TEST PLAN FOR AN IN-BEPTH STUDY
OF THE S3 BATTERY SMELTING FJENACE IN
GI.OSTRUF, DENMARK
CORPORATION
-------
RADIAN
CORPORATION
DCN #78-200-163-09
TEST PLAN FOR AN IN-DEPTH STUDY
OF THE SB BATTERY SMELTING FURNACE IN
GI.OSTRUF, DENMARK
by:
R. T. Coleman Jr., J. R. Hoover, and R. Vandervort
Radian Corporation
8500 Shoal Creek Blvd.
Austin, Texas 78766
Prepared for:
James A. Gidecn, Project Officer
Control Technology Research Branch
National Institute for Occupational Safety and Health
Cincinnati, Ohio 45268
and
Alfred B. Craig, Project Officer
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
July 17, 1978
8500 Shoal Creek Blvd./P.O. Box 9948/Austin, Texas 787667(512)454-4797
-------
TABLE OF CONTENTS
Page
INTRODUCTION 1
PROGRAM OBJECTIVES AND APPROACH 2
Technical Approach - NIOSH 3
Technical Approach - EPA 7
PLANT DESCRIPTION 9
SCOPE OF WORK 14
Test Plan - NIOSH 14
Test Plan - EPA 17
SCHEDULE AND MILESTONES 28
APPENDICES A AND B
-------
SECTION 1
INTRODUCTION
Radian Corporation of Austin, Texas is currently under contract to the
National Institute for Occupational Safety and Health to determine the most
effective controls used by the secondary nonferrous metals industry to pro-
tect workers from hazardous chemical and physical agents. As part of a
preliminary survey of the Secondary Lead Industry, Radian has identified
the SB battery smelting furnace and the flash dust agglomeration furnace
operated by Paul Bergstfe and Sons as two new processes possibly offering the
best control of secondary lead smelting process emissions. The purpose of
this study plan is to quantify, to some degree, the amount of worker protec-
tion afforded by these processes in reducing the fugitive emission of fumes
and dust to the workplace.
A second objective of this test plan is to quantify the extent and nature of
emissions to the atmosphere. This portion of the program is sponsored by the
U.S. Environmental Protection Agency under an interagency grant. The gather-
ing of information for both government organizations simultaneously is pro-
posed for efficiency of time and money and in order to generate more complete
results. The basic philosophy of this test plan is to quantify workplace and
atmosphere emissions during a time period in which the operating parameters
and material flaws of the process are also well defined. This process know-
ledge will be useful in estimating the potential change in emissions due to
process changes (such as different feedstocks) should this technology be
applied in the U.S.
-------
SECTION 2
PROGRAM OBJECTIVES AND APPROACH
The objective of both the Control Technology Research Branch of NIOSH and the
Metals and Inorganic Chemicals Branch of EPA in pursuing this program is to
provide technical support for the current or proposed regulations regarding
lead and the other toxic elements processed by the secondary lead industry.
Both agencies are interested in identifying the best technology available for
controlling lead emissions.
The SB smelting system and the flue dust flash agglomeration furnace used at
the Glostup, Denmark smelter have been identified as two processes which
potentially represent best available control technology (BACT) for the
secondary lead industry. This program will provide some of the answers
needed to judge the effectiveness of these furnaces in meeting both workroom
exposure and atmospheric emission standards.
The interests of these two agencies are different in that NIOSH is concerned
with controlling the workroom environment while EPA focuses on stack or
over-the-fence emissions. Some overlap in interests occurs because EPA is
interested in the effect of fugitive process emissions on the ambient environ-
ment. Fugitive emissions are essentially the same as the workroom contamina-
tion studied by NIOSH except that the effects of the emissions outside the
workplace rather than inside are of concern to EPA.
The general objectives of this program are to:
Quantify the stack emission rates for lead, antimony,
arsenic, chlorine, fluorine, and sulfur dioxide,
-------
• Quantify the exposure to lead and other agents received
by the workers at the smelter,
Complete a material balance around the SB smelting
furnace for lead, arsenic, antimony, chlorine, and
fluorine,
Document the process operating conditions under which
the above measurements were taken, and
• Identify the potential limitations of the process which
may limit or prevent its applicability to the U.S.
smelting industry.
In addition, the operation of several other small furnaces will be studied
by NIOSH. The following discussion describes the specific approach to in-
formation gathering and evaluation used by both agencies in performing this
study.
TECHNICAL APPROACH - NIOSH
Integrated Process Description
An appreciation for materials flow, energy consumption, transformation of
raw materials, etc., must be obtained to completely evaluate process controls.
The observations made and data collected on each of these topics during this
study will be added to the plant description given in Section 3 to give a
complete description of how the SB smelting furnace operates. The three
factors needed to complete this type of process description are:
Chemical Agent/Physical Agent Control
• Process/Control Illustration
Equipment Description
Each of these are described below.
-------
Chemical Agent/Physical Agent Control —
Each control within a process has been designed to eliminate problems with
chemical or physical agents. An important part of the description of existing
controls will be an analysis of which hazards are influenced by the presence
of the control. In some situations a control may be an effective means of
reducing exposure to a chemical agent while offering little or no relief from
exposure to various physical agents. Similarly, a particular control may
afford reduced exposure to particular types of chemical agents while being
virtually ineffective against others. Examples would be different physical
forms of contaminant materials, gases, vapors, particulates, fibers, etc.
The presence of occupational hazards associated with a particular operation
and the effects of the control being studied on these hazards will be docu-
mented by collecting personal, area, and grab samples. This sampling data
will be supplemented with visual observations by a trained industrial hy-
gienist.
Process/Control Illustration —
Description of existing controls would not be complete without the pictorial
representations of control configurations. Pictorial or graphic information
may take the form of engineering drawings similar to those found in the
ACGIH Ventilation Manual; heating and ventilating engineering drawings; or
photographic prints taken of actual control installations. The purpose of
these drawings or photographs is to clarify the physical shape of the control
and its relationship to the operation or process. In some situations where
lighting and contaminant visibility permit it may be possible to use photo-
graphs to visually record the effectiveness of a particular control during
certain portions of an operating cycle. Sole reliance will not be placed on
photographs for documentation of engineering design parameters.
Equipment Description —
The equipment involved in a particular operation will be documented. Where
appropriate, equipment will be described in detail to include model numbers,
-------
horsepower, age, associated attachments, state of repair, etc. A variety
of measurements will be made to describe the control being studied. These
measurements will include physical dimensions of the control structure itself
and its relationship to the operation or process. Materials of construction
will be noted, and if possible, historical capital and operating costs of
control equipment will be obtained. An effort will be made to obtain informa-
tion regarding the expected service life of control equipment, the amount of
downtime associated with this equipment, and specific precautions or problems
associated with making routine repairs to the control. Measurements of this
type will be of crucial importance to the evaluation of a control.
Documentation of Control Operating Parameters
Essential to this project effort will be the quantification of operating
parameters for controls studied. A variety of techniques will be utilized
in the field for this purpose. In some situations, physical measurements
may adequately describe operating parameters. These measurements may
include capture velocity, hood static pressure, duct transport velocity,
etc. In other situations, it may be necessary to perform certain types
of air sampling. This air sampling may involve a point emission source,
sampling within a duct or stack, or work area/personnel monitoring. Sampling
will only be performed where it will aid the analysis of operating parameters
associated with a particular control.
Documentation of operating parameters will also be approached using visual
observation techniques, for example, the use of ventilation smoke tubes.
In some cases, comparison of observed control efficiencies to theoretical
design relationships obtained from engineering manuals and other sources
will be conducted.
Relationships Between Employees, Hazards, Operations, Processes and Controls
In some situations, the actions of the employee may significantly affect the
relative protection afforded by a control device. The subjective area con-
cerning work practices will be examined by observing the interrelationship
between employee actions and control effectiveness. In a similar fashion,
-------
the use of personal protective clothing and equipment will be noted to
determine what additional protection is potentially afforded by these items.
These observations will help determine whether the employees' interaction
with the engineering controls limit its effectiveness.
