PB84^178482
Inter laboratory Comparison
Methodf for Volatile and Seraivotlatile
Battelle Columbus Labs., OH
Prepared for
Environmental Monitoring and Support Lab,
Las Vegas/ NV
Mar 84
U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
NIIS
-------
EE84-178432
EPA-600/4-84-027
March 1984
INTERLABORATORY COMPARISON STUDY:
Methods for Volatile and Semivolatlle Compounds
by
The Battelle Columbus Laboratories
Columbus, Ohio 43201
EPA Contract No. 68-03-3098
EPA Project Officer
Donald F. Gurka
Quality Assurance 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, NEVADA 89114
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-600/4-84-027
JIPIENT'S ACCESSION NO.
% A 178482
4. TITLE AND SUBTITLE
INTERLABORATORY COMPARISON STUDY: Methods for
Volatile and Semi volatile Compounds
5. REPORT DATE
March 1984
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
The BatteHe Columbus Laboratories
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
The Battelle Columbus Laboratories
Columbus, Ohio 43201
10. PROGRAM ELEMENT NO.
ABSD1A
11. CONTRACT/GRANT NO.
Contract No. 68-03-3098
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency--Las Vegas, NV
Office of Research and Development
Environmental Monitoring Systems Laboratory
Las Vegas, Nevada 89114
13. TYPE OF REPORT AND PERIOD COVERED
Response Report 1, 5/82-1/84
14. SPONSORING AGENCY CODE
EPA/600/07
15. SUPPLEMENTARY NOTES
16. ABSTRACT
R0ut1-ne monitoring of the organic constitutents of hazardous waste
priori
task for the Environmental Protection Agency. Ultimately, the goal of routine moni-
toring must be the production of reliable data but the diversity of organic chemicals
and the complexity of hazardous waste forms make monitoring a difficult task. A
fundamental requirement for environmental monitoring is the availability of reliable
analytical methodology for the identification and quantitation of organic compounds.
This methodology must be of proven sensitivity, accuracy and precision; it must also
be facile and applicable to as many organic compounds and hazardous waste types as
possible. Lastly, the methodology should be acceptable to a broad spectrum of the
scientific community. One way to ensure the scientific acceptability of methodology
is to prove its sensitivity, precision, and accuracy utilizing strict guidelines for
conducting an inter! aboratory test program. The guidelines provided to the inter-
laboratory test participants in this program included test protocols for volatile
and semivolatile analysis, test samples and standards, quality assurance guidance
and directions for reading and submitting data. In Phase I and II of this study,
methods for the analytical determination of volatile and semivolatile organic compound^
in hazardous wastes were selected. These methods were then modified and tested in a
single-laboratory evaluation. The final task, Phase III of the overall project,
subjected the protocols and the experience from Phases I and II to an inter! aboratory
ty
toc-fr
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
UNCLASSIFIED
21. NO. OF PAGES
378
SS (This page)
22. PRICE
EPA Form 2220-1 (R«v. 4-77) PREVIOUS EDITION is OBSOLETE
1
-------
NOTICE
THIS DOCUMENT HAS BEEN REPRODUCED
FROM THE BEST CO?? FURNISHED US BY
THE SPONSORING AGENCY. ALTHOUGH IT
IS RECOGNIZED THAT CERTAIN PORTIONS
ARE ILLEGIBLE, IT IS BEING RELEASED
IN THE INTEREST OF MAKING AVAILABLE
AS MUCH INFORMATION AS POSSIBLE,
-------
NOTICE
This report has been reviewed 1n accordance with the U.S. Environmental
Protection agency's peer and administrative review policies and approved
for presentation and publication. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
11
-------
ABSTRACT
Routine monitoring of the organic constituents of hazardous waste 1s a
priority task for the Environmental Protection Agency. Ultimately, the goal
of routine monitoring must be the production of reliable data but the
diversity of organic chemicals and the complexity of hazardous waste forms
make monitoring a difficult task.
A fundamental requirement for environmental monitoring is the
availability of reliable analytical methodology for the identification and
quantitatlon of organic compounds. This methodology must be of proven
sensitivity, accuracy and precision; it must also be facile and applicable to
as many organic compounds and hazardous waste types as possible. Lastly, the
methodology should be acceptable to a broad spectrum of the scientific
community. One way to ensure the scientific acceptability of methodology is
to prove its sensitivity, precision, and accuracy utilizing strict guidelines
for conducting an interlaboratory test program. The guidelines provided to
the interlaboratory test participants in this program included test pro-
tocols for volatile and semivolatile analysis, test samples and standards,
quality assurance guidance and directions for reading and submitting data.
In Phase I and II of this study, methods for the analytical determination
of volatile and semi-volatile organic compounds 1n hazardous wastes were
selected. These methods were then modified and tested in a single-laboratory
evaluation. The final task, Phase III of the overall project, subjected the
protocols and the experience from phases I and II to an interlaboratory test.
The body of this Response Report is composed of the individual reports for
Phases I to III which were separately prepared and submitted to the Project
Officer over a period of several years.
111
-------
EXECUTIVE SUMMARY
INTRODUCTION
The United States Environmental Protection Agency (EPA) is currently
involved in a broad program to assess and improve the reliability of the data
generated from its hazardous waste analytical protocols (1). This program is
essential to maintain the credibility of the EPA as a regulatory agency (2).
Included in this broad program is the evaluation of those analytical protocols
in widest usage by an interlaboratory comparison test. Such inter!aboratory
comparison testing is one of the most important elements of any quality assur-
ance plan (3). Despite the substantial expenditure required in both time and
money, the multilaboratory testing of many EPA methods is currently underway-
(4,5). This report describes the testing of analytical protocols for the
determination of volatile compounds (boiling up to ca. 150°C) and semivolatile
compounds (boiling above 150°C) in hazardous wastes (6) by an interlaboratory
comparison test.
BACKGROUND
Both the ASTM and the AOAC have set guidelines for planning, conducting
and analyzing the data derived from interlaboratory testing (7,8). The Food and
Drug Administration (FDA) has been extensively involved in such testing and has
recently published the results of over fifty tests (9). Formal testing of envi-
ronmental methods by other Federal agencies has lagged behind that of Food and
Drug, however the NBS has recently reported a few (10,11).
For this study it was considered desirable to meet at least the minimum
ASTM requirements for the number of participating laboratories, samples and
analysis replicates. Earlier EPA communication with ASTM Committees D-34, D-19
and E-ll had outlined the workplan, requested input, and requested voluntary
participants (12-16). Some valuable input was received from these exchanges
and where possible suggested modifications were incorporated into the inter-
laboratory test workplan. However, voluntary participation was not forthcoming
and it was necessary to carry out the test solely with contractor laboratories.
WORKPLAN
The workplan for this project consisted of three distinct phases:
1. Selection, evaluation and optimization of the best available methods for
determining volatile and semivolatile organic compounds in solid wastes.
iv
-------
2. Rigorous single laboratory evaluation of the optimized methods and re-
vision (if necessary). Preparation of all materials to be distributed to the
participating laboratories and peer review of the selected methods.
3. Interlaboratory comparison testing.
The guiding workplan philosophy was to severely challenge the selected
methods. Thus, the target analytes and sample matrices were selected to
represent a spectrum of environmental analytical situations. In addition, each
waste sample contained analytically significant levels of naturally incorporated
compounds. This ensured the presence of background interference materials and
provided a "real world" matrix background for the spike compounds.
RESULTS
The within-laboratory variability for analysis of volatiles (VOA) was
generally less than 30 percent but ranged from 5 to 300 percent depending on
the compound. The total VOA variability generally ranged from 20 to 80 per-
cent for most of the waste samples while the between-laboratory variability
was usually less than 70 percent but ranged from 5 to 300 percent. The highest
VOA variabilities, both total and component, were reported for non-spike com-
pounds that are common background contaminants such as methylene chloride,
dichloroethane, chloromethane, and chloroform. Only in the case of sample 4
were high percent RSD's reported for spiked compounds. The effect of labora-
tory on the detectability of target volatile compounds was evident but detecta-
bility was also sample dependent.
The sample-to-sample variability noted for the volatile compounds was also
evident for the semivolatile compound analyses. The differences in amount of
data reported were greater for the determination of semivolatiles than for the
detemination of volatiles. For example, for one sample, the number of com-
pounds reported varied from 1 to 27, with 20 being the average number of com-
pounds reported. The total RSD generally ranged from 30 to 80 percent. The
RSD for the within-laboratory component was less than 30 percent and the
between-laboratory variability was about twice that value. The ranges of both
components were the same as those reported for volatile organics, 5 to 300
percent. No difference in ranges of values is apparent for the spiked versus
non-spiked compounds; however, some of the poorer precision may be attributable
to analyte polarity and to background contaminants, such as phthalates.
CONCLUSIONS AND RECOMMENDATIONS
The non-aqueous neutral extraction followed by fused silica capillary
column GC/MS analysis eliminates separate acid/base extractions and reduces
the required number of 6C/MS runs. The analytical data reported for all samples
will be investigated in greater detail to determine the causes of statistical
outliers. If these outliers result from laboratory deviations from the test
methods and quality control protocols, they can be excluded from the pooled
-------
precision results (17). However, the validity of discarding outlier results is
still a subject of disagreement among statisticians and chemists (18, 19). If
outliers are shown to be a result of protocol ambiguities, the appropriate
protocol sections should be clarified. If they are shown to be a result of de-
ficiencies in the protocols, the appropriate protocol sections will be modified.
These steps may reduce the number of outliers generated in future applications
of these methods. The between-laboratory to within-laboratory ratio of about
two for the precision of these test methods compares favorably with the recently
compiled results of over fifty AOAC interlaboratory comparison tests (20).
The wide variation in individual laboratory performance demonstrates the
need for strong QA/QC monitoring of laboratories performing routine environ-
mental analyses. If these samples had been submitted as "blinds" to the par-
ticipating laboratories, the interlaboratory agreement would have probably
been poorer (21). Although each participating laboratory had some prior
experience with the test protocols, it is anticipated that further experience
will lead to improved analytical performance (22).
VI
-------
LITERATURE CITED
(1) Gurka, D. F., Meier, E. P., Beckert, W. F., and A.F. Haeberer. Paper
presented at the 3rd National Conference on Management of Uncontrolled
Hazardous Waste Sites in Washington, D. C., Nov. 1982.
(2) "Improving Analytical Chemical Data used for Public Purposes"; C. and
E. N., Oune 7, 1982, 44.
(3) "Guidelines for Data Acquisition and Data Quality Evaluation in Environ-
mental Chemistry," Anal. Chem. 52:2242, 1980.
(4) Warner, J. S., Slivon, L. E., Meehan, P. W., Landes, M. C., and A. T.
Bishop. Paper presented at the Division of Environmental Chemistry of
the American Chemical Society in Las Vegas, NV, Mar. 1982.
(5) McMillan, C. R., Hilemen, F. D., Kirk, D. E., Mazer, T., Warner, J. J.,
Longbottom, J., and R. Wesselman. Paper presented at the Division of
Environmental Chemistry.of the American Chemical Society in Kansas City,
MO, Sept. 1982.
(6) Fed. Regist. 1979, 44(223), 69464.
(7) "Standard Practice for Conducting an Interlaboratory Test Program to
Determine the Precision of Test Methods"; ASTM Part 41, E691, 1980, 959.
(8) "Collaborative Study Procedures of the AOAC"; Prepared for the Joint
International Symposium "The Harmonization of Collaborative Studies" in
London, Eng. Mar 1978. Published by the A. C. S. 1978.
(9) Horwitz W., Kamps, L. R. and K. W. Boyer. J.A.O.A.C. 63:1344, 1980.
(10) Hilpert, L. R., May, W. E., Wise, J. A., Chesler, S. N., and H. S. Hertz.
Anal. Chem. 50:458, 1978.
(11) Wise, S. A., Chesler, S. N., Guenther, F. R., Hertz, H. S., Hilpert,
L. R., May, W. E., and R. M. Parris, Anal. Chem. 52:1828, 1980.
(12) Private Communication from J. S. Warner to members of ASTM D34 task group,
Feb. 9, 1981.
(13) Private Communication from J. S. Warner to L. H. Howe of ASTM D19 Committee,
Apr. 22, 1982.
(14) Private Communication from J. S. Warner to R. C. Paule, NBS, Apr. 22, 1982.
vii
-------
(15) Private Communication from D. Friedman to William Webster, ASTM D34
Committee, Dec. 7, 1981.
(16) Private Communication from J. S. Warner to members of ASTM D34.02.04 task
group, Dec. 14, 1981.
(17) "Standard Practice for Conducting an Inter!aboratory Test Program to
Determine the Precision of Test Method;" ASTM Part 41, E691, 1980, 959.
(18).Horwitz, W. J.A.O.A.C. 1977, 60, 1355.
(19) Horwitz, W. J.A.O.A.C. 1983, 66, 455.
(20) Horwitz, W., Kamps, L.R. and Boyer K.W., J.A.O.A.C. 1980, 63, 1344.
(21) Maugh II, T. H. Science 1982, 215, 490.
(22) Sherma, J. "Manual of Analytical Quality Control for Pesticides and
Related Compounds," EPA-600/1-79-008, 1979.
viii
-------
CONTENTS
Abstract iii
Executive Summary iv
Literature Cited vi1
Figures xi
Tables xii
Phase I Studies 1
Introduction 2
Technical Discussion . 2
References 54
Phase I Appendices
A. Analytical Results from Target Wastes 55
B. Quality Assurance/Quality Control for the Interlaboratory
Comparison Study 64
Phase II Studies 77
Introduction 78
Summary 79
Technical Discussion 80
Quality Assurance/Quality Control 123
Proposed Additional Studies 128
References 132
Phase III Studies 133
Introduction 134
Conclusions 135
Recommendations 138
Experimental Procedures 140
1x
-------
CONTENTS (Continued)
Phase III Studies (Continued)
Analytical Test Procedure 140
Collaborative Test Procedures 140
Preparation of Calibration Solutions 140
Preparation of Spiked Wastes 141
Design of Collaborative Test 141
Data Processing Procedures 142
References 204
Phase III Appendices 205
A. Method for the Determination of Semivolatile Organic Compounds
in Solid Wastes 205
B. Method for the Determination of Volatile Organic Compounds in
Solid Wastes 223
C. Quality Control Protocol. 245
D. Description of Standard Solutions 270
E. Manual for Collaborators 276
F. Volatiles Report Forms 330
G. Semivolatiles Report Forms 342
-------
FIGURES
Number Page
Phase I Studies
1 Gas chromatogram of extract from poly still bottoms before
and after GPC cleanup 41
2 Gas chromatogram of extract from Ethanes I spent catalyst
before and after GPC cleanup 42
3 Gas chromatogram of extract from Coal Gasification Tar before
and after GPC cleanup 43
4 Total 1on chromatogram obtained from sample S0186 using a DB-5
Fused Silica Capillary Column 47
5 Total ion chromatogram obtained from sample S0186 using an
SE-52 Glass Capillary Column 48
Phase II Studies
1 TIC of water sample EPA/05344, run 2 72
2 Mass spectrum of scan no. 7 in semi volatile sample no.
EPA/05344, run 2 73
3 Library search results for scan no. 7 in semivolatile sample no.
EPA/05344, run 2 74
4 Mass spectrum of scan no. 14 in semivolatile sample no.
EPA/05344, run 2 75
5 Library search results for scan no. 14 1n semivolatile sample
no. EPA/05344, run 2 76
Phase III Studies
1 Purging chamber 242
2 Trap packings and construction to include desorb capability . . . 243
3 Schematic of purge and trap device - purge mode 244
4 Schematic of purge and trap device - desorb mode 244
xi
-------
TABLES
Number
Phase I Studies
1 200 Compounds to be Determined in Solid Wastes 6
2 Candidate Spiking Compounds 9
3 Varied Spiking Scheme 11
4 Approximate Quantities of Each Semivolatile Spiking Compound
to be Added to Various Waste Samples 12
5 Approximate Quantities of Each Purgeable Spiking Compound
to be Added to Various Waste Samples 13
6 Repeatability of GC/MS Absolute Retention Times, Set 2 19
7 Repeatability of GC/MS Absolute Retention Times for
Deuterated Internal Standards, Set 2 - 21
8 Repeatability of GC/MS Absolute Retention Times, Set 3 22
9 Repeatability of GC/MS Absolute Retention Times for
Deuterated Internal Standards, Set 3 24
10 Relative Retention Time Data for Bis(2-Chloroethyl) Ether .... 25
11 Relative Retention Time Data for 2-Chlorophenol 25
12 Relative Retention Time Data for Nitrobenzene 26
13 Relative Retention Time Data for Isophorone 26
14 Relative Retention Time Data for 2,4-Dimethylphenol 27
15 Relative Retention Time Data for Hexachlorobutadiene 27
16 Relative Retention Time Data for Acetanilide 28
17 Relative Retention Time Data for 3,4-Dichloroaniline 28
18 Relative Retention Time Data for 2,4-Dinitrotoluene 29
19 Relative Retention Time Data for Hexaethylbenzene 29
20 Relative Retention Time Data for Pentachlorophenol 30
21 Relative Retention Time Data for Decafluorotriphenylphosphine . . 30
22 Relative Retention Time Data for Anthraquinone 31
23 Relative Retention Time Data for Fluoranthene 31
24 Relative Retention Time Data for 4,4'-DDE 32
25 Relative Retention Time Precision Data 33
26 Repeatability of DFTPP Ion Abundances, Set 2 34
27 Repeatability of DFTPP Ion Abundances, Set 3 35
28 Precision of GC/MS Response Factors of Compounds Used as
Internal standards and for Spiking on an SE-52 Fused Silica
Capillary Column 36
29 Repeatability of GC/MS Response Factors for Deuterated
Internal Standards 37
30 Repeatability of Area Count of an Internal Standard (D-10-
Phenanthrene) for 25 Consecutive GC/MS Analyses 38
xii
-------
TABLES (Continued)
Number
Phase I Studies
31 Comparison of Glass and Fused Silica Capillary Columns for
GC/MS Analysis of Sample S-0190 45
32 Comparison of Glass and Fused Silica Capillary Columns for
GC/MS Analysis of Sample S-0180 45
33 Comparison of Glass and Fused Silica Capillary Columns for
GC/MS Analysis of Sample S-0186 46
34 Performance of Commercial Fused Silica Capillary Columns 50
Phase II Studies
1 Waste Samples Selected for the Inter!aboratory Study 87
2 Spiking Compounds 90
3 Semi volatile Compound Spike Levels 92
4 Volatile Compound Spike Levels 94
5 Analysis Results for Spiked Sample ILS-1 (Creosote Contaminated
Soil) for Volatile Organic Compounds 96
6 Analysis Results for Spiked Sample ILS-2 (Latex Paint Waste)
for Volatile Organic Compounds 96
7 Analysis Results for Spiked Sample ILS-3 (Ethanes Spent
Catalyst) for Volatile Organic Compounds 97
8 Analysis Results for Spiked Sample ILS-4 (Coal Tar) for
Volatile Organic Compounds 97
9 Analysis Results for Spiked Sample ILS-5 (Oxychlorination
Catalyst) for Volatile Organic Compounds 98
10 Analysis Results for Spiked Sample ILS-6 (Cincinnati Dewatered
Sludge) for Volatile Organic Compounds 98
11 Analysis Results for Spiked Sample ILS-7 (Herbicide Manufac-
turing Acetone-Water Waste) for Organic Compounds 99
12 Analysis Results for Spiked Sample ILS-8 (Chlorinated Ethanes
Waste) for Volatile Organic Compounds 99
13 Average Recovery and % RSD of All Spiked Volatile Compounds
(High and Low) in Each Waste Sample 100
14 Summary of Analysis Data for Spiked Sample ILS-1 (Creosote
Contaminated Soil) for Semivolatile Organic Compounds 102
15 Summary of Analysis Data for Spiked Sample ILS-10 (Extract of
Creosote Contaminated Soil) for Semivolatile Compounds 103
16 Summary of Analysis Data for Spiked Sample ILS-2 (Latex Paint)
for Semi volatile Organic Compounds 104
17 Summary of Analysis Data for Spiked Sample ILS-3 (Ethanes
Spent Catalyst) for Semivolatile Organic Compounds 105
18 Summary of Analysis Data for Spiked Sample ILS-4 (Coal Tar)
for Semivolatile Organic Compounds 106
19 Summary of Analysis Data for Spiked Sample ILS-5 (Oxychlorinated
Spent Catalyst) for Semivolatile Organic Compounds 107
xiii
-------
TABLES (Continued)
Number Page
Phase II Studies
20 Summary of Analysis Data for Spiked Sample ILS-6 (Cincinnati
Dewatered Sludge) for Semivolatile Organic Compounds 108
21 Recovery of Spiked Compounds from POTW Sludge 110
22 Recovery of Spiked Compounds from Latex Paint Waste Ill
23 Recommended Compound Response Factors to be Monitored During
the Interlaboratory Comparison Study 113
24 EPA Recommended Compound Response Factors to be Monitored
During the Interlaboratory Comparison Study 116
25 List of Standard Solutions 117
xiv
-------
TABLES (Continued)
Number Page
Phase III Studies
1 Waste Samples Used 1n the Study 142
2 Spiking Compounds 143
3 Concentration of Volatile Compounds Spiked into Waste
Samples 145
4 Concentration of Semi volatile Compounds Spiked into Waste
Samples 146
5 Specific Compounds Searched for in Samples 147
6 Variability of Relative Retention Times and Response Factors
for the Determination of Semi volatile Compounds 152
7 Variability of Retention Times and Response Factors for the
Determination of Volatile Compounds 154
8 Recovery Variability of Volatile Compounds from Sample
ILS-2 155
9 Recovery Variability of Volatile Compounds from Sample
ILS-3 156
10 Recovery Variability of Volatile Compounds from Sample
ILS-4. . . 157
11 Recovery Variability of Volatile Compounds from Sample
ILS-5. ' 158
12 Recovery Variability of Volatile Compounds from Sample
ILS-6 159
13 Recovery Variability of Volatile Compounds from Sample
ILS-7 160
14 Recovery Variability of Volatile Compounds from Sample
ILS-8 161
15 Sample-to-Sample Recovery Dependencies-VOA 162
16 Recovery Variability of Semi volatile Compounds from Sample
ILS-2 163
17 Recovery Variability of Semi volatile Compounds from Sample
ILS-3 165
18 Recovery Variability of Semi volatile Compounds from Sample
ILS-4 167
19 Recovery Variability of Semi volatile Compounds from Sample
ILS-5 170
20 Recovery Variability of Semi volatile Compounds from Sample
ILS-6 172
21 Recovery Variability of Semi volatile Compounds from Sample
ILS-9 174
22 Sample-to-sample Recovery Dependencies-SV 175
23 Effect of Removing Outliers on Data Quality for the Deter-
mination of Semi volatile Compounds in Sample ILS-9 176
24 Recovery of Volatile Compounds from ILS-2 177
25 Recovery of Volatile Compounds from ILS-3 178
26 Recovery of Volatile Compounds from ILS-4 179
27 Recovery of Volatile Compounds from ILS-5 .180
28 Recovery of Volatile Compounds from ILS-6 181
xv
-------
TABLES (Continued)
Number Page
Phase III Studies
29 Recovery of Volatile Compounds from ILS-7 182
30 Recovery of Volatile Compounds from ILS-8 183
31 Recovery of Semi volatile Compounds from ILS-2 184
32 Recovery of Semi volatile Compounds from ILS-3. 186
33 Recovery of Semi volatile Compounds from ILS-4 188
34 Recovery of Semi volatile Compounds from ILS-5 190
35 Recovery of Semi volatile Compounds from ILS-6 192
36 Matrix of Total Volatile Compounds Reported 194
37 Matrix of Total Semi volatile Compounds Reported 194
38 Effect of Compound, Sample and Spike Level on Detectability
of Volatile Compounds 196
39 Effect of Compound, Sample and Spike Level on Detectability
of Semi volatile Compounds 197
40 Effect of Laboratory on Detectability of Volatile Compounds. . . 199
41 Effect of Laboratory on Detectability of Semi volatile
Compounds. 200
42 Amount of Tetraglyme Extract Analyzed for the Determination
of Volatile Compounds 202
43 Methylene Chloride Extract Concentration Factor Used
for Semi volatile Compound Determination 203
xvi
-------
Phase I Studies
-------
INTRODUCTION
Phase I Studies
A variety of analytical methods have been developed and applied to the
determination of organic constituents 1n solid wastes. Each research group
associated with the development of a given method Instituted appropriate
Intralaboratory quality control to ensure applicability and appropriate
accuracy and precision. However, few of these methods have been evaluated In a
well designed Interlaboratory comparison study.
Currently, two solid waste analysis methods are 1n use 1n a large number
of environmental analysis laboratories:
• Modified NEIC (National Enforcement Investigation Center). See page 50.
• Modified MRI (Midwest Research Institute)
An Evaluation study was Initiated at the Battelle-Columbus laboratory
which consisted of three phase's:
I. Methods Evaluation, Modification and Selection
II. Intralaboratory Evaluation
III. Interlaboratory Evaluation.
The objective of Phase I was the selection of two methods for the
analysis of solid wastes: one method for determination of volatile organic
compounds and one method for determination of semivolatile organic compounds.
The Phase I report contains a discussion of the samples and compounds to
be used and a discussion of the spiking scheme. Quality control criteria are
discussed, followed by descriptions of evaluation and modification studies and
method selection. Finally, detailed descriptions of the methods and results
of analyses of wastes are provided.
TECHNICAL DISCUSSION
Overview
Phase I results provided the basis for Phase II and III efforts. The
intralaboratory and Interlaboratory validations studies were based on waste
materials, search list compounds, spiking compounds and the spiking scheme
selected in Phase I. The quality assurance parameters, the modifications and
-------
the methods selected all have direct and crucial bearing on how the later
Phases are conducted. The following discussion addresses each of the Phase I
decisions.
Materials and Compounds
Selection of Wastes
The analysis methods selected for validation must be applicable to as
wide a range of waste types as possible. Therefore, 1n order to challenge the
methods, a broad range of waste types is needed. At the same time the number
of waste samples to be analyzed must be kept to a workable, economically
feasible level.
In general, the number and types of materials to be Included in an
interlaboratory study will depend on the following:
• The range of values and how the precision varies over that range.
§ The number of different types of materials to which the test method 1s
to be applied.
• The difficulty and expense Involved in obtaining, processing and
distributing samples.
• The difficulty of, length of time required for, and the expense
of performing the tests.
• The commercial or legal need for obtaining a reliable and
comprehensive estimate of precision.
• The uncertainty of prior information on any of these points.
With these criteria 1n mind, wastes were selected:
• That represent as many waste types as possible.
• That contain a variety of compounds that are potentially hazardous.
• That represent a challenge to extraction.
• That include significant amounts of many compounds.
• That are available in quantities sufficient for program needs.
The following 10 waste materials met the above criteria:
Olentangy River Sediment
-------
Machine Oil Waste
Herbicide Manufacturing Acetone-Water Waste
Coal Gasification Tar
Spent Caustic
Oxychlorination Spent Catalyst
Cincinnati Dewatered Sludge
Latex Paint Waste
Ethanes I Spent FeCl2 Catalyst
Drying Bed Solids (S0190).
The results of analyses of these waste samples are shown in Tables A-l
through A-10 in Appendix A.
It is clear that a broad variety of wastes is represented: solid,
liquid, semisolid, inorganic and organic. Further, each waste contains many
potentially hazardous compounds that are present in detectable amounts. The
National Bureau of Standards Standard Reference Material No. 1645 (SRM 1645)
was used as a reference material, not because it contains certified amounts
of organic compounds, but because it is broadly available and does contain
certifiable amounts of organic compounds. The results from this program will
provide data that will increase the usefulness of SRM 1645.
Search List
The number and kind of compounds to be determined in this program is
subject to practical constraint. The total number of compounds that may be
selected for study was very large and depended on which of the Agency target
lists were used. Compounds were considered from a number of sources including:
(1) the EPA Priority Pollutant list, (2) a list of 900 compounds prepared by
EPA's Information Clearing House, (3) the lists of compounds submitted to EMSL-
Las Vegas by the EPA Regions, (4) the PURPL 11st, (5) Appendix VIII from page
33,132 of the May 19, 1980 Federal Register, (fi) the list of anilines for
which Battelle Is developing wastewater analyses methods for EPA, and (7) the
compounds available from Aldrich Chemical Company in 1-kg lots. In general,
compounds were selected on the basis of expected ability for analysis by
capillary column gas chromatography, appropriateness for being representative
of a particular class of compounds, ready availability and stability. Com-
pounds that are gases, nonvolatile, insoluble in organic solvents, highly
reactive or highly unstable in water were omitted. Ultimately, the selection
was limited to 200 compounds as a reasonable compromise between the length of
the GC/MS search time and class representativeness. The use of 200 compounds
served to evaluate the analysis methods for most Important compound classes.
In future application of the evaluated method, the method must be evaluated for
-------
specific compounds, other than these 200 compounds, through precision and
recovery studies in the laboratory doing the work.
The list of 200 compounds to be determined by GC/MS analyses is given in
Table 1. Included are 60 volatile compounds and 140 semivolatile compounds
representing a wide range of chemical classes. The compounds to be used for
spiking and most of the compounds identified as being present in the above
wastes are included.
Spiking
Spiking Scheme
The initial design of the interlaboratory study involved 10 solid wastes
and 25 pairs of spiking compounds. The candidate spiking .compounds recommended
are shown in Table 2. Each of 10 waste samples was spiked with one compound
from each pair at a high level and one at a low level. The high and low levels
used will correspond to those levels that will give 100 and 20 ng on the GC
column during analysis if 100% recovery were achieved.
The simplest scheme for spiking samples in keeping with the above design
would be to prepare one spiking solution that contains all of the compounds to
be spiked at the high level and a second spiking solution that contains all of
the compounds to be spiked at .the low level. Each waste sample would be
spiked with each of the two spiking solutions. By using the above scheme the
ratio between the high level and low level would be the same for all 25 pairs
of compounds and for all waste samples. However, this pattern could soon
become apparent to the partlcipitating laboratories and some analysts may be
tempted to force the data to fit the pattern by repeating runs, recalibrating,
or even falsifying data.
In an effort to avoid the above possibility, a varied spiking scheme was
devised. The varied spiking scheme for the semivolatile components Involves
preparing 10 different spiking solutions, each containing four compounds from
four different pairs. The 10 spiking solutions were added to the waste in
a varied manner to give a varied ratio of 4:1, 5:1, 6:1, 1:4, 1:5, or 1:6 for
the two compounds in a pair. A summary of the varied spiking scheme is given
1n Table 3. A similar scheme was used for the volatile components. Although
the use of the varied spiking scheme should discourage analysts from forcing or
falsifying the data, there are least three important disadvantages to the
varied spiking scheme: (1) The varied spiking scheme requires more effort
than a simpler scheme for the preparation of the spiked samples and for the
processing of the data; (2) The data should be perturbed by the varied ratios
used for the two compounds of each pair; (3) A pilot study was needed to
demonstrate that the compounds 1n each pair are similar 1n respect to recoveries
achievable at high and low levels. If the recoveries achieved in the pilot study
for some pairs are not similar, the concentrations of the particular compounds
involved will not be varied 1n the spiked samples or replacement compounds will
be selected and studied. An outline of the pilot study follows:
-------
TABLE 1. 200 COMPOUNDS TO BE DETERMINED IN SOLID WASTES
PURGEABLES
Purgeable Halocarbons
Chloromethane
Bromomethane
Vinyl chloride
Chloroethane
Methylene chloride
Ethyl bromide
1,1-Dichloroethane
1,1-Dichloroethane
trans-l,2-D1ch1oroethene
Chloroform
1,2-Dichloroethane
1,1,1-Trichloroethane
Carbon tetrachloride
Bromodi chloromethane
1,2-Di chloropropane
trans-l,3-Dichloropropene
Trichloroethene
Dibromochloromethane
1,1,2-Trichloroethane
ci s-1,3-Oichloropropene
2-Chloroethyl vinyl ether
Bromoform
1,1,2,2-Tetrachloroethane
Tetrachloroethene
Chlorobenzene
1,1,2-Trichlorotri f1uoroethane
Dibromomethane
Ally! chloride
Ethylene dibromide
Chloropicrin
2-Chloropropane
1-Chlorobutane
Purgeable Hydrocarbons
Benzene
Toluene
Ethyl benzene
o-Xylene
Styrene
Purgeable Oxygen,
Compounds'
2-Butanone
Cyclopentanone
4-Methyl-2-pentanone
2-Hexanone
Carbon disulfide
Dimethyl disulfide
Acrylonitrile
Epichlorohydrin
2-Chloroacrylonitrile
Acetonitrile
Dichloroacetonitrile
n-Propionitrile
Acrolein
Chloroacetaldehyde
2-Chloroethanol
N-Ni trosodimethylami ne
Vinyl acetate
Dimethyl sulfide
Diethyl ether
Acetone
Methyl chloroacetate
Methyl acrylate
Methyl methacrylate
Sulfur, or Nitrogen
SEMIVOLATILES
Aliphatic Halocarbons
1,4-Dichlorobutane
Pentachloroethane
Hexachloroethane
Hexachloropropene
Hexachlorobutadiene
Aromatic Halocarbons
4-Chlorotoluene
Bromobenzene
1,2-Dichlorobenzene
1,4-Di chlorobenzene
1,2,4-Trichlorobenzene
(continued)
-------
TABLE 1. (Continued)
SEMIVOLATILES
Aromatic Halocarbons (continued)
1,2,4,5-Tetrachlorobenzene
Pentachlorobenzene
Hexachlorobenzene
3,3'-Dichlorobiphenyl
4,4'-Di chloroblphenyl
2,2',4,4'-Tetrachloroblphenyl
Benzal chloride
2,2',4,4',6,6'-Hexachlorobiphenyl
Benzyl chloride
1-Chloronaphthalene
2-Chloronaphthalene
a,a,a-Trichlorotoluene
Aromatic Hydrocarbons
Naphthalene
1,2,4-Tr1methyl benzene
1,2,4,5-Tetramethylbenzene
Biphenyl
Acenaphthylene
Acenaphthene
2-Methylnaphthalene
2-Ethylnaphthalene
2,3-Dimeth'yl naphthal ene
1,2,3,4-Tetrahydronaphthalene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(k)fluoranthene
Benzo(a)pyrene
D1benzo(a,h)anthracene
Benzo(g,h,1)perylene
Indeno(1,2,3-cd)pyrene
Aromatic Nitro Compounds
Nitrobenzene
1,3-01 nitrobenzene
Aromatic Nitro Compounds (Continued)
2-N1trotoluene
4-N1trotoluene
2,4-Dinitrotoluene
2,6-D1m'trotoluene
1-Chloro-4-ni trobenzene
2,4-Di ni trochlorobenzene
Phenols
2-Chlorophenol
2-N1trophenol
Phenol
2,4-D1methy!phenol
2,4-D1chlorophenol
2,4,6-Trichlorophenol
4-Chloro-3-methylphenol
2-Methylphenol
4-Methylphenol
Thiophenol
4-Chlorophenol
2,6-Dichlorophenol
2,4,5-Tri chlorophenol
Hexachlorophene
4-Hydroxyb1phenyl
2-Naphthol
4-t-Butylphenol
2-Chloro-4-nitrophenol
2,4-Din1trophenol
2-Methyl-4,6-dinitrophenol
Pentachlorophenol
4-N1trophenol
2,6-D1-t-butyl-4-methylphenol
2,4-D1-t-butylphenol
Dlethylstilbestrol
Amines
Aniline
4-Chloroaniline
4-BromoaniHne
2-Nitroanlline
3,4-D1chloroaniline
(continued)
-------
TABLE 1. (Continued)
Amines (Continued)
2,4,5-Tr1chloroanll 1 ne
3-Nitroanil1ne
4-Chloro-2-n1troanl11ne
4-N1troan1l1ne
2,6-01chloro-4-n1troanl11ne
2-Chloro-4-nltroanl11ne
2,4-D1n1troan1l1ne
N-Methylaniline
4-Chloro-o-2-methylan111ne
4-Methylaniline
2,6-Dimethylaniline
4-Am1nob1phenyl
l-Am1nonaphthalene
N,N-D1methy1aniline
Phenanthr1d1ne
4-Methylpyrldlne
2,4-01methylpyrld1 ne
4-t-Butylpyrldlne
1,2,7,8-01benzocarbazole
2,4,6-Tr1methylpyrld1ne
Qu1nol1ne
4-Methylqu1nol1ne
Acr1d1ne
Carbazole
3,3'-D1chlorobenz1d1ne
D1phenylam1ne
Chlorinated Pesticides
4,4'-ODD
4,4'-ODE
4,4'-DDT
Methoxychlor
Tr1flural1n
Pentachloronltrobenzene
Phthalates
Dimethyl phthalate
01-n-butyl phthalate
D1(2-ethylhexyl) phthalate
SEMIVOLATILES
Phosphates
TrKp-tolyl) phosphate
Tr1phenyl phosphate
A1dehydes
Benzaldehyde
4-Chlorobenzal dehyde
Ethers and Sulfides
Anlsole
Phenyl ether
Dlbenzofuran
Ketones
Anthraqulnone
2-Methylanthraqul none
Proplophenone
Acetophenone
4-Chlorobenzoic add
Benzole add
4-Bromobenzo1c add
2,4-D1chlorophenoxyacet1c acid
2,4,5-Tr1chlorophenoxyacetic ad d
Haloethers
B1s(2-chloroethyl)ether
B1s(2-chloroethoxy)ethane
4-Chlorophenyl phenyl ether
Miscellaneous
Azobenzene
AcetanlUde
Benzyl alcohol
Di(2-ethylhexyl) sebacate
===================================================================:
-------
TABLE 2. CANDIDATE SPIKING COMPOUNDS
:==========================================================================
VOLATILE COMPOUNDS
Low-bo111ng Halocarbons Ketones
1,1,1-Trlchloroethane Cyclopentanone
1,2-Dlchloropropane 2-Hexanone
Higher-boiling Halocarbons N1tr1les
Bromoform Prop1on1tr1le
1,1,2,2-Tetrachloroethane 2-Chloroacryloni tr11e
Aromatic Hydrocarbons
Ethyl benzene
Chiorobenzene
SEMIVOLATILE COMPOUNDS
Aliphatic Halocarbons M1d-bo1l1ng PAHs
Hexachloroethane Fluoranthene
Hexachloropropene Pyrene
Low-bo1Hng Aromatic Halocarbons H1gh-bo1Hng PAHs
4-Chlorotoluene 1,2,5,6-Dlbenzoanthracene
1,4-01chlorobenzene 1,2,7,8-D1benzocarbazole
H1gh-bo111ng Aromatic Halocarbons Aromatic N1tro Cpds.
Pentachlorobenzene l,3-D1n1trobenzene
Hexachlorobenzene 2,6-D1n1trotoluene
Chlorinated Pesticides Low-ac1d1ty Phenols
p,p'-DDD 2-Chlorophenol
p.p'-DDT 2,6-D1methylphenol
Low-bo1Hng PAHs H1gh-ac1d1ty Phenols
2-Ethylnaphthalene 4-N1trophenol
1-Chloronaphthalene 2,4-01n1trophenol
(continued)
-------
TABLE 2. (Continued)
===========================================================================
SEMIVOLATILE COMPOUNDS (Continued)
Phosphates Benzole Acids
Trlphenyl phosphate 4-Ch1orobenzo1c add
Tri-p-tolyl phosphate 4-Bromobenzoic add
Quinones Phenoxyacetlc Adds
Anthraquinone 2,4-D
2-Methylanthraqul none 2,4,5-T
Aromatic Ketones
Acetophenone
Proplophenone
===================:
t Three aliquots of each of 4 wastes were spiked with the 'a' compound
from each of the 25 pairs of spiking compounds at a high level and
spiked with the V compound at a low level. Three additional aliquots
of each waste were spiked with 'b' compounds at the high level and with
the 'a' compounds at a low level. Thus, six aliquots of each waste
were analyzed, a total of 24 analyses.
Spiking Method
Ten samples identified in Table 4 were spiked with both purgeable and
semivolatile compounds; candidate spiking compounds are listed in Table 2. Of
the ten samples, nine may be described as either liquids or non-liquids. The
term non-liquid refers to solid and semi-solid material, including dry powder,
moist-powder, wet filter cake, or tar. The tenth sample was a pelletized solid.
For purposes of this study, it could not be crushed or pulverized during the
spiking procedure. Thus, this material was handled using a different procedure
than that employed with either the liquid or non-liquid samples. The spiking
procedures employed for each category of material are described under the
headings below. The rationale, upon which these procedures are based, is pre-
sented as each procedure is described. Approximate quantities of semivolatile
and purgeable spiking compounds employed are shown in Tables 4 and 5. The manner
in which the quantities of spike added were calculated is presented in the
section beginning on page 17.
Spiking Liquid Samples. Approximately 1.5-kg of sample were taken for
splklngiThis will provide approximately twenty 70-gm samples to be analyzed
for semivolatile compounds, and twenty 5-gm samples to be analyzed for purge-
able compounds. The 1.5-kg portion of sample will be placed in a 3-liter,
10
-------
TABLE 3. VARIED SPIKING SCHEME
===============================================3==============================
Set of -Spiking Level, ug/m1 of Final Extract, for Given Waste Sample
Compounds(a)
A
B
C
D
E
F
G
H
I
1
40
10
50
10
60
10
10
40
10
2
10
60
40
10
50
10
60
10
10
3
10
50
10
60
40
10
50
10
60
4
10
40
10
50
10
60
40
10
50
5
60
10
10
40
10
50
10
60
40
6
50
10
60
10
10
40
10
50
10
7
40
10
50
10
60
10
10
40
10
8
10
60
40
10
50
10
60
10
10
9
10
50
10
60
40
10
50
10
60
10
10
40
10
50
10
60
40
10
50
J 50 40 10 10 10 60 50 40 10 10
============================================================3==========3======
(a)Set A = la + 2a + 3a + 4a Set B = Ib + 2b + 3b + 4b
Sec C = 5a + 6a + 7a + 8a Set D » 5b + 6b + 7b + 8b
Set E = 9a + lOa + lla + 12a Set F = 9b + lOb + lib + 12b
Set G = 13a + 14a + 15a + 16a Set H = 13b + 14b + 15b + 16b
Set I = 17a + 18a + 19a + 20a Set J = 17b + 18b + 19b + 20b
la and Ib are the two compounds 1n the first pair, 2a and 2b are the two
compounds in the second pair, etc.
2-neck round bottom flask. The flask was equipped with a 45/50 standard
taper center joint and an angled 24/40 standard taper side joint. The shaft
of a high-speed homogenizer (Tekmar model SDT-1810, or equivalent) will be
Inserted through the angled side joint. The shaft was Inserted through an
appropriately bored solid-Teflon 24/40 standard taper stopper (Fisher No.
14645-10F, or equivalent). The large diameter center joint facilitated
addition of solids to the liquid sample. Provision was also be made for
magnetic stirring.
11
-------
INS
TABLE 4. APPROXIMATE QUANTITIES OF EACH SEMI VOLATILE SPIKING COMPOUND
ADDED TO VARIOUS WASTE SAMPLES**'
Waste
Olentangy river
sediment
Machine Oil Waste
Herbicide acetone-
water
Coal-gas tar
EDC spent caustic
Cincinnati sludge
Latex paint
Ethane spent catalyst
(S-0100)
Drying bed solids
Approximate TSEC^b) Spiking Level ("High-Level")*0)
Physical gm extractable matter gm spiking compound
Description gm
i
Dry powder
Liquid
Liquid
Tar
Liquid
Wet filter cake
Semi -sol id
Moist and/or oily powder
Semi-solid, wet and/or
pasty cake
waste sample gm waste sample
0.005
1.0
0.05
0.5
0.005
0.01
0.05
0.15
0.10
Oxychlori nation spent
atalyst Pelletized solid 2 x 10-4
2.5 x 10-5
5.0 x 10-3
2.5 x 10-4
2.5 x 10-3
2.5 x 10-5
5.0 x 10-5
2.5 x 10-4
5.0 x 10-4
1.0 x 10-6
ja|See pages 93 to 95 for final spike quantities.
jb|TSEC - Total solvent extractable matter.
IC'The calculated spiking level shown here is that quantity desired for the so-called "high-level", (see
page 7). The low-level spike concentration will be approximately 20% of this value.
-------
TABLE 5. APPROXIMATE QUANTITIES OF EACH PURGEABLE SPIKING COMPOUND
ADDED TO VARIOUS WASTE SAMPLES
==============================================================================
Approximate TVC(a) Spiking Level ("High-Level
gm volatile matter gm spiking compound
Waste gm waste samplegm waste sample
EDC spent caustic 0.02 5 x 10-*
Cincinnati sludge^)
Olentangy river sediment
Ethane spent catalyst
(S-0100) -0.001 <2.5 x 10-5
S-0190
Oxychlorination spent
Catalyst
===================
- Total volatile content. The values shown are preliminary
estimates.
calculated spiking level shown here is that quantity desired for the
"high-level" spike (see page 5). The "low-level" spike concentration will
be approximately 20% of this value.
e page 95 for final spike level for these samples.
Semi volatile spiking compounds were added to the liquid at ambient
temperature in neat form. In the case of Liquid semivolatile compounds, the
volume required was calculated based on the reported or measured density, and
the material added using an appropriate syringe or micro syringe. If some
liquids proved difficult to dispense accurately in this fashion, the weight of
material added was determined by weighing the syringe before and after liquid
1s dispersed.
Solid semivolatile compounds were reduced to a fine powder in a mortar
and pestle. Weighed portions were then added directly to the liquid. In
making the addition, care was be taken to avoid depositing powder on the
flask-wall in an area that may not be swept by liquid during the subsequent
13
-------
homogenlzatlon step. If it was necessary to add portions of powder weighing
less than 10 mg, the compound was weighed into a small aluminum-foil boat,
and the solid was rinsed into the flask with methanol. The total quantity
of methanol to be added to the waste sample (for all purposes) did not exceed
30 mg/kg waste.- After each semi volatile compound was added, the homogenizer
was operated for a brief period (approximately 15 seconds). After all the
semivolatile compounds were added, the flask was securely stoppered and the
homogenizer was operated for 1-2 minutes. At the end of this period the
apparatus was immersed in an ice bath, and the sample was allowed to cool for
15-30 minutes. Preceding the sample can be expected to aid in avoiding loss
of purgeable compounds. A methanol solution containing the purgeable spiking
compounds was added to the sample, the flask was stoppered, and the mixture
stirred for 10 minutes using the magnetic stirrer.
The purgeable compounds were added in methanol solution to minimize loss
of these volatile compounds during handling and transfer. Use of a magnetic
stirrer in place of the high speed homogenizer was intended to prevent loss of
the purgeable spiking compounds. The semi volatile compounds were added in neat
form to minimize addition of solvent.
At the conclusion of the mixing period described above, a siphon-tube
equipped with a Teflon stopcock was inserted into the center joint. Portions
of spiked sample was then dispensed into vials; the stirrer was operated
throughout the period during which samples were being dispensed. As noted
above, separate portions of sample were provided for determination of purgeable
and semi volatile compounds. The 5-gm samples used for determination of purge-
able compounds were dispensed so that the vial is filled completely, and head-
space is eliminated. These vials were equipped with Teflon-lined silicone
rubber screw-cap liners. The 70-gm samples used for determination of semi-
volatile compounds were dispensed in wide-mouth jars equipped with Teflon-lined
screw caps.
The procedure was based on the assumption that some of the semivolatile
spiking compounds may have limited solubility in the waste sample, and thus
may precipitate or undergo other phase separation after the spiked sample has
been stored for some period. It was assumed that a homogeneous suspension is
generated using the high-speed homogenizer, and that this suspension remains
homogeneous for the brief Interval required to dispense the sample. Based on
these assumptions, the analyst was Instructed to re-homogenize the sample
intended for determination of semlvolatHe compounds immediately before an
aliquot is taken for analysis. The sample was provided in a jar large enough
to permit insertion of a homogenizer shaft directly Into the sample.
Problems of limited solubility and sample inhomogeneity were not antici-
pated for the non-polar purgeable spiking compounds. It was expected that
these compounds were completely miscible in the liquid samples. Thus, the
analyst was instructed to weigh and dispense the spiked sample using the same
procedures described in the protocol entitled, Method for Determination of
Purgeable Organic Compounds in Solid Waste.
14
-------
Spiking Non-Liquid Samples
Approximately 1.5-kg of each sample was spiked with both purgeable
and semlvolatlle spiking compounds. Semlvolatlle compounds were added 1n
neat form, whije purgeable compounds were added as a methanol solution.
The spiking compounds were added using the same technique described above
for spiking of liquid samples. However, in place of a flask fitted with a
high-speed homogenizer, samples were mixed using a blender. The blender
container was equipped with a cooling jacket, and the sample was cooled with
liquid nitrogen. Use of cryogenic blending served two purposes. First, loss
of volatile compounds during blending was minimized. Second, tarry or tacky
samples were cooled to a temperature at which they formed brittle solids.
Thus, it was possible to reduce these materials to a fine powder, and thereby
blend the spiking compounds uniformly Into the sample.
The first stage of the blending procedure was carried out using a 250-ml
capacity stainless steel container (Eberbach model 8590 jacketed container, or
equivalent). The second stage was carried out using a 1-gallon stainless steel
container (Waring model 2610-C container with jacketed base. The container
was modified to extend the cooling jacket over the container side-wall.)
The blending assemblies in both containers were equipped with Teflon seals and
pure carbon bearings. The entire procedure was carried out under a nitrogen
atmosphere in a glove box. This prevented contamination of the sample with
liquid oxygen or water of condensation.
The procedure was commenced by placing 50-g of sample in the 250-ml con-
tainer. The sample was cooled, after which the semlvolatlle and purgeable
spiking compounds were added. Here, compounds in methanol solution were
added first, after which those in neat form were added. After each addition,
the blender was operated briefly (approximately 15 seconds). After all com-
pounds were added, the blender was operated for 15 minutes. At the conclusion
of this period, an additional 50-g of precooled sample was added to the con-
tainer, and the blender was again operated for 15 minutes. This procedure
was repeated using a 50-g portion of sample to give a total of 150-g of spiked
sample. The spiked sample was then transferred to the 1-gallon container. At
this point, additional portions of sample were added 1n 100-g increments until a
total of 1.5-kg of sample was added and blended.
In order to aid in complete transfer of material from the small to the
large container, the 100-g portions of sample were first placed in the small
container, blended for 1 minute, and transferred to the large container. After
a total of 1.5-kg of sample was added to the large container, the blender was
operated for a period of 10 minutes.
It was assumed that spiked samples prepared according to this procedure
were homogeneous with respect to both purgeable and semivolatile compounds, and
that the samples remained homogeneous during the period between spiking and
analysis. Thus, the analyst was Instructed to transfer and weigh the spiked
samples using the same procedures described in the Method for Determination of
Purgeable Organic Compounds in Solid Waste and the Method for Determination of
Semivolatile Compounds in Solid Waste.
15
-------
At the conclusion of the blending period, samples were transferred to
vials equipped with Teflon-lined silicone rubber screw cap liners. Approxi-
mately 70-g samples were provided for determination of semi volatile compounds
and approximately 5-g samples were provided for determination of purgeable
compounds. The 70-g and 5-g portions were packaged separately to minimize
sample cross contamination and to maintain sample integrity.
Spiking of Pelletized Solids
The physical structure of the pelletized solid (oxychlorination spent
catalyst) should not be modified during the spiking procedure. Thus, it was
not feasible to prepare homogeneous samples of this material. In this case,
individual spiked samples were prepared. Both purgeable and semivolatile
compounds were added in methanol solution.
Approximately 1-g portions of the solid to be used for the determination
of purgeable compounds were transfered to tared glass ampoules and weighed
to ±1 mg; the sample weights were reported to the analyst. Methanol solu-
tions were added directly onto the pellets in the ampoule using appropriate
micro syringes. The ampoule was then cooled in a Dry-ice/methanol bath,
and flame sealed. For determination of purgeable compounds, the analyst was
instructed to place the ampule directly into a vessel containing polyethy-
lene glycol (PEG) and then to crush the thin-walled ampoule in situ (see Method
for Determination of Purgeable.Organic Compounds in Solid Vlastes).
For determination of semivolatile compounds 20-g portions of the solid
were transfered to glass vials having Teflon-lined screw caps. The methanol
spiking solutions were added directly onto the pellets. The analyst was in-
structed to transfer the entire sample to the extraction vessel (see Method
for the Determination of Semivolatile Organic Compounds in Solid Waste) and
to rinse the vial several times with methylene chloride.
Calculating Quantities of Spiking Compounds to be Added to Waste Samples
Approximate quantities of semivolatile and purgeable spiking compounds
added to various waste samples are shown in Tables 4 and 5. The values shown
were calculated based on the following parameters:
1. The final quantity o^ spiking compound required for GC/MS analysis.
2. The estimated concentration of total solvent extractable matter (TSEC)
or total volatile content (TVC) in the sample.
Using these parameters, 1t is possible to calculate the quantity of spike
needed per unit mass of waste material. Consider first, the calculation as
applied to spiking with semi volatile compounds. It is desired that approxi-
mately 50-ng of each spiking compound be present in the portion of sample ex-
tract taken for GC/MS analysis. This portion typically contains approximately
16
-------
10-iig of extractable matter. Thus, the sample must be spiked to obtain 5-ng
of spiking compound per microgram of extractable matter, or .005-g of spiking
compound per g of extractable matter. If this value is multiplied by the value
for TSEC expressed in g of extractable matter per g of waste sample, a value of
g of semivolatfle spiking compound per g of waste sample will be obtained.
Consider next the calculation as applied to spiking with purgeable com-
pounds. It is desired that approximately 250-ng of each spiking compound be
present in the portion of sample extract (in PEG) taken for the purge- and trap
GC/MS analysis procedure. Typically, this portion of the PEG extract contains
approximately 10-ug of volatile matter. Thus, the sample must be spiked to
obtain 25-ng of each spiking compound per microgram of volatile matter, or
0.025-g of spiking compound per g of volatile matter. If this value is multi-
plied by the value for TVC expressed in g of volatile matter per g of sample,
a value for g of purgeable spiking compound per g of waste sample will be
obtained.
In both of the examples illustrated above, the quantity of spiking compound
calculated is that quantity desired for the so-called "high-level" spike con-
centration (see page 7). The "low-level" spike concentration will be approxi-
mately 20% of this value.
Quality Assurance
A quality assurance plan which will be a part of the complete protocol for
the inter!aboratory study (Phase III) has been developed. That plan includes
quality control performance criteria that are part of the methodologies de-
scribed in this report. The performance criteria are based on precision studies
conducted for:
• Retention time and relative retention time
• Decafluorotriphenylphosphine (DFTPP) ion abundance
• Response factors
t Internal standard response.
Absolute Retention Time
Retention time repeatability for two sets of 26 representative compounds
is shown in Tables 6 to 9. The typical standard deviation is about 0.5 minutes
with a range of 0.1 to 0.6.
Relative Retention Time
The absolute retention time data for Set 3 were converted to relative
retention time data to determine to what extent the retention time window used
for the reverse search routine can be decreased by using relative retention
times. The retention times of 15 compounds detected in seven runs involving a
17
-------
TABLE 6. REPEATABILITY OF GC/MS ABSOLUTE RETENTION TIMES, SET 2
=================================================================:
Retention Time, M1n:Sec. For Given Compound^)
Run No.
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
BCEE
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
:43
:46
:22
:20
:23
:46
:22
:20
:21
:19
:21
:46
:21
:30
:16
:30
:55
:49
:28
:26
CP
12:24
12:35
13:01
12:23
12:17
12:51
12:51
12:25
NDPA
16
17
16
16
16
17
16
16
16
17
16
16
16
17
17
16
16
16
:40
: 12
:48
:16
:49
:13
:48
:16
:46
:13
:48
:53
:42
:02
:15
:44
:18
:26
NB
17:18
16:24
16:00
16:47
16:01
16:26
16:00
16:54
16:54
16:49
16:52
17:19
16:54
17:13
16:48
17:18
16:04
17:21
17:23
18:44
16:56
18:11
16:29
12:22
13
:04
12:22
16
:16
16:54
ISO
18:44
18:20
18:20
18:20
18:39
18:19
18:45
18:21
18:15
18:47
18:48
18:55
18:22
17:55
18:52
18:49
18:20
DMP
19:48
19:49
19:23
19:23
19:23
20:28
19:21
20:28
19:48
19:24
19:17
18:55
18:57
19:50
19:51
19:25
19:23
HCBu
22:36
22:36
22:11
22:11
22:12
22:37
22:11
22:11
22:12
22:11
22:10
22:12
22:37
22:12
21:53
21:34
22:06
21:41
21:46
22:38
22:39
22:17
22:16
22:12
22:17
22:14
22:11
22:11
DCA
30:18
31.08
29:53
29:58
29:51
29:54
29:50
29:54
30:17
29:50
29:54
29:49
29:57
30:28
28:50
DNT
33:32
33:47
34:41
33:06
33:13
33:59
34:31
33:06
33:06
33:04
33:32
33:06
34:53
33:01
33:34
33:36
33:09
33:08
Mean 12:33 12:38 16:56 16:72 18:29 19:34 22:08 29:70 33:40
0.20 0.26 0.36 0.63 0.24 0.45 0.29 0.56 0.50
==============================================================================
(a)BCEE=B1s(2-ch1oroethyl) ether; CP=2-Chlorophenol; NDPA=N-Nitrosodi-
propylamine; NB=N1trobenzene; ISO=Isophorone; DMP=2,3-Dimethylphenol;
HCBu=Hexachlorobutad1ene; DCA=3,4-Dichloroaniline, DMT-2,4-D1nitrotoluene.
(b)standard Deviation.
18
-------
TABLE 6. (Continued)
Retention Time, M1n:Sec. For Given Compound(a)
Run No.
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
Mean
SD(b)
HEB
38:27
38:24
38:00
38:00
38:01
38:27
37:59
38:00
38:00
38:00
37:58
38:01
38:26
38:26
38:03
38:07
37:58
37:56
38:07
37:59
38:33
38:29
38:26
38:04
38:04
38:00
38:05
38:02
37:58
37:59
37:99
0.25
HCB
38:49
38:46
38:22
38:22
38:23
38:49
38:21
38:22
38:22
38:19
38:22
38:47
38:33
38:55
38:42
38:26
38:54
38:44
39:19
38:59
38:51
38:28
38:27
38:24
38:28
38:25
38:22
38:23
38:37
0.21
PCP
40:07
40:04
39:40
39:39
39:37
39:36
39:35
39:31
39:34
40:00
39:35
39:43
39:28
39:40
40:07
40:01
39:39
39:39
39:40
39:34
39:36
39:53
0.30
DFTPP
43:40
43:37
43:12
43:12
43:14
43:40
43:12
43:12
43:13
43:14
43:11
43:13
43:38
43:38
43:14
43:16
43:07
43:07
43:16
43:07
43:42
43:39
43:39
43:16
43:16
43:13
43:17
43:14
43:11
43:12
43:20
0.12
F
47:55
47:52
47:28
47:28
47:28
47:29
47:31
47:26
47:28
47:54
47:28
47:32
46:39
47:24
46:49
46:40
47:16
47:58
47:58
48:55
49:01
47:31
48:56
48:59
48:48
47:30
47:49
0.66
DDE
50:34
50:31
50:06
50:07
50:06
50:07
50:08
50:08
50:05
50:08
50:33
50:08
50:10
50:01
50:09
50:37
50:34
50:36
50:12
50:09
50:13
50:06
50:08
50:15
0.12
DCB
56:50
56:22
56:23
56:26
56:25
56:52
56:25
56:25
57:18
56:44
56:41
0.29
33333333
BKF
62:57
62:54
62:28
62:30
62:31
62:27
62:56
62:30
62:28
63:18
63:06
62:38
62:36
62:49
0.30
3333333333
(a)HEB=Hexaethylbenzene; HCB=Hexach1orobenzene; PCP=Pentachloropheno1;
DFTPP=Decafluorotr1phenylphosph1ne; F=Fluoranthene; DDE=4,4'-DDE;
DCB=3,3'-D1ch1orobenz1d1ne; BkF=Benzo(k)fluoranthene.
(b'Standard deviation.
19
-------
TABLE 7. REPEATABILITY OF GC/MS ABSOLUTE RETENTION TIMES
FOR DEUTERATED INTERNAL STANDARDS, SET 2
=============================================================
Retention Time, MinrSec. For Given Compound^3)
Run No.
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
Mean
BB
9:
9:
9:
9:
9:
9:
9:
9:
9:
9:
9:
9:
9:
9:
9:
9:
9:
9:
9:
9:
10:
10:
10:
9:
9:
9:
9:
9:
55
57
34
31
35
58
33
31
33
32
30
33
57
59
33
41
36
28
32
37
05
02
00
41
41
37
41
37
9:35
9:
33
9:41
Anil
12:18
12:21
11:58
11:51
11:58
12:23
11:58
11:59
11:59
11:55
11:58
11:52
12:20
12:17
11:49
12:29
11:55
11:43
12:04
11:56
12:26
12:15
12:24
11:59
11:57
11:56
11:58
11:56
11:52
11:52
12:02
Phol
12:
12:
11:
11:
11:
12:
11:
11:
12:
11:
12:
11:
12:
12:
11:
11:
11:
11:
11:
11:
12:
12:
12:
11:
11:
11:
11:
11:
16
18
55
53
55
19
54
52
06
53
03
55
18
22
59
57
52
55
59
52
23
17
21
59
58
55
59
57
11:53
11:
52
12:01
NB
17:12
17:12
16:48
16:47
16:49
17:13
16:47
16:47
16:48
16:47
16:46
16:48
17:12
17:13
16:48
17:01
16:50
16:42
17:02
16:56
17:25
17:15
17:16
16:54
16:54
16:50
16:54
16:52
16:48
16:47
16:57
Naph
21:14
21:14
20:50
20:49
20:51
21:16
20:50
20:49
20:49
20:48
20:46
20:49
21:14
21:14
20:51
21:20
21:13
20:35
21:23
21:25
21:37
21:16
21:17
20:54
20:54
20:51
20:54
20:52
20:48
20:49
21:01
DBB
21:27
21:28
21:03
21:02
21:04
21:30
21:04
21:02
21:02
21:02
21:00
21:03
21:27
21:27
21:03
21:07
21:11
20:57
21:39
21:02
21:35
21:30
21:31
21:08
21:08
21:04
21:09
21:06
21:02
21:02
21:11
NN
35:41
35:38
35:14
35:15
35:15
35:15
35:13
35:14
35:14
35:14
35:12
35:15
35:40
35:40
35:15
35:20
35:11
35:09
35:20
35:13
35:48
35:43
35:43
35:20
35:20
35:17
35:21
35:18
35:15
35:15
35:22
Phen Chry
40:36
40:51
40:26
40:09
40:27
40:54
40:08
40:09
40:10
40:28
40:07
40:28
40:53
40:35
40:10
40:17
40:07
40:05
40:17
40:09
40:44
40:39
40:38
40:15
40:15
40:11
40:16
40:13
40:09
40:10
40:22
56:41
56:38
56:13
56:13
56:14
56:42
56:13
56:14
56:15
56:17
56:12
56:16
56:41
56:40
56:15
56:17
56:09
56:11
56:19
56:10
56:49
56:49
56:46
56:25
56:25
56:19
56:26
56:23
56:19
56:19
56:24
0:11 0:13 0:11 0:12 0:16 0:13 0:12 0:15 0:13
============================================================================:
(a)BB=D5-Bromobenzene; Anil=D5-Aniline; Phol=D5-Pheno1;
NB=D5-Nitrobenzene; Naph=D8-Naphthalene; DBB=D4-l,4-Dibromobenzene;
NN=D7-l-Nitronaphthalene; Phen=D10-Phenanthrene; Chry=D12-Chrysene
'^Standard deviation.
ND = Not detected.
20
-------
TABLE 8. REPEATABILITY OF GC/MS ABSOLUTE RETENTION TIMES, SET 3
3S3===========================================================================
Retention Time, M1n:Sec. For Given Compound^)
Run No.
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
Mean
Std. Dev.
% RSD
Range
BCEE
11.
ND
ND
ND
10.
NO
ND
ND
ND
10.
ND
ND
ND
11.
ND
ND
ND
11.
11.
11.
11.
11.
11.
11.
10.
11.
11.
0.
1.
0.
05
87
97
17
20
13
13
13
13
17
13
72
15
07
14
3
48
CP
11.12
ND
ND
ND
10.93
ND
ND
ND
ND
11.03
ND
ND
ND
11.25
ND
ND
ND
11.27
11.29
11.20
11.18
11.22
11.22
11.20
10.78
11.22
11.14
0.14
1.3
0.49
NDPA
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
• ND
ND
ND
15.
15.
15.
ND
15.
15.
14.
ND
14.
0.
1.
02
02
03
03
02
62
96
17
1
0.41
NB
15.52
ND
ND
ND
15.42
ND
ND
ND
ND
15.52
ND
ND
ND
15.62
ND
ND
ND
15.65
15.58
15.58
15.60
15.58
15.60
15.58
15.17
15.60
15.54
0.13
0.81
0.48
ISO
17.03
ND
ND
ND
16.93
ND
ND
ND
ND
17.03
ND
ND
ND
17.10
ND
ND
ND
17.13
17.07
17.08
17.08
17.07
17.08
17.08
16.67
17.08
17.03
0.12
0.70
0.46
DMP
18.17
ND
ND
ND
18.12
ND
ND
ND
ND
18.20
ND
ND
ND
18.25
ND
ND
ND
18.28
18.23
18.23
18.23
18.22
18.25
18.23
17.83
18.23
18.19
0.12
0.63
0.45
HCBu
20.98
ND
ND
ND
20.93
ND
ND
ND
ND
21.03
ND
ND
ND
21.07
ND
ND
ND
21.08
21.07
21.10
21.08
21.02
21.00
21.02
20.62
21.02
21.00
0.12
0.59
0.48
DCA
28.47
ND
ND
ND
28.45
ND
ND
ND
ND
28.53
ND
ND
ND
21.52
ND
ND
ND
28.55
28.50
28.50
28.50
28.48
28.50
28.52
28.10
28.50
28.47
0.11
0.40
0.45
DNT
31.78
ND
ND
ND
31.78
ND
ND
ND
ND
31.85
ND
ND
ND
31.87
ND
ND
ND
31.87
31.85
31.83
31.83
31.78
31.82
31.83
31.42
31.82
31.79
0.12
0.37
0.45
==============================================================================
(a)BCEE=B1s(2-chloroethy1) ether; CP=2-Chlorophenol; NDPA=N-N1trosod1-
propylamlne; NB=Nitrobenzene; ISO=Isophorone; DMP=2,3-D1methylphenol;
HCBu=Hexachlorobutad1ene; DCA=3,4-D1chloroan1Hne, DNT-2,4-D1n1trotoluene.
b Standard Deviation. ND=Not detected.
(c'Values 1n parentheses are those obtained 1f the data from Run 25 1s
excluded.
21
-------
================
TABLE 8. (Continued)
:==============================================:
Retention Time, MinrSec. For Given Compound(a)
Run No.
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
AA
26.77
ND
ND
ND
26.75
ND
ND
ND
ND
26.83
ND
ND
ND
26.83
ND
ND
ND
26.85
26.82
26.80
26.80
26.78
26.78
26.78
26.38
26.80
NA
33.90
ND
ND
ND
33.88
ND
ND
ND
ND
33.97
ND
ND
ND
33.97
ND
ND
ND
33.97
33.93
33.83
33.87
33.88
33.92
33.95
33.52
33.92
HEB
36:82
36:80
36:83
36:83
36:80
36:82
36:83
36:85
36:50
36:88
36:90
36:90
36:90
36:90
36:90
. 37:05
36:90
36:88
36:82
36:80
36:80
36:78
36:82
36:85
36:45
36:82
PCP
38.23
ND
ND
ND
38.22
ND
ND
ND
ND
38.30
ND
ND
ND
38.32
ND
ND
ND
38.30
38.25
38.23
38.22
38.20
38.23
38.27
37.87
38.23
DFTPP
42:05
42:02
42:05
42:07
42:03
42:03
42:05
42:08
41:72
42:10
42:12
42:13
42:13
42:13
42:13
42:20
42:13
42:10
ND
ND
ND
42:02
ND
ND
ND
42:03
ANQ
43.73
ND
ND
ND
43.77
ND
ND
ND
ND
43.87
ND
ND
ND
43.87
ND
ND
ND
43.83
43.80
43.77
43.77
43.77
43.77
43.82
43.42
43.80
F
45.98
ND
ND
ND
45.97
ND
ND
ND
ND
46.05
ND
ND
ND
46.07
ND
ND
ND
46.03
45.97
45.95
45.93
45.93
45.95
46.02
45.60
45.97
DDE
48.73
ND
ND
ND
48.70
ND
ND
ND
ND
48.80
ND
ND
ND
48.80
ND
ND
ND
48.77
48.68
48.67
48.67
48.67
48.67
48.75
48.33
48.70
Avg. 26.77 33.89 36:82 38.22 42:07 43.77 45.96 48.69
SD(b) 0.12 0.12 0.12 0.11 0.095 0.11 0.12 0.12
% RSD 0.45 0.35 0.32 0.29 0.23 0.26 0.25 0.24
Range 0.47 0.45 0.45 0.45 0.48 0.45 0.47 0.47
===========================================================================:
(a)AA=Acetanilide; NA=4-N1troaniline; HEB=Hexaethylbenzene; PCP=Pentachloro-
phenol; DFTPP=Decafluorotriphenylphosphine; ANQ=Anthraquinone;
F=Fluoranthene; DDE=4,4'-DDE;
(b)standard deviation. ND=Not Detected.
22
-------
TABLE 9. REPEATABILITY OF GC/MS ABSOLUTE RETENTION TIMES
FOR OEUTERATED INTERNAL STANDARDS, SET 3
==============================================================================
Retention Time, M1n:Sec. For Given Compound^3)
Run No.
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
Average
SD(b)
%RSD
Range
========
BB
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
0
1
.37
.10
.27
.17
.12
.47
.43
.27
.12
.22
.50
.53
.53
.52
.48
.60
.43
.55
.45
.45
.45
.45
.50
.47
.05
.48
.38
.16
.9
ANIL
10.52
10.32
10.43
10.35
10.30
10.58
10.57
10.47
10.25
10.42
10.63
10.67
10.65
10.63
10.62
NO 1C)
10.60
10.67
10.58
10.58
10.60
10.58
10.62
10.60
10.18
10.62
10.52
0.14
1.3
PHOL
10.70
10.50
10.63
10.53
10.50
10.77
10.75
10.63
10.42
10.60
10.80
10.83
10.82
10.82
10.80
11.00
10.80
10.83
10.78
10.78
10.78
10.80
10.83
10.78
10.37
10.82
10.72
0.15
1.4
NB
15.42
15.33
15.38
15.35
15.32
15.47
15.47
15.43
15.13
15.42
15.52
15.53
15.52
15.52
15.50
15.57
15.52
15.55
15.48
15.47
15.33
15.48
15.50
15.48
15.07
15.50
15.43
0.12
0.78
0.55 0.49 0.46 0.50
===============================3===
NAPH
19.50
19.47
19.48
19.47
19.43
19.52
19.55
19.53
19.20
19.53
19.58
19.60
19.58
19.58
19.57
19.65
19.60
19.60
19.55
19.53
19.55
19.53
19.55
19.53
19.13
19.55
19.51
0.11
0.58
DBB
19.72
19.70
19.73
19.72
19.67
19.75
19.78
19.77
19.43
19.75
19.80
19.83
19.82
19.82
19.80
19.85
19.87
19.83
19.78
19.78
19.78
19.77
19.78
19.60
19.37
19.78
19.74
0.12
0.59
NN
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
34
33
33
33
33
33
33
33
33
33
33
0
0
0.52 0.50 0
=================
.90
.92
.93
.93
.88
.90
.92
.93
.58
.97
.97
.98
.98
.97
.98
.83
.00
.97
.93
.83
.93
.88
.90
.93
.53
.92
.90
.11
.33
.47
==•==
PHEN
38.77
38.75
38.77
38.78
38.75
38.75
38.80
38.80
38.45
38.83
38.85
38.87
38.85
38.85
38.85
39.00
38.85
38.83
38.77
38.75
38.75
38.75
38.77
38.80
38.40
38.77
38.78
0.12
0.30
0.60
=======
CHRY
54.62
54.55
54.62
54.62
54.57
54.57
54.62
54.65
54.28
54.67
54.70
54.72
54.70
54.67
54.70
54.67
54.70
54.62
54.53
54.52
54.50
54.52
54.53
54.62
54.22
54.55
54.59
0.12
0.22
0.50
=======
An1l=D5-Annine; Phol=Dr-Phenol ; NE=D
benzene; Naph=D8-Naphthalene; DBB=D4-l,4-D16romobenzene;
NN=D7-l-N1troanphthalene; Phen=Dio-Phenanthrene; Chry=Di2-Chrysene.
bStandard Deviation
detected
23
-------
standard solution and six runs Involving solid waste extracts spiked with
the compounds were calculated relative to each of eight different deuterated
Internal standards. The mean relative retention time (RRT), standard devia-
tion (SO), percent relative standard deviation, and 3-times-standard-
deviation window in seconds were calculated. These data are presented in
Tables 10 to 24. As expected the 3x SD window in seconds was generally con-
siderably lower when the RRT was closer to 1.0. The 3x SD windows for all
cases in which the RRT falls within the range of 0.85 to 1.25 are given in
Table 25. If no internal standard gave an RRT within that range the Internal
standard that gave an RRT closest to that range was used.
It was encouraging to note that for this routine set of runs performed
over a period of several days, the 3x SD windows were frequently less than 5
seconds and 1n all cases were less than 15 seconds. The use of multiple in-
ternal standards in conjunction with FSCC GC/MS analyses is therefore strongly
recommended for the Inter!aboratory study. By using RRT data to narrow the
window required for the reverse search routine, the time required for the
searches can be decreased and the number of compounds that can be searched
during a GC/MS run can be increased.
DFTPP Ion Abundances
The repeatability of DFTPP ion abundances were determined for two sets of
GC/MS runs to provide the basis for control limits. The two sets of runs gave
relative standard deviations for the major ions of 5 to 20% as shown in Tables
26 and 27.
Response Factors
Response factor precision for Internal standards and compounds are shown
in Tables 28 and 29. The relative standard deviations are generally within
less than 15% and average about 10%. These data will also help establish
performance criteria for the Inter!aboratory efforts.
Internal Standard Response
The repeatability of response of an internal standard is useful to monitor
the performance of the analytical system. If there are gross changes in an
Internal standard response, confidence in that analytical run is reduced.
Gross changes may indicate that the internal standards were added at improper
levels or that the GC/MS system is not functioning properly. Monitoring the
response is another flag to Indicate system performance.
Internal standard (DIO-Phenanthrene) response was measured over several
routine analysis runs. The standard deviation was found to be 10% (RSD).
Applying the 3x standard deviation convention, the internal standard can be
expected to be within a range of a factor of 2. Thus, if the average response
is 450,000 the performance range is 300,000 to 600,000. Supporting data are
shown in Table 30.
24
-------
TABLE 10. RELATIVE RETENTION TIME DATA FOR BIS(2-CHLOROETHYL) ETHER
Internal Sta'ndard
Used for Calculation
Ds-Bromobenzene
Ds-Ani 1 i ne
Ds-Phenol
Ds-Nitrobenzene
Dg-Naphthalene
04-! , 4-Di bromobenzene
Dj-l-Nitronaphthalene
Dio-Phenanthrene
Di2-Chrysene
Relative
Retention
Time (RRT)
1.3198
1.0515
1.0327
0.7178
0.5676
0.5613
0.3267
0.2857
0.2030
TABLE 11. RELATIVE RETENTION
Internal Standard
Used for Calculation
Ds-Bromobenzene
Ds-Ani line
Ds-Phenol
Ds-Nitrobenzene
Dg-Naphthalene
04-1 , 4-Di bromobenzene
Dy-l-Nitronaphthalene
DiQ-Phenanthrene
Di2-Chrysene
Relative
Retention
Time (RRT)
1.3286
1.0585
1.0396
0.7226
0.5715
0.5651
0.3289
0.2876
0.2043
Standard
Deviation
0.0092
0.0016
0.0015
0.0045
0.0042
0.0047
0.0034
0.0030
0.0023
TIME DATA FOR
Standard
Deviation
0.0092
0.0032
0.0027
0.0050
0.0048
0.0051
0.0037
0.0033
0.0025
Percent
Relative
Standard
Deviation
0.69
0.15
0.14
0.63
0.75
0.83
1.04
1.06
1.15
3x SO
Window,
sec.
18
3
3
9
15
17
21
21
23
2-CHLOROPHENOL
Percent
Relative
Standard
Deviation
0.69
0.30
0.26
0.70
0.84
0.90
1.11
1.14
1.23
3x SD
Window,
sec.
