Supercritical Fluid Extraction in Environmental Analysis - ACS

May 8, 1992 - 3 Current address: Midwest Research Institute, 625—B Clyde Avenue, Mountain View, CA 94043. Supercritical Fluid Technology. Chapter 14...
0 downloads 0 Views 2MB Size
Chapter 14

Supercritical Fluid Extraction in Environmental Analysis V. Lopez-Avila

1,3

2

and W. F. Beckert

1

Mid-Pacific Environmental Laboratory, 625-B Clyde Avenue, Mountain View, CA 94043 Environmental Monitoring Systems Laboratory, U.S. Environmental Protection Agency, 944 East Harmon Avenue, Las Vegas, NV 89109

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

2

Samples of sand spiked with 36 nitroaromatic compounds, 19 haloethers, and 42 organochlorine pesticides, and a standard reference soil (certified for 13 polynuclear aromatic hydrocarbons, dibenzofuran, and pentachlorophenol) were extracted with supercritical carbon dioxide in a two- or four-vessel supercritical fluid extractor to establish the efficiency of the extraction and the degree of agreement of the parallel extraction recoveries. Furthermore, the many variables that influence the extraction process (e.g., flowrate, pressure, temperature, moisture content, cell volume, sample size, extraction time, modifier type, modifier volume, static versus dynamic extraction, volume of solvent in the collection vessel, and the use of glass beads to fill the void volume) were investigated.

In this paper, the supercritical fluid extraction (SFE) of organic compounds from sand spiked with 36 nitroaromatic compounds, 19 haloethers, and 42 organochlorine pesticides, and from a standard reference material certified for 13 polynuclear aromatic hydrocarbons (PAH), dibenzofuran, and pentachlorophenol was examined using a two- and a four-vessel extractor. Although the results achieved by S F E for the sand and the standard reference soil samples were very encouraging, previous data obtained in our laboratory on the standard reference soil and a few other standard reference marine sediments were less favorable. It was therefore decided that an investigation of seven variables for their influence on the analyte recoveries from the standard soil sample would be useful. Two tests were conducted in which these variables were investigated. In Test 1, the seven variables selected were pressure, temperature, moisture content, cell volume, sample size, extraction time, and modifier volume. In Test 2, the seven variables were pressure, temperature, volume of toluene added to the matrix, volume of solvent in the collection vessel, 3

Current address: Midwest Research Institute, 625-B Clyde Avenue, Mountain View, C A 94043 0097-6156/92AM88-0179$07.75A) © 1992 American Chemical Society

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

180

SUPERCRITICAL

FLUID

TECHNOLOGY

moisture content, presence/absence of glass beads, and static extraction time. For each variable we chose two values. Eight experiments were performed per test, and the results of these experiments are summarized in this paper as the relative changes in recoveries when going from low to high values across the 15 target compounds. In addition to these method optimization experiments, we investigated the effect of varying flowrate, pressure, and temperature using the standard reference soil and a Hewlett-Packard extractor. Experimental

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

Apparatus. •

Supercritical fluid extraction system - Suprex Model SE-50 consisting of a 250-mL syringe pump with the necessary valves and connecting lines to the extraction vessel, a control module containing a microprocessor for controlling the SFE system and able to store up to 25 extraction programs, and an oven module consisting of the extraction oven, the extraction vessel, a 4-port and a 12-port valve configured with electronic actuators for automated operation. The system was set up either with two or four 2-mL extraction vessels in a horizontal position for simultaneous extractions. The extraction vessels (0.9-cm ID x 3-cm length) were obtained from Alltech Associates (Deerfield, Illinois). Supercritical pressures were maintained inside the extraction vessels by using 60-cm length of uncoated fused-silica tubing (50-/xm ID x 375-^tm OD) from J & W Scientific as restrictors. Collection of the extracted material was performed by inserting the outlet restrictors into 15-mm ID x 60-mm glass vials (Supelco Inc., Bellefonte, Pennsylvania) containing hexane spiked with a known amount of an internal standard (terphenyl-di4). Figures 1 and 2 show schematic diagrams of the two- and the fourvessel extraction systems. For the four-vessel extraction system, we used eight restrictors mounted in the 12-port valve to allow collection of two fractions per sample. Fractions 1A, 2A, 3A, and 4A were collected when the 12-port valve was in the load position. Fractions IB, 2B, 3B, and 4B were collected when the 12-port valve was in the inject position. The SFE conditions are given in Table I.



Supercritical fluid extraction system ~ Hewlett Packard Model 7680A totally automated system with unlimited-capacity reciprocating pump, specially designed extraction chamber with safety interlocks, a variable restrictor nozzle and analyte collection trap. The operation of the extractor is controlled by a personal computer which is a Microsoft Windows-based system. A n animated status screen provides real-time monitoring of the extraction process. Table II gives the SFE conditions for the HP extractor.

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

14. LOPEZ-AVILA & B E C K E R T

SFE in Environmental Analysis

Fr.lA

181

• uu

Fr. 2A Fr. IB Fr. 2B

Figure 1. Schematic representation of two-vessel suprex extraction system. Table I. S F E Conditions for the Suprex Extraction System Value

Parameter Pressure (atm) Temperature (°C) Flowrate (mL/min)

Number of extraction vessels Position of extraction vessel Extraction time (min) Extraction vessel volume (mL) Extraction vessel dimensions Restrictor temperature (°C) Restrictor dimensions Collection solvent Temperature of collection vial

a

Varied Varied Not known unless volume of compressed C O 2 used in extraction was recorded 2 to 4 a

30 to 60 per fraction 2

9 mm ID x 30 mm length Room temperature 50 /xm ID x 60 cm length Hexane (5 mL) Room temperature b

a

The pressure and temperature were varied as indicated in the tables of results, temperature decreased during collection due to cooling of carbon dioxide upon expansion.

