Abstract
The differences of hygroscopic property among six tree species with four thicknesses were examined. The density, chemical composition, crystallinity, equilibrium moisture content (EMC), and moisture absorption rate were measured by static saturated salt solution method, and the isothermal moisture absorption curve was fitted by the H-H model to analyze monolayer molecular adsorption and multilayer molecular adsorption. The results show that under the same relative humidity (RH), the EMC of Picea asperata and Populus deltoides were increased with increasing thickness, while that of Quercus spp. and Xanthostemon melanoxylon were decreased. The moisture absorption rate of P. asperata was the largest and that of X. melanoxylon was the smallest. When RH ranged from 0 to 97%, the monolayer molecular adsorption water amount of samples with different thicknesses of the six tree species gradually became close. With the increase of tree species density, the monolayer molecular adsorption water amount of the thinner sample is gradually greater than that of the thicker sample. The change of multi-layer molecular adsorption water content is consistent with that of EMC, Moisture absorption rate, monolayer molecular adsorption water, and multilayer molecular adsorption water are related to the chemical composition content, density, and thickness of tree species.
Download PDF
Full Article
Hygroscopic Properties of Six Tree Species with Four Tangential Thicknesses
Chang Liu,a Shuai Cheng,a Shuai Cao,a Pengyu Wang,a and Jiabin Cai a,b,*
The differences of hygroscopic property among six tree species with four thicknesses were examined. The density, chemical composition, crystallinity, equilibrium moisture content (EMC), and moisture absorption rate were measured by static saturated salt solution method, and the isothermal moisture absorption curve was fitted by the H-H model to analyze monolayer molecular adsorption and multilayer molecular adsorption. The results show that under the same relative humidity (RH), the EMC of Picea asperata and Populus deltoides were increased with increasing thickness, while that of Quercus spp. and Xanthostemon melanoxylon were decreased. The moisture absorption rate of P. asperata was the largest and that of X. melanoxylon was the smallest. When RH ranged from 0 to 97%, the monolayer molecular adsorption water amount of samples with different thicknesses of the six tree species gradually became close. With the increase of tree species density, the monolayer molecular adsorption water amount of the thinner sample is gradually greater than that of the thicker sample. The change of multi-layer molecular adsorption water content is consistent with that of EMC, Moisture absorption rate, monolayer molecular adsorption water, and multilayer molecular adsorption water are related to the chemical composition content, density, and thickness of tree species.
DOI: 10.15376/biores.17.2.2959-2976
Keywords: Tree species; Wood thickness; Hygroscopic property; Hailwood Horrobin Model
Contact information: a: Nanjing Forestry University, Nanjing 210037 China; b: Co-Innovation Center of Efficient Processing and Utilization of Forest Resources, Nanjing 210037 China;
* Corresponding author: nldfloor@163.com
INTRODUCTION
Wood is a renewable material that is widely used in construction engineering, furniture manufacturing, and interior decoration. Due to its hygroscopic properties, it often leads to quality problems such as cracking and deformation during the process of production, processing, or use. Due to the different thickness of tree species and wood used in furniture products and interior decoration materials, the quality problems are also different. For instance, thick parts in solid wood furniture rarely have problems, while thin parts with the same tree species and moisture content are prone to cracking, joint pulling, and other problems. The problems caused by the hygroscopicity of wood of different tree species and different thicknesses have been studied (Neimsuwan et al. 2008; Fernandez et al. 2014; Podlena et al. 2017; Murr and Lackner 2018; Kalita et al. 2019). There has been research on the effects of different tree species and samples of different specifications (Skaar 1988; Ma et al. 2010; Fernandez et al. 2014; Yang et al. 2018; Majka et al. 2019; Mvondo et al. 2021). In addition, effects of wood chemical composition (Rowell 2005) on wood moisture absorption, and the correlation between wood moisture absorption and wood species, wood thickness, and wood chemical composition content have been investigated. Although the hygroscopicity of a variety of tree species and wood thickness has been studied, there are few systematic comparative studies on the hygroscopic differences between coniferous wood and broad-leaved wood, and the hygroscopic differences between the same tree species and wood thickness.
