Isotherm, Kinetics, and Adsorption Mechanism Studies of DTPA-Modied Banana/Pomegranate Peels as E�cient Adsorbents for Removing Cd(II) and Ni(II) from Aqueous Solution

16 Two novel absorbents were synthesized for the first time by banana and pomegranate 17 peels using diethylenetriaminepentaacetic acid (DTPA) modification to eliminate Cd(II) 18 and Ni(II) of sewage. The DTPA-modified peels performed significantly higher 19 adsorption capacity than unmodified materials. Adsorption isotherm and kinetics 20 models were simulated to determine their removal efficiency and potential for recovery 21 of these two heavy metals. As the results, the adsorption reached equilibrium within 5 22 minutes and was well described by the pseudo-second order model and Langmuir 23 isotherm. The surface morphology analysis of the synthetic materials by Scanning 24 Electron Microscopy-Energy Dispersive X-ray spectroscopy, Fourier Transform 25 Infrared spectroscopy and X-ray Photoelectron Spectroscopy, implied that ion 26 exchange, complexation and physical adsorption may together contribute to Cd(II) and 27 Ni(II) loading on DTPA-modified peels. This study demonstrates the feasibility of 28 waste peels as cost-efficient bio-absorbents to remove Cd(II) and Ni(II) in sewage 29 systems, and discovers potential adsorption mechanism of efficiency improvements 30 after DTPA modification. 31


Introduction
Various non-degradable heavy metal ions (HMIs) have received increasing attention on account of their serious threat to the ecosystems and human health by water pollution and food chain accumulation (Haroon et al. 2020, Liu et al. 2012, Nguyen et al. 2021, Oliveira et al. 2021, Zhao et al. 2021).The contamination of cadmium (Cd) and nickel (Ni) have always been a matter of great concern to mankind in many areas (He et al. 2021, Hezarjaribi et al. 2020, Teng et al. 2021).From the current data, the global concentrations of Cd and Ni in rivers and lakes have reached approximately 25.33 and 80.99 μg/L from 2010 to 2017, respectively (Zhou et al. 2020).As one of the most toxic HMIs in wastewater, Cd intake could lead to severe toxicity even at low concentrations (0.001-0.1 mg/L) (Gao et al. 2019), such as infertility, immune deficiency, kidney damage and femoral pain (Ruan et al. 2021), while exposure to Ni could cause allergies and cancer (Yoon et al. 2020).Eliminating Cd and Ni of sewage systems, therefore, becomes an urgent demand to prevent them from entering the human body.
Removal of HMIs by adsorption based on bio-materials, such as agri-wastes, activated sludges, plant tissues and their derivatives, has a broad application due to the low cost, high yield, renewability, less variables to control, and strong metal recovery capacity (Chen et al. 2016, Tan et al. 2016).Fruit peels, feathered with various functional groups (-OH, -COOH, -NH2, etc.) that exist in lignin, cellulose, hemicellulose, pectin and other tissues, can be used as the binding sites of HMIs (Vilardi et al. 2018).Banana and pomegranate are two popular and widely cultivated fruits in the world.The total production of banana per year is about 1.2×10 8 tons around the world, and the banana peel accounts for 30-40% (Albarelli et al. 2011).In 2017, the world's total pomegranate production was approximately 3.8×10 6 tons and half of pomegranate was composed of peel (El Barnossi et al. 2021).As reported previously, banana and pomegranate peels can be widely used as adsorbents to eliminate HMIs, to maximize the waste utilization and resolve environmental issues.However, the adsorption performance of raw peels was not satisfactory, and thus many chemical treatments were implemented to increase the adsorption capacity, such as acrylonitrile grafted cellulose, carbonization, alkalization and esterification (Zhou et al. 2017).The maximum adsorption capacity of Cd(II) and Ni(II) was only 35.52 and 27.40 mg/g for banana peel (BP) (Memon et al. 2008, Van Thuan et al. 2017), while 22.72 and 10.82 mg/g for pomegranate peel (PP) (Abedi et al. 2016, Khawaja et al. 2015).To further enhance the adsorption performance of agricultural waste peel on HMIs, a low-cost and efficient modification approach needs to be developed.
As a powerful metal chelating agent, diethylenetriaminepentaacetic acid (DTPA) can bind with a variety of metals to form stable complexes, and was frequently employed to enhance the performance of HMI adsorption through chemical modification (Repo et al. 2009, Repo et al. 2010).In this study, DTPA was utilized for the first time to modify banana/pomegranate peel through an esterification reaction with hydroxyl, after defatting and deproteinizing.Abundant-COO from DTPA were loaded to the surface of the adsorbents, providing binding sites for HMIs.To this end, we (1) synthesized two novel bio-adsorbents modified by DTPA to improve the removal performance of HMIs; (2) optimized the parameters in the adsorption process by adjusting pH, initial concentration, and contact time; (3) explored the potential reaction mechanisms of HMIs loaded onto bio-materials in depth via Electron Microscopy-Energy Dispersive X-ray spectroscopy (SEM-EDX), Fourier Transform Infrared spectroscopy (FTIR) and X-ray Photoelectron Spectroscopy (XPS).This study confirmed the feasibility of developing new bio-based HMI adsorbent with fruit peels and its potential application in agricultural waste treatment and environmental purification.

