Carbon Composite Derived from Spent Bleaching Earth for Rubbery Wastewater Treatment

Abstract

The industrial production of palm oil generates substantial amounts of Spent Bleaching Earth (SBE), a waste byproduct from the bleaching process. In Malaysia and Indonesia, SBE is typically landfilled, causing environmental risks such as greenhouse gas emissions and contamination. Wastewater from the rubber industry also contains harmful pollutants that require effective treatment. This study proposes a sustainable solution by converting SBE into carbon composites (CCs) for treating rubber industry wastewater. Characterization of CCs using XRD, BET, FESEM, and FTIR revealed its porous structure, high surface area, and functional groups, contributing to excellent adsorption properties. Response Surface Methodology (RSM) optimized treatment conditions, determining 90.56 min of contact time and 0.75 g of adsorbent weight as optimal for maximum chemical oxygen demand (COD) and turbidity removal. Quadratic models showed R² values of 0.8828 for COD removal and 0.8336 for turbidity reduction, with numerical optimization achieving 90.30% COD reduction and 49.02% turbidity removal. Verification experiments confirmed model reliability with minimal deviation (0.37%). These findings demonstrate the potential of SBE-derived CCs as an eco-friendly solution for environmental challenges in the palm oil and rubber industries.

1. Introduction

The palm oil industry has undergone significant expansion over the past few decades, solidifying its position as a cornerstone of Malaysia’s economy. In 2023, crude palm oil production rose to 18.5 million tons, slightly exceeding the 18.4 million tons recorded in 2022 [1]. As a leading global producer, Malaysia plays a significant role in meeting the worldwide demand for palm oil, an essential ingredient in various food products, cosmetics, and industrial applications [2]. The palm oil refining process includes essential stages like bleaching, where bleaching earth is used to eliminate impurities such as color pigments, metal contaminants, phospholipids, and oxidized compounds [3]. However, this process generates Spent Bleaching Earth (SBE), a byproduct containing 20–40% residual oil and organic substances [4]. In 2022, with approximately 18.45 million tons of crude palm oil refined, around 185,000 tons of SBE were produced, as 5–10 kg of bleaching earth are required per ton of oil, highlighting the significant volume of waste generated by the industry.

The disposal of SBE poses a considerable environmental challenge. Landfilling, the primary method used in Malaysia, poses risks such as spontaneous combustion due to residual oil, greenhouse gas emissions, and soil contamination [4]. Consequently, the development of sustainable solutions to manage SBE is critical. One promising strategy is the regeneration and conversion of SBE into value-added products, such as adsorbents for wastewater treatment [5]. This approach not only addresses disposal issues but also aligns with circular economy principles by transforming waste into functional materials.

In parallel, the rubber industry, another pillar of Malaysia’s industrial landscape, generates vast amounts of wastewater during its processing activities. Reports indicate that approximately 80 million liters of untreated rubber effluent are discharged daily into nearby streams and rivers in Malaysia [6]. This wastewater is characterized by high concentrations of organic and inorganic pollutants, including nitrogen, sulfate, heavy metals, and various toxic compounds [7]. These contaminants can severely impact aquatic ecosystems and human health if discharged untreated. High levels of biochemical oxygen demand (BOD) and chemical oxygen demand (COD) in the effluent significantly reduce dissolved oxygen levels in receiving water bodies, creating hypoxic conditions that threaten aquatic life [8]. Excess nitrogen in the wastewater further exacerbates ecological disturbances by causing eutrophication, which stimulates excessive algal growth, disrupts the ecological balance, and intensifies oxygen depletion [9]. Additionally, high sulfate concentrations in the effluent, resulting from the use of sulfuric acid in latex coagulation, can lead to the release of hydrogen sulfide (H₂S). This compound not only generates malodorous emissions but also inhibits biological digestion processes in wastewater treatment systems, thereby diminishing their efficiency. The odor of H₂S is perceptible even at minimal concentrations, making water unpalatable far downstream from rubber processing factories [10]. Hence, it is imperative to implement effective and innovative treatment strategies to address the environmental and health challenges associated with untreated rubber wastewater.

