FINAL REPRT
gef Sound Estuary Progrstn
SAMPLING
'AND • ANALYZING SUBTIDllSBfNTHlC;
MACROMYERTEBR^&E
IN PUGE»SOUNB
Prepared bys
TETRA
Prepared for:
U.S.
Region 10^-
Seattle, WA
Sound
January, 1987
11820
Bellevue, WA
-------
CONTENTS
Page
LIST OF FIGURES iii
LIST OF TABLES iv
ACKNOWLEDGEMENTS v
INTRODUCTION 1
STUDY DESIGN CONSIDERATIONS 3
KIND OF SAMPLER 3
AREA OF SAMPLER 6
SAMPLE REPLICATION 6
SIEVE MESH SIZE 6
SIEVING LOCATION 7
USE OF RELAXANTS 8
USE OF STAINS 8
LEVEL OF TAXONOMY 9
SAMPLING SEASON 9
PROTOCOLS FOR SAMPLING AND ANALYSIS 11
FIELD PROCEDURES 11
Pre-Collection Preparation 11
Collection 14
Processing 19
LABORATORY PROCEDURES 22
Equipment and Supplies 22
Preservative Preparation 22
Analytical Procedures 23
QA/QC PROCEDURES 27
Calibration and Preventive Maintenance 27
Quality Control Checks 28
Corrective Action 29
DATA QUALITY AND REPORTING REQUIREMENTS 29
REFERENCES 30
ii
-------
FIGURES
Number Page
1 Construction of a sieve box 12
2 Deployment of a grab sampler 16
3 Examples of acceptable and unacceptable grab samples 18
4 Example of a sieving stand 20
-------
TABLES
Number Page
1 Contributors to the benthos protocols 2
2 Summary of the major design characteristics used most
frequently in historical surveys of subtidal benthic
macroinvertebrate assemblages in Puget Sound 4
ACKNOWLEDGEMENTS
This chapter was prepared by Tetra Tech, Inc., under the direction
of Dr. Scott Becker, for the U.S. Environmental Protection Agency in partial
fulfillment of Contract No. 68-03-1977, Dr. Thomas Ginn of Tetra Tech was
the Program Manager. Mr. John Underwood and Dr. John Armstrong of U.S.
EPA were the Project Officers. Much of this chapter was modified from
material prepared originally by Tetra Tech, Inc. for the Marine Operations
Division, Office of Marine and Estuarine Protection, U.S. EPA, Washington, DC
as part of U.S. EPA Contract No. 68-01-6938, Allison J. Duryee, Project
Director. The primary authors of this chapter were Drs. Gordon Bilyard
and Scott Becker of Tetra Tech, Inc., Mr. Peter Striplin of Evans Hamilton,
Inc., and Mr. Jack Word of Battelle Northwest.
IV
-------
Benthic Infauna
Introduction
January 1987
INTRODUCTION
Recommended methods for sampling and analyzing subtidal soft-bottom
benthic macroinvertebrate assemblages in Puget Sound are presented in this
chapter. The methods are based on the results of a workshop and written
reviews by representatives from most organizations that fund or conduct
environmental studies in Puget Sound (Table 1). The purpose of developing
these recommended protocols is to encourage all Puget Sound investigators
conducting monitoring programs, baseline surveys, and intensive investigations
to use standardized methods whenever possible. If this goal is achieved,
most data collected in the Sound should be directly comparable, and thereby
capable of being integrated into a sound-wide database. Such a database
is necessary for developing and maintaining a comprehensive water quality
management program for Puget Sound.
Before the recommended protocols are described, a section is presented
on study design considerations. This section discusses some major elements
of the design of subtidal benthic macroinvertebrate studies that were considered
at the workshop but left unresolved. Following this initial section, specifi-
cations are provided for the field, laboratory, quality assurance/quality
control (QA/QC), and data reporting procedures that are recommended for
most future benthic macroinvertebrate studies in Puget Sound.
Although the following protocols are recommended for most studies
conducted in Puget Sound, departures from these methods may be necessary
to meet the special requirements of individual projects. If such departures
are made, however, the funding agency or investigator should be aware that
the resulting data may not be comparable with most other data of that kind.
In some instances, data collected using different methods may be compared
if the methods are intercalibrated adequately.
-------
TABLE 1. CONTRIBUTORS TO THE BENTHOS PROTOCOLS
Name
Organization
Rick Albright3
John Armstrong3
Scott Beckera,b
Alice Benedict3
Gordon Bilyard3
Steven Brocco3
Peter Chapman3
Faith Cole3
Wally De Ben3
Tom Ginn3
Evan Hornig3
David Kendall3
Ed Long3
Michael Matta3
Gary Mauseth3
Brian Melzian
Jeff Osborn3
John Palmisano3
Walter Pearson3
Deborah Penny3
Tony Roth3
Liko Self
Kathy Sercu3
Craig Smith
Margaret Stinson3
Rick Swartz3
David Terpening3
Ron Thorn3
Jeff Ward3
Bert Webber3
Don Weston3
Jack Word3
U.S. EPA
U.S. EPA
Tetra Tech
NOAA
Tetra Tech
Nortec
EVS Consultants
U.S. EPA
U.S. EPA
Tetra Tech
U.S. EPA
U.S. COE
NOAA
U.S. EPA
Nortec
U.S. EPA
Parametrix
CH2M HILL
Battelle Northwest
University of Washington
Cooper Consultants
University of Washington
U.S. EPA
University of Washington
Washington Dept. Ecology
U.S. EPA
U.S. EPA
University of Washington
URS Engineers
Western Washington University
SAIC
Battelle Northwest
3 Attended the workshop held on December 12, 1985.
b Workshop moderator.
-------
Benthic Infauna
Study Design Considerations
January 1987
STUDY DESIGN CONSIDERATIONS
The designs of different benthic macroinvertebrate studies can vary
substantially, depending upon study-specific objectives. Therefore, it
is not possible to standardize all of the elements that constitute such
a study design. Because variations in some of these elements can influence
the comparability of different data sets, it is preferable that as many
of these elements as possible be similar among studies.
Nine study design elements that may vary among different studies in
Puget Sound and may. limit data comparability are described in this section.
They include:
• Kind of sampler
• Area of sampler
• Sample replication
• Sieve mesh size
• Sieving location
t Use of relaxants
• Use of stains
• Level of taxonomy
• Sampling season.
The specifications for these nine elements that are used most frequently
in surveys of subtidal benthic macroinvertebrate assemblages in Puget Sound
are summarized in Table 2.
KIND OF SAMPLER
A wide variety of devices can be used to sample benthic macroinvertebrates,
including trawls, dredges, grabs, box corers, suction samplers, and hand-
held corers (Eleftheriou and Holme 1984). Because most of these devices
sample the benthos in a unique manner, comparability of data collected
using different devices may be questionable. Trawls and dredges generally
collect organisms over a variable and relatively large area. By contrast,
the remaining devices generally collect organisms over a fixed and relatively
small area. Data collected using the former devices are semi-quantitative
at best, and detailed comparisons with data collected using the latter,
-------
TABLE 2. SUMMARY OF THE MAJOR STUDY DESIGN CHARACTERISTICS USED
MOST FREQUENTLY IN HISTORICAL SURVEYS OF SUBTIDAL BENTHIC
MACROINVERTEBRATE ASSEMBLAGES IN PUGET SOUND
Study Design Variable
Most Common Specification
Kind of sampler
Area of sampler
Sample replication3
Sieve mesh size
Initial sieving location
Use of relaxants
Use of stains
Level of taxonomy
Sampling season
Modified van Veen bottom grab
0.1 m2
4-5 per station
1.0 mm
On vessel
No
Yes - rose bengal
Species, if possible
Variable
3 For variance-related comparisons.
-------
Benthic Infauna
Study Design Considerations
January 1987
more quantitative, devices generally are questionable. Differences among
data collected using the latter devices generally are more subtle.
The most common device used to sample subtidal soft-bottom benthic
macroinvertebrates in Puget Sound is the modified van Veen bottom grab
(Kahlsico 1986). Penetration depth (i.e., the maximum depth sampled below
the sediment surface) can be as great as 15-16 cm when using this device.
The major advantages of this device are its ease of deployment from small
vessels, its reliable operation in a wide range of sediment types (from
clays through sands), and its frequent use in Puget Sound in the past (affording
a large database for comparison). Its principal disadvantages are that
its penetration depth varies from sample to sample with sediment properties,
that it can land at an angle (providing varying penetration depth within
the same sample), and that the sample inevitably is folded by the closing
motion and geometry of the device (with resulting loss of information on
vertical structure within the sediments).
Most of the disadvantages identified for the van Veen grab are shared
by all grabs. The Smith-Mclntyre grab's characteristics differ only slightly
from those of the van Veen. It is spring loaded and encased in a frame
that ensures vertical entry of the grab into the sediments. This combination
of features slightly reduces variability in penetration, both within and
between samples. Its major disadvantages relative to the van Veen grab
are slightly greater difficulty in handling and general lack of intercali brat ion
studies with the more widely used (in Puget Sound) van Veen. No other
grabs have been used commonly in the Sound.
Box corers (Messier and Jumars, 1974; Eleftheriou and Holme, 1984)
have a surrounding frame that ensures vertical entry. Although most have
stops and weighting systems that allow depth of penetration to be set,
most workers adjust the devices for maximum penetration (roughly 45 cm
in the most common models) and then slice the resulting core to a standard
depth (e.g., 10 cm) for sieving. Thus, imprecision due to variable penetration
depth is much reduced in comparison to grab samples. Using box corers,
in situ horizontal partitioning of samples for gaining further spatial
information or for unbiased subsampling is routine. Box corers are widely
recognized as the tools of choice for maximal accuracy and precision of
sampling in soft sediments below diving depths. Their disadvantages are
large size and weight, requiring a large vessel for deployment and large
expense for construction. In addition, their relatively recent introduction
and lack of intercalibration studies with the van Veen grab make comparability
with historical data in Puget Sound an issue.
Suction samplers and hand-held corers avoid some of the problems identified
for grabs and box corers by being operated in situ using SCUBA. Suction
corers can penetrate sediments as deeply as box corers, but they can draw
animals (vacuum-cleaner-like) from surrounding sediments, inflating abundance
estimates. Some suction methods are extremely rough on organisms, turbulently
abrading them with drawn-in sediments. Hand-held corers, on the other
-------
Benthic Infauna
Study Design Considerations
January 1987
hand, are limited in penetration depth. Both kinds of devices are restricted
to SCUBA depths and thus are not of general utility in Puget Sound.
AREA OF SAMPLER
Because different species of benthic macro invertebrates may have different
scales of horizontal spatial distribution (Elliott 1971), data comparability
generally is enhanced if sampling devices sample the same area of sediment
surface. The major reason that trawls and dredges are considered semi-
quantitative devices is that they do not sample consistently the same area
of sediment surface. Although most grabs and corers sample sediment surface
area relatively consistently, comparisons among samples with different
surface areas may be questionable. At present, it is uncertain how such
comparisons would be affected.
The most common sediment surface area sampled by the quantitative
bottom devices used historically in Puget Sound is 0.1 m2 (van Veen grab,
Smith-Mclntyre qrab). Other surface areas sampled using these devices
include 0.06 m? (van Veen grab, box corer), 0.002 m2 (hand-held corer),
and 0.001 m2 (hand-held corer).
SAMPLE REPLICATION
Because the appropriate level of sample replication is determined
largely by study objectives, it cannot be standardized for all studies
in Puget Sound. Given the potentially large within-station variability
of benthic macroinvertebrate assemblages, it generally is advisable to
use more than one sample to represent a station. However, single samples
may be acceptable for some kinds of investigations, such as preliminary
surveys. For statistical comparisons that rely on within-station variance
of benthic infaunal variables, Swartz (1978) recommends that five replicates
be collected at each station, if possible, and that the minimum number
of replicates per station be three. Most historical studies in Puget Sound
that have used variance-related statistical analyses have collected four
to five replicate samples per station.
SIEVE MESH SIZE
Perhaps more than any of the other elements discussed in this section,
the mesh size with which benthic infauna are sieved can limit data comparability
among studies (e.g., Reish 1959; Lewis and Stoner 1981; Schwinghamer 1981;
Rees 1984). In some cases, study objectives may require that a specific
mesh size be used. For example, studies of infaunal recruitment or predation
patterns of juvenile fishes generally require very small mesh sizes (i.e.,
0.3 mm or smaller). However, in other cases (e.g., general characterization
of benthic infaunal assemblages for impact assessment or monitoring), the
study objectives do not narrowly constrain the choice of mesh size. Data
comparability among such studies can be ensured by using a common mesh
size, whenever possible.
-------
Benthic Infauna
Study Design Considerations
January 1987
The mesh size used most frequently to characterize benthic macroinverte-
brate assemblages in Puget Sound is 1.0 mm. A mesh size of 0.5 mm has
also been used in a small number of Puget Sound investigations and is commonly
used in studies of benthic macroinvertebrates on the east coast of the
U.S. Eleftheriou and Holme (1984) recommend that a mesh size of 0.5 mm
be used for most macroinvertebrate studies. A major advantage to using
a 0.5-mm mesh size rather than a 1.0-mm mesh size is increased retention
of total macroi nvertebrates (e.g., by a factor of 130-180 percent; Lewis
and Stoner 1981), including adults of smaller species and juveniles of
larger species (see also Rees 1984). A major disadvantage is increased
cost (e.g., by as much as 200 percent) of sample processing (i.e., primarily
sorting and taxonomic identifications).
For future characterizations of benthic macroinvertebrate assemblages
in Puget Sound, it is recommended that either a 1.0- or 0.5-mm mesh size
be used to sieve samples. If a 0.5-mm mesh size is used, it is recommended
that each sample first be screened using a 1.0-mm mesh size and that the
two fractions (i.e., 0.5 and 1.0 mm) be processed separately. In this
manner, the 1.0-mm results can be compared with data based on a 1.0-mm
mesh size from other studies. Data from the two fractions also can be
pooled during data analysis to represent the full fraction of organisms
>0.5 mm in size.
SIEVING LOCATION
Sieving can be conducted either aboard the survey vessel as samples
are collected or onshore after a sampling excursion has been completed.
In the first case, sieving usually precedes fixation and is conducted primarily
on live organisms. This is the method used by most studies in Puget Sound.
In the second case, sieving generally occurs after fixation and is therefore
conducted on dead organisms. Comparability between the results of these
two techniques may be influenced by at least two factors. First, because
fixation may cause some taxa to distort their shape or autotomize (i.e.,
cast off body parts), the sieving characteristics of those taxa may change
following fixation. Second, sieving characteristics of live organisms
may differ from those of dead individuals. This bias occurs primarily
for soft-bodied organisms (e.g., polychaetes) that can crawl through mesh
openings or entangle themselves on the screen when they are sieved live.
A major problem that may be encountered when organisms are fixed in
sediment before being sieved is that the fixative either will not reach all
buried organisms or will not reach them in time or in sufficient concentration
to prevent some deterioration. Because deteriorated individuals may decompose
completely or fragment upon sieving, their sieving characteristics can be
modified substantially by inadequate fixation. Therefore, if samples are
fixed in sediment, extra care should be taken to ensure that organisms
are fixed adequately. For example, the sample container can be rotated
-------
Benthic Infauna
Study Design Considerations
January 1987
gently immediately after fixation and again after 12-24 h to ensure adequate
fixative penetration.
From a logistical standpoint, sieving of samples in the field is generally
preferred for surveys in which a large number of samples are collected
during each cruise. Field sieving results in a considerable reduction
in the volume of material that must be stored on the vessel (i.e., where
space is often limiting) and later transported to the laboratory. Most
historical large-scale studies in Puget Sound have sieved samples in the
field.
USE OF RELAXANTS
Relaxants are often used when processing benthic macroinvertebrate
samples for at least two major reasons. First, relaxants facilitate, taxonomic
identifications (and morphometric measurements) by reducing the tendency
of organisms to distort their shape or autotomize when exposed to a fixative
(Gosner 1971). Complete organisms having a natural appearance are easier
to identify correctly than are fragmented and/or distorted specimens.
For some taxonomic groups (e.g., Maldanidae), complete organisms are required
for species-level identification.
A second reason for using a relaxant is to ensure that animals are
sieved whole, if sieving follows fixation. The tendency for some taxa
(especially polychaetes) to autotomize if not relaxed can influence sieving
by reducing the size of individuals.
Because relaxation can influence taxonomic identification and sieving,
data comparability between studies that use a relaxant and those that do
not use one may be affected. The magnitude of these effects is unknown,
but probably is greatest for soft-bodied taxa that are difficult to identify
(e.g., some polychaetes) and smallest for taxa encased in a hard enclosure
such as a calcareous shell (e.g., most molluscs) or an exoskeleton (e.g.,
crustaceans), particularly if the hard parts are the primary taxonomic
characters used for identification. To date, most studies in Puget Sound
have not used a relaxant prior to sieving and fixation.
USE OF STAINS
A vital stain (primarily rose bengal) is often added to samples to
facilitate sorting. The stain colors most infauna and thereby enhances
their contrast with the debris from which they are sorted. Taxa that do
not always stain adequately include ostracods and gastropods.
Some taxonomists have found that staining may interfere with the identifi-
cation of certain taxa, and therefore discourage its use. Although it
generally is agreed that staining aids the sorting process (particularly
for inexperienced sorters), a proper quality control program should ensure
8
-------
Benthic Infauna
Study Design Considerations
January 1987
that sorting efficiency is adequate whether or not staining is used. Most
past studies in Puget Sound have used rose bengal stain to facilitate sorting.
LEVEL"OF TAXONOMY
Depending on the objectives of different studies, taxonomic identifications
can range from the phylum to the species level. Identifications to higher
taxonomic levels can provide gross characterizations of benthic infaunal
assemblages, but sacrifice the potential wealth of information available
using species-level identifications (e.g., species composition, species
indicative of impacted or reference conditions, species diversity and evenness,
species replacements, interspecific interactions). The primary drawback
to identifying organisms to the species level is cost, which can be 200-300
percent greater than identifications to the two highest taxonomic levels
(i.e., phylum and class).
Although data based on different taxonomic levels generally cannot
be compared directly, data based on lower taxonomic levels can be pooled
upward (e.g., species to genus, genus to family) for comparisons with higher
level taxa. Data based on highei—level taxa can be compared with lower-
level taxa only if additional taxonomic identifications are made to lower
the level of taxonomy of the former data set. Because future comparisons
may make it desirable to lower the taxonomic level of a data set, it is
strongly recommended that all samples identified only to higher taxonomic
levels be properly archived (indefinitely if possible). Most historical
studies in Puget Sound have identified organisms to either the species
level or the lowest taxonomic level possible (i.e., based on the physical
condition of specimens).
SAMPLING SEASON
Benthic assemblages are constantly changing over time. Probably the
most common temporal patterns observed in benthic assemblages are those
associated with seasonal changes (Gray 1981). Seasonal variation in benthic
assemblages can result from changes in physical or chemical environmental
variables such as temperature, light, salinity, dissolved oxygen, and habitat
disturbance. In general, the influence of these kinds of variables is
greatest in shallow water (Gray 1981). Seasonal variation can also result
from changes in biological variables (e.g., competition, predation, recruit-
ment).
The season in which benthic assemblages are sampled depends largely
on study objectives. Past studies in Puget Sound have sampled benthic
assemblages during a variety of time periods. Although seasonal variations
of benthic macroinvertebrate assemblages are not well characterized for
Puget Sound, information presented by Lie (1968) suggests that both numbers
of individuals per sample and variability among stations is lowest during
the late winter and highest during the late summer. This pattern may reflect
the recruitment cycles of many, but not necessarily all, species. For
-------
Benthic Infauna
Study Design Considerations
January 1987
characterizing adult populations of benthic macroinvertebrates it generally
is preferable to sample when population estimates are least variable.
Data collected by Lie (1968) suggest that late winter may be the most appro-
priate time to sample adult populations of benthic macroinvertebrates in
Puget Sound.
Given the seasonal variation characteristic of benthic assemblages
in general, it is recommended that direct comparisons between samples collected
during different seasons be made with appropriate caution, or avoided com-
pletely. Therefore, studies investigating interannual variation in the
characteristics of benthic assemblages should be conducted during the same
season (preferably the same month) each year.
10
-------
Benthic Infauna
Protocols for Sampling and Analysis
January 1987
PROTOCOLS FOR SAMPLING AND ANALYSIS
FIELD PROCEDURES
Pre-Collection Preparation
Construction of Sieve Boxes—
If sieving will be conducted in the field, it is recommended that
sieve boxes be used to facilitate processing. Sieve boxes should be sturdy,
and have high sides to minimize the possibility of material washing out
of the box. They should also be large enough to receive the benthic sample
and wash water without completely clogging. Swartz (1978) recommends boxes
40 cm x 40 cm. The boxes should also be constructed to permit nesting
of the sieves, especially if more than one mesh size will be used. A typical
sieve box might be constructed as shown in Figure 1. Note the application
of silicone sealant at the mesh/wood interface. This sealant will prevent
organisms from crawling into the space where the mesh enters the box frame.
All wood pieces used in construction of the sieve boxes should be treated
with fiberglass or epoxy resin (of the types used in boat building), sanded,
and painted.
It is imperative that the mesh used in the sieve boxes meet specifications
outlined in ASTM E-ll, USA Standard Z23.1, AASHO M92, and Fed. Spec. RR-S-366b.
Such mesh is available from scientific supply houses. Inferior mesh will
not have uniform openings and will not be durable.
Before each sample is sieved, all sieves should be examined for damage
and wear. Look for rips in the mesh, irregular mesh spacing, and sand
grains caught in the mesh. Use water pressure or a nylon brush to dislodge
the sand. Do not use sharp objects or stiff brushes, as the mesh may be
damaged or the mesh spacing may be altered.
Fixative Preparation—
The fixative most commonly used for benthic macroinvertebrate samples
is formalin, an aqueous solution of formaldehyde gas. Under no circumstances
should ethyl or isopropyl alcohol (i.e., preservatives) be used in place
of the formalin. Penetration of the alcohol into body tissues is too slow
to prevent decomposition of the specimens.
Caution should be exercised when handling formalin mixtures because
formalin is toxic and carcinogenic (Kitchens et al. 1976). It can cause
irritation to the eyes, nose, and throat at concentrations as low as 1 ppm.
Sensitivity in humans varies with the individual, but in general, the detection
limit is around 2 ppm. Anyone working with formalin mixtures should therefore
11
-------
/o
•SCREEN LAPS OVER BOTTOM SIDEPIECE
CONSTRUCTION
1 Construct upper and lower box frames (A,B)
2. Nail or staple mesh over lower frame.
3. Mount upper frame on lower frame.
4. Nail and glue side pieces (C) on upper and
lower frames (A.B). offsetting to permit
sieves to be nested.
5. Use waterproof glue (resorcinol. epoxy) and
nails throughout construction.
Figure 1. Construction of a sieve box.
12
-------
Benthic Infauna
Protocols for Sampling and Analysis
January 1987
wear protective clothing, rubber gloves, and safety goggles, and should
work under a properly ventilated fume hood. A protective vapor mask should
be worn, even if working near open windows or under a ventilation hood.
Formalin solutions of 5-20 percent (v/v) strength are recommended
for fixing marine organisms (Gosner 1971; Birkett and Mclntyre 1971; Smith
and Carl ton 1975; Swartz 1978). Solutions of 10-15 percent are used most
commonly. It is recommended that at least 2 L of diluted formalin solution
be on hand for each replicate sample to be collected, unless experience
has shown otherwise.
The formalin solution should always be buffered to reduce acidity.
Failure to buffer may result in decalcification of molluscs and echinoderms.
Ideally, pH should be at least 8.2, as calcium carbonate dissolves in more
acidic solutions. Borax (sodium borate, ^38407) should be used as the
buffer because other buffering agents may hinder identification by leaving
a precipitate on body tissues and setae.
To prepare a 10-percent buffered formalin solution, add 4 oz of borax
to each gallon of concentrated formalin (i.e., a 40-percent solution of
formaldehyde in water). This amount will be in excess, so use the clear
supernatant when making seawater dilutions. Dilute the concentrate to
a ratio of one part concentrated formalin to nine parts seawater. Seawater
will further buffer the solution. Seawater also makes the fixative isotonic
with the tissues of the animals, thereby decreasing the potential for animal
tissues to swell and break apart, as often happens with freshwater dilutions
of formalin.
It is recommended that fresh fixative be prepared prior to each sampling
excursion, as formalin will eventually consume all the buffering capacity
of the borax. Formalin solution of any strength should not be exposed
to freezing temperatures, because the formaldehyde polymers will degrade
into paraformaldehyde and the solution will have to be discarded.
Rose Bengal Preparation—
If staining is used, rose bengal may be added to samples either as
a powder or a solution. Both are effective. However, it is easier, and
perhaps less expensive, to use a solution. A rose bengal concentration
of 4 g/L of concentrated formalin commonly is used (Eleftheriou and Holme
1984).
Relaxant Preparation—
If a relaxant is to be used, several kinds are available for use with
benthic organisms. However, a solution of magnesium chloride in tap water
is effective on a wide variety of taxa (Gosner 1971), and is easily prepared
and used. The MgCl2 solution should be isotonic with seawater. To prepare,
dissolve 73 g MgCl2'6H20 per liter of tap water. Anhydrous MgClg can be
13
-------
Benthic Infauna
Protocols for Sampling and Analysis
January 1987
purchased (optionally and at a considerably higher cost) and used to prepare
the relaxant solution. However, accurate determinations of mass are very
difficult because of the propensity of the crystals to absorb atmospheric
moisture. Hence, use of the hydrated form is recommended.
Sample Containers—
Samples can be stored in a variety of containers including glass or
plastic jars, and plastic or muslin bags. If jars are used, plastic lids
are preferable to metal lids because formalin corrodes metal. If glass
jars are used, extra care should be taken when handling, shipping, and
storing them to prevent breakage. If plastic or muslin bags are used,
extra care should be taken to prevent them from tearing.
In general, a single 1- or 2-quart container is large enough to hold
a sieved sample from a 0.1-m2 sampler. However, more or larger containers
may be required if large quantities of gravel, peat, wood chips, or other
large items occur in the sample.
Labels—
A complete label should be placed inside each sample container, as
well as on the side of each container. An abbreviated label may be placed
on the caps of jars to identify them when in shipping or storage cases.
All labels should be waterproof and preprinted. The internal label should
be made of at least 100 percent waterproof rag paper and the external labels
should be gummed. External labels may be filled out using waterproof ink,
but internal labels should be filled out using only a pencil.
Collection
Design of Sampler—
Collection of an acceptable sediment sample for infaunal analysis
generally requires that the sampler 1) create a minimal bow wake when de-
scending, 2) form a leakproof seal when the sample is taken, and 3) prevent
winnowing (i.e., loss of fine-grained material) and excessive sample disturbance
when ascending. A desirable feature of a sampler is easy access to the
sample surface. Reduction of the bow wake is critical to ensuring that
small, lightweight, surface-dwelling organisms are not blown away before
the sampler contacts the sediment. A leakproof seal is necessary to ensure
that organisms are not lost when the sampler is being retrieved. Preventing
sample disturbance is necessary for accurately characterizing the sediment
and measuring penetration depth. Easy access to the sample surface facilitates
sediment characterization and measurement of penetration depth.
The bow wake of several kinds of sampler is reduced by having hinged
solid doors or rubber flaps cover the open upper face of the device. The
rubber flaps generally cover screened doors, which prevent organisms from
14
-------
Benthic Infauna
Protocols for Sampling and Analysis
January 1987
escaping as the sampler is retrieved. Upon descent of the sampler, the
solid doors or rubber flaps are cocked open or held open by water pressure.
Upon ascent, the solid doors are held closed by springs or elastic cords,
whereas the rubber flaps are held closed by water pressure.
Although most samplers seal adequately when purchased, the wear and
tear of repeated field use eventually reduces this sealing ability. A
sampler should therefore be monitored constantly for sample leakage. If
unacceptable leakage occurs, the sampler should be repaired or replaced.
