Kinetics of Aqueous Cu(II) Biosorption onto Thevetia peruviana Leaf Powder

Abstract

Copper is an essential micronutrient; however, as a result of its increasing demand, subsequent mining followed by its direct discharge into the environment has led to the contamination of our ecosystem. Thevetia peruviana (TP) is an ornamental herb of medicinal interest and is extensively used as an antipyretic and anticancer agent due to the presence of cardiac glycosides. In this work, we have explored the TP leaf powder as a biosorbent for Cu(II) removal from aqueous media and observed that it yields better results in comparison to other reported biosorbents for the removal of Cu(II). This work also emphasizes on the biosorption kinetics along with its plausible mechanism of interactions. The leaf powder characterized by FT-IR spectroscopy confirmed the diverse surface functionalities including hydroxyl, carbonyl, glycosides, etc. The morphology and elemental composition of the plant material have been investigated using SEM-EDAX analysis that confirms the heterogeneity and porosity of the biosorbent surface. The encouraging results revealed that the TP leaf powder could be used as a cost-effective biosorbent with an adsorption capacity of 187.51 mg g^–1 for Cu(II) in aqueous media at pH ∼ 5 and a temperature of 303 K. The complex functionality of the TP surface most likely played a significant role in attaining fast equilibrium within 60 min by following pseudo-second-order kinetics, having a rate constant of 2 × 10³ mg g^–1 min^–1 that has been confirmed with statistical tools such as regression coefficient, chi-squared, and sum of error square tests. The adsorption mechanism is controlled by diffusion of Cu(II) from the liquid phase to the solid phase of the TP biosorbent followed by the chemical interaction between the biosorbent and the adsorbate with slow intraparticle diffusion on the biosorbent surface. The adsorption of Cu(II) on TP has been observed to rise from 59.29 to 197.63 mg g^–1 with the rise in the pH of the medium from 2 to 7. The adsorption of Cu(II) has been found to increase from 176.80 to 191.33 mg g^–1 with increasing temperature from 293–308 K, confirming the endothermic nature of the adsorption process. The thermodynamic study revealed the adsorption process to be spontaneous with negative ΔG (−10.43 to −13.74 kJ mol^–1) and that it has an endothermic nature with positive ΔH (54.24 kJ mol^–1). The isotherm study for Cu(II) on TP followed the Langmuir adsorption isotherm model with the maximum monolayer adsorption capacity of 303.03 mg g^–1 rather than Freundlich and Temkin isotherm models, which confirmed the chemical interaction between the sorbent and sorbate. FT-IR and SEM-EDAX analyses have also been used to confirm the adsorption of Cu(II) onto the TP surface. The present study revealed 99.7% Cu(II) desorption using 0.8 N HCl as the desorbent accompanied by a 69.71% regeneration efficiency of the TP biosorbent. After desorption of Cu(II), the regenerated TP could be disposed of in soil. The encouraging results revealed that TP could be used as an alternative and low-cost biosorbent for the removal of heavy metals from aqueous solutions.

1. Introduction

Metals with density greater than or equal to 6.0 g cm^–3 are classified as heavy metals. These metal ions have attracted much attention due to their enhanced toxicity. Contamination by heavy metals occurs in the aqueous wastes of several industries, such as metal finishing, mining and mineral processing, coal mining, oil refining, and tannery among others.¹⁻³ The most familiar metals are Cd (8.65 g cm^–3), Cr (7.19 g cm^–3), Co (8.90 g cm^–3), Cu (8.95 g cm^–3), Pb (11.34 g cm^–3), Hg (13.53 g cm^–3), Ni (8.91 g cm^–3), and Zn (7.14 g cm^–3). Among them, Cu is the most widely used metal in different industries and household items and is also used as ornaments. It is primarily used in different industries, such as mining and smelting, brass manufacturing, plating, electroplating, and petroleum refining, and in Cu-based agrochemical syntheses. Cu(II) ions released from these industries with various concentrations into the environment cause pollution. Although Cu is an essential micronutrient for both animals and plants, soluble Cu compounds beyond the permissible limit become a threat to aquatic lives, animals, plants, and human health.⁴⁻⁶

According to the WHO, the maximum recommended level of Cu(II) in drinking water is 2.0 mgL^–1, while in accordance with BIS and USEPA standards, this limit is restricted to 1.5 and 1.3 mgL^–1, respectively (S1; Supporting Information). Acute symptoms of Cu poisoning involve vomiting, hematemesis, hypotension, melena, coma, jaundice, and gastrointestinal distress, and chronic exposure to Cu can damage the liver and kidneys, cause Wilson’s disease, and may lead to genetic disorders. As a result of the heavy metal toxicity all around the world, researchers pay significant attention to studying the extent of water pollution by heavy metals and to developing effective remediation methods and materials for recovery of water systems from pollution to develop a sustainable society in the future. Among the different heavy-metal removal methods, such as membrane separation, electrochemical precipitation, ion exchange, preconcentration, adsorption, and fertilization, adsorption techniques have become more popular in regard to their efficiency in the removal of heavy metal ions in aqueous media. Adsorption techniques have several advantages over the other methods, and some of their advantages include their simplicity in operation, low cost, and the lack of sludge (S2; Supporting Information).⁷⁻¹⁰ Extensive research studies have been carried out for the effective removal of heavy metals from aquatic ecosystems using different adsorbents, such as clays, modified clay nanocomposites, bio-based materials and their modified forms, etc.^5,7,11−14 Different parts of plants and their related wastes are naturally available as well as inexpensive in comparison to other chemically modified materials. Thus, most of the adsorption studies have been focused on untreated plant materials and their wastes.^9,10,15,16 In this paper, yellow oleander leaf powder (Thevetia peruviana, TP) is selected as the biosorbent for adsorption of Cu(II) from aqueous media. Its leaves are willowlike, linear-lanceolate, and glossy green in color. They are covered in a waxy coating to reduce water loss (typical of oleanders). Its stem is green and turns silver/gray as it ages. It is reported that the TP plant contains cardiac glycosides (S3; Supporting Information), which are toxic to most vertebrates.¹⁷⁻¹⁹ Although all parts of the TP plant are toxic to most vertebrates, it is cultivated as an ornamental plant and planted as a large flowering shrub in gardens and parks in temperate climates. The TP plant belongs to the family Apocynaceae and is widely distributed throughout the tropical to sub-temperate zones of the world.²⁰ This plant also may be found growing wild as it tolerates most soils as well as drought seasons.^17,20 It is often planted in numbers as a hedge.²¹ It is available in various states of India, such as Assam, West Bengal, Delhi, Bihar, Gujarat, Rajasthan, Madhya Pradesh, Tamil Nadu, and Uttar Pradesh, where semi-arid climate is predominant. It is represented by various common names, such as Kolke (Bengal), Kaneir or Kaner (in Hindi, India), Mexican oleander (Mexico), Yellow Oleander (America), Lucky Nut (West Indies), etc.¹⁷ The plant’s toxins have been tested for uses in biological pest control, besides possessing antipyretic, antibacterial, and anticancer properties.¹⁷⁻²⁰ It is expected that the presence of ligands such as carboxyl, carbonyl, and hydroxyl groups, etc. on the cell wall of biological materials immobilizes the metal ion and then uptake occurs.²²⁻²⁴ Thus, the presence of cardiac glycosides with polyhydroxyl groups in the TP leaf powder may play a significant role in the biosorption of heavy metal ions from aqueous media.

