Introduction

Products developed from wood derivatives are widely used in civil construction, among them wood-cement panels, since they present satisfactory mechanical performance, high durability, resistance to xylophages and weathering agents, reduced production cost, thermal properties appropriate for buildings, dimensional stability and sound insulation (Quiroga; Marzocchi; Rintoul, 2016; Çavdar; Yel; Torun, 2022). The replacement of materials of fossil origins by sustainable raw material allowed the diffusion of wood-cement panels (Wei; Zhou; Tomita, 2000; Hasan; Horvath; Alpar, 2021).

These panels are produced by wood, cement, water, and small amounts of chemical additives (Ashori; Tabarsa; Sepahvand, 2012; Rana et al., 2020). Wood is the reinforcement material and can be used in different sizes, from fibers to wood yarns (Yel, 2022). Among the main advantages of using wood in cement panels, the increase in mechanical strength stands out (Rana et al., 2020; Ferraz et al., 2012; Caprai et al., 2018).

Erythrina poeppigiana is a fast-growing alien species belonging to the family Leguminosae, being used in agroforestry systems in consortium with cocoa. However, the excessive shading provided by Erythrina poeppigiana resulted in low productivity of cocoa clones resistant to witches' broom (Marques; Monteiro, 2016). Aiming to guarantee the productivity of cocoa trees, the government of Bahia (Brazil) published an ordinance allowing the removal of trees that are not native to the Atlantic Forest, without the need for prior authorization (SMA; IMARH, 2019).

The stock of wood of this species in the Atlantic Forest of Brazil is estimated at 1,000,000 m³, which has no indication of use as solid wood, and the production of wood blades and panels is an alternative for its final destination, because studies of new species with rapid growth is extremely important to diversify the supply of raw material in the forest market (Sá et al., 2012), in addition to being an alternative to cushion pressure on slow-growing forests (Cabral; Nakanishi; Fiorelli, 2018).

The incompatibility of cement with wood is one of the challenges of the production of the panels, since there is interference in the hydration and curing of the cement. Extractives and impurities present in lignocellulosic materials can affect the balance of the cement hydration reaction (Boulos et al., 2017). More recent studies have demonstrated that the effect of extractives on reducing cement hydration is due to their absorption on the surface of the grains of hydrated cement products (Cabral; Nakanishi; Fiorelli, 2018).

However, pretreatments such as cold/hot water and NaOH extractions have potential for improvements in hydration (Hasan; Horvath; Alpar, 2021; Rana et al., 2020; Nasser et al., 2014).

The degradation of the lignocellulosic material over time, due to the alkalinity of the cement, is another problem to solve in the production of the panels. Carbonation is an alternative to reduce this alkalinity, being a natural phenomenon that affects cementitious materials (Almeida et al., 2013). This phenomenon consists of the reaction of the products of the hydration of the cement with the carbon dioxide (CO₂). Over time, cement absorbs CO₂ from the atmosphere and this process occurs during the life of the structure (Filomeno et al., 2020). In order to optimize the process, providing greater and faster carbonation to cementitious composites, climatic chambers can be used with application of CO₂ and with control of temperature and relative humidity of the air, a process called accelerated carbonation. In addition, accelerated carbonation can decrease porosity and water absorption and improve the mechanical performance of wood-cement panels (Tonoli et al., 2010). Cabral, Nakanishi and Fiorelli (2018) found that wood cement panels produced with sugarcane bagasse and subjected to accelerated carbonation and accelerated aging tests showed inferior physical properties and superior mechanical properties compared to the non-carbonated version.

Accelerated carbonation of cement-based materials is one of the effective methods for preparing construction materials and is therefore an innovative curing procedure that results in the sequestration of carbon dioxide gas (CO₂) and its conversion into stable products, reducing CO₂ emissions from the construction industry. Combined with the fact that Erythrina poeppigiana wood has a low calorific value, which makes it unsuitable even for firewood and charcoal, this work seeks to find an alternative use that adds value to this wood. Thus, the objective of the study was to evaluate the potential of Erythrina poeppigiana wood in the production of cement-wood panels with accelerated carbonation process.

