissions from a building or
complex of buildings. This technique uses ambient sampling stations upwind
and downwind from the source to determine the impact of emissions from the
source. A dispersion model is then used to calculate the contribution of the
source. Because of the low level of accuracy currently available from most
dispersion models, this method is of questionable value if greater than order-
of-magnitude accuracy is required* In addition, the method does not have the
capability to isolate most fugitive sources within an iron foundry.
The exposure profiling technique is a method which has been developed
by MRI for sampling of fugitive dust emissions from roads and storage piles.
It is also possible to apply the technique to indoor sources which have an
emission stream with a relatively constant convective direction. The method
requires a sampler upstream from the emissions source and a grid of isokinetic
samplers downstream from the source which completely define the emissions plume.
The system has the advantage that it is not dependent upon dispersion models.
However, the accuracy of the system is dependent upon constant directional
plume which can be isolated from the emission plumes from other sources. These
conditions may be.difficult to attain for most foundry sources.
•" As a .part of this-study, >a dilution profiling-technique was examined for'.;:
use on high temperature sources. The basic concept is the same as the exposure
profiling technique. However, rather than defining the plume by using a grid
of mass samplers, a minimum of two mass samplers are used to determine center-
line concentrations and the plume is profiled with temperature sensing devices.
The system was tested in laboratory experiments and appear to be a viable means
of testing high temperature sources with buoyant plumes. However, it has not
been tested on a full-scale operation. The system is limited to those sources
which emit a buoyant plume which has a temperature profile significantly higher
than ambient temperature.
An analysis of the applicability of the various emissions measurement
techniques for major fugitive emission sources in the foundry industry is pre-
sented in Table 3-1. The results indicate that only a limited number of tests
of fugitive emissions sources have been performed. In addition, little is known
about the capability of the various emissions measurement techniques for test-
ing iron foundry fugitive emissions sources with a high level of accuracy.
Further work on test methodology, both development and verification, is needed
if reliable fugitive emission inventories are to be developed.
34
-------
TABLE 3-1. ANALYSIS OF FUGITIVE EMISSION MEASUREMENT METHODS
U)
U1
Method applicability
Emissions source
Raw materials input
Storage piles
Coke
Sand
Scrap
Materials handling and transfer
New sand handling
Sand screening
Coke handling
Melting and casting
Electric arc furance
Induction furnace
Cupola
Iron inoculation
Iron pouring
Casting cooling
Product finishing
Shakeout
Grinding
Core and mold preparation
Mulling
Shellcore or hot box heating
Shellcore or hot box cooling
Cole set core
Core wash
Molding
.'••'.••• Roof
Quasi- stack monitor
.•.'•*.;• ^
F •: :- ;
F .• •;•' '
F
F '.'
E
E
E - " • •
- • - -
D ,
D
D
D.
A' " , ' '
c
C ; •
D . •
D
E
E .
E '
D -.
D
F
F
F
F
F
F
F
E
E
F
E
E
E
F
F
F
F
F
F
F
F
Upwind-
downwind
B
B
B
B
E
E
E
F
F
F
F
F
F
F
F
F
F
F
F
F
F
Exposure
profile
B
B
B
B
D
D
D
E
E
E
E
E
E
D
F
F
E
E
E
E
F
Dilution
profile .
F
F
F
F
F
F
F
E
E
E
E
E
F
F
F
F
E
F
E
F
(continued)
-------
TABLE 3-1. (continued)
Method applicability
Emissions source
Waste handling
Slag quench
Waste sand transfer
Sand and slag storage
Roof
Quasi-stack monitor
F F
F ' F
F F
Upwind-
downwind
B
F
B
Exposure
profile
D
B
B
Dilution
profile
F
F
F
u>
Note: A = Has been used to test a full-scale iron foundry operation,
B = Has been used to test a similar operation.
C = Has been used to test process on a bench scale.
D = Has not been used, but should be a viable method.
E = Has not been used; may be a viable method.
F — Not a viable method.
-------
If available test data are insufficient to develop a reliable emission
factor, it is necessary to use engineering judgment to develop estimates for
these factors. The estimated data presented in this report were generally.
developed in one of three ways: (a) as a fixed percent of uncontrolled stack
emissions, (b) by extrapolation of data for similar processes, and (c) based
on input/output data from a process and knowledge of the reactions involved
in the process. It should be noted that in all cases where estimated data are
used, a relatively low confidence should be placed in these values.
3.2 EMISSION FACTORS FOR THE FOUNDRY INDUSTRY
Available data on iron foundry fugitive emissions were obtained through
an extensive literature search, contact with knowledgeable EPA and industry
personnel, and comparison of iron foundry operations with similar processes
for other industries. It was found that almost no substantive test data are
available from full scale foundry operations. Thus, most of the emission factors
are engineering estimates based on limited data. Many of the emission factors
Q/ ?2/ 1 1 79/
were derived from information presented by Gutow,— Bohn,—' and Bates. •L»J^/
There are difficulties involved in the use of the studies in estimating foundry
emissions as described below.
Gutow presented emission factors for many of the fugitive sources within
.iron foundries.,.The methodology the Gutow used to develop these .emission fac-
tors was riot presented -in Reference 9. When further information was requested,
the author indicated that the factors were developed as a part of EPA Contract
No. CPA 22-69-106, Systems Analysis of Emissions and Emissions Control in the
Iron Foundry Industry. However, due to the time span since the completion of
the study in 1971, it is no longer possible to access the original data or
methodology.
Bohn, et al., developed emission factor equations for materials handling
operations in integrated iron and steel plants. These equations are based on
a limited number of exposure profiling tests at integrated iron and steel
plants. These operations differ from those in iron foundries with respect to
both size and location. However, since no test data are available for foundry
materials handling, the equations were used to estimate emission factors.
But, the reliability of these emission factors is questionable.
Bates and Scott have determined emissions from several foundry sources
through the use of quasi-stack sampling. However, each of the tests was con-
ducted on a bench or pilot scale operation, rather than a full-scale process
in an operating foundry. It is not possible to ascertain the effect of scale
on these emissions. Hence, the reliability of the emission factors developed
from these data is uncertain.
37
-------
Even though the data obtained from the sources described above and simi-
lar sources are of questionable reliability, they are the best available data.
Hence, these data were used to develop the emission factors in the following
sections*
3.2.1 Raw Materials Storage and Handling
No test data are available for materials handling operations in foundries.
Hence, data from Gutow2/ and Bonn, et al.H/ are used to estimate emission factors,
9/
Gutow~ gives an emission factor of 12.3 Ib/ton of melt for emissions
from dry sand handling, prepared sand handling and drying, and sand reclama-
tion. However, 90% of these emissions are greater than 50 /^m in diameter and
are likely to fall out in the foundry. Hence, only 10% or 1.2 Ib/ton melt of
sand handling emissions escape to the atmosphere.
9/
Gutow~ also presents an emission factor of 10 Ib/ton of melt for sand
screening, again with 90% of the particles greater than 50 /im in diameter.
Thus, the emission factor is 1.0 Ib/ton of melt for particulate escaping to
the atmosphere.
Based on these data from Gutow, total emissions from return sand handling
and screening is estimated to be 2.2 Ib/ton of melt, the total of the above
two values. Since the sand-to-metal ratio is on the average 7':1, this represents
0.3 Ib/ton of sand handled.. - . . •
Based on data from tests on operations at integrated iron and steel plants,
the emission factor equations presented in Figures 3-1 through 3-3 were developed
for particulate emissions from loading of materials onto storage piles and
wind erosion from the piles. The emission factor equations were derived for
particles smaller than 30 ^ra in Stokes diameter. The quality assurance rating
scheme for the emission factors is shown in Figure 3-4.The paragraphs below
present the estimates of particulate emissions from storage and handling of
sand, coke, and scrap that were derived from the equations.
As indicated in Section 2.2.1, sand is usually stored in enclosed bins
and is often transported pneumatically. If closed storage and transport are
utilized, the emissions from the system will be negligible. If conveyors are
used to transfer sand to storage and from storage to the mold and core prepara-
tion area, emissions of particulate less than 30 jim are estimated to be 0.04
Ib/ton of sand transferred for each transfer point.
The emission factor for conveyor transfer of sand to storage was calcu-
lated using the equation for conveyor transfer to storage presented in Figure
3-1. Since the transfer is conducted inside a building, the correction factor
for wind speed was disregarded. The silt content was assumed to be 7% based on
specifications received during a plant visit.—' Sand purchased by foundries is
generally cleaned and dried; hence, a low moisture content of 0.5% was assumed.
38
-------
OPEN DUST SOURCE: Storage Pile Formation by Means of
Translating Conveyor Stacker
QA RATING: B
EF =0.0018
I
Determined by profiling of emissions
from pile stacking of pelletized and
lump iron ore.
where: :EF = suspended parti cu I ate emissions
(Ib/ron of material transferred)
s = silt content of aggregate (%). .
M= moisture content of aggregate (%)
U = mean wind speed (mph )
Figure 3-1. Predictive emission factor equation for storage
pile formations by means of translating
conveyor stacker.
39
-------
OPEN DUST SOURCE: Transfer of Aggregate from Loader to Truck
QA RATING: B
lb/ton
Determined by profiling of emissions
from load-out of crushed steel slag
and crushed limestone.
where: EF = suspended particulate emissions
(Ib/ton of material transferred)
s = silt content of aggregate (%)
M "moisture content of aggregate (%)
.U =mean wind speed (mph)
Y = effective loader capacity (yd )
Figure 3-2. Predictive emission factor equation for transfer of
aggregate from front-end loader to
40
-------
OPEN DUST SOURCE: Wind Erosion from Storage Piles
QA RATING: C
Based on upwind/downwind Estimated factors to
sampling of emissions from correct measured
inactive storage piles of emissions to other
sand and gravel. source conditions.
where: EF = suspended particulate emissions
(Ib/ton of material put through storage cycle)
s = silt content of aggregate (%)
. D = duration of. storage> (.days-)
d = dry days per year
f = percentage of.time wind speed exceeds 12 mph
Figure 3-3. Predictive emission factor equation for wind
erosion from storage piles.^2'
41
-------
QUALITY ASSURANCE RATING SCHEME
A = FORMULATION BASED ON STATISTICALLY REPRESENTATIVE
NUMBER OF ACCURATE FIELD MEASUREMENTS (EMISSIONS.
METEOROLOGY AND PROCESS DATA) SPANNING EXPECTED
PARAMETER RANGES
B = FORMULATION BASED ON LIMITED NUMBER OF ACCURATE
FIELD MEASUREMENTS
C = FORMULATION OR SPECIFIC VALUE BASED ON LIMITED
NUMBER OF MEASUREMENTS OF UNDETERMINED ACCURACY
OR —
EXTRAPOLATION OF B-RATED DATA FROM SIMILAR PROCESSES
D = ESTIMATE MADE BY KNOWLEDGEABLE PERSONNEL
E = ASSUMED VALUE
22/
Figure 3-4. Qua!Ity assurance (QA) rating scheme for emission factors.—
-------
No test data are available on sand handling in iron foundries. The equa-
tion in Figure 3-1 was used to develop an emission factor of 0.04 Ib/ton of
sand per transfer point for particulates less than 30 /zm as described above.
In addition, Vandegrift et al. developed a sand handling emission factor of
0.3 Ib/ton of sand; however, neither the method of determination nor the opera-
tions covered by the factor were identified.—
A diagram of a sand handling system at one foundry is shown in Figure
3-5*-*= It appears that after the shakeout there are a minimum of 10 trans-
fer points plus a separator, aerator, and drum sand cooler. If the equation
from Figure 3-1 is used with the values for silt and moisture content given
earlier, total emissions from the transfer operations of the return sand are
calculated to be 0.4 Ib/ton of sand. Emissions from the screening and cooling
are visually estimated to be equivalent to transfer operations. Hence, total
emissions from the return sand system are estimated to be 1 Ib/ton sand. How-
ever, many of these emissions are controlled by hooding systems, and thus
actual emissions may be much lower.
No data are available on emissions from the handling and storage of foundry
coke, nor are data available on silt content or moisture content of the coke.
