Digital filtering in stripping analysis

Stripping voltammetry has been shown to be an extremely valuable technique for the trace analysis of both organic and inorganic species and capable of very low limits of detection (LODs). In practice, the LOD in stripping voltammetry is dictated by the magnitude of noise present in the actual measurement; this is so because, at the limit of very low concentrations, the amplitude of the noise may be comparable to, or even greater than, the magnitude of the stripping peak current, this situation resulting in low signal-to-noise (SN) ratios. In this work, an attempt to tackle the problem of noise in some applications of stripping voltammetry in flowing solutions is reported. The proposed scheme is based on the implementation of digital filtering that makes use of a variety of filtering techniques. Both infinite impulse response filters (IIRFs) and finite impulse response filters (FIRFs) were utilised and assessed. The programming task was implemented in a commercially available, novel, icon-based language that is more operator-friendly than most text-based traditional alternatives. It was found that digital filtering provides a powerful and accessible tool for discriminating against noise even in cases where the SN ratios are very unfavourable. The relevant parameters, involving the choice of the type of filter, cutoff frequencies, leakage effects and comparison of various filtering windows, are discussed.