The present study reports the use of TP leaf powder as a biosorbent for Cu(II) removal from aqueous media with an emphasis on its biosorption kinetics, plausible mechanism of interactions, thermodynamics, adsorption isotherm studies, influence of environmental factors and desorption, and safe disposal of the used adsorbent.

2. Results and Discussion

2.1. Fourier Transform Infrared Spectroscopy (FT-IR) Analysis

FT-IR has been employed to determine the surface functionality of the powder TP material (Figure 1). The broad band around 3450 cm^–1 corresponds to the bonded −OH groups.^5,7 The small band at ∼3070 cm^–1 corresponds to =C–H and =CH₂ stretching vibrations. The bands at ∼2924 and ∼2856 cm^–1 could be attributed to asymmetric and symmetric C–H stretching, respectively, corresponding to the presence of methyl or methylene groups. The strong as well as broad band at ∼1647 cm^–1 corresponds to the asymmetric ester C–O stretching mode. The weak band at ∼1541 cm^–1 is due to the symmetric ester C–O stretching mode. The bands at ∼1383 and ∼1319 cm^–1 could be due to C–H scissoring modes. The weak band at ∼1267 cm^–1 could be assigned to the ether/alcohol C–O stretching mode. The broad band comprising several absorption bands at around 1063 cm^–1 has been assigned to C–O–C glycosidic ether vibrations present in cellobiose, such as cardiac glycosides and cellulose. The structures of the glycosides are presented in Figure S1 (Supporting Information).^12,25−28 The absorption peaks at ∼891 cm^–1 could be due to =C–H bond vibrations. The bands at ∼777, ∼667, and ∼517 cm^–1 correspond to the various metal–oxygen (M–O) bond vibrations.^7,29 The major bands correspond to the presence of −OH (hydroxyl), >C=O (carbonyl), and C–O–C (ether, glycosides, esters, etc.) functional groups within the material. The complex nature of the surface functionality of the material is most likely to play a significant role in binding water pollutants and helps it exhibit its biosorbent properties for water treatment. The IR data file is presented in Supporting Information 1.

Figure 1.

FT-IR spectrum of TP leaf powder.

2.2. Morphology and Elemental Composition Analysis

The heterogeneity associated with the surface morphology of TP is further confirmed with SEM analysis at 10 μm (Mag = 1.64 KX) using current at 5 kV voltage (Figure 2A). The biosorbent has cavities and pores, which has been confirmed by the micrographs obtained at scales 10 μm (Mag = 1.64 KX) and 1 μm (Mag = 7.00 KX) (Figure 2B), respectively. As a result of its porous structure, a higher adsorption behavior is most likely to be expected from the material.

Figure 2.

SEM images at (A) 10 μm (Mag = 1.64 KX) and (B) 1 μm (Mag = 7.00 KX) dimensions; (C) EDAX spectrum of TP.

The EDAX indexing of elements showed the presence of elements C, N, O, S, and P, which are considered to be the main constituents of organic residues, Figure 2C. The biosorbent TP leaf powder is prepared by drying the leaves followed by several washings to remove the chlorophyll pigments and water-soluble contents and then again drying; it is reflected in the EDAX data that the prepared TP powder showed 0.00% Na, 0.04% Mg, and 0.98% K by weight. The negligible Si has most likely arisen due to the presence of the glass substrate on which the TP sample was spread over. The amount of Ca is most likely due to the water-insoluble Ca compounds present in the plant residue and/or the glass substrate used for mounting the sample. It confirmed that the biosorbent TP initially possessed a negligible (<0.5% by weight) Cu micronutrient content. The elemental composition is presented in Table S5 (Supporting Information). The EDAX data file has been presented in Supporting Information 2.

2.3. Batch Adsorption Studies

2.3.1. Effect of Time

Metal uptake as a function of contact time has been observed to have occurred in four phases. The first phase has been extremely rapid. The TP leaf powder showed fast adsorption efficiency, 147.01 mg g^–1 [73.5% Cu(II) removal capacity], in the first 5 min of sorbent–sorbate contact time followed by a slow phase of metal pore diffusion and adsorption. This step is followed by desorption, and adsorption occurs over a period until the equilibrium is reached (Figure 3). The adsorption equilibrium has been attained after 50 min of contact time, indicating saturation of the adsorbent surface with metal ions. Before reaching the saturation point, adsorption followed desorption and has occurred due to the repulsion between like charges on the fully occupied surface of the adsorbent. During equilibrium time, the adsorption capacity of TP for the removal of Cu(II) has reached 187.51 mg g^–1, and the percentage of Cu(II) removal is 93.76%. Many researchers have studied the heavy metal removal capacity of different biosorbents at different times and places. The effectiveness of TP has been compared with other important adsorbents (Table 1), and it was found that TP exhibited a much higher biosorption capacity for Cu(II) in comparison to other biosorbents.

Figure 3.

(A) % of adsorption vs t (min) and (B) adsorption capacity: q_t (mg/g) vs t (min) for Cu(II) removal by TP with the maximum standard error, SE, ± 1 (error bar is marked red).

