Optimization of Particle Size and Ramie Fiber Ratio on Hybrid Bio Panel Production from Oil Palm Trunk as Thermal Insulation Materials

The abundant availability of waste oil palm trunks is one of the potential fibers for new thermal insulation mate - rials. While focusing on the manufacturing of thermal insulation materials, the main points to be considered are particle size, reinforcement fiber ratio, and press durations, besides binders type and temperature. This study aimed to optimize the manufacturing process of hybrid bio panels based on oil palm trunks as thermal insulation mate - rial. The response surface methodology (RSM), with a Box-Behnken Design (BBD), was used to model and op - timize the manufacturing process variables. A total of 17 hybrid bio panels were in operation and the independent variables used were particle size, ramie ratio, and press duration. The dependent variables were water absorption, thickness swelling, MOR, and thermal conductivity. The hybrid bio panel obtained under the optimum conditions was characterized by thermogravimetric analysis to observe thermal stability. On the basis of analysis of variance and the contour plot, it was discovered that the interaction between particle size and ramie fiber ratio was a signifi - cant variable to optimize hybrid bio panel manufacture. The thermal resistance and modulus of rupture of hybrid bio panels also improved with higher particle size and ramie fiber ratio. The optimum manufacturing process was obtained at OPT particle size of 0.248 mm, ramie fiber ratio of 19.775, and press duration of 25 min. This condition produces a thermal conductivity of 0.079 W/mK, modulus of rupture of 17.702 MPa, water absorption of 54.428%, and thickness swelling of 21.974%. In addition, the hybrid bio panel resulted in thermal stability of 341 °C.


INTRODUCTION
Energy efficiency has gained attention in recent years due to the concern about the crisis and global warming. In building construction, the investigation on the need of energy efficiency for thermal insulation is rapidly increasing [Awoyera et al., 2022]. This is because thermal insulation is one of the methods of reducing energy loss, specifically in commercial and residential buildings. It is also used to ensure the thermal comfort of people in building structures [Bünning, Hu- The building insulation materials are typically synthetic polymer products such as fiberglass, glass wool, polyurethane, and expanded polystyrene with low thermal conductivity. These materials such as glass wool and fiberglass can cause health issues and adverse effects, including biodegradation on the environment during production [Abu-Jdayil, Barkhad Pásztory, 2021]. Therefore, the replacement of traditional insulation materials with natural ones that have a low impact on the environment is of great concern in the Optimization of Particle Size and Ramie Fiber Ratio on Hybrid Bio Panel Production from Oil Palm Trunk as Thermal Insulation Materials construction industry. Natural fibers are a kind of renewable resource. M Ramesh, Palanikumar, & Reddy, 2017 reviewed the overall characteristics of natural fibers in bio-composites, including source, type, structure, composition, and properties. The properties of the bio-composites are based on the influence of natural fibers combinations. Furthermore, it was also found that there was a negative and positive impact on the environment during the cultivation stage due to the use of pesticides. On the other hand, the disposal of these bio-composites had a clear advantage for the environment. Using natural fiber for reinforcement bio-composites and/or biopolymers has such advantages as low cost, low relative density, high specific strength, renewable nature, and biodegradability.
According to [Tettey, Dodoo, & Gustavsson, 2014], the use of natural fiber as a substitute for synthetic insulating materials such as rock and glass wools reduced 6-8% carbon dioxide emissions and 39% of fossil fuels. Moreover, several reasons that need to be hybridized, such as higher moisture absorption and poor compatibility characteristics, have forced hybridizing with other synthetic/natural fibers. Review properties of hybrid composites such as mechanical, thermal, water absorption, morphological characteristics, tribological behaviour, and other properties have been reported by [Gupta, Ramesh, & Thomas, 2021]. In the field of sustainable thermal insulation materials, several investigations are focused on using renewable resources and waste with lower costs, better insulation properties, and fewer environmental effects [Cetiner & Shea, 2018 [Indonesia, 2020]. Generally, these wastes are usually burned or the biomass pellet is used [Bakar et al., 2013]. They have also been used as non-structural products such as particleboard [Komariah et al., 2019], lumber [Hashim, Sarmin, Sulaiman, & Yusof, 2011], and plywood [Loh et al., 2010]. The conversion of OPT to natural fiber to substitute synthetic ones as thermal insulation materials provides more benefit and potency. In addition, ramie fiber is one of the natural fibers with a lot of cellulose content and better mechanical properties. Several researchers have previously discussed the reinforcement of ramie fiber for green composites with several focus reviews such as; mechanical, thermal stability, and thermal conductivity [Manickam Ramesh, Rajeshkumar, & Balaji, 2021].
Previous reports were carried out on OPT as one of the potential raw materials for bio panel thermal insulation [Mawardi et al., 2021a[Mawardi et al., , 2021bMawardi, Aprilia, Faisal, & Rizal, 2022]. It was discovered that the thermal insulation of hybrid bio panels made from OPT is affected by manufacturing process variables such as particle size, raw material properties, binders and composition, press durations, and fiber ratio. Therefore, there is a need to maximize the selection of manufacturing process variables to obtain bio-panel thermal insulation with good heat resistance performance without compromising physical and mechanical properties. According to [Montgomery, 2017], response surface methodology (RSM) is a statistical strategy for generating, enhancing, and optimizing response variables applicable to various manufacturing processes. RSM is used to analyze problems that optimize all the process parameters collectively [ This study aimed to determine the optimal parameters for manufacturing hybrid bio panels based on OPT as thermal insulation using RSM statistical design. The particle size of OPT, ramie fiber ratio and press duration were variables applicable to the manufacturing process. The accuracy of the variables selection will facilitate the analysis of the interaction of physical and mechanical properties to determine the coefficient of thermal conductivity, which is not easily obtained by experimental testing.

