PARAMETRIC OPTIMIZATION OF OIL PALM MESOCARP FIBER VALORIZATION WITH HYBRID OZONATION-ULTRASONIC PRETREATMENT METHOD

: Oil palm mesocarp fiber is a promising lignocellulosic biomass as a raw material for valorizing biomass into more valuable products such as second-generation biofuels, biocomposites, or bioenergy. However, the lignin composition present in lignocellulosic biomass provides resistance to the valorization process and protects the cellulose composition, thereby limiting the conversion of cellulose into more valuable products. The hybrid ozonation-ultrasonic method as a lignin-degrading method is starting to be considered an effective method. Additionally, a Box-Behnken Design (BBD) was employed to investigate each independent variable's effect on pretreatment process conditions using the response surface methodology (RSM), namely reaction time (30-90) min, reaction temperature (20 -40) o C and ozone flow rate (1-3) L/min to the response of the percentage of lignin degradation (%). The optimum condition of the pretreatment process is determined using the desirability function graph. The results showed that reaction time, reaction temperature, and ozone flow rate had a significant effect on lignin degradation (p <0.05). The optimum conditions obtained the highest percentage of lignin degradation, namely 92.08% at a reaction temperature of 30 o C with an ozone


INTRODUCTION
Mesocarp fiber is one of the by-products produced in large quantities by the palm oil industry in the palm oil production process. Total palm oil production continues to increase yearly due to the increasing world market demand for vegetable oil. In 2020, the total world palm oil production reached 73 million tonnes per year, then it will increase in the following year 2021 to reach 75 million tonnes per year, and now the total worldwide palm oil production has reached 77 to 79 million tonnes per year [1,2,3]. Then, from the total amount of palm oil production, it produces around 12-15% by-products in the form of mesocarp fiber which, if not handled seriously, can become a problem in waste management and endanger the surrounding environment [4]. In line with the problems that can be caused, oil palm mesocarp fiber is also one of the interesting potential lignocellulosic biomasses because of its abundant availability at a low cost. It has a high main component of cellulose which is around 40.12% by weight, hemicellulose 20.12% by weight, and 30.33% lignin by weight, so if utilized properly, it can be promising in terms of the economy [5,6,7,8].
The valorization of oil palm mesocarp fiber biomass, which entails converting lignocellulosic biomass into second-generation biofuels or other valuable products such as biocomposites, has become a current trend. However, the complex structure of lignocellulosic biomass has received special attention because of the lignin component that hinders the accessibility of cellulose, thereby inhibiting the valorization of lignocellulosic biomass [9]. Therefore, the pretreatment process is an important step before valorizing lignocellulosic biomass. The pretreatment process's primary goal is to degrade the complex structure of lignocellulosic biomass into simpler fibrous components and remove some of the lignin components in what is commonly referred to as the biomass delignification process [10]. Ozone-based advanced oxidation technology is one of the oxidation methods that has recently gained popularity for degrading complex compounds in waste and biomass [11]. In degrading complex compounds effectively and optimally, advanced ozone-based oxidation technology is based on the use of radical generation such as the hydroxyl radical (OH * ) as an oxidizing agent which has a strong oxidation potential value and is non-selective so that it shows good potential in the treatment of lignocellulosic biomass [12,13].
Ozone-based advanced oxidation processes (AOPs) can degrade complex compounds by two mechanisms that depend on the pH of different solutions, namely directly and indirectly. The indirect oxidation of ozone involves the hydroxyl radical (OH * ) as an oxidizing agent, while the direct oxidation of ozone represents oxidation with ozone molecules [14]. Both ozone-based advanced oxidation processes rely on different pH solutions to react with complex compounds. Ozone oxidation directly oxidizes complex compounds when the pH of the solution is acidic (< 5), while indirect ozone oxidation can take place when the pH of the solution is alkaline (> 8) [15]. The oxidation potential of ozone is 2.07 eV, which makes it a selective and strong oxidizer [11,14]. The ozonation process as a pretreatment method for lignocellulosic biomass has several advantages, such as a strong oxidation potential value and a characteristic of ozone molecules that they react readily with conjugated double bonds found https://doi.org /10.31436/iiumej.v24i2.2717 in complex compounds and functional groups with high density, which makes ozone able to oxidize lignin which has a high C=C bond [16]. In addition, ozone can be easily decomposed into compounds with greater oxidizing abilities, such as hydroxyl radicals (OH * ), which have a potential oxidation value of 2.80 eV and are also non-selective so that they can easily be better in degrading complex compounds of lignocellulosic biomass [11].
