vvEPA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Cincinnati OH 45268
EPA-600/2-78-065a
March 1978
Research and Development
Trace Element Study
at a Primary Copper
Smelter
Volume I
-------
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
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-78-065a
March 1978
TRACE ELEMENT STUDY AT
A PRIMARY COPPER SMELTER
Volume I
by
Klaus Schwitzgebel, Richard T. Coleinan,
Robert V. Collins, Robert M. Mann,
and Carol M. Thompson
Radian Corporation
8500 Shoal Creek Blvd.
Austin, Texas 78766
Contract No. 68-01-4136
Project Officers
Margaret J. Stasikowski and John 0* Burckle
Metals and Inorganic Chemistry Branch
Industrial Environmental Research Laboratory
Cincinnati, Ohio 45268
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Industrial Environmental Research
Laboratory, Cincinnati* 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.
ii
-------
FOREWORD
When energy and material resources are extracted, processed,
converted, and used, the related pollutional impacts on our environment
and even on our health often require that new and increasingly more
efficient pollution control methods be used. The Industrial Environmental
Research Laboratory - Cincinnati (lERL-Ci) assists in developing and
demonstrating new and improved methodologies that will meet these needs
both efficiently and economically.
This report presents the findings of an investigation of air
pollutant emissions from the reverberatory furnace pollution control
system at a primary copper smelter. The study was performed to assess
the degree of particulate emissions control and control problems
associated with the application of electrostatic precipitators in the
nonferrous metals production industry. The results are being used within
the Agency's Office of Research and Development as part of a larger
effort to define the potential environmental impact of emissions from
this industry segment and the need for improved controls. The findings
will also be useful to other Agency components and the industry in
dealing with environmental control problems. The Metals and Inorganic
Chemicals Branch of the Industrial Pollution Control Division should be
contacted for any additional information desired concerning this program.
David G. Stephan
Director
Industrial Environmental Research Laboratory
Cincinnati
ill
-------
ABSTRACT
The present study was sponsored by the Industrial -Environmental
Research Laboratory of EPA, Cincinnati. It describes results obtained by
sampling at a primary copper smelter with a production rate of 200 tons
Cu/day. Concentrate, matte and slag streams of the reverberatory furnace
were analyzed. The main effort of the study, however, focused on the elec-
trostatic precipitator (ESP) controlling particulate emissions from the
reverberatory furnace. The main findings of the study are presented below.
SMELTER MINOR ELEMENT FLOW
1) Titanium, potassium, magnesium, and sodium enter the smelter
as major components besides the elements of importance in the
smelting operation itself (Cu, Fe, Si, Ca, and Al).
2) Minor elements of environmental concern are:
arsenic,
cadmium,
molybdenum,
lead,
antimony,
selenium,
zinc, and
fluorine.
3) The rest of the elements surveyed enter in the low ppm
concentration range or were not detected by spark source mass
spectrometry analysis.
4) Molybdenum entering the reverberatory furnace with the
concentrate is almost completely discharged in the reverber-
atory furnace slag.
5) Nearly 50 percent of the selenium and 30 percent of the
fluorine are discharged together with the reverberatory
furnace off-gases.
6) Nearly all of the fluorine escapes in the gaseous state.
IV
-------
ARSENIC SAMPLING
The arsenic flows found by Radian have been compared in this report to
tests made by smelter personnel during the periods of February 28 through
March 4, 1977. In addition, integral arsenic balances were established for
the months of January and March 1977 from the plant data. The findings are
incorporated in this report.
1) Approximately 50 percent of the arsenic entering the
smelter leaves in the reverberatory furnace off-gas based
on both plant and Radian sampling data.
2) Data gathered during this sampling study indicate that
as much as 90 percent of the arsenic entering the reverber-
atory furnace ESP may leave in the off-gas.
3) Arsenic and selenium escaping the electrostatic pre-
cipitator are partly in the vapor state.
4) The arsenic concentration in the reverberatory furnace
feed varies within a few days from 700 ppm to 7000 ppm. This
variation is not only a function of the arsenic concentration
in the ore, but also depends on the amount of electrostatic
precipitator dust recycled.
5) The following arsenic rates were determined at the
ESP outlet:
140 Ib/hr
76 Ib/hr
54 Ib/hr
45 Ib/hr
51 Ib/hr
45 Ib/hr
40 Ib/hr
25 Ib/hr
34 Ib/hr
49 Ib/hr
July 11, 1976 -
July 13, 1976 -
July 13, 1976 -
July 14, 1976 -
July 14, 1976 -
January, 1977 -
March, 1977
March 1, 1977
March 2, 1977
March 3, 1977
Radian, direct measurement
Radian, direct measurement
Radian, direct measurement
Radian, direct measurement
Radian, direct measurement
- Smelter, balance difference
- Smelter, balance difference
- Smelter, direct measurement
- Smelter, direct measurement
- Smelter, direct measurement
6) The arsenic detected was present as arsenolite,
was determined by X-ray diffraction.
This
-------
ESP PERFORMANCE
1) ESP temperature (600°F) and gas flow rate (160,000 acfm)
correspond to design parameters.
2) ESP inlet grain loading determined at duct temperature is
0.60 grains/scf, and the outlet grain loading is 0.020 grains/scf.
The overall particulate collection efficiency is calculated to
be 96.7%. Design efficiency was 96.8%.
3) The amount of condensibles between duct temperature 600°F
and 250°F (EPA recommended) accounts for material corresponding
in mass to an additional 1.60 gr/scf at the inlet and 1.0 gr/scf
at the outlet.
4) The element flow at the reverberatory furnace ESP outlet
measured July 11, 1976 is as follows:
Sulfur 2260 Ibs/hr
Arsenic 140 Ibs/hr
Fluorine 7.8 Ibs/hr
Copper 1.0 Ibs/hr
Selenium .97 Ibs/hr
Antimony .33 Ibs/hr
Molybdenum .2 Ibs/hr
Nickel .1 Ibs/hr
Lead .1 Ibs/hr
Zinc .1 Ibs/hr
5) Good material balance closure around the ESP was established
for:
Arsenic Nickel
Beryllium Sulfur
Chromium Antimony
Copper Selenium
Fluorine Vanadium
Mercury Zinc
Molybdenum
The balance for cadmium is marginal. The balance for barium and
lead did not close due to precipitation of insoluble sulfates in
the collection device.
6) Cadmium, lead, and zinc entering the smelter are transported
with the matte to the converters. Cadmium most likely leaves the
smelting circuit with the converter off-gases. Some of the zinc
and lead will be volatilized in the converter. Some will also
be returned to the reverb in the converter slag.
VI
-------
CONTENTS
Section Page
Foreword ill
Abstract v
Figures ix
Tables x
1 Introduction 1
2 Plant Description 4
Material Flow 4
Reverberatory Furnace. .,.,.., , , 10
Reverberatory Furnace ESP. 12
Plant Operation During Sampling 13
3 Sampling Techniques... . f 15
Wet Electrostatic Precipitator 15
Integral WEP Sampling Train. 16
Vapor Phase Sampling Train. T . . . 18
Arsenic Sampling Train 19
In-Stack (600°F)/Out-of-Stack (250°F) Grain Loadings 21
Cyclone Sampling Train 22
4 Results and Discussion of Results 23
Reverberatory Furnace Feed 23
Reverberatory Furnace Slag 27
Reverberatory Furnace Matte 30
Reverberatory Furnace Element Balance (July 12-14, 1976).. 33
Arsenic Material Balance Around the Smelter
(Smelter Data) 36
Arsenic Flow at ESP Inlet and Outlet (Radian Data) 39
Collection Efficiency of the Electrostatic Precipitator... 42
Electrostatic Precipitator Material Balance
(July 11, 1976) , 45
Identification of Major Crystalline Species 47
TABLE OF CONVERSION FACTORS 50
vii
-------
FIGURES
Number Page
2-1 Schematic Diagram of Primary Copper Smelter 5
2-2 Reverberatory Furnace Feed Preparation 6
2-3 Reverberatory Furnace Flow Diagram , 7
2-4 Converters and Anode Furnace 9
2-5 Converter Dust Recovery and HaSCH Plant 11
2-6 Electrostatic Precipitator and Dust Handling System 14
3-1 Wet Electrostatic Precipitator 16
3-2 Schematic of the Integral WEP Sampling Train 17
3-3 Schematic of the Vapor-Phase Trace Element Sampling Train.... 18
3-4 Arsenic Sampling Train (Modified EPA-5 Train) 19
3-5 In-Stack/Out-of-Stack Sampling Train 21
3-6 Schematic of the Cyclone Sampling Train 22
4-1 Boundary of Arsenic Balance 37
4-2 Traverse Points Layout for Reverberatory Electrostatic
Precipitator Inlet 40
4-3 Traverse Points Layout for Reverberatory Electrostatic
Precipitator Outlet , . 41
viii
-------
TABLES
Number Page
1-1 A Summary of the Sampling Report (July 7 through
July 16, 1976) 2
3-1 Implnger Solutions Used for ESP Material Balance 17
3-2 Impinger Solutions Used for Arsenic Sampling 20
4-1 Element Concentration and Flows in the Reverberatory
Furnace Feed 23
4-2 Survey Analysis of the Reverberatory Furnace Feed by SSMS
(Samples Collected July 12, 1976 through July 14, 1976).,. 25
4-3 Variation of Arsenic Concentration in Reverberatory
Furnace Feed (Data by Smelter Personnel) 26
4-4 Reverberatory Furnace Slag Composition and Element Flow
Rates (July 12-14, 1976) 27
4-5 Survey Analysis of Reverberatory Furnace Slag by SSMS
(Samples Collected July 12, 1976 through July 14, 1976)... 28
4-6 Arsenic Concentration in the Reverberatory Slag During
March 1, 1977 through March 3, 1977 (Data by Smelter
Personnel) , 29
4-7 Concentration of Selected Elements in Matte (Composite
Sample July 12 through July 14, 1976) 30
4-8 Survey Analysis of Matte by Spark Source Mass Spectrometry
(Composite Sample for July 12 through July 14, 1976) 31
4-9 Fluctuation of the Arsenic Concentration in the Matte for
the Period March 1 through March 3, 1977 (Data determined
by Smelter Personnel) 32
4-10 Element Flow Rates in the Feed and Discharge Streams of the
Reverberatory Furnace (Ibs/hr) (Radian Data) 34
ix
-------
TABLES (Cont'd)
Number Page
4-11 Arsenic Concentration in the Waste Heat Boiler Dust,
Reverb ESP Dust, Converter Dust and Acid Plant Purge
Water (Data by Plant Personnel) 35
4-12 Arsenic Concentrations in Incoming and Outgoing Streams of
the Reverberatory Furnace During the Month of January
1977 (Data by Plant Personnel) 36
4-13 Summary of Arsenic Emission Data from the Reverberatory
Electrostatic Precipitator (Radian Data) 40
4-14 Grain Loading Data Inlet , 43
4-15 Grain Loading Data Outlet 44
4-16 In-Duct/Out-of-Duct Grain Loading (gr/scf) (July 15, 1977)... 45
4-17 Elemental Flow Rates in ESP Inlet and Outlet Streams 46
4-18 Summary of Crystalline Species Identified by X-Ray
Diffraction., 49
x
-------
SECTION 1
INTRODUCTION
The primary copper smelter studied in this report started production
about 25 years ago. Gas cleaning facilities were added in the early seventies
to treat the off-gases from the converters. Waste heat boilers and electro-
static precipitators are currently used to generate steam and remove partic-
ulates from both reverberatory furnace and converter gas streams. The cleanup
and cooling of converter gases is achieved in a humidifier, cooler, wet
electrostatic precipitator arrangement followed by a contact sulfuric acid
plant.
