EMISSIONS FROM HOT-DIP GALVANIZING PROCESSES
FINAL REPORT
-------
EMISSIONS FROM HOT-DIP GALVANIZING PROCESSES
FINAL REPORT
EPA - 905/4-76-002
by
Peter J. Drlvas
Pacific Environmental Services, Inc.
1930 14th Street
Santa Monica, California 90404
213-393-9449
Contract No. 68-01-3156
Task Order No. 7
Project Officer: Will 1am E, Beyer
U.S. Environmental Protection Agency
Region V
230 South Dearborn Street
Chicago, Illinois 60604
Warcht 1976
-------
This report has been reviewed by the Environmental Protection
Agency, Region V, and approved for publication. Approval does
not signify that the contents necessarily reflect the views
and policies of the Environmental Protection Agency, nor does
mention of trade names or commercial products constitute en-
dorsement or recommendation for use.
In keeping with the stated policy of the EPA, the metric
system 1s used almost exclusively 1n this report. However,
equivalent English units are presented, in addition to metric
units, for the emission factors derived 1n this work. Also,
the following conversion factors are presented for convenience
tn making other conversions:
TO Convert From To Multiply by
kg
metric ton
kg/metric ton
meters
lb
2.205
short ton
1.102
lb/short ton
2
feet
3.281
sq. meters
sq. feet
10.764
lb/ft2
0.205
-------
TABLE OF CONTENTS
Page
TABLE OF CONTENTS 1ii
LIST OF FIGURES . v
LIST OF TABLES vi
1. INTRODUCTION 1
1.1 Project Background 1
1.2 The Hot-Dip Galvanizing Process 2
1-.3 Galvanizing Emissions 2
2. SUMMARY, CONCLUSIONS AND RECOMMENDATIONS 5
3. LITERATURE REVIEW . . . , 7
3.1 Extent of Review ...... 7
3.2 General Process and Emission Control Information ... 8
3.3 Specific Particulate Emission Data 10
4. OBSERVATION OF CURRENT GALVANIZING PRACTICES. ... 12
4.1 Los Angeles Area 12
4.2 Chicago Area 14
5. PARTICULATE EMISSION SOURCE TEST DATA 17
5.1 Existing Source Test Data ^
5.2 Source Tests Conducted by Pacific
Environmental Services, Inc 18
6. DATA ANALYSIS 22
6.1 Controlled vs. Uncontrolled Emissions 22
6.2 Emission Factor Based on Process Weight 23
6.3 Emission Factor Based on Hours of Operation 26
6.4 Emission Factor Based on Surface Area 30
6.5 Total Galvanizing Emissions in the U.S 31
APPENDICES
A-l DESCRIPTION OF ZACLON GALVANIZING FLUXES A-l
A-2 DETAILS OF PARTICULATE EMISSION SOURCE TESTS
CONDUCTED BY THE PES STAFF A-5
111
-------
TABLE OF CONTENTS (continued)
Page
A-3 LINEAR LEAST SQUARES REGRESSION ANALYSIS A-15
A-4 ADDRESSES OF TESTED GALVANIZING PLANTS ' A"18
LIST OF REFERENCES . . . A-20
1v
-------
LIST OF FIGURES
Page
6-1 PARTICULATE EMISSIONS VS. GALVANIZED PRODUCT
PROCESS WEIGHT 28
v
-------
LIST OF TABLES
Page
1-1. 1973 U.S. ZINC CONSUMPTION BY GALVANIZING 3
5-1. SUMMARY OF GALVANIZING SOURCE TESTS 20
5-2. RESULTS OF PES SOURCE TESTS 21
6-1. SOURCE TESTS WHICH ESTIMATED AMOUNT OF
ZINC ADOED 24
6-2. EMISSION FACTORS BASED ON PROCESS WEIGHT ...... 27
6-3. EMISSION FACTORS BASED ON -SURFACE AREA 30
vi
-------
1. INTRODUCTION
1.1 Project Background
A number of stat-e and local agencies, concerned with visible
emissions from galvanizing plants, have asked the Environmental
Protection Agency for more information on the nature and amounts of
galvanizing emissions. A limited survey by the EPA revealed that
very little had been published in respect to galvanizing operations
or evaluation of the particulate emissions from galvanizing plants.
Thus, although a need for control had been indicated by soifie agencies,
the enforcement position was weakened by a lack of information.
In response to a request from the EPA, Region V, Pacific
Environmental Services, Inc. (PES) conducted a thorough investigation
of the hot-dip galvanizing process to determine the quantity and
characteristics of particulate emissions from the process. The main
purpose was to develop an emission factor as a function of process
rates, procedures, or equipment. The study focused on the actual
galvanizing step, and was not concerned with the cleaning operations
preparatory to galvanizing.
The scope of effort was as follows:
(1) Perform a complete literature search to locate
and Identify all sources of information.
(2) Observe current galvanizing practices, procedures,
and equipment through plant visits.
(3) Collect all data from actual source tests which measured
emission rates from hot-dip galvanizing operations.
(4) . Perform source testing of current galvanizing
operations, using EPA Method 5, to determine
accurate emissions of particulates.
(5) Analyze and review all available literature and test
data to develop an accurate emission factor.
-1-
-------
1.2 The Hot-Dip Galvanizing Process
Hot-dip galvanizing is the art of coating clean, oxide-free iron
or steel with a thin layer of zinc by immersion of the iron or steel in
a bath of molten zinc. The article to be coated is properly cleaned, com-
pletely immersed in the molten zinc bath, and withdrawn with enough of
the molten zinc adhering as a surface film to give the desired coating.
The molten zinc bath is usually operated at temperatures between 445° and
460° C. (1).
In 1973, a total of 511,505 metric tons of zinc were consumed by
galvanizing in the United States; this amount is 37.5% of the total
U.S. zinc consumption in 1973 (2). The zinc consumption by galvanized
product category is given in Table 1-1. Sheet and strip iron and steel
accounted for over half of the total zinc used in galvanizing.
Typically, the basic steps which are followed in cleaning and
coating an iron or steel article are the following:
v
(1) Degreasing in a hot, alkaline solution;
(2) Rinsing thoroughly in a water rinse;
(3) Pickling in a hot, acid bath;
14) Rinsing thoroughly in a water rinse;
(5) Prefluxlng 1n a zinc ammonium chloride solution;
(6) Immersing the article 1n the molten zinc through a
molten flux cover (usually zinc ammonium chloride);
(7) Finishing (dusting with ammonium chloride to produce
a smooth finish).
Some steps may be omitted- in individual galvanizing processes; such
differences are noted 1n Section 4, "Observation of Current Galvanizing
Practices."
1.3 Galvanizing Emissions
The main emission problem 1n the galvanizing process 1s caused by the
last two steps, (6) and (7) above. It has been observed that grayish-white
-2-
-------
Table 1-1
1973 U.S. ZINC CONSUMPTION BY GALVANIZING (2)
Galvanized Product Category
Zinc Consumed,Metric
Ton?;
Percent of Total
Sheet and strip
292,047
57.1
Tubes and pipe
61,732
12.1
Wire and wire rope
31,130
6.1
Fencing, cloth, and netting
23,059
4.5
Structural shapes
19,699
3.9
Fittings (for tubes and pipe)
10,858
2.1
Pole-line hardware
7,433
1.5
Fasteners
4,338
0.8
Tanks and containers
2,668
0.5
Other and unspecified uses
58,541
11.4
TOTAL
511,505
100.0
3-
-------
particulate fumes are discharged whenever the kettle flux cover is
disturbed, when fresh flux is added, and when galvanized articles are
dusted with ammonium chloride (1).
Flux agitation occurs to some extent each time an object is intnersed
in the molten zinc through the flux cover. If the objects are smooth and
dry, the agitation is not great and the amount of fuming is low. When
the agitation of the flux cover is severe, a correspondingly larger
amount of fumes is discharged.
When fresh flux is placed on a kettle, it takes some time to
form a foaming cover. During this time dense fumes escape. Also, when
fresh flux is stirred into the existing flux cover, fumes are discharged
due to both the agitation and the time necessary for the fresh flux to
be absorbed by and become part of the foam.
To obtain brighter, smoother finishes, especially on small items,
they are dusted with finely.ground ammonium chloride immediately after
being removed from the molten zinc bath. The articles dusted are still
at a temperature well above the decomposition temperature of ammonium
chloride, thus, much of the NH^ CI is converted to fumes by the operation.
Although only small amounts of dusting fluxes are used, dense fumes
are normally created.
-4-
-------
2. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS
A comprehensive literature search was performed to locate and
Identify all sources-of information on emissions from hot-dip galvanizing
processes. Observations of current galvanizing practices, procedures,
and equipment were made in a total of five plant visits - two in
the Los Angeles area and three in the Chicago area. To determine
accurate emissions of particulates from current galvanizing operations,
the PES staff performed three separate source tests, using EPA Method
5, on one galvanizing plant in the Los Angeles area.
