United States Solid Waste and
Environmental Protection Emergency Resonse EPA530-R-95-036
Agency (5305W) June 1995
vvEPA Guidance for the
Sampling and
Analysis of
Municipal Waste
Combustion Ash
FortheToxicity
Characteristic
-------
EPA Pub No.: EPA530-R-95-036
GUIDANCE FOR THE SAMPLING AND ANALYSIS OF
MUNICIPAL WASTE COMBUSTION ASH
FOR THE TOXICITY CHARACTERISTIC
June 1995
Office of Solid Waste
U.S. Environmental Protection Agency
401 M Street, SW
Washington, DC 20460
-------
-------
TABLE OF CONTENTS
Section Page No.
I. INTRODUCTION: PURPOSE OF THIS GUIDANCE DOCUMENT . . 1
II. THE SAMPLING APPROACH 3
III. ANALYSIS FOR THE TOXICITY CHARACTERISTIC 8
IV. QUALITY ASSURANCE AND QUALITY CONTROL 13
V. TCLP DATA EVALUATION 14
VI. REFERENCES 19
APPENDIX: DEFINITIONS OF TERMS USED IN THE GUIDANCE
-------
-------
I. INTRODUCTION: PURPOSE OF THIS GUIDANCE DOCUMENT
The purpose of this document is to assist generators of ash
from municipal waste combustion facilities in determining whether
their ash is hazardous because it exhibits the Toxicity
Characteristic (TC). This document is guidance, and ash generators
are not required by regulation or otherwise to follow its approach
to sampling and analysis for the TC. It is also not intended that
this guidance be used to replace the guidance or requirements for
TC determinations developed by authorized States or for the
sampling and analysis of ash for any other purpose under State or
local programs.
The Toxicity Characteristic (40 CFR § 261.24) is one of four
characteristics described in Subpart C of 40 CFR part 261 by which
a hazardous waste is identified. A hazardous waste identified by
any one of these characteristics, including the TC, is subject to
the notification requirements of section 3010 of the Resource and
Recovery Act (RCRA) and all applicable requirements under parts 262
through 265, 268 and 270 of the RCRA regulations. A TC
determination is the responsibility of the generator, and is
generally made by either testing using the Toxicity Characteristic
Leaching Procedure (TCLP), or by using knowledge of the process
(pursuant to 40 CFR 262.11). All solid waste, unless excluded by
40 CFR § 261.4, is subject to this determination.
Ash generated by municipal waste combustion (MWC) facilities
with resource recovery is not exempt from RCRA, Subtitle C,
regulation. Therefore, persons who generate such ash must
determine whether their ash is hazardous because it exhibits the
TC. Ash first becomes subject to this hazardous waste
determination at the point that the ash leaves the "resource
recovery facility", defined as the combustion building (including
connected air pollution equipment). For further information
regarding the Agency's interpretation of when RCRA, Subtitle C,
jurisdiction begins for MWC ash at waste-to-energy facilities, see
60 FR 6666, February 3, 1995 and EPA's "Revised Implementation
Strategy for City of Chicago vs. EOF Municipal Waste Combustion Ash
(MWC) Supreme Court Decision" found in the public docket for this
guidance (Docket No. F-95-MRIF-FFFFF). Any ash that exhibits the
TC when exiting the combustion building must be managed in
compliance with all applicable Subtitle C requirements.
This guidance assumes that the generator has elected to
conduct testing to determine whether the ash exhibits the TC for
any of the TC contaminants. In addition to general information
regarding this determination, this document includes a sampling and
analysis approach which is one example of a prescriptive sampling
and analysis plan that can be used by MWC facilities, especially
those that do not have the resources to develop their own plans.
The Agency maintains that, where possible, generators should use
-------
the guidance of Chapter Nine in "Test Methods for Evaluating Solid
Waste, Physical/Chemical Methods" (SW-846), the Agency's guidance
for all RCRA sampling (including TC determinations), to develop a
sampling and analysis plan tailored to site-specific conditions and
to meet the data quality objectives (DQOs) of the study.
This document contains the following sections:
II. The Sampling Approach; Discusses typical concerns during
development of a sampling plan, and presents one example
of an approach to ash sampling.
III. Analysis for the Toxicity Characteristic; Describes
analysis using the TCLP from "Test Methods for Evaluating
Solid Waste, Physical/Chemical Methods"(SW-846) for the
contaminants listed in 40 CFR § 261.24.
IV. Quality Assurance and Quality Control; Discusses the
importance of quality assurance/quality control (QA/QC)
and references portions of SW-846 that contain
information regarding QA/QC.
V. TCLP Data Evaluation; Describes criteria for evaluating
data to determine whether ash is hazardous for the TC.
