EPA/560/5-82/014
I States
mmental Protection
Office of
Toxic Substances
Washington DC 20460
EPA-560/5-82-014
January, 1982
Substances
METHODS MANUAL
Sampling and
Analysis Procedures
for Assessing
Organic Emissions
from Stationery ^
Combustion Sources
in Exposure
Division Studies
Hf
-------
DISCLAIMER
This document has been reviewed and approved for publication by the
Office of Toxic Substances, Office of Pesticides and Toxic Substances,
U.S. Environmental Protection Agency. Approval does not signify that
the contents necessarily reflect the views and policies of the Environ-
mental Protection Agency, nor does the mention of trade names or com-
mercial products constitute endorsement or recommendation for use.
-------
SAMPLING AND ANALYSIS PROTOCOL FOR ASSESSING ORGANIC
EMISSIONS FROM STATIONARY COMBUSTION SOURCES IN
EXPOSURE EVALUATION DIVISION COMBUSTION STUDIES
by
John S. Stanley
Clarence L. Haile
Ann M. Small
Edward P. Olson
"X,
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
METHODS MANUAL
)
EPA Contract No. 68-01-5915
MRI Project No. 4901-A(36)
^
Prepared for:
U.S. Environmental Protection Agency
Office of Pesticides and Toxic Substances
Field Studies Branch
401 M Street, S.W.
Washington, D.C. 20460
Attn: Dr. Frederick Kutz, Project Officer
Mr. David Redford, Task Manager
-------
PREFACE
This sampling and analysis document was prepared for the Environmental
Protection Agency under EPA Contract No. 68-01-5915. The methods described
in this document were designed for use by Midwest Research Institute in
assessments of stationary combustion source emissions. They may also be
used as guidelines by other researchers who wish to conduct comparable stud-
ies. This document was prepared by Dr. John S. Stanley, Dr. Clarence L.
Haile (MRI Task Manager), Ms. Ann M. Small, and Mr. Edward P. Olson.
MIDWEST RESEARCH /INSTITUTE
John E. Going
Program Manager
Approved:
fl
James L. Spigarelli, Director
Analytical Chemistry Department
December 1, 1981
11
-------
CONTENTS
Preface ii
Figures iv
Tables iv
1. Introduction 1
2. Sampling 2
Media selection 2
Presampling site visit 3
Flue gas sampling equipment and materials 5
Solid and liquid sampling 11
Combustion air sampling 11
Continuous monitoring 12
Process data collection 12
Quality assurance procedures 12
Sample control and shipping 13
3. Sample Analysis 19
General analytical procedures 19
Extraction 21
Extract fractionation/cleanup 22
Extract analysis 22
Quality assurance (QA) procedures 24
4. Data Reporting 28
Sample tracking (analytical) 28
Data management and reporting 28
5. References 33
-------
FIGURES
Number Page
1 Modified Method 5 train for organics sampling 6
2 Condensed and sorbent trap modification of the Method 5 sampling
train for collection of organic vapors from flue gas 7
3 Eight-digit label code for stationary combustion source samples. 14
4 Sixteen character sample code for stationary combustion source
samples 15
5 Sample custody transfer sheet used by MRI 18
6 Analysis scheme for sample extracts 20
TABLES
Number Page
1 Sample Capture Requirements 4
2 GC/MS-SIM Ions 24
3 Sampling and Analysis Quality Assurance 26
4 Sample Tracking Sheet 29
5 Analytical Data Reporting Sheet 30
6 Total Inputs and Emissions 31
7 Daily Flue Gas Sampling Data 32
IV
-------
SECTION 1
INTRODUCTION
The sampling and analysis methods described in this report were specifi-
cally designed for use in an ongoing nationwide survey of emissions of organic
pollutants from stationary combustion sources. The primary focus of this sur-
vey is on polynuclear aromatic hydrocarbons (PAHs) and polychlorinated aromatic
hydrocarbons including polychlorinated biphenyls (PCBs), polychlorinated di-
benzo-£-dioxins (PCDDs), and polychlorinated dibenzofurans (PCDFs). To date,
these procedures have been used by Midwest Research Institute (MRI) to survey
emissions from coal-fired utility boilers, a co-fired (coal + refuse-derived
fuel) utility boiler, and a municipal refuse incinerator. This document was
prepared by MRI solely as a guideline for other laboratories who may partici-
pate in the same study, and for other researchers who wish to use these methods.
-------
SECTION 2
SAMPLING
The procedures described in this section have been used by MRI in ob-
taining representative samples of the inputs and the emissions from station-
ary conventional combustion sources. These procedures encompass the specific
requirements for site surveys, pretest preparations, and actual sampling pro-
cedures. A quality assurance program and sample control and custody docu-
mentation procedures are also presented.
The precise sampling procedures used for a specific plant may vary some-
what depending on the specific configuration and operation of the facility.
The samples that should be collected include gaseous, solid, and liquid ma-
terials. Gaseous emissions should be collected by the EPA Method 5 pro-
cedures1 modified for the capture of trace organic compounds as described in
this section. The solid and liquid samples should be collected according to
a sound, statistically designed 24-hr schedule.
These methods were designed to provide both qualitative and quantitative
information on polycyclic and chlorinated organic compounds. Therefore, it
is imperative that the sampling procedures should be followed as closely as
possible to prevent contamination or compromise the integrity of the samples.
MEDIA SELECTION
The sample collection program should be designed to allow accurate as-
sessments of the organic pollutants in both input to and emissions from the
combustion process. Although the focus of this project is on organic com-
pounds that likely undergo considerable chemical changes in the combustion
process, the media selection criteria are the same as would be used to deter-
mine a mass balance for a conservative pollutant, e.g., a nonvolatile metal.
It is of paramount importance that the collection procedures provide the most
representative specimens of the media selected.
The specific media and sampling points for each plant will depend some-
what on the specific design of the plant. However, the media can be described
in three categories: inputs, emissions, and miscellaneous media. The princi-
pal inputs to the combustion process are fuels and combustion air. All pri-
mary fuels should be sampled. Fuels used only for unit startup should be ex-
cluded from the sampling program. Combustion air can be sampled near the air
intake for the unit. However, care should be taken to avoid collection of
fugitive dust that may be suspended by the activities of the sampling equip-
ment or personnel.
-------
The principal emission media are the bottom ash, i.e., the residue from
the combustion process, and the flue gases with associated fly ash. Flue gas
emissions may have the most widespread impact on the quality of the surround-
ing environment. Hence, flue gas samples must be collected at a point down-
stream of the unit's particulate emissions control systems. However, the
materials collected by air pollution control devices must also be sampled to
allow accurate characterization of the total plant emissions. For example,
many coal-fired utility boilers employ electrostatic precipitators (ESPs) to
control particulate emissions. The fly ash collected in the ESP hoppers of
these plants must be included in the sampling program. In addition, ash that
is removed from flue gases by economizers in many plants must also be sampled
where practical.
Miscellaneous sample media include other materals that may have direct
contact with the combustion products. Examples are the input and overflow
waters from a bottom ash wet quenching system. These secondary emission media
may pose environmental hazards depending on the plant disposal practices.