Particulate emission data from a total of seventeen source tests on
hot-dip galvanizing plants were considered relevant in developing an
emission factor for galvanizing kettles. Fourteen of the source tests
measured both kettle emissions and emissions from control devices; the
control devices tested were mainly water scrubbers and baghouses. There
was no significant difference between the kettle and control device
emissions.
An averaged emission factor based on galvanized product process
weight was calculated, considering both kettle and control device
emissions as one data base:
Emission factor ¦ 0.26 kg/metric ton galvanized product (
-------
A linear regression analysis indicated that a slightly more
accurate emission factor is one based on the hours of kettle operation.
Considering both kettle and control device emissions as one data base,
an emission factor which is independent of process weight was cal-
culated:
Emission factor = 0.51 kg/hr per kettle ( ° - 0.44)
(1.12 lb/hr per kettle)
The rather large standard deviation (cr) indicates the considerable
scatter in the data base, and this approach is only slightly more
accurate than the emission factor based on process weight.
From the source tests conducted by the PES staff, a relationship
can be derived between galvanizing emissions and the surface area of
the galvanized product. However, this data base is too limited for
confident characterization of an emission factor.
Due to the nature of the galvanizing process, Intermittent visible
emissions are released when the kettle flux cover 1s disturbed and when
galvanized articles are dusted with ammonium chloride. These Intermittent
emissions, when averaged over a one-hour period, result in fairly low
uncontrolled emission rates. Using the derived emission factors, the
total annual amount of particulate emissions produced by hot-dip gal-
vanizing operations In the United States 1s estimated to be about 1,600
metric tons (1,800 short tons) which is negligible when compared with
the total particulate emissions from other sources.
Since available control devices have not been shown to be
particularly effective 1n reducing galvanizing emissions, 1t 1s
recommended that control devices for galvanizing operations be required
only 1n the case of severe visible emissions. A reduction 1n the
amount of ammonium chloride used for finishing galvanized articles
will significantly reduce visible emissions from galvanizing kettles.
-6-
-------
3. LITERATURE REVIEW
3/1 Extent of Review
The literature review, which was concerned mainly with information
on galvanizing emissions, encompassed a wide variety of sources. Books
and journal articles were obtained from a number of library systems,
including:
• University of California, Los Angeles (UCLA) Libraries
• University of Southern California (USC) Libraries
• California Institute of Technology (Caltech) Libraries
• Los Angeles Public Library.
A literature search was requested from the Air Pollution Technical
Information Center (APTIC) on zinc and metal fabrication, which resulted
in a total of 44 references. However, only five of these references
pertained specifically to emissions from hot-^ip galvanizing operations.
Other abstracting services which were checked year-by-year Included:
• Air Pollution Control Association Abstracts, 1952-1970
• A1r Pollution Abstracts, 1970-1S75
• Metal Finishing Abstracts, 1565-1975
• Z1nc Abstracts, 1965-1975.
Contact with Industrial organizations was quite rewarding. The
American Hot-Dip Galvanizers Association (AHDGA) provided a number of
unpublished preprints concerned with galvanizing emissions, in add'itlon
to a directory of AHDGA members in the United States. The DuPont Industrial
Chemicals Department provided a data sheet on the various types of kettle
fluxes and prefluxes which they manufacture (see Appendix A-l).
The Los Angeles County Air Pollution Control District (APCD) provided
by far the. most relevant data for determining an emission factor - the
-7-
-------
results of 21 source and chemical tests on galvanizing kettles conducted
over a 25-year period. Not all tests were relevant and some information
which was considered confidential was not provided; however, these data
provided by the Los Angeles County APCD were the largest single source
of particulate emission data from galvanizing kettles.
The San Francisco Bay Area Air Pollution Control District provided
the results of one particulate emission source test on a galvanizing
kettle. A computer search of the National Emission Data System (NEDS)
for all source tests on galvanizing kettles produced only one test
result; this test was performed on a galvanizing furnace, which is not
a hot-dip operation. Neither the Los Angeles County tests nor the
San Francisco test were listed in the NEDS computer search.
3.2 General Process and Emission Control Information
There are a number of books which describe the hot-dip galvanizing
process in some detail (3, 4, 5). However, these sources do not even
discuss the emission problem. The most recent (1974) and probably the
most technical discussion of the galvanizing process is "The Galvanizing
Manual" published by the St. Joe Minerals Corporation (6). A good tech-
nical discussion of the various alloy layers which make up the zinc
coating in galvanizing is given by Mohler (7). Most galvanizers use a
zinc ammonium chloride kettle flux known as "ZACLON," which is manufactured
by DuPont; a data summary listing the chemical composition and properties
of the various types of "ZACLON" galvanizing fluxes is given 1n Appendix A-l.
Most available publications discussed the galvanizing emission
problem In fairly general terms and did not list specific particulate
emission data. The consensus was that grayish-white particulate fumes
are discharged whenever the kettle flux cover is disturbed, when fresh
flux 1s added, and when galvanized articles are dusted with ammonium
chloride (1). Lemke et al_. (8) presented the results of a chemical analysis
of particulate emissions from a galvanizing kettle. The probable
-8-
-------
composition of the emissions was estimated as follows:
NH4 CI
¦
cr>
CO
•
0
by
weight
Zn 0
15.8%
by
weight
Zn
4.9%
by
weight
Zn C12
3.6%
by
weight
C
2.8%
by
weight
h2o
2.5%
by
weight
Oil
1.4%
by
weight
nh3
1.0%
by
weight
Since ammonium chloride (NH^ CI) vaporizes at350° C and the zinc
bath temperature is usually between445° and 460° C, it is not surprising
that NH^ CI makes up the bulk of the particulate emissions. However,
zinc and zinc chloride have very low vapor pressures at normal galvanizing
temperatures, and one would expect neither of them to vaporize to any
great extent. It is believed that the discharge of these materials is
the result of mechanical entrainment, and occurs when wet objects are
galvanized or when objects are immersed rapidly through the flux layer (8).
A number of articles discuss the size distribution of the particles
emitted from the hot-dip galvanizing process and methods of controlling
the emissions. Size distributions of galvanizing kettle emissions have
been estimated as averaging 2.0 microns (8), between 0.5 and 2.0 microns
(9), and between 0.1 and 1.0 microns (10). For these extremely fine
particulate emissions, three types of control equipment have been
recommended (9, 10, 11):
(1) Electrostatic precipitators;
(2) High pressure-drop water scrubbers;
(3) Fabric filters (baghouses).
Of these three methods, fabric filters are considered the most economical
in terms of operating cost.
-9-
-------
Lynam (12) discusses the health problem of the particulate
galvanizing emissions. The main health problems are zinc chloride
(Zn CIg) and zinc oxide (Zn 0), which can cause metal fume fever
when high concentrations are inhaled. The Occupational Safety and
Health Association (OSHA) threshold limit values are:
3
Zn Clg - 1 mg/m ;
' 3
Zn 0 - 5 mg/m.
However, the literature search did not reveal any measurements of the air-
borne concentrations of these compounds in the vicinity of a galvanizing
kettle.
3.3 Specific Particulate Emission Data
Attempts to find data from actual source tests in the literature
did not prove very successful. As discussed in Section 3.2 , all
available books, journal articles, industrial publications,' and preprints
on hot-dip galvanizing processes discussed the emission problem in
fairly general terms and did not list specific emission data. Two
possibly relevant journal articles, one in German (13) and one in
Japanese (14), were unavailable in Los Angeles area libraries and in
the statewide University of California library system. Vandegrift
el_ al_. (15) estimated the annual particulate emissions from secondary
zinc processes in the United States as 4,500 metric tons/year; however, the
percentage of this amount due to galvanizing emissions was not stated.
The emission factor listed in "Compilation of Air Pollutant
Emission Factors" (16) for galvanizing kettles is 2.5 kg particulates/ metric ton
of zinc used or 5 lb/short ton of zinc used. The listed reference
for this figure is a report by the Los Angeles County Air Pollution
Control District (APCD) written 1n 1966 (17). Contact with the Los
Angeles County APCD revealed that this report, probably unbound, is no
longer available; also, the reported data were most probably source
tests conducted on galvanizing plants in the Los Angeles area before
1960 (18).
-10-
-------
As previously mentioned, the Los Angeles County APCD provided
the results of 21 source and chemical tests on galvanizing kettles
conducted since 1950. One test result was provided by the San Francisco
Bay Area Air Pollution Control District. These data are reported and
discussed in Sections, "Particulate Emission Source .Test Data."
A NEDS computer search for all source tests on galvanizing kettles
produced only one test result; this test was performed on a galvanizing
furnace, which is not a hot-dip operation. Neither the Los Angeles
County tests nor the San Francisco test were listed in the NEDS
computer search.
