Abstract

Materials

Six-year-old E. pellita wood was obtained from the Eucalyptus Research Center in Guangdong Province, China. E. pellita wood samples were oven-dried below 80°C before vacuum heat treatment, and specimens measuring 330 by 40 by 40 mm (length by trangential by radial) were prepared from the ovendried wood. Specimens were divided into two groups. One group was heat treated at 80°C, 120°C, 160°C, 200°C, 240°C, and 280°C for 4 hours (time for temperature increasing not included) under pressures of −0.08 to −0.09 MPa, and the other group was used as a control.

Vacuum heat treatment equipment (HJ-ZK60; Guangzhou Hengjun Company), with a minimum pressure of −0.09 MPa and a maximum temperature of 350°C, was used to treat wood. The equipment was composed of two parts: a processing box and a vacuum pump. During treatment, the vacuum condition was realized by a vacuum pump, and a high-temperature environment was realized by a heating cord around the processing box. There were three layers for the vacuum box, heating cords were located between the two layers of wall, and when the cords were heating, the wood in the vacuum box was also heating.

Wood powder was used to test the chemical composition, crystallinity, and FTIR spectra. First, samples were divided into two groups: one as control and the other prepared for vacuum heat treatment at 80°C, 120°C, 160°C, 200°C, 240°C, and 280°C, respectively, for 4 hours. Second, all of the wood, including both treated and untreated, was chopped into small strips and then smashed into powder in a grinder. Finally, wood powder was graded by sieves; powder between 40 and 60 mesh was used for chemical composition testing and that between 80 and 100 mesh was used for crystallinity testing. Wood powder tested for FTIR spectra should be shaved from the wood by a sharp blade (the smaller the better). The moisture content of all the powder was dried to 0 percent.

Testing methods

Chemical composition

Chemical composition tested in this experiment and its corresponding testing methods are as follows:

Lignin: Tested by Chinese Standard GB/T 2677.8-94 (Standardization Administration of the People's Republic of China [SAC] 1994b), which equals the American standard TAPPI T222 om-11.

Holocellulose: Tested by Chinese Standard GB/T 2677.10-1995 (SAC 1995) were as follows: (1) Weighed 2 g of powder and recorded the weight as m₀. (2) Bagged the power in filter paper and then extracted it with a benzene and alcohol solution; after extracting, the power bag was air-dried. (3) Removed the powder from the filter paper bag, put it in a 250-mL container, and then added 65 mL of distilled water, 0.5 mL of glacial acetic acid, and 0.6 g of sodium chlorite, in that order, and heated the mixture for 1 hour at 75°C. (4) Added 0.5 mL of glacial acetic acid and 0.6 g of sodium chlorite in the mixture and heated them for 1 hour at 75°C again. (5) Repeated Step 4 until the power turned white. (6) Washed the residue with distilled water until the residue showed no acid, and then washed it with acetone three times. (7) Removed the solvent and dried the residue at 105°C to a constant weight (labeled as m₁). The content of the holocellulose was calculated by the following equation:

where

• X ₁ = content of holocellulose in wood (%),

• m ₁ = weight of holocellulose after drying (g), and

• m ₀ = weight of wood powder sample under ovendried condition (g).

α-Cellulose: Tested by the nitric acid–ethanol method according to the following steps: (1) Prepared 500 mL of the 68 percent nitric acid and absolute alcohol mixture with a ratio of 1:4. (2) Weighed 1 g of powder (labeled as m₀). (3) Put the powder in the prepared 25-mL mixed solution. (4) Heated the 68 percent nitric acid and absolute alcohol mixture in boiling water for 1 hour. (5) Removed the solvent and kept the residue. (6) Repeated Steps 3, 4, and 5 until the residue turned white. (7) Washed the whitened residue with the 68 percent nitric acid and absolute alcohol mixture, hot water, and absolute alcohol, in that order. (8) Dried the residue at 105°C until the moisture content was reduced to 0 and then weighed it (labeled as m₁). (9) Treated the residue at 500°C and then weighed it (labeled as m₂). The content of the α-cellulose was calculated by the following equation:

where

• C = content of α-cellulose in wood (%),

• m ₀ = weight of the wood powder sample (g),

• m ₁ = weight of residue after drying at 105°C (g), and

• m ₂ = weight of residue after treating at 500°C (g).

Benzene-alcohol extractive: Tested by Chinese Standard GB 2677.6-94 (SAC 1994a), which equals the American standard TAPPI T204-om97.

Crystallinity

The crystallinity of wood was determined by an X-ray diffractometer (Philips X'Pert PRO) with a tube voltage of 40 kV and a current of 40 mA. Cu targets were used as the X-ray tube, a nickel layer was used to eliminate radiation from CuKα, a θ/2θ link scan was used as a measuring method, and the emission slit was 0.76 mm.

