R E S E A R^C H TRIANGLE INSTITUTE
PRELIMINARY STUDY ON TOXIC CHEMICALS IN ENVIRONMENTAL AND HUMAN SAMPLES
PART II: PROTOCOLS FOR ENVIRONMENTAL AND HUMAN
SAMPLING AND ANALYSIS
EPA Contract No. 68-01-3849
RTI/1521/00 - 26 S
WORK PLAN (PHASE I)
by
E. D. Pellizzari, M. D. Erickson, M. T. Giguere, T. D. Hartwell,
R. W. Handy, S. R. Williams, C. M. Sparacino, H. Zelon and
R. D. Waddell
Program Manager
Lance Wallace
Total Exposure and Assessment Methodology Study
Quality Assurance and Monitoring Systems Division
Office of Monitoring and Technical Support
U. S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
WASHINGTON, DC 20460
RESEARCH TRIANGLE PARK. NORTH CAROLINA 27709
-------
PREFACE
This volume of the work plan for the Total Exposure and Assessment
Methodology (TEAM) Study contains analytical protocols that will be examined
in the twelve-person Phase I Pilot Study. At the time of Volume II's draft
preparation a few protocols (PNAs in particulate, metals in particulate,
PCBs/Pesticides in air) had not been received from outside laboratories
which had been tapped for information or participation. Upon receiving
these protocols they were submitted to the TEAM Committee members for review
and comment and are incorporated here.
ii
-------
TABLE OF CONTENTS
Preface ii
1. Introduction 1
2. Summary 3
3. Evaluation of Protocols 4
4. Quality Assurance 14
Appendices
A. Analytical Protocols for Toxic Chemicals in Environmental Media . 18
B. Analytical Protocols for Toxic Chemicals in Body Burden Samples . 227
iii
-------
SECTION 1
INTRODUCTION
GENERAL PROGRAM CONCEPT
The goal of this program is to establish an ongoing human exposure
monitoring capability within EPA. The ultimate output will be a field-tested
methodology for measuring human exposure to toxic substances in a defined
geographical area.
The basic approach is to measure individual exposures through the main
exposure pathways of air, drinking water, and food. Instrumentation for
individual measurements (personnel monitors) will either be evaluated in the
field or, where necessary, development programs will be initiated. These
monitors will be used to provide direct measurements of exposure throughout
an individual's 24-hour daily routine. Monitoring in industrial and residen-
tial neighborhoods and in individual households to determine potential expo-
sures will be carried out. To determine long-term cumulative or equilibrium
dosages, body burden (concentrations in body fluids, excreta, etc.) will be
determined for a range of individuals. Using individual activity pattern
studies and probability sampling methods, a frequency distribution of exposures
will be studied. This will permit examination of the relevence of extending
these results to the entire population of the study area.
PROGRAM DESIGN
The program is divided into two phases. Phase I (the subject of this
work plan) is to test methodology (sampling, survey, monitoring, chemical
analysis, and statistical analysis) prior to Phase II, which involves a large
number of samples. Detailed discussions of the program are in Part I of
this work plan.
METHODOLOGY EVALUATION
A major objective of this program is to provide a methodology for obtain-
ing valid estimates of multimedia exposures of individuals to any pollutant
1
-------
of concern which has potential for transmission via the food, vapor, inhalable
particulate, or liquid routes. Therefore, methods for collection and analysis
of the pollutants must be available. To obtain valid estimates of exposure,
samples must be collected using personnel monitors. This means that the most
advanced techniques in this area must be tested. Furthermore, the small
sample volumes collected (as opposed to a "HiVol", for example) will require
the most sensitive chemical analytical methods.
Phase I will select the most likely candidate methods and test these
methods. Based upon these results, the best monitoring protocols will be
selected and validated for use in Phase II.
This volume describes the protocols to be investigated and the criteria
for evaluation in Phase I.
-------
SECTION 2
SUMMARY
This second volume of the work plan presents the chemical analytical
protocols to be utilized in Phase I of the Total Exposure and Assessment
Methodology Monitoring program. Protocols are proposed for the analysis of
air (volatiles, pesticides, PCBs, PNAs, and Metals), breath (volatiles),
water (volatiles, pesticides, PCBs, PNAs, and metals), blood (volatiles,
pesticides, PCBs, PNAs, and metals), urine (pesticides, PCBs, PNAs, and
metals), and food (volatiles, pesticides, PCBs, PNAs, and metals). In some
cases, two or more protocols will be evaluated during Phase I. Based on
this evaluation, the "best" protocol will be selected for use in Phase II.
Selection critieria are discussed.
It is anticipated that some of the compound classes for some or all
media will not provide sufficient data to warrent inclusion in Phase II.
These analyses will be deleted from Phase II efforts.
This volume also discusses the status of efforts to set up a Quality
Assurance Program for this study.
-------
SECTION 3
EVALUATION OF PROTOCOLS
INTRODUCTION
As discussed below and in the Appendices, several protocols are available
for a given analysis. Where possible a preferred protocol has been selected
and will be used in Phase I. Presuming no major problems are encountered and
the analysis is required, the protocol will be used in Phase II. In many
cases, none of the available protocols is clearly superior. In this case,
two or more protocols have been selected for comparative evaluation in Phase
I. If one protocol proves superior in laboratory validation, it will be used
for the field samples. If not, all candidate protocols will be field-tested
(possibly on only part of the samples) and compared. Based on the side-by-
side performance, a protocol will be selected for use in Phase II.
For some analyses, no validated protocol was found. In these cases a
procedure is proposed which will be validated in the laboratory prior to
field testing.
SELECTION OF CANDIDATE PROTOCOLS
Tables 3-1 and 3-2 list the toxic and hazardous chemicals which have
been selected for study in Phase I of this program. Section 6 of Part I
provides the rationale leading to their selection. The analytical protocols
which provide for their collection and analysis in the various matrices are
discussed here.
Halogenated Hydrocarbons and Benzene in Air
Criteria—
The methods must be amenable to personnel monitoring. This restricts
collection devices to either active personnel monitors (pumps with a sorbent
cartridge or filter) and passive monitors (permeation dosimeter). Many
present serious detection limit problems due to their design for work place
-------
Table 3-1. TOXIC CHEMICALS SELECTED FOR MONITORING IN ENVIRONMENTAL MEDIA (PHASE I)
Air
Vapors
Benzene
Chloroform
1,2-Dichloroethane
1, 1, 1-Trlchloroethane
Carbon tetrachloride
Vinylldene chloride
Trichloroethylene
Tetrachloroethylene
BroBodlchloroaethane
Chlorobenzene
1,1, 2-Trlchloroethane
Vinyl chloride
a-BHC
Llndane
Heptachlor
Chlordane
HCB
DDT/DDD/DDE
PCBs
Partlculatc
Arsenic
Cadmium
Lead
Benzo(a)pyrene
Pyrene
Chrysene
Benzo(a)anthrene
Fluoranthene
Benzo (k) f luoran thene
Drinking water
Benzene
Chloroform
1 , 2-Dlchloroethane
1,1, 2-Trlchloroe thane
Carbon tetrachloride
Vlnylldene chloride
Trichloroethylene
Tetrachloroethylene
Bromodlchlorome thane
Chlorobenzene
Vinyl chloride
1,1,1-Trichloroe thane
Arsenic
Cadmium
Lead
Benzo(a)pyrene
Fluoranthene
Benzo(k)fluoranthene
Fluoranthene
Beverages
Chloroform
1,2-DIchloroethane
1,1, 1-Trichloroe thane
1,1, 2-Trlchloroethane
Vlnylldene chloride
Trichloroethylene
Tetrachloroethylene
Bromodlchlorome thane
Chlorobenzene
Vinyl chloride
Food (tent.)
Arsenic
Cadmium
Lead
a-BHC
Llndane
Heptachlor
Heptachlor epoxlde
Chlordane
t-Nonanchlor
Oxychlordane
HCB
DDT/DDD/DDE
PCBs
Household dust
Arsenic
Cadmium
Lead
Benzo(a)pyrene
Fluoranthene
Benzo (k)f luoran thene
Pyrene
Chrysene
Benzo (a) anthrene
a-BHC
Llndane
Heptachlor
Chlordane
HCB
DDT/DDD/DDE
PCBa
Benzo(k)fluoranthene
Pyrene
Chrysene
Benzo(a)anthrene
a-BHC
Llndane
Heptachlor
Heptachlor epoxlde
Chlordane
j:-Nonachlor
Oxychlordane
HCB
DDT/DDD/DDE
PCBs
-------
Table 3-2. TOXIC CHEMICALS SELECTED FOR MEASUREMENT IN HUMAN BODY FLUIDS
AND TISSUES (PHASE I)
Breath
Blood
Urine
Mother's Milk (tent.) Hair
Benzene
Chloroform
1,2-Dichloroethane
1,1,1-Trichloroethane
1,1,2-Trichloroethane
Carbon tetrachloride
Vinylldene chloride
Trichloroethylene
Tetrachloroethylene
Bromodichloromethane
Chlorobenzene
Benzene
Chloroform
1,2-Dichloroethane
1,1,1-Trichloroethane
1,1,2-Trichloroethane
Carbon tetrachloride
Vinylidene chloride
Trichloroethylene
Tetrachloroethylene
Bromodichloromethane
Chlorobenzene
Arsenic
Cadmium
Lead
6-BHC
Lindane
Heptachlor expoxide
Jt-Nonachlor
Oxychlordane
HCB
DDT/DDD/DDE
PCBs
Benzo(a)pyrene
Fluoranthene
Benzo(k)fluoranthene
Pyrene
Chysene
Benzo(a)anthrene
Arsenic
Cadmium
Lead
B-BHC
^-Nonachlor
Oxychlordane
Heptachlor epoxide
PCBs
Benzo(a)pyrene
Fluoranthene
Benzo(k)fluoranthene
Pyrene
Chrysene
Benzo(a)anthrene
Benzene
Chloroform
1,2-Dichloroethane
1,1,1-Trichloroethane
1,1,2-Trichloroethane
Carbon tetrachloride
Vinylidene chloride
Trichloroethylene
Tetrachloroethylene
Bromodichloromethane
Chlorobenzene
Arsenic
Cadmium
Lead
PCBs
Arsenic
Lead
Cadmium
B-BHC
Lindane
Heptachlor epoxide
^-Nonachlor
Oxychlordane
HCB
DDT/DDD/DDE
PCBs
-------
monitoring where levels are much higher than would be encountered during
ambient monitoring.
Candidate Methods—
The method developed by RTI is the only known candidate method for
general volatiles. A passive dosimiter for vinyl chloride appears to be
sufficiently sensitive for this application and will be tested.
Pesticides and PCBs in Air
Criteria—
As with the halogenated hydrocarbons and benzene, collection must be by
a personnel monitor and analysis must be both sensitive and selective.
Candidate Methods--
No validated protocols are currently available. Two candidate methods
are being considered:
(1) Pesticides and PCBs are being collected by EPA/HERL-RTP using per-
sonnel monitoring pumps with a polyurethane foam (PUF) collection
device. The samples can then be solvent-desorbed and analyzed by
GC/ECD (personal communication, R. G. Lewis, EPA, February 1980).
The method has had only limited field testing, but appears suffi-
ciently promising to warrant inclusion as a protocol for use in
this program.
(2) As an adaptation of the method for volatiles in air (Tenax) used
by RTI, pesticides could be collected on a filter ahead of the
Tenax cartridge and analyzed by thermal desorption/GC. This
method has never been investigated, but has one distinct advantage
since with thermal desorption, the whole sample is analyzed,
increasing the sensitivity over solvent desorption by a factor of
100-1000.
Protocols to be Evaluated—
Both methods discussed above will be investigated further during Phase I
PNAs in Air
No validated personnel sampling methods for PNAs in air are available.
Three methods will be assessed in this study:
(1) A NIOSH HPLC/Fluorescence method,
-------
(2) Thermal Desorption/GC/MS of the filters from the volatile personnel
samplers, and
(3) Analysis of the polyurethane foam extracts collected for Pesticides
and PCBs by GC.
Two restrictions hamper this study:
(a) Any PNA-specific sampling would add another monitoring device
to the test subject; and
(b) The concentrations of the PNAs in air are so low that sampling
rates far in excess of those provided by personnel pumps would be
required to obtain sufficient sample. Therefore, the analytical
technique must compensate for this reduction (•^thousand fold) in
sample size.
Due to the lack of data on any of these techniques, all three will be
tested. The first step will involve the determination of the ambient PNA
levels which would be observable. If these levels are higher than generally
reported ambient levels, the Project Officer will be consulted.
The methods which provide adequate sensitivity will be evaluated for
selectivity, ease of use, and parity with the QA laboratory.
Metals in Air
The standard protocols for metals in air involve high volume air samplers
generally weighing tens or hundreds of kg, clearly unsuitable to personnel
monitoring. Three methods were considered for this program.
(1) Atomic Absorption Spectrometry (AA) of glass fiber filters from
personnel samples, (RTI-in-house methodology)
(2) X-ray Fluorescence Spectrometry (XRF) of Teflon filters from person-
nel samples.
(3) Anodic Stripping Voltametry (SAV) of filters from personnel samples.
o
XRF was selected because the technique is very sensitive (50 ng/m
should be observable); it is non-destructive, so the filters may be retrieved
for Pesticide, PCB, and/or PNA analysis; and automated analytical instrumenta-
tion available in Dr. Robert Stevens' laboratories (ERL-RTP) can perform the
analyses efficiently.
-------
Halogenated Hydrocarbons and Benzene in Water
This area has been well-researched and a large number of methods have
been used in recent years. Several methods were considered:
(1) EPA/EMSL-CI (Bellar and Lichtenberg, Purge and Trap)
(2) EPA/HERL-CI (Closed Loop Stripping)
(3) RTI-Halogenated Hydrocarbon Five City Study (3-1) (Purge and Trap)
(4) RTI-Master Analytical Scheme
Evaluation Criteria—
The first method above is an "on-line" method, whereas the others generate
an extract or sorbent cartridge which is stored for later analysis. This is
advantageous in that samples may be extracted independent of instrumental
availability.
A major option which must be considered is the GC detection system. The
Bellar and Lichtenberg technique may be performed using either GC/Selective
detector (HECD and PID are currently favored) or MS. The other three utilize
MS. There are two distinct advantages to MS: greater compound identification
confidence and archival of data (magnetic tape) for subsequent re-examination
for non-target compounds. On the other hand, non-mass spectrometric detection
is less expensive, since the data interpretation time is greatly reduced.
The techniques will be evaluated for reproducibility, sensitivity,
selectivity, purge efficiency and parity with the QA laboratory.
Selection of Methods for Evaluation—
All methods, except the "RTI-Halogenated Hydrocarbon Five City Study"
method (number 3), will be evaluated. Experiments at RTI (under the Master
Analytical Scheme contract) have shown that the RTI-MAS method is superior to
the other RTI method. This method compensates for the poorer sensitivity of
MS by using a larger volume (200 mL) of water and increases purge efficiency
through addition of sodium sulfate to the water sample to give a high ionic
strength.
The Bellar and Lichtenberg technique remains the "standard for comparison"
due to its wide utilization in the U.S. Its major drawback is the use of
packed columns. Another disadvantage for this study is the need for PID or
FID detection for benzene in addition to the HECD for the halogenated compounds
-------
The major advantages of the closed loop stripping are the large sample
concentration factor (4000 ml to .01 ml) and the use of glass capillary GC.
Pesticides and PCBs in Water
The candidate pesticides have been determined in water using a variety
of methods which involve solvent extraction and GC analysis (usually GC/ECD).
The principle option is in GC detector: ECD, HECD, positive ion mass spectro-
metry and negative ion chemical ionization mass spectrometry being the most
sensitive and selective.
Evaluation Criteria—
Extraction efficiency and levels of interferences are the major evalua-
tion criteria for the extraction/cleanup aspects of the protocols. The
analytical evaluations will encompass selection of GC column and potential
alternatives to ECD. This latter option will be pursued only if ECD proves
inconclusive due to high background or ambiguous peak identifications.
Candidate Methods—
Two validated EPA methods have been considered:
(1) The EMSL-CI (Method 608) method, and
(2) A U. of Miami (Dr. Pfaffenberger) method.
The major difference is in choice of extraction solvent. Both methods
utilize ECD as the GC detection system.
Protocols to be Evaluated—
Both the methods will be evaluated. Depending upon complexity of
samples and sensitivity required, HECD, PIMS, and NICIMS may be substituted
for ECD if distinct advantages are offered.
PNAs in Water
It is hoped that the same extracts used for analysis of Pesticides and
PCBs in water may be analyzed for PNAs. This will require validation.
Three analytical procedures were considered for evaluation:
(1) glass capillary GC/MS
(2) glass capillary GC/FID, and
(3) HPLC/Fluorescence (Method 610, EMSL-CI)
The HPLC/fluorescence method will be used in Phase I since it is already
developed and can be used without a lengthy development/validation effort.
10
-------
Metals in Water
Metals in water will be analyzed by graphite furnace Atomic Absorption
spectrometry. This technique is routinely performed at RTI and no known
alternatives present distinct advantages.
Halogenated Hydrocarbons and Benzene in Blood
Two methods have been considered:
(1) The RTI-Halogenated Hydrocarbon Five City study method and
(2) The U. of Miami method.
The former entails a headspace purge of the blood onto a Tenax cartridge for
later thermal desorption/GC/MS analysis. The latter is a modification of
the Bellar and Lichtenberg method for water. The foaming of the blood is
controlled by a chemical antifoam. The detector usually employed is HECD.
Neither method has been validated by interlaboratory testing.
Both methods will be evaluated in this study. Criteria for selection of
a final protocol include sensitivity, selectivity, efficiency, and parity
with QA laboratory.
Benzene in Blood
As discussed in Part I of this work plan, benzene levels in blood have been
shown to be generally near the limit of detection for available analytical
methodology. The protocol developed by RTI is very sensitive and has been
previously validated, so it will be used to analyze for benzene in blood.
Pesticides and PCBs in Blood
Two methods are under consideration:
(1) The HERL-RTP (Thompson) method, and
(2) An RTI modification which substitutes glass capillary GC/MS
analysis for GC/ECD.
The complexity of the samples (and therefore the need for MS selecti-
vity), target compound concentration (and therefore need for ECD sensitivity)
and the need for maintaining parity with the QA laboratory will determine
which method is preferable.
PNAs in Blood
No standard procedures have been obtained for consideration. It is
hoped that the sample extracts for Pesticides and PCB determination may be
11
-------
analyzed for PNAs. This will require validation. Three analytical procedures
were considered for evaluation:
(1) glass capillary GC/MS
(2) glass capillary GC/FID, and
(3) HPLC/Fluorescence
Metals in Blood
Blood will be analyzed for metals using graphite furnace Atomic Absorption
spectrometry. This technique is routinely performed at RTI and no known
alternatives present distinct advantages.
Halogenated Hydrocarbons and Benzene in Urine
Two methods have been considered:
(1) The RTI-Halogenated Hydrocarbon Five City Study Method, and
(2) The U. of Miami Method.
These methods are analogous to those for blood and evaluation criteria
will be the same. Both methods will be evaluated.
Pesticides and PCBs in Urine
Two methods are under consideration:
(1) The HERL-RTP (Thompson) method, and
(2) An RTI modification which substitutes glass capillary GC/MS
for the GC/ECD analysis.
As with blood, sample complexity, target compound concentration, and the need
for maintaining parity with the QA laboratory will determine which method is
preferable.
PNAs in Urine
As with blood, no standard procedures have been obtained for consi-
deration. It is hoped that the sample extracts for Pesticide and PCB determi-
nation may be analyzed for PNAs. This will require validation. These analy-
tical procedures were considered for evaluation:
(1) glass capillary GC/MS
(2) glass capillary GC/FID, and
(3) HPLC/Fluorescence
12
-------
Metals in Urine
Metals will be analyzed by graphite furnace AA. No known methods are
available which are more effective.
References
3-1 Pellizzari, E. D., M. D. Erickson and R. A. Zweidinger, "Analytical
Protocols for Making a Preliminary Assessment of Halogenated Organic
Compounds in Mass and Environmental Media", EPA-560/13-79-010, September
1979.
13
-------
SECTION 4
QUALITY ASSURANCE
Quality assurance (QA) has been recognized as an integral part of this
research project since its inception. The Quality Assurance Program consists
of quality control (blanks and spiked controls to determine artifacts, losses,
etc.)> internal quality assurance (procedure monitoring), and external quality
assurance (duplicate samples analyzed by an independent laboratory to compare
results). The details of the QA program for each analysis are in each protocol
(see Appendices).
Tables 4-1 and 4-2 list the number of samples and sample types to be
obtained in Phase I of this study.
The general objectives of the external QA program are:
(1) The QA lab will analyze about 10% of the samples (duplicate vials,
cartridges, filters, etc., where possible).
(2) Encoded (blind) blanks and spiked controls will be included with
the field samples.
(3) Both qualitative and quantitative results will be compared between
the principle and QA lab.
(4) The results of the QA will be reported along with the primary
results. Any discrepancies will be investigated and remedied prior to Phase
II.
Commitments from laboratories to perform external QA are being obtained.
The current status of this effort is listed in Table 4-3.
14
-------
Ln
Table 4-1. NUMBER OF SAMPLES AND SAMPLE TYPES PLANNED FOR ENVIRONMENTAL MEDIA
IN PHASE I (PER TRIP)
Matrix
Alr-Volatlles
Air- Aerosols
Water
Beveragea &
Foodstuffs
Dust
Food
3RTI.
bREAL.
°EMSL/RTP.
dHERL/RTP.
6FSU.
fUC/Davis.
^ISOH.
hEMSL/CI.
Chemical group
Vinyl chloride
Benzene + halocarbons
Pesticides
PCBs
Metals
PNAa
Benzene + halocarbons
Metala
PNAa
halocarbons
Metala
PNAa
Pesticides
PCBa
Metals
Pesticides
PCBa
No. of field
samples
<
>a
f
<
81,"
271
\
1
No. of duplicate
samples
?
la
la
1
1
8
8
1
31
i;
-
No. of QC No. of QA No. of broad Total
samples samples spectrum samples samples
9" 9b 8U,1C 104
9» 9rf - 113
1 ld - 12
la ld - 12
le lf - 12
la 1« - 12
9° 9^ 8*.8J 123
9a 9* - 107
la lh - 12
- - 30
9 19
l" -. - 11
ll 1A ~ "
la ld - 12
9
- - 9
9
Total 618
j
FDA.
HERL/CI.
-------
Table 4-2. NUMBER OF ANALYSES PLANNED FOR BIOLOGICAL FLUIDS AND TISSUE IN PHASE I
(PER TRIP)
Matrix
Breath
Blood
Urine
Hair
Chemical group
Benzene -*• Volatile Halogenated Compounds
Benzene
Volatile Halogenated Compounds
Pesticides
PCBs
PNAs
Elements
Pesticides
PCBs
PNAs
Elements
PCBs
Elements
No. of field
samples
21a
18a
18d
10a
18a ,
6a,f
18a
18a
18a ,
6a.f
18a
3a
3a
No. of duplicate
samples
21a
2a
2d
2C
2a
2a
2a
2a
2a
-
No. of QC
samples
8a
8a
8d
8a
8a
8a
8a
8a
8a
8a
8a
8a
8a
No. of QA
samples
3b
2d
2a
2e
2e
2
28
2e
2e
2
28
le
1
No. of broad Total
spectrum samples samples
2a,c „
30
la 31
30
30
16
30
30
30
16
30
12
12
Total 352
Analysis by RTI.
Analysis by IITRI (requires IITRI sampling cartridges).
Full interpretation of GC/MS data from selected samples. Does not require additional sampling.
Analysis by U. Miami.
6Analysis by HERL/RTP.
Selected samples analyzed. If positive results are obtained, all samples will be analyzed.
gHERL/Cl.
-------
Food
Table 4-3. TEAM LEADERS IN PRIMARY AND QA LABORATORIES FOR
PUBLIC HEALTH INITIATIVE CHEMICAL ANALYSIS
Medium
Air
Breath
Water
Blood
Urine
Hair
Class
Volatiles
Pesticides/PCBs
PNAs
Metals
Volatiles
Volatiles
Pesticides/PCBs
PNAs
Metals
Volatiles
Pesticides
PNAs
Metals
Volatiles
Pesticides/PCBs
PNAs
Metals
PCBs
Elements
Primary Lab (Team Leader)
RTI (Sparacino)
RTI (Sparacino)
RTI (Sparacino)
FSU (Nelson)
RTI (Erickson)
RTI (Sparacino)
RTI (Sparacino)
RTI (Sparacino)
RTI (Handy)
U. Miami (Pfaf fenberger)
RTI (Erickson)
RTI (Erickson)
RTI (Handy)
RTI (Erickson)
RTI (Erickson)
RTI (Erickson)
RTI (Handy)
RTI (Erickson)
RTI (Handy)
QA Lab (Team Leader)
EMSL-RTP (Clements)
HERL-RTP (Lewis)
NIOSH (Larkin)
U. C. at Davis (Cahill)
IITRI (Knotoszynski)
HERL-CI (Kopfler)
HERL-RTP (Lewis)
EMSL-CI (Booth)
EMSL-CI (Booth)
RTI (Erickson)
U. Miami (Pfaf fenberger)
To be determined
HERL-CI (Kopfler)
U. Miami (Pfaf fenberger)
U. Miami (Pfaf fenberger)
To be determined
HERL-CI (Kopfler)
HERL-RTP (Lewis)
HERL-CI (Kopfler)
Volatiles
Pesticides/PCBs
PNAs
Metals
FDA (Lombardo)/RTI (Pellizzari)
FDA
FDA(?)
FDA
RTI
RTI
RTI
-------
APPENDIX A
ANALYTICAL PROTOCOLS FOR TOXIC CHEMICALS IN ENVIRONMENTAL MEDIA
Page
1. Personal Monitoring of Vapor-Phase Organic Compounds in Ambient
Air (RTI) 19
2. Polynuclear Aromatic Hydrocarbons in Drinking Water 52
3. Vinyl Chloride From Personal Monitoring Device 65
4. Arsenic, Cadmium and Lead in Ambeint Air Particulate 72
5. Organochlorine Pesticides and PCBs in Air 73
6. Volatile Organochlorides and Benzene By the Purge and Trap
Method 81
7. Determination of Organic Contaminants by Grob Closed-Loop-
Stripping Analysis (CLSA) 110
8. Analysis of Purgeable Organic Compounds in Water (Master
Analytical Scheme) 12Q
9. Polynuclear Aromatic Hydrocarbons in Air 135
10. Organochlorine Pesticides and PCBs in Drinking Water 142
11. Determination of Organochloride Pesticides and Metabolites in
Drinking Water (U. of Miami) 163
12. Sampling and Analysis of Arsenic, Cadmium, and Lead in Drinking
Water (RTI) 174
13. Sampling and Analysis of Arsenic, Cadmium and Lead in Water
(EMSL-CI) 182
14. Volatile Halogenated Hydrocarbons in Beverages and Foodstuffs
(RTI) 197
15. Polynuclear Aromatic Hydrocarbons from Household Dust 215
16. Sampling and Analysis of Arsenic, Cadmium, and Lead in House
Dust (RTI) 216
17. Organochlorine Pesticides and PCBs in Household Dust 226
18
-------
ANALYTICAL PROTOCOL: PERSONAL MONITORING OF VAPOR-PHASE ORGANIC COMPOUNDS
IN AMBIENT AIR (RTI)
1.0 Principle of Method
Recovery of volatile organics from Tenax GC is accomplished by thermal
desorption and purging with helium into a liquid nitrogen cooled nickel
capillary trap (1-3) and then the vapors are introduced into a high resolution
glass gas chromatographic column where the constitutents are separated from
each other (2,4). Characterization and quantification of the constituents
in the sample are accomplished by mass spectrometry either by measuring the
intensity of the total ion current signal or by extracted ion current profile
(2,5,6). The analysis system is shown in Figure 1.
2.0 Range and Limits of Detection
The linear range for the analysis of volatile organic compounds depends
upon two principal features. The first is a function of the breakthrough
volume of each specific compound which is trapped on the Tenax GC sampling
cartridge and the second is related to the inherent limits of detection of
the mass spectrometer for each organic (2,5-8). Thus, the range and the
maximum limit of detection are a direct function of each compound which is
present in the original ambient air. The linear range for quantitation
using glass capillaries on a gas chromatograph/mass spectrometer/computer
(GC/MS/COMP) is generally three orders of magnitude [5-5,000 ng (5-8)].
Table 1 lists the overall detection limits for some examples of volatile
organics which are based on these two principles.
3.0 Interferences
For the target compounds in Table 1, no interferences have been observed.
Particular attention must be paid to the preparation of clean collection
devices and the use of appropriate blanks and controls to establish that the
background contaminants have been removed. Otherwise, false positive detec-
tion of chloroform, toluene or benzene may occur.
Precautions must be taken for sampling caustic atmospheres which contain
levels of NO and molecular halogens greater than 2-5 ppm and 25 ppb,
A
19
-------
PURGE
GAS
TWO
POSITION
VALVE
THERMAL
DESORPTION
CHAMBER
CAPILLARY
GAS
CKFOMATOGRAPH
x\
CARRIER
GAS
1
yx
"N
CAPII I AF
HEATED
BLOCKS
EXHAUST
TRAP
Figure 1. Analytical system for analysis of organic vapors in ambient air.
20
-------
Table 1. APPROXIMATE MEASURED LIMITS OF DETECTION AND QUANTIFIABLE
LIMITS FOR SELECTED VAPOR-PHASE ORGANICS IN AMBIENT AIR
Compound
Benzene
Chloroform
1, 2-Dichlo roe thane
1, 1, 1-Trichloroethane
Carbon tetrachloride
Vinylidene chloride
Trichloroethylene
Tetrachloroethylene
Bromodichloromethane
Chlorobenzene
1,1,2-Trichloroethane
m-Di chlorobenzene
m/z
78
83/85
98/62
97/99
117/119
96/98
130/132
164/166
127/83
112/114
97/99
146/148
LOD3
, 3
yg/m
0.08
0.08
0.12
0.16
0.24
0.12
0.16
0.24
0.24
0.16
0.16
0.20
ppt
26
17
31
31
39
31
31
36
37
36
31
34
QL
yg/m
0.40
0.40
0.60
0.80
1.20
0.60
0.80
1.20
1.20
0.80
0.80
1.00
PPt
128
85
153
154
195
156
154
179
184
179
154
169
Limit of Detection (LOD) was defined as S/N = 4 for mAz ion selected
for quantification. Quantification Limit (QL) was defined as 5 x LOD
or S/N = 20. Limits are based on a collection volume of 25 Si or
breakthrough volume (70°F), which ever is smaller, for 1.5 cm x 8.0
cm Tenax GC bed volume and mass spectrometer response to that compound.
21
-------
respectively (5,9,10). Quenching agents incorporated into the particulate
filter should be employed in these cases (5,9,10).
4.0 Precision and Accuracy
The reproducibility of this method has been determined to range from
+10 to +30% of the relative standard deviation for different substances when
replicate sampling cartridges are examined (4-10). The inherent analytical
errors are a function of several factors: [I] the ability to accurately
determine the breakthrough volume and its relation to field sampling condi-
tions for each of the organic compounds identified; [2] the accurate measure-
ment of the ambient air volume sampled; [3] the percent recovery of the
organic from the sampling cartridge after a period of storage; [4] the
reproducibility of thermal desorption for a compound from the cartridge and
its introduction into the analytical system; [5] the accuracy of determining
the relative molar response ratios between the identified substance and the
external standard used for calibrating the analytical system; [6] the repro-
ducibility of transmitting the sample through the high resolution gas chroma-
tographic column; and [7] the day-to-day reliability of the MS/COMP system
(1-12).
The accuracy of analysis is generally +10-30% but depends on the chemical
and physical nature of the compound (2,5,6,7,12).
5.0 Apparatus and Reagents
5.1 Collection and Analysis Devices
5.1.1 Personnel Monitor Pump
A personal monitor pump (MSA Co. - Model C-200) is used for sample
collection. Flow rates are adjusted to ^0.05 £/min for an 8 hr collection
o
period. Flows are adjusted such that a total volume of ^0.025 m air is
sampled for a given collection period.
5.1.2 Sampling Cartridges
The sampling tubes are prepared by packing a ten centimeter long by 1.5
cm i.d. glass tube containing 8 cm of 35/60 mesh Tenax GC with glass wool in
the ends to provide support (2,11). Virgin Tenax (or material to be recycled)
is extracted in a Soxhlet apparatus for a minimum of 18 hours each time with
methanol and n-pentane prior to preparation of cartridge samplers (2,11).
After purification of the Tenax GC sorbent and drying in a vacuum oven at
22
-------
120°C for 3 to 5 hours at 28 inches of water, all the sorbent material is
meshed to provide a 35/60 particle size range. Meshing and all further
cartridge preparation is conducted in a "clean" room. Cartridge samplers
are then prepared and conditioned at 270°C with a purified helium flow of 30
ml/min for 120 min. Prior to entering the Tenax GC cartridge the helium is
purified by passing through a liquid N~ cooled cryogenic trap. The conditio-
®
ned cartridges are transferred to Kimax (2.5 cm x 150 cm) culture tubes,
immediately sealed using Teflon-lined caps and cooled. This procedure is
performed in order to avoid recontamination of the sorbent bed (2,12).
5.1.3 Inlet Manifold
An inlet manifold for thermally recovering vapors trapped on Tenax
sampling cartridges is used and is shown in Figure 1 (1-4).
5.1.4 Gas Chromatograph
A Varian 1700 or a Pye Unicam 102 gas chromatograph is used to house
the glass capillary column and is interfaced to the inlet manifold on the
Varian MAT CH-7 or LKB 2091 systems, respectively. A mass flow controller
(Tylan) is used to precisely control the carrier gas. Such an analytical
system was presented schematically in Figure 1.
A jet separator is employed to interface the glass capillary column to
the mass spectrometer on the Varian MAT CH-7 GC/MS/COMP or LKB 2091 systems.
The separator is maintained at 240°C (2,5).
5.1.5 Mass Spectrometer/Computer
A Varian MAT CH-7 or LKB 2091 mass spectrometer capable of a resolution
of 1500-2,000 equipped with single ion monitoring capability is used in
tandem with the Varian 1700 or Pye Unicam 102 gas chromatograph and interfaced
to a Varian 620/L or PDP 11/04 computer, respectively (Figure 1).
5.2 Reagents and Materials
All reagents used are analytical reagent grade. All solvents (Burdick
& Jackson) are redistilled before their use.
6.0 Procedure
6.1 Cleaning of Glassware
All glassware is washed in Isoclean/water, rinsed with deionized distil-
led water, acetone and air dried. Glassware is heated to 450-500°C for 2
hours to insure that all organic material has been removed prior to its use.
23
-------
6.2 Collection of Volatile Organics in Ambient Air
For large sample volumes, it is important to realize that the total
volume of air may cause the elution of compounds through the sampling tube
if their breakthrough volume is exceeded. The breakthrough volumes of some
of the volatile organics are shown in Table 2 (2,3,6,7). These breakthrough
volumes have been determined and verified by previously described techniques
(2,5,6). The breakthrough volume is defined as that point at which 50% of a
discrete sample introduced into the cartridge is lost. Although the identity
of a compound during ambient air sampling is not known (therefore, also its
breakthrough volume), the compound can still be quantified after identifica-
tion by GC/MS/COMP once the breakthrough volume has subsequently been esta-
blished. Thus, for calculating concentration, the last portion of the
sampling period which represents the volume of air sampled prior to break-
through is selected. For cases in which the identity of a volatile organic
compound is not known until after GC/MS, the breakthrough volume is subse-
quently determined.
Previous experiments have shown that the organic vapors collected on
Tenax GC sorbent are stable and can be quantitatively recovered from the
cartridge samplers up to 4 weeks after sampling when they are tightly closed
in cartridge holders and placed in a second container that can be sealed,
protected from light and stored at -20°C [Table 3 (1,2,6,7)].
6.2.1 Deuterated Standards
The use of deuterated compounds provides for an assessment of any
premature breakthrough (and thus reduced collection efficiency) which may
occur if the total vapor-phase organic load exceeds 1/10 of the cartridge
capacity during field sampling, d -Bromoethane (B.P. 34°C), dc-tetrahydro-
D 8
furan (B.P. 65), dg-benzene (B.P. 80), d1()-cyclohexene (B.P- 83) and d -
chlorobenzene (B.P. 132) are loaded as a discrete zone onto at least 10% of
all Tenax GC sampling cartridges prior to sampling. Using GC/MS/COMP the
exogenous deuterated compounds are differentiated from the endogenous vapor-
phase organics in ambient air.
The addition of these deuterated standards is performed by injecting
1.0 ml air/vapor of the substances onto the "front" end of the cartridge.
24
-------
Table 2. TENAX GC BREAKTHROUGH VOLUMES FOR TARGET COMPOUNDS3
Compound
chloroform
carbon tetrachloride
1, 2-dichloro ethane
1,1,1-trichloroethane
tetrachloroethylene
trichloroethylene
chlorobenzene
b.p.
(°C)
61
77
83
75
121
87
132
Temperature (°F)
50
56
45
71
31
481
120
1989
60
41
36
55
24
356
89
871
70
32
28
41
20
261
67
631
80
24
21
31
16
192
51
459
90
17
17
24
12
141
37
332
100
13
13
19
9
104
28
241
aFor a Tenax GC bed of 1.5 x 8.0 cm.
25
-------
Table 3. RECOVERY OF TARGET COMPOUNDS AFTER STORAGE
Compound
Benzene
Chloroform
1, 2-Dichloro ethane
1,1,1-Trichloroethane
Carbon tetrachloride
Vinylidene chloride
Trichloroethylene
Tetrachloroethylene
Bromodichloromethane
Chlorobenzene
1,1, 2-Tr ichloroethane
m-Dichlorobenzene
1 day
100
100
100
100
ND
100
100
100
ND
100
100
100
1 wk
107 + 26
83 + 14
100 + 5
87 + 17
ND
92
87 + 17
92 + 1
ND
87 + 3
95 + 3
89 + 15
2 wk
69+8
49 + 4
100 + 7
71+7
ND
ND
71+7
78 + 4
ND
80+7
92 + 2
85 + 3
26
-------
An air/vapor mixture stream containing 200-400 ng/ml is generated using
permeation tubes of each compound with a permeation system (4,5,8,9).
The quantity of each substance on the sampling cartridges is determined
by GC/MS/COMP after field sampling and the percent recovery is compared to
control (unused) cartridges carried to and from the field sampling site and
subjected to the same storage regime. Statistically significant differences
are attributed to premature breakthrough.
6.2.2 Quantification Standards
Unique substances may be added as internal standards during sampling.
Examples are the deuterated compounds listed under 6.2.1. However, the
volume of air sampled is accurately known and thus external standards may be
introduced into the cartridge prior to its analysis. Three standards,
hexafluorobenzene, octafluorotoluene, and iodotoluene are used for the
purpose of calculating RMRs and the levels in ambient air. Previous research
has shown that their retention times span the chromatographic range of
analysis (SE-30 coated capillary) and they do not interfere with the analysis
of unknown compounds in ambient air samples.
The external standards (300-400 ng) are injected into the sampling
cartridges as a 1.0 ml air/vapor mixture using a gas sampling syringe. The
air/vapor mixture is synthesized using permeation tubes and a permeation
system (4,5,8,9).
6.3 Analysis of Samples
The instrumental conditions for the analysis of volatile organics on
the sorbent Tenax GC sampling cartridge is shown in Table 4. The thermal
desorption chamber and the six port Valco valve are maintained at 270°C.
The jet separator is maintained at 245°. The mass spectrometer is set to
scan the mass range from approximately 20-350. The helium purge gas through
the desorption chamber is adjusted to 15-20 mL/min. The nickel capillary
trap on the inlet manifold is cooled with liquid nitrogen. In a typical
thermal desorption cycle, a sampling cartridge is placed in the preheated
desorption chamber and the helium gas is channeled through the cartridge to
purge the vapors into the liquid nitrogen capillary trap [the inert activity
efficiency of the trap has been shown in a previous study (4,10)]. After
the desorption has been completed, the six-port valve is rotated and the
27
-------
Table 4. OPERATING PARAMETERS FOR GLC-MS-COMP SYSTEM
Parameter
Setting
Inlet-manifold
desorption chamber and valve
capillary trap - minimum
maximum
thermal desorption time
He purge flow
GLC
85 m glass WCOT BaC03 SE-30 (0.8-1.0 y film)
carrier (He) flow
separator/transfer line
MS
Varian MAT CH-7
scan range
scan cycle, automatic-cyclic
filament current
multiplier
ion source vacuum
LKB 2091
270°C
-195°C
240°C
8 min
15 ml/min
24-240°C, 4°C/min
vL.25 ml/min
245°C
m/z 20 -»• 350
1 sec/decade
300 yA
4.0
4 x 10~6 T
scan range
scan cycle, automatic
filament current
multiplier
ion source vacuum
m^z 20 -»- 500
2 sec total
300 yA
4.5
\4 x 10~6 T
28
-------
temperature on the capillary loop is rapidly raised (greater than 10°/min);
the carrier gas then introduces the vapors onto the high resolution GC
column. The glass capillary column is temperature programmed from ambient
to 240°C at 4°C/min and held at the upper limit for a minimum of 10 rain.
After all the components have eluted the column is then cooled to ambient
temperature and the next sample is processed (2).
An example of the analysis of volatile organics in ambient air is shown
in Figure 2 and the background from a blank cartridge is shown in Figure 3.
The high resolution glass capillary column was coated with SE-30 stationary
phase which is capable of resolving a multitude of compounds to allow their
subsequent identification by MS/COMP techniques; in this case over 110
compounds were identified in this chromatogram.
6.3.1 Qualitative Analysis
The mass spectral data are processed in the following manner. First,
the original spectra are scanned and the reconstructed ion chromatogram
(RIC) is extracted and examined. The intensity (RIC) is plotted against the
spectrum number using the software package available. The information will
generally indicate whether the run is suitable for further processing, since
it provides some idea of the number of unknowns in the sample and the resolu-
tion obtained using the particular gc column conditions.
If mass conversion of spectral peak times to peak masses has not been
performed on-the-fly during data acquisition by hardware methods than this
function is next performed by software methods (magnetic systems). In
either case the mass conversion is accomplished by the use of the calibration
table obtained prior to sample analysis for perfluorokerosene. In general
the calibration data are sufficient for an entire day's data processing;
however, it is verified every eight hours.
After the spectra are obtained in mass converted form, processing
proceeds either manually or by computer by comparison to a Library (13).
Compound identification can involve various degrees of certainty. These
levels of identification have been defined as follows:
29
-------
U)
o
till
Mass Spectrum No.
Figure 2. Profile of ambient air pollutants obtained using high resolution gas chrotnatography/
•mass spectrometry/computer.
-------
CO
80-
TOH
«
90-
20-
44
56 68
80 92
TEMPERATURE (*C)
104 116 128
192
Figure 3. Background profile for Tenax GC cartridge blank.
-------
Level I Computer Interpretation. The raw data generated from the
analysis of samples are subjected to computerized deconvolu-
tion/library search and compound identification made using
this approach has the lowest level of confidence. In general
Level I is reserved for only those cases where compound
verification is the primary intent of the qualitative analysis
Level II Manual Interpretation. The plotted mass spectra are manually
interpreted by a skilled interpreter and compared to those
spectra compiled in a data compendium. In general a minimum
of five masses and intensities (+5% S.D.) should match be-
tween the unknown and library spectrum. This level does not
utilize any further information such as retention time since
many compounds the authentic compound may not be available
for establishing retention times.
Level III Manual Interpretation Plus Retention Time/Boiling Point of
Compound. In addition to the effort as described under Level
II, the retention time of the compound is compared to the
retention time which has been derived from previous chromato-
graphic analysis. Also the boiling point of the identified
compound is compared to the boiling points of other compounds
in the near vicinity of the one in question when a capillary
coated with a non-polar phase has been used.
Level IV Manual Interpretation Plus Retention Time of Authentic Com-
pounds . Under this level, the authentic compound has been
chromatographed on the same capillary column using identical
operating conditions and the mass spectrum of the authentic
compound is compared to that of the unknown.
Level V Level IV Plus Independent Confirmation Techniques. This
Level utilizes other physical methods of analysis such as
GC/fourier transform/IR, GC/high resolution mass spectrometry,
or NMR analysis. This Level constitutes the highest degree
of confidence in the identification of organic compounds.
32
-------
6.3.2 Quantitation
The quantitation of constituents in ambient air samples is accomplished
either by utilizing the total ion current monitor or, where necessary, from
extracted ion current profiles. In order to eliminate the need to obtain
complete calibration curves for each compound for which quantitative infor-
mation is desired, the method of relative molar response (RMR) factors
(5-10) is used. Successful use of this method requires information on the
exact amount of standard added and the relationship of RMR (unknown) to the
RMR (standards).
6.3.2.1 RMR Determination
The compounds to be quantified are loaded onto Tenax GC cartridges
using a permeation system (4,5,8,9) or in cases where permeation tubes are
not available the vaporization system shown in Figure 4 is used (5,14).
With the vaporization method helium is purified by passing it through a
cryogenic trap followed by two carbon traps. The standards and substances
to be quantified are prepared in methanol and a 2.0 |j£ solution is injected
through the septum of the heated loading tube (250°C). The vaporized com-
ponents are swept onto the Tenax GC cartridge at a rate of 200 ml/min for 6
min (total He 1.2 £). Because of the low breakthrough volume for methanol
(0.8 H at 70°F), the majority is passed from the cartridge. This system is
used to load relatively non-volatile compounds with breakthrough volumes
>50 2.
The method of calculating RMRs is as follows:
A . /Moles .
(1) RMR - unk unk
unknown/standard A ,/Moles ,
A = system response, height or area determined by integration
or triangulation.
unk = unknown
std = standard
The value of RMR is determined from at least three independent analyses
during analysis of samples (5). Linearity over the dynamic range and an
intercept of zero has been previously described (5,14).
A ,/g ,/GMW ,
0-\ PMR - unk unk unk
C } " Astd/8Btd/GMWBtd
33
-------
METER
TENAXGC CARTRIDGE CARBON TRAPS
Z SEPTUM 3-WAY STOPCOCK / I
/ I •• HeFLOW(30mL/mln)/ [
—vrm—IT-——— rt» n—m4—n0n—
\ /
TEFLON UNIONS
LOADING TUBE WRAPPED
WITH ALUMINUM FOIL
AND HEATING TAPE
Figure 4. Schematic of vaporization unit for loading organics dissolved in methanol onto Tenax
GC cartridges.
-------
A = system response, as above
g = number of grams present
GMW = gram molecular weight
A , -GMW . -g ,
_ unk unk °std
8unk A . .-GMW ....•RMR , , _
std std unk/std
6.3.2.2 Calculation of Ambient Air Concentrations
Since the volume of air taken to produce a given sample is accurately
known and an external standard is added to the sample, then the weight per
cartridge and hence the concentration of the unknown can be determined. The
approach for quantitating ambient air pollutants in this study requires that
the RMR be determined for each constituent of interest during the analysis
of field samples. Every sixth cartridge is a control cartridge for deter-
mining RMRs for each compound (calibration of instrument, storage and
recovery). This means that when an ambient air sample is taken, the exter-
nal standard is added at a known concentration prior to analysis. It is not
imperative at this point to know what the RMR of each of the constituents in
the sample happens to be. However, after the unknowns are identified then
the RMR can subsequently be determined and the unknown concentration calcu-
lated in the original sample using the RMR. In this manner it is possible
to obtain qualitative and quantitative information on the same sample with a
minimum of effort.
Once the quantity of substance per cartridge has been determined, the
level in ambient air is given by
ug , • 1000 L
&unk
m3 • Volume Sampled (L)
7.0 Quality Assurance Program
7.1 Reagent and Glassware Control
Reagent and glassware control is required in order to minimize contamina
®
tion. Sample containers, glassware, etc. are cleaned with Isoclean , rinsed
with distilled/deionized water and heat treated at 450-500°C to insure the
removal of all traces of organic compounds.
35
-------
7.2 Sampling Protocol and Chain of Custody
As part of the quality control procedures, sampling protocols and chain
of custody forms are prepared for each sampling cartridge. Examples of
these forms are given in Tables 5 and 6. The fate of each sampling cartridge
is tracked from the time they are prepared until the data has been reduced
to a finished form.
7.3 Blanks, Controls, Standards, and System Performance Samples
7.3.1 Blanks
Ten percent of the sampling cartridges from each batch are set aside to
serve as blanks to be analyzed for background contamination. After the
preparation of a set of sampling cartridges, one cartridge is checked for
background prior to their committment to field sampling. Blank (unused)
cartridges travel to the field site returned to the laboratory and stored
along with the field samples at -20°C until ready for analysis.
7.3.2 Controls
Ten percent of the sampling cartridges are loaded with the deuterated
compounds listed in 6.2.1. Sampling with control cartridges allow for an
assessment of premature breakthrough if it occurs. Control cartridges are
analyzed along with other samples and since deuterated compounds are employed
the qualitative and quantitative analysis of these air samples proceeds
unimpeded.
7.3.3 Standards for RMR Determination
The compounds listed in Table 1 are loaded onto Tenax GC cartridges
from a permeation system. A minimum of three analyses is required for
determining RMRs for a set of samples which are quantitatively analyzed.
7.3.4 System Performance Mixtures
The system performance standards listed in Table 7 are loaded onto
Tenax GC cartridges (using the vaporization method) to determine mass calibra-
tion and intensity and chromatographic performance of the GC/MS/COMP system.
7.4 Sample Analysis
To insure the accuracy and precision of the data acquired instrument
and chromatographic performance are monitored on a daily basis.
36
-------
Table 5. FIELD SAMPLING PROTOCOL SHEET - A
Date:
Project No. ( )
Operator ( 7
Trip No. (3.5*
Sampler (_"^_)
Area "" ~
Address
Period ()
Site
L J
Location ( _
Sample Code"
IKitten 221 (N)
DC aas«
Saaplinz rate (tea)
V.CUUB ( " Be)
Cad: Tiae Ft3
Start: Tiae Tt3
total: teia) »t3
Pu?ent (D)
Saapllat rate (lalt.)
Saapllni rate (final)
End: Tl*«
Start: Tiae
Total: (ain)
MSA (M)
(Ipn) Sampling rate (init.)
(tpa) Saoplinf rate (final)
End: Tiae Count
Start: Tiae Count
Total: (min) Count
•I/count
(tF=)
(ts=)
Remarks
Volume Air/Cartridge
Time Temp. Wet. Dry
Rel. Humid * Wind Dir./Speed __/_
Cloud Odor ~_
Remarks
Time Temp. Wet. Dry
Fel. Humid * Wind Dir./Speed _/_
Cloud Odor
Remarks
Time Temp. Wet.
Dry
Rel. Humid % Wind Dir./Speed __/_
Cloud Odor,
Remarks
Time Temp. Wet Dry
Rel. Humid I Wind Dir./Speed _/_
Cloud Odor
Remarks
37
-------
Table 6. CHAIN OF CUSTODY RECORD
Research Triangle Institute
Analytical Sciences Division
Chemistry and Life Sciences Group
Research Triangle Park, NC 27709
SAMPLE CODE:
:
* ^«-
Sample Type:
No. of Containers:
Volume Collected:
Volume Analyzed:
Relinquished
By:
Received
By:
Time
•
Date
Operation Performed (aliquot, std. cone.,
remarks, etc.)
38
-------
Table 7. GC/MS/COMP SYSTEM PERFORMANCE STANDARDS
Compound Quantity (ng)
Perfluorokerosene 350
Perfluorotoluene 350
Ethylbenzene 300
£-Xylene 300
n-Octane 300
n-Decane 300
1-Octanal 300
5-Nonanone 300
Acetophenone 300
2,6-Dimethylaniline 300
2,6-Dimethylphenol 300
39
-------
7.4.1 Instrument Calibration
Calibration of mass and intensity of magnetic systems employ perfluoro-
kerosene. Table 8 lists the tolerances for each mass and intensity which
the mass spectrometer must achieve.
Perfluorotoluene in the performance mixture is employed for determining
instrument stability as related by mass resolution and relative ion abun-
dance under GC conditions (5,14). The masses and intensities listed in
Table 8 are compared to the results obtained on a daily basis and for each
set of samples analyzed.
7.4.2 Assessment of Chromatographic Performance
The quality of the chromatography is of utmost importance since the
accuracy and precision of qualitative and quantitative analysis are directly
affected (5,14). Glass capillary columns are evaluated according to the
following criteria:
(1) percent peak asymmetry factor (PAT)
% PAF = | x 100
r
where B = the area of the back half of a chromatographic peak
F = area of the front half of the chromatographic peak both
measured 10% above baseline
(2) effective Height Equivalent to a Theoretical Plate (HETP )
HETP ., = ir-FT-
eff 5.54
where X = the corrected retention distance for sweep time of the
compound,
Y = chromatographic peak width at 1/2 peak height,
L = column length (mm)
(3) separation number (SN)
SN = „ D „ -1
where D = the distance between two peaks,
W1'W2 = widths at !/2 height
40
-------
Table 8. MASS AND INTENSITY TOLERANCES ACCEPTABLE FOR
CALIBRATION OF MAGNETIC INSTRUMENTS FOR QUANTITATION
Perfluorotoluene8
m/_z
69
79
93
117
167
186
217
236
Perfluorokerosene
ZI (C.V.) m/^
33
11
16
43
15
59
100
66
(5)
(10)
(8)
(8)
(7)
(5)
(0)
(4)
51
100
119
131
169
181
219
231
ZI (C.V.)
39
22
100
89
58
62
24
29
(10)
(8)
(0)
(5)
(4)
(6)
(5)
(8)
Q
To be achieved in the chromatography mode.
-------
(4) resolution (R)
2 A W
R =
where AW = average base width,
W = peak width at base
(5) Acidity and Basicity
. . , _ weak base (peak area or height)
acetophenone (peak area or height)
R . . _ weak acid (peak area or height)
y - acet0pnenone (peak area or height)
The use of the compounds listed in Table 7 provides information as to the
degree of adsorption and the type of adsorption mechanisms. 1-Octanol and
5-nonanone serves to determine the extent of deactivation of the glass sur-
face (PAF). The acidity and basicity of the glass capillary column are
assessed by the adsorption of weak bases and acids, respectively (5,14).
The resolution and separation number are determined for the compound
pairs ethylbenzene:£-xylene and octane:decane, respectively. HETP f is
based on octane. Table 9 lists the minimum performance specifications accept-
able for ambient air analysis (5). Figures 5-11 depict extracted ion current
profiles used for calculating performance specifications.
7.4.3 Sequence of Sample Analysis
A strict step-sequence of analysis is followed. Upon mass and intensity
calibration of the MS system, a Tenax GC cartridge loaded with the perfor-
mance mixture is first analyzed. Following the performance mixture a blank
and an RMR standard mixture is analyzed next, then five samples. The cycle
is then repeated. At the beginning of each day the analysis cycle begins with
the performance mixture, blank and RMR standards. Thus, 30% of the cart-
ridges analyzed consist of control samples.
8.0 References
1. Pellizzari, E. D., Development of Method for Carcinogenic Vapor Analy-
sis in Ambient Atmospheres. Publication No. EPA-650/2-74-121, Contract
No. 68-02-1228, 148 pp., July, 1974.
42
-------
Table 9. SPECIFICATIONS OF PERFORMANCE FOR GLASS CAPILLARY COLUMN5
Parameter Test Compound(s) Value + S.D. (C.V.)
Resolution Ethylbenzene:£-Xylene 1.30 + 11 (8)
Separation No. Octane:Decane 73+6 (8)
% Peak Asymmetry Factor 1-Octanol 239 + 141 (59)
Nonanone 130 + 32 (25)
Acetophenone 260 + 34 (13)
Acidity 2,6-Dimethylaniline:Acetophenone 1.00 + 0.07 (9)
Basicity 2,6-Dimethylphenol:Acetophenone 0.82 + 0.03 (5)
8SE-30 WCOT/BaCOa, 0.48 mm i.d. x 75 m, 1 \i film thickness.
-------
V) S>
z u>
IS
ID
CM
E> K
B Z
CE a
iu u
J LU Z
*-CL O
I
c
c
e:
Scan No.
Figure 5. Extracted ion current profile of m/£ 91 for £-xylene
and ethylbenzene used in calculating resolution.
44
-------
in CD
z us
u.
ti
m
o: a:
u. u.
• •
SBJNI-
R55
e
m
iii u
_JbUZ Z
- (LOO
it m M M
Scan No.
Figure 6. Extracted ion current profile of m/z 70 and 84 for
1-octanol used for calculating percent peak asymmetry
factor.
45
-------
Scan No.
Figure 7. Extracted ion current profile of in/z 57 for nonanone
used in calculating percent peak asymmetry factor.
46
-------
e
«
u
«
•e
r
yv^ WA-v-J W
Figure 8. Extracted ion current profile of m/z 43 for octane and decane used in calculating
separation number.
-------
0
ID
E
in o
^f 1^
c
El
L.1
a o.
in
£ £ ?.
c c c.
UJ U
. _i u —
«- Cu O
U. U) —
m/z 105
i
c
Scan No.
Figure 9. Extracted ion current profile of m/z 105 for acetophenone
in calculating percent peak asymmetry factor.
48
-------
vo
lW.ft~
i«7 .
199.9-
121 .
199.9-
122.
1(
4
1
(X
rH
1 1 1
-------
01
IW.
121 _
199.9-
122 .
100
4:10
200
8:20
300
12:30
400
16:40
Scan Time
1
500
25:00
600
25:00
4448.
m. t32
* ft.5M
4M8.
121.936
* 9.599
4448.
122.937
k 9.599
56128.
Figure 11. Extracted ion current profile of m^z 106, 121, and 122 for 2,6-dimethyl-
aniline in performance mixture.
-------
2. Pellizzari, E. D., "Development of Analytical Techniques for Measuring
Ambient Atmospheric Carcinogenic Vapors", Publication No. EPA-600/2-75-
075, Contract No. 68-02-1228, 187 pp., November, 1975.
3. Pellizzari, E. D., J. E. Bunch, B. H. Carpenter and E. Sawicki, Environ.
Sci. Technol., 9, 552 (1975).
4. Pellizzari, E. D., "The Measurement of Carcinogenic Vapors in Ambient
Atmospheres", Publication No. EPA-600-7-77-055, Contract No. 68-02-
1228, 288 p., June, 1977.
5. Pellizzari, E. D. , "Evaluation of the Basic GC/MS Computer Analysis
Technique for Pollutant Analysis", Final Report, EPA Contract No. 68-
02-2998.
6. Pellizzari, E. D. and L. W. Little, "Collection and Analysis of Pur-
geable Organics Emitted from Treatment Plants", Final Report, EPA
Contract No. 68-03-2681, 216 pp.
7. Pellizzari, E. D., unpublished results.
8. Pellizzari, E. D. , "Analysis of Organic Air Pollutants by Gas Chromato-
graphy and Mass Spectroscopy", EPA-600/2-77-100, June 1977, 114 pg.
9. Pellizzari, E. D., "Analysis of Organic Air Pollutants by Gas Chromato-
graphy and Mass Spectroscopy", EPA-600/2-79-057, March 1979, 243 pg.
10. Pellizzari, E. D., "Ambient Air Carcinogenic Vapors Improved Sampling
and Analytical Techniques and Field Studies", EPA-600/2-79-081, May
1979, 340 pg.
11. Pellizzari, E. D., J. E. Bunch, R. E. Berkley and J. McRae, Anal.
Chem., 48, 803 (1976).
12. Pellizzari, E. D., J. E. Bunch, R. E. Berkley and J. McRae, Anal.
Lett., 9, 45 (1976).
13. "Eight Peak Index of Mass Spectra", Vol. I, (Tables 1 and 2) and II
(Table 3), Mass Spectrometry Data Centre, AWRE, Aldermaston, Reading,
RF74PR, UF, 1970.
14. Pellizzari, E. D., et al., "Master Scheme for the Analysis of Organic
Compounds in Water Part III: Experimental Development and Results",
EPA Contract No. 68-03-2704, March 1980.
Written analytical protocol prepared 5/6/80.
51
-------
ANALYTICAL PROTOCOL: POLYNUCLEAR AROMATIC HYDROCARBONS IN DRINKING WATER
1.0 Principle of the Method
This method is applicable for the determination, from drinking water, of
the following polynuclear aromatic hydrocarbons (PAH):
Acenaphthene
Acenaphthylene
Anthracene
Benzo(a)anthracene
Benzo(b)fluoranthene
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(ghi)perylene
Benzo(k)fluoranthene
Chrysene
Dibenzo(a,h)anthracene
Fluoranthene
Fluorene
Indeno(1,2,3-cd)pyrene
Naphthalene
Phenanthrene
Pyrene
This method contains both liquid and gas chromatographic approaches,
depending upon the needs of the analyst. The gas chromatographic procedure
cannot adequately resolve the following four pairs of compounds: Anthracene
and phenanthrene; chrysene and benzo(a)anthracene; benzo(b)fluoranthene and
benzo(k)fluoranthene; and dibenzo(a,h)anthracene and indeno(l,2,3-cd)pyrene.
Unless the purposes of the analysis can be served by reporting a sum for an
unresolved pair, the liquid chromatographic or capillary gas chromatographic
approaches must be used for these compounds. The liquid chromatographic
method will resolve all of the 16 compounds listed above.
This method is recommended for use only by experienced residue analysts
or under the close supervision of such qualified persons.
52
-------
2.0 Range and Detection Limit
The sensitivity of this method is usually dependent upon the level of
interferences rather than instrumental limitations. The limits of detection
listed in Table 1 for the liquid chromatographic approach represent sensi-
tivities than can be achieved in wastewaters in the absence of interferences.
Gas chromatography (packed column) can produce detection limits at least as
low as the HPLC approach, while capillary gas chromatography will produce
significantly lower limits.
3.0 Interferences
Solvents, reagents, glassware, and other sample processing hardware may
yield discrete artifacts and/or elevated baselines causing misinterpretation
of the chromatograms. All of these materials must be demonstrated to be free
from interferences under the conditions of the analysis by running method
blanks. Specific selection of reagents and purification of solvents by
distillation in all-glass systems may be required.
Interferences coextracted from the samples will vary considerably from
source to source, depending upon the diversity of the industrial complex or
municipality being sampled. While a general clean-up technique is provided
as part of this method, unique samples may require additional clean-up
approaches to achieve the sensitivities stated in Table 1.
The extent of interferences that may be encountered using liquid chroma-
tographic techniques has not been fully assessed. Although the chromatographic
conditions described allow for a unique resolution of the specific PAH compounds
covered by this method, other PAH compounds may interfere.
Capillary gas chromatographic methods, with inherently greater resolution
than either HPLC or packed column GC, will minimize the extent of interferen-
ces .
4.0 Precision and Accuracy
The U. S. EPA Environmental Monitoring and Support Laboratory in Cincin-
nati is in the process of conducting an interlaboratory method study to
determine the accuracy and precision of this test procedure.
53
-------
Table 1. HIGH PERFORMANCE LIQUID CHROMATOGRAPHY OF PAHs
Compound
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo (a) anthracene
Chrysene
Benzo (b)f luoranthene
Benzo (k)fluoranthene
Benzo(a)pyrene
Dibenzo(a ,h)anthracene
Benzo (ghi)perylene
Indeno(l,2,3-cd)pyrene
Retention time
(min)
16.17
18.10
20.14
20.89
22.32
23.78
25.00
25.94
29.26
30.14
32.44
33.91
34.95
37.06
37.82
39.21
Detection limit
(Hg/L)3
UV
2.5
5.0
3.0
0.5
0.25
0.10
0.50
0.10
0.20
0.20
1.0
0.30
0.25
1.0
0.75
0.30
Fluorescence
20.0
100.0
4.0
2.0
1.2
1.5
0.05
0.05
0.04
0.5
0.04
0.04
0.04
0.08
0.2
0.1
Detection limit is calculated from the minimum detectable HPLC response
being equal to five times the background noise, assuming an equivalent
of a 2 ml final volume of the 1 liter sample extract, and assuming an
HPLC injection of 2 microliters.
HPLC conditions: Reverse phase HC-ODS Sil-X 2.6 x 250 mm Perkin-
Elmer column; isocratic elution for 5 min using 40% acetonitrile/60%
water, then linear gradient elution to 100% acetonitrile over 25 min-
utes, flow rate is 0.5 mL/min.
54
-------
5.0 Apparatus and Reagents
5.1 Sampling equipment, for discrete or composite sampling
5.1.1 Grab sample bottle-amber glass, 1-liter or 1-quart volume.
French or Boston Round design is recommended. The container must be washed
and solvent rinsed before use to minimize interferences.
5.1.2 Bottle caps-threaded to screw on to the sample bottles. Caps
must be lined with Teflon. Foil may be substituted if sample is not
corrosive.
5.2 Compositing equipment-automatic or manual compositing system
Must incorporate glass sample containers for the collection of a minimum
of 250 ml. Sample containers must be kept refrigerated during sampling. No
Tygon or rubber tubing may be used in the system.
5.3 Separatory funnel-2000 ml, with Teflon stopcock
5.4 Drying column-20 mm ID pyrex chromatographic column with coarse
frit
5.5 Kuderna-Danish (K-D) apparatus
5.5.1 Concentrator tube-10 ml, graduated (Kontex K-570050-1025 or
equivalent). Calibration must be checked. Ground glass stopper (size 19/22
joint) is used to prevent evaporation of extracts.
5.5.2 Evaporative flask-500 ml (Kontes K-57001-0500 or equivalent).
Attach to concentrator tube with springs (Kontes K-66270-0012).
5.5.3 Snyder column-three-ball macro (Kontes K503000-0121 or equiva-
lent) .
5.5.4 Snyder column-two-ball micro (Kontes K-569001-0219 or
equivalent).
5.5.5 Boiling chips-solvent extracted, approximately 10/40 mesh.
5.6 Water bath-heated, with concentric ring cover, capable of tempera-
ture control (±2°C). The bath should be used in a hood.
5.7 HPLC Apparatus:
5.7.1 Gradient pumping system, constant flow.
5.7.2 Reverse phase column, 5 micron HC-ODS Sil-X, 250 mm x 2.6 mm ID
(Perkin Elmer No. 809-0716 or equivalent).
5.7.3 Fluorescence detector, for excitation at 280 nm and emission at
389 nm.
55
-------
5.7.4 UV detector, 254 nm, coupled to fluorescence detector.
5.7.5 Strip chart recorder compatible with detectors, (a data system
for measuring peak areas is recommended).
5.8 Gas chromatograph-Analytical system complete with gas chromatograph
suited for on-column injection and all required accessories including dual
flame ionization detectors, column supplies, recorder, gases, syringes. A
data system for measuring peak areas is recommended.
5.9 Chromatographic column-250 mm long x 10 mm ID with coarse fritted disc
at bottom and Teflon stopcock.
5.2 Reagents
5.2.1 Preservatives
5.2.1.1 Sodium hyroxide-(ACS) ION in distilled water.
5.2.1.2 Sulfuric acid-(ACS) Mix equal volumes of cone. H^SO^ with
distilled water.
5.2.1.3 Sodium thiosulfate-(ACS) Granular.
5.2.2 Methylene chloride, Pentane, Cyclohexane, High Purity Water-HPLC
quality, distilled in glass.
5.2.3 Sodium sulfate-(ACS) Granular, anhydrous (purified by heating at
400°C for 4 hrs in a shallow tray).
