Jump to main content or area navigation.

Contact Us

EPA's Region 6 Office

Serving: Arkansas, Louisiana, New Mexico, Oklahoma, Texas, and 66 Tribal Nations

Automated Liquid-Liquid Extraction of Semivolitle Analytes


Poster paper presented at the 1995 WTQA meeting in Washington, D.C.

Rick McMillin, Mike Daggett, Diane Gregg, and Lisa Hurst, U.S. Environmental Protection Agency, Region 6 Lab, Houston, Texas, 77099;

Kevin Kelly, Ph.D., David L. Stalling, Ph.D., Nancy L. Schwartz, Laboratory Automation, Incorporated, Columbia, Missouri, 65202.

Abstract

Organic extractions have traditionally been very labor intensive and time consuming. Many variables such as bath temperatures, emulsions, and rate of concentration can directly affect analyte recoveries by traditional methods. The Houston Lab has been involved in evaluating equipment and methodology that would reduce turn-around time while maintaining a high level of quality control. The lab's current shift has been away from traditional methods to improved automated techniques. Their investigation of new extraction techniques have included an automated liquid-liquid extraction device produced by Laboratory Automation, Inc. (ABC Instruments, now a part of O.I. Analytical), called the ExCellTM.

The ExCell is being compared in this study with various other extraction techniques. It was determined that the comparison would focus more on the Accelerated One-Step (AOS) than the other techniques since AOS is now the most prevalent technique used in this lab.

Electrically assisted extraction (ExCell) more closely mimics separatory funnel extractions (equilibrium based) than the continuous extraction device in physical interaction. It uses an innovative electric field to provide the water/solvent mixing action as the water sample passes up through aliquots of solvent. Because of the replacement of mechanical mixing with electrically actuated dispersion, emulsions were less likely to be produced with this technique than with separatory funnel extractions. This technique is automated for most of the extraction process.

For the purposes of this test, the Labconco RapidVap N2TM was used to concentrate all samples from the ExCell, traditional continuous extractor, and separatory funnel extractions. The main goal of the lab was to evaluate the productivity enhancements that this device could provide, and attempt to measure how equivalent this technique is to other techniques currently employed. Equivalency was measured by spiking various matrices with 54-64 semi-volatile target compounds, extracting the samples, and comparing the results. Precision, accuracy, and MDL data are presented. The extractions were all carried out at a single acid pH (< 2). Analyses of the extracts were performed by method 8270.

For each matrix evaluated, seven replicates at low level (10ug/L) were extracted for MDL determination, and three replicates were extracted at a high level (400 or 500ug/L) for extraction capacity evaluation. The matrices evaluated were DI water, ground water, waste water, and TCLP buffer #1 (pH 4.93, +0.05). Not all extraction techniques could be compared in all matrices due to time constraints.


Table 1:

ESTIMATED TIMES FOR SAMPLE PREPARATION

(in minutes -- for six samples)

Function

ExCell

AOS

SF

CE

 

Tech Time

Total Time

Tech Time

Total Time

Tech Time

Total Time

Tech Time

Total Time

Equipment Prep (wash, etc)

35

35

60

60

60

60

60

60

Time of Extraction

0

210

0

360

0

45

0

1080

Cooling Step

0

0

0

0

0

0

0

60

Drying Step (+ prep)

20

20

0

0

15

15

45

45

Breakdown / Washing

20

20

60

60

35

35

35

35

Concentration Step*

30

180

1

10

30

180

30

240

Nitrogen Blow-down Step**

30

120

30

120

30

120

30

120

Total Time

135

585

151

610

170

455

200

1640

Amount of Solvent Used

240

100

250

500

* The RapidVap N2 was used for all methods that required separate concentration.

** Same nitrogen blow-down method was used for all extraction methods.

Results

Recoveries for all compounds averaged for all replicates are presented in table 2. Precision for the same data is presented in table 3.


Table 2:

Average Percent Recovery for All 56 Compounds Low Level

Matrix

ExCell

AOS

SF

CE

DI water

54

70

64

67

Ground Water

71

84

na

na

Waste Water

62

73

na

na

TCLP Buffer

56

74

na

na


Table 3:

Average %RSD for All 56 Compounds Low Level

Matrix

ExCell

AOS

SF

CE

DI water

7

13

8

10

Ground Water

9

8

na

na

Waste Water

8

9

na

na

TCLP Buffer

16

12

na

na


CONCLUSIONS

Although some recoveries were slightly lower using the ExCell extraction, this technique can reduce the cost and complexity of sample preparation while increasing worker safety. Therefore instrument method optimization is being attempted to bring recoveries closer to those obtained using other extraction techniques. Already modifications have been made that may improve recoveries while maintaining the automation features. Most compounds that did not perform well by the ExCell were not CLP target compounds, or did not perform well by the SF method. For those that were CLP compounds (4-chloroaniline), this is a concern that will be investigated further. Ruggedness of this method needs to be investigated further with a wider variety of real sample matrices. This instrument could be a real benefit to the overall productivity of the organic lab.


Important note

Further studies proved that spiking these compounds with MeCl instead of MeOH lead to a negative bias of these results for the ExCell device. See study #2, "Performance Evaluation and Method Equivalency Study for Automated Electrically Assisted Liquid-Liquid Extraction", presented at the 1996 Pittsburgh Conference.

Jump to main content.