It was assumed that the values of 1% silt content and 1% moisture developed
in Reference 22 for metallurgical coke are appropriate for foundry coke. (Based
on the similarity in the production process of these cokes, this assumption
appears to be reasonable.) Using these assumptions, an emission -factor of 0.0014
Ib/ton of coke per conveyor transfer point was calculated using,the equation
from Figure 3-1. Thus, load-in and load-out of storage piles is estimated, to
have an emission factor of 0.003 Ib/ton.
If a high loader is used to transfer coke, the equation from Figure 3-2
can be used to calculate an emission factor of 0.0005 Ib/ton coke dumped. Finally,
if coke is stored in outdoor storage, wind erosion emissions are calculated
to be 0.02 Ib/ton coke using the equation in Figure 3-3. This assumes the average
storage duration to be 60 days and assumes the default values apply for dry
days and percentage of the time that wind erosion exceeds 12 mph. Thus, total
coke storage emissions are at most 0.03 Ib/ton which are negligible compared
to other operations.
The amount of emissions from scrap storage is directly dependent on the
amount of dirt contained in the scrapj for well-cleaned scrap, emissions will
be negligible. If we assumed as worst conditions 5% dirt in the scrap and a
storage time of 60 days, an emission factor of 0.1 Ib/ton scrap can be estimated
for wind erosion from scrap piles. Data are insufficient to estimate emissions
from transfer of scrap.
The emission factors developed above were based on limited data, and several
assumptions used to calculate the factors. Hence, low reliability (QA level
D or E) should be placed on the emission factors for sand, coke, and scrap
storage and handling.
43
-------
Floor
Sand
Hopper
Floor
Shake -
out
200
1
Ton Sand 2
Storage Bin
. Squeezer Line
J
105
• F
^ 1
1
1
. 100 Ion
Capacity
. • Drum
r*.i . r»i • • Muller
Dilution Plow
1
—
160° F
x
»-
^-^
Prepared - -. Sand •»
/ f
¥
—
Dilution Plow
r A.jlj,,— il-
Une Urw
Stakeout Shale
I
L
90° F at
Machines
1
c
^ a 160° F ^ ^
i
3
£ 1
I
Prepared
pout 1 • - • /
"""" | 'III Id
•^—
^
r i
y 171 1 " i
— »•• Shakeout Sand Bell
Sand
1 1
Bucket
Elevator
00 Ton Sand
torage Bin
Automatic
^achifie Line
05° F
Drum Sand
Cooler
(Not Used)
100 Ton Mew
Capacity 150 Ton
Dium Muller
Muller Planned
1
Aerator
Magnetic
Separator
Slorage Bin Temperature
>- . Varies with Ambient:
g ^^> ' 85-90'F AMB. * M5°F Bin (9-11-69)
% 73° f AMB. = 105° F Bin (12-9-69)
j 90° F al . "
2 Machines
c • • '
5
Figure 3-5. Line drawing of Canton Malleable's sand system showing plowoff
points and resultant sand temperatures.—
-------
3.2.2 Melting and Casting
3.2.2.1 Electric Arc Furnace--
Data in the literature indicate that total emissions from electric arc
furnace melting range from 3.0 to 40.0 Ib/ton of charge, with an average emis-
sion factor of 13.8 Ib/ton.— The manufacturer of an efficient close capture
hooding system has indicated that total baghouse catch for emissions from all
stages of an EAF at foundries charging dirty and oily scrap is on the order
of 35 to 40 Ib/ton of charge^!' Based on visual observation, it is claimed
that the hooding system captures most of the emissions from the EAF from charging
through tapping. If it is assumed that the baghouse captures 99+% of the emis-
sions, then the total uncontrolled particulate emissions from that facility
are slightly more than 35 to 40 Ib/ton of metal charged.
The level of fugitive emissions from an EAF is dependent on the capture
effectiveness of the primary system and the cleanliness of the scrap charged.
If no primary control system is used, total emissions are fugitive and are
estimated to be in the range of 4 to 40 Ib/ton of metal charged. If, as is
the typical case, a fourth-hole duct or side draft hood is used .to capture
melting emissions, 5 to 10% of total emissions are estimated to be fugitive.
Thus, an emission factor of 0.5 to 3.0 Ib/ton of charge with an average of
2.0.Ib/ton is estimated for typical EAFs. Again, this is based'on limited data
and is considered to have relatively low reliability." . . .."..
3.2.2.2 Electric Induction Furnaces— . .
Reference 28 indicates that electric induction furnaces have a total emis-
sion factor of 1.5 Ib/ton charge. Since these furnaces generally have no capture
system, the total was assumed to be the effective fugitive emission rate.
3.2.2.3 Cupola Tapping—
Based on visual observations during plant visits, cupola tapping appears
to be a source of fugitive particulate emissions. However, no data are availa-
ble to determine the extent of these emissions.
3.2.2.4 Iron Inoculation—
An engineering estimate of emissions from inoculation can be developed
from mass balances of the magnesium used for inoculation. Data from Reference
29 indicate that the amount of magnesium added to inoculate iron will vary
from 0.12 to 0.30% of the iron treated, or from about 2-4 to 6.0 Ib of mag-
nesium per ton of iron. About 1.3 Ib of magnesium are consumed in reaction
leaving about 1.1 to 4.7 Ib of magnesium per ton of iron. This will react to
form 2 to 8 Ib of MgO, which is emitted to the atmosphere. Since MgO accounts
for only 60 to 80% of the emissions from iron inoculation^:^' the fugitive
emission rate may range from about 2.5 to 13 Ib/ton of iron inoculated.
45
-------
Test data from an inoculating station indicate an emission rate of 3.3
Ib/ton of iron inoculated for _an operation similar to the one described above.—
For this station, a total of 20 to 22 Ib of inoculant, comprised of soda ash,
MgFeSi (10% Mg), and 75% Fe, were used per ton of iron. Only 32% of the emissions
produced were MgO. In this case, it appears that about 70% of the magnesium
was retained in the iron. Huelson— indicates that with current practices
50 to 90% of the magnesium is expected to be retained in the iron.
Based on these data, 0.4 to 2 Ib of MgO will be formed for each ton of
iron treated. Assuming the quantity of other particulate emissions to be constant
at about 2 Ib/ton, an emission rate range of 2.4 to 4.0 Ib/ton with a typical
value of 2*5 Ib/ton is estimated.
3.2.2.5 Pouring and Cooling—
The most significant testing data for foundry fugitive emissions have
been generated for the iron pouring and cooling operations. A series of quasi-
stack emissions tests was conducted on actual pouring operations in an iron
fr.iinH-ry.21/ The data from these tests yielded emissions factors for iron pouring
ranging from 0.55 to 4.5 Ib/ton of metal poured. However, difficulties were
encountered in this test in separating background particulate from emissions.
In another study ll/ quasi-stack tesss were run on both bench-scale and
pilot-scale pouring and cooling operations. The pouring and cooling of a 30-
Ib'cube casting resulted in a total of 54.61 g of particulate, — which gives,
an emission rate.of 8.3 Ib/ton. Based upon concentration profile data, this
has been separated into 4.0 Ib/ton for pouring and 4.3 Ib/ton for cooling.—
Data on concentrations of organic gases evolved during pouring and cooling
are also presented. However, data are insufficient to determine emission factors.
9/
Gutow~ has also developed emission factors for iron pouring and cooling.
The emission factor given for pouring is 5.10 Ib/ton of melt with 60% of the
particles greater than 50 /urn. Hence, the factor for emissions escaping to the
atmosphere is 2.0 Ib/ton of melt. The cooling emission factor is 10.30 Ib/ton
of melt with 90% of the particles being greater than 50 /zm. The emission which
escape to the atmosphere are 1.0 Ib/ton.
3.2.3 Product Finishing
The only major sources of fugitive emissions in the product finishing
area are shakeout and grinding. As a part of the testing discussed earlier,
Bates and Scott used the quasi-stack to develop an emission factor of 3.15
Ib/ton of cast iron with 98% of the mass less than 15 ^m in diameter.^:' Gutow
estimated an emission factor of 32.20 Ib/ton melt with 90% of the particles
being greater than 50 /urn in diameter. Hence, the emission rate is 3.2 Ib/ton
for those particles escaping to the atmosphere.
46
-------
Data are insufficient to determine an emission factor for grinding.
3.2.4 Core and MoId-Making
9/
The only data available for core and mold-making are presented by Gutow.""
He indicates that mulling has an emission rate of 20.60 Ib/ton of melt with
90% of the particles greater than 50 /urn. It is assumed that particles larger
than 50 /im will not reach the atmosphere. Thus, emission rate is 2.1 Ib/ton
melt. This factor seems high but no other data are available.
According to Gutow, molding has an emission rate of 0.50 Ib/ton melt with
90% having diameter greater than 50 /xm»^ Hence, the effective emission rate
(those particles less than 50 /ira) is 0.05 Ib/ton melt.
Based on observations during plant visits and conversations with indus-
try personnel, particulate emissions from shell-core and mold and cold set
processes appear to be negligible. Data are insufficient to determine emis-
sions from core washing.
3.2.5 Waste Handling
Since no data are available on emissions from waste handling in the foun-
dry industry,, the following estimates were developed based on equations in
Ref. 22. Based on a silt content of 7% (see Section 3.2.1) and an average,.stor-
age period of 90 days before covering, sand storage emissions are estimated.
to be 0.24 Ib/ton. Based on the assumption that foundry.slag storage emissions
are comparable to iron and steel slag storage, an emission factor of 0.18 Ib/ton
slag stored*-^ is estimated. Data are insufficient to estimate emissions from
slag quenching.
3.3 INVENTORY OF IRON FOUNDRY EMISSIONS
The data on production rates presented in Section 2.3 and the best emis-
sion factors from Section 3.2 were used to generate an inventory of fugitive
emissions from iron foundries. This inventory is presented in Table 3-2. The
third column presents the total annual particulate emissions from each source.
It should be noted that some of the values are for particulate of less than
30 or less than 50 Mm in diameter and some are for.TSP. This is a result of
the variation in the reporting practice in the literature and the fact that
particle size data are insufficient to translate the values to a common basis.
In addition, estimates of the fine particle content (particles less than
5 ra in diameter) were estimated and a fine particle inventory developed.
These data are presented in the last two columns of Table 3-2.
It is again stressed that the emission factors used in this table are
.based on very limited data and as such have a low reliability. Consequently
this emission inventory should be applied with caution.
47
-------
00
Dnlpsi.on Source
Sl.or.ipn .
Coke
Scrip
ll;in'llinf» -inil Tr.insfer
Spenl S.ni'l
<:olte
II- 1 ling iii'l i:.ir-l. Inp.
Cnpol.l T.ln-lllE
KAF Cli-Mp.'np. \
T.lpplllP. \
(.e.Tix.ipc- )
Iti'ltM1 1 1 on Fnrll-ir,"
Iiioctll.it 1 on
Pouring
Cool lite
Finl slij up.