VI. References; Provides a listing of resources for
designing a sampling and analysis plan.
Appendix: Definitions of Terms Used in the Guidance
-------
II. THE SAMPLING APPROACH
As explained in "Introduction: Purpose of this Guidance
Document", the generator (e.g., facility owner/operator) is
responsible for determining whether his ash exhibits the TC. Ash
first becomes subject to this determination at the point that it
leaves the "resource recovery facility", defined as the combustion
building (including connected air pollution control equipment).
All handling of the ash within the building is exempt from RCRA
Subtitle C regulations (see 60 FR 6666, February 3, 1995). Thus,
a facility can treat the ash or combine fly ash and bottom ash
within the combustion building before collecting samples outside
the building for the hazardous waste determination.
Nearly every resource recovery facility is configured
differently. In several instances, these facilities are not
confined within a single structure enclosed by four walls. A few
facilities, in fact, exist where the combustion device is not
enclosed at all within a building structure. However, in WTE
facilities where the ash always moves between structures in
enclosed conveyors, such configurations fall within the common
sense meaning of "resource recovery facility". In contrast, some
facilities may collect bottom ash within the combustion building
housing the combustion device and collect the fly ash outside the
combustion device building in a manner that exposes that ash to the
environment. In that case, RCRA Subtitle C jurisdiction begins at
the two exit points from the resource recovery facility;
specifically at: (1) the point where the bottom ash leaves the
combustion device building, and (2) the point where the fly ash
becomes exposed to the environment as it is discharged from the air
pollution equipment into the containers. Thus, the generator
should collect samples for a TC determination at each exit point.
In any case, should a generator determine that either bottom
ash, fly ash, or combined ash is hazardous based on the TC,
management of that ash must be conducted pursuant to RCRA Subtitle
C. [For further information regarding the Agency's interpretation
of when RCRA Subtitle C jurisdiction begins for MWC ash at waste-
to-energy facilities, see 60 FR 6666, February 3, 1995 and EPA's
"Revised Implementation Strategy for City of Chicago vs. EOF
Municipal Waste Combustion (MWC) Supreme Court Decision" found in
the public docket for this guidance (Docket No. F-95-MRIF-FFFFF.)]
The generator also is responsible for ascertaining ash
variability over time and has a continuing responsibility for
knowing whether the ash is hazardous at any point in time. Thus,
given this responsibility for making an accurate TC determination
and insuring appropriate ash management (e.g., management as a
nonhazardous or hazardous waste), it is up to the generator to
decide how frequently retesting or reevaluation of ash for a TC
determination should be made. Generators should also consult with
-------
their authorized States regarding any State requirements or
recommendations regarding both the initial characterization and any
recharacterization of their ash for the TC.
The Agency recommends retesting or reevaluation of MWC ash for
the TC whenever the generator suspects that the leachability of ash
for the TC contaminants may have significantly changed, e.g., such
that a previous determination that the ash is nonhazardous for the
TC may no longer be accurate. In determining whether to
recharacterize ash, the generator should consider all facility-
specific and external factors that could cause ash leachability to
vary. For example, a facility may retrofit pollution control
equipment by adding a scrubber acid gas removal system and a fabric
filter to replace an existing electrostatic precipitator. This
equipment change may significantly change the leachability of
certain TC metals of concern in the combustion residue. The
leachability of ash may be affected by changes in ash treatment
(e.g., lime addition) or conditioning practices which occur before
it is subject to a TC determination. These factors also can be
used to identify which contaminants of the TC should be the
analytes of concern during the retesting or reevaluation.
The sampling plan in this section is just one example of a
prescriptive sample collection approach that can be used by MWC
facilities, especially those that do not have the resources to
develop their own plans. The Agency maintains that, where
possible, generators should use Chapter Nine of SW-846, the
Agency's guidance for all RCRA sampling (including TC
determinations), to develop a sampling and analysis plan tailored
to site-specific conditions and to meet the DQOs of the study.
Regardless of whether SW-846, this guidance, or State guidance
is used, common objectives include the need to obtain
representative samples which exhibit the average properties of the
ash as a whole, and to make a correct determination regarding the
status of the ash under RCRA. Determining whether ash passes or
fails the TC requires reliable information on the leachability of
the TC contaminants of concern in the ash. Several factors
contribute to this reliability, including accuracy, precision, and
the prevention of bias. (See Chapter Nine of SW-846 for an in-
depth discussion regarding the consideration of these factors in
any sampling effort.)
The approach in this guidance is designed to determine the
concentration of TC contaminants in the ash leachate through the
collection and analysis of fourteen (14) composite ash samples over
a minimum of one-week of operation. This approach is one example
of a sampling approach that might be taken by some facilities.
