A green technology for the synthesis of cellulose succinate for efficient adsorption of Cd(II) and Pb(II) ions

Abstract

Cellulose succinate, which was used for efficient adsorption of heavy metals, was directly prepared by mechanical activation (MA)-assisted solid-phase synthesis in a stirring ball mill with bagasse pulp and succinic acid as materials without the use of organic co-reagents and solvents. FTIR, XRD, SEM, and specific surface area analysis were used to characterize the structural characteristics of cellulose succinate. Furthermore, the effects of different degrees of esterification of modified cellulose on the adsorption of Cd²⁺ and Pb²⁺ were investigated. A surface charge characteristic was used to prove the effect of pH on the adsorption ability of cellulose succinate. It was found that the adsorption kinetics of Cd²⁺ and Pb²⁺ onto cellulose succinate fitted well with the pseudo-second-order model. The adsorption of Cd²⁺ and Pb²⁺ onto cellulose succinate was well described by the monolayer adsorption of the Langmuir isotherm model rather than the multilayer adsorption of the Freundlich isotherm model. The E values for the adsorption of Cd²⁺ and Pb²⁺ by cellulose succinate calculated by the Dubinin–Radushkevich equation were all in the range of 8–16 kJ mol⁻¹, suggesting that the adsorption process mainly proceeded by ion exchange. The MA-assisted solid-phase synthesis method can produce efficient and environmental-friendly adsorbents.

---

1. Introduction

The rapid development of industry has made our lives more comfortable, but it has also resulted in heavy metal pollution from industrial discharges, such as mining, smelting, electroplating, etc.^1,2 Heavy metals are highly toxic and nonbiodegradable,³ which pose great threats to human health and can be considered as a serious pollution problem.^4,5 Therefore, efficient treatment of hazardous heavy metals is urgent and important.

Traditional treatment methods for the removal of heavy metals include chemical precipitation, redox, ion exchange, membrane separation, and adsorption.⁶ Among these technologies, adsorption is considered to be the most efficient and suitable method for removing heavy metals from wastewater.^7,8 Agricultural wastes, which are rich in lignocellulose, can be used as cheap adsorbents for heavy metals. Wheat straw, rice bran, wood sawdust, pineapple peel, etc. have been successfully applied to remove heavy metals from water.^9,10

However, the capacity of natural cellulose to adsorb metal ions is limited. Therefore, natural cellulosic materials need to be modified to increase their adsorption capacities for heavy metal ions.¹¹ There exist a large number of intermolecular and intramolecular hydrogen bonds between the molecular chains of cellulose, forming a network structure and severely inhibiting the contact and reaction between reagents and the hydroxyl groups in cellulose molecules.¹² Researches on modifying cellulose for the adsorption of heavy metals normally involve the pretreatment of raw cellulose with sodium hydroxide to improve its reactivity and the grafting of functional groups with strong adsorption for metal ions, such as amine and carboxyl groups, to the cellulose surface in organic solvent reaction system.^11,13–15 However, these conventional preparation methods are complex and usually require a long time, and the used organic solvent is harmful to environment. It has important significance to develop a simple, fast, and solvent-free technology for the preparation of cellulose-based adsorbents.

Mechanical activation (MA) changes the crystal structure and the physicochemical properties of solid materials by using high-energy ball milling and is normally used in the pretreatment of solid materials to improve their reactivity. Because MA does not require any solvent or organic medium, it is considered to be a simple and environmentally friendly pretreatment method.¹⁶ In recent years, our laboratory has developed a high-energy ball mill as a solid-phase reactor for the modification of natural polymers, and it has been successfully applied to the modification of lignocellulose with combining MA and chemical reaction in the same equipment.^17–20 Based on these previous studies, this study investigated the preparation of cellulose succinate in the solid phase for simultaneous activation and esterification induced by MA. Fourier transform infrared spectroscopy (FTIR), X-ray diffractometry (XRD), scanning electron microscopy (SEM), and specific surface area analysis were used to measure the changes in the functional groups, crystal structure, microstructure, and specific surface area of the cellulose before and after modification. The cellulose succinates with different degrees of esterification, which were prepared in different MA time, were used as adsorbents for the adsorption of Cd(II) and Pb(II) ions from aqueous solution. By investigating the effect of different MA time on the adsorption ability of modified cellulose, the optimal cellulose-based adsorbent was selected, and its adsorption behavior and kinetics for heavy metal ions were detailedly investigated.

2. Experimental

2.1. Materials

Bleached pulp (BP), which had been treated by pulping and bleaching processes to remove lignin and chromophores, was obtained from a local paper mill (Nanning, China). It had about 86% cellulose and a degree of polymerization of 1300. Other chemical reagents were of analytical grade without further purification and obtained commercially.