14
6
5
18
17
31
22
23
5
25
-------
TABLE 12. RELATIVE RETENTION TIME DATA FOR NITROBENZENE
Internal Standard
Used for Calculation
D5-Bromobenzene
D5-Anil1ne
Ds-Phenol
D5-Nitrobenzene
Dg-Naphthalene
04-1 ,4-Di bromobenzene
Dy-l-Nitronaphthal ene
DiQ-Phenanthrene
Di2"Chrysene
Relative
Retention
Time (RRT)
1.8524
1.4758
1.4494
1.0074
0.7953
0.7885
0.4586
0.4010
0.2849
Standard
Deviation
0.0229
0.0098
0.0094
0.0031
0.0050
0.0030
0.0024
0.0023
0.0019
TABLE 13. RELATIVE RETENTION TIME DATA
Internal Standard
Used for Calculation
Ds-Bromobenzene
Ds-Aniline
Ds-Phenol
Ds-Nitrobenzene
Ds-Naphthalene
04-! ,4-Di bromobenzene
Dy-l-Nitronaphthalene
DiQ-Phenanthrene
Dig-Chrysene
Relative
Retention
Time (RRT)
2.0304
1.6168
1.5887
1.1042
0.8733
0.8635
0.5026
0.4395
0.3122
Standard
Deviation
0.0272
0.0131
0.0121
0.0033
0.0011
0.0026
0.0021
0.0020
0.0017
Percent
Relative
Standard
Deviation
1.23
0.66
0.65
0.30
0.63
0.38
0.53
0.57
0.66
FOR ISOPHORONE
Percent
Relative
Standard
Deviation
1.34
1.67
0.76
0.30
0.12
0.30
0.42
0.45
0.55
3x SD
Window,
sec.
64
27
26
9
11
8
11
16
19
3x SD
Window,
sec.
52
25
23
10
3
8
13
14
17
26
-------
TABLE 14. RELATIVE RETENTION TIME DATA FOR 2,4-DIMETHYLPHENOL
Internal Standard
Used for Calculation
Ds-Bromobenzene
Ds-Aniline
D5-Phenol
D5-Nitrobenzene
Da-Naphthalene
D4-l,4-Dibromobenzene
Dy-1-Nitronaphthalene
Dio-Phenanthrene
Di2-Chrysene
TABLE 15. RELATIVE
Internal Standard
Used for Calculation
Ds-Bromobenzene
Ds-Ani 1 i ne
Ds-Phenol
Ds-Nitrobenzene
Dg-Naphthalene
04-1 ,4-Di bromobenzene
Dy-l-Ni t ronaphthal ene
Dio-Phenanthrene
Di2-Chrysene
Relative
Retention
Time (RRT)
2.1683
1.7282
1.6961
1.1792
0.9326
0.9221
0.5368
0.4694
0.3335
RETENTION
Relative
Retention
Time (RRT)
2.5036
1.9945
1.9589
1.3615
1.0768
1.0647
0.6197
0.5419
0.3850
Standard
Deviation
0.0309
0.0149
0.0143
0.0038
0.0006
0.0024
0.0019
0.0018
0.0016
TIME DATA FOR
Standard
Deviation
0.0037
0.0187
0.018
0.0059
0.0016
0.0027
0.0020
0.0019
0.0017
Percent
Relative
Standard
Deviation
1.42
0.86
0.84
0.33
0.06
0.26
0.35
0.38
0.49
3x SD
Window,
sec.
47
28
28
12
2
8
11
13
16
HEXACHLOROBUTADIENE
Percent
Relative
Standard
Deviation
1.50
0.94
0.92
0.43
0.15
0.25
0.32 .
0.35
0.45
3x SD
Window,
sec.
57
35
35
16
6
10
12
13
17
27
-------
TABLE 16. RELATIVE RETENTION TIME DATA FOR ACETANILIDE
Relative
Internal Standard Retention
Used for Calculation Time (RRT)
Ds-Bromobenzene
Ds-Aniline
Ds-Phenol
Ds-Nitrobenzene
Ds-Maphthalene
D4-l,4-Dibromobenzene
Dy-l-Nitronaphthalene
DiQ-Phenanthrene
Di2-Chrysene
TABLE 17. RELATIVE
Internal Standard
Used for Calculation
Ds-Bromobenzene
Ds-Aniline
Ds-Phenol
Ds-Nitrobenzene
Ds-Naphthalene
D4-l,4-Dibromobenzene
Dy-l-Ni t ronaphthal ene
DiQ-Phenanthrene
Di2-Chrysene
3.1909
2.5421
2.4967
1.7352
1.3723
1.3570
0.7899
0.6907
0.4907
RETENTION
Relative
Retention
Time (RRT)
3.3941
2.7039
2.6556
1.8457
1.4597
1.4434
0.8402
0.7347
0.5216
Standard
Deviation
0.0522
0.0274
0.0264
0.0083
0.0027
0.0038
0.0012
0.0012
0.0014
TIME DATA FOR
Standard
Deviation
0.0562
0.0247
0.0286
0.0092
0.0034
0.0048
0.0009
0.0010
0.0024
Percent
Relative
Standard
Deviation
1.64
2.54
1.06
0.48
0.20
0.28
0.15
0.18
0.28
3x SD
Window,
sec.
79
52
51
23
10
18
7
4
14
3,4-DICHLOROANILINE
Percent
Relative
Standard
Deviation
1.67
1.10
1.08
0.50
0.24
0.33
0.11
0.13
0.46
3x SD
Window,
sec.
85
56
55
26
12
17
5
7
24
28
-------
TABLE 18. RELATIVE RETENTION TIME DATA FOR 2,4-DINITROTOLUENE
Internal Sta'ndard
Used for Calculation
Ds-Bromobenzene
Ds-Aniline
D5-Phenol
Ds-Nitrobenzene
Dg-Naphthalene
0^-1 ,4-Di bromobenzene
Dy-l-Nitronaphthalene
DiQ-Phenanthrene
Di2-Chrysene
Relative
Retention
Time (RRT)
3.8058
3.0295
2.9755
2.0657
1.6329
1.6119
0.9382
0.8204
0.5829
TABLE 19. RELATIVE RETENTION
Internal Standard
Used for Calculation
Ds-Bromobenzene
Ds-Aniline
Ds-Phenol
Ds-Nitrobenzene
Dg-Naphthalene
04-!, 4-Di bromobenzene
Dy-l-Nitronaphthalene
DiQ-Phenanthrene
Dip-Chrysene
Relative
Retention
Time (RRT)
4.3873
3.4952
3.4909
2.3858
1.8869
1.8657
1.0860
0.9496
0.6747
Standard
Deviation
0.0767
0.0440
0.0424
0.0174
0.0101
0.0055
0.0009
0.0009
0.0012
TIME DATA FOR
Standard
Deviation
0.0764
0.0416
0.0439
0.0139
0.0065
0.0075
0.0008
0.0002
0.0008
Percent
Relative
Standard
Deviation
2.01
1.45
1.42
0.84
0.62
0.34
0.10
0.11
0.21
3x SD
Uindow,
sec.
116
83
82
48
36
19
5
5
12
HEXAETHYLBENZENE
Percent
Relative
Standard
Deviation
1.74
1.19
1.26
0.58
0.34
0.40
0.07 .
0.02
0.12
3x SD
Window,
sec.
145
80
122
49
23
32
5
1
8
29
-------
TABLE 20. RELATIVE RETENTION TIME DATA FOR PENTACHLOROPHENOL
Internal Sta'ndard
Used for Calculation
Ds-Bromobenzene
D5-Aniline
Ds-Phenol
Ds-Nitrobenzene
Ds-Naphthalene
04-!, 4-Di bromobenzene
Dy-l-Nitronaphthalene
Dig-Phenanthrene
Di2-Chrysene
Relative
Retention
Time (RRT)
4.5565
3.6300
3.5651
2.4778
1.9596
1.9377
1.1279
0.9863
0.7007
TABLE 21. RELATIVE RETENTION TIME
Internal Standard
Used for Calculation
D5-Bromobenzene
Ds-Aniline
D5-Phenol
Ds-Nitrobenzene
Da-Naphthalene
04-1 , 4-Di bromobenzene
Dy-l-Nitronaphthalene
DiQ-Phenanthrene
Di2-Chrysene
Relative
Retention
Time (RRT)
5.0212
3.9986
3.9255
2.7234
2.1549
2.1289
1.2402
1.0843
0.7702
Standard
Deviation
0.0797
0.0434
0.0420
0.0146
0.0069
0.0080
0.0010
0.0002
0.0008
Percent
Relative
Standard
Deviation
1.75
1.20
1.18
0.59
0.35
0.41
0.08
0.02
0.11
3x SD
Window,
sec.
120
82
81
41
24
28
6
2
8
DATA FOR DECAFLUOROTRIPHENYLPHOSPHINE
Standard
Deviation
0.0921
0.0469
0.0489
0.0136
0.0066
0.0059
0.0019
0.0006
0.0006
Percent
Relative
Standard
Deviation
1.83
1.17
1.24
0.50
0.30
0.28
0.15
0.05
0.08
3x SD
Window,
sec.
139
89
94
38
23
21
11
5
6
30
-------
TABLE 22. RELATIVE RETENTION TIME DATA FOR ANTHRAQUINONE
Internal Standard
Used for Calculation
Ds-Bromobenzene
Ds-Aniline
Ds-Phenol
Ds-Nitrobenzene
Ds-Naphthalene
D4-l,4-Dibromobenzene
Dy-l-Nitronaphthalene
Dio-Phenanthrene
Di2-Chrysene
Relative
Retention
Time (RRT)
5.2179
4.1569
4.0826
2.8374
2.2441
2.2189
1.2916
1.1294
0.8024
TABLE 23. RELATIVE RETENTION
Internal Standard
Used for Calculation
Ds-Bromobenzene
Ds-Aniline
Ds-Phenol
Ds-Nltrobenzene
Dg-Naphthalene
D4-l,4-Dibromobenzene
D7-l-Nitronaphthal ene
DiQ-Phenanthrene
Di2-Chrysene
Relative
Retention
Time (RRT)
5.4786
4.3646
4.2866
2.9792
2.3566
2.3298
1.3560
1.1858
0.8424
Standard
Deviation
0.0927
0.0509
0.0491
0.0176
0.0087
0.0097
0.0014
0.0006
0.0008
TIME DATA
Standard
Deviation
0.0979
0.0540
0.0524
0.0188
0.0095
0.0106
0.0017
0.0006
0.0005
Percent
Relative
Standard
Deviation
1.78
1.23
1.20
0.62
0.39
0.44
0.11
0.06
0.10
FOR FLUORANTHENE
Percent
Relative
Standard
Deviation
1.79
1.24
1.22
0.63
0.40
0.41
0.13
0.05
0.06
3x SD
Window,
sec.
140
96
95
49
31
34
9
4
8
3x SD
Window,
sec.
148
102
101
52
33
38
11
4
5
31
-------
TABLE 24.
.
Internal Standard
Used for Calculation
Ds-Bromobenzene
05-Aniline
Ds-Phenol
Ds-Nitrobenzene
Dg-Naphthalene
D4-l,4-Dibromobenzene
Dy-l-NHronaphthalene
Djo-Phenanthrene
Di 9— Chrysene
RELATIVE RETENTION TIME DATA FOR 4, 4 '-DDE
Relative
Retention
Time (RRT)
5.8043
4.6241
4.5415
3.1563
2.4970
2.4683
1.4368
1.2564
0.8925
S33 83=33=3= = 33:
Standard
Deviation
0.1045
0.0578
0.0560
0.0204
0.0102
0.0115
0.0018
0.0008
0.0004
Percent
Relative
Standard
Deviation
1.80
1.25
1.23
0.64
0.41
0.46
0.12
0.07
0.04
3x SD
Window,
sec.
158
109
108
57
36
41
11
6
4
The performance criteria discussed above are interrelated. If one
criterion is outside specification, the expectation is that others may
demonstrate an out-of-control situation as well. In that regard during
Phase II or Phase III it may be discovered that not all criteria are
necessary. Nevertheless, the usefulness of these performance criteria
should be assessed.
The quality assurance plan that is expected to be a part of the
complete protocol is shown in Appendix B.
32
-------
TABLE 25. RELATIVE RETENTION TIME PRECISION DATA
3x SO Window (sec)
Using Given Internal
StandardU)
Compound
Bis(2-chloroethyl) ether
2-Chlorophenol
Nitrobenzene
Isophorone
2, 4-Dimethyl phenol
Hexachlorobutadiene
Acetanilide
3,4-Dichloroaniltne
2,4-Dinitrotoluene
Hexaethyl benzene
Pentachlorophenol
Decaf luorotriphenylphosphine
Anthraquinone
Fluoranthene
4,4'-DDE
1
2.9
5.3
8.6
10.2
1.9
10.1
5.6
4.9
5.2
1.3
1.5
4.8
4.4
5.2
3.9
2 3
3.2 --(b)
6.0
-_
3.3 8.0
10.0 12.6
7.0
—
—
—
5.3
5.8
11.3
—
4.2
5.7
(a)The data given here are limited to that obtained
using internal standards -that gave RRT values of
0.85 to 1.25. If none of the internal standards
give a RRT in that range, the data using the
internal standard that gave an RRT closest to that
range was used. Internal standard No. 1 is that
which gave an RRT closest to 1.0; No. 2 gave an
RRT second closest to 1.0; No. 3 gave an RRT third
closest to 1.0.
(b)— RRT values outside the 0.85 to 1.25 range.
33
-------
TABLE 26. REPEATABILITY OF.DFTPP ION ABUNDANCES, SET 2
(b)
Relative Abundance 'of Given Ion (m/e)
Run No.
1
2
3
4
5
6
7
8(e)
9
10
11
12
13
14
15
16
17
18
19
20
21
22(
-------
TABLE 27. REPEATABILITY OF DFTPP ION ABUNDANCES, SET 3
Run No. Sample^)
1 Calibrated Solution
2 S-0386-UN-1
3 S-0386-UN-2
4 S-0386-UN-3
5 Calibrated Solution
6 Oil blank
7 S-0380-UN-1
8 S-0382-UN-1
9 S-0383-UN-1
10 Calibrated Solution
11 S-0381-UN-1
12 S-0381-UN-2
13 S-0381-UN-3
14 Calibrated Solution
IS Water Blank
16 S-0384-UN-1
17 S-0385-UN-1
18 Calibrated Solution
19 S-0386-SP-1
20 S-0386-SP-2
21 S-0386-SP-3
22 Calibrated Solution
23 S-0381-SP-1
24 S-0381-SP-2
25 S-0391-SP-3
26 Calibrated Solution
Average
Standard Deviation
Percent Relative Std. Dev.
Relative Abundance of Given Ion (m/e)
51
43.9
44.7
43.7
48.3
44.2
NOT
46.2
50.1
45.9
42.3
49.2
50.8
44.7
NOT
NOT
52.0
47.7
NOT
NOT
NOT
54.9
NOT
NOT
NOT
52.5
47.7
3.6
7.6
68 69
2.9 49.7
6.0 67.4
5.4 64.5
5.7 69.8
3.2 51.3
DETECTED
7.2 75.2
6.8 76.1
6.2 73.3
3.1 52.1
7.4 83.8
6.1 75.5
3.6 54.5
DETECTED
DETECTED
8.3 83.9
3.0 53.1
DETECTED
DETECTED
DETECTED
3.5 60.6
DETECTED
DETECTED
DETECTED
3.2 57.7
5.2 66.0
1.8 11.5
35.2 17.4
70
4.1
9.9
8.9
9.9
4.7
10.8
11.2
10.7
4.9
12.9
11.0
5.3
13.5
4.2
4.7
4.3
8.2
3.4
40.6
127
45.9
45.9
45.3
46.2
46.0
49.2
49.2
46.6
42.9
50.7
47.3
43.9
49.2
44.4
49.2
47.5
47.0
2.2
4.7
19;
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
ooo ||
198
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
0
0
199 275
6.6 20.9
6.8 21.3
6.9 21.1
6.3 19.4
6.5 20.4
6.1 23.3
6.7 20.2
6.0 19.7
6.5 20.1
6.0 20.5
6.4 20.1
6.1 20.4
6.2 20.9
6.5 20.7
6.1 19.9
6.4 21.8
6.4 20.7
0.3 0.9
4.3 4.4
(b)
365
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
ooo ||
441
6.3
7.0
6.3
5.3
6.8
5.0
5.6
5.3
6.2
5.8
4.6
6.0
5.7
6.2
5.3
6.3
5.9
0.6
10.8
442
60.9
62.6
64.9
53.3
59.1
55.3
52.5
58.9
63.9
57.3
50.7
61.0
58.9
61.0
52.0
62.7
58.3
4.4
7.5
443
10.3
10.9
11.0
8.7
10.3
9.3
8.5
10.3
11.0
9.8
8.1
10.6
9.8
10.4
8.7
10.6
9.9
0.9
9.4
(*)SP-1, SP-2, and SP-3 designate spike replicates No. 1, 2, and 3, respectively, for the given sample
number. UN-2 and UN-3 designate unsplked replicates No. 2 and 3, respectively, for the given
sample number.
(b)DFTPP (Decafluorotrlphenylphosphlne) has a molecular 1on of 442.
35
-------
TABLE 28. PRECISION OF GC/MS RESPONSE FACTORS OF COMPOUNDS USED AS INTERNAL
STANDARDS AND FOR SPIKING ON AN SE-52 FUSED SILICA CAPILLARY COLUMNU)
to
ot
Compound
Ds-Bronbenzene
Ds-Aniline
Aniline
Dt-fhenol
Bls(Z-chloroethyl Jether
2-Chlorophenol
2.4,6-Trlmethylpyrldlne
1 ,4-Olchl orobenzene
N-N1trosodlpropylMlne
Oj-Nltrobenzene
Nitrobenzene
Isophorone
2.4-OlMthylphenol
Benzole acid
1,2,4-Trlchlorobenzene
Dg-Naphthalene
D4-1.4-D1bnMMbenzene
Hexachlorobutadlene
Acetanlllde
Biphenyl
Dtphenylether
3,4-Oichloroanlline
2,4-Oinitrophenol
4-Nitrophenol
2.4-Oinltrotoluene
4-N1troan1l1ne
0;-l-Nltronaphtha1ene
Hexaethylbenzene
Pentachl orophenol
Hethylparathlon
Quantift-
Retentlon cation
Tine. Ion, Rep. 1
•1n:sec »/e 4/21/81
8:22 82
10:31 98
10:35 93
10:42 99
11:03 93
11:07 128
11:13 121
12:10 146
15:01 70
15:25 82
15:31 123
17:02 82
18:10 122
19:11 122
19:21 180
19:30 136
19:43 240
20:59 225
26:46 93
26:49 154
27:36 170
28:28 161
30:56 184
31:48 139
31:51 165
33:54 138
33:55 134
36:49 231
38:14 266
41:48 125
Oecafluorotriphenylphosphlne 42:03 198
Anthraqulnone 43:47 208
Fluoranthene 45:59 202
4.4'-DDE
Oi2-Chrysene
{•'Response factor relat
("'Standard Deviation.
RF - 5i x Eli ^ere A =•
48:44 246
54:37 240
0.315
0.774
0.794
0.629
0.477
0.339
0.670
0.566
0.138
0.406
0.363
0.952
0.596
0.122
0.421
1.286
0.241
0.158
0.787
1.181
0.647
0.421
0.037
0.086
0.238
0.087
0.231
0.621
0.084
0.107
0.260
0.262
0.984
0.175
0.467
Rep. 2 Rep. 3 Rep. 4
4/22/81 4/23/81 4/23/81
0.331
0.808
0.832
0.664
0.479
0.349
0.679
0.566
0.146
0.423
0.368
0.988
0.604
0.068
0.430
1.301
0.253
0.161
0.782
1.187
0.659
0.418
0.034
0.037
0.234
0.110
0.245
0.632
0.085
0.100
0.247
0.251
0.953
0.173
0.473
jve to DiQ-Phenanthrene. (°)tach compound
(e'Percent relative standard deviation.
Area counts and C - Concentration.
0.342
0.798
0.825
0.644
0.478
0.340
0.685
0.559
0.137
0.413
0.362
0.965
0.589
0.057
0.417
1.220
0.246
0.153
0.773
1.112
0.632
0.413
0.029
0.081
0.232
0.092
0.243
0.609
0.085
0.111
0.242
0.274
1.025
0.182
0.529
present at
0.343
0.807
0.835
0.642
0.475
0.461
0.780
0.555
0.127
0.416
0.354
0.965
0.583
0.056
0.411
1.272
0.243
0.147
0.745
1.116
0.605
0.394
0.029
0.033
0.217
0.116
0.238
0.575
0.078
0.104
0.243
0.271
0.996
0.179
0.515
50 na/wl.
Rep. 5
4/24/81
0.332
0.796
0.802
0.637
0.465
0.468
0.727
0.553
0.124
0.405
0.356
0.951
0.569
0.068
0.415
1.221
0.232
0.153
0.667
1.133
0.629
0.357
0.015
0.061
0.222
0.078
0.228
0.592
0.083
0.105
0.349
0.274
1.027
0.183
0.513
Two pi was
Rep. 6
4/27/81
0.339
0.810
0.842
0.625
0.506
0.488
0.546
0.600
0.119
0.461
0.371
1.025
0.509
0.063
0.439
1.427
0.271
0.170
0.570
1.161
0.637
0.264
0.024
0.056
0.138
0.048
0.207
0.654
0.069
0.043
0.264
0.243
0.899
0.189
0.280
Injected
Rep. 7
4/27/81
0.369
0.810
0.842
0.635
0.489
0.475
0.619
0.578
0.131
0.447
0.360
1.025
0.568
0.099
0.425
1.242
0.253
0.167
0.653
1.111
0.614
0.341
0.019
0.044
0.191
0.100
0.236
0.643
0.100
0.105
0.259
0.257
0.990
0.190
0.504
on column
Rep. 8 ., ,
4/28/81 ilc)
0.357 0.341
0.824 0.790
0.832 0.813
0.637 0.639
0.493 0.484
0.369 0.421
0.653 0.668
0.564 0.568
0.128 0.131
0.443 0.430
0.364 0.362
1.017 0.991
0.577 0.571
0.108 0.074
0.420 0.422
1.207 1.270
0.256 0.251
0.165 0.159
0.692 0.697
1.079 1.128
0.615 0.627
0.329 0.359
0.022 0.026
0.089 0.057
0.210 0.205
0.096 0.091
0.250 0.235
0.648 0.622
0.108 0.087
0.116 0.098
0.255 0.266
0.267 0.262
1.050 0.991
0.199 0.185
0.514 0.475
for each analysts.
SD(d)
0.016
0.039
0.034
0.012
0.013
0.065
0.082
0.016
0.008
0.021
0.006
0.031
0.030
0.020
0.010
0.080
0.010
0.009
0.076
0.036
0.018
0.055
0.007
0.021
0.036
0.021
0.014
0.030
0.013
0.025
0.038
0.012
0.051
0.008
0.088
(c)Hean.
RSO(e)
4.8
4.9
4.2
1.9
2.8
15.0
12.0
2.9
6.6
4.8
1.7
3.2
5.3
28.0
2.3
6.0
4.8
5.4
11.0
3.2
2.9
15.0
29.0
38.0
18.0
23.0
6.1
4.9
15.0
25.0
14.0
4.7
5.2
4.6
19.0
-------
TABLE 29. REPEATABILITY OF GC/MS RESPONSE FACTORS FOR
DEUTERATED INTERNAL STANDARDS
—==========================================================================
RF(a) USED FOR GIVEN DEUTERATED COMPOUND(b)(m/e)(c)
NB
82
0.41
0.45
0.41
0.46
0.42
0.39
0.40
0.43
0.43
0.41
0.39
0.41
0.41
0.41
0.41
0.22
0.42
0.41
0.46
0.45
0.43
0.45
0.49
0.45
0.49
0.44
Mean 0.42
SD 0.049
% RSD 12
ANIL
98
0.77
0.83
0.74
0.83
0.81
0.76
0.73
0.81
0.80
0.80
0.67
0.66
0.76
0.81
0.80
ND
0.26
0.80
0.42
0.56
0.79
0.81
0.80
0.92
0.88
0.82
0.75
0.14
19
PHOL
99
0.63
0.71
0.65
0.72
0.66
0.60
0.64
0.67
0.68
0.64
0.67
0.68
0.68
0.64
0.64
0.61
0.64
0.64
0.68
0.64
0.67
0.64
0.70
0.77
0.71
0.64
0.66
0.037
5.6
NAPH
136
1.29
1.52
1.36
1.51
1.30
1.30
1.26
1.41
1.38
1.22
1.23
1.02
1.30
1.27
1.34
0.69
1.30
1.22
1.48
1.39
1.37
1.24
1.45
1.24
1.43
1.21
1.30
0.17
13
BB
82
0.32
0.36
0.33
0.36
0.33
ND
0.32
0.35
0.34
0.34
0.33
0.35
0.34
0.34
ND
0.24
0.33
0.33
0.37
0.35
0.35
0.37
0.39
0.43
0.41
0.36
0.35
0.035
10
NN
134
0.23
0.22
0.21
0.22
0.25
0.19
0.21
0.20
0.20
0.24
0.20
0.20
0.19
0.24
0.19
0.25
0.24
0.23
0.26
0.26
0.26
0.24
0.24
0.24
0.28
0.26
0.23
0.023
10
DBB
240
0.24
0.28
0.25
0.28
0.25
0.24
0.25
0.26
0.25
0.25
0.23
0.25
0.23
0.24
0.24
0.19
0.25
0.23
0.28
0.26
0.25
0.25
0.27
0.25
0.28
0.26
0.25
0.019
7.8
CHRY
240
0.47
0.48
0.44
0.49
0.47
0.49
0.50
0.53
0.53
0.53
0.50
0.50
0.50
0.52
0.51
0.50
0.46
0.51
0.47
0.50
0.49
0.50
0.56
0.50
0.56
0.51
0.50
0.028
5.5
====== ============3==== ============================= 3= =3= ====== ======= ====3
(^Response relative to that of m/e of 188 for Din-Phenanthrene.
(b)NB-D5-Nitrobenzene' Anil=D5-Aniline; Phol=Dc-Pnenol ; Naph=DQ-
Naphthalene; BB=D5-Bromobenzene; NN=D7-l-N1tronaphtahlaene;
DBB=d4-l,4-Dibromobenzene; Chry=D12-Chrysene.
vc)The m/e of the base peaks used for determining the response 1s given
below the compound abbreviation.
As 'Cis
RF = --- x — where A = Area counts and C = Concentration.
37
-------
TABLE 30. REPEATABILITY OF AREA COUNT OF AN
INTERNAL STANDARD (DiQ-PHENANTHRENE) FOR
25 CONSECUTIVE GC/MS ANALYSES^)
Run Area Count
1 477865
2 496069
3 564487
4 507940
5 409343
6 607792
7 451664
8 534921
9 514559
10 524567
11 491671
12 415231
13 453006
14 414140
15 516642
16 . 474358
17 449165
18 466601
19 456778
20 453641
21 467746
22 476233
23 437642
24 462681
25 468228
Mean 479800
Standard Deviation 46600
Relative Standard Deviation 9.7%
==================================================
data were acquired using a GC/MS system
fitted with an automatic sample injector.
38
-------
Evaluation and Modification
Prescreening
In the GC/MS analysis of either purgeable or semivolatile components it
is essential that the amount of material introduced into the GC/MS system does
not overload the system so that the desired determinations cannot be made. In
order to avoid such overloading and at the same time have methods that accommo-
date waste samples that vary widely in their purgeable or semivolatile content,
the amount of sample used must be variable. The variation can occur in either
the weight of the original sample taken or in the size of a subsequent aliquot.
In order to minimize problems associated with sample inhomogeneity or loss of
volatile components caused by weighing small samples, the final procedures
selected specify a single sample and a variable analysis aliquot.
The size of the aliquot used is based upon the result of a prescreening
procedure. For the determination of purgeable components, the prescreening
procedure involves extraction with n-hexadecane and GC analysis to determine
the total volatile content. For the determination of semivolatile components,
the prescreening procedure involves weighing an extract aliquot to determine
the total solvent extractable content and also a GC analysis of the extract to
determine the total semivolatile content.
The details of these prescreening procedures are incorporated into the
analytical protocols. Both GC analyses are performed using a 30 m x 0.25 mm
DB-5 fused silica capillary column. This column is the same one that is recom-
mended for the GC/MS determination of semivolatile components; thus the GC
results will be directly applicable to estimating the optimum concentration to
use for the GC/MS analysis. The DB-5 column was selected for screening volatile
compounds because its high temperature limit permits most high-boiling components
to be purged from the column at the end of each run. Over a dozen samples of
standards dissolved in hexadecane and hexadecane extracts of wastes were run
on an SE-52 glass capillary column without significant column deterioration to
demonstrate that a si 11 cone-phase capillary column is suitable for this type
of a screening analysis.
GPC Cleanup
Gel permeation chromatography serves very well as a cleanup method to
separate long-chain molecules from smaller condensed molecules. In a method
that was developed by Battelle several years ago for determining organic
priority pollutants in municipal sludge, GPC was found to be very useful for
removing the triglycerides, fatty acids, and long-chain molecules that
comprised most of the extractable material. However, most industrial solid
wastes studied, unlike municipal sludge, do not contain large amounts of such
long chain molecules. Satisfactory GC/MS data can be obtained by analyzing
crude extracts with no cleanup. The major advantage of using GPC cleanup is
that less high-boiling or nonvolatile high-molecular-weight material may
accumulate in the GC injector liner or in the injector end of the GC column.
However, numerous sets of 25 to 50 industrial waste extracts have been
39
-------
analyzed without GPC cleanup and no analytical difficulties have been
encountered.
In an effort to assess the possible value of GPC cleanup for obtaining
cleaner gas chromatograms with better-resolved peaks, three solid waste
extracts that had particularly large amounts of unresolved components were
cleaned up by GPC and reanalyzed by GC analysis. The before and after gas
chromatograms are shown in Figures 1 to 3. In each case GPC cleanup had no
effect on the quality of the gas chromatogram.
Undoubtedly there will be specific instances in which GPC cleanup would
be beneficial. However, on the basis of our observations to date, we do not
believe the additional effort required for GPC cleanup of every sample is
warranted. A GPC cleanup step has been included in the analysis procedure for
use only with extracts that contain high levels of solvent extractable
material that do not elute from the GC column.
Bonification Studies
Four waste samples which are considered difficult to disperse, namely
coal tar, ethanes spent FeCl2 catalyst, latex paint, and organic still
bottoms, were homogenized in a mixture of methylene chloride and water using
three different techniques. The three techniques Involved the use of: (1) a
Brinkmann Polytron Model PT-20ST, (2) a Heat Systems Model W-375 Sonicator
with a 0.5-Inch diameter probe-type horn, and (3) a Heat Systems Model W-375
Sonicator with a 3-inch diameter water-cooled cup horn. In each case 2-g of
sample was mixed with 50-mL of methylene chloride and 50-mL of 10% NaCl. The
effectiveness of the dispersion was assessed by observation of the efficiency
of the mixing procedure and the rapidity of phase separation that occurred
thereafter. The use of the cup horn was of particular interest because it
could provide a means of homogenizing abrasive samples without damaging the
homogenlzatlon apparatus. The cup horn was not very effective in comparison
with the use of the sonic probe or Polytron. The use of a thin-walled Teflon
container in the cup horn may be more effective than the tall-form beaker
used, but the technique does not appear to be very promising for general use.
The Brinkmann Polytron gave excellent homogenization in all cases and was
recommended for these solid waste extractions. In most cases, the waste
samples will not contain large amounts of abrasive materials that would cause
rapid deterioration of the probe generator. If large numbers of sandy soil,
sandy sediment or other abrasive samples are extracted, a shorter homogeniza-
tion time may be necessary or other methods may need to be considered.
Although the cup horn did not work very well for preparing oil-in-water
dispersions, it may be very effective for promoting the extraction of organic
components into methylene chloride or acetonitrile in the absence of added
water.
Evaluation of Fused Silica Capillary Column (FSCC)
Battelle has made extensive use of glass capillary columns for gas
chromatography for the past five years. We set up a special laboratory for
40
-------
F , I. LP*Q3S
. B
E-O'SC Er«-.sr
In»t:1S
10.0
80
-IB
Tray • 37
10 52.18 5/05/1921
Pol » P»g« 1
Before GPC Cleanup
I ' ' ^^ l^ ' ^ ' ^T ^T 1 ] T T , , i T—I-T -T J I—I .—T-J-T—I I I |
20.0 30 0 40.0 50.0 60.0 70.0 80.0 90 0 10C.0
t Smx: 100.OC3
is™-. -1 cc;
403 n«
85* my
MINUTES
S-Q18Q WTER GPC
In»t'18 Ch-0 Tr«y • IB
After GPC Cleanup
23:40:56 S/CS/1931
Pol * Pagt 1
10.0 20.0 30 0 40.0 50.0 60.0 70.0 80.0 90 0 100.0
M«x= 58 00a m» MINUTES * *"*( 1t» CCS3
Mir,- -2.807 «. MINUTES I Sr~: -1C23
Figure 1. Gas chromatogram of extract from poly still bottoms
before and after GPC cleanup.
41
-------
S-QC1Q "isne- **-
In«t: 18 D- 0 Trmy • 35
ia.3S S3 5/05/1991
Pol + Pa.g« 1
Before GPC Cleanup
' ' I ' ' ' ' I ' > ' ' I ' ' I ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' I I I ' ' ' 1 i ' > • ' I
ie.0 so.a 30.0 40.B 50.a 60.0 70.a so 0 90.a 100.0
X Snx: 10(3 BOB
X S™n -1 Q33
23 341 nv
-2 6*4 m.
MINUTES
F . I * LPidSR
0
o=c
Ch 0 Tr«y
17:13:30 5/05/1981
Pol +
After GPC Cleanup
10.0 20. a 30.0 43 a 50. a sa a 73. a aa a 9o.a 103. a
M«x; 1?.184 m» J Smx: 100 O30
, 5^ ; -1.030
1?.184 m»
-2473-..
MINUTES
Figure 2. Gas chromatogram of extract from Ethanes I spent
catalyst before and after GPC cleanup.
42
-------
F.I..LP4B2R
FF • 0
14 30 36 5/05/198'
Fo 1
' ' 'IB
i '
ll .1
32
1,
1 '
.0
Jit
30
1
.0
40
Before GPC Cleanup
JjLU ll.l_ J«« -" *•*"--
0 50. 0 60.0 700 800 90.0100.0
Max• 117.420 mv
Mtn: -3.469 m.
* Smx: 108.030
I &*n. -1.D3S
FE • 0
ll
CCW. tfTEK GPC
In«x:18 Ch:0 Tr«y «
)B-.Z7;S3 5/02/1981
Pol + P»gt 1
After GPC Cleanup
Jll
r ,,,,-,,-,, -r , -p
10.0 33.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0
Ctax: 111.470 mv MTMiiTrs *9nw: 100 B00
Mir,: -3.46E m. MINUTES f Snw : . -1 .0CE
Figure 3. Gas chromatogram of extract from Coal Gasification Tar
before and after GPC cleanup.
43
-------
preparing our own glass capillary columns three years ago and consistently
prepared SE-52 WCOT columns that were superior to those purchsed from J W,
Supelco, SGE, and Quadrex. The Battene-made SE-52 columns gave good peak
characteristics for both benzidine and nitrophenols at the 20-ng level. Pre-
viously Battelle used the columns for the GC/MS analysis of nearly a hundred
combined acid and neutral fractions from industrial wastewaters for an EPA
study in 1978 and for the GC/MS analysis of several hundred combined fractions
from solid wastes and solid waste leachates for an EPA study in May and June,
1980. The columns withstood heating to 300°C and injection of over a hundred
very dirty samples.
The biggest problem in the use of the capillary columns for GC/MS analysis
was a suitable transfer line. This transfer line provides a link between the
GC column end and the MS detector. A specially-prepared glass capillary transfer
line was used and connected with a low-dead-volume union. The transfer line
was very fragile and difficult to install. Thus, when fused silica capillary
columns were introduced by Hewlett-Packard we hoped they would be the ideal
solution to the transfer line problem. Unfortunately, when we checked the H-P
columns for performance characteristics we found the columns to be very inferior
to the Battelle-made glass capillary columns in terms of efficiency, polarity,
and stability. However, fused silica capillary columns prepared by J&W
Scientific this year, performed as well as Battelle-made glass capillary
columns. Thus, we strongly endorse the use of fused silica capillary columns
for which satisfactory performance can be demonstrated. We recognize their
superiority for GC/MS work since they can serve as the transfer line as well
as the column and eliminate the need for a union. They are easy to install,
are commercially available, and should lead to very significant improvements
in inter!aboratory reproducibility. The Phase III inter!aboratory study should
not be attempted without the use of fused silica columns. It is very likely
that the DB-5 bonded-phase columns produced by J&W, which are recommended for
the interlaboratory study, are superior to SE-52 or SE-54 fused silica columns
in terms of stability.
A further important advantage of FSCCs over glass capillary columns may
be the precision of relative retention times. Although it is not possible to
determine whether the effect is due to the elimination of the union or to the
nature of the fused silica columns, a significant improvement has been observed
in the relative retention times obtained in GC/MS analyses when an FSCC was
used.
GC/MS analyses using a fused silica capillary column (30 m x 0.25 mm DB-5
purchased from J&W) were performed on three solid waste extracts that had
previously been analyzed using a glass capillary column (30 m x 0.25 mm SE-52,
made by Battelle). The three extracts, which were from Set 2 of the Residual
Waste Analysis project, were selected as the extracts most likely to cause
column overloading problems. Comparisons of the quantities of the major com-
pounds found in each sample using the two different columns are given in Tables
31 to 33. Although there are significant differences between the two sets of
data that may be caused by changes in the extracts during storage, differences
in column adsorption effects, or differences in the mass spectrometer tuning,
there were no consistent differences that would indicate that the FSCC performs
better than the glass column.
44
-------
TABLE 31. COMPARISON OF GLASS AND FUSED SILICA CAPILLARY COLUMNS
FOR GC/MS ANALYSIS OF SAMPLE S-0190
===========================================================================
Compound
Amount-Found Using Given Column, ug/g
GlassFused Silica
Naphthalene
Hexachl orobutadlene
Trichlorobenzene
Blphenyl
Diphenyl ether
Cis-Alkane
Hexachl orobenzene
Cn-Alkane
==================================:
11
88
10
13
35
28
18
8
48
210
3
26
90
7
65
7
TABLE 32. COMPARISON OF GLASS AND FUSED SILICA CAPILLARY COLUMNS
FOR GC/MS ANALYSIS OF SAMPLE S-0180
===========================================================================
Compound
Amount-Found Using Given Column, ug/g
GlassFused Silica
Chlorinated unknown-87, 123
D1 ethyl octatrl ene
Unknown-172, 79
Dimethyl octatrl ene
Unknown-ISO
Unknown-150 (CioHi/iO)
Unknown-164, 135
Unknown-178, 149
Cs-Alkyl phenol -206, 177
580
82
180
78
74
68
100
44
23
180
18
60
24
45
64
60
30
28
=====================================================================3=====
45
-------
TABLE 33. COMPARISON OF GLASS AND FUSED SILICA CAPILLARY COLUMNS
FOR GC/MS ANALYSIS OF SAMPLE S-0186
Compound
1,4-01 chl orobenzene
Hexachl oroethane
1,2,4-Trichlorobenzene
Hexachl orocycl opentadi ene
Tetrachl orobenzene
Tetrachl orobenzene
Chlorinated unknown-179
Pentachl orobenzene
Chlorinated unknown
Hexachl orobenzene
Hexaethyl benzene
Hexachl orobutadi ene Isomer
Hexachl orobutadi ene
Amount Foun<
Glass
700
29,000
320
29,000
1,100
1,300
3,300
2,800
1,900
13,000
230
9,400
81,000
d Using Given Column, pg/g
Fused Silica
280
8,600(a)
3,200
23,000
l.lOo(a)
l,10o(a)
l,700(a)
l,100(a)
l,300(a)
4,400
290
2,900
43,000
(a)saturated mass spectra. Several of the chlorinated unknowns
appeared different in this analysis (higher molecular weight).
The greatest amount of overloading occurred when the extract from Sample
S-1086 was analyzed. The total ion chromatograms obtained from the GC/MS
analysis of this sample using fused silica and glass capillary columns are
shown in Figure 4 and 5, respectively. No differences in the degree of
overloading were found.
A narrow-bore (0.25 mm I.D.) and a wide-bore (0.32 mm I.D.) DB-5 FSCC
obtained from J&W Scientific were evaluated to compare the sample loading
capacities of the two columns. The evaluation was performed by analyzing
various concentrations of the Grob2 column performance test mixture. The
mixture contains the following compounds:
46
-------
SMI
TIME
Figure 4. Total ion chromatogram obtained from sample S0186
using a DB-5 Fused Silica Capillary Column.
-------
lOO.O-i
CO
RIC
3IAS
2506
41:40
SCftl
TIME
Figure 5. Total ion chromatogram obtained from sample S0186
using an SE-52 Glass Capillary Column.
-------
n-decane 2,3-butanediol
n-undecane 2,6-dimethylphenol
methyl decanoate 2,6-dimethylaniline
methyl undecanoate dicyclohexylamine
methyl dodecanoate 2-ethylhexanoic add
1-octanol
Separation numbers were calculated from the retention time and peak width at
half-height obtained for the fatty add methyl esters at different Injected
amounts. The separation number provides a numerical basis for assessing the
column capacity. As the separation number decreases the efficiency or resolv-
ing ability of the column also decreases. The results obtained (see Table 34)
indicate that there was little difference in the capacities of the two columns
evaluated. The separation numbers for both columns decrease as amounts above
40-ng are injected; however acceptable performance (SN = 30) was achieved at
levels up to 200 ng. Since the capacities of the two columns are similar, the
narrow bore column has been selected for Phase II and Phase III studies on the
basis of its somewhat better performance at low concentrations, its greater
availability, and a lower pressure drop at the detector end.
Method Selection
Requirements of Methods
Phase I involved the selection of two methods for the analysis of solid
wastes, one method for the determination of purgeable organic compounds and
one method for the determination of semi volatile compounds. The requirements
of the methods are as follows:
1. Applicable to as broad a range of matrices as possible with a minimum
of waste-specific modifications.
2. Applicable to as broad a range of potentially hazardous organic
compounds as possible.
3. Capable of detecting and quantifying compounds at concentrations down
to at least 1 pg/g of solid waste. It 1s recognized that the detec-
tion limits will be proportionately higher (poorer) for waste samples
that contain more than 1000 pg/g of total purgeable compounds or total
solvent extractable compounds. The acceptability of these detection
limits will depend on the Intended application of the analytical data
generated from these methods.
4. Known detection limits.
5. As simple as possible.
6. Acceptable repeatability and reproducibility; both within (intra-
laboratory) and between (interlaboratory) precision.
49
-------
TABLE 34. PERFORMANCE OF COMMERCIAL FUSED SILICA CAPILLARY COLUMNS
Amount
Injected,
ng/compound
Separation Number(a)(b) Using Given column
Narrow Bore (0.25 mm I.D.)Wide Bore (0.32 mm I.D.)
Methyl Decanoate and Methyl Undecanoate
40
100
200
400
800
40
100
200
400
800
48
38
33
25
19
44
38
31
25
19
Methyl Undecanoate and Methyl Dodecanoate
45
36
ND(C)
24
18
38
36
30
24
18
U)Separation Number =
RT(b)-RT(a)
- 1 where RT = retention time
Wl/2(a)+Wi/2(b)
and Wi/2 = width at half height for compounds (a) and (b) which are two
compounds in a homologous series which differ from each other by one-
CH2- group.
(b)GC Conditions:
Carrier gas - Helium at 28 cm/sec at ambient
Sample injection - 1 yL, splitless
Column temperature - 40°C to 150°C at 0.8°C per minute
(c)Not determined; temperature program was terminated prematurely.
Sample Preparation Procedures for Determining Semi volatile Compounds
The two solid waste analysis methods considered for the determination of
semivolatile organic compounds are referred to as the NEIC method (National
Enforcement Investigation Center) and a BMS method Battelle-Midwest-
Southern).
The NEIC method involves the addition of a one-gram sample to 1000 ml of
water followed by successive extractions under basic and acidic conditions
with 15% methylene chloride in hexane. The effect of using a more polar
solvent, e.g. 100% methylene chloride, to obtain better recovery of the more
polar compounds such as nitrophenols, and the effect of using an acetonitrile,
50
-------
methanol, or isopropyl alcohol extract of the waste to achieve better extrac-
tion efficiency and fewer emulsification problems were studied. The BMS method
involves homogenization and successive extractions of the wet sample with
methylene chloride under basic and acidic conditions. The effects of decreasing
the sample/methylene chloride ratio, using different solvents, and adding 10%
aqueous sodium chloride to minimize emulsion formation and the effect of adding
neutralizing agents and drying agents with little or no water to eliminate
emulsion formation was also studied.
Three separate procedures for sample preparation, designated A, B, and C,
were written and evaluated in terms of the parameters discussed above. Proce-
dures A and B are based on the BMS techniques and Procedure C is based on the
NEIC techniques. Each of these sample preparation procedures is described in
Appendix D.
Procedure A involves the addition of water and sequential extraction with
an organic solvent under acidic and basic conditions. Procedure B involves
the addition of neutral buffer salts and a drying agent and the simultaneous
extraction of acids, bases, and neutrals. Procedure C involves extraction
with a water-miscible organic solvent followed by addition of the extract to
water and sequential extraction with a water-immiscible organic solvent.
We investigated the effects -of using different solvents when the proce-
dures were applied to the extraction of coal gasification tar, ethanes I spent
FeC12 catalyst, drying bed solids, and Cincinnati dewatered sludge. The results
obtained indicated the following:
1. Diethyl ether works much better than methylene chloride in Procedure
A since in all cases the ether layer can be readily separated by
centrifugation. A stable emulsion is formed when methylene chloride
is used for the extraction of either the drying bed solids or the
sludge.
2. Methanol or acetone works better than acetonitrile in Procedure C
since the extract can be more readily concentrated.
3. Acetone extracts more color and possibly more resinous nonvolatile
material, in some cases, than methanol in Procedure C.
4. Diethyl ether does not work as well as methylene chloride in
Procedure C in that the extract contains much more water.
5. The determination of residue weight is not adequate for determining
the optimum concentration of material for the GC/MS analysis. A GC
screening, estimation of total semivolatile content (TSVC), using
FSCC was therefore, added.
6. The TSVC in conjunction with the residue weight data can probably be
used to determine whether GPC cleanup can lower the detection limit.
51
-------
Procedure B is favored because of its relative simplicity. The procedure
as originally described in which no water was added, was revised to avoid the
homogenization of dry materials which caused extensive damage to the
homogenizer.
In the revised procedure the homogenization was performed in the presence
of a minimum amount of water and the incorporation of a drying agent was
achieved separately during stirring with a spatula. A separate neutralization
step was included to determine the amount of acid or base required to achieve
pH 7. The amount of sample used for the estimation of total solvent extract-
able material was increased and the extract obtained was adequate for all
subsequent steps in the procedure.
One of our major concerns with Procedure B was to determine if satis-
factory recoveries for both acidic and basic compounds can be achieved in a
single step. Spiking experiments were therefore performed using a standard
solution that contained a wide range of compounds including alkylphenols,
nitrophenols, anilines, quinoline, 2,4-D, and 2,4,5-T. Quantitative recoveries
were obtained for all compounds except for the carboxylic acids, 2,4-D and
2,4,5-T. The recovery of the carboxylic acids was about 20 percent. The
procedure is, therefore, quite promising for the extraction of all of the
neutral, acidic, and basic compounds that are normally expected to be deter-
mined by GC/MS analysis without derivatization.
The changes mentioned above in the estimation of total solvent extract-
bles were also incorporated into Procedure A. Another revision in Procedure A
was made to provide for the ether layer separating out by standing and to
specify the use of a refrigerated explosion-proof centrifuge if centrifuga-
tion was required to facilitate phase separation. It may be advisable to
provide for the use of methylene chloride or ether as two separate options,
with ether to be selected as the solvent only if methylene chloride is first
found to be unsatisfactory for the particular waste being studied.
Procedure C doesn't provide any advantage over Procedure A and suffers
from the disadvantage of interference by the water-miscible solvent during the
concentration step. As stated above, because of the relative simplicity of
the method, Procedure B was favored for the inter!aboratory study. It is
recognized, however, that in some cases, e.g. when aliphatic amines are
present along with reactive components such as aldehydes, esters, haloethers,
and other reactive halogen compounds, there would be a distinct advantage to
keeping the basic fraction separate from the neutral and acidic fraction as is
possible by using Procedure A with a slight variation. It should be noted,
however, that procedures commonly used for wastewater analysis, e.g. EPA
Method 625, do not provide for keeping basic compounds separate from reactive
neutral compounds.
Procedure for Determination of Volatile Organic Compounds
The following six procedures were considered for the GC/MS determination
of volatile compounds:
1. Dispersion in water followed by a purge-trap-desorb procedure.
52
-------
2. Solution or extraction into methanol followed by direct injection.
3. Solution or extraction into methanol followed by addition of an
aliquot of the extract to water and a purge-trap-desorb procedure.
4. Solution or extraction into n-hexadecane followed by direct injection.
5. Sonification of the sample in a septum-sealed VOA vial at an elevated
temperature and direct injection of a headspace sample.
6. Solution or extraction into polytethylene glycol), m.w. 400, followed
by addition of an aliquot of the extract to water and a purge-trap-
desorb procedure.
Procedure 1 was shown to work well for municipal sludges which disperse
well in water and which contain relatively low levels of volatiles. It would
not be suitable for a sample such as coal tar which would not disperse in water.
It would not be suitable for a sample that contains more than a few ppm of
each volatile component since the small sample size that would be required to
avoid overloading the GC/MS system could not be accurately weighed and trans-
ferred without major losses of components by volatilization.
Procedure 2 avoids the problem of using a small sample since a larger
sample, e.g. one gram, can be taken and appropriate dilutions made. The proce-
dure can work well for many sampl'es containing volatile components at levels
greater than about 50 pg/g. It is not suitable for low levels. It is not
very suitable for samples that contain high levels of extractable semivolatile
components that remain on the GC column and elute over a period of several
hours. Also, if the solvent is vented to protect the mass spectrometer, the
lower-boiling volatiles are lost.
Procedure 3 works well for determining volatile compounds at levels down
to about 10 ug/g. However, since 10 uL of methanol is about the maximum amount
that the purge-trap-desorb system can tolerate, the procedure is not suitable
for levels of 1 to 10 ug/g. Also, methanol is frequently not a very good
dispersing medium.
Procedure 4 is subject to the same concentration limits and semi volatile
problems as procedure 2; however, none of the lower boiling volatiles are
lost. The solvent is vented after all volatiles of interest are eluted. This
procedure requires the use of a si 11 cone column, preferably a capillary column,
rather than the packed column used with the Method 624 purge-trap-desorb proce-
dure. The same extract that is used for estimating the total volatiles content
by GC analysis can be used for the GC/MS analysis.
Procedure 5, like a purge-trap-desorb procedure, is very sensitive and
avoids problems caused by high-boiling components. However, since it relies on
an equilibrium between the liquid and gas phase, recovery is highly matrix-
dependent especially if the samples contain high levels of oils and tars.
53
-------
Procedure 6 is similar to Procedure 3; however, since a nonvolatile
solvent is used a larger amount of extract can be added to the water and a
detection limit of 1 ug/g can be readily achieved if necessary. Also,
poly(ethylene glycol), m.w. 400, seems to be a better dispersing medium than
methanol.
Procedure 6 was applied to coal tar and several still bottoms and was
found to work better than other procedures used with these samples. The proce-
dure appears to be more universally applicable than any other procedure tested.
This procedure was selected as the basis for the method for determination of
purgeable compounds.
Recommended Methods
The final methods descriptions are given in the Appendices of the Phase
III report. The method for determination of semivolatile compounds uses
procedure B. The methods are written in a routine applications format. These
descriptions were modified for the inter!aboratory study protocol to reflect
the demands of the study. For instance the method will include necessary
elements of the quality assurance protocol.
REFERENCES
1. ASTM E691, Part 41, 1980.
2. Grob, K. Jr., Grob, G., and K. Grob, J. Chromatogr., 156, 1-20 (1978).
54
-------
APPENDIX A
ANALYTICAL RESULTS FROM TARGET WASTES
55
-------
APPENDIX A
COMPOSITION OF WASTES
Results of the analysis of wastes selected for the interlaboratory
comparison study are given in Tables A-l through A-10.
TABLE A-l. RESULTS OF GC/MS ANALYSIS
OF OLENTANGY RIVER SEDIMENT
Amount,
Tentative Identification ug/g
Acenaphthylene 0.1
Acenaphthene 0.1
Fluorene 0.1
Dibenzothiophene 0.2
Phenanthrene 1.8
Fluoranthene 2.2
Pyrene 1.7
Chrysene 1.0
Benzo(b/k)fluoranthene 1.0
Benzo(a)pyrene 0.7
Perylene 0.4
Bis(2-ethylhexyl) phthalate 1.5
56
-------
TABLE A-2. RESULTS OF GC/MS ANALYSIS
OF MACHINE OPERATING WASTE
==================================================
Amount,
Tentative Identification yg/g
n-Undecane 600
n-Dodecane 3,600
n-Tridecane 6,200
n-Tetradecane 6,000
n-Pentadecane 1,400
n-Hexadecane 4,900
n-Heptadecane 3,800
Pristane 1,600
n-Hexadecane 2,400
Phytane 1,100
n-Nonadecane 1,400
==================================================
TABLE A-3. RESULTS OF GC/MS ANALYSIS
OF HERBICIDE ACETONE-WATER
Amount,
Tentative Identification yg/g
Alkylamine (C8H19N) 75
Alky!amine 9
Alkyl ketone (MW = 128) 8
Unknown 20
Unknown 30
Unknown 9
==================================================
57
-------
TABLE A-4. RESULTS OF GC/MS ANALYSIS
OF COAL GASIFICATION TAR
==================================================
Amount,
Tentative Identification yg/g
Phenol 110
2,4-Dimethylphenol 9
Naphthalene 110
Benzothiophene 13
Isoquinoline 7
2-Methyl naphthalene 60
1-Methyl naphthalene 35
Biphenyl 29
Cg-Alkylnaphthalene 11
C2-Alkylnaphthalene 10
Acenaphthylene 160
Dibenzofuran 40
MW 166 PAH 60
Fluorene 25
Methyldibenzofuran 12
DibenzotMophene 13
Phenanthrene • 90
Anthracene . 60
Carbazole 20
Methylphenanthrene 7
Methylphenanthrene 11
Cyclopentaphenanthrene 20
Phenylnaphthalene 6
Fluoranthene 46
MW 202 PAH 14
Pyrene 190
Methylpyrene 11
Benzo(a)anthracene 30
Chrysene 140
Benzo(b)fluoranthene 110
Benzo(a)pyrene 135
Benzo(g,h,i)perylene 120
D1benzo(a,h)anthracene 50
Indeno(l,2,3-cd)pyrene 93
==================================================
58
-------
TABLE A-5. RESULTS OF GC/MS ANALYSIS
OF EDC SPENT CAUSTIC
==================================================
Amount,
Tentative Identification ug/g
Methyl chloride 2
Ethyl chloride 17
2-Propanol 120
1,1-Dichloroethylene 12
Brotnochlororaethane 100
1,2-Dichloroethylene 1
Diethyl ether 3
1,2-Dichloroethane 4,500
Trichloroethylene 1
Unknown 6
1,2-Dichloroethane 28
Unknown 3
Unknown 7
Unknown 8
1,1,2-Trichloroethane 16
Tetrachloroethylene 2
Dichlorobutene 1
Tetrachlorobutane • 2
Chlorinated unknown 1
B1s(2-chloroethyl) ether 3
Chlorinated unknown 10
Hexachloroethane 1
Chloroacetic acid 9
Chlorinated unknown 22
Chlorinated unknown 2
Hexachlorobutadiene 2
59
-------
TABLE A-6. RESULTS OF GC/MS ANALYSIS
OF OXYCHLORINATION SPENT CATALYST
Tentative Identification
Amount,
pg/g
Ethyl benzene
63 Alkylbenzene
Propylbenzene
03 Alkyl benzene
Trimethyl benzene
1,4-Dichlorobenzene
03 Alkylbenzene
Cn Alkane
04 Alkylbenzene
Naphthalene
Ci2 Alkane
Hexachlorobutadi ene
Dichloromethylphenol
Chlorinated unknown
4-Chloro-3-methylphenol
Methyl naphthalene
Methyl naphthalene
C; Alkylbenzene
2,4,6-Tri chlorophenol
C.Q Alkyl benzene
Biphenyl
Ci4 Alkane
Diphenylether
Cy Alkyl benzene
C; Alkylbenzene
Cy Alkylbenzene
Tetraethyl benzene
GS Alkyl benzene
Hydrosymethoxymethyl
2H-l-benzopyran-2-one
Cy Alkylbenzene
Cg Alkylbenzene
Ca Alkylbenzene
C15H24
Unknown
Ethyl ester of dimethyldihydro indene
carboxylic acid
Tripropylbenzene
====3===============================:
0.1
0.1
0.1
0.2
0.2
0.7
0.3
0.2
0.2
0.3
0.2
0.1
0.3
0.3
0.1
0.3
0.1
0.2
0.5
0.2
0.6
0.2
0.6
0.2
0.8
0.4
1.0
1.2
0.5
0.6
0.9
3.0
2.6
1.3
1.3
1.4
60
-------
TABLE A-7. RESULTS OF 6C/MS ANALYSIS
OF CINCINNATI DEWATERED SLUDGE
===========================================
Amount,
Tentative Identification ug/g
1,2,4-Trichlorobenzene 40
n-Dodecane 3
Trichlorobenzene 12
n-Tridecane 9
n-Tetradecane 20
n-Pentadecane 30
n-Hexadecane 30
n-Heptadecane 30
Pristane 30
Biphenyl 2
Diphenyl ether 4
N-Ethyl-N-benzylaniline 20
n-Octadecane 25
Phytane 17
n-Nonadecane 16
Di-n-butyl phthalate 1
n-Eicosane • 9
PAH (MW = 254) 3
Fluoranthene 4
Pyrene 4
Butyl benzyl phthalate 3
Bis(2-ethylhexyl) phthalate 40
l,l-B1s(5-t-buty1-4-hydroxy- 5
2-methylphenyl)-butane
61
-------
TABLE A-8. RESULTS OF GC/MS ANALYSIS
OF LATEX PAINT
==================================================
Amount,
Tentative Identification yg/g
C3~Alkylbenzene 7
CjQ-Alkane 30
C^-Alkane 18
C^Alkyl benzene 3
C^-Alkylbenzene 5
Cn-Alkane 50
Decahydromethylnaphthalene 6
Naphthalene 10
Cjo-Alkane 37
Unknown alcohol 12
TABLE A-9. RESULTS OF GC/MS ANALYSIS
ETHANES -I SPENT FeCl2 CATALYST
============================================,======
Amount,
Tentative Identification ug/9
Chlorinated unknown 320
Chlorinated unknown or mixture 980
Chlorinated unknown or 1,200
Trichloromethylpropene +
Chlorinated unknown
Chlorinated unknown 890
n-Pentadecane 290
n-Hexadecane 790
n-Heptadecane 1,700
Pristane 570
n-Octadecane 2,200
Phytane 1,300
n-Nonadecane 2,400
Alkane 470
n-Eicosane 1,500
==================================================
62
-------
TABLE A-10. RESULTS OF GC/MS ANALYSIS
OF DRYING BED SOLIDS
======================
Tentative Identification
Amount,
yg/g
Vinyl chloride
Chloroethane
Methyl ene chloride
Acetone
Acrolein
Carbon disulfide
Trichlorofluoromethane
1,1-Dichloroethylene
1,1-Dichloroethane
1,2-Dichloroethylene
Chloroform
1,2-01 chloroethane
2-Butanone
Trichloroethylene
Benzene
Diethyl sulfide
Toluene .
Chlorobenzene
Diethyl disulfide
Chlorinated unknown
GH Alkane
1, 2, 4-Tri Chlorobenzene
Naphthalene
Hexachlorobutadiene
Methyl naphtha! ene
Tetrachlorobenzene
Biphenyl
n-Tetradecane
Diphenyl ether
Alkane
n-Pentadecane
Pentachlorobenzene
n-Hexadecane
n-Pentadecane
Hexachlorobenzene
===============================
0.4
5.4
1.3
3.6
1.4
0.4
2
14
0.2
0.3
0.3
0.2
0.1
0.3
0.1
0.1
0.5
0.1
2
4
7.5
10
11
88
5
40
13
30
35
9.0
28
7.0
30
20
18
===========
63
-------
APPENDIX B
QUALITY ASSURANCE/QUALITY CONTROL FOR THE
INTERLABORATORY COMPARISON STUDY
64
-------
QUALITY ASSURANCE/QUALITY CONTROL >•
(i
The QA/QC document presented here 1s primarily Intended for the deter-
mination of semi volatile compounds. A similar document was generated for
use In volatile compound determination and was completed during Phase II. i
QA/QC documentation 1s a part of the protocol development for Phase II !
of this study. However, the progress thus far 1s Included 1n this Phase I
report for review. The plan was subject to change as the result of Phase II I
effort. For Instance, performance criteria was relaxed or made more severe
or the number of performance criteria were changed. Also, the report forms ,
(not Included 1n this package) were tested and designed 1n a final form for ,
Inclusion 1n the test protocol. i
i
Quality Assurance/Quality Control (Semi volatile Compounds) i
*
?
i
Overview !
The objective of QA/QC activities for any analytical program 1s to pro- i;
vide data of known quality. In addition, 1f the results of these analyses
are contested, their quality must be demonstrable. ;
, !
While the data from this study are not apt to be directly contested !.
1n a litigation, the validity of the method may be contested and needs to
be substantiated. Further, the results of the study will form the basis ,<
for quality control requirements when the method is applied routinely. |
i>
QA Objectives
The objectives of QA/QC activities on this program were to assure that the j
work carried out to evaluate the analytical protocols was done under controlled ,
conditions, that those controls were uniformly applied by all collaborators and [
that all work done was recorded for archival storage. f
In addition, when the analysis method is fully evaluated and 1s applied
for routine analyses of hazardous wastes, the method description should 1n- !
elude the necessary and appropriate quality control elements and requirements.
Part of the requirements for those quality assurance and quality control will
be based on experiences and knowledge derived from the evaluation exercise.
Therefore, 1t is expected that all persons involved 1n this program who are
aware of the ultimate use of the methods will be alert to the QA/QC guidelines
which can help to improve the data quality.
Documentation and Records
The documents for this program Include the Manual of Instructions for
Collaborators (of which this QA/QC plan is a part), the Program Review
65
-------
Inquiry, data reports, letters of transmittal, records of telephone
conversations relative to this program and all data and records associated
with this program. Copies of these documents will be kept on file by the
principal investigator for audit purposes and for possible submission to the
subcontractor .at the study conclusion.
A record shall be kept of all efforts and events associated with the
laboratory work and of all data as follows:
• Samples
Date received
Volume and or weight of samples
Condition of samples
Location and temperature of storage
Date removed from storage
t Analysis
Reference method used
Date analysis started
Sample size used for analysis
Final extraction volume
Amount of internal standard added
Internal standard area response
Injection volume
Relative response factors used for quantification
Calculations (examples)
Replicate results
Data acquisition dates
Scan number
Absolute retention time
Most intense ion
Compound identification
Probable molecular weight
Fit and reverse fit
Total ion current chromatographs
Mass spectrum of each scan number
Library search results
9 track tape records
Search system used
Performance check results
Mass spectrometer tuning results
Maintenance record
Almost all of the above information are among the "required data" sub-
mitted by the participating laboratories. These data were either submitted
on supplied forms or in a previously specified format.
66
-------
Quality Control
This section specifies the operations necessary to know and document that
the analysis system is in control before initiating the inter!aboratory test
of Phase III. jThese operations were implemented in the work described in
that Phase.) QC starts with a description of the method which must be followed
without exception. Before any laboratory work is done, the method must be read
and understood by all who will use it. Questions regarding what is to be done
must be discussed with Battelle persons before laboratory work starts. In that
way uncertainties can be corrected or clarified among all participants and all
will possess the same information. Thus, to the extent possible, all partici-
pants will carry out all operations in exactly the same manner.
Control will be maintained by monitoring the performance of mass spectrom-
eter tuning compound (DFTPP), process blanks, calibrating standards and spiked
samples.
The start up will consist of mass spectrometer tuning followed by calibra-
tion of the GC/MS system. Calibration will be done using the supplied standards
in duplicate in the sequence of standard 1 through standard 4 two times. The
daily analysis routine will consist of:
Mass spectrometer tuning (DFTPP)
Process blank
Check standard
Samp!e
Sample
Sample
Sampl e
Check standard.
Each of these analyses will generate data to be reported and to serve as
control.
The four calibration standards will contain 10 internal standard compounds
plus DIO-Phenanthrene, and the tuning compound, decafluorotriphenylphosphine.
The initial tuning will be performed according to EPA criteria (shown in
Form 1). Each time a standard is analyzed (either during calibration or during
a check run) the mass intensities from DFTPP will be recorded on Form 1. The
standard's peak intensities must be within 25% of the intensities generated
during tuning.
The peak area intensity of DIO-Phenanthrene will also provide control.
Its area intensity in every analysis after the calibration runs must be
between 0.7 and 1.4 times the mean peak area intensity of DIO-Phenanthrene
found in the calibration runs. For example, if the mean area intensity for
DIO-Phenanthrene found during calibration is 355,000 area counts, then each
subsequent sample, blank or standard analyzed must have a area intensity for
DIO-Phenanthrene of between approximately 250,000 and 500,000. If an area
intensity is outside this range, the sample shall be reanalyzed and if the area
intensity is still outside the range, the analysis will be stopped in order to
investigate and correct the problem. These data will be recorded on Form 2.
67
-------
Relative retention times will be kept in the range of 0.85 to 1.25
through use of the appropriate internal standard from among the 11 present. A
mean relative retention time will be calculated for each compound during
calibration along with a standard deviation for each mean. These data will be
recorded on Form 3. The relative retention times for each subsequent standard
analyzed must-be within two standard deviations of the mean relative retention
time calculated from the data obtained during the initial calibration. These
data will be recorded on Form 3a. The standard deviation should be <5 second
and therefore the'search window can be kept to ±10 seconds. For purposes of
control, rather than checking retention times on all compounds in the
standard, every fifth compound may be checked.
The initial calibration will also be used to determine response factors.
Once the system has been calibrated, the calibration must be verified with one
standard at least twice each 8-hour shift. If the percent change of response
factor for the calibration check compounds (every fifth compound) averages
greater than twenty percent, the system must be recalibrated:
Percent Change = ERFv - RFcl x 100
Rr
RFc = mean response factor from initial calibration
RFv = response factor from current verification check standard.
Response factors for all compounds are to be recorded using Form 4 for initial
calibration data and Form 5 for daily check standards data.
All samples and blanks will be fortified with two surrogate spiking
compounds before purging or extracting by procedures described in the method.
This will help determine acceptance limits. Calculate percent surrogate
recovery by:
% recovery = «. x 100
Qa
where: Qd - quantity determined
Qa = quantity added.
In this scheme it is assumed that the compound added is not present, in a
detectable amount, before addition of the compound. If a sample does contain
natural amounts of the spiked compound, the % recovery will be in error
because the natural amount present will not be determined. Report the
uncorrected results.
Process blank analysis must be performed once every day. The reagent
blank is used in all analyses to verify that the determined concentrations do
not reflect contamination. The results of blank analysis must be evaluated
immediately after data have been acquired and recorded on Form 7.
If a compound in the standards is detected in the blank, the blank value
is utilized in the sample calculation according to the following options:
68
-------
1. If the concentration in the blank is less than or equal to 1/2 of the
method detection limit, the blank value is ignored.
2. If the concentration in the blank is greater than 1/2 of the method
detection limit and is less than or equal to 1/2 the concentration
detected in a sample, subtract the concentration in the blank from
the concentration in the sample. Record the correct value and
indicate that this had been done by placing a "C" in the flag column
of the data sheet.
3. If the concentration in the blank is greater than 1/2 the method
detection limit and if the blank concentration is greater than 1/2
the concentration detected in a sample, correction is not possible
and the blank must be rerun. If the results are again high, the
samples associated with the blank must be resampled and reextracted.
If the blank problem is associated with volatile compound determina-
tion, the compound should be reported as "NO" but with a "B" in the
flag column of the data sheet. The cause of this high blank must be
determined and corrected before additional samples are analyzed.
Other Forms for data recording include Form 8 on which results of sample
analyses are reported, and Form 9 on which is recorded the samples analyzed
each day.
Data Management
The above section specified 9 data forms to be used to record data
necessary to demonstrate control of the system. The forms are:
Form 1 DFTPP Tuning Criteria
Form 2 DIO-Phenanthrene Peak intensities from initial calibration
Form 2a DIO-Phenanthrene Peak intensities from daily use of standards
Form 3 Results of determination of relative retention time including
average and standard deviation for each compound/internal
standard combination from 2 calibration runs.
Form 3a Results of daily check of RRT from use of check standards
Form 4 Results of calculation of response factors from Initial
Calibration
Form 5 Result of check of response factor from daily check standards
Form 6 Result of spike recovery of three spiked compounds
Form 7 Result of analysis of process blank
Form 8 Results of duplicate analyses of samples
Form 9 A log of samples analyzed daily.
The data entered shall be reviewed by the person who is knowledgeable of
the procedures other than the person who entered the data. The reviewer will
document the review by his signature.
In addition to the above data a supplemental data package will be
submitted to the coordinating laboratory (Battelle). For each sample run this
package will include:
69
-------
• A table of results that includes scan number, compound identification,
absolute retention times, most intense ion, probable molecular weight,
response factor used, amount of compound, and some indication of
goodnes.s-of-fit. Table B-l provides an example of the required data.
• Total ion chromatograph for each sample. See Figure 1.
• Mass spectra for each peak. See Figures 2 and 4.
• Library search results of each scan. See Figures 3 and 5.
• A narrative description of the process used to obtain information from
the GC/MS system -- i.e. were manual or automatic techniques applied,
principle of automatic systems etc.
70
-------
TABLE B-l. RESULTS OF GC/MS ANALYSIS OF SAMPLE NO. EPA/05344, REPLICATE 2
Scan
No.
7
14
30
69
98
173
200
211
216
227
240
280
330
362
378
465
487
521
570
676
529
Absolute
Retention
Tentative Time, tR,
Identification minutes
Bis(chloromethyl ) ether
Butoxyethanol
Dichloropropanol
Chloroethyl benzene + silicone
Chlorinated unknown
Silicone
a-terpineol
Trimethyloxo-hexenoic acid
Unknown alcohol
Unknown alcohol
Unknown
Silicone
Di phenyl ether
Halogenated Unknown
Silicone
Silicone
Hexaethyl benzene(IS)
Unknown
Alkane
Fatty acid
diQ-anthracene( IS)
7.1
7.5
8.4
10.6
12.2
16.3
17.9
18.5
18.8
19.4
20.1
22.4
25.2
27.0
28.0
33.0
34.2
36.3
39.2
45.5
36.8
Most Probable
Intense Molecular
Ion Weight
79
57
62
43
62
73
59
43
59
55
81
73
170
93
73
73
231
69
57
60
188
122
118
128
140
188
S
154
170
>117
>131
NE
S
170
208
NE
NE
246
NE
NE
NE
188
Response
Factor
Used
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
0.79
1.00
1.00
1.00
1.00
Amount,
yg/L
340
120
42
4.6
3.2
S
4.1
8.9
18
28
18
S
11
3
S
S
14
3
3
0.2
100
Fit
912
738
972
828
912
S
953
766
U
U
U
S
996
U
S
S
983
U
A
862
972
Reverse
Fit
710
908
883
280
563
S
764
635
U
U
U
S
895
U
S
S
774
U
A
596
938
NE - Molecular weight not estimable from mass spectrum.
U - Library search list does not match unknown mass spectrum.
S - Silicone bleed peaks not quantified.
A - Alkanes of wrong molecular weight generally selected by computer search.
IS - Internal standard.
-------
TOTAL ION CHROMATOGRAM (X2)
ro
500
1000
Figure 1. TIC of water sample EPA/05344, run 2.
-------
MASS SPECTRUM
CO
q
o>
100.0-1
50.0
o
O T^
r= f*
10 I
..I. 1.1, .1.
ll
o
o>
M/E
I
60
100
140
200
Figure 2. Mass spectrum of scan no. 7 in semi volatile sample no. EPA/05344, run 2.
-------
LIBRARY SEARCH
ACETYL BROMIDE
METHANE. OXYBISCHLORO-
* 9
1, 2-PROPANEDIOL. 3-CHLORO-
_. IT
ETHANONE, 1 -(2-METHYL-2-CYCLOPENTEN-1 -YLJ-
.». - .li
1-PROPANOL. 2, 3-DICHLORO-. ACETATE
I
1.
1
1479
SAMPLE
RANK1
RANK 2
RANK 3
RANK 4
RANK 5
M/E 40 60 80 100 120 140 160 180 200
Figure 3. Library search results for scan no. 7 in semivolatile sample no. EPA.05344, run 2.
-------
MASS SPECTRUM
lOO.Oi
q
N
10
—i
in
50.0-
in
q
K
CO
q
id
M/E 50
8
100
q
oo
oo
n
q q
co K
2 8
eo
-------
LIBRARY SEARCH
1000.
SAMPLE
•Ill
RANK1
•lid
RANK 2
RANKS
RANK 4
LJ
RANKS
M/E 50
ETHANOL. 2-BUTOXY
BUTANE. 1.1 '-OXYBIS
ETHANOL. 1-{2-BUTOXYETHOXY)-
J. .
MORPHOLINE
PROPANOIC ACID. 2-METHYLPROPYL ESTER
100
150
200
250
Figure 5. Library search results for scan no. 14 in semivolatile sample no. EPA/05344, run 2.
-------
PHASE II STUDIES
77
-------
INTRODUCTION
Phase II Studies
Many analytical methods have been developed and applied to the analysis
of organic constituents in solid wastes. Each research group associated with
the development of a given method instituted appropriate intralaboratory
quality control to provide method accuracy and precision. However, few of
these methods have been evaluated by conducting a well-designed interlaboratory
comparison study.
At the time this project was initiated two solid waste analysis methods
were in use in a large number of laboratories:
• Modified NEIC (National Enforcement Investigation Center)
• Modified MRI (Midwest Research Institute).
The interlaboratory program being conducted by Battelle includes three phases:
I. Evaluation of Procedures Based on These Two Methods.
II. Intralaboratory Evaluation.
III. Interlaboratory Evaluation.
The objective of Phase I studies was the evaluation of these two methods for
the analysis of solid wastes: one method for volatile organic analysis and
one method for semivolatile organic analysis. The results of Phase I are
discussed in the Phase I Section of this Report.
The objectives of Phase II included the following:
t Critical review, application, and modification of the methods selected
and described for the determination of volatile and semivolatile
organic compounds in waste samples.
• Development and generation of a broad based review of a protocol for
the conduct of an interlaboratory collaborative study to be carried
out as Phase III of this program.
78
-------
SUMMARY
This report on Phase II of a three phase program for "Evaluation of
Methods for Hazardous Waste Analysis" provides discussions of efforts toward
preparing for an Inter!aboratory test of the selected methods.
The efforts or results for this phase can be summarized as follows:
1. The volatile organic and semlvolatile organic analysis methods were
tested in Battelle's laboratory to determine 1f the methods were
technically satisfactory and clearly defined.
2. The major modification of the method for determination of volatile
compounds was the use of tetraglyme as the solvent for extraction of
volatile organic compounds.
3. The major modification of the method for determination of
semivolatile compounds was the use of an ultrasonic device during
extraction with methylene chloride and neutralization of the matrix
prior to extraction.
4. Many performance checks were Incorporated into the methods to enhance
quality control.
5. The Inter!aboratory experimental design, based on ASTM Methods E
691-69 and E 2777-77, was ultimately based on use of 9 laboratories
and 8 waste samples analyzed in triplicate.
6. A Manual for Collaborators for an Inter!aboratory study was prepared
giving details of the method descriptions and data requirements.
7. Review of the methods and design by technical experts within
Battelle, EPA, and by participants resulted in some design change.
8. Samples were prepared for the participating laboratories. The
samples were spiked with 40 semlvolatile and 10 volatile compounds.
79
-------
TECHNICAL DISCUSSION
Overview
The objective of the Phase II effort was to develop protocols for the
analysis of hazardous wastes. The intent of these methods was to identify and
quantify a broad spectrum of organic compounds rather than to optimize
recovery of specific constituents.
The objective was achieved through a series of simultaneous, intercon-
nected experiments with iterations, regroupings, and reviews within Battelle
and with EPA. The protocols were developed for the inter!aboratory test to be
evaluated during the Phase III effort. The inter!aboratory test results,
along with the program review inquiry results, will provide the basis for
specific language and for quality control policy, events, and criteria for
broadly applicable analytical protocols.
Analysis Methods
The analysis methods as described in Phase I were modified during Phase
II efforts as the result of editorial reviews and application of the methods
in Battelle's laboratories. Some of the modifications were necessary to
remove procedural ambiguities. Other modifications involved technical change
that were a result of experience or external review. The significant changes
are discussed below.