A list of typical sample capture requirements for several materials is
given in Table 1. Included are sample size, storage container type and size,
sampling frequency, and total samples obtained each day. No compositing
should b£ done in the field. All samples should be placed in the appropriate
containers prepared as outlined in the pretest preparation and setup proce-
dures .
PRESAMPLING SITE VISIT
Representatives from EPA and the sampling crew chief must consult with
the plant supervisor to determine where and how each type of sample may be
captured. The crew chief should obtain data on key parameters related to
flue gas sampling. These parameters include the stack dimensions, flue gas
temperatures, moisture content, static pressure, and flue gas velocities.
Sampling points for grab samples should be located as close as possible
to the actual combustion process to avoid sampling combined streams (e.g.,
from multiple units) or combined waste media (e.g., ash-water mixtures) and
to prevent dilution of the desired sample. These precautions should allow
simpler data assessments. It is also advantageous to centralize the sampling
locations if possible such that the sampling schedules can be followed accu-
rately by the sampling crew. Where possible, special aids for obtaining the
samples in a safe and efficient manner should also be considered. Plant staff-
operated equipment, limited access areas, special tools, electrical outlets,
and periodic safety calls are some possible considerations.
Once all possible sampling points are determined, a statistically sound,
random sampling scheme should be provided for solid/liquid sampling that cor-
responds with the flue gas sampling activities. The sampling schedule should
be constructed to provide the sampling team with the specific time and loca-
tion a sample will be taken. It may be necessary to follow an assigned grid
pattern or port selection scheme in order to effectively subsample large sur-
face areas. In addition, sampling schedules may be subject to change each day.
-------
TABLE 1. SAMPLE CAPTURE REQUIREMENTS
Material
Solid
Coal
Refuse or RDF3
Bottom ash
Fly ash
Other solid waste
Gaseous
Dry particulate
Reeve Angel 934 AH
filter
Nozzel, probe, cyclone
and flask combined
rinses
Sorbent trap
First impinger with
rinses
h
Control 934 AH filter
Combustion air
Storage
size/type
1 qt amber glass
1 qt amber glass
1 pt amber glass
1 pt amber glass
1 pt amber glass
1 pt amber glass
150 mm X 15mm glass
petri dish
1 qt amber glass (may
require additional
250 ml of same)
Traps capped with plugs.
950 ml amber glass (may
require additional
250 ml of same)
150 mm X 15 mm glass
petri dish
Sorbent trap capped with
plugs
Sample
frequency
Twice per shift
Twice per shift
Twice per shift
Twice per shift
Twice per shift
c
One per train
One per train
One per train
One per train
One per train
One per day
One per day
Total
samples
(24 hr)
6
6
6
6
6
2
2
2
2
2
1
1
Liquid
Effluent water
Influent water
1 qt amber glass
1 qt amber glass
Twice per shift 12
(duplicates)
One per day 1
(duplicate)
a Refuse - derived fuel.
b Or equilvalent.
c Dry particulate will be collected only from trains using a cyclone trap.
This trap may not be necessary if particulate loading is light.
d Additional filters may be necessary if particulate loading is high.
-------
The presampling site visit should also allow the crew chief to determine
local sources for expendable sampling supplies. In addition, the most con-
venient accommodations for the sampling crew during the testing period should
be located.
FLUE GAS SAMPLING EQUIPMENT AND MATERIALS
The modified Method 5 train, shown schematically in Figure 1, is used to
collect samples for organic compounds from the stack. (See Note below.) Ad-
ditional empty impingers may be added just after the first impinger to retain
water from high moisture gases. This train should be operated according to
Method 5 specific procedures modified by the additional cleanup and recovery
procedures required for organic compounds.
The sampling probe liners must be glass or TFE, depending on the flue
gas temperature. A glass cyclone should be provided for high particulate
gases to avoid excessive filter loading. Vaporous organics are collected by
a sorbent trap (Figure 2). This trap is located in the sample line down-
stream of the heated oven and upstream of the first impinger. The trap is
packed with precleaned XAD-2. The module that houses the sorbent trap is
water-jacketed. Cold water from an ice bath surrounding the impingers is
pumped through the jacket to maintain an outlet temperature of ^ 16°C (60°F).
Because of the possible sensitivity of potential analytes to ultraviolet
light all sorbent traps should be kept wrapped in aluminum foil.
All solvents used for preparing the sampling train for testing and for
field laboratory cleaning of sample trains should be stored in glass or TFE
bottles. All solvents should be Burdick and Jackson Distilled-in-Glass or
equivalent grade. TFE or stainless steel forceps should be used for handling
filters. The train and train components that contact the sample should be
handled with clean, bare hands, i.e., without gloves.
Flue Gas Sampling Pretest Preparation
All train components that will contact the sample (probe, cyclone,
filter holder, resin cartridge, and connecting tubes) must be clean of all
potentially interfering materials. Component joints that have been previously
treated with a sealant, such as silicone grease, must be thoroughly cleaned
before use. The recommended procedure for removing Dow Corning High Vacuum
NOTE: The collection efficiencies for PCBs, PCDDs, PCDFs, and PAHs have not
been evaluated for this sampling train. However, the train design was based
on a validated particulate emissions collection system (EPA Method 5) with
the addition of an adsorbent cartridge (packed with XAD-2 resin) to collect
vaporous emissions of semivolatile organics. The collection efficiencies of
XAD-2 have been evaluated for a large number of compounds including PCBs and
PAHs, and XAD-2 was selected for use in the EPA source assessment sampling
system (SASS) train.2'3 XAD-2 was also evaluated for use in a train specially
designed for PCB sampling.4
-------
Cyclone
(optional)
Thermocouple
Reverse-Type
Pitot Tube
Condenser
& Resin
Cartridge
Console
Si Impingers 1,3 and 4 are of the Modified Greenburg-Smith Type
Impinger 2 is of the Greenburg-Smilh Design
Impinger 1 and 2 Contain 100 ml Water
Impinger 3 Empty
Impinger 4 Contains 200-300 Grams Silica Gel
Figure 1. Modified Method 5 train for organics sampling.
-------
a.
e
c
V
-e
o
o
8
9)
•o
C
O
u
60
C
a
ca
o u
•H -i-l
14-1 C
•i-l CO
•a M
O M
e o
CO O
M
4-> C
O
4-1 -i-l
C 4J
01 U
,£) 0)
!-i t-l
O i-l
CO O
O
T)
c X
03 «2
(U
4J C
CO -H
ca ca
C H
0) 4-1
ai
n
00
-------
Silicone Grease involves several steps. Removal of excess grease is accom-
plished by wiping clean with a rag. The joint is then dipped in warm concen-
trated KOH for 10 min, rinsed with distilled water, and wiped dry. The last
traces of sealant are removed by dipping in chromic acid, followed by rinsing
with distilled water and drying. Subsequently, the entire component should
be washed in soap (such as Alconox®) and water, followed by at least three
liberal rinsings of tap water and then distilled water. All apparatus are
then rinsed with acetone (reagent grade) until all traces of water are re-
moved. Finally, the glassware is rinsed sequentially with acetone and cyclo-
hexane (Burdick and Jackson Distilled-in-Glass or equivalent), taking care to
contact all sampling surfaces with the solvent. The components should be al-
lowed to dry in a dust-free and organic-free area to minimize contamination
of cleaned surfaces. The dried glassware should be protected by covering all
open joints and exposed sample-contacting surfaces with solvent-rinsed alumi-
num foil and by subsequent storage in a closed airtight shipping container
until use.
