Synthesis of poly(m-phenylenediamine)/iron oxide/acid oxidized multi-wall carbon nanotubes for removal of hexavalent chromium

Abstract

Poly(m-phenylenediamine)-coated Fe₃O₄/o-MWCNTs nanoparticles (PmPD/Fe₃O₄/o-MWCNTs) were synthesized by one-step chemical oxidation polymerization. The materials were characterized by transmission electron microscopy, Fourier-transform infrared spectroscopy, X-ray diffraction, vibrating sample magnetometry, and Brunauer–Emmett–Teller surface area measurement. Hexavalent chromium adsorption by PmPD/Fe₃O₄/o-MWCNTs was found to be strongly dependent on the solution pH. The adsorption isotherm of Cr(VI) onto PmPD/Fe₃O₄/o-MWCNTs fitted the Langmuir isotherm model at different temperatures and the maximum adsorption capacity was determined to be 346 mg g⁻¹. The adsorption process of different initial concentrations can be described by the pseudo-second-order kinetic model. The intraparticle diffusion study revealed that external mass transfer and intraparticle diffusion apparently influenced the overall rate process. The values of thermodynamic parameters indicated that the adsorption process was spontaneous and endothermic. In addition, the adsorption mechanism included both the physical and the chemical adsorption mechanisms. After adsorption, PmPD/Fe₃O₄/o-MWCNTs could be conveniently separated from the media by an external magnetic, and the adsorption capacity can remain at up to 52% after five times of usage. These results demonstrate that PmPD/Fe₃O₄/o-MWCNTs are a good candidate for the removal of Cr(VI) from aqueous solution.

---

1. Introduction

Chromium compounds are widely used in industries such as leather, textile, metal plating, battery and pigments. In aqueous media, chromium exists in trivalent and hexavalent states. However the hexavalent form has been considered more hazardous to public health due to its mutagenic and carcinogenic properties.¹ According to the World Health Organization, it has limited 5 μg L⁻¹ as the maximum allowable emission standard of Cr(VI).² Therefore, it is necessary to scavenge Cr(VI) ions from industrial effluents before their discharge into the environment in order to prevent the deleterious impact of Cr(VI) ions on the ecosystem and public health. Several conventional methods for removing heavy metal ions wastewater include, physical and physicochemical treatment technologies, such as ion-exchange, electro dialysis, membrane filtration, reverse osmosis, chemical precipitation and adsorption.^3–7 Among these methods, adsorption is one of the most economically favorable and a technically easy method. However, biosorbents often cause potential secondary pollution in extreme conditions, have weak mechanical strength, are not easily separated, and are not sufficiently effective. Thus, developing new materials with simple synthesis, high adsorption capacity, easy separation, and good stability in extreme conditions, is highly desired.⁸

Recently, the applications of carbon nanotubes (CNTs) as potential adsorbents have been widely studied for the removal of metallic pollutants.^9–15 CNTs have been widely used as a potential adsorbent because of their high surface area to weight ratios, small size, hollow/layered structures, and hydrophobic surfaces. Considering their large surface areas and π–π electrostatic interactions, CNTs exhibit exceptional adsorption capabilities for gas and liquid phases, such as organic pollutants,¹⁶ inorganic pollutants¹³ and several heavy metal ions.^17–19 However, the major difficulties with small size CNTs are the formation of agglomerates and separation from the aqueous medium because of the strong intermolecular van der Waals interaction between tubes. Recently, magnetic separation technology has been introduced as a particular separation method capable of quickly separating magnetic adsorbents from an aqueous solution via the application of an external magnetic field.²⁰ Furthermore, due to the existence of synergistic effects, the combination of magnetic structures and adsorbents has extended the possibilities for applications. Carbon nanotubes filled with magnetic nanoparticles are very interesting as new materials for applications in adsorption because they provide a rapid and effective technology for separating materials from the aqueous phase.^21,22 However, due to agglomeration, the active sites of CNTs are blocked resulting in decreased adsorption capacity. One way to alleviate this problem is the modification of CNTs, using surfactant, polymer, etc.^23,24

More recently, polymeric material have been widely studied for the removal of metallic pollutants. Poly(m-phenylenediamine)s, the diamine derivatives of polyaniline, have been given special concerns due to the higher content of functional groups in conjugated structure and more intriguing multifunctionality.²⁵ It is the most important scavenger of aqueous trace metals due to its seemingly dominant adsorptive behavior. Nevertheless, existing studies seldom concern their usage on Cr(VI) removal. To our knowledge, this is the first report of the preparation of synthesize PmPD/Fe₃O₄/o-MWCNTs magnetic nanocomposites for the removal of Cr(VI).