Maintenance activities often involve employee exposure to levels of
contaminants or energy produced by a process which are atypical to the
normal working exposure. Often, engineering controls are not in operation
during maintenance periods. For these reasons, each control will be exam-
ined to determine how maintenance is performed, how often it is performed
and what procedures are utilized in performing routine and emergency main-
tenance.
A similar analysis or evaluation will be performed concerning process upsets
or particularly energetic cycles in a production process. It is possible
that some controls studied during these in-depth evaluations will offer
complete or nearly complete control during major portions of operating
cycles. However, there may be certain periods during an operating cycle
where a rapid emission of comtaminants or burst of physical energy is lib-
erated for which the control is not capable of handling, The importance
of these momentary upsets will depend on their frequency, duration, and
the relative intensity of energy release or toxicity of contaminant emitted.
Where these conditions are observed, an attempt to evaluate the importance
of such conditions will be made. Realistically, however, this evaluation
will have to be qualitative in nature.
Critique of Control Application
Possibly the most important product to be derived from these in-depth
surveys is the evaluation of the control with emphasis on how it may be
improved, what other forms of control may be equally effective, and what
potential exists for transferring the particular control technology to
other operations. A discussion of possible improvements will be included
with each control evaluation. The possible improvements are expected to
cover a wide range. In some cases, it may be necessary to offer more
-------
enclosure or to cl'nge the shape or structure of a particular exhaust hood.
Potential improvements will not be limited to the physical construction of
the control. In some circumstances, it may be possible to improve employee
work practices, use of personal protective equipment, or management of the
process or operation to minimize the impact of upsets or energetic process
cycles.
A discussion of alternative controls will be approached with caution. It
will be difficult and demanding for the survey team to gain complete under-
standing of the trade-offs offered by various means of control and how they
impact the working environment in the production of product in a particular
smelter. Heavy reliance will be placed on the experience of smelting opera-
tors to critique any alternative approach which may be forwarded. The intent
of this project is to suggest and document reasonable and effective controls
for specific operations in secondary smelting.
TECHNICAL APPROACH - EPA
The approach used by EPA in this study is designed to quantify the environ-
mental emissions generated by the SB smelting furnace and relate those emis-
sions to a specific set of process operating conditions. This will establish
a basis for comparing emissions from this process to other lead smelting
processes. To do this, two sets of experiments will be performed. First,
the standard EPA stack sampl :fn.fr will be performed (EPA Method 5) to quantify
the atmospheric emissions. Second, material balances for six of fhp imfinr-
tant toxic ele^nts processed, lead, antimony, arsenic, fluorine, chlorine,
and sulfur will be completed in order to define the flow of material through
the smelter. These measurements rill be examined along rith the data genera-
ted by NIOSH in order to establish a basline for the total emissions which
can be expected from this process. Section 4 presents a detailed list of
each measurement to be :.i = de for the experiments described above.
The preparation of a detailed process description will also be a part of the
EPA scope of work. It is of prime interest to relate atmospheric emissions
-------
from the SB smelting furnace to the furnace design and operation. An
integral process description similar to that described under the NIOSH
TECHNICAL APPROACH will be prepared.
-------
SECTION 3
PLANT DESCRIPTION
Secondary lead smelting invuljts three major operations: scrap pretreatment,
smelting, and refining. Figure 1 outlines the material flnr in a secondary
leid smelter and lists the major proces^s, raw materials, and products.
V/hole battery smelting is an impQrtant nrnrpss "in the Rm«T»fiiag operation
because it eliminates the battery breaking step. In addition, sino3 flue
dust agglomeration is an integral part of the whole battery smelter, a
major part of the smelter fugiti^t dust emissions are eliminated.
The rnncept of whole battery smelting includes luth environmental emissions
and n -cupational health as key design parameters. Plant layout, raw material
storage and handling, process and hygiene ventilation, housekeeping, process
rnntrol, flue dust agglomeration, particulate rnllection, and the smelting
parameters are all included in the smelter design. This concept has been
implemented in Europe where stricj: fTWi rpnmental and c-^np^tional health
regulations have forced lead smelters fcn mndp.rnifp fllP-V; nrnr.e,s.fif,fi .
The SB shaft furnace has an oblong crnss section unlike most cylindrical
secondary lead blast furnaces used in the United States. Figure 2 is a
diagram of the furnarp and associated gas treatment Astern. There are
tr-.u r«"iia of tuyeres, one on either side of the furnace, design;.! to use
air preheated to 500°C. The ("instruction is similar to a primary lead
blast furnace. The tuyeres have special coders which minimize fugitivt
emissions during punching.
The furnace is constructed so as to isolate the charging fln^r from the
bottom of the furnace. Thus, only the front :.nd loader operator forks in
a "dirty" area. Howe^tr, the front c.nd loader dop^ havt a filtered air
-------
3-
-r;T
LJ
itTT
_ r ..„
Figure 1. General process flowsheet for the
secondary lead smelting industry.
-------
STREAM KAMI
0+©
Sl/fiVACf IOf>
SEC.
'Af VEMT
ftOCfSS GAS
(*)+(+)
MZLOMfSATtON
•0
couawfo
EXIT GAS
t&C, LOMEKATIOfJ
IIIKFS VINT
Figure 2. Whole battery smelter furnace and flue gas treatment system.
-------
supply. In addition, the top of the blast furnace is well hooded so as to
minimize fugitive emissions.
At the bottom of the furnace, the four slag taps and one lead well are
hooded. The strong draft on these hoods virutally eliminates any fugitive
emissions. In addition, a special clay/wood plug minimizes the time
required for slag tapping.
The sanitary ventilation and process gases are mixed and all gases pass to
a single baghouse (at 100-125°C). The baghouse is a Swedish design using
felted polyester cloth. The dust is collected op, the outstrip flf, The fn&
and only a mild cleaning air jet is required to dislodge the d.us.l\ This
gives the bags an exceptionally long life.
The collected dust is conveyed in an enclosed system to one of two small
flash agglomeration furnaces. In this patented process, the dust is melted
and greatly reduced in volume. The agglomerated dust represents only 7 nr
3 percent by weight of the furnace charge rather than the normal 8 to 10
percent. This reduces the dust load circulating in the gas cleaning sys-
tem, increases production per square foot of furnace cross section, and
lowers energy consumption. The flash agglomerator furnace is oil fired
and consumes 24.6 liters of oil per hour.
The entire smelter area is p
-------
This smelter is serviced with two additional sewer collection systems
including one for rainfall and one for sanitary sewage. The rainfall col-
lected is of sufficient quality to be used directly as makeup cooling water
for the furnace. Additional makeup cooling water is obtained from the muni-
cipal water supply. This water is first softened in an ion exchange unit
before use. Sanitary sewage is discharged directly to the municipal col-
lection system.
13
-------
SECTION 4
SCOPE OF WORK
Separate sampling efforts are required for the NIOSH and EPA portions of
the study at the Glostrup, Denmark smelter. Both efforts use a similar
strategy with the following subtasks:
• Test Plan
Trip Preparation
• Sampling at Glostrup
Sample Analysis
Data Evaluation
• Reporting
Presentation of Results
Program Management
The following discussion describes the specific test plans for both the
NIOSH and EPA efforts. The other subtasks are described in a general
fashion.
TEST PLAN - NIOSH
A combination of personal breathing zone, work area, and grab sampling
techniques will be utilized to help characterize the effectiveness of
the hooding over the copper induction furnaces, the rotary "furnaces,
and the SB battery smelting furnace. An attempt will be made to evaluate
control during various phases of the smelting operations. Accordingly,
f
samples will be taken during specific time intervals to describe variations
related to operating phases. The following paragraphs describe the sampling
to be performed.
14 .
-------
Personal Monitoring
Copper Induction Furnaces—
Personal breathing zone samples will be obtained for the two furnacemen
normally assigned to the copper induction furnace. Sampling will commence
with the initial charging of the furnace and continue through the complete
furnace operating cycle. Sampling will terminate when pouring of the fur-
nace has been concluded. Two complete furnace cycles occurring during the
day shift of two separate working days will be monitored.
Each employee selected to participate in personal sampling will be asked to
wear a small battery powered pump with plastic hose and filter cassette
during his workshift. Filters will be changed at the completion of each
major phase in furnace operation (i.e. charging, refining, pouring).