The primary goal of the present study was to evaluate the performance
of the electrostatic precipitator collecting particulate matter escaping
the reverberatory furnace. Several sampling techniques were chosen to de-
termine:
• gas flow rate,
• gas composition,
• grain loading at the inlet duct,
• grain loading at the outlet duct,
• particle size distribution,
• electrical performance of the ESP,
• trace element material balance,
• trace element flow rates as vapor,
• the amount of material condensible between the gas
temperature of 600°F and the EPA recommended cool-off
point of 250°F, and
• arsenic emission rates using an EPA suggested sampling
device.
The sampling effort was conducted from July 8 through July 16, 1976. A
detailed sampling schedule is presented in Table 1-1.
-------
TABLE 1-1. A SUMMARY OF THE SAMPLING EFFORT
(JULY 7 THROUGH JULY 16, 1976)
Date
Location/Stream Sampled
Parameter
Technique
Evaluate Reverb ESP performance:
7-8 to 7-10
7-10
7-8 to 7-10
7-10
7-10
ESP Outlet
ESP Outlet
ESP Inlet
ESP Inlet
ESP Control Room
Complete a material balance around Reverb ESP:
7-11
7-11
7-11 to 7-13
ESP Outlet
ESP Inlet
ESP Dust
grain loading
particle size distribution
grain loading
particle size distribution
electrical performance
trace element flow rates
trace element flow rates
trace element flow rates
Complete an approximate material balance around the Reverb furnace:
7-13
7-12 to 7-14
7-11 to 7-13
7-12 to 7-14
7-12 to 7-14
ESP Outlet
Reverb feed
ESP Dust
Reverb Slag
Matte
trace element flow rates
trace element flow rates
trace element flow rates
trace element flow rates
trace element flow rates
Collect particulate by particle size for trace element analysis:
7-16 ESP Outlet particulate by size fraction
Collect vapor phase emissions:
7-16 ESP Outlet
trace element flow rates
as vapor
Determine amount of condensible material and SO emitted:
7-15
7-15
ESP Outlet
ESP Outlet
Determine arsenic emission rates:
7-13 to 7-14
7-13 to 7-14
ESP Outlet
ESP Inlet
condensed material
(between 600°F-250°F)
and SOa-SOa concentrations
condensed material
(between 600°F-250°F)
and S02- SOs concentrations
arsenic emission rate
arsenic emission rate
in-stack filter
Andersen cascade impactor (SRI)
instack filter
Brinks cascade impactor (SRI)
monitor operating parameters (SRI)
integral WEP
integral WEP
periodic grab sample
integral WEP
compositing slinger bin catches at the
end of each shift
periodic grab sample
periodic grab sample
periodic grab sample
3 out of stack cyclones in series
plus filter
filter at duct temperature followed
by impingers
EPA 5 train with in-stack filters
EPA 5 train with in-stack filters
modified EPA Method 5 train
modified EPA Method 5 train
-------
Plant management released composites of the following samples:
• concentrate (reverberatory furnace feed),
• matte, and
• reverberatory furnace slag.
In addition, all critical flow rates were released to Radian. A complete
material balance around the reverberatory furnace was still not possible
because compositions for the following streams are not known:
• converter slag,
• dust collected in the converter waste heat boiler,
balloon flue, and ESP, and
• material condensed on the surface of the reverberatory
furnace waste heat boiler.
The study was performed in parallel with an effort by Southern Research
Institute, Birmingham, Alabama (SRI). SRI determined the electrical charac-
teristics of the ESP during the sampling period and also determined the
particle size distribution at the inlet and outlet using Brink's impactors.
Details of these investigations can be found in SRI's report: "Performance
Evaluation of an Electrostatic Precipitator Installed on a Copper Reverbera-
tory Furnace," SORI-EAS-76-511, EPA Order No. CA-6-99-2980-J.
The important results and conclusions are summarized in this volume
of the report. Descriptions of the smelter, the reverberatory furnace, and
the electrostatic precipitator are presented in Section 2. Section 3 de-
scribes the sampling techniques used to perform each experiment listed in
Table 1-1.
Section 4 presents the data generated by this study. The results and
data limitations are also explained in this section. More detailed descrip-
tions of the sampling and analytical procedures are presented in Volume II
along with the actual sampling data.
-------
SECTION 2
PLANT DESCRIPTION
This section first presents a general process description of the smel-
ter studied. The reverberatory furnace and the hot electrostatic precipita-
tor are described next. The plant operation during the sampling program is
covered last.
MATERIAL FLOW
The primary copper smelter studied in this report started production
about 25 years ago. Gas cleaning facilities were added in the early seven-
ties to treat the off-gases from the converters. The copper smelting facil-
ity sampled can be divided into the following process steps (see Figure 2-1) :
• mining,
• concentrate preparation,
• reverberatory furnace feed preparation,
• matte production in the reverberatory furnace,
• production of blister copper in the converters and
copper refining in the anode furnace, and
• cleaning of the converter off-gases with sulfuric
acid production.
An open pit mine feeds the smelter. Typical copper concentrations in the
ore range from 0.5 to 1.0%. All ore with less than 0.3% copper content is
rejected as waste rock.
The process flow is depicted in Figures 2-2 and 2-3. Several trains
transport the ore from the mine to the crushing and grinding unit. The first
step is the reduction of the larger boulders in a gyratory crusher. Contin-
uous reduction of the rock size is achieved first in cone crushers than
finally in wet ball mills.
The smelter uses a conventional flotation process. Water and flotation
reagents are added to the ball mill effluent stream and the slurry is aerated.
Copper bearing particles adhere to the ascending air bubbles. The "froth" is
-------
Reformed Gas
Anode
Furnace
Anode
Copper
/ Reverb Slag \._
\ to Slag Dumpy'"
Blister
Copper
Slag
Concentrate and
Flux Material
(
Natural Gas, Diesel Oil
and/or No. 6 Fuel Oil
One Reverberatory
Furnace
Copper
Matte
Three Converters,
One Great Falls
Oxidizing Furnace
93% H2SO/(
Tail Gas
Stack
HumidifyingX
Tower /
Slowdown /
Figure 2-1. Schematic Diagram of Primary Copper Smelter.
-------
Ore from
Open Pit
Mine
Water
and
Flotation Reagents
Calcium Carbonate
Quartz
FEED STREAMS
Refuse to
Tailings
Pile
\
Green Feed
to
Reverb Furnace
PRODUCT STREAMS
Figure 2-2. Reverberatory Furnace Feed Preparation.
-6-
-------
I Reverb Furnace
I Feed
Natural Gas,
Diesel Oil and/c
6 Fuel
/Natural Gas, A
(Diesel Oil and/orV-
VNo. 6 Fuel OiV
FEED STREAMS
Reverb
Furance
Converter &
Reverb Dust
Two Waste
Heat Boilers
Dry Electrostatic
Precipitator
PROCESSING STEPS
issions to
Atmosphere
PRODUCT STREAMS
Figure 2-3. Reverberatory Furnace Flow Diagram.
-7-
-------
skimmed from the surface. Solid-liquid separation is achieved next by vacuum
filtration. The aqueous phase is recycled, 'and the concentrate containing
20% or more copper goes to storage. The particulates low in copper ore sink
to the bottom in the flotation step and are pumped as a slurry to the tail-
ings pond. Again, supernatant water from this pond is recycled to the ore
enrichment facility. The ore and concentrate are analyzed for copper, and
major constituents. Calcium carbonate and quartz are mixed with the concen-
trate to produce an optimum reverberatory furnace feed.
The major steps in operating the reverberatory furnace are shown in
Figure 2-3. The purpose of this furnace is to reduce the sulfur and iron
in the concentrate to the point where FeS and CuaS are present in approxi-
mately stoichiometric amounts. This mixture is called matte. Excess iron
sulfide is converted in an exothermic reaction to FeO and S02. Iron oxide
combines with silica to form a slag of lower density than the matte. The
slag layer is removed from the reverberatory furnace and transported in a
slag train to the slag dumping site. The off-gases contain sulfur dioxide,
particulate matter and fumes of metal compounds. Heat is recovered in two
waste heat boilers.
Particulates are removed in two waste heat boilers and in a hot electro-
static precipitator. These collected dusts are recycled to the reverberatory
furnace feed preparation area. The gas leaving the reverb ESP is discharged
to the atmosphere through the reverberatory furnace stack.
Streams entering the reverberatory furnace in addition to the concen-
trate (reverb furnace feed) are (1) converter slag, (2) converter dusts
collected in both the balloon flue and the ESP treating the converter off-
gases, and (3) reverb ESP and reverb waste heat boiler dusts.
Three Peirce-Smith converters process the matte produced in the rever-
beratory furnace. The schematic of the material flow is shown in Figure 2-4.
Charging of the matte to the converters is a batch process using ladles
transported by an overhead crane. In the converter, iron sulfide is oxidized
to FeO and SOa. Quartz addition to the converter binds the FeO in the form
of a silicate slag (4% Cu) which is recycled to the reverb. Again, the oper-
ation is discontinuous and is carried out using ladles and the overhead crane.