However, it may not be appropriate for all facilities. It is
largely based on the assumption that one week is an adequate
sampling period for the collection of samples that are fully
representative of any temporal variability in the ash. If the
-------
above assumption is not valid or if the generator finds for any
reason that the approach may compromise the collection of
representative samples, then a facility-specific sampling and
analysis program designed by knowledgeable personnel should instead
be employed. Examples of "knowledgeable personnel" that might be
involved in designing a sampling plan include the end user of the
data, an experienced member of the sample collection team, an
analytical chemist familiar with the analytical requirements, an
engineer or other person familiar with the process, a statistician,
and a quality assurance representative.
The sampling procedure is as follows:
1; Determine the most convenient location for sampling at the
point the ash exits the combustion building and is subject to
RCRA Subtitle C jurisdiction. For example, sampling can be
conducted either from transport vehicles, the ash conveyance
device, or an ash pile.
2. Obtain or construct a sampling device (trough, bucket,
shovel, thief, etc.) to be used to gather a grab sample of the
entire depth of the hopper, pile, or truck load, or the entire
depth and width of the belt conveyor, drag chain flight, or
vibrating conveyor.
3. If a conveyor is to be the sample location point, collect
the entire width and depth of the conveyor at a fixed point
each hour for eight (8) hours. If trucks are to be sampled,
randomly select 8 trucks to sample during the eight (8) hour
period. (In certain situations, where less than 8 truckloads
are generated, a different schedule may be necessary, e.g.,
less than 1 truck per hour.) Composite all samples for the
period into an eight (8) hour composite. Containerize, label,
and set aside for reduction.
4. Collect a second eight (8) hour composite during the
course of the work day. The second composite should be
collected during a different shift from the first composite.
5. For an initial ash characterization, samples should be
collected each day for a minimum of one week's operation. Two
daily composite samples over the course of one week will yield
a total of 14 composite samples.
6. Each composite should be mixed (a cement or other
mechanical mixer is acceptable for this purpose), and then a
representative subsample should be obtained from the
composite. The subsample should be obtained by taking a full
core or "slice" of equal proportions through the mixed
composite. To reduce the size of this subsample and obtain a
1000-gram aliquot suitable for shipment to the laboratory, the
sample may need to be riffled or coned and quartered. Another
-------
acceptable procedure (for example, if riffling is difficult
due to ash moisture content) is alternate shoveling, whereby
the sample is divided (subsampled) using a system of alternate
shoveling wherein the large composite sample is apportioned
into two or more smaller piles. One of the small piles
(subsamples) is then randomly chosen for analysis.
7. In accordance with the TCLP, each composite sample (e.g.,
the 1000-gram aliquot obtained during step 6) should be passed
over a 3/8-inch (9.5 mm) screen. Materials which do not pass
through the screen should be subjected to a particle size
reduction step. Materials can be reduced by crushing,
cutting, or grinding. A mechanical crusher can be used.
Sometimes, ash contains large pieces of structurally intact
material that cannot be crushed or otherwise reduced by means
available to generators for sample reduction. The hammer blow
test can be used to determine whether a particle is a
candidate for size reduction. (Note: the hammer test is not
itself a method of particle size reduction, but rather is a
method of determining whether the material can be reduced.)
In that test, the material is subjected to blows with a 5-
pound sledge hammer dropped from one foot above the pieces.
The hammer blow test should be performed on a hard surface
(e.g., iron or steel plate) which will not break upon impact
by the hammer and will not cause sample contamination or loss.
Particles that do not pass the 3/8-inch screen after the
particle size reduction step are discarded. It is not
necessary to weigh the discarded material and, in the case of
MWC ash analysis for the TC, TCLP results should not be
adjusted based on the weight of discarded material.
8. Samples should be properly labelled and stored. Submit
samples for analysis by the TCLP (Method 1311 of SW-846).
Once a sample has been collected, it must be stored and
preserved to maintain the chemical and physical properties that it
possessed at the time of collection. The sample type, type of
containers and their preparation, possible forms of contamination,
and preservation methods are all items which must be thoroughly
examined in order to maintain the integrity of samples. SW-846
contains guidance in its chapters and methods regarding these
important considerations.
For the purposes of a TC determination, the ash and ash
samples should not be dried before sample reduction and analysis
unless that represents the actual state of the ash at the point of
Subtitle C jurisdiction. (Sample drying may be appropriate for
other types of ash characterizations; in those instances, the
generator should check with his State or other appropriate
regulatory authority for guidance.) Also, volatile organics are
not expected to be detected in ash because municipal waste
-------
combustion facilities typically operate at temperatures higher than
the boiling points of the compounds. However, should the generator
have reason to believe that TC volatile organic contaminants are
present in the ash, practical measures should be taken to avoid the
loss of those volatile organics. Guidance regarding such measures
can be found in SW-846.
