EPA-450/4-84-014v
NATIONAL DiOXIN STUDY
TIER 4 — COMBUSTION SOURCES
Final Test Report — Site 13
Residential Wood Stove WS — A
By
Michael W. Hartman
Deborah J. Benson
Lawrence E. Keller
Radian Corporation
Research Triangle Park, North Carolina 27709
Contract Number: 68-03-3148
Donald Oberacker, Project Officer
Hazardous Waste Engineering Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
U.S. Environmental Protection Agency
Office Of Air And Radiation
Office Of Air Quality Planning And Standards
Research Triangle Park, North Carolina 27711
And
Office Of Research And Development
Washington DC 20460
April 1987
-------
This report has been reviewed by the Office Of Air Quality Planning And Standards, U.S.
Environmental Protection Agency, and approved for publication as received from the
contractor. Approval does not signify that the contents necessarily reflect the views and
policies of the Agency, neither does mention of trade names or commercial products
constitute endorsement or recommendation for use.
EPA-450/4-84-014V
-------
FOREWORD
This report is the result of a cooperative effort
between the Office of Research and Development's Hazardous
Waste Engineering Research Laboratory (HWERL) and the
Office of Air Quality Planning and Standard's Monitoring
and Data Analysis Division (MDAD). The overall management
of Tier 4 of the National Dioxin Study was the responsi-
bility of MDAD. In addition, MDAD provided technical
guidance for the source test covered by this report.
HWERL was directly responsible for the management and
technical direction of the source test.
m
-------
-------
TABLE OF CONTENTS
Section
Page
1.0 Introduction ..............
2.0 Summary and Conclusions ..........
2.1 Source Sampling and Analysis Overview . ? i
2.2 Summary of Results .......... !.'!."'*'' 2-4
3.0 Process Description ...... ..... -31
*****"****•** *J - 1
4.0 Test Description ...............
4.1 Field Sampling ....... .
4.2 Laboratory Analyses .....'.' ............ ?"i
4.2.1 Dioxin/Furan Analyses !.".'.' ......... 43
4.2.2 Dioxin/Furan Precursor Analyses' ! .' ." .' .' .." .' ' Jlj
5.0 Test Results
5.1 Process Data ........
\'l £°.Dt1nuous Missions Monitoring Data' ! ........ l~\
5.3 MM5 Dioxin/Furan Emissions Data . . ....... i,
5.4 Woodstove Ash and Flue Wipe . .......... To
Sample Dioxin/Furan Data ............ B"y
5.5 Wood Feed Precursor Data ..... K Q
••"......... a-y
6.0 Sampling Locations and Procedures fi ,
............. o-i
6.1 Gaseous Sampling .......
6.1.1 Gaseous Sampling Locations . ! ......... I ~\
6.1.2 Gaseous Sampling Procedures ....... fi f
6.1.2.1 Modified Method 5 (MM5J ;••••-••• o-i
6 1.2.2 Volumetric Gas Flow Rate Determination' '. 6-5
6.2 Wood Sampling ""* ^ M°1StUre Dete^"ation ...... 5.5
6.3 Ash Sampling .... ............. .... 6-6
6.. 4 Stack Wipe Sampling . '. \ \ '. \ \ '. '. \ \ \ \ \ ' ' ° ° \~_\
7.0 Analytical Procedures ..........
7.1 Dioxins/Furans ..... 7 .
7.2 Dioxin/Furan Precursors . .............. l~\
7.2.1 GC/MS Analyses ..'.'.'.'.'.'.'. ........ l~\
7.2.1.1 Sample Preparation . . . \ ....... 7"o
7.2.1.2 Analyses ..... ....... l'\
7.3 Total Chlorine Analysis ...... .'.*.'.".*.*."*'' .
-------
Section
8.0
TABLE OF CONTENTS (continued)
Page
Quality Assurance/Quality Control (QA/QC) 8-1
8.1 Manual Gas Sampling ............
8.2 Laboratory Analysis ,
8.2.1 Dioxin/Furan Analysis . „ ,
8.2.1.1 Recovery of Labelled Surrogate Compounds
8.2.1.2 Sample Blanks ,
8.2.2 Precursor Analyses -.-....
8-2
8-4
8-4
8-4
8-6
8-6
Appendix A Field Sampling Data
A.I Modified Method 5 and EPA Methods 1-4 Field Results
A.2 Continuous Emission Monitoring Results
vi
-------
LIST OF TABLES
Number
2-1
4-1
5-1
5-2
5-3
5-4
6-1
7-1
7-2
8-1
8-2
8-3
8-4
Source Sampling and Analysis Overview
Source Sampling and Analysis Matrix for Site WS-A
Summary of Woodstove Operating Parameters During the
ner 4 Test Runs , . ' ' '
Mean Values and Standard Deviations of Continuously
Monitored Combustion Gases at the Outlet Location ' ' '
Dioxin/Furan Content of Woodstove Ash and Flue
Wipe Samples
Summary of Dioxin/Furan Precursor Data for Site WS-A
Feed Samples ' " '
Summary of Gas Sampling Methods for Site WS-A
Instrument Conditions for GC/MS Precursor Analyses
Components of the Calibration Solution
Glassware Precleaning Procedure
nnneS- Surr°9?t(LRec°veries for Site WS-A Ash and ...
Flue Wipe Dioxin/Furan Analyses
Analysis Results for Quality Control Samples
Percent Surrogate Recoveries for Site WS-A Feed Samples .
Page
2-3
4-2
5-2
5-4
5-10
5-11
6-2
7-7
7-8
8-3
8-5
8-7
8-8
VI1
-------
I
LIST OF FIGURES
Number
2-1
5-1
5-2
5-3
5-4
6-1
7-1
Page
Simplified Diagram of Woodstove WS-A . . . . ........ 2-2
Oxygen Concentration Data .......... * ...... 5-5
Carbon Monoxide Concentration Data
(corrected to 3% 02)
Carbon Dioxide Concentration Data
(corrected to 3% O)
5-6
5-7
Total Hydrocarbon Concentration Data . ........... 5-8
(corrected to 3% 02)
Modified Method 5 Train with Back-up Sorbent Module .... 6-4
Sample Preparation Flow Diagram for Site WS-A Precursor
Analyses
7-4
vm
-------
1.0 INTRODUCTION
This report summarizes the results of a dioxin/furan3 emissions test of a
residential woodstove conducted by Radian Corporation. The stove is a
freestanding noncatalytic model manufactured by Atlanta Stove"Works" and offered
for sale in the Sears Catalog (#42G84156N). During testing oak and pine were
burned at low burn rates, which is representative of normal residential use.
The test was the thirteenth in a series of dioxin/furan emissions tests
conducted under Tier 4 of the National Dioxin Study. The primary objective of
Tier 4 is to determine if various combustion sources are sources of dioxin
and/or furan emissions. If any of the combustion sources are found to emit
dioxin or furan, the secondary objective of Tier 4 is to quantify these
emissions.
Residential woodstoves are among 8 combustion source categories that have
been tested in the Tier 4 program. The tested woodstove, hereafter referred to
as Woodstove WS-A, is a test unit located at an EPA contractor facility. This
stove was selected for inclusion in the Tier 4 program due to its location in
the Research Triangle Park area and because simultaneous testing of the stove
was already being conducted for another EPA program (Integrated Air Cancer
Project). The woodstove tested is considered representative of woodstoves
built in the last 5 to 10 years.
This test report is organized as follows. A summary of test results and
conclusions is provided in Section 2.0, followed by a process description in
Section 3.0. The source sampling and analysis plan is outlined in Section 4.0,
and the dioxin test data are presented in Section 5.0. Sections 6.0 through
8.0 present various testing details. These include descriptions of the
refer tn tho nnl^hi •*"? ih*acronyms PCD° and PCDF as used in this report
JouTor°mofe SlSllritJS? dlbenz°-P-d1ox1n and dibenzofuran isomers with
1-1
-------
sampling locations and procedures (Section 6.0),, descriptions of the analytical
procedures (Section 7.0), and a summary of the quality assurance/quality
control results (Section 8.0). The appendices contain data generated during
the field sampling and analytical activities.
1-2
-------
2.0 SUMMARY AND CONCLUSIONS
2.1 SOURCE SAMPLING AND ANALYSIS OVERVIEW
A simplified diagram of Hoodstove WS-A is shown in Figure 2-1. The stove
is considered to be a typical residential wood stove. The fuel was oak and
Pine aged approximately 1 year. The stove was operated at low burn rates and
low operating temperatures for maximum wood use efficiency, which is typical
of residential operation.
Sampling for dioxin/furan emissions was performed by Radian at the outlet
exhaust stack in each of a series of three test runs conducted on May 10, 17,
and 24, 1985. The dioxin/furan sampling was based on the October 1984 draft
of the Modified Method 5 (MM5) procedure developed by the American Society of
Mechanical Engineers (ASME) for measuring emissions of chlorinated organic
compounds. Modifications to the draft ASME protocol used at this test site
are discussed in Section 6.1.2. MM5 train components and train rinses were
analyzed for dioxins and furans by ECL-Bay St. Louis and EMSL-RTP, two of
three EPA laboratories collectively known as Troika. The dioxin/furan
analysis attempted to quantify the 2378 TCDD/TCDF isomers and the tetra-
through octa- dioxin/furan homologues present in the samples.
Dioxin/furan precursor analyses were performed by Radian on samples of
the wood fed to the stove. The specific dioxin precursors analyzed for were
chlorophenols, chlorobenzenes, polychlorinated biphenyls, and total chlorides
Woodstove ash and flue wipe samples were also taken and analyzed by Troika for
dioxin/furan content.
In addition to the above sampling and analysis efforts conducted
specifically for the Tier 4 program, simultaneous testing of the woodstove was
2-1
-------
C.-C Bag -*3*> n «==*" To continuous gas analysers
Sample Line
- Probe —
re"
20"
FID
Thermocouple
wires T deg C
_ Asbestos
' Aluminum
— Scale
30"
Front
6
Figure 2-1. Simplified Diagram of Woodstove WS-A
2-2
-------
TABLE 2-1. SOURCE SAMPLING AND ANALYSIS OVERVIEW
Item
Item Description
1. Number of test runs
2. Gaseous Sampling
Solids Sampling
Three test runs
(Runs.01, 02, 03)
MM5 sampling at woodstove outlet
(Runs 01, 02, 03).
