United States Solid Waste and
Environmental Protection Emergency Response EPA530-R-94-020
Agency (5305) May 1994
vvEPA
Sampling and Analysis
Of Municipal Refuse
Incinerator Ash
DRAFT
Recycled/Recyclable
"). ~O Printed on paper that contains at
7").
\_]f_/ least 50% post-consumer recycled fiber
-------
INTRODUCTION
DRAFT
The purpose of this document is to assist generators of ash
from municipal incinerators in designing a plan to determine
whether any constituent in the ash exceeds the levels specified
in EPA's Toxicity Characteristic (TC).
Determining whether a waste passes or fails the TC requires
reliable information on the chemical properties of the waste.
Several factors contribute to this reliability. Accuracy can be
achieved by collecting and testing enough unbiased samples to
determine the average properties of the waste and its
variability. Bias can be prevented by incorporating randomness
into the sampling collection process, and precision can be
obtained by selecting an appropriate sampling interval and number
of samples.
This document provides general guidelines on approaches for
achieving these goals. Users are reminded, however, that this
guidance must be adapted to the particular facility under
consideration to assure accuracy in determining whether wastes
pass or fail the TC. This document is based on limited
information on municipal waste resource recovery incinerators.
The sampling and testing described represents the minimum that
the Agency considers as being appropriate. As the Agency obtains
more information on these facilities, this document may be
revised. In the meantime, facility owners should use this
document as a general guidance in developing a facility-specific
plan.
This document contains the following sections:
1. Sampling: Describes how to design a sampling plan,
including type and frequency of sampling, as well as
location.
2. Analysis: Describes the specific procedures from "Test
Methods for Evaluating Solid Waste" (SW-846) for the
species listed in 40 CFR 261.24.
3. Strategy for Evaluating Samples: Describes criteria
for evaluating data to determine which wastes pass or
fail the TC and the frequency of recharacterization.
4. Quality Assurance and Quality Control: References
procedures to follow to insure the quality of the
sampling and analysis data.
ENVIRONMENTAL
PROTECTION
AGENCY
DALLAS, TEXAS
LIBRARY
-------
Di
SAMPLING
The objective of sampling and analysis is to assess the
properties of the waste being generated. Each residual waste
stream that is stored, transported, or disposed of as a separate
unit is considered to be a discrete waste and must be evaluated
accordingly. You must determine whether the ash from your
facility passes or fails the TC.
The sampling plan described in this document represents the
Agency' s current thoughts on what constitutes the minimum amount
of sampling needed to determine the average property of the ash
from the combustion of municipal refuse. As noted later,
facility operators are responsible for evaluating the waste to
ascertain its variability over time. Currently, the Agency
believes that, at a minimum, seasonal rechecking of residual
properties is necessary to insure that the waste has not
substantially changed.
The sampling strategy for evaluating the ash is based on the
assumptions that 1) the waste feed prior to incineration is not
segregated by type of generator (e.g., household, commercial,
industrial) and 2) the ash generated is not separated by size
during storage or disposal. The strategy is designed to
determine the average concentration of TC analytes. If the above
assumptions are not valid, then a facility-specific sampling and
analysis program designed by knowledgeable personnel should be
employed.
For characterizing ash from facilities for which all of the
above assumptions apply, the sampling procedure is as follows:
1. Determine the most convenient location for sampling. In
situations where the sampling can be conducted either from
transport vehicles or from the waste conveyance device, the
Agency recommends sampling from the transport vehicle (e.g.,
dump truck, barge) .
2. 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 width of the belt conveyor, drag chain flight, or
vibrating conveyor. A draft practice for sampling
unconsolidated waste materials from trucks is currently
under development by ASTM Committee D-34. Pending
completion of the standard, this draft practice may be used
for guidance on appropriate procedures for sampling ash from
trucks .
3. If a conveyor is to be the sample location, collect the
entire width 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 further processing.
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 waste characterization, samples should be
collected each day for a minimum of one week's operation.