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

SUPERCRITICAL

FLUID T E C H N O L O G Y

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

cos

VIAL3B

Load Position

Figure 2. Schematic representation of four-vessel suprex extraction system.

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

b

b

c

C

C

HP-2

d

0.85 0.85 335 335 1.0 2.0 60 60 30 30 9.3 4.7 60 60 20 20 Hexane Hexane 1.0 1.0 1.0 1.0 30 30 25 25 7 7 10 m m ID 10 mm ID x 90 mm x 90 mm length length Vertical Vertical 1.0 1.0 ODS ODS 2 2

HP-1 0.85 335 3.0 60 30 14.0 60 20 Hexane 1.0 1.0 30 25 7 10 mm ID x 90 mm length Vertical 1.0 ODS 2

HP-3 0.85 335 4.0 60 30 18.7 60 20 Hexane 1.0 1.0 30 25 7 10 mm ID x 90 mm length Vertical 1.0 ODS 2

HP-4 0.6 149 2.0 60 30 13.7 60 20 Hexane 1.0 1.0 30 25 7 10 mm ID x 90 mm length Vertical 1.0 ODS 2

HPS 0.75 218 2.0 60 30 10.6 60 20 Hexane 1.0 1.0 30 25 7 10 mm ID x 90 m m length Vertical 1.0 ODS 2

HP-6

HP-9 0.90 281 2.0 40 30 8.8 60 20 Hexane 1.0 1.0 30 25 7 10 mm ID x 90 mm length Vertical 1.0 ODS 2

HP-7 0.85 329 2.0 60 30 9.3 60 20 Hexane 1.0 1.0 30 25 7 10 mm ID x 90 mm length Vertical 1.0 ODS 2

21

0.90 350 2.0 50 30 8.8 60 20 Hexane 1.0 1.0 30 25 7 10 mm I D x 90 mm length Vertical 1.0 ODS 2

HP-10

HP-12

0.85 0.80 329 314 2.0 2.0 60 70 30 30 9.3 9.9 60 60 20 20 Hexane Hexane 1.0 1.0 1.0 1.0 30 30 25 25 7 7 10 m m ID 10 mm ID x 90 m m x 90 m m length length Vertical Vertical 1.0 1.0 ODS ODS 2 2

HP-11

d

c

b

* A l l experiments were performed with SRS103-100; the sample size was 1.0 g. A n additional experiment identified as H P - 8 was planned to be performed at 381 bars and 6 0 ° C ; however, the prototype system which w e used did not allow us to reach 381 bars. D u r i n g the extraction step. During the rinse step. ODS—ocucecyl-bonded silica.

Position o f extraction vessel Equilibration time (min) Trap material Number o f rinses per trap

c

Fluid density (g/mL) Pressure (bar) Flowrate ( m L / m i n ) Temperature ( ° C ) Extraction time (min) Thimble volumes swept N o z z l e temperature ( ° C ) Trap temperature ( ° C ) Rinse solvent Volume ( m L ) Rate ( m L / m i n ) N o z z l e temperature ( ° C ) Trap temperature ( ° C ) Extraction vessel volume (mL) Extraction vessel dimensions

Parameter

Table II. S F E Conditions for H P Extractor Evaluation

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

SUPERCRITICAL FLUID

TECHNOLOGY

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

Gas Chromatograph — A Varian 6000 equipped with two constantcurrent/pulsed-frequency electron capture detectors, a 30-m x 0.53mm ID DB-5 fused-silica open-tubular column (1.5-/*m film thickness), and a 30-m x 0.53-mm ID DB-1701 fused-silica opentubular column (l.Q-fim film thickness), both connected to a press-fit Y-shaped fused-silica inlet splitter (Restek Corporation, Bellefonte, Pennsylvania), was used to analyze for the nitroaromatic compounds. The columns were temperature-programmed from 120°C (1.0-min hold) to 200°C (1-min hold) at 3°C/min, then to 250°C (4-min hold) at 8°C/min; injector temperature 250°C; detector temperature 320°C; helium carrier gas 6 mL/min; nitrogen makeup gas 20 mL/min. Gas Chromatograph - A Varian 6500 equipped with two constantcurrent/pulsed-frequency electron capture detectors, a 30-m x 0.53mm ID DB-5 fused-silica open-tubular column (0.83-/xm film thickness), and a 30-m x 0.53-mm ID DB-210 fused-silica opentubular column (1.0-/xm film thickness), both connected to an 8-in injection tee (Supelco, Bellefonte, Pennsylvania), was used to analyze for the haloethers. The columns were temperature-programmed from 180°C (0.5-min hold) to 260°C (1-min hold) at 2°C/min; injector temperature 250°C; detector temperature 320°C; helium carrier gas 6 mL/min; nitrogen makeup gas 20 mL/min. Gas Chromatograph - A Varian 6000 equipped with two constantcurrent/pulsed-frequency electron capture detectors, a 30-m x 0.53mm ID DB-5 fused-silica open-tubular column (0.83-^m film thickness), and a 30-m x 0.53-mm ID DB-1701 fused-silica opentubular column (1.0-/xm film thickness), both connected to a press-fit Y-shaped fused-silica splitter (Restek Corporation, Bellefonte, Pennsylvania), was used to analyze for the organochlorine pesticides. The columns were temperature-programmed from 140°C (2.0-min hold) to 270°C (15-min hold) at 2.8°C/min; injector temperature 250°C; detector temperature 320°C; helium carrier gas 6 mL/min; nitrogen makeup gas 20 mL/min. Gas Chromatograph/Mass Spectrometer - A Finnigan 4510B (Finnigan Mat, San Jose, California) interfaced with a data system for data acquisition and processing and equipped with a 30-m x 0.32-mm ID DB-5 fused-silica open-tubular column (l-/xm film thickness) was used to analyze the extract of the standard reference material SRS 103-100. The column was temperature-programmed from 40°C (4-min hold) to 300°C (6-min hold) at 8°C/min; injector temperature 270°C; interface temperature 270°C.

