Synthesis and characterization of ground biochar (GB) reinforced composites for removal of heavy metal from palm oil mill effluent (POME)

Rahman, M. R., James, A. A., Marwani, H. M., Hamdan, S. Said, K. A. M., Bakri, M. K., Uddin, J., Matin, M. M., Madkhali, O., Aljabri, M. D., and Rahman, M. M. (2023). “Synthesis and characterization of ground biochar (GB) reinforced composites for removal of heavy metal from palm oil mill effluent (POME),” BioResources 18(3), 5548-5573.

Abstract

Heavy metal contamination ruins the ecosystem and water quality. The adsorption method for heavy metal remediation is preferred because of its low cost and high efficiency. This work created eco-friendly ground biochar (GB) biomass-based derivatives reinforced polylactic acid (PLA) with titanium dioxide (TiO₂). The composites improved palm oil mill effluent (POME) conditions, and H₂SO₄ activation increased pores by 80%. PLA and TiO₂ altered GB characteristics, according to FTIR analysis. A significant adhesion interaction showed that GB, PLA, and TiO₂ particles were compatible. Ball milling’s shear force increased surface area, according to Brunauer-Emmett-Teller (BET) research. Particle size reduction increased GB porosity. Scanning electron microscopy (SEM) was used to study the porous structure of GB and the synergistic effect of PLA and TiO₂ on POME during treatment. The SEM showed several components on the composite surface, demonstrating its efficacy. Atomic absorption spectroscopy (AAS) showed that sample C’s composite, which had the most GB, decreased POME heavy metals by 94.4% manganese (Mn), 88.4% cadmium (Cd), and 94.4% zinc (Zn). The resulting POME met the Malaysian Department of Environment’s POME discharge limit by reducing chemical oxygen demand (COD), total suspended solids (TSS), turbidity, and pH.

Download PDF

Full Article

Synthesis and Characterization of Ground Biochar (GB) Reinforced Composites for Removal of Heavy Metal from Palm Oil Mill Effluent (POME)

Md Rezaur Rahman,^a,* Anthonette Anak James,^a Hadi M. Marwani,^bSinin bin Hamdan,^a Khairul Anwar bin Mohamad Said,^a Muhammad Khusairy Bin Bakri,^a,c Jamal Uddin,^dMohammed Mahbubul Matin,^e O. Madkhali,^f Mahmood D. Aljabri,^g and Mohammed M. Rahman ^b,*

DOI: 10.15376/biores.18.3.5548-5573

Keywords: POME; Heavy Metals; Adsorption; Nano-activated Carbon; Composites

Contact information: a: Department of Chemical Engineering and Energy Sustainability, Faculty of Engineering, Universiti Malaysia Sarawak (UNIMAS), 94300 Kota Samarahan, Sarawak, Malaysia; b: Department of Chemistry, King Abdulaziz University, Jeddah 21589, Saudi Arabia and Center of Excellence for Advanced Materials Research (CEAMR), King Abdulaziz University, Jeddah 21589, Saudi Arabia; c: Composite Materials and Engineering Center, Washington State University, 2001 East Grimes Way, 99164, Pullman, Washington, U.S.A.; d: Department of Natural Science, Coppin State University, Science and Technology Center, Baltimore, Maryland, 21216, U.S.A.; e: Bioorganic and Medicinal Chemistry Laboratory, Department of Chemistry, Faculty of Science, University of Chittagong, 4331 Chittagong, Bangladesh; f: Department of Physics, College of Science, Jazan University, P.O. Box 114, Jazan 45142, Kingdom of Saudi Arabia; g: Department of Chemistry, University College in Al-Jamoum, Umm Al-Qura University, Makkah 21955, Saudi Arabia; *Corresponding author: rmrezaur@unimas.my

INTRODUCTION

Based on the current global production volume, Malaysia is the second largest palm oil producer, with one-third of the world’s palm oil (Hirschmann 2022; Shahbandeh 2022). The increased demand for palm oil has increased the generation of palm oil mill effluent (POME). There are several challenges regarding the discharge of POME to water bodies. Such effluents often do not comply with the Environmental Quality Act 1974 (EQA 1974) discharge limit (Zahari et al. 2017; Kamyab et al. 2018; Mosunmola and Olatunde 2020; Mohammad et al. 2021). Although POME is classified as a non-toxic industrial effluent, raw POME contains a high level of chemical oxygen demand (COD), biochemical oxygen demand (BOD), organic and inorganic contaminants, as well as suspended solids, which can pollute the environment and harm the aquatic ecosystem (Lim et al. 2020; Mohammad et al. 2021). In Malaysia, most POME treatments use the traditional ponding system because it has low capital and operating costs. However, this method raises concerns about the low treatment effectiveness and the required large surface area (Kaman et al. 2017).

