Activated banana peel macrocomposite adsorbent for river water treatment: isotherm and kinetic studies

Banana is one of the most typical and consumed fruits in many countries. Disposal of banana peel waste can cause serious problems to the environment because it contains large quantities of nitrogen, phosphorus, and water that attract many microorganisms and vector-borne diseases. However, studies on banana peels as an adsorbent have found remarkable function of this waste for water treatment via the adsorption process. Hence, the current study investigated the hybrid process combining sand filtration column with activated banana peel macrocomposite (ABPM) adsorbent for river water treatment that can significantly reduce the fruit waste disposal into landfills and practice the waste-to-wealth concept. Moreover, ABPM has not been investigated by researchers in the past. This study proved ABPM has a valuable potential as a natural adsorbent in river water treatment. This is economical and reduces the negative impact on the environment. The isotherm and kinetic studies also provide the best overall model for experimental data.

Water sources play an important role to sustain all living organisms in this world. About 97% of water sources is saline water with only 1% of freshwater supply primarily from the rivers and lakes for the living purpose (Ahmed & Siwar 2014). Waterbodies such as rivers and lakes are the most sensitive environments for the activities of living organisms (Prakasam et al. 2021). Any changes in the ecosystem due to anthropogenic activities will endanger aquatic life in the water (Gholizadeh et al. 2016). Disposal of pollutants from the domestic, agricultural, and industrial sectors has degraded the quality of the river water and led to various negative impacts on human health as well as the environment (Naubi et al. 2016; Iwar et al. 2021).

Rapid population growth, urbanization, and industrial development have increased annual waste generation exponentially where solid wastes are commonly dumped in the landfill (Parvin & Tareq 2021). It is reported that more than 50% of food waste generated from municipal solid waste is disposed in the landfills in Malaysia (Mohd Yatim et al. 2019). The Malaysia Agricultural Research and Development Institute (MARDI) had found that 20–50% of this food waste consists of fruit waste and vegetable waste generated in Malaysia (The Star 2016). Generally, 70% out of 50% of these food wastes are disposed in the landfill for a composting process (Ghafar 2017). The landfilling process can harm the environment as the emission of landfill gases will pose risks to human health and contribute to the global warming phenomenon (Danthurebandara et al. 2012).

Banana is one of the highest consumed fruits in the world with about 54.4% of the world production of banana from Asian countries (Khoozani et al. 2019). The disposal of banana peels can contribute to environment problems as it contains nitrogen, phosphorus, and high water content that may promote the growth of microorganisms (Ibrahim et al. 2017). Fruit waste from banana peels is a commonly accessible, underused, and very effective biosorbent in water treatment (Akpomie & Conradie 2020; Rudi et al. 2020; Shukla et al. 2020; Nathan et al. 2021). In addition, banana peels are now being used extensively as an efficient waste to cleanse waters that have been contaminated with various pollutants (Akpomie & Conradie 2020). The banana peel contains the porous structure and functional groups such as amine (-NH2), alcohol (-OH), and carboxyl (-COOH) that are responsible for the adsorption process (Zheng & Wang 2013; Ali 2017; Oladipo et al. 2019; Çatlıoğlu et al. 2021). However, the porous adsorbent is fragile and brittle due to its low mechanical properties in terms of tensile stress induced in sorbent or catalyst bulk (David 2015). To overcome this problem, some researchers had improved the mechanical strength of macrocomposite adsorbents by mixing the cement with the adsorbent (Wang et al. 2017; Rosli et al. 2020).

Apart from the adsorbent, a sand filter is a conventional method that has been used in the water treatment for the removal of organic and inorganic contaminants (Ibrahim et al. 2021; Mokhtari et al. 2021). Patel et al. (2020) show that the combination of sand filtration followed by adsorption had remarkable COD and BOD removal of 95.6 and 93.3%, respectively. Filtration and adsorption are low-cost and simple processes to treat the river water. This study attempts to investigate a new hybrid process that combines a sand filtration column with activated banana peel macrocomposite (ABPM) adsorbent for the treatment of river water. The ABPM adsorbent is characterised by SEM, energy-dispersive X-ray (EDX), and Fourier transform infrared analysis for surface morphology, elemental composition, and functional groups. The column was developed by filing coarse sand and gravel. The performance of the sand filter column with and without ABPM adsorbent was compared for the river water treatment.

To date, the effectiveness of ABPM adsorbent has not yet been investigated with the details of the Langmuir and Freundlich model in a lab-scale sand filter column. In the present study, the adsorption isotherms of COD, BOD, total suspended solids (TSS), AN, and turbidity from river water treatment using sand filtration column with ABPM adsorbent were investigated using Langmuir and Freundlich to elucidate the adsorption process.