-------
III. ANALYSIS FOR THE TOXICITY CHARACTERISTIC
In order to determine whether the ash exhibits the TC by
conducting analysis for the TC contaminants, the Toxicity
Characteristic Leaching Procedure (TCLP, Method 1311 of SW-846)
must be used pursuant to 40 CFR § 261.24 (a). Use of the TCLP
generates an extract (also called a leachate), which is then
analyzed for the TC contaminants listed in Table 1 of 40 CFR §
261.24 (see Exhibit 1) using the appropriate determinative methods.
Determinative methods for all of the TC contaminants can be found
in SW-846 (see Exhibit 2). Depending on the analytes of concern
and other factors, more than one aliquot of at least 100 grams from
each composite sample may be needed for use in the TCLP. Process
or waste knowledge can be used in lieu of testing for a TC
determination regarding any of the TC constituents. This guidance
assumes that the generator has elected to at least initially test
for most (inorganic and organic) of the TC contaminants.
After sample preparation and selection of an extraction fluid,
the TCLP consists of mixing 100 grams of sample with an acetic acid
extraction fluid in a liguid-to-solid ratio of 20:1. The sample
extract is then agitated end-over-end for 18 hours, after which it
is filtered through a 0.7 /urn filter and the filtrate is analyzed
for the contaminants found in Table 1 of 40 CFR § 261.24. The
analyst must follow section 8.5 of the TCLP regarding the maximum
sample holding times for all stages of a TC determination.
Given the low probability of their occurrence in municipal
waste combustion ash, it is recommended that, should the generator
believe TC organic contaminants are present, analysis for the TC
organic compounds only occur in the first one or two extracts of
each sampling and analysis event (during the initial TC
determination testing event and subsequently if the generator
believes that the leachability of the organic contaminants of
concern in the ash may have changed). It is further recommended
that only if one or more TC organic compounds (semivolatile or
volatile) are detected in the first extracts should the remaining
extracts be analyzed for the TC organics.
Prior to analysis of the extracts using atomic absorption
spectrometry (AA), inductively coupled plasma spectroscopy (ICP),
gas chromatography (GC) or other appropriate determinative
procedure, the extracts should be prepared using the appropriate
methods (e.g., Methods 3010 and 3510, see Exhibit 2). The SW-846
manual contains several analytical techniques for trace metal
determinations: ICP atomic emission spectroscopy (ICP-AES) and ICP
mass spectrometry (ICP-MS), direct aspiration flame atomic
absorption (FAA), graphite furnace atomic absorption (GFAA),
hydride-generation atomic absorption (HGAA) and cold-vapor atomic
absorption (CVAA). Each of these is briefly discussed below in
terms of their advantages and disadvantages.
8
-------
1. ICP's primary advantage is that it allows simultaneous or
rapid sequential determinations of many elements in a short
time. The primary disadvantage of ICP is interference by
background radiation from other elements and the plasma gases.
Although all ICP instruments use high-resolution optics and
background correction to minimize these interferences,
analysis for traces of metals in the presence of a large
excess of a single metal is difficult. Examples would be
traces of metals in a limed (high calcium) waste. ICP and FAA
have comparable detection limits (within a factor, of 4) ,
except that ICP exhibits greater sensitivity for refractory
elements (e.g., aluminum, barium). GFAA, in general, exhibits
lower detection limits than does either ICP or FAA. However,
all these techniques have adequate sensitivity. Detection
limits are improved when ICP-MS is used. In general, ICP-MS
exhibits greater sensitivity than either GFAA or FAA for most
elements. The greatest disadvantage of ICP-MS is isobaric
elemental interferences. Mathematical correction for inter-
fering ions can minimize these interferences.
2. FAA determinations, as opposed to ICP determinations, are
normally completed as single-element analyses and are
relatively free of inter-element spectral interferences.
Either a nitrous-oxide/acetylene or an air/acetylene flame is
used as an energy source for dissociating the aspirated
samples into the free atomic state, making analyte atoms
available for absorption of light. In the analysis of some
elements, the temperature or type of flame used is critical.
If the proper flame and analytical conditions are not used,
chemical and ionization interferences can occur.
3. GFAA replaces the flame with an electrically heated
graphite furnace. This allows gradual heating of the sample
in several stages. Thus, the processes of dissolution,
drying, decomposition of organic and inorganic molecules and
salts, and formation of atoms, which must occur in FAA or ICP
in a few milliseconds, may be allowed to occur over a much
longer time and at a controlled temperature in the furnace.
