United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 2771 1
EPA-600/2-80-021
January 1980
Research and Development
vvEPA
Environmental
Assessment of
Iron Casting
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and 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
•A,
~% This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
~jf NOLOGY series. This series describes research performed to develop and dem-
"*; * ^ onstrate instrumentation, equipment, and methodology to repair or prevent en-
-_3f * vironmental degradation from point and non-point sources of pollution. This work
fH| provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the contents necessarily
reflect the views and policy of the Agency, 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.
-------
EPA-600/2-80-021
January 1980
Environmental Assessment
of Iron Casting
by
V.H. Baldwin, Jr.
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, North Carolina 27709
Contract No. 68-02-2630
Task No. 2
Program Element Nos. 1AB604C and 1BB610C
EPA Project Officer: Robert V. Hendriks
Industrial Environmental Research Laboratory
Office of Environmental Engineering and Technology
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
itavtronrnema] Pr®t*ctl0n Agpnsy
Sc:ih DtMi'born Street
60804
-------
ENVIRONMENTAL ASSESSMENT OF IRON CASTING
ABSTRACT
Sampling of ductile iron casting in green sand molds with phenolic iso-
cyanate cores and in phenol-formaldehyde bound shell molds did not provide de-
finitive proof that environmentally hazardous organic emission occur. Both
molding systems produced the same type of major emissions, alkyl halides, car-
boxylic acid derivatives, amines, substituted benzenes, nitrogen hetero-
cyclics, and fused aromatics in quantities that slightly exceed the lowest
Minimum Acute Toxicity Effluent (MATE) values for the categories, but probably
not for individual compounds. GC-MS analysis revealed the major fused aromat-
ics to be naphthalene compounds. Quantitative analysis of specific PNA's
showed no significant level of concern. Inorganic dust emissions are haz-
ardous if uncontrolled because of silicon, chromium, and nickel. The dust is
sufficiently high in 12 metals to render it a hazardous waste if collected as
a sludge and landfilled, but leachate testing may change that categorization.
Relatively high levels of Sr, Ba, Ce, Pr, and Nd in the dust indicate that in-
oculation smoke should be examined.
-------
DISCLAIMER
This report has been reviewed by the Industrial Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views and
policies of the U.S. Environmental Protection Agency, nor does mention of
trade names or commercial products constitute endorsement or recommendation
for use.
-------
TABLE OF CONTENTS
Page
TABLES v
FIGURES ' viii
ACKNOWLEDGEMENT ix
1.0 SUMMARY 1
1.1 Particulate Analysis . 2
1.2 Organic Analysis. . 2
1.3 Inorganic Analysis 7
2.0 CONCLUSIONS 9
3.0 INTRODUCTION 10
4.0 INDUSTRY DESCRIPTION 11
5.0 PROCESS ANALYSIS 15
5.1 Casting Methods . 15
5.1.1 Green Sand . 18
5.1.2 Inorganically Bound Molds 19
5.1.3 Organically Bound Sand 19
Shell Molding 20
Hot Box Molds 21
Cold Set Binders 21
No-Bake Resins 21
Oils 22
Full Mold Process 22
5.1.4 Permanent Molds 22
5.1.5 Physically Bonded Molds 22
5.2 Supporting Processes 24
5.2.1 Pattern Making 24
5.2.2 Sand Processing 24
5.2.3 Iron Melting 25
Cupola 25
Induction Furnaces 26
Electric Arc Furnaces 27
5.2.4 Inoculation 28
The Nature of Inoculation Smoke 32
5.2.5 Pouring 33
5.2.6 Cooling 37
5.2.7 Shakeout '. 37
5.2.8 Finishing 32
6.0 WASTE STREAM CHARACTERISTICS 39
6.1 Solid Wastes og
6.2 Particulate Emissions rn
6.3 Water Effluents 52
-------
TABLE OF CONTENTS (cont. )
Page
6.4 Potential Pouring and Shakeout Discharges ........ 52
6.5 Decomposition Products of Substances Used in
Molds and Cores ..................... 55
7.0 ENVIRONMENTAL DATA ACQUISITION ................ 59
7.1 Sampling and Analytical Strategy ............. 59
7.2 Test Site Selection ................... 60
7.3 Source Assessment Sampling System Acquisition
of Samples ....... . ................ 52
8.0 ENVIRONMENTAL DATA ANALYSIS ..... . ............ 57
8.1 Analysis of SASS Train Sampling of Green Sand
Shakeout Effluent; Sample 1 ............... 70
8.1.1 Total Particulate Loading ......... . . . . 7]
8.1.2 Level 1 Organic Analysis ............. 71
8.1.3 Inorganic Analysis ................ g-j
8.2 Analysis of SASS Train Sampling of Scrubber
Effluent from Shakeout of Green Sand Molding With
Isocyanate Cores .................. ... 33
8.2.1 Total Particulate Loading. . . .......... 83
8.2.2 Level 1 Organic Analysis ............. g3
8.2.3 Inorganic Analysis ................ 8^
8.3 Analysis of Fugitive Emissions in the Shakeout
Room of a Phenolic Shell Molding Foundry, Sample 3. ... 86
8.4 Comparison of Organic Emissions to MATES ......... 88
9.0 DISCUSSION OF RESULTS ..................... 90
9.1 Analysis of Physical-Chemical Mechanisms
Affecting Emissions ................... gg
9.2 Comparison of Emissions From Different Chemical
Sources ......................... g2
9.3 Comparison of Laboratory Versus Field
Measurements ............... ........ g5
9.4 Recommendations ..................... gj
9.4.1 Control of Shakeout Emissions ........... gy
9.4.2 Pouring Emissions ................. gy
9.4.3 Inoculation Smoke ................. g8
9.4.4 Chromium Emissions ................ g8
REFERENCES ............................... 100
APPENDIX
A Decomposition Products of Some Substances Used
in Molds and Cores ..... ................. -
B Level 1 Organic Analysis Data of Samples 1-3, and
Inorganic Analysis Data ............. . ...... 123
-------
LIST OF TABLES
Page
1 Foundries and Iron Foundries in Each State as of 1976 12
2 Organic Core Binder 20
3 Magnesium Treatment Systems Emissions Report for
Ductile Iron Production and Gray Iron Desulfurization 31
4 Characteristics and Sources of Emissions in Various
Foundry Departments 41
5 Pounds of New Material Purchased Per Year By Category 43
6 Percentage of Material Purchased By Category Excluding
Metal Melted . . 44
7 Pounds of New Material Consumed Annually Per Ton of
Metal Melted 45
8 Estimated Pounds of Material to Landfill Per Year By
Category 46
9 Estimated Percentage of Material to Landfill Per Year
By Category 47
10 Estimated Pounds of Material to Landfill Per Ton of
Metal Melted . 48
11 Estimated Pounds of Material to Landfill Per Ton of
Metal Shipped 49
12 Particulate Size Distributions of Green Sand Emissions
for 4" Cube Pattern 51
13 Ranges of Pollutants in Selected Wastes. 51
/
14 Lysimeter Results—18 Simulated Months 53
15 Pyrolysis Products of Some Binder Materials 57
16 Summary of Particulate Data 57
17 Summary of Organic Data 67
18 Production During Sampling 53
19 Particulate Concentration 7]
20 Summary of Sampling Data for Green Sand Shakeout,
Sample 1 72
21 Organic Extractables, Sample I 72
22 Summary of Organic Vapor Analysis From Green Sand
Shakeout, Sample 1 73
-------
LIST OF TABLES (cont.)
Page
23 Quantitative Determination of PNA Compounds Present in
Green Sand Shakeout, Sample 1 74
24 Identities of Major Organic Compounds in Air, Sample 1 .... 76
25 Metal Content of <3 Micron Dust from Green Sand Shakeout ... 82
26 Cyanide Analysis Sample 1; Green Sand Shakeout 83
27 Particulate Loading, Sample 2, Post Scrubber 84
28 Summary of Sampling Data for Scrubber Effluent, Sample 2 ... 84
29 Summary of Organic Vapor Analysis From Green Sand
Shakeout After Wet Scrubbing, Sample 2 85
30 Cyanide Analysis, Sample 2 85
31 Summary of Organic Vapor Analysis from Phenolic Shell
Shakeout, Sample 3 . 87
32 Parti cul ate Loading, Sample 3. . 88
33 Comparison of Organic Effluents 88
34 Comparison of Percent of Each Liquid Chromatograph
Fraction 93
35 Percentage of Each Component in Samples 94
36 Ranges of Decomposition Product Concentrations in the
Effluent Collected from Sealed Flask Experiments 96
B-l Stack Data, Samples 1 and 2 124
B-2 ' SASS Train Data, Sample 1 125
B-3 Velocity Traverse Data and Calculations, Sample 1 126
B-4 SASS Train Data, Sample 2 .127
B-5 Velocity Traverse Data and Calculations, Sample 2 128
B-6 SASS Train Data, Sample 3 129
B-7 LC Analysis Report, Sample 1 130
B-8 Organic Extract Summary, Sample 1 131
B-9 Compound Categories Possible in Different LC Fractions .... 133
B-10 IR Report—Sample 1, Cut LC-1 134
B-ll IR Report—Sample 1, Cut LC-2 134
B-l2 IR Report—Sample 1, Cut LC-3 135
B-13 IR Report—Sample 1, Cut LC-4 135
B-14 IR Report—Sample 1, Cut LC-5 135
B-15 IR Report—Sample 1, Cut LC-6 137
B-16 IR Report--Sample 1, Cut LC-7 137
-------
LIST OF TABLES (cont.)
Page
B-17 Mass Spectroscopy Report—Sample T, Cut LC-1 138
B-18 Mass Spectroscopy Report—Sample 1, Cut LO2 138
B-19 Mass Spectroscopy Report—Sample 1, Cut LC-3 139
B-20 Mass Spectroscopy Report—Sample 1, Cuts LC 4-7 139
B-21 Metal Content of <3 Micron Dust, Sample 1 140-
B-22 LC Analysis Report, Sample 2 141
B-23 Organic Extract Summary, Sample 2 142
B-24 • IR Report—Sample 2, Cut LC-1. 144
B-25 IR Report—Sample 2, Cut LC-2 144
B-26 IR Report—Sample 2, Cut LC-3. . 144
B-27 IR Report—Sample 2, Cut LC-4 145
B-28 IR Report—Sample 2, Cut LC-5 145
B-29 IR Report—Sample 2, Cut LC-6 146
B-30 IR Report—Sample 2, Cut LC-7 . 145
B-31 Mass Spectroscopy Report--Sample 2, Cut LC-1 147
B-32 Mass Spectroscopy Report—Sample 2, Cut LC-2 147
B-33 Mass Spectroscopy Report—Sample 2, Cut LC-3 147
B-34 Mass Spectroscopy Report—Sample 2, Cuts LC 4-7 148
B-35 LC Analysis Report, Sample 3 149
B-36 Organic Extract Summary, Sample 3 15Q
B-37 IR Report—Sample 3, Cut LC-1 152
B-38 IR Report—Sample 3, Cut LC-2 152
B-39 IR Report—Sample 3, Cut LC-3 153
B-40 IR Report—Sample 3, Cut LC-4 153
B-41 IR Report—Sample 3, Cut LC-5 154
B-42 IR Report—Sample 3, Cut LC-6 154
B-43 IR Report—Sample 3. Cut LC-7 -|55
B-44 Mass Spectroscopy Report—Sample 3, Cut LC-1 155
B-45 Mass Spectroscopy Report—Sample 3, Cut LC-2 15g
B-46 Mass Spectroscopy Report—Sample 3, Cut LC-3 ^g
B-47 Mass Spectroscopy Report—Sample 3, Cuts LC 4-7
vn
-------
LIST OF FIGURES
Figure page
1 1978 Density Distribution of Iron Foundries .......... 13
2 Casting Production in the U.S ........ . ........ 14
3 Typical Foundry Production Flow Chart ............. • 15
4 Iron Foundry Process Flowsheet, Emission Sources ....... ]7
5 Typical Green Sand Mold .................... ig
6 Illustration of Magnesium Treatment Methods for Producing
Ductile Iron
7 Hooded Pouring Station .................... 3*
8 Moveable Pouring Hood ............. . ....... 36
9 Balance of Major Solid Materials Entering and Leaving
the Sand Foundry ....................... ,„
10 Temperature Levels in Sand at Various Distances From the
Metal/Sand Interface ................. C/I
11 Quantity of Gases Evolved from a Phenol -formaldehyde
No-Bake Core at Various Temperatures ............. 56
12 Evolution of Gases from Molding Sands ........... rr
bo
13 Sampling of Shake-Out Emissions ................ 51
14 SASS Train Sampling Procedures ................ 53
15 SASS Train Sample Recovery Procedures ............. 54
16 SASS Train Sample Recovery Procedures. .... ........ g5
17 Gas Chromatogram of Organic Effluents, Sample 1 ........ 75
18 Emissions from Shakeout Compared with MATEs .......... 78
vm
-------
ACKNOWLEDGEMENT
This report was prepared for the U.S. Environmental Protection Agency to
present the results of work performed under Contract No. 68-02-2630, Task 2.
The research was conducted in the Energy and Environmental Research
Division and the Chemistry and Life Sciences Group of the Research Triangle
Institute. Mr. Ben H. Carpenter, Head, Industrial Process Studies Section,
served as Program Manager and Dr. Vaniah H. Baldwin, Jr. was the principal
investigator. Dr. Robert Handy directed the chemical analysis effort and Mr.
Frank Phoenix of Entropy Environmentalists directed the plant sampling effort.
The sampling program could not have been accomplished without the generous
help of managers and engineers at certain foundries who wish to remain anonymous,
Mr. William B. Huelsen of the American Foundrymen's Society gave invaluable
assistance in these arrangements.
The assistance of Robert V. Hendriks, EPA Project Officer, is gratefully
acknowledged.
-------
1.0 SUMMARY
This report presents the findings of the environmental effects of iron
castings in organically bound sand molds, with particular emphasis on the
organic vapors produced. The purpose of the study was to investigate the
potential hazards of the process from available literature, acquire new data
by sampling and analysis, and draw conclusions about the environmental accept-
ability of the process.
The iron-castng industry ranks sixth in value added among all manufactur-
ers, with 1,367 foundries that can cast 19 million tons of iron per year.
Sand constitutes 75 percent of the solid waste produced. While the foundry
now appears as a less smoky neighbor, there is still concern for the invisible
organic vapor emissions that are the result of using organic binders and
additives in the sand molds. The works of Bates and Sott revealed the presence
of benzo(a) pyrene and other substances of concern to human health in the
emissions from, iron casting.
The present study began with a review of the chemical literature to
determine the possible chemical products from the pyrolysis of the organic
substances used in foundry molds. This listing indicated that phenolic-
isocyanate and green sand with seacoal have the highest pollution potential of
the commonly used substances." Previous studies indicated that half or more of
the pouring-to-shakeout emissions occur in the shakeout; therefore, this
operation was selected for sampling.
' Three sites were sampled:
1) A duct drawing air from the shakeout of green sand and phenolic-
isocyanate core molding.
2) The exhaust stack from the wet scrubber downstream of the previous
s i te
3) Fugitive emissions in the shakeout room of a phenolic-shell molding
foundry.
-------
The samples were analyzed using methodologies based on the Environmental
Protection Agency's Level 1 protocols. Indications of possible carcinogenic
material triggered a quantitative analysis by gas chromatography-mass spec-
trometry for a standard list of PNA compounds. The dust collected was analyzed
for all the elements by spark source mass spectrometry.
1.1 PARTICIPATE ANALYSIS
The results of particulate analysis were found to be:
Before
scrubbing
After
scrubbing
dust
1-3 urn dust
3-10 urn dust
> 10 urn dust
Total, including
probe rinse
19.2 g/tonne cast
(17.4 g/ton)
213.6 g/tonne cast
(193.9 g/ton)
863.5 g/tonne cast
(783.9 g/ton)
5.874 kg/tonne cast
(5.333 kg/ton)
7.017 kg/tonne cast
(6.37 kg/ton)
19.8 g/tonne cast
(18.0 g/ton)
23.6 g/tonne cast
(21.4 g/ton)
(unmeasurable)
(unmeasurable)
43.4 g/tonne cast
(39.4 g/ton)
Thus, using a wet scrubber, better than 99 percent control is obtained
for total particulates.
1.2 ORGANIC ANALYSIS
The total organic emissions from the shakeout of green sand molds prior
to wet scrubbing was found to be 99.5 percent in the vapor state with the
remainder concentrated on the larger particulates, divided as follows:
-------
On 0-3 urn dust : not measurable
On 3-10 urn dust : 0.42 g/tonne cast
(0.38 g/ton)
On >10 urn dust : 1.32 g/tonne cast
(1.2 g/ton)
In air : 610 g/tonne cast
(554 g/ton)
Cyanide in air : 7.13 g/tonne cast
(6.47 g/ton)
The cyanide concentration was 1.68 vppm, considerably less than the MATE value
of 10 vpmm.
The MATE is the Minimum Acute Toxicity of Effluent and is the
concentration level at which undesirable environmental or health effects
become apparent.
The organic emissions found in the shakeout emissions were tenta-
tively identified and quantified by IR spectrophotometry according to
Level 1 protocol. This produced the following results for the unscrubbed
emissions from green sand casting:
TCO, mg/m3 : 163.8
GRAY, mg/m3 : 9.85
Total Organics, : 173.7
mg/m
-------
Category
Aliphatics
Alky Thai ides
Substituted Benzenes
Halobenzenes
Fused aromatics
Hetero N compounds
Hetero 0 compounds
Hetero S compounds
Alky! S compounds
Nitriles
Aldehydes, ketones
Nitroaromatics
Ethers, epoxides
Alcohols
Phenols
Amines
Amides
Esters
Carboxylic Acids
Sulfonic Acids
/ 3
mg/m
0.72
0.22
2.45
0.24
2.45
0.56
0.10
0.10
0.06
0.01
0.1
0.01
0.1
0.56
0.56
0.56
0.47
0.15
0.46
0.05
Lowest
MATE for
category
mg/m
20
0.1
1.0
0.7
0.0001
to 230
0.1
300
2
1
1.8
0.25
1.3
16
10
2
0.1
1.0
5.0
0.3
0.8
Ratio
cone, found
MATE
< 1
2.2
2.45
< 1
24000
5.6
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
5.6
< 1
< 1
1.5
< I
-------
Low resolution mass spectrometry failed to confirm significant levels
of alkyl halides, carboxylic acids, amines, or nitrogen heterocyclics.
This leaves fused polycyclics and substituted benzenes as possible areas of
concern. Of the substituted benzenes listed in the MEGs, only one of the
18 has a MATE lower than the analysis for the category, namely biphenyl.
This is exceeded by a factor of 2.5 only if it is the entire constituent of
that fraction, which is not probable. The other category of possible concern
is that of fused polycyclics. These were quantified for a standard
set of PNA's by capillary gas chromatography-mass spectrometry (GC-MS).
The PNA levels tested for are well below the MATE values, specifically:
Compound
Naphthalene
Dibenzofuran
Anthracene
Phenanthrene
Fluoranthene
Pyrene
Chrysene
Cone. ~
ug/m
1,484
, 9.8
36.8
7.6
0.7
0.7
15.4
Air, health,
MATE, (jg/m
50,000
--
56,000
1,600
90,000
230,000
2,200
The GC-MS analysis produced a complete mass spectrum for each GC peak,
some of which were analyzed, revealing the 36 compounds that composed 79 per-
cent of the material. The compounds identified in the ventilating air from
the green sand shakeout are listed in the table on the following page. It is
notable that the majority of the compounds are one- and two-ring compounds,
and only one three-ring polycyclic, anthracene, was found. This indicates a
trend toward minimal quantities of large polycyclic compounds. In summarizing
the organic analysis, the level 1 procedure provides no definitive evidence
that the substances present exceed their MATE values in the shakeout effluent
5
-------
Chromatographic
peak no.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Relative
peak height
.34
.18
.14
.49
.11
.18
.12
.12
.19
1.00
.73
.28
.68
.59
.34
.18
.21
.13
.14
.15
.11
Percent
of
sample
4.2
2.2
1.7
6.0
1.3
2.2
1.4
1.5
2.4
12.3
9.1
3.5
8.3
7.2
4.2
2.2
2.6
1.6
1.7
1.8
1.3
Compound
Aniline
Phenol
Cresol isomer
C11H24 1'somer
Naphthalene
Cj-alkyl benzene isomer
^12^26 1'soraer
Dimethyl indan isomer
Dimethyl indan isomer
Cg alkylbenzene isomer
Cg alkylbenzene isomer
C14H30 isomer
Dimethyl indan isomer
(3-methy 1 naphthal ene
Unsaturated Cg alkylbenzene
i somer
Cg alkylbenzene isomer
C"13^28 isomer
a-methyl naphthal ene
Ethylnapthalene isomer
Tri methyl indan isomer
Ethyl naphthal ene isomer
C14H30 isonier
Dimethyl naphthal ene isomer
Diphenylmethane
Dimethyl naphthal ene isomer
Dimethyl naphthal ene isomer
C15H32 isomer
C, alkyl naphthal ene isomer
C3 alkyl naphthal ene isomer
C3 alkyl naphthal ene isomer
C,gH3- isomer
Di-p-tolymethane (tent.)
C,,H isomer
Anthracene-d10
-------
from green sand molding in a well-ventilated foundry. The results can be
viewed as borderline because some categories have concentrations equal or
slightly greater than the lowest MATE in the category, but the large number
of compounds reduces the probability that any specific compound is present
above its MATE level. This indicates that Level 2 analysis is required to
determine if the pollutant levels are above the MATE levels. The analytical
results did indicate, as discussed later, that the pouring process is a more
probable source of high molecular weight polycyclic compounds and should be
given higher priority than the shakeout in future investigations.
1.3 INORGANIC ANALYSIS
The res.pirable portion of the particulate (<3pm) was subjected to spark
source mass spectrometry. Aluminum, magnesium, and silicone dominated the
analysis, which is consistent with the major composition of the dust being
clay and silica. The analysis shows quantities of Si, Cr, and Ni, in the
unscrubbed shakeout emissions greater than the air, health MATE values. The
worst case, Cr, can be held within the MATE level by 98.6 percent removal of
all particulates; however, only 25 percent of the < I urn particulates are
removed by the wet scrubber. Assuming the total particulates from the scrubber
have the same analysis as the < 3 [jm particulates that were analyzed, the
following results were computed:
2
Total scrubber exhaust particulates: 8.92 mg/m
Cr concentration: 1100 ug/g particulate
Cr emission: 9.8 pg/m
Cr air, health MATE: 1 ug/m3
TLV: 100 ug/m3.
The TLV or Threshold Limit Value is the level of contaminants considered safe
for the workroom atmosphere,, as established by the American Conference of
Governmental Industrial Hygienists (ACGIH). Ten hours per day or 40 hours
91
per week exposure is assumed.
This shows that while chromium is safe by TLV standards, it exceeds the
MATE standards, thus it is difficult to definitively assess the situation.
Although small amounts of chromium is sometimes added to the metal, there
was not an identifiable source of chromium at the time of testing. The
presence of impurities in the selected scrap used is always a possibility.
1
-------
An unexpected finding of the inorganic analysis was the presence of
Zr, Ba, La, Ce, Pr, and Nd at levels above a background of other metals not
normally a part of the system (i.e., Zr-140 ppm; Ba-150 ppm; La-28 ppm;
Ce-100 ppm; Pr-4.7 ppm; Nd-17 ppm). These are additives to the magnesium
inoculation alloy and were not expected to show up at the shakeout. This
indicates that the nature of the inoculation smoke should be examined more
closely.
-------
2.0 CONCLUSIONS
This study was a Level 1 assessment which indicated that most emissions
were less than MATE values but some may exceed MATE values, although there
is no definitive proof that is the case. Several areas of concern were
identified, however, such as:
1. Chromium emissions after scrubbing exceed the MATE value although
they are well under the TLV. The source of the chromium could not
be determined.
2. If the sludge from the wet scrubbers is landfilled it may be
classified as hazardous in Ca, Ti, Cr, Mn, Ni, Cu, In, As, Se, Cd,
and Pb, based on particulate analysis. Leach testing will have to
be performed to determine if the sludge is unacceptable for land-
filling under RCRA.
3. The shakeout particulates contained notable amounts of Zr, Ba, Ce,
Pr, and Nd. These are common additives to magnesium inoculation
alloys. The inoculation smoke can be expected to contain much
higher concentrations of these elements.
4. If the shakeout emissions are not collected and scrubbed or other-
wise subjected to pollution control processes, the emissions of
silicon, nickel and chromium exceed the health MATE values.
5. Positive identification of carcinogens in notable quantities will
require level 2 testing. The results of the present study indicate
that the pouring and early cooling stages are more probable sources
than the shakeout.
6. The emissions from the shakeout are a function of the metal tempera-
ture at the time of shakeout, according to a theoretical model
derived in this report. This signals an additional parameter to
be monitored if emissions are monitored, and a possible way of
controlling emissions.
-------
3.0 INTRODUCTION
The foundry industry is basic to an industrial society. Since the 19th
century it has been an important producer of farm implements, water pipe,
and valves. In this century, all power-producing machines, electric motors,
internal combustion engines, and steam turbines are made by the foundries
from castings. Most of these castings are made in sand molds that either
contain organic additives (for casting purposes) or are bound together by
organic polymers. Over 220 million pounds of organic polymers were used by
the foundry industry in 1971, and their use is increasing because of the
better castings obtained.
The organic additives and binders used in iron casting decompose under
the heat of molten iron to produce smoke and vapors of unknown composition.
These were studied in the laboratory by Bates and Scott21 who collected the
emissions and subjected them to partial analysis. Their work identified
benzo(a)pyrene but quantities were not reported.
The objective of this study was to determine if potentially hazardous
organic materials are generated by pyrolysis of mold materials used in iron
casting. The problem of smoke on particulate emissions from foundries has
been reduced by the employment of air pollution control devices, namely wet
scrubbers and baghouses. While foundries were now visually cleaner, the
organic vapor emission levels were unknown and needed determination. Al-
though the initial interest was the organic emissions, following Level 1
protocol resulted in important discoveries about inorganic particulate
emissions.
10
-------
4.0 INDUSTRY DESCRIPTION
In 1976 there were 4,517 foundries in the United States. Of these
2
1,367 were iron foundries (Table 1). Over the past decade the industry has
shown a trend toward fewer but larger foundries with an average annual
attrition rate of approximately 75 plants, most of which are small, closely
held operations. Today, the industry is in a state of transition from one
that has been labor-intensive to one that is capital-intensive. As a result,
the foundry industry now ranks sixth among all manufacturing industries
based on value added by manufacture, increasing from $476 per ton in 1966 to
$1,011 per ton in 1976. A density distribution of U.S. iron foundries is
given in Figure 1. The highest concentration of foundries is in Pennsylvania,
Ohio, Michigan, Illinois, Wisconsin, New York, and Indiana, accounting for
more than half of the iron-casting capacity of the nation. Two-thirds of
the iron foundries are located in metropolitan areas. The decline in foun-
dries has taken place mostly in the smaller metropolitan areas with only a
4
slight change in the larger areas. Figure 2 gives the status of casting
2
production in the United States from 1965 to 1977. As shown on the figure,
there has been an overall decline, some of which has been caused by production
changes as the steel industry perfects methods of sheet metal fabrication.
The major change in the industry in the past decade has been a decline
in the use of the cupola for iron melting, with an increase in the use of
electric induction furnaces and electric arc furnaces. There is also a
continuing trend toward automated casting lines, which adversely affects
many smaller foundries. Chemically bound sand is easy to handle on auto-
mated equipment and the economic pressure to reduce cost, along with automa-
tion, is causing a continual increase in the use of chemically bound sand.
Another major reason for increasing reliance on chemically bound sand is the
declining availability of highly skilled labor and the fact that chemically
bound sand produces a better product, even with less skilled labor.
11
-------
TABLE 1. FOUNDRIES AND IRON FOUNDRIES IN EACH STATE AS OF 1976
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
D. C.
Florida
Georgia
Hawai i
Idaho
n-i * •
nnois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Foundries
90
1
19
43
440
50
101
2
2
60
44
3
6
333
198
77
57
30
24
16
26
141
351
84
16
Iron
Foundries
64
1
3
9
8
12
20
1
1
12
25
2
4
81
75
35
23
13
8
8
10
43
m
35
7
State
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
TOTAL
Foundries
108
3
24
4
29
134
8
282
57
3
465
45
54
386
57
29
1
76
175
19
4
48
53
28
200
4,517
Iron
Foundries
26
£_ w
3
*J
8
2
£.