The filter housing gaskets should be cleaned using the same basic proce-
dure with the following modifications. Do not allow non-TFE gaskets to soak
in the 15% HN03 solution. Following the air drying, place the gaskets in a
225°F oven for 30 min to remove any moisture/solvents. Remove gaskets from
the oven and, store in a clean, covered container.
It is highly recommended that TFE filter housing gaskets be used when-
ever possible as contact with silicon and rubber gaskets can cause significant
sample contamination. If non-TFE gaskets must be used, care must be taken to
avoid contact with the organic solvents, during sample recovery.
Sample storage containers must also be cleaned prior to use. All sample
containers must be amber glass (or wrapped with aluminum foil) with TFE-lined
caps. All bottles and sample recovery apparatus must be cleaned with soap
and water, water rinsings, acetone rinsing and cyclohexane rinsing as outlined
above.
Sorbent resin used in the sampling trains should be precleaned and its
cleanliness verified prior to use. The recommended protocol for XAD-2 resins
is outlined in the EPA Level 1 Procedures Manual.3
All aspects of sampling train assembly should be conducted under the
cleanest laboratory conditions possible. To accomplish this, a limited-access
field laboratory should be maintained at the site to minimize the possibility
of airborne dust problems. Similarly, activities not directly related to train
preparation or sample recovery should be done elsewhere. Finally, smoking
should not be? permitted in the laboratory.
Prior to assembly, all sample-contacting train surfaces should be rinsed
with cyclohexane (Burdick and Jackson, Distilled-in-Glass or equivalent).
Care must be taken to contact all surfaces with solvent. During assembly it
is of vital importance that sealants, such as silicone grease, are not applied
to any connecting joints. All train parts must be closely examined for any
visual signs of contamination or defects that might induce sample error or
downtime problems; corrections will be made if necessary. Leak sealing should
-------
be accomplished using a material that has a high boiling point and high
thermal stability, such as the gas chromatography phase, Dow Corning DC 200.
Sorbent cartridges must be protected from exposure to light during sampling,
sample recovery, and shipping by wrapping each cartridge with aluminum foil.
Pretest Checkout of Sampling Apparatus
Briefly, the checkout involves assembling the entire sampling train as
shown in Figure 1 without the probe. The fitting at the inlet of the filter
box is sealed and the oven brought to operating temperature. The pump is
turned on, and the flow meter gauges are observed for the existence of any
appreciable flow. The train must pass the Method 5 standard leak test of
less than 0.02 cfm at 15 in. of mercury or 4% of the sampling rate, whichever
is less. If an unacceptable leak rate is observed, the operator should
(starting at the pump and moving in the direction of the probe) tighten each
fitting in order to assure that a loose fitting is not responsible for the
leak. If this action does not solve the leak, the system should be leak
checked on a modular basis until the problem is pinpointed. Under no condi-
tion should a sampling test be conducted with a leak rate in excess of 0.02
cfm at 15 in. Hg.
Flue Gas Sampling Procedures
Standard U.S. EPA methodology for particulate sampling, Method 5, as
specified in the Federal Register1 will be followed.
Two modified Method 5 sampling trains operating simultaneously should be
used to traverse points at the center of equal areas within the stack. The
number of traverse points and duration of sampling at each point should be
provided to the sampling crews. The sampling rates should be adjusted to ob-
tain samples at isokinetic conditions. The sum of flue gas collected each
day in the two trains should total 20 m3 ± 10%.
After the sampling trains are properly assembled and an acceptable pre-
test leak checkout has been made, preheat the probe and oven to 250°F. The
stack temperature, moisture content, and velocity profiles must be determined.
Compute the appropriate sampler flow rate and the proper nozzle size using
the procedures and calibration curves supplied by the equipment manufacturer.
During the course of the sampling run, scheduled parameter checks should
be made on flow rates, temperatures, and pressures. These data should be
logged in a sampling record book. Sufficient ice must be kept in the impinger
box to chill the condenser and resin trap to keep the impingers cool. At the
conclusion of the sampling run, a post leak rate check should be performed.
Sample Recovery
Proper cleanup procedure begins as soon as the probe is removed from the
stack at the end of the sampling period. During all rinsing, the approximate
volumes of glass-distilled water, acetone, and cyclohexane used should be re-
corded. This is necessary for the determination of background contributions
from the solvents. All organic solvents should be Burdick and Jackson
-------
Distilled-in-Glass or equivalent quality. The wash bottles used for all rins-
ings should be clean glass or TFE. Other plastic materials are unacceptable
due to their potential for sample contamination.
When the probe can be safely handled, wipe off all external particulate
matter near the tips of the probe nozzle. Remove the probe from the train
and cap off the mating joints of both the probe and the train with solvent-
rinsed aluminum foil. Also, cap the outlet of the train assembly after dis-
connection from the pump. Transfer the probe and train assemblies to the
field laboratory for cleanup. This area should be clean and protected to
minimize the chance of sample contamination or loss. Inspect the train prior
to and during disassembly and note any abnormal conditions. Remove the sor-
bent trap from the train and cap it off. The cartridge should be transferred
to the analytical laboratory intact for further sample recovery.
Rinse the probe with three portions each of water, acetone, and cyclo-
hexane. Brush the entire length of the probe with a natural bristle brush
during each rinse. The connecting tube between the sorbent module and the
filter housing should then be subjected to sequential rinsings using acetone
and cyclohexane, respectively. These rinses should be combined with the probe
and filter holder rinses.
The filter particulate is recovered by carefully removing the used filter
from the filter housing. Care must be taken to avoid tearing the filter or
losing particulate sample. The filter should be stored in a suitable sealed
glass container, such that the filter and its contents may be readily removed
for weighing in the lab. After removal of the filter, both halves of the fil-
ter housing should be subjected to sequential rinsing with acetone and cyclo-
hexane. These rinses should be combined with the preceding rinses. Non-TFE
filter housing gaskets should not be rinsed during sample recovery.
When cyclones are employed, the cyclone particulate catch should be re-
covered and stored in a separate sealed glass container. The cyclone should
be rinsed with water, acetone, and then cyclohexane. The rinses should be
combined with the other rinses. Similarly, all remaining interconnecting
tubing should be rinsed with acetone and cyclohexane. These rinses should
be combined with previous rinses.
The contents of the first impinger (aqueous condensate) should be poured
into a tared sample bottle. The bottle should be reweighed to ± 1 g and the
weight recorded in the sampling record book. The impinger jar should be rinsed
with acetone and cyclohexane and the rinses added to the sample bottle. Water
accumulated in the remaining impingers should also be determined gravimetrically
to ± 1 g.