Wood is a renewable material widely used in construction, furniture manufacturing, and interior decoration. Due to its hygroscopic properties, when the ambient temperature and humidity change, it will absorb or release moisture, leading to swelling or shrinkage, resulting in quality problems such as cracking and deformation during production, processing, or use. Due to the different tree species and wood thicknesses used in furniture products and upholstery materials, there are also differences in the quality problems that arise. For example, thick parts in solid wood furniture rarely have problems, while thin plate parts of the same tree species and moisture content are prone to cracking, pulling, and other problems. Therefore, the problems caused by the hygroscopicity of wood of different tree species and different thicknesses have attracted the attention of some scholars (Neimsuwan et al. 2008; Podlena et al. 2017; Murr and Lackner 2018).
Scholars have carried out a large number of experiments to test the effects of different tree species, samples of different specifications (Mvondo et al. 1988; Ma et al. 2010; Yang et al. 2018; Majka et al. 2019), and wood chemical composition (Rowell 2005) on the hygroscopicity of wood. The difference in wood moisture absorption of different tree species is related to the content of internal chemical components and crystallinity (Huang and Wang 2014). Fernandez et al. (2014) used the saturated salt solution method to study the difference in adsorption of the two tree species at different temperatures. They analyzed the reasons for the difference in combination with softwood, hardwood, functional groups, and crystallinity. It was found that the hemicellulose content of the hardwood was higher than that of the softwood, and the crystallinity was lower, the adsorption sites were more, and the hygroscopic equilibrium moisture content that could be achieved was higher.
Murr and Lackner (2018) studied the adsorption kinetics of wood with different particle sizes and thicknesses. They found that the smaller the particle size and the smaller the thickness of the sample, the faster the mass increase of the sample, and the faster the time for the sample with smaller thickness to reach equilibrium for thicker samples. Yang and Ma (2013) studied the change of water content of different tree species and samples of different sizes under the condition of cyclic change of relative humidity and found that the samples with larger thickness achieved lower water content when using the same time and temperature and humidity conditions. The magnitude of radial and chordwise dimension changes is small, and the chordwise change is always greater than the radial dimension. This is because wood is an anisotropic material with different shrinkage and expansion in different directions.
The largest dimensional change occurs in the direction tangent to the growth rings, and the shrinkage is smaller along the wood ray direction. Shrinkage is minimal in the longitudinal direction (Kalita et al. 2019). Many scholars have studied the relationship between the hygroscopicity of wood and wood species, wood thickness, and wood chemical composition content. However, there are few systematic comparative studies on the hygroscopic difference between softwood and hardwood, and the difference in hygroscopicity between woods of the same species but different thicknesses.
This study determined the differences of equilibrium moisture content (EMC) and moisture absorption rate of six kinds of tree species with four thicknesses of wood by the static saturated salt solution adsorption method. The Hailwood Horrobin (H-H) model was used to fit the isothermal moisture absorption curve of wood, analyze the monolayer molecular adsorption and multi-layer molecular adsorption, and explore the reasons for the differences in moisture absorption of different tree species with different thickness of wood. This data will provide solutions to the quality problems of furniture products and interior decoration materials in production and use.
EXPERIMENTAL
Test Materials
Populus deltoides was provided by Jiangsu Siyang Meizhi Wood Industry Co., Ltd. (Siyang, China); Picea asperata, Carpinus L., Quercus spp., and Zelkova schneideriana were provided by Ningbo Senhe musical instrument Co., Ltd. (Ningbo, China). Xanthostemon melanoxylon was provided by Zhejiang Huzhou Lezai Wood Industry Co., Ltd. (Huzhou, China).
Sample preparation: The sample is heartwood, the specifications were 20 mm (L) × 20 mm (R) × 1 mm (T), 20 mm (L) × 20 mm (R) × 4 mm (T), 20 mm (L) × 20 mm (R) × 8 mm (T), 20 mm (L) × 20 mm (R) × 20 mm (T). Initial moisture content: Carpinus L 10.53%, Picea asperata 10.49%, Quercus spp. 12.49%, Populus deltoides 12.77%, Zelkova schneideriana 12.45%, Xanthostemon melanoxylon 9.79%.
The wood of six tree species was sawn into 20 mm × 20 mm × 1 mm (T), 20 mm × 20 mm × 4 mm (T), 20 mm × 20 mm × 8 mm (T ), and 20 mm × 20 mm × 20 mm (T) samples. The thickness was in the tangential direction, and there were five static saturated salt solution moisture absorption test samples conducted for each tree species and thickness.