This allows an experienced analyst to remove unwanted matrix
components by using temperature programming and/or matrix
modifiers. The major advantage of this technique is that it
affords extremely low detection limits. It is the easiest to
perform on relatively clean samples. Because this technique
is so sensitive, interferences can be a real problem with
complex matrices. Finding the optimum combination of
digestion, heating times and temperatures requires an analyst
experienced in the use of a GFAA.
4. HGAA uses a chemical reduction to reduce and separate
arsenic or selenium selectively from a sample digestate. The
technique therefore has the advantage of being able to isolate
these two elements from complex samples that may cause
-------
interferences for other analytical procedures. However,
significant interferences have been reported when any of the
following is present: an easily reduced metal (copper,
silver, mercury); a high concentration (>200 ag/L) of
transition metals; or an oxidizing agent (oxides of nitrogen)
remaining after sample digestion.
5. CVAA uses a chemical reduction to reduce mercury
selectively. The procedure is extremely sensitive but is
subject to interferences from some volatile organics,
chlorine, and sulfur compounds.
10
-------
'EXHIBIT 1.
LIST OF TC CONTAMINANTS AND REGULATORY LEVELS FOUND
IN 40 CFR §261.24
EPA HW No.'
D004
0005
D018
D006
D019
0020
D021
D022
D007
D023
D024
D025
D026
0016
D027
D028
D029
D030
D012
003 1
0032
0033
0034
0008
0013
0009
0014
0035
0036
0037
0038
0010
0011
0039
0015
0040
0041
0042
0017
0043
Contaminant
Arsenic
Barium
Benzene
Cadmi um
Carbon tetrachloride
Chlordane
Chlorobenzene
Chloroform
Chromi um
o-Cresol
m-Cresol
p-Cresol
Cresol
2,4-0
1 , 4-Oi chl orobenzene
1,2-Dichloroethane
1 , 1-Di chl oroethy 1 ene
2,4-Dinitrotoluene
Endrin
Heptachlor (and its epoxide)
Hexachl orobenzene
Hexachl orobutadi ene
Hexachl oroethane
Lead
Lindane
Mercury
Methoxychlor
Methyl ethyl ketone
Nitrobenzene
Pentachlorophenol
Pyridlne
Selenium
Silver
Tet rachl oroethy 1 ene
Toxaphene
Trlchloroethylene
2,4,5-Trichlorophenol
2,4,6-Trichlorophenol
2,4,5-TP (Silvex)
Vinyl chloride
CAS No.1
7440-38-2
7440-39-3
71-43-2
7440-43-9
56-23-5
57-74-9
108-90-7
67-66-3
7440-47-3
95-48-7
108-39-4
106-44-5
--
94-75-7
106-46-7
107-06-2
75-35-4
121-14-2
72-20-8
76-44-8
118-74-1
87-68-3
67-72-1
7439-92-1
58-89-9
7439-97-6
72-43-5
78-93-3
98-95-3
87-86-5
110-86-1
7782-49-2
7440-22-4
127-18-4
8001-35-2
79-01-6
95-95-4
88-06-2
93-72-1
75-01-4
Regulatory Level (mg/L)
5.0
100.0
0.5
1.0
0.5
0.03
100,0
6.0
5.0
'200.0
'200.0
'200.0
'200.0
10.0
7.5
0.5
0,7
30.13
0.02
0.008
'0.13
0.5
3.0
5.0
0,4
0.2
10.0
200,0
2.0
100.0
'5.0
1.0
5.0
0.7
0.5
0.5
400.0
2.0
1,0
0.2
1 Hazardous waste number.
1 Chemical abstracts service number.
3 Quantltation limit is greater than the calculated regulatory level. The quantitation limit therefore
becomes the regulatory level.
' If o-, m-, and p-Cresol concentrations cannot be differentiated, the total cresol (D026)
concentration 1s used. The regulatory level of total cresol is 200 mg/1.
11
-------
EXHIBIT 2. SW-846 METHODS OF ANALYSIS FOR TCLP EXTRACTS
3010
6010
Ba-
Cr-
Ag-
-As
-Cd
-Pb
-Se
7470
Hg
Sample
TCLP
3510
Neutral
8260 3510
Volatile (Acidic
Organics and
Basic)
8081
Pestic-
ides
8270
Semivol-
atile
Organics
8151
Herbic-
ides
12
-------
IV. QUALITY ASSURANCE AND QUALITY CONTROL
To ensure that the analytical data used for the TC
determination are of known and desired quality, all activities
associated with sampling and analysis should be conducted under
strict quality assurance and quality control (QA/QC) procedures.