Biodegradable and sustainable adsorbents have the potential to reclaim a higher quality of treated POME. The carbonaceous adsorbent is an eco-friendly material that has a great pore structure with a large surface area, allowing them to remove high percentages of organic and inorganic contaminants (Ahmad and Azam 2019). Jahromi and Ghahreman (2019) discovered that acid pre-treatment improves adsorption properties by increasing adsorbent porosity and surface area. One of the advantages of carbonaceous adsorbent is that it offers regeneration and reuse for multiple cycles in adsorption processes (Alsawy et al. 2022; Baskar et al. 2022). The regeneration and reuse approach of adsorbents plays a vital role in improving the cost-effectiveness and sustainable approach that aligns with the Sustainable Development Goals (SDGs). Although the concept of adsorbent regeneration and reusability appears promising, more research and validation are required to assess the economic feasibility of adsorbent regeneration, especially on the operational cost and energy requirements. In addition, there must be a plan for how to deal with concentrated brines collected during the regeneration process. Studies have considered the effect of the regeneration cycle on the adsorbent properties and potential savings compared to using new adsorbents before assessing the overall performance of adsorbent regeneration (Moosavi et al. 2020; Alsawy et al. 2022; Santos et al. 2022).

The search for effective natural-based adsorbent materials has consistently been a vibrant field of study. Several studies have demonstrated the effectiveness of carbonaceous-based composites as adsorbents in removing heavy metals (e.g., cadmium (Cd), lead (Pb), chromium (Cr), mercury (Hg), arsenic (As), zinc (Zn), iron (Fe)) from industrial effluent via adsorption (Gul et al. 2021; Kumar et al. 2021; Mahesh et al. 2022). However, the existing studies still show a lack of information regarding the adsorption of various heavy metals from POME using composites derived from polylactic acid (PLA) and titanium dioxide (TiO₂) with ground biochar-adsorbent to overcome the limitation of conventional adsorbents where the interaction between the process parameters has been restricted. This research aims to create ground biochar (GB) carbonaceous adsorbent from pine sawdust via a pyrolysis process and sulphuric acid (H₂SO₄) activation. The ball milling process is performed to reduce the particle size of the adsorbent for better porosity development and a larger surface area. The synergic effects of GB-adsorbent-reinforced PLA with TiO₂ photocatalytic behaviour were examined by looking at the efficacy of the adsorption process. PLA was chosen due to its biodegradable derived properties, while TiO₂ can help minimize the brittleness, fading, and cracking that can occur in plastics and other materials as a result of light exposure. TiO₂ also has high heavy uptake, chemical stability, wide bandgap, high refractive index, nontoxic, and high catalytic activity. The analyses included Fourier transform infrared spectroscopy (FTIR), scanning electron microscopy (SEM) with energy dispersive X-ray spectroscopy (EDX), and the Brunauer-Emmett-Teller (BET) method. Together, these methods made it possible to observe the functional groups, morphology, elemental composition, and porosity of the composite. Furthermore, the study also employed atomic absorption spectroscopy (AAS) to observe the performance of the composites in eliminating the pollutants from raw POME. In addition, in-depth analysis and comparison are also performed to validate the performance of the prepared composites against other results.

EXPERIMENTAL

Materials

The poly(lactic) acid (CAS Number: 26100-51-6), titanium (IV) oxide (CAS Number: 13463-67-7), sulphuric acid (CAS Number 13813-19-9), and sodium bicarbonate (NaHCO₃) (CAS Number 144-55-8) were supplied by Sigma-Aldrich Sdn (Petaling Jaya, Malaysia). The pine sawdust was provided by Vyatka State University, Kirov, Russia. The POME was obtained from FELCRA, Kota Samarahan, Sarawak.

