United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S2-83-092 Dec. 1983
v°/EPA Project Summary
Multimedia Environmental
Assessment of Electric
Submerged Arc Furnaces
Producing Ferroalloys
C.W. Westbrook
The report gives results of sampling
and analysis at five ferroalloy plants to
determine the amounts of particulates
and organics generated during
manufacture of several products, the
effect of furnace type on the amounts
generated, and the amounts of these
materials escaping to the environment.
Furnace emission reduction systems
tested were: baghouses serving open
furnaces producing silicon metal (Si),
high-carbon ferromanganese (H.C.FeMn),
and 50 percent grade ferrosilicon
(FeSi); and scrubbers (primary
emissions) and baghouses (secondary
emissions) serving semi-enclosed
furnaces producing H.C.FeMn and 50
percent grade FeSi. All air pollutant
sampling was conducted using
modified EPA Method 5. Simultaneous
samples were obtained before and after
the capture device except for the
scrubbers, where samples were taken
only after the device. For scrubber
systems, composite scrubber discharge
water samples were also taken
simultaneously with the air samples.
Samples were also taken of the influent
to and effluent from two plant
wastewater treatment systems and
groundwater samples around the waste
disposal area of one plant.
Particulate emissions ranged from
0.078 kg/MW-hr for the open FeMn
furnace to 1.22 kg/MW-hr for the Si
furnace. More organics were emitted to
the atmosphere than particulates. Open
furnaces emitted as much as or more
organics to the atmosphere than did
semi-enclosed furnaces (kg/MW-hr
basis). Benzo(a)pyrene (BaP) was
detected in the emission to the
atmosphere from both semi-enclosed
furnaces and from the Si furnace. The
BaP content of solids from the pollution
reduction equipment of all furnaces,
except the open FeSi furnace, exceeded
0.006 /ug/g. Wastewater treatment
systems were very effective in
removing particulates, organics, and
BaP. No metals were extractable
(EPA/OSW extraction procedure) from
any of the solid waste in sufficiently
high concentration to classify the waste
as hazardous. Groundwater testing at
one site showed no leaching of metals
or organics.
This Project Summary was developed
by EPA's Industrial Environmental
Research Laboratory, Research
Triangle Park, NC, to announce key
findings of the research project that is
fully documented in a separate report of
the same title (see Project Report
ordering information at back).
Introduction
Ferroalloys are mixtures of iron and
alloying elements which, when added to
molten iron, give it the unique character
and properties associated with different
grades of steel. The ferroalloy used
amounts to about 2 percent of the weight
of the finished steel. CO-nch gases
generated by the smelting reactions
contain fume from the high temperature
region and also entrain finer material
contained in the furnace charge. In open
furnaces, which have no top cover, the
escaping gases burn on the surface of the
-------
charged material. Typically fumes from
these furnaces are collected in hoods,
above the furnace and the particulate
matter removed in a baghouse, but wet
scrubbers are used on a few furnaces For
covered (semi-enclosed and sealed)
furnaces, on which the covers fit tightly
over the furnace top and raw materials
are added either through annular spaces
around the electrodes (semi-enclosed) or
through sealed chutes (sealed), little
combustion normally occurs under the
cover. The space under the cover is
evacuated through a scrubber. The
cleaned CO-nch gas is then either flared
or used as a fuel. Fumes escaping around
the electrodes (secondary emissions) are
captured by a hood above the furnace and
discharged to the atmosphere either
directly or after cleaning in a baghouse.
The purpose of this study was to
determine the amounts of particulates
and organics generated by various
ferroalloy furnaces, and the amounts of
these materials escaping to the
environment. To this end, one furnace at
each of five plants was sampled. EPA
Methods 1-5 were used for stack
sampling wherever possible. The units
tested included two covered (semi-
enclosed) furnaces (one producing 50
percent FeSi and one producing standard
FeMn) and three open furnaces
(producing, respectively, Si, standard
FeMn, and 50 percent FeSi). (The term
"covered furnace" in this report refers to
a semi-enclosed furnace, unless
otherwise stated.)
Note that, in some cases, test runs
were made which do not meet all criteria
of EPA Methods 1-5. The project
emphasized obtaining the best data
possible without expending great sums of
money on construction of facilities (e.g.,
baghouse exhaust stacks) that comply
exactly with these criteria.
