Pulp yield is an important measure of pulpwood quality, which is used regularly by the pulp and paper industry for which the possibility of using rapid methods to predict pulp yield would be very useful for screening and quality control. This work addresses the prediction of Kraft pulp yield under standard identical conditions and targeted to a kappa number of 15, using near-infrared (NIR) partial least squares regression modelling. A total of 75 pulp samples of Acacia melanoxylon R. Br. (blackwood) with a pulp yield variation range of 47.0–58.2 % were used. Very good correlations between NIR spectra and pulp yield were obtained. Ten methods were used for PLS analysis (cross-validation with 62 samples), and an external validation was made with 13 samples. The 2ndDer pre-processed spectra coupling two wavenumber ranges from 9087 to 5440 and 4605 to 4243 cm allowed the best model with a standard error of prediction of 0.4 %, a r of 98.1 %, and the ratios of performance to deviation (RPD) of 4.8. According to AACC Method 39-00, the present model is sufficiently accurate to be used in screening programs and in quality control (RPD = 6.9).
For further details log on website :