Methods

Figure 1 displays a flowchart outlining the various steps taken during the experiment. The subsequent elaboration of each step is described on the following page.

Pine sawdust adsorbent preparation

The pine sawdust was imported from Russia by Vyatka State University, Kirov, Russia. It was washed thoroughly with deionized water to remove excess dirt and impurities. Next, the samples were kept in an oven at 50 °C overnight. Later it was stored in an airtight container. The sample was ground and sieved to reduce the particle size between 0.38 to 0.3 mm to improve the activation reaction (Senturk and Yildiz 2020).

Pre-treatment of pine sawdust

First, 20 g of sample was pre-treated and activated with 200 mL of H₂SO₄ acid solution in a 500 mL conical flask. The mixture was then heated in a muffle oven at 150 °C for a period of 24 h. Once the carbonization was completed, the sample was washed with deionized water and soaked overnight in a 1% NaHCO₃ solution to reduce the sample acidity (Senturk and Yildiz 2020). The sample was filtered using vacuum filtration to obtain the residue and then washed using deionized water several times until it reached pH 7. The residue was distributed evenly on a plate to ensure equal heat distribution from the oven. The operating conditions were the same as before.

Fig. 1. Overview of the experimental steps

Carbonization of pine sawdust

Upon carbonization, the sample was first placed into a ceramic bowl wrapped with aluminium foil and then covered with a layer of clay before putting it in LT Furnace (Selangor, Malaysia). The temperature was ramped from room temperature to 200 °C, then increased to 400 °C at a heating rate of 10 °C/min for 2 h, during which the carbonization process occurred. After the carbonization process, LT Furnace was cooled at a cooling rate of 6.16 °C/min until it reached room temperature. The sample was collected the next day and further sieved using a 300 µm sieve. The sample was labelled as biochar sample.

Preparation of ground biochar (GB) via ball milling process

Approximately 10 g of biochar were ground for 10 min at a speed of 250 rpm using a grinder. The biochar was poured into a half-capacity cylindrical steel ball mill jar. Ten stainless-steel balls weighing 0.5 g each were used to reduce the particle size of biochar in the Planetary Ball Mill (PM 400) jar. The process took about 30 h at a speed of 600 rpm, and the sample was removed for testing at regular intervals to ensure it reached the nanoscale. The sample was labelled as ground biochar (GB) sample.

PLA/TiO₂/GB composites preparation

The PLA, TiO₂, and GB were mixed based on the composition weightage tabulated in Table 1.

Table 1. Composites Weightage Factors

Application of PLA/TiO₂/GB composites with POME via solution mixing

The POME was collected from Felcra Samarahan Jaya Palm Oil Mill, Kota Samarahan, Sarawak, and the initial conditions of POME, such as pH, COD, turbidity, and TSS, were evaluated. In each conical flask, 100 mL of POME was poured and added to the sample. Each conical flask was labelled as samples A, B, and C. The samples were placed in the auto-shaker for 48 h of contact time at 180 rpm at room temperature. After the end of the contact time of 48 h, the samples were removed from the auto-shaker and were filtered using 0.55 µm filter paper with a vacuum pump to obtain the residues for further analysis.

Characterizations

Fourier transform infrared (FTIR) spectroscopy

The functional groups of the samples were examined using an IRAffinity-1 FTIR spectrophotometer (Shimadzu, Kyoto, Japan) via the observation of IR spectrum bands with 32 scannings ranging from 400 to 4000 cm^-1. First, a sample pellet for FTIR spectroscopy was made using a 1:5 ratio of the specimen to potassium bromide (KBr) powder on a small pestle. The pelletized sample was extracted and placed inside the FTIR spectroscopy sample holder. The FTIR analysis was carried out following ASTM E1252-98 (2021) and ASTM E168-16 (2016) standards.

Scanning electron microscopy (SEM)

The morphological images of the samples, especially the pore structure, were evaluated via a Hitachi analytical tabletop SEM (benchtop) (TM-3030, Hitachi High Technologies, Mannheim, Germany). The specimens were mounted on aluminium stubs and were fine-coated using auto fine coater JFC-1600 (Jeol, Tokyo, Japan). The morphological images of the sample were captured using a field emission gun with an acceleration voltage of 5 and 15kV. The procedure for the SEM analysis was conducted following the ASTM E2015-04 (2014) standards.