Materials and reagents
Banana from Nanning, Guangxi Province, and the pomegranate from Xian, Shaanxi Province of China, were purchased from the local market.The peels were cleaned for five times by deionized water to remove the contaminants attached to the surface, and then dried at 50℃.Next, the obtained peels were cut into small pieces of about 1 cm, then crushed with a grinder and screened using Taylor standard sieve of 40 mush.In this work, reagents for preparing ionic solution were all analytically pure without additional processing, including Cd(NO3)2 and Ni(NO3)2, provided by Shanghai Sinopharm Chemical Reagent Co., Ltd.Diethylenetriaminepentaacetic acid (DTPA, purity 99%) and N-N-dimethylformamide (DMF, purity 99.5%) and were provided by ThermoFisher Scientific Co., Ltd.Deionized water (18.25 MΩ cm) was adopted throughout this study.
Finally, DTPA modified banana and pomegranate peels (assigned as DMBP and DMPP) were obtained after drying to constant weight in an oven at 50°C.

Adsorption experiments
The standard solutions (500 mg/L) of Cd(II) and Ni(II) were formulated by dissolving the Cd(NO3)2 and Ni(NO3)2 into deionized water, respectively, which were  (Zhang et al. 2020).All these experiments were repeated in triplicate and the equilibrium adsorption capacity (Qe) was expressed by the Eq. ( 1).
Where Qe is the equilibrium adsorption capacity (mg/g), C0 and Ce are the initial and equilibrium concentrations of Cd(II) and Ni(II) in solution (mg/L), respectively; V is the solution volume (L), and W is the weight of adsorbent (g).
The dimensionless separation constant RL is often employed to characterize the Langmuir adsorption isotherm, which can be used to judge whether the adsorption process is favorable.When RL = 0, the adsorption is irreversible; when 0 < RL <1, the reaction is favorable; the reaction is unfavorable when RL > 1, and RL can be expressed by the following equation:

Surface characterization
The surface features of the adsorbents were characterized by a field emission high resolution Scanning Electron Microscope (SEM; Apreo, FEI Inc., USA), equipped with EDX for semi-quantitative elemental analysis (Zhang et al. 2018).The information of surface functional groups was detected by Fourier Transform Infrared spectroscopy (FTIR; Nicolet IS50, ThermoFisher Scientific Inc., USA) within 4000-400 cm −1 with a resolution of 4 cm −1 (Huang et al. 2019).Composition and binding states of major elements were acquired using X-ray Photoelectron Spectroscopy (XPS; ESCALAB 250Xi, ThermoFisher Scientific Inc., USA).The C1s peak at 284.8 eV was performed as the standard of calibration (Ding et al. 2020).

Modified banana and pomegranate peels improved HMI removal efficiency
As presented in Fig. 1, the removal rate of unmodified peels on Cd(II) and Ni(II) were intensely limited, especially for pomegranate peel (33.23% and 32.50% for UBP; only 8.84%, 8.67% for UPP).Therefore, a series of modification operations including mercerization and esterification reaction were performed to release -OH and introduce Obviously, the adsorption effect of Cd(II) and Ni(II) on DMBP is more favorable than that on DMPP, and the removal rate of both materials on Cd(II) is much higher than Ni(II).