"White metal" consisting of CuaS is left after the removal of the iron. At
this stage, the copper blowing cycle starts. Copper sulfide is converted to
a 98% pure copper called "blister copper". Cold copper-containing scrap is
added periodically to absorb the heat of the exothermic reaction.
The blister copper is transported to the anode furnace using ladles.
Air blowing serves to remove the sulfur left in the blister copper. Copper
containing CuaO as an impurity is the result of this processing step. The
-8-
-------
1
1
f „ A
1 Matte 1 *.
1
(Silica 1 !-«.
y i
/ >\ !
1 ScrsD \- — -•- bt
I p j i
i
( Ur 1 '»
1 ' j I
1
|
!
1
1
1
1
1
1
1
j ^
\
f .. ^ '
I Air J r
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
FEED STREAMS |
Three
Peirce-
Smith
Converters
'
M
91
1 ?
« en
1 2
J5 ^
- 3
CO
Furnace
Anode
Casting
Wheel
PROCESSING STEPS
i
1
I
/>
Converter Slag
""N. to Keverb
\_ , j
/ ^
~~ \ And Dust
V X
/ N
M f Anode Furnare f!asps
\To Atmosphere
/
/ \
i _ / Anodes to Electrolvtic
\Refining
/
PRODUCT STREAMS
Figure 2-4. Converters and Anode Furnace.
-9-
-------
copper oxide is then reduced with reformed gas. The resulting copper with
a purity of 99+% is ready for the anode pouring step which is done once a
day. An anode casting wheel is used. The anodes weigh about 750 Ibs apiece
and are shipped for electrolytic refining. Output of the plant is 180-200
tons of anode copper a day.
Sulfur dioxide off-gas and dust and fumes composed of metal compounds
are the result of the CuaS and FeS oxidation in the converters. The converter
dust collection and gas cleaning steps are shown in Figure 2-5. The SOa con-
centration of the converter off-gases varies between 4 and 8 percent (v) and
as such is amenable to the manufacturing of sulfuric acid using the contact
process. The off-gases from the converters are first cooled in waste heat
boilers. Particulate removal is achieved in an electrostatic precipitator.
The dust collected in the balloon flue and the electrostatic precipitator is
recycled to the reverberatory furnace. Dust from the waste heat boilers is
fed back to the converters.
Gas cooling and cleaning is the first step in sulfuric acid production.
A humidifier, a cooler and a wet electrostatic precipitator are used to cool
and clean the gas. The humidifying tower has a blowdown stream consisting of
dilute sulfuric acid containing the particulate matter escaping the ESP, as
well as condensable fumes. The HaSOi* plant consists of drying towers, a
vanadium pentoxide catalytic bed, and SOs absorption towers. Heat exchangers
remove the heat of the exothermic SOa oxidation process. The product of the
sulfuric acid plant is a 93% HaSOi). acid stream and a tail gas with low
(presently unknown) concentrations of contaminants.
REVERBERATORY FURNACE
The single reverberatory furnace, 30 feet wide and 100 feet long
(inside dimensions), fired with burners which are capable of burning natural
gas, diesel oil and/or No. 6 fuel oil, processes concentrates from the milling
and flotation plant. The concentrates, which comprise over 90% of the furnace
charge are received from the filter plant and dryer by conveyor or from stock
piles by truck. Each charge is weighed on scales and is carried by overhead
crane to a charge point above and to the side of the furnace. Six charge
ports are provided, three on each side of the furnace. A concrete bin of
approximately 1600 tons capacity is used for surge storage of concentrates.
When necessary, the concentrates are reclaimed by clamshell from the storage
bin.
The bath smelting reverberatory furnace was designed for wet (green
charge) smelting. The crucible of the furnace is constructed in sections of
rammed periclase. The top and sides of the furnace crucible are cooled by
10
-------
Converter Gas
&
Dust
Makeup
Water
FEED STREAMS
\ L
/
|
t i
r
ik
p
Three Waste Heat
Boilers
i
r
Balloon Flue
Collection System
1
^/ Diisl- fn \
\ Converters /
1 r. ^/ Dust to \
l k \ Reverb /
Electrostatic
Precipitator
i
r
Gas Cooling
&
Cleaning
i
r
Contact
Sulfuric
Acid
Plant
^ Humidifying Tower\
^\ Slowdown /
^/ cn% Snlfnrir \
"A Acid /
< Clean Gas To \
Atmosphere y
PROCESSING STEPS PRODUCT STREAMS
Figure 2-5. Converter Dust Recovery and HjSOi). Plant.
11
-------
copper water jackets. The working bottom of the furnace is magnetite
accretions fused in place. The bath smelting furnace requires reaction
between the new charge and the liquid bath to cause spreading of the new
charge. When calcines are smelted, charging them through an inclined pipe
imparts sufficient velocity to allow a reaction with the liquid bath to
cause spreading of the new charge. This gravity charging method cannot be
used with wet concentrates, however, since they will not flow through an
inclined pipe as will calcines. Considerable research was necessary to
develop proper charging equipment. It is necessary to charge the wet con-
centrates to the furnace at a considerable velocity and at a relatively flat
angle. This enables the wet concentrate to penetrate the liquid bath suffi-
ciently to start the spreading reaction and prevents the concentrate from
piling up in one place.
A high-speed belt slinger was developed for this operation and has
performed well. Six slingers are provided, three on each side of the furnace.
Each has a port through the sidewall covered with a vertically movable door
when not in use. When charging begins, the door is raised by an air cylinder,
the slinger started, and the concentrates drop onto the slinger from the
charge can overhead. The slinger angle is adjustable both horizontally and
vertically. Belt speeds are variable, allowing the charging rate to range
from 1.5 to 2 tons/min for each slinger. Furnace charging is done by one
man and is controlled by visual inspection.
The reverb is a sprung silica arch type. The brickwork is maintained
by silica slurry hot patching. The silicious material of over 90% Si02 is
ground wet in a ball mill and pumped to a storage tank above the furnace
from which it is blown and sprayed onto the interior surface of the brickwork.
A curtain damper has been used successfully in the furnace and its use has
resulted in better working conditions and decreased maintenance expense in
and around the boiler uptakes and flues. The curtain damper sections are
made of cast refractory supported by loops of water-cooled steel tubing and
are maintained by slurry patching.
REVERBERATOR! FURNACE ESP
Gases from the reverberatory furnace enter two waste heat boilers in
parallel. The boilers cool the gas to approximately 600°F and remove some
particulate and condensable material. Steam produced by the boilers is used
to generate either electricity or low-pressure air for the converters. Gases
exiting the waste heat boilers enter a plenum chamber for mixing prior to
treatment in an electrostatic precipitator (ESP).
The reverberatory furnace electrostatic precipitator consists of two
independent horizontal parallel units designed to handle a total 160,000 acfm
(600°F and 13.8 psia). Gas treatment time is 6 seconds; pressure drop across
12
-------
the precipitator is 0.5 inches w.c. maximum. Collecting surfaces are plates;
the discharge electrodes are spring steel wires. The transformer-rectifiers
are silicon full wave for 45KV (DC) average; there is an automatic voltage
control system to maintain the optimum precipitator operating efficiency.
Dust collected is transported via screw conveyor and chain drag to a storage
bin from which it is pneumatically conveyed to the charge mixing area for the
reverberatory furnace by a fluid flow automatic pump. This pump can handle
20,000 pounds per hour when supplied with 280 scfm compressed air at 30 psig.
The electrostatic precipitator and dust handling system are shown schemati-
cally in Figure 2-6.
An induced draft fan was installed to draw the reverberatory gases from
the reverberatory furnace through the waste heat boilers, the flue system,
and the electrostatic precipitator. This fan can handle 160,000 acfm @ 600°F
with a suction of -1.75" H20 and a discharge of 3.25" H20. The gases are
discharged through a 360-foot concrete stack.
PLANT OPERATION DURING SAMPLING
The reverberatory furnace operated continuously during the entire
sampling program. One upset in the ESP operation was recorded on July 12.
This upset was attributed to wet concentrate charged to the furnace. The
concentrate dryer was repaired that day and no further upsets were recorded.
High dust loadings were noted during the periods when the reverberatory
furnace was being charged. This, however, is normal operation.
13
-------
ESP Outlet
Sampling Ports
_n
Electrostatic
Precipitator
n n n n
Dust Hopper
ESP Inlet
Sampling Ports
Figure 2-6. Electrostatic Precipitator and Dust Handling System.
-------
SECTION 3
SAMPLING TECHNIQUES
Special sampling trains had to be built to meet the objectives set
forward in the introduction. This section is intended to give a succinct
survey of what sampling techniques were used. The equipment is described in
greater detail in Volume II. The descriptions of the following sampling
techniques are presented below:
• Wet Electrostatic Precipitator (WEP)
• Integral WEP Sampling Train
• Vapor Sampling Train
• Arsenic Sampling Train
• In/Out Stack Grain Loading Train
• Cyclone Sampling Train
All sampling was done isokinetically. Strict adherence to the EPA
procedures as set forward in Method 1 was not possible at the ESP inlet due
to the duct configuration and the location of the sampling ports. Sampling
at the outlet was more ideal. Bends in the duct occurred approximately 4.5
stack diameters upstream and two stack diameters downstream of the sampling
port location. In addition, the duct ran vertically thus avoiding stratifi-
cation of particulate matter. The duct configuration at the inlet and outlet
of the hot electrostatic precipitator as well as the traversing steps are
described in detail in Volume II.
WET ELECTROSTATIC PRECIPITATOR
The wet electrostatic precipitator (WEP) used to collect particulates
and acid mist is described first. It is part of the Integral WEP Sampling
Train designed to collect solids, mist and gaseous components from a gas
stream and is shown in Figure 3-1. An electrolyte is circulated through a
round bottom flask and a vertical glass cylinder by a peristaltic pump. The
walls of the cylinder are wetted by the falling film of electrolyte. A thin
platinum wire is suspended in the center of the glass cylinder. A high volt-
age of 12-15 KV-DC causes a corona discharge at the center electrode.
15
-------
The gas entering the WEP is first scrubbed and cooled in the round
bottom flask. Particulates and mist not retained here are electrically
charged in the glass cylinder, collected in the falling film and washed into
the electrolyte reservoir. This sampling device does not clog like a filter
or a thimble and no analytical background corrections are necessary since no
extraneous material is introduced as is the case with filters.
Platinum Electrode
Falling Film of
"Electrolyte
-WEP Body
Peristaltic
P-ump
Circulating
Electrolyte
Reservoir
Figure 3-1. Wet Electrostatic Precipitator.