Ozonation is also a green pretreatment method because, in the process of oxidizing lignocellulosic biomass, it does not produce side products that can harm the surrounding environment. Ozone can decompose into oxygen molecules, so it is safe if directly discharged into the surrounding environment. The filtrate produced from the lignin degradation process can be utilized for lignin recovery [17]. Praptyana and Budiyono [18] reported that the pretreatment of mahogany sawdust biomass using ozone in alkaline conditions could degrade lignin and increase biohydrogen production up to 86.9% over other pretreatments. Similar results were also reported by Hassan et al. [19], that ozone pre-treatment of green algae Ulva lactuca could degrade complex biomass compounds, thereby significantly increasing biogas production. In addition, ozone has a few drawbacks that limit its industrial use, which is related to the insufficient solubility and stability of ozone molecules in solution and their limited selectivity in the oxidation process resulting in the low mass transfer of ozone in solution [20,21].
In recent years, researchers have become interested in finding ways to improve ozone mass transfer. The ultrasonic process provides several advantages compared to other technologies in increasing the efficiency of ozone-based advanced oxidation processes (AOPs), which include a simple operational process that can increase the solubility of ozone and does not produce harmful pollutants [22]. Several studies have found that ultrasonic cavitation has the ability to increase the mass transfer of ozone in water and also increase the production of free radicals such as hydroxyl radicals (OH * ). This is because the cavitation phenomenon in the ultrasonic process can cause the formation, growth, and collapse of acoustic bubbles, which can produce great energy with very high temperatures (> 4000 K) and local pressure (> 500 bar) as a result of the hotspot phenomenon that is formed when the rupture cavitation bubbles to increase the surface area of contact between ozone and solution and can cause dissociation of water molecules which leads to an increase in the mass transfer of ozone and an increase in the formation of radical compounds such as hydroxyl radicals (OH * ) [23]. The hybrid of ozonation and ultrasonic methods can accelerate the oxidation process of lignin by hydroxyl radical compounds (OH * ) into oxyaromatics compounds, which then turn into short-chain aliphatic acids. Additionally, the collapse of cavitation bubbles in the ultrasonic process also produces hydromechanical shear forces in solution which can interfere with Surface morphology of lignocellulosic biomass [24]. The synergistic effect between the ozonation and ultrasonic processes can be described as follows [25].
Shen et al. [26] investigated the effectiveness of using ozonation and ultrasonic processes simultaneously to provide a significant degradation effect of X-3B reactive red dye up to 99.2%. In addition, Weavers and Hoffmann [27] also reported that the mass transfer of ozone was much larger at a frequency of 20 kHz, which resulted in a mass transfer coefficient value (KLa 0.28 -0.41 min -1 ), while at a frequency of 500 kHz, the value of the displacement coefficient was mass is only equal to (KLa 0.05 -0.1 minutes -1 ). Therefore, the combination of ozonation and ultrasonic processes is an approach that has great potential and promise. The synergistic effect of the combination of the two processes increases the efficiency of the degradation of complex compounds such as lignocellulosic biomass. In line with the synergistic effect of the hybrid ozonation-ultrasonic method, process variables also affect the effectiveness of the lignocellulosic biomass pretreatment process. Optimization of the hybrid ozonation-https://doi.org/10.31436/iiumej.v24i2.2717 ultrasonic process variables using statistical and mathematical approaches such as the Response Surface Methodology (RSM) method in the biomass pretreatment process have not been reported so far.
Thus, it is necessary to optimize the process variables to provide information about the effect of the pretreatment process variables on the response, namely lignin degradation, and to save processing time and pretreatment costs [5]. This study applied a hybrid ozonationultrasonic process as a pretreatment method for lignocellulosic biomass in oil palm mesocarp fibers. Three different operational variables, operating temperature, operating time, and ozone flow rate were evaluated using a three-factor box-behnken design (BBD) approach, which is one type of design of experiment (DOE) in the response surface methodology (RSM) method to determine the most influential variable on the response and to obtain optimal operating conditions in the biomass pretreatment process.