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

14.

LOPEZ-AVILA & B E C K E R T

SFE in Environmental Analysis

185

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

Materials. •

Standards -analytical reference standards of the PAHs, nitroaromatics, haloethers, and organochlorine pesticides were obtained from the U.S. Environmental Protection Agency, Pesticides and Industrial Chemicals Repository (Research Triangle Park, North Carolina); Aldrich Chemical (Milwaukee, Wisconsin); UltraScientific Inc. (Hope, Rhode Island); Chem Service (West Chester, Pennsylvania), and University of Illinois. Purities were stated to be greater than 98 percent. Stock solutions of each test compound were prepared in pesticide-grade hexane at 1 mg/mL. Working calibration standards were prepared by serial dilution of a composite stock solution prepared from the individual stock solutions.



Carbon dioxide, SFC-grade, Plumsteadville, Pennsylvania).



Sample matrices ~ sand and PAH-contaminated soil SRS 103-100 (Fisher Scientific, Pittsburgh, Pennsylvania). The SRS 103-100 PAH-contaminated soil was certified by Fisher Scientific for 13 PAHs, dibenzofuran, and pentachlorophenol. The certified values are presented in the Results and Discussion section.

liquid

(Scott

Specialty

Gases,

Sample Extraction for the Two-Vessel Setup. The extraction of the 36 nitroaromatic compounds from sand spiked at 600 ng/g (per analyte) was begun at 150 atm/50°C/10 min (dynamic), continued at 200 atm/60°C/10 min (dynamic), and completed at 250 atm/70°C/10 min (dynamic) using carbon dioxide only. Sample size was 2.5 g. For all the experiments reported in this paper, the sample was sandwiched between two plugs of silanized glass wool to fill out the void volume. Two spiked samples identified as Experiments 1 and 2 in Figures 3a and 3b were extracted in parallel using the same conditions. Two additional sand samples spiked at the same concentration were extracted in parallel at 300 atm/70°C/30 min (dynamic) and are identified in Figures 3a and 3b as Experiments 3 and 4. To verify the completeness of the SFE technique, we collected an additional fraction for Experiments 1 and 2 at 300 atm/70°C/30 min (dynamic). No compounds were detected in the second fraction. The extraction of the 19 haloethers from sand spiked at 600 ng/g (per analyte) was performed at 250 atm/70°C/60 min (dynamic) using carbon dioxide only (Experiments 5 through 8 in Figure 4). Sample size was 2.5 g. In Experiments 5 and 6, two sand samples (spiked in an aluminum cup with 150 /xL of 10 ng//xL spiking solution) were extracted in parallel. In Experiments 7 and 8, two other sand samples (spiked directly in the extraction vessel to avoid losses due to compound volatilization and sample transfer) were extracted in parallel. In Experiments 9 and 10, two soil samples (spiked in an aluminum cup as in Experiments 5 and 6) were extracted in parallel at 150 atm/50°C/60 min (dynamic).

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

Exp 1

^

6

8

9

10

Compound Number

7

11

12

13

14

15

P

16

17

18

Exp 2 £222 Exp 3 [ s s gEX 4 min; 200 atm/60°C/10 min; 250 atm/70°C/10 min. Exp. 3 and 4:300 atm/70°C/30

5

min

13. 4-Chloro-3-nitrotoluene 14. 3,4-Dichloronitrobenzene 15. 2,3-Dichloronitrobenzene 16. 2,4,6-Trichloronitrobenzene 17.1,4-Naphthoquinone 18.1,2,4-Trichloro-5-nitrobenzene

1. Nitrobenzene 2. o-Nitrotoluene 3. m-Nitrotoluene 4. p-Nitrotoluene 5.1-Chloro-3-nitrobenzene 6.1-Chloro-4-nitrobenzene 7.1-Chloro-2-nitrobenzene 8. 2-Chloro-6-nitrotoluene 9. 4-Chloro-2-nitrotoluene 10. 3,5-Dichloronitrobenzene 11. 2,5-Dichloronitrobenzene 12. 2,4-Dichloronitrobenzene

Figure 3a. Percent recoveries of 18 nitroaromatics (compounds 1 through 18) extracted from spiked sand (Z5 g). The experimental conditions are given in the experimental section.

Exp. 1 and 2:150 atm/50°C/10

ZZI

10

20

30

40

50

60

70

80

90

100

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

-

20

Exp 1

19

21

22

26

27

28

30

Exp 3

29

31

min; 250 atm/70°C/10

Compound Number

25

Exp 2

24

min; 200 atm/60°C/10

[XXI

23

33

min.

^

32

35

36

Exp. 3 and 4:300

Exp 4

34

atm/70°C/30

min

19.1,4-Dinitrobenzene 20. 2,6-Dinitrotoluene 21.1,3-Dinitrobenzene 22.1,2,3-Trichloro-4-nitrobenzene 23. 2,3,5,6-Tetrachloronitrobenzene 24.1,2-Dinitrobenzene 25. 2,4-Dinitrotoluene 26.1-Chloro-2,4-dinitrobenzene 27. 2,3,4,5-Tetrachloronitrobenzene 28.1-Chloro-3,4-dinitrobenzene 29. Trifluralin 30. Benefin 31. Pentachloronitrobenzene 32. Profluralin 33. Dinitramine 34. Butralin 35. Isopropalin 36. Penoxalin

Figure 3b. Percent recoveries of additional 18 nitroaromatics (compounds 19 through 36) extracted from spiked sand (25 g). The experimental conditions are given in the experimental section.

