oEPA
United States
Environmental Protection
Agency
Office of
Solid Waste and
Emergency Response
DIRECTIVE NUMBERr 9445.02(85)
TITLE: Notes on RCRA Methods and QA Activities
APPROVAL DATE: 4-23-85
EFFECTIVE DATE: *-23-85
ORIGINATING OFFICE: office of solid waste
O FINAL
O DRAFT
STATUS.
[ 1
A- Pending OMB approval
I ] B- Pending AA-OSWER approval
I } C- For review &/or consent
I ] D- In development or circulating
REFERENCE (other documents):
headquarters
OSWER OSWER OSWER
VE DIRECTIVE DIRECTIVE Dl
-------
PART 261 - APPENDICES DOC: 9445.02(85)
Key Words: Test Methods
Regulations:
Subject: Notes on RCRA Methods and QA Activities
Addressee: N/A
Originator: David Friedman, Manager, Methods Program
Source Doc: #9445.02(85)
Date: 4-23-85
Summary:
The memo discusses new testing methods and reevaluates existing methods.
-------
y445.02 (85)
JPR23
MEMORANDUM *6
f - " / '
DATES April 1985 '
SOBJBCTt Notes on RCRA Methods and QA Activities
FROMs David Friedman, Manager
Methods Program (WH-562B)
TOi Addreasees
Thank you again for your response to my previous RCRA Methods
and QA Activities nemos* In this memo we will inform you about
the following recent initiatives in the Agencyi
o Public Meetings on Delisting
o Intra-agency Work Group on Update to SW-846
o Development of New Testing Methods and
Reevaluation of Existing Methods.
We appreciate your convents and suggestions.
Public Meetings on Delisting
Approximately 200 persons attended public meetings in washing*
ton D.C. and Dallas, Texas, organized by the Delisting Program.
The meetings were conducted to describe in detail how future
delisting petitions will be affected by the Hazardous Waste
Amendments of 1984 and to provide instruction to petitioners on
the preparation of supplemental material that should be included
in future delisting petitions. Presentations were made on the
reauthorization of RCRA, the general petition review process,
information resources and special procedures that will apply to
petitions concerning wastes from petroleum refineries and multiple
waste treatment facilities. A discussion session followed the
presentations in which attendees were given the opportunity to
question the delisting staff about their particular concerns.
-------
Intra-aqency Hork Group on Update to SW-846
KM part of tile process of finalising the October 1, 1984
proposal to update SW-846, "Test Methods for Evaluating Solid
Waste,* an intra-agency work group ia in tha process of considering
the comments received by tbe Agency in raaponaa to tha proposed
ehangaa and ia rewriting soaa of tba methods to reflect these
oosuaants and auggaationa. OSW expects to complete this process
by the end of this summer, with Federal Register publication
anticipated in the Spring of 1986.
Development of Haw Teatinq Methods and Reevaluetion of existing
Methods .. '.
As part of ita efforta to develop new testing methods and to
determine the accuracy* precision and sensitivity of the existing
methods, tha SPA Office of Research and Development has completed
a nuaber of test methods studies including!
o Method 9022 and Interim Method 450.1 Determining total
organic halide (TQX) in ground water and waate oil
o Method 8030 Determining acrolein, acrylonitrile and
acetonitrila in ground water, solids and organic liquid
matrices
o Method 8090 Procedure for analysis of nitroaromatics
and cyclic ketonea in ground water* liquid waste and
solid aanple matrices
o Method 8280 Procedure for analysis of hazardous wastes
containing tetra-, penta- and hexa-chlorinated dibenao-
dioxina and -furana . . .
o Application of a gas chromatography/Fourier transform
infrared protocol to the determination of semivolatile
organic compounds in waste water, soils, sediments and
solid wastes
o Analytical methods for compounds that do not gas chromato
graph.
The following sections describe these recent research
developments.
