Morphological analysis of pulps from orange tree trimmings and its relation to mechanical properties

Abstract

To optimize the pulping and refining processes of new alternative raw materials for papermaking, researchers generally perform tests that consume considerable time and large amounts of sample. We propose measuring the morphological properties of pulps from orange tree trimmings by image analysis systems, which are fast and user-friendly, to develop models relating key mechanical properties to the dimensions, the deformation and the population of fibers. Data modeling involves multiple linear regression, as used in other studies, and support vector regression, not used before for this purpose, achieving higher R2 values (up to 0.90). Although tensile, tear and burst tests are still required to obtain accurate values, a quick morphological characterization allows for a rough but satisfactory prediction of paper strength. In this case, chemical pulping and moderate refining are shown to be necessary to obtain pulps of acceptable quality from orange tree trimmings.