Material and methods

Raw material

The Erythrina poeppigiana wood came from experimental plantations in the city of Ilheus, Bahia, Brazil. The 4 m long logs were where they sectioned into 58 cm long logs and then stored in a tank containing hot water for 24 h for subsequent lamination.

Sheets of 2 mm thickness were obtained in a rolling lathe and were processed in a hammer mill for generation of silver type particles. The particulate material was sieved through a set of overlapping sieves, whose openings were 12 mesh (upper) and 40 mesh (lower), respectively.

Particles pre-treatment

The wood particles were treated with hot water initially at 100 °C. For this treatment it was used water: particle ratio of 15:1. After the hot water treatment, the particles were washed in running water and dried in an oven at 60 °C for 72 h, until reaching a relative humidity of 8%.

Characterization of the wood

The basic density of the wood was determined according to a methodology adapted from the NBR 11941 standard (ABNT, 2003). The wood was chemically characterized according to the levels of Extractives: NBR 14853 (ABNT, 2010a), Lignin: NBR 7989 (ABNT, 2010b), Holocellulose (cellulose + hemicelluloses): (Browning, 1963); Cellulose: (Kennedy; Phillips; Williams, 1987). The hemicellulose content was quantified by the difference between the holocellulose and cellulose contents; and Ashes: NBR 13999 (ABNT, 2017). Mean values for all analysis were obtained in triplicate.

Characterization of the wood

To produce the panels, Portland cement CPV-ARI (Portland Cement of High Initial Strength) was used as binder. The chemical additive used was calcium chloride (CaCl₂), to reduce the total setting and hardening time of the mixture. Hot water treated and in nature wood particles were used. The materials for panel production were homogenized in a mixer, following the methodology applied by Souza et al. (2021). The variables used for the production of the panels are shown in Table 1.

After homogenization, the resulting mass of the mixture for each panel was properly weighed, separated and randomly distributed in aluminum plates greased with low viscosity oil to facilitate the removal of the panel after its pressing and stapling, and with the aid of iron rims with dimensions of 48 x 48 x 1.5 cm to control the dimensions and thickness of the panels.

The panels were arranged in a cold press, loaded with three panels and the apparatus for stapling, to reach a pressure of 4 MPa. The panels remained stapled for 24 h, after the removal of the staples the panels were conditioned in an acclimatized chamber with a temperature of 20 ± 2 °C and a relative humidity of 65 ± 3%, for a period of 28 days.

Accelerated carbonation

After 3 days of healing, panels produced with particles treated with hot water and in nature, were destined to the accelerated carbonation process (Figure 1). Samples taken using a circular saw were subjected to accelerated carbonation according to the methodology proposed by Arantes et al. (2023). For this, a vertical autoclave was used connected by a hose to an industrial grade CO₂ cylinder with a purity of 99%, maintaining a constant pressure of 0.75 kgf/cm². The temperature (T), relative humidity (RH), and CO₂ concentration [CO₂] in the chamber used for the accelerated carbonation test were set at 25 ± 2 °C, 70 ± 5%, and 20 ± 3%, respectively.

The samples remained in this system for 10 hours at a constant pressure of 0.75 kgf/cm², and were then removed and placed in a climate-controlled chamber with a temperature of 20 ± 2 °C and relative humidity of 65 ± 3%, remaining in this condition for 25 days for curing to occur.

Table 2 shows the experimental plan that was used in this study. The control treatment represents the panels that did not have the particles treated with hot water and were not subjected to the accelerated carbonation process.

Evaluation of the physical and mechanical properties of the panels

To determine the physical and mechanical properties, the panels were tested after reaching 28 days of curing. Table 3 describes the tests and standards used in this work to evaluate the panels. Mechanical tests were carried out on a Universal Testing Machine (AROTEC, model WDW-20E) with a 2 ton load cell.

It was also determined the compaction ratio of the produced panels, obtained by the bulk density of the panels over the basic density of the wood.

Microstructural analysis

The fractured wood cement panels were visualized in a Leica DM4000B (LM) compound optical light microscope coupled with a Moticam X - Motic Europe CMOS digital camera with 100x magnification. The fractured region and the surface of the panels were evaluated, in order to evaluate the surface porosity of the panels and the particle-matrix interaction.

