Лако-красочные материалы — производство

1. Crack Initiation

Fracture of wood and bonded joints and materials begins at a geometric or material discontinuity where displacement of the adherends (due to external or internal stress) creates the greatest stress concentration and where either the adherend or the adhesive is the weakest. Examples of geometric discontinuities in adhesive-bonded wood joints are the square-cut ends of overlapped adherends, voids at the tips of fingers in finger joints, voids in reconstituted boards, voids in the adhesive layer, and even the square-cut ends of individual fibers. Examples of material discontinuities are the juncture of adherends of different density, the interface between adhesive and adherends of differing moduli, earlywood and latewood bands of widely different density, and the transition zone between the low fibril angle S1 and high fibril angle S2 layers of the cell wall. When adhesive bonds near this zone are sheared, the microfibrils in the S1 layer appear to undergo a rolling-shear failure [43]. Adhesive penetration of the cell wall was shown to affect fracture positively in the vicinity of the S2-S3 interphase [44]. An epoxy adhesive applied soon after mixing was of sufficiently low molecular weight to penetrate the cell wall from the lumen. Subsequently, when the adhesive layer was stripped from the wood surface, fracture occurred in the S2 layer. The same adhesive applied some hours after mixing was higher in viscosity (and thus molecular weight) and did not penetrate the cell wall as deeply. In this case, fracture occurred in the S3 layer and S2-S3 interphase.

The idea of an intrinsic or inherent flaw size in wood was explored by Schniewind and Lyon [32] who found the intrinsic flaw to be 3 mm. The same idea was applied to wood-based panels by Ilcewicz and Wilson [45] and to solid-wood joints by Kyokong and others [28]. Ilcewicz and Wilson used a modified fracture model based on Eringen’s non­local theory [46] to determine the fracture toughness of flakeboard in tension perpendi­cular to the panel. According to their model, the critical stress intensity factor of the flakeboard is a function of the intrinsic flaw size (which they determined to be 8.6 mm), the intrinsic strength of the board (determined to be 4.5 MPa), and the “characteristic dimension.’’ The characteristic dimension in the original model for the fracture behavior of metal is the atomic distance of the metal. Ilcewicz and Wilson [45] substituted the flake thickness for the atomic distance in their modified model for flakeboard. They found the critical stress intensity factor (KIc) of the flakeboard was indeed a function of the char­acteristic dimension as well as the resin content of the board. Furthermore, the effect of flake thickness decreased as the resin content in the board increased from 5% to 11%. Based on this relationship, the authors predicted that KIc would become independent of resin content at about 17% and at this point the dependency of KIc would shift from the flake thickness to some anatomical substructure, independent of resin content, such as the average lumen diameter of the cells in the flakes. Similar relationships of fracture tough­ness to board density, resin content, and particle size were reported by Niemz and Schadlich [47]. It seems clear that the geometric discontinuities in reconstituted materials can be minimized by using lower-modulus, more conformable woods such as aspen rather than oak, thinner flakes or strands, higher compaction ratios, and higher resin content.

Research by Kyokong and others [28] lent credibility to Ilcewicz and Wilson’s hypothesis. They applied Eringen’s nonlocal theory to solid poplar (Populus tremuloides) joints bonded with resorcinol adhesive, substituting the average vessel lumen diameter of aspen (100 mm) as the characteristic dimension. They were able to show that the nonlocal theory using this dimension correlated very closely with the fracture toughness of the joints as determined by classic (local) theory.

In solid wood members, considerable effort is devoted to minimizing geometric discontinuities through the use of scarf and finger joints instead of butt and lap joints. Scarf joints of sufficiently low slope can achieve 85 to 90% of the strength of solid wood [48]. Scarf joints effectively minimize material discontinuities between earlywood and late — wood as well as geometric discontinuity. However, uniform-density wood, such as white pine, is likely to produce more efficient scarf joints than wood with growth rings of widely varying density, such as Southern pine. However, 100% efficiency is unlikely to be obtained in any case because of mismatched wood and adhesive properties. Finger joints are less efficient because the flat portion of each finger tip represents a small butt joint and geometric as well as material discontinuity. These can be effectively minimized by cutting sharp rather than blunt tips [49]. Tool wear presents a practical limitation to tip sharpness in machined finger joints, especially in higher-density woods or woods with high-density latewood bands.

Impression finger joints take tip sharpness to the extreme and would seem to approach a well-made scarf joint in freedom from geometric discontinuity. Impression joints are formed by pressing a heated die with knife-edged serrated surfaces into the end — grain surfaces to be joined. This process eliminates damage caused by cutting and has the advantage of producing essentially a side-grain surface for gluing. But because of the maximum compressibility of the wood at the finger tip by the die, impression joints are limited to woods with density less than about 0.5 [50]. Even though many structural woods are lower in density, they possess latewood bands of much higher density. However, elimination of the geometric discontinuity by the impression process densifies the finger tips but not the valleys; this results in a material discontinuity and thus stress concentra­tion. Fracture typically occurs across the roots of the fingers as a result of these closely spaced stress concentrations.