Material and Methods
The oil palm mesocarp fibers in this study were obtained from Indonesian palm oil companies. Mesocarp fiber raw materials were washed with water to remove particulates and dried in an oven at 105 o C for 24 hours. Then the size was made uniform at 60 mesh and stored in a closed container at room temperature. The chemicals used in this study were H2SO4 with a purity of 95% -97% (E. Merck Cat No. 100731), NaOH with a purity of 99% (E. Merck Cat. No. 106498), ozone gas obtained from an ozone generator (Dipo Technology Indonesia) and ultrasonic irradiation was carried out using an ultrasonic bath type KLS 303365 equipped with a thermostatic water bath with a frequency of 42 kHz.

Oil Palm Mesocarp Fiber (OPMF) Characterization
The characterization of the composition of cellulose and hemicellulose in oil palm mesocarp fibers was determined by the cellulose isolation and bleaching methods that were slightly modified from Candido [28]. While the characterization of lignin composition in oil palm mesocarp fibers was determined using the Chesson Datta method, namely by weighing 1 g of dry sample (weight a), then reacting with 150 mL of distilled water for 1 hour at a temperature of 100 o C. The resulting residue was then washed with hot aquadest, dried in an oven for 24 hours, and weighed to a constant (weight b). Then the solid was reacted again with 150 mL H2SO4 1N for 1 hour at 100 o C, filtered and washed with distilled water, dried in the oven for 24 hours at 100 o C, and then weighed to a constant (weight c). The weighed solid was then reacted again with 10 mL of 72% H2SO4 for 4 hours at room temperature, then added to 150 mL of 1N H2SO4 and the mixture was refluxed for 1 hour. The residue was washed, dried, and weighed to a constant weight (weight d). Following that, the final solid is subjected to dry ashing using a muffle furnace at 525 o C, and the dry ash resulting from that process is weighed until constant (weight e). The percentage composition of cellulose, hemicellulose and lignin can be calculated by equations (1 -4). Changes in the mesocarp fiber surface morphology were confirmed using Scanning Electron Microscopy (SEM). In contrast, the crystallinity value of the OPMF before pretreatment and after pretreatment was confirmed by X-Ray Diffraction (XRD) rays, and the change in the chemical structure after the pretreatment process was confirmed by using Fourier-Transform Infrared Spectroscopy (FTIR). (4) In which a represents the initial dry weight (g) of the oil palm fiber sample, b represents the dry weight of the sample residue after refluxing with hot aquadest (g), c represents the weight of sample residue after refluxing with H2SO4 1 N (g), d represents the weight of sample residue after 72% H2SO4 treatment (g), and e represents the weight of ash from sample residue (g).

Hybrid Ozonation-Ultrasonic Pretreatment
The hybrid ozonation-ultrasonic pretreatment process for oil palm mesocarp fibers was carried out using a glass reactor equipped with a gas sparger and an ultrasound bath, as shown in  Oil palm mesocarp fibers weighing 15 g were placed into a glass reactor connected to an ozone generator and ultrasound bath, then 225 mL of distilled water was added, which was adjusted to pH 9 by adding 0.1 M NaOH solution, and the reaction process was carried out by flowing ozone gas and ultrasonic irradiation at an intensity of 42 kHz. After the reaction process is complete, the ultrasound bath and ozone generator were turned off, and the reaction products were filtered and then washed with distilled water until the pH was neutral. The filtered residue was dried using an oven at 105 o C for 24 hours and then stored for analysis of the composition of lignin, hemicellulose, and cellulose using the Chesson-Datta method.