8
29
]
66
27
2
152
22
12
157
8
12
1
40
66
12
1 £.
4
29
18
13
88
1,367
12
-------
DC:1
>r\
F--
r**3,~vr
Figure 1. 1978 density distribution of iron foundries.
-------
C/3
O
20
I
16
Li.
O 10
en
z
O
—•TOTAL FERROUS
"• GRAY IRON
~*STEEL
"a MALLEABLE
•~* DUCTILE
J_
1965 66 67 68 69 70 71 72 73 74 75 76 77
YEAR
FIGURE 2. Casting production in the U.S.
14
-------
5.0 PROCESS ANALYSIS
The methods of sand casting used by foundries today are sophisticated
compared to those of 50 years ago. Nevertheless, the principal processes
remain the same. A model is made of wood, metal, or plastic, and placed in
a container, which is then packed with sand. Clay and other substances are
added to increase the shape-holding ability of the sand. After this, the
model is removed and molten metal is poured into the cavity and allowed to
cool. Once cool, the mold is broken and discarded leaving a cast iron copy
of the desired object.
Upon cooling from the molten state, cast iron (iron containing 3 to 5
percent carbon) can form seven basic metallurgical structures. Five of
these structures result from the plain metal containing sulfur impurities
and the other two result from desulfurized metal. Ordinary cast iron,
containing sulfur impurities and frequently silicon and manganese, forms
white, pearlitic gray, or ferretic gray cast iron according to the rate of
cooling. The white cast iron can be further transformed into either pearl-
itic, or ferritic malleable forms by heat treatment. If the hot metal is
desulfurized, either pearlitic ductile, or ferritic ductile cast iron is
formed according to cooling rate. The outstanding characteristic of the
"ordinary" gray cast iron is the presence of graphite in the form of carbon
flakes that causes the metal to exhibit brittleness. Graphite is also
present in the malleable and ductile cast irons but in the form of spherical
nodules. In addition to the types of cast iron previously discussed, hybrid
forms are often created for special purposes by varying the cooling rates
involved, sometimes by oil quenching.
Figures 3 and 4 present a flow sheet and a graphic presentation of the
1 4
major operations and equipment involved in the foundry industry. '
5.1 CASTING METHODS
There are two basic casting methods utilized by the foundry industry.
One is to pour the molten metal into the mold and the other is pressure
15
-------
CT)
Finishing
Figure 3. Typical Foundry production flow chart.1
-------
METALUCS
COKE
COOLING AND
CLEANING
SAND
PREPARATION
Figure 4. Iron Foundry process flowsheet, emission sources.
-------
injection of the metal, usually by throwing it into the mold on a centrifuge.
However, the industry refers to casting methods according to the type of
mold used and, sometimes, according to the type of mold binder utilized.
This results in a large number of so-called methods. The methods that will
be discussed herein are (1) green sand, (2) inorganic bound sand, (3) organic
bound sand, (4) permanent molds, and (5) physically bonded molds (sometimes
called the third generation method).
5.1.1 Green Sand
Green sand is the original mold type and is still.the predominant
material in the foundry industry today. Originally, naturally binding sands
or pure silica sand with desirable grain size, shape, and flow properties
were employed with the addition of clay and water as a binder. Later it was
learned that the addition of organic materials to the sand improved the
casting quality.
The term "green sand" is applied when the chief bonding agent is clay,
usually western or southern bentonite (montmorillonite). The clay is
plasticized with about 3 to 5 percent water and organic materials such as
sea coal, wood flour, oat hulls, and substances that are the "pot ends" of
organic chemical production are added in amounts up to 8 percent. The
purpose of the organic addition is to cushion the thermal expansion, provide
a reducing atmoshere, and promote graphite formation at the sand-metal
interface to give a better finish to the metal.
Once the pattern or blank is fabricated half of it is placed in the
bottom (called the drag) of a flask and the green sand mixture is packed
around and on top of the pattern either by hand or hydraulic press. In
similar fashion, the other half of the blank is placed in another flask
(called the cope), filled, and then the drag and cope are put together as a
complete mold (Figure 5). In modern foundries, machines make up the cope
and drag simultaneously at a rate of about one every ten seconds and hy-
draulic pressure is applied through a large number of small metal feet to
compress the sand into the mold. A major disadvantage to this mold is that,
although it can withstand the casting process, it is easily damaged.
18
-------
Sprue pin
Bottom board'
(c)
(a)
Bottom board
Flask
Gate
Bottom board
Figure 5. Typical green sand mold, (a) machined blank,
(b) drag, (c) cope, (d) finished mold.
5.1.2 Inorganically Bound Molds
There are foundries that use plaster of paris, sand and plaster of
paris, or a form of port!and cement mixed with sand to create this small
mold category. The molds that are produced make very high quality castings,
but the manufacturing time involved makes them expensive and, therefore,
limited to speciality work.
The most promising type of inorganic binder in present use is sodium
silicate. When this material is mixed with sand, a solid gel is formed as
carbon dioxide gas is blown through the mold. Mold formation is identical
Q
to the green sand process and is virtually nonpolluting. However, technical
difficulties are involved with the binders because they are too strong and
do not weaken from hot metal addition. Therefore, removal of the mold from
the metal can be difficult.
5.1.3 Organically Bound Sand
The availability of synthetic resin organic binders has resulted in a
19
-------
large number of mold making techniques, some of which are shell molding, hot
box molds, cold set binders, no bake resins, oils, and full mold process.
Table 2 presents a listing of the more popular organic binders.
Shell Molding
In this technique a mold, about 3/4 in. thick, is made in two pieces
which are clamped together forming a shell. Since the shell alone would not
withstand the weight of the molten metal, it is set in a large flask (typi-
cally a small rail car) and surrounded with iron shot for added support
before the iron is poured into it.
Shell molding is used for high precision casting such as small engine
parts. An advantage to this type of mold is that it promotes faster metal //"
cooling, which is metallurgically desirable. Nearly all shell molds are
made from phenol-formaldehyde which requires baking for about one minute.
TABLE 2. ORGANIC CORE BINDERS
OILS
Core oils (oven-baked)
Oil-oxygen (no-bake)
URETHANES
Alkyd isocyanate (no-bake)
Phenolic isocyanate
a. Gassed
b. Ungassed (no-bake)
HOT BOX (heated core box)
Urea-formaldehyde
Phenol-formaldehyde
a. Novalak
b. Resole
Furan
Modified
a. Urea-formaldehyde/furfuryl alcohol (UF/FA)
b. Phenol-formaldehyde/furfuryl alcohol (PF/FA)
c. Phenol-formaldehyde/urea-formaldehyde (FF/UF)
ACID NO-BAKES
Furan
Phenol-'formal dehyde
20
-------
Hot Box Molds
These molds are fabricated of sand bound with thermosetting resins such
as phenol-formaldehyde, and require baking to set the resin. Using this def-
inition, shell molds may also be categorized as hot box if phenol-formaldehyde
is used as the binder. Other modifications such as urea-formaldehyde,
furan, and phenol-formaldehyde/furfuryl alcohol resins are used for hot box
molds. These resins are well established as heat stable polymers that do
not soften, but under extreme temperatures, do degrade and weaken, the ideal
characteristic for a sand binder.
In recent years, however, these molds have become unpopular because of
the time and energy consumption required, the high equipment cost, and also
cold set binders have been found to be time and energy efficient. In some
cases, however, these disadvantages are outweighed because of the strength
of the cores produced by this method. For example, the automotive industry
has found an actual cost reduction and production increase of intricate,
fragile water jacket cores by using furfuryl-phenolic resins and hot box
9
technique.
Cold Set Binders
Cold set binders, developed about 1967, are urethane resins hardened by
passing a catalyst gas (triethyl amine (TEA) or dimethyl ethylamine (DMEA))
through the mold. The mold itself is actually made of two resins, a phenolic
resin and a polyisocyanate mixture which is incorporated with the sand. The
mold making machine clamps together two metal molds shaped so as to cast the
sand mold o,r core desired, and the sand mixture is blown into the mold by a
pneumatic process. The catalyst gas is then blown through the mold and the
resins harden in about 3 to 15 seconds depending on size. The metal molds
then separate and the sand mold is ejected from the machine. This system
is used almost exclusively for core making.
No-Bake Resins
No-bake resins are polymer systems which are catalyzed while mixing
with the sand and harden over a relatively short period of time but suffi-
ciently long to enable the sand to be packed into a pattern to make a mold.
The materials used for this process can be either certain urethane or certain
phenol-formaldehyde resins. The earliest no-bakes were drying oils.
21
-------
Oils
Oils were the earliest form of chemical binder. Core oils were various
oil mixtures that hardened when baked. The drying oils, such as linseed and
tung oil, that are used in no-bake operations are oils that react with
oxygen in the atmosphere and harden. While similar to varnish in composi-
tion, these contained large amounts of lead and cobalt drying catalyst.
Full Mold Process
In the full mold process, the pattern is made of styrofoam using a
standard plastic molding machine. The metal molds are clamped together,
styrofoam beads are poured into the mold, and steam is blown through which
causes the beads to expand and fuse together forming a solid block in the
shape of the cavity. The styrofoam is placed in a flask and either organ-
ically bound or physically bonded sand is packed around it. The completed
mold is sent to the pouring station where the molten metal is poured direct-
ly on top of the styrofoam. The styrofoam either vaporizes or turns to
graphite, which promotes a fine finish, and the metal comes to rest in the
sand mold.12'17
5.1.4 Permanent Molds
From an environmental viewpoint, permanent molds are the ultimate
casting method since there is no pollution involved. In this system, the
mold is made of steel, cast iron, or ceramic and, therefore, there are no
substances to decompose under the heat of the metal. The disadvantages to
this system, however, are that they are expensive and time consuming.21
5-1.5 Physically Bonded Molds
This is the newest casting method and holds the greatest promise for
low environmental effects in the future. Physically bonded molds are molds
in which sand is not bound together chemically.14 Also in this method sand
is not always used; powdered iron can be used instead. An example of this
type of mold process is the ice bonded mold which is used by a company in
England. Wet sand is packed around the pattern halves in the cope and
drag, placed in a freezer and frozen. The mold halves are then removed,
assembled, and cast iron poured in. -It results in no pollution since after
22
-------
the metal is removed, the moist sand which remains can be reclaimed. From
an environmental viewpoint, this system is ideal; however, currently there
is little application beyond a few users. Perhaps its unorthodox nature and
the freezing time required are inhibiting factors. It appeared that this
method more readily lends itself to a small foundry rather than a large,
high production facility, assuming that a practical rapid freeze method
cannot be found.
The term "physically bonded molds" is becoming synonymous with the term
"third generation molds" and includes molds in which the sand is held in
position by air pressure, or powdered iron is used in place of sand and
frozen into position by a magnetic field. In both of these processes, a
styrofoam pattern is made, placed in a flask, and surrounded by sand or
granular iron. In the magnetic process, the flask is placed in a magnetic
field, bonding the iron particles together, and the hot metal is poured on
top of the styrofoam, vaporizing it. After the metal has cooled, the mag-
netic field is turned off and the cast object is removed. In the sand
process, after the sand is placed around the styrofoam pattern in the flask,
it is stabilized by applying a vacuum through vents p.n the bottom of the
14
flask pulling the sand down and packing it tightly. Another variation
utilizes a layer of plastic over the top of the mold giving maximum pressure
from the atmosphere against the sand.
There is also another third generation process called the "V" molding
process.18'19 In this, a sheet of ethylene vinyl acetate 0.002 in. thick,
is heated to its softening point and vacuum molded around the pattern. Sand
is placed on top of this and another sheet of plastic is laid over the top
of the flask. A vacuum is applied to the flask through side vents attract-
ing the two sheets of plastic and compressing and binding the sand. The
flask and mold are removed from the pattern, the two halves are assembled,
and casting proceeds under vacuum. After the metal is cooled, the vacuum is
released and the sand is fluidized and poured out of the mold.
These processes involve no chemical binders and are relatively pollution
free. Although a small amount of polymer material is vaporized, the nature
and quantity is such that the pollutants expected from them is minor compared
with the chemically bonded molds. It has been proposed that either magnetic
or vacuum molding processes can be utilized for any of the desired molding
23
-------
problems within the foundry industry with a few exceptions. Because of the
low capital investment, low pollution involved, and the potential for high
speed production, these practices are recommended for the future.
5.2 SUPPORTING PROCESSES
5.2.1 Pattern Making
As stated previously, all molds are made from patterns of almost any
material. However, most often the material used is aluminum because it is
easy to fabricate and handle, lightweight, and wears well. Sometimes the
aluminum patterns are nickel plated to further increase their wear resist-
ance. All foundries have a small group employed in pattern making, the
environmental aspects of which are similar to a woodworking or metal working
shop.
5.2.2 Sand Processing
Previously, when naturally bonded molding sand was universally used
for green sand molding, the only preparation required was the addition of
water to the sand along with some make-up sand. However, reliance on a
naturally occurring product of highly variable properties does not allow for
high production of precision parts and, therefore, modern foundries no
longer utilize naturally bonded sand. Today, the sand is mixed to order
according to the recipe of the caster. Pure, clean silica sand is sized
and mixed with the desired quantities of specific types of clay, water,
binders, and additives in a device called a sand muller and then conveyed to
the molding units. After the molding is completed, the sand is cooled and
recycled. Lumps, pieces of iron, and other debris are screened out, and
the sand is screened to the desired size range. The reclaimed sand is
analyzed and make-up sand plus other additives are introduced according to
chemical and physical analysis. Then the sand is ready for reuse. In a
typical large foundry about 20 percent of the sand is replaced with new sand
each day. The build-up of carbonaceous materials as well as the production
of fines and other mechanical degradation prevent continual reuse of the
sand.
24
-------
5.2.3 Iron Melting
There are three major methods of iron melting for foundry use—the
cupola, the electric induction furnace, and the electric arc furnace (EAF).
There are other methods most of which involve the reverberatory furnace.
However, this furnace constitutes less than two percent of the industry
although it merits environmentally because of its low particulate emissions.
Cupola
The cupola is a vertical furnace having the appearance of a miniature
blast furnace but distinctly different. The number of cupolas in the United
States is declining despite some claims of economic advantage over the
electric furnaces. The reasons given by foundries for this decline is the
pollution problem and greater operational ease of the EAF and induction
furnace.
In the cupola, scrap metal and coke are top loaded into the furnace and
blasts of air from the bottom burn the coke and melt" the metal. Fluxing
material is also added, producing a slag. The conventional cupolas are made
of sheet metal and lined with refractory brick; a water-cooled cupola is
lined with carbon blocks and has a continual flow of water covering the
outside. The cupola is operated with a blast of hot air at the bottom
similar to the blast furnace and is amenable to many different techniques
for controlling the manner, temperature, and position of the air emission.
There are some instances of successful operation with natural gas injection,
as well as utilization of pure oxygen which has the advantage of reducing
stack gas volume. Because the cupola is charged through a hole in its side,
the manner of operation of the doors in the charging hole determines whether
or not air is mixed with the offgases. If the charging door is open contin-
uously, large amounts of air infiltrate, increasing the volume of gas to be
handled by the air pollution control system. On the other hand, if the door
is closed, insufficient air is introduced to complete combustion of the
carbon monoxide in the offgas. In this case, a common practice is to delib-
erately add adequate air and install an after-burner above the charging hole
to insure the ignition of the carbon monoxide laden offgas.
A typical cupola producing medium strength cast iron from a cold charge
will utilize the following quantities of material: (as percentage of iron
25
-------
input) scrap steel - 42 percent; foundry returns - 58 percent; FeSi - 1.1 per-
cent; FeMn - 0.2 percent; total coke - 14 percent, limestone - 3 percent;
and melting loss - 2 percent. In addition, the following materials are used
in operation:
Refractories, cupola 3.3 kg/metric ton
Refractories, slag skimmer 2.2 kg/metric ton
Cooling water 1.2 m3/metric ton
Water for slag/granulation 0.11 mVmetric ton
Fuel for preheating 2.2 kg/metric ton
As with the blast furnace, the cupola is under continual development.
Coke consumption can be as high as 352 pounds per ton but with hot blast
design, this can be reduced to 150 pounds per ton. Cokeless cupolas have
been designed but are not in common use. Supplementary hydrocarbons and
oxygen enrichment are also under research and development, as well as systems
for recovering the heat from the cupola and utilizing it to heat the entire
factory.
Induction Furnaces
The simplest induction furnace is a cylindrical or cup-shaped vessel
lined with a refractory material and with water-cooled electrical wires
around its circumference. The coil of wire is energized with an alternating
current and the magnetic field set up by this process causes the metal in
the furnace to reach melting temperature. When the metal has melted, the
magnetic fields generated by the exciting coil interact with magnetic fields
generated within the metal by the circulating current. This results in the
metal undergoing a strong stirring action. This type of furnace is referred
to as a coreless furnace because it contains only an electrical coil wrapped
around a cylindrical container.
The channel induction furnace differs from the coreless furnace in that
a tube, positioned above the bottom, passes horizontally through the furnace.
Within this tube there is an iron core wound with wire. The core extends
outside the furnace and loops back making connection with itself. The
channel furnace requires that a continuous circuit of iron or metal exists
around this core within the furnace, and only the iron in the lower portion
26
-------
of the furnace immediately surrounding the channel is heated. Some residual
metal must always be left in the furnace for it to operate.
Induction furnaces are best suited for batch type operations although
some have been recently designed for continuous operation. The coreless
type is better adapted for melting whereas the channel type is better suited
for holding or superheating metal. These furnaces operate at frequencies of
60 and 180 and sometimes up to 1,000 cycles per second. Generally only the
very small furnaces operate at high frequency. Laboratory furnaces of a few
ounces capacity require radio frequency current but the frequency can be
reduced as the size of the furnace increases. Most industrial sized furnaces
operated on 180 or 60 cycles.
The induction furnaces are very efficient, exhibiting very low melting
losses and very high recovery of alloy additions. They are usually charged
with scrap steel and cast iron scrap, foundry returns, and ferrosilicon and
carbon according to the compositional requirements. If channel furnaces or
furnaces containing molten metal are being charged, the charge is dried so
as to prevent explosions that would occur if wet metal was charged into
molten metal. No chemical actions take place in the furnace, so it is not a
refining furnace. After the metal has melted, additions of pelletized coke
are made to adjust the carbon content. Because it is not a refining furnace,
great care must be taken to control the composition of the scrap metal
charged into it to prevent metal contamination. The major pollution problems
that can occur from induction furnaces are those that would result from the
charging of dirty and oily scrap metal. This can be obviated with a hood
system over the furnace which then traps the emissions in a fabric filter
system.
Electric Arc Furnaces
The EAF is considerably different from other types of electrical furn-
aces both in operating characteristics and in environmental concerns. The
furnace consist of a refractory lined, cup-shaped steel shell with a refrac-
tory lined roof through which three graphite electrodes are inserted. As
used in iron foundries, the holding capacities vary from about 500 pounds to
65 tons, with 25 tons being more common size. The roof of the furnace is
removable to allow charging and pouring. The furnace is usually charged
27
-------
with a bottom dump bucket. The roof is replaced and three electrodes,
connected to a system of transformers fed by 3-phase alternating current,
are lowered into the metal. Upon contact, there is a short period of time
during which the electrodes are arcing to various pieces of scrap metal.
Soon a smooth electrical discharge is formed between the electrode and metal
and the melting proceeds smoothly. The distance between the electrodes and
the metal, the voltage, and current parameters are continuously adjusted to
maintain an optimum electrical arc. This arc is a plasma in which reactions
take place, virtually all of which produce air emissions. Iron oxide is
produced and, if zinc is present in the scrap, a zinc ferrite is likewise
produced. The oxides formed in the electrical arc tend to be of the ferrite
structure. At the present time it is normal practice for an air pollution
control system to be utilized with EAFs to capture and filter (baghouses)
the dust produced. When the metal has melted, the carbon content is adjusted
by the addition of petroleum coke or other carbon material. When the metal
is at the desired temperature and composition, the electrodes are raised out
of the furnace and the entire furnace is tilted to pour the metal from it.
It is common practice to add a small amount of calcium carbonate to act as a
flux.
5.2.4 Inoculation
Inoculation is the process of introducing certain alloying elements
into the iron thereby causing the graphite in the iron to form spheroidal
particles resulting in ductile iron. No other metal alloy has had as rapid
an increase in production as ductile iron. Shipments of ductile iron cast-
ings increased from 200,000 tons in 1963 to 2,200,000 tons in 1973. The in-
creased emphasis of high strength to weight ratio in the automative industry
is a major factor in this growth.88
Ductile iron is based on innoculation with magnesium but other elements
such as Ba, Ca, Ce, Nd, Pr, Sr, and Zr are also added. The magnesium may be
added as a wire or block submerged in the molten iron, but increasingly the
practice is to use ferrosilicon alloys containing the magnesium, or porous
blocks of steel turnings impregnated with magnesium. The final cast iron
must have 0.035 percent Mg for the alloying to be effective, but 0.04 to 0.8
percent is added, depending on the chemistry of the metal and the operational
28
-------
nature of the foundry, because of fading. (Since the melting point of iron
is above the boiling point of magnesium, the magnesium added to the iron is
lost in a short period of time. This phenomenon is called fading.) The
effectiveness of inoculation (retained magnesium) fades 50 percent every
five minutes after magnesium introduction until the metal has cooled sub-
89
stantially.
A common method of innoculation is to load the magnesium or magnesium
containing ferrosilicon into a graphite "bell". The bell contains holes and
a rod is placed across the bottom to retain a container of inoculant. This
bell, mounted at the bottom of a vertical graphite rod is then plunged deep
into a ladle of molten iron. A turbulent reaction ensues because the mag-
nesium boils under the heat of molten iron. As much as 65 percent of the
magnesium may be lost in this process, and the Mg vapor that issues from the
iron ignites in air, creating large quantities of smoke. This is presumed
to be MgO, but many other possibilities exist, as will be discussed below.
Numerous methods of inoculation have been tried, and the problem of effici-
ently accomplishing the alloying is still under active investigation. Some
of these are shown in Figure 6. European foundries are trying closed
ladles under pressure to improve efficiency. In most foundries the inocula-
tion smoke is vented through the roof as with other emissions in the melt
shop.
The control of emissions has been recommended by the American Foundry-
pn
men's Society (AFS). However, no references have been found, in this or
other studies, indicating the extent of emission control systems for inocu-
lation in actual use. The AFS book on environmental control shows local
exhaust hoods fitted to cupolas that pour the iron directly into small
ladles, presumably using the pour over technique of adding iron to an empty
ladle containing the inoculant. This would be such an inefficient method of
inoculation that economics would prohibit its use in large scale production.
Other sources have suggested control devices that would be applicable only
to small scale, infrequent inoculation practice.
A. T. Kearney has reported one case of measured inoculation emissions,
which are presented in Table 3. The analysis was reported to them by a
foundry they visited.
29
-------
PRESSURE LADLE
PRESSURE CHAMBER
mium
DETACHABLE BOTTOM LADLE
(MAC-COKE)
INJECTION
TRICKLING-IN (GAZAL)
PLUNGING
POUR-OVER
THROW-IN
PLUNGING
Figure 6. Illustration of magnesium treatment methods for
producing ductile iron.4
30
-------
TABLE 3. MAGNESIUM TREATMENT SYSTEMS EMISSIONS REPORT FOR DUCTILE
IRON PRODUCTION AND GRAY IRON DESULFURIZATIOM4
Iron Treated 30 tons per hour
Inoculant Added - 20-22 pounds per ton Iron
Inoculants Used - MgFeSi - (10% Mg)
75% FeSi
Soda Ash
Emissions Produced - TOO pounds per hour
3.3 pounds per ton iron
Emissions Analysis - 32% MgO
18:7% Fe90-
9.5% CO, J
4.2% S109
2.5% S *
1.1% C
0.6% CaO
Balance NaO
In large operations of ductile iron production the metal is desulfur-
ized before inoculation. This is frequently done by calcium carbide addi-
tions. Failure to desulfurize results in desulfurization by the magnesium,
which can be a very expensive method.
Mold inoculation is practiced to a lesser degree, when possible. In
some cases, a powder of magnesium or its alloys is spooned into the mold
cavity in the drag mold. More elaborate methods involve using "plugs" of
inoculant, made of iron,, magnesium, ferrosilicon and additive elements,
which are anchored into the mold. The mold is specially designed for this
type of casting. Since the inoculation occurs during the casting process,
fading is not a problem so less material can be used.
At the present time, inoculation seems to involve as much art as science,
for procedures that work at one foundry do not work at another because of
variables in operating time, temperature, casting size, and metal chemistry.
The industry's prime concern is the metallurgical result. Environmental
pollution from inoculation is being indirectly attacked by seeking more
efficient methods that would result in reduced need for control. Some have
suggested that control can be effected with hoods and fabric filters. In
the case of very small operations this may be true, but the larger operations
31
-------
are not physically amenable to conventional control techniques and may
require new engineering designs for the inoculation facilities with the
intent of making them amenable to control.
The Nature of Inoculation Smoke
The burning magnesium from inoculation is commonly referred to as MgO.4
According to the chemical literature,3 burning magnesium in air will also
produce:
a. MgQ2 magnesium peroxide,
b. Mg3N2 magnesium nitride.
The fact that magnesium burns in nitrogen, as well as several fire extin-
guishing gases and liquids, is known, and one can expect to find a consider-
able amount of Mg3N£ in the inoculation smoke. This could have adverse
environmental or health effects because on contact with water the magnesium
nitride produces ammonium hydroxide and magnesium hydroxide:
Mg3N2 + 8H20 -» 3 MgO-^O + 2NH4OH.
If this reaction occurs in the lungs or breathing passages, the Mg3N2 dust
would deposit NH4OH (PH > 11.6) and Mg(OH)2 (PH 10.5) which are caustic to
the mucous membranes.
The magnesium oxide, MgO, formed can exist in two forms.13 MgO formed
at "low temperature" will hydrolyze readily by the reaction MgO + H 0 ->•
MgO-H20 (or Mg (OH)2), and the hydroxide dissolves slightly forming a solu-
tion of pH 10.5. This is known to be corrosive to paint. While the alka-
linity may be undesirable, it is conceivable that small quantities of MgO in
the lungs could be eliminated from the body because of its solubility.
When MgO is formed at "high temperatures," (commercially known as "dead
burnt") it does not hydrolyze or react within reasonable times, such as one
year. This suggests that it would be classified as insoluble inhalable
particulate. Which form of MgO is emitted from the inoculation process is
not known.
Magnesium also reacts with oxygen to form the peroxide, MgO?. There is
no data on the quantity of this substance that can be expected to form from
inoculation.
32
-------
Another topic of environmental concern that has not been addressed is
the fate of inoculation additives. The effects of fading are reduced by
adding Ba, Ce, Ca, Nd, Pr. In addition, metallurgical problems with heavy
sections that require up to 3 hours to cool are alleviated by adding Sr and
op
Zr. These metals have been detected in the shakeout smoke, as will be
noted in the sampling analysis section of this report. It is reasonable to
assume that much larger quantities are present in the smoke from inoculation
itself.
5.2.5 Pouring
In nearly all cases, iron castings are made by pouring the liquid
metal into the molds under human guidance. Totally automatic systems have
been designed but are seldom used, even in the large automotive foundries.
Each different job, or type of casting, will require pouring different
amounts of metal into a hole that has different positions. If the gate is
blocked, or other faults occur within the mold during the pouring operation,
an operator can detect such problems visually and stop the metal flow. Such
ability has not yet been programmed into a machine.
In the simplest case, iron is tapped from the cupola or electric furnace
into a small ladle of 1/3 to 1 ton capacity. The ladle usually hangs from
an overhead conveyor controlled by a switch box on or near the ladle carrier.
The pouring man moves the ladle along the conveyor line of moving molds, and
when he has positioned the ladle with respect to the mold, turns a large
steering wheel tilting the ladle and pouring the metal into the sprue hole.
In foundries that do extensive ductile iron casting, the metal is tapped
from the furnace to a desulfurizing ladle, then to an inoculation ladle.
After inoculation the large ladle is transported to a point adjacent to the
pouring station and is used to refill the pouring ladles, several of which
may be in operation at a given time. In foundries that do limited ductile
iron casting, inoculants may be added to the pouring ladle just prior to
tapping the furnace.
Emissions from pouring can be successfully captured by two methods.
The most convenient method for a large foundry is the hooded pouring station,
90
shown in Figure 7. In this type of hood, air is blown down from the front
edge and sucked up by the lower grill. A push pull system utilizing an
33
-------
Ladle
To We*
Scrubber
Figure 7. Hooded pouring station.