2.2. Preparation of adsorbents

The preparation of modified BP (MBP) was performed in a high-energy stirring ball mill. In a typical experiment, 10.0 g of BP, 20.0 g of succinic acid, and 1.5 g of sodium hypophosphite were evenly mixed and then were put in a milling tank. The milling was carried out at the speed of 300 rpm and a constant temperature of 90 °C for all the batch experiments. After milled for different reaction times (reaction time = milling time), the resulting powdered products were washed in sequence with 95% ethanol, deionized water, 98% ethanol, deionized water, and acetone. After drying at 50 °C for 2 h, the modified cellulose with three different degrees of esterification, which named MBP1, MBP2, and MBP3 (milling time = 0.5, 1.0, and 2.0 h, respectively), were obtained. To obtain more carboxylate functions for a better chelating function than the carboxylic groups, the cellulose succinates were treated with a saturated sodium bicarbonate solution for 30 min under a constant stirring, and then they were filtered and rinsed with deionized water to neutrality. The final products were dried and stored in a desiccator.

2.3. Determination of degree of esterification

The degree of esterification of the MBPs was calculated by measuring the amount of carboxylic groups (C_COOH). The C_COOH was measured as follows: 0.1 g of MBP was precisely weighed and placed in a 250 mL conical flask, and then 100 mL of 0.01 mol L⁻¹ sodium hydroxide solution was added to the flask. The solution was stirred at a constant temperature of 30 °C for 1 h. Finally, the solution was filtered, and 25 mL of filtrate was titrated with standard 0.01 mol L⁻¹ HCl. The C_COOH (mmol g⁻¹) in the adsorbent was calculated by using the following equation:²¹where C_NaOH (mmol L⁻¹) is the concentration of NaOH solution, C_HCl (mmol L⁻¹) is the concentration of HCl, V_NaOH (L) is the volume of NaOH solution, V_HCl (L) is the volume of HCl solution used for the titration, and m_mat (g) is the weight of MBP.

2.4. Characterization of structures

FTIR spectra of the samples (BP, MBP1, MBP2, and MBP3) were recorded with a Spectrum One FTIR Spectrometer (FTIR-8400S, SHIMADZU Corporation, Japan) at a resolution of 4 cm⁻¹ from 400 to 4000 cm⁻¹. Pressed pellets were prepared by grinding the powder specimens with KBr in an agate mortar.

Wide angle XRD measurement was carried out on a D/MAX2500V diffractometer (Rigaku, Japan). The patterns with Cu K_α radiation (λ = 0.154 nm) at 40 kV and 30 mA were recorded in the range from 5° to 40°. The crystallinity (CrI) of the samples was estimated by using the following equation:²²

where I₀₀₂ is the diffraction intensity for the crystalline portion of cellulose at about 2θ = 22.5°, and I_am represents the diffraction intensity for the amorphous portion of cellulose at about 2θ = 18°.

The surface morphologies of the samples (BP, MBP1, MBP2, and MBP3) and energy dispersive X-ray (EDX) spectra of MBP2 before and after the adsorption of heavy metal ions were observed by using a scanning electron microscope (S-3400N, Hitachi, Japan). The surfaces of the samples were coated with carbon and gold to be observed and photographed.

BET surface area analyzer (Gemini VII 2390, Micromeritics Corporation, USA) was used to determine the specific surface area of BP, MBP1, MBP2, and MBP3.

2.5. Characteristics of surface charge

Samples of 1.0 g of BP or MBP2 were mixed with 50 mL of 0.1 mol L⁻¹ NaNO₃ solution. HNO₃ or NaOH solution (0.1–1.0 mol L⁻¹) was added to the mixed solution to adjust the initial pH (pH₀) to 2.0, 4.0, 6.0, 8.0, 10.0, and 12.0. Then, the suspensions were intermittently shaken to mix and allowed to stand for 48 h until they reached equilibrium. The final pH (pH_f) was measured, and then ΔpH = pH_f − pH₀ was calculated. The curve of ΔpH versus pH₀ was plotted to determine the point of zero charge (pH_ZPC) of the samples.²³

2.6. Adsorption experiments

Samples of 25.0 mg of BP, MBP1, MBP2, or MBP3 were placed in the 50 mL conical flasks, which 25 mL of metal ion solution (Cd²⁺ or Pb²⁺) was added. NaOH or HCl solution (0.01–1.0 mol L⁻¹) was added to adjust the pH. The flasks were then placed in a constant-temperature shaking bath. The rotation speed and temperature were set at 150 rpm and 25 °C, respectively. Sufficient time was allowed for complete mixing and adsorption. After adsorption, the solution was filtered, and single sweep oscillopolarography was used to determine the concentration of metal ions in solution.²⁴ The amount of metal ions adsorbed onto adsorbents was calculated by using the following equation:where q (mg g⁻¹) is the adsorption capacity of adsorbent, C₀ (mg L⁻¹) is the initial concentration of metal ions, C_e (mg L⁻¹) is the equilibrium concentration of metal ions, V (L) is the volume of the metal ion solution, and m_mat is the weight of adsorbent.