Upon completion of the train recovery, at least four and possibly five
samples should be recovered: (a) the resin cartridge, (b) filter particulate,
(c) the first impinger contents, and (d) combined water, acetone, and cyclo-
hexane rinses of the entire train forward of the sorbent trap. A cyclone
catch will be the fifth sample if cyclones are employed.
10
-------
The importance of thorough rinsing cannot be overstressed. Care must be
taken to completely contact the interior surfaces of the train with each rinse
to quantitatively remove the analyte material. Approximate rinsing volumes
of each solvent used for each component should be recorded to allow the accu-
rate determination of solvent background contributions. After all rinsing is
completed, the sample bottle should be sealed and the volume of the contents
marked to identify possible losses during shipment. All samples should be
labeled and logged in the sampling record book as they are recovered. All
samples and rinses should be refrigerated at 4°C (or stored in an ice chest)
and exposure to light should be minimized during storage and shipment.
SOLID AND LIQUID SAMPLING
All sampling site locations should be clearly and appropriately labeled
for easy identification. Also posted at the sample site should be an expla-
nation of any subsample grid scheme to be followed. This serves as a reminder
of specific details in subsampling. The crew chief should tour the sampling
locations with the sampling personnel prior to the test to verify the collec-
tion procedure.
Crew chiefs should provide copies of all sampling schedules for the
plant supevisor to post with the plant operations staff. This should provide
for any necessary plant staff supervision or assistance in obtaining samples,
or in the event of an emergency.
The solid and liquid sampling schedule will start at 0000 hr on the
first day of flue gas sampling. Visits should to be made to sample sites as
scheduled, and samples taken and placed into prelabeled bottles. Sample and
container size required for typical media which should be sampled are given
in Table 1. Also included are the recommended number of samples to be col-
lected each day. All samples should be labeled and logged in the sampling
record book as they are collected.
Stainless steel trowels, cups, and tongs and glass bottles should be
used as necessary in sample capture. Long-handled extensions may be needed
to reach some specific areas. Sampling tools should be kept free from con-
taminants and cleaned with methodology described in this manual. Collection
of some samples may require special safety measures such as lab coat, work
suit, plant staff or assistance. Safety should be a primary consideration in
all sampling operations.
Solid and liquid sampling should continue through all three shifts each
day. Sampling staff will have to determine a suitable schedule so all shifts
are covered.
COMBUSTION AIR SAMPLING
Combustion air samples should be collected on 75 g of prepurified XAD-2
resin using vacuum pumps equipped with dry gas meters. A sampling rate of
0.75 cfm should be sustained until a total of 10-20 m3 has been sampled. The
resin should be placed in a cartridge similar to that in the sampling train,
but without the condenser. The resin cartridge should be wrapped in aluminum
11
-------
foil to prevent photodegradation of the adsorbed organic compounds. The sor-
bent cartridge should be capped immediately after sampling has been completed
to prevent contamination. This sample should be labeled and stored on ice as
soon as possible after collection.
CONTINUOUS MONITORING
Continuous monitoring of the flue gas should be conducted during the
period of flue gas sampling to aid in characterizing the efficiency of the
combustion process and to provide an indicator of dramatic changes in the
unit performance. The parameters monitored should include oxygen, carbon
monoxide, carbon dioxide, and total hydrocarbons. The continuous monitoring
probe should be inserted into the gas stream inlet to the air pollution con-
trol device to mitigate the influence of dilution by ambient air infiltration.
PROCESS DATA COLLECTION
In order to fully characterize the operation of the particular combus-
tion facility it is necessary to collect the engineering data during flue gas
sampling. A member of the sampling crew should be assigned to obtain perti-
nent information concerning the general description and design data and param-
eters for the power plant and air pollution control equipment that is not suf-
ficiently described from the presite visit. In addition, details and sched-
ules for soot blowing and ash removal during the actual testing period should
be recorded. A member of the sampling crew should document any plant break-
down, maintenance, or operating problem during each day's test period that
may have an impact on that day's test results. Process engineering data should
also be recorded for the megawatt output, steam flow rate, coal loading (rate
or total during each test period), and the operation of the electrostatic pre-
cipitators during each test period. The electrostatic precipitators should
be monitored for operating voltages and amperages, rapping frequency, spark
rate, and the number of inoperable units, if any. Information should also be
obtained concerning the electrostatic precipitator installed on the unit.
Specific parameters include the design volume, temperature, inlet concentra-
tion, number of precipitators, field array, gas passages per field, collecting
surfaces, collecting surface spacing, face area per precipitator (ft2), total
surface (ft2), gas velocity, and retention.
QUALITY ASSURANCE PROCEDURES
Calibrations
All sampling equipment should be calibrated prior to testing according
to the procedures outlined for Method 5 sampling trains.1 This should include
probe nozzle diameter measurements, pitot tube, and dry gas meter calibrations
as well as dial and liquid-filled thermometers and thermocouple-potentiometer
system calibrations.
12
-------
Field Blanks
The collection of field blanks is mandatory to indicate the cleanliness
of all sample handling phases. A complete flue gas sampling train should be
assembled in the field laboratory for use as a train blank at each plant.
The probe and filter oven should be maintained at operating temperatures for
a period equivalent to a typical sampling run. At the end of this period the
train should be disassembled and the sample components recovered in the same
manner as a flue gas sample.
Blanks of each rinsing solvent, as well as unused filters and resin car-
tridges, must be collected for blank determinations. The resin cartridges
and filter blanks should be exposed to the same laboratory environment for
the same time intervals as the sample filter and sorbent resin. Similarly,
individual samples of acetone and cyclohexane rinse solvents must be collected
from the wash bottles for each lot number solvent used. The volume of each
solvent blank collected should be approximately equal to the solvent volumes
used during the recovery of a sampling train. At least three unused sample
bottles of each type should be designated as bottle blanks. All blank materials
must be stored in clean sealed glass or TFE containers and treated as samples.
Liquid volumes should be marked on the containers to monitor possible shipping
losses.
SAMPLE CONTROL AND SHIPPING
An area designated for sample control and shipment preparation should be
close to the field laboratory. This area should also have limited personnel
traffic. Some stations may require lock and key access if outside of regular
plant surveillance. The following shipment preparations should be done by
staff on solid/liquid sampling duty.
Properly labeled bottles should be supplied to the sample train recovery
team. This team should then complete the label and return the full sample
bottle and any necessary sampling or recovery remarks to the sample control
and shipment operator.
Labels should be provided on computer printout paper and should be
grouped by date. Each date should be subgrouped into air sample labels and
solid and liquid sample labels. Extra labels should be provided at periodic
intervals of the label packet.
All labels should be provided in duplicate. One label should be placed
along the left column of a log book page each time a sample is labeled and
secured in an ice chest. This will provide chronological entry of sample
codes in the log book. Any sampling remarks should be recorded along side
the label. Also to be included in this right margin is the sample cooler
number in which the corresponding labeled sample has been placed.
