Wood material

A systematic recording of the wood properties from different locations is not possible because of the control measures of the tree of heaven in Austrian forests (Brandner and Schickhofer 2010, 2013). Therefore, individual, noncontrolled trees from forested areas were used as test material, which were assumed to be more representative than trees from combated stocks with tree of heaven. In cooperation with Österreichische Bundesforste, a total of eight trees (A. altissima (Mill.) Swingle), with an average breast height diameter of 30 cm (27 to 69 cm), was felled from a forest area in Donaudorf, Austria. The trees were cut into logs of a length of 3 m each. The first two logs of the eight trees were used for the study (i.e., N = 16 logs). The logs were processed into sawn timber (thickness of the boards in green condition 1.5 inches = 38 mm). The small sawmill (Preisberger, Gföhl, Austria) used a live (through and through) sawing pattern. A total of 1.45 m³ of sawn timber was produced. The boards were largely knot free. The remaining tree material was processed into wood chips using an industrial standard chipper. The material was subsequently used for paper, fiber, and chipboard production.

Wood chip drying

Technical drying of 40 kg of wood chips took place in a Weiss drying chamber (Weiss Klimatechnik GmbH, Germany). After 7 days of storage in the climate chamber, a wood MC of approximately 4 percent was achieved. The material for the refiner tests was stored at −19.7°C until 24 hours before milling.

Sawn timber drying

The board material was divided into three stacks of the same size and fed to three different drying processes, air drying and kiln drying without and with steaming. The material before drying had an average MC of approximately 65 percent (range 40% to 85%). Investigation of drying stresses after both drying experiments was carried out according to ÖNORM ONR EN/TS 14464 (2010).

Material characterization

The sawn timber obtained from both controlled drying experiments was used for the production of samples according to the various standards and publications. The type of test, standard, or publication describing the specimen geometry and number of samples by test series are summarized in Tables 1 and 2. All mechanical tests were performed using a universal testing machine (Zwick/Roell Z100 or Zwick/Roell Z20). A mechanical extensometer (makroXtens; Zwick/Roell, Ulm, Germany) was used to measure the specimen deformation in the DIN 52188 (1979) tensile and DIN 52185 (1976) compression tests. In the bending tests according to DIN 52186 (1978), the MOE was measured in addition to the bending strength using the same mechanical extensometer, whereby the deflection in the middle of the specimen was determined with an accuracy of ±1.5 μm. The speed of the crosshead for all mechanical experiments was selected in accordance with the cited standards. Impact bending was tested with a pendulum impact tester (Otto Wolpert-Werke, Germany) corresponding to DIN 52189-1 (1981).

Table 1.Overview of the tests, sample sizes, and references (standards) according to which the various tests were performed, as well as the number (N) of samples per test series for tree of heaven (Ailanthus altissima).

View Fullscreen

Table 2.Parameters for the production of laboratory particle- and fiberboard. Degree of gluing refers to the dry mass of the wood, the proportion of hardener to the mass of the solid resin.

View Fullscreen

The tensile (t), compressive (c), bending (m), and shear (ν) strengths (_f) were determined at an equilibrium MC of approximately 12 percent after conditioning in a climate chamber at 20°C and 65 percent RH for at least 60 days. The indices (_L) and (_T) describe the load direction in and perpendicular to the grain, respectively. The MOEs determined on compression (E_c) and tensile (E_t) specimens were calculated for the center of the specimen after loading in accordance with the standards DIN 52188 (1979) and DIN 52185 (1976) by Equation 1: (1) $$E=\frac{\text{Δσ}}{\left(\frac{\text{Δ}l}{l}\right)}$$where Δσ is the difference between the stresses of the sample at 10 and 40 percent of the maximum stress at failure. Δl corresponds to the elongation of the sample within the measuring base of length l at the two load levels.