Fourier Transform Infrared Spectroscopy (FTIR)

Fourier transform infrared spectroscopy was performed to verify the efficiency of accelerated carbonation through the analysis of calcium carbonate (CaCO₃) peaks. Analyses of the panels were performed on a Varian 600-IR Fourier transform (FTIR) FT-IR spectrometer, with GladiATR accessory from Pike Technologies coupled for measurements by total attenuated reflectance (ATR) at 45° with a zinc selenide crystal. The spectral range analyzed was 450 to 4.000 cm^-1, 2 cm^-1 resolution and 32 scans.

Analysis by thermogravimetry

The thermal analysis of the panels was performed using a TA Instruments (Delaware, USA) TGA Q500 thermal analyzer, to verify the influence of hot water treatment and accelerated carbonation on the thermal stability of the panels. Samples of 10 mg of the ground panels were added to the crucible of the equipment and heated from 25 to 800 °C. The established analysis condition was in synthetic air atmosphere (80% N₂ and 20% O₂) flowing at 50 mL/min and heating rate of 10 °C/min (Raabe et al., 2015).

Analysis of results

The treatments were analyzed according to experimental design entirely randomized. To verify the influence of hot water treatment and accelerated carbonation on the physical and mechanical properties of the panels, the means of the treatments were compared using the Tukey test at 5% significance level. The statistical analyses were performed using the SISVAR statistical program.

Results and discussion

Wood characterization

The basic density found for Erythrina poeppigiana (0.262 g/cm³) (Table 4) was lower compared to those found in the literature for Pinus spp. and Shizolobium amazonicum (Amaral; Ferreira; Couto, 1977; Mendes et al., 1999; Silva et al., 2016; Lobão et al., 2012) which are commercial woods used for veneer and wood panel production. Matos et al. (2019) find for Pinus oocarpa and Shizolobium amazonicum woods a basic density of 0.530 and 0.310 g/cm³, respectively. According to Iwakiri (2005) the density of wood for the production of wood-cement panels should be medium to low to ensure the compaction ratio of the panel within adequate levels for densification and consolidation of the material.

The treatment with hot water provided different contents of total extractives, lignin and ash. The most important chemical component for the present study is the quantification of the extractives content, since the presence in large amounts of extractives can affect the cement curing and consequently affect the interaction between the wood particle and the cementitious matrix (Almeida et al., 2013; Iwakiri et al., 2012; Ferreira et al., 2012). In the study by Borges et al. (2022) the value of extractives found was 13.69% for soybean residues. Souza et al. (2021), a high amount of extractives in coconut fibers (19.8%) was observed, being the main responsible for hindering and delaying the cement curing process, reducing the mechanical resistance of the panels. However, the treatment with hot water was efficient in reducing the content of total extractives, which is desirable for the production of this type of panel.

Physical properties of the panels

In the apparent density of the panels there were no statistical differences between the treatments (Figure 2a). The values showed a low standard deviation, which demonstrates homogeneity of production within each situation evaluated. The treatments presented bulk density below the nominal density (1.25 g/cm³), due to the losses that occur during the stages of the panel formation process and the return in thickness after the pressing.

The compaction ratio showed no variation between treatments (Figure 2a). According to Scatolino et al. (2017), the low density of wood is one of the key factors for the use of lignocellulosic materials in the production of reconstituted panels with this a higher compaction ratio can result in better mechanical properties.

There was a reduction in apparent porosity for treatments T2 and T4 (Figure 2b). The hot water treatment increased the interaction of the particles with the cement, consequently reducing the porosity of the panels. Accelerated carbonation provided lower apparent porosity in treatment T4 when compared to the panels evaluated under the different conditions. In their studies, Tonoli et al. (2010) and Soroushian, Won and Hassan (2012) verified a decrease in porosity in carbonated cementitious composites, corroborating the present study.

Figure 3a shows that the surface of treatment T1 is more porous in relation to the other conditions evaluated. Larger pores are observed for treatment T3 (Figure 3c). On the other hand, Figures 3b and 3d show less porous surfaces for treatments T2 and T4.

