Abstract

Wood material

Maple (Acer spp.) and red oak (Quercus rubra Michx.f.) samples were tested in this study. The mean basic densities of maple and red oak based on ovendried weight and green volume were 599.8 and 696.9 kg/m³, respectively. Two series of end-matched specimens were prepared from air-dried defect-free boards of each species, one for radial tests and the other for tangential tests (hereafter referred to as “LR” and “LT” specimens, respectively). The dimensions of LR specimens were 125 by 12 by 2 mm (length [L] by radial [R] by tangential [T]), and those of LT specimens were 125 by 12 by 2 mm (L by T by R).

Moisture conditioning

Samples for the vibration tests at constant MC were divided into five groups, each consisting of five samples. Air-dried specimens were moisture conditioned in three conditioning rooms (20°C; RH levels of 30%, 65%, and 90% ± 2%) for 4 weeks. Saturated samples were prepared by soaking air-dried samples in water for 4 weeks at room temperature. Ovendried samples were dried using a constant-temperature drying chamber (24 h at 105°C). The EMCs of the samples are shown in Table 1.

Table 1. Conditions for equilibrium moisture content (EMC) of samples.

View Fullscreen

Five ovendried specimens of each species were subjected to multistep adsorption in three rooms with relative humidity of 30, 65, and 90 percent. During the moisture sorption processes, specimens were taken out of the conditioning rooms at different times (2, 5, 10, 16, or 26 h) to determine their MC (Table 2) and vibrational behavior. After completion of the adsorption experiments, samples were saturated over distilled water for 10 days. These samples were subjected to multistep desorption mode in three rooms with relative humidity of 90, 65, and 30 percent; they were then placed in a desiccator containing P₂O₅ (0% RH). The specimens were weighed and measured for the vibrational properties at different times (2, 5, 10, 16, or 26 h) of each desorption process (Table 2).

Table 2. Conditions for moisture content (MC) changes of samples.

View Fullscreen

Measurement of vibrational properties

The test samples that were moisture conditioned for 2, 5, 10, 16, or 26 hours were sealed with a polyethylene film to avoid changes in MC during the test. A cantilever beam vibration testing apparatus (Forest Products Laboratory, USA) was used for vibration tests under constant MC and during moisture changes. The specimen was gripped securely at one end and excited near the free end with a sharp flick from a shooting valve. The vibration was detected with a laser displacement transducer and recorded as a function of time. The frequency and amplitude attenuation were then used to calculate the DMOE and the δ of the cantilever samples.

DMOE can be calculated by the following equation (Turk et al. 2008):

where

•  M = mass of the specimen (kg),

•  L = complete length of the specimen (m),

•  b = base width of the specimen (m),

•  t = thickness of the specimen (m),

•  f = detected frequency of the first natural mode of vibration (Hz), and

•  l = unclamped length of the cantilever beam (m).

As far as the internal damping is concerned, the law governing a damped vibration has the following form:

where h is the damping constant. The amplitude is a function of time, and it decreases toward zero.

A typical specimen free vibration response is shown in Figure 1. In order to determine the δ, the following equation can be used:

View Figure

• Download as Powerpoint
• Download Figure

Vibrational properties of wood at constant moisture contents

Figure 2 shows the DMOE and δ values of LR and LT specimens as a function of constant MC for maple and red oak. Each dot represents the averaged values for five samples, and the 95 percent confidence interval for the plots was less than 0.03 GPa and 0.006 for DMOE and δ, respectively. The DMOE reached a maximum from 0 to 6 percent MC and then decreased with increasing MC. The δ value decreased as the MC changed from 0 to 6 percent MC and then rose with increasing MC.

View Figure

• Download as Powerpoint
• Download Figure

These results agree with those obtained by Kollmann and Krech (1960), James (1961, 1964), Suzuki (1962), Obataya et al. (1998), and Takahashi et al. (2006). In their research, the elastic modulus showed a maximum value, and δ showed a minimum value of about 5 to 8 percent MC. These phenomena can be explained by the hypothesis that in an ovendried state, molecular chains in the amorphous regions of the cell wall are unnaturally distorted. In addition, microvoids exist between the molecular chains. This unnatural structure of the cell wall in an ovendried state causes a smaller DMOE value and a greater δ value. With the increasing MC from the ovendried state, water molecules are embedded in the microvoids, and the distorted molecular chains are rearranged. As a result, at about 5 to 8 percent MC, a more stable state for the cell-wall structure is obtained. Above 8 percent MC, water acts as a plasticizer that allows more molecular movement, and the cohesive forces between molecules are decreased, resulting in decreasing DMOE values and increasing δ values.