For the bending specimens, the MOE (E_m) was determined according to DIN 52186 (1978) using Equation 2: (2) $$E=\frac{\text{Δ}Fl3}{48\text{ Δδ }I}$$

ΔF corresponds to the difference between the load levels of the specimen at 10 and 40 percent of the maximum load level at break, l is the span length of the specimen, Δδ is the difference in deflection at the corresponding load levels 10 and 40 percent, and I is the moment of inertia of the specimen.

To investigate the suitability of tree of heaven for flooring applications, not only the strength properties but also the hardness (ÖNORM EN 1534 2010) were examined and compared with 23 samples each of ash and oak.

To determine the equilibrium MC of the samples after storage in the standard climate chamber, density samples were used. Before the mechanical tests and before measuring the kiln-dry density and MC corresponding to DIN 52182 (1975) and DIN 52183 (1977), all samples were stored for at least 60 days in a climate chamber (65% RH and 20°C).

For determining tensile shear strength (τ) corresponding to DIN EN 302 (2013) in dry (A1) and wet conditions (A5) of the material, a one-component polyurethane (PUR; Purbond HBS309, Henkel, Switzerland), urea–formaldehyde (UF; Aerolite UP4535, Synthesa, Austria), melamine–urea–formaldehyde (MUF; Metadynea, Krems, Austria), phenol–resorcinol–formaldehyde (PRF; Aerodux 185, Friebe Luftfahrt-Bedarf, Germany), and emulsion polymer isocyanate (EPI; Jowat, Switzerland) were used. All the resins examined are approved for load-bearing structures and are also used in other studies (e.g., Bockel et al. 2019), making it easy to compare the results with those of other hardwood species. The adhesive spreading quantity was chosen according to the technical data sheets and was between 150 and 220 g/m². The samples were pressed at room temperature at about 0.8 MPa for 2 hours. The same number of samples with the abovementioned adhesives was also produced for European beech wood (F. sylvatica), tested and compared with the results for tree of heaven.

All experiments performed to determine the mechanical properties and wood MC and density are summarized in Table 1, with a list of the corresponding standards.

Pulping

The pulp tests were conducted in the Mondi laboratory (Mondi Frantschach, St. Gertraud, Austria). All pulp and paper properties were gained by using following standards: (1) sheet formation (ISO 2004), (2) pulping pulp freeness index (PFI) and buffer used (ISO 2002), (3) degree of grinding (ISO 1999), (4) sheet weight (ISO 2012), (5) air resistance (ISO 2003), (6) strength properties (ISO 2005), (6) bursting strength (ISO 2001), and (7) whiteness (ISO 2004). To obtain a basis for comparison with the tests carried out on tree of heaven, measurements were also taken on Norway spruce (Picea abies). Comparative investigations of raw material resources with regard to their pulp properties are only meaningful if they are carried out at similar kappa numbers (i.e., residual lignin contents). A total of 15 tries had to be performed to reach a comparable kappa range for tree of heaven and Norway spruce.

A so-called PFI laboratory refiner was used for the pulping. The cooking conditions were as follows: (1) tree of heaven: dry content 93.2 percent, alkali content 14.5 to 16.5 percent, temperature 165 to 170°C, pressure 6 to 6.3 bar, residual alkali 0 to 1.3; (2) Norway spruce: dry content 90 percent, alkali content 22 to 25 percent, temperature 165 to 170°C, pressure 6 to 6.3 bar, residual alkali 10.6 to 18.5 percent.

Material was obtained from the pulp produced at different rotation speeds (2,000 to 9,000 revolutions per minute [rpm]) and laboratory sheets were made from it. On the basis of these laboratory sheets, the basis weight (g/m²), air resistance according to Gurley (s per liter [m² s]⁻¹), tensile strength index (Nm/g), elongation at break (%), elongation stiffness (kN/m), specific tear propagation resistance (kPam²/g), specific tensile fracture work (J/g), and the degree of whiteness according to ISO (%) were determined. The fiber morphology was also determined.