90
34
-------
incoming draft from a floor grating which is drawn out by the hood is also
very effective. Smaller foundries can use a portable exhaust hood as shown
in Figure 8.20
Pouring and cooling are areas of concern from an emissions standpoint.
During the pouring operation, the mold and core are usually enclosed in a
flask. Within seconds of pouring, emissions are evolved. A controlled
laboratory test with an uncored, green-sand mold containing 5 percent seacoal,
in which a 30 pound 4 in. cube was cast was performed by Bates and Scott.
The carbon monoxide concentrations peaked at about 1900 ppm after 5 minutes
and the total hydrocarbons maximized at 1225 ppm after 6 minutes. The sand
to metal ratio was 3:1. .,
The same study used green sand molds with various formulations of core
sand. Maximum values were reached after 1 to 5 minutes for carbon monoxide,
1 minute for carbon dioxide, 1 to 5 minutes for methane, and 1 to 6 minutes
for total hydrocarbons. Particulate emissions were 0.0625 grains/scf (142
q 4.
mg/m ) during solidification. Peak particle counts (3x10 ) of 0.35 to
1.0 mm sized particles occured approximately 11 minutes after pouring.
The experiments of Bates and Scott that most closely approximate the
pouring conditions were the sealed flask experiments. The effluent they
collected from flasks after pouring, was analyzed by GC-MS and several
carcinogenic compounds were identified. Unfortunately no quantification was
performed.
Section 9 of this report discusses the findings of RTI's sampling in
terms of the mechanisms involved in the emission of organic vapors from the
casting processes. According to the operative mechanism discussed, the
maximum emission of higher molecular weight (HMW) substances should occur
during pouring and initial cooling, with the release of HMW substances in
shakeout being a function of metal temperature. There are moderating factors:
in the first instance, the major organic vapor emission on pouring will be
from the top'surface of the sand around the sprue hole. The majority of the
gases formed at the sand metal interface will have to pass through the sand
to escape, with the HMW compounds being trapped, as explained in Section 9.
Secondly, large quantities of H2, CO, and CH4 are produced and at the time
of pouring these ignite. The burning gases may be seen for several minutes
after pouring. Since the HMW compounds that escape will be entrained in
35
-------
Crane
Ladle
To Wei-
Scrubber
Flexible
Hose
Hood
Mold
Figure 8. Moveable pouring hood.
36
-------
this release of gases, they will be burned along with the lighter gases,
thereby destroying some of them. Thus the unignited emissions from pouring
are the most probable source of HMW organic emissions.
5.2.6 Cooling
After pouring, on an automated casting line, the molds are conveyed
to a cooling room. In this room the conveyor system is designed to provide
maximum track length, or in terms of operating conditions, time delay.
Cooling time varies from 45 minutes to 2 hours on the automated lines and
may extend to Overnight in small nonautomated foundries. In some places the
cooling occurs in a tunnel rather than a room. No literature data have been
found on cooling times but obviously it will vary with the size of the
casting and the degree to which production is "pushed." Foundries have been
observed operating at twice their design capacity, which means the cooling
time has been reduced from the original design value.
This study has learned, as indicated in Section 9, that cooling time is
a major factor in shakeout emissions. One foundry visited was casting at
less than design capacity and cooling for 2 hours. The shakeout emissions
were wet scrubbed and blown into the cooling room, from which they were
vented through the roof. No noticeable odor was present in the cooling
room. It should be noted that the foundry had an unusually large ventilating
system that-changed the air in the building 20 times per hour. The ventilat-
ing system, however, was a major noise source.
5.2.7 Shakeout
The most elementary method of removing castings from a mold is to
dump the mold, and hook, or pull out, the casting from the sand. When
significant production is required, the molds are automatically inverted and
dumped onto a vibrating grating which shakes out the sand and separates the
casting. The sand falls through the grating and onto a conveyor belt which
carries it to the conditioning and reprocessing system. In some cases the
shakeout can be a long vibrating grate (30 meters), such as for gasoline
engine blocks and heads, where much internal core sand must be removed.
There are many variations of shakeout systems, including heavy screen drums
that rotate batches of castings and long cylindrical perforated cylinders
that tumble the parts and process parts continuously.
37
-------
The shakeout has the potential to generate the most fumes of the many
foundry operations. By the time the mold assembly reaches the shakeout, the
bulk of the thermal decomposition of the mold/core materials has occurred.
The products of thermal decompositon will tend to be lower molecular weight
materials and will vaporize and diffuse away from the hot metal-sand inter-
face into the cooler sand. The physical chemistry of the situation predicts
that some of the organic emissions will condense and adsorb on the cooler
sand of the mold. Most compounds boiling below 100°C will be lost in cooling.
During shakeout, the cooler sand comes into contact with the hot sand sur-
rounding the metal, and the metal itself. This causes a flash boiling,
thereby producing an emission of the pyrolysis products. In addition, there
will be a lesser amount of decomposition (than occurs during pouring) of the
organic constituents. This is discussed fully in Section 9. The experiments
of Bates and Scott showed higher peak hydrocarbon emissions (1500 ppm)
during shakeout than during pouring and cooling, although the average con-
centrations were lower during shakeout. The particulate emissions during
these laboratory tests were 55 percent higher with a 10 fold particle count
increase over those of pouring. Toeniskoetter and Schafer sampled many
foundries for selected emissions from different binder systems.93 Their
results show that the isocyanate concentration is frequently greater at
shakeout than at the pouring station.
5.2.8 Finishing
After castings are removed from the molds the sprues, gates, and
risers must be broken off. If the separate parts of the mold did not mate
perfectly, there may be a "flash" or sharp edge. The final finishing is
done by grinding off these imperfections. The surface of the casting may
also be cleaned by shot blasting.
The emissions from these processes are relatively coarse and easily
controlled by dry mechanical collectors and baghouses.4
38
-------
6.0 WASTE STREAM CHARACTERISTICS
Foundries have long been recognized for their visible air emissions,
and sometimes for their obnoxious odors. In terms of quantity, solid waste
in the form of sand is the major pollutant emitted, but there are many other
emissions. (Table 4). After solid waste, particulate emissions are the
most prevalent with water pollution generally a secondary problem to particu-
lates control. Water that is used to scrub the air picks up contaminants,
most of which can be removed by settling tanks and the remaining soluble
organics are removed by digestion in holding ponds.
6.1 SOLID WASTES
The solid wastes that are produced by a foundry consist of used core
and molding sand, slag and refractories from iron melting, and dust and
other particulates collected by the air scrubbers (Figure 9).
Over 75 percent of the foundry generated solid waste is from the core
making and molding operations with the remainder coming from melting opera-
23-?7
tions and emissions control processes. This waste can be divided into
the following categories:
Refractories
System sand (including molding and core sand dilution)
Core sand (butts and sweepings not entering the
system sand)
Annealing room waste (in malleable iron foundries)
Cleaning room waste
Slag
Coke ash (collected particulates)
Scrubber discharge
Dust collector discharge
Miscellaneous
23~25
Details of the material balances of these wastes have been determined.
Tables 5 through 11 present data on the magnitude of materials movement
77
from three foundries. Foundry 1 is a malleable iron operation using
39
-------
MAJOR
MATERIALS
IN
CORE AND MOLD
MATERIALS:
SAND
BINDERS
ADDITIVES
MELTING MATERIALS:
METALLICS
REFRACTORIES
FUELS
FLUXES
OTHER MATERIALS:
GRINDING WHEELS
SHOT
ABRASIVES
ETC.
FOUNDRY
INTERNAL PROCESS
RECYCLING:
METALLICS
MOLDING SAND
CORE SAND
MAJOR
MATERIALS
OUT
Figure 9.
CASTINGS-PRODUCT
SOLD
SOLID WASTE
TO LANDFILL
USED CORE AND
MOLDING SAND: SWEEPINGS
CORE BUTTS
MELTING WASTE: SLAG,
REFRACTORIES FLUX
SCRUBBERS
DUST COLLECTOR
PARTICULATES,iABHASWES,
SHOT, ETC.
• COMBUSTION GASES,
WATER
Balance of major solid materials entering and leaving the sand
foundry."
40
-------
TABLE 4. CHARACTERISTICS AND SOURCES OF EMISSIONS IN VARIOUS FOUNDRY DEPARTMENTS
21
Department
Molding, Pouring,
and Shakeout
Cleaning and
Finishing
Operation
Molding
Pouring
Cray and
ductile iron
Malleable
Shakeout
Abrasive
cleaning
Grinding
Type
Sand
Dust
Vapor
Core oil vapors
Facing Fumes
Metal oxides
Flouride fumes
Magnesium oxide
fumes
Synthetic binder
smoke and fumes
Sand fines
Smoke
Steam
Dust
Dust
Metal dust
Sand fines
Abrasives
Wheel Bond material
Vitrified resins
Emissions
Concentration
Light
Heavy
Heavy
Light
Heavy
Heavy
Moderate
to heavy
3 to 5 gr/
cu ft
Heavy
Heavy
3 to 5 gr/
cu ft
3 gr/cu ft
and up
5 gr/cu ft
and up
3 to 5 gr/
cu ft
0.5 to 2 gr/
cu ft
Light
Light
Particle Relative
Size Control -
(Microns) lability
Coarse Easy
Moderate
Fine to
medium
0.01 to 0.4
50%- 2 to Moderate
15
0.01 to 0.4
50%- 2 to
15
50%-2 Easy
to 15
Above 7 Medium
Fine to
medium
50%- 2 to 7
Fine
50%- 2 to 15
_ , ^*rt«i"
Relative
Cost
Low
Medium
Medium
Low
Low
-J m iur{ I
-------
TABLE 4. (cont'd)
ro
Department Operation
Annealing and
heat treating
Painting
spray and dip
Sand Conditioning New sand storage
Sand handling
system
Screening
Mixing
Drying and
reclamation
Coremaking Sand storage
Coremaking
Baking
...
Type
Oil vapors
Volatile fumes
Paint spray carryover
Water spray carryover
Fines
Fines
Steam
Fines
Fines
Flour
Bentonites
Sea Coal
Cellulose
Dust
Oil vapors
Sand fines
Flour
Binders
Sand fines
Dust
Vapors
Smoke
Emissions
Concentration
0.5 to 2 gr/
cu ft
3 to 5 gr/
cu ft
3 to 5 gr/
cu ft
3 to 5 gr/
cu ft
3 to 5 gr/
cu ft
Moderate
Moderate
Moderate
Moderate
1/2 to 2
gr/cu ft
Heavy
3 to 5 gr/
cu ft
Heavy
Light
Particle*
Size
(Microns)
0.03 to 1
50%-2 to 7
50%- 2 to 15
50%- 2 to 15
50%- 2 to 15
50%- 2 to 15
Fine to
medium
Fine to
medium
Fine to
medium
Fine to
medium
50%- 7 to 15
0.03 to 1
Fine
50%- 7 to 15
Fine to
medium
Fine to
medium
Relative
Control-
lability
Moderate
Easy
Moderate
Moderate
Easy
Easy
Easy
Moderate
Moderate '
Easy
Relative
Cost
Low
Low
High
Medium
Low
Medium
Medium
High
Medium
Medium
^Represents the view of Bates and Scott, reference 21.
-------
TABLE 5. POUNDS OF NEW MATERIAL PURCHASED PER YEAR BY CATEGORY
,22
Foundry
A. Refractories
B. Sand used directly
in molding system
1 . New Sand
2. Clay
3. Carbon
Subtotal
C. Sand used as Cores
1. Shell Sand
2. Oil Sand
3. No-Bake
4. C02 Sand
Subtotal
Total Sand Binder
and Additives
D. Metal
E. Miscellaneous
F. Annealing Room
G. Cleaning Room
1. Grinding
2. Steel Shot
3. Other
Subtotal
H. Slag Floculant
I. Flux
J. Scrubber Line
K. Coke
Other
TOTAL
Malleable
1
200,200
3,492,000
1,012,800
387,300
4,892,000
558,000
2,243,800
2,801,800
7,693,000
27,805,000
25,800
220,000
13,800
49,100
5,400
68,300
38,900
101,800
36,153,800
Ductile Iron
2
728,100
20,546,000
3,677,700
734,300
24,938,000
3,976,000
4,076,000
8,052,000
32,990,000
63,209,000
129,000
126,000
255,000
1,396,000
5,658,000
32,500
8,672,000
1,200
112,941,800
Gray and
Ductile Iron
3
530,000
4^25,800
2,160,000
1,584,000
8,469,800
1,800,700
15,200,600
3,540,000
2,688,000
23,236,300
31,707,100
122,205,000
29,300
216,000
6,000
251,300
8,544,000
400,000
27,516,000
185,153,900
43
-------
TABLE 6- PERCENTAGE OF MATERIAL PURCHASED BY CATEGORY EXCLUDING METAL MELTED22
Foundry
A.
B.
C.
Total
and
E.
F.
G.
H.
I.
J.
K.
Refractories
Sand used directly
in molding system
1. New Sand
2. Clay
3. Carbon
Subtotal
Sand used as Cores
1. Shell Sand
2. Oil Sand
3. No-Bake
4. C02 Sand
Subtotal
Sand Binder
Additives
Miscellaneous
Annealing Room
Cleaning Room
1. Grinding
2. Steel Shot
3. Other
Subtotal
Slag Floculant
Flux
Scrubber Line
Coke
TOTAL
Malleable
1
2
41
12
4.
58,
6.
26.
33.
92.
1.
2.
0.
0.
0.
0.
0.
100.
.40
.83
.13
.64
.60
,68
,87
55
15
53
64
17
59
06
82
46
00
Ducti
1
41
7
1
50
7.
8,
16.
66.
0.
0.
0.
2.
11.
0.
17.
100.
le Iron
2
.46
.31
.39
.48
.14
.99
.20
.19
,33
26
25
51
81
38
07
44
00
Gray and
Ductile Iron
3
0.84
7.
3.
2.
13.
24.
5.
4.
34.
50.
0.
0.
0.
0.
13.
0.
34.
100.
.51
,43
.52
,46
.16
62
27
05
37
05
34
01
40
57
64
18
00
44
-------
TABLE 7. POUNDS OF NEW MATERIAL CONSUMED ANNUALLY PER TON OF METAL MELTED
(BASED ON NEW PURCHASES)
Foundry
A.
B.
C.
Total
and
D.
E.
F.
G.
H.
I.
J.
K.
Refractories
Sand used directly
in molding system
1 . New Sand
2. Clay
3. Carbon
Subtotal
Sand used as Cores
1. Shell Sand
2. Oil Sand
3. No-Bake
4. C02 Sand
Subtotal
Sand Binder
Additives
Metal
Miscellaneous
Annealing Room
Cleaning Room
1. Grinding
2. Steel Shot
3. Other
Subtotal
Slag Floculant
Flux
Scrubber Line
Coke
TOTAL
Malleable
1
14.40
251.18
72.85
27.85
351.88
40.14
161.40
201.54
553.42
2000.00
9.17
15.82
0.99
3.53
0.39
4.91
2.80
600.52
Ductile Iron
2
, 23.04
649.51
116.39
23.20
789.10
125.80
128.96
254.76
1043.86
2000.00
0.04
4.08
3.99
8.07
44.17
179.03
1.03
274.39
1573.63
Gray and
Ductile Iron
3
8.67
77.34
35.35
25.93
138.62
29.47
248.90
57.94
43.99
380.30
518.92
2000.00
0.48
3.54
0.10
4.12
139.83
6.55
352.13
1030.22
45
-------
TABLE 8. ESTIMATED POUNDS OF MATERIAL TO LANDFILL PER YEAR BY CATEGORY22
Foundry
A.
B.
C.
D.
E.
F.
G.
H.
I.
J.
TOTAL
Refractories
System Sand
1. Molding Sand from
New Material
2. Degraded Shell
3. Degraded C0?
4. Degraded Oil
5. Degraded No-Bake
Subtotal
Core Sand Total
1. Core Butts
1. Core Room
Sweepings
Subtotal
Total Sand
Annealing Room
Waste
Cleaning Room Waste
1. Grinding
2. Steel Shot
3. Other
Subtotal
Slag
Coke Ash
Scrubber Discharge
Dust Collector
Miscellaneous
Malleable
1
200,200
1,924,100
195,300
617,600
503,000
6.240,000
1,315,900
250,000
1,565,900
7,805,900
220,000
13,800
49,100
5,400
68,300
480 , 000
100,000
25,200
8,899,600
Ductile Iron
2
728,100
23,600,000
6,623,200
30,222,200
1,168,800
260,000
1,428,800
31,652,000
1,205,900
5,460,000
882,800
39,928,800
Ductile Iron
3
530,000
20,351,600
382,000
570,300
3,226,700
751,000
25,281,600
4,929,900
1,790,400
6,720,300
32,001,900
29 300
t- «/ } +J\J\J
216,000
6,000
251,300
7,968,000
2,190,000
1,032,000
4,800,000
48,773,200
46
-------
*22
TABLE 9. ESTIMATED PERCENTAGE OF MATERIAL TO LANDFILL PER YEAR BY CATEGORY
Foundry
A. Refractories
B. System Sand
1. Molding Sand
Materials
2. Degraded Shell
3. Degraded Oil
4. Degraded C02
5. Degraded No-Bake
Subtotal
C. .Core Sand Total
1. Core Butts
1. Core Room
Sweepings
Subtotal
Total Sand
D. Annealing Room
Waste
E. Cleaning Room Waste
1. Grinding
2. Steel Shot
3. Other
Subtotal
F. Slag
G. Coke Ash
H. Scrubber Discharge
I. Dust Collector
Discharge
J. Miscellaneous
TOTAL
Total Sand Percentage
Excluding Slag, Coke
Ash and Refractories
Malleable
1
2.25
55.34
2.19
5.65
6.94
70.12
14.79
2.81
17.60
87.72
2.47
0.16
0.55
0.06
0.77
5.39
1.12
0.28
100.00
95.0
Ductile Iron
2
1.82
59.11
16.59
75.70
0.65
2.93
3.58
79.28
3.02
13.67
2.21
100.00
96.3
Gray and
Ductile Iron
3
1.09
41.72
0.78
6f* *\
.62
1_ «.
. 17
1.54
51.83
10.11
3.67
13.78
65.61
0.06
0.44
0.01
0.51
16.34
4.49
2.12
9.84
100.00
83.0
*This table is expressed as a percentage of Table 8 adjusting to exclude
losses resulting from processes such as coke conversion, etc.
47
-------
TABLE 10- ESTIMATED POUNDS OF MATERIAL TO LANDFILL PER TON OF METAL MELTED22
Foundry
A.
B.
C.
D.
E.
F.
G.
H.
I.
J.
TOTAL
Refractories
System Sand
1. Molding Sand -
New Material
2. Degraded Shell
3. Degraded Oil
4. Degraded C0?
5. Degraded No-Bake
Subtotal
Core Sand Total
1. Core Butts
1. Core Room
Sweepings
Subtotal
Total Sand
Annealing Room
Waste
Cleaning Room Waste
1. Grinding
2. Steel Shot
3. Other
Subtotal
Slag
Coke Ash
Scrubber Discharge
Dust Collector
Discharge
Miscellaneous
Malleable
1
14.40
354.19
14.05
36.18
44.42
112.63
94.65
17.98
112.63
561.47
15.82
0.99
3.53
0.39
4.91
34.53
7.19
1.81
640.13
— ==
Gray and
Ductile Iron Ductile Iron
2 3
23.05 8.67
537.16 333.08
209.57 6 25
•"*-<^*v* \j . t*+j
52.81
9.33
12.29
45.21 109.98
36.98 80.68
8.23 29.30
45.21 109.98
791.94 523.74
0.48
3.54
0.10
38.16 4.12
172.76 130.50
27.93 35.85
16.89
78.56
1053.84 798.33
48
-------
TABLE 11. ESTIMATED POUNDS OF MATERIAL TO LANDFILL PER TON OF METAL SHIPPED22
Malleable Ductible Iron
Foundry 1 2
A.
B.
C.
D.
E.
F.
G.
H.
I.
Refractories
System Sand
1. Molding Sand -
New Material
2. Degraded Shell
3. Degraded Oil
4. Degraded (XL
5. Degraded No-Bake
Subtotal
Core Sand Total
1. Core Butts
1 . Core Room
Sweepings
Subtotal
Total Sand
Annealing Room
Waste
Cleaning Room Waste
1. Grinding
2. Steel Shot
3. Other
Subtotal
Slag
Coke Ash
Scrubber Discharge
Dust Collector
Di charge
J. Miscellaneous
TOTAL
40.25
989.76
39.26
101.11
124.14
1254.27
264.50
50.25
314.75
1569.08
44.22
2.77
9.87
1.09
13.73
96.48
20.10
5.06
1788.86
52.80
1711.32
480.27
2191.59
84.75
18.85
103.60
2295.19
87.44
395.92
64.02
2895.36
Gray Iron
Ductible Iron
3
17.96
689.68
12.94
109.34
19.33
856.74
167.06
60.67
227.73
1084.45
0.99
7.32
0.20
8.51
270.02
74.22
34.97
162.67
1652.83
49
-------
induction melting; Foundry 2 produces gray and ductile iron using basic
practice cupola melting; and Foundry 3 also produces gray and ductile iron
using a cupola for primary melting and duplexing into induction furnaces.
6.2 PARTICULATE EMISSIONS
The effect of cupola emissions on the surrounding environment caused
serious examination of particulate emissions by the Public Health Service in
1968 and the A. T. Kearney Co. in 1971.4 At that time the major furnaces in
operation were cupolas and EAFs. It was determined that 10.4 kg/metric ton
(20.8 Ib/ton) of particulate emissions were produced by the cupolas and 6.9
kg/metric ton (13.8 Ib/ton) from the EAFs. There are no reasons for these
emission factors to be different today, but the emissions to the environment
have been reduced by the addition of air pollution control devices on the
cupolas and EAFs and also some foundries have changed to the induction
furnace. When charged with clean metal, the induction furnace produces
virtually no emissions.
Particulate emissions have been measured in a laboratory apparatus by
Bates and Scott, , whose data are presented in Table 12. Interpretation of
their data requires care. In the first instance, as revealed by the columns
of cumulative summation (summed by RTI), for particles greater than 0.54 urn,
the total mass of particulate from pouring exceeds that from the shakeout.
Bates and Scott also determined the dust loadings, over the 30 minute cool-
ing interval after pouring and the 25 minute interval after shakeout. This
o
exhibited an average of 142 mg/m of pouring and cooling emissions and 221
mg/m of shakeout emissions, 56% higher than the pouring emissions. An
optical particle counter was used to determine the time profile of dust
concentration from 0.35 to 1.00 micron particles. This showed a peak concen-
4 *->
tration of 3 x 10 particles per cubic centimeter after pouring and 3 x 10
particles/cm after shakeout. These laboratory results coincide in princi-
ple with A. T. Kearney's estimates of shakeout emissions (32 Ib/ton melt or
16 kg/tonne) being greater than pouring emissions (5.1 Ib/ton melt or 2.55
kg/tonne).
AS a result of the visible nature of particulate emissions and the
imposition of environmental control regulations, most foundries have installed
particulate control systems. These systems do not control organic
50
-------
TABLE 12. PARTICLE SIZE DISTRIBUTIONS OF GREEN SAND
EMISSIONS FOR 4" CUBE PATTERN21
Size
(microns) Mass (g)
Less -than 3.98
0.54
0.54-0.83 8.35
0.84-1.34 23.01
1.35-2.67 16.69
2.68-4.14 1.86
4.15-6.08 .97
6.09-8.95 .53
8.96-14.36 .40
More than
14.36 .68
TABLE 13.
Component
Organic carbon (mg/1)
COD (mg/1)
Phenol (ng/1)
Cyanide (ug/1)
Sulfate (mg/1)
Fluoride (mg/1)
Iron (mg/1)
Zn (mg/1)
Ni (mg/1)
Cu (mg/1)
pH
Pouring
I % of Total
3.98 7.0
12.33 14.8
35.34 40.7
52.03 29.5
53.89 3.3
54.86 1.7
55.39 0.9
55.79 0.7
56.47 1.2
RANGES OF POLLUTANTS
Foundry Leachate
4-185
25-1,100
12-400
20-80
30-1,200
3-120
0.1-0.5
0.1-15
0-0.6
0.02-1.6
7.2-10.0
Shakeout
Mass (g) I
5.14 5.14
2.28 7.42
1.36 8.78
0.36 9.14
0.56 9.7
0.24 9.94
10.88 20.82
0.34 21.16
0.28 21.44
IN SELECTED WASTES22
Urban Landfill
Leachate
250-28,000
100-51,000
—
___
25-1,500
—
200-1,700
1-135
0.01-0.8
0.1-10
4-9
% of Total
24.0
10.6
6.3
1.7
2.6
1.1
50.7
1.6
1.3
Septic Tank
Effluent
25-200
250-1,000
0-300*
-— —
10-600
0-10
0-20
0.15*
0.02
0.1*
6.8-8.5
*Municipal Wastewater Effluents
51
-------
vapor emissions however, and that is a problem of concern. Some foundries,
especially high capacity companies operating in densely populated areas,
have installed chemical scrubbers. These not only reduce pollution but also
allow the air to be recycled within the plant, which in some cases saves
energy. Chemical scrubbers are not in significant use and add to the eco-
nomic burden on a company.
Further discussion of organic emissions to the air is presented in
Section 6.4.
6.3 WATER EFFLUENTS
The only effluents from foundries are indirect, i.e., resulting from
the air pollution control systems. Larger foundries remove the sand and
dust from the scrubber discharge in a clarifier tank and landfill it. The
remaining water goes to a settling pond and often flows from the pond to a
river. Sometimes some of the pond water is recirculated to the scrubbers.
Although no specific data was found, it is known that there is a problem
with phenols in foundries using phenol-based chemical binders, unless their
ponds provide adequate holding time for biological action.
The major source of industry water pollution is in the form of leachate
from discarded sand. An extensive study undertaken by the American Foundry-
men's Society showed that the major emission occurs within a 1-2 year period.
Table 13 is a comparison of the pollutant ranges for selected wastes and
Table 14 is a summary of the AFS laboratory analyses.22
6.4 POTENTIAL POURING AND SHAKEOUT DISCHARGES
The major concern of the sampling effort undertaken during this study
was the determination of the nature and quantity of discharges resulting
from the pyrolysis of the organic materials used in sand casting. The
results of a literature study presented in this section and Appendix A and
indicate that environmentally undesirable organic compounds could be released
as a consequence of using organic binders and additives in the molds.
When molten iron is poured into a sand mold, the temperature reached by
the sand varies according to the distance from the sand-metal interface.
Figure 10 presents time-temperature curves for the metal and sand at various
distances from the metal determined in a laboratory study of clays.28
52
-------
OJ
( )-Estimate value
t4- -Increase/decrease
•* - Steady
TABLE 14. LYSIMETER RESULTS-18 SIMULATED MONTHS
22
Component.
Organic carbon (mg/1)
COD (mg/1)
Phenol (ug/1)
Cyanide (ug/1)
Fluoride (mg/1)
Sul fates (mg/1)
pH Range
max
14
75
25
--
3
30
Foundry
1
1 yr
5
30
14
--
--
--
7.6-8.0^
Concentrations in Leachate/Foundries
Foundry
18 mo
4
25
12
--
--
--
max
31
240
78
80
32
1220
2
1 yr
15
100
16
—
25
—
8.0-8.8-*
18 mo
13
90
15
<20
20
(800)
max
185
1100
52
<20
3
78
Foundry
1 yr
35
260
18
— — —
_ _ —
—
7.3-8.0^
18 mo
21
260
15
<20
___
— — —
-------
3000
u.
o
cc
i
Ul
Q.
Ill
J-
2000
1000
10
20 30 40
TIME-MINUTES
50
Figure 10. Temperature Levels in Sand at Various Distances from the Metal/
Sand Interface' (Reprinted from AFS Transactions, 1976)28
It can generally be assumed that organic compounds will begin to decom-
pose above 400° C. Thus, binders and additives will undergo some degree of
thermal decomposition at the sand-metal interface and for a distance of 1.9 -
to 2.5 cm (3/4 to 1 in) away from the interface. Some of the decomposition
products may be gaseous at room temperature, 25° C (77° F) and will pass
through the sand escaping into the atmosphere. Other pyrolysis products
will pass into the cooler sections of the sand and condense to solids or
liquids. Examination of Figure 10 reveals two temperature arrests. The top
curve, for metal, exhibits a temperature arrest just above 1093° C (2000° F),
which is the freezing point of the metal. Once the metal is frozen the
temperature declines further. The sand temperature (other curves in Figure 10)
exhibits an arrest at 100° C (212° F) 2.5 cm (1 in.) from the metal surface.