Materials
OPT collected from a local oil palm plantation in Aceh and ramie fibers from an agricultural in Yogyakarta, Indonesia, were used as lignocellulosic raw materials. The material was sawn into rectangles with dimensions of 50×10×10 before being reduced manually into chips with 3×3×0.5 cm. The chips were ground into particles using a Disc Mill with three size levels, namely <0.07 mm (fine), 0.42-0.07 mm (medium), and 0.84-0.42 mm (coarse). The ramie fibers were chopped into strands less than 5 mm long. Subsequently, OPT particles were boiled in hot water for 30 minutes, and ramie fibers were soaked in a 5% NaOH solution for 1 hour. The materials were dried in an oven with a temperature of 80 °C to a moisture content of 10-12%. The chemical composition and the mechanical and physical properties of OPT and ramie fiber are given in Table 1. A natural binder, namely Tapioca was used as an adhesive in the bio panel manufacturing process (amylopectin: 83%, amylose: 17%) [Asrofi, Syafri, Sapuan, & Ilyas, 2020].

Manufacture of hybrid bio panel and testing
A total of 17 hybrid bio panels with 15×15×1 cm was manufactured using hot press equipment manually with the independent variable and OPT, ramie fibers as a co-reinforcement at three levels ratio, namely 0%, 10%, 20%, and tapioca bio binder of 30%. OPT particle and ramie fibers were combined with tapioca starch, followed by 100 ml of hot water. They were stirred with a mixer for 5 minutes until completely mixed and poured into a mold. Furthermore, the materials were prepressed for 5 minutes before being compressed at a temperature of 150 °C for 15 and 25 minutes to a target density of 0.80 g/cm 3 . The manufacturing variables were specifically selected to optimize bio panel production conditions using RSM. Before testing, the hybrid bio panels were conditioned for one week at room temperature of 25 ± 2 °C and relative humidity of approximately 60%.