Fiber- and particleboard

Norway spruce is the standard material used by the Central European timber industry and is well researched. A comparison with this wood species with respect to particle- and fiberboard provides important practical information for the possible future use of the tree of heaven in Europe. For the production of particleboard, 20 kg of tree of heaven chips and Norway spruce chips from a sawmill (Binderholz GmbH, Fügen, Austria) were processed into particles (Gebr. Jehmlich GmbH, Nossen, Germany) using a cutting mill with a 2-mm sieve insert. Before further processing, the particles were dried to a wood MC of 3 percent using a self-made tunnel dryer.

The fibers for the fiberboards were produced using laboratory refiners. The 12-inch refiner (connected load = 45 kW, motor speed 3,200 rpm) was equipped with a rotating disc and a fixed disc. Unidirectional grinding discs with three rows of bars spaced at 7, 4, and 1 mm (Andritz AG, Graz, Austria) were used for refining experiments. The grinding disc spacing was 0.25 mm and the screw conveyor was operated at minimum speed (∼5 rpm). Before milling, the wood chips of tree of heaven and Norway spruce were cooked in an integrated pressure boiler at 165°C (heating rate 10°C/min). When the set temperature was reached, the screw conveyor was activated and the blow valve opened. During milling, the boiler temperature dropped to 130°C. The power consumption was measured during milling. The fibers were dried to a wood MC of 3 percent, analogous to the particle material.

For the production of all boards, a commercial E 0.5 MUF resin (Metadynea Austria GmbH) was used. This product is widely used in the wood industry and in applied research. Gluing was done in a standard laboratory gluing drum by spraying in the glue. During the production of the laboratory panels, the hydrophobic treatment customary in the industry was dispensed with to enable a better differentiation between the two types of wood regarding the thickness swelling. The key data for production of the particle- and fiberboard are summarized in Table 2. The same amounts of adhesive and production processes were used for a comparison of the particleboard and fiberboard produced.

Thickness swelling (N = 20), internal bond strength (N = 20), and bending strength (N = 9) of the boards were tested according to European standards (DIN EN 317 1993, ÖNORM EN 319, 310 2005).

Natural durability

Tree of heaven can be used as a substitute for European ash in solid-wood applications. Ash is also frequently used in outdoor applications (e.g., decking). Consequently, it is recommended to examine the natural durability to assess the technical potential of the tree of heaven. To determine the natural durability of this wood species in ground contact, prismatic specimens (500 by 50 by 25 mm) were cut from dried and conditioned boards and tested following the field test method in ÖNORM (2014). The material was stored in a climatic chamber (20°C and 65% RH) until equilibrium MC was reached. The boards were then buried in the soil on a suitable test site in the Vienna area (48°20′N, 16°3′E, 180 m above sea level). With an annual average precipitation of >700 mm at the site, there is sufficient moisture in the soil to promote the activity of fungi and other wood-destroying microorganisms. This method is therefore well suited to determining the natural durability in practice. Every 6 months, samples were taken from the soil and visually assessed for fungal and insect degradation.

Other samples were exposed to natural weathering according to ÖNORM EN 927-3 (2020) and ÖNORM EN ISO 2810 (2020). For this purpose, samples 300 mm in length, 100 mm in width, and 24 mm in thickness were mounted outdoors at the same site in a southerly direction at an angle of 45°. During the 3-year weathering period, the samples were regularly visually assessed and the wood color was determined by a spectrophotometer (Phyma CODEC 400) at wavelengths ranging from 400 to 700 nm. The light source was defined by the standard illuminant D65 with an observer of 2° (Teischinger et al. 2012, Meints et al. 2017). The diameter of the field of view was 20 mm, and the analysis of the color data was expressed according to the CIE L*a*b* color space (ÖNORM EN ISO/CIE 11664-4 2020).

