A. Rygula and P. Miskowiec
286
the growing interest in using of IR and Raman spectroscopies instead of commonly applied chromatographic and uv-vis methods to quantitative and qualitative
analysis of dyes, has been dictated by a number of factors. the most important of
them are rapidity, simplicity, safety to the environment (extremely low chemicals
consumption) and low operational costs [42–45]. In most cases infrared methods
can be also applied in the non-destructive way [45, 46]. the mid-IR spectroscopy
has been successfully used to differentiate the color ink and paint entries [44–47].
however, application of mid-IR for dyes analysis in forensic science as well as in
cultural heritage studies is beyond the scope of this chapter.
the assistance of chemometrics to mid-IR spectroscopy is often related with
the evaluation of the results against hPLC and even uv-vis data. the latter are
typically used in dyes determination and have been developed as reference methods. mid-IR spectroscopy is also used in the food quality control. the results are
in most cases comparable to those obtained with traditional uv-vis and hPLC
methods [48–50].
10.2.2 Chemometric Methods Used for Quality, Quantity
and Distribution Analysis
According to [51], two chemometric approaches can be named: uni- (or bivariate)
and multivariate analysis. uni- or bivariate analysis uses one or two variables to
extract information from a system e.g. peak area or peak height. this approach can
be difficult to use when many peaks overlap but is the easiest and the most intuitive.
It is the most popular approach to analyse mapping results.
multivariate analysis uses many (up to thousands) of variables to obtain quantitative data. very small, even non-distinguishable by human eye spectral changes
can give significant meaning and provide important results. the main advantage
is the possibility to analyse the overlapping peaks and simultaneous detection of
several components in a sample.
Figure 10.7 shows the classification of the major methods used in analysis of
spectroscopic data [52]. Presented in Fig. 10.7 graph was dedicated to NIR spectroscopy, but the same rules could be applied to other spectroscopic techniques.
In recent years, NIR spectroscopy combined with chemometrics has attracted
considerable attention in chemical content analysis. the main advantage of NIR
comparing to mid-IR spectroscopy is that using NIR spectroscopy one can typically
penetrate sample much deeper. NIR spectroscopy can be therefore, very useful in
probing bulk material with very little or even no sample preparation.
most of absorption bands in the NIR range are overtones or combination of the
fundamental vibrations of the molecules. the typical NIR reflectance spectra and
its first derivative are shown in Fig. 10.8 [53]. they were obtained for banana pulps
with high content of carotenoids, prepared from different genome groups. one can
observe that in the complex mixtures, such as biological material, food or medicines, the multiple bands and the effect of peak-broadening result in broad signal
286
the growing interest in using of IR and Raman spectroscopies instead of commonly applied chromatographic and uv-vis methods to quantitative and qualitative
analysis of dyes, has been dictated by a number of factors. the most important of
them are rapidity, simplicity, safety to the environment (extremely low chemicals
consumption) and low operational costs [42–45]. In most cases infrared methods
can be also applied in the non-destructive way [45, 46]. the mid-IR spectroscopy
has been successfully used to differentiate the color ink and paint entries [44–47].
however, application of mid-IR for dyes analysis in forensic science as well as in
cultural heritage studies is beyond the scope of this chapter.
the assistance of chemometrics to mid-IR spectroscopy is often related with
the evaluation of the results against hPLC and even uv-vis data. the latter are
typically used in dyes determination and have been developed as reference methods. mid-IR spectroscopy is also used in the food quality control. the results are
in most cases comparable to those obtained with traditional uv-vis and hPLC
methods [48–50].
10.2.2 Chemometric Methods Used for Quality, Quantity
and Distribution Analysis
According to [51], two chemometric approaches can be named: uni- (or bivariate)
and multivariate analysis. uni- or bivariate analysis uses one or two variables to
extract information from a system e.g. peak area or peak height. this approach can
be difficult to use when many peaks overlap but is the easiest and the most intuitive.
It is the most popular approach to analyse mapping results.
multivariate analysis uses many (up to thousands) of variables to obtain quantitative data. very small, even non-distinguishable by human eye spectral changes
can give significant meaning and provide important results. the main advantage
is the possibility to analyse the overlapping peaks and simultaneous detection of
several components in a sample.
Figure 10.7 shows the classification of the major methods used in analysis of
spectroscopic data [52]. Presented in Fig. 10.7 graph was dedicated to NIR spectroscopy, but the same rules could be applied to other spectroscopic techniques.
In recent years, NIR spectroscopy combined with chemometrics has attracted
considerable attention in chemical content analysis. the main advantage of NIR
comparing to mid-IR spectroscopy is that using NIR spectroscopy one can typically
penetrate sample much deeper. NIR spectroscopy can be therefore, very useful in
probing bulk material with very little or even no sample preparation.
most of absorption bands in the NIR range are overtones or combination of the
fundamental vibrations of the molecules. the typical NIR reflectance spectra and
its first derivative are shown in Fig. 10.8 [53]. they were obtained for banana pulps
with high content of carotenoids, prepared from different genome groups. one can
observe that in the complex mixtures, such as biological material, food or medicines, the multiple bands and the effect of peak-broadening result in broad signal