No influence of the hot water treatment was verified on the WA2h and WA24h properties of the T2 panels (Figure 4a), as the values were statistically equal to that of the control (T1). Iwakiri et al. (2015) also found no influence of hot water treatment for 6h on WA24h in paricá wood-cement panels. Iwakiri et al. (2015) did not find statistical differences in WA2h and WA24h values for Eucalyptus benthamii wood cement panels produced with in nature and hot water treated particles for 6h.

The panels that underwent accelerated carbonation (treatments T3 and T4) showed lower values for WA2h and WA24h properties compared to the treatments without carbonation (T1 and T2). The accelerated carbonation enhances the formation of calcium carbonate in the pores of cementitious materials. The reduction in water absorption in treatment T3 can be explained by the filling of the pores of the panels with calcium carbonate. Regarding treatment T4, the decrease in water absorption can be explained by the lower porosity found for this panel (Figure 2B).

For TS2h and TS24h, there was also no influence of the hot water treatment (T2), because the values were statistically equal to the control (T1) (Figure 4B). The thickness swelling of wood-cement panels is influenced by the amount of wood and its adequate covering by the cementitious matrix (Mendes et al., 2017). Due to the fact that carbonation accelerates cement curing, the covering of wood particles by the cement matrix was more efficient for the panels submitted to accelerated carbonation (T3 and T4), resulting in the lowest TS2h and TS24h values after immersion in water.

According to the BISON process (2017) wood-cement panels should present TS2h values lower than 1.0% and for TS24h lower than 1.5%. In this study only the panels submitted to accelerated carbonation (treatments T3 and T4) complied with the recommended by BISON (2017) for TS2h. As for TS24h, only the control (T1) did not reach the values recommended by BISON (2017).

Mechanical properties

The treatments T1 and T2 showed statistically similar values of MOR and MOE (Figure 5a and 5b). Wood is used in cementitious materials to optimize flexural properties and post-cracking behavior (Cabral et al., 2021). Extractives inhibit cement solidification, negatively influencing the mechanical properties of wood cement panels, which was not verified in this study, since the hot water treatment reduced the total extractives contents (Table 4). According to Quiroga, Marzocchi and Rintoul (2016) the setting inhibition does not depend only on the amount of extractives but also on the type.

The highest values of MOR and MOE were found for treatments T3 and T4, which were statistically similar. In Figure 6C and D it is observed that the carbonated panels (treatments T3 and T4) had no torn and fractured particles, providing a dense and compact cement matrix that favors the adhesion of the particles and results in higher mechanical strength. According to Tonoli et al. (2010) carbonation causes a reduction in the alkalinity of the matrix, reducing the degradation of the lignocellulosic material. Thus, the strength of the particle is preserved due to the lower alkalinity, improving the mechanical performance and bonding of the particle in the matrix of carbonated panels. The BISON (2017) determines a minimum value of 9.0 MPa for MOR and 3000 MPa for MOE. Thus, all treatments studied met the marketing standard for MOR, and for MOE only treatments T3 and T4. Cabral et al. (2018), examining cementitious panels produced with sugarcane bagasse subjected to accelerated carbonation, found an increase in the flexural rupture modulus of 78.69% and in the flexural elasticity modulus of 125.13%, compared to the non-carbonated cementitious panel.

The panels presented in their fracture the tenacification mechanisms associated to the particles, which act in the fracture process of the composite, such as debond, pull-out and bridging, capable of improving the capacity of the material to absorb energy and deform permanently without fracturing, through the slipping of the particles, instead of their rupture (Figure 6). The images revealed panels with particles adhered and anchored to the matrix, with transverse fissures, interrupted by the Erythrina poeppigiana particles, and covered by the hydration products, which may contribute to the petrification process of the particles. The panels showed micro cracks that can be attributed to calcium hydroxide crystals of lower strength, responsible for filling the spaces left by water, and that presented an altered morphology due to impurities present in the system. These characteristics can present themselves with different intensity and reflect in the mechanical behavior of the composite. Therefore, it is possible to visualize in the images the hydration products adhered to the particles, the displacements of some particles in the matrix and cracks in the transition zone.