Method for Determination of Volatile Organic Compounds
The changes made in the protocol for determination of volatile organic
compounds since the Phase I report was written are largely the result of using
tetraglyme (tetraethylene glycol dimethyl ether) rather than polyethylene
glycol (PEG 400) as the solvent for extraction of volatile organic compounds.
The greatest advantage to the use of tetraglyme is that tetraglyme is a
more universal solvent than PEG. In addition, tetraglyme is miscible with
hydrocarbon solvents as well as with water, whereas PEG is a very poor solvent
for hydrocarbons. Although PEG was found to be suitable for many of the
particular waste samples included 1n the inter!aboratory study, it would not
be suitable for samples containing high levels of mineral oil or paraffin wax.
A second advantage to using tetraglyme as the extraction solvent is Its
lower viscosity relative to PEG. It was believed that the relatively high
viscosity of PEG would make it difficult to draw up into a syringe. We found
that the PEG could be handled satisfactorily by using wide-bore syringe
needles; however, the large volume contained in the needle made it difficult
to accurately dispense less than 20 uL. The viscosity of tetraglyme is
similar to alcohol and small volumes can be dispensed using standard micro-
syringes. A tetraglyme extract can be Introduced Into the purging chamber by
piercing the septum with the syringe needle. The septum is replaced after
80
-------
each run to avoid any carry-over of components that might have been trapped In
the septum during the Injection.
Tetraglyme 1s available from Aldrlch or Eastman. We have purified tetraglyme
by vacuum stripping to remove volatile components as previously recommended
for the purification of PEG. The batch of tetraglyme purified Initially was
found to be free of Interfering volatile components when analyzed by the purge
and trap GC/MS procedure and was readily dispensed with a microsyrlnge. More,
recent experience with other batches of tetraglyme has indicated that additional
purification 1s necessary. Bonification of tetraglyme solutions of representative
volatile compounds, (methylene chloride, trichloroethylene, and tetrachloroethane)
resulted 1n no significant loss of the compounds. The one-half inch probe
that is described for the dispersion of samples in the method for determination
of semivolatile compounds is used for Bonification. The change to tetraglyme
led to other changes, such as making of standard solutions, permissible use of
narrow gauge syringe needles, and changes throughout the sections on extraction.
The method was written to be applicable to a broader range of solid waste
types than the representative sample types chosen for this study. These changes
are reflected in more general statements on means and implements for effective
sample transfer, a clearer statement on obtaining and using response factors,
and a more general statement on record keeping for quality control purposes.
On the other hand, specificity was increased in instructions for making cali-
bration standards, for daily performance requirements, and for estimation of
major volatile compound content. Lastly, surrogate and internal standards
were selected from among fluorinated (surrogate) and deuterated (internal)
compounds to reduce the possibility of adding compounds for control that may
occur naturally in a waste.
Method for Determination of Semivolatile Compounds
The method for determination of semivolatile compounds evolved out of
efforts to Improve extraction procedures and quality control. The homogenlza-
tlon step was changed from using a Polytron homogenlzer to using an ultrasonic
device. The changes in aspects of the method that relate to performance checks
were all directed toward keeping the method in control so that reliable data
of known quality are generated.
The solvent for extracting semivolatile compounds, methylene chloride,
must be brought into contact or homogenized with the sample to be extracted.
For some samples a Brinkman Polytron homogenizer Is satisfactory for the
extraction process. The Polytron is a mechanical shearing device that operates
at sonic frequency and can homogenize most samples. However, the device is
somewhat difficult to clean and the Teflon bearing deteriorates rapidly when
used on a sample that has high amounts of abrasive matter such as sand or
ceramic catalyst material. As discussed in the Phase I report, Bonification
was examined as a means for homogenization but under the conditions used was
found not to be generally applicable. Nevertheless, because of the rapid
wear of the Polytron (sometimes the Teflon bearing needed changing after only
three or four samples) it was necessary to reexamlne the use of a sonic probe
to homogenize and disperse samples in solvent.
81
-------
On the basis of conversations with EPA, NEIC personnel and
representatives of Heat Systems, the manufacturer of the sonic probe, it
appeared that the sonic probe operating conditions were not optimized in Phase
I efforts. The procedure used by NEIC involved the sonification of a 1-gram
sample with 10-mL of methylene chloride in a 20-mL glass scintillation vial.
By using a micro probe the dispersion was reportedly to be entirely adequate.
The use of a 1:10 ratio of sample to solvent instead of a 1:50 ratio probably
permits a significantly better energy transfer from probe to sample. In Phase
I, a larger amount of solvent was used in an attempt to minimize saturation of
the solvent with the extracted components. We also included a neutralization
step and a drying step to permit acids and bases to be extracted simultaneously,
Nevertheless, the manufacturer reported to Battelle that by selecting the
proper geometry, reducing the power output, and perhaps using a pulse mode it
should be possible to achieve excellent dispersion.
Sonification studies were continued using a standard 1/2-inch probe, a
1/2-inch probe with a 5-inch half-wave extender, and an 1/8-inch microprobe.
The waste samples that were studied included a wet municipal sludge, an oily
soil, ethanes spent catalyst, drying bed solids, and oxychlorination catalyst
pellets. An effort was made to optimize the power input, duty cycle, probe
position and sample volume.
The sonification protocol that appeared to work best involved:
1. The addition of 15 ml. of methylene chloride and 1 ml of buffer to 3g
of sample in a 250-mL centrifuge tube.
2. Sonification for one-minute using a 1/2-inch probe with an extender
and operated at a power setting of 50% and a pulsed duty cycle of
50%.
3. Addition of 135 ml of methylene chloride and sonification as above
for one minute.
4. Addition of the specified amount of anhydrous sodium sulfate followed
immediately by vigorous shaking.
5. A final sonification as described in 2, for one minute.
For all sonifications the probe is positioned to place the tip 0.5 to 1 cm
beneath the surface of the liquid. The above protocol worked well for all
samples; however, the catalyst pellets were not broken up. Although the
extraction of the pellets may be complete with sonification alone, we
specified that the particle size of all such samples be reduced to less than
0.1 mm diameter before the required 3-g sample is taken. A glass mortar and
pestle is recommended for grinding the sample; however, many alternate methods
would be suitable.
The method for the determination of semivolatile organic compounds was
also revised slightly to accommodate wet sludges that do not disperse well in
methylene chloride. The revision specifies the use of homogenization after
the addition of sodium sulfate as well as before.
82
-------
In several cases the extraction of waste samples with methylene chloride
gave turbid extracts that form precipitates when concentrated. In some cases
the turbidity gradually decreased upon standing. Centrifugation or filtration
through a fritted funnel or through a dense glass fiber filter did not clarify
the extracts. However, clarification can be achieved by filtration through a
0.2-um membrane filter. This step was subsequently added to the procedure.
This change improved the reproducibility of the residue weight determination,
decreased precipitation during the concentration step, decreased the amount of
nonvolatile material injected into the GC/MS system, and improved the overall
reproducibility of the method.
In most cases the membrane filtration can be accomplished by using a
filter holder fitted to a syringe. For extracts that are extremely difficult
to filter, the filtration assembly used for the Extraction Procedure (1) works
well if a fluoropolymer membrane filter and a Viton cylinder gasket are used.
Only one of the samples studied required this larger assembly. Based on the
above findings the method for the determination of semi volatile organic com-
pounds in solid wastes was revised to include the use of membrane filtration.
Instructions for the use and calibration of gel permeation chromatography
were added for applications of these methods beyond the scope of this study.
The analyses of waste materials selected for the collaborative study do not
require use of 6PC cleanup.
• •
Several procedures to enhance quality control and data quality were
incorporated into the method. These include:
• Check of ion intensity ratios of decafluorotriphenylphosphine (DFTPP)
in every analysis run. (This procedure was subsequently abandoned as
the degradation rate of DFTPP in these waste samples was too rapid).
• Use of eight internal standards.
• Use of surrogate standards.
• Performance criterion based on response of an internal standard
(DIO-Phenanthrene).
t Use of column performance standards.
These procedures were incorporated in the method to provide a check of
instrument performance in every analysis run. Even though performance
criteria are given, the data from participating laboratories will be analyzed
to determine which criteria are useful and what acceptable performance limits
are useful.
Decafluorotriphenylphosphine (DFTPP) is used as the mass spectrometer
tuning compound and it was considered desirable to monitor the stability and
correctness of tune in every analysis run. The ions to monitor (m/e 69, 127,
275 and 442) were selected from among those used for the initial tune and on
the basis of avoiding interference from ions of compounds apt to be found in
extracts of wastes. However, as is discussed in the section of this report
entitled Proposed Additional Studies, DFTPP solutions frequently degrade
83
-------
within days after a solution is exposed to air. The result is that the amount
of DFTPP remaining in solution is insufficient to provide good relative ion
abundance data when analyzed, therefore, we do not recommend this criterion
be used in Phase III. Recommendations are made in that section for studies to
eliminate the degradation problem.
The semivolatile internal standard compounds used in the program are:
Ds-bromobenzene Ds-naphthalene
DiQ-phenanthrene Dio-biphenyl
Dio-acenaphthene Dio-pyrene
Di2-chrysene
Use of these 8 internal standards is based on the demonstrated benefit of
associating an analyte compound with an internal standard compound that has a
similar retention time. It was shown in the Phase I report that the repeata-
bility of relative retention time (RRT) is highest when the RRT is near 1.0.
Thus, we have selected internal standards that will give RRTs of between 0.8
and 1.2. The greater the precision of the RRT, the smaller the retention time
window and the purer the mass spectra obtained. The use of internal standards
having characteristic ions similar to those of the analytes produces more
reproducible response factors than in cases where .there are gross differences
in characteristic ions.
Surrogate standards ( decaf 1 uorobi phenyl , 2-fluoroaniline and pentafluoro-
phenol) are added to the sample before extraction in order to facilitate
monitoring of the extraction process. If surrogate standards are recovered
poorly while other indicators are within criteria, the indication is that the
extraction process is out of control (see page 233).
The method specifies that the response for DIO-phenanthrene must be
within set limits. This criterion forms the basis for setting the electron
multiplier gain of the mass spectrometer within specified limits that provides
the required dynamic range.
Column performance is assessed through use of representative acidic,
basic and polar neutral compounds. These are:
Ds-aniline Ds-phenol
Ds-nitrobenzene D3-2,4-dinitrophenol
The response factors of these deuterated compounds relative to DIO-phen-
anthrene will provide information on possible deterioration of the GC column.
A further check on the column is accomplished with every check standard by
monitoring pentachlorophenol and 2,4-dinitro-3-chloroaniline which are present
in the standards. The response factors of these compounds (relative to
DIO-phenanthrene) should be at least 0.05. Any degradation of the column
will be immediately apparent by loss of detection of these compounds.
84
-------
ASTM Review
The InterTaboratory study protocol was designed to comply with ASTM
Method E 691-79 (Standard Practice for Conducting An Interlaboratory Test
Program to Determine the Precision of Test Methods) (2) and ASTM Method D
2777-27 (Standard Practice for Determination of Precision and Bias of Methods
of Committee D-19 on Water) (3). The design was reviewed by Dr. Lyman Howe,
Methods Advisor for ASTM D-19, Committee on Water, and Dr. Robert Paule of
ASTM D-34, Committee on Wastes and ASTM E-ll, Committee on Statistics. The
results of those reviews initiated a change in the number of replicate
analyses (from 2 to 3) and the number of samples analyzed (from 10 to 8).
The change to triplicate rather than duplicate analyses was based in
part by guidelines in ASTM D2777-77. That practice contains an inequality
fr >1 + (30/P)] to calculate the number of replicates (r) to be made for a set
of conditions. In the inequality P is the product of number of concentration
levels (2), number of operators (1), number of apparatuses (1) and number of
laboratories (10). Thus P for this program 2x1x1x10 or 20 r >1 + 30 or r
>2.5. 20~
There were several strategies possible to make the test protocol satisfy
the inequality. However, in order to keep the participant effort cost effec-
tive it was suggested that the number of waste samples analyzed be reduced and
the number of replicates be increased to 3. That suggestion was adopted, the
protocol was changed accordingly and the number of waste samples reduced to 8
samples for the determination of volatile compounds and 6 samples for the
determination of semi volatile compounds.
Other comments resulting from review by ASTM can be summarized by stating
that while the conditions of the test obey the principles of E 691, the program
study goals require that many answers be determined from few measurements and
that some assumptions must be made. The assumptions mentioned by reviewers
Included the following:
• Assumption of zero blank level.
• Assumption that paired spiked compounds behave identically during
analysis.
• Assumption that the spike compounds do not Interact with each other.
t Impossibility of obtaining a smoothed function relating precision and
accuracy to concentration level.
It was not intended to determine the repeatability and reproducibility inter-
vals for the method as a smoothed function of the level but rather for only two
representative levels. It may be sufficient to determine the repeatability and
reproducibility at a single level, but a second level was included simply to
provide an indication of the effect of concentration.
35
-------
To provide blank measurements during Phase III, the participating
laboratories wi.ll analyze several process blanks but not unspiked samples.
Single laboratory replicate analyses of unspiked samples have been conducted
and in most cases the spike compounds were not detected in the unspiked
samples. Thus, when zero is used as the blank value, it will not be an
assumed value but will be based on experimental data. Admittedly, there
would be greater certainty if each laboratory were to establish that the
blank value is zero, but it is believed that the additional expense involved
in the required measurements is not justifiable.
The assumption that the paired compounds are identical in nature is a
rather critical assumption. Therefore, a pilot study was conducted as part
of Phase II to provide an indication of the validity of this assumption.
However, since we are not attempting to establish a smooth curve for recovery
versus spike level, the assumption is not as critical as it would be otherwise.
This study is discussed in a later section of this report.
We recognize that the study goals are ambitious in attempting to study
both a broad range of waste types and a broad range of chemical classes.
We also recognize that interactions among the spike compounds, interactions
between the spike compounds and the wastes, and sample homogeneity are uncon-
trolled variables. The spike compounds were chosen on the basis of expected
chemical stability in an effort to minimize interactions.
Waste Sample Selection
The wastes being used for the inter!aboratory study are shown in Table 1.
These wastes were selected:
• To represent as many waste types as possible.
• To contain a variety of compounds that are potentially hazardous.
• To represent a challenge in extraction.
t To include significant amounts of many target compounds.
This list departs somewhat from that given in the Phase I section to reduce
the number of samples.
The drying-bed solids and oil refining wastes that were listed in the
Phase I report were deleted because these wastes contain an oily matrix that
was somewhat similar to the organic content of the contaminated soil sample.
The herbicide acetone-water and TCE spent caustic samples were retained for
the analysis of volatile components because they were distinctly different from
86
-------
TABLE 1. WASTE SAMPLES SELECTED FOR THE INTERLABORATORY STUDY
Program
Identification
Number
ILS-2
ILS-3
ILS-4
ILS-5
ILS-6
ILS-7
ILS-8
ILS-9
Waste Name
Latex paint waste
Ethanes spent catalyst
Coal-gasification tar
Oxychlori nation spent catalyst
POTW sludge
Herbicide acetone-water
Chlorinated ethanes waste
Contaminated river sediment
Physical
Description
Semi -sol id
Oily powder
Tar
Pelletized solid
Wet filter cake
Liquid
Liquid
Powder
==S=3=============================================
any of the other samples 1n respect to volatile content. However, these two
samples were not analyzed for semi volatile components because they present
no extraction problems and have no unique semivolatile components.
The selected wastes were spiked and were analyzed for semivolatile and
volatile compounds as shown below:
ILS-2 Semivolatile and volatile
ILS-3 Semivolatile and volatile
ILS-4 Semivolatile and volatile
ILS-5 Semivolatile and volatile
ILS-6 Semivolatile and volatile
ILS-7 Volatile only
ILS-8 Volatile only
ILS-9 Semivolatile only.
Phase III of the program included a laboratory performance evaluation.
The samples for the performance evaluation include ILS-1, a methylene chloride
extract of ILS-1 for semi volatile compound determinations and a tetraglyme
extract of ILS-2 for volatile compound determinations. These extracts were
identified as ILS-10 and ILS-11, respectively, and were included as perfor-
mance evaluation samples to evaluate the reproducibllity of the analysis In-
dependent of the variability of extraction. These waste samples were spiked,
homogenized, and packaged in 25-ml vials with Teflon-lined screw caps. ILS-9
was the NBS Standard Reference Material No. 1645, which was already homogenized
and was analyzed for reference purposes. ILS-9 provided recovery data on
naturally incorporated compounds.
87
-------
Use of Analysis Protocols
Target list of 200 compounds. The 60 volatile compounds and 140
semi volatile compounds on the "target 11st" were selected by us to provide a
rigorous test of the applicability of the methods. Compounds are Included on
the 11st that were unlikely to be detected due to extraction, chromatographlc
or Instrumental difficulties. The evaluation of the methods 1n Phase II has
shown that some of the compounds are Indeed not determined. Additional com-
pounds will be excluded from the scope statement as a result of Phase III
activities. The problem compounds Identified to date should be excluded from
the list for use of the method at this time. It is understood that the methods
are considered to be the best available methodology and are thus currently in
use within the Agency. Since the list of 200 compounds was selected to
challenge the method during the course of the Collaborative Study, these com-
pounds should not be expected to constitute a hazardous waste priority list at
this time. It is suggested that the Agency applications of these methods be
reviewed to determine If the method discussions are appropriate at this stage
of the development and evaluation program.
Spiking
The usefulness of adding a compound (spiking) to a sample to be analyzed
is a well established technique of analytical chemistry. With this technique
one can obtain information on precision and recovery (accuracy) at various
concentrations of the compound added. Typically, this information is obtained
through replicate analysis of an unsplked sample and of the sample spiked at a
minimum of 2 concentrations of the compound or compounds of Interest.
A goal of this program was to test two methods, one for determination of
semlvolatile compounds and one for determination of volatile compounds, on a
wide variety of wastes that represent all waste types. Six wastes have been
selected for testing the determination of semlvolatiles organic compounds and
eight wastes have been selected for testing the determination of volatile
organic compounds.
Because of budgetary constraints, a spiking scheme was devised that
minimizes the number of analyses per test waste. The scheme involves spiking
each waste 1n advance with 25 pairs (a total of 50 compounds) of chemically
compatible reference compounds. Each of the 25 pairs represents different
compound classes. The two compounds in each pair were selected on the basis of
similar properties (volatility, solubility, polarity, or acidity) that would
lead to the same recovery efficiencies. One compound from each pair was spiked
at a relatively low level and the other at a considerably higher level. Most
of the spiked compounds were different from the components in the unspiked
waste. The samples were homogenized after spiking and all quoted for shipment
to the participating laboratories.
Each laboratory has been requested to analyze three replicates of each
sample to determine volatile components and three replicates to determine
88
-------
semi volatile components. This approach was designed to provide a cost effec-
tive program. With this approach, data on the determination of unspiked
components and data on the recoveries of different classes of compounds at two
spike levels will be obtained simultaneously with a single run.
The high and low levels used correspond to those levels that will give
approximately 250-ng and 50-ng of volatile compounds or 50-ng and 10-ng of
semi volatile compounds on the GC column during analysis if 100% recovery is
achieved. Since the degree of dilution or concentration required for each
waste varies widely from waste-to-waste, the actual spike level used also
varies widely from waste-to-waste.
A varied spiking scheme was used so that the ratio between the two
compounds in each pair is not always 5:1. Six different ratios, 4:1, 5:1,
6:1, 1:4, 1:5, and 1:6, were used in a varied manner for the 25 pairs of
compounds. In this way there is not any analyst-recognizable spiking pattern
from waste-to-waste that can lead to a bias in the data. However, the high:low
ratios are always close enough to 5:1 to avoid any significant effect on
recoveries. The need for a study to ascertain that the two compounds in each
pair are similar in respect to recoveries achievable at high and low
concentrations was discussed in the Phase I report. That study (Pilot Study)
is discussed in a later section of this report.
The spiking compounds shown in Table 2 were spiked into waste samples at
the concentrations shown in Table 3 (semivolatile compounds) and Table 4
(volatile compounds).
Although the ten volatile compounds listed in Table 2 were spiked into
each waste, the compound paired with 2-hexanone, namely cyclopentanone, is not
purgeable under the conditions of the method, and was not determined in the
inter!aboratory study. The performance of cyclopentanone was not discovered
in time to select and add a substitute compound.
Since the wastes were spiked at relatively high levels, it was not possi-
ble to use solutions of the spiking compounds without drastically changing the
nature of the sample and interfering with the determination of volatile com-
pounds. The only exception was in the spiking of oxychlorination catalyst
pellets in which the semivolatile spike compounds were dissolved in methanol
and the methanol was removed prior to spiking with volatile compounds. The
volatile compounds were readily miscible with each other and were mixed to
give a single solution prior to spiking. The neat semivolatile compounds were
added to a glass mortar and pestle and ground to a thin slurry of finely
divided particles prior to spiking.
The spiking of each waste was conducted in a manner that would yield a
spiked sample that was as homogeneous as possible and not subject to phase
separation. The spiking was conducted during the month of September, 1981.
The procedures used for each waste are described below.
Sample ILS-1 (Contaminated Soil). Semivolatile spike compounds (4.64 g
total) were measured into a mortar and ground and blended. Soil (300 g) was
added to the mortar and mixing was continued to blend compounds and soil. The
89
-------
TABLE 2. SPIKING COMPOUNDS
Low-boiling Halocarbons
1,1,1-Trichloroethane
1,2-Dichloropropane
High-boiling Halocarbons
Bromoform
1,1,2,2-Tetrachloroethane
Aromatic Hydrocarbons
Ethyl benzene
Chlorobenzene
Volatile Compounds
Ketones
Cyclopentanone
2-Hexanone
Nit riles
Propionitrile
2-Chl oroacryl oni tri 1 e
Semi volatile Compounds
Aliphatic Halocarbons
Hexachloroethane
Hexachloropropene
Low-boiling Aromatic Halocarbons
4-Chlorotoluene
1,4-Di chlorobenzene
High-boiling Aromatic Halocarbons
Pentachlorobenzene
Hexachlorobenzene
Chlorinated Pesticides
p.p'-DDD
p.p'-DDT
Low-boiling PAHs
2-Ethylnaphthalene
1-Chloronaphthalene
Chioroanilines
4-Chloroaniline
4-Chloro-2-methy 1ani 1 i ne
Nitroanilines
3-N1troaniline
2-Chloro-4-nitroani 1 i ne
Pyri dines
2,4,6-Trimethylpyridine
4-t-Butylpyridine
Qu inclines
Quinoline
4-Methylquinoline
Haloethers
Bi s(2-chloroethyl)ether
Bi s(2-chloroethoxy)ethane
(continued)
90
-------
TABLE
2. (Continued)
Semi volatile Compounds (continued)
Mid-boiling PAHs
Fluoranthene
Pyrene
High-boiling PAHs
1,2,5,6-Dibenzoanthracene
1,2,7,8-Dibenzocarbazole
Aromatic Nitro Cpds.
1,3-Di nitrobenzene
2,6-Dinitrotoluene
Low-acidity Phenols
2-Chlorophenol
2,6-Dimethyl phenol
High-acidity Phenols
4-Nitrophenol
2,4-Dinitrophenol
Phosphates
Triphenyl phosphate
Tri-p-tolyl phosphate
Quinones
Anthraquinone
2-Methylanthraquinone
Aromatic Ketones
Acetpphenone
Propiophenone
Benzoic Acids
4-Chlorobenzoic acid
4-Bromobenzoic acid
Phenoxyacetlc Acids
2,4-D
2,4,5-T
==============================================
mix was transferred with 100 ml of water to a glass jar and homogenized using
a Polytron homogenizer for 10 minutes. The water was used to rinse the mortar
and to create a slurry to promote homogenization. The contents of the jar
were kept at <40°C through use of an ice bath. The homogenized slurry was
transfered quantitatively to a Bramley Mill containing 595 g of the soil, the
neat volatile spike compounds added, the mill sealed air tight, and the
contents milled for 20 minutes. A Bramley Mill is a heavy-duty mill used for
mixing rubber formulations; it consists of a stainless steel bowl, two
horizontal sigmoidal blades that turn in opposite direction to provide high
shear, and a gas-tight Teflon-lined cover.
Sample ILS-2 (Latex Paint Waste). A 1000 g portion of the waste and 110
g of water were homogenized with a Polytron homogenizer. A 988-g portion of
this homogenized waste was added along with semivolatile spike compounds (11.9
g total) to a jar and homogenized with a Polytron for 20 minutes while keeping
the temperature at <40°C with an ice bath. The volatile spiking compounds
(5.8 g total) were added, the jar closed, and the contents mixed by rolling
the jar for 24 hours.
91
-------
TABLE 3. SEMIVOLATILE COMPOUND SPIKE LEVELS
ro
Spike Concentration,
Compound
Number(l)
5
12
13
15
16
22
25
28
33
36
43
45
50
51
56
57
64
65
66
69
74
76
77
80
84
85
98
Compound
4-Chlorotoluene
Bi s (2-chl oroethyl )ether
2-Chlorophenol
2,4,6-Trimethylpyridine
1 ,4-Di chl orobenzene
Acetophenone
Hexachl oroethane
4-t-Butylpyridine
2, 4-Dimethyl phenol
Propiophenone
4-Chloroaniline
Hexachl oropropene
Qu incline
Bi s (2-chl oroethoxy Jethane
4-Chl oro-3-methyl ani 1 i ne
Chlorobenzoic acid
1-Chl oronaphthal ene
4-Methylqu incline
2 Ethyl naphthalene
4-Bromobenzoic acid
1,3-Di nitrobenzene
2,6-Dinitrotoluene
3-Nitroaniline
2 , 4-Di ni t rophenol
4-Nitrophenol
Pentachl orobenzene
2 -Chi oro-4-nit roani 1 i ne
Spiked
Pair
No. (2)
2
15
9
13
2
18
1
13
9
18
12
1
14
15
12
19
5
14
5
19
8
8
11
10
10
3
11
ILS-1
Soil
40
40
40
40
200
40
40
200
200
240
40
160
40
160
200
40
160
160
40
160
40
160
40
240
40
200
160
ILS-2
Latex
600
500
600
600
100
100
500
100
100
400
600
100
500
100
100
500
100
100
500
100
500
100
500
400
100
100
100
ILS-3
EtCat
800
200
800
800
200
1,000
200
200
200
200
800
1,200
200
1,200
200
200
1,200
1,200
200
1,200
200
1,200
200
200
1,000
200
1,200
ILS-4
Coal Tar
1,900
11,000
1,100
1,900
7,600
1,900
11,400
7,600
7,600
9,500
1,900
1,900
11,400
1,900
7,600
11,400
1,900
1,900
11,400
1,900
11,400
1,900
11,400
9,500
1,900
9,600
1,900
yg/g
ILS-5
Oxycat
4
24
4
4
16
4
24
16
16
20
4
4
24
4
16
24
4
4
24
4
24
4
24
20
4
16
4
ILS-6
POTW
20
20
20
20
100
20
20
100
100
120
20
80
20
80
100
20
80
80
20
80
20
80
20
120
20
100
80
(continued)
-------
TABLE 3. (Continued)
CO
Spike Concentration,
Compound
Number(l)
100
102
113
117
118
120
121
123
125
126
132
137
139
Compound
Hexachlorobenzene
2,4-Dichlorophenoxyacetic acid
2,4,5-Trichlorophenoxyacetic acid
Anthraquinone
Fluoranthene
2-Methy 1 anthraqui none
Pyrene
4, 4' -ODD
4, 4 '-DDT
Triphenyl phosphate
Tri-(p-to1y)phosphate
Dibenzocarbazole
Di benz o ( a , h ) a nth racene
Spiked
Pair
No. (2)
3
20
20
17
6
17
6
4
4
16
16
7
7
ILS-1
Soil
40
40
200
40
40
200
240
40
240
40
200
160
40
ILS-2
Latex
600
600
100
600
100
100
400
100
400
600
100
100
500
ILS-3
EtCat
800
800
200
800
1,000
200
200
1,000
200
800
200
1,200
200
ILS-4
Coal Tar
1,900
1,900
7,600
1,900
1,900
7,600
7,500
1,900
9,500
1,900
7,600
1,900
11,400
ug/g
ILS-5
Oxycat
4
4
16
4
4
16
20
4
20
4
16
4
24
ILS-6
POTW
20
20
100
20
20
100
120
20
120
20
100
80
20
(1) From form 2S-Study Manual
(2) From Table 2 - This report.
-------
TABLE 4. VOLATILE COMPOUND SPIKE LEVELS
Spike Concentration, yg/g
Compound
Number(l) Compound
12
24
25
33
44
46
47
50
51
Propionitrile
2-Chloroacrylonitrile
1,1,1-Trichloroethane
1 , 2-Di chl oropropane
Bromofonm
2-Hexanone
1,1,2,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
Pair
No. (2)
E
E
A
A
B
0
B
C
C
ILS-1
Soil
5
25
20
5
25
20
5
5
30
ILS-3
Latex
200
1,000
800
200
1,000
800
200
200
2,500
ILS-3
EtCat
500
2,000
500
3,000
2,000
500
500
500
2,500
ILS-4
ILS-5
Coal Tar Oxycat
500
100
100
400
100
100
500
600
100
25
5
5
20
5
5
25
30
5
ILS-6
POTW
5
30
25
5
30
25
5
20
5
ILS-7 ILS-8
Herb Cl Et
Acet
600
100
100
500
100
100
600
100
400
Waste
16,000
4,000
24,000
4,000
4,000
24,000
16,000
20,000
4,000
(1) From form 2V-Study Manual
(2) From Table 2 - This report.
-------
Sample IlS-3 (Ethanes Spent Catalyst). The semi volatile spike compounds
(22.6 g total) were ground and blended In glass mortar and mixed thoroughly
with 130.4 g of the waste. This mix, along with 823 g of waste and the
volatile spike.compounds (15 g total), was mixed in the Bramley Mill for 20
minutes.
Sample ILS-4 (Coal Tar). The semivolatile spike compounds (22.6 g total)
were ground and blended In a glass mortar. Three increments of the waste
sample totaling 774 g were mixed 1n the same mortar and transferred incremen-
tally to a glass jar where the mix was homogenized with the Polytron at 55°C
for 20 minutes. The volatile spike compounds (5.8 g total) were added and
blended by rolling the jar for 12 hours. Upon standing for a few hours, a
liquid layer developed comprised mainly of previously emulsified water. Fifty
grams of mlcrocrystalline cellulose was added to the mix to absorb the water
and the jar was rolled for an additional 12 hours. This mix remained as a
single phase.
Sample ILS-5 (Oxychlorination Catalyst). The semi volatile spike com-
pounds (0.5 g total) were dissolved In 500 mL of methanol and the solution
added to a 2-L round-bottom flask along with 1000 g of the sample. The
methanol was evaporated at room temperature by placing the flask on a rotating
evaporator. The sample was transferred to a wide-mouth glass jar. A
cellulose extraction thimble was added to the jar, the volatile spike
compounds (0.145 g total) were added to the cellulose thimble, and the jar was
Immediately sealed with a Tefl'on-lined cap and rolled for 30 hours, after
which the thimble was discarded.
Sample ILS-6 (POTVJ). The semlvolatile spike compounds (2.3 g total) were
ground and blended in a glass mortar. Three 50-g allquots of sample were
added incrementally to the mortar, mixed and transferred to a glass jar along
with more sample to total 997.7 g. This mix was homogenized for 10 minutes
with a Polytron homogenlzer. The homogenized mix was transferred to the
Bramley Mill along with the volatile spike compounds and milled for 20
minutes.
Sample ILS-7 (Herbicide Acetone Haste). The volatile spike compounds
(3.1 g total) were added to 997 g of f11tered waste in a glass jar and the jar
rolled for 4 hours.
Sample ILS-8 (Chiorinated Ethanes Waste). The volatile spike compounds
(180 g total) were added to 1320 g of the waste in a glass jar and the jar
rolled for 4 hours.
Sample Analyses
Determination of Volatile Organic Compounds
Analyses of waste samples spiked with volatile compounds were conducted
within about five months after spiking. Results for these analyses are shown
1n Tables 5 through 12 in terms of amount spiked, amount found (average of
triplicates), percent recovery (not corrected for background) and relative
95
-------
TABLE 5. ANALYSIS RESULTS FOR SPIKED SAMPLE ILS-1 (Creosote
Contaminated Soil) FOR VOLATILE ORGANIC COMPOUNDS
uq/q
Compound
Propionitrile
2-Chloroacrylom'trile
1,1,1-Trichloroethane
1,2-Dichloropropane
Bromoform
1,1,2,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
2-Hexanone
CPD No.
12
24
25
33
44
47
50
51
46
Spike
5
25
20
5
25
5
5
30
20
Found(l)
N(2)
1.6
6.3
1.9
10
3.2
4.7
19
7.8
% Recovery
N
6
31
38
40
63
95
62
31
% RSD
ND(3)
6
9
5
17
5
1
2
10
(1) Average of triplicate results.
(2) Compound not detected.
(3) Not determined.
TABLE 6. ANALYSIS RESULTS FOR SPIKED SAMPLE ILS-2 (Latex
Paint Waste) FOR VOLATILE ORGANIC COMPOUNDS
uq/q
Compound
Propionitrile
2-Chloroacrylonitrile
1,1,1-Trichloroethane
1,2-Dichloropropane
Bromoform
1,1,2 , 2-Tet rachl oroethane
Chlorobenzene
Ethyl benzene
2-Hexanone
CPD No.
12
24
25
33
44
47
50
51
46
Spike
200
1,000
800
200
1,000
200
200
1,200
800
Found(l)
265
580
1,090
250
1,150
258
300
>1,900
720
% Recovery
132
58
136
125
115
129
150
>150
90
% RSD
11
3
9
4
14
1
4
ND(2)
6
(1) Average of triplicate results.
(2) No meaningful value available.
96
-------
TABLE 7. ANALYSIS RESULTS FOR SPIKED SAMPLE ILS-3 (Ethanes
Spent Catalyst) FOR VOLATILE ORGANIC COMPOUNDS
yg/g
Compound
Propionitrile
2-Chl oroacryl oni tril e
1 ,1 ,1-Trichl oroethane
1,2-Dichloropropane
Rromoform
1,1, 2, 2-Tetrachl oroethane
Chlorobenzene
Ethyl benzene
2-Hexanone
CPD No.
12
24
25
33
44
47
50
51
46
Spike
500
2,000
500
3,000
2,000
500
500
2,500
500
Found(l)
N(2)
370
1,670
620
550
160
183
394
110
% Recovery
N
18
330
21
28
32
37
16
22
% RSD
ND(3)
36
36
40
31
19
8
45
10
(1) Average of triplicate results.
(2) Compound not detected.
(3) Not determined.
TABLE 8. ANALYSIS RESULTS FOR SPIKED SAMPLE ILS-4 (Coal Tar)
FOR VOLATILE ORGANIC COMPOUNDS
Compound
Propionitrile
2-Chl oroacryl oni tri 1 e
1 ,1 ,1-Trichl oroethane
1,2-Dichloropropane
Bromoform
1 ,1 ,2 , 2-Tetrachl oroethane
Chlorobenzene
Ethyl benzene
2-Hexanone
CPD No.
12
24
25
33
44
47
50
51
46
Spike
500
100
100
400
100
500
600
100
100
Found(l)
390
N(2)
140
480
87
N
690
120
85
% Recovery
79
N
140
120
87
N
120
120
85
% RSD
19
ND(3)
14
11
16
ND
9
8
11
(1) Average of triplicate results.
(2) Compound not detected.
(3) Not determined.
97
-------
TABLE 9. ANALYSIS RESULTS FOR SPIKED SAMPLE ILS-5 (Oxychlorination
Catalyst) FOR VOLATILE ORGANIC COMPOUNDS
uq/q
Compound
Propionitrile
2-Chloroacrylonitrile
1 , 1 , 1 -Tri chl oroethane
1,2-Dichloropropane
Bromoform
1,1, 2, 2-Tetrachl oroethane
Chlorobenzene
Ethyl benzene
2-Hexanone
CPD No.
12
24
25
33
44
47
50
51
46
Spike
25
5
5
20
5
25
30
5
5
Found(l)
14
1.2
0.9
1.6
0.7
11
10
1
2.2
% Recovery
56
25
18
8
15
45
33
20
43
% RSD
19
20
37
39
44
20
18
23
3
(1) Average of triplicate results.
TABLE 10. ANALYSIS RESULTS FOR SPIKED SAMPLE ILS-6 (Cincinnati
Dewatered Sludge) FOR VOLATILE ORGANIC COMPOUNDS
uq/q
Compound
Propionitrile
2-Chl oroacryl onitri 1 e
1,1,1-Tri chl oroethane
1,2-01 chl oropropane
Bromoform
1,1, 2, 2-Tetrachl oroethane
Chlorobenzene
Ethyl benzene
2-Hexanone
CPD No.
12
24
25
33
44
47
50
51
46
Spike
5
30
25
5
30
5
20
5
25
Found(l)
N(2)
N
1.2
1.3
N
N
10
3
1.6
% Recovery
N
N
5
27
N
N
53
60
6
% RSD
ND(3)
ND
5
40
ND
ND
32
39
72
(1) Average of triplicate results.
(2) Compound not detected.
(3) Not determined.
98
-------
TABLE 11. ANALYSIS RESULTS FOR SPIKED SAMPLE ILS-7 (Herbicide Manufac-
turing Acetone-Water Waste) FOR VOLATILE ORGANIC COMPOUNDS
ug/q
Compound
Propionitrile
2-Chloroacrylom'trile
1,1,1-Trichloroethane
1,2-Dichloropropane
Bromoform
1,1,2,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
2-Hexanone
CPD No.
12
24
25
33
44
47
50
51
46
Spike
600
100
100
500
100
600
100
400
100
Found(l)
790
N(2)
40
330
43
200
70
180
87
% Recovery
130
N
40
65
43
33
70
45
87
% RSD
3
ND(3)
8
4
7
9
1
0
3
\" i
81
(1) Average of triplicate results.
2) Compound not detected.
Not determined.
TABLE 12. ANALYSIS RESULTS FOR SPIKED SAMPLE ILS-8 (Chlorinated
Ethanes Waste) FOR VOLATILE ORGANIC COMPOUNDS
uq/q
Compound
Propionitrile
2-Chl oroacryl onitri 1 e
1,1,1-Trichloroethane
1,2-Dichloropropane
Bromoform
1,1,2,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
2-Hexanone
CPD No.
12
24
25
33
44
47
50
51
46
Spike
16,000
4,000
24,000
4,000
4,000
16,000
20,000
4,000
24,000
Found(l)
13,700
3,900
24,700
3,000
2,700
14,000
17,700
3,500
17,700
% Recovery
85
98
102
75
68
88
88
88
74
% RSD
8
3
2
ND(2)
7
ND
3
ND
3
(1) Average of triplicate results.
(2) Not determined.
99
-------
standard deviation (RSD). The relative standard deviations appear consistent
among spiked compounds within a particular waste sample but vary across waste
types. Also, the RSDs are, in general, good when the % recovery is high.
High-level and low-level spike averages for all volatile compound spikes
are shown in Table 13. These averages suggest that there is no correlation
between spike level and percent recovery or RSD, but seem to suggest a
relationship between sample type and analytical performance. However, there
are many variables and more data will be needed before firm correlations can
be made. For example, the ability to spike a sample effectively with volatile
compounds 1s largely untested. Thus, variable recoveries among samples may
relate to spiking efficiency. The ability to distribute spike compounds may
be revealed by the average relative standard deviation data. For instance, as
indicated in Table 13, the waste catalysts are associated with the highest
RSD's.
The spike compounds with the poorest performances were propionitrile
2-chloroacrylonitrile and hexanone. The good performers in general were
1,1,1-trichloroethane, 1,2-dichloropropane, chlorobenzene and ethylbenzene.
These performances tend to confirm expectations that are based on experiences
with the application of Method 624 for quantifying acrylonitrile, halocarbons,
and aromatic hydrocarbons 1n water. Highly polar compounds such as
propionitrile, 2-chloroacrylon1tr1le, and 2-hexanone are readily solvated by
water and not expected to purge well. These polar compounds may also degrade
on the active surfaces of solid wastes.
TABLE 13. AVERAGE RECOVERY AND % RSD OF ALL SPIKED VOLATILE
COMPOUNDS (HIGH AND LOW) IN EACH WASTE SAMPLE
% Recovery. Average
High Spike Low Spike
% RSD
High Spike Low Spike
Spike Level Range
ug/g
ILS 1
ILS 2
ILS 3
ILS 4
ILS 5
ILS 6
ILS 7
ILS 8
35
103
21
106
30
29
68
90
49
134
34
110
20
44
38
82
8
9
38
13
22
18
2
3
3
5
21
13
31
40
5
3
5-30
200-1200
500-3000
100-600
5-30
5-30
100-600
4000-24000
100
-------
Determinations of Semi volatile Organic Compounds
The summary data for determinations of semivolatlle compounds spiked in
the waste samples are shown in Tables 14 through 20. These data include:
0 Amount "of a spiking compound naturally incorporated in a waste before
any compounds were spiked.
I The amount of the spike.
• The amount found after spiking.
0 The percent recovery.
0 The relative standard deviation results from triplicate analysis of
spiked samples.
The data show that the recoveries and RSDs appear to be matrix dependent
not concentration-dependent. The recoveries for the most part are about
50-70% with ranges from zero to about 200%. The relative standard deviations
typically range from 10% to 50%, with total range from 1% to greater than
100%. When one considers that the spiking compounds were selected to provide
a severe challenge to the analysis methodology, the data appear reasonable.
In fact, these data appear similar to data obtained using EPA Method 625 on
similar type samples (4). For instance, previous analysis of a sample identified
as Tar Column Bottoms and spiked with 10 compounds showed recoveries of from 15%
to 96% and RSDs of from 15% to about 50%. Furthermore, these 10 compounds
spiked in that study did not include some of the more challenging compounds
included here.
Certain classes of compounds were not recovered or were poorly recovered.
The compound classes that performed poorly are:
0 Nitrophenols
0 Carboxylic acids
0 High m.w. polynuclear aromatic hydrocarbons
0 Pyridines
0 Anilines
• Perch!orinated hydrocarbons
In general, the poor recovery performance was correlated to poor chromato-
graphic properties, i.e., high polarity.
The poor recoveries of perch!orinated hydrocarbons, hexachloroethane and
hexachloropentene, may indicate a stability problem. The former exhibited
satisfactory recovery in nearly every spiked sample, while the latter was
found in only one sample.
101
-------
TABLE 14. SUMMARY OF ANALYSIS DATA FOR SPIKED SAMPLE ILS-1 (CREOSOTE
CONTAMINATED SOIL) FOR SEMIVOLATILE ORGANIC COMPOUNDS
Compound
4-Chlorotoluene
B1s(2-chloroethyl) ether
2-Chlorophenol
2 ,4 ,6-Tri methyl pyrl d1 ne
1 ,4-D1 chl orobenzene
Acetophenone
Hexachloroethane
4-t-Butylpyr1d1ne
2,4-D1methyl phenol
Proplophenone
4-Chloroan1l1ne
Hexachloropropene
Quinoline
B1s(2-chloroethoxy) ethane
4-Chl oro-2-methyl anil 1 ne
4-Chlorobenzoic add
1-Chl oronaphthal ene
4-Methylqulnollne
2-Ethyl naphtha! ene
4-Bromobenzoic acid
1,3-01 nitrobenzene
2,6-D1n1trotoluene
3-N1troan1Hne
2,4-D1nitrophenol
4-N1trophenol
Pentachl orobenzene
2-Chl oro-4-n1troan1l 1 ne
Hexachl orobenzene
2,4-01 chl orophenoxyacetlc add
2,4,5-Tr1chlorophenoxyacet1c add
Anthraqulnone
Fluoranthene
2-Methyl anthraqul none
Pyrene
4,4'-DDD
4, 4 '-DDT
Trlphenyl phosphate
Trl-(p-tolyl) phosphate
Dlbenzocarbazole
01 benzo( a, h) anthracene
Present
(I)'
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
2
N
N
N
N
M
N
M
N
N
N
N
N
N
1,060
N
890
N
N
N
N
M
92
ng/g
Spike
40
40
40
40
200
40
40
200
200
240
40
160
40
160
200
40
160
160
40
160
40
160
40
240
40
200
160
40
40
200
40
40
200
240
40
240
N
200
160
40
Found
(2)
29
41
38*
N
135
18
27
148
220
120
N
N
21
180
26
N
250
100
36
158*
N
200
N
N
N
287
61*'
58*
N
N
65
1,270
166
1,260
88
375
N
144*
N
N
Recovery
%
79
103
94
N
68
45
68
74
110
50
N
N
53
113
13
N
156
• 62
90
99
N
125
N
N
N
143
* 38
145
N
N
160
115
83
112
220
130
N
72
N
N
%RSD
9
10
ND
ND
11
11
5
23
9
8
ND
ND
13
7
3
ND
9
12
5
ND
ND
60
ND
ND
ND
9
24
ND
ND
ND
21
6
18
6
12
17
ND
ND
ND
ND
(l)-Result of single analysis of unsplked sample; (2)
analyses of spiked sample; N-Compound not detected
**-Dupl1cate results; ND-Not determined.
-Result of triplicate
*-S1ngle result;
102
-------
TABLE 15. SUMMARY OF ANALYSIS DATA FOR SPIKED SAMPLE ILS-10 (EXTRACT OF
CREOSOTE CONTAMINATED SOIL) FOR SEMIVOLATILE COMPOUNDS
= ========= = ===:===== === = = ==== = === = ====: = = ===== = ===: ======== ===== = ==== ====== =
Amount
Compound Found(l) tig/g % Recovery %RSD
4-Chlorotoluene
B1s(2-chloroethyl ) ether
2-Chlorophenol
2,4,6-Tr1methylpyr1d1ne
1 , 4-DI chl orobenzene
Acetophenone
Hexachloroethane
4-t-Butylpyrldlne
2, 4-Dimethyl phenol
Proplophenone
4-Chloroan1l1ne
Hexachloropropene
Qu Incline
B1s(2-chloroethoxy) ethane
4-Chl oro-2-methyl anil 1 ne
4-Chlorobenzo1c add
1-Chloronaphthalene
4-Methylqu Incline
2-Ethyl naphthalene
4-Bromobenzo1c add
1,3-Di nitrobenzene
2,6-D1n1trotoluene
3-N1troan1l1ne
2,4-D1n1trophenol
4-N1trophenol
Pentachl orobenzene
2-Chl oro-4-n1 troanll 1 ne
Hexachl orobenzene
2,4-D1chlorophenoxyacet1c add
2,4,5-Tr1chlorophenoxyacet1c acid
Anthraqulnone
Fluoranthene
2-Methyl anthraqulnone
Pyrene
4, 4 '-ODD
4, 4 '-DDT
Trlphenyl phosphate
Trl-(p-tolyl) phosphate
Dlbenzocarbazole
22
44
33
50
92
36
27
200
166
183
N{1 )
N
42
160
76
N
298
119
61
132
27
154
N
223
N
190
142
41
N
N
72
58
194
338
59
220
48
214
360
D1benzo(a,h)anthracene 63
=====================================================
54
110
83
125
46
90
67
100
83
76
N
N
105
100
38
N
119
119
153
77
67
96
N
93
N
95
89
102
N
N
180
145
97
141
147
92
120
107
227
3
5
5
24
2
3
2
9
2
3
ND
ND
4
0
15
ND
ND
ND
6
12
8
4
ND
3
ND
ND
*
4
ND
ND
3
4
3
4
5
8
10
3
3
157 13
=======================
[1) Result of triplicate analyses of one extract. Amount present 1n
unsplked sample (see Table 14) has been subtracted from total amount
found; N-Compound not detected; ND-Not determined; *-S1ngle analytical
result.
103
-------
TABLE 16. SUMMARY OF ANALYSIS DATA FOR SPIKED SAMPLE ILS-2 (LATEX
PAINT) FOR SEMIVOLATILE ORGANIC COMPOUNDS
Present
Compound (1)
4-Chlorotoluene
Bis(2-chloroethyl) ether
2-Chlorophenol
2,4,6-Trimethylpyridine
l,4-D1chlorobenzene
Acetophenone
Hexachloroethane
4-t-Butylpyrid1ne
2 ,4-Dimethyl phenol
Proplophenone
4-Chloroaniline
Hexachloropropene
Quinoline
Bis(2-chloroethoxy) ethane
4-Chl oro-2-methyl am' 1 1 ne
4-Chlorobenzoic acid
1-Chl oronaphthal ene
4-Methylquinol1ne
2-Ethyl naphtha! ene
4-Bromobenzoic add
l,3-D1n1trobenzene
2,6-Din1trotoluene
3-N1troaniline
2,4-Dinitrophenol
4-N1trophenol
Pentachl orobenzene
2-Chl oro-4-n1troan1l 1ne
Hexachl orobenzene
2,4-Dichlorophenoxyacetic acid
2,4,5-Trichlorophenoxyacetic acid
Anthraquinone
Fluoranthene
2-Methylanthraquinone
Pyrene
4, 4 '-ODD
4, 4 '-DDT
Triphenyl phosphate
Tri-(p-tolyl) phosphate
Dibenzocarbazole
Dibenzo(a,h)anthracene
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
uq/q
Spike
600
500
600
600
100
100
500
100
100
400
600
100
500
100
100
500
100
100
500
100
500
100
500
400
100
100
100
600
600
100
600
100
100
400
100
400
600
100
100
500
Found Recovery
(2) %
520
460
510
1,370
60
103
400
113*
106
240
250
N
394
96
27
650
143
111
588
N
310
88
220
N
N
100
N
790
N
N
490
83
80
198
296
400
400
93
N
N
87
92
85
230
60
103
87
113
106
60
41
N
79
96
27
130
140
. Ill
118
N
62
88
44
N
N
100
N
130
N
N
81
83
80
49
296
100
66
93
N
N
%RSD
35
34
30
118
45
1
40
ND
52
2
37
ND
9
52
18
97
31
33
12
ND
78
66
ND
ND
ND
69
ND
60
NO
ND
5
7
ND
105
47
113
62
ND
ND
ND
(I)-Results of single analysis of unspiked sample;(2)-Average result of
duplicate analyses; *-Single result only; N-Compound not detected;
ND-Not determined.
104
-------
TABLE 17. SUMMARY OF ANALYSIS DATA FOR SPIKED SAMPLE ILS-3 (ETHANES
SPENT CATALYST) FOR SEMIVOLATILE ORGANIC COMPOUNDS
Compound
4-Chlorotoluene
Bis(2-chloroethy1) ether
2-Chlorophenol
2,4,6-Trimethylpyridine
1,4-Dichlorobenzene
Acetophenone
Hexachloroethane
4-t-Butylpyrid1ne
2,4-Dimethyl phenol
Propiophenone
4-Chloroaniline
Hexachloropropene
Qu1nol1ne
Bis(2-chloroethoxy) ethane
4-Chl oro-2-methyl am' 1 i ne
4-Chlorobenzo1c add
1-Chl oronaphthal ene
4-Methylquinoline
2-Ethyl naphthalene
4-Bromobenzo1c add
1,3-Dinitrobenzene
2,6-D1n1trotoluene
3-N1troan1l1ne
2,4-Dinitrophenol
4-Nitrophenol
Pentachlorobenzene
2-Chl oro-4-ni troani 1 1 ne
Hexachl orobenzene
2,4-Dichlorophenoxyacetic acid
2,4,5-Trichlorophenoxyacetic add
Anthraquinone
Fluoranthene
2-Methyl anthraqui none
Pyrene
4,4'-DDD
4, 4 '-DDT
Triphenyl phosphate
Trl-(p-tolyl) phosphate
Dlbenzocarbazole
D1benzo(a,h)anthracene
vg/g
Present
(1) Spike
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
200
N
N
N
N
N
N
N
N
N
N
N
N
N
N
800
200
800
800
200
1,000
200
200
200
200
800
1,200
200
1,200
200
200
1,200
1,200
200
1,200
200
1,200
200
200
1,000 '
192
1,200
- 800
800
200
800
1,000
200
200
1,000
200
800
200
1,200
200
Found Recovery
(2) %
808
196
526
N
138
755
138
90
N
108
N
77**
128
1,330
N
N
N
1,740
173
560
N
1,160
N
N
N
96
135**
896
N
N
750
940
184*
38*
1,270
N
738
143*
N
N
101
98
66
N
67
76
69
45
N
54
N
6
64
111
N
N
N
145
86
47
N
97
N
N
N
44
11
112
N
N
94
94
92
19
127
N
92
71
N
N
%RSD
21
14
16
ND
23
16
34
9
ND
19
ND
4
9
26
ND
ND
ND
43
60
64
ND
21
ND
ND
ND
1
4
ND
ND
22
23
ND
ND
31
ND
14
ND
ND
ND
(l)-Result of single analysis of unspilced sample;
triplicate analyses; M-Compound not detected;
**-Duplicate results; ND-Not determined.
(2)-Average result of
*-Single result;
105
-------
TABLE 18. SUMMARY OF ANALYSIS DATA FOR SPIKED SAMPLE ILS-4 (COAL TAR)
FOR SEMIVOLATILE ORGANIC COMPOUNDS
Compound
4-Chlorotoluene
Bis(2-ch1oroethyl) ether
2-Chlorophenol
2 ,4 ,6-Trimethyl pyridi ne
1 ,4-01 chl orobenzene
Acetophenone
Hexachloroethane
4-t-Butylpyr1d1ne
2, 4-Dimethyl phenol
Propiophenone
4-Chloroan1l1ne
Hexachloropropene
Qulnoline
Bis(2-chloroethoxy) ethane
4-Chloro-2-methyl aniline
4-Chlorobenzo1c add
1-Chl oronaphthal ene
4-Methylqu1noline
2-Ethyl naphtha! ene
4-Bromobenzoic add
l,3-D1n1trobenzene
2,6-Din1trotoluene
3-N1troan1line
2,4-Din1trophenol
4-N1trophenol
Pentachl orobenzene
2-Chl oro-4-n1troan1 1 1ne
Hexachl orobenzene
2,4-01 chl orophenoxyacetlc acid
2,4,5-Trichlorophenoxyacetic add
Anthraqulnone
Fluoranthene
2-Methyl anthraqul none
Pyrene
4,4'-DDD
4, 4 '-DDT
Trfphenyl phosphate
Tri-(p-tolyl) phosphate
Dlbenzocarbazole
Dibenzo(a ,h)anthracene
w
Present
(1)
N
N
N
N
N
N
N
N
900
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
4,600
N
19,000
N
N
N
N
N
5,000
uq/q
Spike
1,900
11,400
1,900
1,900
7,600
1,900
11,400
7,600
7,600
9,500
1,900
1,900
11,400
1,900
7,600
11,400
1,900
1,900
11,400
1,900
11,400
1,900
11,400
9,500
1,900
7,600
1,900
1,900
1,900
7,600
1,900
1,900
7,600
9,500
1,900
9,500
1,900
7,600
1,900
11,400
Found Recovery
(2) %
1,810
12,410
2,800*
N
6,900
1,300
10,800
7,500
9,170
6,400
1,060*
N
2,100
1,500
N
11,930**
2,700
2,900
1,600
N
8,600
2,100**
N
8,200**
N
8,300
N
900*
N
N
1,200**
15,600
6,600
19,800
2,500
8,600
1,200**
6,360**
N
N
95
109
147
N
90
68
95
99
108
67
56
N
18
81
N
105
140
•150
14
N
75
108
N
86
N
109
N
48~
N
N
65
240
86
70
130
91
63
84
N
N
%RSD
49
34
*
NO
35
14
38
31
44
9
*
ND
11
62
ND
128
27
29
10
ND
58
56
ND
48
ND
50
ND
*
ND
ND
34
7
8
7
60
79
45
12
ND
ND
(l)-Result of single analysis of unsplked sample;
triplicate analyses; N-Compound not detected;
**-Duplicate results; ND-Not determined.
(2)-Average result of
*-Single result;
106
-------
TABLE 19. SUMMARY OF ANALYSIS DATA FOR SPIKED SAMPLE ILS-5 (OXY-
CHLORINATED SPENT CATALYST) FOR SEMIVOLATILE ORGANIC COMPOUNDS
Present
Compound (1)
4-Chlorotoluene
Bis(2-chloroethyl) ether
2-Chlorophenol
2,4,6-Trimethylpyridine
1,4-Dichlorobenzene
Acetophenone
Hexachloroethane
4-t-Butylpyridine
2, 4-Di methyl phenol
Propiophenone
4-Chloroam'line
Hexachloropropene
Quinoline
Bis(2-chloroethoxy) ethane
4-Chl oro-2-methyl ani 1 i ne
4-Chlorobenzoic acid
1-Chloronaphthalene
4-Methylquinoline
2-Ethyl naphthalene
4-Bromobenzoic acid
1,3-Dinitrobenzene
2,6-Dinitrotoluene
3-Nitroan1line
2,4-Dinitrophenol
4-Nitrophenol
Pentachlorobenzene
2-Chl oro-4-nitroani 1 ine
Hexachlorobenzene
2,4-Dichlorophenoxyacetic acid
2,4,5-Trichlorophenoxyacetic acid
Anthraquinone
Fluoranthene
2-Methyl anthraquinone
Pyrene
4,4' -ODD
4,4' -DDT
Triphenyl phosphate
Trl-(p-tolyl) phosphate
Dibenzocarbazole
D1benzo( a, h) anthracene
N
N
N
N
0.7
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
yg/s
Spike
4
24
4
4
16
4
24
16
16
20
4
4
24
4
16
24
4
4
24
4
24
4
24
20
4
16
4
4
4
16
4
4
16
20
4
20
4
16
4
24
Found Recovery
(2) % %RSD
0.17*
8.6**
N
4.3*
1.3**
0.9**
1.6**
2.8*
N
5**
N
N
2.7**
2.7**
N
8*
2.4**
0.6*
11**
N
18**
3.4**
N
16**
N
13**
0.7*
3
N
N
1
1
5**
6**
5
28
2
8
N
N
4
36
N
108
8
23
7
18
N
26
N
N
9
67
N
34
60
15
46
N
76
86
N
81
N
80
17
77
N
N
26
29
30
28
116
139
46
49
N
N
ND
2
ND
ND
11
12
12
ND
ND
9
ND
ND
120
6
ND
ND
.7
ND
6
ND
0.8
9
ND
15
ND
6
ND
1.5
ND
ND
9
13
4
2
30
20
24
5
ND
ND
(D-Result of single analysis of unspiked sample; (2)-Average result of
triplicate analyses; N-Compound not detected; *-Single result;
**-Duplicate results; ND-Not determined.
107
-------
TABLE 20. SUMMARY OF ANALYSIS DATA FOR SPIKED SAMPLE ILS-6 (CINCINNATI
DEWATERED SLUDGE) FOR SEMIVOLATILE ORGANIC COMPOUNDS
Present
Compound (1)
4-Chlorotoluene
Bis(Z-chloroethyl) ether
2-Chlorophenol
2,4,6-Trimethylpyridine
1,4-Dichlorobenzene
Acetophenone
Hexachloroethane
4-t-Butylpyridine
2, 4-D1methyl phenol
Proplophenone
4-Chloroan1Hne
Hexachloropropene
Qu Incline
Bis(2-chloroethoxy) ethane
4-Chl oro-2-methyl ani 1 1 ne
4-Chlorobenzoic add
1-Chl oronaphthal ene
4-Methylqu1nol1ne
2-Ethyl naphthalene
4-Bromobenzo1c add
1,3-Di nitrobenzene
2,6-Dinitrotoluene
3-NitroaniHne
2,4-Dinitrophenol
4-N1trophenol
Pentachlorobenzene
2-Chl oro-4-ni t roani 1 1 ne
Hexachlorobenzene
2,4-Dichlorophenoxyacetic add
2,4,5-Trichlorophenoxyacetic add
Anthraquinone
Fluoranthene
2-Methyl anthraqui none
Pyrene
4, 4 '-ODD
4, 4 '-DDT
Triphenyl phosphate
Tri-(p-tolyl ) phosphate
Dlbenzocarbazole
D1benzo(a,h)anthracene
N
N
N
N
N
N
• N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
4
N
4
N
N
N
N
N
N
uq/q
Spike
20
20
20
20
100
20
20
100
100
120
20
80
20
80
100
20
80
80
20
80
20
80
20
120
20
100
80
20
20
100
20
20
100
120
20
120
20
100
80
20
Found Recovery
(2) %
10
16
14**
N
46
8
N
84
88
45
N
N
10
60
24
N
92
80
21
N
N
N
N
N
N
96
N
38*
N
N
9
19
70
103
58
N
14*
90*
48*
N
47
78
73
N
46
42
N
84
88
38
N
N
50
74
24
N
116
• 100
108
N
N
N
N
N
N
96
N
193
N
N
48
78
70
83
290
N
70
90
60
N
%RSD
50
28
45
N
39
7
N
42
52
6
N
N
8
42
27
N
33
37
13
N
N
N
N
N
N
62
N
N
N
N
23
21
10
10
63
N
N
N
N
N
(l)-Result of single analysis of unspiked sample; (2)-Average result of
triplicate analyses; N-Compound not detected; *-Single result;
**-Dupl1cate results; ND-Not determined.
108
-------
The above data were obtained early in the program and later were checked
by analysis in our laboratory of Performance Evaluation Sample ILS-10, which
is an extract of sample ILS-1. The analysis results of ILS-10 are shown in
Table 15 immediately following the ILS-1 data. The results of analysis of
sample ILS-10 support the conclusion that nitro phenols, carboxylic acids,
hexachloropropene and anilines perform poorly.
Several compounds specifically 2,4,6-trimethylpyridine, 1,3-dinitro-
benzene, triphenyl phosphate, dibenzocarbozole, and dibenzo(a,h)anthracene,
that were not found initially in ILS-1 were found in ILS-10. The improved
detection may have resulted from better GC/MS performance as well as from a
better extraction technique (use of ultrasonic energy).
Not only are the recoveries better for many of the compounds in ILS-10
than for ILS-1 but also the relative standard deviations appear to be better.
It may be premature to make that judgment, in as much as the analysis of ILS-1
was done on extractions of three separate sample aliquots, whereas the
analysis of ILS-10 represented replicate analyses of an extract of one sample
aliquot. Thus the RSD for the ILS-10 data includes only analysis variation
and omits extraction variations.
Pilot Study
The spiking scheme used for this program is unique. The scheme involves
spiking each waste with a series of compound pairs. Compounds in each pair
were chosen to be chemically similar and it was assumed that analytical
performance is similar. One compound of each pair was spiked at a high
concentration and the other at a low concentration. The benefit of such an
arrangement is that one sample so spiked will, in a single analysis, provide
precision and recovery (accuracy) information at two different concentrations.
This would lead to a considerable saving in analyses costs.
Because the scheme is a departure from standard ways of obtaining
information on precision and recovery as a function of concentration, it
seemed prudent to investigate the effectiveness of such a design in a small
single laboratory study before using it in the inter!aboratory study. The
objective of this pilot study was to determine whether spiked compound pairs
will have similar analytical recoveries at both high and low concentrations.
The pilot study was conducted on 2 waste samples, latex paint waste and
POTW sludge. The design of the study included triplicate analyses of unspiked
wastes, triplicate analyses of the wastes with one of each compound pair (see
Table 2) spiked at a high concentration and the other spiked at one fifth that
level, and triplicate analyses of the wastes with the compound pair concentra-
tion reversed. The latex paint waste was spiked at 1250 yg/g and 250 yg/g and
the POTW sludge was spiked at 250 yg/g and 50 yg/g.
The results obtained from this study are shown in Tables 21 and 22. In
most cases, with the exception of the benzole acids, phenoxyacetic acids,
109
-------
TABLE 21. RECOVERY OF SPIKED COMPOUNDS FROM POTW SLUDGE
Compound
Hexachloroethane
Hexachloropropane
4-Chlorotoluene
1 , 4-Dichl orobenzene
Pentachlorobenzene
Hexachl orobenzene
4, 4 '-ODD
4, 4 '-DDT
2-Ethyl naphthalene
1-Chl oronaphthal ene
Fluoranthene
Pyrene
l,3-D1nitrobenzene
2,6-Dinitrotoluene
2-Chlorophenol
2, 4-Dimethyl phenol
4-N1trophenol
2,4-Dinitrophenol
4-Chloroan1line
4-Chl oro-2-methyl an1 1 i ne
3-Nitroanil1ne
2-Chl oro-4-ni t roam 1 i ne
2,4,6-Tr1methylpyrid1ne
4-t-Butylpyridine
Quinoline
4-Methylqu1noline
B1s(2-chloroethyl)ether
Bis(2-chloroethoxy)ethane
THphenyl phosphate
Tn-p-tolyl phosphate
Anthraquinone
2-Methyl anthraqul none
Acetophenone
Propiophenone
4-Chlorobenzolc acid
4-Bromobenzoic add
2,4-0
2,4,5-T
% Recovery at
MeanU)
70
50
60
57
74
64
79
57
69
63
72
70
49
75
63
77
42
41
55
66
50
41
108
62
56
62
78
53
77
78
78
86
61
59
59/ x
ND(b)
ND
NO
50 uq/q
SO
11
6
4
5
7
2
12
8
11
5
6
3
8
16
9
4
5
3
8
3
1
12
7
1
5
3
3
5
5
3
12
6
5
4
9
ND
ND
ND
% Recovery at
Mean(a)
54
50
54
51
60
60
66
56
63
57
63
62
56
59
55
64
60
48
46
55
61
65
116
56
53
58
63
55
65
70
64
66
56
56
55
56
ND
ND
250 uq/q
SD
8
2
6
5
5
6
6
4
3
4
2
6
2
5
3
5
4
4
4
5
4
5
4
5
3
3
7
3
2
7
4
7
7
3
5
7
ND
ND
(^Average of 3 replicates.
= compound not detected.
110
-------
TABLE 22. RECOVERY OF SPIKED COMPOUNDS FROM LATEX PAINT WASTE
. %
Compound
Hexachloroethane
Hexachloropropene
4-Chlorotoluene
1,4-Dichlorobenzene
Pentachl orobenzene
Hexachlorobenzene
4, 4 '-ODD
4, 4 '-DDT
2-Ethyl naphthalene
1-Chloronaphthalene
Fluoranthene
Pyrene
l,3-D1nitrobenzene
2,6-Dinitrotoluene
2-Chlorophenol
2, 4-Dimethyl phenol
4-NHrophenol
2,4-Dinitrophenol
4-Chloroaniline
4-Chloro-2-methyl aniline
3-Nitroaniline
2-Chloro-4-nitroani 1 1 ne
2,4,6-Trimethylpyr1d1ne
4-t-Butylpyridine
Quinoline
4-Methylquinoline
Bis(2-chloroethyl )ether
Bis(2-chloroethoxy)ethane
Triphenyl phosphate
Tri-p-tolyl phosphate
Anthraquinone
2-Methylanthraquinone
Acetophenone
Propiophenone
4-Chlorobenzoic acid
4-Bromobenzo1c add
2,4-D
2,4,5-T
Recovery at
MeanU)
71
51
72
74
69
64
73
54
85
94
84
89
16
82
72
76
' 44,^
ND(b)
75
84
15
NO
57
78
89
82
83
64
54
49
64
47
96
90
ND
NO
ND
ND
250 ug/q
SO
5
4
2
7
1
4
3
5
1
8
1
5
11
6
1
9
19
ND
10
4
3
ND
4
5
5
3
2
8
8
10
2
3
2
6
ND
ND
ND
ND
% Recovery at
Mean(a)
84
95
81
83
91
94
80
96
89
94
89
92
84
102
86
97
80
113
94
99
• 69
107
64
88
91
104
86
101
89
97
90
93
94
94
85
ND
ND
ND
1250 uq/q
SD
5
2
8
3
3
3
6
1
2
3
5
1
9
8
5
1
12
27
6
4
7
4
5
1
8
4
4
2
3
5
7
2
4
2
42
ND
ND
ND
(^Average of 3 replicates.
(b)ND = compound not detected.
Ill
-------
pyridines, and some of the nitro compounds, the two compounds in each set give
similar recoveries at both levels within the precision of the method. Under-
ivatized benzoic and phenoxyacetic acids were expected to have poor chromato-
graphic properties, even on good fused silica capillary columns. These com-
pounds were included to challenge the method and thereby to establish method
limitations. The pyridines and nitro compounds and especially the nitrophenols
have poor chromatographic properties and for these compounds it will be dif-
ficult to separate analytical and extraction variables.
Response Factor Study
The quality assurance protocol for this collaborative study included a QC
Protocol for Fused Silica Capillary Columns that was developed by Acurex under
the direction of Mr. Drew Sauter, EPA-LV and modified by Battelle for this
study. The protocol included specific instructions for calculating response
factors and criteria for acceptance.
In order that response factor criteria given in the QA protocol for
collaborators might be realistic in absolute value and acceptable variation,
the response factors for the semivolatile compounds in the calibration
standard were determined in three different laboratories: Battelle, EMSL-LV,
and Raltech. A three laboratory study was considered minimum to provide
reasonable certainty on an average response factor and a standard deviation
from which QA limits could be ,set.
Because the limits established will need to be met by participants in the
interlaboratory study, compounds that may present chromatography problems or
compounds having quantification ions with masses that differ too much from the.
mass of the internal standard quantification ion were not included in an
initial review of the RF study data.
By using an initial mass difference criterion of ±20 atomic mass units 43
of the 140 compounds in the calibration standards were acceptable as compounds
for which to set RF limits. These 43 compounds were evaluated on the basis of
the RF RSD from the three laboratory study. As a result of this study 11
compounds were eliminated from further consideration as compounds for RF limit
criteria.
Further examination showed reasonably good agreement for response factors
for several compounds that have quantification ions with masses that differ by
more than 20 a.m.u. from mass of the internal standard quantification ion. We
therefore, reconsidered the ±20 a.m.u. range that was previously selected as
an acceptable range and selected 18 additional compounds to be subject to
specified RF limits. The additional compounds include primarily anilines and
phenols. The total list of 50 compounds with response factor data, recommend-
ed specified response factors, and a.m.u. difference between the quantifica-
tion ion and the internal standard quantification ion is given in Table 23.
We recommended that the participants in the interlaboratory study be
required to meet the specified response factors for all 50 compounds in all
calibration runs at the 50 ug/ml level within a precision of ±40%. For many
112
-------
TABLE 23. RECOMMENDED COMPOUND RESPONSE FACTORS TO BE MONITORED
DURING THE INTERLABORATORY COMPARISON STUDY
No.
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
31
32
33
Compound
Bromobenzene
Bis(2-chloroethyl) ether
Benzyl chloride
1,2-Dichlorobenzene
N-Methyl aniline
4-Methyl aniline
Hexachl oroethane
N,N, -Dimethyl aniline
1,2, 4, 5-Tetramethyl benzene
2-Nitrophenol
2, 4-Dimethyl phenol
2, 6-Di methyl aniline
2,6-Dichlorophenol
1,2,4-Trichlorobenzene
Naphthalene
4-Chloroaniline
2,4-Dichlorophenol
Hexachl orobutadi ene
Quinoline
3,4-Dichloroaniline
2-Chloro-4-nitrophenol
2, 3-Dimethyl naphthalene
Dimethyl phthalate
Acenaphthene
2, 4-Di-t-butyl phenol
Dibenzofuran
2,4-Dinitrotoluene
2,4,5-Trichloroaniline
Fluorene
Diphenylamine
Azobenzene
Hexachl orobenzene
Recom-
mended
RF
1
1
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
•
•
•
•
•
•
•
•
•
•
•
*
•
•
•
•
•
•
•
0.
0
1
1
1
1
1
0
0
1
0
0
0
•
•
•
t
•
•
•
•
•
•
•
•
Pentachlorophenol 0.
00
10
90
30
65
65
16
60
70
18
34
40
30
32
00
40
30
14
70
50
15
00
30
00
20
60
35
42
00
68
24
23
13
RF from
Given Source(a)
A
-
-
-
-
_
-
-
-
0.22
0.32
-
>
0.32
1.08
_
0.30
0.13
-
-
_
-
-
-
-
• _
-
-
-
0.58
. -i.''
0.24
0.13
B
1.04
1.13
1.89
0.31
0.57
0.51
0.15
0.58
0.72
0.19
0.34
0.42
0.29
0.32
0.99
0.45
0,30
0.16
0.73
0.54
0.15
0.95
1.29
0.95
1.36
1.64
0.32
0.42
1.09
0.68
0.24
0.23
0.11
0.
1.
1.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
1.
0.
0.
0.
0.
0.
0.
1.
0.
0.
1.
1.
0.
0.
0.
0.
0.
0.
0.
c
94-1.
08-1.
59-2.
29-0.
49-0.
64-0.
12-0.
50-0.
69-0.
15-0.
26-0.
33-0.
21-0.
26-0.
05-1.
29-0.
22-0.
10-0.
15-0.
37-0.
10-0.
01-1.
87-1.
08-1.
05-1.
47-1.
24-0.
38-0.
99-1.
53-0.
18-0.
20-0.
09-0.
%RSD
of RF a.m.u.
From Differ-
Source ence
18
36
03
39
70
88
21
65
85
28
45
49
31
40
42
51
33
17
95
65
20
11
52
12
38
89
42
52
28
78
40
36
16
B
12
17
16
12
16
37
16
18
13
30
13
10
17
14
7
11
19
19
11
17
36
11
9
11
30
10
32
13
9
16
26
11
33
(b)
5
11
9
10
20
29
19
16
17
3
14
15
26
44
8
9
26
89
7
3
9
8
1
10
27
4
1
31
2
19
6
96
78
(continued)
113
-------
TABLE 23. (Continued)
No.
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
Compound
Phenanthrene
Anthracene
Acridine
Phenanthridlne
Anthraqulnone
Fluoranthene
2-Methylanthraqufnone
Pyrene
Triphenyl phosphate
Benzo(a)anthracene
Chrysene
Benzo(k)fluoranthene
Benzo(a)pyrene
Dfbenzocarbazole
Indeno(l,2,3-CD) pyrene
Dibenzo( a, h) anthracene
Benzo{ g,h , 1 )peryl ene
=========
Recom-
mended
RF
1.00
1.00
0.60
0.70
0.30
1.00
0.25
1.00
0.20
1.00
1.00
1.00
1.00
0.50
0.80
0.60
0.70
%RSD
of RF a.m.u.
RF from From Differ-
G1ven Source(a) Source ence
A
1.16
1.15
m
_
-
-
-
_
-
1.11
1.02
1.10
1.00
-
0.45
0.58
0.64
B
0.99
1.01
0.61
0.67
0.29
0.98
0.25
0.97
0.20
1.08
1.03
1.05
0.99
0.54
0.80
0.60
0.72
C
0.96-1.21
0.99-1.17
0.59-0.91
0.65-0.92
0.26-0.41
0.16-1.16
0.22-0.33
0.93-1.23
0.17-1.83
0.87-1.26
0.80-1.29
1.07-1.20
0.91-1.07
0.34-0.52
0.64-0.88
0.48-0.70
0.58-0.76
B
13
10
7
6
8
12
4
11
10
17
8
12
, 15
30
18
19
21
(b)
10
10
9
9
4
10
10
10
86
16
12
12
12
3
12
14
12
= ==3====== = 3==:= ssassssssr 3=33==== =====:: = ==3==3=3=================== ====== =
(a) A = FSCC QC Protocol prepared by Acurex for Drew Sauter of EMSL-LV.
B = Response factor study Involving EMSL-LV, Raltech, and Battelle.
C = Range of values from calibration data reported by six laboratories
participating in the interlaboratory study. (Data from two
laboratories with sensitivity and tuning problems are not
included.)
(b) Difference between quantification ion mass for listed compound and
Internal standard.
of the compounds listed, the percent RSD for response factors from the 3-laboratory
study was 15-20% or higher.
The recommended ±40% range is based on using 2x RSD for compounds having
RSDs of up to 20%. This range gives at least a 95% confidence level. The
establishment of tighter RF limits at this time would be unrealistic. For
114
-------
example, the percent RSD obtained in the 3-laboratory study for benzo(a)anthracene
was ±17%. This result would indicate that if a ±20% range was used there
would be about a 30% chance of being outside the range.
The list of 50 compounds recommended for response factors monitoring was
reviewed by EPA and was refined to 20 compounds as shown in Table 24. It is
believed that the 20 compounds listed represent a range of compound polarities
and response factors that will be adequate to monitor the analytical system.
In addition, the use of 20 compounds rather than 50 compounds will aid cost
effectiveness.
Calibration Standards
Eleven separate standard solutions labeled A through J were prepared for
the Phase II and Phase III studies. The 140 semivolatile compounds to be used
for calibration were prepared in four separate solutions (A through D) to
minimize chemical reactions between components. The least soluble components
were put in one solution, Solution A, at a concentration of 0.1 mg/ml for each
component. The other three semivolatile calibration solutions were prepared
at a five-fold or ten-fold higher concentration. The solutions were to be
mixed and diluted just prior to use to give calibration solutions that contain
2, 10, 50, or 250 ng/uL of all 140 components. The components in Solution A,
however, were not included in the high level mixture because of solubility
limitations.
The semivolatile surrogate and internal standard compounds were prepared
as solutions E and F, respectively. Volatile compounds used for standards were
In solutions G, J, and K, and Supelco's Purgeable A, B, and C and were at
concentrations of 0.2 mg/ml. These solutions were to be mixed and diluted to
give calibration solutions suitable for adding 25, 100, 250 and 1000 ng of each
compound to the purging apparatus.