This temperature is the boiling point of water and represents the drying of
the sand-clay-water mixture. Unfortunately, data are not available for sand
54
-------
temperatures at distances greater than 1 in. from the metal surface, but
thermodynamic principles predict that at greater distances the 100° C
(212° F) thermal arrest will last longer and at even further distances it
will dictate the maximum achievable temperature. Therefore, in large molds
there is considerable amount of material available as a condensing receiver
for pyrolysis products. The pyrolysis products will condense and be
"stored" on the cooler sand surrounding-the metal, as discussed in Sec-
tion 9.
When the mold is shaken out and the cooler sand comes into contact with
the warmer sand and metal, condensed pyrolysis products will be boiled off,
forming a second emission.
In one laboratory study, the quantity of gases involved from a no-bake
core was investigated at various temperatures. Figure 11 shows the results
30
for a phenol-formaldehyde resin and a toluene sulfonic acid catalyst. A
molding sand containing both Western and Southern bentonite as well as
seacoal was tested at 1010° C (1850° F) and emitted gas as shown by the top
curve of Figure 12.30 Although base sands are not generally considered as
emission sources, small quantities of gas were evolved from Illinois silica
sand (=1 cm3/g) and silica sand mixed with dolomite (= 7 cm /g) at 1010° C
O A
(1850° F) during laboratory experiments. The only quantitative literature
data available on organic emissions was that of Bates and Scott. In tests
with green sand molds they found total hydrocarbons to peak at 1200 ppm
after pouring and 1500 ppm after shakeout. On the other hand, the time
average emissions reported for hydrocarbons was 1780 ppm for pouring and
640 ppm for shakeout.
6.5 DECOMPOSITION PRODUCTS OF SUBSTANCES USED IN MOLDS AND CORES
Moldmaking involves the use of organic and inorganic chemical addi-
tives. These substances can pyrolyze or decompose during use of the mold.
The decomposition products may react to produce further products. The high
temperature that these products may attain and their exposure to oxygen in
the exit gases are important in determining the final pollutant composition
in any particular case.
Most binders are blends of several substances that, together meet
desired processing characteristics. Many formulations are proprietary,
55
-------
o
p
O
tu
£50
8
= 40
IX
§30
20
UJ 10
to
<
o
0
-5
NO-BAKE CORE
PF/TSA
1600
T?oo—
I20O
05 1.0 1.5 2.0 2 4
TIME-MINUTES
10
12
Figure 11.
Quantity of gases evolved from a phenol-formaldehyde no-bake
core at various temperatures (in °F)?0 (reprinted from AFS
Transactions, 1976).
>60
(=50
o
UJ
tc.
§40
o 20
5 IO
lu
CO
o
O.5
!.5 20 2 4
TIME-MINUTES
10
12
Figure 12. Evolution of gases from molding sands30 (reprinted from AFS
Transactions, 1976)
56
-------
nevertheless some 46 substances are reported as components of currently used
binders (including complex mixtures such as pitch).
A study was made of the chemical literature to determine the known
pyrolysis products from chemicals used in moldmaking. Appendix A is a
complete listing of the findings of this study. A listing of the pyrolysis
products expected from the resins used by the foundries sampled is given in
Table 15.
TABLE 15. PYROLYSIS PRODUCTS OF SOME BINDER MATERIALS
Substance
Decomposition Products
Phenol-Formaldehyde
Phenolic Resins
(Novalak and Resole)
53,54
At 620° C:
Carbon monoxide and dioxide
Hydrogen
Methane
Phenol
Formaldehyde
Ammonia
Hydrogen cyanide
Acetylene
Ethylene
Ethane55
Same as phenol-formaldehyde plus:
Allene
Methyl acetylene
Propylene
Acetaldehyde
Methyl chloride
Acrolein
Acetone
Propionaldehyde
Vinyl chloride
Ethyl chloride
Cyclopentadiene
(continued)
57
-------
TABLE 15. (cont'd)
Substance
Decomposition Products
Phenolic Resins (continued)
Phenol Urethane
Benzene
Methyl eyelopentadiene
Toluene
Cresols
Methylenediphenol
^2 phenols
Ethylene diphenol
CUHo phenol
^fi
Propene
Acetylene
Carbon monoxide and dioxide
Ethane
Ethylene
Hydrogen
Methane
The nitrogen in the isocyanate should
yield:57
Ammonia
Simple amines
Aniline
Hydrogen cyanide
The phenolic component should produce:
Formaldehyde
Substituted phenols
58
-------
7.0 ENVIRONMENTAL DATA ACQUISITION
Reviewing the literature on the environmental aspects of foundries
reveals incomplete evaluation of the emission of organic chemicals by chemi-
»
cal binders, although laboratory studies have been performed verifying that
21
a potential problem exists.
7.1 SAMPLING AND ANALYTICAL STRATEGY
Two decisions were made prior to performing environmental tests at a
foundry, namely; which operation to test and which chemical formulation to
test. Discussions with the American Foundrymen's Society, and the study
presented in Section 6.5 identified five process areas and five molding
systems as candidates for environmental sampling. The process areas are
pouring and cooling, shakeout, return sand belts, coke ovens, and hot box
and shell coke making. The chemical formulations of concern are seacoal,
isocyanate, phenol-formaldehyde, polyphosphate esters, and polystyrene as
®
used in the Full Mold system.
Considering the large quantity of pollutants estimated to be produced
from shakeout, the relative ease of sampling and sampling cost, the shakeout
was selected as a suitable site for measuring organic emissions.
The phenolic-isocyanate and seacoal systems were selected due to their
common use and potential for pollution. This was pursued by sampling an
operation that used phenolic-isocyanate cores in green sand molds with
seacoal added. The second system selected was a shell molding foundry using
phenol-formaldehyde binder.
The philosophy of the phased approach developed by the Process Measure-
ments Branch of the Industrial Environmental Research Laboratory at Research
Triangle Park, N.C. was employed as a guide in the sampling and analysis.
The Level 1 Procedure Manual outlines this approach and describes the Level 1
sampling and analytical techniques. The goal of Level 1 sampling and analy-
sis is to identify the pollution potential of a source in a quantitative
manner within a factor of ±2 to 3. This does not require a statistically
59
-------
representative sample. The sample is acquired with the Source Assessment
Sampling System which collects particulates by size range and removes organic
and inorganic vapors from the air.
A more sensitive although not comprehensive analysis was planned if the
Level 1 analysis indicated possible PNA compounds, which did occur. Other-
wise the analytical techniques were as described in the Level 1 manual.
7.2 TEST SITE SELECTION
The selection of sampling sites was based on the binders used, the
level of air pollution control employed, and permission to sample. The AFS
suggested possible sites and the companies contacted were cooperative and
friendly. The preferred sample site experienced a change in level of opera-
tion which necessitated replanning and selection of an alternate site. Two
foundries were selected.
Foundry A is a large modern installation producing ductile iron
castings. After melting, the iron is desulfurized, then inoculated
by the magnesium plunging technique, and transferred to the pouring
ladles.
The molding lines are automated, producing a mold every 12 seconds on
each line. The green sand drags are fitted with phenolic isocyanate cores
prior to placement of the copes. After pouring the molds make a 47 minute
tour of the cooling room and are then "punched out" onto a vibrating grate
to separate the sand from the castings. The "punch out" shakeout operation
(hereafter referred to as shakeout) is completely enclosed and air is drawn
through it by a 32 inch duct to a 30,000 cfm wet Ventri-Rod™ scrubber made
by Riley Environeering Inc. Three independent scrubber systems are used on
each molding line, with one dedicated to the shake out. Figure 13 shows the
general nature of the structure and the sample points. Samples one and two
were obtained at Foundry A.
Foundry B is a shell molding foundry using phenol formaldehdye bound
sand sheTTTmbunted in boxes and surrounded with iron shot. The foundry has
virtually no free floor space except a minimum amount for fork lift trucks
to transport materials. The air control system is mostly general ventila-
tion. The shakeout room is a large room in which the railcars are inverted
135°, dumping the contents onto a shakeout table. Exhaust fans are located
60
-------
v
V
Sample 2
Point
SASS
Roof
A
Fan
Ventri-Rod Scrubber
Sample 1
Point
Floor
Hood
Punch Out
Mold Box
Return
Sand Belt
Casting
- Vibrating Grate
" Sand
Figure 13. Sampling of Shake-out Emissions.
61
-------
at a considerable elevation in the room's wall and are essentially inacces-
sible for sampling purposes. The room has an open door and the emitted
smoke occasionally took that exit. Fugitive sampling was all that could be
accomplished at that location, but the density of the smoke in the room was
such as to make observation of the process difficult, leading to the conclu-
sion that a reasonable quantity of organic vapors could be obtained. Sample 3
was obtained in the shake out room of Foundry B.
7.3 SOURCE ASSESSMENT SAMPLING SYSTEM ACQUISITION OF SAMPLES
The sample were acquired with the Source Assessment Sampling System,
commonly called the SASS train, built by Acurex Corporation. This unit
draws in air through a nozzle, at a velocity matching that of the stream
being sampled, and conveys it via a heated tube to a series of three cyclones
in an oven. The cyclones separate the >10M, >3|j, and >1M particulates. The
sample is then passed through a fiberglass filter to remove the <1M particu-
lates, and then is cooled and passed through a cartridge of XAD-2 resin to
adsorb organic materials. After the organic vapors are removed, the collect-
ed air passes through a series of reagent bubblers to remove inorganics.
All reagents and procedures were according to the recommended practices
found PB-257850, IERL-RTP Procedures Manual Level 1 Environmental Assessment
except that a NaOH bubbler was used for determining cyanide. Figure 14 is
the flow scheme showing steps taken in the sampling procedure, and Figures 15
and 16 show the sample recovery procedures.
Foundry A had pre-existing ports on the roof stacks for the SASS probe.
The company installed ports in a duct drawing air from the shakeout hood to
enable traverse measurements and sampling upstream of the scrubber. Obtain-
ing the proper distance downstream from a bend resulted in the sampling
probe being located 8 feet above the floor. Figure 13 shows the sampling
points relative to the process. Sampling was at a single point in the ducts
at a flow rate through the SASS train of about 0.11 scmm (4 scfm) to insure
proper operation of the cyclones. The sampling probe and oven were maintain-
ed at 121° C (250° F) instead of the usual 204° C (400° F) because it was
known that the particulates probably contain coal dust and carbonaceous
petroleum residues, which would distill volatile organics at higher tempera-
tures, thereby biasing the measurements of organic vapors emitted. The
production records were obtained, giving full information on metal, sand,
62
-------
• ATTACH NOZZLE TO PROSE
• ATTACH PROBE TO OVEN
• ATTACH CYCLONES ANO FILTER HOLDER
• ATTACH TEFLON HOSS TO FILTER HOLDER
ASSEMBLE SASS TRAIN COMPONENTS
AT SAMPLING SITE
LEVEL ANO ZERO MAGNEHELIC
GAUGES IN CONTROL MODULE
> CONNECT TEFLON HOSE TO ORGANIC MODULE
• CONNECT ORGANIC MODULE TO IMP1NGERS
• CONNECT IMPINGERS
• CONNECT IMPINGES TRAIN TO TUMPS
• CONNECT PUMPS TO CONTROL MODULE
LEAK CHECK FROM FRONT ON
tOi. CYCLONE AT 20" H
TAKE BLANKS
RECORD LEAD RATE AND FILTER
NUMBER ON FIELD OATA SHEET
PREPARE OXIDIZING IMPINGER
SOLUTIONS IN OFFICE
• IMPINGER '1 750 ml. 30% HiO:
> IMPINGER '2. *2 750 mi. 0 2M
INH4l2S2Oa and 0 02M AgNO3
• IMPINGER »4 750 gm. SILICA GEL
CHARGE IMPINGER TRAIN AT
SAMPLING SITE ANO HEAT
UP TRAIN TO 400° f
ADO ICE TO IMPINGEH
TRAIN AS NEEDED
TEAM LEADER CHECK WITH
PROCESS OPERATOR
INSURE PROCESS
OPERATING PROPERLY
i
POSITION PRO8E AT SINGLE
SAMPLING POINT IN DUCT
• RECORD CLOCK TIME
• RECORD DRV GAS METER READING
• RECORD iP. Tm. T,
• SET iH C 2.00 I - 4 sctml
• READ REMAINING GAUGES
i
START SASS TEST
GATHER PROCESS DATA
SAMPLE AT 4ictm DURING
HOT METAL ADDITION
RECORD STOP TIME ANO OTHER OATA
RECORD OATA ON FIELO
OATA SHEET
STOP SAMPUNG. REMOVE
PROBE FROM CUCT WAIT
FOR NEXT ADDITION
INSERT PROSE IN OUCT
ANO CONTINUE SAMPLING
REPEAT UNTIL TEST IS COMPLETE
RECORD FINAL READINGS
DISASSEMBLE SASS TRAIN.
SEAL COMPONENTS IN FOIL
ANO TRANSPORT TO OFFICE
Figure 14. SASS train sampling procedures.
63
-------
SAMPLING HOlllt.
10)1 CVCtONE OUSI
HINVE AND UHUbH WIIH 1:1
M£THANOL
-------
en
tn
UHUM IIOiE AND INliRNAl
SURIACES OF OHGANIC
MODULE
RINSE AND BRUSH WIIH
MtTIIVlENE CHIOHIDE
IHANSFEH WASHINGS
TO LAUELCD AMBER
GLASS BOTTLE
SEAL I OH SllirMENI TO
RESEARCH IHIANGLE INSIITUIC
XAO 1 CARTRIDGE
IHANSFEH XAO 2 TO
AMBER GLASS JAR
RINSE CARTRIDGE WIIH
ME1HVIENE CHLORIDE
IRANSf EH WASHINGS TO
LABELED AMBtfl GLASS JAH
CONTAINING XAO 2
SEAL Kill SHIPMLNI Ul
RESEARCH IMlANGlt INblllUIL
HINSE WIIH II
IPA.'DI WATER
TRANSFER WASHINGS
TO LABELED AMBEH
GLASS HOI HE
SEAL FOR Stlll-MENI IO
RESEARCH TRIANGLE INSIIIUIE
MEASURE VOLUME
AND RECORD
HINSE CONNECTOR.
SIEM eOIILE WIIH
1 1 leA'EO WATER
• V
IIIANSI HI IMI'IN(,IH
CONIEN1S IO LABELED
POLVEIHVLENE BOIUE
SEAL FOR SHIPMENI IO
RESEARCH IRIANGIE INSTUUIE
IMPINGEHI2 13
}
MEASURE VOLUME
AND BECOHO
RINSE CONNECTOR.
SUM. DOIIIE WIIM
II IPAilD WATCH
IRANSrCR IMPINUER
CONTENIS IU LABELED
POXVEIIIVIENE BOIILE
SEAL FOR
HESEAHCI
FOR SHIPMENI TO I
II IHIANGLEdNSIIIUIE I
WEIGH SILICA GCt
AND DISCARD
•NOCONDtNSAIt COILECIEO IN GLASS CONDCNSAIE JAB
Figure 16. SASS train sample recovery procedures.
-------
and cores on an hourly basis except when the line went down. Full records
were available on a minute by minute basis of work stoppage and work accom-
plished. These were provided by the companies. When the scrubber outlet
was sampled, the water flow and operation of the scrubber was continuously
monitored to insure that sampling only occurred while the scrubber was
operating. Likewise, periodic checks were made of the production line, but
the down-time was for pattern changes.
Foundry B had a shakeout room which was evacuated by inaccessible fans
at the top of the room. Considerable smoke emanated from the shakeout and
no flow pattern of air was discernable at the floor level. The SASS train
was used with only the filter and XAD-2 cartridge to obtain a fugitive
sample about 10 feet from the shakeout.
r T
L J
66
-------
8.0 ENVIRONMENTAL DATA ANALYSIS
Three samples were collected using the SASS train. Tables 16 and 17
summarize the results of particulate and organic data obtained for the three
samples. The source of the samples is detailed below.
TABLE 16. SUMMARY OF PARTICULATE DATA
Sampling Site
Air flowrate m /min
Particulate concentration,
mg/m
Particulate generated
kg/tonne cast
Sample 1
Green sand
shakeout
before scrubber
635
1,996
7.01
Sample 2
Scrubber
outlet
867
8.92
0.0434
Sample 3
Shell mold
shakeout
(fugitive)
49.59
-
TABLE 17. SUMMARY OF ORGANIC DATA
Sampling Site
Air flow rate, m /min
Total organic concentration
mg/m
Total organic generated,
kg/tonne cast
Sample 1
Green sand
shakeout
before scrubber
635
174. 61
0.614
Sample 2
Scrubber
outlet
857
105.3
0.512
Sample 3
Shell mold
shakeout
-
29.7
"•
67
-------
TABLE 18. PRODUCTION DURING SAMPLING
Total metal , tonnes
Metal/hr, tonnes
Total cores, tonnes
Total sand, tonnes
Total sand + cores, tonnes
(Sand + core)/metal ratio
Sample volume, m3
Air flow/ton cast:
Shell + cores, tonnes
Shot, tonnes
Sample 1
27.841
10.789
9.945
114. 519
124.464
4.471
15.23
3,516m3
-
-
Sample 2
59.809
10.556
18,678
279.682
298.359
4.989
26.15
4,865m3
-
-
Sample 3
25.445
15.118
_
_
_
0.0365
12.47
_
9.285
262.529
Table 18 summarizes the production data during the sampling periods.
The stack and SASS train data are listed in the Appendix.
Production and material data pertinent to samples 1 and 2 are as follows:
Normal casting rate: 11-17 tonnes per hour
Weight of iron per mold: 63.6-72.7 kg (140 to 160 Ibs)
Weight of individual pieces: about 4.1 kgs (9 Ibs)
Maximum rate of casting: 250 molds/hour
Minimum cooling time: 47 minutes
Weight of green sand per mold: 340-364 kg (750-800 Ib)
Weight of cores per mold: 18-23 kg (40-50 Ibs)
Sand to metal ratio: 5:1
Percentage core sand: 6%
Size of molds: 61 x 81 x 41 cm (24" x 32" x 16")
Temperature of fresh return sand: 121-177° C (250-350° F)
Temperature of cooled return sand: 30.7° C (97.2° F ± 2.5)
Carbon content of return sand: 1.16 ± 0.15%
Moisture in molding sand: 2.96 ± 0.36%
Analysis of green sand:
68
-------
New sand: 5%, Compression Strength: 20 psi
Clay: 7.5% (bentom'te)
Water: 3.0%
Combustibles: 4.0%
Volatiles (at 482° C): 2.0% (1.9% during test)
Organic components:
"Charbo" - charred oat hulls
"Kleankasf'-Asphalt Emulsion.
(Due to changeover from seacoal to kleankast, the noncharbo organic
content was 70% seacoal (0.57% of sand), 30% kleankast).
Analysis of Cores:
Percent binder: 1.75%
Composition of binder:
315 Phenolic 0.9625%
615 Isocyanate 0.7875%
Catalyst: TEA 0.10%-0.20% of Sand Weight
Density: 95 Ibs/cu. ft.
Tensile Strength: 100-200 psi.
Sand: Lake; 50 GFN; ADV 0-5
The collected samples were subjected to analysis by the following
procedure outline:
Organic Vapors collected by XAD-2 resin and rinses of SASS train:
Soxhlet Extraction
TCO and Gravimetry
LC; IR; LRMS; TCO; GRAV
Particulates collected in cyclones and filter:
Gravimetry
Soxhlet Extraction
Parr/Acid Digestion
SSMS
As/Hg/Sb
NaOH Impinger:
CN analysis.
Further, a portion of the organic extract of the XAD-2 was subjected to
GC-MS analysis.
69
-------
Sample 1
This is the "master sample" the uncontrolled effluent from shakeout.
The molding line was using phenolic isocyanate bound cores in green sand
molds with seacoal and "kleankast"® additions.
This sample was taken from a duct on the floor above the shakeout hood
as shown in Figure 13 by standard SASS train procedures. The air flow in
the duct was 10.526cm3/sec, which was 3,516 mVtonne of metal cast during
the sample period.
Sample 2
This sample came from the same source and conditions as sample 1 with
the difference that it was obtained after the air had passed through a wet
scrubber. This sample was obtained the day following sample 1. This sample
is the controlled atmosphere discharge. During the collection of sample 2,
the air flow was 4,865 m3/tonne of metal cast, at a rate of 14.375 mVsec.
This flow is greater than for sample 1. The only observable reason for
this is the presence of leaks in the system. The air is drawn by suction
from the shakeout hood up through the wet scrubber. The air ducts had been
damaged by erosion—corrosion, and other factors. The damage was between
the take off duct from the shakeout hood and the scrubber, allowing ambient
air from above the casting line to enter the system.
Sample 3
Sample 3 was taken in a room in which phenol-formaldehyde shell molds
were dumped onto a shakeout table. The shells were held in flasks and
surrounded with iron shot for the casting operation. The process weight
during the test was 194.66 tons/hour, consisting of 6.08 tons/hr shells and
cores, 16.65 tons/hr iron poured, and 171.82 tons/hr of supporting shot.
The shot temperature was 232° C (450° F).
8.1 ANALYSIS OF SASS TRAIN SAMPLING OF GREEN SAND SHAKEOUT EFFLUENT:
SAMPLE 1
Sample 1 is the shakeout effluent from green sand molds containing
phenolic isocyanate cores. Both seacoal and petroleum additives were used
in the green sand. The importance of this sample is that it represents a
typical casting operation and the environmental emissions before any air •
pollution control efforts are made.
70
-------
TABLE 19. PARTICULATE CONCENTRATION
Sample: 1, Shakeout, Green sand, Line 5
, Total Emission,
Category Weight, mg Load, mg/m g/tonne cast
10u dust
Probe rinse
Total
83.2
925.1
3,740.4
25,447.0
196.1
30,3918
5.46
60.74
245.59
1,670.85
12.88
1,995.5
19.2
213.6
863.5
5,874.7
45.28
7,017'
3
Sample volume at 15.5° C and 76.1 cm Hg: 15.23m
Total load in grains/ft3: 0.8720
Metal cast during sample period: 27.841 tonnes
Air flow/tonne cast: 3,516m3 (Std. dry)
8.1.1 Total Particulate Loading
The total mass of particulates from an uncontrolled shakeout is given
in Table 19. Included in this table are the values of particulate emission
per ton of metal cast. Since the sand to metal ratio was 5:1, a common
target value, these values could be extrapolated to obtain an order of
magnitude estimate for similar plants. It should be noted that particulates
would be emitted even if the production line was operating temporarily
without iron being poured, since shaking out molds containing no iron will
still produce dust. The quantity of fine particles would probably be smaller
in that case. Table 20 summarizes the sampling conditions.
8.1.2 Level 1 Organic Analysis
Table 21 presents the organic extractables. The distribution among the
sizes of the particulates might be correlated with the fact that the larger
particles are likely to be made up of coal dust and carbonized petroleum
additive, which contain significant amounts of organic material. The fine
particulates were probably clay, as indicated by the inorganic analysis.
The organics in the vapor phase were 94.3 percent TCO material, that is, low
boiling and smaller molecules.
-------
TABLE 20. SUMMARY OF SAMPLING DATA FOR GREEN
SAND SHAKEOUT, SAMPLE NO. 1
Date of test:
Volume of gas sampled: 15.23m3
Duct gas temperature: 68.9° C
Duct gas pressure: 75.95cm
Duct gas molecular weight:
Duct gas moisture:
Duct gas velocity: 15.46m/sec
Duct gas flowrate: 10.53m3/sec (22,304 dscfm)
Total sampling time: 9300 sec (155 minutes)
SASS train flowrate: 0.001638m3/sec
(3.47 dscfm)
27.841 tonnes
(30.667 tons)
6/28/78
(537.81 dscf)
(156° F)
(29.90 inches Hg)
28.84
3%
(50.72 ft/sec.)
Iron cast during sampling:
TABLE 21. ORGANIC EXTRACTABLES, SAMPLE 1
Type of Sample
Filter: (>lu)
>3u:
>10u:
XAD-2:
Total
Emission -
• cone, mg/m
0
0.12
0.82
173.67*
174.61
Emission cone.
g/tonne cast
0
0.42
1.32
610
612
*94.3% TCO
Table 22 summarizes the LC and IR analysis of the vapor phase organics
collected by the XAD resin. The detailed summary by LC fractions is found
in the Appendix.
72
-------
TABLE 22. SUMMARY OF ORGANIC VAPOR ANALYSIS FROM GREEN
SAND SHAKEOUT, SAMPLE 1.
Emission rate: 554 g/ton cast
Category
Aliphatics
Haloaliphatics
Substituted benzenes
Halobenzenes
Fused aromatics
Hetero N compounds
Hetero 0 compounds
Hetero S compounds
Alkyl S compounds
Nitriles
Aldehydes, ketones
Nitroaromatics
Ethers, Epoxides
Alcohols
Phenols
Amines
Amides
Esters
Carboxylic acids
Sulfonic acids
GRAV cone.
TCO cone.
Found mg/m
0.72
0.22
2.45
0.24
2.45
0.56
0.10
0.10
0.06
0.01
0.10
0.01
0.10
0.56
0.56
0.56
0.47
0.15
0.46
0.05
9.85
163.8
Min. MATE
value in 3
category mg/m
20
0.1
1.0
0.7
0.001 to 200
0.1
300
2
1
1.8
0.25
1.3
16
10
2
0.1
1.0
5.0
0.3
0.8
Ratio
cone, found
MATE
0.04
2.2
2.45
0.34
24,000
5.6
0.00
0.05
0.06
0.01
0.4
0.01
0.01
. 0.06
0.28
5.60
0.47
0.03
1.53
0. 06
The MATE values are the Minimum Acute Toxicity of Effluent values, or
the minimum quantity that has been determined to be detrimental to the
environment. These are "Air, Health MATE" values from the "MEGs" or Multi-
media Environmental Goals91. The MEGS give a MATE value for each individual
compound. The values listed in this report are the lowest MATE values in
each category of compounds. Thus, unless the specific compound having this
MATE value is actually in the sample, the MATE value shown would be too low
and the concern ratio too high.
The LRMS data (Appendix) indicated possible PNA's. The sample was
analyzed by GC-MS for confirmation. Known compounds, listed in Table 23,
were introduced to the GC-MS to obtain calibration factors, which were then
used to quantify the same compounds in the sample. The results in Table 23
73
-------
show that the concentration of PNA tes.ted for are well below the MATE values.
The highest concentration found (for naphthalene) is only 3 percent of the
MATE value. The GC-MS system used can identify PNA's with molecular weights
below about 270. No PNA's between 229 and 270 (which includes benzo(a)pyrene)
were found. Since BaP and the high molecular weight PNA's are from the same
source (the shakeout) as the PNA's tested for, the low values found by GC-MS
analysis indicate an equal or lower concentration of the higher molecular
weight PNA's. The identity of the fused aromatics indicated by LRMS and not
listed in Table 23 is not known. If the Level 1 analysis is correct, then '
1/3 of the fused aromatics are unaccounted for, by GC-MS. However, the
technique used by Level 1 procedures is too inaccurate to firmly establish
the quantitative level.
The GC-MS analysis produced a complete set of mass spectra for each GC
peak. Figure 17 is the gas chromatogram of sample one. The 21 chromato-
graphic peaks that exceeded 9.6% of -the highest concentration components,
(p-methylene naphthalene and an unsaturated Cg akyl benzene isomer) were
interpreted. Table 24 lists the substances identified along with the rela-
tive peak heights of the 21 peaks analyzed. The peak height is proportional
to concentration and can therefore be used to measure relative concentrations
to a first approximation. (Accurate determinations require comparison with
a known quantity of the substance of concern.) By summing all peak heights
it was estimated that the 21 peaks analyzed represent 79% of the total
quantity of material analyzed. 62 peaks (representing 21% of the material)
TABLE 23. QUANTITATIVE DETERMINATION OF PNA COMPOUNDS PRESENT
IN GREEN SAND SHAKEOUT, SAMPLE 1
Compound
Naphthalene
Dibenzofuran.