Sli.ikooi.it
GrlnriliiR
Mold .ind Core Prepar.il Ion
Mnl lint:
Slinl t or Hot. nox
llc.il Inp
llol.llnp, P.illel
ColJsei
Cnre W.l-,h
Kilning
Waste. HamJHnR
StlH IJlielt.'ll
Warte S.intl Transfer
Sl.or.ig"
S.ind
Slag
Reproduced from i^^^
best available copy. ^S^
TADLF. 1-2. FIICITIVE PART1CIIMTF. F.MISSIONS HIVKNTORV
AniiM.il Fliir- Fin.- P.nu.Jr|p
Process Rile. Kinl.isloiir. P.irllr.le rml ssl oiin
F.PllnwtPtl l)i-lsr.l"ii K.icior (t.on/yr) (ll./yr) (X T HI > (II. /yr)
M'-g - - - -
O.I Ib/ton .=cr.i|. A.O x IO7 2.0 ,: Id6 JO2' f,.i| „ KI'
N.'u - - -
2.2 Ib/ton meU . 2.7 x I"7 -i.i x l.o7. Id'-' l.n •• |o'
d.2 lli/lnn c-.ok,. 1.1 x III' 2.6 x Id' I03 7.11 ,, lip''
Ho ll.iL.i -
2.0 Ib/t.-n cli'i-R.' 'i.° x IO6 <>.l| x IO6 Hl»-^ 7.M x 10*
' ft ft It ' ft
1.6 Ib/ton cli.ir,;.- l.o x lir 2.fl x III BlVf. 2.' x 1"
2.5 Ih/ton lnncnl.il.pil 1.7 x Id' 4.2 x in" 8O^ • 7.'i \ Id'
/..O Ib/lon puirnil 2.2 x Id' 8.11 x In Q-^i fl.'i x Id'
'i.2 lb/|oii pnurpil 2.2 x 10 1.5 x 1(1 Qr— O.O x Id'
7 7 Jl' 7
J.I 5 Ib/lon c.ist 2.2 x Id ft. 1.1 x Id 3(>— !.'• x 10
No ll.it.-i ' . -
7 7 -I/ J
2.1 Hi/ton moll : 2.7 x IO '5.7 x III |l>- 1.; x III
Ho Dal.l 'I.'.. - -
No »!t.1 -
Ho D.it.i . -
No |,.1t;, - - - II -
d.05 lb/l:on m.-ll 2.7 x III l.'i x 111 llr- , 1,.}. s III
No Ililrn - - f - i - ,
0.1 Ib/lon s.111-1 /i. 7 x III 1. 7 x in' ll^ ',.! x III'
fi n a' •">
0.2'i I. !>/ Inn Sfinrl •",.; .x KK l.'i x I0r llr" 'i.2 x Id,
O.IB Ib/lon slap. I.5 x IO 2.7 x I0' VI2 o.l x Id'
_a/ Est im;ttntl b-isrtl on prev! nits m.itrri.'i! s Itandl J.MR mni ssInns
J>/ Ansinnrd same .TS nlfctrlc -ire fnrn.-icr.
-------
SECTION 4.0
FUGITIVE EMISSIONS CONTROL
As a result of the internal plant environmental problems created by fugi-
tive emissions sources in iron foundries, effective control methods have been
developed for many of these sources. These control methods may be separated
into the following three basic components:
1. Preventive process and operating changes.
2. Capture methods for containment of the fugitive emissions stream.
3. Devices for removal of pollutants from the captured emission streams.
Preventive process and operating changes act to control fugitive emissions
:,by either eliminating, or,.reducing emissions at the source. These preventive •
measures may consist of minor changes in operating procedures, such as wetting
of dusts oh storage piles, or better monitoring of input materials to ensure
more consistent feed properties. Increased maintenance and more, efficient use
of existing equipment can also.lead to decreased emissions. Finally, basic re-
design of equipment or processes may be practicable, especially in new or re-
modeled plants.
When preventive measures are not practicable, fugitive emissions must
first be captured by a hood and ducting system or by containment within a
closed building or special enclosure with a venting system. Three types of
hoods may be considered: (a) fixed, standard-type hoods and ducts (such as
standard laboratory hoods), (b) portable hoods with flexible ducts (such as
those used currently in some machining and pilot-plant processing areas), and
(c) custom hoods with moving closures (such as those used on copper converters).
The removal devices for fugitive emissions are .similar to those used for
primary confined source emissions. These devices must be used in conjunction
with a capture or collection method and blower to capture and force the emis-
sion stream through the removal device. Thus, a total fugitive emission control
system must include capture and collection methods and one or more removal de-
vices. For example, particulate emissions from a high-temperature fugitive
source might be captured by a hood and duct with a blower, the large particu-
lates removed by a cyclone separator, the carrier gases cooled in an exhaust
49
-------
gas cooler pipe, and the fine particulates removed by a baghouse. The optimum
choices of removal devices depend on the types of emissions, the size distri-
bution of the particulate emissions and the properties of the carrier gases.
For some sources, the fugitive emissions may be ducted directly back into
the emission control system for the primary source. However, if large amounts
of air are required to capture and collect the fugitive emission in a hood or
enclosure, then the concentration of emissions (gases or particulate) may be
much lower than for the primary source. This decrease in concentration must be
considered in evaluation and selection of potential removal devices^
As indicated earlier, control systems have been developed for many of the
major sources in the foundry industry. A summary of possible control systems
for these sources is presented in Table 4-1. Also included in Table 4-1 are
estimates of the effectiveness of the various abatement, capture and removal
methods and an indication of some problems associated with the application of
some of the methods.
To the extent that source operations vary from plant to plant, it is un-
likely that a single control option would be most suitable for uniform appli-
cation throughout the industry. Added to this is the need for determining the
degree to which individual fugitive sources at a given plant are to be con-
trolled in order to meet plant-specific control strategy objectives. The most
cost-effective control strategy, for a particular plant entails the application
of the most efficient controls to the largest contributing sources. .
In the sections below, control- system options are presented for the fol-
lowing fugitive source categories:
1. Raw material input
. Coke and scrap piles
. New sand handling
. Coke handling
. Spent sand handling
2. Melting and casting
. Cupola tapping
. Electric arc furnaces (leakage, charging, and tapping)
. Electric induction furnaces
. Iron inoculation
. Pouring and cooling
3. Finishing
. Shakeout
. Grinding
50
-------
TABLE 4-1. FUGITIVE EMISSIONS CONTROL TECHNOLOGY SUMMARY
Fugitive emissions
source
Haw material input
Storage
'Coke
Sand
Scrap
Handling and transfer
Now sand handling
'
Coke handling
Spent sand handling
He It Ing and casting
Cupola tapping
Electric arc furnace
Induction furnace
Iron inoculation
Iron pouring
Floor pouring
Fuui ing station
Iron cooling
Floor pouring
Pouring station
Capture or abatement system
Hut hod £f tectivMiess Problems
Enclosed storage
Pneumatic transfer
Covered belts and
enclosed transfer
Hooded screens
'
and covered belts
Covered belts and
enclosed transfer
Schumacher process
Stationary hood
Hoveablc hood
Canopy hood
losed clar in
c ,arg ng system
Hooded charging bucket
Furnace enclosure
Close capture hooding
Close capture hooding
Tapping hood
Booth
Building evacuation
Mobile vent
Touring hood
Building evacuation
Mold funnel
991
Good
Good
Good
Good
99% ^
ni
01
01
1)1
DI
DI
60-851
90-95%
01
01
01
01
Good
01
Good
None ,-
. None •. . - -
" None
• ", Mine
None
None
Some. capital investment
Interference with operations
Capture problems with
cross draft, high flows
docs not control tapping
Does not control tapping
Interference with operation
. None - '^
Mine
May interfere with
melting operations or
impossible to Interface
May be safety hazard
High cost
Quest Innahle effectiveness
None
High cost
None
Removal system
Method Effectiveness
Wet scrubber
Baghouse
Wet scrubber
Baghouse
Baghouse
Wet scrubber
•
Primary capture sysrem
Primary capture system
Baghouse
Baghouse
M-iin melting system
Baghouse
Baghouse
Baghouse
Main melting system
B
-------
TABLE 4-1. (continued)
Ui
Fugitive emissions
source
Product finishing
Vibrating shakeout
Revolving shakeout
Grinding
Core and mold
preparation
Hulling
Shellcore or hot box
Heating
Holding pallet
Cold set
Core wash
Molding
Waste handling
Slag quench
Waste sand transfer
Storage piles
Capture or abatement system
He thud
Total enclosure
Side draft hood
Enclosure
Swing grinder booth
Oowndraft table
Hooded charging
Overhead hood
Movcable hood
Closed system
Spray booth
None needed
Dl
Welting
Wetting
Effectiveness ' Problems
Good None
Moderate None .
1)1 None '
ni None
Dl Size limitations
Cood None
Dl May be Ineffective
ni
ni
Dl M.iy be a safety hazard
.during torching
• • -
-
90-957. " -
Dl
Method
Wet scrubber
Wet scrubber
Wet scrubber
Wet scrubber
Baghouse
Wet scrubhrr
Wet scrubber
Chemical scrubber
Chemical scrubber
Chemical scrubber
Dl
-
Dl
-
Removal system
Effectiveness Problems
98-997. None
9B-9VZ None
96-997. None
997. 4 None
991 4 None
997. 4 None
99Z 4 Noil"
90-1007. Hone
90-100Z None
90- lOOt Hone
-
-
-
-
Notel Dl = Data Insufficient.
-------
4. Core and mold preparation
. Mulling
. Shell or hot box heating
. Shell or hot box cooling
. Cold set preparation
. Core wash
. Molding
5* Waste material handling
. Slag quench
. Waste sand transfer
. Slag and sand storage
For each of the above source categories, control options (including both
emissions capture methods and pollutant removal methods) are presented. Process
changes which act to limit or eliminate emissions are also considered as con-
trol options. For each control method, equipment operating parameters, expected
level of performance, associated operational problems, and estimated capital
and operating costs are given when available.
4.1 RAW MATERIALS INPUT
This section describes the preventive measures and control systems that
can be used to limit fugitive particulate emissions from raw material handling
and preparation. Those sources examined include coke and scrap storage piles,
new sand handling, spent sand handling, sand reclaimers, and coke handling.
With the exception of some storage operations, most of these sources are cur-
rently controlled within the foundry industry.
4.1.1 Material Storage
As indicated in Section 2.2.1, the only possibly significant emissions
from material storage occur in the outdoor storage of coke and scrap. No data
are available on measures which can reduce emissions from outdoor storage with-
out degradation of materials. It appears that the only practicable measure to
control storage emissions is the use of an enclosed storage area. No data are
available on cost of enclosures. However, many foundries now have covered stor-
age to avoid degradation of coke and scrap.
4.1.2 Materials Handling
Fugitive emissions problems from handling of raw materials have been di-
vided into three basic areas: (a) new sand handling and storage; (b) coke
handling; and (c) spent sand handling and reclamation. Controls for each of
these areas are described in the following subsections.
53
-------
4.1.2.1 New Sand Handling —
As new sand is received at the foundry, it is transferred to storage bins
by pneumatic feed or mechanical conveyor. The pneumatic system is totally en-
closed and in effect is well-controlled. However, with mechanical systems, fur-
ther controls are necessary.
In most foundries visited by MRI personnel, sand handling conveyors ap-
peared to be adequately controlled. Reference 37 suggests that the following
ventilation system is adequate to control dust emissions,
At transfer points with less than a 3-ft drop, the transfer point should
be enclosed and air should be exhausted from the top of the enclosure at the
rate of 350 cfm/ft of belt width for belt speeds less than 200 fpm. An exhaust
rate of 500 cfm/ft of belt width should be used for belts with speeds in excess
of 200 fpm. If the drop is greater than 3 ft, an additional exhaust should be
used at the lower level with a flow rate of 700 cfm for 12- to 36- in. belts and
1,000 cfm for belts wider than 36 in.
Belts should also be covered between transfer points with additional ex-
haust points at 30-ft intervals. Exhaust rates of 350 cfm/ft of belt width are
sufficient to control dust emissions between transfer points.
Huelsen — has indicated that wet scrubbers (6- to 10-in. pressure drop)
are normally used to remove particulate from sand handling ventilation systems.
In addition, fabric filters for removal were observed at several plants visited
by MRI. These systems appeared totally effective in capturing and removing par-
ticulates at the plants visited by MRI.' Reference 36 indicates that particulate
removal efficiency for the scrubber is 99^7,*
4.1.2.2 Coke Handling-
Coke is generally transferred manually or mechanically within the foundry.
Little can be done to control the negligible emissions from manual handling.
With mechanical conveyors the systems described in Section 4.1.2.1 should be
effective in controlling emissions.
At some foundries coke is screened before going to the cupola to eliminate
fines which may have been created. The following control measures are recommended
for use with screening devices.
Fugitive dust emissions from vibrating screens can generally be captured
in an overhead hood. Particulate can then be removed from the emissions stream
by either a scrubber (6- to 10-in. pressure drop) for humid or dry emissions
streams or by baghouses for dry
A typical hooding system for a vibrating screen is shown in Figure 4-1.
Reference 37 suggests that a flow rate of 70 cfm/sq ft of screen area is needed
for adequate capture emissions. It has been estimated by MRI that capture ef-
ficiencies of 75 to 90% can be attained with this system.
54
-------
To Scrubber or
Fabric Filter
Vibrating
Screen
18" Minimum
I
37/
Figure 4-1. Hooding system for a vibrating screen.—
55
-------
For control of fugitive dust emissions from rotating drum screening opera-
tions, hoods enclosing the drum as completely as possible should be utilized.