-11-
-------
4. OBSERVATION OF CURRENT GALVANIZING PRACTICES
4.1 Los Angeles Area
4.1.1 Los.Angeles Galvanizing Company
Visits were made to five different galvanizing plants, two in the
Los Angeles area and three in the Chicago area, to observe current
hot-dip galvanizing practices, procedures, and equipment. The first
plant visited was Los Angeles Galvanizing Company, 2524 East 52nd
Street, Huntington Park, California, on October 15, 1975. As discussed
in Sections, three source tests for particulate emissions were,
carried out by the PES staff on this plant.
Galvanizing kettles are usually about 1.2 meters In width and
1.5 meters deep, and vary only in length. This plant has one kettle
9.5 meters long which 1s run at a temperature between 443° and 449° C.
The kettle flux used was zinc ammonium chloride, ZACLON Type 2N (See
Appendix A-l). Contrary to typical galvanizing procedures, no rinsing
or prefluxing steps were used 1n article preparation; the sequence of steps
In preparation for galvanizing were as follows:
(1) Degreasing in a hot, alkaline solution;
(2) Pickling in a hot» sulfuric acid bath;
(3) Pickling in a hot, hydrochloric acid bath.
The main particulate emissions, consisting of grayish-white fumes,
were observed when an article was iirmersed through the kettle flux layer,
when the flux layer was agitated, and when a galvanized article was
dusted with NH^ CI. This plant uses a canopy hood over the kettle and
a lime-injected baghouse for emission control. The baghouse was designed
by Industrial Clean Air (ICA) and employs 1,152 nylon bags. The bags
are coated with FUntkote Type S lime, which 1s a mixture of 60% Ca (OH^
and 40% Mg (0H)2. No visible emissions were observed from the baghouse
exhaust. As discussed in Section 5, this baghouse was tested for
efficiency and particulate emissions by'the PES staff.
-12-
-------
4.1.2 Joslyn Manufacturing and Supply Company,
Dominguez Galvanizing Plant
A visit was made to the Dominguez Galvanizing Plant of Joslyn
Manufacturing and Supply Company, 2226 East Dominguez Street, Long
Beach, California on October 31, 1975. This is a large plant which
has three galvanizing kettles: one 13.7 meters long, one 7.9 meters long,
and one 4.5 meters long. The kettle flux used for all three kettles was
zinc armionium chloride, ZACLON Type 2N (see Appendix A-l). All kettles
were run at a temperature between 450° and 454° C.
The sequence of steps in preparation for galvanizing were as
follows:
(1) Degreasing in a hot, alkaline solution;
(2) Rinsing thoroughly in a water bath;
(3) Pickling in a hot, sulfuric acid bath;
(4) Rinsing thoroughly in a water bath;
(5) Prefluxing in a zinc ammonium chloride
solution (ZACLON Type F;. see Appendix A-l).
The main particulate emissions, consisting of grayish-white fumes,
were observed when an article was immersed through the kettle flux layer
or the flux layer was agitated, and especially when a galvanized object
was sprayed with NH^ CI. All other plants which were observed in
finishing articles sprinkled NH^ CI on the galvanized article by hand;
this plant used a spray-gun technique with large quantities of NH^ CI,
creating dense grayish-white fumes.
For emission control, this plant uses three individual lime-injected
baghouses, all designed by ICA. The smallest kettle ( 4.6 meters long) had
a direct hood over thie kettle leading to a small baghouse. The two larger
kettles, which were in individual large rooms, employed roof-level exhausts
leading to a baghouse (a separate baghouse for each kettle). Thus, the
entire room air was exhausted for the two larger kettles. The three
-------
baghouses all used Dacron bags and were injected with Flintkote Type S
lime (60% Ca (0H)2, 40% Mg (OH)2)• No visible emissions were observed
from any of the baghouse exhausts.
4.1 Chicago Area
4.1.1 Reliable Galvanizing Company
A visit was made to Reliable Galvanizing Company, 8800 South
Genoa Avenue, Chicago, Illinois on November 6, 1975. This plant has
one galvanizing kettle 12.8 meters long which is run at about 454° C. The
unusual aspect of this plant's operation was that no kettle flux was
used; in the galvanizing literature, this is termed "dry" galvanizing,
as opposed to the normal "wet" galvanizing technique using a kettle
flux. A preflux was used just before the galvanizing kettle.
The sequence of steps in preparation for galvanizing were as
follows:
(1) Degreasing in a hot, alkaline solution;
(2) Pickling in a hot, sulfuric acid bath;
(3) Prefluxing in a zinc anmonium chloride solution
(manufactured by Stauffer Chemical Company).
The main particulate emissions, consisting of grayish-white fumes,
were observed when an article was dipped into the molten zinc, apparently
vaporizing some of the zinc ammonium chloride preflux. No finishing of
galvanizing articles with NH^ CI was done at this plant. No particulate
emission control device was used; emissions were simply vented to the
atmosphere.
4.2.2 Joslyn Manufacturing and Supply Company,
Morgan Street Division
A visit was made to the Morgan Street Division of Joslyn Manufacturing
and Supply Company, 3700 South Morgan Street, Chicago, Illinois on
-14-
-------
November 6, 1975. This plant had three kettles, each 4.9 meters long, which
were 1n operation; one kettle 7.9 meters long, was being repaired. The
kettles were operated at a temperature between 449° to 454° C, and used
a zinc amnonium chloride kettle flux, ZACLON Type F (see Appendix A-l).
The sequence of steps in preparation for galvanizing were as
follows:
(1) Degreasing in a hot, alkaline solution;
(2) Rinsing thoroughly in a water bath;
(3) Pickling in a hot, sulfuric acid bath;
(4) Rinsing thoroughly in a water bath;
(5) Prefluxing in a zinc ammonium chloride
solution (ZACLON Type F; see Appendix A-l).
The main particulate emissions, consisting of grayish-white fumes,
were observed when an article was immersed through the kettle flux layer
or the flux layer was agitated, and when galvanized objects were dusted
with NH^ CI. Canopy or side hoods were used over the galvanizing kettles;
however, no control device was used and the particulate emissions were
simply vented to the atmosphere from roof vents.
4.2.3 Josl.yn Manufacturing and Supply Company,
Empire Galvanizing Division
A visit was made to the Empire Galvanizing Division of Joslyn
Manufacturing and Supply Company, 10909 Franklin Avenue, Franklin Park,
Illinois on November 7, 1975. This plant had one galvanizing kettle
15.9 meters long, which was kept at a temperature of 449° - 454° C. The
kettle flux used was zinc ammonium chloride, ZACLON Type F (see Appendix A-l).
The sequence of steps in preparation for galvanizing were as
follows:
(1) Degreasing 1n a hot, alkaline solution;
-15-
-------
(2) Rinsing thoroughly in a water bath;
(3) Pickling in a hot, sulfuric acid bath;
(4) Rinsing thoroughly in a water bath;
(5) Prefluxing in a zinc ammonium chloride
solution (ZACLON Type F; see Appendix A-l).
The main particulate emissions, consisting of grayish-white fumes,
were observed when an article was immersed through the kettle flux layer
or the flux layer was agitated, and when a galvanized article was dusted
with NH^ CI.' As in Ooslyn's Dominguez Galvanizing Plant, the entire
room air was vented to a lime-injected baghouse for emission control.
The baghouse was designed by Industrial Clean Air (ICA), and the
polyester bags were injected with Hydrofol Type S lime. No visible
emissions were observed from the baghouse exhausts.
It should be noted that an observation that a plant is operating with
or without controls or visible emissions 1s not sufficient information to
determine whether a plant is in compliance with all applicable emissions
regulations. Normally, emission regulations are given 1n terms of a specific,
particulate emission rate, and source tests are usually necessary to de-
termine whether a plant 1s 1n compliance.
-16-
-------
5. PARTICULATE EMISSION SOURCE TEST- DATA
5.1 Existing Source Test Data
5.1.1 Los Angeles County Air Pollution Control District
The Los Angeles County Air Pollution Control District (APCD)
provided by far the most relevant data for determining an emission
factor - the results of 21 source and chemical tests on galvanizing
kettles conducted in the Los Angeles area over a 25-year period. Some
information which was considered confidential was not provided; however,
these data provided by the Los Angeles County APCD were the largest
single source of particulate emission data from galvanizing kettles.
As discussed in Section 3.3, the Los Angeles County data before 1960 were
the probable source of the emission factor listed in "Compilation of
Air Pollutant Emission Factors" (16).
Of the 21 source and chemical tests provided, three tests were
concerned only with chemical analysis and one source test was on a
galvanizing furnace, which is not a hot-dip operation. Of the remaining
17 source tests on galvanizing kettles, two tests did not measure process
weight, and two very early tests, done in 1950 and 1951, listed extremely
low and questionable process weights. Thus, there were thirteen source
tests which were considered relevant 1n determining an emission factor
based on process weight.
