Clean Technology Approach For Ciprofloxacin Sequestration Using Modified Banana Stalk


 The adsorption mechanisms and capacity of adsorbent derived from banana stalks in removing ciprofloxacin (CIP) from aqueous solutions were investigated in this study. The Banana Stalk Activated Carbon (BSAC) was prepared through acid activation and were characterised via SEM, FTIR and pHpzc. The SEM micrograph of BSAC revealed prominent well-developed pores which aid the adsorption process. The FTIR spetra shows the presence of various functional groups for efficient CIP adsorption. The effects of temperature (30 – 70 oC), adsorbent dosage (0.01 – 0.05 g/L), pH (2 - 10),initial adsorbate concentration (1 - 5 mg/L) and contact time (10 - 360 min) were evaluated using batch operations. Optimum CIP adsorption was obtained at pH 6. The maximum monolayer adsorptive capacity of the acid-treated banana stalk was 222.22 mg/g at 30 oC. Langmuir isotherm was the most appropriate isotherm model while pseudo second order kinetics best describe the kinetic process. Its thermodynamic studies indicated endothermic and spontaneous process. The cost implication of BSAC preparation (12.35 USD/kg) indicates that it is about 21 times cheaper than Commercial activated carbon (270.8 USD/kg) while the desorption studies reflects the reusability efficiency of the spent adsorbent. BSAC was found to be effective, efficient and economically viable for ciprofloxacin removal from aqueous solution.


Introduction
The release of trace level chemicals classi ed as emerging contaminants into the environment through daily human activity such as domestic, industrial and agricultural activity has become a source of concern [1,2]. The chemicals persist in the environment thereby pose a public health risk and threatened the availability of potable water [3]. A subset of emerging contaminant known as Endocrine Disruptive Chemicals (EDCs) deters the regular synthesis of the endocrine system, inhibits the metabolism and secretion of natural hormones thereby affecting the reproduction of man, plants and animals [4,5].
Numerous chemicals are classi ed as emerging contaminants due to adverse health hazards on the ecosystem through continual direct contact [6]. Some of these chemicals are pharmaceuticals [7], (Li et al., 2020), personal care products [8], pesticides [9] and plastic products [10]. Researchers have used various techniques to remove these trace compounds from water and waste water. Some of these removal techniques include electrochemical treatment [11], biological treatment [12], ozonation [13]and adsorption [14][15][16].
Cipro oxacin (C 17 H 18 FN 3 O 3 ) a uoroquinolone antibiotic is an example of pharmaceutical often used to treat broad spectrum bacterial infectious diseases [17]. It is a yellow to light yellow crystalline powder, 331.346 g/mol molecular weight, soluble in 0.1 M hydrochloric acid and water. It is the most prescribed or widely used antibiotics due to its therapeutic purpose. Its concentration in wastewater ranges between and was noticed to increase about 31 mg/L times in the e uents of pharmaceutical plant [18]. Cipro oxacin is pH-dependent and zwitterionic in nature, at pH ≤ 4.0 it exists in cationic (CIP + ) nature,4.0 pH ≤ 8.0 it exists in zwitterionic (CIP ± ) while pH > 8 it is in anionic (CIP − ) nature [19].
Cipro oxacin is released into the water environment either by defecation of unmetabolized chemical or direct disposal into the environment [20]. The effect of minimal concentration of CIP in the environment is a prolonged allergic reaction in the ecosystem and antibiotic resistance in bacteria [21][22][23].
Agricultural residues are ligno-cellulosic materials with different functional groups and highadsorptive capacity. They are generated in high quantity as a result of industrialization and commercialization [24].
Different research work has been carried out using agricultural residues as adsorbent. Some of these includes: eggshell [25], date seeds [26], bean pod [27], Barbara nut chaff [28] amongst others. Banana (Musa paradisiaca) is a perennial fruit, one of the most widely grown which originates from Southeastern Asia. It is cultivated in both tropical and sub-tropical regions. It is widely traded fresh fruit as a result of its high nutrient level and sweet taste [29]. In medicine, it is mostly recommended because of its antimicrobial [30] and antioxidant [31] characteristics. Two main residues are derived from banana fruit plantation which is stalks and peel. Improper disposal of these residues constitutes environmental menace hence, the need for alternative usage.
Banana stalks are readily available at little or no cost hence a single step acid activation and carbonization was carried out to produce low cost activated carbon in this study. Five different operational parameters were investigated into namely: solution pH, contact time, initial CIP concentration, temperature and adsorbent dose. Sequel to batch adsorption studies, the data were analyzed using four isotherm models namely: Freundlich [32], Dubinin-Radushkevich [33],Langmuir [34] and Temkin [35]. Also, the process kinetics model was examined by analyzing the data using the pseudo rst order, pseudosecond order and intraparticle diffusion model. Also, the cost of commercial activated carbon (CAC) was compared with the cost of preparation of the adsorbent to justify its economic viability. Desorption studies was also investigated into so as to determine the e ciency of reuse of spent adsorbent.