Samples will be analyzed for metal particulates according to the methods
described in the appendix. Specifically, samples will be analyzed for copper,
lead, zinc and arsenic.
Rotary Furnaces—
Personal breathing zone samples will be obtained for the two furnacemen
normally assigned to charge preparation and the two or three furnacemen
assigned to tapping and pouring the rotary furnaces. Sampling will be con-
ducted for the entire furnace cycle with filters being changed at the com-
pletion of each major furnace operating phase. Two complete furnace cycles
occurring on two separate working days will be monitored.
The same type of sampling equipment as described for the copper induction
furnace will be used. Samples will be analyzed for lead, antimony, and
arsenic. A few selected samples will also be analyzed for copper, chlorine,
and fluorine.
15
-------
SB Battery Smelting Furnace—
Personal breathing zone samples will be obtained for the furnace charge
mixer/charger on the upper level of the furnace building, the approximately
five furnacemen who are assigned to slagging, tapping, and pouring on the
lower level. In addition, any other employees working in the whole battery
furnace area will be monitored. Sampling will be conducted for the entire
furnace cycle with filters being changed at the completion of each major
furnace operating phase. At least two complete furnace cycles occurring
on separate working days will be monitored.
The same type of sampling equipment as described for the copper induction
furnace. Samples will be analyzed for lead, arsenic, antimony, chlorine
and fluorine.
Area Sampling
Copper Induction Furnaces—
Work area samples will be obtained at two locations around the copper
induction furnaces. These locations will be selected upon arrival at
the smelter.
Area samples will be collected using similar equipment as described for
personal monitoring, located in stationary positions around the process of
interest. Area sampling will extend over the same work periods during the
same days as personal sampling. Filters will be changed at each station at
the conclusion of each major phase of furnace operation. Area samples will
be analyzed using the methods described in the appendix. Samples will be
analyzed for copper, lead, zinc and arsenic.
Rotary Furnaces—
Work area samples will be obtained at approximately two locations around
the rotary furnaces. These locations will be selected upon arrival at the
smelting site.
16
-------
The technique used for area sampling is similar to the personal monitoring
described for the copper induction furnaces. The samples collected will be
analyzed for lead, antimony, and arsenic. A few selected samples will be
analyzed for copper, chlorine, and fluorine.
SB Battery Smelting Furnace—
Work area samples will be obtained at a minimum of four locations around
the whole battery furnace. These locations include the upper charging level,
around the slag tapping area and around the bullion tapping area. Filters
will be changed at each station at the conclusion of each major phase of
furnace operation. In addition to the area samples taken around the whole
battery furnace, area samples may be taken around the flash agglomeration
furnace located adjacent to the baghouse.
These samples will be collected using similar equipment as described for
personal monitoring. Methods for chemical analysis of these samples are
described in the appendix. These samples will be analyzed for lead, arsenic,
antimony, chlorine and fluorine.
Grab Sampling
During the personal and area sampling periods described in the previous
sections, grab samples will be taken to determine the presence of gaseous
contaminants. These gaseous contaminants include carbon monoxide and
oxides of nitrogen. These contaminant concentrations will be measured
using NIOSH certified gas detector tubes.
TEST PLAN - EPA
The EPA portion of this program will focus only on the SB smelting furna.ce_
and the agglomeration furna.ce^, Table 1 is a schedule describing the order
and number of measurements which will be made for this characterization.
Four types of samples will be collected: stack gases and particulates,
solid inlet ajifl outlet streams, area gas particulates, and personal exposure^
17
-------
TABLE 1. SAMPLING SCHEDULE
STREAM NUMBER**
MEASUREMENT
Temperature Profiles
Velocity Profiles
Grain Loadings
in-stack filter
• impingers
Material Balance Samples (WEP)
Ambient Air Area Samples
Personal Monitoring
Agglomeration Hood Ventila-
tion Measurements
Slag Tap Hood Measurements
Lead Well Hood Measurements
Composite Solid Samples
Coke
Iron
Lime
Drosses
Batteries*
Plates
Mud
Acid
Case
Bullion
Slag
Speiss
Flue Dust
Day 1
11
11
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Day 2
11
11
11
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Day 3
11
11
11
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
* Battery component composition data will also be obtained from the
manufacturers.
** Stream numbers refer to those indicated in Figure 2.
18
-------
p articulates. Some of these samples will be those taken by NIOSH. In
addition, process stream temperature and flow rate measurements will
be made to determine how the process is performing.
The stack samples will allow controlled process emission factors to be cal-
culated. The solid stream samples will allow material balances to be conh-
pleted for six key elements: lead? antimony, arsenic, fluorine, chlorine,
and sulfur. The combined data will provide the information necessary to
determine the fate of these toxic materials as they pass through the
smelting process.
Fugitive emissions.,are,Mexp.er,r.Prt- to
However, both the area and personal samples taken by NIOSH will be used to
estimate over- the- fence lead levels which can be compared to similar measure-
ments made at smelters using different control technology.
The analysis of the samples taken should provide the data needed to calculate
emission factors, close material balances, and identify other potentially
toxic emissions not currently known. Table 2 lists the types of survey
and detailed techniques needed to complete these analyses for the furnace
feed. Particulate and gas samples taken at the stack will be analyzed using
the same techniques.
Grain Loadings - EPA Method 5
The standard EPA Method 5 pictured in Figure 3 will be used to measure
total stack particulate emissions. Each filter will be weighed for total
particulate mass and then analyzed for Pb, As, Sb, F, and Cl. An on-line
monitor will be used to measure SOa emissions.
Material Balance
Integral solid samples of each of the feed and discharge streams will be
collected during the sample period. The stack gas will be sampled using a
special sampling train which employs a wet electrostatic precipitator as
the major collection device.
19
-------
TABLE 2. ANALYTICAL SCHEDULE FOR COMPOSITE SOLID STREAM SAMPLES
Stream
Furnace Feed
Coke
Iron
Lime
Drosses
Batteries
Plates
Mud
Acid
Case
Bullion
Slag
Speiss
Flue Dust
Survey
SSMS***
SSMS
SSMS
SSMS
SSMS
SSMS
SSMS
SSMS
SSMS
SSMS
SSMS
" Analytical Techniques
X-Ray
Diffraction AA*
Pb,Sb,As
Pb,Sb,As
Pb,Sb,As
Pb,Sb,As
Pb,Sb,As
Yes Pb,Sb,As
Detailed
SIE**
E,CL
F,C1
F.C1
F,CL
F.C1
F,C1
ICEDt
S
S
S
S (GBP)tt
S
S
S
S
S
* Atomic Absorption
** NaaCOa fusion, HaO leach, specific ion electrode
*** Spark source mass spectrometry
t Ion chromatography of an Eschka digestion
tt Gravimetric barium precipitation
-------
Vapor Collection
Probe Assembly
PROBE
REVEHSE-TVPE
P1TOT TUOE
PITOT MANOMETER
Particulate
Collection
THERMOMETER
FILTER
HOLDER
THERMOMETERS
IMPINOER TRAIN OPTIONAL. MAY BE REPLACED
BY AN EQUIVALENT CON06NSER
VACUUM
LINE
•MAIN
VALVE
DRY QAS METER
Metering
Figure 3. EPA 5 sampling train.
The wet electrostatic precipitator (WEP) is used to collect particulates
and acid mist. 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 4. .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 voltage of
12-15 KV-DC causes a corona discharge at the center electrode.
21
-------
HIGH VOLTAGE
POWEfl SUPPLY
Falling Film o
Elaccrolvca
PERISTALTIC
PUMP
CIRCULATING
ELECTROLYTE
RESERVOIR
Figure 4. Wet electrostatic precipitator.
The gas entering the WEP is first scrubbed and cooled in the round bottom
flask. Participates 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.
22
-------
The wet electrostatic precipitator was integrated into the sampling device
shown in Figure 5. The probe consists of a pyrex nozzle and is pyrex lined.
Teflon tubing is used to connect it with the WEP. All the lines are rinsed
after sampling. The WEP is followed with eight impingers in an ice bath.
The charge of the impingers is summarized in Table 3. A pump and a dry gas
meter complete the assembly.