5.2.4 Stock standards-Prepare stock standard solutions at a concentra-
tion of 1.00 [Jg/(JL by dissolving 0.100 grams of assayed reference material in
pesticide quality isooctane or other appropriate solvent and diluting to
volume in a 100 ml ground glass stoppered volumetric flask. The stock
solution is transferred to ground glass stoppered reagent bottles, stored in
a refrigerator, and checked frequently for signs of degradation or
evaporation, expecially just prior to preparing working standards from them.
5.2.5 Acetonitrile-Spectral quality.
5.2.6 Silica gel-100/120 mesh desiccant (Davison Chemical grade 923 or
equivalent). Before use, activate for at least 16 hours at 130°C in a foil
covered glass container.
6.0 Procedure
6.1 Collection of Sample
Grab samples must be collected in glass containers. Conventional
sampling practices should be followed, except that the bottle must not be
56
-------
prewashed with sample before collection. Composite samples should be collec-
ted in refrigerated glass containers in accordance with the requirements of
the program. Automatic sampling equipment must be free of Tygon and other
potential sources of contamination.
The samples must be iced or refrigerated from the time of collection
until extraction. Chemical preservatives should not be used in the field
unless more than 24 hours will elapse before delivery to the laboratory. If
the samples will not be extracted within 48 hours of collection, adjust the
sample to a pH range of 6.0-8.0 with sodium hydroxide or sulfuric acid and
add 35 mg sodium thiosulfate per part per million of free chlorine per liter.
All samples must be extracted within 7 days and completely analyzed
within 30 days of collection.
6.2 Sample Extraction
Mark the water meniscus on the side of the sample bottle for later
determination of sample volume. Pour the entire sample into a two-liter
separatory funnel. Check the pH of the sample with wide-range pH paper and
adjust to within the range of 5-9 with sodium hydroxide or sulfuric acid.
Add 60 mL methylene chloride to the sample bottle, seal, and shake 30
seconds to rinse the inner walls. Transfer the solvent into the separatory
funnel, and extract the sample by shaking the funnel for two minutes with
periodic venting to release vapor pressure. Allow the organic layer to
separate from the water phase for a minimum of ten minutes. If the emulsion
interface between layers is more than one-third the size of the solvent
layer, the analyst must employ mechanical techniques to complete the phase
separation. The optimum technique depends upon the sample, but may include
stirring, filtration of the emulsion through glass wool, or centrifugation.
Collect the methylene chloride extract in a 250-mL Erlenmeyer flask.
Add a second 60-mL volume of methylene chloride to the sample bottle and
complete the extraction procedure a second time, combining the extracts in
the Erlenmeyer flask.
Perform a third extraction in the same manner. Pour the combined
extract through a drying column containing 3-4 inches of anhydrous sodium
sulfate, and collect it in a 500-mL Kuderna-Danish (K-D) flask equipped with
a 10 mL concentrator tube. Rinse the Erlenmeyer flask and column with 20-30
mL methylene chloride to complete the quantitative transfer.
57
-------
Add 1-2 clean boiling chips to the flask and attach a three-ball Snyder
column. Prewet the Snyder column by adding about 1 mL methylene chloride to
the top. Place the K-D apparatus on a hot water bath (60-65°C) so that the
concentrator tube is partially immersed in the hot water, and the entire
lower rounded surface of the flask is bathed in vapor. Adjust the vertical
position of the apparatus and the water temperature as required to complete
the concentration in 15-20 minutes. At the proper rate of distillation the
balls of the column will actively chatter but the chambers will not flood.
When the apparatus volume of liquid reaches 1 mL, remove the K-D apparatus
and allow it to drain for at least 10 minutes while cooling. Remove the
Snyder column and rinse the flask and its lower joint into the concentrator
tube with 1-2 mL of methylene chloride. A 5 mL syringe is recommended for
this operation. Stopper the concentrator tube and store refrigerated if
further processing will not be performed immediately.
Determine the original sample volume by refilling the sample bottle to
the mark and transferring the liquid to a 1000 mL graduated cylinder.
Record the sample volume to the nearest 5 mL.
If the sample requires cleanup before chromatographic analysis, proceed
to Section 6.3. If the sample does not require cleanup, or if the need for
cleanup is unknown, analyze an aliquot of the extract according to Section
6.5.
6.3 Cleanup and Separation
Before the silica gel cleanup technique can be utilized, the extract
solvent must be exchanged to cyclohexane. Add a 1-10 mL aliquot of sample
extract (in methylene chloride) and a boiling chip to a clean K-D
concentrator tube. Add 4 mL cyclohexane and attach a micro-Snyder column.
Prewet the micro-Snyder column by adding 0.5 mL methylene chloride to the
top. Place the micro K-D apparatus on a boiling (100°C) water bath so that
the concentrator tube is partially immersed in the hot water. Adjust the
vertical position of the apparatus and the water temperature as required to
complete concentration in 5-10 minutes. At the proper rate of distillation
the balls of the column will actively chatter but the chambers will not
flood. When the apparent volume of the liquid reaches 0.5 mL, remove K-D
apparatus and allow it to drain for at least 10 minutes while cooling.
58
-------
Remove the micro-Snyder column and rinse its lower joint into the
concentrator tube with a minimum of cyclohexane. Adjust the extract volume
to about 2 ml.
6.3.2 Silica Gel Column Cleanup for PAHs
Prepare a slurry of 10 g activated silica gel in methylene chloride and
place this in a 10 mm ID chromatography column. Gently tap the column to
settle the silica gel and elute the methylene chloride. Add 1-2 cm of
anhydrous sodium sulfate to the top of the silica gel.
Preelute the column with 40 ml pentane. Discard the eluate and just
prior to exposure of the sodium sulfate layer to the air, transfer the 2 mL
cyclohexane sample extract onto the column, using an additional 2 mL of
cyclohexane to complete the transfer. Just prior to exposure of the sodium
sulfate layer to the air, add 25 mL pentane and continue elution of the
column. Discard the pentane eluate.
Elute the column with 25 mL of 40% methylene chloride/60% pentane and
collect the eluate in a 500 mL K-D flask equipped with a 10 mL concentrator
tube. Elution of the column should be at a rate of about 2 mL/min.
Concentrate the collected fraction to less than 10 mL by K-D techniques as in
6.2, using pentane to rinse the walls of the glassware. Proceed with HPLC or
gas chromatographic analysis.
6.4 Calibration
Prepare calibration standards that contain the compounds of interest,
either singly or mixed together. The standards should be prepared at
concentrations covering two or more orders of magnitude that will completely
bracket the working range of the chromatographic system. If the sensitivity
of the detection system can be calculated from Table 1 as 100 pg/L in the
final extract, for example, prepare standards at 10 pg/L, 50 |Jg/L, 100 Mg/L,
500 |Jg/L, etc. so that injections of 1-5 (JL of each calibration standard will
define the linearity of the detector in the working range.
Assemble the necessary HPLC or gas chromatographic apparatus and
establish operating parameters equivalent to those indicated in Table 1 or
2. By injecting calibration standards, establish the sensitivity limit of
the detectors and the linear range of the analytical systems for each
compound.
59
-------
Before using any cleanup procedure, the analyst must process a series of
calibration standards through the procedure to validate elution patterns and
the absence of interferences from the reagents.
6.5 Analysis
6.5.1 HPLC
To the extract in the concentrator tube, add A mL acetonitrile and a new
boiling chip, then attach a micro-Snyder column. Increase the temperature of
the hot water bath to 95-100°C. Concentrate the solvent as above. After
cooling, remove the micro-Snyder column and rinse its lower joint into the
concentrator tube with about 0.2 mL acetonitrile. Adjust the extract volume
to 1.0 mL.
Table 1 summarizes the recommended HPLC column materials and operating
conditions for the instrument. Included in this table are estimated
retention times and sensitivities that should be achieved by this method. An
example of the separation achieved by this column is shown in Figure 1.
Calibrate the system daily with a minimum of three injection of calibration
standards.
Inject 2-5 (JL of the sample extract with a high pressure syringe or
sample injection loop. Record the volume injected to the nearest 0.05 pL,
and the resulting peak size, in area units. If the peak area exceeds the
linear range of the system, dilute the extract and reanalyze. If the peak
area measurement is prevented by the presence of interference, further
cleanup is required. The UV detector is recommended for the determination of
naphthalene and acenaphthylene and the fluorescence detector is recommended
for the remaining PAHs.
6.5.2 Gas Chromatography
The gas chromatographic procedure will not resolve certain isomeric
pairs as indicated in Table 2. The liquid chromatographic procedure (6.5.1)
must be used for these materials.
To achieve maximum sensitivity with this method, the extract must be
concentrated to 1.0 mL. Add a clean boiling chip to the methylene chloride
extract in the concentrator tube. Attach a two-ball micro-Snyder column.
Preset the micro-Snyder column by adding about 0.5 ml of methylene chloride
to the top. Place this micro K-D apparatus on a hot water bath (60-65°C) so
60
-------
COLUMN: HC-ODS SIL-X
MOBILE PHASE: 40% TO 100% ACETONITR1LE IN WATER
DETECTOR: FLUORESCENCE
40
RETENTION TIME-MINUTES
Figure 1. Liquid chromatogram of polynuclear aromatics.
61
-------
Table 2. GAS CHROMATOGRAPHY OF PAHs
Compound Retention time (min)
Naphthalene 4.5
Acenaphthylene 10.4
Acenaphthene 10.8
Fluorene 12.6
Phenanthrene 15.9
Anthracene 15.9
Fluoranthene 19.6
Pyrene 20.6
Benzo(a)anthracene 20.6
Chrysene 24.7
Benzo(b)fluoranthene 28.0
Benzo(k)fluoranthene 28.0
Benzo(a)pyrene 29.4
Dibenzo(a,h)anthracene 36.2
Indeno(l,2,3-cd)pyrene 36.2
Benzo(ghi)perylene 38.6
GC conditions: Chromosorb W-AW-DMDCS 100/120 mesh coated with 3% 0V-
17, packed in a 6* x 2 mm ID glass column, with nitrogen carrier gas
at 40 mL/min flow rate. Column temperature was held at 100°C for 4
minutes, then programmed at 8°/minute to a final hold at 280°C.
62
-------
that the concentrator tube is partially immersed in the hot water. Adjust
the vertical position of the apparatus and water temperature as required to
complete the concentration in 5 to 10 minutes. At the proper rate of
distillation the balls will actively chatter but the chambers will not flood.
When the apparent volume of liquid reaches 0.5 mL, remove the K-D apparatus
and allow it to drain for at least 10 minutes while cooling. Remove the
micro-Snyder column and rinse its lower joint into the concentrator tube with
a small volume of methylene chloride. Adjust the final volume to 1.0 mL and
stopper the concentrator tube.
Table 2 describes the recommended gas chromatographic column material
and operating conditions for the instrument. Included in this table are
estimated retention times that should be achieved by this method. Calibrate
the gas chromatographic system daily with a minimum of three injections of
calibration standards.
Inject 2-5 |JL of the sample extract using the solvent flush technique.
Smaller. (1.0 |JL) volumes can be injected if automatic devices are employed.
Record the volume injected to the nearest 0.05 (JL, and the resulting peak
size, in area units. If the peak area exceeds the linear range of the
system, dilute the extract and reanalyze. If the peak area measurement is
prevented by the presence of interferences, further cleanup is required.
6.6 Calculations
Determine the concentration of individual compounds according to the
formula:
(A)(B)(Vt)
Concentration, (Jg/L =
(v.)(vs)
where:
A = calibration factor for chromatographic system, in nanograms
material per area unit
B = peak size in injection of sample extract, in area units
V. = volume of extract injected (|JL)
V* = volume of total extract (|JL)
V = volume of water extracted (ml)
s
63
-------
Report results in micrograms per liter without correction for recovery
data. When duplicate and spiked samples are analyzed, all data obtained
should be reported.
7.0 Quality Assurance Program
7.1 Quality Control
Before processing any samples, the analyst should demonstrate through
the analysis of a distilled water method blank, that all glassware and
reagents are interference-free. Each time a set of samples is extracted or
there is a change in reagents, a method blank should be processed as a
safeguard against laboratory contamination.
Standard quality assurance practices should be used with this method.
Field replicates should be collected to validate the precision of the
sampling technique. Laboratory replicates should be analyzed to validate the
precision of the sampling technique. Laboratory replicates should be
analyzed to validate the precision of the analysis. Fortified samples
should be analyzed to validate the accuracy of the analysis. Where doubt
exists over the identification of a peak on the chromatogram, confirmatory
techniques such as fraction collection and GC-mass spectroscopy should be
used.
8.0 References
8.1 "Development and Application of Test Procedures for Specific Organic
Toxic Substances in Wastewaters, Category 9-PAHs", Report for EPA
Contract 68-03-2624 (in preparation).
-------
ANALYTICAL PROTOCOL: VINYL CHLORIDE FROM PERSONAL
MONITORING DEVICE
1.0 Principal of the Method
The method for measuring the exposure of personnel to vinyl chloride
utilizes the permeation technique for sampling. The vinyl chloride that
permeates the membrane is trapped on activated charcoal which is removed
for subsequent determination by gas chromatography. The monitor is
about the size of a standard film badge, weighs less than 35 g, and
requires no source of power. The method is insensitive to temperature
and humidity. It is ideally suited to personal monitoring programs
because the analytical data represent a time-weighted-average exposure
and require no further data reduction step.
2.0 Range and Detection Limit
The method has been validated for a concentration range (8 hr air
sample) of 5 ppb to 50 ppm. The lower value represents the dilution
limit when analysis is accomplished via thermal desorption techniques.
When interferences preclude the use of this approach, solvent desorption
is used. The limit of detection is then 20 ppb.
3.0 Interferences
Studies have shown that sulfur dioxide, nitrogen dioxide, ozone and
chlorine do not interfere with the analysis of vinyl chloride using the
personal monitor device. The only known interferent is ethylene chloride
which is converted to vinyl chloride by heat treatment of the trapping
medium during assay. The solvent desorption technique, described below,
precludes this artifactual formation of vinyl chloride. Ambient humidity
does not affect the operation and assay of the permeation device.
4.0 Precision and Accuracy
The level of precision at the detection limit (5 ppb) is within +
50%. At concentrations greater than 100 ppb the precision is 5% for the
thermal desorption mode of analysis; the solvent desorption procedure
yields a precision of V>% at 500 ppb. At this concentration (500 ppb)
the accuracy for replicate determinations was + 2.5%.
65
-------
5.0 Apparatus and Reagents
5.1 Sampling Device
The sampling device is a badge having dimensions of 41 mm by 48 mm
and a thickness of 7 mm (Figure 1). An internal cavity is covered by a
permeable membrane through which vinyl chloride passes at a rate propor-
tional to the external concentration. The vinyl chloride that permeates
through the membrane is adsorbed on activated charcoal which is later
removed from the device after completion of the exposure, and the amount
adsorbed is then determined by gas chromatography.
5.2 Adsorber
The adsorber used is Darco Activated Charcoal, 20-40 mesh, supplied
by Matheson, Coleman and Bell. It is pretreated to remove moisture and
any organics present by heating to 350 C under a flow of inert gas.
5.3 Reagents
All gases used in the preparation of permeation tubes are chemically
pure grade. All liquids used are reagent grade. Purity of each organic
reagent is checked by gas chromatography.
5.4 Instrumentation
5.4.1 Calibration Device
The permeation device used in this work is the same as that described
previously (1), except the absorbing solution is replaced by a known
quantity of activated charcoal which serves as an adsorber for the vinyl
chloride that permeates the membrane.
The membrane material used for the device is a disk of single-
backed dimethylsilicone rubber (General Electric Co., One River Road,
Schnectady, N. Y. 12305). The standard permeation tube described by
O'Keeffe and Ortman (2) and Scaringelli et al. (3) is used. These tubes
are unsuitable for studies at very low levels; therefore, for concen-
trations below 0.05 ppm it is necessary to use a low-level permeation
tube which is similar to the reservoir device of Saltman et al (4) .
A reservoir of 3 ml is used with an active tube length of as little
as 1 mm. This device, which has been used successfully for other gases,
has a very short active area and is capable of accurately dispensing
levels as low as 20 ng/min of vinyl chloride. Permeation tubes containing
vinyl chloride
66
-------
have a tendency to form a solid material, presumably a polymeric species,
despite precautions taken to entrap inhibitor within the tube. This is not a
serious problem as long as there is liquid visible in the tube.
5.4.2 Analytical Instrumentation
A Varian Model 1200 gas chromatograph along with a Honeywell-Brown
recorder equipped with a Disc integrator, or equivalent instrumentation, is
used in all investigations. The gas chromatograph is equipped with a flame
ionization detector and is modified to permit the use of an external sampling
system (Figure 2) consisting of an interchangeable sampling loop in a thermally
shielded compartment which can be rapidly heated to 300°C. The sample loop
consists of a 15 cm x 6.35 mm o.d. stainless steel tube. The tube is placed
inside a brass tee and is heated by the exhaust from a heat gun, (Master
Appliance Corp., Model No. HG-501) being directed through the side of the
tee. A four-way valve provides a bypass for the helium carrier gas to permit
changing of the sample loop without interrupting the determination.
6.0 Procedure
6. 1 Calibration
6.1.1 Permeation Device
The membrane material used in the device is not entirely of uniform
thickness; hence, each permeation device under study is individually calibrated
using the apparatus illustrated in Figure 3. Calibration is accomplished by
cleaning and dehumidifying air from the laboratory by pumping it though
columns of activated charcoal and silica gel. Finally, the clean, dry air is
passed at a predetermined flow rate over a permeation tube, of the design of
O'Keeffe and Ortman (2). The permeation tube emits vinyl chloride at a known
constant rate, thereby providing a primary standard for the calibration
procedure. The standard concentration of vinyl chloride in air is passed
through the exposure chamber where the permeation devices are exposed for
calibration.
The determination of the amount of vinyl chloride adsorbed on the
charcoal from the device allows the calculation of a permeation constant for
each device. This constant is calculated from the following equation:
w
where k = constant (about 0.5); C = concentration of vinyl chloride, ppm; t=
67
-------
time of exposure, and w = amount of vinyl chloride adsorbed, pg.
6.1.2 Gas Chromatograph
The response factor of the gas chromatograph is determined by preparing
a standard containing a known amount of vinyl chloride adsorbed on 1 g of
activated charcoal. This is accomplished by collecting the vinyl chloride
from a stream of known concentration by passing the standard through a tube
containing the charcoal sample. A backup tube is used to assure that no
breakthrough of vinyl chloride occurs. These are then analyzed along with
the unknown samples. This provides an accurate measure of response of the
instrument to the levels of vinyl chloride being determined. It also compen-
sates for the fact than vinyl chloride may not be completely desorbed from
the charcoal. No breakthrough of vinyl chloride is observed for any standard
studied over the range of 0.2 to 200 ng.
6.2 Sample Collection
The method utlizes a permeable membrane to collect the air sample at a
rate proprotional to concentration. The device is worn by individuals as a
badge in a manner that permits free access of ambient air to the device.
6.3 Storage of Samples
The monitor may be stored in an air-tight container for at least 8 h
without loss of vinyl chloride. However, for longer storage, the charcoal
should be transferred to a sealed vial. Samples stored in vials for periods
up to 6 months have shown no significant losses. It is not advisable to
store samples collected in a complex environment for this length of time. It
should also be noted that with monitors that are exposed to uncontaminated
air for 8 h, no detectable losses are observed indicating that desorption
through the membrane is very slow.
6.4 Analytical Method
The chromatographic determination of vinyl chloride used for these
studies primarily employs the thermal desorption process. The charcoal
sample is placed in a sample loop and the loop mounted in the desorption
oven. Standards are first transferred to a vial, then to the sample loop.
The column is cooled to 25 C and the sample loop purged with the helium
carrier gas. Then, with the carrier gas still flowing through the sample
loop, the loop is heated by the hot air from the heat gun. The conditions
necessary for optimum desorption were found to be 5 min desorption at 300°C.
68
-------
After the completion of the thermal desorption process, the vinyl
chloride from the sample is trapped on the head of the analytical column.
This column is a 2-m by 2.4 mm i.d. stainless steel tube packed with Chromosorb
102 (Johns-Manvilie). The carrier gas flow is shunted around the sample loop
by the bypass valve, and the column oven heated to 90°C. Under these condi-
tions, vinyl chloride exhibits a rentention time of ~ 4 min. After the
emergence of the vinyl chloride, the column is heated to 200°C to remove
other substances and ready the column for the next determination. Carbon
disulfide extraction and analysis procedures are also applicable for use with
the permeation sampling technique. The procedure involves extraction of the
charcoal with 5 ml of CS in a vial sealed with a Teflon-backed rubber septum.
The mixture is allowed to develop for 30 min with periodic agitation. Both
the sample and solvent are chilled to dry ice temperatures before extraction,
and sample vials are sealed with Teflon-backed rubber septa.
The permeation constant, which with care can be determined to a precision
of 3%, is used for the calculation of the average concentration of vinyl
chloride in an unknown atmosphere by using the equation:
r _ wk
L ~ t
where C = time-weighted-average vinyl chloride concentration, ppm.
7.0 Quality Control Program
The exercise of control of the quality of the data generated during the
sampling and analysis of vinyl chloride using the permeation device is achieved
through the use of sampling blanks, and lab controls. The analysis of blanks
that have experienced field environments, without exposure, determines back-
ground levels. Control samples allow for updating gas chromatographic response
factors. Both types of blanks will signal interferences, either from the
atmosphere or from the instrumentation (septa, fittings, etc).
The permeation rate for a given personal monitor should be checked
periodically using the calibration apparatus described above (6.1.1). The
analysis of each sample extract should be conducted in duplicate. For all
sample handling, extraction, and storage, only cleaned materials (glass,
Teflon, stainless steel) should be used.
69
-------
8.0 References
8-1 Reiszner, K. D. West, P- W. , Environ. Sci. Technol., T_t 526 (1973).
8-2 O'Keeffe, A. E., Ortman, G. C., Anal. Chem., 38, 760 (1966).
8-3 Scaringelli, F. P., Frey, S. A., Saltzman, B. E., Anal. Chem., 42, 871
(1970).
8-4 Saltzman, B. E., Burgh, W. R., Romaswamy, G. K., Environ. Sci. Technol.,
5, 1121 (1971).
70
-------
MCMMANE
T
41*
i
41mm
TEHON
-41mm-
tlATE
Figure 1. Personal monitoring device
Slolnlttl Slitl
!^ 00 f 6" long
lomlcst Sltvl
Mtmo»oblc
Somplt Moldti
Figure 2. Apparatus for thermal desorption of vinyl chloride
EXPOSURE CHAMBER
i PERMEATION
ITUBE HOLDER
J CONSTANT
TEMPERATURE BATH
EXHAUST
Figure 3. Calibration apparatus
Adapted from "Personal Vinyl Chloride Monitoring Device with Permeation
Technique for Sampling", L. II. Nelms, K. D. Reiszner and P. W. West Anal
Chem., 49, 994 (1977).
71
-------
ANALYTICAL PROTOCOL: ARSENIC, CADMIUM AND LEAD IN AMBIENT AIR PARTICULATE
The analyses for the target metals will be accomplished by proton
induced X-ray emission (PIXE) analysis under the direction of Dr. J. W.
Nelson at Florida State University. The technique has been validated for a
large number of metals from air aerosol from a number of different sources.
Sensitivites are sufficient for the analysis of some 20 metals from aerosol
obtained from 120 L of air. Complete specifications and methodology are
provided in the following references:
1. Hudson, G. M., H. C. Kaufmann, J. W. Nelson and M. A. Bonacci, Nuclear
Instruments and Methods, 168, 259 (1980).
2. Johansson, T. B., R. E. VanGrieken, J. W. Nelson and J. W. Winchester,
Anal. Chem., 47, 855 (1975).
3. Jensen, B. and J. W. Nelson, "Nuclear Methods in Environmental Research",
CONF-740701 (Oak Ridge, Tennessee:USERDA Technical Information Center,
1974), p. 366.
72
-------
ANALYTICAL PROTOCOL: ORGANOCHLORINE PESTICIDES AND PCBS IN AIR
1.0 Principle of the Method
Air is sampled using a personal monitoring pump (2-4 L/min), and a
cartridge of polyurethane foam (PUT) for trapping the target compounds. The
PUF is Soxhlet extracted with hexane-diethyl ether (95:5 v/v). Chlorinated
pesticides and PCBs are measured by EC-GC after column chromatographic
clean-up on alumina. PCBs are separated from technical chlordane and other
pesticides by column chromatography on silicic acid deactivatived with 3%
distilled water.
2.0 Range and Detection Limit
The limits of detection for selected organochlorine pesticides and PCBs
as extracted from PUF plugs according to this protocol are given in Table 1.
Levels vary from several hundred picograms up to ~10 ng for the pesticides,
and from *°25-35 ng for various Arochor mixtures. Also shown are the limits
of measurement which consist of the low detectable quantities as determined
by injection of standards onto the GC system. The range of detection is
unspecified in the EPA protocol.
3.0 Interferences
Unspecified in EPA protocol.
4.0 Precision and Accuracy
The precision and accuracy of the recovery and collection efficiency
are shown in Table 2 for selected pesticides and 3 Arochlors. Recoveries
were determined by spiking PUF plugs via syringe with solutions of known
concentrations of target compounds. The collection efficiencies were
measured by vaporizing known quantities of analytes into the PUF, and
are corrected by the recovery values.
5.0 Apparatus and Reagents
5.1 Sampling
5.1.1 Pumps
DuPont Constant Flow Sampling Pump, Model P4000A (includes charger),
Catalog No. 66-241. DuPont, Applied Technology Division, Wilmington, DE
TM
19898 or MSA Monitatire Sampler, Model S, Catalog No. 458475 and charger
No. 456059. Mine Safety Applicances Company, 600 Penn Center Boulevard,
73
-------
Table 1. LIMITS OF DETECTION (EC) AND AMOUNTS
COMMONLY FOUND ON FOAM
Compound
Y-BHC
aldrin
p_,E'-DDE
£,E'-DDT
mirex
tech chlordane
Aroclor 1242
Aroclor 1254
Aroclor 1260
Measurement limit
P8
3
5
10
30
10
55
90
150
520
Amount on control foam
ng
0
0
1
10
2
5
30
25
34
.3
.6
.5
This is defined as the amount necessary to give a peak 10% of full
scale deflection at attenuation X2. For most compounds this is 10-
15 times noise (or 5-7 times the detection limit).
74
-------
Table 2. EFFICIENCY OF FOAM AS A COLLECTOR OF PESTICIDES
Collection
efficiency
(collected for
recovery
aldrin
£,2' -DDE
£,E'-DDT
mirex
tech chlordane
Aroclor 1242
Aroclor 1254
Aroclor 1260
X
59
102
98
86
-
96
95
109
% RSD
12
11
21
22
-
15
7
5
n
12
12
12
7
-
6
6
11
Recovery
efficiency
X
91
96
101
89
103
100
99
98
% RSD
2
2
2
1
7
3
5
7
n
4
4
4
4
5
5
5
5
Retention
efficiency
X
67
100
97
94
-
97
101
% RSD
13
6
8
5
-
14
11
n
4
4
4
4
-
7
7
75
-------
Pittsburgh, PA 15235. Both of these small, battery operated pumps are
capable of pumping air through an 18 mm diameter x 50 mm cylindrical PUT
plus at 2.5 to 4 liters/minute for at least 8 hours with a fully charged
battery pack. The DuPont pump has the advantage that it will automatically
adjust its pumping rate to compensate for changes in flow resistance (e.g.,
due to accumulation of particulate matter at the intake of the collection
module). It also operates more quietly than the MSA and can be programmed
to stop sampling after a prescribed period.
5.1.2 Collection Device
A glass tube, 10 cm x 1.5 cm I.D. is used which contains ^15 ml of PUF
(8 x 1.5 cm). The foam plug is cut slightly oversized for a compression
fit. An all Teflon filter housing allows for collection of particulate
matter at the intake end of the cartridge. The cartridge is attached via a
compression fitting and short length of flexible tubing to the pumping
device.
5.2 Analysis
5.2.1 Gas chromatograph, Tracer 222 or 560, equipped with linearized
/-o
Ni FPD, and electrolytic conductivity detectors, or equivalent.
5.2.2 Glassware. Centrifuge tubes, 15 mL, graduated; separatory
funnel, 500 ml; Buchner filtration device.
5.2.3 Extractors, Soxhlet, 1000, 500 and 250 mL.
5.2.4 Clean-up microcolumn, 10 cm x 5 mm i.d. disposable pipet or
Chromaflex column, size 22, 20 cm x 7 mm, Kontes, Vineland, NJ, K 420100-
0022.
5.2.5 Chromatoflo chromatography column, 25 cm x 9 mm i.d., Pierce
#29020, equipped with a Teflon mesh support membrane, Pierce #29268, lower
end plate, adapter, and 500 mL solvent reservoir (Ace #5824-10).
5.2.6 Rotary vacuum evaporator, e.g., Buchi, with 250, 500, and 1000
mL round bottom flasks.
5.2.7 Solvents, glass distilled, pesticide quality, or equivalent.
Diethyl ether, analytical reagent grade, Mallinckrodt #0850, containing 2%
ethanol.
5.2.8 Pesticide standards and commercial PCB mixtures, 98-100% pure,
obtainable from the Pesticide Repository, U. S. EPA, EDT, HERL, Research
76
-------
Triangle Park, NC (MD-69). Individual PCBs, obtainable from RFR Corp.,
Hope, RI.
5-2.9 Column chromatography. stationary phase.
5.2.9.1 Alumina, basic, 60 mesh, Alfa Products. Adjust to Brockmann
activity IV by adding 6% (w/v) distilled water to the adsorbent in a flask,
stoppering, and shaking well; allow to equilibrate for at least 15 hours
before use. Discard after two weeks.
5.2.9.2 Silicic acid, Mallinckrodt AR, 100 mesh; heat at 130°C for at
least 7 hours and cool to room temperature in a desiccator; to deactivate,
weigh into a bottle, add 3% (w/w) distilled water, seal tightly, shake well,
and place in a desiccator for at least 15 hours. Discard any adsorbent not
used within one week.
6.0 Procedure
6.1 Extraction of PUT Plug
Place the foam plug in a Soxhlet extractor, handling with forceps
rather than hands.
NOTE: After sampling, the foam plugs should have been wrapped in
aluminum foil until analysis. Use plugs carried to the field
along with those employed for sampling as controls.
Extract with an appropriate volume of n-hexane-acetone-diethyl ether
(47:47:6 v/v) for 8-12 hours at 8 cycles per hour with the smaller Soxhlet.
Remove the boiling flask to a rotary evaporator and reduce the solvent
volume to approximately 5 mL. Transfer the concentrate to a 15 mL graduated
centrifuge tube with rinsing.
6.2 Analysis
6.2.1 Sample Preparation
Reduce the volume in the 15 mL tube to below 1 mL by careful evaporation
under a gentle stream of nitrogen at room temperature. Carry out alumina
cleanup as follows:
Place a small plug of pre-extracted glass wool in the Chromaflex column
and wash with 10 mL of hexane. Pack the column with 10 cm of activity grade
IV alumina. Transfer the sample from the centrifuge tube to the top of the
column; rinse the tube three times with 1 ml portions of n-hexane, adding
each rinse to the column. Elute the column at a rate of ca. 0.5 mL per
77
-------
minute with 10 ml of n-hexane, collecting the eluate in a 15 ml centrifuge
tube. Adjust the final volume of the eluate to 10 ml for gas chromatographic
analysis.
When necessary, separate PCBs from technical chlordane by silicic acid
chromatography as follows:
Place 3 grams of deactived silicic acid in a Chromatoflo column assembly.
Wash the column with hexane. Place the sample, concentrated to less than 1
ml, on the column and add 130 mL of hexane to the reservoir. Apply nitrogen
pressure to the column to increase the flow rate to ca. 1 mL/minute. Collect
the eluate in three fractions: Fraction I (0-30 ml) contains all the HCB
and Aroclor 1254 and most of the Aroclor 1242; Fraction II (31-50 mL) contains
the remainder of Aroclor 1242, £,£'-DDE, some of the o,£'-DDT and toxaphene,
and the early eluting peaks of technical chlordane; Fraction III (51-130
mL) contains the remainder of the technical chlordane, including all of the
cis- and trans-chlordane, £,£'-DDT, and 30% of the toxaphene. Elute dieldrin,
£,£*-DDD, 6% of the toxaphene, and the remaining pesticides with 15 mL of
dichloromethane. Adjust the fraction volumes and analyze by GC. Blank
values of unused plugs determined by extraction and alumina clean-up of the
3
extract should be equivalent to <1 pg/m .
6.2.2 Gas Chromatography
Determine target pesticides on a 183 cm x 4 mm i.d. glass column packed
with 1.5% OV-17/1.95 % OV-210 and/or 4% SE-30/6% OV-210 on 80-100 mesh Gas
Chrom Q; column, 200°C; injection port, 215°C; nitrogen carrier gas, 60-85
mL/minute; electron capture detector.
Determine PCBs by EC-GC under the above conditions on a similar column
packed with 3% OV-1 on Gas Chrom Q at 180°C. Alternatively, use columns
containing 3% OV-225 on Supelcoport, 80-100 mesh or 4% SE-30/6% OV-210 on
Gas Chrom Q, 100-200 mesh at 200°C.
Quantitate peaks in the usual way, i.e., by measuring peak heights to
the nearest mm when the base width is <1 cm or via peak areas by integration
or triangulation for broader peaks. Confirm results as required by combined
GC/MS or some other appropriate procedure (EPA Pesticide Analytical Quality
Control Manual, Chapter 8).
78
-------
Commercial PCB mixtures are quantitated by comparisons of the total
heights or areas of GC peaks with the corresponding peaks in the standard
used. The absolute retention times on the 3% OV-1 column for the peaks used
were as follows:
Aroclor 1242 - 2.39, 2.65, 3.11, 3.33, 3.94, 4.37, 4.67, 5.59, and
6.25 minutes
Aroclor 1254 - 3.81, 4.28, 4.61, 5.55, 6.68, 7.76, 8.23, 9.83, 11.47,
and 13.67 minutes.
With the SE-30/OV-210 column, the total peak heights of the peaks shown
in Figure 1 can be used for quantitation. (Make Aroclor standards by dissol-
ving the Aroclor in isooctane, and prepare dilutions in hexane. Store stock
solutions in brown bottles at -10°C. Remake working standards periodically
from these and store in a refrigerator when not in use).
7.0 Quality Assurance Program
7.1 Quality Control
In-house control of all instruments, solvents, and procedures will
follow the detailed guidelines specified in "Manual for Analytical Quality
Control for Pesticides and Related Compounds in Human and Environmental
Samples", EPA Report No. EPA-600/1-79-008.
Blank PUF plugs and plugs loaded with known amounts of standards are
prepared for each sampling trip. A portion of the samples are designated
"lab blanks/controls" and remain in the laboratory; a futher portion are
designated "field blanks/controls", and are carried to the field in the same
containers as the sample plugs. This procedure not only provides a check on
possible contamination during transport and storage, but also sllows calcula-
tion of overall recoveries during the storage and analysis phases.
7.2 Quality Assurance
A specified number of duplicate samples (Table 6-7, Part I of this Work
Plan) will be provided to a separate laboratory (HERL/RTP) for analysis.
The results will be compared with those obtained at the primary lab (RTI).
79
-------
UJ
X
UJ
a.
UJ
UJ
tc
T I I
A. AROCLOR 1242. 0.50 ng
B. AROCLOR 1254. 0.50 ng
C. AROCLOR 1260. 0.50 ng
I
I
10 15
TIME, min
20
JO
Figure 1. Chromatograms showing peaks used in quantifying PCS,
SE-30/OV-210.
80
-------
ANALYTICAL PROTOCOL: VOLATILE ORGANOCHLORIDES AND BENZENE BY THE
PURGE AND TRAP METHOD
1.0 Principle of the Method
This method is applicable in the determination of the halogenated
compounds listed in Table 1 contained in carbon filtered drinking water
or raw source water. It is also applicable for the determination of
benzene.
Organohalides are extracted by an inert gas which is bubbled through
the aqueous sample. The organohalides, noted in Table 1 along with
other organic constituents which exhibit low water solubility and boil
less than 200°C, are efficiently transferred from the aqueous phase to
the gaseous phase. These compounds are swept from the purging device
and are trapped in a short column containing a carefully selected sorbent
combination. After a predetermined period of time, the trapped components
are thermally desorbed and backflushed onto the head of a gas chromato-
graphic column and separated under programmed conditions. The effluent
is split and measurement is accomplished with both a halogen specific
detector, which eliminates interference problems commonly encountered
with universal or semispecific detectors, and a flame ionization or
photoionization detector. The extraction/concentration technique enhances
the quantities of organohalides injected into the gas chromatograph by a
factor of 1000 over direct injection gas chromatography.
This method is recommended for use only by analysts experienced in
the measurement of purgeable organics at the |Jg/L level or by experienced
technicians under the close supervision of a qualified analyst.
2.0 Range and Detection Limit
The actual detection limits are highly dependent upon the ability
of the analyst to properly maintain the entire analytical system. Using
carefully optimized equipment the method has been proven to be useful
for the detection and measurement of multicomponent mixtures spiked into
carbon-filtered finished water and raw source water at concentrations
between 0.2 and 0.4 |Jg/I" The method as described is capable of accura-
tely measuring those compounds mentioned in Table 1 over a concentration
range of 0.10 to 5.0 pg/L. Additionally, it is possible to measure
81
-------
Table 1. ORGANOHALIDES TESTED USING PURGE AND TRAP METHOD
Retention Time
Compound (SEC)
Column I Column II
chloromethane
bromomethane
dichlorodlfloromethane
vinyl chloride
chloroethane
d1 chloromethane
fluorotrlchloromethane
allylchlorlde
1,1-dichloroethylene
bromoch loromethane
1,1 -d1 chloroethane
clstrans 1,2-d1chloro-
ethylene
ds-1,2-d1chloro-
ethylene
chloroform
l,2-d1chloroethane
dlbromomethane
1,1,1-trl chloroethane
carbon tetrachlorlde
bromod 1ch loromethane
2,3-d1chloropropene
l,2-d1chloropropane
I,l-d1chloropropene
trans-1,3 dichloropropene
1,1,2-TrlChloroEthylene
1,3-d1chloropropane
chlorodlbromomethane
1, 1, 2 -tr1 chloroethane
ds-1,3-d1ch1oropropene
1,2-dibromoethane
2-chloroethyl ethyl ether
2-ch1oroethylv1nyl ether
bromof orm
1,1, 1 ,2-tetrach loroethane
1,2,3-trlchloropropane
chlorocyclohexane
1,1,2,2-tetrachloroethane
1,1,2,2-tetrachloroethylene
pentachloroethane
1 -ch 1 orocyc 1 ohexene- 1
chlorobenzene
90
130
157
160
200
315
431
475
476
509
558
605
605
641
684
698
756
781
819
891
895
904
913
948
973
989
991
992
1046
1056
1080
1154
1163
1279
1283
1297
1300
1300
1345
1451
317
423
?
317
521
607
?
?
463
760
754
563
726
725
921
895
786
664
877
?
997
7
997
787
?
997
1084.
1078
1131
?
?
1150
1302
?
?
?
898
?
1186
1130
Estimated A lower
Limit of detection
UQ/1
0.009
0.03
0.03
0.01
0.01
0.01
0.01
0.02
0.006
0.02
0.004
0.006
0.006
0.006
0.006
0.03
0.005
0.007
0.006
0.007
0.004
0.01
0.006
0.005
0.01
0.01
0.006
0.008
0.01
0.03
0.06
0.02
0.003
0.007
0.01
0.006
0.007
0.01
0.03
0.03
(continued)
82
-------
Table 1 (cont'd.)
1-chlorohexane 1499 1229 0.05
b1s-2-chloroethy1ether 1500 ? ' Jj«J
bromobenzene 1626 ? °-^
o-chlorotoluene 1927 1320 0-07
b1s-2-ch1oro1sopropy1 n nA
ether 1931 ? °-°J
m-d1ch1orobenzene 2042 1346 o.uj
o-d1ch1orobenzene 2094 1411 O-UJ
p^ichlorobenzene 2127 1340 0.04
A » See 10.3
83
-------
individual organohalides up to 1500 Mg/L. However, the ability to
measure complex mixtures containing co-eluting or partially resolved
organohalides with concentration differences larger than a factor of 10
is hampered.
3.0 Interferences
Impurities contained in the purge gas and organic compounds out-gas-
ing from the plumbing ahead of the trap usually account for the majority
of contamination problems. The presence of such interferences is easily
monitored using the quality control program described herein. Sample
blanks are normally run between each set of samples. When a positive
organohalide reponse is noted in the sample blank, the analyst should
analyze a method blank in order to identify the source of contamination.
Method blanks are run by charging the purging device with organic-free
water and analyzing it in the normal manner.
Whenever organohalides are noted in the method blank, the analyst
should change the purge gas source and regenerate the molecular sieve
purge gas filter. Subtracting blank values from sample results is not
recommended. The use of non-TFE plastic tubing, non-TFE thread sealants,
or flow controllers with rubber components in the purging device should
be avoided since such materials out-gas organic compounds which will be
concentrated in the trap during the purge operation. Such out-gasing
problems are common whenever new equipment is put into service. With
use, minor out-gasing problems generally cure themselves.
Several instances of accidental sample contamination have been
noted and attributed to diffusion of volatile organics through the
septum seal and into the sample during shipment and storage. The sample
blank is used as a monitor for this problem.
For compounds that are not efficiently purged, such as bromoform,
small variations in sample volume, purge time, purge flow rate, purging
device geometry, or purge temperature can effect the analytical result.
Therefore, samples and standards must be analyzed under identical condi-
tions .
Cross contamination can occur whenever high level and low level
samples are sequentially analyzed. To reduce the liklihood of this, the
84
-------
purging device and sample syringe should be rinsed out twice between
samples with organic-free water. Whenever an unusually concentrated
sample is encountered, it is necessary that it be followed by a sample blank
analysis to check for sample cross contamination. For samples containing
large amounts of water soluble materials, suspended solids, high boiling
compounds or high organohalide levels it may be necessary to wash out the
purging device with a soap solution, rinse with distilled water, and then dry
in a 105°C oven between analyses.
Qualitative misidentifications are a problem in gas chromatographic
analysis. Whenever samples whose qualitative nature is unknown are analyzed,
the following precautionary measures should be incorporated into the
analysis.
1. Perform duplicate analyses using the two recommended columns
which provide different retention order and retention times for
many organohalides.
2. Whenever possible use GC/MS techniques which provide unequivocal
qualitative identifications.
4.0 Precision and Accuracy
4.1 Control Samples
Both Ohio River water (ORW) and carbon-filtered tap (CFT) water were
spiked with known amounts of organohalides. The spiked solutions were then
sealed in septum-seal vials both with and without sodium thiosulfate (thio)
and sodium sulfite (sulfite) then stored on the bench top for up to four
weeks. Samples were randomly analyzed on four occasions. When matrix
effects were noted or suspected then that data was not included in the
following single laboratory precision and accuracy statement. Table 2 shows
the accuracy and precision data obtained from this study.
4.2 Blank Samples
Organic-free water was dosed with mixtures of organohalides. The dosed
water was used to fill septum seal vials which were stored under ambient
conditions. The dosed samples were randomly analyzed over a 2-week period of
time. The data listed in Table 3 reflect the errors due to the analytical
procedure and storage.
85
-------
Table 2. SINGLE LAB ACCURACY AND PRECISION FOR PURGE AND TRAP METHOD
HALL 700A ELECTROLYTIC CONDUCTIVITY DETECTOR
oo
chloromethane
bromomethane
vinyl chloride
dl chlorodl fl uoromthane
chloroethane
dl chloromethane
fl uorotrt chloromthane
1 ,1-dlchloroethylene
•llyl chloride
bromochloromethane
1,1-dl chloroethane
cls+trans-1 ,2-dlchloro-
ethylene
cls-1 ,2-dlchloro-
ethylene
1 ,2-dlchloroethane
dlbronome thane
1 ,1 ,1-trl chloroethane
carbon tetrachloMde
bromdlchloroMethane
2.3-dlchloropropene
1 ,2-dlchloropropane
1 ,1-dlchloropropene
trans-1,3 dlchloropro-
pene
cls-1 ,3-d1ch1oropro-
pene ' <
1 ,1 ,2-trlchloroethane
1 ,3-dlchloropropane
Spike
Level
0.40
0.40
0.20
0.40
0.40
0.20
0.40
0.40
0.40
0.40
0.20
0.20
0.40
0.20
0.40
0.40
0.20
0.20
0.20
0.40
0.40
0.40
0.40
0.40
0.40
C.F.T.
yesc
yes*"
no
yes
yes
no
yes
yes
no
no
yes
no
yes
yes
no
no
no
yes
yes
yes'
yes'
yes
yes
O.R.W.
J?
yes
yes
yes
yes
yes
yes-
yes'
yes
yes
yes
yes
yes
yes
yes
yes
yes
yesC
yes
yes
yes'
yes'
yes
yes
Preserved
Sample
Thlo.
Sulflde
yes'
no
yes
yes
no
yes
yes
no
no
yes
no
yes
yes
no
no
no .
yes.
yes*
yes'
yes'
yes
yes
Mean t
Recovery
93
85
110
103
93
85
90
88
85
90
95
95
88
110
100
93
90
100
95
95
88
88
90
95
98
Number of
Samples
16
8
12
12
20
17
21
18
8
19
17
17
20
17
5
20
17
17
14
20
18
4
4
15
21
Std.
Devia-
tion
.034
.025
.029
.081
.071
.024
.037
.037
.046
.038
.012
.011
.028
.014
.032
,032
.014
.013
.012
.014
.037
.000
.050
.024
.026
Rel.
Std.
Devia-
tion
8.5
6.3
15
20
1 A
18
12
9 A
.3
9.3
12
9.5
6.0
5.5
7.0
7.0
8g\
.0
8.0
74&
.0
6.5
64*
.0
3.5
9.3
000
12.5
6.0
6.5
Maximum
Holding
Time
(Days)
21
2
6
27
91
C\
27
97
27
27
2
21
27
27
21
27
91
Zl
41
21
M^
27
27
21
27
1
1
27
27
(continued)
-------
Table 2 (cont'd.)
oo
Spike
Level
Compound gq/l
chlorodibromomethane 0.20
1 ,1 ,2-trlchloroethylene 0.20
1.2-dlbromoethane 0.40
2-chloroethyl ether ether 0.40
2-chloroethylvlnyl ether °-*°
bromoform 0.20
1 ,1 .1 ,2-tetrachloro-
ethane 0.40
1,2,3-trlchloropropane 0.40
chlorocyclohexane 0.40
1 .1 .2 ,2-tetrachloro-
ethane 0.40
1 ,1 ,2.2-tetrachloro-
ethylene 0.20
pentachloroethane 0.40
l-chlorocyclohexene-1 0.40
chlorobenzene 0.40
1-chlorohexane 0.40
bls-2-chloroethyl ether 0.40
bromobenzene 0.40
o-chlorotoluene 0.40
bls-2-chlorolsopropyl
ether 0.40
m-dlchlorobenzene 0.40
o-dtchlorobenzene 0.40
p-dlchlorobenzene 0.40
C.F.T.
no
no
yes
yes
yes
no
yes
yes
yes
yes
no
yes
yes
yes-
yes'
yesu
yes
yes
yes
yes
yes
yes
O.R.N.
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
yes
J3
yes"
yes
yes
yes
yes
yes
yes
Preserved
Sampl e
Thlo.
Sulflde
no
no n
yes"
yesA
yes
no
yes
yes
yes
yes*
no
yes
yes
yesc
yesc
yes
yes
yes
yes
yes
yes
yes
Hean t
Recovery
95
94
93
95
100
95
93
100
93
95
90
98
93
88
83
100
93
85
125
95
95
90
Number of
Samples
17
17
18
IB
21
17
20
20
21
18
17
21
21
18
4
16
20
20
21
21
21
20
Std.
Devia-
tion
.014
.012
.050
.030
.031
.030
.032
.038
.033
.036
.019
.039
.051
.037
.022
.065
.047
.037
.11
.033
.053
nCl
.051
Rel.
Std.
Devia-
tion
7.0
6.0
12.5
7.5
7.8
15.0
8.0
9.5
8.3
9.0
9.5
9.8
12.8
9.3
5.5
16.
12
9.3
28.
8.3
13.
• <>
13.
Maximum
Holding
Time
(Days)
27
ttf
27
21
27
27
27
21
41
21
27
21
27
27
27
21
1
9
21
21
27
*4
27
27
*M
21
A - matrix effect noted due to the presence of sodium sulflte
B - matrix effect noted due to the presence of sodium thlosulfate * , . , »
C - matrix effect noted due to the sample storage (recommended storage time noted In maximum age column)
-------
Table 3. SINGLE LABORATORY ACCURACY AND PRECISION FOR
TRIHALOMETHANES HALL 700 ELECTROLYTIC CONDUCTIVITY DETECTOR
Dose
(ug/1)
1.19
11.9
119
Chloroform
Number
samples
12
8
11
Mean
(ug/1)
1.21
11.3
105
Standard
deviation
0.14
0.16
7.9
Bromodlchloromethane
Dose
(ug/1)
1.60
16.0
160
Dose
(ug/1)
1.96
19.6
196
Dose
(ug/1)
2.31
23.1
231
Number
samples
Mean
(ug/1)
12 1.52
8 15.1
11 145
Chlorodlbromomethane
Number
samples
12
8
11
Bromoform
Number
samples
12
8
11
Mean
(ug/1)
1.91
19.1
185
Mean
(ug/1)
2.33
22.5
223
Standard
deviation
0.05
0.39
10.2
Standard
deviation
0.09
0.70
10.6
Standard
deviation
0.16
1.38
16.3
88
-------
5-° Apparatus and Reagents
5>1 Purge and Trap Device
The purge and trap equipment consists of three separate pieces of
apparatus: the purging device, trap, and desorber. Construction details
for a purging device and an easily automated trap-desorber hybrid which
has proven to be exceptionally efficient and reproducible are shown in
Figures 1 through A.
5.1.1 Purging Device
Construction details are given in Figure 1 for an all-glass 5 mL
purging device. The glass frit installed at the base of the sample
chamber allows finely divided gas bubbles to pass through the sample
while the sample is restrained above the frit. Gaseous volumes above
the sample are kept to a minimum to eliminate dead volume effects, yet
allowing sufficient space for most foams to disperse. The inlet and
exit ports are constructed from heavy walled 1/4 inch glass tubing so
that leak-free removable connections can be made using "finger-tight"
compression fittings containing Teflon ferrules. The removable foam
trap is used to control samples that foam. Purging device design has
been found to be critical for low level analyses. For this reason
variations from Figure 1 will require method revalidation.
5.1.2 Trapping Device
The trap (Figure 2) is a short gas chromatographic column which at
22°C retards the flow of the compounds of interest while venting the
purge gas. The trap is constructed with a low thermal mass so that it
can be rapidly heated for efficient desorption, then rapidly cooled to
room temperature for recycling. Variations in the trap ID, wall thickness,
sorbents, sorbent packing order, and sorbent mass adversely affect the
trapping and desorption efficiencies for certain compounds shown in
Table 1. For this reason, any changes in the trap design will require
method revalidation. Pack the trap according to Figure 2. In order to
function properly the trap must be packed in the following order: Place
the glass wool plug in the inlet end of the trap, follow with the OV-1,
Tenax, silica gel, charcoal, and finally, the second glass wool plug.
89
-------
OPTIONAL
FOAM TRAP
1 4 IN. O.D. EXIT
EXIT 1/4
IN. O.O.
*• 14MM. O.D.
INLET 1/4
IN. O.D.
SAMPLE INLET
2-WAY SYRINGE VALVE
17CM. 20 GAUGE SYRINGE NEEDLE
6MM.O.D. RUBBER SEPTUM
M 10MM. O.D.
INLET
1/4 IN. O.D.
1/16 IN. O.D.
STAINLESS STEEL
13X MOLECULAR
SIEVE PURGE
GAS FILTER
PURGE GAS
FLOW CONTROL
10MM. GLASS FRIT
MEDIUM POROSITY
Figure 1. Purging device.
90
-------
PACKING PROCEDURE
CONSTRUCTION
MULTIPURPOSE TRAP
GLASS WOOL 5MM
ACTIVA1ID CHARCOAL 77CM
ORADI IS
SIUCA Otl 7.7CIM
UMAX 77CM
37. OV-I KM
GLASS WOOL
SMM
I
'.V
V//////////////S.
fe'
$
-->>:'
'i.
RISISTANCC
WIRt WRAPPIO SOIIO
(DOUBLE IAYIR)
I5CM
JU/FOOT RESISTANCC
WIRE WRAPPED SOUD
(SINGLE LAYER)
•CM—*
COMPRESSION rilTINO NUT
AND FERRUIES
THERMOCOUPIE/CONTROUER
SENSOR
ELECTRONIC
TEMPERATURE
CONTROL
AND
PYROMETER
TUBINO 25CM O.IOS IN. I.D.
O.US IN. 0.0. STAINLESS STEEL
TRAP INlfl
Figure 2. Trap.
-------
VO
CARRIER OAS FLOW CONTROL
PRESSURE REGULATOR
LIQUID INJECTION PORTS
PURGE GAS
FLOW CONTROL
13X MOLECULAR
SIEVE FILTER
OPTIONAL 4-PORT COLUMN
SELECTION VALVE
TRAP INLET (TENTAX END)
6-PORT VALVE / ^ RESISTANCE WIRE
COLUMN OVEN
— CONFIRMATORY COLUMN
TO DETECTOR
ANALYTICAL COLUMN
HEATER CONTROL
NOTE: ALL LINES BETWEEN
TRAP AND GC
SHOULD BE HEATED
PURGING DEVICE TO 80°C
Figure 3. Purge-trap system (purge-sorb mode).
-------
LO
CARRIER GAS FLOW CONTROL
PRESSURE REGULATOR
LIQUID INJECTION PORTS
COLUMN OVEN
PURGE GAS
FLOW CONTROL
13X MOLECULAR
SIEVE FILTER
C0lim"
ANALYTICAL COLUMN
OPTIONAL 4-PORT COLUMN
SELECTION VALVE
TRAP INLET (TENAX END)
6-PORT VALVE I RESISTANCE WIRE
HEATER CONTROL
NOTE:
ALL MNFS BETWEEN
TRAP AND GC
SHOULD BE HEATED
TO 80"C
PURGING DEVICE
Figure 4. Purge-trap system (desorb mode).
-------
Reversing the packing order (placing the charcoal in the trap first) will
cause the silica gel and Tenax layers to become contaminated with charcoal
dust causing poor desorption efficiencies. Be sure to install the trap so
that the effluent from the purging device enters the Tenax end of the Trap.
5.1.3 Desorber Assembly
Details for the desorber are shown in Figures 3 and 4. With the
6-port valve in the Purge-Sorb position (See Figure 3), the effluent from
the purging device passes through the trap where the flow rate of the
organics is retarded. The GC carrier gas also passes through the 6-port
valve and is returned to the GC. With the 6-port valve in the Purge-Sorb
position, the operation of the GC is in no way impaired; therefore, routine
liquid injection analyses can be performed using the gas chromatograph in
this mode. After the sample has been purged, the 6-port valve is turned to
the desorb position (See Figure 4). In this configuration, the trap is
coupled in series with the gas chromatographic column allowing the carrier
gas to back-flush the trapped materials into the analytical column. Just as
the valve is actuated the power is turned on to the resistance wire wrapped
around the trap. The power is supplied by an electronic temperature control-
ler. Using this device, the silica gel/charcoal area of the trap is rapidly
heated to 180°C with minimal temperature overshoot and then maintained at
180°C. The trapped compounds are released as a "plug" to the gas chromato-
graph by this heat and backflush step. Normally, packed columns with theore-
tical efficiencies near 500 plates/foot under programmed temperature condi-
tions can accept such desorb injections without altering peak geometry.
Substituting a non-controlled power supply, such as a manually-operated
variable transformer, will provide non-reproducible retention times and poor
quantitative data unless Injection Procedure is used.
A commercial device manufactured by Tekmar has been tested and shown to
be equivalent: Tekmar, P. 0. Box 37202, Cincinnati, Ohio 45202. This
device or its equivalent may be used as long as the proper trap and purging
device are installed on the instrument.
5.2 Gas Chromatograph
The chromatograph must be temperature programmable and equipped with a
halide specific detector.
94
-------
5.2.1 Column I is a highly efficient column which provides outstanding
separations for a wide variety of organic compounds. Because of its ability
to resolve complex mixtures of organochlorine compounds, Column I should be
used as the primary analytical column (See Figure 5).
5.2.1.1 Column I parameters: Dimensions - eight feet long x 0.1 inch
ID stainless steel or glass tubing. Packing - 1% SP-1000 on Carbopack-B
(60/80) mesh. Carrier Gas - helium at 40 mL/minute. Temperature program
sequence: 45°C isothermal for 3 minutes, program at 8°C/minute to 220°C
then hold for 15 minutes or until all compounds have eluted. NOTE: It has
been found that during handling, packing, and programming, active sites are
exposed on the Carbopack-B packing. This results in tailing peak geometry
and poor resolution of many constituents. To correct this, pack the first 5
cm of the column with 3% SP-1000 on Chromosorb-W 60/80 followed by the
Carbopack-B Packing. Condition the precolumn and the Carbopack columns with
carrier gas flow at 220°C overnight. Pneumatic shocks and rough treatment
of packed columns will cause excessive fracturing of the Carbopack. If
pressure in excess of 60 psi is required to obtain 40 mL/minute carrier flow
then the column should be repacked.
5.2.2 Column II provides unique organohalide-separations when compared
to those obtained from Column I (see Figure 6). However, since the resolution
between various compounds is generally not as good as those with Column I,
it is recommended that column II be used as a qualitative confirmatory
column for unknown samples when GC/MS confirmation is not possible.
5.2.2.1 Column II parameters: Dimensions: six feet long x 0.1 inch
ID stainless steel or glass. Packing: n-octane on Porasil-C (100/120
mesh). Carrier Gas: helium at 40 mL/minute. Temperature program sequence:
50°C isothermal for 3 minutes, program at 6°/minutes to 170°C, then hold for
4 minutes or until all compounds have eluted.
5.2.3 Detector - A halogen specific detector must be used in order to
eliminate misidentifications due to non-organohalides which are coextracted
during the purge step.
95
-------
I!'
V"~
! j
i
COmr»* I* »-MW
WOMAN: 41*C-t MMMII t*/«WMI tO 1W<
OPtl«IM« M »••
i i
• M
II
11
t«
tl
Figure 5. Chromatogram of 0.4 yg/L standard.
-------
COIWMM: H-OCtANI ON PORAIIl-C
PIOOIAM: JO'C-J MIMUtfS 4*/MINUtl tO lfO*C
MHCIOt: IIICIIOITIIC CONOUCIIVItT
VO
-J
10 II 14 U
RIIINIION 1IMI MINUKS
II
II
14
Figure 6. Chromatogram of organohalides.
-------
5.2.3.1 A Hall model 700-A available from Tracor has been tested and
found to provide the sensitivity needed to produce meaningful analyses down
to 0.10 |Jg/L for most organohalides with a relative standard deviation of
less than 10%.
Operating conditions for Hall 700-A Detector:
Reactor tube: nickel 1/16" O.D.
Reactor temperature: 810°C
Reactor base temperature: 225°C
Electrolyte: 100% n-propyl alcohol
Electrolyte flow rate: 0.8 mL/minute
Reaction gas: hydrogen at 40 mL/minute
Carrier gas: helium at 40 mL/minute
5.2.3.2 Other halogen specific detectors including electrolytic conduc-
tivity and microcoulometric titration can be used. However, the stability
and sensitivity of these detectors limit the method to measurements down to
1.0 pg/L with a relative standard deviation near 10%.
5.3 Sample Containers - 40 mL screw cap vials sealed with Teflon faced
silicone septa.
Vials and caps - Pierce #13075 or equivalent
Septa - Pierce #12722 or equilvalent
5.4 Syringes - 5 mL hypodermic with luerlok tip (2 each).
5.5 Micro syringes- 10, 100 pL.
5.6 Micro syringe - 25 |JL with a 2" by 0.006 inch I.D. needle (Hamilton
#702N or equilvalent).
5.7 2-way syringe valve with Luer ends (3 each).
5.8 Modified 500 and 1000 mL volumetric flasks. See Figure 7.
5.9 Syringe - 5 mL gas-tight with shut-off valve.
5.10 Trap Materials
5.10.1 Porous polymer packing 60/80 mesh chromatographic grade
Tenax GC (2,6-diphenylene oxide).
5.10.2 OV-1 (3%) on Chromosorb-W 60/80 mesh.
5.10.3 Silica gel-(35/60 mesh) - Davison, grade-15 or equivalent.
5.10.4 Coconut charcoal (26 mesh) Barnaby Chaney, CA-580-26
lot #M-2649 or equilvalent.
98
-------
6MM O.D. HALF-HOLE
CYLINDRICAL SEPTUM
8MM O.D.TUBING
9MM LONG
Figure 7. Modified volumetric flask.
99
-------
5.11 SP-1000 (1%) on Carbopack-B 60/80 mesh available from Supelco.
5.12 n-Octane on Porasil-C (100/120 mesh) avilable from Waters Associates.
5.13 SP-1000 (3%) on Chromosorb-W (60/80 mesh).
5.14 Dechlorinating compound-crystalline sodium thiosulfate, A.C.S.
Reagent Grade.
5.15 Activated carbon (for preparation of organic-free water) - Filtra
sorb-200, available from Calgon Corp., Pittsburgh, PA, or equivalent.
5.16 Organic-free water
5.16.1 Organic-free water is defined as water free of interference
when employed in the purge and trap procedure described
herein. It is generated by passing tap water through a
carbon filter bed containing about 1 Ib. of activated carbon.
5.16.2 A Millipore Super-Q Water System or its equivalent may
be used to generate organic-free deionized water.
5.16.3 Organic-free water may also be prepared by boiling water
for 15 minutes. Subsequently, while maintaining the tempera-
ture 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 with a Teflon seal. NOTE: Test
organic free water daily by analyzing according to paragraph
8.
5.17 Standards
5.17.1 Obtain 97% pure reagent grade reference standards.
5.18 Standard Stock Solutions (compounds boiling above room temperature).
NOTE: Because of the toxicity of organohalides, it is necessary
to prepare primary dilutions in a hood. It is further recommended
that a NIOSH/MESA approved toxic gas respirator be used when the
analyst
handles high concentrations of such materials.
5.18.1 Place about 9.8 ml of methyl alcohol into a 10 mL a ground
glass stoppered volumetric flask.
5.18.2 Allow the flask to stand, unstoppered, for about 10 minutes
or until all alcohol wetted surfaces have dried.
5.18.3 Weigh the flask to the nearest 0.1 mg.
100
-------
5.18.4 Using a 100 pL syringe, immediately add 2 drops of the
reference standard to flask, then reweigh. Be sure that
the 2 drops fall directly into the alcohol without contacting
the neck of the flask.
5.18.5 Dilute to volume, stopper, then mix by inverting the flask
several times.
5.18.6 Calculate the concentration in micrograms per microliter
from the net gain in weight.
5.18.7 Transfer the standard solution to a 10 ml screw-cap bottle
with a Teflon cap liner.
5.18.8 Store the solution at 4°C.
NOTE: With the exception of 2-chloroethylvinyl ether
standard solutions prepared in methyl alcohol are stable up to
4 weeks when stored under these conditions. They should be
discarded after that time has elapsed.
5.19 Standard Stock Solutions (Gaseous Compounds)
5.19.1 Place about 9.8 ml of methyl alcohol into a 10.0 mL ground
glass stoppered volumetric flask.
5.19.2 Allow the flask to stand unstoppered about 10 minutes or
until all alcohol wetted surfaces have dried.
5.19.3 Weigh to the nearest 0.1 mg.
5.19.4 Fill a 5 mL valved gas-tight syringe with the reference
standard to the 5.0 mL mark.
5.19.5 Lower the needle to 5 mm above the methyl alcohol menicus.
5.19.6 Slowly inject the reference standard into the neck of flask
(the gas will rapidly dissolve into the methyl alcohol).
5.19.7 Immediately reweigh the flask to the nearest 0.1 mg.
5.19.8 Dilute to volume, stopper, then mix by inverting the flask
several times.
5.19.9 Transfer the standard solution to a 10 mL screw - Cap
bottle with a Teflon cap-liner.
5.19.10 Store stock solutions at 0°C.
101
-------
5.19.11 Stock solutions prepared from gaseous compounds are generally
not stable for periods exceeding 1 week. They should be
discarded after that time.
5.19.13 Calculate the concentration in micrograms per microliter
from the net gain in weight.
5.20 Calibration Standards
5.20.1 In order to prepare accurate aqueous standard solutions the
following precautions must be observed.
a. Do not inject more than 20 |JL of alcoholic standards
into 100 mL of organic-free water.
b. Use a 25 pLl Hamilton 702N microsyringe or equivalent.
(Variations in needle geometry will adversely effect
the ability to deliver reproducible volumes of methanolic
standards into water).
c. Rapidly inject the alcoholic standard into the expanded
area of the filled volumetric flask. Figure 7. Remove
the needle as fast as possible after injection.
d. Mix aqueou standards by inverting the flask three times
only.
e. Discard the contents contained in the neck of the
flask. Fill the sample syringe from the standard solution
contained in the expanded area of the flask as directed
in paragraph 8.5.
f. Never use pipets to dilute or transfer samples or
aqueous standards.
g. Aqueous standards are not stable and should be
h. Aqueous standards are not stable and should be discarded
after one hour unless stored and sealed according to
6.4.
5.20.2 Prepare, from the standard stock solutions, secondary dilution
mixtures in methyl alcohol so that a 20 (jL injection into 100,
500, or 1000 ml of organic-free water will generate a calibration
standard which produces a response close (+ 10%) to that of
the unknowns.
102
-------
5.20.3 Purge and analyze the aqueous calibration standards in the
same manner as the unknowns.
5.21 Quality Check Standard (0.40 Mg/L)
5.21.1 From the standard stock solutions, prepare a secondary dilution
in methyl alcohol containing 10 ng/pL of each compound
normally monitored. NOTE: It may be necessary to prepare two
or
more quality check standards so that all of the compounds in
each mixture are adequately resolved for quantitative measure
ment.
5.21.2 Daily, inject 20.0 |JL of this mixture into 500 mL of organic-free
water and analyze according to the Procedure Section 6.
6.0 Procedure
6.1 Sample Collection and Handling
6.1.1 The sample containers should have a total volume in excess of
40 mL. Narrow mouth screw cap bottles with the TFE fluorocarbon
faced silicone septa cap liners are strongly recommended.
Crimp-seal serum vials with TFE fluorocarbon faced septa are
acceptable if the seal is properly made and maintained during
shipment.
6.1.2 Sample Bottle Preparation
Wash all sample bottles and TFE seals in detergent. Rinse
with tap water and finally with distilled water. Allow the
bottles and seals to air dry at room temperature, then place
in a 105°C oven for one hour, then allow to cool in an area
known to be free of organics.
NOTE: Do not heat the TFE seals for extended periods of time
(i.e., more than 1 hour) because the silicone layer slowly
degrades at 105°C. When cool, seal the bottles with the TFE
seals that will be used for sealing the samples.
6.2 Sample Preservation - Sodium thiosulfate, a chemical dechlorinating
agent, is added to samples containing free chlorine in order to arrest
the formation of trihalomethanes after sample collection (1). If
chemical preservation is employed, the preservative is also added to
103
-------
the blanks. The chemical preservative (2.5 to 5 mg/40 ml) is added to
the empty sample bottles just prior to shipping to the sampling site.