Chemicals Used
Four main chemicals were used in the adsorption process and these include cipro oxacin (C 17

Sample Collection and Pretreatment
The banana stalk was collected from a local market in Omu-Aran Kwara State, Nigeria. Specks of dirt and impurities on the stalks were removed by washing with distilled water. The stalks were further reduced in size by cutting with a knife after it was transferred into an oven operated at 105 o C until a consistent weight was attained. The dried samples were screened to 106 µm particle size sequel to crushing and subsequently kept in a closely tight container for further experimental use.

Activated Carbon Preparation
The acid-treated activated carbon was prepared by mixing 600 cm 3 of 0.3 M ortho-phosphoric acid (H 3 PO 4 ) with 30 g banana stalk powder in1000 cm 3 graduated beaker. The combination was stirred continuously and heated up on a hot plate until a paste is formed [27]. Four clean crucibles were heated in a furnace at 250 o C for 30 min to remove impurities therein and it was later placed in a desiccator for it to cool. The paste was transferred into the crucibles and heated at 650 o C in a furnace for 35 min and allowed in a desiccator. The pH of the activated carbon was adjusted to neutrality by washing severally with distilled water. After which it was dried to a consistent weight in an oven operated at 105 o C. Subsequently, for further use, BSAC was kept in an air tight container.

Fourier transform infrared spectroscopy
BSAC characteristic functional group was described by Fourier Transform Infrared Spectroscopy (FTIR) analysis. These functional groups were interpreted by the various bands on the adsorbent surface for e cient adsorption process.

pH point of zero charges (pH pzc )
The point on the adsorbent surface with which pH net charge is zero is referred to as the pH point of zero charges. The adsorbent surface is positively charged at pH less than pH pzc and negatively charged at pH greater than pH pzc [25].

Batch equilibrium studies
The batch adsorption studies of cipro oxacin over BSAC were investigated by analyzing the effect of initial cipro oxacin concentration, solution pH, adsorbent dosage, temperature and contact time. For this study, cipro oxacin concentration was varied between 1-5 mg/L. 100 ml of cipro oxacin solution was added to 0.01 g BSAC in a 250 cm 3 volumetric ask for each concentration. Afterward, the ask was placed in a water bath shaker set at 26 o C and 130 rpm agitation speed. The sample was withdrawn at a different time interval and the supernatant was separated. The absorbance of the supernatant was taken using a single cell holder Jenway UV Visible spectrophotometer operated at 277 nm [36]. The process of sample withdrawal and absorbance was repeated until equilibrium was attained. The percentage removal of cipro oxacin adsorbed was determined using Eq. 1.
where C i and C t (mg/L) are the concentration of cipro oxacin inaqueous solution at the initial time and speci c time t respectively.

Characteristics of the prepared adsorbent
The physicochemical and proximate analysis of BSAC was determined with results obtained presented in Table 1. The bulk density, ash content and moisture content of the adsorbent is very low. The low moisture and ash content obtained from BSAC made banana stalk an appropriate precursor for activated carbon preparation [37]. The high speci c surface area of BSAC indicates its good potential for cipro oxacin adsorption in an aqueous solution. Acid activation and carbonization of the precursor enhanced the development of the pores and subsequently resulted in a high speci c area of BSAC [38].

Fourier transform infrared spectroscopy (FTIR)
The speci c functional group on banana stalk activated carbon (BSAC) and cipro oxacin laden banana stalk activated carbon (CLBSAC) as shown by FTIR spectral is depicted in Fig. 1a [38]. After cipro oxacin adsorption, some peaks shifted, new peaks emerged and some disappeared. Table 2 shows an explicit presentation of the various bands' assignment of the spectra.

pH point of zero charge (pH pzc )
The value of pH pzc is at pH 4 and this was determined at the point on the x-axis where the curve cut through the pH 0 axis. The net charge is in the acidic medium with pH pzc at 4. At pH < 4 the adsorbent surface is positively charged and this aids the adsorption of anions. Also, at pH > 4 the adsorbent surface is negatively charged and this aids the adsorption of cations. This result shows that BSAC has a positive charged surface up till pH 4 [25,39]. The pH pzc of BSAC as investigated is shown in Fig. 3.