Exp. 1 and 2:150 atm/50°C/10

ZZ

10 - f

20

30

40

50

60

70

80

90

100

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

EXP-6

10

20

30

40

50

60

70

atm/70°C/60

EXP-7

min.

Exp. 9 and 10:150

COMPOUNDS ^ EXP-8

atm/50°C/60

KX

min

EXP-9

EXP-10

Figure 4. Percent recoveries of 19 haloethers (compounds 1 through 19 in Table V) extracted from spiked sand (Z5 g). The experimental conditions are given in the experimental section.

Exp. 5, 6, 7, and 8:250

[XX

u Q.

u

O u

>

>-

80

90

100

1 10

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

5

o

o a

3

g

C0

00 00

14.

LOPEZ-AV1LA & BECKERT

SFE in Environmental Analysis

189

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

Sample Extraction for Four-Vessel Setup. A l l extractions were conducted with supercritical carbon dioxide. The extraction of the 19 haloethers from the four spiked sand samples (2.5 g each, spiked at 600 ng/g per compound) was performed at 250 atm/60°C/60 min (dynamic). The extraction of 42 organochlorine pesticides from the four spiked sand samples (3 g each, spiked at the levels indicated in Table VI) was begun at 150 atm/50°C/10 min (static), continued at 200 atm/60/10 min (dynamic), and completed at 250 atm/70°C/30 min (dynamic) using 2.5-g samples. The SRS 103-100 standard reference soil (2.5 g) was extracted at 300 atm/70°C/30 min (dynamic) for Fraction 1 and 350 atm/70°C/30 min (dynamic) for Fraction 2. Method Optimization. The experimental conditions for method optimization focused on seven variables, each chosen at two levels (high and low). Two separate tests were performed to see how these variables influence the method performance. The seven variables and their values chosen for each test are presented in Tables III and IV. The group differences for Test 1 (Vp through Wq) were calculated using Equations 1 through 7; those for Test 2 were calculated in a similar manner. Vp V D

V M

Vy Vs Vj Vc

= = = = = = =

l/4(w + x + y + z) l/4(u + v + y + z) l ' ( t + v + x + z) l/4(u + v + w + x) l/4(t + v + w + y) l/4(t + u + x + y) l/4(t + u + w + z) 4

- l/4(s + t + u + v) - l/4(s + t + w 4- x) - l/4(s + u + w + y) - l/4(s + t + y + z) - l/4(s + u + x + z) - l/4(s + v + w + z) - l/4(s + v + x + y)

= p-P = d-D = m-M = v-V = s-S = t-T = c-C

(1) (2) (3) (4) (5) (6) (7)

The relative changes as 100 x change/average recovery at low level were calculated and the data were sorted by variable and compound. The variables were ranked in the increasing order of absolute relative change for each compound and the sum of ranks was calculated. The effect of varying flowrate, pressure, and temperature was investigated using the SRS 103-100 standard reference soil and the Hewlett Packard extractor. The experiments were performed under the conditions given in Table II. Results and Discussion Reproducibility of Two-Vessel-System Extractions. Figures 3a and 3b show the recoveries of the 36 nitroaromatic compounds that were spiked on sand at 600 ng/g per compound and extracted with supercritical carbon dioxide using the Suprex extractor. The data indicate that 17 of the 36 compounds (Compounds 1 through 17 in Figure 3a) exhibited higher recoveries when extracted at 150 to 250 atm and 50 to 70°C. For the remaining 19 compounds, the recoveries were above 60 percent, with most of the values around 80 to 85 percent. It is quite possible that the more volatile nitroaromatic compounds get extracted in the first 5 or 10 min of the extraction; however, at higher pressures, the flowrate of the carbon dioxide is higher and these compounds are lost from the collection vessel by volatilization. The

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

1

Exp. 1 s

Pressure (atm) 300 Temperature (°C) 70 Moisture content (%) 10 Extraction vessel vol. (mL) 4.7 Sample size (g) 2.5 Time (min) 60 Modifier volume (jiL) 250

Variable 300 70 0 4.7 1.0 30 50 300 50 10 2 2.5 30 50

Exp. 2 Exp. 3 t u 300 50 0 2 1.0 60 250

V

Exp. 4

150 70 10 2 1.0 60 50

Exp. 5 w 150 70 0 2 2.5 30 250

X

Exp. 6

150 50 10 4.7 1.0 30 250

Exp. 7 y

Table III. Experimental Conditions for Method Optimization (Test 1)

150 50 0 4.7 2.5 60 50

Exp. 8 z

a

SOURCE: Reprinted with permission from reference 1. Copyright 1990 Preston Publications. Hexane was used as modifier. This experiment was initially set up to investigate not only the polynuclear aromatic hydrocarbons, but also a set of organochlorine pesticides, which was spiked on the standard reference material at a low and a high concentration. The low and high concentrations were obtained by spiking 50 fiL and 250 fiL, respectively, of a hexane solution containing the compounds. Since the organochlorine pesticides could not be detected due to high background, the parameter "analyte spike" was changed to "modifier volume". The eight experiments were performed with the Suprex extractor.

P D M V S T C

Code Letter

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

a

Exp. 1 s

Pressure (atm) 300 Temperature (°C) 70 Volume of toluene (jiL) 500 Collection volume (mL) 5.0 Moisture (%) 10 w Glass beads Static extraction time(min) 30

Variable 300 70 0 5.0 0 w/o 15

300 50 500 1.0 10 w/o 15

Exp. 2 Exp. 3 u / 300 50 0 1.0 0 w 30

V

Exp. 4

The eight experiments were performed with the Suprex extractor.