Anthracene
Phenanthrene
Fluoranthene
Pyrene
Chrysene
Total
MW
128
168
178
178
202
202
228
ng/ul
452.0
3.0
11.2
2.3
0.2
0.2
4.7
Wt. in total
extract (ug)
22,600
150
560
115
10
10
235
Cone.
ug/m
1,484
9.8
36.8
7.6
0.7
0.7
15.4
1,555 mg/m
Air health
mate ug/m
50,000
N
56,000
1,600
90,000
230-,000
2,200
74
-------
CO
QD
V
*-
tr> o
Ul
H-
*""*
• SJJ
~
cr
uu
Qi
p
(M
O3
CO
to j—
in
cc
'x ' •
SCAN IDENTIFICATION
10 RUN IDENTIFICATION 1434-2 628-2G CONC.
SPECTRA FILE NAME AAA753
FILE POSITION
RETENTION TIME
601 SPECTRUM HO. COO
287S2
' NO BACKGROUND SUBTRACTION
MAXIMUM INTENSITY iB72 V. OF TOTAL ION 13.8
INTENSITY FACTOR 1 RANGE 1 TO 339
SCAN DATE 13-BEC-1978
OUTPUT MASS RANGE 1 TO 492
NORMALIZATION RANGE 1 TO 999
1 . SCAN SPEED
1
\^ ii. J
2
V
' ill P'Y'j 1111 i
UJ ~ZL —i O
no o
u: cr — —
3
1
J h U
9
6 1
i
a
i I
i
W
MAXIMUM MASS
2
503
. 13 ION SOURCE TEM?. 210 ELECTRON ENERGY 70 TRAP CURRENT
ACC. VOLTAGE 3500 MULTIPLIER SETTING 400
START TIME
.0 INTERVAL TIME .0 FACTOR .0
14 COLUMN NAME 1M SE-33 22M
12
I
VP
FLOU 3.S2MU 33SP13:!
INJ. TEMP. 250
COLUMN TEM?. lQe=.V2' S>x/' TO 265*
SEP. TEMP. 200
15
17
IS
> i
_J
i
2
<
U)
V-
i
o
5 I
1 o '
1 s5
oi
i
1319 '? v 1
I 1 '1
| . ! j
IIP'
T 1 1 1 1 1 1 1 | 1 1 1 1 1 1 'I' 1 '1 | 1 1 1 1 1
0 O
O O
w m :
2l' *
jjAi i!| n
1 1 1 1 1 1 1 1 1 1 1 1
O
O
if
1 1 | 1 1 1 1 1 1 1 1 1 | 1 1 1 1 1 1 IT 1 | I 1 1 1 1 1 li 1 |"
.0 O 0 0
0 D 0 0
ID ' VO 1". OC
50
uj o en
_1 UJ h-
« 0. O
UL. cn P
Figure 17. Gas chromatogram of organic effluents, sample 1.
-------
TABLE 24. IDENTITIES OF MAJOR ORGANIC COMPONENTS IN AIR
SAMPLE 1
Chroma tographic
peak no.
1
2
3
4
5
6
7
8
Relative
peak height
.34
.18
.14
.49
.11
.18
.12
.12
Percent
of
sample
4.2
2.2
1.7
6.0
1.3
2.2
1.4
1.5
Compound
Aniline
Phenol
Cresol isomer
C11H24 1'soraer
Naphthalene
Ce-alkyl benzene isomer
CV II •
12 26 lsomer
Dimethyl indan isomer
Dimethvlindan isomer
Cg alkylbenzene isomer
•19 2.4 Cg alkyl benzene isomer
C14H30 isomer
Dimethylindan isomer
10 l-OQ 12.3 p-methylnaphthalene
Unsaturated Cg alkylbenzene
isomer
Cg alkylbenzene isomer
11 -73 9-1 C13H28 1somer
a-methylnaphtha 1ene
12
13
14
15
16
17
18
19
20
21
.28
.68
.59
.34
.18
.21
.13
.14
.15
.11
3.5
8.3
7.2
4.2
2.2
2.6
1.6
1.7
1.8
1.3
Ethylnapthalene isomer
Trimethyl indan isomer
Ethylnaphthalene isomer
C-i/iH^rt isomer
in iO
Dimethyl naphthalene isomer
Diphenylmethane
Dimethyl naphthalene isomer
Dimethylnaphthalene isomer
^15^32 lsoraer
C3 alkyl naphthalene isomer
C3 alkyl naphthalene isomer
C3 alkyl naphthalene isomer
^16^34 1'somer
Di-p-tolymethane'(tent. )
C,yH isomer
Anthracene-d1f)
76
-------
were not analyzed. Table 24 also lists the percentage of each substance in
the sample. This is based on the assumption that equal quantities of any
substance produce equal peak heights, which is not true, therefore these
values are an approximation for comparison only. The GC spectrum.is pre-
sented in the appendix as Figure Al. The 36 predominant compounds in the
sample were identified. Nine are benzene compounds, 18 are two ring poly-
cyclics, 11 of which are naphthalenic compounds, and one, anthracene, is a 3
ring polycyclic. Seven are aliphatic compounds. Thus a trend toward lower
quantities of greater than two ring compounds is seen. The list in Table 23
can be added to this, identifying five >2 ring PNA's. As seen in Table 23,
and the small peak heights for higher boiling substances in Figure Al, the
quantity of >2 ring PNA's is very small.
Figure 18 compares the emissions from the shakeout, before scrubbing,
with the MATE value ranges. The values given in the organic extract summary
table for sample 1 were inserted into this figure as triangles. Level I
analysis does not discriminate the subcategories and therefore in a case
such as amines, the emission value is safe by an order of magnitude if the
amines are primary, but not if they are secondary or aromatic. This table
indicates that there may be problems with:
1. Alkyl halides (or Haloaliphatics)
2. Carboxylic acids; derivatives
3. Amines
4. Substituted Benzene Hydrocarbons
5. Fused polycyclics
6. Nitrogen heterocyclics
A closer examination however, remembering that Level 1 analysis seeks
only a factor of 3 accuracy, reveals the following:
1. Alkyl halides:
Of concern only if they are unsaturated. -LRMS data indicates a
much lower concentration, about 0.03 mg/m . Therefore they are
not likely a problem.
2. Carboxylic Acids:
The level only slightly exceeds the MATE for a few members of
"Acids with other functional groups". There is no LRMS confirma-
tion. It would be most difficult to propose that a level of
concern exists.
77
-------
C»
MEG's Category
1. ALIPHATIC HYDROCARBONS
A. Alkanes and Cyclic Alkanes
B. Alkenes, Cyclic Alkenes, Dienes
C. Alkynes
2. ALKYLHALIDES
A. Saturated
B. Unsat Lira ted
3. ETHERS
A. Noncyclic Aliphatic or Aromatic
B. Cyclic
4. HALOGENATED ETHERS
A. Monohalogenated
B. Dihalogenated, Polyhalogenated
5. ALCOHOLS
A. Primary
B. Secondary
C. Tertiary
6. GLYCOLS, EPOXIDES
A. Glycols.
B, Epoxides
7. ALDEHYDES, KETONES
A. Aldehydes
B. Ke tones
8. CARBOXYLIC ACIDS; DERIVATIVES
A. Carboxylic Acids
B, Acids With Other Functional Groups
C. Amides
D. Esters
9. NITRILES
A. Aliphatic
B. Aromatic
MATE VALUES, »qfm3
0.1
1.0
10
100
104 105 106
w
'' '*'' '
\r wv&v^ **CTf
\ r «„.;.,„.... j,.,^..,^...^
I
e
a
Figure 18. Emissions from shakeout compared with MATEs.
-------
MEG's Category
10. AMINES
A. Primary Aliphatic
B. Secondary Aliphatic
C. Aromatic
D. Tertiary
11. AZO COMPOUNDS;
HYDRAZINE DERIVATIVES
A. Azo Compounds
B. Hydrazine Derivatives
12. NITROSAMINES
A. Aliphatic
B. Aromatic
13. THIOLS.-SULFIDES
A. Thiols
B. Sulfides; Disulfides
14. SULFONIC ACIDS; SULFOXIDES
A. Sulfonic Acids
B. Sulfoxides
15. BENZENE; SUBSTITUTED
BENZENE HYDROCARBONS
A. Benzene; Monosubstituted
B. Disubstituted, Polysubstituted
16. HALOGENATED AROMATICS
A. Ring Substituted
B. Halogenated Alky! Side Chain
17. AROMATIC NITRO COMPOUNDS
A. Simple
B. With Additional Functional Groups
18. PHENOLS
A. Monohydrics
B. Dihydrics; Poiyhydrics
C. Fused Ring Hydroxy Compounds
MATE VALUES, j
0.1
1.0
10
100
10"
103
10°
E
5;fcS18fc.c .a^r j«. .^:^Aiaaa
V
Figure 18. (Continued.)
-------
CO
o
MEG's Category
19. HALOGENATED PHENOL1CS
A. Halphenois
B. Halocresols
20. NITROPHENOLICS
A. Nitrophenols
B. Nitrocresols
21. FUSED POLYCYCLICS
A. Two or Three Rings
B. Four Rings
C. Five Rings
D. Six or More Rings
22. FUSED NON-ALTERNANT POLYCYCLICS
A,B. Two, Three, or Four Rings
B. Five Rings
C. Six or More Rings
23. NITROGEN HETEROCYCLICS
A. Pyridino; Substituted Pyridinos
B. Fused Six-Membered Rings
C. Pyrrole; Fused-Ring Pyrrole Derivatives
D. With Additional Hetero Atoms
24. OXYGEN HETEROCYCLES
A,B. One, Two, Three, or More Rings
25. SULFUR HETEROCYCLES
A. One Ring
B. Two or More Rings
26. ORGANOPHOSPHORUS COMPOUNDS
A. Aliphatic-
B. Aromatic
MATE VALUES, /ig/m3
0.1
1.0
10
100
10J
10"
10°
10"
f
&
i
Figure 18. (Continued.)
-------
3. Amines:
No LRMS confirmation. Level exceeds the lowest MATE value by less
than an order of magnitude, and then only if they are aromatic
amines. It should be noted that aromatic amines are probable in
this system and the level of amines is the highest level of con-
cern in the results, with the exception of fused polycyclics.
4. Substituted Benzene Hydrocarbons
This system of pyrolysis products is expected to give the greatest
concern in this family of compounds, but the level is less than an
order of magnitude above the lowest MATE value.
5. Fused Polycyclics:
Because of the used of seacoal and asphaltic substances, this was
the area of greatest concern at the onset of the sampling program.
The results indicate very definite problems if the polycyclics are
of four or more rings. The GC-MS analysis however did not reveal
any concern level in that category but revealed a predominance of
naphthalene compounds. The level found is near the lowest MATE
values for two ring systems (naphthalene compounds) and is of less
concern than amines.
6. Nitrogen Heterocyclics:
Again, these do not show up in the LRMS analysis. The level
indicated is less than an order of magnitude above the lowest MATE
for pyroles.
In summary, no definitive statement can be made to the effect that the
organic emissions are hazardous. There is a possibility that some organic
compounds are being emitted above the MATE levels. This is only a reason-
able possibility if (a) the entire quantity of family substance found of
concern is made up of less than 10 chemical compounds and (b) the compounds
present have the lowest MATE values in their category. The probability of
both (a) and (b) being true is quite low, certainly less than 10 percent- if
not less than 1 percent. High resolution studies would show over 1000
chemical compounds, as Bates21 has indicated, and this factor alone pre-
cludes the probability of proposition (a) being true.
8.1.3 Inorganic Analysis
The respirable portion of the particulate (<3u) was subjected to spark
source mass spectrometry. The complete analysis is found in the Appendix.
Table 25 presents the portion of the results that indicates a possible
81
-------
00
TABLE 25. METAL CONTENT OF <3 MICRON DUST FROM GREEN SAND SHAKEOUT
Ele-
ment
Si
Ca
Tr
Cr
Mn
Fe
Ni
Cu
As
Se
Cd
Pb
Observed
|jg/m
12E4**
655
36.4
73
31.1
1,260
26.5
3.8
0.79
0.54
0.38
2.6
Air MATE Concern Ratio Control
Mg/m (Value/Mate) Level %
1E4 12 91.7
16E3
6,000
1 73 98.6
5,000
700 to 9,000 1.8 to 0.1 44 to 0
15 1.77 43.4
200
2
200
10
150
Observed
M9/9
18E4
9,900
550
1,100
470
19E3
400
99
12
<8.2
5.7
40
Land MATE Concern Ratio
pg/g (Value/Mate)
None
3,200
160
50
20
50
2
20
10
5
0.2
10
_
3.10
3.44
22
23.5
380
200
4,95
1.2
1.64
28.5
4
Required
Control
Level %
—
67.7
70.9
95.5
95.8
99.7*
99.5
79.8
16.7
39
96.5
75
*Not firmly established yet.
**To economize space, E is used to mean "positive power of 10", thus 1E4 means 1 x io4 or 10,000.
-------
environmental concern. In this, the Air, Health MATE and the Land, ecology
MATE values are compared with the sample analysis. A "Concern ratio" was
then calculated. This is defined as the ratio of the value found to the
MATE value. The concern ratio can be used to determine the degree of control,
i.e., the percentage of removal required to reduce the concentration to the
MATE value. The Land, ecology- values do not apply to the air sample but
would apply to the collected dust for landfill considerations. The dominance
of Al, Mg, Si in the analysis is consistent with the major composition of
the dust being clay and silica.
Of significant concern are Zr, Ba, and the rare earths Ce, Pr, Nd.
These are additives to the Mg inoculant. Their appearnce as far down the
processing line as the shakeout was not expected. This indicates that the
inoculation process should be investigated further.
Since the isocyanate in the binders can conceivably decompose to HCN, a
special NaOH bubbler was used on the SASS train to trap cyanides. This
analysis is given in Table 26.
TABLE 26. CYANIDE ANALYSIS SAMPLE 1; GREEN SAND SHAKEOUT
Volume NaOH in impinger:
CN~ analysis 31-5 PPm
CN~ content 30-87 m9 , 3
CN~ load 2.027 mg/m
MATE, Air Health, value: 11 mg/m
CN~ emissions per ton cast: _ 6.470 g
8.2 ANALYSIS OF SASS TRAIN SAMPLING OF SCRUBBER EFFLUENT FROM SHAKEOUT OF
GREEN SAND MOLDING WITH ISOCYANATE CORES
8.2.1 Total Particulate Loading
Sample 2 was taken from the roof stack after the exit of a wet scrubber
of the venturi rod type. The scrubber was 99.54 percent efficient in remov-
ing parti culates, thus the parti cul ate catch was small. Due to the small
catch, the probe and all cyclone catches were rinsed out and combined in the
field. The results are presented in Table 27. The summary of Sampling Data
is given in Table 28.
8.2.2 Level 1 Organic Analysis
Table 29, the organic analysis summary, gives the LC and IR analysis of
vapor phase organics. The detailed LC data is found in the appendix.
83
-------
TABLE 27. PARTICULATE LOADING, SAMPLE 2, POST SCRUBBER
Emission
. . concentration Total emission
Category Weight, mg mg/mj g/tonne cast
lu dust 126.7 4.85 23.6
(Probe rinse and all cyclone catches were combined in field due to small
qty.)
Total 233.2 8.92 43.4
Sample volume at 15.5° C, dry: 26.15m3 Total load in grains/ft3: 0.00390
Metal cast during sample period: Air flow/tonne cast: 4,865m3
59.81 tones (std., dry)
TABLE 28. SUMMARY OF SAMPLING DATA FOR SCRUBBER
EFFLUENT, SAMPLE NO. 2
Date of test: 6/29/78
Volume of gas sampled: 26.151m3
Stack gas temperature: 42.77° C
Stack gas pressure: 75.54cm Hg
Stack gass molecular weight: 28.84
Stack gas moisture: 12%
Stack gas velocity: 25.79m/sec.
Stack gas flowrate: 14.375m3/sec
Total sampling time: 337 minutes
SASS train flowrate: 0.001293m3/sec
Iron cast during sampling: 72.567 tonnes
As with sample 1, substituted benzenes and fused aromatics predominante.
The wet scrubber did little to remove organic vapors.
84
-------
TABLE 29. SUMMARY OF ORGANIC VAPOR ANALYSIS
FROM GREEN SAND SHAKEOUT AFTER WET
SCRUBBING, SAMPLE 2
Category
Aliphatics
Haloaliphatics
Substituted benzenes
Halobenzenes
Fused aromatics
Hetero N compounds
Hetero 0 compounds
Hetero S compounds
Alkyl S compounds
Nitriles
Aldehydes, ketones
Nitroaromatics
Ethers, Epoxides
Alcohols
Phenols
Amines
Amides
Esters
Carboxylic acids
Sulfonic acids
TCO
GRAV
3
Found mg/m
0
0.27
2.82
0.29
2.82
0.61
0.12
0.12
0.05
0.08
0.08
0. 08
0.08
0.49
0.49
0.49
0.49
0.09
0.49
0.05
95.17
10.13
MIN. MATE
value in 3
category mg/m
20
0.1
1.0
0.7
0.001 to 200
0.1
300
2
1
1.8
0.25
1.3
16
10
2
0.1
1.0
5.0
0.3
0.8
Ratio
cone, found
MATE
0
2-7
. 7
2 GO
. 82
0.4
28,000
6.1
0.00
Or\f
. 06
OO f"
.05
On /i
.04
0.32
Or\r
. Ub
0.01
Or\ c.
. 05
0.25
4.90
0.49
0.02
1.63
0/1 y~
. 06
The cyanide emissions were 19 percent less after the scrubber on a per
ton cast basis, as indicated in Table 30.
TABLE 30. CYANIDE ANALYSIS, SAMPLE 2
Volume of NaOH in impinger
CN analysis
CN content
CN load
MATE, Air Health
CN emissions per ton cast
810 ml
38.0 ppm
30.78 mg 3
1.18 mg/m
11 mg/m
5.212 g
8.2.3 Inorganic Analysis
Inorganic analysis an sample 2 was not performed because of the small
quantity and the reasonable assumption that the analysis would be essentially
85
-------
the same as that of Sample 1. Since the scrubber is highly efficient (99.5%)
for large participates but not for smaller particulates (25% for
-------
ing the mold core and could be emitted when the shakeout exposed this sand
to the hot metal and hot sand. The higher molecular weight substances are
considered to be of greater environmental concern. In the case of shell
molding, the shell is thin enough that even the sand on the outside suffers
extreme heat.
The iron shot is more permeable than sand and does not present the
large surface area for adsorption that clay and sand do. It is therefore
reasonable to. expect a larger portion of the low boiling volatiles to escape
and also burn during the initial period after pouring. These mechanisms
would predict a lower yield of TCO material, as was found.
In spite of the differences in sampling conditions, the values for
substituted benzenes and fused aromatics are about equal to those in sample 1.
A notable difference is the high value of aliphatics, and a nitrile level
nearly 50 times that of green sand shakeout.
TABLE 31. SUMMARY OF ORGANIC VAPOR ANALYSIS FROM PHENOLIC
SHELL SHAKEOUT, SAMPLE 3
Category
Aliphatics
Haloaliphatics
Substituted benzenes
Halobenzenes
Fused aromatics
Hetero N compounds
Hetero 0 compounds
Hetero S compounds
Alkyl S compounds
Nitriles
Aldehydes, ketones
Nitroaromatics
Ethers, Epoxides
Alcohols
Phenols
Amines
Amides
Esters
Carboxylic acids
Sulfonic acids
GRAV cone.
TCO cone. x
Found mg/m
2.14
0.40
2.46
0.24
2.46
0.75
0.27
0.27
0.05
0.47
0.27
0.03
0.27
0.54
0.14
0.54
0.49
0.54
0.48
0.04
12.84
16.86
Min. MATE
value in 3
category mg/m
20
0.1
1.0
0.7
0.001 to 200
0.1
300
2
1
1.8
0.2
1.3
16
10
2
0.1
1.0
5.0
0.3
0.8
Ratio
cone, found
MATE
0.11
4/«»
.0
2.46
0.34
25,000
7.5
0.00
0.14
0.05
0.26
1.35
0.02
0.02
0.05
0.07
5.4
0.49
0.11
1.60
0.05
87
-------
Category
All dust
Sample volume at
Load in grains/ft
TABLE 32. PARTICULATE
60° F: 12.47m3
3: 0.02167
LOADING, SAMPLE 3
Weight, mg
618.3
3
Load, mg/m
49.59
8.4 COMPARISON OF ORGANIC EMISSIONS TO MATES
Table 33 lists the major categories of compounds, the values found in
samples 1 and 3, and the lowest MATE values for some member of the category.
From this it is seen that the only possible problems are with alky! halides,
amines, fus.ed polycyclics, and nitrogen heterocyclics. As stated earlier,
the GC-MS results for sample 1 showed that the major carcinogenic members of
TABLE 33. COMPARISON OF ORGANIC EFFLUENTS
Substance Category
Aliphatic hydrocarbons
Alky! halides
Ethers
Alcohols
Aldehydes, ketones
Carboxylic acids
Nitriles
Amines
Sulfonic Acids
Substituted Benzenes
Halogenated Aromatics
Phenols
Fused polycyclics
Nitrogen heterocyclics
Sample 1
mg/m
0.7
0.2*
0.1
0.6
0.1
0.5
0.01
0.5*
0.05
2.4
0.2
0.6
2.4*
0.6*
Sample 3
mg/m
2.1
0.4*
0.3
0.5
0.3
0.5
0.5
0.5*
0.04
2.5
0.2
0.1
2.5*
0.8*
Lowest MATE
for category
mg/m
20
0.1
16
10
0.2
0.3
1.8
0.1
0.8
1
0.7
2
0.0001 to 200
0.1
*Possible problem exists.
88
-------
the fused polycyclics are not present at levels of more than 3 percent of
the MATE values and naphtha!enic compounds predominate. The fact that
similar results were obtained for substituted benzene and fused polycyclics
in the case of green sand with seacoal and synthetic asphalt and also in the
case of phenol-formaldehyde and sand, indicates that seacoal and heavy
organic additives are of no greater concern than any other organic material.
When making the comparisons it must be carefully observed that the
values of substance found is the sum of all the members of the category that
were present. On the other hand, the MATE values are the lowest value
applicable to one member of the category.
With this caveat in view, there is a high probability that the uncon- \
trolled organic emissions from the shakeout do not pose a threat to the
environment in foundries that operate in a manner similar to the ones tested.
Evaluating the results from Level 1 testing also requires cognizance of '
the purpose and philosophy of Level 1 testing. The analytical accuracy
expected is only within a factor of three. Thus the true answers could well
be less by a factor of three, which would remove most of the categories that
reach MATE values. On the other hand the true values could be three times
greater than the analytical report. In the present case, this would still
result in only the hetero N, amine and fused aromatics exceeding the MATE by
a factor of ten. Thus the analytical results do not definitively describe
the pollutant level as either unacceptable or safe. To resolve this problem
Level 2 testing will be required.
89
-------
9.0 DISCUSSION OF RESULTS
9.1 ANALYSIS OF PHYSICAL-CHEMICAL MECHANISMS AFFECTING EMISSIONS
A notable result of the testing was the low quantity of high molecular
weight compounds in the effluent revealed by the analysis of sample 1,
shakeout of green sand with seacoal molds and isocyanate cones. This war-
rants an explanation since high molecular, weight compounds were expected.
This will be presented as a mechanistic analysis of the fate of the organic
compounds emitted during casting.
Consider a large block of moist sand, clay and high molecular weight
organic material, containing a cavity into which iron is poured. The molten
iron will heat the sand mixture from the inside toward the outside, pro-
ducing a high thermal gradient. Figure 10, page 53, shows, by the curves
for Jrinch and 1 inch from the sand-metal interface, that the temperature of
the sand mixture cannot rise above 212°F (100°C) until after the water
content has vaporized. Thus the moisture content helps absorb the heat of
the cooling iron and minimizes the distance from the metal-sand interface at
which the temperature can rise above 212°F. Since, in addition, dry sand is
a good insulator there is a high thermal gradient in the sand surrounding
the casting, throughout the cooling period.
The introduction of the molten iron causes the organic material to
pyrolyze into lower molecular weight substances. Some of this organic
material graphitizes forming the "lustrous carbon" layer next to the metal
that produces a good metal finish. The laboratory test of Bates & Scott, }
found 50% hydrogen, 22% carbon monoxide, 6.4% carbon dioxide, 4.5% methane,
and 4.8% higher hydrocarbons in the gases emitted from a sealed mold. The
vaporized substances thus formed travel away from the metal-sand interface
and into the cooler sand, both by gas pressure and by thermal transpiration.
As the vaporized organic material travels away from the sand-metal
interface, it is adsorbed on the clay particles and may condense to a liquid
when it encounters sand that is below the boiling point of the substance
90
-------
involved at the partial pressure of the substance. The first action will be
adsorption on the clay, since adsorption of a compound onto a solid will
occur above its boiling point. This process will be of lower significance
relative to the sand, but clay has a very high surface area and can adsorb
considerable quantities of material per unit weight. The second action to
occur is simple condensation. The permanent gases will, of course, pass on
through the sand mixture. Thus, the sand clay mixture will act as a selec-
tive trap, adsorbing the higher molecular weight materials (e.g., benzene
and larger) more readily than the more volatile materials.
Immediately after pouring iron into a sand mold, gases are observed
burning at the seams of the flask and other places that allow escape. The
analysis given by Bates & Scott indicates that the majority of burning gases
will be hydrogen, carbon monoxide and methane.
Upon shaking out the mold, the cooler sand and clay that have trapped
or condensed the hydrocarbons will come into contact with the hot metal and
the layer of hot sand surrounding the metal. This will result in vaporizing
.some of the condensed organics. There are two processes that favor emis-
sions of the lower molecular weight material. The first is the generation
stage of pyrolysis, which by its nature breaks larger molecules into smaller
molecules, thereby tending to produce more low molecular weight substances.
The second is the revolatization of the condensed hydrocarbons during
shakeout. The heating of the cooler sand by the metal and hotter sand is
limited, therefore the boiling off of the lower molecular weight and higher
vapor pressure compounds will be favored.
If the mold is completely cooled before shakeout, then no secondary
boiloff emissions will occur. Thus both the quantity of shakeout emissions
and the ratio of the high to low boiling compounds emitted will vary with
metal temperature at the time of shakeout. This strongly indicates that
cooling time can be used as a technique to control shakeouts emission.
Further, foundrymen report that in cases where a casting is cooled over-
night, there the emissions on shakeout are nearly completely eliminated.
91
-------
9.2 COMPARISON OF EMISSIONS FROM DIFFERENT CHEMICAL SOURCES
Two chemical systems were tested:
Samples 1 & 2: Shakeout of green sand molds containing seacoal
and phenolic isocyanate cores.
Sample 3: Shakeout of phenol formaldehyde bound shell molds.
The green sand with seacoal molds were expected to emit substances similar
to those emitted by coke ovens and other coal processes. The phenolic shell
system was expected to emit the decomposition products of phenol -formalde-
hyde, especially since the area around the foundry smelled of phenol.
It is reasonable to expect differences in the emissions from these
sources, but since Level 1 analysis is by category of compounds, the dif-
ferences may not appear significant. In addition, Sample 3 was a fugitive
sample, thus the concentrations cannot be related to the quantity of casting.
The most obvious difference is in the ratios of high boiling (GRAV
material) to low boiling (TCO material) in each sample.
GRAV x 100 J Sampjej
(GRAV + TCO) 5'7% 9-6% 43.2%
The differences between samples 1 and 2 are within experimental error
but sample 3 exhibits 5.8 times the GRAV material as the average of sam-
ples 1 and 2.
In shell molding the shell is about Jg-inch thick and is supported in a
flask of iron shot. The shell is thin enough for even the outter portions
to become very hot. Thus a significant amount of condensation of low boil-
ing compounds on the sand is not expected. There is no moisture in the
system to absorb heat and the iron shot has a very low surface area relative
to sand or clay, thereby reducing its capacity to trap or condense low
boiling organics before they pass through the interticies of the shot and
escape in the air of the cooling room. Thus, at the time of shakeout, the
proportion of higher boiling compounds in the sand and iron shot is expected
to be greater than the low boiling compounds. This explains the experi-
mental results.
Another method of comparing the samples is to examine the quantity of
material in each of the LC fractions and express this as the percentage of
92
-------
the total LC material for the sample of concern. This is presented in
Table 34, which shows a larger proportion of aromatic hydrocarbons from
samples 1 & 2 (Green sand with seacoal and isocyanate cores) than sample 3
(phenol formaldehyde). On the other hand, the phenolic shell molding pro-
duced a larger proportion of phenols as seen in fraction 6.
