Effective biosorption of arsenic from water using La(III) loaded carboxyl functionalized watermelon rind

2.1 Materials

The watermelon rind was collected from local juice vendor located at Ratnapark, Kathmandu, Bagmati Province, Nepal. It was washed several times with distilled water, cut into pieces and dried in sun for several days. Then, the dried watermelon rind pieces were pulverized using a mechanical grinder, sieved to allow pass through 150-µm mesh, finally dried in a hot air oven at 70 °C for 24 hrs. The dried powder so obtained was termed as raw watermelon rind henceforth abbreviated as RWR. The chemicals lanthanum chloride heptahydrate (LaCl₃·7H₂O), calcium hydroxide (Ca(OH)₂), and arsenic trioxide (As₂O₃) were of analytical grade and purchased from Sigma-Aldrich, New Delhi, India. The hydrochloric acid (HCl), sodium hydroxide (NaOH), sodium chloride (NaCl), potassiumdihydrogenphosphate (KH₂PO₄), sodium sulphate (Na₂SO₄) and potassium nitrate (KNO₃) supplied by Merck Chemicals Co. Ltd, were used. The stock solution (1000 mg/L) of As(III) was prepared by dissolving 1.32 g of arsenic trioxide (As₂O₃) in 20 mL of 10 M NaOH solution followed by the neutralization using nitric acid and diluted to 1000 mL. The double distilled water was used for the frequent dilutions of the stock solutions at the time of experiment.

2.2 Analytical and characterization techniques

The scanning electron microscopy (SEM, JEOL JSM – 6700F instruments, Jeol Ltd, Tokyo, Japan) was operated to analyze the surface morphology of the biosorbents. An energy-dispersive X-ray spectrometer (EDX, JEOL model JSM) was used for the elemental analysis and elemental mapping for the sample of SWR, La(III)-SWR and As-La(III)-SWR. Fourier transform infrared spectrometer (IR Affinity −1S-SHIMADZU spectrometer, Kyoto, Japan) was run to analyze the surface functional groups of the RWR, SWR and La(III)-SWR biosorbents in the wavenumber range from 4000 to 600 cm⁻¹. X-ray photoelectron spectrometer (Nexsa XPS system, Thermo-Fisher Scientific, UK) with an Al Kα excitation source was used to analyze the surface composition, valence state of adsorbed species and binding nature of As(III) onto La(III)-SWR. Zeta potential analyzer (HORIBA Scientific SZ- 100) was used to analyze the surface charge and identify the pH_PZC value of the investigated biosorbent. The inductively coupled plasma mass spectrometer (ICP-MS, Agilent 7900, Santa Clara, CA, USA) was operated to analyze the arsenic concentration in the solution before and after biosorption whereas pH meter (CHEMI LINE CL-180) was used to measure the pH of the solution.

2.3 Preparation of La(III) loaded saponified watermelon rind

Incipiently, about 100 g of powdered RWR was mixed with a 500 mL saturated solution of Ca(OH)₂ in a conical flask. The pH of the suspension was maintained to 12.0 by adding 3/4 pellets of NaOH to enhance saponification of methyl ester part of watermelon pectin to the carboxyl group as shown in scheme 1 (step 1). The mixture was agitated for 24 hrs at a speed of 190 rpm to complete the saponification reaction. The product was washed almost upto neutral pH and was dried at 343 K which is referred to as saponified watermelon rind henceforth abbreviated as SWR. SWR act as cation exchanger carrying exchangeable Ca(II), where La(III) was loaded according to the procedure mentioned elsewhere with modification (Paudyal et al, 2011; Ghimire et al., 2007; Poudel et al., 2021). About 3 g of SWR was mixed with 500 mL of 0.1 M LaCl₃·7H₂O solution at neutral pH to load La(III) ion onto SWR. The pH of the mixture was adjusted by addiing diluted solutions (0.1 M each) of HCl and NaOH. Afterwards, the mixture was agitated for 24 hrs at a speed of 200 rpm. La(III)-loaded SWR was filtered and washed several times with distilled water almost to neutral pH and dried in an oven at 343 K for 48 hrs. After all, prepared La(III)- loaded SWR biosorbent henceforth abbreviated as La(III)-SWR. It is inferred that La(III) ion existed in LaCl₃·7H₂O solution substitutes Ca(II) present in the SWR by cation exchange reaction during La(III) loading to give La(III)-SWR as shown in Scheme 1 (step 2).

Close

Scheme 1

Synthetic route of La(III)- SWR)adsorbent from watermelon waste.