PREPARATION OF SAMPLING TRIP
Radian will supply all the equipment to perform the sampling at the Glostrup
smelter. The gear will be assembled, checked, packed and shipped to
Glostrup. Changes will be made and any special equipment necessary will
be purchased or built during this program phase.
SAMPLING AT GLOSTRUP
Sampling at Glostrup is expected to take approximately seven days, with two
days for equipment set-up, check-out and take down. The main sampling effort
will take four days.
SAMPLE ANALYSES
A spark source mass spectrometry survey analysis will be performed on the
feed material. Level of concentration and toxicity will be used as guide-
lines for selecting those elements to be included in the material balance
calculations. At present, six substances are thought to be of interest.
These are: lead, arsenic, antimony, fluorine, chlorine, and sulfur.
Radian recommends the analytical procedures discussed in Appendix B.
These methods proved to be successful in a balance attempt around the
electrostatic precipitator servicing a reverberatory furnace. Selected
samples will be subjected to spark source mass spectrometry analysis.
Scanning electron microscopy and X-ray diffraction will be used to further
characterize the samples if warranted.
23
-------
Acid Itnplngers.
Caustic Impingers
Pyrex Pyrex Lined Probe
Nozzle
Teflon
Tubing
Wet Electrostatic
Precipitator
llydroRcnperoxido
Impingcr
VVT;. "•'&/??',-• rV
•: ?K- '.• .•.••:.><; ::•:•.': :v J
Ice Bath
\x
Dry
Impingers
Silca Gel
Impinger
Fine
Adjustment Valve
Coarse
Adjustment Valve
Pump
Figure 5. Schematic of the integral WEP sampling train.
TABLE 3. - IMPINGER SOLUTIONS COMPOSITIONS
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
potassium hydroxide in a Smith-Greenburg
impinger
hydrogen peroxide in a Smith-Greenburg impinger
preweighed silica gel in a modified Smith-
Greenburg impinger
24
-------
DATA EVALUATION
The analytical results and the flow rates will be used to establish the
material balances. Error propagation calculations will bracket the con-
fidence limits of the results generated for the EPA measurements.
The EPA data evaluation part of the study includes the selection of elements
present in minor concentrations in the smelter feed for detailed analysis.
Lead, the major element important in the smelting process, will be used in
the analytical scheme in order to check the accuracy of flow rate data.
Spark source mass spectrometry will be used as the technique to analyze
the element combination in the smelter feed. This technique provides semi-
quantitative results. It has the advantage of covering 73 elements of the
periodic chart. From these data, elements will be selected for quantitative
determination.
Atomic absorption, x-ray fluorescence, ion specific electrodes, fluorometry,
spectrophotometry as well as wet chemical methods will be used for quantita-
tive elemental analysis. Detailed analytical methods employed by Radian are
described in Appendix B.
Errors are inherent to both the chemical analysis results and the flow
rates of the streams sampled. Both errors propagate if the chemical con-
centration of an element is multiplied by the flow rate of the stream to
give the flow rate of an element. Similarly, the errors propagate if a
mass balance is established for a minor element.
The error propagation analysis used to establish error limits for the
calculated total inlet and outlet mass rates of each element is described
below. The values indicate the degree of variance to be expected due to
random errors and should not be interpreted as variance over a set of data
points.
25
-------
A 95 percent confidence interval of 2S(Q) is calculated for a given value
of Q according to the following standard definition:
s'(d> - s f^ s*(qi>
where, S(Q) = the variance in Q,
Q = the mas flow rate of a given element into or out
of the system,
q. = the i— independent measured value
(stream flow or elemental concentration),
and
S(q.) - the variance of q..
The analytical results from the air samples taken, the engineering data
gathered, and the visual observations made will be used to qualitatively
evaluate the engineering controls for NIOSH at the smelter. Quantitative
results expected will be the actual employee exposure to hazardous agents
at various workplaces and the general workroom concentrations measured during
different phases of operations. Comparisons of actual engineering data to
design specifications will determine if the controls are operating as ex-
pected. Documentation of maintenance procedures as well as observations of
employee interaction with the equipment will also contribute to the overall
control equipment evaluation.
REPORTING
A final report will be produced conforming to the format and instructions
provided by the NIOSH and EPA project officers. A copy of the draft report
will be reviewed by the Bergs^e smelter management prior to release to insure
the technical accuracy of all process descriptions and to avoid disclosing
any proprietary information. The report will contain both the NIOSH and EPA
sampling results. A final copy of the report will be issued after all tech-
nical and editorial comments are received.
26
-------
PRESENTATION OF RESULTS
Radian's project technical director will present the results of the sampling
effort to both EPA and NIOSH after submittal of the draft final report.
The reason for this meeting is to clarify all questions concerning the
draft final report.
PROGRAM MANAGEMENT
Effort devoted in this task is aimed to keep close liaison to the NIOSH and
EPA program officers, to the Bergs^e smelter management and also to coordi-
nate Radian's in-house effort.
27
-------
SECTION 5
SCHEDULE AND MILESTONES
Timely completion of this study will depend heavily on the operation of the
plant during the sampling period. Plant upsets or delays caused by customs
or sample analysis could cause delays. At present, the plan calls for the
schedule shown in Figure 6. The milestones and expected completion dates
are shown in Table 4.
TABLE 4. MILESTONES AND COMPLETION DATE
Milestone Description Completion Date
M-l Preliminary Test Plan July 17, 1978
M-2 Final Test Plan July 31, 1978
M-3 Trip Preparation August 31, 1978
M-4 Field Sampling September 22, 1978
M-5 Sample Analysis October 30, 1978
M-6 Data Analysis November 30, 1978
M-7 Draft Final Report December 15, 1978
M-8 Final Report January 31, 1978
M-9 Presentation of Results February 9, 1978
28
-------
VD
Subtask
Test Plan
Trip Preparation
Sampling
Sample Analysis
Data Analysis
Reporting
Presentation of Results
Program Management
M-l M-2
? T
M-3
f
M-4
T
M-5
T
M-6
T
M-7
f
M-8
T
M-9
f
July'Aug.' Sept.Oct.'Nov. 'Dec.'Jan.Feb.
Figure 6. Test program schedule.
-------
APPENDIX A
NIOSH SAMPLING AND ANALYTICAL TECHNIQUES
SAMPLING TECHNIQUES AND EQUIPMENT
Personal and Area Monitoring
Personal and area monitoring is performed using small battery powered pumps
fitted with filters to collect any metal fume which is present in the work-
room atmosphere. For personal monitoring, the pump is worn by the employee
and the filter worn in the vicinity of the breathing zone. For area sampling,
the pump and attached filter are located in a stationary position. The
sampling techniques for the elements of interest in this study, i.e., copper,
lead, arsenic, zinc, antimony, tin are described in the NIOSH Manual of
Sampling Data Sheets as follows:
General Procedure for Metals —
Method - A known volume of air is drawn through a cellulose membrane filter
to trap the metal fume present. The filter is digested with an acidic solu-
tion to destroy the filters and any organic material. The analyte is then
solubilized in a suitable acid solution (see Analytical Methods), and aspi-
rated into an atomic absorption spectrophotometer for determination of con-
centration.
Sampling Equipment-- A calibrated personal sampling pump whose flow can be
determined to an accuracy of ±5% over the range of 1.0 to 2.0 liters per
minute; a 37-mm three-piece cassette filter holder and a 37-mm diameter
0.8-micrometer pore size mixed cellulose ester membrane filter (MCEF) sup-
ported by a cellulose backup pad.
Sample Size - A sample size of 250 liters is recommended. Sample at a flow
rate of 1.5 liters per minute.
A-l
-------
Sampling Procedure - Assemble the filter and three-piece filter cassette and
close firmly to insure that the center ring seals the edge of the filter.
Examine the holder for a good filter seal. If the cassette will not seal
tightly, it should be discarded. Secure the cassette holder together with
tape or shrinkable band.
Remove the cassette plugs and attach to the personal sampling pump tubing.
Clip the cassette to the worker's lapel.
Air being sampled should not be passed through any hose or tubing before
entering the filter cassette.