The infrared analysis can be compared for the samples from Tables 22,
29, & 31 by determining the percentage of the total sample for each compound
class. This is presented in Table 35, which shows that samples 1 and 2
produced five times the proportion of phenols as sample 3. This discrepency
may be caused by the technique of analysis in which the sample extract is
applied to a NaCl plate, blown dry, and the IR spectrum measured, thereby
losing nearly all TCO material. Samples 1 and 2 were over 90% TCO material,
but the Level 1 analysis only identifies functional groups for the 10% of
material that did not evaporate. Another difficulty involved is that the
procedure requires reading IR spectra of mixtures, which prohibits compound
identification and introduces considerable interference. The technique
specified is such that a compound with a high extinction coefficient (abil-
ity to absorb energy) may be present in small quantities and cause an indi-
cation of high concentration while another compound may be present in large
TABLE 34. COMPARISON OF PERCENT OF EACH LIQUID
CHROMATOGRAPH FRACTION
LC
fraction
1
2
3
4
5
6
7
Fraction percent
Sample 1
16.35
60.10
8.98
0.92
1.50
12.09
0
Sample 2
17.66
60.59
8.45
0.76
1.14
11.59
0
Sample 3
16.50
30.30
7.07
8.42
8.42
28.28
1.01
Compound class types*
Paraffins
Aromatic Hydrocarbons
Polyaromatic Hydrocarbons
Polyaromatic Hydrocarbons
Heterocyclic Sulfur Compounds,
Esters, Ketones, Alcohols
Esters, Ketones, Alcohols, Phenols,
Amides, Carboxylic Acids
Phenols, Amides, Carboxylic Acids,
Sulfonates
92
*Chemical class type found in each fraction.
93
-------
TABLE 35. PERCENTAGE OF EACH COMPONENT IN SAMPLES
(Based on GRAY analysis)*
Category
1. Aliphatics
2. Haloaliphatics
3. Substituted Benzenes
4. Halobenzenes
5. Fused Aromatics
6. Hetero N Compounds
7. Hetero 0 Compounds
8. Hetero S Compounds
9. Alkyl S Compounds
10. Nitriles
11. Aldehydes, Ketones
12. Nitro aromatics
13. Ethers, Epoxides
14. Alcohols
15. Phenols
16. Amines
17. Amides
18. Esters
19. Carboxylic Acids
20. Sulfonic Acids
Total Organ ics, mg/m3
TCO, mg
GRAV, mg
GRAV, mg/m3
TOC, mg/m3
Sample 1
7.3
2.2
24.7
2.4
24.7
5.6
1.0
1.0
0.6
0.1
1.0
0.1
1.0
5.6
5.6
5.6
4.7
1.5
4.6
0.5
173.7
2495
150
9.85
163.8
Sample 2
0
2.7
28.2
2.9
28.2
6.1
1.2
1.2
0.5
0.8
0.8
0.8
0.8
4.9
4.9
4.9
4.9
0.9
4.9
0.5
105.3
2490
265
10.13
94.17
Sample 3
16.7
3.1
19.1
1.9
.19.1
5.8
2.1
2.1
0.4
3.7
2.1
0.2
2.1
4.2
1.1
4.2
3.8
4.2
3.7
0.3
29.7
210
160
12.84
16.86
^Quantities of substances per cubic meter were used to determine the per-
centage of each substance in the samples.
94
-------
concentrations but be read as being present in low concentration due to a
low extinction coefficient. Thus the level one procedure is only an approx-
imation, as was intended.
9.3 COMPARISON OF LABORATORY VERSUS FIELD MEASUREMENTS.
In the work of Bates and Scott, emissions from green sand molding were
measured by two techniques. In the first, a mold was made in a flask con-
sisting of an iron pipe. After pouring the pipe was capped. Gases produced
®
were vented by a tabulation through a cold trap at 0°C and into a Mylar
bag. The second technique utilized an open mold and a portable sampling
hood. After pouring the sampling hood was placed over the mold. This hood
provided a known draft and was equipped with a sampling manifold. The gases
were drawn from the hood through reagent bubblers and grab samples were also
obtained with glass bulbs.
The emission samples were analyzed for cyanide, ammonia, carbon monoxide,
carbon dioxide, methane, ethane, ethylene, acetylene, hydrogen and total
hydrocarbons. In the hood experiments several compounds were so diluted
that they were reported as total hydrocarbons. Ammonia and cyanide were
determined with specific ion electrodes and the other compounds were de-
termined by gas chromatography. Total hydrocarbons were determined by gas
chromatography with an unpacked column, and calibrated with methane-air •
mixtures.
The cold trap condensate contained the higher molecular weight com-
pounds. The organic fraction (about 2%) was separated by silica gel liquid
chromatography into three fractions, aliphatic hydrocarbons, aromatic hydro-
carbons, and solar compounds. These were analyzed by GC-Mass Spectrometry.
Green sand containing 4-6% clay, 1-2% cereal binder, 3-5% seacoal and
other organic additives and 3.5-4% water was used for the tests. The re-
sults of the sealed flasks experiments are given in Table 36.21 The value
of total hydrocarbons includes methane, therefore the higher hydrocarbons
averaged 4.8%. The volume of gas evolved was 5.5 liters per kg cast thus
the emissions of hydrocarbons other than methane was 317 grams per tonne
cast.
By comparison, the sampling performed for this assessment found 610 g
per tonne for sample 1, green sand shakeout. The greater amount found may
95
-------
TABLE 36. RANGES OF DECOMPOSITION PRODUCT CONCENTRATIONS
INTHE EFFLUENT COLLECTED FROM SEALED FLASK EXPERIMENTS
Compound
Hydrogen
Carbon Monoxide
!
Carbon Dioxide
Methane
Total Hydrocarbon
Ammonia
Cyanide
=====:==:=:=========::====================================:
• -: .. ~
Range
32.0%
16.6%
5.2% -
3.9% -
6.8% -
—
- 60.0%
- 23.4%
8.4%
5.5%
11.3%
========
Average
50.2%
21.9%
6.4%
4.5%
9.3%
3__ _ M
ppm
125 ppm
be the result of a good air flow that enabled a free release of vapors in
the shakeout, whereas the sealed flask experiments were limited to those
vapors carried out with the steam evolved. The sealed flask experiments can
only be compared with pouring emissions in foundry practice.
In the work of Bates and Scott, the heavy organics, obtained from the
cold trap, were analyzed by GC-MS. Fourteen polynuclear aromatic and five
polar compounds were identified from over 100 GC peaks obtained. No quanti-
tative data was given.
RTI's sampling and analysis identified 16 compounds not identified in
Bates & Scott's report, but Bates and Scott identified 14 compounds not
identified by RTI's report. In both cases only a fraction of the substances
present were identified. RTI specifically quantified the PNA compounds of
environmental concern, as given in Table 23. Benzo(a and e)pyrenes and
perylene, which were reported by Bates and Scott were not found by RTI.
This may be the consequence of the GC column used, and the fact that only
one column was used rather than a series of columns. Benzo(a, or e)pyrenes
have a molecular weight of 252. RTI did find Chrysene (MW 228) at a concen-
tration of 0.0154 mg/m3 which is 0.007 of the Air Health Mate value. Since
Benzo(a)pyrene has a higher boiling point than chrysene (510° vs 448°C), an
argument can be made that a lower concentration would be expected from the
shakeout.
The comparison of field tests with laboratory tests involves several
difficulties. The best comparison can be made for pouring emissions which
96
-------
can be appropriately simulated in the laboratory. Shakeout emissions will
vary in both quantity and quality with the size and shape of castings, time
required to remove all sand from the casting, air flow over the return sand
belt, and most of all, casting temperature at the time of shakeout. If the
casting is cooled to room temperature, then it can safely be predicted that
no significant quantities of organic vapor will be evolved.
9.4 RECOMMENDATIONS
The findings of this research indicate the need for further data acqui-
sition and a strong recommendation regarding pollution control from shakeout.
9.4.1 Control of Shakeout Emissions
The test results were explained by a proposed mechanism of emissions.
The mechanism presented predicts that shakeout emissions will be reduced
with the temperature of the metal at the time of shakeout. This also coin-
cides with observations of industry personnel. Consequently the industry
should consider extended cooling time as a method of assisting pollution
control and should compare the cost of extended cooling time against the
cost of more extensive air pollution control measures that would be required
if minimum cooling time is allowed. Such considerations will be affected by
the type of casting, quantity and shape of cores, and physical situation of
the individual foundry. Estimating the relative cost and merits of cooling
as a pollution control measure will require testing to determine emissions
as a function of metal temperature. This can be done with "typical types"
of castings, and a graph made of emissions versus metal temperature at
shakeout. From this the metal temperature required to keep emissions below
a target value can be determined. After that, measuring the temperature
versus time during the cooling of a specific casting system will identify
the cooling time required, and from that the required cooling facilities can
be determined.
9.4.2 Pouring Emissions
As indicated previously, the maximum emissions of high molecular weight
(>250) substances, such as benzo(a)pyrene and other PNA's should occur at
pouring. The degree to which these substances are destroyed by the burning
of H2, CH3, and CO emissions that occur shortly after pouring is unknown.
97
-------
When pouring emissions are collected for animal testing, as has been pro-
posed by OSHA, the organization involved could provide samples of the mater-
ial to EPA which should be subjected to GC-MS and other tests specific for
PNA's. If these are found at levels of concern then further research on
pouring is indicated. This should start as a laboratory test, possibly
implemented by hiring the services of a small foundry, in which a flask is
surrounded with a hood, bearing an asbestos board top with a hole for pour-
ing. Provisions should be made to supply nitrogen to the air inlets and to
flood the pouring hole with nitrogen. An appropriate fan system will ven-
tilate the hood and provide for sampling with a high volume sampler. Samples
of pouring emissions can then be obtained under conditions that do not allow
combustion of the emissions. This should be followed with a similar test
using air, with gas flames to ignite the pouring emissions. If indeed the
unignited emissions have an unacceptably high PNA content, and ignition
reduces- this to an acceptable value, then the design of flasks to provide
ventilation of emissions at holes or tubes that allow deliberate ignition of
the gases may be indicated. Under production conditions, the ignition of
pouring emissions may or may not be a dependable event. In cases in which
it it not a dependable event, special arrangements to force the ignition may
provide a substantial reduction in emissions of unacceptable substances.
9.4.3 Inoculation Smoke
As indicated earlier in this report, there is a virtual certainty that
inoculation emissions consist of more than MgO. Furthermore, the nature and
solubility of the MgO produced is not known. Since inoculation emissions
may contain Mg3N2 and Mg02, and definitely must contain oxides of the rare
earth additives to the magnesium alloy, collection and characterization of
inoculation emissions is indicated.
9.4.4 Chromium Emissions
The high concentration of chromium and nickel in the fine (<3(j) partic-
ulates was an unexpected finding. The foundry tries to minimize the level
of these elements and does not known what might be their source. This
indicates that all further testing of iron and steel foundries should pay
careful attention to the metal analysis and an effort to relate the concen-
tration of emitted Cr and Ni to the metal analysis should be made. Labora-
98
-------
tory experiments would determine whether or not Cr and Ni are selectively
volatilized by reaction with organic materials. If indeed certain organic
binders react with Cr and Ni forming volatile metal!oorganics or otherwise
causing Cr and Ni emissions, then the burden of producing binders that do
not enhance these emissions would be upon the chemical binder industry. On
the other hand, if seacoal or simply any organic material produces the same
result, then the emissions problem must be solved by air pollution control
systems. The effect of temperature at the time of shakeout should also be
investigated relative to these metals.
A necessary step in future studies of Cr and Ni emissions should be
verification of the quantity of these metals "extracted" from the stainless
steel SASS train. Published results are needed on the Cr and Ni pick-up by
abrasive particulates and any corrosive attack by the chlorinated solvents
used for rinsing the system.
99
-------
REFERENCES
1. Foundry M&T, "Metal Casting Industry Census Guide," Cleveland, OH, 1976.
2. Data from American Foundrymen's Society, 1978.
3. Moffatt, W. G., G. W. Pearsall, and J. Wulff, The Structure and
Properties of Materials. Vol. 1, J. Wiley and Sons, New York, 1964.
4. A. T. Kearney Company, "Systems Analysis of Emission Control in the
Iron Foundry Industry." Exhibit IV!, USEPA, CPA 22-69-106, February 1971.
5. Smith, C. 0., The Science of Engineering Materials. Prentice-Hall, Inc ,
New Jersey, 196ST
6. Dietert, H. W., "Foundry Core Practice," American Foundrymen's Society,
1966.
7. Heine, H. J., "The Cupola-Dying or Reviving?," Foundry M&T. June 1975,
p. 27.
8. Hancox, D. E. and A. G. Fennel!, "Why Not Make It By the CO. Process?,"
BCIRA Chemical Binders in Foundries Conference. University of Warwick,
1976.
9. Awbery, M., "Automobile Castings by the Hot-Box Process," BCIRA Chemical
Binders in Foundries Conference, University of Warwick, 1976.
10. Truchelut, J., "Ashland Cold Box Process," ibid.
11. Heine, H. J., "Molding with Expendable Polystyrene Patterns," Foundry M&T
May 1977, p. 84.
12. Draper, A. B., "Lustrous Carbon in Molding Sand," AFS Transactions, 84
749, (1976). —
13. Mealier, T., Inorganic Chemistry—An Advanced Textbook, J. Wiley and Sons,
New Jersey, 1952.
14. Wittmoser, A., "The New Third Generation of Molding Processes," AFS
Transactions. 83, 63 (1975).
15. Hoult, F. H. t "The 'Effset' Process," BCIRA Chemical Binders in Foundries
Conference, University of Warwick, 1976.
100
-------
16. Brokmeier, K. H., "Magnetic Moulding Process: Present Position," ibid.
17. Shroyer, H. F., "Cavityless Casting Mold and Method of Making Same,"
U.S. Patent No. 2,830,343 (April 15, 1978).
18. Munton, A. E. and R. K. Ruhr, "Air Evaluation of the Vacuum Molding Process,"
AFS Transactions. 84, 233 (1976).
19- Miura, T. , "Casting Production by V-Process Equipment," AFS Transactions,
84, 233 (1976).
20. Control of Internal Foundry Environment, Vol. 1, American Foundrymen's
Society, Chicago, 1972.
fill Bates, C. W. and W. D. Scott, Better Foundry Hygiene Through Permanent
Mold Casting, NIOSH Final Report, Contract No. 1-R01-OH-00456-01,
January 30, 1976.
22. Boyle, W. C. et al, Foundry Landfill - Leachates from Solid Wastes,
American Foundrymen's Society, 1978.
23. Kunes, T., M. S. Thesis, University of Wisconsin, Madison (Aug. 1975).
24. Kunes, T., AFS Paper, 1975 AFS Convention, St. Louis, Missouri.
25. Heine, R. W., P. C. Rosenthal, C. R. Loper, Jr., Principles of Metal
Casting, McGraw-Hill Publishing Co., (1965).
26. Santa Maria, C., "A Study of Foundry Waste Materials," MS Thesis, University
of Wisconsin, Madison (Jan 1974).
27. Gutow, B. S., "An Inventory of Iron Foundry Emissions," Modern Casting,
January 1972.
28. Heine, R. W., J. S. Schumacher, and R. A. Green, "Bentonite Clay Consump-
tion in Green Sand," AFS Transactions, 1976, pp. 97-100.
29. Butow, B. S., "An Inventory of Iron Foundry Emissions," Modern Casting,
January 1972.
30. Dietert, H. W., A. L. Graham, and R. M. Praski, "Gas Evolution in Foundry
Materials—Its Source and Measurement," AFS Transactions, 1976, pp. 221-228.
31. Elliott, J. F., Thermochemistry for Steel making, Addison-Wesley Publishing
Co., New York, 1960.
32. Eckey, E. W., Vegetables, Fats and Oils, Reinhold Publishing Co., New
York, 1954.
33. Liepins, R. and E. M. Pearce, "Chemistry and Toxicity of Flame Retardants
for Plastics," Environmental Health Perspectives, Vol. 17, 1976, pp. 55-63.
101
-------
34. Sidgwick, N. W., Organic Chemistry of Nitrogen, Oxford University Press
London, 1937. — '
35. Hurd, C. D., The Pyrolysis of Carbon Compounds, The Chemical Catalog
Company, Inc., New York, 1929, p. 607.
36. Coffey, S., Rodd's Chemistry of Carbon Compounds. 2nd ed., Elsevier
Publishing Co., Amsterdam, New York, Vol. 1, Part C, 1964, p. 357.
37. Pryor, W. A., Free Radicals. McGraw-Hill Book Co., New York, 1966, p. 174.
38. Walling, C., Free Radicals In Solution. John Wiley and Sons, New York,
1967.
39. Coffey, S., Rodd's Chemistry of Carbon Compounds. 2nd Ed., Elsevier
Publishing Co., Amsterdam, New York, Vol. 1, Part C, 1964, p. 357.
40. Markely, K. S., Fatty Acids. Interscience Publishers, New York, 1947.
41. Chemical Abstracts. 74:29040 b (1971)..
42. Journal Am. Chem. Soc.. 80, 3301 (1958).
43. Hurd, C. D., The Pyrolysis of Carbon Compounds, The Chemical Catalog
Company, Inc., New York, 1929, p. 707.
44. J. Chem. Soc.. Perkins Trans., 2(6), 1976, p. 704.
45. Encyclopedia of Polymer Science and Technology. Vol. 1, Interscience
Publishing Co., New York, 1964, p. 864.
46. Chemical Abstracts. 70:406t (1969).
47. Hurd, C. D., The Pyrolysis of Carbon Compounds. The Chemical Catalog Company,
Inc., New York, 1929, p. 160.
48. Ott, E., Cellulose and Cellulose Derivatives, Part 1, Interscience Pub-
lishers, New York, 1954, p. 46, 174.
49. Kirk and Othmer, Encyclopedia of Chemical Technology, Vol. 22, Inter-
science Publishers, New York, 1970, p. 169.
50. Ettre, E. and P. F. Varaki, Anal. Chem.. 35, 1963, p. 69.
51. Isuchiyn, Y. and K. Sumi, J. Polym. Sci. (A-l), Vol. 7, 1969, p. 3151.
52. "Thermal Decomposition of Furfuryl Alcohol Resins," NASA Technical Report
N63-22125, 1963.
53. Salsberg, H. K. and J. J. Greaves, "Phenolic Resin Bond in Solid Sand
Cores," AFS Transactions. Vol. 68, American Foundrymen's Society, Des
Plaines, 111., 1960, pp. 387-396.
102
-------
54. Bates, E. E. and L. D. Scheel, "Processing Emissions and Occupational
Health in the Ferrous Foundry Industry," Am. Ind. Hyg. Assoc. J., August
1974, pp. 452-462.
55. Boettner, E. A., G. L. Ball, and B. Weiss, Combustion Products From The
Incineration of Plastics, NTIS Pb-222 001, EPA-670/2-73-049, 1973.
56. Lee, H. and K. Neville, Handbook of Epoxy Resins, McGraw-Hill, New York,
1967.
57. Bates, C. E. and W. D. Scott, "The Decomposition of Resin Binders and
the Relationship Between Gases Formed and the Casting Surface Quality.
Part 2—Gray Iron," AFS Research Progress Report, AFS Transactions,
American Foundrymen's Society, Des Plaines, 111., 1976, pp. 793-804.
58. Tubich, G. E. , "The Potential Health Hazards of the New Oil Base No-Bake
Binders," AFS Transactions, American Foundrymen's Society, Des Plaines,
111., 1966, p. 448-453.
59. Bott, B., J. G. Firth, and T. A. Jones, Br. Polymer. J., Vol. 1, 1969,
p. 203.
60. Hirasa, 0., Bulletin of Research Institute for Polymers and Textiles
(Japan), No. 112, 1976, p. 11.
61. Mark, H. F. (ed.)., Encyclopedia of Polymer Science and Technology,
Interscience Publishers, New York, 1969, p. 28.
62. Ruff, W. J., Fibers Plastics and Rubbers, Butterworth Scientific
Publications, London, 1956., p. 14.
63. Hurd, C. D., The Pyrolysis of Carbon Compounds, The Chemical Catalog
Company, New York, 1929, p. 278-9, 284.
64. Chemical Abstracts, 67:65576v, 1967.
65. Stanbridge, R. P., "The Replacement of Seacoal in Iron Foundry Molding
Sands," AFS Transactions, American Foundrymen's Society, Des Plaines,
111. , 1974, pp. 169-180.
66. Commins, B. T. , Atmospheric Environment, Vol. 3, 1969, p. 565.
67. Mark, H. F. (ed.), Encyclopedia of Polymer Science and Technology, Vol. 1,
Academic Press, Inc., New York, 1965, p. 421.
68. Horton, B., "Pyrolysis of Starch," Starch: Chemistry and Technology,
Vol. 1, Academic Press, Inc., New York, 1965, p. 421.
69. Banerfee, D. K., Bull. Nat. Inst. Sci. India, Vol. 37, 1968, p. 114.
70. Walling, C., Free Radicals In Solution, John Wiley and Sons, New York,
1957, p. 446.
103
-------
71. Gough, T. A., R. Tarrest, and E. A. Walker, "The Pyrolysis and Hydro-
genation of Alkylbenzenes," Journal of Chromatography, 48(3), 1970,
pp. 521-3.
72. Kirk and Othmer, Encyclopedia of Chemical Technology. Vol. 10, Inter-
science Publishers, 1968, p. 279.
73. Ganman T. and J. Hoigne, Aspects of Hydrocarbon Radiolysis. Academic
Press, New York, 1968, Chapter 3.
74. Hurd, C. D., The Pyrolysis of Carbon Compounds. The Chemical Catalog
Company, New York, 1929, p. 188.
75. Ibid.. p. 98.
76. Ibid., p. 687.
77. Ibid., p. 101.
78. Ibid., p. 104.
79. Chemical Abstracts, 84:P138260a, 1976.
80. Hurd, C. D., The Pyrolysis of Carbon Compounds, The Chemical Catalog
Company, New York, 1929, p. 115-116.
81. Ralston, A. W., Fatty Acids and Their Derivatives. John Wiley and Sons,
New York, 1948.
82. Encyclopedia of Polymer Science and Technology. Vol. 7, Interscience
Publishers, New York, 1967, p. 41.
83. Chemical Abstracts, 66:18818v, 1967.
84. Chemical Abstracts. 69:76323c, 1968.
85. Stanbridge, R. P., "The Replacement of Seacoal in Iron Foundry Molding
Sands," AFS Transactions. American Foundrymen's Society, Des Plaines,
111., 1974, pp. 169-180.
86. Commins, B. T., Atmospheric Environment. Vol. 3, 1969, p. 565.
87. Morton, B., "Pyrolysis of Starch," Starch: Chemistry and Technology.
Vol. 1, Academic Press, Inc., New York, 1965, p. 421.
88. Heine, H. J., "Joint AFS/OIS Conference Studies Ductile Iron,"
Foundry M&T, December 1975, P. 72.
89. Hillner, G. F. and K. H. Kleeman, "Mold Inoculation of Gray and Ductile
Cast Iron—New Solutions to Old Problems," AFS Transactions. 83, 167 (1975),
104
-------
90. Midwest Research Institute, "A Study of Fugitive Emissions from
Metallurgical Processes - (Iron Foundries)," EPA Contract 68-02-2120.
91. Cleland, J. G. and G. L. Kingsbury, "Multimedia Environmental Goals for
Environmental Assessment, Vol 1 and Vol 2. EPA-600/7-77-136 a&b,
November 1977.
92. Dorsey, J. A., L. D. Johnson, R. M. Statnick and C. H. Lochmuller,
"Environmental Assessment Sampling and Analysis: Phased Approach and
Techniques for Level 1." • EPA-600/2-77-115, June 1977.
93. Toeniskoetter, R. H. and R. J. Schafer, "Industrial Hygiene Aspects
of the Use of Sand Binders and Additives," BCIRA Report 1264, 1977.
105
-------
APPENDIX A
DECOMPOSITION PRODUCTS OF SOME SUBSTANCES
USED IN MOLDS AND CORES
106
-------
APPENDIX A
DECOMPOSITION PRODUCTS OF SOME SUBSTANCES 'USED IN MOLDS AND CORES
Substance
Decomposition Products
INORGANIC-ORGANIC COMPOUNDS:
Tetraethyl Silicate
Polydimethylsi 1oxane
(silicone)
Calcium Stearate
Polyphosphate Esters
0,0-diethyl-n,n,-bis (2-hydroxy-
ethyl) aminomethyl phosphonate
At 300°C:
300°C:
400°C:
A ketone
formaldehyde
silica
ethylene
water and carbon dioxide
formaldehyde
silica
31
Carbon dioxide
Methane
Ethane
Ethylene
32
Propylene
Phosphine
Toluene
Benzene
Phosphorous pentoxide
Carbon dioxide
Water
Carbon monoxide (in absence of Op)
Potential for highly toxic materials
Upon burning: 4-ethyl-l-phospha-
2,6,7 rioxabicyclo
(2,2,2) octane-1-oxide1
(a toxic organophosphorus compound)
107
(continued)
-------
Substance
Decomposition Products
ORGANIC MONOMERS:
Urea
Thiourea
Ammonium Thiocyanate or
Thiourea
Pseudocumene
Ethyl Alcohol
Cyanic acid and ammonia
At 132°C: biuret which then forms
tricyanourea
(CN-NH-CO-N(CH)* )
or ammonia + cyanic acid
Ammonium cyanate-in absence of water
Alkyl isocyanates
35
Ammonia
Thiocyanic acid
High temperature, oxidizing conditions:
ammonia
carbon dioxide
sulfur dioxide and/or
hydrogen sulfide
At 140°C in the presence of water:
ammonium Thiocyanate
At 180-190°C: Guanidine thiocyanate
At 200-300°C: me!am
carbon disulfide
Benzene
Toluene
Methane
Dimerization products such as:
1,3-(3,4-dimethylphenyl) ethane
2,3-4-trimethylphenyl-3,4-dimethyl-
phenyl methane
3,3',4,4'-tetramethyl biphenyl
Below 400°C: ethylene
methane
glycols (e.g., 2-3 butane
glycol)
(continued)
108
-------
Substance
Decomposition Products
Ethyl Alcohol (cont'd)
Above 800°C:
Stearic Acid
Toluenesulfonic Acid
Benzenesulfonic Acid
Oleic Acid
Hexamethy1enetetrami ne
ethylene
water
acetaldehyde then methane
and carbon dioxide
hydrogen
carbon dioxide (in oxidizing
conditions)
Above 300°C: hydrocarbons (including
40
methane)
At 650°C under nitrogen:
homologus series of mono-
alkenes. Highest is heptadec-
1-ene.
At 400°C: sulfur dioxide
substituted phenols (o,m,p
cresols)
biphenyl derivatives (e.g.,
2-methyl biphenyl,
• 3-methyl biphenyl,
4-methyl biphenyl)
41
possibly toluene
Sulfur dioxide
Substituted phenols
Biphenyl derivatives
Benzene
Distillation yields hydrocarbon and
43
phenyl sulfones
Azelaic acid
Carbon dioxide
44
Hydrocarbons
Ammonia
Formaldehyde
45 46
Carbon-rich residue '
(continued)
109
-------
Substance
Decomposition Products
Hydro! (Tetramethyldia-
mine-benzhydeol)
Binaphthyl
ORGANIC POLYMERS:
Graphite
Dextrin
Waxes (long chain alcohol
esters of fatty acids)
Polyvinyl Alcohol
Furan Resins (furfuryl
alcohol resins)
No information available
Dimers of binaphthyl47
Oxidizes above 400°C
Carbon monoxide
Above 500°C:
carbon dioxide
carbon monoxide
p - blucosan
methane
ethane
Linoleic acid
ethylene
48
Myristic acid
Oleic acid
Hexadodecane
Dodecene
1,9 - Octodecadiene
Ethylene
Ethane methane
J\(\
Carbon dioxide
At 500-800°C: acetaldehyde
crotonaldehyde
benzaldehyde
acetophenone
carbon monoxide
benzene50 _.
tolueneou'01
Carbon monoxide and dioxide
Ethylene
Ethane
Propylene
(continued)
110
-------
Substance
Decomposition Products
Furan Resins (furfuryl
alcohol resins) (cont'd)
Phenol Formaldehyde
Phenolic Resins
(Novalak and Resole)
Propane
Furan
Methanol
Ethanol
Methane
Hydrogen and water
52
At 620°C: carbon monoxide and dioxide
hydrogen
methane
phenol
formaldehyde
ammonia ro 54
hydrogen cyanide '
acetylene
ethylene
55
ethane00
Same as phenol-formaldehyde plus:
Allene
Methyl acetylene
Propylene
Acetaldehyde
Methyl chloride
Acrolein
Acetone
Propionaldehyde
Vinyl chloride
Ethyl chloride
Cyclopentadiene
Benzene
Methylcyclopentadiene
Toluene
Cresols
Methylenediphenol
(continued)
111
-------
Substance
Decomposition Products
Phenolic Resins
(Novalak and Resole) (cont'd)
Phenolic-Urethane
Alkyd-Urethane
Cp phenols
Ethylene diphenol
C~H9 phenol
56
Propene
Acetylene
Carbon monoxide and dioxide
Ethane
Ethylene
Hydrogen
Methane
The nitrogen in the isocyanate should
yield:57
ammonia
simple amines
aniline
hydrogen cyanide
The phenolic component should produce:
formaldehyde
substituted phenols
Carbon monoxide and dioxide
Nitrous oxide
Hydrogen cyanide
Benzene
Toluene
Methane
Acetylene
Hydrogen
Ethane
Ethylene
Ammonia
Simple amines
Possibly aniline
57,58
(continued)
112
-------
Substance
Decomposition Products
Alkyd-Urethane (cont'd)
Urea-Formaldehyde
Polystyrene
Alkyd Resins (mixture of poly-
functional alcohols, dibasic
acids, styrene, and filler)
Methylene dephenyl isocyanate has been
CO
identified in shakeout
At 610°C: carbon monoxide and dioxide
hydrogen cyanide
methane
ammonia
nitrogen oxides ^
unidentified substances '
At 450°C; benzene
toluene
ethylene
styrene
benzaldehyde
crmel thy! -styrene
phenol
methylstyrene
n-propyl styrene
i ndene
acetophenone
methyl indine
naphthalene
cinnamyl alcohol
methyinaphtha!ene
biphenyl or acenaphrhene
methylbiphenyl
diphenylethane
methane
ethyl benzene
hydrogen55'60
Phthalic anhydride
Maleic acid
Fumaric acid
Ethylene glycol
Ethylene
Propylene
Cyclohexane
Carbon dioxide
(continued)
113
-------
Substance
Decomposition Products
Wood Flour
Pitch
Methane
Products of benzoic acid if it is in the
resin.