If a sampler is to be borrowed or leased for a project, its sealing ability
should be confirmed prior to sampling. Also, it is prudent to have a back-up
sampler on board the survey vessel in case the primary sampler begins leaking
during a cruise.
Penetration depth (i.e., maximum distance below the sediment surface
that is sampled) generally varies with sediment character for most samplers,
being greatest in fine-grained sediments and least in coarse-grained sediments.
The penetration depth achieved by a particular sampler can often be increased
by attaching lead weights to the device.
Operation of the Sampler—
The sampler should be attached to the hydrowire using a ball-bearing
swivel (Figure 2). The swivel will minimize the twisting forces on the
sampler during deployment and ensure that proper contact is made with the
bottom. For safety, the hydrowire, swivel, and all shackles should have
a load capacity at least 3 times greater than the weight of a full sampler.
The sampler should be deployed and retrieved with a minimum amount
of swinging when out of the water. Excessive swinging can cause the sampler
to trigger prematurely upon deployment and can disturb the sediment sample
upon retrieval. Swinging can be minimized by heading the survey vessel
into any waves when the sampler is out of the water and by attaching handling
lines to the cable that can then be operated by the sampling team (Figure 2).
Because form drag and skin friction of the sampler can produce a bow
wave when the device is lowered too quickly, it is essential that the sample
enter the sediment at a relatively slow speed. It is recommended that
the lowering speed at sediment entry be <0.3 m/sec (<1 ft/sec). Lowering
rates through the water column can be much faster until several meters
from the bottom, as long as the speed at sediment entry is £0.3 m/sec.
Entry at faster speeds requires demonstration that bow waves are not a
problem. Swell and chop can significantly degrade samples due to effects
on entry speed (i.e., vertical ship motion alternately adds to and subtracts
from entry velocity). These additional factors must therefore be taken
into account when they are present.
After the sampler has contacted the bottom, it initially should be
retrieved slowly to permit the device to close properly. After the jaws
15
-------
DAVIT
SNATCH BLOCK
HANDLING LINE
WITH SNAP HOOKS
Figure 2. Deployment of a grab sampler.
16
-------
Benthic Infauna
Protocols for Sampling and Analysis
January 1987
are closed, a constant retrieval speed should be maintained to avoid jerking
the sampler and possibly disturbing the sample. When the sampler approaches
the water surface (i.e., when first sighted), the winch should be stopped
to permit the handling lines to be clipped onto the cable. The sampler
can then be raised slowly, and the handling lines can be used to minimize
swinging of the device. When brought on board, the sampler should be properly
secured as soon as possible.
Sample Acceptability Criteria—
After the sampler has been secured, the sediment sample should be
inspected carefully before being accepted. The following acceptability
criteria should be satisfied:
• Sediment is not extruded from the upper face of the sampler
such that organisms may have been lost
• Overlying water is present (indicates minimal leakage)
• The sediment surface is relatively flat (indicates minimal
disturbance or winnowing)
• The entire surface of the sample is included in the sampler
• The following penetration depths (i.e., the maximum depth
of sediment sampled) are achieved at a minimum
4-5 cm for medium-coarse sand
6-7 cm for fine sand
>10 cm for muddy sediment.
If a sample does not meet any one of these criteria, it should be rejected.
Examples of some acceptable and unacceptable grab samples are presented
in Figure 3.
Sample Characterization-
After a sample is judged acceptable, the following observations should
be noted on the field log sheet:
0 Station location
0 Depth
0 Gross characteristics of the surficial sediment
Texture
Color
Biological structures (e.g., shells, tubes, macrophytes)
17
-------
ACCEPTABLE IF MINIMUM
PENETRATION REQUIREMENT MET
AND OVERLYING WATER IS PRESENT
UNACCEPTABLE (WASHED, ROCK
CAUGHT IN JAWS)
oo
UNACCEPTABLE (CANTED
WITH PARTIAL SAMPLE)
UNACCEPTABLE (WASHED)
Figure 3. Examples of acceptable and unacceptable grab samples.
-------
Benthic Infauna
Protocols for Sampling and Analysis
January 1987
Presence of debris (e.g., wood chips, wood fibers,
manmade debris)
Presence of oily sheen
Odor (e.g., hydrogen sulfide, oil, creosote)
• Gross characteristics of the vertical profile
Changes in sediment characteristics
Presence and depth of redox potential discontinuity
(rpd) layer (if visible)
• Maximum penetration depth (nearest 0.5 cm)
• Comments relative to sample quality
Leakage
Winnowing
Disturbance.
Processing
It is recommended that the entire sample be sieved for benthic infaunal
analyses. If subsamples are removed for physical or chemical analyses,
they should be very small relative to the size of the entire sample (i.e.,
<5 percent) because organisms would be lost from the sample in the process.
If large numbers of organisms are lost at this stage, subsequent abundance
determinations could be biased substantially. Subsamples, other than those
made in situ by box-core partitions, are not recommended for benthic infaunal
analyses because it is unknown what effect the sampling process has on
the spatial distribution of motile organisms. For example, suface-dwelling
organisms may move to the edges of the sample as the grab is being retrieved.
If the sampling process disrupts the natural spatial patterns of the organisms,
collection of a representative subsample for infaunal analysis may not
be possible.
After qualitative characteristics of the sample have been recorded,
sediments should be washed on the designated sieve(s). Sediment adhering
to the outside of the sampler should not be mixed with the sample. When
being sieved, sediments may be gently sprayed with water from above, gently
agitated by hand in a washtub of water (in an up-and-down, not swirling,
motion), or washed using a combination of these techniques. For all methods,
it is imperative that the samples be washed gently to minimize specimen
damage. A few minutes extra care in the field can save hours of time for
the taxonomist, and will result in a better data set.
For many surveys, it is easiest to wash the samples from above with
a gentle spray, because efficient, easy-to-use gear may be constructed
to hold the sampler and sieve boxes. An example of a stand designed to
hold a van Veen grab is shown in Figure 4. The top section is designed
19
-------
SPOUT
SIEVE TRAY
EYE BOLT
REFERENCE: STRIPLIN AND MAUPIN (1982)
Figure 4. Example of a sieving stand. Screen boxes
(not shown) are placed in sieve tray.
20
-------
Benthic Infauna
Protocols for Sampling and Analysis
January 1987
to accept the grab sampler. Wash water and sediment drain through the
openings in the bottom of the top tray and into the lower section of the
sieving stand, where the screen box(es) is (are) located.
All wash water should be filtered (using a cartridge-filter system)
or screened through mesh with openings less than one-half the size of those
used in the survey, so as not to introduce planktonic or bentho-pelagic
organisms into the samples. Failure to screen in this way can result in
increased sorting time. It can also compromise the quality of the resulting
data, because it is impossible to distinguish bentho-pelagic organisms
caught by the grab from those entrained in the wash water.
Sieving stands should have attachment points (e.g., eyebolts) at appro-
priate places with which the stand may be lashed to the deck or rail.
As shown in Figure 4, all wastewater should exit the sieve tray via a spout,
to which a hose can be attached. The wash water can then be discharged
overboard through a scupper. This is especially important in cold weather,
when wash water may otherwise freeze on the deck and safety may be compromised.
Once sieving is completed, the screen box should be held at an angle
and the remaining material gently washed into one corner. The sample may
then be transferred to a container for relaxation, if desired, or for immediate
fixation, using as little water as possible. Place a permanent internal
sample label in the container at this time. If more than one screen fraction
is generated, be sure to keep them separate throughout all phases of field
and laboratory processing. Be sure to check the screen for organisms trapped
in (or wound around) the mesh wires. If they cannot be dislodged with
gentle water pressure, use a pair of jewelers forceps. Be careful not
to damage the wire mesh. After the screen has been checked for remaining
animals and sample removal is complete, back-wash the screen with a high-
pressure spray to dislodge any sediment grains that may be caught in the
mesh.
As mentioned earlier, a 10-15 percent solution of borax-buffered formalin
usually is sufficient to fix benthic organisms. However, samples containing
large amounts of fine-grained sediments, peat, or woody plant material may
require higher concentrations. The volume of fixative should be at least
twice the volume occupied by the sample. The formalin solution should be
added to the sample container until it is completely filled. This will minimize
abrasion during shipping and handling. If the sample volume exceeds one half
of the container volume, more than one container should be used. Use of
multiple containers for single samples should be recorded on the log sheet.
After fixative has been added to a sample container, it is critical
that the contents be mixed adequately. This usually can be accomplished
by inverting the container several times. After mixing, sample containers
should be placed in protective containers for storage and transport to
the laboratory. After being stored for approximately 1 h, samples should
be inverted several times again to ensure adequate mixing.
21
-------
Benthic Infauna
Protocols for Sampling and Analysis
January 1987
On board ship, samples should be stored so as to minimize exposure
to sunlight and temperature extremes. They should also be stored in a
stable part of the ship to minimize agitation.
LABORATORY PROCEDURES
Equipment and Supplies
The laboratory should be equipped with both stereo dissection and
compound microscopes. Magnifying lamps also can be available for sorting
samples. Compound microscopes should be capable of magnifications up to
1,000-power. The optics of the dissection and compound microscopes should
be of the highest quality. Apparent savings realized by purchasing lower
quality optics are quickly consumed by increased labor costs during the
sorting and identification processes. The probability of misidentifying
organisms also is increased. Other recommended laboratory supplies include
jewelers forceps, fine scissors, small scalpels, fine needles, flat and
depression microscope slides, cover slips, small dissection trays, immersion
oil, and glycerol alcohol (half glycerol and half 70-percent alcohol).
Preservative Preparation
After the specimens are fixed, alcohol should be used as a long-term
preservative. Either 70-percent ethanol (v/v) in water or 70-percent isopro-
panol (v/v) may be used (Fauchald 1977). Although isopropanol is less
expensive than ethanol, it is more unpleasant to work with. Specimens
preserved in isopropanol are unsuitable for histological examination.
If future studies of anatomy or reproductive biology are anticipated, ethanol
should be used.
It is most cost-effective to purchase isopropanol and ethanol in bulk
solutions of 5-percent water and 95-percent alcohol. Purer grades are
available, but more costly. To prepare 1 L of a 70-percent solution of
either alcohol, add 263 mL of water to 737 ml of 95-percent alcohol solution.
It may be necessary to use distilled water to dilute the alcohol solution,
because hard water mixed with alcohol creates a milky precipitate that
makes examination of the samples difficult.
Use of the 70-percent alcohol/30-percent water solution is adequate
for the preservation of most infaunal organisms (Fauchald 1977; Eleftheriou
and Holme 1984). For long-term storage of crustaceans, however, it is
recommended that glycerine be substituted for some of the water. The glycerine
helps keep the exoskeletons supple, thereby facilitating examination and
manipulation. This is especially critical for crustaceans archived in
the reference collection (see below). An appropriate alcohol-glycerine
solution would be 70-percent alcohol, 25-percent water, and 5-percent glycerine
(Eleftheriou and Holme 1984).
22
-------
Benthic Infauna
Protocols for Sampling and Analysis
January 1987
Analytical Procedures
Transfer to Alcohol —
Samples should remain in the formal in-seawater solution for a minimum
of 24 h to allow proper fixation (Fauchald 1977). A maximum fixation period
of 7-10 days is recommended to reduce the risk of decalcifying molluscs
and echinoderms. After fixation, the samples should be washed (i.e., re-
screened) on a sieve with mesh openings half the size (at most) of those
used in the field. The smaller screen size ensures that specimens collected
in the field will be retained in the sample regardless of shrinkage or
breakage resulting from contact with the formalin. It is desirable to
wash the formalin from the samples as soon as possible after the initial
24 h, because the buffering capacity of the borax in the formalin solution
decreases continually.
If the sample consists of multiple containers, locate all containers
prior to rescreening and wash them at the same time. Carefully pour the
contents of each container into the appropriately sized screen and rinse
the container to remove adhering organic material, sediment, or organisms.
Do not fill the screen more than half full to avoid spilling or splashing
the sample.
As mentioned earlier, caution should be exercised when handling formalin
mixtures because formalin is toxic and carcinogenic (Kitchens et al. 1976).
It can cause irritation to the eyes, nose, and throat at concentrations
as low as 1.0 ppm. Sensitivity in humans varies with the individual, but
in general, the detection limit is around 2 ppm. Therefore, by the time
formalin generally is detected, it has already caused some irritation.
The technician doing the rescreening should wear protective clothing, rubber
gloves, and safety goggles, and should work under a properly ventilated
fume hood. A protective vapor mask should be worn, even if working near
open windows or under a ventilation hood.
There are several acceptable methods for rinsing formalin from a sample.
One method is to gently flush the sample with large quantities of fresh
water from a low-pressure faucet or hose, being careful not to splash any
sample material. A second method is to partly immerse the sieve in a plastic
tub filled with fresh water and wash the sample by moving the sieve in
an up and down motion. Care must be taken not to let the water rise above
the top level of the sieve.
Allow the rinse water to completely drain from the sieve and lightly
rinse the sample with a solution of 70-percent ethanol from a squirt bottle.
Carefully wash the sample material into a sample jar filling it no more
than three-quarters full. Rinse the last bit of material into the jar
using the squirt bottle of alcohol. Fill the jar to the top with the 70-percent
alcohol solution and screw the lid on tightly. Gently shake and invert
the jar several times to ensure proper mixing.
23
-------
Benthic Infauna
Protocols for Sampling and Analysis
January 1987
Each jar should have one internal label and two external labels.
The internal label should be made of waterproof, 100-percent (at least)
rag paper and filled out using a pencil. Paper with less than a 100-percent
rag content or that is not waterproofed will disintegrate in the 70-percent
alcohol mixture. The two external labels should be preprinted and should
be labeled with an indelible marking pen. One label should be attached
to the side of the jar and the second should be attached to the lid of
the jar. All three labels should include all information recorded on the
field data tag, plus all other information needed to ensure proper identifi-
cation of the sample.
Keep all jars of a given sample together (if more than one), and all
replicate samples from a given station together. As the samples are shelved
prior to sorting, each should be cross-referenced to the field log sheet.
At this point the sample custodian should date and initial the rescreening
section of the sample tracking form for each station. Store washed samples
in an upright position at a cool temperature, and away from direct sunlight.
Storage should be in a secure place, where sample containers are not exposed
to breakage, and samples should be checked periodically to ensure that
adequate levels of preservative are maintained.
Sample Sorting—
Several techniques can be used to sort organisms from sediment. The
most common technique involves placing a small amount of the sample into
a glass or plastic petri dish and'using a pair of jewelers forceps to sort
through the sample in a systematic manner, removing each organism. This
entire process should be done while viewing the sample through a 10-power
dissecting microscope or a magnifying lamp. Care must be taken that enough
liquid is present in the petri dish to completely cover the sample; otherwise,
reflections from the sediment/liquid interface will cause distortions and
the sorter may miss some organisms. Each petri dish of material should
be sorted twice to be sure that all organisms are removed.
A second sorting technique is a flotation method, which is particularly
effective when the sediment residue is primarily coarse sediment grains
containing small amounts of organic matter (e.g., wood fragments, leaf
debris, sewage sludge). The sample is first washed with fresh water in
a large flat tray. The less dense material that becomes suspended in the
fresh water (organic material, arthropods, and most soft-bodied organisms)
is carefully poured into a sieve, and is sorted using the standard technique
described above. The remaining material is covered with liquid and sorted
using a 5-power self-illuminated hand lens. Organisms remaining in this
portion of the sample generally include molluscs and some tube-dwelling
or encrusting organisms that are associated with sand grains. Because
it is difficult to see extremely small organisms with the 5-power hand
lens, the sorter must remove all molluscs and polychaete tube fragments
for closer inspection. All material collected from this portion is placed
24
-------
Benthic Infauna
Protocols for Sampling and Analysis
January 1987
into a labeled sample jar and viewed under a 10-power dissecting microscope
to remove organisms from tubes and to ensure that the molluscs were alive
when captured.
Whichever technique is used, the sorter is exposed to alcohol fumes.
Because these fumes can be irritating to some people, the sorting process
can be done using fresh water. However, as each portion of the sample
is sorted, it should be drained and returned to the alcohol solution im-
mediately.
Each sample should be sorted by only one person. At a minimum, organisms
should be sorted into the following major taxonomic groups: Annelida,
Arthropoda, Mollusca, Echinodermata, and miscellaneous phyla (combined).
All organisms should be placed in large vials containing 70-percent alcohol
solution. The exception is Ophiuroidea, which require air-drying for identifi-
cation. Removal of the majority of arms from certain Ophiuroidea (e.g.,
Amphiuridae) permits easier identification. This preparation may be performed
by experienced sorters to minimize identification time. Special handling
of Ophiuroidea should be conducted after biomass analyses, if biomass analyses
are performed. Each vial containing a major taxonomic group should have
an internal label listing the survey name, station designation, water depth,
date sampled, and field screen size. All vials from the same sample should
be stored in a common container and immersed in the 70-percent alcohol
solution. To reduce evaporation of alcohol, vial and container lids can
be sealed with plastic tape.
Biomass Determination—
When required, biomass estimates for the major taxonomic groups should
be made prior to identifying the organisms to the species level. It is
recommended, however, that taxonomists examine the major taxonomic groups
before biomass measurements are made, to ensure that sorters have correctly
grouped all individuals and fragments and that the remains of dead organisms
(e.g., empty mollusc shells) are not included. Biomass should be estimated
to the nearest 0.1 g (wet weight). All specimens of taxa within the following
major groups should be composited for biomass analyses: Annelida (principally
polychaete worms), Mollusca (principally bivalves, gastropods and aplaco-
phorans), Arthropoda (principally crustaceans), Echinodermata (principally
asteroids, ophiuroids, echinoids, and holothuroids), and miscellaneous
taxa (combined). These five categories generally are adequate to characterize
the standing stocks of the major infaunal groups. They also are sufficiently
distinct from each other to permit proper assignment of fragments to each
of the groups. All fragments should be placed in their respective major
taxonomic groups prior to weighing.
There are several major problems associated with the collection and
interpretation of biomass information. Some taxa lose weight when immersed
in preservative fluids, while others gain weight (Howmiller 1972; Lappalainen
and Kangas 1975; Wiederholm and Eriksson 1977; Mills et al. 1982). For
25
-------
Benthic Infauna
Protocols for Sampling and Analysis
January 1987
this reason, the most accurate biomass estimates are performed on live
material. However, it is rarely practical to sort and weigh live specimens.
Accurate measurements of biomass may be compromised further by evaporation
from the specimens while they are on the balance. Lastly, biomass measurements
are only estimates of standing crop. They do not reflect estimates of
production because all organisms are treated in the same manner whether
they are large and long-lived, or small and short-lived. Because of these
problems, biomass measurements should be interpreted carefully.
Several methods of measuring biomass are possible. One technique
is to estimate the difference in weight of a tared beaker filled with preserva-
tive before and after organisms are placed in the beaker. The individual
organisms are not blotted prior to weighing, and as few individuals as
possible are transferred to the weighing container. These procedures minimize
the transfer of fluids held within a pile of individuals. This technique
can be used for preserved or live animals, and appears to introduce the
least amount of variation into the weighing process.
A second technique for biomass determination consists of air-drying
the organisms on absorbent paper for a specific length of time (e.g., 5 min).
Because 70-percent ethanol is volatile, small variations in drying time
may increase the errors associated with the weight measurements. A container
open at one end and covered at the other end with a 0.25-mm mesh screen
(maximum mesh opening) can be used to hold the organisms for weighing.
After the tare weight of the container is measured, the animals are carefully
placed into the container. The container with organisms is then placed
on a paper towel and allowed to air dry for exactly 5 min prior to weighing.
The weight of the organisms is obtained by subtracting the weight of the
container with the organisms from the tare weight of the container. Extremely
large organisms (e.g., large molluscs or asteroids) should be weighed indi-
vidually.
Taxonomic Identification—
After biomass estimates are completed, identification and counting
of the organisms may begin. Unless otherwise specified, identifications
should be to the lowest taxonomic level possible, usually the species level.
For incomplete specimens, enumerate only the anterior or posterior ends,
depending upon the taxon. All identifications should be made using binocular
dissecting or compound microscopes. If possible, at least two pieces of
literature should be used for each species identification. Moreover, each
species identification should be checked against a reference specimen from
a verified reference collection (see QA/QC Procedures).
After completing taxonomic identifications, all organisms should be
placed in vials containing 70-percent alcohol. All vials for a single
sample should be stored in common jars and immersed in 70-percent alcohol.
Each vial should contain an internal label with the following information:
survey name, station number, replicate number, collection gear, water depth,
26
-------
Benthic Infauna
Protocols for Sampling and Analysis
January 1987
and date of collection. Any specimens removed from the sample jar and
placed in the reference collection should be so noted (species, number)
on the sample identification sheet.
Each taxonomist should record initial identifications and counts in
a notebook, which should also include notes and comments on the organisms
in each sample. Upon completion of the sample, the data should be transferred
to the sample data sheets and double-checked. The taxonomist should then
sign and date the sample data sheet. All notebooks should be kept in the
laboratory at all times so the laboratory supervisor can check questionable
identifications and follow the progress of each sample.
QA/QC PROCEDURES
Calibration and Preventive Maintenance
The analytical balance used for biomass determinations should be calibrated
weekly, at a minimum. The balance and all microscopes should be serviced
at regular intervals. Annual service and inspection is adequate in most
cases, unless the manufacturer recommends otherwise.
Taxonomic identifications should be consistent within a given laboratory,
and with the identifications of other regional laboratories. To that end,
at least three individuals of each taxon should be sent for verification
to recognized experts. The verified specimens should then be placed in
a permanent reference collection. Continued collection of a verified species
does not require additional expert verification, because the reference
collection can be used to confirm the identification. Participation of
the laboratory staff in a regional taxonomic standardization program (if
available) is recommended, to ensure regional consistency and accuracy
of identifications.
All specimens in the reference collection should be held in labeled
vials that are segregated by species and sample. For example, there may
be three labeled vials of Gemma gemma, one from each of three samples.
More than one specimen may be in each vial. The labels placed in these
vials should be the same as those used for specimens in the sample jars.
It is important to complete these labels, because future workers may not
be familiar with the survey, station locations, and other details of the
work in progress. In addition, the reverse side of the label should contain
information about the confirmation of the identification by experts in
museums or other institutions (if appropriate). Such information would
include the name and institution of the outside expert, and date of verifi-
cation. All vials for a given species should be placed in a single jar
filled with alcohol. To reduce evaporation of alcohol, the lids of vials
and jars can be sealed with plastic tape wrapped in a clockwise direction.
The species (or other taxonomic designation) should be written clearly
on the outside and on an internal label. Reference specimens should be
27
-------
Benthic Infauna
Protocols for Sampling and Analysis
January 1987
archived alphabetically within major taxonomic groups. A listing of each
species name, the name and affiliation of the person who verified the identi-
fication, the location of the individual specimen in the museum, the status
of the sample if it has been loaned to outside experts, and references
to pertinent literature should be maintained by the laboratory performing
the identifications.
Reference specimens are invaluable, and should be retained at the
location where the identifications were performed, in the offices of the
funding agencies, or at a museum with long-term storage capabilities.
In no instance should this portion of the collection be destroyed. A single
person should be identified as the curator of the museum collection and
should be responsible for its integrity. Its upkeep will require periodic
checking to ensure that alcohol levels are adequate. When refilling the
jars, it is advisable to use ful1-strength alcohol (i.e., 95 percent),
because the alcohol in the 70-percent solution will tend to evaporate more
rapidly than the water.
Quality Control Checks
It is recommended that at least 20 percent of each sample be re-sorted
for QA/QC purposes. Re-sorting is the examination of a sample or subsample
that has been sorted once and is considered free of organisms. The 20-
percent aliquot should be taken after the entire sample has been spread
out in a pan or tray. It is critical that the aliquot be a representative
subsample of the total sample. Care should be taken to include any organisms
that may be floating in the preservative. Re-sorting should be conducted
using a dissection microscope capable of magnification to 25-power. A
partial re-sorting of every sample should ensure that all gross sorting
errors are detected. In addition, it should give added incentive to sorters
to process every sample accurately. Re-sorting should be conducted by
an individual other than the one who sorted the original sample.
In addition to efficient sample sorting, consistent identification
of organisms among individuals and among sampling programs are critical
to the collection of high quality data. Consistent identifications are
achieved by implementing the procedures discussed below and by maintaining
informal, but constant, interaction among the taxonomists working on each
major group. One important procedure is to verify identifications by comparison
with the reference collection. To ensure that identifications are correct
and consistent, 5 percent of all samples identified by one taxonomist should
be re-identified by another taxonomist who is also qualified to identify
organisms in that major taxonomic group. It is the duty of the senior
taxonomist to decide upon the proper identification(s). The senior taxonomist
may also decide whether the taxonomic level to which a given organism is
identified is appropriate. If it is not, the senior taxonomist may decide
to drop back to a higher taxonomic level, or to further refine the taxonomy
of that group through additional study.
28
-------
Benthic Infauna
Protocols for Sampling and Analysis
January 1987
When all identification and QA/QC procedures are completed, the jars
containing the vials of identified species should be topped off with 5-
percent glycerine/70-percent alcohol. The lids should then be sealed tightly
with black electrical tape to prevent evaporation. All sample jars should
be placed in containers filled with 70-percent alcohol for long-term storage.
The containers should be fitted with a tightly sealed lid, and electrical
tape should again be used to seal the joints. Each container should be
labeled clearly with the survey name, date, and number and type of samples
within it.
Corrective Action
Following QA/QC procedures discussed earlier, each 20-percent sample
aliquot should be checked for complete or nearly complete removal of organisms.
Thus, each sample elicits a decision concerning a possible re-sort. When
a sample is found that does not meet the recommended 95-percent removal
criterion (see Data Quality and Reporting Requirements below), it should
be re-sorted.
When a taxonomic error or inconsistency is found, it is necessary
to trace all of the work of the taxonomist responsible for the error, so
as to identify those samples into which the specific error or inconsistency
may have been introduced. This process can be very time-consuming. However,
upon completion of all taxonomic work, few (if any) taxonomic errors or
inconsistencies should remain in the data set. Avoiding errors and inconsis-
tencies through the constant interchange of information and ideas among
taxonomists is the best way to minimize lost time due to faulty identification.
DATA QUALITY AND REPORTING REQUIREMENTS
A sample sorting efficiency of 95 percent of total number of individuals
generally is considered acceptable. That is, no more than five percent
of the organisms in a given sample are missed by the sorter. Similarly,
species identifications by each taxonomist can reasonably be expected to
be accurate for at least 95 percent of the total number of species. Unless
otherwise specified, all organisms should be identified to the lowest possible
taxon; to species level whenever possible. In cases where the identity
of a species is uncertain, a species number will suffice (e.g., Macoma
sp.l, Macoma sp.2). Numerical designations must be consistent throughout
each study. To facilitate comparability among different studies, the dis-
tinguishing characteristics of each unidentified species should be recorded.
Data for each replicate sample should be reported as numbers of individuals
per sample for each species and as biomass (nearest 0.1-g wet weight per
sample) for each major taxonomic group.
29
-------
REFERENCES
Birkett, L., and A.D. Mclntyre. 1971. Treatment and sorting samples.
pp. 156-168. In: Methods for Study of Marine Benthos. N.A. Holme and
A.O. Mclntyre (eds). IBP Handbook No. 16. Blackwell Scientific Publications,
Oxford, UK.
Eleftheriou, A., and N.A. Holme. 1984. Macrofauna techniques, pp. 140-216.
In: Methods for the Study of Marine Benthos. N.A. Holme and A.D. Mclntyre
(eds). Blackwell Scientific Publications, London.
Elliott, J.M. 1971. Some methods for the statistical analysis of samples
of benthic invertebrates. Scientific Publication No. 25, Freshwater Biological
Assn., Ferry House, UK. 148 pp.
Fauchald, K. 1977. The polychaete worms; definitions and keys to the
orders, families, and genera. Science Series 28. Natural History Museum
of Los Angeles County. Los Angeles, CA. 188 pp.
