Esterification of solid wood for plastic forming

Abstract

Plastic formation from wood has recently attracted immense attention as a new method of producing three-dimensional materials by only one press. In this study, hydroxyl groups in (block-shaped) wood were replaced with ester groups to realize plastic formation. FT-IR measurements indicated that most of the hydroxyl groups in the block-shaped wood were replaced by ester groups after 4 h of treatment. The thermal properties of the esterified wood were investigated; the modulus of elasticity decreased remarkably when heated. Particularly, propionylated wood demonstrated a distinct softening point at relatively low temperatures, demonstrating its thermoplasticity. Additionally, the propionylated wood was stretched extensively by pressing with heat, showing its fluidity. Observation of the end grain of the esterified wood samples after pressing revealed that slippage occurred between the cells, resulting in the bulk flow of the wood. Finally, plastic formation using propionylated wood as the raw material was attempted, and a cup-shaped three-dimensional product was successfully formed. Contact angle measurements demonstrated that the cup-shaped molded product exhibited high water resistance. Thereafter, re-forming was attempted, using crushed fragments of the cup-shaped molded product, and a plate-shaped product was successfully re-formed. Thus, esterification is a promising method for increasing fluidity in solid wood, helping to enable its utilization in plastic forming.

Full Article

Esterification of Solid Wood for Plastic Forming

Mitsuru Abe,^a,* Yukiko Enomoto,^b Masako Seki,^a and Tsunehisa Miki ^a

Plastic formation from wood has recently attracted immense attention as a new method of producing three-dimensional materials by only one press. In this study, hydroxyl groups in (block-shaped) wood were replaced with ester groups to realize plastic formation. FT-IR measurements indicated that most of the hydroxyl groups in the block-shaped wood were replaced by ester groups after 4 h of treatment. The thermal properties of the esterified wood were investigated; the modulus of elasticity decreased remarkably when heated. Particularly, propionylated wood demonstrated a distinct softening point at relatively low temperatures, demonstrating its thermoplasticity. Additionally, the propionylated wood was stretched extensively by pressing with heat, showing its fluidity. Observation of the end grain of the esterified wood samples after pressing revealed that slippage occurred between the cells, resulting in the bulk flow of the wood. Finally, plastic formation using propionylated wood as the raw material was attempted, and a cup-shaped three-dimensional product was successfully formed. Contact angle measurements demonstrated that the cup-shaped molded product exhibited high water resistance. Thereafter, re-forming was attempted, using crushed fragments of the cup-shaped molded product, and a plate-shaped product was successfully re-formed. Thus, esterification is a promising method for increasing fluidity in solid wood, helping to enable its utilization in plastic forming.

Keywords: Wood; Flow Forming; Thermoplasticity; Esterification; Re-molding; Water resistance

Contact information: a: Multi-Materials Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), 463-8560, Nagoya, Japan; b: Department of Biomaterial Sciences, Graduate School of Agricultural and Life Sciences, The University of Tokyo, 113-8657, Tokyo, Japan

* Corresponding author: m.abe@aist.go.jp

INTRODUCTION

Cutting, scraping, bending, and compression techniques are typically used in the fabrication of two-dimensional and/or three-dimensional materials from wood. Plastic formation from wood has recently attracted immense attention as a new method of producing three-dimensional materials. For example, wood-plastic composites (WPCs), produced by kneading wood flour and plastic, are being actively studied as raw materials in plastic formation (Clemons 2002; Chan et al. 2018; Mazzanti and Mollica 2019). However, to produce such WPCs, wood is used in its powder form. This powdering pretreatment is both time- and energy-consuming; further, unfortunately, the mechanically beneficial fiber structure of wood is lost during this pretreatment.

To overcome these shortcomings, a plastic formation method from wood has been developed. This method does not require the use of powdered wood or kneading of plastic and wood flour. In this method, a certain amount of water is flowed into the wood by heating and compression (Yamashita et al. 2006, 2009). In conventional wood compression, lumens of wood cells disappear, but the positional relation between the wood cells does not change. In the new method, however, the wood flow is caused by the sliding of each cell wall under certain conditions. Owing to the potential of substantial deformability, one-shot production of a complex molded body becomes possible. Recently, it was clarified that the fluidity of the wood, which is a measure of how easily the wood flows under low temperature and/or pressure, could be improved by impregnating the wood with a monomer, such as a phenol or melamine resin as a binder, rather than water (Miki et al. 2014a,b). Furthermore, by curing the resin monomer during heat treatment, water resistance was successfully imparted to the flow-formed product, and the dimensional stability was remarkably improved. However, because these resins are thermoset and cannot be re-formed, their recyclability remains a challenge. Therefore, a monomer of methyl methacrylate (MMA), a thermoplastic resin, was used as a binder to produce a composite material of wood and poly(methacrylic acid) (PMAA). As a result, thermoplastic flow-formed products that could be re-formed repeatedly were developed (Seki et al. 2014, 2016, 2017). However, as water cannot be used as a solvent in the impregnation solution because MMA is highly hydrophobic, it is necessary to employ an organic solvent, such as alcohol. Chemical modification of the wood surface can also be considered. By substituting the hydroxyl groups in the wood with the acetyl groups, the affinity between the wood and MMA is improved. Although these pretreatments demonstrated several effects, they are complicated and require further improvement. In addition to the above-described examples, using a thermoset, the resin composite-type flow forming of wood utilizes a large amount of petroleum-derived resins. Therefore, the development of another novel method is required.

