Removal E ﬃ ciencies of Manganese and Iron Using Pristine and Phosphoric Acid Pre-Treated Biochars Made from Banana Peels

: The purpose of this study was to compare the removal e ﬃ ciencies of manganese (Mn) and iron (Fe) using pristine banana peel biochar (BPB) and phosphoric acid pre-treated biochars (PBPB) derived from banana peels. The removal e ﬃ ciencies of Mn and Fe were investigated under di ﬀ erent adsorbent dosages (0.4–2 g L − 1 ), temperatures (15–45 ◦ C), and ionic strengths (0–0.1 M), and were directly correlated to the di ﬀ erences in physicochemical properties of BPB and PBPB, to identify the removal mechanisms of heavy metals by adsorption processes. The removal of Mn by PBPB obeyed the Freundlich isotherm model while the removal of Mn and Fe by BPB followed the Langmuir isotherm model. However, the removal of Fe by PBPB followed both Freundlich and Langmuir isotherm models. The removal e ﬃ ciencies of Mn and Fe by BPB and PBPB increased with increasing temperatures and decreased with increasing ionic strengths. PBPB more e ﬀ ectively removed Mn and Fe compared to BPB due to its higher content of oxygen-containing functional groups (O / C ratio of PBPB = 0.45; O / C ratio of BPB = 0.01), higher surface area (PBPB = 27.41 m 2 g − 1 ; BPB = 11.32 m 2 g − 1 ), and slightly greater pore volume (PBPB = 0.03 cm 3 g − 1 ; BPB = 0.027 cm 3 g − 1 ). These observations clearly show that phosphoric acid pre-treatment can improve the physicochemical properties of biochar prepared from banana peels, which is closely related to the removal of heavy metals by adsorption processes. prepared from banana peels. First, the physicochemical characteristics of pristine and phosphoric acid pre-treated biochars derived from banana peels were rigorously characterized. Then, various adsorption experiments were conducted to investigate optimum adsorbent dosages, adsorption kinetics, adsorption isotherms, and e ﬀ ects of temperature and ionic strengths on adsorption. The improved adsorption e ﬃ ciencies by phosphoric acid treatment were analyzed based on the physicochemical characteristics of the biochar, and the adsorption mechanisms of Mn and Fe were discussed. This study improves our understanding of the e ﬀ ect of phosphoric acid treatment on modiﬁcation of the surface structure and functional groups of biochars for heavy metal removal. acid pre-treatment on banana peel biochar was investigated for enhancement of Mn and Fe removal e ﬃ ciencies. The physicochemical characteristics of pristine and phosphoric acid pre-treated biochars were characterized using the elemental analyzer, BET, and ATR-FTIR. These characterizations revealed that phosphoric acid pre-treatment facilitated the formation of oxygen-containing functional groups (i.e., hydroxyl and carboxyl functional groups), which could enhance the adsorption of heavy metals on the biochar surfaces. In addition, phosphoric acid pre-treatment improved speciﬁc surface area and pore volume for more adsorption sites for heavy metals. The adsorption experiments showed that phosphoric acid pre-treatment can improve the removal of Mn and Fe signiﬁcantly. The results of the adsorption kinetics of Mn and Fe by BPB and PBPB were well-matched to the pseudo-second-order, indicating that adsorption was predominantly governed by chemical adsorption. The adsorption of Mn and Fe by BPB obeyed the Langmuir


Introduction
Water pollution by heavy metals released from various industrial activities such as metal plating and cleaning, mining, refineries, coatings, batteries, and automobile radiators, is an emerging environmental issue in water treatment engineering since heavy metals may pose adverse effects on human health and aquatic ecosystems due to their high toxicity, carcinogenicity, and non-biodegradability [1,2]. Among various heavy metals used in industrial activities, manganese (Mn) and iron (Fe) are known to be major inorganic pollutants affecting water quality [3]. Although Mn is essential to activate enzymes in the human system, high Mn concentrations can generate respiratory diseases, and continuous administration can cause neurotoxicity risk in humans [3,4]. In the case of Fe, it can cause undesirable aesthetic concerns (i.e., metallic tastes) and lead to the growth of ferrobacteria

