Matrix-Free Calibration Solutions

Data for FAAS are from Welz and Sperling (1999) (limits of detection are defined as 3 s and all were determined using matrix-free calibration solutions) Stoeppler (1991) (1st edition of this book data are reported... 

The first step of a method development in GF AAS is usually an optimization of the pyrolysis and atomization temperatures by establishing pyrolysis and atomization curves using a matrix-free calibration solution as well as at least one representative sample or reference material. The pyrolysis curve exhibits the integrated absorbance signal obtained at a fixed atomization temperature as a function of the pyrolysis temperature, as shown schematically in Figure 8.13. 

Calibration Solutions (Matrix Free) Clean solutions of the analytical standard (and SIS if appropriate) used to calibrate instrument response use of this clean solutions can lead to failure to correct for variations in extraction efficiency, matrix effects (ionization suppression), instrument response etc. (see Chapter 8). 

For both methods (A, B), a calibration should be carried out in matrix-free solution over a defined working range, in order to be able to establish the procedural characteristic data (SxoA> the determination... 

The slop of a calibration curve expressed in the appropriate units (e.g. microamps, millivolts, or hertz) per millimole or micromole should always be provided both for a matrix-free solution and for a sample in its matrix. The corresponding LOD should also be reported in both cases. In the absence of a blank the signal 3 X larger than the noise determines the LOD. The limit of quantification is typical 10 x LOD. 

For calibration, microlitre amounts of a gravimetrically prepared standard solution of DMS in ethanediol are injected via the T-fitting into the helium line (see Fig. 24-1), as a sample of degassed (DMS-free) seawater is loaded into the purge vessel. This procedure provides a matrix-matched calibration and minimizes systematic errors by automatically correcting for degassing efficiency and potential DMS losses within the purge and trap unit. To avoid subsequent DMS release from particulates, the DMS-free seawater should be prepared from a filtered sample, ideally from deep water of low DMS concentration. 

The main drawback is the existence of matrix effects and the possible non-ideality of the solvents mixture. These imply that ideally one should determine the calibration curve by adding standard solutions of the solvents of interest to the sample matrix, free of solvents. Because it is very difficult to obtain such a sample matrix, the classical standard addition method is recommended. It consists of adding to the sample matrix to be analyzed a known amount of the solvents to be determined. This method requires two analyses for the final calculation but the main advantage is that the matrix effect is overcome. 

The approach presented above is referred to as the empirical valence bond (EVB) method (Ref. 6). This approach exploits the simple physical picture of the VB model which allows for a convenient representation of the diagonal matrix elements by classical force fields and convenient incorporation of realistic solvent models in the solute Hamiltonian. A key point about the EVB method is its unique calibration using well-defined experimental information. That is, after evaluating the free-energy surface with the initial parameter a , we can use conveniently the fact that the free energy of the proton transfer reaction is given by... 

Figure 5.60 Calibration curves for the diarrhetic shellfish poisons in (i) standard solutions in methanol (O), and (11) standard solutions in poison-free scallop extract solutions ( ) (a) pectenotoxin-6 (b) okadaic acid (c) yessotoxin (d) dinophysistoxin-1. Reprinted from J. Chromatogr., A, 943, Matrix effect and correction by standard addition in quantitative liquid chromatographic-mass spectrometric analysis of diarrhetic shellfish poisoning toxins , Ito, S. and Tsukada, K., 39-46, Copyright (2002), with permission from Elsevier Science.

The mode of injection in GC-based methods can affect the recoveries of diazinon. In a study of the determination of organophosphorus pesticides in milk and butterfat, it was found that the recoveries of diazinon from butterfat, calculated relative to organic solutions of standard compounds, were 125% and 84% for splitless and hot on-column injections, respectively (Emey et al. 1993). Recoveries from milk were not dependent on the mode of injection. It was concluded that the sample matrix served to increase diazinon transfer to the GC column by reducing thermal stress imposed on the analytes and by blocking active sites within the injector. Therefore, on-column injection should be used in order to prevent bias when organic solutions of standard compounds are used for quantitation if this is not possible, the matrix must be present at low concentrations or the calibration standards must be prepared in residue-free samples to avoid unknown bias. 

By definition, an interference effect occurs when the analytical signal is changed by the sample matrix compared with the reference or calibration standard, typically an acidified aqueous solution. This article is only concerned with nonspectral interferences in ET-AAS spectral interferences are considered elsewhere. It has been demonstrated in ET-AAS that the atomization efficiency (conversion of analyte to free atoms) is 100% for the majority of elements in simple solutions, which means that, in most cases, only negative nonspectral interferences can occur, i.e., the signal can only be reduced by the presence of... 

A number of equilibrium ion-exchange methods have been developed to measure free metal ion concentrations in waters. This approach involves the equilibrations of a small volume of resin with a sample. Following the attainment of equilibrium, the amovmt of metal adsorbed is measmed. Calibration is achieved by using matrix-matched solutions of known free metal ion concentration. 

The detection limit for a particular application of interest is easy to determine. One starts with solution free of measured ion, whose matrix resembles that of the sample solutions to be tested, and adds successive portions of measured ion to span the range 10 to 10" M. A calibration curve is constructed from the data thus obtained, and the detection limit is evaluated as shown in Fig. 51. If conditioning solutions are later to be used (to free bound measured ions, bind interfering ions, etc.), they should also be used here, of course. If, as is to be expected, the EMF value in the measured ion-free solution drifts, the value after three minutes should be recorded. 

---