Do not add sodium thiosulfate to samples when data on maximum trihalo-
methane formation is desired. See Table II in order to determine the
stability of various organohalides in the presence of sodium thiosulfate
6.3 Sample Collection
Collect a miminum of two replicates from each sample source. Fill the
sample bottles in such a manner that no air bubbles pass through the
sample as the bottle is being filled. Seal the bottles so that no air
bubbles are entrapped in it. Maintain the hermetic seal on the sample
bottle until time of analysts.
6.3.1 Sample from a water tap.
Turn on water and allow the system to flush. When the temperature of
the water has stabilized, adjust the flow to about 500 mL/minute and collect
duplicate samples from the flowing stream.
6.3.2 Sampling from an open body of water.
Fill a 1-quart wide-mouth bottle or 1-liter beaker with sample from a
representative area. Carefully fill a minimum of two sample bottles from
the sampling container as noted above. If preservative has been added to
the sample bottles, then fill with sample just to overflowing, seal the
bottle, and shake vigorously for 1 minute.
6.3.3 Sealing practice for septum seal screw cap bottles.
Open top bottle and fill to overflowing, place on a level surface,
position the TFE side of the spetum seal upon the convex sample meniscus and
seal the bottle by screwing the cap on tightly. Invert the sample and
lightly tap the cap on a solid surface. The absence of entrapped air indi-
cates a successful seal. If bubbles are present, open the bottle, add a few
additional drops of sample and reseal bottle as above. NOTE: If the
septum seals are inverted (i.e., the silicone side against the sample) then
significant organohalide losses will occur in shipment and storage.
6.4 Preparation of Blanks
Sample blanks must be prepared and accompany the samples wherever the
samples are shipped or stored. If the samples are immediately analyzed at
the sampling site then blanks are not required.
104
-------
Prepare blanks in replicate at the laboratory by filling and sealing a
minimum of two sample bottles with pre-tested organic-free water just prior
to shipping the sample bottles to the sampling site. If the sample is to be
preserved, add an identical amount of preservative to the blanks. Ship the
blanks to and from the sampling site along with the sample bottles. Store
the blanks and the samples collected from a given source (sample set),
together. A sample set is defined as all the samples collected from a given
source (i^e., at a water treatment plant, the replicate raw source waters,
the replicate finished waters and the replicate blank samples comprise the
sample set). Store the sample set in an area known to be free of organic
vapors. See Table II for maximum storage time.
6.5 Conditioning Traps
6.5.1 Condition newly packed traps overnight at 200°C by backflushing
with an inert gas flow of at least 20 mL/min. Vent the trap effluent to the
room, not to the analytical column. Prior to daily use, condition traps 10
minutes while backflushing at 180°C. The trap may be vented to the analytical
column; however, after conditioning the column must be programmed prior to
use.
6.6 Extraction and Analysis
Adjust the purge gas (nitrogen or helium) flow rate to 40 mL/min.
Attach the trap inlet to the purging device. Turn the valve to the purge-sorb
position (Figure 3). Open the syringe valve located on the purging device
sample introduction needle. Remove the plungers from two 5 mL syringes and
attach a closed syringe valve to each. Open the sample bottle (or standard)
and carefully pour the sample into one of the syringe barrels until it
overflows. Replace the syringe plunger and compress the sample. Open the
syringe valve and vent any residual air while adjusting the sample volume to
5.0 ml. Close the valve. Fill the second syringe in an identical manner
from the same sample bottle. This second syringe is reserved for a replicate
analysis, if necessary.
Attach the syringe-valve assembly to the syringe valve on the purging
device. Open the syringe valve and inject the sample into the purging
chamber. Close both valves. Purge the sample for 11.0 + .05 minutes.
After the 11 minute purge time, attach the trap to the chromatograph (turn
105
-------
the valve to the desorb position) and introduce the trapped materials to the
GC column by rapidly heating the trap to 180°C while backflushing the trap
with an inert gas between 20 and 60 mL/min for 4 minutes. If the trap can
be rapidly heated to 180°C and maintained at this temperature, the GC analysis
can begin as the sample is desorbed, i.e., the column is at the initial 45°C
operating temperature. The equipment described in Figure 4 will perform
accordingly.
With other types of equipment where the trap is not rapidly heated or
is not heated in a reproducible manner, it is necessary to transfer the
contents of the trap into the analytical column at 30°C where it is once
again trapped. Once the transfer is complete (4 minutes), the column is
rapidly heated to the initial operating temperature for analysis. Note: In
some cases it may be necessary to cool the column down to 0°C. If injection
procedure 8.9.1 is used and the early eluting peaks in the resulting chroma-
togram have poor geometry or variable retention times, then method 8.9.2
should be used.
While the extracted sample is being introduced into the gas chromatograph,
empty the purging device using the sample introduction syringe, follow by
two 5 ml flushes of organic-free water. After the purging device has been
emptied, leave the syringe valve open to allow the purge gas to vent through
the sample introduction needle. After desorbing the sample for approximately
four minutes recondition the trap by returning the valve to the sorb position,
wait 15 seconds then close the syringe valve on the purging device. Maintain
the trap temperature at 180°C. After approximately seven minutes turn off
the trap power and open the syringe valve. NOTE: If the operations described
in 8.11 are omitted then large amounts of water will be injected into the
column which will cause large narrow peaks to appear in the early elution
area of the chromatogram from the following analysis.
Analyze each sample and sample blank from the sample set in an identical
manner (see 6.4.9.5) on the same day. Prepare calibration standards from
the standard stock solutions (5.20) in organic-free water that are close to
the unknown in composition and concentration. The concentrations should be
such that no more than 20 |jl of the secondary dilution need be added to 100
106
-------
to 1000 mL of organic-free water to produce a standard at the same level as
the unknown.
6.7 Calculations
Quantify the unknowns by comparing the peak height of the unknowns to
the standard peak height (obtained from calibration standards). Round off
the data to the nearest .01 (Jg/L.
= peak height sample n }
peak height standard ' r6
Report the results obtained from the EMSL Quality Control Sample and the
lower limit of detection estimates along with the data for the unknown
samples. Calculate the limit of detection (LOD) for each organohalide not
detected using the following criteria:
LOD CMg/L) = 0.4 <
where: A = 5 times the noise level in (mm) at the exact retention
time of the organohalide or the baseline displacement in
(mm) from the theoretical zero at the exact retention time
of the organohalide. B = peak height (mm) of 0.4 (Jg/L quality
check standard ATT = Attenuation factor.
7.0 Quality Assurance Program
7.1 Analytical Quality Control
Analyze the 0.40 (Jg/L quality check sample daily before any samples are
analyzed. Instrument status checks and olower limit of detection estimators
based upon response factor calculations at five times the noise level are
obtained from these data. In addition, respone factor data obtained from
the 0.40 (Jg/L quality check standard can be used to estimate the concentration
of the unknowns. From this information the appropriate standard dilutions
can be determined.
Analyze the EMSL-Cincinnati volatile organics quality control samples
or their equivalent on a quarterly basis. Analyze the sample blank or a
method blank to monitor for potential interferences as described in Section
3.0. Perform the following instrument status checks using the data gathered
from blanks duplicate analyses and the quality check sample.
107
-------
7.1.1 Peak Geometry Check
7.1.1.1 All of the peaks contained in the quality check chromatogram
must appear to be sharp and symetrical. Peak tailing in
excess of that shown in the method chromatogram Figure 5 must
be corrected.
Tailing problems are generally traceable to:
A. Active sites on column - repack
B. Reactor temperature too low.
C. Reactor base temperature too low.
D. Contaminated reactor tube - recondition/replace.
E. Contaminated reactor transfer line - replace
F. Detector flow too low.
G. Spent ion exchange column - replace.
7.1.1.2 If only the compounds eluting before chloroform give random
responses, unusually wide peak widths, are poorly resolved, or are missing,
the problem is usually traceable to the trap/desorber.
7.1.1.3 If only brominated compounds show poor peak geometry or do not
properly respond at low concentrations, repack the trap.
7.1.1.4 If negative peaks appear in the chromatogram replace the ion
exchange column and replace electrolyte.
7.1.1.5 Retention times for the organohalides should remain constant
throughout the day (less than 10% variance).
7.1.2 Replicate analyses
A properly operating gas chromatograph should perform with an average
relative standard deviation of less than 6% over a concentration range of
0.1 to 100 |Jg/L. Poor precision is generally traceable to pneumatic leaks
especially around the detector reactor inlet and exit.
The method blank analysis should represent less than a 0.1 |jg/L response
or less than a 10%. Interference for those compounds that occur routinely.
Any instrument not performing according to 7.1 specifications should be
considered "out of control". The instrument must be "in control" before
acceptable data can be generated.
108
-------
7.1.3 Confirmatory Analyses
Confirmatory analyses are performed using dissimilar columns. If
sufficient material is present then confirmatory analyses are performed by
gas chromatography-mass spectrometry.
Aqueous standards and unknowns are always extracted and analyzed under
identical conditions in order to compensate for extraction losses.
8.0 References
1. Identification and Analysis of Organic Pollutants in Water, Keith, L. H.,
Ann Arbor Science, p. 87 (1976).
109
-------
ANALYTICAL PROTOCOL: DETERMINATION OF ORGANIC CONTAMINANTS BY
GROB CLOSED-LOOP-STRIPPING ANALYSIS (CLSA)
1.0 Principle of the Method
In 1973 in Zurich, Switzerland, Grob (1) reported on CLSA for the
measurement of semivolatile, intermediate molecular weight organics in
drinking water at the part-per-trillion (nanogram-per-liter) level. Grob
CLSA is accomplished by making the sample vessel, containing a headspace,
part of a closed system in which the entrapped air is continually circulated
through the sample and an activated carbon filter. As the air passes through
the sample, organic compounds are purged from the water into the headspace
air. The carbon filter then adsorbs these compounds from the air. The
carbon filters are extracted with a small amount of solvent and the extracts
are analyzed by capillary GC/MS.
2.0 Range and Detection Limit
The CLS technique is capable of analyzing water samples containing
purgeable organic contaminants at the low ng/L (ppt) level. Reported analyses
(2,3) indicate an acceptable analytical concentration range of ^1-100
ng/L.
3.0 Interferences
Impurities in the purge gas and plumbing ahead of the trap can lead to
contamination problems. The analytical system can be demonstrated to be
free from contamination under the conditions of the analysis by running
method blanks. Method blanks are run by charging the purging device with
organic-free water and analyzing it in a normal manner. The use of non-TFE
plastic tubing, non-TFE thread sealants, or flow controllers with rubber
components in the purging device is avoided.
Samples can be contaminated by diffusion of volatile organics (particu-
larly freons and methylene chloride) during shipment and storage. A sample
blank prepared from organic-free water and carried through the sampling and
handling protocol can serve as a check on such contamination.
Cross contamination can occur whenever high level and low level samples
are sequentially analyzed. To reduce the likelihood of this, the purging
device is rinsed between samples with organic-free water. Whenever an
110
-------
unusually concentrated sample is encountered, it is followed by an analysis
of organic-free water to check for cross contamination.
The use of a high resolution separation technique (capillary GC) coupled
with a highly specific detection device (MS) reduces the likelihood of
interference problems. Published reports (2,3,4) have shown that the method
is not seriously impaired by interferences when environmental samples were
analyzed.
4.0 Precision and Accuracy
Statistical analyses (SA) to determine standard deviations have been
performed on RRTs, GC peak areas, response factors, amounts, and recovery
efficiencies. SA for both total ions and single ion quantitation have been
compared. Greater qualitative and quantitative accuracy was achieved by
single ion quantitation for complex environmental samples due to co-eluting
GC peaks. Table 1 shows the SA results of the repetitive direct injections
with quantitation results based on a single ion. This table shows that peak
areas vary an average of 13% whereas the amounts vary an average of 9%.
Since the sensitivity of the mass spectrometer fluctuates daily, these area
variations are expected. However, most importantly, the internal standard
method of quantitation compensates for these changes when the amounts are
calculated by the computer.
5.0 Apparatus and Reagents
The original closed-loop stripping apparatus of Grob (1) has been
modified for increased durability and ease of sample handling. With the
exception of the sample container, all glass parts have been eliminated and
§
the apparatus constructed of type 304 stainless steel and Teflon as seen in
®
Figures 1 and 2; tubing connections are made with Swagelok fittings containing
Teflon ferrules.
The sample container is a one-gallon jug on which the neck opening has
been ground flat to ensure a leak-proof seal during the vapor phase stripping.
With this system, a sample can be collected and analyzed in the same container.
Contamination is minimized, spillage and errors introduced by sample transfer
are eliminated, sample changeover is simplified, and the time required is
greatly reduced. Figure 3 shows a schematic diagram of the closed-loop
stripping apparatus.
Ill
-------
Table 1. STATISTICAL ANALYSIS OF REPETITIVE DIRECT INJECTION STUDY WITH
QUANTITATION BASED ON A SINGLE ION
Compound
Quantisation Mass
(m/e)
Relative Retention
Time (RRT)*
Area
Amount*
(ng)
Bis-(2-chloroethyl)ether
1,4-Dichlorobenzene
2-Ethy1-1,4-dimethylbenzene
1,2,4-Trichlorobenzene
Hexachloro-1,3-butadiene
1-Chlorododecane (I.S.)
2,2*,4,5,5'-Pentachlorobiphenyl
93
146
119
180
225
91
254
0.525 + .001
0.557 + .001
0.635 + .001
0.728 + .001
0.776 + .000
1.000 + .000
1.447 + .002
125,339 + 20,978
174,043 + 21,770
231,704 + 30,954
183,366 + 22,306
79,039 + 8,047
37,512 + 7,667
17,564 + 1,505
61.0 + 3.1
50.0 + 4.6
42.8 + 3.4
72.6 + 6.5
84.0 + 9.6
52.0 + 0.0
50.0 + 7.1
Average value + standard deviation/based on 10 injections.
-------
Silver Soldered
38-400 Plastic Cap
with Hose Clamp
Jug-type 1 Gallon Bottle
•«—1/8" Stainless Steel Tubing
Stainless Steel Swagelock Union
'20u Porosity Stainless Steel Sparger
Figure 1. Sample container for CLSA.
113
-------
Silver Soldered
24 Gauge Stainless Steel
Disc 35mm Diameter
PTFE Washer
38-400 Plastic Cap
with Hose Clamp
— 1/8" Stainless Steel Tubing
Stainless Steel Swagelock Union
20u Porosity Stainless Steel Sparger
Figure 2. CLSA sample container plumbing and fittings.
114
-------
ACTIVATED CARBON
FILTER HOLDER
GAS
HEATER
THERMOSTATIC WATERBATH
Grob ond Zurcher
J. Chromatogr. 1976
Figure 3.' Schematic diagram of closed loop stripping apparatus
115
-------
6.0 Procedure
6.1 Sample Collection
Water samples are collected directly in muffled (400°C) bottles contain-
ing 100 mg of sodium sulfite to quench residual chlorine and 40 rag of mercury
chloride (HgCl2) to retard bacterial growth. The bottle is filled completely
without any headspace and the Teflon-lined caps are applied securely.
Samples are stored at 5°C until analyzed.
6.2 Sample Stripping
The HERL closed-loop stripping apparatus is shown schematically in
Figure 3. The water bath temperature is thermostatically controlled at 30°C;
the preheater temperature is maintained at 80°C to prevent condensation of
vapors on the carbon filter; the filter holder is insulated from the surrounding
atmosphere and its temperature is maintained about 40°C.
CLSA is initiated by decanting a small amount of the water sample down
to the "one-gallon" mark, adding 0.6 pL of internal standard solution (1-
chloroalkanes in acetone) and assembling the purging apparatus. The 0.6 |JL
of internal standard solution when added to the sample gives a concentration
of 52 ng/L each of 1-chlorohexane, 1-chlorooctane, 1-chlorododecane and
1-chlorohexadecane and 260 ng/L of 1-chlorooctadecane. The latter amount is
added due to poorer recovery efficiency of 1-chlorooctadecane. With a
clean, dry, carbon filter in place, the sample is purged by recirculating the
headspace for 2 hours.
6.3 Carbon Filter Extraction
The carbon filter extraction procedure using carbon disulfide is very
similar to that used by Grob and others (1,2,3,5). The carbon filter is
removed and assembled into the extraction apparatus (5). The filter is
butted up tightly to the receiver tube and secured with a Teflon sleeve. The
first 6 pL aliquot of CS2 is added directly to the top of the carbon and is
refluxed back and forth through the carbon 3 times by alternately cooling the
filter with ice to pull the solvent down just below the carbon, then warming
with the hand to push the solvent back through. When the solvent is pulled
through a fourth time, the solvent is transferred to the receiver by swinging
the tube in a long downward arc. As most of the solvent aliquot will be
required to wet the carbon, only about 2-3 [iL will be recovered. The
116
-------
process is repeated with three additional aliguots of 2 pL each which results
in a total sample volume of about 8 pL. Two microliters of this final
2 2
volume are used for (GC) or (GC) /MS analysis. The entire extraction
procedure takes about 10 minutes to complete.
6.4 Analysis
2
For this protocol all (GC) /MS analyses are performed on a Finnigan
Model 3300 mass spectrometer equipped with an Incos Model 2300 data system
computer. A Finnigan Model 9500 gas chromatograph equipped with a Grob
designed splitless injector (6) and a 60m x 0.25 mm i.d. WCOT glass capil-
lary SP2100 column was interfaced to the mass spectrometer with a glass-lined
stainless steel dual transfer line held at 230°C.
Two microliters (2 |JL) of sample in CS? are injected with the split
closed. The split is opened (20:1 ratio) after 30 sec. Grob's hot needle
technique (7) is used for injection of all samples and standards. The
helium carrier gas flow rate is about 3 mL/min at 25°C; the linear velocity
is 25 cm/sec with a helium head pressure of 20 psi (no flow controller).
The injector is maintained at 260°C. The column is temperature programmed:
20°C isothermal (cooled with liquid nitrogen) for approximately 8 min, then
2°C/min to 250°C. Mass spectra are acquired at the rate of one per 2 sec
from 14-450 amu at 70 eV and 90°C source temperature. Source pressure is
_/- 2
approximately 5 x 10 torr. For a typical (GC) /MS analysis of this type,
3900 scans are recorded (130 minutes from injection to termination). Con-
firmation of identifications are made by comparison of GC and GC/MS proper-
ties with those of authentic standards or library spectra.
6.5 Quantitation
Automatic computerized quantitation procedures based on the internal
standard method of quantitation are used in this method. The quantitation
formula used by the computer is:
. ^ Area x Internal Standard Amount
Amount =
Internal Standard Area x Response Factor
where
p _ Area x Internal Standard Amount
" Amount x Internal Standard Area
117
-------
7.0 Quality Assurance
Since many organics in water can be measured at 5 ng/L or less with
CLSA, it is extremely necessary to incorporate procedural blanks into the
analytical scheme. Prior to analyzing actual samples of drinking water,
filter blanks and CLSA blanks of low organic water (Milli-Q Water) containing
the same internal standard mix are performed. When spiking water with low
levels of organic chemicals for recovery studies one must consider the
background level of these chemicals in the spiking medium for correct quanti-
tation results.
Blank water samples are run between highly contaminated environmental
samples. If a series of samples is to be analyzed, the least contaminated
samples are run first.
Standard quality assurance practices should be used with this method.
Field replicates should be collected to validate the precision of the sampling
technique. Laboratory replicates should be analyzed to validate the precision
of the analysis. Fortified samples should be analyzed to validate the
accuracy of the analysis.
8.0 References
8-1 Grob, K., Organic Substances in Potable Water and in its Precursor,
Part I. Methods for Their Determination by Gas-Liquid Chromatography,
J. Chromatogr., 84, 255 (1973).
8-2 Stieglitz, L., Roth, W., Kuhn, W., and Leger, W., The Behavior of
Organohalides in the Treatment of Drinking Water, Vom Wasser, 47, 347
(1976).
8-3 Schwarzenbach, R. P., Molnar-Kubica, E., Giger, W. , and Wakeham, S. G.,
Distribution, Residence Time, and Fluxes of Tetrachloroethylene and
1,4-Dichlorobenzene in Lake Zurich, Zwitzerland, Envir. Sci. and Tech.,
13(11), 1367 (Nov. 1979).
8-4 Zurcher, F. and Giger, W., The Study of Volatile Organic Compounds
in the Glatt River, Vom Wasser, 47, 37 (1976).
8-5 Grob, K. and Zurcher, F., Stripping of Trace Organic Substances from
Water Equipment and Procedure, J. Chromatogr., 117, 285 (1976).
8-6 Grob, K. and Grob, K., Jr., Splitless Injection and the Solvent Effect,
HRC & CC, 1(1), 57 (July 1978).
118
-------
8-7 Grob, K. and Grob, G., Practical Capillary Gas Chromatography - a
Systematic Approach, HRC & CC, 2(3), 109 (Mar. 1979).
119
-------
ANALYTICAL PROTOCOL: ANALYSIS OF PURGEABLE ORGANIC COMPOUNDS IN WATER
(MASTER ANALYTICAL SCHEME)
1.0 Principle of the Method
This method describes the determination of purgeable organic compounds
in drinking water, surface waters and treated municipal and industrial
wastewater effluents. The procedure utilizes gas stripping analysis to
effectively partition semi-soluble and insoluble volatile organic compounds
between gaseous and aqueous phases followed by trapping on a Tenax GC cartridge
Exposed cartridges are thermally desorbed and the compounds released are
analyzed by GC/MS/COMP using capillary columns. Summarized recovery data
relating to the procedure below is presented in Figure 1 and Table 1.
2.0 Range and Detection Limits
Detection limits are:
0.1 ppb in drinking water
1.0 ppb in surface water
10 ppb in industrial wastewater effluent
3.0 Interferences
Background from the Tenax cartridge can present problems with identifi-
cation and quantitation. Phenolics and amines can mask other compounds of
interest (because of chromatographic tailing). If the pH of sample is
adjusted to seven the phenols and amines are in an ionic state and will not
purge.
4.0 Precision and Accuracy
Table 1 presents mean % recovery and confidence limits for five compounds
in six different water matrices. Minor variations in recovery data are due
to the fact that the experimental procedure was modified slightly during the
study.
5.0 Apparatus and Reagents
For Purging --
1. Purge flask, 200 mL capacity (Figure 2).
2. Tenax GC sorbent cartridges.
3. Flowmeter, (10-100 mL/min).
120
-------
WATCH .
MATRIX
— I * * B *
liliii
w
i
i
frt
HI
lii
t
f:
Hill!
¥
m
VI
Hi
COMPOUND
METHYL BROMIDE
CHLOROFORM
TrllurnENE
TOLUENE
BROMOBENZENE
cartridge* •cored overnight give low recovery.
Duplicate data only.
^Experimental procedure altered:aemi-clo«ed «ysteB.
Experlaents performed with old atock of brovobenzene.
pfean of 2 triplicate experlnenta.
Single determination only.
Figure 1. Recoveries for selected compounds in various water matrices and particulate types.
-------
Table 1. SUMMARY OF RECOVERY DATA FOR SELECTED COMPOUNDS IN
VARIOUS WATER MATRICES3'
Compound Mean % recovery + standard deviation (C.V.)
methyl bromide0 76+5 (7) 81 (30)
chloroform 91+6 (7) 91 (12)
thiophene 92+6 (7) 92 (11)
toluene 97 + 4 (5) 97 (14)
bromobenzene 66+9 (14) 71 (8)
a
Experiments were performed at the 10 ppb level with the exception of
those at the limit of detection (0.2 to 2 ppb).
Water matrices examined were: distilled water, 10% municipal waste-
water effluent, 10% energy effluent, distilled water (limit of detec-
tion), 500 ppm activated sludge, and 500 ppm river bottom particulates.
Q
Does not include activated sludge results.
122
-------
1/4" ao.
200ml Capacity
Medium Porosity Frit
Figure 2. Purge Flask, 200 mL capacity.
123
-------
4. Reagents
a. Sodium sulfate, anhydrous powder (ACS grade).
b. Phospate buffer, 2.0M, pH7-
c. High purity water, purged.
d. Helium gas, scrubbed to remove volatiles.
5. Liquid-nitrogen trap (3' of 1/4" O.D. copper tubing,
coiled and immersed in liquid nitrogen).
6. Pipets, volumetric, 1 ml and 20 mL.
7. Beckmann fittings, #416.
8. Quick-connect fittings, 1/4" tube, double end shut-off.
9. Water bath, 30°C.
6.0 Procedure
6.1 Sample Collection, Preservation and Handling
6.1.1 Sample Collection
Grab samples must be collected in glass containers having a total
volume in excess of 40 mL. Fill the sample bottles in such a manner that no
air bubbles pass through the sample as the bottle is being filled. Seal the
bottle so that no air bubbles are entrapped in it. Maintain the hermetic
seal on the sample bottle until time of analysis.
6.1.2 Sample Preservation
The samples must be iced or refrigerated from the time of collection
until extraction. If the sample contains free or combined chlorine, add
sodium thiosulfate preservative (10 mg/40 mL will suffice for up to 5 ppm
Cl_) to the empty sample bottles just prior to shipping to the sampling
site, fill with sample just to overflowing, seal the bottle, and shake
vigorously for 1 minute.
6.1.2 Sample Handling
All samples should be analyzed within 14 days of collection.
6.2 Extraction/Purge/Etc
6.2.1 Tenax GC Cartridge Preparation
Tenax GC is prepared by extraction with redistilled, pesticide analysis
grade methanol in a Soxhlet apparatus for 48 hours with a cycle time of
approximately fifteen minutes followed by a similar extraction with redistil-
led, pesticide analysis grade pentane. After air drying, the Tenax is
124
-------
placed in a vacuum oven at 100°C for at least 24 hours. Cartridges (1.3 x
6.0 cm; 35/60 mesh) are prepared in 1.6 x 10 cm Pyrex tubes with 1 cm glass
wool (silanized) plugs at each end and thermally desorbed at 260°C for a
minimum of 2 hours under a helium flow of 30 mL/min. Following thermal
desorption the cartridges are transferred, hot, to Pyrex culture tubes (25 x
®
150 mm) with Teflon -lined screw caps and cooled to room temperature. The
Teflon cap-liners are a major source of contamination and must be carefully
solvent rinsed and vacuum-dried overnight. All gases which come in contact
with the Tenax GC cartridges, as well as water samples must be passed through
liquid-nitrogen traps to remove volatile components. Cartridge preparation
and thermal desorption is done in a lab free of solvent vapors.
Tenax GC background is checked in a manner identical to the procedure
for analyzing loaded cartridges. Injection of adsorbed materials from Tenax
GC onto a gas chromatographic column is accomplished using the thermal
desorption system illustrated in Figure 3 and 4. This system consist of
four main components: a desorption chamber, a six-port, two-position,
high-temperature, low-volume, valve (Valco Instruments, Inc.), a Ni capillary
cyrogenic trap, and a temperature controller. The stainless steel thermal
desorption chamber and six-port valve are encased in a common aluminum
sandwich which serves as a heating block. The chamber itself has an overall
length of 12 cm and accommodates a Pyrex sampling cartridge of dimensions 13
mm i.d. x 10 mm o.d. and 10 cm length. Two, 150 W, 115V heating cartridges
are used to heat the aluminum sandwich and the temperature is controlled and
monitored with iron-constantan thermocouples and output on a pyrometer
(Omega Engineering, Inc.). The desorption chamber is connected to the valve
with a short section of Ni capillary tubing (0.50 mm i.d., 1.47 mm o.d.).
Similar capillary tubing is used for the cryogenic trap (Figure 3). This
trap, constructed of aluminum, is cooled to -195°C by passing cryogenically
cooled nitrogen gas through stainless steel tubing (18" x 1/6" o.d. x 0.04"
i.d.) prior to entry into the body of the trap. This allows collection and
concentration of vapors desorbed from the sorbent cartridge. The vapors are
injected onto the GC column by rapid heating of the capillary trap to 270°C
provided by a 150 W cartridge heater located inside the aluminum cylinder.
The desorption chamber is interfaced to the capillary GC column with a
125
-------
10
o\
PLATINUM
PROBE SENSOR
TEFLON SEAL
STRING
METAL SEAL
HEATING
CARTRIDGE
CARRIER
GAS
TOOIC
CAPILLARY
HEATING
CARTRIDGE
V VENT TO
CARBON TRAP
TWO POSITION
SIX PORT
Hum. VAIVE
CARTRIDGE
— HEATING AND
COOLING BATH
LIQUID NITROGEN
VALVE POSITION A
(SAMPLE DESORPTION)
VALVE POSITION B
(SAMPLE INJECTION)
Vttrt
Figure 3. Thermal desorptlon Inlet-manifold for Tenax GC cartridges,
-------
004 In
Iflt la. tv*|tlek Kt
L— LCcm —J
Figure 4. Cryo-heater module for inlet-manifold.
127
-------
minimal length of gold-plated Ni capillary tubing (0.40 mm i.d., 1.57 mm
o.d.), deactivated with OV-17. Connection between the Ni transfer line and
glass capillary is made with 1/16" stainless steel "zero dead volume" union
(8 ®
(Swagelok , Crawford Fitting Co.) using stainless steel and Vespel Ferrules.
In a typical thermal desorption/injection cycle, an exposed cartridge
is placed in the preheated (240°C) chamber with a flow of He gas (15 mL/min)
through the cartridge to purge the desorbed vapors into the cryogenic trap;
this constitutes valve position A (Figure 3). After 8 min of thermal desorp-
tion, the six-port valve is rotated to position B, the temperature on the
capillary trap is rapidly raised (>100°/min), and carrier gas sweeps the
vapors onto the gas chromatographic column. Upon reaching the maximum trap
temperature, the trap heater is turned off; however, the valve is retained
in position B with the cartridge in the desorption chamber. After approxima-
tely 30 min the cartridge is removed and the valve is returned to Position
A. The removal of the cartridge and the switching of the valve should be
done simultaneously or as closely together as possible.
The criteria for Tenax GC cartridge acceptability are, for the most
part, empirical, and are based on the following chromatographic evidence
-12
obtained at 128 x 10 amps full scale on a Varian 3700 gas chromatograph
using a 50 m x 0.25 mm SE-30 WCOT column temperature programmed from 40 to
220°C at 4°/min:
1. No peaks with the exception of pentane and methanol,
over 10% full scale. Pentane and methanol are generally
30% full scale.
2. No more than 15 peaks over 3% full scale.
6.2.2 Distilled/Deionized Water Purification
Interfering volatile substances in distilled and/or deionized water are
removed by purging with purified helium at 90°C and 50 mL/min for at least
two hours. Water purified in this manner should be used immediately.
6.2.3 Sample Extraction
1. Assemble the purge flask, except for the Tenax GC cartridge,
and place in a 30°C bath.
128
-------
2. Weigh out 60 g of anhydrous Na_SO, and transfer to the
purge flask. Purge with helium* at 20 raL/min for approxi-
mately 15 minutes.
3. With the Tenax GC cartridge in place, the flask and
contents at 30°C, the helium flow stopped and the gas
inlet tube stoppered, transfer exactly 200 mL of sample to the
flask as quickly as possible. Immediately, close the
stopcock just ahead of the cartridge, remove the flask
from the batch and shake vigorously to dissolve the salt.
4. Return the flask to the water bath, reconnect the gas
line, open the stopcock and purge the sample with 500 mL
helium using a flow rate between 10 and 100 mL/min.
5. Remove the exposed cartridge to its screw-capped culture
tube and store, if necessary, at -10°C in a sealed metal
container.
6. Analyze cartridges by capillary GC/MS/COMP-
6.3 Analysis
Analyses of the catridges as done on a glass capillary column - 50
m x 0.5 mm i.d., coated with SE-30 (tentatively) by a GC/MS/COMP system.
6.4 Qualitative Identification
Qualitative analysis is to be performed either by computer searches
or manual interpretation of information acquired by utilizing a full
scan mode.
6.5 Quantitation
A software program will provide for calculating the quantity of
each component in each of the sample extracts using relative molar
response factors (ratioed to the internal deuterated standards). Also
it will correct for the recovery of each organic compound from the water
matrix based on the recovery data which was generated under this program
or by the user. Finally, it will calculate the concentration of each
organic in the original water sample.
*The cryogenic trap must be used at all times to prevent contamination of
the sample with volatile components in the purge gas.
129
-------
7.0 Quality Assurance Program
An on-going quality assurance program is required to assure the
data quality. Quality control (QC) procedures determine artifacts,
losses, etc. through a system of blanks and controls. Quality assurance
(QA) procedures monitor the execution of the procedure and check data
interpretations and calculations.
7.1 Quality Control
Prior to a field sampling trip, chough blanks and controls are
prepared to equal 10%, each (2 minimum) of the anticipated number of
field samples. Blanks consist of 40 mL of purged distilled water in the
same type of sampling container as is used in the field. Controls
consist of the 40 mL of purged distilled water and are spiked with known
compounds. These blanks and controls are carried to the field and
receive the same handling as the field samples. Workup and analyses of
field blanks and controls is interpreted with field samples on a regular
basis. This method allows assessment of sample storage stability.
Table 2 presents a typical set of blanks and controls for quality
control on a field trip were 50 tap water samples are to be collected.
7.1.2 Procedural Blanks and Controls
7.1.2.1 Extraction Blanks
With each set of samples, a procedural blank is run. This consists
of 10 mL of prepurged distilled water which is extracted under the same
conditions as the samples. These blanks are designed to detect artifacts
from dirty glassware, laboratory atmosphere instrusion, and other sources
7.1.2.2 GC/MS Procedural Control
At the start of each working day, a mixture of 2,6-dimethylphenol,
2,6-dimethylaniline, and acetophenone (PA mixture) is analyzed to monitor
the capillary GC column performance. This also serves to check the mass
spectrometer tuning.
Field samples, field controls, field blanks, and procedural blanks
are queued up for GC/MS analysis such that at least one QC sample is run
each working day. In addition, a standard solution is analyzed each day
to serve as a procedural control and also to update the RMR value.
130
-------
Table 2. QC SAMPLES FOR A TYPICAL SAMPLING TRIP
Sample type
Number
Comment
Field Blank
Field Control
Lab Blank
5
5
Lab Control
Freeze after preparation
vary to field, store with
field samples
Store with field blanks
Freeze after preparation,
store in same freezer as
field samples will be
stored
Store with lab blanks
131
-------
Thus, in a typical working day, 4 field samples, 1 blank or control, and
1 RMR standard are run.
7.2 Quality Assurance
Both internal and external quality assurance procedures are to be
followed. Internal quality assurance procedures assure the continuity
and consistency of the data. External quality assurance procedures
(interlaboratory checks) verify or dispute the accuracy of the data.
7.2.1 Internal Quality Assurance
7.2.1.1 Supervision and Monitoring of Activities
There are three levels of quality assurance (QA). The primary
quality assurance is the person conducting the sampling and/or analysis.
This person must be aware of their actions, observe events which may
effect the data, and maintain appropriate records. At the second level,
the chemist's supervisor monitors their daily activities, reviews the
notebooks, checks data and calculations, and assists in "troubleshooting"
problems. At the tertiary level, a QA coordinator interviews all person-
nel on the project. The interviews cover the operations they perform
(precisely), the data they obtain, a spot-check of their calculations,
and any problems they have had.
7.2.1.2 Documentation
Chain of Custody
From the initial preparation of a sample container through reporting
of the analytical results, each sample is accompanied by a chain of
custody sheet. Each person signs in the time of receipt, operations
performed, and transmittal of the sample. This record is important for
tracing a contaminant, bad standard, or some other problem.
Sampling Protocol Sheets
When a sample is collected, a sampling protocol sheet is filled in
which contains a discrete sample code which identifies project number,
area, site, locations, trip number, sampling period, and sample type.
Also included are sample times, volumes, addresses, meteorology, and
other pertinent information. Where appropriate, a map is made to precise-
ly identify the location.
132
-------
Sample Log
Upon return from a sampling trip, each sample code is entered into
a sample log book. This log is updated as samples proceed through
workup and analysis. Thus, at a glance, project personnel can tell the
status of each sample and find out how many are at different stages in
the analytical protocol.
GC/MS Log
Each sample run by GC/MS is logged into a notebook, detailing
analysis conditions, where the data are archived, and what hardcopy data
has been produced.
7.2.2 External Quality Assurance
Samples will be analyzed by a designated quality assurance laboratory
Samples, controls and blanks will be shipped directly from the field to
the laboratory for analysis. They will report the result to the primary
laboratory for correlation with primary data.
7.2.2.1 Selection of Samples for Quality Assurance
Approximately 10% of the field samples (2 minimum) will be collected
in duplicate for shipment to the quality assurance laboratory. The
selection process will be random with the following restrictions: If
any stratefication of sites is known, purposive, selection of quality
assurance sites may be used to get representative samples (up river vs
down river sites).
7.2.2.2 QC Samples
Along with the QA duplicate samples, at least one QC control and
one QC blank must be included with the QA samples. An example is shown
in Table 3 for a trip collecting 50 samples.
7.2.2.3 Sample Codes
All samples shipped to the QA laboratory should be encoded so that
the laboratory cannot discern between samples and controls. This "blind"
protocol will add validity to the results.
7.2.2.4 Shipping
Samples should be shipped on dry ice directly to the QA laboratory
by an appropriate air carrier (£•£•, Federal Express, Eastern Sprint) in
well insulated and packed cartons.
133
-------
Table 3. SAMPLES TO BE COLLECTED AND SHIPPED TO QA LABORATORY
FOR A TYPICAL SAMPLING TRIP
Sample Number Comment
Duplicate Sample 5 Random selection unless
prior information stratifies
sites
Field Blank 1 Ship with samples
Field Control 1 Ship with samples
134
-------
ANALYTICAL PROTOCOL: POLYNUCLEAR AROMATIC HYDROCARBONS IN AIR
1.0 Principle of the Method
Air is drawn through a sampling train consisting of a filter followed
by a polyurethane foam (PUT) plug. Only particulate material trapped by the
filter is analyzed for PNAs. The filter is extracted with cyclohexane. The
resulting solution is fractionated by chromatography on a short silica gel
bed, followed by thin layer chromatography on cellulose. The solution
resulting from this clean-up is analyzed by liquid chromatography with
fluorescence detection; if sufficient sensitivity and/or resolution is not
attained, the solution will be analyzed by capillary gas chromatography with
flame ionization detection (GC-FID). This methodology is based on procedures
developed at NIOSH (Cincinnati), Division of Physical Sciences and Engineering,
Wesssrements Research Branch, Organic Methods Development Section.
2~.0 Range and Detection Limit
For this procedure the lower limit of measurement for all target PNAs
is approximately 100 ng per sample. The concentration range embraced by the
procedure was not specified in the NIOSH protocol.
3.0 Interferences
Unspecified in NIOSH protocol.
4.0 Precision and Accuracy
Unspecified in NIOSH protocol.
5.0 Apparatus and Reagents
5.1 Sampling
5.1.1 Filters - Glass filter, 25 mm diameter (Gelman Type A/E);
Teflon, 25 mm diameter (Gelman Type TF).
5.1.2 Filter holder - Gelman, open, Delrin; Beckman union reducer,
Cat. No. 830517.
5.1.3 Personal sampling pumps capable of sampling at 2 L/min with the
sampling train in line (MSA Model G or equivalent). The pumps must be
calibrated with a representative sampling train in line.
5.2 Analysis
5.2.1 High pressure liquid chromatograph, automated (Two Waters M6000A
pumps, a Waters 660 solvent programmer, and a Waters WISP 710 auto sampler).
135
-------
5.2.2 LC column, 25 cm x 6.4 mm inside diamter, packed with Vydac
201TP reverse phase, 10 pm particle size, or equivalent.
5.2.3 Constant temperature water jacket for LC column.
5.2.4 Fluorescence detector, monitoring emission at 425 nm with excita-
tion at 340 nm, 300 [JL flow cell (Farrand Mark I spectrophotofluorometer), or
equivalent.
5.2.5 Laboratory data system (Hewlett-Packard 3354B), or equivalent.
6.2.6 Scintillation vials, 20 mL, with polyethylene-lined screw caps.
5.2.7 Ultrasonic water bath (Cole-Palmer Model 8845-60), or equivalent.
5.2.8 TLC plates, 20 cm x 30 cm, precoated with MN 300 cellulose
normal, layer thickness 250 pm (Analtech).
5.2.9 TLC tank, 27 cm x 21.5 cm x 7.5 cm, with glass-plate cover and
filter paper lining one large side.
5.2.10 Polyperfluoroethylene filter, 25 mm diameter, unlaminated,
0.5 pm pore size (Millipore fluoro pore #FHUP 025-00).
5.2.11 Holder, 25 mm filter, glass (Millipore //XX-10-025-00).
5.2.12 Cartridges, silica gel, 2 cm x 1 cm diameter (Waters SEP-PAK).
5.2.13 Syringe, 10 mL, glass.
5.2.14 Vials, 5 mL, conical-cavitied, with screw caps and Teflon-lined
septa.
5.2.15 Vial heater, block type (Supelco Blok Heater), or equivalent.
5.2.16 Sample concentrator, 6-port, positive-gas-flow type.
5.2.17 Pipets, Eppindorf, 10- and 25-|jL.
5.2.18 UV light box.
5.2.19 TLC plate zone collectors, made in-house as follows: The tip of
a Pasteur pipet is removed to give a glass tube 8 cm long x 7 mm in diameter.
A quarter of a 25 mm polyperfluoroethylene filter (Item I.A.10) is wrapped
over and around the smooth end of the tube, which is then gently inserted
into flexible tubing connected to house vacuum.
5.2.20 Pipets, 1-, 2-, 5-, 9- and 10 mL.
5.2.21 Flasks, volumetric, 10- and 100 mL, low-actinic, tinted red.
5.2.22 Culture tubes, 13 mm x 100 mm, round-bottomed, with screw caps.
5.2.23 Centrifuge.
136
-------
5.2.24 Centrifuge tubes, tapered-bottom, 15 mL, graduated with 0.1 mL
subdivisions.
5.2.25 Solvents (Burdick and Jackson) - acetonitrile, methanol, cyclo-
hexane, methylene chloride, 2-propanol, acetone, water (deionized, distilled)
5.2.26 Mobile phases.
Solvent A: 48.4:43.8:7.8 Acetonitrile-water-methanol by
volume
Solvent B: 61.5:38.5 methanol-acetonitrile by volume
Solvent C: 2:2:1 water-2-propanol-acetone by volume
Solvent D: 9:1 hexane-methylene chloride by volume
5.2.27 Reference standards - fluoranthene, pyrene, benz(a)anthracene,
chrysene, benzo(a)pyrene, benzo(e)pyrene.
6.0 Procedure
6.1 Sampling
Immediately before sampling, connect the sampling train to a piece of
flexible tubing connected to a personal sampling pump. Position the sampling
device vertically with the face of the filter cassette down. This will
prevent channeling of the sorbent beds. Upon completion of the sampling,
transfer the particulate filter to a clean glass container, covered with
foil and sealed with Teflon-lined screw-cap vials. Note sampling flow,
length of sampling period, and ambient temperature and pressure. Store the
samples in a refrigerator and protect from light until they are analyzed.
6.2 Analysis
Prior to use for sample fractionation the TLC plates must be conditioned
by exposure to solvent C. Develop a plate with solvent C. In about 3 hours
the solvent front will be within 1 cm of the top of the plate.
Transfer the filter to a 20 mL scintillation vial. Add 5 mL of cyclo-
hexane. After insuring that the filter is covered by the solvent, seal the
vial and agitate it in an ultrasonic water bath for 1 hour. Filter the
sample solution through a 0.5 |jm pore size polyperfluoroethylene filter
using positive nitrogen pressure. Collect the filtrate in a clean 20 mL
scintillation vial. About 4 ml of the solution is recovered.
Transfer a 3 mL aliquot of the sample solution to the barrel of a 10 mL
syringe fitted with a silica gel cartridge. Insert the plunger and force
137
-------
the solution through the silica gel bed over a 10-s period. Discard the
eluate. In order to prevent disturbance of the silica gel bed, before
removing the plunger for the addition of solvent D, remove the silica gel
bed. Add 3 ml of solvent D to the syringe, and force it through the silica
gel bed, collecting the eluate in a 5 ml vial. Add an additional 2 ml of
solvent D to the syringe and force it through the silica gel bed, combining
the eluate with the previous 3 mL.
With the water bath set around 60°C, concentrate the sample solution to
about 2 mL under a gentle stream of nitrogen. Add 1 ml of acetonitrile and
continue the concentration until the volume has been reduced to about 0.1
mL. In order to minimize sample loss, DO NOT ALLOW RESIDUE TO FORM ON THE
SIDES OF THE VIAL OR ALLOW THE SAMPLE TO GO TO DRYNESS. Remove the developed
blank TLC plate from the developing chamber and allow it to dry approximately
20 min. Spot the sample in 10 (jL or smaller aliquots at the origin, centered
about 4 cm from the bottom of the plate. Rinse the vial two times with 25
|JL portions of acetonitrile, each time spotting the washing with the sample.
Five samples can be run on a plate. Include in these samples one blank and
one control sample of sufficient concentration that the analyte will be
visible when viewed under short-wave UV light.
At 30 minutes past the time the blank TLC plate was removed from the
developing chamber, return the spotted plate to the chamber and develop the
chromatogram with solvent C. If the TLC plate becomes too dry prior to
development of the chromatogram, the polycyclic aromatics will move with the
solvent front. After removing the developed TLC plate from the tank, visua-
lize the spots of the chromatograms by exposing the plate to short-wave UV
light. There is no need to dry the TLC plate. Using the control sample as
a reference, mark spots or areas corresponding to the analytes for each of
the five samples. The polycyclic aromatics usually are located between Rf
0.62 and R 0.81.
Using the zone collector described in Section 5.2.19, remove the cellu-
lose containing the polynuclear aromatic fraction of the samples from the
TLC plate. Transfer the cellulose and the filter and glass tube of the spot
collector to a culture tube. Add 2 mL of acetonitrile. After insuring that
138
-------
all of the cellulose is covered with acetonitrile, agitate the test tube in
an ultrasonic bath for 1 hour.
Centrifuge with sample. Transfer a 200 pL aliquot of the supernatant
solution to an auto sampler vial for LC analysis.
Set up the LC system for the following conditions:
Mobile phase flow 1.0 L/min
Mobile phase solvents A and B
Mobile phase program
injection to 37 min linear increase in solvent B from
30% to 100%
37 min to 55 min 100% solvent B
55 min to 75 min 30% solvent B
Column temperature 28°C
Under these conditions the retention times were:
fluoranthene 12.6 min
pyrene 14.0 min
benz(a)anthracene 20.1 min
chrysene 22.1 min
benzo(e)pyrene 25.5 min
benzo(a)pyrene 32.2 min
Inject a 10 [iL aliquot of the sample into the LC system and begin the
solvent program. Use of an auto sampler to run a series of samples and
standards at regular intervals maximizes the reproducibility of the chroma-
tography. Determine the area of the analyte chromatographic peaks using the
laboratory data system. Use the detector monitoring emission >370 run for
the chrysene measurement; use the other detector for obtaining measurements
of the other five analytes. Check to see that the level of each analyte
falls within the range of the standard curve.
If the levels of one or more components is above the range of the
standard curve, dilute an aliquot to an appropriate concentration and rerun
the sample as above. If the levels of one or more components is below the
range of the standard curve, continue with the procedure below. Recombine
the remainder of the 200 pL aliquot in the auto sampler vial with the bulk
of the sample in the culture tube. Filter the sample under positive nitrogen
139
-------
pressure through a polyperfluoroethylene filter, collecting the filtrate in
a graduated centrifuge tube. Rinse the residue with four 2 mL portions of
acetonitrile, adding the rinsings to the first filtrate. Using the vial
heater set around 120°C, reduce the volume of the solution to below 0.5 mL
under a gentle stream of nitrogen. Cool the sample to room temperature.
Adjust the sample volume to 0.5 ml with acetonitrile. Inject a 125 |JL
aliquot into the LC system and begin the solvent program. Analyze the
samples under the same chromatographic conditions as described above.
6.3 Calibration and Standards
Prepare stock solutions of each polycyclic aromatic in acetonitrile at
the following concentrations:
fluoroanthrene 1 mg/mL
pyrene 1 mg/mL
benz(a)anthracene 1 mg/mL
chrysene 0.5 mg/mL
benz(e)pyrene 1 mg/mL
benzo(a)pyrene 0.5 mg/mL
Prepare a combined stock solution at the concentration 0.1 mg/mL by
transferring 2 mL aliquots of the chrysene and benzo(a)pyrene stock solutions
and 1 mL aliquots of the others to a 10 mL volumetric flask and diluting to
the mark with acetonitrile. This solution may be stored in a refrigerator
if transferred to a 20 mL scintillation vial, which is capped and wrapped in
aluminum foil. Using the combined stock solution, prepare five or six
standards covering the range 1-25 M8/mL- Use these standards for the analysis
of the dilute samples. With the same combined stock solution, prepare five
or six standards covering the range 0.1-5 (Jg/mL. Use these standards for
the analysis of the concentrated samples.
Analyze the standards with the samples. Construct calibration curves
of peak area plotted against concentration for each analyte. Prepare separate
curves for the analysis of dilute and concentrated samples.
6.4 Calculations
Read the concentration c (pg/mL) of the analyte in the solution from
the appropriate calibration curve. Calculate the amount w (pg) of analyte
in the sample by the equation:
140
-------
c x v x F
w = —676—
where
v = volume of solution containing the sample workup (either
2 ml or 0.5 ml)
F = dilution factor, if sample diluted in step III-C.14
0.6 = fraction of total sample submitted to workup and analysis
Subtract from w contributions of the blank to give W.
Q
Calculate the air concentration C (mg/m ) of the analyte as follows:
C - ^
L - V
where
V = volume (L) of air sampled
7.0 Quality Assurance Program
7.1 Quality Control
Laboratory and field blanks are prepared and analyzed as a check on
background contamination. Two of each will be analyzed. The filters for
the blanks will be obtained from the same lot as the filters used for field
samples. One duplicate sample will be taken and analyzed as a means of
checking method reproducibility.
7.2 Quality Assurance
Of the eleven samples to be collected and analyzed, two will be shipped
to the QA laboratory (NIOSH, Cincinnati) for analysis, and comparison of
results with those obtained at the primary lab.
141
-------
ANALYTICAL PROTOCOL: ORGANOCHLORINE PESTICIDES AND PCBs IN DRINKING WATER
1.0 Principal of Method
This method covers the determination of certain organochlorine pesticides
and polychlorinated biphenyls (PCBs) found in drinking water. The following
compounds may be determined by this method:
Aldrin Endrin
a-BHC Endrin Aldehyde
b-BHC Heptachlor
d-BHC Heptachlor Epoxide
g-BHC Toxaphene
Chlordane PCB-1016
4,4'-ODD PCB-1221
4,4'-DDE PCB-1232
4,4'-DDT PCB-1242
Dieldrin PCB-1248
Endosulfan I PCB-1254
Endosulfan II PCB-1260
Endosulfan Sulfate
A 1-liter sample of wastewater is extracted with methylene chloride
using separatory funnel techniques. The extract is dried and concentrated to
a volume of 10 mL or less. Chromatographic conditions are described which
allow for the accurate measurement of the compounds in the extract. If
interferences are encountered, the method provides selected general purpose
cleanup procedures to aid the analyst in their elimination.
This method is recommended for use only by experienced residue analysts
or under the close supervision of such qualified persons.
2.0 Range and Detection Limit
The sensitivity of this method is usually dependent upon the level of
interferences rather than instrumental limitations. The limits of detection
listed in Table 1 represent sensitivities that can be achieved in wastewaters.
3.0 Interferences
Solvents, reagents, glassware, and other sample processing hardware may
yield discrete artifacts and/or elevated baselines causing misinterpretation
142
-------
Table 1. GAS CHROMATOGRAPHY OF PESTICIDES AND PCB's
Parameter
Retention time (min)
Column 1 Column 2
* Multiple peak response. See Figures 2-10.
Detection limit
(yg/L)**
Aldrln
a-BHC
b-BHC
d-BHC
g-BHC
Chlordane
4, 4 '-ODD
4,4'-DDE
4, 4'- DDT
Dieldrin
Endosulfan I
Endosulfan II
Endosulfan sulfate
Endrin
Endrin aldehyde
Heptachlor
Heptachlor epoxide
Toxaphene
PCB-1016
PCS- 1221
PCB-1232
PCB-1242
PCB-1248
PCB-1254
PCB-1260
2.40
1.35
1.90
2.15
1.70
*
7.83
5.13
9.40
5.45
4.50
8.00
14.22
6.55
11.82
2.00
3.50
*
*
*
*
*
*
*
*
4.10
1.82
1.97
2.20
2.13
*
9.08
7.15
11.75
7.23
6.20
8.28
10.70
8.10
9.30
3.35
5.00
*
*
*
*
*
*
*
*
0.003
0.002
0.004
0.004
0.002
0.04
0.012
0.006
0.016
0.006
0.005
0.01
0.03
0.009
0.023
0.002
0.004
0.40
0.04
0.10
0.10
0.05
0.08
0.08
0.15
**
Detection limit is calculated from the minimum detectable GC response
being equal to five times the GC background noise, assuming a 10 ml
final volume of the 1 liter sample extract, and assuming a GC injection
of 5 micro liters.
Column 1 conditions: Supelcoport 100/120 mesh coated with 1.5X
SP-2250/1.95% SP-2401 packed in a 180 cm long x 4 mm ID glass column
with 5X Methane/95% Argon carrier gas at 60 ml/min flow rate. Column
temperature is 200°C.
Column 2 conditions: Supelcoport 100/120 mesh coated with 3% OV-1 in a 180
cm long x 4 mm ID glass column with 5X Methane/95% Argon carrier gas at
60 ml/min flow rate. Column temperature is 200°C.
143
-------
of gas chromatograms. All of these materials must be demonstrated to be free
blanks. Specific selection of reagents and purification of solvents by
distillation in all-glass systems may be required. Interferences coextracted
from the samples will vary considerably from source to source, depending upon
the source of the water being sampled. While general cleanup techniques are
provided as part of this method, unique samples may require additional cleanup
approaches to achieve the sensitivities stated in Table 1.
Glassware must be scrupulously clean. Clean as soon as possible after
use by rinsing with the last solvent used. This should be followed by detergent
washing in hot water. Rinse with tap water, distilled water, acetone and
finally pesticide quality hexane. Heavily contaminated glassware may require
treatment in a muffle furnace at 400°C for 15 to 30 minutes. Some high
boiling materials, such as PCBs, may not be eliminated by this treatment.
Volumetric ware should not be heated in a muffle furnace. Glassware should
be stored immediately after drying or cooling to prevent any accumulation of
dust or other contaminants. Store inverted or capped with aluminum foil.
Interferences by phthalate esters can pose a problem in the 15% and 50%
fractions from the Florisil fractionation. These interferences can be minimized
by avoiding contact with any plastic materials. The presence of phthalate
esters is indicated in samples that respond to electron capture detectors but
not to microcoulometric or electrolytic conductivity (halogen mode) detectors.
4.0 Precision and Accuracy
The results of studies delineating the precision and accuracy of this
method have not yet been determined. The sensitivity of this method with the
EPA (HERL-RTP) protocol for response in river water, which is similar to the
protocol outlined here, indicates acceptable levels of precision for recovery
of the chlorinated pesticides and PCBs would be obtained.
5.0 Apparatus and Reagents
5.1 Sampling Equipment (for discrete or composite sampling)
5.1.1 Grab sample bottle - amber glass, liter or quart volume.
French or Boston Round design is recommended. The container
must be washed and solvent rinsed before use to minimize
interferences.
144
-------
5.1.2 Bottle Caps - Threaded to screw on sample bottles. Caps
must be lined with Teflon. Foil may be substituted if
sample is not corrosive.
5.1.3 Compositing equipment - Automatic or manual compositing
system. Must incorporate glass sample containers for the
collection of a minimum of 250 ml. Sample containers must
be kept refrigerated during sampling. No Tygon or rubber
tubing or fittings may be used in the system.
5.2 Separatory funnel - 2000 mL, with Teflon stopcock.
5.3 Drying column - A 20 mm ID pyrex chromatographic column with coarse
frit.
5.4 Kuderna-Danish (K-D) Apparatus
5.4.1 Concentrator tube - 10 ml, graduated (Kontes K-570050-1025
or equivalent). Calibrations must be checked at 1.0 and
10.0 mL level. Ground glass stopper (size 19/22 joint) is
used to prevent evaporation of extracts.
5.4.2 Evaporative flask - 500 mL (Kontes K-57001-0500 or equivalent).
Attach to concentrator tube with springs. (Kontes K-662750-012)
5.4.3 Snyder column - three-ball macro (Kontes K503000-0121 or
equivalent).
5.4.4 Boiling chips-extracted, approximately 10/40 mesh.
5.5 Water Bath - Heated, with concentric ring cover, capable of temper-
ature control (+ 2°C). The bath should be used in a hood.
5.6 Gas chromatograph - Analytical system complete with gas chromatograph
suitable for on-solumn injection and all required accessories
including electron capture or halogen-specific detector, column
supplies, recorder, gases, syringes. A data system for measuring
peak areas is recommended.
5.7 Chromatographic column - Pyrex, 400 mm x 25 mm OD, with coarse
fritted plate and Teflon stopcock (Kontes K-42054-213 or equivalent).
5.8 Preservatives:
5.8.1 Sodium hydroxide - (ACS) 10 N in distilled water.
5.8.2 Sulfuric acid - (ACS) Mix equal volumes of/ cone.
H-SO, with distilled water.
145
-------
5.9 Methylene chloride - Pesticides quality or equivalent.
5.10 Sodium Sulfate - (ACS) Granular, anhydrous (purified by heating at
400°C for 4 hrs.) especially just prior to preparing working stan-
dards from them.
5.11 Boiling chips - Hengar granules (Hengar Co.; Fisher Co.) or equiva-
lent.
5.12 Mercury - triple distilled.
5.13 Aluminum oxide - basic or neutral, active.
5.14 Hexane - pesticide residue analysis grade.
5.15 Isooctane (2,2,4-trimethyl pentane) - pesticide residue analysis
grade.
5.16 Acetone - pesticide residue analysis grade.
5.17 Diethyl ether - preserved with 2% ethanol.
5.17.1 Must be free of peroxides as indicated by EM Quant test
strips (EM Laboratories, Inc., 500 Executive Blvd.,
Elmsford, N.Y. 10523).
5.17.2 If test indicates, remove peroxides by eluting over basic
or neutral grade aluminum oxide. Retest before using.
5.18 Florisil - PR grade (60/100 mesh); purchase activated at 1250°F
and store in glass containers with glass stoppers or foil-lined
screw caps. Before use activate each batch at least 16 hours at
130°C in a foil covered glass container.
5.19 Stock standards - Prepare stock standard solutions at a concentra-
tion of 1.00 (Jg/(Jl by dissolving 0.100 grams of assayed reference
material in pesticide quality isooctane or other appropriate
solvent and diluting to volume in a 100 ml ground glass stoppered
volumetric flask. The stock solution is transferred to ground
glass stoppered reagent bottles, stored in a refrigerator, and
checked frequently for signs of degradation or evaporation.
6.0 Procedure
6.1 Sample Collection
Grap samples must be collected in glass containers. Conventional
sampling pratcies should be followed, except that the bottle must not be
prewashed with sample before collection. Composite samples should be
146
-------
collected in refrigerated glass containers in accordance with the require-
ments of the program. Automatic sampling equipment must be free of Tygon and
other potential sources of contamination. The samples must be iced or refrig-
erated from the time of collection until extraction. Chemical preservatives
should not be used in the field unless more than 24 hours will elapse before
delivery to the laboratory. If the samples will not be extracted within 48
hours of collection, the sample should be adjusted to a pH range of 6.0-8.0
with sodium hydroxide or sulfuric acid. All samples must be extracted within
7 days and completely analyzed within 30 days of collection.
6.2 Sample Extraction
Mark the water meniscus on the side of the sample bottle for later
determination of sample volume. Pour the entire sample into a two-liter
separatory funnel. Check the pH with wide-range paper and adjust to within
the range of 5-9 with sodium hydroxide or sulfuric acid. Add 60 mL methylene
chloride to the sample bottle and shake 30 seconds to rinse the walls.
Transfer the solvent into the separatory funnel, and extract the sample by
shaking the funnel for two minutes with periodic venting to release vapor
pressure. Allow the organic layer to separate from the water phase for a
minimum of ten minutes. If the emulsion interface between layers is more
than one-third the size of the solvent layer, the analyst must employ mechanical
techniques to complete the phase separation. The optimum technique depends
upon the sample, but may include stirring, filtration of the emulsion through
glass wool, or centrifugation. Collect the methylene chloride extract in a
250-mL Ehrlenmeyer flask.
Add a second 60-mL volume of methylene chloride to the sample bottle and
complete the extraction procedure a second time, combining the extracts in
the Ehrlenmeyer flask.
Perform a third extraction in the same manner. Pour the combined
extract through a drying column containing 3-4 inches of anhydrous sodium
sulfate, and collect it in a 500-mL Kuderna-Danish (K-D) flask equipped with
a 10 mL concentrator tube. Rinse the Ehrlenmeyer flask and column with 20-30
mL raethylene chloride to complete the quantitative transfer.
147
-------
Add 1-2 clean boiling chips to the flask and attach a three-ball
Snyder column. Prewet the Snyder column by adding about 1 mL methylene
chloride to the top. Place the K-D apparatus on a steaming hot (60-
65°C) water bath so that the concentrator tube is partially immersed in
the hot water, and the entire lower rounded surface of the flask is
bathed in steam. Adjust the vertical position of the apparatus and the
water temperature as required to complete the concentration in 15-20
minutes. At the proper rate of distillation the balls of the column
will actively chatter but the chambers will not flood. When the apparent
volume of liquid reaches 1 ml, remove the K-D apparatus and allow it to
drain for at least 10 minutes while cooling.
Increase the temperature of the hot water bath to about 80°C.
Momentarily remove the Snyder column, add 50 ml of hexane and a new
boiling chip and reattach the Snyder column. Pour about 1 mL of hexane
into the top of the Snyder column and concentrate the solvent extract as
before. Elapsed time of concentration should be 5 to 10 minutes. When
the apparent volume of liquid reaches 1 mL, remove the K-D apparatus and
allow it to drain at least 10 minutes while cooling. Remove the Snyder
column and rinse the flask and its lower joint into the concentrator
tube with 1-2 mL of hexane, and adjust the volume to 10 mL. A 5-mL
syringe is recommended for this operation. Stopper the concentrator
tube and store refrigerated if further processing will not be performed
immediately. If the sample extract requires no further cleanup, proceed
with gas chromatographic analysis. If the sample requires cleanup,
proceed to 6.3
Determine the original sample volume by refilling the sample bottle
to the mark and transferring the liquid to a 1000 mL graduated cylinder.
Record the sample volume to the nearest 5 mL.
6.3 Cleanup and Separation
Cleanup procedures are used to extend the sensitivity of a method
by minimizing or eliminating interferences that mask or otherwise dis-
figure the gas chromatographic response to the pesticides and PCBs.
The Florisil column allows for a select fracionation of the compounds
148
-------
and will eliminate polar materials. Elemental sulfur interferes with the
electron capture gas chromatography of certain pesticides and can be
removed by the techniques described below.
6.3.1 Florisil Column Cleanup
Add a weight of Florisil, nominally 21 g, but predetermined by
calibration, to a chromatographic column. Settle the Florisil by tapping
the column. Add sodium sulfate to the top of the Florisil to form a
layer 1-2 cm deep. Add 60 ml of hexane to wet and rinse the sodium
sulfate and Florisil. Just prior to exposure of the sodium sulfate to
air, stop the draining of the hexane by closing the stopcock on the
chromatography column. Discard the eluate.
Add the sample extract from the K-D concentrator tube to the Florisil
column. Rinse the tube twice with 1-2 ml hexane, adding each rinse to
the column. Place a 500 ml K-D flask and clean concentrator tube under
the chromatography column. Drain the column into the flask until the
sodium sulfate layer is nearly exposed. Elute the column with 200 mL of
6% ethyl ether in hexane (Fraction 1) using a drip rate of about 5
mL/min. Remove the K-D flask and set aside for later concentration.
Elute the column again, using 200 mL of 15% ethyl ether in hexane (Fraction
2), into a second K-D flask. Perform the third elution using 200 mL of
50% ethyl ether in hexane (Fraction 3). The elution patterns for the
pesticides and PCBs are shown in Table 2.
Concentrate the eluates by standard K-D techniques (6.2), substitu-
ting hexane for methylene chloride and using the water bath at about
85°C. Adjust final volume to 10 mL with hexane. Analyze by gas chroma-
tography.
Elemental sulfur will usually elute entirely in Fraction 1. To
remove sulfur interference from this fraction or the original extract,
pipet 1.00 mL of the concentrated extract into a clean concentrator tube
or Teflon-sealed vial. Add 1-3 drops of mercury and seal. Agitate the
contents of the vial for 15-30 seconds. Place the vial in an upright
position on a reciprocal laboratory shaker and shake for 2 hours.
Analyze by gas chroraatography.
149
-------
Table 2. DISTRIBUTION AND RECOVERY OF CHLORINATED PESTICIDES
AND PCBs USING FLOROSIL COLUMN CHROMATOGRAPHY
Recovery (%) by fraction*
Parameter 1(6%) 2(15%) 3(50%)
Aldrin
a-BHC
b-BHC
d-BHC
g-BHC
Chlordane
4, 4 '-ODD
4.4--DDE
4, 4 '-DDT
Dieldrin
Endosulfan I
Endosulfan II
Endosulfan sulfate
Endrin
Endrin aldehyde
Heptachlor
Heptachlor epoxide
Toxaphene
PCB-1016
PCB-1221
PCB-1232
PCB-1242
PCB-1248
PCB-1254
PCB-1260
100
100
97
98
100
100
99
98
100
0
37
0
0
4
0
100
100
96
97
97
95
97
103
90
95
100
64
7
0
96
68
4
91
106
26
*From: "Development and Application of Text Procedures for Specific Organic
Toxic Substances in Wastewaters. Category 10-Pesticides and PCB's.
Report for EPA Contract 68-03-2606.
150
-------
6.4 Analysis
6.4.1 Gas Chromatography
Table 1 summarizes some recommended gas chromatographic column
materials and operating conditions for the instrument. Included in this
table are estimated retention times and sensitivites that should be
achieved by this method. Examples of the separations achieved by these
columns are shown in Figures 1 through 10. Calibrate the system daily
with a minimum of three injections of calibration standards.
Inject 2-5 |JL of the sample extract using the solvent-flush technique.
Smaller (1.0 |jL) volumes can be injected if automatic devices are employed.
Record the volume injected to the nearest 0.05 |JL, and the resulting
peak size, in area units. If the peak area exceeds the linear range of
the system, dilute the extract and reanalyze. If the peak area measure-
ment is prevented by the presence of interferences, further cleanup is
required.
6.4.2 Calibration
Prepare calibration standards that contain the compounds of interest,
either singly or mixed together. The standards should be prepared at
concentrations covering two or more orders of magnitude that will com-
pletely bracket the working range of the chromatographic system. If the
sensitivity of the detection system can be calculated from Table 1, as
100 |Jg/L in the final extract for example, prepare standards at 10 M8/L,
50 |Jg/L, 100 pg/L, 500 pg/L, etc., so that injections of 1-5 pL of each
calibration standard will define the linearity of the detector in the working
range.
Assemble the necessary gas chromatographic apparatus and establish
operating parameters equivalent to those indicated in Table 1. By
injecting calibration standards, establish the sensitivity limit of the
detector and the linear range of the analytical system for each compound.
Before using any cleanup procedure, the analyst must process a
series of calibration standards through the system to validate elution
patterns and the absence of interferences from the reagents.