Treatment T1 showed fractured particles, pulled and detached from the matrix, indicating low adherence (Figure 6a). Treatments T2 and T4 showed particles forming bridges and slipping in the matrix, indicating that the hot water treatment causes interface improvement (Figure 6b and 6d), which justifies the reduction of porosity in these treatments (Figure 2b).

The accelerated carbonation provided higher compressive strength values for treatments T3 and T4 compared to treatments T1 and T2 (Figure 5c). Accelerated carbonation improves the durability of cementitious composites, reducing porosity and making them less susceptible to cracks and fissures, resulting in higher compressive strength.

Higher internal bond values were found for treatments T3 and T4 compared to treatments T1 and T2 (Figure 5d), demonstrating that there was a better coverage of the particles in the cement matrix of the carbonate panels. The BISON (2017) determines a minimum value for internal bond of 0.40 MPa, thus all treatments met the marketing standard for the internal bond property.

Fourier Transform Infrared Spectroscopy (FTIR)

The band sizes of 940 and 1415 cm^-1 (Figure 7) are typical peaks of calcium carbonate (Reig; Adelantado; Moreno, 2002; Santos et al., 2021). A higher intensity of the peaks is verified for the treatment T3, indicating a greater amount of calcium carbonate, and a lower intensity for the T1 treatment.

The carbonation reactions in cementitious composites can be divided into three steps: First occurs to CO₂ diffusion in the matrix, then CO₂ is dissolved in the pores and Ca²⁺ ions are migrated from the cement particles. The third step is the formation of calcium carbonate (Teir; Eloneva; Zevenhoven, 2005; Wang et al., 2019). Accelerated carbonation promotes greater formation of calcium carbonate, which influenced the physical and mechanical properties of Erythrina poeppigiana cement-wood panels.

The FTIR technique allows revealing characteristic peaks of functional groups. In Figure 7, in addition to the calcium carbonate peaks, the peak in the 3367 cm^-1 range also stands out for its intensity and is characteristic of OH^- groups (Popescu et al., 2013).

Thermogravimetry

Figure 8 depicts the thermogravimetry (TG) and differential weight loss (DTG) of the panels in the different conditions evaluated. Endothermic events can be identified in the temperature zones of 100-250 °C (hydrated C-S-H phases) (Morandeau; Thiery; Dangla, 2014), 300-400 and 450-550 °C (Ca(OH)₂) and calcium carbonate), and 700-800 °C (stable and less stable carbonates) (Figure 8b).

The mass loss between 250-350 °C corresponds to the degradation of wood cellulose (Protásio et al., 2015), already between 400-450 °C occurs the degradation of Ca(OH)₂ and calcium carbonate and in the range of 550-700 °C only calcium carbonate (Choi et al., 2017) (Figure 8B). Among the evaluated conditions, treatments T1, T2 and T3 present similar thermal degradation curves in the range 50-600 °C (Figure 8a), but between 600-800 °C (calcium carbonate degradation range) the mass loss of treatment T2 is lower, due to the lower amount of calcium carbonate in this treatment (Figure 7). The T4 treatment presents the lowest thermal stability among the evaluated conditions, presenting the highest mass loss in the range of 300-800 °C (Figure 8a). In their studies with oriented cement slabs using particles of Eucalyptus spp., Cabral et al. (2023) observed smaller peaks of Ca(OH)₂ in carbonated plates, indicating greater consumption of Ca(OH)₂ for the formation of CaCO₃ caused by curing with CO₂.

Conclusions

The treatment with hot water decreased the extractive content and improved the interaction of the Erythrina poeppigiana particles with the cement, reducing the porosity of the wood-cement panels.

The accelerated carbonation allowed better physical and mechanical properties of the Erythrina poeppigiana Wood-cement panels, and all the quality requirements of BISON's standard were achieved.

The Erythrina poeppigiana wood is technically feasible for the production of wood-cement panels, a fact that might allow adding value and an appropriate destination for this specie.

References

Edited by

Publication Dates

History