For selection of plant material for tree improvement, both growth and material properties have to be considered. Poplar wood is a good raw material for pulp and paper. Its fibres have suitable dimensions to form a smooth paper surface with good optical properties, but it also contains vessels, which may create problems at printing.
Research in wood and fibres has been limited by lack of efficient measurement methods. STFI has developed several methods useful for evaluation of the industrial potential of softwoods. One task within of the project has been to establish a set of characterization methods, useful in tree improvement programs for evaluation of properties of young hardwood materials. This includes investigation of age-to-age relationships, based on measurements of property differences between individual growth rings. Measurements of length and width of fibres and vessel elements in poplar have been emphasized.
Prior to the project, STFI had developed a system, STFI FiberMaster, for characterization of fibers in pulp, primarily softwood fibers. The method had also been used for analysis of the radial development of the fiber length in stem cross-sections. Cross-sections were cut into radial sub-samples. The fibers were liberated through maceration and analysed. The same procedure has now been tested on poplar, with positive results.
To allow the study of sub-samples representing individual growth rings, a new procedure has been developed: Radial sample strips are cut from larger-size wood samples and the growth rings to be analyzed are marked out. Sample strips are placed in especially designed containers and a kraft cook is performed. After cooking, the strips are soft but intact. The marked out growth rings are removed and the material is agitated to liberate the fibers, vessel elements, etc.
With the FiberMaster, it was not possible to analyse the vessel elements in the pulp. A new method for this has been developed in parallel with and inspired by the POPWOOD project. Pulp is diluted and images are recorded from its different objects: fibers, vessel elements, fines, etc. Image analysis is applied to locate the individual objects, classify them by type and collect the objects of each type in separate files. The objects are then characterized with dedicated routines. When ¡§minority objects¡¨ are characterized, it may happen that too few objects for good statistics are analysed. Routines have been added to handle this and good measurement results are now obtained.
In many cases it is, however, interesting to measure on the objects in situ in the original wood matrix, to allow a better interpretation of the result. In the POPWOOD project, a new technique has been developed for this. The cross-sectional dimensions of individual fibres and vessel elements are analysed with microscopy over large areas in wood (mm x cm). A solid wood sample is sawn from the wood material. The sample is planed, sanded and prepared to offer an even surface with high contrast between the fibre wall and lumen. The sample is mounted on a motorized stage under a microscope equipped with a video camera. The set-up is operated by a computer control system. Many small, elementary images are taken and stitched together into larger images. Image analysis is applied to identify each object (fibres and vessels) and to determine features such as widths, perimeter, area, roundness. A functioning system has been built. There were difficulties with the image quality, which were to a large extent solved by using of microtome sliced samples instead of solid wood. There were also difficulties with long calculation times. Eventually, it was decided that the method for measurement on pulp would be used when large numbers of poplar samples would be analysed. However, many experiences valuable also in other measurement applications have been gained and will be reported.
The evaluation of age-to age-relationships of different plant materials (result 19795) has, thus, been performed with the new methods for production of growth ring pulp samples and for analysis of fibers and vessel elements in pulp, with good results.
Large interest has been shown from industry and researchers. The methods are already applied in other research projects and contract work on aspen and eucalyptus. In one project, the natural variability within and between trees and clones is investigated for better matching of wood raw materials and products. In other projects the objectives are to select suitable clones for plantation forests. Other possible applications are trouble-shooting in mills, if problems related to vessel elements or other raw material properties are suspected, process investigations and optimisation.
The new procedures are now in operation in the STFI Wood and Fiber Measurement and Simulation Laboratory. Extensive use is expected in own research, R&D co-operations and contract work with research groups and companies.