151
-------
Figure 1. EC Gas Chromatography of Organochlorlne Pesticides on Column 1
For Conditions, See Table I.
152
-------
Figure 2. EC Gas Chromatography of Chlordane on Column 1
For Conditions, See Table I.
153
-------
Figure 3. EC Gas Chromatography of Toxaphene on Column 1
cFor Conditions, See Table I.
154
-------
-:«J— {- •-• ••'
7H1-H---T
r ;rs H-
• —!»n j
•"trr
"ilcr
' it: I
'Ht:"j-|"r 2 ir:TT '-.r^^y*
frjjl;"'"' ......^.
.: 1
'r- -'-
J.r
•1 :
! i | j'Ui ; | !
TSflS
J-L^j. ";-> i i
-
. i >.. :
-•_..1.._'
4| N
!.,{..;
TTT
_. _ j»
SLa.
d|
1H
—48 . ,',1'to. • •""
* l4-i±'
By:.;|;
-•It. '_:_I.~
II • ;• • • T i"fi < i * ,
toti^lpT.
OTrfnf
,»••> - - i •• J ;•
!. j. •_ i » i .. * -i.
r^lr"; i :Sl
• t f •
;UrT: .IV I.':.!.
T
•|-7 •
_-r
r™
zl-rr^:
~i*"~""" *
-
'
Miwjtis-it"i4:~i Jn~i1:~H
Figure 4. EC Gas Chromatography of PCB 1016 on Column 1.
conditions, see Table 1.
For
155
-------
' •
•• 1"
CJ-
.
!VS-
•\ -
_; ...
.Ul... L.
:-CET.
44Tf
xti:
•rr
*
.-:^z::LqL.
-— -+H-11-1
...
.._.__
m
i
..T
t
•|.j :
LL.J:rLi
JL^.-MtNUTESii:
Figure 5. EC Gas Chromatography of PCS 1221 on Column 1 For
conditions, see Table 1.
156
-------
!.
•-i-
-'•hr—3-
.-+-- —
•=nzr
-..:.: 1 i-.L
-r-rrj—.I'-T
~-T—H r—
"t ;""" ~ ' " .
^ i
...J...
T't
-I—
.~j'~~ """ "™"i!"_'' ~iT2i'"j-:":3ir.j.'.~_r.."iTir ..".~r_!
»
••I-
I
- \-
i
f-- --I
V*- —
m
"!
:l
T~~r
-l i-H
;.ti>_
: I •
"T T"."
— j.
J5f^±i^F
..-j-^^zr^:
_'p.i __. •
_4._i^..,_.2 r ••:{??--— .-i
4:1 i dt il.: J $ j. -:±:~^v::iSr_::i:
qrJi^lJt^^^
:."». "t r^.rjUi;vlil|r^p- j|-"|; ^£j"-:-'~—
"! : -\- '!i- ill*;'1- T-lM' J •" 'i-^!" ~-::
--T
S^fe1
!"—; i - -
.._..:...j.:: -:•:[: ::
~i : • •
iV4|feX^
Figure 6. EC Gas Chromatography of PCB 1232 on Column 1.
conditions, see Table 1.
For
157
-------
^-.—^ .... .._.
v-i -ri-H :;'; !
-
4Ji:.'4^-"""-p:r
J—/1 — .:.:.-!...:..•_... !._-_+
—.j. I ....'.I.'..'.. ... T .;
£^
Figure 7. EC Gas Chromatography of PCB -1242 on Column 1.
For Conditions, See Table I.
158
-------
4t
" % - •—*
.. >....
i
• t
.....
•3.
..L: .....
i • •
..4.: :..;j.
-tl;: -:•»•-.-:-r • -.-_=•
1
1
".. :l:L-b:
v:: i. •..•.j.'i.sivi;.-'..:.:M...~'{'^r.1:.
: ..._: • ;...
:-r
-vv
1:
.... i
si^
~_j.t
• ^~f ^^^^^3^^r^\~
•__• ,._\_ .'__'.._. _ «.!."« —. . * . .__..!—_..-- I '
:l-i
_2 i"
h J . .C. '• . O . i . Yh
* ... J... y . ...' i j..:. !*.- ...j..i.!-.*
•: :jVr>iiNutES j v.:-|r:*::.|;:! _;fc:.cl
Figure 8. EC Gas Chromatography of PCS 1248 on Column 1. For
conditions, see Table 1.
159
-------
I LJ
Figure 9. EC Gas Chromatography of PCB-1254 on Column 1
For Conditions, See Table I.
160
-------
Figure 10. EC Gas Chromatography of PCB - 1260 on Column 1
For Conditions, See Table I.
161
-------
6.4.3 Calculations
Determine the concentration of the individual compounds according
to the formula:
(A) (B) (Vt)
Concentration, (Jg/L =
(V.) (Vs)
where A = Calibration factor for chromatographic system, in nanograms
material per area unit.
B = Peak size in injection of sample extract, in area units
V. = Volume of extract injected ((JL)
V = Volume of total extract (|JL)
V = Volume of water extracted (ml)
s
Report results in micrograms per liter without correction for recovery
data. When duplicate and spiked samples are analyzed, all data obtained
should be reported.
7.0 Quality Control
Before processing any samples, the analyst should demonstrate through
the analysis of a distilled water method blank, that all glassware and
reagents are interference-free. Each time a set of samples is extracted or
there is a change in reagents, a method blank should be processed as a
safeguard against chronic laboratory contamination.
Standard quality assurance practices should be used with this method.
Field replicates should be collected to validate the accuracy of the analysis
Where doubt exists over the identification of a peak on the chromatogram,
confirmatory techniques such as mass spectroscopy should be used.
8.0 References
8-1 "Development and Application of Test Procedures for Specific Organic
Toxic Substances in Wastewaters. Category 10-Pesticides and PCBs."
Report for EPA Contract 68-03-2606.
162
-------
ANALYTICAL PROTOCOL: DETERMINATION OF ORGANOCHLORIDE PESTICIDES AND
METABOLITES IN DRINKING WATER ( U. OF MIAMI)
1.0 Principle of the Method
Drinking water is extracted with methylene chloride and the extract is
concentrated. The concentrate is eluted through deactivated silica gel first
with hexane as the solvent and then with benzene/hexane, 3:2. The two fractions
are analyzed by GC/ECD for twelve pesticides and metabolites: hexachloroben-
zene, p-hexachlorocyclohexane ({J-BNC), heptachlor, oxychlordane, heptachlor
epoxide, trans-nonachlor, j>,p'-DDE, dieldrin, o,£-DDT, £,p'-DDT, a-chlordane
and 6-chlordane.
2.0 Range and Detection Limit
Although the linear dynamic range of an ECD is not very large the quanti-
tative range of detection can be greatly expanded by successive dilution of
the extracts. Theoretically this can make the range virtually infinite. The
detection limits for the pesticides sought in this study are listed in
Table 1.
3.0 Interferences (1)
Interferences in sample analysis and quantification using GC/ECD are
manifested in the electron capturing ability of the given contaminant. The
relative purity of the water matrices (i_.e., residential drinking water)
minimizes large interferences.
4.0 Precision and Accuracy
No data applicable to precision measurement was not presented in this
study. Recovery studies performed in duplicate on the pesticides of interest
is presented in Table 2. The blank control for this study showed none of the
compounds of interest in concentrations exceeding the limit of detection.
5.0 Apparatus and Readgents
5.1 Apparatus
3
1. A Tracer Model 220 gas chromatograph equipped with a H electron
capture detector.
2. Column: 6 ft x 6.35 mm I.D. glass U-tube packed with 1.5% OV-17
+ 1.95% QF-1 on 100/120 Chromosorb W.
163
-------
Table 1. PERCENT RECOVERY, DETECTOR SENSITIVITY AND LIMITS
OF DETECTABILITY OF PESTICIDE AND METABOLITES IN WATER
Compound
Detector
Sensitivity
Limit of
Detectability
(pptr)
^-HCH
Heptachlor
Oxychlordane
Heptachlor epoxide
trans-Nonachlor
p,p'-DDE
Dieldrin
o,p'-DDT
p,p'-DDT
HCB
•y-Chlordane
O-Chlordane
6
3
3
5
4
4
8
11
14
2
6
7
25
10
13
18
15
17
30
44
56
6
23
28
164
-------
Table 2. PERCENT RECOVERY OF OC'S AND OC METABOLITES IN WATER
Compound Fraction Recovery
P-HCH 11 88
Haptachlor 1 81
Oxychlordane 1 89
Heptachlor epoxide 11 86
trans -Nona chlor 1 93
p,p'-DDE 1 97
Dieldrin 11 93
o,p'-DDT 1 96
p,p'-DDT 1 96
HCB 1 79
y-Chlordane 1 86
a-Chlordane 1 94
Mean 90
165
-------
3. 2.2 cm x 30 cm Pyrex filtering column
4. 125 ml separatory funnel
5. 15 ml centrifuge tubes
6. 7 mm i.d. ChromaFlex column
5.2 Reagents
1. Methylene Chloride, Nanograde from Mallinckrodt Chemicals
®
2. Hexane, Nanograde from Mallinckrodt Chemicals
3. Deionized water washed twice with benzene
®
4. Benzene, Nanograde from Mallinckrodt Chemicals
5. Na_SO, aunydrous
6. "Keeper" solution of 1% USP paraffin oil/hexane
7. Silica Gel, Woelm, activity grade I from Waters Associates, Inc.,
deactivated with 20% water.
6.0 Procedure
6.1 Collection of Samples
Drinking water samples are collected from kitchen water faucets. Each 1
L sample is contained in a glass bottle with a ground glass stopper and stored,
removed from any source of organo-chlorine pesticide, at 4°C until analyzed.
6.2 Extraction
1. Add 5 ml of methylene chloride to 50 mL of water in a 125 mL
sep funnel and shake vigorously for 2 rain.
2. Drain the organic (lower) phase through a 2.2 cm x 30 cm filter
column with a 2 cm layer of Na SO, into a 15 mL conical centrifuge
tube.
3. repeat 1 & 2
4. Rinse column with an additional 4 mL of solvent and combine all
elutes.
5. Add 5 drops of "Keeper" (see Reagents) and concentrate to 0.5 mL
under a stream of dry nitrogen at 40°C.
6. Prepare a silica gel column by adding to a 7 mm i.d. column
plugged with a small pad of glass wool, 1 g of deactivated (20%
water) silica gel. Settle the silica gel with firm tapping and top
with 2.5 cm NaSO,.
4
166
-------
7. Wash column with 10 ml of hexane and discard. As the hexane
reaches the Na2SO, place a 15 mL centrifuge tube under the column.
8. Transfer sample to column using a disposable pipet, rinse concentrating
tube with 1 ml of hexane and transfer wash to the column using same
disposable pipet.
9. repeat 8 wash step 2 more times
10. Add 6.5 mL of hexane to the column and combine effluents.
11. Add 5 drops of "keeper", concentrate to 0.2 ml and redilute to
1 mL. This is fraction I to be analysed by GC/ECD. (see Table 2
for compounds eluted in this fraction
12. As soon as fraction I has eluted change to another centrifuge
tube and add 15 mL of 3:2 benzene:hexane (V/V) to the column and
elute. This is fraction II
13. Add 5 drops of keeper and concentrate to 0.2 ml, and redilute
with hexane to 1 mL. This solution is then ready for analysis by
GC/ECD.
6.3 Analysis
The GC operating conditions for the analysis of pesticides is as follows:
1. N_ carrier flow rate of 60 mL/min
2. Injection port temperature: 215°C
3. Column temperature: 200°C (isothermal)
4. Detector temperature: 210°
These parameters were chosen to maximize column efficiency and minimize analysis
time, the latest eluting pesticide, p,p'-DDT eluting in 18 min.
The usual injection was 10-|jL and convenient electrometer settings were
10 x 8 and 10 x 16.
6.4 Qualitative Identification
Qualitative identification is made based on relative retention times and
is subject to the limitations of this methodology (2).
6.5 Quantitation
Quantitation is based on peak heights for the early eluting sharp peaked
compounds and on peak areas for the later eluting broader peaks (e.g., £,£*-
DDT) (2).
167
-------
7.0 Quality Assurance Program
An on-going quality assurance program is required to assure the data
quality. Quality control (QC) procedures determine artifacts, losses, etc.
through a system of blanks and controls. Quality assurance (QA) procedures
monitor the execution of the procedure and check data interpretations and
calculations.
7.1 Quality Control
7.1.1 Field Blanks and Controls
Prior to a field sampling trip, enough blanks and controls are prepared
to equal 10% each (2 minimum) of the anticipated number of field samples.
Blanks consist of 50 mL of water in the same type of sampling container as is
used in the field. Controls consist of 50 ml of water spiked at 50-202 ng
with the compounds listed in Table 3. These blanks and controls are carried
to the field and receive the same handling as the field samples. Workup and
analysis of field blanks and controls is interspersed with the field samples
on a regular basis. This method allows assessment of sample storage stability
Table 4 presents a typical set of blanks and controls for QC on a field
trip where 50 water samples are to be collected.
7.1.2 Procedural Blanks and Controls
7.1.2.1 Extraction Blanks
With each set of samples, a procedural blank is run. This consists of 50
mL of prepurged distilled water which is extracted under the same conditions
as the samples. These blanks are designed to detect artifacts from dirty
glassware, laboratory atmosphere intrusion, and other sources.
7.1.2.2 GC/ECD Procedural Control
Field samples, field controls, field blanks, and procedural blanks are
queued up for GC/ECD analysis such that at least one QC sample is run each
working day. In addition, a standard solution is analyzed each day to serve
as a procedural control. Thus, in a typical working day, 4 field samples and
1 blank or control are run.
In addition standard solutions are run at regular intervals at least
early, middle and later in a day) to monitor detector response and update the
analytical curve.
168
-------
Table 3. PESTICIDES USED AS CONTROLS IN THIS STUDY
Pesticide
Amount
100 ng
60 ng
51 ng
95 ng
59 ng
103 ng
100 ng
100 ng
202 ng
200 ng
19 ng
100 ng
101 ng
Reference I
Reference II
3-Hexachlorocyclohexane
Heptachlor
Aldrin
Oxychlordane
Heptachlor epoxide
trans-Nonachlor
pp'-DDE
Dieldrin
o-p'-DDT
pp'-DDT
Hexachlorobenzene
Of-Chlordane
•y-Chlordane
169
-------
Table 4. WATER QC SAMPLES FOR A TYPICAL SAMPLING TRIP
Sample Type
Field Blank
Number
5
Comments
refrigerate (40) after pre-
paration, carry to field,
store with field samples
Field Control 5 Store with field blanks
Lab Blank 5 Refrigerate after preparation
store in same freezer as
field samples will be stored
Lab control 5 Store with Lab Blanks
170
-------
7.2 Quality Assurance
Both internal and external QA procedures are to be followed. Internal
QA procedures assure the continuity and consistancy of the data. External QA
procedures (interlaboratory checks) verify or dispute the accuracy of the
data being generated in the primary laboratory.
7.2.1 Internal Quality Assurance
7.2.1.1 Supervision and Monitoring of Activities
There are three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This person
must be aware of their actions, observe events which may effect the data, and
maintain appropriate records. At the second level, the chemist's supervisor
monitors their daily activities, reviews the notebook, checks data and calcu-
lations, and assists in "troubleshooting" problems. At the tertiary level, a
QA coordinator interviews all personnel on the project. The interviews cover
the operations they perform (precisely), the data they obtain, a spot-check
of their calculations, and any problems they have had.
7.2.1.2 Documentation
Chain of Custody
From the initial preparation of a sample container through reporting of
the analytical results, each sample is accompanied by a chain of custody
sheet. Each person signs in the time of receipt, operations performed, and
transmittal of the sample. This record is important for tracing a contaminant,
bad standard, or some other problem.
Sampling Protocol Sheets
When a sample is collected, a sampling protocol sheet is filled in which
contains a discrete sample code which identifies project number, area, site,
locations, trip number, sampling period, and sample type. Also included are
sample times, volumes, addresses, meteorology, and other pertinent information.
Where appropriate, a map is made to precisely identify the location.
Sample Log
Upon return from a sampling trip, each sample code is entered into a
sample log book. This log is updated as samples proceed through workup and
analysis. Thus, at a glance, project personnel can tell the status of each
sample and find out how many are at different stages in the analytical protocol
171
-------
7.2.2 External Quality Assurance
Samples will be analyzed by a designated QA laboratory. Samples, controls,
and blanks will be shipped directly from the field to the QA laboratory for
analysis. They will report the results to the primary laboratory for correlation
with the primary data.
7.2,2.1 Selection of Samples for QA
Approximately 10% of the field samples (2 minimum) will be collected in
duplicate (i^e. , 2 50 ml water samples) for shipment to the QA laboratory.
This selection process will be random with the following restrictions:
(1) The subject must consent to the additional water collection
(2) If any stratification of subjects is known, purposive selection of
QA subjects may be used to get representative samples (e.g., occupa-
tionally exposed vs. "normal" individuals)
7.2.2.2 QC Samples
Along with the QA duplicate samples, at least one QC control and one QC
blank must be included with the QA samples. An example is shown in Table 5
for a trip collecting 50 samples.
7.2.2.3 Sample Codes
All samples shipped to the QA laboratory should be encoded so that the
laboratory cannot discern between samples and controls. This "blind" protocol
will add validity to the results.
7.2.2.4 Shipping
Samples should be shipped on ice directly to the QA laboratory by an
appropriate air carrier (e.g., Federal Express, Eastern Sprint) in well
insulated and packed cartons.
8iO References
(1) Pellizzari, E.D., M. D. Erickson and R.A. Zweidinger, "Analytical
Protocols for Making a Preliminary Assessment of Halogenated Organic
Compounds in Man and Environmental Media", Appendix G, Pg 165. Revised
April 79.
172
-------
Table 5. SAMPLES TO BE COLLECTED AND SHIPPED TO QA LABORATORY
FOR A TYPICAL SAMPLING TRIP
Sample Number Comment
Duplicate Sample 5 Random selection unless
prior information stratifies
subjects
Field Blank 1 Ship with samples
Field Control 1 Ship with samples
173
-------
ANALYTICAL PROTOCOL: SAMPLING AND ANALYSIS OF
ARSENIC, CADMIUM, AND LEAD IN DRINKING WATER (RTI)
1.0 Principle of Method
The analysis of arsenic, cadmium, and lead in drinking water is carried
out using atomic absorption spectrophotometry. Increased sensitivity is
achieved by atomizing the metal in a graphite furnace with continuous deu-
terium background correction. Arsenic determinations are performed on
solutions containing 1000 ppm nickel.
2.0 Range and Detection Limit
The minimum detection limit (MDL) and range for the metal assays in
drinking water are shown below.
Metal MDL (|Jg/L) Max. Cone. (pg/L)
Arsenic 3.50 200.0
Cadmium 0.04 50.0
Lead 0.35 50.0
Samples containing higher metal concentrations may be analyzed by
suitable dilution with 0.5 of nitric acid. Dilution for arsenic determi-
nations is made with 1.0% nitric acid containing 1000 ppm nickel.
3.0 Interferences
No known chemical or spectral interferences exist in the analysis of
arsenic, cadmium, or lead in drinking water.
4.0 Precision and Accuracy
The precision and accuracy associated with these analyses is a function
of sample metal concentration at the detection limit, the total measurement
error is + 100%. Based on the results of a previous study (1), the metal
analyses are performed with the following precision (relative standard
deviation) and accuracy (relative error). The total analysis error is also
given (2).
Metal Range (|Jg/L) Precision (% RSD) Accuracy (% RE) Total Error (%)
Arsenic 10-30 10 10 30
Cadmium 0.5-1.0 10 10 30
Lead 10-30 5 10 20
174
-------
5-0 Apparatus and Reagents
A commercially available stock solution containing 1000 ppm metal is
used for the preparation of the calibration standards. The concentrated
nitric acid is reagent grade quality and the deionized water used in this
study is prefiltered and subjected to the action of an activated carbon
cartridge and two sequential ion exchange units.
The glassware used for the preparation of the calibration solutions
must be subjected to a nitric acid cleaning protocol.
All volumetric flasks should be soaked overnight in 20% nitric acid,
rinsed with deionized water, soaking for an additional 15-18 hours in a 5%
nitric acid bath, followed by a deionized water rinse. The flasks are
completely filled with 0.5% nitric acid and stored in this manner. Prior to
use, each flask is emptied and rinsed well with deionized water. Pipets are
soaked in 5% nitric acid, rinsed well with deionized water, rinsed well with
deionized water, air-dried, and stored in a clean, dust-free environment.
Sample cups for the graphite furnace autosampler may be made of polysty-
rene or Teflon. The former type requires overnight soaking in 1% nitric
acid and followed by rinsing with deionized water. The latter type may be
soaked overnight in 20% nitric acid, rinsed, and dried in a 105°C oven.
Nickel chloride hexahydrate is used for adjusting the nickel concen-
tration in all samples and standards to 1000 ppm.
6.0 Procedure
6.1 Collection of Samples
Drinking water samples are collected in 4-ounce wide mouth polyethylene
bottles with a Polyseal cap and spiked with 0.5 ml concentrated nitric
acid/100 ml sample. The sample is labeled and its location and other perti-
nent data recorded on a protocol sheet.
6.2 Extraction, Cleanup, and Concentration
None
6.3 Instrumental
A Perkin-Elmer Model 403 Spectrophotometer, equipped with a HGA-2000
furnace attachment with deuterium background correction is used for this
analysis. An electrodeless discharge lamp is used as the light source and
the furnace atmomization response traced on a Perkin-Elmer Model 056 recorder
175
-------
An AS-1 Autosampler is used to increase throughput and/or to improve pack
reproducibility and sensitivity.
Arsenic: Wavelength - 193.7 nm
Gas Interrupt (N-) - Auto Furnace Cycle Conditions -
Dry: 200°C for 20 sec.
Char: 1200°C for 30 sec.
Atomize: 2500°C for 8 sec.
Injection Volume - 20 pi
Cadmium: Wavelength - 228.8 nm
Gas Interrupt (N«) - Auto
Furnace Cycle Conditions -
Dry: 200°C for 20 sec.
Char: 400°C for 20 sec.
Atomize: 2000°C for 8 sec.
Injection Volume - 20 (Jl
Lead: Wavelength - 217.0 nm
Gas Interrupt (N,,) - Auto
Furnace Cycle Conditions -
Dry: 200°C for 20 sec.
Char: 550°C for 20 sec.
Atomize: 2000°C for 8 sec.
Injection Volume - 20 |Jl
6.4 Qualitative and Quantitative Analysis
6.4.1 Qualitative Analysis
N/A
6.4.2 Quantitative Analysis
The instrument is calibrated with four aqueous standards and a reagent
blank.
Calibration Range:
Arsenic - 0.0 to 4.0 ng/20 pi
Cadium - 0.0 to 1.0 ng/20 pi
Lead - 0.0 to 1.0 ng/20 pi
176
-------
An exponential of the form y = Ae -M provides the best representation
of the analytical curve. The values of the x,y calibration pairs are ente-
red into a Monroe Calculator Model 1880 programmed to regress the data to
the exponential and to provide values for the constants A, b, and M.
Sample peak heights are measured manually and expressed in units of
millivolts. The calibration constants, A, b, and M are entered into the
storage banks of a Texas Instrument Calculator Model 57 and the metal
concentration results obtained by keying in peak height data. Sample peak
measurements and concentration results are recorded on a calculation work-
sheet
y = Aebx-M, ng/20 |jl
Units Conversion: ng/20 (j •* (jg/liter
MS \ p__|Jl\ _ |jg
°
__
20 pi 1000 ng liter liter
y = 50 (Aebx-M), (Jg/liter
y = SOD (Aebx-M)
y = metal concentration in sample, pg/liter,
x = sample peak height, mv
D = dilution factor
7.0 Quality Assurance Program
An on-going quality assurance program is required to assure the data
quality. Quality control (QC) procedures determine artifacts, losses, etc.,
through a system of blanks and controls. Quality assurance (QA) procedures
monitor the execution of the procedure and check data interpretation and
calculations .
7.1 Quality Control
7.1.1 Field Controls
Prior to field sampling, several blanks and spiked water samples (10%
of anticipated number of field samples, are prepared. These field controls
are placed in a 4-ounce polyethylene bottle (Polyseal cap), adjusted to 0.5%
nitric acid and carried to the sampling site. They are subjected to the
same handling and storage conditions as field samples. The analysis of
these samples is a part of each water analytical run. Within the precision
177
-------
of the assay, the calculated metal concentration of these controls is a
measure of the contamination/loss during field storage and transit to RTI.
7.1.2 Internal Quality Control
7.1.2.1 Calibration Standards and Blanks
The instrument is calibrated before aech analytical run with four
standard solutions and a reagent blank. Evidence of contamination or
instrument malfunction is evident at this time. Such problems are resolved
before initiating sample analysis.
7.1.2.2 Conditioning of Graphite Tube
The instrument is calibrated before each analytical run with four
standard solutions and a reagent blank. Evidence of contamination or
instrument malfunction is evident at this time. Such problems are resolved
before initiating samples analysis.
7.1.2.2 Conditioning of Graphite Tube
Before each analytical run, the graphite tube is conditioned by injecting
10 to 20 20 pi aliquots of one of the calibration standards. This operation
insures acceptable precision during sample analysis.
7.1.2.3 Duplicate Injections
Reproducibility of peak response is continuously monitored during
sample analysis. All standard and sample solutions receive two successive
injections into the graphite furnace. Signal agreement between the duplicate
injections is evaluated according to the following criterion:
First Signal % Maximum Permissible Permissible Range of
% of Full Scale Variation (% MPV) Second Signal, % of Full Scale
90 ± 4% 86-94
80 ± 5% 76-84
70 ± 6% 66-74
60 ± 1% 56-64
50 ± 8% 46-54
40 ± 10% 36-44
30 ± 30% 26-34
20 ± 20% 16-24
10 ± 30% 7_13
5 ± 60% 2.8
2 ±100% 0-4
178
-------
If the second injection gives a signal which falls outside the per-
missible range, a third injection is performed. The peak measurement not
in agreement with the maching pair is discarded.
All calibration and sample calculations are based on the mean of the
duplicate determinations.
7.1.2.4 Standard Checks
Instrument performance is monitored during each analytical run. After
the analysis of every 12-16 samples, one of the calibration standards is
reinjected into the furnace. The standard which most closely matches the
sample peak heights is selected as the check solution. A metal concentra-
tion is calculated for the check standard based on its peak height during
the calibration run. Similar calculations are carried out for each check
response and the observed changes in metal concentration expressed in terms
of standard deviation units (SDU).
ST)T] _ (Calibration Value - Check Value) 100
(Calibration Value) (% of RSD)
The analysis is under control when the SDU < 2.0. Standard checks
which indicate a variation in peak response greater than 2.0 SDU are unac-
ceptable. In this event, the graphite tube is changed, conditioned, and the
system recalibrated. Quality control charts are graphed to show this change
in instrument performance with time.
7.2 Quality Assurance
Both internal and external QA procedures are to be followed. Internal
QA procedure assume the continuity and consistancy of the data. External QA
procedures (interlaboratory checks) verify or dispute the accuracy of the
data being generated in the primary laboratory.
7.2.1 Interal Quality Assurance
7.2.1.1 Supervision and Monitoring Activities
There are three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This
person must be aware of their actions, observe events which may effect the
data, and maintain appropriate records. At the second level, the chemist's
supervisor monitors their daily activities, reviews the notebook, checks
data and calculations, and assists in "troubleshooting" problems. At the
179
-------
tertiary level, a QA coordinator interviews all personnel on the project.
The interviews cover the operations they perform (precisely), the data they
obtain, a spot-check of their calculations, and any problems they have had.
7.1.2.2 Documentation
Chain of Custody
From the initial preparation of a sample container through reporting of
the analytical results, each sample is accompanied by a chain of custody
sheet. Each person signs in the time of receipt, operations performed, and
transmittal of the sample. This record is important for tracing a contaminant,
bad standard, or some other problems.
Sampling Protocol Sheets
When a sample is collected, a sampling protocol sheet is filled in
which contains a discrete sample code which identifies project number, area,
site, locations, trip number, sampling period, and sample type. Also included
are sample times, volumes, addresses, meteorology, and other pertinent
information. Where appropriate, a map is made to precisely identify the
location.
Sample Log
Upon return from a sampling trip, each sample code is entered into a
sample log book. This log is updated as samples proceed through workup and
analysis. Thus, at a glance, project personnel can tell the status of each
sample and find out how many are at different stages in the analytical
protocol.
Instrument Log
Each sample analysis is logged into a notebook, detailing analysis
conditions.
7.2.2 External Quality Assurance .
Samples will be analyzed by a designated QA laboratory. Samples,
controls, and blanks will be shipped directly from the field to the QA
laboratory for analysis. They will report the results to the primary
laboratory for correlation with the primary data.
180
-------
7.2.2.1 Selection of Samples for QA
Approximately 10% of the field samples (2 minimum) will be collected in
duplicate for shipment to the QA laboratory. This selection process will be
random with the following restrictions:
(1) The donor must consent to the additional collection.
(2) If any stratification of donors is known, purposive selection
of QA donors may be used to get representative samples (e.g.,
occupationally exposed vs "normal" individuals or upwind vs downwind
residents).
7.2.2.2 QC Samples
Along with the QA duplicate samples, at least one QC control and .one
QC blank must be included with the QC samples. An example is shown in Table
7 for a trip collecting 50 samples.
7.2.2.3 Sample Codes
All samples shipped to the QA laboratory should be encoded so that the
laboratory cannot discern between samples and controls. This "blind"
protocol will add validity to the results.
7.2.2.4 Shipping
Samples should be shipped on dry ice directly to the QA laboratory by
an appropriate air carrier (e.g., Federal Express, Eastern Sprint) in well
insulated and packed cartons.
8.0 References
8-1 "Epidemiologic Study Conducted in Populations Living Around Non-Ferrous
Smelters", Final Report for Contract No. 68-02-2442 (in preparation).
8-2 McFarren, E. F. , Lishka, R. J. and Parker, J. H., Criterion for Judging
Acceptability of Analytical Methods, Anal. Chem., 42(3), 358 (1970).
181
-------
ANALYTICAL PROTOCOL: SAMPLING AND ANALYSIS OF ARSENIC, CADMIUM
AND LEAD IN WATER (EMSL-CI)
1.0 Principle of Method
Drinking water is analysed for arsenic, cadmium and lead by atomic
absorption spectroscopy (A.A.). Metals in solution are readily determined
by A.A. The procedure used here is furnace atomic absorption because of
greater sensitivity and less interference from chemicals and sample matrices
than with a flame.
2.0 Range and Detection Limit
Arsenic and Lead - Optimal Concentration Range 5-100 (Jg/L
Detection Limit - 1 |Jg/L
Cadmium - Optimal Concentration Range .5-10 |Jg/L
Detection Limit - 0.1 pg/L
With quantitative dilution the upper limit of range can be increased
indefinitely.
3.0 Interferences
This technique is subject to chemical and matrix interference, the
presence or absence of which should be verified by a standard additions
method on a spiked and unspiked sample diluted to a least half of its original
concentration.
There are several other potential interferences associated with furnace
AA such as presence of organics in the matrix, absorptive gas generation,
spectral interference from other elements present, carbide formation, and
external contamination of the sample. The analysis therefore should be done
by an analyst experienced with the potential sources of error associated
with the method, and the techniques for correcting them.
4.0 Precision and Accuracy
In a single laboratory (EMSL), using Cincinnati, Ohio tap water spiked
at concentrations of 20, 50 and 100 |Jg As/1, the standard deviations were
±0.7, ±1.1 and ±1.6 respectively. Recoveries at these levels were 105%,
106% and 101%, respectively, Cincinnati, Ohio tap water spiked at concentra-
tions of 2.5, 5.0 and 10.0 (jg Cd/1, gave standard deviations of ±0.10, ±0.16
and ±0.33, respectively. Recoveries at these levels were 96%, 99% and 98%,
182
-------
respectively, and Cincinnati, Ohio tap water spiked at concentrations of 25,
50, and 100 \ig Pb/1, the standard deviations were ±1.3, ±1.6, and ±3.7,
respectively. Recoveries at these levels were 88%, 92%, and 95% respect-
ively.
5.0 Apparatus and Reagents
1. Atomic absorption spectrophotometer: Single or dual channel,
single-or double-beam instrument having a grating monochromator,
photomultiplier detector, adjustable slits, a wavelength range of
190 to 800 run, and provisions for interfacing with a strip chart
recorder.
2. Hollow cathode lamps: Single element lamps are to be preferred
but multi-element lamps may be used. Electrodeless discharge
lamps may also be used when available.
3. Graphite furnace: Any furnace device capable of reaching the
specified temperatures is satisfactory.
A. Strip chart recorder: A recorder is strongly recommended for
furnace work so that there will be a permanent record and any
problems with the analysis such as drift, incomplete atomization,
losses during charring, changes in sensitivity, etc., can be
easily recognized.
5. Pipets: Microliter with disposable tips. Sizes can range
from 5 to 100 microliters as required. NOTE: Pipet tips which
are white in color, do not contain CdS, and have been found suitable
for research work are available from Ulster Scientific, Inc. 53
Main St. Highland, NY 12528 (914)691-7500.
7. Separatory flasks: 250 ml, or larger, for extraction with organic
solvents. NOTE: Glassware; all glassware, linear polyethylene,
polyproplyene or Teflon containers, including sample bottles,
should be washed with detergent, rinsed with tap water, 1:1 nitric
acid, tap water, 1:1 hydrochloric acid, tap water and deionized
distilled water in that order.
8. Borosilicate glass distillation apparatus.
183
-------
5.2 Reagents
1. Deionized distilled water: Prepare by passing distilled water
through a mixed bed of cation and anion exchange resins.
2. Use deionized distilled water for the preparation of all reagents,
calibration standards, and as dilution water.
3. Nitric acid (cone.): If metal impurities are found to be present,
distill reagent grade nitric acid in a borosilicate glass distilla-
tion apparatus or use a spectrograde acid. Caution: Distillation
should be performed in hood with protective sash in place.
4. Nitric Acid (1:1): Prepare a 1:1 dilution with deionized, dis-
tilled water by adding the cone, acid to an equal volume of water.
5. Argon and Nitrogen: Used as furnace purge gas.
6. Arsenic Analyses
Arsenic Trioxide, As^C- , analytical reagent grade
Nickel Nitrate, Ni^CLK^H-O, analytical reagent grade
7. Cadmium Analysis
Cadmium Sulfate (3CdSO,'SH^O); analytical reagent grade
Ammonium phosphate (NH,)_HPO, analytical reagent grade.
8. Lead Analysis
Lead Nitrate (Pb(NO«)2) analytical reagent grade
Lanthanum Oxide (LaJD.) analytical reagent grade
6.0 Procedure
6.1 Sample Handling and Preservation
For the determination of trace metals, contamination and loss are of
prime concern. Dust in the laboratory environment, impurities in reagents
and impurities on laboratory apparatus which the sample contacts are all
sources of potential contamination. For liquid samples, containers can
introduce either positive or negative errors in the measurement of trace
metals by (a) contributing contaminants through leaching or surface desorption
and (b) by depleting concentrations through adsorption. Thus the collection
and treatment of the sample prior to analysis requires particular attention.
The sample bottle whether borosilicate glass, linear polyethylene, polypro-
plyene or Teflon should be thoroughly washed with detergent and tap water;
rinsed with 1:1 nitric acid, tap water, 1:1 hydrochloric acid, tap water and
184
-------
finally deionized distilled water in that order. Before collection of the
sample a decision must be made as to the type of data desired, i.e., dissolved,
suspended, total or total recoverable. For container preference, maximum
holding time and sample preservation at time of collection see Table 1.
To determine total recoverable metals, acidify the entire sample at the
time of collection with cone, redistilled HNCL, 5 ml/1. At the time of
analysis a 100 ml aliquot of well mixed sample may be transferred to a
beaker or flask and heated on a steam bath or hot plate until the volume has
been reduced to 15-20 ml, making certain the samples do not boil. After
this treatment the sample is filtered to remove silicates and other insoluble
material and the volume adjusted to 100 ml. The sample is then ready for
analysis. Concentrations so determined shall be reported as "total".
6.2.1 Preparation of Standard Arsenic Solution
1. Stock solution: Dissolve 1.320 g of arsenic trioxide, As-O.,
(analytical reagent grade) in 100 ml of deionized distilled water
containing 4 g NaOH. Acidify the solution with 20 ml cone. HNCL
and dilute to 1 liter. 1 ml = 1 mg As (1000 mg/1).
2. Nickel Nitrate Solution, 5%: Dissolve 24.780 g of ACS reagent
grade Ni(NO_),?'6H,)0 in deionized distilled water and make up to
100 ml.
3. Nickel Nitrate Solution, 1%: Dilute 20 ml of the 5% nickel nitrate
to 100 ml with deionized distilled water.
4. Working Arsenic Solution: Prepare dilutions of the stock solution
to be used as calibration standards at the time of analysis.
Withdraw appropriate aliquots of the stock solution, add 1 ml of
cone. HN03, 2 ml of 30% H^, and 2 ml of the 5% nickel nitrate
solution. Dilute to 100 ml with deionized distilled water.
6.2.2 Arsenic Sample Preparation
1. Transfer 100 ml of well-mixed sample to a 250 ml Griffin beaker,
add 2 ml of 30% H-Cv and sufficient cone. HNO. to result in an
acid concentration of 1% (v/v). Heat for 1 hour at 95°C or until
the volume is slightly less than 50 ml.
2. Cool and bring back to 50 ml with deionized distilled water.
185
-------
Table 1. RECOMMENDATION FOR SAMPLING AND PRESERVATION OF SAMPLES
ACCORDING TO MEASUREMENTS
Metals
Dissolved
Suspended
Total
Volume
Required (m£)
200
200
100
Container
polyethylene
with propylene cap
no liner
M
Preservative
Filter onsite
HN03 to pH < 2
Filter onsite
HNO to pH < 2
Maximum
Holding Time
6 mos
6 mos
6 mos
00
-------
3. Pipet 5 ml of this digested solution into a 10-ml volumetric
flask, add 1 ml of the 1% nickel nitrate solution and dilute to 10
ml with deionized distilled water. The sample is now ready for
injection into the furnace.
6.2.3 Preparation of Standard Cadmium Solution
1. Stock Solution: Carefully weigh 2.282 g of cadmium sulfate
(SCdSO,-SH^O, analytical reagent grade) and dissolve in deionized
distilled water. 1 ml = 1 rag Cd(1000 mg/1).
2. Ammonium Phosphate solution (40%): Dissolve 40 grams of ammonium
phosphate, (NH,)_HPO, (analytical reagent grade) in deionized
distilled water and dilute to 100 ml.
3. Prepare dilutions of the stock cadmium solution to be used as
calibration standards at the time of analysis. To each 100 ml of
standard and sample alike add 2.0 ml of the ammonium phosphate
solution. The calibration standards should be prepared to contain
0.5% (v/v) HN03.
6.2.4 Cadmium Sample Preparation
Prepare as described under Sample Handling and Preservation (above).
Sample and solutions for analyses should contain 0.5% (v/v) HNCL.
6.2.5 Preparation of Standard Lead Solution
1. Stock Solution: Carefully weigh 1.599 g of lead nitrate, Pb(N03)2
(analytical reagent grade), and dissolve in deionized distilled
water. When solution is complete, acidify with 10 ml redistilled
HNO« and dilute to 1 liter with deionized distilled water. 1 ml =
1 mg Pb (1000 mg/1).
2. Lanthanum Nitrate Solution: Dissolve 58.64 g of ACS reagent grade
La0C- in 100 ml cone. HNO., and dilute to 1000 ml with deionized
£ 3 3
distilled water. 1 ml = 50 mg La.
3. Working Lead Solution: Prepare dilutions of the stock lead solution
to be used as calibration standards at the time of analysis. Each
calibration standard should contain 0.5% (v/v) HNOg. To each 100
ml of diluted standard add 10 ml of the lanthanum nitrate solution.
187
-------
6.2.6 Lead Sample Preparation
1. Prepare as described under Sample Handling and Preservation
(above). Sample solutions for analyses should contain 0.5% (v/v)
HN03.
2. To each 100 ml of prepared sample solution add 10 ml of the lanthanum
nitrate solution.
6.2.7 Calibration Standards and Curves
Calibration standards are prepared by diluting the stock metal solutions
at the time of analysis. For best results, calibration standards should be
prepared fresh each time an analysis is to be made and discarded after use.
Prepare a blank and at least four calibration standards in graduated amounts
in the appropriate range. Beginning with the blank and working toward the
highest standard, aspirate the solutions and record the readings. Repeat the
operation with both the calibration standards and the samples a sufficient
number of times to secure a reliable average reading for each solution.
6.3 Analysis
6.3.1 Furnace Procedure
Furnace devices (flameless atomization) are a most useful means of
extending detection limits. Because of differences between various makes and
models of satisfactory instruments, no detailed operating instructions can be
given for each instrument. Instead, the analyst should follow the instructions
provided by the manufacturer of his particular instrument and use as a guide
the temperature settings and other instrument conditions listed for each
metal. The listing for each metal are the recommended settings working on a
Perker-Elmer HGA-2100.
6.3.2 Arsenic Analysis
Instrument Parameters (General)
1. Drying Time and Temp: 30 sec-125°C.
2. Ashing Time and Temp: 30 sec-1100°C.
3. Atomizing Time and Temp: 10 sec-2700°C.
4. Purge Gas Atmosphere: Argon
5. Wavelength: 193.7 nm
6. Other operating parameters should be set as specified by
the particular instrument manufacturer.
188
-------
6.3.3 Cadmium Analysis
Instrument Parameters (General)
1. Drying Time and Temp: 30 sec-125°C.
2. Ashing Time and Temp: 30 sec-500°C.
3. Atomizing Time and Temp: 10 sec-1900°C.
4. Purge Gas Atmosphere: Argon
5. Wavelength: 228.8 nm
6. Other operating parameters should be set as specified by the
particular instrument manufacturer.
6.3.4 Lead Analysis
Instrument Parameters(General)
1. Drying Time and Temp: 30 sec-125°C.
2. Ashing Time and Temp: 30 sec-500°C.
3. Atomizing Time and Temp: 10 sec-2700°C.
4. Purge Gas Atmosphere: Argon
5. Wavelength: 283.3 nm
6. Other operating parameters should be set as specified by the
particular instrument manufacturer.
6.3.5 General Notes
The following points may be helpful in the analyses of samples. With
flameless atomization, background correction becomes of high importance
especially below 350 nm. This is because certain samples, when atomized, may
absorb or scatter light from the hollow cathode lamp. It can be caused by
the presence of gaseous molecular species, salt particules, or smoke in the
sample beam. If no correction is made, sample absorbance will be greater
than it should be, and the analytical result will be erroneously high. If
during atomization all the analyte is not volatilized and removed from the
furnace, memory effects will occur. This condition is dependent on several
factors such as the volatility of the element and its chemical form, whether
pyrolytic graphite is used, the rate of atomization and furnace design. If
this situation is detected through blank burns, the tube should be cleaned by
operating the furnace at full power for the required time period as needed at
regular intervals in the analytical scheme. Some of the smaller size furnace
devices, or newer furnaces equipped with feedback temperature control
189
-------
(Instrumentation Laboratories MODEL 555, Perkin-Elmer MODELS HGA 2200 and
HGA 76B, and Varian MODEL CRA-90) employing faster rates of atomization, can
be operated using lower atomization temperatures for shorter time periods
than those listed in this manual. Although prior digestion of the sample in
many cases is not required, providing a representative aliquot of sample can
be pepeted into the furnace, it will provide for a more uniform matrix and
possibly lessen matrix effects. Inject a measured microliter aliquot of
sample into the furnace and atomize. If the concentration found is greater
than the highest standard, the sample should be diluted in the same acid
matrix and reanalyzed. The use of multiple injections can improve accuracy
and help detect furnace pipetting errors. To verify the absence of interfer-
ence, follow the procedure as given in part 5.2.1. A check standard should
be run approximately after every 10 sample injections. Standards are run in
part to monitor the life and performance of the graphite tube. Lack of
reproducibility or significant change in the signal for the standard indicates
that the tube should be replaced. Even though tube life depends on sample
matrix and atomization temperature, a conservative estimate would be that a
tube will last at least 50 firings. A pyrolytic-coating would extend that
estimate by a factor of 3.
6.4 Qualitative Identification
Not applicable. (determined by wavelength being monitored).
6.5 Quantitation
For determination of metal concentration by the furnace: Read the
metal value in |jg/l from the calibration curve or directly from the readout
system of the instrument.
If different size furnace injection volumes are used for samples than
for standards:
|Jg/l of metal in sample = Z(S/U)
where:
Z = Mg/1 of metal read from calibration curve or readout system
S = ul volume standard injected into furnace for calibration curve
U = ul volume of sample injected for analysis
If dilution of sample was required but sample injection volume same
as for standard:
190
-------
pg/1 of metal in sample = Z[(C+B)/C]
where:
Z = |Jg/l metal in diluted aliquot from calibration curve
B = ml of deionized distilled water used for dilution
C = ml of sample aliquot
7.0 Quality Assurance Program
An on-going quality assurance program is required to assure the data
quality. Quality control (QC) procedures determine artifacts, losses, etc.
through a system of blanks and controls. Quality assurance (QA) procedures
monitor the execution of the procedure and check data interpretations and
calculations.
7.1 Quality Control
7.1.1 Procedural Quality Control
Minimum Requirements
All quality control data should be maintained and available for easy
reference or inspection. An unknown performance sample (when available)
must be analyzed once per year for the metals measured. Results must be
within the control limit established by EPA. If problems arise, they should
be corrected, and a follow-up performance sample should be analyzed.
Minimum Daily Control
After a calibration curve composed of a minimum of a reagent blank and
three standards has been prepared, subsequent calibration curves must be
verified by use of at least a reagent blank and one standard at or near the
MCL. Daily checks must be within ±10 percent of original curve. If 20 or
more samples per day are analyzed, the working standard curve msut be verified
by running an additional standard at or near the MCL every 20 samples.
Checks must be within ±10 percent of original curve.
Optional Requirements
A current service contract should be in effect on balances and the
atomic absorption spectrophotometer. Class S weights should be available to
make periodic checks on balances. Chemicals should be dated upon receipt of
shipment and replaced as needed or before shelf life has been exceeded. A
known reference sample (when available) should be analyzed once per quarter
for the metals measured. The measured value should be within the control
191
-------
limits established by EPA. At least one duplicate sample should be run every
10 samples, or with each set of samples to verify precision of the method.
Checks should be within the control limit established by EPA. Standard
deviaiton should be obtained and documented for all measurements being conducted
Quality Control charts or a tabulation of mean and standard deviation should
be used to document validity of data on a daily basis.
7.1.2 Field Blanks and Controls
Prior to a field sampling trip, enough blanks and controls are prepared
to equal 10% each (2 minimum) of the anticipated number of field samples.
Blanks consist of 100 ml of water in the same type of sampling container as
is used in the field. Controls consist of 100 ml of water spiked at 1 M8
arsenic and lead and 0.1 (J cadmium. These blanks and controls are carried to
the field and receive the same handling as the field samples. Workup and
analysis of field blanks and controls is interspersed with the field samples
on a regular basis. This method allows assessment of sample storage stability.
Table 2 presents a typical set of blanks and controls for QC on a field
sampling trip when 50 water samples are to be collected.
7.2 Quality Assurance
Both internal and external QA procedures are to followed. Internal QA
procedure assure the continuity and consistency of the data. External QA
procedures (interlaboratory checks) verify or dispute the accuracy of the
data being generated in the primary laboratory.
7.2.1 Internal Quality Assurance
7.2.1.1 Supervision and Monitoring of Activities
There are three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This person
must be aware of their actions, observe events which may effect the data, and
maintain appropriate records. At the second level, the chemist's supervisor
monitors their daily activities, reviews the notebook, checks data and calcula-
tions, and assists in "troubleshooting" problems. At the tertiary level, a
QA coordinator interviews all personnel on the project. The interviews cover
the operations they perform (precisely), the data they obtain, a spot-check
of their calculations, and any problems they have had.
192
-------
Table 2. WATER QC SAMPLES FOR METALS ANALYSIS--
TYPICAL SAMPLING TRIP (50 SAMPLES)
Type
Lab Blank
Lab Control
Field Blank
Field Control
Number
5
5
5
5
o
Comments
Store at 4°C in refrigerator
to be used for field samples
Store with Lab Blanks
Accompanies field samples
from lab to field & return
Same as field blanks.
All preserved at pH< 2 with HN03
193
-------
Documentation
Chain of Custody
From the initial preparation of a sample container through reporting of
the analytical results, each sample is accompanied by a chain of custody
sheet. Each person signs in the time of receipt, operations performed, and
transmittal of the sample. This record is important for tracing a contaminant,
bad standard, or some other problem.
Sampling Protocol Sheets
When a sample is collected, a sampling protocol sheet is filled in which
contains a discrete sample code which identifies project number, area, site,
locations, trip number, sampling period, and sample type. Also included are
sample times, volumes, addresses, meteorology, and other pertinent information.
Where appropriate, a map is made to precisely identify the location.
Sample Log
Upon return from a sampling trip, each sample code is entered into a
sample log book. This log is updated as samples proceed through workup and
analysis. Thus, at a glance, project personnel can tell the status of each
sample and find out how many are at different stages in the analytical protocol.
7.2.2 External Quality Assurance
Samples will be analyzed by a designated QA laboratory. Samples, controls,
and blanks will be shipped directly from the field to the QA laboratory for
analysis. They will report the results to the primary laboratory for correla-
tion with the primary data.
7.2.2.1 Selection of Samples for QA
Approximately 10% of the field samples (2 minimum) will be collected
randomly in duplicate for shipment to QA laboratory.
7.2.2.2 QC Samples
Along with the QA duplicate samples, at least one QC control and one QC
blank must be included with the QA samples. An example is shown in Table 3
for a trip collecting 50 samples.
7.2.2.3 Sample Codes
All samples shipped to the QA laboratory should be encoded so that the
laboratory cannot discern between samples and controls. This "blind" protocol
will add validity to the results.
194
-------
Table 3. SAMPLES TO BE COLLECTED AND SHIPPED TO QA
LABORATORY FOR A TYPICAL SAMPLING TRIP
Sample Number Comments
Duplicate Sample 5 Random selection unless
prior information stratifies
subjects
Field Blank 1 Ship with samples
Field Control 1 Ship with samples
195
-------
7.2.2.4 Shipping
Samples should be shipped directly to the QA laboratory by an appropriate
air carrier (e.g. Federal Express, Eastern Spint) in well insulated and
packed cartons.
Adapted from Environmental Protection Agency, Environmental Monitoring and
Support Laboratory, Cincinnati, OH, "Methods for Chemical Analysis of Water
and Wastes", EPA publication No.-EPA-600/4-79-020, March, 1979.
* Indicates section not in original protocol. Added here for application to
this research project.
196
-------
ANALYTICAL PROTOCOL: VOLATILE HALOGENATED HYDROCARBONS IN
BEVERAGES AND FOODSTUFFS (FDA)
The target compounds will be analyzed from beverages and foods as
obtained from the food sources used by the study volunteers. Specific items
to be obtained will be determined from the individual diaries maintained for
each respondent during the sampling period. The collections and compositing
of food and beverage samples will be carried out in accordance with the FDA
protocols included herein (Attachment A). The collection of samples will be
the responsibility of RTI; compositing will be carried out by Howard Univer-
sity personnel under separate contract. All samples will be analyzed by
FDA/Washington as indicated in the following letter of intent (Attachment
B). The addition of an analytical chemist on loan from RTI to FDA (6 months)
will allow for the implementation of a full analytical program for the Phase
I samples.
197
-------
ATTACHMENT A
198
-------
ATTACHMENT A
INSTRUCTIONS FOR COLLECTION OF FOOD PRODUCTS
Food items are to be purchased as listed in Composites I, II, X and
XII of the Shopping/Compositing Guide for the Northeast region.
Attempt to collect each market basket on a one stop basis. Where
certain food items are not available at one store, visit as many stores as
necessary to collect all of the items required under the Shopping/Compositing
Guide. List each store visited.
Use local chains or large independent grocery stores as sample sources.
National chains may be used if local stores are not available, or to supple-
ment collection of items not found at local stores.
Give preference to the sampling of locally produced food items, whenever
possible.
The quantities to be collected for the individual products in the
Shopping/Compositing Guide are sufficient for most analyses to be performed.
If these products are not available in the unit size shown, collect the next
larger size or sufficient smaller units to meet the requirement. Never
collect less than the amount stated.
Where more than one form of a single food item is listed on the Shopping/
Compositing Guide and one form is preferred over the other, it has been
underlined. Collect the preferred form if available, £•§_•, Adult Sub Number
"63 Broccoli fresh or frozen" indicates that fresh broccoli will be collected
if it is available. If fresh broccoli is not available, collect frozen
broccoli.
Collect units with legible coding if the food product bears a code.
Make sure all units bear the same code if more than one unit is collected to
obtain the amount required by the Shopping Compositing Guide.
List the store name and address, date collected, sample number and brand
names for each item on the attached Shopping/Compositing Guide sheets for
each composite.
The food items should then be packaged for shipping as outlined in the
attached section on packaging food products. This should be done as quickly
as possible, maintaining refrigerated or frozen products in that state until
they can be further packaged.
A set of heat scalable polymer pouches will be necessary to protect
certain food items during shipment. A heat sealing device is also necessary.
KAPAK/SCOTCHPAK BRAND scalable pouches are recommended.
Coolers will be necessary to ship frozen or chilled items. Ice and dry
ice will also be necessary. The coolers and mailable packages should be
199
-------
shipped no later than the day following the collection by Federal Express or
similar delivery service to the address listed below. Please call Mr. Entz
at 202-245-1380 immediately upon shipping samples and provide the shipper
and invoice number for future tracking.
Richard Entz
Division of Chemical Technology (HFF-424)
Food and Drug Administration
Room 4818
200 C Street, S.W.
Washington, DC 20204
202-245-1380
Send completed copies of the Shopping/Compositing Guide for each compo-
site and samples of unused packaging materials to Mr. Entz.
200
-------
PACKAGING OF FOOD PRODUCTS FOR TOTAL DIET TYPE COMPOSITES FOR VOLATILES
ANALYSIS (PACKAGING PRIOR TO SHIPMENT TO DCT)
Composite I
Subsample #
Food
Remarks
4
5
6
7
8
9
2
3
Milk
Ice Cream
Cottage Cheese
Processed Cheese
Natural cheese
Butter
Skim milk
Evaporated milk (canned)
Non fat Dry Milk
Original container sealed
in plastic pouch
shipped in cooler on Dry Ice
Original container sealed
in plastic pouch
Shipped in cooler with
dry ice (C02>
Ship in original
container. No refrigeration
necessary
Heatsealable Plastic Pouch: KAPAK/SCOTCHPAK BRAND or equivalent to be used
Bags should be heat sealed, use the extra heavy weight bags.
201
-------
Composite II
Subsample #
Food
Instructions
12,13,14,15,16,17
19,20,21,23,24
25,26,27
18
22
Fish & meats, frozen
and fresh
Canned tuna or salmon
Eggs
Wrap original package in
prewashed aluminum fpil,
seal in plastic pouch
ship in cooler with dry ice.
Ship as is, no refrigeration
necessary.
Seal package in plastic
pouch, wrap with packaging
material to avoid breakage,
ship in cooler on ice.
Prewashed aluminum foil - wash foil with acetone and air dry.
3Heatsealable Plastic Pouch: KAPAK/SCOTCHPAK BRAND or equivalent to be
used. Bags should be heat sealed, use the extra heavy weight bags.
202
-------
Composite X
Subsample #
Food
Remarks
108
109
110
111
112
113
Salad dressing (mayonnaise)
Salad dressing - French
Salad dressing - other
Shortening
Peanut butter
Margarine
Ship in original
container (glass)
with adequate
protection from breakage.
Place in plastic bag.
No refrigeration necessary
Seal original container in
plastic pouch, ship
frozen on dry ice (CO,,)
in cooler.
Heatsealable Plastic Pouch: KAPAK/SCOTCHPAK BRAND or equivalent to be
used. Bags should be heat sealed, use the extra heavy weight bags.
203
-------
Composite XII
Subsample # Food Remarks
123
123
128
125
126
127
Tea
Coffee, ground
Coffee, instant
Cocoa
Cola soft drink
Non-cola soft drink
Ship in original
container.
necessary.
Purchase in
Ship in orij
No refrigeration
glass bottles.
»inal container.
No refrigeration necessary.
129 Water Let 2 gallons run from tap
before collecting 1 gallon
sample in prewashed bottle
with Teflon seal. Fill bottle
to overflowing and seal with
Teflon faced cap. Ship in
cooler with ice and packaging
material to prevent breakage.
204
-------
Shipping Containers Summary
Cooler - (H20) ice:
Subsamples 22, 129
Cooler (C02) Dry Ice:
Subsamples 1,9,4,5,6,7,8,12,13,14,15,16,17,19,20,21,23,24,25,26,27,113
Containers - Packaged to eliminate breaking:
Subsamples 2,3,18,108,109,110,111,112,123,124,125,126,127,128
Canned items need not be specially packaged.
Ship items to:
Richard Entz
Division of Chemical Technology, HFF-424
Food and Drug Administration
200 C St., S.W., Room 4818
Washington, DC 20204
205
-------
PROTOCOL FOR FOOD PREPARATION AND COMPOSITING FOR JOINT FDA/EPA STUDY ON
VOLATILE HALOCARBONS IN FOODS
This draft protocol provides instructions for preparation, compositing,
and packaging of selected foods prior to analysis for volatile halocarbons.
The major goals of this work are to provide an uncontaminated, stable,
homogeneous food composite for analysis. These instructions are based on
the Food and Drug Administration's (FDA) Total Diet Study.
The attached Total Diet Studies Shopping/Composite Guide for composites
I, II, X, and XII will be used as instructions for collection and the amount
of each food item to add to the final composites. Two sets of composites I,
II, X, and XII are to be collected and prepared by personnel from Howard
University in the Washington, DC area during the first or second week of
June and the first week of July 1980. Instructions for sample collection
and storage are attached. These same food items will be collected from
Northern New Jersey in the middle of June, July, and August and be shipped
to the Division of Chemical Technology (DCT). DCT will arrange for transpor-
tation and participate in overseeing of the initial food preparation and
compositing at Howard University.
Specific food preparation and packaging instructions are included with
each of the attached composite sheets. These instructions are intended to
provide guidance for obtaining homogeneous composites with minimum loss of
volatile halocarbons. Written records of specific chopping/homogenization
equipment, homogenization times, food mixing and equipment cleaning procedures,
and changes from this protocol should be maintained and copies provided to
DCT.
206
-------
INSTRUCTIONS FOR SAMPLE PREPARATION, COMPOSITING, AND STORAGE FOR FOOD ITEMS
FROM TOTAL DIET STUDIES - ADULT SHOPPING/COMPOSITING GUIDE
Composite I
Frozen items, (milk, skim milk, ice cream and butter subsamples)
are allowed to soften (ice cream and butter) or thaw (milk, skim
milk) in the original packages.
The cheese should be grated/ground while still hard (cold) and
then let come to near xoom temperature (cottage cheese need not be
ground).
Individual subsamples (food components of the composite) are
weighed in amounts listed in the compositing guide.
These subsamples are then added to the blender and mixed until a
homogeneous mixture is attained.
It may be necessary to mix the weighed portions of butter and
cheese in a polytron head blender until homogeneous before addition
to remainder of composite.
The final composite mixture is placed in the provided 1 quart
mason jar, (prewashed) with aluminum foil liner and two 40 mL
screw cap vials with Teflon lined septa. These composite mixtures
should be labeled and frozen.
Portions of the subsamples are to be stored frozen, in the manner
listed as follows:
Subsample
Number
1
2
9
4-8
Food
Item(s)
Milk
Evaporated
Skim Milk
Ice Cream, Cheeses,
Butter
Nonfat Dry Milk
Storage
Retain 2x1 Quart bottle (Frozen)
Retain 2 x 40 mL vials (Frozen)
Retain 1 x 250 mL (pint) bottle (frozen)
Retain ca. 100 g in 4 oz bottle (frozen)
Retain ca. 100 g in 4 oz bottle
207
-------
Comments
Recommend blender sufficient to hold 2 gallons to composite these
food items.
Specific type of blender (Hobart or Waring, etc.) not important
except that it be able to develop a homogeneous mixture (no settling
out or separation for at least 15 minutes).
All composites/subsamples to be frozen.
Bottles are to have Teflon seal/cap cover or (pre-cleaned) aluminum
foil where necessary (A 02 bottle).
All bottles are to be prewashed (by DCT).
Individual items can be weighed onto aluminum foil (prewashed with
isopropanol) or beakers (prewashed).
Use of a top loading balance accure to 0.1 g is necessary.
208
-------
INSTRUCTIONS FOR SAMPLE PREPARATION, COMPOSITING, AND STORAGE FOR FOOD ITEMS
FROM TOTAL DIET STUDIES - ADULT SHOPPING/COMPOSITING GUIDE
Composite II
Some items (subsamples 12,13,14,15,16,17,21,23,24,25,26,27) will be
cooked before compositing. See instructions for food preparation (Attachment
A). Those items needing cooking are listed in the Shopping/Compositing
Guide under "Diet" as processor. (Note that bacon drippings need not be
retained).
After cooking, individual food items should be ground/homogenized
after bones are removed. Other food items: canned fish (18),
luncheon meat (19), frankfurters (20) and eggs (22) should be
homogenized. Meats may be ground in a Hobart or similar meat
chopper.
Individual items should be weighed prior to addition to a blender
for compositing (Waring commercial blender recommended).
Blend for minimum time necessary to obtain a homogenous mixture.
Composite to be stored frozen in 1 quart jar with prewashed aluminum
cap liner and in 2 x 40 mL screw cap vials.
After homogenization and removal of a portion for the composites,
individual subsample components (ca. 100 g each) are to be stored
in 4 oz bottles with aluminum foil liners or in heat sealed retort
pouches with appropriate labels. These should then be frozen.
Compositing will follow the cooking/homogenization of individual
components as quickly as possible (no intermediate storage needed).
Indvidual samples can then be retained in 100 g quantities and the
remainder discarded.
209
-------
INSTRUCTIONS FOR SAMPLE PREPARATION, COMPOSITING, AND STORAGE FOR FOOD ITEMS
FROM TOTAL DIET STUDIES - ADULT SHOPPING/COMPOSITING GUIDE
Composite X
Frozen items, after thawing sufficiently to be handled, should be
weighed individually according to the compositing guide. Approximately 100
grams should be saved for storage in 4 oz wide mouth jars with washed aluminum
foil liners for lid or 125 mL bottles. The composite can be blended using a
(Hobart or similar) chopper/blender and placed in a prewashed 1 quart jar
and 2 x 40 mL Teflon sealed vials. All composites and subsamples should be
frozen after packaging.
NOTE:
Chunky style peanut butter (#112) may require blending to chop the
peanuts.
210
-------
INSTRUCTIONS FOR SAMPLE PREPARATION, COMPOSITING, AND STORAGE FOR FOOD ITEMS
FROM TOTAL DIET STUDIES - ADULT SHOPPING/COMPOSITING GUIDE
Composite XII
Tea and coffee are to be prepared with the supplied drinking water
according to the instructions at the bottom of the compositing
guide. Other components are weighed and added in a blender,
homogenized and the resulting composite stored in 2 x 1 quart
bottle and 2 x 40 ml vials and refrigerated.
Individual components should be stored as follows: Tea leaves
(123), coffee (124), cocoa (125), coffee instant (128), in 4 oz
bottles (refrigeration not necessary).
The drinking water and cola soft-drinks should be stored in the
provided Teflon sealed 1 quart and 125 mL bottles respectively and
refrigerated.
211
-------
ATTACHMENT B
212
-------
DEPARTMENT OF HEALTH. EDUCATION. AND WELFARE
PUBLIC HEALTH SERVICE
FOOD AND DRUG ADMINISTRATION
WASHfrlGTON. D.C. 20204
February 25, 1980
Lance Wallace, Ph.D.
EPA RD-680
401 M Street, SW
Washington, DC 20460
Dear Dr. Wallace:
As suggested in our recent meeting with Dr. Jelinek and Mr. Burke, we are
providing you with our thoughts on now the food of Northern New Jersey
residents mignt be collected, prepared and analyzed for certain organic
contaminants.
As previously pointed out, the problems associated with the collection and
transportation of the actual table-ready diets of 9 families over a period of
several months appear to be overwhelming. Additionally, this matter was
discussed with our Epideniology Unit, whicn concluded that the only practical
and statistically valid approach would be to sample one general diet for the
area, such as tnat used oy FDA in its Total Diet studies. A listing of the
foods in the total diet for the Northeast is enclose;!.
If this approach is selected, you might consider the analysis of Composites I,
II, X and XII. As can be observed, these for the most part comprise the fatty
food and beverage portions.
Another very important reason for recommending a general (vs. individual) diet
is that it is the roost practical in an analytical sense. A changing matrix
would result in an analytical nightmare, as recovery/validation studies would
be required for most samples. On the other hand, the selection of a general
diet would necessitate but a single investigation for each diet composite.
You also requested an estimate of cost, time frames, expected acconplishments,
etc., if FDA were to carry out- the food portion of the pilot study.
For the group of compounds including chloroform, carbon tetrachloride, tri-
and tetrachloroethylene, methyl chloroform, 1,2-dichloroethane, ethylene
dibrcmide and bromodichlorcmetliane, approximately 4 person months would be
needed to adapt/validate our methodology for the 4 food composites. If 10 diet
samples were collected (a total of 40 analytical samples), analysis would
213
-------
require about 3 person months. If the chlorobenzenes, vinyl chloride (VC) and
vinylidene chloride (VC1-) were included, trie regjircd time and resources
would roughly double. Analysis for the chlorobcnzanes necessitates a different
procedure, while additional methodology would have to be developed for VC and
VC1-. The same would be true for benzene.
We are prepared to absorb part of the cost of the study, as the investigation
would provide us useful information. In addition, wa would need to add a
chemist to our staff to carry out the bulk of the work. If limited to the
first group of chemicals, it is estimated that the pilot project could be
completed in about 6 months after start-up. A total of a year or more would be
needed if chlorobenzenes, VC, VC1? and benzene wer_ included.
Items such as sample collection, preparation and shipping are not included in
the time/resource estimates. It is assumed that these activities would be
carried out or arranged for by KTI, your contractor for the other phases of
the project.
We will be happy to discuss these matters further at your convenience.
Sincerely yours,
../>
P. Lombardo
Chemical Industry Practices Branch
Division of Chemical Technology
Bureau of Poods
214
-------
ANALYTICAL PROTOCOL: POLYNUCLEAR AROMATIC HYDROCARBONS
FROM HOUSEHOLD DUST
The determination of the target PNAs from household dust will involve
application of the EMSL (Cincinnati) analysis protocol for PNAs from water.
Dust samples will be collected on Teflon filters, (25 mm diameter) using the
modified vacuum cleaner described for metal analysis. Each Teflon filter
will be halved, one half to be used for organochlorine pesticides and PCB
determination, the other half for PNAs analysis. The assay specifications
(range, sensitivity, etc.) are as delineated in the EMSL procedures.
215
-------
ANALYTICAL PROTOCOL: SAMPLING AND ANALYSIS OF ARSENIC, CADMIUM,
AND LEAD IN HOUSE DUST (RTI)
1.0 Principle of Method
The analysis of arsenic, cadmium, and lead in house dust is carried out
using atomic absorption spectrophotometry. Increased sensitivity is achieved
by atomizing the metal in a graphite furnace with continuous background
correction. Arsenic determinations are performed on solutions containing
1000 ppm nickel.