Production of Biochar
The banana peel wastes were dried in an oven at 105 • C for 24 h, ground using a mortar, washed several times with deionized (DI) water to remove impurities on their surfaces, and then dried in the oven at 105 • C for 12 h. From the resulting banana peel powder, 50 g was immersed in a 500 mL phosphoric acid solution (20 wt. %) for 2 h to activate adsorption sites, and dried at 105 • C for 12 h. Pristine and phosphoric acid pre-treated banana peel wastes (weight of each banana peel waste = 20 g) were pyrolyzed in the tubular furnace (PyroTech, Namyangju, Gyeonggi-do, Korea) at 600 • C (heating rates = 0.2 • C min −1 ) under N 2 conditions (N 2 flow rate = 0.25 L min −1 ) for 2 h and then cooled to room temperature. Biochars prepared from banana peel wastes were washed several times with DI water, filtered with a 0.7 µm GF/F filter (Whatman, Maidstone, UK), and then dried in the oven at 105 • C for 12 h. Biochars produced from pristine and phosphoric acid pre-treated banana peel wastes are defined as BPB (banana peel biochar without pre-treatment) and PBPB (pre-treated banana peel biochar), respectively.

Characterization of Biochar
Element composition of BPB and PBPB was analyzed using an elemental analyzer (EuroEA3000 CHNS-O, Euro Vector S.p.A, Via Tortona, Milan, Italy). The ash content was calculated by subtracting the quantities of carbon (C), hydrogen (H), nitrogen (N), and oxygen (O) from the total mass fraction of the adsorbents. The specific surface area of biochar was determined with a Bronauer-Emmett-Teller (BET) analyzer (BELSORP-mini II, MicrotracBEL, Japan). The functional group composition of the adsorbents was identified using attenuated total reflectance-Fourier transform infrared spectroscopy (ATR-FTIR) (Frontier Optica, Perkin Elmer, Waltham, MA, USA).

Optimal Adsorbent Dosages
Prior to the adsorption kinetics experiments, the optimum dosage of BPB and PBPB for each heavy metal was determined. The adsorbents (dosage = 0.02-4 g L −1 ) were added to 25 mL of heavy metal solution (each metal concentration = 10 mg L −1 , pH = 7.0) and then mixed at 150 rpm and 25 • C using a shaking incubator (VS-8480, Vision Scientific, Daejeon-Si, Korea) for 3 h. All the adsorption tests were repeated three times to minimize experimental errors.

Adsorption Kinetics Analysis
For adsorption kinetics experiments, the optimum dosage of each adsorbent was added to 25 mL of sample solutions (each heavy metal concentration = 10 mg L −1 ) and mixed at 150 rpm using the shaking incubator for 0-24 h (temperature = 25 • C, pH = 7.0). The concentrations of Mn and Fe at the initial and equilibrium states were measured using colorimetric methods with a UV-Vis spectrophotometer (UV-1280, Shimadzu, Kyoto, Japan) at UV absorbances of 525 and 510 nm, respectively. All the Water 2020, 12, 1173 4 of 13 adsorption tests were repeated three times to minimize experimental errors. The amount of adsorbed heavy metals at time t (q t (mg g −1 )) was calculated as follows [14]: where C 0 and C are the initial and final concentrations (mg L −1 ) of heavy metals in the solutions, V is the volume (L) of the solution, and m is the mass (g) of the used adsorbents. The removal efficiencies of heavy metals were calculated using Equation (2) [14]: Removal efficiency of heavy metal (%) = where C e represents the concentrations of each heavy metal (mg L −1 ) at equilibrium of the solutions. The adsorption kinetics of Mn and Fe were investigated using Equations (3) and (4): Pseudo-second-order model : where k 1 (min −1 ) is the constant of the pseudo-first-order equation, and k 2 (g mg −1 min −1 ) is the constant of the pseudo-second-order equation. q e (mg g −1 ) is the adsorption capacity at equilibrium.