Above 400°C: formaldehyde
acetone
glyoxal
formic acid
acetic acid
lactic acid
glycolic acid 52
glycolaldehyde
Pyrolysis of lignin produces:
acetic acid
methanal
phenol ethers (e.g., methyl
phenyl ether, ethyl phenyl
ether, diphenye ether).
phenol derivatives (e.g.,
cresol isomers, ethyl phenols)
carbon
tars
hydrocarbons 6?-fi4
carbon monoxide and dioxide
Pyrene
Fluoranthene
Benzo(a)pyrene
Benzo(e)pyrene
Benzo(ghi)pery1ene
Anthanthrene
Coronene
Methane
< Cg hydrocarbons
Benzene
3-Methyl hexane
Toluene
(continued)
114
-------
Substance
Decomposition Products
Linseed Oil
Cereal (corn and wheat flours)
3-Ethyl hexane
3-Methyl heptane
Nonenes
Cumene
Pseudocumene
Prophenyl benzene
1-3-Diethyl benzene
Ethyl-m-xylene
Amy! benzene
Hexahydro naphthalene
Isohexyl benzene
Napththalene
Pentamethyl benzene
1,2,4,5-Tetraethyl benzene
0-m-Bitoyl
Acenapthene
Phenanthrene65'66
Acrolein
Myristic acid
Palmitic acid
Stearic acid
Oleic acid
Linoleic acid
Linolenic acid
C, to C,fi hydrocarbons
Methane (predominant)
Carbon dioxide
67
Carbon monoxide and dioxide
Acetaldehyde
Acrolein
(continued)
115
-------
Substance Decomposition Products
Acetone
Butanone
2-or 3-Methyl furan
2,3-', 2,4; or 2,5-Demethyl furan
Acetic acid
Methyl ethyl or ethyl ethyl furan
Aliphatic amines (methyl-ethyl-, propyl-,
and butyl-amine)
Phenolics (e.g., cresols, ethyl phenols,
xylenols, and dihydroxybiphenyls)62
Rosin l,2-Dimethyl-l,2,3-trans, trans-
CO
cyclohexanetricarboxlic acid
(Rosin pitch) Benzene
< Cg hydrocarbons
Methyl eyelohexene
2,4-Heptadiene
Toluene
1,4-Dimethylocyclohexane
3-Methylheptane
2,6-Dimethytheptane
Xylenes
Methyloctadiene
Cumene
Isopropylcyclohexane
Ethyl toluene
Mesitylene
Isopropyltoluene
Diethyl benzenes
Ethyl Xylenes
3,4-Diethyltoluene
0-Butyltoluene
(continued)
116
-------
Substance
Decomposition Products
Kerosene
alkane components
aromatic components
Fuel Oil (C14 to C26 hydro-
carbons
Coal Tars
(toluene and naphthalene
produced form)
(phenols produced forms)
(fluorene produced forms)
(n-methylcarbozole content
forms)
Hexahydronaphthal ene
Pentamethy1 benzene
Phenylcyclohexane
2-Methylnaphtha! ene
1,2,4,5-Tetraethyl benzene
0,m-Bitolyl
65
Phenanthrene
Low molecular weight hydrocarbons
predominantly methane
In low oxygen environment:
pyrolyze to dimers
dibenzyl ethane
biphenyl
alkylbenzene series
(i.e., methyl, ethyl, propyl, butyl,
amyl substitutions)
alkylcyclohexane series
(i.e., methyl, ethyl, propyl, butyl,
amyl substitutions)
Lower hydrocarbons
In presence of oxygen:
oxygenated derivatives of hydrocarbons,
(e.g., acetaldehyde, acetic acid, etc.)
See products of pitch 1,3-Binaphthylethane if
enough CL present
Phenyl-1-naphthy1 methane
74
p-Hydroxy-diphenyl
Difluorenylene
Rubicene
Dihydrorubicene
Phenthridine
73
(continued)
117
-------
Substance
Decomposition Products
(anthracene content forms)
(p-xylene content forms)
Synthetic Asphalt
Gilsonite (one of the purest
natural bitumins)
.78
Dianthryl77
p-Dixylyl dimethyl
anthracene
p,p'-Dimethyl-stiIbene'
Benzene
2,5-Dimethyl-l,5-hexadiene
Toluene
Octadiene
Ethyl benzene
Hydrocarbons ^-C^, C^, C
Styrene
Ethyl toluene
Misitylene
Pseudocumene
Butyl toluene
Tetrahydronaphthalene
a-Hexahydroanthracene
Phenanthcene
Anthracene
Pyrene
Fluoranthene
Benzo(a) and (e)pyrene
Benzo(ghi)perylene
Anthanthrene
Coronene
Carbon monoxide and dioxide
Benzene insolubles
Quinoline insolubles66'79
Benzene
< CQ hydrocarbons
C olefins
(continued)
118
-------
Substance Decomposition Products
3,4-Dimethyl hexane
1,4-Dimethyl cyclohexane
0-Xylene
Styrene
Toluene
Ethyl benzene
Propyl benzene
Ethyl toluene
Misitylene
Isobutyl benzene
Isopropyl toluene
Diethyl benzene
Butyl benzene
Ethyl xylene
p-Butyl toluene
1-Methyl anthracene
Naphthalene
Penta methyl benzene
1-Methyl naphthalene
2-Methyl naphthalene
1-Ethyl naphthalene
Diphenylmethane
Acenaphthene
m-m'-Bitolyl
Fluorene
Stilbene
Phenanthrene
Pyrene
Fluoranthene
Benzo(a) and (e) pyrene
Benzo(ghi)perylene
(continued)
119
-------
Substance
Decomposition Products
Petroleum Oil
Mineral Spirits
Seacoal (finely ground coal)
Anthanthrene
Coronene
Methane
Carbon monoxide65'66
Lower chain aliphatics
Lower chain olefins
Alkyl substituted benzenes
Products similar to kerosene
Benzene
Toluene
Xylene
Naphthalene
Anthracene
At 750-1000°C: methane (44.8%)
hydrogen (20.5%)
ethylene (16.2%)
propylene (11.9%)
on
other products
Low member hydrocarbons and olefins
List approaches 1000. Literature identifies:
< C5 hydrocarbons
hexene
benzene
trimethyl benzenes
2,3-dimethyl pentane
3-methyl hexane
toluene
3-ethyl hexane
m- and p-xylene
4-ethly-O-xylene
3-methyl octane
pseudocumene
phenol
indene
napththalene
4-ethyl-O-xylene
120
(continued)
-------
Substance Decomposition Products
~~~ ~ cresols
xylenols
dicylo-hexyl
1 -ethyl naphthalene
1 ,4-di methyl naphthalene
acenaphthane
1-naphthol
1,1-binaphthyl
fluorene
anthracene
phenanthrene
binaphthyl
tetraphenyl ethane
9-phenyl anthracene
tetraphenylethylene
pyrene
fluoranthene
benzo(a)pyrene
benzo(e)pyrene
benzo(ghi)perylene
anthanthrene
coronene
-A 65,66
carbon dioxide
Gluten Carbon dioxide
Acetic acid
Aliphatic amines (methyl or ethyl)
36
Phenolics (cresols or ethyl phenols)
Soy Oil Acrolein
Methane
Ethane
Ethyl ene
Malonic acid
32
Other oxygenated derivatives
Fish Oil Carbon dioxide
Methane series hydrocarbons
Olefins (principally ethylene)
81
Unsaturated acids
(continued)
121
-------
Substance ' Decomposition Products
Molasses (sugar content) Formaldehyde
Acetone
Glyoxal
Glycolaldehyde
Glycolic acid
Lactic acid
Formic acid
oo
Acetic acid
At 330°C: Furfural At 700°C:
5-methyl furfural FLO
carbonyl compounds CCL
po
acids, others
122
-------
APPENDIX B
LEVEL 1 ORGANIC ANALYSIS DATA OF SAMPLES 1-3 AND
INORGANIC ANALYSIS DATA
123
-------
TABLE B-l. STACK DATA, SAMPLES 1 and 2
Properties of
Sampling Locations
Purpose of stack
Width ft.
Length ft
IIM.lll.il . . ,
Diameter ft, I. D.
"
Wall thickness in.
——————^__
Material of construction
3.083 Dia.
~l/8 in. I -1/8 in.
Steel
a, mode
3k in
a. Existing
b. Size opening
e. Distance from platform
8 ft. above floor
Straight distance before port
Type of restriction
Straight distance after port
Type of restriction
Ambient temperature °F
Average pitot reading H70, in Hg
Approximate stack velocity ft/min.
Approximate stdft3/min.
Approximate moisture % by volume
—• .
Approximate stack temperature °F
.
Approximate paniculate loading gr/SCF
——• _
Approximate particle size
Approximate composition gases present
Approximate stack pressure H-0. in Hg
~30 in Hg
Used for Sample #
124
-------
TABLE B-2. SASS TRAIN DATA, SAMPLE 1
Compar
.y/Location
Sampling Location Duct leading
Green sand, prescrubber
. *• .
from shakeout,
Run # •
Date 6-28 Test Participants FJP, BH. EES
• H
•'3
^ •<
ro
en
"•.,,'11.
P n
5 5
n r
H
2
Anibient Temp. 90 Bar. Pressure
Clock
Time
0
Dry Gas
Mctor/Cu.Ft.
120.185
20 194.2
40
269.6
fin 344.8
80
Pitot
Manometer
Ap
0.69
n.69
0.69
0.69
422.1 0.69
100 502.9
120
140
155
582.2
0.69
0.69
661.5 0.69
720.605 I
Diff. 600.420
t
Average
(=15.23 m3
29.90
Est. Xois
Pitot Lea
jure 3 Hozzle (in) U.49/
< Test Good
Sampling Train Leak Test 0.035 - 0.085
Averarjo A;
SaiT.pl ir.g :
Start \L
0.696
?oint A-3
kOO Finish 16:50
TEMPSRATuRE
Setting . MODUL2 IKPINGERS
GAS lYSTflK
I.XLKT
1.69 ! 69 94 inq
1.69 69 91
1?3
1.69 69 87 126
1.86 70 82
1.89 j 61 80
126
129 -
1.89 ! 65 i 78 130
1.89 71 77 i 132
1.89 64 73
i i
!
\
!
0.69
134
• CUTLET
.102
inq
STACK OVEM
155 220
134 95?
114 144 251
116
118
120
121
122
165 252
176 253
147 252
161 251
-> "~j
257
250
250
250
250
250 !
170 252 250
i i
!
!
i i
1.81 67 83 126
'Flowrate= 3.874 cfm - 3.47 dscfm '
115 j 156 249 251
i i
1 i
111% isokinetjic •
j
i
i
' ! i
!
. ! '
1
i
'
-------
TABLE B-3. VELOCITY TRAVERSE DATA AND CALCULATIONS, SAMPLE 1
Plant
Location just, Shakeout, Green sand,
Date 6-^8-78
71 me 1Q:QQ
Prescrubber
Initials pjp
Barometric Pressure 29.90 Moisture Content 3%
Duct Dimensions _4Q.5" Dia. Pitot Tube-Factor .84
POINT
NUMBER
1
2
3
4
5
6
7
8
9
10
11
12
sum
DISTANCE:
IN
i
I 3/4
4 3/4
7 1/8
10 1/8
14 3/8
- 26
30 3/8
33 3/8
35 3/4
37 3/4 ,
39 5/8
A
0.48
0.69
0.71
0.75
0.81
0.80
0.65
0.60
0.61
0.57
0.57
0.57
/
B
0.56
0.62
0.67
a^8___
0.63
0.62-
0.78
0.84
0.86'
0,90
0.94
0.93
P
C
D
A
0.693
0.831
0.843
n.Hfifi
0.900
0.894.
0.806
0.774
0.781
0.755
0.755
0.755
9.653
7A
B
0.748
0.787
0.818
0 8?5
0.794
0-787
0.883
0.916
0.927
JL.949
0.969
0.964
HL3fiZ_
P
C
D
Average /&"p
0.834
Molecular Weight _2£L84
Gas Velocity
3043 _ ft/r.iin
~U-S! -
Average Temperature 156 °p
Ib/lbmol Duct Area 3. 94 ft2
,Ts.,,n
Volumetric Flow Rate 27218
Volumetric Flow Rate 22304
ft /nn'n @ stack conditions
ft/min 0 standard conditions
126
-------
TABLE B-4. SASS TRAIN DATA, SAMPLE 2
Company/Location
Est. Moisture
12
Nozzle (in) 0.370
ro
— :i1
&-„
s :>
O r
m -
Pitot Leak Test
sampling Location Scrubber Outlet Stack,
from Green sand shakeout
Date 6-29 Test Participants FJP, BH, EES
Ambient Temp. 90 Bar. Pressure
Clock
Time
0
40
Dry Gas .
Meter, Cu. Ft.
724.210
840.500
80 961.50
. Pitot
Map.or.ctcr
Ap
2.10
2.10
2.10
JL25 092.40 | 2.10
16b
205.80 2.10
^00 308,00
?Kn 4fi7 on
290
i 330
337.'
579.90
702.99
725.035
2.10
'/.in
2.10
2.2
29.74
Run # •
Sampling Train Leak
Averaqo A:> 2.10
Good
rest 0.050 0.080
Sampling Point A-3
Start 11
:27 Finish 17:07
TEMPERATURE
Sotting j . MODI
1.00 6
1.00 6
JLS IKPINGEKS
7 104
9 86
CAS K:/:T!:"<
INLET
98
103
.1.00 i 64 77 109
1.00 ! 75 79
1.00 6
4 71
1.00 65 73
1 . 00 €
1.00 1
1.00 7
»3 66
6 66
1 i 70
102
101
107
106
106
104-
OUTLET
97
101
105
105
102
104
106
105
105
STACK
109
110
105
OVEN
242
256
254
PROBE
256
254
245 j
104 2b2 2bb
109
113
112
112
108
253
253
251
248
247
253
250 250
250
246
i
i
A=1000.83 (=2$. 15m3)
Avg.
I
j 2.10
Flowrate = 2.966 c
78% isokinetjic
i
|
l
1.00 j 68 77
Fm
104
103
;
109
251 250
i
-
-------
TABLE B-5. VELOCITY TRAVERSE DATA AND CALCULATIONS, SAMPLE 2
Plant
Location Scrubber Outlet Stack
Date 6-2-78
Time 1Q:3Q
Initials
FJP
Barometric Pressure J29..75 Moisture Content 12%
Duct Dimensions 37" dia.
Pitot Tube-Factor 0.84
POINT
NUMBER
1
2
3
4
5
6
7
8
9
10
11
12
sum
DISTANCE:
IN
1.6"
5.4"
10.9"
26.0"
31.6"
35.4"
i
A
203
2.70
2.15
1.70
1.75
1.65
t
B
2.25
2.45
2.00
2.00
2.55
2.45
P
C
D
. . ..
A
1.42
1.64
1.47
1.30
1.32
1.28-
8.42
Jl
B
1.50
1.56
1.41
1.41
1.50
1.56
9.04
P
C
o
D
Average /A~p 1.45
Average Temperature
109 °F
Molecular Height 28. R4 Ih/lbniol Duct Area 7
" "'
Gas Velocity 5077 _ ft/nin
- - --
ft
yT£> IM
Volumetric Flow Rate 37873
V o 1 uine t r i c Flow Ra t e _J3Q459
128
ft /min @ stack conditions
ft /min P standard conditions
-------
TABLE B-6. SASS TRAIN DATA, SAMPLE 3
Company/Location Phenolic Shell Molding
Foundry
sampling Location Shakeout room, fugitive
Date fi-30 Tost Particigants_FJP. BH, TT
Ambient Ter?.p. 95 Bar. Pressure 29.80
Run
Est. Xoisuuro
1-Joazle (in)
Pitot Leak Test
Sampling Train Leak Test_
Average A;.> ~ .
Sampling Point ^
Start 11:05
Finish 12:52
Clock
Tine
0
15
30
. 45
60
75
90
101
Diff.
Averac
Dry Gas
Motor, Cu. Ft.
734.885
814.6
886.34
958.10
030.85
104.25
177,84
234.905
500.02
ie
Pitot
Manometer
AD
In.Jl20
-
(=12.47 mj
m
Setting
2.8
2.8
2.8
2.8 |
2.9
3.0
3.0
)
2.87
. ;/iODUL"
65
67
62
64
65
65
66
65
IXPINGERS
103
92
88
89
82
76
72
86
THy^*v
GAS t
INLET
125
138
135
146
146-
148
149
-
140
"uYTuro:-:
• ••/rr:R
OUTLET
103
112
118
125
129
131
133
121
STACK
ovs>;
1
?***rt •"»»•*
i\OiJ£.
1
1
i
ro
-------
TABLE B-7. LC ANALYSIS REPORT, SAMPLE 1
contractor Research Triangle Institute
Sample Site Duct 5
Sample Acquisition Date 28 June 1978
Type of Source Shakeout, Greensand-Isoc.yanate molding
Test Number
Sample ID Number 6282-G (XR)
Sample Description Sorbent extract, shakeout, green sand, line 5
Original Sample Volume or Mass IS. 23m3 Std. , dry
Responsible Analyst
Date Analyzed
Calculations and Report Reviewed By
Report Date 31 August 1978
Column Flow Rate.
Observations
Column Temperature.
Fraction
1
2
3
4
5
6
7
Sum
Total Sample''
Taken for LC^
Recovered3 -r
TCO
mg
2495
74.9
78.8
GRAV
mg
150
4.5
4.1
TCO^inmg
Total
421
1539
209
25
25
276
0
Blank
0
0
0
0
0
0
0
1
Corrected '
421
' 1539
1 209
- 25
25
276
" ' 0
2495
Total
mg
2645
79.4
82.9
Concentration^;
mg/m3 i
173.7
5 2
5.4
GRAV4 in mo
Total
11
66
51
15
15
66
15
Blank
15
22
15
0
22
22
?
Corrected '
11
51
29
0
15
44
0
150
Total* I
mg I
i
432
1590
238
25
40 ,
320
0
2645
Concentration**
mg/m3
28.4
i rid A
Ti fi
i 7
2 6
21 0
0
173.7
1. Quantity in entire sample, determined before LC
2. Portion of whole sample used for LC, actual mg
3. Quantity recovered from LC column, actual mg
4. Total mg computed back to total sample
5. Total mg divided by total volume
130
-------
TABLE B-8. ORGANIC EXTRACT SUMMARY
1. Shakeout, Green Sand, Line 5
3
Total Organic;, mg/m
TCO.mg (94.33%)
GRAV, mg (5.67%)
LC1 '
28.4
421
11
LC2
104.4
1539
51.2
LC3
15.6
209
29.3'
LC4
' 1.6
25
0
LC5
2.6
25
14.6
LCG
21.0
276
43.9
LC7
0
:.0
0 v
S
173.7
2495
750
GRAV, mg/m3 . _ 0.7 3.4 1.9 0 1.0. -2.9 0 9.85
u>
Category Int/mrj/m
Aliphatics
Haloaliphatics
Substituted Benzenes
Halobenzenes
Fused Aromatics
Hetero N Compounds
Hetero 0 Compounds
Hetero S Compounds
Alkyl S Compounds '
Nitriles
•Aldehydes, Ketones
Nitroaromatics
Ethers, Epoxides
Alcohols
Phenol s
Am i n es
100/0.72
10/0.07
*
10/0.15
•100/1.53
10/0.15
100/1.53
100/0.92
10/0.09
100/0..92
100/0. 1C
100/0.10
100/0.10
10/0.01
10/0.01
100/0.10
10/0.0
100/0.10
100/0.10
100/0.10
100/0.10
100/0.46
10/0. OE
100/0.46
100/0.46
100/0.46
0.72
0.22
2.45
0.24
•2.45
0.56
0.10
0.10
0.06
0.01
0.10
0.01
0.10
0.56
0.56
0.56
-------
TABLE B-8 (cont'd)
Sample 1.-. Shakeout, Green Sand, Line 5
Total Organics, mg/m
TCO, rng
GRAV, mg
VUIU^JII. .11., ,
LC1 '
LC2
if
LC3
LC4
•
LC5
LCG
LC7
J
''•
2
Category
Int/mg/m3
Amides
Esters
Carboxylic Acids
Sulfonic Acids
p
•
-------
TABLE B-9. COMPOUND CATEGORIES POSSIBLE IN DIFFERENT LC FRACTIONS
(NUMBERS IN PARENTHESIS REFER TO LC FRACTION DESIGNATION)
LC FRACTION 1
Aliphatic Hydrocarbon (1)
Halogenated Aliphatics (1,2)
LC FRACTION 2
Halogenated Aliphatics (1,2)
Monoaromatic Hydrocarbons (2,3)
Halogenated Monoaromatic Hydrocarbons (2,3)
Polyaromatic Hydrocarbon, MW < 216 (2,3)
Polyaromatic Hydrocarbon, MW > 216 (2,3)
LC FRACTION 3
Monoaromatic Hydrocarbons (2,3)
Halogenated Monoaromatic Hydrocarbons (2,3)
Polyaromatic Hydrocarbons, MW < 216 (2,3)
Polyaromatic Hydrocarbons, MW > 216 (2,3)
LC FRACTION 4
Heterocyclic N Compounds (4,6)
Heterocyclic 0 Compounds (4)
Heterocyclic S Compounds (4)
Nitriles (4)
Ethers and Epoxides (4)
Aldehydes and Ketones (4)
Nitroaromatic Hydrocarbons (4,5)
LC FRACTION 5
Heterocyclic N Compounds (4,6)
Heterocyclic 0 Compounds (4)
Heterocyclic S Compounds (4)
Alky! Sulfur Compounds (6)
Nitriles (4)
Aldehydes and Ketones (4)
Ethers and Epoxides (4)
Nitroaromatic Hydrocarbons (4,5)
Alcohols (6)
Phenols (6)
Amines (6)
Amides (6)
Esters (6)
133
-------
TABLE B-9. (cont'd)
LC FRACTIONS 6 AND 7
Phenols (6)
Esters (6)
Amines (6)
Heterocyclic N Compounds (4,6)
Sulfonic Acids and Sulfoxides (7)
Carboxylic Acids (6,7)
Alcohols (6)
Amides (6)
TABLE B-10. IR REPORT— SAMPLE NO. 1, CUT
v> cm
Quantity
I
Not
Total Sample
Assignment
Sufficient
*Since Aliphatics are consistently
is assigned to that category.
GRAV = 11.0 mg
Possible Categories
Aliphatics
Haloal iphatics
shown in this fraction,
TABLE B-ll. IR REPORT— SAMPLE NO. 1, CUT
v» cm
3030,3053
2865-2971
1632
1603
1509
1445-1456
1308
1034
699-875
I
S
S
W
S
S
S
M
M
S
Total Sample
Assignment
CH, aromatic/
olefinic
CH, aliphatic
CH, olefinic
C=C, aromatic
C=C, aromatic
CH, aliphatic
CH, aliphatic
CH, aromatic
Multiplet
GRAV = 51.2 mg
Possible Categories
Haloal iphatics
Substituted Benzenes
Halobenzenes
Fused Aromatics
LC-1
Max. Wt. in
I Total Sample
100 10.0 mg*
10 1.0
the total GRAV weight
LC-2
Max. Wt. in
I Total Sample
10 2.33 mg
100 23.27
10 2.33
100 23.27
134
-------
TABLE B-12- IR REPORT-SAMPLE NO. 1, CUT LC-3
v, cm"
3024,3065
2871-2971
1603
1497
1456
1380
1034
670-881
I
S
S
S
S
S
w
w
S
Total Sample GRAV = 29.3 mg
Max. Wt. in
Assignment Possible Categories I Total Sample
CH, aromatic Substitute Benzenes 100 13.95
CH, aliphatic Halobenzenes 10 1.40
CH, aromatic Fused Aromatics 100 13.95
CH, aromatic
CH, aliphatic
CH, aliphatic
CH, aromatic
Multiplet
TABLE 3-13. IR REPORT—SAMPLE NO. 1, CUT LC-4
v, cm
-1
Total Sample GRAV = 0.0 mg
Assignment
Possible Categories
Max. Wt. in
Total Sample
Quantity Not Sufficient
135
-------
TABLE B-14. IR REPORT-SAMPLE NO. I, CUT LC-5
v, cm~
3150,3500
3034,3065
2859,2963
1732
1602
1495
1457
1276
1221
1028-1124
701,720
I
W
W
S
s
M
M
S
S
M
M
S
Total Sample GRAV = 14.6 mg
Assignment Possible Categories
OH
CH,
CH,
CO
NH,
CH,
CH,
CH,
C=0
NH,
C=0
CO,
C=0
CO,
CH,
or NH
aromatic/
olefinic
aliphatic
, ketone/
ester
amine;
aromatic
aromatic
aliphatic
, ester/ether
amine
, ester/ether
phenol
, ether;
alcohol
substitute
Hetero N Compounds
Hetero 0 Compounds
Hetero S Compounds
Alkyl S Compounds
Nitriles
Aldehydes, Ketones
Nitroaromatics
Ethers, epoxides
Alcohols
Phenols
Amines
Amines
Esters
I •
100
100
100
10
10
100
10
100
100
100
100
10
100
Max. Wt. in
Total Sample
1.55
1.55
1.55
0.16
0.16
1.55
0.16
1.55
1.55
1.55
1.55
0.16
1.55
136
-------
TABLE B-15. IR REPORT--SAMPLE NO. 1, CUT LC-6
Total Sample GRAV = 43.9 mg
v, cm
3336
3028,2065
2868-2967
1687
1601
1508
1459
1379
1275
1121
1028
696-812
I
M
M
S
S
S
S
M
VI
S
M
W
S
Assignment
OH or NH
CH, aromatic
CH, aliphatic
C=0, amide/car-
boxylic aci
OO, aromatic;
NH, amine
C=0, aromatic
CH, aliphatic
CH, aliphatic
Amide, carboxyli
acid
CH, aromatic
COH, alcohol
Multiplet
TABLE B-16. IR
Possible Categories
Phenols
Esters
Amines
Hetero N Compounds
d
Alky! S Compounds
Sulfonic Acids,
Sulf oxides
Carboxylic Acids
Alcohols
c Amines
REPORT— SAMPLE NO. 1, CUT
Max. Wt. in
I Total Sample
100
10
100
'100
10
10
100
100
100
LC-7
6.97
0.70
6.97
6.97
0.70
0.70
6.97
6.97
6.97
v, cm
-1
Total Sample GRAV = 0.0 mg
Max. Wt. in
Assignment Possible Categories I Total Sample
Quantity Not Sufficient
137
-------
TABLE B-17. MASS SPECTROSCOPY REPORT—SAMPLE NO. 1, CUT LC-1
XAD-2 EXTRACT
Total Sample GRAV = 11.0 mg
Quantity Not Sufficient
TABLE B-18. MASS SPECTROSCOPY REPORT—SAMPLE NO. 1, CUT LC-2
XAD-2 EXTRACT
Total Sample GRAV » 51.2 mg
Categories
Relative Intensity
Haloaliphatics
Substitute Benzenes
Halobenzenes
Fused Aromatics (MW <216)
Fused Aromatics (MW >216)
Possible Identifications
Naphthalene
Phenanthracene, Antharacene
Pyrene, Fl uoranthene
Chrysene, Benzanthracene
Benzof 1 uoranthene, Benzopyrene
Dibenzofluorene
Indenopyrene, Benzoperylene
Mol. Wt.