Effect of initial pH
The effects of initial pH on the adsorptive performance of DMBP and DMPP were explored.pH 3-7 was selected since the alkaline conditions could induce ion precipitation (Zhang et al. 2020).As depicted in Fig. 2 a, DMBP and DMPP exhibited poor loading capacities of Cd(II) and Ni(II) at pH 3, which may be ascribed to that the existence of large quantity of protons compete with HMIs for binding sites (Bulin et al. 2020, Zhang et al. 2021).The adsorption capacity gradually increased with the pH from 3 to 6, whereas a downward trend in Cd(II)-adsorption curve appeared at pH=7.
Meanwhile, the removal efficiency of both materials for Cd(II) was more excellent than Ni(II), and DMBP has better adsorption performance for Cd(II) and Ni(II).Langmuir and Freundlich isotherm models were applied to assess the relationship between loading capacity and initial concentrations.As shown in the model parameters in Table 1, the correlation coefficients R 2 of the Langmuir isotherm model were 0.974-0.998,which were higher than those of the Freundlich model (R 2 were 0.757-0.874), indicating that the adsorption of DMBP and DMPP on Cd(II) and Ni(II) might be monolayer and homogeneous (Yuan et al. 2017).In this work, the values of Kc were 0.037-0.250for DMBP and DMPP, and the values of RL were between 0-1, suggesting that the adsorption of Cd(II) and Ni(II) on both materials were favorable.Furthermore, the calculated Qm of Cd(II) and Ni(II) for DMBP were 46.729 and 29.240 mg/g, respectively, while 46.296 and 16.611 mg/g for DMPP (Table 1), significantly higher than the previous researches (Abedi et al. 2016, Khawaja et al. 2015, Memon et al. 2008, Van Thuan et al. 2017).

Effect of contact time and kinetic models
To investigate the adsorption kinetics of the adsorbents (DMBP and DMPP) on Cd(II) and Ni(II), a time-coursed adsorption experiments were performed (Fig. 2 c), and the obtained kinetic data were fitted by the pseudo-first-order and pseudo-secondorder kinetic models, which were displayed in Table 2.The adsorption process reached equilibrium within 5 min, which was much faster than previous studies (at least 60 min) (Abedi et al. 2016, Khawaja et al. 2015).The combination of rapid adsorption and high adsorption capacities demonstrated that DMBP and DMPP were desirable for HMI removal.
As shown in Table 2, for both DMBP and DMPP, the values of the correlation coefficient (R 2 ) calculated from the pseudo-second-order kinetic model (R2 2 = 0.994-1) were larger than that from the pseudo-first-order kinetic model (R1 2 =0.825-0.936),and the theoretical values of Qe were closer to the actual measured values.This phenomenon suggested that the adsorption process could be better described by the pseudo-second-order model.As we all know, the pseudo-first-order model representatives the physical adsorption that the diffusion between the adsorbate and the binding sites dominants the adsorption rate, while the pseudo-second-order model      (Lessa et al. 2017).The stretching vibration of C=O at 1617 cm -1 confirmed the presence of acidic oxygen-containing groups in the adsorbents, which might contribute to the HMI adsorption (Fan et al. 2020).The peaks at 1037 and 1049 cm -1 were attributed to C-O stretching (Shen et al. 2021), and the bands around 1734 and 1732 cm -1 were ascribed to N-H stretching.After mercerization, the disappearance of N-H signal probably was caused by the removal of protein, meanwhile, the reduction of C-H and C=O peak area might be attributed to the degreasing treatment.Moreover, the increase of peak intensity at 1037 and 1049 cm -1 illustrated the increase in the proportion of cellulose in MBP and MPP, which provided another evidence for the effectiveness of mercerization (Zhang et al. 2018).After modification with DTPA, the peaks around 1628 and 1020 cm -1 were broadened, which suggested that DTPA was successfully grafted onto the materials and large quantities of -COO were introduced, providing more binding sites for the capture of Cd(II) and Ni(II).After binding with HMIs, blue shift of C=O and C-O peaks occurred (from 1628 cm -1 and 1020 cm -1 to 1630 cm -1 and 1023 cm -1 for DMBP, from 1629 and 1025 cm -1 to 1636 and 1050 cm -1 for DMPP, respectively), which might be due to the interaction between high electron density of HMIs and oxygen-containing groups (Ding et al. 2016).Meanwhile, the reduction of the peak area near 3440, 1630 and 1020 also provided a theoretical basis for the surface complexation between these oxygen-containing groups and metal ions.

