Estimation of background continuum emission intensity of inductively coupled plasma for correction of fast changing background

Abstract

A method to estimate background continuum emission in inductively coupled plasma-atomic emission spectrometry (ICP-AES) is proposed. The method assumes that the ICP background continuum emission is mainly due to recombinative radiation in the plasma. The emission intensity at an arbitrary wavelength is related to the temperature and electron number density of the plasma. Furthermore, the ratio of background continuum emission intensity at two arbitrary plasma excitation conditions is a constant for all wavelengths. Therefore, if there is a change in plasma excitation conditions, the new background continuum emission intensity at an arbitrary wavelength can be calculated from the initial background emission intensity and the emission intensity ratio that can be measured accurately at another convenient wavelength. In this paper, rather than the emission intensity ratio, the ratio of the differences in intensities is used to remove the contribution of dark current in the measured intensity, if any. The ratio is termed 'f-factor'. Firstly, the background emission intensities at two reference plasma excitation conditions are measured. The f-factors of these two background emission intensities are assigned values 0 and 1, respectively. When the plasma excitation condition changes, the f-factor relative to the reference conditions at an arbitrary wavelength can be calculated from the new emission intensity. The background emission intensity at other wavelengths can then be determined from their corresponding emission intensities at the reference conditions and the new f-factor. Background emission intensity of the ICP (λ=200-460 nm) is measured using an ICP spectrometer with a segmented-array imaging detector. The ICP forward power was varied from 800 to 1500 W. In agreement of the model, the f-factors at any wavelengths remain constant at all forward power used. The proposed method was compared to off-peak background correction method. Analyte concentrations were selected to give signal-to-background ratio of ∼1. The corrected analyte emission intensities using both background correction methods agree, except for Ca 396.847 nm, which suffers from spectral overlap from a nearby hydrogen emission line. The ability of the proposed method for background correction of fast changing background emission was demonstrated using direct sample insertion and a direct reading multi-channel spectrometer. © 2002 Elsevier Science B.V. All rights reserved.