Adsorption Isotherm Analysis
The adsorption isotherms of Mn and Fe by BPB and PBPB were identified with 6 different initial concentrations (each heavy metal concentration = 1-10 mg L −1 ) under controlled conditions (agitation time = 24 h, mixing speed = 150 rpm, temperature = 25 • C, pH = 7.0). The adsorption results were analyzed using the Langmuir and Freundlich isotherm models.
Langmuir isotherm : q e = q max K L c e 1 + K L c e (5) where q e (mg g −1 ) is the maximum monolayer adsorption capacity of heavy metals, and K L (L mg −1 ) is the equilibrium constant of the Langmuir equation.
Freundlich isotherm : where K F (mg 1−(1/n) L 1/n g −1 ) is the Freundlich adsorption constant, and n is the dimensionless empirical coefficient related to adsorption strength, which depends on the surface heterogeneity.

Effects of Temperatures and Ionic Strengths
The effects of temperature and ionic strength on the adsorption of heavy metals by BPB and PBPB were investigated at various temperatures (15-45 • C) and ionic strength conditions (0.005-0.1 M) (each heavy metal concentration = 10 mg L −1 , agitation time = 6 h, mixing speed = 150 rpm, temperature = 25 • C, pH = 7.0). The removal efficiencies of Mn and Fe using BPB and PBPB were calculated using Equations (2) and (3) as described in the previous section. the functional group composition affecting the adsorption of heavy metals were different. The IR peaks related to O-H stretching and C-O stretching of alcohols appeared at 3500-3000 cm −1 (BPB = 3443 cm −1 , PBPB = 3431 cm −1 ) and 1210-1100 cm −1 (BPB = 1103 cm −1 , PBPB = 1180 cm −1 ) due to the presence of alcohol functional groups (-CH 2 OH-) derived from the cellulose component of the banana peels [33]. The IR peaks related to N-H stretching of amides were found in the range of 1650-1550 cm −1 (BPB = 1565 cm −1 , PBPB = 1575 cm −1 ). Furthermore, C-H stretching of alkanes and C-H stretching of aromatics exhibited relatively strong IR peaks in the range of 1500-1300 cm −1 and 900-670 cm −1 , respectively [34]. The IR peak derived from C-O stretching of alcohols exhibited a relatively high IR intensity in PBPB compared to BPB in the range of 1210-1100 cm −1 because the phosphoric acid pre-treatment promoted the formation of oxygen-containing functional groups closely associated with the adsorption of heavy metals [34,35].