Dioxin/furan analysis.
EPA Reference Methods 2 and 4 at
woodstove outlet exhaust stack
(Runs 01, 02^ 03). Gas velocity
and moisture.
Continuous monitoring of CO, C0?, 09,
and total hydrocarbons at woodstove
outlet (Runs 01, 02, 03).
Oak feed sampling (Runs 01, 02).
Dioxin/furan precursor analysis.
Pine feed sampling (Run 03).
Dioxin/furan precursor analysis.
Bottom ash sampling (Runs 01, 02,
03) Dioxin/furan analysis.
Stack wipe sampling at outlet exhaust
stack (Runs 02, 03)
2-3
-------
performed by Research Triangle Institute (RTI) as part of the Integrated Air
Cancer Project. Continuous emissions monitoring (CEM) was performed by RTI
personnel at the stove exhaust location for CO, C02, THC, and 02- RTI also
conducted Modified Method 5 sampling tests for polycyclic aromatic
hydrocarbons (PAH)(GC-FID with GC-MS confirmation), gravimetric and total
chromatographable hydrocarbons sampling, spot dilution tests for PAH
concentration, and retene analysis. Manual recordings included wood weight,
burn time, stack flow, room temperature and humidity, inlet flows through the
dampers and general operating conditions and occurrences. Continuous
measurements of stack temperature and stove temperature were also recorded.
The data from these tests are reported in reference 1.
2.2 SUMMARY OF RESULTS
No valid flue gas dioxin/furan emissions data were obtained for Woodstove
WS-A. Labelled internal standards spiked onto the MM5 sample train components
were not recoverable due to the large amounts of hydrocarbons present.
Analyses of woodstove ash and flue wipe samples from this test site
showed minimal dioxin/furan content. Octa-CDD was the only dioxin/furan
homologue detected in the three ash samples analyzed, and the values reported
were near the analytical detection limit. The maximum octa-CDD content of the
ash samples was 0.09 parts-per-billion (ppb). Small quantities of octa-CDD
were found in each of the two flue wipe samples analyzed, with hepta-CDD also
being detected in one of the two samples. The maximum octa-CDD content of the
flue wipe samples was 0.6 ppb, and the measured hepta-CDD content was 0.04
ppb.
The woodstove was operated at low burn rates for all test runs, which is
representative of normal consumer operation. Burn rates for individual test
runs ranged from 1.3 kg/hr (Run 01) to 3.5 kg/hr (Run 02). Average
1. Leese, K. E., and S. M. Harkins, RTI. Integrated Air Cancer Project-
Source Measurement. Draft Final Report. RTI/3065-07. March 1986.
2-4
-------
as-measured emission concentrations of CO and THC during the test runs were
8,000 ppmv and 10,800 ppmv, respectively. The average oxygen content of the
flue gas was 17.0 vol %.
Chlorobenzenes, chlorophenols, and polychlorinated biphenyls were not
detected in the oak and pine samples analyzed. The total chloride contents of
the oak and pine samples were 125 ppm and 49 ppm, respectively.
2-5
-------
-------
1
3.0 PROCESS DESCRIPTION
Woodstove WS-A is a Sears Catalog No. 42G84156N free-standing radiant
woodstove. The stove is rectangular with a set of hinged interlocking doors
on the front and one hinged door on the right side which houses three
screw-down air dampers. The air dampers are each three inches in diameter and
form a triangle on the door. These dampers are used to control the amount of
air entering the stove, and thus, the burn rate. The upper damper was sealed
shut with silicone high temperature sealant to allow better control of the
burn rate. A steel grate is normally located on the front of the stove just
inside the two interlocking front doors, but it was removed to facilitate
loading of kindling before each run and removal of ash after each run. The
interior of the stove is lined with firebrick. The stove is baffled and the
flue exit is in the rear, opposite the hinged interlocking front doors. The
legs of the stove were removed, and the stove and flue sections were mounted
on a Detecto 5850 scale which was calibrated to 1000 pounds just prior to this
study. Before the stove and flue were mounted, the scale was leveled and a
3' x 3' x 1/2" sheet of aluminum was placed on top of the platform along with
an asbestos board of the same approximate dimensions. The flue exit from the
rear of the stove is 6 inches in diameter and the inside diameter of the flue
is 8 inches. A single-walled 6-inch/8-inch adaptor was used to connect the
stove exit to an 8-inch inside diameter double-walled Metalbestos R insulated
tee. All sections of the flue from the tee upward consisted of the same type
of double-walled insulated flue. Two 30-inch and two 9-inch vertical flue
sections were mounted above the tee.
1. Leese, K.E., and S. M. Harkins, RTI. Integrated Air Cancer Project-
Source Measurement. Draft Final Report. RTI/3065-07. March 1986.
3-1
-------
-------
4:0 TEST DESCRIPTION
This section describes the Tier 4 field sampling, process monitoring, and
analytical activities that were performed for test Site WS-A. The purpose of
the section is to provide sufficient descriptive information about the test so
that the data presented in Section 5.0 can be easily understood. Specific
testing details (sampling locations and procedures) will be presented later,
in Section 6.0.
This section is divided into three parts. Section 4.1 summarizes field
sampling activities, Section 4.2 summarizes process monitoring activities, and
Section 4.3 summarizes analytical activities performed during the test
program.
4.1 FIELD SAMPLING
Table 4-1 shows the source sampling and analysis matrix for test Site
WS-A. Three dioxin/furan emissions tests (Runs 01, 02, 03) were performed at
the woodstove outlet exhaust stack. Dioxin/furan sampling was based on the
MM5 sampling protocol developed by ASME for measuring emissions of chlorinated
organic compounds. Testing was performed at the woodstove exhaust stack for a
period corresponding to 240 minutes of on-line sampling. The ASME protocol
was modified for woodstove use by the addition of a second XAD-2 R resin trap
due to the high total hydrocarbon concentration in the exhaust gas. The
protocol was also modified to allow for velocity readings taken by a vane
anemometer every 15 minutes rather than with a pitot tube. The extremely low
velocity of the woodstove exhaust gas precluded normal flow measurement and
isokinetic sampling. Additional details on sampling procedures and deviations
from the ASME protocol are contained in Section 6.2.1.
4-1
-------
I
Ul
V)
1
X
i
t/t
in
|
12
OT
Ul
i— t
1
Ul
Ul
Ii
58-
CD
•*- U
O U.
Z CO
•§
i
Analytical
o
i
i
!
CO
Sample Type or
Parameter
e
o
8
0
or
O ^3
c *—
.«•*-
U CN •* *3
01 - «
£3S£
§«**->
coo
is
2 s
si
SI
U CL
£ CO
Ul Ul
T3
SS
Is X
Ul U
*J
«— £
Ul
s
Dloxln/furan enlssl
I
•M
o tn
>
O •!•*
•M Ul
433
II
^4
3
(A
i
L.
0)
CL
0
O
1
u
CL
O.
fO
o
z
1
e
e
5
19
U
U)
"u
i
L 2 en
S3 4J 5
O U ^
3 n m _J =
*" u S "3 t.
o* e t- o 4J
|^2 S. i«
U (D •*•» O ** '
O C Ul U C i«
Ul O
L. O
O C «-
s ^ s
"c « o> >»
S m 2*«
u Tn c
ui •*- t- o
a &• O
o *> a.-o
I I 11
•s 5=t
0 U 0 _C
tn ** ^ ®
Ul A
< < 1 O
uj LU i-t a.-
c
o
Molecular Height
Moisture
CO, C02 concentrati
0, 01
i 1
Ul U>
>» c >» c
*ui E In L,
O +J O 4-»
a ui 3 ui
£S S-
11 11
,
e
O
j 1
1 lu
S1 -5
i IS
t. 10 *»
a •— »
a. u. "a
N. <^
! 8 8
:8 5 o
) — 0 O
~ *" Ul ul
3 9 19
fi
i
O +*
**s
>*
CL O
2S
-I
§8
t- CL
,C CO
° Ul
Ul Ul
in «j
a a:
|
Ul
0 U)
CL i9
3 -U
U. »
4J Ul
C O
Dloxln/furan conte
of flue wipe sampl
o
CL
•^
O
u.
•
m
-------
/
Continuous emissions monitoring of 02, CO, C02, and THC was performed by
RTI during the three MM5 test runs. These data were obtained to assess
variations in combustion conditions during the sampling periods.
Instantaneous concentration values for each species monitored were determined
every five minutes by the CEM system.
Three types of process samples were taken at Site WS-A: the wood feed,
the bottom ash after the burn and the stack flue creosote deposits. The wood
samples were taken from a representative log of each type of wood selected at
random after the test period. Three identical portions of each wood sample
were prepared: one for potential dioxin/furan analysis by Troika, one for
dioxin/furan precursor analysis by Radian/RTP, and one for total chlorides
analysis by RTI. The ash samples were taken from the woodstove after each
test burn and were sent to Troika for dioxin/furan analysis. The flue
deposits were taken after the oak and pine tests by wiping the inside of the
flue with precleaned glass wool. These samples were sent to Troika for
dioxin/furan analysis.
4.2 LABORATORY ANALYSES
Laboratory analyses performed on samples from test Site WS-A included
dioxin/furan analyses, dioxin/furan precursor analyses and chloride analyses.
Samples analyzed for dioxin/furan are discussed in Section 4.3.1 and samples
analyzed for dioxin precursors are discussed in Section 4.3.2. Samples
analyzed for chloride are discussed in Section 4.3.3.
4.2.1 Dioxin/Furan Analyses
All dioxin/furan analyses for Site WS-A samples were performed by Troika.
Field samples requiring dioxin/furan analysis were prioritized by Tier 4 based
on their relative importance to the program objectives. The priority levels,
in order of decreasing importance, were designated Priority 1 through
Priority 3; however, during this test all samples were designated Priority 1.
4-3
-------
Priority 1 samples were sent to Troika with instructions to perform
immediate extraction and analysis. These included the MM5 train components
for the outlet sampling locations, the ash and flue wipe samples, and an MM5
train field blank.
4.2.2 Dioxin/Furan Precursor Analysis _
Dioxin/furan precursor analyses of wood feed samples were performed by
Radian/RTP. The specific dioxin/furan precursors analyzed for included
chlorophenols, chlorobenzenes, and PCB's. Total chlorine analyses of the wood
feed samples were performed by RTI. ;
4-4
-------
5.0 TEST RESULTS .,-._,.,.
The results of the Tier 4 dioxin/furan emissions tests of Woodstove WS-A
are presented in this section. The individual test runs are designated as
Runs 01-03 in this report and as Runs 2, 3, and 5 in the RTI report.
Process data obtained during the test runs are presented in Section 5.1,
and continuous monitoring results for 02, CO, C02, and THC are presented in
Section 5.2. Dioxin/furan emissions data are contained in Section 5.3. Ash
and flue wipe sample analyses are presented in Section 5.4, and precursor
analyses of the wood feed samples are presented in Section 5.5.
5.1 PROCESS DATA
An overview of the woodstove operating data obtained during the Tier 4
test runs is presented in Table 5-1. Additional operating data (e.g., stove
temperatures, detailed wood analyses, inlet air flow rates, etc.) are
contained in the RTI report.
The feed during all test runs was split cord wood. Oak was burned during
Runs 01 and 02, and pine was burned during Run 03. Cured wood was burned
during Runs 01 and 03, and uncured wood was burned during Run 02. Burn rates
were low for all test runs, ranging from 1.25'kg/hr to 3.5 kg/hr. Low burn
rate test runs were purposely selected for the Tier 4 program to maximize the
potential for dioxin/furan formation. The wood load ranged from 8.8 kg of
initial charge to 20.3 kg of initial charge, which is close to the capacity of
the stove. Flue gas flow rates were consistent between test runs, ranging
from 19.6 to 27.5 dscfm. These flow rates were typical for this stove at low
burn rates.
5-1
-------
TABLE 5.1 SUMMARY OF WOODSTOVE OPERATING PARAMETERS DURING
THE TIER 4 TEST RUNS
Run
Number
Fuel
Type
Wood
Moisture
(wt %)
Wood Load
(kg)
Burn Rate
(kg/hr)
Flue Gas
'. FT ow
i(dscfm)
1 Oak 18.7 (cured) 8.8 1.25 19.6
2 Oak 34.9 (uncured) 20.3 3.53 25.1
3 Pine 15.1 (cured) 12.0 1.87 27.5
5-2
-------
5.2 CONTINUOUS EMISSIONS MONITORING DATA
Mean values and standard deviations of the continuously monitored
combustion gases at the stack location (02, CO, C02, and THC) are shown for
each MM5 test run in Table 5-2. The data show that most-of-the runs, have ^
similar mean concentration values for the individual gases. The overall mean
values for the three test runs are as follows: oxygen, 17.0 percent by volume
(dry); carbon monoxide, 3.7 percent by volume (dry @ 3% 02); carbon dioxide,
15.8 percent by volume (dry @ 3% 02); and total hydrocarbons, 4.9 percent by
volume (wet @ 3% 02, as propane).
Instantaneous concentration values obtained at 5-minute intervals for
each of the continuously monitored combustion gases are tabulated in Appendix
A-2 and are shown graphically as functions of time in Figures 5-1 through
Figures 5-4. These graphs show that in general the measured 02 values were
fairly constant within runs and between runs. During all three runs conducted
on the woodstove, the hydrocarbon analyzer was in the maximum reading
position. The THC values are to be considered lower bound values, since the
instrument's upper limit on the highest range is 10,000 ppmv as methane.
5.3 MM5 DIOXIN/FURAN EMISSIONS DATA
No valid flue gas dioxin/furan emissions data were obtained for the
woodstove. The four labelled internal standards spiked in the MM5 train
samples could not be recovered. This indicates that both the aqueous and
XAD-2 portions of the samples caused serious sample preparation problems. The
sample extracts were reported to be yellow in color, and exhibited evidence of
significant hydrocarbon contamination. This resulted in peak broadening and
overloading of the alumina and carbon GC/MS columns. The Troika laboratory
report concluded that the analytical results did not yield any valid
indication of whether dioxins/furans were present in the MM5 samples.
5-3
-------
TABLE 5-2. MEAN VALUES AND STANDARD DEVIATIONS OF CONTINUOUSLY
MONITORED COMBUSTION GASES AT THE OUTLET LOCATION
Parameter (a,b,c)
02 (% vol)
Standard deviation
CO (ppmv @ 3% 02)
Standard deviation
C02 (% vol @ 3% 02)
Standard deviation
THC (ppmv @ 3% 02)
Standard deviation
Run 01
16.8
(0.4)
32464.6
(5366.3)
14.3
(2.0)
40965.2
(4057.0)
Run 02
17.5
(0.3)
41730.2
(5592.3)
16.9
(1.2)
46577.9
(10620.4)
Run 03
16.6
(0.5)
36419.9
(7050.0)
16.3
(1.8)
60879.2
(8385.1)
Average
17.0
37000
15.8
49000
Mean values shown on top, with standard deviation below in parenthesis.
a6as sampling for the continuous monitors was performed at the outlet
location.
concentrations expressed on a dry volume basis except for total
hydrocarbon concentrations, which are expressed on a wet volume basis.
°Total hydrocarbon data are expressed in units of ppmv (wet) as methane.
5-4
-------
SITE WOOD STOVE - TEST 1
TEST TIMC (HOURS)
SITE WOOD STOVE - TEST 2
TST TIUC (H00«*>
SITE WOOD STOVE - TEST 3
ao
!
1*
13
14
13
ia
11
to
TOT TIMC (HOUMS)
Figure 5-1. Oxygen Concentration Data
5-5
-------
SITE WOOD STOVE - TEST 1
CARBON MONOXIDE PROFILE
~ so
8
X
n
9
20
TEST TIMC (HOURS)
SITE WOOD STOVE - TEST 2
CARSON
8 *
7O
a 3
TEST TIMC (HOOKS)
SITE WOOD STOVE - TEST 3
CAMMM MOMOXIOC
^ 30'
8
.
a 3
TOT TIMC (HOURS)
Figure 5-2.
Carbon Monoxide Concentration Data
(corrected to 3% 02)
5-6
-------
SITE WOOD STOVE - TEST
TBST
(HOURS)
i
i
17
1*
19
14.
13
13
<1
1O
*
SITE WOOD STOVE - TEST 2
C»*»ON OIOMOC
TOT Txe (HOURS)
SITE WOOD STOVE - TEST 3
GOMCENTMMION (*/ • 3X 03)
• • in — ., -^.^^
ZT^^ssass^^.:
TOT -HMC (HOURS)
Figure 5-3.
Carbon Dioxide Concentration Data
(corrected to 3% 02)
5-7
-------
SITE WOOD STOVE - TEST 1
100
TOTAL
TCTT TIMC (MOUII3)
SITE WOOD STOVE - TEST 2
rerr TIMC (MOU*S>
SITE WOOD STOVE - TEST 3
8
S
•
*l
38
r
•0-
f
\
j* ^VA^_*\/\. f^fn
"**\ JL -MM.
.... ,.., , , _
TEXT TIMC (HOUM)
Figure 5-4.
Total Hydrocarbon Concentration Data
(corrected to 3% 02)
5-8
-------
5.4 WOODSTOVE ASH AND FLUE WIPE SAMPLE DIOXIN/FURAN DATA
Table 5-3 summarizes the results of dioxin/furan analyses of woodstove
ash samples from Runs 01-03 and flue wipe samples from Runs 02 and 03.
Octa-CDD was the only dioxin/furan homologue detected in the woodstove ash
samples, with the maximum value for any test run being 0.09 ppb octa-CDD.
Both hepta-CDD and octa-CDD were detected in the flue wipe samples, with the
maximum values for any test run being 0.04 ppb hepta-CDD and 0.3 ppb octa-CDD.
5.5 WOOD FEED PRECURSOR DATA
Table 5-4 summarizes the dioxin/furan precursor data for Site WS-A feed
samples. Chlorobenzenes, chlorophenols, and polychlorinated biphenyls were
not detected in the oak and pine samples analyzed. The total chloride
contents of the oak and pine samples were 125 ppm and 49 ppm, respectively.
5-9
-------
TABLE 5-3. DIOXIN/FURAN CONTENT OF WOODSTOVE ASH
AND FLUE WIPE SAMPLES
Dioxin/Furan Content, ppb
Dioxin/Furan
Homologue
Dloxins
2378 TCDD
all tetra CDD
penta CDD
hexa CDD
hepta CDD
octa CDD
total PCDF
Furans
2378 TCDF
all tetra CDF
penta CDF
hexa CDF
hepta CDF
octa CDF
total PCDF
Woodstove Ash Samples
Run 01
a
ND(0.03)
ND(0.02)
ND(0.03)
ND(0.03)
0.01
0.01
a
ND(0.03)
ND(O.Ol)
ND(O.Ol)
ND(O.Ol)
ND(O.Ol)
ND(0.07)
Run 02
a
ND(O.Ol)
ND(0.03)
ND(0.04)
ND(0.03)
0.01
0.01
a
ND(0.02)
ND(O.Ol)
ND(0.02)
ND(0.02)
ND(O.Ol)
ND(0.08)
Run 03
a
ND(0.04)
ND(0.02)
ND(0.03)
ND(0.2)
0.09
0.09
a
ND(0.09)
M0(0.03)
MD(0.04)
MD(0.08)
MD(0.02)
MD(0.26)
Flue Wipe Samples
Run 02
a
ND(O.Ol)
ND(0.02)
ND(O.Ol)
ND(0.02)
0.06
0.06
a
ND(0.02)
ND(0.02)
ND(0.03)
ND(0.02)
ND(0.006)
ND(O.IO)
Run 03
a
ND(0.08)
ND(0.02)
ND(0.02)
0.04
0.3
0.34
a :
ND(0.02)
ND(0.03)
ND(O.Ol)
ND(0.04)
ND(0.03)
ND(0.13)
ND-species not detected (detection limit In parenthesis)
Analytical detection limit for the 2378 isomers is estimated to be less than half
of the detection limits shown for all tetra CDDs arid all tetra CDFs.
5-10
-------
TABLE 5-4. SUMMARY OF DIOXIN/FURAN PRECURSOR DATA
FOR SITE WS-A FEED SAMPLES
Precursor Category
Total Chlorinated Benzenes
Total Chlorinated Biphenyls
Total Chlorinated Phenols
Total Chloride
ND = not detected
Oak
ND
ND
ND
125 ug/g
Pine
ND
ND
ND
49 ug/g
5-11
-------
-------
6.0 SAMPLING LOCATIONS AND PROCEDURES
Details on the sampling locations and methods at Site WS-A are discussed
in Sections 6.1 through 6.3. Analytical procedures for continuous monitoring
of CO, C02, 02, and THC are included in Section 6.1. All other analytical
procedures are discussed in Section 7.0.
6.1 GASEOUS SAMPLING
Two types of gaseous samples were taken during this test program:
Modified Method 5 (MM5) and continuous emission monitoring (CEM). The
sampling locations and methods are further discussed in this section.
6.1.1. Gaseous Sampling Locations
The Tier 4 MM5 samples were taken at a single point in the center of the
8-inch diameter stack. The sampling location was approximately 14 feet (21
duct diameters) downstream of the stove/stack coupling and 2 feet (3 duct
diameters) upstream of the stack discharge. The RTI continuous monitoring
probe, which was the nearest upstream flow disturbance, was located 6 feet (9
duct diameters) from the Tier 4 MM5 probe.
6.1.2 Gaseous Sampling Procedures
Gaseous sampling procedures used during the testing are listed in
Table 6-1. These procedures are discussed in detail in the Tier 4 Quality
Assurance Project Plan (QAPP). A brief description of each method and any
necessary deviations from the procedures outlined in the QAPP are provided in
the following section.
6-1
-------
TABLE 6-1. SUMMARY OF GAS SAMPLING METHODS FOR SITE WS-A
Sample Location
Sample Type
or Parameter
Sample
Collection Method
Woodstove outlet
exhaust stack
Dioxin/Furan
Volumetric Flow
Molecular Weight
Moisture
Modified EPA Method 5
Vane Anemometer
Obtained from CEM Analysis
EPA Method 4
6-2
-------
6'1-2-1 Modified Method 5 (MM5). Gas sampling for dioxins and furans
was based on the October 1984 draft of the ASME chlorinated organic compound
sampling protocol. Minor deviations from the ASME protocol are discussed
later in this section. This sampling method is a modified version of EPA
Method 5 that includes a solid sorbent module-for-trapping vapor phase
organics. The MM5 sampling train was used to collect samples at the wood
stove outlet exhaust stack sampling location. Following sample recovery, the
various parts of the sample (filter, solvent rinses, sorbent traps, etc.) were
sent to the EPA's Troika laboratories to quantify 2,3,7,8-TCDD, tetra- through
octa- dioxin homologues, and tetra- through octa- furan homologues. A total
of three MM5 test runs was conducted at the sampling location, with one test
run being conducted at each location per test day. The MM5 samples were
collected at a constant rate over a 240-minute on-line sampling period outlet
with a sample flow rate of approximately 0.5 scfm.
A schematic diagram of the MM5 sampling train is shown in Figure 6-1.
Flue gas is pulled from the stack through a nozzle and a heated glass probe.
Particulate matter is removed from the gas stream by means of a glass fiber
filter housed in a teflon-sealed glass filter holder maintained at 248 + 25°F.
The gas passes through a sorbent trap for removal of organic constituents.
The trap consists of separate sections for (1) cooling the gas stream and (2)
adsorbing the organic compounds on Amberlite XAD-2 R resin. During the
woodstove testing the THC level was extremely high and a backup XAD module was
inserted to collect any compounds that might have broken through the first
module. A chilled impinger train following the sorbent trap is used to remove
water from the flue gas, and a dry gas meter was used to measure the sample
gas flow.
Modifications to the ASME protocol that were instituted for this test
site include the following:
6-3
-------
0)
3
-o
o
c
O)
ja
s-
o
GO
Q.
=3
o
CO
-C
(0
s-
LT)
•o
o
O)
O)
<£
•o
o
L/T !
o
o
o
M
T •*
II
I
6-4
-------
1. Sample recovery was modified to include water in the sample train
rinsing scheme. Water, acetone, and methylene chloride were used in
series to recover the probe, back half/coil, and first impinger
samples.
2. A back-up XAD trap was placed between the first and second
impingers.
3. Traversing was not performed since the flow rate on a woodstove is
approximately 30 ACFM, making measurement by a pitot tube
impossible. The flows were taken by a four-inch vane anemometer
located in the center of the eight-inch diameter duct. This was the
only point of representative flow.
4. The MM5 train condenser was oriented horizontally.
5. The volume sampled was maintained at 0.5 CFM through the meter in
order to ensure sufficient sample to analyze, which resulted in high
isokinetics.
6. The probe brush specified in the ASME protocol is inert material
with a stainless steel handle. To ensure cleanliness, a separate
nylon bristle brush attachable to a stainless steel handle was used
for each probe cleaning.
6-1-2-2 Volumetric Gas Flow Rate Determination The average flue gas
flow rate was calculated from the average gas velocity as directly measured by
the vane anemometer, the average flue gas temperature, the wet molecular
weight, and the absolute static pressure. Anenometer readings were taken
every 20 minutes during the test.
6-1-2-3 Hue Gas Moisture Determination The moisture content of the
flue gas was determined at the outlet sampling locations using EPA Method 4.
6-5
-------
Based on this method, a measured volume of particulate-free gas was pulled
through a chilled impinger train. The quantity of condensed water was
determined gravimetrically and then related to the volume of gas sampled to
determine the moisture content.
6.2 WOOD SAMPLING
Samples were taken of both types of wood burned (oak and pine). Oak was
burned during tests one and two and an oak sample was taken from a randomly
selected log in the woodpile. Chips were prepared from the sample using a
precleaned circular dato saw. Two identical oak samples were obtained from
tests one and two: a one-liter container was sent to RTI for total chlorine
analysis and a one-liter composite was retained at Radian/RTP for precursor
analysis. A pine sample for test three was treated similarly.
6.3 ASH SAMPLING
After each test run, samples were obtained of the bottom ash or wood
residue left in the firebox. This sample was collected and sent to Troika for
dioxin analysis.
6.4 STACK WIPE SAMPLING
After the oak burn and the pine burn, the exhaust stack was wiped with a
precleaned portion of glass wool to remove the creosote buildup from the stack
wall. This material is thought to contain the heavier organic compounds which
condense out on the stack lining. The glass wool containing the creosote was
placed in a wide mouth amber jar (precleaned) and sent to Troika for dioxin
analysis.
6-6
-------
7.0 ANALYTICAL PROCEDURES
Laboratory procedures u^d-to-qtrantify-dToxtus/furans- ami dtoxin/furan
precursors in the Tier 4 samples are described in this section. Samples
analyzed by the EPA's Troika laboratories for dioxin/furan content included
MM5 train samples and the woodstove ash and flue wipe samples. Procedures
used for the dioxin/furan analyses are described in detail in the Analytical
Procedures and QA Plan for the Analysis of Tetra through Octa Chlorinated
Dibenzo-p-dioxins and Dibenzofurans in Samples from Tier 4 Combustion and
Incineration Processes Addendum to EPA 600/3-85/019, May 1986 These
procedures are summarized in Section 7.1. Wood feed samples were analyzed by
Radian to determine concentrations of chlorinated phenols (CP), chlorobenzenes
(CB), and polychlorinated biphenyls (PCB's). Total chlorine analyses of the
wood feed samples were performed by RTI. Procedures used for these analyses
are detailed in Section 7.2.
7.1 DIOXINS/FURANS
The analytical procedures described in this section were used by Troika
for dioxin/furan analysis of MM5 train samples and the woodstove ash and flue
wipe samples from Site WS-A. Samples consisting of organic solvents, aqueous
solutions, and solids were prepared for analysis using slightly different
procedures. The organic solvent samples consisted of rinses from the MM5
probe, nozzle, filter housing, and condenser coil. Aqueous samples consisted
of mipinger catch solutions; solid samples included filters, XAD resin
woodstove ash samples, and flue wipe samples. Isotopically-labeled surrogate
compounds were added to all samples prior to extraction to allow determination
of method efficiency.
Organic liquid samples (e.g., acetone and methylene chloride-based MM5
train rinses) were concentrated using a nitrogen blowdown apparatus. The
residue, which contained particulate matter from the MM5 train probe and
7-1
-------
nozzle, was combined with the filter and handled: as a solid sample. Solid
samples were extracted with toluene in a Soxhlet. apparatus for a period of at
least 16 hours. The extract was concentrated by nitrogen blowdown and
subjected to chromatographic cleanup procedures. I
Aqueous solutions such as the MM5 train impinger samples were extracted;
with hexane by vigorous shaking for a 3-hour period. This extraction
procedure was repeated three times, with the organic fractions ultimately
being combined and concentrated for chromatographic cleanup.
The cleanup procedure involved using liquid chromatographic columns to !
separate the compounds of interest from other compounds present in the ;
samples. Four different types of columns were used: a combination acid andi
base modified silica gel column, a basic alumina column, a PX-21 carbon/celite
545 column, and a silica/diol micro column. These were used in successive
steps, with the last two being used only if necessary.
The cleaned samples were analyzed using high resolution gas
chromatography/mass spectrometry (GC/MS). The GC/MS conditions for the
analyses were as follows: !
Gas Chromatoqraph - Injector configured for capillary column, split!ess
injection, injector temperature 280°C, helium carrier gas at 1.2 ml/min
initial column temperature 100°C, final column temperature 240°C, interface
temperature 270°C.