Scanning electron micrograph (SEM)
The surface characteristics of acid-activated BSAC and cipro oxacin -laden BSAC are shown in Fig. 3a and b. Numerous well-developed pores were shown on the adsorbent surface which can be attributed to the activating agent (H 3 PO 4 ) used for its modi cation [27]. Rough cavities and openings were also visible on the surface which is a result of lignocellulosic material breakdown coupled with the evaporation of volatile organic compounds at high temperature [40]. From Fig. 3b, the adsorbent surface was completely altered sequel to adsorption of cipro oxacin. The well-developed pores observed before adsorption were presumed to be lled with cipro oxacin molecules after adsorption [41].

Effects of contact time and initial cipro oxacin concentration
The adsorbate-adsorbent interaction as related to varying cipro oxacin concentration at the initial stage and contact time is shown in Fig. 4. From this gure, the quantity adsorbed at a speci c time increases as the initial concentration increased from 1 mg/L to 5 mg/L. For these concentrations, the adsorption process reached an equilibrium between 90 min to 360 min. Rapid adsorption of cipro oxacin was observed from the gure and consequently, the adsorption became slower until equilibrium was reached.
The speedy uptake could be ascribed to numerous widely open active sites on the adsorbent surface. A likely report was given in the study of modi ed activated carbon as a good adsorbent for quinoline sequestration [42]. Also, it was observed that the quantity of cipro oxacin adsorbed for the initial concentration consideredincreased from 61.24 to 170.30 mg/g at equilibrium. Adsorption rate increases as the diffusion process increased at equilibrium [43].

Effects of solution pH
From Fig. 6, adsorption at pH 6 and pH 2 gave the maximum and minimum percentage removal e ciency of 92.85% and 72.72%. This observation shows the zwitterionic nature of cipro oxacin that is at different pH values its molecule exists in different forms. Solution pH < 5.9 cipro oxacin molecules exist as cationic form, at solution pH > 8.89 it exists in an anionic form while at pH between 6.1 and 8.7 the zwitterionic and neutral form is prevalent [45]. The electrostatic interaction that exists between cipro oxacin and BSAC determines the adsorption rate. The lower percentage removal observed at pH < 4 could be as a result of the positively charged surface of the adsorbent and the cationic nature of the molecule. At pH 4-5 a slight increase was observed and this can be described as the effect of reduction in the cationic charge on the adsorbent surface. The maximum adsorption recorded at pH 6 can be presumed to be as a result of the zwitterionic nature of cipro oxacin and the negative sites of the adsorbent. Researchers have reported related results for oat hull and bamboo-based carbon [20,44].

Effects of temperature
The quantity of cipro oxacin adsorbed increased with an increase in temperature, quantity adsorbed increased from 41.08 to 70. 54 mg/g from 30 o C to 70 o C temperature range (Fig. 7). This speci es that the system considered is an endothermic system that can be related to an increase in the adsorptive sites in which cipro oxacin molecules penetrate [44]. This phenomenon also is presumed the movement of adsorbate molecules is increased with increasing temperature of the solution thereby cipro oxacin molecule may have generated energy that overcomes the activation barrier [44].

Kinetic studies
The rate-determining step and adsorption mechanism of this process were determined. This is achievable via the analysis of the process adsorption data using three different kinetic models. These models include pseudo rst order, pseudo-second order and intraparticle diffusion model. Considering the values reported in Table 3 pseudo second order best described the kinetics of this process. There is close agreement between q calculated and q experimental , with low SSE value and correlation coe cient (R 2 ) close to unity. The intraparticle diffusion model interprets the adsorption mechanism of this process. The q t value shows a linear correlation with t 1/2 (Fig. 8 iii), the slope and intercept of the plot are rate constant (K diff ) and boundary layer effect (C) respectively. In examining the intraparticle diffusion model in the adsorption process a linear regression plot of qt against t 1/2 that passes through the origin shows that the model is the main rate-determining step. Also, there is a boundary layer resistance at the solid-liquid interface of the process if it does not pass through the origin [46]. In this study, the adsorption rate is also controlled by the boundary layer because the plot did not pass through the origin that is the two processes occurred simultaneously [47].

