Detector Electrodes



In electrochemical detection, the signal being monitored is a direct response to an ACTUAL CHEMICAL REACTION, as compared to the physical measurement occurring in other LC detectors (e.g., refractive index, absorbance, fluorescence). Electrochemical detectors sometimes get the reputation of being "finicky" compared to optical detectors, but it is important to realize that they behave in a very different manner, and this difference is responsible for the higher degree of sensitivity you can expect from EC methods. You will learn to handle electrochemical detection the same way you would handle any chemical reaction, by considering several variables which can influence the outcome of the reaction. In LCEC, the reaction product that concerns us is the current (i.e., the response, or signal).




ELECTRODES AS ACTIVE SURFACES

The response of an electrode is dependent on the chemical (electrochemical) reaction variables. These include the electrode surface where the reaction is taking place, the mobile phase (reaction medium), and the compound undergoing the reaction. The fundamental electrochemical relationships for this mode of detection were discussed in the BASIC PRINCIPLES section. This section will go into greater detail on some of the practical aspects of the electrodes, including electrode materials, detector cell design, solvent considerations, and the maintenance, service, and performance of each.

The general requirements for electrochemical detection are that the mobile phase be conducting, the working (detector) electrode be chemically inert, and the analyte be electrochemically oxidizable or reducible at the electrode surface in the chosen solution. Solution conductance is met by having an electrochemically inert salt (an ionic conductor) dissolved in the mobile phase. This places some restrictions on the mobile phase composition. Usually, aqueous or partially nonaqueous solutions are used, though a totally nonaqueous solution can be used as long as an appropriate salt is dissolved in it. Since the majority of the liquid chromatographic separations being performed today use reverse-phase packing materials, this requirement is easily met. It is also advisable that the mobile phase be a buffer solution for both electrochemical and chromatographic reasons.

Ideally, the working electrode should be inert to the electrolytic solution and only respond to the analyte in a thermodynamically defined, potential-dependent fashion. Many times this is not the case. The kinetics of heterogeneous charge transfer between the electrode and the analyte, in addition to the reactivity of the electrode itself, enter into the situation. For example, Figure 2.1 shows actual current-voltage curves (normalized hydrodynamic voltammograms; click here for further discussion of these waveforms) for the oxidation of a substituted o-hydroquinone on four carbon-based electrodes.

Figure 2.1. Hydrodynamic voltammograms for various carbon pastes and glassy carbon working electrodes.

The sharpest break with potential occurs on the CP-O while the broadest voltammogram occurs on the CP-W, indicating that faster electron-transfer kinetics occur on CP-O relative to CP-W. The CP-S and glassy carbon electrode materials exhibit kinetics similar to CP-O. Fast electron-transfer kinetics characterized by sharply rising voltammograms improve the selectivity of the overall determination. Some carbon paste formulations (graphite mixed with hydrocarbon and fluorocarbon polymers) show slower kinetics, similar to the CP-W described above. The sensitivity or response for a given amount injected is approximately the same in the diffusion-limited region, but for these materials, a greater potential is usually required for equivalent response.

Electrochemical reactivity can be altered considerably by changing the electrode material. In many cases, this can be highly advantageous. The large hydrogen overpotential characteristic of mercury electrodes in protic solutions extends the attainable negative potential range (past carbon) and makes difficult reduction reactions possible. For this reason, mercury remains the material of choice in these potential regions. However, the reduction of dissolved oxygen, a very facile reaction on mercury over a wide potential range, does not occur until well into the negative-potential region on a glassy carbon electrode. Thus, the oxygen overpotential of glassy carbon is much better than that of mercury (Figure 2.2) and precludes the need for oxygen purging at moderately negative potentials.

Figure 2.2. Hydrodynamic voltammogram for oxygen on glassy carbon and mercury electrodes.

Not all electrode materials will withstand solvents. All carbon-paste formulations are limited to varying degrees (usually not more than 10% organic solvent), depending on the graphite binder. Glassy (vitreous) carbon, platinum, and mercury (amalgamated gold) are far more resistant to organic solvents.

All electrode materials require some surface conditioning or modification before they stabilize to a constant background current level. The conditioning process is observed as a slowly decaying current output from the detector after the electrode is turned on. This may take only a few minutes for an electrode that has been switched off momentarily to as long as a few hours for a freshly prepared glassy carbon electrode at high negative or positive potentials. Longer stabilization times will be required for high-sensitivity operation.




UNIJET DETECTOR CELL

The UniJet detector is a new addition to the BASi line of amperometric detectors for liquid chromatography. The detector has been designed with microbore chromatography in mind. Due to the stringent requirements of minimal dead volume in microbore systems, the UniJet detector was designed as the end fitting of the SepStik microbore column. In order to keep the overall size and the internal volume to a minimum, a radial flow pattern was used (Figure 2.3; this is not wall-jet hydrodynamics).

Figure 2.3. Flow patterns for cross-flow and radial cells.

In comparison to the more traditional approach of a cross-flow cell, the radial flow cell gives improved response at microbore flow rates (< 200 mL/min) and less dilution of the sample before the detector. Figure 2.4 illustrates the improved response for the UniJet cell compared to the classical BASi detector cell under microbore conditions. In addition to the above improvements, the radial flow profile allows for more rapid equilibration of the electrode. The UniJet detector cell does allow for a variety of reference electrodes used without a liquid junction (salt bridge).

Figure 2.4. Comparison of chromatograms from BASi classic and UniJet cells at low flow rates.




LEAVING THE ELECTRODE "ON"

If the LC system is being used on a daily basis, the electrode can be left "ON" continuously. Make sure you have plenty of mobile phase in the reservoir, or route the outlet from the LCEC cell to the reservoir so you recycle the mobile phase until you return. The startup time the following morning is substantially reduced.