3. Results and discussion

3.1. Characterization

3.1.1. FTIR. As shown in Fig. 1, the IR spectra show two absorption regions at low wavenumbers in 1800–700 cm⁻¹ and high wavenumbers in 3500–2700 cm⁻¹ for all of the samples, which agree with the FTIR spectrum of natural cellulose.²⁵ The spectrum of BP exhibits a strong broad band centered at 3422 cm⁻¹ and the peaks at 2900, 1634, 1370, and 1056 cm⁻¹, which are attributed to O–H stretching vibration (hydroxyl groups of cellulose), C–H stretching vibration, H–O–H bending of adsorbed water, C–H bending, and C–O ether vibration, respectively. After the ball-milling assisted modification, new strong absorption peaks appear at 1729 cm⁻¹ and 1165 cm⁻¹, attributed to the stretching vibrations of C=O and C–O–C in ester groups, respectively, which suggest that the cellulose had been successfully esterified. Moreover, with the increase of milling time, the absorption intensity of the peak at 1729 cm⁻¹ also increased, indicating that the degree of esterification improved with increasing the reaction time. A new weak absorption peak appearing at 1427 cm⁻¹ could be attributed to the bending vibration of OH in carboxyl groups, suggesting that –COOH was introduced to cellulose after the modification. The wide vibration absorption peak at 3401 cm⁻¹ did not show a significant change in intensity because of the presence of unreacted hydroxyl groups in MBPs.¹⁷ However, the area at this peak position became wide, likely caused by the introduced –COOH groups. The FTIR analysis reveals that BP was successful esterified by MA-assisted solid-phase reaction with different milling times, and a large amount of –COOH groups had been introduced to cellulose, which possess strong adsorption for heavy metal ions.

Fig. 1 FTIR spectra of (a) BP, (b) MBP1, (c) MBP2, and (d) MBP3.

3.1.2. XRD. As shown in Fig. 2, there is strong diffraction at 2θ values of 15.2°, 16.3°, and 22.2° for all of the samples, corresponding to the 101, 101̄, and 002 crystal planes for cellulose I, respectively.¹⁸ The diffraction intensity was weak for the 101 and 101̄ crystal planes and these two peaks overlap with each other because BP contains a certain amount of lignin and semicellulose, resulting in the amorphous state of BP.¹⁷ It can be directly observed from the XRD patterns that with the increase of milling time, the 002 crystallization peak gradually decreased and the 101̄ crystallization peak widened with weak intensity. The crystallinity of BP, MBP1, MBP2, and MBP3 can be calculated by eqn (2), and the results are shown in Table 1. The crystallinity of BP was 76.4%, and it decreased to 52.6%, 42.2%, and 33.7% after milled for 0.5, 1.0, and 2.0 h, respectively. These results indicate that MA can significantly disrupt the intramolecular and intermolecular hydrogen bonds in cellulose and effectively destroy the crystal structure of cellulose, leading to the decrease of crystallinity and the increase of amorphous regions.²⁶ Cellulose consists of crystalline and amorphous regions. Chemical reagents are difficult to enter into the crystalline region because of its stable structure, and chemical reactions mainly occur in the amorphous region. On one hand, the intense ball-milling destroys the stable hydrogen bonds, generating free hydroxyl groups and increasing the reactivity of cellulose. On the other hand, ball-milling transforms the stable crystalline structure to an easily accessible and penetrable amorphous structure, thereby enhancing the accessibility of cellulose.

Fig. 2 XRD patterns of (a) BP, (b) MBP1, (c) MBP2, and (d) MBP3.

Table 1 The CrI, surface area, C_COOH, and q_e of BP, MBP1, MBP2, and MBP3

Sample	Milling time (h)	CrI (%)	Surface area (m^2 g^−1)	CCOOH (mmol g^−1)	qe (mg g^−1)
					Cd^2+	Pb^2+
BP	—	76.4	1.5609	0.65	9.67	13.82
MBP1	0.5	52.6	1.7168	3.72	40.96	82.40
MBP2	1.0	42.2	2.0992	5.35	87.56	187.65
MBP3	2.0	33.7	1.9451	5.86	82.85	180.96

---

3.1.3. SEM. As shown in Fig. 3, the surface morphology is dramatically different before and after the MA-assisted esterification. BP exhibits a flat, stripe-like cellulose structure, and the surface is smooth (Fig. 3a). After the modification, the fiber bundle was sheared and broken. The stripe-like fibers gradually disappeared, and new small particles were produced (Fig. 3b–d). This damage is beneficial because it can increase the contact between the esterifying agent and the hydroxyl groups of BP, thereby improving its reactivity.²⁷ In addition, the surfaces of the ball-milled powders were coarse and folded with numerous cracks. These characteristics can enhance the retention of metal ions and the ability to capture these ions. It is notable that the aggregation between the particles occurred after 2.0 h of ball-milling due to the increase of products, thus likely leading to the encapsulation of previous exposed surfaces and the reduction of the adsorptive capacity for metal ions.