A typical system is shown in Figure 4-2. Reference 37 suggests that a flow rate
of 125 cfm/sq ft of cross-sectional area of the screen should be used. In cases
where removal of large quantities of fines is necessary, 150 cfm/sq ft should
be used. The particulate removal devices for these systems are the same as those
described for vibrating screens and removal efficiencies are of the same magni-
tude.
4.1.2.3 Spent Sand Handling--
After the sand leaves the shakeout, it is dry and has potential for high
levels of fugitive emissions during transfer and reclamation. If mechanical
conveyors are used,it is essential that the controls described in Section
4.1.2.1 be utilized. No data were obtained which described systems for control-
ling emissions from sand reclaimers. However, based upon observations during
plant visits it appears that a system such as that described for vibrating con-
veyors will adequately control reclaimer emissions.
A patented concept (U.S. Patent No. 3,461,941) has been developed which
has the potential for control of fugitive dust emissions from most sand handling
operations other than shakeout. The process is called the Schumacher Sand Pro-
cess System. The normal sand-to-metal ratio in a green sand foundry is between
5 and 7:1. The Schumacher process utilizes a sand processed to metal ratio of
20:1. This is the quantity of sand put through the muller. However, the extra
•sand is not utilized to produce molds. It is diverted to an inundator. Here
the hot dry sand taken off the shakeout is mixed with the moist sand from the
muller to produce a moist cool sand. This sand is then taken through the nor-
mal sand handling processes. However, the now moist sand presents no emissions
problems. Tests near transfer stations indicate that dust concentrations are
reduced by as much as 99% by application of the system.—'
The system requires little additional equipment (the inundator and a small
amount of additional mulling equipment) and is estimated to cost substantially
less than equivalent collection systems. The system is claimed to have the ad-
ditional advantages of saving binder loss and producing cooler sand for the
molding line.
4.2 MELTING AND CASTING
The operations that occur in an iron foundry from the time scrap is charged
into a furnace for melting until the time the casting is to be removed from
the mold constitute the greatest fugitive emission sources for which generally
applicable control measures have not been found. The primary control problems
from melting are (a) cupola tapping, (b) electric arc furnace charging, tapping,
and leakage, and (c) induction furnace charging. Other major emission sources
in this area include (a) inoculation of ductile iron, (b) pouring hot metal
into molds, and (c) cooling the filled molds before shakeout. Possible fugi-
tive emissions control methods for these sources are .discussed below.
56
-------
To Scrubber or
Fabric Filter
Rotating
Screen
Hood
37/
Figure 4-2. Hooding system for revolving screen.—
57
-------
4.2.1 Cupola Tapping
In general, the tapping of hot metal from any melting furnace, and in par-
ticular from a cupola, appears to be one of the lesser sources of fugitive emis-
sions in the melting and casting areas. However, control of these emissions may
be necessary on new plants located in regions with high ambient particulate lev-
els, or in plants melting scrap containing significant amounts of hazardous
metals.
Limited data are available on capture methods for particulate emissions
from cupola tapping. A local exhaust hood such as that shown in Figure 4-3 is
suggested in Reference 39 for cupolas having toxic fumes„ A velocity of 150
ft/min through the hood opening is suggested. If this type of permanent sys-
tem is utilized, it will be necessary to use a rail arrangement to move the
ladle to the cupola's spout. It is suggested that new systems be designed in
such a way that this system can be vented to the removal device used for pri-
mary control of cupola emissions.
When a permanent system described above is not feasible, a movable sys-
tem such as that shown in Figure 4-4 is suggested,, In this system, a telescop-
ing duct with a funnel-type hood is suspended near the cupola. An additional
hood, which is joined with the funnel hood on the duct, is attached to the
ladle. This system has the advantage that the ladle can remain attached to the
crane during tapping, which is more efficient from a standpoint of time,, The
disadvantage is that the hood is located a' greater distance from the ladle,
making it more difficult to capture .emissions. It is suggested that these emis-
sions be vented to the primary removal device,, It should be noted that particu-
late emissions from tapping may be in the submicron size range,, If this is the
case, the primary removal device would possibly not be effective and alterna-
tive control measures may be necessary.
No data are available on flow rates necessary for efficient performance
of the movable hood system. Cost data are not available on either the fixed or
the movable system,,
4.2.2 Electric Arc Furnaces
The use of electric arc furnaces (EAFs) for scrap melting is expanding
rapidly, especially in high production foundries. The most serious fugitive
emissions problem associated with EAFs is the charging of scrap, particularly
dirty or oily scrap. Thus, the methods chosen for fugitive emissions capture
should first address the charging problem. The system used for tapping and
leakage emissions should then be interfaced with the charging and primary
melting emissions capture systems.
58
-------
Charging
Door
To Wet
Scrubber
39/
Figure 4-3. Fixed hood for cupola tapping.—
59
-------
Exhaust Duct
Telescopic Section
of Duct with
Attached Funnel
Hood Suspended
from Lad le
I— Forehearth
397
Figure 4-4. Movable hood for cupola tapping.—
60
-------
Over the past several years many attempts have been made to design sys-
tems which can adequately control EAF charging emissions. The following five
general types of systems can be utilized to capture charging emissions:
1. Canopy hooding
2. Closed charging systems
3. Hooding charging buckets
4. Total furnace enclosures
. 5. Close hooding
The type of system chosen for a particular EAF may vary depending on such
parameters as building design, size of furnace, necessary degree of control
and whether the system is for new or existing furnaces. The closed charging
system and the close hooding system are discussed below. The other systems are
discussed fully in Reference 22.
4.2.2.1 Closed Charging System—
One conceptual method for control of EAF charging emissions which has not
been applied commercially utilizes an automated charging system that does not
.require removal of the furnace hood. This system uses the primary melting emis-
sions control system as a capture device, eliminating the need for another cap-
. turing system. However, a major disadvantage of the system is that it requires
sized scrap with diameter less than 3 in., which may increase system cost ex-
cessively. Since no systems of this type are currently operating, data on cost
and effectiveness of the system are not available.
4.2.2.2 Close Capture Hooding System—
Hawley Manufacturing Corporation has patented a close capture hooding
system for electric arc furnaces which controls emissions during charging,
melting, slagging, and tapping operations. The system (shown in Figure 4-5)
uses a large plenum (mixing chamber) connected to four separate hoods and a
removal device by movable ductwork to allow continuous exhaust during all
phases of furnace operation. The four hoods are positioned (a) around the
electrodes for control during meltdown, (b) over the tapping spout, (c)
around the slag door for control during slagging or oxygen lancing, and (d)
suspended off the side of a movable hood for control during charging. Each
of the hoods is automatically controlled to allow ventilation to be directed
toward the area of greatest emissions. Typically, 30 to 35% of the total flow
is to the slag hood during all phases of the operation. During charging and
meltdown the other 70 to 757, of the flow is directed to the charging hood and
electrode hood, respectively. During tapping 25% -of the flow is directed to
the electrode hood and the remaining 45 to 50% to the tapping
61
-------
Figure 4-5. Close capture hooding system for
electric arc furnaces.
62
-------
The system, called the TOT-L-VENT system, has been installed on EAFs rang-
ing in size from 6.0 to 12.0 ft in diameter. Furnaces in this size require air
volumes ranging from 20,000 to 35,000 acfm,^P/ xhe parameters affecting the re-
quired flow are: (a) furnace shell diameter, (b) transformer rating, (c) nomi-
nal heat size, (d) oil content in charge, (e) charge composition, (f) rate of
oxygen lance, (g) size of charge bucket, and (h) size of ladle.—' The cost for
a capture system of this size range is on the order of $20,000 to $30,000. This
cost includes all engineering design, manufacturing, electrical control panel
and installation of the capture system. It does not include removal device
costs.
Capture efficiencies on the order of 90 to 95% for tapping, 60 to 90% dur-
ing charging and 99% (no visible emissions) during melting can be expected from
the system. Normally capture efficiency is on the order of 80 to 90%. However,
on some systems design difficulties and charging of extremely dirty scrap have
led to lower capture efficiencies«-t!i'
The removal system suggested by the manufacturer is a shaker-type fabric
filter with an air-to-cloth ratio of 2:1 and no more than 3:1. The pressure
drop across the system is 14 in,, 6 in. in the hooding device and the remainder
in the baghouse and blower.—
Similar systems were suggested by other manufacturers but no specific data
were obtained.
4.2.3 Electric Induction Furnace
Electric induction furnaces have fewer emissions than other types of melt-
ing equipment. Hence, in the past induction furnaces have often had no pollution
control. A close capture hooding system such as the one shown in Figure 4-6 is
suggested for control of induction furnace emissions during charging, tapping,
and melting. This close capture system is patented by Hawley Manufacturing Com-
pany. Other manufacturers suggested that they do supply similar hoods on a cus-
tom basis. However, no specific data were provided.
The hood is built into the furnace platform and has a telescopic duct to
allow the hood to move with the furnace and to maintain flow during tapping.
The top of the hood and one side swivel away from the furnace to allow access
to the furnace during charging. The hood is normally used with a system of hoods
for two or three furnaces attached to a common removal device* This allows the
user to take advantage of the different flows required for charging and melting.
Required flow rates for various size furnaces aret-i2'
1. Less than 15. tons 7,500 cfm for charging
2,000 cfm for melting
2. 15 tons to 50 tons 20,000 cfm for charging
5,000 cfm for melting
63
-------
Figure 4-6, Close capture hooding system for
electric induction furnace.
64
-------
3. 40 tons to 50 tons 25,000 cfm for charging
5,000 cfm for melting
One plant engineer indicated that no visible emissions had been observed during
either charging or melting on three 5-ton induction furnaces.^'
The suggested removal device is a shaker-type fabric filter with an air-
to-cloth ratio of 2.1:1 and not more than 3:1. The pressure drop across the
hood is about 4 in. of water with a 6-in. drop across the baghouse.
4.2.4 Iron Inoculation
Approximately 157<> of the iron castings produced in the United States util-
ize ductile iron. Generally, ductile iron is produced by the addition of mag-
nesium or other alloying substances to molten iron after it has been tapped
into the ladle. The primary methods of capture for iron inoculation emissions
are (a) utilization of the furnace tapping control system, and (b) use of sep-
arate enclosures for iron inoculation. If operations permit, the most efficient
method to control iron inoculation is to add the alloying agents while the
ladle is still in place from tapping. In this instance, the tapping control
system can be used for control of iron inoculation.
If furnace operations do not permit inoculation at the tapping station,
.a. separate enclosure, such as that shown in Figure 4-7 is suggested. Air cur-
tains or door at the roof opening for the crane may be utilized to prevent es-
:cape, of the buoyant emissions. A velocity of 100 .to 125 fpm should be maintained
at each opening. A fabric filter is suggested for pollutant removal.
The authors are unaware of any commercial system such as that described
above. Hence, no substantive data on cost or efficiency of the system are avail-
able. •
As described in Section 2.2.2.4, emissions from iron inoculation are de-
pendent upon the process chosen for inoculation. By the application of tech-
niques which yield high magnesium capture efficiency, it is possible that emis-
sions may be reduced to the extent that add-on control measures are unnecessary.
However, data are not sufficient to substantiate this view.
4.2.5 Iron Pouring and Cooling
The mold pouring and cooling floor has been one of the dirtiest and most
difficult areas to control in the iron foundry. Much of the difficulty in de-
veloping economically feasible controls has arisen from the variation of pour-
ing methods from foundry to foundry and from the large areas over which emis-
sions occur in a particular foundry. In fact, Huelsen—has indicated that
some foundries cannot absorb the cost of controlling these sources. While this
.may be true for small foundries, several alternatives appear to.be feasible
for larger foundries, especially new foundries. •
65
-------
Crane
Air Curtain
To Fabric
Filter
Figure 4-7. Enclosure for ladle inoculation.
66
-------
Three types of control measures appear to be technically viable. For new
or redesigned foundries which produce many copies of the same castings, "perma-
nent" or reusable metal or graphite molds can be used in place of green sand
molds to reduce emissions, A second alternative i"s the use of a stationary
hooded pouring station in conjunction with an enclosed cooling conveyor. Fi-
nally, for those foundries in which the mold is placed on the floor and the
ladle is moved to the mold, some type of movable hooding system or building "
evacuation will be required. Each of these systems is discussed in the follow-
ing paragraphs.