Existing methods for treating rubber wastewater typically combine physical, chemical, and biological processes, including coagulation-flocculation, activated sludge systems, and advanced oxidation processes [11]. While effective to some extent, these approaches often suffer from limitations such as high operational costs, secondary pollutant generation, and sensitivity to variations in wastewater composition [12]. For instance, chemical treatments can produce residual chemicals, while biological processes often require long retention times and are less effective under shock loads. These limitations have prompted the search for alternative methods that are efficient and environmentally friendly. Among the alternatives, adsorption has emerged as a promising technique for wastewater treatment. This method offers simplicity, low operational costs, and high removal efficiency for a wide range of pollutants [13]. Carbon-based adsorbents, in particular, have shown significant potential, with studies reporting up to 96% zinc removal using palm shell activated carbon [14] and a 95% reduction in COD through the combined use of Fenton reagent and activated carbon [15], underscoring the potential of adsorption as a reliable and effective solution for treating rubber wastewater.

Carbon composites (CCs), a type of carbon-based adsorbent, represent a significant advancement in adsorption technology. These materials combine the adsorptive properties of carbon with the structural and functional benefits of other materials, resulting in enhanced mechanical strength, larger surface areas, and improved pollutant removal capabilities [16]. Recent studies have demonstrated the efficacy of CCs in various wastewater treatment applications. For instance, Saeidi, Parvini, and Niavarani [17] reported that a graphene/activated carbon composite achieved a maximum adsorption capacity of 217 mg/g for Pb(II) from wastewater, while Elessawy et al. [18] highlighted the ability of magnetic fullerene nanocomposites with functionalization to remove 99.61% of methylene blue from dye-contaminated wastewater. Additionally, magnetic carboxyl-functionalized attapulgite/carbon nanocomposites, derived from SBE, reached maximum adsorption capacities of 254.83 mg/g for methylene blue and 312.73 mg/g for Pb(II) [19]. Recently, Abu et al. (2024) [20] reported on the removal of ammonia from rubber wastewater using biochar derived from rubber sludge (RSB) and successfully removed 7.5 mg/l of ammonia from the rubber wastewater. On the other hand, Wang et al. (2024) [21] reported on the treatment of high-concentration antibiotic wastewater by using an iron–carbon composite combined with the Fenton oxidation process. They found that 66.9% of COD removal was achieved. While these studies focus on different types of wastewater, the findings highlight the versatility and potential of CCs for advanced treatment applications. Although substantial works have been reported on utilizing carbon-based materials in wastewater treatment as well as heavy metal removal, the use of CCs derived from SBE for rubber wastewater treatment, particularly, is hardly found in the literature and remains a relatively new area of research, highlighting the novelty of the present work. In fact, our research group has recently reported the continuous treatment of wastewater from a natural rubber factory using an attapulgite/carbon nanocomposite pellet derived from SBE [22].

This study aims to investigate the feasibility of utilizing CCs derived from SBE for the treatment of rubber wastewater. The large production and abundant of the SBE from the palm oil industries and the mass production of rubbery wastewater from rubber industries in Malaysia have motivated the selection of CCs derived from SBE as an adsorbent for the rubber wastewater treatment in this work. By exploring this waste-to-resource approach, the research seeks to address dual environmental challenges: the effective management of SBE and the treatment of rubber wastewater. The outcomes of this study could contribute to the development of sustainable treatment solutions, minimizing the environmental footprint of the rubber industry while promoting the regeneration and reuse of industrial byproducts. Such advancements align with global efforts to achieve sustainable industrial practices and safeguard ecological and human health.

2. Materials and Methods

2.1. Materials

The carbon composite (CC) used in this study was prepared by the Universitas Lampung, Indonesia. The preparation method of CCs is reported in our previous study [22]. Firstly, the SBE was ground and sieved with a 200-mesh sieve. Then, the sieved SBE was calcined at 300 °C for 2 h to obtain a CC sample. Rubber wastewater samples were collected from a rubber processing facility located at Perak, Malaysia. Then, the collected wastewater was diluted prior to experimental analysis.