Table 1. List of Equilibrium Time of Cu(II) Adsorption Capacities Associated with Some Important Biosorbents.

adsorbent	Co (mg L–1)	te(exp) (min)	qe(exp) (mg g–1)	refs
Platanusorientalis leaf powder	50	180	44.94	(9)
coconut shell	5–300	180	19.89	(10)
neem leaves	5–300	180	17.49
hyacinth roots	5–300	150	21.79
rice straw	5–300	150	18.35
rice bran	5–300	180	20.98
rice husk	5–300	150	17.87
wheat shell	250	120	10.84	(30)
methylsulfonated Lagenaria vulgaris shell	50	120	10.36	(31)
KOH-treated pine cone powder	120	15	19.02	(32)
spent tea leaves	20	30	62.80	(33)
Cinnamomum camphora leaf powder	25	30	5.85	(34)
Tectonagrandis L.f	900	300	100.28	(35)
rubber leaves	10	60	3.38	(36)
cassava peel	200	30	39.08	(37)
sawdust	10	30	1.50	(38)
groundnut seed powder	10	30	4.82	(39)
sesame seed powder			4.24
coconut seed powder			4.32
mercaptoacetic acid-bamboo powder	50	20	9.70	(40)
carbon disulfide-bamboo powder			9.91
tectonagrandis leaf powder	20	180	42.99	(41)
banana trunk fibers	10	60	2.39	(42)
raw Ipomoea carnea	20	120	2.13	(43)
Ipomoea carnea-ZnCl2 (1:0.5)		240	3.13
Ipomoea carnea-ZnCl2 (1:1)			3.06
H3PO4-activated rubber wood sawdust	20	240	3.75	(44)
chemically modified orange peel	50	120	23.3	(45)
chemically treated tomato waste	50	120	9.87	(46)
sugar beet pulp	25.6	30	11.8	(47)
T. peruviana leaf powder	20	50	187.51	this study

2.3.2. Kinetics Study

The kinetics of adsorption were studied for zero-order, pseudo-first-order, and pseudo-second-order models, and the adsorption kinetics are presented as follows.

2.3.2.1. Zero-Order Kinetics

The zero order model (abbreviated as M₀), which highlighted the independence of the adsorption behavior on the concentration of the adsorbate and/or adsorbent, has the following differential form

1

After integration, eq 1 takes the form⁴⁸

2

where q_t (mg g^–1) represents the adsorption capacity at time t min; q_e (mg g^–1) represents the adsorption capacity at the time of equilibrium, and k₀ is the zero-order rate constant.The model M₀ could be E from its solution at equilibrium and is represented as

3

The experimental results were fitted to M₀ by plotting q_t versus time (Figure 4A) for the adsorption of Cu(II) onto TP. The q_e(calc) value (calculated equilibrium adsorption capacity) was determined to be 155.38 mg g^–1, which is not close enough to the q_e(exp) value (experimental equilibrium adsorption capacity), 187.51 mg g^–1. The calculated time of equilibrium, t_e(calc)(443.94 min), was calculated using eq 3 and was found to be quite higher than the experimental time of equilibrium, t_e(exp) (50 min). Again, the k₀ value, 0.35 mg g^–1 min^–1, obtained from the plot, Figure 4A, is quite low for explaining the fast equilibrium observed within 5 min of the adsorption process. The plot for M₀ had a regression coefficient (R² = 0.69) < 1, confirming the nonlinearity of the experimental data with M₀.

Figure 4.

(A) q_t (mg/g) vs t (min) plot for M₀, (B) ln (q_e– q_t) vs t (min) plot for M₁, (C) t/q_t (min g/mg) vs t (min) plot for M₂, (D) q_t (mg/g) vs (min^1/2) plot for M₃, (E) −ln (1 – F) vs t (min) for M₄, and (F) q_t (mg/g) vs ln t for M₅.

2.3.2.2. Pseudo-first-Order Kinetics

The Lagergren pseudo-first-order kinetics model (represented as M₁) has the following differential form

4

After integration, eq 4 takes the form⁵

5

where k₁ represents the first-order rate constant.

The experimental results were fitted to linear eq 5 for determination of the regression coefficient, slope, and intercept by plotting ln(q_e – q_t) versus time (Figure 4B). The regression coefficient was found to be 0.99, which revealed that the experimental data fitted to M₁ had good linearity. However, q_e(calc) obtained from the plot, 57.40 mg g^–1 (Table 2), is quite a low value compared to the q_e(exp) value of 187.51 mg g^–1 for Cu(II) adsorption onto TP. Thus, M₁ is not in good agreement with the experimental findings.

Table 2. Kinetics Data for Cu(II) Adsorption on TP^a.

						validity tests
model	rate constants	qe(exp)	qe(calc)	te(exp)	R2	χ2	SES
M0	(k0) 0.35	187.51	155.38	50	0.69	0.04	3.21
M1	(k1) 0.04		57.40		0.99	5.14	13.01
M2	(k2) 0.002		192.31		1.00	6.23 × 10–4	0.48

^aThe units of k₀, k₁, and k₂ are mg g^–1 min^–1, min^–1, and g mg^–1 min^–1, respectively; those of q_e(calc) and q_e(exp) are mg g^–1; that of t_e(exp) is min.

Again, the first-order rate constant, k₁(0.04 min^–1), calculated from the plot (Figure 4B) is not a measurable value for explaining the fast equilibrium found within 5 min for Cu(II) adsorption onto TP. Thus, the pseudo-first-order kinetic model has been observed to be unable to describe the order of Cu(II) adsorption onto TP.

2.3.2.3. Pseudo-second-Order Kinetics

Ho and McKay’s pseudo-second-order kinetics model (denoted as M₂) has the following differential form

6

After integration, eq 6 takes the following form⁵

7

where k₂ is the second order rate constant.

The experimental results have been fitted to M₂, eq 7, by plotting t/q_t versus time for Cu(II) adsorption onto TP, Figure 4C. The regression coefficient, slope, and intercept values obtained from the plot were used to determine q_e(calc) and k₂. q_e(calc) (192.31 mg g^–1), obtained from the plot (Figure 4C), is close enough to q_e(exp) (187.51 mg g^–1). Again, the obtained second-order rate constant k₂ is 0.002 g mg^–1 min^–1 (= 2 × 10³ mg g^–1 min^–1), presented in Table 2, and is used to calculate the initial adsorption rate, and h (= k₂q_e²)^9,49,50 was observed to be 70.42 mg g^–1 min^–1. The value of “h” has been in good agreement with fast Cu(II) adsorption at the initial stage of the adsorption process. Moreover, M₂ plots presented good linearity with the experimental findings as the regression coefficient has been found to be ∼1. Thus, M₂ has been considered to be the best fit model indicating the adsorbate–adsorbent strong interactions, thereby confirming the existence of chemisorption of Cu(II) ions on TP surfaces. The pseudo-second-order kinetics for Cu(II) adsorption onto biosorbents are supported by several findings, such as Cu(II) adsorption by Platanus orientalis leaf powder,⁹ cassava peels,³⁷ sulfur-modified bamboo powder,⁴⁰Tectona grandis leaf,⁴¹ banana trunk fiber,⁴² H₃PO₄-activated rubber wood sawdust,⁴⁴ sugar beet pulp,⁴⁷ herbaceous peat,⁴⁹ sour orange residue,⁵¹ etc.