Determination of Wood Density and Chemical Composition Content
The measured density is the full dry density and the method for determining the density of wood was in accordance with the standard Method for determination of the density of wood (GB/T 1933-2009 Method for determination of the density of wood). The contents of benzene alcohol extract, acid insoluble lignin, holocellulose, and pentose in wood were determined as described (Luostarinen and Hakkarainen 2019; Domingos et al. 2020).
Determination of Crystallinity
The crystallinity of absolute dry wood powder passing through a 100-mesh sieve was determined. The combined multi-functional horizontal X-ray diffractometer (XRD; UItima IV, Nippon Science Company, Japan) was used with the following parameters: wavelength input 0.154058 nm, slit device Ds=1°, Ss=1°, Rs=0.3mm, voltage 40 kV, current 30 mA, scanning angle 4 to 50°, sampling spacing 0.1°, and scanning speed 5 °/min.
Static Saturated Salt Solution Humidity Absorption Test
The dryer was placed in a constant temperature and humidity box with a set temperature of 30 °C. The saturated salt solution inside the dryer stabilizes the relative humidity (RH) (Li et al. 2016). The saturated salt solution is shown in Table 1.
Table 1. Relative Humidity Corresponding to 30 °C Saturated Salt Solution
The moisture absorption test of the sample consists of four stages: stage A: 0% to 32% RH; Stage B: 32% to 56% RH; Stage C: 56% to 75% RH; Stage D: 75% to 97% RH. The sample was dried to absolute dryness, weighed (M0), and then subjected to isothermal adsorption test phase A. During the moisture absorption process, the sample was weighed again in different time periods and recorded as Mi. The measurement time i was 2, 4, 6, 18, 30, 42, 54, 66, and 78 h after the sample is put into the dryer. The sample mass was measured every 48 h until moisture absorption equilibrium was reached. After the sample reached the isothermal moisture absorption equilibrium in each stage, the saturated salt solution was replaced for the moisture absorption test in the next stage.
The moisture content was calculated by Eq. 1,
(1)
where MCi is the moisture content of wood at moisture absorption i time, Mi is the mass of wood sample with moisture absorption i time, and M0 is the absolute dry mass of hygroscopic sample.
The moisture absorption rate is the change of hourly moisture content of the sample per unit of time, based on two adjacent measurements, calculated as follows,
(2)
where Vi1~i2 is the average hourly moisture absorption rate in the i1 to i2 time period, Mi1, Mi2 is the moisture content of sample at measuring time i1, i2, and i1 and i2 are adjacent time points for sample quality measurement.
H-H Model Isothermal Adsorption Curve Fitting
The H-H model was used analyze the isothermal adsorption of wood moisture (Hailwood and Horrobin 1946); it can predict the change of monolayer molecular adsorption moisture content in a wide RH range, which is in line with the characteristics of wood dry shrinkage and wet expansion (Hill et al. 2010).
The EMC of wood in different RH environments can be expressed as follows,
(3)
where M is the equilibrium moisture content of the sample, Mh is the amount of water adsorbed by monolayer molecules, Md is the amount of water adsorbed by multilayer molecules, W is the mass of absolutely dry wood containing adsorption sites per mole (g·mol-1), H is relative humidity, K1 is the equilibrium constant between monolayer molecular adsorbed water and multilayer molecular adsorbed water, and K2 is the equilibrium constant between water adsorbed by multilayer molecules and environment temperature and humidity.
Equation 3 can be sorted into the following polynomial form,
(4)
where A, B, and C are polynomial fitting parameters. The relationship between fitting parameters and W, K1, and K2 is shown below.
(5)
(6)
(7)
In this study, Origin 2021 software (Origin Lab, America) was used to fit the water isothermal adsorption curve.
RESULTS AND DISCUSSION
Wood Density, Chemical Composition Content, and Crystallinity
The density values of six kinds of wood were determined as follows: Picea asperata, 0.36 g/cm3; Populus deltoides, 0.44 g/cm3; Zelkova schneideriana, 0.67 g/cm3; Carpinus L, 0.71 g/ cm3; Quercus spp., 0.75 g/cm3; and Xanthostemon melanoxylon, 1.23 g/cm3.