Gosner, K.L. 1971. Guide to identification of marine and estuarine inver-
tebrates. Wiley-Interscience, New York, NY. 693 pp.
Gray, J.S. 1981. The ecology of marine sediments. Cambridge University
Press, London. 185 pp.
Hessler, R.R., and P.A. Jumars. 1974. Abyssal community analysis from
replicate box cores in the central North Pacific. Deep-Sea Res. 21:185-209.
Howmiller, R.P. 1972. Effects of preservatives on weights of some common
macrobenthic invertebrates. Trans Am. Fish. Soc. 101:743-746.
Kahlsico. 1986. Catalogue for Kahl Scientific Instrument Corporation.
P.O. Box 947, El Cajon, CA 92022.
Kitchens, J.F., R.E. Casner, G.S. Edwards, W.E. Harward III, and B.J. Macri.
1976. Investigation of selected potential environmental contaminants:
formaldehyde. EPA-560/2-76-009. U.S. Environmental Protection Agency,
Washington, DC.
Lappalainen, A., and P. Kangas. 1975. Littoral benthos of the northern
Baltic Sea. II. Interrelationships of wet, dry, and ash-free weights of
macroinfauna in the Tvarminne area. Int. Rev. Gesamten Hydrobiol. 60:297-312.
Lewis, F.G. Ill, and A.W. Stoner. 1981. An examination of methods for
sampling macrobenthos in seagrass meadows. Bull. Mar. Sci. 31:116-124.
Lie, U. 1968. A quantitative study of benthic infauna in Puget Sound,
Washington, USA, in 1963-1964. Fisk Dir. Skr. Ser. HavUnders. 14:229-556.
30
-------
Mills, E.L., K. Pittman, and B. Munroe. 1982. Effect of preservation
on the weight of marine benthic invertebrates. Can. J. Fish. Aquat. Sci.
39:221-224.
Rees, H.L. 1984. A note on mesh selection and sampling efficiency in
benthic studies. Mar. Pollut. Bull. 15:225-229.
Reish, O.J. 1959. A discussion of the importance of screen size in washing
quantitative marine bottom samples. Ecology 40:307-309.
Schwinghamer, P. 1981. Characteristic size distributions of integral
benthic communities. Can. J. Fish. Aquat. Sci. 38:125-1263.
Smith, R.I., and J.T. Carlton (eds). 1975. Light's manual: intertidal
invertebrates of the central California coast. University of California
Press, Berkeley, CA. 716 pp.
Striplin, P.L., and S.H. Maupin. 1982. Custom-designed sieving stations
for small research vessels. Task report to the U.S. Environmental Protection
Agency. Corvallis, OR.
Swartz, R.C. 1978. Techniques for sampling and analyzing the marine macro-
benthos. EPA-600/3-78-030. U.S. Environmental Protection Agency, Corvallis,
OR. 27 pp.
Wiederholm, T., and L. Eriksson. 1977. Effects of alcohol preservation
on the weights of some benthic invertebrates. Zoon 5:29-31.
31
-------
FINAL REPORT
TC-3991-04 «- * Estuary Program
RECOMMENDED PROTOCOLS FOR MEASURING
ORGANIC COMPOUNDS IN PUGET SOUND
SEDIMENT AND TISSUE SAMPLES
Prepared by:
TETRA TECH, INC.
Prepared for:
U.S. ENVIRONMENTAL PROTECTION AGENCY
Region 10 - Office of Puget Sound
Seattle, WA
December, 1986
TETRA TECH, INC.
11820 Northup Way
Bellevue, WA 98005
-------
CONTENTS
Page
LIST OF FIGURES iv
LIST OF TABLES iv
ACKNOWLEDGEMENTS iv
INTRODUCTION 1
ORGANIC COMPOUNDS IN SEDIMENTS 8
USES AND LIMITATIONS 8
SAMPLING PREPARATION AND FIELD PROCEDURES 9
LABORATORY ANALYTICAL PROCEDURES 13
ORGANIC COMPOUNDS IN TISSUE 21
USES AND LIMITATIONS 21
SAMPLING PREPARATION AND FIELD PROCEDURES 22
LABORATORY ANALYTICAL PROCEDURES 24
INSTRUMENTAL PROCEDURES 28
QA/QC PROCEDURES AND REQUIREMENTS 29
SURROGATE SPIKE COMPOUNDS (RECOVERY INTERNAL STANDARDS) 30
INJECTION INTERNAL STANDARDS 34
METHOD BLANKS 36
STANDARD REFERENCE MATERIALS (SRM) 38
MATRIX SPIKES 40
METHOD SPIKES 42
ANALYTICAL REPLICATES 43
FIELD REPLICATES 45
ii
-------
INITIAL CALIBRATION 46
ONGOING CALIBRATION 47
DATA REPORTING REQUIREMENTS 51
RECOVERY AND BLANK CORRECTIONS 52
LOWE"R LIMIT OF DETECTION 53
COST IMPLICATIONS 55
REFERENCES 59
GLOSSARY 61
APPENDIX A - U.S. EPA CONTRACT LABORATORY PROGRAM: PROCEDURES FOR
ANALYSIS OF EXTRACTABLE ORGANIC COMPOUNDS IN SOILS/SEDIMENT
APPENDIX B - U.S. EPA CpNTRACT LABORATORY PROGRAM: PROCEDURES FOR
ANALYSIS OF PURGEABLE ORGANIC COMPOUNDS
APPENDIX C - ESTABLISHED U.S. EPA ADVISORY LIMITS FOR PRECISION AND
ACCURACY AND METHOD PERFORMANCE LIMITS FOR ANALYTICAL
PROCEDURES
APPENDIX D - GC/MS IDENTIFICATION OF TARGET AND LIBRARY SEARCH COMPOUNDS
111
-------
FIGURES
Number Page
1 Cost implication of minimum recommended QA samples for Puget
Sound programs (as a function of numbers of field samples
analyzed) 58
TABLES
Number Page
1 Organics workshops attendees 2
2 Summary of analytical procedures and detection limits for
organic compound analyses 4
3 Summary of quality control samples 5
4 Summary of sample collection and preparation QA/QC
requirements for organic compounds 10
5 Summary of warning and control limits for quality control
samples 31
6 Compounds that must meet ongoing calibration control limit 49
7 Approximate cost range of analyses as a function of matrix,
detection limits, and precision 56
ACKNOWLEDGEMENTS
This chapter was prepared by Tetra Tech, Inc., under the direction
of Dr. Scott Becker, for the U.S. Environmental Protection Agency in partial
fulfillment of Contract No. 68-03-1977. Dr. Thomas Ginn of Tetra Tech
was the Program Manager. Mr. John Underwood and Dr. John Armstrong of
U.S. EPA were the Project Officers. The primary authors of this chapter
were Mr. Robert Barrick, Mr. Harry Seller, and Ms. Julia Wilcox of Tetra
Tech.
iv
-------
ORGANIC COMPOUNDS
INTRODUCTION
DECEMBER 1986
INTRODUCTION
This protocol for analysis of organic compounds is one of a series
of protocols for the measurement of environmental variables in Puget Sound.
Its purpose is to encourage all investigators to use acceptable and comparable
methods when measuring contaminants in Puget Sound. No procedure has been
formally approved by a regulatory agency for the analysis of low parts-
per-billion concentrations of organic contaminants in estuarine sediments
and tissue samples. Multiple procedures for the analysis of different
compound classes used by laboratories and the choice of different options
could yield equivalent results. This document summarizes procedures that
will enable an assessment of the comparability of data sets when analytical
techniques vary among laboratories or within a laboratory over time.
The analytical techniques and quality assurance/quality control (QA/QC)1
guidance provided in this report have been abstracted and combined from
multiple written sources [e.g., U.S. EPA 1984a, 1984b; Horwitz et al. 1980;
NUS 1985; Municipality of Metropolitan Seattle (Metro) 1981; MacLeod et
al. 1984; Brown et al. 1985; Tetra Tech 1985, 1986a, 1986b, 1986c], from
discussions at a series of Puget Sound Estuary Program (PSEP) workshops
on organics protocols, and from a national quality assurance workshop sponsored
by the National Oceanic and Atmospheric Administration (NOAA) and the National
Bureau of Standards (NBS). Participants at the PSEP workshops (Table 1)
included representatives from regional commercial and research laboratories,
government agencies, and private contractors responsible for generation
and interpretation of environmental chemistry data.
IA glossary is included at the end of the document to define key words and
abbreviations.
1
-------
TABLE 1. ORGANICS WORKSHOPS ATTENDEES
Name
Organization
Workshop
1 2 3
John Armstrong
Bob Barrick
Tim Bates
Scott Becker
Harry Seller
Jim Bentley
Joe Blazevich
Don Brown
Jim Bruya
Rob Deverall
Bob Dexter
Andrew Friedman
Mark Fugiel
Tom Ginn
Burt Hamner
Mike Hiatt
Mike Higgins
Dick Huntamer
Roger Kadeg
Carl Kassebaum
Peggy Knight
Catherine Krueger
Jim Krull
Larry LaFleur
Rob Lowe
Dick Lucke
Bill MacLeod
Bob Matsuda
Merley McCall
Barbara McNatt
Alan Mearns
Dave Mitchell
Shawn Moore
Paulette Murphy
Mike Nelson
Bob Ozretich
John Park
Bob Pastorok
Gordon Pol ley
Bob Randall
Bob Rieck
Bob Riley
Mike Schlender
Jim Thornton
Mark Weidner
Julia Wilcox
Bill Yake
John Yee
Larry Young
U.S. EPA Seattle x
Tetra Tech x x
NOAA/PMEL x
Tetra Tech x
Tetra Tech x x
Analytical Technologies x
U.S. EPA Manchester x x
NOAA/NMFS x x
Farr, Friedman, and Bruya x x
ASL Laboratories x
EVS Consultants x
NOAA/NMFS
Am Test x x
Tetra Tech x
U.S. C9E x x
Analytical Laboratories x x
Analytical Laboratories x x
WDOE x x
Envirosphere x
U.S. EPA Seattle
Weyerhaeuser Company x x
U.S. EPA Seattle x
WDOE
NCASI x x
DSHS x
Battelle x
NOAA/NMFS x
Metro
WDOE x x
Laucks Testing Labs x
NOAA/OAD x
Metro x
Am Test x x
NOAA/PMEL x
Laucks Testing Labs x x
U.S. EPA Newport x x
ASL Laboratories x
Tetra Tech x
Can Test x
U.S. EPA Newport x
U.S. EPA Manchester x
Battelle x x
WDOE x
WDOE
Analytical Resources x
Tetra Tech x x
WDOE
Can Test x x
Can Test x
1 = July 29, 1985
2 = September 25, 1985
3 = November 15, 1985.
-------
ORGANIC COMPOUNDS
INTRODUCTION
DECEMBER 1986
A summary of the basic analytical techniques and detection limits
addressed in this document is given in Table 2. A summary of quality control
samples that enable verification of the adequacy of these procedures is
given in Table 3. Specific guidance on limits of acceptability (i.e.,
warning and control limits requiring corrective action) is summarized for
each type of quality control sample in the "QA/QC PROCEDURES AND REQUIREMENTS"
section.
This document is intended primarily as a reference document for analytical
laboratory staff and technical experts. Managers who plan studies and
request analyses should take the following steps with the assistance of
these experts:
• Determine compounds of interest (e.g., phenols, hydrocarbons,
PCBs, chlorinated benzenes, phthalates, pesticides, PCBs,
volatiles) and required detection limits for each sample
matrix (e.g., sediments, tissues; see Table 2) that are
consistent with project objectives and available resources.
• Issue a statement of work that 1) details the analytical
procedures selected from Table 2, 2) incorporates the number
of QC samples recommended in this document (Table 3), 3)
specifies data reporting requirements (see page 42), and
4) specifies corrective action that will be required if
the QA requirements are not met. Technical advice may be
required to select the appropriate analytical procedure
because several alternative techniques for each stage of
analysis' were recommended (Table 2) by the participants
in the workshops.
-------
TABLE 2. SUMMARY OF ANALYTICAL PROCEDURES AND DETECTION LIMITS
FOR ORGANIC COMPOUND ANALYSES3
Detection Limit
Compound Type
Sediment Tissue
(ug/kg dry wt) (ug/kg wet wt)
Analytical Procedure
Volatiles
Semivolatiles'
Semivolatiles
Sample size
10-20
500-1,000
1-50
5-10
10-20
Sample drying
Extraction
Extract drying
Extract concentration
Extract cleanup
Extract analysis
Heated purge-and-trap, or vacuum
extraction/purge-and-trap
Use U.S. EPA
procedure as
analyses
CLP "low-level"
screeni ng level
50 to 100 g wet weight sediment
25 g wet weight tissue
Centrifugation or sodium sulfate
Shaker/roller; Soxhlet; or sonicationb
Separatory funnel partitioning as
needed to remove water (pH must
be controlled); sodium sulfate
used for all other extract drying
Kuderna-Danish apparatus (to
1 mL) or rotary evaporation (to
2 mL); purified N2 stream for
concentration to smaller volumes
- Remove elemental sulfur (sediments
only) with mercury or activated copper
- Remove organic interferents
with GPC, Sephadex, bonded octadecyl
columns, HPLC, silica gel, alumina
gel (for PCB/pesticides)
GC/MS; GC/FID; GC/ECD
The U.S. EPA CLP procedure was developed for "low level" analysis of hazardous
wastes (i.e., hundreds of parts per billion); these procedures are used only as "screening
level" analyses in Puget Sound environmental samples.
The steps described generally apply to low parts-per-billion, full scan analyses.
Some of the options for extract cleanup and analysis are best suited for certain
compound groups rather than full scan analyses. See "Laboratory Analytical Procedures"
sections for full description of options as applied to sediments and tissues.
-------
TABLE 3. SUMMARY OF QUALITY CONTROL SAMPLES
Analysis
Type
Recommended Frequency of Analysis
Surrogate spikes
Method blank
Standard reference
materials
Matrix spikes
Spiked method blanks
Analytical replicates
Field replicates
Required in every sample - minimum 3 neutral, 2 acid spikes,
plus 1 spike for pesticide/PCB analyses, and 3 spikes for
volatiles. Isotope dilution technique (i.e., with all available
labeled surrogates) is recommended for full scan analyses
and to enable recovery corrections to be applied to data.
One per extraction batch (semivolatile organics)
One per extraction or one per 12 hour shift, whichever is
most frequent (volatile organics)
<50 samples:
>SO samples:
one per set of samples submitted to lab
one per SO samples analyzed
Not required if complete isotope dilution technique used
<20 samples: one per set of samples submitted to lab
>20 samples: 5 percent of total number of samples.
As many as required to establish confidence in method before
analysis of samples (i.e., when using a method for the first
time or after any method modification).
<20 samples: one per set of samples submitted to lab
>20 samples: one triplicate and additional duplicates
for a minimum of 5 percent total replication.
At the discretion of the project coordinator.
a The definition of each type of quality control sample is given in the "QA/QC Procedures
and Requirements" section and Glossary.
b Frequencies listed are minimums; some programs may require higher levels of effort.
See "QA/QC and Requirements section for full descriptions of recommended frequencies.
c As available (see "Standard Reference Materials" section).
-------
ORGANIC COMPOUNDS
INTRODUCTION
DECEMBER 1986
• Secure bids from laboratories based on the analytical procedure
selected, kinds of analytes desired (e.g., if the entire
range of U.S. EPA priority pollutant acid/neutral/PCB/pesticide
compounds is required, then full-scan extractable analyses
should be requested), the desired level of effort for QA/QC
(minimum recommended requirements are listed in Table 3),
and the support documentation required to evaluate the quality
of the data (e.g., new data including quantitation reports;
chromatograms; calculation algorithms; instrument calibration,
fine tuning, and mass calibration; surrogate percent recovery
summaries).
• Review returning data packages for completeness to ensure
that the sample data and QA/QC information requested are
present.
• Review QA/QC data packages to determine if data quality
objectives specified in the "QA/QC PROCEDURES AND REQUIREMENTS"
section have been satisfied.
Minimum QA/QC requirements should be established before any laboratory
work is begun and discussed with the laboratory staff. Program managers
and project coordinators may wish to monitor laboratory performance before
analysis of the actual samples is initiated. For example, project coordinators
may evaluate accuracy by comparing analytical results for standard reference
materials (SRM-test materials of known composition) with established acceptance
limits. The overall precision of replicate analyses (the measure of sample
variability) can also be compared with a summary of precision expected
for similar analyses. Based on this review, some corrective actions (i.e.,
measures undertaken to correct a laboratory or data problem) may be recommended
by program managers or project coordinators. Laboratories or QA reviewers
6
-------
ORGANIC COMPOUNDS
INTRODUCTION
DECEMBER 1986
also may contact the project coordinator concerning specific project or
program goals before implementing corrective actions. Most technical guidelines
and QA documentation (e.g., original quantification reports) recommended
in this document are intended for use by professional chemists and technical
experts in independent QA data review.
The PSEP organic protocols workshops emphasized discussion of technically
acceptable minimum QA/QC requirements that would allow assessment of inter-
laboratory comparability, regardless of the specific analytical procedure
used. The QA/QC measures recommended in this protocol are consistent with
those of the established U.S. EPA Contract Laboratory Program (CLP), with
the exception of a few modifications suggested by regional experts specifically
for application to Puget Sound related research. It is expected that the
overall QA/QC protocol will evolve over time to allow appropriate changes
in techniques and quality control limits (i.e., criteria used to indicate
unacceptable results of laboratory analysis).
Appropriate detection limits and minimum basic procedural steps (e.g.,
extraction, removal of selected interferences, quantification, QA/QC) that
would promote reliable recovery of analytes (i.e., compounds of interest)
were agreed upon at the workshops and are summarized in this document.
Chemicals covered by this document include U.S. EPA priority pollutants
that can be analyzed by "full-scan" techniques, with the exceptions discussed
in the next section, "USES AND LIMITATIONS." Techniques judged by workshop
participants to be appropriate at different stages of these "full-scan"
analyses are summarized. Procedures described in this document also apply
to the analysis of some additional compounds (e.g., coprostanol).
Workshop participants agreed that a single step-by-step analytical
procedure analysis of organic compounds could not be recommended over all
others at this time. Such an analytical protocol will require interlaboratory
comparisons and review of results for most analytes of interest.
7
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN SEDIMENTS
DECEMBER 1986
ORGANIC COMPOUNDS IN SEDIMENTS
USES AND LIMITATIONS
The various techniques described in this section are suitable for
the analysis of some or all of the semi volatile and volatile organic priority
pollutants in sediments. Based on discussion at the PSEP workshops, three
compound groups are not suitable target compounds for routine full-scan
analysis: organic bases, halogenated ethers, and hexachlorocyclopentadiene.
These compound groups were excluded because they have rarely been detected
in Puget Sound, they have been detected at very low levels, or they present
exceptional analytical difficulties. However, the organic base N-nitrosodi-
phenylamine has been detected in several sediment samples and is included
in this discussion.
Additional (i.e., not priority pollutant) analytes recommended by
workgroup participants include coprostanol (a human and anin^ fecal indicator);
9(H)-carbazole (a component of creosote and coal tar); polychlorinated
butadienes (detected at high levels in certain areas of Puget Sound); and
polychlorinated styrenes. Some of these compounds may require specific
analytical methods for detection at low parts-per-bi 11 ion concentrations.
Thus program coordinators should confer with technical experts to evaluate
project objectives and the cost-effectiveness of including these compounds
in each environmental study.
As noted in the introduction, a single step-by-step laboratory procedure
has not been recommended. For this reason, considerable weight is placed on
the recommended QA/QC procedures to establish a basis for comparing data. OA/QC
procedures assess performance relative to specific compounds. For example,
8
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN SEDIMENTS
DECEMBER 1986
results from the repeated analysis of SRM are useful for assessing inter-
and intralaboratory precision and accuracy, but no SRM are available that
contain all potential analytes. Hence, the compounds used in QA/QC comparisons
among different laboratory procedures must always be specified.
The emphasis of this protocol is on full-scan analyses rather than
dedicated analysis for one group of related compounds (e.g., hydrocarbons).
However, the QA/QC procedures are applicable to most types of analyses.
Improved detection limits and performance may be attained by focusing on
certain compounds or related compound classes, but at the cost of considerably
more laboratory effort. Focused analyses may be preferable when analyzing
for a limited number of analytes whose expected concentrations are low
(e.g., hydrocarbons at an oil spill site and a reference area).
SAMPLE PREPARATION AND FIELD PROCEDURES
Collection
Guidelines for the collection and acceptance of surficial sediment
samples are provided elsewhere in this series of Puget Sound protocols
(see Protocols for Measuring Conventional Sediment Variables). Before
removing a representative subsample from acceptable grab samples, the overlying
water in the sampling device should be siphoned off with minimum disturbance
to the surface layer of sediment. To avoid potential contamination from
the sampling device, sediment in contact with the sides of the device should
be excluded. Other potential sources of contamination include grease from
ship winches or cables, ship engine exhaust, dust, and ice used for cooling
(if containers are not properly sealed). Utensils for removing sediment
from the sampler should be made of glass, stainless steel, or polytetrafluoro-
ethylene [PTFE, e.g., Teflon (TM)]. Utensils should be solvent-rinsed
and air-dried before each use. Sample collection and preparation requirements
are summarized in Table 4.
9
-------
TABLE 4. SUMMARY OF SAMPLE COLLECTION AND PREPARATION
QA/QC REQUIREMENTS FOR ORGANIC COMPOUNDS
Variable
Sediments
Semivolatiles
Volatiles
Tissues (Whole)
Tissues
(After Resection)
Semivolatiles
Volatiles
Sample
Size3 Contained
50-100g G
40 ml Gd
A
25g G,T
5g G,T
Preservation
Freeze
Cool, 4° ce
Freeze
Freeze
Freeze
Maximum
Holding
Time
1 yrc
14 days
6 moc
6 moc
14 daysc
a Recommended field sample sizes for one laboratory analysis. If additional
laboratory analyses are required (i.e., replicates). The field sample
size should be adjusted accordingly.
b G = Glass, A = Wrapped in aluminum foil, placed in watertight plastic
bags, T = PTFE (Teflon).
c This is a suggested holding time. No U.S. EPA criteria exist for the
preservation of this variable.
d No headspace or airpockets should remain.
e Freezing these samples will likely cause breakage of the sample container,
because no airspace for expansion is provided.
10
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN SEDIMENTS
DECEMBER 1986
Thorough mixing of the initial sample is required when removing subsamples
for different chemical analyses. Homogenization is also important when
the combined contents of several sediment grab samples are required to
provide sufficient material for testing. Compositing may be performed
by first transferring sediment to a dry, solvent-rinsed stainless steel
or glass bowl and then stirring with a clean stainless steel spoon or spatula
until textural and color homogeneity are achieved. All decisions regarding
items that should be removed from the sample (e.g., twigs, leaves, shells,
rocks) should be made in the field and recorded in the field sampling logbook.
The bowl and all utensils should be solvent-rinsed between composites,
and kept covered with aluminum foil to prevent airborne or other contamination.
Sample containers must be carefully cleaned prior to sample collection.
Separate samples are typically required for volatile and semivolatile organic
compounds. Container preparation and collection techniques differ for
these two chemical groups.
Sediment samples for analysis of semivolatile compounds should be
collected in 240-mL (8-oz) or larger, wide-mouth glass jars with PTFE-lined
screw lids. The container should be washed with detergent, rinsed twice
with tap water, rinsed at least twice with distilled water, rinsed with
acetone, and, finally, rinsed with high-purity methylene chloride. The
PTFE liner of the lid should be similarly cleaned. Firing of the glass
jar at 450° C may be substituted for the final solvent rinse only if precautions
are taken to avoid contamination as the container is dried and cooled.
Container blanks should be analyzed periodically to assess container contamina-
tion. A sample weighing at least 200 g (wet weight; i.e., approximately three-
quarters of the volume of a wide-mouth 8-oz jar) should provide enough material
for a full analysis and all required QC analyses. The remaining headspace
(i.e., approximately one-quarter of the 8-oz jar) will facilitate mixing
of the sample in the laboratory and enable the sample to be frozen if necessary.
11
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN SEDIMENTS
DECEMBER 1986
If analysis of volatile compounds is required, two separate 40-mL
glass containers should be filled and no headspace should be left. It
is recommended that two samples are collected to ensure that an acceptable
sample with no headspace is submitted to the laboratory for analysis.
The containers, screw caps, and cap septa (silicone vapor barriers) should
be washed with detergent, rinsed once with tap water, rinsed at least twice
with distilled water, and dried at >105° C. A solvent rinse is avoided
because it may interfere with the analysis. Samples for analyses of volatile
organic compounds should be taken directly from single grab samples prior
to any subsampling for other analyses. Many of the volatile compounds
of interest could be lost while compositing. Sample containers can be
filled without leaving headspace in one of two ways, depending on the water
content of the sediment. If there is adequate water in the sediment, the
vial should be filled to overflowing so that a convex meniscus forms at
the top. If there is little or no water in the sediment, jars should be
filled as tightly as possible, eliminating obvious air pockets. With the
liner's PTFE side down, the cap should be carefully placed on the opening
of the vial, displacing any excess material. Once sealed, the bottle should
be inverted to verify the seal by demonstrating the absence of air bubbles.
Samples collected in this manner cannot be frozen.
Storage
Samples should be stored in the dark at 4° C, on ice, or frozen (except
volatiles, as described above) until extraction. Analyses for volatile
compounds should be performed within 14 days of collection as recommended
by U.S. EPA (1984a). Freezing is preferred for samples to be analyzed
for semi volatile organic compounds if the analysis will not be performed
within the recommended 7-day holding time. Care must be taken with frozen
samples to prevent container breakage by leaving headspace for the interstitial
water to expand and by freezing containers at an angle rather than in an
12
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN SEDIMENTS
DECEMBER 1986
upright position. Appropriate holding times have not been established
for frozen sediments. For sediment samples held at -20° C, workshop partici-
pants discussed a general guideline of 6-12 mo. In an unpublished study
at the University of Washington School of'Oceanography, replicate samples
of sediment homogenates analyzed for hydrocarbons were frozen for as long
as 5 yr. No significant differences in hydrocarbon concentrations were
found. Reproducible results have been reported by the Northwest NOAA/
National Marine Fisheries Service (NMFS) for hydrocarbons and PCBs in frozen
sediment homogenates of Duwamish River reference sediments analyzed over
a period of >1 yr. As part of the Duwamish Head Baseline Study, Metro
is examining the degradation of trace organic compounds and metals in archived
sediment samples. Samples have been stored in PTFE-lined sealed double
bags at -40° C. The samples were analyzed for organic compounds initially
6 and 18 mo later, but the results have not been compliled.
LABORATORY ANALYTICAL PROCEDURES
Volatiles
The routine U.S. EPA CLP heated purge-and-trap procedure (Appendix B)
is cost-effective as a screening procedure and can attain the 10-20 ppb
[dry weight (DW)] detection limits that are considered appropriate for
low-level analyses.
The vacuum extraction/purge-and-trap technique described by Hiatt
(1981) and Hiatt and Jones (1984) has also produced acceptable results.
Recovery of several compounds was better with the vacuum extraction technique
than with the U.S. EPA CLP heated purge-and-trap technique. The vacuum
extraction technique is under consideration for U.S. EPA validation as
a standard method and has been recommended for U.S. EPA-approved marine
monitoring programs (Tetra Tech 1986b).
13
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN SEDIMENTS
DECEMBER 1986
Semivolatiles (Extractable Organic Compounds)
Screening Level Analyses—
Laboratory procedural requirements are a function of required detection
limits. For semivolatile compounds, two levels of detection limits (for
screening analyses and sensitive analyses) were recommended by workshop
participants as appropriate for most anticipated project goals. Recommended
screening levels are 500-1,000 ppb DW for acid and neutral compounds and
15-300 ppb DW for pesticides and PCBs. Recommended sensitive (low-level)
detection limits are 1-50 ppb DW for acid and neutral compounds and 0.1-15 ppb
DW for pesticides and PCBs.