P D F G M B E

Code Letter 150 70 500 1.0 0 w/o 15

Exp. 5 w 150 70 0 1.0 10 w 30

X

Exp. 6

a

150 50 500 5.0 0 w 30

y

Exp. 7

Table IV. Experimental Conditions for Method Optimization (Test 2)

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

150 50 0 5.0 10 w/o 15

Exp. 8 z

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

192

SUPERCRITICAL

FLUID

TECHNOLOGY

agreement between the duplicate extractions performed in parallel was excellent (within 5 percent). Figure 4 shows the recoveries of the 19 haloethers. Overall, the recoveries were quite good (averaging around 70 percent across the 19 compounds) but the agreement between the duplicate extractions performed in parallel was not as good as in the case of the nitroaromatics, however, it was within 15 percent for most compounds. To determine if the pressure had any effect upon recovery, we compared experiments 5,6 with experiments 9,10 because they were pairs (performed in parallel at different pressures) and they were performed with sand samples spiked under identical conditions. A l l recoveries were slightly higher when extractions were performed at lower pressures. This seems to be in contradiction with what we obtained for experiments 7,8; however, in this case the sand was spiked directly in the extraction vessel and therefore, the data from experiments 7,8 cannot be compared with the data from experiments 5,6. Reproducibility of Four-Vessel-System Extractions. Tables V and VI summarize the data for 19 haloethers and 42 organochlorine pesticides, respectively, that had been spiked onto sand (at the levels indicated in the tables) and extracted with supercritical carbon dioxide using the Suprex extractor. The experimental conditions (pressure, temperature, extraction time) had been established previously using our one-vessel and two-vessel extraction arrangements (1,2). This time, we were primarily interested in establishing the degree of agreement of the four parallel extraction recoveries using the four-vessel extraction system. The results presented in Table V indicate that of the 19 haloethers extracted from spiked sand, 10 compounds exhibited recoveries greater than 84 percent, five compounds were in the range of 75 to 84 percent, and four compounds were in the range of 46 to 74 percent. The agreements of the four parallel extraction recoveries, expressed as the percent relative standard deviations (RSDs) of the four extraction recoveries, ranged from 1.2 to 36 percent, with 15 compounds having RSDs under 10 percent. In reality, the agreements of these parallel extraction recoveries are better than the RSD values in the table indicate because these tabulated values include not only variations due to the extraction, but also variations associated with matrix spiking, transfer of the sample extracts to the autosampler, extract injection, and peak quantitation. The results presented in Table VI for the 42 organochlorine pesticides extracted from spiked sand indicate that SFE with carbon dioxide worked reasonably well for 35 of the 42 compounds (recovery > 50 percent); from those 35 compounds, 30 compound had recoveries > 70 percent. Among the seven compounds with recoveries < 50 percent, two are very volatile (DBCP and hexachlorocyclopentadiene). Three of the remaining five compounds (chlorthalonil, captan, and endosulfan sulfate) gave very poor recoveries with carbon dioxide alone, probably because of their polarity; however, in other experiments we have recovered them quantitatively from spiked sand with carbon dioxide modified with 10 percent methanol (1). Chlorobenzilate recoveries were poor, even when we used carbon dioxide with 10 percent methanol, and captafol was not tested with the carbon dioxide/methanol combination.

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

14.

Table V .

193

SFE in Environmental Analysis

LOPEZ-AVILA & B E C K E R T

Average Recoveries and Percent RSDs for 19 Haloethers Extracted from Spiked Sand with Supercritical Carbon Dioxide

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

a

Compound no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

Compound

Average recovery

% RSD

93.8 84.5 66.4 85.5 81.7 85.3 89.4 89.4 80.6 78.8 87.8 91.5 84.7 88.8 78.8 78.7 59.2 74.0 46.2

1.2 3.3 5.8 3.7 6.5 4.6 3.9 3.9 4.1 7.5 8.6 1.7 12 2.0 3.3 7.7 29 10 36

4-Bromophenyl-phenyl ether Phenyl-4'-nitrophenyl ether 2-Chlorophenyl-4'-nitrophenyl ether 3-Chloropheny 1-4' -nitrophenyl ether 4-Chloropheny 1-4' -nitrophenyl ether 2,6-Dichloropheny 1-4' -nitrophenyl ether 3,5-Dichloropheny 1-4'-nitrophenyl ether 2,5-Dichloropheny 1-4' -nitrophenyl ether 2,4-Dichloropheny 1-4' -nitrophenyl ether 2,3-Dichloropheny 1-4' -nitrophenyl ether 3,4-Dichlorophenyl-4 -nitrophenyl ether 2,4,6-Trichloropheny 1-4' -nitrophenyl ether 2,3,6-Trichloropheny 1-4' -nitrophenyl ether 2,3,5-Trichloropheny 1-4'-nitrophenyl ether 2,4,5-Trichloropheny 1-4' -nitrophenyl ether 2,4-Dibromopheny 1-4' -nitrophenyl ether 3,4,5-Trichlorophenyl-4 -nitrophenyl ether 2,3,4-Trichloropheny 1-4' -nitrophenyl ether 2,4-Dichlorophenyl-3'-methyl-4'-nitrophenyl ether b

b

,

,

a

The number of samples extracted in parallel was four. The spiking level was 600 ng/g. The experiments were performed with the Suprex extractor, b These compounds coelute.

