The natural gypsum used in this research was purchased from Omid Company (Semnan, Iran). Agricultural waste used in this research was bagasse obtained from Sugar cane workshop (Bahnemir, Iran). Chemical composition of bagasse was in Table 1. Commercial wollastonite was purchased from Vard manufacturing company of mineral and industrial products (Tehran, Iran). Dimensions of the nano wollastonite and specifications of the wollastonite composition are given in Figure 1 and Table 2, respectively. Density wollastonite was 2920 kg/m³.

Table 1:
Chemical composition of bagasse.

Figure 1:
SEM micrograph of nano wollastonite.

Table 2:
Composition of the Wollastonite Gel.

Bagasse (Figure 2a) were applied at three proportions (15 %, 25 %, or 35 %, with respect to the total content of binder in each mixture and different wollastonite substitution levels 0 %, 5 % or 10 %) for gypsum. The moisture content of bagasse was 50 % before dryer and 6 % after dryer, also oven dry density bagasse was 390 kg/m³.

First, bagasse was washed hot water and then milled into particles using a hammer mill (Figure 2a). Particle size average was 2 cm to 3 cm length 1 mm to 2 mm thickness. Oven dry density and dimensions of all boards were set at 1100 kg/m³ 550 × 270 × 20 mm³. Therefore, for the production of boards, the density (oven dry) of raw materials in each treatment was calculated first using density equation. Then the weight ratios of each board were determined (Table 3). Next, using a digital scale, the ratios of water, nano-wollastonite and calcium chloride were determined, and after mixing them in the mixing apparatus (laboratory mixer), these materials were added to the bagasse and gypsum paste that had already been weighted. Subsequently, the cake from the mixing machine was poured uniformly into a 55 × 27 × 6 cm³ metal mold and pressed under pressure (Burkle-LA-160) under cold (room conditions for 24 hours at a pressure of 30 kg/cm² to reach final thickness 20 mm). After pressing, the boards were kept for 24 hours conditions (20 ºC ± 1 ºC, 65 % ± 2% RH). After initial cooling, the boards were kept in a special chamber at a temperature of about 20 °C and relative humidity above 90 % for 20 days in order to minimize drying speed. After that, the boards were removed using a circular saw and placed in an air-conditioned room at 20 °C and 65 % relative humidity for 28 days (Figure 2b).

Figure 2:
(a) Bagasse chip, (b) Gypsum board with bagasse chip

Table 3:
Weight ratios of each board.

The samples were then evaluated for static banding three points edgewise such as modulus of rupture (MOR), modulus of elasticity (MOE) and internal bonding (IB) according to DIN EN 634-1:1195-04 (1995) using a universal testing machine (model GT-TCS-2000; Taichung Industry Park, Taichung, Taiwan) at the speed of 10 mm/min. Three replicates were made at a dimension of 550 × 270 × 20 mm³ for each board type. Samples dimensions for static banding and internal bonding test were 450 (loading span: 400 mm) length × 50 width × 20 thickness (mm³) and 50 length × 50 width × 20 thickness (mm³), respectively. For IB test, first, cut the specimens to 5 cm × 5 cm dimensions and then we glue them to the special jaws of the machine. Next we place the jaws inside the machine. The apparatus applies tensile force to the jaws, causing fractures in the specimen's brain, resulting in the amount of internal bonding being determined (Figure 3).

Figure 3:
Testing machine for internal banding.

The load and deflection were continuously recorded and the data were used to calculate modulus of rupture (MOR), modulus of elasticity (MOE) and internal bonding (IB) in based on Equation 1, Equation 2 and Equation 3:

(1)

(2)

(3)

Where P is the maximum force (N), L is the span length (mm), b is the sample width (mm), d is the sample thickness (mm), D is the deflection, A is the felling area (mm²).