Energy X-Ray Dispersive Spectroscopy (EDX/EDS)

EDX analysis (Quantax75TM Series Energy Dispersive X-Ray Spectrometer by Hitachi Ltd., Tokyo, Japan) was used to determine the elementary material composition present on the surface of the samples, which was correlated with SEM images produced for the respective sample.

Brunauer-Emmett-Teller (BET)

The BET analyzer was used to measure the specific surface area of each sample based on the ASTM D6556-14 (2014) standard. The degassing condition of the sample was 1 to 3 h at 300 °C. Several repeated tests were carried out to obtain the most representative outcomes. The micropore volume was calculated using the Dubinin-Radushkevich (DR) equation, while the pore volume was calculated using the plotted graph of N₂ adsorption isotherms of biochar and ground biochar. The equations are as follows (Tran et al. 2017).

Total pore volume (V_Total) is given by Eq. 1,

(1)

Average pore diameter (L₀), is given by Eq. 2,

(2)

The Dubinin-Radushkevich (DR) equation for characteristic adsorption energy (E₀) is presented as Eq. 3,

(3)

Gaussian distribution for micropore volume estimation is given by Eq. 4,

(4)

where (5)

In the foregoing equations, V_Total is the total pore volume (cm³g^-1); Q_0.99 is the quality absorbed volume (cm³g^-1STP) at 0.99 of the relative pressure ; S_BET is the BET specific surface area (m²g^-1); L₀ is the average pore diameter (nm); E₀ is the characteristic adsorption energy (kJmol^-1); V_Micro is the micropore volume (cm³g^-1); V_TotalMicro is the total micropore volume (cm³g^-1); and E is based on the Polanyi equation (Buttersack et al. 2016).

Atomic absorption (AAS) spectroscopy

The heavy metal concentrations (Mn, Cd, Zn, Fe) in raw and final POME were investigated using a flame AAS ‘AA-7000’ spectrometer (Shimadzu, Japan). The calibration solution was obtained from Sigma-Aldrich Sdn (Petaling Jaya, Malaysia). The observations were made according to the electromagnetic radiation wavelength emitted by the light source produced by the equipment. The test was done according to the ASTM D7740-20 (2020) standard.

Chemical oxygen demand (COD)

A total of 2.0 mL of POME solution and distilled water were added to two different vials. The distilled water was spared as a “blank sample”. The caps of the vials were closed tightly, and both were shaken gently to mix the samples. The vials were placed in a COD digester (DRB200 Reactor) for heating. The temperature was set at 150 °C for 2 h. The vials were removed after 2 h from the COD digester and were cooled at room temperature. The COD values were measured using HACH DR900 and recorded. It is noted that potassium dichromate (K₂Cr₂O₇) in combination with boiling sulfuric acid (H₂SO₄) has been commonly used as an oxidant in COD assays. This test was done according to ASTM D1252-06 (2020).

Total Suspended Solid (TSS) and Turbidity Test

First, 10 mL of deionized water was poured into a sample cell and served as a blank sample to calibrate the TSS and turbidity reading to 0.00 mg/L and 0.00 FAU. After calibration, 10 mL of raw POME was poured into a sample cell. Before inserting the sample cell into the cell holder, it was thoroughly cleaned. The “read” button was pressed to measure the TSS and turbidity value of the sample. This procedure was repeated for all treated POME samples. This test was done according to ASTM D5907-18 (2018).

RESULTS AND DISCUSSION

Infrared Spectral Properties

The FTIR analysis presented in Fig. 2 shows the presence of functional groups in both biochar and ground biochar (GB), representing their complex characteristics. A strong and sharp peak was observed at 3741.9 cm^-1, indicating strong O-H stretching bands in both biochar and ground biochar samples (Nyuk Khui et al. 2020, 2021). According to Senturk and Yildiz (2020), this finding was indeed due to the presence of O-H groups in lignocellulosic materials. It corresponded to water molecules containing O-H hydroxyl groups in corn stalk biomass, characterizing peaks between 4000 cm^-1 and 3584 cm^-1 (Salema et al. 2017). The medium peaks appearing at 2956.9 cm^-1 (biochar), 2951.1 cm^-1 (GB), 2544.1 cm^-1 (biochar), and 2549.9 cm^-1 (GB) showed the presence of CH stretching vibration in the methyl and methylene groups. This finding was supported by the rising of symmetric and asymmetric peaks at 2850 cm^-1and 2918 cm^-1 in the FTIR analysis of Leucaena leucocephala bark, which refers to the methoxyl groups of lignin (Salim et al. 2021).