2.2. Characterization of Carbon Composite (CC)

The characterization of the carbon composite (CC) sample involved multiple analytical techniques to assess its structural and surface properties. The crystalline structure was analyzed using X-ray diffraction (XRD), conducted on an X-ray diffractometer (Xpert3 Powder, Panalytical, Malvern, UK). The specific surface area was determined via the Brunauer–Emmett–Teller (BET) method using a surface area analyzer (ASAP 2020, Micromeritics, Norcross, GA, USA), while the pore size distribution was assessed using the Barrett-Joyner-Halenda (BJH) approach at 77 K. The surface morphology and nano-structural features were examined through Field Emission Scanning Electron Microscopy (FESEM, Clara, Tescan, Brno, Czech Republic) at an accelerating voltage of 5 kV. To prepare the sample for FESEM analysis, it was mounted on an aluminum stub using adhesive tape and subsequently coated with a thin layer of gold to mitigate charging effects during imaging. Furthermore, the functional groups present in the CC sample were identified through Fourier Transform Infrared (FTIR), performed using Frontier 01, Perkin Elmer spectrometer (Waltham, MA, USA).

2.3. Design of Experiments

Response Surface Methodology (RSM) was employed to evaluate the effects of operating parameters on chemical oxygen demand (COD) and turbidity removal from rubber wastewater using the carbon composite (CC) sample. This study was conducted based on a central composite design CCD with 13 experimental runs that included 5 center points and 8 cube points. The independent variables were contact time (30 to 90 min) and adsorbent weight (0.5 to 1.0 g), while the response variables measured were COD and turbidity removal efficiencies. Each adsorption experiment was conducted in triplicate to ensure reliability. The experimental data were modeled using statistical software to analyze the influence of contact time and weight through ANOVA results and three-dimensional response surface plots. Optimization was performed to identify the minimum contact time and adsorbent weight required to maximize the removal efficiency. The optimized conditions were validated with additional experiments, and the predicted removal efficiencies were compared with the actual results.

Table 1. Experimental runs suggested by the statistical software.

Run	Factor 1: Adsorbent Weight (g)	Factor 2: Contact Time (min)
1	0.5	90
2	0.75	60
3	1	60
4	0.5	60
5	0.75	60
6	0.75	30
7	1	90
8	0.75	60
9	0.75	90
10	0.5	30
11	0.75	60
12	1	30
13	0.75	60

summarizes the experimental design data in this work.

2.3.1. Determination of Turbidity Removal Efficiency

In the batch adsorption experiment, the procedure for each experimental run (refer to Table 1) involved mixing the carbon composite (CC) sample with 50 mL of diluted rubber wastewater in a 100 mL beaker using a volumetric pipette. A blank sample was prepared by adding only 50 mL of diluted wastewater to a beaker without the adsorbent. The initial turbidity of the blank sample was measured using a turbidimeter.

To begin each run, the mixture in the beaker was stirred using a magnetic stirrer. Following the adsorption process, the adsorbent was separated from the treated wastewater using vacuum filtration. The filtered mixture was then transferred into small vials using a clean syringe fitted with a filter, and the turbidity of the treated samples was measured using a turbidimeter.

The same procedure of mixing, equilibration, filtration, and turbidity measurement was repeated for all 13 experimental runs. The turbidity removal efficiency for each run was calculated using Equation (1) [23]

$$\mathrm{T}\mathrm{u}\mathrm{r}\mathrm{b}\mathrm{i}\mathrm{d}\mathrm{i}\mathrm{t}\mathrm{y} \mathrm{r}\mathrm{e}\mathrm{m}\mathrm{o}\mathrm{v}\mathrm{a}\mathrm{l} \mathrm{e}\mathrm{f}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{y} \left(\%\right)={\displaystyle \frac{\left(\mathrm{I}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l}-\mathrm{F}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{l}\right) \mathrm{t}\mathrm{u}\mathrm{r}\mathrm{b}\mathrm{i}\mathrm{d}\mathrm{i}\mathrm{t}\mathrm{y}}{\mathrm{I}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l} \mathrm{t}\mathrm{u}\mathrm{r}\mathrm{b}\mathrm{i}\mathrm{d}\mathrm{i}\mathrm{t}\mathrm{y}}}\times 100\%.$$

