This research study investigates the effect of micronized copper azole type C (MCA-C) preservative system on the rolling shear (RS) properties of CLT. In the first part of research, bonding performance of CLT panels treated at two retention levels, 0.96 kg/m3 and 2.5 kg/m3, were evaluated. Three structural adhesive systems, melamine formaldehyde (MF), resorcinol formaldehyde (RF) and one-component polyurethane (1C-PUR) were used to assemble visually graded No. 2 2×6 southern yellow pine (SYP) lumber while manufacturing CLT panels. For treated CLT panels, 1C-PUR provides better bonding performance test results. The RS properties of MCA-C treated CLT panels were studied in the second part of the research. The CLT panels were subjected to out-of-plane loading according to the EN 16351 standard. The mean values of RS strength and modulus of treated CLT were 1.89 MPa and 289.4 MPa respectively.

Knots are usually regarded as defects when grading lumber. In order to evaluate a member under out-of-plane loading, shear strength is one of the major mechanical properties, specifically, rolling shear (RS) strength is one of the critical mechanical properties of Cross-Laminated Timber (CLT), which determines the flexural strength of CLT under short-span bending loads. Lower grade lumber with a higher percentage of knots is recommended to be utilized for the cross-layer laminations which are mainly responsible for resisting shear stresses. Firstly, shear tests were performed in order to evaluate the effect of knots on longitudinal shear strength using shear blocks. After that, the effect of knots on the RS strength of 3-ply southern yellow pine CLT were investigated by experimental tests and an analytical model. Center-point bending tests with a span-to-depth ratio of 6 and two-plate shear tests with a loading angle of 14° were conducted on six CLT configurations composed of different types of cross layer laminations: clear flatsawn lumber with/without pith, lumber with sound knots with/without pith, and lumber with decayed knots with/without pith. The shear analogy method was implemented to evaluate the RS strength values from the bending test results, which were also compared against the results from the two-plate shear tests. It was found that: (1) The shear blocks containing sound knots had higher shear strength than matched clear shear blocks, the shear blocks containing unsound knots had lower shear strength than the matched clear shear blocks. (2) CLT specimens with cross-layer laminations with either sound knots or decayed knots had higher RS strength. (3) In general, the shear analogy method underestimated the RS strength of CLT specimens containing knots and pith.

Cross-Laminated Timber (CLT) is an engineered wood product made by laminating dimensional or structural composite lumber in alternating orthogonal layers. Compared to Canada and Europe, CLT is a novel product to the US. With the additions included in the 2021 International Building Code (IBC), CLT material properties, especially rolling shear, would need to be explored. The increasing demand for softwood lumber, along with the increase of demand of CLT panel production, could place a burden and surpass the domestic softwood supply. Rolling shear is a phenomenon that occurs when the wood fibers in the cross-layers roll over each other because of the shearing forces acting upon a CLT panel when it is loaded out-of-plane. This study used the two-plate shear test from ASTM D2718 to measure the rolling shear properties of various southeastern US wood species: southern pine, yellow-poplar, and soft maple. A secondary study was conducted, using the same two-plate shear test, to measure the rolling shear properties of re-manufactured southern pine for CLT cross-layer application. The soft maple had the greatest average rolling shear strength at 5.93 N/mm2 and southern pine had the lowest average rolling shear strength at 2.51 N/mm2. Using a single factor analysis of variance (ANOVA), the rolling shear strength values from soft maple were significantly greater than yellow-poplar, which was significantly greater than the southern pine. For the rolling shear modulus, the southern pine and soft maple were of equal statistically significant difference, and both were greater statistically significant different compared to the yellow-poplar. The most common failure found from testing was rolling shear. / Master of Science / Cross-Laminated Timber (CLT) is an engineered wood panel product, similar to plywood, constructed with solid-sawn or structural composite lumber in alternating perpendicular layers. The additions included in the incoming 2021 International Building Code (IBC) has placed an importance in expanding the research related to the mechanical and material properties of CLT. Also, with the increasing demand for softwood lumber and CLT panel production, the demand for the domestic softwood lumber could place a burden and surpass the domestic softwood supply. Rolling shear is a failure type that occurs when the wood fibers in the cross-layers roll over each other because of the shearing forces acting upon a CLT panel. This study used the two-plate shear test to measure the rolling shear properties of various southeastern US wood species: southern pine, yellow-poplar, and soft maple. A secondary study was conducted, using the same two-plate shear test, to measure the rolling shear properties of re-manufactured southern pine for CLT cross-layer application. The soft maple had the greatest average rolling shear strength at 5.93 N/mm2 and southern pine had the lowest average rolling shear strength at 2.51 N/mm2. Using a single factor analysis of variance (ANOVA), the rolling shear strength values from soft maple were significantly greater than yellow-poplar, which was significantly greater than the southern pine. For the rolling shear modulus, the southern pine and soft maple were of equal statistically significant difference, and both were greater statistically significant different compared to the yellow-poplar. The most common failure found from testing was rolling shear.