INTEGRAL WEP SAMPLING TRAIN
The wet electrostatic precipitator was integrated into the sampling
device shown in Figure 3-2. The probe consisted of a pyrex nozzle and was
pyrex lined. Teflon tubing was used to connect it with the WEP. All the
lines were rinsed after sampling. The WEP was followed with eight impingers
in an ice bath. The charge of the impingers is summarized in Table 3-1. A
pump and a dry gas meter completed the assembly. Samples were collected on
16
-------
July 11, 1976 at the inlet and outlet duct of the electrostatic precipitator.
The samples were used to establish an element balance around the ESP.
Acid Impingers
Caustic Impingers
HydroRcnpcroxide
Impingor
Pyrex Pyrex Lined Probe
Nozzle
Teflon
Tubing
Wet Electrostatic
Precipitator
T n
Ice Bath
\/
Dry Silca Gel
Impingers Impinger
Fine
Adjustment Valve
Coarse
Adjustment Valve
Pump
Figure 3-2. Schematic of the Integral WEP Sampling Train.
TABLE 3-1. IMPINGER SOLUTIONS USED FOR ESP MATERIAL BALANCE
Impinger
Number
Solutions
1, 2
3, 7
4, 5
6
8
1:1:1 sulfuric acid, nitric acid, deionized
water in a Smith-Greenburg impinger
dry modified Smith-Greenburg impingers
20% potassium hydroxide in a Smith-Greenburg
impinger
hydrogen peroxide in a Smith-Greenburg impinger
preweighed silica gel in a modified Smith-
Greenburg impinger
-17-
-------
An additional sample was collected on July 13, 1977 with this train.
This sample was used to study the element balance around the reverberatory
furnace.
The sampling train shown in Figure 3-2 collects an integral sample
consisting of particulates, mist and gaseous species which are collected
in the WEP electrolyte and the impinger train.
VAPOR PHASE SAMPLING TRAIN
The temperature of matte and slag in a reverberatory furnace ranges
typically from 1950-2300°F. Certain chemical compounds are vaporized at
this temperature. The temperature of the flue gas is lowered to 600°F in
waste heat boilers.
A vapor train as shown in Figure 3-3 was built to determine which com-
pounds are still in the flue gas as vapors at this temperature. This train
was used at the outlet of the electrostatic precipitator on July 16, 1977 to
collect a vapor sample. A pyrex nozzle was attached to a pyrex-lined, heat
traced probe. The probe was connected to an out-of-stack oven kept at duct
temperature. The oven housed a cyclone followed by a filter. The impinger
train following the out-of-stack oven was the same as that used for the in-
tegral WEP train. The cyclone-filter arrangement retained particulates. The
impinger train collected species condensable between 600°F and 32°F.
Acid Itnplngers
Caustic Itnnlngern
\ Uydrogenperoxl
'.\ ~\r:;:
5 6 7 n
r r " i
: .' ' - ':
° ••'£?;•''''•*•"•" *"-•,
Dry Sllcn Cel
Ilnplngers ImpiiiRer
Fine
Ad)listment Valve
Pump
Figure 3-3. Schematic of the Vapor-Phase Trace Element Sampling Train.
-18-
-------
ARSENIC SAMPLING TRAIN
The arsenic sampling train shown in Figure 3-4 was used to collect
three integral arsenic samples each at the inlet and outlet. The inlet
samples were obtained on July 13 (0930-1225) and July 14 (0555-0907; 1132-
1505). Corresponding times for the outlet samples were July 13 (0951-1336)
and July 14 (0610-0855; 1130-1415).
The content of the impinger solutions are described in Table 3-2.
Two sampling ports each at the inlet west duct and inlet east duct were
traversed. Strict adherence to EPA Method 1 was followed for sampling at
the outlet. This duct was traversed from the north and west port. A total
of 18 points on each traverse was covered. Wall condensation caused by
cooling the gas from 600°F to 250°F was observed. The deposit having a
reddish appearance could not be completely removed causing the analytical
results to be potentially low. The samples collected with this device were
analyzed by Battelle, Columbus Laboratories, Columbus, Ohio.
-CYCLONE
STACK TEMPERATURE T.C.
IMPINGERS
°r
< Mi
VACUUM
LINE
ORIFICE AP
MAGNEHf LIC GAGE ^
DRY TgST METER
VACUUM
GAGE
COARSE
ADJUSTMENT
Ain TIGHT VALVE
VACUUM
PUMP
*Actually six impingers.
Figure 3-4. Arsenic Sampling Train (Modified EPA-5 Train).
-19-
-------
TABLE 3-2. IMPINGER SOLUTIONS USED FOR ARSENIC SAMPLING
Impinger
Number
1
2
3
Content
10%
10%
10%
H202
H2 02
H202
(250
(250
(250
Impinger
Number
ml) 4
ml) 5
ml) 6
Content
.1 N NaOH (250 ml)
.1 N NaOH (250 ml)
Silica Gel
Arsenic sampling at the ESP outlet was repeated by plant personnel during
the period of March 1, 1977 through March 3, 1977. The sampling procedure
they used incorporated a combination of the EPA Methods 5 (particulate
determination) and Method 8 (sulfuric acid mist and S02). Additionally,
the method was designed to collect arsenic or arsenic compounds on the
particulate filter, in the first two impingers (each containing 200 ml
1% HNOs) and on the inline (or Method 8) filter.
A major deviation from normal testing occurred during the ESP
outlet tests conducted by plant personnel. A full traverse could not be
accomplished. This deviation was caused by an abnormally high loading on the
inline filter located down stream from the first two impingers. A yellow-
orange to orange-red particulate material of a very fine nature would collect
on this filter and subsequently raise the sample train vacuum until the iso-
kinetic sampling rate could not be maintained. This condition would develop
in a very short time and, consequently, the sample sizes of the tests are
below the acceptable amount for a reliable test. A major modification of the
sampling equipment would have been necessary to continue testing.
20
-------
IN-STACK (600°F)/OUT-OF-STACK (250°F) GRAIN LOADINGS
The gas temperature at the ESP outlet is approximately 600°F. Certain
compounds, especially sulfuric acid mist condense if the gas is cooled to
250°F as is recommended in EPA Method 5. The increase in grain loading de-
termination caused by condensation was determined using the in-stack/out-of-
stack sampling train shown in Figure 3-5. The equipment is the same as that
shown in Figure 3-4 with the exception that a filter was inserted into the
duct to collect the material solid at duct temperature. Three determinations
were made on July 15, 1977.
In Duct Filter
(600°F)
-CYCtONE
i
IMPINOERS
FINE ADJUSTMENT
BY PASS VALVE
VACUUM
LINE
OniFICE AP
MAGNEHCLICGAGEV
DRY TEST METEO
- VACUUM
GAGE
1 COARSE
ADJUSTMENT
AIRTIGHT VALVE
VACUUM
PUMP
Figure 3-5. In-Stack/Out-of-Stack Sampling Train.
21
-------
CYCLONE SAMPLING TRAIN
The train shown in Figure 3-6 was constructed to separate the
particulate matter into three size fractions using cyclones. A difference
in the chemical composition of each fraction would have indicated that an
evaporation condensation mechanism and/or a selective removal of particles
as a function of composition occurs in the ESP.
The oven was kept at duct temperature. However, a separation was
unsuccessful due to the sticky nature of the deposit in the cyclones.
fll
3 Cyclones
n
Wet Electrostatic
Precipitator
Ice Bath
Silica Gel
Impinger
Oven
Fine
Adjustment Valve
1. . J-
-V-4
Coarse
Adjustment Valve
Pump
Figure 3-6. Schematic of the Cyclone Sampling Train.
22
-------
SECTION 4
RESULTS AND DISCUSSION OF RESULTS
The results are presented in such a manner as to follow the flow of
the material through the smelter. The compositions of concentrate, matte
and slag are discussed first. The reverberatory furnace element balance is
discussed next. A complete material balance around the reverberatory furnace
could not be established due to the lack of samples from several streams.
Some general conclusions, however, are drawn.
Two sections discussing experimental results of arsenic sampling
follow. Data taken by smelter personnel are presented first. Data gener-
ated by Radian are presented in the following section. Discussions of the
electrostatic precipitator efficiency, the electrostatic precipitator material
balance, and the major crystalline species present at the smelter conclude
the section.
REVERBERATORY FURNACE FEED
A composite of the concentrate feed to the reverberatory furnace was
collected during July 12, 1976 through July 14, 1976 by emptying the slinger
bin catches at the end of each shift. Measurements of the converter slag and
recycle dusts fed to the reverb were not possible. The measured concentration
of an element was multiplied by the flow rate of the concentrate. The average
element flows obtained in this manner for the July 12 through July 14, 1976
period are shown in Table 4-1.
-TABLE 4-1. ELEMENT CONCENTRATION AND FLOWS IN THE REVERBERATORY FURNACE FEED
Element
Arsenic
Cadmium
Molybdenum
Lead
Antimony
Selenium
Zinc
Fluorine
Analysis
Technique
AA
AA
AA
AA
AA
F
AA
SIE
Concentration in
Feed (ppm)
3900
1200
1600
980
120
200
830
67
Flow Rate (Ib/hr)
Calculated
190
59
79
49
6
10
42
3
AA - atomic absorption spectrophotometry
F - fluorometric determination
SIE - specific ion electrode using the method of standard additions
23
-------
The concentrate was also analyzed by semlquantitative spark source
mass spectrometry (SSMS) analysis. This technique shows a detection limit
of 0.1 to 1 ppm for solids. Elements present in a concentration higher
than 1000 ppm are listed as major components (MC). The results are summar-
ized in Table 4-2. Species entering as major components (>1000 ppm) are:
Arsenic Calcium Aluminum
Copper Potassium Magnesium
Iron Sulfur Sodium
Titanium Silicon
Species showing a concentration in the order of 50-1000 ppm are:
Bismuth Antimony Zinc
Lead Molybdenum Nickel
Barium Selenium Fluorine
The rest of the 71 elements surveyed show concentrations in the low ppm
ranee or are not detectable.
The high arsenic concentration in the concentrate is noteworthy. Data
measured by plant personnel during February 28, 1977 through March 4, 1977
are summarized in Table 4-3. They show values in the same order of magnitude.
The fluctuations from day to day and even from shift to shift, however, are
significant. The maximum concentration was 7360 ppm during the B shift on
March 1, 1977, the minimum was 780 ppm during the B shift on March 4, 1977.