Beyond these common trends, the differences in both DMOE and δ are apparent between two grain orientations for maple and red oak. The DMOE and δ of LR samples were lower and higher, respectively, than those of LT samples for the maple specimens. However, the opposite trends were found for red oak. The variation of vibrational behavior between two species is probably because of the difference in the arrangement of the cell wall and the lumen in addition to the anisotropy of cell-wall substances.

Changes in vibrational properties during moisture adsorption processes

Figure 3 shows changes in DMOE and δ values for the wood that was moisture conditioned by the adsorption process in both grain directions of maple and red oak specimens. The tests were carried out five times for each condition. Because the behavior of DMOE and δ with MC change was similar for all conditions, only one of the results is shown here. In each adsorption condition, regardless of species and grain direction, DMOE decreased in the early stage and then increased or flattened out in the higher MC ranges. This is attributed to the fact that the samples were more stabilized as MC changes finished in the later stages. The changing behavior of DMOE from the ovendried state to approximately 5 percent MC (0% → 30% RH) was compared with that of the other two adsorption processes (30% → 65% and 65% → 90% RH). A remarkable decrease in DMOE with increasing MC was apparently observed in the former case. The adsorption rate should affect the decreasing behavior of DMOE because a greater adsorption rate leads to greater destabilization during an adsorption process (Takahashi et al. 2005).

View Figure

• Download as Powerpoint
• Download Figure

The δ values of maple and red oak during the moisture adsorption processes for LR and LT samples are also presented in Figure 3. Regardless of species and grain direction, the δ value had a relative maximum at about 3 percent MC and then increased almost linearly with an increase in MC from around 5 percent MC. In the ovendried state, the hydroxyl groups of the cellulose cell wall closed together, allowing the formation of weak cellulose-to-cellulose bonds. When adsorption of water then occurred in lower MC (less than 5% MC), a part of these weak bonds was broken down by entering water molecules, and some hydroxyl groups were released (Bowyer et al. 2003). Therefore, the reorientation of molecular chains may cause energy dissipation during the initial adsorption processes, resulting in an increasing δ value. In general, the δ value of wood increases with an increase of MC within the limits of the MC examined in this study.

With respect to the comparison of Figures 2 and 3, the DMOE and δ values during the adsorption processes were relatively lower and higher, respectively, than those of the samples measured at constant MC. This is because wood in a nonequilibrium moisture state shows a lower strength and higher damping properties than wood in the equilibrium state (Takahashi et al. 2004, 2005).

Changes in vibrational properties during moisture desorption processes

Representative results of DMOE and δ values of LT and LR samples for maple and red oak during the moisture desorption processes are shown in Figure 4. Regardless of species and grain direction, DMOE increases with decreasing MC without showing a maximum value. The DMOE during the desorption processes was lower than that in a stable state (Fig. 2) at a similar corresponding MC.

View Figure

• Download as Powerpoint
• Download Figure

In Figure 4, δ decreases with a decrease in MC during the desorption processes. Furthermore, wood conditioned by desorption showed lower δ values than wood conditioned by adsorption (Fig. 3) regardless of the conditioning period. Considering the characteristics of the unstable state of wood, this phenomenon can be attributed to the fact that wood that is moisture conditioned by adsorption is more unstable than wood that is conditioned by desorption. This interpretation is reasonable and can be explained as follows. Trees live in a state swollen by water. and this state is the most stable state for wood. During the drying process, a highly unstable state is induced. Such an unstable state is not easily modified by the adsorption of moisture. Accordingly, wood moisture that is conditioned by the adsorption process remains in a more unstable state than wood that is conditioned by the desorption process to the same MC (Ishimaru et al. 2001a, 2001b).