Using computed tomography to predict the elastomechanical properties of wood

Sintič, Jan (author), Straže, Aleš (mentor), Huber, Johannes (comentor)

In this study, we investigated the predictive ability of density measured by X-ray computed tomography (CT) and gravimetry for determining the modulus of elasticity (MOE) in Norway spruce (Picea abies (L.) Karst.), European beech (Fagus sylvatica L.), and English oak (Quercus robur L.). We used oriented specimens with four cross-sectional sizes: 20 × 20, 30 × 30, 40 × 40, and 50 × 50 mm. We determined the modulus of elasticity in three anatomical directions of wood (L – longitudinal, R – radial, and T – tangential) using static compression testing (ESL, ESR, EST) and dynamically by measuring ultrasound velocity (EDL, EDR, EDT). The study showed that CT scanning provides a more reliable prediction of the dynamic modulus of elasticity. Using multidimensional analysis with partial least squares regression (PLS-R), we developed a robust predictive model for the dynamic stiffness of the tested wood species, with moderate reliability (R²Y = 0.53, RMSECV = 1.60). We found that mechanical stiffness is primarily determined by specific structural properties of wood, such as the number of pores per unit length, with the most important predictive variables being the maximum density within the pore (Max_LW) and the proportion of latewood, while ring width is a less reliable predictor. Part of the unexplained variance can be attributed to influential variables that were not measured, such as microfibril angle, degree of cellulose crystallinity, and the possible presence of juvenile or reaction wood. The study confirms that CT scanning is a useful tool for non-destructive structural characterisation of wood and for predicting its mechanical stiffness.

• Language: English
• Keywords: wood, computed tomography (CT), density, mechanical properties, modulus of elasticity
• Work type: Master's thesis/paper
• Typology: 2.09 - Master's Thesis
• Organization: BF - Biotechnical Faculty
• Publisher: [J. Sintič]
• Year: 2026
• PID: 20.500.12556/RUL-182945
• UDC: 630*812(043.2)
• COBISS.SI-ID: 279958275
• Publication date in RUL: 29.05.2026

Secondary language

• Language: Slovenian
• Title: Uporaba računalniške tomografije za napovedovanje elastomehanskih lastnosti lesa

Abstract

V tej raziskavi smo preučevali napovedno sposobnost gostote, izmerjene z rentgensko računalniško tomografijo (CT), in gravimetrično, za določanje modula elastičnosti (MOE) pri navadni smreki (Picea abies (L.) Karst.), evropski bukvi (Fagus sylvatica L.) in hrastu dobu (Quercus robur L.). Uporabili smo orientirane preizkušance, 4 velikosti prečnega prereza, tj. 20 × 20, 30 × 30, 40 × 40 in 50 × 50 mm. Modul elastičnosti smo določali v 3 anatomskih smereh lesa (L-vzdolžno, R-radialno in T-tangencialno) s statičnim tlačnim testiranjem in dinamično (ESL, ESR, EST), in dinamično, z določanjem hitrosti preleta ultrazvoka (EDL, EDR, EDT). Raziskava je pokazala, da s CT skeniranjem zanesljivo lahko napovemo dinamični modul elastičnosti. Z večrazsežno analizo z regresijo delnih najmanjših kvadratov (PLS-R) smo izdelali robusten napovedni model za togost testiranih lesnih vrst, s srednje dobro zanesljivostjo (R²Y = 0,53, RMSECV = 1,60). Ugotovili smo, da mehansko togost določajo predvsem specifične strukturne lastnosti lesa, kot so število branik na dolžinsko enoto, pri čemer sta najpomembnejši napovedni veličini najvišja gostota znotraj branike (Max_LW) in delež kasnega lesa, širina branik pa je manj zanesljiva napovedna spremenljivka. Del nepojasnjene variance lahko pripišemo vplivnim veličinam, ki jih nismo določali, kot so denimo mikrofibrilni kot, stopnja kristaliničnosti celuloze, ter morebitna prisotnost juvenilnega oz. reakcijskega lesa. Študija potrjuje, da je CT-skeniranje uporabno orodje za neporušno strukturno karakterizacijo lesa ter napovedovanje njegove mehanske togosti.

• Keywords: les, računalniška tomografija (CT), gostota lesa, mehanske lastnosti, modul elastičnosti