Fig. 3 SEM micrographs of (a) BP, (b) MBP1, (c) MBP2, and (d) MBP3.

3.1.4. BET surface area. Specific surface area is an important index for evaluating the adsorption capacity of adsorbents. Table 1 lists the specific surface areas of BP, MBP1, MBP2, and MBP3. The BET surface area was 1.5609 m² g⁻¹ for unmodified BP, and it increased for all of the samples after ball-milling assisted reaction. MBP2 with the milling time of 1.0 h had the maximum BET surface area of 2.0992 m² g⁻¹ because the strip-like structure of BP was sheared and broken during ball milling, and the BP was thinned to powder particles which were rich in folds on surfaces and cracks, leading to an increase in the contact area.¹⁸ However, when the milling time was extended, the specific surface area decreased: excessively long ball milling leads to the aggregation of particles, which caused the encapsulation of particle surfaces, thus decreasing cracks and, in turn, decreasing the specific surface area.

3.1.5. EDX. The EDX spectra of MBP2 before and after the adsorption of heavy metal ions are shown in Fig. 4. MBP2 contains C, O, and Si, among which C and O are the main constituents. After the adsorption of heavy metal ions, the absorption peaks of the corresponding metals appeared in the energy spectra, in which the absorption intensity of Pb²⁺ is markedly higher than that of Cd²⁺.

Fig. 4 Energy distribution spectra of (a) MBP2 and (b) Cd²⁺ and (c) Pb²⁺ loaded MBP2.

3.2. Adsorption studies

3.2.1. Effect of degree of esterification. Table 1 shows that the equilibrium adsorption uptakes of MBP1, MBP2, and MBP3 were 40.96, 87.56, and 82.85 mg g⁻¹ for Cd²⁺ and 82.40, 187.65, and 180.96 mg g⁻¹ for Pb²⁺, respectively. Compared with unmodified cellulose, the adsorption capacity of esterified cellulose enhanced significantly. MBP2 showed the best adsorption among these adsorbents, and the adsorption capacity was 9.1 times that of BP for the adsorption of Cd²⁺ and 13.6 times that of BP for the adsorption of Pb²⁺. On one hand, esterification introduced a large number of carboxyl groups, leading to the increased capability of the adsorbents to capture heavy metal ions. On the other hand, ball milling resulted in the thinning of adsorbent particles, and the surface became more folded with more cracks, leading to an increase in specific surface area. Meanwhile, the increase of active sites enhanced the ability of the adsorbents to retain heavy metal ions. Although the degree of esterification of MBP3 was higher than that of MBP2, the adsorption capacity of MBP3 was lower than that of MBP2. This may due to that excessively long milling time led to the aggregation of particles, which in turn caused the encapsulation of particle surfaces and the decrease of cracks, thus decreasing the exposed active sites.

3.2.2. Effect of pH. Solution pH has a significant effect on the adsorption capacity of adsorbents for heavy metal ions because it is closely related to the surface charge characteristics of the adsorbents, the ionization degree, and the ion types of the adsorbates.²⁸

Fig. 5 shows the pH_ZPC curve for BP and MBP2. The pH_ZPC values obtained at ΔpH = 0 for BP and MBP2 were 5.3 and 4.2, respectively. For pH < pH_ZPC, the adsorbent surface carries a positive charge, whereas for pH > pH_ZPC, the adsorbent surface carries a negative charge.²³ The results suggest that the potential at the point of zero charge for BP decreased after esterification, resulting in more negative charges on the surface of MBP2, which enhanced the adsorption of metal cations.

Fig. 5 pH_ZPC of BP and MBP2.

For the pH in the range of 2–7 and 1.5–6.5, the adsorption of Cd²⁺ and Pb²⁺ by MBP2 is shown in Fig. 6a. The adsorption capacities of Cd²⁺ and Pb²⁺ by MBP2 increased with increasing pH. The largest adsorption capacities were 91.73 and 196.67 mg g⁻¹ at pH 7.0 and 6.5, respectively, which agree well with the results obtained by Gurgel et al.¹⁵ For pH < 3.0, the adsorption capacity of Cd²⁺ and Pb²⁺ was very small because a large amount of H⁺ compete with heavy metal ions at low pH, leading to protonation reactions with the functional groups on the surface of adsorbents.¹⁴ Under this condition, pH < pH_ZPC, and the adsorbent surfaces carried positive charges, which generated electrostatic repulsion to metal cations.²⁹ With increasing pH, this competition weakened; when pH > pH_ZPC, the adsorbent surfaces carried negative charges and easily adsorbed positively charged metal ions. For pH > 6, Cd²⁺ and Pb²⁺ may form hydroxide precipitates, which cannot be efficiently adsorbed by the adsorbents.³⁰ Therefore, the optimum pH for investigating the adsorption ability and behavior of MBP2 is around 6.0.