Physical Properties of Biochar
Water 2020, 12, x FOR PEER REVIEW 5 of 13 acid pre-treatment promoted the formation of oxygen-containing functional groups closely associated with the adsorption of heavy metals [34,35].  Table 1 shows the elemental composition and surface properties of BPB and PBPB. Although the hydrogen, nitrogen, and ash content of BPB and PBPB were similar (H content of BPB = 1.6%, H content of PBPB = 1.9%; N content of BPB = 2.9%, N content of PBPB = 2.2%; ash content of BPB = 19.0%, ash content of PBPB = 16.4%), PBPB had lower carbon content (54.7%) and higher oxygen content (24.8%) compared to BPB (C content = 75.4%, O content = 1.1%). Therefore, the H/C, O/C, and N/C ratios of PBPB were much higher than those of BPB. These observations indicate that increases in hydroxyl and carboxyl functional groups in PBPB after phosphoric acid pre-treatment may enhance electrostatic attractions between heavy metals and the adsorbent surfaces intimately related to the adsorption of heavy metals [36]. Furthermore, the specific surface area and total pore volume of PBPP (specific surface area = 27.41 m 2 g −1 ; total pore volume = 0.032 cm 3 g −1 ) were considerably greater compared to BPB (BPB = 11.32 m 2 g −1 ; total pore volume = 0.027 cm 3 g −1 ). These results suggest that heavy metals can be adsorbed more readily by PBPB than BPB due to its abundance of adsorption sites [12].   Table 1 shows the elemental composition and surface properties of BPB and PBPB. Although the hydrogen, nitrogen, and ash content of BPB and PBPB were similar (H content of BPB = 1.6%, H content of PBPB = 1.9%; N content of BPB = 2.9%, N content of PBPB = 2.2%; ash content of BPB = 19.0%, ash content of PBPB = 16.4%), PBPB had lower carbon content (54.7%) and higher oxygen content (24.8%) compared to BPB (C content = 75.4%, O content = 1.1%). Therefore, the H/C, O/C, and N/C ratios of PBPB were much higher than those of BPB. These observations indicate that increases in hydroxyl and carboxyl functional groups in PBPB after phosphoric acid pre-treatment may enhance electrostatic attractions between heavy metals and the adsorbent surfaces intimately related to the adsorption of heavy metals [36]. Furthermore, the specific surface area and total pore volume of PBPP (specific surface area = 27.41 m 2 g −1 ; total pore volume = 0.032 cm 3 g −1 ) were considerably greater compared to BPB (BPB = 11.32 m 2 g −1 ; total pore volume = 0.027 cm 3 g −1 ). These results suggest that heavy metals can be adsorbed more readily by PBPB than BPB due to its abundance of adsorption sites [12].  Figure 2 presents the change in the removal efficiencies of Mn and Fe as a function of adsorbent dosages. The removal efficiencies of Mn and Fe using BPB and PBPB were increased with increasing adsorbent dosages because of the increased availability of adsorption sites on the adsorbent surfaces [14]. The lower removal efficiencies of Mn by BPB and PBPB compared to those of Fe might be attributed to differences in the electronegativity and ion radius of Mn and Fe. Since Fe has a higher electronegativity than Mn (Fe = 1.8, Mn = 1.5) and a small ion radius, it can more easily diffused into the pores of PBPB and BPB [37,38]. In the equilibrium state, PBPB exhibited higher removal efficiencies of Mn and Fe than BPB (removal efficiency of Mn by BPB = 32%, removal efficiency of Mn by PBPB = 46%, removal efficiency of Fe by PBPB = 96%, removal efficiency of Fe by BPB = 85%). These results mean that phosphate pre-treatment promoted the formation of functional groups on the biochar surfaces capable of adsorbing heavy metals [27,39]. The removal efficiencies of Mn by BPB and PBPB reached the steady state at an adsorbent dosage of 3 g L −1 while the steady states of Fe adsorption occurred at 0.1 g L −1 for BPB and 0.3 g L −1 for PBPB, respectively. Hence, the PBPB and BPB dosages obtained at their steady states for Mn and Fe were applied for further adsorption experiments.