2.3.2.4. Validity Test

The validities of the models M₀, M₁, and M₂ are tested with the chi-square (χ²) test (eq 8) and the sum of error square (SES) test (eq 9).⁵²⁻⁵⁴

8

9

where N represents the number of data points.

Both the tests compare the closeness between q_e(calc) and q_e(exp). The smaller the difference between q_e(calc) and q_e(exp), the smaller the χ² and SES values for a particular kinetics model, thereby imparting validity to the experimental results.^52,53

The χ² and SES values were calculated to be smaller for the M₂ model (χ² = 6.23 × 10^–4 and SES = 0.48) than to those of M₀ and M₁ (Table 2). Thus, the M₂ model is the best fit model for the Cu(II) adsorption process on TP rather than the M₀ and M₁.

2.3.2.5. Mechanistic Study

The mechanism of adsorption generally involves three steps: (i) mass transfer across the external boundary layer film of liquid surrounding the outside of the particle; (ii) adsorption at a site on the surface (internal or external)—the energy is most likely to be dependent on the binding process (physical or chemical), and this step is often assumed to be extremely rapid; and (iii) diffusion of the adsorbate molecules to an adsorption site either by a pore diffusion process through the liquid-filled pores or by a solid surface diffusion mechanism. One or any combination of these steps could be the rate-controlling mechanism for the adsorption process.⁵⁵ Here, the sorption mechanism studies have been carried out using kinetics-based models such as Weber’s intraparticle diffusion model, the liquid film diffusion model, and the Elovich model and have been presented below.

2.3.2.5.1. Weber’s Intraparticle diffusion model

Weber’s intraparticle diffusion model (abbreviated as M₃) has the form^5,50,56,57

10

where k_id represents the intraparticle diffusion rate constant.

M₃ gives an insight into the mechanisms and rate-controlling steps affecting Cu(II) adsorption onto TP surfaces. The intercept, I in eq 10, corresponds to the boundary layer effect. The larger the value of intercept (I > 0), the greater the contribution of surface sorption in the rate-controlling step would be. When the linear q_t versus t^1/2 plot passes through the origin (i.e., I = 0), the intraparticle diffusion is considered to be the sole rate-limiting step.⁵ The experimental results were fitted to M₃ by plotting q_t versus t^1/2 for Cu(II) adsorption on TP (Figure 4D). Linear fitting to the plot gave a straight line with I (= 141.48 mg g^–1; Table 3) > 0 (the red line in Figure 4D). Thus, the sorption process of Cu(II) on TP could not be explained only with the help of the intraparticle diffusion mechanism, and larger I values indicate greater contribution of the surface sorption in the rate-controlling step. The slope of the single linear plot (the red line, Figure 4D) corresponds to the intraparticle diffusion rate constant, k_id, that is related to the porous nature of the sorbent surface.¹² The higher value of k_id reflects the higher porosity of the sorbent material. The k_id obtained is 4.91 mg g^–1 min^–1/2 for Cu(II) adsorption on TP, confirming the porous nature of the material surface in conformity with the SEM results.

Table 3. Parameters Associated with M₃, M₄, and M₅ Models.

model	parameters
intraparticle diffusion (M3)	kid (mg g–1 min–1/2)	intercept	R2
	4.91	141.48	0.83
liquid-film diffusion (M4)	kfd (min–1)	intercept	R2
	0.04	1.18	0.99
Elovich (M5)	α (mg g–1 min–1)	β (g mg–1 min)	R2
	1.57 × 1013	0.20	0.89

A single linear Weber–Moris plot (q_t versus t^1/2) depicted the adsorption process to have been controlled by intraparticle diffusion only. However, the experimental data are most likely to exhibit multilinear plots, which describe the involvement of more than two steps in the adsorption process.^58,59 The experimental data actually yield multilinear plots, as in Figure 4D (green line), and hence the adsorption process involving more than two steps and simple intraparticle diffusion could no longer be described as the only rate-controlling step. There are almost four steps involved in the adsorption process as observed from the multilinear plot, Figure 4D.⁵⁹ The first step is the fastest diffusion of the adsorbate from the liquid phase to the solid phase, which is completed within 5 min, followed by the second step, controlled by chemical interactions among the adsorbate–adsorbent, followed by the third step of intraparticle and pore diffusion. Finally, in the fourth step, adsorption and desorption occur until the surface is completely saturated.

2.3.2.5.2. Liquid film diffusion model

The liquid film diffusion model (abbreviated hereafter as M₄) has the form^7,60

11

12

where k_fd represents the intraparticle diffusion rate constant.

The M₄ model has been used to study the role of metal ion transport from the liquid phase up to the solid phase boundary in the adsorption process. A linear plot of {−ln (1 – F)} versus time passing through the origin would suggest that the kinetics of the adsorption process are controlled by ion diffusion through the liquid surrounding the solid adsorbent.^6,7 The experimental results are fitted to M₄ by plotting {−ln (1 – F)} versus time, Figure 4E. The film diffusion rate constant, k_fd, has been determined, with the slope of the plot, to be 0.04 min^–1 (Table 3). For the Cu(II) adsorption process on TP, the k_fd value is very low, indicating the slow rate of metal ion transport from the liquid phase up to the solid phase boundary during the adsorption process. The intercept of the M₄ plot, Figure 4E, is 1.18 > 0 but comparatively less than that of the M₃ model, indicating that M₄ might have a greater role in the kinetics than M₃.⁵⁸