2.0 Range and Detection Limit
The minimum detection limit (MDL) and range for the metal assays in
house dust are shown below (assume 25 mg sample weight).
Metal MDL Max. Cone.
Arsenic 0.50 |Jg/g 100.0 [Jg/g
Cadmium 0.50 100.0
Lead 0.50 400.0
Samples containing higher metal concentrations may be analyzed by
suitable dilution with 1.0% nitric acid. Dilution for arsenic determinations
is made with 1.0% nitric acid containing 1000 ppm.
3.0 Interferences
No known chemical or spectral interferences exist in the analysis of
arsenic, cadmium, or lead in house dust.
4.0 Precision and Accuracy
The precision and accuracy associated with these analyses is a function
of sample metal concentration. At the detection limit, the total measurement
error is ± 100%. Based on the results of a previous study (1), the metal
analyses are performed with the following precision (relative standard
deviation) and the total analysis error (estimated) is also given (2).
Estimated
Precision (%RSD) Accuracy (%RE) Total Error (%)
10 10-20 30-40
10 10-20 30-40
10 10-20 30-40
216
-------
5.0 Apparatus and Reagents
A commercially available stock solution containing 1000 ppm metal is
used for the preparation of the calibration standards. The concentrated
nitric acid is reagent grade quality and the deionized water used in this
study will be prefiltered and subjected to the action of an activated carbon
cartridge and two sequential ion exchange units.
The glassware used for sample workup and the preparation of the calibra-
tion solutions must be subjected to a nitric acid protocol.
All volumetric flasks and beakers should be soaked overnight in 20%
nitric acid, rinsed with deionized water, soaking for an additional 15-18
hours in a 5% nitric acid bath, followed by a copious deionized water rinse.
The flasks are completely filled with 0.5% nitric acid and stored in this
manner. Prior to use, each flask is emptied and rinsed well with deionized
water. Pipets are soaked in 5% nitric acid, rinsed well with deionized
water, air-dried, and stored in a clean, dust-free environment.
Sample cups for the graphite furnace auto sampler may be made of polysty-
rene or Teflon. The former type requires overnight soaking in 1% nitric
acid and followed by rinsing with deionized water. The latter type may be
soaked overnight in 2p% nitric acid, rinsed, and dried in a 105°C oven.
Nickel chloride hexahydrate is used for adjusting the nickel concentra-
tion to 1000 ppm in solutions slated for arsenic analysis.
6.0 Procedure (3)
6.1 Collection of Samples
House dust samples are collected on a filter medium (glass fiber,
Teflon, etc.) with a modified, portable vacuum cleaner. The sample plus
filter is placed in a Zip-Loc bag, labelled, and all pertinent information
recorded on a protocol sheet. The collection device is cleaned after every
sampling with a canister of pressurized Freon (Dust-Off).
6.2 Extraction, Cleanup, and Concentration
The Zip-Loc bag containing the filter plus collected dust is placed in
a dessicator (less than 50% relative humidity), equilibrated overnight and a
weight recorded for the filter plus sample. This is accomplished by cutting
the creases of the bag and scrapping the dust material onto a sheet of tared
powder paper.
217
-------
The net sample weight is determined by subtracting the tare filter
weight. Each collection (filter plus dust) is placed in a 3-dram vial with
a Teflon- lined cap and allowed to stand overnight under an accurately
delivered 10.0 ml volume of 1% HNCy The mixture is heated at 50°-65°C for
45 minutes and the digestion finished by treatment with 0.2 ml of 30% I^C^
at 50-65°C for 15 minutes. The vial is capped during this operation. A
blank filter is carried through the same workup. Each digest is clarified
by passing the supernatant through a plug of acid-washed glass wool and
stored in a 2-dram vial with a Teflon- lined cap until analyzed.
6.3 Instrumental
A Perkin-Elmer Model 403 Spectrophotometer, equipped with a HGA-2000
furnace attachment with deuterium background correction is used for this
analysis. An electrodeless discharge lamp is used as the light source and
the furnace atomization response traced on a Perkin-Elmer Model 056 recorder
AS-1 Autosampler may be used to increase throughput and/or to improve peak
reproducibility and sensitivity.
Arsenic: Wavelength - 193.7 nm
Gas Interrupt (N9) - Auto
Furnace Cycle Conditions -
Dry: 200°C for 20 sec.
Char: 1200°C for 30 sec.
Atomize: 2500°C for 8 sec.
Injection Volume - 20 (Jl
Cadmium: Wavelength - 228.8 nm
Gas Interrupt (N ) - Manual
Furnace Cycle Conditions -
Dry: 200°C for 20 sec.
Char: 400°C for 30 sec.
Atomize: 2000°C for 8 sec.
Injection Volume - 20 (Jl
Lead: Wavelength - 217.0 nm
Gas Interrupt (N2) - Auto
218
-------
Furnace Cycle Conditions -
Dry: 200°C for 20 sec.
Char: 500°C for 20 sec.
Atomize: 2000°C for 8 sec.
Injection Volume - 20 pi
6.4 Qualitative and Quantitative Analysis
6.4.1 Qualitative Analysis
N/A
6.4.2 Quantitative Analysis
The instrument is calibrated with a four aqueous standards in 1.0%
nitric acid, a reagent blank, and a flask filter subjected to the same workup
as samples.
Calibration Range -
Arsenic - 0.0 to 5.0 ng/20 pi
Cadmium - 0.0 to 5.0 ng/20 pi
Lead - 0.0 to 20.0 ng/20 pi
bx
An exponential of the form y = Ae -M provides the best representation
of the analytical curve. The values of the x, y calibration pairs are
entered into a Monroe Calculator Model 1880 programmed to regress the data
to the exponential and to provide values for the constants A, b, and M.
Sample peak heights are measured manually and expressed in units of
millivolts. The calibration constants A, b, and M are entered into the
storage banks of a Texas Instrument Calculator Model 57 and the metal con-
centration results obtained by keying in peak height data. Sample peak
measurements and concentration results are recorded on a calculation work-
sheet.
hv
y = Ae - M, ng/20 pi
y = y (uncorr. metal cone, in sample) - y (metal cone, in matrix)
bx
yg = Ae S - M
bx
. m M
y = Ae - M
'm
219
-------
bx bx
y = (Ae S - M) - (Ae m - M)
bx bx
. / s ms
y = A(e - e )
Units Conversion: ng/20 |Jl •* pg/gm
ng OOO n3 Mg
20 pi ml UOOO ng/ W gm/ 20 W gm
V = volume of dust digest, ml
W = weight of dust sample, gm
bx bx
y = (V/20W) A(e S - e m) ,
y = metal concentration in sample, (Jg/gm,
x = sample peak height, mv,
S
y = uncorrected metal concentration in sample,
S
x = matrix peak height, mv,
y = matrix blank (metal concentration in dust digest due to impurities
in filter medium) ,
7 .0 Quality Assurance Program
An on-going quality assurance program is required to assure the data
quality. Quality control (QC) procedures determine artifacts, losses, etc.
through a system of blanks and controls. Quality assurance (QA) procedures
monitor the execution of the procedure and check data interpretations and
calculations .
7.1 Quality Control
7.1.1 Field Controls
Prior to field sampling, several control dust collections (10% of antici-
pated number of field samples) are obtained. Each dust sample is mixed well
and divided into two portions. One aliquot is placed in a container identical
to that used for field samples, sent to the site, and subjected to the same
handling and storage conditions as field samples. The other aliquot is
stored at RTI in a dust-free environment. On receipt of samples at RTI , both
portions of the control dust collections are worked up and analyzed as a part
220
-------
of each dust analytical run. Within the precision of the assay, the difference
in calculated metal concentrations of the two control dust aliquots is a
measure of the contamination/loss during field storage and transit to RTI.
7.1.2 Internal Quality Control
7.1.2.1 Calibration Standards and Blanks
The instrument is calibrated before each analytical run with four
standard solutions and a reagent blank. Evidence of contamination or instru-
ment malfunction is evident at this time. Such problems are resolved before
initiating sample analysis.
7.1.2.2 Conditioning of Graphite Tube
Before each analytical run, the graphite tube is conditioned by inject-
ing 10 to 20 20 |Jl aliquots of one of the calibration standards. This
operation insures acceptable precision during sample analysis.
7.1.2.3 Duplicate Injections
Reproducibility of peak response is continuously monitored during
sample analysis. All standard and sample solutions receive two successive
injections into the graphite furnace. Signal agreement between the dupli-
cate injections is evaluated according to the following criterion:
First Signal, % Maximum Permissible Permissible Range of
% of Full Scale Variation (% MPV) Second Signal, % of Full Scale
90 ± 4% 86-94
80 ±5 76-84
70 ±6 66-74
60 ±7 56-64
50 ±8 46-54
40 ± 10 36-44
30 ± 13 26-34
20 ± 20 16-24
10 ± 30 7-13
5 ±60 2-8
2 ±100 0-4
If the second injection gives a signal which falls outside the permis-
sable range, a third injection is performed. The peak measurement not in
agreement with the matching pair is discarded.
221
-------
All calibration and sample calculations are based on the mean of the
duplicate determinations.
7.1.2.4 Standard Checks
Instrument performance is monitored during each analytical run. After
the analysis of every 12-16 samples one of the calibration standards is
reinjected into the furnace. The standard which most closely matches the
sample peak heights is selected as the check solution. A metal concentration
is calculated for the check standard, based on its peak height during the
calibration run. Similar calculations are carried out for each check response
and the observed changes in metal concentration expressed in terms of standard
deviation units (SDU),
SDU =
(Calibration Value-Check Value)100
(Calibration Value)(% RSD)
The analysis is under control when the SDU <2.0. Standard checks which
indicate a variation in peak response greater than 2.0 SDU are unacceptable.
In this event, the graphite tube is changed, conditioned, and the system
recalibrated. Quality control charts are graphed to show this change in
instrument performance with time.
7.2 Quality Assurance
Both internal and external QA procedures are to be followed. Internal
QA procedures assume the continuity and consistency of the data. External QA
procedures (interlaboratory checks) verify or dispute the accuracy of the
data being generated in the primary laboratory.
7.2.1 Internal Quality Assurance
7.2.1.1 Supervision and Monitoring of Activities
There are three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This person
must be aware of their actions, observe events which may effect the data, and
maintain appropriate records. At the second level, the chemist's supervisor
monitors their daily activities, reviews the notebook, checks data and calcula-
tions, and assists in "troubleshooting" problems. At the tertiary level, a
QA coordinator interviews all personnel on the project. The interviews
222
-------
cover the operations they perform (precisely), the data they obtain, a
spot-check of their calculations, and any problems they have had.
7.2.1.2 Documentation
Chain of Custody
From the initial preparation of a sample container through reporting of
the analytical results, each sample is accompanied by a chain of custody
sheet. Each person signs in the time of receipt, operations performed, and
transmittal of the sample. This record is important for tracing a contiminant,
bad standard, or some other problem.
Sampling Protocol Sheets
When a sample is collected, a sampling protocol sheet is filled in
which contains a discrete sample code which identifies project number, area,
site, locations, trip number, sampling period, and sample type. Also included
are sample times, volumes, addresses, meteorology, and other pertinent
information. Where appropriate, a map is made to precisely identify the
location.
Sample Log
Upon return from a sampling trip, each sample code is entered into a
sample log book. This log is updated as samples proceed through workup and
analysis. Thus, at a glance, project personnel can tell the status of each
sample and find out how many are at different stages in the analytical
protocol.
Instrument Log
Each sample analysis is logged into a notebook, detailing analysis
conditions.
7.2.2 External Quality Assurance
Samples will be analyzed by a designated QA laboratory. Samples,
controls, and blanks will be shipped directly from the field to the QA
laboratory for analysis. They will report the results to the primary labora-
tory for correlation with the primary data.
7.2.2.1 Selection of Samples for QA
Approximately 10% of the field samples (2 minimum) will be collected in
duplicate for shipment to the QA laboratory. This selection process will be
random with the following restrictions:
223
-------
(1) The donor must consent to the additional collection
(2) If any stratification of donors is known, purposive selection of
QA donors may be used to get representative samples (e.g.,
occupationally exposes vs. "normal" individuals or upwind vs.
downwind residents).
7.2.2.2 QC Samples
Along with the QA duplicate samples, at least one QC control and one QC
blank must be included with the QA samples. An example is shown in Table 1
for a trip collecting 50 samples.
7.2.2.3 Sample Codes
All samples shipped to the QA laboratory should be encoded so that the
laboratory cannot discern between samples and controls. This "blind" protocol
will add validity to the results.
7.2.2.4 Shipping
Samples should be shipped on dry ice directly to the QA laboratory by
an appropriate air carrier (e.g., Federal Express, Eastern Spint) in well
insulated and packed cartons.
8.0 References
8-1 "Epidemiologic Study Conducted in Populations Living Around Non-Ferrous
Smelters", Final Report for Contract No. 68-02-2442 (in preparation).
8-2 McFarren, E. F., Lishka, R. J., and Parker, J. H., Criterion for Judging
Acceptability of Analytical Methods, Anal. Chem., 42(3), 358 (1970).
8-3 Handy, R. W., et al., "Analysis of Housedust for Trace Metal Consent by
Atomic Absorption Spectrophotometry", Paper No. 33 presented at the
31st ACS Southeastern Regional Meeting at Roanoake, VA, Oct. 24-26,
1979.
224
-------
Table 1. SAMPLES TO BE COLLECTED AND SHIPPED TO QA LABORATORY
FOR A TYPICAL SAMPLING TRIP
Sample Number Comment
Duplicate sample 5 Random selection unless
prior information strati-
fies subjects
Field blank 1 Ship with samples
Field control 1 Ship with samples
225
-------
ANALYTICAL PROTOCOL: ORGANOCHLORINE PESTICIDES AND PCBs IN
HOUSEHOLD DUST
This protocol is identical to "Organochlorine Pesticides and PCBs in
Drinking Water (Method 608)" except for sampling and sample extraction.
Dust samples will be collected using the modified vacuum cleaner with
filter described in the protocol for metals in dust. Each filter is halved
(clean surgical scissors) and the filter half designated for pesticides/PCBs
analysis is immersed in ^ 5 ml toluene (glass distilled) and sonicated
(ultrasonic bath) for 5 minutes. The toluene is decanted, and a fresh
portion is added and the Bonification process is repeated. A further repeti-
tion provides ^ 15 ml toluene extract. The combined extracts are concentrated
to "v- 1 ml via the K-D apparatus (Method 608), and further concentrated to
dryness via a gentle stream of nitrogen (using a 3-ball Snyder column -
balls removed - to minimize sample losses). The residue is dissolved, with
sonication, in ^ 1 ml hexane and treated according to the procedures outlined
in Method 608.
226
-------
APPENDIX B
ANALYTICAL PROTOCOLS FOR TOXIC CHEMICALS IN BODY BURDEN SAMPLES
Page
1. Sampling and Analysis Procedure for Human Breath Samples (RTI). . . 228
2. Sampling and Analysis for Benzene in Blood 262
3. Sampling and Analysis of Volatile Purgeable Halogenated Hydrocar-
bons in Human Blood Serum and Urine (U. Miami) 272
4. Sampling and Analysis of Purgeable Halogenated Hydrocarbons
in Blood (RTI) 281
5. Sampling and Analysis of Arsenic, Cadmium and Lead in Whole
Blood (RTI) 293
6. Analysis of Human Serum and Urine for Extractables (HERL-RTP) . . . 303
7. Sampling and Analysis of Extractable Halogenated Organics in
Blood and Urine (RTI) 322
8. Polynuclear Aromatic Hydrocarbons in Blood and Urine (RTI) 338
9. Sampling and Analysis of Purgeable Halogenated Organics in
Urine (RTI) 352
10. Sampling and Analysis of Arsenic, Cadmium, and Lead in Urine (RTI). 364
11. Sampling and Analysis of Extractable Halogenated Organics
in Hair (RTI) 374
12. Sampling and Analysis of Polynuclear Aromatic Hydrocarbons
in Hair (RTI) 391
13. Sampling and Analysis of Arsenic, Cadmium, and Lead in Scalp
Hair (RTI) 404
227
-------
ANALYTICAL PROTOCOL: SAMPLING AND ANALYSIS PROCEDURE FOR HUMAN BREATH
SAMPLES (RTI)
1.0 Principle of Method
The breath sample is collected on a Tenax GC cartridge using a specially
designed spirometer and low hydrocarbon air ("ultrapure") or equivalent (1).
The Tenax cartridge is then dried over CaSO, and analyzed by thermal desorp-
tion into a gas chromatograph (WCOT glass column)-mass spectrometer [(GC/MS),
Figure 1].
2.0 Range and Limits of Detection
The linear range for the analysis of volatile organic compounds depends
upon two principal features. The first is a function of the breakthrough
volume (Table 1) of each specific compound which is trapped on the Tenax GC
sampling cartridge and the second is related to the inherent limits of
detection of the mass spectrometer for each organic (3,6-9). Thus, the
range and the maximum limit of detection are a direct function of each
compound which is present in the original breath samples. The linear range
for quantitation using glass capillaries on a gas chromatograph/mass spectro-
meter/computer (GC/MS/COMP) is generally three orders of magnitude [5-5,000
ng (5-8)]. Table 2 lists the overall detection limits for some examples of
volatile organics which are based on these two principles (1,8).
3.0 Interferences
For the target compounds in Table 2, no interferences have been observed.
Particular attention must be paid to the preparation of clean collection
devices and the use of appropriate blanks and controls to establish that the
background contaminants have been removed. Otherwise, false positive detec-
tion of chloroform may occur.
4.0 Precision and Accuracy
The reproducibility of this method has been determined to range from
+10 to +30% of the relative standard deviation for different substances when
replicate sampling cartridges are examined (5-11). The inherent analytical
errors are a function of several factors: [1] the ability to accurately
determine the breakthrough volume and its relation to field sampling condi-
tions for each of the organic compounds identified; [2] the accurate
228
-------
PURGE
CAS
ION
CURRENT
RECORDER
CLASS
JET
SEPARATOR
TWO
POSITION
VALVE
THERMAL
DESORPTION
CHAMBER
CAPILLARY
GAS
CHRDMATOGRAPH
HEATED
BLOCKS
EXHAUST
CAPILLARY
TRAP
Figure 1. Analytical system for analysis of organic vapors in ambient air.
229
-------
Table 1. TENAX GC BREAKTHROUGH VOLUMES FOR TARGET COMPOUNDS3
Compound
chloroform
carbon tetrachloride
1, 2-dichloroethane
1, 1, 1-trichloroethane
tetrachloroethylene
trichloroethylene
chlorobenzene
b.p.
(°C)
61
77
83
75
121
87
132
Temperature (°F)
50
56
45
71
31
481
120
1989
60
41
36
55
24
356
89
871
70
32
28
41
20
261
67
631
80
24
21
31
16
192
51
459
90
17
17
24
12
141
37
332
100
13
13
19
9
104
28
241
For a Tenax GC bed of 1.5 x 8.0 cm.
230
-------
Table 2. APPROXIMATE MEASURED LIMITS OF DETECTION AND
QUANTIFIABLE LIMITS FOR SELECTED VAPOR-PHASE ORGANICS IN BREATH
LOD3
Compound
Benzene
Chloroform
1,2-Dichloroethane
1, 1, 1-Trichloroethane
Carbon tetrachloride
Vinylidene chloride
Trichloroethylene
Tetrachloroethylene
Bromodichloromethane
Chlorobenzene
1,1, 2-Trichloroethane
m-Dichlorobenzene
m/^
78
83/85
98/62
97/99
117/119
96/98
130/132
164/166
127/83
112/114
97/99
146/148
Ug/m
0.11
0.11
0.16
0.22
ND
0.16
0.22
0.33
0.33
0.22
0.22
0.27
ppt
35
23
41
42
ND
42
42
49
51
49
42
46
QL
yg/m
0.55
0.55
0.82
1.10
ND
0.82
1.10
1.65
1.65
1.10
1.10
1.37
ppt
176
116
209
208
ND
214
211
245
253
245
208
231
Limits of detection (LOD) was defined as S/N = 4 for ion selected
for quantification. Quantification Limit (QL) was defined as 5 x LOD
or S/N = 20. Limits are based on a collection volume of 20 SL or break-
through volume (70°F), whichever is smaller, for 1.5 xm x 8.0 cm
Tenax GC bed volume and mass spectrometer response to that compound.
231
-------
measurement of the ambient air volume sampled; [3] the percent recovery of
the organic from the sampling cartridge after a period of storage; [4] the
reproducibility of thermal desorption for a compound from the cartridge and
its introduction into the analytical system; [5] the accuracy of determining
the relative molar response ratios between the identified substance and the
external standard used for calibrating the analytical system; [6] the
reproducibility of transmitting the sample through the high resolution gas
chromatographic column; and [7] the day-to-day reliability of the MS/COMP
system (2-13).
The accuracy of analysis is generally +10-30% but depends on the
chemical and physical nature of the compound (3,6,7,8,13).
5.0 Apparatus and Reagents
5.1 Collection and Analysis Devices
5.1.1 Spirometer
The spirometer is diagrammed in Figure 2. The valves in the Douglas
mouth piece must be replaced with Tedlar or similar material. A bubbler
filled with distilled deionized water is placed in-line with the air tank
to humidify the air for subject comfort (1).
5.1.2 Sampling Cartridges
The sampling tubes are prepared by packing a ten centimeter long by
1.5 cm i.d. glass tube containing 8 cm of 35/60 mesh Tenax GC with glass
wool in the ends to provide support (3,12). Virgin Tenax (or material to
be recycled) is extracted in a Soxhlet apparatus for a minimum of 18 hours
each time with methanol and n-pentane prior to preparation of cartridge
samplers (3,12). After purification of the Tenax GC sorbent and drying in
a vacuum oven at 120°C for 3 to 5 hours at 28 inches of water, all the
sorbent material is meshed to provide a 35/60 particle size range. Meshing
and all further cartridge preparation is conducted in a "clean" room.
Cartridge samplers are then prepared and conditioned at 270°C with a purified
helium flow of 30 ml/min for 120 min. Prior to entering the Tenax GC
cartridge the helium is purified by passing through a liquid N9 cooled
IS\
cryogenic trap. The conditioned cartridges are transferred to Kimax (2.5
cm x 150 cm) culture tubes, immediately sealed using Teflon-lined caps and
232
-------
to
ULTRAPURE
AIR TANK
TEFLON
CONNECTORS \/
TEDLAR BAG A
TENAXGC
TEDLAR BAG B
DOUGLAS
VALVE AND
MOUTHPIECE
Figure 2. Schematic of spirometer for collection of breath samples.
-------
cooled. This procedure is performed in order to avoid recontamination of
the sorbent bed (3,13).
Kimax® culture tubes are cleaned as described under Section 6.1.
Teflon liners are washed in Isoclean/water, rinsed with deionized distilled
water, acetone and air dried. Subsequently the liners are soaked in methanol
for 2-3 hr, then air dried and placed into the culture tube caps (which are
already aluminum foil lined).
Calcium sulfate is cleaned by heating in a muffle furnace at 400-
®
500°C. Approximately 2 g is placed in the Kimax culture tube, followed by
glass wool and the thermally conditioned Tenax GC cartridges.
5.1.3 Inlet Manifold
An inlet manifold for thermally recovering vapors trapped on Tenax
sampling cartridges is used and is shown in Figure 1 (2-5).
5.1.4 Gas Chromatograph
A Varian 1700 or a Pye Unicam 102 gas chromatograph is used to house
the glass capillary column and is interfaced to the inlet manifold on the
Varian MAT CH-7 or 1KB 2091 systems, respectively. A mass flow controller
(Tylan) is used to precisely control the carrier gas. Such an analytical
system was presented schematically in Figure 1.
A jet separator is employed to interface the glass capillary column to
the mass spectrometer on the Varian MAT CH-7 GC/MS/COMP or 1KB 2091 systems.
The separator is maintained at 240°C (3,6).
5.1.5 Mass Spectrometer/Computer
A Varian MAT CH-7 or LKB 2091 mass spectrometer capable of a resolution
of 1500-2,000 equipped with single ion monitoring capability is used in
tandem with the Varian 1700 or Pye Unicam 102 gas chromatograph and interfa-
ced to a Varian 620/L or PDP 11/04 computer, respectively (Figure 1).
5.2 Reagents and Materials
All reagents used are analytical reagent grade. All solvents (Burdick
& Jackson) are redistilled before their use.
6.0 Procedure
6.1 Cleaning of Glassware
All glassware is washed in Isoclean/water, rinsed with deionized
distilled water, acetone and air dried. Glassware is heated to 450-500°C
234
-------
for 2 hours to insure that all organic material has been removed prior to
its use.
6.2 Collection of Samples
The subject is seated in a comfortable chair and the mouthpiece height
adjusted to a convenient level. A long spring clamp is used to seal the
air flow from Bag A to the mouthpiece (Fig. 2). A plug is placed in the
mouthpiece opening until the test begins to prevent room air contamination.
Air flow from the air tank is started and when the 50 L Bag A is about half
full, the mouthpiece is attached, the clamp and plug removed and the subject
may begin to breathe on the apparatus. The nose clips must also be in
place at this time. After a minute or two, the Nutech Model 221 sampler
pump (Nutech, Durham, NC) is started with the flow at approximately 7 L/min.
The flow may be adjusted to match the individual subject's respiration
rate. It is useful to furl Bag B using spring clips to avoid using this
bag as an exponential dilutor but retaining the safety factor of the 25 L
bag capacity. After a predetermined volume (20 L) of breath has been
sampled, the test is terminated. The subject is removed from the apparatus
and the nose clips removed from the subject. The Tenax cartridges are
removed and stored in culture tubes. The entire apparatus is then flushed
with pure air to decontaminate it for the next use. This is best done by
successively filling and evacuating the bags. The mouthpiece is sterilized
by placing in alcohol after each use. The Tenax cartridges are desiccated
over CaSO, before analysis by placing ^2 g in the bottom of a culture tube
and covering with glass wool. The cartidge is then sealed in the culture
tube for 2-4 hr.
Previous experiments have shown that the organic vapors collected on
Tenax GC sorbent are stable and can be quantitatively recovered from the
cartridge samplers up to 4 weeks after sampling when they are tightly
closed in cartridge holders and placed in a second container that can be
sealed, protected from light and stored at -20°C [Table 3 (2,3,7,8)].
6.2.1 Deuterated Standards
The use of deuterated compounds provides for an assessment of any
premature breakthrough (and thus reduced collection efficiency) which may
occur if the total vapor-phase organic load exceeds 1/10 of the cartridge
235
-------
Table 3. RECOVERY OF TARGET COMPOUNDS AFTER STORAGE
Compound
Benzene
Chloroform
1, 2-Dichloroethane
1,1,1-Trichloroethane
Carbon tetrachloride
Vinylidene chloride
Trichloroethylene
Tetrachloroethylene
Bromodichloromethane
Chlorobenzene
1,1, 2-Trichloroethane
m-Di chl o rob enz ene
1 day
100
100
100
100
ND
100
100
100
ND
100
100
100
1 wk
107 + 26
83 + 14
100 + 5
87 + 17
ND
92
87 + 17
92 + 1
ND
87 + 3
95 + 3
89 + 15
2 wk
69 + 8
49 + 4
100 + 7
71+7
ND
ND
71 + 7
78 + 4
ND
80+7
92 + 2
85+3
236
-------
capacity during sampling of breath, d -Bromoethane (B.P. 34°C), d0-
D o
tetrahydrofuran (B.P. 65), dg-benzene (B.P. 80), d,--cyclohexene (B.P. 83)
and d -chlorobenzene (B.P. 132) are loaded as a discrete zone onto at least
10% of all Tenax GC sampling cartridges prior to sampling. Using GC/MS/COMP
the exogenous deuterated compounds are differentiated from the endogenous
vapor-phase organics in breath samples.
The addition of these deuterated standards is performed by injecting
1.0 ml air/vapor of the substances onto the "front" end of the cartridge.
An air/vapor mixture stream containing 200-400 ng/mL is generated using
permeation tubes of each compound with a permeation system (5,6,9,10).
The quantity of each substance on the sampling cartridges is determined
by GC/MS/COMP after breath sampling and the percent recovery is compared to
control (unused) cartridges carried to and from the field sampling site and
subjected to the same storage regime. Statistically significant differences
are attributed to premature breakthrough.
6.2.2 Quantification Standards
Unique substances may be added as internal standards during sampling.
Examples are the deuterated compounds listed under 6.2.1. However, the
volume of air sampled is accurately known and thus external standards may
be introduced into the cartridge prior to its analysis. Three standards,
hexafluorobenzene, octafluorotoluene, and iodotoluene are used for the
purpose of calculating RMRs and the levels in human breath. Previous
research has shown that their retention times span the chromatographic
range of analysis (SE-30 coated capillary) and they do not interfere with
the analysis of unknown compounds in human breath samples.
The external standards (300-400 ng) are injected into the sampling
cartridges as a 1.0 ml air/vapor mixture using a gas sampling syringe. The
air/vapor mixture is synthesized using permeation tubes and a permeation
system (5,6,9,10).
6.3 Analysis of Samples
The instrumental conditions for the analysis of volatile organics on
the sorbent Tenax GC sampling cartridge is shown in Table 4. The thermal
desorption chamber and the six port Valco valve are maintained at 270°C.
The jet separator is maintained at 245°. The mass spectrometer is set to
237
-------
Table 4. OPERATING PARAMETERS FOR GLC-MS-COMP SYSTEM
Parameter
Setting
Inlet-manifold
desorption chamber and valve
capillary trap - minimum
maximum
thermal desorption time
He purge flow
GLC
85 m glass WCOT BaC03 SE-30 (0.8-1.0 y film)
carrier (He) flow
separator/transfer line
MS
Varian MAT CH-7
scan range
scan cycle, automatic-cyclic
filament current
multiplier
ion source vacuum
LKB 2091
270°C
-195°C
240°C
8 min
15 ml/min
24-240°C, 4°C/min
M..25 ml/min
245°C
m/z 20 •* 350
1 sec/decade
300 yA
4.0
^4 x 10~6 T
scan range
scan cycle, automatic
filament current
multiplier
ion source vacuum
m/_z 20 •* 500
2 sec total
300 yA
4.5
^4 x 10~6 T
238
-------
scan the mass range from approximately 20-350. The helium purge gas through
the desorption chamber is adjusted to 15-20 mL/min. The nickel capillary
trap on the inlet manifold is cooled with liquid nitrogen. In a typical
thermal desorption cycle, a sampling cartridge is placed in the preheated
desorption chamber and the helium gas is channeled through the cartridge to
purge the vapors into the liquid nitrogen capillary trap [the inert activity
efficiency of the trap has been shown in a previous study (5,11)]. After
the desorption has been completed, the six-port valve is rotated and the
temperature on the capillary loop is rapidly raised (greater than 10°/min);
the carrier gas then introduces the vapors onto the high resolution GC
column. The glass capillary column is temperature programmed from ambient
to 240°C at 4°C/min and held at the upper limit for a minimum of 10 min.
After all the components have eluted the column is then cooled to ambient
temperature and the next sample is processed (3).
6.3.1 Qualitative Analysis
The mass spectral data are processed in the following manner. First,
the original spectra are scanned and the reconstructed ion chromatogram
(RIC) is extracted and examined. The intensity (RIC) is plotted against
the spectrum number using the software package available. The information
will generally indicate whether the run is suitable for further processing,
since it provides some idea of the number of unknowns in the sample and the
resolution obtained using the particular gc column conditions.
If mass conversion of spectral peak times to peak masses has not been
performed on-the-fly during data acquisition by hardware methods than this
function is next performed by software methods (magnetic systems). In
either case the mass conversion is accomplished by the use of the calibration
table obtained prior to sample analysis for perfluorokerosene. In general
the calibration data are sufficient for an entire day's data processing;
however, it is verified every eight hours.
After the spectra are obtained in mass converted form, processing
proceeds either manually or by computer by comparison to a Library (14).
Compound identification can involve various degrees of certainty. These
levels of identification have been defined as follows:
239
-------
Level I Computer Interpretation. The raw data generated from the
analysis of samples are subjected to computerized deconvolu-
tion/library search and compound identification made using
this approach has the lowest level of confidence. In general
Level I is reserved for only those cases where compound
verification is the primary intent of the qualitative analysis
Level II Manual Interpretation. The plotted mass spectra are manually
interpreted by a skilled interpreter and compared to those
spectra compiled in a data compendium. In general a minimum
of five masses and intensities (+5% S.D.) should match be-
tween the unknown and library spectrum. This level does not
utilize any further information such as retention time since
many compounds the authentic compound may not be available
for establishing retention times.
Level III Manual Interpretation Plus Retention Time/Boiling Point of
Compound. In addition to the effort as described under Level
II, the retention time of the compound is compared to the
retention time which has been derived from previous chromato-
graphic analysis. Also the boiling point of the identified
compound is compared to the boiling points of other compounds
in the near vicinity of the one in question when a capillary
coated with a non-polar phase has been used.
Level IV Manual Interpretation Plus Retention Time of Authentic Com-
pounds . Under this level, the authentic compound has been
chromatographed on the same capillary column using identical
operating conditions and the mass spectrum of the authentic
compound is compared to that of the unknown.
Level V Level IV Plus Independent Confirmation Techniques. This
Level utilizes other physical methods of analysis such as
GC/fourier transform/IR, GC/high resolution mass spectrometry,
or NMR analysis. This Level constitutes the highest degree
of confidence in the identification of organic compounds.
240
-------
6.3.2 Quantitation
The quantitation of constituents in breath samples is accomplished
either by utilizing the total ion current monitor or, where necessary, from
extracted ion current profiles. In order to eliminate the need to obtain
complete calibration curves for each compound for which quantitative infor-
mation is desired, the method of relative molar response (RMR) factors
(6-11) is used. Successful use of this method requires information on the
exact amount of standard added and the relationship of RMR (unknown) to the
RMR (standards).
6.3.2.1 RMR Determination
The compounds to be quantified are loaded onto Tenax GC cartridges
using a permeation system (5,6,9,10) or in cases where permeation tubes are
not available the vaporization system shown in Figure 3 is used (6,15).
With the vaporization method helium is purified by passing it through a
cryogenic trap followed by two carbon traps. The standards and substances
to be quantified are prepared in methanol and a 2.0 p£ solution is injected
through the septum of the heated loading tube (250°C). The vaporized com-
ponents are swept onto the Tenax GC cartridge at a rate of 200 ml/min for 6
min (total He 1.2 JH). Because of the low breakthrough volume for methanol
(0.8 S, at 70°F), the majority is passed from the cartridge. This system is
used to load relatively non-volatile compounds with breakthrough volumes
>50 H.
The method of calculating RMRs is as follows:
A ,/Moles .
(1) RMR =
unknown/standard A ^,/Moles . ,
std std
A = system response, height or area determined by integration
or triangulation.
unk = unknown
std = standard
The value of RMR is determined from at least three independent analyses
during analysis of samples (6). Linearity over the dynamic range and an
intercept of zero has been previously described (6,15).
A ,/g ,/GMW .
WMR - unk7 &unk' unk
" A.td>«Std/GMWBtd
241
-------
TENAX GC CARTRIDGE CARBON TRAPS
Z SEPTUM 3-WAY STOPCOCK / I
/ 1 * H*FLOW(30mLAnhi)X I
—new—(1 rt« n—«»*—Wto—
j> \ / LOADING TUBE WRAPPED
\ / WITH ALUMINUM FOIL
TEFLON UNIONS AND HEATING TAPE
Figure 3. Schematic of vaporization unit for loading organlcs dissolved in methanol onto Tenax
GC cartridges.
-------
A = system response, as above
g = number of grams present
GMW = gram molecular weight
A , *GMW . «g ,,
>-oN unk unk 6std
(3) g
unk A ,.,-GMW ,-RMR , . . .
std std unk/std
6.3.2.2 Calculation of Organic Vapor Concentrations in Breath
Since the volume of breath taken to produce a given sample is accurately
known and an external standard is added to the sample, then the weight per
cartridge and hence the concentration of the unknown can be determined. The
approach for quantitating ambient air pollutants in this study requires that
the RMR be determined for each constituent of interest during the analysis
of field samples. Every sixth cartridge is a control cartridge for deter-
mining RMRs for each compound (calibration of instrument, storage and
recovery). This means that when a breath sample is taken, the external
standard is added at a known concentration prior to analysis. It is not
imperative at this point to know what the RMR of each of the constituents in
the sample happens to be. However, after the unknowns are identified then
the RMR can subsequently be determined and the unknown concentration calcu-
lated in the original sample using the RMR. In this manner it is possible
to obtain qualitative and quantitative information on the same sample with a
minimum of effort.
Once the quantity of substance per cartridge has been determined, the
level in breath is given by
ug , • 1000 L
M°
_
m3 • Volume Sampled (L)
7.0 Quality Assurance Program
Both internal and external JQA procedures are to be followed. Internal
QA procedures assure the continuity and consistency of the data. External
QA procedures (interlaboratory checks) verify or dispute the accuracy of the
data being generated in the primary laboratory.
243
-------
7.1 Internal Quality Assurance
7.1.1 Supervision and Monitoring of Activities
There are three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This
person must be aware of his actions, observe events which may effect the
data, and maintain appropriate records. At the second level, the chemist's
supervisor monitors his daily activities, reviews the notebook, checks
data and calculations, and assists in "troubleshooting" problems. At the
tertiary level, a QA coordinator interviews all personnel on the project.
The interviews cover the operations they perform (precisely), the data they
obtain, a spot-check of their calculations, and any problems they have had.
7.1.2 Reagent and Glassware Control
Reagent and glassware control is required in order to minimize contami-
®
nation. Sample containers, glassware, etc. are cleaned with Isoclean ,
rinsed with distilled/deionized water and heat treated at 450-500°C to
insure the removal of all traces of organic compounds.
7.1.3 Sampling Protocol and Chain of Custody
As part of the quality control procedures, sampling protocols and
chain of custody forms are prepared for each sampling cartridge. Examples
of these forms are given in Tables 5 and 6. The fate of each sampling
cartridge is tracked from the time they are prepared until the data has
been reduced to a finished form.
7.4 Blanks, Controls, Standards, and System Performance Samples
7.1.4.1 Blanks
Ten percent of the sampling cartridges from each batch are set aside
to serve as blanks to be analyzed for background contamination. After the
preparation of a set of sampling cartridges, one cartridge is checked for
background prior to their committment to field sampling. Blank (unused)
cartridges travel to the field site returned to the laboratory and stored
along with the field samples at -20°C until ready for analysis.
7.1.4.2 Controls
Ten percent of the sampling cartridges are loaded with the deuterated
compounds listed in 6.2.1. Sampling with control cartridges allow for an
assessment of premature breakthrough if it occurs. Control cartridges are
244
-------
Table 5. FIELD SAMPLING PROTOCOL SHEET - A
Date:
Project No. (
Operator ( 7 ~
Trip No. C^f
Sampler (J"^)
Area ' _ ""
)
~ ~
Address
Period ( )
Site __
location ( _
Sampje Code
L J
»utreb 221 (H)
PC aapt
Pu?ent (P)
Saapli&t ratt (Ixtlt.)
SaapllAt rata (final)
Cad: TiM
Start: Ti»e _
Total: (Bin)
HSA 00
(Ipa) Sanplint rate (inlt.)
(Ipa) Saaplint rate (final)
tad: Tine Count
dps)
Start: TiM _
Total: (sin)
•I/count
Count
Count
Remarks
Volume Air/Cartridge
Tine Temp. Wet. Dry
Rel. Humid * Wind Dir./Speed __
Cloud Odor
Remarks '_
Time Temp. Wet. Dry
Bel. Humid * Wind Dir./Speed __/.
Cloud Odor
Remarks
Time Temp. Wet. Dry
Rel. JHSid % Wind Dir./Speed __/,
Cloud Odor .
Remarks __.
Time Temp. Wet Dry
Rel. Hmnid 1 Wind Dir./Speed __
Cloud Odor
Remarks
245
-------
Table 6. CHAIN OF CUSTODY RECORD
Research Triangle Institute
Analytical Sciences Division
Chemistry and Life Sciences Group
Research Triangle Park. NC 27709
SAMPLE CODE:
s .
* •••
Sample Type:
No. of Containers:
Volume Collected:
Volume Analyzed:
Relinquished
By:
Received
By:
Time
•
Date
Operation Performed (aliquot, ctd. cone
remarks, etc.)
246
-------
analyzed along with other samples and since deuterated compounds are employed
the qualitative and quantitative analysis of these air samples proceeds
unimpeded.
7.1.4.3 Standards for RMR Determination
The compounds listed in Table 2 are loaded (250-450 ng each) onto
Tenax GC cartridges from a permeation system. A minimum of three analyses
is required for determining RMRs for a set of samples which are quantitati-
vely analyzed.
7.1.4.4 System Performance Mixtures
The system performance standards listed in Table 7 are loaded onto
Tenax GC cartridges (using the vaporization method) to determine mass
calibration and intensity and chromatographic performance of the GC/MS/COMP
system.
7.1.2 Sample Analysis
To insure the accuracy and precision of the data acquired instrument
and chromatographic performance are monitored on a daily basis.
7.1.2.1 Instrument Calibration
Calibration of mass and intensity of magnetic systems employs perfluoro-
kerosene. Table 8 lists the tolerances for each mass and intensity which
the mass spectrometer must achieve.
Perfluorotoluene in the performance mixture is employed for determining
instrument stability as related by mass resolution and relative ion abun-
dance under GC conditions (6,15). The masses and intensities listed in
Table 8 are compared to the results obtained on a daily basis and for each
set of samples analyzed.
7.1.2.2 Assessment of Chromatographic Performance
The quality of the chromatography is of utmost importance since the
accuracy and precision of qualitative and quantitative analysis are directly
affected (6,15). Glass capillary columns are evaluated according to the
following criteria:
(1) percent peak asymmetry factor (PAF)
% PAF = | x 100
247
-------
Table 7. GC/MS/COMP SYSTEM PERFORMANCE STANDARDS
Compound Quantity (ng)
Perfluorokerosene 350
Perfluorotoluene 350
Ethylbenzene 300
£-Xylene 300
n-Octane 300
n-Decane 300
1-Octanal 300
5-Nonanone 300
Acetophenone 300
2,6-Dimethylaniline 300
2,6-Dimethylphenol 300
248
-------
Table 8. MASS AND INTENSITY TOLERANCES ACCEPTABLE FOR
CALIBRATION OF MAGNETIC INSTRUMENTS FOR QUANTITATION
Perfluorotoluene*
n/z
69
79
93
117
167
186
217
236
ZI (C.V.)
33 (5)
11 (10)
16 (8)
43 (8)
15 (7)
59 (5)
100 (0)
66 (4)
Perfluorokerosene
m/_z
51
100
119
131
169
181
219
231
21 (C.V.)
39 (10)
22 (8)
100 (0)
89 (5)
58 (4)
62 (6)
24 (5)
29 (8)
To be achieved in the chromatography mode.
249
-------
where B = the area of the back half of a chromatographic peak
F = area of the front half of the chromatographic peak both
measured 10% above baseline
(2) effective Height Equivalent to a Theoretical Plate (HETPeff)
5.54 (X/Y)*
where X = the corrected retention distance for sweep time of the
compound,
Y = chromatographic peak width at 1/2 peak height,
L = column length (mm)
(3) separation number (SN)
D
SN =
where D = the distance between two peaks,
W1'W2 =
(4) resolution (R)
W ,W = widths at 1/2 height
_ 2 A W
l\ — rr ; rr~
where AW = average base width,
W = peak width at base
(5) Acidity and Basicity
. weak base (peak area or height)
Acidity = —r ^—p ; ? . ,-•.
acetophenone (peak area or height)
_ . . weak acid (peak area or height)
Basicity = —r 5J1-7 ; ? . ,- ^
3 acetophenone (peak area or height)
The use of the compounds listed in Table 7 provides information as to the
degree of adsorption and the type of adsorption mechanisms. 1-Octanol and
5-nonanone serves to determine the extent of deactivation of the glass sur-
face (PAF). The acidity and basicity of the glass capillary column are
assessed by the adsorption of weak bases and acids, respectively (6,15).
The resolution and separation number are determined for the compound
pairs ethylbenzene:£-xylene and octane:decane, respectively. HETP is
based on octane. Table 9 lists the minimum performance specifications accept-
able for breath analysis (6). Figures 4-10 depict extracted ion current
profiles used for calculating performance specifications.
250
-------
Table 9. SPECIFICATIONS OF PERFORMANCE FOR GLASS CAPILLARY COLUMN*
to
Parameter Test Compound(s) Value + S.D. (C.V.)
Resolution Ethylbenzene:£-Xylene 1.30 + 11 (8)
Separation No. Octane:Decane 73+6 (8)
% Peak Asymmetry Factor 1-Octanol 239 + 141 (59)
Nonanone 130 + 32 (25)
Acetophenone 260 + 34 (13)
Acidity 2,6-Dimethylaniline:Acetophenone 1.00 + 0.07 (9)
Basicity 2,6-Dimethylphenol:Acetophenone 0.82 + 0.03 (5)
8SE-30 WCOT/BaCOa, 0.48 mm i.d. x 75 m, 1 p film thickness.
-------
IO ID
U>
a
rf
'i
I S> »-
i B Z
N. —
a a
x£
Ul U
C&2
c
c
c:
Scan No.
Figure 4. Extracted ion current profile of a/£ 91 for £-xylene
and ethylbenzene used in calculating resolution.
252
-------
IM
OC QC
pu. u.
• •
BK»-
inr z
e
Scan No.
Figure 5. Extracted ion current profile of m/z 70 and 84 for
1-octanol used for calculating percent peak asymmetry
factor.
253
-------
in
o>
Scan No.
Figure 6. Extracted ion current profile of tt/z 57 for nonanone
used in calculating percent peak asymmetry factor.
254
-------
r
d
u
«
•o
Figure 7. Extracted Ion current profile of m/z 43 for octane and decane used in calculating
separation number.
-------
£
in o
0)
o
I
a
o
4J
0)
u
ce
c
(V
m/z 105
c
r.
c
c
Scan No.
Figure 8. Extracted ion current profile of m/_z 105 for acetophenone
in calculating percent peak asymmetry factor.
256
-------
Cn
-o
Itt.ft-
W7 .
121 _
IW.fr-
122.
100
4:10
4 k
200
8:20
300
12:30
400
16:40
Scan Time
500
20:50
600
25:00
3309.
IC7.032
* 0.!
33Q9.
121.936
* 0.5W
33M.
122.037
k 0.5M
Figure 9- Extracted ion current profile of m/z 107, 121 and 122 of 2,6-dimethylphenol in
mixture.
-------
IM
121 .
K)
tn
00
!••.
122.
100
4:10
200
8:20
300
12:30
400
16:40
Scan Time
01
c
8
•o
500
25:00
600
25:00
199.932
* 9.5W
121.036
* t.see
I22.W7
* t.5W
56128.
Figure 10. Extracted ion current profile of m/z 106, 121, and 122 for 2,6-ditnethyl-
aniline in performance mixture.
-------
7.1.2.3 Sequence of Sample Analysis
A strict step-sequence of analysis is followed. Upon mass and intensity
calibration of the MS system, a Tenax GC cartridge loaded with the perfor-
mance mixture is first analyzed. Following the performance mixture a blank
and an RMR standard mixture is analyzed next, then five samples. The cycle
is then repeated. At the beginning of each day the analysis cycle begins
with the performance mixture, blank and RMR standards. Thus, 30% of the
cartridges analyzed consist of control samples.
7.2 External Quality Assurance
Samples will be analyzed by a designated QA laboratory. Samples,
controls, and blanks will be shipped directly from the field to the QA
laboratory for analysis. They will report the results to the primary
laboratories for correlation with the primary data.
7.2.1 Selection of Samples for QA
All of the field samples will be collected in duplicate for replicate
analyses. Approximately 10% of the duplicates will be shipped to the QA
laboratory. This selection process will be random, unless any stratification
of donors is known. If so, purposive selection of QA donors may be used to
get representative samples (e.g., occupationally exposed vs. "normal" indivi-
duals or upwind vs. downwind residents).
7.2.2 QC Samples
Along with the QA duplicate samples, at least one QC control and one
QC blank must be included with the QA samples. An example is shown in
Table 10 for a trip collecting 50 samples.
7.2.3 Sample Codes
All samples shipped to the QA laboratory should be encoded so that the
laboratory cannot discern between samples and controls. This "blind"
protocol will add validity to the results.
7.2.4 Shipping
Samples should be shipped on dry ice directly to the QA laboratory by
an appropriate air carrier (£•£•, Federal Express, Eastern Sprint) in well
insulated and packed cartons.
259
-------
8.0 References
1. Zweidinger, R., et al., Final Draft Report on Benzene, EPA Contract No.
68-01-3849, 1980, in preparation.
2. Pellizzari, E. D., "Development of Method for Carcinogenic Vapor Analy-
sis in Ambient Atmospheres", Publication No. EPA-650/2-74-121, Contract
No. 68-02-1228, 148 pp., July, 1974.
3. Pellizzari, E. D., "Development of Analytical Techniques for Measuring
Ambient Atmospheric Carcinogenic Vapors", Publication No. EPA-600/2-75-
075, Contract No. 68-02-1228, 187 pp., November, 1975.
4. Pellizzari, E. D., J. E. Bunch, B. H. Carpenter and E. Sawicki, Environ.
Sci. Technol., 9, 552 (1975).
5. Pellizzari, E. D., "The Measurement of Carcinogenic Vapors in Ambient
Atmospheres", Publication No. EPA-600-7-77-055, Contract No. 68-02-
1228, 288 p., June, 1977.
6. Pellizzari, E. D., "Evaluation of the Basic GC/MS Computer Analysis
Technique for Pollutant Analysis", Final Report, EPA Contract No. 68-
02-2998.
7. Pellizzari, E. D. and L. W. Little, "Collection and Analysis of Pur-
geable Organics Emitted from Treatment Plants", Final Report, EPA
Contract No. 68-03-2681, 216 pp.
8. Pellizzari, E. D., unpublished results.
9. Pellizzari, E. D., "Analysis of Organic Air Pollutants by Gas Chromato-
graphy and Mass Spectroscopy", EPA-600/2-77-100, June 1977, 114 pg.
10. Pellizzari, E. D., "Analysis of Organic Air Pollutants by Gas Chromato-
graphy and Mass Spectroscopy", EPA-600/2-79-057, March 1979, 243 pg.
11. Pellizzari, E. D., "Ambient Air Carcinogenic Vapors Improved Sampling
and Analytical Techniques and Field Studies", EPA-600/2-79-081, May
1979, 340 pg.
12. Pellizzari, E. D., J. E. Bunch, R. E. Berkley and J. McRae, Anal.
Chem., 48, 803 (1976).
13. Pellizzari, E. D., J. E. Bunch, R. E. Berkley and J. McRae, Anal.
Lett., 9, 45 (1976).
260
-------
14. "Eight Peak Index of Mass Spectra", Vol. I, (Tables 1 and 2) and II
(Table 3), Mass Spectrometry Data Centre, AWRE, Aldermaston, Reading,
RF74PR, UF, 1970.
15. Pellizzari, E. D., et al., "Master Scheme for the Analysis of Organic
Compounds in Water Part III: Experimental Development and Results",
EPA Contract No. 68-03-2704, March 1980.
Written analytical protocol prepared 5/6/80.
261
-------
ANALYTICAL PROTOCOL: SAMPLING AND ANALYSIS FOR BENZENE IN BLOOD
1.0 Principle of Method
A blood sample is equilibrated at 37°C with an air space of determined
volume until equilibrium is attained. The entire headspace is then purged
into a cryogenic trap which can be placed in line with a gc as a sample loop
and heated. In this manner, the recovery is determined by the partition
between fluid and air and avoids the many artifacts and other problems
introduced by purging (i.e., foaming, precipitation occulsion, and sorbent
background).
2.0 Detection Limit
The range is limited by the limit of detection on one extreme and by
the chromatographic capacity of the capillary on the other or ^-lO4. Minimum
detectable concentration for the method is estimated to be 1.6 |Jg/l (95%
confidence level).
3.0 Interferences
No interferences have been observed, however high levels of other
hydrocarbons in the sample could cause the benzene peak to be obscured.
4.0 Precision and Accuracy
Precision at 500 |Jg/l is 8% relative standard deviation increasing to
33% at 1.8 |Jg/l. Recovery of control samples spiked at 5 Mg/1 was 98%.
5.0 Apparatus and Materials
1. A thermostated two-position, six-port valve with nickel capillary
trap as indicated in Figure 1.
2. A gas chromatograph with flame ionization detector.
3. Thermostated oven (37°C).
4. Glass capillary gas chromatography column with SE30 liquid phase.
5. Glass hypodermic syringes (10 ml) and needles.
6. Silicone rubber septum material.
7. Liquid nitrogen.
8. Ultrapure air (<0.1 ppm total hydrocarbon).
9. Vacuum blood sampling tubes (Venoject KT200SKA, Kimble).
262
-------
CO
VRLVE POSITION A
(SAMPLE PURGE)
VALVE POSITION B
(SAMPLE INJECTION)
CARRIER
GAS
PURGE
GAS ••
• TO GLC
VENT
I
U.
Figure 1. Six-port, 2 position valve for the introduction of headspace samples,
-------
6.0 Procedure
6.1 Collection of Blood Sample
Blood samples are collected from selected participants from a brachial
vein by venipuncture using a 10 ml Venoject tube. These blood samples are
collected by experienced medical personnel using accepted medical proce-
dures .
6.2 Analysis of Samples
Pre-equilibrate a 10 ml glass syringe at 37°C (>30 rain), remove the
needle from the syringe and inject 1.0 ml of blood (sample, standard or
blank) into the syringe which has been sealed around the plunger with sa-
turated lithium chloride. Adjust the volume to 10 ml by filling the syringe
with "ultra pure" air, replace the needle on the syringe and seal it by
inserting into a piece of silicone septum material. Incubate the entire
syringe assembly at 37°C for 20 min. After the incubation, the needle is
removed and the syringe connected to the cryogenic trap via an 18 gauge
needle. The total air space in the syringe is purged through the trap. An
additional 1 ml of air is purged through the trap from another syringe. The
latter step is to prevent sample holdup in the transfer lines to the trap.
At this point, the coolant (liquid nitrogen) is removed from the trap,
the valve rotated and the trap rapidly heated to 175°C. The GC operating
parameters are given in Table 1.
Calibration of the GC is obtained by analyzing blood spiked with known
amounts of benzene and blood blanks under identical conditions to the sample.
6.3 Quantitation
Peak areas of benzene in unknown samples are compared to calibration
curves generated with known amounts of added benzene. This results in the
following relationship:
Concentration of benzene (|Jg/l) = —-——
std
where
S t*le Peak area of the sample,
ggtd is the amount of benzene added to the standard
is the peak area of the standard.
264
-------
Table 1. OPERATING PARAMETERS FOR GC/FID ANALYSIS OF BENZENE
IN BLOOD AND URINE
Parameter
Value
Column
47 M glass WCOT SE-30, BaC03
Helium carrier gas flow rate
50°C initial for 3 min
then 4°/min to 200°C
2.1 ml/min
265
-------
7.0 Quality Assurance Program
An on-going quality assurance program is required to assure the data
quality. Quality control (QC) procedures determine artifacts, losses, etc.
through a system of blanks and controls. Quality assurance (QA) procedures
monitor the execution of the procedure and check data interpretations and
calculations.
7.1 Quality Control
7.1.1 Field Blanks and Controls
Prior to a field sampling trip, enough blanks and controls are prepared
to equal 10% each (2 minimum) of the anticipated number of field samples.
Blanks consist of 10 ml of water in the same type of sampling container as
is used in the field. Controls consists of 10 ml of plasma spiked at 50 ng
with benzene. These blanks and controls are carried to the field and receive
the same handling as the field samples. Workup and analysis of field blanks
and controls is interspersed with the field samples on a regular basis.
This method allows assessment of sample storage stability.
Table 2 presents a typical set of blanks and controls for QC on a field
trip where 50 blood samples are to be collected.
7.1.2 Procedural Blanks and Controls
7.1.2.1 Extraction Blanks
With each set of samples, a procedural blank is run. This consists of
10 ml of prepurged distilled water which is extracted under the same condi-
tions as the samples. These blanks are designed to detect artifacts from
dirty glassware, laboratory atmosphere intrusion, and other sources.
7.1.2.2 GC Procedural Control
At the start of each working day, a standard test mixture is analyzed
to monitor the capillary GC column performance.
Field samples, field controls, field blanks, and procedural blanks are
queued up for analysis such that at least one QC sample is run each working
day. In addition, a standard solution is analyzed each day to serve as a
procedural control and also to update the calibration value. Thus, in a
typical working day, 4 field samples, 1 blank or control, and 1 calibration
standard are run.
266
-------
Table 2. BLOOD QC SAMPLES FOR A TYPICAL SAMPLING TRIP
Sample Type
Number
Comments
Field Blank
Field Control
Lab Blank
Lab Control
5
5
Freeze after preparation,
carry to field, store with
field samples.
Store with field blanks.
Freeze after preparation,
store in same freezer as field
samples will be stored.
Store with Lab Blanks
267
-------
7.2 Quality Assurance
Both internal and external QA procedures are to be followed. Internal
QA procedures assure the continuity and consistency of the data. External
QA procedures (interlaboratory checks) verify or dispute the accuracy of the
data being generated in the primary laboratory.
7.2.1 Internal Quality Assurance
7.2.1.1 Supervision and Monitoring of Activities
There are three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This
person must be aware of their actions, observe events which may affect the
data, and maintain appropriate records. At the second level, the chemist's
supervisor monitors their daily activities, reviews the notebook, checks
data and calculations, and assists in "troubleshooting" problems. At the
tertiary level, a QA coordinator interviews all personnel on the project.
The interviews cover the operations they perform (precisely), the data they
obtain, a spot-check of their calculations, and any problems they have had.
7.2.1.2 Documentation
Chain of Custody
From the initial preparation of a sample container through reporting of
the analytical results, each sample is accompanied by a chain of custody
sheet. Each person signs in the time of receipt, operations performed, and
transmittal of the sample. This record is important for tracing a contaminant,
bad standard, or some other problem.
Sampling Protocol Sheets
When a sample is collected, a sampling protocol sheet is filled in
which contains a discrete sample code which identifies project number, area,
site, locations, trip number, sampling period, and sample type. Also included
are sample times, volumes, address, and other pertinent information. Where
appropriate, a map is made to precisely identify the location.
Sample Log
Upon return from a sampling trip, each sample code is entered into a
sample log book. This log is updated as samples proceed through workup and
analysis. Thus, at a glance, project personnel can tell the status of each
268
-------
sample and find out how many are at different stages in the analytical
protocol.
GC Log
Each sample run by GC is logged into a notebook, detailing analysis
conditions.
7.2.2 External Quality Assurance
Samples will be analyzed by a designated QA laboratory. Samples,
controls, and blanks will be shipped directly from the field to the QA
laboratory for analysis. They will report the results to the primary labora-
tory for correlation with the primary data.
7.2.2.1 Selection of Samples for QA
Approximately 10% of the field samples (2 minimum) will be collected in
duplicate (i.e. , two 10 ml blood samples) for shipment to the QA laboratory.
This selection process will be random with the following restrictions:
(1) The donor must consent to the additional hair collection.
(2) If any stratification of donors is known, purposive selection of
QA donors may be used to get representative samples (£•£• occupa-
tionally exposed vs. "normal" individuals or upwind vs. downwind
residents).
7.2.2.2 QC Samples
Along with the QA duplicate samples, at least one QC control and one QC
blank must be included with the QA samples. An example is shown in Table 3
for a trip collecting 50 samples.
7.2.2.3 Sample Codes
All samples shipped to the QA laboratory should be encoded so that the
laboratory cannot discern between samples and controls. This "blind" protocol
will add validity to the results.
7.2.2.4 Shipping
Samples should be shipped on dry ice directly to the QA laboratory by
an appropriate air carrier (£.£• Federal Express, Eastern Sprint) in well
insulated and packed cartons.
269
-------
Table 3. SAMPLES TO BE COLLECTED AND SHIPPED TO QA LABORATORY
FOR A TYPICAL SAMPLING TRIP
Sample Number Comments
Duplicate Sample 5 Random selection unless prior
information stratifies subject
Field Blank 1 Ship with samples
Field Control 1 Ship with samples
270
-------
8.0 References
1. Zweidinger, R. A., S. D. Cooper, B. S. H. Harris, III, T. D. Hartwell,
R. E. Folsom, Jr., E. D. Pellizzari, A. W. Sherdon, T. K. Wong and
H. S. Zelon, "Measurement of Benzene Body-Burden for Populations
Potentially Environmentally Exposed to Benzene", EPA Contract No.
68-01-3849, Final Report (in preparation).
Revised April, 1980
271
-------
ANALYTICAL PROTOCOL: SAMPLING AND ANALYSIS OF VOLATILE PURGEABLE
HALOGENATED HYDROCARBONS IN HUMAN BLOOD SERUM AND URINE (U. MIAMI)
1.0 Principle of the Method
Human blood serum or urine is analysed for volatile purgeable halogenated
hydrocarbons (VPHH's) by a purge/trap/desorb procedure based on that of
Bellar and Lichtenberg. No extraction or clean-up step is required and the
cost per analysis is reasonable. Each analysis is completed in about 30
minutes.
Successfully analyzing blood and urine is dependent on circumventing
foaming problems. To circumvent or greatly reduce foaming problems, 0.5 mL
of a 1% solution of Dow Corning Antifoam Emulsion B is purged 20 min with
pure inert gas prior to the addition of the blood serum.
A small condenser, attached to the effluent arm of the purging device
also aids in keeping foaming serum out of the Tekmar instrument. The serum,
urine/antifoam mixture is purged for 30 min at room temperature and then at
115°. The elevated temperature presumably denatures the proteins, one of
the causes of foaming; it is also necessary to counteract the inherent
binding capacity of serum for halogenated organic compounds. Iri vitro
spiked compounds can be recovered completely at room temperature or 115°.
Samples obtained from dosed animals (in vivo work) require 115° to effect
complete removal of volatile purgeable halocarbons.
2.0 Range and Detection Limits
Detection limits are:
chloroform - 0.05 Mg/L
carbon tetrachloride - 0.05 M8/L
bromodichloromethane - 0.1 (Jg/L
trichloroethylene - 5 |Jg/L
3.0 Interferences
Co-eluting halogenated compounds present problems with identification
and quantitation. Co-elution can be circumvented by the use of different
columns such as Carbopack C or Tenax GC. There is some concern that chloro-
form is being generated in situ from trichloroacetic acid. This is currently
under investigation.
272
-------
4.0 Precision and Accuracy
One serum sample was analyzed ten times over a two-day period. The
chloroform concentration ranged from 23 to 36 Mg/1 with a mean value of 27
|jg/l and a standard deviation of 4. Accuracy has not been determined.
5.0 Apparatus and Reagents
5.1 Apparatus
1. A Tekmar Model LSC-1 liquid sample concentrator
2. Tracer Model 222 gas-liquid chromatograph (GLC)
3. Hall electrolytic conductivity detector which is operated in
the halide specific mode.
4. Chromatographic column, 2-m x 6.4-mm I.D. glass U-tube containing
n-octane on 100-120 mesh Porasil C packing.
5. A Finnigan Model 4000 gas -liquid chromatograph/mass spectrometer
(GLC/MS) analytical system interfaced to a Tekmar liquid sample
concentrator
6. Both the GLC and GLC/MS systems utilized a hot plate stirrer and a
glyceral bath to heat the sample in the Tekmar purging device.
5.2 Solvents and Reagents
1. Chloroform, pesticide grade, Fisher Scientific Co.
2. Carbon tetrachloride, pesticide grade.
3. Hexane, pesticide grade, Fisher Scientific Co.
4. Trichloroethylene, Aldrich Chemical Co.
5. 1,2-dichloroethane, Aldrich Chemical Co.
6. Bromoform, Aldrich Chemical Co.
7. Bromodichloromethane, Columbia Organic Chemical Co.
8. Dibromochloromethane, Columbia Organic Chemical Co.
9. Dow Corning Antifoam Emulsion B, Fisher Schientific Co.
10. n-octane on 100-120 mesh Porasil C chromatographic packing, Supelco
Inc.
6.0 Procedure
6.1 Collection and Handling of Samples
Blood is collected in vacuum tubes suitable for use in gc applications,
®
viz., Venoject L428 (Kimble) tubes (see page 208). Blood plasma is separated
from the cells and stored at 4° prior to analysis.
273
-------
First morning urine samples are collected in wide-mouthed bottles,
sealed and stored at 4° until analysis.
6.2 Preparation of Standards and Samplers
6.2.1 Preparation of Standards
6.2.1.1 Two ml each of carbon tetrachloride, dibromochloromethane and
bromoform and 1 ml each of chloroform, trichloroethylene, bromodichloromethane
and 1,2 dichloroethane are diluted with hexane to final volume of 100 mL
(Solution 1). Appropriate amounts of 1,1,1-trichloroethane, 1,1,2-trichloro-
ethane, vinylidene chloride, and chlorobenzene can also be diluted to prepare
standards.
6.2.1.2 One ml of solution 1 is quantitatively diluted to 100 ml with
hexane (solution 2), and a convenient working standard prepared by diluting
0.1 ml of solution 2 to 25 ml with hexane. Use of 5 [Jl of this solution
leads to acceptable peak heights when the Hall detector attenuation is 10 x
8.
6.2.1.3 A standard curve is obtained by using three hexane dilutions
of solution 2: the working standard, 0.1 ml diluted to 50 ml (for 10 x 4
attenuation) and 0.2 ml diluted to 25 ml (for 10 x 16 attenuation). Although
solutions 1 and 2 are stable at room temperature, fresh working standards
must be made daily.
6,2.2 Procedure for Blood Serum and Urine
6.2.2.1 One ml of 1% aqueous antifoam is added to the 5 ml purging
device and the sample concentrator is operated in the Trap Bake Mode for 20
min while the trap temperature was 200°. The trap is then cooled to the
ambient temperature.
6.2.2.2 By means of a gas tight syring, 0.5 ml of serum or 2 ml of
urine is introduced into the purging device and a purge flow-rate of 10
ml/min was started. During analyses the stripping gas volume is 0.3 L.
6.2.2.3 The lower portion of the purging device is immersed in a 115°
stirred glycerol bath for 30 min. To prevent steam contamination of the
Tenax/silica gel trap, a small glass vapor trap or interceptor is placed
between the purging device and the adsorbent trap.
274
-------
6.2.2.4 After the purge/trap period is complete, the adsorbed compounds
are desorbed and transferred to the analytical column (60°) by heating the
trap at 150° for 6 min.
6.3 Analysis
6.3.1 The GLC operating conditions include: a nitrogen carrier gas
flow-rate of 30 ml/min, an inlet temperature of 140°, and a transfer line
temperature of 210°. The GLC column is temperature programmed 7°/min to
140°.
6.3.2 The Hall detector furnace is maintained at 900° with a hydrogen
flow-rate of 40 ml/min and a solvent (1:1 n-propanol:distilled water) flow
of 0.4 ml/min.
6.4 Qualitative Identification
Identifying of compounds for LSC/GLC/HECD system is done by retention
times. This is confirmed by LSC/GLC/MS methodology. The confirmations are
based on relative retention data and mass fragmentation data.
6.5 Quantitation
LSC/GLC/HECD chromatograms are quantified by using the external standard
method described in section 6.2.1.3.
*7.0 Quality Assurance Program
An on-going quality assurance program is required to assure the data
quality. Quality control (QC) procedures determine artifacts, lossed, etc.
through a system of blanks and controls. Quality assurance (QA) procedures
monitor the execution of the procedure and check data interpretations and
calculations.