4.2.5.1 Permanent Mold Casting--
The standard process for production of gray iron castings has utilized
green sand molds with sand cores bound by organic binders. For many years "per-
manent" or reusable molds have been used to produce small cast iron parts. How-
ever, recent developments have made it possible to extend the process to other
high volume castings.—
The permanent mold-casting technique uses reusable molds of iron, steel
or graphite which are held together by a machine. The mold is coated with an
insulating material and cores are set into place. After the hot metal is poured
into the mold and allowed to solidify, the mold is opened. The maximum time
from the beginning of a pour until the mold is released is about 3 min.
- Tests, of emissions from.13-Ib cast iron blocks produced from a permanent
mold and a green sand mold, were reported in Reference 43. The results .of- these
tests are given in Table 4-2. The permanent mold technique resulted in a 99%
•reduction in particulate emissions and a 99% reduction in hydrocarbon emissions.
If capture of the remaining pollutant is deemed necessary, the stationary mold
machines are relatively easy to hood. The technique has the additional advantage
of reducing emissions in the sand handling and core-making areas and virtually
eliminating shakeout problems.
The cost analysis in Reference 43 indicated that production costs would
be equal to or slightly less than costs incurred with conventional high volume
molding processes. Cost reductions were estimated to be $0.005 to $0.015/lb'
for gray iron and $0.02/lb for ductile iron. This did not include adjustment
for the reduced capital cost obtained from elimination of much of the sand
handling and storage equipment. Hence, for a new high volume foundry, permanent
mold-casting methods appear to be a viable alternative to sand mold casting.
It should be noted that this method can be economically applied only in
those foundries producing adequate volume of identical castings. Reference 44
suggests that a minimum of 2,000 castings per mold is required to make this
system competitive with green sand molding. In addition, this control method
is more appropriate for new or significantly modified foundries than for found-
ries currently in operation with adequate process equipment. Finally, it has
been suggested that there may be size limitation in the use of the process.
67
-------
TABLE 4-2. COMPARISON OF EMISSIONS FROM GREEN SAND AND PERMANENT
MOLD PROCESSES FOR PRODUCING A 13-LB UNCORED
CASTING UNDER VENTILATED CONDITIONS^/
Green sand
(S:M ratio = 7:1)
Permanent mold
Time of emissions
Dust loading
Calculated total weight of
particulate evolved on a
one-casting basis
Maximum hydrocarbon concentration
Average hydrocarbon concentration
Maximum carbon monoxide concentration
Average, carbon monoxide concentration
1 hr
0.04052 gr/scf
5.5 g
1,800 ppm
460 ppm
1,350 ppm
250 ppm
3 min
0.01017 gr/scf
0.15 g
125 ppm
100 ppm
100 ppm
> 50 ppm
68
-------
4.2.5.2 Mold Pouring Hood/Conveyor System—
If green sand molds are used, the best method of control consists of sta-
tionary hooded pouring stations and covered conveyors for mold cooling. Ref-
erence 39 suggests that the hot metal be placed in the ladle and then be cov-
ered with a steel lid to limit emissions during transport. The ladle is then
transported to a hooded pouring area such as that shown in Figure 4-8. The
molds are moved through the pouring area on a conveyor and, upon leaving the
pouring area, are transported through a mold tunnel such as that shown in Fig-
ure 4-9 to shakeout. Commercially available models of these systems are de-
scribed in References 45 and 46.
Reference 37 suggests a flow rate of about 150 to 175 cfm/linear foot of
hood with slot velocities of 1,500 fpm for the pouring hood. Exhaust takeoffs
every 8 to 10 ft are recommended. The enclosed smoke hood for the conveyor will
require about 75 to 100 cfm/linear foot of hooding with a minimum flow of 200
fpm through all openings. Exhaust takeoffs should be located on approximately
60-ft centers.
One manufacturer indicates that the velocity through control areas for
the pouring hood will generally be in the range of 150 to 200 ft/min.—- The
system has an air supply system to properly distribute flows across open areas.
Hood lengths range from 10 to 200 ft. Most pouring hoods are 50 to 70 ft long.
Exhaust connections to the plenum are on approximately 20-ft centers and supply
connections are usually midway between the exhaust connections.—
Significant numbers of the installation described above have been installed
in the United States. The installed cost of these systems in 1977 range from
about 60 to 800/cfm of exhausted air. Although no data on capture effectiveness
are available, manufacturers representatives indicate that visually, capture ap-
pears essentially complete.
No data have been obtained on the type of removal devices to be used for
pouring and cooling. However, the moist nature of the emissions stream seems
to indicate that wet scrubbers (probably high pressure drop) would be the most
suitable type of system. However, information from industry personnel indicates
that scrubbers are not sufficient to control these emissions. A secondary
scrubber such as that described in Section 4.4 may be necessary to control the
hydrocarbon emissions generated during pouring and cooling. No data are avail-
able on the efficiency and cost of this system.
4.2.5.3 Portable Exhaust Hoods--
For those pouring operations where a stationary pouring area is not feas-
ible, the most efficient solution appears to be a portable exhaust hood attached
to the pouring ladle (see Figure 4-10). Reference 39 suggests that this exhaust
system can be used with either monorail or crane and can capture emissions with
a flow of 1,500 cfm/ladle. A scrubber (probabLy medium or high pressure drop)
is suggested as the most appropriate removal device, possibly with a secondary
device to remove hydrocarbon emissions as described in Section 4.4.
69
-------
Ladle
46/
Figure 4-8. Hooded pouring station.—
70
-------
46/
Figure 4-9. Mold tunnel.—
71
-------
Crane
Ladle
To Wet
Scrubber
Flexible
Hose
Hood
Mold
Figure 4-10. Movable pouring hood.
72
-------
The primary disadvantage of this system is that it provides for no control
of emissions during cooling. There appears to be no effective capture system
other than building evacuation for cooling emissions if stationary molds and
movable ladles are used. No data on operating parameters or costs are available
for building evacuation.
4.3 PRODUCT FINISHING
Castings can be removed from a sand molding: (a) mechanically removing
the casting from the mold with a fork or payloader, (b) vibrating shakeout,
or (c) revolving shakeout. The latter two are of primary interest from a con-
trol standpoint. Grinding may also be significant in a fugitive emissions.
control problem.
4.3.1 Casting Shakeout
In the foundry, removal of the casting from the sand mold has traditionally
been accomplished by placing the mold and casting on a heavy vibrating screen.
This causes the sand to be shaken from the casting and dropped onto a conveyor
where it is returned for reuse. Recently, however, a revolving shakeout system
has been developed which may lead to easier control of dust emissions. The fol-
lowing subsections describe possible control measures for each of these sources.
4.3.1.1 Vibrating:. Shakeout— . ... , -
• As described above, the shakeout. operation in foundries has generally con-
sisted of a vibrating screen which shakes the sand from the casting. The mold
is generally placed on the shakeout by either an overhead crane, manual place-
ment or by conveyor. The capture method used is dependent on the operation.
The three most common methods are a type of side draft, double side draft (or
push and pull), overhead hood (generally used with crane or manual placement)
or a complete tunnel enclosure of the shakeout (generally used with conveyor
systems).
Reference 15 describes the side draft hood for shakeout shown in Figure
4-11. For shakeouts less than 6 ft wide, a flow rate of 500 cfm/sq ft of shake-
out grate is recommended. It is also suggested that sufficient air be exhausted
from the shakeout hopper to provide a downdraft of 40 ft/min through the grate.
For shakeouts greater than 6 ft in width, hoods should be used on any two adja-
cent sides with a flow rate of 500 cfm/sq ft of grate area. It is suggested
that.flow rates need to be increased if (a) castings are quite hot, (b) sand-
to-metal ratio is low, or (c) cross drafts are high.^' However, no data are
provided as to effectiveness of higher flow rates under the above conditions.
A wet scrubber (1- to 10-in. pressure drop) is suggested as the best particulate
removal device for this system, due to the moisture level of gases from shake-
out. Reference 48 indicates that 90% effectiveness can be attained with proper
side draft hooding. If we assume a grain loading of 0.5 to 1.0 gr/cu ft—and
attain the outlet loading of 0.01 gr/scf suggested by Reference 35, this yields
73
-------
To Wet
Scrubber
157
Figure 4-11. Side draft hooding for vibrating shakeout.-^
74
-------
a scrubber efficiency of 98 to 99%. Thus, total collection efficiency is in
the range of 88 to 89%.
When operations permit, an enclosed tunnel system is suggested. Reference
46 suggests for similar sources that an air velocity of 105 ft/min through the
openings should be maintained. Reference 48 suggests that a total flow of
15,000 to 30,000 cfm is needed. It may be possible to acquire part of this
velocity through the use of air curtains at openings. As with the sidedraft
hood, a wet scrubber is suggested as removal device. With the enclosed hood,
collection efficiencies on the order of 97% can be expected*-^' With lower
grain loadings of 1 to 3 gr/scf due to increased air flow, removal efficiencies
of 994% can be attained.— Thus, a total collection efficiency of 96% can be
obtained.
4.3.1.2 Rotary Shakeout—
A recent development in the foundry industry is the rotary shakeout pic-
tured in Figure 4-12. This system is described in Reference 49. Since no spe-
cific data on control systems have been found, a removal system similar to
that described for revolving screens (Section 2.2.2) is suggested. Reference
49 suggests that air flows on the order of 2,000 to 4,500 cfm are adequate for
dust control. Since shakeout exhaust gases have a high moisture content, a wet
scrubber is suggested as the best removal device. This device claims to be ad-
vantageous because of a decreased level of emissions due to less turbulence.
However, • it,'is limited to small to-medium sized castings.
4.3.2 Grinding . ... . •
» '
As indicated in Section 3.4.2 four basic types of grinders are used- in
foundries: bench grinders, floor stand grinders, portable grinders, and swing
grinders. In general, most grinding operations are controlled with the type of
control dependent upon the type of grinder and casting and other possible op-
erations which may have interfacing control devices. Many control operations
are custom designed so data on generally applicable systems are limited. Those
systems for which data were obtained are described below.
Reference 50 suggests that emissions from a swing frame grinder are best
controlled by an exhaust hood such as that shown in Figure 4-13. A control air
volume of 100 cfm/sq ft of opening is minimum and 150 cfm/sq ft of opening is
preferred. Reference 36 suggests that grinding emissions can be adequately re-
moved (99+7, efficiency) with either a wet scrubber (6- to 30-in. pressure drop)
or a fabric filter. No data on the cost or effectiveness of these systems were
obtained.
For bench and some portable grinding operations, self-contained capture
and removal systems such as the one shown in Figure 4-14 are available. At
least one manufacturer has systems capable of handling up to a 10,000-lb
General flow rates are on the order of 300 ft/min downdraft through the grating.
No details on cost or effectiveness of the system- were obtained.
75
-------
49/
Figure 4-12. Rotary shakeout.—
76
-------
-Branch Take-Off at Top or Back
45° Slope
Grinder lo Operate in or
Close to Face Opening
lote - Verify Dimensions in the Field
Jib Crane Rail Inside Hood
Keep Width as
Small as Practical
4'-6'- Large Opening-Face
Velocity = 100 to 150 fpm -
Never Below 100fpm
2'-0" - 2'-6" - Small Opening-
Grinder in Front-Face Velocity = 200fpm
Transfer Car
• Booth Encloses Grinder
Frame and Suspension
Figure 4-13. Swing frame grinder booth with transfer car.
50/
-------
517
Figure 4-14. Dovndraft grinding control.—
78
-------
Reference 50 suggests that enclosures can generally be used with grinders.
However, care must be taken to ventilate the enclosure in such a way as to
maintain clean worker breathing zones. It is assumed that wet scrubbers or fab-
ric filters can be used with such enclosures.