2.3.2.5.3. Elovich model

The Elovich model (abbreviated hereafter as M₅) has the form^6,7

13

where α and β are the Elovich constants. M₅ has been first developed to describe the kinetics of chemisorption of a gas onto solid adsorbents, but it has been found to be useful in other forms of adsorption as well as to understand the mechanism of adsorption in the liquid phase. The experimental data are fitted into M₅, Figure 4F. The Elovich parameters α (mg g^–1 min^–1) and β (g mg^–1 min) are calculated from the intercept and slope of the plot, respectively. The Elovich parameter α represents the rate of chemisorption at zero coverage or the initial adsorption rate, and β is a desorption constant related to the extent of surface coverage and the activation energy of chemisorption.^7,61 The higher the value of α, the greater is the extent of chemisorption. Cu(II) adsorption on TP showed large values of α, 1.57 × 10¹³ mg g^–1 min^–1 (Table 3), indicating chemisorption as one of the rate-controlling steps, confirming the pseudo-second-order type of adsorption. Again, the lower value of β indicates the lower value of activation energy for chemisorption, hence indicating fast adsorption.⁶¹ In this study, it is found that β for Cu(II) adsorption on TP is quite small, 0.20 g mg^–1 min, indicating a lower activation energy of chemisorption.The comparison of R² values (Table 3) confirms that the adsorption of Cu(II) on the biosorbent TP is mainly controlled by liquid to solid diffusion (M₄ model) followed by the chemical interaction (M₅ model) and then followed by a slow intraparticle diffusion (M₃ model) on the biosorbent surface.

2.3.3. Effects of the pH, Temperature, Amount of Adsorbent, and Cu(II) Concentration on Adsorption

The effect of environmental factors such as pH (2–7), temperature (293–308 K), and the concentration of Cu(II) (10–50 mg L^–1) on the adsorption by TP has been studied keeping the other factors constant.

The effect of pH (2–7) on adsorption has been studied with 20 mg L^–1 of Cu(II) solution using 1 g L^–1 TP biosorbent at 303 K. The adsorption capacity of TP is found to increase (59.29–197.63 mg g^–1) with an increase in pH (Figure 5A). At a lower pH, the functional groups, such as hydroxyl, carboxyl, carbonyl, etc., present over the surface of TP get protonated with H⁺ ions (as shown in Figure 5B), which leads to the lower adsorption affinity of TP toward Cu(II), hence showing low adsorption capacity. With the rise in pH, the surface functionality remains free to adsorb Cu(II) and thus leads to higher adsorption. A similar trend in the increase of Cu(II) adsorption (from 42.00 to 49.94 mg g^–1) with increasing pH from 2 to 6 has also been reported for the biosorbent Platanus orientalis leaf powder.⁹ The effect of pH beyond 7 has not been investigated because beyond pH 7, precipitation of Cu(II) as Cu(OH)₂ predominates over adsorption.^7,9,10

Figure 5.

Effect of (A) pH on q_t (mg/g), maximum SE ± 2, (B) schematic representation for the protonated functionality of TP surface at lower pH, (C) temperature (K) on q_t (mg/g), maximum SE ± 1, (D) C₀ (mg/L) on q_t (mg/g), maximum SE ± 3, and (E) TP amount (g/L) on % of adsorption, maximum SE ± 1.

The effect of temperature on adsorption by TP has been carried out at 293, 298, 303, and 308 K by keeping the other parameters constant, such as pH (5), the concentration of Cu(II) (20 mg L^–1), and the adsorbent amount (1 gL^–1). With the rise in temperature (293–308 K), the adsorption capacity of TP is found to increase slightly from 176.80 to 191.33 mg g^–1 (Figure 5C), which implies that the adsorption process of Cu(II) on TP is endothermic in nature.

The effect of the concentration of Cu(II) on adsorption by TP has been determined using 10, 20, 30, 40, and 50 mg L^–1 Cu(II) solutions keeping the other factors constant, such as 303 K temperature, pH 5, and 1 gL^–1 TP amount. It is observed that with the rise in the concentration of Cu(II) in aqueous medium, the adsorption capacity of TP is increased from 96.73 to 309.59 mg g^–1 (Figure 5D). However, the adsorption process is found to slow down comparatively after 40 mg L^–1 Cu(II) concentration (291.59–309.59 mg g^–1), which may be due to accomplishment of surface saturation of the TP amount used.

The effect of TP amount on Cu(II) adsorption has been determined using 0.5, 1.0, 1.5, and 2.0 gL^–1 TP for 20 mg L^–1 Cu(II) solution at 303 K temperature and pH 5. It is observed that 1.5 g L^–1 TP amount is sufficient for the complete removal of 20 mg L^–1 Cu(II) from the aqueous solution (Figure 5E). An amount of 1 g L^–1 TP adsorbed ∼94% of the Cu(II) from 20 mg L^–1 Cu(II) solution at 303 K temperature and pH 5, and using 1.5 g L^–1 TP showed ∼100% removal of Cu(II) from the aqueous medium under the same conditions.

2.3.4. Thermodynamics of Adsorption

The thermodynamic parameter Gibbs free energy change (ΔG) is an important parameter to depict the spontaneity of a process. A process with ΔG<0 is a spontaneous process; otherwise, the process is nonspontaneous if ΔG>0. The other thermodynamic parameters of adsorption, change in enthalpy (ΔH) and change in entropy (ΔS), are also important to depict whether energy is released or absorbed during the interaction of the adsorbate to the adsorbent and to depict the affinity of the adsorbate toward the adsorbent. ΔG (kJ mol^–1) is related to ΔH (kJ mol^–1) and ΔS (kJ K^–1 mol^–1) as shown in eq 14 and to the distribution coefficient (K_d = q_e/C_e) of the solute as shown in eq 15.

14

15

Therefore, using eq 14 in eq 15 implies that

16

Equation 16 is known as the van’t Hoff equation, where R is the universal gas constant (8.314 J K^–1 mol^–1) and T (K) is the temperature of adsorption, which is in the form of a straight line equation. Thus, the plot of ln K_d versus 1/T produces a straight line with a slope and intercept equal to −ΔH/R and ΔS/R, respectively (Figure 6). The slope and intercept values are used to determine the ΔH and ΔS values, respectively, and hence the ΔG value for the adsorption process is obtained using eq 14 (Table 4). The negative value of ΔG confirms the spontaneous nature of Cu(II) adsorption on TP in aqueous medium. The positive value of ΔH confirms the endothermic nature of the adsorption process, and the positive value of ΔS confirms the affinity of the TP for Cu(II), which results in the randomness in the solid–solution interface.^6,7,9,10

Figure 6.

ln K_d vs 1/T (K⁻¹) plot for Cu(II) adsorption on TP.