MRI's labeling system provides an 8-digit number on each label to be
used for primary sample tracking. The label number is defined in Figure 3.
The label will also contain a 16-character sample code which provides for
easy, accurate identification. This sample code is defined in Figure 4.
13
-------
LABEL CODE
Plant Number
e.g., 01-
Saraple* number
within sample
type, e.g., 01-10
Train Component-
see key
Sample type-
see key
Sampling Day
e.g., 01-05
or 90 for blanks
SAMPLE TYPE
0 - Flue Gas Outlet
1 - Bottom Ash
2 - Control Device Ash (Fly Ash)
3 - Combustion Air
4 - Coal
5 - Refuse-Derived Fuel
Water sources will be numbered and defined as they
are taken.
TRAIN COMPONENT
0 - No Component
I - Probe Rinse
2 - Cyclone Catch
3 - Filter
4 - Resin
5 - Aqueous Condensate
Figure 3. Eight-digit label code for stationary combustion source samples.
14
-------
SAMPLE CODES
task
Plant Number -
e.g., 01-
Sample type - See 1
ey to tables
Date-MMDD
Sample Type
BA - Bottom Ash
FA - Fly Ash
CA - Combustion Air
FO - Flue Gas Outlet
CO - Coal
RF - Refuse-Derived Fuel
OW - Overflow Water
SW - Sluice water
RW - River water
Train Component
P - Probe Rinse
C - Cyclone Catch
F - Filter
X - Resin
W - Aqueous Condensate
Time
Train
Compo
ent:
Duct Code
(Optional)
Figure 4. Sixteen character sample code for stationary
combustion source samples.
15
-------
All labels should be prepared so that minimum amount of additional in-
formation must be entered in the field. The time of sample capture, any sub-
sampling location designations, and the sampler's name are typically the only
entries. All entries must be made on the duplicate label as well.
Labels should be self-adhesive. In addition, 1-1/2 to 2 in. transparent
tape should be overlaid on the completed label placed on sample containers as
a precautionary measure.
All liquid samples should have the volume marked on the side of the ship-
ping container with a permanent marker. This should allow losses of sample
from handling and shipping to be noted.
Samples other than those in amber glass bottles will require special
packaging. These are given in the following two paragraphs.
Modified Method 5 Particulate
The particulate filters should be returned to their original containers
(petri dishes) when sampling is completed. Each petri dish should be taped
shut using masking tape. The identification label should be placed on the
top center of each dish. The filter and dish number should be included on
both labels. The dish should be wrapped in aluminum foil and sample ID number
(from label) copied onto top side. The petri dish should be carefully sealed
in a ziplock bag, with a minimum air space. Care must be taken to ensure that
filters are returned to the original containers since it is necessary to know
the predetermined weight of each specific filter.
Resin Cartridge
The sample identification label should be affixed to the aluminum foil
covering. The entire cartridge should be wrapped in an adequate amount of
bubble pack (bubbles to inside) with cartridge ends capped securely with
glass balls. A lab marker should be used to copy the label ID number on to
the outside of the wrap.
Sample Custody Documentation
A chain-of-custody record should be prepared for every sample. The
custody sheet should be initiated in duplicate immediately after the sample
has been labeled. It should include the sample label number and the sampler's
signature. At the time of sample shipment, the record should be signed and
the time and date should be noted. The original copy of the chain-of-custody
record should be enclosed in the sample container. The yellow copy should be
retained by the crew chief until the samples are received and logged in at
the analytical laboratory.
When the container arrives at the laboratory, the person who will be
preparing the samples should receive (take custody of) it. That person should
then open the shipping container and check each sample for damage or tamper-
ing. This person should then sign all the enclosed chain-of-custody records
and note any damage or indication of tampering.
16
-------
Whenever custody of the sample is transferred from one person to another,
the person relinquishing custody of the sample should sign the chain-of-custody
form and note the time and date. The person receiving the sample should do
the same. The person having custody of the sample should have sole control
of access to the sample.
Figure 5 is an example of a chain-of-custody record used by MRI.
17
-------
4900-A36
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
CHAIN OF CUSTODY RECORD
Label No.:
Samplers:
(Signature)
Relinquished by:
(Signature)
Date/Time
Received by:
(Signature)
Date/Time
Comments (Changes, Volume
Removed, Dilutions, etc.)
Figure 5. Sample custody transfer sheet used by MRI.
18
-------
SECTION 3
SAMPLE ANALYSIS
The analytical procedures described in this section were developed
during a pilot study of stationary combustion source facilities. The primary
objective of this section is to ensure that the extraction and analysis of
samples from other such facilities is coordinated and consistent for all labo-
ratories involved in possible future Exposure Evaluation Division programs.
Each of the different types of samples should be combined into daily
composites, extracted, and analyzed by capillary gas chromatography using
flame ionization and Hall (halide mode) electrolytic conductivity detectors
(HRGC/Hall-FID) and by capillary gas chromatography/mass spectrometry
(HRGC/MS) using fused silica columns. The analytical scheme presented in
Figure 6 should be followed to determine the presence of various compounds in
sample extracts. Both qualitative and quantitative results are expected for
the range of polycyclic and chlorinated organic compounds determined by these
procedures. A rigorous quality assurance/quality control program has been
outlined and should be considered by other laboratories participating in sim-
ilar analytical efforts. In addition, a procedure to determine total organic
chlorine (TOC1) is described. This technique may be beneficial for pilot
studies to provide a more sensitive means of identifying the presence of
chlorinated polycyclic compounds.
GENERAL ANALYTICAL PROCEDURES
All solvents should be Burdick and Jackson, Distilled-in Glass, pesti-
cide grade quality, or equivalent. Glass wool, boiling chips and anhydrous
sodium sulfate should be pre-extracted with a hexane-acetone mixture or ben-
zene. The anhydrous sodium sulfate should be extracted with the hexane-
acetone azeotropic mixture, air dried, heated at 110°C for several hours, and
finally baked at 650°C for at least 2 hr. It is important to allow the ex-
tracted Na2S04 to dry thoroughly before baking at high temperature to avoid
explosions in the high temperature oven.
All glassware that will be used in handling the samples and extracts
should be cleaned first with soap and hot water, rinsed thoroughly with hot
water, followed by distilled water. Acetone (reagent grade) should be used
to rinse glassware for removal of all traces of water and final rinses with
acetone and cyclohexane should be required. If blanks are a problem, the
glassware should be baked at 400°C for at least 8 hr prior to use.
19
-------
ANALYSIS SCHEME
Sample Extract
Capillary GC/HALL & FID
Screen
Add Internal Standard
Anthracene - d]Q
Capillary GC/MS
(Scanning ) Surrogates +
Polycyclic Organic Compounds
Add Internal Standard
i 2,3,7,8 - Tetrachlorodibenzo-p-dioxin- C^or C6
Capillary GC/MS (SIM)
Chlorinated Polycyclic Organic
Compounds (Biphenyls, Dioxins, Furans)
Tentative N> '"" „ Hold
ID ?