Export to PPT

2.4 Adsorption tests

The performance of As(III) biosorption onto RWR and La(III)-SWR were investigated in batch equilibration mode. The biosorption studies were carried out in the pH range of 2–13 to figure out the optimum pH. The pH of the solution was adjusted by micro-addition of diluted solution of HCl or NaOH. The effect of the initial concentration of As(III) in the range of 5 – 400 mg/L was studied by agitating the biosorbents at solid liquid ratio of 1 g/L. The effect of As(III) concentation for the uptake of As(III) was tested at 298 K, 303 K and 308 K temperatures to investigate the thermodynamic parameters. Arsenic removal performance of La(III)-SWR from actual arsenic polluted groundwater was studied at different solid liquid ratio ranging from 0.5 to 5 g/L. All the biosorption experiment for As(III) removal were studied by stirring the mixture for 24 hrs at the speed of 200 rpm in a rotary shaker except kinetic studies. The effect of contact time was investigated by equilibrating 25 mg of La(III)-SWR with 25 mL of As(III) solution (10 mg/L) in the range of 5 up to 360 min. After shaking, the suspensions were instantly filtered using a Whatman 42 filter paper (Whatman International Ltd Maidstone, England). The concentration of As(III) prior to and after biosorption were evaluated. All the experiments were repeated in triplicate, and the average of the triplicate values are presented in this paper. The experimental precision is higher than 98%. In all tests, the adsorbed amount of As(III) per unit mass of biosorbent, q_e,(mg/g) and the removal efficiency (R %) were enumerated from the following mass balance relationships,

where, C_o and C_e represent the original and equilibrium concentration of As(III) in the solution (mg/L), V is the volume of solution (L) and M is the mass of biosorbent used in gram.

2.5 Biosorption isotherm studies and modelling

The evaluation of biosorption isotherm for the specific sorption system is significant to determine the maximum sorption capacity, adsorption equilibrium, investigate the closely fitted biosorption model (Cruz et al., 2020). The batch data for biosorption of As(III) onto La(III)-SWR were fitted using two commonly used Langmuir and Freundlich isotherm models. The Langmuir isotherm model assumed that all the biosorption sites have equivalent affinity in which uniform monolayer biosorption takes place. The Langmuir isotherm can be expressed in its non linear form by the following equation as

The linear form of the Langmuir equation can be represented as,

where, C_e is the equilibrium concentration of arsenic (mg/L), q_m is the maximum As(III) biosorption capacity (mg/g), q_e is the biosorption capacity at equilibrium (mg/g) and b is Langmuir constant related to equilibrium constant (L/mg). The parameters q_m and b, were evaluated from the plot of C_e/q_e versus C_e.

Another imperative dimensionless parameter (R_L) was applied to analyze the validity of the biosorption process in Langmuir model, which can be represented by the following equation

where, b is Langmuir equilibrium constant (L/mg) and C₀ is the original (initial) concentration of As(III) in the solution. If the value of R_L is 0 < R_L less than 1 (favorable), whereas R_L > 1 (unfavorable), R_L = 1(linear), and R_L = 0 (irreversible) sorption process.

The Freundlich isotherm model assumed multilayer biosorption of adsorbate onto the heterogeneous biosorbent surfaces with non-uniform energy distribution on the binding sites. The Freundlich isotherm is expressed by the following relation as

The linear equation can be obtained after taking the natural logarithm of Eq. (6) on both sides as

where, C_e is the equilibrium concentration (mg/L), q_e is the biosorption capacity at equilibrium (mg/g) and K_F is Freundlich constant (mg/g) (L/mg)^1/n related to biosorption capacity and 1/n is dimensionless heterogeneity factor indicates the biosorption intensity of the biosorbent. These parameters can be evaluated from the plot of log q_e versus log C_e.

The distribution coefficient (K_D) signifies the binding capacity of the biosorbent to retain the As(III), which is expressed as the ratio of As(III) concentration adsorbed onto La(III)-SWR and equlibrium concentration of As(III) remain in a solution in mg/L. It is represented in the equation as (Poudel et al., 2020).

where, C₀ and C_e are the initial and equilibrium concentrations (mg/L) of As(III), respectively, V is the volume of solution in litre and M is the mass of biosorbent in gram. Greater value of K_D prevails the effectiveness of the biosorbents as the solid phase distribution of As(III) increased with the increase of K_D value (Yantasee et al., 2007).

2.6 Biosorption kinetic studies and modelling

The kinetic study is an important characteristic to identify the biosorption performance of biosorbent estimating the equilibrium time, biosorption rate and plausible mechanism for the investigated biosorption process (Lagergren 1998; Simonin, 2016). The pseudo first order model in the non-linear form is expressed as follows,

The Eq. (9) can be rearranged into linear form as

where, q_e and q_t are the amounts of As(III) adsorbed per unit mass of the biosorbent (mg/g) at equilibrium, and at time t, respectively, and k₁ is the first order rate constant (min⁻¹). A plot of log (q_e- q_t) versus t can be used to evaluate q_e and k₁ values.

Similarly, the pseudo second order model can be expressed in the non-linear form as,

The Eq. (11) can be rearranged into linearized form as (Blanchanchard, 1984; Ho, 2006),

Where k₂ is the second order rate constant (g/(mg min)). A plot of t/q_t versus t can be used to evaluate q_e and k₂ values.

2.7 Interference of coexisting ions

The interference on the biosorption of As(III) onto La(III)-SWR in the solution including coexisting ions such as chloride, nitrate, sulphate and phosphate have been studied. 25 mg of La(III)-SWR with a different concentration range of (0–200) mg/L each coexisting ions with 10 mg/L of As(III) concentration were equilibrated for 24 h at optimum pH. The suspensions were filtered and the equilibrium concentration of As(III) were analyzed for all the samples.