Preparing a wood thermoplastic by derivatization appears to be one of the most promising solutions. Wood flour has been reported to exhibit thermal fusibility after esterification treatment (Funakoshi et al. 1979; Shiraishi et al. 1982; Matsuda and Ueda 1985; Hon and Xing 1992). An example of film-forming that utilized thermal fusibility and esterification was also reported (Shiraishi et al. 1982). Recently, several types of thermoplastic wood flour with ester substituents produced by efficient esterification processes of wood flour have been developed. Xie et al. (2007) reported an efficient esterification process for wood-based lignocellulosic materials in ionic liquids. They also reported that acetylated wood showed clear glass transition temperatures at 131.7 °C. Su et al. (2015) successfully obtained injection-molded wood-based plastics through a simple reaction of wood flour with phthalic anhydride in 1-methylimidazole at 15 °C for 2 h. Chen et al. (2017) developed highly translucent wood plastics by the esterification of ball-milled mulberry wood in a mixed solution of ionic liquid/dimethyl sulfoxide. Hassan et al. (2000a, b) succeeded in subjecting bagasse fibers to an esterification treatment with carboxylic anhydride and acetone to achieve thermal fusibility. Further, they stated that the use of fibrous biomass instead of wood flour affords the possibility of producing a novel material that utilizes the strength of fibers in plant biomasses. However, they only evaluated the meltability by observing the surface of the esterified bagasse after heat treatment with pressure and did not proceed to develop a three-dimensional flow forming of bagasse. Furthermore, for bulk wood, only some evaluations within the elastic range, such as the dimensional stability and dynamic viscoelasticity, have been performed (Nakano and Nakamura 1986a,b,c, 1987; Nakano 1988; Rowell 2013). Three-dimensional materials by flow forming have yet to be developed. In the flow phenomenon of bulk wood, it is necessary to soften the wood and induce a change in the mutual position of the cells owing to slippage between the cells. Therefore, an analysis of the plastic region is necessary. Additionally, after intercellular slippage occurred, the cells were required to recombine to maintain their shapes after the flow.

In this study, the hydroxyl groups in block-shaped bulk woods were replaced with ester groups using a mixture of trifluoroacetic acid and carboxylic anhydride, and the physical, chemical, and thermal properties of the esterified woods were evaluated. Subsequently, a three-dimensional flow-formed product was developed. Furthermore, in addition to the assessment of the water resistance of the obtained product, its re-formation was also attempted to evaluate its recyclability.

EXPERIMENTAL

Original Materials

Pieces of Japanese cypress (Chamaecyparis obtusa, softwood) (Yamani, Co., Ltd., Nagano, Japan) and beech (Fagus crenata, hardwood) (Yamani, Co., Ltd., Nagano, Japan) with dimensions of 5.0 mm (longitudinal direction, L) × 20 mm (radial direction, R) × 20 mm (tangential direction, T) were used as wood specimens. They were cut from a larger stick with a transverse section of 20 mm (R) × 20 mm (T). The initial mass of these specimens was ~0.7 g for cypress and ~1.2 g for beech. Trifluoroacetic acid, acetic anhydride, propionic anhydride, and methanol were purchased from FUJIFILM Wako Pure Chemical Corp. (Osaka, Japan) and utilized as obtained.