Capillary GC/MS (SIM]
HIRES
Confirmation
Hold
Interlaboratory
Verification
Capillary GC/MS (SIM)
HIRES
Figure 6. Analysis scheme for sample extracts.
20
-------
EXTRACTION
The sample extraction methods described below have been developed to max-
imize recovery of a wide range of PAH compounds and polychlorinated aromatics.
Each sample should be spiked with labeled surrogate compounds prior to extrac-
tion for component recovery determinations.
Solid Samples
All solid samples should be Soxhlet extracted for 8 to 24 hr using ben-
zene (Burdick and Jackson, Distilled-in-Glass or equivalent) as the solvent.
Solid samples include XAD-2 resin, filters from the filter catch, and the
cyclone catch from the modified Method 5 train, control device ash (fly ash),
bottom ash, and fuel.
Samples from the Modified Method 5 sampling train should be extracted
simultaneously. For example, filters from the filter catch taken for a single
day's operation may be combined and extracted in the same Soxhlet apparatus.
However, sampling train materials should not be mixed (e.g., XAD-2 resin,
filters, cyclone material) since XAD-2 resins may be reclaimed. The filters
from the filter catch should be weighed prior to extraction. The weight of
the collected particulate matter should be calculated from measurements for
each filter made prior to sampling the field. No attempt should be made to
achieve constant weights for the filter samples. Also, the particulate ma-
terial obtained from the cyclone and probe rinses should be weighed prior to
extraction.
Control device ash and bottom ash should be individually composited for
each sampling day and 20 g each of these media Soxhlet extracted with benzene
for 8 to 24 hr. Prior to extraction, 10 ml of organic-free water should be
added to the control device ash. If the bottom ash is dry, 10 ml of water
should also be added to wet the material before beginning the extraction. An
inert material, such as Chromosorb W can be added to ash samples to promote
more efficient solvent flow through the sample.
Coal (10 g) should be Soxhlet extracted with benzene (8 to 16 hr). Large
mesh coal samples should first be ground to a powder using a ceramic mill with
stainless steel balls. Refuse-derived fuel will be evaluated when homogeneous
samples are available. This material (10 to 20 g) should also be milled and
Soxhlet extracted with benzene as the solvent for periods of 8 to 24 hr. This
extract should be washed three to four times with 100-ml aliquots of organic
free water.
All extracts should be dried by passage through short columns of an-
hydrous sodium sulfate. The dry extracts should be concentrated in Kuderna-
Danish appartus to approximately 5 ml. The extracts from the various com-
ponents of each day's flue gas sample should be combined and reduced to 5 ml
in a Kuderna-Danish evaporator. The extracts should be further concentrated
to 1 ml with a gentle stream of purified nitrogen. If solids precipitate from
extracts during concentration, slowly dilute and stir the extract to redissolve
the solids. Do not attempt to further concentrate the extract.
21
-------
Aqueous Samples
Aqueous samples may be obtained as plant influents and effluents, as well
as from the combined train rinses from flue gas sampling and the aqueous con-
densate from the first impinger. The combined train rinses should include
water, acetone, and cyclohexane. This sample should be shaken vigorously and
the organic layer removed using a separatory funnel. Two 60-ml aliquots of
cyclohexane should be used to extract the aqueous solution. The cyclohexane
extracts should be combined, dried by passing through a short column of an-
hydrous sodium sulfate, and finally combined and reduced with other extracts
from the modified Method 5 sampling train. If emulsions are formed during
the extraction, 2 to 3 g of sodium sulfate can be added to the mixture to pro-
mote adequate separation. The extracts should be dried and concentrated in
the same manner as extracts from solids.
EXTRACT FRACTIONATION/CLEANUP
Sample extracts, particularly the flue gas samples, may contain consider-
able interferences that may present problems in the effective analysis of such
compounds as polychlorinated dibenzo-p-dioxins and polychlorinated dibenzo-
furans or the surrogate compounds. These extracts should be cleaned by EPA
Method 613 to simplify the sample matrix prior to analysis for polychlorinated
dibenzo-£-dioxins and dibenzofurans. Since other polycyclic organic compounds
may be lost by this cleanup procedure, and it may be necessary to fractionate
a portion of sample by other techniques, such as silica or florisil adsorption
chromatography, prior to GC/MS analysis. Only half of a sample extract should
be submitted to any fractionation scheme. It is also important to measure
the exact volume of a sample extract subjected to cleanup to ensure valid
quantitation of specific compounds on the final aliquot.
EXTRACT ANALYSIS
HRGC/Hall-FID
Each extract should be screened first for the presence of halogenated
organic compounds by capillary gas chromatography separation with a Hall
electrolytic conductivity detector operated in the halogen specific mode.
Fused silica capillary columns, 30 m in length and coated with SE-54 [1% vinyl
in poly(methylphenylsiloxane) previously deactivated by silylation], 0.25 mm
ID, will be used for gas chromatography separations of all extracts.
On-column and direct injection techniques are preferred, but Grob-type
splitless injectors may be used, if necessary. The extracts should be chro-
matographed using the following temperature program: isothermal at 60°C for
2 min, increase temperature at 10°/min to 300°C and hold isothermally for 15
min. The qualitative results from these analysis should be useful in identi-
fying the presence of halogenated compounds by gas chromatography/mass spec-
trometry analyses. Large halogen responses at specific retention times should
identify the regions of the chromatograms where interpretation of the mass
spectra data may lead to the identification of halogenated compounds.
22
-------
HRGC/MS (Scanning)
The correlation of the mass spectral data with halogen responses from
the Hall electrolytic conductivity detector is necessary for positive identi-
fication of chlorinated organic compounds in the sample extract. Therefore,
it is necessary to duplicate gas chromatography conditions for the two methods.
Fused silica capillary columns coated with SE-54, 15 to 30 m in length, should
be used for all scanning HRGC/MS studies. On column, direct, or Grob-type
splitless injections should be used and the same temperature program used for
HRGC/Hall-FID should be followed.
Mass spectra should be acquired over the range of m/e 40 to 500 at a rate
of 1 to 1.2 sec/scan. The spectral data from these analyses should be used
for both qualitative and quantitative determinations. The compounds that are
positively identified and are of sufficient concentration should be quantitated
by peak area from the total ion chromatogram versus the peak area of the ap-
propriate internal standard.
HRGC/MS Selected Ion Monitoring (HRGC/MS-SIM)
The selected ion monitoring (SIM) technique should be used to determine
the presence of chlorinated dibenzodioxins or dibenzofurans. Preliminary in-
dications of the presence of these compounds may be evident from results of
the HRGC/Hall-FID and HRGC/MS (scanning) experiments. However, the HRGC/MS-
SIM technique has greater sensitivity for determination of these compounds in
the sample extracts. Selected ions characteristic of the mono- through octa-
chloro PCDDs, PCDFs, and PCBs should be monitored by this technique. The cri-
teria for the identification of these analytes in any extract will be dependent
on the coincidence of peaks in the extracted ion current plots of the char-
acteristic ions at the appropriate retention times and on the characteristic
relative intensity ratios of these selected ions. Table 2 lists the HRGC/MS-
SIM ions that should be used to identify the presence of the dibenzodioxins,
dibenzofurans, and biphenyls. Fused silica capillary columns coated with Car-
bowax 20M or other polar materials may be used for isomer specific SIM analyses
for polychlorinated dibenzodioxins and dibenzofurans.