2.8 Elution studies

Firstly, the As(III) sorbed La(III)-SWR was prepared by equilibrating 2 g of La(III)-SWR with 200 mL of As(III) solution (120 mg/L) at an optimum pH of 12.0 for 24 h. Then the suspension was filtered and the As(III) concentration from the filtrate was analyzed whereas, the residue was washed to remove un-sorbed As(III) for repeated times and finally dried at 343 K. Then, 50 mg of As(III) loaded La(III)-SWR was agitated with 15 mL of different concentration range of NaOH (0.001– 2 M) for 24 h. After filtration, the filtrate was examined separately for desorbed amount of As(III). The elution efficiency (% D) of As(III) was obtained from the following equation,

where A_r and A_a are the amounts of As(III) released and adsorbed (mg), respectively.

After elution, the residue was water washed to removed ing until neutral pH, the La(III)-SWR is regenerated. The biosorption of As(III) using this regenerated adsorbnet shows insignificant reduction of As(III) uptake capacy (less than 2%) with no leakage of loaded La(III). The result strongly suggest the high stability of La(III)-SWR during As(III) biosorption and elution process.

2.9 X-ray photoelectron spectroscopic (XPS) analysis

The oxidation state of biosorbed species of arsenic and elemental composition after the sorption of arsenic onto La(III)-SWR were analyzed by an X-ray photoelectron spectrometer (Nexsa XPS system, Thermo Fisher Scientific, UK) equipped with a monochromatic Al Kα X-ray source (1486.6 eV).

2.10 Statistical analysis

To evaluate the accuracy and reproducibility of sorption process, various statistical treatment such as regression coefficient (R²), Chi-square test (χ²), root-mean square error (RMSE) and mean absolute error (MAE) were done for kinetic and isotherm data. The mathematical expression for their determination are represented by the supplimentary equations SE1, SE2, SE3 and SE4 as shown in supplimentary document (Ncibi et al., 2007).

3.1 Characterization of the biosorbents

3.1.1

3.1.1 Surface morphology of the biosorbents

Fe-SEM was applied to examine the surface morphological characteristics of the biosorbent. Fe-SEM images of RWR, SWR, La(III)-SWR prior and after As(III) biosorption are shown in Fig. 1 (a), (b),(c) and (d), respectively. As seen in the images, RWR surface is smooth with small white patches (Fig. 1a) which may be due to the existence of sugar, limonene and low molecular weight organic compounds which are expected to be leached out after base treatment, the SWR surface is little bit rough and has limited hollow cavities (Fig. 1b) which can be reasonablly interpreted due to the leaching of soluble organic components like sugar and limonene molecule from the watermelon rind surface during Ca(OH)₂ treatment while La(III)-SWR (Fig. 1c) has uneven cavities with irregular hollow protuberances. After sorption of As(III) onto La(III)-SWR (Fig. 1d), its surface became little bit uniform, smooth and some shining or whitish layer was observed on the surface, which is attributed due to the coverage of adsorbed As(III) anion on the surface of La(III)-SWR during the biosorption process.

Close

Fig. 1

SEM images of (a) RWR (b) SWR (c) La(III)-SWR prior biosorption (d) La(III)- SWR after biosorption of As(III).

Export to PPT

3.1.2

3.1.2 Energy dispersive X-ray (EDX) spectra with elemental mapping

The clear evidence for the loading of La(III) in SWR and biosorption of As(III) onto La(III)-SWR were investigated from elemental analysis using EDX spectroscopy. An EDX spectra of SWR, and La(III)-SWR before and after As(III) biosorption were recorded and results are presented in Fig. 2 together with elemental mapping. From the results of elemental mapping, C, O and Ca are the major elements in SWR (Fig. 2a), C, O, and La are in maximum amount in La(III)-SWR (Fig. 2b) whereas the elements such as C, O, La, and As are detected in large proportion in As-La(III)-SWR (Fig. 2c). The result from the EDX spectrum and elemental composition of SWR (Fig. 2d and Table 1) revealed that SWR contains 54.22%, 41.04%, and 0.41% of C, O, and K at binding energy around 0.27, 0.51, and 3.31 keV, respectively in addition to the peaks of Ca (4.33%) at 3.69 and 4.12 keV. After La(III) loading, a new peak corresponding to elemental La(10.67%, Table1) appeared in the EDX spectrum of La(III)-SWR at binding energy around 0.82, 4.71, 5.12, 5.44, 5.82 and 6.01 keV together with the peaks of SWR as shown in Fig. 2(e), which clearly provides the strong evidence of loading La(III) onto SWR. Moreover, the percentage of calcium decreased from 4.33% to 1.08 % after La(III) loading reaction, which is reasonably attributed due to the formation of lanthanum pectate according to the Scheme 1 by replacing calcium from SWR with La(III) ion. After As(III) sorption (As-La(III)-SWR), new peak corresponding to As element was observed at binding energy around 1.28 keV (Fig. 2f, Table1) together with all other elements existed in La(III)-SWR, which strongly provide the direct evidence for the bisorption of As(III) onto La(III)-SWR. The percentage of As (III) adsorbed onto La(III)- SWR was found to be 1.32 % under examined experimental conditions. The weight ratio of Ca element in La(III)-SWR to SWR was less than unity (0.11), which demonstrated that La(III) loading occured by substituting Ca(II) from SWR via a cation exchange mechanism as expressed by the reaction in Scheme 1. The weight percentage (wt%) of oxygen increased from 38.07 to 44.88% after As(III) biosorption by La(III)-SWR in addition to the appearence of new peaks of arsenic, suggesting the biosorption of complex ion containing arsenic and oxygen (oxo-anionic species of As) during As(III) biosorption process. These results clearly provide the direct evidence of loading La(III) by cation exchange mechanism and biosorption of As(III) anion by ligand exchange mechanism.