4.4 MOLD AND CORE PREPARATION
Although particulate emissions are minimal in the core and mold prepara-
tion, the use of organic binders may create a hydrocarbon fugitive emissions
problem. In cold box and green sand molding and core-making, emissions can be
adequately controlled by proper direct venting of the process and hence are
not a fugitive emissions problem. However, production of shell cores and molds
and washing of all types of cores may.produce fugitive emissions problems. Con-
trols for these sources are discussed in the following paragraphs.
4.4.1 Sand Mixing or Mulling
The first step in the preparation of molds and cores is the mixing of sand
and water or binders in a mixer or muller. Most emissions from mulling occur
during loading. The moisture in the muller generally suppresses emissions once
mixing begins. Reference 37 presents capture data for three types of mullers:
no cooling, blow-through cooling and draw-through cooling. These data are pre-
sented in Figure 4-15. It is suggested that a wet scrubber (6- to 10-in. pres-
sure drop) be used to remove the particulate from the emission stream. In gen-
.eral, emissions from sand mulling are well controlled to prevent exposure of
operators to silica dust.
4.4.2 Shell Core and Shell Mold Making .
Control of the shell core-making and molding process requires adequate
ventilation of the shell-making machine and of the cooling area where cores
and molds may continue to emit vapor for periods of up to 30 min.—' In addi-
tion, an effective system for the removal of organic vapors from the pollutant
stream is required. Since sand may be dry mixed, it may be necessary to control
particulate emissions from this source.
4.402.1 Dry Sand Mixing-
Reference 15 suggests that dust produced from dry mixing of sand can be
controlled by either exhaust ventilation or by using a wetting agent to sup-
press dusting. Addition of kerosene at the rate of 0.25% by weight is sufficient
to minimize dust emissions.—
If a dust suppressant is not used, Reference 37 suggests that a closed
type hood with flow rate of 1,000 cfra be used to capture emissions. Dust can
be removed from the pollutant stream by use of either a scrubber (8- to 10-in.
pressure drop) or a fabric filter.
79
-------
Loading
ho,
i-To prevent condensation,
I insulation or strip heaters may
be necessary or use
dilution fitting.
Tight enclosure
Side Stood or
booth
/-Enclosing hood
Bond hopper
Muller
/-Low-velocity duct
used with cooling
type muller.
Cooling fan
blow-through
arrangement
Minimum exhaust volume
Location
Batch hopper .
Bond hopper. .-
Muller:
4't diameter
6' diameter
7'diometer
8'diameter
lO'diometer
Muller type
No cooling
Note 1
600
Note 2
750
900
I05O
1200
1575
Blow-thru
cooling
600
, 6OO
Note 3
II
II
tt
»i
II
Or aw -thru
cooling
Note 1 .
- 600
Note 3
tt
ii
it
it
u
Duct velocity - 4500 fpm minimum
Entry loss - O.25VP
Notes:
I. Botch hopper requires separate exhaust with blow-thru cooling. With other fan arrangement,
(muller under suction) separate exhaust may not be required. (If skip hoist is used, see VS -107)
2. Maintain 150 fpm velocity through all openings in muller hood. Exhaust volume shown are
the minimum to be used.
3. Cooling mullers do not require exhaust if maintained in dust tight condition. Blow-thru fan
must be off during loading. If muller is not dust tight, exhaust as in note 2 plus cooling
air volume.
37/
Figure 4-15. Mixer and muller ventilation.—
80
-------
4.4.2.2 Shell or Hot Box Core and Mold Heating--
The fugitive emission problem from the core-making and molding operations
consists primarily of gaseous organics generated during heating and cooling of
the cores and molds.
Since the core or mold machine cannot be adequately controlled by direct
evacuation, adequate ventilation near the molding or core-making machine and
in the mold or core cooling area is essential for effective control. The cap-
tured emissions must then be vented to a removal device capable of control of
the organic gases.
An exhaust system for a shell-molding system as described as typical in
Reference 15 is shown in Figure 4-16. Air velocities on the order of 150 to
250 fpm into all hood openings are suggested. It is likely that the hooding
system shown is relatively effective in capturing emissions from the molding
machine. However,.it appears that the emissions generated during the transfer
of the mold from the machine to the cooling table are not captured. In addi-
tion, the hood design shown in Figure 4-16 may be of limited effectiveness for
controlling cooling emissions due to deviation of emissions from the shell and
the need to remove shells from the table to make room for newly formed shells.
Lack of data on the total emissions from the molding process and on the
temporal variation of these emissions makes estimation of capture efficiencies
difficult. Reference 52 suggests that some molding area control systems have
effectively captured odorous emissions :with'flow rates on the order of 5,000
acfm for a single molder. However, no data as to size of the area or points
of pickup.were reported.
Primary methods of removal of the organic pollutants from the emission
stream are incineration and chemical absorption systems. However, it should
be noted that many incinerators are being phased out due to energy shortages.
Reference 52 briefly describes currently used design parameters for incineration
and a possible chemical absorption system. These are described in the following
paragraphs.
Reference 52 indicates that until recently, incinerators had been designed
for 0.3 sec retention of shell fumes at 1200°F. However, tests in early 1975
showed that pollutant removal in these incinerators was not sufficient to ef-
fectively reduce odors in the surrounding area. Testing indicates that a reten-
tion time of 0.85 to 1 sec at 1500°F is necessary for effective hydrocarbon re-
moval. The relative costs and efficiencies of these two designs for a 5,000-cfm
system are given in Table 4-3.
Reference 52 goes on to describe a chemical absorption system for hydro-
carbon removal which has been applied to shell core and mold operations at
several foundries. The system, known as System 1, consists of single scrubbing
tower using an acidic absorption solution in conjunction with a proprietary
81
-------
To Wet
Scrubber
Flexible
Tube
Movable
Hood
Figure 4-16. Control system for shell molding.
82
-------
TABLE 4-3. POLLUTANT REMOVAL SYSTEMS FOR SHELLCORE AND MOLD MACHINES-^
52/
System
Costs
Pollutant concentrations (ppm)
CO
__^__________»___ Equipment Operating
Formaldehyde Amines Ammonia Phenol $/cfm $/hr/l(P cfm
No control
700
10
250
200
20
00
System 1 350
Standard
incineration
(1200°F,
0.3 sec) 3,000
200
150
20
3.40
4.25
0.01
1.68
Proper
incineration
(ISOCfF,
0.85-1 sec) < 100
0
8.60
3.85
-------
packing tempering agent that creates an ion exchange reaction in acidic solu-
tion. System maintenance requirements include filling the acid tank every 2
weeks, partial draining of the system once per week, checking the pH control
every 2 weeks and completely draining and refilling the system twice per year.—
Relative efficiency and cost of the system are given in Table 4-3.
4.4.3 Cold Box Core and Mold Making
No data were obtained on systems to capture and remove emissions from cold
box systems.
4.4.4 Core Washing
Both organic and particulate emissions may be generated from core washing.
Reference 37 suggests that spray booths are commercially available in all sizes.
A flow rate of 100 to 175 cfm/sq ft of hood opening is suggested. It is currently
general practice to duct the air stream from such booths directly to the atmo-
sphere. Sufficient data on emission characteristics have not been obtained to
recommend a removal device for this source.
4.5 WASTE HANDLING
Each of the waste handling operations in the foundry industry has the po-
tential to generate particulate emissions. It appears that for the most .part
control methods should be aimed at minimizing emissions. Possible methods are
described below. ' "•••'•'-•. . . • '! .
4.5.1 Slag Quenching
No data were obtained on methods for control of emissions from slag
quenching.
4.5.2 Waste Sand Transfer
Transfer of waste sand both within and away from the foundry can be con-
trolled by adequate use of watering or chemical wetting systems. It is esti-
mated that proper wetting can control internal transfer points with 90 to 95%
efficiency.— A combination of wetting and covered transfer mode (conveyor or
truck) can attain the same level of efficiency for external transfer. The ini-
tial cost of a spray system is estimated to be $10,000 to $15,000,-^' No data
are available on operating costs.
4.5.3 Sand and Slag Storage
The primary points of emission from waste storage are load-in and wind
erosion from storage areas. If proper wetting is attained for waste transfer,
load-in emissions should be minimized. The following paragraphs describe meth-
ods for the elimination of emissions from wind erosion.
84 .
-------
The process of stabilizing the surface layer of a pile consists of bind-
ing the surface particulates into a nonerodible crust. Occasional watering of
the pile surface or the addition of chemical crusting agents will accomplish
this task.
The control efficiency associated with periodic watering of the pile sur-
face is estimated to be 80%, assuming that wetting of storage piles occurs on
a regular basis.— Water spray systems may consist of stationary ground level
sprinkler systems, tower-mounted sprinklers, or mobile tank-truck sprayer sys-
tems. An operating example is a stationary ground level system wetting two 900-
ft long coal piles utilizing sprinkler heads spaced every 180 ft."Under dust
producing meteorological conditions, the system of 32 sprinklers surrounding
the piles sprays about 13,000 gal. of water per day. This system adequately
controls wind erosion generation of fugitive dust^-t'
A sprinkler system mounted on a 30-ft tower producing a dense, 40-ft wide
cloud of water mist has been used to minimize storage pile wind erosion at a
quarry site. This system, which is both wind speed and direction activated,
has produced favorable results.—
The control efficiencies associated with the spraying of surface crust-
ing agents upon storage piles can extend to 9970, as derived from wind tunnel
tests«^£' Surface crusting agents can be applied by either stationary or mobile
sprinkler systems. Example chemicals and application rates for different types
of these crusting agents are presented in Table 4-4. . • . '
The initial cost of erecting a stationary elevated water spray system,
.which controlled one relatively large stockpile, was estimated to be about
$11,000, including sprays, piping, pumping, wind instruments, and installation
costsjii' No annual operating costs were obtained for this system.
The cost of applying surface crusting agents to storage piles from sta-
tionary equipment is assumed to be slightly more costly. This assumption is
based on the need for additional mixing chambers and proportioners to dilute
the crusting agents with water. The cost of applying these various surface
crusting agents is presented in Table 4-4.
85
-------
TABLE 4-4. EXAMPLE SURFACE CRUSTING AGENTS FOR STORAGE PILES
AND EXPOSED AREASS/
Surface crusting
agent (concentrate)
Dilution
Application
rate
Application
costll/
A. Organic polymers
• Johnson-March,
SP-301
• Houghton,
Rexosol 5411-B
Full 1 gal. concentrate
strength per 100 ft2
2% solution 1 gal. concentrate
per 300
1.2C
0.7c
B. Petroleum resin
water emulsion
• Witco Chemical,
Coherex
Latex type-synthetic
liquid adhesive
• Dowell M145
chemical binder
20%
solution
. water
solution
1 gal. concentrate
per 50 ft2
4 gal. of 47. solution
per 100 ft2
0.4c
0.4C
a/ Reference 53.
W Cost per square foot of surface area.
86
-------
SECTION 5.0
RESEARCH AND DEVELOPMENT RECOMMENDATIONS
Specific research areas which need investigation before adequate control
of fugitive emissions in iron foundries can be accomplished are identified in
this section. The flow diagram in Figure 5-1 depicts the methodology used to
determine the research needs. Although the ultimate purpose of a research and
development (R&D) program is the provision of appropriate technology for fugi-
tive emissions sources, preliminary programs which identify levels and charac-
teristics of these emissions may be necessary.
The first step in determining R&D needs is the identification of those
sources which have emissions of sufficient severity to be considered a serious
problem. For those sources identified in the initial step, availability of
control techniques must be determined. If possible control techniques are
available, the efficiency and cost of .the techniques must be analyzed.^ Finally,
for those research needs identified at each stage of the process, the adequacy
of ongoing research to meet these needs must be determined.
The following subsections present information used to determine research
needs in each of the above areas. Critical sources of emissions are determined
and data gaps are identified; sources for which no control is available are
identified; deficiencies in information on performance and cost of control
techniques are evaluated; and ongoing research is examined. Finally, R&D pro-
grams are recommended.
5.1 DETERMINATION OF CRITICAL SOURCES
This section identifies those sources which have been determined to have
the most critical emissions problem.
The severity of a fugitive emissions source in an iron foundry is dependent
on: (a) the total particulate emission level; (b) the percentage of fine par-
ticulate (i.e., particles smaller than 5 A"n)j (c) the current extent of control;
and (d) the presence of organic emissions. Each of these parameters is discussed
and quantified below.