Testing of hybrid bio panel
The characteristics of the hybrid bio panels

Box-Behnken design of response surface methodology
The software package Design Expert (version 6.0.11) was used for the statistical analysis of the experimental data. This program is used for various purposes, namely regression analysis of experimental data to fit an empirical mathematical equation, analysis of variance (ANOVA), and 3D visualizations of the response surface. In this study, response surface methodology (RSM) was used to investigate the effect of independent variables, which include particle size (X 1 ), press duration (X 2 ), and ramie fiber ratio (X 3 ), on response variables such as water absorption, thickness swelling, MOR, and thermal conductivity. Subsequently, the Box-Behnken Design (BBD) model was used to design the optimum variables of particle size, ramie fibers ratio, and press duration during the manufacturing of the hybrid bio panel as thermal insulation materials. The RSM is one of the statistical programs for a model building to optimize independent variables [Homayoonfal, Khodaiyan, & Mousavi, 2015] and calculate the best-operating factors and the area it meets the operating requirements [Montgomery, 2017]. The number of independent variables and levels used are shown in Table 2. The range of the variables was established and coded to be +1 for axial, center 0, and factorial points -1. According to BBD, 17 manufacturing conditions were randomly carried out as summarized in Table 3. Table 2 shows the correlation results between the independent and dependent variables obtained from the BBD for the suggested design of manufacturing hybrid bio panels. Furthermore, a quadratic model was selected to show the interactions between all variables. The result quadratic models form the effect of independent variables water absorption (Y 1 ), thickness swelling (Y 2 ) and MOR  Table 3. Design of independent and dependent variables according to BBD Run Independent variabel Dependent value Note: X 1 -particle size (mm), X 2 -press duration (min.), X 3 -ramie fiber ratio (%), Y 1 -water absorption (%), Y 2 -thickness swelling (%), Y 3 -MOR (MPa), Y 4 -thermal conductivity (W/mK).

Regression model of analysis of variance
(Y 3 ), and thermal conductivity (Y 4 ) as a function of particle size (X 1 ), press duration (X 2 ), and ramie fiber ratio (X 3 ) are given in regression Equations 1-4.

2
(1) Thickness swelling (Y 2 ) = 55.58 The actual and predicted values of the experimental results were calculated based on Equations 1-4 ( Table 4). The result showed no significant Table 4. Result from experimental design to manufacture hybrid bio panels based on OPT

Eff ects of independent variables on the response
The RSM analysis can interpret the interaction among the variables by generating three-dimensional (3D) response surface plots [Danish, Hashim, Ibrahim, & Sulaiman, 2014]. Figures 2 to 5 show the 3D response surface plots of the eff ect of interaction between independent and dependent variables formed from hybrid bio panels.

Water absorption
The 3D response surface plots for water absorption of hybrid bio panels were generated (Figure 1) to explore the eff ect of OPT particle size, ramie ratio, and press duration. Figure 1a shows that increased OPT particle size can signifi cantly improve the water absorption performance of hybrid bio panels. This indicated that the larger particle size absorbs more water than the smaller ones. Increasing the press duration during hybrid bio panels manufacturing did not aff ect the water absorption properties (Figure 1b). However, an increase in the quantity of ramie fi ber aff ected the decrease in water absorption.
The presence of ramie fi ber and a higher quantity also led to an increase in water absorption resistance. Hybrid bio panels with OPT particle sizes of 0.074 mm and 20% of ramie fi ber showed the lowest range of water absorption value compared to other bio boards. A slight decrease in water absorption was noticed by increasing the ramie fi ber ratio (Figure 1c). Hybridization of ramie fi ber can increase compactness consequence or densities of hybrid bio panels. This is indicated by the 20% hybridization which showed a lower range value for water absorption properties than others. Highly densifi ed hybrid bio panels had lower water absorption capability compared to those with low compaction ratios. Meanwhile, panels with lower compactness had more void spaces between fi bers, which leads to easy absorption of water. The addition of co-reinforcement has reduced water absorption due to suitable characteristics of the mixed material, decreasing the void and surface area exposed to water.