The water sampling was conducted at Panchor River, located at Panchor, Johor, Malaysia (2.1730° N and 102.7106° E). The water sample was collected in a 30 L high-density polyethylene bottle by composite sampling technique. The collected river water was stored in a refrigerator at 40 °C at the laboratory to preserve the water sample before the experiment. Characterisation of the river water sample including pH, chemical oxygen demand (COD), dissolved oxygen (DO), biochemical oxygen demand (BOD), and TSS was conducted according to the American Public Health Association Standard Method (APHA 2012). Turbidity of the water sample was determined using TL2300 Tungsten Lamp Turbidimeter (Hach) based on the procedure of Method 180.1 (O'Delll 1993), while ammonia nitrogen (AN) was determined by the Nessler Method using UV-Vis Spectrometer (DR 6000) (Hach). The characteristics of the river water sample are shown in Table 1.

The surface morphology and elemental composition of ABPM adsorbent were determined by SEM and EDX analysis COXEM EM-30AX. The SEM was operated at magnification and accelerating voltage of 5 kV. The samples were inserted into the SEM system that is gold coated for the scanning process as described by Faria et al. (2018). The functional groups for ABPM adsorbent were analysed using Fourier-Transform Infrared (FTIR) Spectrometer (Agilent Cary 630) that was operated between the wavelength region of 600 and 4,000 cm⁻¹. The detector measures the intensity of the radiation transmitted and converts it into FTIR spectra by a computer using the Fourier transform method (Munajad & Subroto 2018). The FTIR spectra give information about the characteristic of the functional group of the adsorbent.

Figure 6 shows turbidity removal for sand filter with and without banana peel macrocomposite adsorbent at a flowrate of 1.15, 5.02, and 15.8 ml/s. Overall, the turbidity removal of the water sample from the sand filter with adsorbent is higher than the sand filter without adsorbent. For the flowrate of 1.15 ml/s, the turbidity removal of the sand filter with adsorbent was maintained above 90% throughout the experiment. While for sand filter without adsorbent, the removal was maintained above 50% throughout the 80 min filter run time. For the flowrate of 5.02 ml/s, the turbidity removal fluctuated throughout the filter run time with the highest removal efficiency of 87.6% for the sand filter with adsorbent compared to the sand filter without adsorbent with only 62.64% removal efficiency. The turbidity removal of the sand filter with and without adsorbent under the flowrate of 15.8 ml/s was the lowest. The trend of removal efficiency shows that sand filter with and without adsorbent was declining as the flowrate increases. This is because the increase in the flowrate had resulted in greater shearing strength at the media pore and enhanced the transportation of deposited turbidity matters through the filter media (Mahanna et al. 2018).

Figure 7 shows COD removal for sand filter with and without banana peel macrocomposite adsorbent at a flowrate of 1.15, 5.02, and 15.8 ml/s. The COD removal of 44.4% (1.15 ml/s), 38.3% (15.8 ml/s), and 27.8% (5.02 ml/s) was obtained in sand filter with macrocomposite banana peel adsorbent, while the reduction of COD in the sand filter without adsorbent was slightly low with only 35% (15.8 ml/s), 22.2% (1.15 ml/s), and 16.7% (5.02 ml/s). Hence, the effective flowrates of the sand filter with and without adsorbent for COD removal were 1.15 and 15.8 ml/s, respectively. Yamina et al. (2013) also showed that the removal of COD by sand filter with the activated carbon adsorbent was more effective compared to the sand filter without adsorbent. This is due to the characteristic of adsorbents that have a larger specific surface for the development of biofilm, resulting in higher COD removal.

Figure 8 shows BOD removal for sand filter with and without banana peel macrocomposite adsorbent at a flowrate of 1.15, 5.02, and 15.8 ml/s. High BOD removal was displayed in the sand filter with banana peel macrocomposite adsorbent with 86.7% at 1.15 ml/s followed by 82% at 15.8 ml/s and 81% at 5.02 ml/s. This indicates that more than 80% of organic matter was removed after the filtration process. On contrary, BOD removal for sand filter without adsorbent was low with 26.2% at 5.02 ml/s, followed by 23.5% at 1.15 and 21.3% at 15.8 ml/s. According to Munagapati et al. (2018), the presence of carboxyl, hydroxyl, and amine groups at the surface of banana peels contributes to the adsorption of contaminants, thus increasing the removal rate. The addition of ABPM adsorbent in the sand filter was effective in removing organic pollutants in the river water. This also may be contributed to the porous surface of adsorbent after carbonization, which traps the organic matter by physical adsorption (Dalahmeh 2016).