A listing of the standard solutions is shown in Table 25. Seven of the
volatile compounds listed in Solution G were not among the compounds to be
quantified because they were found not to be purgeable. All of the semi-
volatile compounds except hexachlorophene, 2,4-D and 2,4,5-T were detected
in the intralaboratory study.
Laboratory Selection
Laboratories were selected for participation in this program on the basis
that each must have the GC/MS analytical instrumentation specified in the
methods and must be staffed by persons experienced in GC/MS analysis of complex
environmental samples. Ten laboratories were selected to provide bid requests.
They were:
• Acurex
• Arthur D. Little, Inc.
t California Analytical Laboratories
• GCA Technology
115
-------
TABLE 24. EPA RECOMMENDED COMPOUND RESPONSE FACTORS TO BE MONITORED
DURING THE INTERLABORATORY COMPARISON STUDY
No.
Compound
Recommended
RF
% RSD of RF
From 3 Lab Study
1 Hexachlorobutadiene
2 Hexachloroethane
3 Hexachlorobenzene
4 2,6-Dichlorophenol
5 1,2,4-Trichlorobenzene
6 2,4-D1chlorophenol
7 2,4-Dimethylphenol
8 2,6-Dimethylaniline
9 Dibenzo(a,h)anthracene
10 N-Methylaniline
11 Diphenylamine
12 Benzo(g,h,i)perylene
13 Qulnoline
14 Benzo(a)anthracene
15 Chrysene
16 Phenanthrene
17 Bis(2-chloroethyl)ether
18 Dimethyl phthalate
19 Dibenzofuran
20 Benzyl chloride
0.14
0.16
0.23
0.30
0.32
0.34
0.34
0.40
0.60
0.65
0.68
0.70
0.70
1.00
1.00
1.00
1.10
1.30
1.60
1.90
19
16
11
17
14
19
13
10
19
16
16
21
11
17
8
13
17
9
10
16
Mead CompuChem
Midwest Research Institute
Southwest Research Institute
Southern Research Institute
Systems, Science, and Software
West Coast Technical Services
Bid requests were sent to these laboratories and on the basis of the
responses, four laboratories were requested to submit revised bids based on a
work statement and analysis procedures both of which were modified as the
result of questions from the ten initial respondents. The four laboratories
GCA, Southwest Research Institute, Acurex, and Midwest Research Institute were
subsequently accepted as subcontractors to Battelle to conduct the
inter!aboratory study work.
Five laboratories namely: Arthur D. Little, Mead CompuChem, Southwest
Research Institute, California Analytical Laboratory and West Coast Technical
Services were selected to participate and were funded through agreements
116
-------
TABLE 25. LIST OF STANDARD SOLUTIONS
Semi volatile Calibration Solution A
(all compounds at 0.1 mq/ml in methylene chloride)
Aromatic Halocarbons
3,3'-D1chlorobiphenyl
4,4'-Dichlorobiphenyl
2,2'-4,4'-Tet rachlorobi pheny1
2,2'-4,4',6J6'-Hexachlorobiphenyl
Aromatic Hydrocarbons
Benzo(a)anthracene
Benzo(k)fluoranthene
Benzo(g,h,i)perylene
Aromatic Hydrocarbons (con't)
Indeno(l,2,3-cd)pyrene
Chrysene
Benzo(a)pyrene
D1benzo(a,h)anthracene
Ami nes
1,2,7,8-Dibenzocarbazole
Semivolatile Calibration Solution B
(all compounds at 1.0 mg/ml in benzene)
Aromatic Hydrocarbons
Naphthalene
1,2,4-Trimethyl benzene
1,2,4,5-Tetramethylbenzene
Biphenyl
Acenaphthylene
Acenaphthene
2-Methy 1naphthalene
2-Ethylnaphthalene
2,3-Dimethylnaphthalene
1,2,3,4-Tetrahydronaphthalene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Ami nes
Aniline
4-Chloroan1l1ne
4-Bromoan1line
2-H1troaniline
3,4-Dichloroaniline
2,4,5-Trich!oroani11 ne
Amines (con't)
3-Nitroaniline
4-Chloro-2-methyl an111ne
4-Nitroaniline
2,6-D1chloro-4-nitroaniline
2-Chloro-4-nitroanl11ne
2,4-D1n1troanil1ne
N-Methylaniline
4-Chloro-2-n1troani11ne
4-Methylaniline
2,6-Dimethylaniline
4-Aminobipheny1
1-Aminonaphthalene
N,N-Dimethylaniline
Phenanthridine
4-Methylpyridine
2,4-Dimethylpyridine
4-t-Butyl pyri dine
2,4,6-Trimethylpridlne
Qu incline
4-Methylquinoline
Ac ri dine
Carbazole
3,3'-D1chlorobenz1dine
Diphenylamine
(continued)
117
-------
TABLE 25. (Continued)
Semi volatile Calibration Solution C
(all compounds at 0.5 mg/ml in methylene chloride)
Aliphatic Halocarbons
1,4-Dichlorobutane
Pentachloroethane
Hexachloroethane
Hexachloropropene
Hexachlorobutadiene
Aromatic Halocarbons
4-Chlorotoluene
Bromobenzene
1,2-Dichlorobenzene
1,4-Dichlorobenzene
1,2,4-Trichlorobenzene
1,2,4,5-Tetrachl orobenzene
Pentachlorobenzene
Hexachlorobenzene
Benzal chloride
Benzyl chloride
1-Chloronaphthalene
2-Chl oronaphthalene
a,a,a-Trichlorotoluene
Aromatic Nitro Compounds
Nitrobenzene
1,3-Di ni t robenzene
2-Nitrotoluene
4-Nitrotoluene
2,4-Dinitrotoluene
2,6-Dinitrotoluene
1,-Chloro-4-nitrobenzene
2,4-Dinitrochlorobenzene
Phenols
2-Chlorophenol
2-Nitrophenol
Phenol
2,4-Dimethylphenol
2,4-Dichlorophenol
2,4,6-Trichlorophenol
Phenols (con't)
4-Chloro-3-methylphenol
2-Methylphenol
4-Methylphenol
Thiophenol
4-Chlorophenol
2,6-Dichlorophenol
2,4,5-Trichlorophenol
Hexachlorophene
4-Hydroxybiphenyl
2-Naphthol
4-t-Butyl phenol
2-Chloro-4-nitrophenol
2,4-Dinitrophenol
2-Methyl-4,6-dinitrophenol
Pentachlorophenol
4-Nitrophenol
2,6-Di-t-butyl-4-methylphenol
2,4-Di-t-butyl phenol
Di ethyl stil best rol
Chlorinated Pesticides
4,4'-DDD
4,4'-DDE
4,4'-DDT
Methoxychlor
Trifluralin
Pentachloronitrobenzene
Acids
acid
4-Chlorobenzoic
Benzoic acid
4-Bromobenzoic acid
2,4-Dichlorophenoxyacetic acid
2,4,5-Trichlorophenoxyacetic acid
Haloethers
Bi s(2-chloroethyl )ether
Bis(2-chloroethoxy)ethane
4-Chlorophenyl phenyl ether
(continued)
118
-------
TABLE 25. (Continued)
Semi volatile Calibration Solution D
(all compounds at 1.0 mg/ml in methylene chloride)
Phthalates Ethers and Sulfides
Dimethyl phthalate Am'sole
Di-n-butyl phthalate Phenyl ether
Di(2-ethylhexyl) phthalate Dibenzofuran
Phosphates Ketones
Tri(p-tolyl) phosphate Anthraquinone
Triphenyl phosphate 2-Methylanthraquinone
Propiophenone
Acetophenone
Aldehydes
Miscellaneous
Be nz aldehyde
4-Chlorobenzaldehyde Azobenzene
Acetanilide
Benzyl alcohol
Di(2-ethylhexyl) sebacate
Semi volatile Surrogate Standard Solution E
(all comppunds at 10 mq/ml in methyl ene chloride)
Decaf luorobiphenyl
2-Fluoroaniline
Pentafluorophenol
Semi volatile Internal Standard Solution F
(all compounds at 200 ug/ml in methylene chloride)
Di2-Benzo(a)pyrene
Ds-Naphthalene Ds-Aniline
DiQ-Phenanthrene Ds-Phenol
DlO-Biphenyl Ds-Nitrobenzene
DiQ-Acenaphthene D3-2,4-Dinitrophenol
D]_Q-Pyrene . Decafluorotriphenylphosphine
(continued)
119
-------
TABLE 25. (Continued)
Volatile Calibration Solution G
(all compounds at 0.2 mg/ml in methanol)
2-Chloroethylvinyl ether Dimethyl disulfide
1,1,2-T ri chl o rot ri f 1 uoroethane Epi chl orohyd ri n
Dibromomethane 2-Chloroacrylonitrile
Ally! chloride Acetonitrile
Ethylene dibromide Dichloroacetonitrile
Chloropicrin n-Propionitrile
2-Chloropropane Chloroacetaldehyde
1-Chlorobutane 2-Chloroethanol
o-Xylene N-Nitrosodimethylamine
Styrene Vinyl acetate
2-Butanone Dimethyl sulfide
Cyclopentanone Diethyl ether
4-Methyl-2-pentanone Acetone
2-Hexanone Methyl chloroacetate
Carbon disulfide Methyl acrylate
Methyl methacrylate
Volatile Surrogate Standard Solution H
(all compounds at 10 mq/ml in methanol)
1,2-Di bromotetraf1uoroethane
Bis(perfluoroisopropyl) ketone
Fluorobenzene
m-Bromobenzotri fluoride
Volatile Internal Standard Solution I
(all compounds at 0.2 mg/ml in methanol)
D4-l,2-Dichloroethane
Ds-Benzene
Ds-Ethylbenzene
4-Bromofluorobenzene
Volatile Calibration Solution J
(all compounds at 0.2 mg/ml in water)
Acrolein
Acrylonitrile
(continued)
120
-------
TABLE 25. (Continued)
Volatile Calibration Solution K
2-Chloroethyl vinyl ether
0.2 rug/ml in tetraglyme
Supelco's Purgeable A
(all compounds at 0.2 nig/ml 1n methanol)
Methylene chloride
1,1-Dichloroethene
l,l-D1chloroethane
Chloroform
Carbon tetrachloride
1,2-01chloropropane
Trichloroethylene
1,1,2-Trlchloroethane
Dibromochl oromethane
Tetrachloroethene
Chlorobenzene
Supelco's Purgeable B
(all compounds at 0.2 mg/ml in methanol)
trans-1,2-01chloroethene
1,2-Dichloroethane
1,1,1-Trichloroethane
Bromodichl oromethane
trans-1,3-Dichloropropene
ci s-1,3-Oichloropropene
Benzene
Bromoform
1,1,2,2-Tetrachloroethane
Toluene
Ethyl benzene
Chloromethane
Bromomethane
Supelco's Purgeable C
(all compounds at 0.2 mg/ml in methanol)
Vinyl chloride
Chloroethane
===========================================================================
directly to EPA. From among the laboratories that volunteered to participate
in the interlaboratory study, two were selected; these laboratories are
Environment Canada and Battelle Northwest Laboratories but these laboratories
eventually declined participation 1n the study.
121
-------
Names and addresses of the participating laboratories follow:
Acurex Corporation
405 Clyde Avenue
Mountain View, California 94043
Contact: Viorica Lopez-Avila
Telephone: (415) 964-3200
Arthur D. Little, Incorporated
20 Acorn Park
Cambridge, Massachusetts 02140
Contact: Linda Sadowski
Telephone: (617) 864-5770
California Analytical Laboratories
5895 Powrer Inn Road
Sacramento, California 95824
Contact: Mike Miilie
Telephone: (916) 381-5105
GCA Corporation
Technology Division
213 Burlington Road
Bedford, Massachusetts 01730
Contact: Gary Hunt
Telephone: (617) 275-5444
Mead-CompuChem
P.O. Box 12652
Research Triangle Park,
North Carolina 27709
Contact: Paul Mills
Telephone: (919) 549-8263
Midwest Research Institute
425 Volker Boulevard
Kanas City, Missouri 64110
Contract: Jim Spigarelli
Telephone (816) 753-7600
Southern Research Institute
2000 Ninth Avenue South
Brimingham, Alabama 35205
Contract: Ruby James
Telephone: (205) 323-6592
Southwest Research Institute
6220 Culebra Road
San Antonio, Texas 78238
Contact: Carter Nulton
Telephone: (512) 684-5111
West Coast Technical Services
Incorporated
17605 Fabrica Way
Cerritos, California 90701
Contact: Rich Amano .
Telephone: (213) 921-9831
Environment Canada
Laboratory Services Division
Air Pollution Technology Center
River Road Laboratories
Ottawa, Ontario K1A 1C8
Contact: Judy Lockwood
Telephone: (613) 998-3671
Battelle
Pacific Northwest Laboratories
Battelle Boulevard
Richland, Washington 99352
Contact: Don Schoengold
Telephone: (509) 376-0005
122
-------
QUALITY ASSURANCE/QUALITY CONTROL
The objective of Quality Assurance (QA)/Quality Control (QC) activities
conducted for a'ny chemical analysis program is to provide data of known
quality. In case the results of analyses are contested in any way, the
quality of these data must be demonstrable.
While the data from this study are not likely to be directly contested in
a pollution assessment case, the validity of the method may be contested and
does need to be substantiated. The results of this study will form the basis
for quality control requirements when the method is applied routinely.
QA Objectives
The objectives of QA/QC activities in the interlaboratory test were
to make certain that the laboratory work conducted to evaluate the chemical
analysis methods was done under controlled conditions, that those controls
were uniformly applied by all collaborators and that all experimental work
was recorded for archival storage.
In addition, when the analysis method 1s fully evaluated and is applied
for routine analyses of hazardous wastes, the method description will Include
the necessary and appropriate quality control elements and requirements. Part
of the requirements for that quality assurance and quality control will be
based on experience and knowledge derived from this evaluation. Therefore, 1t
is expected that all persons involved in this program will be aware of the
ultimate use of the methods and will be alert, sensitive and critical to
controls instituted to provide high quality data.
Quality Control - Performance Criteria and Checks
Quality control activities start with a description of the method which
must be followed without exception. Before any laboratory work, the method
must be read and understood by all personnel who use 1t. Questions regarding
what is to be done in the interlaboratory test must be discussed with Battelle
persons before laboratory work starts. In that way uncertainties can be
corrected or clarified among all cooperating laboratories and all will possess
the same information prior to analysis. Thus, to the extent possible, the
collaborators will conduct all operations in exactly the same manner.
Control was maintained by monitoring the mass spectrometer tuning (using
DFTPP or BFB), and analyzing process blanks and calibration standards. The
details of the required quality control measures are described in the methods
and in the QC Protocol for Fused Silica Capillary Columns.
In application of the methods it was expected that care was exercised
to use properly calibrated and clean apparatus such as balances and volumetric
glassware for extract preparation and to ascertain that the GC/MS system was
functioning properly.
123
-------
On-Site Audits and Performance Evaluation Samples
After the study participants received and reviewed the Instruction manual
each laboratory underwent two audits. The first was a systems audit conducted
at each laboratory to ascertain that the facilities necessary to conduct the
study were available and to review the instruction manual. If there was any
misunderstanding of the manual by one or more laboratories, clarifications were
distributed to all participants before the study continued.
The second audit consisted of performance evaluation samples analyzed by
each participant. This exercise not only provided laboratory qualification
but offered another opportunity to clarify misunderstandings and correct un-
clear statements in the instruction manual and gave an excellent opportunity
for participants to become intimately familiar with the test methods.
GC/MS Run Logs
Several forms were supplied for reporting analytical information and a
log of GC/MS runs. A separate log (Forms IV and IS) was required for each
mass spectrometer used to determine volatile compounds and for each mass
spectrometer used to determine semi volatile compounds. Run numbers should
be consecutive from the very first calibration run to the last sample or
calibration run. We intend that the analysis runs for this program be
consecutive and without interruption by other programs.
• •
GC/MS Calibration
The daily run routine must include at least one calibration run at the
beginning of the day and additional calibration runs during the day if more
than 8 sample runs were made in one day.
Surrogate Standards
All samples must be spiked with surrogate compounds before extraction.
The surrogate compounds are:
For Volatile Compounds
1,2-Dibromotetrafl uoroethane
Bis(perfluoroisopropyl) ketone
Fluorobenzene
m-Bromobenzotrif1uoride
For Semivolatile Compounds
Decaf1uorobi phenyl
2-Fluoroaniline
Pentafluorophenol
124
-------
The spiking level used should be that which gave a concentration In the final
extract used for GC/MS analyses that was equal to the level of the Internal
standard added, assuming 100% recovery. This level was determined as described
1n Section 8.4 of the methods. Thus two allquots of each sample must be screened,
one for volatile compounds and one for semlvolatile compounds, before allquots
can be spiked with surrogates and analyzed.
Blanks
Blanks were defined as system or process blanks and consisted of all
reagents used 1n sample preparation and carried through the entire preparation
process and finally analyzed by GC/MS. This activity assessed the purity of
reagents and cleanliness of apparatus and environment. It was required that
a system blank be generated and analyzed with every batch of samples prepared
or with every new batch of reagent material. A minimum of two process blanks
for the volatile analyses and two process blanks for the semivolatile analyses
must be run and the data reported.
Documentation and Records
The documents for this program include the Manual of Instructions for
collaborators of which this QA/QC plan was a part, the Program Review Inquiry
form, data reports, letters of transmlttal.pertinent records of telephone con-
versations and all data and records associated with this program. Copies of
these documents will be kept on file by Battelle for audit purposes and for
possible submission to the EPA at the conclusion of the study.
A record shall be kept by participants of all efforts and events
associated with the laboratory work and of all data such as:
• Sample Handling
Date received
Volume and/or weight of samples
Condition of samples
Location and temperature of storage
Date removed from and returned to storage
• Analytical Data
Date of extraction and GC/MS analysis
All volumes and weights used
Dilution and concentration factors
Amount of internal standard added —
Internal standard area response
Injection volume
Relative response factors used for quantification
Total solvent extractable content
125
-------
Major volatile compounds content
Scan number
Absolute and/or relative retention time
Most intense ions
Compound identification
Proba'ble molecular weight
Total ion current chromatograms
Library search results
9-Track tape files
Search system used
Calibration results
Mass spectrometer tuning results
Maintenance records
Most of the above information is "required data" to be reported on forms
supplied or in specified format. The deliverables required from each
participating laboratory are as follows:
• Amount, absolute retention time, relative retention time, and response
factor used for each of the target (up to 200) compounds found in each
of three replicates of 10 waste samples at levels greater than
0.02 times ,the level of internal standard used.
• Amount, tentative identification most intense ion, absolute retention
time and relative retention time for each of the 20 major volatile
compounds and the 20 major semi volatile compounds found in each of the
three replicates of 10 waste samples (some or all of these may be
among the list of 200).
• Relative ion abundances found in each GC/MS run (including blanks,
calibration runs and sample runs) for the tuning compound (BFB or
DFTPP) included in the internal standard solution.
• Peak area counts found for De-benzene or Dio-phenanthrene in each
GC/MS run.
t Relative retention times and response factors found for each of the
200 compounds in initial GC/MS calibration runs using
Battelle-supplied calibration solutions.
• Relative retention times and response factors found for each of 20
volatile compounds and each of 80 semi volatile compounds in each GC/MS
calibration run using Battelle-supplied calibration solutions.
• Percent recovery of three surrogate compounds spiked into each
replicate analyzed.
• The total ion chromatogram (8-1/2 x 11) from each GC/MS run with a
scan number scale indicated and the internal standard peaks and
surrogate standard peaks indicated.
126
-------
• The enhanced mass spectrum and library search results (each on one
8-1/2 x 11 page) for each compound reported that 1s not on the 11st of
200 compounds.
• A narrartlve description of the GC/MS system used to acquire data and
the computer software and/or manual processes used to obtain the
qualitative and quantitative results reported.
« A completed program review Inquiry to critique the methods and
Indicate where clarifications, modifications, or additional
specifications would be useful 1n the methods.
• In addition to the above, all of the GC/MS raw data must be archived on
9-track tape for a period of one year.
Performance criteria associated with these functions have been established
for the Intsrlaboratory test phase of the program and are a part of the Instruc-
tion manual.
QC For Fused Silica Capillary Columns
A QC Protocol for fused silica capillary columns was developed by Acurex
under the direction of Mr. Drew Sauter, EPA-LV and modified by Battelle for
application to this Interlaboratory study. This protocol for the use of fused
silica capillary columns must be followed when the method for the determination
of semlvolatlle organic compounds was used. No deviations from this protocol
were permitted. This protocol (revision as of July 1982) 1s presented 1n
Appendix A.
Quantification Limits
The method descriptions cite an Ideal detection limit of 1 ppm with higher
limits expected for complex samples. However, there was a need to cite detec-
tion limits for defenslblllty of data.
Although a study of quantification limits has been conducted on a separate
contract with EMSL-C1nc1nnat1, some additional comments regarding quantifica-
tion limits that can be expected may be helpful at this time. For the purposes
of this discussion, we assume that detection limit means quantification limit.
As stated 1n the methods under the section on Scope and Application, the detec-
tion limit 1s approximately 1 ug/g but proportionately higher for samples that
contain more than 1 mg/g of total solvent extractable material (semlvolatHes
method) or total volatile material (volatHes method). This estimated quanti-
fication limit 1s based on the premise that the amount of the primary Internal
standard used was 50 times the lowest amount that can be quantified. Thus, 1n
the method for senrlvolatHes, 1n which 50 ug/ml of Dio-phenanthrene was used as
the primary Internal standard, the lowest amount that can be quantified was
taken as 1 ug/ml which would be 1 ng Injected. For samples containing 1 mg/g
or less of total solvent extractable content (TSEC), the extract from 1 g of
the waste would be concentrated to 1 ml (to give a solution containing 1 mg
TSEC/ml) and treated with 50 ug of Internal standard'. Thus, the Internal
standard would be present at a level representing 50 ug/g of waste and the
127
-------
quantification limit would be 1 ug/g. If the sample contained 100 mg/g of
TSEC, the extract from 0.01 g of the waste would be diluted to 1 ml (to give
a solution containing 1 mg TSEC/ml) and treated with 50 ug of the internal
standard. In this case, the Internal standard would be at a level representing
5000 ug/g of wa'ste and the quantification limit would be 100 ug/g.
The quantification limit is based on the lowest mass spectrometer
response that can be quantified. If a Finnigan 4000 is used and an area of
100,000 1s obtained for 50 ng of DiQ-phenanthrene, the above rationale means
that an area of 2000 is being taken as a practical quantification limit in a
complex sample matrix. The rationale also Implies that the quantification
limit of a compound of Interest will vary with Its response factor relative to
010-phenanthrene. Thus, if the response factor of a compound relative to
DIO-phenanthrene is 0.1, the area obtained would be 10 times lower and the
quantification limit would be 10 times higher.
The following formula may be used for estimating the quantification limit
of the method when applied to the determination of a particular compound In a
particular waste:
1
Quantification Limit, ug/g = —————— x TSEC, mg/g.
RF rel . to
A similar treatment of the method for volatiles leads to a similar formula in
which D6-benzene is the primary internal standard and total volatile content
(TVC) 1s used to determine the volume of extract analyzed:
1
Quantification Limit, ug/g = ---- X TVC, mg/g.
RF rel to D6-benzene
The above formulas provide reasonable but sometimes conservative estimates of
the quantification limits. For example, in some cases, the concentration of
the extract analyzed may be 10 times greater than that considered in the above
calculations and result in a quantification limit that is 10 times lower than
the estimate given by the formulas.
PROPOSED ADDITIONAL STUDIES
During the course of conducting the Phase II studies and reviewing the
performance evaluation data from Phase III, several additional studies that
would strengthen the overall program have become apparent to us. Some of the
additional studies that would benefit the program, e.g., homogeneity of the
spiked wastes, quantification limits of the methods, comparison of the tetraglyme
method to direct purge and trap methods, methyl t-butyl ether as an extraction
solvent, and extract matrix effects, have been discussed previously and were
conducted on other contracts. Additional studies recommended are discussed
below.
128
-------
DFTPP Stability
In order to determine whether the mass spectrometer tuning meets the EPA
criteria for use 1n the determination of semivolatile compounds, DFTPP must be
Introduced to the mass spectrometer via the gas chromatograph. This GC/MS
analysis Is generally made as a separate run at the beginning of each day
prior to the analysis of calibration samples. In the method prepared for the
collaborative study, DFTPP was specified to be Included in the internal
standard solution and to be analyzed in every run. With this protocol the
DFTPP check of tuning could be conducted as part of the first calibration run
of the day and thereby decrease the time and costs involved. Unfortunately,
DFTPP solutions frequently degrade within a few days, so that the resulting
DFTPP mass spectrum obtained is too weak to give good relative ion abundance
data. When DFTPP is analyzed by itself, a fresh solution is frequently pre-
pared every few days to avoid the degradation problem. It is not practical to
prepare fresh multicomponent Internal standard solution or calibration solu-
tions containing DFTPP every few days. The multicomponent Internal standard
solution containing DFTPP that was prepared for the program was found to be
stable for several months at room temperature when stored in sealed glass.
After the ampoules were opened, the DFTPP seemed to decompose in a few days at
variable rates. The instability of DFTPP Is very likely caused by an air
oxidation of the phosphine to the phosphine oxide. It may be possible to
stabilize DFTPP solutions by the addition of an oxygen scavenger. In order
to be of use, the oxygen scavenger must not Interfere with the GC/MS analysis,
which in general means that it should be either nonvolatile or very volatile.
Two volatile compounds that may serve as effective oxygen scavengers and DFTPP
stabilizers are triethylphosphine and trimethyl phosphite.
Because of the time and cost advantages to be gained by Incorporating the
DFTPP tuning check with a calibration run, it 1s recommended that DFTPP
"poi
^ "DTI
stability and the effectiveness of oxygen scavengers be studied ~
Improved Internal Standard Solution
The internal standard solution that has been prepared for the collabora-
ive study contains eight deuterated compounds as internal standards, four
deuterated polar compounds as column performance standards, and DFTPP. It has
been found that one of the compounds originally suggested as an Internal
standard, DiQ-biphenyl, cannot be used in the study because of the frequent
interference of a coeluting compound, l-chloronaphthalene. Because of the
interference, neither Dig-biphenyl or l-chloronaphthalene can be quantified
reliably. In addition two column performance compounds in the internal
standard solution, namely De-phenol and Ds-anillne, occasionally coelute
and cannot be quantified reliably. In order to improve the usefulness of the
Internal standard solution it 1s recommended that DiQ-biphenyl be deleted as
an internal standard and that D3-2,4-dichlorophenol and Dg-2-am1nobiphenyl
be used 1n place of Ds-phenol and Ds-aniline as column performance
standards.
The above change will permit Ds-phenol to be available as an acidic
surrogate. In general, fluorinated compounds Instead of deuterated compounds
were selected as surrogates because many of the more common deuterated
129
-------
compounds were desired as internal standards and because many of the other
deuterated compounds available were too expensive. In many cases, as much as
20 to 50 mg of surrogate 1s required for the 3-gram sample analyzed. Although
most of the deuterated compounds available that are not used as Internal
standards would" be too costly, Ds-phenol, Dy-quinoline, and Dg-Fluorene
cost less than $10/50 mg and would be suitable as acidic, basic, and neutral
surrogates, respectively.
It 1s recommended therefore, that the deuterated compounds discussed
abovelte studied In respect to potential Interferences and, if found suitable,
be used In the remainder of the inter!aboratory study, it will also be of
Interest to study the usefulness of including additional compounds to monitor
the mass spectrometer tuning, e.g., pentafluorobromobenzene, octafluoro-
naphthalene, or trlch!orotrif!uorobenzene.
Gas Chromatography of Pyr1dines
Four pyrldlnes have been Included 1n the 11st of representative semi-
volatile compounds selected for the Inter!aboratory study. Even though the
pyrldlnes are relatively low boiling, they usually give very broad peaks under
the GC conditions used. Frequently there are two or more humps to the peaks
which Indicate that the pyrldines may be released from different salts 1n the
injection port at different rates, e.g. from salts with 2,4-D, 2,4,5-T,
chlorobenzolc acid, etc., which are included in the solutions. It has also
been noted in other work that the pyridine peak becomes much narrower 1f the
injector temperature is decreased. This latter observation may indicate a
reaction of the pyridine with the methylene chloride solvent to form a hydro-
chloride salt from which the pyridine is slowly released. It is recommended
that studies using other solvents, e.g., methyl t-butyl ether, and studies o'f
the effects of adding various acidic compound be conducted to Improve the
Chromatography of pyridines.The results may have a significant bearing on
the reliability of data from the quantification of both acids and bases. It
may be that the recommended GC/MS conditions are not suitable for the
simultaneous determination of some acids and bases. It may be found that
methylene chloride 1s not suitable as a solvent for compounds as strongly
basic as pyrldlnes.
Tetraglyme Method Interferences
Potentially interfering peaks can be obtained when the tetraglyme method
is used for the determination of purgeable compounds. Most of these peaks of
concern come from impurities in the tetraglyme while others may come from
antifoam agents used. When large aliquots of tetraglyme extracts are used,
significant amounts of tetraglyme may purge, foam, or both from the water and
show up as a very broad peak several runs later. It is recommended that
tetraglyme purification, tetraglyme substitutes, and antifoam agents bT"
studied to Improve the tetraglyme procedure.Tetraglyme purification would
Involve fractional distillation or prolonged sparging at an elevated tempera-
ture. A chemical supply house may be willing to provide purified tetraglyme.
A sparging procedure and the use of antlfoams 1s being evaluated for the
Cincinnati hazardous waste program.
130
-------
As far as a tetraglyme substitute is concerned, any solvent of interest
will need to be less volatile than tetraglyme and contain lower levels of
purgeable impurities than tetraglyme but retain all of the advantages of
tetraglyme; namely low viscosity, water-miscibility, oil-miscibility, and
availability. Some potential candidates as tetraglyme substitutes are
hexaglyme, trie'thylene glycol propionate, 15-crown-5 ether, and 18-crown-8
ether. Sources of such compounds have been investigated.
On-Column Injection
Variable injector discrimination can occur when the usual split or split-
less injection technique is used in capillary column GC work. The variable
discrimination leads to variable quantitative results. Injector discrimina-
tion can be avoided by using on-column injection. Various on-column injectors
for capillary column work are now available and their use is becoming more
common. In addition to avoiding injector discrimination, on-column injection
decreases compound degradation that may occur in the injector and cause poor
precision and accuracy. It is recommended that a comparison of the precision
and accuracy of on-column injection versus splitless injection be made using
extracts of the spiked waste samples being studied in the interlaboratory
program.After completion of this project, such a study was reported in J.
Chromatogr. Scl. 21, 512, 1983. Although it would not be practical to require
the participatingTaboratories to switch to on-column injection at this point,
the information obtained from the comparison study could be very useful in
future hazardous waste analysis programs.
Re-Analysis of Samples to Determine Semivolatile Compounds
Intralaboratory data recently collected at Battelle with the revised
method were better than the data collected in September. The extractions of
the six spiked wastes for the determination of semivolatile compounds in the
intralaboratory study were conducted last fall using the first generation
method. That method involved the use of a Polytron homogenizer. The method
was later revised to use sonification. In order to strengthen the background
data base for justification of the initiation of Phase III it Is recommended
that the six spiked waste samples be extracted and analyzed in triplicate at
Battelle using the revised method.
Statistical Evaluation of Data
The Phase III program as originally proposed includes a statistical
evaluation of the amounts of individual compounds found in the eleven ILS
samples by the ten participating laboratories. The data involved are those to
be reported on Form 2 and 3, i.e., amounts of listed and unlisted compounds.
However, a large body of additional information has been requested that may
provide explanations to variabilities encountered. This additional informa-
tion includes area and retention time of the primary internal standards,
De-benzene or DiQ-phenanthrene, relative ion abundances of DFTPP or BFB,
retention times and response factors of internal standard components, surro-
gate recovery data, retention times or relative retention times of all com-
pounds in all calibration runs, and response factors of all compounds in all
calibration runs. Some of this additional information from the performance
evaluation runs has already been evaluated to a limited extent. It is
131
-------
recommended that a more thorough evaluation be made of these additional data
from the performance evaluation runs and from the remainder of the study.
REFERENCES
1. Federal Register Monday, December 18, 1978, Part IV, Vol. 43, No. 243 p.
58956.
2. ASTM, Annual Book of Standards, 1981, Part 41, p. 964.
3. ASTM Standards on Precision and Accuracy for Various Applications 1st Ed.
1977.
4. Final Report on Residual Waste Analyses to EPA, March 31, 1981, Contract
68-02-3628, Battelle.
132
-------
Phase III Studies
133
-------
INTRODUCTION
The Environmental Protection Agency, under the authority of the Resource
Conservation and Recovery Act of 1976, is charged with the responsibility of
assessing the potential hazard of municipal and industrial solid wastes. One
of the characteristics of a waste that must be determined in order to assess
the potential hazard associated with the disposal of the waste, is the organic
chemical content. A variety of analytical methods have been developed and
applied to the determination of organic constituents in solid waste. Each
research group associated with the development of a given method instituted
appropriate Intralaboratory quality control to determine the applicability,
accuracy and precision of the method developed. However, none of the methods
has been evaluated by conducting a well-designed Interlaboratory comparison
study. The interlaboratory collaborative study conducted by Battelle involved
three phases:
I Methods Evaluation, Modification and Selection
II Intralaboratory Evaluation and Methods Revision
III Interlaboratory Evaluation
The objective of Phase I was the selection of two methods using combined
gas chromatography/mass spectrometry (GC/MS) for the analysis of solid wastes,
one method for determination of volatile organic compounds and one method for
determination of semi volatile organic compounds. The results of the Phase I
effort were presented in the Phase I Report dated July 14, 1981 (1). The
objective of Phase II was the single laboratory application, critical review
and revision of the methods selected in Phase I. Phase II also included the
development of a protocol for conducting the Interlaboratory collaborative
study. The results of the Phase II effort were presented in the Phase II report
dated September 24, 1982 (2).
The objective of Phase III, the subject of this report, was the performance
of a collaborative study to assess the precision and accuracy of the revised
methods that resulted from Phase II research efforts. The collaborative study
involved triplicate analyses of seven different spiked waste samples by nine
participating laboratories using the two revised GC/MS analysis methods. The
data were used in a statistical analysis to determine the between-laboratory
and within-laboratory components of the total variability.
134
-------
CONCLUSIONS
Phase I
t Gel permeation chromatography was of marginal value and was not
necessary because industrial wastes generally do not contain
high molecular weight substances,
• Homogenization of wastes can in most cases be carried out through use
of a Brinkmann Polytron Homogenizer although sonication was the ultimate
method of choice,
t Fused silica capillary columns were recommended for GC/MS studies and
have been specified in the methods description.
Phase II
Tetraglyme was found to be superior to polyethylene glycol as the
solvent for extraction of volatile organic compounds because of lower
viscosity and better solvent properties.
Ultrasonic energy was found to be superior to mechanical energy for
dispersing a sample in methylene chloride for extraction of semi-
volatile compounds.
The methods selected in Phase I were, with the above modifications,
found to be suitable for use in Phase III.
The methods are adequate but some additional work is suggested as
follows:
Stability of DFTPP and the effectiveness of oxygen scavengers to
stabilize DFTPP should be studied.
Because of interferences in the use of the recommended internal
standard compounds (for semivolatile compounds) the use of other
internal standards should be studied in respect to interferences.
To improve chromatography of pyridines and based on observations made
during Phase II studies, it appears beneficial to investigate the use-
fulness of solvents other than methylene chloride.
Potentially interfering peaks were frequently obtained when the
tetraglyme method was used for the determination of purgeable com-
pounds. Tetraglyme purification, tetraglyme substitutes and antifoam
agents need to be studied to improve the procedure.
135
-------
The precision and accuracy obtained using on-column Injection versus
split!ess Injection should be compared.
The mea.ns suggested to calculate detection limit provides a conser-
vative estimate of quantification limits. Some work was needed to
provide a rigorous, yet simple means, for estimating detection and
quantification limits.
Phase III
• Sample-to-sample variations were evident for both analysis procedures.
t Precisions of relative retention times for semivolatiles were excellent
(average of 1.9%).
t Precisions of retention times for volatiles were good (average of 13%).
• Response factor relative standard deviations were less than 40% for SVs
and ranged from 15 to 80% for VOAs.
• Laboratory-to-laboratory variability was evident for the total number
of compounds detected.
• Within laboratory component of the spike recovery variability averaged
about 30% for both volatlles and semivolatiles.
t Between laboratory component of the variability was about twice the
within laboratory component for both volatiles and semivolatiles.
t Seventy percent of the potential positive results of high level spike
data were reported compared to 50 to 60 percent of potential low level
spike data.
t Tetraglyme VOA method gave best results for high boiling compounds and
poorer results for more polar compounds.
• Semi volatile method gave best results for aromatic hydrocarbons and
halocarbons and poorer results for benzole acids and nitrophenols.
The most striking difference 1n the Interlaboratory application of the two
analysis methods was evident for the screening phase of both protocols. The
amount of tetraglyme extract used for the purge and trap analysis varied as
much as three orders of magnitude. The concentration factor used for the
semi volatile methylene chloride extract also varied by as much as three orders
of magnitude. Interpretation of instructions for these decisions by each of
the participating laboratories should be examined.
The quantity and type of organic compounds found for each of the waste
samples indicated that some of the samples could have undergone change with
time. More time elapsed between preparation of spiked samples and analysis
136
-------
than planned in the experimental design. Performance evaluation samples were
analyzed by each of the participating laboratories to provide evidence of
capability. The planned time of three weeks for this event was exceeded by a
factor of 3. The time originally planned for interpretation of Performance
Evaluation data was also exceeded since it was necessary to evaluate a number
of factors prior to approval by EPA to initiate Phase III activities. Finally,
the Phase III interlaboratory results were requested in two months, however,
four months were required by the participants due to both scheduling and
contractual difficulties. The primary technical problem was the purity of
tetraglyme required by the VGA protocol. The tetraglyme batch used by Battelle
1n the Phase II developmental work did not cause interferences. Some of the
other laboratories could not obtain sufficiently pure tetraglyme for Performance
Evaluation sample analysis causing time delays. EMSL-LV arranged for pure
tetraglyme, supplied by Radian, to be made available for Phase III of the
study.
Examples of suspected sample degradation include ILS-5 and ILS-6 which
contained fewer volatile organic constituents than originally found. Also
degradation of DDT and absence of nitro compounds were observed for ILS-6.
137
-------
RECOMMENDATIONS
Further examination of certain results and observations of the collabora-
tive study for evaluation of hazardous waste methodology will significantly
enhance the value of this complex study. The volume of data generated probably
far exceeds data obtained for any previous interlaboratory method evaluation.
The scope of the study covered statistical evaluation of only a small portion
of this data base. In addition, no provisions were included for statistical
reanalysis excluding outlier results. Based on the interpretation of the
analytical data the following recommendations were presented for consideration:
• Expand the volume of statistically analyzed data to include further
analysis of data to identify observed variances.
• Repeat statistical analysis of percent recovery data excluding obvious
outliers.
• Investigate sources of'Significant variability such as
-- Prescreening
~ Extraction
-- Instrumental Analysis
-- Response factor determination
-- Confirmation of compound identification and quantification procedures
-- Calculation
• Conduct in-depth interview at collaborators laboratories to evaluate
interpretation of instructions (both excellent and poor performances
would provide valuable insight).
0 Modify method instructions and quality control protocols to eliminate
potential sources of variance where possible.
t Investigate and confirm suspected sample degradation for ILS-5 and
ILS-6.
The conduct of the tasks recommended above will result in a more thorough
evaluation of the data generated by the collaborative study. It is antici-
pated that the revised comparative statistics will be significantly improved
in terms of precision and accuracy. Further, the resulting revisions in the
protocols for volatile and semivolatile organic analysis will result in more
138
-------
uniform application of these methods in future EPA analysis programs. Many of
the recommendations are being implemented on U.S. EPA Contract No. 68-03-3122
by the EMSL-CIN.
139
-------
EXPERIMENTAL PROCEDURES
ANALYTICAL TEST PROCEDURE
Two analytical test procedures were evaluated in the Interlaboratory study.
One of the procedures was a GC/MS analysis procedure for the determination of
volatile organic compounds. The procedure involved the extraction of a waste
sample with tetraethyleneglycol dimethyl ether (tetraglyme), the addition of a
portion of the extract to water, and analysis by the purge and trap technique
of EPA Method 624 (3). The other method, a GC/MS analysis method for the
determination of semi volatile organic compounds, involved the extraction of the
waste by sonification with methylene chloride under neutral conditions 1n the
presence of anhydrous sodium sulfate, and GC/MS analysis of the extract using
a fused silica capillary column. Complete descriptions of the two methods
are given 1n Appendix A and Appendix B. A quality control protocol for GC/MS
analyses involving fused silica capillary columns was specified as a supplement
to the method for the determination of semi volatile organic compounds. The
quality control protocol was based on a protocol developed by Acurex for EMSL-
Las Vegas and was revised by Battelle for applicability to the collaborative
study. The revised protocol is presented in Appendix C.
COLLABORATIVE TEST PROCEDURES
Preparation of Calibration Solutions
All of the standard solutions required for the study were purchased or
prepared by Battelle 1n sealed glass ampoules and distributed to the partici-
pating laboratories. The standards included standard calibration solutions,
surrogate recovery standard solutions, and internal standard solutions. The
standard solutions distributed for the determination of volatile compounds
included six calibration solutions containing a total of 53 representative
volatile compounds, a surrogate solution containing four fluorlnated compounds
as recovery standards, and an Internal standard solution containing bromofluoro-
benzene as a mass spectrometer tuning standard and three deuterated Internal
standards. The standard solutions distributed for the determination of semi-
volatile compounds included four calibration solutions containing a total of
140 representative semi volatile compounds, a surrogate solution containing
three fluorlnated recovery standards, and an internal standard solution con-
taining decafluorotriphenylphosphlne as a mass spectrometer tuning standard
and twelve deuterated internal standards and GC/MS performance standards. The
compositions of the standard solutions are given in Appendix D.
140
-------
Preparation of Spiked Wastes
The wastes used for the study were selected to challenge the extraction
procedure and to represent a broad range of waste types. The wastes are listed
in Table 1. Each waste was spiked with the 9 representative volatile compounds
and 36 representative semi volatile compounds listed in Table 2. The concentra-
tions spiked into the wastes are given in Tables 3 and 5 showing the volatile
and semivolatile compounds, respectively. These spike concentrations were
chosen to be similar to the levels of other volatile and semivolatile components
present in the unspiked wastes. Most of the spiked compounds were different
from those in the unspiked wastes.
Because of budgetary considerations, a spiking scheme was used to minimize
the number of analyses required for each waste. The scheme involved spiking
each waste in advance with a number of pairs of chemically compatible compounds.
Each pair represents a different compound class as indicated in Table 2. The
two compounds of each pair were selected on the basis of similar properties
(volatility, solubility, polarity, or acidity) that would lead to similar
recovery efficiencies. A study reported in the Phase II report demonstrated
that the two compounds 1n each pair did indeed give comparable recovery effi-
ciencies. One compound from each pair was spiked at a relatively low level
and the other at an approximately five-fold higher level. With this approach,
data for the determination of unspiked components and data for the recoveries
of different classes of compounds at two spike levels were obtained simultan-
eously with a single GC/MS run.
The high and low levels used corresponded to those levels that would give
approximately 50 ng and 250 ng of volatile compounds or 10 ng and 50 ng of
semi volatile compounds on the GC column during analysis 1f 100% recovery were
achieved. Since the degree of dilution or concentration required for each
waste varied widely from waste-to-waste, the actual spike level used also varied
widely from waste-to-waste, as shown in Tables 3 and 4.
Since the spike concentrations were relatively high, 1t was not possible
to spike the wastes with solutions of the spiking compounds without significantly
changing the nature of the sample and interfering with the determination of
volatile compounds. The volatile compounds were readily miscible with each
other and were mixed to give a single neat solution prior to spiking. The neat
semivolatHe compounds, both liquid and solid materials, were added to a glass
mortar and pestle and ground to a thin slurry of finely divided particles prior
to spiking. The spiking of each waste was conducted in a manner that resulted
in a homogeneous spiked sample which was not subject to phase separation. The
details of the spiking and homogenizing procedures were described in the Phase
II report.
Design of Collaborative Test
Each participating laboratory was sent detailed descriptions of the .two
analysis methods, ampoules of the standard solutions, samples of all of the
spiked wastes and a sample of a contaminated river sediment, ILS-9, which is NBS
Standard Reference Material No. 1645. ILS-9 was included to obtain reference
data for future work. Each laboratory was requested to analyze three replicates
141
-------
TABLE 1. WASTE SAMPLES USED IN THE STUDY
Program
Identification Physical
Number . Waste Name Description
Analyses
Performed(
ILS-2
ILS-3
ILS-4
ILS-5
ILS-6
ILS-7
ILS-8
ILS-9
Spiked latex paint waste
Spiked ethanes spent catalyst
Spiked coal gasification tar
Spiked oxychlorination spent
catalyst
Spiked POTW sludge
Spiked herbicide acetone-water
Spiked chlorinated ethanes waste
Contaminated river sediment
Semi-sol id
Oily powder
Tar
Pelletized solid
Wet filter cake
Liquid
Liquid
Oily powder
s,v
s,v
s,v
s,v
s,v
v
v
s
(a)s = determination of semivolatile compounds
V = determination of volatile compounds
of each sample to determine volatile compounds and three replicates to determine
semivolatile compounds as indicated In Table 1. The laboratories were requested
to search for 53 specific volatile compounds and 140 specific semivolatile com-
pounds and quantify the amounts of any compounds found. The specific compounds
are listed in Table 5. These compounds were selected as representative of
various classes of compounds and to challenge the method and its applicability.
The laboratories were also requested to report retention times or relative
retention times found, response factors used, and to quantify up to 10 major
unlisted compounds found 1n each sample. The detailed instructions sent to
each laboratory are given in Appendix E. The laboratories were sent special
report forms for reporting all data requested. Copies of the forms for reporting
data from the determination of volatile and semivolatile compounds are given in
Appendix F and Appendix G, respectively.
Data Processing Procedures
The data reported by the participating laboratories of this collaborative
study include the amount of compound found in the sample (AF), the response
factor (RF) and the retention time (RT) or relative retention times (RRT).
These data were reported by nine laboratories for 53 volatile compounds in each
of seven sample types, and for 140 semivolatile compounds in each of six sample
types. Data on the amount-found AF have been summarized in terms of percent
of compound recovered, percent of data recovered and within and between
laboratory relative precision.
Since each of the nine laboratories were required to make three replicate
analyses, there were potentially 27 values to be reported, for each compound, in
each sample type. However, due to various causes, data were not always reported
for each replication.
142
-------
TABLE 2. SPIKING COMPOUNDS
Volatile Compounds
Low-bo111ng Halocarbons
1,1,1-Trichloroethane
1,2-01chloropropane
High-boiling Halocarbons
Bromoform
1,1,2,2-Tetrachloroethane
Aromatic Hydrocarbons
Ethyl benzene
Chlorobenzene
Semi volatile Compounds
Aliphatic Halocarbon
Hexachloroethane
Low-boiling Aromatic Halocarbons
4-Chlorotoluene
1,4-01chlorobenzene
High-boil1ng Aromatic Halocarbons
Pentachlorobenzene
Hexachlorobenzene
Chlorinated Pesticides
p.p'-DDD
p,p'-DDT
Low-bom ng PAHs
2-Ethylnaphthalene
1-Chloronaphthalene
Ketone
2-Hexanone
N1tr11es
Prop1on1tr1le
2-Chloroacrylon1tr1le
ChloroaniUnes
4-Chloroan1l1ne
4-Chloro-2-methylan111ne
NitroanHlnes
3-N1troan111ne
2-Chloro-4-n1troan111ne
Pyr1 dines
2,4,6-Trimethylpyr1d1ne
4-t-Butylpyr1d1ne
Qu1nolines
Quinollne
4-Methylqu1nol1ne
Haloethers
B1s(2-chloroethyl)ether
B1s(2-chloroethoxy)ethane
(contained)
143
-------
TABLE 2. (Continued)
Semi volatile Compounds
Mid-boiling PAHs
Fluoranthene
Pyrene
Carbazole
1,2,7,8-Dibenzocarbazole
Aromatic Nltro Cpds.
1,3-01 nitrobenzene
2,6-Dinitrotoluene
Low-acidity Phenols
2-Chlorophenol
2,6-Dimethylphenol
High-acidity Phenols
4-N1trophenol
2,4-Dinitrophenol
Phosphates
Triphenyl phosphate
Tri-p-tolyl phosphate
Qul nones
Anthraquinone
2-Methylanthraqulnone
Aromatic Ketones
Acetophenone
Proplophenone
Benzole Acids
4-Chlorobenzoic acid
4-Bromobenzo1c acid
33833333aa3SBae33Bas33B3a3SBBa33Ba3338B33333B333aaaa333BBaB3asa33333s
The percent of compound recovered was calculated by
% Recovery
AF
x 100
(1)
where IAF is the initial amount found in the sample by Battelle analysis, and SLQ
is the spike level quantity. The summation is taken over all AF data reported.
The percent of AF data reported was calculated by dividing the number of data
reported by 27, the maximum number of available data points.
For each combination of sample type and compound, the data were analyzed
using the random components analysis of variance model
AF1d
i = 1,2 ..... 9
(2)
144
-------
TABLE 3. CONCENTRATION OF VOLATILE COMPOUNDS SPIKED INTO WASTE SAMPLES
en
Spike Concentration
Compound
Number
12
24
25
33
44
46
47
50
51
Compound
Proplonltrile
2-Chloroacrylonltrile
1 ,1 ,1-Trichloroethane
1 , 2-D1 chl oropropane
Bromoform
2-Hexanone
1,1,2 ,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
ILS-2
Latex
200
1000
800
200
1000
800
200
200
1200
ILS-3
EtCat
500
2000
500
3000
2000
500
500
500
2500
ILS-4
Coal Tar
500
100
100
400
100
100
500
600
100
ILS-5
Oxycat
25
5
5
20
5
5
25
30
5
:========:
, ug/g
ILS-6
POTW
5
30
25
5
30
25
5
20
5
=======
ILS-7
Herb
Acet
600
100
100
500
100
100
600
100
400
==========
ILS-8
Cl Et
Waste
16000
4000
24000
4000
4000
24000
16000
20000
4000
-------
TABLE 4. CONCENTRATION OF SEMIVOLATILE COMPOUNDS SPIKED INTO WASTE SAMPLES
Compound
Number
5
12
13
15
16
22
25
28
33
36
43
45
50
51
56
57
64
65
66
69
74
76
77
80
84
85
98
100
117
118
120
121
123
125
126
132
137
Compound
4-Chlorotoluene
Bis(2-chloroethyl )ether
2-Chlorophenol
2 ,4 ,6-Tri methyl py ri di ne
1,4-Dichl orobenzene
Acetophenone
Hexachloroethane
4-t -Butyl py ri di ne
2, 4-Dimethyl phenol
Propiophenone
4-Chloroaniline
Hexachloropropene
Quinoline
Bi s ( 2-chl oroethoxy Jethane
4-Chl oro-3-methy 1 ani 1 i ne
4-Chlorobenzoic acid '
1-Chloronaphthalene
4-Methylquinoline
2-Ethyl naphthalene
4-Bromobenzoic acid
1,3-Di nitrobenzene
2,6-Dinitrotoluene
3-Nitroaniline
2,4-Dinitrophenol
4-Nitrophenol
Pentachl orobenzene
2-Chloro-4-nitroaniline
Hexachl orobenzene
Anthraquinone
Fluoranthene
2-Methylanthraquinone
Pyrene
4,4' -ODD
4 ,4 '-DDT
Triphenyl phosphate
Tri -(p-toly)phosphate
Dibenzocarbazole
S 83 33 3883338833838888 3 3 888 33338
ILS-2
Latex
600
500
600
600
100
100
500
100
100
400
600
100
500
100
100
500
100
100
500
100
500
100
500
400
100
100
100
600
600
100
100
400
100
400
600
100
Spike
ILS-3
EtCat
800
200
800
800
200
1000
200
200
200
200
800
1200
200
1200
200
200
1200
1200
200
1200
200
1200
200
200
1000
200
1200
800
800
1000
200
200
1000
200
800
200
Concentration, ug/9
ILS-4
Coal Tar
1900
11000
1100
1900
7600
1900
11400
7600
7600
9500
1900
1900
11400
1900
7600
11400
1900
1900
11400
1900
11400
1900
11400
9500
1900
9600
1900
1900
1900
1900
7600
7500
1900
9500
1900
7600
ILS-5
Oxycat
4
24
4
4
16
4
24
16
16
20
4
4
24
4
16
24
4
4
24
4
24
4
24
20
4
16
4
4
4
4
16 .
20
4
20
4
16
ILS-6
POTW
20
20
20
20
100
20
20
100
100
120
20
80
20
80
100
20
80
80
20
80
20
80
20
120
20
100
80
20
20
20
100
120
20
120
20
100
100 1200 1900 4 80
8888888333333838833833883338383883883838
146
-------
TABLE 5. SPECIFIC COMPOUNDS SEARCHED FOR IN SAMPLES
8333333333=3 3:^:i--i:^
3I = -=^-— 3 = = = 2 3:3 3=3 '.3 = 3333333333333333333 3=53 33 3 3 -3=3333 = 3
VOLATILES
Purgeable Halocarbons
Chloromethane
Bromomethane
Vinyl chloride
Chloroethane
Methylene chloride
Tri chlorof1uoromethane
1,1-Dichloroethene
l,l-D1chloroethane
trans-l,2-Dichloroethane
Chloroform
1,2-Dichloroethane
1,1,l-Tr1chloroethane
Carbon tetrachloride
BromodiChloromethane
1,2-DI chloropropane
trans-l,3-Dichloropropene
Trlchloroethene
01bromochloromethane
1,1,2-Tri chloroethane
ds-l,3-Dichloropropene
2-Chloroethylvinyl ether
Bromoform
1,1,2,2-Tetrachloroethane
Tetrachloroethane
Chlorobenzene
1,1,2-Tr1chlorot ri f1uoroethane
Dlbromoethane
Allyl chloride
Ethylene dibromide
2-Chloropropane
1-Chlorobutane
Aliphatic Halocarbons
l,4-D1chlorobutane
Pentachloroethane
Hexachloroethene
Hexachloropropene
Hexachlorobutadlene
SEMIVOLATILES
Purgeable Hydrocarbons
Benzene
Toluene
Ethyl benzene
o-Xylene
Styrene
Purgeable Oxygen. Sulfur.
or Nitrogen Compounds
2-Butanone
4-Methyl-2-pentanone
2-Hexanone
Carbon disulflde
Dimethyl disulflde
Acrylonltrile
2-Chloroacrylom'trile
Acetonltrile
01chloroacetoni tri1e
n-Prop1on1trile
Acroleln
Vinyl acetate
Dimethyl sulfide
D1ethyl ether
Acetone
Methyl acrylate
Methyl methacrylate
Aromatic Halocarbons
4-Chlorotoluene
Bromobenzene
1,2-Dichlorobenzene
1,4-01chlorobenzene
1,2,4-Tri chlorobenzene
1,2,4,5-Tetrachlorobenzene
Pentachlorobenzene
(continued)
147
-------
TABLE 5. (Continued)
SEMIVOLATILES (continued)
Aromatic Halocarbons (continued)
Hexachlorobenzene
3,3'-Dichlorobiphenyl
4,4'-Dichlorobiphenyl
2,2',4,4'-Tetrachlorobiphenyl
Benzal Chloride
2,2',4,4',6,6'-Hexachlorobiphenyl
Benzyl chloride
1-Chloronaphthalene
2-Chloronaphthalene
a,a,a-Trichlorotoluene
Aromatic Hydrocarbons
Naphthalene
1,2,4-Trimethylbenzene
1,2,4,5-Tetramethylbenzene
Biphenyl
Acenaphthylene
Acenaphthene
2-Methy1 naphtha!ene
2-Ethylnaphthalene
2,3-Di methylnaphthalene
1,2,3,4-Tetrahydronaphthalene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(k|fluoranthene
Benzo(a)pyrene
Dibenzo(a,h)anthracene
Benzo(g,h,i)perylene
Indeno[l,2,3-cd)pyrene
Amines
Aniline
4-Chloroaniline
4-Bromoaniline
2-Nitroaniline
Aromatic Nitro Compounds
Nitrobenzene
1,3-Dinitrobenzene
2-Nitrotoluene
4-Nitrotoluene
2,4-Dinitrotoluene
2,6-Dinitrotoluene
l-Chloro-4-nitrobenzene
2,4-Di nitrochlorobenzene
Phenols
2-Chlorophenol
2-Nitrophennol
Phenol
2,4-Dimethylphenol
2,4-Dichlorophenol
2,4,6-Tri chlorophenol
4-Chloro-3-methyl phenol
2-Methy1 phenol
4-Methylphenol
Thiophenol
4-Chlorophenol
2,6-Dichlorophenol
2,4,5-Trichlorophenol
Hexachlorophene
4-Hydroxybiphenyl
2-Naphthol
4-t-Butyl phenol
2-Chloro-4-nitrophenol
2,4-Dinitrophenol
2-Methyl-4,6-dinitrophenol
Pentachlorophenol
4-N1trophenol
2,6-Di-t-butyl-4-methylphenol
2,4-Di-t-butylphenol
Diethylstilbestrol
Chlorinated Pesticides
4,4'-ODD
4,4'-DDE
4,4'-DDT
(continued)
148
-------
TABLE 5. (Continued)
========
SEMIVOLATILES (continued)
Amines (continued)
3,4-Dichloroaniline
2,4,5-Trichl oroani 1 i ne
3-Nitroaniline
4-Chl oro-2-nitroani1i ne
2,4-Dinitroaniline
N-Methylaniline
4-Chloro-o-toluidine
4-Methyl aniline
2,6-Dimethylaniline
4-Aminobiphenyl
1-Aminonaphthalene
N.N-D1methyl aniline
Pyridine
4-Methyl pyri dine
2,4-D1methylpyridine
4-t-Butyl pyri dine
1,2,7,8-Dibenzocarbazole
2,4,6-Trimethylpyridine
Quinoline
4-Methyl quincline
Acridine
Carbazole
3,3'-Dichlorobenzidine
Diphenylamine
Ketones
Anthraquinone
2-Methylanthraquinone
Propiophenone
Acetophenone
Haloethers
Bis(2-chloroethyl) ether
B1s(2-ch1oroetho*y)ethane
4-Chlorophenyl phenyl ether
Chlorinated Pesticides (continued)
Metho^ychlor
Trifluralin
Pentachloronitrobenzene
Phthalates
Dimethyl phthalate
Di-n-butyl phthalate
Di(2-ethylhe*yl) phthalate
Phosphates
Tri(p-tolyl) phosphate
Triphenyl phosphate
Aldehydes
Benzaldehyde
4-Chlorobenzaldehyde
Esters and Sulfides
Anisole
Phenyl ether
Dibenzofuran
Acids
acid
4-Chlorobenzoic
Benzoic acid
4-Bromobenzoic acid
2,4-01chlorophenoxyacetlc acid
2,4,5-Dichloropheno>Q'acet1c acid
Miscellaneous
Azobenzene
Acetanilide
Benzyl alcohol
Di(2-ethylhexyl) sebacate
149
-------
where AF^ was the amount of compound found on the j^ replication by the i_
laboratory. L-j was the random systematic error due to laboratory i and EJJ was
the random within-laboratory error. It was assumed that Lj was distributed
normally with mean zero and variance and <*e/u x 100, respectively, were estimated using
the AF data obtained from the laboratories. Since there were instances of
missing data, appropriate formulas for the construction of one-way analysis
of variance tables with unequal samples sizes were employed (4, 5).
In addition to the above analyses, the percent recovery of each compound
and sample type was calculated for each laboratory as
E
% Recovery = .A + IjL x 100 (4)
150
-------
RESULTS AND DISCUSSION
The participating laboratories were required to report the retention
times and response factors for the semi volatile and volatile compounds found
In each waste sample. A summary of these data with variability expressed as
percent relative standard deviation (RSD) is provided in Tables 6 and 7. The
percent RSD for the relative retention times of semi volatile compounds was
generally less than five percent and ranged from 15 to 30 percent for the
volatile compounds. The response factor data generated by this program repre-
sent the largest reported volume of comparative data for this parameter. The
response factors RSDs for the semivolatile compounds were generally less
than 40 percent and the volatile compound RSDs ranged from 15 to 80 percent.
The average percent recoveries and the total, between laboratory, and
within laboratory relative standard deviations for the spiked volatile compounds
in the seven wastes analyzed to determine volatile components are given 1n
Tables 8 to 14. These data tables include the amount of compound spiked and
amount-found (ng/g) for each of the reported spiked compounds and the amount-
found for listed nonsplked compounds reported by a significant number of lab-
oratories. The amount of data reported by the nine participants is also
provided. If data were reported for fewer than 20 percent of the analytical
runs the percent RSDs were not determined. The variability of volatile organic
analysis (VOA) results exhibited a dependence on sample matrix. For example,
the average of the average percent recovery and range for four samples are
summarized in Table 15. Variability of data reported by the laboratories was
evident and was not reproducible on a per sample basis, i.e., a certain labo-
ratory could report a larger volume of data for one sample than the other
laboratories and report the fewest number for another sample. For example, for
ILS-2 the number of compounds reported ranged from 6 to 9 with an average of 8.
The within laboratory variability for analysis of volatiles was generally
less than 30 percent but ranged from 5 to 300 percent depending on compound.
The total variability generally ranged from 20 to 80 percent for most of the
waste samples. The between-laboratory variability was usually less than 70
percent but ranged from 5 to 300 percent. The higher variabilities, both total
and component, were reported for non-spiked compounds such as methylene chloride,
dichloroethane, chloromethane, and chloroform. Only in the case of sample
ILS-4 were high percent RSDs reported for spiked compounds.
The corresponding summary data for the spiked semi volatile compounds in
the six wastes analyzed to determine semlvolatile components are given in Tables
16 to 21. The sample-to-sample variability noted for the volatile compounds
was also evident for the semlvolatile compound analyses. Summary data for
three samples are presented in Table 22. The differences in amount of data
reported were more remarkable for the semi volatile determinations than for the
151
-------
TABLE 6. VARIABLITY OF RELATIVE RETENTION TIMES AND RESPONSE FACTORS FOR THE
DETERMINATION OF SEMIVOLATILE COMPOUNDS
Ul
ro
Cpd.
No.
6.
12.
13.
15.
16.
22.
25.
28.
33.
36.
43.
50.
51.
56.
57.
64.
65.
66.
69.
74.
76.
77.
84.
85.
Compound
4-Chlorotoluene
B1s(2-chloroethyl )ether
2-Chlorophenol
2,4, 6-Trimethylpyri dine
1 ,4-Di chl orobenzene
Acetophenone
Hexachl oroethane
4-t-Butylpyridine
2 ,4-Dimethyl phenol
Propiophenone
4-Chloroaniline
Qu incline
Bi s (2-chl oroetho^y Jethane
4-Chloro-2-methyl am* 1 i ne
4-Chlorobenzoic acid
4-Chloronaphthalene
4-Methylquinoline
2-Ethyl naphthalene
4-Bromobenzoic acid
1 ,3-Di nitrobenzene
2,6-Dinitrotoluene
3-Nitroaniline
4-Nitrophenol
Pentachl orobenzene
Internal Standard
Ds-Bromo benzene
D5~Bromobenzene
Ds-Bromobenzene
Ds-Bromobenzene
Ds-Bromobenzene
Dg-Naphthalene
Dg-Naphthalene
Dg-Naphthalene
Dg-Naphthalene
Dg-Naphthalene
Dg-Naphthalene
Dg-Naphthalene
Dg-Naphthalene
D^Q-Acenaphthene
DiQ-Acenaphthene
DjQ-Acenaphthene
DjQ-Acenaphthene
DjQ-Acenaphthene
DiQ-Acenaphthene
DiQ-Acenaphthene
DjQ-Acenaphthene
DiQ-Acenaphthene
DjQ-Acenaphthene
DjQ-Acenaphthene
Relative
Retention
Time(a)
1.084
1.173
1.177
1,237
J.233
0.838
0.850
0.892
0.951
0.978
1.030
1.077
1.128
0.845
0.877
0.918
0.919
0.939
0.967
0.973
0.983
1.005
1.031
1.033
% RSD(b) of
Relative
Retention
Time
2.7
5.0
5.0
7.4
6.8
3.4
2.6
1.5
1.1
0.4
0.8
1.1
2.4
2.4
3.5
1.4
1.1
3.3
2.2
0.4
0.3
0.7
1.1
0.4
Response
Factor (c)
0.62
1.31
1.05
0.83
1.47
0.58
0.19
0.48
0.37
0.89
0.34
0.82
0.42
0.82
0.10
1.27
0.92
1.28
0.04
0.25
0.29
0.19
0.31
0.49
% RSD(b) of
Response
Factor
26
20
25
67
25
23
19
28
29
25
51
11
38
88
27
29
39
29
34
22
16
62
112
27
(continued)
-------
TABLE 6. (Continued)
Cpd.
No. Compound
Internal Standard
% RSD(°) of
Relative Relative
Retention Retention
Time(a) Time
Factorvc)
% RSD(b) of
Response
Factor
en
CA>
98.
100.
117.
118.
120.
121.
123.
125.
126.
132.
137.
2-Chloro-4-m"troani 1 i ne
Hexachlorobenzene
Anthraquinone
Fluoranthene
2-Methylanthraquinone
Pyrene
4,4'-ODD
4,4'-DDT
Triphenyl phosphate
Tri-(p-tolylJphosphate
Dibenzocarbazole
Djo-Phenanthrene
Djg-Phenanthrene
DjQ-Pyrene
Djg-Pyrene
DjQ-Pyrene
D^-Chrysene
D^-Chrysene
Dj2-Benzo(a)pyrene
)py rene
0.953
0.963
0.940
0.979
0.997
1.001
0.934
0.962
0.976
0.955
1.106
0.3
0.3
0.6
0.2
0.1
0.1
0.4
0.2
0.2
1.3
4.9
0.12
0.26
0.30
0.94
0.24
0.94
0.61
0.42
0.21
0.21
0.42
11
26
11
32
21
33
47
52
39
37
34
(*)Average of all of the relative retention times, relative to the internal standard, reported by the
participating laboratories for the determination of the compound in ILS-2.
(b)Percent relative standard deviation.
(OAverage of all of the response factors, relative to the internal standard, reported by the
participating laboratories for the determination of the compound in ILS-2.
-------
en
TABLE 7. VARIABILITY OF RETENTION TIMES AND RESPONSE FACTORS FOR THE
DETERMINATION OF VOLATILE COMPOUNDS
Cpd.
No.
12.
24.
25.
33.
44.
46.
47.
50.
51.
Compound
Propionitrile
2-Chloroacrylonitrile
1,1,1-Trichloroethane
1 ,2-Di chl oropropane
Bromoform
2-Hexanone
1 ,1 ,2 ,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
Retention
Time (a)
sec
551
760
787
952
1170
1309
1308
1472
1618
% RSD(°) of
Retention
Time
32
22
20
19
17
15
14
13
14
Response
Fact or vc)
0.15(DCE)
2.31(DCE)
6.90(DCE)
0.30(Bz)
0.24(Bz)
0.06(EBz)
0.27(EBz)
0.58(EBz)
0.88(EBz)
% RSD(b) of
Response
Factor
74
67
75
33
80
71
47
15
20
(a)Average of all of the retention times reported by the participating laboratories for the determination
of the compound in ILS-8.
(b)percent relative standard deviation.
(c)Average of all of the response factors, relative to the internal standard given in parentheses,
reported by the participating laboratories for the determination of the compound in ILS-8.
DCe=D4-l,2-dichlo ethane; Bz=Ds-benzene; and EBz=D5-ethylbenzene.
-------
TABLE 8. RECOVERY VARIABILITY OF VOLATILE COMPOUNDS FROM SAMPLE ILS-2
Cpd.
No.
12.
24.
25.
33.
44.
46.
47.
50.
51.
£ 36.
01 48.
Compound
Propionltrlle
2-Chloroacrylonitrile
1,1,1-Trichloroethane
1,2-Dichloropropane
Bromoform
2-Hexanone
1,1,2,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
Trichloroethylene
Tet rachl oroethy 1 ene
53. o-Xylene
Amount
Spiked,
u9/9
200
1000
800
200
1000
800
200
200
1200
0
0
0
Amount
Percent Relative Standard
Deviation of Amount-Found
Found, (a) Percent
ug/g
(b)
431
728
190
1170
680
179
202
1740
25
158
1130
Recovery (a)
(b)
43
91
95
117
85
90
101
145
(c)
(c)
(c)
Total
(b)
34
31
25
77
31
34
23
19
93
72
76
Between Lab. Within Lab.
Component
(b)
28
23
18
69
25
26
17
6
88
69
73
================
Component
(b)
20
20
17
34
19
22
16
18
30
22
19
Percent of
Amount -Found
Data
' Reported (d)
11
100
100
100
100
89
89
96
100
56
100
89
(a) Average of the values reported by the
(maximum of 3 per laboratory) in which
participating laboratories for all analytical runs
the compound was detected and quantified.
(b) Not determined because amount-found data were reported for fewer than 20 percent of the analytical runs.
(c) Not applicable because the compound was not used as a spike compound.
(d) The maximum number of analytical runs for which data could be reported in the study was 27
(replicates x 9 labs).
The percent of amount-found data reported =
total number of analytical runs for which data were reported x 100
27
-------
TABLE 9. RECOVERY VARIABILITY OF VOLATILE COMPOUNDS FROM SAMPLE ILS-3
en
Cpd.
No.
12.
24.
25.
33.
44.
46.
47.
50.
51.
1.
16.
17.
19.
Compound
Propionitrile
2-Chloroacrylonitrile
1 ,1 ,1-Trichloroethane
1 ,2-Dichloropropane
Bromoform
2-Hexanone
1,1,2,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
Chloromethane
trans, 1,2-Dichloroethylene
1,1-Dichloroethane
Amount
Spiked,
yg/g
500
2000
500
3000
2000
500
500
500
2500
0
0
0
Amount
Found,
ng/g
(b)
1570
5730
2090
1340
353
450
511
1660
486
870
3940
Chloroform 0 260
Percent Relative Standard
Deviation of Amount-Found
(a) Percent
Recovery (a)
(b)
78
1150
70
67
71
90
102
67
(c)
(c)
(c)
Between Lab.
Total
(b)
104
67
48
51
56
40
49
38
161
70
70
Component
(b)
30
54
42
39
31
27
45
33
154
69
65
Within Lab.
Component
(b)
100
39
23
33
47
29
19
19
47
14
27
(c) 70 65 25
Percent of
Amount -Found
Data
1 Reported (d)
0
96
93
96
85
52
93
96
96
63
74
96
74
(a) Average of the values reported by the participating laboratories for all analytical runs
(maximum of 3 per laboratory) in which the compound was detected and quantified.
(b) Not determined because amount-found data were reported for fewer than 20 percent of the analytical runs.
(c) Not applicable because the compound was not used as a spike compound.
(d) The maximum number of analytical runs for which data could be reported in the study was 27
(replicates x 9 labs).
The percent of amount-found data reported * .
total number of analytical runs for which data were reported x 100
27
-------
TABLE 10. RECOVERY VARIABILITY OF VOLATILE COMPOUNDS FROM SAMPLE ILS-4
Cpd.
No. Compound
Amount
Spiked,
Amount
Found, (a) Percent
ug/g Recovery (a)
Percent Relative Standard
Deviation of Amount-Found
Between Lab.
Total Component
Within Lab.
Component
Percent of
Amount-Found
Data
1 Reported(d)
en
12. Propionitrile 500 1110
24. 2-Chloroacrylonitrile 100 (b)
25. 1,1,1-Trichloroethane 100 73
33. 1,2-Dichloropropane 400 304
44. Bromoform 100 742
46. 2-Hexanone 100 825
47. 1,1,2,2-Tetrachloroethane 500 385
50. Chlorobenzene 600 492
51. Ethylbenzene 100 90
19. Chloroform 0 141
36. Trichloroethylene 0 168
40. Dibromoethanethane 0 46
48. Tetrachloroethylene 0 2630
49. Toluene ; 0 95
52. Styrene 0 113
53. o-Xylene i 0 111
223(e)
(b)
73
76
742(e)
825(e)
77
82
90(c)
(c)
(c)
(c)
(c)
(c)
(c)
(c)
133
(b)
86
33
204
178
68
73
63
317
187
39
143
87
65
70
127
(b)
72
27
204
171
64
66
52
64
187
23
137
79
60
60
41
(b)
47
19
16
49
22
32
35
310
12
32
42
34
27
37
22
4
78
100
56
37
67
100
89
100
52
63
100
89
78
78
(a)
(b)
Average of the values reported by the participating laboratories for all analytical
(maximum of 3 per laboratory) in which the compound was detected and quantified.
Not determined because amount-found data were reported for fewer than 20 percent of
Not applicable because the compound was not used as a spike compound.
The maximum number of analytical runs for which data could be reported in the study
(replicates x 9 labs).
The percent of amount-found data reported =
total number of analytical runs for which data Were reported x 100
27
runs
the analytical runs.
was 27
-------
TABLE 11. RECOVERY VARIABILITY OF VOLATILE COMPOUNDS ROM SAMPLE ILS-5
Ul
Cpd.
No.
12.
24.
25.
33.
44.
46.
47.
50.
51.
5.
19.
(a)"
Compound
Propionltrile
2-Chloroacrylonitrile
1,1,1-Trichloroethane
1,2-Dichloropropane
Bromoform
2-Hexanone
1,1,2,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
Methyl ene chloride
Chloroform
Average of the values repor
Amount
Spiked,
ug/g
25
5
5
20
5
5
25
30
5
0
0
ted by the
Amount
Percent Relative Standard Percent of
Deviation of Amount-Found Amount-Found
Found, (a) Percent
pg/g
(b)
(b)
(b)
4
3
(b)
13
10
3
124
4
Recovery (a)
(b)
(b)
(b)
18
48
(b)
51
34
52
(c)
(c)
Between Lab.
Total
(b)
(b)
(b)
53
77
(b)
51
63
30
266
219
participating laboratories for
Component
(b)
(b)
(b)
51
75
(b)
47
63
29
189
Within Lab.
Component
(b)
(b)
(b)
17
17
(b)
18
7
4
187
Data
Reported (d)
11
11
11
56
67
11
100
89
41
78
88 200 56
all analytical runs
(maximum of 3 per laboratory) in which the compound was detected and quantified.
(b) Not determined because amount-found data were reported for fewer than 20 percent of the analytical runs.
(c) Not applicable because the compound was not used as a spike compound.
(d) The maximum number of analytical runs for which data could be reported in the study was 27
(replicates x 9 labs).
The percent of amount-found data reported =
total number of analytical runs for which data were reported x 100
27
-------
TABLE 12. RECOVERY VARIABILITY OF VOLATILE COMPOUNDS FROM SAMPLE ILS-6
Ol
VO
Cpd.
No.
12.
24.
25.
33.
44.
46.
47.
50.
51.
5.
17.
38.
48.
53.
Compound
Propionitrile
2-Chloroacrylonitrile
1 , 1 , 1-Tri chl oroethane
1 ,2-Di chl oropropane
Bromoform
2-Hexanone
1,1,2,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
Methyl ene chloride
1 , 1-Di chl oroethane
1,1,2-Tri chl oroethane
Tetrachloroethylene
o-Xylene t
=============================
Amount Amount
Spiked, Found, (a) Percent
iig/g ug/g Recovery (a)
5
30
25
5
30
25
5
20
5
0
0
0
0
0
(b)
(b)
(b)
1
(b)
11
(b)
11
4
27
1
5
2
2
(b)
(b)
(b)
26
(b)
44
(b)
56
76
(c)
(c)
(c)
(c)
(c)
Percent Relative Standard
Deviation of Amount-Found
Total
(b)
(b)
(b)
22
(b)
124
(b)
47
35
307
59
34
27
75
Between Lab. Within Lab.
Component Component
(b)
(b)
(b)
21
(b)
109
(b)
44
15
19
56
25
26
55
i=======— =—=====
(b)
(b)
(b)
8
(b)
59
(b)
18
32
307
17
23
8
52
Percent of
Amount -Found
Data
' Reported (d)
0
4
7
48
4
26
4
100
89
63
52
78
67
67
(a) Average of the values reported by the participating laboratories for all analytical runs
(maximum of 3 per laboratory) in which the compound was detected and quantified.
(b) Not determined because amount-found data were reported for fewer than 20 percent of the analytical runs.
(c) Not applicable because the compound was not used as a spike compound.
(d) The maximum number of analytical runs for which data could be reported in the study was 27
(replicates x 9 labs).
The percent of amount-found data reported =
total number of analytical runs for which data were reported x 100
27
-------
O»
o
TABLE 13. RECOVERY VARIABILITY OF VOLATILE COMPOUNDS_FROM_SAMPLE_ILS-7=
Cpd.
No.