Close

Fig. 2

Energy dispersive X-ray (EDX) analysis of SWR, La(III)-SWR, and As-La(III)-SWR showing respective (a), (b), and (c) elemental mapping, (d), (e), and (f) EDX spectra.

Export to PPT

Table 1 Elemental compositions of SWR, La(III)-SWR , and arsenic loaded La(III)-SWR (As-La(III)-SWR).

Elements	BE (keV)	As- La(III)-SWR (Wt%)	La(III)-SWR (Wt%)	SWR (Wt%)
C	0.27	42.73	49.67	54.22
O	0.52	44.88	38.07	41.04
P	2.01	0.74	0.42	ND
Si	1.73	ND*	0.09	ND
K	3.31	ND*	ND*	0.41
Ca	3.69, 4.12	0.07	1.08	4.33
As	1.28	1.32	ND*	ND*
La	0.82, 4.71, 5.12, 5.44, 5.82, 6.01	10.26	10.67	ND*
Total:		100.00	100.00	100.00

* ND = not detected

3.1.3

3.1.3 Functional groups analysis using FTIR spectroscopy

Fig. 3 displayed the Fourior transform infrared (FTIR) spectra of RWR, SWR and La(III)- SWR recorded in the range from 4000 to 600 cm⁻¹. In case of RWR, broad and strong band observed at around 3320 is due to OH stretching vibration of cellulose, hemicellulose, pectin and lignin exist in watermelon biopolymer. The peak at around 2924 cm⁻¹ is due to –CH stretching vibrations of methyl and methoxy groups (Guptaa et al 2015). The peak at around 1730 cm⁻¹ may be ascribed due to -C = O stretching vibrations of carboxylic acid or esters and symmetric and asymmetric vibrations of ionic carboxylic groups (–COO-). Small peaks around 1412, 1012 cm⁻¹, respectively may be attributed due to C-O stretching vibrations of carboxylic acids and alcohols (Ben-Ali 2021). After saponification, the peak observed at around 1730 cm⁻¹ is largely diminished in SWR at around 1608 cm⁻¹ and La(III)-SWR at around 1595 cm⁻¹ which may be due to the formation of metal carboxylate that is calcium carboxylate in case of SWR during saponification and lanthanum carboxylate in case of La(III)-SWR via exchange of calcium with lanthanum during La(III) loading process through cation exchange reaction. The shifting of carboxyl peak in La(III)-SWR towards lower frequency region is due to the substitution of higher molecular weight La(III) to lower molecular weight calcium. A part of FTIR spectra from 1500 to 2000 cm⁻¹ was replotted and presented in Supplementary Fig. 1 (Fig. S1). It shows the clear vision for chemical modifications and spectral shift. This fact intimates carboxyl group is main functional group which was modified where favorable sorption sites for anionic species like As(III) anion have developed by La(III) loading reaction.

Close

Fig. 3

Functional groups analysis of RWR, SWR and La(III)-SWR using FTIR spectroscopy.

Export to PPT

3.2 Biosorption of As(III) in batch wise mode

3.2.1

3.2.1 Effect of solution pH and As(III) biosorption mechanism

The surface charge of the biosorbent and speciation of the adsorbate in the solution is the pH-dependent fundamentals that determine the uptake capacity during biosorption (Xu et al., 2021). pH is the hallmark for the biosorption of As(III) onto the biosorbent. The biosorption of As(III) onto RWR and La(III)-SWR has been studied in the pH range of 2.0 – 13.0 as displayed in Fig. 4(a). The biosorption of As(III) onto RWR is insignificant (less than 0.5 mg/g even at optimal pH) while biosorption of As(III) onto La(III)-SWR exhibits remarkable removal efficiency. The result declared that the adsorbed amount of As(III) gradually increases from 0.64 mg/g at pH 2.19 to a maximum of 4.68 mg/g till pH 12.08 and decreases to 2.53 mg/g at pH 13.02. Therefore, the optimum equilibrium pH was marked to be 12.08. Depending on the chemistry of aqueous solution, As(III) exists predominantly in the pH range of 2.0–7.0 as neutral species (H₃AsO₃), at pH range of 7.0–12.0 significantly as anionic species (H₂AsO₃^-) and (HAsO₃^2-) while at pH > 12.0, AsO₃^3- is prevalent. At lower pH between 2.19 and 7.03, the ligand exchange reaction between hydroxyl ligand of La(III)-SWR and neutral species of As(III) i.e. H₃AsO₃ is not favorable resulting insignificant biosorption of As(III) anion. The biosorption of As(III) gradually increased from pH 7.62 – 12.08 and declined at pH 13.02. Evaluated optimum pH in this study is similar to the starch stablized ferrromanganese binary oxide adsorbent (Xu et al., 2021).The monovalent species of As(III) is predominate at pH around 12 thus the exchange of OH by this monovalent As(III) species (H₂AsO₃^-) is more favorable resulting enhanced uptake of As(III) at this pH as shown in Scheme 3. The decrease in As(III) biosorption at pH > 12.0 may be due to competition between anionic species of As(III) and hydroxyl ion as shown in elementary reaction ‘a’. P≡La-OH + H₂AsO₃^- → P≡La- H₂AsO₃ + OH^– (a)

Close

Fig. 4

Biosorption of As(III) onto La(III)-SWR (a) Effect of pH and (b) logK_D versus pHe conditions: weight of biosorbent = 25 mg, volume of solution = 25 mL, temperature = 298 K, As(III) concentration = 10 mg/L, shaking time = 24 hrs, shaking speed = 200 rpm.