Appropriate procedures for screening level analyses are described in the
U.S. EPA CLP protocol (Appendix A). These procedures entail use of a 30-g (wet
weight) sample, optional cleanup of extracts by gel permeation chromatography
(GPC) prior to gas chromatography/mass spectroscopy (GC/MS) analyses, and
required alumina column cleanup and optional elemental sulfur removal prior to
GC/electron capture detection (GC/ECD) analysis of pesticides and PCBs.
Low-Level Analyses—
Sensitive (low-level) analyses of sediments require multiple extract
cleanup steps to remove biological macromolecules, elemental sulfur, and
unresolved complex mixtures of nonpolar compounds. The steps described
below apply to low-level, full-scan (i.e., acid/neutral compounds) analysis
of sediment samples.
Sample Size—
A sample size of approximately 50-100 g (wet weight) with a minimum
final dilution volume of 0.5 mL is considered adequate to attain the low-
14
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN SEDIMENTS
DECEMBER 1986
level detection limits for full-scan analyses of semivolatile organic
compounds. Smaller sample sizes can yield similar detection limits providing
that the final extract volume can be reduced proportionately without loss of
analytes. Attainable detection limits will be adversely affected by signifi-
cantly smaller sample sizes when all acid/neutral analytes must be determined
in a single instrumental analysis. If final dilution volumes of less than
approximately 0.4 mL are used, the analyst should verify that the solvent
reduction technique does not result in co-distillation of target compounds.
Sample Drying—
Drying of large sediment samples by the addition of sodium sulfate
is considered inefficient because it greatly increases sample volume and
is of limited effectiveness in the presence of polar solvents (e.g., methanol).
The following drying procedures were discussed at the workshops:
• Sample drying with sodium sulfate is feasible for samples
of approximately 10 g (after overlying and interstitial
water from the sample is centrifuged and decanted). The
dried sediment/sodium sulfate mixture is then extracted
with pure methylene chloride and the extract is purified
without a separatory funnel step (a procedure similar to
U.S. EPA CLP developed at the Northwest NOAA/NMFS).
• Alternatively, wet samples are mixed (at room temperature)
with methanol, the slurry is centrifuged, and the water/methanol
supernatant is decanted and saved for later extraction.
The sample is then more exhaustively extracted with less
polar solvents (e.g., dichloromethane or equivalent). The
water/methanol and solvent extracts are combined and subjected
to separatory funnel partitioning (a laboratory procedure
used by Battelle Northwest).
15
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN SEDIMENTS
DECEMBER 1986
For some projects, the concentration of analytes in the interstitial
water associated with the solid phase may be of interest (e.g., oiled sedi-
ments). Decanting, centri fugation, and discarding of this water may bias
the results. If concentrations in the whole sample (i.e., including inter-
stitial water) are of interest, the decanted water should be fractionated
and the resulting extract added to the sediment extract. The desired procedure
should be specified in the statement of work to the laboratory to ensure
the generation of data appropriate to project goals. Similarly, laboratories
should be notified that sieving or other alterations should not be performed
prior to subsampling for analysis. All decisions on what constitutes the
sample should be made in the field by project personnel.
Extraction—
Several extraction procedures are routinely used in Puget Sound analyses.
All procedures are acceptable pending demonstration that the compounds
of interest can be recovered within specified QA/QC guidelines when spiked
in method blanks or sample replicates or when analyzed in a reference material.
The following procedures were discussed at the workshops:
• A shaker or roller extraction is routinely used by several
regional laboratories (e.g., Metro and Northwest NOAA/NMFS)
and is considered to be as effective as exhaustive Soxhlet
extraction. The roller technique used by NOAA/NMFS National
Status and Trends Program uses pure methylene chloride as
the solvent when water is removed as described in "Sample
Drying," above.
• Soxhlet extraction (e.g., U.S. EPA Method 3550) is also
used by a number of regional laboratories (e.g., Northwest
NOAA/Pacific Marine Environmental Laboratory and University
16
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN SEDIMENTS
DECEMBER 1986
of Washington). Various effective solvent mixtures are
methylene chloride/methanol (2:1), methylene chloride/methanol
(9:1, azeotropic), and benzene/methanol (3:2, azeotropic).
(Benzene is an excellent cosolvent with methanol but its
use is not encouraged in government laboratories for health
reasons.) The methanol is slurn'ed with the sediment in
the extraction thimble prior to extraction. The sediments
are stirred several times during extraction to prevent solvent
channeling. An initial brief Soxhlet extraction with pure
methanol is not considered necessary or even desirable because
problems with solvent superheating (i.e., "bumping") have
been reported. However, water should be removed as described
in the "Sample Drying" section in advance of Soxhlet extraction.
• The U.S. EPA CLP has documented and validated the use and
performance of sonication with specified solvent mixtures
and a 30-g subsample of sediment (see Appendix A). Sample
sizes of 100 g may be too large for efficient sonication.
Hence, other adjustments in the procedure (e.g., reducing
the final dilution volume) may be required to attain low
detection limits using this extraction procedure. Each
laboratory must validate its sonication technique because
extraction efficiency can vary considerably as a function
of probe size and shape, the power and efficiency of the
probe motion, and the solvents used.
Drying of Extract—
Separatory funnel partitioning can be used to remove water (and methanol
or acetone) from organic extracts. The sample pH must be carefully controlled
(maintain pH<2) to ensure the recovery of polar analytes. As necessary,
distilled water acidified with a non-oxidizing acid (i.e., hydrochloric acid)
17
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN SEDIMENTS
DECEMBER 1986
should be used to adjust pH. Sulfuric acid (an oxidizing acid) has been
shown to cause losses of some target compounds in sample extracts and should
not be used. The pH should be monitored in the separator^ funnel because
the natural acidity/basicity of sample extracts can vary. Sodium sulfate
columns are typically used to dry extracts following partitioning.
Extract Concentration—
Kuderna-Dani sh and rotary evaporation (to a minimum of 2 mL) are both
considered acceptable methods of sample concentration.
Extract Cleanup—
Sediments - Elemental Sulfur Removal—Elemental sulfur removal with
metallic mercury can be accomplished within minutes if vigorous mechanical
agitation is used (e.g., Vortex Genie). However, endrin aldehyde is reported
to be susceptible to degradation in the presence of mercury; its recovery
should be documented when quantitative analyses of this pesticide are required.
Activated copper columns are also considered acceptable by several
workshop participants for elemental sulfur removal. Disadvantages of this
method include confirmed losses of mercaptans and possible losses of hepta-
chlor. Recovery of these compounds should be documented when quantitative
results are required.
Gel Permeation Chromatography (GPC)—The U.S. EPA CLP guidelines (see
Appendix A) for GPC using a methylene chloride solvent system and Bio Beads (TM)
SX-3 are acceptable. An S-X2 column used with a methylene chloride/pentane
(1:1) solvent system is an acceptable alternative. The extract volume
appropriate for GPC application is 2-4 mL. Calibration of columns for
optimal separation and recovery of analytes is required.
18
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN SEDIMENTS
DECEMBER 1986
Sephadex (TM) is used in combination with normal phase column Chroma-
tography to isolate several analytical fractions. There are published
data on use of a Sephadex (TM) LH-20 column with tertiary (cyclohexane/
methanol/methylene chloride) solvent system for aromatic hydrocarbons,
PCBs, hexachlorobenzene (HCB-partially recovered in a silica gel fraction),
chlorinated butadienes (CBD), and 1,2-dichlorobenzene (MacLeod et al. 1984).
Calibration of the columns for all analytes is required.
Column Chromatography Cleanup (for GC/MS fractions)—Reverse-phase column
cleanup with bonded octadecyl columns is one recommended technique to reduce
the amounts of nonpolar, chromatographically unresolvable compounds in
sediment extracts for the analysis of acid/base/neutral compounds (Tetra
Tech 1985b). The U.S. EPA Marine Division, Environmental Research Laboratory,
Newport, OR has also conducted tests with bonded octadecyl columns for
sediments, and in conjunction with an aminopropyl column for tissues. Recovery
data for neutral compounds using these columns are available (Ozretich
and Schroeder 1985). Precautions are required to avoid incomplete solvent
exchange (from methylene chloride to methanol) and column overloading.
Use of a prior gel permeation step has been effective in preventing overload
problems (Tetra Tech 1985b). A cleanup step is recommended if the sample
is contaminated with petroleum and the removal of paraffinic hydrocarbon
constituents that contribute to the unresolved complex mixture is necessary.
Solvent exchanges with acetone instead of methanol as the elution
solvent may be easier to accomplish, but comparative tests of the separation
efficiency of acetone and methanol on C\Q columns would need to be conducted
before recommendations for use can be made.
Normal Phase Column Chromatography—Metro (1981) and others have success-
fully used normal phase high performance liquid Chromatography (HPLC) to
generate separate fractions containing neutral, base, and acid compounds
for analysis by GC/MS. Normal phase separations on activated silica, cesium
19
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN SEDIMENTS
DECEMBER 1986
silicate (used to retain organic acids), and aminopropyl phases are possible
alternatives to reverse phase extract cleanup. It has been reported that
strong adsorption of un-derivi tized organic acids to activated silica can
be overcome after the silica is penetrated by heating it to 700° C for
18 h and conditioning with methylene chloride prior to use. Northwest
NOAA/NMFS reports testing this technique with substituted and unsubstituted
phenols but results have not been published.
Alumina Column Chromatograplry (for GC/ECD interferences)—Alumina column
cleanup for PCBs and pesticides with hexane as an eluting solvent is used in the
U.S. EPA CLP (see Appendix A). Twenty percent methylene chloride/hexane
may be suitable as an eluting solvent but there may be problems recovering
endrin aldehyde. Any solvent system must be calibrated for the compounds
of interest prior to conducting analyses.
20
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN TISSUE
DECEMBER 1986
ORGANIC COMPOUNDS IN TISSUE
USES AND LIMITATIONS
The various techniques described in this section are suitable for
the analysis of some or all of the semivolatile and volatile organic priority
pollutants in tissues. Based on discussion at the Puget Sound workshops,
three compound groups are not suitable target compounds for routine full-
scan analysis: organic bases, halogenated ethers, and hexachlorocyclo-
pentadiene. These compound groups were excluded because they are rarely
detected in Puget Sound biota or they present exceptional analytical difficul-
ties. However, the organic base N-nitrosodiphenylamine is of potential
concern because it has been detected in several sediment samples and thus
could accumulate in tissue.
Additional (i.e., not priority pollutant) analytes recommended for
potential analysis include coprostanol, 9(H)-carbazole, polychlorinated
butadienes, and polychlorinated styrenes. Some of these compounds may
require specific analytical methods for detection at low parts-per-bil 1 ion
concentrations. Thus, program coordinators should confer with technical
experts to evaluate project objectives and the cost-effectiveness of including
these compounds in an environmental study.
As noted in the introduction, a single step-by-step laboratory procedure
has not been recommended. For this reason, considerable weight is placed
on the recommended QA/QC procedures to establish a basis for comparing
data. To the extent that QA/QC procedures cannot completely assess performance
(e.g., no SRM are available that contain all potential analytes), different
laboratory procedures cannot be verified as absolutely comparable. Results
21
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN TISSUE
DECEMBER 1986
from the repeated analysis of SRM containing at least some of the analytes
of concern are useful for assessing inter- and intralaboratory precision
and accuracy.
The emphasis of this protocol is on full-scan analyses rather than
dedicated analysis for one group of related compounds (e.g., hydrocarbons).
However, the QA/QC procedures are applicable to most types of analyses.
Improved detection limits and performance may be attained by focusing on certain
compounds or related compound classes, but at the cost of considerably more
laboratory effort. These focused analyses may be preferable when analyzing
for a limited number of analytes whose expected concentrations are low.
SAMPLE PREPARATION AND FIELD PROCEDURES
Collection
In the field, sources of contamination include sampling gear, grease
from ship winches or cables, ship engine exhaust, dust, and ice used for
cooling. Efforts should be made to minimize handling and to avoid sources
of contamination. For example, to avoid contamination from ice, the whole
samples (e.g., molluscs in shell, whole fish) should be wrapped in aluminum
foil, placed in watertight plastic bags, and immediately cooled in a covered
ice chest. Many sources of contamination can be avoided by resecting (i.e.,
surgically removing) tissue in a controlled environment (e.g., a laboratory).
Organisms should not be frozen prior to resection if analyses will be conducted
on only selected tissues (e.g., internal organs) because freezing may cause
internal organs to rupture and contaminate other tissue. If organisms
are eviscerated on board the survey vessel, the remaining tissue may be
wrapped as described above and frozen. Tissue sample collection and preparation
requirements are summarized in Table 4.
22
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN TISSUE
DECEMBER 1986
Processing
To avoid cross-contamination, all equipment used in sample handling
should be thoroughly cleaned before each sample is processed. All instruments
must be of a material that can be easily cleaned (e.g., stainless steel,
anodized aluminum, or borosilicate glass). Before the next sample is processed,
instruments should be washed with a detergent solution, rinsed with tap
water, soaked in high-purity acetone or methylene chloride, and finally
rinsed with distilled water. Work surfaces should be cleaned with 95 percent
ethanol and allowed to dry completely.
The removal of biological tissues should be carried out by or under
the supervision of an experienced biologist. Tissue should be removed
with clean stainless steel or quartz instruments (except for external sur-
faces). The specimens should come into contact with precleaned glass surfaces
only. Polypropylene and polyethylene (plastic) surfaces and implements
are a potential source of contamination and should not be used. To control
contamination when resecting tissue, technicians should use separate sets
of utensils for removing outer tissue and for resecting tissue for analysis.
For fish samples, special care must be taken to avoid contaminating
targeted tissues (especially muscle) with slime and/or adhering sediment
from the fish exterior (skin) during resection. The incision "troughs"
are subject to such contamination and should not be included in the sample.
In the case of muscle, a "core" of tissue is taken from within the area
bordered by the incision troughs, without contacting them. Unless specifically
sought as a sample, the dark muscle tissue that may exist in the vicinity
of the lateral line should not be mixed with the light muscle tissue that
constitutes the rest of the muscle tissue mass. This dark tissue is not
typically consumed by humans and because of a higher lipid (fat) content,
may contain concentrations of organic chemicals at levels greater than
the remaining muscle tissue.
23
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN TISSUE
DECEMBER 1986
The tissue sample should be placed in a clean glass or PTFE container
that has been washed with detergent, rinsed at least once with tap water,
rinsed at least twice with distilled water, rinsed with acetone, and, finally,
rinsed with high-purity methylene chloride. Firing of the glass jar at
450° C may be substituted for the final solvent rinse only if precautions
are taken to avoid contamination as the container is dried and cooled.
The solvent rinse could contaminate sample jars used for volatile organics
analysis of tissues. Instead, sample jars for volatiles analysis should
be heated to >105° C as a final preparation step.
Storage
Recommended holding times for frozen tissue samples have not been
established by U.S. EPA, but a maximum 6-mo to 1-yr holding time similar
to the sediment holding times is recommended for Puget Sound studies.
(For extended sample storage, precautions should be taken to prevent
desiccation). NBS is testing the effects of long-term storage of tissues
at temperatures of liquid nitrogen (-120° to -190° C). At a minimum, the
samples should be kept frozen at -20° C until extraction. This will slow
biological decomposition of the sample and decrease loss of moisture.
Liquid associated with the sample when thawed must be maintained as part
of the sample.
LABORATORY ANALYTICAL PROCEDURES
Some of the analytical procedures are identical to those discussed
for sediment matrices and have been incorporated by reference where appropriate.
24
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN TISSUE
DECEMBER 1986
Volatiles
Analysis of tissue samples for volatiles can be performed by using
the U.S. EPA CLP procedure with a purge-and-trap apparatus (Appendix B).
The use of a vacuum extraction device (Hiatt 1981; Hiatt and Jones 1984)
may improve recovery of selected compounds. A brief discussion of the
two techniques is provided in the sediment section.
Semivolatiles (Extractable Organic Compounds)
Laboratory procedural requirements for semivolatile organic compounds
are a function of required detection limits. Detection limits in tissue
samples were recommended after considering technical constraints and concentra-
tions that may result in potential human health effects (from the ingestion
of edible tissues).
Screening level detection limits similar to those recommended for
sediments were not considered appropriate for tissue analysis because they
would not allow for detection of contaminants at levels that may be estimated
to pose significant human health risks. Sensitive analyses should have
lower limits of detection of 10-20 ppb (wet weight) for acid and neutral
compounds and 0.1-20 ppb for pesticides and PCBs. The definition of lower
limits of detection is given below in "DATA REPORTING REQUIREMENTS." Tissue
extracts contain high concentrations of lipids and require removal of biological
macromolecules (e.g., by GPC) prior to analysis.
Sample Size—
A laboratory sample of approximately 30 g (wet weight) is adequate
to attain the recommended detection limits. A smaller sample size may
adversely affect detection limits. At least 60 g (wet weight) is reconmended
25
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN TISSUE
DECEMBER 1986
for samples that must be analyzed in duplicate. Note that tissue detection
limits in this document are listed on a wet weight basis rather than dry
weight (as with sediment detection limits).
Sample Drying—
Sodium sulfate is frequently used for drying of tissue samples. It
also helps macerate the tissue and produces a paste that facilitates extrac-
tion. Cleaned sand can also be used for maceration.
Extraction—
The following extraction procedures were discussed at the workshops:
• The Soxhlet method described for sediment samples is also
appropriate for biological tissue samples.
0 Grinding and homogenization (e.g., with a Tekmar Tissuemizer
or Brinkman Polytron) is recommended. Son i cat ion of tissue
samples has been found to produce emulsions (Ecology/Manchester
laboratory), and is not recommended.
0 Hydrolytic digestion with hydrochloric acid has been found
to double recoveries of phenolic compounds from tissue matrices.
The enhanced recovery is thought to result not only from
pH-induced changes in the solvent solubility of phenols,
but also from the hydrolytic degradation and physical breakdown
of the tissue matrix (University of Washington, Environmental
Health Department laboratory). This procedure has the greatest
potential applicability when a focused analysis for phenols
is desired. It is possible that the vigorous digestion
procedure could hydrolyze compounds that had been biologically
26
-------
ORGANIC COMPOUNDS
ORGANIC COMPOUNDS IN TISSUE
DECEMBER 1986
altered back to phenols. The use of this procedure could
limit the detection of compounds with important biological
roles (e.g., detoxification).
When analysis for analytes in addition to phenols is of importance, hydrolytic
digestion with hydrochloric acid may require an initial extraction at neutral
pH. For example, numerous chlorinated pesticides would be degraded by
strong acid refluxing. A less extreme acidification technique proposed
for Soxhlet extraction is adding acetic acid to the extraction thimble
before starting the extraction. Acetic acid would establish a low extraction
pH but, unlike hydrochloric acid, would not hydrolyze the tissue matrix
or degrade chlorinated pesticides.
Drying of Extract—
See Sediment section.
Extract Concentration—
See Sediment section.
Extract Cleanup—
GPC cleanup is always required for extracts to be analyzed by GC/MS.
Extracts to be analyzed by GC/ECD for pesticides and PCBs also require
preparative chromatography (e.g., HPLC or column chromatography).
Analyses by GC/FID are strongly affected by fatty acids and related
natural interferents unless the extracts are subjected to additional cleanup
[e.g., using Sephadex (TM) LH-20 chromatography]. The removal of these
interferents may also result in the loss of some acid/neutral compounds
of interest.
27
-------
ORGANIC COMPOUNDS
INSTRUMENTAL PROCEDURES
DECEMBER 1986
INSTRUMENTAL PROCEDURES
Standard U.S. EPA CLP procedures (see Appendix D) for calibration
and quantification of analytes by GC/MS.are considered appropriate for
the analysis of organic compounds. Legally defensible data (e.g., for
enforcement action) require GC/MS confirmation for the analysis of U.S. EPA
acid/neutral priority pollutants. This requirement stems from the need
to document and confirm the presence or absence of co-eluting interferences
for each target compound in each sample. Such confirmation is not possible
by alternative detection systems (e.g., GC/FID), although it can be approached
by requiring verification of results by replicate analyses on gas chroma-
tographic columns of differing polarity. For PCBs and pesticides, legally
defensible data can be obtained with capillary column GC/ECD, which can
achieve low detection limits for these compound groups. High concentrations
detected by GC/ECD should be confirmed by GC/MS to prevent reporting of
"false positives" (i.e., compounds thought to be present but actually absent).
Enforcement work requires a minimum of dual column verification of PCBs
and pesticides by GC/ECD.
For other data uses (e.g., research studies of pollutant fates, transport,
and biological effects), acid and neutral compounds can be quantified by
GC/FID or GC/ECD with GC/MS confirmation for at least a portion of the
samples. Analytes co-eluting with interferents cannot be quantified by
GC/FID. Use of isotopically labeled surrogate spikes is limited when analyses
are conducted by GC/FID because only a few labeled compounds can be chromato-
graphically resolved for proper quantification.
28
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
QA/QC PROCEDURES AND REQUIREMENTS
QA/QC requirements are the foundation of this protocol because they
provide information necessary to assess the comparability of data generated
by different laboratories or different analytical procedures. The following
QA/QC variables are discussed in the order noted:
• Surrogate spike compounds (used to evaluate the analytical
recovery of each sample)
• Injection internal standards (used in the quantification
of samples and added immediately prior to instrumental analysis)
• Method blanks (used to evaluate possible sources of laboratory
contamination)
• Standard reference materials (used to provide an evaluation
of laboratory accuracy)
• Matrix spikes (used to evaluate the effect of sample matrix
on the compound of interest)
• Method spikes (used as a procedural check to eliminate sample
matrix interferences)
• Analytical replicates (used to evaluate precision of the
analytical method and instrumentation)
• Field replicates (used to evaluate total precision)
29
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
• Initial and ongoing calibrations (used to establish and
verify the quantification technique).
Data for all QA/QC variables should be submitted by the laboratory as part
of the data package. Program managers and project coordinators should
verify that requested QA/QC data are included in the data package as supporting
information for the summary data, and may review key QA/QC data (e.g.,
analytical replicate data or surrogate spike recoveries). Acceptable limits
for these variables are discussed in the following sections and summarized
in Tables 3 and 5. A detailed QA/QC review of the entire data package
(especially original quantification reports and standard calibration data)
should be conducted by a technical expert.
Screening level analyses (see Table 2) should be conducted according
to the QA/QC requirements of the most recent U.S. EPA CLP program document
(provided in Appendix A). The guidance provided in this section is applicable
to low parts-per-billion analyses of both sediment and tissue analyses
unless 'specifically noted. Warning limits are numerical criteria that
serve as flags to data QA reviewers and data users. When a warning limit
is exceeded, the laboratory is not obligated to halt analyses, but the
reported data may be qualified during subsequent QA/QC review. Control
limits are numerical criteria that, when exceeded, require specific action
by the laboratory before data may be reported. Control limits are intended
to serve as contractual controls on laboratory performance. The warning
and control limits are summarized in Table 5.
SURROGATE SPIKE COMPOUNDS (RECOVERY INTERNAL STANDARDS)
Surrogate spike compounds are added to each sample prior to extraction
or purging. Because surrogate spikes are the only means of checking method
performance on a sample-by-sample basis, they are always required.
30
-------
TABLE 5. SUMMARY OF WARNING AND CONTROL
LIMITS FOR QUALITY CONTROL SAMPLE
Analysis Type3
Recommended
Warning Limit
Recommended
Control Limit
Surrogate spikes
Method Blank
Phthalate, Acetone
Other Organic
Compounds
Standard Reference
Materials
Matrix spikes
Spiked Method Blanks
Analytical Replicates
Field Replicates
Ongoing Calibration
10 percent recovery
30 percent of the analyte
1 ug total or 5 percent
of the analyte
95 percent confidence
interval
(50-65 percent recovery)
(50-65 percent recovery)
See Appendix C
Table C-l
(50 percent recovery)
5 ug total or 50 percent
of the analyte
2.5 ug total or 5 percent
of the analyte
95 percent confidence
interval for Certified
Reference Material
(50 percent recovery)
(50 percent recovery)
+ 100 percent coefficient
of variation
25 percent of initial
calibration
a The definition of each quality control sample is given in the "QA/QC Procedures
and Requirements" section.
b Values in parenthesis are limits set for analyses with fewer surrogate compounds
than those used for the isotope dilution technique.
31
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
Frequency
Surrogate spikes should be added to each sample.
Compound Type
A minimum of five surrogate spikes must be added to each sample (three
neutral and two acid compounds) when the more extensive GC/MS isotope dilution
technique (e.g., U.S. EPA Method 1625B) is not used. These surrogate spikes
should cover a wide elution range and include one of the more volatile
compounds (e.g., ds-phenol) as well as a degradable PAH (e.g., di2-perylene
or di2-benzo(a)pyrene). Isotopically labeled analogs of the analytes are
'strongly recommended as surrogate spikes (over 50 isotopically labeled
compounds are commercially available).
Isotopically labeled analogs of the U.S. EPA priority pollutants have
been shown to behave like the unlabeled priority pollutant compounds in
several method tests. An isotope dilution technique (i.e., U.S. EPA Method
1625, Revision B) can be used to correct for potential losses of analytes
during the cleanup and analysis of extracts. The technique does not necessarily
account for the efficiency of extraction (i.e., the amount of chemicals
actually recovered from the sample matrix during sample processing) because
some pollutants may be more tightly bound to particles in the sample than
are the surrogate compounds spiked into the sample. Recovery corrections
made using the isotope dilution technique enables a correction for losses
of compounds after extraction. There is no completely accurate way to
account for extraction efficiency. Hence, exhaustive solvent extraction
procedures are considered essential to optimize extraction efficiency.
At least one pesticide/PCB surrogate spike is required as a check on
recovery. This compound must be well-resolved, not co-elute with any PCB
32
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
or pesticide analytes, and behave similarly to the analytes. This surrogate
will likely not be a perfect PCB/pesticide analog. Possible standards
are dibutylchlorendate (used in the U.S. EPA CLP), isodrin (endo-endo isomer
of aldrin), and dibromooctofluorobiphenyl (used routinely by NOAA/NMFS).
Three surrogate spikes are required for the analysis of volatile compounds.
Limitations
Recovery corrections for full-scan analyses by GC/MS should be made
only if the isotope dilution technique is used with all the available (i.e.,
approximately 50) stable labeled acid/neutral surrogate spikes.
If fewer than five acid/neutral surrogate spikes (i.e., recovery standards)
are used for full-scan analyses, recoveries should be reported, but no
corrections should be applied to sample results. However, when analyzing
for a specific compound class (e.g., hydrocarbons), recovery corrections
may be applied based on a limited number of labeled analogs if similar
behavior of all target compounds has been demonstrated (MacLeod et al. 1984).
Warning and Control Limits
If all the available labeled acid/neutral compounds in the isotope
dilution technique are not used, much more significance is put on the results
of a limited number of labeled surrogate spikes. Therefore a control limit
of at least 50 percent recovery is reasonable. When all target compounds
are associated with labeled analogs, the isotope dilution technique is
self-validating for each compound in each sample. The major requirement
of this technique is adequate recovery to yield a reliable signal for analysis
(e.g., per requirements of the extensively reviewed U.S. EPA dioxin/furan
program). A large number of labeled surrogate spikes gives more confidence
in the results over a wide range of recoveries. A warning limit of less
than 10 percent recovery is reasonable for these analyses.
33
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
Corrective Action
All data from the isotope dilution technique reported with recoveries
below 10 percent should be qualified as estimates because of some concern
that the large correction factor has introduced uncertainty. If analytes
are detected in the sample but the labeled surrogate spikes have a recovery
of less than 1 percent, the concentration of the analyte should be calculated
using a 1 percent recovery correction. The data should then be qualified
as greater than the reported value. Data generated without using the complete
isotope dilution technique (e.g., with only five of the acid/neutral surrogate
spikes) should be qualified as estimated values, if surrogate recoveries
are less than 50 percent.
Report
A summary of percent recovery values in sample and method blanks for
all surrogate compounds analyzed should accompany the data.