Thickness swelling
The interaction plots between OPT particle size, ramie fi ber ratio, and press duration for thickness swelling of hybrid bio panels are shown in Figure 2. These plots were similar trends as shown in Figure 2, where a decrease in particle size and increase in ramie fi ber ratio, reduce the thickness swelling values of hybrid bio panels. Figure 1. Response surface 3D plots on water absorption of hybrid bio panels (a) particle size and press durations, (b) ramie fi ber ratio and press durations, and (c) ramie fi ber ratio and particle size a) c) b) On the basis of Figure 2a which showed the lowest value at a particle size of 0.074 mm and press duration ranging from 19 to 21 minutes, higher particle size increases the thickness swelling of the hybrid bio panels. Although these properties were not signifi cantly aff ected by press duration, there was a slight improvement in their performance as the amount of ramie fi ber increased (Figure 2b), which was observed when the panels' OPT particle size was reduced (Figure 2c). This showed that the thickness swelling properties of hybrid bio panels infl uenced OPT particle size and ramie fi ber ratio. However, the press duration only marginally aff ected the properties. On the basis of the results, the thickness swelling of the bio panel was aff ected by the water absorption performance. Figure 3 showed the 3D relationship between OPT particle size, ramie fi ber ratio, and press duration on MOR properties of hybrid bio panels. From Figure 3a, the OPT particle size signifi cantly infl uenced the modulus of rupture compared to the press duration, where MOR increased with higher OPT particle size. The bio panel of 0.841 mm seems to have higher MOR properties than the samples with a size of 0.074 mm. This is due to larger particle size, leading to a more comprehensive interfacial bonding, or higher glue lines, which increases the fl exural strength of panels. This was also discovered in the interaction between ramie fi ber and press duration on MOR response, as shown in Figure 3b. A decrease in the quantity of ramie fi ber resulted in a strong reduction in MOR strength. Figure 3c shows the interaction between OPT particle size and ramie fi ber ratio for MOR properties of hybrid bio panels. The 3D response surface plots indicated that the ramie fi ber and OPT particle size signifi cantly infl uence the modulus of rupture, which is increased with higher OPT particle size. Moreover, an improvement in the MOR was also observed when there was an increase in the ratio of ramie fi ber in the bio panels.

Modulus of rupture
On the basis of the results, it can be concluded that MOR of hybrid bio panels was infl uenced by OPT particle size and ramie fi ber ratio, but not by press duration for bio panels manufacturing. Moreover, bio panels with large OPT particle size and the hybridization of ramie fi ber as coreinforcement require a greater load to make them rupture. Previous studies observed a similar tendency with diff erent reinforcements such as wood, bamboo, and rice straw using natural resin as a binder [Nguyen,   Response surface 3D plots on thickness swelling of hybrid bio panels (a) particle size and press durations, (b) ramie fi ber ratio and press durations, and (c) ramie fi ber ratio and particle size c) b) a) Figure 3. Response surface 3D plots on MOR of hybrid bio panels (a) particle size and press durations, (b) ramie fi ber ratio and press durations, and (c) ramie fi ber ratio and particle size c) b) a) The decrease in particle size led to a reduction in porosity, which indicated that a low porosity will improve the density of bio panels and reduces the MOR value.

Thermal conductivity
The 3D response surface plot between OPT particle size, ramie fi ber ratio, and press duration on thermal conductivity is shown in Figure 4. On the basis of the results, there is a signifi cant relationship between the particle size, co-reinforcement ramie fi ber ratio composition, and thermal conductivity coeffi cient of bio panels. When the ramie fi ber ratio is the highest and the particle size is the largest, the thermal conductivity coeffi cient of bio panels is lowest. The thermal conductivity coeffi cient increases with the smaller the particle size, which improves density and decreases the porosity. An increase in porosity can aff ect the thermal conductivity susceptibility of the material. This fi nding is similar to the research studied by [Pundiene, Vitola, Pranckeviciene, & Bajare, 2022] and [Dębska, Lichołai, & Krasoń, 2017]. However, larger particle size indicated a positive eff ect on thermal resistance (Figure 4a). Figure 4b shows that the interaction between the quantity of ramie fi ber and the press duration has little eff ect on the thermal conductivity value of the bio panels. The addition of ramie fi ber to OPT particles in the bio panels reduced the thermal conductivity coeffi cient (Figure 4c).
The pores formed contain gas and became the center of thermal scattering, reducing the heat passing through bio panel. The increase in the thermal resistance of the hybrid bio panel with higher ramie fi ber ratio and particle size of OPT led to a decrease in physical properties and an improvement in the MOR properties. The results showed that the thermal conductivity of hybrid bio panels is infl uenced by several parameters such as the type of the material, the thermal stability, the density associated with the compactness level, the number of pores formed, and the adhesive.