Prior to initiating a sampling and analysis program,
generators should prepare a detailed quality assurance project plan
(QAPjP) describing the QA/QC steps and controls to be followed. In
addition, a person knowledgeable regarding the QA/QC procedures
should oversee the sampling and analysis effort to insure that all
QAPjP procedures are followed. For more information on preparing
and implementing quality assurance programs, see Chapter One of SW-
846.
In addition, the appropriate use of data generated under the
great range of analytical conditions encountered in waste
characterization requires reliance on the quality control practices
incorporated into the various testing methods. The Agency has, in
many cases, issued approved methods for sampling and analysis
operations that are intended to fulfill regulatory requirements.
However, the mere use of approved methods does not guarantee
accurate results. Inaccuracies can result from many causes,
including unanticipated matrix effects, equipment malfunctions, and
operator error. Therefore, the quality control component of each
method is indispensable. The TCLP and the determinative methods of
SW-846 contain method-specific quality assurance procedures that
should be followed during a TC determination. All of the QA
procedures found in .section 8.0 of the TCLP must be followed during
the determination.
13
-------
V. TCLP DATA EVALUATION
To evaluate the analytical data and determine whether the ash
exhibits the TC, the data evaluation approach below can be
followed. For the statistical formulas and equations (e.g.,
"equation 2a", "equations 3a and 4", etc.), refer to Exhibit 3.
Exhibit 4 contains tabulated values of Student's t distribution.
1. Determine the mean TC concentration (x~) of the fourteen
eight-hour composite samples for each regulated analyte
(equation 2a).
2. Determine the standard deviation (s) of the data employed
to calculate the mean (i.e., the individual composite extract
results) (equations 3a and 4).
3. Determine the upper limit of a two-sided 80 percent
confidence interval, which is equivalent to a 90 percent (one-
sided) confidence interval, for the mean for each analyte
(equation 6). (Note: Exhibit 3 does not include equations
for arcsine or square root transformations. The Agency is
currently revising Chapter Nine of SW-846 whereby arcsine and
square root transformation discussions are being considered
for removal from the chapter. Transformations should only be
used if the data distribution and valid statistical practices
indicate such transformations are warranted. If transfor-
mations are used, methods for obtaining unbiased estimates of
the mean and confidence limits should be employed.)
4. If the 80 percent upper confidence limit is less than the
applicable regulatory threshold for all analytes listed in 40
CFR § 261.24, then the waste (ash) passes the TC. If the 80
percent upper confidence limit is greater than or equal to the
applicable regulatory threshold for any contaminant listed in
Table 1 of 40 CFR § 261.24, then the waste (ash) fails the TC
and is a hazardous waste.
Results from multiple events.may be combined for evaluation
under certain limited conditions. Data sets representing two or
more sampling events can be combined (pooled) into one data set,
and a new confidence interval calculated, only if all of the
following conditions apply:
Sampling data are for the same waste (e.g., for the
bottom ash and not for any other waste (or ash type)
generated by the facility).
Field sampling and laboratory analysis procedures were
the same for all sampling and analysis events (e.g., the
data are from use of the TCLP and the same procedures
14
-------
were followed by the laboratory during testing, including
the same preparative and determinative methods).
There are no other reasons to believe that the waste and
sampling events were different (e.g., the ash has not
changed over time or space due to changes in pollution
control equipment).
If there is some doubt whether the two data sets can be
combined, statistical tests are available for testing the
assumption that the samples were drawn from identical populations.
For example, the "t" test methods to compare population means can
be used if the underlying populations have normal distributions,
and the Wilcoxon Rank Sum Test (also known as the Mann-Whitney U
Test) can be used to test whether the two populations are identical
but not normal. Generators should seek assistance from a statis-
tician prior to combining results from multiple sampling events.
Regarding the treatment of "non-detects", a number of
approaches are available and the appropriate treatment will depend
on characteristics of the data (e.g., what percentage of the data
is reported as less than the detection limit) . Some of these
approaches are described in references 2, 3, and 5 listed in
section VI of this manual. Generators should consult their
appropriate authorized State or, if in an unauthorized state, their
EPA Regional Offices regarding the evaluation of data sets which
include values reported as less than the analyte detection limit.
Regarding identification and handling of "outliers", testing
for outliers should be done only if an observation seems
particularly high or low compared to the rest of the data set. If
an outlier is identified, the result should not be treated as such
until a specific reason for the abnormal measurement can be
determined. Valid reasons may, for example, include:
Contaminated sampling equipment.
Laboratory contamination of the sample.
Errors in transcription of the data values.
Once a specific reason is documented, the result should be excluded
from any further statistical analysis. If a plausible reason
cannot be found, the observation should be treated as a true, but
extreme value, and not be excluded from the data evaluation.