(1)

2.3.2. Determination of Chemical Oxygen Demand (COD) Removal Efficiency

For each sample, 2 mL of diluted rubber wastewater was pipetted into a chemical oxygen demand (COD) reagent vial using a clean 10 mL pipette at a 45-degree angle. For the blank, 2 mL of deionized water was added to a COD reagent vial using the same method. Both vials were then securely sealed, wiped clean with a paper towel, and gently inverted multiple times to ensure thorough mixing. The prepared vials were then placed in a preheated DRB200 reactor and heated for 2 h. After the reaction period, the reactor was turned off, and the vials were allowed to cool for approximately 15 min before being placed in a rack to reach room temperature. The COD concentration was measured using a colorimetric determination system set to the 435 COD HR program. To standardize the instrument, the blank sample vial was cleaned externally and inserted into the 16 mm cell holder, where the system was calibrated to zero, displaying a reading of 0.0 mg/L COD. Subsequently, the vial containing the rubber wastewater sample and reagent was placed in the cell holder, and the COD concentration was measured and recorded.

This procedure was repeated for all 13 experimental runs as specified in Table 1. COD removal efficiency was calculated using Equation (2) [24]

$$\mathrm{C}\mathrm{O}\mathrm{D} \mathrm{r}\mathrm{e}\mathrm{m}\mathrm{o}\mathrm{v}\mathrm{a}\mathrm{l} \mathrm{e}\mathrm{f}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{y} \left(\%\right)={\displaystyle \frac{\left(\mathrm{I}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l}-\mathrm{F}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{l}\right) \mathrm{C}\mathrm{O}\mathrm{D}}{\mathrm{I}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l} \mathrm{C}\mathrm{O}\mathrm{D}}}\times 100\%.$$

(2)

3. Results and Discussion

3.1. Characterization

3.1.1. X-Ray Diffraction (XRD)

Figure 1 illustrates the X-ray diffraction (XRD) pattern of the carbon composite (CC) sample. The diffraction patterns revealed the coexistence of amorphous and crystalline phases, as indicated by the presence of both weak peaks’ characteristic of amorphous regions and sharp peaks associated with crystalline structures. This dual-phase nature aligns with previously reported characteristics of carbon composites [25]. Prominent diffraction peaks identified at 2θ values of 20.8°, 21.7°, 26.6°, 36.5°, 39.5°, 59.9°, 62.3°, and 68.2° were attributed to SiO₂, likely originating from the silica content in the spent bleaching earth [26]. The distinct and well-defined nature of these peaks suggests that the high-temperature carbonization process did not significantly alter its crystalline phase [27]. Additionally, peaks observed at 2θ values of 6.2°, 35.5°, 43.3°, and 50.1° were associated with montmorillonite, a clay mineral commonly found in similar materials [26]. Several smaller diffraction peaks distributed across the spectrum further suggest the presence of additional crystalline minerals, along with residual ash content on the CC surface [25].

3.1.2. BET Surface Area

The Brunauer–Emmett–Teller (BET) method was employed to determine the surface physical characteristics of the carbon composite (CC) sample derived from spent bleaching earth (SBE).

Table 2. BET result of carbon composite (CC) sample.

Parameter	Result
BET surface area (m2/g)	33.6695
Pore volume (cm3/g)	0.1051
Pore size (nm)	12.8717

summarizes the BET analysis results, including specific surface area, pore volume, and pore size. The pore size of the CC sample was measured as 12.8717 nm, classifying it as a mesoporous material according to the International Union of Pure and Applied Chemistry (IUPAC) standards [28] (p. 2). Furthermore, the BET surface area and pore volume were found to be 33.6695 m²/g and 0.1051 cm³/g, respectively. Meanwhile, Figure 2 shows the N₂ adsorption–desorption isotherms curve of the CC sample. Referring to Figure 2, the CC sample exhibits Type IV (IUPAC) hysteresis loops of mesoporous material, which is consistent with the result reported in our previous work [22].