2.4.Process Variable Optimization
Experimental analysis of the mesocarp fiber pretreatment optimization process was carried out using statistical software 6.0 with RSM and BBD design, which has three factors, one block, and fifteen trials, was chosen to identify the optimal parameters that have a significant influence on the response and investigate the optimum conditions for the pretreatment process of oil palm mesocarp fibers [29]. Three independent variables in the pretreatment process, namely reaction time (X1), reaction temperature (X2), and ozone flow rate (X3), each of which has three levels of value, were selected based on Abdurrahman [25]. The ranges of independent variables in the experimental process are summarized in Table 1, while the experimental designs given by the RSM are presented in Table 2.  The data obtained were analyzed using analysis of variance (ANOVA) and the response surface to determine the functional relationship between the independent process variables and the desired response [30]. The second-order polynomial regression equation describes the significance of the relationship between the independent variables in the response. The polynomial equation is described as follows: Where Y is the desired response, βO represents a constant coefficient, Xi and Xj represent independent variables, k represents the number of independent variables, and βi, βii, and βij represent linear coefficients, quadratic coefficients, and interaction coefficients. Using the RSM method in the experimental process can help summarize the number of trials needed to optimize the system and response so that the optimization process can be more efficient and save the cost of the experiments used [31].

2.5.Analysis of Cellulose Crystallinity with X-Ray Diffraction (XRD)
Crystallinity index (CrI) values of oil palm mesocarp fiber biomass without pretreatment and pretreatment and commercial cellulose were analyzed using X-rays at an angle of 2θ ranging from 5 o to 100 o . The detector detects the recorded X-ray diffraction spectrum results to form a diffraction pattern with peak intensity, which can be used to calculate cellulose crystallinity. The crystallinity index (CrI) value can be calculated using the following methods [32].
Where (I002) denotes the crystal peak intensity, (CrI) the crystallinity index, and (Iam) the amorphous phase peak intensity.

Functional Group Analysis with Fourier-Transform Infrared Spectroscopy (FTIR)
Fourier-transform infrared spectroscopy (FTIR) analysis aims to determine the chemical composition of lignocellulosic biomass of oil palm mesocarp fiber before pretreatment and after pretreatment with commercial cellulose composition as the standard value. Infrared spectroscopy (FTIR) examination of lignocellulosic biomass was performed at wavelengths between 400 cm -1 and 4000 cm -1 . The results of all formed spectra are reported in reflection mode at a predetermined wavelength range.

2.7.Cellulose Morphology Analysis by Scanning Electron Microscopy (SEM)
In order to study how the oil palm mesocarp fiber changed morphologically before and after pretreatment with lignin degradation, scanning electron microscopy (SEM) was conducted. Micrographs of oil palm mesocarp fiber biomass samples were observed at 7500 x magnification.

Oil Palm Mesocarp Fiber Composition
The main composition of oil palm mesocarp fiber consists of carbohydrates, namely holocellulose (cellulose and hemicellulose) and lignin. Oil palm mesocarp fiber composition can be determined using the Chesson-Datta method. The results of measurements using the https://doi.org/10.31436/iiumej.v24i2.2717 Chesson Datta method show that the composition of the oil palm mesocarp fiber in this study is in accordance with previous studies that have been reported in the literature ( Table 3). The lignin composition in mesocarp fiber, which is relatively high, gives toughness and stiffness to the cell walls of biomass so that it can become an obstacle in the valorization process of oil palm mesocarp fiber biomass both in the bioenergy and biocomposite fields [33]. From the data reported, the composition of lignin in oil palm mesocarp fibers is relatively different. It is influenced by differences in the background from plantation areas and types of oil palm as well as the level of maturity of oil palm fruit when the sample is used [37]. Therefore, it is very important to analyze the initial composition of lignocellulosic biomass to determine the composition of the raw materials to be used, considering that cellulose, hemicellulose, and lignin are the main compositions that can affect the efficiency in the valorization of oil palm mesocarp fiber biomass. Additionally, the results showed that the lignin and hemicellulose content of palm mesocarp fiber decreased after the hybrid ozonizationultrasonic pretreatment, as lignin and hemicellulose had been degraded.