The feed concentration of arsenic varies widely. Therefore, sampling must
be time phased to permit meaningful comparison of measurements. Material
balances can only be completed using time phased measurements.
a-Quartz and chalcopyrites, CuFeSz, were found to dominate the X-ray
diffraction pattern taken from the concentrate.
24
-------
J3LE 4-2. SURVEY ANALYSIS OF THE REVERBERATORY FURNACE FEED
(SAMPLES COLLECTED JULY 12, 1976 THROUGH JULY 14,
CONCENTRATION IN PPM WEIGHT
ELEMENT CONC.
Uranium 2
Thorium 4
Bismuth 80
Lead 910
Thallium 250
MC
>600
=160
NR
NR
NR
2
<0.1
1
NR
NR - SSMS technique is not suitable.
No value listed: Element concentration is below the detection limit of
SSMS (0.1-1 ppm for solids, 1-10 ppb for liquids)
MC - Major component, read out is off the scale on SSMS.
* - Sample not homogenous with respect to these elements
25
-------
TABLE 4-3. VARIATION OF
FURNACE FEED
ARSENIC CONCENTRATION IN REVERBERATORY
(DATA BY SMELTER PERSONNEL)
Sample
A Shift 2/28/77
B Shift 2/28/77
C Shift 2/28/77
A Shift 3/1/77
B Shift 3/1/77
C Shift 3/1/77
A Shift 3/2/77
B Shift 3/2/77
C Shift 3/2/77
A shift 3/3/77
B Shift 3/3/77
C Shift 3/3/77
A Shift 3/4/77
B Shift 3/4/77
C Shift 3/4/77
% Arsenic
0.132
.148
Average =
0.120
.120
Average =>
0.290
.290
Average =
0.384
.352
Average =
0.713
.714
.749
.769
Average =
0.310
.300
Average =
0.134
.125
.148
.156
Average =
0.240
.240
Average »
0.320
.348
Average =
0.306
.300
.306
.306
Average =
0.235
.220
Average -
0.401
.400
Average =
0.185
.175
Average =
0.080
.075
Average =
0.369
.373
Average »
0.140
0.120
0.290
0.369
0.736
0,305
0.141
0.240
0.334
0.305
0.228
0.401
0.180
0.078
0.371
26
-------
REVERBERATORY FURNACE SLAG
The concentration and flows of the elements found in high concentra-
tion in the feed are listed in Table 4-4 for the reverberatory furnace slag.
These results are extremely interesting. They indicate that the slag con-
tains 25 percent of the arsenic, 27 percent of the lead and only a trace of
the cadmium which enters the reverberatory furnace. Only half of the in-
coming antimony and selenium leave the smelter in this stream. Two-thirds
of the fluorine and zinc are discharged with the slag, while molybdenum is
almost completely contained in the slag.
Elements depleted in the slag must be present in other smelter exit
streams.
TABLE 4-4. REVERBERATORY FURNACE SLAG COMPOSITION AND ELEMENT FLOW RATES
(JULY 12-14, 1976)
Element
Arsenic
Cadmium
Molybdenum
Lead
Antimony
Selenium
Zinc
Fluorine
Analysis
Technique
AA
AA
AA
AA
AA
F
AA
SIE
Concentration in
Slag (ppm)
1300
9
2400
350
92
150
810
55
Flow Rate (Ib/hr)
Calculated
48
.3
89
13
3
5
30
2
AA - atomic absorption spectrophotometry
F - fluorometric determination
SIE - specific ion electrode using the method of standard additions
The survey analysis of the reverberatory furnace slag by SSMS is shown in
Table 4-5. Major components (>1000 ppm) are the following:
Molybdenum
Arsenic
Zinc
Copper
Iron
Titanium
Calcium
Potassium
Sulfur
Phosphorus
Silicon
Aluminum
Magnesium
Sodium
-27-
-------
TABLE 4-5. SURVEY ANALYSIS OF REVERBERATORY FURNACE SLAG
BY SSMS (SAMPLES COLLECTED JULY 12, 1976 THROUGH
JULY 14, 1976)
CONCENTRATION IN PPM WEIGHT
ELEMENT
Uran i urn
Thorium
Bismuth
Lead
Thallium
Mercury
Gold
Platinum
Iridium
Osmium
Rhenium
Tungsten
Tantalum
Hafnium
Lutetium
Ytterbium
Thulium
Erbium
Hoi mi urn
Dysprosium
CONC.
9
24
5
370
NR
4
3
0.3
1
0.3
1
1
5
ELEMENT
Terbium
Gadol inium
Europium
Samarium
ffeodymium
Praseodymium
Cerium
Lanthanum
Barium
Cesium
Iodine
Tellurium
Antimony
Tin
Indium
Cadmi urn
Silver
Palladium
Rhodi urn
CONC.
0.5
3
1
10
12
10
55
50
270
4
160
15
STD
4
1
ELEMENT
Ruthenium
Molybdenum
Niobium
Zirconium
Yttrium
Strontium
Rubidium
Bromine
Selenium
Arsenic
Germanium
Gallium
Zinc
Copper
Nickel
Cobal t
Iron
Manganese
Chromium
CONC.
MC
10
130
18
130
44
2
25
MC
0.6
8
MC
MC
14
38
MC
260
120
ELEMENT
Vanadium
Titanium
Scandium
Calcium
Potassium
Chlorine
Sulfur
Phosphorus
Silicon
Aluminum
Magnesium
Sodium
Fluorine
Oxygen
Nitrogen
Carbon
Boron
Beryl 1 ium
Lithium
Hydrogen
CONC.
15
MC
4
MC
MC
95
MC
MC
MC
>250
MC
>640
=60
NR
NR
NR
2
0.2
2
NR
NR - SSMS technique is not suitable.
No value listed: Element concentration is belov the detection limit of
SSMS (0.1-1 pptn for solids. 1-10 ppb for liquids)
MC - Major component, read out is off the scale on SSMS.
28
-------
Species showing a concentration in the order of 50-1000 ppm are:
Lead Strontium
Barium Manganese
Antimony Chromium
Zirconium Fluorine
The concentration of arsenic in the slag is shown in Table 4-6. These data
were measured by plant personnel during the period of March 1, 1977 through
March 4, 1977. The arsenic concentration in the reverberatory slag of 700-
800 ppm reported by smelter personnel compares to 1300 ppm found during the
July 12 through July 14, 1976 period.
TABLE 4-6. ARSENIC CONCENTRATION IN THE REVERBERATORY SLAG
DURING MARCH 1, 1977 THROUGH MARCH 3, 1977
(DATA BY SMELTER PERSONNEL)
Sample % Arsenic
B Shift 3/1/77 0.065
.065
Average - 0.065
A Shift 3/2/77 0.083
.093
Average = 0.088
A Shift 3/3/77 0.066
.076
Average = 0.071
B Shift 3/3/77 0.085
.086
Average = 0.086
It is noteworthy that the fluctuations are small if compared to those
observed in the concentrate (see Table 4-3). The melt in the reverberatory
furnace evidently acts as damper. The residence time is in the order of 12
to 20 hours. Fluctuations in the feed concentration can therefore be ex-
pected to be smoothed in the matte and reverberatory furnace slag. This is,
however, not true for volatile compounds like AsaOs, which can be expected
to leave the furnace immediately upon formation due to the high vapor pres-
sure at furnace temperature.
29
-------
REVERBERATORY FURNACE MATTE
The concentrations and flow rates of the same species in the matte
are listed in Table 4-7.
TABLE 4-7. CONCENTRATION OF SELECTED ELEMENTS IN MATTE (COMPOSITE SAMPLE
JULY 12 THROUGH JULY 14, 1976)
Element
Arsenic
Cadmium
Molybdenum
Lead
Antimony
Selenium
Zinc
Fluorine
Analysis
Technique
AA
AA
AA
AA
AA
F
AA
SIE
Concentration in
Matte (ppm)
1100
880
200
2000
110
4
750
<.3
Flow Rate (Ib/hr)
Calculated
48
37
8
84
5
.2
31
<.01
This data indicate that only minor amounts of the incoming molybdenum,
selenium and fluorine, are contained in the matte. This combined with the
results of the reverberatory slag analysis indicates that major parts of
the elements:
e arsenic,
• selenium, and
e fluorine,
leave the reverberatory furnace with the reverberatory furnace off-gases.
Molybdenum is primarily retained in the reverb and discharged with
the slag.
The major amounts of cadmium, lead and zinc are transported with the
matte to the converters.
The semiquantitative results of the SSMS analysis are given in
Table 4-8. Again, the samples are composites for the days July 12 through
July 14, 1976.
-30-
-------
TABLE 4-8. SURVEY ANALYSIS OF MATTE BY SPARK SOURCE MASS
SPECTROMETRY (COMPOSITE SAMPLE FOR JULY 12 THROUGH
JULY 14, 1976)
CONCENTRATION IN PPM WEIGHT
ELEMENT CONC.
Uranium 13
Thorium 13
Bismuth 120
Lead MC
Thallium 3
Mercury NR
Gold 3
Platinum
Iridium
Osmium
Rhenium
Tungsten
Tantalum
Hafnium
Lutetium
Ytterbium
Thulium
Erbium
Hoi mi urn
Dysprosium
ELEMENT CONC.
Terbi urn
Gadolinium
Europium
Samarium
Neodymi urn
Praseodymi urn 0,4
Cerium 0.7
Lanthanum 0.9
Barium 36
Cesium 0.2
Iodine
Tellurium 6
Antimony 180
Tin 30
Indium STD
Cadmium 50
Silver 110
Palladium
Rhodium
ELEMENT
Ruthenium
Molybdenum
Niobium
Zirconium
Yttrium
Strontium
Rubidium
Bromine
Selenium
Arsenic
Germanium
Gallium
Zinc
Copper
Nickel
Cobal t
Iron
Manganese
Chromi urn
CONC.
170
5
2
14
3
1
410
MC
0.2
MC
MC
60
570
MC
34
15
ELEMENT
Vanadium
Titanium
Scandi urn
Calcium
Potassium
Chlorine
Sulfur
Phosphorus
Silicon
Aluminum
Magnesium
Sodium
Fluorine
Oxygen
Nitrogen
Carbon
Boron
Beryllium
Lithium
Hydrogen
CONC.
1
26
10.4
MC
MC
19
MC
6
470
>350
15
MC
=25
NR
NR
NR
0.4
1
NR
NR - SSMS technique is not suitable.