Table 2 shows the average value of chemical composition of six tree species. The content of holocellulose of X. melanoxylon, P. asperata, Z. schneideriana, Carpinus L, Quercus spp. and P. deltoides decreased successively. The content of benzene alcohol extract of X. melanoxylon was higher, and the content of benzene alcohol extract of Z. schneideriana was the lowest.
Table 2. Average Content of Chemical Components of 6 Tree Species
Figure 1 shows the diffraction intensity curves of six tree species by X-ray diffraction. The crystallinity of six tree species was obtained as follows: Z. schneideriana, 41.65%; Quercus spp., 42.26%; P. asperata, 42.35%; P. deltoides, 43.39%; Carpinus L, 45.38%; and X. melanoxylon, 49.32%.
Fig. 1. Diffraction intensity curve of wood X-ray diffraction method
Equilibrium Moisture Content and Moisture Absorption Rate of Wood
Balanced moisture content of wood
Figure 2 shows the relationship between mean moisture absorption EMC and thickness of six tree species in four stages. In stages A, B, and C, the EMC of P. asperata and P. deltoides increased with the increase of sample thickness. The reason may be that P. asperata and P. deltoides have low density and high porosity (Plotze and Niemz 2011), and water molecules more easily enter the interior of wood and combine with free hydroxyl groups (Jang et al. 2020). During moisture absorption, hydrogen bonds within the polysaccharide structure break, and volume expansion creates cracks between wood microfibrils (Hartley et al. 1992). More adsorption sites are produced, which increases the wood EMC. The EMC of Z. schneideriana and Carpinus L had little correlation with the change of sample thickness. The EMC of thin samples of Quercus spp. and X. melanoxylon was higher than that of thick samples. The reason is that the density of Z. schneideriana and Carpinus L. species is large. The structure is relatively compact, and the total porosity is about 60% of P. asperata (Plotze and Niemz 2011). When the sample thickness increases, it is relatively difficult for water to enter the wood. A number of hydrogen bonds break, and the extent of volume expansion decreases, resulting in small changes in EMC of samples with different thickness. Quercus spp. and X. melanoxylon have higher wood density, high extractives content, and compact structure. When the sample thickness increases, it is difficult for water to enter the wood, with less hydrogen bond fracture and volumetric expansion produces fewer cracks (Mantanis et al. 1994; Morisato et al. 2002), resulting in reduced EMC.
The increase of EMC in phase D of the moisture absorption test was much greater than that in phases A, B, and C. The possible explanation is the adsorption of free water in the capillary (Hill 2006). The other explanation is related to the glass transition of hemicellulose (Vrentas and Vrentas 1991; Thygesen et al. 2010) and the water molecular monomer first binds to the adsorption site at low RH, converted to dimer at high RH (Willems 2018).
Figure 3 shows the relationship between EMC mean value of four kinds of wood thickness and tree species. In stages A, B and C, the EMC of P. asperata was always higher than that of other tree species, which may be due to the high content of holocellulose and low crystallinity of P. asperata, making the moisture absorption of P. asperata larger. The average EMC of P. deltoides was the lowest, which was related to the lowest content of cellulose and hemicellulose. With the increase of RH, the EMC of Quercus spp. and X. melanoxylon gradually became smaller than that of Z. schneideriana and Carpinus L, which may be due to the high density and extract content, resulting in low expansion rate and reduced multi-layer molecular adsorption (Morisato et al. 2002). The EMC of Carpinus L was always lower than that of Z. schneideriana because the content of Z. schneideriana holocellulose was high and the crystallinity was lower than that of Carpinus L. In phase D, the EMC of the four kinds of wood decreased with the increase of tree species density, which may be due to the minimum density of Z. schneideriana and the high content of holocellulose. In addition, the wood density and extract content limited the amount of water adsorbed by multi-layer molecules.
Fig. 2. Equilibrium moisture content of six tree species wood with four thicknesses
Fig. 3. Equilibrium moisture content of six tree species wood with four thicknesses (4 sample thicknesses of 1mm-20 mm)
Differences of moisture absorption rate of wood
Figures 4 through 7 show the relationship between moisture absorption rate and moisture absorption time of samples of six tree species and four thicknesses in stages, A, B, C, and D (take 4 mm, 20 mm thick sample and P. asperata, Quercus spp. as examples). At the beginning of each stage, the sample was put into the dryer.