Compound
Amount Amount
Spiked, Found,(a) Percent
ug/g Recovery(a)
Percent Relative Standard Percent of
ppviat™" nf Amount-Found Amount-Found
Between Lab.Within Lab. Data
Total Component Component ' Reported(d)
12.
24.
25.
33.
44.
46.
47.
50.
51.
6.
36.
Propionitrile
2-Chloroacrylonitrile
1,1,1-Trichloroethane
1,2-Dichloropropane
Bromoform
2-Hexanone
1,1,2,2-Tetrachlo
Chlorobenzene
Ethyl benzene
Acetone
Trichloroethylene
600
rile 10°
hane 10°
lid lib
ne 500
lie
100
100
rnpthane 600
roetnane ^
400
0
0
(b)
(b)
28
250
30
102
212
49
230
10800
87
:==========:
(b)
(b)
28
50
30
102
35
49
57
(c)
(c)
(b)
(b)
68
38
96
119
44
45
40
92
56
(b)
(b)
65
37
92
71
36
43
38
91
55
(b)
(b)
20
9
24
95
26
12
10
13
14
11
4
78
100
56
52
100
100
89
78
100
(b)
(c)
(d)
or me va,u« , cp by the participating laboratories for all analytical runs
of 3 per laboratory) in which the compound was detected and quantITiea.
Not determined because amount-found data were reported for fewer than 20 percent of the analytical runs.
Not applicable because the compound was not used as a spike compound.
The maximum number of analytical runs for which data could be reported in the study was 27
(replicates x 9 labs).
The percent of amount-found data reported =
number of analytical runs for which data were reported x 100
-------
TABLE 14. RECOVERY VARIABILITY OF VOLATILE COMPOUNDS ROM SAMPLE ILS-8
CTl
Cpd
No.
12.
24.
25.
33.
44.
46.
47.
50.
51.
15.
16.
17.
19.
26.
36.
•
Compound
Proplonitrile
2-Chloroacrylonitrile
1,1,1-Trichloroethane
1,2-Dichloropropane
Bromoform
2-Hexanone
1,1,2,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
1,1-Dichloroethane
t rans-1 ,2-Di chl oroethylene
1,1-Dichloroethane
Chloroform
Carbon tetrachloride
Tri chl oroethylene
ss==s==:s
Amount
Spiked
ug/9
16000
4000
24000
4000
4000
24000
16000
20000
4000
0
0
0
0
0
0
=======
Amount
Percent Relative Standard
Deviation of Amount-Found
, Found, (a) Percent
ng/g
17000
5020
24300
2960
3290
20900
12300
18100
3760
7290
13200
3760
1770
520000
Recovery (a)
106
126
101
74
82
87
77
91
94
(c)
(c)
(c)
(c)
(c)
136000 (c)
Between Lab. Within Lab.
Total
28
50
46
47
60
53
49
56
61
176
84
52
49
66
65
Component
27
48
43
45
58
48
48
53
57
174
82
49
46
63
63
Component
8
13
16
16
16
24
10
18
21
28
19
18
17
19
16
Percent of
Amount -Found
Data
' Reported (d)
26
78
89
78
89
89
100
100
89
89
100
78
100
100
100
(a) Average of the values reported by the participating laboratories for all analytical runs
(maximum of 3 per laboratory) in which the compound was detected and quantified.
(b) Not determined because amount-found data were reported for fewer than 20 percent of the analytical runs,
(c) Not applicable because the compound was not used as a spike compound.
(d) The maximum number of analytical runs for which data could be reported in the study was 27
(replicates x 9 labs).
The percent of amount-found data reported =
total number of analytical runs for which data were reported x 100
27
-------
TABLE 15. SAMPLE-TO-SAMPLE RECOVERY DEPENDENCIES-VGA
=====================================================
Sample Average Percent Recovery
ILS-2
ILS-3
ILS-5
ILS-8
96
78
41
93
40-150
67-102
13-52
74-126
=====================================================
Excludes obvious outliers
volatile determinations. For example, for ILS-3, the number of compounds
reported varied from 1 to 27 with 20 being the average number of compounds
reported. The total relative standard deviation generally ranged from 30 to 80
percent. The relative standard deviation for the within-laboratory component
was less than 30 percent and the between-laboratory variability was about twice
that value. The ranges of both components were the same as that reported for
volatile results, namely 5 to 300 percent. No difference 1n range of values
was apparent for the spiked versus non-spiked compounds, however, some of the
higher recoveries may be attributable to differential extraction efficiencies
between analytes and Internal standards and to background contaminants such as
phthalates.
The results for ILS-9, the National Bureau of Standards SRM, presented in
Table 21 were surprisingly variable. The total variabilities ranged from 28 to
338 percent and averaged 160 percent. The quantities of compounds found were
relatively lower in many of the samples, however, the quantities of compounds
found in ILS-5 and ILS-6 were comparable. The degree of homogeneity of ILS-9
should be better than the other program samples since NBS prepared this sample.
An Inspection of the data reported by each individual laboratory revealed
that most of the variability could be attributed to outlier values reported by
one laboratory. In some cases a significant portion of the variability was
also caused by outlier values reported by a second laboratory. The improvement
1n the total variabilities obtained by omitting the data from one laboratory or
two laboratories 1n the calculation are shown in Table 23. By omitting the
data from two laboratories the total variabilities ranged from only 13 to 90
percent and averaged 31 percent. An average total variability of 31 percent
was considered excellent. The within-laboratory component of the variability
would be much less.
The above treatment could undoubtedly be applied to the data from other
samples and provide significant improvements 1n the precision. A very impor-
tant next step would be to investigate the supplemental data in detail to
determine the causes of other outliers. The information on the probable causes
of outliers would be used to modify the method instructions and quality control
protocols to reduce the number of outliers generated in any future application
of the methods.
162
-------
TABLE 16. RECOVERY VARIABILITY OF SEMIVOLATILE COMPOUNDS FROM SAMPLE ILS-2
co
Cpd.
No.
6.
12.
13.
15.
16.
22.
25.
28.
33.
36.
43.
50.
51.
56.
57.
64.
65.
66.
69.
74.
76.
77.
80.
84.
85.
98.
Amount
Spiked,
Compound pg/g
4-Chl orotoluene
Bi s (2-chl oroethy 1 Jether
2-Chlorophenol
2,4,6-Trimethylpyridine
1, 4-Di chl orobenzene
Acetophenone
Hexachloroethane
4-t-Butylpyridi ne
2, 4-Di methyl phenol
Propiophenone
4-Chl oroani line
Qu incline
Bis( 2-chl oroetho^y) ethane
4-Chl oro-2 -methyl ani 1 i ne
4-Chl orobenzoi c acid
1-Chloronaphthalene
4-Methylquinoline
2-Ethyl naphthal ene
4-Bromobenzoic acid
1 ,3-Di ni t robenzene
2,6-Dinitrotoluene
3-Nitroaniline
2,4-Dinitrophenol
4-Nitrophenol
Pentachl orobenzene
2-Chl oro-4-nit roani 1 i ne
600
500
600
600
100
100
500
100
100
400
600
500
100
100
500
100
100
500
100
500
100
500
400
100
100
100
Amount
Percent Relative Standard
Deviation of Amount-Found
Found, (a) Percent Between Lab.
jig/g Recovery (a) Total Component
363
310
358
692
36
103
257
47
71
254
337
313
62
61
393
63
88
373
52
199
64
406
(b)
54
71
62
61
62
60
115
36
103
51
47
71
64
56
63
62
61
79
63
88
74
52
40
64
81
(b)
54
71
62
63
62
52
91
63
47
55
53
34
46
257
38
(b)
73
(b)
51
76
51
67
49
37
50
(b)
(b)
43
42
60
60
47
87
62
41
29
48
20
41
134
27
(b)
66
(b)
47
67
41
64
44
32
44
(b)
(b)
32
35
Within Lab.
Component
17
17
21
25
12
23
47
22
28
20
219
27
(b)
33
(b)
20
35
30
19
22
18
25
(b)
(b)
28
23
Percent of
Amount -Found
Data
' Reported (d)
96
100
100
59
70
89
89
44
56
89
74
100
56
41
33
33
63
100
22
63
44
44
0
11
74
22
(continued)
-------
TABLE 16. (Continued)
Cpd.
No. Compound
Amount
Spiked,
yg/g
Amount
Found, (a) Percent
yg/g Recovery (a)
Percent Relative Standard
Deviation of Amount-Found
Between Lab.
Total Component
Within Lab.
Component
Percent of
Amount-Found
Data
Reported (d)
100. Hexachlorobenzene 600 460
117. Anthraquinone 600 403
118. Fluoranthene 100 71
120. 2-Methylanthraquinone 100 71
121. Pyrene 400 306
123. 4,4'-ODD 100 67
125. 4,4'-DDT 400 225
126. Triphenyl phosphate 600 524
132. Tri-(p-tolyl) phosphate 100 57
137. Dibenzocarbazole 100 109
29. 1,2,4,5-Tetramethylbenzene 0 25
41. Naphthalene 0 174
114. Di-n-butyl phthalate 0 80
77
67
71
71
77
67
56
87
57
109
(c)
(c)
(c)
43
57
41
55
40
77
68
87
64
70
69
231
675
38
49
36
49
36
14
37
58
56
58
38
231
165
21
28
20
26
18
75
57
65
30
39
57
14
82
100
100
100
78
100
78
85
89
63
56
63
78
56
(a) Average of the values reported by the participating laboratories for all analytical runs
(maximum of 3 per laboratory) in which the compound was detected and quantified.
(b) Not determined because amount-found data were reported for fewer than 20 percent of the analytical runs.
(c) Not applicable because the compound was not used as a spike compound.
r
(d) The maximum number of analytical runs for which data could be reported in the study was 27
(replicates x 9 labs).
The percent of amount-found data reported =
total number of analytical runs for which data were reported x 100
27
-------
TABLE 17. RECOVERY VARIABILITY OF SEMIVOLATILE COMPOUNDS FROM SAMPLE ILS-3
en
Cpd.
No.
6.
12.
13.
15.
16.
22.
25.
28.
33.
36.
43.
50.
51.
56.
57.
64.
65.
66.
69.
74.
76.
77.
80.
84.
85.
98.
Compound
4-Chlorotoluene
Bis(2-chloroethyl )ether
2-Chlorophenol
2,4 ,6-Trimethylpyridl ne
1 ,4-DI chl orobenzene
Acetophenone
Hexachl oroethane
4-t -Butyl py ri di ne
2, 4-Dimethyl phenol
Propiophenone
4-Chloroanlline
Quinoline
Bis(2-chloroetho*y) ethane
4-Chl oro-2-methyl ani 1 i ne
4-Chlorobenzolc acid
1-Chloronaphthalene
4-Methylquinoline
2-Ethyl naphthalene
4-Bromobenzoic acid
1 ,3-Di ni trobenzene
2,6-Dinitrotoluene
3-Nitroaniline
2,4-Di nitrophenol
4-Nitrophenol
Pentachl orobenzene
2-Chloro-4-nitroani 1 i ne
Amount
Spiked,
wg/g
800
200
800
800
200
1000
200
200
200
200
800
200
1200
200
200
1200
1200
200
1200
200
1200
200
200
1000
200
1200
Amount
Found, (a) Percent
ug/g Recovery (a
877
166
158
2460
169
543
243
92
(b)
181
(b)
281
1230
86
(b)
(b)
1790
IS
652
1670
(b)
(b)
(b)
279
467
110
83
20
307
84
54
122
46
(b)
90
(b)
141
103
43
(b)
(b)
149
ffl
326
139
(b)
(b)
(b)
71
39
Percent Relative Standard
Deviation of Amount-Found
Between Lab.
) Total Component
96
49
27
136
103
88
160
44
(b)
185
(b)
169
138
83
(b)
(b)
134
ft
111
152
(b)
(b)
(b)
114
227
80
48
19
131
86
51
159
37
(b)
143
(b)
137
103
83
(b)
(b)
104
(b)
(b)
111
109
(b)
(b)
(b)
91
226
Within Lab.
Component
54
6
20
36
57
71
17
23
(b)
117
(b)
99
91
10
(b)
(b)
85
ft
7
105
(b)
(b)
(b)
68
13
Percent of
Amount -Found
Data
' Reported (d)
89
67
67
78
89
89
85
56
4
78
11
78
100
22
0
0
67
11
11
19
56
7
0
11
78
15
(continued)
-------
TABLE 17. (Continued)
en
Cpd.
No.
100.
117.
118.
120.
121.
123.
125.
126.
132.
137.
Compound
Hexachl orobenzene
Anthraquinone
Fluoranthene
2 -Methyl anthraqui none
Pyrene
4, 4 '-ODD
4, 4 '-DDT
Triphenyl phosphate
Tri-(p-tolyl) phosphate
Dibenzocarbazole
Amount
Spiked,
yg/g
800
800
1000
200
200
1000
200
800
200
1200
Amount
Percent Relative Standard
Deviation of Amount-Found
Found, (a) Percent
wg/g
1170
1240
907
352
229
1550
508
1180
339
263
Recovery (a)
146
155
90
176
115
155
254
147
170
22
Total
123
128
92
100
188
118
211
137
126
100
Between Lab. Within Lab.
Component
103
109
75
87
96
97
211
110
107
78
Component
68
67
55
49
162
69
13
81
67
62
Percent of
Amount-Found
Data
• Reported(d)
89
78
89
67
22
89
74
89
41
11
(a) Average of the values reported by the participating laboratories for all analytical runs
(maximum of 3 per laboratory) in which the compound was detected and quantified.
(b) Not determined because amount-found data were reported for fewer than 20 percent of the analytical runs.
(c) Not applicable because the compound was not used as a spike compound.
(d) The maximum number of analytical runs for which data could be reported in the study was 27
(replicates x 9 labs).
The percent of amount-found data reported =
total number of analytical runs for which data were reported x 100
27
-------
TABLE 18. RECOVERY VARIABILITY OF SEMIVOLATILE COMPOUNDS FROM SAMPLE ILS-4
Cpd.
No.
6.
12.
13.
15.
16.
22.
25.
28.
33.
36.
43.
50.
51.
56.
57.
64.
65.
66.
69.
74.
76.
77.
80.
84.
85.
98.
Compound
4-Chlorotoluene
Bis(2-chloroethyl )ether
2-Chlorophenol
2,4, 6-Trimethylpyri dine
1 ,4-DI chl orobenzene
Acetophenone
Hexachloroethane
4-t-Butylpyrldine
2, 4-Di methyl phenol
Propiophenone
4-Chloroaniline
Qu incline
Bis(2-chloroetho\y) ethane
4-Chloro-2-methyl aniline
4-Chlorobenzoic acid
1-Chloronaphthalene
4-Methylqui nol i ne
2-Ethyl naphthal ene
4-Bromobenzoic acid
1 ,3-Di ni trobenzene
2,6-Dinitrotoluene
3-Nitroaniline
2,4-Dinitrophenol
4-Nitrophenol
Pentachl orobenzene
2-Chloro-4-nitroani 1 i ne
Amount
Spiked,
vg/g
1900
11400
1900
1900
7600
1900
11400
7600
7600
9500
1900
11400
1900
7600
11400
1900
1900
11400
1900
11400
1900
11400
9500
1900
7600
1900
Amount
Found, (a)
yg/g
1490
8500
2310
2270
4950
1880
8980
6790
8840
9610
2420
2680
1870
6920
15900
1260
2430
1410
1210
12900
2280
14200
2010
(b)
7090
2400
Percent Relative Standard
Deviation of Amount-Found
Percent Between Lab.
Recovery (a) Total Component
78
75
122
119
65
99
79
89
104
101
127
24
98
91
139
66
128
12
64
113
120
125
21
(b)
93
127
65
86
120
29
47
82
91
61
83
74
63
72
86
60
154
32
91
100
40
105
105
95
93
(b)
74
125
50
84
115
23
42
78
86
57
80
73
60
68
81
54
139
27
86
95
34
103
103
93
89
(b)
72
121
Within Lab.
Component
41
20
32
18
22
25
30
23
23
17
17
23
30
24
67
17
30
32
22
18
21
23
27
(b)
19
29
Percent of
Amount-Found
Data
' Reported (d)
89
96
89
37
100
89
100
89
89
89
56
89
89
78
78
44
89
89
22
78
78
78
56
11
100
59
(continued)
-------
TABLE 18. (Continued)
en
oo
Cpd.
No.
100.
117.
118.
120.
121.
123.
125.
126.
132.
137.
139.
11.
14.
20.
23.
41.
54.
63.
72.
73.
78.
83.
90.
Amount
Spiked,
Compound pg/g
Hexachl orobenzene
Anthraquinone
Fluoranthene
2-Methyl anthraqui none
Pyrene
4, 4 '-ODD
4 ,4 '-DDT
Triphenyl phosphate
Tri-(p-tolyl) phosphate
Dibenzocarbazole
Dibenzo( a, h) anthracene
Phenol
1 ,2 ,4-Trimethyl benzene
2-Methyl phenol
4-Methyl phenol
Naphthalene
2-Methyl anthraqui none
Bi phenyl
Acenaphthylene
2 ,3-Dimethyl naphthal ene
Acenaphthene
4-Aminobi phenyl
4-Chlorophenyl phenyl ether
1900
1900
1900
7600
9500
1900
9500
1900
7600
1900
0
0
0
0
0
0
0
0
0
0
0
0
0
Amount
Found, (a)
wg/g
2060
2180
14400
8420
17400
2120
9270
2230
10530
(b)
880
3810
370
1040
3220
26400
16600
1730
62000
850
720
5530
8760
Percent Relative Standard
Deviation of Amount-Found
Percent
Recovery(a)
108
115
87(e)
111
61(f)
111
98
117
139
(b)
(c)
(c)
(c)
(c)
(c)
(c)
(c)
(c)
(c)
(c)
(c)
(c)
(c)
Between Lab.
Total Component
87
94
59
92
54
88
108
95
92
(b)
82
78
58
78
79
31
191
79
98
77
75
74
67
82
91
39
86
45
81
106
92
85
(b)
45
59
52
77
70
26
189
76
97
72
71
72
64
Within Lab.
Component
28
27
44
32
30
35
22
24
35
(b)
69
51
27
11
37
17
28
21
10
27
22
16
19
Percent of
Amount-Found
Data
Reported(d)
89
89
100
89
100
89
89
78
89
11
67
89
67
74
89
100
89
78
89
78
78 /
100
100
(continued)
-------
TABLE 18. (Continued)
Percent Relative Standard Percent of
Amount Amount Deviation of Amount-Found Amount-Found
Cpd.
No.
107.
109.
110.
111.
112.
Compound
Phenanthrene
Anthracene
Acridine
Phenanthridine
Carbazole
Spiked,
ug/g
0
0
0
0
0
Found, (a) Percent
yg/g
19100
6910
650
720
3520
Recovery (a)
(c)
(c)
(c)
(c)
(c)
Between Lab.
Total
54
60
77
93
87
Component
49
53
74
91
84
Within Lab.
Component
23
28
21
21
22
Data
' Reported (d)
100
100
67
78
89
10
(a) Average of the values reported by the participating laboratories for all analytical runs
(maximum of 3 per laboratory) in which the compound was detected and quantified.
(b) Not determined because amount-found data were reported for fewer than 20 percent of the analytical runs.
(c) Not applicable because the compound was not used as a spike compound.
(d) The maximum number of analytical runs for which data could be reported in the study was 27
(replicates x 9 labs).
The percent of amount-found data reported =
total number of analytical runs for which data were reported x 100
27
(e) Corrected for the 14,600 pg/g present in the unspiked sample.
(f) Corrected for the 19,000 pg/g present in the unspiked sample.
-------
TABLE 19. RECOVERY VARIABILITY OF SEMIVOLATILE COMPOUNDS FROM SAMPLE ILS-5
Cpd.
No.
6.
12.
13.
15.
16.
22.
25.
28.
33.
36.
43.
50.
51.
56.
57.
64.
65.
66.
69.
74.
76.
77.
80.
84.
85.
98.
Amount Amount
Spiked, Found, (a) Percent
Compound yg/g yg/g Recovery (a
4-Chlorotoluene
Bis(2-chloroethyl )ether
2-Chlorophenol
2 ,4 ,6-Trimethyl py ridi ne
1,4-Dichlorobenzene
Acetophenone
Hexachloroethane
4-t-Butylpyridine
2, 4-Di methyl phenol
Propiophenone
4-Chloroanillne
Qu incline
Bis(2-chloroetho*y) ethane
4-Chloro-2-methyl ani 1 i ne
4-Chlorobenzoic acid
1-Chl oronaphthal ene
4-Methylquinoline
2-Ethyl naphtha! ene
4-Bromobenzoic acid
1,3-Di nitrobenzene
2,6-Dinitrotoluene
3-Nitroaniline
2,4-Dinitrophenol
4-Nitrophenol
Pentachl orobenzene
2-Chloro-4-nitroani 1 i ne
4
24
4
4
16
4
24
16
16
20
4
24
4
16
24
4
4
24
4
24
4
24
20
4
16
4
(b)
7.2
(b)
5.3
2
2
2
2
(b)
9
(b)
7
2
(b)
7
2
1
11
(b)
11
2
(b)
2
(b)
9
(b)
(b)
30
(b)
131
8
26
7
15
(b)
43
(b)
29
48
(b)
27
49
26
47
(b)
46
54
(b)
11
(b)
53
(b)
Percent Relative Standard
Deviation of Amount -Found
Between Lab.
) Total Component
(b)
51
(b)
145
48
75
62
87
(b)
51
(b)
48
68
(b)
65
43
125
47
(b)
78
68
(b)
76
(b)
53
(b)
(b)
50
(b)
118
43
67
54
82
(b)
48
(b)
38
66
(b)
49
18
49
44
(b)
74
55
(b)
71
(b)
52
(b)
WfthinTaDT
Component
(b)
14
(b)
85
22
32
29
31
(b)
18
(b)
29
17
(b)
42
40
114
16
(b)
25
40
(b)
29
(b)
11
(b)
Percent of
Amount-Found
Data
Reported (d)
11
78
0
22
78
67
78
67
0
89
0
89
56
0
44
44
67
89
0 ,
78
52
0
44
11
78
11
(continued)
-------
TABLE 19. (Continued)
Percent Relative Standard
Amount Amount Deviation of Amount -Found
Cpd.
No.
100.
117.
118.
120.
121.
123.
125.
126.
132.
137.
39.
63.
67.
114.
Compound
Hexachl orobenzene
Anthraquinone
Fluoranthene
2 -Methyl anthraqui none
Pyrene
4, 4 '-ODD
4, 4' -DDT
Triphenyl phosphate
Tri-(p-tolyl) phosphate
Dibenzocarbazole
2,6-Dichlorophenol
Biphenyl
Phenyl ether ,
Di-n-butyl phthalate
Spiked,
ug/g
4
4
4
16
20
4
20
4
16
4
0
0
0
0
Found, (a) Percent
ug/g
2
2
3
7
9
3
10
2
10
(b)
1
1
1
7
Recovery (a)
50
48
63
46
44
70
49
60
59
(b)
(c)
!c)
c)
(c)
Between Lab.
Total
58
80
84
64
56
68
76
44
89
(b)
48
73
86
137
Component
55
69
55
61
53
60
59
39
86
(b)
37
40
52
122
Within Lab.
Component
18
38
64
19
19
33
47
20
23
(b)
30
61
69
62
Percent of
Amount-Found
Data
Reported (d)
78
81
93
78
100
81
89
59
70
0
56
78
67
56
(a) Average of the values reported by the participating laboratories for all analytical runs
(maximum of 3 per laboratory) in which the compound was detected and quantified.
(b) Not determined because amount-found data were reported for fewer than 20 percent of the analytical runs.
(c) Not applicable because the compound was not used as a spike compound.
(d) The maximum number of analytical runs for which data could be reported in the study was 27
(replicates x 9 labs).
The percent of amount-found data reported =
total number of analytical runs for which data were reported x 100
27
-------
TABLE 20. RECOVERY VARIABILITY OF SEMIVOLATILE COMPOUNDS FROM SAMPLE ILS-6
ro
Amount Amount
Cpd. Spiked, Found, (a)
No. Compound pg/g yg/g
6.
12.
13.
15.
16.
22.
25.
28.
33.
36.
43.
50.
51.
56.
57.
64.
65.
66.
69.
74.
76.
77.
80.
84.
85.
98.
4-ChTorotoluene
B1s(2-chloroethyl )ether
2-Chlorophenol
2, 4,6-Tri methyl pyridine
1 ,4-Di chl orobenzene
Acetophenone
Hexachloroethane
4-t -Butyl pyridine
2, 4-Dimethyl phenol
Propiophenone
4-Chloroaniline
Qu incline
Bis(2-chloroetho*y) ethane
4-Chl oro-2-methy 1 ani 1 i ne
4-Chlorobenzoic acid
1-Chl oronaphthal ene
4-Methylquinoline
2-Ethyl naphthalene
4-Bromobenzoi c acid
1 ,3-Di ni t robenzene
2 ,6-Di ni trotol uene
3-Nitroaniline
2,4-Dinitrophenol
4-Nitrophenol
Pentachl orobenzene
2-Chloro-4-nitroani 1 i ne
20
20
20
20
100
20
20
100
100
120
20
20
80
100
20
80
80
20
80
20
80
20
120
20
100
80
5
7
10
(b)
29
5
(b)
45
50
25
(b)
5
31
40
20
41
38
11
28
ffl
(b)
(b)
(b)
52
(b)
Percent Relative Standard
Deviation of Amount-Found
Percent Between Lab.
Recovery (a) Total Component
26
33
51
(b)
29
24
(b)
45
50
21
(b)
23
38
40
100
52
48
56
35
Si
(b)
(b)
(b)
52
(b)
97
55
46
(b)
60
55
(b)
54
42
53
(b)
34
43
67
110
50
78
54
65
ft
(b)
(b)
(b)
48
(b)
95
50
41
(b)
54
50
(b)
50
39
46
(b)
23
37
65
38
47
77
49
51
ft!
(b)
(b)
(b)
42
(b)
Within Lab.
Component
20
23
20
(b)
26
24
(b)
22
18
27
(b)
25
22
19
103
17
13
23
41
81
(b)
(b)
(b)
23
(b)
Percent of
Amount -Found
Data
' Reported (d)
33
22
44
11
93
44
0
85
89
89
11
44
78
44
22
59
78
78
96
0
0
0
0
0
100
0
(continued)
-------
TABLE 20. (Continued)
Cpd.
No. Compound
Amount
Spiked,
ug/g
Amount
Found, (a) Percent
vg/g Recovery (a)
Percent Relative Standard
Deviation of Amount Found
Between Lab.
Total Component
Within Lab.
Component
Percent of
Amount Found
Data
' Reported (d)
100.
117.
118.
120.
121.
123.
125.
126.
132.
137.
11.
23.
40.
67.
107.
114.
131.
Hexachlorobenzene
Anthraquinone
Fluoranthene
2-Methylanthraqui none
Pyrene
4,4'-ODD
4,4'-DDT
Triphenyl phosphate
Tri-(p-tolyl) phosphate
Dibenzocarbazole
Phenol
4-Methylphenol
1,2,4-Tri chlorobenzene
Phenyl ether
Phenanthrene
Di-n-butyl phthalate
Di-(2-ethylhexyl) phthalate
20
20
20
100
120
20
120
20
100
80
0
0
0
0
0
0
0
13
7
14
36
76
32
(b)
15
68
72
18
29
39
5
3
21
30
63
35
59
36
61
162
(b)
76
68
89
(c)
III
(c)
(c)
(c)
(c)
46
58
45
57
52
61
(b)
99
46
78
33
40
49
82
54
135
75
37
55
41
51
51
54
(b)
93
41
75
25
36
43
73
34
94
69
27
19
18
25
10
27
(b)
33
22
18
21
17
25
37
42
97
29
44
33
100
89
100
89
0
33
85
52
67
81
89
52
56
78
100
(a) Average of the values reported by the participating laboratories for all analytical
(maximum of 3 per laboratory) in which the compound was detected and quantified.
(b) Not determined because amount-found data were reported for fewer than 20 percent of
(c) Not applicable because the compound was not used as a spike compound.
(d) The maximum number of analytical runs for which data could be reported in the study
(replicates x 9 labs).
The percent of amount-found data reported =
total number of analytical runs for which data were reported x 100
runs
the analytical runs.
was 27
-------
TABLE 21. RECOVERY VARIABILITY OF SEMIVOLATILE COMPOUNDS FROM SAMPLE ILS-6
Cpd.
No.
41.
72.
78.
90.
107.
109.
114.
117.
118.
121.
127.
130.
134.
136.
138.
139.
140.
Compound
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Di-n-butyl phthalate
Anthraquinone
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(k)fluoranthene
Benzo(a)pyrene
Indeno(l,2,3 cd)pyrene
Dibenzo(a,h)anthracene
Benzo(g,h,i jperylene
Amount
Spiked,
wg/g
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Amount
Found, (a)
pg/g
4
45
1
14
28
10
22
15
53
93
40
70
40
14
14
10
18
Percent Relative Standard
Deviation of Amount-Found
Between Lab. Within Lab.
Total Component Component
263
278
28
305
338
287
218
102
154
107
79
44
127
95
127
87
93
125
87
16
197
131
109
158
84
65
50
41
29
57
92
78
77
64
231
265
23
233
311
265
150
59
139
94
68
33
113
21
101
40
67
Percent of
Amount-Found
Data
' Reported (b)
75
75
63
59
75
75
63
71
100
78
100
100
88
88
88
88
83
(a) Average of the values reported by the participating laboratories for
(maximum of 3 per laboratory) in which the compound was detected and
all analytical runs
quantified.
(b) The maximum number of analytical runs for which data could be reported in the study was 27
(replicates x 9 labs).
The percent of amount-found data reported =
total number of analytical runs for which data were reported x 100
27
-------
TABLE 22. SAMPLE-TO-SAMPLE RECOVERY DEPENDENCIES-SV
====================================================
Sample Average Percent Recovery Range
ILS-5 68 30-115
ILS-4 95 20-140
ILS-5 44 7-131
====================================================
The percent recoveries of spiked volatile compounds reported by each
laboratory including the total number of spike compounds reported are given in
Tables 24 to 30. The corresponding data for the spiked semi volatile compounds
are given in Tables 31 to 35. These compilations of data are based on numerical
data reported; missing data are not included in the averages as zeros. These
data tables were prepared in this format to enable the reader to ascertain
specific laboratory results and to facilitate the identification of trends.
The total number of spike compounds reported by each laboratory is recorded for
purposes of comparison. The individual percent recovery entries for each
laboratory may be examined for variability from compound-to-compound, from
sample-to-sample, and of course, from laboratory-to-laboratory. These data are
particularly useful for identifying anomalous data that may skew the summary
statistical data presented in the previous tables. For example, for ILS-3 the
data in Table 25 show that all 9 laboratories consistently found larger
quantities of 1,1,1-trichloroethane than were spiked. However, only one
laboratory, Lab 5, reported excessive quantities of 2-chloroacrylonitrile and
2-hexanone in sample ILS-4 (Table 26). Examination of the raw data shows that
abnormally low response factors were used for the quantification; thus, the
high average values reported in Table 10 were explained.
The variability in the total number of volatile compounds reported is
presented in Table 36 in a matrix for samples versus laboratory. The totals of
these columns can be used as a guide to identify difficult samples and to
identify overall laboratory performance differences. The two samples for which
the poorest results were obtained were ILS-5 (oxychlorination spent catalyst)
and ILS-6 (POTW sludge). These two sample matrices would be expected to be the
least retentive of volatile compounds and amounts of the components may have
been lost during sample storage prior to analysis. Excellent results were
obtained with ILS-2, ILS-3 and ILS-8, while ILS-4 and ILS-7 provided intermediate
results. The differences in laboratory performance are more difficult to
measure in such a complex study. However, the performance of Lab 8 was definitely
poorer than the other laboratories for the VOA determinations. The summary data
for semi volatile compounds presented in Tables 31-35 are presented in matrix
form In Table 37. The poorest performance was experienced by Lab 7 for semi-
volatiles analyses. An excellent performance may be noted for Lab 2. The
problem samples for VOA determinations were also repeated with ILS-5 and ILS-6
being the most difficult. These were closely followed by ILS-3. Excellent
results were obtained for ILS-4 (except for Lab 7) and ILS-2. Lab 4 experienced
considerable difficulty with ILS-3 while Labs 1 and 5 had unusual detection
problems with ILS-5.
175
-------
en
TABLE 23. EFFECT OF REMOVING OUTLIERS ON DATA QUALITY FOR THE DETERMINATION OF
SEMIVOLATILE COMPOUNDS IN SAMPLE ILS-9
Percent Relative
Number of Data Amount Found, vg/g, Standard Deviation
Cpd. Using Given Data Set(a) Using Given Data Set(3) Using Given Ddta Set(a)
No.
41.
72.
18.
10.
107.
109.
114.
117.
118.
121.
127.
130.
134.
136.
138.
139.
140.
Compound
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Di-n -butyl phthalate
Anthraquinone
Fluoranthene
Pyrene
Benzo( a) anthracene
Chrysene
Benzo(k )f 1 uoranthene
Benzo(a)pyrene
Indeno(l,2,3 cdjpyrene
Dibenzo(a,h)anthracene
Benzo(g,h,i)perylene
A
18
18
15
14
18
18
15
17
24
24
24
24
21
21
21
21
20
B
15
15
12
11
15
15
12
14
21
21
21
21
18
18
18
18
17
:====z
C
12
12
9
8
12
12
9
11
18
18
18
18
15
15
15
15
14
A
4
45
1
14
28
10
22
15
53
93
40
70
40
14
14
10
18
B
1
13
1
2
3
2
4
10
34
69
33
62
27
9
9
7
13
C
1
17
1
1
3
2
5
9
32
61
30
60
26
8
8
5
11
A
263
278
28
305
338
287
218
102
154
107
79
44
127
95
127
87
93
B
20
111
27
42
26
87
48
30
23
23
38
18
28
33
48
73
48
C
18
90
24
23
13
74
30
15
18
15
32
17
28
18
36
48
29
(a) Data set A includes all reported data. The maximum number of data possible is 24 (3 replicates x 8
laboratories). One of the 9 laboratories, with EPA's concurrence, did not analyze ILS-9.
Data set B includes all reported data except that from one laboratory.
Data set C includes all reported data except that from two laboratories.
None of the data reported by Lab 5 was included in sets B and C.
-------
TABLE 24. RECOVERY OF VOLATILE COMPOUNDS FROM ILS-2
Amount
Cpd. Spiked, Percent Recovery Reported by Given Laboratory(a)
No. Compound ng/g Lab 1 Lab 2 Lab 3 Lab 4 Lab 5 Lab 6 Lab 7 Lab 8 Lab 9
12. Propionitrile 200 (b) (b) (b) (b) (b) (b) 36 (b) (b)
24. 2-Chloroacrylonitrile 1000 44 66 59 35 38 42 21 38 55
25. 1,1,1-Trichloroethane 800 78 124 112 63 77 78 68 94 124
33. 1,2-Di chl oropropane 200 87 137 100 96 78 86 99 66 106
44. Bromoform 1000 73 134 104 51 328 103 125 60 74
46. 2-Hexanone 800 79 (b) 91 65 137 77 72 94 75
47. 1,1,2,2-Tetrachloroethane 200 86 (b) 95 95 138 85 81 93 45
50. Chlorobenzene 200 95 80 112 105 97 90 84 95 146
51. Ethyl benzene 1200 112 168 139 135 148 133 152 154 163
Total no. of spike compounds
reported 868888988
(a) Average of the values reported for all analytical runs (maximum of 3) in which the compound was
detected and quantified.
(b) Not determined because the compound was not detected and quantified in any of the three analytical
runs performed by the laboratory.
-------
TABLE 25. RECOVERY OF VOLATILE COMPOUNDS FROM ILS-3
CD
Cpd.
No. Compound
12.
24.
25.
33.
44.
46.
47.
50.
51.
Propionitrile
2-Chloroacrylonitrile
1,1,1-Trichloroethane
1 ,2-Di chl oropropane
Bromoform
2-Hexanone
1,1,2,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
Amount
Spiked.
M9/9
500
2000
500
3000
2000
500
500
500
2500
Lab 1
(b)
47
1285
64
(b)
(b)
60
71
60
====================================
Percent Recovery Reported by^ Given
Lab 2
(b)
195
1137
78
98
(b)
116
123
67
Lab 3
(b)
61
2448
113
97
(b)
134
151
82
Lab 4
(b)
55
940
80
35
56
99
94
79
Lab 5
(b)
50
697
32
45
120
58
35
31
Lab 6
(b)
42
940
65
82
46
74
76
61
Laboratory (a)
Lab
(b)
34
71
21
54
61
68
61
42
7 Lab 8
(b)
112
1152
79
35
(b)
123
117
110
Lab 9
(b)
96
1790
106
96
62
81
183
67
Total no. of spike compounds
reported 6 7 7 8 88 8 7 8
(a) Average of the values reported for all analytical runs (maximum of 3) in which the compound was
detected and quantified.
(b) Not determined because the compound was not detected and quantified in any of the three analytical
runs performed by the laboratory.
-------
TABLE 26. RECOVERY OF VOLATILE COMPOUNDS FROM ILS-4
IO
Cpd.
Amount
Spiked,
Percent Recovery Reported by Given Laboratory!3)
No.
12.
24.
25.
33.
44.
46.
47.
50.
51.
Compound
Propionitrile
2-Chloroacrylonitrile
1 ,1 ,1-Trichloroethane
1 ,2-Dichloropropane
Bromoform
2-Hexanone
1,1,2,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
Total no. of spike compounds
reported
===============================
ug/g
500
100
100
400
100
100
500
600
100
:========
Lab
(b)
(b)
15
71
(b)
(b)
75
95
127
5
=====
1 Lab 2
(b)
(b)
161
3450(d)
7
13
8
8
7
Lab 3
426(c)
(b)
103
86
97
(b)
84
89
95
7
Lab 4
(b)
(b)
25
38
19
34
(b)
45
53
6
Lab 5 Lab
(b), . (b)
295(c) (b)
18 109
55 82
88 (b)
2685(d) (b)
159 95
71 98
125 107
8 5
6 Lab
20
(b)
77
81
58
29
38
37
46
8
7 Lab 8
» •
(b)
(b)
(b)
64
(b)
(b)
(b)
89
(b)
2
Lab 9
(b)
(b)
(b)
114
(b)
(b)
(b)
208
161
3
(a) Average of the values reported for all analytical runs (maximum of 3) in which the compound was
detected and quantified.
•V _
(b) Not determined because the compound was not detected and quantified in any of the three analytical
runs performed by the laboratory.
(c) Compound was probably misidentified.
(d) Abnormally low response factor was used for the quantification.
-------
TABLE 27. RECOVERY OF VOLATILE COMPOUNDS FROM ILS-5
CO
o
Cpd.
No.
12.
24.
25.
33.
44.
46.
47.
50.
51.
Compound
Propionltrile
2-Chl oroacryl oni t rll e
1 ,1 ,1-Trichloroethane
1 ,2-Dichloropropane
Bromoform
2-Hexanone
1,1,2 ,2-Tetrachl oroethane
Chlorobenzene
Ethyl benzene
Total no. of spike compounds
reported
Amount
Spiked,
vg/g
25
5
5
20
5
5
25
30
5
=====:
:=========:
Percent Recovery Reported by Given
Lab
(b)
(b)
(b)
20
(b)
(b)
52
47
47
4
1 Lab 2
(b)
42
12
21
94
(b)
79
59
51
7
:=========:
Lab 3
(b)
(b)
(b)
27
85
(b)
72
52
(b)
4
Lab 4
(b)
(b)
(b)
(b)
(b)
(b)
20
13
20
3
Lab 5
281 (O
(b)
(b)
(b)
(b)
(b)
50
(b)
52
3
Lab 6
(b)
(b)
(b)
23
43
67
53
48
69
6
•
Laboratory (a)
Lab 7
(b)
(b)
(b)
(b)
18
(b)
44
12
(b)
3
Lab 8
(b)
(b)
(b)
2
45
(b)
78
41
(b)
4
Lab 9
(b)
(b)
(b)
(b)
2
(b)
8
2
(b)
3
(a) Average of the values reported for all analytical runs (maximum of 3) in which the compound was
detected and quantified.
(b) Not determined because the compound was not detected and quantified in any of the three analytical
runs performed by the laboratory.
(c) Compound was probably misidentified.
-------
TABLE 28. RECOVERY OF VOLATILE COMPOUNDS FROM ILS-6
00
Cpd.
No.
12.
24.
25.
33.
44.
46.
47.
50.
51.
Amount
Spiked,
Compound
Proplonitrile
2-Chloroacryl oni tri 1 e
1 ,1 ,1-Trichloroethane
1 ,2-Di chl oropropane
Bromoform
2-Hexanone
1,1,2 ,2-Tetrachl oroethane
Chlorobenzene
Ethyl benzene
ug/g
5
30
25
5
30
25
5
20
5
Lab 1
(b)
(b)
(b)
(b)
(b)
(b)
(b)
46
62
Percent Recovery Reported by Given
Lab 2
(b)
(b)
1
31
(b)
2
(b)
50
67
Lab 3
(b)
(b)
(b)
(b)
(b)
(b)
(b)
72
(b)
Lab 4
(b)
(b)
(b)
31
(b)
12
(b)
53
79
Lab 5
(b).
25(c)
(b)
??(0
89
188(c)
39
85
Lab 6
(b)
(b)
(b)
22
(b)
(b)
(b)
42
81
»
Laboratory (a)
Lab 7
(b)
(b)
(b)
(b)
(b)
(b)
(b)
31
47
Lab 8
(b)
(b)
(b)
20
(b)
(b)
(b)
54
74
Lab 9
(b)
(b)
(b)
(b)
(b)
(b)
(b)
115
109
Total no. of spike compounds
reported 251473232
(a) Average of the values reported for all analytical runs (maximum of 3) in which the compound was
detected and quantified.
(b) Not determined because the compound was not detected and quantified in any of the three analytical
runs performed by the laboratory.
(c) Compound was probably misidentified.
-------
00
ro
TABLE 29. RECOVERY OF VOLATILE COMPOUNDS FROM ILS-7
Amount
Cpd. Spiked, Percent Recovery Reported by Given Laboratory!9)
No. Compound yg/g Lab 1 Lab 2 Lab 3 Lab 4 Lab 5 Lab 6 Lab 7 Lab 8 Lab 9
12.
24.
25.
33.
44.
46.
47.
50.
51.
Propionitrile
2-Chloroacrylonitr1le
1,1,1-Trichloroethane
1 ,2-Di chl oropropane
Bromoform
2-Hexanone
1,1,2,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
6uO
100
100
500
100
100
600
100
400
114
(b)
16
54
(b)
(b)
43
51
58
(b)
(b)
25
47
42
50
25
30
20
(b)
(b)
60
71
75
(b)
53
84
78
(b)
(b)
(b)
33
15
(b)
33
39
36
(b)
19
22
66
7
257
51
46
51
(b)
(b)
28
55
(b)
81
38
55
61
(b)
(b)
41
66
(b)
89
29
63
66
(b)
(b)
(b)
46
(b)
(b)
39
66
90
(b)
(b)
3
12
13
30
8
10
(b)
Total no. of spike compounds
reported 676586646
(a) Average of the values reported for all analytical runs (maximum of 3) in which the compound was
detected and quantified.
(b) Not determined because the compound was not detected and quantified in any of the three analytical
runs performed by the laboratory.
-------
TABLE 30. RECOVERY OF VOLATILE COMPOUNDS FROM ILS-8
00
CO
Cpd.
No. Compound
Amount
Spiked,
yg/g Lab
Percent Recovery Reported by Given Laboratory(a)
1 Lab 2 Lab 3 Lab 4 Lab 5 Lab 6 Lab 7 Lab 8 Lab 9
12.
24.
25.
33.
44.
46.
47.
50.
51.
Propionltrile
2-Chloroacrylonitrile
1 ,1 ,1-Trichloroethane
1 ,2-Dichloropropane
Bromoform
2-Hexanone
1,1 ,2,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
16000
4000
24000
4000
4000
24000
16000
20000
4000
127
143
132
102
126
73
106
103
106
(b)
174
110
88
100
66
99
94
91
(b)
(b)
124
103
107
97
99
111
109
58
105
55
88
44
104
115
88
89
(b)
203
(b)
59
155
162
96
30
55
(b)
91
97
68
82
75
94
91
88
103
130
138
71
69
94
87
116
90
(b)
(b)
119
(b)
(b
(b)
35
90
(b)
(b)
116
135
(b)
51
90
45
176
203
Total no. of spike compounds
reported
8
(a) Average of the values reported for all analytical runs (maximum of 3) in which the compound was
detected and quantified.
(b) Not determined because the compound was not detected and quantified in any of the three analytical
runs performed by the laboratory.
-------
TABLE 31. RECOVERY OF SEMIVOLATILE COMPOUNDS FROM ILS-2
CO
Cpd.
No. Compound
5.
12.
13.
15.
16.
22.
25.
28.
33.
36.
43.
50.
51.
56.
57.
64.
65.
66.
69.
74.
76.
77.
80.
84.
85.
98.
100.
4-Chlorotoluene
Bis(2-chloroethyl )ether
2-Chlorophenol
2 ,4 ,6-Trimethyl pyridi ne
1 ,4-Dichl orobenzene
Acetophenone
Hexachloroethane
4-t-Butylpyridine
2 ,4-Dimethyl phenol
Propiophenone
4-Chloroaniline
Quinoline
Bi s ( 2-chl oroethoxy Jethane
4-Chl oro-2-methy 1 am' 1 i ne
4-Chlorobenzoic acid
1-Chloronaphthalene
4-Methy Iqui nol i ne
2-Ethyl naphthalene
4-Bromobenzoic acid
1,3-Di nitrobenzene
2,6-Dinitrotoluene
3-Nitroaniline
2,4-Dinitrophenol
4-Nitrophenol
Pentachl orobenzene
2-Chloro-4-ni t roam* 1 i ne
Hexachl orobenzene
Amount
Spiked,
600
500
600
600
100
100
500
100
100
400
600
500
100
100
500
100
100
500
100
500
100
500
400
100
100
100
600
Percent Recovery Reported byJ5iven Laboratory(fl)
Lab 1
38
24
36
47
29
(b)
41
45
(b)
(b)
(b)
24
(b)
(b)
(b)
(b)
(b)
46
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
42
Lab 2
56
67
69
64
46
149
51
76
94
92
40
82
68
112
85
98
102
137
75
71
91
92
(b)
59
84
78
114
Lab 3
13
48
50
(b)
15
23
19
22
(b)
16
6
46
(b)
(b)
(b)
(b)
22
24
(b)
(b)
(b)
(b)
(b)
(b)
26
(b)
24
Lab 4
24
27
22
(b)
17
75
39
44
57
39
26
53
51
21
55
42
46
57
28
29
40
56
(b)
50
49
45
56
Lab 5
35
30
36
(b)
9
74
(b)
(b)
56
56
34
61
55
(b)
(b)
(b)
(b)
59
(b)
19
(b)
48
(b)
(b)
76
(b)
87
Lab 6
125
143
116
305
(b)
108
44
(b)
(b)
95
32
73
(b)
(b)
(b)
(b)
(b)
84
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
85
Lab 7
93
83
77
99
(b)
127
67
(b)
(b)
72
(b)
71
(b)
(b)
(b)
(b)
134
94
(b)
30
(b)
(b)
(b)
(b)
86
(b)
94
Lab 8
t
62
58
51
96
45
109
82
(b)
61
54
344
69
60
34
96
50
57
71
(b)
35
56
(b)
(b)
(b)
96
(b)
80
Lab 9
88
77
82
67
73
158
70
(b)
87
83
6
87
75
69
(b)
(b)
185
99
(b)
49
68
129
(b)
(b)
84
(b)
109
(continued)
-------
TABLE 31. (Continued)
CO
en
Cpd.
No. Compound
Amount
Spiked, '
w9/g Lab
Percent Recovery Reported by Given Laboratory(a)
1 Lab 2 Lab 3 Lab 4 Lab 5 Lab 6 Lab 7 Lab 8 Lab 9
117.
118.
120.
121.
123.
125.
126.
132.
137.
Anthraquinone
Fluoranthene
2-Methy 1 anthraqui none
Pyrene
4, 4 '-ODD
4, 4 '-DDT
Triphenyl phosphate
Tri-(p-tolyl) phosphate
Dibenzocarbazole
600
100
100
400
100
400
600
100
100
27
46
(b)
50
30
23
21
(b)
(b)
129
104
130
105
62
60
180
16
171
19
28
20
26
30
17
24
19
(b)
47
50
50
54
59
63
63
63
35
77
64
78
68
110
96
161
80
85
81
81
(b)
98
(b)
(b)
(b)
(b)
65
55
74
63
98
(b)
43
97
85
(b)
72
84
61
81
88
69
42
(b)
(b)
98
109
103
110
91
91
109
90
189
Total no. of spike compounds
reported
16
35
22
34
24
15
20
28
29
(a) Average of the values reported for all analytical runs (maximum of 3) in which the compound was
detected and quantified.
(b) Hot determined because the compound was not detected and quantified in any of the three analytical
runs performed by the laboratory.
-------
TABLE 32. RECOVERY OF SEMIVOLATILE COMPOUNDS FROM ILS-3
00
CTl
Cpd.
Amount
Spiked,
Percent Recovery Reported by Given Laboratory(a)
No.
5.
12.
13.
15.
16.
22.
25.
28.
33.
36.
43.
50.
51.
56.
57.
64.
65.
66.
69.
74.
76.
77.
80.
84.
85.
98.
100.
Compound
4-Chlorotoluene
Bis(2-chloroethyl)ether
2-Chlorophenol
2 ,4 ,6-Trimethy 1 pyri di ne
1 ,4-Di chl orobenzene
Acetophenone
Hexachloroethane
4-t-Butylpyridi ne
2 ,4-Dimethy 1 phenol
Propiophenone
4-Chloroaniline
Quinoline
Bis(2-ch1oroethoxy)ethane
4-Chloro-2-methyl ani 1 i ne
4-Chlorobenzoic acid
1-Chloronaphthalene
4-Methyl qui nol i ne
2-Ethy 1 naphtha! ene
4-Bromobenzoic acid
1,3-Di nitrobenzene
2 ,6-Di ni trotol uene
3-Nitroaniline
2 ,4-Di nitrophenol
4-Nitrophenol
Pentachl orobenzene
2-Chloro-4-nitroani 1 i ne
Hexachl orobenzene
pg/g
800
200
800
800
200
1000
200
200
200
200
800
200
1200
200
200
1200
1200
200
1200
200
1200
200
200
1000
200
1200
800
Lab 1
30
14
15
22
23
12
37
19
(b)
(b)
(b)
21
23
(b)
(b)
(b)
23
(b)
(b)
(b)
(b)
(b)
(b)
(b)
17
(b)
38
Lab 2
83
90
24
97
62
76
84
54
(b)
54
(b)
95
103
18
(b)
(b)
110
(b)
21
(b)
(b)
(b)
(b)
(b)
45
8
92
Lab 3
97
(b)
16
(b)
74
10
102
64
(b)
16
(b)
(b)
66
(b)
(b)
(b)
99
(b)
(b)
(b)
89
(b)
(b)
38
65
(b)
135
Lab 4
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
47
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
Lab 5
337
(b)
(b)
15
272
122
705
(b)
138
413
39
609
412
(b)
(b)
(b)
494
357
(b)
633
449
419
(b)
(b)
228
133
539
Lab 6
81
139
25
602
50
55
41
(b)
(b)
39
(b)
60
77
68
(b)
(b)
98
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
79
Lab 7
95
88
19
1100
85
54
73
(b)
(b)
47
(b)
84
81
(b)
(b)
(b)
(b)
(b)
it!
46
(b)
(b)
(b)
58
(b)
100
Lab 8
70
81
20
46
51
49
64
37
(b)
29
(b)
67
40
(b)
(b)
(b)
72
(b)
(b)
(b)
48
(b)
(b)
(b)
46
(b)
87
Lab 9
84
86
(b)
270
59
58
62
57
(b)
35
(b)
68
75
(b)
(b)
(b)
(b)
(b)
(b)
122
64
(b)
(b)
(b)
51
(b)
102
(continued)
-------
TABLE 32. (Continued)
Amount
Cpd. Spiked, Percent Recovery Reported by Given Laboratory(a)
No.
117.
118.
120.
121.
123.
125.
126.
132.
137.
Compound
Anthraquinone
Fluoranthene
2-Methyl anthraqui none
Pyrene
4, 4 '-ODD
4, 4 '-DDT
Triphenyl phosphate
Tri -( p-toly 1 ) phosphate
Dibenzocarbazole
yg/g
800
1000
200
200
1000
200
800
200
1200
Lab 1
27
32
(b)
(b)
37
30
23
(b)
(b)
Lab 2
95
77
120
(b)
110
109
117
67
(b)
Lab 3
117
72
111
(b)
109
106
100
90
(b)
Lab 4
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
Lab 5
557
269
506
223
549
1740
578
522
32
Lab 6
101
67
110
(b)
82
(b)
79
(b)
(b)
Lab 7
96
75
107
(b)
107
79
92
116
(b)
Lab 8
»
(b)
63
(b)
6
136
109
92
(b)
3
Lab 9
93
71
102
(b)
113
102
99
(b)
(b)
oo Total no. of spike compounds
^ reported 18 24 20 1 27 18 20 21 20
(a) Average of the values reported for all analytical runs (maximum of 3) in which the compound was
detected and quantified.
(b) Not determined because the compound was not detected and quantified in any of the three analytical
runs performed by the laboratory.
(c) There apparently is a calculation error in this set of data from Lab 5.
-------
TABLE 33. RECOVERY OF SEMIVOLATILE COMPOUNDS FROM ILS-4
CO
00
Cpd.
No.
5.
12.
13.
15.
16.
22.
25.
28.
33.
36.
43.
50.
51.
56.
57.
64.
65.
66.
69.
74.
76.
77.
80.
84.
85.
98.
100.
Compound
4-Chlorotoluene
Bis(2-chloroethyl)ether
2-Chlorophenol
2 ,4 ,6-Trlmethyl pyridi ne
1 ,4-Di chl orobenzene
Acetophenone
Hexachl oroethane
4-t-Butyl pyri di ne
2 ,4-Dimethyl phenol
Proplophenone
4-Chloroaniline
Quinoline
B1s(2-chloroethoxy)ethane
4-Chl oro-2-methyl ani 1 i ne
4-Chlorobenzoic acid
1-Chl oronaphthal ene
4-Methyl qui nol i ne
2-Ethyl naphthal ene
4-Bromo benzole acid
1 ,3-Di nitrobenzene
2 ,6-Di ni trotoluene
3-Nitroaniline
2,4-Dinitrophenol
4-Nitrophenol
Pentachl orobenzene
2-Chl oro-4-ni troani 1 i ne
Hexachl orobenzene
Amount
Spiked,
yg/g
1900
11400
1900
1900
7600
1900
11400
7600
7600
9500
1900
11400
1900
7600
11400
1900
1900
11400
1900
11400
1900
11400
9500
1900
7600
1900
1900
Percent Recovery Reported b:
Lab 1
59
53
64
93
49
53
52
63
62
47
51
16
60
78
67
62
75
8
47
(b)
67
66
2
77
63
49
52
Lab 2
80
77
88
104
62
104
64 *
73
91
99
66
23
88
84
(b)
93
96
9
(b)
104
100
87
1
(b)
88
99
111
Lab 3
85
41
97
(b)
76
98
76
136
106
115
114
25
118
125
75
(b)
42
9
80
82
93
122
ft
81
68
103
Lab 4
39
49
44
153
40
43
34
30
43
39
(b)
11
38
(b)
22
47
46
6
(b)
27
(b)
29
(b)
(b)
58
(b)
55
Lab 5(d.
176
273
471
(b)
133
287
259
187
307
271
242
63
290
130
591
(b)
264
42
(b)
371
396
374
(b)
(b)
270
415
327
/ Given Laboratory (a)
)Lab 6
51
73
65
(b)
50
69
57
61
79
80
163
18
61
25
96
(b)
103
8
(b)
57
63
52
26
(b)
69
37
65
Lab 7
(b)
34
(b)
(b)
50
(b)
42
(b)
(b)
(b)
(b)
(b)
(b)
33
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
48
(b)
(b)
Lab 8
51
60
62
142
49
58
45
42
62
57
(b)
14
56
(b)
93
64
55
7
(b)
50
37
(b)
31
(b)
68
31
64
Lab 9
86
79
81
(b)
78
83
80
115
83
101
(b)
19
75
162
32
(b)
342
10
(b)
99
83
143
46
(b)
95
(b)
90
(continued)
-------
TABLE 33. (Continued)
00
IO
Cpd.
No.
117.
118.
120.
121.
123.
125.
126.
132.
137.
Compound
Anthraquinone
Fluoranthene(c)
2 -Methyl anthraquinone
Pyrenevc)
4, 4 '-ODD
4, 4 '-DDT
Trlphenyl phosphate
Tri-(p-tolyl) phosphate
Dibenzocarbazole
Amount
Spiked,
ug/g
1900
1900
7600
9500
1900
9500
1900
7600
1900
Percent Recovery Reported ty
Lab 1
39
51
42
31
52
33
31
160
(b)
Lab 2
132
102
127
69
86
89
176
223
(b)
Lab 3
110
101
79
57
112
62
99
98
(b)
Lab 4
50
58
44
42
51
50
(b)
(b)
(b)
Lab 5(d;
364
176
339
126
331
351
336
394
16
f Given
(Lab 6
59
68
77
48
48
44
66
71
(b)
Laboratory U)
Lab 7
(b)
49
(b)
42
(b)
(b)
(b)
39
(b)
Lab 8
71
69
63
49
123
61
25
29
(b)
Lab 9
95
111
116
82
88
90
90
94
(b)
Total no. of spike compounds
reported
34
32
31
26
31
31
8
30
29
(a) Average of the values reported for all analytical runs (maximum of 3) in which the compound was
detected and quantified.
(b) Not determined because the compound was not detected and quantified in any of the three analytical
runs performed by the laboratory.
(c) The recoveries are corrected for the amount present in the unspiked sample.
(d) There apparently is a calculation error in this set of data from Lab 5.
-------
TABLE 34. RECOVERY OF SEMIVOLATILE COMPOUNDS FROM ILS-5
VO
o
Cpd.
Amount
Spiked,
Percent Recovery Reported by Given Laboratory(a)
No.
5.
12.
13.
15.
16.
22.
25.
28.
33.
36.
43.
50.
51.
56.
57.
64.
65.
66.
69.
74.
76.
77.
80.
84.
85.
98.
100.
Compound
4-Chlorotoluene
Bis(2-chloroethyl )ether
2-Chlorophenol
2 ,4 ,6-Trimethy 1 py rl di ne
1 ,4-DI chl orobenzene
Acetophenone
Hexachl oroethane
4-t -Butyl pyri dine
2, 4-Di methyl phenol
Propiophenone
4-Chloroaniline
Qu incline
Bi s ( 2-chl oroethoxy ) ethane
4-Chloro-2-methyl ani 1 i ne
4-Chlorobenzoic acid
1-Chloronaphthalene
4-Methylquinolfne
2-Ethyl naphthalene
4-Bromobenzoic acid
1,3-Di nitrobenzene
2,6-Dinitrotoluene
3-Nitroaniline
2,4-Dinitrophenol
4-Nitrophenol
Pentachl orobenzene
2-Chl oro-4-ni t roam' 1 i ne
Hexachl orobenzene
ug/g
4
24
4
-. 4
16
4
24
16
16
20
4
24
4
16
24
4
4
24
4
24
4
24
20
4
16
4
4
Lab 1
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
44
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
Lab 2
7
55
(b)
13
12
61
12
9
(b)
79
(b)
29
77
(b)
(b)
58
15
85
(b)
116
69
(b)
9
(b)
92
80
92
Lab 3
(b)
20
(b)
(b)
10
20
8
38
(b)
59
(b)
40
85
(b)
(b)
(b)
24
42
(b)
56
90
(b)
8
183
73
(b)
74
Lab 4
(b)
30
(b)
(b)
6
23
6
5
(b)
28
(b)
20
(b)
(b)
23
29
9
40
(b)
23
(b)
(b)
(b)
(b)
37
(b)
31
Lab 5
(b)
(b)
(b)
(b)
(b)
(b)
(b)
8
(b)
36
(b)
22
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
39
(b)
(b)
Lab 6
(b)
35
(b)
(b)
7
23
4
18
(b)
50
(b)
35
31
(b)
42
(b)
22
49
(b)
40
(b)
(b)
12
(b)
(b)
(b)
35
Lab 7
(b)
32
(b)
249
10
(b)
9
(b)
(b)
39
(b)
35
(b)
(b)
(b)
61
68
49
(b)
39
61
(b)
(b)
(b)
53
(b)
55
Lab 8
\
(b)
29
(b)
(b)
7
27
5
22
(b)
42
(b)
41
40
(b)
35
49
(b)
58
(b)
39
38
(b)
34
(b)
71
(b)
58
Lab 9
(b)
6
(b)
(b)
2
5
1
7
(b)
8
(b)
6
8
(b)
8
(b)
17
8
(b)
8
8
(b)
7
(b)
9
(b)
8
(continued)
-------
TABLE 34. (Continued)
UD
Amount
Cpd.
No.
117.
118.
120.
121.
123.
125.
126.
132.
137.
Spiked,
Compound
Anthraquinone
Fluoranthene
2-Methylanthraquinone
Pyrene
4, 4 '-ODD
4, 4 '-DDT
Triphenyl phosphate
Tri-(p-tolyl) phosphate
Dibenzocarbazole
Total no. of spike compounds
reported
wg/g
4
4
16
20
4
20
4
16
4
Lab
(b)
61
(b)
33
(b)
21
(b)
(b)
(b)
4
Percent Recovery Reported by Given Laboratory(a)
1 Lab
88
101
96
92
85
104
89
154
(b)
29
2 Lab 3
68
122
60
47
108
50
86
77
(b)
24
Lab 4
18
22
22
22
40
45
39
35
(b)
22
Lab 5
140
159
(b)
53
172
(b)
(b)
53
(b)
9
Lab
28
38
48
41
32
37
38
(b)
(b)
21
6 Lab 7
46
65
45
49
85
41
58
67
(b)
21
Lab 8
53
57
43
50
94
85
35
17
(b)
24
Lab 9
6
9
6
7
8
8
(b)
9
(b)
13
(a) Average of the values reported for all analytical runs (maximum of 3) in which the compound was
detected and quantified.
(b) Not determined because the compound was not detected and quantified in any of the three analytical
runs performed by the laboratory.
-------
TABLE 35. RECOVERY OF SEMIVOLATILE COMPOUNDS FROM ILS-6
UD
ro
Cpd.
Amount
Spiked,
Percent Recovery Reported by Given Laboratory(a)
No.
5.
12.
13.
15.
16.
22.
25.
28.
33.
36.
43.
50.
51.
56.
57.
64.
65.
66.
69.
74.
76.
77.
80.
84.
85.
98.
100.
Compound
4-Chlorotoluene
Bis(2-chloroethyl )ether
2-Chlorophenol
2 ,4 ,6-Trimethyl pyri di ne
1 ,4-Di chl orobenzene
Acetophenone
Hexachloroethane
4-t-Butyl pyri dine
2, 4-Di methyl phenol
Propiophenone
4-Chloroaniline
Quinoline
Bi s( 2-chl oroethoxy ) ethane
4-Chl oro-2-methylani 1 i ne
4-Chlorobenzoic acid
1-Chloronaphthalene
4-Methylquinoline
2-Ethyl naphthalene
4-Bromobenzoic acid
1 ,3-Di nitrobenzene
2 ,6-Di nitrotol uene
3-Nitroaniline
2,4-Dinitrophenol
4-Nitrophenol
Pentachl orobenzene
2-Chloro-4-ni t roam" 1 i ne
Hexachl orobenzene
W9/9
20
20
20
20
100
20
20
100
100
120
20
20
80
100
20
80
80
20
80
20
80
20
120
20
100
80
20
Lab 1
19
(b)
58
(b)
17
20
(b)
44
57
16
(b)
27
42
33
(b)
29
26
39
58
(b)
(b)
(b)
(b)
(b)
28
(b)
(b)
Lab 2
55
45
69
60
47
42
(b)
68
82
41
(b)
20
26
76
(b)
86
49
102
11
(b)
(b)
(b)
(b)
(b)
89
(b)
100
Lab 3
(b)
(b)
58
(b)
22
(b)
(b)
40
63
19
(b)
16
55
13
(b)
(b)
27
29
41
(b)
(b)
(b)
(b)
(b)
41
(b)
46
Lab 4
(b)
(b)
(b)
(b)
25
(b)
(b)
39
42
17
(b)
(b)
39
(b)
(b)
(b)
24
48
18
(b)
(b)
(b)
(b)
(b)
50
(b)
48
Lab 5
(b)
(b)
(b)
(b)
4
(b)
(b)
15
30
17
(b)
(b)
(b)
(b)
(b)
63
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
31
(b)
(b)
Lab 6
(b)
(b)
(b)
(b)
53
17
(b)
35
55
25
(b)
29
52
40
(b)
66
68
71
47
(b)
(b)
(b)
(b)
(b)
69
(b)
58
Lab 7
(b)
(b)
(b)
(b)
33
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
48
(b)
(b)
(b)
(b)
(b)
(b)
(b)
(b)
68
(b)
(b)
Lab 8
t
6
21
19
(b)
7
16
(b)
16
19
6
(b)
(b)
12
(b)
50
225
19
24
(b)
(b)
(b)
(b)
(b)
(b)
22
(b)
(b)
Lab 9
(b)
(b)
(b)
(b)
34
(b)
(b)
83
50
25
(b)
(b)
44
(b)
150
(b)
123
75
(b)
(b)
(b)
(b)
(b)
(b)
67
(b)
(b)
(continued)
-------
TABLE 35. (Continued)
CO
Cpd.
Amount
Spiked,
Percent Recovery Reported by Given Laboratory(a)
No.
117.
118.
120.
121.
123.
125.
126.
132.
137.
Compound
Anthraquinone
Fluoranthene
2-Methyl anthraqui none
Pyrene
4, 4 '-ODD
4, 4 '-DDT
Triphenyl phosphate
Tri-(p-tolyl) phosphate
Di benzocarbazol e
yg/g
20
20
100
120
20
120
20
100
80
Lab 1
(b)
28
17
30
78
(b)
(b)
30
33
Lab 2
56
74
67
100
311
(b)
160
116
195
Lab 3
18
40
24
20
107
(b)
29
43
(b)
Lab 4
(b)
49
(b)
63
198
(b)
41
66
36
Lab 5
(b)
63
24
49
85
(b)
(b)
56
90
Lab 6
30
79
47
81
184
(b)
(b)
71
93
Lab 7
(b)
82
39
84
(b)
(b)
(b)
91
(b)
Lab 8
i
(b)
25
14
25
68
(b)
(b)
23
(b)
Lab 9
(b)
94
54
97
262
(b)
(b)
81
(b)
Total no. of spike compounds
reported
21
27
20
16
12
21
19
14
(a) Average of the values reported for all analytical runs (maximum of 3) in which the compound was
detected and quantified.
(b) Not determined because the compound was not detected and quantified in any of the three analytical
runs performed by the laboratory.
-------
TABLE 36. MATRIX OF TOTAL VOLATILE COMPOUNDS REPORTED
Sampl e
ILS-2
ILS-3
ILS-4
ILS-5
ILS-6
ILS-7
ILS-8
•
1
8
6
5
4
2
6
9
2
6
7
7
7
5
7
8
3
8
7
7
4
1
6
7
Laboratory
4 5
8 8
8 8
6 8
3 3
4 7
5 8
9 7
Number
6
8
8
4
6
3
6
8
7
9
8
8
3
2
6
9
Total Per
Laboratory 40 47 40 43 49 44 45
3333333333333333333333333333333333333333333333333333333
TABLE 37
3333333333333333333.
Sampl e
ILS-2
ILS-3
ILS-4
ILS-5
ILS-6
1
16
18
34
4
21
. MATRIX
3333333333
2
35
24
32
29
27
3
22
20
31
24
20
OF TOTAL SEMIVOLATILE
Laboratory
4 5
34 24
1 27
26 31
22 9
16 12
Number
6
15
18
31
21
21
COMP
7
20
20
8
21
7
8
8
7
2
4
3
4
3
9
8
8
3
3
2
6
7
Total Per
Sampl e
71
67
51
37
29
54
67
31 37
33 S3 33= 3 3 33333333333333 3
OUNDS REPORTED
33333333333333:
8
28
21
8
21
7
9
29
20
29
13
14
1333333333
Total Per
Sampl e
223
168
252
167
157
Total Per
Laboratory 93 147 117 99 103 106 76 122 105
33333333333333333333=3333333333333333333S333S333S333333333333333333333333333333
194
-------
However, caution should be used 1n attributing poor results with particu-
lar samples as a failure of the method. Appreciable compound losses may have
occurred prior to analysis as a result of spiking procedures, degradation during
storage or handling losses.
A presentation of percent detection data, arranged according to specific
compound, for each waste sample 1s provided 1n Table 38 for volatile compounds
and Table 39 for semlvolatHe compounds. Interpretation of these two data
tables provides Information regarding the effect of compound and further
Information regarding effect of sample. In addition, the data are Identified as
either low spike level (underlined values) or high spike level. The average
percent of reported compounds was higher for both volatile and semlvolatHe
compounds spiked at the high level (73 and 70 percent respectively). The low
level spikes were reported 57 percent of the time for volatile determinations
and 47 percent for semi volatile determinations. The most difficult waste
samples are also apparent using these data, namely ILS-5 and ILS-6 for VGA and
ILS-3, ILS-5 and ILS-6 for semlvolatHe analyses.
The ability of a laboratory to detect specific volatile and semi volatile
compounds may be discerned from the data presented 1n Tables 40 and 41. Labs
3 and 8 reported the fewest volatile compounds. Lab 7 reported only 42 percent
of the semivolatile compounds compared to 81 percent for Lab 2 and an average
reporting value of 59 percent for all the participants.
The most troublesome compounds may be Identified by comparing data pre-
sented 1n Tables 38 through 41. The best VOA results were obtained for high
boiling compounds such as chlorobenzene, tetrachloroethane, and ethylbenzene.
The more polar compounds, prop1on1tr1le, chloroacrylon1tr1le and hexanone,
were not detected as often as the less .polar compounds. The semlvolatHe
compounds most suitable for determination by the methodology were aromatic
hydrocarbons and halocarbons such as pyrene, fluoranthene, and hexachloro-
benzene. Poor results were generated for benzole adds and nltrophenols.
The laboratory-to-laboratory variability for detection, Identification and
quantification of the volatile and semi volatile compounds may provide key
Information for methodology applicability and revisions. In some cases, errors
1n calculation are the primary cause of variability and some misunderstanding
of reporting requirements may have occurred. Trends that were apparent Include:
Lab 2 consistently reported more SV and VOA compounds
Lab 7 reported the fewest SV compounds
Lab 8 reported the fewest VOA compounds
Lab 5 reported high recoveries for both SV and VOA compounds
Lab 4 reported lower recoveries for SV compounds and the most variabil-
ity 1n total detection on a sample-to-sample basis
The causes for some of the trends may be dlscernable by Implementing a more
exhaustive approach to data Interpretation.
The most dramatic difference 1n application of the analysis methods was
experienced with the prescreening protocols. The amount of tetraglyme extract
analyzed for each waste by specific participants 1s presented 1n Table 42.
The volumes vary by three orders of magnitude for samples ILS-4, ILS-5, ILS-6
and ILS-8. The laboratory reporting fewest VOA compounds, Lab 8, consistently
195
-------
vo
TABLE 38. EFFECT OF COMPOUND, SAMPLE AND SPIKE LEVEL ON DETECTABILITY OF VOLATILE COMPOUNDS
Percent of Data Reported For Given Sample(a)(k) Average Percent
No.
12.
24.
25.
33.
44.
46.
47.
50.
51.
Compound
Propionitrile
Chloroacrylonitrile
1,1,1-Trichloroethane
1,2-Dichloropropane
Bromoform
2-Hexanone
1 , 1 ,2 ,2-Tet rachl oroethane
Chlorobenzene
Ethyl benzene
Average
(a) The maximum number of analyt
.(3 replicates x 9 laboratori
The percent of data reported
ILS-2 ILS-3 ILS-4 ILS-5 ILS-6
11
100
100
100
100
89
89
96
100
0
96
93
96
85
52
93
96
96
87 79
22
4_
78
100
56
37
67
100
89
11
H
H
56
67
H
100
89
il
61 44
Q
4
7
48
4
26
4
100
89
31
ical runs for which data could be
es).
= total number of analytical runs
ILS-7
11
4
78
100
56
52
100
100
89
66
ILS-8 Low(c) High(a)
26
78
89
78
89
89
100
100
89
82
4
24
65
75
67
38
62
97
77
57
*
18
67
65
88
63
68
92
97
95
73
reported for each sample was 27
for which data were reported x
Total (e)
12
42
65
83
65
51
79
97
87
64
100
27
Percent of data reported for compounds spiked at the low level are underlined.
(c) Average of the percents of data reported for samples spiked at the low level.
(d) Average of the percents of data reported for samples spiked at the high level.
(e) Average of the percents of data reported for all samples.
-------
TABLE 39. EFFECT OF COMPOUND, SAMPLE AND SPIKE LEVEL ON SEMIVOLATILE COMPOUNDS
UD
Percent of Data Reported For Given Sample(a)(b) Average Percent
No.
6.
12.
13.
15.
16.
22.
25.
28.
33.
36.
43.
50.
51.
56.
57.
64.
65.
66.
69.
74.
76.
77.
80.
84.
85.
98.
100.
117.
Compound
4-Chlorotoluene
Bis(2-chloroethyl Jether
2-Chlorophenol
2,4, 6-Tri methyl py ri di ne
1,4-Dlchlorobenzene
Acetophenone
Hexachloroethane
4-t -Butyl py ri di ne
2, 4-Dimethyl phenol
Propi ophenone
4-Chloroaniline
Qu incline
Bis(2-chloroetho>ty) ethane
4-Chl oro-2-methyl ani 1 i ne
4-Chlorobenzoic acid
1-Chloronaphthalene
4-Methy Iqui nol i ne
2-Ethyl naphthalene
4-Bromobenzoi c acid
1,3-Di nitrobenzene
2,6-Dinitrotoluene
3-Nitroaniline
2,4-Dinitrophenol
4-Nitrophenol
Pentachlorobenzene
2-Chloro-4-nitroani 1 i ne
Hexachl orobenzene
Anthraquinone
ILS-2
96
100
100
59
70
89
89
44
56
89
74
100
56
41
33
33
63
100
22
63
44
44
0
11
74
22
100
100
ILS-3
89
67
67
78
89
89
85
56
4
78
IT
78
100
22
0
ff
67
11
TT
19
56
7
0
11
78
15
89
78
ILS-4
89
96
89
37
100
89
100
89
89
89
56
89
89
78
78
44
89
89
22
78
78
78
56
11
100
59
89
W
ILS-5
11
78
0
2?
78
67
7B
67
0
89
0
89
56
0
44
44
67
89
0
78
52
~0
44
11
78
11
78
IT
ILS-6
33
22
44
TT
93
44
0
85
89
89
11
44
78
44
22
W
78
78
96
0
0
0
0
0
100
0
44
33"
Low(C)
44
45
44
23
80
72
43
50
30
78
22
61
67
32
11
40
73
45
15 .
10
58
4
0
8
76
31
70
68
High(Q)
93
91
84
69
90
89
89
80
59
89
43
93
89
41
52
30
73
93
54
73
28
41
33
11
93
8
95
89
Total (e)
64
73
60
41
86
76
70
68
48
87
30
80
76
37
35
36
73
73
30
48
46
26
20
9
86
21
80
76
(continued)
-------
TABLE 38. EFFECT OF COMPOUND, SAMPLE AND SPIKE LEVEL ON DETECTABILITY OF VOLATILE COMPOUNDS
10
cr>
Percent of Data Reported For Given
No.
12.
24.
25.
33.
44.
46.
47.
50.
51.
Compound
Proplonitrile
Chloroacrylonitrile
1,1,1-Trichloroethane
1,2-Dlchloropropane
Bromoform
2-Hexanone
1,1,2,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
Average
ILS-2
_U
100
100
100
100
89
89
96
100
87
ILS-3
0
96
93
96
85
52
93
96
96
79
ILS-4
22
_4
78
100
56
37
67
100
89
61
ILS-5
11
n.
n.
56
67
JU
100
89
41
44
ILS-6
0
4
7
48
4
26
4
100
89
31
Sample(a)(b)
ILS-7
11
4
78
100
56
52
100
100
89
66
ILS-8
26
18
89
78
89
89
100
100
8£
82
Average Percent
Low(c)
4
24
65
75
67
38
62
97
77
57
High(Q)
18
67
65
88
63
68
92
97
95
73
Total (e)
12
42
65
83
65
51
79
97
87
64
(a) The maximum number of analytical runs for which data could be reported for each sample was 27
(3 replicates x 9 laboratories).
The percent of data reported = total number of analytical runs for which data were reported x 100
27
(b) Percent of data reported for compounds spiked at the low level are underlined.