Methods

Esterification of wood

Trifluoroacetic acid and carboxylic anhydride were mixed at a ratio of 1 mol:1.1 mol, and the mixture was aged for 30 min and used as an esterification reagent. The esterification reagent (30 mL) was added to a one-neck reactor flask (300 mL, separate round-bottom flasks), and the specimen (dried wood) was placed into it. A stainless-steel net was placed over the wood to prevent the wood from floating in the solution. Thereafter, a reflux tube was set up to form a closed system. The degassing operation was performed at -80 kPa, and the injection of nitrogen gas was repeated three times to remove residual air from the wood and replace the whole reaction system with nitrogen. The flask was immersed in an oil bath that was heated to 60 °C, and the reaction solution was gently stirred for 1 h or 4 h. Thereafter, the wood chips were collected, rinsed with 20 mL of methanol, and immersed in 300 mL of methanol to stop any further reaction. The methanol was replaced with unused methanol every 12 h, and the operation was repeated three times to effectively wash the wood samples.

Table 1. Sample Codes and Differences among the Samples

The obtained esterified wood was dried at 25 °C for 24 h, after which it was dried at 105 °C for 24 h. The samples are abbreviated as Cy_Ac_1, Cy_Ac_4, Be_Ac_1, Be_Ac_4, Cy_Pr_1, and Be_Pr_4, where the first codes, Cy and Be, refer to the types of wood (Cypress, Cy; beech, Be); the next codes, Ac and Be, indicate the introduced ester groups (acetyl, Ac; propionyl, Pr), and the numbers represent the durations of the esterification reactions (1 h or 4 h) (see Table 1).

Before measurements were taken, the wood samples, which included unesterified woods, were abbreviated as Cy_untreated and Be_untreated, and then dried at 105 °C for ≥ 2 h and cooled to 25 °C in a tight box with dried silica gels.

Scanning electron microscopy

Cross-sections of the wood samples were cut using a rotary microtome (RMS, Nihon Microtome Laboratory, Inc., Osaka, Japan). Then the inner structures of the wood samples were observed using a scanning electron microscope (JSM-IT500; JEOL Ltd., Tokyo, Japan) with an Au coating over the cut surface (Quick Coater VPS-020; ULVAC Ltd., Yokohama, Japan). The thicknesses of the cell walls of the woods were measured using ImageJ image analysis software (U.S. National Institutes of Health, ver. 1.8.0_112, Bethesda, MD, USA) with the SEM images at 1700-fold magnification. The thicknesses at 50 points were measured for each wood sample, and the average values were calculated.

FT–IR measurement

Fourier-transform infrared (FT-IR) measurements were performed on a Nicolet 6700 spectrometer (Thermo Scientific Inc., Waltham, MA, USA), and the standard attenuated total reflectance (ATR) method was employed. Flat pressed pieces obtained from the compression test described below were used as test pieces. The measurement was carried out at three or more points on the test piece, and it was observed that there were no noticeable differences in the results.

Dynamic mechanical analysis

A small test piece (1 mm × 1 mm × 1 mm) cut from the wood sample was subjected to DMA (TMA SS6100; Hitachi High-Tech Science Corp., Tokyo, Japan). An oscillating compressive load of 200 ± 100 mN (frequency of 0.03 Hz) in the R direction was applied to the test piece, and dry nitrogen gas was purged during the measurement. The measurement was conducted with increasing temperature from 30 °C to 200 °C at a constant temperature ramping rate of 3 °C/min. The results of the heating process in the second cycle were computed.

Compression test

A small test piece (5 mm × 5 mm × 5 mm) cut from the wood sample was subjected to a compression test (Autograph AG-X plus 20 kN; Shimadzu Corp., Kyoto, Japan; load cell SFL-20KNAG; Shimadzu Corp., Kyoto, Japan). It was set on a press surface that was heated to 160 °C and pressed in the R direction at a speed of 1 mm/min. The press was stopped when the force on the specimen reached 10 kN. The apparent compressive stress was calculated based on the initial surface area of the specimen.

Molding test

Disk-shaped pieces of wood ( = 35 mm; thickness, 5 mm; R direction) were subjected to a molding test. Press molding was performed using a metal mold for the cup-shaped product ( = 50 mm) following the report of Yamashita et al. (2009). The temperature of the mold was raised to 160 °C by heating, and a release agent was applied inside the mold. Thereafter, a wood sample at 25 °C was placed into the mold and pressurized to 200 MPa at a speed of 1 mm/s. After reaching 200 MPa, pressurization was allowed to proceed for 5 min. Thereafter, the mold was air-cooled to ≤ 70 °C, after which the molded sample was removed from the mold.

The molded sample was re-formed as follows: the cup-shaped product obtained by the above molding process was crushed into flakes ( approximately 2 cm), which were then molded into a plate-like product (50 mm (length) × 50 mm (width) × 10 mm (thickness)). The subsequent molding procedure was similar to the one used to obtain the cup-shaped product.