INJECTION INTERNAL STANDARDS
Injection internal standards are added just prior to injection to
enable optimal quantification, particularly of complex extracts subject
to retention time shifts relative to the analysis of standards.
Frequency
An injection internal standard should be added to each sample.
Compound Type
Injection internal standards are essential if the actual recovery
of standards added prior to extraction is to be calculated. These injection
34
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
internal standards can be used to detect and correct for problems in the
GC injection port or other parts of the instrument. The injection internal
standards should include at least one early eluting compound (e.g., hexamethyl-
benzene or 2,2'-difluorobiphenyl) and one late eluting compound (e.g.,
5-alpha-cholestane or an isotopically labeled compound not already used
as a surrogate spike).
More than two injection internal standards are not required, because
a minimum of three neutral and two acid surrogate spikes are already required
in each sample. However, if the isotope dilution technique is not used
to calculate compound concentrations, additional injection internal standards
should be used to assure proper quantification and correction for differential
GC loading of a range of analytes in the analysis of complex extracts.
The U.S. EPA CLP recommends three to six injection internal standards when
the internal standard quantification technique is used.
Corrective Action
The analyst should monitor injection internal standard retention times
and recoveries to determine if instrument maintenance or repair is indicated
(e.g., to replace injection port septum or columns) or if changes in analytical
procedures are indicated (e.g., change ramping specifications). Corrective
action is often initiated based on the experience of the analyst and not
because warning limits were exceeded.
Report
All routine or corrective maintenance procedures should be noted in
either a logbook kept with the instrument or in the analyst's daily operations
logbook. Any instrument problems that may have affected the data collected
or resulted in the reanalysis of the sample should be documented in the
35
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
analyst's logbook and on the raw data report. Justification for reanalysis
should be discussed in the cover letter accompanying the data.
METHOD BLANKS
Method blanks are used to assess laboratory contamination during all
stages of the preparation and analysis of sample extracts.
Frequency
At a minimum, one method blank should be run for every extraction
batch (e.g., 10-20 samples) or for every 12-h shift, whichever is more
frequent.
Phthalate..Methylene Chloride, and Acetone Warning Limit
The warning limit for phthalate, methylene chloride, and acetone is
30 percent of the concentration of analyte in the sample. Ideally, if
these contaminants are detected at all in method blanks, sufficient method
blanks should be run to determine a confidence interval for laboratory
contamination. Environmental analytical chemists have not universally agreed
upon a convention for determining and reporting the effects of blank contam-
ination on sample results. However, Keith et al. (1983) define a "limit of
detection" as the lowest concentration level that can be determined to be
statistically different (e.g., three standard deviations for a 99 percent
confidence interval) from the blank. Detection of these compounds in a sample
should be reported at the concentration above the upper confidence interval.
Phthalate. Methylene Chloride, and Acetone Control Limit
The control limit for phthalate, methylene chloride, and acetone is
5 ug total in the blank after correction for recovery (this control limit
36
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
corresponds to approximately 100 ppb assuming a 50-g dry weight sample
and a final volume of 0.5 mL) or 50 percent of the amount of analyte in
the sample, whichever is greater. This control limit is less strict than
that for other contaminants because of the recognized difficulty in controlling
for phthalates and common laboratory solvents. Determination of a confidence
interval for laboratory contamination will enable a more accurate quantifi-
cation of results for these contaminants.
Other Contaminants Warning Limit
The warning limit for other contaminants is 1 ug total in the blank
after correction or recovery (this warning limit corresponds to approximately
20 ppb dry weight in a sediment sample) or 5 percent of the amount of analyte
in the samples, whichever is greater.
Other Contaminants Control Limit
The control limit for the other contaminants is 2.5 ug total in the
blank after correction for recovery in method blank (this control limit
corresponds to approximately 50 ppb assuming a 50-g dry weight sample and
a final dilution volume of 0.5 mL), or 5 percent of the total amount in
the least contaminated sample, whichever is greater. See discussion of
confidence intervals in the "Corrective Action" section.
Corrective Action
If any warning limit is exceeded, likely sources of contamination
should be discussed in the cover letter of the data report. If the control
limit for phthalates, methylene chloride, or acetone is exceeded, the source
of contamination should be identified and controlled. If control limits
for other contaminants are exceeded, analysis should be halted until the
contaminant source is eliminated or greatly reduced.
37
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
Replicate method blank analyses should always be conducted when laboratory
contaminants other than phthalates, methylene chloride, and acetone are
detected. Replicate method blanks should be used to determine confidence
intervals for the observed contamination. Detection limits for analytes
in samples should be adjusted to account for the confidence intervals estab-
lished for laboratory contaminants in replicate method blank analyses (Keith
et al. 1983). Data may be qualified by data managers if warning limits
are exceeded, or rejected if the control limit is exceeded.
Report
Method blank results should be reported with the sample data, except
that the reporting units should be total ng/sample. Laboratories should
not blank-correct any data.
STANDARD REFERENCE MATERIALS (SRM)
Standard reference materials are the most useful of QC samples that
are used for assessing the accuracy of an analysis (the closeness of a
measurement to its true value). Other QC samples used to approximate measures
of laboratory accuracy are method spikes and matrix spikes. SRM have not
been readily available for marine sediments, especially for fresh-frozen
sediments. However, Northwest NOAA/NMFS has prepared a fresh-frozen marine
sediment sample (from Sequim Bay) spiked with PCBs, PAH, and selected pesticides
for use by U.S. EPA, NOAA, and other agencies and laboratories in Puget
Sound studies. This SRM is available from the U.S. EPA Office of Puget
Sound Estuary Program. Routine analysis of a regional SRM is recommended
to provide data for interlaboratory comparisons. Several tissue homogenate
SRM are already available (e.g., Mega Mussel, U.S. EPA, Environmental Research
Laboratory, Narragansett, RI).
38
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
Frequency
If five or fewer samples are submitted for analysis, one SRM is recom-
mended, at the discretion of the project coordinator. If analysis of an
available reference material is not included, a lower quality ranking may
be assigned to the data in regional databases. If 6-50 samples are submitted,
at least one SRM should be analyzed. For more than 50 samples, one SRM
should be analyzed for each 50 samples.
Warning Limits
The warning limit for SRM is the 95 percent confidence interval for
the certified values (if available). "Certified values" are those provided
for certified reference materials (CRM) through an agency validation testing
program (e.g., NBS). Reference materials provided by other sources may
be used to assess warning limits when a 95 percent confidence interval is
developed from multiple laboratory analyses of the reference material using
more than one analytical method.
Control Limits
Control limits are only appropriate for analysis of SRM that have
been certified. If more than two analytes fall outside the 95 percent
confidence interval, corrective action should be taken.
Corrective Action
It is recommended that the SRM, if available, be analyzed prior to
analysis of any samples. If values are outside the control limits, the
SRM should be reanalyzed to confirm the results. If the values are still
outside control limits in the repeat analysis, the samples may be analyzed
and reported with statements that describe the possible bias of the results
39
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
in the cover letter accompanying the data. Alternatively, the laboratory
may be required to repeat the analyses until control limits are met before
continuing with sample analyses. Determination of the appropriate corrective
action is the responsibility of the program manager or project coordinator
and should be specified a priori in the statement of work for the laboratory.
Corrective action requiring the laboratory to repeat analyses may
be appropriate only for reference materials that have certified values
(e.g., NBS certified or equivalent). NBS-certified SRM for organic compounds
(except PCBs) in sediments are not currently available.
Report
The laboratory should keep a running record of results obtained for
each analysis of an SRM. Observed results should be compared to the mean
provided by the originator of the SRM, the observed mean obtained from
repeated analyses by the laboratory, and acceptable range limits. Minimum
reporting of SRM results with laboratory data should include observed and
expected values and the acceptable range limits. The steps for corrective
action and observed bias relative to existing SRM values should be reported
and discussed in the cover letter.
MATRIX SPIKES
Matrix spike results are currently the most common form of recovery
data provided by laboratories, and are required by the U.S. EPA CLP protocol
for screening level analyses. Matrix spike results are of less value than
SRM results. The efficiency of the extraction of the compounds of interest
from the sample matrix is not accounted for in matrix spike results. Matrix
spike results are preferred as QC samples for low-level analyses only in
the absence of a suitable SRM. Matrix spikes should include a wide range
of representatives of analyte types. Spikes should be added at 1-5 times
40
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
the concentration of compounds in the sample. Use of the isotope dilution
technique precludes the need for additional QC matrix spike samples.
Frequency
Matrix spike samples.are not required when the isotope dilution technique
is used with labeled analogs for analytes. If five or fewer samples are
submitted to the laboratory, a minimum of one matrix spike is recommended,
at discretion of program manager or project coordinator. If 6-19 samples
are submitted, at least one matrix spike should be run. If 20 or more
samples are submitted, one matrix spike should be run for each 20 samples
(5 percent of total number of samples).
Warning and Control Limits
Matrix spikes are not required when the complete isotope dilution
technique is used. When the complete isotope dilution technique is not
used, a warning limit range of 50-65 percent recovery and a control limit
of 50 percent recovery for matrix spikes would not be unreasonable for
the reasons stated in the "SURROGATE SPIKES" section. Control limits for
matrix spikes recommended in the U.S. EPA CLP are provided in Appendix
C, Table C-l.
Corrective Action
Matrix spike results should be monitored during the analysis. If
percent recovery limits exceed the warning limit range of 50-65 percent,
the chromatograms and raw data quantitation reports should be reviewed.
If an explanation for a low percent recovery value is not discovered, a
continuing calibration check should be made to assure that the instrument
is responding within acceptable limits. Low matrix spike recoveries may
be a result of matrix interferences in the sample itself and further instrument
41
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
response checks may not be warranted. Analysis of samples should not resume
until acceptable instrument response has been verified.
Report
An explanation of low percent recovery values for matrix spike results
should be discussed in the cover letter accompanying the data package.
Corrective actions taken and verification of acceptable instrument response
should be included.
METHOD SPIKES
Method spikes (i,e., method blanks spiked with surrogate compounds
and analytes) are useful in verifying acceptable method performance prior
to routine analysis of samples (i.e., post extraction). Method spikes
do not take into account all possible sample matrix effects, but can be
used to identify basic problems in procedural steps. Method spikes can
also provide minimum recovery data when no suitable SRM is available or
when insufficient sample size exists for matrix spikes. Standard analytes
and surrogate spikes are added prior to extraction.
Frequency
A method spike should be analyzed before analysis of samples when
a method is used for the first time in a project and after any method modifi-
cation.
Warning and Control Limits
Warning and control limits that apply to surrogate spikes also apply
to method spikes. Tables C-5 and C-8 from U.S. EPA Methods 1624B and 1625B
are provided for comparison (see Appendix C).
42
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
Corrective Action
Analysis of samples should not begin until results are within control
limits.
Report
Detailed notes should be kept in a laboratory notebook. The notes
should discuss method spike results exceeding recommended limits, corrective
action, and verification of instrument response within acceptable criteria.
This information need not be included with data package results because
analysis cannot continue until all results are within control limits.
ANALYTICAL REPLICATES
Analytical replicates provide precision information on the actual
samples, or in the case of CLP, on matrix spike samples. Replicate analyses
are useful in assessing potential sample heterogeneity and matrix effects.
Frequency
If five or fewer samples are submitted for analysis, a minimum of
one replicate is recommended, at the discretion of program manager or project
coordinator. Without at least one replicate, a lower quality ranking may
be assigned to the data by data managers. If 6-19 samples are submitted,
one laboratory replicate should be analyzed. If at least 20 samples are
submitted, one blind (i.e., unknown to the laboratory) triplicate analysis
and additional duplicate analyses should be required, for a minimum of
replication 5 percent overall. One of the replicates in the triplicate
may be designated as the laboratory duplicate, but the third replicate
sample should be submitted as an unknown to the laboratory.
43
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
Warning and Control Limits
Based on data of Horwitz et al. (1980), who chart precision as a function
of concentration, a 30 percent coefficient of variation (c.v., a statistical
measure of precision) is expected for concentrations ranging between 1 and
50 ppb dry weight. Compound-specific advisory limits excerpted from the
U.S. EPA CLP are provided in Appendix C; Table C-l.
These advisory limits are recommended as warning limits (i.e., possibly
requiring qualification by data users). Extensive discussion of precision
requirements occurred at a final Puget Sound organics workshop of envi-
ronmental chemists and program administrators. It was decided that a laboratory
control limit of +100 percent coefficient of variation would be required (i.e.,
exceedance of the control limit would require automatic reanalysis to confirm
the results). Many compound analyses will be more precise. There was discus-
sion about easing the control limit if it is well under or well above some
regulatory guideline for acceptable contamination, and tightening the control
limit if it is close to the regulatory guideline. However, most data will
have multiple uses and adjustable limits will be difficult to apply as
a laboratory control.
Corrective Action
If results fall outside the warning limits, the reported data may
be qualified in regional databases as suitable for limited use pending
QA review of the probable laboratory or field sources of variation. If
results fall outside the control limit for more than two compounds, a replicate
analysis is required to confirm the problem before the data are reported.
Subsequent corrective action (i.e., if results continue to exceed control
limits) is at the discretion of the program manager or project coordinator
because matrix effects or laboratory error may be contributing factors. The
44
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
advice of a technical expert may be necessary to recommend appropriate
action.
A discussion of the results of duplicate sample analysis should include
probable sources of laboratory error, evidence of matrix effects, and an
assessment of natural sample variability. If data are to be qualified
on the basis of duplicate results, justification for assigning the data
qualifier should be provided.
FIELD REPLICATES
Field replicates are separate samples collected at the identical station
in the field and submitted for analysis. These QA samples are useful in
determining total variability (i.e., analytical plus field variability).
Frequency
The program manager or project coordinator will determine the frequency
with which field replicates are collected. Laboratory replicates must
be coordinated with field replicates so that sampling and analytical variability
will be measured for the same station.
Control Limits
Control limits are not appropriate when measuring environmental and
field sampling variability.
Corrective Action
No corrective action is recommended for field replicate analyses.
45
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
Report
If it is determined that variability observed in field duplicate results
can be partially explained by analytical or sampling variability, it should
be noted and discussed in a QA/QC evaluation of the data.
INITIAL CALIBRATION
Relative response factors (RRF) of analytes to standards are established by
calibration. The standards may be surrogate spikes or injection internal
standards.
Frequency
Equipment should be subject to initial calibration at the beginning
of the project before any samples are analyzed, after each major equipment
disruption, and when ongoing calibration does not meet criteria.
Number of Calibration Points
RRF must be determined for at least three concentration levels. The
standard concentrations tested should cover the range of expected sample
concentrations. One standard concentration for each target chemical must
be within 150 percent of the lower limit of detection.
Control Limit (Linearity)
For most compounds, control limits are established when the ratio
of the RRF to concentration does not differ by more than 20 percent coefficient
of variation over the range of concentrations tested (minimum of a three-
point calibration curve). Hence, the response of the instrument is assumed
to increase in direct proportion to the concentration of the sample when
46
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
less than a 20 percent deviation in the response is observed over the concen-
tration range represented by the calibration curve. A wider control limit
of 30 percent coefficient of variation is recommended for butylbenzyl phthalate,
bis 2-ethylhexyl phthalate, 4-nitrophenol, 2,4-nitrophenol, N-nitroso-di-n-
propylamine, and hexachlorocyclopentadiene (U.S. EPA CLP).
Corrective Action
If linearity is not established, the range for reporting data must
be reduced to within the observed linear range. Alternatively, the instrument
may be adjusted and linearity tests repeated before any samples are analyzed.
Failure to perform the proper calibration before analysis of samples will
be cause for omitting the data from regional databases.
Report
Initial calibration results within acceptable limits must be verified
prior to the analysis of samples. Summary data documenting initial calibration
and any episodes requiring recalibration and the corresponding recalibration
data should be included with analytical results.
ONGOING CALIBRATION
The ongoing calibration (single point) is used to check the assumption
that the original three-point calibration curve continues to be valid.
Frequency
For GC/MS or GC/FID analyses, calibration should be checked at the
beginning of each work shift, during the analyses at least once every 12 h
(or every 10-12 analyses, whichever is more frequent), and after last sample
of each work shift.
47
-------
ORGANIC COMPOUNDS
QA/QC PROCEDURES AND REQUIREMENTS
DECEMBER 1986
For GC/ECD analyses, calibration should be checked at the beginning
of each shift, during the work shift every 6 h (or every 6 samples, whichever
is less frequent), and after last sample of each shift.
GC/MS Tuning
Tuning criteria for GC/MS calibration standards [i.e., decafluorotriphenyl-
phosphine (DFTPP) and p-bromofluorobenzene (BFB)] are given in Tables C-2
and C-3 of Appendix C. Tuning must be performed and verified before each
12-h shift (as per U.S. EPA CLP and Methods 624/625).
Control Limit
RRF determined for the specific compounds listed in Table 6 should
agree within 25 percent of the initial calibration. This contro.l limit
is used by the U.S. EPA CLP for compounds analyzed by GC/MS. Acceptance
criteria for performance tests (Tables C-5 and C-8) from U.S. EPA Methods
1624B and 1625B are provided for comparison in Appendix C. RRF determined
for PCBs and pesticides with GC/ECD should agree within 15 percent of the
initial calibration as specified in U.S. EPA CLP.
Corrective Action
If control limit is not met, the initial three-point calibration will
have to be repeated. The last sample analyzed before the standard analysis
that failed criteria should then be reanalyzed. The results are expected
to agree within 15 percent (e.g., the expected reproducibility for replicate
injections of complex extracts). However, if the results exceed a 25 percent
control limit, the instrument is assumed to have been out of control during
the original analysis and the earlier data would be rejected. Reanalysis
of samples should progress in reverse order until it is determined that
48
-------
TABLE 6. COMPOUNDS THAT MUST MEET
ONGOING CALIBRATION CONTROL LIMIT
Semivolatiles
Phenol
1,4-dichlorobenzene
2-nitrophenol
2,4-dichlorophenol
Hexachlorobutadiene
4-chloro 3-methylphenol
2,4,6-trichlorophenol
Acenaphthene
N-ni trosodi phenylami ne
Pentachlorophenol
Fluoranthene
Di-n-octyl phthalate
Benzo(a)pyrene
Volatiles
Vinyl chloride
1.1-di ehloroethane
Chloroform
1,2-di chloropropane
Toluene
Ethyl benzene
49
-------
ORGANIC COMPOUNDS
DATA REPORTING REQUIREMENTS
DECEMBER 1986
there is <25 relative percent difference (RPD) between initial and reanalysis
results. In some cases reanalysis results may exceed a 25 percent control
limit because of a matrix effect. If the next sample reanalyzed meets
RPD requirements, evidence exists for assuming a matrix effect. Requirements
for additional reanalysis should be at the discretion of the program manager
or project coordinator.
Report
Samples requiring reanalysis should be identified. Reanalysis results
should be provided with the sample results. A discussion of the values
causing exceedance of limits and corrective actions taken should also be
provided.
50
-------
ORGANIC COMPOUNDS
DATA REPORTING REQUIREMENTS
DECEMBER 1986
DATA REPORTING REQUIREMENTS
The following standard items should be provided by the analytical
laboratory. The items listed below include most but not all of the standard
documentation required by the U.S. EPA CLP. The documentation below is
required for independent QA/QC review of the data and should always be
specified in the original statement of work:
• A cover letter discussing analytical problems (if any) and
referencing or describing the procedure used
• Reconstructed ion chroma tog rams for GC/MS analyses for each
sample
• Mass spectra of detected target compounds (GC/MS) for each
sample
• GC/ECD and/or GC/FID chromatograms for each sample
• Raw data quantification reports for each sample
• A calibration data summary reporting calibration range used
(and DFTPP and BFB spectra and quantification report for
GC/MS analyses)
• Final dilution volumes, sample size, wet-to-dry ratios,
and instrument detection limit
51
-------
ORGANIC COMPOUNDS
DATA REPORTING REQUIREMENTS
DECEMBER 1986
• Analyte concentrations with reporting units identified (to
two significant figures unless otherwise justified)
• Quantification of all analytes in method blanks (ng/sample)
• Method blanks associated with each sample
• Tentatively identified compounds (if requested) and methods
of quantification (include spectra)
• Recovery assessments and a replicate sample summary (laboratories
should report all surrogate spike recovery data for each
sample; a statement of the range of recoveries should be
included in reports using these data)
• Data qualification codes and their definitions.
RECOVERY AND BLANK CORRECTIONS
Recovery corrections will be applied only if the stable isotope dilution
technique is used with multiple standards. (Recovery correction algorithms
are incorporated into the quantification software.)
Blank corrections should not be applied by the laboratory. Values
for analytes in method blanks should be reported by the laboratory with
the data; corrections may then be made by program or project data managers.
Whether data are corrected or not, the concentration of analytes in method
blanks should always be given in reports. Results for several analytes
are often suspect because they are commonly reported in method blanks.
Hence, reported concentrations of phthalates, methylene chloride, acetone,
chloroform, and benzene in the sample should be carefully compared to those
52
-------
ORGANIC COMPOUNDS
DATA REPORTING REQUIREMENTS
DECEMBER 1986
in the method blank before the compounds are assessed as environmental
contaminants of concern.
LOWER LIMIT OF DETECTION
Lower limits of detection (LLD) are defined in this section. LLD
are established for GC/MS analyses by analysts based on their experience
with the instrumentation and with interferences in the sample matrix being
analyzed. LLD are greater than-the instrumental detection limits because
the former take into account sample interferences. To estimate LLD, the
noise level should be determined within the range of retention times represen-
tative of the analytes to be quantified. These determinations should be
made for at least three field samples in the sample batch under analysis.
The signal required to exceed the average noise level by at least a factor
of two should then be estimated. This signal is the minimum response required
to identify a potential signal for quantification. The LLD is the concentration
corresponding to the level of this signal based on the response factors
determined with standards. Based on best professional judgment, this LLD
would then be applied to samples in the batch with comparable or lower
interference. Samples with much higher interferences (e.g., at least a
factor of two higher) should be assigned a higher LLD (usually a multiple
of the original LLD).
These LLD values may be less than the rigorously defined method detection
limits specified in the revised "Guidelines Establishing Test Procedures
for the Analysis of Pollutants" (40 CFR Part 136, 10/26/84). This latter
procedure requires the analysis of seven replicate samples and a statistical
determination of the method detection limit with 99 percent confidence.
Data quantified between the LLD and the rigorous method detection limit
are valid and useful in environmental investigations of low-level contamination,
but have a lower statistical confidence associated with them than data
quantified above the method detection limit.
53
-------
ORGANIC COMPOUNDS
DATA REPORTING REQUIREMENTS
DECEMBER 1986
LLD quantification approaches for GC/FID and GC/ECD instrumentation
should be determined analogously (i.e., a minimum 2:1 signal:noise ratio
taking into account representative interferences in field samples).
54
-------
ORGANIC COMPOUNDS
COST IMPLICATIONS
DECEMBER 1986
COST IMPLICATIONS
Higher analytical costs may be required to achieve lower detection
limits and to increase the precision of results (Table 7). Lowering detection
limits to achieve project goals can increase costs, particularly if additional
sample cleanup is required. Additional sample cleanup can also improve
precision because interferences are removed. However, the range of precision
expected at a given detection limit in Table 7 reflects primarily differences
in the analytical variability of a set of diverse compound types. For
example, hydrocarbons can typically be recovered at the lower end of each
range of precision estimates shown, while phthalates and some acid compounds
are often analyzed much less precisely (i.e., higher coefficient of variation).
Hence, a wide range of precision may be found at constant cost when analyses
cover a wide range of compounds.
The major determinants of the range of analytical costs at a given
detection limit are individual laboratory efficiencies and the specific
analytical technique used (i.e., methods having large differences in cost
can yield similar detection limits and precision of results). Because
of these factors, lowering the required detection limits tends to raise
the minimum cost expected for the analysis; a range in costs can still
be expected above this minimum for different laboratories. Substitution
of a tissue matrix for a sediment matrix also increases costs because tissue
extracts can be more difficult to manipulate in the laboratory.
Major goals of QA/QC activities are to provide feedback to minimize
the quantity of data that are rejected (a waste of sampling and analysis
resources), to improve the legal defensibi1ity of the data set, and to
enable an assessment of comparability among data sets. Additional analytical
55
-------
TABLE 7. APPROXIMATE COST RANGE OF ANALYSES AS A FUNCTION
OF MATRIX, DETECTION LIMITS, AND PRECISION3
Matrix
Sediments
o Extractable
acid/base/neutrals/
PCBs/pesticides
o Volatile;
Approximate
Detection Limit
500 ppb
<1-50 ppb
5-20 ppb
Typical
Precision
±20% - >±100%
<±5% - >±50%
<±5% - >±50%
Approximate
Cost Rangeb
>$600
$800 - >$2,000
$170-$300
Tissues
o Extractable
acid/base/neutrals/
PCBs/pesticides
o Volatile*
<1-20 ppb
<5-20 ppb
<±5% - >±100%
<±10% - >±100%
$900 - >$2,000
$250-$350
a NOTE: Cost range is based on multiple quotes complied in 1986 for specific applications
and >5 samples. The actual costs may vary from the range shown. The table provides
a general perspective of the relative difference in costs.
b Each cost range is mainly the result of laboratory differences in technique and
pricing, NOT the range in precision or detection limits shown.
56
-------
ORGANIC COMPOUNDS
COST IMPLICATIONS
DECEMBER 1986
costs are incurred to achieve these goals because QC samples must be analyzed
with each sample set. The percent of the total analytical cost attributed
to QC samples as a function of the number of samples submitted for analysis
is shown in Figure 1. The number of QC samples for each sample set is
based on the minimum frequency of analysis recommended in the "QA PROCEDURES
AND REQUIREMENTS" section. The percent of total costs attributed to QC
samples rapidly declines when more than one sample is submitted for analysis.
The percent QA cost is constant at 7 to 12 percent of total costs (depending
on whether matrix spike analyses are conducted) in sets of greater than
50 samples.
57
-------
100 -,
I1J
UJ
CL
IL
°g
o) in
80 -
60 -
40 -
CD
UJ
O
DC
20 -
1 SRM
1REP
(1 M.S.)
SRM : STANDARD REFERENCE MATERIAL
REP: REPLICATE ANALYSIS
M.S.: MATRIX SPIKE SAMPLE (if isotope dilution
technique and/or SRM is not used)
Assumes calibration runs and method blanks included
in per sample cost, and method has been checked with
spiked blanks.
1SRM
2 REP
(1 M.S.)
1SRM
3 REP
(3 M.S.)
2 SRM
5 REP
(5 M.S.)
5 20 50 100
NUMBERS OF FIELD SAMPLES IN SET
4 SRM
10 REP
(10 M.S.)
200
Figure 1. Cost implication of minimum recommended QA samples for Puget Sound programs
(as a function of numbers of field samples analyzed)
-------
ORGANIC COMPOUNDS
REFERENCES
DECEMBER 1986
REFERENCES
Brown, D., A. Friedman, and W. MacLeod, Jr. 1985. Quality assurance guidelines
for chemical analysis of aquatic environmental samples. Draft. Prepared
by National Analytical Facility Region X and National Oceanographic and
Atmospheric Administration, for Seattle District, U.S. Army Corps of Engineers,
Seattle, WA.
Hiatt, M.H. 1981. Analysis of fish and sediment for volatile priority
pollutants. Anal. Chem. 53:1541-1543.
Hiatt, M.H., and T.L. Jones. 1984. Isolation of purgeable organics from
solid matrices by vacuum distillation. U.S. Environmental Protection Agency,
Region IX, Environmental Monitoring Systems Laboratory, Las Vegas, NV.
Horwitz, W., L. Kamps, and K. Boyer. 1980. Quality assurance in the analysis
of foods for trace contaminations. Anal. Chem. 63:1344-1354.
Keith, L.H., W. Crommett, J. Deegan, Jr., R.A. Libby, J.K. Taylor, and
G. Wentler. 1983. Principles of environmental analysis. Anal. Chem.
55:2210-2218.
MacLeod Jr., W., D. Brown, A. Friedman, 0. Maynes, and R. Pierce. 1984.