Statistical Analysis and Empirical Models for Lignin Degradation
In this study, the results of the lignin degradation pretreatment and the predicted values generated by the experimental design of the statistical software are shown in Table 4.
The percentage of lignin degradation in mesocarp fiber obtained from this study showed varying values. The lowest percentage of lignin degradation was obtained at 79.27%, and the highest percentage of lignin degradation was obtained at 92.08%. While the process of lignin degradation using the ozonation and ultrasonic methods used individually has been reported by previous studies and shows that lignin degradation is only 87.9% in the ozonation process and 20.11% in the ultrasonic process, respectively [5,38]. Thus, the pretreatment of lignocellulosic biomass using the hybrid ozonation-ultrasonic method showed an increase in the lignin degradation process. Regression analysis uses second-order polynomial equations to explain the relationship between the independent and dependent variables. The polynomial equation is presented in Eqn. percentage of lignin degradation. It was reflected in the negative value of the model. The significance of the polynomial equation can be validated by the resulting regression coefficient (R 2 ) and statistical analysis such as analysis of variance (ANOVA). The regression coefficient value formed from this study is 99.86%. It proves that the predicted and observed values are close together, so the model can be represented in predicting lignin degradation with an error value of less than 5%. The significance of the independent variables of the hybrid ozonationultrasonic pretreatment consisting of reaction time, reaction temperature, and ozone flow rate to the response, namely the percentage of lignin degradation, is shown in Table 5. P values (probability) can be used to differentiate statistical analysis results in ANOVA tables. Where if the value of P (probability) obtained is less than 5% or (P <0.05), then it is considered significant because it has a probability level of 95%. It can be inferred that the independent variable and response have a greater significance if the P value (probability) is smaller [39]. Statistical analysis in this study showed that the linear function and quadratic function of the variable reaction time (X1), reaction temperature (X2), and ozone flow rate (X3) showed a significant effect on the response of the percentage of lignin degradation because it had a P value <0.05. The interaction function between the reaction time variable and the ozone flow rate (X1X3) showed similar P value. Meanwhile, the interaction function between the variable reaction time and reaction temperature (X1X2) and the reaction temperature variable and ozone flow rate (X2X3) did not have a significant effect on the response (P > 0.05). This hypothesis is strengthened by the F-test (Fisher), where the results obtained from the F-test show that the independent variables (reaction time, reaction temperature, and ozone flow rate), both in linear and quadratic form, have a higher F value than the F table (F-value > 4.772) so that it can be said that the independent variables in the linear and quadratic forms have a significant effect on lignin degradation. In the interaction between variables, the F-value has a smaller value than the F-table (F-value < 4.772), so the interaction between variables does not have a significant effect on lignin degradation.  The residual mean square (MS) value in the analysis of variance (ANOVA) is used to describe the difference between the experimental data and the predicted value of the model [40]. This study produced a residual mean square (MS) value of 0.0173, so it can be concluded that the model used is good enough and accurate in explaining the closeness between the experimental results and the predicted values obtained from the experimental design. The closeness between the experimental data and the predicted value is clarified by the coefficient of determination (R 2 ) obtained, which is equal to 0.9996, so it can be stated that 99.96% of the model can explain the experimental data. The reliability and significance of the model can be validated by looking at the average probability percentage in Figure 3a and the relationship between the experimental results obtained, and the predicted value of the model can also be validated by the observer vs predicted diagram illustrated in Figure 3b.
The diagonal line on the normal probability plot shows the data predicted by the experimental design. In contrast, the dots around the diagonal line show the value of the experimental results obtained. The closer the points, which are the experimental data, to the diagonal line, the residual values are normally distributed [41]. This normal distribution analysis aims to observe the magnitude of the deviation from the model. In this study, the normal probability plot shows dots that spread and approach the diagonal line, so it can be said that the model given by the experimental design can fulfill the assumption of normality.