HRGC/MS High Resolution Mass Spectrometry (HRGC/HRMS)
The tentative identification of dibenzodioxins and dibenzofurans by
HRGC/MS-SIM in any extract should be confirmed by HRGC/HRMS. This procedure
should be used to verify the presence of these compounds in any sample extract.
Positive identifications of chlorinated dibenzodioxins or chlorinated dibenzo-
furans by HRGC/HRMS should be supplemented by verification by other accredited
laboratories with HRGC/HRMS capabilities. All HRGC/HRMS studies should employ
fused silica capillary columns coated with either SE-54, Carbowax 20M, or
other materials capable of providing equivalent or better chromatographic
resolution.
23
-------
TABLE 2. GC/MS-SIM IONS
Number of chlorines
Biphenyls
Dibenzofurans
Dibenzo-jD-dioxins
1
2
3
4
5
6
7
8
9
10
188/190
222/224
256/258
290/292
324/326
360/362
394/396
428/430
462/464
498/500
202/204
242/244
270/272
304/306
338/340
374/376
408/410
442/444
-
-
218/220
252/254
286/288
320/322
354/356
390/392
424/426
458/460
-
-
Total Organic Chlorine (TOC1) Measurements
TOC1 measurements may be used in a primary sensitive screen of the chro-
matographable organically bound halide contents of extracts prepared for GC/MS
analysis in tiered analytical schemes. The TOC1 procedure is a simplified
gas chromatographic method using a Hall electrolytic conductivity detector in
the halide mode. A short packed GC column (typically 1-2 in. x 1/4 in. ID)
and a rapid temperature program are used to elute all chromatographable com-
pounds with volatiles equal to or greater than dichlorobenzenes as a single
peak. The area of this peak constitutes the TOC1 response which is quanti-
tated as chloride against a mixture of chlorinated compounds (typically a PCB
mixture such as Aroclor 1254). The typical method senstivity is 0.25-2 ng
chloride.
QUALITY ASSURANCE (QA) PROCEDURES
The positive identification and quantitation of specific compounds in
this assessment of stationary conventional combustion sources is highly de-
pendent on the integrity of the samples received and the precision and accu-
racy of all analytical procedures employed. The QA procedures described in
this section were designed to monitor the performance of the analytical meth-
ods and to provide information to take corrective actions if problems are
observed. These procedures are summarized in Table 3.
Field Blanks
The field blanks should be submitted as part of the samples collected at
each particular testing site. These blanks should consist of materials that
are used for sample collection and storage and are expected to be handled with
exactly the same procedure as each sample medium.
24
-------
Method Blanks
A method blank should be prepared for each set of analytical operations.
This will evaluate contaminations and artifacts that are derived from glass-
ware, reagents and sample handling in the laboratory. Method blanks should
be evaluated by each laboratory for solid and aqueous sample extractions.
Recovery Spikes
Surrogate compounds should be added to all samples prior to extraction
to provide an accurate record of analyte recovery. Specific analytes should
be used for method development procedures. In either case, duplicate samples
should be prepared. The surrogate compounds should include napthalene-d8 and
chrysene-d12- The other compounds that will be used as surrogates include
pentachlorophenol-13C6> 1,2,4,5-tetrachlorobenzene-13C6, and 3,4,3',4'-tetra-
chlorobiphenyl-d6.
Internal Standards
Each concentrated extract should be spiked with anthracene-d10 prior to
analysis by HRGC/MS in the scanning mode. This should allow for adequate quan-
titation of specific analytes in the extracts once proper response ratios have
been established using standard solutions. This internal standard should be
added to the extracts to yield a concentration in the range of the analytes
and the surrogate compounds. This internal standard can also be used to de-
termine relative retention times in any particular chromatograms, and this
provides another means of analyte identification. Stable isotope labeled
isomers of tetrachlorodibenzo-j>-dioxin and tetrachlorodibenzofuran should be
used for selected ion monitoring methods for these specific compounds. The
isotope label should provide sufficient distinction of the internal standard
and the actual isomers present in the sample extracts.
Reference Materials
A reference ash will be prepared by compositing ash from several facili-
ties by MRI. Portions of this ash may be sent to interested laboratories as
a means of evaluating interlaboratory performance. Two samples of spiked ref-
erence ash and two samples of unspiked reference ash will be submitted to all
laboratories. These samples will be extracted and analyzed with the same pro-
cedure used for all other samples.
Capillary Column Performance Tests
The optimum performance of the fused silica capillary columns coated with
SE-54 is an integral function of the separation and identification of specific
compounds in the sample extracts. Therefore, each laboratory should frequently
evaluate the performance of capillary columns used for extract analysis. Grob-
type test mixtures should be used to evaluate each column used for GC/Hall
and GC/MS studies. A test mixture prepared with halogenated compounds should
be used to test capillary columns with Hall electrolytic conductivity detectors.
25
-------
TABLE 3. SAMPLING AND ANALYSIS QUALITY ASSURANCE
Field blanks
Method blanks
Recovery spikes -
analytes
surrogates
dg-napthalene
13C6-1 ,2,4,5-tetrachlorobenzene
d6-3,4,3' ,4' -tetrachlorobiphenyl
13C6-pentachlorophenol
Internal standards - d10-anthracene
37C1-TCDD or 13C
37C1-TCDF
Reference materials
Capillary column performance checks
Interlaboratory verification
The parameters that should be monitored include separation number (Tz)
for a homologous series of compounds, the height equivalent for theoretical
plates (HETP), the number of theoretical plates (N), peak asymmetry, adsorp-
tion ratios, and pH of the column. Peak asymmetry is calculated for each peak
of the test mixture from the formula:
W
AS = -~ x 100
Where W and W are the back and front baseline widths of the peak measured
from a line bisecting the peak maximum. Adsorption ratios are determined by
comparison of the peak height for a compound susceptible to adsorption with
that of an inert compound. The pH of a column can be determined from the
ratio of the peak heights of equivalent quantities of an acid and a base in
the test mixtures.
The capillary columns should be evaluated immediately upon installation
and at least once per week. The capillary columns should be rejected for poor
performance as related to separation number, adsorptivity, and pH. The columns
should be tested more frequently if drastic deterioration of the column is
noted in a 1-week time span.
26
-------
Interlaboratory Verification
All extracts in which polychlorinated dibenzodioxins and dibenzofurans
are identified by HRGC/MS or HRGC/HRMS should be submitted to other labora-
tories for confirmation of these identifications by HRGC/HRMS.
