TECHNICAL USE OF RESIDUES FROM THE CELLULOSE INDUSTRY AND SUGARCANE BAGASSE IN PARTICLEBOARDS

The use of alternative materials such as lignocellulosic residues in the production of particleboards has increased considering that these residues are produced in large volumes and often do not have an appropriate destination. This research studied the use of residues from cellulose industries, sugarcane bagasse, and castor oil-based polyurethane resin in the production of wooden panels and evaluated the influence of using these residues on the physical and mechanical properties of the panels. The products were manufactured according to the Brazilian standard ABNT NBR 14810 and the requirements of the panels were evaluated based on national and international standards. All treatments partially met the regulatory requirements. The addition of bagasse led to an improvement in physical and mechanical properties, with treatment 2 (50% wood residue and 50% bagasse) presenting the best performance, which indicates the possibility of using panels with residues with non-structural purpose in environments to improve the thermoacoustic performance of rural buildings. The statistical analysis indicated that the percentage of bagasse was significant, improving the evaluated properties.


INTRODUCTION
The use of engineered wood-based products, such as particleboards, medium-density fiberboard (MDP), oriented strand boards (OSB), and plywood panels has increased in rural structures, being an alternative to the use of sawn wood in buildings for structural use and ambiance, improving thermal and acoustic comfort in buildings ( (Bertolini et al. 2019a;Bertolini et al. 2019b;Garzón-Barrero et al. 2016;Labans et al. 2017;Scatolino et al. 2017;Souza et al. 2014;Zhou & Pizzi 2014).
Brazil has a vast sugarcane production area, with annual cultivation of approximately 620 million tons (CONAB 2019), producing a large volume of bagasse, which is largely used for energy production by burning it (Hofsetz & Silva 2012). It is also a major producer of paper and cellulose, with an annual production of 10 and 18 million tons annually, respectively (Indústria Brasileira de Árvores -IBÁ 2017). This industrial production process emits a large volume of residues, such as bark and branches of Eucalyptus, which are mostly used for energy production Engenharia Agrícola, Jaboticabal, v.41, n.1, p.107-111, jan./feb. 2021 by burning (66%) and destined to landfills (5%) (Indústria Brasileira de Árvores - IBÁ 2017).
An important component in the production of panels is the resin, as it influences the physical and mechanical properties of the engineered wood-based product, varying the adhesive weight and the resin chemical composition. Commercial resins such as urea-formaldehyde and phenolformaldehyde release formaldehyde gas throughout the production process, which is toxic to humans (Kusumah et al. 2017;Muttil et al. 2014;Pan et al. 2007;Zhou & Pizzi 2014). An alternative to the use of these adhesives is the castor oil-based bicomponent polyurethane resin, with oil of vegetable origin and without emission of toxic gases (Ferro et al. 2018;Fiorelli et al. 2019;Zau et al. 2014).
The use of particleboards for thermal and acoustic insulation is a possibility due to its porosity. The presence of air in the voids promotes thermal resistance and acoustic absorption, leading to thermal and acoustic comfort in rural and urban structures (Bertolini et al. 2019a).
This research aimed to evaluate the physical and mechanical properties of particleboards made with residues collected from paper and cellulose industries from Eucalyptus urophylla, Eucalyptus grandis, and Eucalyptus camaldulensis, sugarcane bagasse, and castor oil-based bicomponent polyurethane resin, as well as check the possibility of using these panels based on normative requirements.