Notably, the CCs derived from SBE in this study exhibits significantly enhanced adsorption properties compared to the raw SBE material. For instance, the BET analysis of raw SBE reported by Yulikasari et al. [26] exhibited a surface area of 10.868 m²/g, a pore volume of 0.0056 cm³/g, and a pore size of 2.066 nm, all of which were substantially lower than those values obtained for CC sample. This enhancement can be attributed to the pyrolysis process, which improved the material’s porous structure and adsorption potential [25]. Moreover, the BET results of the CC prepared in this work are comparable to those values reported in the literature for CC samples. For instance, another carbon composite derived from spent bleaching earth demonstrated a surface area of 29.71 m²/g [25], which is lower than that of the CC prepared in this work. This indicates that the CC sample prepared in this work possesses a greater number of adsorption sites. Additionally, in terms of pore volume, the CC prepared in this work is competitive with other carbon composites, such as those derived from waste sawdust, which exhibited a pore volume of 0.1517 cm³/g [29]. These findings highlight the suitability of the CC prepared in this work as a potential adsorbent for treating wastewater.

3.1.3. Field Emission Scanning Electron Microscope (FESEM)

The surface morphology of the CC was examined using Field Emission Scanning Electron Microscopy (FESEM), and the results are shown in Figure 3. The micrographs of the CC sample at different magnifications revealed a highly porous and irregular surface structure, featuring extensive cavities and openings across the CC particles. This morphology was similar to that of carbon composites derived from spent bleaching earth (SBE), as reported by Ke et al. [25]. The formation of these porous and rough surface features is attributed to the carbonization process during the high-temperature treatment of SBE. Such a porous structure provides a large surface area which could facilitate the interaction between the adsorbent and adsorbate molecules.

3.1.4. Fourier Transform Infrared Spectroscopy (FTIR)

Figure 4 presents the Fourier Transform Infrared (FTIR) spectrum of the CC. The broad absorption band at 3435.37 cm⁻¹ is associated with O–H stretching vibrations, characteristic of hydroxyl groups [25]. This indicates the presence of surface-bound hydroxyl functionalities, which play a vital role in adsorption. The peak at 2925.65 cm⁻¹ corresponds to the asymmetric stretching vibration of the C–H bond [30]. The absorption band at 1630.87 cm⁻¹ is attributed to the bending vibration of O–H groups, with possible overlap from the C=C stretching and C–H bending vibrations in aromatic rings. A prominent peak at 1054.06 cm⁻¹ is the characteristic of Si–O–Si stretching vibrations, confirming the presence of silica, which originates from the SBE [31]. This silica content remains largely unaffected during the pyrolysis process, contributing to the structural stability and integrity of the CC material. Additional peaks at 796.87 cm⁻¹ and 468.34 cm⁻¹ correspond to Si–O vibrations, indicating the coexistence of crystalline and amorphous silica phases [19]. The band at 693.76 cm⁻¹ is attributed to the asymmetric stretching of aromatic C–H groups [32], while the absorption near 523.85 cm⁻¹ represents the bending vibration of Si–O–Al bonds [33]. These peaks collectively indicate the presence of SiO₂ and montmorillonite within the CC material.

3.2. Treatment of Rubber Wastewater

3.2.1. Visual Analysis of Carbon Composite (CC) Adsorption Performance

The adsorption performance of the carbon composite (CC) in treating rubber wastewater was evaluated visually. This approach is a widely accepted qualitative method for assessing the effectiveness of adsorbents [23]. As shown in Figure 5a, the untreated diluted wastewater exhibited a light blue color, indicative of high concentrations of contaminants such as organic matter and suspended solids contributing to the coloration. Following treatment with the CC, the effluent appeared clear and nearly colorless, demonstrating the effective adsorption and removal of color-causing compounds by using a CC sample. However, a slight pale grayish hue was observed in the treated effluent, as shown in Figure 5b. This residual coloration is attributed to the inherent dark color of the CC adsorbent. Moreover, the effluent underwent only a single filtration step, which further contributed to the residual coloration in the treated effluent.