Figure 9 shows TSS removal for sand filter with and without banana peel macrocomposite adsorbent at a flowrate of 1.15, 5.02, and 15.8 ml/s. The sand filter with banana peel macrocomposite adsorbent has higher TSS removal efficiency compared to the sand filter without the adsorbent. The highest TSS removal efficiency of the sand filter with adsorbent was recorded at 75.2% (1.15 ml/s), 74.2% (5.02 ml/s), and 62% (15.8 ml/s). On the other hand, the sand filter without the adsorbent was able to remove 66.1% TSS under a flowrate of 5.02 ml/s followed by 53.1 and 40.3% with a flowrate of 1.15 and 15.8 ml/s, respectively. The TSS removal efficiency declined with the increase of the flowrate in the sand filter due to the shearing of particle deposit. Increasing the flowrate in the filter promoted the turbulence level of the untreated river water. This reduces the deposited pollutant in the pore of the sand filter and adsorbent media, which leads to higher TSS (Yaseen et al. 2019).

Figure 10 shows AN removal for sand filter with and without banana peel macrocomposite adsorbent at a flowrate of 1.15, 5.02, and 15.8 ml/s. The highest AN removal was recorded at 53.9% (1.15 ml/s), followed by 48% (15.8 ml/s) and 35% (5.02 ml/s) for the sand filter with adsorbent. The sand filter without the adsorbent shows the highest AN removal of 45% at the flowrate of 15.8 ml/s, while the lowest AN removal of 17% at the flowrate of 5.02 ml/s. The sand filter with ABPM adsorbent had higher AN removal compared to the sand filter without the adsorbent due to the high porosity of micropores and acidic functional groups such as hydrocarbon, hydroxyl, and alkene group, which enhances the cation exchange capacity of the adsorbent (Liang et al. 2016).

Langmuir and Freundlich isotherms were obtained by plotting 1/q_e versus 1/C_e and log q_e versus log C_e, showing that each model is applicable to sorption data. The result revealed that the correlation coefficient value for all parameters in the Langmuir isotherms was more than in the Freundlich isotherms as shown in Figures 11–15(a) and 15(b). The Langmuir and Freundlich constant isotherms in Table 2 display the parameters of each model on the established correlation coefficient of R² from the basis of the modelling curved. The coefficient of turbidity, COD, BOD, TSS, and AN in both isotherms revealed that the correlation for Langmuir had R² > 0.98 and Freundlich isotherms with R² > 0.95. From the results, the Q_max values and KL (g/L) for all parameters corresponding to the adsorption capacity of ABPM are in line with the studies by Kassim et al. (2018) and Hor et al. (2016) which reported that all parameter adsorption experiments by the Langmuir model were performed and the experimental data have fitted R² values more than 0.98.

Based on Table 2, the low value of n in Freundlich isotherm directed to more adsorption for TSS and COD. Also, the high value of K_f indicated the better adsorption capacity on the parameter, where TSS achieved 0.8041 of K_f and n = 0.5525. These results confirm the effectiveness of banana peel macrocomposite adsorbent in the filtration column for the removal of all the parameters. The outcome revealed that the selection of Langmuir adsorption isotherm was the best model for the removal of all the tested parameters with R² > 0.98.

Table 3 describes that pseudo-first order is mainly the dominant kinetic model for the adsorption by ABPM adsorbent since the values of R² are higher than those in the pseudo-second order. Hence, the calculated q_e values in the pseudo-first order are in good agreement with the experimental q_e values compared to that in the pseudo-second-order kinetic model. These results also indicate that the interaction mechanism by ABPM acts simultaneously with the adsorption process.

The hybrid process combining the sand filtration column with the ABPM adsorbent reduces BOD, COD, TSS, AN, and turbidity from river water. The sand filter with ABPM adsorbent operated at 1.15 ml/s demonstrates the highest removal efficiency of more than 90% turbidity, 44.4% COD, 86.7% BOD, 75.2% TSS, and 53.9% AN removal. This is due to the addition of ABPM adsorbent that consists of high porosity and surface area which provide better adsorption performance. In addition, the presence of functional groups such as hydrocarbon group, hydroxyl, and alkene enhanced the cation exchange capacity of the adsorbent. Consequently, greater pollutant reduction was achieved compared to the sand filter without the adsorbent. Besides, the low flowrate of sand filter has more tendency to deposit the pollutant within the pore of filter media and adsorbent media. The Langmuir isotherm model described the adsorption process better than the Freundlich isotherm model with a high coefficient of determination (R² > 0.98), while the kinetic model of pseudo-first order takes place as an adsorption function for all parameters. The sand filtration process with ABPM adsorbent can be used as an alternative to the commercial expensive filter and adsorbent for river water treatment.

This study was funded by the Ministry of Higher Education (MOHE), Malaysia through the Fundamental Research Grant Scheme grant (K219) (FRGS/1/2019/TK10/UTHM/03/3).

All procedures performed in the present study involving human participants were in accordance with the ethical standards of the institutional and/or national research committee.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.