No value listed: Element concentration is below the detection limit of
SSMS (0.1-1 ppm for solids, 1-10 ppb for liquids)
MC - Major component.
31
-------
Species found In major concentrations (>1000 ppm) are:
Lead Copper Potassium
Arsenic Iron Sulfur
Zinc Calcium Sodium
Elements found in the concentration in the order of 100-1000 ppm are:
Bismuth Molybdenum Cobalt
Antimony Selenium Silicon
Cadmium Nickel Aluminum
Silver
The rest, of the 71 elements surveyed show concentrations below the detection
limit or are present in the low ppm range.
The fluctuation of the matte arsenic concentration was measured by
smelter personnel during the period March 1 through March 3, 1977. The
average concentration found for the three days was approximately 900 ppm
(see Table 4-9). The fluctuations observed in the concentrate are dampened
as was observed for the slag. Again, the residence time of the melt in the
reverberatory furnace of approximately 12-20 hours being the reason. The
average value of 900 ppm for the March 1977 period compares to 1100 ppm
determined for the July 12 through July 14, 1976 period determined by Radian.
TABLE 4-9. FLUCTUATION OF THE ARSENIC CONCENTRATION IN THE MATTE FOR THE
PERIOD MARCH 1 THROUGH MARCH 3, 1977 (DATA DETERMINED BY
SMELTER PERSONNEL)
Sample Sample No. % Arsenic
B Shift 3/1/77 497 0.079
.093
Average = 0.086
A Shift 3/2/77 505 0.102
.108
Average = 0.105
A Shift 3/3/77 501 0.080
.032
Average = 0.081
B Shift 3/3/77 0.092
.120
.100
Average = 0.104
32
-------
REVERBERATORY FURNACE ELEMENT BALANCE (JULY 12-14, 1976)
The original sampling plan was designed to encompass the electrostatic
precipitator servicing the reverberatory furnace only, but plant personnel
made concentrate, reverberatory furnace slag and matte samples available for
analysis during the July sampling effort. Flow rates around the reverb were
released to Radian in February 1977. The values are given in Volume II.
A rigorous material balance around the reverberatory furnace is still
not possible due to the lack of the following samples:
• converter slag,
• converter dust, and
• reverberatory furnace waste heat boiler dust.
Certain conclusions concerning the fate of the elements entering the plant
as minor components with the feed are still possible.
Arsenic, cadmium, molybdenum, antimony, selenium, zinc and fluorine
were the trace elements appearing in the highest concentration. The concen-
tration of these elements in the matte, reverberatory slag and gas streams
indicate their pathway through the smelter.
Molybdenum
It is evident from Table 4-10 that nearly all the molybdenum is trapped
in the reverb slag and discharged with this stream. Reverb ESP dust and matte
contain only a minor percentage of the molybdenum entering the furnace.
Arsenic
From the 190 Ib/hr of arsenic entering the furnace roughly 1/4 is trans-
ported with the matte to the converters and roughly 1/4 is discharged with
the slag. About 1/2 leaves the reverberatory furnace with the flue gas. The
majority of this amount is not retained by the electrostatic precipitator.
Arsenic values measured by plant personnel in the electrostatic pre-
cipitator dust, the waste heat boiler dust, the converter dust and the acid
plant purge water during the March 1977 sampling effort are summarized in
Table 4-11. The low values of 1.7 percent of As in the converter dust and
0.013 percent in the acid plant purge water are consistent with the obser-
vations made by Radian during the July 1976 sampling effort.
-33-
-------
TABLE 4-10. ELEMENT FLOW RATES IN THE FEED AND DISCHARGE STREAMS OF THE REVERBERATORY
FURNACE (Ibs/hr) (RADIAN DATA)
Element
Al
As
Ba
Be
Ca
Cd
Cr
Cu
F
Fe
Hg
Mo
Ni
Pb
Sb
Se
Si
V
Zn
Incoming Streams
Reverb Converter Converter
Feed Slag Oust
400
190
31
0.072 " "
770 « «
59 J J
0.076 n w
1.6x10" * ^
3.4 < •«
1.0x10* H H
0.018 o o
79 *
0.70 « "
: :
6.0 a a
10 « «
1100
0.92
42
Total
>400
>190
SO. 072
>770
>59
>0.076
>1.6xlO"
>3.4
>1.0xlO*
>0.018
>79
X).70
>49
>6.0
>10
>1100
>0.92
>42
Matte
<17
48
33
4. 1x10" 3
12
37
0.64
1.8x10*
0.012
1 . IxlO*
0.020
8.5
2.0
84
4.6
0.17
<42
0.33
31
Slag
700
48
43
0.032
].9xl03
0.33
3.8
2.2xl02
2.2
1.2x10"
9.1xlO"3
89
0.76
13
3.4
5.5
4.8xl03
0.86
30
Outgoing Stream
Flue
Outlet
0.10
76
0.64
3.4xlO"3
0.011
.076
0.044
1.8
9.4
0.55
0.033
0.17
0.011
0.38
0.030
0.65
1.7
0.027
0.22
3
ESP* Waste Heat*
Catch Boiler Dust
1.1
30
0.023
-t. w
4.0x10 " ,j
2.1 «
0.74 \j
0.018 w
62 >
0.032 <
42.6 H
1.5x10"" °
6.6
0.059 w
5.3 ^
1.3 *
0.16 ^
1.7
0.011
4.6
Total
•WOO
•x-200
V6.2
M).039
%2000
•»46
^4.5
•^1.8x10*
%12
^2.3x10"
'\-0.062
^100
%29
^100
^•9.4
^6.5
M800
M.2
^66
* These streams were not recycled during the time of ESP sampling.
Concentrate
,., „, sample not
Converter Slag - -
available
Converter Dust
sample not
available
Reverberatory
Furnace
sample not
available
-» Matte
_». Slag
—> Flue Outlet
ESP Catch
-.Waste Heat Boiler Dust
-------
TABLE 4-11. ARSENIC CONCENTRATION IN THE WASTE HEAT BOILER DUST, REVERB
ESP PUST, CONVERTER DUST AND ACID PLANT PURGE WATER
(DATA BY PLANT PERSONNEL)
Date
3/1/77
3/2/77
3/3/77
3/4/77
3/5/77
3/6/77
3/7/77
3/1/77
3/2/77
3/1/77
3/2/77
4/3/77
Sample
Reverb ESP Dust
Reverb ESP Dust
Reverb Dust
Reverb ESP Dust
Reverb ESP Dust
Reverb ESP Dust
Reverb ESP Dust
Waste Heat Boiler Dust
(Reverb)
Waste Heat Boiler Dust
(Reverb)
Converter Dust
Converter Dust
Acid Plant Purge
Water
% Arsenic
5.5
5.6
5.7
Average
5.0
5.1
Average
4.1
4.9
Average
5.3
5.0
Average
4.5
4.8
Average
3.3
3.5
Average
3.3
6.5
6.5
Average
4.7
4.6
Average
1.8
1.9
Average
1.3
1.4
Average
0.0013
.0012
.0014
Average
= 5.6
= 5.1
= 4.5
« 5.2
= 4.7
= 3.4
= 6.5
= 4.7
- 1.9
- 1.4
= 0.0013
Selenium
The selenium data are not quite conclusive. Fifty percent of the in-
coming selenium is discharged with the slag. The flows found in the matte,
flue outlet and ESP catch do not account for the rest. A potential expla-
nation could be condensation of a volatile selenium compound in the reverber-
atory waste heat boilers.
The phase of the selenium compound escaping the reverb ESP could not
be identified. The X-ray diffraction pattern did not match the lines of
elemental selenium, known selenium oxides, or selenium sulfide.
35
-------
ARSENIC MATERIAL BALANCE AROUND THE SMELTER (SMELTER DATA)
An arsenic material balance around the smelter was established by plant
personnel. Inputs to this balance were composite samples and flows for the
whole month of January. The analytical values used in this balance are shown
in Table 4-12 with exception of the acid plant purge water. That part of the
smelter included in the balance is shown in Figure 4-1.
TABLE 4-12. ARSENIC CONCENTRATIONS IN INCOMING AND OUTGOING STREAMS OF THE
REVERBERATORY FURNACE DURING THE MONTH OF JANUARY 1977
(DATA BY PLANT PERSONNEL)
Sample
% Arsenic
Smelter Concentrate
Composite, Jan. 1977
Matte Composite
Jan. 1977
Reverberatory Slag
Composite, Jan. 1977
Reverberatory Precipitator
Dust Composite, Jan. 1977
Converter Precipitator
Dust Composite, Jan. 1977
Converter Flue Dust
Composite, Jan. 1977
Converter Slag
Composite, Jan. 1977
Anode Copper
Composite, Jan. 1977
0.280
.270
.250
.248
Average
0.074
.080
Average
0.061
.064
Average
4.4
4.4
Average
7.7
1.7
1.8
Average
0.014
.015
Average
0.029
0.029
Average
= 0.262
= 0.077
= 0.063
= 4.4
= 1.8
= 0.015
= 0.029
36
-------
Reverb SlagNi
Slag Dump/'
Reformed Gas
Anode
Furnace
Blister
Copper
Slag \
V /
Concentrate and A
Flux Material j
(Natural Gas, Diesel Oil
and/or No. 6 Fuel Oil
One Reverberatory
Furnace
Copper
Matte
/ Converter
I Slag
Three Converters,
One Great Falls
Oxidizing Furnace
Single
Absorption
Acid Plant
(5)/ Anode
\ Copper
93%
Tail Gas
Acid Plant
Purge Water
Figure 4-1. Boundary of Arsenic Balance.
-------
101.9 Ib/hr
The following flows were found based on plant data:
Incoming Streams: (Average for January 1977)
Point©: Concentrate Feed 101.9 Ib/hr
Point©: Lime Rock Feed 0_.0 Ib/hr
TOTAL ARSENIC IN:
Outgoing Streams: (Average for January 1977)
Point(|): Reverb Furnace Slag 25.7 Ib/hr
Point©: Acid Plant Purge Water 27 Ib/hr
Point(5): Anode Copper 4.6 Ib/hr
Point (6): Reverb ESP Outlet X
TOTAL ARSENIC OUT
57.3 + X Ib/hr
Reverb ESP Outlet = 101.9 - 57.3 = 44.6 Ib/hr
The value of 27 Ib/hr used in the balance for acid plant purge water was de-
termined from a sample taken March 4, 1977. Streams not included in this
scheme comprise the tail gas and the sulfuric acid from the acid plant. The
arsenic flow in these streams, however, can be expected to be very much
smaller than that in the purge water stream.
