SUPERCRITICAL FLUID EXTRACTION OF ORGANIC
COMPOUNDS FROM VARIOUS SOLID MATRICES
Susan Warner> Carole Tulip, Philip Shreiner and Joseph Slayton
USEPA Central Regional Laboratory, 839 Bestgate Road, Annapolis, MD,21401
INTRODUCTION
Supercritical fluid extraction (SFE) utilizes compounds such as
carbon dioxide at supercritical temperatures and pressures. The
supercritical fluid has properties of a liquid (solvating power)
and the properties of^ a gas (low viscosity and high diffusivity) .
The analysis of soil and sediment samples are routinely performed
using either Soxhlet extraction or sonication. Both procedures
utilize large quantities of organic solvents (methylene chloride,
hexane, and acetone) . These solvents are expensive to purchase and
dispose of properly. In addition, these procedures are time
consuming and tedious. Supercritical fluid extraction is rapid
(minutes as opposed to hours), and uses very little solvent (a few
mLs as opposed to hundreds of mLs). This technique employs small
quantities of harmless gases, e.g., carbon dioxide, to extract the
sample and collect target compounds in small volumes of solvent
(about 10 mLs).
Pawliszyn of the University of Waterloo in Ontario, Canada
concludes that two steps are necessary to accomplish rapid
extraction. First, the conditions need to be such that the analytes
are not retained on the matrix. This can be done by increasing
their solubility in the extraction fluid as compared to the matrix.
Secondly, the rapid transfer of the analytes from the matrix to the
fluid must be ensured. This transfer can be accomplished during
static extraction.
Once the analytes are extracted from the matrix, the choice of
trapping media can play a significant role in the accuracy and
precision of the results. Different trapping approaches are
commonly used. These include cryogenic, absorbent bed, and liquid
trapping.
This work was funded by USEPA Office of Research and Development.
The authors recognize the contribution of Angela Cogswell, NNEM
fellowship student, on the preliminary studies performed with the
Supercritical Fluid Extractor.
-------
The SFE technique has been available for several years. The
Agency's use of this technique has been hampered by the extensive
scope of work necessary to develop this technique for use in the
many possible environmental matrices, and for use in extracting the
many target analytes mandated by the Agency's various programs. An
Office of Research and Development Leopard Team (a group of
specialists) outlined a "Strategy Plan" for the development of SFE
(October 31, 1992). Section 4.18 of this plan indicates that a
method protocol for the extraction of Base/Neutral/Acidic compounds
from soils would be developed and that this project is planned to
be conducted in FY '95. In discussions with the Methods Research
Branch, USEPA EMSL-Las Vegas, the complexity and extensive nature
of the effort involved to overcome the development obstacles could
benefit from a coordination of research efforts between ORD and
EPA'S Regional laboratories (EMSL-LV and CRL Region III). Work on
the extraction of Acidic and Base/Neutral compounds done by CRL
could serve as the framework upon which EMSL-LV could build
additional complementary capacities (more target compounds) and
further improvements.
This work was part of a general effort by the USEPA Central
Regional Laboratory (Region III) to minimize the solvent necessary
for extraction of semi-volatile compounds. A series of experiments
were conducted to optimize various operating conditions for SFE.
These conditions included varying the pressure, flow, extraction
modifiers, time, and temperature to optimize the recovery of the
target compounds. One of the goals was to determine if there were
a set of conditions that would extract all semi-volatile compounds
(NPDES, Superfund) which vary widely in physical and chemical
characteristics. Extractions were first attempted at low tempera-
tures and pressures. These parameters were increased during the
progress of this study in order to improve extraction recoveries.
The initial phase of this work involved the recovery of the target
compounds spiked onto diatomaceous earth. This material has become
a commonly used matrix in published work on SFE. It is inexpensive,
clean and absorbent. Modifier was added and extractions were
performed at various temperatures and pressures. The extractions
were directly performed off-line (not coupled to a detector).
GC/FID, GC/ECD or GC/MS analyses were performed to determine
efficiency of recovery as part of the method validation. Spiked
cocktails containing a variety of target compounds were employed.
During this on-going process various modifications were made in the
extraction procedures to determine their effect on the efficiency
of recovery. Work was also done on recovering spikes from soils
and similar materials, as well as the extraction of reference soils
(known PNA contaminant levels) for comparison to Soxhlet extrac-
tions. Generally, compounds that1were difficult to extract from
diatomaceous'earth exhibited the same properties during extraction
from the contaminated soil used in the study. Other researchers
have found that native analytes present on solid matrices have
-------
different extraction rates than those spiked onto the matrices.
The authors are aware that some matrices may be more difficult to
extract than others and that spiked sample recoveries may not
reflect those of native analytes. However, reference soils are not
available for the great majority of semi-volatile target compounds.
DISCLAIMER
Although the research described in this document has been supported
by the U.S. Environmental Protection Agency and is awaiting Agency
wide review, it does not necessarily reflect the views of the
Agency, and no official endorsement should be inferred. Mention of
trade names or commercial products in this report is for illustra-
tive purposes and does not constitute endorsement or
recommendation.
MATERIALS AMD APPARATUS
THERMOLYNE MUFFLE FURNACE was used to clean glassware at a
temperature of 450 C for 8 hours.
CONCENTRATION OF EXTRACTS: used 10 mL concentrator tubes with a
19/22 joint (Labglass, Vineland N.J.) and 3 ball micro Snyder
columns.
COLLECTION VIALS USED: Screw cap vials with teflon faced silicone
septa, 1.8 mL, used to store references and extracts (Supelco #2-
3277; screw cap vials with hole caps and septa, 5 mL (Supelco #2-
3249)and 7.4 mL (Supelco #2-3218)used to collect extracts; and
Environmental Concentrating Autosampler vials (Supelco #3-3255)
used in analysis.
DRUMMOND WIRETROL disposable micropipettes; 10 uL, 50 uL, and 100
uL were used to dispense spike material, add modifiers, and make
reference solutions.
CLASS A volumetric flasks: 1.0 mL, 5.0 mL and 10 mL were used to
prepare standards and bring extracts to volume.
BRANSONIC 52 sonicator was used to clean extraction vessels and
frits.
SUPERCRITICAL FLUID EXTRACTION SYSTEM [Figure 1] The system used
was a Suprex Model MPS 225 (Multipurpose SFE-SFC system) . The SFC
was not employed during this work. The system consisted of: a pump
module containing a 250 mL syringe pump, phase valves etc.; control
-------
module, containing electronic circuitry, CRT, keyboard necessary to
control the system; and an oven module containing a GC oven,
extraction vessel and two valves, a 10 and a 4 port valve. The
modules were connected by a mobile phase fluid line from the pump
module to the oven module and electronic control and power cables.
The extraction vessels used were one and one-half mL stainless
steel Suprex EX62010, maximum pressure of 500 atm. In the initial
phase of this work a restrictor was employed to collect the
extracted target compounds in methylene chloride. The restrictor
was 40 urn ID fused silica approximately 40 cm in length. The
extracted material was collected by insertion of the restrictor
through the septum of a 7.4 mL screw cap vial (Supelco #2-3218)
into 5 mL of methylene chloride below the solvent surface.
Methylene chloride was used as a collection solvent in all
extractions.
FIGURE 1
DIAGRAM OF 8FE FLOW SYSTEM
In a later phase of this work a Suprex Accutrap™ was added to the
system TFigure 2]. This unit consisted of: a variable restrictor
of PEEIT tubing, a heated restrictor module, a cryogenic collection
trap and a collection system. During the dynamic extraction
phase, the variable restrictor, heated to 50 C, allowed the
extraction gases to pass through the SFE system at flow rates of
0.5 to 2 mL per minute. The cryogenic trap, cooled to -48 C with
carbon dioxide, captured the target compounds onto glass beads. The
trap was heated to 30 C during the desorption phase and was flushed
with a measured amount of methylene chloride, which passes into a
collection vial. The trap was then purged with nitrogen to clear
out any remaining solvent.
-------
NTSeure*
FIGURE 2
-TH
DIAGRAM OF FLOW WITH CRYOGENIC TRAP
The Accutrap1" system as provided by the manufacturer was altered
by extending the stainless steel delivery tube further into the
collection vessel below the solvent surface [Figure 3].
FIGURE 3
COMPARISON OF COLLECTION SYSTEMS
GAS CHROMATOGRAPH. An HP 5890 Series II gas chromatograph was
equipped with: an electron capture detector; a flame ionization
detector; an HP 7673 Automatic sampler; and an HP 3365 Chemstation
data system. The FID was equipped with a Supelco SPB-5 fused
silica capillary column, 60 meters in length, 0.32 mm ID with a
film thickness 0.25 urn (catalogue # 2-4050). The BCD was equipped
with a Supelco SPB-608, fused silica capillary column of 30 m,
0.53 mm ID with a film thickness of 0.50 urn. For FID analysis, the
GC was programmed from 50 C to a final temperature of 280 C at 5
C/minute with a final holding time of 10 to 25 minutes. The (-JC
temperature program for the pesticides (BCD) was from 150 C to 280 C
at 10 degrees /minute with a final holding time -of 10 minutes.