Fig. 6 Effects of pH (a) and contact time (b) on the adsorption of metal ions by MBP2.

3.2.3. Effect of time. Equilibrium time is an important parameter for adsorption. The effect of adsorption time on the adsorption by MBP2 is shown in Fig. 6b. The adsorption rate of Cd²⁺ and Pb²⁺ by MBP2 is fast, and the adsorption amounts were 77.61 and 175.77 mg g⁻¹ within 20 min, respectively. Adsorption equilibrium was reached for Cd²⁺ at 45 min and for Pb²⁺ at 40 min. The initial stage of adsorption rate was very fast, and then gradually slowed down, finally reached equilibrium. Initially, the concentration of metal ions was high, and the effective adsorption area on the adsorbent surface was large. There are many vacant binding sites for metal ions, all of which gave rise to easy adsorption. As the adsorption progressed, the concentration of metal ions and the active sites decreased, thus resulting in a decreased adsorption rate and ultimately adsorptive equilibrium.¹³

3.3. Adsorption kinetics

To investigate the kinetic mechanism of the adsorption of Cd²⁺ and Pb²⁺ by MBP2 and the adsorption rate and rate-controlling steps, the experimental data were fitted by using pseudo-first order model, pseudo-second order model, intra-particle diffusion model, and Elovich model.

The expression for the pseudo-first order model is shown in eqn (4):³¹

ln(q_e − q_t) = ln q_e − k₁t (4)

The expression for the pseudo-second order model is shown in eqn (5):³²

where q_t and q_e (mg g⁻¹) are the adsorption capacities at time t (min) and at equilibrium, respectively, and k₁ and k₂ (min⁻¹) are the rate constants of pseudo-first order and pseudo-second order adsorption, respectively. k₁, k₂ and q_e can be calculated from the plots of ln(q_e − q_t) versus t (Fig. 7a) and t/q_t versus t (Fig. 7b) using the adsorption data.

Fig. 7 Fitting experimental date of absorption kinetics: (a) pseudo-first order model, (b) pseudo-second order model, (c) intra-particle diffusion model, and (d) Elovich model.

The expression for the intra-particle diffusion model is shown as follows:³³

q_t = K_it^0.5 + C (6)

where the diffusion rate constant K_i (mg g⁻¹ min⁻¹) can be obtained from the slope of the linearized plot of q_t versus t^0.5 (Fig. 7c), and C (mg g⁻¹) is the intercept of this line, which is related to the thickness of the boundary layer.

The Elovich model is a two-parameter model, and its expression is shown in eqn (7):³⁴

where α (mg g⁻¹ min⁻¹) and β (mg g⁻¹) are the rate constants for adsorption and desorption, respectively, which can be calculated from the plot of q_t versus ln t (Fig. 7d).

The parameters of these four aforementioned kinetic models are shown in Table 2. Based on the correlation coefficients (R²), the pseudo-second order model fitted the experimental kinetic data best (R² > 0.999). In addition, compared with the pseudo-first order model, values of q_e(cal) calculated from the pseudo-second order model were in better agreement with the experimental values, q_e(exp). Moreover, the pseudo-first order model can only be applied to describe the initial stage of adsorption. In contrast, the pseudo-second order model can predict the entire period of the adsorption behaviors. Therefore, the pseudo-second order model can better describe the adsorption behavior of heavy metal ions by MBP2.

Table 2 Estimated kinetic model constant for Cd²⁺ and Pb²⁺ adsorption onto MBP2 adsorbent

Kinetic model	Parameters	Metal ion
		Cd^2+	Pb^2+
Pseudo-first order	qe(exp) (mg g^−1)	87.56	187.65
	qe(cal) (mg g^−1)	90.78	273.34
	k1 (min^−1)	0.117	0.177
	R^2	0.9841	0.9552
Pseudo-second order	qe(exp) (mg g^−1)	87.56	187.65
	qe(cal) (mg g^−1)	89.37	190.48
	k2 (g mg^−1 min^−1)	0.0034	0.0022
	R^2	0.9993	0.9994
Intra-particle diffusion	Ki (mg g^−1 min^−1)	12.50	28.40
	C (mg g^−1)	15.06	32.69
	R^2	0.9079	0.8986
Elovich model	α (mg g^−1 min^−1)	42.932	106.164
	β (mg g^−1)	0.0535	0.0250
	R^2	0.9077	0.8745

---

The R² value was very small for fitting the whole process by using the intra-particle diffusion model, suggesting that the whole adsorption process did not agree with the intra-particle diffusion model. If separate fittings were performed, an excellent linear relationship could be obtained. These results suggest that the adsorption process consisted of two stages.³⁵ If the fitted line passes through the origin, the adsorption is controlled by internal diffusion. However, whether whole-process fitting or separate-stage fitting was performed, the line did not pass through the origin, suggesting that internal diffusion was not the only controlling step for the adsorption.