Conclusion
Two novel adsorbents were prepared by mercerizing and then esterification using DTPA for the first time.
diluted to specified concentrations by deionized water if needed.The effects of pH, initial concentration, contact time and competitive adsorption of Cd(II) and Ni(II) were investigated.1 M HCl and NaOH were adopted to adjust pH with 3-7.The adsorption isotherms were explored by assessing the loading capacity of Cd(II) and Ni(II) under different initial concentration gradients of Cd(II) (30-250 mg/L) and Ni(II) (20-150 mg/L).The samples of adsorption kinetics experiments were harvested at different time points within 180 min.All batch experiments were performed in Erlenmeyer flasks containing 50 mL HMI solution in a shaker at 200 rpm.Besides, adsorbent dose was maintained at 1 mg/ml, initial ion concentrations at 70 mg/L (Cd(II)) and 50 mg/L (Ni(II)), at 25°C, unless specified otherwise.Next, the mixture of solid and liquid was separated by centrifugation at 10,000×g and filtered with 0.22 μm Nylon filter.Both initial and residual concentrations of Cd(II) and Ni(II) were detected by an Inductively Coupled Plasma-Atomic Emission Spectroscopy (ICP-OES, Avio 200, PerkinElmer, USA)

Fig. 1
Fig.1The fabrication of DTPA modified material composites and the removal rate before and after

Fig. 2
Fig. 2 Effects of solution pH, initial concentration, contact time on adsorption capacity.(a) banana reveals the chemical adsorption such as ion exchange and covalent bonding (Hydari et al. 2012).The better fitting with the pseudo-second-order model indicated that the capture behaviors of both DMBP and DMPP towards Cd(II) and Ni(II) were dominated by chemical adsorption.
To compare the adsorption effects of DMBP and DMPP on Cd(II) and Ni(II), a mixed adsorption experiment was implemented.The loading capacities of DMBP and DMPP for Cd(II) were stronger than Ni(II) in the mixed ion solution (Fig.3), demonstrating that DMBP and DMPP had higher absorptivity to Cd(II) than Ni(II), which were consistent with previous results(Cutillas-Barreiro et al. 2016, Liu et al. 2013, Wen &Hu 2021).Moreover, the amount of HMIs captured by DMBP and DMPP in the mixed solution of Cd(II) and Ni(II) was close to the maximum loading capacity in the presence of a single Cd(II), suggesting that the binding sites were fully occupied.

Fig. 3
Fig. 3 Comparison of the mixed ion adsorption and the single ion adsorption.(a) banana peel; (b)

Fig. 4 Fig. 5
Fig. 4 Surface morphology characterization of the adsorbents by SEM

Fig. 7
Fig. 7 High-resolution XPS spectra of adsorbents.(a) full spectrum of adsorbents before and after

Figures Figure 1
Figures COO groups onto the material surface.It was clear that the removal rate to Cd(II) and Ni(II) were significantly increased by 11.22% and 4.76%, to 44.45% and 37.26% for DMBP, while increased by 19.40% and 18.24%, to 28.24% and 26.91% for DMPP. -

Table 1
Parameters of the Langmuir and Freundlich isotherm models in single ion system under different initial concentrations

Table 2
Kinetics parameters for the adsorption of Cd(II) and Ni(II) on DMBP and DMPP Compared with UBP/UPP, DMBP/DMPP performed greater adsorption capacities (46.729/46.296mg/g for Cd(II), and 29.240/16.611mg/g for Ni(II)), which could be employed as efficient adsorbents to remove HMIs from sewage.Isotherm and kinetic modeling.Journal of Water Process Engineering.https://doi.org/10.1016/j.jwpe.2020.101577He J, Lu Y, Luo G (2014) Ca(II) imprinted chitosan microspheres: An effective and green adsorbent for the removal of Cu(II), Cd(II) and Pb(II) from aqueous Hezarjaribi M, Bakeri G, Sillanpää M, Chaichi MJ, Akbari S (2020) Novel adsorptive membrane through embedding thiol-functionalized hydrous manganese oxide into PVC electrospun nanofiber for dynamic removal of Cu(II) and Ni(II) ions from aqueous solution.Journal of Water Process Engineering.