3.3.Effect and Interaction of Variables Process on Lignin Degradation
The influence of process variables and their interactions during the pretreatment of lignin degradation with the ultrasonic hybrid ozonation method can be explained by employing response surface plots. Response surface analysis can provide a 3-dimensional profile that explains the relationship between the independent variables and the response. The response surface plot of the percentage of lignin degradation is presented in Figure 4. Based on the response surface plot images, the percentage of lignin degradation increases with increasing reaction time, reaction temperature, and ozone flow rate up to a certain critical point. The highest percentage of lignin degradation is represented by the contour plot area of the response surface, which has a solid red color. In contrast, the smallest percentage of lignin degradation depicted on the contour plot is colored green. In this study, the contour plots showed that more than 90% of lignin degradation was obtained when it was reacted at a temperature of 24-31 o C, with an ozone flow rate of 1-2 L/minute, for 50-65 minutes. According to Ramados et al. [42] and Pujakaroni et al. [43], the highest percentage of lignin degradation occurs when the temperature is 40 o C for 60 minutes. Increasing the reaction time can extend the contact time between radical compounds and a little ozone with lignocellulosic biomass, thus increasing lignin degradation. However, using a reaction time of more than 60 minutes does not have a significant effect because most of the lignin has been degraded so that it opens the surface of lignocellulosic biomass, and ozone could attack the holocellulose component [44]. Likewise, an increase in temperature causes an increase in the solubility and diffusion coefficient of the lignin component. On the other hand, although the concentration and solubility of ozone in water decrease with an increase in the reaction temperature, more hydroxy radical compounds are also produced under the same conditions, which have a greater oxidation potential value than ozone so that they are more oxidative which leads to an increase in the rate of degradation [45].
According to Chiha et al. [46], the degradation products increased from 20°C to 30°C as temperature increased and tended to be constant and even decreased when the temperature was higher than 30 o C. The decrease in the efficiency of the organic compound degradation process using the hybrid ozonation-ultrasonic method is caused by a decrease in ozone solubility and a decrease in the formation of radical compounds due to an increase in water vapor in the bubbles, which provides a cushioning effect on the collapsed bubbles so that the bursting of cavitation bubbles only produces low energy [47]. A flow rate of 2 L/min of ozone has a significant influence on the results, as shown by Wang et al. [48], that the ozone flow rate is increased, and degradation efficiency increases because cavitation bubbles form under ultrasonic irradiation, accelerating the mass transfer of ozone and causing radical compounds to form that can accelerate organic compound degradation. However, excessive ozone flow rates can cause greater turbulence, decreasing the contact time between ozone and water and shorter biomass, thereby reducing the degradation percentage [49].

3.4.Optimization of The Hybrid Ozonation-Ultrasonic Pretreatment Process Variables
Optimizing the lignin degradation pretreatment process using the hybrid ozonationultrasonic method was conducted by analyzing the desirability function represented by the response desirability profiling contained in statistical software 6. This tool consists of a series of graphs that represent every independent variable to determine whether the independent variables provide significant responses to the desired outcome. The results of optimizing the lignin degradation pretreatment process with the hybrid ozonation-ultrasonic method are shown in Figure 5.
The results of the analysis of the desirability function for each independent variable and the predicted response at optimal conditions can be observed from the response desirability profiling graph. The desirability profile graph shows that the percentage of lignin degradation in the hybrid ozonation-ultrasonic pretreatment has increased along with the increasing values of independent process variables such as reaction time, reaction temperature, and ozone flow rate until they reach a certain critical value. The optimal condition of the lignin degradation process can be seen from the intersection of the independent variables with the highest response values. In this study, the optimum conditions were obtained at 30 o C, with an ozone flow rate https://doi.org/10.31436/iiumej.v24i2.2717 of 2 L/min, and for 60 minutes of reaction time, the optimum response was in the form of a lignin degradation percentage of 92.08%.