27
-------
SECTION 4
DATA REPORTING
The following section provides examples of the pertinent data that should
be reported for characterization of polycyclic organic matter from stationary
combustion sources. Examples are given for methods of sample tracking through
the entire organic compound analyses, reporting of analytical data for surro-
gate compounds and the inputs and emissions of particular analytes, and the
reporting of the engineering process data necessary for describing flue gas
sampling methods and the actual combustion process. It is highly recommended
that all laboratories involved in combustion facility characterizations for
polycyclic organic materials adhere as closely to these reporting guidelines
to facilitate comparision of data from several sources.
SAMPLE TRACKING (ANALYTICAL)
Sample tracking sheets should be used by all analytical laboratories to
monitor the status of sample analyses. An example of the tracking sheet is
shown in Table 4. The sample numbers illustrated are truncated when samples
for 1 day's operation are composited. The pertinent information presented in
this sample number includes the task number (36), plant number, and the date
sampled.
The sample tracking sheets should be initiated upon receipt of the sam-
ples from the field stations. The information on the sheets should include
dates samples were received and dates the samples were composited and ex-
tracted. The other designations should indicate the extent of analysis for
each composited sample, i.e., screening sample by HRGC/Hall-FID, HRGC/MS, and
HRGC/MS-SIM. Other remarks can be added as required. For example, PAH, PCDD,
or PCDF might be added to indicate that polynuclear aromatic hydrocarbons,
chlorinated dibenzodioxins, or chlorinated dibenzofurans have been tentatively
identified in particular extracts. Likewise, other abbreviations might be
added for identification of chlorinated benzenes or phenols. Target may be
used to indicate that analysis for specific compounds has been completed.
Major will be used to specify that the major components of a sample have been
investigated.
DATA MANAGEMENT AND REPORTING
The data generated for the target compounds and the major components of
each sample extract should be presented to MRI in two forms. Assessment of
the QA program should be accomplished by reporting percent recovery of the
surrogate compounds in a specific extract together with the concentrations of
the target and major compounds in the extracts. Concentrations for the target
28
-------
H
|V1
s
CO
O
fe
w
Q_}
^J
P5
H
W
eu
co
w
CQ
^
H
t— i
§ CJ 2.
o §
H §
2
~-. t— 1
C_> CO
O I
06 CO
pr{ ^r]
^— \
CO 00
2 a
— . -H
cj a
o a
p-i oj
S o
CO
V. — ^
-^ 1
U r-t Q
O rH M
pi co PZ^
H w
T3
OJ
X u
W "3
4J
M 13
O OJ
p. 4.)
a -H
O W
U
V
>
OJ
u
D
•
O
a
(U
1
OH
a
03
CO
CN
1
*
01
00
J-(
BJ
H
r^ r-
CN CN
i i
j-i
0) S-l
00 O
i-( T-)
CO CO
H 2
CO CO
CN CN
1 1
^3" •*3'
O O
CN CN
1 1
O>
st 5- ••?
o o o
00 CT\ ON
rH rH i— *
%j" x^- ^"
000
LO LO LO
-------
compounds should be reported for all extracts. If a compound is present but
cannot be quantitated, it should be reported as less than the detection limit.
If a compound is not detected, it should be reported as not detected. The
data for QA will be reported as shown in Table 5. The percent recovery should
be reported for the surrogate compounds only.
_ TABLE 5. ANALYTICAL DATA REPORTING SHEET _
Day 1 _ Bottom ash - Composite ID No. _
Surrogate compounds Concentration % Recovery
d8-Naphthalene
13C6-1 ,2,4,5-Tetrachlorobenzene
13C6-3,4,3' ,4' -Tetrachlorobiphenyl
13C6-Pentachlorophenol
Analyte Compounds
1.
2.
3.
4.
The other method for reporting the data should follow the concentration
of a particular compound in all sample matrices. Sample concentrations should
be grouped according to inputs and outputs of the plant. This data report
should be presented as shown in Table 6, representative for particular com-
pounds over five composited sampling days.
The data supplied by the engineering report will be used to determine
the mass flow inputs and emissions for the various sample media. The summary
of the flue gas sampling and the continuous monitoring data should be tabu-
lated as shown in Table 7.
30
-------
en oo
S-
• 00
o —
C 00
o a
cj ^
3 A
: -- O -—
(0 rH 00
2
O l-l
v
CO 00
•rl S
S
• o
u w
c T:
o ~-
o oo
In) vl
33 -H
co a<
o a
O -i-l
U 00
c —
O 00
0 C
I •
< o.—
c: oo
•H E
I X a
•a
a c3
1 I— ( HH (— I >
i-i E
a
o
6-1
I
•H J3
r-(~M
«: S
*J
o
C
o
•H J3
O w
'3.2 I*
S -
-------
z
I—
a
3 J2
<
32
-------
SECTION 5
REFERENCES
1. Federal Register, 41(111), 23060-23090 (1976).
2. Adams, J., K. Menzies, and P. Levins, "Selection and Evaluation of Sorbent
Resins for the Collection of Organic Compounds," EPA 600/7-77-044
(April 1977).
3. IERL-RTP Procedures Manual: Level 1 Environmental Assessment (Second
Edition), EPA 600/7-78-201 (October 1978).
4. Haile, C. L., and E. Baladi, "Methods for Determining the Total Poly-
chlorinated Biphenyl Emissions from Incinerator and Capicator- and
Transformer-Filling Plants," NTIS No. PB-276 745/761 (1977).
33
-------
Sample C, Technical Report Data Sheet, EPA Form 2220- 1
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
56Q/5-82-014
2.
3. RECIPIENT'S ACCESSION NO.
A. TITLE AND SUBTITLE
Sampling and Analysis Procedures for Assessing Organic
Emissions from Stationary Combustion Sources for EED
Studies. Methods Manual
s. REPORT DATE
12/1/81 Preparation Date
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
J. Stanley, C. Haile, A. Small, and E. Olson
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Midwest Research Institute
425 Volker Boulevard
Kansas City, MO 64110
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-01-5915
12. SPONSORING AGENCY NAME AND ADDRESS
Exposure Evaluation Division (TS-798)
Office of Pesticides and Toxic Substances
401 M Street, SW
Washington. DC 2Q46Q
13. TVPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The sampling and analysis methods described in this report were specifically designed
for use in an ongoing nationwide survey of emissions of organic pollutants from
stationary combustion sources. The primary focus of this survey is on polynuclear
aromatic hydrocarbons (PAHs) and polychlorinated aromatic hydrocarbons including
polychlorinated biphenyls (PCBs), polychlorinated dibenzo-p-dioxins (PCDDs), and
polychlorinated dibenzofurans (PCDFs). To date, these procedures have been used by
Midwest Research Institute (MRI) to survey emissions from coal-fired utility boilers,
a co-fired (coal + refuse-derived fuel) utility boiler, and a municipal refuse in-
cinerator. This document was prepared by MRI as a guideline for laboratories who
may participate in this study, and for other researchers who wish to use these
methods.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Fieid/Grouo
Sampling and Analysis
Methodology
Combustion
Emissions
PAH
PCDD
PCDF
POM
18. DISTRIBUTION STATEMENT
'Release to Public
19. SECURITY CLASS / This Report)
Unclassified
21. NO. OF PAGES
34
20. SECURITY CLASS \Thu pa%e>
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |