Measurement of Gaseous Emission Rates from
Land Surfaces Using an Emission
Isolation Flux Chamber. User's Guide
Radian Corp., Austin, TX
Prepared for
Environmental Monitoring Systems Lab.
Las Vegas, Jiv
Peb 86
-------
EPA/600/8-86/008
February 1986
MEASUREMENT OF GASEOUS EMISSION
RATES FROM LAND SURFACES
USING AN EMISSION
ISOLATION FLUX CHAMBER
USER'S GUIDE
by
M. R. Klenbusch
Radian Corporation
P.O. Box 9948
Austin, Texas 78766
EPA Contract No. 68-02-3889
Work Assignment 18
Project Officer: Shelly J. Williamson
Exposure Assessment Division
Environmental Monitoring Systems Laboratory
Las Vegas, Nevada 89114
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS, NV 89114
-------
- TECHNICAL RETORT DATA ^
(Htnt nti InttnttHam on Mr rrrcne trfort compltlittl
REPORT no.
EPA/600/8-86/008
5. RECIPIENT? ACCESSION NO.
P9 n o o -1 /?
TITLE AND SUBTITLE
MEASUREMENT OF GASEOUS EMISSION RATES FROM LAND
SURFACES USING AS EMISSION ISOLATION FLUX CHAMBER
USER'S GUIDE
S. REPORT DATE
February 1986
t. PtHf OftUINO ORGANIZATION COOC
AUTHOR(S)
M. R. Klenbusch
a. PERFORMING ORGANIZATION REPORT NO,
1O. PROGRAM ELEMENT NO.
ABSD1A
. PERFORMING ORGANIZATION NAME AND ADDRESS
Radian Corporation
P.O. Box 9948
Austin, TX 78766
n. CONTRACT/GRANT NO.
Contract, Radian Corp.
Ho. 69-02-3889
t2. SPONSORING AOENCY NAME ANO ADDRESS
Environmental Monitoring Systems Laboratory - LV, NV
Office of Research and Development
U.S. Environmental Protection Agency
Las Vegas, NV 89114
1*. TYPE OF REPORT AND PERIOD COVERED
Uaer'g Gu-Mo .
14. SPONSORING AGCMCY COOE
EPA/600/07
15. SUPPLEMENTARY NOTES
6. ABSTRACT
A promising method for monitoring ground emissions involves the use of an
emission isolation flux chamber. The method is simple, easily available, and
inexpensive. Applications would include RCRA and CERCLA facilities. To date,
a uniform method operations does not exist. For this reason, an operations
guide has been developed. This guide presents literature surveys, operation
protocols, a case study, and references for further reading. The use of this
protocol will aid in unifying flux chamber measurements and increase data
comparability.
17.
KEY WORDS ANO DOCUMENT ANALYSIS
DESCRIPTORS
b,IOENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Croap
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
1>. SECURITY CLASS (T
UNCLASSIFIED
» NO* Vr PAQ6S
60
JO. SECUBITYCLASS (THt M*J
UNCLASSIFIED
12. PRIC
EPA F*ml220-t (••». 4-77) PKBVIOU* KOITIOM i» OMOLKTC
-------
mKTRUCTIOItt
I. DEPORT NUMBER
Imrrt the EPA report number at it appears on the cam of At pubtkttk*.
2. LEAVE BLANK
J. RECIPIENTS ACCESSION NUMBER
Reserved fot usf by «cfi report recipient.
4, TITLE AND SUBTITLE
Till* should indicate dearly and briefly tht subject coverage of the report, and be disnlarrd prmuanrtr. Srt wblftle. If uard. i* mutter
type ot otherwise subordinate it to mam title. When a report a prepared in more ihM OIK vohntt. repeat UK primary litk. add volume
number and mctude subtitle for the specific title r- » -«
S. REPORT DATE
Each report shall carry a date indicating at kail month and year. Indicate the hob on which it wa* ithcttd (r*., dttr ofiatr dot of
tpprottl. date o{ fttfantton. tie.I. ^ ' ' '
(. PERFORMING ORGANIZATION CODE
Lean blank.
r. AUTHORS)
Give lumen) m ronvraiioaal otde» (John R. Dot. J. Robtn Dot. rle.J. Ul auUioc'i ifTUiiikM (f H dMTen from (be perfo«miB» otfUt-
t. PERFORMING ORGANIZATION REPORT NUMBER
Inxn if perfortninf otpnizatioa viitici to aitifn thu numbct.
«. PERFORMING ORGANIZATION NAME AND AOOREGS
oive name, (tnet, city, itate. and ZIP code, list no more thin (wo levrb of an organizational (Ureaitky.
tO. PROGRAM ELEMENT NUMBER
UK UK ptojrwn element number under which the report was prepared. Subordinate number* any be MkHudcU in parenlbcict.
11. CONTRACT/GRANT NUMBER
InieTt contnci or grant number under which report was prepared.
12. SPONSORING AGENCY NAME AND ADDRESS
Include IIP code,
13. TYPE OF REPORT AND PERIOD COVERED
Indicate interim final, etc., and if applicable, dates covered.
14. SPONSORING AGENCY CODE
Insert appropriate code.
IS. SUPPLEMENTARY NOTES
Enwr information not included elsewhere but usetwl, such as: Prepared in cooperation with. Translation of, Ptcvnlcd at COOIVU-IHX «>r.
To be published in. Supersedes. Supplement!, etc.
1». ABSTRACT
Include a bntf (200 »crdi or lea/ factual wmmary of the mmt sipiiTKant infonaalion contained in the report. If the report cunmim a
Ufni/icjm bibjkijnphy o> literature survey, mention it be»e.
17. KEY «VOROS AND DOCUMENT ANALYSIS
(a) DESCRIPTORS - Select from the Theuunii of Enguieermg and Sdenlific Term (he proper a»lhoened tcrm» that identify the major
concept of the research and are sufficiently specific and precise to be tued as index entries for cataloging.
(b) IDENTIFIERS AND OPEN-ENDED TERMS • Use identifiers for project IUIMS, cede names, equipment dctifjulots, etc. UK open-
ended tttros written in descriptor form.foi ihoie subjects for which BO descriptor exist*.
(c) COSAT1 HLLD GROUT - Field and group assignment! are to be taken from the 196$ COSATI Subject Category Usl. Since the ma-
jority of documents are moltidflaplinary in natunt, the Primary FieM/Group asugnmetilfi) iriB 6e wdTic disoplmc. are* of hum™
endearor, or type of pbyskal object. The appUc*Uon
-------
NOTICE
The information in this document has been funded wholly or in part
by the United States Environmental Protection Agency under Contract
No. 68-02-3889 to Radian Corporation. It has been subject to the Agency's
peer and administrative review, and it has been approved for publication
as an EPA document.
11
-------
OF CONTENTS
faction. Page
1 INTRODUCTION ........................ 1-1
2 BACKGROUND ......................... 2-1
2.1 Emission Processes ................... 2-1
2.2 Measurement Techniques ......... ........ 2-2
2.3 Flux Chamber Operation. ... ........ ..... 2-3
3 MEASUREMENT OF GASEOUS EMISSION RATES FROM LAND SURFACES
USING AN EMISSION ISOLATION FLUX CHAMBER - PROPOSED METHOD . 3-1
3.1 Applicability and Principle .............. 3-1
3.1.1 Applicability .................. 3-1
3.1.2 Principle .................... 3-1
3.2 Precision, Accuracy, Sensitivity, and Range ...... 3-1
3,2.1 Precision .................... 3-1
3.2.2 Accuracy .................... 3-2
3.2.3 Sensitivity ................... 3-2
3.2.4 Range. ....... ..... ......... 3-2
3.3 Interferences ..................... 3-2
3.3.1 Flux Chamber Method ............... 3-2
3.3.2 Emission Process ................ 3-4
3.4 Apparatus and Materials ................ 3-4
3.4.1 Flux Chamber and Supporting Equipment ...... 3-4
3.4.2 Discrete Sample Collection ........... 3-6
3.4.3 Analysis , ................... 3-6
3.5 Procedure ....................... 3-11
3.5.1 Flux Chamber Operation .... ..... .... 3-11
3.5.2 Sample Collection. ................ 3-11
3.5.3 Sample Analysis ..... ....... ..... 3-13
3.5.4 Sampl Ing Strategy ................ 3-15
3.6 Calibration ...................... 3-18
3.6.1 Equipment .................... 3-18
3.6.2 Analyzers. . .................. 3-19
3.7 Qua I Ity Control .................... 3-20
3.7.1 Sampling Equipment ............... 3-20
3.7.2 Sampl ing .................... 3-22
3.7.3 Analytical ................... 3-22
3.8 Calculations ...................... 3-23
3.8.1 Definitions ................... 3-23
3.8.2 Percent Recovery ........ . ....... 3-25
3.8.3 Calculation of the Dilution Factor Involved
In Gas Canister Analysis ............ 3-27
3.8.4 Area Source Emission Rate Equations ....... 3-27
4 CASE STUDY ......................... 4-1
5 ADDITIONAL INFORMATION ................... 5-1
REFERENCES ......................... R-1
APPENDIX A - SELECTION OF A RANDOM SAMPLE. . . ....... A-t
-------
SECTION 1
INTRODUCTION
Volatilization of organic compounds from contaminated soil or ground-
water Into the air represents a major potential source of exposure which has
not been assessed. In order to assess this exposure potential, a method Is
needed to directly measure gas emission rates frcm contaminated soils and/or
groundwater. Additionally, It Is recognized that an understanding of the
volatilization, transport, and emission processes could lead to a predictive
tool for exposure assessment. The Information provided by direct mea-
surement *nd/or predictive modeling will allow state and local regulatory
agencies to develop programs to assess and define the need to control gas
emissions from area sources contaminated by organic compounds.
The purpose of this User's Guide Is to present an approach and proto-
col , namely the emission Isolation flux chamber (or flux chamber) technique,
for measuring emission rates of volatile organic compounds from contaminated
soils and/or groundwater. Presented Is the theory of operation, specifica-
tions, sensitivities, method of operation, and data reduction procedures for
this technique. It Is assumed that the Individuals who will use the proto-
col are. In general, familiar with sample collection and analysis of vola-
tile organic compounds. Also Included In this document Is a case study that
demonstrates the measurement and data reduction processes around a spill
site.
The flux chamber technique Is applicable to the measurement of emission
rates frcm Resource Conservation and Recovery Act (RCRA) facilities (hazard-
ous waste land-treatment, and landfill facilities), and from Comprehensive
Environmental Response, Compensation, and Liability Act (CERCLA) area
sources contaminated by losses of volatile organic compounds from spills,
from leaking underground storage tanks, frcm pipelines, and/or from surface
Impoundments.
This protocol does itot present the vast amount of work that was required
to develop this document. Rather, the protocol Is a result of IIterature
reviews selecting a measurement technique and field applications demon-
strating the technique and developing a data base and validation studies
Identifying the method of flux chamber operation. References to the other
area sources where this technique was applied, the work performed to vali-
date the technique, and the Investigations of variables which control the
emission process are also given for those Individuals desiring further
Information.
1-1
-------
SECTION 2
BACKGROUND
The following subsections discuss the process by which volatile organic
compounds are emitted froa contaminated land surfaces, the basis upon which
the flux chamber technique was selected as an approach for measuring such
emission rates, and the principle of the technique.
2.1 Emission Processes
The rate of volatile organic compound (VOC) emissions from contemlnated
soils Is generally believed to be controlled by the diffusion rate of the
chemical compound through the air-filled pore spaces of the soil.(1,2,3)
The exception occurs when the contaminated material Iles on or very near the
sotl surface. Such Is the case when spills occur or Immediately after waste
Is surface-applied to a landtreatment site. In these cases, the emission
process will be controlled by the rate of evaporation.
Evaporation Is a surface phenomenon, and trte parameters that affect the
evaporation process are the properties of the waste Itselt as well as those
that have an effect on the air-surface Interface (I.e. wind, surface rough-
ness). The Important parameters Include the volatility or vapor pressure of
the waste, ambient meteorological conditions (solar Insolation, air and
waste temperature, surface wind speed, relative humidity), surface coarse-'
ness, and the bulk concentration of the volatile components In the air
(although this Is usually very low and generally assumed to be negligible).
There are two major types of soil emission processes. Each are treat-
ment dependent. One type occurs In Iandtreatment facilities and the other
at underground facilities such as landfills. In land-treatment applications,
the emission rate Is generally highly time-dependent, When a fixed amount
of waste Is applied to the soil surface. It penetrates the soil to a certain
depth. The vaporization rate Is maximum Immediately after waste applica-
tion, as the material nearest the surface Is vaporized and diffuses through
a very thin layer of soil. As the waste near the surface Is depleted of Its
VOC content, the volatile material deeper In the soil aust diffuse through
an Increasingly thick soil layer. The soil presents a resistance to VOC
diffusion In direct proportion to the VOC depth. Thus, the rate of emis-
sions from the surface decreases with time.
It Is common practice In Iandtreatnent to periodically till the sofl to
provide oxygen for bacterial activity. The tilling effectively mixes the
remaining waste In a homogeneous layer near the soil surface. The emission
rate Is at a maximum Immediately following each titling episode since
2-1
-------
volatile waste Is again present very near the surface, and resistance to
diffusion Is at a minimum.
Although also diffusion controlled, the emission process from under-
ground sources such as landf11 led waste or material present as a "lens" on
the water table has significantly different characteristics than that from
surface or near-surface sources. The depth of the emission source Is
usually quite substantial. Therefore, the emission rate Is Initially lover
due to the resistance to diffusion produced by the coloumn of soil. The
Initial emission rate Is zero, since It takes some time for the volatile
material to diffuse through the soil layer. The adsorptlve sites on the
soil particles must also be Initially saturated. Once the emission rate has
equilibrated, the rate Is relatively constant with time until the under-
ground source Is exhausted.
The diffusion process Itself through the soil Is the same for both
types of sources, landtreatment (surface) and landfill (underground). Con-
sequently, many of the parameters Important to the emission processes are
the same, Including dtffuslvlty of the VOC In air, soil properties (particle
size distribution, soil type, moisture content, particle density, porosity),
soil/waste temperature, and volatility of the VX In the waste. Additional
parametars Important to the near surface emission processes are the amount
of material present In the contaminated soil layer, the Initial depth of the
contamination, the elapsed ttme from application (or tilling) and, possibly,
ambient conditions such as surface wind speed and relative humidity. The
depth of the soil layer above the waste is a very Important parameter In the
em sslon process from subsurface sources. Additionally, the adsorptlve
properties of the soil may also have a significant effect on the emission
rate from this latter source type.
An understanding of the emission processes and the Important parameters
Is necessary In the measurement of emission rates from soil surfaces and In
the proper Interpretation of the test results. As an example, the emission
rate from a source Is affected by rain since the porosity and, hence, the
diffusion rate are reduced with Increasing moisture content of the soil.
Thus, emission rates Immediately after a rainfall will be lower than those
from drier soils and may take substantial periods of tine to return to the
emission rate prior to the rain,(4) Emission rates may vary with the time
of day and season, as a result of changes In ambient and sol I/waste tempera-
tures. (4) Emission rates from soil areas containing fissures can be higher
and much less homogeneous than those from unfractured areas. Thus, consi-
derable care must be taken In planning and implementing a measurement pro-
gram to determine representative emission rates from such soli surfaces.
2.2 Measurement Techniques
Based on a literature review (5>, the techniques for determining gas
emissions rates from land surfaces contaminated with organic compounds can
be divided Into three approaches: Indirect measurements, direct measure-
ments, and laboratory simulations, indirect techniques typically require
measurements of ambient air concentrations at or near the site. These
2-2
-------
measurements are related to the surface area of the area source and local
meteorological conditions using a dispersion model to determine an emission
rate. The second approach is to directly measure emission rates using for
example the flux chamber. The third approach Is to create an emission
source In the laboratory and model the emissions by various techniques for
application to field sites. These three approaches were compared for preci-
sion, accuracy, and sensitivity. Other considerations Included applicabil-
ity, complexity, manpower requirements, and costs.
The most promising technique for measuring gas emission rates from land
surfaces was determined to be the emission isolation flux chamber technique.
The advantages are:
o lowest (most sensitive) detection limit of the methods
examined;
o easily obtained accuracy and precision data;
o simple and economical equipment relative to other
techniques;
o minimal manpower and time requirements;
o rapid and simple data reduction; and
o applicable to a wide variety of surfaces.
2.3 Flux Chamber Operation
The flux chamber technique has been used by researchers to measure
emission fluxes of sulfur, nitrogen, and volatile organic species
(6,7,8,9,10). The approach uses a flux chamber (enclosure device) to sample
gaseous emissions from a defined surface area. Clean dry sweep air Is added
to the chamber at a fixed, control led rate. The volumetric flow rate of
sweep air through the chamber Is recorded and the concentration of the
species of interest Is measured at the exit of the chamber. The emission
rate is calculated as:
E, - YtQ/A (2-1)
where: Ej • emission rate of component I (mass/area-time),
Y| » concentration of component I In the air flowing fron the chamber
(mass/volume),
0. « flow rate of air Into the chamber (volume/time),
A » surface area enclosed fay the chamber (area).
All parameters In Equation 2-1 are measured directly.
Most of the emission rate assessments are of area sources much larger
than the enclosed surface area of the flux chamber (0.130 m2). In these
2-3
-------
cases, an overall emission rate for the area source Is calculated from
•ultlple measurements based on random sampling and statistical analysts.
2-4
-------
SECTION 3
MEASUREMENT OF GASEOUS EMISSION RATES FROM LAND SURFACES
USING AN EMISSION ISOLATION aUX CHAMBER - PROPOSED METHOD
3.1 Applicability and Principle
3.1.1 Applicability
The flux chamber technlquo Is applicable to the measurement of emission
rates from Resource Conservation and Recovery Act (RCRA) facilities such as
hazardous waste Iandtreatment and landfill facilities. This technique Is
also applicable for emission rate measurements from Comprehensive Environ-
mental Response, Compensation, and Liability Act CCERCLA) waste sites such
as areas contaminated by losses of volatile organic compounds from spills,
from leaking underground storage tanks, frctn pipelines, and/or from surface
Impoundments.
3.1.2 Principle
Gaseous emissions are collected from an Isolated surface area with an
enclosure device called an emission Isolation flux chanter (or flux cham-
ber). The gaseous emissions are swept through an exit port where the con-
centration Is monitored and/or sampled. The concentration Is monitored
and/or sampled either continuously (I.e., "real-time") or discretely. Real-
time measurements are typically made with portable total hydrocarbon ana-
lyzers and are useful for relative measurements (I.e.? the determination of
flux chamber steady-state operation, zoning). Discrete samples are taken
when absolute measurements are necessary (I.e., steady-state concentrations,
emission rate levels). Tha emission rate Is calculated based upon the sur-
face area Isolated, the sweep air flow rate, and the gaseous concentration
measured. An estimated average emission rate for the area source Is calcu-
lated based upon statistical sampling of a defined total area.
3.2 Precision, Accuracy, Sensitivity, and Range
3.2.1 Precision
Single Camber precision (I.e., repeatabllIty) of the method Is approx-
imately 5 percent at measured emission rates of 3,200 ug/mln-m2. Variabil-
ity between different flux chambers (I.e., reproduclbtllty) Is, approximately
9.5 percent within a measured emission rate range of 39,000 to
65,000 ug/mln«m2.(4>
3-1
-------
The roproduclb!IIty results were determined from a bench-scale study.
The tests were designed to eliminate temporal variations from the flux
chamber reproduclbllIty. However, using the same bench-scale facility, a
test design was not possible for measuring flux chamber repeatability with-
out bias from temporal variations. As a result, the repeatability tests
were performed In the laboratory. The differences therefore between the
stated emission rates for repeatability and reproduclbllfty reflect the
differences In laboratory simulated emission rates and those meausred from
the bench-scale facility.
3.2.2 Accuracy
Flux chamber recovery (Section 3.6.1.4.2) results show a recovery range
of 77 percent to 124 percent. Table 3-1 lists measured recoveries for a
number of compounds tested. The average recovery for the 40 compounds
tested Is 103 percent.
Flux chamber emission rate measurements made on the soil cells range
from 50 percent to 100 percent of the predicted emission rates. That Is,
the measured emission rates can be expected to be within a factor of one-
half of the "true" emission rates.(4) The flux chamber accuracy based upon
both the recovery tests and predictive modeling ranges from 50 percent to
124 percent.
3.2.3 Sensitivity
The sensitivity of this method depends on the detection limit of the
analytical technique used. When discrete samples are collected using gas
canisters and analyzed by gas chromatographIc methods, the estimated emis-
sion rate sensitivity Is 1.2 ug/mln-m^ for an analytical detection limit of
10 ppbv benzene. When emission rates are measured In a continuous {real-
time) method, the estimated sensitivity Is 124 ug/mln-m^ for an analytical
detection Unit of 1 ppmv benzene.
3.2.4 Range
The range of this method depends upon the analytical technique used.
High level emission rates are analyzed by Introducing proportional amounts
of gas sample to the analyzer. Using this technique, high level emission
rates of 120,000 ug/mln-m* have been measured.(4) Low levels are limited by
the sensitivity of the analytical technique. Gas chroraatographIc techniques
have been used to measure low level emission rates of 1.2 ug/mln-m2 for mea-
sured concentrations of 10 ppbv benzene.
3.3 Interferences
3.3.1 Flux Chamber Method
Impurities In the sweep atr and/or organic compounds outgasslng from
the transfer tines and acrylic chamber top may causa background contamina-
tion. The emission Isolation flux chamber must be demonstrated to be free
3-2
-------
TABLE 3-1
COMPOUNDS TESTED IN THE EMISSION ISOLATION FLUX CHAMBER
AND THE MEASURED PERCENT RECOVERY
Compound
Percent
Recovery*
Compound
Percent
Recovery*
Total C2
Total C3
Isobutane
1-toutene
n-butane
t-2-butene
c-2-butene
Isopentane
1-pentene
2-«ethyI-1-butene
n-pentane
n-pentene
c-2-pentene
Cyclopentene
n-hexane
Isohexane
3-methyIpentane
MethyIcycIopentane
Benzene
1,2-DImethyIpentane
100 3-«ethyIhexane 106
108 2,2,4-trlmethyl pentane 106
109 n-heptane 103
108 Methyfcyciohexane 103
106 Toluene 103
107 Ethyl benzene 94.7
109 ntp-xylene 88.?
112 o-xylene 97.3
105 n-nonane 99.4
124 n-propyI benzene 95.5
107 p-ethy I toluene 92.5
103 1,3,5-trlroethylbenzene 93.5
105 1,2,4-trJmethylbenzene 88.7
105 2-«iethyl-2-butene 103
95.1 Methyl mercaptan 107
107 Ethyl nercaptan 107
106 Butyl nercaptan 101
105 Tetrahydrothlophene 115
106 Trlchloroethylene 77.1
105 Ethylene
-------
from significant (
-------
FIGURE 3-1
A CUT/WAY DIAGRAM OF THE EMISSION ISOLATION FLUX CHAMBER AND
SUPPORT EQUIPMENT
TEMPERATURE
READOUT
THERMOCOUPLE
\J»
SYRINGE/CANISTER
V SAMPLING PORT
REAL TIME
ANALYZER
CARRIER
GAS
OUTLET LINE
STAINLESS STEEL
OR PLEXIGLAS
CUT AWAY TO SHOW
SWEEP AIR INLET LINE
AND THE OUTLET LINE
-------
Table 5-2. A construct ton diagram of the flux chamber Is shown In
Figure 5-2.
The sweep air carrier gas should be dry, organic free air equal to or
better than commercial ultra high purity grade «0.1 ppmv THC). A gas flow
meter with no Internal rubber parts and adjustable wfthln the range of 1-10
L/mln should be used to control gas flow. Temperature measurements should
be made with an accuracy of ±1.0°C. A fine-wire thermocouple with elec-
tronic readout Is recommended. Caution should be taken to avoid any contact
of a thermocouple with metal. This would give Inaccurate air temperature
readings. A pressure release port Is required to avoid pressure bulid-up
Inside the flux chamber during operation. This port should never be
blocked. For system blanks, a clean Teflon" sheet should bo used to provide
a clean surface for the flux chamber.
5.4.2 Discrete Sample Collection
Discrete grab samples should be collected with air-tight. Inert con-
tainers. For on-slte analysis, 100 ml precision lock, glass syringes are
recommended. Glass plungers are recommended over Teflon" tip plungers. If
Teflon" tip plungers are used, then special controls must be followed to
avoid cross-contamination (Section 3.7.1.1). For samples to be transported
or to be stored for periods longer than 1 hour, 2L stainless steel gas
canisters are recommended.
5.4.5 Analysis
5.4.5.1 Real Time
Analyzer
For real-time, continuous monitoring of the exit gas concentration,
analyzers with precision of ±10 percent of the measured value and a detec-
tion I Imlt of 1 ppmv are recommended.
Calibration Gases
The portable, real-time analyzers will require the following levels of
calibration gases:
o High-level Gas: Concentration within 50 percent to 90 per-
cent of the span value (maximum expected concentration or
upper limit of Instrument linear range).
o Low-Level Gas: Concentration less than or equal to O.Ot
percent of the span value.
o Zero Grade Gas: Ultra high purity (UHP) air «0.1 ppmv THC).
The calibration gas for these jnalyers-can be the same as that used for
the on-slte discrete analyzer (Section 5.4.5.2.2).
3-6
-------
TABLE 3-2
CHAMBER MATERIALS SPECIFICATIONS
Item
Description
Specification
Carrier Gas Lines:
Inlet/Outlet Teflon" (clear)
Sweep Afr Wrap
Perforation8
FIttlngsb
Thermocouples
Air (!)
Flux Chamber:
Base
Support ring
flange
Dome
Seal
Dome to Base
Statnle-js Steel
four equidistant holes
jetting direction
Stainless steel
Stainless steel
Fine wire
K type
Stainless steel
column
Stainless steel
Acrylic
four holes
Inlet/outlet
Air temperature
Pressure release
Top gasket
Dome IIp
1/4" 00, 5« to 8' long, thin
walled, 1/4" stainless steel
fittings
1/4" OD, 54" long, perforated
hole No. 1 (nearest Input), 5/64"
ID, holes No. 2-4, 3/32" ID,
axially, horizontally
1/4" bulkheads with teflon
washers for chamber penetration
1/4" cap to seal wrap I Ine end
36" long, bead tfp, teflon coated
(extensions optional), penetrate
flux chamber 3", support with
1/4" bulkhead with septa
16" ID x 7" tall, welded to a
support ring flange
16" ID x 20" 00 x 1/4" thick
Spherical, 4" displacement at
center, 16" 10 at seal, 2" lip
for seal, 1/4* thick, molded
Equidistant, 4* from aluminum
gasket
1/2" ID with 1/4" stainless steel
bulkhead
1/2" ID with 1/4" stainless steel
bulkhead
13/16" ID with 3/4" stainless
steel bulkhead
Aluminum 16" ID, 20" OD, 1/4"
thick
Below alumlnu* gasket Is the
acrylic lip of dome
(Continued)
3-7
-------
TABLE 3-2
(Continued)
Item
Description
Specification
Seal Ing washer
Botton gasket
Fasteners
Volume
Surface Area
Exit Line Probe Teflon
Perforation
With 1" soil
penetration
Enclosed by chamber
2 rows of holes
Teflon, 16" 10, 20" 00, 1/32"
thick
Stainless steel support ring
20, 1/4" bolts equidistant around
Up ,
0.03 m3 (30L)
0.130 m2
1/4" OD, 6" long, stainless
steel fitting, perforated
3/32" ID, 5 holes per row, 1"
separation, rows are positioned
orthogonally
"Avoid placement of exit line probe In Jetting path of sweep air Inlet holes
bAII fittings are manufactured by Swagelok* or equivalent manufacturer
(bulkheads use Teflon" washers for sealing)
5-e
-------
FIGURE 3-2
EXPLODED VIEW OF THE FlUX CHAMBER
THERMOCOUPLE
INLET CARRIER
GAS LINE
SEPTA
V." BULKHEAD
& TEFLON WASHER
FLUX CHAMBER
DOME
DOME UP
SEALING
WASHER
1 OF 20
BOLTS
TOP GASKET
OUTLET CARRIER
GAS UNE
BULKHEAD
4 TEFLON WASHER
BULKHEAD
& TEFLON WASHER
(PRESSURE RELEASE)
20 HOLES.
EVENLY SPACED
(SEE TOP GASKET)
PERFORATED
SWEEP AIR WRAP
INLET HOLES
EXIT UNE PflCJBE
PERFORATED
20 HOLES,
EVENLY SPACED
(SEE «DP GASKET)
SUPPORT
RING FLANGE
FLUX CHAMBER
BASE
3-9
-------
Quality Control (QC) Gas
The portable, real-time analyzer will require a quality control «3C)
gas concentrated to fall within the span range. The QC gas for this analy-
zer can be the same as that used for the on-slte discrete analyzer,
3.4.3.2 Discrete
Analyzer
The analyzer should be sensitive with low detection limits. For on-
slte analysis of grab samples, Instrumentation having precision of ±5 per-
cent of the measured value with a detection limit of 1 ppm Is recommended.
Analyzers with Injection loops are recommended to reproduce the sample
volumes Injected. For off-site analysis, Instrumentation with precision of
±30 percent at detection IImlts of J ppbv are recommended.
CalIbratIon Gases
The concentrations and composition of the calibration gases to be used
will vary depending on the species of Interest. Preferably, the following
gas concentrations should be used for each species of Interest:
o High-level Gas: 90 percent of the span vaJue.
o Mid-Level Gas: Average expected concentration.
o Low-Level Gas: 0.01 percent of the span value.
o Zero Grade Gas: Ultra high purity (UHP) air, (
-------
3.5 Procedure
3.5.1 Flux Chamber Operation
The flux chamber Is operated Identically for real-tine and discrete
sampI Ing.
3.5.1.1 Preparation
All exposed chamber surfaces should be cleaned with water and wiped dry
prior to use. Assemble the samp I Ing apparatus and check for malfunctions
and leaks.
3.5.1.2 Operation
Place the flux chamber over the surface area to be sampled and work It
Into the surface to a depth of 2-3 on. Initiate the sweep air and set the
flow rate at 5 L/mln. Record data at time Intervals defined by residence
times or T (tau), where U - flux chamber volume (30L)/sweep air flow rate
(5L/mln), One T then has the value of 6 minutes under normal operating
conditions. At T » 0 (flux chamber placement), record the following: time,
sweep air rate, chamber Inside air temperature, ambient air temperature, and
exit gas concentration (real-time analyzer). The data should be recorded on
the data sheet shown In Figure 3-3. At each residence time (T, 6 minutes),
the sweep air rate shall be checked (and corrected to 5 L/mln If necossary),
and the gas concentration shall be recorded (real-time analyzer). After 4
residence times (24 minutes). Initiate sample collection. At this time,
record the following data: time, sweep air rate, air temperatures Inside and
outside, exit gas concentration, and sample number(s). If sulfonated
organic compounds are of specific Interest, then meesurenents should be
taken after 10 residence times (1 hour).
3.5.2 Sample Collection
3.5.2.1 Real Time
When real-time monitoring Is required, the sample Is collected by the
real-time analyzer directly from the exit gas line.
3.5.2.2 Discrete Sample Collection
Sample collection should not exceed a flow rate of 2 L/mln.
Gas Syringes
Sample collection with syringes should be performed after purging the
syringe three times with the sample gas. This should be performed without
removing the syringe from the sampling line manifold. To ensure fresh
sample at each purge, a sampling manifold should be positioned prior to a
real-time analyzer (Figure 3-1). The analyzer will then draw the sample
past the manifold for sampling.
3-11
-------
FIGURE 3-3
FLUX CHAMBER GAS EMISSION MEASUREMENTS FIELD DATA SHEET
FLUX CHAMBER EMISSIONS MEASUREMENT DATA
Date.
.Samplers).
Location.
.Zone/Grid Point
Surface Description
Concurrent Activity..
Time
Sweep Air
Rate,Q
(UMIn)
Residence
No.
(Q/V)
0
1
2
3
4
5
Gas
Cone.
(ppmv)
Air Temperature
Chamber Ambient
(C) (C)
Sample
Type/No.
Comments:
Comments:.
74*24843
3-12
-------
Gas Canister
Sample collection with evacuated gas canisters should be performed with
the real-time analyzer replaced by the gas canister (Figure 3-4).
To collect canister samples, remove the real-time analyzer froa the
exit line sampI Ing manifold. Securely fasten the canister sampling manifold
to the exit line manifold. Open the flow control valve
-------
FIGURE 3-4
STAINLESS STEa GAS CANISTER AND SAMPLING MANIFOLD
(NOT TO SCALE)
CAP, SAMPLE ELUTION/
PRESSURE MEASUREMENT
1/4' S.S. SWAGELOCK*-
EXIT LINE
FROM FLUX i
CHAMBER
VALVE
EXIT LINE
SAMPLING
MANIFOLD
VALVE
(V)
2L CANISTER
-------
gas canister, the sample Is then Introduced Into the gas chronatograph
through cryogenic traps. The dilution factor fs calculated by Equation 3-2
(Section 3.8.3).
3.5.4 Sampling Strategy
The following sampling strategy provides an accurate and precise esti-
mate of the emission rate for a total area source through random sampling In
which any location within the area source has a theoretically equal chance
of being sampled. The sampling strategy described below provides an esti-
mated average emission rate within 20 percent of the true mean with 95
percent confidence.
3.5.4.1 Zones
Based on area source records and/or preliminary survey data, subdivide
the total area source Into zones If nonrandom chemical distribution Is
exhibited or anticipated. The zones should be arranged to maximize the
between-zone variability and minimize the wI thin-zone variability.
3,5.4.2 Grids
Divide each zone by an Imaginary grid with unit areas that depend on
zone area size (Z) as follows:
If Z 1 500 m2, then divide the rone area Into units with areas
equal to 5 percent of the total zone area (I.e., 20 units total).
If 500 m2 < Z £ 4,000 m2, then divide the zone area Into units of
aroa 25 m2.
If 4000 m2 < Z <. 32000 m2, then divide the zone area Into 160
units.
If Z > 32000 w2, than divide the zone area Into units with area
equal to 200 in2.
Assign a series of consecutive numbers to the units In each zone.
3,5.4.3 Sample Number
Using Equation 3-3 (Section 3.8.4), calculate the number of units (grid
points) to be sampled for the Kth zone (n«).
3,5.4.4 Sample Locations
Using the random numbers table (Appendix A), Identify n^ grid points
(units) that will be sampled In zone K. A grid point shall be selected for
measurement only once. (This Is not to be confused with duplicate sampling.
Section 3.7.2.2.)
3-15
-------
3.5.4.5 Emission Rate Calculations
After sample col lection, use Equations 3-4, 3-5, 5-6, 5-7, and 5-8 to
calculate the measured emission rate (Ec«|) for each grid point (I) In each
zone (K). Research has shown an emission rate dependency upon the air
temperature Inside the flux chamber.(4) Through a statistical analysis of
both laboratory and field data, a correction factor for temperature varia-
tions has been developed. The correction factor compensates a measured
emission rate for chamber air temperature variations from the nominal cham-
ber air temperature.
The nominal chamber air temperature can be defined In two ways de-
pending on the purpose of emission rate measurements. If emission rate
measurements are for an estimate of an area source, then the nominal chamber
air temperature should be the mean chamber air temperature of all the mea-
surements made at that area source. If emission rate measurements are
compared between area sources, then the nominal chamber air temperature
should be 25°C (298K).
3.5.4.6 PrelImlnary Estimates
W_Ith Equations 5-9, 3-10, and 3-11, calculate the zone mean emission
rate n^, then N^-IK additional
samples must be collected from zone K. Locate these additional samples
using a random numbers table. Do not duplicate previously sampled loca-
tions.
If N£ » n«. It may be most effective to rezone using the preliminary
measured emission rates as a guide. If new zones are established, then
these new zones will need to be grldded accordingly (Section 3.5.4.2).
3.5.4.8 Final Estimates
Collect any additional samples and recalculate the emission estimates
for the sample mean (EK) and variance (Sj£> lor each zone (Section 3.5.4.6).
Then compute the overall area source mean (£> and variance (S*) for the
total site area using Equations 3-13 and 3-14, respectively. Determine the
95 percent confidence interval for each zone
-------
TABLE 3-3
TOTAL SAMPLE SIZE REQUIRED BASED ON THE PRELIMINARY
SAMPLE COEFFICIENT OF VARIATION ESTIMATE*
Coefficient of Number of Samples
Variation - CV (?)** Required (Ng) per Zone K
0 - 19.1
19.2 - 21.6
21.7 - 24.0
24.1 - 26.0
26.1 - 28.0
28.1 - 29.7
29.8 - 31.5
31.6 - 33.1
33.2 - 34.6^
34.7 - 36.2
36.3 - 37.6
37.7 - 38.9
39.0 - 40.2
40.3 - 41.5
41.6 - 42.8
42.9 - 43.9
44.0 - 45.1
45.2 - 46.2
46.3 - 47.3
47.4 - 48.4
48.5 - 49.5
49.6 - 50.7
50.8 - 51.6
51.7 - 52.3
52.4 - 53.4
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
given Is the sample size required to estimate the average emission
rate with 95 percent confidence that the estimate will be within 20 per-
cent of the true mean.
**Foc CVs greater than 53.4, the sample size required Is greater or equal to
CVV100.
3-17
-------
3.6 Calibration
3.6.1 Equipment
3.6.1.1 Flow Meters
The flow meter should be calibrated against an NBS-traceable bubble
meter before sampling. The flow meter should have a working range of 2-10
L/mln.
3.6.1.2 Thermocouple
Fine wire K-type Insulated thermocouples are recommended for tempera-
ture measurements. Prior to field use, the thermocouple and readout should
be calibrated against a mercury-In-glass thermometer meeting ASTM E-1 No.
63C or 63F specifications. The thermocouple should have an accuracy within
±1°C.
3.6.1.3 Ca11 bratIon Gases
For checking the concentrations of the calibration gases, use calibra-
tion gases that are documented traceable to National Bureau of Standard
Reference Materials. Use Traceablltty Protocol for Establishing True Con-
centrations of Gases Used for Calibrations and Audits of Continuous Source
Emission Monitors (Protocol Number 1) that Is available from the Environ-
mental Monitoring and Support Laboratory, Quality Assurance Branch, Mall
Drop 77, Environmental Protect Ior Agency, Research Triangle Park, North
Carolina 27711. Obtain a certification frcm +he gas manufacturer that the
protocol was followed.
3.6.1.* Flux Chamber System
Several tests should be performed to characterize a new flux chamber
prior to use. These tests should be repeated If a chamber Is exposed to
severe conditions such as corrosive gases, extremely high levels of organic
vapors, or organic I(quids.
Blanks
Check the flux chamber for background by placing the chamber over a
clean Teflon" surface and running a test using ultra high purity sweep air
and routine operating conditions. Sample col lection and analysis should be
as previously described (Sections 3.5.2 and 3.5.3).
Recovery Efficiency
Check the flux chamber sample recovery efficiency by placing the cham-
ber over a flat Teflon" surface containing an Inlet port at the canter for
Introduction of a cat(oration gas(es). The calibration gas should be that
used for the on-slte analyzer at a concentration of at least 1,000 ppnv
(high-level gas). The calibration gas should be Introduced Into the chamber
3-18
-------
at a flow rate of no greater than 0.5 L/mfn. Add ultra high pur/ty weep
air concurrently through the enclosure sweep air Inlet (5 l/mltt) and (Jeter-
nine the concentration exiting the enclosure under routine operating condi-
tions. Compare the measured concentration to the true concentration (cor-
rected for dilution), and calculate a percent recovery using Equation 3-1.
Results for a variety of volatile organic compounds are presented In Table
3-1. Results should be within 10 percent of the true concentration. The
United data characterizing the recovery efficiency for halogenated com-
pounds Indicate an acceptance level tnet may be larger than 10 percent.
Corrective Action
If the background levels of tho flux chamber are greater than 10 per-
cent of the measured concentrations or 10 ppmv, whichever fs smaller, then
rerun the blank sample. If high levels persist, then disassemble the flux
chamber, clean all Internal parts with water and replace those suspected to
be contaminated, and rccrsemble for another blank run. Repeat above until
satisfactory levels are reached.
If the recovery efficiency Is below 90 percent for non-halogenated
compounds, then rervn the recovery test. It low recoveries persist, check
for poor sealing and/or Inlet gas shortcuttlng directly from the Input line
to the exit line and/or mlsadjusted flow rate settings.
3.6.2 Anelyzers
The following procedures should be performed at the recommended fre-
quency during the analysis of flux chamber samples.
3.6.2.1 Real Time
Real-time analyzers are used more for relative, continuous measurements
than for absolute measurements. If these analyzers are Intended for abso-
lute measurements, then they should be calibrated according to Section
3.6.2.2. Real-time analyzers may be used when data qua!Ity requirements are
less stringent (Section 3.1.2). As such, these analyzers require less
stringent quality control practices.
Each day prior to sampling, a three-point calibration should be per-
formed on each analyzer (Section 3.4.3.1.2). Consider the calibration
acceptable If responses are within ±20 percent of the expected response. If
the responses are not acceptable, then recalibrate the Instrument.
3.6.2.2 Discrete Analyzer
Discrete analyzers are those that are the most relied upon for abso-
lute, quantitative data of the analyzers used on site. As su~.h, these
analyzers require more stringent quality control practices (Section 3.1.2).
The calibration procedure suggested here is for linear detectors (I.e., FID,
PID). Compensations for non-linear detectors used for analysis of sulfo-
nated compounds (flame photometric detectors) must bo made.
3-19
-------
Prior to each field Investigation, a multipoint calibration Including
zero and at least three upscale concentrations CSubsectlon 3.4.3.2.2) should
be performed to establish the linearity of the analyzer. The results may be
used to prepare a calibration curve for each compound. Alternatively, If
the ratio of GC response to amount Injected (response factor) Is a constant
over tha multipoint range «10 percent coefficient of variation, standard
deviation/mean), linearity through the origin can be assumed, and the aver-
age response factor can be used In place of a calibration curve.
Each day prior to sampling and after every fifth sample, the working
calibration curve (or response factor) must be verified by the measurement
of one or more calibration standards. If the response for any standard
varies from the predicted response by 20 percent, the test must be repeated
using a fresh calibration standard. If the analyzer response Is still
unacceptable, a new calibration curve (or response factor) must be prepared
for tha" compound. A new calibration curve (or response factor) should be
calculated after each verification of calibration using the acceptable
result* of the one or more calibration standards Injected.
3.7 V allty Control
3.7.1 Sampling Equipment
3.7.1.1 Syringes
Prior to use for sample collection, all syringes should be challenged
with one or more of the calibration standards. An acceptable response Is
within ±10 percent of the predicted response. If the response Is unaccept-
able, then repeat the test. Alternatively, check for leakage around the
plunger or lock valve by pressurizing the syringe and submerging It under
water. Syringes should be checked after every 25 to 30 uses or whenever
leakage Is suspected. If Teflon" tip plungers are used, then suspect memory
effects after exposure to high levels of organIcs. In Instances when memory
effects are apparent, the Teflon" tips should be replaced.
3.7.1.2 Gas Canisters
Gas canisters should be cleaned and evacuated before each use. The
pressure should be recorded after each evacuation. Prior to sample collec-
tion, check the pressure and compare It to that recorded after cleaning.
Acceptable differences are <10 percent of the post evacuation pressure.
Canisters having unacceptable pressure differences should not be used for
sample collection.
To Identify gas canisters and record pressure values, each gas canister
should have a chaIn-of-custody form (Figure 3-5). Copies of this font
should be retained for the sampler, laboratory, and sample control.
-------
FIGURE 3-5
CHAIN-OF-CUSTODY FORM FOR GAS CANISTER SAMPLES
STAINLESS STEB. CANISTER
CHAIN OF CUSTODY
TtNEt
- -T08E COMPLETED BY HBJ) SAMPLE*
SAMPLE CONTROL NUMBER)
CANISTER KUNBERi
DATE SANPLEDi
Wat/STATION NUMBER]
OVA READING (PEAK)i
AOORCSS/RCF i tern LOOT roNt
(CJCKT/OEPTH/ROOMi
SAKPUR'S INlTIALSt
TASK!
TYPE (C(ROEOIC)i
COMCNTSi
AI6IENT or POINT SOURCE (specify)*
TO BE COMPIETB) .BY LAB (PART ONE)
OPEPATION
DATE
INITIALS
COMMENTS
1. Canister cleaned
2. Filter cleaned
3. Canister evacuated
4. Canister shipped
5. Canister received
6. Analysis-coBpletetf
7. Staple discarded
TO BE COMPUTED BY LAB (PART TM>
PARAMTTER
DILUTION 1 DILUTION 2 DILUTION 3 DILUTION 4
Initial PTMSUT*
Final PTMSW*
Add UK» Air
Dilution Factor
FINAL OlUtlon Factor
3-21
-------
3.7.2 SampI Ing
These tests should be performed at the specified frequency during use
of the flux chamber.
3.7.2.1 Sample Blanks
Sample blanks should be performed once dally or after extremely high-
level samples. The flux chamber should be cleaned and blanks rerun until
exit concentrations are <10 ppmv or <10 percent of expected concentrations,
whichever Is smaller.
3.7.2.2 DupI leate Samples
A minimum of 10 percent of the sampling points should be sampled In
duplicate. Take the two samples over as brief a time span as feasible to
minimize any temporal variations In the emitting source.
3.7.2.3 Control Point Samples
One sampling location (grid point or unit) In each zone should be
resampled after every ten Individual measurements (or a minimum of once per
day) when an area source Is being Investigated. Preferably, this control
point should be measured at different times during the diurnal cycle (maxi-
mum difference In ambient temperatures). These values provide a measure of
temporal variability of the emission rate from the area source.
3.7.3 Analytical
3.7.3.1 Real-Time Analyzers
Real-time measurements are typically made with portable total hydrocar-
bon analyzers. Real-time analyses are useful for relative measurements
(I.e., to determine If steady-state operation of the flux chamber has been
attained or to determine the zoning boundaries). Each day following cali-
bration, the analyzer should be challenged with the QC gas (Section
3.4.3.1.3). Analyzer performance should be considered acceptable If the
measured concentration Is within 20 percent of the certified concentration.
If this criterion Is not met, the QC analysis should be repeated. If the
criterion Is still not met, then dally calibration should be repeated.
At the conclusion of each day, the QC gas should be reIntroduced to the
analyzer. The difference between pretesting and posttestlng responses pro-
vides a measure of upscale drift. Drifts >30 percent should be flagged and
not relied upon. If these data are necessary, then resample the grid points
sampled on that day.
3.7.3.2 Discrete Analyzers
Each day after calibration, the analyzer should be challenged with the
QC gas (Section 3.4.3.2.3). Analyzer performance should be considered
3-22
-------
acceptable If the measured concentration Is within 1.0 percent of the certi-
fied concentration. Jf this criterion !s not met, repeat the QC gas analy-
sis. If the criterion still cannot be met, then repeat the dally calibra-
tion (Section 3.6.2.2).
At the conclusion of each day's testing, the QC gas and zero grade gas
should be relntroduced to the analyzer. The differences between pretesting
and posttestlng values provide a measure of upscale and zero drifts. Dally
drift results that show >20 percent should be flagged and tests repeated If
determined necessary.
3.7.3.3 Analysis of Integrated Samples
Quality control for the analysis of Integrated samples should Include a
minimum of 10 percent analytical blanks and 10 percent duplIcate analysts.
It Is recommended that duplicate samples each be analyzed In duplicate to
provide Information on analytical as well as sampling variation. A con-
venient technique Is the use of a nested sampling scheme as shown In Figure
3-6.
3.8 CALCULATIONS
3.8.1 Definitions
A m surface area enclosed by the flux chamber (0.130 m2)
a * number of carbon atoms per compound molecule
Ct * confidence Interval for the area source emission rate mean
(±ug/mln-m2)
confidence Interval for the zone K emission rate mean (±ug/mln*m^)
measured concentration of species I (ppmv) corrected for dilution
CJT « theoretical concentration of species I (ppmv)
* measured concentration for point I In zone K, total NWC (pprav-C)
« coefficient of variance for zone K 1%)
E * mean emission rate for tha area source (ug/mln-m2)
K * zone K emission rate mean (ug/nln^m2)
* measured emission rate for point I In zone K (ug/mln-m2)
« measured emission rate for point I In zone K (ug/mln-m2) corrected
for temperature variations
Mf « molecular weight of compound (g/mole)
3-23
-------
FIGURE 3-6
NESTED SAMPLING SCHEME
Event
Sample 1
Duplicate
saapleB
(sanpling
variability)
Analysis 1-1 Analyst* 1-2
Analysis 2-1
Analysis 2-2
Duplicate
analyaia
(analytical
variability)
-------
N « total number of grid points sampled In the area source (alt zones)
K " final number of grid points (units) sampled In zone K
nj( « Initial number of grid points (units) sampled fn zone K
P » atmospheric pressure (atm)
0 » sweep air f lor rate (L/»!n)
R » gas constant (0.08205 L-atm/mol «K)
S « standard error of the overall area source emission rate mean
Sg " zone K emission rate variance
T * temperature of laboratory where analyzer Is located (K)
TEMP » temperature of the flux chamber sir <°C)
f-Q.025 * tne 97.5th percentage pofnt of a student's t-dfstrfbutfon (Table
5-4)
V * volume enclosed by the flux chamber (301)
WK » the fraction of the site represented by the zone K (zone area
(m2)/slte area (m?))
YKJ m measured concentration for point I In zone K, total tWC (ug/L)
o« parameter defining the level of confidence. 100X1-20) percent
Y » total number of zones In the total area source
P- confidence Interval (f)
T » measure of residence time Y/Q (mln)
3.8.2 Percent Recovery
The percent recovery measurements used to characterize the flux chamber
performance are calculated accordingly:
Percent Recovery « (CIM/CIT) x '00 (5-1)
where: CJM * the measured concentration of species I (ppmv) corrected for
dilution as follows;
C,M « (1/DF) x C (5-ta)
3-25
-------
TABLE 3-4
TABULATED VALUES OF STUDENT'S »t"
Degrees of Tabulated Degrees of Tabulated
Freedom* "t" Value** Freedom* »t" Value**
1 12.706 21 2.080
2 4.303 22 2.074
3 3.182 23 2.069
4 2.776 24 2.064
5 2.571 25 2.060
6 2.477 26 2.056
7 2.365 27 2.052
8 2.306 28 2.048
9 2.262 29 2.045
10 2.228 30 2.042
11 2.201 40 2.021
12 2.179 60 2.000
13 2.160 120 1.980
14 2.145 » 1.960
15 2.131
16 2.120
17 2.110
18 2.101
19 2.093
20 2.086
•Degrees of freedom (df) are equal to the number of samples collected less
one.
••Tabulated "t" values are for a two-tailed confidence Interval and a
probability of 0.05 (the same values are applicable to a one-tailed
confidence Interval and a probability of 0.025).
3-26
-------
where C Is the sample concentration (ppav) and
OF Is the dilution factor calculated as follows:
S1/(S2+S1)
where 8j Is the flow rate of the trace gas and
$2 fs the sweep air flow rate
the true concentration of species I, gas cylinder value (pp«v)
3.8.3 Calculation of the Dilution Factor Involved In Gas Canister
Analysis
Analyzing the gas canisters requires pressurizing the canister with
nitrogen. This Intrloduces a dilution which must be accounted for as
fol lows:
DF « (P2 - Pi)/(14.7 + P3) C3-2)
where: P\ » the measured pressure after cleaning and canister evacuation
prior to sampling (pslg)
?2 * the measured pressure after sample collection (pslg)
?3 = the measured pressure after pressurizing with nitrogen (pstg)
The temperature Is not required If all pressure Measurements used In this
equation are performed In the same laboratory (I.e., same temperature) after
the canisters have thermally equilibrated.
3.8.4 Area Source Emission Rate Equations
The number of units or grids (n«) to be sampled per zone (K) is depen-
dent upon the zone area as follows:
nK « 6 + 0.15 Varea of zone K (m2) (3-3)
Flux chamber measurements taken at each of the n« sampling units are
measured In terms of ppmv-C. To calculate an emission rate representing the
sampled unit, the measured concentration (Cgf) oust first be converted from
ppmv-C to ug/L as fol lows:
YK| - (P/CR'TWMf/aXfci (3-4)
where P Is pressure (atm), R Is Ryd berg's gas constant (L-atm/mole-K), T It
the flux chamber air temperature (K) (Section 3.5.4.5), Ml Is the species*
molecular weight Cg/mole), a Is the number of notes of carbon per note, CKJ
Is the measured concentration of sampled unit I In zone K (ppmv-C), and Y«|
Is the measured concentration of sampled unit I In zone K (ug/L).
The emission rate for point I In zone K (Ej<|) Is then calculated using
the converted gas concentration (ug/L) as follows:
3-27
-------
(0-YKj)/A (3-5)
where Q Is the flux chamber sweep air flow rate (L/mln), A Is the enclosed
surface area measured (m2), and EK| Is the emission rate measured for point
I In zone K (ug/m^-mln).
Prior to calculating a mean emission rate for the zone neasured, the
emission rates measured for 1he Individual sampling points need to be cor-
rected for fluctuations In chamber air the temperature (I.e., atnosphere
temperature).
The approach used to develop the correction procedure Involved devel-
oping an empirical equation to predict emission rates as a function of
chamber air temperature. (4) The resulting emission rate equation was then
used to define the correction factor (C), as follows:
C - EFs/EFa (3-6)
where: EFS =• emission factor calculated at the nominal chamber air tempera-
ture (Section 3.5.4.5)
EFa * emission factor calculated at the measured chamber air tem-
perature
Both EFS and EFa are predicted using the proper chamber air tempera-
tures and the following equation:
EF(s or a) * exp CO.Ot3(TEMP(s or a)): (3-7)
where TEMP Is measured In °C.
The measured emission rate (EF«|) Is then corrected to the nominal
emission rate (EFCK|) accordingly:
EcKi = C-EFKi (3-8)
The above procedure has a significance level (I.e., probability that
the correlation between chamber air temperature and emission rate measured
ts due to chance) of 0.4 percent. The standard error of the coefficient In
Equation 3-7 Is ±0.003.
The mean emission rate for each zone Is then calculated accordingly:
E . -L ? E (3-9)
where E^i Is the temperature corrected emission rates (Equation 3-8) and
Is the number of points sampled In zone K (Section 3.9.4,7).
For each zone (K) sampled, the zone variance ($£) and coefficient of
variance (CVg) must be determined as follows:
3-28
-------
(ECKI' - "K
1-1
(3-10)
cvK - too • SK/EK (5-n>
where OK, E^m, and EK are defined In Equations 3-3, 3-8, and 3-9, respec-
tively. The standard deviation
require* from a given zone. If NK > n«, then Ng-ng additional samples oust
foe collected from zone K.
Collect any additional samples and recalculate the emission estimates
for the jone mean (EK) and variance (3£) using Equations 3-9 and 3-10,
respectively. If ty-ry^ additional samples were collected., then use ^
samples Jnstead_of n^ In the recalculations. The overall area source mean
emission rate (E) is then calculated as follows:
Y
E« E vL (3-13)
where EK Is defined by Equation 3-9, WK 's tne fraction of site covered by
zone K (zone area/site area) and-Y Is the total number of zones sampled.
3-29
-------
Ffnaffy, calculate the variance of the overat( area source mean
ami the confidence Intervals for each zone K (C!K) and area source (CD
emission rate mean as follows:
CIK
Cl - E ± t0.025'S
3-30
-------
SECTION 4
CASE STUDY
To supplement the protocol presented In Section 3, a case study will be
reviewed. This study will Illustrate an actual application of the protocol.
Calculations and pertinent decisions will be presented.
The site, referred to as the Bon I fay Spill Site, was the scene of an
accidental spill of 5500 gallons of JP-4 aviation fuel. The spill site
occurred near the Intersection of two roads. The majority of the contami-
nated soil was excavated. The residual product extended over two areas, 30
feet of unvegetated right-of-way along the highway and Into a pine forest
containing dense underbrush.
The free surface of the water table was three feet below the land sur-
face. The thickness of the uneonso11 dated sediments that comprised the
water table aquifer at the site ranged from 20 to 50 feet. The state
aquifer underlaid this sediment layer. Contamination of the free water
table surface was expected since It was only 3 feet below I and surf ace.
However, the state aquifer was not considered threatened due to the contami-
nants net upward hydraulic gradient.
A preliminary survey was performed to define the contaminated area. A
series of ten borings Indicated that the contaminants had percolated down-
ward to the capillary fringe and moved laterally down gradient. A lens of
product several Inches thick was detected at a depth of seven feet below
land surface. The estimated extent of contamination at the time of the
survey study was 7,000 square feet (Figure 4-1).
Results from a preliminary emissions survey performed with a portable
real-time analyzer (organic vapor analyzer) held a few Inches above ground
were used to divide the area source Into emission zones for grlddlng pur-
poses. The survey Indicated only one zone was present, and the site was
gridded accordingly. The field data for the survey Is shown In Table 4-1.
The grid system used Is shown In Figure 4-2.
Surface emission measurements were made Initially at eight sampling
grid points. The protocol, at that tine, called for the Minimum number of
samp I Ing points per zone, n«, to be selected according to the following
equation (note, this equation has since been changed to Equation 3-3).
6 + 0.1-yj zone area (m2)
4-1
-------
FIGURE 4-1
SCHEMATIC DIAGRAM OF BONIFAY SPILL SITE, MONITOR WELLS, AND
EXPLORATORY BORINGS {BROWN AND KIRXNER, INC., 1983)
o
o
t-
APPROXIMATE EXTENT OF
AREA
PROJECT AREA OF
SUBSURFACE CONTAMINATION
.3
KEY
0 AREA OF MAXIMUM PRODUCT ACCUMULATION
IMMEDIATELY AFTER SPILL
2m WATER-LEVEL MONITOR WELL AND
IDENTIFICATION NUMBER
P62 EXPLORATORY BORING AND
IDENTIFICATION NUMBER
SCALE]
M
75 WO FEET
4-2
-------
TABLE 4-1
DATA SHEET FOR UNDISTURBED SURFACE SURVEY
Operators
Heathen
Grid
Point
01
02
04
08
14
20
21
22
23
24
25
16
18
19
Well P-3
Well P-4
Well P-7
Well P-7
BME
Temperature - 45°F. Light t^reez
-------
FIGURE 4-2
SCHEMATIC DIAGRAM OF SAMPLING GRID AT BOMIFAY SPILL SITE
X
~l
KEY
POINT RANDOMLY SELECTED FOR SAMPLING
USING EMISSION ISOLATION FLUX CHAMBER
R
|
0
i
Ul
s
W
/^/\
f f \
' 01 ,/
///
///.
/&
/or/i o
f%
p-i
O MONITOR WELL AND IDENTIFICATION NUMBER
P3 AREA OF MAXIMUM PRODUCT ACCUMULATION
^ IMMEDIATELY AFTEH SPILL
\ 7000 FT* SITE AREA TOTAL
\ 280 FT3 ORID AREA (4H OF TOTAL)
\ SCALE: 3/4' «
' \
/V^ 05 06* 0\
^22z..--.—_i____j
O|
\l
08* 1 09 10 11
1
-X 1__
_ _j
\ :
14* ! IS* 16 17
\*
1
.P-2, ^-5.._
?v
20 JNV21 22 23*
! X
w
^
\ 1
N 1
12 | 13*
\ |
...A ..
18 \| 19*
-poe-4--pi7
24 J\ 25*
S
^r \
5
- I
o
P-3
P-4
O
TO
BONiFAY
4-4
-------
For the single zone at Bonffay, this reduced to»
n« >. 6 * 0.1 -^650 m2 - 8.5
The 8 locations were selected through the use of a random number table.
Appendix A. Grid point 08 was selected to be the control point (I.e., a
sampling point to be repeated each day) since It was believed that emissions
would be of the largest magnitude at that location. At each sampling loca-
tion a gas syringe sample was taken for on-slte analysts. At several sam-
p| Ing locations, a gas canister was collected tn addition to the syringe
samples for off-site detailed analysis. A sample field data sheet Is shown
In Figure 4-3. The results of the emission rate measurement are given In
Table 4-2, and a sample calculation Is given In Table 4-3.
Total non-methane emission rates were calculated for each grid point
based on the on-slte analytical date. These emission rates are also pre-
sented In Table 4-2. The variation (spatially and temporally) In measured
emission rates over the extent of the contaminated area was large (93.8
percent coefficient of variation). Replicate sampling at tfie control point
allowed an estimate of the emission rate temporal variability. The temporal
variability was also large (96.0 percent). The major contributor to the
variation In measured emission rates from point-to-point can, therefore, be
attributed to day-to-day (temporal) variability. The spatial variability
was then estimated to to negligible. Using Table 3-3 to determine the total
number (%) of samples to be collected based upon the spatial variability
shows that at least 17 samples should have been collected. Although addi-
tional samples were required to be collected, sampling was terminated due to
rain. It was real Ized that the lack of a complete data sot would then
result tn a larger emission rate confidence Interval.
Using the following equation, the 95 percent confidence Interval (Cl)
for the zone emission rate was estimated.
Cl « ER t1,
where ER Is the mean emission rcte of the zone, s2 Is the zone variance, NK
Is the total number erf sites sampled, and t0tj)25 is obtained from Table 3-4.
The 95 percent confidence Interval for the zone emission rate Is from 11.3
ug/mln**2 to 55.2 uy/mln-w2.
4-5
-------
FIGURE 4-3
FIELD DATA SHEET FOR ISOLATION FLUX CHAMBER SAMPLING AT GRID POINT 08
Date 1-13-84
. Sampler s_
Location Bonlfay Spill Site, Grid Point 08
KtE
Concurrent Activity None
Surface Description Sand
Time
0858
0902
3906
0910
0914
0913
0933
Purge Air
or
FlovraCe
4.86 L/mln
4.86 L/min
4.86 L/min
4.86 L/min
4.86 L/min
4.86 L/mln
4.66 L/min
Residence
Time Num-
ber (T)
0
1
2
3
4
5
9
Temp. "F
Surface Air
46 48
Gas Data
OVA J>pnv KNU ppmv
0.15
0.16
0.16
4.0 0.16
4.0
-
4.0 0.16
Air
Sample
Number
Canister B003
Gas Svrtnge
B002
Comments OVA background - 4 ppm. Some trouble with syringe needle
plugging
4-6
-------
TABLE 4-2
RESULTS OF GC ANALYSIS OF GAS SYRINGE TAKEN DURING aUX CHAMBER SAMPLING
Grid
Point
4
6
8
8
8
14
15
19
23
25
Sample
No.
8004
801 7-A
8001
B002
8016
B006
8013
B009
8011
8008
Total NMHC Syringe
r* — «..»
(ppmv-C)
1/13/84 1.0
1/14/84 6.8
1/12/84 2.0
1/13/84
1/14/84
1/13/84
1/13/84
1/13/84
1/13/84
.0
.0
.0
.0
.0
.0
1/13/84 8.8
(ug/L)
0.62
4.2
1.2
0.62
0.62
0.62
0.62
0.62
0.62
5.4
Sweep Air
Rate
(L/mln)
2.60
2.60
5.00
4.86
2.60
2.60
2.60
2.60
2.60
2.60
Atmospheric
Temperature
•p oC
47
51*
42
48
52
45
51
51
50
53
8.3
10.6
5.5
8.9
11.1
7.2
10.6
10.6
10.0
11.7
Average Emission
Rate
(ug/mz*mln)
14.4
72.6
79.6
24.9
10.0
16.6
10.7
10.7
11.5
81.4
Variability
Spatial and Temporali
Mean
Standard Deviation
CV(Jf)
Temporals (Control Points)
Mean
Standard Deviation
33.24
31.17
93.8
38.2
36.6
96.0
•Surface temperature used rather than the chamber air temperature due to a large temperature
differential not present In the other measurements. This Is suggestive of an error In chamber air
temperature measurement.
-------
TABLE 4-3
SAMPLE CALCULATIONS OF THE EMISSION RATE FOR GRID POINT 08 ON 1/13/84
Concentration Conversion:
Y, - (P/(R-T))(MW/a)(C|) (Equation 3-4)
where: P
R
T
MM
a
1 atm
0.08205 L-a-hn/mole-K
282.6K (average area site air temperature)
86.18 g/fflole (referenced to hexane)
6 moles of carbon/mole of hexane
1.0 ppmv-C
1 atm 86.18 g/mole
YI * • . x ———————- x 1.0 DDfliv-C
' (0.08205 L-atm/mole-KM282.6K) 6 mole C/mote
YI « 0.6194 ug/L
Emission Rate (uncorrected)
EI - (Q'Y|)/A (Equation 3-5)
where: 0 " 4.86 L/mln
YI » 0.6194 ug/L
A * 0.130 m2
4.86 l/mln-0.6194 ug/L
El * 0.130 m2
E| » 23.15 ug/mln-mZ
Emission Rate Correction Factor
EFS - expC0.13(TEMPs)] (Equation 3-7)
where: TEMPS » 9.45*C (nominal chamber air temperature *C)
EFS « emission factor at nominal chamber atr temperature
EFS - exp (0.13*9.45)
EFS - 3.416
(Continued)
4-8
-------
TABLE 4-3
(Continued)
EFa " exp[0.13(TEMPa)D
where: TEMPa « 8.9*C (measured chamber air temperature *C)
EFa » emission factor at the measured chamber air temperature.
EFa - exp(0.13-8.9)
EFa « 3.160
C « EFs/EFa (Equation 3-6)
C » 3.416/3.180
C » 1.074
Emission Rate (corrected for temperature variation)
EC| » C'Ej (Equation 3-8)
Eg] « 1.074-23.15 ug/mtn-m2
- 24.86
• 24.9 ug/mln-n»2
4-9
-------
SECTION 5
ADDITIONAL INFORMATION
For further Information on vapor/liquid equilibria (VIE) for organic
systems, the following reference Is suggested. The Intent of this bibliog-
raphy was to provide a ready listing of the references for data on VLE.
Nelson, T.P., N.P. Meserole, Annotated Bibliography of Published
Material on Vapor/Liquid Equilibria. EPA, July 1983.
For further Information on the selection of the flux chamber enclosure
method for direct measurement of gas emission rates from contaminated soils
and/or groundwater, the following reference Is suggested.
Radian Corporation. Soil Gas Sampling Techniques of Chemicals for
Exposure Assessment, Interim Report. EPA Contract No. 68-02-3513, Work
Assignment 32, August 1983.
For further Information on the actual field applications of this tech-
nique, the following references are suggested:
Radian Corporation, Soil Gas Sampling Techniques of Chemicals for
Exposure Assessment: Tustln Spill Site Data Volume. EPA Contract No.
68-02-3513, Work Assignment 32. July 27, 1984.
Radian Corporation, Soil Gas Sampling Techniques of Chemicals for
Exposure Assessment, BonI fay Spill Site Data Volume. EPA Contract No.
68-02-3513, Work Assignment 32, 1984.
For further Information on the validation of the flux chamber technique
for emission rate measurements on soil surfaces, the following reference Is
suggested!
Ktenbusch, M.R., D. Ranum, Validation of Flux Chamber Emission
Measurements on Soil Surfaces. EPA Contract No. 68-02-3889, Work
Assignment 18, December 1985.
For Information concerning the emission process Including diffusion and
adsorption, the following reference Is suggested:
5-1
-------
Manos, C.G., Jr., Effects of Clay Mineral Organic Hatter Complexes on
YOG Adsorption, Draft Report. EPA Contract No. 68-02-3889, Work
Assignment 18, October 3, 1985.
Radian Corporation. Soil Gas Sampling Techniques of Chemicals for
Exposure Assessment; Laboratory Study of Emission Rates from Soil
Columns, Draft Final Report, EPA Contract No. 68-02-3513, Work
Assignment 32, October 1984.
5-2
-------
REFERENCES
1. Farmer, W.J., M.S. Yang, and J. Letey. Land Disposal of Hexachloro-
benzene Wastes—Controlling Vapor Movement In Soil. EPA-600/2-80-119,
Municipal Environmental Research Laboratory, Cincinnati, Ohio, August
1981.
2. Shen, T.T. Estimating Hazardous Air Emissions from Disposal Sites,
Pollution Engineering, August 1981.
3. U.S. EPA, Office of Solid Waste. Guidance Document for Subpart F, Air
Emission Monitoring, Land Disposal Toxic Air Emissions Evaluation
Guideline. December 1980.
4. Klenbusch, M.R, D. Ranum. Validation of Flux Chamber Emission Measure-
ments on Soil Surfaces. EPA, EMSL, Contract No. 68-02-3889, Work
Assignment 18, December 1985.
5. Radian Corporation, Soil Gas Sampling Techniques of Chenlea Is for
Exposure Assessment, Interim Report, EPA Contract No. 68-02-3513, Work
Assignment 32, U.S. EPA EMSL, EAO, August 1983.
6. Adams, O.F., M.R. Pack, W.L. Bamesberger, and A.E. Sherrard,
"Measurement of Blogenlc Sulfur-Containing Gas Emissions from Soils and
Vegetation." In: Proceedings of 71st Annual APCA Meeting, Houston, TX,
1978, p. 78-76.
7. Adams., D.F., Sulfur Gas Emissions from Flue Gas Desul fur Izatlon Sludge
Ponds. J. Air Pol. Contr. Assoc. Vol. 29, No. 9, p. 963-968, 1979.
8. Denmead, O.T. Chamber Systems for Measuring Nitrous Oxtde Emission
from Soils In the Field. Soil Sciences Soc. of Am. J., 43, p. 89-95,
1979.
9. Bat four, W.D. and C.E. Schmidt, Sampling Approaches fcr Measuring
Emission Rates from Hazardous Haste Disposal Facilities. In:
Proceedings of 77th Annual Meeting of the Air Pollution Control
Association, San Francisco, California, June 1984.
10. Zimmerman, P. Procedures for Conducting Hydrocarbon Emission
Inventories of Blogenlc Sources and Sane Results of Recent
Investigations. In: Proceedings of 1977 Environmental Protection
Agency Emission Inventory/Factor Workshop, Raleigh, NC, 1977.
R-1
-------
APPENDIX A
SaECTION OF A RANDOM SAMPLE
An illustration of the method of use of tables of randan numbers
follows. Suppose the population consists of 87 I tans, and we wish to select
a random sample of 10, Assign to each Individual a separate two-digit
number between 00 and 86. In a table of random numbers, pick an arbitrary
starting place and decide upon the direction of reading the numbers. Any
direction may be used, provided the rule Is fixed In advance and Is Indepen-
dent of the numbers occurring. Read two-digit numbers from the table, and
select for the sample those Individuals whose numbers occur until 10 Indi-
viduals have been selected. For example, In Table A-1, start with the
second page of the table, column 20, line 6, and read down. The 10 Items
picked for the sample would thus be numbers 38, 44, 13, 73, 39, 41, 35, 07,
14, and 47.
The method described Is applicable for obtaining simple random samples
from any sampled population consisting of a finite set of Individuals. In
the case of an Infinite sampled population for the target population of
weighings as comprising all weighings which might conceptually have been
made du-Ing the time while weighing was done. Wo cannot, by mechanical
randomization, draw a random sample from this population, and so must recog-
nize that we have a random sample only by assumption. This assumption will
be warranted If previous data Indicate that the weighing procedure Is In a
state of statistical control; unwarranted If the contrary Is Indicated; and
a leap In the dark If no previous data are available.
-------
TABLE A-1
SHORT TABLE OF RANDOM NUMBERS
46 96 85 77 27 92 8« M 45 21 89 91 71 42 M 64 58 22 TS SI 74 »1 48 4f II
44 19 15 32 63 55 87 T7 3J 29 43 00 31 34 84 05 72 90 44 27 TJ 22 07 62 IT
3439806224338167281I3479263S342309940080SS31632791
74 97 80 30 (i VI 71 30 01 M 47 45 89 70 74 13 M 90 SI 27 «1 34 63 87 44
22 U 81 60 M 3S 33 71 13 13 72 M 16 13 SO M U SI 29 U 30 93 45 M 29
40 M M 40 03 47 24 60 Of 21 M U 00 0$ II 52 M 40 73 73 57 68 3< 33 91
52 33 7« 44 54 IS 47 TS T» T3 78 19 87 M U 47 43 02 62 03 42 0$ 32 Si 02
37 59 20 40 M )V 82 24 19 90 80 87 32 74 59 84 24 49 79 17 23 7$ 83 42 00
11 02 55 57 43 84 74 U 22 47 19 20 IS 92 53 37 13 7J 54 (9 M 73 23 39 07
10 33 79 2« 34 M 71 33 89 74 68 48 23 17 49 18 81 OS $2 85 70 05 73 11 17
17 59 28 25 47 89 II 65 «5 20 42 23 96 41 64 20 30 89 87 64 37 93 M M 3S
93S07S2009I8S4S46B
-------
TABLE A-1
(CONTINUED)
OS
37
a
42
OS
57
78
rt
67
W
a
16
u
98
03
M
OC
M
41
84
24
87
82
«7
32
23
12
64
44
«2
M
46
«T
28
83
M
22
29
71
27
1C
90
01
45
48
U
97
»
OS
u
7S 28 81 it
78 (7 31 06
46 T2 OS SO
I» 47 76 30
M I> 84 M
U
63
I>
2C
20
62
60
27
72
20
82
SI
47
33
50
45
02
1&
69
87
63
07
76
92
74
to
16
SI
SI
83
36
7S
S8
>S
SI
02
12
67
23
C
76
90
06
26
10
«
41
M
41
a
63
16
W
76
30
60 44 18 41 23 74 T3 51 72 90 40 52 95 41 » 89 48 98 27 J3 81 33 83 *2 94
3280647S91M09406489299946356991S0737S9290S6829324
79MS37778Mt237438271007821656S884582447893227309
4$ (3 23 32 01 09 46 36 43 M 37 IS 3$ 04 88 79 83 S3 1* 13 91 S> 61 31 87
2060»7432141S422727799818330441S9024S17S663499t040
67914483432S$4332>a0995327S4I9807432S39607UM6I98
845O76938435t84S3781474492576659M144339249454644«
M 73 38 38 23 M 10 95 U 01 10 01 59 71 55 99 24 U 31 41 00 T3 1> 80 62
55 11 SO 29 17 73 97 04 20 39 20 22 71 11 4* 00 16 10 12 35 09 U 00 8f OS
23543387929204497394S7S3S7M9309M87U«T4629SOrr85
41 43 67 7» 44 57 40 29 10 34 58 63 SI 18 07 41 02 39 79 14 46 68 10 01 61
03 97 71 72 43 27 3* 24 S» 88 « 87 26 31 11 44 28 S8 99 47 83 21 35 22 88
90 24 83 48 07 41 56 68 11 14 77 75 48 68 08 90 89 63 87 00 06 18 63 21 91
98 98 97 42 2T II 80 51 13 13 03 42 91 14 51 22 IS 48 67 S2 09 40 34 60 M
74 20 94 21 49 9« SI 69 99 U 43 76 SS 81 M 11 88 68 32 43 08 14 78 OS 34
94 67 48 87 U 84 00 8S 93 S« U 99 21 74 84 13 S4 41 90 M 30 04 19 68 73
58 18 84 82 71 23 $6 33 19 2S 65 17 90 M 24 91 75 M 14 83 86 22 70 M W
31 47 28 24 88 49 28 69 78 62 23 «S S3 38 78 &S 87 44 91 W 91 62 78 09 20
45 62 31 06 70 92 73 27 83 57 15 (4 40 57 M 54 42 3S 40 93 55 82 08 78 87
31 49 87 12 27 41 07 91 72 64 63 «2 OC « 82 71 28 34 4S 31 99 01 03 3S 76
69 37 22 » 46 10 75 83 62 94 44 6S 46 23 65 71 69 20 89 12 16 56 61 70 41
93 67 21 54 98 4* S2 53 14 M 24 70 2S 18 23 a 56 24 03 M 11 06 46 10 23
77 54 18 37 01 32 20 18 70 79 20 U 77 89 28 17 77 IS S2 47 IS 30 35 12 7S
37 07 47 79 60 75 24 IS 31 63 25 93 27 66 19 S3 52 49 98 45 12 12 06 00 32
7208710173463960375322252084M0203B6S383S04M6994
55 12 48 46 72 SO 14 « 47 47 84 37 32 84 82 64 »7 13 69 M 20 09 80 41 75
692498907029342S33231269MS0389384322896036S70901t
01 86 77 18 21 91 « U 84 C$ 4» 75 21 94 SI 40 SI S3 36 39 77 69 06 25 «
51 40 94 0« 80 61 M 23 46 28 11 U U M 60 65 0« 63 71 06 1J SS OS 32 56
SI7I02«SM296727440T6723202StZe97S962134172T07107
3375685I003354I584»42*S0166S12615443S4U63377497S9
S860J745«20»»5M18»3422»17804t7988020M3l93139t30
721)129S3287993283«S401792S722t8MT91«23S3S4M J* 09 85 24 59 46 OS 91 U 38 62 SI 71 47 37 38
81 94 71 90 47 41 M 3« 33 93 04 90 *» 72 85 23 0 * TO SI 56 03 23 84 80
44 «2 20 81 21 57 57 M 00 47 36 10 87 22 4S 72 03 31 75 23 38 38 54 77 97
63 91 i: IS 08 02 1* 74 it ?» 21 S3 « 41 77 IS 07 » 87 11 1» 25 62 19 30
29337760290»yM42»07IS4067S629S87J84041IS43116S3
54 13 39 19 2» 64 tr 7J 71 41 T8 03 24 02 M M 69 74 74 28 08 98 84 08 23
75 18 85 «4 84 93 &5 « 08 84 IS <1 S7 84 4S U 70 13 17 « 47 80 10 13 00
34 4? 17 08 79 03 92 8* IS 42 93 48 27 37 99 98 81 M 44 72 04 95 4Z 31 IT
29
-------
TABLE A-1
CCOWTINUHJ)
«S 44 38 M M II 20 89 a S2 45 41 01 Tl SS 14 II OS II W 74 94 SO M 07
14U14572ll2U8l«704tS«»052723Mt423S20721l7I3S2ll
OS2S735123929S05S412»44<«13392133091731130442171tt
9t211024793»llOIM20l2M479«0704*29S01S CT 9* 27 70 71 04 43 IS 44 II 75 11 TO 53 21 <0 71 30 n 54 21 02 42
U 14 « 4< 15 SI 91 23 U 24 71 19 «7 11 79 90 S3 47 M 32 U « 97 10 17
«322MU10020S0347934<702S27M329l4I45M3*M91Tt79
42 U 20 M 19 U M 91 21 M U 91 74 92 31 97 tt 24 20 55 19 54 « « M
» 90 7« 51 51 49 25 51 21 Cl 55 SS 73 10 22 M 79 23 *0 03 SI 11 00 11 17
20 12 t7 40 25 45 94 3S 11
-------
TABLE A-1
(CONTINUED)
24 *1 04 U 98 24 93 58 43 M 56 26 24 45 *S »I 42 W 67 42 II 74 TT II 41
75 55 54 29 67 02 81 01 47 54 08 81 34 00 79 42 38 52 14 U 38 44 S9 41 97
41 71 80 54 37 73 34 U 74 U 91 64 82 41 02 74 12 34 71 38 43 72 84 34 27
04 19 43 35 54 M 00 41 47 44 63 13 27 SO 18 75 14 72 40 90 02 45 67 82 IS
44 IS 52 42 22 91 22 96 38 41 03 27 IS 67 26 34 81 75 U 82 94 U 62 08 94
10 80 17 17 83 05 31 23 08 07 40 00 80 44 U 70 II 31 73 OS 41 41 47 14 a
40 42 27 55 71 82 88 42 71 51 58 4t 58 75 38 23 57 04 14 19 44 M Ot U tt
ff5 57 21 21 25 12 05 41 70 28 03 M 91 37 (4 4* *> 48 St «0 8* 71 35 83 05
57 27 M M M 88 M 70 8* 59 41 84 08 32 31 75 II 1» 41 11 28 41 71 79 28
80 54 W 4$ 13 83 78 78 71 38 8) 51 U 47 35 84 It M M 88 >1 22 47 24 84
44 51 T5 51 08 17 43 53 31 09 60 34 34 81 9S M 01 M 37 13 24 M 75 29 21
55 42 48 76 SO 13 89 69 00 05 99 45 82 01 53 86 68 81 34 50 75 20 17 94 47
80 SO 67 83 01 97 78 21 44 34 62 43 02 64 38 13 60 26 32 34 81 43 17 56 41
03 64 65 44 02 75 41 13 91 28 82 97 57 38 «9 27 26 97 34 44 26 12 00 M 24
14 53 75 37 91 43 95 IS 13 2( 33 27 45 48 33 80 80 21 19 71 04 V 83 58 32
01 M 43 34 30 71 24 75 92 73 07 81 13 35 48 88 42 80 84 69 M 25 73 92 98
39 38 79 42 17 77 99 J5 32 85 13 35 48 49 80 83 59 04 34 94 04 03 41 85 02
74 9« 24 94 89 54 64 29 35 M 50 4$ (5 SO 28 62 45 80 81 95 07 9» 57 10 54
21 18 54 55 77 48 38 33 88 Si 21 56 18 93 « 94 24 80 97 03 7» 39 73 87 70
S8SI99S39S736077«10476S978W949907S391959940S441T»
44 98 27 95 19 22 29 41 54 78 83 48 49 82 79 79 20 00 24 40 22 SO 14 30 73
58 44 34 76 19 IS 00 60 50 28 32 44 IS 35 »9 28 91 SO S3 62 21 61 24 44 81
43 OS 50 00 20 39 2i 44 84 39 17 39 92 42 59 04 54 15 09 35 07 II 25 51 17
84 07 33 83 87 14 03 75 07 « 60 43 66 57 57 57 59 01 78 80 13 77 43 58 10
93 54 23 72 70 09 36 IS 24 04 74 05 85 29 M 67 37 28 13 98 01 48 29 75 89
54 44 72 02 34 52 81 38 $2 94 14 54 27 32 41 74 84 83 90 01 97 59 87 44 41
43 80 84 28 32 93 91 78 70 31 50 22 09 40 89 44 85 82 78 91 14 71 99 98 70
44 80 SO 18 92 44 42 48 47 22 87 14 20 45 82 01 45 21 49 80 17 39 70 74 03
78 70 39 30 04 59 65 14 64 04 62 28 44 64 05 89 81 80 09 89 54 11 27 81 44
14884703593215830401208292253488848076492510048402
49 28 04 18 54 78 97 49 14 85 01 58 31 18 20 53 74 03 27 05 80 39 15 47 49
99480994345410779SS89044S214S2S887SI31716SS3H8550
01 84 22 15 54 43 63 44 15 30 21 84 48 17 11 48 92 14 17 49 36 OS 17 80 24
47 85 26 91 23 14 28 01 76 47 45 12 58 24 27 41 59 43 20 15 93 47 30 54 27
I39!14749197834l99S04094M449782M04M»7452t4498M
95 82 20 95 52 45 9S 03 48 75 44 25 04 13 85 80 13 37 08 18 09 28 43 07 49
44 04 82 49 28 27 34 53 42 35 44 12 40 84 35 04 28 14 37 23 97 38 07 40 80
99 22 24 44 15 71 04 94 22 93 H 44 73 57 SI 22 54 82 37 99 94 27 25 87 77
08 44 26 12 87 72 42 13 57 77 61 07 94 24 62 17 76 19 45 18 98 11 47 40 31
14 94 74 04 37 32 09 72 81 22 87 70 81 93 78 93 37 22 32 25 38 43 38 OS 81
27 66 ft S3 58 16 49 99 19 03 62 98 79 81 98 IS 03 63 32 93 68 24 14 44 SO
99 47 81 61 25 52 97 87 98 15 85 99 01 84 59 00 M 39 32 S3 49 18 42 51 45
89 14 37 94 03 22 W 45 42 61 97 83 04 26 30 48 49 40 9» 99 6» H 13 94 21
34 13 53 15 32 42 02 58 32 14 83 73 02 82 49 25 42 91 U M 70 n 44 SO 51
72 11 79 75 79 31 07 12 92 61 89 93 77 82 08 23 74 75 67 54 37 45 35 13 44
19 72 57 II 99 08 42 02 24 82 52 90 72 51 94 84 59 79 84 19 95 71 21 49 91
9« 99 71 43 90 27 M 94 15 70 17 74 92 31 85 24 47 55 44 SI 91 47 13 31 49
44 15 86 76 18 IS 57 29 51 62 95 84 20 W 01 11 90 66 80 81 40 43 65 87 35
34 83 94 07 SO 11 89 84 14 SO 09 97 04 74 51 41 20 54 M 20 33 53 70 10 22
S3 07 OC 16 M M 43 40 $7 32 11 09 47 U 69 41 03 3* 24 02 16 4t SI 3* 58
A-5
------- |