Effects of Adsorbent Dosages
Water 2020, 12, x FOR PEER REVIEW 6 of 13 attributed to differences in the electronegativity and ion radius of Mn and Fe. Since Fe has a higher electronegativity than Mn (Fe = 1.8, Mn = 1.5) and a small ion radius, it can more easily diffused into the pores of PBPB and BPB [37,38]. In the equilibrium state, PBPB exhibited higher removal efficiencies of Mn and Fe than BPB (removal efficiency of Mn by BPB = 32%, removal efficiency of Mn by PBPB = 46%, removal efficiency of Fe by PBPB = 96%, removal efficiency of Fe by BPB = 85%). These results mean that phosphate pre-treatment promoted the formation of functional groups on the biochar surfaces capable of adsorbing heavy metals [27,39]. The removal efficiencies of Mn by BPB and PBPB reached the steady state at an adsorbent dosage of 3 g L −1 while the steady states of Fe adsorption occurred at 0.1 g L −1 for BPB and 0.3 g L −1 for PBPB, respectively. Hence, the PBPB and BPB dosages obtained at their steady states for Mn and Fe were applied for further adsorption experiments.  Figure 3 depicts the adsorption kinetics of Mn and Fe by BPB and PBPB. The adsorption proceeded rapidly in the beginning for both heavy metals (Mn ≤360 min; Fe ≤180 min) and almost reached equilibrium at 10 h. A possible explanation for these results is that the availability of adsorption sites plays key roles in the adsorption of Fe and Mn by BPB and PBPB [19]. Table 2 presents the kinetic model parameters for the adsorption of Mn and Fe by BPB and PBPB. Based on the correlation coefficient values (R 2 ), the pseudo-second-order model better described the adsorption of Mn and Fe by BPB and PBPB than the pseudo-first-order model. These observations indicate that the adsorption of Mn and Fe by BPB and PBPB is predominantly governed by chemical adsorption (i.e., covalent bonding or ion/electron exchange) [27]. As shown in Figure 3, the adsorption capacities at equilibrium (qe) of Mn and Fe by BPB were 1.14 and 31.61 mg g -1 , respectively. Meanwhile, the qe of Mn and Fe by PBPB were 2.03 and 32.99 mg g -1 , respectively. The higher qe values by PBPB support the assumption that PBPB is more effective for the adsorption of heavy  Figure 3 depicts the adsorption kinetics of Mn and Fe by BPB and PBPB. The adsorption proceeded rapidly in the beginning for both heavy metals (Mn ≤360 min; Fe ≤180 min) and almost reached equilibrium at 10 h. A possible explanation for these results is that the availability of adsorption sites plays key roles in the adsorption of Fe and Mn by BPB and PBPB [19]. Table 2 presents the kinetic model parameters for the adsorption of Mn and Fe by BPB and PBPB. Based on the correlation coefficient values (R 2 ), the pseudo-second-order model better described the adsorption of Mn and Fe by BPB and PBPB than the pseudo-first-order model. These observations indicate that the adsorption of Mn and Fe by BPB and PBPB is predominantly governed by chemical adsorption (i.e., covalent bonding or ion/electron exchange) [27]. As shown in Figure 3, the adsorption capacities at equilibrium (q e ) of Mn and Fe by BPB were 1.14 and 31.61 mg g −1 , respectively. Meanwhile, the q e of Mn and Fe by PBPB were 2.03 and 32.99 mg g −1 , respectively. The higher q e values by PBPB support the assumption that PBPB is more effective for the adsorption of heavy metals than BPB due to its higher content of oxygen-containing functional groups [34,35].

Adsorption Isotherms of Mn and Fe
The adsorption mechanisms of Mn and Fe by BPB and PBPB were analyzed using the Langmuir and Freundlich adsorption isotherm models ( Figure 4 and Table 3). The Langmuir isotherm model was well-fitted to the adsorption of Mn and Fe by BPB (R 2 of Mn = 0.972, R 2 of Fe = 0.869). These results imply that monolayer adsorption is responsible for the adsorption of Mn and Fe by BPB [7]. The adsorption of Mn by PBPB followed the Freundlich isotherm model (R 2 = 0.993) more closely than the Langmuir isotherm model (R 2 = 0.898). It is evident that multilayer adsorption strongly contributes to the adsorption of Mn by PBPB [13]. However, the adsorption of Fe by PBPB followed both Freundlich (R 2 = 0.933) and Langmuir isotherms (R 2 = 0.949). The n value of the Freundlich isotherm model was used to examine the adsorption affinity of Mn and Fe onto BPB and PBPB (Table 3): (i) n > 1 (favorable), (ii) n = 1 (linear), and (iii) n < 1 (unfavorable) [30]. The adsorption of Mn and Fe by BPB (n value of Mn = 7.267, n value of Fe = 1.069) and the adsorption of Mn by PBPB (n value = 2.471) were favorable whereas the adsorption of Fe by PBPB (n value = 0.977) was not favorable. The separation parameter R L value, based on R L = 1/(1 + K L C 0 ) of the Langmuir isotherm model, was also calculated to assess the adsorption preference of Mn and Fe toward BPB and PBPB: (i) R L > 1 (unfavorable), (ii) R L = 1 (linear), (iii) 1 > R L > 0 (favorable), and (iv) R L = 0 (irreversible) [40].