Senturk and Yildiz (2020) and Wijaya and Wilharto (2020) stated that the alkane functional groups of cellulose, hemicellulose, and lignin introduced C-H bending vibrational modes giving rise to absorbances between 2500 and 3000 cm^-1. The peak of 1591.3 cm^-1observed in both samples indicated the presence of lignin in pine sawdust, in line with the observations of Wijaya and Wilharto (2020) in cocoa shell pyrolysis and Abel et al. (2021) in biochar samples.

The peaks at 1255.7 cm^-1 (GB) and 1207.4 cm^-1(biochar) represented the depolymerization and dehydration of hemicellulose and cellulose in the respective samples. The C≡C stretching vibration that specified the alkyne was identified at peaks of 2166.1 and 2164.1 cm^-1 of biochar and GB. The peaks between 1400 cm^-1 and 1750 cm^-1 presented the features of C=C and C=O of an aromatic ring, which also appeared at 1442 cm^-1 (Feng et al. 2014) and 1587 cm^-1 (Alfattani et al. 2022). Carboxylic acids, esters, and alcohol have been introduced by absorption bands at 1255.7 and 1207.4 cm^-1, distinguishing the C-O functional groups. Senturk and Yildiz (2020) explained that these groups were generally found in the molecular structures of the biochar that had been activated by H₂SO₄. Previous researchers mentioned that the formation of peaks between 1250 and 1020 cm^-1 can also be attributed to the C-N stretching vibration indicating the presence of amines in the samples (Bilal et al. 2013; Salim et al. 2021).

According to Salim et al. (2021) and Ali and Latip (2020), the cyclic nature of the ether by forming an aliphatic C-O stretching band at 1030 cm^-1has been identified for both samples at an intensity of 1033.8 cm^-1. The absorbance band appears between 675 cm^-1 and 1000 cm^-1, corresponding to the C-H stretching vibration of alkenes, while the bands between 550 cm^-1 and 730 cm^-1 represent the C-Cl bending stretching. The results in Fig. 2 indicated that the study of biochar and GB did not reveal any significant changes, as the spectra patterns appeared the same regardless of their transmittance intensities. This result was predicted by Nyuk Khui et al. (2020) due to the formation of new pores and increased surface area after the ball milling process. However, the ball milling does not affect the presence of functional groups in the sample, which shows similar effects to the chemical pre-treatment analysis.

Fig. 2. IR spectra of biochar (black-line) and GB (red-line)

Figure 3 depicts the FTIR spectra of titanium dioxide (TiO₂) and polylactic acid (PLA), while Fig. 4 illustrates the FTIR spectra of three different GB/ TiO₂ /PLA composites labelled as samples A, B, and C. The addition of TiO₂ and PLA to GB was found to influence band intensity characteristics. The functional groups present in the region of 695 to 760 cm^-1 of Fig. 4 were the features of C=O in PLA. PLA contains C=O groups because it was derived from lactic acid, which has a high carbon chain (Shaharuddin et al. 2019). However, the findings of Naayi et al. (2018) contradicted that interpretation, defining the adsorption peak between 430 cm^-1 and 781 cm^-1 as the Al-O stretching vibration band in alumina. It was detected in all samples at weak absorbance bands of 678.9, 761.9, and 763.8 cm^-1, as shown in Fig. 4.