These data indicate that roughly 50% (44.6 Ib/hr) of the incoming ar-
senic (101.9 Ib/hr) leaves the smelter with the reverberatory furnace flue
gas, 25% (25.7 Ib/hr) are discharged with the reverberatory furnace slag and
25% (27 Ib/hr) are transported with the matte to the converters and leave the
converters with the off-gases.
The same 50-25-25 split was indicated in the reverberatory balance for
arsenic based on the samples obtained during the July 1976 sampling effort
(see Table 4-10). The only major difference is that the arsenic flow rates
in the streams is roughly double that measured by plant personnel during
January. The difference can be explained if it is attributed to the wide
arsenic fluctuations as indicated in Table 4-3.
The same balance attempt was repeated by plant personnel during March
1977. Average flows were as follows:
38
-------
Incoming Streams:
Point(T) : Concentrate Feed
Point(2): Lime Rock Feed
TOTAL ARSENIC IN
Outgoing Streams:
Point (5): Reverb Furnace Slag
Point (4): Acid Plant Purge Water
Point(5): Anode Copper
Point©: Reverb ESP Outlet
TOTAL ARSENIC OUT
103 Ib/hr
0 Ib/hr
103 Ib/hr
31.1 Ib/hr
27.1 Ib/hr
5.2 Ib/hr
X
63.4 + X Ib/hr
By difference:
Reverb ESP Outlet X
103 - 63.4 = 39.6 Ib/hr
The actual values obtained by direct measurements during March were:
25 Ib/hr (March 1, 1977), 34 Ib/hr (March 2, 1977), and 49 Ib/hr (March 3,
1977). Of these emissions, 16, 17 and 23 Ib/hr was in the form of gaseous
arsenic on March 1, March 2 and March 3, 1977 respectively. Difficulties
were encountered due to the deposit of a yellow-orange material clogging the
filter.
ARSENIC FLOW AT ESP INLET AND OUTLET (RADIAN DATA)
The arsenic sampling train described in Section 3 was used by Radian
to collect samples for arsenic analysis on July 13 (9:51 a.m.-13:36 a.m.)
and July 14, 1976. Both the inlet and outlet ducts were traversed. The
exact sampling times, traverse points used and the flows determined are
summarized in Table 4-13. Figures 4-2 and 4-3 help as visual guides to
locate the sampling points indicated in Table 4-13.
Arsenic flow rates of 59.3, 72.9 and 75.4 Ibs/hr giving an average of
69.2 Ibs/hr were found at the inlet. The corresponding values at the outlet
were 53.7, 44.8 and 51.3 Ibs/hr or an average of 49.9 Ibs/hr.
These values are slightly higher than the flows determined by difference
by Smelter personnel (44.6 Ib/hr for January and 39.6 Ibs/hr for March).
Direct measurements by plant personnel for March 1, 2 and 3, 1977 determined
flows of 25, 34 and 49 Ibs/hr.
39
-------
TABLE 4-13. SUMMARY OF ARSENIC EMISSION DATA FROM THE REVERBERATORY
ELECTROSTATIC PRECIPITATOR (RADIAN DATA)
INLET
Gas
Time Flow Rate
Run Date (mln.) (acfm)
1 7-13-76 0930-1226 155200
2 7-14-76 0555-0907 156000
3 7-14-76 1132-1505 157900
Average - 156400
Sampling
Points
4-6
4-6
4-6
4-6
4-6
4-6
E;
W;
E;
W;
E;
W;
13-15
13-15
13-15
13-15
13-15
13-15
Arsenic
Flow Rate
Ubs/hr)
E 59.3
W
E 72.9
W
E 75.4
W
69.2
OUTLET
Gas
Time Flow Rate Sampling
(min.) (acfm) Points
0951-1336 170700 N
W
0610-0855 168900 N
W
1130-1415 166500 N
W
168700
7-16
1-17
1-17
1-17
1-17
1-17
Arsenic
Flow Rate
(Ibs/hr)
53.7
44.8
51.3
49.9
Other determinations by Radian showed a flow of 140 Ib/hr (ESP balance
on July 11) and 75 Ib/hr (reverb balance July 13, 1976, 6:35 a.m.-10:28 a.m.).
The 76 Ib/hr are slightly more than the 53.7 Ib/hr found with the arsenic
train the same day only 4 hours later.
West Duct
East Duct
3- 6X 9.
2. 5x 8.
12. 15 X 18-
11. 14 X 17.
1 . 4 X 7 . 10 . 13 x 16-
1 1
j«i2»>;
1
Figure 4-2. Traverse Points Layout
3. 6X 9.
2. 5X 6.
I. 4X 7.
i ,„.,„ i
1 >
I
12. 15 * 18-
11. 14 X 17.
10, 13 x 16.
innr
tator Inlet.
40
-------
*17
*16
•"15
x 14
* 13
v 12
x 11
* 10
north
port |12345 6
X < X X if
8
*
9
v
10
^
North port extention = 5.25"
West port extention = 5.25"
11 12 13 14 15 17
* if <
16 18
Inside diameter = 72.5"
Figure 4-3. Traverse Points Layout for Reverberatory Electrostatic
Precipitator Outlet.
41
-------
COLLECTION EFFICIENCY OF THE ELECTROSTATIC PRECIPITATOR
In Stack Grain Loadings
The particulate sampling conditions at the inlet were quite nonideal
due to the duct configuration. The problem was enhanced due to a surge in
particulate concentration each time the reverberatory furnace was fed. In
addition, the material collected clogged the in-stack sampling devices.
Sampling with thimbles had to be abandoned in favor of in-duct Gelman filters.
Duct temperature ranged typically from 575 to 650°F during these experiments.
The six sampling ports of both the inlet east and inlet west duct were
traversed during the times and dates indicated in Table 4-14. A schematic
showing the sample point numbers is included in Table 4-14. The grain load-
ings varied from .358 grains/scf to 1.552 grains/scf and showed an average
of 0.60 grains/scf.
The measurements at the ESP outlet showed much less fluctuation. The
ESP evidently dampened the surge in grain loading observed at the inlet.
Dates, times and duct locations are summarized in Table 4-15. Average grain
loading was 0.020 grains/scf.
The overall particulate removal efficiency of the ESP was calculated
to be 96.7% @ 600°F which is in excellent agreement with the design efficiency
of 96.8%.
In-Duet/Out-of-Duct Grain Loading (July 15, 1976)
These experiments were designed to determine the amount of particulate
which condenses by cooling the flue gas from duct temperature to 250°F, which
is the temperature recommended by EPA for particulate sample collection.
The amount of particulate which condensed corresponds to an additional
1.6 gr/scf at the inlet and 1 gr/scf at the outlet. The dry electrostatic
precipitator is not able to collect species in the vapor state. The results
are summarized in Table 4-16.
During the three in-stack, out-of-stack grain loading determinations
(July 15, 1976, 8:50 a.m.-16:30 p.m.), a reddish condensate formed at the
inlet tube to the first impinger. This is between the 250°F oven and the
first impinger in the ice bath. A sample which accumulated in the inlet pipe
was subjected to X-ray fluorescence and SSMS analysis.
42
-------
TABLE 4-14. GRAIN LOADING DATA INLET
Date
July 10
July 10
July 10
July 10
July 10
July 9
July 9
July 9
July 9
July 8
July 8
Point
W 1-12
W 4-6
W 1-12
E 1-12
E 1-12
W 1-8
W 1-12
E 1-5
E 1-12
W 1-9
E 1-6, 9
Time
0640-0845
0908-0923
1005-1135
0645-0830
1010-1230
0657-0929
1117-1315
0650-0852
1120-1350
1035-1303
1030-1302
Average
SCF Grains/SCF
17.17
4.51
18.59
19.12
18.76
20.91
18.20
10.81
26.51
45.33
28.57
Grain Loading
1.552
.501
.519
.358
.647
.647
.433
.365
1.013
.422
.173
.60
7%
3
• —
2
• —
1
1
— |—
12"
6
—
5
—
4
— 1-
1
, 12"
9
—
8
7
H-
12"
12
—
11
—
10
H-
i
12"
15
— •
14
13
I
1
1
r 12"
8
7
6
• i ;%"
I
ll
T
a"
_L
3" Openings
Male Threads
Cross Section and Sample Point Identification at Inlet
East and West Ducts
43
-------
TABLE 4-15. GRAIN LOADING DATA OUTLET
ESP
Outlet
7-10
7-9
7-8
Time
0630-0814
1010-1215
0650-0915
1135-1315
1030-1305
SCF
25
33
25
25
.32
.05
.03
.12
32.88
AVERAGE
Grain Loading
.018
.017
.019
.022
.025
.020
Sampling
W
W
W
W
W
1-6;
1-6;
1-6;
1-6;
1,2;
S
s
N
N
N
Points
12-9
12-7
7-12
7-12
7-12
SAMPLE POINT
1.7
2,8
3,9
4,10
5,11
e.12
SOUTH PORT
3". FEMALE
DISTANCE FROM
INSIDE WALL
3.1"
10.6"
4'2.8"
5'8.9"
WEST PORT
3", FEMALE THREAD
W
Cross Section and Sample Point Identification of Exit Duct,
-44-
-------
TABLE 4-16. IN-DUCT/OUT-OF-DUCT GRAIN LOADING (gr/scf)
(JULY 15, 1977)
Run
1
2
3
Time
0845-1028
1300-1420
Lost Sample
In-Due t
600'F
.58
.31
INLET
Out-of-Duct
250°F Combined
1.89 2.47
1.25 1.56
—
Run
1
2
3
Time
0850-1020
1300-1400
1530-1630
In-rDuct
600°F
.027
.072
.019
OUTLET
Out-of-Duct
250°F
.94
.78
1.35
Combined
.97
.84
1,37
Similar deposits were observed during in-stack grain loading measure-
ments and during arsenic sampling. They also were reported from plant per-
sonnel during their effort to determine the arsenic flow rate at the ESP
exit.
X-ray fluorescence indicated that arsenic and selenium are the pre-
dominant heavy elements in this sample. The presence of smaller amounts of
zinc and potentially lead were indicated.
The results of the SSMS analysis confirm the presence of arsenic and
selenium as major species (>1000 ppm). Elements found in smaller concen-
tration are lead, rhenium and zinc.