The results of the thirteen relevant source tests conducted by
the Los Angeles County APCD are given 1n Table 5-1, along with the
type of control device tested. The main control devices used for
galvanizing emissions were baghouses and water scrubbers; .as shown 1n
Table 5-1, with a few exceptions, the control devices did not work very
well for galvanizing emissions. Reasons for the relatively poor
efficiency are discussed In section 6.1. Four of the thirteen tests,
as Indicated 1n Table 5-1, were on continuous process operations
which galvanized chain-link fencing. The other tests were on batch-
operation, hot-dip galvanizing processes. Names and addresses of the
-17-
-------
plants tested in Table 5-1 are given in Appendix A-4.
The Los Angeles County APCD did not use EPA Method 5 in their
testing, but used a similar method. The APCD test procedure prior to
1971 differed from later procedures in the method of collection of
particulates which are volatile at the exhaust temperature (19).
However, since galvanizing emissions are mainly solid particles and
are normally at ambient temperature, all of the Los Angeles County
APCD tests can be considered to accurately represent total particulate
emissions.
5.1.2 San Francisco Bay Area Air Pollution Control District
One source test on emissions from a galvanizing kettle was
received from the San Francisco Bay Area Air Pollution Control District.
This source test, done in 1970 in the San Francisco area, measured
particulate emissions only from the exhaust of a water scrubber control
system. The results of this test, which used not EPA Method 5 but a
similar method, are shown in Table 5-T.
5.2 Source Tests Conducted by Pacific Environmental Services, Inc.
On December 2 and 3, 1975, the PES staff conducted three separate
source tests on one galvanizing plant in the Los Angeles area, Los
Angeles Galvanizing Company in Huntington Park, California. This plant
uses a canopy hood over the galvanizing kettle and a lime-injected bag-
house for emission control. The three source tests used EPA Method 5,
and two of the tests measured particulates simultaneously from both the
Inlet and exhaust of the baghouse; on one test, only the baghouse exhaust
was tested due to equipment malfunction. Details of the source tests
are given in Appendix A-2.
-18-
-------
The main results of the PES source tests are given in Table 5-1,
for ready comparison with the other source test results. The baghouse
did not work very well; this fact is discussed in Section 6. One
parameter which was measured in the PES tests was the surface area of
the galvanized product. Table 5-2 compares the product surface area
with the measured emission rates.
19-
-------
Table 5-1 SUMMARY OF GALVANIZING SOURCE TESTS
Test Date
Agency
Galvanized Product
, Process Weight
(metric tons/hr)
Control Device
Kettle
Emissions
(kg/hr)
Emissions From
Control Device
(kg/hr)
Efficiency
(*).
2/13/57
L.A. APCD
4.04
Water scrubber
0.24
0.26
(Negative)
3/17/58
L.A. APCD
0.69
Water scrubber
0.54
0.73
(Negative)
10/22/58*
L.A. APCD
2.33
Electrostatic precipitator
0.64
0.033
95
3/10/59*
L.A. APCD
1.00
Water scrubber
0.32
0.007
98
3/26/59
L.A. APCD
0.56
Baghouse (heated)
0.30
0.15
51
4/16/59
L.A. APCD
3.71
Water scrubber
0.52
0.71
(Negative)
5/11/59
L.A. APCD
1.36
None
0.25
m
-
9/14/59*
L.A. APCD
2.59
Electrostatic precipitator
0.70
0.15
79
4/30/70
L.A. APCD
4.57
Water scrubber
0.077
0.064
18
2/24/72*
L.A. APCD
1.77
Baghouse (heated)
0.091
0.10
(Negative)
6/6/73
L.A. APCD
11.81
Baghouse (I1me-1njected)
0.65
0.30
53
10/15/73
L.A. APCD
4.87
Baghouse (lime-injected)
0.95
o.n
88
4/4/74
L.A. APCD
3.58
Baghouse (Hme-injected)
0.46
0.43
6
2/5/70
S.F. APCD
2.72
Water scrubber
m
0.66
-
12/2/75
PES
2.11
Baghouse (lime-injected)
1.47
1.46
0
12/3/75
PES
2.27
Baghouse (lime-injected)
-
0.75
-
12/3/75
PES
2.08
Baghouse (lime-injected)
1.23
1.48
(Negative)
'continuous process (chain-link fencing)
-------
Table 5-2. RESULTS OF PES SOURCE TESTS
Test Date
Product
Surface Area
(m2/hr)
Kettle
Emissions
(kg/hr)
Emissions from
Control Device
(kg/hr)
! 12/2/75
79.0
1.47
1.46
12/3/75
52.5
-
0.75
12/3/75
72.3
1.23
1.48
i
In section 6-4, the results of Table 5-2 are discussed in the possible
development of an emission factor based on surface area.
-21-
-------
6. DATA ANALYSIS
6.1 Controlled vs. Uncontrolled Emissions
The most striking feature about Table 5-1 1s the fact that
there was not a significant difference between kettle emissions and
emissions from control devices. The control devices tested, mainly
water scrubbers and baghouses, did not work very well for galvanizing
emissions. Of the fourteen cases in which efficiency of the control
device was tested, eight cases had efficiencies less than 20%, and
five cases had apparently negative efficiencies.
The probable explanation for the low efficiencies is the fact
that the control devices are not designed to control the fairly low
and unsteady emission rates characteristic of galvanizing processes,
but are designed to control higher, continuous emission rates. Also,
in the case of lime-injected baghouses, it is possible that some of
the added powdered lime esca'pes into the atmosphere from the bag house
exhaust. In the tests conducted by PES, a qualitative chemical
analysis indicated a different chemical composition of the material
collected at the inlet and exhaust of a lime-1njected baghouse
(see Appendix A-2).
From observations, of galvanizing plants, it 1s evident that
visible emissions are released only at certain intervals, when the
kettle flux cover is disturbed or when galvanized articles are
dusted with ammonium chloride. These-intermittent emissions,-when
averaged over a one-hour period, result 1n fairly low kettle emission
rates, as shown 1n Table 5-1. Thus, control devices are likely to
function mainly to reduce visible emissions rather.than to achieve
significant quantitative reductions In average emissions from gal-
vanizing processes.
Since no significant differences were found between kettle emissions
and emissions from control devices, 1n the analyses below, both kettle
and control device emissions will be considered together as one data
base. As will be demonstrated, the differences Involved 1n considering
22-
-------
kettle and control device emissions separately are much less than
the differences between individual data points.
6.2 Emission Factor Based on Process Weight
In an assessment of local air pollution, there is a need to
estimate the emissions from galvanizing plants as a function of
process rates, procedures, or equipment. For such an estimation, a
useful tool is an emission factor which relates the quantity of
pollutants emitted to some easily measurable activity level or rate.
Typically, emission factors are estimated as arithmetic averages of
a number of data points from source tests.
In "Compilation of Air Pollutant Emission Factors" 06), the
emission factor listed for galvanizing kettles is 2.5 kg particulates/metric
ton of zinc used or 5 lb/short ton of zinc used. The listed
reference for this figure is a report by the Los Angeles County
APCO, written in 1966 (17). Contact with the Los Angles County
APCD revealed that this report, probably unbound, is no longer
available; also, the reported data were most probably source tests
conducted on galvanizing plants in the Los Angeles area before 1960
(18).
The problem with having an emission factor based on the amount
of zinc used is that this quantity is rarely measured and usually must
be estimated. In all the galvanizing plants which the PES staff has
visited and in the majority of the source tests, only the total
galvanized product weight was measured. Galvanizing plants base
their financial charges on measurement of galvanized product weight,
and do not normally measure the amount of zinc used, except over a
long period of time, such as monthly or yearly. Thus, it is recommended
that an emission factor in terms of process weight be defined per weight
of galvanized product and not per weioht of zinc used.
-------
Only four of the source test results estimated the amount of
zinc added. It should be noted that even in these four tests, the
amount of zinc added was estimated, not measured, as a function of
the total amount galvanized. The results of these four tests, all
conducted by the Los Angeles County APCD, are given in Table 6-1.
As can be seen, typically the amount of zinc in the finished galvanized
product is about 835 by weight. All other source tests measured only
the galvanized product process weight.
Table 6-1
SOURCE TESTS WHICH ESTIMATED AMOUNT OF ZINC ADDED
Test Date
Galvanized Prbduct
, Propess Weiqht.
(metric ton/hour)
Estimated
. Zinq Added, .
(metric ton/hr)
Zinc in
Galvanized Product
(% by weight)
2/24/72*
1.77
0.132
7.44
4/4/74
3.58
0.326
9.09
10/15/73
4.87
0.376
7.73
6/6/73
11.81
0.826
6.99
Ave. = 7.81
(
-------
From the data in Table 5-1, emission factors based on galvanized
product process weight can be calculated. The results are shown in
Table 6-2, with the emission factors in units of kg particulates/
metric ton galvanized product. As can be seen, there is not a sig-
nificant difference between the kettle and control device emission
factors. Considering both kettle arid control device emissions as one
data base, an averaged emission factor can be calculated:
Emission factor = 0.26 kg/metric ton galvanized product
(0.52 lb/short ton galvanized product)
o * 0.28 kg/metric ton galvanized product
It should be noted that the standard deviation (a) is of the
same order of magnitude as the averaged emission factor, Indicating
the considerable scatter 1n the data base. The standard deviation
(a) 1s a measure of the spread of a statistical distribution, and
is defined as,
I 1 (Xi - x)2
V N-1
where N is the total number of data points and 7. 1s the arithmetic
average. If a Gaussian distribution 1s assumed, approximately 95% of
the data points can be expected to He within + Za of the arithmetic
average.