MATERIAL AND METHODS
The panels were manufactured from debarking residues of logs of Eucalyptus urophylla, Eucalyptus grandis, and Eucalyptus camaldulensis (mixture of barks), collected in the pulp and paper industry Eldorado Brasil, Três Lagoas, Mato Grosso do Sul, Brazil ( Figure 1a). The material was sun-dried until the apparent moisture reached between 10 and 12% ( Figure 1b). After drying, the material was crushed to obtain particles with a size ranging from 2 to 6 mm. Sugarcane bagasse was collected at the Vale do Paraná S/A alcohol and sugar mill, being sun-dried until the apparent moisture reached between 10 and 12%. The material was also crushed to obtain a particle size between 4 and 10 mm (Sugahara et al. 2019). The resin consisted of a castor oil-based bicomponent polyurethane resin (PU), with polyol (1.1 g/cm 3 ), made from castor oil and polyfunctional isocyanate (1.24 g/cm 3 ) at a 1:1 ratio (polyol and isocyanate) (Ferro et al. 2014). The adhesive content used was 10% relative to the dry mass of particles (Sugahara et al. 2019).
The PU resin was mechanically homogenized with Eucalyptus particles and sugarcane bagasse. Subsequently, the mixture was taken to a mold to form the particle mattress for a pre-pressing at 0.015 MPa. Afterward, the panel was hot-pressed at 100 °C and 5 MPa pressure for 3 minutes. Decompression was performed for 30 seconds to eliminate gases that can cause voids in the panel and then a compression was applied for another 7 minutes (Sugahara et al. 2019). After that, the panel was squared and stored in an appropriate place.
Six panels with nominal dimensions 40 × 40 × 1 cm (1600 cm 3 ) and a nominal density of 0.8 g/cm 3 were produced for each treatment. Twelve specimens were extracted from each treatment in each property to evaluate physical (apparent density, thickness swelling, water absorption, and average moisture) and mechanical properties under static bending (modulus of strength and elasticity) and stress perpendicular to the faces. All properties were determined following the precepts of the Brazilian standard ABNT NBR 14810 (ABNT 2018). Table 1 shows the treatments performed in the present study. An analysis of variance (ANOVA) at a 5% significance level was performed using the software Minitab ® to investigate the influence of bagasse and industrial residue content on physical and mechanical properties.
The normality and homogeneity of the residual distribution were evaluated using the Anderson-Darling test to validate ANOVA (α = 5%) and the Tukey test (α = 5%). For the test formulation, a p-value equal to or higher than 0.05 implies that the sample distribution is normal and the variance between treatments is homogeneous, which validates the ANOVA model. According to the Tukey test, A denotes the treatment associated with the highest mean value, B the second highest mean value, and so on, and equal letters imply treatments with statistically equivalent means.  Tables 2 and 3 show the physical (water absorption -WA, thickness swelling -TS, apparent density -AD, and moisture content -MC) and mechanical properties (modulus of strength -MOS, modulus of elasticity -MOE, and perpendicular stress -PS), mean values, coefficient of variation (CV), and the result of the Tukey test, considering the percentage of addition of sugarcane bagasse.  The apparent density values ranged from 877 and 934 kg/m 3 , similar to the values found by Sugahara et al. (2019) using Eucalyptus particles (60%) and sugarcane bagasse (40%) (880 kg/m 3 PU). The moisture values of all treatments met the Brazilian normative requirement (ABNT 2018), which recommends that moisture should be between 5 and 13%.

RESULTS AND DISCUSSION
Only one study dealing with the use of industrial residues from tropical wood and sugarcane bagasse to produce particle boards and polyurethane resin can be found in the literature. Yano et al. (2020) evaluated the use of industrial residues from the woods of Cariniana micrantha (Tauari), Goupia glabra (Cupiúba), Vochysia guianensis (Cambará), Tabebuia alba (Ipê), and Apuleia leiocarpa (Garapa), sugarcane bagasse, and castor oil-based PU resin for the manufacture of panels. Five treatments were proposed with varying contents of sugarcane bagasse and industrial residues, and the adhesive content was 10% relative to the dry mass of particles. The treatment with the best performance had 50% bagasse and 50% industrial residue, with MOS equal to 11.09 MPa, MOE equal to 2034 MPa, and apparent density equal to 0.71 g/cm 3 , which are values similar to those obtained in this research. Table 4 shows the normative requirements for particleboards.

Standard Thickness (mm) AD (kg/m³) MOS (MPa) MOE (MPa) PS (MPa) TS 24h (%)
NBR 14810 (2018)   In general, the addition of sugarcane bagasse improved physical and mechanical properties. In this sense, Treatment 2 presented the highest dimensional stability, with the lowest values for WA and TS compared to the other treatments. Moreover, this treatment presented good performance for PS and MOE although the MOS value was lower than the other treatments. Further research is required to define the optimum bagasse content in the panels.
The results of this research indicate the possibility of using industrial residues from paper and cellulose companies and sugarcane bagasse in commercial panels with a non-structural function aiming at the thermoacoustic insulation of rural buildings.

CONCLUSIONS
The results of this research allow us to conclude that: 1. The treatments partially met the normative requirements, with Treatment 2 being the best fit between residues for panel production, with values similar to those found in the literature.
2. The incorporation of sugarcane bagasse promoted improvements in physical and mechanical properties, indicating the possibility of using industrial residues from paper and cellulose companies and sugarcane bagasse in commercial panels with a non-structural function aiming at the thermoacoustic insulation of rural buildings.