Mass Spectrometer - Varian/HAT Model 311A, electron energy 70ev, filament
emission IMA, mass resolution 8,000 to 10,000, ion source temperature 270°C.
7.2 DIOXIN/FURAN PRECURSORS
Feed samples for Site WS-A were analyzed by Radian/RTP for CP, CB, and
PCB's by GC/MS. Analytical procedures are discussed in the following
sections.
7-2
-------
7.2.1 GC/MS Analyses
The analytical procedures used for determining CP, CB, and PCB
concentrations in feed samples are modified versions of procedures typically
used for the analysis of MM5 train components. These procedures involve
initial extraction of the sample with an appropriate solvent, preliminary
separation of the compounds of interest by solvent partitioning and liquid
chromatography, and analysis of the processed fractions. Solutions containing
CB and PCB are injected directly into the GC/MS, and solutions containing CP
are derivatized prior to injection, Details on the procedures used for Site
WS-A samples are provided in the sections below.
7'2-1-1 Sample Preparation. A flow chart for the sample preparation
procedure used for Site WS-A feed samples is shown in Figure 7-1. The first
step in the procedure involved adding labeled surrogate compounds to provide a
measure of extraction method efficiency. The next step involved adding a
mixture of 0.5 N NaOH and MeCl,, to the sample and sonicating the sample for 30
minutes. The NaOH and MeCl2 mixture converts the acid compounds to their
salts and collects base/neutrals in the organic solvent. The sonicated sample
was filtered and rinsed with 0.5 N NaOH. The filtrate was extracted three
times in a separatory funnel with MeCl,, and the aqueous and organic fractions
were saved for derivatization and/or further cleanup. The aqueous fraction
(or acids portion) was acidified to pH2 with HC1 and then extracted three
times with MeCl2. The MeCl,, from this extraction was dried with anhydrous
Na2S04, exchanged to benzene, and concentrated using a nitrogen blowdown
apparatus. Acetylation of any CP present in the sample involved the following
steps:
2.
' 2*° mL acetonitrile, 50 uL pyridine, and 20 uL
acetic anhydride were added to the extract. The test tube
containing the extract was placed in a 60°C water bath for 15
minutes and was shaken 30 seconds every 2 minutes.
LThjLV1 ? H?P°* W6re added to the test tube' and the sample was
agitated for 2 minutes on a wrist action shaker.
2rga?iC layer was removed ^d the quantisation standard was
added. The sample was concentrated in a Reacti-Vial at room
temperature (using prepurified N£) to 1 ml prior to GC/MS analysis.
7-3
-------
aog Sample
t.OmL Bate/Neutral Surrogates
1.0mL Acid Surrogate*
Sonleata with 2SOmL
0.5 H NaOH and ISmL MaClie
Flltar thru Buehnar and
Rlnaa with 0.5 N. NaOH
Extract 3x with MaClj
In Saparatory Funnal
Aquaoua
Organic
Adjust to pH2 with HCl;
Extract 3x with MaCI2
Olacard Aquaoua
Flltar with Na2S04
Dlacard All
AcM/HjO Lay«>ra
Add 3OmL Cone.
Shaka 4 mln; Attarnata
with 30mL dtotltlad H2O;
Rapaat until acid la claar.
FHtar with
Add 10ml. Benzene
Concantrata to 1mL
To ImL Banzana add:
2.0ml. lao oetana
2.0ml. Acatonltrlla
SOmL Pyrldma
20ml. Acatlc Anlydrlda
Add lOmL Haxanaa;
Concantrata to ImL
Pta-wat Column
with 2OmL Haxaiwa
Chromatography column with:
1.0g SlMca
2.0g 33% NaOH SIHea
2.0g SIHea
Put In 6O C Hf bath
for 15 mlnutaa. Shaking
3O aaconda avary 2 mlnutaa.
Ehita with OOmL Haxanaa;
Concantrata to ImL
Add 8mL of 0.01 N
H3PO4; Shaka 2 mlnutaa.
Mmt-cotuntn with
1.0g Alumina
Ekita with 2OmL 50/50
MaCI2/Haxanaa
Add Quantttatton Standarda;
Concantrata to ImL
QC/US Anaryala
Figure 7-1. Sample Preparation Flow Diagram for
Site WS-A Precursor Analyses.
7-4
-------
Cleanup of the organic (or base/neutrals) layers from the first MeCl
extraction involved successively washing the extract with concentrated H?SO,
and double-distilled water. The acid or water was added in a 30 ml portion
and the sample was shaken for 2 minutes. After the aqueous (or acid) and
organic layers were completely separated, the aqueous (or acid) layer was
discarded. The acid washing procedure was repeated until the acid layer was
colorless. The organic fraction from the final wash was dried with anhydrous
Na2S04, exchanged to hexane and concentrated. Final cleanup of the sample by
column chromatography involved the following procedure.
A glass micro-column, 20 mm o.d. x 230 mm in length, tapered to 6 mm o.d
on one end was prepared. The column was packed with a plug of silanized glass
wool, followed successively by 1.0 g silica, 2.0 g silica containing
33 percent (w/w) 1 N NaOH, and 2.0 g silica. The concentrated extract was
quantitatively transferred to the column and eluted with 90 ml hexane. The
entire eluate was collected and concentrated to a volume of 1 ml in a
centrifuge tube.
A disposable liquid chromatography min-column was constructed by cutting
off a 5-mL Pyrex disposable pipette at the 2.0 ml mark and packing the lower
portion of the tube with a small plug of silanized glass wool, followed by 1 g
of Woehlm basic alumina. The alumina had been previously activated for at
least 16 hours at 600°C in a muffle furnace and cooled in a desiccator for 30
minutes just before use. The concentrated eluate from above was
quantitatively transferred onto the liquid chromatography column. The
centrifuge tube was rinsed consecutively with two 0.3 ml portions of a
3 percent MeCl2:hexane solution, and the rinses were transferred to the liquid
chromatography column.
The liquid chromatography column was eluted with 20 ml of a 50 percent
(v/v) MeCl2:hexane solution, and the eluate was concentrated to a volume of
approximately 1 ml by heating the tubes in a water bath while passing a stream
of prepurified NZ over the solutions. The quantitation standard was added and
the final volume was adjusted to 1.0 ml prior to GC/MS analysis.
7-5
-------
7.2.1.2 Analyses. Analyses for CP, CB, and PCB's present in the feed
sample extracts were performed with a Finnigan Model 5100 mass spectrometer
using selected ion monitoring. A fused silica capillary column was used for
chromatographic separation of the compounds of interest. Analytical
conditions, for-the-GC/MS analysis are shown in Table 7-1.
Tuning of the GC/MS was performed daily as specified in the Tier 4
Quality Assurance Project Plan. An internal-standard calibration procedure
was used for sample quantitation. Compounds of interest were calibrated
against a fixed concentration of either d,2-chrysene (for CO) or
dg-naphthalene (for CB, PCB). Components of the calibrations solution are
shown in Table 7-2. For multi-point calibrations, this solution was injected
at levels of 10, 50, 100, and 150 ng/mL.
The instrument detection limit for the analytes of interest (i.e., CP,
CB, and PCB) was estimated to be approximately 500 pg on column. For a 50 g
sample and 100 percent recovery of the analyte, this corresponds to a feed
sample detection limit of 10 ppb.
7.3 TOTAL CHLORINE ANALYSIS
Total chlorine concentrations in feed samples were determined by Parr
Bomb combustion followed by ion chromatography (1C). An 0.5 g sample was
placed in the Parr Bomb with 10 mL of a 50 g/L Na2C03 solution. After
combustion of the samples according to standard procedures (ASTM 2015), the
contents of the bomb were rinsed into a 100 mL flask and diluted to 100 mL.
The resulting solution was analyzed for chloride concentration (Cl')by 1C
using standard anion conditions.
7-6
-------
TABLE 7-1. INSTRUMENT CONDITIONS FOR GC/MS PRECURSOR ANALYSES
Parameter
Column" *" " '
Injector Temperature
Separator Oven Temperature
Column Head Pressure
He flow rate
GC program
Emission Current
Electron Energy
Injection Mode
Mode
Chlorobenzenes/
Polychlorinated biphenyls
30 m WB DB-5 (1.0 u film
thickness) fused silica
capillary
290°C
290°C
Chlorophenols
40(4)-290°C,
10°/min & hold
0.50 ma
70 ev
290°C
290°C
9 psi
1 mL/min
9 psi
1 mL/min
40(1)-290°C,
120/min & hold
0.50 ma
70 ev
Splitless 0.6 min, then 10:1 split
Electron ionization, Selected Ion
Monitoring
7-7
-------
TABLE 7-2. COMPONENTS OF THE CALIBRATION SOLUTION
Base/Neutrals
4-chlorobiphenyl
3,3'-dich1orobiphenyl
2,4',5-trichlorobiphenyl
3,3*4,4'-tetrachlorobi phenyl
2,2',6,6'-tetrachlorobiphenyl
2,2,4,5,6-pentachlorobiphenyl
2,2',4,4',5,5'-hexachlorobiphenyl
2,2',3,4,4',5',6-heptachlorobiphenyl
2,2',3,3',4,4',5,5'-octachlorobiphenyl
2,2',3,3',4,4',5,6,6'-nonachlorobiphenyl
decachlorobi phenyl
p-di chlorobenzene
1,2,4-tri chlorobenzene
1,2,3,5-tetrachlorobenzene
pentachlorobenzene
hexachlorobenzene
d4-l,4-dichlorobenzene (SS)
3-bromobiphenyl (SS)
2,2',5,5'-tetrabromobiphenyl (SS)
2,2',4,4',6,6'-hexabromobiphenyl (SS)
2
octachloronaphthalene (QS)
d,Q-phenanthrene (QS)
(QS)
2,5-dichlorophenol !
2,3-dichlorophenol
2,6-dichlorophenol
3,5-dichlorophenol
3,4-dichlorophenol
2,3,5-trichlorophenol
2,3,6-trichlorophenol
3,4,5-trichlorophenol
2,4,5-trichlorophenol
2,3,4-trichlorophenol
2,3,5,6-tetrachlorophenol
pentachlorophenol
dg-phenol (SS)
dg-2-chlorophenol (SS)
Cg-pentachlorophenol (SS)
dg-naphthalene (QS)
2,4,6-tribromophenol (QS) '.
d,Q-phenanthrene (QS) '
(QS)
1
Surrogate standard.