-2-
-------
Method fOtl and Interim Method
Varioms methede .tor determining total organic halide (TOX)
in samples of ground water and waste oil were evaluated by EH8L-
CZH (Tbomaa Frees ley, S13-4ie-74M)« Ol three inorganic halide
species generation approaches and Xhree Inorganic halide determin-
ative techniques evaluated for ground water aaalyaee, one combined
approach uaiag Sehoeniger flaek oxidation with oolorimetrie
chloride determination was chosen for laboratory validation and
method detection limit stmdies* Ground water samples were also
analysed for TO* using moutrft* activation analysis (Hotaod 9022).
Th« roaulta of this valuation indioata that Nothod §022 ia
an appropriate taohniqua for TOK analyaia and of fa ra tha additional
advantaga of providing individual ohlorina and broaina valuaa for
tlM aaHplav at lavola aqual to TOO detection liaita aoaievable
with the ierocottlomotvie deteml native technique which haa a
reliable liait of sensitivity at 5 ug/L ( Inter ia Hethed 4SO»1J.
the achoeniger flaad/oo lor i»e trie chloride and neutron activation
analysis Methods were found to give TOI results for spiked ground
waters that were comparable to those obtained using the ierocou-
loaetric method* However* the Schoeniger f lask/eolorimstrie .
halide method was useful only for halide levels above 0.2 mg/L
due to the high halide background levels encountered in the
reagents* .., /-. .- . . '>... . -.:... i :*.-».;- ..
.,,-...*..--..«.'
A method for analyais of oil samples that uses the sodium
biphenyl reagent and a color im* trie chloride method was found to
be unsatisfactory for the analysis of various oils spiked with
PCBs due to interferences in the color i me trie determinative step
that resulted in recoveries greater than 100 percent* further
valuation of the sodium biphenyl reagent using alternative
determinative techniques, .such as miorocoulcmetrio detection or
an additional clean-up step for the extract prior to the, color i-
metric determinative step, will have to be performed to establish
the uaefulaeaa of the technique at halide levels in the range of
20 to 350 mg/g»-;!. :.:..>' r L^-.V-. i .;--...., .-, T i .».: -. ,...:>; *;;;-. '...-: '. .
i ^Cst.'".'; ~ -.-.- r.r,:' *>;-.:;. rf..:^l:'>.- '.'.- ^f*«. ci (; . . -,-.r-r*^:. '-.- -
;'- Method 8030 ' ,."j ^. > «- '! c i .-; f.fc^^-t.r, -_->/>=.>« t"-^ *»-...:. .--..
.;c- ^Method MM -was modified uaing revised chroma tographic
conditions for determining acrolein, acrylonitrile and acetonitrlla
in ground water, -solids and organic liquid matrices* --In order to
evaluate these ^modifications, method validation 'studies were
conducted. on eaoh< of the, four sample. preparation techniques
included in Method §030, namely (1) heated purge-and-trapt (2)
jpolyethylene giycol (PBa> extraction followed by heated purge-arid-
trapt (3) direct liquid inject ion t and (4 > .manual heated headspace,
Bach method validation involved the determination of the method
detection limit -(HDL) .and seven replicate analyses of one or two
matrices unspiked and spiked at two different levels. Ground
water waa analysed by the heated purge-and-trap method* a solid
-------
waate wae analysed by; .*» FflQ/boated purge-end-trap method* and '
liquid organic wae te waa analysed by direct liquid injection and
by tha haatad headapaoa mathod* ^ . - ^ -j-. - - > -3 - ,
..>> TIM heated purge-and-trap, method gave excellent
(tS fco M p*ro*ot) and praoisioii for tJ* throo «aalyt*« in ground
«Mt«r« HM MG/h*«t«d purgo-and-tvap Method O«T« good r*cor«ri«s
C7,« ta t pareoftt) tor aoetoaitril* and aeryloaltril* to th«
olid waata but poor racovariaa for aerolaia (10 paroant).