Adsorption Isotherms of Mn and Fe
The adsorption mechanisms of Mn and Fe by BPB and PBPB were analyzed using the Langmuir and Freundlich adsorption isotherm models (Figure 4 and Table 3). The Langmuir isotherm model was well-fitted to the adsorption of Mn and Fe by BPB (R 2 of Mn = 0.972, R 2 of Fe = 0.869). These results imply that monolayer adsorption is responsible for the adsorption of Mn and Fe by BPB [7]. The adsorption of Mn by PBPB followed the Freundlich isotherm model (R 2 = 0.993) more closely than the Langmuir isotherm model (R 2 = 0.898). It is evident that multilayer adsorption strongly contributes to the adsorption of Mn by PBPB [13]. However, the adsorption of Fe by PBPB followed both Freundlich (R 2 = 0.933) and Langmuir isotherms (R 2 = 0.949). The n value of the Freundlich isotherm model was used to examine the adsorption affinity of Mn and Fe onto BPB and PBPB (Table  3): (i) n > 1 (favorable), (ii) n = 1 (linear), and (iii) n < 1 (unfavorable) [30].

Effects of Temperature on Adsorption of Mn and Fe
The effects of temperature on the adsorption of Mn and Fe by BPB and PBPB are compared in Figure 5. The removal efficiencies of Mn and Fe by both BPB and PBPB were increased with increasing temperatures from 15 (removal efficiency of Mn by BPB = 17%, removal efficiency of Mn by PBPB = 46%, removal efficiency of Fe by BPB = 88%, removal efficiency of Fe by PBPB = 97%) to 45 • C (removal efficiency of Mn by BPB = 32%, removal efficiency of Mn by PBPB = 65%, removal efficiency of Fe by BPB = 97%, removal efficiency of Fe by PBPB = 99%). These results suggest that high temperatures may provide sufficient energy for the adsorption of heavy metals on the surficial and interior layers of biochars [41]. From the higher removal efficiencies of Mn and Fe by PBPB than BPB, it can be concluded that the abundance of oxygen-containing functional groups facilitates the adsorption of heavy metals by carbonaceous adsorbents [36].
Water 2020, 12, x FOR PEER REVIEW 10 of 13 by PBPB (removal efficiency of Mn = 32%-54%; removal efficiency of Fe = 96%-99%) than BPB (removal efficiency of Mn = 21%-35%; removal efficiency of Fe = 84%-95%) at all the tested ionic strengths provide evidence that the surface structural features and oxygen-containing functional group abundance govern the adsorption of heavy metals by carbonaceous adsorbents [12,21,22].  Figure 6 illustrates the effects of ionic strength on the removal of heavy metals by BPB and PBPB. The removal efficiencies of Mn and Fe by BPB and PBPB were gradually decreased with increasing ionic strengths. For example, when ionic strength was increased from 0 to 0.1 M, the removal efficiencies of Mn by BPB and PBPB were decreased from 35% and 54% to 21% and 32%, respectively. Meanwhile, the removal efficiencies of Fe by BPB and PBPB were decreased from 95% and 99% to 84% and 96%, respectively. These observations imply that increases of ionic strength may reinforce the electrostatic repulsion between heavy metals and the biochar surfaces and reduce the availability of adsorption sites on the biochar surfaces through the aggregation of biochars [42]. The removal efficiencies of Fe by BPB and PBPB were less affected by the changes in ionic strength compared to the removal efficiencies of Mn by BPB and PBPB because Fe has a smaller ion radius and higher electronegativity than Mn [37,38]. These inherent natures allowed Fe to exhibit higher attractive charges in the nucleus on the electron orbital [37,38]. Therefore, Fe more easily penetrated into the pores of biochars compared to Mn [37,38]. In addition, the higher removal efficiencies of Mn and Fe by PBPB (removal efficiency of Mn = 32-54%; removal efficiency of Fe = 96-99%) than BPB (removal efficiency of Mn = 21-35%; removal efficiency of Fe = 84-95%) at all the tested ionic strengths provide evidence that the surface structural features and oxygen-containing functional group abundance govern the adsorption of heavy metals by carbonaceous adsorbents [12,21,22].

