Adhesives for Wood and Lignocellulosic Materials. R. N. Kumar

migrate to the surface and might reduce adhesion due to weak boundary layer (see Chapter 9).

      Bonding of wood is described as a chain of several links, comprising wood, wood surface and its boundary layer, interphase of wood and adhesive and interface between wood and adhesive, and the adhesive bond line itself.

One useful method for understanding the abovementioned links that control the adhesive strength as well as weakness is the chain link analogy proposed by Marra [5]. Different areas of the substrate and adhesive are likened to a series of chain links, with the weakest link being the site of fracture. This is depicted in Figure 2.7. These links are as follows:

Figure 2.7 Different links in adhesive bonding.

      Link 1, adhesive film; links 2 and 3, intra-adhesive boundary layers; links 4 and 5, adhesive–adherend interface (in this region, the weak boundary layer exist); links 6 and 7, adherend subsurface; links 8 and 9, adherend proper.

Another cause for weak link is due to the stresses caused at the bond line due to swelling and shrinkage due to moisture changes. If the adhesive bond has to be durable, it has to adapt itself to the dimensional changes and the consequent strain due to swelling and shrinkage accompanying the changes in moisture conditions. Two distinct classes of wood adhesives have different ways to distribute this strain: (a) rigid in situ polymerized adhesives relieves this stress in many cases by distributing the strain through the wood interphase region. (b) The other class, the more flexible pre-polymerized adhesives, can distribute the strain through the adhesive interphase. Failure to adequately perform either of these strain distribution processes can lead to high strains and subsequent failure zones. As wood dries, it shrinks back to near its original dimensions. These failure zones can expand and become more visible as delamination areas [36].

      Mechanically weak wood surfaces can be another source of weak boundary layer. The causes of this are many [5, 35]. One cause is physical crushing of the surface, especially by abrasive planing or by too high of a bonding pressure. This occurs when more pressure is applied than the thin-walled earlywood cells can withstand. The strength of these cells is reduced due to deformation and fracture of the cell walls. A second cause is sanding dust or other dust accumulation on surfaces. A third cause is tearing of the surface that occurs during planing and sanding. A fourth cause, associated with high-density wood species, is cells becoming separated from one another due to the force of planning [36].

2.16 Measurement of the Wetting Parameters for Wood Substrate

      Unlike other substrates, wood exhibits complex anatomical features that make it heterogeneous and porous. It is basically hygroscopic. It contains extractives.

As a result of these unique physicochemical characteristics, wetting measurements on wood is difficult. Direct measurement of contact angle of adhesives on wood surface may not be reproducible and hence not satisfactory. For example, a drop of water on a wood surface will, in most cases, quickly change its size and shape over time, which will thus lead to a change in the contact angle.

      New methods for determining the wetting parameters of wood are necessary. A detailed and critical review of the various methods to determine the wetting parameters have been reported by Magnus Wålinder. The readers may refer to the same [37]. A brief summary from the above review is given below.

      “One way to address some of the difficulties in wood wetting measurements may be to apply the Wilhelmy (1863) method [38]. In contrast to direct measurement of contact angles, as in the drop method, the Wilhelmy method involves determining the force acting on a specimen when it is immersed in and withdrawn from a liquid. An apparent contact angle can then be estimated from an analysis of the recorded force”.

      “Other promising techniques for estimating the surface energetics of wood may be the Axisymmetric Drop Shape Analysis-contact diameter (ADSA-CD) technique. Contact angle measurements determined as constant wetting rate angle values (cwra) and also a capillary rise technique (column wicking) applied to wood”.

      “Inverse gas chromatography (IGC) is a useful technique for determining surface energetics of particle surfaces. By using appropriate gas probes, IGC can provide information on the surface thermodynamic characteristics of particles including surface free energy, acid-base interactions, enthalpy, and entropy. IGC has been applied to many materials such as polymers, wood pulp and wood particles”.

      Some spectroscopic methods, namely, X-ray photoelectron spectroscopy (XPS) and FT Raman, have also been recommended by the author.

      Both IGC and wicking methods rely on wood powder, which will give different results compared to measurements on solid wood.

      2.16.1 Some Results on Surface Energy of Wood

As discussed in Chapter 1, wood is a hierarchical cellular material and is therefore anisotropic in nature. Further wood surfaces are topographically different in radial, tangential, and transverse sections [31]. The wood surface consists of earlywood and latewood. The water contact angle of earlywood is often different from that of latewood. At the microscale, the wood surface consists primarily of lumen surfaces and cross-sectional walls. Thus, wetting of wood by the adhesives is therefore complex. Freeman was the first to report on the wettability of wood [39].

      Gray carried out an extensive investigation on the wettability of 20 species of wood [7]. Gray was first to determine the surface free energy of wood. Sessile drop method was used to determine the contact angle, and the critical surface tension (γ_c) was obtained by the Zisman method. Species used by Gray were Western Hemlock (Tsuga heterophylla), Douglas-fir (Pseudotsuga taxifolia), Afrormosia (Afrormosia elata), Parana pine (Araucaria angustifolia), European redwood (Pinus sylvestris), English oak (Quercus robur), and Beech (Fagus sylvatica). The values of critical surface tensions ranged from 34.5 to 81.0 mJ/m². One of the important findings was that freshly sanded surfaces were approximately 20 mJ/m² higher in surface energy than un-sanded, aged surfaces. It was also shown that surface contamination occurs rapidly on freshly cleaned surfaces.

      Herczeg reported on the wettability of Douglas-fir wood [40]. The critical surface tensions, γ_c, were found to be between 44 and 50 dynes/cm for summerwood and springwood, respectively. The surface-free energy and the maximum work of adhesion were also reported. It was also reported that aging increased contact angle, showing that wood wettability was reduced. Chen reported that removal of extractives from some tropical woods improved wettability [41]. Hse measured the wettability of southern pine veneers by measuring the contact angles formed with 36 phenol-formaldehyde resins [42]. The contact angle of resins on earlywood was less than that on latewood, apparently because earlywood surfaces were rougher. Also, the contact angle was positively correlated with the glue bond quality as tested by wet shear strength, percent of wood failure, and percent of delamination. Nguyen and Johns found that the surface free energy of Douglas-fir and redwood decreased with aging time [43]. Extractives and aging had significant influence on the surface energy. The surface free energy of Douglas-fir was 48.0 mJ/m², and after extraction, it increased to 58.9 mJ/m². These results emphasize the influence of extractives on the wood surface energy.

      Kalnins et al. [44] employed a video-type technique to measure the dynamic contact angle of distilled water as test liquid on wood with measurements conducted at the elapsed time of 3 to 5 s.

      Gardner et al. found dynamic contact angle measurements to be useful for determining the effect of wood processing and environmental