3.2.2. Analysis of Statistical Model

A total of 13 experimental runs were conducted using the face-centered central composite design (CCD) method as suggested by the statistical software. The experimental conditions involved varying the adsorbent weight between 0.5 g and 1 g and the contact time from 30 to 90 min. The results of these experimental runs are presented in

Table 3. Results of 13 experimental runs.

Run	Factor 1: Weight (g)	Factor 2: Time (min)	Response 1: COD Removal Efficiency (%)	Response 2: Turbidity Removal Efficiency (%)
1	0.5	90	81.34	50.70
2	0.75	60	81.07	38.31
3	1	60	83.80	46.20
4	0.5	60	84.42	47.48
5	0.75	60	80.90	41.41
6	0.75	30	84.24	33.24
7	1	90	89.88	46.48
8	0.75	60	68.75	31.27
9	0.75	90	91.29	48.45
10	0.5	30	90.32	47.32
11	0.75	60	86.71	40.85
12	1	30	81.69	37.18
13	0.75	60	89.61	36.06

.

The statistical analysis of variance (ANOVA) results for chemical oxygen demand (COD) removal, as shown in

Table 4. ANOVA result and fit statistics for Response 1: COD removal efficiency.

ANOVA Result
Source	Sum of Squares		df	Mean Square		F-Value	p-Value		Remarks
Model	336.28		5	67.26		5.05	0.0281		Significant
A: Weight	23.84		1	23.84		1.79	0.2229
B: Time	40.04		1	40.04		3.00	0.1266
AB	226.65		1	226.65		17.01	0.0044
A2	40.21		1	40.21		3.02	0.1260
B2	0.0575		1	0.0575		0.0043	0.9494
Lack of fit	16.93		3	5.64		0.2957	0.8277		Not significant
Pure error	76.37		4	19.09
Cor total	429.58		12
Fit statistics
R2		0.8828			Adjusted R2			0.7277
Adeq precision		8.1531			Predicted R2			0.7513

, were conducted to evaluate the significance of the process parameters and validate the empirical model generated using Response Surface Methodology (RSM). A quadratic model was identified as the best fit for COD removal, with a significant model F-value of 5.05. This indicates that the model is statistically significant, with only a 2% likelihood that the observed F-value occurred due to random noise. The p-value of 0.0281 (<0.05) confirms the overall significance of the model [34]. While the individual factors of adsorbent weight (A) and contact time (B) demonstrated p-values of 0.2229 and 0.1266, respectively, which are greater than 0.05, their interaction (AB) showed a p-value of 0.0044, indicating statistical significance. This suggests that the combined effect of adsorbent weight and contact time significantly influences the COD removal, implying that the impact of one parameter depends on the level of the other. Moreover, an R² value of 0.8828 demonstrates that the model accounts for 88.28% of the variability in COD removal. The predicted R² value of 0.7513 of the quadratic model was within 20% of the adjusted R² value of 0.7277, indicating good agreement between these values and confirming that the model is not overfitted. Additionally, an adequate precision value of 8.1531, exceeding the threshold of 4, confirms an acceptable signal-to-noise ratio, supporting the model’s reliability for navigation within the design space [35].

Building on the COD removal analysis, the ANOVA results for turbidity removal, as presented in

Table 5. ANOVA result and fit statistics for Response 2: turbidity removal efficiency.

ANOVA Result
Source	Sum of Squares		df	Mean Square		F-Value	p-Value		Remarks
Model	390.71		5	78.14		7.01	0.0119		Significant
A: Weight	52.45		1	52.45		4.71	0.0667
B: Time	111.89		1	111.89		10.04	0.0157
AB	47.61		1	47.61		4.27	0.0776
A2	0.4142		1	0.4142		0.0372	0.8526
B2	146.47		1	146.47		13.14	0.0085
Lack of fit	50.64		3	16.88		2.47	0.2018		Not significant
Pure error	27.38		4	6.84
Cor total	468.73		12
Fit statistics
R2		0.8336			Adjusted R2			0.7147
Adeq precision		6.8502			Predicted R2			0.6389