6. Each composite is to be passed over a 2 inch screen.
Material passing the 2 inch screen is to be set aside.
Material >2 inches is to be subjected to repeated blows with
a five (5) pound sledge hammer dropped from one foot above
the >2 inch size pieces. If a piece does not break after
eing subjected to three (3) blows of the hammer, it is to
be weighed and discarded. Material that breaks is then
reduced in size to pass the 2 inch screen and recombined
with the original <2 inch sample. As an alternative,
facilities may reduce all material in a composite to pass a
3/8 inch (9.5 mm) screen and not discard any components of
the composite.
7. Each composite should be crushed to pass a 3/8 inch (9.5
mm) screen, and riffled or coned and quartered to obtain a
1000 gram aliquot. Properly label the sample and store it
in a clean, dry, cool secure area. For further details, see
ASTM standard D346.
8. Analyze the composite samples for the properties of
interest.
-------
RAi
i.r
ANALYSIS
In order to determine whether the ash exhibits the TC, 100
gram or larger size aliquots of each eight hour composite shall
be tested using Method 1311 and the extract analyzed for the
species listed in 40 CFR 261.24. All testing shall be performed
following the specific procedures described in "Test Methods for
Evaluating Solid Waste" (SW-846).
Briefly, Method 1311 (TCLP) consists of mixing 100 grams of
sample with an acetic acid extraction fluid in a liquid-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 analyzed for the species found in Table 1 of 40
CFR 261.24 (attached).
Given the low probability of their occurrence in incinerator
ash, it is recor -ended that the organic compounds be analyzed for
only in the first 1 or 2 extracts. Only if one or more organic
compounds are detected should the remaining extracts be routinely
analyzed for the organics.
Prior to analysis of the extracts using atomic absorption
spectrometry (AA), inductively coupled plasma spectroscopy (ICP),
or gas chromatography (GC), the extracts must be prepared using
the appropriate methods. Recommended SW-846 methods are shown in
Figure 2-3B of Chapter Two of SW-846 (attached).
SW-846 contains several analytical techniques for trace
metal determinations: ICP, 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 advantages, disadvantages, and cautions for analysis of
municipal incinerator wastes and TCLP extracts.
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 for determining whether the wastes pass or fail
the TC.
-------
2. FAA determinations, as opposed to ICP determinations,
are normally completed as single-element analyses and are
relatively free of interelement 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. The furnace allows for 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 -' i a few milliseconds, may be allowed to
occur over a much longer time and at a controlled
temperature in the furnace. This allows an experiences
analyst to remove unwanted matrix modifiers. The major
advantage of this techniques 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, and matrix modifiers can be
a challenge.
4. HGAA uses a chemical reduction to reduce and separate
arsenic or selenium selectively from a sample digest. The
techniques 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 mg/L)
of a transition metal; 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 and
sulfur compounds.
-------
f nvtoonuwntol Protection Agency
§261.30
quantity sufficient to present a danger
to human health or the environment.
(5) It is a cyanide or sulflde bearing
waste which, when exposed to pH con-
ditions between 2 and 12.5, can generate
toxic gases, vapors or fumes in a quan-
tity sufficient to present a danger to
human health or the environment.
(6) It is capable of detonation or ex-
plosive reaction if it is subjected to a
strong initiating source or if heated
under confinement.
(7) It is readily capable of detonation
or explosive decomposition or reaction
at standard temperature and pressure.
(8) It is a forbidden explosive as de-
fined in 49 CFR 173.51, or a Class A ex-
plosive as defined in 49 CFR 173.53 or a
Class B explosive as defined in 49 CFR
173.88.
(b) A solid waste that exhibits the
characteristic of reactivity has the
EPA Hazardous Waste Number of D008.
[45 FR 33119, May 19, I960, aa amended at 55
FR 22684, June 1. 1990]
$261.24 Toxicity characteristic.