The following meteorological data were recorded during the weathering period from 23 June 2015 to 28 August 2019 (1,532 days ≈ 4 yr): average daily temperature 10.5° (+36 to –17.2°C); average daily precipitation 1.5 mm (summed over the entire duration of weathering, 2,300 mm); average daily global radiation 115 W/m² (summed over the entire duration of weathering, 175,375 W/m²); average daily wind speed 5.7 km/h (maximum 95 km/h). The meteorological data were taken from the weather station, which is positioned directly at the location of the samples.

Solid-wood properties

Starting at first with some more general data, the assumption, indicated by some literature sources (e.g., Hu 1979), that tree of heaven is subject to widespread use in its homeland China could not be confirmed. As part of this study, a native Chinese speaker, using Chinese keywords, also conducted an intensive internet search for wood technology information on tree of heaven but found no significant new results. This is because the roundwood supply in China comes mainly from plantation cultivation, poplar species in the north, and Southern blue gum (Eucalyptus globulus) mainly in the south. In total, only four companies were found to use tree of heaven for wood panel production. Fourteen companies used it in flooring production. However, the most widespread application in China seems to be in furniture production. Apart from this, the decorative, visually appealing wood of tree of heaven differs only slightly from native European ash (Berki 2014). A large amount of tree of heaven wood was processed during the study. The joinery work and the assessment of the workability of tree of heaven were carried out by a master joiner who has sufficient experience in working and processing hardwood species. The good machinability was in line with much-cited studies (Hölbling 1989, Panayotov et al. 2011, Brandner and Schickhofer 2010, 2013) and considered uncomplicated and comparable with European ash. Only boards with higher internal stresses (see drying properties) showed certain limitations during sawing and planning owing to stronger warping. However, it possible that these higher internal stresses originate from the lack of forest management.

Table 3 summarizes the values of the tests carried out on solid wood. Regarding the bending, compression, and tensile tests, the results were in good agreement with the values published by Brandner and Schickhofer (2010, 2013). The bending strength (m_f) of 100 ± 9 N/mm² was very similar to that of the samples examined by Hölbling (1989). The tensile strength along the grain (t_fL) showed a high range of variation, whereas the bending (m_f) and compressive strengths along the grain (c_fL) showed a much smaller variation. Except for the values published by Panayotov et al. (2011), the compressive strengths obtained showed good agreement with Hölbling (1989) and Brandner and Schickhofer (2010, 2013).

Table 3.Summary of the test results corresponding to Table 1 (tests performed), showing mean values (mean), standard deviation (SD), minimum (min.), and maximum (max.) values.

View Fullscreen

Regarding the investigated samples of tree of heaven, a MC of 12.7 ± 0.9 percent was observed after standard climate storage. At this MC, a density of 611 ± 50 kg/m³ (SD) was determined. The density values were thus slightly below the values of earlier studies reported by Hölbling (1989), Panayotov et al. (2011), and Brandner and Schickhofer (2010, 2013).

The determination of the MOEs from bending (E_m), compression, and tensile tests provided slightly higher values for the tensile tests. The observed MOEs were relatively uniform at 12,000 to 13,300 N/mm² and showed high agreement with the values published by Hölbling (1989) and Panayotov et al. (2011). Compared with the investigations by Brandner and Schickhofer (2010, 2013), there was a high level of agreement with the tensile MOEs, but not for the MOEs from the compression tests, which were significantly lower than in this study. The reason could possibly lie in a higher measurement scatter of samples that were not optimally axially oriented. For perfectly compressed specimens, no significant difference between the MOE in compression and bending should be observed (Kollmann 1951). The deviations between the compression MOEs between the two studies should therefore be put into perspective. However, the significantly higher MOEs reported by Brandner and Schickhofer (2010, 2013) in the bending test of 14,590 N/mm² compared with their own measurements of 11,590 N/mm² cannot be explained by the different test methods (three-point vs. four-point bending test) alone. The differences may be due to different laboratory standards, lower density, or other natural scatter of the sample material. Overall, the investigations conducted in the present study confirmed the previous studies.