GAS CHROMATOGRAPH-MASS SPECTROMETER. The Finnigan MAT 4500 was
equipped with: a quadrupole analyzer and El source; an HP 7673
automatic sampler and an Incos data system. The FSCC column was a
DB-5, J&W Scientific, 30m x 0.32 mm with a film thickness of 1 urn.
The GC temperature program was: 30 C for 2 minutes, ramped to 300 C
at 10 C/minute.
-------
STANDARDS and SPIKING MATERIAL. Analytical reference standards for
the Base/Neutral Extractables, Pesticides, Toxic Substances and
Phenol Mixtures were obtained from AccuStandard, Inc., New Haven,
CT. The following compounds are now available through EMSL-CI, but
were originally obtained through the EPA Quality Assurance
Materials Bank (Analytical Reference Standards, Las Vegas, Nevada):
Toxaphene; Chlordane; Naphthalene; PCBs; 2,4,6- Tribromophenol; p-
Terphenyl-du; Organic surrogate mix; 1,4- Dichlorobenzene; N-
Nitrosodi-n-propylamine; 2,4-Dinitrotoluene; Di-n-butyl Phthalate;
Acenaphthene; 1,2,4-Trichlorobenzene; Pyrene; 4-Nitrophenol; 2-
Chlorophenol; Phenol; 4-Chloro-3-methylphenol; Pentachlorophenol;
Benzo(a)pyrene; Fluoranthene; Indeno(l,2,3-c,d)pyrene; Bis(2-
chloroiso-propyl) ether; and 2-Chloroethyl vinyl ether.
SUPERCRITICAL FLUIDS. SFC grade carbon dioxide, with helium
headspace and a dip tube, was obtained from Scott Specialty Gases,
Inc. of Plumsteadville, Pa. SFC grade carbon dioxide with ten
percent methanol obtained through Scott was also used in some
extractions.
SAMPLE MATRICES. Matrices used included: diatomaceous earth, PAH-
contaminated soil SRS 103-100 (Fisher Scientific), NIST SFE Round
Robin Exercise sediments and a test soil containing a large
percentage of clay which had been dried, ground and sieved.
EXTRACTIONS OF DIATOMACEOUS EARTH WITH FUSED SILICA RESTRICTOR
All extractions were performed by adding spike and modifier to
approximately 0.2 g. diatpmaceous earth in the bottom of the
extraction vessel. The solvent was not allowed to evaporate before
the modifier was added and the spike solvent probably acted as an
additional modifier. The sample was then covered with a glass
fiber filter (to eliminate any possible dead space and to prevent
any diatomaceous earth from passing through the frit into the
restrictor). The supercritical fluid containing the extracted
material passed through the restrictor and bubbled into the
methylene chloride solvent in the collection vial. Upon completion
of the extraction, the collection solvent was adjusted to 1.0 mL.
in class "A" volumetric flasks either by evaporation using a micro
Kuderna-Danish apparatus and/or by diluting with methylene
chloride.
Compound spike recoveries were computed against the response of a
reference standard prepared the. same day that the samples were
extracted. The reference standards and extracts were analyzed on
the same GC/FID or GC/ECD.
The temperatures and pressures used in initial extractions were
similar to those used by others as cited in the literature in
similar extractions. The densities obtained from each tempera-
ture/pressure combination were displayed as part of the run
-------
parameters during the extraction. Information was obtained that
provided Hildebrand solubility parameter data on organic solvents
and extraction conditions for supercritical carbon dioxide
(temperature, pressure and resulting density) that emulated the
solubility parameters or solvent strengths of various organic
solvents. Later extractions used temperature and pressure
combinations that gave carbon dioxide densities that were similar
to the solvent strengths of various organic solvents (Table 1).
TABLE 1
SUPERCRITICAL CARBON DIOXIDE EXTRACTION CONDITIONS
EXTRACTION TENPERATURE/PRESSURE
ACTUAL CARBON
DIOXIDE DENSITY
THEORETICAL CARBON
DIOXIDE DENSITY
CORRESPONDING SOLVENT
40° C./150 atm
40° C./210 atm
40° C./390 atm
60° C./300 atm
80° C. /ISO atm
80° C./210 atm
80° C./300 atm
80° C./330 atm
0.78 g/mL
0.85 g/mL
0.95 g/mL
0.83 g/mL
0.44 g/mL
0.62 g/mL
0.75 g/mL
0.85 g/mL
0.75 g/mL
0.86 g.mL
0.96 g/mL
0.82 g/mL
0.75 g/mL
0.86 g/mL
heptane
hexane
cyclohexane
pentane
unlisted
unlisted
heptane
hexane
PROTOCOLS USED IN EXTRACTIONS
The APPENDIX lists composition and concentration of standards and
spiking solutions.
The Acid/Base Matrix spike extractions were begun at 40° C and 150
atm, continued at 210 atm for 7.5 min (static) and 15 minutes
dynamic extraction using carbon dioxide. No modifier was added.
Base/Neutral Mix 2 - MOO1E was extracted under a variety of
conditions as indicated in Tables 2 & 3. All extractions were
done with 50 uL of 1:1 MeCl2-MeOH as the modifier with an initial
pressurization at 150 atm for 1 minute, followed by a static
extraction for 7.5 minutes. Pressure, temperature and time were
varied. Extractions were run under three sets of conditions: at
80 C with a dynamic extraction for 45 minutes at 210 atm; at 80 C
with a dynamic extraction at 250 atm for 45 minutes; and at 100 C
for 45 minutes at 210 atm.
-------
Base/Neutral Mix 1-MOOID, MOOlG/Naphthalene, Mix 3 MOO1F, Mix 4 -
the Phenol Mix - MOO1P as well as Chlordane, Toxaphene and PCB
1260 were extracted with 50 uL of 1:1 MeCl2-MeOH modifier at 80°C,
with an initial pressurization at 150 atm for one minute, followed
by a static extraction at 210 atm for 7.5 min and dynamic extrac-
tion at 210 atm for 45 minutes. Collection was done using a 40
micron fused silica restrictor with methylene chloride as collec-
tion solvent. The spikes of Phenol mix MOO1P were also analyzed
using the Suprex Accutrap .
Spikes of Toxic Substances Mixes 2- Z-014E/Z-014F were extracted at
80 C with 50 uL of 1:1 MeCl2-MeOH modifier. Initial pressurization
took place at 150 atm for one minute, followed by static extraction
at 210 atm for 7.5 minutes and dynamic extraction at 210 atm for 15
minutes. The compounds were collected in methylene chloride with
the 40 micron fused silica restrictor. Spikes of this mix were
also analyzed using the Suprex Accutrap .
Pesticides Mix MOO1H/M-608-1 and the Surrogate Mix were extracted
at 80 C, with initial pressurization at 150 atm for one minute,
static extraction at 210 atm for 7.5 min and dynamic extraction at
210 atm for both 30 and 45 minutes. A modifier of 50 uL of 1:1
methanol-methylene chloride was used. Collection was in methylene
chloride using a 40 micron fused silica restrictor.
PCBs: 1260, 1254, 1248, 1242, 1232, 1221, and 1016 were extracted
using 50 uL of 1:1 MeCl2.MeOH as modifier at 80 C. The extraction
included an initial pressurization at 150 atm for one minute,
followed by static extraction at 210 atm for 7.5 minutes and
dynamic extraction at 210 atm for 30 minutes.
GENERAL RESULTS AND DISCUSSION
Technical problems associated with the restrictor technique that
have been quoted by SFE researchers include: vaporization and
increased loss of solvents due to heated restrictors and loss of
volatile compounds due to high flow rates . We encountered these
problems and many of the difficulties cited by Engelhardt and Haas
and others. These included: clogged restrictors; poor reproduc-
ibility; and variable flow rates with changing restrictor size.
This was especially true with mixes that contained high molecular
weight compounds such as late-eluting PNAs. Apparently, the
compounds were not passing entirely through the restrictor and were
ultimately causing it to plug. There were many unsuccessful
attempts to solve this problem. Base/Neutral mix 2 (M001E) proved
to be the ultimate challenge as it contained both volatile
dichlorobenzenes and late-eluting dibenzo(a,h)anthracene.
Believing that the compounds were plating out when the gases left
the restrictor and entered the cooler methylene chloride in the
collection vial, the collection vial was placed in a beaker of
warmed water. This resulted only in increased evaporation of the
collection solvent. Clipping the end of the restrictor after each
8
-------
use did not eliminate the problem either. Another approach was to
put the restrictor in a teflon tube in such a way that only the
teflon tube entered the collection solvent; with the restrictor not
extending past the cap of the collection vial. The clogging was
reduced, but the compounds plated out on the teflon tube. Table 2
compares these extractions with those obtained when the restrictor
was placed directly in the collection solvent. Recoveries were
comparable, but the standard deviation increased greatly when using
the teflon tube.