The absorbance bands between 869.9 and 889.2 cm^-1 were attributed to the ring formation and deformation of C-H out-of-plane of lignin. A wide and broader peak of symmetric and asymmetric -CH₃ vibrations was revealed at peaks 2902.9, 2883.6, and 2933.7 cm^-1 in all samples. These peaks at 2995 and 2947 cm^-1 represent the alkyl stretching vibrations of C-H in the PLA polymer chain. A strong O-H stretching vibration was found at peaks of 3211.5 to 3493.1 cm^-1 in sample A, including 3498.9 to 3414 cm^-1 in sample B, indicating the presence of alcohol compound in the mixtures. In addition, the peaks at 3425.6 and 3498.9 cm^-1 also seem to be visible in sample C. Datta et al. (2016) stated that the peaks were caused by the increasing TiO₂ content in the prepared composites, as evidenced by the additional band at 3370 cm^-1 in their research. Another observed band at 1377.17 cm^-1 implies that the carboxylate compound occurred with medium vibration of stretching -C-O, as described by Sayuti et al. (2021). The medium vibrations at 3226.91 cm^-1 in sample A and 3234.62 cm^-1 in sample C were attributed to the stretching bond of N-H, which arise in the same region of alcohol O-H, similar to the previous findings by Kowalczuk and Pitucha (2019). A weak vibration found at peak intensities of 1728.25 cm^-1 and 1971.25 cm^-1 shows the aromatic compound properties of PLA. The existence of 1635.64 cm^-1 in sample C indicates a strong C=O vibration of aldehyde groups where a strong adhesion interaction between the activated carbon and PLA and TiO₂ particles has been observed. Table 2 provides the IR spectra summary of the relevant information obtained from the analysis of the samples.

Fig. 3. IR spectra of (a) TiO₂ and (b) PLA

Fig. 4. IR spectra of samples A, B, and C composites.

Table 2. IR Spectra Table by Frequency Range

Morphological and Porosity Properties

SEM was used to examine the samples to determine the surface morphologies of both biochar and ground biochar (GB). The SEM image of biochar shown in Fig. 5(a) at 400x magnification displayed multiple microfibrils with irregular sizes, showing open pores with diameters ranging from 15 to 20 µm, whereas Fig. 5(b) shows numerous appearances of small particulates, but they are too small to be seen clearly in the micrograph due to the micro-nano scales pores after ball milling. The outcome of the SEM analysis shown in Fig. 5(b) shows more pronounced particles’ deformation at 30 h of milling times, changing the sample morphology due to the shear force applied by the metallic balls on the surface of biochar particles in the mono planetary stainless-steel jar. This result followed other researchers who showed significant particle size reduction after periods of ball milling (Phanthong et al. 2018; Lyu et al. 2018; Qanytah et al. 2020). However, the particle size remained constant after 30 h of milling, as no further changes were observed. The result was similar to findings by Shan et al. (2016), who reported that coconut shell-based activated carbon particle size remained the same even after increasing the milling time. Qanytah et al. (2020) added that a longer milling process reduces the sample crystallinity, producing an amorphous sample, as displayed in Fig. 5. This may initiate agglomeration of biochar to form large particulates. Despite the disadvantage of a longer milling time, this approach can increase the adsorption capabilities of biochar by creating numerous pores on its surface (Hu et al. 2019), leading to an increase in reactivity as the volume ratio of the sample increases (Qanytah et al. 2020). Besides, it also increases the contact surface area and leaves numerous vacant pores (Qanytah et al. 2020; Abbas et al. 2021). The BET analysis in Table 3 reflected this assertion.

Fig. 5. SEM images at x400 magnification for a) biochar and b) GB

Table 3 shows the specific surface area, total pore volume, micropore volume, and average pore diameter of samples that were obtained from the N₂ adsorption isotherms graph in Fig. 6. According to the results, GB was fivefold higher in surface area than biochar, rising from 5.09 m²g^-1 to 25.8 m²g^-1. This result demonstrated the efficiency of ball milling in reducing particle size from the micro to the micro-nano scale (Nyuk Khui et al. 2020). Based on the comparison, GB particles have a higher specific surface area than biochar particles, indicating a better porosity of the composite material. The specific surface area of the sample was calculated using the amount of adsorbate gas corresponding to a monomolecular layer on the surface. The results showed that the GB sample had a significant micropore volume, accounting for approximately 90% of the total pore volume. This finding implies that the GB sample contained numerous micropores and some mesopores, which is similar to the findings of Hidayu et al. (2013).

