Jump to main content.


Trace Organic Analysis

Capillary Electrophoresis/laser-induced Fluorescence in Groundwater Migration Determination

William C. Brumley1 and Clare L. Gerlach2

1 Dr. Brumley is Research Chemist, U.S. EPA, National Exposure Research Laboratory,
P.O. Box 93478, Las Vegas, NV 89193-3478, U.S.A.; tel.: 702-798-2684; fax: 702-798-2142
e-mail:brumley.william@epa.gov
2 Ms. Gerlach is Senior Scientist, Lockheed Martin, Las Vegas, NV

Original citation:
W. C. Brumley and C. Gerlach, "Capillary Electrophoresis/Laser-induced Fluorescence in Groundwater Migration Determination," Am. Lab., 31, 45-49 (1999).

Some of the more difficult problems facing environmental scientists involve analyzing phenomena that occur underground, such as plumes from leaking underground storage tanks, seepage from buried waste drums, and the complex features that influence groundwater migration.1 The U.S. EPA, through its Office of Research and Development, designs, modifies, validates, and evaluates new methods for quantifying compounds of environmental concern.

The National Exposure Research Laboratory Environmental Sciences Division in Las Vegas, NV (ESD-LV) has conducted research in capillary electrophoresis (CE) and has applied this research successfully to a groundwater migration study.2, 3 Results indicate that CE with laser-induced fluorescence (LIF) detection can be a valuable quantitative method for detecting the fluorescent dyes that are commonly used as tracers in groundwater migration studies. Using LIF, detection limits in the ng/L range can be achieved. Liquid chromatography (LC), LC-LIF, and CELIF have previously been applied in various tracer experiments.4-6

FUNDAMENTALS OF CAPILLARY ELECTROPHORESIS

Capillary electrophoresis is a powerful separation technique that is based on the characteristic electromigration of ions through a capillary column. Capillary zone electrophoresis (CZE), also known as free zone electrophoresis, is a separation that is based on the tendency of mixtures of ions to separate into sharply defined zones of identical ions, which then migrate at a unique velocity as a result of their characteristic mobility. CZE is easily interfaced with optical detection methods (such as UV-VIS absorption), indirect detection (based on UV or fluorescence), or LIF, where on-column detection is usually carried out.

In basic CZE, a fused-silica capillary (typically 10 Ám i.d.--approx. 150 Ám i.d., with a length of about 20-100 cm) filled with buffer connects two buffer reservoirs containing electrolyte. A high-voltage power supply (up to approx. 30 kV) connects the reservoirs via the buffer-filled capillary. Analyte bands in CE travel with flat profiles and produce extremely narrow peaks. Reported values range from 250,000 to 1,000,000 theoretical plates, with exceptional values in the millions. This efficiency is considerably higher than that usually observed with HPLC or ion chromatography.

One of the issues facing all capillary format separations is the achievable concentration detection limit. In general, capillary separations exhibit excellent mass sensitivity where on-column detection of amounts of low pg to 100 attograms (ag) can realistically be achieved. However, only very small volumes can be injected under normal conditions (10-50 nL), and this results in relatively high concentration detection limits, especially for UV detection. A complicating factor in all capillary separations is the relatively short optical path available (no greater than the column diameter for on-column detection) together with the limited ability to focus available light onto the small region of the detection window and the inner channel of the capillary where separations occur.

As a basis for comparison, the figures of merit for an analytical technique in analysis range from about 1 ng/ÁL (ca. ppm) injected to about 1 pg/ÁL (ca. ppb) typical of flame ionization detectors (FID) and electron capture detectors (ECD), for example. Thus, UV detection in CE is near the upper limit of this range at about 10-6 M with 10-7 M (ca. 10 ppb, 10 ng/mL, or 10 pg/ÁL) achievable for commercial extended light path detectors.

LASER-INDUCED FLUORESCENCE

LIF detection overcomes the optical limitations of the capillary in two ways. Intense light is available from the laser, and this light can be efficiently focused into the narrow channel of the capillary in the window. Typically, detection limits range from 10-7 to 10-11 M and theoretically/experimentally down to a single molecule.7

CE/LIF

Fluorescent dyes, one class of popular groundwater tracers, are ideal substrates for CE-LIF techniques. They are usually anionic dyes (aromatic sulfonic acids or aromatic carboxylic acids) with large extinction coefficients (80,000 cm-1M-1 or higher) and high quantum efficiencies (fluorescent quantum yields near 1.0). Thus, a simple CZE separation presents itself for their analysis, and LIF is uniquely appropriate for their detection because many laser lines correspond to the absorption maxima of dyes.

Every analytical problem presents two limiting conditions for its resolution. The first concerns the absolute sensitivity (and related detection limit) of the technique. The second is the "chemical noise" presented by the particular matrix of the sample. In the case of CZE-LIF, a very powerful separation is coupled with a very sensitive detection technique. Thus, improved selectivity, reliability, and detection limits are realized with the coupled system.