TABLE 2
COMPARISON OF MOO1E EXTRACTIONS
USING TEFLON TUBE AND FUSED SILICA RESTRICTOR
FROM DXAXOM&CEOUS EARTH
Compounds
Conditions t
Ave = %, Recovery N = 4
1 , 3-Dichlorobenzene
1,2-Dichlorobenzene
Bis (2-chloroethoxy ) methane
Naphthalene
Hexachlorobutadiene
Acenaphthene
2 , 4-Dinitrotoluene
Diethylphthalate
Fluorene
Anthracene
Hexachlorobenzene
Pyrene
Benzo (a) anthracene
Chrysene
Dibenzo ( a, h) anthracene
Teflon
7.S/45/
£^^S5S55SS
Average
97.8
98.4
100.4
99.3
99.8
97.8
93.4
98.0
96.8
91.5
87.4
88.0
82.0
81.7
68.5
tube
80/210 *
8
5.7
5.2
3.6
3.8
4.4
3.8
4.5
5.0
4.7
5.7
5.6
7.0
12.5
17.8
39.8
Fused silica
7.S/45/
Average
93.1
93.5
96.3
94.4
94.3
95.6
97.5
101.7
96.1
95.5
93.8
90.8
86.3
83.2
53.8
80/210 *
=
8
1.3
1.4
1.8
1.3
1.1
1.2
1.5
3.8
1.6
0.9
1.4
0.5
1.3
2.6
5.1
* static time(min) /dynamic time(min)/temperature( C) /pressure(atm)
Extracting at higher temperatures accelerated the rate of evapora-
tion of collection solvent, causing some of the extracts to go to
dryness. Adding more solvent in mid-run was difficult and
undesirable. Extending the time of the dynamic extraction from 15
-------
minutes to 45 minutes resulted in less plugging and better
recoveries. Increasing the pressure also resulted in better
recoveries. Table 3 compares the recoveries with increased
temperature and pressure. Overall, recoveries decreased and
standard deviation increased when the temperature was increased.
While increasing the pressure did increase the standard deviation,
the recovery of dibenzo(a,h)anthracene increased by 20% when
compared to the recovery obtained at lower atmospheric pressure.
However, other researchers have reported better extraction
efficiencies when the extraction temperature is increased. They
also found that for sample matrices having tightly bound analytes,
temperature was more knportant than pressure in achieving better
extraction recoveries.
TABLE 3
COMPARISON OF MOO1E EXTRACTIONS FROM DIATOMACEOUS EARTH
VARYING TEMPERATURE AND PRESSURE
Compounds
Conditions
Ave = % recovery N = 4
1 , 3-Dichlorobenzene
1,2-Dichlorobenzene
Bis (2-chloroethoxy)methane
Naphthalene
Hexachlorobutadiene
Acenaphthene
2 , 4-Dinitrotoluene
Diethylphthaiate
Fluorene
Anthracene
Hexachlorobenzene
Pyrene
Benzo ( a ) anthracene
Chrysene
Dibenzo ( a , h ) anthracene
Increased Tempera-
ture
7.5/45/100/210 *
Average s
83.1 12.3
83.5 12.1
84.7 13.3
84.0 13.2
85.0 12.1
76.0 21.2
67.1 21.9
72.2 21.2
72.7 21.3
63.4 23.7
60.9 24.3
61.6 24.5
85.6 28.2
56.6 32.6
49.5 34.2
Increased Pres-
sure
7.5^45/80/250 *
Average s
80.5 8.5
81.1 8.4
86.5 7.9
82.9 7.4
81.8 7.2
83.0 2.1
81.5 0.9
85.5 1.6
86.7 6.1
94.8 25.8
81.7 5.1
77.8 2.3
76.2 3.5
78.7 2.2
77.0 8.2
* static time(min)/dynamic time(min)/temperature(°C)/pressure(atm)
10
-------
Table 4 summarized the results of the ninety-seven compounds that
were extracted from spikes onto diatomaceous earth using a fused
glass restrictor. It should be noted that five different sets of
extraction conditions were used during efforts to optimize the
maximum recoveries. Pesticides and PCBs were analyzed by 6C/ECD.
All others were analyzed by 6C/FID. To compare these recoveries
with an approved EPA method, the QC parameters for Method 625 are
included. Thirteen of the compounds had a recovery of 100-107%,
forty-seven fell between 90 and 100%, thirty-two between 80 and
90%, and seven between 70 and 80%, and one below 70%. Fifteen of
the averages had a standard deviation of greater than 10%. The
lowest recovery was that of benzo(g,h,i)perylene at 68.8% and a
standard deviation of 11.86. This compares favorably with the 625
method QC limits of Detected-195.0 percent recovery and a standard
deviation of 58.9. This set of extractions was performed at a
pressure of 210 atm., possibly, a similar increase in recovery
similar to that of dibenzo(a,h)anthracene could have been obtained
if extracted at 250 atmospheres. Five of the seven compounds with
recoveries between 70 and 80% were also late eluting PNAs. The
volatile compound 2-chloroethyl vinyl ether was recovered at 80%.
Failure to recover a greater percent of benzidine and 3,3'-
dichlorobenzidine may be a function of their very basic nature.
11
-------
TABLE 4
COMPARISON OF SPIKE RECOVERIES FROM DIATOMACEOUS EARTH WITH METHOD 625 QC
Target Compounds
Recovery = %
Alpha-BHC
Delta-BHC
Gamma-BHC
Beta-BHC
Heptachlor
Aldrin
Heptachlor epoxide
Endosulfan I
4,4'-DDE
Dieldrin
Endrin
4,4'-DDD
4, 4 '-DDT
Endrin aldehyde
Endosulfan sulfate
Endosulfan II
D.E.
Average
91.1
92.3
91.8
88.8
85.9
88.7
91.3
91.9
88.1
96.4
90.0
106.7
91.4
94.4
91.6
92.8
8
n = 4
2.7
3.1
3.2
5.9
11.7
1.4
3.5
3.6
1.8
5.7
5.1
4. '4
3.8
6.8
5.5
2.0
Method 625 QC
Average
D-172.2
7.2-152.2
70.9-109.4
19.2-119.7
44.3-119.3
D-134.5
D-170.6
D-188.8
D-103.5
s
37.2
39.0
54.7
32.0
30.7
31.0
61.6
32.5
16.7
Conditions
s/d/t/p
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210 .
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
Mix
Pesticides, MOO1H
Pesticides, MOO1H
Pesticides, MOO1H
Pesticides, MOO1H
Pesticides, MOO1H
Pesticides, MOO1H
Pesticides, MOO1H
Pesticides, MOO1H
Pesticides, MO01H
Pesticides, MOO1H
Pesticides, MOO1H
Pesticides, MOO1H
Pesticides, MOO1H
Pesticides, MOO1H
Pesticides, MOO1H
Pesticides, MOO1H
s/d/t/p = static extraction(min)/dynamic extraction (min)/temperature (C)/pressure (atm)
Analysis of pesticides and PCBs by GC/ECD, all others by GC/FID.
12
-------
Target Compounds
Recovery = %
Bis ( 2-chloroethy 1 ) ether
1 , 4-Dichlorobenzehe
Bis ( 2-chloroisopropyl ) ether
Nitrobenzene
D imethy Iphthalate
2 , 6-Dinitrotoluene
Acenaphthylene
4-Bromophenyl phenyl ether
Di-n-butyl phthalate
3 , 3 '-Dichlorobenzidine
Bis ( 2-ethylhexy 1 ) phthalate
Benzo(b)fluoranthene
1,2-Dichlorobenzene
1 i 3-Dichlorobenzene
Bis(2-chloroethoxy)methane
Naphthalene
Hexachlorobutadiene
Acenaphthene
D.E.
Average
89.7
88.0
90.7
91.6
99.3
98.1
96.7
100.0
104.5
88.9
106.4
94.9
80.5
81.1
86.5
82.9
81.8
83.0
8
n = 4
6.9
6.6
6.8
7.3
10.7
11.4
8.4
10.7
11.1
17.9
11.2
17.5
8.5
8.4
7.9
7.4
7.2
2.1
Method 625 QC
Average
42.9-126.0
37.3-105.7
62.8-138.6
54.3-157.6
D-100.0
68.1-136.7
53.5-126.0
64.9-114.4
8.4-111.0
8.2-212.5
28.9-136.8
42.0-140.4
48.6-112.0
16.7-153.9
49.2-164.7
35.6-119.6
37.8-102.2
60.1-132.3
8
55.0
32.1
43.6
39.3
23.2
29.6
40 ..2
23.0
16.7
71.4
41.1
38.8
30.9
41.7
34.5
30.1
26.3
27.6
Conditions
s/d/t/p
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/250
7.5/45/80/250
7.5/45/80/250
7.5/45/80/250
7.5/45/80/250
7.5/45/80/250
Mix
MOO1D
MOO1D
MC01D
M001D
MOO1D
KOO1D
M001D
M001D
MOO1D
MOO1D
MOO1D
MOO1D
MOO1E
MOO1E
M001E
MCO1E
MOO1E
MOO1E
s/d/t/p = static extraction (min)/dynamic extraction (min)/temperature ( C)/pressure (atm)
Analysis of pesticides and PCBs by GC/ECD, all others by GC/FID.
