CONTINUOUS FLOW ANALYSIS (CFA)

Introduction

Continuous analyses of the tracers ammonium (an indicator for biological activity), calcium (a proxy parameter for dust), formaldehyde, and hydrogen peroxide (both compounds closely linked to the oxidation capacity of the atmosphere) were performed on the GRIP ice core (Sigg 1994). This method provides on-line information on a series of chemical species in the ice, yields records with high temporal resolution (sub-seasonal) and uses minimal amounts of ice.

To take further advantage of this technique, we have developed analytical methods for the determination of sodium (a sea salt component), nitrate (a major impurity in polar ice), and conductivity (indicating the total amount of ions in the ice). Additionally, detectors for dust and sulphate provided by collaborating institutes from Copenhagen and Florence will be joined to the existing system to extend its possibilities.

Method

Continuous Flow Analysis (CFA) is based on a steady sample flow combined with an immediate detection of the compound of interest. To obtain a continuous sample stream, the ice core (typically a subcore of 3x3 cm cross section) is melted continuously on a specially developed melt device where the sample is permanently drained off by a peristaltic pump and split into the analytical systems (see Figure 1). The design of the melter prevents contamination of the sample, because only the innermost part of the core section, which does not come in contact with the atmosphere, is used for the analyses. A specific reagent is then injected into each sample stream to produce fluorescent or absorbing complexes with the species of interest. After a certain reaction time (approx. 1 minute) the concentrations of these complexes, which depend directly on the concentration of the original species, are determined.

Figure 1: Schematic Setup of our CFA-System

Thus, an analytical method for CFA measurements has to meet the following requirements:

• small sample flow rate (~1 ml/min),
• flow through operation,
• fast response of the detector to achieve a high resolution,
• sufficient sensitivity and selectivity for polar ice core samples.

Furthermore, we aim to minimise size and weight of the detectors to facilitate field analyses. Therefore, all the spectrometers we use are individually designed and built in our institute.

The CFA setup has the following advantages:

• continuous ice core record with a high spatial resolution,
• field measurements, producing immediate results and preventing concentration changes in the ice core during long term storage,
• highly automated analyses.

An example of the results obtained with our CFA-System is shown in Figure 2. The shallow core was drilled by a german expedition at Kottas Camp, Dronning Maud Land, Antarctica, in 1996/97. As can be seen in Figure 2, sodium, ammonium, calcium, nitrate, and conductivity show a clear annual cycle. Furthermore, there is a phase shift between nitrate and the cations, which indicates that they are subjected to different transport mechanisms.

Figure 2: Firn Core from Kottas Camp, Dronning Maud Land, Antarctica

Method for Continuous Sodium Analyses

After evaluating several methods for sodium determination, we focused on a method based on absorption spectrometry. An enzymatic reaction used for sodium determination of clinical samples which is described by Quiles (1993) has been adapted to concentrations in the range of some ppb's.

To obtain a quantitative measure of the sodium concentration of an aqueous solution, the method takes advantage of the sodium dependent activity of b-Galactosidase, an enzyme which catalyses the hydrolysis of o-Nitrophenyl-b-D-Galactopyranoside. The higher the sodium concentration is, the more o-Nitrophenol is produced, which is detected with an absorption spectrometer at 410 nm.

With our setup the detection limit is approximately 5 ppb. The linear range can be adjusted with the concentration of the substrate and the length of the cuvette. The selectivity of this method over potassium, ammonium, magnesium, and calcium, ions which might also act on the enzyme activity (Boyer 1972), has been measured. The selectivity factors are >100 for magnesium, ~70 for calcium, and ~30 for ammonium and potassium, which is just about sufficient considering the expected concentrations of these ions in Antarctic snow. However, the method needs further improvement concerning its selectivity.

References

Sigg et al., Environ. Sci. Technol. (1994), Vol. 28, pp. 204-209
Quiles et al., Clin. Chem. (1993), Vol. 39, pp. 500-503
Boyer, The Enzymes, 3rd Ed. Vol VII, New York Academic Press, 1972

For more information please send an email to roethlisberger@climate.unibe.ch or bigler@climate.unibe.ch