Standard analytical procedures of the NOAA National Analytical Facility,
1984-85, extractable toxic organic compounds. Prepared for the NOAA National
Status and Trends Program. NOAA Technical Memorandum NMFS F/NWC-64.
Metro. 1981 (revised 1983). Analytical support and data validity: organics.
Prepared for toxicant pretreatment planning study. Municipality of Metropolitan
Seattle, Seattle, WA.
NUS. 1985. Laboratory data validation functional guidelines for evaluating
organics analyses. Technical Directive Document No. HQ-8410-01. Prepared
by the U.S. EPA Data Validation Workgroup for U.S. EPA Hazardous Site Control
Division.
Ozretich, R.J. and W.P. Schroeder. 1985. Determination of priority pollutant
organic pollutants in marine sediment, tissue, and reference materials
utilizing bonded-phase sorbants. U.S. Environmental Protection Agency,
Environmental Research Laboratories, Narragansett, RI and Newport, OR.
Tetra Tech. 1985. Commencement Bay nearshore/tideflats remedial investiga-
tion. Final Report. Prepared for Washington Department of Ecology and
U.S. EPA. Tetra Tech, Inc., Bellevue, WA.
59
-------
ORGANIC COMPOUNDS
REFERENCES
DECEMBER 1986
Tetra Tech. 1986a. Bioaccumulation monitoring guidance: 4. Analytical
methods for U.S. EPA priority pollutants and 301(h) pesticides in tissues from
estuarine and marine organisms. Final Report. Prepared for U.S. Environmental
Protection Agency, Office of Marine and Estuarine Protection, Washington, DC.
Tetra Tech. 1986b. Analytical methods for U.S. EPA priority pollutants
and 301(h) pesticides in estuarine and marine sediments. Final Report.
Prepared for U.S. Environmental Protection Agency Office of Marine and
Estuarine Protection, Washington, DC.
Tetra Tech. 1986c. Quality assurance/quality control for 301(h) monitoring
programs: guidance on field and laboratory methods. Final Report. Prepared
for U.S. Environmental Protection Agency, Washington, DC. 267 pp. +
appendices.
U.S. Environmental Protection Agency. 1984a (revised January, 1985).
U.S. EPA contract laboratory program statement of work for organics analysis,
multi-media, multi-concentration. IFB WA 85-T176, T177, T178. U.S. EPA,
Washington, DC.
U.S. Environmental Protection Agency. 1984b. Method 1625 Revision B.
Semivolatile organic compounds by isotope dilution GC/MS. Federal Register
Vol. 49, No. 209. October 26, 1984. pp. 43416-43429.
60
-------
ORGANIC COMPOUNDS
GLOSSARY
DECEMBER 1986
GLOSSARY
Accuracy - The closeness of a measured or computed value to its true value.
Analyte - The specific component measured in a chemical analysis.
Batch - Usually refers to the number of samples that can be prepared or
analyzed at one time. A typical commercial batch size is 20 samples for
extraction of organic compounds.
Blank-Corrected - The concentration of a chemical in a sample adjusted
for the concentration of that chemical in the method blank carried through
the procedure concurrently with the sample.
Calibration - The systematic standardization of either the response of
instruments used for measurements or the chemical separation achieved by
a laboratory cleanup procedure.
Certified Reference Material - A reference material accompanied by, or
traceable to, a certificate stating the concentration of chemicals contained
in the material. The certificate is issued by an organization, public
or private, which routinely certifies such material (e.g., National Bureau
of Standards, American Society for Testing and Materials).
Coefficient of Variation - The standard deviation expressed as a percentage
of the mean.
Control Limit - Defines the minimum quality of data as measured by some
indicator (e.g., recovery) required to assume that the system or method
is performing as expected. Exceedance of a control limit triggers action
by the laboratory to correct the problem before data are reported.
Corrective Action - Measures taken to remove, adjust, remedy, or counteract
a malfunction or error so that a standard or required condition is met.
Duplicate Analysis - A second analysis made on the same (or identical)
sample of material to assist in the evaluation of measurement variance.
GC - Gas chromatography. An instrumental technique used to separate a
complex mixture into its component compounds by partitioning the compounds
between a mobile gaseous phase (under pressure) and a stationary solid
or liquid phase.
61
-------
ORGANIC COMPOUNDS
GLOSSARY
DECEMBER 1986
GC/ECD - Gas chromatography/el ectron capture detection. An instrumental
technique useful for the determination of compounds containing halogens
(e.g., chlorine).
GC/FID - Gas chromatography/f 1 ame ionization detection. An instrumental
technique useful for the detection of organic compounds that can be converted
to ions during exposure to a flame.
GC/MS - Gas chromatography/mass spectroscopy. An instrumental technique
useful for breaking organic compounds into characteristic fragments that
can be used to determine the original structure of the compound.
GPC - Gel permeation chromatography. A cleanup procedure used to remove
interfering biological macromolecules from sample extracts.
HPLC - High pressure (or high performance) liquid chromatography. An instru-
mental technique used to separate a complex mixture into its component
compounds by partitioning the compounds between a mobile liquid phase (under
high pressure) and a stationary solid phase.
Injection Internal Standard - A standard added to a sample extract just
prior to instrumental analysis. This standard is used to determine the
actual percent recovery of the surrogate spike compounds. When the isotope
dilution technique is not used, the injection internal standard also is
used to quantify compounds of interest in the sample relative to standards.
Isotope Dilution Technique - A technique for quantification of organic
compounds that uses a large number of stable isotopically labeled compounds
(i.e., compounds for which some hydrogen atoms have been replaced with
deuterium, or some carbon-12 atoms have been replaced with carbon-13) spiked
in the sample before sample extraction to correct for compound losses during
sample workup. The labeled compounds are analogs of the compounds of interest
and behave similarly.
Matrix - The sample material in which the chemicals of interest are found
(e.g., water, sediment, tissue).
Matrix Spike - An analysis conducted by adding a known amount of chemicals
of interest to an actual sample (i.e., matrix), usually prior to extraction
or digestion, and then carrying the spiked sample through the analytical
procedure. The final matrix spike results are reduced by the amount of
each chemical found in a replicate analysis of the sample conducted without
spikes. A comparison of these results with the known concentration of
spike added to the sample enables an evaluation of the effect of the particular
sample matrix on the recovery of compounds of interest.
62
-------
ORGANIC COMPOUNDS
GLOSSARY
DECEMBER 1986
Method Blank - A measure of the contribution of analytes from all laboratory
sources external to the sample. The method blank value is determined by
proceeding through all phases of extraction and analysis with no addition
of sample.
Method Spike - A method blank to which a known amount of surrogate standards
and analytes (compounds of interest) have been added.
Metro - Municipality of Metropolitan Seattle.
NBS - National Bureau of Standards.
NOAA - National Oceanic and Atmospheric Administration.
NQAA/NMFS - NOAA/Northwest National Marine Fisheries Services (Montlake
facility in Seattle).
NOAA/PMEL - NOAA/Pacific Marine Environmental Laboratories (Sand Point
facility in Seattle).
Noise - The electronic signal intensity attributed to instrument "background"
or electronic current from chemical interferents (i.e., any part of an
electrical signal that cannot be related in a known way to the electronic
current from a target compound).
Precision - The degree of mutal agreement characteristic of independent
measurements as the result of repeated application of a method under specified
conditions.
Priority Pollutant - Toxic pollutants defined by the U.S. EPA in 1976 that
are the primary subject of regulation of the Clean Water Act. A list of
these substances can be found in the Code of Federal Regulations Vol. 40,
Section 401.15.
PTFE - Polytetrafluoroethylene; the generic chemical name for materials
such as Teflon, a registered trademark of the du Pont Corporation.
QA/QC - Quality assurance/quality control (see below).
Quality Assurance - The total integrated program for assuring the reliability
of monitoring and measurement data. A system for integrating the quality
planning, quality assessment, and quality improvement efforts to meet user
requirements.
Quality Control - The routine application of procedures for obtaining prescribed
standards of performance in the monitoring and measurement process.
63
-------
ORGANIC COMPOUNDS
GLOSSARY
DECEMBER 1986
Quantification - The determination or expression of the number or amount
of a variable.
Reconstructed Ion Chromatogram - A graphical display of the total ionization
current resulting from all mass fragments detected over time during a mass
spectral analysis. The Chromatogram can be used to indicate the relative
composition of components in the sample mixture analyzed by GC/MS.
Recovery - The amount of a chemical detected in a sample extract at the
end of a procedure relative to the total amount present in a sample before
the procedure was begun. Also, the amount of a chemical detected in a
sample relative to the amount added (i.e., spike) or known to be present
(i.e., in a naturally derived standard reference material). Recovery is
usually expressed as a percentage.
Relative Percent Difference - Difference of two measurements xj and x£,
divided by the mean of the measurements, multiplied by 100.
Replicate - One of several identical experiements, procedures, or samples.
Duplicate is a special case of replicates consisting of two samples or
measurements.
Reproducibility - The ability to produce the same results for a measurement.
Often measured by calculation of relative percent difference or coefficient
of variation.
Resection - The surgical removal of tissue from an organism during sampling
(dissection is the sectioning of tissues within the organism, but does
not entail removal of the tissues).
Response Factor - Generally, the ratio of the amount (mass) of a substance
to a measurement of its response over time measured by the detector of
an analytical instrument. The ratio of response factors for a chemical
and a surrogate spike in a sample, or a chemical in a sample and a standard
calibration are used to quantify the concentration of chemicals in a sample.
Semi volatile Organic Compounds - Organic compounds with moderate vapor
pressures that can be extracted from samples using organic solvents and
analyzed by gas chromatography. In this document, semivolatile organic
compounds include the U.S. EPA acid/base/neutral compounds (including pesticides
and PCBs) as well as numerous other neutral and organic acid compounds
of regional interest (e.g., carbozole, retene, coprostanol, 4-methylphenol).
Sensitivity - Capability of a method or instrument to discriminate between
samples having differing concentrations of a chemical. The degree to which
an instrument responds to low concentrations of a chemical.
64
-------
ORGANIC COMPOUNDS
GLOSSARY
DECEMBER 1986
Significant Figure - A figure(s) that remains to a number or decimal after
the zeros to the right or left are cancelled.
Spike - The addition of a known amount of a substance to a sample.
Standard - A substance or material, the properties of which are believed
to be known with sufficient accuracy to permit its use to evaluate the
the same property of a sample. In chemical measurements, a standard often
describes a solution of chemicals, commonly prepared by the analyst, to
establish a calibration curve or the analytical response function of an
instrument.
Standard Reference Material (SRM) - A material or substance for which one
or more properties are sufficiently well established to be used for the
assessment of a method or the calibration of an instrument.
Surrogate Spike Compound - A known amount of a compound that has characteristics
similar to that of a compound of interest, added to a sample prior to extrac-
tion. The surrogate compound can be used to estimate the recovery of chemicals
in the sample. These compounds are also called "recovery internal standards".
Target Compounds - The chemicals of interest in a sample that can be quantified
relative to response factors of reliable standards (in contrast to tentatively
identified compounds).
Tentatively Identified Compounds - Chemicals identified in a sample on the
basis of mass spectral characteristics held in common with a reference
mass spectra of a known chemical. These compounds cannot be more confidently
identified unless a reliable standard of the compound is obtained and is
confirmed to co-elute with the tentatively identified compound and generate
similar mass spectra using the same gas chromatograph/mass spectrometer.
U.S. EPA CLP - United States Environmental Protection Agency Contract
Laboratory Program.
Volatile Organic Compounds - Organic compounds with high vapor pressures
that tend to evaporate readily from a sample. In this document, volatile
organic compounds are the 29 U.S. EPA priority pollutants considered as
volatiles (e.g., benzene).
Warning Limit - In Puget Sound programs, a value either above or below
which data returned by a laboratory are subjected to qualification before
inclusion in a regional database. The principle is identical to that of
a control limit, but is less stringent and serves as a warning that the
system or method may become out of control.
65
-------
APPENDIX A
U.S. EPA CONTRACT LABORATORY PROGRAM:
PROCEDURES FOR ANALYSIS OF EXTRACTABLE
ORGANIC COMPOUNDS IN SOILS/SEDIMENT
[base/neutrals and acids, and pesticides/PCBs
at detection limits of 500-1,000 ppb (dry weight)]
Note: The most recent U.S. EPA/CLP Statement of Work (Spring 1987)
will be included in this appendix when available.
-------
This appendix has been extracted from the U.S. EPA Contract Laboratory
Program (CLP) Statement of Work for revised January 1985. There are some
differences in nomenclature that should be noted here. The term "screening"
as used in the CLP refers to an optional 20 ppm screen to determine the
appropriate preparation technique within the confines of the CLP. The
"low-level" analysis referred to in the CLP is the method that will yield
500-1,000 ppb detection limits. This is what has been referred to in this
Puget Sound protocol as screening level analyses.
2. Low Level Preparation for Screening and Analysis of Extractable Base/Neutrals
and Acids (Semlvolatiles SNA), and Pesticides/ PCBs (PEST) in Sediment/Soil.
2.1 Summary of Method
2.1.1 A 30 gram portion of sediment is mixed with anhydrous sodium
sulfate and extracted with 1:1 methylene chloride/acetone using
an ultrasonic probe. If the optional low level screen is used,
a portion of this dilute extract is concentrated fivefold and is
screened by GC/FID or GC/MS. If peaks are present at greater
than 20,000 ug/kg, discard the extract and prepare the sample
by the medium level method. If no peaks are present at greater
than 20,000 ug/kg, the extract is concentrated and split into two
fractions. An optional gel permeation column cleanup may be used
before splitting the extract. One fraction is for GC/MS analysis
of BNA. The other fraction is cleaned up using a micro alumina
column and analyzed by GC/EC for pesticides.
D - 32
Rev: 9/84
-------
II. C
2.2 Interferences
2.2.1 Method interferences may be caused by contaminants in solvents,
reagents, glassware, and other sample processing hardware that
lead to discrete artifacts and/or elevated baselines in the total
ion current profiles. All of these materials must be routinely
demonstrated to be free from interferences under the conditions
of the analysis by running laboratory reagent blanks. Matrix
interferences may be caused by contaminants that are coextracted
from the sample. The extent of matrix interferences will vary
considerably from source to source, depending upon the nature and
diversity of the industrial complex or municipality being sampled.
2.3 Apparatus and Materials
2.3.1 Apparatus for determining percent moisture
2.3.1.1 Oven, drying
2.3.1.2 Desiccator
2.3.1.3 Crucibles, porcelain
2.3.2 Disposable Pasteur glass pipets, 1 mL
2.3.3 Sonic cell disrupter, Heat Systems - Ultrasonics, Inc. Model
375C or equivalent- (375 watt with pulsing capability and 3/4"
disrupter horn).
2.3.4 Beakers, 400 mL
2.3.5 Vacuum filtration apparatus
2.3.5.1 Buchner funnel.
2.3.5.2 Filter paper, Whatman No. 41 or equivalent.
2.3.6 Kuderna-Danish (K-D) apparatus-
2.3.6.1 Concentrator tube - 10 mL, graduated (Kontes K-570040-
1029 or equivalent).
D - 33
Rev: 9/8-
-------
II. C
2.3.6.2 Evaporative flask - 500 mL (Kontes K-570001-0500 or
equivalent).
2.3.6.3 Snyder column - three-ball macro (Kontes K-503000-0121
or equivalent).
2.3.6.4 Snyder column - two ball micro (Kontes K-569001-0219)
or equivalent).
2.3.7 Silicon carbide boiling chips - approximately 10/40 mesh. Heac
to 400°C for 30 minutes or Soxhlet extract with methylene chloride.
2.3.8 Water bath - heated, with concentric ring cover, capable of
temperature control (+2°C). The bath should be used in a hood.
2.3.9 Top loading balance, capable of accurately weighing 0.01 gm.
2.3.10 Vials and caps, 2 mL for GC auto sampler.
2.3.11 Balance - Analytical, capable of accurately weighing 0.0001 gm.
2.3.12 Nitrogen evaporation device equipped with a water bath that can be
maintained at 35-40°C. The N-Evap by Organomation Associates, Inc.
South Berlin, MA (or equivalent) is suitable.
2.3.13 Gel permeation chromatography cleanup device.
2.3.13.1 Automated system
2.3.13.1.1 Cel permeation chromatograph (GPC) Analytical
Biochemical Labs, Inc. GPC Autoprep 1002 or
equivalent Including:
2.3.13.1.2 25 mm ID X 600 - 700 mm glass column packed
with 70 gm of Bio-Beads SX-3.
0-34
5/84
-------
II. C
2.3.13.1.3-Syringe, 10 mL with luer lok fitting.
2.3.13.1.4 Syringe filter holder and filters - stainless
steel and TFE, Gelman 4310 or equivalent.
2.3.13.2 Manual system assembled from parts.*
2.3.13.2.1 25 mo ID X 600 - 700 mm heavy wall glass
column packed with 70 go of BIO-Beads SX-3.
2.3.13.2.2 Pump: Altex Scientific, Model No. 1001A,
senipreparative, solvent metering system.
Pump capacity - 28 mL/min.
2.3.13.2.3 Detector: Altex Scientific, Model No. 153,
with 254 no UV source and 8-ul semi-prepar-
ative flowcells (2-nn pathlengths)
2.3.13.2.4 Microprocessor/controller: Altex Scientific,
Model No. 420, Microprocessor System Con-
troller, with extended memory.
2.3.13.2.5 Injector: Altex Scientific, catalog No.
201-56, sample injection valve, Tefzel,
with 10 mL sample loop.
2.3.13.2.6 Recorder: Linear Instruments, Model No. 385,
10-Inch recorder.
2.3.13.2.7 Effluent Switching Valve: Teflon slider
valve, 3-way with 0.060" ports.
*Wlse, R.H., Bishop, D.F., Williams, R.T. & Austern, B.M. "Gel Permeation
Chromatography In the GC/MS Analysis of Organlcs in Sludges" U.S. EPA,
Municipal Environmental Research Laboratory - Cincinnati, Ohio 45268
D - 35
5/84
-------
II. C
2..3.13.2.8 Supplemental Pressure Gauge with connecting
Tee: U.S. Gauge, 0-200 psl, stainless steel.
Installed as a "downstream" monitoring device
between column and detector.
Flow rate was typically 5 mL/min. of methylene
chloride. Recorder chart speed was 0.50 ca/min.
2.3.14 Chromatography column for alumina. 8 mL (200 mm & 8 mm ID) Poly-
propylene column (Kontes K-420160 or equivalent) or 6 mL (150 mm
X 8 mm ID) glass column (Kontes K-420155 or equivalent) or 5 mL
t
serological pipets plugged with a small piece of Pyrex glass wool
in the tip. (Pyrex glass wool shall be pre-rinsed with approp-
riate solvents to Insure its cleanliness). The Kontes columns
may be plugged with Pyrex glass wool or a polyethylene porous
disk (Kontes K-420162).
2.3.15 Pyrex glass wool.
2.3.16 Bottle or test tube, 50 mL with Teflon lined screw cap for sulfur
removal.
2.3.17 Pasteur pipets, disposable.
2.4 Reagents
2.4.1 Sodium Sulfate -anhydrous and reagent grade, heated at 400°C
for four hours, cooled in a desiccator, and stored in a glass
bottle. Baker anhydrous powder, catalog #73898 or equivalent.
2.4.2 Methylene chloride, hexane, acetone, isoooctane, 2 propanol and
benzene pesticide quality or equivalent.
D - 36
5/84
-------
II. C
2.4.3 Alumina - neutral, super I Woeln or equivalent (Universal Scien-
tific, Atlanta, GA or equivalent). Prepare activity III by adding
72 (v/w) reagent water to the Super I neutral alumina. Tumble or
shake on a wrist action shaker for a minimum of 2 hours or prefer-
ably overnight. There should be no lumps present. Store in a
tightly sealed glass container. A 25 cycle soxhlet extraction of
the alumina with methylene chloride is required if a solvent blank
analyzed by the pesticide techniques indicate any interferences
for the compounds of interest. See page D-28, paragraph 1.5.11.
2.4.4 Reagent water - Reagent water is defined as a water in which an
interferent is not observed at the method detection limit of each
parameter of interest.
2.4.5 Tetrabutylammonium (TEA) - sulfite reagent. Dissolve 3.39 g
tetrabutylaramonium hydrogen sulfate in 100 mL distilled water.
To remove impurities, extract this solution three times with 20
mL portions of hexane. Discard the hexane extracts, and add 25
g sodium sulfite to the water solution. Store the resulting
solution, which is saturated with sodium sulfite, in an amber
bottle with a Teflon-lined screw cap. This solution can be
stored at room temperature for at least one month.
2.4.6 GPC calibration solutions:
2.4.6.1 Corn oil - 200 mg/mL in methylene chloride.
2.4.6.2 Bis(2-ethylhexylphthalate) and pentachlorophenol - 4.U
mg/mL in methylene chloride.
2.4.7 Sodium Sulfite, reagent grade.
2.4.8 Surrogate standard spiking solution.
2.4.8.1 Base/neutral and acid surrogate solution.
D - 37
Rev: 9/84
-------
II. C
2.4.8.1.1 Surrogate standards are added to all samples,
blanks, matrix spikes, matrix spike duplicates,
and calibration solutions; the compounds speci-
fied for this purpose are phenol-dg, 2,4,6-
trlbromophenol, 2-fluorophenol, nitrobenzene-d5,
terphenyl-di4, and 2-fluorobiphenyl. Two
additional surrogates, one base/neutral and one
acid may be added.
2.4.8.1.2 Prepare a surrogate standard spiking solution
at a concentration of 100 ug/1.0 mL for BN and
200 ug/1.0 mL for acids in methanol. Addition
of 0.5 mL of this solution to 30 gm of sample
is equivalent to a concentration of 1700 ug/kg
for base/neutrals and 3,330 ug/kg for acids of
each surrogate standard. Store the spiking
solutions at 4°C in Teflon-sealed containers.
The solutions must be replaced after six months,
or sooner if comparison with quality control
check samples' indicate a problem.
2.4.8.2 Pesticide surrogate standard spiking solution.
2.4.8.2.1 The surrogate standard is added to all samples,
blanks, matrix spike, matrix spike duplicates,
and calibrations solutions; the compound spec-
ified for this purpose is dibutyl chlorendate.
2.4.8.2.2 Prepare a surrogate standard spiking solution
at a concentration of 20 ug/1.0 mL in methanol.
D - 38
Rev: 10/84
-------
II. C
Addicion of 100 uL of this solution to 30 go
of sample is equivalent to a concentration
of 67 ug/kg of surrogate standard. Store
the spiking solutions at 4°C in Teflon-sealed
containers. The solutions should be checked
frequently for stability. These solutions
oust be replaced afte.r six months, or sooner
if comparison with quality control check
samples indicate a problem.
2.4.9 Matrix standard spiking solutions.
2.4.9.1 Base/neutral and acid matrix spiking solution consists of:
Base/Neutrals (100 ug/1.0 mL) Acids (200 ug/1.0 mL)
1,2,4-trichlorobenzene pentachlorophenol
acenaphthene phenol
2,4-dinitrotoluene 2-chlorophenol
di-n-butylphthalate 4-chloro-3-methylphenol
pyrene 4-nitrophenol
N-nitroso-di-n-propylamine
1,4-dichlorobenzene
Prepare a spiking solution that contains each of the
above in methanol.
Matrix spikes also serve as duplicates, therefore,
add 0.5 mL to each of two 30 gm portions from one
sample chosen for spiking.
2.4.9.2 Pesticide matrix standard spiking solution. Prepare a
spiking solution in methanol that contains the following
pesticides in the concentrations specified.
Pesticide ug/1.0 mL
lindane 2.0
heptachlor 2.0
aldrin 2.0
dleldrin 5.0
endrin 5.0
4,4* DDT 5.0
D - 39 Rev: 1/85
-------
II. C
Matrix spikes are also to serve as duplicates, therefore,
add 400 uL to each of two 30 gm portions from one sample
chosen for spiking.
2.5 Sample Extraction
2.5.1 Decant and discard any water layer on a sediment sample. Mix
samples thoroughly, especially composited samples. Discard any
foreign objects such as sticks, leaves, and rocks.
2.5.1.1 Transfer 50 g of soil/sediment to 100 ml beaker. Add 50
ml of water and stir for 1 hour. Determine pH of sample
with glass electrode and pH meter while stirring. Report
pH value on appropriate data sheets. If the pH of the soil
is greater than 11 or less than 5, contact the Project
Officer or Deputy Project Officer for instructions on how
to handle the sample. Document the instructions in the
Case Narrative. Discard this portion of sample. NOTE:
Recovery of dibutylchlorendate will be low if pH is
outside this range.
2.5.2 The following step should be performed rapidly to avoid loss of
the more volatile extractables. Weigh approximately 30 gins of
sample to the nearest 0.1 gram into a 400-mL beaker and add 60
gms of anhydrous sodium sulfate. Mix well. The sample should
have a sandy texture at this point. Immediately, add 100 mL of
1:1 methylene chloride - acetone to the sample.
2.5.2.1 Immediately after weighing the sample for extraction,
weigh 5-10 g of the sediment into a tared crucible.
Determine the percent moisture by drying overnight at
105°C. Allow to cool in a desiccator before weighing.
Concentrations of individual analytes will be reported
relative to the dry weight of sediment.
Percent moisture
gm of sample - gm of dry sample
gm of sample X 100 - 2 moisture
D - 40
Rev: 1/85
-------
II. C
2.5.2.2 Weigh out four 30 gra (record weight to nearest O.lg)
portions for use as matrix and matrix spike duplicates
and as matrix spikes. Follow 2.5.2 and then add 1.0 mL
of the base/neutral and acid matrix spike to each of
two portions and 400 uL of the pesticide matrix spike
to each of the other two portions.
2.5.2.3 Add 1.0 mL of base/neutral and acid surrogate standard
and 100 uL of pesticide surrogate to the sample.
2.5.3 Place the probe about 1/2" below the surface of the solvent but
above the sediment layer.
2.5.A Sonicate for 3 min., using 3/4" horn, at full power with pulse
set at 50%. Do not use microtip.
2.5.5 Decant and filter extracts through Whatman 041 filter paper using
vacuum filtration or centrifuge and decant extraction solvent.
2.5.6 Repeat the extraction two more times with 2 additional 100 mL
portions of 1:1 methylene chloride - acetone. Decant off the
extraction solvent after each sonication. On the final sonica-
tion, pour the entire sample into the Buchner funnel and rinse
with 1:1 methylene chloride - acetone.
2.5.6.1 If the sample is to be screened from the low level
method, take 5.0 mL and concentrate to 1.0 mL following
paragraph 2.7.2 or 2.7.3. Note that the sample volume
in this case is 5.0 mL not 8.0 mL as given in 2.7.2.
Screen the extract as per Section III, paragraph 2,
"Screening of Extractable Organic Extracts". Transfer
the remainder of the 1 mL back to the total extract
from paragraph 2..5.6 after GC/FID or GC/MS screening.
(CAUTION: To minimize sample loss, autosamplers which pre-
flush samples through the syringe should not be used.)
D - 41
Rev: 1/85
-------
II. C
2.5.7 Transfer the extract to a Kuderna-Danish (K-D) concentrator con-
sisting of a 10-mL concentrator tube and a 500-mL evaporative
flask. Other concentration devices or techniques may be used if
equivalency is demonstrated for all extractable and pesticide
compounds listed in Exhibit C.
2.5.8 Add one or two clean boiling chips to the evaporative flask and
attach a three-ball Snyder column. Pre-wet the Snyder column by
adding about 1 mL methylene chloride to the top. Place the K-D
apparatus on a hot water bath (80 to 90°C) so that the concentrator
tube is partially immersed in the hot water and the entire lower
rounded surface of the flask is bathed with hot vapor. Adjust
the vertical position of the apparatus and the water temperature
as required to complete the concentration in 10 to 15 minutes.
At the proper rate of distillation the balls of the column will
actively chatter but the chambers will not flood with condensed
solvent. When the apparent volume of liquid reaches 1 mL, remove
the K-D apparatus and allow it to drain and cool for at least 10
minutes, and make up to lOmL volume with methylene chloride.