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

194

SUPERCRITICAL FLUID

TECHNOLOGY

Table VI. Average Recoveries and Percent RSDs for 42 Organochlorine Pesticides Extracted from Spiked Sand with Supercritical Carbon Dioxide*

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

Compound no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Compound DBCP Hexachlorocyclopentadiene Etridiazole Chloroneb Propachlor Hexachlorobenzene Trifluralin Diallate alpha-BHC PCNB gamma-BHC Heptachlor Chlorthalonil Alachlor Aldrin beta-BHC delta-BHC DCPA Isodrin trans-Permethrin Heptachlor epoxide Endosulfan I gamma-Chlordane alpha-Chlordane trans-Nonachlor p,p*-DDE Captan Dieldrin Perthane Chlorobenzilate Endrin Chloropropylate p,p'-DDD Endosulfan II p,p'-DDT Endrin aldehyde Endosulfan sulfate Kepone Mirex Methoxychlor Captafol Endrin ketone

Spike level (ng/g) 62.5 62.5 62.5 1250 1250 62.5 125 1875 62.5 62.5 62.5 62.5 125 62.5 62.5 62.5 62.5 62.5 62.5 625 62.5 62.5 62.5 62.5 62.5 62.5 125 62.5 1875 125 62.5 62.5 125 62.5 62.5 125 62.5 62.5 125 62.5 62.5 62.5

Average recovery

% RSD

7.0 22.6 76.8 85.2 56.8 73.5 86.7 89.2 79.4 75.1 80.8 82.2 49.0 65.1 82.5 77.4 87.3 77.8 107 69.1 90.2 84.1 86.1 85.5 85.4 89.1 45.8 88.6 78.1 0 83.3 72.5 83.3 74.2 74.0 75.9 0 64.9 86.1 60.7 36.2 70.1

31 10 4.5 4.1 11 5.3 4.7 4.5 5.4 13 6.9 5.2 10 10 5.1 5.4 4.8 6.8 4.3 23 12 4.7 5.2 4.6 4.3 21 7.6 3.1 5.4 ~

5.0 5.5 10 14 15 3.7 —

6.1 4.8 20 19 6.1

•The number o f samples extracted in parallel was four. The experiments were performed with the Suprex extractor.

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

14.

LOPEZ-AVILA & B E C K E R T

SFE in Environmental Analysis

195

Overall, the agreement of the extraction recoveries (as percent RSDs) of the 42 organochlorine pesticides, obtained for the spiked sand with the four-vessel extraction system, was better than what we had achieved earlier (1) by sequential extraction with the one-vessel extraction system. For example, 26 compounds have RSDs under 10 percent with the four-vessel extraction system while only 12 compounds had RSDs under 10 percent with the one-vessel extraction system. Standard Reference Material. The standard reference soil SRS 103-100 certified for 13 PAHs, dibenzofuran, and pentachlorophenol at concentrations ranging from 19.1 to 1618 mg/kg was extracted with carbon dioxide. The results are presented in Table VII. These results are from two experiments in which we collected two fractions per sample. The data show clearly that under the conditions used a 30-min extraction will not completely remove the target compounds from the matrix. Even after 60 min, the recoveries of benzo(a)anthracene, chrysene, benzo(b+k)fluoranthene, and benzo(a)pyrene were still below 40 percent (in Fractions 1 and 2 combined). The percent RSDs for this soil were higher than those obtained for the spiked sand samples; RSDs for the soil ranged from 9.9 to 29 percent for Fraction 1 in Trial 1 and from 4.9 to 36 percent for Fraction 1 in Trial 2; the ranges were higher for Fractions 2 in both trials because much smaller amounts were recovered in Fraction 2. Method Optimization. The experimental conditions focused on two sets of seven variables listed in Tables III and IV, each chosen at two levels. A full factorial design would have required 128 experiments. Instead, we performed only eight experiments. This design allowed us to estimate the main effects of the two sets of seven variables; however, we could not test for statistical significance of any of these effects because there were no degrees of freedom for the error terms. Tables VIII and IX summarize the recovery data for the eight experiments performed for each test. From these tables, one can calculate the effect on the recovery of each variable at its low and its high value using Equations 1 through 7. To get an overall picture of which variables affect the recovery of the highest number of compounds, the group differences were ranked and the ranks across the 15 compounds were summed up. From these sums, we concluded that in Test 1 recovery was most affected by time (sum of ranks 88) and least affected by the volume of modifier (sum of ranks 31). Pressure and moisture ranked second (sum of ranks 70) and third (sum of ranks 69) in importance after time. The sum of ranks for sample size, temperature, and cell volume are 65, 50, and 47, respectively. Figure 5 shows the relative change in recovery for each compound and for each of the seven variables. When the experiments were performed at the same pressure, temperature, and moisture content but with toluene as modifier and with a static extraction time of 15 or 30 min prior to the dynamic extraction step, then recovery was most affected by moisture content (sum of ranks 88) followed by pressure (sum of ranks 70) and the toluene volume (sum of ranks 68). The fourth variable to influence was the static extraction time (sum of ranks 57). Temperature, volume of collection solvent, and the presence/absence of glass beads were the least important. Figure 6 shows the relative changes in recovery for each compound and for each of the seven variables investigated in Test 2.

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

a

32.4 62.1 19.1 632 307 492 1618 422 1280 1033 252 297 152 97.2 965

±

+ +

± ± ±

+

±

+

± ±

+

± ±

+ 8.2 11.5 4.4 105 49 78 348 49 220 289 38 26 22 17.1 374

Certified value (mg/kg)

Trial 2

Trial 1

Fraction 2 Trial 2

54.0 79.0 52.4 96.3 105 81.2 92.1 79.4 73.1 48.3 32.4 31.1 0 0 52.1 24

---

12 12 10 10 9.9 12 23 17 29 23 22 22

56.9 66.4 44.2 75.9 80.6 63.2 72.1 62.0 52.0 44.7 27.8 27.6 0 0 52.3 34

---

13 4.9 28 15 14 13 17 23 29 36 34 31

0 5.9 0 9.7 9.8 10.4 21.0 16.7 16.0 16.2 11.6 10.2 7.3 4.5 14.8

17 23 21 18 27 15 8.8 7.8 6.0 12 24 17

--

28

13.5 17.1 13.1 19.4 19.9 18.9 34.0 24.7 26.0 27.1 19.2 19.0 12.3 9.3 31.3

23 43 20 24 27 15 24 15 26 22 28 29 46 80 30

Average Average Average Average recovery % RSD recovery % RSD recovery % RSD recovery % RSD

Fraction 1

The number of samples extracted in parallel in each trial run was four. The experiments were performed with the Suprex extractor.