13
-------
Target Compounds
Recovery = %
2 , 4-Dinitrotoluene
Diethyl phthalate.
Fluorene '
Anthracene
Hexachlorobenzene
Pyrene
Benzo ( a ) anthracene
Chrysene
Dibenzo( a, h) anthracene
Hexachloroethane
N-Nitroso-dipropylamine
Isophorone
1 , 2 , 4-Tr ichlorobenzene
Hexachlorocyclopentadiene
2-chloro-naphthalene
1 , 2-Diphenylhydrazine
N-Nitrosodiphenylamine
Phenanthrene
D.E.
Average
81.5
85.5
86.7
94.8
81.7
77.8
76.2
78.7
77.0
99.6
100.8
102.5
100.8
100.7
102.7
101.5
102.7
102.0
s
n = 4
0.9
1.6
6.1
25.8
5.1
2.3
3.5
2.2
8.2
3.2
2.5
5.2
2.7
1.4
3.3
3.5
3.4
4.1
Method 625 QC
Average
47.5-126.9
D-100.0
71.6-108.4
43.4-118.0
7.8-141.5
69.6-100.0
41.8-133.0
44.1-139.9
D-199.7
55.2-100.0
13.6-197.9
46.6-180.2
57.3-129.2
64.5-113.5
65.2-108.7
8
21.8
26.5
20.7
32.0
24.9
25.2
27.6
48.3
70.0
24.5
55.4
63.3
28.1
13.0
20.6
Conditions
s/d/t/p
7.5/45/80/250
7.5/45/80/250
7.5/45/80/250
7.5/45/80/250
7.5/45/80/250
7.5/45/80/250
7.5/45/80/250
7.5/45/80/250
7.5/45/80/250
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
Mix
MOO1E
MOO1E
MOO1E
MOO1E
MOO1E
MOO1E
MOO1E
MOO1E
MOO1E
M001F
M001F
M001F
MOO1F
MOO1F
M001F
MOO1F
MOO1F
MOO1F
s/d/t/p = static extraction(min)/dynamic extraction (min)/temperature ( C)/pressure (atm)
Analysis of pesticides and PCBs by GC/ECD, all others by GC/FID.
14
-------
Target Compounds
Recovery = %
Fluoranthene
Butyl benzyl phthalate
2-chloroethyl vinyl ether
N-Nitrosodimethyl amine
Naphthalene
4-Chlorophenylphenyl ether
Benzidine
Di-n-octyl phthalate
Benzo(k) f luoranthene
Benzo ( a ) pyrene
Indeno(l,2,3-cd) pyrene
Benzo ( g , h , i ) pery lene
Phenol
2-Chlorophenol
2-Nitrophenol
2 / 4-Dichlorophenol
2 , 4-Dimethylphenol
4-chloro-3-methyl phenol
D.E.
Average
102.0
104.4
80.0
85.7
88.8
82.5
83.5
90.4
85.4
81.3
80.4
68.8
92.5
91.2
93.2
92.6
93.2
95.0
8
n = 4
5.6
6.6
14.0
9.8
7.4
6.1
9.7
6.4
15.9
8.8
10.5
11.9
4.2
3.8
3.9
4.1
4.2
4.4
Method 625 QC
Average
42.9-121.3
D-139.9
35.6-119.6
38.4-144.7
18.6-131.8
25.2-145.7
31.7-148.0
D-150.9
D-195.0
16.6-100.0
36.2-120.4
45.0-166.7
52.5-121.7
41.8-109.0
40.8-127.9
8
32.8
23.4
30.1
33.4
31.4
32.3
39.0
44.6
58.9
22.6
28.7
35.2
26.4
26.1
37.2
Conditions
s/d/t/p
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
Mix
MOO1F
M001P
MOOlG/Naphthalene
MOOlG/Naphthalene
MOOlO/Naphthalene
MOOlG/Naphthalene
MOOlG/Naphthalene
MOOlG/Naphthalene
MOOlG/Naphthalene
MOOlG/Naphthalene
MOOlG/Naphthalene
MOOlG/Naphthalene
MOO1P
MCO1P
MOO1P
M001P
MOO1P
MOO1P
s/d/t/p = static extraction(min) /dynamic extraction (min)/temperature (C)/pressure (atm)
Analysis of pesticides and PCBs by GC/ECD, all others by GC/FID.
15
-------
Target Compounds
Recovery = %
2,4, 6-Trichlorophenol
2 , 4-Dlnitrophenol
4-Nitrophenol
2-Methyl-4, 6-dinitrophenol
Pentachlorophenol
Toxaphene
Chlordane (alpha isomer)
Chlordane (gamma isomer)
PCB 1260
PCB 1254
PCB 1248
PCB 1242
PCB 1232
PCB 1221 '
PCB 1016
Phenol
2-Chlorophenol
1 , 4-Dichlorobenzene
D.E.
Average
93.4
84.3
97.8
89.6
91.1
85.5
92.1
91.4
95.9
96.5
93.9
87.0
94.5
92.5
97.3
89.7
89.2
89.0
8
n = 4
4.2
14.7
5.5
9.8
5.0
6.9
2.3
1.5
2.9
7.1
2.3
7.5
7.3
2.9
9.2
9.5
9.5
9.6
Method 625 QC
Average
52.4-129.2
D-172.9
13.0-106.5
53.0-100.0
36.1-151.8
19.3-121.0
16.6-100.0
36.2-120.4
37.3-105.7
B
31.7
49.8
47.2
93.2
48.9
54.2
22.6
28.7
32.1
Conditions
s/d/t/p
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/45/80/210
7.5/45/80/210
7.5/45/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.7/15/40/210
7.7/15/40/210
7.7/15/40/210
Mix
MOO1P
M001P
M001P
MOO1P
MO01P
CAS: 8001-35-2
CAS: 57-74-9
CAS: 57-74-9
CAS: 11096-82-5
CAS: 11097-69-1
CAS: 12672-29-6
CAS: 53469-2 1-9
CAS: 11141-16-5
CAS: 11104-28-2
CAS: 12674-11-2
Matrix
Matrix
Matrix
s/d/t/p = static extraction (min) /dynamic extraction (min)/temperature (°C)/pressure (atm)
Analysis of pesticides and PCBs by GC/ECD, all others by GC/FID.
16
-------
Target Compounds
Recovery = %
N-Nitroso-di-n-propylamine
1 , 2 L4-tr ichlorobenzene
4-Chloro-3-»ethyl phenol
Acenaphthene
4-Nitrophenol
2 , 4-Dinitrotoluene
Pentachlorophenol
Di-n-butylphthalate
Pyrene
2-Fluorophenol
D5-Phenol
D5-Nitrophenol
2-Fluoro-l-l ' -biphenyl
2,4, 5-Tribromophenol
D14-Terphenyl
Alpha BHC
Delta BHC
Gamma BHC
D.E.
Average
89.0
89.2
90.5
89.0
88.1
87.8
96.1
83.8
87.2
81.5
83.7
83.7
85.6
99.9
88.2
91.0
92.8
91.5
8
n = 4
10.4
9.6
6.6
9.8
11.4
9.6
28.1
12.4
11.0
2.0
1.3
1.2
1.3
4.6
4.7
4.5
2.9
2.8
Method 625 QC
Average
13.6-197.9
57.3-129.2
40.8-127.9
60.1-132.3
13.0-106.5
47.5-126.9
38.1-151.8
8.4-111.0
69.6-100.0
8
55.4
28.1
37.2
27.6
47.2
21.8
48.9
16.7
25.2
Conditions
s/d/t/p
7;7/15/40/210
7.7/15/40/210
7.7/15/40/210
7.7/15/40/210
7.7/15/40/210
7.7/15/40/210
7.7/15/40/210
7.7/15/40/210
7.7/15/40/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
Mix
Matrix
Matrix
Matrix
Matrix
Matrix
Matrix
Matrix
Matrix
Matrix
Surrogates
Surrogates
Surrogates
Surrogates
Surrogates
Surrogates
Pesticides
Pesticides
Pesticides
s/d/t/p = static extraction (min)/dynamic extraction (min)/temperature ( C)/pressure (atm)
Analysis of pesticides and PCBs by GC/ECD, all others by GC/FID.
17
-------
Target Compounds
Recovery = %
Beta BHC
Heptachlor
Aldrin
Heptachlor epoxide
Endosulfan I
4,4'-DDE
Dieldrin
Endrin
4,4'-DDD
4, 4 '-DDT
Endrin aldehyde
Endosulfan sulfate
Endosulfan II
Aniline
Benzyl alcohol
4-Chloroaniline
2-methylnaphthalene
2-Nitroaniline
D.E.