(a) A solid waste exhibits the char-
acteristic of toxicity if, using the test
methods described in appendix n or
equivalent methods approved by the
Administrator under the procedures set
forth in §§260.20 and 260.21. the extract
from a representative sample of the
waste contains any of the contami-
nants listed in table 1 at the concentra-
tion equal to or greater than the re-
spective value given in that table.
Where the waste contains less than 0.5
percent filterable solids, the waste it-
self, after filtering using the methodol-
ogy outlined in appendix n, is consid-
ered to be the extract for the purpose
of this section.
(b) A solid waste that exhibits the
characteristic of toxicity has the EPA
Hazardous Waste Number specified in
Table I which corresponds to the toxic
contaminant causing it to be hazard-
ous.
TABLE 1—MAXIMUM CONCENTRATION OF CON-
TAMINANTS FOR THE Toxicmr CHARACTERIS-
TIC
TABLE 1—MAXIMUM CONCENTRATION OF CON-
TAMH4ANTS FOR THE TOXIOTY CHARACTERIS-
TIC—Continued
EPAHW
No.'
0004
Contaminant
Arsanc _
CAS No.'
7440-38-2
Raov
latory
Laval
0006
0018
0006
DO20
0021
0022
0007
0023
0024
0023
0026
0016
0027
0028
0030
0012
0031
0033
0034
0008
0013
0009
0014
0035
0036
0038
0010
0011
ryvM
001 5
0017
0043
Contamnant
Barium
Banzana ,,,,.,.,,... .__
Cadmium
Chtordana .. .
Chtorebanzana
Chromium _„_....._
O*€faBOl r-T-lrlll
m-C/aaol ,
p-Crwol
OMOI
2.4-O .._........—...— ~.
1,4-Oicnlorobaruana -..
2-Oichloroathana
2 4-Oinilrotaluana
Endrin
HaptacNor (and to ag-
enda*.
HayacNorobutadtena ».
Haxachtaroatiana
I*!*!
Lindana .
Marcury
Mathoncninr
Mawyl alnyl katorta »«..
NMrobanzana »....«...».
aniiacnHrapnanoi —
Pyridina
Sttanium
Sivar
Toxapnana
z,4.!>-TiKnioropnanoi _.
2,4,5-TP (Sifeax)
Vinyl chloride
CASNtxt
7440-39-3
71-43-2
7440-43-9
56 23-6
57-74-9
108-90-7
67-66-3
7440-47-3
96-48-7
106-39-4
106-44-6
94-75-7
106-46-7
107-06-2
Tg *K A
121-14-2
72-20-8
76-44-8
87-68-3
67-72-1
7439-92-1
58-89-9
7439-97-6
7"?— J*W»
78-93-3
96-95-3
87-oft-o
110-86-1
7782-49-2
7440-22-4
1 97—1 ° -*
8001-35-2
35- 0O 4
93-72-1
75-01-4
Ragw
latory
Laval
5.0
1.0
5.0
ft 7
05
9 ft
1.0
0.2
1 Hazardout waste numbar.
z Chemical at»liaU» sarvica numbar.
3Quan«alion limrt a graatar than tha calcUatad ragulatory
(aval. Tha quamWabon limit tharatora bacomas tha
4 If o-, m-t and p-Crasol concantrabons cannot ba drftoran-
tiatad, tha totaJ enact (0026) i
itratan a usad. Tha ragu-
latory (aval of total craaoi « 200 mgA.
[55 FR 11862, Mar. 29, 1990, as amended at 55
FR 22684, June 1, 1990; 55 FR 26987, June 29,
1990]
Subport D—Lists of Hazardous
Wastes
§261.30 General.
(a) A solid waste is a hazardous waste
if it is listed in this subpart, unless it
has been excluded from this list under
§§260.20 and 260.22.
(b) The Administrator will indicate
his basis for listing the classes or types
-------
FIGURE 2-3B.