Problaaa raaultiag froa raaidoal aaounta of F80 iadieatad that
additional Bodifieationa of this aathod ara noadad* Tha direct
liquid iajaetio* taehniqua gava axeallaat raaulta for tba datar-
1 nation of all thraa analyt«a (« to 111 paroant raeovariaa) in
tha liquid organic vaatai nowavor* lato a luting aatarial may -
praaant aariooa problava ia aoaa eaaoa« Tha anual haatad haadapaee
othod gava diatiaotly Jlaaa aoourata and laaa praeiao raaulta
than tha PM/purga-and-trap aathod for tha determination of :
acatonitrila and aorylonitrila ia tha aolid waste sample. Ac role In
waa not raoorarad at ail by tha mat hod duo* apparently, to decoa-
Doeitioa during, tha ona*^iour equilibration at M*C« tha haadapaee
method gave extremely erratic raaulta for tha analyaia of tha
organic liquid vaata and uaar therefore* eonaidorad completely
unaatiafactory for auoh aamplea. , < v- :
Thia atudy waa conducted by tha msL-CIff laboratory (James
Longbottom, 5U-484-730»).
Method 8090
..» '.'* .?,
Another atudy recently completed by tha 913L-CXM laboratory
(James Longbottorn, 519-684-7308) waa a single laboratory evaluation
of Method 8090* which ia a procedure for the analyaia of nitro-
aromatiea and cyclic ketones in ground water, liquid waste and
aolid aample matrices. The method spike recoveries in authentic
matrices at two challenge concentrations were determined* Authentic
matrices ware obtained from induatrial sources with manufacturing
processes that, might raquira tha use of thla method for monitoring,
and aa auch these authentic samples provided the analytical
conditions and background intarferenta that might be- expected in
actual implementation of thia method. . :
The method detection limit (HDL) in reagent water for all the
analytaa ranged from 1.2 ug/L to a low of 0*26 ug/L. Tha experi-
mental method detection limits ware lower than the published
method detections in Method 8090 for 2,4-dinitrotoluene, 2,6-di-
nitrotoluene, nitrobenzene and iacphorona. Tha. recoveries for
all the analytea in. reagent water were greater than 75 percent*
~; The recoveriea for the majority of the analytaa in tha three
matrices (ground water* nonaquaoua liquid waste, and aolid waste)
were generally good* . All the analytea gave good recoveries
(greater than 70 percent) and good precision (relative standard
-------
deviation lass than 12 percent) in the ground water matrix.
2,4-Dinitrotoluene, 2,6-dinitrotoluene, 1,3-dinltrobenaane and
1,4-naphthoquinone gave good recoveries (greater than 70 percent)
in the nonaqueous liquid waste, poor recoveriee were observed
Cor nitrobenzene and isophorone in the nonaqueous liquid waste.
For the solid vasts matrix, l,4-napht&oquinone gave the worst
recovery (SO psresnt) of the six analytes, while the remaining
analytes gave) recoveries ranging from 0 to 70 percent at the
high concentration level and greater than 80 percent at the low
concentration level.
4 «;
Method 82SO ' ,
In 1983, EPA proposed a ruling affecting disposal of hazardous
wastes containing tetra-, penta-, and hexa-chlorinated dibenso-
dioxins and -furana. As a result, the 3*81. ~LV laboratory initiated
a validation study of the method proposed to detect these substances.
Th* RCRA Method 8280 consists of four major sectionss (1)
extraction of the analytes from the environmental samplei (2)
"open* chromatographic clean-up with alumina using methylene
chloride/hexane eluentj (3) UPLC clean-up* and (4) analysis by
capillary column gas chromatography/low resolution mass spectronetry
(HRGC/LRttS), In order to test Method 8280 efficiently and to
develop appropriate modifications with minimal effort, each
section of the methodology was tested separately. Initial tests
were performed on a simple (pottery clay soil) sample matrix and
upon standard solutions, necessarily, the first step evaluated
was the measurement technique. Both GC/M8 and QC/BC (electron
capture detection) were tested using guidelines from the published
RCRA method. Because the analytes could not be measured at the
published mass-to-charge ratio (m/s) values by GC/MS, these
values were corrected prior to further work. .!
