Background correction erroneous

Incorrect standard reference material, incorrect standard preparation, and incorrect sample preparation can cause incorrect results. In addition, if background correction is not employed when required, it will give rise to erroneous results. 

A significant effort was necessary to obtain accurate results for selenium in blood, plasma, and erythrocytes [27]. Iron causes a spectral interference at the selenium line at 196 nm if a continuum source background corrector is used, causing erroneous results particularly in blood and erythrocytes. Zeeman effect background correction is therefore mandatory for the determination of selenium in clinical samples. In addition, some of the previously recommended chemical modifiers were found to stabilize the different selenium species differently, so that some of them may be lost in the pyrolysis stage. A mixture of palladium and magnesium nitrates was found to solve the problem and prevent any preatomization losses [28]. 

As already mentioned in Section 2.3, molecules show complex and fine-structured spectra. This is due to the numerous rotational, vibrational and electronic transitions possible between the different energy levels. While the first two types of transitions are present in the microwave to IR region, the electronic ones mostly lie in the visible to UV wavelength range and therefore are the dominant causes of spectral interference with atomic lines or erroneous background correction (BC). 

Collective excitation losses occur as broad features about 20 eV above the core level fines. If these features are neglected during Shirley background subtraction highly erroneous correction results are obtained. 

Table 23.4 contains the results of analysis of a urine sample spiked with 50 ng/L natural uranium with an Elan DRC2 optimized as described with a PC3 Peltier-cooled SSI spray chamber. Acquisition included 0.010 s dwell times and five replicates of 1000 scans. Note that simple division of cps by cps by the Elan software gives an erroneously high ratio because of the lack of internal standard corrected background subtraction (see Figure 23.6), whereas taking these into account in Excel gives the correct natural uranium isotope... 

Note that simple division of cps by cps by the Elan software gives an erroneously high ratio because of lack of IS-corrected background subtraction, whereas taking these into account in Excel gives the correct natural uranium U/ U isotope ratio in this case. 

Both polystyrene samples contained an ester and a mineral oil type of lubricant together with a phenolic antioxidant. The lubricants have little absorption in the 280-300 nm region and do not interfere in either method of analysis at the 5-10% concentrations at which they are used in polystyrene formulations. The absorption spectrum of the phenolic antioxidant, however, shows a sharply decreasing non-linear absorbance in the 280-300 nm region and contributes significantly to the background absorption of the test solution in the direct UV spectroscopic method. This invalidates the baseline correction procedure and leads to erroneous styrene monomer values. In the distillation procedure, however, the test solution used for spectroscopy does not contain the phenolic antioxidant and there is no interference in the determination of styrene monomer. 

---