RECOMMENDED SW-846 METHODS OF ANALYSIS FOR TCLP LEACHATES
I Sample I
TCLP
1
3010
1
7470
Hg
3510
Neutral
1
8240
8260
Volatile
Orqanics
1
3510
(Acidic
and
Basic)
1
8150
8151
Herbic-
ides
6010
Ba -
Cr -
Ag -
- As
- Cd
- Pb
- Se
TWO - 49
Revision 2
November 1992
-------
D!
QUALITY ASSURANCE AND QUALITY CONTROL
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 and sampling procedures. The
Agency has, in many cases, issued approved methods for sampling
and analysis operations fulfilling 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.
To ensure that the test data used to evaluate the ash are of
known and appropriate quality, all activities associated with
sampling and measurement should be conducted under strict quality
control. Prior to initiating a sampling/testing prograr
facilities should prepare a detailed quality assurance project
plan describing the steps and controls to be followed. In
addition, a knowledgeable person should be appointed to oversee
the program to insure that all procedures are followed. For more
information on preparing and implementing quality assurance
programs, see Chapter One of the Third Edition of SW-846.
8
-------
DRAFT
DATA EVALUATION
In evaluating the data to determine whether or not the ash
passes or fails the TC, use the following approach (see attached
Tables 9-1 and 9-2 of Chapter Nine of SW-846 for statistical
formulas to use in the following calculations):
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 bound of the 90 percent (one-sided)
confidence interval for the mean for each analyte (equation
6) .
4. If the upper bound of the interval is below the
applicable regulatory threshold for all analytes listed in
40 CFR 261.24, then the waste passes the TC. If the upper
bound of the interval is above the applicable regulatory
threshold for any analyte listed in 40 CFR 261.24, then the
waste fails the TC.
5. Given the variability inherent in the ash generation
process, facilities should recharacterize their ash at least
four (4) times each year to insure that accurate
characterization is obtained. In determining how often to
recharacterize the ash, the generator should consider all
facility-specific and external factors that could cause the
ash properties to vary. These factors include changes in
the composition of the waste (e.g., new types of industries
moving into the area, institution of recycling programs in
the collection area, seasonal changes affecting population
or waste composition); changes in plant design (e.g.,
addition of dry scrubber, addition of quench tank); and
significant changes in plant operating conditions (e.g.,
increase in combustion time or temperature, change in lime
utilization rate).
REFERENCES
1. Test Methods for Evaluating Solid Waste, Physical/Chemical
Methods, (SW-846), Third Edition, US EPA, Washington, DC, August,
1993.
-------
TABLE 9-1. BASIC STATISTICAL TERMINOLOGY APPLICABLE TO SAMPLING PLANS FOR SOLID WASTES
Terminology
Symbol
Mathematical equation
(Equation)
Variable (e.g., barium
or endrin)
Individual measurement
of variable
Mean of all possible
measurements of variable
(population mean)
Mean of measurements
generated by sample
(sample mean)
N
E
with N « number of
possible measurements
Simple random sampling and
systematic random sampling
n
E x
1=1
n
1
, with n = number of
sample measurements
(1)
(2a)
Stratified random sampling
r
E
with X|< = stratum (2b)
mean and W|< - frac-
tion of population
represented by Stratum
k (number of strata
[k] range from 1 to r)
Variance of sample
Simple random sampling and
systematic random sampling
n „ n
E x
- (E x,)2/n
1=1 1
n - 1
(3a)
Stratified random sampling
*% i o *i
E
k-1
with sr = stratum
variance and WV =
fraction of population
represent by Stratum k
(number of strata [k]
ranges from 1 to r)
(3b)
NINE - 2
Revision 0
Date September 1986
10
-------
TABLE 9-1. (Continued)
Terminology
Symbol
Mathematical equation
(Equation)
Standard deviation of
sample
Standard error
(also standard error
of mean and standard
deviation of mean)
of sample
sx
= 17
s L.
