Weather degradation of thin wood samples assessed with NIR hypersepctral imaging in transmission mode

Introduction
The use of wood outdoor is widespread in various applications from ancient times. Nowadays, untreated wooden surfaces are frequently utilized as cladding in modern buildings. The choice of wood is often related to wood/structure aesthetics and there is an increasing demand for a better understanding of the deterioration mechanisms of wood during the outdoor exposure. Wood subjected to weathering is degraded by various environmental agents such as solar radiation, cyclic wetting, atmospheric temperature and relative humidity changes, environmental pollutants and certain micro-organisms.
Near infrared spectroscopy and hyperspectral imaging are perfect scientific tools for rapid and nondestructive characterization of wood surfaces. The great advantage of both techniques is the possibility of determining both chemical and physical properties of a large number of samples.
The goal of this work was to assess the degradation of weathering and to model its kinetics on the base of thin wood samples exposed outdoors by means of NIR hyperspectral imaging. The work is part of the pan-European Round Robin experiment established within the COST Action FP1006.

Materials and methods
Experimental samples were prepared from one piece of Norway spruce wood (Picea abies) on a slicing planner (Marunaka) to a thickness of ~100μm, and a surface of 30mm x 35mm. Sets of 12 samples were exposed outdoors at 18 different locations with identical setup. One sample was collected from the set after 0, 1, 2, 4, 7, 9, 11, 14, 17, 21, 24 and 28 days of weathering. Hyperspectral images in NIR wavelengths (1000 – 2500 nm) were obtained from the samples with a linescan camera (Specim, SWIR). The hyperspectral image acquisition was carried out in transmission mode using a custom setup; backside illumination with halogen lamps below a semi opaque glass plate, and another transparent glass plate transmitting NIR radiation above the samples.
The images were analyzed to study the evolution of earlywood and latewood as a function of weather degradation doses by using PLS regression techniques and PLS-DA classification. MATLAB (Mathworks) and PLS Toolbox (Eigenvector) were used as software platforms.

Results and discussion
Spectra corresponding to the earlywood and latewood zones of samples at varying degradation stages were extracted from hyperspectral images. Changes in the spectra were assessed separately (for early- and late-woods) due to significant differences to the morphological, chemical and physical structures of these woody constituents.
The deterioration progress due to weathering was modeled by means of a custom algorithm based on PLS. It was found that the latewood spectra appear more similar to the earlywood spectra after several days of outdoor exposure. These differences are partly due to degradation processes that changes the thickness of the samples and partly due to changes in chemical composition occurring during exposure. The chemical modifications are interpreted in correlation to the other analytical techniques as explored within the Round Robin experiment. The whole set of available information was also considered when developing numerical models. Such models will be used to simulate the future performance wooden surfaces exposed to varying doses of natural weathering.