Average
92.4
90.5
92.0
92.6
93.2
90.4
97.0
92.9
106.7
92.6
95.2
93.3
96.1
95.8
96.4
95.7
96.1
95.6
s
n = 4
4.3
15.2
3.5
3.4
3.0
4.6
3.2
2.6
4.5
4.3
2.9
1.2
1.6
1.0
1.2
1.2
1.4
1.4
Method 625 QC-
Average
D-172.2
7.2-152.2
70.9-109.4
19.2-119.7
44.3-119.3
D-134.5
D-170.6
D-188.8
D-103.5
s
37.2
39.0
54.7
32.0
30.7
31.0
61.6
32.5
16.7
Conditions
s/d/t/p
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/30/80/210
7.5/15/80/210
7.5/15/80/210
7.5/15/80/210
7.5/15/80/210
7:5/15/80/210
Mix
Pesticides
Pesticides
Pesticides
Pesticides
Pesticides
Pesticides
Pesticides
Pesticides
Pesticides
Pesticides
Pesticides
Pesticides
Pesticides
Z014E,Z014P
Z014E,Z014F
Z014E,Z014P
Z014E,Z014F
Z014E,Z014F
s/d/t/p = static extraction (min)/dynamic extraction (min)/temperature (°C)/pressure (atm)
Analysis of pesticides and PCBs by GC/ECD, all others by GC/FID.
18
-------
Target Compounds
Recovery = %
3-Nitroaniline
Benzidine
3,3' -Dichlorobenzidine
D.E.
Average
94.2
70.4
71.8 .
s •
n «a 4
1.2
13.5
17.8
Method 625 QC
Average
s
Conditions
s/d/t/p
7.5/15/80/210
7.5/15/80/210
7.5/15/80/210
Mix
Z014E,Z014F
Z014E,Z014F
Z014E,Z014F
s/d/t/p = static extr act ion (min) /dynamic extraction (min)/temperature (C)/pressure (atm)
Analysis of pesticides and PCBs by GC/ECD, all others by GC/FID.
19
-------
5PM
EXTRACTION OF DIATOMACEOU8 EARTH USING THE ACCUTRAP
With the installation of the Accutrap™ cryogenic trap, it was
decided to test the new system with the standard mix, M001E, which
includes a variety of compounds. Initial extractions suggested that
a CO2 flow rate of 2-3 mLs per minutes increased the recoveries of
dibenzo(a,h)anthracene, but resulted in poor or no recovery of the
two volatile dichlorobenzenes in the mix.
We reprogrammed the extraction (using the micro processor con-
trolled multiport valve) so that during the first minute the flow
was directed through a 50 micron glass restrictor into a vial
containing methylene chloride. The remainder of the dynamic flow
period was directed to the cryogenic trap. This was the reverse of
the configuration used by Richards and Campbell while extracting
priority pollutants in soil. They installed a cryogenic trap after
the collection flask to improve the recoveries of volatile
compounds. Analysis showed that the dichlorobenzenes were being
extracted during the first minute, but were not being captured in
the cryogenic trap.
Consequently, we reconfigured the collection system again as shown
previously in Figure 3 by installing a piece of 1/16 inch stainless
steel tubing after the trap in place of the Accutrap collection
system. The tubing passed through the rubber septum of a 7.5 mL
screw cap collection vial and down into five mL. of methylene
chloride. The carbon dioxide would now bubble through the solvent
during the dynamic extraction period, capturing the volatile
compounds that had eluted from the cryogenic trap. It was also
necessary to restrict the flow of the carbon dioxide through the
solvent during the first few minutes of the dynamic phase of the
extraction to allow the more volatile compounds to be trapped more
efficiently in the solvent. It is planned to replace this
restrictor with 1/32 inch ID stainless steel tubing which should
reduce the size of the bubbles and perhaps enhance recoveries
because of the greater area for exchange between the carbon dioxide
and solvent. The smaller tubing should also eliminate the necessity
to vary the flow through the restrictor during the dynamic stage.
This is tedious and difficult to reproduce the flows with each run.
The most optimum extraction conditions using the Accutrap M proved
to be 80°C at 300 atmospheres with a static extraction of 10
minutes and a dynamic extraction of 45 minutes. These conditions
produced a density emulating that of pentane. For the first three
minutes, the flow was restricted and for the balance of the
extraction, the flow was set at approximately 1.0 mL/min. A
comparison of the recoveries•achieved from extraction under the
original configuration as provided by the manufacturer with those
obtained with the modified collection system are shown in Table 5.
20
-------
The Phenol mix MOO1P was also extracted. Recoveries were about 10%
lower than those achieved using the configuration where the fused
silica restrictor bubbled through the collection solvent.
The Combined Toxic Substances mix Z-014E and Benzidines mix Z014F
were extracted using the same parameters. The recovery of the
toxic substances were generally down by about 9% from those
obtained using the fused silica restrictor configuration, while the
benzidines were up by about 10%.
TABLE 5
COMPARISON OF RECOVERIES FROM DIATOMACEOUS EARTH OF COMPOUNDS
FROM M001E USING DIFFERENT RECOVERY CONFIGURATIONS OF THE ACCUTRAP
,TM
Compounds
Recovery in %
Original n = 2
Altered n = 4
1,3 Dichlorobenzene
1,2 Dichlorobenzene
Bis ( 2-chloroethoxy ) methane
Naphthalene
Hexachlorobutadiene
Acenaphthene
2 , 4-Dinitrotoluene
Diethylphthalate
Fluorene
Anthracene
Hexachlorobenzene
Pyrene
Benzo ( a ) anthracene
Chrysene
Dibenzo ( a, h) anthracene
Original
System *
Average
16.8
22.2
33.6
35.3
31.8
53.8
81.8
81.8
69.6
85.1
85.8
89.8
97.4
105.6
108.4
Altered
System *
Average
80.1
88.6
92.9
89.1
91.3
92.7
91.2
92.5
91.4
100.2
91.6
91.1
88.9
88.8
94.6
Altered
System
8
3.2
2.0
1.8
1.7
1.6
2.7
6.3
5.6
4.6
18.8
12.3
8.0
10.6
13.4
14.2
* at 80°C/300 atm
This new configuration achieved recoveries of greater than 80% for
all compounds, both semi-volatile and late-eluting PNAs. The
recovery of the last four late-eluting PNAs increased from an
average of 77% to an average of 100%.
21
-------
EXTRACTION OF SOILS USING FUSED SILICA RESTRICTOR
Attempts at extraction of Fisher PAH Contaminated Soil (SRS 103-
100) proved trying to both the analysts and extraction apparatus.
This soil is widely used in SFE studies and is one of the few
"real-world" reference materials that is available at the necessary
concentrations. An acceptance range of recoveries for each of the
compounds is provided by Fisher.
Initial attempts at extraction of soil involved a three step
program that increased both the temperature and pressure from 40
C/150 atm initially to 80 C at 350 atm. Problems encountered
included: the inability to pressurize the extraction chamber past
300 atm (blown lines); clogged restrictors; and excessive evapora-
tion of collection solvent. Much effort was focused on: the causes
and possible solutions for restrictor clogging; increasing the
recoveries of late eluting compounds; improved reproducibility; and
accuracy. Numerous modifications included: shortening the
restrictor length; increasing time and temperature in the extrac-
tion program; increasing the temperature of the collection solvent
and use of various modifiers. Eventually, with the addition of 1:1
MeCl2-MeOH modifier, multiple extractions were accomplished without
restrictor clogging at a temperature of 80 C with dynamic extrac-
tion of 45 minutes at 210 atm. These were the conditions used in
the extraction of Base/neutral compounds. The extracted compounds
were collected by the insertion of the 40 micron restrictor through
the septum of a 10 mL screw cap vial into approximately seven
milliliters of methylene chloride. After extraction, the volume of
the collection solvent was diluted to 5.0 mL volumetrically.
Analysis was done by GC/MS. The results appear in Table 6 under
Fused Silica extractions.
EXTRACTION OF PNA CONTAMINATED SOILS USING ACCUTRAP™
With the addition of the cryogenic trap and the availability of the
SFE grade carbon dioxide with 10% methanol, the Fisher Soil was
again extracted. A one mL extraction vessel was used so that the
sample size could be increased to approximately one gram. This set
of extractions was initially carried out at 80 C at 300 atmo-
spheres, a density of 0.75 g/mL, with a static extraction for 10
minutes and a dynamic extraction of 45 minutes (restricting the
flow during the first few minutes to insure that the volatile
compounds were captured in the solvent) . The collection vessel was
a 20 mL. screw top vial with approximately 10 mL of methylene
chloride. The trap was cooled to -45 C, and was desorbed at 30 C
with 3 mL of methylene chloride. The extract was diluted to 10 mL
volumetrically. Interestingly, during the desorption process, a
solid phase appeared in the collection vial which went back into
solution upon warming to room temperature. Analysis was done by
GC/MS.
22
-------
Levy and Langenfeld reported greater recoveries than we were able
to attain, Langenf eld by increasing pressures to as much as 650 atm
at 200 C and Levy to a maximum of 450 atmospheres at 75°C. However,
they did not report recoveries on all fifteen compounds present.
The second set of extractions took place at 40°C. and 390 atmo-
spheres (a density of 0.95 g/mL, equivalent to cyclohexane).