Especially for load-bearing structures such as glued laminated timber, the shear strengths and the strengths transverse to the grain represent a limiting parameter (Liu et al. 2016). For this reason, these two material properties were analyzed as well. Table 3 shows the block shear strength according to ASTM 143 D (ASTM 1992) and the transverse compressive strength according to DIN 52192 (1979). Regarding the shear strength (ν; 12.8 N/mm²), slightly higher values were observed than in the study by Brandner and Schickhofer (2010, 2013; 11.5 N/mm²), who determined the shear strength using DIN 52 187 (1979). The small differences can be explained on the one hand by the natural variability of the investigated sample material and on the other hand by the different measurement methods (DIN 52187 1979 vs. ASTM 143D 1992; Stretenovic et al. 2004).

With a mean Brinell hardness of 28.1 ± 3.4 N/mm², the values of tree of heaven were significantly lower than those of European ash (37.7 ± 6.7 N/mm²) and oak (35.0 ± 8.4 N/mm²). Tree of heaven is therefore of limited suitability for highly stressed floors in the commercial sector. However, from a technological point of view, the similar optical, technological, and mechanical–physical properties would allow tree of heaven to be used as a substitute for European ash.

For the bending specimens and the specimens for Brinell hardness, the density was determined in addition to the mechanical properties. The coefficient of determination for the bending strength vs. density was R² = 0.39 and for the hardness vs. density R² = 0.48. The correlation between the mechanical properties and the density was thus within the range of comparable studies (Verkasalo and Liban 2002).

Wood drying and effects on wood processing

After completion of the two kiln-drying processes, the final wood MC, new cracking and tension in the wood, change of shape and color, and machinability were determined. To assess the workability of dried or steamed tree of heaven wood, several boards were cut on a circular saw and planed with a jointer. An experienced carpenter carried out the assessment using practical evaluation criteria.

Gluing

Solid-wood applications, regardless of whether they are used for load-bearing or nonload-bearing purposes, require not only optically appealing properties and advantageous mechanical/physical properties but also good bondability. Experience has shown that European ash is rather difficult to bond with conventional adhesives (Knorz et al. 2014). Better bonding properties than those of European ash would therefore be an advantage for tree of heaven as a substitute wood species. In the present tests, the adhesive properties according to DIN EN 302 (2013) were investigated for tree of heaven and European beech. In addition to testing the dry samples (A1), tests of the tensile shear strength after water storage (A5) were also conducted (6-h storage in water at 100°C, 2-h storage in water at 20°C, conditioning of the samples in standard climate until equilibrium MC was reached). No A5 tests were carried out for the nonwaterproof UF adhesive. Overall, tree of heaven showed very good bonding properties. Most of the values determined did not differ from beech (see Figs. 1a and 1b). In the dry state, only the samples bonded with EPI showed lower shear tensile strengths. However, the tests did not show any significant differences compared with European beech (one-way analysis of variance, P = 0.05). Significantly higher shear strengths (ν) were only found for European beech when bonded with PRF. Overall, the tensile shear strength was in the range of the determined block shear strength (ν) of 12.75 N/mm². Owing to the high percentage of wood failure in tree of heaven (EPI 80%, PRF 90%, PUR 75%, UF 80%, MUF 100%), it can be assumed that the strength of the wood is the limiting factor of the determined tensile shear strengths. Overall, the tests showed that the tree of heaven has good bondability. The results are in good agreement with the results reported by Brandner and Schickhofer (2013), although these were determined on the basis of ÖNORM EN 392 (1995).