Effects of Ionic Strength on Adsorption of Mn and Fe
Water 2020, 12, x FOR PEER REVIEW 10 of 13 by PBPB (removal efficiency of Mn = 32%-54%; removal efficiency of Fe = 96%-99%) than BPB (removal efficiency of Mn = 21%-35%; removal efficiency of Fe = 84%-95%) at all the tested ionic strengths provide evidence that the surface structural features and oxygen-containing functional group abundance govern the adsorption of heavy metals by carbonaceous adsorbents [12,21,22].

Conclusions
In this study, phosphoric acid pre-treatment on banana peel biochar was investigated for enhancement of Mn and Fe removal efficiencies. The physicochemical characteristics of pristine and phosphoric acid pre-treated biochars were characterized using the elemental analyzer, BET, and ATR-FTIR. These characterizations revealed that phosphoric acid pre-treatment facilitated the formation of oxygen-containing functional groups (i.e., hydroxyl and carboxyl functional groups), which could enhance the adsorption of heavy metals on the biochar surfaces. In addition, phosphoric acid pre-treatment improved specific surface area and pore volume for more adsorption sites for heavy metals. The adsorption experiments showed that phosphoric acid pre-treatment can improve the removal of Mn and Fe significantly. The results of the adsorption kinetics of Mn and Fe by BPB and PBPB were well-matched to the pseudo-second-order, indicating that adsorption was predominantly governed

Conclusions
In this study, phosphoric acid pre-treatment on banana peel biochar was investigated for enhancement of Mn and Fe removal efficiencies. The physicochemical characteristics of pristine and phosphoric acid pre-treated biochars were characterized using the elemental analyzer, BET, and ATR-FTIR. These characterizations revealed that phosphoric acid pre-treatment facilitated the formation of oxygen-containing functional groups (i.e., hydroxyl and carboxyl functional groups), which could enhance the adsorption of heavy metals on the biochar surfaces. In addition, phosphoric acid pre-treatment improved specific surface area and pore volume for more adsorption sites for heavy metals. The adsorption experiments showed that phosphoric acid pre-treatment can improve the removal of Mn and Fe significantly. The results of the adsorption kinetics of Mn and Fe by BPB and PBPB were well-matched to the pseudo-second-order, indicating that adsorption was predominantly governed by chemical adsorption. The adsorption of Mn and Fe by BPB obeyed the Langmuir isotherm model, whereas the Freundlich isotherm model described the adsorption of Mn by PBPB well. The adsorption of Fe by PBPB followed both Freundlich and Langmuir isotherm models. Furthermore, the adsorption of Mn and Fe by both BPB and PBPB increased with increasing temperature. However, the removal efficiencies were decreased with increasing ionic strength. From the excellent adsorption performance of Mn and Fe by PBPB compared to BPB, it can be concluded that phosphoric acid pre-treatment is a promising method to enhance Mn and Fe removal by banana peel biochar.

Conflicts of Interest:
The authors declare no conflict of interest.