Naphthalene 2-Methylnaphthalene Acenaphthylene Acenaphthene Dibenzofuran Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benzo(a)anthracene Chrysene Benzo(b+k)fluoranthene Benzo(a)pyrene Pentachlorophenol

Compound

Trial 1

a

Table VII. Average Recoveries and Percent RSDs for 15 Target Analytes Extracted from a Certified Soil (SRS103-100) with Supercritical Carbon Dioxide (Fraction l )

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

SOURCE:

a

Naphthalene 2-Methylnaphthalene Acenaphthylene Acenaphthene Dibenzofuran Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benzo(a)anthracene Chrysene Benzo(b+k)fluoranthene Benzo(a)pyrene Pentachlorophenol

Compound 3.1 0.8 13.1 19.9 31.4 37.8 74.5 76.8 60.9 55.7 35.5 32.0 21.4 11.3 40.3

Exp. 1 1.5 13.1 20.9 35.9 44.6 39.4 66.4 37.0 28.8 30.8 15.7 15.0 9.5 4.6 16.4

Exp. 2

Exp. 4 1.5 7.2 31.4 45.7 51.1 47.8 91.5 14.9 60.9 65.8 43.7 41.1 30.9 17.0 50.8

Exp. 3 1.5 2.4 10.5 16.5 16.6 16.7 42.0 27.7 39.5 44.2 37.9 36.0 27.6 18.5 37.8

2.5 8.1 30.4 45.9 50.8 45.9 59.2 44.5 46.6 43.2 29.3 25.9 17.0 9.5 44.1

Exp. 5

Exp. 7 1.5 2.9 6.3 7.5 6.5 7.3 19.3 20.1 20.3 21.1 18.9 18.0 15.5 8.2 18.4

Exp. 6 4.6 7.1 34.6 55.4 64.2 63.4 107.0 63.5 46.3 42.8 19.4 17.6 10.0 5.6 40.2

Percent Recovery

Reprinted with permission from reference 1. Copyright 1990 Preston Publications. The experimental conditions are given in Table III.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Compound No.

a

4.3 27.1 54.5 77.5 80.8 71.5 123.0 80.1 71.6 61.9 29.8 25.0 6.2 7.6 37.1

Exp. 8

Table VIII. Extraction of the SRS103-100 by S F E with Carbon Dioxide under Various Conditions (Test l )

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

a

Naphthalene 2-Methylnaphthalene Acenaphthylene Acenaphthene Dibenzofuran Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benzo(a)anthracene Chrysene Benzo(b+k)fluoranthene Benzo(a)pyrene Pentachlorophenol

Compound 67.9 90.2 57.6 98.6 105 87.0 124 85.1 81.3 57.5 36.5 37.7 30.3 16.5 89.4

Exp. 1

The experimental conditions are given in Table IV.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Compound No. Exp. 3 18.5 67.6 47.1 80.9 87.9 78.0 129.0 90.8 97.7 74.7 59.1 63.6 52.0 31.9 55.9

Exp. 2 30.9 24.2 26.2 4.7 53.7 48.4 78.5 59.5 51.4 37.9 24.2 23.6 9.2 2.1 46.5

3.1 1.6 5.2 8.9 7.5 10.8 49.2 42.2 74.4 58.7 38.1 36.7 17.8 7.2 69.2

Exp. 4

Exp. 7 52.5 75.7 47.1 80.2 85.3 70.3 92.1 60.7 50.3 35.6 19.4 21.9 7.2 8.2 42.7

Exp. 6 49.4 74.1 57.6 62.7 67.4 64.0 130 96.4 113 85.0 54.8 49.8 7.9 2.0 117

Exp. 5 77.2 101 62.8 108 110 95.1 153 101 109 74.2 46.4 46.1 29.6 17.5 147

Percent Recovery

a

12.3 33.8 47.1 88.0 104.0 93.1 174 123 137 96.8 59.9 56.9 13.2 15.4 144

Exp. 8

Table IX. Extraction of the SRS103-100 by SFE with Carbon Dioxide under Various Conditions (Test 2)

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

-

-

-40

-60

-200

1

-180 -

-160 -

-140 -

-120 -

-100 -

-80

— I)

-20

P C V

Compound Number

+

M

p s

~i— 13

M

11

T

p

§ T

8

IT

15

Figure 5. Relative change in percent recovery (when going from low to high) for the 15 target compounds identified in Table V m (test 1). The variables are identified as follows: pressure (P), temperature (D), moisture (M), cell volume (V), sample size (S), time (T), and volume of modifier (C). The 15 compounds are those listed in Table DC (Reproduced with permission from reference 1. Copyright 1990 Preston Publications.)

a:

o o

o

c o

9

0

20 -

40 H>

60 -

80 -

100

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

i 1

D

F

E

F

D

M

E

§

M

e

P

p

3

0 M F

|

P

i

F

fl

B

E

i 5

F

M

B D G

I

7

M

8

G

1

»

M

B

o

u F

g

i 9

M

Compound Number

i

M F

e

B

1



M

B

D

P

9

11

M

B

§

G E

i

M

B

F

P

i

i 13

F

P

M

B

0

g

i

F

M

P

E 0 G

i 15

B

El

f

P

Figure 6. Relative change in percent recovery (when going from low to high) for the 15 target compounds identified in Table IX (test 2). The variables are identified as follows: pressure (P), temperature (D), volume of toluene as modifier (F), volume of collection solvent (G), moisture (M), glass beads (B) and static extraction time (E). The 15 compounds are those listed in Table DC

-

-

-

-

-

-

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

14.

