Multicomponent determinations

It has been seen that UVWIS absorption peaks are generally broad, so if there are two compounds, X and Y, in solution, it is likely that they will not be completely resolved from each other. That is, both X and Y contribute to the absorbance at most wavelengths. It is possible to calculate the concentrations of X and Y from a series of measuranents. Measurements must be made at a number of wavelengths equal to the number of components in the mixture. In this case, there are two components, so two wavelengths are needed. Caveats for this approach are that the two components each follow Beer s law and that the two components must not interact in solution their absorbances must be additive. 

This same approach can be used for a mixture of three components. More complex mixtures can be unraveled through computer software that uses an iterative process at multiple wavelengths to calculate the concentrations. Mathematical approaches used include partial least squares (PLS), multiple least squares, principle component regression, and other statistical methods. Multicomponent analysis using UV absorption has been used to determine how many and what type of aromatic amino acids are present in a protein and to quantify five different hemoglobins in blood. 

---

R. Tauler, A.K. Smilde, J.M. Henshaw, L.W. Burgess and B.R. Kowalski, Multicomponent determination of chlorinated hydrocarbons using a reaction-based chemical sensor. 2 Chemical speciation using multivariate curve resolution. Anal. Chem., 66 (1994) 3337-3344. 

C.A. Scolari and S.D. Brown, Multicomponent determination in flow-injection systems with square-wave voltammetric detection using the Kalman filter. Anal. Chim. Acta, 178 (1985) 239-246. 

Tauler R., Smilde A.K., Hemshaw J.M., Burgess L.W., Kowalski B.R., Multicomponent Determination of Chlorinated Hydrocarbons Using a Reaction-based Chemical Sensor. Part 2. Chemical Speciation Using Multivariate Curve Resolution, Anal. Chem. 1994 66 3337-3344. 

The multicomponent determination of the three xylene isomers and ethylbenzene in a cyclohexane solution has been described by Baumann (1962). The bands at 795.2, 768, 741.2, and 696.3 cm were selected as analytical bands. Table 5.1-1 lists the ab.soiption coefficients of the pure compounds, measured in cyclohexane solution, as well as the absorbance values of the unknown at the chosen wavenumbers. With the procedure of Sec. 5.1.3.5 the concentrations of the components can be calculated. The actual concentrations in mol L were ethylbenzene 0.07912, o-xylene 0.07661, m-xylene 0.05105, p-xylene 0.04672. 

Multivariate curve resolution can be used for the analysis of augmented sets as a way of reinforcing conclusions on peak purity, improving the resolution of overlapping compounds, and performing multicomponent determinations in the presence of interferences. 

Ghasemi J, Abbasi B, Niazi A, Nadat E, Mordai A. Simultaneous spectrophotometric multicomponent determination of folic acid, thiamine, riboflavin and pyridoxal by using double divisor-ratio spectra derivative-zero crossing method. Anal Lett 2004 37 2609-23. 

L. L. Juhl, J. H. Kalivas, Evaluation of experimental designs for multicomponent determinations by spectrophotometry, Anal. Chim. Acta 207 (1988) 125. 

The use of kinetic methods of analysis based on noncatalytic reactions has leveled off in the last few years, especially as regards practical applications in any case some of them are the best choices available to analytical chemists for individual and multicomponent determinations of species in a wide variety of real samples. This growing use of noncatalytic kinetic methods can largely be ascribed to the increased automation of reaction rate-based determinations, especially in organic analysis. Applications of these methods to real samples lie in various areas those of environmental, clinical, pharmaceutical, and nutritional interest are discussed in some detail below on account of their great significance. 

An important field of application of multicomponent determination is in the continuous monitoring of nitrogen and phosphorous in wastewater treatment plants. Figure 2.16 in Chapter 2 illustrates an SIA setup designed as an environmental wastewater monitor for the speciation of different forms of nitrogen (ammonium, nitrites, nitrates, and total nitrogen) and phosphorous (orthophosphates and total... 

Simultaneous multicomponent determinations, 361 Sine waves, 33, 34, 508,956 mathematical description, 135 superposition. 137 Single-beam computerized spectrophotometers, 358 grating instruments, 357 instruments, 240, 241, 351 for atomic spectroscopy, 240 efl ect of mismatched cells, 343 for molecular absorption measurements, 351 advantages, 358 Singlet state, 216, 400-402,411 excited, 216,400,401,405 Size distributions, measuring, 951. 955, 961... 

Oleksy-Frenzel J. and Jekel M. 1996. Online multicomponent determination of organic compounds in water following gel permeation chromatographic separation. Anal. Chim. Acta 319 165-175. 

Fast-scanning PDA detectors and powerful signalprocessing techniques, such as multiwavelength analysis and derivative spectrophotometry, greatly facilitate multicomponent determinations and eliminate or reduce interferences. However, the precision is usually degraded for mixture analysis requiring multivariate techniques when compared to analyses where interferences are not observed. 

The use of the Kubelka-Munk equation for quantitative analysis by diffuse reflectance spectroscopy is common for measurements in the visible, mid-IR and far-IR regions of the spectrum, but not in the near-IR region. As has been pointed out [187, 188], almost all near-IR diffuse reflectance spectra have been converted to log(l// ) R = reflectance of the sample relative to that of a non-absorbing sample). The use of log(l// ) instead of the K-M function provides a more linear relationship between reflectance and concentration. Olinger et al. [189] explain this behaviour by the effective penetration depth of the beam, which is very short, when absorption is strong. For many of the algorithms developed to achieve multicomponent determinations from the diffuse reflectance spectra of powdered samples, a linear dependence of band intensity on analyte concentration is not absolutely mandatory for an analytical result to be obtained. 

Felice, L.J., Chalermchaikit, T., Murphy, M.J., 1991. Multicomponent determination of 4-hydroxycoumarin anticoagulant rodenticides in blood serum by liquid chromatography with fluorescence detection. J. Anal. Toxicol. 15 (3), 126-129. 

Fig. 2 Scheme of electronic tongue multicomponent determination of the two compounds A and B from the readings of an array formed by potentiometric sensors... 

Fig. 3 Simple scheme of the voltammetric electronic tongue, processing a whole voltammogram and performing a multicomponent determination of three compounds sought. Erom [13], with permission from Elsevier...

The first system, devised in the University of Porto, Portugal, was a sensor array for the measurement of creatinine in urine [18]. It consisted of a creatinine iminohydrolase enzyme immobilized by entrapment using a chitosan membrane onto a nonactin ISE. Catalytic hydrolysis by the enzyme generated ammonium ion which was then directly detected. The system was completed with ISEs for ammonium, potassium, sodium and calcium which allowed to correct for any endogenous ammonium (by the first ISE) or for alkaline and alkaline-earth interference in the ammonium-based biosensor (the other three). Linear response ranges were between 0.1 and 10 mM, what permitted the resolution of the multicomponent determination by PLS method. Comparison with reference Jaffe method showed a satisfactory correlation, although the slope of obtained vs. reference values was rather low the obtained value was 0.87, whereas theoretical value should be 1.0. 

Fig. 22 Modelling ability of the optimized ANN for the multicomponent determination of phenolic compounds present in wine. Training filled circles), external test empty circles) and spiked wine samples empty diamonds) set adjustments of the expected concentration vs. obtained concentrations for top) catechol, middle) caffeic acid and bottom) ( )-catechin. Dashed lines correspond to theoretical diagonal line. Error bars correspond to five different retrainings with random reinitialization of weights for the final architecture. Polyphenolic structures attached to the graphs and fitted regression lines for the three subsets stated on top of each plot. Adapted from [63], with permission from the Royal Society...

---