The R² values were 0.9077 and 0.8745 for fitting the Cd²⁺ and Pb²⁺ adsorption by using the Elovich model, suggesting that the adsorption did not agree with the Elovich model.

3.4. Adsorption isotherms

Usually, adsorption isotherms describe the relationship between the amount of the solute that is adsorbed and solute concentration of the solution at equilibrium at a certain temperature and pH. Three adsorption isotherm models were used to fit the experimental data of the adsorption of Cd²⁺ and Pb²⁺ by MBP2.

The expression for the Freundlich isotherm is shown in eqn (8):³⁶

where q_e is the equilibrium adsorption capacity, and 1/n and K_f (L mg⁻¹) are empirical constants. The 1/n value reflects the effect of the concentration of metal ion on the adsorption capacity. According to the plots of ln q_e versus ln C_e shown in Fig. 8a, the parameters of the Freundlich isotherm can be obtained (as shown in Table 3).

Fig. 8 Fitting experimental date of absorption isotherms: (a) Freundlich isotherm, (b) Langmuir isotherm, and (c) Dubinin–Radushkevich isotherm.

Table 3 The Freundlich, Langmuir, and Dubinin–Radushkevich isotherms parameters for Cd²⁺ and Pb²⁺ adsorption onto MBP2 adsorbent

Equilibrium model	Parameters	Metal ion
		Cd^2+	Pb^2+
Freundlich isotherm	1/n	0.254	0.128
	Kf (L mg^−1)	26.32	8.12
	R^2	0.9777	0.9742
Langmuir isotherm	Qmax (mg g^−1)	109.41	207.90
	KL (L mg^−1)	0.041	0.106
	R^2	0.9999	1.0000
Dubinin–Radushkevich isotherm	Qmax (mg g^−1)	112.91	227.15
	Kd (mol^2 kJ^−2)	0.0029	0.0020
	E (kJ mol^−1)	13.13	15.81
	R^2	0.9908	0.9914

---

The R² values were 0.9777 and 0.9742 as using Freundlich isotherm model to fit the adsorption of Cd²⁺ and Pb²⁺ by MBP2, respectively. The 1/n values were 0.254 and 0.128 for the adsorption of Cd²⁺ and Pb²⁺, respectively. The 1/n values between 0.1 and 0.5 indicate that the adsorption is easy to occur; by contrast, values of 1/n > 2.0 indicate that it is difficult to adsorb. This can effectively explain the phenomenon that the adsorption could quickly reach equilibrium. However, the Freundlich isotherm is an empirical model that cannot explain the adsorption mechanism.

The expression of the Langmuir isotherm is shown as follows:³⁷

where C_e is the equilibrium concentration of metal ions, q_e is the equilibrium adsorption capacity, Q_max (mg g⁻¹) is the saturated adsorption capacity of adsorbent, and K_L (L mg⁻¹) is the Langmuir adsorption constant related the affinity of binding sites. The parameters of the Langmuir isotherm can be obtained from the plots of C_e/q_e versus C_e as shown in Fig. 8b.

Table 3 shows that the R² values were 0.9999 and 1.0000 as using the Langmuir isotherm to fit the adsorption of Cd²⁺ and Pb²⁺ by MBP2, respectively, which suggest that the adsorption was in excellent agreement with the Langmuir isotherm model. However, the calculated Q_max values for Cd²⁺ and Pb²⁺ were 109.41 and 207.90 mg g⁻¹, respectively, which were larger than the experimental values. The Langmuir isotherm can well explain the phenomenon that the adsorption rate was fast in the initial stage but slowed down thereafter. The Langmuir isotherm assumes that the adsorption mechanism is single-layer adsorption, and the binding sites accordingly decrease with the increase of adsorption time.

The expression for the Dubinin–Radushkevich isotherm is shown in eqn (10):³⁸

ln q_e = ln Q_max + K_dε² (10)

where q_e is the equilibrium adsorption capacity, Q_max (mg g⁻¹) is the saturated adsorption capacity of adsorbent, K_d is a constant related to the adsorption energy, and ε is the Polanyi potential, which is determined as follows:where R, T, and C_e denote the gas constant (8.314 J mol⁻¹ K⁻¹), the absolute temperature of adsorption (K), and the equilibrium concentration of the adsorbates (mg L⁻¹), respectively. K_d (mol² kJ⁻²) and Q_max can be obtained from the slope and intercept of the plots of ln q_e versus ε² in Fig. 8c. The mean adsorption energy (E) can be calculated with the following equation by using K_d:³⁹

The parameters and the linear correlation coefficient for the Dubinin–Radushkevich isotherm are listed in Table 3. The E values for the adsorption of Cd²⁺ and Pb²⁺ by MBP2 were 13.13 and 15.81 kJ mol⁻¹, respectively. The E value is very useful for determining the adsorption mechanism. Usually, a value of E < 8 kJ mol⁻¹ indicates the physical adsorption, while a E value between 8 and 16 kJ mol⁻¹ is associated with ion exchange adsorption.^40,41 In this study, the E values for the adsorption of Cd²⁺ and Pb²⁺ by MBP2 were between 8 and 16 kJ mol⁻¹, suggesting that the adsorption was mainly governed by ion exchange, which is attributed to the introduction of a large number of carboxyl groups to BP by MA-assisted modification.