Also note that the report frequently
refers to New Source Performance
Standards (NSPS). No furnace in the
industry is currently subject to NSPS, and
few (if any) of the emissions control
facilities were built to meet NSPS. NSPS
are used to provide uniform comparisons,
rather than using individual state
regulations
Conclusions
In this project, environmental data
were collected for five furnaces. Each test
period lasted less than 5 days. It should be
recognized, therefore, that the data are
based on a limited test program and
represent the performance of individual
systems under specific operating
conditions. Frequently, especially when
comparisons are made, the data are used
to represent typical performance of
similar furnaces producing similar
products. These statements are more
properly indicators of trends: undue
significance should not be attached to
numerical ratios.
An effort has been made to use
sufficiently specific language to prevent
mistaking data taken before control
devices for actual emissions to the
environment. The term "generated" refers
to materials entering a collection or
control device. The term "emitted" refers
only to material actually entering the
atmosphere.
1. The total amount of particulates
generated by the furnaces per MW-
hr of furnace power input increases
for different alloys in the order
FeMn, FeSi, and Si in the
approximate ratios of 1:2:9,
respectively. Based on the limited
test data, there appears to be little
significant difference in the
particulate generation rates of open
and covered furnaces producing
similar products.
2. Particulate emissions to the
atmosphere from the Si furnace,
and the covered furnaces
producing FeMn and FeSi exceeded
the NSPS. NSPS do not apply to
these furnaces and are used here
only as a reference.mark for
discussion. The data indicate that
similar new furnaces could achieve
NSPS compliance with similar well
designed and operated control
equipment.
3. The test data presented in this
report indicate that, in general, the
ferroalloy industry does a good job
in preventing discharge of
generated pollutants into the
environment. However, despite
efforts by at least one company,
further development is needed to
find a solution to the rapid failure of
the bags in baghouses handling
fume from Si metal production.
These failures are not only
expensive, in both time and money
expended, but also result in periods
of increased particulate emissions
to the atmosphere. This product
line is the largest, in terms of
installed MW capacity (but not
product tonnage), in the industry;
baghouses are the common
method of fume control
4. Current wording of the NSPS leads
to some confusion as to what the
applicable reference mark should
be for the FeSi furnaces tested. In
the NSPS, FeSi is defined as an
alloy meeting a certain ASTM
designation and containing 50
percent or more Si. Silvery iron is
likewise defined as an alloy of Fe
and Si containing less than 30
percent Si. There is no definition of
or reference to a FeSi with Si
content between 30 and 50 percent
or to products not meeting all
aspects of the ASTM designations.
For FeSi, the NSPS value is 0.45
kg/MW-hr; for silvery iron, 0.23
kg/MW-hr. The two FeSi furnaces
tested in this study were producing
50 percent grade FeSi; however,
the actual Si content was low, on
the order of 47 percent. This is
within the ASTM designation
A100-69 grade E (47-51 percent);
however, it does not meet either the
50-percent-or-greater or the 30-
percent-or-less criterion of the
NSPS. Thus, it seemed neither
NSPS value was applicable as a
reference mark. For this study, it
was decided that the lower NSPS
value (0.23 kg/MW-hr) would be
used as a reference.
5. The amount of organics generated
per MW-hr of furnace power
increases in the following order of
furnace and product types: covered
FeMn with substantial undercover
combustion, open FeSi, covered
FeSi with little undercover
combustion, open FeMn, and open
Si. The data suggest that organic
generation rates are significantly
lower in covered furnaces with
substantial undercover combustion
than covered furnaces in which this
does not occur, and may be as low
as or lower than organic generation
rates of open furnaces producing
equivalent products.
6. The data indicate that open
furnaces emit as much as or more
organics to the atmosphere than do
covered furnaces producing
equivalent products. A major
reason for this is that the scrubbers
used on covered furnaces remove a
greater percentage of the organics
from gas from covered furnaces
than baghouses remove from gas
from open furnaces. No testing was
done to determine if scrubbers
would remove a greater percentage
of organics from gas from open
furnaces than is achievable with
baghouses. In general, organic
-------
emissions exceeded inorganic
paniculate emissions.
7 The percentage capture of organics
by baghouses seems to decrease
with increasing operating
temperature of the baghouse. The
industry uses several methods (e.g ,
introduction of dilution air, heat
exchangers, water sprays) to
reduce gas temperatures to protect
baghouse components. Further
reductions in gas temperatures
would be expensive, possibly
impractical for retrofit application,
and may still not reduce organics
significantly.