x Jn
(4)
(5)
Confidence Interval
for /»a
CI
« - * ± t.20 s7, with t.go
"led
obtained from
Table 2 for
appropriate
degrees of freedom
(6)
Regulatory threshold3
RT
Defined by EPA (e.g., 100 ppm for (7)
barium 1n elutriate of EP toxldty)
Appropriate number of
samples to collect from
a solid waste (financial
constraints not considered)
RT - x
(8)
• Degrees of freedom
df
df = n - 1
(9)
aThe upper limit of the CI for /i 1s compared with the applicable regulatory
threshold (RT) to determine 1f a solid waste contains the variable (chemical
contaminant) of concern at a hazardous level. The contaminant of concern 1s not
considered to be present 1n the waste at a hazardous level 1f the upper limit of the CI
1s less than the applicable RT. Otherwise, the opposite conclusion 1s reached.
NINE - 3
Revision 0
Date September 1986
11
-------
TABLE 9-2. 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" valueb
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.
tabulated "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).
NINE - 4
Revision 0
Date September 1986
12
-------
D:
APPENDIX A
DEFINITIONS OF TERMS APPROPRIATE TO
ASH GENERATION/CHARACTERIZATION
MSW: Municipal Solid Waste including domestic solid wastes,
commercial solid wastes, light demolition and renovation waste
and non-hazardous industrial wastes usually removed from
dumpsters.
Bottom Ash: Coarse, relatively dense (40-70 lbs/ft3 dry) ash
remaining on the furnace grate after controlled combustion of MSW
(usually less than 2" OD).
Clinkers: Large (greater than 2" OD) glass-like fused
noncombustible materials which form during the high temperature
combustion on the furnace grate and are discharged from the
furnace in this fused, agglomerate form.
Slagging: Usually larger (greater than 6" OD) fused and layered
silicates and semi-volatile metals which build up on the
refractory and convective surfaces of the furnace internal
fireside walls and either break off during furnace operation or
are manually removed during furnace and boiler cleaning.
Riddlings: Sand-like material which passes down through the
furnace grate system during normal facility operation. This
material is quite often silica-based and noncombustible.
Non-Combustibles: Variable sized materials which are only
partially volatile at their surfaces and maintain a majority of
their structural integrity through the combustion process.
Examples are car wheels, springs, machine parts, frames, rocks,
and bottle tops.
Soot Blows: Very fine (10 micron +) agglomerated particles and
particle-entrapped gases which collect by condensation,
impingement and/or attachment onto boiler tube faces, usually in
boiler regions where gas velocities suddenly decrease or form
eddy turbulent flow. These layered materials are removed during
boiler maintenance usually by high pressure hydraulics or steam
spargers or by vibratory, pneumatic or impact devices.
Fly Ash: Light (usually less than 20 lbs/ft3 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 flyash hoppers
mostly in solid particle form, with some smaller gaseous
fractions entrapped in gaseous form.
Scrubber By-Product: Unreacted lime and lime-gas reaction
products resulting from the injection of lime prior to an
ESP/Cyclone/FF; used primarily for control of acid gases. By-
13
-------
u , <... .." .:
products include CaSO3, CaSO4, CaCl2, CaF2, unreacted lime (CaO or
Ca(OH)2) and a small amount of solid-to-solid phase reaction
compounds.
The most common collection points for fly ashes and soot
blows are as follows:
1. Superheater hopper: Similar to boiler hopper and many
times discharged directly back onto the furnace grate.
Collects flyash in the superheater section of the boiler.
2. Boiler hopper: Usually collects heavier particles in
the main boiler tubed sections; often discharged onto
bottom-ash.
3. Economizer hopper: Collects flyash in the economizer
sec xon of the boiler. This is sometimes collected separate
from the bottom ash within the plant process.
4. ESP/cvclone/FF hoppers: Three possible particulate
collection devices, Electrostatic Precipitators (ESP),
cyclone collection devices and fabric filters (FF) are used
primarily to reduce fine particle emissions to the
atmosphere. These devices usually account for 75% of the
flyash weight basis collected by all hoppers, with the
boiler, superheater and economizer accounting for the
remaining 25%.
14
------- |