Export to PPT

Close

Scheme 2

Generation of exchangeable hydroxyl ligands in SWR after La(III) loading for As(III) removal.

Export to PPT

Close

Scheme 3

Inferred mechanism for the biosorption of As(III) onto La(III)-SWR via ligand exchange reaction.

Export to PPT

Furthermore, As(III) is partly oxidized into As(V) due to aerial oxidation during the biosorption process (Poudel et al., 2020) which will be further confirmed by recording the XPS spectra of La(III)-SWR after As(III) sorption. The oxidized species H₂AsO₄^- may be biosorbed onto La(III)-SWR according to reaction ‘b’. P≡La-OH + H₂AsO₄^- → P≡La- H₂AsO₄ + OH^– (b)

The solid phase distribution of As(III) onto the investigated La(III)-SWR can be clearly understood from the knowledge of distribution coefficient at different equilibrium pH. Fig. 4(b) shows the natural logarithm value of distribution coefficient (logK_D) evaluated for the sorption of As(III) onto La(III)-SWR as a function of equilibrium pH of the solution. The result of this figure showed that logK_D value increases from −1.17 to −0.87 with the increase of equilibrium pH of the solution from 2.19 to 7.26 and reached maximum value (-0.31) at 12.08 then it was decreased with further increase of equilibrium pH. The highest value of logK_D was observed at pH 12.08 where solid phase distribution of As(III) is maximum during sorption process, which also provides the strong evidence of optimum biosorption of As(III) onro La(III)-SWR at this pH.

3.2.2

3.2.2 Effect of contact time

Since biosorption of As(III) onto RWR was poor at wide pH ranging from 2 to 13, so the kinetic experiment was performed only with La(III)-SWR at optimum pH of 12.08 to estimate the rate of biosorption, which is an important characteristic to identify the biosorption performance of the biosorbent. Fig. 5 shows the As(III) uptake capacity of La(III)-SWR as a function of contact time. It shows that the removal of As(III) increases sharply at the initial phase of contact whereas it becomes steady and finally achieved equilibrium within 6 hrs. The biosorption rate is found to be proportional to the number of unoccupied adsorption sites. It may be reasonably attributed that at the initial phase of biosorption, a significant number of unoccupied sites were available for As(III) biosorption resulting the increase of biosorption rate whereas at equilibrium all of the unoccupied sites were successively saturated. The kinetic behaviour of biosorption of As(III) were analyzed to examine the best fit kinetic model according to Eqs. (9) and (11) and the results were portrayed in Supplementary Fig. 3 (Fig. S3) and Supplementary Fig. 4 (S4). The estimated pseudo first order and pseudo second order kinetic parameters are shown in Table 2 along with error functions.

Close

Fig. 5

Biosorption kinetics of As(III) onto La(III)-SWR, conditions: weight of biosorbent = 25 mg, volume of solution = 25 mL, As(III) concentration = 10 mg/L, pH 12.08, shaking speed = 200 rpm and temperature = 298 K.

Export to PPT

Table 2 Kinetics parameters for the adsorption of As(III) onto La(III)-SWR.

Kinetic Models	Parameters
Pseudo first order	k1 × 10-3 (min−1)	14.3 ± 0.43
	qe, cal. (mg/g)	5.80 ± 0.91
	R2 χ2 RMSE MAE	0.92 0.18 ± 0.36 3.12 ± 0.54 0.89 ± 0.35
Pseudo second order	k2 × 10-3 (g/(mg min))	3.68 ± 0.05
	qe, cal.(mg/g)	5.40 ± 0.13
	R2	0.97
	qe, exp. (mg/g) χ2 RMSE MAE	4.78 ± 0.09 0.0076 ± 0.01 0.36 ± 0.05 0.105 ± 0.02

The correlation coefficient (R²) value for pseudo second order model is found to be (0.97) is far better compared to pseudo first order model (R² = 0.92). Similarly the value of q_e obtained from pseudo second order kinetic model (is in proximity with experimentally determined q_e,). The values of χ², RMSE and MAE for pseudo first order were 0.18 ± 36, 3.12 ± 54 and 0.89 ± 35 whereas these values for pseudo second order model were determined only as 0.0076 ± 01, 0.36 + 05 and 0.105 ± 02, respectively. For further confirmation, the nonlinear fitting of pseudo first order and pseudo second order kinetic model were made and plotted together with experimental data which are also shown in Fig. 5. The value of As(III) uptake capacity obtained from pseudo second order model (4.39 ± 0.18) is more close to the experimental value (4.78 ± 0.09) then pseudo first order model (5.80 ± 0.91). The result inferred that biosorption of As(III) onto La(III)-SWR followed pseudo second order kinetics (see Scheme 4).