3.5.Proposed Hybrid Ozonation-Ultrasonic Reaction (AOPS) Mechanism
The process of degradation of organic compounds such as lignocellulosic biomass using a combined method of ozonation and ultrasonic processes provides a higher degradation rate than the ozonation and ultrasonic processes when carried out individually. The possible reaction mechanism in the hybrid ozonation-ultrasonic process in degrading lignocellulosic biomass is shown in Figure 6. Decomposed ozone molecules initiate the mechanism of the hybrid ozonation-ultrasonic reaction, and water molecules dissociate in the presence of cavitation bubbles under ultrasonic irradiation to produce the free radical compounds shown in equations (7) and (8), then oxygen atoms produced from the ozone decomposition process. It will react directly with steam from water molecules in the liquid phase to form radical hydroxy compounds, as in equation (9). In addition, a small part of the ozone dissolved during the ultrasonic process can react with lignin directly on lignocellulosic biomass as shown in Figure 7 and then the lignin is degraded. The hybrid ozonation-ultrasonic process generally provides high efficiency in the lignocellulosic biomass pretreatment process because each ozone molecule will produce twice the hydroxyl radicals, which can better oxidize lignin and also the products resulting from the dissociation of water molecules and ozone decomposition will react with each other in the bubble interface and then will diffuse into the aqueous phase (Eq. (10)- (14)) [25]. Zhao et al. [50] reported that the enhanced decomposition of the ozone molecule in the cavitation bubbles under ultrasonic irradiation is due to mechanical effects such as the collapse of the cavitation bubbles. The reaction mechanism for the degradation of lignin compounds by hydroxyl radicals formed during the decomposition process of ozone and the dissociation of water molecules due to the effects of the acoustic cavitation phenomenon in the hybrid ozonation-ultrasonic pretreatment process is shown in Figure 8. However, it is important to note that the mechanism of the proposed lignin degradation reaction by hydroxyl radical compounds is only a speculative mechanism based on the reaction mechanism that has been proposed previously in the literature [51,52]. Pretreatment of lignin degradation with hydroxyl radical compounds involves several processes, such as demethylation or demethoxylation, side chain oxidation, and aromatic hydroxylation processes. Lignin degradation begins with breaking aryl ether bonds in lignin and forms water-soluble phenolic groups [53]. Furthermore, the highly electrophilic hydroxyl radical compounds will also attack other lignin groups rich in electrons through side chain oxidation processes and aromatic hydroxylation.

3.6.X-Ray Diffraction (XRD) Characterization of Lignocellulosic Biomass
XRD analysis of oil palm mesocarp fiber aims to determine the crystal structure of the biomass and structural changes both before and after the lignin degradation process. Cellulose in lignocellulosic biomass has crystalline and amorphous fractions, while the hemicellulose and lignin components have amorphous fractions. The crystallinity index (CrI) can be determined by referring to the XRD pattern, which is the diffraction peak of the cellulose I structure at around 20 o and the lowest diffraction peak at around 18 o , which is an amorphous cellulose region. Most industries have used analysis of cellulose crystallinity to determine elasticity, absorption, and other physical properties, which are important parameters in the production of biocomposites and bioenergy [54]. The diffraction patterns of non-pretreated and pretreated oil palm mesocarp fiber biomass and the diffraction patterns of commercial alpha cellulose are presented in Figure 9.  The diffraction peaks detected in the XRD pattern show high peaks at an angle of 2θ, namely at 15 o -16 o and 2θ = 20 o -22.5 o , which indicates that the oil palm mesocarp fiber biomass has a crystal structure of cellulose I and is a natural plant cellulose [55]. The diffraction peak at angle 2θ, about 22.5 o , is associated with the typical peak portion of the cellulose crystal structure. Whereas at an angle of 2θ, which is 15 o , is part of the amorphous region in the biomass. Rosli [56] stated that the diffraction peak of the cellulose crystal structure is at an angle of 2θ, which is around 22-23 o . In the biomass that was not pretreated, it showed a broad peak which was associated with the amorphous cellulose area. The crystallinity index (CrI) in mesocarp fibers can be calculated using the Segal method, which is summarized in Table 6. The crystallinity index (CrI) value of mesocarp fibers undergoing hybrid ultrasonic ozonation pretreatment increased due to the loss of amorphous structures associated with lignin and hemicellulose components. The crystal structure of cellulose is recalcitrant, so it is not easy to remove in the pretreatment process. The low crystallinity index (CrI) value of the raw material for palm oil mesocarp fiber is due to the lignin and hemicellulose components which are cross-linked and have an amorphous structure. Thus, increasing the crystallinity index (CrI) value may imply that pretreatment with the hybrid ozonation-ultrasonic method can degrade components with amorphous structures such as lignin and hemicellulose properly and efficiently. On the other hand, increasing the crystallinity index (CrI) can also encourage an increase in the mechanical properties of cellulose for advanced industrial use [37].