LOPEZ-AVILA & B E C K E R T

SFE in Environmental Analysis

201

The effect of varying flowrate, pressure, and temperature was investigated using the HP extractor. This is the only commercially available extractor that allows flowrates of up to 4 mL/min (as compressed fluid) through the extractor. The recovery data for the 15 target compounds extracted from SRS 103-100 as a function of flowrate (experiments HP-1 through HP-4), pressure (experiments HP-5 through HP-7), and temperature (HP-9 through HP-12) are presented in Figures 7 through 9, respectively. These data indicate that the highest recoveries (except for naphthalene and 2-methylnaphthalene) were obtained at flowrates of 4 mL/min; this was quite evident for higher-molecular-weight PAHs. Since we are seeing increased recoveries by increasing the flowrate (or the amount of carbon dioxide used for extraction), it appears that the S F E rates in this case are at least to some extent controlled by compound solubility in carbon dioxide. Pressure affected compound recovery in several ways: for naphthalene and 2-methylnaphthalene, we observed a slight decrease in recovery by increasing the pressure from 149 to 329 bars. Acenaphthylene, acenaphthene, dibenzofuran, and fluorene recoveries were not affected at all by the change in pressure, and the rest of compounds, except benzo(a)pyrene, showed higher recoveries at 218 bars than at 149 and 329 bars. It is possible that we may have reached a maximum for compound solubility at that intermediate pressure. Finally, recoveries were almost identical at 40°, 60°, and 70°C, but they were consistently lower by about 30 percent at 50°C. We repeated the experiment and were able to duplicate the original data at 50°C. Thus, we concluded that pressures of 350 bars and temperatures of 50 °C give lower recoveries regardless of the fact that the density of the supercritical fluid was at its maximum under these conditions. Conclusion We have demonstrated in this paper that two and four samples can be extracted in parallel with supercritical carbon dioxide without significant impact on data quality. Modifications made to an off-line extractor involved addition of a multiport manifold for the distribution of supercritical fluid to four extraction vessels and of a 12-port, two-way switching valve that allowed collection of two fractions per sample in unattended operation. The only limitation that we have experienced with the fourvessel extraction system was in the duration of the extraction. When working with 2-mL extraction vessels and 50-/*m restrictors, and using the pressure/temperature conditions mentioned above, the 250-mL syringe pump allows us a maximum extraction time of 60 min. During this time, two 30-min fractions can be collected with the present arrangement. Notice Although the research described in this paper has been supported by the U . S . Environmental Protection Agency, it has not been subjected to Agency review and therefore does not necessarily reflect the views of the Agency, and no official endorsement should be inferred. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

IZZI F1

COMPOUNDS F2 WZA

F3

^

F4

Figure 7. Recovery as a function of flowrate for the 15 target compounds identified in Table IX; F l is 1 mL/min; F2 is 2 mL/min; F3 is mL/min; and F4 is 4 mL/min. Experimental conditions are given in Table II.

o or

u > O o

130.0 -

140.0 -

150.0

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

10.0 -

ZZl PI

COMPOUNDS

Figure 8. Recovery as a function of pressure for the 15 target compounds identified in Table DC; PI is 149 bar; P2 is 218 bar; and P3 is 329 bar. Experimental conditions are given in Table II.

01

0

o o u

>Q: UJ >

120.0

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

8

3

w w o

I s

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.

(ZZ) T1

COMPOUNDS

EX3

T4 #

Figure 9. Recovery as a function of temperature for the 15 target compounds identified in Table IX; T l is 40 C; T2 is and T4 is 70 * C. Experimental conditions are given in Table II.

u

z u

\-

o:

o o

>a: u >

150.0

160.0

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

14.

LOPEZ-AVILA & B E C K E R T

SFE in Environmental Analysis

205

Acknowledgment

Downloaded by UNIV OF ARIZONA on January 18, 2013 | http://pubs.acs.org Publication Date: May 8, 1992 | doi: 10.1021/bk-1992-0488.ch014

We acknowledge the assistance of Nikhil S. Dodhiwala who performed most of the SFE work, of Janet Benedicto and Lisa Balch, who performed all gas chromatographic and gas chromatographic/mass spectrometric analyses reported in this manuscript, and of Karin Bauer of Midwest Research Institute, who provided input on the statistical analysis of the preliminary results from our method optimization study. We would also like to acknowledge the assistance of Ashok Shah and Carl Stadler of Suprex Corporation who helped with the design of the dualextraction setup on the Suprex SE-50 system, and of Hewlett-Packard Company who made available to us a Hewlett Packard extractor. Literature Cited 1. 2.

Lopez-Avila, V.; Dodhiwala, N.S.; and Beckert, W.F. J. Chrom. Sci. 1990, 28, 468. Lopez-Avila, V.; Dodhiwala, N.S.; and Beckert, W.F. Off-line Supercritical Fluid Extraction Technique for Difficult Environmental Matrices Contaminated with Compounds of Environmental Significance. Paper presented at the Sixth Annual Waste Testing and Quality Assurance Symposium, Washington, D . C . , July 1990.

R E C E I V E D December 2, 1991

In Supercritical Fluid Technology; Bright, F., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1992.