Table 4. Thermodynamic Parameters.

			ΔG (kJ mol–1)
ΔH (kJ mol–1)	ΔS (kJ K–1 mol–1)	R2	293 K	298 K	303 K	308 K
54.24	0.22	0.98	–10.43	–11.53	–12.63	–13.74

2.3.5. Isotherm Study on Adsorption

In adsorption processes, the interactions between the adsorbate and adsorbent follow one particular type of interaction that is predominant over the other types for a particular situation. The adsorption isotherm study helps to depict the type of interaction in a particular case of the adsorption process. Among the various mathematical forms of adsorption isotherm models, herein, the experimental data have been fitted to three isotherm models, namely, Langmuir, Freundlich, and Temkinin, in order to verify which model presented the best adjustment.

2.3.5.1. Langmuir Adsorption Isotherm Model

The Langmuir adsorption isotherm model (abbreviated hereafter as M₆, eq 17) depicts monolayer formation of the adsorbate molecules/ions over the surface of the adsorbent molecule through mainly chemical interactions between the adsorbate and adsorbent.⁷

17

where q_m is the monolayer adsorption capacity of the adsorbent (mg g^–1) and K_L is the Langmuir adsorption constant (L mg^–1) related to the affinity of the adsorption. The plot of 1/q_e versus 1/C_e is shown in Figure 7A, and the slope and intercept obtained from the plot are used to determine K_L (1.43 L mg^–1) and q_m(303.03 mg g^–1), respectively, (listed in Table 5). A higher K_L value (2.68 L mg^–1) with a q_m value of 169.49 mg g^–1 has also been reported for Cu(II) adsorption onto the Platanusorientalis leaf powder.⁹

Figure 7.

(A) 1/q_e (g/mg) vs 1/C_e (L/mg) plot of M₆; (B) ln q_e vs ln C_e plot of M₇; and (C) q_e (mg/g) vs ln C_e plot of M₈ for Cu(II) adsorption on TP.

Table 5. Isotherm Models Representing Different Parameters.

isotherm models	parameters
M6	KL (L mg–1)	qm (mg g–1)	RL	R2
	1.43	303.03	0.01–0.06	1.0
M7	KF (mg(1–1/n) L1/n g–1)		N	R2
	150.13		3.60	0.94
M8	KT (L g–1)		b (J mol–1)	R2
	23.47		48.45	0.99

The Langmuir adsorption constant, K_L, is related to the separation factor, R_L, as shown in eq 18

18

R_L is a measure of feasibility of the adsorption process. If R_L = 0, then the adsorption process is categorized to be irreversible; 0 < R_L < 1 implies a favorable process, R_L = 1 implies a linear process, and R_L > 1 implies an unfavorable adsorption process.^7,9,10 In this study, the R_L value ranges from 0.01 to 0.06 (0 < R_L < 1), and it corresponds to favorable adsorption of Cu(II) on TP. The regression coefficient (R²) of the plot for M₆ is ∼1.0, which confirms that the adsorption process follows the M₆ model, and the adsorption process mainly takes place via chemical interactions between the adsorbate and adsorbent.

2.3.5.2. Freundlich Adsorption Isotherm Model

The Freundlich adsorption isotherm model (abbreviated herein as M₇, eq 19) depicts heterogeneous adsorption as adsorbents possess a heterogeneous surface with various types of adsorption sites bearing different energies.

19

where K_F is the Freundlich coefficient, which reflects the adsorption capacity, and n is the Freundlich coefficient, which reflects the degree of heterogeneity.^6,7,9,10 For the favorable adsorption process, the value of n should be 1 < n < 10. The plot of ln q_e versus ln C_e (Figure 7B) yields the slope (1/n) and intercept (ln K_F). From the slope and intercept values, the n (3.60) and K_F (150.13 mg^(1–1/n) L^1/n g^–1) values are determined, respectively (Table 5). 1 < n (3.60) < 10 confirms the favorable nature of the adsorption process. The regression coefficient (R²) of the plot for M₇ is 0.94, which is close to 1.0 but compared to M₆, R² is quite less than 1.0. This confirms that the adsorption process follows the M₆ model over M₇.

2.3.5.3. Temkin Adsorption Isotherm Model

The Temkin isotherm model (abbreviated hereafter as M₈, eq 20) assumes that the adsorption energy decreases linearly with the surface coverage due to adsorbent–adsorbate interactions.⁵²

20

where T (K) is the absolute temperature, R is the universal gas constant [8.314 J K^–1 mol^–1], b (J mol^–1) is the Temkin constant related to the enthalpy of adsorption, and K_T (L g^–1) is the equilibrium binding constant. The linear plot of q_e versus ln C_e (Figure 7C) results in the slope, RT/b, and the intercept, (RT/b) ln K_T. From the slope and intercept, Temkin constants b (48.45 Jmol^–1) and K_T (23.47 L g^–1) are calculated, respectively (Table 5). The positive value of b confirms the endothermic nature of the adsorption of Cu(II) on TP. The regression coefficient (R²) of the plot for M₈ is 0.99, which is very much close to 1.0. This confirms that the adsorption process follows the M₈ model, which conveys the chemical interactions between the adsorbate and adsorbent.

3. Characterizations of the Biosorbent TP after Adsorption

3.1. Comparative FT-IR Analyses of the Biosorbent TP before and after Adsorption

The FT-IR spectra of the TP leaf powder after adsorption of Cu(II) (Figure 8; blue spectra; Cu-TP) showed considerable changes in intensities of the different characteristic bands as well as shifts could be observed in the band positions with respect to that of the TP leaf powder before adsorption (Figure 8; red spectra; TP). The observed differences associated with the bands are listed in the Supporting Information (Table S6). It is also seen that after adsorption, the leaf powder showed prominently high transmission intensities in almost all the vibration bands, compared to the leaf powder before adsorption, thereby signifying its good adsorbent characteristics. Specifically, the increase in intensity and broadening of the −OH stretching band after adsorption confirms the formation of complexes such as (OH₂)₅Cu(II)—O-(TP residue) and hence confirms the presence of more H₂O molecules in Cu(II)-adsorbed TP than that in TP before adsorption. Again, a new peak in Cu(II)-adsorbed TP appeared at around 822 cm^–1, which is most likely to be referred to the newly formed Cu(II)—O-(TP residue) bond vibration. Many findings showed similar shifts in FT-IR peaks/bands of the biosorbents after the adsorption process.^{34,37,39,46,49,51} The relevant IR data of the biosorbent TP after adsorption are presented in Supporting Information 3.