Adsorption isotherm studies
From Figure 8 (i), a linear graph with 0.9774correlation coe cient R 2 value was obtained from evaluating the data using the Langmuir model. The dimensionless parameter R L determines the favourability of the process. A process is favourable if 0 < R L < 1, it is an irreversible process when R L = 1 and unfavourable when R L > 1 [48]. However, in this study the adsorption process is favourable. The monolayer adsorption capacity at the maximum level (q m ) was experimentally determined to be 222.22 mg/g. Further analysis of the data obtained for the Freundlich model with adsorption capacity (n) greater than 1 point to the favourability of the process [48].Although monolayer adsorption is prevalent in the cipro oxacin uptake onto BSAC, the Freundlich isotherm model has moderately high R 2 value which indicated some degree of multilayer adsorption in the system. The positive value of heat of adsorption obtained in the Temkin model analysis indicates an endothermic process. Also, the process mechanism is physisorption with E less than 8 kJ/mol. With reference to correlation coe cient R 2 the e ciency of the isotherms can be described thus: Langmuir (R 2 = 0.9774) > Freundlich (R 2 = 0.9671) > Temkin (R 2 = 0.8995) >Dubinin-Radushkevich (R 2 = 0.8383). This result shows that Langmuir isotherm best ts the adsorption of cipro oxacin onto BSAC. Similar trends of the e cacy of Langmuir model were reported using date pal lea ets [49], bamboo [44], different biomass wastes [50], Entermorphaprolifera [45]and bean husk [27] as adsorbent.

Adsorption thermodynamics studies
The interaction of the solid-liquid interface, spontaneity and feasibility of the process was investigated by

Cost Analysis
The summary of the cost of production of BSAC and the purchase of commercial activated carbon (CAC) is as presented in Table 7. The major signi cant cost in the production of BSAC includes the cost of energy, transportation, ortho-phosphoric acid and deionized water. The total cost of preparation and production of BSAC is 12.35 USD/kg (12,350 USD /ton). This amount is economically friendly compared with the cost of commercially activated carbon (270.8 USD/kg; 270,800 USD/ton). From this analysis, one kg of BSAC is about twenty-one (21) times cheaper than CAC. Hence, BSAC can be a good substitute for the commercial activated carbon.

Desorption and regeneration studies
Sequel to adsorption process the regeneration of exhausted adsorbent is very important in the water treatment process. The possibility of reuse of the adsorbent heightens the economic viability of the process [51,52]. For this study, ve different eluents were used for the evaluation of desorption e ciency which includes acetic acid (CH 3 COOH), water (H 2 O), potassium hydroxide (KOH), hydrochloric acid (HCl) and sodium hydroxide (NaOH). The percentage desorption e ciency was evaluated to be less than 10% with all the eluents. Acetic acid and water have the highest desorption e ciency (Table 6) whereas HCl desorbed 3.2%, NaOH desorbed 1.6% and KOH has the least desorption e ciency of 1.3%. The desorption e ciency follows the order CH 3 COOH = H 2 O > HCl > NaOH > KOH. Numerous active sites on the adsorbent surface aid large adsorption energy and this can be related to the low desorption e ciency. This in turn makes it di cult for the adsorbed molecule to be released from the surface and hence low desorption e ciency [53].Also, complexes might have been formed via interaction of the active sites and the functional groups of the adsorbates. This suggests that some amount of the adsorbates might have been chemisorbed onto the surface of the adsorbents [54].

Conclusions
This study considered the e ciency of the adsorptive capacity of the banana stalk in sequestering cipro oxacin in aqueous solutions. The feature of BSAC was investigated using SEM, FTIR, and pH pzc .
Considering the four isotherm models, the adsorption process was best described by the Langmuir model with a maximum monolayer adsorption capacity of 222 mg/g. Likewise in the kinetic study, the process was best interpreted by pseudo second order kinetics. The results obtained from the thermodynamic studies revealed that the adsorption process was endothermic, spontaneous and high randomness rate at the solid-liquid interphase. The cost analysis of the preparation and production of BSAC revealed that the adsorbent is about 21 times cheaper than CAC which makes it economically viable. Acid treated banana stalk is economically viable, e cient and effective for sequestering cipro oxacin from aqueous solutions.        Effect of contact time and initial cipro oxacin concentration Effect of adsorbent dosage on cipro oxacin adsorption Effects of solution pH Effects of temperature on cipro oxacin adsorption Cipro oxacin adsorption kinetics plot of (i) pseudo rst order (ii) pseudo second order (iii) intra particle diffusion modelat 303Kusing BSAC Figure 9 Cipro oxacin adsorption isotherms plots of (i); Langmuir (ii); Freundlich (iii), Temkin (iv) Dubinin-Radushkevich at 303 K using BSAC Figure 10 Vant Hoff plot of cipro oxacin adsorption onto BSAC

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. GRAPHICALABSTRACT.doc