15
-------
EXHIBIT 3. BASIC STATISTICAL TERMINOLOGY APPLICABLE TO SAMPLING
PLANS FOR SOLID WASTES
Terminology
Symbol
Mathematical Equation
(Equation)
• Variable {e.g.,
barium or endrtn)
• Individual measurement
of variable
• Mean of possible measurements
of variable (population mean)
.i-l
with N » number of
possible measurements
(1)
• Mean of measurements generated
by sample (sample man)
Simple random sampling and systematic
random sampli nq
_ . •* with N • number of
x » —— , sample measurements
n
(2a)
Stratified random sampling
with x, > stratum mean and V, - (2b)
£ fit xl fraction of population represented
.... by Stratum k (number of strata [k]
K * ' range from 1 to r)
• Variance of sample
Simple random sampling and
systematic random sampling
n
E xt
i-l
i-l
,) Vn
n - 1
(3a)
Stratified random sampling
r
£ W s '
* *
(c*1 '
with s', » stratum variance (3b)
and ^ " fract1on of population
represent by Stratum k (number of
strata [k] ranges from 1 to r)
(Source: Adoption of Table 9-1 of Chapter Nine, SW-846,
Third Edition as Amended by Updates I, II, IIA, and IIB)
16
-------
EXHIBIT 3.
BASIC STATISTICAL TERMINOLOGY APPLICABLE TO SAMPLING
PLANS FOR SOLID WASTES (Continued)
Terminology
Symbol
Mathematical Equation
(Equation)
» Standard deviation of sample
(4)
» Standard error (also standard
error of mean and standard
deviation of mean) of sample
i/5
(5)
» Confidence interval for
» Regulatory threshold"
CI
RT
with t.JO obtained from
CI « ~x ± t 20 s= , Exhibit 4 for appropriate
degrees of freedom
Defined by EPA
(6)
(7)
» Appropriate number of samples to
collect from a solid waste (financial
constraints not considered)
with A = RT - -x
(8)
• Degrees of freedom
df
df = n - 1
0)
' The upper limit of the CI for /i is compared with the applicable regulatory threshold (RT) to determine if a solid
waste contains the variable (chemical contaminant) of concern at a hazardous level. The contaminant of concern is not
considered to be present in the waste at a hazardous level if the upper limit of the CI is less than the applicable
RT. Otherwise, the opposite conclusion is reached.
(Source: Adoption of Table 9-1 of Chapter Nine, SW-846,
Third Edition as Amended by Updates I, II, HA, and IIB)
17
-------
EXHIBIT 4. TABULATED VALUES OF STUDENT'S "t" FOR EVALUATING
SOLID WASTES
Degrees of
freedom (n-l)a
1
2
3
4
5
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
40
60
120
Tabulated
"t" value"
3.078
1.886
1.638
1.533
1.476
1.440
1.415
1.397
1.393
1.372
1.363
1.356
1.350
1.345
1.341
1.337
1.333
1.330
1.328
1.325
1.323
1.321
1.319
1.318
1.316
1.315
1.314
1.313
1.311
1.310
1.303
1.296
1.289
1.282
aDegrees of freedom (df) are equal to the number of samples (n)
collected from a solid waste less one.
bTabulated "t" values are for a two-tailed confidence Interval
and a probability of 0.20 (the same values are applicable to a one-tailed
confidence Interval and a probability of 0.10).
(Source: Adoption of Table 9-2 of Chapter Nine, SW-846,
Third Edition, as Amended by Updates I, II, IIA, and IIB)
18
-------
VI. REFERENCES
The references listed below represent resources that may be
helpful during the development of a QAPjP and a sampling and
analysis plan. The TCLP, preparative and determinative methods for
analysis of the TCLP extract, and Agency guidance regarding quality
assurance/quality control, the development of sampling and analysis
plans, and data evaluation can all be found in reference number 1
(SW-846). The other listed references provide information
regarding statistical evaluations of data that may prove useful
during a TC determination.
1. USEPA, Test Methods for Evaluating Solid Waste,
Physical/Chemical Methods (SW-846), Third Edition as amended
by Updates I, II, IIA, and IIB, Washington, DC.
2. USEPA, 1989, Statistical Analysis of Ground Water Monitoring
Data at RCRA Facilities, Interim Final Guidance and Draft
Addendum to Interim Final Guidance (July, 1992). EPA Office
of Solid Waste, Washington, DC.
3. Gilbert, R.O., 1987, Statistical Methods for Environmental
Pollution Monitoring, New York: Van Nostrand Reinhold, 320
pp.
4. Helsel, D.R., 1990, "Less Than Obvious, Statistical Treatment
of Data Below the Detection Limit", in Environ. Sci. Technol,
24(12): 1766-1774.