3.7.Fourier-Transform Infrared Spectroscopy (FTIR) Characterization of Biomass
Fourier-transform infrared spectroscopy (FTIR) analysis observed functional groups of oil palm mesocarp fiber before and after hybrid ozonation-ultrasonic pretreatment. The characteristic features of the FTIR spectrum of oil palm mesocarp fiber biomass are presented in Figure 10. The FTIR spectrum of cellulose in oil palm mesocarp fiber both before pretreatment and after pretreatment of lignin degradation produced different types and the number of absorption peaks indicating that the hybrid ozonation-ultrasonic pretreatment could destroy the structure of lignocellulosic biomass. The broad absorption peaks in the region from 3200 cm -1 to 3400 cm -1 are caused by the stretching of the O-H functional groups, which is the main characterization of cellulose [57,58]. Pretreated cellulose and commercial cellulose samples showed absorption peaks in the area of 2890 cm -1 , which indicated sp 3 stretching carbon in the methyl (CH3) and methylene (CH2) groups. In contrast, in that area, there were no absorption peaks for non-pretreated oil palm mesocarp fibers [59]. The absorption peaks in the 1638 cm -1 area and 1427 cm -1 area, respectively, represent the structure of the aromatic ring in lignin according to the C=C and -CH2 bonding functional groups. The intensity of this absorption peak was reduced in pretreated cellulose and commercial cellulose samples, so it can be concluded that the hybrid ozonation-ultrasonic pretreatment process can damage the structure of lignocellulosic biomass and degrade lignin [60,61]. The absorption peak in the 1315 cm -1 area represents a typical cellulose absorption peak with the CH2 rocking functional group. In this area, the intensity of the absorption peak increases in pretreated cellulose [54,62]. Increased intensity of absorption peaks in pretreated cellulose samples and commercial cellulose also occurs in the areas 1027 cm -1 and 557 cm -1 , which are the stretching of the βglycosidic bond, namely C-O-C stretching and the lowest C-H vibration, which refers to an increase in the value of cellulose crystallinity [63,64]. An analysis of Fourier-transform infrared spectroscopy (FTIR) results confirmed that the lignin composition and amorphous areas in lignocellulosic biomass were successfully degraded because similar trends were also shown in commercial cellulose.

Scanning Electron Microscopy (SEM) Characterization of Biomass
Changes in surface morphology and roughness of the lignocellulosic biomass of oil palm mesocarp fiber before and after pretreatment can be identified through scanning electron  Figure  11.
The SEM micrographs demonstrated that the surface of untreated mesocarp fiber lignocellulosic biomass had a dense and smooth structure without any cell wall cracks or fiber damage. It is shown by the red square, which indicates that the lignin structure was still present on the surface of the lignocellulosic biomass. However, in the biomass that was pretreated using the hybrid ozonation-ultrasonic method, as indicated by the red square in section (b), shows that the surface structure of the biomass experienced perforations such as damage caused by some of the degraded lignin structure and also due to shear forces and the sudden burst of cavitation bubbles that collided with the structure. Lignin in the ultrasonic process causes lignocellulosic biomass surface damage [65]. In commercial cellulose, it is also seen that the structure of lignocellulosic biomass is no longer dense and smooth but has been damaged, as shown by the red square in section (c). It is because lignin degradation has occurred, and some of the structure has been removed.

CONCLUSION
The hybrid ozonation-ultrasonic pretreatment process for lignin degradation in oil palm mesocarp fiber biomass shows the potential for application, with the acquisition of a high percentage of lignin degradation, up to 98.02%. The RSM was used based on the BBD to obtain the highest percentage of lignin degradation. The optimum conditions were obtained at 30°C, with a flow rate of 2 L/min for 60 minutes. A decrease in the absorption peak of the aromatic ring structure in the FTIR spectrum confirms the occurrence of a lignin degradation process. It was validated by an increase in the crystallinity index (CrI) of mesocarp fiber biomass subjected to a hybrid ozonation-ultrasonic pretreatment process and reinforced by changes in the surface structure of lignocellulosic biomass as shown in the results of SEM micrographs. Therefore, the results of this study can be used as a reference to increase the percentage of lignin degradation in lignocellulosic biomass.