In summary, the following revisions were made to the published
methodi (1) correct m/s values were substituted for those in the
published method so that the mass spectrometer could detect
analytes and standards introduced via the interfaced gas chromato-
graphf (2) the extract clean-up with an alumina column was revised
so that all desired analytes eluted in a single fraction, with the
bulk of the analytical interferences removedi (3) the method wan
revised so that wet samples could be accomodatedt and (4) HPLC pro-
cedures that could be performed effectively and be reproduced con-
sistently were developed* The extraction revision allows accaraoda-
tion of wet samples and improves recovery of spiked analytes in
dry soil samples/ . . ....
The RCRA method with revisions discussed above was subjected
to performance tests that includeds (1) analysis of reference
materials containing 2,3,7,8-TCOO and Interferencesy and (2)
precision and accuracy determinations on samples having known
composition through spiking the sample matrix at this laboratory.
-5-
-------
Two independent teams of analysts investigated the performance
of the revised method asing precision and accuracy determinations
(standard deviation of results) and by recovery of epiked analytea
and iaotopically labeled standards. Effects of experimental
parameters, such as OC column type (coating) and alumina activation
level were also determined.
After incorporating necessary revisions, satisfactory method
performance has been demonstrated oh soil-type samples* ,Much
precision and accuracy data obtained/to date were based on GC/lC
determinations. Comparisons to the precision and accuracy attainable
by utOC/ums is underway. Performance of the method on relatively
ocmplex. matrices, such as sludges, still bottoms and fly ash was
determined. " - ' " ' ^ -..: \ ', ' , '-.-. .-....'
-'. .,-. ' ''-- "'" - ' -. - . -". . """"* "' " ' :--. »..' .''
Although the method was-found suitable Cor soils, fly. ash
and other relatively-clean matrices* serious interferences were
obtained during the analysis of still bottoms. Additional clean-up
steps are now being studied. A copy of the revised method will be
attached to the July memo (Ron Nitchum, 702-798-2103).
* **** Chromatography/fourier Transform Infrared
The application of gas chromatography/Fourier transform
infrared (aC/FT-IR) data to regulatory decisions requires the
availability of validated analytical protocols. A GC/fT-lR
protocol was developed by EM8L-LV (Donald Qurfca, 702-798-2113)
that is applicable to the determination of s«nivolatile organic
compounds in waste water, soils, sediments and solid waates. The
protocol is designed for automated analysis of multicomponent
environmental and havardous waste .extracts, waste water analysis
for semivolatile organic compounds is based upon extracting 1
L of sample with methylens chloride and concentrating.the sample
extract to 1.0 mL. The analysis of the semivolatile fraction
derived from solid waste analysis is based upon extracting 50
grams of sample and concentrating the sample extract to 1.0 mL.
A ge>l permeation option is included to further purify those
extracts which cannot be concentrated to the specified final
.volume* ',._. '. r^ - ^ '-' >-'' -I;.-..-. . . ... , "..^ ;'' ; r>-.- --..'; ,,: ....
Osing capillary GC/FT-IR techniques, waste water identifica-
tion limits of 150 to 400 ppb can be achieved with this method
while the corresponding identification limits for solid samples
are 3 to 8 ppm. Automated packed column GC/PT-IR identification
limits are approximately a factor of five higher than the corres-
ponding capillary CC/FT-IR values. The most frequent obstacle to
achieving these identification limits is expected to be the
presence of large'quantities of interfering high boiling coextrac-
tants. These coextractants would raise the identification limits
by preventing the concentration of extracts to the desired final
volume, thereby necessitating gel permeation clean-up, and/or by
-------
decreasing the spectral signal-to-noise QC-volatile analytic by
raising the spectral background intensity.