Set the flow rate as accurately as possible using the manufacturer's direc-
tions. Record the temperature and pressure of the atmosphere being sampled.
If the pressure reading is not available, record the elevation. Position
the middle of the rotameter ball of the personal sampling pump to the 1.5-
liter per minute calibration mark as accurately as possible. Since it is
possible for the filter to become plugged by heavy particulate loading or
by the presence of oil mists or other liquids in the air, the pump rotameter
should be observed frequently, and readjusted as needed. If the rotameter
cannot be adjusted to correct a problem, terminate the sampling.
Cassettes containing collected samples should be firmly sealed with the
plugs in both the inlets and outlets.
Carefully record sample identity and all relevant sample data.
Blank. With each batch of ten samples, submit one filter from the same lot
of filters which was used for sample collection and which is subjected to
exactly the same handling as the samples except that no air is drawn through
it. Label this as a blank.
A-2
-------
Procedure for Hydrogen Fluoride —
Method - A known volume of air is drawn through a midget bubbler containing
10 ml of 0.1 N sodium hydroxide to trap hydrogen fluoride. The resulting
solution is diluted to 25 ml with 0.1 N sodium hydroxide and buffered with
an equal volume of total ionic strength activity buffer (TISAB). The sample
is analyzed using a fluoride ion specific electrode and an expanded scale
millivolt/pH meter. The method has been validated over the range of 1.33-
4.50 mg/cu m for a 45-liter sample at an atmospheric temperature and pressure
of 22°C and 761 mm Hg.
Sampling Equipment - A calibrated personal sampling pump whose flow can be
determined accurately, ±5%, at 1.5 liters per minute, plus midget bubbler
with 1/8-in O.D. Teflon inlet tube, containing 10 ml of 0.1 N sodium hydrox-
ide. A prefilter unit, consisting of a 37-mm/0.8-micrometer mixed cellulose
ester membrane filter and a polystyrene cassette filter holder, is connected
in front of the bubbler to collect fluoride particulate that may be present.
A backup pad should not be used. Care must be taken to insure that the fil-
ter is sealed tightly to avoid air leaks during sampling. The sampling pump
is protected from splashover by a 5-cm long by 6-mm I.D. glass adsorption
tube loosely packed with a plug of glass wool and inserted between the exit
arm of the bubbler and the pump. (An additional 100 ml of collection medium
should accompany each set of bubblers for use in rinsing the bubbler stems
after sampling).
Sample Size - A sample size of 45 liters, taken at a flow of 1.5 liters per
minute for 30 minutes, is recommended.
Sampling Procedure - Connect the midget bubbler (containing 10 ml of the
collection medium) with a 5-cm glass adsorption tube containing the glass
wool plug to the prefilter unit and the personal sampling pump using short
pieces of flexible tubing. The minimum amount of tubing necessary to make
the joint between the prefilter and bubbler should be used.
A-3
-------
The bubbler must be maintained in a vertical position during sampling.
Air being sampled should not be passed through any hose or tubing before
entering the bubbler.
Set the flow rate as accurately as possible using the manufacturer's direc-
tions. Position the middle of the rotameter ball of the personal sampling
pump to the 1.5 liter per minute calibration mark as accurately as possible.
Record the temperature and pressure of the atmosphere being sampled. If the
pressure reading is not available, record the elevation.
After sampling, the bubbler stems may be removed and cleaned. Tap the stems
gently against the inside walls of the bubbler bottles to recover as much of
the sampling solution as possible. Wash the stem with 1 ml of 0.1 N sodium
hydroxide, adding the wash to the bubbler. Transfer the contents of the
bubbler to a 50-ml polyethylene bottle. Rinse the bubbler vessel with 2-3 ml
of 0.1 N sodium hydroxide and add the wash to the bottle. Seal tightly for
shipment.
Care should be taken to minimize spillage or loss by evaporation.
A "blank" bubbler should be handled in the same manner as the bubblers con-
taining the samples (fill, seal, and transport) except that no air is
sampled through this bubbler.
The filter may be discarded since the method is designed to measure gaseous
hydrogen fluoride only.
\
Procedure for Hydrogen Chloride —
Method - A known volume of air is drawn through a midget bubbler containing
10 ml of 0.5 M sodium acetate to trap hydrogen chloride. The resulting solu-
tion is diluted to 25 ml with distilled water. The sample is analyzed using
a chloride ion specific electrode and an expanded scale millivolt/pH meter.
A-4
-------
The method has been validated over the range of 3.5-14 mg/cu m for a 15-liter
sample at an atmospheric temperature and pressure of 22°C and 764 mm Hg.
Sampling Equipment - A calibrated personal sampling pump whose flow can be
determined accurately, ±5%, at 1.0 liter per minute, plus midget bubbler,
containing 10 ml of 0.5 M sodium acetate. The sampling pump is protected
from splashover by a 5-cm long by 6-mm I.D. glass splashover tube loosely
packed with a plug of glass wool and inserted between the exit arm of the
bubbler and the pump. (An additional 100 ml of collection medium should
accompany each set of bubblers for use in rinsing the bubbler and bubbler
stems after sampling.) Samples are shipped in polyethylene containers.
Sample Size - A sample size of 15 liters, taken at a flow of 1.0 liter per
minute for 15 minutes, is recommended.
Sampling Procedure - Connect the midget bubbler (containing 10 ml of the
collection medium) with a 5-cm glass splashover tube containing the glass
wool plug to the personal sampling pump using short pieces of flexible
tubing.
The bubbler must be maintained in a vertical position during sampling.
Air being sampled should not be passed through any hose or tubing before
entering the bubbler.
Set the flow rate as accurately as possible using the manufacturer's direc-
tions. Position the middle of the rotameter ball of the personal sampling
pump to the 1.0 liter per minute calibration mark as accurately as possible.
Record the temperature and pressure of the atmosphere being sampled. If the
pressure reading is not available, record the elevation.
After sampling remove the bubbler stem and transfer the contents of the
bubbler to a polyethylene container. Rinse the bubbler and bubbler stem
A-5
-------
with 3-5 ml of collection medium, adding the rinse to a polyethylene
container. Seal the polyethylene container with the associated caps just
prior to shipment.
Care should be taken to minimize spillage or loss by evaporation.
A "blank" bubbler should be handled in the same manner as the bubblers con-
taining the samples (fill, seal, and transport) except that no air is
sampled through this bubbler.
Grab Sampling
Grab sampling for gaseous contaminants such as carbon monoxide will be done
using NIOSH approved detector tubes attached to a hand operated pump. This
method is described in the NIOSH Manual of Sampling Data Sheets as follows:
Sampling Equipment —
Certified or NIOSH approved detector tubes and a hand operated pump (not
rubber bulb type) may be used for compliance purposes. (Secondary Method)
Analytical Procedure —
Detector tubes are read for length of stain, strokes of pump are noted, and
appropriate corrections are made for pressure (or altitude above 5,000 feet
(1,500 m). The tubes are not to be used when below 0 degrees F (-18°C) or
above 125 degrees F (52°C).
Sampling Period —
Depends on portable instrument used.
Detector tubes: direct reading, length-of-stain detector tubes (certified
or NIOSH approved) may be used for compliance purposes, if at least one
sample per hour is taken per exposed worker or close group of workers. In
an 8-hour day more frequent sampling per hour is preferred, especially for
A-6
-------
exposures over 50 ppm. Eight different samplings, one per hour, is
considered the minimum acceptable number.
ANALYTICAL METHODS
The following paragraphs describe the chemical analysis methods which will
be used to analyze the filter samples taken during both personal and area
monitoring. These methods are taken from the NIOSH Manual of Analytical
Methods with modifications to the methods as noted. For specific information
for each element of interest, see Table A-l.
General Procedure for Metals
Principle of the Method —
Samples are ashed using nitric acid to destroy the organic matrix and the
metals are solubilized in an acidic solution maintaining a pH of 1.
Samples and standards are aspirated into the appropriate AA flame. A hollow
cathode lamp for each metal of interest provides the characteristic line for
that particular metal. The absorption of this line by the ground state atoms
in the flame is proportional to the metal concentration in the aspirated
sample.
Range and Sensitivity —
The optimum working range for each metal is given in the Table. This value
can be extended to higher concentrations by dilution of the sample.