In order to compare the calculated result with the emission factor,
2.5 kg/metric ton zinc, listed 1n "Compilation of Air Pollutant Emission
Factors" (16), the typical factor of 8% zinc (by weight) In the
finished galvanized product can be used, I.e., the equivalent number
would be (2.5 x 0.08) ¦ 0.2 kg/metric ton galvanized product. Likewise, the
-25-
-------
new averaged emission factor can be expressed in terms of the amount of
zinc added:
Emission factor =3.3 kg/metric ton zinc used
(6.6 lb/short ton "zinc used)
& =3.5 kg/metric ton zinc used
Thus, the new averaged emission factor, based on process weight, is
only slightly higher than the listed number in "Compilation of Air
Pollutant Emission Factors" (16). This fact is not surprising, since
both numbers are based on some of the same data, namely the eight
Los Angeles County APCD tests before 1960.
6.3 Emission Factor Based on Hours of Operation
The data points in Table 5-1, including both kettle and control
device emissions, are plotted in Figure 6-1, as emissions vs. galvanized
product process weight. The solid line Indicates the averaged emission
factor of 0.26 kg/metric ton galvanized product. However, in order to
determine the best fit to the data, a linear least squares regression
analysis was run on the data points; the regression analysis is
described in Appendix A-3.
Considering all the data points in Figure 6-1, if emissions in
kg/hr are designated as y, and the process weight in metric ton/hr of
galvanized product Is designated as x, then the best-fit straight line
(see Appendix A-3) 1s given by:
y = 0.54 -0.009 x (1)
This relationship 1s plotted as the dashed line 1n Figure 6-1. Due to
the very slight dependence on x, 1t 1s essentially a horizontal straight
line.
-26-
-------
Table 6-2. EMISSION FACTORS BASED ON PROCESS WEIGHT
Units: kg parti ctilates/metric ton galvanized product
Test Date
Agency
Kettle
Emission Factor
Control Device
Emission Factor
2/T3/57
L.A. APCD
0.061
0.064
3/17/58
L.A. APCD
0.790
1.053
10/22/58*
L.A. APCD
0.273
0.014
3/10/59*
L.A. APCD
0.323
0.007
3/26/59
L.A. APCD
0.545
0.269
4/16/59
L.A. APCD
0.141
0.191
5/11/59
L.A. APCD
0.187
-
9/14/59*
L.A. APCD
0,270
0.056
4/30/70
L.A. APCD
0.017
0.014
2/24/72*
L.A. APCD
0.052
0.057
6/6/73
L.A. APCD
0.056
0.026
10/15/73
L.A. APCD
0.196
0.024
4/4/74
L.A. APCD
0.128
0.120
2/5/70
S.F. APCD
-
0.244
12/2/75
PES
0.693
0.691
12/3/75
PES
-
0.332
12/3/75
PES
0.592
0.714
'
Ave. = 0.288
(0.58 lb/short
ton)
-------
PARTICULATE
EMISSIONS
(KG/HOUR)
1 ,
1
i i i
o - Kettle Emissions
-
• - Control Device
Emissions
•o
•
o *
" So
• /
c/* #
•
°JS 8
X. . «•
« •
1 1 1
2 4 6 8 10 12
GALVANIZED PRODUCT PROCESS WEIGHT (METRIC TONS/HOUR)
Figure 6-1. PARTICULATE EMISSIONS VS. GALVANIZED PRODUCT PROCESS WEIGHT
-------
The best fit straight line draws the conclusion that particulate
emissions are essentially independent of the amount of material
processed. As discussed in Appendix A-3, similar linear least-squares
regression analyses were run on the kettle and control device
emissions separately, and were also run neglecting the two data
points at the very high process weight (11.81 metric ton/hr). In all
cases, very similar, almost horizontal straight lines were found,
indicating that galvanizing emissions are substantially independent
of process weight. This conclusion is supported by the fact that, in
the hot-dip galvanizing process, the main emissions are caused by
articles entering and exiting the kettle, i.e., the process of
disturbing the kettle flux layer on entering and exiting the kettle
is more important than the absolute weight of the article galvanized.
Equation (1) could be used*as an emission factor equation,
however, due to the very small dependence on process weight, a simpler
approach is to assume an emission factor which is independent of
process weight. Simply averaging both kettle and control device
emission data points in Table 5-1 yields the following result:
Emission factor =0.51 kg/hr per kettle
(1.12 lb/hr per kettle)
v* 0.44 kg/hr per kettle
This approach thus uses the conclusion that galvanizing emissions are
independent of process weight, and relates emissions simply to the total
time of kettle operation and the number of kettles in operation.
Again, the rather large standard deviation (cr) indicates the consider-
able scatter in the data base, even though this approach is essentially
the best linear fit mathematically.
-29-
-------
6.4 Emission Factor Based on Surface Area
From the three source tests conducted by the PES staff, a
relationship can be derived between galvanizing emissions and the
surface area of the galvanized product. From the data given in
Table 5-2, emission factors based on galvanized product surface area
can be calculated. The results are shown in Table 6-3, with the
2
emission factors in units of kg particulate/m galvanized product.
Table 6-3. EMISSION FACTORS BASED ON SURFACE AREA
Units: kg particulates/m2 galvanized product
Test Date
Kettle
Emission Factor
Control Device
Emission Factor
12/2/75
12/3/75
12/3/75
0.0185
•
0.0170
0.0185
0.0143
0.TJ205
Considering both kettle and control device emissions, an averaged
emission factor can be calculated:
Emission factor a 0.0178 kg/m2 2
(0.0036 lb/ft )
<* * 0.0023 kg/m2
It should be noted thit this emission factor Is based on data
from only one galvanizing plant and only five data points. The emission
factors 1n Sections 6.2 and 6.3 are based on data from eleven different
plants and 31 data points. Thus, this approach based on surface area
should only be considered as a preliminary conclusion. Also, since no
records are kept of galvanized product surface area, this type of emission fac
tor would be difficult to use 1n practice.
-30-
-------
6.5 Total Galvanizing Emissions in the U.S.
Using the emission factors generated in Sections 6.2 and 6.3,
it is possible to estimate the total annual amount of particulate
•emissions produced by galvanizing plants in the United States. From
U.S. Btireau of Mines data (2), the amount of zinc consumed by gal-
vanizing in the U.S. was 511,505 metric tons in 1973. Using the
estimated emission factor of 3.3 kg/metric ton of zinc used, the total
galvanizing emissions in the U.S. can be estimated at 1,690 metric tons
in 1973. This number is negligible when compared to the total particulate
emissions from all sources, estimated as about 16,000,000 metric tons/yr
in the U.S. by Vandegrift et.iL (15)
Another way of determining the total annual emission rate for
galvanizing is to consider the total hours of operation and the number
of kettles. Galvanizing kettles are operated, on the average, 16 hours
per day, 6 days per week, and 50 weeks per year, or a total of 4,800
hours per year. Using the emission factor of 0.51 kg/hr, each gal-
vanizing kettle produces 2,450 kg or 2.45 metric tons of particulates
annually. The America Hot Dip Galvanizers Association (AHDGA) lists
about 100 members who operate approximately 200 galvanizing kettles.
However, many galvanizers do not belong to the AHDGA, in particular
large steel companies who have extensive in-house galvanizing operations.
If the assumption is made that the AHDGA membership represents
one-third of the galvanizing kettles in the U.S., I.e., a total of 600
kettles are In operation 1n the U.S., then the annual particulate emissions
would be, (600 x 2.45) = 1,470 metric tons from galvanizing operations.
This number Is very close to the previous estimate of 1,690 metric tons
annually. Thus, a plausible estimate of the total amount of particulate
emissions In the United States from galvanizing operations 1s 1,600 metric
tons (1,800 short tons) per year.
-31-
-------
APPENDICES
-------
A-l. DESCRIPTION OF ZACLQN GALVANIZING FLUXES
A-l
-------
DATA SHEET
MIZENG FLUXES
Pont manufactures a series of ZACLON* galva-
ng fluxes to meet the needs of the industry.
» materials are available with varying degrees of
: ammonium chloride activity, additives, and
sical forms. They are used during the galva-
ng process fn the prefluxing, galvanizing, and
shing steps.
11LON galvanizing fluxes protect steel and zinc
nst oxidation, reduce dross formation, mini-
mize ammonium chloride fume losses and insure
good galvanizing. They are economical and easy to
use.