Trio la Quadruple Haas Spectroscopy
'"* In order to develop analytical methods for compounds that do
not gas chromatograph, BNSL-Or (Steve ftillete, 702*798-2332)
has been studying the use of alternative analytical methods. A
number of dye compounds and dye wastes wets characterised by
MS/MS using thermoepray ioni*ation and triple-qusdrupole mass
epeotrometry. TIM pblymethlhe, eoumsrln, vanthene, erytoethane
and noo-euDonated aso compounds ore claaaea of dye* that, in
general, give collision mass spectra when introduced Into the
spectrometer via the thermosprar interface without chrometogrnphy.
Other claaaea of dyes, such as phthaioeyanines, stilbenes and
bensidinedyes, have not been detected by this method.
Many synthetic intermediates and other dyes were detected in
standards of individual dyes. The precursor compounds are generally
not completely removed from the commercial product. The dyestuffs
are formulations based on color index standards. Dye wastee
(mother liquors from various dye processes) were analysed without
pretreatment before injection into the spectrometer. High percent-
agea of the starting materiala used In the synthesis of specified
dyes were often detected. Other types of dyes spiked into these
wastes could be identified by their collision spectra. The
quantification of these dyes proved difficult because of the
impurity of the dye standards.
Detection limit studies indicate that at leaat 100 ng to 250
ng of injected material are needed for the dye to be identified.
However, specific detection limits are dependent upon the class
of dye and its actual chemical structure.
The use of tripie-quadrupole mass spsctrametry proves very
effective in identifying those dyes that undergo thermoepray
ionisation. The production of a useful spectrum from the single
ion per compound generated from thermospray ion1sation makes the
tripie-quadrupole mass spectrometer useful in both structure and
mixture analysis,
Test Methods Discussed at AOAC Meeting In Pallas
Zn addition to the above studies for which reports have
recently been completed, two other methods under development,
a new Toxic!ty Characteristic Leaching Procedure (TCLP) and a
Sorbent Preasure Test Method, ware discussed at a session of tha
AOAC meeting on April 11, 1985 in Dallas, Texas, A discussion
of these new methods is presented in the following sections.
-7-
-------
Toxic!ty Characteristic Leaching Procedure
The 1984 amendments to the RCRA call for EPA to make the
Extraction Procedure (BP) More accurate and to expand the hazardous
waste characteristics. In response, the Agency is expanding the
list of compounds that can be detected using the Extraction Procedure
and is developing a second generation mobility procedure. The
new TCLP is being designed to be suitable for determining the
leaching behavior* of volatile compounds as well as to be less
costly and sore precise than the ourrent EP.- furthermore, the
features of this new procedure are expected to approximate the
leaching action of a sanitary landfill. The draft TCLP that was
discussed at the AOAC meeting in Dallas makes the use of a sero-
headspaee extraction vessel, the tumbler type- agitator currently
employed in the IF, and an acetic acid/sodium acetate buffer as
the extraction medium. A report is in preparation summarising
the results of the lysimeter. and laboratory experiments that led
to the draft procedure* A copy of the draft method is attached.
Por further information, contact Todd Kimmell (202-382-4795).
Sorbent Pressure Test Method
Compression of materials occurs during routine landfill
operations, and the 1984 amendments to RCRA directed EPA to
prohibit the landfill disposal of liquids absorbed in materials
(sorbents) which can release these liquids when compressed.
In order to determine whether certain sorbents could release
liquids under simulated landfill pressure, the Agency has Initiated
a program to develop a new test method, entitled the Sorbent
Pressure Test* As discussed at the AOAC meeting, eentrifugation
and consolidation are ourrentry being investigated as a means of
simulating landfill pressure. The test will be designed to be
easily applied in the field and to yield both qualitative and
quantitative results*
Attachment
------- |