Figure 8.

FT-IR spectra of TP leaf powder: before (TP) and after (Cu-TP) adsorption.

3.2. SEM Analyses of the Biosorbent TP after Adsorption

After adsorption, the biosorbent TP leaf powder was collected after centrifugation and dried for SEM analysis. The SEM image of the TP after adsorption confirmed the coverage of the surface pores and cavities of TP with Cu(II) ions (Figure 9A). The EDAX spectrum is presented in Figure 9B. The relevant EDAX data of the biosorbent TP after Cu(II) adsorption are presented in Supporting Information 4. The EDAX analysis (Table S7; Supporting Information) of the biosorbent after adsorption showed a significant increase in the Cu content (12.15% by weight) in comparison to that (0.40% by weight) observed before adsorption, thereby confirming the remarkable adsorption capacity of the TP leaf powder for Cu(II) in aqueous medium.

Figure 9.

(A) SEM images of TP after adsorption at 3 μm (Mag = 3.00 KX); (B) EDAX spectrum of TP after adsorption.

4. Plausible Mechanistic Pathway of Biosorption

The adsorbate–adsorbent interactions could be explained on the basis of the chelation properties of the organic functional groups, primarily the hydroxyl groups present at the surface of the sorbent toward the Cu(II) ion.^13,34 Cu(II) ions actually exist as hexaaquocopper(II) complex ions. The hydroxyl groups, present at the sorbent surface, are directly attached to organic molecules, i.e., the carbon chains, and the positive inductive nature of the carbon chains is most likely to enhance the chelation properties of the attached hydroxyl groups, compared to that of simple H₂O molecules. Thus, Cu(II) ions more strongly undergo chelation to the sorbent surface, compared to H₂O molecules. The generalized interactions dominating the adsorption of Cu(II) onto the TP surface are schematically presented in Figure 10.

Figure 10.

Schematic representation of the surface complexation model for Cu(II) adsorption onto the TP biosorbent.

5. Desorption, Regeneration Efficiency, and Disposal of the Used Biosorbent TP

Nowadays, many researchers opt to use bio-adsorbents for the removal of heavy metals instead of chemical adsorbents/chemically modified adsorbents as biosorbents provide promising results with a minimum disposal problem since the bio-adsorbents could be degraded easily by microorganisms.

The desorption of heavy metals from the surface of biosorbents is the ratio between the solid and liquid (S/L): a heavy metal amount ratio where the solid phase is the adsorbent on which heavy metals are adsorbed, and the liquid phase is the desorbing eluent (desorbent).^62,63 A desorption study helps in understanding the reusability of the biosorbents, also reducing the cost of the biosorption process due to the reuse of the sorbents. Solvents such as acids, alkalis, and alcohols are used as desorbents for removing the sorbates from the biosorbents. Desorbents such as NaOH solution, ethanol, and methanol solutions are generally used for desorption to regenerate the biosorbent used in adsorption of dyes.^62,64 Generally, acid solutions are used as desorbents for heavy metal removal from heavy metal-loaded biosorbents.^10,62 The efficiency of heavy metal desorption has been determined using the following equation⁶²

21

where C_d represents the concentration of Cu(II) in the desorbent (mg L^–1), V_d is the volume of the desorbents (L), q_e is the adsorbent’s adsorption capacity (mg g^–1) before the desorption study, and m_a is the mass of the Cu(II)-loaded adsorbent (g) taken for the desorption study.^62,65The main purpose of desorption activity is associated with the regeneration of the sorbent so that it could be reused. The regeneration efficiency of biosorbent is primarily dependent on desorbent pH, since the removal of cations is significant in acidic condition. On the other hand basic desorbent is generally unfavorable for anion recovery. Thus, the selection of a proper desorbent is necessary for obtaining maximum regenaration efficiency.⁶² The regeneration efficacy (RE) of an adsorbent and biosorbent is calculated by the following formula

22

where A_r and A₀ are the adsorption capacity (mg g^–1) of the adsorbent after regeneration and the original adsorption capacity of the adsorbent (mg g^–1), respectively.^62,66

Herein, the desorption experiments have been carried in order to estimate the percentage of Cu(II) desorption by the Cu(II)-loaded biosorbent TP and the RE percentage of the biosorbent TP. These experiments have been carried out using different concentrations of HCl solution (0.1–0.8 N) at 303 K. Table 6 shows that the RE percentage of the biosorbent and desorption percentage of Cu(II) increased with increasing strength of HCl solution. It has been observed that with a 0.8 N HCl desorbent, the Cu(II)-loaded biosorbent TP showed a desorption of 99.70%. Thus, before disposal of the used biosorbent, it was treated with 0.8 N HCl for 1 h. The RE of the used TP was observed to decrease with 0.7 and 0.8 N HCl desorbent, which could be attributed to the denaturation of the surface functionality of TP after acid treatment.

Table 6. Desorption Percentage of Cu(II) from Loaded TP and RE of TP.

HCl strength (N)	desorption percentage of Cu(II)	RE percentage of TP
0.1	62.27	57.42
0.2	64.11	61.34
0.3	67.49	64.35
0.4	72.58	68.21
0.5	78.42	70.54
0.6	81.73	74.93
0.7	90.08	72.46
0.8	99.70	69.71

Biosorbents are reported to be degraded by microorganisms.⁶⁷ However, direct disposal of the Cu(II)-loaded biosorbent in nature may create serious concerns to humans and the environment.^67,68 Hence, the used biosorbent TP could be released into the environment or dumped into the soil only after recovery of Cu(II) through the desorption process.