87
-------
Yes
Yes
Yes
Develop Preliminary Ranking
of Source Control Needs
Based on Emission Rare
Determine Emission Factor
for High Ranking Source
Yes
Are Emission
Data Adequate?
Are There Control Techniques
for the Source?
©
Is the Control Technique
Efficiency Properly Quantified
as a Function of All Important
Independent Variables?
Is the Control Technique Cost
Effective at the Desired
Control Efficiency?
Control is Suited for Source
Is There Ongoing
Research?
There Exists a Need
for Research and
Development
Figure 5-1. Flow diagram to determine the need for R&D.
88
-------
An estimate of total particulate emissions and the percentage of fine
particulate was developed for each of the major sources in iron foundries and
presented in Table 3-2. The level of fine particulate emissions is important
because fine particles are most hazardous to human health and that have great-
est potential for atmospheric dispersion. Table 5-1.lists the major sources in
descending order of uncontrolled fine particulate emissions.
In addition, MRI has developed estimates of the industry-wide extent of
control for each source. This is a combination of capture efficiency and re-
moval efficiency on an industry-wide basis and should not be construed to be
applicable to a particular operation. These estimates are based on data pre-
sented by Gutow,— observations made by MRI personnel during plant visits,
and conversations with industry personnel. These estimates are presented in
Table 5-1 and a ranking of sources on a controlled basis is developed. It
should be noted that these estimates are given a relatively low level of con-
fidence and should be applied with great care to circumstances other than this
ranking.
Based on the data in Table 5-1, those sources having the highest uncon-
trolled fine particle emission levels are:
1. Iron cooling
•2. Iron pouring . .• • . .' ".-...''•'.
3. Shakeput ; . ,
4. Spent sand handling
5. Mulling
Those five sources having the highest ranking based on controlled fine particle
emissions are:
1. Iron cooling
2. Iron pouring
3. Shakeout
4. Electric arc furnace
5. Iron inoculation
To this point, sources have been ranked strictly on the basis of particu-
late emission levels. However, it appears that several processes within the
foundry industry are potential sources of organic emissions. Although industry
89
-------
TABLE 5-1. RANKING OF PARTICULATE EMISSION SOURCES
vO
o
Source
No.-'
26
25
27
28
31
18-20
24
21-23
10
43
44
33
44
5,6,8
16
7
15
30
37
37
38,39
41
42
Uncontrolled fine
particle emissions
Emission source (Ib/yr)
Iron cooling
Iron pouring
Shakeout
Spent sand handling
Mulling
Electric arc furnace
Iron inoculation
Induction furnace
Scrap storage
Waste sand transfer
Waste sand storage
Molding
Slag storage
Coke handling
Cupola tapping
Coke storage
Sand storage
Grinding
Shell or hot box heating
Shell or hot box holding
Cold set mold and core
Core wash
Slag quench
9.0 x 107
8.4 x 107
3.5 x 107
1.8 x 107
1.7 x 107
7.8 x 106
7.4 x 106
2.2 x 106
6.0 x 105
5.1 x 105
4.2 x 105
4.2 x 105
8.1 x 104
7.8 x 104
- "
Neg.
Neg.
-
-
-
-
—
Extent of
control
10
10
75
90
90
10
25
10
0
0
0
0
0
50
-
0
50
50
-
-
-
-
™
Controlled fine
particle emissions
(Ib/yr)
8.1 x 107
7.6 x 107
8.8 x 106
1.8 x 106
1.7 x 106
7.0 x 106
5.6 x 106
2.0 x 106
6.0 x 105
5.1 x 105
4.2 x 105
4.2 x 105
8.1 x 104
3.9 x 104
_
-
-
.
-
-
-
-
—
Ranking of
controlled
emissions
1
2
3
7
8
4
5
6 '
9
10
11
12
13
14
_
,
a/ See Table 2-2.
-------
personnel contend that low in-plant organic concentrations indicate minimal
emissions from some of these sources, data appear insufficient to warrant such
a conclusion. The sources of organic emissions are:
1* Iron pouring
2. Iron cooling
3. Shell and hot box processes
4. Cold set operations
5. Shakeout
Data obtained during the study were not sufficient to quantify levels of or-
ganic emissions. Thus, it is not possible to rank these sources.
Based upon the analysis of both particulate and organic emissions, nine
sources have been identified as representing possible critical control needs.
These can be grouped into the following generic classifications:
1. Iron pouring and cooling
2. Shakeout
3. Sand handling operations (including mulling)
4. Electric arc furnaces
5. Iron inoculation
6. Core and mold preparation using organic binders
Since data throughout the industry are extremely limited, the response to
Step 3 of the R&D flow chart is '.'no" for each source.
5.2 ANALYSIS OF CONTROL AVAILABILITY
Step 5 of the evaluation process asks the question, "are there control
techniques for the sources"? In response, data developed in Section 4.0 have
been used to determine the status of control capability for each major fugi-
tive emission source in the foundry industry. The status of control of each
source falls into one of three categories.
1. Standard methods of control are available and are generally or ex-
tensively applied around the industry.
91
-------
2. Acceptable controls appear to be available but are not yet applied
extensively throughout the industry.
3. Current state-of-the-art appears to be inadequate.
Those sources falling into the first category obviously have little need
for control research. Research programs for the second source will deal with
further analysis and development of available equipment. In addition the pos-
sibilities for the expanded use of the technology must be examined. Finally,
for those sources falling in the third category, a major research effort di-
rected toward the development of adequate and feasible methods may be needed.
The only source falling in the first category is the sand handling and
mulling area. Data in Section 4.0 indicate that not only are control techniques
available but they are also well .engineered to meet performance requirements
and are cost effective. Thus, available controls are suited for the source and
no further research is needed.
Those sources with significant emission levels which fall in the second
category include:
1. Electric arc furnaces
A
2, Iron inoculation
3. Shakeout (particulate emissions)
Those sources falling into the third category are:
1. Core and mold preparation using organic binders
2. Iron pouring and cooling
3. Shakeout (organic emissions)
For electric arc furnaces close capture hooding systems have been util-
ized in conjunction with fabric filters. These appear to adequately control
fugitive emissions. However, further data are needed detailing the effective-
ness of these systems. Additional information also needs to be developed on
the engineering problems which may limit application of these systems.
Several methods have been developed which appear adequate for control of
particulate emissions for iron inoculation. However, no data are available on
either the extent of application of these methods or the effectiveness of the
method.
92
-------
Finally, shakeout particulate appeared to be well-controlled in about
half of the operations observed during plant visits. However, this was based
on visual operation and no substantive data have been found which indicate
effectiveness of the various control systems. In addition, shakeout operations
vary so greatly from plant to plant, that extensive study may be needed to
determine the feasibility of application of shakeout controls to less mecha-
nized foundries or those foundries preparing large castings. Little information
is available on organic emissions or control for shakeout.
Few data, other than these obtained from a single control device manufac-
turer, are available on the availability of control systems for those core and
mold preparation methods which utilize organic binders. In all foundries vis-
ited, these sources were not controlled and it is probable that they are not
controlled in most iron foundries. The data available do not adequately de-
scribe an effective capture system.
No data are available on the adequacy or extent of application of any
type of capture and removal system for either core washing or cupola tapping.
The control problems involved with iron pouring and cooling of molds are
two-fold. For those foundries which utilize floor pouring, either because of
size of castings or economic inability to develop mechanized lines, a tech-
nically and economically feasible capture system is necessary. For all types
of pouring and cooling operations,,an effective removal .system is needed to
handle the combination of moist particulate and organic emissions.
5.3 CURRENT RESEARCH - "•
Knowledgeable personnel in the foundry industry and in governmental agen-
cies were contacted to identify current research programs concerned with fugi-
tive emissions quantification or control. A computer search of the Smithsonian
Scientific Information Exchange files was also conducted to identify pertinent
research programs listed there. Only three relevant research programs were
identified. These are summarized in Table 5-2.
The most significant studies appear to be the NIOSH study conducted by
Envirex and the EPA study conducted by Research Triangle Institute (RTI). The
Envirex study is an extensive analysis of the state of the art of internal
foundry emission control technology. Most sources covered are identical to
those identified as fugitive emissions sources in this report. However, the
Envirex study will be primarily concerned with an analysis of capture or abate-
ment mechanisms. With the help of the foundry industry the 30 to 40 best con-
trol foundries have been identified. Thirty of these plants will be chosen for
further analysis. At each of these plants control equipment design data, effec-
tiveness and application problems will be identified. It is suggested that the
results of this project be carefully analyzed by EPA personnel. Little informa-
tion was obtained on the RTI study„
93 •:
-------
TABLE 5-2. CURRENT RESEARCH
Research
group
Sponsoring
organization
Program summary
Program status
A. D. Little
Company
EPA
Study of economic im-
pacts of air pollu-
tion regulations on
the foundry indus-
try.
To be completed 1979.
Have completed a
macro-analysis.
Beginning a micro-
analysis for spe-
cific processes.
Envirex
NIOSH
A study of the state
of the art of in-
ternal foundry con-
trol technology.
The control systems
are to be analyzed
at the 30 best con-
trolled foundries
in the U.S.
Have identified 30
foundries. Study
to be completed
in late 1978.
Research Triangle
Institute (RTI)
EPA
A multimedia assess-
ment of iron found-
dry processes and
their related pollu-
tion control tech-
nologies with
particular atten-
tion directed toward
organic emissions.
Study to be completed
in mid-1979.
94
-------
5.4 SUGGESTED RESEARCH
Based upon the ranking developed in Section 5.0 and the emissions data
and control technology deficiencies discussed above, R&D programs are sug-
gested for four types of sources in iron foundries. In order of priority these
are:
1. Pouring and cooling
2, Electric arc furnaces
3. Core and mold preparation using organic binders
4. Shakeout
The pouring and cooling of iron castings in sand molds is by far the most
significant problem in the industry and will require the greatest effort to
develop emissions control technology. As indicated earlier two basic methods
are used for pouring and cooling. The first method utilizes a fixed pouring
station which is normally hooded. The molds are moved to the pouring station
by conveyor and into a cooling tunnel which is vented. In almost all cases,
the ventilation systems for pouring and cooling are vented without pollutant
removal to the atmosphere. The second method, floor pouring and cooling is
conducted in an open area with molds placed on the floor and pouring ladles
moving to the mold for pouring. Generally, no control is applied to these
sources. This method is used in smaller, older foundries and in cases.where
the mold and casting are too large to be moved b^ conveyor.
The basic control technology deficiencies for these sources are (a) lack
of emissions data, (b) unavailability of a capture system for floor pouring
and cooling, and (c) insufficient removal systems for either floor pouring or
fixed pouring stations. The tasks which must be accomplished to effectively
deal with the problem are:
1. Definition of the emissions problem. This will include data to define
total particulate and gaseous emissions, type of organic vapors emitted, par-
ticle size distribution, moisture content of emissions stream, temporal vari-
ation of emissions during pouring and cooling. Some of these data have been
developed for a bench scale system by Bates and Scott,— but no data are
available for full-scale operation.
2. Development of a technically and economically feasible capture system
for floor pouring and cooling.
3. Development of an effective removal system for the captured emissions
stream. This should include an analysis of any systems currently used on pour-
ing or cooling lines. No published data are available for such systems.
95
-------
Electric arc furnaces emit fine metallic fumes to the foundry and even-
tually to the atmosphere during charging, tapping, and meltdown from leakage
around the electrodes. A close' capture hooding system is described in Section
4.0 which visually appears to effectively capture these emissions. However,
the system has been applied only to a limited number of EAFs and no test data
are available. Canopy hooding systems have also been applied to EAFs with some
degree of effectiveness, but again no actual data to determine the effective-
ness of canopy hoods are available.
The tasks which must be completed to allow foundry personnel to determine
the best method for controlling EAF fugitive emissions are:
1. Development of an adequate methodology for testing uncontrolled and
controlled emissions from EAFs and testing of uncontrolled emissions and the
effectiveness of close capture and canopy hoods.
2. Determination of technical limitations for application of both close
capture and canopy hood systems.
3. Evaluation of the economic impact of both systems on foundries which
currently have uncontrolled EAFs.
The primary fugitive emissions problem in the core and mold preparation
area is the evolution of organic gases from operations using organic binders.
