Estimativa do Ângulo Microfibrilar em Madeira de Eucalyptusurophyllax E. grandispor Meio da Espectroscopia no Infravermelho Próximo

Abstract

RESUMOA espectroscopia no infravermelho próximo (NIR) é uma técnica para caracterização de materiais. Os espectros no NIR são utilizados para estabelecer modelos preditivos que permitem a caracterização da madeira em grande escala e o relativo baixo custo. O presente estudo teve por objetivo ajustar calibrações baseadas nos espectros e nas estimativas do ângulo das microfibrilas (AMF) de celulose na parede secundária obtidas por difração de raios X. Foi possível desenvolver regressões para predição desta característica da parede celular por meio da espectroscopia no NIR. O coeficiente de determinação da regressão linear entre os valores obtidos por difração de raios X e os preditos pelo modelo NIR para angulo microfibrilar foi de 0,75 e as estimativas apresentam um erro de predição de ±1,31 graus. A partir dessa calibração, foi possível estimar essa característica da fibra da madeira de Eucalyptus em grandes amostragens com base unicamente na informação espectral, que pode ser facilmente e rapidamente obtida.Palavras-chave: ultraestrutura da parede celular, difração de raios X, análise multivariada. Estimation of Microfibrillar Angle in Eucalyptus urophylla × E. grandis Wood by Near Infrared Spectroscopy ABSTRACTNear Infrared (NIR) spectroscopy is a promising tool used for materials characterization. The information of the NIR spectra is used to establish predictive models that allow for wood characterization on a large scale and at low cost. The purpose of this research was to adjust calibrations based on NIR spectra and microfibril angle (MFA) of cellulose on the secondary wall estimated by X-ray diffraction. It was possible to develop regressions to predict this characteristic of the cell wall by means of the NIR spectra measured in the wood. The determination coefficient of the linear regression between the values obtained by X-ray diffraction and those predicted by the NIR model for microfibrillar angle was 0.75 and the estimates presented a prediction error of ±1.31 degrees. From this calibration, it was possible to estimate the fiber characteristics of the Eucalyptus in large samples based solely on spectral information, which can be obtained easily and quickly.

NIR-based models for estimating selected physical and chemical wood properties from fast-growing plantations

“…Concerning the chemical composition of wood, NIR spectroscopy detects variations in the chemical constitution and can be used to assess key properties (Hein et al 2012) and estimate the performance of wood in industrial operations, such as pulping and bleaching. Predictive models for estimating the extractive content with good performance for material selection and tree breeding programmes have been reported (Estopa et al 2017).…”

Determination of Heat-Treated Eucalyptus and Pinus Wood Properties Using Nir Spectroscopy

“…In addition, in Brazil, it is common to find sawmills working with softwood and hardwood. A higher coefficient of determination for models will be obtained when there are greater variety in values of parameters (Hein et al 2012). Heat treatment reduced EMC and radial and tangential swellings as much as 49, 57 and 39% in Eucalyptus grandis heat-treated wood at 220 °C respectively (Calonego et al 2012).…”