2.5.9 If GPC cleanup is not used proceed to paragraph 2.7.
2.6. Extract Cleanup
2.6.1 GPC Setup and Calibration
2.6.1.1 Packing the column - Place 70 g of Bio Beads SX-3 in a
400-mL beaker. Cover the beads with methylene chloride;
allow the beads to swell overnight (before packing the
columns). Transfer the swelled beads to the column and
start pumping solvent through the column, from bottom to
top, at 5.0 raL/min. After approximately 1 hour, adjust
the pressure on the column to 7 to 10 psi and pump an
additional 4 hours to remove air from the column. Adjust
the column pressure periodically as required to maintain
7 to 10 psi.
D - 42
-------
II. C
2.6*1.2 Calibration of the column - Load 5 mL of the corn oil
solution Into sample loop No. 1 and 5 mL of the phthalate-
phenol solution into loop No. 2. Inject the corn oil and
collect 10 mL fraction (i.e., change fraction at 2-minute
Intervals) for 36 minutes. Inject the phthalate-phenol
solution and collect 15 mL fractions for 60 minutes.
Determine the corn oil elution pattern by evaporation of
each fraction to dryness followed by a gravimetric deter-
mination of the residue. Analyze the phthalate-phenol
fractions by GC/FID on the DB-5 capillary column, a UV
spectrophotometer, or a GC/MS system. Plot the concen-
tration of each component in each fraction versus total
eluent volume (or time) from the Injection points. Choose
a "dump time" which allows >^ 852 removal of the corn oil
and 2. 852 recovery of the bis(2-ethylhexyl)-phthalate.
Choose the "collect time" to extend at least 10 minutes
after the elution of pentachlorophenol. "Wash the
column at least 15 minutes between samples. Typical
parameters selected are: Dump time, 30 minutes (150
mL), collect time, 36 minutes (180 mL), and wash time,
15 minutes (75 mL). The column can also be calibrated
by the use of a 254 mm UV detector In place of gravimetric
and GC analyses of fractions• Measure the peak areas at
various elution times to determine appropriate fractions.
The SX-3 Bio Beads column may be reused for several
months, even if discoloration occurs. System calibra-
tion usually remains constant over this period of time
if column flowrate remains constant.
2.6.2 GPC Extract Cleanup
Prefliter or load all extracts via the filter holder to avoid
particulates that might cause flow stoppage. Load one 5.0 mL
aliquot of the extract onto the GPC column. Do not apply
D - 43
5/84
-------
II. C
excessive pressure when loading the GPC. Purge the sample
loading tubing thoroughly with solvent between extracts. After
especially dirty extracts, run a GPC blank (methylene chloride)
to check for carry-over. Process the extracts using the dump,
collect, and wash parameters determined from the calibration and
collect the cleaned extracts in 400 mL beakers tightly covered
with aluminum foil. The phthalatephenol calibration solution
shall be taken through the cleanup cycle with each set of 23
extracts loaded into the GPC. The recovery for each compound
must be >^ 85%. This must be determined on a GC/F1D, using a
DB-5 capillary column, a UV recording spectrophotometer, or a
GC/MS system. A copy of the printouts of standard and check
solution are required as deliverables with each case. Show Z
recovery on the copy.
2.6.2.1 If GPC cleanup of samples is required because of poor GC/
EC chroma'tography in Section IV, dilute the extract to 10
mL with methylene chloride and perform GPC cleanup as per
paragraph 2.6.2. The reagent blank accompanying the
samples should be included, unless only one or a partial
group of samples requires cleanup. In this case, set up
a new reagent blank with 10 mL of methylene chloride and
appropriate surrogate standard added.
2.6.3 Concentrate the extract as per paragraphs 2.5.7 and 2.5.8.
D - 44
Rev: 9/84
-------
II. C.
2.7 Splitting of Extract and Final Concentration
NOTE; If only pesticide or BNA anlaysls is to be performed on a sample
only the appropriate surrogates for that fraction should be added as
per paragraph 2.5.2.3 (and only appropriate matrix spikes for duplicate
matrix spike samples). The 10 mL extract resulting from paragraph 2.5.8
should not be split as described in paragraph 2.7.1, following, but
should be concentrated as follows: to 1.0 mL for BNAs (not to 0.8 mL
as In paragraph 2.7.2). However, for pestlcides/PCBs, follow 2.7.1 as
written, because of the limited cleanup capacity of the micro aluoina
column.
The alumina clean-up for pesticides is still required when BNA
surrogates are not present in order to remove polar Interferents.
2.7.1 Transfer 0.5 mL of the 10 mL methylene chloride extract to a
separate concentrator tube. Add 5 mL of hexane and a silicon
carbide boiling chip and mix using vortex mixer. Attach a
two-ball mlcro-Snyder column. Pre-wet the Snyder column by
adding 0.5 mL of hexane to the top of the column. Place the
K-D apparatus on a hoc water bath (80°-90°C) so that the
concentrator tube
-------
II. C
partially Immersed in Che hot water. Adjust the vertical posi-
tion of the apparatus and the water temperature as required to
complete the concentration in 5 to 10 minutes. Concentrate the
extract to an apparent volume of less than 1 mL. Use Nitrogen
blowdown to reduce the volume to 0.5 mL. Add 0.5 mL of acetone.
The pesticide extract must now be passed through an alumina
column to remove the BNA surrogates and polar interferences.
Proceed to paragraph 2.8.
2.7.2 Reattach the micro-Snyder column to the remaining 9.5 mL of the
extract and add a fresh silicon carbide boiling chip to the
concentrator tube. Pre-wet the Snyder column with 0.5 mL of
methylene chloride. Place the K-D apparatus on the hot water
bath (80°-90°C) so that the concentrator tube in partially
immersed in the hot water. Adjust the vertical position of the
apparatus and the water temperature as required to complete the
concentration in 5 to 10 minutes. When the apparent volume of
the liquid reaches 0.5 mL, remove the K-D apparatus from the
water bath and allow it to drain for at least 10 minutes while
cooling. Remove the Snyder column and rinse the lower joint
into the concentrator tube with 0.2 mL of methylene chloride.
Adjust the final volume to 0.95 mL with methylene chloride. If
GPC cleanup was used, this 0.95 mL represents a twofold dilution
to account for only half of the extract going through the GPC.
2.7.3. Nitrogen blowdown technique (taken from ASTM Method D 3086).
The following method may be used for final concentration of the
BNA extract instead of the procedure in paragraph 2.7.2. Place
the concentrator tube in a warm water bath (35°C) and evaporate
the solvent volume to below 1 mL using a gentle stream of clean,
dry nitrogen (filtered through a column of activated carbon).
Caution; New plastic tubing must not be used between the carbon
trap and the sample, since it may introduce interferences. The
D - 45
Rev: 1/35
-------
II. C
internal wall of the tube oust be rinsed down several times with
methylene chloride (hexane for pesticides analysis) during the
operation, and the final volume brought to 0.8 mL with methylene
chloride (hexane for pesticide analysis). During evaporation,
the tube solvent level must be kept below the water level of the
bath. The extract must never be allowed to become dry. If GPC
cleanup was used, this 0.8 mL represents a 2 times dilution to
account for only half the extract going through the GPC.
2.7.4 Store all extracts at A°C in the dark in Teflon-sealed containers
until all analyses are performed.
2.8 Pesticide/PCB.
2.8.1 Alumina Column Cleanup
All samples must be taken through this cleanup tecnhique to
eliminate BNA surrogates that will interfere in the GC/ECD
analysis.
2.8.1.1 Add 3 gm of activity III neutral alumina to the 10 mL
chromatographic column. Tap the column to settle the
alumina. Do not pre-wet the alumina.
2.8.1.2 Transfer the 1.0 mL of hexane/acetone extract from
paragraph 2.7.1 to the top of the alumina using a
disposable Pasteur pipet. Collect the eluate in a
clean, 10 mL concentrator tube.
2.8.1.3 Add 1 mL of hexane to the original extract concentrator
tube to rinse it. Transfer these rinsings to the alum-
ina column. Elute the column with an additional 9 mL of
hexane. Do not allow the column to go dry during the
addition and elution of the sample.
D - 46
5/84
-------
II. C
2.8.1.4 Concentrate the extract to 1.0 mL following either
paragraph 2.7.1 or 2.7.3, using hexane where methylene
chloride is specified. When concentrating medium level
extracts, the Nitrogen blovdown technique should be used
to avoid contaminating Che micro Snyder column.
2.8.2 Observe the appearance of the extract.
2.6.2.1 If crystals of sulfur are evident or sulfur is expected
to be present, proceed to paragraph 2.8.3.
2.8.2.2 If the sulfur is not expected to be a problem, transfer
the 1.0 mL to a GC vial and label as Pestlcide/PCB
fraction. The extract is ready for GC/ECD analysis.
Proceed to Section IV, paragraph 3. Score the extracts
at 4°C in the dark until analyses are performed.
2.8.3 Sulfur Cleanup
2.8.3.1 Transfer the 1.0 mL from paragraph 2.8.2 to a 50 mL
clear glass bottle or vial with a Teflon-lined screw cap.
Rinse the concentrator tube with 1.0 mL of hexane, adding
the rinsings to the SO mL bottle. If only a partial set
of samples requires sulfur cleanup, set up a new reagent
blank with 1.0 mL of hexane and take it through the
sulfur cleanup. Include the surrogate standards.
2.8.3.2 Add 1 mL TBA-sulfite reagent and 1 mL 2-propanol, cap the
bottle, and shake for at least 1 mln. If the sample is
colorless or if the initial color is unchanged, and if
clear crystals (precipitated sodium sulflte) are observed,
sufficient sodium sulfite is present. If the precipitated
sodium sulflte disappears, add more crystalline sodium
sulfite in approximately 100 mg portions until a solid
residue remains after repeated shaking.
D - 47
5/84
-------
11. C
2.8.3.3 Add 5 ml distilled water and shake for at least 1 min.
Allow the sample to stand for 5-10 min. and remove the
hexane layer (top) for analysis. Concentrate the hexane
to 1.0 mL as per paragraphs 2.7.1 and 2.7.3 using hexane
where methylene chloride is specified. The temperature
for the water bath should be about 80°C for the micro
Snyder column column technique. Continue as outlined
in paragraph 2.8.2.2.
D - 48
5/84
-------
APPENDIX B
U.S. EPA CONTRACT LABORATORY PROGRAM:
PROCEDURES FOR ANALYSIS OF PURGEABLE
ORGANIC COMPOUNDS
-------
This appendix has been extracted from the U.S. EPA Contract Laboratory
Program (CLP) Statement of Work for multi-media, multi-concentration organics
analyses, May 1984, revised January 1985. The sections discussing "medium
level" analyses are not considered applicable and have been removed. Use
of the medium level technique would yield detection limits much higher
than the 10-20 ppb (dry weight) considered appropriate for Puget Sound
environmental studies.
1. GC/MS Analysis of Purgeable Organics
1.1 Summary of Methods
1.1.2 Sediment/Soil Samples
1.1.2.1 Low Level. An inert gas is bubbled through a mixture
of a 5 gm sample and reagent water contained in a sug-
gested specially designed purging chamber (illustrated
on page D-95) at elevated temperatures. The purgeables
are efficiently transferred from the aqueous phase to
the vapor phase. The vapor is swept through a sorbent
column where the purgeables are trapped. After purging
is completed, the sorbent column is heated and back-
flushed with the inert gas to desorb the purgeables
onto a gas chromatographic column. The gas chromato-
graph is temperature programmed to separate the purge-
ables which are then detected with a mass spectrometer.
D - 64
Rev: 9/84
-------
1.2 Interferences
1.2.1 Impurities in the purge gas, organic compounds out-gassing
from the plumbing ahead of the trap, and solvent vapors in the
laboratory account for the majority of contamination problems.
The analytical system must be demonstrated to be free from
contamination under the conditions of the analysis by running
laboratory reagent blanks as described in Exhibit E. The use
of non-TFE tubing, non-TFE thread sealants, or flow controllers
with rubber components in the purging device should be avoided.
1.2.2 Samples can be contaminated by diffusion of volatile organics
(particularly fluorocarbons and methylene chloride) through
the septum seal into the sample during storage and handling.
A holding blank prepared from reagent water and carried through
the holding period and the analysis protocol serves as a check
on such contamination. One holding blank per case should be
analyzed. Data must be retained by laboratory and made avail-
able for inspection during on-site evaluations.
D - 65
Rev: 1/85
-------
IV.
1.2.3 Contamination by carry over can occur whenever high level and
low level samples are sequentially analyzed. To reduce carry
over, the purging device and sampling syringe must be rinsed
with reagent water between sample analyses. Whenever an
unusually concentrated sample is encountered, it should be
followed by an analysis of reagent water to check for cross
contamination. For samples containing large amounts of water-
soluble materials, suspended solids, high boiling compounds
or high purgeable levels, it may be necessary to wash out
the purging device with a detergent solution, rinse it with
distilled water, and then dry it in a 105°C oven between
analyses. The trap and other parts of the system are also
subject to contamination; therefore, frequent bakeout and
purging of the entire system may be required.
1.3 Apparatus and Materials
1.3.1 Micro syringes - 25 uL and larger, 0.006 inch ID needle.
1.3.2 Syringe valve - two-way, with Luer ends (three each), if
applicable to the purging device.
1.3.3 Syringe - 5 mL, gas tight with shut-off valve.
1.3.4 Balance-Analytical, capable of accurately weighing 0.0001 g.
and a top-loading balance capable of weighing O.lg.
1.3.5 Glassware
1.3.5.1 o Bottle - 15 mL, screw cap, with Teflon cap liner.
o Volumetric flasks - class A with ground-glass stopper's.
o Vials - 2 mL for GC autosampler.
D - 66
5/84
-------
IV.
1.3.6 Purge and trap device - The purge and trap device consists of
three separate pieces of equipment; the sample purger, trap
and the desorber. Several complete devices are now commercially
available.
1.3.6.1 The sample purger must be designed to accept 5 mL
samples with a water column at least 3 cm deep. The
gaseous head space between the water column and the
trap must have a total volume of less than 15 mL. The
purge gas must pass through the water column as finely
divided bubbles with a diameter of less than 3 mm at
the origin. The purge gas must be Introduced no more
than 5 mm from the base of the water column. The
sample purger, illustrated in Figure 1, meets these
design criteria. Alternate sample purge devices may
be utilized provided equivalent performance is
demonstrated.
1.3.6.2 The trap must be at least 25 cm long and have an inside
diameter of at least 0.105 inch. The trap must be
packed to contain the following minimum lengths of
absorbents: 1.0 cm of methyl silicone coated packing
(32 OV-1 on Chromosorb W or equivalent), 15 cm of 2,6-
diphenylene oxide polymer (Tenax-GC 60/80 mesh) and 8
cm of silica gel (Davison Chemical, 35/60 mesh, grade
15, or equivalent). The minimum specifications for the
trap are illustrated la Figure 2.
1.3.6.3 The desorber should be capable of rapidly heating
the crap to 180°C. The polymer section of the
trap should not be heated higher than 180°C and
the remaining sections should not exceed 220°C.
The desorber design, illustrated in Figure 2, meets
these criteria.
D - 67
5/84
-------
IV.
1.3.6.4 The purge and trap device may be assembled as a
separate unit or be coupled to a gas chromatograph
as illustrated in Figures 3 and 4.
1.3.6.5 A heater or heated bath capable of maintaining the
purge device at 40"C + 1°C.
1.3.7 GC/MS system
1.3.7.1 Gas chromatograph - An analytical system complete with
a temperature programmable gas chromatograph suitable
for on-column injection and all required accessories
including syringes, analytical columns, and gases.
1.3.7.2 Column - 6 ft long x 0.1 in ID glass, packed with IX
SP-1000 on Carbopack B (60/80 mesh) or equivalent.
1.3.7.3 Mass spectrometer - Capable of scanning from 35
to 260 amu every seven seconds or less, utilizing
70 volts (nominal) electron energy in the electron
impact ionlzation mode and producing a mass spectrum
which meets all the criteria in table 2 when 50 ng
of 4-bromofluorobenzene (BFB) is injected through
the gas chromatograph inlet.
1.3.7.4 GC/MS interface - Any gas chromatograph to mass
spectrometer interface that gives acceptable cali-
bration points at 50 ng or less per injection for
each of the parameters of interest and achieves all
acceptable performance criteria (Exhibit E) may
be used. Gas chromatograph to mass spectrometer
interfaces constructed of all-glass or glass-lined
materials are recommended. Glass can be deactivated
by silanizing with dichlorodimethylsilane.
D - 68
5/84
-------
IV.
1.3.7.5 Data system - A computer system must be interfaced
to the mass spectrometer that allows the continuous
acquisition and storage on machine readable media
of all mass spectra obtained throughout the duration
of the chromatographic program. The computer must
have software that allows, searching any CC/HS data
file for ions of a specified mass and plotting such
ion abundances versus time or scan number. This
type of plot is defined as an Extracted Ion Current
Profile (EICP). Software must also be available that
allows Integrating the abundance in any ECIP between
specified time or scan number limits.
1.4 Reagents
1.4.1 Reagent water - Regent water is defined as water in which an
interferent is not observed at the MDL of the parameters of
interest*
1.4.1.1 Reagent water may be generated by passing tap water
through a carbon filter bed containing about 453 g of
activated carbon (Calgon Corp., Filtrasorb-300 or
equivalent).
1.4.1.2 A water purification system (Hillipore Super-Q or
equivalent) may be used to generate reagent water.
1.4.1.3 Reagent water may also be prepared by boiling water
for 15 minutes. Subsequently, while maintaining the
temperature at 90°C, bubble a contaminant-free inert
gas through the water for one hour. While still hot,
transfer the water to a narrow-mouth screw-cap bottle
and seal with a Teflon-lined septum and cap.
1.4.2 Sodium thiosulfate - (ACS) Granular.
D - 69
5/84
-------
IV.
1.4.3 Methanol - Pesticide quality or eqvuivalent.
1.4.4 Stock standard solutions - Stock standard solutions may be
prepared from pure standard materials or purchased and must
be traceable to EMLS/LV supplied standards. Prepare stock
standard solutions in methano1 using assayed liquids or gases
as appropriate.
1.4.4.1 Place about 9.8 mL of methanol into a 10.0 mL tared
ground glass stoppered volumetric flask. Allow the
flask to stand, unstoppered, for about 10 minutes or
until all alcohol wetted surfaces have dried. Weigh
the flask to the nearest 0.1 mg.
1.4.4.2 Add the assayed reference material as described below.
1.4.4.2.1 Liquids - Using a 100 uL syringe,
immediately add two or more drops of
assayed reference material to the flask
then reweigh. The liquid must fall
directly into the alcohol without
contacting the neck of the flask.
1.4.4.2.2 Gases - To prepare standards for any of
the four halocarbons that boil below 30°C
(bromomethane, chloroethane, chloromethane,
and vinyl chloride), fill a 5 mL valved
gas-tight syringe with the reference
standard to the 5.0 mL mark. Lower the
needle to .5 mm above the methanol meniscus.
Slowly introduce the reference standard
above the surface of the liquid. The
heavy gas rapidly dissolves In the
methanol.
D - 70
5/84
-------
IV.
1.4.4.3 Reweigh, dilute to volume, stopper, Chen mix by
inverting the flask several times. Calculate the
concentration in mlcrograms per mlcrollter from the
net gain in weight. When compound purity is assayed
to be 96% or greater, the weight may be used without
correction to calculate the concentration of the stock
standards may be used at any concentration if they are
certified by the manufacturer. Commercial standards
oust be traceable to QiSL/LV supplied standards.
1.4.4.4 Transfer the stock standard solution into a Teflon-
sealed screw-cap bottle. Store, with minimal head-
space at -10°C to -20°C and protect from light.
1.4.4.5 Prepare fresh standards weekly for the four gases and
2-chloroethyl-vinyl ether. All other standards must
be replaced after one month, or sooner if comparison
with check standards indicate a problem.
1.4.5 Secondary dilution standards - Using stock standard solutions,
prepare secondary dilution standards in methanol that contain
Che compounds of interest, either singly or mixed together.
(See GC/MS Calibration in Exhibit E). Secondary dilution
standards should be stored with minimal headspace and should
be checked frequently for signs of degradation or evaporation,
especially just prior to preparing calibration standards from
them.
1.4.6 Surrogate standard spiking solution. Prepare stock standard
solutions for toluene-d8, p-bromofluorobenzene, and 1,2-
dichloroethane-d4 in methanol as described in Paragraph 1.4.4.
Prepare a surrogate standard spiking solution from these stock
standards at a concentration of 250 ug/10 nL in methanol.
D - 71
Rev: 9/84
-------
IV.
1.4.7 Purgeable Organic Matrix Standard Spiking Solution
1.4.7.1 Prepare a spiking solution in methanol that contains
the following compounds at a concentration of 2SO
ug/10.0 mL:
Purgeable Organics
1,1-dichloroethene
trichloroethene
chlorobenzene
toluene
benzene
1.4.7.2 Matrix spikes also serve as duplicates; therfore, add
an aliquot of this solution to each of two portions
from one sample chosen for spiking.
1.4.8 BFB Standard - Prepare a 25 ng/uL solution of BFB in methanol.
1.4.9 Great care must be taken to maintain the Integrity of all stan-
dard solutions. It is recommended that all standard solutions
be stored at -108C to -20°C in screw cap amber bottles with
teflon liners.
1.5 Calibration
1.5.1 Assemble a purge and trap device that meets the specification
in paragraph 1.3.6. Condition the trap overnight at 180°C in
the purge mode with an inert gas flow of at least 20 mL/min.
Prior to use, daily condition traps 10 minutes while back-
flushing at 180°C with the column at 220°C.
1.5.2 Connect the purge and trap device to a gas chromatograph.
The gas chromatograph must be operated using temperature and
flow rate parameters equivalent to those in paragraph 1.7.1.2
Calibrate the purge and trap-GC/MS system using the internal
standard technique (paragraph 1.5.3).
D - 72
5/84
-------
IV.
1.5.3. Internal standard calibration procedure. The three internal
standards are bromochloromethane, 1,4-difluorobenzene, and
chlorobenzene-d5.
1.5.3.1 Prepare calibration standards at a minimum of five
concentration levels for each HSL parameter. The
concentration levels are specified in Exhibit E.
Aqueous standards may be stored up to 24 hours, if
held in sealed vials with zero headspace as described
in paragraph 1.7. If not so stored, they must be
discarded after an hour.
1.5.3.2 Prepare a spiking solution containing each of the
internal standards using the procedures described in
paragraphs 1.4.4 and 1.4.5. It is recommended that
the secondary dilution standard be prepared at a
concentration of 25 ug/mL of each internal standard
compound. The addition of 10 uL of this standard
to 5.0 mL of sample or calibration standard would
be equivalent of 50 ug/L.
1.5.3.3 Analyze each calibration standard, according to
paragraph 1.7 adding 10 uL of internal standard
spiking solution directly to the syringe. Tabulate
the area response of the characteristic ions against
concentration for each compound and internal standard
and calculate response factors (RF) for each compound
using equation 1.
Ay C
-------
IV.
Where:
Ax » Area of Che characteristic ion for the compound
to be measured.
Ais " Area of the characteristic ion for the
specific internal standard from Exhibit E.
cis " Concentration of the internal standard.
Cx • Concentration of the compound to be measured.
1.5.3.4 The average response factor (RF) must be calculated
for all compounds. A system performance check oust
be made before this calibration curve is used. Five
compounds (the system performance check compounds)
are checked for a minimum average response factor.
These compounds (the SFCC) are chloromethane, 1,1-
dichloroethane, bromoform, 1,1,2.2-tetrachloroethane,
and chlorobenzene. Five compounds (the calibration
check compounds, CCC) are used to evaluate the curve.
Calculate the Z Relative Standard Deviation (ZRSD)
of RF values over the working range of the curve.
A minimum ZRSD for each CCC must be met before the
curve is valid.
ZRSD • Standard deviation x 100
mean
See instructions for Form VI, Initial Calibration
Data for more details.
1.5.3.5 Check of the calibration curve should be performed
once every 12 hours. These criteria are described in
detail in the Instructions for Form VII, Continuing
Calibration Check. The minimum response factor for
the system performance check compounds must be checked.
If this criteria is met, the response factor of all
D - 74
5/84
-------
IV.
compounds are calculated and reported. A percent
difference of the daily response factor (12 hour)
compared to the average response factor from the
initial curve is calculated. The maximum percent
difference allowed for each compound flagged as
'CCC1 in Form Vll is checked. Only after both
these criteria are met can sample analysis begin.
1.5.3.6 Internal standard responses and retention times in
all samples must be evaluated immediately after or
during data acquisition. If the retention time for
any internal standard changes by more than 30 seconds
from the latest daily (12 hour) calibration standard,
the chromatographic system must be inspected for mal-
functions and corrections made as required. If the
extracted ion current profile (E1CP) area for any
internal standard changes by more than a factor of
two (-50% to +100%), the mass spectrometric system
must be inspected for malfunction and corrections
made as appropriate. When corrections are made,
re-analysis of samples analyzed while the system
was malfunctioning is necessary. Retention time and
EICP area records shall be maintained in appropriate
form by the laboratory as a part of its internal
quality control (Exhibit E).
1.6 GC/MS Operating Conditions
1.6.1 These performance tests require the following Instrumental
parameters:
Electron Energy: 70 Volts (nominal)
Mass Range: 35 - 260
Scan Time: to give at least 5 scans per peak
but not to exceed 7 seconds per scan.
D - 75
5/84
-------
IV.
1.7.2 Sediment/Soil Samples
Two approaches may be taken to determine whether the low level
or medium level method may be followed.
o Assume the sample is low level and analyze a 5 gram sample
o Use the X factor calculated from the optional Hexadecane
screen (Section III), paragraph 1.7.2.1.3
If peaks are saturated from the analysis of a 5 gram sample,
a smaller sample size must be analyzed to prevent saturation.
However, the smallest sample size permitted is 1 gra. If smaller
than 1 gram sample size is needed to prevent saturation, the
medium level method must be used.
1.7.2.1 Low Level Method
The low level method is based on purging a heated
sediment/soil sample mixed with reagent water
containing the surrogate and internal standards.
Use 5 grams of sample or use the X Factor to determine
the sample size for purging.
o If the X Factor is 0 (no peaks noted on the
hexadecane screen), analyze a 5 gm sample.
o If the X Factor is between 0 and 1.0, analyze
a 1 gm sample.
1.7.2.1.1 The GC/MS system should be set up as in
1.7.1.2 - 1.7.1.4. This should be done
prior to the preparation of the sample
to avoid loss of volatiles from standards
and sample.
D - 80
Rev: 9/84
-------
IV.
1.7.2.1.2 Remove the plunger from a 5 mL "Luerlock"
type syringe equipped with a syringe valve
and fill until overflowing wich reagent
water. Replace the plunger and compress
the water to vent trapped air. Adjust the
volume to 5.0 mL. Add 10 uL each of the
surrogate spiking solution (1.4.6) and the
internal standard solution to the syringe
through the valve. (Surrogate spiking
solution and internal standard solution may
be mixed together). The addition of 10 uL
of the surrogate spiking solution to 5 gm
of sediment/ soil is equivalent to 50 ug/kg
of each surrogate standard.
1.7.2.1.3 The sample (for volatile organics) consists
of the entire contents of the sample con-
tainer. Do not discard any supernatant
liquids. Mix the contents of the sample
container with a narrow metal spatula.
Weigh the amount determined in 1.7.2.1 into
a tared purge device. Use a top loading
balance. Note and record the actual weight
to the nearest 0.1'gm.
1.7.2.1.3.1 Immediately after weighing the
sample weigh 5-10 g of the
sediment into a tared crucible.
Determine the percent moisture
by drying overnight at 1U5°C.
Allow to cool in a desiccator
before weighing. Concentrations
of individual analytes will be
reported relative to the dry
weight of sediment.
D - 81
Rev: 9/84
-------
IV.