Before testing, wood powder was pressed into a thin flat used for the 2θ scan. The scan range was from 5° to 50°, and the scan speed was 8° min⁻¹.

Crystallinity was calculated by the Segal formula as follows:

where

• C_rI = relative crystalline degree, namely, crystallinity (%);

• I ₀₀₂ = maximum intensity of (002) crystal lattice diffraction angle (arbitrary unit), appearing in scan curve when 2θ was near 22°; and

• I_am = scattering intensity of background diffraction in amorphous region when 2θ was near 18° and whose unit was the same with that of I₀₀₂.

FTIR spectroscopy

For these tests, KBr powder was mixed with the powder of wood samples using a weight ratio of 100:1. Then this mixture was pressed into a small disc with a diameter of 13 mm under a pressure of 75 kN cm⁻². The IR spectra of every sample were analyzed with a PerkinElmer Spectrum One FTIR spectrometer. Averaged curves from 32 repeated scans were obtained between wave numbers of 4,000 and 400 cm⁻¹. Precision of the spectrometer was 4 cm⁻¹.

Chemical composition

Thermal decomposition degrees of chemical components in wood depend on different treatment temperature. Relative content of each component showed different trends with an increase of temperature from 80°C to 280°C (Fig. 1). As shown in Figure 1, compared with untreated samples, content of holocellulose and α-cellulose of treated wood decreased gradually as temperature increased and dropped markedly above 200°C. Content of holocellulose and α-cellulose decreased from 74.25 and 45.43 percent, respectively, under the untreated condition to 49.36 and 12.97 percent, respectively, when treated at 280°C. Lignin content increased as temperature rose and changed from 34.1 percent at 200°C to 54.98 percent at 280°C. Alcohol benzene extractive increased as temperature rose; the maximum value was 4.54 percent at 240°C.

View Figure

• Download as Powerpoint
• Download Figure

The changing trends of these several chemical compositions were consistent with many studies on heat-treated wood. Deng et al. (2009) treated Cunninghamia lanceolata thinning wood in hot air for 4 hours at 200°C, finding that holocellulose, cellulose, Klason lignin, and benzene-alcohol extractive changed from 63.83, 45.73, 32.12, and 3.18 percent to 55.12, 43.89, 39.96, and 3.37 percent, respectively. Among the three main chemical components of wood, hemicellulose is the first structural compound to be thermally affected, even at low temperatures. The degradation starts by deacetylation, and the released acetic acid acts as a depolymerization catalyst that further increases polysaccharide decomposition (Tjeerdsma et al. 1998, Sivonen et al. 2002). Cellulose is less affected by heat treatment below 200°C, probably because of its crystalline nature (Bourgois and Guyonnet 1988), but as temperature increases, the structure of crystalline is destroyed, and the degree of polymerization is reduced at about 240°C (Yin 1996).

Lignin increased as treatment temperature rose, according to many researchers. Lignin content of Pinus pinaster increased from 28 percent to 41, 54, and 84 percent, respectively, at 260°C for 0.5, 1, and 4 hours (Bourgois and Guyonnet 1988); for Scots pine and birch treated between 205°C and 230°C during 4 and 8 hours, the lignin content increased from 24.5 to 38.7 percent and from 21.8 to 35.8 percent, respectively (Zaman et al. 2000). The increase of lignin content as temperature rises might be caused by the following reasons. First, the small molecule detached from hemicellulose or cellulose enclosed in the lignin was hard to detach because of the three-dimensional network structure of lignin. Second, as an amorphous polymer, lignin has a character of glass transition; when heated at about 200°C or above, lignin steps into the glass transition stage, becomes sticky, works as an adhesive, and glues the degradation product detached from hemicellulose or cellulose to lignin molecules, enhancing lignin content as temperature rises (Yin 1996). Finally, lignin determined using the standard methods in heat-treated wood samples is probably not a pure lignin. Several authors have suggested that polycondensation reactions with other cell wall components results in further cross-linking and contributes to the increase in apparent lignin content (Tjeerdsma and Militz 2005, Boonstra and Tjeerdsma 2006).

Crystallinity

Crystallinity of wood exerts a significant effect on the physical-mechanical properties of wood or fiber, such as tensile strength, modulus of elasticity, hygroscopicity, moistening, and so on. Besides age and position, heat treatment also had an effect on the extent of crystallization. Changes of wood X-ray diffraction spectra treated under different vacuum heat treatment temperatures can be seen in Figure 2.