Du Pont's. products for the galvanizing industry
include pickling acids, ammonium chloride fluxes,
and ZACLON galvanizing fluxes. Additional infor-
mation on these products is available from any
sales office of Du Pont's Industrial Chemicals
Department listed on the back page.
'ERAGE ANALYSES - GENERAL PURPOSE GRADES
Grade of ZACLON*
& ammonium chloride
2nCla-2NH4CI,%
;ZnCla-3NH4 Cl,%
|2nCI}-6NH4 CI, %
as ZnCli, %
honia as NH4 CI, %
ZnCU :NH4 CI
I iron as Fe, %
>tes as S03, %
3s PbClj, %
r- insoluble matter, %
fcure, %
c.csa
99.5
56.0
43.5
1.29
0.3
99.4
45.8
53.6
0.85
0.004
0.03
0.02
0.04
0.03
99.7
45.9
53.8
0.85
0.004
0.03
0.02
0.03
0.2
2Nb
99.0
30.0
69.0
0.43
0.004
0.02
0.06
0.04
0.6
bPERTlES
x. density, lb/ft3
yellowish
white
53
69
pj.S. Pat. Off.
I^lent to ZACLON C but includes a wetting agent.
Available in Rod Form.
Off
white
56
63
off
white
57
63
off
white
51
59
information set forth herein is furnished free of charge and is based on technical date thst Du Pont believes to be reliable. It is
*xjed for use by persons having technics! skill end at their own discretion and risk. Since conditions of use ore outside our control
no warranties, express o. implied, ahd assume no liability in connection with any use of this information. Nothinu herein is to
as a liccnfg to 'operate under or a recommendation to infringe any patents.
-------
THE GALVANiZIWG PROCESS
Dean, oxide-free iron or steel is galvanized by
soating it with a thin layer of zinc. This protects
the iron or steel by shielding it from the atmos-
phere, as well ss providing cathodic or sacrificial
Protection. Even if the surface is scratched, ex-
posing the base metal, the more electronegative
fine is slowly consumed while the iron or steel is
Protected. The most important galvanizing method
S the hot dip process which is adaptable to coating
Nearly all fabricated and nonfabricated products
>*jch as wire, tanks, sheets, strip, pipes and tubes,
'ittings, hardware, wire cloth,, hollow ware, and
Trructural assemblies.
lot dip galvanizing consists of these fundamental
teps:
• surface preparation — The surface is de-
greased and pickled to free it of dirt, grease,
rust, and scale.
• prefixing — The surface is coated with a
thin layer of ZACLON* galvanizing flux
which prevents rust and dissolves light
oxides that may have formed since pickling.
• galvanizing — Clean, oxide-free work is im-
mersed into molten zinc. It may be immersed
through a blanket of ZACLON galvanizing
flux.
• finishing - Excess zinc is removed, the piece
is quenched and inspected. ZACLON galva-
nizing fluxes or ammonium chloride may be
used as a dusting sprinkle in some oper-
ations.
APPLICATIONS
bnera! Purpose Fluxes
*{CLON F galvanizing flux is the basic flux used
the industry. It serves either as an aqueous
teflux for work prior to galvanizing or as a direct
Ciition to the kettle to maintain a foaming top
As a preflux it insures uniform zinc coverage,
iluces dross formation and constantly activates
top flux. The foam blanket produced by
KCLON F reduces fuming, zinc spatter, and
kiongs the life of the flux. Galvanizers of struc-
hal shapes, pipe and conduit, tanks, wire, and
fcfcet work generally use this grade.
\CLON K is similar to ZACLON F as a preflux,
does not produce a foaming flux blanket - it
fes a thin, fiuid cover. ZACLON K is generally
Inferred by galvanizers of fittings processed on
1^. U.S. Pat. Off.
automatic continuous-equipment, structural shapes
and nails.
ZACLON 2N is formulated with the high activity
usually obtainable only by separate additions of
ammonium chloride. This flux contains a very
efficient foaming agent. ZACLON 2N can be used
alone as the top flux, or with other fluxes to
maintain an active flux blanket. When using
ZACLON 2N, it is recommended that a few shovel-
fuls of ZACLON HV or the heel of an-old flux be
used to start the top flux. It is not recommended
for use as a preflux.
ZACLON 2N (nonfoaming) is similar to ZACLON
2N, but does not contain a foaming agent. It is
uised to maintain activity in a top flux blanket and
as a replacement for ammonium chloride.
ZACLON CS is specially formulated for use as £n
aqueous preflux in continuous strip galvanizing. It
protects the steel against oxidation at the higher
preheat temperatures required in this process.
ZACLON CS has lower fluxing activity than either
ZACLON F or ZACLON K but generates less
fuming and provides a more stable coating, when
preheat is used.
Solutions of ZACLON CS are also available.
Special Purpose Fluxes
ZACLON A is a nonfuming flux that can be used
as an aqueous preflux or as a fluid cover on molten
zinc. It is useful where strict control of fuming is
necessary. However, ZACLON A is a less active
flux and careful practice is required to insurt
galvanizing quality.
ZACLON AF is a nonfuming flux similar to
ZACLON A. ZACLON AF forms a foamed cover
on the molten zinc and is recommended as a top
flux instead of ZACLON A. ZACLON AF can also
be used in an aqueous preflux solution. As with
ZACLON A, careful galvanizing practice is neces-
sary because of its lower activity.
ZACLON C is generally used in the formulation-of
other fluxes for special tinning and soldering appli-
cations. ZACLON C is similar in fluxing activity to
ZACLON CS.
ZACLON HV is specially formulated to permit
building an active zinc ammonium chloride top
flux without generating excess smoke. The formu-
lation includes an efficient foaming agent that
provides an insulating blanket and further sup-
presses fuming. A more active flux, such as
ZACLON F can then be used as an aqueous
preflux, or as an addition to the top flux for
maintaining activity.
-------
PACKAGES
Du Pont ships ZACLONC, CS, F,.K, 2M and 2N
•(nonfoaming) in 100-lb (net) paper hags. ZACLON
¥ and K are also available in 4001b (net) fiber
drums. The grades containing higher amounts of
iinc chloride such as ZALCON A, AF, and HV are
shipped in 525-lb (net) steel drums. Bulk solution
shipments of certain grades are available.
STORAGE AND HANDLING
solid grades of ZACLON galvanizing fluxes
l^liould be stored so as to avoid moisture pickup.
p"he grades containing higher amounts of zinc
ifcliloride, such as ZACLON A, AF and HV, should
stored in tightly closed containers in a dry
fciace. An inventory turnover rate of 2-3 months is
fhecommended to minimize caking.
§Jnheated solutions of ZACLON slowly corrode
jteel. Rubber-lined steel tanks or fiber-glass rein-
forced polyester tanks are recommended for stor-
age.
PERSONAL SAFETY
AND FIRST AID
Health Hazards
ZACLON galvanizing fluxes, generally zinc ammo-
nium chloride mixtures, are acidic and can cause
severe skin or eye injury. The principal hazard is to
the eye, since even brief contact of the undiluted
product may produce permanent damage.
The U.S. Department of Labor has issued the
regulation that an employee's exposure to zinc
chloride fumes in any 8-hour shift of a 40-hour
work week shall not exceed the 8-hour time-
weighted average of I mg/m3. (Title -29, Part
1910.93 Air Contaminants).
Safety Precautions
All persons handling ZACLON should avoid con-
tact with the powders or solutions. Do not get in
eyes, on skin or on clothing. Avoid breathing dusts,
mists or fumes. Wash thoroughly after handling.
Contaminated clothing should be washed before
reuse. Adequate ventilation should be provided.
First Aid
Iri case of contact of ZACLON with skin, immedi-
ately flush with plenty of water While removing
contaminated clothing and shoes. For eyes, flush
with water for at least 15 minutes and call a
physician. If dusts, mists or fumes are inhaled,
remove person to' fresh air immediately and con-
tact a physician.
All persons handling ZACLON galvanizing
fluxes should be thoroughly familiar with the
additional information on Storage and Hand-
ling and Personal Safety and First Aid in
Du Pont's Galvanizing Handbook. This hand-
book is available from any sales office listed
on the back page.
Iheg. U.S. Pat. Off.
A-4
-------
DETAILS OF PARTICULATE EMISSION SOURCE TESTS
CONDUCTED BY THE PES STAFF
-------
I. DESCRIPTION OF SOURCE TESTS
On December 2 and 3, 1975, Pacific Environmental Services, Inc.
(PES) conducted three separate source tests for particulate emissions
at Los Angeles Galvanizing Company in Huntington Park, California.
This plant uses a canopy hood over the galvanizing kettle and a lime-
injected baghouse for emission control. The three source tests used
EPA Method 5, and two of the tests measured particulates simultaneously
from both the inlet and exhaust of the baghouse; on one test, only the
baghouse exhaust was tested due to equipment malfunction. The test
team was coordinated by Dr. Peter Drivas and included Robert Norton,
Robert Missen, Joseph Boyd, and Bansi Parekh, all staff members of
Pacific Environmental Services.