Soxhlet results were established at the CRL using a 24 hour hexane-
acetone extraction procedure and analyzed by GC/MS.
Comparison of the PAHs extracted using SFE extractions of 0.2 gm
(fused silica restrictor), 1.0 gm extractions using CO2-Methanol
(Accutrap^M), those of Lopez-Avila , and the Soxhlet extraction are
shown in Table 6. It has been indicated in an EPA publication that
higher recoveries than those listed in the table were achieved by
Lopez-Avila using 2.5 grams .
23
-------
TABLE 6
COMPARISON OF RECOVERIES OBTAINED BY VARIOUS METHODS REPORTED FOR 8RS 103-100 REFERENCE SOIL
WITH CERTIFIED VALUES USING 6C/MS ANALYSIS
Recoveries in mg/kg
Weight extracted
Compound
Naphthalene
2-Methylnaphthalene
Acenaphthylene .
Acenaphthene
Dibenzofuran
Fluorene
Pentachlorophenol
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo ( a ) anthracene .
Chrysene
Benzo ( b , k) f luoranthene
Benzo(a) pyr ene
Fused
silica
0.2 g
21.3
61.7
13.2
635
284
427
820
1823
488
1406
1140
179
141
51.3
39.5
Accutrap™
1.0 g
19.4
53.5
9.0
370
217
304
713
869
289
766
769
135
162
43.3
39.6
Lopez-
Avila6
6.0 g
9.5
37.6
5.3
616
307
209
443
368
177
584
592
245
319
135
36.4
Soxhlet
5.0 g
32.6
70.9
14.5
665
307
466
1660
1646
441
1505
1318
231
272
77.0
66.8
Fisher
Cert.
Values
24.2-40.6
50.6-73.6
14.7-23.5
527-737
258-356
414-570
591-1339
1270-1966
373-471
1060-1500
744-1322
214-290
271-323
130-174
80.1-114
Fisher Range
of recoveries
8.9-38.3
37.8-75.6
8.3-24.3
516.3-665.5
253.5-357.7
392.5-558.5
410.8-1357.2
995.6-1903.4
377.4-472.2
1021.3-1591.9
643.7-1278.7
203.9-294.9
265-356
126.3-185.9
77.5-117.5
24
-------
The extraction of 0.2 g of soil was with a static extraction of 7.5
minutes and a dynamic extraction of 45 minutes at 210 atmospheres
at 80 C. A one gram sample was extracted using carbon dioxide with
10% methanol. The static extraction time was lengthened to 10
minutes, with a dynamic extraction of 45 minutes at 80 C and 300
a tin. Lopez-Avila collected the extracted compounds by inserting
the restrictor in a vial containing either methanol or methylene
chloride. The recoveries are the sum of nine extractions over a
time period of 270 minutes. The use of the Accutrap did not
inprove the recovery over that obtained using the fused silica
restrictors. For most of the compounds, the highest recoveries
were obtained by Soxhlet extraction. Low recoveries were obtained
for late-eluting PNAs using the fused silica restrictor and the
Accutrap . More work needs to be done to improve the extraction
recoveries of these compounds.
Initial extraction of NIST SFE Round Robin Sediment A exhibited the
same problems as the Fisher Soil sample, i.e., restrictor clogging.
Extraction of Sediments A and B and the Air Particulate sample did
not result in restrictor plugging using the procedure for
Base/neutral compounds.
EXTRACTION OF TEST SOIL USING ACCUTRAP™
Test soil (see sample matrices) was first extracted using the
cryogenic trap and carbon dioxide with ten percent methanol to
determine its chromatographic characteristics and suitability for
spiking. The resulting extracts were relatively clear and had few
chromatographic peaks. The soil was then spiked with 100 uL of
M001E and extracted at 60 C at 330 atmospheres, a density of 0.83
g/mL or roughly that of pentane. The extract was adjusted to 2.0
mL for analysis with the Finnigan GC/MS. The results are summa-
rized as part of Table 7.
The soil was also spiked with Z014GR, a mixture of 17 Polynuclear
Aromatic Hydrocarbons, of which all but carbazole were analyzed.
The extractions took place at 40 C at 390 atmospheres, a density of
0.95 g/mL, with a 10 minute static extraction and a 45 minute
dynamic extraction. Due to the volatile nature of some of the
compounds, flow was restricted during the first few minutes of
dynamic extraction. The remainder of the dynamic extraction was at
one mL/min. Analysis was by GC/MS.
In Table 7, the test soil extractions are compared with diatoma-
ceous earth extractions obtained using the fused silica restrictor,
a modifier of 1:1 methanol-methylene chloride and SFC grade carbon
dioxide. Generally, test soil extraction recoveries were lower and
standard deviations were higher when compared to the diatomaceous
earth results. This is to be expected since diatomaceous earth is
a "clean matrix".
25
-------
TABLE 7
*nui
COMPARISON OF SPIKE RECOVERIES OF MOO IE AND Z014GR ON TEST SOIL USINO ACCUTRAP WITH SPIKE RECOVERIES ON DIATOMACEOUS EARTH
WITH FtfSED SILICA RESTRICTOR
Matrix
Compounds
Avg in % recovery, n = 4
Acenaphthene
Acenaphthylene
Anthracene
Benzo ( a ) anthracene
Benzo(a)pyrene
Benzo(b,k) f luoranthene
Benzo ( g , h , i ) pery lene
Chrysene
Dibenzo ( a, h ) anthracene
1 , 2-Dichlorobenzene
1 , 3-Dichlorobenzene
Diethyl phthalate
2 , 4-Dinitrotoluene
Test
soil
MOO1E1
Avg.
93.2
92.2
73.4
69.0
102
61.3
75.7
102
83.9
B
5.8
4.2
10.9
12.9
11.8
14.0
16.6.
14.7
6.7
Test
soil
Z014GR2
Avg.
103
106
91.0
102
107
106
109
104
118
8
3.6
9.1
6.3
2.1
8.3
24.0
11.0
2.9
12.2
Diatom. •
Earth
MOO1E3
Avg.
98.1
95.5
86.3
83.2
53.8
93.1
93.5
98.4
98.4
a
1.2
0.9
1.3
2.6
5.1
1.3
1.4
5.0
1.4
Diatom.
Earth
MOO1E4
Avg.
83.0
94.8
76.2
78.7
77.1
80.5
81.1
85.5
81.5
8
2.1
2.1
3.5
7.2
8.2
8.5
8.4
1.6
0.9
1 Extracted at 10 min/45 min/60°C/330 atm with 10% Methanol enhanced CO2.
2 Extracted at 10 min/45 min/60°C/390 atm with 10% Methanol enhanced CO2.
3 Extracted at 7.5 min/45 min/80°C/210 atm with SFC grade CO2 fused silica restrictor.
4 Extracted at 7.5 min/45 min/80°C/250 atm with SFC grade CO2 using teflon tube with restrictor.
26
-------
Matrix
Compounds
Avg in % recovery, n = 4
Fluor ant hene
Fluorene
Hexachlorobenzene
Hexachlorobutadiene
Indeno (1,2, 3-cd ) pyrene
Naphthalene
Bis ( 2 -chloroet hoxy ) methane
Phenanthrene
Pyrene
Test
soil
MOO1E1
Avg.
97.1
93.1
76.5
87.4
87.0
92.6
8
6.1
6.5
5.8
6.4
4.8
5.4
Test
soil
Z014GR2
Avg.
96.0
97.0
124
111
95.0
97.0
s
4.6
5.0
20.6
6.9
7.1
4.6
Diatom.
Earth
MOO1E3
Avg.
96.1
93.8
94.3
94.4
96.3
90.7
8
1.6
1.4
1.1
1.3
1.8
.8
Diatom.
Earth
M001E4
Avg.
86.7
81.7
81.9
82.9
86.5
77.8
s
1.6
5.1
7.2
7.4
7.9
2.3
1 Extracted at 10 min/45 min/60°C/330 atm with 10% Methanol enhanced CO2.
2 Extracted at 10 min/45 min/60°C/390 atm with 10% Methanol enhanced CO2.
3 Extracted at 7.5 min/45 min/80°C/210 atm with SFC grade CO^ fused silica restrictor.
4 Extracted at 7.5 min/45 min/80°C/250 atm with SFC grade CO2 using teflon tube with restrictor.
27
-------
CONCLUSIONS
The use of diatomaceous earth as the matrix for our initial extrac-
tions proved very beneficial. As a "clean" simple substrate, it
could be expected to yield the spiked compounds easily, as we did
not have to deal with the complexities found in "native" pollu-
tants . However, the problems that were encountered in extracting
various high molecular weight compounds from diatomaceous earth
were also encountered in the contaminated soil analyzed in this
study.
The use of an unheated fixed diameter restrictor in conjunction
with the extraction of the high molecular weight compounds
generated problems with restrictor plugging. Attempts to overcome
this included: lengthening the restrictor so that a measured amount
could be cut off after each extraction; removing the clogged
portion; placing the collection vial in warmed water; putting the
restrictor in a copper tube which was heated with electrical
heating tape; and placing the restrictor in a teflon tube. By
increasing the extraction time, the useful life of the restrictor
was increased.