Data on the quantities of organics emitted from these sources and the effec-
tiveness of control systems are not adequate. It is suggested that the problem
should be further defined and the significance of these s.ources should be de-'
termined before control technology research can be justified. The tasks which
need to be accomplished to define the problem are:
1. Development of an inventory of core and mold production processes and
the type of binders used on a geographic basis.
2. Determination of total organics emitted and the compounds present for
each major process.
3. Determination of emissions levels on a geographic basis and evalua-
tion of the significance of the source.
It appears that this program may be covered in part by the RTI study.
Effective controls appear to have been developed for many shakeout opera-
tions. However, lack of accurate emissions data prevents analysis of the effec-
tiveness of these controls. The first task in any program should be the deter-
mination of emission levels on a full-scale shakeout and the effectiveness of
available control systems. After this has been completed, those foundries not
96
-------
employing fugitive controls should be analyzed. Methods should then be deter-
mined for control of emissions at these foundries*
In summary, the four areas for which R&D programs are needed are: (a)
iron pouring and cooling, (b) electric arc furnaces, (c) mold and core prepa-
ration processes utilizing organic binders, and (d) shakeout. Adequate testing
methods must be developed to quantify emissions from these sources and to de-
termine the effectiveness of alternative control methods.
97
-------
SECTION 6.0
REFERENCES
1. Personal communication with Nauro T. Georgieff.
2. AT. Kearney Company. Systems Analysis of Emissions and Emission Control
in the Iron Foundry Industry, Volume II: Exhibits. PB198 349. U.S.
Environmental Protection Agency, February 1971, Exhibit IV-8.
3. Ibid., Exhibit IV-9.
4. Ibid., Exhibit IV-13.
5. Ibid., Exhibit IV-14.
6. Ibid., Exhibit III-7.
7. Sohr, R. T. Economic Air Pollution Systems for Foundries Using Any Va-
riety of Shell. P.C S. Hormel, Chicago, Illinois, p. 6.
8. Sohr, R. T. Cold Box Air Pollution Problems. No More—With an -Economical
Chemical Absorption System. Presented at the 80th Casting Congress and
Exposition of the American Foundrymen's Society, April 1976. pp. 8-9.
9. Gutow, B. S. An Inventory of Iron Foundry Emissions. Modern Casting.
January 1972.
10. A. T. Kearney Company. Systems Analysis of Emissions and Emission Control
in the Iron Foundry Industry, Volume II; Exhibits. PB198 349. U.S. En-
vironmental Protection Agency, February 1971, Exhibit VI-15.
11. Bates, C. E., and W. D. Scott. Better Foundry Hygiene Through Permanent
Mold Casting. Contract No. 1 R01 OH 000456-01, NIOSH. January 1976. p. 68,
12. Ibid., p. 64.
13. A. T. Kearney and Company. Systems Analysis of Emissions and Emission
Control in the Iron Foundry Industry, Volume I, Text. U.S. Environmental
Protection Agency, February 1971.
98
-------
14. Modi, E. K. Comparing Processes for Making Ductile Iron. Foundry, July
1970. p. 42-49.
15. Molding, Goremaking and Patternmaking, American Foundrymen's Society,
1972.
16. Bates, C. E., and L. D. Scheel. Processing Emissions and Occupational
Health in the Ferrous Foundry Industry. American Industrial Hygiene
Association Journal, August 1974.
17. Bates, C. E., and L. D. Scheel. Processing Emissions and Occupational
Health in the Ferrous Foundry Industry. American Industrial Hygiene
Association Journal, August 1974. p. 452-462.
IS. A. T. Kearney Company. Systems Analysis of Emissions and Emission Control
in the Iron Foundry Industry, Volume I, Text. U«S» Environmental Protec-
tion Agency, February 1971, p. IV-28.
19. Current Industrial Reports: Iron and Steel Foundries and Steel Ingot
Producers, Summary for 1974. U.S. Department of Commerce, Series:
M33A974)-13, October 1975.
20. Annual Statistical Report. American Iron and Steel Institute, 1975.
21. Kalika, P. W. Development of Procedures for Measurement of Fugitive
." Emissions. U.S. Environmental Protection Agency, Contract No. 68-02-1815,
July 1975.
22. Bohn, R., T. Cuscino, and C. Cowherd, Jr. Fugitive Emissions from Inte-
grated Iron and Steel Plants. EPA-600/2-78-050, March 1978.
23. Personal cummunication with Mr. Denise, Independence Foundry and Manu-
facturing Company.
24. Vandegrift, A. E., and L. J. Shannon. Handbook for Emissions, Effluents
and Control Practices for Stationary Particulate Pollutant Sources.
Midwest Research Institute, EPA Contract No. CPA 22-69-104, November 1,
1970.
25. Modern Casting. August 1970.
26. A. T. Kearney Company. Systems Analysis of Emissions and Emission Control
in the Iron Foundry Industry, Volume II: Exhibits. PB198 349. U.S. En-
vironmental Protection Agency, February 1971, Exhibit VI-16.
27. Personal communication with Pramohd Nighawan.
99
-------
28. Compilation of Air Pollutant Emission Factors. U.S. Environmental Protec-
tion Agency, Publication No. AP 42, Supplement No. 5, October 1975.
29. A. T. Kearney Company. Systems Analysis of Emissions and Emission Control
in the Iron Foundry Industry, Volume I, Text. U.S. Environmental Protec-
tion Agency, February 1971, pp. VI-42, 43.
30. A. T. Kearney Company. Systems Analysis of Emissions and Emission Control
in the Iron Foundry Industry, Volume II: Exhibits. PB198 349. U.S. En-
vironmental Protection Agency, February 1971, Exhibit VI-19.
31. Personal communication with Mr. W. Huelsen.
32. Bates, C. E., and W. D. Scott. Better Foundry Hygiene Through Permanent
Mold Casting. Contract No. 1 R01 OH 000456-01, NIOSH. January 1976. p. 64.
33. Ibid., p. 66.
34. Bohn, R., T. Cuscino, and C. Cowherd, Jr. Fugitive Emissions from Inte-
grated Iron and Steel Plants. EPA-600/2-78-050, March 1978.. p. 3-43.
35. Personal communication with Mr. William B. Huelsen, Director, Environ-
mental Affairs. American Foundrymen's Society, November 5, 1976.
36. A. T. Kearney Company. Systems Analysis of Emissions and Emission Control
in the Iron Foundry Industry, .Volume II: Exhibits. PB198 349..U.S. En-
vironmental Protection Agency, February 1971, Exhibit VII-21.
37. Design of Sand Handling and Ventilation Systems. American Foundrymen's
Society, 1972.
38. Personal communication with Joseph Schumacher.
39. Melting and Pouring Operations. American Foundrymen's Society, 1972.
40. Personal communication with Pramohd NighavTan.
41. Personal communication with James Overmeyer.
42. Plant visit to Wagner Castings.
43. Bates, C. E., and W. D. Scott. Better Foundry Hygiene through Permanent
Mold Casting. Southern Research Institute, January 30, 1976.
44. Bates, C. E. Profit Potential in Permanent Mold Iron Castings. Foundry,
November 1972.
45. Dust and Fume Control Systems. Catalog 12745-WG, Kirk and Blum Manufac-
turing Company.
-------
46. Dust, Fume and Smoke Hoods for Shakeouts, Pouring Stations, Mold Con-
veyors. Bulletin 574, Schneible Company, Holly, Michigan.
47. Personal communication with Mr. A. S. Lundy, Schneille Company.
48. Kane, J. M. Air Pollution Ordinances. Foundry, October 1952.
49. Does Your Shakeout Comply With OSHA? Bulletin, Didion Manufacturing
Company.
50. Cleaning Room, American Foundrymen's Society, 1972.
51. Brochure, Wolvering Dust Collecting Equipment, Wolvering Equipment Company.
52. Sohr, R. T. Economic Air Pollution Systems for Foundries Using Any Variety
of Shell, Pollution Control System, Division of George A. Hormel and
Company, Chicago, Illinois.
53. Bohn, R., T. Cuscino, and C. Cowherd, Jr. Fugitive Emissions from Inte-
grated Iron and Steel Plants. EPA-600/2-78-050, March 1978. p. 6-24.
54. Price, W. L. Open Storage Piles and Methods of Dust Control. Paper Pre-
sented at the October 1972 Meeting of the' American Institute of Mining
Engineers, Birmingham, Alabama.
55. Personal communication with Richard, R. Cole, Harry T. Campbell Sons'
Company, Baltimore, Maryland, March 11, 1977.
56. Bohn, R., T. Cuscino, and C. Cowherd, Jr. Fugitive Emissions from Inte-
grated Iron and Steel Plants. EPA-600/2-78-050, March 1978. p. 6-10.
101
-------
SECTION 7.0
GLOSSARY
Duration of Storage - The average time that a unit of aggregate material
remains in open storage, or the average pile turnover time.
Dust Suppressant - Water or chemical solution which, when applied to an aggre-
gate material, binds suspendable particulate to larger particles.
Emission Control System, Primary - A control system installed to capture and
remove most of the total emissions prior to atmospheric discharge.
Emission Control System, Secondary - A control system designed to capture and
remove the smaller portion of the total emissions that the primary system
does not collect with the smaller portion usually being fugitive in nature.
Enclosure - A structure which either partially or totally surrounds a fugi-
tive emissions source thereby reducing the amount ,of emissions.
Fugitive Emissions, Total - All particles from either open dust or process
fugitive sources as measured immediately adjacent to the source.
Fugitive Emissions - Emissions not originating from a stack, duct, or flue.
Load-In - The addition of material to a storage pile.
Load-Out - The removal of material from a storage pile.
Materials Handling - The receiving and transport of raw, intermediate and waste
materials, including barge/railcar unloading, conveyor transport and associ-
ated conveyor transfer and screening stations.
Moisture Content - The mass portion of an aggregate sample consisting of un-
bound moisture on the surface of the aggregate, as determined from weight
loss in oven drying with correction for the estimated difference from total
unbound moisture.
102
-------
Partial Diameter, Aerodynamic - The diameter of a hypothetical sphere "of unit
density (~1 g/cnP ) having the same terminal settling velocity as the particle
in question, regardless of its geometric size, shape, and true density.
Particle Diameter, Stokes - The diameter of a hypothetical sphere having the
same density and terminal settling velocity as the particle in question,
regardless of its geometric size and shape.
Particulate, Fine - Airborne particulate smaller than 5 p,m in Stokes Diameter.
Particulate, Suspended - Airborne particulates smaller in Stokes diameter than
30 pm. the approximate cut-off diameter for the capture of particulate
matter by a standard high-volume sampler, based on a particle density
of 2 to 2.5 g/cm3.
Precipitation-Evaporation Index - A climatic factor equal to 10 times the sum
of 12 consecutive monthly ratios of precipitation in inches over evaporation
in inches, which is used as a measure of the annual average moisture of a
flat surface area.
Source, Open Dust - Any source from which emissions are generated by the force
of wind and machinery acting on exposed aggregate materials.
Source, Process Fugitive Emissions - An unducted source of emissions involv-
ing a process step which alters the chemical or physical characteristics
of a material, frequently occurring within a building.
Silt Content - The mass portion of an aggregate sample smaller than 75 M-m in
diameter as determined by dry sieving.
Spray System - a device for applying a liquid dust suppressant in the form
of droplets to an aggregate material for the purpose of controlling the
generation of dust.
Storage Pile Activities - Processes associated with aggregate storage piles,
specifically, load-in, vehicular traffic around storage piles, wind eros-
sion from storage piles, and load-out.
Surface Stabilization => The formation of a resistive crust on an exposed ag-
gregate surface through the action of a dust suppressant, which suppresses
the release of otherwise suspendable particles.
103
-------
SECTION 8.0
ENGLISH TO METRIC UNIT CONVERSION TABLE
English unit
Multiplied by
Metric unit
Ib/ton
lb/vehicle mile
Ib/acre year
lb
Ton
mph
mile
ft
acre
0.500
0.282
112
0.454
0.907
0.447
1.61
0.305
0.00405
kg/ ton »
kg/ vehicle km
kg/km year
ton
m/s
km
m
km2
104
------- |