(c) Average of the percents of data reported for samples spiked at the low level.
°) Average of the percents of data reported for samples spiked at the high level.
e) Average of the percents of data reported for all samples.
-------
TABLE 39. EFFECT OF COMPOUND, SAMPLE AND SPIKE LEVEL ON SEMIVOLATILE COMPOUNDS
VO
Percent of Data Rep
No.
6.
12.
13.
15.
16.
22.
25.
28.
33.
36.
43.
50.
51.
56.
57.
64.
65.
66.
69.
74.
76.
77.
80.
84.
85.
98.
100.
117.
Compound
4-Chlorotoluene
Bis(2-chloroethyl Jether
2-Chlorophenol
2,4,6-Trimethylpyridine
1,4-Dichlorobenzene
Acetophenone
Hexachloroethane
4-t-Butylpyrldine
2, 4-Dimethyl phenol
Propi ophenone
4-Chloroaniline
Qu incline
Bis(2-chloroethoxy) ethane
4-Chl oro-2-methy 1 ani 1 i ne
4-Chlorobenzolc acid
1-Chloronaphthalene
4-Methylquinoline
2-Ethyl naphthalene
4-Bromobenzoic acid
1,3-Di nitrobenzene
2,6-Dinitrotoluene
3-Nitroaniline
2,4-Dinitrophenol
4-Nitrophenol
Pent achlo robe nzene
2-Chl oro-4-ni t roani 1 i ne
Hexachl orobenzene
Anthraquinone
ILS-2
96
100
100
59
70
89
89
44
56
89
74
100
56
41
33
33
63
100
22
63
44
44
0
11
74
22
100
100
ILS-3
89
67
67
78
89
89
85
56
^4
78
11
78
100
22
~0
U
67
11
11
19
56
7
0
11
78
15
89
78
orted For Given
ILS-4
89
96
89
17
100
89
100
89
89
89
56
89
89
78
78
44
89
89
22
78
78
78
56
11
100
59
89
H
ILS-5
11
78
0
21
78
67
78
67
0
89
0
89
56
~0
44
44
67
89
0
78
52
0
44
11
78
11
78
8T
Sampje(a)(bl Average Percent
ILS-6
33
22
44
TT
93
44
~0
85
89
89
11
44
78
44
22
W
78
78
96
0
0
0
0
0
100
0
44
3J
Low(C)
44
45
44
23
80
72
43
50
30
78
22
61
67
32
11
40
73
45
15 .
10
58
4
0
8
76
31
70
68
High(a)
93
91
84
69
90
89
89
80
59
89
43
93
89
41
52
30
73
93
54
73
28
41
33
11
93
8
95
89
Total (e)
64
73
60
41
86
76
70
68
48
87
30
80
76
37
35
36
73
73
30
48
46
26
20
9
86
21
80
76
(continued)
-------
TABLE 39. (Continued)
UD
00
No. Compound
Percent of Data Reported For Given Sample(a)(b) Average Percent
ILS-2 ILS-3 ILS-4 ILS-5 ILS-6 Low(c) High(d) Total (e)
118.
120.
121.
123.
125.
126.
132.
137.
Fluoranthene
2-Methy 1 a nth raqui none
Pyrene
4, 4' -ODD
4, 4 '-DDT
Triphenyl phosphate
Tri-(p-tolyl) phosphate
Dibenzocarbazole
Average
=============================
100
~78
100
78
85
89
63
JJ6
67
============
89
67
22
89
74
89
41
11
51
=======
100
~89
100
89
89
78
89
n.
77
========
93
78
100
81
89
59
70
13
52
:======:
100
"89
100
89
0
33
85
52
48
98
73
22
84
74
57
52
22
47
89
85
100
89
66
89
81
32
70
96
80
84
85
67
70
70
26
59
(a) The maximum number of analytical runs for which data could be reported for each sample was 27
(3 replicates x 9 laboratories).
The percent of data reported = total number of analytical runs for which data were reported x 100
27
(b) Percent of data reported for compounds spiked at the low level are underlined.
(c) Average of the percents of data reported for samples in which the compound was spiked at the low level.
(d) Average of the percents of data reported for samples in which the compound was spiked at the high level.
(e) Average of the percents of data reported for all samples.
-------
TABLE 40. EFFECT OF LABORATORY ON DETECTABILITY OF VOLATILE COMPOUNDS
10
vo
Cpd
No.
12.
24.
25.
33.
44.
46.
47.
50.
51.
(a)
•
Percent of Data Reported by Given Laboratory(a)
Compound Lab 1
Propionitrile
2-Chloroacrylonitrile
1,1, 1-Trl chl oroethane
1 ,2-Di chl crop ropane
Bromof orm
2-Hexanone
1,1,2,2-Tetrachloroethane
Chlorobenzene
Ethyl benzene
! Average
29
43
71
86
29
29
86
100
100
64
The maximum number of analytical runs
(3 replicates x 7 laboratories).
The percent of data reported =
total
Lab 2 Lab 3
0
57
95
100
86
38
67
95
100
72
==========
for which
number of
14
29
71
86
86
29
86
100
71
56
data
Lab 4
5
43
57
86
71
71
71
100
90
68
could be
analytical runs
Lab 5 Lab 6 Lab 7
14
52
57
76
76
86
90
86
90
71
reported
0
38
67
95
52
67
81
95
95
67
for each
43
43
71
71
71
67
86
100
86
72
sample
Lab 8
0
29
43
86
43
14
71
100
57
51
was 21
for which data were reported x
Lab 9
0
43
57
57
71
57
71
100
71
60
100
27
-------
TABLE 41. EFFECT OF LABORATORY ON DETECTABILITY OF SEMIVOLATILE COMPOUNDS
ro
o
o
Cpd.
No.
6.
12.
13.
15.
16.
22.
25.
28.
33.
36.
43.
50.
51.
56.
57.
64.
65.
66.
69.
74.
76.
77.
80.
84.
85.
98.
100.
Compound
4-Chlorotoluene
Bis(2-chloroethyl Jether
2-Chlorophenol
2,4, 6-Trimethylpyri dine
1 ,4-Di chl orobenzene
Acetophenone
Hexachl oroethane
4-t-Buty 1 py ri di ne
2, 4-Dimethyl phenol
Propiophenone
4-Chloroaniline
Quinoline
Bis(2-chloroethoxy) ethane
4-Chloro-2-methyl aniline
4-Chlorobenzoic acid
1-Chloronaphthalene
4-Methylquinoline
2-Ethyl naphthalene
4-Bromobenzoic acid
1,3-Di nitrobenzene
2,6-Dinitrotoluene
3-Nitroaniline
2,4-Di nitrophenol
4-Nitrophenol
Pentachl orobenzene
2-Chl oro-4-ni t roani 1 i ne
Hexachl orobenzene
:=======================================:
Percent of Data Reported by
Lab 1
80
1 60
80
60
80
73
60
80
40
40
20
80
60
40
20
40
60
80
40
0
20
20
20
20
60
20
60
Lab 2
100
100
80
100
100
100
80
100
60
100
60
100
100
80
20
80
100
80
60
60
60
40
40
20
100
80
100
Lab 3
60
60
80
0
100
80
80
100
40
100
40
80
80
40
20
0
100
80
40
40
60
20
13
40
100
20
100
Lab 4
40
60
40
20
80
60
60
80
60
80
20
60
80
20
60
60
80
80
40
60
20
40
0
20
80
20
80
Lab 5
53
33
40
20
53
60
33
47
60
100
60
80
60
20
20
7
40
60
0
47
40
53
0
0
93
20
60
Given Laboratory (a)
Lab 6
60
80
60
40
80
100
80
60
40
100
40
100
80
60
40
20
80
80
20
40
20
20
40
0
40
20
100
Lab 7
40
80
40
53
80
40
80
0
0
53
0
60
20
20
0
40
33
40
0
40
40
0
0
0
100
0
60
Lab 8
80
100
80
40
100
100
80
67
60
100
13
80
100
13
80
80
80
80
0
60
73
0
27
0
100
7
80
Lab 9
60
80
40
40
100
80
80
80
60
100
20
80
100
40
60
0
80
80
0
80
80
40
40
0
100
0
80
(continued)
-------
TABLE 41. (Continued)
ro
o
Cpd. Percent of Data Reported by Given Laboratory(a)
No. Compound Lab~lLab~2 Lab~3 Lab~4Lab 5 La5~6 Lab 7 Lab 8 Lab 9
117.
118.
120.
121.
123.
125.
126.
132.
137.
Ant hraqui none
Fluoranthene
2-Methy 1 ant hraqui none
Pyrene
4, 4 '-ODD
4, 4 '-DDT
Triphenyl phosphate
Tri-(p-tolyl) phosphate
Dibenzocarbazole
60
100
40
80
80
80
60
40
20
100
100
100
80
100
80
100
100
40
100
100
100
80
100
80
100
100
0
60
80
60
80
80
60
60
60
40
67
87
80
100
87
47
60
60
67
100
100
80
80
80
40
60
40
40
60
100
80
80
60
60
60
100
0
60
100
80
100
80
80
67
47
7
80
100
100
80
100
80
60
80
20
Average 52 81 65 55 50 59 42 64 63
(a) The maximum number of analytical runs for which data could be reported for each sample was 21
(3 replicates x 7 laboratories).
The percent of data reported = total number of analytical runs for which data were reported x 100
27
-------
TABLE 42. AMOUN1
Laboratory
Lab 1
Lab 2
Lab 3
Lab 4
Lab 5
Lab 6
Lab 7
Lab 8
Volun
ILS-2
20
5
10
10
5
30
(b)
1
f OF TETRAGLYME EXTRACT ANALYZED FOR THE DETERMINATION
OF VOLATILE COMPOUNDS (
==========================================================
ie of Tetraglyme Extract of Given Waste Analyzed, pL(a)
ILS-3
3
5
1
3
20
10
(b)
1
ILS-4
2
250
3
25
5
5
(b)
3
ILS-5
200
2000
30
200
50
200
(b)
200
ILS-6
200
2000
15
200
200
200
(b)
200
ILS-7
10
100
10
10
50
30
(b)
4
ILS-8
1
1
0.2
0.2
10
5
(b)
0.04
Lab 9 6 3 0.2 100 3 60 0.06
3=3============================================================================
(a)jhe volume given is that portion of the inttial tetraglyme extract repre-
senting Ig of waste per 20 mL of extract. In order to obtain the amount of
sample represented by the smaller volumes, the extract was diluted with
tetraglyme. For example, 1 uL of the Initial extract may have been
obtained by using 20 uL of a 1:20 dilution of the initial extract.
(b^Not reported.
used a smaller volume of tetraglyme extract for the purge and trap part of the
method. The laboratories reporting the highest percent of VOA data (Labs 2
and 5 generally used large volumes of extract. Lab 7 did not report these
data so interpretation of performance could not be made.
For the semivolatHe screening method, variations of up to three orders of
magnitude were observed for the methylene chloride extract concentration factor
as shown 1n Table 43. Lab 2 which reported the most semlvolatile compounds
used the higher concentration factors. The poor performance by Lab 7 for SV
determination could not be assessed, for this parameter, since the concentra-
tion factor was not reported. Additional Information regarding the criteria
used by each of the laboratories to make the decisions affecting the VOA extract
volume and SV concentration factor is essential for standardizing the applica-
tion of the tested methods.
In addition to the comparative data presented in tabular form 1n this re-
port, additional comparisons could be made by fully using the large quantity
of information generated for the program. Statistical analyses excluding obvious
outliers were beyond the scope of the effort reported here, however, valuable
Information for decision making could be gleaned by further evaluation of the
data.
202
-------
TABLE 43. METHYLENE CHLORIDE EXTRACT CONCENTRATION FACTOR USED FOR
SEMIVOLATILE COMPOUND DETERMINATION
oo
Concentration Factor Used
Laboratory
Lab 1
Lab 2
Lab 3
Lab 4
Lab 5
Lab 6
Lab 7
Lab 8
ILS-2
1
100
3
25
4
1
(b)
2
ILS-3
1
3
1
0.1
1
1
(b)
2
For Methylene Chloride Extract of Given Waste Analyzed^3)
ILS-4
1
1
1
0.1
1
0.2
(b)
0.2
ILS-5
10
50
100
100
50
100
(b)
100
ILS-6
25
50
25
100
4
12
(b)
15
ILS-9
50
150
40
(c)
4
50
(b)
100
Lab 9 2 1 0.2 100 4 30
(a) The initial methylene chloride extract representing 1 g of waste per 50 mL of extract was concentrated
by the listed factor prior to analyses. Concentration factors less than one indicate that the initial
extract was diluted accordingly.
(b) Not reported.
(c) This sample was not analyzed by Lab 4.
-------
REFERENCES
1. "Phase I Report on Evaluation of Methods for Hazardous Waste Analysis" to
U.S. Environmental Protection Agency from Battelle Columbus Laboratories,
July 14, 1981.
2. "Phase II Report on Evaluation of Methods for Hazardous Waste Analysis" to
U.S. Environmental Protection Agency from Battelle Columbus Laboratories,
September 24, 1982.
3. "Purgeables — Method 624", Federal Register, Vol. 44, 3 Dec. 1973, p. 69532,
4. Davies, 0. L., The Design and Analysis of Industrial Experiments, 1978,
Longman Group Limited, New York.
5. Snedecor and Cochran, Statistical Methods Sixth Edition, 1974, The Iowa
State University Press, Ames, Iowa.
204
-------
APPENDIX A
METHOD FOR THE DETERMINATION OF SEMIVOLATILE
ORGANIC COMPOUNDS IN SOLID WASTES
205
-------
Battelle - Columbus Laboratories - January, 1982
METHOD FOR THE DETERMINATION OF SEMIVOLATILE
ORGANIC COMPOUNDS IN SOLID WASTES
1. Scope and Application
1.1 This method covers the determination of semivolatile organic compounds
1n a variety of solid waste matrices.
1.2 This method is applicable to nearly all types of samples, regardless
of water content, including aqueous sludges, caustic liquors, acid
liquors, waste solvents, oily wastes, mousses, tars, fibrous wastes,
polymeric emulsions, filter cakes, spent carbons, spent catalysts,
soils, and sediments.
1.3 This method is applicable to the determination of most neutral, acidic,
and basic organic compounds that are soluble 1n methylene chloride and
are capable of being eluted without derivatization as sharp peaks from
a gas chromatographic fused silica capillary column coated with a
slightly polar si 11 cone. Such compounds include polynuclear aromatic
hydrocarbons, chlorinated hydrocarbons and pesticides, phthalate
esters, organophosphate esters, nitrosamines, haloethers, aldehydes,
ethers, ketones, anilines, pyridines, qulnolines, aromatic nitro
compounds, and phenols including nitrophenols.
1.4 The detection limit of the method for determining an Individual compound
is estimated to be approximately 1 pg/g (wet weight). For samples
which contain more than 1 mg/g of total solvent extractable materials,
the detection limit should be proportionately higher.
1.5 This method is based upon a solvent extraction, gas chromatographic/mass
spectrometric (GC/MS) procedure.
1.6 This method Is restricted to use by or under the supervision of analysts
experienced in the use of gas chromatograph/mass spectrometers and
skilled in the interpretation of mass spectra. Each analyst must
demonstrate the ability to generate acceptable results with this method.
2. Summary of Method
2.1 A measured weight of sample, 3.0 g wet weight, 1s neutralized and
sonified with 150 mL of methylene chloride. Anhydrous sodium sulfate
is added to bind the water present. A portion of the methylene chloride
supernatant is concentrated and analyzed by GC/MS using a fused silica
capillary column. Qualitative identification is performed using the
retention time of the compound and the relative abundance of three or
more characteristic ions. Quantitative analysis is performed using an
internal standard technique with a single characteristic ion.
206
-------
3. Interferences
3.1 Method interferences may be caused by contaminants in solvents,
reagents, glassware, and sample processing hardware that produces
discrete artifacts and/or elevated baselines in the total ion current
profiles. All of these problems must be demonstrated to be absent
under the conditions of the analysis by routinely analyzing laboratory
reagent blanks.
3.1.1 Glassware must be scrupulously cleaned. Clean all glassware as
soon as possible after use by rinsing it with the last solvent
used. Heating in a muffle furnace at 450°C for 5 to 15 hours
is recommended whenever feasible. Alternatively detergent
washes, water rinses, acetone rinses, and oven drying may be
used. Cleaned glassware should be sealed and stored in a
clean environment to prevent any accumulation of dust or other
contaminants.
3.1.2 The use of high purity reagents and solvents helps to minimize
interference problems.
3.2 Matrix interferences may be caused by components that are coextracted
from the sample but are not normally of interest. The most common of
such components are petroleum-derived naphthenes, high molecular weight
polymeric components, and long-chain components such as waxes and
triglycerides. The extent of such matrix interferences will vary
considerably from sample to sample. A cleanup procedure using gel
permeation chromatography has been incorporated into the method for
certain cases to remove long-chain and high-molecular-weight material.
No cleanup procedure is available for the removal of naphthenes. When
such matrix interferences are present, the sample extract is diluted
and the detection limit is increased proportionately. Many of the
matrix interferences are solvent-extractable nonvolatile components
which necessitate the more frequent cleaning of the GC injection port
and the more frequent removal of the injection end of the GC capillary
column.
4. Safety
4.1 The toxicity of cardnogeniclty of each reagent used in this method
has not been precisely defined; however, each chemical compound should
be treated as a potential health hazard. From this viewpoint, exposure
to these chemicals must be minimized by whatever means available. The
laboratory is responsible for maintaining a current awareness file of
OSHA regulations regarding the safe handling of the chemicals specified
in this method. A reference file of material data handling sheets
should also be made available to all personnel involved in the chemical
analysis.
4.2 All operations involving the use of methylene chloride, including the
extraction of the waste sample, filtration of the extract, and
207
-------
concentration of the extract, must be performed in a fume hood. Care
should be taken to avoid the skin contact with methylene chloride.
5. Apparatus -
5.1 Sampling equipment - Glass screw-cap vials or jars of at least 100 ml
capacity. Screw caps must be Teflon lined.
5.2 Glassware
5.2.1 Beaker - 400 ml
5.2.2 Centrifuge tubes - approximately 200-mL capacity, glass with
screw cap (Corning No. 1261 or equivalent). Screw caps must
be fitted with Teflon liners.
5.2.3 Concentrator tube, Kuderna-Danish - 25 ml, graduated (Kontes K
570050-2526 or equivalent). Calibration must be checked at the
volumes employed in the test. A ground glass stopper is used to
prevent extract evaporation.
5.2.4 Evaporative flask, Kuderna-Danish 250-mL (Kontes K-570001-0250
or equivalent). Attach to concentrator tube with springs.
5.2.5 Snyder column, Kuderna-Danish - Three-ball macro (Kontes K-
503000-0121 or equivalent).
5.2.6 Snyder column, Kuderna-Danish - Two-ball micro (Kontes K-569001-
0219 or equivalent).
5.4 Micro syringe - 100 yL (Hamilton No. 84858 or equivalent).
5.5 Weighing pans, micro - approximately 1-cm diameter aluminum foil.
Purchase or fabricate from aluminum foil.
5.6 Boiling chips - approximately 10-40 mesh carborundum (A. H. Thomas
No. 1590-D30 or equivalent). Heat to 450°C for 5-10 hr. or extract
with methylene chloride.
5.7 Water bath - Heated, capable of temperature control (±2°C). The bath
should be used in a hood.
5.8 Balance - Analytical, capable of accurately weighing 0.0001 g.
5.9 Microbalance - Capable of accurately weighing to 0.001 mg (Mettler
model ME-30 or equivalent).
5.10 Sonifier - 375 watt, fitted with a 1/2-inch probe and a half-wave
extension, capable of pulsed operation at variable power settings.
(Heat Systems-Ultrasonics Sonicator Model W-375 with No. 200 1/2-inch
disrupter horn and No. 406-HW-050-T half-wave extender, or equivalent).
208
-------
5.11 Centrifuge - Capable of accommodating 200-mL glass centrifuge tubes.
5.12 pH meter and electrodes - Capable of accurately measuring pH to ±0.2
pH unit.
5.13 Spatula - Having a metal blade 1-2 cm in width.
5.14 Heat lamp - 250 watt reflector-type bulb (GE No. 250R-40/4 or equivalent)
in a heat-resistant fixture whose height above the sample may be
conveniently adjusted.
5.15 Gas chromatograph/mass spectrometer system
5.15.1 Gas chromatograph - An analytical system complete with a
temperature programmable gas chromatograph suitable for
splitless injection and all required accessories including
syringes, analytical columns, and gases.
5.15.2 Column - 30 m x 0.25 mm bonded-phase sillcone coated fused
silica capillary column (J W Scientific DB-5 or equivalent).
5.15.3 Mass spectrometer - Capable of scanning from 40 to 450 amu
every 1 second or less, utilizing 70 volts (nominal) electron
energy in the electron impact ionization mode and producing a
mass spectrum'which meets all required criteria when 50 ng of
decafIuorotr1phenylphosph1ne (DFTPP) is Injected through the
GC Inlet.
5.15.4 Data system - A computer system must be interfaced to the mass
spectrometer which allows the continuous acquisition and storage
on machine readable media of all mass spectra obtained throughout
the duration of the chromatographlc program. The computer
must have software that allows searching any GC/MS data file
for Ions of a specific mass and plotting such ion abundances
versus time or scan number. This type of plot 1s defined as
an Extracted Ion Current Profile (EICP). Software must also
be available that allows integrating the abundance in any EICP
between specified time or scan number limits.
5.16 Gel permeation chromatography system.
5.16.1 Chromatographic column - 600 mm x 25 mm I.D. glass column
fitted for upward flow operation.
5.16.2 Bio-Beads S-X8 - 80 g per column.
5.16.3 Pump - capable of constant flow of 0.1 to 5 mL/m1n at up to
100 psi.
5.16.4 Injector - with 5-mL loop
209
-------
5.16.6 Strip chart recorder.
5.16.5 Ultraviolet detector - 254 mm
6. Reagents
6.1 Reagent water - Reagent water is defined as water in which no
interference is observed at the method detection limit for each
compound of interest.
6.2 Potassium phosphate, tribasic (K3P04) - Granular (ACS).
6.3 Phosphate buffer, W, - 2.0 moles of N32HP04 and 2.0 moles of
dissolved in reagent water and diluted to 1000 mL. The solution is
very temperature sensitive; it must be checked carefully before using
and, if necessary, warmed to redissolve any crystals that may have formed.
6.4 Phosphoric acid (^04) - 85% aqueous solution (ACS).
6.5 Sodium sulfate, anhydrous (^504) - Powder (ACS).
6.6 Methylene chloride - Distilled-in-glass quality (Burdick and Jackson,
or equivalent).
6.7 Internal standards - D5-bromobenzene, Ds-naphthalene, Dig-biphenyl,
Dig-acenaphthene, D^g-phenanthrene, D^g-pyrene, D^-chrysene, and
Di2-benzo(a)pyrene.
6.7 Column performance standards - Ds-phenol, Ds-aniline, D5-nitrobenzene,
and D3-2,4-dinitrophenol.
6.8 Surrogate standards - Decafluorobiphenyl, 2-fluoroaniline, and
pentafluorophenol.
6.9 Decafluorotriphenylphosphine (DFTPP).
6.10 GPC calibration solution - Methylene chloride containing 100 mg of
corn oil, 20 mg of di-n-octyl phthalate, 3 mg of coronene, and 2 mg of
sulfur per 100 ml.
7. Calibration
7.1 A multiple internal standard calibration procedure as described by
Sauter, et alU) is used. To use this approach, the analyst must
select five or more internal standards that are similar in analytical
behavior to the compounds of Interest. The analyst must further
demonstrate that the measurements of the internal standard are not
affected by method or matrix interferences. Use the base peak ion as
the primary ion for quantification of the standards. If interferences
are noted, use the next most Intense ion as the secondary ion. The
Internal standards are added to all calibration standards and all
sample extracts analyzed by GC/MS. Column performance standards, and
210
-------
a mass spectrometer tuning standard are included in the internal
standard solution used.
7.1.1 A set of five or more internal standards is selected that will
permit all components of interest in a chromatogram to have
retention times of 0.80 to 1.20 relative to at least one of the
internal standards. The following internal standards are
recommended for general use: Ds-bromobenzene, Ds-naphthalene,
DiQ-biphenyl, DiQ-acenaphthene, Dio-phenanthrene, Dio-pyrene,
Di2-chrysene, and Di2-benzo(a)pyrene.
7.1.2 Representative acidic, basic, and polar neutral compounds are
added with the internal standards to assess the column performance
of the GC/MS system. The following column performance standards
are recommended for general use: Ds-phenol, Ds-aniline, 05-
nitrobenzene, and D3-2,4-dinitrophenol. These compounds can
also serve as internal standards if appropriate and the internal
standards recommended in Section 7.1.1 can serve as column
performance standards if appropriate.
7.1.3 Decafluorotriphenylphosphine (DFTPP) is used to tune the mass
spectrometer on each workshift.
7.1.4 Prepare the internal standard solution by dissolving in 50.0 ml
of methylene chloride 10.0 mg of each standard compound specified
in Sections 7.1.1, 7.1.2, and 7.1.3. The resulting solution
will contain each standard at a concentration of 200 yg/mL. A
solution containing 500 g/mL of each standard can be prepared
by using 5 percent benzene in methylene chloride as the solvent.
7.2 Prepare calibration standards at a minimum of three concentration
levels. Each calibration standard should contain each compound of
interest and each surrogate standard. Each calibration standard should
be mixed with an appropriate amount of the internal standard solution.
One of the calibration standards should be at a concentration of two
to five times the method detection limit, 1 to 10 g/mL; one should be
at a concentration near, but below the concentration that causes
saturation of the mass spectrometer; and, the third should be at a
concentration in the middle of this working range of the GC/MS system.
7.3 Analyze 1 to 2 pL of each calibration standard and tabulate the area
of the primary characteristic ion against concentration for each
compound including the surrogate compounds. Calculate response fac-
tors (RF) for each compound using equation 1.
Eq. 1 RF = (AsC1s)/(A1sCs)
Where:
As = Area of the characteristic ion for the compound to be
measured.
211
-------
A-jg » Area of the characteristic ion for the internal standard;
the internal standard chosen should be such that the
relative retention time of the compound is within the
range of 0.80 to 1.20 and as close as possible to 1.00.
G-JS = Concentration of the internal standard, (yg/mL).
Cs « Concentration of the compound to be measured (yg/mL).
If the RF value over the working range is constant (< 10% RSD), the RF
can be assumed to be invariant and the average RF can be used for
calculations. Alternatively, the results can be used to plot a
calibration curve of response ratios, As/A-jS, vs. RF.
7.4 The RF must be verified on each working day by the measurement of two
or more calibration standards, including one at the beginning of the
day and one at the end of the day. The response factors obtained for
the calibration standards analyzed immediately before and after a set
of samples must be within ± 20% of the response factor used for
quantification of the sample concentrations.
8. Quality Control
8.1 Each laboratory that uses this method is required to operate a formal
quality control program. The minimum requirements of this program
consist of an initial demonstration of laboratory capability and the
analysis of spiked samples as a continuing check on performance. The
laboratory is required to maintain performance records to define the
quality of data that are generated. Ongoing performance checks must
be compared with established performance criteria to determine if the
results of analyses are within the accuracy and precision limits
expected of the method.
8.1.1 Before performing any analyses, the analyst must demonstrate
the ability to generate acceptable accuracy and precision with
this method. This ability is established as described in
Section 8.2.
8.1.2 The laboratory must spike all samples including check samples
with surrogate standards to monitor continuing laboratory
performance. This procedure is described 1n Section 8.4.
8.2 To establish the ability to generate acceptable accuracy and precision,
the analyst must peform the following operations using a representative
sample as a check sample.
8.2.1 Analyze four aliquots of the unspiked check sample according
to the method beginning 1n Section 10.
8.2.2 For each compound to be measured, select a spike concentration
representative of twice the level found in the unspiked check
212
-------
sample or a level equal to 10 times the expected detection
limit, whichever is greater. Prepare a spiking solution by
dissolving the compounds in methylene chloride at the appropriate
- levels.
8.2.3 Spike a minimum of four aliquots of the check sample with the
spiking solution to achieve the selected spike concentrations.
Spike the samples after they have been transferred to centrifuge
tubes for extraction. Analyze the spiked aliquots according to
the method beginning in Section 10.
8.2.4 Calculate the average percent recovery, (R), and the standard
deviation of the percent recovery, (s), for all compounds and
surrogate standards. Background corrections must be made before
R and s calculations are performed. The average percent recovery
must be greater than 20 for all compounds to be measured and
greater than 60 for all surrogate compounds. The percent
relative standard deviation of the percent recovery, (s/R x
100), must be less than 20 for all compounds to be measured and
all surrogate compounds.
8.3 The analyst must calculate method performance criteria for each of the
surrogate standards.
8.3.1 Calculate upper and lower control limits for method performance
for each surrogate standard, using the values for R and s
calculated in Section 8.2.4:
Upper Control Limit (UCL) = R + 3 s
Lower Control Limit (LCL) * R - 3 s
The UCL and LCL can be used to construct control charts that
are useful in observing trends in performance.
8.3.2 For each surrogate standard, the laboratory must maintain a
record of the R and s values obtained for each surrogate standard
in each waste sample analyzed. An accuracy statement should be
prepared from these data and updated regularly.
8.4 The laboratory is required to spike all samples with the surrogate
standards to monitor spike recoveries. The spiking level used should
be that which will give a concentration, in the final extract used for
GC/MS analysis, that is equal to the concentration, of the internal
standard assuming a 100% recovery of the surrogate standards. For
unknown samples, the spiking level 1s determined by performing the
extraction steps in Section 10 on a separate aliquot of the sample and
calculating the amount of internal standard per gram of sample as
described in Section 10.8. If the recovery for any surrogate standard
does not fall within the control limits for method performance, the
results reported for that sample must be qualified as being outside of
control limits. The laboratory must monitor the frequency of data, so
qualified, to ensure that the frequency remains at or below 5%. Three
213
-------
surrogate standards, narrely decafluorobiphenyl, 2-fluoroaniline, and
pentafluorophenol, are recommended for general use to monitor recovery
of neutral, basic, and acidic components, respectively.
8.5 Before processing any samples, the analyst must demonstrate, through
the analysis of a process blank, that all glassware and reagent
interferences are under control. Each time a set of samples is
extracted, or there is a change in reagents, a process blank should be
analyzed to detenrine the level of laboratory contamination.
8.6 It is recommended that the laboratory adopt additional quality assurance
practices for use with this method. The specific practices that are
most productive depend upon the needs of the laboratory and the nature
of the samples. Field replicates may be analyzed to monitor the
precision of the sampling technique. Whenever possible, the laboratory
should perform analysis of standard reference materials and participate
in relevant performance evaluation studies.
8.7 The features that must be monitored for each GC/MS analysis run for
quality control purposes and for which performance criteria must be
met are as follows:
• Relative ion abundances of the mass spectrometer tuning compound
DFTPP
• Response factors of the internal standards and column performance
standards relative to DiQ-phenanthrene
• Relative retention times of column performance standards
relative to the closest internal standard
• Peak area intensity and absolute retention time of DiQ-phenanthrene.
9. Sample Collection. Preservation, and Handling
9.1 Grab samples must be collected in glass containers having Teflon-lined
screw caps. Sampling equipment must be free of oil and other potential
sources of contamination.
9.2 The samples must be refrigerated at 4°C from the time of collection
until extraction.
9.3 All samples must be extracted within 14 days of collection and completely
analyzed within 40 days of extraction.
10. Sample Extraction
10.1 The extraction procedure involves sonification of the sample with
methylene chloride, neutralization to pH 7, and the addition of
anhydrous sodium sulfate to remove the water. The amount of acid or
base required for the neutralization is determined by titration of
the sample. The particle size of all samples, except those comprised
214
-------
of nonporous Inorganic particles, should be reduced to less than 0.1
mm diameter before extraction. A glass mortar and pestle are recommended
for grinding the sample.
10.1.1 Thoroughly mix the sample to enable a representative sample
to be obtained. Weigh 3.0 g (wet weight) of sample Into a
200-mL centrifuge tube. Add 15 ml of methylene chloride and
15 ml of water.
10.1.2 Sonify the mixture for two minutes by inserting the sonlfier
horn 0.5 to 1.0 cm below the surface and using a power setting
of 5 and a 50 percent pulsed-duty cycle.
10.1.3 Transfer the contents of the centrifuge tube to a 400-mL
beaker using 50 ml of methylene chloride followed by 150 ml
of water as rinses.
10.1.4 Adjust the pH of the mixture to 7.0 ± 0.2 by titration with
0.4 M H3P04 or 0.4 M K3P04 using a pH meter. Record the
volume of acid or base required.
10.2 The extraction with methylene chloride is performed using a fresh
portion of the sample. Weigh 3.0 g (wet weight) of sample into a 200-
mL centrifuge tube. Add 15 ml of methylene chloride. Spike the
sample with surrogate standards as described in Section 8.4. Add
1.0 ml of 4 M phosphate buffer, and an amount of 4 M H3P04 or 4 M
K3P04 equal To one tenth of the pH 7 acid or base volume requirement
determined in Section 10.1.4. For example, if the acid requirement
1n Section 10.1.4 was 2.0 ml of 0.4 M ^04, the amount of 4 IN ^04
needed would be 0.2 ml.
10.3 Sonify the mixture for 1 minute by Inserting the sonifier horn 0.5 to
1.0 cm below the surface and using a power setting of 5 and a 50
percent pulsed-duty cycle. Cool the mixture In an ice bath or cold
water bath, if necessary, to maintain a temperature of 20 to 30°C.
Add 135 ml of methylene chloride, adjust the position of the sonlfier
horn to 0.5 to 1.0 cm below the surface and repeat the sonificatlon
for 1 minute. Some samples, especially those that contain much water,
may not disperse well in this step but will disperse after sodium
sulfate is added. Add, all at once, an amount of anhydrous sodium
sulfate powder equal to 15.0 g plus 3.0 g for each ml of the 4 M
H3P04 or 4 f4 KsP04 added in Section 10.2. Immediately cap the cen-
trifuge tube and shake vigorously for 1 minute. Insert the sonlfier
horn 0.5 to 1.0 cm below the surface and sonify for 2 minutes as
described above. Allow the mixture to stand until a clear super-
natant is obtained. Centrifuge if necessary to facilitate the phase
separation, filter the supernatant required for Sections 10.4, 10.5,
and 10.7 (at least 2 ml) through a 0.2 urn Teflon filter.
10.4 Estimate the total solvent extractable content (TSEC) of the sample
by determining the residue weight of an aliquot of the supernatant
215
-------
from Section 10.3 Transfer 0.1 ml of the supernatant to a tared
aluminum weighing dish, place the weighing dish under a heat lamp at
a distance of 8 cm from the lamp for one minute to allow the solvent
to evaporate, and then weigh on a microbalance. If the residue
weight of the 0.1-mL aliquot is less than 0.05 mg, concentrate 25 ml
of the supernatant to 1.0 ml and obtain a residue weight on 0.1 ml
of the concentrate. For the concentration step use a 25-mL evaporator
tube fitted with a micro Snyder column; add two boiling chips and
heat in a water bath at 60-65°C. Calculate the TSEC as milligrams
of residue per gram of sample using Equation 2 if concentration was
not required or Equation 3 if concentration was required.
Eq. 2. mg of Residue = Residue Wt«. mg. of 0.1 ml of Supernatant
g of Sample 0.002
Eq. 3. mg of Residue = Residue Wt.» mg. of 0.1 mL of Coned. Supernatant
g of Sample 0.05
10.5 If the TSEC of the sample (as determined in Section 10.4 is less than
50 mg/g, concentrate an aliquot of the supernatant that contains a
total of only 10 to 20 mg of residual material. For example, if the
TSEC is 44 mg/g, use a 20-mL aliquot of the supernatant, which will
contain 17.6 mg of residual material, or if the TSEC is 16 mg/g, use
a 50-mL aliquot of the supernatant, which will contairv 16.0 mg of
residual material. If the TSEC is less than 10 mg/g use 100 ml of
the supernatant. Perform the concentration by transferring the super-
natant aliquot to a K-D flask fitted into a 25-mL concentrator tube.
Add two boiling chips, attach a three-ball macro Snyder column to
the K-D flask, and concentrate the extract using a water bath at 60
to 65°C. Place the K-D apparatus in the water bath so that the
concentrator tube 1s about half Immersed in the water and the entire
rounded surface of the flask is bathed with water vapor. Adjust the
vertical position of the apparatus and the water temperature as
required to complete the concentration in 15 to 20 minutes. At the
proper rate of distillation the balls of the column actively chatter
but the chambers do not flood. When the liquid has reached an appar-
ent volume of 5 to 6 ml, remove the K-D apparatus from the water
bath and allow the solvent to drain for at least 5 minutes while
cooling. Remove the Snyder column and rinse the flask and the lower
joint of the flask Into the concentrator tube with methylene chloride
to bring the volume to 10.0 mL. Mix the concentrator tube contents
by inserting a stopper and Inverting several times.
10.6 Analyze the concentrate from Section 10.5 or, if the TSEC of the
sample 1s 50 mg/g or more, analyze the supernatant from Section 10.3
using gas chromatography. Use a 30-m x 0.25 mm bonded-phase silicone
coated fused silica capillary column under the chromatographic
conditions described in Section 13. Estimate the concentration factor
or dilution factor required to give the optimum concentration for the
subsequent GC/MS analysis. In general the optimum concentration will
be one 1n which the average peak height of the five largest peaks or
the height of an unresolved envelope of peaks is the same as that of
216
-------
an internal or external standard (e.g., phenanthrene) at a concentra-
tion of 50-100 ug/mL.
10.7 If the optimum concentration determined in Section 10.6 is 20 mg of
residual material per ml or less proceed to Section 10.8. If the
optimum concentration is greater than 20 mg of residual material per
ml and if the TSEC if greater than 50 mg/g, apply the GPC cleanup
procedure described in Section 11. For the GPC cleanup concentrate
90 ml of the supernatant from Section 10.3 or a portion of the
supernatant that contains a total of 600 mg of residual material
(whichever is the smaller volume). Use the concentration procedure
described in Section 10.5 and concentrate to a final volume of 15.0
ml. Stop the concentration prior to reaching 15.0 ml if any oily or
semisolid material separates out and dilute as necessary (up to a
maximum final volume equal to the volume of supernatant used) to
redissolve the material. (Disregard the presence of small amounts of
inorganic salts that may settle out).
10.8 Concentrate further or dilute as necessary an aliquot of the concen-
trate from Section 10.5 or an aliquot of the supernatant from Section
10.3, or if GPC cleanup was necessary, an aliquot of the concentrate
from Section 11.3 to obtain 1.0 ml of a solution having the optimum
concentration, as described in Section 10.6, for the GC/MS analysis.
If the aliquot needs to be diluted, dilute it to a volume of 1.0 ml
with methylene chloride. If the aliquot needs to be concentrated,
concentrate it to 1.0 ml as described in Section 10.4. Do not let
the volume in the concentrator tube go below 0.6 ml at any time.
Stop the concentration prior to reaching 1.0 ml if any oily or semi-
solid material separates out and dilute as necessary (up to a maximum
final volume of 10 ml) to redissolve the material. (Disregard the
presence of small amounts of inorganic salts that may settle out). Add
a volume of the internal standard solution that contains 50 ug each
of the internal standards, column performance standards, and DFTPP,
to 1.0 ml of the final concentrate and save for GC/MS analysis as
described in Section 13. Calculate the concentration in the original
sample that is represented by the internal standard using Equation 4
if an aliquot of the concentrate from Section 10.5 was used in Section
10.8. Equation 5 if an aliquot of the supernatant from Section 10.3
was used in Section 10.8, or Equation 6 if an aliquot of the GPC
concentrate from Section 11.3 was used in Section 10.8.
Eq 4 ug of Int. Std. = 50 x JJQ. x li x Final Vol.. ml
g of Sample 3 Vs(io.5) Vc(i0.8) 1
Eq. 5 ug of Int. Std. a 50 y 150 v Final Vol.. ml
g of Sample 3 Vs(io.8) 1
E 6 ug of Int. Std. _ 50 x 150 x ^F x Final Vol.. ml
M* g of Sample 3 Vs(io.7) VQPC(10.7) 1
217
-------
where:
Vs » Volume of supernatant from Section 10.3 used in
Sections 10.5, 10.8, or 10.7
vc(10 8) s Volume of concentrate from Section 10.5 used 1n
Section 10.8.
Vp(10.7) = Pi"3! volume of concentrate 1n Section 10.7.
VQPC s Volume of GPC concentrate from Section 11.3 used in
Section 10.8.
Use this calculated value for the quantification of individual compounds
as described in Section 15.2.
11. Cleanup Using Gel Permeation Chromatography
11.1 Prepare a 600 mm x 25 mm I.D. gel permeation Chromatography (GPC)
column using a slurry containing 80 g of Bio-Beads S-X8 that have
been allowed to swell 1n methylene chloride for at least 4 hours.
Prior to initial use, rinse the column with methylene chloride at 1
mL/m1n for 16 hours to remove any traces of contaminants. Calibrate
the system by injecting 5 ml of the GPC calibration solution, elut-
1ng with methylene chloride at 5 mL/min for 50 minutes and observing
the resultant UV detector trace. The column may be used indefini-
tely as long as no darkening or pressure increases occur and a column
efficiency of at least 500 theoretical plates is achieved. The
pressure should not be permitted to exceed 50 psi. Recalibrate the
system daily.
11.2 Inject a 5-mL aliquot of the concentrate from Section 10.7 onto the
GPC column and elute with methylene chloride at 5 ml_/m1n for 50
minutes. Discard the first fraction that elutes up to a retention
time represented by the minimum between the corn oil peak and the di-
n-octyl phthalate peak in the calibration run. Collect the next
fraction eluting up to a retention time represented by the minimum
between the coronene peak and the sulfur peak in the calibration
run. Apply the above GPC separation to a second 5-mL aliquot of the
concentrate from Section 10.7 and combine the fractions collected.
11.3 Concentrate the combined GPC fractions to 10.0 ml as described in
Section 10.5. Estimate the TSEC of the concentrate as described 1n
Section 10.4. Estimate the concentration factor or dilution factor
required to give the optimum concentration for the subsequent GC/MS
analysis as described in Section 10.6.
12. Daily GC/MS Performance Tests
12.1 At the beginning of each day that analyses are to be performed, the
GC/MS system must be checked to see that the tuning sensitivity and
overall performance of the system are acceptable. The quality
218
-------
control protocols required for the fused silica capillary column
GC/MS analyses are described in a separate report.(2)
12.2 The .foilowing instrumental parameters are required for all performance
tests and for all sample analyses:
Electron Energy - 70 volts (nominal)
Mass Range - 40 to 450 amu
Scan Time - 0.75 second per scan.
12.3 The mass spectrometer must be tuned to achieve all of the key ion
criteria for the mass spectrum of DFTPP(3) given in Table 1 when a
solution containing 50 yg/mL of DFTPP 1s injected into the GC/MS
system. If the DFTPP analysis meets the criteria in Table 1, the
sample analyses are considered valid. Otherwise, the analyst must
retune the instrument to meet the Table 1 criteria, followed by
reanalysis of the sample in question.
12.4 The electron multlpier of the mass spectrometer must be adjusted
such that the Injection of a solution containing 20 yg/mL of
phenanthrene or Djn-phenanthrene will give a response that 1s
approximately 20 times the detection limit but less than one-tenth
of the value that causes serious saturation of the mass spectrometer.
12.5 The sensitivity of the GC/MS system must be such that a response
factor of at least 0.05 relative to Dig-phenanthrene is obtained for
pentachlorophenol and 2,4-dinitroaniline when a solution containing
50 yg/mL of each component is injected.
13. GC/MS Analysis
13.1 Analyze the 1-mL concentrate from Section 10.8 by GC/MS using a 30 m
x 0.25 mm bonded-phase si 11 cone-coated fused silica capillary column.
The recommended GC operating conditions to be used are as follows:
Initial Column Temperature Hold - 30°C for 4 minutes
Column Temperature Program - 30-300°C at 8 degrees/min
Final Column Temperature Hold - 300°C for 10 minutes
Injector Temperature - 280°C
Transfer Line Temperature - 300°C
Injector - Grob-type, splitless
Sample Volume - 1 yL
Carrier Gas - Hydrogen (preferred) at 50 cm/sec or helium at
30 cm/sec :rr: —--
13.2 If the response for any ion exceeds the working range of the GC/MS
system, dilute the extract and reanalyze.
13.3 Perform all qualitative and quantitative measurements as described
1n Sections 14 and 15. When the extracts are not being used for
analyses, store the extracts at 4°C protected from light 1n screw
cap vials equipped with unpierced Teflon-Hned septa.
219
-------
14. Qualitative Identification
14.1 Obtain an EICP for the primary characteristic ion and at least two
other characteristic ions for each compound when practical. The
following criteria must be met to make a qualitative identification.
14.1.1 The characteristic ions for each compound of interest must
maximize in the same scan or within one scan of each other.
14.1.2 The retention time must fall within ± 15 seconds of the
retention time of the authentic compound..
14.1.3 The relative peak heights of the characteristic ions in the
EICPs must fall within ± 20% of the relative intensities of
these Ions 1n a reference mass spectrum.
15. Quantitative Determination
15.1 When a compound has been identified, the quantification of that
compound will be based on the Integrated abundance from the EICP of
the primary characteristic ion. In general, the primary characteristic
1on selected should be a relatively intense ion, as interference-
free as possible, and as close as possible in mass to the characteristic
ion of the internal standard used.
15.2 Use the internal standard technique for performing the quantification.
Calculate the concentration of each individual compound of interest
1n the sample using Equation 7.
Eq. 7 Concentration, pg/g - yifl of Int. Std. x ^s_ x 1
' Ka/a g of Sample A-js RF
where:
ug of Int. Std. s internal standard concentration factor
g of Sample calculated 1n Section 10.8
As = area of the primary characteristic ion of the
compound being quantified
Ais » area of the primary characteristic ion of the
Internal standard
RF « response factor of the compound being quantified
determined in Section 7.3; use a response factor
of 1.0 for the quantification of an unknown com-
pound.
15.2 Report results 1n micrograms per gram without correction for recovery
data. When duplicate and spiked samples are analyzed, report all
data obtained with the sample results.
220
-------
15.3 If the surrogate standard recovery falls outside the control limits
in Section 8.3, the data for all compounds in that sample must be
labeled as suspect.
TABLE 1. DFTPP KEY IONS AND ION ABUNDANCE CRITERIA
8338888388838333883838333838388883888888883383383883833833888888883888388383883
Mass Ion Abundance Criteria
51 30-60% of mass 198
68 less than 2% of mass 69
70 less than 2% of mass 69
127 40-60% of mass 198
197 less than 1% of mass 198
198 base peak, 100% relative abundance
199 5-9% of mass 198
275 10-30% of mass 198
365 greater than 1% of mass 198
441 present but less than mass 443
442 greater than 40% of mass 198
443 17-23% of mass 442
8833883888333888338883883388333838838333338883883883833338388883838538388388888
221
-------
16. References
1. Sauter, A. D., Betowski, L. D., Smith, T. R., Strickler, V. A., Beimer,
B. G., Colby, B. N., and Wilkinson, J. E., "Fused Silica Capillary
Column GC/MS for the Analysis of Priority Pollutants", J. High Resolut.
Chromatogr. Chromotogr. Commun., 4., 366 (1981).
2. "Quality Control Protocol for the Fused Silica Capillary Column Gas
Chromatography/Mass Spectrometry Determination of Semi volatile Organic
Compounds", Battelle Columus Laboratories revision of a September 1981
priority pollutant protocol prepared by Acurex Corporation for U.S.
Environmental Protection Agency, Environmental Monitoring and Systems
Laboratory, Las Vegas, NV 89114, November, 1981.
3. Eichelberger, J. W., Harris, L. E., and Budde, W. L., "Reference Compound
to Calibrate Ion Abundance Measurement in Gas Chromatography-Mass
Spectrometry Systems", Anal. Chem., 47. 995 (1975).
222
-------
APPENDIX B
METHOD FOR THE DETERMINATION OF VOLATILE
ORGANIC COMPOUNDS IN SOLID WASTES
223
-------
Battelle - Columbus Laboratories - November 10, 1981^
METHOD FOR THE DETERMINATION OF VOLATILE
ORGANIC COMPOUNDS IN SOLID WASTES
1. Scope and Application
1.1 This method covers the determination of volatile organic compounds in
a variety of solid waste matrices.
1.2 This method is applicable to nearly all types of samples, regardless
of water content, Including aqueous sludges, caustic liquors, acid
liquors, waste solvents, oily wastes, mousses, tars, fibrous waste,
polymeric emulsions, filter cakes, spent carbons, spent catalysts,
soils, and sediments.
1.3 The detection limit of the method for determining an individual
compound is estimated to be approximately 1 yg/g (wet weight). For
samples which contain more than 1 mg/g of total volatile material,
the detection limit should be proportionately higher.
1.4 The method is based upon a purge and trap, gas chromatographic/mass
spectrometric (GC/MS)'procedure.
1.5 This method is restricted to use by, or under the supervision of analysts
experienced in the use of purge and trap systems and gas chromatograph/
mass spectrometers and skilled in the interpretation of mass spectra.
2. Summary of Method
A portion of solid waste is dispersed in tetraglyme to dissolve the volatile
organic constituents. A portion of the tetraglyme solution 1s combined
with water 1n a specially designed purging chamber. An Inert gas 1s then
bubbled through the solution at ambient temperature and the volatile com-
ponents are efficiently transferred from the aqueous phase to the vapor
phase. The vapor is swept through a sorbent column where the volatile
components are trapped. After purging 1s completed, the sorbent column is
heated and backflushed with Inert gas to desorb the components onto a gas
chromatographic column. The gas chromatographic column 1s heated to elute
the components which are then detected with a mass spectrometer.U'2)
3. Interferences
3.1 Low molecular weight Impurities in tetraglyme can be volatilized dur-
ing the purging procedure. Thus, the tetraglyme employed in this
method must be purified before use and stabilized to prevent peroxide
formation as described 1n Section 6.2.
3.2 Impurities in the purge gas and organic compounds out-gassing from the
plumbing ahead of the trap account for the majority of contamination
224
-------
problems. The analytical system must be demonstrated to be free from
contamination under the conditions of the analysis by running laboratory
reagent blanks as described 1n Section 8.3. The use of non-TFE plastic
tubing, non-TFE thread sealants, or flow controllers with rubber
components 1n the purging device should be avoided.
3.3 Samples can be contaminated by diffusion of volatile organlcs
(particularly fluorocarbons and methylene chloride) through the septum
seal Into the sample during shipment and storage. A field reagent
blank prepared from reagent water and carried through the sampling and
handling protocol can serve as a check on such contamination.
3.4 Contamination by carry-over can occur whenever high-level and low-
level samples are analyzed sequentially. Whenever an unusually
concentrated sample is encountered, it should be followed by an analysis
of reagent water to check for cross contamination. After each use,
the purging chamber is cleaned as described in Section 11.4.3. The
trap and other parts of the system are also subject to contamination;
therefore, frequent additional heating and purging of the entire system
may be required.
4. Safety
4.1 The toxiclty or carcinogenlcity of each reagent used in this method
has not been precisely defined; however, each chemical compound should
be treated as a potential health hazard. From this viewpoint, exposure
to these chemicals must be minimized by whatever means available; this
warning is particularly important when handling reference compounds.
The laboratory is responsible for maintaining a current awareness file
of OSHA regulations regarding the safe handling of the chemicals used
in applying this method. A reference file of material data handling
sheets should also be made available to all personnel involved in the
chemical analysis. Additional references to laboratory safety are
available and have been 1dentif1ed(3-5) for the information of the
analyst.
4.2 The following volatile organic compounds which may be determined by
this method have been tentatively classified as known or suspected,
human or mammalian carcinogens: benzene, carbon tetrachloride, chloro-
form, and vinyl chloride. Whenever primary standards containing any
of these or other toxic compounds are prepared, the operation should
be performed 1n a hood. A NIOSH/MSHA approved toxic gas respirator
should be worn when the analyst handleshigh-concentrations of these
toxic compounds.
5. Apparatus
5.1 Sampling equipment, for discrete sampling.
5.1.1 V1al - 25 ml capacity or larger, equipped with a screw cap
(Pierce 13075 or equivalent). Detergent wash, rinse with tap
and distilled water, and dry for one hour at 105°C before use.
225
-------
5.1.2 Septum - Teflon-faced sillcone (Pierce 12722 or equivalent).
Detergent wash, rinse with tap and distilled water and dry at
,105°C for one hour before use.
5.2 Purge and trap device - The purge and trap device consists of three
separate pieces of equipment: the purging chamber, trap, and the
desorber. Several complete devices are now commercially available.
5.2.1 The purging chamber must be designed to accept 5-mL or 25-mL
samples with a water column at least 3 cm deep. The gaseous
head space between the water column and the trap must have a
total volume of less than 15 ml. The purge gas must pass
through the water column as finely divided bubbles. The purge
gas must be Introduced no more than 5 mm from the base of the
water column. The purging chamber, illustrated in Figure 1,
meets these design criteria.
5.2.2 The trap must be at least 25 cm long and have an inside diameter
of at least 2.5 mm. The trap must be packed to contain the
following minimum amounts of adsorbents: 1.0 cm of methyl
sillcone coated packing (Section 6.3.2), 15 cm of silica gel
(Section 6.3.3). The minimum specifications for the trap are
Illustrated in Figure 2.
5.2.3 The desorber must be capable of rapidly heating the trap to
180°C within 30 seconds. The polymer section of the trap should
not be heated higher than 180°C and the remaining sections
should not exceed 220°C. The desorber design, illustrated in
Figure 2, meets these criteria.
5.2.4 The purge and trap device may be assembled as a separate unit
or be coupled to a gas chromatograph as Illustrated in Figures
3 and 4.
5.3 Gas chromatograph/mass spectrometer system
5.3.1 Gas chromatograph - An analytical system complete with a
temperature programmable gas chromatograph and all required
accessories including syringes, analytical columns, and gases.
5.3.2 Column - 2 m x 2 mm ID stainless or glass, packed with 1 percent
SP-1000 on 60/80 mesh Carbopack B or equivalent.
5.3.3 Mass spectrometer - Capable of scanning from 40 to 250 amu
every 7 seconds or less, utilizing 70 volts (nominal) electron
energy in the electron impact ionizatlon mode and producing a
mass spectrum which meets all the criteria in Table 1 when 50
ng of 4-bromofluorobenzene (BFB) is injected through the GC inlet.
5.3.4 GC/MS interface - Any GC to MS interface that gives accceptable
calibration points at 50 ng per injection for each compound of
226
-------
interest and achieves acceptable tuning performance criteria
(see Section 10) may be used. GC to MS interfaces constructed
of all glass or glass-lined materials are recommended. Glass
. can be deactivated by silanizing with dichlorodimethylsilane.
5.3.5 Data system - A computer system must be interfaced to the mass
spectrometer which allows the continuous acquisition and storage
on machine readable media of all mass spectra obtained through-
out the duration of the chromatographic program. The computer
must have software that allows searching any GC/MS data file
for ions of a specific mass and plotting such ion abundances
versus time or scan number. This type of plot is defined as
an Extracted Ion Current Profile (EICP). Software must also
be available that allows integrating the abundance in any EICP
between specified time or scan number limits.
5.4 Sample Transfer Implements - Implements are required to transfer por-
tions of solid, semi-solid, and liquid wastes from sample containers
to laboratory glassware. The transfer must be accomplished rapidly to
avoid loss of volatile components during the transfer step. Liquids
may be transferred using a hypodermic syringe with a wide-bore needle
or no needle attached. Solids may be transferred using a conventional
laboratory spatula, spoon, or coring device. A coring device that is
suitable for handling some samples can be made by using a glass tubing
saw to cut away the closed end of the barrel of a glass hypodermic
syrings.
5.5 Syringes - 5-mL and 25-mL gas-tight with shut-off valve, equipped with
narrow-bore needle, at least 15 cm in length.
5.6 Micro syringes - 10-pL, 25-yL, 100-pL, 250-yL, 500-yL, and 1000-yL.
These syringes should be equipped with narrow-bore needles having a
length sufficient to extend from the sample inlet to within 1 cm of
the glass frit in the purging device (see Figure 1). The needle
length required will depend upon the dimensions of the purging device
employed.
5.7 Centrifuge tubes - 50-mL round bottom glass centrifuge tubes with
Teflon-lined screw caps. The tubes must be marked before use to show
an approximate 20-mL graduation (Kimble #45212) or equivalent).
5.8 Centrifuge - Capable of accommodating 50-mL glass tubes.
5.9 Bottle - 15-mL, screw-cap, Teflon cap-liner.
5.10 Balance - Analytical, capable of accurately weighing 0.0001 g.
5.11 Rotary evaporator - equipped with Teflon-coated seals (Buchi Rota-
vapor R-100, or equivalent).
5.12 Vacuum pump - mechanical, two stage.
227
-------
5.13 Sonifier - 375-watt, fitted with a 1/2-in probe and a half-wave
extension, capable of pulsed-operation at variable power settings
(Heat Systems - Ultrasonics Sonicator Model W-375 with 200 1/2-in
disrupter horn and 406-HW-050-T half wave extender, or equivalent).
6. Reagents
6.1 Reagent Water - Reagent water is defined as water in which no inter-
ferent is observed at the method detection limit for the compounds of
interest.
6.1.1 Reagent water may be generated by passing tap water through a
carbon filter bed containing about 500 g of activated carbon
(Calgon Corp., Filtrasorb-300, or equivalent).
6.1.2 A water purification system (Millipore Super-Q or equivalent)
may be used to generate reagent water.
6.1.3 Reagent water may also be prepared by boiling water for 15
minutes. Subsequently, while maintaining the temperature at
90°C, bubble a contaminant-free inert gas through the water
for one hour. While still hot, transfer the water to a narrow-
mouth screw-cap bottle and seal with a Teflon lined septum and
cap.
6.2 Reagent Tetraglyme. Reagent tetraglyme is defined as tetraglyme for
which no interferences are observed at the method detection limit
for the compounds of interest.
6.2.1 Tetraglyme (tetraethylene glycol dimethyl ether, Aldrich 17,
240-5 or equivalent) is purified by treatment at reduced
pressure in a rotary evaporator. The tetraglyme should have a
peroxide content of less than 5 ppm as indicated by EM Quant
Test Strips (available from Scientific Products Co., Catalog
No. P1126-8 and other suppliers). Peroxides may be removed by
passing the tetraglyme through a column of activated alumina.
The tetraglyme is placed is a round bottom flask equipped with
a standard taper joint, and the flask is affixed to a rotary
evaporator. The flask is immersed in a water bath at 90 to
100°C and vacuum is maintained at <10 mm Hg for at least
two hours using a two-stage mechanical pump. The vacuum
system is equipped with an all-glass trap, which is maintained
in a Dry Ice/methanol bath. Cool the tetraglyme to ambient
temperature and add 0.1 mg/mL of 2.6-di-tert-butyl-4-methyl-
phenol to prevent peroxide formation. Store the tetraglyme
in a tightly sealed screw-cap bottle in an area that is free
of solvent vapors.
6.2.2 In order to demonstrate that all interfering volatiles have
been removed from the tetraglyme, a reagent water/tetraglyme
blank must be analyzed.
228
-------
6.3 Trap Materials
6.3.1 2,6-Diphenylene oxide polymer - 60/80 mesh Tenax, chromato-
graphic grade or equivalent.
6.3.2 Methyl silicone packing - 3% OV-1 on 60/80 mesh Chromosorb-W
or equivalent.
6.3.3 Silica gel, Davison Chemical (35/60 mesh), grade-15 or equivalent,
6.4 Calibration standards; stock solutions - Stock solutions of calibration
standards may be prepared from pure standard materials or purchased
as certified solutions. Prepare stock standard solutions of individual
compounds in tetraglyme using assayed liquids or gases as appropriate.
6.4.2.1 Liquids - Using a 100-yL syringe, immediately add 2 drops of
assayed reference material to the flask, then reweigh. The
liquid must fall directly into the tetraglyme without
contacting the neck of the flask.
6.4.2.2 Gases - To prepare standards for any compounds that boil
below 30°C (e.g., bromomethane, chloroethane, chloromethane,
or vinyl chloride), fill a 5-mL valved gas-tight syringe
with a reference standard to the 5.0-mL mark. Lower the
needle to 5 mm above the tetraglyme meniscus. Slowly
introduce the reference standard above the surface of the
liquid. The heavy gas rapidly dissolves in the tetraglyme.
6.4.3 Reweigh, dilute to volume, stopper, then mix by inverting
the flask several times. Calculate the concentration in
micrograms per micro!iter from the net gain in weight. When
compound purity 1s assayed to be 96 percent or greater, the
weight may be used without correction to calculate the
concentration of the stock standard. Commercially prepared
stock standards in methanol may be used at any concentration
if they are certified by the manufacturer or by an independent
source.
6.4.4 Transfer the stock standard solution into a Teflon-sealed
screw-cap bottle. Store, with minimal headspace, at -10 to
-20°C and protect from light.
6.4.5 Prepare fresh standards weekly for gases or for reactive
compounds such as 2-chloroethyTv1nyl ether. All other
standards must be replaced after one month, or sooner if
comparison with check standards indicate a problem.
6.5 Calibration standards; secondary dilution solutions - Using stock
solutions described 1n Section 6.5, prepare secondary dilution
standards 1n tetraglyme that contain the compounds of interest, either
singly or mixed together. The secondary dilution standards should be
229
-------
prepared at concentrations such that the aqueous calibration solutions
prepared as described in Section 7.3.2 will bracket the working range
of the analytical system(2). Secondary dilution standards should be
stored with minimal headspace.
6.6 Surrogate standards - Surrogate standards are added to all samples
and calibration solutions; the compounds employed for this purpose
are 1,2-dibromotetrafluoroethane, bis(perfluoroisopropyl) ketone,
fluorobenzene, and m-bromobenzotr1fluoride added to each sample to
assess the effect of the sample matrix on recovery efficiency.
Prepare tetraglyme solutions of the surrogate standards using the
procedures described in Sections 6.5 and 6.6. The concentrations
prepared and the amount of solution added to each sample should be
those required to give an amount of each surrogate in the purging
device that 1s equal to the amount of each internal standard added
assuming a 100 percent recovery of the surrogate standards.
6.7 Internal standards - In this method, Internal standards are employed
during analysis of all samples and during all calibration procedures.
The analyst must select one or more internal standards that are
similar 1n analytical behavior to the compounds of interest. The
analyst must further demonstrate that the measurement of the internal
standard is not affected by method or matrix Interferences. Because
of these limitations, no internal standard can be suggested that is
applicable to all samples. However, for general use, 04-1,2-
dichloroethane, Ds-benzene, and Ds-ethylbenzene are recommended as
Internal standards covering a wide boiling point range.
6.8 4-Bromofluorobenzene (BFB) - BFB is added to the internal standard
solution to permit the mass spectrometer tuning for each GC/MS run to
be checked.
6.9 Internal standard solution - Using the procedures described in Sections
6.5 and 6.6, prepare a tetraglyme solution containing BFB and each
Internal standard at a concentration of 20.0 yg/mL.
6.10 Sodium monohydrogen phosphate - 2.0 JM in distilled water.
6.11 n-Nonane and n-dodecane, 98+ percent purity.
6.12 n-Hexadecane, d1stnied-1n-glass (Burdick and Jackson, or equivalent).
7. Calibration
7.1 Assemble a purge and trap device that meets the specifications in
Section 5.2 and connect the device to a GC/MS system. Condition the
trap overnight at 180°C by backflushing with an inert gas flow of at
least 20 mL/min. Prior to use, daily condition the trap for 10
minutes while backflushing at 180°C.
230
-------
7.2 Operate the gas chromatograph using the conditions described in
Section 11.4 and operate the mass spectrometer using the conditions
described in Section 10.2.
7.3 Calibration Procedure:
7.3.1 Conduct calibration procedures using a minimum of three
concentration levels for each calibration standard. One of
the concentration levels should be two to five times the method
detection limit (5 to 50 ng); one should be at a level near but
below the concentration that causes saturation of the mass
spectrometer; and the third should be at a level in the middle
of this working range of the GC/MS system.
7.3.2 Prepare a solution containing the required concentrations of
calibration standards, including surrogate standards, to 5 mL
of reagent water contained in a 5-mL gas-tight syringe having a
shut-off valve and fitted with a 15-cm narrow-bore needle.
Add a volume of the secondary dilution solution containing
appropriate concentrations of the calibration standards (see
Section 6.6) to the reagent water using a microsyringe. When
discharging the contents of the microsyringe be sure that the
end of the syringe needle is well beneath the surface of the
water. Similarly, add 12.5 yL of the internal standard solution
(see Section 6.10). Add the contents of the 5-mL syringe to
the purging device by inserting the needle as far as possible
through the septum of the purging device and discharging the
contents slowly.
7.3.3 Conduct the purge and analysis procedure as described in
Sections 11.4. Tabulate the area response of the primary
characteristic ion against concentration for each compound
including the internal standards. Calculate response factors
(RF) for each compound using Equation 1.
Eq. 1 RF = (AsCis)/(Ais Cs)
where:
As = Area of the primary characteristic ion for the
compound to be measured
Ais a Area of the primary characteristic ion of the
internal standard
Cis = Concentration of the internal standard
Cs = Concentration of the compound to be measured.
231
-------
The Internal standard selected for the calculatibn of the RF
of a compound and subsequent quantification of the compound is
generally the Internal standard that has a retention time
closest to that of the compound. It is assumed that a linear
- calibration plot will be obtained over the range of concentra-
tions used. If the RF value over the working range is a
constant (<10% RSD), the RF can be assumed to be invariant,
and the average RF can be used for calculations. Alterna-
tively, the results can be used to plot a calibration curve
of response ratios, As/A-fS, versus RF.
7.3,4 The RF must be verified on each working day by the measurement
of two or more calibration standards, Including one at the
beginning of the day and one at the end of the day. The
concentrations selected should be near the midpoint of the
working range. The response factors obtained for the calibra-
tion standards, analyzed immediately before and after a set
of samples, must be within ± 20% of the response factor used
for quantification of the sample concentrations.
8. Quality Control
8.1 Each laboratory that uses this method is required to operate a formal
quality control program. The minimum requirements of this program
consist of an initial demonstration of laboratory capability and the
analysis of spiked samples as a continuing check on performance.
The laboratory is required to maintain performance records to define
the quality of data that are generated. Ongoing performance checks
must be compared with established performance criteria to determine
if the results of analyses are v/ithin the accuracy and precision
limits expected of the method.
8.1.1 Before performing any analyses, the analyst must demonstrate
the ability to generate acceptable accuracy and precision with
this method. This ability 1s established as described in
Section 8.2.
8.1.2 The laboratory must spike all samples including check samples
with surrogate standards to continuously monitor laboratory
performance. This procedure Is described 1n Section 8.4.
8.2 To establish the ability to generate acceptable accuracy and precision,
the analyst must peform the following operations using a representative
sample as a check sample.
8.2.2 Analyze four aliquots of the unspiked check sample according
to the method in Section 11.
8.2.2 For each compound to be measured, select a spike concentration
representative of twice the level found in the unspiked check
sample or a level equal to 10 times the expected detection
232
-------
limit, whichever is greater. Prepare a spiking solution by
dissolving the compounds in tetraglyme at the appropriate levels.
8.2.3 Spike a minimum of four aliquots of the check sample with the
. spiking solution to achieve the selected spike concentrations.
Spike the samples by adding the spiking solution to the
tetraglyme used for the extraction. Analyze the spiked aliquots
according to the method in Section 11.
8.2.4 Calculate the average percent recovery, (R), and the standard
deviation of the percent recovery, (s), for all compounds and
surrogate standards. Background corrections must be made
before R and s calculations are performed. The average percent
recovery must be greater than 20 for all compounds to be
measured and greater than 60 for all surrogate compounds. The
percent relative standard deviation of the percent recovery,
(s/R x 100), must be less than 20 for all compounds to be
measured and all surrogate compounds.
8.3 The analyst must calculate method performance criteria for each of
the surrogate standards.
8.3.1 Calculate upper and lower control limits for method perfor-
mance for each surrogate standard, using the values for R and
s calculated in Section 8.2.4:
Upper Control Limit (UCL) - R + 3 s
Lower Control Limit (LCD = R - 3 s
The UCL and LCL can be used to construct control charts that
are useful in observing trends in performance.
8.3.2 For each surrogate standard, the laboratory must maintain a
record of the R and s values obtained for each surrogate stan-
dard in each waste sample analyzed. An accuracy statement
should be prepared from these data and updated regularly.
8.4 The laboratory is required to spike all samples with the surrogate
standards to monitor spike recoveries. The spiking level used should
be that which will give an amount in the purge apparatus that is
equal to the amount of the internal standard assuming a 100 percent
recovery of the surrogate standards. For unknown samples, the spiking
level is determined by performing the extraction described In Section
11.1 to estimate the major volatile compounds content and determining
the amount of extract to be analyzed as described in section 11.3.2.
If the recovery for any surrogate standard does not fall within the
control limits for method performance, the results reported for that
sample must be qualified as being outside of control limits. The
laboratory must monitor the frequency of data so qualified to ensure
that the frequency remains at or below 5 percent. Four surrogate
standards, namely l,2-d1bromod1fluorethane, bis(perfluoroisopropyl)
233
-------
ketone, fluorobenzene, and m-broirobenzotrifluoride, are recommended
for general use to rronitor recovery of volatile compounds varying in
volatility and polarity.
8.5 Each day, the analyst irust demonstrate through the analysis of a
process blank that all glassware and reagent interferences are under
control.
8.6 It is recomrended that the laboratory adopt additional quality
assurance practices for use with this method. The specific practices
that are irost productive depend upon the needs of the laboratory and
the nature of the sairples. Field replicates iray be analyzed to
rronitor the precision of the sampling technique. Whenever possible,
the laboratory should perfonr analysis of standard reference mater-
ials and participate in relevant performance evaluation studies.
9. Sarrple Storage
9.1 All samples must be stored in Teflon-lined screw cap vials. Sample
containers should be filled as completely as possible so as to obtain
minimum headspace or void space. Vials contaiing liquid sample should
be stored in an inverted position.
9.2 All samples must be refrigerated from the time of collection to the
time of analysis, and should be protected from light.
10. Daily GC/MS Performance Tests
10.1 At the beginning of each day that analyses are to be performed, the
GC/MS system must be checked tc
criteria are achieved for BFB({
GC/MS system must be checked to see that acceptable performance
10.2 The BFB performance test requires the following instrumental param-
eters.
Electron Energy: 70 volts (nominal)
Mass Range: 40 to 250 amu
Scan Time: to give at least 6 scans per peak but not to
exceed 7 seconds per scan.
10.3 Bleed BFB vapor into the mass spectrometer and tune the instrument
to achieve all the key ion criteria for the mass spectrum of BFB
given in Table 1.
10.4 BFB is included in the internal standard solution added to all
samples and calibration solutions. If any key ion abundance observed
for BFB during the analysis of a sample differs by more than 20%
from that observed during the analysis of the calibration solution,
then the analysis in question is considered invalid. The instrument
must then be retuned or the sample and/or calibration solution
reanalyzed until the above condition is met.
234
-------
10.5 The peak intensity of De-benzene is used to monitor the mass spec-
trometer sensitivity. The peak intensity for De-benzene observed
during each sample analysis must be between 0.7 and 1.4 times the
De-benzene peak intensity observed during the applicable calibra-
tion runs. For example, if the peak intensity of De-benzene observed
during calibration was 170,000 area counts, then each subsequent
sample or blank must give a D6-benzene peak intensity of between
120,000 and 240,000 area counts. If the De-benzene peak intensity
is outside of the specified range, the sample must be reanalyzed.
If the peak intensity is again outside the specified range, the
analyst must investigate the cause of the variability in sensitivity
and correct the problem.
11. Sample Extraction and Analysis
The analytical procedure involves extracting the sample with tetraethylene
glycol dimethyl ether (tetraglyme) and analyzing a portion of the extract
by a purge and trip GC/MS procedure. The amount of the extract to be
taken for the GC/MS analysis is based on the estimated major volatile
compounds content (MVCC) of the sample. The MVCC is estimated by extracting
the sample with n-hexadecane and analyzing the n-hexadecane extract using
gas chromatography.
11.1 The estimated MVCC is based on the area response relative to that
of n-nonane for the. five major components that elute prior to the
retention time of n-dodecane. The response factor and the retention
time of n-nonane are determined by analyzing a 2-pL aliquot of an
n-hexadecane solution containing 100 yg/mL of n-nonane.
11.1.1 The GC analyses are conducted using a flame ionization
detector and a 3 m x 2 mm I.D. glass column packed with
10% OV-101 on 100-120 mesh Chromosorb W-HP. The column
temperature is programmed from 80°C to 280°C at 8 degrees
per minute and held at 280°C for 10 minutes.
11.1.2 Determine the area response for n-nonane and divide by 100
to obtain the area response factor. Record the retention
time of n-nonane.
11.1.3 Add 1.0 gm of sample to 20 ml of n-hexadecane and 2 mL of
0.5 M. NagHP04 contained in a 50-mL glass centrifuge tube
and cap securely with a Teflon-lined screw cap. Shake the
mixture vigorously for one minute. If the sample does not
disperse during the shaking process, sonify the mixture
for two minutes by inserting the sonifier horn 1-2 cm
below the surface of the hexadecane and using a power
setting of 5 and a 25 percent pulsed duty cycle. Cap the
tube tightly and allow the mixture to stand until a clear
supernatant is obtained. Centrifuge if necessary to
facilitate phase separation.
235
-------
11.1.4 Analyze a 2-pL aliquot of the n-hexadecane supernatant using
the conditions described in Section 11.1.1. Determine the
total area response for the five major components that elute
prior to the retention time of n-nonane and subtract any
areas for the same five components that appear in an n-hexa-
decane blank. Using the 11.1.2, calculate the MVCC using
Equation 2.
E,. 2 MVCC . "VCAR ! . HVCARb1 k
n-Nonane Area Response Factor
where:
MVCC = major volatile compounds content of the
sample in yg/g.
MVCARsampie = major volatile compounds area response
obtained for the sample
MVCARfciank = major volatile compounds area response
11.2 The transfer of an aliquot of the sample for extraction with
tetraglyme should be made as quickly as possible to minimize loss of
volatiles from the sample.
11.2.1 To a 50-mL glass centrifuge tube with Teflon-lined cap, add
40 ml of reagent tetraglyme. Add an appropriate volume of
the surrogate standard solution; insert the needle of the
syringe approximately 4 cm below the surface of the tetra-
glyme during the addition. Weigh the capped centrifuge
tube and tetraglyme on an analytical balance.
11.2.2 Using an appropriate implement (see Section 5.4), transfer
approximately 2 gm of sample to the tetraglyme in the
centrifuge tube in such a fashion that the sample is dis-
solved in or submerged in the tetraglyme as quickly as
possible. Take care not to touch the sample-transfer imple-
ment to the tetraglyme. Recap the centrifuge tube immediately
and weigh on an analytical balance to determine an accurate
sample weight.
11.2.3 Disperse the sample by vigorous agitation for one minute.
The mixture may be agitated manually or with the aid of a
vortex-mixer. If the sample does not disperse during this
process, sonify the mixture for two minutes by inserting the
sonifier horn 1-2 cm below the surface of the tetraglyme and
using a power setting of 5 and a 25 percent pulsed duty
cycle. Cap the tube tightly and allow the mixture to stand
until a clear supernatant is obtained as the sample extract.
Centrifuge if necessary to facilitate phase separation.
236
-------
11.2.4 The sample extract may be stored for future analytical needs.
If storage 1s desired, transfer the solution to a 10 ml
screw cap vial with teflon cap Uner. Store at -10 to -208C,
and protect from light.
11.3 Reagent water, Internal standard solution, and the sample extract
are added to a purging chamber that 1s conected to the purge and
trap device and that has been flushed with helium during a seven-
minute trap reconditioning step (see Section 11.4.4). The additions
are made using an appropriate size syringe equipped with a 15-CM
narrow-bore needle. Insert the needle through the septum of the
purging device as far as possible.
11.3.1 Add 5.0 ml of reagent water to which 12.5 uL of the internal
standard solution has been added (see Section 6.9) to the
purging device. Insert the needle of the microsyringe
approximately 4 cm below the surface of the water during the
addition of the Internal standard solution.
11.3.2 Inject an aliquot of the sample extract from Section 11.2.3
below the surface of the water in the purging device. The
total quantity of the five major volatile compounds injected
should not exceed approximately 2 ug. If the major volatile
compound content (MVCC) of the sample as determined in
Section 11.1.4 is 200 ug/9 or less use a 200-uL aliquot of
the sample aliquot. If the MVCC is greater than 200 ug/g, use
an aliquot of the sample extract that contains 2 to 4 ug of
the five major volatile compounds. Dilute the sample extract
with tetraglyme as necessary to prepare a diluted extract
that contains 10 to 400 ug/mL of the five major volatile
compounds. The concentration (in pg/mt) of the five major
volatile compounds 1n the sample extract can be calculated
by dividing the MVCC by 20 times the dilution factor (DF).