The partial pressure of water vapor in the dryer was higher than that on the wood surface. When the wood absorbs water vapor quickly, the moisture absorption rate is large, and the moisture content increases rapidly. With the progress of moisture absorption process, the water vapor differential pressure decreases, and the moisture absorption rate of wood gradually decreases to zero.
The time required for each sample in the figure to reach EMC ranged from 78 h to 1,326 h. When the moisture absorption time was longer than 78 h, the moisture absorption rate was less than 0.05% per h, which was difficult to distinguish, so only the moisture absorption rate in the first 78 h was analyzed.
Fig. 4. Relationship between moisture absorption rate and moisture absorption time of wood with 4 mm thickness (A, B, C, D are four stages of moisture absorption test: A: 0%-32%RH; B: 32%-56%RH; C: 56%-75%RH; D: 75%- 97%RH)
Fig. 5. Relationship between moisture absorption rate and moisture absorption time of wood 20 mm thickness (A, B, C, D are four stages of moisture absorption test: A: 0%-32%RH; B: 32%-56%RH; C: 56%-75%RH; D: 75%- 97%RH)
Fig. 6. Relationship between hygroscopic velocity and hygroscopic time of P. asperata with four thicknesses
Fig. 7. Relationship between hygroscopic velocity and hygroscopic time of Quercus spp. with four thicknesses
At the beginning of the moisture absorption test in Figs. 4 and 5, the moisture absorption rate of P. asperata was the largest. It may be that the axial tracheids of coniferous P. asperata are larger in diameter than those of other five broad-leaved wood fiber cells, and there are many pits on the cell wall with large pore diameter. Water can enter the wood faster, so the water movement speed is faster than that of broad-leaved wood. The moisture absorption rate of P. deltoides was slightly higher than that of Z. schneideriana, Carpinus L and Quercus spp., which may be caused by the joint influence of two aspects. The density of P. deltoides is slightly higher than that of P. asperata and lower than that of the other four tree species. The higher density of broad-leaved wood results in more wood fiber content; the thicker cell wall of wood fiber means there are smaller cell cavities. However, the lignin content and crystallinity of P. deltoides are high, and the content of holocellulose is the lowest, resulting in less free hydroxyl content in P. deltoides wood. Among other broad-leaved trees, the initial moisture absorption rate of Z. schneideriana and Carpinus L was similar and greater than that of Quercus spp., which may be due to the fact that Quercus spp. has more inclusions, which affect the pore circulation of cell wall, block the grain pores, and affect the transmission of water, resulting in a small moisture absorption rate. The lowest moisture absorption rate of X. melanoxylon in the initial stage is because it has the highest density, crystallinity, and benzene alcohol extract contents, and it has the longest time to achieve absorption stability. When the moisture absorption rate of the other five species gradually decreased to zero, X. melanoxylon did not reach the moisture absorption balance. Figures 6 and 7 show that the moisture absorption rate decreased with the increase of sample thickness because the difficulty of water entering the wood increases with the increase of sample thickness.
In Figs. 4 to 7, there are differences in the time required for the samples in stages A, B, C, and D to reach EMC. The time required for the continued moisture absorption to approach zero increased with increasing RH, which may be because the higher RH is associated with the higher EMC of wood. With the increase of sample thickness, the moisture absorption extent of wood decreased as a whole. The time required for moisture absorption to reach EMC was prolonged because the increase of sample thickness requires more pores for moisture diffusion into the wood, resulting in the decrease of velocity.
There were differences in the time for wood samples of different tree species and different thicknesses to reach EMC. The greater density between different tree species and the greater thickness of the sample of the same tree species resulted in a longer time required to reach the moisture absorption equilibrium. This is due to the different moisture absorption rate, which is also one of the indicators to measure the moisture absorption.
H-H Model Isothermal Moisture Absorption Curve Fitting
The parameters of isothermal adsorption curve fitted by H-H model are shown in Table 3. The goodness of fit R2 was greater than 0.99, indicating that the Hailwood-Horrobin model had a good fitting effect on the isothermal moisture absorption curve of six tree species. The k1, k2, and W values were calculated by combining parameters A, B, and C with Eqs. 5, 6, and 7. The parameter values are close to the fitting parameters of previous studies (Papadopoulos and Hill 2003; Zaihan et al. 2009). Figure 8 is the isothermal moisture absorption curve of H-H model of six tree species using the data in Table 3.
Table 3. Isothermal Adsorption Curve Parameters Calculated by H-H Model