128
178
202
228
252
266
276
1
100
1
100
100
Relative Intensity
10
100
100
10
100
10
100
138
-------
TABLE B-19. MASS SPECTROSCOPY REPORT-SAMPLE NO. 1, CUT LC-3
XAD-2 EXTRACT
Total Sample GRAV = 29.3 mg
Categories Relative Intensity
Substitute Benzenes 1°
Halobenzenes 1
Fused Aromatics (MW <216) 10
Fused Aromatics (MW >216) TOO
Possible Identifications Mol. Wt. Relative Intensity
Naphthalene 128 10
Phenanthracene, Antharacene ^ 178 10
Pyrene, Fluoranthene 202 10
Chrysene, Benzanthracene 228 100
Benzofluoranthene, Benzopyrene 252 100
TABLE B-20. MASS SPECTROSCOPY REPORT-SAMPLE NO. 1, CUTS LC-4-7^
XAD-2 EXTRACT
Sample weight of LC-4 and 7 was Quantity Not Sufficient for analysis.
Mass spectra of LC fractions 5 and 6 were too complex for unequivocal
category identification. Assessment of LC-5 and 6 should be based on LC/IR
evaluation.
139
-------
TABLE B-21.
METAL CONTENT OF < 3 MICRON DUST, SAMPLE 1
—SHAKEOUT, GREEN SAND
Element
Li
Be
B
Na
Mg
Al
Si
P _
S
K
Ca
Ti
V
Cr
Mn
Fe
Co
Ni
Cu
Zn
Ca
Ge
As
Se
Br
Rb
Sr
Y
Zr
Nb
Mo
Cd
Sn
Sb
1
Cs
Ba
La
Ce
Pr
Nd
Sm
Dy
Pb
Th
U
Observed
jjg/m3
0.32
0.04
21.9
331
993
Major
12E4
21.2
364
271
655
36.4
0.93
73
31.1
1260
0.79
26.5
3.8
6.6
1.13
0.07
0.79
0.54
0.49
0.79
13.9
0.66
9.27
0.86
8.61
0.38
0.36
0.07
0.11
0.01
9\93
1.9
6.62
0.31
1.1
0.19
0.24
2.6
<0.66
0.12
Air
Health
MATE
^9/m3
22
2
3-10E3
2-53E3
6-10E3
5-10E3
1E4
1-10E2
1E3to4.4E5
2000
16E3
6000
500
1
5000
700-9000
50
15
200
4000
500
560
2
200
1E4
12E4
3100
1000
5000
22E3
5000
10
1E4
500
N'«l
82E3
500
11E4
37E3
51E3
N
53E3
9300
150
420
9.0
Observed
^g/g
4.8
0.61
330
5000
15E3
Major
18E4
320
5500
4100
9900
550
14
1100
470
19E3
1*
400
58
99
17
0.99
12
<8.2
<7.4
12
210
10
140
13
130
5.7
5.5
1.0
1.7
0.15
150
28
100
4.7
17
2.9
<3.6
40
10
1.8
Land
Ecology
MATE(=>
M9/9
75.
11.
5000
NIa)
17E3
200
N
(d)
N
4600
3200
160
30
50
20
50">'
50
2
10
20
N
N
10
5
N
N
N
N
N
N
1400
0.2
N
40
N
N
500
N
N
N
N
N
N
10
N
100
MEG
Category
27
32
37
28
33
33
43
48
53
29
34
62
65
68
71
72
74
76
78
81
39
44
49
54
58
30 -
35
61
63
66
69
82
45
50
59
31
36
84
84
84
84
84
84
46
85
85
'a'N means not determined or not set in the case of MATE values.
'"'The land MATE values are incompletely developed and subject to modification. No MATE value
has been set for hydrated ferric oxide, the most probable equilibrium form of iron in the environ-
ment.
("The land ecology values listed in EPA 600/7-77-136a have been multiplied by 100 to correspond
with new recommendations in development.
'"'MATE for elemental P is 0.1 pg/g but this is unsettled as the occurance of elemental phosphorous
in the environment will be transitory at best. Phosphate, PC>4~3, is listed as "N" or not determined.
140
-------
TABLE B-22. LC ANALYSIS REPORT, SAMPLE 2
ontraetor Research Triangle Institute
ample Site Stack 5 Sample Acquisition Date 29 June 1978
ype of sour™ Shakeout, Green sand, post scrubber, line 5
2St Number Sample ID Number 6293-G (XR)
Sorbent Extract, stack, post scrubber
•iginal Sample Volume or Mass 26.15 m Std., dry
isponsible Analyst. Date Analyzed
Iculations and Report Reviewed By
Report Date
lumn Flow Rate.
. Column Temperature.
sen/aliens
Total Sample'
Taken for LC*
Recovered^ ••-
TCO
mg
2490.0
74.8
81.0
GRAV
mg
265.0
7.9
7.2
Total
mg
2755.0
82.7
88.2
Concentration^:
mg/m^ j
105.4
3.2
3.4
'1
raction :
i
1
2
3
4
5
6
7
Sum '
i
TCO^inmg
- Total
485.7
1512.4
224.4
6.2
24.6
236.7
0
Blank
0
0
0
0
0
0
0
Corrected :
485.7
1512.4
1 224.4
6.2
24.6
236.7
.- -o
2490.0
GRAV4inmg
Total
0
169.3
29.4
29.4
7.4
103.1
122.1
Blank
0
14.7
22.1
14.7
0
22.1
22.1
Corrected
0
154.6
7.4
14.7
7.4
81.0
0
265.1
Total4 I
mg !
485.7
1667.0
231.8
20.9
32.0
317.7
0
2755.1
Concentration^
mg/ni3
18.6
63.8
8.9
0.80
. 1.2
12.2
0
105.5
1. Quantity in entire sample, determined before LC
2. Portion of whole sample used for LC, actual mg
3. Quantity recovered from LC column, actual mg
4. Total mg computed back to total sample
5. Total mg divided by total volume
141
-------
TABLE B-23. ORGANIC EXTRACT SUMMARY
Sample 2- Stack, Post Scrubber, Line 5
Total Organic;, mg/m
TCO.rng 90.38%
GRAV, ma 9.62%
LCI •
18.6
486
0.0
LC2
63.8
1512
154.6
LC3
8.9
224
7.4
LC4
0.8
6.1
14.7
LC5
1.2
25
7.4
LCG
12.2
237
81.0
LC7
0
' •-. . 0
0 '
2
105.3
2490
265
GRAV, mg/m3 . _ 0 5.92 0.28 0.56 0.28 3.09 0 10.13
Category
Int/mg/m^
Aliphatics
Haloaliphatics
Substituted Benzenes
Halo-enzenes '
Fused Aromatics
Hetero N Compounds
Hetero 0 Compounds
Hetero S Compounds
Alkyl S Compounds
Nitriles
Aldehydes, Ketones
Ni troaromatics
Ethers, Epoxides
-
10/0.2
•100/2.69
10/0.2"
100/2.69
.
100/0.1:
10/0.02
100/0.13
10/0.08
10/0.08
10/0.. 08
10/0.08
10/0.08
lo/n.na
in/n.np
100/0.04
100/0.04
100/0.04
10/0.004
10/0.00'
100/0.0'
10/0. nrv
inn/n ru
100/0.49
10/0.05
* f
0.0
0.27
2.82
0.29
2.82
0.61
0.12
0.12
0.05
0.08
n.nR
n -rip
n ns
I\D
-------
TABLE B-23. (cont'd)
Total Crganics, mg/m
TCO, mg
GRAV. mg
«5,«ri. 2- Stack, Post Scrubber, Line 5 •
LC1 '
•-
LC2
,i
LC3
LC4
•
LC5
LCC
LC7
'
2
oo
Category Int/mg/m
Alcohols
.
Phenols
Amines
Amides
Esters
Carboxylic Acids
Sulfonic Acids
>.
,
10/0.004
10/0.004
10/0.004
10/0.004
100/0.04
•
100/0.49
100/0.49
100/0.49
100/0.49
10/0.05
100/0.49
10/0.05
1 *
0.49
0.49
0.49
0.49
0.09
0.49
0.05
-------
TABLE B-24. IR REPORT—SAMPLE NO. 2, CUT LC-1
v, cm"
I
Sorbent Extract, Stack, Post Scrubber, Line 5
Total Sample 6RAV = 0.0 mg
. . Max. Wt. in
Assignment Possible Categories I Total Sample
Quantity Not Sufficient
TABLE B-25. IR REPORT— SAMPLE NO. 2, CUT LC-2
v, cnf
3024,3053
2871-2967
1606
1493
1379
1033
699-800
I
S
S
M
W
M
M
S
Total Sample GRAY * 154.6 mg
. . Max. Wt. in
Assignment Possible Categories I Total Sample
CH, aromatic/ Haloaliphatics 10 7.0 mg
olefinic
CH, aliphatic Substituted Benzenes 100 70.3
C=-0, aromatic Halobenzenes 10 7.0
C=0, aromatic Fused Aromatics 100 70.3
CH, aliphatic
CH, aromatic
Multiplet
TABLE B-26. IR REPORT-SAMPLE NO. 2, CUT LC-3
v, cm
3024
2871-2967
1603
1497
1456
1380
1034
699-881
I
S
S
S
M
S
W
W
S
i Total Sample GRAV = 7.4 mg
Max. Wt. in
Assignment Possible Categories I Total Sample
CH, aromatic Substituted Benzenes 100 3.5
CH, aliphatic Halobenzenes 10 0.4
C=C, aromatic Fused Aromatics 100 3.5
CzC, aromatic
CH, aliphatic
CH, aliphatic
CH, aromatic
Multiplet
144
-------
TABLE B-27. IR REPORT—SAMPLE NO. 2, CUT LC-f
Total Sample GRAV « 14.7 mg
v, cm
Quantity
I
Not
Assignment
Sufficient
Possible Categories
Hetero N Compounds
Hetero 0 Compounds
Hetero S Compounds
Nitriles
Ethers, Epoxides
Aldehydes, Ketones
Nitroaromatics
TAB! F B-28. IR REPORT- -SAMPLE NO. 2, CUT
v, cm
3034
2857-2963
1721
1603
1498
1456
1274
1221
911-1121
668-750
I
W
S
S
M
M
S
S
M
M
M
Total Sampl
Assignment
CH, aromatic
CH, aliphatic
C=0, ester/ketone
C=0, aromatic
C=0, aromatic
CH, aliphatic
COC, ester/ether
COC, ester/ether
CH, aromatic
Multiplet
e GRAV - 7.4 mg
Possible Categories
Hetero N Compounds
Hetero 0 Compounds
Hetero S Compounds
Alkyl S Compounds
Nitriles
Aldehydes, Ketones
Nitroaromatics
Ethers, Epoxides
Alcohols
Phenols
Ami des
Ami nes
Esters
Max. Wt. in
I Total Sample
10 2.1
10 2.1
10 2.1
10 2.1
10 2.1
10 2.1
10 2.1
LC-5
Max. Wt. in
I Total Sample
100 1.1
100 1.1
100 1.1
10 1.1
10 0.1
100 1.1
10 0.1
100 1.1
10 C.I
10 0.1
10 0.1
10 0.1
100 1.1
145
-------
TABLE B-29. IR REPORT-SAMPLE NO. 2, CUT LC-6
v, cm
3319
3035,3070
2857-2952
1687
1604
1509
1450
1373
1273
1119
670-818.
v, cm
I
M
W
s
s
s
s
M
M
S
W
M
I
Total
Assignment
NH or OH
CH, aromatic
CH, aliphatic
C*0, amide/car-
boxylic aci
C=0, aromatic;
NH, amine
C^O, aromatic
CH, aliphatic
CH, aliphatic
Amide/carboxylic
acid
COH, alcohol ;
CH, aromatic
Multiplet
TABLE B-30. IR
Total
Assignment
Sample GRAV * 81.2 mg
Possible Categories
Phenols
Esters
Amines
Hetero N Compounds
d
Alkyl S Compounds
Sulfonic Acids,
Sulf oxides
Carboxylic Acids
Alcohols
Amides
REPORT— SAMPLE NO. 2, CUT
Sample GRAV = 0.0 mg
Possible Categories
I
100
10
100
100
10
10
100
100
100
LC-7
I
Max. Wt. in
Total Sample-
12.9
1.3
12.9
12.9
1.3
1.3
12.9
12.9
12.9
Max. Wt. in
Total Sample
Quantity Not Sufficient
146
-------
TABLE B-31. MASS SPECTRQSCOPY REPORT—SAMPLE NO. 2, CUT LC-1
Total. Sample GRAV = 0.0 mg
Weight of Sample was Quantity Not Suitable for Analysis
TABLE B-32. MASS SPECTROSCOPY REPORT-SAMPLE NO. 2, CUT LC-2
Total Sample GRAV =154.6
Categories Relative Intensity
Haloaliphatics
Substitute Benzenes
Halobenzenes
Fused Aromatics (MW <216)
Fused Aromatics (MW >216)
Possible Identifications
Phenanthracene, Antharacene
Pyrene, Fluoranthene
Chrysene, Benzanthracene
Benzofluoranthene, Benzopyrene
Indenopyrene, Benzoperylene
Mol. Wt.
178
202
228
252
276
1
10
1
100
100
Relative Intensity
10
100
10
100 -
100
TABLE B-33. MASS SPECTROSCOPY REPORT-SAMPLE NO. 2, CUT LC-3
Total Sample GRAV - 7.4 mg
Categories ' Relative Intensity
Substitute Benzenes
Halobenzenes
Fused Aromatics (MW <216)
Fused Aromatics (MW >216)
Possible Identifications
Naphthalene
Phenanthracene, Anthracene
Benzofluoranthene, Benzopyrene
Dibenzofluorene
Mol. Wt.
128
166
252
266
10
1
10
100
Relative Intensity
10
10
100
100
147
-------
TABLE B-34... MASS SPECTROSCOP.Y REPORT—SAMPLE NO. 2, CUTS LC 4-7
Sample weight of LC-4 and 7 was Quantity Not Sufficient for analysis,
Mass spectra of LC fractions 5 and 6 were too complex for unequivocal
category identification. Assessment of LC-5 and 6 should be based on LC/IR
evaluation.
148
-------
TABLE B-35. LC ANALYSIS REPORT, SAMPLE 3
Contractor Research Triangle Institute
;ampie site Shakeout room
Sample Acquisition Date 30 June 1978
'ype of source Shakeout, phenolic shell molding
est Number.
Sample ID Number 6304-G (XR)
ample Description Sorbent extract, Sample 3 Shakeout, phenolic, Line 1
3
riyinal Sample Volume or Mass 1 ? , 47 FTI Stfl . ,
esponsible Analyst _ -
initiations and Report Reviewed By.
Date Analyzed
. Report Date
)!umn Flow Rate.
. Column Temperature.
iservations
I*.
Total Sample^
Taken for LC2
Recovered^
TCO
mg
210.0
42.0
41.4
GRAV
my
160.0
32.0
35.5
Total
mg
370.0
74.0
76.9
Concentration^;
mg/m^ !
29.7
5.9
6.2
i
raetton !
1
2
3
4
5
6
7
Sum '
TCO^inmg
- Total
32.0
59.3
12.7
17.2
23.3
65.4
0
Blank
0
0
0
0
0
0
0
i Corrected •
i
32.0
59.3
1 12.7
17.2
23.3
65.4
' 'Q
210
GRAV4inmg
Total
29.3
55.0
16.2
15.3
7.2
42.4
6.3
Blank
0
1.8
2.7
1.8
0
2.7
2.7
Corrected •'
29.3
53.2
13.5
13.5
7.2
39.7
3.6
160
Total* i
mg I
i
61.3
112.5
26.2
30.8
30.5
105.1
3.6
370
Concentration^
mt|/m3
4.9
9.0
2.1
2.5
2.4
8.4
0.3
29.7
1. Quantity in entire sample, determined before LC
2. Portion of whole sample used for LC, actual mg
3. Quantity recovered from LC column, actual mg
4. Total mg computed back to total sample
5. Total mg divided by total volume
749
-------
TABLE B-36. ORGANIC EXTRACT SUMMARY
Sample 3. Shakeout. Fugitive. Phenolic
TotalOrganics, mg/m^
TCO.mg . (56.76%)
GRAV.ma (43.24%)
LC1 •
4.9
32.0
29.9
. LC2
9.0
59.3
53.2
LC3
:2.1
12.7
13.5
LC4
' 2.5
17.2
13.5
LC5
2.5
23.3
7.2
LCG
8.4
65.4
39.7
LC7
0.3
0
3.6 '
2
29.7.
210
160
GRAV, mg/m3 , _ 2.35 4.27 1.08 1.08 0.58 3.18 0.29 12.83
Category
Int/mg/m3
Aliphatics
Haloaliphatics
Substituted Benzenes
Halobenzenes
Fused Aromatics
Hetero N Compounds
Hetero 0 Compounds
Hetero S Compounds
•Alkyl S Compounds
Nitriles
Aldehydes, Ketones
Nitroaromatics
ion/?. 14
10/0.21
*
10/0.19
100/1.94
10/0.19
100/1.94
TOO/0.52
10/0.05
100/0.52
100/0.21
100/0.21
100/0..21
10/0.02
100/0.21
10/0.02
100/0.06
100/0.06
100/0.06
10/0.01
10/0.01
100.0.06
10/0.01
100/0.44
10/0.04
100/0.44
g ^
100/0.04
10/0.00
? 14
0.40
2.46
0.24
2.46
0.75
0.27
0.27
0.05
0.47
0.27
0:03
tn
o
-------
TABLE B-36. (cont'd)
3. Shakeout, Fugitive, Pehnolic
TotalOrganics, mg/m
TGO. rng
GRAV, mg
Cample ,
LCT
LC2
.if
LC3
LC4
•
LC5
LCG
•
LC7
'
2
•r
Category Int/mn/m
Ethers, Epoxides
Alcohols
Phenols
Amines
Amides
Esters
Carboxylic Acids
Sulfonic Acid's, Sul foxides
*
,
100/0.21
100/0.06
100/0.06
100/0.06
100/0.06
10/0.01
100/0.06
100/0.44
10/0.04
100/0.44
100/0.44
100/0.44
100/0.44
10/0.04
• t
100/0.04
100/0.04
100/0.04
100/0.04
100/0.04
100/0.04
10/0.00
0.27
0.54
0.14
•0.54
0.49
0.54
0.48
0.04
-------
v, cm"
2857-2959
1464
1378
720-971
I
S
S
M
W
*Since there i
is assigned to
Total
Assignment
CH, aliphatic
CH, aliphatic
CH, aliphatic
s evidence of only
that category.
Sample GRAY = 29.3 mg
Possible Categories
Aliphatics
Haloaliphatics
one compound category, the
TABLE B-38. IR REPORT-SAMPLE NO. 3, CUT
v, cm"
3031-3052
2870-2971
1602
1458
1378
698-800
I
M
S
M
S
M
S
Total
Assignment
- CH, aromatic
CH, aliphatic
C*C, aromatic
C=C, aromatic
CH, aliphatic
Multiplet
Sample GRAV = 53.2 mg
Possible Categories
Haloaliphatics
Substitute Benzenes
Halobenzenes
Fused Aromatics
I
100
10
total
LC-2
I
10
100
10
100
Max. Wt. i
Total Samp!
26.6 mg*
2.7
GRAV weight
Max. Wt. i
Total Samp!
2.6 mg
24.2
2.4
24.2
n
e
n
e
152
-------
TABLE B-39. IR REPORT--SAMPLE NO. 3, CUT LC-3
v, cm
3030,3056
2857-2962
1740
1604
1494
1457
1378
702-880
v, cm
3000-3100
2861-2955
1731
1602
1466
1378
1272
1072,1125
713,754
I
M
S
M
M
W
S
M
S
I
W
S
M
M
S
W
M
W
M
Total
Assignment
CH, aromatic
CH, aliphatic
OO, ketone
C=0, aromatic
C=C, aromatic
CH, aliphatic
CH, aliphatic
Multiplet
TABLE B-40. IR
Total
Assignment
CH, aromatic
CH, aliphatic
Ketone
C=C, aromatic
CH, aliphatic
CH, aliphatic
COC, ether
CH, aromatic;
COC, ether
CH, substituted
Sample GRAV = 13.5 mg
Possible Categories
Substituted Benzenes
Halobenzenes
'Fused aromatics
REPORT— SAMPLE NO. 3, CUT
Sample GRAV = 13.5 mg
Possible Categories
Hetero N Compounds
Hetero 0 Compounds
Hetero S Compounds
Nitriles
Ether, Epoxides
Aldehydes, Ketones
Nitroaromatics
I
100
10
100
LC-4
I
100
100
100
10
100
100
10
Max. Wt. in
Total Sample
6.4 mg
0.6
6.4
Max. Wt. in
Total Sample
2.6 mg
2.6
2.6
0.3
2.6
2.6
0.3
153
-------
TABLE. B-41. IR REPORT—SAMPLE NO. 3, CUT LC-5
_1
v, cm
3444
3038,3057
2855-2961
1725
1602
1496
1454
1378
1278
1219
1001-:n25
701 ,748
I
W
W
S
s
M
W
S
W
s
M
M
M
Total Samp!
Assignment
NH or OH
CH, aromatic
CH, aliphatic
C=0, ketone, ester
C=C, aromatic
C*C, aromatic
CH, aliphatic
CH, aliphatic
COC, ester/ether
COH, phenol;
COC, ester
COH, alcohol;
COC, ether
CH, substituted
e GRAV = 7.2 mg
Possible Categories
Hetero N Compounds
Hetero 0 Compounds
Hetero S Compounds
Alky! S Compounds
Nitriles
Aldehydes, Ke tones
Nitroaromatics
Ethers, Epoxides
Alcohols
Phenols
Amines
Amides
Esters
TABLE B-42. IR REPORT— SAMPLE NO. 3,
_1
v, cm
3200-3358
3067
2857-2960
2227
1722
1659
1608
1503
1457
1381
1273
1115
718-825
I
S
W
S
M
S
M
S
M
S
M
S
M
M
Total Samp!
Assignment
NH or OH
CH, aromatic
CH, aliphatic
C = N, nitrile
Carboxylic acid,
ester
C=0, amide
NH, amide;
carboxylic acid
C=C, aromatic
CH, aliphatic
CH, aliphatic
Carboxylic acid,
amide;
CN, amine
OH, alcohol;
CH, aromatic
Multiplet
e GRAV = 39.7 mg
Possible Categories
Phenols
Esters
Amines
Hetero N Compounds
Alky! S Compounds
Sulfonic Acids,
Sulfoxides
Carboxylic Acids
Alcohols
Amides
Nitriles
I
100
100
100
10
10
100
10
100
100
100
100
10
100
CUT
I
10
100
100
100
10
10
TOO
100
100
100
Max.
Total
0.8
0.8
0.8
0.1
0.1
0.8
0.1
0.8
0.8
0.8
0.8
0.1
0.8
LC-6
Max.
Total
0.5
5.4
5.4
5.4
0.5
0.5
5.4
5. '4
5.4
5.4
Wt. in
Sample
mg
Wt. in
Sample
mg
154
-------
TABLE B-43. IR REPORT— SAMPLE NO. 3, CUT
LC-7
Total Sample 6RAV = 3.6 mg
_]
v, cm
3000-3400
2860-2948
1704
1657
1605
1458
1399
1376
1258
1112
666,719
I
M
S
S
S
S
S
M
M
M
S
M
TABLE
Assignment
NH or OH
CH, aliphatic
Carboxylic acid,
ester
C=0, amide
NH, amide;
carboxylate
CH, aliphatic
Sulf oxides
Ami de
CH, aliphatic
CO, ester;
OH, phenol ;
CH, amine
OH, alcohol
CH, substituted
Possible Categories
Phenols
Esters
Amines
Hetero N Compounds
Alkyl S Compounds
Sulfonic acids,
Sulfoxides
Carboxylic Acids
Alcohols
Ami des
B-44. MASS SPECTROSCOPY REPORT— SAMPLE NO.
I
100
100
100
100
10
10
100
100
100
3, CUT
Max. Wt. in
Total Sample
0.5 mg
0.5
0.5
0.5
0.1
0.1
0.5
0.5
0.5
LC-1
Total Sample GRAV = 29.3 mg
Categories
Relative Intensity
Aliphatics
Haloaliphatics
100
1
155
-------
TABLE B-45, MASS SPECTROSCQPY REPORT—SAMPLE NO. 3, CUT LC-2
Total Sample GRAY * 53.2 mg
Categories Relative Intensity
Haloaliphatics 1
Substituted Benzenes 10
Halobenzenes 1
Fused Aromatics (MW < 216) 100
Fused Aromatics (MW > 216) 100
Possible Identifications Mol. Wt. Relative Intensity
Npahthalene
Acenaphthylene
Phenanthracene, Anthracene
Fl uoranthene, Pyrene
Benzanthracene, Chrysene
Benzof 1 uoranthene, Benzopyrene
128
152
178
202
228
252
10
10
100
100
100
10
TABLE B-46. MASS SPECTROSCOPY REPORT-SAMPLE NO. 3, .CUT LC-3
Total Sample GRAY - 13.5 mg
Categories Relative Intensity
Substituted Benzenes 10
Halobenzenes 1
Fused Aromatics (MW < 216) 10
Fused Aromatics (MW > 216) 100
Possible Identifications Mol. Wt. Relative Intensity
Naphthalene 128 10
Phenanthracene, Antrhacene 178 10
Fluoranthene, Pyrene 202 10
Benzanthracene, Chrysene 228 100
•Benzof1uoranthene, Benzopyrene 252 10
156
-------
TABLE B-47. MASS SPECTROSCQPY REPORT-SAMPLE NO. 3, CUTS LC-4-7
Mass spectra of LC fractions 4-6 were too complex for unequivocal category
identification. Assessment of LC fractions 4-7 should be based on LC/IR
evaluation.
157
-------
TECHNICAL REPORT DATA
(flease read Instructions on the reverse before completing)
1. REPORT NO. 2.
EPA-600/2-80-021
4. TITLE AND SUBTITLE '
Environmental Assessment of Iron Casting
7. AUTHOR(S) '
V. H. Baldwin, Jr.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, North Carolina 27709
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
January 1980
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPOR"
10. PROGRAM ELEMENT NO.
1AB604C and 1BB610C
11. CONTRACT/GRANT NO.
68-02-2630, Task 2
13. TYPE OF REPORT AND PERIOD COVE
Task Final: 3/77-3/79
14. SPONSORING AGENCY CODE
EPA/600/13
91^41-2^3" N°TES ffiRL-RTP Pro'ect officer is R°bert V. Hendriks, Mail Drop i
_ _ _ ___ f*r^ciT"iTic
Sampling of ductile iron casting in green sand molds with phenolic isocyanate cor
and in phenol-formaldehyde bound shell molds did not provide definitive proof tha
environmentally hazardous organic emissions occur. Both molds produced the sa
types of major emissions: alkyl halides, carboxylic acid derivatives, amines, su
stituted benzenes , nitrogen heterocyclics, and fused aromatics in quantities that
slightly exceed the lowest Minimum Acute Toxocity Effluent (MATE) values for-ft
categories, but probably not for individual compounds. GC-MS analysis revealed
that the major fused aromatics were naphthalene compounds. Quantitative analys-'
of specific PNA's showed no significant level of concern. Inorganic dust emissior
are hazardous if uncontrolled because of Si, Cr, and Ni. The dust is sufficiently
in 12 metals to render it a hazardous waste if collected as a sludge and landfilled
)ut leachate testing may change that categorization. Relatively high levels of Sr,
10 " Pr, and Nd in the dust indicate that inoculation smoke should be examined.
17- KEY WORDS AND DOCUMENT ANALYSIS
3- DESCRIPTORS
Pollution Leaching
Assessments
Iron Castings
Dust
Sludge
Earth Fills
13. DISTRIBUTION STATEMENT ""* "~ *
Release to Public .
b.lDENTIFIERS/OPEN ENDED TERMS
Pollution Control
Stationary Sources
Particulate
19. SECURITY CLASS (This Report)
Unclassified
20. SECURITY CLASS (This page)
Unclassified
c. COSATl Field/Gr
13B 07;
14B
11F
11G
07A
13C
21. NO. OF PAGES
168
22. PRICE
EPA Form 2220-1 (9-73)
58
------- |