Ce mémoire de thèse aborde le problème du flambement de murs en bois lamellé-croisé. Ces panneaux de bois, constitués de planches collées perpendiculairement, sont de plus en plus utilisés dans la construction. La tendance actuelle du marché est de concevoir des immeubles de grande hauteur, ce qui soulève la question de la résistance en compression de ces murs. Il s'avère que le bois est fortement anisotrope. En particulier, la raideur et la résistance en cisaillement perpendiculaire aux fibres, également appelé cisaillement roulant, sont beaucoup plus faibles que dans la direction parallèle aux fibres. Ce fort contraste nécessite un critère de conception plus élaboré que les outils classiques utilisés dans l'ingénierie du bois. Ce travail est organisé en deux parties. Dans la première partie, la raideur équivalente de cisaillement transverse d'un panneau de bois lamellé-croisé est étudiée. Des bornes sont établies par une approche théorique. Ces bornes sont validées par un nouveau dispositif expérimental qui permet la mesure de la raideur en cisaillement roulant avec une variabilité plus faible que le test textit{single-lap} classiquement utilisé. Dans la deuxième partie, ces données sont utilisées dans l'analyse du flambement de panneaux en bois lamellé-croisé en raffinant progressivement le problème. Dans un premier temps, la charge critique de flambement linéaire d'une plaque épaisse sans imperfections est établie. Cette charge critique est basée sur une nouvelle théorie de plaque d'ordre supérieur et montre que la charge critique de flambement basée sur une théorie de plaque mince (Kirchhoff-Love) ne peut pas estimer correctement la résistance de murs en bois lamellé-croisé. Dans un second temps, l'influence des imperfections est étudiée en adaptant l'approche classique de Ayrton et Perry à une poutre de Timoshenko. Cette extension a révélé qu'un nouveau critère de résistance doit être satisfait lors du flambement qui est spécifique aux murs en bois lamellé-croisé. Dans un dernier temps, cette analyse est étendue aux charges permanentes en supposant une loi de fluage simple qui conduit à un nouveau critère de conception simple qui pourrait facilement être adopté dans les codes de conception actuels / This PhD thesis addresses the issue of CLT wall buckling. These wooden panels, made of boards which are glued cross-wise, are more and more used in construction. The current trend of the market is to design high-rise buildings which raises the issue of the compressive strength of such walls. It turns out that wood is a highly anisotropic material. Especially, the shear stiffness and strength perpendicular to the grain (rolling shear) are much weaker than in the direction parallel to the grain. This high contrast requires more elaborate design criteria than classical tools used in timber engineering. This work is organized in two main parts. First, the equivalent rolling-shear behavior of a CLT layer is investigated. Bounds are established for the stiffness of an equivalent layer using a theoretical approach. These bounds are validated by means of a new experimental set-up which allows the measurement of the rolling shear stiffness with less variability than the classical single lap shear test. In the second part, this data is used in the buckling analysis of CLT walls with increasing refinements. First, the linear buckling load of a thick plate without imperfection is established. This load is based on a new higher-order plate theory and reveals that the critical load based on a thin plate theory (Kirchhoff-Love) cannot predict correctly the strength of CLT walls. Then, the influence of imperfections is introduced adapting the classical approach from Ayrton and Perry to the case of a Timoshenko beam. This extension reveals that a new design criterion has to be satisfied under buckling which is specific to CLT. Finally, this analysis is extended to long term loads assuming a simple creep law and leading to a new simple design criterion which may be easily introduced in current design codes