A. Rygula and P. Miskowiec
300
group under thermal perturbation from 50 to 120 °C. generally, they concluded that
the 2d correlation IR spectra not only enlarges the differences in the Ft-IR spectra,
but also provides useful information for studying the thermal stability of chemical
components. the relative contents of flavonoids in the different extracts could be
deduced according to the IR characteristic peaks base on the shaping and intensity
of peaks near 1,620 cm
−1
.
mid-IR turned out also promising techniques in some other dyes analysis. Among
others color and antioxidant compounds like phenols, flavons and anthocyanins of
different berries were studied [100–102]. the obtained spectral datasets were correlated with technological parameters of ripening or fruit antioxidant activity by
using partial least squares (PLS) regression algorithms both in full cross-validation
and test set validation. the models developed in the mid-IR region again proved to
be well correlated with analytically determinated phenolic compounds.
Similar to the above described, the results for anthocyanins are presented in
work of Sinelli et al. [100]. the authors aimed to examine the potential of NIR (as
well as mid-IR) spectroscopy to evaluate ripening indices and nutraceutical properties of the two cultivars of blueberries. the spectral datasets obtained were correlated with such parameters like: total soluble solids, total phenols, total flavonoids
and total anthocyanins contents by using PLS regression algorithms. In the NIR
region the wavenumbers with the highest influence in the PLS models for predicting the parameters above described were 8,859–8,678 cm
−1
, 7,251–7,154 cm
−1
,
6,063–5,770 cm
−1
, 5,345–5,160 cm
−1
, 4,700–4,500 cm
−1
assigned to the C-h and
o-h overtones and combination bands. As the result NIR methods were characterized by PLS models mainly in full cross-validation by r values from 0.92 to 0.96 for
flavonoids and total anthocyanins and turned out very promising alternative quality control method. moreover, PCA was also successfully performed to evidence
separation according to harvest year and a partial separation according to varieties
of the blueberries.
NIR method was also applied to determine the total flavonoids content in ginkgo
( Ginkgo biloba L.) leafs [103]. In traditional Chinese medicine ginkgo leafs were
used as a means of treating the symptoms of asthma, cough, and skin infections.
today, extracts from this material, rich in flavonoids, are used in the treatment of
diseases of the cardiovascular system as well as supplements for memory and concentration. the total content of flavonoids, however, changes dramatically in fall,
which is the period of harvesting. that is why NIR method was used as rapid and
nondestructive method of determination of total flavonoids content in ginkgo leaf.
Apart from PLS, interval partial least squares (iPLS) and synergy interval partial
least squares (SiPLS) were used to develop calibration models.
the authors described the increasing amount of total flavonoids in the following order: green, green-yellowish and yellow leaves. two characteristic wavelength
regions (5,840–6,090 cm
−1
and 6,620–6,880 cm
−1
), selected by SiPLS were corresponded to the absorptions of two aromatic ring in basic flavonoid structure
(Fig. 10.14). the synergy interval partial least squares method of calibration turned
out the most optimal.
300
group under thermal perturbation from 50 to 120 °C. generally, they concluded that
the 2d correlation IR spectra not only enlarges the differences in the Ft-IR spectra,
but also provides useful information for studying the thermal stability of chemical
components. the relative contents of flavonoids in the different extracts could be
deduced according to the IR characteristic peaks base on the shaping and intensity
of peaks near 1,620 cm
−1
.
mid-IR turned out also promising techniques in some other dyes analysis. Among
others color and antioxidant compounds like phenols, flavons and anthocyanins of
different berries were studied [100–102]. the obtained spectral datasets were correlated with technological parameters of ripening or fruit antioxidant activity by
using partial least squares (PLS) regression algorithms both in full cross-validation
and test set validation. the models developed in the mid-IR region again proved to
be well correlated with analytically determinated phenolic compounds.
Similar to the above described, the results for anthocyanins are presented in
work of Sinelli et al. [100]. the authors aimed to examine the potential of NIR (as
well as mid-IR) spectroscopy to evaluate ripening indices and nutraceutical properties of the two cultivars of blueberries. the spectral datasets obtained were correlated with such parameters like: total soluble solids, total phenols, total flavonoids
and total anthocyanins contents by using PLS regression algorithms. In the NIR
region the wavenumbers with the highest influence in the PLS models for predicting the parameters above described were 8,859–8,678 cm
−1
, 7,251–7,154 cm
−1
,
6,063–5,770 cm
−1
, 5,345–5,160 cm
−1
, 4,700–4,500 cm
−1
assigned to the C-h and
o-h overtones and combination bands. As the result NIR methods were characterized by PLS models mainly in full cross-validation by r values from 0.92 to 0.96 for
flavonoids and total anthocyanins and turned out very promising alternative quality control method. moreover, PCA was also successfully performed to evidence
separation according to harvest year and a partial separation according to varieties
of the blueberries.
NIR method was also applied to determine the total flavonoids content in ginkgo
( Ginkgo biloba L.) leafs [103]. In traditional Chinese medicine ginkgo leafs were
used as a means of treating the symptoms of asthma, cough, and skin infections.
today, extracts from this material, rich in flavonoids, are used in the treatment of
diseases of the cardiovascular system as well as supplements for memory and concentration. the total content of flavonoids, however, changes dramatically in fall,
which is the period of harvesting. that is why NIR method was used as rapid and
nondestructive method of determination of total flavonoids content in ginkgo leaf.
Apart from PLS, interval partial least squares (iPLS) and synergy interval partial
least squares (SiPLS) were used to develop calibration models.
the authors described the increasing amount of total flavonoids in the following order: green, green-yellowish and yellow leaves. two characteristic wavelength
regions (5,840–6,090 cm
−1
and 6,620–6,880 cm
−1
), selected by SiPLS were corresponded to the absorptions of two aromatic ring in basic flavonoid structure
(Fig. 10.14). the synergy interval partial least squares method of calibration turned
out the most optimal.