The baghouse was designed by Industrial Clean Air (ICA) and
employs 1,152 nylon bags. The bags are coated with Flintkote Type S
lime, which is a mixture of 60% Ca(0H)2 and 40% Mg(0H)2. The bag-
house inlet which was tested consists of one 1.1 m by 1.5 m rectangular
duct. The exhaust of the baghouse consists of two identical 0.6 m by
0.6 m square ducts. Only one of the baghouse exhausts (the north one)
was tested; it was assumed that the other exhaust was operating at an
identical velocity and emission rate. This assumption was verified by
a comparison of the inlet and outlet volumetric flow rates. In the
discussions below, all calculated values consider the operation of
both exhausts.
The first test (Run No. 1) on December 2, 1975, tested both the
inlet and exhaust of the baghouse simultaneously. The second test
(Run No. 2) on December 3, 1975, tested only the baghouse exhaust;
due to equipment trouble (a broken glass liner in the probe), the
inlet data were considered invalid. The third test (Run No. 3) on
December 3, 1975 tested both the inlet and exhaust of the baghouse
simultaneously.
-------
II. SUMMARY AND DISCUSSION OF RESULTS
Pertinent data and calculated values for the three source tests
are tabulated in Section III. From the exhaust volumetric flow rates,
the exhaust particulate concentrations can be compared with those cited
in the regulations of the Los Angeles County Air Pollution Control
District (APCD). The allowable particulate concentrations, as given
by Rule 52, are compared with the measured values below:
Thus, the exhaust particulate concentrations comply with the Los
Angeles County APCD regulations (Rule 52). The. particulate concentra-
tions at the baghouse inlet, given in Section III, also comply with
this rule.
From the estimated process weights, the exhaust particulate
emission rates can be compared with those cited in the Los Angeles
County APCD regulations (Rule 54). The allowable particulate
emission rates, as given by Rule 54, are compared with the measured
values below:
Allowable conc., Measured outlet
grains/ft"^ * conc., grains/ft"* *
2
3
0.0622
0.0562
0.0559
0.0176
0.0069
0.0135
Run No.
Allowable emission Measured outlet
rate, Ib/hr emission rate, lb/hr
2
3
6.20
6.49
6.14
3.22
1.66
3.27
Dry gas at standard conditions: T ¦ 15,6° C, p a 76.0 cm Hg.
-------
Thus, the exhaust particulate emission rates comply with the
Los Angeles County APCD regulations (Rule 54). The inlet emission
rates, given in Section III, also comply with this rule.
From the simultaneous testing in Run 1 and Run 3, the
efficiency of the baghouse can be determined. The testing revealed
that the baghouse was very inefficient:
d.,„ Inlet emission Outlet emission 4 /*\
RunNo- rate, kg/hr rate, k9/hr Efficiency »)
1 1.47 1.46 0.3
3 1.23 1.48 Negative
A probable explanation for the poor efficiency is the fact that the
baghouse employs a powdered lime injection system; it is possible
that some of this powdered lime escapes into the atmosphere from the
baghouse exhaust. A qualitative examination of the inlet and outlet
filters resulted in a substantial difference in the pH values of the
collected material, indicating a different chemical composition at
the inlet and outlet. The collected material on the outlet filter
was more basic than the material on the inlet filter, possibly in-
dicating the presence of lime particles on the outlet filter.
-------
III. DATA TABULATION
A. Run No. 1 - 12/2/75
Baghouse
Inlet
Baghouse
Out! et
Time of sampling, PST (24-hour clock)
13:50-14:50
13:50-14:50
Total time of test, min
60
60
Number of sampling points
12
12
2
Area of sampling location, m
1.63
0.74
Sampling nozzle diameter, cm
0.90
0.64
Pitot tube coefficient
0.747
0.790
Process rate, metric ton/hr galvanized product
(approximate)
2.11
2.11
COg in stack gas, % by volume (dry)
0.5
0.5
Og in stack gas, % by volume (dry)
21.0
21.0
Ng in stack gas, % by volume (dry)
78.5
78.5
Total H20 collected in impingers and
silica gel, ml
1.4
7.7
Volume of water vapor collected at
standard conditions, .
0.0019
0.0102
Moisture in stack gas, % by volume
0.18
0.73
Gas molecular weight, dry basis, g/g-mole
28.92
28.92
Gas molecular weight, wet basis, g/g-mole
28.90
28.84
Volume of dry gas sampled at meter
conditions, m^
1.044
1.462
Average gas meter temperature, °C
30.6
37.6
A-9
-------
Run No. 1 (continued)
Baghouse
Inlet
Baghouse
Outlet
Absolute barometric pressure, cm Hg
75.84
75.84
Average orifice pressure drop, cm H2O
3.40
6-86
Volume of dry gas sampled at standard
3
conditions*, m
1.010
1.386
Average stack gas temperature, °C
32.8
35.3
Absolute stack gas pressure, cm Hg
75.84
75.84
Stack gas velocity at stack conditions,
5.19
14.65
m/sec
Dry stack gas volumetric flow rate at
578
614
standard conditions, *m3/nrjn
Isokinetic sampling, %
75
88
Particulate weight: filter, mg
2.0
5.5
Particulate weight: probe and cyclone, mg
33.6
39.6
Particulate weight: 1mp1ngers, mg
7.2
9.8
Total particulate weight, mg
42.8
54.9
Total particulate concentration at standard
* 3
conditions, g/m
0.0423
0.0396
Total particulate emission rate, kg/hr
1.47
1.46
*Standard conditions: T ¦ 20° C, p ¦ 76.0 cm Hg.
A-1Q
-------
B. Run No. 2 - 12/3/75
Baghouse
Outlet
Time of sampling, PST (24-hour clock)
11:00-12:00
Total time of test, min
60
Number of sampling points
12
Area of sampling location, m
0.74
Sampling nozzle diameter, cn\
0.64
Pitot tube coefficient
0.790
Process rate, metric ton/hr galvanized product
2.27
(approximate)
C02 in stack gas, % by volume (dry)
0
02 in stack gas, % by volume (dry)
20
Ng in stack gas, % by volume (dry)
80
Total HgO collected in impingers and
silica gel, ml
11.1
Volume of water vapor collected at
* 3
. standard conditions, m
0.0147
Moisture in stack gas, % by volume
0.92
Gas molecular weight, dry basis, g/g-mole
28.80
Gas molecular weight, wet basis, g/g-mole
28.70
Volume of dry gas sampled at meter
3
conditions,"1
1.632
Average gas meter temperature, °C
29.4
Absolute barometric pressure, cm Hg
75.92
Average orifice pressure drop, cm HgO
9.12
Volume of dry gas sampled at standard
conditions,* nf*
1.594
-------
Run No. 2 (continued)
Baghouse
Outlet
Average stack gas temperature, 0 C
28.9
Absolute stack gas pressure, cm Hg
75.92
Stack gas velocity at stack conditions,
m/sec
18.78
Dry stack gas volumetric flow rate at
* 3
standard conditions, m /m1n
804
Isokinetic sampling, %
78
Particulate weight: filter, mg
0.0
Particulate weight: probe and cyo1one,.mg
21.9
Particulate weight: Implngers, mg
2.9
Total particulate weight, mg
24.8
Total particulate concentration at
standard conditions, g/m3
0.0156
Total particulate emission rate, kg/hr
0.753
~Standard conditions: T « 20° C, p a 76.0 cm Hg.
-------
C. Run No. 3 - 12/3/75
Baghouse
Inlet
Baghouse
Outlet
Time of sampling, PST (24-hour clock)
15:00-15:50
15:00-16:00
Total time of test, min
50
60
Number of sampling points '
10
1.63
0.90
12
Area of sampling location, .m
Sampling nozzle diameter, cm
0.74
0.64
Pi tot tube coefficient
0.804
0.790
Process rate, metric ton/hr galvanized product
(approximate)
2.08
2.08
CO2 1n stack gas, % by volume (dry)
0
0
O2 1n stack gas, % by volume (dry)
20
20
N2 In stack gas, % by volume- (dry
80
80
Total H2O collected 1n 1mp1ngers and
silica gel, ml
11.7
11.3
Volume of water vapor collected at
standard conditions, * nr
0.0156
0.0150
Moisture in stack gas, % by volume
1.14
1.15
Gas molecular weight, dry basis, g/g-mole
28.80
28.80
Gas molecular weight, wet basis, g/g-mole
28.68
28.68
Volume of dry gas sampled at meter
conditions, m3
1.409
1.316
Average gas meter temperatue, °C
35.0
27.3
Absolute barometric pressure, cm Hg
75.79
75.79
Average orifice pressure drop, cm H2O
11.94
7.57
Volume of dry gas sampled at standard
conditions, m^
1.353
1.290
-------
Run No. 3 (continued)
Baghouse Baghouse
Inlet Outlet
Average stack gas temperature, °C
26,7
28.3
Absolute stack gas pressure, cm Hg
75.79
75.79
Stack gas velocity at stack conditions,
m/sec
7.84
19.02
Dry stack gas volumetric flow rate at
standard conditions, * m3/m1n
736
813
Isokinetic sampling, %
94
62
Particulate weight: filter, mg
15.3
0.0
Particulate weight: probe and cyclone, mg
13.0
29.6
Particulate weight: impingers, mg
9.3
9.6
Total particulate weight, mg
37.6
39.2
Total particulate concentration at
standard conditions, * g/m3
0.0277
0.0304
Total particulate emission rate, kg/hr
1.23
1.48
*Standard conditions: T ¦ 20° C,
p ¦ 76,0 cm Hg.