7.1 Quality Control
7.1.1 Field Blanks and Controls
Prior to a field sampling trip, enough blanks and controls are prepared
to equal 10% each (2 minimum) of the anticipated number of field samples.
Blanks consist of 10 ml of water in the same type of sampling container as
is used in the field. Controls consist of 10 ml of plasma or urine spiked
at 10-15 ng with chloroform, 1,2-dichloroethane, 1,1,1-trichloroethane,
1,1,2-trichloroethane, carbon tetrachloride, vinylidene chloride, trichloro-
ethylene, tetrachloroethylene, bromodichloromethane, chlorobenzene and m-
dichlorobenzene. These blanks and controls are carried to the field and
275
-------
receive the same handling as the field samples. Workup and analysis of
field blanks and controls is interspersed with the field samples on a regular
basis. This method allows assessment of sample storage stability.
Table 1 presents a typical set of blanks and controls for QC on a field
trip where 50 blood or urine samples are to be collected.
7.1.2 Procedural Blanks and Controls
With each set of samples, a procedural blank is run. This consists of
2 ml of prepurged distilled water which is analyzed under the same conditions
as the samples. These blanks are designed to detect artifacts from dirty
glassware, laboratory atmosphere intrusion, and other sources.
7.1.2.2 GC/MS Procedural Control
Field samples, field controls, field blanks, and procedural blanks are
queued up for GC/MS analysis such that at least one QC sample is run each
working day. In addition, a standard solution is analyzed each day to serve
as a procedural control and also to update the RMR value. Thus, in a typical
working day, 4 field samples, 1 blank or control, and 1 RMR standard are
run.
The Finnigan GC/MS is a quadrupole mass spectrometer which requires
frequent tuning. Daily tuning is achieved using FC-43 and decafluorotri-
phenylphosphine (DFTPP).
7.2 Quality Assurance
Both internal and external QA procedures are to be followed. Internal
QA procedure assure the continuity and consistancy of the data. External QA
procedures (interlaboratory checks) verify or dispute the accuracy of the
data being generated in the primary laboratory.
7.2.1 Internal Quality Assurance
7.2.1.1 Supervision and Monitoring of Activities
There are three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This
person must be aware of their actions, observe events which may affect the
data, and maintain appropriate records. At the second level, the chemist's
supervisor monitors their daily activities, reviews the notebook, checks
data and calculations, and assists in "troubleshooting" problems. At the
tertiary level, a QA coordinator interviews all personnel on the project.
276
-------
Table 1. BLOOD QC SAMPLES FOR A TYPICAL SAMPLING TRIP
Sample Type
Number
Comments
Field Blank
Field Control
Lab Blank
Lab Control
5
5
Freeze after preparation,
carry to field, store with
field samples.
Store with field blanks.
Freeze after preparation,
store in same freezer as field
samples will be stored.
Store with Lab Blanks
277
-------
The interviews cover the operations they perform (precisely), the data they
obtain, a spot-check of their calculations, and any problems they have had.
7.2.1.2 Documentation
Chain of Custody
From the initial preparation of a sample container thrpugh reporting of
the analytical results, each sample is accompanied by a chain of custody
sheet. Each person signs in the time of receipt, operations performed, and
transmittal of the sample. This record is important for tracing a contaminant
bad standard, or some other problem.
Sampling Protocol Sheets
When a sample is collected, a sampling protocol sheet is filled in
which contains a discrete sample code which identifies project number, area,
site, locations, trip number, sampling period, and sample type. Also included
are sample times, volumes, address and other pertinent information. Where
appropriate, a map is made to precisely identify the location.
Sample Log
Upon return from a sampling trip, each sample code is entered into a
sample log book. This log is updated as samples proceed through workup and
analysis. Thus, at a glance, project personnel can tell the status of each
sample and find out how many are at different stages in the analytical
protocol.
GC/MS Log
Each sample run by GC/MS is logged into a notebook, detailing analysis
conditions, where the data are archived, and what hardcopy data has been
produced.
7.2.2 External Quality Assurance
Samples will be analyzed by a designated QA laboratory. Samples,
controls, and blanks will be shipped directly from the field to the QA
laboratory for analysis. They will report the results to the primary
laboratory for correlation with the primary data.
7.2.2.1 Selection of Samples for QA
Approximately 10% of the field samples (2 minimum) will be collected in
duplicate (i-.e. , two 10 ml blood or two 120 mL urine samples) for shipment
278
-------
to the QA laboratory. This selection process will be random with the
following restrictions:
(1) The donor must consent to the additional blood and urine
collection.
(2) If any stratification of donors is known, purposive selection
of QA donors may be used to get representative samples (£.g.,
occupationally exposed vs. "normal" individuals or upwind vs.
downwind residents).
7.2.2.2 QC Samples
Along with the QA duplicate samples, at least one QC control and one
QC blank must be included with the QA samples. An example is shown in Table
2 for a trip collecting 50 samples.
7.2.2.3 Sample Codes
Sll samples shipped to the QA laboratory should be encoded so that the
laboratory cannot discern between samples and controls. This "blind"
protocol will add validity to the results.
7.2.2.4 Shipping
Samples should be shipped on dry ice directly to the QA laboratory by
an appropriate air carrier (£•£•> Federal Express, Eastern Sprint) in well
insulated and packed cartons.
8.0 References
1. Sherma, J., "Manual of Analytical Quality Control for Pesticides in
Human and Environmental Media" EPA -600/1-76-017, 2K, 33 (1979).
2. ibid, Section 4, A, (6), p 2.
3. ibid, Section 4, A, (4), p 4.
Adapated from: "Determination of Volatile Purgeable Halogenated Hydrocarbons
in Human Adipose Tissue and Blood Serum: Peoples, A. J., Pfaffenberger, C.
D., Shafik, T. M., Enos, H. F., University of Miami School of Medicine,
Department of Epidemiology and Public Health, Chemical Epidemiology Division,
Miami Florida.
* Indicates sections not in original protocol, added here for application to
this research project.
279
-------
Table 2. SAMPLES TO BE COLLECTED AND SHIPPED TO QA LABORATORY
FOR A TYPICAL SAMPLING TRIP
Sample Number Comments
Duplicate Sample 5 Random selection unless prior
information stratifies subjects
Field Blank 1 Ship with samples
Field Control 1 Ship with samples
280
-------
ANALYTICAL PROTOCOL: SAMPLING AND ANALYSIS OF PURGEABLE
HALOGENATED HYDROCARBONS IN BLOOD (RTI)
1.0 Principle of the Method
Volatile compounds are recovered from a blood sample by warming the
sample and purging an inert gas over the warm sample. The vapors are then
trapped on a Tenax cartridge which is analyzed by thermal desorption interfa
ced to GC/MS.
2.0 Range and Sensitivity
For a typical organic compound approximately 30 ng are required for a
mass spectral identification using high resolution glass capillary GC/MS
analysis. Based on a 10 ml blood sample, the limit of detection is about 3
4
ng/ml or 3 ppb. The dynamic range for a purged sample is ^10 , however,
smaller samples may be purged and the range increased commensurately.
3.0 Interferences
Two possible types of interference must be considered: (1) material
present in the sample which physically prevents the effective purge of the
sample, and (2) a material which interferes with the analysis of the purged
sample. In the former case, several techniques have been developed to
handle such problems (e.g., foaming) by diluting and stirring the sample, or
the use of chemical antifearning agents. The second case is minimized by the
use of GC/MS for the analysis since unique combinations of m/£ and retention
times can be selected for most compounds. This permits the analysis of
compounds even though chromatographic resolution is not obtained.
4.0 Precision and Accuracy
14
The purge and trap technique was validated using four C-labeled model
compounds and six "cold" model compounds with an average recovery of 98.3%
±14.2%. Based on these data, expected recoveries of purgeable halogenated
organics from blood are about 80% or better. Within the precision require-
ments of this study, these recovery values indicate that the method is
essentially quantitative.
281
-------
5.0 Apparatus and Reagents
5.1 Sampling Apparatus
Vacutainers "suitable for GC", Venoject L 428 (Kimble, or other high
purity vacutainers; ice bath; disposable Pasteur pipettes and bulbs, and
cleaned and oven-treated shell vials with Teflon-lined screw caps.
5.2 Purge Apparatus
The apparatus required is shown in Figure 1.
5.3 Tenax Cartridges
Tenax cartridges are prepared and the background checked as described
in Section 6.1.1 of Protocol A-8 (Analysis of Purgeable Organic Compounds in
Water [Master Analytical Scheme]).
5.4 GC/MS/COMP
The volatile halogenated hydrocarbons purged from water are analyzed on
either an LKB 2091 GC/MS with an LKB 2031 data system or a Varian MAT CH-7
GC/MS with a Varian 620/1 data system. The sample, concentrated on a Tenax
GC cartridge is thermally desorbed using an inlet manifold system (2-6) .
The operating conditions for the thermal desorption unit and the analysis
Tenax GC cartridges are given in Table 1.
5.5 Reagents and Solvents
1. Pentane, Burdick and Jackson distilled in glass, redistilled prior
to use,
2. Methanol, Burdick and Jackson distilled in glass, redistilled
prior to use.
6.0 Procedure
6.1 Collection of Samples
Blood samples are collected in replicate 10 ml vacutainer tubes contain-
ing an anticoagulant. Using a qualified phlebotomist, the samples are
collected by brachial venipuncture. Glass syringes represent the optimal
collection device, since no polymeric material which may contaiminate the
sample comes in contact with the blood. However, sterilization of large
numbers of glass syringes in the field is not practical, so vacutainers will
be used.
Possible contamination by permeation through the rubber septum caps of
the vacutainers is a cause for concern. Teflon-lined vacutainers are not
282
-------
THERMOMETER
-20tol50cc
THERMOMETER ADAPTER
with 0—ring
I 10/18
TENAX CARTRIDGE
HELIUM
PURGE
HELIUM INLET
TUBE
LIQUID LEVEL
100 ml ROUND BOTTOM FLASK
MAGNETIC STIRRING BAR
Figure 1. Headspace using apparatus for blood, urine and tissue
samples.
283
-------
Table 1. INSTRUMENTAL OPERATING CONDITIONS
LKB 2091
Varian MAT CH-7
00
Desorption chamber temperature
Desorption chamber He flow
Desoprtion time
Capillary Trap Temperature during desorption
Temperature of capillary trap during injection
onto column
Time of He flow through capillary trap
He flow through column [sweep time]
Carrier flow
Capillary column
Column temperature
Scan range
Scan rate
Scan cycle time
Scan mode
Trap current
Filament current
Accelerating volatage
270
15 ml/rain
8.0 rain
-196°C
265
10 ml/min
8.0 Bin
-1968C
-196°C to 250°C - then held at 190°C
12 3/4 min
9.5 min
2.0 ml/min
100 m SE-30 SCOT
30°C for 2 min,
then 4°/min to 240°
5-490 dalton
2 sec full scale
2.4 sec
parabolic
4A
50|jA
3.5 kV
12 3/4 min
4 min
1.0 ml/min
20 m SE-30 WCOT
20 -» 240° at 4°/min
20 •» 500 dalton
1 sec/decade
4.5 sec
exponential
300pA
2kV
-------
available but manufacturers recommended special vacutainers "suitable for
GC" (Venoject L 428, Kimble). Validation experiments for benzene under a
separate contract (EPA No. 68-01-3849, Task I) found the background of these
tubes to be acceptable.
Following collection, shipping and storage procedures must assure that
the purgeable halogenated organics remain intact in the blood sample.
Leakage of the vacutainer caps has been observed and permeation through the
cap material is suspected. Accordingly, these containers are not suitable
for storage. In the procedure used for this project, the blood sample is
chilled and transferred to a clean shell vial with a Teflon-lined screw cap.
The cap is then taped shut. This procedure has been validated and field
tested on EPA Contract No. 68-01-3849, Task I.
6.2 Purge of Volatile Organics
1. Measure a 10 ml aliquot of whole blood (if available), previously
chilled to 4°C, into the purge flask (Fig. 1).
2. Dilute sample to 50 ml with purged distilled water and add stir
bar.
3. Assemble apparatus, start stirring and raise the temperature
to 50°C.
4. Adjust helium flow to 25 ml/min and purge for 90 min.
5. After 90 min disassemble apparatus and transfer Tenax cartridge
to a Kimax culture tube with 2 g calcium sulfate dessicant for 4
hours of drying.
6. Transfer Tenax cartridge to an identical Kimax culture tube
without calcium sulfate, seal in a paint can and store in freezer
until analysis.
6.3 Analysis of Sample Purged on Cartridge
The insturmental conditions for the analysis of halogenated hydrocarbons
of the sorbent Tenax GC sampling cartridge is shown in Table 1. The thermal
desorption chamber and six-port valve are maintained at 270° and 200°C,
respectively. The helium purge gas through the desorption chamber is adjusted
to 15-20 ml/min. The nickel capillary trap at the inlet manifold is cooled
with liquid nitrogen. In a typical thermal desorption cycle a sampling
cartridge is placed in the preheated desorption chamber and helium gas is
285
-------
channeled through the cartridge to purge the vapors into the liquid nitrogen
cooled nickel capillary trap. After desorption the six-port valve is rota-
ted and the temperature on the capillary loop is rapidly raised; the carrier
gas then introduces the vapors onto the high resolution GLC column. The
glass capillary column is temperature programmed from 20° to 240°C at 4°/min
and held at the upper limit for a minimum of 10 min. After all of the
components have eluted from the capillary column the analytical column is
then cooled to ambient temperature and the next sample is processed.
6.4 Quantitation
All data are acquired in the full scan mode. Quantitation of the
halogenated compounds of interest is accomplished by utilizing selected ion
plots, SIPs, which are plots of the intensity of specific ions (obtained
from full scan data) vs. time. Using SIPs of ions characteristic of a given
compound in conjunction with retention times permits quantitation of compo-
nents of overlapping peaks. Two external standards, perfluorobenzene and
perfluorotoluene, were added to each Tenax GC cartridge in known quantities
just prior to analysis. In order to eliminate the need to construct com-
plete calibration curves for each compound quantitated, the method of
relative molar response (RMR) is used. In this method the relationship of
the RMR of the unknown to the RMR of the standard is determined as follows:
A , /Moles .
unk' unk
unknown A ../Moles . ,
std std
A ./g . /GMW .
_ unk &unk unk
unk/std ~ A ,,/g _./GMW ,
std'6std' std
where A = peak response of a selected ion,
g = number of grams present, and
GMW = gram molecular weight.
Thus, in the sample being analyzed:
(A ,)(GMW ,)(g .)
unk/v unk'V6unk^
8unknown (A ..)(GMW 4.,)(RMR , , , ,
std'v std ^ unk/std
286
-------
The value of an RMR is determined from at least three independent analyses
of standards of accurately known concentration prepared using a gas permeation
system (5). The precision of this method has been determined to be generally
±10 percent when replicate sampling cartridges are examined.
7.0 Quality Assurance Program
An on-going quality assurance program is required to assure the data
quality. Quality control (QC) procedures determine artifacts, lossed, etc.
through a system of blanks and controls. Quality assurance (QA) procedures
monitor the execution of the procedure and check data interpretations and
calculations.
7.1 Quality Control
7.1.1 Field Blanks and Controls
Prior to a field sampling trip, enough blanks and controls are prepared
to equal 10% each (2 minimum) of the anticipated number of field samples.
Blanks consist of 10 ml of water in the same type of sampling container as
is used in the field. Controls consist of 10 ml of plasma spiked at 100-450
ng with chloroform, 1,2-dichloroethane, 1,1,1-trichloroethane, 1,1,2-tri-
chloroethane, carbon tetrachloride, vinylidene chloride, trichloroethylene,
tetrachloroethylene, bromodichloromethane, chlorobenzene, m-dichlorobenzene,
and vinyl chloride. These blanks and controls are carried to the field and
receive the same handling as the field samples. Workup and analysis of
field blanks and controls is interspersed with the field samples on a regular
basis. This method allows assessment of sample storage stability.
Table 2 presents a typical set of blanks and controls for QC on a field
trip where 50 blood samples are to be collected.
7.1.2 Procedural Blanks and Controls
7.1.2.1 Extraction Blanks
With each set of samples, a procedural blank is run. This consists of
10 ml of prepurged distilled water which is extracted under the same condi-
tions as the samples. These blanks are designed to detect artifacts from
dirty glassware, laboratory atmosphere intrusion, and other sources.
7.1.2.2 GC/MS Procedural Control
At the start of each working day, a mixture of 2,6-dimethylphenol,
2,6-dimethylaniline, and acetophenone (PA mixture) is analyzed to monitor
287
-------
Table 2. BLOOD QC SAMPLES FOR A TYPICAL SAMPLING TRIP
Sample Type
Number
Comments
Field Blank
Field Control
Lab Blank
Lab Control
5
5
Freeze after preparation,
carry to field, store with
field samples.
Store with field blanks.
Freeze after preparation,
store in same freezer as field
samples will be stored.
Store with Lab Blanks
288
-------
the capillary GC column performance. This also serves to check the mass
spectrometer tuning.
Field samples, field controls, field blanks, and procedural blanks are
queued up for GC/MS analysis such that at least one QC sample is run each
working day. In addition, a standard solution is analyzed each day to serve
as a procedural control and also to update the RMR value. Thus, in a typical
working day, 4 field samples, 1 blank or control, and 1 RMR standard are
run.
The Finnigan GC/MS is a quadrupole mass spectrometer which requires
frequent tuning. Daily tuning is achieved using FC-43 and decafluorotri-
phenylphosphine (DFTPP).
7.2 Quality Assurance
Both internal and external QA procedures are to be followed. Internal
QA procedure assure the continuity and consistancy of the data. External QA
procedures (interlaboratory checks) verify or dispute the accuracy of the
data being generated in the primary laboratory.
7.2.1 Internal Quality Assurance
7.2.1.1 Supervision and Monitoring of Activities
There are three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This
person must be aware of their actions, observe events which may affect the
data, and maintain appropriate records. At the second level, the chemist's
supervisor monitors their daily activities, reviews the notebook, checks
data and calculations, and assists in "troubleshooting" problems. At the
tertiary level, a QA coordinator interviews all personnel on the project.
The interviews cover the operations they perform (precisely), the data they
obtain, a spot-check of their calculations, and any problems they have had.
7.2.1.2 Documentation
Chain of Custody
From the initial preparation of a sample container through reporting of
the analytical results, each sample is accompanied by a chain of custody
sheet. Each person signs in the time of receipt, operations performed, and
transmittal of the sample. This record is important for tracing a contaminant,
bad standard, or some other problem.
289
-------
Sampling Protocol Sheets
When a sample is collected, a sampling protocol sheet is filled in
which contains a discrete sample code which identifies project number, area,
site, locations, trip number, sampling period, and sample type. Also included
are sample times, volumes, address and other pertinent information. Where
appropriate, a map is made to precisely identify the location.
Sample Log
Upon return from a sampling trip, each sample code is entered into a
sample log book. This log is updated as samples proceed through workup and
analysis. Thus, at a glance, project personnel can tell the status of each
sample and find out how many are at different stages in the analytical
protocol.
GC/MS Log
Each sample run by GC/MS is logged into a notebook, detailing analysis
conditions, where the data are archived, and what hardcopy data has been
produced.
7.2.2 External Quality Assurance
Samples will be analyzed by a designated QA laboratory. Samples,
controls, and blanks will be shipped directly from the field to the QA
laboratory for analysis. They will report the results to the primary labora-
tory for correlation with the primary data.
7.2.2.1 Selection of Samples for QA
Approximately 10% of the field samples (2 minimum) will be collected in
duplicate (i^.e. , two 10 ml blood samples) for shipment to the QA laboratory.
This selection process will be random with the following restrictions:
(1) The donor must consent to the additional blood collection.
(2) If any stratification of donors is known, purposive selection of
QA donors may be used to get representative samples (e.g.,
occupationally exposed vs. "normal" individuals or upwind vs.
downwind residents).
7.2.2.2 QC Samples
Along with the QA duplicate samples, at least one QC control and one QC
blank must be included with the QA samples. An example is shown in Table 3
for a trip collecting 50 samples.
290
-------
Table 3. SAMPLES TO BE COLLECTED AND SHIPPED TO QA LABORATORY
FOR A TYPICAL SAMPLING TRIP
Sample Number Comments
Duplicate Sample 5 Random selection unless prior
information stratifies subjects
Field Blank 1 Ship with samples
Field Control 1 Ship with samples
291
-------
7.2.2.3 Sample Codes
Sll samples shipped to the QA laboratory should be encoded so that the
laboratory cannot discern between samples and controls. This "blind" protocol
will add validity to the results.
7.2.2.4 Shipping
Samples should be shipped on dry ice directly to the jQA laboratory by
an appropriate air carrier (e.g., Federal Express, Eastern Sprint) in well
insulated and packed cartons.
8.0 References
1. Pellizzari, E.D., M.D. Erickson and R.A. Zweidinger, "Analytical
Protocols for Making a Preliminary Assessment of Halogenated Organic
Compounds- in Man and Environmental Media", Appendix C, Pg 116-117,
Revised April 79.
2. Pellizzari, E.D., Development of Method for Carcinogenic Vapor Analysis
in Ambient Atmospheres. Publication No. EPA-650/2-74-121, Contract No.
68-02-1228, 148 pp., July, 1974.
3. Pellizzari, E.D., Development of Analytical Techniques for Measuring
Ambient Atmospheric Carcinogenic Vapors. Publication No. EPA-600/
2-76-076, Contract No. 68-02-1228, 185 pp., November, 1975.
4. Pellizzari, E.D., J.E. Bunch, B.H. Carpenter and E. Sawicki, Environ.
Sci. Tech., 9, 552 (1975).
5. Pellizzari, E.D., B.H. Carpenter, J.E. Bunch and E. Sawicki, Environ.
Sci. Tech., 9, 556, (1975).
6. Pellizzari, E.D., Development of Analytical Techniques for Measuring
Ambient Atmospheric Carcinogenic Vapors, Publication No. EPA-600/
7-77-055, 288 pp., June 1977.
Revised April 1980
292
-------
ANALYTICAL PROTOCOL: SAMPLING AND ANALYSIS OF
»
ARSENIC, CADMIUM, AND LEAD IN WHOLE BLOOD (RTI)
1.0 Principle of Method
The analysis of arsenic, cadmium, and lead in urine is carried out
using atomic absorption spectrophotometry. Increased sensitivity is achieved
by atomizing the metal in a graphite furnace with continuous deuterium
background correction. Sample workup for arsenic analysis includes an extrac-
tion from the blood matrix and furnace atomization of solutions containing
1000 ppm nickel.
2.0 Range and Detection Limit
The minimum detection limit (MDL) and range for the metal assays in
urine are shown below.
MDL Max. Cone.
not established
0.05 10.0
1.00 100.0
Samples containing higher metal concentrations may be analyzed by
suitable dilution with 0.5% nitric acid. Dilution for arsenic determina-
tions is made with 0.005 M dichromate solution containing 1000 ppm nickel in
1.0% nitric acid.
3.0 Interferences
No known chemical or spectral interferences exist in the analysis of
arsenic, cadmium or lead in whole blood. Severe matrix interferences with
arsenic analysis are minimized by incorporating the toluene extraction step
into the workup procedure.
4.0 Precision and Accuracy
The precision and accuracy associated with these analyses is a function
of sample metal concentration. At the detection limit, the total measurement
error is + 100%. Based on the results of a previous study (1), the metal
analyses are performed with the following precision (relative standard
deviation) and accuracy (relative error). The total analysis error is also
given (2).
293
-------
Metal Range Precision (% RSD) Accuracy (% RE) Total Error (%)
Cadmium 0.3-5.0 15 10 40
Lead 5.0-100.0 10 5 25
The precision and accuracy for the arsenic analysis has not been deter-
mined .
5.0 Apparatus and Reagents
A commercially available stock solution containing 1000 ppm metal is
used for the preparation of the calibration standards. The concentrated
nitric acid is reagent grade quality and the deionized water used in this
study is prefiltered and subjected to the action of an activated carbon
cartridge and two sequential ion exchange units.
The glassware used for sample workup and the preparation of the calibra-
tion solutions must be subjected to a nitric acid cleaning protocol.
All volumetric flasks should be soaked overnight in 20% nitric acid,
rinsed with deionized water, soaking for an additional 15-18 hours in a 5%
nitric acid bath, followed by a copious deionized water rinse. The flasks
are completely filled with 0.5% nitric acid and stored in this manner.
Prior to use, each flask is emptied and rinsed well with deionized water.
Pipets are soaked in 5% nitric acid, rinsed well with deionized water,
air-dried and stored in a clean, dust-free environment.
All beakers used for blood operations require additional pretreatment.
Clean beakers (soaked in 20% and 5% nitric acid) are "predigested" by
heating 10-25 ml of cone, nitric to reflux (with watchglass), cooling, and
discarding the acid. The beakers are rinsed thoroughly with deionized water
and used for a sample digestion within 30 minutes. The beakers are never
allowed to go dry.
Sample cup for the graphite furnace autosampler may be made of polysty-
rene or Teflon. The former type requires overnight soaking in 1% nitric
acid and followed by rinsing with deionized water. The latter type may be
soaked overnight in 20% nitric acid, rinsed, and dried in a 105°C oven.
Nickel chloride hexahydrate is used for adjusting the nickel concen-
tration to 1000 ppm in all solutions slated for arsenic analysis.
294
-------
6.0 Procedure
6.1 Collection of Samples
Whole blood samples are collected by veinpuncture from a brachial (arm)
vein. The blood is drawn into a Vacutainer tube (B.D. No. 4727, low trace
metal content) containing an EDTA anticoagulant. The tube is labeled and
all pertinent information recorded on a protocol sheet. • -
6.2 Extraction, Cleanup, and Extraction
6.2.1 Cadmium and Lead Analysis
One ml of whole blood is treated with 10.0 ml of cone. HN03 at 85-90°C
in a predigested beaker. The heating is continued for 2 hours 45 minutes
with a watchglass placed on top of the beaker to minimize losses. The
beaker is cooled and a total of 1.0 ml of 30% H.O- is added in 0.2 ml portions
The contents of the beaker is heated at 85-90°C for an additional 15 minutes.
The watchglass is removed and the heating continued to reduce the digest
volume to 1-2 ml. The residue is transferred to a 10 ml volumetric flask
and diluted to the mark with 0.5% HNO». The sample solution is stored in 1
oz. polyethylene bottle (with screw cap) at ambient temperatures until
analyzed.
6.2.2 Arsenic Analysis
4
One ml of whole blood is treated with 0.1 ml of a 10 ppm nickel solution
and digested with 10.0 ml of cone. fflTCL as described in Section 6.2.1.
After treatment with 30% H.O. the digestate is cautiously evaporated to
dryness. The residue is transferred to a clean 4 oz glass bottle with 25 ml
of 4:1 cone. HC1 and the mixture allowed to stand at room temperature for
3-7 days. .At the end of this period, 10 ml of 0.5 M SnCl2 and 5 ml of 30%
KI is added to the sample digest and the total allowed to stand at room
temperature and for 30 minutes. Forty ml of cone. HC1 and 10.0 ml of toluene
is added to the mixture and the arsenic bodies extracted into the organic
phase. Half of the toluene layer (5.0 ml) is withdrawn and mixed with 2.0
ml of a 0.005 M dichromate solution containing 1000 ppm nickel in 1.0% HNOg.
The arsenic compounds are back-extracted into the aqueous phase and stored
in polypropylene bottles until ready for analysis.
295
-------
6.3 Instrumental
A Perkin-Elmer Model 403 Spectrophotometer, equipped with a HGA-2000
furnace attachment with deuterium background correction is used for this
analysis. An electrodeless discharge lamp is used as the light source and
the furnace atomization response traced on a Perkin-Elmer Model 056 recorder
An AS-1 Autosampler may be used to increase throughput and/or to improve
peak reproducibility and sensitivity.
Arsenic: Wavelength - 193.7 nm
Gas Interrupt (N2) Auto
Furnace Cycle Conditions -
Dry: 200°C for 30 sec.
Char: 15000°C for 35 sec.
Atomize: 2500°C for 6 sec.
Injection Volume - 20 pi
Cadmium: Wavelength - 228.8 nm
Gas Interrupt (N~) - Manual
Furnace Cycle Conditions -
Dry: 200°C for 20 sec.
Char: 400°C for 30 sec.
Atomize:- 2000°C for 8 sec.
Injection Volume - 20 (jl
Lead: Wavelength - 217.0 nm
Gas Interrupt (N_) - Auto
Furnace Cycle Conditions -
Dry: 200°C for 20 sec.
Char: 500°C for 30 sec.
Atomize: 2000°C for 8 sec.
Injection Volume - 20 pi
6.4 Qualitative and Quantitative Analysis
6.4.1 Qualitative Analysis
N/A
296
-------
6.4.2 Quantitative Analysis
The instrument is calibrated with a digested control blood spiked -at
four different concentration, an unspiked control urine, and a reagent
blank.
Calibration Range (spike concentration):
Arsenic - 0.0 to 10.0 Mg/100 ml •
Cadmium - 0.0 to 10.0
Lead - 0.0 to 100.0
An exponential of the form y = Ae -M provides the best representation
of the analytical curve. The values of the x,y calibration pairs are
entered into a Monroe Calculator Model 1880 programmed to regress the data
to the exponential and to provide values for the constants A, b, and M.
Sample peak heights are measured manually and expressed in units of
millivolts. The standard additions calibration constants A, b, and M are
entered into the storage banks of a Texas Instrument Calculator Model 57 and
the metal concentration results obtained by keying in peak height data.
Sample peak measurements and concentration results are recorded on a calcu-
lation worksheet.
y = Aebx-M, |Jg/100 ml
y = y (metal cone, in sample relative to control blood) +
y (metal cone, in control blood)
bx
y = Ae S-M
s
bx
y = -y - -(Ae -M)
bx bx
A f S °\
y = A(e -e )
y = metal concentration in sample, (Jg/100 ml,
y = reagent blank peak height, mv
y = standard additions metal concentration corresponding to reagent
blank signal (y < 0),
y = -y = metal concentration in control blood, pg/100 ml,
297
-------
x = sample peak height, mv
s
y = concentration differential between sample and control blood.
5 This blood may be either negative (y < y£) or positive (yc < y)
7.0 Quality Assurance Program
An on-going quality assurance program is required to assure the data
quality. Quality control (QC) procedures determine artifacts, losses, etc.,
through a system of blanks and controls. Quality assurance (QA) procedures
monitor the execution of the procedure and check data interpretations and
calculations.
7.1 Quality Control
7.1.1 Field Control
Prior to "field sampling, several control whole blood collections (10%
of anticipated number of field samples) are obtained. Each blood sample is
collected in duplicate. One tube from each collection is sent to the site
and subjected to the same handling and storage conditions as field samples.
The other tube is stored at RTI in a dust-free environment. On receipt of
samples at RTI, both tubes of the control blood collections are worked up
and analyzed as a part of each blood analytical run. Within the precision
of the assay, the difference in calculated metal concentrations of the two
control blood tubes is a measure of the contamination/loss during field
storage, and transit to RTI.
7.1.2 Internal Quality Control
7.1.2.1 Calibration Standards and Blanks
The instrument is calibrated before each analytical run with four
standard solutions and a reagent blank. Evidence of contamination or
instrument malfunction is evident at this time. Such problems are resolved
before initiating sample analysis.
7.1.2.2 Conditioning of Graphite Tube
Before each analytical run, the graphite tube is conditioned by inject-
ing 10 to 20-20 (Jl aliquots of one of the calibration standards. This
operation insures acceptable precision during sample analysis.
298
-------
7.1.2.3 Duplicate Injections
Reproducibility of peak response iscontinuously monitored during
sample analysis. All standard and sample solutions receive two successive
injections into the graphite furnace. Signal agreement between the duplicate
injections is evaluated according to the following criterion:
First Signal % Maximum Permissible Permissible Range of
% of Full Signal Variation (% MPV) Second Signal, % of Full Scale
90 ± 4% 86-94
80 ± 5% 76-84
70 ± 6% 66-74
60 ± 7% 56-64
50 ± 8% 46-54
40 ± 10% 36-44
30 ± 13% 26-34
20 ± 20% 16-24
10 ± 30% 7-13
5 ± 60% 2-8
2 ±100% 0-4
If the second injection gives a signal which falls outside the permis-
sible range, a third injection is performed. The peak measurement not in
agreement with the matching pair is discarded.
All calibration and sample calculations are based on the mean of
the duplicate determinations.
7.1.2.4 Standard Checks
Instrument performance is monitored during each analytical run.
After the analysis of every 12-16 samples one of the calibration standards
is reinjected into the furnace. The standard which most closely matches
the sample peak heights is selected as the check solution. A metal
concentration is calculated for the check standard based on its peak
height during the calculation run. Similar calculations are carried out
for each chech response and the observed changes in metal concentration
expressed in terms of standard deviation units (SDU).
299
-------
(Calibration value - check Value) 100
SU ~ (Calibration Value)(% of RSD)
The analysis is under control when the SDU < 2.0. Standard checks
which indicate a variation in peak response greater than 2.0 SDU are un-
acceptable. In this event, the graphite tube is changed, conditioned, and
the system recalibrated. Quality control charts are graphed to show this
change in instrument performance with time.
7.2 Quality Assurance
Both internal and external QA procedures are to be followed. Internal
QA procedures assume the continuity and consistency of the data. External
QA procedures (interlaboratory checks) verify or dispute the accuracy of the
data being generated in the primary laboratory.
7.2.1.1 Supervision and Monitoring of Activities
There are three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This
person must be aware of their actions, observe events which may effect the
data, and maintain appropriate records. At the second level, the chemist's
supervisor monitors their daily activities, reviews the notebook, checks
data and calculations, and assists in "troubleshooting" problems. At the
tertiary levels, a QA coordinator interviews all personnel on the project.
The interviews cover the operations they perform (precisely), the data they
obtain, a spot-check of their calculations, and any problems they have had.
7.2.1.2 Documentation
Chain of Custody
From the initial preparation of a sample container through reporting of
the analytical results, each sample is accompanied by a chain of custody
sheet. Each person signs in the time of receipt, operations performed, and
transmittal of the sample. This record is important for tracing a contami-
nant, bad standard, or some other problem.
Sampling Protocol Sheets
When a sample is collected, a sampling protocol sheet is filled in
which contains a discrete sample code which identifies project number, area,
site, locations, trip number, sampling period, and sample type. Also
300
-------
included are sample -times,, ^volumes,, ^addressjes.^ meteorology," and other perti-
nent information. Where appropriate, a map is made to precisely identify
the location.
Sample Log
Upon return from a sampling trip, each sample code is entered into a
sample log book. This log is updated as samples proceed-through workup and
analysis. Thus, at a glance, project personnel can tell the status of each
sample and find out how many are at different stages in the analytical
protocol.
Instrument Log
Each sample analysis is logged into a notebook, detailing analysis
conditions.
7.2.2 External Quality Assurance
Samples will be analyzed by a designated QA laboratory. Samples,
controls, and blanks will be shipped directly from the field to the QA
laboratory for analysis. They will report the results to the primary
laboratory for correlation with the primary data.
7.2.2.1 Selection of Samples for QA
Approximately 10% of the field samples (2 minimum) will be collected in
duplicate shipment to the QA laboratory. This selection process will be
random with the following restrictions:
(1) The donor must consent to the additional collection.
(2) If any stratification of donors is known, purposive selection
of QA donors may be used to get representative samples (e.g.
occupationally exposed vs_ "normal" individuals or upwind vs downwind
residents).
7.2.2.2 QC Samples
Along with the QA duplicate samples, at least one QC control and one
QC blank must be included with the QA samples. An example is shown in Table
7 for a trip collecting 50 samples.
7.2.2.3 Sample Codes
All samples shipped to the QA laboratory should be encoded so that the
laboratory cannot discern between samples and controls. This "blind"
protocol will add validity to the results.
301
-------
7.2.2.4 Shipping
Samples should be shipped on dry -ice directly to the QA laboratory by
an appropriate air carrier (£•£•, Federal Express, Eastern Sprint) in well
insulated and packed cartons.
8.0 References
8-1 "Epidemiologic Study Conducted in Populations Living Around Non-Ferrous
Smelters", Final Report for Contract No. 68-02-2442 (in preparation).
8-2 McFarren, E. F., Lishka, R. J. and Parker, J. H., Criterion for Judging
Acceptability of Analytical Methods, Anal. Chem., 42(3), 358 (1970).
302
-------
ANALYTICAL PROTOCOL: ANALYSIS OF HUMAN SERUM AND URINE
FOR EXTRACTABLES (HERL-RTP)
*1.0 Principle of the Method
Semi-volatile hydrocarbons are extracted from blood plasma and urine
with organic solvents, dried, and concentrated to an appropriate volume" for
analysis using gas chromatography/electron capture detection (GC/ECD).
*2.0 Range and Detection Limit
The lower limit of detectability as established by the EPA Analytical
Chemistry Committee for chlorinated pesticides in serum is as follows:
B-BHC, lindane, aldrin, heptachlor, heptachlor epoxide, o,£'-DDE, £,p_'-DDE,
dieldrin .— -— 1 ppb.
Endrin, o,p_'-DDT, £,£'-DDD, £,£!-DDT 2 ppb.
Although the linear dynamic range of an electron capture detector is
not very large the range of this method can be greatly enhanced by dilution
of the sample.
*3.0 Interferences
Interferences in sample analysis and quantification using GC/ECD are
manifested in the electron capture ability of the given contaminant. Blood
and urine extracts which have not been cleaned up contain repidly eluting
components which exhibit high electron capturing properties; these interferen-
ces can be largely removed by a fractionation process with an activated
Florisil column as discussed below.
*4.0 Precision and Accuracy
Recoveries from spiked serum samples in a three year study gave average
recoveries of 96% ± 13% (1). The four compounds used were not specified.
5.0 Apparatus and Reagents
1. A rotary mixer so designed as to accomodate the 16 mm culture
tubes and which may be operated at a rotary speed of 50 rpm.
TM
Fisher Scientific Company, Roto^Rack , Cat. No. 14-456.
2. Gas chromatograph fitted with electron capture detector. Recom-
mended GLC columns and operating parameters are given in reference
(2).
303
-------
3. Tubes, Culture, 16 x 125 JKa^CitigA JHJLh Jifi£AW^fiflI!«*JU?£ JJu-415..-
with Teflon-faced rubber liners, Corning No. 9826.
4. Micro-Snyder column modified, with 10/22 "S joint, Kontes No.
K-569251.
5. Concentrator tube, 10 ml, grad. 0 to 0.1 and 2 to 10 x 1, 19/22
V joint, size 1025, Kontes No. K-570050.
6. Syringe, 100 pi, Hamilton No. 710 or equivalent.
7. Vortex Genie mixer.
8. Pipet, Mohr type, 1 ml grad. in 0.01 ml increments. Corning
No. 7063 or equivalent.
9. Pipets, transfer, 2-, 5-, and 6-ml Corning No. 7100 or the
equivalent.
10. Beads, solid, glass, 3 mm, Corning No. 7268 or the equivalent.
11. Six-place tube carrier, stainless steel. May be fabricated at
local tin shop (see original reference).
12. Water bath capable of holding temp, of 95 to 100°C.
13. Centrifuge with head to accommodate the Corning No. 9826 tube,
capable of speed of 2,000 rpm.
14. Hexane, distilled in glass, pesticide grade.
If Florisil column cleanup is necessary the additional following apparatus
and reagents are necessary:
1. Funnels, glass, ca 60 mm diameter.
2. Separatory funnels - 125 ml and 1 liter, Kimble 29048-F, or
equiv.
3. Chromatographic columns -25 mm o.d. x 300 mm long, with Teflon
stopcocks, without fritted glass plates, Kontes 420530, Size 241.
4. Erlenmeyer flasks -500 ml capacity.
5. Kuderna-Danish concentrator fitted with grad. evaporative concen-
trator tube. Available from the Kontes Glass Company, each com-
ponent bearing the following stock numbers:
a. Flask, 500 ml, stock //K-570001
b. Snyder Column, 3-ball, stock #K-503000
304
-------
c. Steel springs, 1/2", stock-#K-662750
d. Concentrator tutes,~ JO ml, size 1025, stock #K-570050
6. Glass wool - Corning #3950 or equivalent.
5.2 Reagents
1. Petroleum ether - Pesticide Quality, redistilled in glass, b.p.
30°C - 60°C.
NOTE: If this method is used for the detection and quantitation of
organophosphorous compounds, some special factors must be considered. The
presence of any peroxides in the ethyl ether and/or impurities in the pet.
ether can result in extremely low recoveries. Recovery efficiency should be
predetermined on standard mixtures containing the specific compounds of
interest. If -low recoveries are obtained, it may be necessary to try an
alternate manufacturer's pet. ether.
2. Diethyl ether - AR grade, peroxide free, Mallinckrodt #0850 or
the equivalent. The ether must contain 2% (v/v) absolute ethanol.
Some of the AR grade ethers contain 2% ethanol, added as a stabili-
zer, and it is therefore unnecessary to add ethanol unless peroxides
are found and removed.
NOTE: To determine the absence of peroxides in the ether, add 1 ml of
freshly prepared 10% KC1 solution to 10 ml of ether in a clean 25-ml cylinder
previously rinsed with the ether. Shake and let stand 1 minute. A yellow
color in the ether layer indicates the presence of peroxides which must be
removed before using. See Misc. Note 4 at end of procedure. The peroxide
test should be repeated at weekly intervals on any single bottle or can as
it is possible for peroxides to form from repeated opening of the container.
3. Eluting mixture, 6% (6+94) - purified diethyl ether (60 ml) is
diluted to 1000 ml with redistilled petroleum ether and anhydrous
sodium sulfate (10-25 g) is added to remove moisture.
4. Eluting mixture, 15% (15+85) - purified diethyl ether (150 ml) is
diluted to 1000 ml with redistilled petroleum ether and dried as
described above.
NOTE: Neither of the eluting mixtures should be held longer than 24
hours after mixing.
5. Florisil, 60/100 mesh, PR grade, to be stored at 130°C until used.
305
-------
NOTE: (1) In a high humidity -room,;-the -column may pick up enough
moisture during packing to influence the elution pattern. To insure unifor-
mity of the Florisil fractionation, it is recommended to those laboratories
with sufficiently large drying ovens that the columns be packed ahead of
time and held (at least overnight) at 130°C until used. (2) Florisil fur-
nished to the contract laboratories by the RTF, NC laboratory on order, has
been activated by the manufacturer, and elution pattern data is included
with each shipment. However, each laboratory should determine their own
pesticide recovery and elution pattern on each new lot received, as environ-
mental conditions in the various laboratories may differ somewhat from that
in RTF, NC. Each new batch should be tested by the procedure described
previously for assurance that the operator can obtain recoveries and compound
elution patterns comparable to the data given on the accompanying table.
6. Acetonitrile, pesticide grade, saturated with pet. ether.
NOTE: Occasional lots of CH^CN are impure and require redistillation.
~~~—~~~ .3
Generally, vapors from inpure acetonitrile will turn litmus paper blue
when the moistened paper is held over the mouth of the bottle.
7- Anhydrous sodium sulfate, reagent grade granular, Mallinkrodt
stock #8024 or the equivalent.
NOTE: When each new bottle is opened, it should be tested for con-
taminants that will produce peaks by Electron Capture Gas Liquid Chromato-
graphy. This may be done by transferring ca 10 grams to a 125 ml Erlenmeyer
flask, adding 50 ml pet. ether, stoppering and shaking vigorously for 1
minute. Decant extract into a 100 ml beaker and evaporate down to ca 5 ml.
Inject 5 |Jl into the Gas Liquid Chromatograph and observe chromatogram for
contaminants. When impurities are found, it is necessary to remove them by
extraction. This may be done by using hexane in a continuously cycling
Soxhlet extraction apparatus or by several successive rinses with hexane in
a beaker. The material is then dried in an oven and kept in a glass-stoppered
container.
NOTE: See Note for sodium sulfate, Step 7, above.
8. MgO-Celite mixture (1:1) weigh equal parts of reagent grade MgO
and Celite 545 and mix thoroughly, (optional)
306
-------
6.0 Procedure
*6.1 Collection of Samples
6.1.1 Blood
Blood samples are obtained in 10 ml vacutainer tubes "suitable for GC"
(Venoject L428, Kimble) containing 15 rag tripotassium salt of EDTA and 20 (Jg
potassium sorbate.
6.1.2 Urine
Place whole blood sample in the refrigerator for about 30 minutes for a
settling period and then centrifuge for a sufficient time for the separation
of at least 3 ml of clear serum - generally 10 minutes at 2,500 r.p.m.
Whether or not the analysis is to be conducted immediately, it is desirable
at this point to transfer the 2 ml sample aliquot to the 16 x 125 mm culture
tube used for extraction. If analysis cannot be run immediately, place in
refrigerator at 2-5°C for periods of up to 24 hours before analysis. If
time interval to analysis exceeds 24 hours, the tube should be stored in a
deep freeze at -15 to -25°C. Stored in this manner, analysis may be delayed
for periods up to a month without undue effects on the chlorinated pesticides
present.
First morning urines are collected in a 120 ml wide-mouthed glass
bottles. The sample is stored at 4° awaiting analysis.
6.2 Extraction, Cleanup, and Concentration
6.2.1 Serum Samples
1. Mix blood serum sample thoroughly and, with a volumetric pipet,
transfer 2 ml to a 15 ml round bottom culture tube.
NOTE:. .In case of the presence of any flocculent or sedimentary material,
it is strongly recommended that the sample be centrifuged ca 5 minutes @
2,000 r.p.m. before pipetting the 2 ml aliquot. Failure to observe this
point may result in. poor reproducibility of replicated analyses of the same
sample.
2. Add 6 ml hexane from a volumetric pepet. Tightly stopper the
. culture tube with a Teflon-lined screw cap. Place tube on rotator.
3. Set rotator speed at 50 r.p.m. and rotate for 2 hours.
NOTE: (1) This speed may vary from 50 to 44 r.p.m. but should be
confined to this range. (2) Unless the sample is extremely old, emulsion
307
-------
formation should present no problem,-la case it-occurs, -centrifuge-at 2,000^
r.p.m. 4 to 5 minutes, or longer if necessary, to -effertr-sufficrentr separation
to permit withdrawal of the 5 ml aliquot of clear extract.
4. With a volumetric pipet, transfer 5 ml of the hexane extract to a
10 ml grad. concentrator tube, add one 3 mm glass bead, and attach
a modified micro-Snyder column. Evaporate the .extract in a steam
or hot water bath at 100°C to a volume slightly less than that
which is estimated as appropriate to accommodate (1) the current
level of electron capture detector sensitivity, and (2) the expected
residue range in the particular sample. When working with general
population blood of low pesticide levels, it may be necessary to
evaporate to ca 0.5 ml.
NOTE: (1) With some experience the operator can complete the evapora-
tion step in less than 5 minutes. The tube must be withdrawn from the water
when boiling agitation becomes too vigorous. Immersion and withdrawal are
alternated based on observation of boil agitation. (2) Up to six tubes of
extract may be evaporated simultaneously 'by using the special rack shown in
Fig. 2. Time and motion studies have shown that the time required for the
evaporation period is equal to that required for a single tube. (3) When
working with blood from high exposure donors, the 5 ml aliquot may require
dilution rather than concentration. This can be determined by a preliminary
analysis of the 5-ml aliquot. (4) With lower concentrations, use higher
degree of concentration of samples.
5. Allow the tube to cool (3 to 5 minutes), remove the micro-Snyder
column and rinse down the sides of the tube and the column joint
with hexane. The volume used will depend on the desired dilution.
NOTE: (1) When a minimal dilution is required after evaporation, a 100
|Jl syringe is useful in performing the hexane rinse. (2) To obtain a suitable
extract concentration for p_,p_'-DDE, it is generally necessary to adjust the
extract volume to a level in excess of 1 ml. In this case, add hexane until
the meniscus is exactly at the 1-ml mark on the concentrator tube. Then use
a 1-ml Mohr pipet for total volumes up to 3 ml. For larger volumes, use a 5
ml Mohr pipet, carefully measuring the volume of hexane delivered. Above
308
-------
the 1 ml graduation mark, the concentrator tube calibrations are-not suffi-
ciently accurate for use in this- analysis. 'It is also good practice to
check the graduation marks up to 1 ml for all concentrator tubes used in
this analysis.
6. Stopper the concentrator tube and hold on the Vortex mixer, set
for high speed for ca. 30 seconds for volumes of. 6 ml or less. It
is safer practice to mix a full minute for larger volumes.
6.2.2 Urine Samples
The urine is treated exactly as the blood serum except that 5 ml are
analysed rather than 2 mL.
6.2.3 Optional Florisil Cleanup
If interferences from unwanted electron capturing materials hinder
analysis and quantitation of desired peaks, a Florisil column cleanup can be
incorporated into the extraction scheme. The procedure for pooled serum is
as follows:
6.2.3.1 Extraction
1. Measure 50 ml of serum into a 1 L sep. funnel containing 190
ml of CH-CN, 200 ml of aqueous 2% Na.SO, and 50 ml of hexane.
2. Stopper, shake funnel vigorously 2 minutes, and allow the
layers to separate.
3. Draw off the a'queous (lower) layer into a second 1-L sep.
funnel and percolate the hexane layer through a 2-inch
column of anhydrous Na0SO, into a 500 ml Kuderna-Danish
24
flask fitted with a 10-ml grad., evap. concentrator tube
containing one 3-mm glass bead.
4. Add another 50-ml portion of hexane to the aqueous solution
in the second 1-L separator; stopper and shake vigorously
another two minutes. When layers have separated, draw
aqueous layer back into the first 1 L separator and percolate
the hexane layer through the Na^SO, into the K-D flask.
Repeat the extraction twice more resulting in a total hexane
extract of 200 ml.
309
-------
5. Assemble K-D-evaporator and concentrate extract to ca 3 ml.
Disassemble evaporator rinsing tube joint with a small volume
of hexane and dilute extract to exactly 5 ml. Stopper and
shake on Vortex mixer 2 minutes.
6.2.3.2 Florisil Fractionation
1. Prepare a chromatographic column containing. 4 inches (after
setting) of activated Florisil topped with 1/2 inch of anhy-
drous, granular Na.SO,. A small wad of glass wool, preextrac-
ted with pet. ether, is placed at the bottom of the column to
retain the Florisil.
NOTE: (1) Florisil is activated by heating for at least 5 hr at 130°.
(2) If the oven is of sufficient size, the columns may be prepacked and
stored in the oven, withdrawing columns a few minutes before use. (3) The
amount of Florisil needed for proper elution should be determined for each
lot of Florisil.
2. Place a 500-ml Erlenmeyer flask under the column and prewet
the packing with pet. ether (40-50 ml, or a sufficient volume
to completely cover the Na^SO, layer).
NOTE: From this point and through the elution process, the solvent
level should never be allowed to go below the top of the Na?SO, layer. If
air is introduced, channeling may occur, making for an inefficient column.
3. Using a 5-ml Mohr or a long disposable pipet, immediately
transfer the tissue extract (ca 5 ml) from the evaporator
tube onto the column and permit it to percolate through.
4. Rinse tube with two successive 5-ml portions of pet. ether,
carefully transferring each portion to the column with the
pipet.
NOTE: Use of the Mohr or disposable pipet to deliver the extract
directly onto the column precludes the need to rinse down the sides of the
column.
5. Prepare two Kuderna-Danish evaporative assemblies complete
with 10 ml graduated evaporative concentrator tubes. Place
one glass bead in each concentrator tube.
310
-------
6. Replace the 500-ml Erlemneyer flask -under .each column with a
500 ml Kuderna-Danish assembly -and commence elution with 200
ml of 6% diethyl ether in pet. ether (Fraction I). The
elution rate should be 5 ml per minute. When the last of the
eluting solvent reaches the top of the Na2SO, layer, place a
second 500 ml Kuderna-Danish assembly unde.r the column and
continue elution with 200 ml of 15% diethyl ether in pet.
ether (Fraction II).
7- To the second fraction only, add 1.0 ml of hexane containing
200 nanograms of aldrin, place both Kuderna-Danish evaporator
assemblies in a water bath and concentrate extract until ca
. 5.ml. remain in the tube.
8. Remove assemblies from bath and cool to ambient temperature.
9. Disconnect collection tube from Kuderna-Danish flask and
carefully rinse joint with a little hexane.
10. Attach modified micro-Snyder column to collection tubes,
place tubes back in water bath and concentrate extracts to 1
ml. If preferred, this may be done at room temperature under
a stream of nitrogen.
11. Remove from bath, and cool to ambient temperature. Disconnect
tubes and rinse joints with a little hexane.
NOTE: The extent of dilution or concentration of the extract at this
point is dependent on the pesticide concentration in the substrate being
analyzed and the sensitivity and linear range of the Electron Capture Detector
being used in the analysis.
• '* •»
12. Should it prove necessary to conduct further cleanup on the
15% fraction, transfer 10 grams MgO-Celite mixture to a
chromatographic column using vacuum to pack. Prewash with ca
40 ml pet. ether, discard prewash and place a Kuderna-Danish
receiver under column. Transfer concentrated Florisil eluate
to column using small portions to pet. ether. Force sample
and washings into the MgO-Celite mixture by slight air pressure
and elute column with 100 ml pet. ether. Concentrate to a
suitable volume and proceed with Gas Liquid Chromatography.
311
-------
NOTE: Standard Recoveries should be made through
quantitative recoveries.
6.3 Analysis
Detection and quantification of semi-volatile halogenated hydrocarbons
is made using a model MT-220 gas chromatograph manufactured by Tracor, Inc.
Austin, TX, equipped with a tritium foil electron capture detector. Separa-
tion is achieved using a 6 foot x 5/32" i.d. silanized glass column packed
with 1.5% OV-17/1.95% OV-210 on a 80/100 mesh silanized support. Confirmatory
separation may be effected using an identical column packed with 5% OV-210
on a 100/120 mesh silanized support. A flow rate of 50-70 ml/min is recommen-
ded for the first column at an operating temperature of 200° (isothermal)
while 45-60 ml-/min is recommend for the second column at an operating tempera-
ture of 180° (isothermal). Both columns should be operated with a detector
temperature of 205°C.
NOTE: Laboratories may substitute GC/ECD equipment from other manufac-
tures and use 2 mm i.d. glass columns with a carrier gas flow rate of 20-30
mL/min.
6.4 Qualitative Identification
The peaks obtained are qualitatively identified by relative retention
times which are listed in Tables 1 & 2 for the primary and the confirmatory
columns. If the relative retention times are correct for both columns a
nearly positive identification has been made. Identification can be improved
by such techniques as TLC and/or electrolytic conductivity detection (3).
6.5 Quantitation
Quantisation is accomplished using either peak height if the peaks are
tall and narrow or peak ht. x width at half height if the peaks are symmetric
and rather broad (e.g., late eluting peaks). These peak measurements are
compared to peak measurements of standard solutions of known concentration.
It must be remembered that the peak heights of the sample and standard
should be within 25% of one another to assure precise quantitation due to
the small linear dynamic range of an ECD. It is also recommended that a
sample injection of at least 5 pi be employed to minimize injection error.
312
-------
Table 1. RELATIVE RETENTION TIMES ON
1.5% OV-17/1.95% OC-210
Column T«mp«ratur« , *C.
170
0.25
0.32
'0.34
0.30
0.44
0.42
0.48
O.S4
O.S6
O.S4
0.67
0.65
0.66
0.76
0.82
0.82
0.94
1.00
1.17
1.17
1.49
1.41
1.49
1.71
1.70
1.82
2.07
1.92
2.02
2.14
2.32
2.15
2.20
2.75
2.97
2.80
3.34
3.26
3.47
3.98
4.65
4.4S
5.57
6.1
6.4
10.7
13.1
12.4
16.9
22.1
1
170
|
0.25
0.32
0.34
0.31
0.45
0.42
0.48
0.54
0.56
0.54
0.67
0.65
0.67
0.76
0.82
0.82
0.94
1.00
1.16
1.16
1.41
1.40
1.49
1.69
1.69
1.80
2.04
1.91
2.00
2.12
2.28
2.13
2.16
2.72
2.93
2.77
3.29
3.23
3.43
3.94
4.§7-
4.39
5.48
5.97
6.2
10.5
12.7
12.1
16.1
21.5
I
174
|
0.26
0.32
0.35
0.38
0,4$
0.43
0.49
0.54
0.56
0.55
0.66
0.65
0.67
0.76
0.82
0.82
0.93
r.oo
1.15
1.16
1.45
1.39
1.47
1.67
1.68
1.78
2.01
1.89
1.99
2.09
2.25
2.11
2.16
2.68
2.88
2.75
3.25
1.19
1.40
3.88
4.49
4.34
5.39
S.8S
6.1
10.3
12.4
11.8
U.I
20.9
1
174
,
0.26
0.32
0.35
0.38
0.45
0.43
0.49
O.M
0.56
0.55
0.66
0.65
0.67
0.76
0.82
0.82
0.92
1.00
1.14
1.15
1.43
1.38
1.47
1.66
1.67
1.76
1.98
1.88
1.97
2.07
2.22
2.09
2.1S
2.64
2.84
2.72
1.20
3.16
1.36
1.83
4.41
4.28
S.29
5.73
5.99
10.1
12.0
11.6
15.7
20.3
•
17*
1
0.26
0.32
0.16
0.39
0.45
0.44
0.50
O.SS
O.S6
0.56
0.66
0.66
0.67
0.75
0.81
0.82
0.92
1.00
1.13
1.14
1.41
1.36
1.46
1.64
1.66
1.74
1.95
1.86
1.95
2.0S
2.19
2.07
2.13
2.61
2.79
2.69
3.15
3.13
3.33
3.77
4.33
4.23
5.20
5.61
5.88
9.9
11.6
11.3
15.3
19.6
1
178
j
0.27
0.32
0.16
O.M
0.45
0.44
0.50
O.SS
0.56
0.56
0.66
0.66
0.67
0.75
0.81
0.82
0.91
1.00
1.11
1.14
1.40
1.35
1.45
1.62
1.65
1.72
1.92
1.85
1.93
2.03
2.16
2.05
2.11
2.58
2.75
2.67
3.11
3.09
3.29
3.71
4.26
4.17
5.11
5.49
5.76
9.7
11.2
11.0
14.9
19.0
1
-482-
|
0.27
0.32
0.36
0.39
0.45
0.44
0.50
0.55
0.56
0.56
0.66
0.66
0.67
0.75
0.81
0.82
0.90
1.00
1.10
1.13
1.38
1.34
1.44
1.60
l.M
1.70
1.89
1.83
1.92
2.01
2.13
2.03
2.10
2.M
2.71
2.64
3.06
3.06
3.26
3.66
4.18
4.11
5.01
5.36
5.64
9.5
10.8
10.7
14.5
18.4
1
182
1«6~ MO-- 19* -
, I , I ,
0.27 0.28 0.28 0.28 0.: ) 0.29 (
0.32 0.32 0.12 0.33 0.:3 0.33 I
0.17 0.37 0.38 0.38 0.38 .39 1
0.19 0.39 0.40 0.40 0.4*. .40 I
0.46 0.46 0,46 0.46 0.46 .46
0.45 0.45 0.45 0.46 0.46 .47 I
0.51 0.51 0.52 0.52 O.Si. .53 I
0.55 O.SS 0.55 O.SS O.SS .56
0.56 0.56 0.56 0.56 0.56 .56 I
0.57 0.57 O.SB 0.58 0.58 .59 1
0.65 0.65 0.65 0.65 0.65 .65 (
0.66 0.66 0.67 0.67 0.67 0.67 I
0.68 0.68 0.68 0.68 0.68 0.68 I
0.75 0.75 0.74 0.74 0.74 0.74 I
0.81 0.81 0.81 0.81 0.80 0.80
0.82 0.82 0.82 0.82 0.82 0.82
0.90 0.89 0.88 0.88 0.67 0.87
1.00 1.00 .00 1.00 1.00 1.00
1.09 1.08 .07 1.06 l.OS 1.03
1.12 1.12 .11 1.10 1.09 .09
1.36 1.34 .32 1.31 1.29 .27
1.33 1.32 .31 1.30 1.29 .28
1.44 1.43 .42 1.42 1.41 .40
1.59 1.57 .55 1.53 1.52 .50
1.63 1.62 .61 1.59 1.58 .57
1.68 1.66 .64 1.62 1.60 .58
1.87 1.94 .81 1.78 1.75 .72
1.81 1.80 .78 1.77 1.75 .74
1.90 1.88 .86 1.85 1.83 .82
1.98 1.96 .94 1.92 1.90 .88
2.09 2.06 2.03 2.00 1.97 .93
2.01 1.99 1.97 1.96 1.94 .92
2.08 2.06 2.05 2.03 2.01 2.00
2.51 2.47 2.43 2.40 2.37 2.33
2.66 2.62 2.57 2.53 2.49 2.44
2.61 2.59 2.56 2.53 2.51 2.48
3.01 2.97 .92 2.88 2.83 2.77
3.03 1.00 .96 2.93 .90 .87
3.22 3.18 .15 3.12 .08 .04
3.60 3.54 .46 3.43 .38 .32
4.10 4.02 .94 3.87 .79 .71
4.05 3.99 .94 3.88 .82 .76
4.92 4.83 .74 4.64 .55 .46
5.24 5.12 .00 4.88 .76 .64
5.52 5.40 5.28 5.16 .04 .92
9.3 9.1 8.9 8.7 .5 .3 1
10.4 10.0 9.7 9.3 .9 .5 1
10.4 10.1 9.8 9.5 .3 .0 1
14.1 13.7 13.3 12.9 12.5 12.1 1
17.7 17.1 16.5 15.8 15.2 14.6 1'
1 | l | l
186 190 194
1
1
1.29 (
).33 I
).39 I
3.41 I
1.47 I
1.47 I
1.53. I
).56 1
3.56 I
3.59 I
1.64
1.67 (
).69 (
3.73 <
1.80 1
9.82
3.86 <
.00
.02
.08
.25
.27
.39
.48
.56
.56
.69
.72
.79
.86
.90
.90
.98
.30
.40
.45
2.74
2.83
J.01
J.27
J.M
1.71
1.36
1.52
1.80
1.1.
1.1
1.7 1
.7 1
1.0 1
1
98- I
1
1.30 O.M
).33 0.33
3.40 0.40
3.41 0.41
3.47 0.47
3.48 0.48
LM O.M
1.56 0.56
3.56 0.56
3.60 0.60
1.64 0.64
3.68 0.68
1.69 0.69
3.73 0.73
3.80 0.80
3.82 0.82
1.85 0.85
.00 .00
.01 .00
.07 .07
.23 .22
.26 .25
.39 .38
.47 .45
.55 .54
.54 .52
.66 .63
.71 .69
.78 1.76
.84 1.82
.87 1.84
.88 1.86
.97 1.95
1.27 2.23
2.35 2.31
2.43 2.40
2.69 2.65
2.80 2.77
2.97 2.93
3.21 3.16
3.61 3.46
1.65 3.59
1.27 4.18
1.40 4. 28
1.66 4.56
E 9 £ 1
1. fc O.I
1.9 7.7
1.7 7.3
1.4 8.1
.3 10.9
J.4 12.7
I '
96
20? "" 204 '
| |
0.30 0.31
0.33 0.33
0.40 0.41
0.41 0.41
0.47 0.47
0.48 0.49
O.M O.SS
0.56 0.56
0.56 0.56
0.60 0.61
0.64 0.64
0.68 0.68
0.69 0.69
0.73 0.73
0.80 0.80
0.82 0.82
0.84 0.83
1.00 1.00
0.99 0.98
.06 1.05
.20 1.18
.24 1.23
.37 1.36
.43 1.41
.53 1.52
.50 1.48
.60 1.57
.68 1.66
.74 1.73
.79 1.77
.81 1.78
.84 1.82
.93 1.91
2.20 2.17
2.^7 2.22
2.37 2.35
2.60 2.56
2.74 2.70
2. «0 2.87
3.10 3,04
3.40 3.32
3.54 3.48
4.0$ 4.00
4.16 4.04
4.44 4.3:
60 C Ike
. W 9. O3
7.5 7.3
7.0 6.6
7.8 7.5
10.5 lO.i
12.1 11.5
1 '
202 204
•--
Compound
•OTwthyl PMhalite
HtMnphoi
TlA4Ztfl<
Olcthyl PhthaUu
2.4-l)(ME)
HcMcMoroBcntMi
e-EMC
COEC
2.4-OdeC)
CMorde«>
01»l1non
PCNB
Llndane
2.4.S-HME)
I-8HC
Htpttchl^
2.4.S-T(Irt)
Aldrtn (REFERENCE)
DlMthoate
Ronnel
01 butyl PMhalate
1 -Hydroiyehl ordene
Oxychlordane
H. ParathlOP
Heptachlor Epoitde
DCPA
Nalathton
Chlordane, Gftjirj
lyvu-Monachlor
o.p'-DDE
E. Pirathlon
Chi ordene, Alpha
Endosulfan I
p.p'-OOE
DOA(«)
DUIdrln
o.p'-DOO
Cltlordtcont
Endrln
e.p'-DCT
p.p'-DDO
Endosulfin 11
p,p'-30T
EtMen
Carhop c :• ' i
Endrln Ketont -153
Dloctyl Phthaliti
Nttnoxychlor
Yttrad1fon
Otphcnyl Phthalate
Retention re tie*, re let IT* to eldiln. of 4* compound. «t te*v»reture* froej
1*0 to }04*Ci wpfort of 0*e Chra (l> 100/UO Beehi »l»ctxon capture detected
trttluB eouroe, pexellel pletei ell ebeolute retention* neeeured froo Injection
point, krrav Indicated optlenai coliewi operetlne, tee^erature with carrier flow
•t M ml per minute.
313
-------
Table 2. RELATIVE RETENTION TIMES ON 5% OV-210
Column Tompcrotur* , CC.
177
(
0.43 0.43
0.51 0.51
0.52 0.53
0.58 O.S9
0.58 0.58
0.65 0.66
0.69 0.69
0.73 0.73
0.75 0.75
0.78 0.79
O.SS 0.85
0.83 0.83
0.86 9.86
0.93 0.93
1.00 1.00
1.41 1.39
1.44 1.43
1.59 1.58
1.66 1.64
1.88 1.87
1.88 1.87
1.96 1.94
2.05 2.03
2.25 2.21
2.21 2.18
2.21 2.19
2.54 2.52
2.60 2.58
2.69 2.65
2.83 2.79
2.97 2.92
1.00 2.95
2.95 2.91
3.08 3.05
1.71 1.66
4.01 1.94
4.45 4.31
4.15 4. Of
4.38 4.31
4.78 4.70
5.28 5.17
5.90 5.77
7.3 7.1
13.6 13.1
12.9 12.5
29.0 19.4
21.0 20.4
'
170
3.44 |
0.51
0.53 (
0.59 I
0.59 (
0.66 (
0.69 (
0.72 (
0.75 (
0.79 (
O.SS (
0.34 (
0.87 (
0.92 (
1^00 1
.38 1
.42 1
.57 1
.63 1
.85 1
1.85 1
1.92 1
2.02 i
2.17 i
2.16 i
2.16 i
2.49
2.55
2.61
2.76
2.86
2.89
2.87
3.01
3.61
1.88
4.17
4.03
4.23
4.63
5.06
5.63
6.9
12.7 i;
12.3 12
18.9 11
19.7 1!
