Improved performance using internal standardization in inductively-coupled plasma emission spectroscopy

Abstract

Significantly improved performance in ICP emission spectroscopy is possible using internal standardization given the proper choice of operating conditions. A dual monochromator detection system was used to study the correlation between two emission signals recorded simultaneously as several ICP parameters were varied. The behavior of the signal formed by dividing the analyte and reference signals was also measured. The noise behavior of twenty analytically important elements was compared to that of a manganese ion line used as the internal standard for a single, compromise choice of ICP operating parameters (RF power = 1250 W, carrier gas flow rate = 0.7 or 0.91 min , and viewing height = 20mm). The elements differed widely in excitation and ionization energies. Sample concentrations were chosen so that the analyte emission signals were sufficiently intense that the noise and drift contributions due to plasma background and shot noise could be ignored. Under these conditions very good signal correlation ( r ⩾ 0.95) and similar noise proportionality factors were observed leading to noise reductions of ⩾ 10 × after signal division in almost every instance. After signal division the noise, defined as the relative standard deviation of the emission signal, (time constant = 1.0 s, n = 20) was less than 0.1 %. The closely similar noise behavior is attributed to sample density fluctuations arising from corresponding fluctuations of the sample aerosol density. Good correlation is also observed between fluctuations in the plasma background simultaneously detected at two widely separated wavelengths when appropriate steps are taken to reduce shot noise. At carrier gas flow rates above 1.01 min and at viewing heights below 15mm or above 20mm, the signal correlation is reduced or absent, yielding little or no improvement using internal standardization. The use of internal standardization leads to other improvements in analytical performance in addition to noise reduction. Drift is reduced, and nebulizer performance is improved in that it is less sensitive to variations in liquid sample uptake rate, or nebulizer instabilities associated with high solids content sample introduction. The divided signal takes much less time to reach equilibrium following introduction of a new sample.