In the scope of recycling timber, we are faced with a number of challenges, primarily from a quality assurance perspective. That is, assuring that only material of appropriate quality is re-used. The quality of timber that is seen in recycling centres varies wildly, and certain types of wood simply can’t be re-used due to environmental and legislative limitations, or simply because of poor quality due biological aspects (fungal decay, insect attack, presence of mould). The former definitely includes timber treated with certain types of preservatives, as their use in Europe has been prohibited yet they are still in use and will turn up in recycling centres for the foreseeable future. Thus, having a fast method to identify timber contaminated with preservatives would ensure better quality control in a continuous sorting process as well as prevent contaminated materials ending up in products utilizing recycled timber, or in timber used as an energy source.
Here we present the results of preliminary research focused on using hyperspectral imaging in order to identify and distinguish 3 common types of wood preservatives, i.e. copper-ethanolamine (CuEA), copper chromium arsenic (CCA), and copper chromium boron (CCB) as they are typically encountered in a greenish-grey tint and difficult to tell apart. Laboratory samples were prepared and analysed with a hyperspectral camera in the range of 1000 to 2500 nm, outside the visible range of light. In addition to laboratory specimens, commercial samples with a known composition were obtained, to see if identification of the preservatives is possible on non-ideal samples.
In the laboratory specimens the differences between the samples are clearly visible. The obtained spectra show differences between 2000 and 2300 nm, further analysis shows the presence of 3 distinct materials, but there is significant overlap between CCA and CCB samples, making distinction more difficult.
Overall, hyperspectral imaging is a powerful technique for the qualitative assessment of materials, and it is already a common technique in the food industry. The potential in the timber industry certainly exists. As was highlighted in this preliminary work, the machine is capable of distinguishing between 3 common wood preservatives. However, numerous challenges in this process remain. Firstly, a comprehensive library of such materials is needed. Due to the inhomogeneity of the materials and plethora of different tree species that are used, the spectra obtained as part of this preliminary research, on Scots pine, can not be used on other tree species treated with the same preservatives. A large databank ensures a higher probability that the machine is correctly identifying the material. Secondly, most materials going through a recycling centre will undoubtedly not be in as good a shape as the laboratory specimens, they will be dirty, the wood will have undergone some aging. This is a major challenge that will have to be overcome.
Keywords: Hyperspectral imaging, Wood preservatives, Recycling
Authors
Jaka Levanič
University of Ljubljana, Biotechnical faculty, Department of wood science and technology
Eli Keržič
University of Ljubljana, Biotechnical faculty, Department of wood science and technology
Boštjan Lesar
University of Ljubljana, Biotechnical faculty, Department of wood science and technology
Miha Humar
University of Ljubljana, Biotechnical faculty, Department of wood science and technology
Login to download the PDF