1
174
).45
1.51
1.54
).60
).60
1.66
1.69
1.72
).74
1.79
.84
1.84
1.87
1.92
.00
.37
.41
.56
.61
.83
.83
.91
t.OO
E.13
E.13
E.13
!.*6
.53
.57
.72
.81
.84
.82
.98
.56
.81
.04
.98
.16
.55
.95
.50
.7
'.3
E.O
1.3
1.1
178
0.45
0.52 I
0.54 (
0.60 (
0.61 (
0.67 I
0.69 (
0.72 (
0.74 (
0.79 (
0.34 (
0.84 (
0.87 I
0.92 I
1.00
1.35
1.40
1.55
1.60
1.81
1.82
1.89
1.98
2.10
2.10
2.11
2.44
2.50
2.55
2.69
2.75
2.79
.78
.94
.51
.74
.90
.92
.08
.48
.84
.36
.55
11.9 1
11.8 11
17.8 V
18.5 i;
1
178
»
]
1.46
).S2
1.55
1.61
1.62
1.67
1.69
J.72
).74
).79
>.84
1.84
1.87
).92
.00
.34
.39
.54
.SB
.80
.80
.87
.96
.06
.09
.09
.41
.4«
.49
.66
.70
.73
.74
.91
.44
.67
.76
.86
.01
.40
.73
.23
.4
.4
.5
r.i
r.s
1
ie:
I
0.46
O.S2
0.55
0.61
0.62
0.67
0.69
0.72
0.74
0.80
0.84
0.84
0.87
0.92
1.00
1.33
1.38
1.53
1.S6
1.78
1.78
1.85
1.94
2.02
2.04
2.0S
2.38
2.46
2.45
2.62
2.65
2.68
2.70
2.87
3.41
3.60
3.62
1.80
1.93
4.13
4.62
S.09
6.2
11.0
11.2
18.7
17.2
1
182
186 190 194
11 I
'111
0.47 0.48 0.48 0.49 ".49 0.50 O.S1
0.53 0.53 0.53 0.53 0.54 0.54 0.54
0.55 0.56 0.56 0.57 0.57 0.58 0.58
0.62 0.62 0.63 0.63 0.64 0.64 0.65
0.63 0.64 0.65 0.66 0.66 0.67 0.68
0.67 0.68 0.68 0.68 0.69 0.69 0.69
0.69 0.69 0.68 0.68 0.68 0.68 0.68
0.71 0.71 0.71 0.71 0.71 0.70 0.70
0.73 0.73 0.73 0.72 0.72 0.72 0.71
0.80 0.80 0.80 0.80 0.81 0.81 0.81
0.84 0.83 0.83 0.83 0.83 0.83 0.82
0.84 0.34 0.84 0.84 0.84 0.85 0.8S
0.87 0.*' 0.87 0.87 0.87 0.88 0.88
0.92 0.9? 0.92 0.92 0.92 0.91 0.91
.00 .00 .00 .00 1.00 1.00 1.00
.32 .30 .29 .28 1.26 1.25 .24
.37 .36 .36 .35 1.34 1.33 .32
.52 .SI .50 .49 1.48 1.47 .45
.55 .S3 .52 .50 1.48 1.47 .45
.76 .74 .73 .71 1.69 1.68 .66
.77 .75 .73 .72 1.70 1.68 .67
.84 .83 .80 .79 1.77 1.75 .74
.92 .90 .89 .87 1.85 1.83 .81
.99 .95 .91 .87 1.84 1.80 .76
2.01 .98 .95 .92 1.89 1.87 .84
r.02 2.00 .97 .94 1.92 1.89 .85
2.35 2.33 2.30 2.27 2.25 2.22 2.19
2.43 2.41 2.38 2.36 2.34 2.31 2.29
2.41 2.37 2.33 2.29 2.25 2.21 2.17
2.59-i.SS 2.52 2.49 2.45 2.42 2.38
2.59 2.54 2.48 2.43 2.38 2.32 2.27
2.63 2.58 2.52 2.47 2.42 2.36 2.31
2.66 2.61 2.57 2.53 2.49 2.45 2.40
2.84 2.80 2.77 2.71 2.70 2.66 2.63
3.36 3.31 1.26 1.21 1.12 1.11 1.06
1.51 1.4C 3.39 3.32 3.25 1.19 3.12
1.49 3.35 3.21 3.08 2.94 2.80 2.67
3.74 3.69 3.63 3.57 1.51 1.45 1.40
1.85 1.78 1.70 1.61 1.55 1.47 1.40
4.25 4.18 4.11 4.03 3.96 1.88 1.81
4.51 4.40 4.28 4.17 4.06 1.95 1.84
4.96 4.82 4.69 4.55 4.42 4.28 4.15
(.0 5.84 5.66 5.49 5.31 5.13 4.95
10.6 10.2 9.7 9.3 8.9 8.5 8.0
10.9 10.6 10.3 10.1 9.8 9.5 9.2
16.2 1S.6 15.1 14.6 14.0 13.5 13.0
16.5 15.9 1S.2 14.6 13.9 13.3 12.6
1 | 1 | 1 | 1
186 190 194
i
i
O.S1
0.54
0.59
0.65
0.69
0.70
0.68
0.70
0.71
0.81
0.82
0.85
0.88
0.91
1.00
1.22
1.31
1.44
1.44
1.64
1.65
1.72
1.79
1.72
1.81
1.84
2.17
2.26
2.13
2.35
2.21
2.26
2.36
2.59
3.01
3.05
2.53
3.34
3.32
3.73
3.73
4.01
4.77
7.6
8.9
12.4
12.0
1
1M
1
0.52
O.SS
0.59
0.66
0.70
0.70
0.68
0.70
0.71
0.81
0.82
0.85
0.88
0.91
.00
.21
.30
.43
.42
.62
1.63
1.70
1.77
1.69
1.78
1.81
2.14
2.24
2.09
2.32
2.16
2.21
2.32
2.56
2.96
2.98
2.39
.28
.25
.66
.62
.89
.60
7.2
8.6
11.9
11.4
— r—
:*
o.s?
O.SS
0.60
0.67
0.70
0.70
0.68
0.70
0.71
0.82
0.82
0.8S
0.88
0.91
.00
.20
.29
.42
.40
.61
.62
.68
.75
.65
.75
.78
2.11
2.22
2.05
2.28
2.11
2.15
2.28
.52
.91
.91
.25
.22
.17
.58
.51
.74
.42
.8
.4
11.3
10.7
202
.'04
1
O.S3
O.SS
0.60
0.67
0.71
0.70
0.68
0.69
0.70
0.82
0.82
O.SS
0.83
0.91
1.00
1.19
1.28
1.41
1.39
1.59
1.60
1.67
1.74
1.61
1.72
1.75
2.08
2.19
2.01
2.25
2.05
2.10
2.24
2.49
2.86
2.84
2.12
3.16
3.09
3.50
3.40
3.61
4.2*
6.4
8.1
10.8
10.1
— 1
204
C J^V- !*_** 3
Nt-Mcmon**,,*,
Olisetnyl Phtftal.]*.
T*cnazene
Chlordene
•-BHC
CDEC
Nevinpnos
Dlethyl Phtnalate
Olazinon
Undane
2.4-0(!PE)
PCNB
Ntptacnlor
I-BHC
AldMn (REFERCM;!]
ftonne 1
1 -Hydroaychl ordf ""
Oiycnlarair.e
0,fl '-DIE
rrorj-NonacMor
Chloriane, Garni
Hcotaehlor Eponte
Chlordane, Alom
Dlbutyl Phthalitt
Olmetnoate
p.p'-cor
EndotuUan I
o.p'-DDO
OCPA
Chlordecone
0,p'-OOT
NalatrHon
M. ParatMon
Oleldrin
Enarln
p.p'-DOO
E. ParttMcn
Mtrex
p,p'-OOT
Endosulfar. 11
Carboonenotnlon
EtMen
Metnsiyc^l"-
Dloctyi Pntna'..*.«
Endrin Ketone '15''
Tttradtfpn
Olphenyl Withalati
Retention ratlot, relatl** to aldrtn, of 47 eoneoundi tt tenoerituroi fro*
JJO to 204*C; Support of fit Chren Q, 80/100 «**«. oltetron capture detector;
•™ source; all abwluU retentions matured from Injection point. Arrow
Indicated optliwe calm operating tmerature with carrier f1on at SO •!
314
-------
*7.0 Quality Assurance .Program ...
An on-going quality assurance program is required to assure the data
quality. Quality control (QC) procedures determine artifacts, losses, etc.
through a system of blanks and controls. Quality assurance (QA) procedures
monitor the execution of the procedure and check data interpretations and
calculations. - ,
7.1 Quality Control
7.1.1 Field Blanks and Controls
Prior to a field sampling trip, enough blanks and controls are prepared
to equal 10% each (2 minimum) of the anticipated number of field samples.
Blanks consist of 10 ml of water in the same type of sampling container as
is used in the- field. Controls consist of 10 ml of plasma spiked at 10-15
ng/ml with the compounds listed in Table 3. For urine 10 ml of water serves
as a blank and 10 ml of water spiked with 10-15 ng/ml of the compounds in
Table 3 serves as a control sample. These blanks and controls are carried
to the field and receive the same handling as the field samples. Workup and
analysis of field blanks and controls is interspersed with the field samples
on a regular basis. This method allows assessment of sample storage stability
Table A presents a typical set of blanks and controls for QC on a field
trip where 50 blood samples are to be collected.
7.1.2 Procedural Blanks and Controls
With each set of samples, a procedural blank is run. This consists of
10 ml of prepurged distilled water which is extracted under the same condi-
tions as the samples. These blanks are designed to detect artifacts from
dirty glassware, laboratory atmosphere intrusion, and other sources.
7.2 Quality Assurance
Both internal and external QA procedures are to be followed. Internal
QA procedure assure the continuity and consistancy of the data. External QA
procedures (interlaboratory checks) verify or dispute the accuracy of the
data being generated in the primary laboratory.
7.2.1 Internal Quality Assurance
7.2.1.1 Supervision and Monitoring of Activities
There are three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This
315
-------
Table 3. SEMI -VOLATILE _HAL0_GEJJATEH HYDROCARBONS -JR.
METHANOL SPIKING SOLUTION '
Compound
Compound
4-Chlorobiphenyl
Cf-BHC
3-BHC
Y-BHC
4,4'-Dichlorobiphenyl
2,4,5-TrIchlofobiphenyl
Heptachlor
Aldrin
Heptachlor epoxide
Dieldrin
P_,E'-DDE
£,£'-DDT
2,2',3,3',6,6'-Hexachlorobiphenyl
trans-Nonachlor
Oxychlordane
HCB
316
-------
Table 4. - -BLQOD ,QC ^SAMPLES- EOR^A-XYPICAL SAMELIMG -TRIE
Sample Type
Number
Comments
Field Blank
Field Control
Lab Blank
Lab Control
5
5
Freeze after preparation,
carry to field, store with
field samples.
Store with field blanks.
Freeze after preparation,
store in same freezer as field
samples will be stored.
Store with Lab Blanks
317
-------
person must be aware ofT_their_a.c^^^Qb$e,^^^
data, and maintain appropriate records. At the second level, the chemist1s
supervisor monitors their daily activities, reviews the notebook, checks
data and calculations, and assists in "troubleshooting" problems. At the
tertiary level, a QA coordinator interviews all personnel on the project.
The interviews cover the operations they perform (precisely), the data they
obtain, a spot-check of their calculations, and any problems they have had.
7.2.1.2 Documentation
Chain of Custody
From the initial preparation of a sample container through reporting of
the analytical results, each sample is accompanied by a chain of custody
sheet. Each person signs in the time of receipt, operations performed, and
transmittal of the sample. This record is important for tracing a contaminant
bad standard, or some other problem.
Sampling Protocol Sheets
When a sample is collected, a sampling protocol sheet is filled in
which contains a discrete sample code which identifies project number, area,
site, locations, trip number, sampling period, and sample type. Also included
are sample times, volumes, address and other pertinent information. Where
appropriate, a map is made to precisely identify the location.
Sample Log
Upon return from a sampling trip, each sample code is entered into a
sample log book. This log is updated as samples proceed through workup and
analysis. Thus, at a glance, project personnel can tell the status of each
sample and find out how many are at different stages in the analytical
protocol.
GC/MS Log
Each sample run by GC/MS is logged into a notebook, detailing analysis
conditions, where the data are archived, and what hardcopy data has been
produced.
7.2.2 External Quality Assurance
Samples will be analyzed by a designated QA laboratory. Samples,
controls, and blanks will be shipped directly from the field to the QA
318
-------
laboratory for analysis. -They will report, the results to the primary labora-
tory for correlation with the prima-ry data.
7.2.2.1 Selection of Samples for QA
Approximately 10% of the field samples (2 minimum) will be collected in
duplicate (i^e., two 10 ml blood or two 120 ml urine samples) for shipment
to the QA laboratory. This selection process will be random with the follow-
ing restrictions:
(1) The donor must consent to the additional blood collection.
(2) If any stratification of donors is known, purposive selection of
QA donors may be used to get representative samples (e.g.,
occupationally exposed vs. "normal" individuals or upwind vs.
downwind -residents).
7.2.2.2 QC Samples
Along with the QA duplicate samples, at least one QC control and one QC
blank must be included with the QA samples. An example is shown in Table 5
for a trip collecting 50 samples.
7.2.2.3 Sample Codes
Sll samples shipped to the QA laboratory should be encoded so that the
laboratory cannot discern between samples and controls. This "blind" protocol
will add validity to the results.
7.2.2.4 Shipping
Samples should be shipped on dry ice directly to the QA laboratory by
an appropriate air carrier (e.g., Federal Express, Eastern Sprint) in well
insulated and packed cartons.
8.0 References
1. Sherma, J., "Manual of Analytical Quality Control for Pesticides in
Human and Environmental Media" EPA -600/1-76-017, 2K, 33 (1979).
2. ibid, Section 4, A, (6), p 2.
3. ibid, Section 4, A, (4), p 4.
319
-------
Table 5. SAMPLES TO BE COLLECTED AND SHIPPED-TO QA LABORATORY
FOR A TYPICAL SAMPLING TRIP
Sample Number Comments
Duplicate Sample 5 Random selection unless prior
information stratifies subjects
Field Blank 1 Ship with samples
Field Control 1 Ship with samples
320
-------
Adapted from:
Thompson, J. F., Analysis of Pesticide Residues in Human and Environmental
Samples, A Compilation of Methods Selected for Use in Pesticide Monitoring
Programs, Environ. Toxicol. Div., Health Effects Research Lab., USEPA, RTF,
NC, December, 1977. Sections 4A, 5A(1), and 5A(3)(a).
* Indicates a section not in original protocol, added here for application
to this research project.
321
-------
ANALYTICAL raQTQCpLjL^AlffiU^^M^H^^ ":>
ORGANICS IN BLOOD AND URINE (RTI)
1.0 Principle of Method
Semi-volatile halogenated hydrocarbons are extracted from blood plasma
or urine with organic solvents, dried, and concentrated to. an appropriate
volume for quantification using a gas chromatograph/electron capture detector
(GC/ECD). Identifications are confirmed by GC/ECD using a second column
and, when sufficiently concentrated, by GC/MS/COMP- Blood and urine samples
are optionally subjected to liquid chromatographic cleanup on Florisil if
severe interferences are encountered. This procedure was adapted from that
of Thompson (t).
2.0 Range and Limit of Detection
The sensitivity of response to GC/ECD is a function of the instrument,
the compound, and the matrix from which it is extracted. Acceptable recove-
ries from human plasma at approximately 5 ppb (parts per billion) have been
achieved.
3.0 Interferences
Interferences in sample analysis and quantification using GC/ECD are
manifested in the electron capturing ability of the given contaminant.
Blood extracts which have not been cleaned up contain rapidly eluting compo-
nents which exhibit high electron capturing properties; these interferences
can largely be removed by gradient liquid-solid chromatography on 2% aqueous
deactivated Florisil, as discussed below.
4.0 Precision and Accuracy
Recovery studies were initiated with a wide variety of model halogenated
hydrocarbons (e.g., trifluralin, a-BHC, p-BHC, y-BHC, 2,4,5-trichlorobiphenyl,
heptachlor, aldrin, heptachlor epoxide, endosulfan, p_,p_'-DDE and dieldrin).
For 2.0 ml aliquots of human plasma and 3.0 ml aliquots of urine spiked with
14 ng of the above halogenated hydrocarbons and equilibrated for 19 hr at
4°C, a mean recovery of 53.1 ± 12.6% was obtained for blood plasma (2).
322
-------
5.0 Apparatus and Reagents
5.1 Sampling Apparatus
Blood samples are obtained in 10 ml vacutainer tubes "suitable for GC"
(Venoject L428, Kimble) containing 15 rag tripotassium salt of EDTA and 20 pg
potassium sorbate.
5.2 Extraction Apparatus
1. Glass culture tubes (16 x 125 mm) and caps equipped with Teflon
liners, reciprocal shaker (ca. 40 oscillations/minute);
2. 500 ml Kuderna-Danish evaporators, receiving tubes and three ball
Snyder columns;
3. glass bottles and caps equipped with Teflon liners;
®
4. reactivials ;
5- centifuge
6. 22 mm i.d. chromatography columns.
5.3 Solvents and Reagents
1. Hexane (Burdick and Jackson) distilled in glass, redistilled prior
to use.
2. Methylene chloride (Burdick and Jackson) distilled in glass,
redistilled prior to use.
3. Anhydrous sodium sulfate, (extracted with pentane in Soxhlet for
24 hr and stored in oven at 140°C).
4. Florisil, 60/100 mesh (activated by heating for at least 5 hr
at 130QC).
NOTE: For a more complete treatment on the handling and charac-
teristics of Florisil batches refer to the "EPA Manual of Analy-
• '* i*
tical Methods for the Analysis of Pesticide Residues in Human and
Environmental Samples", Section 3, D.
6.0 Procedure
6.1 Collection of Samples
Blood is obtained from volunteers by a qualified phlebotomist using
vacutainers. Samples are inverted to ensure dispersal of the anticoagulant
and are then centrifuged for 45 min at 9000 rpm. Plasma is then pipetted
off, stored in a shell vial with teflon-lined screw caps and frozen as soon
as possible.
323
-------
First morning urines are collected by the volunteers-^Ln 1-20 ml wi-de-mou-
thed bottles. The samples are then picked up by a member of the sampling
team, and stored in a refrigerator or on ice until ready for analysis.
6.2 Extraction, Cleanup, and Concentration
6.2.1 Plasma
1. Transfer 2.0 ml of plasma to a 12 x 125 mm culture tube.
2. Pipet in 7.0 ml of hexane and cap.
3. Shake on a shaker bath at 40 oscillations/minute for 24 hr.
4. Remove sample from bath, centrifuge for v5 minutes to improve
separation of layers, transfer 5.0 ml of organic extract to glass
bottle with Teflon liner.
5. Pipet 5.0- ml additional hexane into serum and repeat (3).
6. Repeat (4), combine extracts and dry for at least 30 min over
0.5 g anhydrous Na.SO,.
7. Transfer to a 500 ml K-D flask (micro K-D optional), concentrate
to 2-4 ml and cool to ambient temperature.
8. Rinse sides of K-D with 1.5 ml hexane and blow under dry nitrogen
to about 1 ml.
®
9. Transfer to a reactivial previously calibrated to a specific
volume and further concentrate.
6.2.2 Urine
1. Transfer 5.0 ml of urine to a 12 x 125 mm culture tube.
2. Follow steps 2 through 9 of the procedure for the extraction of
blood serum for organochloride pesticides.
6.3 Instrumental
The detection and quantification of semi-volatile halogenated hydrocar-
bons is made using a Series 4400 Fisher/Victoreen Gas Chromatograph equipped
with a tritium foil electron capture detector. Separation is effected on a
40 m, 0.38 mm i.d., glass SCOT capillary column coated with 1% SE-30 on
0.32% Tullanox (3,4). Maximum efficiency is obtained with a flow rate of
2.5 ml/min of nitrogen gas with makeup nitrogen gas adjusted to a total flow
of 25.0 ml/min, column 220°C (isothermal), and detector 285°C.
As a confirmatory column a 190 cm x 0.2 cm i.d. 1.5% OV-17/1.95% QF-1
on 80/100 Chromosorb W-HP packing is employed. Efficient responses are
324
-------
obtained for flow rates of 18 ml/rain at identical column and detector tempera-
tures .
Final confirmation of the identity of the components of sufficiently
concentrated extracts (generally greater than 10 ng/pl) can be made using
gas chromatography/mass spectrometry/computer (GC/MS/COMP) (Finnigan 3300).
The GC/MS/COMP systems used are a Finnigan 3300 GC/MS/COMP and an 1KB
2091 GC/MS equipped with an 1KB 2031 data system. Chromatographic conditions
for the Finnigan 3300 are 20 m x 0.38 mm i.d., 1% SE-30 SCOT capillary
operated isothermally at 235°C and a flow rate of 2.0 ml/min helium. Split-
less injection (0.2-0.3 Ml) is used, with standard electron impact (70 eV)
ionization conditions.
The 1KB 2D91 is operated using a 18 m 1% SE-30/BaC03 WCOT capillary
column at 240°, isothermal for PCBs and a 40 m x 0.38 mm i.d. 1% SE-30
SCOT capillary column at 230° isothermal for the pesticides. In both cases,
the column flow rate is 2 ml/min with 20 ml/min split off at the injector.
The mass spectrometer is operated under standard electron impact conditions.
NOTE: Similar GC columns may be substituted for those prescribed. GC
conditions may change accordingly.
6.4 Qualitative and Quantitative Analysis
6.4.1 Qualitative Analysis
Alternate single injection of extracts and standard solutions is the
routine procedure for processing samples. If the retention time of a given
component of an extract suggests the presence of a standard compound, a
repetitive injection is then made. Tentative identification is made if the
deviation between the two respective means is no greater than three percent.
• '" •*
A similar criterion is then applied to the retention times of both extract
and standard component upon a second confirmatory column. Qualitative
identification of a component is made if both criteria are satisfied.
6.4.2 Quantitative Analysis
A mean linear response range of 5-160 pg/Ml has been established for
the compounds trifluralin and Y~BHC on a 1% SE-30/0.32% Tullanox 40 m, 0.38
mm i.d. SCOT capillary column installed in a Series 4400 Fisher/Victoreen
Gas Chromatograph. Quantification of a given component is made by a compari-
son of the means of recorder trace areas of two extract and two standard
325
-------
solutions within this linear response range. The precision of the concentra
tion of a given component is normally less than ten percent of the mean
concentrations and is obtained by propagation of the standard deviations of
the responses of both the extract and standard solutions. The effective
concentration multiplied by the volume of extract results in the total
amount of extracted material.
The areas employed in the quantification of blood extracts must be
normalized to compensate for the volume of extract that is not analyzed.
The removal of the first 5.0 ml aliquot from the 7.0 ml extract leaves 2/7
of the extracted material behind. Since the partition coefficients for
semi-volatiles in human plasma and urine are large, one approximates that
the second extraction contributes more as a dilution gradient than as an
extracting solvent. Under this approximation, 5/7 of the original 2/7
remaining in the organic extract would be obtained from the second aliquot.
One would then have 5/7 + (2/7)-(5/7) = 45/49 of the total initial extract
to analyze, and the area responses for the extracts would necessarily require
normalization by 49/45 = 1.089 prior to comparison to standard solutions.
This methodology was initiated due to the formation of complex emulsions
between the plasma and organic solvent which prevented the easy removal of
all of the organic extract. This procedure assumes an infinite partition
coefficient and that the second extraction does not influence the total
recovery.
If the extracts are colored, the presence of lipids may interfere with
either analysis or concentration of the extracts due to precipitation. In
this case, the sample should be transferred to a 22 mm i.d. column containing
1.6 g of 2% aqueous-deactivated Florisil and eluted with 10 ml each of
hexane, 5% MeCl2/hex, 10% MeCl2/hex, 15% MeCl2/hex, 20% MeCl2/hex, 30%
MeCl2/hex, 50% MeCl2/hex, and MeCU. The extracts are concentrated and
analyzed. Most semi-volatile halogenated hydrocarbons are expected to
appear in the first five fractions. This estimate is based upon elution
data of pesticides on Florisil (5) and has not been subjected to full experi-
mental verification.
326
-------
6.4.3 GC/MS/COMP Confirmation
The chromatography conditions are similar to those used for GC/ECD.
The samples for this study are to be screened by GC/ECD and confirmed (if
sufficiently concentrated) by GC/MS/COMP. Therefore, the retention times of
the two techniques must be similar. GC/ECD must operate isothermally, so
the GC/MS/COMP conditions reflect this restriction.
The Finnigan 3300 and the LKB 2091 systems may be operated in both the
full scan and selected ion monitoring (SIM) modes. In the full scan mode,
full spectra are collected. Spectra or mass fragmentograms (single ion
plots) may be plotted for interpretation. In the SIM mode, only a small
number (up to 9 for the Finnigan 3300 and up to 16 for the LKB 2091) of ions
are monitored." Full spectra are not collected. The advantage of this
method is that the detector spends more time "looking" at the selected ion
and therefore better (generally 10-50 times) sensitivity is obtained.
To determine the limits of detection, standard solutions of selected
pesticides and PCB isomers have been analyzed on the Finnigan 3300 and LKB
2091. In the full scan mode, the limit of detection was the amount of
compound required for an interpretable spectrum. - In the SIM mode, the limit
of detection was the amount of compound required to yield a peak 2-4 times
the noise level.
The estimated limits of detection for the Finnigan 3300 and LKB 2091
are presented in Table 1.
Quantitation using GC/MS/COMP is achieved by comparing the computer-cal-
culated integrated area of the unknown with the integrated response for a
known amount- of standard. To compensate for differences in ionization
cross-section, the relative molar response of authentic compounds is obtained
The calculation of the relative molar response (RMR) factor allows the
estimation of the levels of sample components without establishing a calibra-
tion curve. The RMR is calculated as the integrated peak area of a known
amount of the compound, A° , , with respect to the integrated peak area of a
UHK
known standard, A° , (in this case d.A-pyrene), according to the equation
std iu
327
-------
Table 1. ESTIMATED LIMITS OF DETECTION FOR EXTRACTABLE HALOGENATED
ORGANICS ANALYSIS3
LKB 2091b
Full scan
Compound ng/yl
Y-BHC (lindane) >12<20
heptachlor 12
chlordane -30
£,£' -DDE 12
2-chlorobiphenyl . -1
hexachlorobiphenyl <1
decachlorobiphenyl 12
m/z
181
272
375
246
188
360
498
SIM
ng/yl
0.10-0.4
0.10-0.4
5
>0.3
0.004
-0.016
0.42
Finnigan 33003
T, -,-, SIM
Full scan
ng/yl
5-10
10-20
25-50
5-10
-2.5
25-50
150
5/z
181
272
375
246
188
360
498
ng/yl
1
1-1.5
5-10
0.5-1
-0.025
-0.15
-0.3
See text for conditions.
15:1 split injection, only 1/15 of injection is on column.
>n
"0.2 yl injected with no split.
328
-------
.
-
From this calculated value, the concentration of an identified compound in a
sample is calculated by rearranging Equation 1 to give
(mWunk) (
(RMR)
The use of RMR for quantitation by GC/MS has been successful in repea-
ted applications to similar research problems.
The RMRs for the compounds were calculated from the numerical integra-
tions of peaks observed in the appropriate MID channel. Typical RMRs listed
in Table 2 and 3 are mean values of three injections of each of three repli-
cate standard mixtures.
The RMRs given here are to be regarded as typical values. Not only
must they be determined for each instrument, but day-to-day variations are
sometimes large enough to require daily calibration.
7.0 Quality Assurance Program
An on-going quality assurance program is required to assure the data
quality. Quality control (QC) procedures determine artifacts, losses, etc.
through a system of blanks and controls. Quality assurance (QA) procedures
monitor the execution of the procedure and check data interpretations and
calculations.
7.1 Quality Control
7.1.1 Field Blanks and Controls
Prior to a field sampling trip, enough blanks and controls are prepared
to equal 10% each (2 minimum) of the anticipated number of field samples.
Blanks consist of 10 ml of water in the same type of sampling container as
is used in the field. Controls consist of 10 ml of plasma or water (urine
control) spiked at 10-15 ng/ml with the compounds listed in Table 4. These
blanks and controls are carried to the field and receive the same handling
as the field samples. Workup and analysis of field blanks and controls is
interspersed with the field samples on a regular basis. This method allows
assessment of sample storage stability.
329
-------
Table 2. RMRs FOR PCBs AND PESTICIDES OF INTEREST
TO THIS PROGRAM3
Compound
2- chlo rob ipheny 1
hexachlorobiphenyl
decachlorobiphenyl
lindane
heptachlor
£,£'-DDE
chlordane (peak 1)
chlordane (peak 2)
Concentration
104 ng/yl
3.8 ng/yl
570 ng/yl
10.4 ng/yl
1156 ng/yl
8. -4 ng/yl
100 ng/yl
100 ng/yl
100 ng/yl
100 ng/yl
100 ng/yl
Ion
188
360
498
181
183
272
246
373
375
373
375
RMR
elutes with solvent
and' was two scans wide-
not determinable
.38+3%
.35 + 10%
.14 + 7%
not determinable
.74+9%
.62 + 12%
. 74 + 6%
.45 + 6%
.71+5%
.65+5%
.051 + 6%
.045 + 13%
Standard is din-pyrene (m/z= 212)
330
-------
Table 3. RMR FACTORS FOR STANDARD PCB SOLUTIONS, SELECTED ION MONITORING MODE
Standard
RMR
m/z 188
2-Chlorobiphenyl
RMR
m/z 358
Hexachlorobiphenyl
RMR
m/z 498
Decachlorobiphenyl
I PCB-STD-20
II PCB-STD-2
0.60
0.257
III PCB-STD-0.2
IV PCB-STD-0.04
0.566
0.840
0.637
0.597
0.705
0.640 +.171
0. 699 +.171
1.020~\
0.692 >0. 763 +.257
0.576J
0.325 + .009
0. 294+ -072
0.320"^
0.528 } 0. 459 +.072
0.528J "
0.341
0.430"^
0 462 >°-456 ± -018
0.431J
0.372"^
0.361 I
0.373 ) 0.361 + .033
0.303 f
0.394J
0.287"^
0.543 ) 0.401 + .142
0.372J
Standard is d1f)-pyrene
212).
-------
Table 4. SEMI-VOLATILE HALOGENATED HYDROCARBONS IN
METHANOL SPIKING SOLUTION
Compound
Compound
ct-BHC
6-BHC
Y-BHC
4,4'-Dichlorobiphenyl
2,4,5-Trichlorobipenyl
Heptachlor
Aldrin
Heptachlor epojcide
Dieldrin
£,£'-DDE
£,£'-DDT
2,2',3,3',6,6'-Hexachlorobiphenyl
trans-Nonachlor
Oxychlordane
HCB
332
-------
Table 5 presents a typical set of blanks and controls for QC on a field
trip where. SO .blood samples are to be collected.
7.1.2 Procedural Blanks and Controls
7.1.2.1 Extraction Blanks
With each set of samples, a procedural blank is run. This consists of
10 ml of prepurged distilled water which is extracted under the same condi-
tions as the samples. These blanks are designed to detect artifacts from
dirty glassware, laboratory atmosphere intrusion, and other sources.
7.1.2.2 GC/MS Procedural Control
At the start of each working day, a mixture of 2,6-dimethylphenol,
2,6-dimethylaniline, and acetophenone (PA mixture) is analyzed to monitor
the capillary J3C column performance. This also serves to check the mass
spectrometer tuning.
Field samples, field controls, field blanks, and procedural blanks are
queued up for GC/MS analysis such that at least one QC sample is run each
working day. In addition, a standard solution is analyzed each day to serve
as a procedural control and also to update the RMR value. Thus, in a typical
working day, 4 field samples, 1 blank or control, and 1 RMR standard are
run.
The Finnigan GC/MS is a quadrupole mass spectrometer which requires
frequent tuning. Daily tuning is achieved using FC-43 and decafluorotriphenyl
phosphine (DFTPP).
7.2 Quality Assurance
Both internal and external QA procedures are to be followed. Internal
QA procedures assure the continuity and consistency of the data. External
QA procedures (interlaboratory checks) verify or dispute the accuracy of the
data being generated in the primary laboratory.
7.2.1 Internal Quality Assurance
7.2.1.1 Supervision and Monitoring of Activities
There are three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This
person must be aware of their actions, observe events which may affect the
data, and maintain appropriate records. At the second level, the chemist's
supervisor monitors their daily activities, reviews the notebook, checks
333
-------
Table 5. BLOOD QC SAMPLES FOR A TYPICAL.SAMPLING TRIP
Sample type
Number
Comments
Field Blank
Field Control
Lab Blank
Lab Control
5
5
Freeze after preparation, carry
to field, store with field sam-
ples.
Store with field blanks
Freeze after preparation, store
in same freezer as field sam-
ples will be stored
Store with lab blanks
334
-------
data and calculations, ^nd~assi54>s--iB..iVtroubi^shoot-iag^'^ppoblems.--'At the- •'-
tertiary level, a QA coordinator interviews all personnel on the project.
The interviews cover the operations they perform (precisely), the data they
obtain, a spot-check of their calculations, and any problems they have had.
7.2.1.2 Documentation
From the initial preparation of a sample container through reporting of
the analytical results, each sample is accompanied by a chain of custody
sheet. Each person signs in the time of receipt, operations performed, and
transmittal of the sample. This record is important for tracing a contaminant,
bad standard, or some other problem.
Sampling Protocol Sheets—When a sample is collected, a sampling protocol
sheet is filled in which contains a discrete sample code which identifies
project number, area, site, locations, trip number, sampling period, and
sample type. Also included are sample times, volumes, addresses, and other
pertinent information. Where appropriate, a map is made to precisely identify
the location.
Sample Log—Upon return from a sampling trip, each sample code is
entered into a sample log book. This log is updated as samples proceed
through workup and analysis. Thus, at a glance, project personnel can tell
the status of each sample and find out how many are at different stages in
the analytical protocol.
GC/MS Log—Each sample run by GC/MS is logged into a notebook, detailing
analysis conditions, where the data are archived, and what hardcopy data has
been produced.
7.2.2., External Quality Assurance
Samples will be analyzed by a designated QA laboratory. Samples,
controls, and blanks will be shipped directly from the field to the QA
laboratory for analysis. They will report the results to the primary labora-
tories for correlation with the primary data.
7.2.2.1 Selection of Samples for QA
Approximately 10% of the field samples (2 minimum) will be collected in
duplicate (i..e. two 10 ml blood samples and two 25 ml urine samples) for
shipment to the QA laboratory. This selection process will be random with
the following restrictions:
335
-------
(1) The donor must consent ot the additional blood and urine collection.
(2) If any stratification of donors is known, purposive selection of
QA donors may be used to get representative samples (e.g. occupa-
tionally exposed vs. "normal" individuals or upwind vs. downwind
residents).
7.2.2.2 QC Samples - -
Along with the QA duplicate samples, at least one QC control and one QC
blank must be included with the QA samples. An example is shown in Table 6
for a trip collecting 50 samples.
7.2.2.3 Sample Codes
All samples shipped to the QA laboratory should be encoded so that the
laboratory cannot discern between samples and controls. This "blind" protocol
will add validity to the results.
7.2.2.4 Shipping
Samples should be shipped on dry ice directly to the QA laboratory by
an appropriate air carrier (e.g. Federal Express, Eastern Sprint) in well
insulated and packed cartons.
8.0 References
1. Pellizzari, E. D., M. D. Erickson, and Zweidinger, R. A., "Analytical
Protocols for Making a Preliminary Assessment of Halogenated Organic
Compounds in Man and Environmental Media", EPA-560/13-79-010, September
1979, Appendix F, p. 151.
2. Thompson, J. F., "Analysis of Pesticide Residues in Human and Environ-
mental Samples, A Compilation of Metods Selected for Use in Pesticide
Monitoring Programs", Environ. Toxicol. Div., Health Effects Research
Lab., USEPA, RTP, NC, December (1974).
3. Hines, J. R., Shaprio, R., Pellizzari, E. and Schwartz, A., HRC and
CC, submitted for publication (1978).
4. Hines, J. R., Shapiro, R., Schwartz, A. and Pellizzari, E., HRC and
CC, submitted for publication (1978).
5. Sherma, J., "Manual of Analytical Quality Control for Pesticides and
Related Compounds", EPA-600/1-76-017, Table 7-1.
Revised, April 1980
336
-------
Table 6. SAMPLES TO BE COLLECTED AND SHIPPED TO QA LABORATORY
FOR A TYPICAL SAMPLING TRIP
Sample
Number
Comment
Duplicate sample
Field blank
Field Control
1
1
Random selection unless
prior information stra-
tified subjects
Ship with samples
Ship with samples
337
-------
ANALYTICAL PROTOCOL: POLYNUCLEAR AROMATIC HYDROCARBONS IN BLOOD
AND URINE (RTI)
1.0 Principle of Method
Polynuclear aromatic hydrocarbons (PAHs) are extracted from blood
plasma or urine with organic solvents, dried and concentrated to an appropri-
ate volume for quantification using gas chromatographic/flame ionization
detection (GC/FID). An alternate method for analysis if high performance
liquid chromatography (HPLC). Blood or urine samples are optionally subjected
to liquid chromatographic cleanup if severe interferences are encountered.
2.0 Range and Limit of Detection
The sensitivity of response to GC/FID is a function of the instrument,
the compound and the sample matrix from which it is extracted. The limit
of detection for this method is dependent on the level of interferences from
sample matrix rather than instrumental limitations. With no interferences
the detection limits of the PAHs of interest are about 1 ng with packed
column gas chromatography. Detection limits for HPLC analysis are given in
Table 1.
3.0 Interferences
Solvents, reagents, glassware, and other sample processing hardware may
yield discrete artifacts and/or elevated baselines causing misinterpretation
of the chromatograms. All of these materials must be demonstrated to be
free from interferences under the conditions of the analysis by running
method blanks. Specific selection of reagents and purification of solvents
by distillation in all-glass systems may be required.
Interferences can be coextracted from the samples. Thus samples may
need a clean-up step to approach the 1 ng detection limit.
Capillary gas chromatographic methods, with inherently greater resolution
than packed column GC, minimizes the extent of interferences.
4.0 Precision and Accuracy
The method has not been validated.
338
-------
Table 1. HIGH PERFORMANCE LIQUID CHROMATOGRAPHY OF PAHsc
c
Compound
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo (a ) anthracene
Chrysene
Benzo (b) f luoranthene
Benzo (k) f luoranthene
Benzo(a)pyrene
Dibenzo(a,h)anthracene
Benzo (ghi)perylene
•'* •»
Indeno (1,2 , 3-cd)pyrene
Retention time
(min)
16.17
18.10
20.14
20.89
22.32
23.78
25.00
25.94
29.26
30.14
32.44
33.91
34.95
37.06
37.82
39.21
Detection limit
(M8/L)b
UV
2.5
5.0
3.0
0.5
0.25
0.10
0.50
0.10
0.20
0.20
1.0
0.30
0.25
1.0
0.75
0.30
Fluorescence
20.0
100.0
4.0
2.0
1.2
1.5
0.05
0.05
0.04
0.5
0.04
0.04
0.04
0.08
0.2
0.1
Source: "Development and Application of Test Procedures for Specific
Organic Toxic Substances in Wastewaters, Category 9-PAHs", Report for
EPA Contract 68-03-2624 (in preparation).
Detection limit is calculated from the minimum detectable HPLC res-
ponse being equal to five times the background noise, assuming an
equivalent of a 2 ml final volume of the 1 liter sample extract, and
assuming an HPLC injection of 2 microliters.
CHPLC conditions: Reverse phase HC-ODS Sil-X 2.6 x 250 mm Perkin-Elmer
column; isocratic elution for 5 min using 40% acetonitrile/60% water,
then linear gradient elution to 100% acetonitrile over 25 minutes,
flow rate is 0.5 ml/min.
339
-------
5.0 Apparatus and Reagents
5.1 Sampling Apparatus
5.1.1 Blood
Blood samples are obtained in 10 ml vacutainer tubes made of borosilicate
glass with silicone stoppers (Venoject-KT-200-SKA) containing heparin as an
anticoagulant.
5.1.2 Urine
Urine samples are obtained in 120 ml wide-mouthed bottles washed,
solvent rinsed once oven dried before use. Screw caps with Teflon liners
must be used on sample bottles.
5.2 Extraction Apparatus
1. Glass culture tubes (16 x 125 mm) and caps equipped with Teflon
liners, reciprocal shaker (ca. 40 oscillations/minute);
2. 500 ml Kuderna-Danish evaporators, receiving tubes and three ball
Snyder columns;
3. glass bottles and caps equipped with Teflon liners;
®
4. reactivials ;
5. centrifuge;
6. 22 mm i.d. chromatography columns.
5.3 Solvents and Reagents
1. Hexane (Burdick and Jackson) distilled in glass, redistilled prior
to use.
2. Methylene chloride (Burdick and Jackson) distilled in glass,
redistilled prior to use.
3. Anhydrous sodium sulfate, (extracted with pentane in Soxhlet for
24 hr and stored in oven at 140°C).
4. Florisil, 60/100 mesh (activated by heating for at least 5 hr at
130°C).
NOTE: For a more complete treatment on the handling and characteristics
of Florisil batches refer to the "EPA Manual of Analytical Methods for the
Analysis of Pesticide Residues in Human and Environmental Samples", Section
3, D.
340
-------
6.0 Procedure
6.1 Collection of Samples
6.1.1 Blood
Blood is obtained from volunteers by a qualified phlebotomist using
vacutainers. Samples are inverted to ensure dispersal of the anticoagulant
and are then centrifuged for 45 min a-t 2500 rpra. Plasma is then pipetted
off, stored in a shell vial with teflon-lined screw caps and frozen as soon
as possible.
6.1.2 Urine
First morning urines are collected by the volunteers in 120 ml wide-
mouthed bottles. The samples are then picked up by a member of the sampling
team, and stored in a refrigerator or on ice until ready for analysis.
6.2 Extraction, Cleanup, and Concentration
6.2.1 Plasma
1. Transfer 2.0 ml of plasma to a 12 x 125 mm culture tube.
2. Pipet in 7.0 ml of hexane and cap.
3. Shake on a shaker bath at 40 oscillations/minute for 3 hr.
4. Remove sample from bath, centrifuge for ~5 minutes to improve
separation of layers, transfer 5.0 ml of organic extract to glass
bottle with Teflon liner.
5. Pipet into serum 5.0 ml additional hexane and repeat (3).
6. Repeat (4), combine extracts and dry for at least 30 min over
0.5 g anhydrous Na-SO,.
7. Transfer to a 500 ml K-D flask (micro K-D optional), concentrate
to 2-4 ml and cool to ambient temperature.
8. Rinse sides of K-D with 1.5 ml hexane and blow under dry nitrogen
to about 1 ml.
®
9. Transfer to a reactivial previously calibrated to a specific
volume and further concentrate.
6.2.2 Urine
1. Transfer 5.0 ml of urine to a 12 x 125 mm culture tube.
2. ' Follows steps 2 through 9 of the procedure for the extraction of
blood plasma for organochlorine pesticides.
341
-------
6.2.3 Optional Florisil Cleanup
If interferences hinder analysTsa5d~qaantitati6n bf"de"sired peaks, a
Florisil column clean-up can be incorporated into the extraction scheme.
6.2.3.1 Florosil Fractionation
1. Prepare a chromatographic column containing 4 inches (after settling)
of activated Florisil topped with 1/2 inch of anhydrous, granular
Na2SO,. • A small wad of glass wool, preextracted with hexane, is
placed at the bottom of the column to retain the Florisil.
NOTE: (1) Florisil is activated by heating for at least 5 hr at 130°.
(2) If the oven is of sufficient size, the columns may be prepacked and
stored in the oven, withdrawing columns a few minutes before use. (3) The
amount of Flor-isil needed for proper elution should be determined for each
lot of Florisil.
2. Place a 500 ml Erlenmeyer flask under the column and prewet the
packing with hexane (40-50 ml, or a sufficient volume to completely
cover the Na_SO, layer).
NOTE: From this point and through the elution process, the solvent
level should never be allowed to go below the top of the Na2SO, layer. If
air is introduced, channeling may occur, making for an inefficient column.
3. Using a 5 ml Mohr or a long disposable pipet, immediately transfer
the tissue extract (ca. 5 ml) from the evaporator tube onto the
column and permit it to percolate through.
4. Rinse tube with two successive 5 ml portions of hexane, carefully
transferring each portion to the column with the pipet.
NOTE: Use of the Mohr or disposable pipet to deliver the extract
directly onto the column precludes the need to rinse down the sides of the
column.
5. Prepare two Kuderna-Danish evaporative assemblies complete with 10
ml graduated evaporative concentrator tubes. Place one glass bead
in each concentrator tube.
6. Replace the 500 ml Erlenmeyer flask under each column with a 500
ml Kuderna-Danish assembly and commence elution with 200 ml of 6%
diethyl ether in hexane (Fraction I). The elution rate should be
5 ml per minute. When the last of the eluting solvent reaches the
342
-------
top of the Na^SO^ layer, place .a second 500 ml Kuderna-Danish
assembly under vth^-C<3luma~4ft
-------
Table 2. GAS CHROMATOGRAPHY OF PAHs
Compound3 Retention timejmin)
Naphthalene 4'5
Acenaphthylene 10'4
Acenaphthene 10'8
Fluorene 12'6
Phenanthrene *•* •"
Anthracene 15.9
Fluorantfiene 19.6
Pyrene 20.6
Benzo(a)anthracene 20.6
Chrysene 24.7
Benzo(b)fluoranthene 28.0
Benzo(k)fluoranthene 28.0
Benzo(a)pyrene 29.4
Dibenzo(a,h)anthracene 36.2
Indeno(l,2,3-cd)pyrene 36.2
Benzo(ghi)perylene 38.6
3GC conditions: Chromosorb W-AW-DMDCS 100/120 mesh coated with 3% 0V-
17, packed in a 6' x 2 mm ID glass column, with nitrogen carrier gas
at 40 mL/min flow rate. Column temperature was held at 100°C for 4
minutes, then programmed at 8°/minute to a final hold at 280°C.
344
-------
estimated retention taaes^that -sbould-'be^chieve'd by this method-. ^Calibrate
the gas chromatographic system daily with a minimum of three injections of
calibration standards.
Prepare calibration standards that contain the compounds of interest,
either singly or mixed together. The standards should be prepared at concen-
trations covering two or more orders of magnitude that will completely
bracket the working range of the chromatographic system.
Inject 2-5 pi of the sample extract using the solvent flush technique.
Smaller (1.0 pi) volumes can be injected if automatic devices are employed.
Record the volume injected to the nearest 0.05 pi, and the resulting peak
size, in area units. If the peak area exceeds the linear range of the
system, dilute the extract and reanalyze. If the peak area measurement is
prevented by the presence of interferences, further cleanup is required.
Table 1 describes the recommended HPLC column material and operating
conditions for the instrument.
6.4 Calculations
Determine the concentration of individual compounds according to the
formula :
(A) (B) (Vt)
Concentration, pg/1 = /-v •> ,y •.
i; s
where :
A = calibration factor for chromatographic system, in nanograms
material per area unit
B = peak size in injection of sample extract, in area units
V. = volume of extract injected (pi)
V* = volume of total extract (pi)
V = volume of water extracted (ml)
s
Report results in micrograms per liter without correction for recovery
data. When duplicate and spiked samples are analyzed, all data obtained
should be reported.
7.0 Quality Assurance Program
An on-going quality assurance program is required to assure the data
quality. Quality control (QC) procedures determine artifacts, losses, etc.
through a system of blanks and controls. Quality assurance (QA) procedures
345
-------
monitor the execution of the procedure and check data interpretations and
calculations.
7.1 Quality Control
7.1.1 Blanks and Controls
Prior to a field sampling trip, enough blanks and controls are prepared
to equal 10% each (2 minimum) of the anticipated number o.f. field samples.
Lab blanks and controls are stored in the laboratory and constitute "storage
control". Field blanks and controls are carried to the field and receive
the same handling as the field samples and constitute "transportation control"
Workup and analysis of blanks and controls is interspersed with the field
samples on a regular basis. This method allows assessment of sample storage
stability.
Table 3 presents a typical set of blanks and controls for QC on a field
trip where 50 blood or urine samples are to be collected.
7.1.1.1 Blood
Blood blanks consist of 10 ml of water in a vacutainer tube. Controls
consist of 10 ml of water spiked with 10-15 ng of the compounds listed in
Table 4.
7.1.1.2 Urine
Urine blanks consist of 120 ml of distilled water. Controls consist of
120 ml of distilled water spiked with 10-15 ng of the compounds listed in
Table 4.
7.1.2 Procedural Blanks and Controls
7.1.2.1 Extraction Blanks
With each set of samples, a procedural blank is run. This consists of
10 ml of prepurged distilled water which is extracted under the same condi-
tions as the samples. These blanks are designed to detect artifacts from
dirty glassware, laboratory atmosphere intrusion, and other sources.
7.2 Quality Assurance
Both internal and external QA procedures are to be followed. Internal
QA procedures assure the continuity and consistency of the data. External
QA procedures (interlaboratory checks) verify or dispute the accuracy of the
data being generated in the primary laboratory.
346
-------
Table 3. BLOOD QC SAMPLES.. EQRiAJIYPICAL -SAMPLING TRIP
Sample type
Number
Comments
Field Blank
Field Control
Lab Blank
Lab Control
5
5
Freeze after preparation, carry
to field, store with field sam-
ples.
Store with field blanks
Freeze after preparation, store
in same freezer as field sam-
ples will be stored
Store with lab blanks
347
-------
Table 4. POLYNUCLEAR AROMATIC HYDROCARBONS IN
METHANOL SPIKING SOLUTION
Compound Compound
Fluoranthene Chrysene
Benzo(a)pyrene Benzo(a)anthracene
Pyrene
348
-------
7.2.1 Internal Quality Assurance
7.2.1.1 Supervision and Monitoring of Activities
There are three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This
person must be aware of their actions, observe events which may affect the
data, and maintain appropriate records. At the second level, the chemist's
supervisor monitors their daily activities, reviews the notebook, checks
data and calculations, and assists in "troubleshooting" problems. At the
tertiary level, a QA coordinator interviews all personnel on the project.
The interviews cover the operations they perform (precisely), the data they
obtain, a spot-check of their calculations, and any problems they have had.
7.2.1.2 Documentation
From the initial preparation of a sample container through reporting of
the analytical results, each sample is accompanied by a chain of custody
sheet. Each person signs in the time of receipt, operations performed, and
transmittal of the sample. This record is important for tracing a contaminant,
bad standard, or some other problem.
Sampling Protocol Sheets—When a sample is collected, a sampling protocol
sheet is filled in which contains a discrete sample code which identifies
project number, area, site, locations, trip number, sampling period, and
sample type. Also included are sample times, volumes, addresses, meteorology,
and other pertinent information. Where appropriate, a map is made to preci-
sely identify the location.
Sample Log—Upon return from a sampling trip, each sample code is
entered into a sample log book. This log is updated as samples proceed
through workup and analysis. Thus, at a glance, project personnel can tell
the status of each sample and find out how many are at different stages in
the analytical protocol.
7.2.2 External Quality Assurance
Samples will be analyzed by a designated QA laboratory. Samples,
controls, and blanks will be shipped directly from the field to the QA
laboratory for analysis. They will report the results to the primary labora-
tories for correlation with the primary data.
349
-------
7.2.2.1 Selection of Samples for QA
Approximately 10% of the field samples (2 minimum) will be collected in
duplicate (i.e. two 10 ml blood samples) and two 25 ml urine samples for
shipment to the QA laboratory. This selection process will be random with
the following restrictions:
(1) The donor must consent to the additional hair collection.
(2) If any stratification of donors is known, purposive selection of
QA donors may be used to get representative samples (e.g. occupa-
tionally exposed vs. "normal" individuals or upwind vs. downwind
residents).
7.2.2.2 QC Samples
Along with the. QA duplicate samples, at least one QC control and one QC
blank must be included with the QA samples. An example is shown in Table 5
for a trip collecting 50 samples.
7.2.2.3 Sample Codes
All samples shipped to the QA laboratory should be encoded so that the
laboratory cannot discern between samples and controls. This "blind" protocol
will add validity to the results.
7.2.2.4 Shipping
Samples should be shipped on dry ice directly to the QA laboratory by
an appropriate air carrier (£.£. Federal Express, Eastern Sprint) in well
insulated and packed cartons.
8.0 References
1. Thompson, J. F., "Analysis of Pesticide Residues in Human and Environ-
mental Samples, A Compilation of Methods Selected for Use in Pesticide
Monitoring Programs", Environ. Toxicol. Div., Health Effects Research
Lab., USEPA, RTP, NC, December (1974).
2. Pellizzari, E. D. , M. D. Erickson, and Zweidinger, R. A., "Analytical
Protocols for Making a Preliminary Assessment of Halogenated Organic
Compounds in Man and Environmental Media", EPA-560/13-79-010, September
1979, Appendix F, p. 151.
3. "Development and Application of Test Procedures for Specific Organic
Toxic Substances in Wastewaters, Category 9-PAHs", Report for EPA
Contract 68-03-2624 (in preparation).
Revised, May, 1980
350
-------
Table 5. SAMPLES TO BE COLLECTED AND SHIPPED TO QA LABORATORY
FOR A TYPICAL "SAMPLING TRIP
Sample Number Comment
Duplicate sample 5 Random selection unless
prior information stra-
tified subjects
Field blank 1 Ship with samples
Field Control 1 Ship with samples
351
-------
ANALYTICAL PROTOCOL: SAMPLING AND ANALYSIS OF PURGEABLE
HALOGENATED ORGANICS IN URINE (RTI)
1.0 Principle of the Method
Volatile compounds are recovered from a urine sample by warming the
sample and purging an inert gas over the warm sample. The vapors are then
trapped on a Tenax cartridge and subsequently can be analyzed by thermal
desorption interfaced to GC/MS.
2.0 Range and Sensitivity
For a typical organic compound approximately 30 ng are required for
mass spectral identification using high resolution glass capillary GC/MS
analysis. Based on a 25 ml urine sample, a limit of detection of about 1.2
Mg/1 (1.2 ppb) may be expected. The dynamic range for a purged sample is
VLO4; however, smaller samples may be purged and the range increased commen-
surately.
3.0 Interferences
Two possible types of interference must be considered: (1) material
present in the sample which physically prevents the effective purge of the
sample, and (2) a material which interferes with the analysis of the purged
sample. In the former case, several techniques have been developed to
handle such problems (e.g., foaming) by diluting and stirring the sample.
The second case is minimized by the use of GC/MS for the analysis since
unique combinations of m/z and retention time can be selected for most
compounds. This permits the analysis of compounds even though chromatographic
resolution is not obtained.
4.0 Precisibn and Accuracy
The purge and trap technique was validated using four C-labeled model
compounds and six "cold" model compounds yielding an average recovery of
82.9% + 20.8% (1).
Based on these data, expected recoveries of purgeable halogenated
organics from urine are about 80% or better. Within the precision require-
ments of this study, these recovery values indicate that the method is
essentially quantitative.
352
-------
5.0 Apparatus and Reagents
5.1 Sampling
Urine samples are collected in 120 ml cleaned and oven-treated glass
bottles and sealed with Teflon-lined caps.
5.2 Purge Apparatus
The apparatus required is shown in Figure 1.
5.3 Tenax Cartridges
Tenax cartridges are prepared and the background checked as described
in Section 6.1.1 of Protocol A-8 (Analysis of Purgeable Organic Compounds in
Water [Master Analytical Scheme]).
5.4 GC/MS/COMP
The volatile halogenated hyrocarbons purged from water are analyzed on
either an 1KB 2091 GC/MS with an 1KB 2031 data system or a Varian MAT CH-7
GC/MS with a Varian 620/i data system. The sample, concentrated on a Tenax
GC cartridge, is thermally desorbed using an inlet manifold system.
The operating conditions for the thermal desorption unit and the
analysis of Tenax GC cartridges are given in Table 7-
5.5 Reagents and Solvents
1. Pentane, Burdick and Jackson distilled in glass, redistilled.
2. Methanol, Burdick and Jackson distilled in glass, redistilled
prior to use.
6.0 Procedure
6.1 Collection of Samples
Each participant is provided with a clean 120 ml (4 oz.) bottle and
asked to collect the first urine sample in the morning. In addition, spot
urine samples are collected from selected participants when tissue, breath
and blood samples are collected. These samples serve both as backups and to
determine individual variability.
6.2 Purge of Volatile Organics
1. Measure a 25 ml aliquot of urine, previously chilled to 4°C,
into the purge flask (Fig. 1).
2. Dilute sample to 50 ml with purged distilled water and add stir
bar.
353
-------
THERMOMETER
-20tol50°c
THERMOMETER ADAPTER
with 0—ring
I 10/18
TENAX CARTRIDGE
HELIUM
'PURGE
HELIUM INLET
TUBE
LIQUID LEVEL
CO ml ROUND BOTTOM FLASK
MAGNETIC STIRRING BAR
Figure 1. Headspace using apparatus for blood, urine, and tissue
samples.
354
-------
3. Assemble apparatus, start -stirring and raise the temperature
to 50°C.
4. Adjust helium flow to 25 ml/min and purge for 90 min.
5. After 90 min disassemble apparatus and transfer Tenax cartridge
to a Kimax culture tube with 2 g calcium sulfate dessicant for 4
hours of drying. „ .
6. Transfer Tenax cartridge to an identical Kimax culture tube
without calcium sulfate, seal in a paint can and store in freezer
until analysis.
6.3 Analysis of Sample Purged on Cartridge
The instrumental conditions for the analysis of halogenated hydro-
carbons of the-sorbent Tenax GC sampling cartridge is shown in Table 1. The
thermal desorption chamber and six-port valve are maintained at 270° and
200°C, respectively. The helium purge gas through the desorption chamber is
adjusted to 15-20 ml/min. The nickel capillary trap at the inlet manifold
is cooled with liquid nitrogen. In a typical thermal desorption cycle a
sampling cartridge is placed in the preheated desorption chamber and helium
gas is channeled through the cartridge to purge the vapors into the liquid
nitrogen cooled nickel capillary trap. After desorption the six-port valve
is rotated and the temperature on the capillary loop is rapidly raised; the
carrier gas then introduces the vapors onto the high resolution GLC column.
The glass capillary column is temperature programmed from 20° to 240°C at
4°/min and held at the upper limit for a minimum of 10 min. After all of
the components have eluted from the capillary column the analytical column
is then cooled to ambient temperature and the next sample is processed.
6.4 Quantitation
All data are acquired in the full scan mode. Quantitation of the
halogenated compounds of interest is accomplished by utilizing selected ion
plots, SIPs, which are plots of the intensity of specific ions (obtained
from full scan data) vs time. Using SIPs of ions characteristic of a given
compound in conjunction with retention times permits quantitation of compo-
nents of overlapping peaks. Two external standards, perfluorobenzene and
perfluorotoluene, were added to each Tenax GC cartridge in known quantities
just prior to analysis. In order to eliminate the need to construct complete
355
-------
Table 1. INSTRUMENTAL OPERATING CONDITIONS
LKB 2091
Varian MAT CH-7
to
Desorption chamber temperature
Desorption chamber He flow
Desorption time
Capillary trap temperature during desorption
Temperature of capillary trap during injection
onto column
Time of He flow through capillary trap
He flow through column [sweep time]
Carrier flow
Capillary column
Column temperature
Scan range
Scan rate
Scan cycle time
Scan mode
Trap current
Filament Current
Accelerating voltage
270
15 ml/min
8.0 min
-196°C
265
10 ml/min
8.0 min
-196°C
-196°C to 250°C - then held at 190°C
12 3/4 min
9.5 min
2.0 ml/min
100 m SE-30 SCOT
30°C for 2 min
then 4°/min to 240°
5-490 dalton
2 sec full scale
2.4 sec
parabolic
4A
50 pA
3.5 kV
12 3/4 min
4 min
1.0 ml/min
70 m SE-30 WCOT
20 -> 240° at 4°min
20 -»• 500 dalton
1 sec/decade
4.5 sec
exponential
300 |JA
2 kV
-------
calibration curves for each compound quantitated,- the method of relative
molar response (RMR) is used. In this method the relationship of the RMR as
the unknown to the RMR of the standard is determined as follows:
A /Moles .
unk
_
unknown/ standard A , /Moles
std' std
= Aunk/8unk/GMWunk
A — Ti — TGMW —
std/8std/GMWstd
where A '= peak response of a selected ion,
g = number of grams present, and
GMW = gram molecular weight
Thus, in the sample analyzed:
(A , )(GMW . )(g _,)
unk unk/V6std
8unknown ~ (A 4.,)(GMW , , , ..
std/v std unk/std
The value of an RMR is determined at least three independent analyses of
standards of accurately known concentration purged using a gas permeation
system (D5). The precision of this method has been determined to be gene-
rally + percent when replicate sampling cartridges are examined.
7.0 Quality Assurance Program
An on-going quality assurance program is required to assure the data
quality. Quality control (QC) procedures determine artifacts, losses, etc.,
through a system of blanks and controls. Quality assurance (QA) procedures
«'* .*
monitor the execution of the procedure and check data interpretations and
calculations.
7.1 Quality Control
7.1.1 Field and Lab Blanks and Controls
Prior to a field sampling trip, enough blanks and controls are prepared
to equal 10% each (2 minimum) of the anticipated number of field samples.
Blanks consist of 100 ml of water in the same type of sampling container as
is used in the field. Controls consist of 100 ml of urine spiked at
357
-------
100-150 ng with chloroform, 1,2-pichloroethane, 1,1,1-Trichloroethane,
Carbon tetrachloride, Trichloroethylene, Tetrachloroethylene, Chlorobenzene,
m-dichlorobenzene, Bromodichloromethane, and vinyl chloride. Field blanks
and controls are carried to the field and receive the same handling as the
field samples. Lab blanks and controls are prepared at the same time as the
field blanks and controls and are stored at -4°C. Workup and analysis of
field blanks and controls is interspersed with the field samples on a
regular basis. This method allows assessment of sample storage stability.
Table 2 presents a typical set of blanks and controls for QC on a field
trip where 50 urine samples are to be collected.
7.1.2 Procedural Blanks and Controls
7.1.2.1 Extraction Blanks
With each set of samples, a procedural blank is run. This consists of
10 ml of purged, distilled water which is extracted under the same conditions
as the samples. These blanks are designed to detect artifacts from dirty
glassware, laboratory atmoshpere intrusion, and other sources.
In addition Tenax cartridge blanks are analyzed to determine cartridge
background.
7.1.2.2 GC/MS Procedural Control
At the start of each working day, a mixture of 2,6-dimethylphenol,
2,6-dimethylaniline, and acetophenone (PA mixture) is analyzed to monitor
the capillary GC column performance. This also serves to check the mass
spectrometer tuning.
Field samples, field controls, field blanks, and procedural blanks are
queued up for GC/MS analysis such that at least one QC sample is run each
working day. In addition, a standard solution is analyzed each day to serve
as a procedural control and also to update the RMR value. Thus, in a typical
working day, 4 field samples, 1 blank or control, and 1 RMR standard are
run.
The Finnigan GC/MS is a quadrupole mass spectrometer which requires
frequent tuning. Daily tuning is achieved using FC-43 and decafluorotri-
phenylphosphine (DFTPP).
358
-------
Table 2. URINE QC SAMPLES FOR A TYPICAL SAMPLING TRIP
Sample Type Number Comments
Field Blank
Field Control
Lab Blank
5
5
Lab Control
Freeze after preparation,
carry to field, store
with field samples
Store with field blanks
Freeze after preparation,
store in same freezer
as field samples will be
stored
Store with lab blanks
359
-------
7.2 Quality Assurance
Both internal and external QA procedures are to be followed. Internal
QA procedures assume the continuity and consistency of the data. External
QA procedure (interlaboratory checks) verify or dispute the accuracy of the
data being generated in the primary labortory.
7.2.1 Internal Quality Assurance
7.2.1.1 Supervision and Monitoring Activities
There are three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This
person must be aware of their actions, observe events which may affect the
data, and maintain appropriate records. At the second level, the chemist's
supervisor monitors-their daily activities, and reviews the notebook, checks
data and calculations, and assists in "troubleshooting" problems. At the
tertiary level, a QA coordinator interviews all personnel on the project.
The interviews cover the operations they perform (precisely), the data they
obtain, a spot-check of their calculations, and any problems they have had.
7.2.1.2 Documentaion
Chain of Custody
From the initial preparation of a sample transfer through reporting of
the analytical results, each sample is accompanied by a chain of custody
sheet. Each person signs in the time of receipt, operations performed, and
transmittal of the sample. This record is important for tracing a contaminant.
bad standard, or some other problem.
Sampling Protocol Sheets
When a sample is collected, a sampling protocol sheet is filled in
which contains a discrete sample code which identifies project number, area,
site, locations, trip number, sampling period, and sample type. Also included
are sample times, volumes, addresses, and other pertinent information.
Where appropriate, a map is made to precisely identify the location.
Sample Log
Upon return from a sampling trip, each sample code is entered into a
sample log book. This log is updated as samples proceed through workup and
analysis. Thus, at a glance, project personnel can tell the status of each
360
-------
sample and find out how many are at different stages in the analytical
protocol.
GC/MS Log
Each sample run by GC/MS is logged into a notebook, detailing analysis
conditions, where the data are archived, and what hardcopy data has been
produced.
7.2.2 External Quality Assurance
Samples will be analyzed by a designated QA laboratory. Samples,
controls, and blanks will be shipped directly from the field to the QA
laboratory for analysis. They will report the results to the primary
laboratory for correlation with the primary data.
7.2.2.1 -Selection of Samples for QA
Approximately 10% of the field samples (2 minimum) will be collected in
duplicate (i..e. , two 100 ml min samples) for shipment to the QA laboratory.
This selection process will be random with the following restrictions:
1. The donor must consent to the additional urine collection.
2. If any stratification of donors is known, purposive
selection of QA donors may be used to get representative
samples (e.g., occupationally exposed v£ "normal" individuals
or upwind vs_ downwind residents).
7.2.2.2 QC Samples
Along with the QA duplicate samples, at least one QC control and one
QC blank must be included with the QA samples. An example is shown in Table
3 for a trip collecting 50 samples.
7.2.2.3> Sample Codes
All samples shipped to the QA laboratory should be encoded so that the
laboratory cannot discern between samples and controls. This "blind"
protocol will add validity to the results.
7.2.2.4 Shipping
Samples should be shipped on dry ice directly to the QA laboratory by
an appropriate air carrier (e.g., Federal Express, Eastern Sprint) in well
insulated and packed cartons.
361
-------
Table 3. SAMPLES TO BE COLLECTED AND SHIPPED TO QA LABORATORY"
FOR A TYPICAL SAMPLING TRIP
Sample Number Comments
Duplicate sample 5 Random selection unless
prior information stratifies
subjects
Field Blank 1 Ship with samples
Field Control 1 Ship with samples
362
-------
8.0 References
1. Pellizzari, E. D., Erickson, M. D., and Zweidinger, R. A. "Analytical
Protocols for Making a Preliminary Assessment of Halogenated Organic
Compounds in Man and Environmental Media". Appendix pg. Revised
April 1979.
2. Pellizzari, E. D., Development of Method for Carcinogenic Vapor
Analysis in Ambient Atmospheres, Publication No. EPA-650/2-74-121,
Contract No. 68-02-1228, 148 pp., July, 1974.
3. Pellizzari, E. D., Development of Analytical Techniques for Measuring
Ambient Atmospheric Carcinogenic Vapors. Publication No. EPA-600/2-76-
076, Contract No. 68-02-1228, 185 pp., November, 1975.
4. Pellizzari, E. D., J. E. Bunch, B. H. Carpenter and E. Sawicki,
Environ. Sci. Tech., 9, 556 (1975).
