Determination of the in-plane wetting of a board structure using NIR spectroscopy

Abstract

This thesis reports on an investigation of the mechanisms of edge wicking in retortable paperboard. Retortable board is used for packaging preserved food, a process which requires that the package and its contents be sterilised by exposure to high temperature steam for up to three hours. The board used must thus have higher water repellence than traditional liquid packaging. Water vapour that condenses on the cut edges on the outside of the board causes particular concern. The paperboards studied were made from refined and unrefined bleached softwood kraft pulp and from refined unbleached softwood kraft pulp in one or two plies with different sizing levels and structures. Samples of each of the boards were immersed in a water bath at 95 °C. Other samples from the same boards were exposed to a combination of saturated steam at 130 °C and condensed steam in a special apparatus developed for this purpose. The board samples were placed on a cooling plate to simulate the temperature gradient when a container of cold food is sterilised with heated steam. Still other samples were exposed to heated steam in the absence of condensation by hanging them freely in the apparatus. The extent of edge wicking in the boards was determined by gravimetric measurements and also by near-infrared (NIR) spectroscopy, a technique that yields more information about the moisture content profiles and the mechanisms of liquid sorption. The moisture content profiles of boards exposed to saturated steam in the absence of condensation show more uniform water uptake. This phenomenon can be explained by rapid vapour phase transport throughout the pore structure, followed by slower water uptake in the fibres. By contrast, in samples exposed to both heated steam and condensed steam, there was both liquid sorption in the fibre network and rapid vapour phase transport of the steam. The moisture content in those samples was much higher close to the edge and lower behind the liquid frontier. The edge wicking of high temperature water was greatest in board that was not fully sized, in low density board, and in board made from unrefined pulp. The greater edge wicking in board made from unbleached pulp can be accounted for in terms of its greater swelling potential. In the low density board and the board made from unrefined pulp, the lumens at the unsized edge and the weaker bonding strength are suggested to affect the results. Capillary sorption takes place in lumens and delaminated bonds at the edge and then any further propagation takes place entirely by fibre diffusion. Capillary sorption may also be important in damaged areas where broken fibre-fibre bonds are exposed to liquid. The weaker bonded area in low density board therefore also contributes to water sorption. Edge wicking in the boards placed on the cooling plate and exposed to a combination of saturated and condensed steam was of a different nature. For these samples, density was not an important factor. The difference is due to the different mechanism at work in this case, where the relative area of the edge exposed to condensed steam was larger in the high density boards than in the low density boards. In the high density boards, condensed steam was sorbed in the structure by diffusion in the fibre structure, whereas in low density boards, the dominant effect was vapour phase transport in the void spaces followed by water uptake in the fibres.