Close

Scheme 4

Inferred mechanism of As(III) elution using NaOH solution.

Export to PPT

3.2.3

3.2.3 Isotherm modellings

Fig. 6(a) shows the biosorption isotherm of As(III) using La(III)-SWR biosorbent at different temperatures. It is reveals from the results of this figure that the biosorption of As(III) increases with increasing equilibrium concentration in the lower concentration range and reaches plateau or nearely constant value at higher concentrations for all the temperature studied. The increase of biosorption capacity of As(III) with the increase of temperature suggesting endothermic nature of biosorption, which will be further confirmed by evaluating the thermodynamic parameters in section 3.3. The observation of plateau at higher concentration may be due to the saturation of limited vacant adsorption sites existed on the biosorbent surface. To evaluate the best fit isotherm model for the uptake of As(III) onto La(III)-SWR, two commonly used isotherms models namely Langmuir isotherm (Eq. (4)) and Freundlich isotherm models (Eq. (7)) were implemented. The isotherms parameters were determined from the slope and intercept of the respective straight-line plots as demonstrated in Supplementary Fig. 5 (Fig. S5) and Supplementary Fig. 6 (Fig. S6) and the evaluated isotherm parameters are portrayed in Table 3 together with various error functions. The experimental data showed an agreement with Langmuir isotherm model. The experimental data described by the Langmuir isotherm model exhibited a higher correlation coefficient (R² > 0.99) compared to the Freundlich isotherm model (R² not higher than 0.96). The value of maximum uptake capacity (q_m) of La(III)-SWR from the Langmuir isotherm model are determined to be 37.73 ± 0.12, 48.78 ± 0.09, 62.50 ± 0.11 mg/g respectively at temperature 298 K, 303 K, and 308 K which are in proximity with the experimental value determined directly through the plateau region in Fig. 6(a).

Close

Fig. 6

Biosorption isotherm of As(III) onto La(III)-SWR (a) isotherm plot at different temperatures (b) comparision of experimental data with data evaluated using non linear modeling of Langmuir and Freundlich isotherm, conditions: weight of biosorbent = 25 mg, volume of solution = 25 mL, pH = 12.08, shaking time = 24 hrs, shaking speed = 200 rpm and temperature = 298 K.

Export to PPT

Table 3 Isotherm parameters for the As(III) biosorption onto La(III)-SWR.

	Isotherm parameters	298 K	303 K	308 K
Langmuir isotherm	qexp(mg/g) qm(mg/g)	35.15 ± 0.12 37.73 ± 0.15	45.28 ± 0.09 48.78 ± 0.14	58.12 ± 0.11 62.50 ± 0.17
	b (L/mg) × 103	28.16 ± 0.03	39.88 ± 0.08	48.34 ± 0.06
	R2 χ2 RMSE MAE	0.99 0.19 ± 0.05 2.96 ± 0.002 0.93 ± 0.08	0.99 0.27 ± 0.08 4.08 ± 0.005 0.94 ± 0.05	0.99 0.33 ± 0.06 6.39 ± 0.004 0.91 ± 0.03
Freundlich isotherm	KF (mg/g) (L/mg)1/n	2.26 ± 0.87	2.76 ± 0.63	2.82 ± 0.58
	1/n	0.48 ± 0.26	0.68 ± 0.18	0.70 ± 0.39
	R2 χ2 RMSE MAE	0.96 0.49 ± 0.47 8.41 ± 0.09 2.65 ± 0.10	0.94 0.53 ± 0.31 10.34 ± 0.07 2.92 ± 0.15	0.89 0.51 ± 0.28 9.21 ± 0.05 2.87 ± 0.13

In order to further confirmation, the As(III) uptake capacity of La(III)-SWR was determined using non-linear fitting for both the models which also shows the good agreement of Langmuir isotherm with the experimental results compared to Freundlich isotherm model as displayed in Fig. 6(b). The values of error functions at three different temperatures (298 K, 303 K and 308 K): χ² (0.19 ± 0.05, 0.27 ± 0.08, 0.33 ± 0.06), RMSE (2.96 ± 0.002, 4.08 ± 0.005, 6.39 ± 0.004) and MAE (0.93 ± 0.08, 0.94 ± 0.05, 0.91 ± 0.03) for Langmuir isotherm model and χ² (0.49 ± 0.47, 0.53 ± 0.31, 0.51 ± 0.28), RMSE (8.41 ± 0.09, 10.34 ± 0.07, 9.21 ± 0.05) and MAE (2.65 ± 0.10, 2.92 ± 0.15, 2.87 ± 0.13) for Freundlich isotherm models were obtained. The results shows that the value of alll the errors paramters are smaller in case of Langmuir isotherm then Freundlich isotherm model for the all the temperatures, which suggest the better fitting of experimental data with Langmuir isotherm model. From these result, it is manifested that the uptake of As(III) onto La(III)-SWR took place according to the Langmuir theory by the formation of a monolayer of As(III) on the surface of biosorbent. In addition, the values of R_L analyzed experimentally from all the concentrations tested ranges for all the temperatures lies in between 0.08 and 0.89 as shown Supplementary Fig. 7(a) to 7(c) (Fig. S7). All these values lie between 0 and 1 (0 < R_L less than 1) which signify La(III)-SWR has a high biosorption affinity for As(III).