5. Pellizzari, E. D., B. H. Carpenter, J. E. Bunch and E. Sawicki,
Environ. Sci. Tech., 9, 556 (1975).
6. Pellizzari, E. D., Development of Analytical Techniques for Measuring
Ambient Atmospheric Carcinogenic Vapors. Publication No. EPA-
600/7-77-055, 288 pp., June, 1977.
Revised April 1980
363
-------
ANALYTICAL PROTOCOL: SAMPLING AND ANALYSIS OF
ARSENIC, CADMIUM, AND LEAD IN URINE (RTI)
1.0 Principle of Method
The analysis of arsenic, cadmium, and lead in urine is carried out
using atomic absorption spectrophotometry. Increased sensitivity is achieved
by atomizing the metal in a graphite furnace with continuous deuterium
background correction. Sample workup for arsenic analysis includes an
extraction from the urine matrix and furnace atomization of solutions
containing 1000 ppm nickel.
2.0 Range and Detection Limit
The minimum detection limit (MDL) and range for the metal assays in
urine are shown below.
Metal MDL Max. Cone.
Arsenic 0.10 Hg/100 ml 6.0 |Jg/100 ml
Cadmium 0.01 0.50
Lead 0.20 10.0
Samples containing higher metal concentrations may be analyzed by
suitable dilution with 0.5% nitric acid. Dilution for arsenic determinations
is made with 0.005 M dichromate solution containing 1000 ppm nickel in 1.0%
nitric acid.
3.0 Interferences
No known chemical or spectral interferences exist in the analysis of
arsenic, cadmium or lead in urine. Severe matrix interferences in the
arsenic analysis are minimized by incorporating the toluene extraction step
into the workup procedure.
4.0 Precision and Accuracy
The precision and accuracy associated with these analyses is a function
of sample metal concentration. At the detection limit, the total measurement
error is + 100%. Based on the results of a previous study (1), the metal
analyses are performed with the following precision (relative standard
deviation). The total analysis error (estimated) is also given (2).
364
-------
Metal
Arsenic
Cadmium
Lead
Range
0.5-2.0 H8/100 ml
0.03-0.07
0.5-1.5
Precision
(% RSD)
15
20
15
Accuracy
(%RE)
5-10
5-10
5-10
Estimated
Total Error (%)
35-40
45-50
35-40
5.0 Apparatus and Reagents
A commercially available stock solution containing 1000 ppm metal is
used for the preparation of the calibration standards. The concentrated
nitric acid is reagent grade quality and the deionized water used in this
study will be prefiltered and subjected to the action of an activated carbon
cartridge and two sequential ion exchange units.
The glassware used for sample workup and the preparation of calibration
solutions must be subjected to a nitric acid cleaning protocol.
All volumetric flasks, beakers, and digestion bottles should be soaked
overnight in 20% nitric acid, rinsed with deionized water, soaking for an
additional 15-18 hours in a 5% nitric acid bath, followed by a copious
deionized water rinse. The flasks are completely filled with 0.5% nitric
acid and stored In this manner. Prior to use, each flask is emptied and
rinsed well with deionized water. Pipets are soaked in 5% nitric acid,
rinsed well with deionized water, air-dried, and stored in a clean, dust-free
environment.
All beakers used for urine digestions require additional, pretreatment.
Clean beakers (soaked in 20% and 5% nitric acid) are "predigested" by
heating 10-25 ml of cone, nitric to reflux (with watchglass), cooling, and
discarding 'tne acid. The beakers are rinsed thoroughly with deionized water
and used for a sample digestion within 30 minutes. The beakers are never
allowed to go dry.
Sample cups for the graphite furnace autosampler may be made of poly-
ptyrene or Teflon. The former type requires overnight soaking in 1% nitric
acid and followed by rinsing with deionized water. The latter type may be
soaked overnight in 20% nitric acid, rinsed, and dried in a 105°C oven.
Nickel chloride hexahydrate is used for adjusting the nickel concentra-
tions to 1000 ppm in all solution slated for arsenic analysis.
365
-------
6.0 Procedure
6.1 Collection of Samples
Urine samples are collected in a 4-ounce polyethylene bottle with a
Polyseal cap. To insure sample stability and to improve handling during
sample workup, the specimen is spiked with concentrated nitric acid shortly
after collection (1.0 ml/100 ml urine). The bottle is labeled and all
pertinent information recorded on a protocol sheet.
6.2 Extraction, Cleanup, and Concentration
6.2.1 Cadmium and Lead Analysis
Two ml of urine, acidified at the site, is added to a 50 ml predigested
beaker and treated with 10.0 ml of cone. HNO_ at 85-90°C for one hour. The
watchglass is removed from the beaker and the digest volume reduced to
approximately 1.0 ml. The residue is cooled and transferred to a 10 ml
volumetric flask and diluted to the mark with 0.5% HNOg. The sample solution
is stored in a 1 oz. polypropylene bottle (with screw cap) at ambient
temperatures until ready for analysis.
6.2.2 Arsenic Analysis (3)
A 5.0 ml aliquot of urine is added to 20 ml of low arsenic cone. HC1 in
a clean 4 oz. glass bottle and the mixture allowed to stand at room tempera-
ture for 3-7 days. At the end of this period, 10 ml of 0.5 M SnCl- and 5 ml
of 30% KI is added to the sample digest and the total allowed to stand at
room temperature and for 30 minutes. Forty ml of cone. HC1 and 10.0 ml of
toluene is added to the mixture and the arsenic bodies extracted into the
organic phase. Half of the toluene layer (5.0 ml) is withdrawn and mixed
with 2.0 ml of a 0.005 M dichromate solution containing 1000 ppm nickel in
1.0% HNO . The arsenic compounds are back-extracted into the aqueous phase
and stored in polypropylene bottles until ready for analysis.
6.3 Instrumental
A Perkin-Elmer Model 403 Spectrophotometer, equipped with a HGA-2000
furnace attachment with deuterium background correction is used for this
analysis. An electrodeless discharge lamp is used as the light source and
the furnace atomization response traced on a Perkin-Elmer Model 056 recorder.
An AS-1 Autosampler may be used to increase throughput and/or to improve
peak reproducibility and sensitivity.
366
-------
Arsenic: Wavelength - 193.7 nm
Gas Interrupt (N_~) - Auto
Furnace Cycle Conditions -
Dry: 200°C for 20 sec.
Char: 1200°C for 30 sec.
Atomize: 2500°C for 8 sec.
Injection Volume - 20 pi
Cadmium: Wavelength - 228.8 nm
Gas Interrupt (NO - Auto
Furnace Cycle Conditions -
Dry: 200°C for 20 sec.
Char: 400°C for 30 sec.
Atomize: 1500°C for 8 sec.
Injection Volume - 20 (jl
Lead: Wavelength- 217.0 nm
Gas Interrupt (N») - Auto
Furnace Cycle Conditions -
Dry: 200°C for 20 sec.
Char: 500°C for 30 sec.
Atomize: 2000°C for 8 sec.
Injection Volume - 20 pi
6.4 Qualitative and Quantitative Analysis
6.4.1 Qualitative Analysis
N/A
6.4.2 Quantitative Analysis
The instrument is calibrated with a digested control urine spiked at
four different concentrations, an unspiked control urine, and a reagent
blank.
Calibration Range (spike concentration):
Arsenic - 0.0 to 6.0 M8/100 ml
Cadmium - 0.0 to 0.5
Lead - 0.0 to 10.0
An exponential of the form y = Ae X-M provides the best representation
of the analytical curve. The values of the x,y calibration pairs are
367
-------
entered into a Monroe Calculator Model 1880 programmed to regress the data
to the exponential and to provide values for the constants, A, b, and M.
Sample peak heights are measured manually and expressed in units of
millivolts. The standard additions calibration constants A, b, and M are
entered into the storage banks of a Texas Instrument Calculator Model 57 and
the metal concentration results obtained by keying in peak height data.
Sample peak measurements and concentration result are recorded in a calcula-
tion worksheet.
y = Aebx-M, MS/100 ml
y = y (metal cone, in sample relative to control urine +
y (metal cone, in control urine)
bx
y = (Ae S-M)
S
bx
yc = ~yo = ~
bx bx
y = (Ae S-M) - (Ae °-M)
bx bx
A f S °\
y = A(e -e )
x = reagent blank peak height, mv
y = standard additions metal concentration corresponding to reagent
blank signal (y £ 0) ,
y = -y = metal concentration in control urine, (jg/100 ml,
x = sample peak height, mv
S
y = concentration differential between sample and control urine.
This value may be either negative (y < y) or positive (y < y)
7.0 Quality Assurance Program
An on-going quality assurance program is required to assure the data
quality. Quality control (QC) procedures determine artifacts, losses, etc.,
through a system of blanks and controls. Quality assurance (QA) procedures
monitor the execution of the procedure and check data interpretations and
calculations.
368
-------
7.1 Quality Control
7.1.1 Field Controls
Prior to field sampling, several control urine collections (10% of
anticipated number of field samples) are obtained. Each urine sample is
divided into two portions. One aliquot is placed in a container identical
to that used for field sample, sent to the site, and subjected to the same
handling and storage conditions as field samples. The other aliquot is
stored at RTI in a dust-free environment. On receipt of samples at RTI,
both portions of the control urine collection is worked up and analyzed as a
part of each urine analytical run. Within the precision of the assay, the
difference in calculated metal concentrations of the two control urine
aliquots is a measure of the contamination/loss during field storage, and
transit to RTI.
7.1.2 Internal Quality Control
7.1.2.1 Calibration Standards and Blanks
The instrument is calibrated before each analytical run with four
standard solutions and ,a reagent blank. Evidence of contamination or instru-
ment malfunction is evident at this time. Such problems are resolved before
initiating sample analysis.
7.1.2.2 Conditioning of Graphite Tube
Before each analytical run, the graphite tube is conditioned by injecting
10 to 20 20 pi aliquots of one of the calibration standards. This operation
insure acceptable precision during sample analysis.
7.1.2.3 Duplicate Injections
Reproducibility of peak response is contiuously monitored during
sample analysis. All standard and sample solutions receive two successive
injections into the graphite furnace. Signal agreement between the duplicate
injections is evaluated according to the following criterion:
First Signal % Maximum Permissible Permissible Range of
% of Full Scale Variation (% MPV) Second Signal, % of Full Scale
90 ± 4% 86-94
80 ± 5% 76-84
70 ± 6% 66-74
60 ± 7% 56-64
369
-------
First Signal % Maximum Permissible Permissible Range of
% of Full Scale Variation (% MPV) Second SignalT % of Full Scale
50 ± 8% 46-54
40 ± 10% 36-44
30 ± 13% 26'34
20 ± 20% -16-24
10 ± 30% 7-13
5 ± 60% 2-8
2 ±100% 0-4
If the second injection gives a signal which falls outside the permis-
sible range, a third injection is performed. The peak measurement not in
agreement with the matching pair is discarded.
All calibration and sample calculations are based on the mean of the
duplicate determinations.
7.1.2.4 Standard Checks
Instrument performance is monitored during each analytical run. After
the analysis of every 12-16 samples one of the calibration standards is
reinjected into the furnace. The standard which most closely matches the
sample peak heights is selected as the check solution. A metal concentra-
tion is calculated for the check standrad based on its peak height during
the calibration run. Similar calculations are carried out for each check
response and the observed changes in metal concentration expressed in terms
of standard deviation units (SDU).
__., (Calibration Value - Check Value)100
• ' (Calibration Value) (% RSD)
The analysis is under control when the SDU < 2.0. Standard checks
which indicate a variation in peak response greater than 2.0 SDU are un-
acceptable. In this event, the graphite tube is changed, conditioned, and
the system recalibrated. Quality control charts are graphed to show this
change in instrument performance with time.
7.2 Quality Assurance
Both internal and external QA procedures are to be followed. Internal
QA procedures assume the continuity and consistency of the data. External
370
-------
QA procedures (interlaboratory checks) verify or dispute the accuracy of the
data being generated-in-the-primary labotfa'trory.
7.2.1 Internal Quality Assurance
7.2.1.1 Supervision and Monitoring of Activities
There are three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This
person must be aware of their actions, observe events which may effect the
data, and maintain appropriate records. At the second level, the chemist's
supervisor monitors their daily activities, reviews the notebook, checks
data and calculations, and assists in "troubleshooting" problems. At the
tertiary level, a QA coordinator interviews all personnel on the project.
The interviews, cover the operations they perform (precisely), the data they
obtain, a spot-check of their calculations, and any problems they have had.
7.2.1.2 Documentation
Chain of Custody
From the initial preparation of a sample container through reporting of
the analytical results, each sample is accompanied by a chain of custody
sheet. Each person signs in the time of receipt, operations performed, and
transmittal of the sample. This record is important for tracing a contaminant,
bad standard, or some other problem.
Sampling Protocol Sheets
When a sample is collected, a sampling protocol sheet is filled in
which contains a discrete sample code which identifies project number, area,
site, locations, trip number, sampling period, and sample type. Also included
are sample times, volumes, addresses, meteorology, and other pertinent
.•• .*
information. Where appropriate, a map is made to precisely identify the
location.
Sample Log
Upon return from a sampling trip, each sample code is entered into a
sample log book. This log is updated as samples proceed through workup and
analysis. 'Thus, at a glance, project personnel can tell the status of each
sample and find out how many are at different stages in the analytical
protocol.
371
-------
Instrument Log
Each sample analysis is logged into a notebook, detailing analysis
conditions.
7.2.2 External Quality Assurance
Samples will be analyzed by a designated QA laboratory. Samples,
controls, and blanks will be shipped directly from the freld to the QA
laboratory for analysis. They will report the results to the primary
laboratory for correlation with the primary data.
7.2.2.1 Selection of Samples for QA
Approximately 10% of the field samples (2 minimum) will be collected in
duplicate shipment to the QA laboratory. This selection process will be
random with the following restrictions:
(1) The donor must consent to the additional collection.
(2) If any stratification of donors is known, purposive selection
of QA donors may be used to get representative samples (e.g.,
occupationally exposed vs "normal" individuals or upwind vs downwind
residents).
7.2.2.2 QC Samples
Along with the QA duplicate samples, at least one QC control and one
QC blank must be included with the QA samples. An example is shown in Table
7 for a trip collecting 50 samples.
7.2.2.3 Sample Codes
All samples shipped to the QA laboratory should be encoded so that the
laboratory cannot discern between samples and controls. This "blind"
protocol will add validity to the results.
7.2.2.4 Shipping
Samples should be shipped on dry ice directly to the QA laboratory by
an appropriate air carrier (e.g., Federal Express, Eastern Spring) in well
insulated and packed cartons.
8.0 References
8-1 "Epidemiologic Study Conducted in Populations Living Around Non-Ferrous
Smelters", Final Report for Contract No. 68-02-2442 (in preparation).
8-2 McFarren, E. F-, Lishka, R. J., and Parker, J. H., Criterion for Judging
Acceptability of Analytical Methods, Anal. Chem., 42(3), 358 (1970).
372
-------
8-3 Handy, R. Wl and Natschke^D. -F., ."Analysis -xaf^Arsenic in Whole Blood
(Urine)", Paper Nor 43 presented at the '30th'ACS Southeastern Regional
Meeting at Savannah, GA, Nov. 8-10, 1978.
373
-------
ANALYTICAL PROTOCOL: SAMPLING- AND .ANALYSIS OF- EXTRACTABLEHIALOGENATED
ORGANICS IN HAIR (RTI)-
1.0 Principle of Method
Semi-volatile halogenated hydrocarbons are extracted from hair with
organic solvents, dried, and concentrated to an appropriate volume for
quantification using a gas chromatograph/electron capture detector (GC/ECD).
Identifications are confirmed by GC/ECD using a second column and, when
sufficiently concentrated, by GC/MS/COMP. Hair samples are optionally
subjected to liquid chromatographic cleanup on Florisil if severe interfer-
ences are encountered.
2.0 Range and- Limit of Detection
The sensitivity of response to GC/ECD is a function of the instrument,
the compound, and the matrix from which it is extracted. If an instrumental
sensitivity of 1.0 pg/|Jl is achieved and 15 g of hair can be extracted, a
detection limit of 7 ppb may be realized. This is probably a lower limit.
3.0 Interferences
Interferences in sample analysis and quantification using GC/ECD are
manifested in the electron capturing ability of the given contaminant.
Additional specificity may be obtained by the use of GC/MS (especially the
negative ion chemical ionization mode), which is very sensitive and selective
toward halogenated organics.
4.0 Precision and Accuracy
This method has not been validated.
5.0 Apparatus and Reagents
5.1 Sampling Apparatus
1. Scissors (cleaned with a solvent such as toluene or alcohol).
2. Bottles (120 ml, wide-mouth), foil-lined caps and teflon-liners.
5.2 Extraction Apparatus
1. Soxhlet extractor (Fisher Cat. No. 9-556B this includes condenser,
Soxhlet tube and flask);
2. Variac and heating mantles;
3. Extraction thimbles;
374
-------
4. 500 ml Kuderna-Danish-evaporators, receiving tubes and three ball
Snyder columns;
5. glass bottles and caps equipped with Teflon liners;
®
6. reactivials ;
7. 10 mm i.d. chromatography columns;
8. 125 ml Erlenmeyer flasks.
5.3 Solvents and Reagents
1. Hexane (Burdick and Jackson) distilled in glass, redistilled prior
to use.
2. Toluene (Burdick and Jackson) distilled in glass,
redistilled prior to use.
3. Anhydrous.granular sodium sulfate, (extracted with pentane in
Soxhlet for 24 hr and stored in oven at 140°C).
4. Florisil, 60/100 mesh (activated by heating for at least 5 hr
at 130°C).
NOTE: For a more complete treatment on the handling and charac-
teristics of Florisil batches refer to the "EPA Manual of Analy-
tical Methods for the Analysis of Pesticide Residues in Human and
Environmental Samples", Section 3, D (1).
6.0 Procedure
6.1 Collection of Samples
Hair samples are collected using solvent-cleaned (toluene or alcohol)
scissors and transferred directly to a precleaned glass bottle (120 ml). It
is prefereable to collect long shrands clipped close to the scalp to integrate
over the hair growth period. Second preference is clippings from short-
haired regions such as the nape of the neck to get the most recent growth.
As a final resort, any clippings are useable.
As much hair as possible is desired up to that which will conveniently
fill the sample bottle. A minimum would be about 5 g.
6.2 Extraction, Cleanup, and Concentration
6.2.1 Extraction
1. The hair sample is weighed and up to 15 g placed in a precleaned
Soxhlet extraction thimble. The sample is extracted with toluene
for M.6 hr (overnight).
375
-------
2. The extract is concentrated .in.a Kuderna-Danish jevaporator. ^.-^
this point, an aliquot is gently taken to dryness with a stream
nitrogen and weighed to determine "oil content". The aliquot is
returned to the main sample.
3. Cool the KD to ambient temperature, rinse the sides with 1.5 ml
hexane, and concentrate by micro KD to 1 ml.
4. Transfer extract to a reactivial , previously calibrated to a
desirable volume and make to volume.
5. As an option, proceed with Florisil cleanup.
6.2.2 Optional Florisil Cleanup (semi-micro)
If interferences from unwanted electron capturing materials hinder
analysis and quantitation of desired peaks, a Florisil column cleanup can be
incorporated into the extraction scheme.
1. Prepare a chromatographic column containing 10 cm (after settling)
of activated Florisil topped with 1 cm of anhydrous, granular
Na^SO,. A small wad of glass wool, preextracted with hexane, is
placed at the bottom of the column to retain the Florisil.
NOTE: (1) Florisil is activated by heating for at least 5 hr at 130°.
(2) If the oven is of sufficient size, the columns may be prepacked and
stored in the oven, withdrawing columns a few minutes before use. (3) The
amount of Florisil needed for proper elution should be determined for each
lot of Florisil.
2. Place a 125 ml Erlenmeyer flask under the column and prewet the
packing with hexane (40-50 ml, or a sufficient volume to completely
cover the Na0SO, layer).
-' •• ^ H
NOTE: From this point and through the elution process, the solvent
level should never be allowed to go below the top of the Na«SO, layer. If
air is introduced, channeling may occur, making for an inefficient column.
3. Using a long disposable pipet, immediately transfer the hair
extract (ca. 1 ml) from the evaporator tube onto the column and
permit it to percolate through.
4. Rinse tube with two successive 1 ml portions of hexane, carefully
transferring each portion to the column with the pipet.
376
-------
NOTE; Use of the disposable :pipet to deliver the extract directly onto
the column precludes the Treed"to Tinse" dowia the'sides of the column.
5. Prepare two Kuderna-Danish evaporative assemblies complete with 10
ml graduated evaporative concentrator tubes. Place 2 or 3 boiling
chips in each concentrator tube.
6. Replace the 125 ml Erlenmeyer flask under each .column with a 500
ml Kuderna-Danish assembly and commence elution with 50 ml of 6%
diethyl ether in hexane (Fraction I). The elution rate should be
5 ml per minute. When the last of the eluting solvent reaches the
top of the Na.SO, layer, place a second 500 ml Kuderna-Danish
assembly under the column and continue elution with 50 ml of 15%
diethyl ether in hexane (Fraction II).
7. To the second fraction only, add 1.0 ml of hexane containing 200
nanograms of aldrin, place both Kuderna-Danish evaporator assemblies
in a water bath and concentrate extract until ca. 5 ml remain in
the tube.
8. Remove assemblies from bath and cool to ambient temperature.
9. Disconnect collection tube from Kuderna-Danish flask and carefully
rinse joint with a little hexane.
10. Attach modified micro-Snyder column to collection tubes, place
tubes back in water bath and concentrate extracts to 1 ml. If
preferred, this may be done at room temperature under a stream of
nitrogen.
11. Remove from bath, and cool to ambient temperature. Disconnect
tubes and rinse joints with a little hexane.
NOTE; The extent of dilution or concentration of the extract at this
point is dependent on the pesticide concentration in the substrate being
analyzed and the sensitivity and linear range of the Electron Capture Detector
being used in the analysis.
12. Should it prove necessary to conduct further cleanup on the 15%
fraction, transfer 12 grams MgO-Celite mixture to a chromatographic
column using vacuum to pack. Prewash with ca. 10 ml hexane,
discard prewash and place a Kuderna-Danish receiver under column.
Transfer concentrated Florisil eluate to column using small portions
377
-------
of hexane. Force -sample-and washings- into-.-the JlgO-Celite mixture-
by slight air pressure and elute column with 25 ml hexane. Concen-
trate to a suitable volume and proceed with gas liquid chromato-
graphy.
NOTE: Standard recoveries should be made through column to ensure
quantitative recoveries.
6.3 Instrumental
The detection and quantification of semi-volatile halogenated hydrocar-
bons is made using a Series 4400 Fisher/Victoreen Gas Chromatograph equipped
with a tritium foil electron capture detector. Separation is effected on a
40 m, 0.38 mm i.d., glass SCOT capillary column coated with 1% SE-30 on
0.32% Tullanox- (3,4.). Maximum efficiency is obtained with a flow rate of
2.5 ml/min of nitrogen gas with makeup nitrogen gas adjusted to a total flow
of 25.0 ml/min, column 220°C (isothermal), and detector 285°C.
As a confirmatory column a 190 cm x 0.2 on i.d. 1.5% OV-17/1.95% QF-1
on 80/100 Chromosorb W-HP packing is employed. Efficient responses are
obtained for flow rates of 18 ml/min at identical column and detector tempera-
tures .
Final confirmation of the identity of the components of sufficiently
concentrated extracts (generally greater than 10 ng/pl) can be made using
gas chromatography/mass spectrometry/computer (GC/MS/COMP) (Finnigan 3300).
The GC/MS/COMP systems used are a Finnigan 3300 GC/MS/COMP and an 1KB
2091 GC/MS equipped with an 1KB 2031 data system. Chromatographic conditions
for the Finnigan 3300 are 20 m x 0.38 mm i.d., 1% SE-30 SCOT capillary
operated isothermally at 235°C and a flow rate of 2.0 ml/min helium. Split-
less injection (0.2-0.3 (Jl) is used, with standard electron impact (70 eV)
ionization conditions.
The LKB 2091 is operated using a 18 m 1% SE-30/BaC03 WCOT capillary
column at 240°, isothermal for PCBs and a 40 m x 0.38 mm i.d. 1% SE-30
SCOT capillary column at 230° isothermal for the pesticides. In both cases,
the column flow rate is 2 ml/min with 20 ml/min split off at the injector.
The mass spectrometer is operated under standard electron impact conditions.
NOTE: Similar GC columns may be substituted for those prescribed. GC
conditions may change accordingly.
378
-------
6.4 Qualitative and Quantitative Analysis
6.4.1 Qualitative Analysis
Alternate single injections of extracts and standard solutions is the
routine procedure for processing samples. If the retention time of a given
component of an extract suggests the presence of a standard compound, a
repetitive injection is then made. Tentative identification is made if the
deviation between the two respective means is no greater than three percent.
A similar criterion is then applied to the retention times of both extract
and standard component upon a second confirmatory column. Qualitative
identification of a component is made if both criteria are satisfied.
6.4.2 Quantitative Analysis
A mean linear response range of 5-160 pg/(Jl has been established for
the compounds trifluralin and y-BHC On a 1% SE-30/0.32% Tullanox 40 m, 0.38
mm i.d. SCOT capillary column installed in a Series 4400 Fisher/Victoreen
Gas Chromatograph. Quantification of a given component is made by a compari-
son of the means of recorder trace areas of two extract and two standard
solutions within this linear response range. The precision of the concentra-
tion of a given component is normally less than ten percent of the mean
concentrations and is obtained by propagation of the standard deviations of
the responses of both the extract and standard solutions. The effective
concentration multiplied by the volume of extract results in the total
amount of extracted material.
6.4.3 GC/MS/COMP Confirmation
The chromatography conditions are similar to those used for GC/ECD.
The samples for this study are to be screened by GC/ECD and confirmed (if
sufficiently concentrated) by GC/MS/COMP. Therefore, the retention times of
the two techniques must be similar. GC/ECD must operate isothermally, so
the GC/MS/COMP conditions reflect this restriction.
The Finnigan 3300 and the 1KB 2091 systems may be operated in both the
full scan and selected ion monitoring (SIM) modes. In the full scan mode,
full spectra are collected. Spectra or mass fragmentograms (single ion
plots) may be plotted for interpretation. In the SIM mode, only a small
number (up to 9 for the Finnigan 3300 and up to 16 for the 1KB 2091) of ions
are monitored. Full spectra are not collected. The advantage of this
379
-------
method is that the detector spends more time "looking" at the selected ion
and therefore better (generally 10-50 times) -sensitivity is obtained.
As an option, GC/Negation Ion Chemical ionization mass spectrometry may be
employed. This technique is still under validation, and must be employed
with caution.
To determine the limits of detection, standard solutions of selected
pesticides and PCB isomers have been analyzed on the Finnigan 3300 and 1KB
2091. In the full scan mode, the limit of detection was the amount of
compound required for an interpretable spectrum. In the SIM mode, the limit
of detection was the amount of compound required to yield a peak 2-4 times
the noise level.
The estimated limits of detection for the Finnigan 3300 and 1KB 2091
are presented in Table 1.
Quantitation using GC/MS/COMP is achieved by comparing the computer-
calculated integrated area of the unknown with the integrated response for a
known amount of standard. To compensate for differences in ionization
cross-section, the relative molar response of authentic compounds is obtai-
ned.
The calculation of the relative molar response (RMR) factor allows the
estimation of the levels of sample components without establishing a cali-
bration curve. The RMR is calculated as the integrated peak area of a known
amount of the compound, A° ,, with respect to the integrated peak area of a
known standard, A° , (in this case d-^-pyrene) , according to the equation
,,
(A°unk> (
From this calculated value, the concentration of an identified compound in a
sample is calculated by rearranging Equation 1 to give
(Aunk) (mwunk)
8unk ~ (Agtd) (mw) (RMR)
The use of RMR for quantitation by GC/MS has been successful in repea-
ted applications to similar research problems.
380
-------
Table 1. ESTIMATED LIMITS OP- DETECTION-TOR EXTRACTABLE
HALOGENATED ORGANIC-ANALYSIS3-
LKB 2091b
Full scai
Compound ng/Ml
Y-BHC (lindane) >12<20
heptachlor 12
chlordane ~30
£,£'-DDE 12
2-chlorobiphenyl ~1
hexachlorobiphenyl <1
decachlorobiphenyl 12
i
m/z
181
272
375
246
188
360
498
SIM
ng/Ml
0.10-0.4
0.10-0.4
5
>0.3
0.004
M).oi6
0.42
Finnigan 3300a
Full scat
5-10
10-20
25-50
5-10
^2.5
25-50
150
SIM
m/z
181
272
375
246
188
360
498
ng/Ml
1
1-1.5
5-10
0.5-1
M) . 025
-0.15
M).3
See text for conditions.
15:1 split injection, only 1/15 of injection is on column.
C0.2 pi injected with no split.
381
-------
The RMRs for the compounds were calculated from the numerical integra-
tions of peaks observed in the appropriate MID channel. Typical RMRs listed
in Table 2 and 3 are mean values of three injections of each of three repli-
cate standard mixtures.
The RMRs given here are to be regarded as typical values. Not only
must they be determined for each instrument, but day-to-day variations are
sometimes large enough to require daily calibration.
7.0 Quality Assurance Program
An on-going quality assurance program is required to assure the data
quality. Quality control (QC) procedures determine artifacts, losses, etc.
through a system of blanks and controls. Quality assurance (QA) procedures
monitor the execution of the procedure and check data interpretations and
calculations.
7.1 Quality Control
7.1.1 Field Blanks and Controls
Prior to a field sampling trip, enough blanks and controls are prepared
to equal 10% each (2 minimum) of the anticipated number of field samples.
Blanks consist of an empty sampling container. Controls consist of 15 g of
hair spiked at 100-150 ng with the compounds listed in Table 4. These
blanks and controls are carried to the field and receive the same handling
as the field samples. Workup and analysis of field blanks and controls is
interspersed with the field samples on a regular basis. This method allows
assessment o-f sample storage stability.
Table 5 presents a typical set of blanks and controls for QC on a field
trip where 9 hair samples are to be collected.
7.1.2 Procedural Blanks and Controls
7.1.2.1 Extraction Blanks
With each set of samples, a procedural blank is run. This consists of
a Soxhlet apparatus and thimble which is extracted under the same conditions
as the samples. These blanks are designed to detect artifacts from dirty
glassware, laboratory atmosphere intrusion, and other sources.
7.1.2.2 GC/MS Procedural Control
At the start of each working day, a mixture of 2,6-dimethylphenol,
2,6-dimethylaniline, and acetophenone (PA mixture) is analyzed to monitor
382
-------
Table 2. RMRs FOR PCBs AND PESTICIDES.OF INTEREST
. TO THIS PROGRAM3
Compound
2 - chl o rob ipheny 1
hexachlorobiphenyl
decachlorobiphenyl
lindane
heptachlor
£,£'-DDE
chlordane (peak 1)
chlordane (peak 2)
Concentration
104 ng/pl
3.8 ng/|Jl
570 ng/jJl
10.4 ng/|Jl
1156 ng/Ml
8.4 ng/pl
100 ng/pl
100 ng/|Jl
100 ng/|Jl
100 ng/pl
100 ng/|Jl
Ion
188
360
498
181
183
272
246
373
375
373
375
RMR
elutes with solvent
and was two scans wide-
no t determinable
0.38 + 3%
0.35 + 10%
0.14 + 7%
not determinable
0.74 + 9%
0.62 + 12%
0.74 + 6%
0.45 + 6%
0.71 + 5%
0.65 + 5%
0.051 + 6%
0.045 + 13%
Standard is d,Q-pyrene (m/z = 212)
383
-------
Table 3. RMR FACTORS FOR STANDARD PCB SOLUTIONS, SELECTED ION MONITORING MODE
RMR RMR RMR
m/z 188 m/z 358 m/z 498
Standard 2-Chlorobiphenyl Hexachlorobiphenyl Decachlorobiphenyl
I PCB-STD-20
II PCB-STD-2
0.60
0.620^
0.257
0.29O
0.341
0.430^
0.6*0 ±. 1,1 0; .32S ± .009 S;J» 0.456 ± .01.
0.643 0.32lJ 0.43lJ
III PCB-STD-0.2 0.566*^ 0.366*^ . 0.372*^
0.840 I 0.293 I 0.361 I
0.637 > 0.699 + .171 0.301 > 0.294 + .072 0.373 > 0.361 + .033
0.597 I 0.239 I 0.303 I
w 0.705J 0.273J 0.394J
oo
*^ IV PCB-STD-0.04 1.020"^ 0.320^ 0.287*^
0.692 } 0.763 + .257 0.528 >0.459 + .072 0.543 > 0.401 + .142
0.576 J 0.528 J 0.372 J
Standard is d1n-pyrene (m/z = 212).
— —
-------
Table 4. - - SEMI-VOLATILE^HAtOGENATED -HYDROCARBONS IN
METHANOL SPIKING SOLUTION
Compound Compound
0-BHC Heptachlor epoxide
P-BHC Dieldrin
Y-BHC £,£'-DDE
4,4'-Dichlorobiphenyl £,j>'-DDT
2,4,5-Trichlorobiphenyl 2,2',3,3',6,6'-Hexachlorobiphenyl
Heptachipr trans-Nonachlor
Aldrin Oxychlordane
HCB
385
-------
Table 5. HAIR QC SAMPLES FOR A TYPICAL SAMPLING TRIP
Sample type
Number
Comments
Field Blank
Field Control
Lab Blank
Lab Control
2
2
Freeze after preparation, carry
to field, store with field sam-
ples
Store with field blanks
Freeze after preparation, store
in same freezer as field sam-
ples will be stored
Store with lab blanks
386
-------
the capillary GC column performance. This also serves: to. check the;mass
spectrometer tuning.
Field samples, field controls, field blanks, and procedural blanks are
queued up for GC/MS analysis such that at least one QC sample is run each
working day. In addition, a standard solution is analyzed each day to serve
as a procedural control and also to update the RMR value. Thus, in a typical
working day, 4 field samples, 1 blank or control, and 1 RMR standard are
run.
The Finnigan GC/MS is a quadrupole mass spectrometer which requires
frequent tuning. Daily tuning is achieved using FC-43 and decafluorotriphenyl-
phosphine (DFTPP).
7.2 Quality Assurance
Both internal and external QA procedures are to be followed. Internal
QA procedures assure the continuity and consistency of the data. External
QA procedures (interlaboratory checks) verify or dispute the accuracy of the
data being generated in the primary laboratory.
7.2.1 Internal Quality Assurance
7.2.1.1 Supervision and Monitoring of Activities
There are three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This
person must be aware of their actions, observe events which may affect the
data, and maintain appropriate records. At the second level, the chemist's
supervisor monitors their daily activities, reviews the notebook, checks
data and calculations, and assists in "troubleshooting" problems. At the
tertiary level, a QA coordinator interviews all personnel on the project.
The interviews cover the operations they perform (precisely), the data they
obtain, a spot-check of their calculations, and any problems they have had.
7.2.1.2 Documentation
From the initial preparation of a sample container through reporting of
the analytical results, each sample is accompanied by a chain of custody
sheet. Each person signs in the time of receipt, operations performed, and
transmittal of the sample. This record is important for tracing a contaminant,
bad standard, or some other problem.
387
-------
Sampling Protocol Sheets—When a sample is collected, a sampling protocol
sheet is filled in which contains a discrete sample code which identifies
project number, area, site, locations, trip number, sampling period, and
sample type. Also included are sample times, volumes, address, and other
pertinent information. Where appropriate, a map is made to precisely identify
the location. . .
Sample Log—Upon return from a sampling trip, each sample code is
entered into a sample log book. This log is updated as samples proceed
through workup and analysis. Thus, at a glance, project personnel can tell
the status of each sample and find out how many are at different stages in
the analytical protocol.
GC/MS Log.--Each sample run by GC/MS is logged into a notebook, detailing
analysis conditions, where the data are archived, and what hardcopy data has
been produced.
7.2.2 External Quality Assurance
Samples will be analyzed by a designated QA laboratory. Samples,
controls, and blanks will be shipped directly from the field to the QA
laboratory for analysis. They will report the results to the primary labora-
tories for correlation with the primary data.
7.2.2.1 Selection of Samples for QA
Approximately 10% of the field samples (2 minimum) will be collected in
duplicate for shipment to the QA laboratory. This selection process will be
random with the following restrictions:
(1) The donor must consent to the additional hair collection.
(2) If any stratification of donors is known, purposive selection of
QA donors may be used to get representative samples (e.g. occupa-
tionally exposed vs. "normal" individuals or upwind vs. downwind
residents).
7.2.2.2 QC Samples
Along with the QA duplicate samples, at least one QC control and one QC
blank must be included with the QA samples. An example is shown in Table 6
for a trip collecting 50 samples.
388
-------
Table 6. SAMPLES TO BE COLLECTED AND SHIPPED 10.QA-LABORATORY
FOR A TYPICAL SAMPLING TRIP
Sample Number Comment
Duplicate sample 5 Random selection unless
prior information stra-
tified subjects
Field blank 1 Ship with samples
Field Control 1 Ship with samples
389
-------
7.2.2.3 Sample Codes
All samples shipped to the QA laboratory should be encoded so that the
laboratory cannot discern between samples and controls. This "blind" protocol
will add validity to the results.
7.2.2.4 Shipping
Samples should be shipped on dry ice directly to the QA laboratory by
an appropriate air carrier (e.g. Federal Express, Eastern Sprint) in well
insulated and packed cartons.
8.0 References
1. Thompson, J. F., "Analysis of Pesticide Residues in Human and Environ-
mental Samples, A Compilation of Methods Selected for Use in Pesticide
Monitoring Programs", Environ. Toxicol. Div., Health Effects Research
Lab., USEPA, RTP, NC, June (1977).
Revised, April, 1980
390
-------
ANALYTICAI PROTOCOL: SAMPLING AND ANALYSIS OF POLYNUCLEAR AROMATIC
HYDROCARBONS IN HAIR (RTI)
1.0 Principle of Method
Polynuclear aromatic hydrocarbons (PAHs) are extracted from hair with
an organic solvent, dried, and concentrated to an appropriate volume for
quantification by gas chromatography/flame ionization detector (GC/FID).
Analysis could also be done by high performance liquid chromatography (HPLC).
Hair samples are optionally subjected to liquid chromatographic clean-up on
silica gel if severe interferences are encountered. This method determines
PAHs in "total hair" (including oil, dust, etc.), not "washed hair".
2.0 Range and. Limit of Detection
In the absence of interferences from the sample matrix, detection
limits of about 1 ng/pl are anticipated with GC/FID. A detection limit of 7
ppm may be realized if 15 g of hair is extracted. Detection limits for HPLC
analysis are given in Table 1.
3.0 Interferences
Solvents, reagents, glassware, and other sample processing apparatus may
yield discrete artifacts and/or elevated baselines causing misinterpretation
of the chromatograms. All of these materials must be demonstrated to be
free from interferences under the conditions of the analysis by running
method blanks. Specific selection of reagents and purification of solvents
by distillation in all-glass systems may be required.
Capillary gas chromatographic methods, with inherently greater resolution
than packed column GC, minimize the extent of interferences.
4.0 Precision and Accuracy
This method has not been validated.
5.0 Apparatus and Reagents
5.1 Sampling Apparatus
1. Scissors (cleaned with a solvent such as toluene or alcohol).
2. Bottles (120 ml, wide-mouth), foil-lined caps and teflon-liners.
5.2 Extraction Apparatus
1. Soxhlet extractor (Fisher Cat. No. 9-556B this includes condenser,
Soxhlet tube and flask);
391
-------
Table I. HIGH PERFORMANCE LIQUID CHROMATOGRAPHY OF PAHs'
Compound
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo (a) anthracene
Chrysene
Benzo (b)fluoranthene
Benzo (k) f luoranthene
Benzo(a)pyrene
Dibenzo(a,h)anthracene
Benzo (ghi)perylene
Indeno (1 ,2 , 3-cd)pyrene
Retention time
(min)
16.17
18.10
20.14
20.89
22.32
23.78
25.00
25.94
29.26
30.14
32.44
33.91
34.95
37.06
37.82
39.21
Detection limit
(M8/L)b
UV
2.5-
5.0
3.0
0.5
0.25
0.10
0.50
0.10
0.20
0.20
1.0
0.30
0.25
1.0
0.75
0.30
Fluorescence
20.0
100.0
4.0
2.0
1.2
1.5
0.05
0.05
0.04
0.5
0.04
0.04
0.04
0.08
0.2
0.1
a
Source: "Development and Application of Test Procedures for Specific
Organic Toxic Substances in Wastewaters, Category 9-PAHs", Report for
EPA Contract 68-03-2624 (in preparation).
Detection limit is calculated from the minimum detectable HPLC res-
ponse being equal to five times the background noise, assuming an
equivalent of a 2 ml final volume of the 1 liter sample extract, and
assuming an HPLC injection of 2 microliters.
TffLC conditions: Reverse phase HC-ODS Sil-X 2.6 x 250 mm Perkin-Elmer
column; isocratic elution for 5 min using 40% acetonitrile/60% water,
then linear gradient elution to 100% acetonitrile over 25 minutes,
flow rate is 0.5 ml/min.
392
-------
2. Variac and heating mantles;
3. Extraction thimbles;
4. 500 ml Kuderna-Danish evaporators, receiving tubes and three ball
Snyder columns;
5. Glass bottles and caps equipped with Teflon liners;
gt
6. Reactivials ;
7. 10 mm i.d. chromatography columns;
5.3 Solvents and Reagents
1. Toluene (Burdick and Jackson) distilled in glass.
2. Cyclohexane (Burdick and Jackson) distilled in glass.
3. Pentane (Burdick and Jackson) distilled in glass.
4. Methylene chloride (Burdick and Jackson) distilled in glass.
5. Anhydrous, granular sodium sulfate (purified by heating at 400°C
for 4 hrs in a shallow tray).
6. Silica gel-100/120 mesh desiccant (Davison Chemical grade 923 or
equivalent). Before use, activate for at least 16 hours at 130°C
in a foil covered glass container.
6.0 Procedure
6.1 Collection of Samples
Hair samples are collected using solvent-cleaned (toluene or alcohol)
scissors and transferred directly to a precleaned glass bottle (120 ml). It
is preferable to collect long strands clipped close to the scalp to integrate
over the hair growth period. Second preference is clippings from short-
haired regions such as the nape of the neck to get the most recent growth.
As a final resort, any clippings are useable.
As much hair as possible is desired up to that which will conveniently
fill the sample bottle. A minimum would be about 5 g.
6.2 Extraction, Cleanup, and Concentration
6.2.1 Extraction
1. The hair sample is weighed and up to 15 g placed in a precleaned
Soxhlet extraction thimble. The sample is extracted with toluene
for ^16 hr (overnight).
393
-------
2. The extract is concentrated in a Kuderna-Danish evaporator. At
this point, an aliquot is gently taken to dryness with a stream
nitrogen and weighed to determine "oil content". The aliquot is
returned to the main sample.
3. Cool the KD to ambient temperature, rinse the sides with 1.5 ml
hexane, and concentrate by micro KD to 1 ml.
4. Transfer extract to a reactivial , previously calibrated to a
desirable volume and make to volume.
5. As an option, proceed with silica gel.
6.2.2 Optional Silica Gel Cleanup
Before the silica gel cleanup technique can be utilized, the extract
solvent must be exchanged to cyclohexane. Add a 1-10 ml aliquot of sample
extract (in toluene) and a boiling chip to a clean K-D concentrator tube.
Prewet the micro-Snyder column by adding 0.5 ml toluene to the top. Place
the micro K-D apparatus on a boiling (100°C) water bath so that the concen-
trator tube is partially immersed in the hot water. Adjust the vertical
position of the apparatus and the water temperature as required to complete
concentration in 5-10 minutes. At the proper rate of distillation the balls
of the column will actively chatter but the chambers will not flood. When
the apparent volume of the liquid reaches 0.1 ml, remove K-D apparatus and
allow it to drain for at least 10 minutes while cooling. Remove the micro-
Snyder column and rinse its lower joint into the concentrator tube with a
minimum of cyclohexane. Adjust the extract volume to about 2 ml.
Prepare a slurry of 10 g activated silica gel in methylene chloride and
place this in a 10 mm ID chromatography -column. Gently tap the column to
settle the silica gel and elute the methylene chloride. Add 1-2 cm of
anhydrous sodium sulfate to the top of the silica gel.
Preelute the column with 40 ml pentane. Discard the eluate and just
prior to exposure of the sodium sulfate layer to the air, transfer the 2 ml
cyclohexane sample extract onto the column, using an additional 2 ml of
cyclohexane to complete the transfer. Just prior to exposure of the sodium
sulfate layer to the air, add 25 ml pentane and continue elution of the
column. Discard the pentane eluate.
394
-------
Elute the column with 25 ml-o£._4fl%, methylene .chlo-r.ide/60%-pentane and
collect the eluate in a-500-ml K-& flask—eqiii-pped-vi'th-'a- -10 ml "concentrator
tube. Elution of the column should be at a rate of about 2 ml/min. Concen-
trate the collected fraction to less than 10 ml by K-D techniques as in
6.2.1, using pentane to rinse the walls of the glassware. Proceed with HPLC
or gas chromatographic analysis.
6.3 Calibration
Prepare calibration standards that contain the compounds of interest,
either singly or mixed together. The standards should be prepared at
concentrations covering two or more orders of magnitude that will completely
bracket the working range of the chromatographic system. If the sensitivity
of the detection system can be calculated as 100 |Jg/l in the final extract,
for example, prepare standards at 10 |Jg/l, 50 |Jg/l, 100 |JgA> 500 (Jg/1, etc.
so that injections of 1-5 pi of each calibration standard will define the
linearity of the detector in the working range.
Assemble the necessary gas chromatographic apparatus and establish
operating parameters equivalent to those indicated in Table 1 or 2. By
injecting calibration standards, establish the sensitivity limit of the
detectors and the linear range of the analytical systems for each compound.
6.4 Analysis
To achieve maximum sensitivity with this GC, the extract must be
concentrated to 1.0 ml. Add a clean boiling chip to the methylene chloride
extract in the concentrator tube. Attach a two-ball micro-Snyder column.
Preset the micro-Snyder column by adding about 0.5 ml of methylene chloride
to the top. Place this micro K-D apparatus on a hot water bath (60-65°C)
so that the concentrator tube is partially immersed in the hot water.
Adjust the vertical position of the apparatus and water temperature as
required to complete the concentration in 5 to 10 minutes. At the proper
rate of distillation the balls will actively chatter but the chambers will
not flood. When the apparent volume of liquid reaches 0.5 ml, remove the
K-D apparatus and allow it to drain for at least 10 minutes while cooling.
Remove the micro-Snyder column and rinse its lower joint into the concentrator
tube with a small volume of methylene chloride. Adjust the final volume to
1.0 ml and stopper the concentrator tube.
395
-------
Table 2 describes the recommended gas chromatographic column material
*
and operating conditions for the instrument. Included in this table are
estimated retention times that should be achieved by this method. Calibrate
the gas chromatographic system daily with a minimum of three injections of
calibration standards.
Inject 2-5 pi of the sample extract using the solvent flush technique.
Smaller (1.0 pi) volumes can be injected if automatic devices are employed.
Record the volume injected to the nearest 0.05 pi, and the resulting peak
size, in area units. If the peak area exceeds the linear range of the
system, dilute the extract and reanalyze. If the peak area measurement is
prevented by the presence of interferences, further cleanup is required.
Table 1 describes the recommended HPLC column material and operating
conditions for the instrument.
6.5 Calculations
Determine the concentration of individual compounds according to the
formula:
(A)(B)(Vt)
Concentration, pg/1 = (y . , .
i ^ sj
where:
A = calibration factor for chromatographic system, in nanograms
material per area unit
B = peak size in injection of sample extract, in area units
V. = volume of extract injected (pi)
V = volume of total extract (pi)
V = volume of water extracted (ml)
9
Report results in micrograms per liter without correction for recovery
data. When duplicate and spiked samples are analyzed, all data obtained
should be reported.
7.0 Quality Assurance Program
An on-going quality assurance program is required to assure the data
quality. Quality control (QC) procedures determine artifacts, losses, etc.
through a system of blanks and controls. Quality assurance (QA) procedures
monitor the execution of the procedure and check data interpretations and
calculations.
396
-------
Table 2. , GAS-CHROMATOGRAPHX'OF PAHsa
Compound Retention time (min)
Naphthalene 4.5
Acenaphthylene 10.4
Acenaphthene 10.8
Fluorene 12.6
Phenanthrene 15.9
Anthracene 15.9
Fluoranthene 19.6
Pyrene 20.6
Benzo(a)anthracene 20.6
Chrysene 24.7
Benzo(b)fluoranthene 28.0
Benzo(k)fluoranthene 28.0
Benzo(a)pyrene 29.4
Dibenzo(a,h)anthracene 36.2
Indeno(l,2,3-cd)pyrene 36.2
Benzo(ghi)perylene 38.6
aSource: "Development and Application of Test Procedures for Specific
Organic Toxic Substances in Wastewaters, Category 9-PAHs", Report for
EPA Contract 68-03-2624 (in preparation).
bGC conditions: Chromosorb W-AW-DMDCS 100/120 mesh coated with 3% 0V-
17, packed in a 6' x 2 mm ID glass column, with nitrogen carrier gas
at 40 ml/min flow rate. Column temperature was held at 100°C for 4
minutes, then programmed at 8°/minute to a final hold at 280°C.
397
-------
7.1 Quality Control
7.1.1 Field Blanks and Controls
Prior to a field sampling trip, enough blanks and controls are prepared
to equal 10% each (2 minimum) of the anticipated number of field samples.
Blanks consist of an empty sampling container. Controls consist of 15 g of
washed hair spiked at 100-150 ng with the compounds listed in Table 3.
These blanks and controls are carried to the field and receive the same
handling as the field samples. Workup and analysis of field blanks and
controls is interspersed with the field samples on a regular basis. This
method allows assessment of sample storage stability.
Table 4 presents a typical set of blanks and controls for QC on a field
trip where 9 hair samples are to be collected.
7.1.2 Procedural Blanks and Controls
With each set of samples, a procedural blank is run. This consists of
a Soxhlet apparatus and thimble which is extracted under the same conditions
as the samples. These blanks are designed to detect artifacts from dirty
glassware, laboratory atmosphere intrusion, and other sources.
7.2 Quality Assurance
Both internal and external QA procedures are to be followed. Internal
QA procedures assure the continuity and consistency of the data. External
QA procedures (interlaboratory checks) verify or dispute the accuracy of the
data being generated in the primary laboratory.
7.2.1 Internal Quality Assurance
7.2.1.1 Supervision and Monitoring of Activities
There are three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This
person must be aware of their actions, observe events which may affect the
data, and maintain appropriate records. At the second level, the chemist's
supervisor monitors their daily activities, reviews the notebook, checks
data and calculations, and assists in "troubleshooting" problems. At the
tertiary level, a QA coordinator interviews all personnel on the project.
The interviews cover the operations they perform (precisely), the data they
obtain, a spot-check of their calculations, and any problems they have had.
398
-------
Table 3. POLYNUCLEAR AROMATIC
METHANOL SPIKING SOLUTION
Compound Compound
Fluoranthene Chrysene
Benzo(a)pyrene Benzo(a)anthracene
Pyrene
399
-------
Table 4. HAIR QC SAMPLES FOR A TYPICAL SAMPLING TRIP
Sample type
Number
Comments
Field Blank
Field Control
Lab Blank
Lab Control
2
2
Freeze after preparation, carry
to field, store with field sam-
ples.
Store with field blanks
Freeze after preparation, store
in same freezer as field sam-
ples will be stored
Store with lab blanks
400
-------
7.2.1.2 Documentation
From the initial preparation of a sample container through reporting of
the analytical results, each sample is accompanied by a chain of custody
sheet. Each person signs in the time of receipt, operations performed, and
transmittal of the sample. This record is important for tracing a contaminant,
bad standard, or some other problem.
Sampling Protocol Sheets—When a sample is collected, a sampling protocol
sheet is filled in which contains a discrete sample code which identifies
project number, area, site, locations, trip number, sampling period, and
sample type. Also included are sample times, volumes, address, and other
pertinent information. Where appropriate, a map is made to precisely identify
the location. -
Sample Log—Upon return from a sampling trip, each sample code is
entered into a sample log book. This log is updated as samples proceed
through workup and analysis. Thus, at a glance, project personnel can tell
the status of each sample and find out how many are at different stages in
the analytical protocol.
GC/MS Log—Each sample run by GC/MS is logged into a notebook, detailing
analysis conditions, where the data are archived, and what hardcopy data has
been produced.
7.2.2 External Quality Assurance
Samples will be analyzed by a designated QA laboratory. Samples,
controls, and blanks will be shipped directly from the field to the QA
laboratory for analysis. They will report the results to the primary labora-
tories for.correlation with the primary data.
7.2.2.1 Selection of Samples for QA
Approximately 10% of the field samples (2 minimum) will be collected in
duplicate for shipment to the QA laboratory. This selection process will be
random with the following restrictions:
(1) The donor must consent to the additional hair collection.
(2) If any stratification of donors is known, purposive selection of
QA donors may be used to get representative samples (e.g. occupa-
tionally exposed vs. "normal" individuals or upwind vs. downwind
residents).
401
-------
ANALYTICAL PROTOCOL: SAMPLING AND ANALYSIS OF ARSENIC, CADMIUM,
AND LEAD IN SCALP HAIR (RTI)
1.0 Principle of Method
The analysis of arsenic, cadmium, and lead in scalp hair is carried out
using atomic absorption spectrophotoraetry. Increased sensitivity is achieved
by atomizing the metal in a graphite furnace with continuous deuterium back-
ground correction. Arsenic determinations are performed on solutions contain-
ing 1000 ppm nickel.
2.0 Range and Detection Limit
The minimum detection limit (MDL) and range for the metal assays in
scalp hair are shown below.
Metal MDL (Mg/g) Max. Cone, (pg/g)
Arsenic 0.08 3.0
Cadmium 0.06 3.0
Lead 0.25 30.0
Samples containing higher metal concentration may be analyzed by
suitable dilution with 0.5% nitric acid. Dilution for arsenic determina-
tions is made 0.5% nitric acid containing 1000 ppm.
3.0 Interferences
No known chemical or spectral interferences exist in the analysis of
arsenic, cadmium, or lead in scalp hair.
4.0 Precision and Accuracy
The precision and accuracy associated with these analyses is a function
of sample metal concentration. At the detection limit, the total measurement
error is + 100%. Based on the results of a previous study (1), the metal
analyses are performed with the following precision (relative standard
deviation). The total analysis error (estimated) is also given (2).
Precision Accuracy Estimated
Metal Range (ug/g) (% RSD) (% RE) Total Error (%)
Arsenic 0.1-0.8 15 5-10 35-40
Cadmium 1-3 15 5-10 35-40
Lead 10-30 10 5-10 25-30
404
-------
Table 5. SAMPLES TO BE COLLECTED AND SHIPPED TO QA LABORATORY
FOR A TYPICAL SAMPLING TRIP
Sample Number Comment
Duplicate sample 5 Random selection unless
prior information stra-
tified subjects
Field blank 1 Ship with samples
Field Control 1 Ship with samples
403
-------
6.2 Evaporation, Cleanup, and Concentration (3)
6.2.1 Washing of Hair Samples
The total hair sample is placed on a clean sheet of paper, cut into
four approximately equal sections with a pair of stainless steel scissors
and then placed in an acid-washed 100 ml beaker.
The hair sections are covered with 50 ml of a 1:1 methanol/ether mixture
and held 30 minutes at room temperature with occasional stirring. This
operation is successful in removing fat-soluble surface contaminants.
The supernatant liquid is decanted and the hair cleansed with approxi-
mately 50 ml of a 10% Prell solution. This is carried out by mixing the
contents of the beaker in an ultrasonic bath for 30 minutes. A square
section of panty hose, secured with a rubber bank, is placed over the beaker
and the detergent mixture decanted through the hose. The hair is rinsed
repeatedly with deionized water, until the rinse water is suds-free.
To insure complete removal of all surface contaminants, the Prell
washing and the deionized water rinsing operations are repeated. The
sample is then dried overnight in a 100°C oven. At this point, the hair
material is further sectioned into a 1 to 2 cm pieces and a 250 mg portion
is weighed for digestion and subsequent analysis.
6.2.1 Digestion of Hair Samples
The weighed hair sample is added to a predigested beaker and heated in
10 ml of cone. HNO-. A watchglass is placed on each beaker and the total
heated at 85-90°C for 3 hours.
The watchglass is removed and the solution evaporated to 1-2 ml. The
residue is transferred to a 10 ml volumetric flask and diluted to the mark
with 0.5% HN03.
Instrumental
A Perkin-Elmer Model 403 Spectrophotometer, equipped with a HGA-2000
furnace attachment with deuterium background correction is used for this
analysis. An arsenic electrodeless discharge lamp is used as the light
source and the furnace atomization response traced on a Perkin-Elmer Model
406
-------
5.0 Apparatus and Reagents
A commercially available stock solution containing 1000 ppm metal is
used for the preparation of the calibration standards. The concentrated
nitric acid is reagent grade quality and the deionized water used in this
study is prefiltered and subjected to the action of an activated carbon
cartridge and two sequential ion exchange units.
The glassware used for sample workup and the preparation of the calibra-
tion solutions must be subjected to a nitric acid cleaning protocol.
All volumetric flasks should be soaked overnight in 20% nitric acid,
rinsed with deionized water, soaking for an additional 15-18 hours in a 5%
nitric acid bath, followed by a copious deionized water rinse. The flasks
are completely- filled with 0.5% nitric acid and stored in this manner.
Prior to use, each flask is emptied and rinsed well with deionized water.
Pipets are soaked in 5% nitric acid, rinsed well with deionized water,
air-dried, and stored in a clean, dust-free environment.
All beakers used for hair digestions require additional pretreatment.
Clean beakers (soaked in 20% and 5% nitric acid) are "predigested" by
heating 10-25 ml of cone, nitric to reflux (with watchglass), cooling, and
discarding the acid. The beakers are rinsed thoroughly with deionized water
and used for a sample digestion within 30 minutes. The beakers are never
allowed to go dry.
Sample cups for the graphite furnace autosampler may be made of polysty-
rene or Teflon. The former type requires overnight soaking in 1% nitric
acid and followed by rinsing with deionized water. The latter type may be
soaked overnight in 20% nitric acid, rinsed, and dried in a 105°C oven.
Nickel chloride hexahydrate is used for adjusting the nickel concen-
tration in all samples and standards to 1000 ppm.
6.0 Procedures
6.1 Collection of Samples
Full strand hair samples are collected with a comb and a pair of
thinning shears. The collection is placed in a zip-loc bag, labeled, and
all pertinent information recorded on a protocol sheet.
405
-------
Calibration Range (spike concentration):
Arsenic - 0.0 to 3.0 |Jg/g
Cadmium - 0.0 to 3.0 |Jg/g
Lead - 0.0 to 30.0 pg/g
An exponential of the form y = Ae X-M provides the best representation
of the analytical curve. The values of the x,y calibration pairs are entered
into a Monroe Calculator Model 1800 programmed to regress the data to the
exponential and to provide value for the constants, A, b, and M.
Sample peak heights are measured manually and expressed in units of
millivolts. The standard additions calibration constants, A, b, and M are
entered into the storage banks of a Texas Instrument Calculator Model 57 and
the metal concentration results obtained by keying in peak height data.
Sample peak measurements and concentration results are recorded on a calcula-
tion worksheet.
. bx w ,
y = Ae -M, pg/g
y = y (metal cone, in sample relative to control hair) +
y (metal cone, in control hair)
bx
. y = Ae S-M
s
bx
yc = ~yo = "(Ae °"M)
bx bx
y = (Ae S-M) - (Ae °-M)
' ' bx bx
A e s °\
y = A(e -e )
y = metal concentration in sample |Jg/g
x = reagent blank peak height, mv
y = standard additions metal concentration corresponding to
reagent blank signal (y £ 0),
y = -y = metal concentration in control hair, |Jg/g
x = sample peak height, mv,
y = concentration differential between sample and control hair.
This value may be either negative (y < y ) or positive
(yc < y) c
C 408
-------
056 recorder. An AS-1 Autosampler may be used to increase throughput and/or
*
to improve peak reproducibility and sensitivity.
Arsenic; Wavelength - 193.7 nm
Gas Interrupt (N ) Auto
Furnace Cycle Conditions -
Dry:- 200°C for 30 sec.
Char: 1500°C for 35 sec.
Atomize: 2500°C for 6 sec.
Injection Volume - 20 |Jl
Cadmium: Wavelength - 228.8 nm
Gas Interrupt (N ) - Manual
. Furnace Cycle Conditions -
Dry: 200°C for 20 sec.
Char: 400°C for 20 sec.
Atomize: 1900°C for 6 sec.
Injection Volume - 20 pi
Wavelength - 217.0 nm
Gas Interrupt (N?) - Auto
Lead:
Furnace Cycle Conditions -
Dry: 200°C for 20 sec.
Char: 500°C for 20 sec.
Atomize: 2000°C for 20 sec.
Injection Volume - 20 pi
6.4 Qualitative and Quantitative Analysis
6.4.1 Qualitative Analyses
N/A
6.4.2 Quantitative Analysis
The instrument is calibrated with a digested control of hair material
spiked at four different concentrations, an unspiked control hair, and a
reagent blank.
407
-------
7.1.2.3 Duplicate Injections
Reproducibility of peak response is continuously monitored during
sample analysis. All standard and sample solutions receive two successive
injections into the graphite furnace. Signal agreement between the dupli-
cate injections is evaluated according to the following criterion:
First Signal % Maximum Permissible Permissible Range of
% of Full Signal Variation (% MPV) Second Signal, % of Full Scale
90 ± 4% 86-94
80 ± 5% 76-84
70 ± 6% 66-74
60 ± 7% 56-64
50 ± 8% 46-54
40 ± 10% 36-44
30 ± 13% 26-34
20 ± 20% 16-24
10 ± 30% 7-13
5 ± 60% 2-8
2 ±100% 0-4
If the second injection gives a signal which falls outside the permis-
sible range, a third injection is performed. The peak measurement not in
agreement with the matching pair is discarded.
All calibration and sample calculations are based on the mean of the
duplicate determinations.
7.1.2.4 Standard Checks
Instrument performance is monitored during each analytical run. After
the analysis of every 12-16 samples one of the calibration standards is
reinjected into the furnace. The standard which most closely matches the
sample peak heights is selected as the check solution. A metal concentra-
tion is calculated for the check standard based on its peak height during
the calculation run. Similar calculations are carried out for each check
response and the observed changes in metal concentration expressed in terms
of standard deviation units (SDU).
(Calibration value - Check Value) 100
(Calibration Value)(% RSD)
410
-------
Arsenic calculations.may.be-J)ased. on_a-linear-regression-of-.»the calibra-
tion data.
7.0 Quality Assurance Program
An on-going quality assurance program is required to assure the data
quality. Quality control (QC) procedures determine artifacts, losses, etc.,
through a system of blanks and controls. Quality assurance (QA) procedures
monitor the execution of the procedure and check data interpretations and
calculations.
7.1 Quality Control
7.1.1 Field Controls
Prior to field sampling, several control hair collections (10% of
anticipated number of field samples) are obtained. Each hair sample is cut
into small pieces, mixed well, and divided into two portions. One aliquot
is placed in a container identical to that used for field samples, sent to
the site, and subjected to the same handling and storage conditions as field
samples. The other aliquot is stored at RTI in a dust-free environment. On
receipt of samples at RTI, both portions of the control hair collection are
worked up and analyzed as a part of each hair analytical run. Within the
precision of the assay, the difference in calculated metal concentrations of
the two control hair aliquots is a measure of the contamination/loss during
field storage, and transit to RTI.
7.1.2 Internal Quality Control
7.1.2.1 Calibration Standards and Blanks
The instrument is calibrated before each analytical run with four
standard solutions and a reagent blank. Evidence of contamination or
instrument malfunction is evident at this time. Such problems are resolved
before initiating sample analysis.
7.1.2.2 Conditioning of Graphite Tube
Before each analytical run, the graphite tube is conditioned by injecting
10 to 20 20 |Jl aliquots of one of the calibration standards. This operation
insures acceptable precision during sample analysis.
409
-------
Sample Log
Upon return from a sampling trip, each sample code is entered into a
sample log book. This log is updated as samples proceed through workup and
analysis. Thus, at a glance, project personnel can tell the status of each
sample and find out how many are at different stages in the analytical
protocol,
Instrument Log
Each sample analysis is logged into a notebook, detailing analysis
conditions.
7.2.2 External Quality Assurance
Samples will be analyzed by a designated QA laboratory. Samples,
controls, and blanks will be shipped directly from the field to the QA
laboratory for analysis. They will report the results to the primary
laboratory for correlation with the primary data.
7.2.2.1 Selection of Samples for QA
Approximately 10% of the field samples (2 minimum) will be collected in
duplicate shipment to the QA laboratory. This selection process will be
random with the following restrictions:
(1) The donor must consent to the additional collection.
(2) If any stratification of donors is known, purposive selection
of QA donors may be used to get representative samples (e.g.
occupationally exposed vs "normal" individuals or upwind vs downwind
residents).
7.2.2.2 QC Samples
Along with the QA duplicate samples, at least one QC control and one
QC blank must be included with the QA samples. An example is shown in Table
1 for a trip collecting 50 samples.
7.2.2.3 Sample Codes
All samples shipped to the QA laboratory should be encoded so that the
laboratory cannot discern between samples and controls. This "blind"
protocol will add validity to the results.
412
-------
The analysis is under control when .the SDU < 2.0. Standard checks
which indicate a variation in peak response greater than 2.0 SDU are unaccept-
able. In this event, the graphite tube is changed, conditioned, and the
system recalibrated. Quality control charts are graphed to show this change
in instrument performance with time.
7.2 Quality Assurance
Both internal and external QA procedures are to be followed. Internal
QA procedures assume the continuity and consistency of the data. External
QA procedures (interlaboratory checks) verify or dispute the accuracy of the
data being generated in the primary laboratory.
7.2.1.1 Supervision and Monitoring of Activities
There are. three levels of quality assurance (QA). The primary quality
assurance is the person conducting the sampling and/or analysis. This
person must be aware of their actions, observe events which may effect the
data, and maintain appropriate records. At the second level, the chemist's
supervisor monitors their daily activities, reviews the notebook, checks
data and calculations, and assists in "troubleshooting" problems. At the
tertiary levels, a QA coordinator interviews all personnel on the project.
The interviews cover the operations they perform (precisely), the data they
obtain, a spot-check of their calculations, and any problems they have had.
7.2.1.2 Documentation
Chain of Custody
From the initial preparation of a sample container through reporting of
the analytical results, each sample is accompanied by a chain of custody
sheet. Each person signs in the time of receipt, operations performed, and
transmittal of the s'ample. This record is important for tracing a contaminant,
bad standard, or some other problem.
Sampling Protocol Sheets
When a sample is collected, a sampling protocol sheet is filled in
which contains a discrete sample code which identifies project number, area,
site, locations, trip number, sampling period, and sample type. Also included
are sample times, volumes, addresses, meteorology, and other pertinent
information. Where appropriate, a map is made to precisely identify the
location.
411
-------
Table 1. SAMPLES TO BE COLLECTED AND SHIPPED TO QA LABORATORY
FOR A TYPICAL SAMPLING TRIP
Sample Number Comment
Duplicate sample 5 Random selection unless
prior information strati-
fies subjects
Field blank 1 Ship with samples
Field control 1 Ship with samples
414
------- |