6. Conclusions

The present study revealed the possibility of using a low-cost and easily available TP biosorbent for treatment of water systems contaminated with heavy metals, which exhibited much better results in comparison to similar biosorbents reported earlier. Characterizations with SEM and FT-IR reflected the existence of heterogeneity, porosity and −OH (hydroxyl), >C=O (carbonyl), and C–O–C (ether, glycosides, esters, etc.) functional groups within the biosorbent TP. It is most likely that the complex nature of surface functionalities, heterogeneity, and porosity of the biosorbent TP surface play significant roles in the binding of water pollutants, thereby exhibiting its adsorption properties. The results presented herein showed that TP could be used as a highly efficient biosorbent for the removal of Cu(II) from aqueous media. The kinetics of biosorption followed the pseudo-second-order reaction, and its mechanism has been primarily controlled by diffusion of Cu(II) from the liquid phase to the solid phase of the biosorbent TP. The complex interaction of Cu(II) on the biosorbent TP could be depicted by the surface complexation model, which involved liquid to solid surface diffusion, chemical interactions, and intraparticle diffusion on the biosorbent surface. Studies on the effect of pH revealed that the adsorption capacity of TP (59.29–197.63 mg g^–1) increases with increasing solution pH (2–7). The thermodynamics of biosorption revealed the adsorption of Cu(II) on TP to be a spontaneous process with ΔG (−10.43 to −13.74 kJ mol^–1) and to be endothermic in nature (ΔH = 54.24 kJ mol^–1). The present study further confirmed that the adsorption process follows the Langmuir adsorption isotherm model rather than the Freundlich and Temkin models, thereby representing the chemically controlled nature of the adsorption process of Cu(II) onto the TP biosorbent. Characterizations using FT-IR and SEM-EDAX confirmed the adsorption process. The desorption study presented 99.70% desorption of Cu(II) with a 0.8 N HCl desorbent accompanied by a 69.71% regeneration efficiency. The encouraging results of biosorption presented by the TP leaf powder with respect to adsorption of Cu(II) reveal its future scope of utilization in the removal of other heavy metals on a large scale from aqueous media.

7. Experimental Section

7.1. Preparation of the TP Biosorbent

Thevetia peruviana (TP) leaves (Figure 11A) have been collected and washed with deionized water. The leaves were dried under sunlight followed by oven drying at 80 °C for 2 h (Figure 11B). The dried leaves were further crushed to powder and washed several times with deionized water until the supernatant liquid is free from green-colored pigments. The powder was dried in an oven at 80 °C overnight and stored in an airtight glass container after sieving through a 100 mesh (149 μm) sieve (Figure 11C). Hence, the particle size of the leaf powder is −149 μm, and it possesses a bulk density of 0.28 gcm^–3.

Figure 11.

(A) Mature TP leaves, (B) dried TP leaves, and (C) TP biosorbent powder.

7.2. Preparation of Adsorbate Solution

A 1000 ppm Cu(II) stock solution was prepared by dissolving 3.929 g of CuSO₄·5H₂O in a 1000 mL volumetric flask, and the volume was made up to the mark with deionized water. From the stock solution, Cu(II) solutions of various concentrations were prepared and used directly for batch adsorption studies.

7.3. Characterization Techniques

The surface functionality of the prepared TP adsorbent was investigated by Fourier transform infrared (FT-IR) spectroscopy (Shimadzu IR Affinity-1) using a KBr pressed disc technique (1.0 mg of sample per 200 mg of KBr). The morphology along with the elemental composition of the samples was evaluated using a Zeiss Gemini SEM (5 kV)/EDX (15 kV) instrument. Gold coatings of the specimen were prepared at ∼6 nm min^–1 (at a pressure of 7 × 10^–2 Pa and a current of 20 mA) with a Quorum (Q 150R ES). The residual metal ion concentration in the supernatant liquids was analyzed using a Perkin Elmer AAnalyst 200 atomic absorption spectrophotometer (AAS).

7.4. Batch Adsorption Studies

Adsorption experiments were carried out in 100 mL Erlenmeyer flasks, which were stirred using an Almicrorotary shaker at a stirring speed of ∼200 rpm. In order to maintain the accuracy of the results, each experiment was performed twice. The adsorption efficiency was measured by exposing 1.0 g L^–1 adsorbent TP to 10 mL of 20 mgL^–1 Cu(II) model solution in the batch process for contact time periods from 5 to 120 min at 293–303 K and pH 2–7. The effects of Cu(II) concentration on adsorption were investigated using 10–50 mgL^–1 Cu(II) solutions. The pH (2–7) of the Cu(II) solution was maintained using 0.1 N HCl and 0.1 N NaOH. The effects of TP amount were investigated using 0.5–2.0 g L^–1 TP leaf powder. The solutions obtained after the adsorption process have been centrifuged using a Lyzer instrument, and the supernatant liquids were kept in airtight bottles for the analyses. Moreover, after the adsorption process, the used adsorbent was collected, dried, and kept in an airtight bottle for further analyses. The concentrations of residual metal ions in the supernatant liquids were analyzed using an atomic absorption spectrophotometer (Perkin Elmer AAnalyst 200). The adsorption capacity (q_t mg g^–1) and percentage of adsorption of the adsorbent TP at time t are calculated according to the following equations^5,7

23

24

where C_o represents the initial Cu(II) concentration (mgL^–1), C_t represents the Cu(II) concentration at time t in the filtrate solution (mg L^–1) after adsorption, V (mL) represents the volume of the Cu(II) solution, and m has been associated with the weight of the leaf powder added into the solution (g L^–1).

The adsorption studies were performed using the zero-order model, Lagergren pseudo-first-order model, Ho and McKay pseudo-second-order model, intraparticle diffusion model, liquid–film diffusion model, Elovich model, and Langmuir, Freundlich, and Temkin isotherm models.^5−7,25,26 Adsorption on porous adsorbents has been described as a three-stage process consisting of external diffusion through the liquid–film over the material, internal diffusion into the pores, and the actual adsorption (physical or chemical). The kinetics of the entire process were determined on the basis of the slowest step (rate-controlling step). The validity of the kinetics models was tested by comparison of the regression coefficients, chi-squared (χ²) tests, and sum of error square (SES) tests.⁵²⁻⁵⁴

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.9b04032.

• Environmental fate of copper; current technologies available and their advantages and disadvantages for removal of heavy metals from aqueous media; cardiac glycosides present in Thevetia peruviana; table for EDAX elemental composition of TP; table for FT-IR bands of TP before and after adsorption; table for EDAX elemental composition of TP after adsorption; and supporting references (PDF)

• Supporting Information 1: FT-IR data of TP (PDF)

• Supporting Information 2: SEM-EDAX data of TP (PDF)

• Supporting Information 3: FT-IR data of TP after adsorption (PDF)

• Supporting Information 4: SEM-EDAX data of TP after adsorption (PDF)

Author Present Address

^# K.B.G.: Department of Chemistry, Assam Don Bosco University, Sonapur 782,402, Assam, India.

Author Contributions

All authors contributed equally.

The authors declare no competing financial interest.