16 Wooden Material and Environmental Sciences
363
Fig. 16.2 Number of publications due to “NIR” and “Wood,” searching by Web of Science
that of IR light; hence samples can be measured nondestructively. Another important
advantage of NIRS is that many properties could be evaluated simultaneously which
use the same NIR data. It is superior to sort wood, which generally is based on several
criteria together.
However, as overtones or combinations of fundamental vibrations are multiply
overlapped each other at the NIR range, measured spectra are “opaque spectroscopic
information.” They are generally analyzed with the aid of multivariate analysis, such
as principal component analysis (PCA), partial least squares regression (PLS-R),
to observe “useful material information.” Some other NIR spectra collection and
data analysis methods have also been developed since wood samples have much
higher light scattering than light absorption. The spectral contributions of them were
evaluated independently, for constructing robust calibration models of various wood
properties without relying on complex multivariate statistical analysis.
Investigating NIRS for the quality assessment of pulp and paper has been going on
for a number of years, Brikett and Gambino estimated pulp kappa number with NIRS
approximately 30 years ago [1]. After that, the publication related to wood science
and technology with NIRS dramatically increased. Figure 16.2 shows the number of
research publications, including the keywords “NIR” and “Wood,” searching by Web
of Science. Nowadays, NIRS has already been exploited successfully by many papermaking companies to evaluate moisture content (MC) online. Below we introduce
some representative studies separated into different wood properties.
16.2.1 Wood Chemical Composition
Wood is typically composed of three main chemical components: cellulose, hemicellulose, and lignin. Their percentages are roughly 45%, 25%, and 25%, respectively [2]. Wood also contains extractives that can be removed by solvents, which
363
Fig. 16.2 Number of publications due to “NIR” and “Wood,” searching by Web of Science
that of IR light; hence samples can be measured nondestructively. Another important
advantage of NIRS is that many properties could be evaluated simultaneously which
use the same NIR data. It is superior to sort wood, which generally is based on several
criteria together.
However, as overtones or combinations of fundamental vibrations are multiply
overlapped each other at the NIR range, measured spectra are “opaque spectroscopic
information.” They are generally analyzed with the aid of multivariate analysis, such
as principal component analysis (PCA), partial least squares regression (PLS-R),
to observe “useful material information.” Some other NIR spectra collection and
data analysis methods have also been developed since wood samples have much
higher light scattering than light absorption. The spectral contributions of them were
evaluated independently, for constructing robust calibration models of various wood
properties without relying on complex multivariate statistical analysis.
Investigating NIRS for the quality assessment of pulp and paper has been going on
for a number of years, Brikett and Gambino estimated pulp kappa number with NIRS
approximately 30 years ago [1]. After that, the publication related to wood science
and technology with NIRS dramatically increased. Figure 16.2 shows the number of
research publications, including the keywords “NIR” and “Wood,” searching by Web
of Science. Nowadays, NIRS has already been exploited successfully by many papermaking companies to evaluate moisture content (MC) online. Below we introduce
some representative studies separated into different wood properties.
16.2.1 Wood Chemical Composition
Wood is typically composed of three main chemical components: cellulose, hemicellulose, and lignin. Their percentages are roughly 45%, 25%, and 25%, respectively [2]. Wood also contains extractives that can be removed by solvents, which