View Figure

• Download as Powerpoint
• Download Figure

As shown in Figure 2, the X-ray diffraction intensity of wood treated at 160°C and 200°C was larger than the intensities at other temperatures. Crystallinity changes at different temperatures are shown in Figure 3. Crystallinity mainly increased as temperature increased from 80°C to 200°C, and then decreased once the temperature was above 200°C (Fig. 3). As the temperature rose from 240°C to 280°C, crystallinity increased again from 43.76 to 45.86 percent. The maximum crystallinity was 52.57 percent at 200°C, an improvement of 16 percent compared with the untreated samples. Crystallinity is an important factor affecting the physical, mechanical, and chemical properties of wood. The larger the crystallinity, the higher the tensile strength, bending strength, and dimensional stability of wood (Yin 1996). In a previous study, when we heated E. pellita wood under vacuum, dimensional stability increased by about 30 percent, and modulus of elasticity increased by 25.2 percent at 200°C (Wang et al. 2014).

View Figure

• Download as Powerpoint
• Download Figure

A similar changing tendency of crystallinity was also found in other studies; for example, the crystallinity of heat-treated Larix spp. first increased, then decreased, and finally increased, and the value of crystallinity was 50.77, 56.46, 43.59, and 64.15 percent when treated at 180°C, 200°C, 220°C, and 240°C, respectively, for 4 hours in air (Sun and Li 2010).

Cellulose is composed of two phase structures: a crystalline region and an amorphous region. The increase in crystallinity as temperature rises might be caused by two reasons. First, degradation of amorphous cellulose, as reported earlier, results in a decreased accessibility of hydroxyl groups to water molecules (Wikberg and Liisa 2004, Bhuiyan and Nobuyuki 2005). Second, crystallization in quasi-crystalline amorphous regions is due to the rearrangement or reorientation of cellulose molecules inside these regions (Akgül et al. 2007). In crystalline regions, as the temperature increases, the segment motion of cellulose chain is enhanced, causing part of the space between the cellulose chain to decrease and a new hydrogen bond to form so that wood crystallinity is enhanced. In this study, when heated at over 240°C, crystallinity of wood decreased again; when the temperature increased to over 240°C, segment motion was so acute that the hydrogen bond was destroyed, the intermolecular force between the cellulose chain was reduced, wood crystallinity decreased, and in the macroscopic view, the mechanical property dropped.

FTIR analysis

Figure 4 shows FTIR spectra of untreated and heat-treated E. pellita wood at 120°C, 200°C, and 280°C. Assignments of main absorption are listed in Table 1.

View Figure

• Download as Powerpoint
• Download Figure

Table 1. Absorption band in Fourier transform infrared spectra under different temperatures.

View Fullscreen

Figure 4 and Table 1 show that absorption peaks of some bands shifted and that intensity changed in the FTIR spectra when heated. The absorption peak band of the benzene skeleton at 1,595 and 1,505 cm⁻¹ for the untreated samples shifted to 1,608 and 1,514 cm⁻¹ at 280°C. The stretching vibration peaks of O–H and N–H at 3,423 cm⁻¹, C–H at 2,917 cm⁻¹, C=O at 2,917 cm⁻¹, and –OCOR at 1,235 cm⁻¹ for untreated samples shifted to 3,379, 2,903, 1,713, and 1,215 cm⁻¹, respectively, when treated at 280°C.

The shift to lower wave numbers of C–H and C=O at 2,917 cm⁻¹ and –OCOR at 1,235 cm⁻¹ for heat treatment samples was probably caused by the intramolecular π-π conjugator effect (Li 2003). When wood was heated at high temperatures, new double or triple bonds formed because of an active group reaction under the thermal effect; if the new double or triple bonds interval linked with double bonds, such as acetyl groups on xylan or –OCOR groups by single bonds, then characteristics of double or triple bonds would decrease (Liu et al. 2000), reducing the absorption peak to a lower wave number.

Absorption intensity of C=O stretching vibration at 1,739 cm⁻¹ decreased, consistent with hemicellulose decomposition. When hemicellulose began to decompose under the thermal effect, the number of carboxyl and carbonyl groups, located mainly in hemicellulose (Yin 1996), and the absorption peak, which was at 1,739 cm⁻¹ in the infrared spectra (Faix 1991), were reduced, further causing the absorption intensity of C=O stretching vibration at 1,739 cm⁻¹ to decrease.

The band absorption intensity at 1,595 and 1,505 cm⁻¹, assigned to benzene skeleton vibration (Faix 1991), increased as the heat treatment temperature rose, indicating that lignin content rose as temperature increased. Absorption peak position at 3,423 cm⁻¹ assigned to O–H stretching vibration (Faix 1991, Schwanninger et al. 2004) shifted to lower wave numbers, and the band broadened as temperature increased, probably caused by hydrogen bond association (Li 2003) under the thermal effect, reducing numbers of hydrophilic hydroxyl groups in wood and increasing crystallinity, hygroscopicity, and dimensional stability.