Close

Fig. 7

Van’t Hoff plot showing the relation between lnK_C versus reciprocal of absolute temperature.

Export to PPT

3.3 Evaluation of thermodynamic parameters

The investigation of thermodynamic parameters for the biosorption system is very important for the determination of spontanety, fesibility and nature of biosoption process. The experimental equilibrium parameters (b) at different temperature are determined from Langmuir isotherm model. Langmuir equilibrium parameters (b, L mg⁻¹) obtained is related to dimentionless thermodynamic equilibrium constant (K_C) by using equation (14) if biosorption experiment is performed in aqueous medium (Xu et al., 2019; Tran et al., 2017; Zhou and Zhou 2014; Xu et al 2021).

Where, M_w is the molar mass of arsenic (74.92 g mol⁻¹), and the factor 55.5 is the number of moles of pure water in one liter of solution. The Gibb’s free energy change (ΔG°) of the system can be determined by using equation (15) as

Where, R is universal gas constant (8.314 J mol⁻¹ K⁻¹), T is temperature in kelvin and K_C is dimentionless thermodynamic equilibrium constant. From thermodynamics, ΔG⁰ is also related to enthalpy change (ΔH⁰) and entropy change (ΔS⁰) at a given temperature according to the relations shown in equation (16) as

Now equating Eqs. (15) and (16) then we obtained a van’t Hoff equation as

The value of ΔG⁰ was directly calculated from Eq. (15) whereas those of ΔH⁰ and ΔS⁰ were determined from slope and intercept of van’t Hoff plot (lnK_C versus 1/T) as shown in Fig. 7. It gives a straight line plot with R² = 0.98. All the evaluated values of thermodynamic parameters are listed in Table 5. It is reveals from the results of this table that the value of ΔG° is negative for all the cases and negative was observed to be increased with increasing temperatures, suggesting that As(III) biosorption onto La(III)-SWR is feasible, thermodynamically favorable, and spontaneous (Pangeni et. al., 2014; Chiban et al 2016). The positive value of ΔH° confirms the endothermic nature of the biosorption process (Pangeni et al 2014) which is further supported from the increase of As(III) uptake capacity of La(III)-SWR with the increase of temperatures. Moreover, the value of positive ΔS° for the investigated biosrption system shows that the entropy of the system in the interfacial region increases which is probably due to the release of coordinated hydroxyl and water ligands via ligand exchang reaction with As(III) during biosorption process.

3.4 Effects of co-existing interfering anions

Naturally, the arsenic polluted water not only contains arsenic but also contains other co-existing ions such as phosphate, chlorides, sulphates and nitrates anions which may potentially interfere As(III) sorption process. To analyze the effect of these anions on the removal efficiency of As(III) by La(III)-SWR, the sorption performance were analyzed at varying concentration (10–200 mg/L) of these coexisting ions keeping As(III) concentration (10 mg/L) constant and the result is presented in Fig. 8. As can be observed from the result of this figure that the interference caused by these ions increases with the increases of their concentration which can be resobably interpreted due to the competition of these ions to the As(III) anions. Interfering effected of tested co-existing ion observed for the removal of As(III) is in the following order: PO₄^{- - -} > SO₄^{- -}> NO ₃^–> Cl ^-. Since chloride ion has insignificant interference, nitrate shows the amid effect while sulphate and phosphate imparted the most inhibitory effect for the sorption of As(III) ion. The phosphate and sulphate ions have a higher stability with La(III) ion thus La(III)-SWR possess more affinity with these ion resulting the enhance of inhibitory effect during As(III) sorption process. The tremendous decline of As(III) biosorption in the existence of phosphate ion is due to the chemical and structural similarity of oxoanionic species of As(III) with phosphate anion in an aqueous solution (Sahu et al., 2019).

Close

Fig. 8

Effect of co-existing ions in the biosorption of As(III) onto La(III)-SWR at varying concentration of co-existing ions conditions: weight of biosorbent = 25 mg, As(III) concentration = 10 mg/L, volume of solution = 25 mL, pH = 12.08, shaking time = 24 hrs, shaking speed = 200 rpm and temperature = 298 K.

Export to PPT

3.5 Creation of ligand exchange site for As(III) removal

The La(III)-SWR is an exemplary adsorbent for the uptake of As(III) anion. The loading of La(III) onto SWR facilitates the carboxyl group and oxygen atom of pyranose ring of watermelon pectic acid to build a stable five-membered chelated ring with loaded La(III) anion (scheme 2), which makes La(III)-SWR more stable and rigid (Paudyal et al., 2012; Aryal et al., 2021). Lanthanides have coordination numbers of 6, 8, 9, or even more 10 or 11, which allows various geometries. Loaded La(III) tend to be extensively hydrolyzed and coordinated several hydroxyl and water ligands on its coordination sphere which are easily exchanged with anionic or neutral species and therefore suitable for the ligand exchange reaction of anionic/neutral species of As(III) with hydroxyl/water ligand in La(III)-SWR as shown in scheme 3. These coordinated hydroxyl and water ligands are responsible for the exchange of anionic and neutral species of As(III) during biosorption process (Biswas et al., 2008).