Side-by-side comparison of the synchronous scan spectrofluorimeter with CZE-LIF illustrates these points clearly. Tables 1 and 2 show spectrofluorimetry results and CELIF results for fluorescein and Tinopal (Ciba-Geigy Corp., Greensboro, NC), respectively, for six pad extracts. These pads, common in groundwater tracer studies, are suspended deep in the monitoring well to collect sample for subsequent extraction and analysis. The CE-LIF instrumentation was customized for the two compounds. For fluorescein, the spectrofluorimeter used was a Spex Fluorolog 2 (Instruments, S.A., Edison, NJ), a double monochrometer instrument (emission and excitation) with 450-W xenon lamp. A synchronous scan with an offset of 20 nm was used. Figure 1 shows the spectrofluorimetry profiles for the receptor pad extracts containing Tinopal and fluorescein. A P/ACE model 5000 CE system (Beckman Instruments, Fullerton, CA) was fitted with a 57 cm x 75 Ám i.d. capillary. The detection method was an argon laser at 488 nm with a notch filter (also at 488 nm) and a bandpass filter. The analysis was conducted at room temperature. The capillary was equilibrated with buffer at the beginning of the experiment and washed with alkali, water, and buffer between the sample injections with direct on-column detection. An internal standard, erythrosin B, was used to normalize corrected peak areas that may have had some variation due to the 5-sec pressure injections.8

Image of Spectrofluorimetry results - contact brumley.william@epa.gov for further information

Figure 1. Spectrofluorimetry results for extracts of receptor pads containing Tinopal and fluorescein (at about 100 and 200 ppt, respectively).

Table 1. Spectrofluorimetry results versus CE-LIF results for fluorescein.

Sample
Spectrofluorimetry results
(ppt in pad extract)
CE-LIF results
(ppt in pad extract)
1
90.0
95.0
2
200.0
153.0
3
90.0
180.0
4
100.0
62.0
5
200.0
147.0
6
30.0
ND*
*ND = not detected

Table 2. Spectrofluorimetry results versus CE-LIF results for Tinopal.

Sample
Spectrofluorimetry results
(ppt in pad extract)
CE-LIF results
(ppt in pad extract)
1
70.0
196.0
2
70.0
-
3
100.0
171.0
4
ND*
88.3
5
ND*
112.0
6
ND*
59.0
* ND = not detected  

For Tinopal, the same spectrofluorimeter was used for the fluorescein work. An optical bench was constructed with a cylindrically symmetric capillary holder to ensure alignment with laser beam, slit, and capillary window. The bench was fitted with a fused-silica capillary (Polymicro Technologies, Phoenix, AZ), 57 cm ┤ 75 Ám i.d., with 50 cm to the detector; LIF detection using the 354-nm line of the HeCd ion laser, model 7203N (Liconix, Santa Clara, CA); and two emission filters (Omega Optical, Battleboro, VT) in series.

Groundwater migration studies

This recent evaluation of CZE-LIF for use in groundwater migration studies provides an example of practical applications of an innovative technology. The study was conducted in a region in which a U.S. EPA Resource Conservation and Recovery Act (RCRA) is located next to a Superfund site. The purpose was to determine migration (if any) from the RCRA site to the Superfund site. The contamination at the two sites was similar in some ways: organic solvents, tars, and various chlorinated hydrocarbons were present. If seepage from the RCRA site were misidentified as having come from the Superfund site, financial and legal responsibility would be laid in the wrong quarter. Groundwater migration studies typically involve the injection of fluorescent dyes into injection wells, followed by timed sampling of monitoring wells nearby. In the study, the tracer dyes fluorescein, Tinopal, rhodamine WT, and eosine Y were used. These groundwater migration studies involve a tremendous and rather imprecise dilution of the injected dyes due to the volume of the groundwater. For example, in this case, 30 lb of Tinopal CBS-X (fluorescent whitening agent 351) was injected into the well following 800 L of potable primer water and followed by about 8000 L of potable chaser water. Thus, very low concentrations of tracer are expected in the groundwater samples. It is necessary to reconcentrate samples taken at the monitoring well and to have analytical methods capable of detecting the tracer at the parts-per-trillion level in the measured solution. Considering the day of injection as Day 0, samples were taken at days 0, 3, 7, 14, 21, 28, and then bi-weekly until completion of the project. Samples were taken three ways: on activated coconut charcoal detector pads (for fluorescent dyes), on cotton dye detector pads (for the optical brightener), and water samples (to verify the representativeness of the dye concentrations found on the detector pads). All samples were kept under chain-of-custody conditions.1 The pads were suspended by fishing line far enough into the well to maximize water flow through the detector. The cotton pads were analyzed on site using long-wave UV. The charcoal detectors were brought to the ESD-LV for sample preparation and analysis. Figures 2 and 3 show the CZE-LIF electropherograms for Tinopal and fluorescein, respectively. The internal standard peaks can be seen (6.45 of the Tinopal profile; 3.30 on the fluorescein profile). Solid-phase extraction (SPE) was performed to concentrate the dilute samples. Figure 4 shows the CZE-LIF profile of the SPE extract of fluorescein. Here the need for separation is apparent. An additional component in water that is not fluorescein fluoresces at 488 nm, but exhibits a different migration time than fluorescein.