As comparing the parameters of the isothermal adsorption models for the adsorption of metal ions by MBP2, it can be found that the values of n, Q_max, K_L, and E for Pb²⁺ were all greater than those for Cd²⁺, suggesting that the adsorption of Pb²⁺ is stronger than that of Cd²⁺. There are two possible reasons for this difference. Firstly, compared with Cd²⁺, the ionic radius and molecular weight of Pb²⁺ are larger, so Pb²⁺ has a larger physical surface area and is easy to be captured.⁴² Secondly, the chemical stability constants (log K) of metal–MBP2 are Pb²⁺ (7.45) > Cd²⁺ (5.73), which lead to the easier adsorption of Pb²⁺.⁴³

4. Conclusions

BP was successfully modified to produce cellulose succinate adsorbents by MA-assisted solid-phase reaction, and large amounts of carboxyl groups were introduced. FTIR, XRD, SEM, and specific surface area analysis were used to characterize the structural characteristics of cellulose succinate and investigate the process and mechanism of the MA-assisted modification. By investigating the effects of different degrees of esterification of modified cellulose on the adsorption of Cd²⁺ and Pb²⁺, MBP2, which had the best adsorption, was selected as the adsorbent to study the factors that influenced the adsorption and the adsorption mechanism. The results of equilibrium adsorption showed that the adsorption rate of MBP2 was fast and reached equilibrium for Cd²⁺ and Pb²⁺ at 45 and 40 min, respectively. The kinetic study indicated that the adsorption process of MBP2 was well-described by the pseudo-second order model, and internal diffusion was not the only controlling step for adsorption. The studies of adsorption isotherms suggested that the Langmuir isotherm model well explained the experimental data of Cd²⁺ and Pb²⁺ adsorption onto MBP2, and the adsorption capacities Q_max for Cd²⁺ and Pb²⁺ were 109.41 and 207.90 mg g⁻¹, respectively. The E values for the adsorption of Cd²⁺ and Pb²⁺ by MBP2 calculated by the Dubinin–Radushkevich equation were all in the range of 8–16 kJ mol⁻¹, suggesting that the adsorption was mainly governed by ion exchange. In summary, the MA-assisted solid-phase synthesis method can produce efficient and environmental-friendly adsorbents.

Acknowledgements

This research was supported by National Natural Science Foundation of China (No. 51163002 and 51463003), Guangxi Natural Science Foundation of China (No. 2013GXNSFDA019004, 2013GXNSFBA019028, and 2014GXNSFBA118057), the Foundation of Guangxi Department of Education, China (YB2014015), Guangxi Distinguished Experts Special Foundation of China, and the Scientific Research Foundation of Guangxi University, China (Grant No. XTZ140787).