The sensitivity of this method for each metal in aqueous solution is also
given in the Table. This value will vary somewhat depending upon the
instrument used.
A-7
-------
Table A-l. DATA TABLE FOR METAL ANALYSIS
Element Sensitivity
(yg/ml)
Be
Ca
Cd
Co
Cr
Cu
Fe
K
Mg
Mn
Mo
Na
Ni
Pb
Sn
Sn (alt)
Sb
Zn
As
As (alt)
F
Cl
0.03
0.07
0.04
0.15
0.2
0.1
0.1
0.1
0.007
0.05
1.0
0.02
0.2
0.5
5.0
0.025
Hydride
Range of
Method
(yg/ml)
0.03-8
0.07-10
0.04-5
0.15-8
0.2-10
0.1-10
0.1-10
0.1-10
0.007-0.7
0.05-4
1.0-60
0.02-5
0.2-20
0.5-30
5.0-300
0.025-27
Generation
Method No.
P & CAM 173
P & CAM 173
P & CAM 173
P & CAM 173
P & CAM 173
P & CAM 173
P & CAM 173
P & CAM 173
P & CAM 173
P & CAM 173
P & CAM 173
P & CAM 173
P & CAM 173
P & CAM 173
P & CAM 173
5183
P & CAM 173
P & CAM 173
P & CAM 173
Method
P & CAM 117
5246
Analytical
Wavelength
(A)
2349
4227
2288
2407
3579
3247
2483
7665
2852
2795
3133
5890
2320
2170,2833
2246,2354
5183
2139
Modifications
to Method
perchloric acid used
to aid digestion
perchloric acid used
to aid digestion
perchloric acid used
to aid digestion
terniary acid used
to aid digestion:
4:4:1 of H2SOi» , HN03 ,
perchloric acid used
to aid digestion
A-8
-------
Interferences —
Known interferences may occur when analyzing for metals listed in the table.
Therefore, procedures for eliminating or overcoming these interferences are
listed. Whenever additions are made to the samples to overcome or eliminate
interferences, similar additions must be made to the standards.
Chemical interferences in the flame prevent conversion of the metal being
determined to the atomic state. Higher flame temperatures (NaO-CaHa) can
overcome this problem in some cases.
Precision and Accuracy —
In general, this method will provide a coefficient of variation for the
analysis of approximately 2% depending upon the instrument used and the
absorbance of the samples. (If absorbence is less than 0.1, the coefficient
of variation is higher.)
No data on accuracy is available at this time.
Advantages and Disadvantages of the Method —
The sensitivity is adequate for all metals in air samples but only for cer-
tain metals in biological matrices. The sensitivity of this direct aspira-
tion method is not adequate for Be, Cd, Ca, Cr, Mn, Mo, Ni, and Sn in
biological samples.
A disadvantage of the method is that at least 1 to 2 m£ of solution is
necessary for each metal determination. For small samples, the necessary
dilution would decrease sensitivity.
Apparatus —
Sampling Equipment - The sampling unit for the collection of personal air
samples for the determination of metal content has the following components:
A-9
-------
• The filter unit, consisting of the filter media (Section 6.2)
and appropriate cassette filter holder, either a 2- or 3-piece
filter cassette (Millipore Filter Corporation, Bedford,
Massachusetts).
• A vacuum-pump such as a personal sampling pump. This pump
must be properly calibrated so the volume of air sampled
can be. measured as accurately as possible. The pump must
be calibrated with a representative filter unit in the line.
Flow rate, times, and/or volume must be known.
• Thermometer.
• Manometer.
• Stopwatch.
• Various clips, tubing, spring connectors, and belt for
connecting sampling apparatus to worker being sampled.
0.8y cellulose membrane filter (or equivalent), 37 mm.
Hollow cathode lamps for each metal.
Atomic absorption spectrophotometer, having a monochromotor with a reciprocal
o
linear dispersion of about 6.5 A/mm in the ultraviolet region. The instru-
ment must have the necessary burner heads for air-acetylene and nitrous
oxide-acetylene flames.
Oxidant:
• Air, which has been filtered to remove water, oil, and
other foreign substances is the usual oxidant.
• Nitrous ,oxide is required as an oxidant when higher
temperatures are required in the analysis of
refractory-type metals.
A-10
-------
Fuel - Acetylene, commercially available for atomic absorption use.
Pressure-reducing valves - A 2-gauge, 2-stage pressure reducing valve and
appropriate hose connections are needed for each compressed gas tank used.
When using nitrous oxide, heating tape, with the temperature controlled by
a rheostat, is wound around the second stage of the regulator and connecting
hose to prevent freeze-up of the line.
Glassware, borosilicate:
• 125 mH Phillips beakers with watchglass covers
• 15 m£ graduated centrifuge tubes
• 10 and 100 m£ volumetric flasks
• 125 mJl polyethylene bottles
Three-switch hot plates capable of reaching 400°C.
Reagents —
All reagents used must be ACS Reagent Grade or better.
Double Distilled or deionized water
Redistilled concentrated nitric acid
Distilled 1:1 hydrochloric acid
Commercially prepared aqueous stock standards (1000 yg/m£) for each metal
listed in the Table.
Lanthanum stock solution (for Ca determination) - 5% La in 25% HC1 (v/v).
Wet 29.33 g La20a with double distilled water. Add 125 mJl concentrated HC1.
Dilute to 500 mJL
A-ll
-------
Aluminum solution (for Mo determination) - commercially prepared aqueous
aluminum stock containing 1000 yg Al/mS,.
Procedure —
Cleaning of Equipment:
• Before initial use, glassware is cleaned with a saturated
solution of sodium dichromate in concentrated sulfuric acid
(Note: Do not use for chromium analysis) and then rinsed
thoroughly with warm tap water, concentrated nitric acid,
tap water and deionized water, in that order, and then dried.
• All glassware is soaked in a mild detergent solution
immediately after use to remove any residual grease or
chemicals.
• For glassware which has previously been subjected to the
entire cleaning procedure, it is not necessary to use the
chromic acid cleaning solution.
Collection and Shipping of Samples:
• Dusts and fumes containing metal can be sampled with a 0.8 y
cellulose membrane filter. The filters must not be loaded
to the point where portions of the sample might be dislodged
from the collecting medium during handling. Personal filter
samples should be sealed in individual plastic filter holders
during shipment. A two-hour sampling period at 1.5 liters
per minute will provide enough sample for air concentrations
of 0.2 x TLV. Beryllium requires a full eight-hour sample
at 0.2 x TLV.
• Blood samples - 10 mH should be collected in chemically
clean, heparinized vacutainers. If the vacutainers have
not been pretreated, 1-2 mJl of a heparin sodium solution
should be injected into the vacutainer. Refrigerate for
shipment if possible.
A-12
-------
Urine samples - 50 m£ should be collected in chemically clean,
borosilicate or polyethylene bottles. The urine samples should
be preserved by the addition of approximately 2.5 mg of thymol
and refrigerated for shipment if possible.
Tissue samples should be collected in chemically clean jars
and preserved in dry ice. Tissue samples must be shipped
back to the laboratory immediately.
Analysis of Samples:
Samples are transferred to clean 125 m& Phillips beakers and
several mil of concentrated HNOa is added to each. Each beaker
is covered with a watchglass and heated on a hot plate (140°C)
in a fume hood until the sample chars or until a slightly
yellow solution remains. Several additions of HNOs may be
needed to completely ash and destroy the organic material.
Completion of the digestion procedure is indicated by a
white residue in the beaker.
For samples containing arsenic (a fairly volatile element),
the residue is dissolved in 2 m£ of 6N HC1 using the low
temperature hot plate (140°C) and quantitatively transferred
to a graduated centrifuge tube with deionized water.
For samples not containing arsenic, the residue is heated
several minutes on the high temperature hot plate (400°C)
and converted to a salt by three successive evaporations
with 1:1 HC1 or concentrated HNOa. The ash is then dissolved
with 1:1 HC1 or concentrated HNOa and deionized water and
quantitatively transferred to a graduated centrifuge tube.