, further validate the significance of the process parameters and the reliability of the empirical model. The analysis identified a quadratic model for turbidity removal, with a significant model F-value of 7.01 and a corresponding p-value of 0.0119, indicating the model’s statistical significance [34]. This suggests that the model is effective in describing the relationship between the process parameters and turbidity removal. Among the individual factors, contact time (B) was statistically significant, with a p-value of 0.0157 (<0.05), while adsorbent weight (A) had a p-value of 0.0667, which is slightly above the 0.05 threshold, suggesting a moderate influence on turbidity removal. The interaction term (AB) had a p-value of 0.0776, indicating that the interaction between weight and contact time did not significantly impact turbidity removal. However, the quadratic term for contact time (B²) was significant, with a p-value of 0.0085 (<0.05), highlighting its importance in the model [36]. In addition, an R² value of 0.8336 demonstrates that the model can explain 83.36% of the variability in turbidity removal, indicating a good fit. The adjusted R² value of 0.7147 is closer to the predicted R² value of 0.6389, indicating good agreement between these values and confirming that the model is not overfitted. Furthermore, the adequate precision value of 6.8502 exceeds the threshold of 4, confirming an adequate signal-to-noise ratio for the model’s application within the design space [35].

Figure 6 and Figure 7 display the three-dimensional surface plots for COD and turbidity removal efficiencies, respectively, illustrating the combined effects of adsorbent weight (A) and contact time (B). The maximum COD removal efficiency, approximately 91%, was observed at higher adsorbent weights (0.9–1.0 g) and longer contact times (90 min). This behavior aligns with adsorption theory, as increasing the adsorbent weight provides more active adsorption sites for pollutant molecules, while longer contact times allow sufficient interaction between the adsorbent and adsorbate, facilitating the attainment of equilibrium [37]. However, at shorter contact times (30–60 min), increasing the adsorbent weight resulted in a slight reduction in COD removal efficiency, as shown by the curvature of the surface plot. This phenomenon may be attributed to site saturation or diffusion limitations, where the adsorbate molecules were unable to effectively access all the adsorption sites within the limited time frame [38].

In Figure 7, which represents turbidity removal efficiency, the plot shows that turbidity removal increased with contact time, peaking at approximately 50% efficiency at 80 min and adsorbent weights between 0.9 and 1.0 g. Beyond 80 min, a decline in removal efficiency was observed, which could be attributed to the dark coloration of the adsorbent due to pyrolysis, potentially contributing to increased turbidity and offsetting removal efficiency gains. At shorter contact times, increasing the adsorbent weight had minimal impact on turbidity removal, likely because there was insufficient time for adsorbate particles to interact with the adsorbent surface [39]. However, after 60 min, increasing the adsorbent weight resulted in a clear improvement in turbidity removal efficiency. This trend is consistent with adsorption theory, which suggests that a higher amount of adsorbent offers more active sites for adsorbate attachment, thereby enhancing removal efficiency [35].

The observed trends in both plots are consistent with the ANOVA results, which indicated that COD removal efficiency is influenced by the interaction between adsorbent weight and contact time, whereas turbidity removal efficiency is primarily impacted by contact time.

3.3. Numerical Optimization Study

A numerical optimization study was conducted to determine the optimal contact time and adsorbent weight for maximizing the removal efficiencies of chemical oxygen demand (COD) and turbidity.

Table 6. Optimization conditions generated using statistical software.

Run	Weight	Time	COD	Turbidity	Desirability
1	0.750	90.560	90.296	49.017	0.958
2	1.000	90.000	90.124	48.448	0.953
3	1.000	89.646	90.173	48.517	0.919

presents three potential solutions with high desirability scores. Among these, the model identified the optimal conditions as a contact time of 90.56 min and an adsorbent weight of 0.75 g, predicting removal efficiencies of 90.30% for COD and 49.02% for turbidity. To validate the model’s predictions, triplicate experiments were performed under optimized conditions, and the experimental results were compared to the values predicted by the model. The average experimental COD removal efficiency was 90.67%, closely matching the predicted value of 90.30%, with a minimal deviation of 0.37%. Similarly, the average experimental turbidity removal efficiency was 49.39%, slightly exceeding the predicted value of 49.02%, with the same deviation of 0.37%. These minor deviations are well within acceptable limits and can be attributed to experimental variability, measurement inaccuracies, and the inherent limitations of empirical modeling. The close agreement between experimental and predicted results highlights the robustness and predictive accuracy of the RSM model, confirming its reliability in optimizing the adsorption process for treating rubber wastewater.