View Figure

• Download as Powerpoint
• Download Figure

A tensile shear strength corresponding to that of beech was also demonstrated for tree of heaven in the dry state after wet storage and reconditioning (A5; see Fig. 1b). As the standard tests for adhesives for load-bearing purposes in accordance with DIN EN 302-1 (2013) for tensile shear strength are carried out using beech wood, the tests with tree of heaven demonstrate excellent bonding properties. For all adhesives, tensile shear strengths in the range of the level of beech were found. Only for PUR were significantly lower values determined for tree of heaven. It can be concluded that the technological use of tree of heaven solid wood is not restricted by bondability.

Natural durability

Natural durability without soil contact.—

Before the weathering period, the initial color of the wood was determined, and the following values were found: lightness of the color L* = 81.0; red–green axis value a* = 3.6; yellow–blue axis value b* = 21.6; chroma C* = 22.0; hue angle h° = 80.6°. The color was identified in three measurement spots per sample, and a mean value was calculated. The same spots were used for all measurement points within the whole weathering period. The color measurements of the individual samples were determined at intervals of approximately 1 month. This means that 50 color measurements (including the determination of the initial color) were taken during the 4-year weathering period. After only 1 year, all samples had turned gray and showed only slight seasonal fluctuations in color. The color measurement showed the following values at the end of the experiment (after 1,532 days): L* = 48.8; a* = 1.3; b* = 2.6; chroma C* = 2.9; hue angle h° = 63.3. The wood color showed a color distance (ΔE*) of 37.4, became darker (ΔL* = −32.2), and lost much of its color intensity (ΔC* −19.0).

The specimens showed crack formation on the faces and on the surface (maximum crack depth 2 mm). The surface hardness was slightly reduced on the weathered surface, whereas the unweathered side showed no change. No wood-destroying organisms (e.g., fungi or insects) were found. A good indicator of the presence of wood-damaging organisms is to check the impact bending strength, according to DIN 52189-1 (1981). For this purpose, appropriate test specimens were cut out of the weathered samples and stored at 20°C and 65 percent RH for 30 days before the tests were conducted. Table 4 shows the comparison of unweathered (data from Table 3) and weathered (1,532 days) samples; no significant difference in strength could be determined.

Table 4.Impact bending test on unaffected tree of heaven (Ailanthus altissima) specimens (values from Table 3) and after 1,532 days of weathering, showing number of samples (N), mean values (mean), standard deviation (SD), and minimum (min.) and maximum (max.) values.

View Fullscreen

Pulping and paper properties

Thanks to the high fiber content and the long fibers of tree of heaven, good paper properties have been postulated (Ferreira et al. 2013; Baptista et al. 2014). Nevertheless, the suitability of tree of heaven for pulp and paper production has not yet been sufficiently investigated (Adamik and Brauns 1957, Isenberg 1981).

The results are presented here as mean values for the kappa numbers 19.5 and 20, respectively. Norway spruce: fiber length 2.62 mm, fines content 3.37 percent, fiber width 28.1 μm, and fiber wall thickness 10 μm; tree of heaven: fiber length 0.98 mm, fines content 8.96 percent, fiber width 21.5 μm, and fiber wall thickness 8.9 μm. Tree of heaven thus differs primarily by shorter fibers and by a higher fines content.

In addition, tree of heaven is characterized by a higher degree of grinding at grindings up to 9,000 rpm and thus a higher energy input during raw material pulping. At grindings up to 2,000 rpm, tree of heaven provided similar paper properties in terms of tensile strength and specific bursting resistance to Norway spruce. Overall, the air resistance increased more steeply with tree of heaven than with Norway spruce, which can also be attributed to the higher fines content. Similar results were also observed for the higher kappa number 29.6, where slightly less favorable paper properties were found for tree of heaven than for Norway spruce. Overall, the trial yielded relatively good pulp and paper properties for tree of heaven. However, relatively high freeness is required for this. High freeness means higher energy consumption, longer dewatering times, and thus lower production speeds (Mandlez 2017).