Aliquots of this can be diluted if necessary or the volume
can be reduced by evaporation to get the metal concentration
within the working range of the method.
A-13
-------
• The sample solution is then aspirated into the appropriate
flame for each metal as indicated in the Table. The analytical
wavelength is also listed in the Table. The other operating
parameters are set according to the instrument manufacturer's
conditions for each metal being determined. The absorbance of
each sample is recorded. When very low metal concentrations
are found in the sample, scale expansion can be used to increase
instrument response.
Calibration and Standards —
From each of the 1000 yg/m£ stock metal standard solutions, prepare working
standards to cover the range for each metal as indicated in the Table. All
standard solutions are made 0.3 N in HC1 and are stored in polyethylene
bottles. The low concentration standards may deteriorate and should be
remade each day.
When analyzing for any of the metals where interferences are known to occur,
as is indicated in the Table, standards should be prepared according to the
"Remedy" listed in the Table.
Aspirate the series of standards and record the percent absorption.
Prepare a calibration curve by plotting on linear graph paper the absorbance
versus the concentration of each standard in yg/m£. It is advisable to run
a set of standards both before and after a sample run to insure that condi-
tions have not changed.
Calculations —
From the calibration curve, read the concentration (yg/mJl) in the analysis
sample.
Blank values, if any, are subtracted from each sample.
A-14
-------
The concentration of the metal in the original sample is:
jag metal/mil sample = yg/m£ x dilution factor
where:
yg/m£ = metal concentration .determined from the appropriate
calibration curve (Section 10.1).
Fluorides and Hydrogen Fluoride in Air
Principle of the Method —
Atmospheric samples are taken using midget impingers containint 10 m£ of
0.1M NaOH.
Samples are diluted 1:1 with Total Ionic Strength Activity Buffer (TISAB).
The diluted samples are analyzed using the fluoride ion specific electrode.
Range and Sensitivity —
The range and sensitivity have not been established at this time. The
recommended range of the method is 0.009-95 mg/m air.
Interferences —
Hydroxide ion is the only significant electrode interference; however,
addition of the TISAB eliminates this problem. Very large amounts of
complexing metals such as aluminum may result in low readings even in
the presence of TISAB.
Precision and Accuracy —
The accuracy and precision of this method have not been completely determined
at this time. No collaborative tests have been performed on this method.
A-15
-------
Advantages and Disadvantages of the Method —
Advantages over previous methods include simplicity, accuracy, speed,
specificity and elimination of distillation, diffusion and ashing of
the samples.
No significant disadvantages are known at present.
Apparatus —
Sampling Equipment - The sampling unit for the impinger collection method
consists of the following components:
• A prefilter unit (if needed) which consists of the filter
media and cassette filter holder.
• A midget impinger containing the absorbing solution or reagent.
• A pump suitable for delivering desired flow rates. The sampling
pump is protected from splashover or water condensation by an
adsorption tube loosely packed with a plug of glass wool and
inserted between the exit arm of the impinger and the pump.
• An integrating volume meter such as a dry gas or wet test meter.
• Thermometer.
• Manometer.
• Stopwatch.
Orion Model 94-09 Fuoride Specific Ion Electrode, or equivalent.
Reference Electrode, Orion 90-01 single junction, or equivalent calomel or
silver/silver chloride electrode.
Expanded Scale Millivolt-pH Meter, capable of measuring to within 0.5
millivolt.
A-16
-------
Polyethylene Beakers, 50-m£ capacity.
Laboratory Glassware.
Magnetic Stirrer and Stirring bars for 50-m£ Beakers.
Reagents —
All chemicals must be ACS reagent grade or equivalent. Polyethylene beakers
and bottles should be used for holding and storing all fluoride-containing
solution.
Double Distilled Water.
Glacial Acetic Acid.
Absorbing Solution: 0.1M Sodium Hydroxide Solution. Dissolve 4 g sodium
hydroxide pellets in 1 liter distilled water.
Sodium Hydroxide, 5M Solution, Dissolve 20 g sodium hydroxide pellets in
sufficient distilled water to give 100 m£ of solution.
Sodium Chloride.
Sodium Citrate.
Total Ionic Strength Activity Buffer (TISAB). Place 500 m£ of double dis-
tilled water in a 1-liter beaker. Add 57 m£ of glacial acetic acid, 58 g
of sodium chloride and 0.30 g of sodium citrate. Stir to dissolve. Place
beaker in a water bath (for cooling) and slowly add 5M sodium hydroxide
until the pH is between 5.0 and 5.5. Cool to room temperature and pour
into a 1-liter volumetric flask and add double distilled water to the mark.
Sodium Fluoride, for preparation of standards.
A-17
-------
Standard Fluoride Solution
• Dissolve 4.2 g of sodium fluoride in double distilled water
and dilute to 1 liter. This solution contains 10"1 M[F~]
(1900 ygF~/m£). The 0.1M fluoride solution may also be
purchased from Orion Research, Inc., Cambridge, Mass.
• Prepare 10~2 M[F~] by diluting 10 m£ of 10-1 M[F~] to
100 m£ with double distilled water (190 ygF~/m£).
• Prepare 10~3 M[F~] by diluting 10 mi of 10~2 M[F~] to
100 m£ with double distilled water (19 ygF~/m£).
• Prepare lO"" M[F~] by diluting 10 m£ of 10~3 M[F~] to
100 m£ with double distilled water (1.9 ygF~/m£).
• Prepare 10~5 M[F~] by diluting 10 mH of IQ-* M[F~] to
100 m£ with double distilled water (0.19 ygF~/m£).
Procedure —
Cleaning of Equipment. All glassware and plastic ware are washed in
detergent solution, rinsed in tap water, and then rinsed with double
distilled water.
Collection and Shipping of Samples:
• Pour 10 mH of the absorbing solution (section 7) into the midget
impinger, using a graduated cylinder to measure the volume.
• Connect the impinger (via the adsorption tube) to the vacuum
pump and the prefilter assembly (if needed) with a short
piece of flexible tubing. The minimum amount of tubing
necessary to make the joint between the prefilter and impinger
should be used. The air being sampled should not be passed
through any other tubing or other equipment before entering
the impinger.
A-18
-------
• Turn on pump to begin sample collection. Care should be
taken to measure the flow rate, time and/or volume as
accurately as possible. The sample should be taken at a
flow rate of 2.5 Upm. A sample size of not more than 200
liters and no less than 10 liters should be collected. The
minimum volume of air sampled will allow the measurement at
least 1/10 times the TLV, 0.2 mg/m3 (760 mm Hg, 25°C).
• After sampling, the impinger stem can be removed and cleaned.
Tap the stem gently against the inside wall of the impinger
bottle to recover as much of the sampling solution as possible.
Wash the stem with a small amount (1-2 m ) of unused absorbing
solution and add the wash to the impinger. Then the impinger
is sealed with a hard, non-reactive stopper (preferably Teflon).
Do not seal with rubber. The stoppers on the impingers should
be tightly sealed to present leakage during shipping. If it is
preferred to ship the impingers with the stems in, the outlets
of the stem should be sealed with Parafilm or other non-rubber
covers, and the ground glass joints should be sealed (i.e.,
taped) to secure the top tightly.
• Care should be taken to minimize spillage or loss by evaporation
at all times. Refrigerate samples if analysis cannot be done
within the day.
• Whenever possible, hand delivery of the samples is recommended.
Otherwise, special impinger shipping cases designed by NIOSH
should be used to ship the samples.
• A "blank" impinger should be handled as the other samples
(fill, seal and transport) except that no air is sampled
through this impinger.
• Where a prefilter has been used, the filter cassettes are capped
and placed in an appropriate cassette shipping container. One
filter disc should be handled as the other samples (seal and trans-
port) except that no air is sampled through, and this is labeled as
a blank.
A-19
-------
Analysis of Samples :
• The sample is transferred from the impinger to a 50-cc
plastic beaker; an equal volume of TISAB is added and the
solution is stirred.
• The fluoride ion electrode and the reference electrode are
lowered into the stirred solution and the resulting millivolt
reading recorded (to the nearest 0.5 millivolt) after it has
stabilized (drift less than 0.5 millivolt per minute).