The volume (in uL) of the aliquot of the diluted extract to
be taken can be calculated by dividing 20,OOOxDF by the MVCC
and multiplying by 2 to 4. For example, if the MVCC is
14,000 ug/g and a DF of 10 is used, 20,000 DF/MVCC = 14.3;
multiplying by 2 to 4 given 28.6 to 57.2; therefore, a 40-uL
aliquot (representing 0.0002 g of original sample) would be
an appropriate volume of the diluted extract to take for
analysis. If the major volatile compounds are expected to
be halogenated compounds, use only one fifth of the amount
of the diluted extract determined above. If the MVCC is
less than 200 ug/g and greaterjsens1t1vity 1s desired, use a
large purging chamber containing 25-mL of reagent water and
use a 1.0 mL aliquot of the sample extract. The volume of
diluted extract to be taken for analysis can be used to
determine the appropriate volume of surrogate standard
solution to be used in Section 11.2.1. The appropriate
volume of surrogate standard solution to add to the 2-gram
sample can be calculated using Equation 4. Equation 4 is
derived from Equation 3.
237
-------
Eq. 3 V c = Vn ., x Ve x DF
^ ss 0.25 Va
where:
VSs = Appropriate volume, uL, of surrogate standard
solution to add to a 2-gram sample.
Vo.25 = Volume of surrogate standard solution that
contains 0.25 yg.
Ve a Volume of tetraglyme extract, yL, from
original extraction of a 2-gram sample.
DF = Dilution factor.
Va = Volume of diluted extract to be taken for analysis
Since Vo.25 equals 0.25 yg divided by the concentration,
yg/yL, of the surrogate standard solution and Ve:B40,000 yl_,
Equation 4 can be obtained.
Ea. 4 V = 0.25 v 40.000 DF = 10.000 DF
T SS " " " -- "
where:
-ss va css
Concentration, yg/yL, of the surrogate standard
solution.
11.4 The sample in the purging device is purged with helium to
transfer the volatile components to the trap. The trap is
then heated to desorb the volatile components which are
swept by the helium carrier gas onto the GC column for analysis.
11.4.1 Adjust the gas (helium) flow rate to 40 ± 2 mL/min.
Set the purging device to purge and purge the sample
for 11.0 ± 0.1 minutes at ambient temperature.
11.4.2 At the conclusion of the purge time, adjust the
device to the desorb mode, and begin the GC/MS
analysis and data acquisition using the following GC
operating conditions:
Column - 6 ft x 2 mm ID glass column packed
with 1% SP-1000 on Carbopack B (60-80
mesh)
Temperature - Isothermal at 45°C for 3 minutes, then
increased at 8°C per minute to 240°C,
and maintained at 240°C for 15 minutes.
238
-------
Concurrently, introduce the trapped materials to the
GC column by rapidly heating the trap to 180°C while
backflushing the trap with helium at a flow rate of
30 mL/min for 4 minutes. If this rapid heating
requirement cannot be met, the GC column must be
used as a secondary trap by cooling 1t to 30°C or
lower during the 4-minute desorb step and starting
the GC program after the desorb step.
11.4.3 Return the purge trap device to the purge mode and
continue acquiring GC/MS data.
11.4.4 Allow the trap to cool for 8 minutes. Replace the
purging chamber with a clean purging chamber fitted
with a new septum. The purging chamber is cleaned
after each use by sequential washing with acetone,
methanol, detergent solution, and distilled water
and drying at 105°C.
11.4.5 Purge the trap at ambient temperature for 4 minutes.
Recondition the trap by heating 1t to 180°C. Do not
allow the trap temperature to exceed 180°C, since
the sorption/desorption is adversely affected by
heating the trap to higher temperatures. After
heating the trap for approximately seven minutes,
turn off the trap heater. When cool, the trap is
ready for use with the next sample.
11.5 If the response for any ion exceeds the working range of the
system, repeat the analysis using a correspondingly smaller
aliquot of the sample extract described in Section 11.2.3.
12. Qualitative Identification
12.1 Obtain an EICP for the primary characteristic ion and at least two
other characteristic ions for each compound when practical. The
following criteria must be met to make a qualitative identification.
12.1.1 The characteristic Ions of each compound of Interest must
maximize in the same scan or within one scan of each other.
12.1.2 The retention time must fall within ± 30 seconds of the
retention time of the authentic compound.
12.1.3 The relative peak heights of the characteristic ions in the
EICPSs must fall within ± 20% of the relative intensities of
these ions in a reference mass spectrum.
239
-------
13. Quantitative Determination "
13.1 When a compound has been Identified, the quantification of that
compound will be based on the Integrated abundance from the EICP of
the.primary characteristic 1on. In general, the primary characteristic
1on selected should be a relatively Intense 1on, as Interference-
free as possible, and as close as possible 1n mass to the characteristic
1on of the Internal standard used.
13.1.1 Use the Internal standard technique for performing the
quantification. Calculate the concentration of each Individual
compound of Interest 1n the sample using Equation 3.
Eq. 3 Concentration ug/g = (AsCfs)/(Ais)(RF)
where:
As = Area of the primary characteristic 1on of the
compound to be measured.
A-}S - Area of the primary characteristic 1on of the
Internal standard.
C-fS = Concentration of the Internal standard 1n ug/g.
RF = Response factor of the compound being quantified
determined 1n Section 7.3.3.
13.2 Report results 1n mlcrograms per gram without correction for recovering
data. When duplicate and spiked samples are analyzed, report all
data obtained with the sample results.
13.3 If the surrogate standard recovery falls outside the control limits
of Section 8.3, the data for all compounds 1n that sample must be
labelled as suspect.
14. References
1. Bellar, T. A., and J. J. Uchtenberg, Journal American Waterworks
Association, 66:739, 1974.
2. Bellar, T. A., and J. J. Llchtenberg, "Semi-Automated Headspace Analysis
of Drinking Waters and Industrial Waters for Purgeable Volatile Organic
Compounds," Measurement of Organic Pollutants 1n Water and Wastewater, C.
E. Van Hall, editor, American Society for Testing and Materials, Philadelphia,
PA. Special Technical Publications 686, 1978.
3. "Carcinogens—Working With Carcinogens", Department of Health, Education,
and Welfare, Public Health Service, Center for Disease Control, National
Institute of Occupational Safety and Health, Publications Mo. 77-206, Aug.
1977.
240
-------
4. "OSHA Safety and Health Standards, General Industry", (29CFR 1910),
Occupational Safety and Health Administration, OSHA 2206, (Revised, January
1976).
5. "Safety 1n Academic Chemistry Laboratories", American Chemical Society
Publication, Committee on Chemical Safety, 3rd Edition, 1979.
6. Budde, W. L., and J. W., Eichelberger. "Performance Tests for the Evaluation
of Computerized Gas Chromatography/Mass Spectrometry Equipment and
Laboratories", EPA-600/4-80-025, U.S. Environmental Protection Agency,
Environmental Monitoring and Support Laboratory, Cincinnati, OH 45268, p.
16, April 1980.
TABLE 1. BFB KEY ION ABUNDANCE CRITERIA
=======================3========================
Mass Ion Abundance Criteria
50 15 to 40% of mass 95
75 30 to 60% of mass 95
95 Base Peak, 100% Relative Abundance
96 5 to 9% of mass 95
173 < 2% of mass 174
174 > 50% of mass 95
175 5 to 9% of mass 174
176 > 95% but < 100% of mass 174
177 5 to 9% of mass 176
==3=3===========================================
241
-------
OPTIONAL
FOAM
TRAP
EXIT K IN.
0.0.
14MM 0. 0.
INLET K IN.
0.0.
VilN.
0. 0. EXIT
SAMPLE INLET WITH
SILICONE SEPTUM
10MM GLASS FRIT
MEDIUM POROSITY
_10MM. 0. 0. 1/16 IN. 0.0.
\ySTAINLESS ST
INLET
1/4 IN. 0.. D.
13X MOLECULAR
SIEVE PURGE
GAS FILTER
PURGE GAS
FLOW
CONTROL
Figure 1. Purging chamber.
242
-------
PACKING PROCEDURE
GLASS
WOOL
GRADE 15 ._.
SILICA GEL80"
TENAX 15CVI
3% OV-1 1CM
GLASS 5MM
WOOL
TRAP INLET
CONSTRUCTION
COMPRESSION
. FITTING NUT
14FT. 7A/FOOT
RESISTANCE WIRE
WRAPPED SOLID
THERMOCOUPLE/
CONTROLLER
SENSOR
(ELECTRONIC
•J TEMPERATURE
-^CONTROL
AND
PYROMETER
TUBING 25O!
0.105 IN. I.D.
0.125 IN. O.D.
STAINLESS STE
Figure 2. Trap packings and construction to include desorb capability,
243
-------
CARRIER GAS PLOW CONTROL LIQUID INJECTION PORTS
PRESSURE REGULATOR "^"pl /, . *- COLUMN OVEN
\ l(\ ' ' /^rTSTlTIILIin-nr" CONFIRMATORY COLUMN
__j. .—ii \ /Ttvj eim. u j u i x TQ 06TlcTog
**- ANALYTICAL COLUMN
_\._»\ .jCrBW
^iXf^
OPTIONAL 4-*orr COLUMN
SCUCT10N VALVt
t
L srnl V
H[
'HEATER CONTROL
PURGING
o evict.
Now.-AU. UNE5 IETWEEN
TRAP AND GC
SHOUL3 BE HEATED
TO
Figure 3. Schematic 'of purge and trap device - purge mode.
CARRIES GAS
FLOW CONTROL
PKESSURS
REGULATOR.
PURGE GAS v _,
FLOW CONTROLQ,
UQUIO INJECTION PORTS
13X MOLECULAR
SIEVE FILTER
COLUMN OVEN
_^CONFIRMATORY COLUMN
COLUMN
IONAL 4W>ORT COLUMN
SaECTION VALVE
S^ORT TRAP INLET
VALVE J RESISTANCE WIRE HEATB,
CONTROL
Notr.
ALL LINES BETWEEN
PURGING TRAP AND GC
DEVICE SHOULD BE HEATED
TO 95* C.
Figure 4. Schematic of purge and trap device - desorb mode.
244
-------
APPENDIX C
QUALITY CONTROL PROTOCOL
245
-------
QC Protocol for Fused Silica Capillary Columns
The QC Protocol presented here was developed by Acurex under the Direction
of Mr. Drew Sau'ter, EPA-LV and modified by Battelle for application to this
interlaboratory study. This protocol for the use of fused silica capillary columns
must be followed when the method for the determination of semivolatile organic
compounds is used. No deviations from this protocol are permitted.
246
-------
QUALITY CONTROL PROTOCOL FOR THE FUSED SILICA
CAPILLARY COLUMN GAS CHROMATOGRAPHY/MASS SPECTROMETRY
DETERMINATION OF SEMIVOLATILE ORGANIC COMPOUNDS
Revision of September 1981 Acurex Corporation Protocol
For
• Andrew D. Sauter
Environmental Protection Agency
Environmental Monitoring and Systems Laboratory
Las Vegas, Nevada 89114
By
Battelle Columbus Laboratories
Analytical Chemistry Section
247
-------
1. SUMMARY
This protocol describes the use of fused silica capillary columns (FSCC)
for analysis of the semivolatile organic compounds by gas chromatography/mass
spectrometry (GC/MS). The protocol gives column specifications, installation
and operating details, and requirements for system calibration and retention
time reproducibility. The quality control is based on multiple internal
standards for column performance evaluation and tuning evaluation and on an
initial multilevel calibration followed by regular single level calibration
verifications.
The successive sections of this protocol describe the purpose of the
protocol, the equipment requirements, column installation, operating conditions,
sample injection, qualitative and quantitative identification, initial
verification, ongoing QC activities, and documentation.
2. PURPOSE AND DESCRIPTION OF THE QC PROTOCOL
This protocol was written to support the efforts of the Environmental
Monitoring Support Laboratory-Las Vegas (EMSL-LV) to standardize the use of
FSCC. The operating conditions and the steps to be followed in implementing
this quality control protocol are given in sufficient detail to obtain reproduc-
ible results. After installation of the FSCC, the GC/MS system is initially
calibrated at three levels. Calibration checks at the middle level are repeated
after the analysis of each set of up to eight samples or at least once daily,
whichever is most frequent. Specifications for the calibration check results
are given to determine the acceptability of the sample analysis data. Multiple
internal standards are employed to verify system performance during each
analysis.
3. EQUIPMENT REQUIREMENTS
3.1 Gas Chromatograph
Capable of isobarlc splitless capillary column injection with quartz
or glass-lined injection port; fused silica capillary column;
temperature program with 30°C hold; and direct interface to mass
spectrometer. The end of the capillary column is Inserted as close
as possible to the ion source of the mass spectrometer, however
without Intercepting the electron beam.
3.1.1 Column
Fused silica capillary column, 30 m x 0.25 mm ID, coated with
0.25 micron thickness of DB-5 bonded phase sllicone. (J&W
Scientific DB-5-30N, or equivalent).
3.1.2 Carrier Gas Supply
Hydrogen or helium can be used as carrier gases. If hydrogen
is employed, safe handling practices must be used.
248
-------
Hydrogen and helium must be oxygen-free. Ultrapure helium
(99.999 percent) can be used directly from the cylinder,
however, it is desirable to install a de-oxo cartridge to
remove residual oxygen in case of leaks.
3.2 Mass Spectrometer
A mass spectrometer with electron impact ion source is required. The
mass spectrometer should be interfaced with the gas chromatograph via
a heated capillary transfer line. The mass spectrometer should be
interfaced to a data system that allows acquisition and storage of
repetitive scan data throughout the GC/MS run.
3.3 Computer and Programs
A computer system that allows searching of the GC/MS data for display
of extracted ion current profiles and integration of peak areas is
required. Furthermore, the capability to generate response factors
from the analysis of standards to accumulate calibration data to
standardize system performance and assure reproducible results is
desirable. If this capability is not available, these data must be
produced manually for every calibration compound analyzed to permit
an updating of response factors to meet the specifications of the
protocol.
4. COLUMN INSTALLATION AND FLOW ADJUSTMENT
4.1 Column Installation
4.1.1 Slide fitting nuts and ferrules onto each end of the column,
(Vespel ferrules treated with graphite are recommended). Cut
2 to 3 cm from each end of the column by scoring the circum-
ference with a sharp edged file or razor blade, and break off
the column ends with the fingers.
4.1.2 Insert one end of the column into the injection port to a
position 0.5 to 2.0 cm from the tip of a fully inserted syringe
needle. Tighten the nut finger tight, then turn it one full
turn with a wrench.
4.1.3 Insert the outlet end of the column into the ion source and
position it. Tighten the nut finger tight, then turn it one
full turn with a wrench.
4.1.4 Leak test the column inlet and outlet fittings with argon
using the mass spectrometer as a leak detector.
4.1.5 Position the column inside the oven such that any contact
between the oven wall and the column is avoided.
249
-------
4.2 Flow Adjustment
4.2.1 Set the column head pressure at 10 psig and the oven temperature
to 30°C.
4.2.2 Measure the linear gas velocity by injecting a small quantity
of air and recording the transit time through the column using
the mass spectrometer as the detector for air (at m/e 28).
4.2.3 Adjust the column head pressure until the linear gas velocity
is 40-60 cm per second at 30°C column temperature if hydrogen
is used as the carrier gas. If helium is used as the carrier
gas the desired flow is 25 to 30 cm per second. For wider bore
columns the gas flow is determined by the vacuum system. Typ-
ical linear gas velocities for wider bore columns (0.32 mm ID)
are 60 to 80 cm per second for hydrogen and 30 to 40 cm per
second for helium.
5. OPERATING CONDITIONS
5.1 Split/Splitless Injector
Dimensions: 2 to 3 mm ID
Material: Quartz or glass
Temperature: • 280°C
Flow rates:
• Sweep flow: 10 mL/min
Split flow: 35 mL/min
Stop flow for 0.5 min at start of program
5.2 Column Temperature Program
Initial temperature: 30°C
Initial hold: 4 min
Rate: 8°C/min
Final temperature: 300°C
Final hold: 10 min
5.3 Mass Spectrometer
Mass range: m/e 40-450
Scan time: 0.75 ± 0.25 second
Electron energy: 70 eV
Source temperature: 280 to 300°C
initiate acquisition of data with start of GC program
6. SAMPLE PREPARATION
6.1 Analytical Standards
Prepare mixed analytical standards from commercially available stock
250
-------
standards that are certified by the manufacturer or an independent
source. The purity of the compounds 1n the original stock standards
should be noted; if greater than 98 percent no correction is needed
estimate the concentration of the specified analyte. Working
solutions containing the compounds of Interest at the three levels,
20, 500, and 200 micrograms per mill 1 liter, should be prepared by
diluting aliquots of the original stock solutions. All analytical
standards should contain the internal standards and system perfor-
mance standards at concentrations of 50 micrograms per mllliliter.
6.2 Internal Standards
The Internal standards (I.S.) which have been used successfully are
bromobenzene-ds, naphtha!ene-dg, biphenyl-dig, acenaphthene-d^g.
phenanthrene-dig, pyrene-dig, chrysene-djg* anc' benzo(a)pyrene-di2-
Other I.S. may also be employed, however, the considerations given in
Reference 1 must be addressed. A solution of the eight I.S. is
prepared in methylene chloride-benzene (9:1) at a concentration of
1.0 milligram per milliliter each.
Each I.S. is spiked Into each sample at 50 micrograms per milllHter.
for complex samples higher levels of I.S. (e.g., 100 micrograms per
mill 1 liter) may be utilized to facilitate detection and identifica-
tion, but care must be taken to avoid saturation.
6.3 System Performance Standards
The standards which have been used successfully to evaluate column
performance are phenol-ds, aniline-d5, nitrobenzene-ds, and 2,4-
d1nitrophenol-d3. DecafTuorotriphenylphosphine (DFTPP) is used to
check the mass spectrometer tuning. These standards are added to the
internal standard mixture and spiked into each sample at the same
concentration as the internal standards. The system performance
standards are also employed as quantification internal standards with
/ respect to the unlabeled counterparts.
7. SAMPLE INJECTION
For the spHt/splitless technique, the flows are turned off for 0.5 minute,
and at 0.1 minute a one-microliter sample is injected quickly. The needle is
removed immediately. Thorough rinsing of the syringe following sample injection
is required. __
The performance of the splitless injection 1s the most critical step in
the procedure. Poor injections lead to discrimination of either the low or the
high boiling compounds and consequently to poor analytical precision. A correct
injection technique should give reproducible response factors for each I.S.
relative to phenanthrene-djo-
251
-------
8. QUALITATIVE AND QUANTITATIVE IDENTIFICATION
8.1 Relative Retention Time Criteria
The use of eight I.S. allows for the accurate measurement of relative
retention times (RRT) for all compounds. The values in Table Q-l
were obtained with new columns under the conditions noted. As columns
are shortened or are modified by the injected components the RRTs
change. Variation in columns, oven temperatures and temperature
programs can also have effects on RRTs. The RRTs should be very
reproducible over several days. The identification criteria require
that the retention time of a listed compound in a unknown agree within
±8 seconds of that in the daily standard.
8.2 Mass Spectral Criteria
All fragment ions with intensities greater than 10 percent of the
base peak in the mass spectrum of the standard must be present in the
mass spectrum of the compound in the sample. Relative abundances
must agree within ±20 percent. Furthermore, the absolute abundance
of all characteristic ions must co-maximize within ± one; i.e. three
consecutive scans.
8.3 Determination of Relative Response Factors
8.3.1 Analyze the 20, 50, and 200 micrograms per milliliter standards
as described in Section 9.3 at least once each week.
8.3.2 Calculate relative response factors of each compound in Table
Q-l using the appropriate Internal standard as indicated in
the table. The relative response factor (RRF) is defined as
follows:
RRF =
where:
Ax a area counts of the quantification 1on of com-
pound X
ATS = area counts of the quantification ion of
internal standard
Wx * amount of compound injected
W-jS * amount of internal standard injected
For the purpose of reproducibility, all compounds must be
quantified using the specific quantification ions listed in
Table Q-l. These Ions have been chosen to minimize Interfer-
ences as well as to minimize mass differences from the Internal
standard. If serious interference occurs at the primary quan-
tification ion, a secondary 1on may be used as described in
Method 625.(2)
252
-------
TABLE Q-lr. RETENTION AND RESPONSE DATA FOR SEMIVOLATILE COMPOUNDS
No.
4
1
2
3
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
48
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Compound _
*Bromobenzene-D5
4-Methylpyridine
1,4-Dichlorobutane
Anisole
Bromobenzene
2,4-Dimethylpr1dine
4-Chlorotoluene
2-Fluoroaniline
Benzaldehyde
Thiophenol
Pentachloroethane
Aniline-D5
Aniline
Phenol -05
Phenol
Bis(2-chloroethyl) ether
2-Chlorophenol
Pentafluorophenol
1,2,4-Trimethylbenzene
2,4,6-Trimethylpyridine
1 ,4-D1 chl orobenzene
Benzyl chloride
*Naphthalene-D8
Benzyl alcohol
1 , 2-Di chl orobenzene
2-Methyl phenol
N-Methyl aniline
Acetophenone
4-Methyl phenol
4-Methyl aniline
Hexachloroethane
Nltrobenzene-05
Nitrobenzene
N,N-D1methyl aniline
4-t -Butyl pyri dine
1,2, 4, 5-Tetramethyl benzene
Decaf luorobiphenyl
Quan
Mass
82
93
55
108
77
107
126
111
106
110
119
98
93
99
94
93
128
184
105
121
146
91
136
108
146
108
106
105
108
107
117
128
123
120
120
119
265
RT
(sec)
639
540
608
622
642
643
681
684
685
691
707
708
710
712
714
723
272
729
733
754
758
760
977
787
789
812
829
831
837
838
841
"855"
858
862
874
894
897
============
RRT(REF)
1.000(4
0.840(4
0.951(4)
0.974(4
1.004(4
1.007(4)
1.066(4)
1.072(4)
1.074(4)
1.082(4)
1.107(4
1.109(4
1.113(4
1.117(4
1.121(4)
1.134(4)
1.141(4)
1.145(4)
1.150 4
1.186(4)
1.186(4)
1.190(48)
1.000 48)
0.805 48)
0.897 48)
0.831(48)
0.848(48)
0.850(48)
0.856 48)
0.857(48)
0.860(48)
0.879(48)
0.878(48)
0.882 48)
0.895(48)
0.914(48)
0.918(48)
RF(REF)
0.59(118)
0.95(4)**
1.16(4)
1.06(4)
1.04(4)
0.98(4)**
0.56(4)
1.35(4)
0.62(4)
0.16(4)
0.35(4)
0.84(118)
1.09(12)
0.76(118)
0.92(14)
1.10(4)
0.99(4)
0.29(4)
1.82 4)
0.89 4)**
1.32(4)
1.90(4)
1.39 118)
0.29 48)
0.31 48)
0.34(48)
0.65(48)
0.50(31)
0.36(48)
0.51(48)
0.16(48)
0.24(118)
1.03(31)
0.58 48
0.50(48)
0.72(48)
0.11(48)
(continued)
253
-------
TABLE Q-lr. (Continued)
===============================================================================
No.
37
38
39
40
41
42
43
44
45
46
47
49
50
51
52
53
54
55
56
57
58
59
87
60
61
62
63
64
65
66
67
68
69
70
71
72
73
Compound '
4-Chl orobenzal dehyde
2-N1trophenol
Benzal chloride
2 ,4-D1methyl phenol
2-N1trotoluene
Tetralin
Propiophenone
2, 6-Dimethyl aniline.
Benzoic acid
2,6-Dichlorophenol
1 ,2 ,4-Tri chl orobenzene
Naphthalene
4-Chl orophenol
4-Chloroaniline
2, 4-Di chl orophenol
Hexachloropropene
4-Nitrotoluene
Hexachl orobutadi ene
Benzotrichloride
l-Chloro-4-nitrobenzene
Quinoline
B1s(2-chloroethoxy) ethane
*Acenaphthene-010
4-Chl oro-3-methyl phenol
4-t-Butyl phenol
4-Bromoaniline
2-Methy 1 naphthal ene
4-Ch 1 oro-2-methy 1 ani 1 i ne
4-Chlorobenzoic acid
1,2, 4, 5-Tetrachl orobenzene
2, 4, 6-Trichl orophenol
2,4,5-Trichlorophenol
Acetanilide
Biphenyl-DlO
2-Chloronaphthalene
Biphenyl
1-Chl oronaphthal ene
Quan
Mass
139
139
125
122
120
104
105
121
122
162
180
128
128
127
162
213
137
225
159
111
129
63
164
142
135
171
142
141
139
216
196
196
93
164
162
154
162
RT
(sec)
907
947
922
928
951
953
955
957
957
938
971
981
985
997
998
1003
1015
1007
1025
1042
1045
1079
1302
1090
1095
1104
1107
1107
1141
1145
1165
1172
1192
1192
1195
1196
1201
RRT(REF)
RF(REF)
0.928(48) 0.21(48)
0.934(48) 0.19(48)
0.931(48) 0.43(48)
0.950(48) 0.34(48)
0.973(48) 0.24(48)
0.974(48) 0.53(48)
0.976(48) 0.98(48)
0.979(48) 0.40(48)
0.980(48) 0.17(48)**
0.979(48) 0.30(48)
0.993(48
1.003(48
1.009(48
0.32(48)
0.99(48)
0.46(48)
1.021(48) 0.45(48)
1.022(48) 0.30(48)
1.026(48) 0.18(48)
1.039(48 0.21 48
1.041(48) 0.14(48)
1.054(48) 0.27(48)
1.072(48) 0.17(48)
1.077(48) 0.70(48)
1.117(48
0.45(48)
1.000(87) 0.74(118)
0.837(87) 0.45(87)
0.841(87) 1.15(87)
0.848(87) 0.42(87)
0.850 87
0.850(87
1.21(87)
0.37(87)
0.876(87) 0.17(87)**
0.879(87) 0.58(87)
0.895(87
0.900(87
0.36(87)
0.35(87)
0.016(87) 0.87(87)
0.916(87) 1.53(118)
0.918(87) 1.49(87)
0.019(87) 1.31(87)
0.922(87) 0.93(87)
(continued)
254
-------
No. Compound -
TABLE Q-lr. (Continued)
:=========s================
Quan RT
Mass (sec)
=========================
RRT(REF) RF(REF)
74
75
76
77
78
79
80
81
82
83
84
85
86
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
118
106
107
108
109
4-Methylqu1nol1ne
2-Ethyl naphthalene
Phenyl ether
2-N1troanil1ne
4-Bromobenzo1c add
3,4-D1chloroan1l1ne
2-Chloro-4-n1trophenol
Acenaphthylene
2, 3-D1methyl naphthalene
1,3-Di nitrobenzene
Dimethyl phthalate
2,6-D1n1trotoluene
3-N1troan1l1ne
Acenaphthene
2,4-D1n1trophenol-D3
2 ,4-Di -t-butyl phenol
2,4-D1n1trophenol
2-Naphthol
2,6-01 -t-butyl -4 -methyl phenol
Dlbenzofuran
4-N1trophenol
Pentachl orobenzene
2,4-D1n1trotoluene
l-Am1nonaphthalene
2 ,4 , 5-Tr1 chl oroanl 1 1 ne
2, 4-D1n1trochl orobenzene
Fluorene
4-Chlorophenyl phenyl ether
4-Chloro-2-n1troan1l1ne
4-N1troan1l1ne
2-Methyl -4 ,6-d1 nltrophenol
*Phenanthrene-D10
D1 phenyl ami ne
Azobenzene
Tr1flural1n
2-Chloro-4-n1troan1 1 1 ne
143
141
170
138
200
161
173
152
156
168
163
165
138
154
187
191
184
144
205
168
109
250
165
143
195
202
166
204
172
138
198
188
169
182
306
172
1203
1213
1218
1222
1232
1246
1254
1259
1260
1257
1261
1278
1300
1307
1318
1318
1319
1323
1325
1338
1339
1342
1349
1354
1336
1376
1402
1405
1413
1416
-1423—
1576
1430
1435
1474
1499
0.924(87) 0.81(87)
0.932(87) 1.15(87)
0.935(87) 0.85(87
0.939(87) 0.36(87)
0.946(87) 0.10(87)**
0.957(87) 0.48(87)
0.964(87) 0.15(87)
0.967(87
0.967(87
0.63(87)
0.95(87)
0.972(87) 0.25(87)
0.974(87) 1.30(87)
0.981(87) 0.27(87)
0.999(87 0.31(87)
1.004(87) 0.95(87)
1.012(87) 0.05(118)
1.012(87) 1.36(87)
1.013(87
1.016(87
1.00(89)
0.42(87)
1.017(87) 1.09(87)
1.027(87) 1.60(87)
1.030(87) 0.13(87)
1.031 87
1.036 87
1.040 87
1.042 87
1.057(87
0.44(87
0.32 87
0.77(87
0.42(87)
0.12(87)
1.076(87) 1.09(87)
1.079(87) 0.42(87)
1.085(87) 0.24 87
1.087(87) 0.26(87)
1.093(87) 0.14(87)
1.000(118) 1.00 118
0.907 118 0.68 118
0.910(118) 0.27 118
0.935 118) 0.15 118
0.951(118) 0.13(118)
(continued)
255
-------
TABLE Q-lr. (Continued)
:======================================================
No.
110
111
112
113
114
115
116
117
119
120
121
122
123
124
125
126
134
127
128
129
130
131
132
133
135
143
136
137
138
139
140
141
142
144
145
146
150
Compound
4-Hydroxyb1 pheny 1
Hexachlorobenzene
2 ,6-D1 chl oro-4-n1 troanl 1 1 ne
2,4-01 chl orophenoxyacetlc add
4-Am1nob1phenyl
Pentachlorophenol
3,3'-D1chlorobiphenyl
Pentachloronltrobenzene
Phenanthrene
4 ,4 ' -D1 chl orobl pheny 1
Anthracene
Acr1d1ne
Phenanthridlne
Carbazole
Decaf 1 uorotrl pheny 1 phosphi ne
Quan
Mass
170
284
206
162
169
266
222
237
178
222
178
179
179
167
198
2,4,5-Trichlorophenoxyacetic add 196
Decaf 1 uorobl pheny 1
4-Chl orobenzal dehyde
2-N1trophenol
Benzal chloride
Anthraqulnone
Fluoranthene
212
149
183
292
208
202
2,2',4,4',6,6'-Hexachlorob1pheny1 360
2-Methyl anthraqul none
Pyrene
*Chrysene-D12
4, 4 '-DDE
4,4 '-ODD
D1 ethyl stilbestrol
4, 4 '-DDT
Tr1 pheny 1 phosphate
Benzo(a)aTthracene
Methoxychlor
3,3'-D1chlorobenz1dine
Chrysene
B1s(2-ethylhexyl) phthalate
*Benzo(a)pyrene-D12
222
202
240
246
235
268
235
326
228
227
252
228
149
264
RT
(sec)
1507
1519
1524
1535
1546
1534
1560
1570
1580
1580
1588
1597
1615
1621
1659
1681
1841
1706
1715
1716
1732
1803
1824
1834
1844
2074
1879
1940
1979
1995
2020
2071
2073
2074
2079
2100
2318
RRT(REF)
0.956(118)
0.964(118)
0.967(118
0.974(118)
0.981(118)
0.986(118)
0.990(118)
0.996(118)
1.002(118)
1.002(118)
1.008(118)
1.013(118)
1.025(118)
1.028(118)
1.052(118)
• 1.066(118)
1.000(134)
0.926(134)
0.931(134)
0.932(134)
0.940(134)
0.979(134)
0.991 134
0.996(134)
1.001(134)
1.000(143)
0.905(143)
0.935(143)
0.954(143)
0.961(143)
0.974(143)
0.998(143)
0.999 143
1.000(143)
1.002(143)
1.012(143)
1;000(150)
RF(REF)
0.70(118)
0.23(118
0.09 118
0.02(118)**
0.27(118)
0.11(118)
0.40(118)
0.07(118)**
1.00(118)
0.42(118)
1.01(118)
0.61 118
0.67 118
0.84 118
0.10(118)
0.01(118)**
0.94(118
1.59(134)
0.11(134)
0.23(134)
0.29(134)
0.98(134)
0.17(134)
0.25(134)
0.97(134)
0.64(118)
0.31(143)
0.47(143)**
0.08(143)
0.35(143)
0.20(143)
1.00(143)
0.68 143
0.18(143)
1.00(143)
1.00(143)
0.52(118)
(continued)
256
-------
TABLE Q-lr. (Continued)
No. Compound
Quan
Mass
RT
(sec)
RRT(REF) RF(REF)
147 Tri-(tolyl) phosphate
148 Bis(2-ethylhexyl) sebacate
149 Benzo(k)fluoranthene
151 Hexachlorophene
152 Benzo(a)pyrene
153 Dlbenzocarbazole
154 Indeno(l,2,3-cd)pyrene
155 Dibenzo(a,h)anthracene
156 Benzo(g,h,i)perylene
0.958(150) 0.20(150)
0.967(150) 0.55(150)
0.977(150) 1.05(150)
368 2222
185 2243
252 2266
196 2319
252 2323
267 2521
276 2571
278 2568
276 2639 1.157(150) 0.70(150)
1.000(150)
1.002(150)
1.099(150)
1.124 150
1.127(150)
0.03(150)**
0.99(150)
0.54(150)
0.80 150
0.60(150)
================================3=33333=======3333=333===============3333=3=33=
Compounds used as Internal standards.
**Compounds which may give poor responses.
8.3.3 Calculate the relative response factors of each of the eight
internal standards, except phenanthrene-dio» and each of the
four column performance standards using phenanthrene-djo as
the internal standard.
8.3.4 From the weekly three level standards data (20, 50, 200 ng
injected) calculate the correlation coefficients of the area
ratio, A-I/A.JS, for each of the counterparts of the eight
labeled internal standards. A correlation coefficient of at
least 0.9990 for each of the eight compounds ensures that the
dynamic range of the method is being achieved.
9. INITIAL VERIFICATION
9.1 Ion Abundance Calibration
Prior to any sample analyses it is necessary to establish standard
mass spectral abundance criteria through the analysis of decaf1uoro-
triphenylphosphine (DFTPP). The ion abundance criteria of Method 625
(see Table Q-2) must be met for the analysis of 50 nanograms of DFTPP
introduced via the GC. Compliance with these requirements may be
determined by examination of the DFTPP spectrum obtained from a daily
calibration check standard.
257
-------
TABLE Q-2. DFTPP KEY IONS AND ION ABUNDANCE CRITERIA
333333333333333333333333333333333333B3S33333S333S333333333S:
Mass Ion Abundance Criteria
51 30-60% of mass 198
68 less than 2% of mass 69
70 less than 2% of mass 69
127 40-60% of mass 198
197 less than 1% of mass 198
198 base peak, 100% relative abundance
199 5-9% of mass 198
275 10-30% of mass 198
365 greater than 1% of mass 198
441 present but less than mass 443
442 greater than 40% of mass 198
443 17-23% of mass 442
•3333333333333333333333333333333333333333333333333333333383
9.2 System Sensitivity and Chromatographic Performance
The desired analytical range for an FSCC GC/MS method is 5 to 500
nanograms Injected. It is difficult to determine amounts within this
range for all compounds because of the large variation in the response
factors of the extractable semi volatile compound. It is critical to
set the system sensitivity so that the highest level (500 ng) of high-
response compounds (such as naphthalene, dimethyl phthalate; and
phenanthene) is slightly saturated. Experience with fused silica
columns has shown that control of the peak area of phenanthrene-dio
is an adequate method for controlling the sensitivity and dynamic
range of the method.
Prior to starting the weekly multilevel calibration procedure, the
overall system sensitivity should be set by injecting phenanthrene
alone or in a mixture. The instrument sensitivity can vary as a
function of multiplier, column, injector and source interface char-
acteristics. The optimum electron multiplier setting is that setting
which will yield approximately 50 X the detection limit but less
than one-fourth of the value where serious saturation occurs for 50
nanograms of phenanthrene. This value must be determined by exper-
ience for each instrument and instrumental configuration. (For a
Finnigan 4000 with an INCOS data system an appropriate sensitivity
258
-------
setting has been found to give 50,000 to 125,000 area counts for 50
nanograms of phenanthrene or phenanthrene-djo).
The chromatographic performance is checked by the on-column injection
of fifty nanograms each of 2,6-d1ch1oro-4-nitroaniline and penta-
chlorophenol. This check can be done concurrently with the tuning
and sensitivity checks, or with the calibration check analysis (see
Section 9.3). The 2,6-dichloro-4-nitroaniline and pentachlorophenol
must both have response factors relative to phenanthrene-d^Q of
greater than 0.05.
9.3 Multilevel Calibration
Analyze mixed standards which contain 20, 50, and 200 yg/mL of the
compounds of interest. The standards should also contain the surro-
gate spiking compounds at the same levels. Spike each calibration
standard with all of the labeled internal standards and column perfor-
mance standards. Inject 1 yL of each standard and acquire a complete
run. Process the data for each compound at the three concentration
levels. The following criteria must be met:
• The relative retention times of each compound in each run should
agree to within 0.05 relative retention time units. Late eluting
compounds usually,will exhibit better agreement.
• The relative response factors should not fall below 50 percent of
those given in Table Q-l.
t All compounds in Table Q-l except those whose response factors are
starred must be detectable at the 20 nanogram level.
• Relative response factors of all compounds in Table Q-l except
those whose response factors are starred should have a relative
standard deviation of <25 percent over the range tested.
• The correlation coefficient for the three-point curve of Ion
current ratio versus amount (i.e. Ax/A-jS versus Wx) must be 0.999
for the labeled/unlabeled analyte pairs. This coefficient must be
calculated and reported for the eight unlabeled analogs of the I.S.
(bromobenzene, naphthalene, blphenyl, acenaphthene, phenanthrene,
pyrene, chrysene, and benzo(a)pyrene).
• The relative response factors for the~Tnternal standards and the
column performance standards using phenanthrene-dio as the internal
standard should not fall below 50 percent of those given in Table
Q-l. These relative response factors can be used to assess the
injector discriminaton and the acidity, basicity, and polarity of
the GC column.
t The relative response factors for the unlabeled analogues of the
eight internal standards, with correction for the purity of the
259
-------
labeled compounds must be 1.0 ± 0.2 at each level, i.e. 20, 50,
and 200 ng.
* The percent relative standard deviations for the relative ion
abundances of m/e 51, 69, 127, 275, and 442 of DFTPP must each be
less than 20 percent for all runs that follow the daily calibration
check standards.
10. ONGOING QC ACTIVITIES
10.1 System Calibration
Samples may be analyzed upon successful completion of the initial QC
activity. A calibration check analysis at the 50 yg/mL level must
be run before the analysis of each set of up to eight samples. A
three level calibration at 20, 50, and 200 ug/mL must be run at
least once a week. The system must be restandardized whenever the
specifications in 9.3 are not met and after any major system main-
tenance such as new column installation or source cleaning.
10.2 Run Verification
During or after each data acquisition, the presence of all eight
I.S. must be verified. The area counts of each internal standard
should be recorded for each standard and sample analyzed. The
percent deviation of the integrated ion current for each internal
standard in the samples run during any one week should remain within
one standard deviation. The area counts for each internal standard
relative to phenanthrene-diQ should not vary by more than a factor
of two from the value obtained in the most recent standard analysis.
Failure to meet this criterion will require minor system maintenance.
Breaking off one foot of the column or cleaning the injector sleeve
will often improve sensitivity for the late eluting compounds;
repositioning the front end of the column will often improve front-
end performance. Poor injection technique can also lead to variable
I.S. ratios.
Each analytical run must also be checked for saturation. The level
at which an individual compound will saturate the detection system
is a function of the overall system sensitivity and the mass spectral
characteristics of that compound. The initial method calibration
(Section 9.3) requires that the system should not be saturated for
high response compounds at 200 micrograms per milliliter. An estimate
of the safe upper limit for any compound can be obtained by dividing
200 by the response factor for that compound. For example, nitro-
benzene, with a relative response factor of 0.25 can be reliably
quantified to 200 divided by 0.25 or 800 micrograms per milliliter.
If more than five compounds in any sample exceed the analytical
range, that sample must be diluted, the internal standard concentra-
tion readjusted, and the sample reinjected.
260
-------
10.3 Dally Calibration Check
The tuning of the Instrument must be checked for each run by
measuring the relative abundances of the DFTPP Ions as described In
Section 9.1. The chromatography and sensitivity checks must also be
successfully completed (Section 9.2). The relative response factors
and relative retention times obtained for each calibration check at
50 jig/ml must meet the specifications of Section 9.3.
If these criteria are met, then the RRF values for all compounds are
added to the 11st of those obtained previously. The RRF values used
for quantification are based on an average of the current dally cali-
bration check standards and those obtained from the two previous
calibration check runs providing the calibration check criteria are
met. If the calibration check criteria are not met, major instrument
repair may be needed. Prior to conducting major repairs make sure
that the standard has not deteriorated or that the system is not
saturated.
10.4 Special Cautions
Skill is required 1n using FSCC GC/MS routinely. Improper sample
Injection technique can lead to poor precision evidenced by vola-
tility discrimination and component cross contamination. The use
of improper solvents can lead to peak splitting for early eluting
components. The sensitivity of the method is very dependent on the
correct positioning of the front end of the column in the injector
and the back end of the column 1n the ion source.
All laboratories using this protocol should be aware of the diffi-
culties and have access to technical advice for troubleshooting.
11. DOCUMENTATION
The following QC documentation is required as support for the analytical
data obtained using this protocol.
• GC/MS Run Log
- A log of all GC/MS runs Including the area counts and absolute
retention times for phenanthrene-dio must be reported using a form
shown as Table R-l.
• Relative Ion Abundance Calibration
- The relative ion abundances for DFTPP in each run must be tabulated
using a form shown as Table R-2.
. t Calibration Checks ,,,.,,,,,
- All RRT and RRF data obtained from multilevel and single level
calibration checks must be reported using forms shown as Tables
R-3 and R-4.
261
-------
TABLE R-l. GC/MS RUN LOG FOR SEMIVOLATILE COMPOUND ANALYSES
UnN*.
«RSO
OtM
KT«(
O»wo1
•A__^A
••m^v
C>«n.
fen* IS
262
-------
TABLE R-2. RELATIVE ABUNDANCES OF DFTPP IONS IN SEMIVOLATILE COMPOUND ANALYSES
NunN*.
•1
70
127
1M
1M
441
442
443
263
-------
TABLE R-3. RELATIVE RETENTIOM TIMES OF SEMIVOLATILE COMPOUNDS FROM CALIBRATION RUNS
ro
-------
TABLE R-4. RESPONSE FACTORS OF SEMIVOLATILE COMPOUNDS FROM CALIBRATION RUNS
«MO
ro
en
en
-------
t Chromatography Checks
- All RT and RRF data obtained for the Internal standards and column
performance standards for each run must be tabulated using forms
such-as Tables R-5 and R-6.
- Hard copies must be produced for the extracted ion current profiles
for 2,6-dichloro-4-n1troan1line, (m/e 206) pentachlorophenol (m/e 266),
and phenanthrene-djo (m/e 188) from each calibration run at the 50
yg/mL level. These profiles should include the peak area counts shown
in Figure 1.
t Data transcription is the responsibility of the participating
laboratory and is subject to audit by Battelle and EMSL-LV.
12. References
1. Sauter, A. D., Betowski, L. D., Smith, T. R., Strickler, V. A., Beimer,
B. G., Colby, B. N., and Wilkinson, J. E., "Fused Silica Capillary
Column GC/MS for the Analysis of Priority Pollutants", J. High Resolut.
Chromatogr. Chromatogr, Commun., j4, 366 (1981).
2. Eichelberger, J. W., Harris, L. E., and Budde, W. L., "Reference
Compound to Calibrate Ion Abundance Measurement in Gas Chromatography-
Mass Spectrometry Systems", Anal. Chem., 47, 995 (1975).
266
-------
TABLE R-5. RETENTION TIMES OF SEMIVOLATILE INTERNAL STANDARDS
ID,
ftMMtten Tinw »f (Mm Imtnul Stwdvd. wwndt
Ml
ONP Phvt PPTPf
Chrv ft*
An.
M •DB
Ohfy
DfTff •
Reproduced from
best available copy.
267
-------
TABLE R-6. RESPONSE FACTORS OF SEMIVOLATILE INTERNAL STANDARDS
HlMlM*.
OM» OrtFf Pyr Owy
AM «0|-AiiWM
Of Iff-
268
Reproduced from
best available copy,
-------
RIC + MASS CHROmTOGRAMS DATA: 13468782265 II
11/82/8121:42:00 CAL1: 134515CAL42 13
SAMPLE: INTEFLftB WLIDAT10N STANDARDS
RANGE: C 484,3650 LABEL: H 5, 2.8 CUAH: A 3, 1.8 BASE: U 28, 3
Zlul
SCANS 2868 TO 2145
Reproduced from Ji»1^
best available copy, ^fl^
168.8-
183.
266.
ro
C7I
to
1.7-
266_
468.3-
RIC_
26
25
IO>JeV.
9I§*2' ." 18560.
/ \ 188.656
dlQ-Phenanthrene 1 \ * 8.588
1 1 — i , . , . t . , ^"1 ~~l • i
2833
1152.
5074. n.9
A
/ \ 206.662
I / \ 2.6-Dichloro-A-nltroaniline ± 8.560
• i • i • i • i • i • i • i
2074
313.
^ " 313.
A
1 Pencachlorophenol / \ * t^'.WQ
1 / ^
1 2107'
86689.
™ 7T 86912'
oa.o »33 2062 31434. . / \ X ™*
"g 14536. 17200. 131697. A/ \ A 123662
J^>" 61488. 84624. A / 1 \ / \ ^f
4U83. Aj/\ J/NV, _j_V V ^XV 1 Vx V y ^.
1 • i *^"T • • i • i • I ' I m ' ^" 1
W0 2020 2640 2obO 26£0 2100 2120 2146 SCAM
:00 25:15 25:30 25:45 26:09 26:15 26:30 26:45 TIME
Figure 1. Chromatography/sensitivity check.
-------
APPENDIX D
DESCRIPTION OF STANDARD SOLUTIONS
270
-------
DESCRIPTION OF STANDARD SOLUTIONS
33338333333333333333333S333333333S:3 33333333333333333333333333333333
Semi volatile Calibration Solution A
jail compounds at 0.1 mg/ml in methylene chloride)
Aromatic Halocarbons
3,3' -Di chlorobi phenyl
4,4'-D1chlorobi phenyl
2,2'-4,4'-Tetrachlorobiphenyl
2,2l-4,4',6,6'-Hexachlorobiphenyl
Aromatic Hydrocarbons
Benzo(a)anthracene
Benzo(k)f1uoranthene•
Benzo(g,h,i)perylene
Indeno(l ,2,3-cd)pyrene
Chrysene
Di benzo(a,h)anthracene
Benzo(a)pyrene
Amines
1,2,7,8-Di benzocarbazole
Semivolatile Calibration solution B
(all compounds at 1.0 mg/ml in benzene)
Aromatic Hydrocarbons
Naphthalene
1,2,4-Trimethylbenzene
1,2,4,5-Tetramethylbenzene
Biphenyl
Acenaphthylene
Acenaphthene
2-Methylnaphthalene
2-Ethylnaphthalene
2,3-Dimethylnaphthalene
1,2,3,4-Tetrahydronaphthalene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Amines
Aniline
4-Chloroaniline
4-Bromoaniline
2-Nitroaniline
3,4-Dichloroaniline
2,4,5-Tr1chloroani1i ne
3-Nitroaniline
4-Chloro-2-methylani1i ne
4-Nitroaniline
2,6-Di chloro-4-nitroani1i ne
2-Chloro-4-n1troan1line
2,4-Dinitroaniline
N-Methylaniline
4-Chloro-2-nitroani1i ne
4-Methylaniline
2,6-D1methylaniline
4-Aminobiphenyl
1-Aminonaphthalene
N,N-Dimethylaniline
Phenanthridine
4-Methylpyridine
2,4-Dimethylpyridi ne
4-t-Butylpyridine
2,4,6-Trimethylpri dine
Quinoline
4-Methylquinoline
Acridine
Carbazole
3,3'-D1chlorobenzidine
Diphenylamine
(continued)
271
-------
DESCRIPTION OF STANDARD SOLUTIONS (Continued)
3=3===3333===============3==a===3==33=a================S=======================
Semi volatile Calibration Solution C
(all compounds at 0.5 mg/ml in methylene chloride)
Aliphatic Halocarbons
l,4-D1chlorobutane
Pentachloroethane
Hexachloroethane
Hexachloropropene
Hexachlorobutadlene
Aromatic Halocarbons
4-Chlorotoluene
Bromobenzene
l,2-D1chlorobenzene
1,4-01chlorobenzene
1,2,4-Tr1chlorobenzene
1,2,4,5-Tetrachlorobenzene
Pentachlorobenzene
Hexachlorobenzene
Benzal chloride
Benzyl chloride
1-Chloronaphthalene
2-Chloronaphthalene
o,o,a-Tr1chlorotoluene
Aromatic Nitro Compounds
Nitrobenzene
1,3-Dinitrobenzene
2-N1trotoluene
4-N1trotoluene
2,4-D1n1trotoluene
2,6-01nltrotoluene
1-Chloro-4-nitrobenzene
2,4-01n1trochlorobenzene
Phenols
2-Chlorophenol
2-N1trophenol
Phenol
2,4-D1methylphenol
2,4-01chlorophenol
2,4,6-Tri chlorophenol
4-Chloro-3-methylphenol
2-Methylphenol
4-Methylphenol
Thiophenol
4-Chlorophenol
2,6-Dichlorophenol
2,4,5-Tri chlorophenol
Hexachlorophene
4-Hydroxybi phenyl
2-Naphthol
4-t-Butylphenol
2-Chloro-4-n1trophenol
2,4-Dinitrophenol
2-Methyl-4,6-d1nitrophenol
Pentachlorophenol
4-N1trophenol
2,6-D1-t-butyl-4-methylphenol
2,4-Di-t-butylphenol
D1 ethyl stil best rol
Chlorinated Pesticides
4,4'-ODD
4,4'-DDE
4,4'-DDT
Methoxychlor
Trifluralln
Pentachloronltrobenzene
Acids
4-Chlorobenzoic acid
Benzole acid
4-Bromobenzo1c acid
2,4-01chlorophenoxyacetic acid
2,4,5-Trichlorophenoxyacetlc aci d
Haloethers
B1s(2-chloroethyl)ether
B1s(2-chloroethoxy)ethane
4-Chlorophenyl phenyl ether
(continued)
272
-------
DESCRIPTION OF STANDARD SOLUTIONS (Continued)
==============================:3=======================S=S===========S=S=
Semi volatile Calibration Solution D
(all compounds at 1.0 mg/mL 1n methylene chloride)
Phthalates Phenyl ether
Dibenzofuran
Dimethyl phthalate
D1-n-butyl phthalate Ketones
D1(2-ethylhexyl) phthalate
Anthraqulnone
Phosphates 2-Methylanthraquinone
Propiophenone
Tri(p-tolyl) phosphate Acetophenone
Triphenyl phosphate
Miscellaneous
Aldehydes
Azobenzene
Benzaldehyde Acetanllide
4-Chlorobenzaldehyde Benzyl alcohol
D1(2-ethylhexyl) sebacate
Ethers and Sulfides
Anisole
Semi volatile Surrogate Standard Solution E
(all compounds at 10 mg/ml in methylene chloride)
Decaf1uorobi phenyl
2-Fluoroanil1ne
Pentafluorophenol
Semivolatile Internal Standard Solution F
(all compounds at 200 ug/ml in methylene chloride)
Ds-Bromobenzene Ds-Aniline
Ds-Naphthalene Ds-Phenol
Din-Phenanthrene D5-N1trobenzene
Dio-Biphenyl D3-2,4-D1nitrophenol
Dio-Acenaphthene Decafluorotriphenylphosphine
DiQ-Pyrene
Di2-Chrysene
Di2-Benzo(a)pyrene
(continued)
273
-------
DESCRIPTION OF STANDARD SOLUTIONS (Continued)
aa»*3a"3=S« =>333333333333333333333S3333333333333333333333333333333333333333333333 3
Volatile Calibration Solution G
(all compounds at 0.2 mq/ml 1n methanol)
2-Chloroethylvinyl ether Dimethyl disulfide
I,l,2-Tr1chlorotr1fluoroethane Epichlorohydrin
Dibromomethane 2-Chloroacrylon1tr1le
Allyl chloride Acetonitrile
Ethylene dibromlde D1chloroacetonitr1le
Chloropicrln n-Prop1onitrile
2-Chloropropane Chioroacetaldehyde
1-Chlorobutane 2-Chloroethanol
o-Xylene N-Nitrosodimethylam1ne
Styrene Vinyl acetate
2-Butanone Dimethyl sulfide
Cyclopentanone D1ethyl ether
4-Methyl-2-pentanone Acetone
2-Hexanone Methyl chloroacetate
Carbon disulfide Methyl acrylate
Methyl methacrylate
Volatile Surrogate Standard solution H
(all compounds at 10 mg/ml in methanol)
1,2-Dibromotetraf1uoroethane
Bls(perfluorolsopropyl) ketone
Fluorobenzene
m-Bromobenzotri f1uorlde
Volatile Internal Standard Solution I
(all compounds at 0.2 mg/ml in methanol)
D4-l,2-D1chloroethane
Ds-Benzene
Ds-Ethylbenzene
4-Bromof1uorobenzene
Volatile Calibration Solution J
(all compounds at 0.2 mq/ml in water)
Acrolein
Acrylonitrile
(continued)
274
-------
DESCRIPTION OF STANDARD SOLUTIONS (Continued)
33=33333333333333333333333333333333333333333333333333333333333333333335
Volatile Calibration Solution K
2-Chloroethyl vinyl ether 0.2 mg/mL in tetraglyme
Supelco's Purgeable A
(all compounds at 0.2 mg/mL 1n methanol)
Methylene chloride Trichloroethylene
l,l-D1chloroethene 1,1,2-Trichloroethane
1,1-Di chloroethane Di bromochloromethane
Chloroform Tetrachloroethene
Carbon tetrachloride Chlorobenzene
1,2-Dichloropropane
Supelco's Purgeable B
(all compounds at 0.2 mg/mL in methanol)
trans-l,2-Dichloroethene Benzene
1,2-01chloroethane Bromoform .
1,1,1-Trichloroethane 1,1,2,2-Tetrachloroethane
Bromodichloromethane Toluene
trans-1,3-Di chloropropene Ethyl benzene
cis-1,3-Dichloropropene
Supelco's Purgeable C
(all compounds at 0.2 mg/mL in methanol)
Chloromethane Vinyl chloride
Bromomethane Chloroethane
275
-------
APPENDIX E
MANUAL FOR COLLABORATORS
276
-------
EVALUATION OF METHODS
FOR ANALYSIS OF
HAZARDOUS WASTES
Manual for Collaborators
Instructions
Analysis Methods
Report Forms
Program Review Inquiry
Battelle
Columbus Laboratories
-------
TABLE OF CONTENTS
Page
PROJECT DISCRETION
Project D1scr1pt1on 279
Study Objective 279
Outline of Study 279
Program Requirements 279
SUPPLIED ITEMS
Instruction Manual 281
Data Report Forms 282
Performance Evaluation Samples • 282
Waste Samples 282
Calibration Standards 282
Fused Silica Capillary Column 282
GC/MS Data 283
GC/MS Library 290
QUALITY ASSURANCE
Quality Assurance/Quality Control 304
QA Objectives ". 304
Documentation and Records 304
Quality Control - Performance Criteria and Checks 305
GC/MS Run Logs 306
GC/MS Calibration 306
Surrogate Standards 306
Blanks 306
CLARIFICATIONS
Clarifications 308
Sample Analysis Order 308
Recording Information on Forms 308
Spiking Surrogates 309
Quality Control 309
Sample Storage 309
Obtaining Aliquots 309
277
-------
TABLE OF CONTENTS (Continued)
FSCC Protocol ClariF1 cations 311
Retention and Response of VolatHes 312
Identification of Unlisted Volatile and Semi volatile Compounds. . . 312
Performance Evaluation Samples 313
Calibration Solutions 313
Prescreenlng Studies 314
Foaming During Purge and Trap 315
Gel Permeation Chromatography . 315
GC/MS Search Information 315
DELIVERABLES
Program Review Inquiry . 318
ADDENDA
Addendum 1 (July 23, 1982) 324
Addendum 2 (August 19, 1982) ..... 328
278
-------
PROJECT DESCRIPTION
This inter!aboratory study is part of a program to select, modify and
evaluate methods for determination of organic constituents in hazardous wastes.
The methods to be examined include one for determination of semi volatile organic
compounds (boiling above ca. 150°C) and one for volatile organic compounds
(boiling up to ca. 150°C), and were selected from several methods for analysis
of solid wastes. The methods were modified to be applicable to a variety of
waste types and were evaluated in one laboratory for applicability, precision
and accuracy. The methods are now to be evaluated via an interlaboratory study.
Study Objective
The objective of this study is to evaluate the methodologies in terms of
the effects of different laboratories and different waste types on data quality.
The study is designed to evaluate the analysis procedure only. Sampling,
standard sources, standard preparation and spiking procedures have been conducted
by Battelle to avoid effects resulting from these sources.
Outline of Study
The study is designed for 10 participating laboratories each to analyze in
triplicate several waste samples. The analyses will be conducted for semivolatile
and volatile organic compounds. The samples delivered to the participants will
have been previously spiked with certain compounds. Each waste sample will be
delivered in one or two septum-sealed screw-cap vials. The sample amounts
will be ample for all analyses required. The data resulting from the partici-
pating laboratories will be analyzed at Battelle to determine repeatability,
reproducibility and recovery all as a function of waste type and laboratory.
The generated data will include analytical performance data that will
provide criteria for quality control parameters in future application of the
methods. Information will be solicited through a Program Review Inquiry Form
regarding any problems encountered in performing the methods and any recommend-
ations for procedural change or procedure wording.
The program includes review of these instructions prior to use in laboratory
work to ascertain that the instructions are understood. If changes result from
the initial review, those changes will be incorporated in all instruction
manuals before laboratory work begins.
Program Requirements
This package of instructions is identical to those sent to all participating
laboratories and it is imperative that all work be peformed and all results be
reported exactly as specified here.
279
-------
It is expected that each participant has the apparatus for conducting the
methods and has personnel that are competent in conducting analyses by GC/MS and
analyses that depart from familiar procedures.
Participants are cautioned against making a number of repeated measurements
and selecting the "best" ones or reporting an average of several measurements.
Each laboratory must report each individual measurement as specified in the
protocol. Runs for which known deviations from the protocol or accidents occur
(spillage, evaporation to dryness, known contamination) should be aborted and
rerun.
280
-------
SUPPLIED ITEMS
The Items supplied to each participating laboratory include:
Instruction Manual
Data Report Forms
Performance Evaluation Samples
Eight Waste Samples
Fourteen Standard Solutions
-- 10 calibration standards
— 2 Internal standards
— 2 surrogate standards
• 1 DB-5 Fused Silica Capillary Column
t GC/MS Data for 200 Organic Compounds
• GC/MS Library for INCOS Users
Any laboratory experiencing equipment or supply problems that might cause
a delay in reporting data on time should contact Battelle program personnel
immediately. Battelle will respond to questions in less than 24 hours. Any
additional supplies that are needed and that can be provided by Battelle will
be shipped within 24 hours of request.
Instruction Manual
This instruction manual was prepared to provide information necessary for
the conduct of the study by each participant. All laboratory personnel to be
associated with the program should become familiar with the manual prior to
Initiation of experimental work. This manual is to be accessible to all per-
sonnel during all phases of the study. Any ambiguities not resolved during
the review of the manual are to be referred to Battelle by telephone and fol-
lowed up by written communication. The Battelle personnel who are to be
contacted include one of the following:
Scott Warner (614) 424-5643
Dick Heffel finger (614)424-5249
or Mary McKown (614) 424-5896
Address written communication to the Battelle person contacted at this address:
Battelle
Columbus Laboratories
505 King Avenue
Columbus, Ohio 43201
281
-------
Data Report Forms
Examples of the data report forms to be used for reporting the data
specified as dellverables are contained in this instruction manual. An appro-
priate supply of each of the forms is provided in the accompanying packages.
Performance Evaluation Samples
The samples to be analyzed by each participating laboratory to demonstrate
acceptable performance include:
1 prepared extract for semivolatile analysis (ILS-10)
1 prepared extract for volatile analysis (ILS-11)
1 solid waste sample (ILS-1) to be extracted and analyzed for
-- semi volatile organics
— volatile organics
The methods to be employed are included in this instruction manual. During the
two-week period allotted for analysis of the performance evaluation samples
each laboratory will be visited by Battelle staff who will observe analysis
techniques and answer any questions posed by participants.
Waste Samples
• ,
The solid waste samples were selected to represent a range of waste types
that are commonly encountered and are distinctly different from each other.
These samples have been spiked, homogenized, and divided into aliquots for
distribution to participating laboratories. The waste samples and the performance
evaluation samples are listed in Table 1.
The samples are to be analyzed for the 200 compounds listed in Table 2 and
for as many as 20 other compounds for which a significant peak is observed. It
may not be possible to Identify and quantify every compound listed in Table 2
for each waste sample. However, any compound not identified should be reported
as not detected. The methods for the determination of semi volatile and volatile
organic compounds Included in this Instruction manual are to be followed for
the extractions and analyses.
Calibration Standards
Fourteen solutions containing chemically compatible calibration standards
are provided to serve as reference points that are Identical for all of the
participating laboratories. The solutions provided are labeled by letters A
through I and contain the compounds listed in Table 3.
Fused Silica Capillary Column
To avoid delays 1n conducting this Interlaboratory study Battelle purchased
10 DB-5 fused silica capillary columms from J&S Scientific Company. Each column
was checked at Battelle and meets performance requirements. This column is the
only column to be used for the semivolatile analyses and the column provided is
not to be used for any other work before the Interlaboratory study is completed.
282
-------
TABLE 1. INVENTORY OF SUPPLIED WASTE SAMPLES
= = -== ================ = = = = === = = =i : s =;-».-! 33= = 5= -33i^1 = ^: = = = ===== =================
Analysis
Battelle . to be
Number Sample Identification Description Performed^)
ILS-1
ILS-2
ILS-3
ILS-4
ILS-5
ILS-6
ILS-7
ILS-8
ILS-9
ILS-10
ILS-11
Contaminated soil Wet solid
Latex paint waste Aqueous d
Ethanes spent catalyst Oily powder
Coal gasification tar Tar
Oxychlorination catalyst Pelletize
Chemically treated POTW sludge Wet solid
Herbicide acetone waste Liquid
Chlorinated ethanes waste Liquid
Contaminated sediment Dry solid
Methylene chloride extract Solution
Tetraglyme extract Solution
3:3= I=-S:3-3.-S:3.-3.= .
35331=3===========
S
persion S
S
S
solid S
S
V
V
S
S
V
, v
, v
, v
, v
, v
, v
(a) S - Semivolatile compounds
V - Volatile compounds
GC/MS Data
GC/MS Data collected by Battelle for the 200 compounds are provided in
Table Q-l from the Quality Control Protocol for Fused Silica Capillary Columns
and Table 4. These data Include:
Response factors
Retention time
Relative retention time
Quantification 1on
Internal standard to be used for quantification
The operating parammeters are specified 1n the methodology description.
283
-------
TABLE 2. 200 COMPOUNDS TO BE DETERMINED IN SOLID WASTES
========3=================================================================;:====
Purgeables Semivolatlies
Purgeable Halocarbons
Chloromethane
Bromomethane
Vinyl chloride
Chloroethane
Methylene chloride
Ethyl bromide
1,1-Dichloroethene
1,1-Dichloroethane
t rans-1,2-01chloroethene
Chloroform
1,2-DiChloroethane
1,1,1-Trichloroethane
Carbon tetrachloride
Bromodi chloromethane
1,2-Di chloropropane
trans-l,3-D1chloropropene
Trichloroethene
01bromochloromethane
1,1,2-Tr1chloroethane
ci s-1,3-D1chloropropene
2-Chloroethylvinyl ether
Bromoform
1,1,2,2-Tetrachloroethane
Terachloroethene
Chlorobenzene
1,1,2-Tri chlorotri f1uoro-
ethane
01bromomethane
Ally! chloride
Ethylene dibromide
Chloropicrln
2-Chloropropane
1-Chlorobutane
Purgeable Hydrocarbons
Benzene
Toluene
Ethyl benzene
o-Xylene
Styrene
Purgeable Oxygen. Sulfur. Aromatic Halocarbons
or
•geabie Oxygen. Sui
Nitrogen Compounds
2-Butannone
Cyclopentanone
4-Methyl-2-pentanone
2-Hexanone
Carbon disulfide
Dimethyl disulfide
Acrylonitrile
Epichlorohydrin
2-Chloroacrylonitr1le
Acetonitrile
D1chloroaceton1trile
n-Prop1onitr1le
Acrolein
Chloroacetaldehyde
2-Chloroethanol
N-Nitrosodimethylamine
Vinyl acetate
Dimethyl sulflde
D1ethyl ether
Acetone
Methyl chloroacetate
Methyl acrylate
Methyl methacrylate
Aliphatic Halocarbons
1,4-Dichlorobutane
Pentachloroethane
Hexachloroethane
Hexachloropropene
Hexachlorobutadi ene
Aromatic Halocarbons
4-Chlorotoluene
Bromobenzene
1,2-Di chlorobenzene
1,4-D1chlorobenzene
1,2,4-Trichlorobenzene
1,2,4,5-Tetrachlorobenzene
Pentachlorobenzene
Hexachlorobenzene
3,3'-D1chlorobiphenyl
4,4' -D1chl orobi phenyl
2,2',4,4'-Tetrachloro-
bi phenyl
Benzal chloride
2,2',4,4',6,6'-Hexa-
chlorobiphenyl
Benzyl chloride
1-Chloronaphthalene
2-Chloronaphthalene
a,a,a-Trichlorotoluene
Aromatic Hydrocarbons
Naphthalene
1,2,4-Trimethyl benzene
1,2,4,5-Tetramethyl-
benzene
Bi phenyl
Acenaphthylene
Acenaphthene
2-Methylnaphthalene
2-Ethylnaphthalene
2,3-D1methylnaphthalene
1,2,3,4-Tetrahydronaph-
thalene
Fluorene
Phenanthrene
anthracene
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(k)fluorannthene
Benzo(a)pyrene
D1benzo(a,h)anthracene
Benzo(g,h,i)perylene
Indeno(l,2,3-cd)pyrene
(continued)
284
-------
TABLE 2. (Continued)
:=================
Senvivolatlles
====================:
Aromatic Nitro Compounds
Nitrobenzene
1,3-DInitrobenzene
2-Nitrotoluene
4-N1trotoluene
2,4-D1nitrotoluene
2,6-Din1trotoluene
1-Chloro-4-n1trobenzene
2,4-Dinitrochlorobenzene
Phenols
2-Chlorophenol
2-N1trophenol
Phenol
2,4-Dimethylphenol
2,4-Dichlorophenol
2,4,6-Trlchlorophenol
4-Chloro-3-methy 1 phenol
2-Methylphenol
4-Methylphenol
Thlophenol
4-Chlorophenol
2,6-D1chlorophenol
2,4,5-Trichlorophenol
Hexachlorophene
4- Hydroxybiphenyl
2-Naphthol
4-t-Butyl phenol
2-Chloro-4-n1trophenol
2,4-Din1trophenol
2-Methyl-4,6-d1nltrophenol
Pentachlorophenol
4-N1trophenol
2,6-01-t-butyl-4-methyl-
phenol
2,4-D1-t-butyl phenol
Dlethylstllbestrol
Amines
Aniline
4-Chloroan1l1ne
4-Bromoan1Hne
Ami nes
2-N1troan1l1ne
3,4-Dichloroanil1ne
2,4,5-Tr1chloroanil1ne
3-N1troan1l1ne
4-Chloro-2-nltroanl11ne
4-N1troaniline
2,6-D1chloro-4-n1troan1l1ne
2-Chloro-4-nitroan111ne
2,4-Dinitroan1line
N-Methylaniline
4-Chloro-o-2-methylan111ne
4-Methylaniline
2,6-Dimethylaniline
4-Am1nobiphenyl
1-Ami nonaphthalene
N,N-D1methylan1line
Phenanthridlne
4-Methylpyridine
2,4-Dimethylpyrldi ne
4-t-Butylpyrldine
1,2,7,8-Di benzocarbazole
2,4,6-Tr1methylpyr1d1ne
Quinollne
4-Methylquinollne
Acridine
Carbazole
3,3-Di chlorobenzi d1ne
Dlphenylamine
Chlorinated Pesticides
4,4'-DDD
4,4'-DDE
4,4'-DDT
Methoxychlor
Trifluralin
Pentachloronitrobenzene
Phthalates
Dimethyl phthalate
D1-n-butyl phthalate
D1(2-ethylhexyl)
phthalate
Phosphates
Tr1 (p-tolyl) phosphate
Tri phenyl phosphate
Aldehydes
Benzaldehyde
4-Chlorobenzaldehyde
Ethers and Sulfides
Anisole
Phenyl ether
Dibenzofuran
Ketones
Anthraquinone
2-Methylanthraquinone
Proplophenone
Acetophenone
4-Chlorobenzoic add
Benzole acid
4-Bromobenzo1c add
2,4-D1chlorophenoxy-
acetlc acid
2,4,5-Trichlorophenoxy-
acetic acid
Haloethers
Bi s(2-chloroethyl)ether
B1s(2-chloroethoxy)ethane
4-Chlorophenyl phenyl
ether
Miscellaneous
Azobenzene
Acetanllide
Benzyl alcohol
D1(2-ethylhexyl)
sebacate
285
-------
TABLE 3. DESCRIPTION OF STANDARD SOLUTIONS
=====================3=====================================================
Semi volatile Calibration Solution A
(all compounds at 0.1 mg/mL in methylene chloride)
Aromatic Halocarbons
3,3'-Dichlorobiphenyl
4,4' -Di chl orobi phenyl
2,2'-4,4'-Tetrachlorob1phenyl
2,2'-4,4',6,6'-Hexachlorobiphenyl
Aromatic Hydrocarbons
Benzo(a)anthracene
Benzo(k)f1uoranthene
Benzo(g,h,i)perylene
Indeno(l,2,3-cd)pyrene
Chrysene
Benzo(a)pyrene
D1benzo(a»h)anthracene
Amines
1,2,7,8-Dibenzocarbazole
Sem1volat1le Calibration Solution B
(all compounds at 1.0 mg/mL in benzene)
Aromatic Hydrocarbons
Naphthalene
1,2,4-Trimethylbenzene
1,2,4,5-Tetramethylbenzene
B1phenyl
Acenapththylene
Acenaphthene
2-Methylnaphthalene
2-Ethylnaphtha!ene
2,3-D1methylnaphthalene
1,2,3,4-Tetrahydronaphthalene
Fluorene
Phenanthrene
Anthracene
Fluroanthene
Pyrene
Amines
Aniline
4-Chloroaniline
4-Bromoan1l1ne
2-N1troan1l1ne
3,4-D1chloroan1l1ne
2,4,5-Trichloroaniline
3-Nitroaniline
4-Chloro-2-methyl ani 1 i ne
4-Nitroaniline
2,6-Di chloro-4-n1troani11nne
2-Chloro-4-ni troani11ne
2,4-D1n1troan1line
N-Methylaniline
2,6-D1methylanil1ne
4-Aminobiphenyl
l-Am1nonaphthalene
N,N-Dimethylaniline
Phenanthridine
4-Methylpyridinne
2,4-Dimethylpyridine
4-t-Butylpyridine
2,4,6-Trimethylpyri di ne
Qulnoline
4-Methylqu1nol1ne
Acridine
Carbazole
3,3'-D1chlorobenz1dine
Diphenylamine
(continued)
286
-------
TABLE 3. (Continued)
========================================3===========S====S===S======
Sem1volat1le Calibration Solution C
(all compounds at 0.5 mg/mL in methylene chloride)
Aliphatic Halocarbons
l,4-D1chlorobutane
Pentachloroethane
Hexachloroethane
Hexachloropropene
Hexachlorobutadlene
Aromatic Halocarbons
4-Chlorotoluene
Bromobenzene
l,2-D1chlorobenzene
l,4-D1chlorobenzene
1,2,4-TH chl orobenzene
1,2,4,5-Tetrachlorobenzene
Pentachlorobenzene
Hexachlorobenzene
Benzal chloride
Benzyl chloride
1-Chloronaphthalene
2-Chloronaphthalene
a,a,o-Tr1chlorotoluene
Aromatic N1tro Compounds
Nitrobenzene
1,3-01 nitrobenzene
2-N1trotoluene
4-N1trotoluene
2,4-D1n1trotoluene
2,6-D1n1trotoluene
1 ,-Chloro-4-n1trobenzene
2,4-D1n1trochlorobenzene
Phenols
2-Chlorophenol
2-N1trophenol
Phenol
2,4-D1methyl phenol
2,4-D1chlorophenol
2,4,6-Trlchlorophenol
4-Chloro-3-methylphenol
2-Methylphenol
4-Methylphenol
Thiophenol
4-Chlorophenol
2,6-Dichlorophenol
2,4,5-Tr1chlorophenol
Hexachlorophene
4-Hydroxyblphenyl
2-Naphthol
4-t-Butylphenol
2-Chloro-4-n1trophenol
2,4-D1n1trophenol
2-Methyl-4,6-din1trophenol
Pentachlorophenol
4-N1trophenol
2,,6-Di-t-butyl-4-methylphenol
2,4-Di-t-butylphenol
01 ethylstllbestrol
Chlorinated Pesticides
4,4'-ODD
4,4'-DDE
4,4'-DDT
Methoxychlor
Trlfluralln
Pentachloronltrobenzene
Adds
4-Chlorobenzo1c add
Benzole acid
4-Bromobenzolc add
2,4-01chlorophenoxyacetlc acid
2,4,5-Trichlorophenoxyacetlc ac1d
Haloethers
B1s(2-chloroethyl)ether
B1s(2-chloroethoxy)ethane
4-Chlorophenyl phenyl ether
(continued)
287
-------
TABLE 3. (Continued)
33333333333333333333333333333333=3333333333333333333333333333333333333=33=3333
Semi volatile Calibration Solution D
(all compounds at 1.0 mg/ml in methylene chloride)
Phthalates Phenyl ether
Dibenzofuran
Dimethyl phthalate
D1-n-butyl phthalate Ketones
D1(2-ethylhexyl) phthalate
Anthraquinone
Phosphates 2-Methylanthraquinone
Proplophenone
Trl(p-tolyl) phosphate Acetophenone
Triphenyl phosphate
Miscellaneous
Aldehydes
Azobenzene
Benzaldehyde AcetanlUde
4-Chlorobenzaldehyde Benzyl alcohol
01(2-ethylhexyl) sebacate
Ethers and Sulfides
Anlsole
Semi volatile Surrogate Standard Solution E
(all compounds at 10 mg/ml 1n methylene chloride)
Decafluoroblphenyl
2-Fluoroan1l1ne
Pentafluorophenol
Semi volatile Internal Standard Solution F
(all compounds at 200 ug/ml 1n methylene chloride)
Ds-Bromobenzene D5-An1Hne
De-Naphthalene D5-Phenol
Dio-Phenanthrene Ds-Nltrobenzene
Din-B1phenyl D3-2,4-D1n1trophenol
Dio-Acenaphthene Decaf1uorotr1phenylphosphlne
DiQ-Pyrene
Di2-Chrysene
Di2-Benzo(a)pyrene
(continued)
288
-------
TABLE 3. (Continued)
s=asa===3!S=s=——3333333333333333333333333333333333333333333333333333335
Volatile Calibration Solution G
(all compounds at 0.2 mg/mL in methanol)
2-Chloroethylvinyl ether Dimethyl disulfide
1,1,2-Tri chlorotri f1uoroethane Epichlorohydri n
D1bromomethane 2-Chloroacrylonitri1e
Allyl chloride Acetonltrile
Ethylene dibromide Dichloroacetonitrile
Chloropicrin n-Propionitrile
2-Chloropropane Chloroacetaldehyde
1-Chlorobutane 2-Chloroethanol
o-Xylene N-N1trosod1methyl amine
Styrene Vinyl acetate
2-Butanone Dimethyl sulfide
Cyclopentanone Diethyl ether
4-Methyl-2-pentanone Acetone
2-Hexanone Methyl chloroacetate
Carbon disulfide Methyl acrylate
Methyl methacrylate
Volatile Surrogate Standard solution H
(all compounds at 10 mg/mL in methanol)
1,2-Dibromotetrafluoroethane
Bis(perfluoroisopropyl) ketone
Fluorobenzene
m-Bromobenzotr1fluoride
Volatile Internal Standard Solution I
(all compounds at 0.2 mg/mL in methanol)
04-1,2-01chloroethane
Ds-Benzene
D5-Ethylbenzene
4-Bromof1uorobenzene
Volatile Calibration Solution J
(all compounds at 0.2 mg/mL in water)
Acroleln
Acrylonitrile
(continued)
289
-------
TABLE 3. (Continued)
3:iaaS3SaSaaaa3a333333a333383a3338SS3a333SSS:i3833S3S8838S3Sa33333SBSS3S33383S888
Volatile Calibration Solution K
2-Chloroethyl vinyl ether 0.2 mg/mL in tetraglyme
Supelco's Purgeable A
(all compounds at 0.2 mg/mL in methanol)
Methylene chloride Trichloroethylene
1,1-Dichloroethene 1,1,2-Trichloroethane
1,1-01chloroethane 01bromochloromethane
Chloroform Tetrachloroethene
Carbon tetrachlorlde Chlorobenzene
1,2-Dichloropropane
Supelco's Purgeable B
(all compounds at 0.2 mg/ml in methanol)
trans-1,2-01chloroethene Benzene
1,2-Dichloroethane Bromoform
1,1,1-Trichloroethane 1,1,2,2-Tetrachloroethane
Bromodichloromethane Toluene
trans-1,3-01chloropropene Ethyl benzene
c1s-1,3-01chloropropene
Supelco's Purgeable C
(all compounds at 0.2 mg/mL in methanol)
Chloromethane Vinyl chloride
Bromomethane Chloroethane
===3333333=3==33=33===3==33333==3==3333====3=3==33==3=========333=333==3==33===
GC/MS Library
The GC/MS library developed by Battelle for the interlaboratory study is
provided on 9-track tape to assist users with reverse search software. This
library can be used only with the INCOS data systems. Hewlett-Packard users
may prepare an equivalent library. Additional information required for the
preparation of a library is given in Tables 5 and 6.