The results prove that tree of heaven can be considered a potentially interesting raw material for pulp and paper production. Whether the good pulp and paper properties justify the higher grinding degrees depends specifically on the respective product properties of the desired papers in production. No general statement about the trials is therefore possible, and it remains to be investigated whether the adaptation of the pulping conditions to tree of heaven does not lead to further improvements. The work of Demirbaş (1998) shows that longer cooking times, higher cooking temperatures, and adapted chemical conditions lead to better results.

Fiber- and particleboard production

In the milling processes with the laboratory refiner for fiberboard production, no differences were found in the measured energy consumption between tree of heaven and Norway spruce in pulp production. In the particleboards produced, there were no significant differences in thickness swelling (TS) for tree of heaven (t test, P = 0.05): Norway spruce TS = 23.1 ± 2.10 percent and tree of heaven TS = 23.0 ± 2.96 percent. On the other hand, the tests on the fiberboards showed a lower TS for tree of heaven (TS = 30.6%) than for Norway spruce (TS = 40.9%). The relatively high TS values can be explained by the absence of a hydrophobic agent in the laboratory production of the boards.

The measurement of the internal bond strength (IB) of the particleboards showed significantly higher values (P = 0.05) for tree of heaven (IB = 1.22 ± 0.35 N/mm²) than for Norway spruce (IB = 0.80 ± 0.18 N/mm²). Analogous to the investigations on the manufactured particleboards, significantly higher values (P = 0.05) were also found for the fiberboards for tree of heaven (IB = 0.60 ± 0.20 N/mm²) than for Norway spruce (IB = 0.19 ± 0.13 N/mm²) despite a relatively high scatter of the measured values.

The target density of 650 kg/m³ was achieved neither for tree of heaven (ρ = 608 ± 30 kg/m³) nor for Norway spruce (ρ = 597 ± 34 kg/m³). However, the statistical evaluation of the density values did not reveal any significant differences between the two wood species (t test, P = 0.05). The glue contents were also approximately the same for all panel types. The differences found in the board properties can therefore not be attributed to a variation in density or different glue contents.

Regarding the bending strength of the tested particleboards, it should be mentioned that only single-layer boards were produced in the laboratory. Industrially produced particleboards usually have a three-layer structure and have bending properties related to the needlelike geometry with a high aspect ratio of the face sheet material. The results presented here therefore allow only a comparison of the laboratory panels with each other. The bending strength of Norway spruce (σ_b = 7.37 ± 1.89 N/mm²) was slightly higher than that of tree of heaven (σ_b = 6.97 ± 1.59 N/mm²). Similarly, slightly higher values for the bending MOE were found for Norway spruce (MOE = 1,355 ± 274 N/mm²) than for tree of heaven (MOE = 1,115 ± 218 N/mm²). The slightly lower bending values of tree of heaven are probably due to the greater amount of cubic chip material. The Norway spruce fibers are comparatively long and it is assumed that they result in particles with a higher aspect ratio. However, a t test performed showed a significant difference only for the MOE (P = 0.05).

In contrast to the particleboards, no significant difference was found for the bending properties of the fiberboards made of Norway spruce (σ_b = 13.2 ± 0.12 N/mm²; MOE = 1,596 ± 297 N/mm²) and tree of heaven (σ_b = 13.9 ± 2.44 N/mm²; MOE = 1,525 ± 317 N/mm²; t test, P = 0.05).

Because of the increasing competition between material and energy use, alternative raw material sources are being sought for the production of wood-based materials. As a result, wood from short-rotation plantations is increasingly becoming the focus of the wood industry. Its fast growth and low requirements make tree of heaven a potential source of raw material for the industry. Its suitability for the production of fiberboard and particleboard has not been sufficiently investigated, even given the present results. However, the results confirm the study by Elbadawi et al. (2015) in which tree of heaven (Ailanthus excelsa Roxb.) was processed into particleboard. From the research conducted so far, it can be concluded that the tree of heaven species should be considered as a potential raw material resource for wood panel production. Whether these results can be directly transferred to industrially produced wood-based materials still needs to be investigated.