Calibration and Standards —
Prepare a series of fluoride standard solutions by diluting equal volumes
of each fluoride standard (7.9) and TISAB in a clear polyethylene beaker.
Insert the "fluoride ion electrode and the reference electrode into each
of the stirred calibration solutions starting with the most dilute solu-
tion and record the resulting millivolt reading to the nearest 0.5 mill-
volt. Plot the millivolt readings vs the fluoride ion concentration of
the standard of semi-log paper. The fluoride ion concentration in yg/m£
is plotted on the log axis. The calibration points should be repeated
twice daily.
Calculations —
The concentration (yg/m£) of fluoride in the sample solution is obtained
from the calibration curve.
Total yg F~ in the sample = sample concentration (yg/m£) x sample solution
volume
The total ygF~ is divided by the volume in liters, of air sampled to
obtain concentration in ygF~/liter or mg3F~/m&.
A-20
-------
mgF~/m3 = UgF~/liter
moir-/m3 _ total
mgF /m - ^— (section 10.4)
Convert the volume of air sampled to standard conditions of 25°C and 760 mm Hg.
P 298
= V X T77T X
760 T + 273
where :
V = volume of air in liters at 25°C and 760 mm Hg
S
V = volume of air in liters as measured
P = barometric pressure in mm Hg
T = temperature of air in degree centigrade
The concentration can also be expressed in ppm, defined as y& of component
per liter of air.
ppmF- = y£F-/Vs = - x ygF-/Vg
=1.29 gF~V
S
Where:
24.45 = molar volume at 25°C and 760 mm Hg
MW = 19 , weight of fluoride ion,
(i.e., 19 UgF~ = 24.45 \il at 25°C, 760 mm Hg)
To calculate the concentration of hydrogen- fluoride as mg HF/m or ppm HF,
simply multiply the corresponding concentration of F~ (from 10.3 or 10.4)
by 1.05.
A-21
-------
APPENDIX B
METHODS FOR CHEMICAL ANALYSIS
This Appendix presents the methods and techniques used for chemical charac-
terization and analysis of the smelter sampling data.
ELEMENTAL METHODS
Quantitative analysis consists of two major steps:
• sample dissolution, and
• chemical analysis.
Sample dissolution techniques include acid reflux digestion, perchloric acid
digestion and lithium borate fusion.
The techniques used for the quantitative determinations of trace elements in
samples collected from the copper smelter are based are:
• atomic absorption,
• ion selective electrode, and
• fluorometry.
Figures 1 through 3 summarize the dissolution and analytical procedures used
for the trace element determinations. The various dissolution and analytical
procedures are described in the remainder of this section.
Sample Preparation —
The dissolution techniques applied to the samples are:
B-l
-------
Klue gas.
Wee Electrostatic Pracigitacor ('JE?) Liquor
___________ Au. Anal. __________ F AA.
-as
AA:
ARC:
DCS:
FAA:
HGA-AA:
ICE:
OE:
PAD:
SA:
SIE:
Solids by
PAD
.3s, Mo
SA/DCS.
ICE.
SA/DCS.
HGA-AA-
.AA.
Emission
'luorotnetT'/
SIE
• 3b, Cd
Si, ?b
_u r,
Zn
.As
-3a
acooic absorpcicn, flans
acid reflux digestion
double capillary syscea
flanelass atonic absorption
heaced graphite analyzer of cha atomic absorption specrrophocoaeca
inorganic coaplex excraccion
organic extraction
perchloric acid digestion
standard additions
specific ion electrode
Figure B-l. Dissolution and analytical scheme of a WEP slurry.
B-2
-------
IMPISGE3. SOLUTION'S
-lapinger Solution
I_:pir.ger Solution
AA: atomic absorption, flasie
ARD: acid reflux digestion
DCS: double capillary systen
FAA: flaaeless atomic absorption
HGA-AA: heated .graphics analyzer of AA
ICE: inorganic complex extraction
OE: organic extraction
PAD: perchloric acid digestion
SA: standard addicio'ns
SIE: specific ion electrode
Acid lapinger:
1:1:1 H.MOj: HZ30-.
3asic Iz
20% KOH
H;02 Impinger:
52
Figure B-2. Analytical scheme of an impinger liquor sample,
B-3
-------
.SOLIDS
Dusc
Samole
Solid Diss
Bg
Se
AA: atomic absorption, flane
ARD: acid reflux digestion
DCS: double capillary system
HGA-AA: heated graphite analyzer of the atomic absorption spectrophotometer
HGA: heated graphite analyzer
ICE: inorganic cocplex extraction
OE: organic extraction
PAD: perchloric acid digestion
SA: standard additions
SIE: specific ion electrode
Figure 3-3. Dissolution and analytical scheme of solid samples.
B-4
-------
• Acid Reflux Digestion - Solid Isamples are dissolved by
refluxing with a mixture of nitric acid, sulfuric acid, and
perchloric acid. Silicates are not attached by this
procedure.
• Perchloric Acid Digestion - The first step is a treatment
using nitric and hydrofluoric acid. Perchloric acid is
added for final oxidation of the sample. A small amount
of hydrochloric is added to insure complete dissolution.
• Lithium Borate Fusion - A small amount of sample is fused
with lithium borate. The cooled melt is dissolved in
hydrochloric acid and hot water. Most elements present
in higher concentrations are analyzed from this digestion.
Elements present in trace concentrations are, in general,
determined from solutions derived from the acid reflux
or the perchloric acid digestion.
ANALYTICAL PROCEDURES
The analytical procedures used were originally developed for the determina-
tion of trace elements in coal, coal ashes, sludges and plant and animal
tissues. The drastic change in the matrix observed in samples collected at
a copper smelter necessitate screening of the procedures for accuracy and
reliability. This task is accomplished using the method of standard addi-
tion and interference studies.
Gas samples are drawn through a plug of gold wool. Deamalgamation is accom-
plished by heating the gold wool. The released mercury is purged through the
absorption cell of an atomic absorption spectrophotometer (AA).
Solid samples are analyzed for mercury by weighing a sample into a platinum
boat and heating the sample slowly in a chamber. The off-gases containing
elemental mercury are purged through a gold plug. Deamalgamation and deter-
mination by AA follow the same procedure as described above.
B-5
-------
Liquid samples are acidified and then the mercury is oxidized with potassium
permanganate. Hydroxylamine hydrochlorid and stannous chloride are used to
reduce the mercury to the metallic state. Air is bubbled through the solu-
tion. The mercury entrained in air is passed through the absorption cell
of an AA.
Lead and Cadmium - Lead and cadmium are extracted simultaneously with MIBK
from the WEP liquor, impinger solutions, and the perchloric acid digestion
of the hopper dust. The double complexing agent of ammonium pyrrolidine
dithiocarbamate and diethylammonium diethyldithiocarbamate chelates lead and
cadmium. The extracted sample is injected into the graphite furnace attach-
ment to the AA.
Antimony - Atimony is extracted as the iodide into a mixture of tributyl-
phosphate and MIBK. Extraction is performed on the WEP liquor, impinger
solutions are boiled to decompose the H202. The extracted solution is
injected into a graphite tube of the AA which has been coated with ammonium
molybdate.
Arsenic - The WEP liquor, impinger solutions and the perchloric acid diges-
tion of the solids are used for the arsenic determination. Arsenic is com-
plexed, in acidic medium, as the heteropoly acid of molybdenum. The aqueous
complex is injected into the heated graphite analyzer (HGA) attachment to
the atomic absorption spectrophotometer. A charring temperature of 1200°C
is used to remove any inteferences in the HGA.
Fluorine - Solid samples are fused with sodium carbonate and the melt
dissolved in deionized water. WEP and impinger solutions are run direct.
Final determination is done with a fluoride specific ion electrode utilizing
the method of known additions to remove the effects of any interfering ions.
Sulfur - Solid samples are dissolved in a solution of hydrochloric acid and
hydrogen perioxide. Solutions from the solid dissolution and the H202
B-6
-------
impingers are boiled to remove excess peroxide. The solutions are then
eluted through a cation exchange resin to convert the sulfate to sulfuric
acid. Volatile acids are removed by heating and the sulfuric acid is
titrated with a standard base.
B-7
------- |