Percent moisture
gm of sample-gin of dry sample
gm of sample x 10° " % moisture
1.7.2.1.4 Add the spiked reagent water to the purge
device and connect the device to the purge
and trap system. NOTE: Steps 1.7.2.1.2 -
1.7.2.1.3, prior to the attachment of the
purge device, must be performed rapidly to
avoid loss of volatile organics. These
steps must be performed in a laboratory free
of solvent fumes.
1.7.2.1.5 Heat the sample to 40°C + 1°C and purge the
sample for 12 +0.1 minutes.
1.7.2.1.6 Proceed with the analysis as outlined in
1.7.1.10 - 1.7.1.13. Use 5 mL of the
same reagent water as the reagent blank.
1.7.2.1.7 For. low level sediment/soils add 1U uL of
the matrix spike solution (1.4.7) to the 5
mL of water (1.7.2.1.2). The concentration
for a 5 gram sample would be equivalent to
SO ug/kg of each matrix spike standard.
D - 82
Rev: 9/84
-------
IV.
1.8 Qualitative Analysis
1.8.1 The target compounds listed in the Hazardous Substances List
(HSL), Exhibit C, shall be identified by an analyst competent in
the interpretation of mass spectra (see Bidder Pre-Award Labora-
tory Evaluation Criteria) by comparison of the sample mass spec-
trum to the mass spectrum of a standard of the suspected compound.
Two criteria must be satisfied to verify the identifications: (1)
elution of the sample component at the same GC relative retention
time as the standard component, and (2) correspondence of the
sample component and standard component mass spectra.
1.8.1.1 For establishing correspondence of the GC relative
retention time (RRT), the sample component RRT must com-
pare within Hh 0.06 RRT units of the RRT of the standard
component. For reference, the standard must be run on
the same shift as the sample. If coelution of interfer-
ing components prohibits accurate assignment of the sam-
ple component RRT from the total ion chromatogram, the
RRT should be assigned by using extracted ion currer
profiles for ions unique to the component of interest..
1.8.1.2 For comparison of standard and sample component mass
spectra, mass spectra obtained on the contractor's GC/
MS are required. Once obtained, these standard spectra
may be used for identification purposes, only if the
contractor's GC/MS meets the daily turning requirements
for BFB or DFTPP. These standard spectra may be
obtained from -the run used to obtain reference RRTs.
1.8.1.3 The requirements for qualitative verification by
comparison of mass spectra are as follows:
(1) All ions present in the standard mass spectra at
a relative intensity greater than 10 % (most abundant
ion in the spectrum equals 100%) must be present in
the sample spectrum.
D - 86
Rev: 9/84
-------
IV.
(2) The relative intensities of ions specified in (1)
must agree within plus or minus 20% between the stan-
dard and sample spectra. (Example: For an ion with
an abundance of 50% in the standard spectra, the
corresponding sample abundance must be between 30
and 70 percent).
(3) Ions greater than 10% in the sample spectrum but
not present in the standard spectrum must be consid-
ered and accounted for by the analyst making the
comparison. In Task III, the verification process
should favor false negatives.
1.8.2 A library search shall be executed for Non-HSL sample components
for the purpose of tentative identification. For this purpose,
the most recent available version of the EPA/NIH Mass Spectral
Library shall be used. Computer generated library search rou-
tines should not use normalization routines that would misrepre-
sent the library or unknown spectra when compared to each other.
1.8.2.1 Up to 10 substances of greatest apparent concentra-
tion not listed in Exhibit C for the purgeable organic
fraction shall be tentatively Identified via a forward
search of the EPA/NIH mass spectral library. (Sub-
stances with responses less than 10% of the internal
standard are not required to be searched in this
fashion). Only after visual comparison of sample
spectra with the nearest library searches will the mass
spectral interpretation specialist assign a tentative
identification.
1.8.2.2 Guidelines for making tentative identification: (I)
Relative intensities of major ions in the reference
spectrum (ions greater than 10% of the most abundant
ion) should be present in the sample spectrum.
D - 87 Rev: 9/84
-------
IV.
(2) The relative intensities of the major ions should
agree within + 20%. (Example: For an ion with an
abundance of 50 percent of the standard spectra, the
corresponding sample ion abundance must be between 30
and 70 percent.)
(3) Molecular ions present in reference spectrum
should be present in sample spectrum.
(4) Ions present in the sample spectrum but not in
the reference spectrum should be reviewed for possible
background contamination or presence of co-eluting
compounds.
(5) Ions present in the reference spectrum but not in
the sample spectrum should be reviewed for possible
subtraction from the sample spectrum because of back-
ground contamination or co-eluting compounds. Data
system library reduction programs can sometimes
create these discrepancies.
1.8.2.3 If In the opinion of the mass spectral specialist,
no valid tentative identification can be made, the
compound should be reported as unknown. The mass
spectral specialist should give additional classif-
ication of the unknown compound, if possible (i.e.
unknown aromatic, unknown hydrocarbon, unknown acid
type, unknown chlorinated compound). If probable
molecular weights can be distinguished, include them.
1.9 Quantitative Analysis
1.9.1 HSL components identified shall be quantified by the internal
standard method. The Internal standard used shall be the one
nearest the retention time to that of a given analyte. The
D - 88
5/84
-------
IV.
EICP area of Che characteristic ions of analytes listed in
Tables 2 and 3 are used. The response factor (RF) from the
daily standard analysis is used to calculate the concentration
in the sample. Use the response factor as determined in para-
graph 1.5.3.3 and the following equations:
Water (low and medium level)
(AX)(IS)
Concentration ug/L - (Ais)(RF)(VQ)
Where:
Ax • Area of the characteristic ion for the compound to be
measured
Ais ™ Area of the characteristic ion for the specific internal
standard from Exhibit E.
Is • Amount of internal standard added in nanograms (ng)
VQ *> Volume of water purged in milliliters (mL) (take into
account any dilutions)
Sediment/Soil (medium level)
Concentration ug/kg • (AX)(IS)(VC)
(Ais)(RF)(Vi)(Ws)(D)
Sediment/Soil (low level)
Concentration ug/kg • 'Ax)(Is)
(Als)(RF)(Ws)(D)
(Dry weight basis)
Where:
AX, Is, Ais = same as for water, above
Vc • Volume of total extract (uL) (use 10,000 uL
or a factor of this when dilutions are made)
V^ « Volume of extract added (uL) for purging
D • 100 - % moisture
100
W8 • Weight of sample extracted (gra) or purged
D - 89
Rev: 9/84
-------
IV.
1.9.2 An estimated concentration for Non-HSL components tentatively
identified shall be quantified by the internal standard method.
For quantification, the nearest internal standard free of inter-
f ereces shall be used.
1.9.2.1 The formula for calculating concentrations is the
same as in paragraph 1.9.1. Total area counts from
the total ion chromatograms are to be used for both
the compound to be measured and the internal standard.
A response factor (RF) of one (1) is to be assumed.
The value from this quantitatlon shall be qualified
as estimated. This estimated concentration should be
calculated for all tentatively Identified compounds
as well as those identified as unknowns.
1.9.2.2 Xylenes (o,m, & p - isomers) are to be reported as
total Xylenes. Since o- and p-Xylene overlap, the
•
Xylenes must be quantitated versus m-Xylene. The
concentration of all Xylene isomers must be added
together to give the total.
1.9.3 Calculate surrogate standard recovery on all samples, blanks
and spikes. Determine if recovery is within limits and report
on appropriate form.
1.9.3.1 Calculation for surrogate recovery.
Percent Surrogate Recovery • Qd_ X 100%
where: Qd = quantity determined by analysis
Qa • quantity added to sample
D - 90
Rev: 9/84
-------
IV.
1.9.3.2 If recovery is not within limits, the following is
required:
o Check to be sure there are no errors in calcula-
tions, surrogate solutions and internal standards,
Also, check instrument .performance. .
o Recalculate the sample data if any of the above
checks reveal a problem.
o Reanalyze the sample if none of the above are a
problem.
o Report the data from both analyses along with
the surrogate data from both.
Table 2
Characteristic Ions for Surrogate and
Internal Standards for Volatile Organic Compounds
Compound . Primary Ion Secondary Ion(s)
SURROGATE STANDARDS
4-Bromofluorobenzene 95 174, 176
1,2-Dichloroethane d-4 65 102
Toluene d-8 98 70, 100
INTERNAL STANDARDS
Bromochloromethane 128 49, 130, 51
1,4-Difluorobenzene 114 63, 88
Chlorobenzene d-5 117 82, 119
D - 91
Rev: 9/84
-------
IV.
Table 3
Characteristic Ions for Volatile HSL Compounds
Parameter
Primary Ion*
Secondary Ion(s)
Chlorooe thane
Bromomechane
Vinyl chloride
Chloroethane
Methylene chloride
Acetone
Carbon disulfide
1 , 1-Dichloroethene
1 , 1-Dichloroethane
trans-1 , 2-Dichloroethene
Chloroform
1 , 2-Di chloroethane
2-Butanone
1,1, 1-Trichloroethane
Carbon tetrachloride
Vinyl acetate
Bromodichlorome thane
1 , 1 , 2,2-Tetrachloroethane
1 , 2-Oichloropropane
trans-1 ,3-Dichloropropene
Trichloroethene
Dibromochlorome thane
1 , 1 ,2-Trichloroethane
Benzene
cis-1 ,3-Dichloropropene
2-Chloroethyl vinyl ether
Bromof orm
2-Hexanone
4-Wethyl-2-pentanone
Tetrachloroethene
Toluene
Chlorobenzene
Ethyl benzene'
Styrene
Total xylenes
50
94
62
64
84
43
76
96
63
96
83
62
72
97
117
43
83
83
63
75
130
129
97
78
75
63
173
43
43
164
92
112
106
104
106
52
96
64
66
49, 51, 86
58
78
61, 98
65, 83, 85, 98, 100
61, 98
85
64, 100, 98
57
99, 117, 119
119, 121
86
85, 129
85, 131, 133, 166
65, 114
77
95, 97, 132
208, 206
83, 85, 99, 132, 134
-
77
65, 106
171, 175, 250, 252, 254, 256
58, 57, 100
58, 100
129, 131, 166
91
114
91
78, 103
91
* The primary ion should be used unless interferences are present, in which
case, a secondary ion may be used.
D - 92
Rev: 9/84
-------
IV.
O.D. exit
„ Sample Inlet
— 2-wey Syringe valve
;•— 17cm 20 gtuge syringe needle
6mm O.D. Rubber Septum
'/„ in. O.D.
Steinlett Steel
13X moleculer
purge
get fitter
Purge get
flow control
10mm glett Ml
medium porosity
Figure 1. Furginf device
Pecking procedure
Construction
Glett
WOOI
Grede IS
Silice gel 8cm
Tone* 15cm
3% QV-1 1em
Glett Smm
wool
inltt
Cemprettion fining
nut end ferrulet
14ft 7-^foot retittence
wire wropped solid
Thermocouple/contnHtt
tensor
Tubing 25 cm.
0.105 in. I.D.
0.1 25 in. O.D.
tteinless tteel
2. Trtp peckingt end comtruction'to include detorb
D - 93
5/84
-------
IV.
Cirrior got flow control Ltqutd mioction pont
Prouun rogulitor
Purgt got
flow control \
13X motffutir
t fiftor
Column ovon
j"jJU"LP—^- ,-— Confirmatory column
rUl/lP") ~-^Analytical column
x optional 4-port column
to/action valve
Trap inlet
Rasistanca win
Trip (OH
22°C
control
Purging
Note:
All linti btrw+on
trip ind CC
ihould bt
to 80°C
Rgur* 3. Schtrmsic of puff* »nd trip dtviet — purgo mod*
g*i fhw controt L«urt inaction pent Cf>/umn
13X mol»cul»r
fillff
til, n "I "1 ~L-1.— Confirmitorf column
i|J.LnUnunJ_>r0rf.f.cfor
optional 4-port column
talaction vatvo
6-port fftp in/tt
v-/v* ^J Rasistanca wira
/ y*^^ ^ ^ Honor control
TrapM- me I On
flow] WC ^ ^
ffota:
AH lino*
trip and CC
ihould oo
to 95'C
Purging
dovico
4. $cf*m*itc of pvrgo ond trip dovico — dotorb mod*
D - 94
5/84
-------
IV.
PURGE INLET FITTING
SAMPLE OUTLET FITTING
3" • 6mm 00. GLASS TUBING
SEPTUM
CAP
Figure 5. Low Soils Impinger
D - 95
5/84
-------
APPENDIX C
ESTABLISHED U.S. EPA ADVISORY LIMITS
FOR PRECISION AND ACCURACY AND METHOD
PERFORMANCE LIMITS FOR ANALYTICAL PROCEDURES
-------
TABLE 1. SUMMARY OF PRECISION AND ACCURACY ADVISORY LIMITS
SET BY THE U.S. EPA CONTRACT LABORATORY PROGRAM
Volatile*
1,1-dichloroethene
trichloroethene
chlorobenzene
toluene
benzene
Semi vol at lies
1 , 2, 4-tr ichl orobenzene
acenaphthene
2,4-dinitrotoluene
di-n-butylphthalate
pyrene
N-nitroso-di-.n-propylamine
1 , 4-dichl orobenzene
pentachlorophenol
phenol
2-chlorophenol
4-chl oro-3-methyl phenol
4-nitrophenol
lindane
heptachlor
aldrin
dieldrin
endrin
4,4'-DDT
Soil
Maximum RPO Between
Duplicates
22
24
21
21
21
23
19
47
47
36
38
27
47
35
50
33
50
50
31
43
38
45
50
/Sediments
Percent Recovery
of Matrix Spikes
59-172
62-137
60-133
59-139
66-142
38-107
31-137
28-29
29-135
35-142
41-126
28-104
17-109
26-90
25-102
26-103
11-114
46-127
3*5-130
34-132
31-134
42-139
23-134
-------
TABLE 2. DFTPP MASS-INTENSITY SPECIFICATION
Mass
Intensity Required
51 30-60% of mass 198
68 Less than 2% of mass 69
70 Less than 2% of mass 69
127 40-60% of mass 198
197 Less than 1% of mass 198
195 Base peak, 100% relative abundance
199 5-9% of mass 198
275 10-30% of mass 198
365 1% of mass 198
441 Less than mass 443
442 Greater than 40% of mass 198
4-43 17-23% of mass 442
TABLE 3. BFB MASS-INTENSITY SPECIFICATION
Mass
Intensity Required
50
75
95
96
173
174
175
176
177
15-40% of mass 95
30-60% of mass 95
Base peak,
100% relative abundance
5-9% of mass 95
<2% of mass 174
>50% of mass 95
5-9% of mass 174
>95% but <101% of mass 174
5-9% of mass 176.
-------
Federal Register / Vol. 49. No. 209 / Friday. October 26. 1984 / Rules and Regulations 181
TABLE 5.—ACCEPTANCE CRITERIA FOR PERFORMANCE TESTS
Acceptance cntena il 20 pg/L
Compound
Inut pieealon and accuracy
Mclion623
••(Mtf/L)
Acetone
Acrolem ._
Acrytonrtnle . . .
Beniene
BromodKNorornethane .. .
Bromolorm
Sromomeihane ....
-:*-*«? •-•'•• -
Methyl ethyl ketone
i. i .2.2-tetrechlcfoeihane
Tatrachtoroetfiene - .
Toluene . . .
1.1.1-inchloroathana
1,1 2'tncMoroothtVM
Tnchloroalhene .
VnylcNonde ..
•e-s
70
'82
to'.
74<
1&2
if*
- »a
"~88
83
59
71
89
279
notel
note 2
note 2
130-282 ne-198
65-315 na-199
7 4-35 I ne-214
0-543 ftt-414
159-246 42-165
142-298
21-487
0-898
116-283
d-S55
1IJ-291
114-314
116-301
tf-498
10 5-31 5
0-488
0-510
0-402
note 1
note 1
156-285
0-498
now 1
107-300
151-285
145-287
105-334
118-297
168-295
0-585
ra-20S
nt-308
IV-SS4
te-172
nt-410
16-165
23-191
12-192
n*-31S
15-195
m-343
n»-381
n»-203
ne-316
5-199
31-181
4-193
12-200
21-184
35-198
m-4S2
4-33
4-34
6-36
0-81
12-30
4-35
0-51
0-79
6-30
0-64
8-32
9-33
6-33
0-52
8-34
0-51
0-58
0-44
5-35
0-50
7-34
11-32
6-33
6-35
9-32
12-34
0-65
0-detectod: reaull mai be greater inan sere
ra.no ipecrnubon. km would be below detection tnM.
Note 1: Spaaficitxxu not ivvlable tor thew compounds tt kme of releaM of tha melnod
Not* 2: Speofictliona not OevetopeO Iw these compounds, use method 603
BILLHM CODE 6580-90-M
-------
Federal Register / Vol. 49. No. 209 / Friday. October 26. 1984 / Rules and Regulations 193
TT3—r-"r*
.sa^'-iE
'•£-*€-
TABLE B —ACCEPTANCE CRITERIA FOR PERFORMANCE TESTS
i-, . \ -
EODNoJ
-.l, Ceavoana
Inbil pracaiifit and ueuiacy
•action 8 2.3 faig/L)
83
(pavanQ.
CalaVaGon
vxrfciuonMC
XI
2OI
377
277
376
278
305
205
372
272
374
274
375
275
373
273
379
279
712
Acanapnoim*
Ac«n«pricnyl«f~H«a .,
Anttvacan*..
AntnracanMIO -
Bannkna
Banntncda .
8*nxo(gh4p«ylenc> ... _
6«nzo
l.3-*n»natd4
21
38
38
31
41
49
lit
269
20
41
181.
168
'. 26
"A
24
21
45
41
43'-
34
33
27
17
27
31
29
44
31
51
70
74
sr
109
33
46
39
59
34
31
11
26
35
35
32
41
M
100
41
37
1VI
13
24
42
52
51
, 6»
18
•7
55
20
31
31
31
15
23
17
3*5
43
46
'42
28
60
12
44
78
"22
,36
toe
16
66
16
37
30
59
79-134
38-147
69-186
38-146
58-174
31-194
16-518
na-na
85-188
25-296
3*-9»* :_ .
11-577
20-270
23-739
na-na
1M05
62-195
35-181
72-180
29-268
7S-I48
28-165
55-166
29-198
43-1 S3
• 1-138
35-149
69-220
32-205
44-140
19-233
24-195
m-298
35-389
n»-331
.ra-MS
•80-162
37-182
42-131
53-263
34-172
45-152
80-139
27-211
35-193
35-183
6' -200
27-242
35-163
48-357
30-168
78-131
30-174'
79-135-
38-182
75-168
4O-I81
14-5M
15-372
16-384
18-308
19006
13479
"i 5-324
33-219
78-140
m-159
23-199
85-138
47-138
7 9- ISO
48-F30
76-1Q5
23-195
7»-t48
14-212
83-201
13-203.
23-255
19-325
13-512
28-220
~29-S1S
"*r"" *"" ~" "•Vt'ili
68-174
nt-562
85-131
38-164.
75-198
m-260
na-494
m-SSO
na-474
na-na
F34"'
n-
22-308
75-158
22-245
8O-I41
44-184
80-125
71-141
60-188
88-152
80-168
58-171
34-298
m-na
70-142
28-357
61-16*
J4-na
, 13-TB
76-129
12-na
•9-145
11-na
58-171
52-192
•1-164
52-194
44-228
67-148
44-228
78-131
43-232
52-193
22-450
42-235
44-227'
60-186
41-242
37.288
72-138
54-188
40-249
54-1 Ml
62-162
40-249
65-154
5O-199
28-392
28-392
68-152-
24-423,
44-227
58-171
72-139
85-115
88-147
78-129
55-180
71-142
57-175
70-142
24-411
79-127-
86-152
13-781
73-136
66-150
72-140
69-145
71-142
52-192
74-135
61-164
65-154
52-192
62-161
65-153
77-130
18-SS8
64-157
74.1H
47-211
67-150
58-172
71-117
50-201
75-113
39-256
19-127
53-187
55-183
36-278
72-144
30-1 80
61-207
33-168
50-199
23-242
11-872
na-na
62-176
22-329
20-na
53-155
na-68S
59-206
32-194
SO-166
25-303
62-178
17-267
50-213
25-222
39-168
77-145
30-I8S
64-212
28-224
15-172
35-170
19-237
na-504
29-424
na-408
71-161
28-202
35-187
46-301
29-198
39-195
78-142
25-22*
31-212
31-212
56-215
23-274
31-188
35-442
24-204
82-159
14-314
76-136
33-176
83-194
29-212
48-221
23-290
72-147
19-340
79-146
39-160
70-186
40-158
74-169
22-209
70-152
11-247
55-225
na-260
53-219
11-245
64-185
na-na
83-135
34-182
65-222
na-na
80-156
14-J42
67-207
na-na
68-141
17-378
72-164
19-275
70-119
31-250
-------
This page was- extracted from the U.S. EPA Contract Laboratory Program
(CLP) S&tmentf Q&tork-ifor mul turnedq~3i multi-concentration organics analyses,
May 1984, revised-'January 1985.
When the surrogate recovery of any one surrogate compound is
the contract required surrogate recovery limits (listed in Table 4.20
reagent blank, the' laboratory"1 must take the following actions:
TABLE 4.2..- CONTRACT REQUIRED SURROGATE SPIKE RECOVERY LIMITS
Fraction
VOA
VOA
VOA
BNA
BNA
BNA
BNA
BNA
BNA
Pest.
Surrogate Compound
Toluene-dg
4-Bromo£luorobenzene
1 ,2-Dichl6roethane-d4
Nitrobenzene-ds
2-Fluordbiphenyl
p-Ter phenyl-d ^4
Phenol-ds
2-Fluorophenol
2,4,6-Trlbromophenol
Dibutyldhlorendate
Low/Medium ,.i*a*H**aMfui*
Water ' S6I3>/5*d3iiLeat=
86-119 5S-16O
85-121 50-ldO
77-120 5^15
41-120 iU^WPgJ.
44-119 7p*Wf9
33-128 2d-t2>0
15-103 20.^X2^'
23-121 2"0'-iT«Ffi(,
10-130 10-t4fl=
( 48-136 )* £2&tSdS$*>;
* These 'limits are for advisory purposes only. They are not used to
if a sample should be reanalyzed. When sufficient data becomes avant&abi-ev
the USEPA may set performance based contract required windowc.
4.4.1.1 Check calculations to assure there are no errors; cKecfc,4jaiSr
ternal standard and surrogate spiking. ^solutions for degradation,
etc; also, check instrument -performance.
4.4.1.2 Recalculate otl rein ject/repurge the blank or extract
in 4.4.1.1 fail to reveal t!)^ 'cause of the non-compliant surrogate
4.4.1.3 Re-extract and reanalyze the blank.
4.4.1.4 If the measures listed in 4.4.1.1 thru 4.4.1.3 fall t.
the problem, the analytical ;system mu£t be considered out of control^
problem MUST be corrected before continuing.
E-30
9/84 Rev
-------
APPENDIX D
GC/MS IDENTIFICATION OF TARGET AND LIBRARY SEARCH COMPOUNDS
-------
This appendix was extracted from the U.S. EPA Contract laboratory
Program (CLP) Statement of Work for multi-media, multi-concentration organics
analyses, May 1984, revised January 1985. Guidance is provided for qualitative
identification of target compounds by GC/MS and tentative identifications
of non-target compounds identified by a spectral library search.
2.6 Qualitative Analysis
2.6.1 The target compounds listed in the Hazardous Substances List
(HSL), Exhibit C, shall be identified by comparison of the sample
mass spectrum to the mass spectrum of a standard of the suspected
compound. Two criteria must be satisfied to verify the identifi-
cations: (1) elution of the sample component at the same GC rela-
tive retention time as the standard component, and (2) correspond-
ence of the sample component and standard component mass spectra.
2.6.1.1 For establishing correspondence of the GC relative
retention time (RRT), the sample component RRT must
compare within +0.06 RRT units of the RRT of the
standard component. For reference, the standard must
be run on the same shift as the sample. The RRT
should be assigned by using extracted ion current
profiles for Ions unique to the component of interest.
D - 101
Rev: 9/84
-------
IV.
2*6.»JU2 .For comparison-.of.standard and sample coaponenc mass
spejpjra,. mass spectrapbtalned on Che contractor's
•G£/MSt are,, jresuire.d. Once obtained, these standard
spectra, jnay. be used fo*_ Identification purposes, only
.if the contractor's GC/MS meets the DFTPP daily tuning
i-pquireoienJts. Iftese standard spectra may be obtained
from £he_juji _used_.to obtain ^reference KRTs.
2..6.1.2.1 The requirements for qualitative verifica-
tion Joy comparison of mass spectra are as
follows:
Cl)LAir iWs1 present in the standard mass
spectra at a relative intensity greater than
I'OZ^Cmbst aBifiia'a'nt loft'ln'the spectrum equals
1002) must be,present in the sample spectrum.
t2VThe Tre'latlve intensities of ions speci-
fied in ti; must agree within plus or minus
2,0% between.the standard and sample spectra.
»V : .
(Exa.mple:.,For _an ion with an abundance of
50% in- the standard spectra, the corres-
DOnd.ing sampJLe ion. abundance must be between
30 and /U percent.)
(3) Ions greater than 10% in the sample
spectrum but not present in the standard
spectttim-oust'tie considered and accounted
fort by s the-.analyst making the comparison.
slrisTaskoIItyothe verification process
should favor false negatives.
* 9 '
2.6.2 A library search shall »e executed for Non-HSL sample components
for the purpose of'tentative' Identification. For this purpose,
the most recent available version or the EPA/NIH Mass Spectral
Library should be used.
D - 102
Rev: 9/84
-------
IV.
2.6.2.1 Up Co 20 substances of greatest apparent concentration
not'Hs'tea~iiT Exhibit'TTTor* Che combined base/neutral/
aci'd traction" shall-be tentatively identified via a
forward searcir~ol~the""EFA7NT& mass spectral library.
TSub'stah'ceVVltff reSpo^se's'less than 10% of the nearest
Internal standaxxi'are riot'required to be searched in
this TashTohT. OhTy"aTt'er""vlsual comparison of sample
spectra with the riearesrt liurary searches will the
mass spectral interpretation specialist assign a ten-
tative Tde"ntificatidn." C6mp"uter generated library
searcti routines should not use normalization routines
that would mTsYepresent the library or unknown spectra
whe.n9c,p,apai?ed,tq each other.
Qu.id£liQes f,or.^making tentative Identification:
'(1) Re4£££ve intensities of major ions in the reference
spectrum Xi-pns greater than 102 of the most abundant
ion) shpulg be present in the sample spectrum.
(^)'Thi' relative intensities of the major ions should
'agree'wlttiTh7^ 20%. (Example: For an ion with an abun-
dance o'f' SOX In' the standard spectra, the corresponding
sample ion a~ounaance must be between 30 and 70 percent.
(3) Molecular ions present in reference spectrum should
be present in sample spectrum.
€*>etoiaioraejS^At in the sample spectrum but not in the
ieferericexs'pecferura should be reviewed for possible back-
groundlccfntSmin'ation or presence of coeluting compounds.
(5) Ions present in the reference spectrum but not in
the sample spectrum should be reviewed for possible
subtraction.from the sample spectrum.because of back-
gr,oundv contamination or coeluting compounds. Data
system library reduction programs can sometimes create
these discrepancies.
D - 103 Rev: 9/84
-------
IV.
If iftoduvcysnieft of. the mass spectral specialist,, no
'tnnr.'ulv* i-'-ntificatiou C/*A be made,, the compound
:lc ba.ref•;'*'*' gs unknown. Vhe naas spectral specJ.al-
iv •{•-fsiir.ion«i!t^6-*aai.iftcati-6h'of the unknown
(•?. • J.jmhlu;oi/n phthalar.e, unknown
cidI••c^ f-.-,.;-*V%•/.-:c ehlcvir,at9d
tx»"i.c-ilarr'weichts can b* dieting-
JL:^.!. y, ineiiide
------- |