FUSED SILICA RESTRICTOR
Generally, the 40 micron restrictor when bubbling into the
collection solvent gave extremely good recoveries and standard
deviations when using diatomaceous earth as shown. Problems
appeared when extracting mixtures of various classes of semi-
volatile compounds as shown in Table 7. The conditions which
produced excellent recoveries and standard deviations for most
compounds were not adequate to extract high-molecular weight
compounds. Conditions which produced better recoveries for high
molecular weight compounds resulted in lower recoveries and higher
standard deviations for the other compounds.
CRYOGENIC TRAP
The cryogenic trap posed its own set of problems. The glass beads
in the trap were not sufficiently adsorbent to capture the volatile
compounds, such as the dichlorobenzenes, so they literally blew
through the collection system. The system was not configured to
have the carbon dioxide pass, through a solvent to collect these
compounds. However,, the trap did hold the heavier compounds until
desorption. The adjustable restrictor has proven difficult to
operate as it works by crimping PEEK^ manually and this has not
proven to be a reproducible method to regulate the flow rate. The
flow rate was that from the pump and not the flow through the
restrictor. After changing the collection system to incorporate
the bubbling of the gases through a solvent, we discovered that the
28
-------
stainless steel tube frosted up and there was a real chance that
melting water would find its way through the septum during
desorption. The water was trapped by an absorbent material placed
around the stainless steel tubing.
GENERAL CONCLUSIONS
Results obtained using the fused silica restrictor collection
system or Accutrap gave acceptable QC data when compared to EPA
method 625. The results produced by the modified Accutrap seemed
to produce the best overall recoveries when working with the
complex matrix of soils. This configuration demonstrated the
capability of recovering both volatile and high-molecular weight
compounds which would be a necessary condition for routine
laboratory soil analysis. It was not subject to the frequent
clogging encountered with the fused silica restrictor. However,
the fused silica restrictor gave better recoveries for certain
compounds, but it was not as rugged as the Accutrap system. At
this time, there is no single set of SFE conditions to extract the
entire class of semi-volatile compounds tested. Avenues of future
inquiry should explore methods that result in better recoveries of
all classes of semi-volatile compounds in the extraction of soils.
These include: increasing the density of the supercritical CO2 to
emulate the solubility parameters of other common solvents; mixing
the soil with diatomaceous earth to determine if this allows better
penetration of the carbon dioxide during static and dynamic
extractions; using a different adsorbent in the cryogenic trap,
such as C-18; continuing to investigate the roles of the size of
the delivery tube (would the smaller ID stainless steel tube allow
the recovery of volatile and high-molecular weight compounds,
without the necessity to vary the flow rate during the extraction) ;
investigating the use of different modifiers and increasing the
sample size to be more reproducible and afford more sensitivity.
Supercritical fluid extraction has been shown to be accurate and
precise in the extraction of semi-volatile compounds from solid
matrices. More work needs to be done to improve the extraction
process and test other types of solid matrices.
29
-------
APPENDIX A STANDARD PREPARATION
ACID/BASE NEUTRAL MATRIX SPIKE SOLUTION
BASE/NEUTRAL SPIKE Combined vial 5000 ug/mL of:
1,4-Dichlorobenzene CAS 106-46-7
CAS 621-0-7
CAS 121-1402
CAS 84-74-2
CAS 83-32-9
CAS 120-82-1
CAS 129-00-0
N-Nitroso-di-n-propylamine
2,4-Dinitrotoluene
Di-n-butyl Phthalate
Acenaphthene
1,2,4-Trichlorobenzene
Pyrene (1000 ug/mL-2 mL)
Diluted 1.5 mL to 15 mL with methanol, approx lOx dilu-
tion (500 ug/mL)
ACID SPIKE Combined 2 vials 5000 ug/mL of:
4-Nitrophenol
2-Chlorophenol
Phenol
4-Chloro-3-methyl phenol
Pentachlorophenol
Diluted contents of 2 vials of each
methanol, approximately a 5x dilution or 1000 ug/mL stock
CAS 100-02-7
CAS 95-57-8
CAS 108-95-2
CAS 59-50-7
CAS 87-86-5
to 15 mL with
REFERENCE SOLUTION: 100 uL of combined stock diluted to 1.0
mL volumetrically.
TARGET COMPOUND CONCENTRATION: 100 ug/mL for acid compounds.
50 ug/mL for base/neutral compounds, except for pyrene, which
is 20 ug/mL.
SPIKE: 100 uL of combined stock solutions.
BASE-NEUTRAL MIX 1 HOOID ACCUSTANDARD
Cone. 100, 200 ug/mL in methanol.
REFERENCE SOLUTION:
with MeCl2
Dilute 100 ul to 1.0 mL volumetrically
TARGET COMPOUND CONCENTRATION:
10 ug/mL Benzo-(b)-fluoranthene
20 ug/mL Acenaphthylene
4-Bromophenyl phenyl ether
Bis(2-chloroethyl) ether
Bis (2-chloroisopropy1) ether
1,4-dichlorobenzene
3,3-dichlorobenzidine
Dimethyl phthalate
Di-n-butyl phthalate
2,6-dinitrotoluene
Bis (2-ethylhexyl) phthalate
Nitrobenzene
SPIKE: 100 uL of stock plus 50 uL 1:1 MeCl2-MeOH.
30
-------
BASE-NEUTRAL MIX 2 MOO1E ACCUSTANDARD
Cone. 100, 200 ug/nL in methanol
REFERENCE SOLUTION: Dilute 100 uL to 1.0 mL volumetrically
with MeCl2.
TARGET COMPOUND CONCENTRATION:
10 ug/mL Benzo(a)anthracene
Chrysene
Dibenzo(a, h)anthracene
20 ug/mL Acenaphthene
Anthracene
1',2-dichlorobenzene
1,3-dichlorobenzene
Diethyl phthalate
2,4-dinitrotoluene
Fluorene
Hexachlorobenzene
Hexachlorobutadiene
Naphthalene
Bis(2-chloroethoxy) methane
Pyrene
SPIKE: 100 uL stock plus 50 uL 1:1 MeCl2-MeOH
BASE-NEUTRAL MIX 3 M001F ACCUSTANDARD
Cone. 100, 200 ug/mL in methanol.
REFERENCE SOLUTION: Dilute 100 uL of stock to 1.0 mL
volumetrically with MeCl2.
TARGET COMPOUND CONCENTRATION:
10 Ug/mL Fluoranthene
20 ug/mL Butyl benzyl phthalate
2-chloronaphthalene
1,2-diphenyIhydraz ine
Hexachlorocyclopentadiene
Hexachloroethane
Isophorone
N-nitroso-di-n-propylamine
N-nitrosodiphenylamine
Phenanthrene
1,2,4-Trichlorobenzene
SPIKE: 100 uL stock plus 50 uL 1:1 MeCl2-MeOH.
31
-------
COMBINED BASE NEUTRAL MIX 4 MOO1G/NAPHTHALENE ACCUSTANDARO
MO01G Cone. 100, 200 ug/mL in 1:1 MeCl2-MeOH
Naphthalene CAS 91-20-3 Cone. 5000 ug/mL
100 uL added to 2.0 mL to 2.0 mL MO01G yielding a 250
ug/mL concentration.
REFERENCE SOLUTION: Diluted 100 uL of combined stock to 1.0
mL volumetrically with HeCl2.
TARGET COMPOUND CONCENTRATION:
10 ug/mL Benzo(k)fluoranthene
Indeno(1,2,3-cd)pyrene
Benz o(g,h,i)perylene
Benzo(a)pyrene
20 ug/mL Benzidine
2-chloroethyl vinyl ether
4-chlorophenyl phenyl ether
Di-n-octyl phthalate
N-nitrosodimethylamine
25 ug/mL Naphthalene
SPIKE: 100 uL of combined stock plus 50 uL 1:1 MeCl,-MeOH.
PHENOL MIX ACCUSTANDARD MOOIP
Cone. 500 TO 2500 ug/mL in methanol
REFERENCE SOLUTION: Diluted 50 uL to 1.0 mL volumetrically
in MeCl2.
TARGET COMPOUND CONCENTRATION:
25 ug/mL 2-Nitrophenol
Phenol
2-Chlorophenol
2,4-dichlorophenol
2,4-dimethylphenol
75 ug/mL 2,4-dinitrophenol
2,4,6-Trichlorophenol
125 ug/mL 4-Chloro-3-methylphenol
4-Nitrophenol
2-Methyl-4,6-dinitrophenol
Pentachlorophenol
SPIKE: 50 uL stock solution plus 50 uL of Methanol.