Optimization of variables
The optimization process was carried out using the desirability function, which ranges from 0 to 1 and combines multiple responses into one part. A very good value is 1, which indicates that the model can use the suggested variables. Table  6 shows an optimum hybrid bio panel manufacturing condition selected from variables with various OPT particle sizes, ramie fi ber ratios, and press duration. The optimum variables for manufacturing were determined using Equations (1)- (4). Since this study aimed to optimize OPT particle size, ramie fi ber ratio, and press duration in the manufacturing hybrid bio panels as thermal insulation, the variables were adjusted to obtain the least values for the minimum thermal conductivity coeffi cient, water absorption, and thickness expansion, while bending strength Figure 4. Response surface 3D plots on thermal conductivity of hybrid bio panels (a) particle size and press durations, (b) ramie fi ber ratio and press durations, and (c) ramie fi ber ratio and particle size (MOR) was maximum. Furthermore, the hybrid bio panels with ideal thermal insulation properties suggested OPT particle size 0.248 mm, ramie fi ber ratio 25, and press duration of 19.775 minutes press duration.

Thermal stability
The large OPT particle size and characteristics of ramie fi ber, with good thermal stability, have promoted the thermal resistance properties on the hybrid bio panels. The thermal stability of the hybrid bio panel was evaluated using thermogravimetry by analyzing the degradation and decomposition of the panel's constituent elements. Figure 5 shows the TGA curve of thermal weight loss and derivative thermogravimetric analysis (DTGA) curve of hybrid bio panels as a function of temperature increase. The TGA curve is divided into three major sections, where the fi rst degradation causes a slight loss of weight due to water evaporation at a temperature of 120 °C. It also shows the maximum decomposition of hybrid bio panels in the range of 310-405 °C, which causes partial oxidation of the starch and degradation of the fi ber. In the fi nal stage, the highest decomposition rate of the bio panel is observed above the 408 °C. Figure 5 also shows the DTGA curve, indicating the thermal stability of bio panel, at peak degradation of 341 °C. This result is better than a bio-panel manufactured from oil palm trunk and treated with ammonium dihydrogen phosphate at 200 °C and 330 °C [Komariah et al., 2019]. The thermal stability is infl uenced by the good interfacial bonding between OPT, ramie fi ber, and tapioca starch, leading to strong hydrogen bonds, thereby reducing weight loss in the sample. Furthermore, the thermal stability behavior of natural-fi ber-reinforced panels is also aff ected by fi ber types, treatment processes, and matrices.

CONCLUSIONS
The optimum manufacturing conditions of hybrid bio panels based on OPT as thermal insulation were determined using a BBD in response surface methodology software. A statistical study of mathematical model equations was built using ANOVA and experimental data. The results showed that the interaction between particle size OPT and ramie fi ber ratio was the largest signifi cant variable in optimizing hybrid bio panel production; however, the press duration is not a substantial variable. The result showed the lowest thermal conductivity of 0.079 W/mK with the modulus of rupture of 17.702 MPa, water absorption of 54.428 %, and thickness swelling of 21.974%. Furthermore, the optimized conditions of the manufacturing process variable at particle size OPT of 0.248 mm, ramie fi ber ratio of 19.775, and press duration of 25 minutes. The DTGA analysis showed that the hybrid bio panel has thermal stability at temperature 341 °C.