3.6 XPS analysis

XPS analysis was carried out to account the mechanism of As(III) sorption onto La(III)- SWR. The wide scan spectrum recorded the peaks of La 3d, O 1 s, C 1 s, and As 3d clearly as displayed in Fig. 9(a). The intense peak of C 1 s observed at binding energy of 284.6 eV is considered as the characteristic peak of the aromatic/or aliphatic carbons of exist in cellulose, pectin, lignin and protein in watermelon rind. The La peaks detected on the La(III)-SWR confirmed the successful loading of La(III) onto SWR (Fig. 9b) in which the existence of As 3d on the sorbed La(III)-SWR revealed the effective sorption of As(III).In addition, the XPS spectrum of As 3d is associated with two peaks of As(III) and As(V) with binding energies of 44.01, 46.2 eV, respectively. This strongly highlights the partial oxidation of As(III) into As(V) by air under the examined experimental condition. The peaks of La 3d_5/2 and La 3d_3/2 were located at 837.85 and 854.74 eV (Fig. 9c) for La(III)-SWR. After As(III) sorption, the binding energies of the La 3d_5/2 and La _{3d 3/2} shifted to higher values. The high-resolution XPS scan of O1s (Fig. 9 d) was deconvoluted into three peaks at 531.0, 529.67 and 532.22 eV assigned to La - O, O – As and O – H bonds from the biosorbent. The presence of La - O bond in sorbent material suggests that the loaded La(III) forms the chemical bond with the functional groups of watermelon biopolymer rich in active oxygen and existence of O – As bond in sorption product indicates the successful sorption of complex ion containing oxygen and arsenic i.e oxoanions of arsenic. This result provides the direct evidence of loading La(III) onto carboxylic acid functional group of watermelon pectin and successful sorption of oxoanionic species of arsenic onto La(III)-SWR sorbent as inferred Scheme 1 and 2.

Close

Fig. 9

XPS spectra of La(III)-SWR after As(III) biosorption (a) Wide scan (b) La 3d (c) As 3d (d) O 1 s.

Export to PPT

3.7 Elution studies

The elution of the adsorbed As(III) from La(III)-SWR at a varying concentration of NaOH is depicted in Fig. 10. The result shows that the elution of As(III) increases from 41.11 to 71.07% with the increase in the concentration of NaOH from 0.001 to 0.5 mol/L, whereas the highest elution of 98.47 % was reached with 2 mol/L of NaOH. It can be inferred that the with the increase of hydroxide ion concentration resulted the increase of ligand exchange reaction of hydroxide ion with adsorbed As(III) anion as shown in scheme 4 during the elution process leading to an increase in elution percentage at higher concentration of NaOH. In addition, there is negligible reduction of As(III) uptake capacity (1.28%) of regenerated La(III)-SWR. Leakage of loaded La(III) is also insignificant (0.21%) even using 1 M NaOH, which strongly suggest the high stability of investigated La(III)-SWR in basic medium that can be used for several adsorption-elution-readsorption cycles.

Close

Fig. 10

Elution of As(III) from As(III) adsorbed La(III)- SWR as a function of NaOH concentration conditions: amount of As(III) sorbed onto La(III)-SWR = 0.25 mg/g, weight of As(III) loaded La(III)-SWR = 50 mg, volume of NaOH = 15 mL, shaking time = 24 hrs, shaking speed = 200 rpm and temperature = 298 K.

Export to PPT

3.8 Application of investigated La(III)-SWR for the treatment of arsenic polluted groundwater

The biosorption performance of La(III)-SWR was examined using arsenic-polluted groundwater (collected from Nawalparasi district, Lumbini Province, Nepal) to assess its feasibility for the dearsenification. From the analysis of the various parameters of the this water sample, the parameters such as pH (7.1), total hardness (387 mg/L), chloride (224 mg/L), fluoride (0.32 mg/L), total dissolved solid (457 mg/L), sulphate (14.7 mg/L), phosphate (1.3 mg/L) and iron (0.57 mg/L) are within the range of drinking water quality whereas arsenic concentration (86.8 μg/L) is 8.68 times higher than WHO recommended amount (10 μg/L). Fig. 11 shows the residual concentration of arsenic in the arsenic polluted groundwater as a function of La(III)-SWR dosage ranging from 0.5 to 5 g/L. The result shows that the residual concentrations of arsenic were found to be lowered considerably with the increase of asorbent dosage abd reached below the Nepalese standard (50 μg/L) by using La(III)-SWR dose of 0.5 g/L whereas 1.5 g/L is required to lowered the arsenic concentration below the WHO and USEPA tolerance level. From these results, it can be clearely suggested that the investigated La(III)-SWR could be used as an alternative material for the remediation of trace amount of arsenic polluted water.

Close

Fig. 11

Application of investigated La(III)-SWR for the removal of arsenic from actual arsenic polluted groundwater conditions: arsenic concentration = 86.8 μg/L, volume of water = 10 mL, native pH of ground water = 7.1, shaking time = 24 hrs, shaking speed = 200 rpm and temperature = 298 K.

Export to PPT