3.4. Comparison with Reported Literature

Table 7. Comparison of COD and turbidity removal efficiencies for rubber wastewater with various materials via adsorption method.

Material	Removal Efficiency (%)		References
	COD	Turbidity
Carbon composite (CC) derived from spent bleaching earth (SBE)	90.3	49.2	Present work
Combination of Fenton reagent and activated carbon	95.0	-	[15]
Activated carbon prepared from Delonix Regia pods	70.7	-	[34]
Bentonite granules	37.5	-	[40]
Moringa oleifera stem bark	80.6	98.2	[41]

highlights the comparison of chemical oxygen demand (COD) and turbidity removal efficiencies achieved by various materials via the adsorption method for rubber wastewater treatment. The carbon composite (CC) derived from spent bleaching earth (SBE) in this study demonstrated removal efficiencies of up to 90.30% for COD and 49.02% for turbidity. While direct comparisons to some of the literature values are unavailable, the CC’s performance stands out, particularly given its waste-derived and sustainable origin. For instance, it shows competitive COD removal compared to activated carbon prepared from Delonix Regia pods [34] and bentonite granules [40], which achieves efficiencies of 70.7% and 37.5%, respectively.

Notably, the CC was tested on real rubber wastewater, which poses greater treatment challenges compared to synthetic wastewater commonly used in some literature studies. This adds to the significance of the CC’s performance, as it effectively treats a complex and difficult-to-handle wastewater matrix. Moreover, the CC offers the advantage of being derived from spent bleaching earth, a waste material, thus aligning with principles of sustainability and circular economy. In addition, the CC sample can be potentially applied for the uptake of iodine from liquid radioactive waste in the nuclear industry and hospital wastewater [42], as an environmental remediation solution.

4. Conclusions

In conclusion, this study successfully demonstrated the potential of carbon composites (CCs) derived from spent bleaching earth (SBE) as an effective adsorbent for treating rubber wastewater. The characterization of CCs revealed its porous structure with a significantly enhanced surface area, as confirmed by BET and FESEM analyses. XRD analysis identified a dual-phase composition of amorphous and crystalline structures, while FTIR analysis confirmed the presence of functional groups essential for adsorption. These findings establish the CCs as a high-performing adsorbent well-suited for the efficient treatment of rubber wastewater.

The application of CCs in treating rubber wastewater achieved remarkable results. Visual analysis confirmed the effective removal of color-causing compounds, while statistical validation through ANOVA highlighted the importance of key parameters such as adsorbent weight and contact time. The quadratic model for COD removal achieved a strong predictive accuracy with an R² value of 0.8828, and the turbidity model demonstrated an R² value of 0.8336. The interaction between adsorbent weight and contact time emerged as a critical factor for optimizing COD removal, whereas contact time alone was found to be crucial for turbidity removal. Numerical optimization indicated that the CCs achieved a maximum COD reduction of 90.30% and turbidity removal of 49.02% under optimal conditions, which included a contact time of 90.56 min and an adsorbent weight of 0.75 g. Verification experiments demonstrated the reliability of these models, with a minimal deviation of only 0.37% for both COD and turbidity predictions. These results highlight the potential of CCs as an efficient and environmentally sustainable solution for addressing the challenges associated with untreated rubber wastewater. Nevertheless, although the use of CCs from SBE offers advantages in terms of environmental benefits by recycling waste material, the reusability and regeneration of CCs remain challenging. In fact, from a real-world ramifications point of view, the CCs derived from SBE have the potential to be scaled up for various industrial usages, for example, in energy storage, construction, filtration etc., after resolving issues including economic feasibility and technological advancements in the regeneration process. Therefore, further research is crucial to address the limitations of CCs, such as chemical stability and performance consistency, to ensure that the CCs can be used in a sustainable manner.