Prediction of important sulphite pulp properties from near infrared spectra: a feasibility study and comparison of methods. An Abstract from Journal of Near Infrared Spectroscopy

Heidi Cecilie Henriksen,^a,d Tormod Næs,^b Rune Rødbotten^c and Are Aastveit^d
^aBorregaard Industries Ltd, R&D Department, PO Box 162, 1701 Sarpsborg, Norway
^bMATFORSK, Osloveien 1, N-1430 Ås, Norway
^cBorregaard Industries Ltd, QC Department, PO Box 162, 1701 Sarpsborg, Norway
^dNorwegian University of Life Sciences, PO Box 5003, 1432 Ås, Norway

In the pulp and paper industry, there is a need to replace some of the time-consuming, expensive pulp analyses used today. These methods could be replaced by rapid, non-destructive analyses. Near infrared (NIR) spectroscopy has proven to be an important tool in that context and has been used for several purposes in the sulphate pulping industry. However, very little work has been published for the sulphite industry. In this work, calibration models were made to predict Kappa number and viscosity of unbleached pulp. Since these pulp analyses are impossible to perform during processing, the NIR spectra were measured on the cooking liquor instead and then used to create calibration models for pulp properties. In order to find models with good predictive ability, four different preprocessing methods of the NIR spectra were tested, along with four different calibration methods. The results from this work demonstrate that NIR on cooking liquor can be used to predict pulp properties and, based on the validation results, the recommended model is partial least squares on 1st derivative spectra.