The effect of composite formulation on the thickness swelling after 2 and 24 hours immersion in water was determined in samples with the dimensions of 50 mm × 50 mm × 20 mm according to DIN EN 634-1:1195-04 (1995). Sample was soaked in distilled water for 2 h and 24 h (room temperature). TS were then calculated according to the following Equation 4:

(4)

Where TS is thickness swelling (%), T₁ is thickness before immersing, T₂ is thickness after each immersion.

The boards measuring 150 length × 100 width × 20 thickness (mm³) were prepared according to the ISO 11925-2 (2010) specifications for the fire resistance tests (Figure 4). Mass loss of the samples due to fire exposure was vertically mounted on a holder up-straight and exposed to a Bunsen-type burner (with the internal diameter of 11 mm) hold at 45 degrees to the surface of the specimen for 120 seconds in accordance with method described by Esmailpour et al. (2017). Then, the burnt area, as well as weight loss, is measured. From the preliminary weight of specimens before fire exposure (W₁) and second weight of specimens after fire exposure (W₂), the Mass loss (ML) was calculated as Equation 5:

(5)

Figure 4:
Schematic picture of fixed fire testing apparatus.

The fracture surfaces were analyzed from samples fractured in the flexural strength test, using images obtained by scanning electron microscopy. The maximum magnification used here was 3000x. The specimens were cut with dimensions of 10 mm × 10 mm, attached to a sample holder and metallized with gold. The microscope used here was a Philips model XL 30 FEG carried out at the School of Electrical and Computer Engineering, University of Amir Kabir, Tehran, Iran.

Statistical analyses were performed using Statistica software v.13 (Dell Inc. 2016). The obtained results were analyzed statistically, and an analysis of variance (ANOVA) was performed to determine the significance of the tested parameter. A duncan’s multiple range test (DMRT) was performed to compare treatment means.

The results from the mechanical testing are shown in Figure 5, Figure 6 and Figure 7. As shown, the MOR, MOE and IB values increased in the gypsum boards that contained wollastonite compared to the gypsum boards without wollastonite. The highest and lowest modulus of rupture (MOR), values were found in the 10 % wollastonite and 15 % bagasse chip (10,8 MPa) and 0 % wollastonite and 35 % bagasse chip (3,25 MPa) treatments, respectively (Figure 5). Results of the modulus of elasticity (MOE) tests indicated that the highest (4870 MPa) and lowest (2370 MPa) MOE values were found in the same treatments as the highest and lowest MOR values (Figure 6). The highest and lowest internal bond (IB) values were found in the 10 % wollastonite and 15 % bagasse chip (2,1 MPa) and 0 % wollastonite and 35 % bagasse chip (0,54 MPa) boards, respectively (Figure 7). The abundance of extractive materials in lignocellulosic materials and agricultural wastes, especially phenolic and sugary materials, reduces gypsum hydration heat, slows the hydration reaction time, and increases bagasse retention time (Rangavar et al. 2016). There are several reasons for the increased mechanical properties with increasing nano-wollastonite in the mix. By measuring the hydration heat of gypsum-bagasse slurry and different amounts of nano-wollastonite, it was found that with increasing nano, the hydration heat was increased, which resulted in faster and better curing of bagasse with gypsum and thus increased board strength. Silica in nano-wollastonite forms a calcium silicate hydrate (Ca-Si-H) gel, preventing the penetration of chlorine ions, sulfates and other harmful chemicals into boards, increasing durability, increasing hydration heat, speeding gypsum and mechanical strength (Li et al. 2004). High substitution levels of wollastonite mitigated the negative effect of the extractive materials (Karimi et al. 2012, Ma and Wang 2012). Consequently, the mechanical properties were improved.

Figure 5:
Effect of varying levels of nano wollastonite on the MOR of composites composed of different percentages of bagasse; letters on each column indicate Duncan’s grouping at the 99 % level of confidence.

Figure 6:
Effect of varying levels of nano wollastonite on the MOE of composites composed of different percentages of bagasse; letters on each column indicate duncan’s grouping at the 99 % level of confidence.

Figure 7:
Effect of varying levels of nano wollastonite on the IB of composites composed of different percentages of bagasse; letters on each column indicate duncan’s grouping at the 99 % level of confidence.