-------
LINEAR LEAST SQUARES REGRESSION ANALYSIS
-------
To fit data with a linear equation of the form,
y - a + bx
the optimal technique is to find the values of a and b which
minimize the sum of the squares of the deviations from the actual
data points (20). The best values of the coefficients a and b are
mathematically,
2*iZ2y, -2*, *,)
Nix,2 - (Sx,)2
ft|£(xj y^) -2x12yj
NSx/ - (Sxj)2
where N is the total number of data points.
The best fit least squares line for four cases was calculated using
the data points listed in Table 5-1.. The results were as fallows:
Case 1. Kettle emissions (Y) vs. Process weight {x):
y * 0.53 + 0.010 x
Case 2. Control device.emissions (y) vs. Process weight (x):
y * 0.55 - 0.028 x
Case 3. Kettle and control device emissions (y) vs.
Process weight (x):
y » 0.54 - 0.009X
-------
Case 4. Kettle and control device emission (y) vs.
Process weight (x), neglecting the (11.81, 0.65)
and (11.81. 0.30) data points:
y = 0.58 - 0.024x
As can be seen, there was no significant difference In consider-
ing the kettle and control device emissions separately (Cases 1 and 2).
Likewise, neglecting the two data points at the very high process
weight (11.81 metric ton/hr) did not change the best fit line
significantly (Case 4). In all cases, there was very little dependence
on the process weight (x), Indicating that galvanizing emissions are
essentially independent of the process weight.
-------
A-4. ADDRESSES OF TESTED GALVANIZING PLANTS
1. Test Date: 2/13/57
Anchor Post Products, Inc.
620 Putman Drive
Whittler, Ca.
2. Test Dates: 3/17/59; 3/26/59
Superior Pacific Galvanizing Company
1711 East 61st Street
Los Angeles, Ca.
3. Test Dates: 10/22/58; 9/14/59
Advance Galvanizing Company
5232 Alcoa Avenue
Vernon, Ca.
4. Test Date: 3/10/59
Western Galvanizing Company.
2701 South Soto Street
Los Angeles, Ca.
5. Test Dates: 4/16/59; 4/30/70
Emsco Steel Products Company
6811 Alameda Street
Los Angeles, Ca.
6. Test Dates: 5/11/59; 12/2/75; 12/3/75
Los Angeles Galvanizing Company
2524 East 52nd Street
Huntington Park, Ca. 90255
7. Test Date: 2/24/72
Davis Wire Corporation
6315 Bandini Boulevard
Los Angeles Ca. 90040
8. Test Date: 6/6/73
Joslyn-Pacific Company
2226 East Dominguez St.
Long Beach, Ca. 90810
9. Test Date: 10/15/73
Western Galvanizing Company
9719 Santa Fe Springs Road
Santa Fe Springs, Ca. 90670
A-18
-------
10. Test Date: 4/4/74
Emsco Galvanizing Company
7001 South Alameda St.
Huntington Park, Ca. 90255
11. Test Date: 2/5/70
Triangle Conduit and Cable Company
Western Steel Conduit Division
1666 Willow Pass Road
Pittsburg, Ca. 94565
A-19
-------
LIST OF REFERENCES
1. Danielson, J. A. (compiler and editor), "Air Pollution Technical
Manual," Second Edition, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, Publication No. AP-40
(May 1973).
2. McMahon, A. n., J. M. Hague, and H. R. Babitzke, "Zinc," in Minerals
Yearbook 1973, Volume I: Metals, Minerals, and Fuels, U.S.
Department of the Interior, Bureau of Mines, U.S. Government
Printing Office, Washington, D.C., pp. 1303-1343 (1975).
3. General Galvanizing, A Manual of Good Practice, Hot Dip Galvanizers
Association, London (August 1957).
4. Fifth International Conference on Hot Dip Galvanizing, June 1958,
European General Galvanizers Association, Sidney Press Limited,
London (1959).
•
5. Burns, R. M. and W. W. Bradley, Protective Coatings for Metals, *
Second Edition, Reinhold Publishing Comp.any, New York (1962).
6. "The Galvanizing Manual," St. Joe Minerals Corporation, 250 Park
Avenue, New York (1974).
7. Mohler, J. B., "Properties of Hot-Dip Galvanizing," Metal Finishing,
71, 37-41 (August 1973).
8. Lemke, E. E., W. F. Hammond, and G. Thomas, "Air Pollution Control
Measures for Hot Dip Galvanizing Kettles," J. Air Pollut.
Control Assoc., 10, 70-77 (I960).
9. Calder, B. L., "Abatement of Air Contaminant Emissions from Hot Dip
Galvanizing Plants," Presented at the 1972 semi-annual meeting
of the American Hot Dip Galvanizers Association, Toronto,
Canada (September 17-20, 1972).
10. Miller, T. S., "The State of the Art in Controlling and Filtering
Hot Dip Galvanizing Kettle Emissions," Presented at the 38th
Annual Meeting of the American Hot Dip Galvanizers Association,
Houston, Texas (March 13-17, 1973).
A-20
-------
List of References (continued)
11. Mantle, E. C., "The Secondary Metals Industry and the Environment,"
Clean Air, 3, 24-29 (Summer 1973).
12. Lynam, D. R., "OSHA Health Requirements' for the Galvanizing Industry,"
Presented at the 38th Annual Meeting of the American Hot Dip
Galvanizers Association, Houston, Texas (March 13-17, 1973).
13. Schwanecke, R., "Emission from Hot Galvanization," Wasser Luft
Betreib., 18 (4), 208-211 (1974). Text in German.
14. Nakano, K., K. Hishida, and M. Takeda, "Characteristics and Treatment
of Waste Gas from Melted Zinc Plating Process,"
-------
TECHNICAL REPORT DATA
(Please read Inu/uctions on lite reverse before completing)
'•mT«ft/4-76-002
2.
3. RECIPIENT'S ACCESSIOWNO.
4. TITLE AND SUBTITLE
Emissions From Hot-Dip Galvanizing Processes
6. REPORT DATE
March, 1976
S. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Peter J. Drivas
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND AOORESS
Pacific Environmental Services, Inc.
1930 14th Street
Santa Monica, California 90404
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-01-3156 .
Task Order No. 7
12. SPONSORING AGENCY NAME ANO AOORESS
U.S. Environmental Protection Agency, Region V
230 South Dearborn Street
Chicago, Illinois 60604
13. TYPE OF REPORT ANO PERIOO COVEREO
Final
14. SPONSORING AGENCY CODE
IS. SUPPLEMENTARY NOTES
16.abstract^ literature review and source tests were performed in order to
accurately characterize emissions from hot-dip galvanizing processes. Particulate
emission data from seventeen source tests on hot-dip galvanizing plants were
considered relevant in developing an emission factor for galvanizing kettles.
There was rfo significant difference between kettle emissions and emissions from
control devices. Considering both kettle and control device emissions as one
data base, two types of emission factors were calculated:
(1) An emission factor based on process weight, equal to
0.26 kg/metric ton galvanized product (o »0.28), or assuming
8% zinc by weight 1n the galvanized product, 3.3 kg/metric ton
zinc used (3.5).
(2) An emission factor based on the hours of kettle operation,
equal to 0.51 kg/hr per kettle (<* ¦ 0.44).
llsing the derived emission factors, the total amount of particulate emissions
produced by hot-dip galvanizing operations 1n the United States 1s estimated
jfco be approximately 1,600 metric tons per year.
I "7. KEY WORDS ANO DOCUMENT ANALYSIS
k DESCRIPTORS
b.identifiers/open ended terms
c.' COSATI Field/Group
Air pollution control
Emissions
Exhaust gases
Hot-dip coatings
Z1nc coatings
13 B
13 H
11 F
mi. distribution statement Document 1s available
to public through the National Technical
Information Service, Springfield# Va.2215'
kA 999A.1 ia.79l
10. SECURITY CLASS (ThitReport)
Unclassified
21.N<^>F PAGES
20. SECURITY CLASS (Thit page)
Unclassified
22. PRICE
»»A Form 2220-1 (••73)
------- |