Table 3. Specific Surface Area (S_BET), Total Pore Volume, Micropore Volume, and Average Pore Diameter of Samples

The nitrogen adsorption/desorption isotherms at 77 K on both biochar and GB at a degassing temperature of 300 °C are shown in Fig. 6. The trend line of the isotherm curves generated by the BET was determined by the pore characteristic and size presented on the surface of the adsorbent (Phothong et al. 2021). Although the adsorbents seem porous, the BET analysis result of biochar exhibited Type V isotherms, specifying a not common and weak interaction between absorbate-absorbate. In comparison, the GB trend line showed a gradually increasing trend to the end of relative pressure, displaying the characteristics of both Types I and V isotherms of microporous and mesoporous materials. In addition, the findings illustrated the hysteresis loop of Type H4 at relative pressures greater than 5, showing the formation of slit-like pores on the surface of GB. This discovery implies that the GB sample has been linked to capillary condensation in well-formed mesoporous structures (Allwar 2012).

Fig. 6. N₂ adsorption isotherms of biochar and ground biochar (GB)

The morphological images in Fig. 7 present the distribution of PLA/TiO₂/GB composites on the surface of each sample. As per observations, all samples had scattered and irregular shape particles with small white flakes referring to TiO₂nanoparticles. The TiO₂ nanoparticles and GB particles on the PLA matrix clumped together and blocked the pores of the mixture composites because of poor dispersion issues. The high surface energy and excessive TiO₂ loading also prompted agglomeration problems (Zain et al. 2021; Xing et al. 2016). It is said that TiO₂nanoparticles were easier to trap in the pores of activated carbon due to their small size (Matteis et al. 2020). Saffar et al. (2014) added to the previous statement that aggregation was also initiated by the O-H hydrophilic groups adhering to each particle via the network of hydrogen bonds.

Additionally, the hydrophobic property of PLA affected the compatibility of mixtures due to the weak interaction. Similar circumstances were encountered by Raquez et al. (2013), where poor mixture dispersion was noticed when adding TiO₂ nanoparticles with PLA. However, according to Kaseem et al. (2019), chemically functionalizing TiO₂ nanoparticles or treating surfaces with alkyl amine and propionic acid can enhance homogeneous dispersion.

The morphological images of the composites shown in Fig. 8 have been evaluated after a 48-h continuous POME adsorption procedure. All sample surfaces shown in Fig. 8 contain impurities, which means that the sample micropores and macropores have been adsorbate-filled.

Fig. 7. SEM images at x1000 magnification for samples A, B, and C before POME treatment

Fig. 8. SEM images at x1000 magnification for samples A, B, and C after POME treatment

Elemental Composition Properties

EDS/EDX analysis

The EDX spectrum results displayed in Figs. 9, 10, and 11 show the presence of elements on the surface of the composites. The comparison of the elemental compositions for each sample based on the EDX peaks has been summarized in Tables 4, 5, and 6.

Fig. 9. EDX diffraction spectra of sample A (a) raw unused sample, (b) treated POME

Fig. 10. EDX diffraction spectra of sample B (a) raw unused sample, (b) treated POME

Fig. 11. EDX diffraction spectra of sample C (a) raw unused sample, (b) POME

Three main elements were identified in all samples, namely carbon (C), oxygen (O), and titanium dioxide (TiO₂). Similar to the findings of other researchers, carbon (C) had the highest appearance percentage on the surface composite, followed by oxygen (O). Alam et al. (2021) recorded 83% of C and 13% of O in the EDX analysis of activated carbon derived from Paulownia Tomentosa wood, indicating that C and O were the predominant elements in the respective sample. The results in Table 5 also revealed the presence of Al, which may have represented the impurities in raw sample B that resulted from ambient contaminants during sample preparation. The Al may have represented the impurities in raw sample B that resulted from ambient contaminants during sample preparation. This finding is reflected in the FTIR analysis of GB/TiO₂/PLA, where the appearance of Al-O stretching vibration has been observed. The AAS analysis provides additional insight into the synergistic impact of adsorption and photocatalytic degradation of the organic pollutants. By looking at the comparison of EDX outcomes tabulated in Tables 4, 5, and 6, it has been possible to identify the contaminants adhering to the samples’ pores shown in Fig. 8 of the SEM result after POME treatment. The C and O made up the highest element but at lower rates after POME treatment. As more pollutants were deposited, the percentages of C and O fell from their initial values, as recorded by all samples. The EDX results were comparable to those from Zain et al. (2021), demonstrating the decrease in C content due to particulate deposition on the surface of the composite. It signifies sample C’s superiority over samples A and B in removing pollutants from raw POME.

Table 4. EDX Comparison Analysis for Sample A

Table 5. EDX Comparison Analysis for Sample B