The physical properties of the panels evaluated were thickness swelling after 2 and 24 hours immersion in water, density and fire resistance (Mass loss). The results of these tests are presented in the Figure 8, Figure 9 and Figure 10. Adding nano wollastonite to gypsum board has been shown to decrease the thickness swelling of boards. The lowest thickness swelling was observed for 10 % wollastonite and 15 % bagasse chip, and the highest was observed for the sample without wollastonite and 35 % bagasse chip (Figure 8). Nano-wollastonite addition to gypsum board and increases the density and decreases the air content in boards. The highest density was found in 10 % wollastonite and 15 % bagasse chip, showing 28 % of increase in comparison to the sample without wollastonite and 35 % bagasse chip (Figure 9). One of the reasons can be attributed to biometrics, slimming coefficient, sponge structure and hydrophilic properties of bagasse. It was also observed that with increasing nano, the dimensional stability of the boards improved. The reason for this can be attributed to the hydrophobicity of nano-wollastonite (Ciullo 1997). Nano-wollastonite, due to its high surface area and small size, fills the possible pores inside the board and blocks the water entering the board, thereby reducing the thickness swelling of the board and increasing the density of the board (Hassanpoor-Tichi et al. 2019). Also, with increasing bagasse content in the boards, their density decreased, which is due to the reversal of the thickness swelling of the boards made from higher ratios of this lignocellulosic material. The results of microscopic imaging are a strong reason for this claim (Figure 11). The results indicated that the fire-retarding properties of the samples were improved by the addition of wollastonite (Figure 10). The lowest mass loss was noted with the sample containing 10 % wollastonite and 15 % bagasse chip (Figure 10). The reason we can say this is because, firstly, nano-wollastonite has a high heat transfer (2,5 w/m∙K) and its heat is not kept at a single point and is rapidly transferred to a point where it has less heat and this causes to reduce the fire proofability of boards and increase their fire resistance (Taghiyari et al 2014), secondly, nano-wollastonite contain silica, titanium, magnesium, calcium and iron that are minerals. This has caused, boards made with higher amounts of nano-wollastonite have higher relative fire resistance.

Figure 8:
Effect of varying levels of nano wollastonite on the thickness swelling (TS) of composites composed of different percentages of bagasse; letters on each column indicate duncan’s grouping at the 99 % level of confidence.

Figure 9:
Effect of varying levels of nano wollastonite on density of composites composed of different percentages of bagasse; letters on each column indicate duncan’s grouping at the 99 % level of confidence.

Figure 10:
Effect of varying levels of nano wollastonite on Mass loss of composites composed of different percentages of bagasse; letters on each column indicate Duncan’s grouping at the 99 % level of confidence.

Scanning Electron Microscopy (SEM) was used to investigate the morphology of the fracture surface of boards. Figure 11 shows different amounts of nano-wollastonite. Increasing nano-wollastonite from 0 % to 10 % improves the adhesion between bagasse and Gypsum, which can lead to higher mechanical and physical strength of boards. Also higher percentages of nano-wollastonite in the boards and reaction of this material with calcium chloride and calcium hydroxide results in high production of hydrated calcium silicate gel (Figure 11a). This gel prevents the penetration of chlorine ions, sulfates and other harmful chemicals into the boards and seals, protects, reduces permeability, durability and self-healing between gypsum and bagasse. The images also show that nano-wollastonite provides a stronger, provides strong bonding between bagasse and gypsum, thus filling the porosity and voids between bagasse and gypsum, thereby increasing the dimensional stability and density of boards. However, as shown in Figure 11a, there is no good bonding between the bagasse chip and the gypsum, and the absence of nanocrystals causes large cavities.

Figure 11:
Scanning electron microscopy (SEM) of composite Gypsum- bagasse in the presence of (a) 10 % nano, (b) 5 % nano and (c) without- nano.

^♠Corresponding author: hasanpoortichi@gmail.com