References

1. Z. A. Al-Othman, M. Naushad and Inamuddin, Chem. Eng. J., 2011, 172, 369–375.
2. M. C. Tomei, D. M. Angelucci, M. C. Annesini and A. J. Daugulis, J. Hazard. Mater., 2013, 262, 31–37.
3. F. Zhao, E. Repo, D. Yin and M. E. T. Sillanpää, J. Colloid Interface Sci., 2013, 409, 174–182.
4. A. M. Alsabagh, M. Fathy and R. E. Morsi, RSC Adv., 2015, 5, 55774–55783.
5. P. C. Nagajyoti, K. D. Lee and T. V. M. Sreekanth, Environ. Chem. Lett., 2010, 8, 199–216.
6. M. Islam and R. Patel, J. Hazard. Mater., 2009, 172, 707–715.
7. E. Repo, J. K. Warchoł, A. Bhatnagar and M. Sillanpää, J. Colloid Interface Sci., 2011, 358, 261–267.
8. C. Cheng, J. Wang, X. Yang, A. Li and C. Philippe, J. Hazard. Mater., 2014, 264, 332–341.
9. T. A. H. Nguyen, H. H. Ngo, W. S. Guo, J. Zhang, S. Liang, Q. Y. Yue, Q. Li and T. V. Nguyen, Bioresour. Technol., 2013, 148, 574–585.
10. M. Ghasemi, M. Naushad, N. Ghasemi and Y. Khosravi-fard, J. Ind. Eng. Chem., 2014, 20, 2193–2199.
11. F. V. Pereira, L. V. A. Gurgel and L. F. Gil, J. Hazard. Mater., 2010, 176, 856–863.
12. A. T. W. M. Hendriks and G. Zeeman, Bioresour. Technol., 2009, 100, 10–18.
13. Z. Sun, Y. Liu, Y. Huang, X. Tan, G. Zeng, X. Hu and Z. Yang, J. Colloid Interface Sci., 2014, 434, 152–158.
14. S. Hokkanen, E. Repo and M. Sillanpää, Chem. Eng. J., 2013, 223, 40–47.
15. L. V. A. Gurgel, R. P. D. Freitas and L. F. Gil, Carbohydr. Polym., 2008, 74, 922–929.
16. P. Baláž and E. Dutková, Miner. Eng., 2009, 22, 681–694.
17. H. Hu, H. Li, Y. Zhang, Y. Chen, Z. Huang, A. Huang, Y. Zhu, X. Qin and B. Lin, RSC Adv., 2015, 5, 20656–20662.
18. Z. Huang, Y. Tan, Y. Zhang, X. Liu, H. Hu, Y. Qin and H. Huang, Bioresour. Technol., 2012, 118, 624–627.
19. Y. Zhang, T. Gan, Y. Luo, X. Zhao, H. Hu, Z. Huang, A. Huang and X. Qin, Compos. Sci. Technol., 2014, 102, 139–144.
20. Y. Zhang, Q. Li, J. Su, Y. Lin, Z. Huang, Y. Lu, G. Sun, M. Yang, A. Huang, H. Hu and Y. Zhu, Bioresour. Technol., 2015, 177, 176–181.
21. K. A. G. Gusmão, L. V. A. Gurgel, T. M. S. Melo, C. D. F. Carvalho and L. F. Gil, J. Environ. Manage., 2014, 133, 332–342.
22. S. Kim and M. T. Holtzapple, Bioresour. Technol., 2006, 97, 583–591.
23. L. B. L. Lim, N. Priyantha and N. H. M. Mansor, Environ. Earth Sci., 2015, 73, 3239–3247.
24. D. Monticelli, E. Ciceri and C. Dossi, Anal. Chim. Acta, 2007, 594, 192–198.
25. A. M. Adel, Z. H. A. El Wahab, A. A. Ibrahim and M. T. Al Shemy, Bioresour. Technol., 2010, 101, 4446–4455.
26. W. Zhang, M. Liang and C. Lu, Cellulose, 2007, 14, 447–456.
27. Z. Huang, N. Wang, Y. Zhang, H. Hu and Y. Luo, Composites, Part A, 2012, 43, 114–120.
28. O. Karnitz, L. V. A. Gurgel, J. C. P. de Melo, V. R. Botaro, T. M. S. Melo, R. P. de Freitas Gil and L. F. Gil, Bioresour. Technol., 2007, 98, 1291–1297.
29. D. H. K. Reddy, K. Seshaiah, A. V. R. Reddy and S. M. Lee, Carbohydr. Polym., 2012, 88, 1077–1086.
30. D. Zhou, L. Zhang, J. Zhou and S. Guo, Water Res., 2004, 38, 2643–2650.
31. Y. Ho, Scientometrics, 2004, 59, 171–177.
32. Y. Ho, J. Hazard. Mater., 2006, 136, 681–689.
33. W. J. Weber and J. C. Morris, J. Sanit. Eng. Div., Am. Soc. Civ. Eng., 1963, 89, 31–60.
34. S. H. Chien and W. R. Clayton, Soil Sci. Soc. Am. J., 1980, 44, 265–268.
35. Q. Zhong, Q. Yue, Q. Li, B. Gao and X. Xu, Carbohydr. Polym., 2014, 111, 788–796.
36. D. H. K. Reddy, Y. Harinath, K. Seshaiah and A. V. R. Reddy, Chem. Eng. J., 2010, 162, 626–634.
37. I. Langmuir, J. Am. Chem. Soc., 1916, 38, 2221–2295.
38. M. M. Dubinin and L. V. Radushkevich, Proc. Acad. Sci. USSR, Phys. Chem. Sect., 1947, 55, 331–333.
39. M. M. Dubinin, Chem. Rev., 1960, 60, 235–266.
40. A. S. Singha and A. Guleria, J. Environ. Chem. Eng., 2014, 2, 1456–1466.
41. L. Mihaly-Cozmuta, A. Mihaly-Cozmuta, A. Peter, C. Nicula, H. Tutu, D. Silipas and E. Indrea, J. Environ. Manage., 2014, 137, 69–80.
42. M. A. Hossain, H. H. Ngo, W. S. Guo, L. D. Nghiem, F. I. Hai, S. Vigneswaran and T. V. Nguyen, Bioresour. Technol., 2014, 160, 79–88.
43. C. R. Adhikari, D. Parajuli, K. Inoue, K. Ohto and H. Kawakita, Chemosphere, 2008, 72, 182–188.

---