Image of Tinopal  - contact brumley.william@epa.gov for further information

Figure 2. Tinopal by CZE-LIF (88 ppt).

Image of Fluorescein - contact brumley.william@epa.gov for further information

Figure 3. Fluorescein in receptor extract (62 ppt).

Image of Fluorescein - contact brumley.william@epa.gov for further information

Figure 4. Fluorescein in SPE sample of groundwater (342 ppt).

The current analytical method for groundwater migration studies is synchronous scans using spectrofluorimetry. This method was the standard against which CZE/LIF results were compared. LIF detection allows somewhat lower detection limits to be achieved.

Though low ppt ranges can be achieved by both LIF and spectrofluorimetry, LIF has considerably more specificity due to the separation associated with characteristic migration times. The separation also allows greater discrimination against background responses. Low detection limits are important in two respects: in determining trace levels in large volumes of water, and in determining background levels. In this case, though background levels might be assumed to be zero, in fact, minute quantities of fluorescing materials (usually from previous groundwater studies) can be misleading.

Other applications of CE in environmental analysis

Capillary ion electrophoresis (CIE) has been used as a multianalyte separation technique for environmental anion and cation analysis.9 It is currently being considered by the U.S. EPA for inclusion in the methods compendium SW-84610 as an anion analysis method (Method 6500). This method is appropriate for the analysis of dissolved inorganic anions in drinking water, ground water, surface water, and wastewater.11 Electrokinetic chromatography (EKC) is another capillary format based on electroosmotic (EO) flow. In this process, C18 derivatized silica particles form the stationary phase in packed chromatographic columns. Methods of this type are particularly suited to neutral compounds and hydrophobic molecules like PNAs (polynuclear aromatic compounds).12 Ongoing research at the ESD-LV includes investigation of the use of CE for several compounds of environmental and human exposure interest. The laboratory expects to use CE in the analysis of tissue samples from fish as part of a national survey. The continued growth and acceptance of CE for environmental applications will depend on successful and diverse applications like these. This technology provides another choice for environmental chemists whose quest for faster, less expensive, and more reliable methods drives commercial development and regulatory acceptance.

REFERENCES

  1. Ferguson, P. The use of spectrofluorimetry and capillary electrophoresis/laser-induced fluorescence for the detection of fluorescent dyes in groundwater migration studies. Thesis. Las Vegas, NV: University of Nevada, 1997.
  2. Brumley, W. C., Ferguson, P. L., Grange, A. H., Donnelly, J. R., Farley, J. W. Applications of capillary electrophoresis/laser-induced fluorescence detection to groundwater migration studies. J Cap Elec 1996; 3(6):295-9.
  3. Ferguson, P. L., Grange, A. H., Brumley, W. C., Donnelly, J.R., Farley, J.W. Capillary electrophoresis/laser-induced fluorescence detection of fluorescein as a groundwater migration tracer. Accepted for publication. Electrophoresis 1998(19).
  4. Suijlen, J., van Leussen, W. Poster presentation. HPCE. Orlando, FL: Jan 25-28, 1993.
  5. Suijlen, J, van Leussen, W. Coastal estuarine studies 1990; 36:181-8.
  6. Van Soest, R.E.J., Chervet, J.P., Ursem, M., Suijlen, J. LC.GC Int 1996; 9:586-93.
  7. Castro A, Shera EB. Appl Opt 1995; 34:3418-22.
  8. Brumley, W. C., Brownrigg, C. M.J Chromatogr 1993; 646:377-89.
  9. Krol, J., Romano, J., Benvenuti, M. Real-world use of capillary ion electrophoresis. Envir Test Anal 1998; 7(1):20-2.
  10. U.S. EPA Test Methods for Evaluating Solid Waste, Rev 2, Dec 1996. Washington, DC: U.S. EPA.
  11. Fordham, Jr., O. M. SW-846 Method 6500: Anions by CIE. Environ Test Anal 1998; 7(1):13.
  12. Yan, C., Dadoo, R., Zare, R. N., Rakestraw, D. J.Anal Chem 1996; 68:2726-30.

Trace Organic Analysis Home Page
Analytical Environmental Chemistry
[ Environmental Sciences ]
[ Office of Research Development]
[ National Exposure Research Laboratory]
Author: William C. Brumley


Local Navigation


Jump to main content.