290
-------
TABLE 4. CHARACTERISTIC MASSES AND INTENSITIES FOR SELECTED
SEMIVOLATILE COMPOUNDS
=========================3=====================================================
No. Compound
Quan
Mass
Characteristic Ions (Intensity)
4 *Bromobenzene-D5
1 4-Methylpyridine
2 1,4-Dichlorobutane
3 Am'sole
5 Bromobenzene
82
93
55
108
77
52(20), 54(25), 82(100, 161(73),
163(70)
51(20), 65(25), 66(33), 67(15),
92(30), 93(100)
41(20). 54(15), 55(100), 62(15),
90(22), 92(8)
65(50). 77(15), 78(55), 79(15),
93(12), 108(100)
50(18), 51(20), 77(100), 156(76),
158(75)
6 2,4-Dimethylpyridine
7 4-Chlorotoluene
8 2-Fluoroaniline
9 Benzaldehyde
10 Thiophenol
107 65(20), 79(50), 92(20), 106(60),
107(100)
126 62(11), 89(12), 91(100), 125(17),
126(40), 128(13)
111 64(12), 83(17), 84(22), 91(10),
111(100)
106 50(15), 51(30), 77(81), 78(12),
105(99), 106(100)
110 66(30), 77(11), 84(18), 109(22),
110(100)
11 Pentachloroethane
12
13
14
15
Aniline-DS
Aniline
Phenol-D5
Phenol
119 117(06), 119(100), 121(30), 123(32),
130(32), 165(76), 167(95), 169(46'
98 70(10), 71(20), 97(10), 98(100)
93 65(15), 66(32), 92(12), 93(100)
99 71(20), 99(100)
94 65(14), 66(13), 94(100)
16 B1s(2-chloroethyl) ether 93
17 2-Chlorophenol 128
18 Pentafluorophenol 184
19 1,2,4-Trimethylbenzene 105
20 2,4,6-Trimethylpyridine 121
63(70), 65(25), 93(100), 95(30)
64(20), 100(20), 128(100), 130(30)
93(10), 117(35), 136(75), 137(10),
155(10), 184(100)
77(10), 91(10), 105(100), 119(12),
120(40)
77(11), 79(23), 106(15), 120(18),
121(100), 122(10)
(continued)
291
-------
No. Compound _
TABLE 4. (Continued)
3-33333333333333883338333333383833383888333383338833338333:
Quan
Mass Characteristic Ions (Intensity)
21 1,4-01chlorobenzene
22 Benzyl chloride
48 *Naphthalene-D8
23 Benzyl alcohol
24 1,2-01chlorobenzene
146 50(11), 74(13), 75(25), 111(40),
113(12), 146(100), 148(65), 159(10)
91 65(13), 91(100), 92(81), 126(28),
128(8)
136 108(11), 135(11), 136(100). 137(11)
108 51(20), 77(55), 78(12), 79(100),
91(15), 107(60), 108(90)
146 50(11), 74(12), 75(25), 111(40),
113(13), 146(100), 148(60), 150(10)
25 2-Methyl phenol
26 N-Methylaniline
27 Acetophenone
28 4-Metnylphenol
29 4-Methylaniline
108
106
105
108
107
27(28), 79(30), 80(13), 90(20),
107(80), 108(100)
51(10), 77(25), 79(20), 106(100),
197(85)
51(20), 77(65), 105(100), 120(25)
77(23), 79(18), 90(8), 107(100),
108(90)
77(15), 79(10), 106(100), 107(70)
30 Hexachloroethane
31 N1trobenzene-D5
32 Nitrobenzene
33 N,N,-Dimethyl aniline
34 4-t-Butylpyrid1ne
117 117(100), 119(95), 121(30), 164(40),
166(50), 199(53), 201(85), 203(55)
128 52(12), 54(45), 70(13), 82(100),
98(15), 128(60)
123 50(12), 51(45), 65(13), 77(100),
93(15), 123(60)
120 77(22), 104(13), 105(13), 120(100),
121(90)
120 92(48), 120(100), 135(46)
35 1,2,4,5-Tetramethylbenzene 119
36 Decafluorobiphenyl 265
37 4-Chlorobenzaldehyde 139
38 2-N1trophenol 139
39 Benzal chloride 125
91(15), 119(100), 134(50)
167(12), 234(12 , 265 40), 296(13),
315(13), 334(100), 335(13)
50(20), 75(25). 111(60). 113(20),
139(100), 140(75), 141(40), 142(30)
53(13), 63(23), 64(18), 65(35),
81(23), 93(10), 109(25), 139(100)
63(13), 89(20), 125(100), 127(30),
160(10)
(continued)
292
-------
No. Compound
Quan
Mass
TABLE 4. (Continued)
Characteristic Ions (Intensity)
40 2,4-Dimethylphenol
41 2-N1trotoluene
42 Tetralin
43 Propiophenone
44 2,6-Dimethylaniline
122 77(23), 79(13), 91(18), 107(06),
121(50), 122(100)
120 63(14), 65(70), 77(16), 89(20),
91(45), 92(46), 120(100), 137(6)
104 78(10), 91(42), 104(100), 115(14),
117(15), 131(12), 132(5)
105 51(13), 77(45), 105(100), 134(14)
121 77(13), 91(12), 106(75), 120(60),
121(100), 122(10)
45 Benzole acid
46 2,6-Dichlorophenol
47 1,2,4-Trichlorobenzene
49 Naphthalene
50 4-Chlorophenol
122 50(28), 51(48), 74(10), 77(78),
105(100), 122(80)
162 63(42), 98(40), 99(15), 100(13),
126(18), 162(100), 164(63), 166(10)
180 74(20), 75(12), 109(22), 145(35),
147(20), 180(100), 182(98), 184(30)
128 102(10), 127(13), 128(100), 129(12)
128 64(12), 65(30), 100(12), 128(100),
139(30)
51 4-Chloroaniline
52 2,4-01chlorophenol
53 Hexachloropropene
54 ' 4-N1troto1uene
55 Hexachlorobutadlene
127 63(10), 65(25), 92(20), 100(15),
127(100), 129(30)
162 63(45), 98(30), 126(30)
213 106(28), 117(32), 119(30), 141(38),
143(28), 211(63), 213(100), 215(67)
137 63(15), 65(70), 77(15), 79(15),
89(15), 91(100), 107(30), 137(88)
225 118(37), 188(30), 190(47), 223(60),
225(100), 227(67), 260(36), 262(30)
56 Benzotri chloride
57 l-Chloro-4-nitrobenzene
58 Quinoline
59 B1s(2-chloroethoxy)ethane
70 *Biphenyl-D10
159 63(12), 89(18), 123(12), 159(100),
161(60), 163(10)
111 50(30), 75(100), 99(38), 111(99),
113(30), 127(50), 157(85), 159(28)
129 51(97), 76(10), 102(28), 128(20),
129(100), 139(10)
63 45(20), 63(100), 65(35), 73(12),
93(63), 95(22), 197(33), 137(6)
164 80(12), 82(10), 169(22), 164(100)
(continued)
293
-------
TABLE 4. (Continued)
33 3 3333 3333 33 33 333333 333 33S333 33333333333 33 333 53 33333 3333 53 3 333333333333333333 3
No. Compound.
Quan
Mass
Characteristic Ions (Intensity)
60 4-Chloro-3-methylphenol
61 4-t-butyl phenol
62 4-Bromoaniline
63 2-Methylnaphthalene
64 4-Chloro-2-methylaniline
142 51(13), 77(40), 79(13), 197(100),
142(08), 144(30)
135 95(13), 107(35), 135(100), 136(11),
150(22)
171 92(80), 171(100), 173(80)
142 71(10), 115(20), 141(70), 142(100)
141 77(25), 79(13), 106(100), 149(33),
141(80), 143(27)
65 4-Chlorobenzoic acid 139
66 1,2,4,5-Tetrachlorobenzene 216
67 2,4,6-Trlchlorophenol 196
68 2,4,5-Trlchlorophenol 198
69 Acetanilide ' 93
75(20), 111(35), 113(15), 139(100),
141(30), 156(65), 158(20)
74(15), 108(15), 143(13), 180(20),
181(15), 214(76), 216(100), 218(46)
97(53), 99(20), 132(50), 134(32),
160(23), 196(100}, 198(99), 200(30)
97(48), 99(18), 132(30), 133(18),
134(20), 196(98), 198(100), 200(30)
43(12), 66(12), 93(100), 135(23)
71 2-Chloroanphthalene
72 Biphenyl
73 1-Chloronaphthalene
74 4-Methylqu1nol1ne
75 2-Ethylnaphthalene
162 126(18), 127(42), 162(100), 164(28)
154 76(12), 152(23), 153(38), 154(100),
155(12)
162 126(18), 127(42), 162(100), 164(28)
143 89(10), 115(30), 142(20), 143(100),
144(10)
141 115(20), 128(10), 141(100), 142(22),
156(50)
76 Phenyl ether
77 2-N1troan1line
78 4-Bromobenzoic acid
87 *Acenaphthene-D10
79 3,4-Dichloroaniline
170 51(22), 77(30), 115(13), 141(50),
142(35), 169(25), 170(100), 171(12)
138 52(10), 65(63), 80(15), 92(50),
108(18), 138(100), 139(10)
200 200(100), 202(98)
164 80(20), 82(11), 158(15), 160(35),
162(90), 163(20), 164(100), 165(13)
161 63(14), 90(18), 99(18), 126(14),
161(100), 163(60), 165(10)
(connntlnued)
294
-------
TABLE 4. (Continued)
No. Compound.
Quan
Mass
= =3 33= =
Characteristic Ions (Intensity)
80 2-Chloro-4-nitrop!ienol
81 Acenaphthylene
82 2,3-Dimethylnaphthalene
83 1,3-Dinitrobenzene
84 Dimethyl phthalate
173 63(45), 91(33),, 99(61), 107(33),
143(60), 145(20), 173(100), 175(33)
152 63(78), 76(18), 150(15), 151(18),
152(100), 153(15)
156 115(20). 128(12), 141(90), 153(10),
155(23), 156(100), 157(12)
168 50(53), 64(23), 72(20), 74(82),
76(85), 92(50), 122(43), 168(100)
163 50(11), 76(14), 77(18), 163(100),
164(10), 194(5)
85 2,6-Dinitrotoluene
86 3-Nitroaniline
88 Acenaphthene
89 '2,4-Oinitrophenol-D3
90 2,4-Di-t-butylphenol
165 63(36), 77(25), 89(40), 90(22),
121(18), 135(16), 148(23), 165(100)
138 65(82), 80(20), 92(92), 108(15),
138(100)
154 76(20), 77(11), 151(12), 152(45),
153(98), 154(100)
187 54(40), 66(37), 82(25), 94(30),
95(20), 110(35), 157(55), 187(100)
191 57(18), 191(100), 192(13), 206(15)
91
92
93
94
95
2,4-Dinitrophenol 184
2-Naphthol 144
2,6-D1-t-butyl-4-methylphenol 205
Dibenzofuran
4-N1trophenol
168
109
53(60), 63(76), 79(42), 91(58),
92(24), 107(44), 154(63), 184(100)
89(10), 115(70), 116(40), 144(100),
145(10)
57(15), 145(72), 177(12), 205(100),
206(12), 220(28)
140(33), 168(100), 169(12)
53(21), 63(18), 65(85), 81(20),
93(30), 109(30), 139(100), 140(10)
96 Pentachlorobenzene 250
97 2,4-Dinitrotoluene 165
98 1-Aminonaphthalene 143
99 2,4,5-Trichloroaniline 195
100 2,4-D1n1trochlorobenzene 202
108(23), 213(18), 215(23), 217(12),
248(60), 250(100), 252(70), 254(21)
63(28), 78(13), 89(50), 90(20),
119(26), 165(100), 182(5)
71(10), 115(35), 116(20), 143(100),
144(10)
97(16), 124(20), 133(18), 135(12),
160(16), 195(100), 197(98), 198(30)
63(38), 74(50), 75(92), 109(45),
110(60), 126(28), 202(100), 204(38)
(continued)
295
-------
TABLE 4. (Continued)
s-s s ?^ 3-3 i-s 3-3 3-3 ss s 33-333 ss
No. Compound
Quan
Mass
Characteristic Ions (Intensity)
101 Fluorene > 166
102 4-Chlorophenyl phenyl ether 204
103 4-Chloro-2-n1troan1l1ne 172
104 4-N1troan1l1ne 138
105 2-Methyl-4,6-d1n1trophenol 198
118 *Phenanthrene-D10 188
106 01phenylamine 169
107 Azobenzene 182
108 Tr1flural1n 306
109 2-Chloro-4-n1troan1l1ne" 172
110 4-Hydro*yt>1 phenyl 170
111 Hexachlorobenzene 284
112 2,6-D1chloro-4-n1troan1l1ne 206
113 2,4-D1chloropheno*yacet1c 162
114 4-Am1nob1phenyl 169
115 Pentachlorophenol 266
116 3,3'-D1chlorob1phenyl 222
117 Pentachloronltrobenzene 237
119 Phenanthrene 178
120 4,4'-D1chlorob1phenyl 222
82(15), 83(13), 14(8), 163(15),
164(12), 165(88), 166(100), 167(13)
51(18), 77(35), 115(12), 141(60),
169(15), 176(10), 204(100), 206(32)
63(32), 90(46), 99(48), 126(80),
128(26), 142(20), 172(100), 174(30)
64(10), 65(81), 80(15), 92(40),
108(45), 138(100)
51(28), 77(17), 93(15), 105(40),
107(23), 121(35), 168(30), 198(100)
94(12), 160(1), 184(11), 187(13),
188(100), 189(13)
83(15), 167(30), 168(60), 169(100),
170(13)
51(18), 77(100), 105(32), 152(10),
182(33)
41(41), 43(100), 52(11), 145(10),
264(40), 290(15), 306(60), 335(2)
63(40), 90(93), 99(30), 126(32),
142(73), 144(22), 172(100), 174(32)
115(15), 141(20), 169(10), 179(100),
171(13)
142(30). 214(18), 249(30), 251(20),
282(50), 284(100), 286(82), 288(35)
124(98), 126(30), 160(50), 162(35),
176(85), 178(52), 206(100), 208(63)
133(30), 162(100), 168(68), 175(23),
220(65), 222(40)
115(8), 141(8), 167(10), 168(20),
169(100), 170(13)
165(40), 167(38), 202(15), 230(15),
264(68), 266(100), 268(65), 270(15)
75(15), 93(11), 151(18), 152(70),
186(80), 222(100), 223(13), 224(65)
142(62). 214(75). 235(66). 237(100),
239(68), 249(77), 293(68), 295(70)
89(12), 15(10), 176(18), 177(10),
178(100), 179(15)
25(14), 93(11), 151(12), 152(70),
186(8), 222(100), 223(12), 224(64)
(continued)
296
-------
=========================3===========================================
No. Compound
TABLE 4. (Continued)
:====:
Characteristic Ions (Intensity)
Quan
Mass
121 Anthracene 178
122 Acridine 179
123 Phenanthridine 179
124 Carbazole 167
125 Decafluorotriphenylphosphine 198
89(12), 152(10), 176(18), 177(10),
178(100), 179(15)
89(20), 151(10), 152(9), 178(23),
179(100), 180(15)
89(8), 151(15), 152(10), 178(23),
179(100), 180(15)
83(15), 139(11), 140(10), 166(18),
167(100), 168(12)
69(60), 77(70), 127(60), 198(100),
255(40), 275(23), 442(50)
126 2,4,5-Trichlorophenoxyacetic 196
134 *Pyrene-D10 212
127 D1-n-butylphthalate 149
128 2,4-Dinitroanil1ne 183
129 2,2',4,4'-Tetrachlorobiphenyl 292
167(28), 196(100), 198(96), 200(30),
254(56), 256(50)
104(10), 106(15), 208(14), 211(12),
212(100), 213(15)
149(100), 150(10), 205(5), 223(5)
52(35), 64(30), 91(58), 107(30),
153(65), 183(100)
110(28), 150(30), 220(80), 222(50),
257(28), 290(73), 292(100), 294(48)
130 Anthraquinone
131 Fluoranthene
132 2,2',4,4l,6,6'-Hexachloro-
biphenyl
133 2-Methylanthraquinone
135 Pyrene
136 4,4'-ODE
137 4,4'-ODD
138 Diethylstilbestrol
139 4,4'-DDT
140 Trlphenylphosphate
208 76(33), 150(18), 151(33), 152(77),
180(96), 207(17), 208(100), 209(16)
202 100(10), 101(18), 200(20), 201(15),
202(100), 203(18)
360 145(30), 218(30), 288(45), 290(60),
358(50), 360(100), 362(85), 364(30)
222 82(20), 165(05), 166(35), 194(50),
222(100)
202 100(11), 101(18), 200(18), 201(13),
202(100), 203(15)
246 105(25), 176(40), 210(18), 246(100),
248(63), 316(43), 318(55), 320(28)
235 165(50), 199(13), 235(100), 236(15),
237(63), 239(12)
268 107(50), 145(53), 159(20), 224(13),
239(60), 253(13), 268(100), 269(20)
235 165(40), 176(10), 199(12), 212(10),
235(100), 236(12), 237(65), 239(10)
326 65(37), 77(65), 169(37), 170(38),
215(38), 233(35), 327(99), 236(100)
(continued)
297
-------
No. Compound -
TABLE 4. (Continued)
S ?S=333838388888888888888838333338:
Characteristic Ions (Intensity)
Quan
Mass
141 Benzo(a)anthracene 228
142 Methoxychlor 227
143 *Chrysene-D12 240
144 3,3'-D1chlorobenz1dine 252
145 Chrysene 228
146 B1s(2-ethylhexyl) phthalate 149
147 Tri-(p-tolyl) phosphate 368
148 B1s(2-ethylhexyl) sebacate 185
149 Benzo(k)fluoranthene 252
150 *Benzo(a)pyrene-D12 264
151 Hexachlorophene 196
152 Benzo(a)pyrene 252
153 Dlbenzocarbazole 267
154 Indeno(l,2,3-C0) pyrene 276
155 01benzo(a,h)anthracene 278
113(15), 114(20), 226(20), 228(100),
229(20)
113(10, 114(14), 120(13), 227(100),
228(15), 236(17)
118(12), 120(15), 236(22), 239(12),
240(100), 241(18)
91(16). 126(18), 127(16), 154(15),
182(9), 252(100), 254(65), 256(12)
113(12), 114(12), 226(28), 227(12),
228(100), 229(18)
43(10), 57(20), 70(15), 71(18),
113(10), 149(100), 167(38), 279(9)
107(35), 108(25), 165(30), 198(23),
261(23), 367(75), 368(100), 369(20)
57(33), 70(23), 71(28), 83(13),
98(11), 112(20), 185(100), 186(11)
125(15). 126(20), 250(22), 252(100),
253(20)
12(13), 13(10), 13(21), 26(17),
26(100), 27(20)
196(100), 198(98), 200(32), 208(43),
210(42), 404(18), 406(36), 408(29)
113(10), 125(12), 126(18), 250(20),
252(100), 253(20)
132(16), 134(12), 256(12), 266(22),
267(100), 268(20)
137(18), 138(30), 274(20), 275(11),
276(100), 277(23)
138(21), 139(34), 276(99), 277(15),
278(100), 279(25)
276 137(22), 138(30), 274(22), 275(11),
276(100), 277(23)
sBsaaaaassaassaaaaaBsaaaaassasssaBBSsssaassaasaaaassassssaaaaaaaasBaas
156 Benzo(g,h,1)Perylene
298
-------
TABLE 5. RETENTION AND RESPONSE DATA FOR SELECTED VOLATILE COMPOUNDS
No.
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
31
32
33
34
35
Compound"
Chloromethane
Bromomethane
Vinyl chloride
Chloroethane
Methyl ene chloride
Acetone
Dimethyl sulfide
Acrolein
Bromoethane
Acrylonitrile
Carbon disulfide
Propionitrile
2-Chloropropane
Allyl chloride
1 , 1-Di chl oroethy 1 ene
t rans-1 ,2-Di chl oroethyl ene
1,1-Dichloroethane
Di ethyl ether
Chloroform
1,1,2-Trichlorotrifluorethane
*D4-l,2-Di Chloroethane
2-Butanone
1,2-Di Chloroethane
Dibromomethane
1,2-Di bromotetrafluorethane
2-Chloroacrylonitrile
1,1,1-Tri Chloroethane
Carbon tetrachloride
Epichlorohydrin
Bromodi chl oromethane
B1s(perfluoroisopropyl)ketone
Methyl acrylate
Di chl oroacetonit rile
Vinyl acetate
1-Chlorobutane
Quan
Mass
50
94
62
64
84
58
62
56
108
53
76
55
78
76
96
96
63
59
83
101
102
72
98
93
179
87
97
117
57
127
69
55
74
86
56
RT,
sec.
200
320
386
472
636
696
682
692
688
734
738
798
782
816
814
930
924
954
954
980
992
1000
996
1002
1016
1032
1072
1094
1108
1116
1124
1148
1160
1178
1182
==================
RF(IS)
2.422(21)
2.502 21)
2.063 21
2.388 21)
16.350(21)
1.078(21)
5.644(21)
2.422 21
6.695 21)
2.422(21)
14.626(21)
0.206(21)
1.082(21)
2.722(21)
3.660(21)
4.136(21
7.628(21
5.549 21)
9.202 21)
10.637 21)
0.061(40)
1.850(21)
0.630 21
9.281(21)
4.916(21)
7.636(21)
7.265(21
3.285(21
0.046(21
0.034(21)
0.211(21)
0.482(40)
0.017 40
0.014(40)
0.774(40)
(continued)
299
-------
TABLE 5.
3333333333333333333333333333333333333
No. Compound
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1,2-Dichloropropane
Dimethyl disulfide
trans-1 ,3-Dichl oropropene
Tr i chl oroethyl ene
*Ds-Benzene
D1 bromochl oromethane
1,1,2-Trichloroethane
cis-l,3-Di chl oropropene
Benzene
Di bromomethane
Fluorobenzene
2-Chl oroethyl vinyl ether
Methyl methacrylate
Bromoform
4-Methyl -2-pentanone
2-Hexanone
1,1 ,2 ,2-Tetrachl oroethane
Tetrachl oroethyl ene
Toluene
Chlorobenzene
*D5-Ethyl benzene
Ethyl benzene
*Bromof 1 uorobenzene
Styrene
o-Xylene
61 m-Bromobenzotri fluoride
aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
(Continued)
S3 333 33333 33333333333333335 333= 3=33 = 33 3=3 3
Quan RT,
Mass sec. RF(IS)
63
94
75
130
84
127
97
75
78
107
96
63
100
173
58
58
83
164
92
112
96
91
174
104
106
1200
1206
1212
1244
1272
1274
1282
1286
1278
1332
1354
1344
1416
1422
1464
1556
1562
1572
1654
1720
1822
1830
1938
2006
2058
0.340(40)
0.212(40)
0.377(40)
0.389(40)
1.000(40)
0.168(40
0.296(40
0.305(40
0.928(40)
0.730(40)
0.923(40)
0.027(40
0.159(40
0.030(40)
0.207(56)
0.192(56)
0.235(56)
0.211(56)
0.537(56
0.600(56
1.116(40)
0.355(56)
0.253(40)
1.118(56)
0.683(56)
145 2182 0.303(56)
aaasaaaaaaaaaaaasaaaaaaaaaaaaaaaaaaaaaaaaa
300
-------
TABLE 6. CHARACTERISTIC MASSES AND INTENSITIES FOR SELECTED
VOLATILE COMPOUNDS
33aa—SS*—=*=333333S3333333333333333333333333333333333333333333333333333
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Compound
Chloromethane
Bromomethane
Vinyl chloride
Chloroethane
Methyl ene chloride
Acetone
Dimethyl sulfide
Acrolein
Bromoethane
Acrylonitrile
Carbon di sulfide
Propionitrile
2-Chloropropane
Ally! chloride
1,1-Dichloroethylene
trans -1 , 2-Di chl oroethy 1 ene
1,1-Dichloroethane
01 ethyl ether
Chloroform
Quan
Mass
50
94
62
64
84
58
62
56
108
53
76
55
78
76
96
96
63
59
83
1,1,2-Trichlorotrifluorethane 101
Characteristic Ions, m/e
(relative intensity)
50(100), 52(33)
94(100, 96(93)
62 100), 64(33)
64(100), 66(33)
49(100), 51(30), 84(55), 86(34)
43(100), 58(35)
47 100 , 61 30), 62(80)
56(100), 55(70)
108(100), 110(97)
53(100), 52(80), 51(36)
76(100), 78(10)
55(100)
63 100), 65(32), 78(81), 80(33)
76(100), 78(42)
61(100), 96(54), 98(35)
61(100), 96(58), 98(37)
63(100), 65(32), 83(11), 85(7),
98(4) 100(3)
45(82), 59(100). 73(10), 74(100)
83(100), 85(65)
85(44), 87(15), 101(100), 103(66),
153(53)
21 *D4-1,2-D1chloroethane
22 2-8utanone
23 1,2-01chloroethane
24 Dibromomethane
102
72
98
93
25 1,2-Dibromotetrafluorethane 179
51(35), 65(100), 67(55), 102(20),
104(12)
43(100), 57(10), 72(20)
62(100), 64(32), 98(10), 100(6)
93(100), 95(80), 172(45), 174(85),
176(40)
129(50), 131(50), 179(100), 181(98)
26 2-Chloroacrylonitrile
27 1,1,1-Trichloroethane
28 Carbon tetrachloride
29 Epichlorohydrin
30 Bromodichloromethane
87 51(45), 52(100), 87(85), 89(28)
97 97(100), 99(63), 117(12), 119(11)
117 117(100), 119(97), 121(32)
57 49(25), 51(8), 57(100)
127 83(100), 85(63), 127(7), 129(9)
(continued)
301
-------
33333333333333333333333:5
No. Compound.
TABLE 6. (Continued)
S S3 S : S2 32.32 32223252 = 333333333333 = 33=33333 = 3:
Quan Characteristic Ions, m/e
Mass (relative intensity)
31 B1s(perfluoro1sopropylJketone 69
32 Methyl aery late
33 D1chloroaceton1trile
34 Vinyl acetate
35 l-Chlorobutane
55
74
86
56
59(100), 69(60), 100(25), 169(20),
197(10)
42(10), 55(100), 85(10)
74(100), 76(33), 82(55), 84(36)
43(100), 86(10)
41(50), 43(30), 56(100)
36 1,2-01 cMoropropane 63
37 Dimethyl dlsulfide 94
38 trans-l,3-D1chloropropene 75
39 Trichloroethylene 130
40 *Ds-Benzene 84
62(70), 63(100), 65(31), 76(25)
45(62), 46(38), 47(25), 79(60),
94(100)
75
95
56
100), 77(27)
100), 97(64), 130(97), 132(97)
12), 82(15), 84(100)
41 Dibromochloromethane
42 1,1,2-Trlchloroethane
43 cis-l,3-D1chloropropene
44 Benzene
45 01bromomethane
127 127(78), 129(100), 206(2), 208(5)
97 83(99), 85(63), 97(100), 99(63),
132(8), 134(8)
75 75(100), 110(22), 112(18)
78 78(100)
107 107(100), 109(98)
46 Fluorobenzene
47 2-Chloroethylvinyl ether
48 Methyl methacrylate
49 Bromoform
50 4-Methyl-2-pentanone
96 50(10), 70(25), 75(8), 96(100)
63 43(70). 44(50), 57(20), 63(100),
65(36), 106(20)
100 69(100), 100(46)
173 171(100), 173(100), 175(49),
250(2), 252(9), 254(8), 256(2),
58 57(58), 58(100), 85(33), 100(25)
51 2-Hexanone 58
52 1,1,2,2-Tetrachloroethane 83
53 Tetrachloroethylene 164
54 Toluene 92
55 Chlorobenzene 112
57(30), 58(100), 100(12)
83(100), 85(66), 95(10)
129(100), 131(62), 164(78),
166(100)
91(100), 92(64)
112(100), 114(32)
302
-------
TABLE 6. (Continued)
aBS = S55^: 53=3= ====:; ====33 ======== =3== = ======= ==3=3= ========= ====== =
No.
56
57
58
59
60
61
2 = 33:
Compound -
*Ds-Ethyl benzene
Ethyl benzene
*Bromof 1 uorobenzene
Styrene
o-Xylene
m-Bromobenzotri f 1 uori de
Quan
Mass
96
91
174
104
106
145
lisas 3 = 2-32
Characteristic Ions, m/e
(relative intensity)
96(100), 111(35)
91(100), 106(31)
75(45), 95(100), 174(50),
176(49)
77(12), 78(20), 103(32),
104(100)
77(10), 91(100), 105(30), 106(60)
75(30), 145(100), 224(61), 226(60)
*An internal standard.
303
-------
QUALITY ASSURANCE/QUALITY CONTROL
The objective of Quality Assurance (QA)/Quality Control (QC) activities
conducted for any chemical analysis program 1s to provide data of known quality.
If the results of analyses are contested 1n any way, the quality of these data
must be demonstrable.
While the data from this study are not likely to be directly contested in
a pollution assessment case, the validity of the method may be contested and
does need to be substantiated. The results of the study will form the basis
for quality control requirements when the method is applied routinely.
QA Objectives
The objectives of QA/QC activities on this program are to make certain
that the laboratory work conducted to evaluate the chemical analysis methods is
done under controlled conditions, that those controls are uniformly applied by
all collaborators and that all experimental work is recorded for archival storage.
In addition, when the analysis method is fully evaluated and is applied
for routine analyses of hazardous wastes, the method description will Include
the necessary and appropriate quality control elements and requirements. Part
of the'requirements for that quality assurance and quality control will be
based on experience and knowledge derived from this evaluation. Therefore, 1t
is expected that all persons involved in this program will be aware of the
ulimate use of the methods andwlll be alert, sensitive and critical to controls
instituted to provide high quality data.
Documentation and Records
The documents for this program include the Manual of Instructions for col-
laborators of which this QA/QC plan 1s a part, the Program Review Inquiry form,
data reports, letters of transmittal, records of telephone conversations relative
to this program and all data and records associated with effort on this program.
Copies of these documents will be kept on file by Battelle for audit purposes
and for possible submission to the EPA at the conclusion of the study.
A record shall be kept by participants of all efforts and events associated
with the laboratory work and of all data such as:
t Sample Handling
Date received
Volume and/or weight of samples
Condition of samples
Location and temperature of storage
304
-------
• Analytical Data
Date of extraction and GC/MS analysis
All volumes and weights used
Dilution and concentration factors
Amount of Internal standard added
Internal standard area response
Injection volume
Relative response factors used for quantification
Total solvent extractable content
Major volatile compounds content
Scan number
Absolute and/or relative retention time
Most intense ions
Compound identification
Probable molecular weight
Total ion current chromatograms
Library search results
9-Track tape files
Search system used
Calibration results
Mass spectrometer tuning results
Maintenance records
Most of the above Information 1s "required data" to be reported on forms sup-
plied or in specified format.
Quality Control - Performance Criteria and Checks
Quality control activities start with a description of the method which
must be followed without exception. Before any laboratory work is done, the
method must be read and understood by all personnel who will use it. Questions
regarding what 1s to be done must be discussed with Battelle persons before
laboratory work starts. In that way uncertainties can be corrected or clar-
ified among all cooperating laboratories and all will possess the same informa-
tion. Thus, to the extent possible, the collaborators will conduct all opera-
tions in exactly the same manner.
Control will be maintained by monitoring the mass spectrometer tuning
(using DFTPP or BFB), and analyzing process blanks and calibration standards.
The details of the required quality control measures are described 1n the
methods and 1n the QC Protocol for Fused Silica Capillary Columns.
In application of the methods it is expected that all normal care will
be exercised to use properly calibrated and clean apparatus such as balances
and volumetric glassware for extract preparation and to ascertain that the
GC/MS system is functioning properly.
305
-------
GC/MS Run Logs
Several forms are being supplied for reporting analytical information and a
log of GC/MS runs. A separate log (Forms IV and IS) is required for each mass
spectrometer used to determine volatile compounds and for each mass spectrometer
used to determine semivolatile compounds. Run numbers should be consecutive
from the very first calibration run to the last sample or calibration run. We
envisage and suggest that the analysis runs for this program be consecutive and
without interruption by other programs.
GC/MS Calibration
The daily run routine must include at least one calibration run at the
beginning of the day and additional calibration runs during the day if more than
8 sample runs are made in one day.
Surrogate Standards
All samples must be spiked with surrogate compounds before extracting.
The surrogate compounds are:
For Volatile Compounds .
1,2-01bromotetraf1uoroethane
Bis(perfluorolsopropyl) ketone
Fluorobenzene
m-Bromobenzotrif1uoride
For Semi volatile.Compounds
Decafluorobiphenyl
2-Fluoroaniline
Pentafluorophenol
The spiking level used should be that which will give a concentration in the
final extract used for GC/MS analyses that is equal to the level of the internal
standard added assuming 100% recovery. This level is determined as described
in Section 8.4 of the methods. Thus two aliquots of each sample must be
screened, one for volatile compounds and one for semi volatile compounds, before
aliquots can be spiked with surrogates and analyzed.
Blanks
Blanks will be defined as system or process blanks and will consist of all
reagents used 1n sample preparation carried through the entire preparation
process and finally analyzed by GC/MS. This activity will assess purity of
reagents and cleanliness of apparatus and environment. It is recommended that
a system blank be generated and analyzed with every batch of samples prepared
or with every new batch of reagent material. A minimum of .two process blanks
for the volatile analyses and two process blanks for the semivolatile analyses
must be run and the data reported.
306
-------
CLARIFICATION OF DEVIATIONS
FROM METHODS
307
-------
CLARIFICATIONS
Sample Analysis Order
The order 1n which samples are analyzed is not specified. However, ali-
quots of a given sample are to be analyzed on the same day and in sequence
if possible. The .order in which samples are analyzed will be recorded on the
daily run log.
Recording Information on Forms
It is expected that data (and information such as sample run sequence)
will be entered without delay when they are available. It is expected that
there will be a daily one-over-one review of recorded data to check for errors
in transcription, for completeness and for legibility. The forms are using
typewriter standard space; typing data is recommended.
The protocol for reporting data is as follows:
Concentration - 3 significant figures but only one figure beyond the
decimal place. For example
12300 yg/g
1230 yg/g
123 yg/g
12.3 yg/g
1.2 yg/g
In rounding data, round down when the digit to be dropped is 1, 2, 3, or
4 and up when 6, 7, 8 or 9. Round to an even number when the digit
dropped is 5. For any compound not found or found but below the detection
limit, enter NO. Detection limit for a compound in a given sample is
defined for this purpose as
Concentration of Internal Standard (yg/g sample)
Response Factor x 50
Response Factor - report 3 significant figures beyond the decimal place,
e.g., 1.214 or 0.789.
Relative Retention Time - report 4 figures beyond the decimal, e.g.,
1.0916 or 0.8674.
308
-------
Retention Time - report total seconds - NOT minutes and seconds or minutes.
Relative Abundance of Tuning Compound Ions - report two significant figures.
Spiking Surrogates
The objective of the use of surrogate compounds 1s to measure losses that
may occur during the sample extraction and to detect Immediate matrix Inter-
actions. No attempt 1s being made to equilibrate the surrogate compounds
with sample matrix 1n an effort to obtain a true extraction efficiency. There-
fore, the surrogates will be added along with Internal standards to the extrac-
ting media rather than to the sample.
Surrogates will be added to all sample allquots, and all calibrating
solutions and blanks as described 1n Section 8.4 of the methods. For the
purpose of this program Sections 8.2 and 8.3 of the methods are not applicable.
Solutions of surrogate compounds are supplied. Therefore, the sections 1n the
methods describing the preparation of surrogate compound solutions do not
apply to this program.
Quality Control
• i
As mentioned above, Sections 8.2 and 8.3 of the methods are not applicable
to the Interlaboratory study. These sections are concerned with the demonstra-
tion of an ability to generate acceptable accuracy and precision and with
establishing control limits before the methods are used for routine analyses.
The Interlaboratory study Is designed to generate that data.
Sample Storage
Waste samples and calibration solutions are to be stored 1n the dark and
at 4°C 1n a refrigerator as soon as possible after receipt. Warm samples to
room temperature 1n the dark for at least two hours but not more than 18 hours
before obtaining allquots for analysis.
Obtaining Allquots
The waste samples shall be mixed to the extent possible before the sample
container 1s opened and once the container 1s opened the triplicate allquots
for determinations of volatile compounds shall be taken without delay. After
the 4 allquots for determinations of volatile compounds have been taken, the
sample shall be mixed further as appropriate before the 5 allquots for determi-
nation of semlvolatlle compounds are taken. The samples vary grossly 1n phys-
ical characteristics and thus the requirements for mixing vary. The following
m1x1ng-a!1quot1ng Instructions are given for each sample supplied.
309
-------
Sample
Identification Instructions
ILS-1 - • Take allquots for volatile compounds determinations
with metal spatula.
0 Stir to mix the sample and take allquots for semivolatile
compound deteminations with metal spatula.
ILS-2 t Shake the sample vial to resuspend sediment and
immediately take allquots for volatile determinations
with a syringe or medicine dripper.
• Take the allquots for semivolatile compound determination
1n the same manner while the sample is being stirred.
ILS-3 • Take allquots for volatile compound determinations
with metal spatula.
• Stir to mix the sample and take aliquots for semivolatile
compound determinations.
ILS-4 • Take allquots for volatile compound determination with
a metal spatula or length of glass tubing.
• Stir to mix the sample and take allquots for semivolatile
compound determinations in the same manner.
ILS-5 • Take aliquots for volatile compound determinations
with metal spatula or spoon.
t Grind the pellets to a fine powder 1n a glass mortar
and pestle and take allquots for semivolatile compound
determinations with a metal spatula.
ILS-6 • Take allquots for volatile compound determinations
with a metal spatula.
• Stir to mix the sample and take aliquots for semi-
volatile compound determinations.
ILS-7 • Shake the sample vial thoroughly. Take aliquots for
volatile compound determinations using a syringe or
medicine dropper.
t (Not to be analyzed for semivolatile compounds)
ILS-8 t Shake sample vial thoroughly. Take allquots for
volatile compound determinations using a syringe or
medicine dropper.
t (Not to be analyzed for semivolatile compounds).
310
-------
Samp]e
Identification Instructions
ILS-9 0 (Not to be analyzed for volatile compounds).
• Take aliquots for semi volatile compound determina-
tions using a metal spatula.
FSCC Protocol Clarifications
This section is intended to clarify various aspects of the FSCC protocol,
as applied to this interlaboratory study. The semivolatile analytes, Internal
standards, and surrogates are listed in retention order in Table Q-l of the
FSCC protocol. For each compound listed, the indicated quantification mass
must be used. For the majority of the analytes, the retention standard and
quantification standard refer to the same internal standard. The exceptions
are the unlabeled counterparts to the labeled column performance standards.
Only for these particular compounds do the labeled column performance standards
serve the additional function of quantification internal standards. Further-
more, each of seven labeled internal standards and the four labeled column
performance standards is referenced to Djo-phenanthrene. The eleven response
factors thus obtained serve as an ongoing guide to injector, column and instru-
ment performance. Since these standards are present in every GC/MS analysis
the potential exists to detect,chromatographic or instrumental problems between
daily calibration check runs.
Although the protocol indicates (3.3 and 8.2) the use of mass chromato-
grams (i.e. extracted ion current profiles) as qualitative identification
criteria, the analyst may elect to utilize a commercial or custom reverse
search algorithm which may not explicitly follow the criteria of using discrete
mass chromatograms. The use of such an algorithm is permissible provided the
reliability of the algorithm approximates that using discrete mass chromatograms.
Quality assurance is met by the analysis of a 3-level set of calibration
standards described on page 8 of this section and a dally calibration check
standard. Proper tuning of the mass spectrometer must be maintained throughout
the study. Tuning 1s verified in the conventional fashion using DFTPP. Ana-
lysts experienced with environmental GC/MS analyses, particularly EPA Method
625 are generally able to approximate the proper DFTPP tuning using a conven-
tional mass calibration compound such as perfluorotrlbutylamine leaked directly
into the source. It is recommended that the requisite mass balance and reso-
lution be tuned 1n this fashion and confirmed using the DFTPP spectrum obtained
from the daily calibration check standard. This procedure should eliminate
the need for repeat or separate injections of DFTPP alone in order to meet the
tuning criteria.
In addition to providing a DFTPP spectrum, the dally calibration check
standard is designed to serve a dual purpose. The standard acts as a perfor-
mance evaluation check and, if these criteria are met, functions as one of the
three calibration checks standards used for the calculation of working average
RRFs as described 1n 10.3 of the protocol. The average RRFs thus obtained are
311
-------
to be applied only to those sample analyses performed during the same day (or
8-hour analysis session) as the dally calibration check standard. This usage
requires that the calibration check standard be analyzed at the beginning of
each day that samples are analyzed.
The three level calibration check (20, 50, 200) described 1n 8.3.5 and 9.3
should be performed at the beginning of the analytical efforts associated with
this study and at subsequent intervals not to internal standards, compute the
A unlab ,
area ratio, A lab, at each of the three levels and calculate the correlation
coefficient with respect to the amount of unlabeled analyte. For a given
unlabeled-labeled pair, the correlation coefficient (r) can be calculated as
follows:
r * 3Zx1y1 - 270£yi
236.22 [3Iy12 -
Where:
y1 = area ratio at amount x1
xi = 20, 50, or 200.
A correlation coefficient of 0.9990 or higher for each of the 8 compounds
verifies that the dynamic range and linearity of the method are not being
compromised by offsets, saturation or non-linear instrument gain. The eight
labeled standards provided are of sufficient isotopic purity and substitution
that isotoplc dilution corrections to the area ratios are not required.
Retention and Response of Volatiles
Several of the listed volatile compounds have demonstrated, 1n a single
laboratory evaluation, weak or Inconsistent recoveries using the purge and trap
method. Those compounds consistently recovered and also the internal standards
and surrogates are shown in retention order in Table 5. The quantification
mass shown for each compound must be used by the participating laboratories.
These masses have been chosen based on a number of factors which Include prox-
imity to the mass of the Internal standard, relative Intensity and chromato-
graphic resolution within the calibration mixture. As with the semivolatile
standards, three of the Internal standards are referenced to the fourth standard
(De-benzene) 1n order to provide a performance check for each analysis. The
response factors listed are to serve as a guide only. An updated 11st will be
provided based on further investigations of these compounds.
Identification of Unlisted Volatile and Semivolatile Compounds
For each sample analyzed, the participant 1s to assign a tentative Iden-
tification and semi quantitative estimate for significant GC peaks not corre-
sponding to one of the listed compounds associated with the method used.
Significance shall be defined as a GC peak whose height is at least 10 percent
of the GC peak corresponding to De-benzene (volatile) or Dio-phenanthrene (seml-
volatlles). The semlquantitative estimate is to be calculated using De-benzene
312
-------
phenanthrene as the internal standard based on the total ion current areas of
the unknown and the Internal standard with an assumed response factor of 1.0.
A maximum of 10 unlisted compounds are to be reported for each semi volatile
analysis and 10 unlisted compounds are to be reported for each volatile analysis.
If more than 10 peaks meet the significance criteria, the 10 largest peaks
should be chosen. Forms S and 3V are to be used to summarize data obtained for
these compounds. The data for each compound to be listed on these forms includes
the estimated amount, retention time in seconds (for volatiles) or relative
retention time (relative to Djo-phenanthrene for semivolatiles) and the integer
masses and relative intensities of the four most abundant ions in the background
corrected mass spectrum. In addition, the participant is to provide a copy of
the total ion chromatogram from each GC/MS run with scan numbers indicated, a
bar graph spectrum of each unlisted compound with the scan number indicated and
a copy of the printout with the scan number indicated obtained from a forward
library search using the resident library search algorithm in the participant's
data system. The search should be conducted using the most recently available
edition of the EPA/NIH mass spectral data base.
Performance Evaluation Samples
The concentrations of compounds in the two prepared extracts have been
adjusted to permit analysis with no further treatment other than the addition
of the specified amount of internal standard. One milliliter of the extract to
be analyzed for semivolatile compounds is to be mixed with 330 pL of Semivolatile
Internal Standard solution F (200 jig/ml) and 1 uL of the mixture is to be
injected into the GC/MS. An 80-nL sample of the extract to be analyzed for
volatile compounds is to be injected into the purging chamber containing 12.5 ML
of internal standard (20 ug/ml_). The internal standard is prepared by diluting
100 uL of Volatile Internal Standard Solution I (200 ng/mL) to 1.0 ml with
reagent tetraglyme.
The methods for the determination of semivolatile and volatile organic
compounds included in this instruction manual are to be followed for the
extraction and analysis of the waste sample, ILS-1, which is being used as a
performance evaluation sample.
Calibration Solutions
The calibration solutions of volatile compounds provided are Supelco's
Purgeable A (4-8815), Supelco's Purgeable B (4-8816), Supelco's Purgeable C (4-
8817), and Volatile Calibration solutions G, J and K. The concentration of
each component in these six calibration solutions is 200 ug/mL. The volatile
surrogate standard solution provided contains 10,000 iig/nt of each surrogate.
A calibration mixture containing 10 ng/nt of each of the volatile compounds is
prepared by taking 50 ML of each of the six calibration solutions and 50 uL of
a 1:50 dilution (in reagent tetraglyme) of the surrogate standard solution and
diluting to 1.0 ml with reagent tetraglyme. Four calibration levels, 25, 100,
250, and 1000 ng, are to be analyzed once at the beginning of each week. These
are obtained by using 2.5, 10, 25 and 100 ML, respectively, of the calibration
313
-------
mixture. The results obtained from the analysis of the low level standard, 25
ng, are to be reported but not used in any average response factor calculations.
A single calibration level, 250 ng, is to be used for all subsequent calibrations.
This calibration level must be anayzed at the beginning of each day that samples
are analyzed.
The calibration solutions of semivolatile compounds provided are Semivola-
t1le Calibration Solutions A, B, C, and D. The concentrations of each component
in these four solutions are 100, 1000, 500, and 1000 yg/mL, respectively. The
Semivolatile Surrogate Standard Solution E provided contains 10,000 yg/mL of
each surrogate. A calibration mixture containing 267 yg/mL of each of the
semivolatile compounds in Solutions B, C, and E and 133 yg/mL of each of the
semivolatile compounds 1n Solution D is prepared by mixing 200 yL of B, 400 yL
of C, 100 yL of D, 20 yL of E and 30 yL of methylene chloride. The resulting
mixture, when mixed with 250 yL of the 200 yl/mL Semivolatile Internal Standard
Solution F, will provide a calibration standard containing 200 yg/mL of each
compound in Solutions B, C, and E, 100 yg/mL of each compound 1n Solution D and
50 yg/mL of each internal standard and performance standard.
A calibration mixture containing 66.7 yg/mL of each of the semivolatile
compounds 1n Solutions A to E 1s prepared by mixing 1.00 mL of A, 100 yL of
B, 200 yL of C, 100 yL of D, 10 yL of E, and 90 yL of methylene chloride.
Calibration mixtures containing 26.7 yg/mL and 6.67 yg/mL are prepared by
diluting the 66.7 yg/mL mixture by factors of 2.5 and 10, respectively. The
resulting mixtures, when 750 yL 1s mixed with 250 yL of the 200 yg/mL Internal
standard solution will provide calibration standards at 5, 20 and 50 yg/mL,
respectively, containing all of the Internal standards and performance standards
at 50 yg/mL. Each of the four calibration concentrations levels is to be
analyzed once at the beginning of each week. The results obtained from the
analysis of the low level standard, 5 yg/mL, are to be reported but not used
1n any average response factor calculations.
Prescreenlng Studies
Prior to the extraction and analysis of three replicate allquots for the
determination of volatile compounds and/or three replicate allquots for the
determination of semivolatile compounds from each waste sample, allquots must
be taken for prescreenlng studies to determine the approximate level of major
volatile compounds and/or major semivolatHe compounds present. The amount of
the surrogate compounds to be added to each replicate will be determined by
these prescreening studies. The prescreenlng procedures are described in the
methods for the determination of volatile and semi volatile compounds. For vola-
tile components, an aliquot is extracted with n-hexadecane and the extract is
analyzed by packed column gas-chromatography. For semi volatile components, an
aliquot 1s extracted with methylene chloride and the extract is analyzed by fused
silica or glass capillary column gas chromatography (DB-5 SE-54 or equivalent).
The prescreenlng for the determination of volatile compounds will give the
estimated major volatile compounds content (MVCC). The prescreening for volatile
content is to be conducted only once and the MVCC is to be reported at the top
of Form 2V. The prescreenlng for the determination of semivolatile compounds
314
-------
will give the total solvent extractable content (TSEC), determined gravimetri-
cally, as well as an indication of the amount of dilution or concentration of
the extract required prior to GC/MS analysis. The prescreenlng for semivola-
tile content >s part of the general extraction scheme. The TSEC value from the
initial prescreenlng does not need to be reported. However, the TSEC is to be
determined on the extract from each of the three replicates extracted for
GC/MS analysis and the TSEC values are to be reported at the top of Form 2S.
The total amount of sample required for the determination of volatile
compounds is 7 grams. This amount Includes one gram for a prescreenlng run and
three 2-gram allquots for extraction and GC/MS analysis. The total amount of
sample required for the determination of semivolatile compounds 1s 15 grams.
This amount Includes three grams for an acid/base screen, 3 grams for a pro-
screening run and three 3-gram aliquots for extraction and GC/MS analysis.
In order to minimize the loss of volatile compounds, we recommend that the
allquots for volatile analyses be taken before the aliquots for semivolatile
analyses are taken.
Foaming During Purge and Trap
The use of tetraglyme increases the degree of foaming that occurs. It is
necessary to use purging chambers that have a bulb at the top or are otherwise
designed to revent foam or liquid from being carried into the transfer line.
Gel Permeation Chromatography
A cleanup step Involving gel permeation Chromatography (GPC) is described
in Section 11 of the method for the determination of semi volatile compounds.
This cleanup step 1s useful for the analysis of samples in which most of the
solvent extractable material 1s polymeric or high molecular weight. GPC should
not be used as a cleanup step for any of the samples in the interlaboratory
study. No GPC capability is required.
GC/MS Search Information
The qualitative criteria specify that a listed compound found in an unknown
must meet the following minimum specifications:
(1) The retention time must be within ± 8 seconds of that calculated from
the relative retention time (RRT).
RRTs should be updated from the daily calibration run...
(2) All Ions above 10% relative abundance in a standard must be present
and within ± 20 percent of the relative abundance in the standard.
(3) All characteristic ions defined 1n (2) above must each exhibit a GC
maximum within ± one scan (i.e., three consecutive scans).
315
-------
It is recognized that a search for this large number of compounds in a
particular set of sample data could require an extensive amount of effort on
the part of the participant, particularly if one attempts to interactively
define and plo't mass chromatograms. Most major GC/MS data systems contain
packaged or modular reverse search programs which are designed to ease this
burden on the user. The degree of automation of these programs varies, how-
ever, many programs are designed to search sequentially for a set of compounds
1n a completely unattended fashion. Although a particular program may not
explicitly Implement the above Identification criteria 1n the search algorithm,
the program may be used for this study provided that the performance of the
program approximates user interactive approach using GC/MS chromatograms. It
is the participant's responsibility to ensure that a reverse search program,
1f used, functions in a reliable fashion.
In addition to the reverse search for listed compounds, a forward library
search is to be performed on background corrected spectra obtained from GC
peaks not associated with one of the listed compounds. Most commercial data
systems provide this capability as well as an on-line mass spectral library
derived from the EPA/NIH data base. Since a variety of forward and reverse
research programs are available among the participating laboratories no further
direction 1s given for conducting the search programs.
316
-------
SUMMARY OF DELIVERABLES
Each participating laboratory is required to meet all due dates listed in
the Schedule. The following data are to be submitted by each participating lab-
oratory.
1. A run log of the GC/MS analyses performed for the study including run
number, identity of run, date of extraction, date of analysis, and the area
counts and retention time of the primary internal standard (Forms IV and IS),
2. Amount, retention time or relative retention time, and response factors
used for each of the 200 listed compounds detected in each sample analyzed,
including process blanks (Forms 2V and 2S).
3. Amount, tentative identification, most intense ions, and retention time or
relative retention time for each of the 10 major unlisted volatile com-
pounds and 10 major unlisted semivolatile compounds found 1n each sample
(Forms 3V and 3S).
4. Relative ion abundances of.the mass spectrometer tuning compound for each
tuning check run. (Forms 4V and 4S).
5. Retention time and response factor for each internal standard in each GC/MS
run (Forms 5/6V, 5S, and 6S).
6. Recovery, concentration added, internal standard level, retention time
(Volatiles) or relative retention time (Semivolatiles) and response factor
used for each surrogate standard 1n each sample (Forms 7V and 7S).
7. Response factor and retention time (Volatiles) or relative retention time
(Semivolatiles) found for each of the listed compounds in each calibration
run (Forms 8V, 8S, 9V, and 9S).
8. Total solvent extractable content and major volatile compounds content
determined during the prescreening of each sample (Forms 2V and 2S).
9. A copy of the total ion chromatogram from each GC/MS run.
10. The background corrected mass spectrum and library search results, both
Identified by scan number, for each unlisted compound reported.
11. A brief narrative description of the GC/MS system used to acquire data and
the computer software and/or manual processes used to obtain the qualita-
tive and quantitative results reported.
12. Comments and recommendations regarding the use of the methods (Program
Review Inquiry Form).
317
-------
PROGRAM REVIEW INQUIRY
FOR
COLLABORATORS IN THE INTERLABORATORY STUDY OF
METHODOLOGY FOR ORGANIC CONSTITUENTS OF
HAZARDOUS WASTE
1981
The purpose of this Inquiry 1s to elicit pertinent Information and
recommendations from operators Involved 1n this study regarding any problems
encountered 1n carrying out their part of the study.
Your response will be an Important aid 1n providing clear and unambiguous
descriptions of the techniques and procedures of the method.
Therefore, respond to all questions and requests. If a particular request
1s outside your experience, enter "no opinion".
A separate Inquiry form 1s to be filled out by every person connected with
the study. Use additional sheets for any additional space necessary for lengthy
comments.
Name of Company
Name
Function or Title
How Involved with this study (Administration, sample preparation, analysis, QC,
etc.)T
Format and Style of Instructions
Were Instructions organized logically and written concisely?
318
-------
What Improvements can you suggest? (You may want to return an edited copy of
the Instructions with your suggestions.)
Sample Packaging and Preservation
Describe any problems encountered?
Describe your treatment of the problems?_
Describe changes that you recommend?_
319
-------
Obtaining AHquots
Were Instructions clear and understandable? If not, please explain.
—''•^
Describe any problems encountered and your treatment of the problems.
•w
Describe changes that you recommend.
Sample Extraction
Were Instructions clear and understandable? If not, please explain.
Describe any problems encountered and your treatment of the problems.
320
-------
Describe changes that you recommend.
GC/MS Procedures
/
Were Instructions clear and understandable? If not, please explain.
Describe any problems encountered and your treatment of the problems.
Describe changes that you recommend.
321
-------
QC Requirements
Were Instructions clear and understandable? If not, please explain.
Was protocol for daily analyses and for 6C/MS performance adequate for control
of analyses? ___
Describe any problems encountered with QC requirements.
Describe changes that you recommend.
322
-------
Data Sheets
Were Instructions clear and understandable? If not, please explain.
What changes do you recommend?
Other Comments
Enter here any comments or recommendations for Items not covered above.
323
-------
ADDENDUM 1
Response Factors of Semivolatile Compounds
Since 1-chloronaphthalene frequently coelutes and interferes with the quanti-
fication of DiQ-biphenyl, we have deleted Dio-b1phenyl as an internal stan-
dard and have replaced 1t with Dio-acenaphtnene. We have revised Table Q-l
that appeared in the Quality Assurance section of the manual to change the
relative retention times and response factors for those compounds that were
previously referenced to DiQ-biphenyl. Your search libraries will need to be
revised to reflect these changes. We have also changed many of the other
response factors and some of the other relative retention times to reflect more
recent and more reliable data. Replace Table Q-l dated December 17, 1981 with
the enclosed Table Q-l dated July 23, 1982.
Performance Criteria for Detemination of Semlvolatile Compounds
Prior to the analysis of samples each laboratory must demonstrate that the
tuning and sensitivity of the GC/MS system are acceptable and that response
factors obtained for calibration standards at the 50 ng/yL level are acceptable.
In order to minimize the amount of effort required to check response factors a
check 11st of 20 representative compounds has been selected. These compounds,
the response factors specified for the compounds (using the reference internal
standards designated in Table Q-l), and the acceptable ranges for the response
factors are given 1n Table 2. The acceptable ranges for the response factors
are given in Table 2. The acceptable ranges represent an allowable variation
of ±40% from the specified response factors. If acceptable response factors
are obtained for these 20 compounds, it 1s reasonable to expect that all
compounds listed in Table Q-l, Including surrogates and deuterated Stamfords,
with the exception of the 11 compounds noted with a double star after their
response factor, will be detectable and will have response factors of at least
50% of the listed values. If any of the unstarred compounds in Table Q-l is
not detected 1n a calibration run at the 50 ng/yL level or 1f a response factor
1s less than 50% of the listed value, the search and quantification should be
checked manually to see 1f there 1s a problem with the computer routine used.
The required dally GC/MS run protocol for the determination of semi volatile
compounds 1s as follows:
• Tune mass spectrometer with DFTPP introduced via the GC column
0 Make first calibration run using a 50 ng/yL standard
0 Make sample analysis runs
0 Make second calibration run using a 50 ng/yL standard
The DFTPP relative ion abundances must meet all EPA (Elchelberger) criteria
(given in Table 1 of the method).
324
-------
The sensitivity setting of the mass spectrometer must be such that the area
obtained for Dio-phenanthrene 1n the first calibration run of the day 1s at
least 50 times the area that represents a reliable quantification Hm1t and
less than one-fourth of the minimum area that represents significant saturation.
The response factors obtained from the first calibration run of each day for
all of the 20 compounds shown 1n Table 2 must agree within ±40% of the RF's
specified (I.e. within the acceptable range given).
The response factors obtained from the first calibration run of the day may be
used for the quantification of the listed compounds found 1n the samples ana-
lyzed during the day 1f the response factors obtained from the second calibra-
tion run for all of the 20 compounds shown In Table 2 agree within ±20% of
those obtained from the first calibration run of the day. (In accordance with
Section 7.4 of the method).
If the response factors obtained from the second calibration run for one or
more of the 20 compounds shown 1n Table 2 varies by more than 20 percent but less
than 40 percent from those obtained from the first calibration run of the day,
the average RFs obtained from the first and second calibration runs must be used
for the quantification of all listed compounds found 1n the samples analyzed
during the day.
If the response factors obtained from the second calibration run for one or
more of the 20 compounds shown.1n Table 2 varies by more than 40 percent from
those obtained from the first calibration run of the day (making It Impossible
to meet the requirement of Section 7.4 of the method), all of the sample analyses
for the day must be rejected.
The 20 percent and 40 percent variations mentioned above refer to relative amounts,
For example, the acceptable range for the response factor of 2,6-d1chlorophenol
1s 0.30 ± 40% or 0.18 to 0.42.
The above performance protocol 1s shown schematically 1n Figure 1. The DFTPP
tuning check can be done either as a separate GC run or 1n combination with the
calibration run.
Dellverables
DFTPP relative 1on abundance data 1s to be reported only for tuning checks made
at least once each day (rather than from every run as previously requested).
Items 6 and 7 of the Summary of Dellverables section of the manual specify that
"retention time or relative retention time" 1s to be reported. Retention time
applied only to determinations of volatile organic compounds and relative
retention time applies only to determinations of semlvolatHe organic com-
pounds. A revised "Summary of Dellverables" section dated July 23, 1982 1s
enclosed to replace the one dated December 17, 1981.
In calculating averages and relative standard deviations, do not Include zeros
nor "not found" data.
325
-------
Because of differences 1n the quality (repeatability) of different data param-
eters there are differences 1n the number of digits to be reported. The pro-
tocol for reporting 1s given on pages 1 and 2 of the Clarifications section of
the Manual.
Be sure to fill out each report form completely; especially remember to Include
the laboratory name and sample number (not run number).
Be sure to Include 1n your final data package to us responses to points 11 and
12 on the "Summary of Dellverables" (narrative description of GC/MS Systems and
comments on the program review Inquiry form).
Check dilution factors, Internal standard levels, and calculations carefully
to Insure that the correct values are reported for the amounts found as ug/g.
Miscellaneous Notes
Poor chromatography 1s frequently obtained for pyrldlnes. For this reason
double stars have been placed after their response factors 1n Table Q-l. Sim-
ply report whatever results you get for pyrldlnes without trying to Improve
their chromatography.
Tetraglyme that 1s satisfactorily pure 1s being sought and availability will
be announced by July 30.
We anticipate that participants have sufficient supplies of standard solutions
to complete this study. However, there 1s a small Inventory of some of these
solutions that will be distributed 1f necessary. Each participant 1s expected
to take care of his own needs for compounds 1n Supelco Purgeable A, B, and C.
Sample ILS-5, for determination of semlvolatHe organic compounds, 1s the only
sample In this study that requires pulverization. Use a mortar and pestle made
of glass or agate Instead of porcelain to avoid losses caused by absorption.
As many of you may have guessed from looking at your results for the perfor-
mance evaluation samples, ILS-10 was an extract of ILS-1. Thus by comparing
your results for the two samples you can get an Indication of how your extrac-
tion efficiency and dilution scheme compared with ours. In most cases 1n
which fewer compounds were reported for ILS-1 than for ILS-10, the extract v/as
diluted considerably more than we diluted. In order to obtain satisfactory
results for the remaining samples, careful attention will need to be given to
the GC/FID screening. We recommend that you screen each extract at the final
concentration Intended for GC/MS analysis (taking Into account the dilution
resulting from the addition of the Internal standard solution) and make sure
that the average peak height of the five largest peaks or the height of a
broad unresolved envelop of peaks 1s at least as great as that of phenanthrene
at a concentration of 100 ugM. recommended 1n Section 10.6 of the method).
If the total 1on current chromatogram Indicated considerably lower levels of
compounds than Indicated by the FID chromatogram, the extract may need to be
concentrated further and reanalyzed. " • • -
326
-------
Many of the detection and quantification problems experienced by several of the
laboratories seemed to be attributable to an Inability of the computer search
routines used to handle such complex and dirty samples. The samples are repre-
sentative of "neal world" samples; but, the extracts are frequently dirtier
and more complex than extracts of water samples. You may find that manual
searching will significantly decrease the number of false negatives and Improve
the reliability of the quantifications 1n many cases.
Several corrections have been made 1n the report forms. These include the
addition of the surrogates to Forms 8V, 9V, 8S, and 9S and the correction of
names or spellings for Cpd. 15 on Form 2V, Cpds. 91, 92, 104, and 138 on form
2S, and Cpds. 66 and 101 on Form 8S. A complete set of the corrected forms,
dated July 23, 1982, 1s enclosed. Additional sets of Forms 2V and 2S are also
enclosed. Please use these latest forms for reporting your data. Do not use
forms for reporting your data. Do not use forms dated December 17, 1981.
We are deeply aware of the amount of effort and attention to detail this study
requires and we appreciate the patience and sense of cooperation exhibited thus
far. We're confident that the results of the study will make the effort worth-
while.
327
-------
ADDENDUM 2
Clarification of Section 11.3.2 of VolatHes Method
Several laboratories have Indicated some difficulty 1n understanding the cal-
culations suggested 1n Section 11.3.2 of the method for the determination of
volatile organic compounds. The confusion has been caused by some errors 1n
the wrlteup that resulted when MVCC was changed from mg/g to ug/g during the
method development. One laboratory has also Indicated that some guidelines for
determining the amount of surrogate standard solution to be used would be help-
ful. We have therefore revised Section 11.3.2 to clarify these matters. The
revised section 1s enclosed. The specific revisions are enclosed 1n brackets.
Screening Protocol for Determining MVCC
One of the samples to be analyzed for volatile compounds has very substantial
amounts of compounds 1n the n-hexadecane extract that elute very close to
n-dodecane. The MVCC value for that sample will vary widely depending upon
whether or not the material that elutes near n-dodecane 1s Included. Since
material that elutes near n-dodecane would have a boiling point of over 200°C
and would not contribute significantly to the purgeable material, we are chang-
ing the screening procedure 1n Section 11.1 to use n-decane Instead of n-
dodecane as the cutoff pont. It 1s also acceptable to use n-decane Instead of
n-nonane for determining an area response factor. Please make these changes
1n Section 11.1 of your copy of the method. Since the changes should be quite
clear, we are not sending a revised Section 11.1.
Surrogate Spiking of Tetraglyme-Soluble Samples
Section 11.2 of the method for determining volatile compounds provides for the
addition of the surrogate standard prior to the extraction of the 2-gram sample.
However, 1f the sample 1s completely soluble 1n tetraglyme and no extraction 1s
Involved, as 1s the case for ILS-8, the chlorinated ethanes waste, 1t 1s accept-
able to add an appproprlate amount of the surrogate standard to an aliquot of
the tetraglyme solution of the sample 1n order to conserve surrogate standard.
Quantification of Unlisted Compounds
The sem1quant1tat1ve estimation of the levels of unlisted compounds 1s addressed
on page 7 of the Clarifications section of the Instruction manual. The Instruc-
tions state that quantification should be based on the total 1on current areas
of the unknown and the Internal standard.
However, 1n some cases there may be large amounts of Interfering compounds
that coelute with the Internal standard and cause large errors 1n the total 1on
328
-------
current areas. Therefore, whenever there are coelutlng compounds present that
affect the total ion current area of the Internal standard, e.g., the presence
of benzene that Interferes with Ds-benzene, use the characteristic ion of the
internal standard for the quantification and correct the amount found by
multiplying by-the ratio of the characteristic ion area to total ion area for
the internal standard. The ratio of the characteristic ion area to total ion
area for the internal standard can be obtained from the data from a process
blank run.
329
-------
APPENDIX F
VOLATILES REPORT FORMS
330
-------
GC/MS RUN LOG FOR VOLATILE COMPOUND ANALYSES
to.
«MO
Urn.
PvwIV
331
-------
LISTED VOLATILE COMPOUNDS FOUND
It.
12.
CO
co
ro
M.
1.1
IT. 1.14
It. Ph
-------
LISTED VOLATILE COMPOUNDS FOUND
37.
l.l.l-TH
41.
CO
CO
CO
SI. E<
U.
-------
UNLISTED VOLATILE COMPOUNDS FOUND
CO
to
-------
RELATIVE ABUNDANCES BFB IONS IN VOLATILE COMPOUND ANALYSES
MvrarylD.
NitfiN*.
M
174
171
17i 177
335
-------
RETENTION TIMES AND RESPONSE FACTORS OF VOLATILE INTERNAL STANDARDS
UtVMwylO
MINN*.
H»»»l»» Tim* «4 Otvwi Imwml taiilvi). i
DC!
Itt
•ra
>" * Olnn Irarml Sunttrt
oet
Ci - 04-1 Jt-t
336
-------
VOLATILE SURROGATE STANDARD DATA
CO
CO
NT •*«>•>••
DTE
ra-i
Faro TV
-------
RETENTION TIMES OF VOLATILE COMPOUNDS FROM CALIBRATION RUNS
1.
•._
*_
»g
n.
12.
IS. 1.1
CJ
to
00
I?. l.t-OCK
X. TCTTEO
21.
22. t.2«C£*
24.
H t. I. I-TCK**
21. •tnOMiUmn
M.
10. OCIAcMaCN
31. Vtnyl
3X l.J
M.
37.
-------
RETENTION TIMES OF VOLATILE COMPOUNDS FROM CALIBRATION RUNS
M. t. t.l-TOCll*
». el.
41. McE«
41. 2-OClV Etfor
». T<
0. T
SO.
11. EH
SI. trim*
oo
u>
vo
M. »Xy«»»
SI.
S2.
SI. F
S4.
-------
RESPONSE FACTORS FROM VOLATILE CALIBRATIONS RUNS
II.
12.
H.
It. I. I-OOE*
CO
-fi,
o
I}. I. l-DOEi
It OMME*
It
2O. TCTFt
21.
22. 1.2-OCC*
n.
24.
IS. 1.1.1-TOC**
T.Td
2?.
2t.
OCMcMCN
32.
32. I.
J4. OMclMwN
37.
-------
RESPONSE FACTORS FROM VOLATILE CALIBRATION RUNS
M. I.I.
42. 2-OElVEfl
4X M*
47. T<
•. Ti
SO.
•I. Ed
•2.
M. •-)
CO
St. !4«»T«rEl
S2.
S3. H
S4.
-------
APPENDIX G
SEMIVOLATILES REPORT FORMS
342
-------
GC/MS RUN LOG FOR SEMIVOLATILE COMPOUND ANALYSES
Run No.
*T«t
Dm of
lMI*f
Inn.
wmo
343
-------
LISTED SEMIVOLATILE COMPOUND FOUND
i t«mm «*ii f
3. A«M»
•. a.
ML
II.
14. 1.1.4-Tt
ti.
II.
W. •im»t tta**
it. i.:
20. aiiiiiii»»imii
It. M
2?. N.
33. 3.
-------
LISTED SEMIVOLATILE COMPOUNDS FOUND
(A)
Ja.
cn
JT. 2.1
4a t. 9. 4-TH
41.
43.
44. a.
ML
51. •«. »!
62. 40««
•4.
t?.
70. 3.4<>lcMo
-------
LISTED SEMIVOLATILE COMPOUNDS FOUND
CT>
n.
X. t.
n.
•J.7.1
M. « NMf^lmnl
n.
J.4.».Trt
M. 2.4
•I. 4-CM*
n.
M.
•?. Ti
M. JO*
101.
MO.
HO.
104,
-------
LISTED SEMIVOLATILE COMPOUNDS FOUND
.M*2_
_•»!-
tot.
107.
1CM.
1M.
lit.
112.
111 a.«.«-Ti
117.
c*>
1201 24*
121.
122. 4.4--OOE
123. 4.4--OOO
124.
125. «.4--OOT
131. CM- »
133.
134. On
*tO*»
116.
-------
UNLISTED SEMI VOLATILE COMPOUWS FOUW)
fan (Ml
00
-------
RELATIVE ABUNDANCES OF OFTPP IONS IN SEMIVOLATILE COMPOUND ANALYSES
to.
TO 117
1M 1M
441 442 443
%MO
Nrmtt
349
-------
RETENTION TIMES OF SEMIVOLATILE INTERNAL STANDARDS
HtrnN*
•I Anl
DNP
.1 .D,*.-*!
•nl • DC-AHUM
I -I
CMy
%f • On*"****!*!'""
350
-------
RESPONSE FACTORS OF SEMIVOLATILE INTERNAL STANDARDS
ID
ft MUM*.
.
i
,
.
w
fell
II
Ftal
nil • Or«*Mw
la>iinNi
W
Mwh
•*h
Ann
•h^*
rnMI
DMP
OFTff
P»»
Chrv
••»
•*k • DM •!»•*< >V» »0i»*v««>» Nrmtt
AMI •D1t.liiii^i«iliii ' Okry -DuOny^M
351
-------
SEMIVOLATILE SURROGATE STANDARD DATA
en
ro
«mo
Off FA
FA
FA«
-------
RELATIVE RETENTION TIMES OF SEMIVOLATILE COMPOUNDS FROM CALIBRATION RUNS
«MO
2. 1.44K»»u«l
& 2.4-OMfPvrM
11.
14. 1.7.4-TMlBlu
It t.
to
U.
11. BmiylAle
It. 1.2-OCMmi
ML
21.
22. At
21
24.
2S. HndEl
27. N. N-OM*An*
79.
31.
32. •MOlCMar
34. 2MT
37.
-------
RELATIVE RETENTION TIMES OF SEMIVOLATILE COMPOUNDS FROM CALIBRATION RUNS
to
en
». 2.6-DOTfcol
40. 1.2.4-TOBw
41.
42.
41
44. 2.4-OCVM
4& 44IT1*—
47.
SI.
irE*
51.
S3. 4 • BuMwnol
54. 24toN«*
65.
57. 4-CmiAdd
SB. TmraOBMt
2.4.S-TOPIMI
61.
67. 2-CWfaph
63. Biplwiyl
64.
65. 4-MiOuinolliii
67.
W. 4-BrBtAdll
TOl
71. 2-CM^NPM
72.
73. 2. 3-OMiNn
-------
RELATIVE RETENTION TIMES OF SEMIVOLATILE COMPOUNDS FROM CALIBRATION RUNS
7S. DMlHill Ihll
2.»ONT
77. I4IAHMM
79L 2.4-MvriMMll
•t.
92.
•3.
94. IMfllMMt
•7.
M. 2.4.S-TCMH*
(*)
in
en
91.
•2. 4-a.INAnMll
93.
94.
97. TrrnvraMn
99.
101. 2.«-OCNA«l
1O2.
IO1
106. 3. )• OCIBipk
109.
107.
109.
109.
110
-------
RELATIVE RETENTION TIMES OF SEMIVOLATILE COMPOUNDS FROM CALIBRATION RUNS
112. CMtanta
111 7.4.S-T
114. ae+M*t*t
IIS. 2.4-ONAnMM
IIC. T«traOB«*
It?. AMkn«MMM
lit. FtMWllMM
110 H.UOM*
120. 7M«»im»»
-------
RESPONSE FACTORS OF SEMIVOLATILE COMPOUNDS FROM CALIBRATION RUNS
2. 1.4-1
.40*
B. PwiuCIEtt
10.
11.
12.
13.
14. 1.2.4-'
18.
u.
IT. Knurl CMor
CO
cn
11. BMurlAlc
It. 1.
20.
21.
22.
23.
24.
25. Ht>OCthM
27. M.M-OM.AIIII
. 44-a.Tyrid
29. TnuMiBim
31. 2Nflmiu<
33. 2.44>MiP>iol
37. 2.IMMIMW1
n
-------
RESPONSE FACTORS OF SEMIVOLATILE COMPOUNDS FROM CALIBRATION RUNS
. •wow Ac*
. J.
40. 1.2.4-T
42. 4-CIHmiul
4). 4-OAntUM
44. 2,
4S. H
48. 4NTnlin.i
47.
49. I-CI 4 N»**l
SO.
51
sa. 44-
54.
CO
01
00
SS. 4-BtOiiiim
57 4OBj Acid
sa. 2. 4. 110*01
ao.
81.
61. 20N«*
63.
M.
65^
aa.
aa. 24MMHM
W. 4-BcBl Add
TO. 3.4-OCMi*
71.
72.
73. 7.3-MMin*
74.
-------
RESPONSE FACTORS OF SEMIVOLATILE COMPOUNDS FROM CALIBRATION RUNS
75,
. 2.84)NT
77
80. 2.4-DNrhMl
•1.
81. Dlmiiuluiia
BY.
88. 2.4.5-TCUMi
90.
CO
en
91.
97. «O-2NAiri»l
94. 24M-4.MMW
97. TcHluratai
98. 2-CMNAnWn
4-l«yJm«yBlp»>
100.
101. 2.8-OCMAnl
102. 2,4-D
IO3.
105. 3.T-OOBW.
106.
108. 4.4--OOBW1
109.
110.
111.
-------
RESPONSE FACTORS OF SEMIVOLATILE COMPOUNDS FROM CALIBRATION RUNS
112.
113. 2.4.S-T
114.
11*. TMmOMph
117.
111. t
lit. M»
170. 7
121.
122. DDE
173. 000
174.
12S. DOT
12*.
127.
171. M«
12*.
130.
131. OCtHil**
132. TT,
133. D£tH«
134. (llttFk
13S. I
13*. ItUIPyniM
137.
131.
140.
SI. 2-F
S3.
S3.
------- |