32
-------
SURROGATE MIX EPA QUALITY ASSURANCE MATERIALS BANK
2,4,6-TRIBROMOPHENOL CAS 118-79-6 (5000 ug/mL) in
methanol (2 ml)
ORGANIC SURROGATE MIX (5000 ug/mL) in methylene
chloride (1 ml)
P-TERPHENYL-D14 CAS 1718-51-0 (5000 ug/mL) in THF(2 mL)
Diluted volumetrically to 5.0 mL resulting in a cone of
2000 ug/mL for 2,4,6-Tribromophenol and p-Terphenyl-
D-14 and 1000 ug/mL for the Organic Surrogate Mix.
REFERENCE SOLUTION: Diluted 100 uL stock to 1.0 mL.
volumetrically with MeCl2.
TARGET COMPOUND CONCENTRATION
40 ug/mL 2-Fluorophenol
D5-Phenol
D5-Nitrobenzene
2-Fluoro-l,1'-biphenyl
200 ug/mL 2,4,6-Tribromophenol
D14-p-terphenyl
SPIKE: 100 uL stock plus 50 uL 1:1 MeCl2-MeOH.
COMBINED TOXIC SUBSTANCE MIX 2 ACCUSTANDARD Z-014E
Cone. 2000 ug/mL in methanol
BENZIDINES MIX ACCUSTANDARD Z-014F
Cone. 2.0 mg/mL (2000 ug/mL) in methylene chloride
REFERENCE SOLUTION: 100 uL of both Z-014E and Z-014F were
diluted volumetrically to 5.0 mL with methylene
chloride.
TARGET COMPOUND CONCENTRATION:
40 ug/mL Aniline
Benzyl alcohol
4-Chloroaniline
Dibenzofuran
2-methyInaphthalene
2-Nitroaniline
3-Nitroaniline
4-Nitroaniline
Benzidine
3,3'-Dichlorobenz idine
SPIKE: 100 uL of combined stock plus 50 uL of 1:1 MeCl2-MeOH
diluted to 5.0 mL with MeCl2 after extraction.
33
-------
PAH MIX Z014GR
Cone. 2.0 mg/mL of each in MeCl2-benzene (50:50)
REFERENCE SOLUTION: diluted 100 uL of stock solution to 2.0 mL
volumetrically with methylene chloride.
TARGET COMPOUND CONCENTRATION
40 ug/mL Acenaphthene
Acenaphthylene
Anthracene
Benz o(a)anthracene
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo (g,h,i)perylene
Ben z o(k)fluoranthene
Chrysene
Dibenzo(a/h)anthracene
Fluoranthene
Fluorene
Indeno(1,2,3-cd)pyrene
Naphthalene
Phenanthrene
Pyrene
Carbazole (not analyzed)
SPIKE: 100 uL stock with no modifier
TOXAPHENE CAS 8001-35-2
CONC 1000 ug/mL +/-100 ug/mL in methanol
REFERENCE SOLUTION: 200 UL of stock diluted to 1.0 mL
volumetrically with 1:1 methylene chloride
yields a 200 ug/mL solution.
SPIKE: 200 uL of stock plus 50 uL 1:1 MeCl2-MeOH.
CHLORDANE CAS 57-74-9
Cone. 1000 ug/mL +/- 100 ug/mL
REFERENCE SOLUTION: 50 uL stock diluted to 1.0 mL
volumetrically with methylene chloride yields
50 ug/mL solution.
SPIKE: 50 uL of stock plus 50 uL of 1:1 MeCl2-MeOH. •
34
-------
PESTICIDE MIX ACCUSTANDARD MOOlH/M-608-1
Cone. 100 TO 600 ug/mL in methanol
INTERMEDIATE REFERENCE STANDARD: Diluted 100 uL to 1.0 mL
volumetrically with methylene chloride.
TARGET COMPOUND CONCENTRATION (Intermediate):
10 ug/mL Aldrin
alpha-BHC
beta-BHC
gamma-BHC
delta-BHC
Heptachlor
Heptachlor epoxide
20 ug/mL p,p'-DDE
Dieldrin
Endosulfan I
Endosulfan II
Endrin
60 ug/m p,p'-DDT
p,p'-DDD
Endosulfan sulfate
Endrin aldehyde
SPIKE: 100 uL of intermediate reference solution plus 50 uL of
1:1 MeCl2-MeOH diluted to 1.0 ml.
PCB 1260 CAS 11096-82-5
Cone. 1000 ug/mL +/-100 ug/mL in iso-octane
REFERENCE SOLUTION: Diluted 50 uL to 1.0 mL volumetrically
with methylene chloride.
SPIKE: 50 uL of stock plus 50 uL of 1:1 MeCl2-MeOH.
PCB 1254 CAS 11097-69-1
Cone. 1000 ug/mL in iso-octane
REFERENCE SOLUTION: 50 uL stock to 1.0 mL volumetrically
with 1:1 MeCl2-MeOH
SPIKE: 50 uL of stock plus 50 uL of 'methylene chloride
35
-------
PCB 1248 CAS 12672-29-6
Cone 5000 ug/mL +/- 500 in iso-octane
REFERENCE SOLUTION: 10 uL of stock diluted to 1.0 mL
volumetrically with methylene chloride yields
50 ug/mL solution.
SPIKE: 10 uL of stock plus 50 uL of MeCl2-MeOH.
PCB 1242 CAS 53469-21-9
Cone. 3000 ug/mL +/- 300 in iso-octane
REFERENCE SOLUTION: Diluted 20 uL of stock to 1.0 mL
volumetrically with methylene chloride yields
a 60 ug/mL solution.
SPIKE: 20 uL of stock plus 50 uL of 1:1 MeCl2-MeOH.
PCB 1232 CAS 11141-16-5
Cone. 5000 ug/mL +/- 500 in methanol
REFERENCE SOLUTION: Diluted 10 uL to 1.0 mL volumetrically
with methylene chloride yielding a 50 ug/mL solution.
SPIKE: 10 uL of stock plus 50 uL of 1:1 MeCl2-MeOH.
PCB 1221 CAS 11104-28-2
Cone. 1000 ug/mL +/-100 in iso-octane
REFERENCE SOLUTION: Diluted 50 uL to 1.0 mL volumetrically
with methylene chloride yielding a 50 ug/mL solution.
SPIKE: 50 uL stock plus 50 uL of 1:1 MeCl2-MeOH.
PCB 1016 CAS 12674-11-2
CONG. 5000 ug/mL +/-500 in iso-octane
REFERENCE SOLUTION: Diluted 10 uL to 1.0 mL volumetrically
with methylene chloride yielding a 50 ug/mL solution.
SPIKE: 10 uL plus 50 uL of 1:1 MeCl2-MeOH.
36
-------
1. J. Pawliszyn,"Kinetic Model of Supercritical Fluid Extraction",
Journal of Chromatoaraphic Science. Vol. 31, pp 31-37,1993.
2. Mark D.Burford, Steven B. Hawthorne, and David Miller,
"Extraction Rates of Spiked Versus Native PAHs from Hetero-
geneous Environmental Samples Using Supercritical Fluid
Extraction an Sonication in Methylene Chloride". Analytical
Chemistry. Vol. 65, No. 11, pp 1497-1505, June 1, 1993.
3. J. Tehrani,"Successful Supercritical Fluid Extraction
Strategies",. American Laboratory. Vol 25, N2 (Feb) ,pp 40HH-
40MM, 1993.
4. Dionex Corporation, Sunnyvale, CA, " Elements of Supercritical
Fluid Extraction", 1992.
5. H. Engelhardt and P. Haas, "Possibilities and Limitations of
SFE in the Extraction of Aflatoxin Bl from Food Matrices",
Journal of Chromatoaraphic Science. Vol 31, pp 13-19, January
1993.
6. John J. Langenfeld, Steven B. Hawthorne, David Miller, and
Janusz Pawliszyn, "Effect of Temperature and Pressure on
Supercritical Fluid Extraction Efficiencies of Polycyclic
Aromatic Hydrocarbons and Polychlorinated Biphenyls" .Analytical
Chemistry. Vol. 65, No. 4, pp 338-344, February 15, 1993.
7. M. Richards and R. M. Campbell, "Comparison of Supercritical
Fluid Extraction, Soxhlet, and Sonication Methods for the
Determination of Priority Pollutants in Soil", LC-GC Magazine
of Separation Science. Vol. 9, No. 5, pp 358-364, 1991.
8. J.M. Levy, L.A. Dolata, and R.M. Ravey, "Considerations of SFE
for GC/MS Determination of Polynuclear Aromatic Hydrocarbons in
Soils and Sediments", Journal of Chromatoaraphic Sciencef
Vol. 31, pp 349-352, September 1993.
9. John J. Langenfeld, 338-344.
10. Viorica Lopez-Avila and N.S. Dodhiwala, Werner F. Beckert,
"Supercritical Fluid Extraction and Its Application to
Environmental Analysis", Journal of Chromatoaraphic Science.
Vol. 28, pp 468-475, September 1990.
11. Viorica Lopez-Avila, N.S. Dodhiwala, and W.S. Beckert,"Method
for the-Suupercritical Fluid Extraction of Soils/Sediments",
USEPA Research and Development Pronect Summary. EPA/600/S4-
90/026, March 1991.
12. Mark D. Burford, 1497-1505.
37
------- |