Vapor Train Gaseous Emissions (July 16, 1976)
Compounds emitted as vapors were: arsenic (16 Ib/hr) , copper (3 Ib/hr) ,
fluorine (11 Ib/hr) and selenium (2 Ib/hr). Arsenolite (AsaOs) was identified
by X-ray diffraction in the condensibles. These emissions found by quanti-
tative analytical methods agree with the X-ray fluorescence analysis and the
SSMS results of the material condensed at the impinger inlet tube. The
vapor train as described in Section 3 was used to establish these results.
ELECTROSTATIC PRECIPITATOR MATERIAL BALANCE (JULY 11, 1976)
The closure of the material balance results within the calculated
error limits increases the confidence in the analytical data and are pre-
sented in Table 4-17. Barium and lead were found in the precipitator dust
in substantial quantities while the amounts found in the WEP samples collected
at the inlet duct were, by comparison low. These results indicate that the
use of sulfuric acid in the WEP's as well as the presence of a high SOs con-
centration in the reverberatory furnace off-gases formed lead and barium
sulfate precipitates which are not dissolved using the digestion techniques
employed. However, the lead concentration found in the exit WEP liquor of
0.6 ppm is below the saturation point of approximately 10 ppm,, which was
45
-------
TABLE 4-17. ELEMENTAL FLOW RATES IN ESP INLET AND OUTLET STREAMS
Element
Arsenic
Barium
Beryllium
Cadmium
Chromium
Copper
Fluorine
-P-
ON Mercury
Molybdenum
Nickel
Lead
Sulfur
Antimony
Selenium
Vanadium
Zinc
Inlet 1st
WEP acid
liquor impinger
190 .036
ND ND
.013 1.2xlO~7
.29 .00068
.027 ND
56 .0054
7.4 .23
.0011 .0025
8.7 .00060
. 092 ND
.92 .0023
Inlet
(Ib/hr)
2nd
acid
impinger
.0060
ND
ND
ND
ND
.0075
.029
.0059
ND
ND
.00072
Outlet
(Ib/hr)
1st 2nd
basic basic
impinger impinger
ND
ND
ND
ND
ND
.0086
.021
.00066
ND
ND
ND
(1320 Ib/hr S as SOa, 30 Ib/hr S as
30 Ib/hr in flue dust)
.81 9.8x10"
.91 ND
.0*1 .036
4.3 .0068
5 .00025
ND
.011
.0068
ND
8.6xlO~s
.0039
.0033
1.8
ND
ND
ND
ND
.018
.0012
ND
ND
ND
ND
S03,
ND
.00078
.0081
ND
£ in
(Ib/hr)
190±95
<.0085±.005
.013±.006
.29±.13
.027+. 012
56±25
7.6±3.2
.010±.006
8.7±4.0
.09±.05
.921.52
1400±640
.81±.36
.91*. 34
.099±.046
4.3±2.2
Outlet 1st 2nd
WEP acid acid
liquor impinger impinger
140 .0084
ND ND
.011 ND
.016 ND
.011 ND
1.0 .0055
7.5 .29
.00087 .0013
.16 .0018
.085 ND
.075 .0035
(2210 Ib/hr S
1.2 Ib/hr in
.33 ND
.97 ND
.0047 ND
.072 .0064
.0016
ND
ND
S.lxlO"5
ND
ND
.012
.0027
ND
ND
ND
as S02, 50
flue dust)
ND
ND
ND
.0032
1st
basic
impinger
-
ND
ND
ND
ND
.0077
.020
.00072
ND
ND
ND
Ib/hr S as
ND
7.2xlO~5
.022
ND
2nd
basic
impinger
.0055
ND
ND
ND
ND
.0077
.016
ND
ND
ND
ND
SOs,
ND
5. 5x10" 5
.035
ND
precipitator
dust
30
.023
.00040
.74
.018
62
.032
.00015
6.6
.059
5.3
30
1.3
.16
.011
4.6
£ out
(Ib/hr)
170±84
<. 0351.016
.011±. 005
.74±.34
.029±.007
63±28
7.8±3.4
.006+. 003
6.8+3.4
.14±.07
5.4±2.2
229011040
1.6±.80
1.1+.5
.073±.032
4.7+2.3
-------
determined by adding Pb(NOs)2 to 5% sulfuric acid and equilibrating the
mixture overnight. The emission rate of the lead which was found from the
WEP outlet analysis is therefore valid. The results of the X-ray fluores-
cence analysis indicate that at least part of the arsenic and selenium leave
the ESP in the vapor state. This is in agreement with the results described
under "Gaseous Emissions."
IDENTIFICATION OF MAJOR CRYSTALLINE SPECIES
X-ray diffraction was used to investigate the nature of the major
crystalline phases. The goniometer technique was applied. Diffraction
patterns from the following samples were obtained:
• reverberatory furnace feed concentrate,
• dust collected by the electrostatic precipitator,
• particulate, collected at the ESP outlet using
an in-stack filter,
• condensed particulate - collected at the ESP outlet
at 250°F preceded by an in-stack filter at 600°F,
• condensed particulate - collected at the ESP inlet
at 250°F preceded by an in-stack filter at 600°F;
this filter was unusual in that the collected material
was black while that on all other filters used in
this manner was very light in color, and
• material which condensed in the glassware following
the filter oven during modified EPA Method 5
sampling (in-stack/out-of-stack grain loading).
Copper K radiation was used in all diffraction investigations.
0(f
A color change on the Gelman filters was observed as function of time.
The same observation was reported by Southern Research Institute during
Brinks impactor measurement. The presence of CuSOif and CuSOit'HaO in the
hopper dust explain these observations. Both compounds pick up water of
crystallization upon exposure to water vapor present in the laboratory at-
mosphere forming copper sulfate trihydrate and pentahydrate. A color as
well as a weight change accompany this transition. Both phases were identi-
fied by X-ray diffraction on the in-stack filters exposed for several months
to laboratory atmosphere.
Arsenolite (AsaOs) was positively identified in the hopper dust, in
the material captured by the in-stack filter, in the deposit on the out-of-
stack filter as well as in the impinger operated at approximately 32°F.
These findings indicate that arsenolite leaves the ESP in both a solid and
a gaseous phase.
47
-------
Other compounds identified positively by X-ray diffraction are a-Quartz
and chalcopyrite in the feed concentrate, and probably hematite and tenorite
in the hopper dust. The findings of the X-ray diffraction studies are sum-
marized in Table 4-18.
-48-
-------
TABLE 4-18. SUMMARY OF CRYSTALLINE SPECIES IDENTIFIED BY X-RAY DIFFRACTION
UD
Sample a-Quartz Chalcopyrite
Designation S102 CuFeSp
Reverberatory present present
Furnace
Feed Concentrate
(Composite July 12-14)
Dust Hopper present
(Composite July 11- 13)
In-Stack Filter
(July 15)
Arsenolite Copper Sulfate
As203
present present
present
Copper
Sulfate
Hydrate
CuSO,,vH20
present
Copper
Sulfate
Bonatite Pentahydrate
CuSO,,-3H20 CuS(V5H20
present present
Hematite Tenorite
ot-FepOa CuO
probably probably
present present
Out-of-Stack Filter present
(July 15)
Out-of-Stack Filter present
(July 15)
Material Condensed present
in Impinger
(July 15)
-------
TABLE OF CONVERSION FACTORS
Multiply
English Unit
by
Conversion
To Obtain
Metric Unit
acres
acre-feet
barrel, oil
British Thermal Unit
British Thermal Unit/pound
cubic feet/minute
cubic feet/second
cubic feet
cubic feet
cubic inches
degree Fahrenheit
feet
gallon
gallon/minute
horsepower
inches
inches of mercury
pounds
million gallons/day
mile
pound/square inch (gauge)
square feet
square inches
tons (short)
yard
0,
1233,
158,
0,
0,
0,
1.
0.
28,
16,
0.555(°F-
0.
3.
0.
0.
2.
0.
0.
3785
1.
(0.06805
0.
6.
0.
(a)
405
5
97
252
555
028
7
028
32
39
•32)
3048
785
0631
7457
54
03342
454
609
psig+1)
0929
452
907
(a)
0.9144
hectares
cubic meters
liters
kilogram-calories
kilogram calories/kilogram
cubic meters/minute
cubic meters/minute
cubic meters
liters
cubic centimeters
degree Centigrade
meters
liters
liters/second
kilowatts
centimeters
atmospheres
kilograms
cubic meters/day
kilometer
atmospheres (absolute)
square meters
square centimeters
metric tons (1000
kilograms)
meters
(a) Actual conversion, not a multiplier.
50
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-78-065a
2.
3. RECIPIENT'S ACCESSION«NO.
4. TITLE AND SUBTITLE
TRACE ELEMENT STUDY AT A PRIMARY COPPER SMELTER
Volume I
5. REPORT DATE
March 1978 i
date
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Schwitzgebel, K., R. T. Coleman, R. V. Collins
R. M. Mann, and C. M. Thompson
8. PERFORMING ORGANIZATION REPORT NO
N.A.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Radian Corporation
8500 Shoal Creek Blvd.
Austin, Texas 78766
10. PROGRAM ELEMENT NO.
1 AB 60k
11. CONTRACT/GRANT NO.
. EPA 68-01-U136
12. SPONSORING AGENCY NAME AND ADDRESS
Industrial Environmental Research Laboratory - Gin. , OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio ^5268
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/12
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This project was undertaken to explore the distribution of trace elements in
environmental emissions from a primary copper smelter. The efforts were
concentrated on the reverberatory furnace and the electrostatic precipitator
controlling emissions from the reverberatory furnace. The following major
conclusions were reached: (l) the electrostatic precipitator effectively
controls all particulate emissions at its design efficiency rating (about 96%)
at the operating gas temperature of 600 degrees F; (2) appreciable material
composed of toxic trace elements pass through the precipitator in the vapor
state at the ESP operating temperature and condense to form particulate upon
cooling. Arsenic trioxide was a major constituent of the emissions passing
the ESP from the reverberatory furnace. The following elements were examined:
Al, As, Ba, Be, Ca, Cd, Cr, Cu, F, Fe, Hg, Mo, Ni, Pb, Sb, Se, Si, V, Zn.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Exhaust emissions
Smelting
Trace elements
99A
Pollution
13. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)'
Unclassified
21. NO. OF PAGES
61
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 22ZO-1 (9-73)
-51-
* U.S. GOVERNMENTPHINTING OFFICE: 1978— 757 - 140 /13 78
------- |