19
-------
-------
APPENDIX:
DEFINITIONS OF TERMS USED IN THE GUIDANCE
Accuracy; The closeness of agreement between an observed value and
a true or accepted reference value.
Bottom Ash; Coarse, relatively dense (40-70 lbs/ft3 dry) solid
material that remains on a hearth or falls off the furnace grate
after thermal processing is complete.
Composite sampling; Sample collection whereby a number of random
samples are initially collected from a waste and then combined into
a single sample, which is then analyzed for the contaminants of
concern. Composite samples are composed of several distinct
subsamples (or grab samples), and are often collected when it is
not economically feasible to analyze a large number of individual
samples.
Confidence interval; The numerical interval constructed around a
point estimate of a population parameter, combined with a
probability statement (the confidence coefficient) linking the
interval to the population's true parameter value.
Data Quality Objectives; Qualitative and quantitative statements
about the data and of the overall level of uncertainty that a
decision-maker is willing to accept in results derived from data.
DQOs should take into account both sampling considerations and
analytical protocols.
Detection limit; The lowest concentration or amount of a target
analyte that can be determined by a single measurement to be
different from zero or background level at a defined level of
probability. The detection limit is generally recognized to be
sample matrix and measurement method dependent.
Disposal; The discharge, deposit, injection, dumping, spilling,
leaking, or placing of any solid or hazardous waste into or on any
land or water.
EPA Hazardous Waste No.; The number assigned by EPA to each
hazardous waste listed in part 261, subpart D of 40 CFR and to each
characteristic identified in part 261, subpart C of 40 CFR.
EPA Region; The states and territories found in any one of ten
regions identified by EPA.
Fly Ash; Light (usually less than 20 lbs/ft dry weight basis)
flue gas-entrainable particle material carried off the furnace
grate during combustion by the updrafting of underfire air.
Depending on the facility design, these flue gas entrained
particles, volatilized elements/compounds, and gaseous fractions
will be partially collected in post combustion fly ash hoppers
-------
mostly in solid particle form, with some smaller gaseous fractions
entrapped in gaseous form.
Generator; Any person, by site, whose act or process produces
waste.
Hazardous Waste; A solid waste, as defined in 40 CFR S 261.2 that
(1) is not excluded from regulation as a hazardous waste under 40
CFR § 261.4(b), and (2) meets any of the criteria under 40 CFR S
261.3(2) (e.g., exhibits one of the characteristics of a hazardous
wastes or is listed as a hazardous waste). A hazardous waste is a
material that poses a substantial present or potential hazard to
human health or living organisms due to its lethal, non-degradable
or persistent nature or because it may cause or tend to cause
detrimental cumulative effects.
Heterogeneous: Consisting of dissimilar or diverse ingredients or
constituents.
Homogeneous; Of uniform structure or composition throughout.
Management or Waste Management; The systematic control of the
collection, source separation, storage, transportation, processing,
treatment, recovery, and disposal of waste.
Operator; The person responsible for the overall operation of a
facility.
Owner; The person who owns a facility or part of a facility.
Pile; Any non-containerized accumulation of solid, nonflowing
waste that is used for treatment or storage.
Precision; The agreement among a set of replicate measurements
without assumption of knowledge of the true value. Precision is
estimated by means of duplicate or replicate analyses.
Project; Single or multiple data collection activities that are
related through the same planning sequence.
Quality Assurance; The process for ensuring that all data and the
decisions based on these data are technically sound, statistically
valid, and properly documented.
Quality Assurance Project Plan; An orderly assemblage of detailed
procedures designed to produce data of sufficient quality to meet
the data quality objectives for a specific data collection
activity.
Quality Control; Procedures employed to measure the degree to
which the quality assurance objectives are met.
Representative Sample; A sample of a universe or whole which (1)
has the properties and chemical composition of the population from
-------
which it was collected, and (2) has them in the same average
proportions found in the universe or whole.
Solid Waste; Any discarded material that is not excluded by 40 CFR
§ 261.4(a) or that is not excluded by a variance granted under 40
CFR §§ 260.30 and 260.31, and as defined by 40 CFR § 261.2.
Storage: The holding of waste for a temporary period, at the end
of which the waste is treated, disposed of, or stored elsewhere.
Transport Vehicle; A motor vehicle or rail car used for the
transportation of cargo (including waste) by any mode. Each cargo-
carrying body (trailer, freight car, etc.) is a separate transport
vehicle.
Treatment; Any method, technique, or process, including
neutralization, designed to change the physical, chemical, or
biological character or composition of any waste, or so as to
recover energy or material resources from the waste, or so as to
render such waste non-hazardous or less hazardous; safer to
transport, store, or dispose of; or amenable for recovery, amenable
for storage, or reduced in volume.
-------
------- |