Contact angle measurements

The cup-shaped product was fixed in a horizontal position, and 2 μL of pure water was dropped on the side surface at 25 °C. The contact angles were measured after 1 s and 20 s of deposition (PCA-11; Kyowa Interface Science Co., Ltd., Niiza, Japan).

RESULTS AND DISCUSSION

Esterification of Wood

Acetylation or propionylation of the wood samples was conducted. Figure 1 shows the changes in their appearances before and after esterification, and Table 2 shows the changes in weight.

Fig. 1. Photographs of the wood samples before and after esterification

The cypress samples shrank slightly with esterification (Figs. 1a, 1b, and 1c), but beech tended to swell (Figs. 1d, 1e, and 1f). The samples were weighed, and the weight per gain (WPG) of both cypress and beech increased with increasing treatment time (Table 2).

Table 2. WPG of the Esterified Woods

Fig. 2. Scanning electron microscopy images of untreated and 4 h propionylated woods

In both the cypress and beech samples, the propionylated woods exhibited larger WPG than the acetylated wood owing to the difference in the molecular weights of their substituents. When one was replaced with an acetyl group, the molecular weight increase was 42, whereas it was 56 when the hydroxyl group was replaced by the propionyl group. Several differences were observed in the tree species. Cypress exhibited higher WPG than beech under any of the reaction conditions. This may have been caused by the differences in the composition of each of the woods.

In the esterification of wood, it has been reported that the hydroxyl groups in lignin react more easily than those in cellulose (Rowell 1982; Rowell et al. 1994). Cypress, a kind of softwood, contains larger amounts of lignin than beech, which is a hardwood. Therefore, the esterification of cypress proceeded more easily than that of beech. It is also possible that the swelling of the cell wall reached its limit and was constrained in the case of beech, therefore, causing the WPG of beech to maintain a relatively low value.

SEM results of the esterified wood are shown in Fig. 2. Compared to the untreated cypress, Cy_untreated, Cy_pr exhibited an increased cell wall thickness (Fig. 2a and Fig. 2b). The average cell wall thickness of Cy_Pr_4 was 4.60 μm, whereas that of Cy_untreated was 1.71 μm. In Cy_untreated, the rectangular cell walls were arranged in an orderly manner. In Cy_Pr, the distortion of the cell walls into pentagons, hexagons, or ellipses was observed.

Similar results were observed in both beech and cypress, and it was confirmed that the average wall thickness of the untreated beech, Be_untreated, was 3.63 μm, but it swelled to 6.82 μm owing to propionylation. Additionally, in Be_Pr_4, it was observed that the cell lumen diminished because of the swelling of the cell wall, and almost no voids were present. This disappearance of the cell lumen likely suppressed further swelling of the cell wall. This may have been one of the reasons why the WPG of beech remained lower than that of the cypress.

FT–IR Measurements

The esterified wood was subjected to FT-IR measurements to confirm the progress of the reaction (Fig. 3). To observe the interior of the esterified wood, ATR measurements were carried out using the pressed samples that were obtained in the compression test. In the data from the analysis (Fig. 3), sharp absorption bands appeared at 1226 cm^-1 to 1236 cm^-1 ( C-O and C=O), 1369 cm^-1 ( C-H), and 1730 cm^-1 to 1737 cm^-1 ( C=O), and a set of these signals is characteristic of the ester group (Mohebby 2008). The absorption at 3500 cm^-1 to 3300 cm^-1 was ascribed to the OH stretching of water adsorbed by the wood cell wall (Pandey 1999; Mohebby 2008). The esterification of the wood remarkably reduced the intensity of this band, which indicated that the hydroxyl groups were replaced by the ester groups.

For the acetylation reaction, the peak intensity of the hydroxyl group decreased remarkably after 1 h of treatment (Cy_Ac_1 in Fig. 3a and Be_Ac_1 in Fig. 3c) and almost completely disappeared after 4 h of treatment (Cy_Ac_4 in Fig. 3a and Be_Ac_4 in Fig. 3c).

For the propionylation treatment, the decrease in the peak intensity of the hydroxyl group was milder than that of the acetylation reaction, and a certain amount of the hydroxyl groups remained in the 1 h-treated wood. However, the peak of the hydroxyl group of the propionylated wood, which was treated for 4 h, was remarkably reduced, which suggested that most of the hydroxyl groups were replaced by the propionyl group. These results showed that the esterification proceeded sufficiently, owing to the hydroxyl groups present in the block-shaped wood piece. However, even with the use of the pressed test piece in this experiment, it may not have been possible to identify the hydroxyl groups contained in the highly crystallized cellulose. Therefore, the above results did not indicate that all the hydroxyl groups present in the wood were completely esterified.