"Quantitation standard.
7-8
-------
8«° QUALITY ASSURANCE/-QUALITY CONTROL (QA/QC),
This section summarizes the results of quality assurance and quality
control (QA/QC) activities for WS-A. No valid flue gas dioxin/furan emissions
data were obtained for this site. The sample extracts exhibited evidence of
significant hydrocarbon contamination. The ash dioxin/furan data for this
site were generally within the QC specifications presented in the Tier 4 QAPP
All of the surrogate recoveries for labeled TCDD's were within the specified
limits of 50 to 120 percent. The surrogate recoveries for the octa-CDD
species from the Run 03 sample was the only run not within the QC limits of 40
to 120 percent for hepta- and octa-CDD's. The results of the analysis of the
fort1fied laboratory QC sample were all within the Tier 4 objective of
±50 percent. These data indicate that the dioxin/furan results are within
accuracy criteria specified for Tier 4.
The dioxin/furan precursor analysis of the feed samples was not as
accurate as the dioxin/furan homologue analysis. Surrogate recoveries of the
six isotopically labeled compounds used as surrogates varied considerably by
specific surrogate species and by wood type. The surrogate recoveries were
generally below the specified 50 percent objective. In spite of the low
recoveries of the surrogates for some feed samples, the resulting analytical
sensitivity for the target analytes was considered acceptable for the purpose
of this study.
The following sections summarize the reselts of all WS-A QA/QC
activities. Manual gas sampling methods are considered in Section 8.1 and the
laboratory analyses QA/QC activities are summarized in Section 8.2.
8-1
-------
8.1 MANUAL GAS SAMPLING
Manual gas sampling methods used at WS-A included Modified Method 5
(MM5), and EPA Methods 1 through 4. These methods are discussed in
Section 6.0. Quality assurance and quality control (QA/QC) activities for the
manual methods centered around 1) equipment calibration, 2) glassware
precleaning, 3) procedural QC checks, and 4) sample custody procedures. Key
activities and QC results in each of these areas are discussed in this
section. Also discussed are problems encountered that may have affected data
quality.
Pre-test calibrations or inspections were conducted on the anemometer,
sampling nozzles, temperature sensors and analytical balances. Both pre-test
and post-test calibrations were also performed on dry gas meters. All of this
equipment met the calibration criteria specified in the QAPP. Differences in
pre- and post-test dry gas meter calibrations were less than 3.0 percent.
An extensive precleaning procedure was implemented for all sample train
glassware and sample containers. This cleaning procedure, which is outlined
in Table 8-1, was implemented to minimize the potential for sample
contamination with substances that could potentially interfere with the
analysis for dioxins and furans. To minimize the potential for contamination
in the field, all sample train glassware was kept capped until use and a
controlled environment was maintained in the recovery trailer during sample
train assembly and recovery. i
Procedural QC activities during manual gas sampling focused on:
inspecting equipment visually,
collecting sampling train blanks,
conducting pre-test, port change, and post-test sample train
leak checks,
maintaining proper temperatures at the filter housing, sorbent
trap and impinger train,
maintaining sampling rates, and
recording all data on Preformatted data sheets.
8-2
-------
TABLE 8-1. GLASSWARE PRECLEANING PROCEDURE
J USE DISPOSABLE GLOVES AND ADEQUATE VENTILATTDN
1. Soak all glassware in hot soapy water (Alconox ) 50°C or higher.
2. Distilled/deionized H20 rinse (X3)a
3. Chromerge rinse if glass, otherwise skip to 6.
4. High purity liquid chromatography grade ti^O rinse (X3).
5. Acetone rinse (X3), (pesticide grade).
6. Hexane rinse (X3), (pesticide grade).
7. Cap glassware with clean glass plugs or hexane rinsed aluminum foil
(X3) = three times.
8-3
-------
Sample custody procedures used during this program emphasized careful
documentation of the samples collected and the use of chain-of-custody records
for samples to be transported. Steps taken to identify and document samples
collected included labeling each sample with a unique alphanumeric code and
logging the sample in a master sample logbook. All samples shipped to Troika
or returned to Radian were also logged on chain-of-custody records that were
signed by the sampler at shipment and then by the receiving laboratory when
the samples arrived. Each sample container was also sealed with
chain-of-custody seal so that the container cou'ld not be opened without
tearing the seal. ;
8.2 LABORATORY ANALYSIS
QA/QC activities were carried out for dioxin/furan and precursor analyses
performed on Site WS-A samples. The dioxin/furan analyses are considered in
Section 8.2.1, and the precursor analyses are considered in Section 8.2.2.
8.2.1 Pioxin/Furan Analvsis
8.2.1.1 Recovery of Labelled Surrogate Compounds. No valid flue gas
dioxin/furan emissions data were obtained for the woodstove. The four
labelled internal standards spiked in the MM5 train samples could not be
recovered. This indicates that both the aqueous; and XAD-2 portions of the
samples caused serious sample preparation problems. The sample extracts were
reported to be yellow in color, and exhibited evidence of significant
hydrocarbon contamination. This resulted in peak broadening and overloading
of the alumina and carbon columns. The Troika laboratory report submitted fpr
this test site concluded that the analytical results did not yield any valid
indication of whether dioxins/furans were present in the MM5 samples.
Analytical recovery values for the labelled TCDD and octa-CDD species
spiked onto woodstove ash and creosote flue wipe samples are shown in
Table 8-2. In general, the surrogate recoveries for these samples are within
the Tier 4 QA objectives of 50 to 120 percent for the TCDD species and 40 to
8-4
-------
TABLE 8-2. PERCENT SURROGATE RECOVERIES FOR SITE WS-A
ASH AND FLUE WIPE DIOXIN/FURAN ANALYSES
Sample
13,
TCDD
13,
OCDD
Woodstove Ash Samples
Run 01
Run 02
Run 03
Creosote Flue Wipe Samples
Run 02
Run 03
94
93
99
61
45
68
72
25
92
52
8-5
-------
120 percent for the octa-CDD species. The only exception was the 25 percent
recovery value for the octa-CDD species for the woodstove ash sample from
Run 03.
8.2.1.2 Sample Blanks. Table 8-3 summarizes the analytical results reported
by Troika for internal laboratory blanks, laboratory fortified QC samples, and
field recovery blank MM5 train samples. In general, these data showed good
surrogate recoveries, with values within the Tier 4 QA objectives of 50 to 120
percent for the tetra-CDD surrogates and 40 to 120 percent for the hepta- and
octa-CDD surrogates. Comparisons of the measured and spiked values for the
laboratory fortified QC samples were all within the Tier 4 QA objectives of ±
50 percent. The field blank MM5 sample was clean except for very small
quantities of hepta-CDD and octa-CDD. The analytical results for the QA
samples at Site WS-A indicate that the sample preparation procedures and GC/MS
detection technique were performing in a satisfactory manner, and that no
significant MM5 train blanking problems occurred.
8.2.2 Precursor Analyses
Table 8-4 presents analytical recovery efficiencies for six isotopically
labeled compounds used as surrogates for the target precursor analytes in the
Site WS-A feed samples. The surrogate recovery values in Table 8-4 vary
considerably by specific surrogate species and by wood type. Several of the
recoveries are below the 50 percent objective stated in the Tier 4 QA Project
Plan and are below those generally considered achievable when analyzing for
similar compounds in water or from MM5 train components. In spite of the
relatively low surrogate recovery values for some of the feed samples, the
resulting analytical sensitivity for the target analytes was considered
acceptable for the purpose of this study.
8-6
-------
TABLE 8-3. ANALYSIS RESULTS FOR QUALITY CONTROL SAMPLES
Flue Gas Qua!
ity Control Samples
Fortified 1 ahnv^torv Or Samnlp
Compound
Dioxins
2378 TCDD
Other TCDD
Penta CDD
Hexa CDD
Hepta CDD
Octa CDD
Furans
2378 TCDF
Other TCDF
Penta CDF
Hexa CDF
Hepta CDF
Octa CDF
37 -Trnn
U4
13- -TCDD
C12
37n -Hepta CDD
U4
13- -Octa CDD
C12
Laboratory
Blank
ND
ND
ND
ND
ND
0.1
NO
ND
ND
ND
ND
ND
Measured
Value
Amount Detected
0.3
ND
ND
1.0
2.6
3.0
0.4
ND
0.6
0.9
2.5
3.0
True ,
Value3' D
— — ^— — ___ — _
(Nanograms per Sample)
0.4 (-25)
ND (0)
ND (0)
1.6 (-37)
2.4 (8)
3.2 (-6)
0.4 (0)
NO (0)
0.8 (-25)
1.6 (-44)
2.4 (8)
3.2 (-6)
Field Blank
MM5 Train
Outlet
'
ND
NO
ND
ND
0.2
0.4
ND
ND
ND
ND
ND
ND
Surrogate Recoveries (Percent)
100
96
41
42
84
104
44
49
NA
NA
NA
NA
80
74
83
90
of each homologue spiked into the laboratory
in parenthesis is the percentage difference between the measured and the
Measured Value - Trim
True Value
x 10°
NR = Not reported by Troika
ND = Not detected.
NA = Not applicable.
8-7
-------
TABLE 8-4. PERCENT SURROGATE RECOVERIES FOR SITE WS-A FEED SAMPLES
Surrogate
Compound
Base Neutral s Fract 1 on
d . -di chl orobenzene
2», 5, 5' tetra
bromobiphenyl
2', 4, 4', 6, 6'
hexabromobi phenyl
Acids Fraction
dg-phenol
d«-2-chlorophenol
Cg-pentachTorophenol
Percent Surrogate Recovery
Oaka Pine
34, 25 18
93, 94 45
50, 35 1
14, 13 6
30, 28 21
26, 20 26
Duplicate analyses were performed on the oak sample .