8. The average total amount of BaP
generated per MW-hr of furnace
power increases in the following
order of furnace type and product:
open FeMn, none detected in any
test; open FeSi, detected in two of
five tests; covered FeMn (but with
substantial undercover combustion),
detected in all tests; open Si,
detected in four of six tests; and
covered FeSi, detected in all tests.
Since the data indicate that the
tightly covered FeSi furnace
generated 4 g/MW-hr, about 180
times as much BaP as did the next
highest ranking furnace, it appears
that tightly covered furnaces would
generate substantially more BaP
than other furnace types producing
equivalent products.
9. BaP was detected in the emissions
to atmosphere from only three of
the five furnaces tested: Si metal,
detected in one of three tests;
covered FeSi, detected in all tests,
and covered FeMn, detected in two
of five tests. Emission rates
decrease in the order given. The
mass of BaP emitted was less than
0.01 g/MW-hr and less than 0.25
g/hr for each furnace.
10. The average organic content of
solid wastes from the pollution
control systems of the covered FeSi
and FeMn furnaces was 12 and 14
percent, respectively, significantly
greater than found for solids from
the open furnaces—0.9 percent
from Si metal and less than 0.2
percent for FeSi and FeMn. In
addition, BaP comprised about 0.4
percent of the organics found on
solids from the covered FeSi furnace,
but less than 0.1 percent of the
organics on solids from the other
four furnaces. Concentration values
found were 0.18,0.78,9.0, and 480
//g/g for Si metal, open FeMn,
covered FeMn, and covered FeSi,
respectively. These data suggest
that solids from open furnaces are
less likely to contain hazardous
amounts of organics, including
BaP, than are solids from covered
furnaces.
11. Wastewater entering the water
treatment system at the two plants
tested that use covered furnaces
contains a wide variety of EPA
designated priority pollutants. More
compounds and higher concentra-
tions were found at the plant using
the tightly covered FeSi furnace.
The treated effluent contained low
concentrations of all detected
priority pollutants, including metals.
No carcinogenic polynuclear aroma-
tic hydrocarbons were detected in
the effluents. The data indicate that
simply removing suspended solids
from the untreated water would
result in 80-90 percent reduction in
discharged organics and 100 per-
cent removal of BaP. These data
indicated that the wastewater
treatment methods in use are
effective in controlling the discharge
of waterborne pollutants.
12. Monochlorinated biphenyl was
detected at <20/ug/L in the untreated
wastewater at the Ashtabula plant
(covered FeSi furnace). This com-
pound, as well as several pesticides
also found, may have been in the
plant intake water from Lake Erie.
Several additional polychlorinated
biphenyls (PCBs), each at less than
2/ug/L, were detected in the treated
discharge. These data suggest the
possibility that the plants' use of a
state-of-the-art technology (chlori-
nation) for phenol and cyanide
destruction results in PCBformation
in this water. The analytical tech-
nique used, however, leaves some
question about the PCB identifica-
tion.
13. The limited test data suggest that
solid wastes removed from the
wastewater treatment system and
placed in on-site landfills would be
about 1 x 10~2 and 8 x 10~5 percent
BaP for the Ashtabula and Marietta
plants, respectively. Analysis of one
pair of up- and down-gradients
groundwater samples around the
landfill-wastewater pond system at
the Ashtabula plant indicates that
neither organic matter, including
BaP, nor heavy metals leach or
percolate into the underground
aquifer. These limited data suggest
that this method of disposal does
not present a clear and present
danger to underground aquifers.
14 No metals were teachable, following
the EPA/OSW extraction procedure,
from any of the solid wastes
(including baghouse dusts, scrubber
sludges, and wastewater treatment
plant sludge) in sufficiently high
concentrations to result in classify-
ing the wastes as hazardous. BaP
concentration in solid wastes from
air pollution systems on all furnaces
tested except the open FeSi furnace
exceeded 0.006 fjg/g. Extraction in
acidic solution by the EPA/OSW
procedure indicated that BaP does
not leach from the solids.
C. W. Westbrook is with Research Triangle Institute, Research Triangle Park, NC
27709.
Robert C. McCrillis is the EPA Project Officer (see below).
The complete report, entitled "Multimedia Environmental Assessment of Electric
Submerged Arc Furnaces Producing Ferroalloys," (Order No. PB 83-262 063; Cost:
$22.00, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
irUS GOVERNMENT PRINTING OFFICE 1984-759-015/7255
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
BULK RATE
U.S. POSTAGE
PAID
Cincinnati, Ohio
Permit No. G35
Official Business
Penalty for Private Use $300
------- |