8-8
-------
APPENDIX A
FIELD SAMPLING DATA
-------
-------
APPENDIX A-l
MODIFIED METHOD 5 AND
EPA METHODS 1-4 FIELD RESULTS
-------
-------
RADIAN SOURCE
EPA METHOD 2 -
( R A W DATA)
PLANT
PLANT SITE
SAMPLING LOCATION
TEST #
DATE
TEST PERIOD
TEST
WOOD STOVE TEST FOR DIOXIN PROJECT
RTI//DURHAM , NORTH CAROLINA
WOOD STOVE EXHAUST
WS-MM5-01
04/11/85
1353-1753
PARAMETER
Sampling time (min.)
Barometric Pressure (in.Hg)
Sampling nozzle diameter (in.)
Meter Volume (cu.ft.)
Meter Pressure (in.H20)
Meter Temperature (F)
Stack dimension (sq.in.)
Stack Static Pressure (in.H20)
Stack Moisture Collected (gm)
Absolute stack pressure(in Hg)
Average stack temperature (F)
Percent C02
Percent 02
Percent N2
Delps Subroutine result
DGM Factor
Pitot Constant
VALUE
240
29.9
.5
132.02
.91
104.04
50.2656
-.00001
133.7
29.9
124.0833
3 .28
16.8
79.92
.3771501
.997
.99
-------
IAN SOURCE
RAD
EPA METHODS 2 -
FINAL RESULTS
PLANT
PLANT SITE
TEST
5
SAMPLING LOCATION
TEST #
DATE
TEST PERIOD
WOOD STOVE TEST FOR DIOXIN PROJECT
RTI//DURHAM , NORTH CAROLINA
WOOD STOVE EXHAUST
WS-MM5-01
04/11/85
1353-1753
PARAMETER
RESULT
Vm(dscf)
Vm(dscm)
Vv gas(scf)
Vw gas (scm)
Z moisture
Md
MWd
MW
Vs(fpm)
Vs (mpm)
Flow(acfm)
Flow(acmm)
Flow(dscfm)
FlowCdscmm)
Z I
Z EA
123 .4069
3.494882
6.303955
.178528
4.860008
.9513999
29.1968
28.65264
65.4332
19.94915
22.84055
.6468443
19.63082
.5559447
670.9438
390.7995
Program Revision:I/16/84
-------
RAD
EPA
( R A W
PLANT
PLANT SITE
SAMPLING LOCATION
TEST #
DATE
TEST PERIOD
TEST
IAN SOURCE
METHOD 2-o
R??//,Snp°uLTESL!OR DIOXIN 'PROJECT
WS-MM5-02
04/17/85
1515-1915
PARAMETER
Sampling time (min.)
Barometric Pressure (in.He)
Sampling nozzle diameter (in.)
Meter Volume (cu.ft.)
Meter Pressure (in.H20)
Meter Temperature (F)
Stack dimension (sq.in.)
Stack Static Pressure (U.H20)
Stack Moisture Collected (gm)
Absolute stack pressureUn Hg)
Average stack temperature (F)
Percent C02
Percent 02
Percent N2
Delps Subroutine result
DGM Factor
Pitot Constant
.452
VALUE
240
29.9
.5
122
.71
113.6
50.2656
-.00001
203.4
29.9
150.8958
3.15
17.54
79.31
.5180025
.997
.99
-------
RADIAN SOURCE
EPA METHODS 2 -
FINAL RESULTS
PLANT
PLANT SITE
SAMPLING LOCATION
TEST f
DATE
TEST PERIOD
TEST
5
WOOD STOVE TEST FOR DIOXIN PROJECT
RTI//DURHAM , NORTH CAROLINA
WOOD STOVE EXHAUST
WS-MM5-02
04/17/85
1515-1915
PARAMETER
RESULT
Vm(dscf)
Vm(dscm)
Vw gas(scf)
Vw gas (scm)
Z moisture
Md
MWd
MW
Vs(fpm)
Vs (mpm)
Flow(acfm)
Flow(acmm)
Flow(dscfm)
Flow(dscmm)
Z I
Z EA
112.5001
3.186004
9.59031
.2715976
7.855087
.9214491
29.2056
28.32539
90.38786
27.55728
31.55139
.8935352
25.11113
.7111472
478.1586
516.2106
Program Revision:I/16/84
-------
RADIAN SOURCE
EPA METHOD 2 -
( R A W DATA)
PLANT
PLANT SITE
.SAMPLING LOCATION
TEST #
DATE
TEST PERIOD
TEST
WOOD STOVE TEST FOR DIOXIN PROJECT
RTI//DURHAM , NORTH CAROLINA
WOOD STOVE EXHAUST
WS-MM5-03
05/01/85
1220-1620
PARAMETER
VALUE
Sampling time (min.)
Barometric Pressure (in.Hg)
Sampling nozzle diameter (in.)
Meter Volume (cu.ft.)
Meter Pressure (in.H20)
Meter Temperature (F)
Stack dimension (sq.in.)
Stack Static Pressure (in.H20)
Stack Moisture Collected (gm)
Absolute stack pressure(in Hg)
Average stack temperature (F)
Percent C02
Percent 02
Percent N2
Delps Subroutine result
DGM Factor
Pitot Constant
240
29.9
.5
123.18
.73
107.5
50.2656
-.00001
140.8
29.9
169.2041
3.88
16.56
79.56
.5723236
.997
.99
-------
SOURCE
RADIAN
EPA METHODS 2 -
FINAL RESULTS
PLANT
PLANT SITE
SAMPLING LOCATION
TEST #
DATE
TEST PERIOD
TEST
5
WOOD STOVE TEST FOR DIOXIN PROJECT
RTI//DURHAM , NORTH CAROLINA
WOOD STOVE EXHAUST
WS-MM5-03
05/01/85
1220-1620
PARAMETER
RESULT
Vm(dscf)
Vm(dscm)
Vw gas(scf)
Vv gas (scm)
Z moisture
Md
MWd
MW
Vs(fpm)
Vs (mpm)
Flow(acfm)
Flow(acmm)
Flow(dscfm)
Flow(dscmm)
Z I
Z EA
114.391
3.239554
6.638721
.1880086
5.485198
.945148
29.2832
28.66429
99.27441
30.26659
34.65339
.981384
27.46613
.7778408
444.5079
372.6507
Program Revision:I/16/84
-------
APPENDIX A-2
CONTINUOUS EMISSION MONITORING RESULTS
-------
-------
C£MS DATA - SITE WS - TEST 1
+*
**
*»
**
**
»•»
**
**
#*
»+
**
*+
**
*•»
*+
*+
*•*
**
*•»
»*
.»*
**
**
**
**
##
*•»
»#
**
•»•»
**
**
*•»
*»
**
**
*+
**
»»
**
**
•»•»
*•»
**
•**
*•»
**
**
**
**
»-»
**
**
**
»*
**
•»•»
**
FACTOR
FOR 3V. 02
NORMALIZATION
OF
J3THER PROCESS
3ASES
•— — — • «___
4. 9571
4.4450
4.8365
4.7217
4.8588
4.8222
4.6871
4.6663
4.4952
4.2731
4. 1027
4.0297
4.2619
4.3247
4.5921
5.0112
4.2167
3.9066
3.7597
3.9646
4.0352
3.8412
3. 921 1
3.8045
3. 9849
3.9117
4. 5593
4. 7354
4.9571
5.3561
5.3290
5.2416
5. 1319
5.0751
5.0823
4.5408
4.2337
4.2337
4.2277
4.2277
4.2447
4.2670
4.2619
4.2337
3.9401
3.9066
3.S922
3. 9358
4. 1502
**
*•*
**
*»
**
**
**
**
**
**
•**
**
**
**
**
**
**
+*
**
»*
*•»
**
»*
**
**
**
**
**
*+
**
**
**
*»
**
**
»*
»*
*•»
*•»
»*
. »*
**
**
**
**
**
**
**
**
**
•»*
*•»
»*
•»*
»*
*•»
**
»*
C°RRECTED DATA -
TIME
1355
1400
1405
1410
1415
1420
1425
1430
1435
1440
1445
1450
1455
1500
.1505
1510
1515
1520
1525
1530
1535
1540
1545
1550
1555
1600
1605
1610
1615
1620
1625
1630
1635
1640
1645
1650
1655
1700
1705
1710
1715
1720
1725
1730
1735
1740
1745
1750
1755
02 CO
C/.V) (PPMV)
@ 37. 02
17.3 39909.4
16.9 40733.9
17.2 40389.9
17.1 40748.4
17.2 38846.5
17.2 41292.5
17.1 40398.0
17.1 39799.0
16.9 40893.1
16.7 41346.5
16.5 39607.3
16.5 36751.0
16.7 33409.1
16.8 33373.8
17.0 33646.3
17.3 32417.4
16.7 35357.2
16.3 34276.4
16.1 33491.5
16.4 32668.0
16.5 28621.4
16.2 32976.7
16.3 37027.3
16.2 35712.5
16.4 30795.0
16.3 37196.5
17.0 32134.3
17.1 29842.8
17.3 27487.0
17-. 6 25468.1
1.7.5 24689.1
17.5 24514.9
17.4 24284. 1
17.4 23619.7
17.4 24278.3
17.0 27908.0
16.7 28662.0
16.7 28662.0
16.7 28714.4
16.7 28291.6
16.7 28218.9
16.7 28413.8
16.7 28614.4
16.7 29322.5
16.4 30713.3
16.3 29319.0
16.3 28646.2
16.4 28530.6
16.6 28744.6
C02 THC
C/.V) (PPMV)
a 3X 02 @ 3X 02
16.8 46151.9
17.4 41384.2
, 16. 5_. 45029.6-
16.9 43960.5
16.3 45237.3
17.0 44896.1
16.9 43638.2
16.6 43444.8
16.5 41851.9
16.8 39783.8
16.5 38197.2
16.0 37517.9
15.3 39679.6
15.6 40264.4
15.5 42489.3
14.4 45282.7
15.8 38472.6
15.9 36371.5
15.5 34986.8
15.3 36911.3
15.0 37243.4
15.5 35762.7
15.5 36507.0
15.7 35420.7
14.8 36852.4
15.6 36419.2
13.7 42448.9
13.0 44088.5
12.4 45889.1
11.2 49533.5
10.7 49148.0
10.5 48534.9
10. 8 47531.5
10.5 46783.6
10.7 47142.8
12.4 41962.8
13.0 39416.8
12.9 39416.8
12.9 39361.0
12.7 .39361.0
12.8 39519.7
12.8 39726.9
13.0 39679.6
13.0 39416.8
13.8 36683.8
13.6 36371.5
13.6 36237. 1
13.1 36643.4
12.8 38640.0
CEMS DATA - SITE WS - TEST 1
NO. PTS.
MEAN
STD. DEV.
*
49
4.4121
0.4
NO. PTS.
MEAN
STD. DEV.
49
16.8
0.4
49
32464.6
5366.3
49
14.3
2.0
49
40965.2
4057.0
2' NOX,and THC vaU.es are corrected to -
-% 0-
-------
CEMS DATA - SITE WS - TEST 2
*•
»»
«*
**
»*
»*
*»
•**
*»
**
•*
«*
»*
**
*»
.**
»*
»»
*»
•»*
FACTOR
FOR T/C 02
NORMALIZATION
OF
OTHER PROCESS
BASES
**
»»
»*
*«
*»
**
**
*»
6.4388
6.3229
6.2478
6.2984
6.4112
6.3860
6.3611
6.2741
6.1176
6.1176
6.0596
5.9906
5.9567
5.6008
5.2771
5.3385
5.2865
5.6916
5.3851
5.0437
5.2959
5.6413
5.5729
5.4857
5.4774
4.8974
4.7367
4.7155
4.9722
4.9722
4.8974
4.9122
4.8235
4.9352
5.1585
5.0594
5.0281
5.0594
4.9352
4.9122
4.9271
4.8509
4.8378
4.8300
4.8235
4.8235
4.8235
4.9189
4.9041
**
**
**
»*
*•»
*»
**
»*
**
**
*•»
»*
•»•»
*•»
•»*
*•»
**
**
**
**
**
*•»
**
NORMALIZED /CORRECTED DATA - WITH
ACTUAL 02 »
»*
**
»*
*»
ME
1514
1519
1524
1529
1534
1539
1544
1549
1554
1559
1604
1609
1614
1619
1624
1629
1634
1639
1644
1649
1654
1659
1704
1709
1714
1719
1724
1729
1734
1739
1744
1749
1754-
1759
1804 '
1809
1814
1819
1824
1829
1834
1839
1844
1849
1854
1859
1904
1909
1914
02
(XV)
18. 1
18. 1
18.0
18. 1
18.1
18. 1
18. 1
18.0
18.0
18.0
17.9
17.9
17.9
17.7
17.5
17.5
17.5
17.8
17.6
17.4
17.5
17.7
17.7
17.6
17.6
17.2
17.1
17.1
17.3
17.3
17.2
17.3
17.2
17.3
17.4
17.4
17.3
17.4
17.3
17.3
17.3
17.2
17.2
17.2
17.2
17.2
17.2
17.3
17.3
CO
(PPMV)
8 3X 02
50B45.0
501538.6
491370. 1
49511.6
50398.2
491349.2
48528.5
48624.3
47753.7
48157.5
481334. 3
47224. 1
47088.0
48110.5
46491.5
47208.4
46166.8
39146.6
41976.7
46886. 0
45020. 1
41667.0
41038.5
40759. 1
41606.0
40335.0
33573.7
38110.6
42432. 9
37053.0
37470.0
38830.8
39938. 6
36885.7
34629.0
34575.6
35251.9
36083.3
36342. 9
36011.2
36120.3
35989.2
35891.9
36635.6
36533.2
36692.3
36210.0
36223.0
34760. 3
C02
C/CV)
3 3V. 02
19.4
19.0
18.7
18.3
18.5
18.0
17.9
17.9
17.8
17.8
17.8
17.8
17.9
18.4
18. 1
18.0
17.8
17.1
17.2
18.0
17.3
16.4
16.4
16.1
16.5
18.1
17.8
17.4
16.6
15.6
16.0
16.0
16.0
15.6
15.2
15.4
15.5
15.4
15.3
15.6
15.6
15.7
15.6
15.8
15.7
15.8
15.7
15.7
16.0
THC
(PPMV)
& 37. 02
41414.7
54111.0
56311.6
58757.6
64663. 1
63266.3
61975.7
63054.7
61261.3
61579.4
60123.2
59744.5
58221.2
60208. 1
55573.4
57581. 1
55534.4
43386.9
52795.3
52363.4
53467.0
49717.2
49999.6
48011.3
49728.9
43160.8
36737.9
39704.4
47072.0
41358.9
40232. 1
40279. 9
41477.3
40745.1
36873.0
33912.9
35709.5
33265.3
37492.8
35937.5
35494.5
34499.9
33221.4
35090.0
33687.3
35365.9
34058. 7
33222.5
30866.5
CEMS DATA - SITE WS - TEST 2
NO. PTS.
MEAN
STD. DEV.
49
5.3843
0.6
NO. PTS.
MEAN
STD. DEV.
49
17.5
0.3
49
41730.2
5592. 3
49
16.9
1.2
49
46577.9
10620.4
* CO, C02 and THC v»lu»a are corrected to 3V. 02.
To obtain actual mca»ured values, divide values in the
table by th« corresponding normalisation -factor.
-------
CEMS DATft - SITE WS - TEST 3
*»
»#
»*
»#
**
**
*»
*#
**
»*
*
*
*
»
*
*
*
»*
**
»*
»*
*•»
**
*#
»*
»*
»*
»* .
*
*
*
*
*
»
*
*
*
**
»*
**
•*
»*
»*
**
**
*•*
«*
»*
**
**
**
»•*
#•
**
FACTOR
FOR 3V. 02
NORMALIZATION
OF
OTHER PROCESS
GASES
-
4.3153
4.4002
4.6182
4.7367
6.4365
5.9966
4.3510
4.5328
4.64S7
4.6530
4.7006
4.4494
4.6B71
4.3753
4.5328
4.4002
4.4307
4.3341
4.2936
4.3510
4.4941
4.3574
4.2457
4.5134
4. 2347
4.2731
4.2731
4.3101
4.1131
4. 0297
"4.0869
• 4. 1244
3.9427
3. 7653
3.7160
3.6060
3. 6479
3.5372
3.4710
3.4238
3.5037
3.3139
3.4636
3. 4489
3.5736
3.6227
3.6643
3.5980
*»
**
**
** ' >
»*
»•*
*#
»•#
»»
#*
**
**
**.
*•»
*»
#*
#»
#»
»*
*•»
*•»
»»
**
•»•»
**
**
»*
*•»
**
**
»*
*••
*•»
**
*»
»*
*«
*•»
**
*«
*«
**
**
»* *
*•»
«*
«*
•»*
*•»
w*
**
**
**
**
NC
-.. TIME - >
1220
1223
1230
1235
1240
1245
1250
1255
1300
1303
1310
1313
1320
1325
1330
1335
1340
1345
1350
1333
1400
1403
1410
1415
1420
1425
1430
1433
1440
1443
1430
1433
1300
1303
1310
1313
1520
1325
1330
1333
1340
1545
1330
1535
1600
1603
1610
1615
JRMALIZE
ACTUAL
02
-------
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing!
EPA-45Q/4-84-014v
3. RECIPIENT'S ACCESSION NO.
National Dioxin Study Tier 4 - Combustion Sources
Final Test Report - Site 13
Residential Wood Stove WS - A
5. REPORT DATE
April 1987
6. PERFORMING ORGANIZATION CODE
Michael W. Hartman, Deborah J. Benson
Lawrence E. Keller
8. PERFORMING ORGANIZATION REPORT NO.
87-222-109-02-25
Radian Corporation
Post Office Box 13000
Research Triangle Park, NC 27709
10. PROGRAM ELEMEN-T-NO.?..,.
1. CONTRACT/GRANT NO.
68-03-3148
AGENCY NAME AND ADORES
U.S. Environmental Protection Agency, OAQPS
Research Triangle Park, NC 27711
Office of Research and Development
Washington, DC 20460
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA Project Officers: Donald Oberacker, ORD
^ William B. Kuykendal, OAQPS
This report summarizes the results of a dioxin/furan emissions test of a residential
woodstove conducted by Radian Corporation. The stove is a freestanding noncatalytic
model manufactured by Atlanta Stove Works and offered for sale in the Sears Catalog
(#42G84156N). During testing oak and pine were burned at low burn rates, which is
representative of normal residential use. The test was the 13th in a series of dioxin/
furan emissions tests conducted under Tier 4 of the National Dioxin Study. The primary
objective is to determine if various combustion sources are sources of dioxin and/or
furan emissions. If any of the combustion sources are found to emit dioxin or furan,
the secondary objective of Tier 4 is to quantify these emissions.
Residential woodstoves are among 8 combustion source categories that have been
tested in the Tier 4 program. The tested woodstove, hereafter referred to as Woodstove
WS-A, is a test unit located at an EPA contractor facility. This stove was selected for
inclusion in the Tier 4 program due to its location in the RTF area and because simul-
taneous testing of the stove was already being conducted for another EPA program (Inte-
grated Air Cancer Project). The woodstove tested is considered representative of wood-
stoves built in the last 5 to 10 years.
No dioxin/furan stack data were obtained from this source. Labelled internal
standards used to determine analytical recovery efficiency were not detected because of
the very high levels of hydrocarbons present in the stack gas. Dioxin/furan ash data
_a T*O T>T*oeoTif'.a^ _ i^*^™>«—^^^—^^^••^^^•^™
presented.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS
COSATl Field/Group
Air Emissions
Combustion Sources
Dioxin
Furans
2,3,7,8 Tetrachlorodibenzo-p-dioxin
Woodstoves
Air Pollution Emissions
Data
8. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (This Report)
Unclassified
20. SECURITY CLASS {Thispage/
21. NO. OF PAGES
74
22. PRICE
EPA Form 2220-1 (R«v. 4-77) PREVIOUS EDITION is OBSOLETE
-------
------- |