In this work, PmPD doped Fe₃O₄/o-MWCNTs (PmPD/Fe₃O₄/o-MWCNTs) nanocomposite was prepared by in situ oxidative polymerization. The magnetic nanocomposite was characterized using transmission electron microscopy (TEM), X-ray diffraction (XRD), Fourier-transform infrared spectroscopy (FTIR) analysis, Brunauer, Emmett, and Teller surface area measurement (BET) analyses, and vibrating sample magnetometry (VSM). Influences of pH, contact time at different initial concentrations and temperature were examined in order to evaluate adsorption performance of PmPD/Fe₃O₄/o-MWCNTs for Cr(VI) removal. In addition, the adsorption kinetics, rate-controlling mechanisms, thermodynamics and the reusability of the adsorbents were also investigated.

2. Experimental

2.1. Materials

MWCNTs having diameter 20–30 nm and length of 5–20 μm, were purchased from Alpha Nano Technology Co., Ltd. (China). Benzyl ether and Fe(acac)₃ (purity, 99.9%) were purchased from Sigma-Aldrich. Potassium dichromate (K₂Cr₂O₇) was used to prepare all hexavalent chromium-containing solution. PmPD adsorbents were prepared by the improved chemical oxidation method. All reagents mentioned above were of analytical grade and were used as received. The biochemical reagent L-tryptophan was provided by Shanghai Biological Technology Development Co. Ltd. (China).

2.2. Preparation of o-MWCNTs

The raw-MWCNTs were immersed in a 250 mL three neck flask with a concentrated nitric acid for 24 h. Then the mixture was stirred and heated to reflux at 130 °C for 4 h. Afterward, the black mixture were collected by filtration, washed by distilled water and dried at 60 °C. These materials were denoted as “o-MWCNTs.”

2.3. Preparation of Fe₃O₄/o-MWCNTs magnetic nanocomposites

Fe₃O₄/o-MWCNTs magnetic nanocomposites were prepared by using a published method with a slight modification.²² Firstly, 1 g Fe(acac)₃, 0.5 g MWCNTs, 0.5 g L-tryptophan and 20 mL benzyl ether were added into a flask followed by ultrasonication for 30 min. The mixture solution was then heated to 300 °C at reflux for 2 h. After cooling to room temperature, the black mixture was magnetically separated by a magnet and cleaned with ethanol several times. Finally the as-prepared Fe₃O₄/o-MWCNTs magnetic nanocomposites were extracted by ethanol for 24 h to remove residual benzyl ether and dried in vacuum at 30 °C.

2.4. Preparation of PmPD doped Fe₃O₄/o-MWCNTs composite

PmPD/Fe₃O₄/o-MWCNTs nanocomposite was synthesized via in situ oxidative polymerization as follows. 1 g Fe₃O₄/o-MWCNTs and 1 g mPD was dissolved in 20 mL of distilled water with equably stirring at 25 °C. Subsequently, a certain concentration of NaOH solution with the same amount and speed as the ammonium persulfate (APS) solution was added dropwise to the suspension, and the reaction mixtures were kept stirring for another 5 h at 25 °C. The products were collected by filtration, and washed by distilled water, ammonia water (1 : 1,v/v), ethanol and dried at 60 °C under vacuum for 24 h.

2.5. Characterization of the adsorbents

The surface morphology of Fe₃O₄/o-MWCNTs and PmPD/Fe₃O₄/o-MWCNTs was examined on a transmission electron microscope. The functional groups of the o-MWCNTs, Fe₃O₄/o-MWCNTs, and PmPD/Fe₃O₄/o-MWCNTs were verified using a NEXUS 670 FTIR spectrometer. X-Ray diffraction was used to investigate the crystal structure of the magnetic nanoparticles. The specific surface area, pore volume, and pore size was calculated by the Brunauer–Emmett–Teller (BET) method. The magnetic properties were characterized by vibrating sample magnetometry (VSM, LAKESH ORE-7304) at room temperature.

2.6. Analysis

The concentration of hexavalent chromium was determined by 1,5-diphenylcarbazide spectrophotometric method. 1,5-Diphenylcarbazide forms a pink complex in the presence of Cr(VI) ions in acidic solutions as a complexing agent for Cr(VI). The concentration of Cr(VI) was calculated from absorbance at 540 nm wavelength using UV spectrophotometer after adsorption experiment.

2.7. Adsorption experiments

Adsorption experiments were carried out in 100 mL stoppered flasks, each of which contained 50 mL of Cr(VI) solution. An amount of PmPD/Fe₃O₄/o-MWCNTs was added to each flask and shaken at 250 rpm min⁻¹ in a temperature-controlled shaker with a shaker. These data were used to calculate the amounts of Cr⁶⁺ adsorbed. The absorption capacity (q_e, mg g⁻¹) can be expressed by the following equation:where, C₀ and C_e are the initial and equilibrium concentration (mg L⁻¹) of Cr(VI) respectively, V is the volume (L) of solution and m is the weight (g) of adsorbent.

2.8. Desorption experiments

In order to investigate the reusability of the PmPD/Fe₃O₄/o-MWCNTs nanocomposites, 0.1 M NaOH was used as the desorption solution for the desorption experiment. 0.032 g of PmPD/Fe₃O₄/o-MWCNTs was first contacted with 100 mL of 300 mg L⁻¹ Cr⁶⁺ for 120 min. After adsorption, desorption was carried out by washing the adsorbents with distilled water several times, then the solution of 100 mL of 0.1 M NaOH was added into the adsorbed Cr⁶⁺ adsorbents, and the suspensions were shaken at for 120 min. The solid and liquid phases were separated by a magnet. The above procedure was repeated for five times to test the reusability of the PmPD/Fe₃O₄/o-MWCNTs.

3. Results and discussion

3.1. Characterization

The morphology of the Fe₃O₄/o-MWCNTs nanocomposites and PmPD/Fe₃O₄/o-MWCNTs nanocomposites were examined by TEM, as shown in Fig. 1. Fig. 1(a) and (b) show the images of the pure Fe₃O₄ nanoparticles under different magnifications. It can be clearly seen that the morphology of the pure Fe₃O₄ nanoparticles is nearly cubic. This interaction of Fe(acac)₃ with L-tryptophan during the nucleation and growth phase may force the crystal growth to become cubic in shape, surrounded by a (100) crystal plane which has the lowest surface free energy and the most thermodynamically stable shape.²⁶ Meanwhile, the o-MWCNTs have been coated with cubic iron oxide nanoclusters that are often aggregated by virtue of their magnetic nature. Fig. 1(c) and (d) show the images of the Fe₃O₄/o-MWCNTs nanoparticles under different magnifications. Fig. 1(c) shows that the MWCNTs were visible debundled and shorted after oxidation with a nitric acid. Fig. 1(d) reveals that the surface of the CNTs was loaded with nanoclusters, which is composed of several nanocrystals. Fig. 1(e) and (f) show the morphology of the PmPD/Fe₃O₄/o-MWCNTs under different magnifications. The PmPD/Fe₃O₄/o-MWCNTs were fully coated with PmPD polymer layers. In addition, the PmPD/Fe₃O₄/o-MWCNTs composite displayed a special uniform and continuous coating layer. These characteristics may provide a good platform for the adsorption of metal ions.

Fig. 1 (a)–(b) are the images of the pure Fe₃O₄ nanoparticles under different magnifications; (c)–(d) are the images of Fe₃O₄/o-MWCNTs nanocomposites under different magnifications; (e)–(f) are the images of PmPD/Fe₃O₄/o-MWCNTs nanocomposites under different magnifications.

The structural information for the Fe₃O₄/o-MWCNTs nanoparticles and the PmPD/Fe₃O₄/o-MWCNTs nanocomposites was obtained by XRD as shown in Fig. 2(a). The new diffraction peak at 2θ = 25.96° can be indexed to the graphite structure (002) reflection of the MWCNTs. After surface modification, the peaks at 2θ values at 30.19°, 35.45°, 43.17°, 53.49°, 56.99°, and 62.57° are respectively consistent with the (220), (311), (400), (422), (511) and (440),which are in line with the standard XRD data for the cubic spinel crystal structure of bulk magnetite (JCPDS file no.19-0629).²⁷ The peak of the Fe₃O₄/o-MWCNTs nanoparticles at 53.49 (422) disappeared in the PmPD/Fe₃O₄/o-MWCNTs, which may be due to the fact that the PmPD is coated on Fe₃O₄/o-MWCNTs.

Fig. 2 XRD patterns of Fe₃O₄/o-MWCNTs and PmPD/Fe₃O₄/o-MWCNTs.

FT-IR analysis is applied to investigate the types of functional groups attached on the surface of the MWCNTs. Fig. 3 compares the FT-IR spectra of o-MWCNTs, Fe₃O₄/o-MWCNTs and PmPD/Fe₃O₄/o-MWCNTs magnetic nanocomposites. The characteristic peaks at 3425 and 1727 cm⁻¹ in the spectrum (o-MWCNTs) was generally attributed to the stretching vibration of ν(OH) and ν(C=O) of the carboxylic groups (COOH), respectively. Peaks were observed at the 1650 cm⁻¹ to 1550 cm⁻¹ correspond to the C=O groups in different environments (ketone/quinone) and to C=C in aromatic rings; whereas the bands in the range of 1300 cm⁻¹ to 950 cm⁻¹ range have proven the presence of C–O bonds in various chemical surroundings.²⁸ In the spectrum (Fe₃O₄/o-MWCNTs), the characteristic peak strongly presented at 586 cm⁻¹ related to the stretching vibration of Fe–O–Fe, characteristic of Fe₃O₄; the comparison of o-MWCNTs and Fe₃O₄/o-MWCNTs FTIR spectra shows a shift of the C=O stretching peak from 1727 to 1733 cm⁻¹, suggesting that magnetite nanoparticles are anchored to the MWCNT carboxyl groups.²¹ For the spectrum of the PmPD/Fe₃O₄/o-MWCNTs magnetic nanocomposites, the vibrational peaks centered at 1618, 1508 and 1238 cm⁻¹ can be attributed to the stretching frequencies of quinoid imine, benzenoid amine, and C–N band respectively.²⁹ Meanwhile, the two broad peaks between 3416 and 2921 cm⁻¹ are attributed to the stretching vibration of –NH–.

Fig. 3 FTIR transmission spectra of o-MWCNTs, Fe₃O₄/o-MWCNTs, and PmPD/Fe₃O₄/o-MWCNTs.

The physical properties of o-MWCNTs, Fe₃O₄/o-MWCNTs, and PmPD/Fe₃O₄/o-MWCNTs are listed in Table 1 (Fig. S1†). The surface area, pore volume, and pore size of Fe₃O₄/o-MWCNTs and PmPD/Fe₃O₄/o-MWCNTs are evidently much lower than the o-MWCNTs. The generation of Fe₃O₄ nanocomposites and PmPD blocks the pore entrances, therefore decreasing the surface area, pore volume, and pore size of the MWCNTs. However, the existence of Fe₃O₄ and PmPD play an important role in the enhancement of the adsorption capacity.

Table 1 Pore structure parameters of different materials

Materials	Surface area (m^2 g^−1)	Pore volume (cm^3 g^−1)	Pore size (nm)
Fe3O4/o-MWCNTs	110.2	0.5917	21.48
PmPD/Fe3O4/o-MWCNTs	45.8	0.1936	16.65

---

The magnetic properties of the Fe₃O₄/o-MWCNTs and PmPD/Fe₃O₄/o-MWCNTs nanocomposites are depicted in Fig. 4. Their saturation magnetization are 22.6 emu g⁻¹ and 11.8 emu g⁻¹, respectively. Furthermore, the two samples exhibit superparamagnetic behavior at room temperature with no coercivity and remanence. In addition, the saturation magnetization value of PmPD/Fe₃O₄/o-MWCNTs is lower than that of the Fe₃O₄/o-MWCNTs, which can be mainly attributed to the existence of PmPD. Fig. 5(a) shows photographs of PmPD/Fe₃O₄/o-MWCNTs nanocomposites dispersed in ethanol. Fig. 5(b) shows the PmPD/Fe₃O₄/o-MWCNTs can be promptly separated from their dispersion by holding the samples close to a magnet, indicating that it is promising to be used as a magnetic adsorbent to remove contaminants in wastewater.

Fig. 4 Magnetic hysteresis curves for Fe₃O₄/o-MWCNTs and PmPD/Fe₃O₄/o-MWCNTs.

Fig. 5 The inset demonstrates that PmPD/Fe₃O₄/o-MWCNTs (a) can be dispersed in ethanol and separated from ethanol by a magnet (b).

3.2. Adsorption capacity of PmPD/Fe₃O₄/o-MWCNTs

The adsorption capacities of PmPD/Fe₃O₄/o-MWCNTs compared with other materials as shown in Fig. 6. It can be seen that PmPD/Fe₃O₄/o-MWCNTs are the most effective removal materials compared with the other materials which indicates PmPD/Fe₃O₄/o-MWCNTs provide more binding sites for Cr⁶⁺.

Fig. 6 Adsorption capacities of different materials with regard to Cr⁶⁺ adsorption.

The adsorption capacity of PmPD/Fe₃O₄/o-MWCNTs is higher than that of raw-MWCNTs, suggesting that the organic functional groups on the o-MWCNTs, such as hydroxyl, carboxyl, and carbonyl groups, are expected to be the active centers for Cr⁶⁺. Furthermore, the adsorption capacity of Fe₃O₄/o-MWCNTs is much higher than that of o-MWCNTs because the Fe₃O₄ coated on the o-MWCNTs and Fe₃O₄ was a good adsorbent for Cr⁶⁺.³⁰ Therefore, the adsorption capacities of different materials mainly depend on the functional groups.

3.3. The effect of solution pH

pH is one of the most important parameters controlling the metal ion adsorption process. In order to optimize the pH for maximum adsorption capacity of Cr⁶⁺ on PmPD/Fe₃O₄/o-MWCNTs magnetic nanocomposites, adsorption experiment was conducted at various initial pH values. Fig. 7 shows that the performance of PmPD/Fe₃O₄/o-MWCNTs became better as lowering the initial pH of Cr(VI) solution, and the adsorption capacity decreases with the increase in pH from 2.0 to 11.0. In the acid media, HCrO₄⁻, Cr₂O₇²⁻ and H₂CrO₄²⁻ are the predominant species. In basic solutions chromium exists in the form of CrO₄²⁻. If pH ranges from 2 to 6, HCrO₄⁻ and Cr₂O₇²⁻ are predominant in equilibrium.³¹ The adsorption free energy of HCrO₄⁻ is lower than that of CrO₄²⁻ and Cr₂O₇²⁻; and consequently, HCrO₄⁻ is more favorably adsorbed than CrO₄²⁻ and Cr₂O₇²⁻ at the same concentration.³² Meanwhile the surface of adsorbents was positive charged at low pH and negative-charged with pH increased. Thus, low pH was favorable for HCrO₄⁻ adsorption because of the electrostatic attraction or weakness of electrostatic repulsion. With the pH increased, the electrostatic repulsion between CrO₄²⁻ and negatively charged surface became strong, resulting in the declined removal efficiency.³³

Fig. 7 Effect of solution pH on chromium ions adsorbed onto PmPD/Fe₃O₄/o-MWCNTs.

3.4. Adsorption isotherms study

Adsorption isotherm studies are important to established the relationship between adsorbent and adsorbate at equilibrium. Several adsorption isotherms originally are used for adsorption equilibria in heavy metals adsorption. Some well-known ones are Freundlich, Langmuir, Temkin, Redlich–Paterson and Sips equation.³⁴ In this study, Langmuir, Freundlich and Temkin models were used for adsorption isotherms.

Langmuir isotherm equation is derived from the assumption that the adsorbent surface has a fixed number of equivalent binding sites, and the monolayer adsorption occurs without transmigration of adsorbate on the surface of adsorbent isotherm.^35,36 This process is represented as follows:

where q_e (mg g⁻¹) is the amount of solution adsorbed per unit mass of the adsorbent, C_e (mg L⁻¹) is the solute equilibrium concentration, q_m (mg g⁻¹) is the maximum adsorbate amount that forms a complete monolayer on the surface, and b (L mg⁻¹) is the Langmuir constant related to adsorption heat. When C_e/q_e is plotted against C_e and the data are regressed linearly, the q_m and b constants can be calculated from the slope and the intercept (Fig. S2†).

The essential characteristics of the Langmuir isotherm can be expressed by the dimensionless equilibrium parameter, which R_L is defined as follows:³⁷

where b is the Langmuir coefficient, C₀ is the initial Cr⁶⁺ concentration (mg L⁻¹). R_L values indicate the type of isotherm. The R_L values for each of the different initial concentrations are between 0 and 1 indicating favorable adsorption of Cr(VI) onto PmPD/Fe₃O₄/o-MWCNTs.

Freundlich isotherm assumes that the binding sites on the surface of adsorbent are heterogeneous, the adsorption is more difficult as more and more binding sites are occupied by adsorbates and multilayer adsorption can occur. The Freundlich isotherm³⁷ can be represented in the following form:

where q_e and C_e are the equilibrium concentrations of chromium(VI) in the adsorbed (mg g⁻¹) and liquid phases (mg L⁻¹) and K_F and n are the Freundlich constants which are related to adsorption capacity and adsorption intensity, respectively. Both constants can be calculated from the slope and intercept of the linear plot of ln q_e versus ln C_e (Fig. S4†).

Temkin isotherm model is applicable to evaluate the adsorption potentials of the applied adsorbents for adsorbate ions, and assumes the effects of the heat of adsorption that decreases linearly with coverage of the adsorbate and adsorbent interactions, and adsorption is characterized by a uniform distribution of binding energies, up to some maximum binding energy.³⁸ The Temkin isotherm can be represented in the following form:

q_e = B_t ln(K_T) + B_t ln(C_e) (5)

where

K_T is the equilibrium binding constant (L mg⁻¹) corresponding to the maximum binding energy and the constant B_t is related to the heat of adsorption. R is the universal gas constant (8.314 J K⁻¹ mol⁻¹), T is the absolute temperature (K) and b is the Temkin constant related to the heat of sorption (J mol⁻¹). A plot of q_e versus ln(C_e) enables the determination of the isotherm constants K_T and B_t (Fig. S5†).

Table 2 shows that the Langmuir model fit the adsorption data better than the Freundlich model and Temkin isotherm model. This result may be due to the homogeneous distribution of active site energies on the surface of the PmPD/Fe₃O₄/o-MWCNTs magnetic nanocomposites, where no interaction has been observed between the adsorbed molecules and the surface of PmPD/Fe₃O₄/o-MWCNTs. Meanwhile, the R_L values in the range of 0–1, suggesting the suitability of adsorbent. In addition, the maximum adsorption capacity of adsorbent increased from 219.78 mg g⁻¹ to 346 mg g⁻¹ with the increase in temperature from 275 K to 335 K, which indicates the possibility that Cr⁶⁺ adsorption by PmPD/Fe₃O₄/o-MWCNTs is an endothermic process.

Table 2 Parameters of the Langmuir, Freundlich and Temkin isotherms for Cr⁶⁺ adsorption onto PmPD/Fe₃O₄/o-MWCNTs

T(K)	Langmuir isotherm				Freundlich isotherm			Temkin isotherms
	qm (mg g^−1)	b	r1^2	RL	n	KF	r2^2	Bt	KT	r3^2
273	219.8	0.0387	0.983	0.079–0.341	4.87	64.00	0.944	31.42	2.20	0.903
303	293.3	0.0484	0.978	0.064–0.292	5.12	3.59	0.943	38.40	1.96	0.884
333	346.0	0.0670	0.984	0.047–0.229	5.18	117.30	0.980	43.13	1.61	0.935

---

3.5. Effect of contact time and initial concentration

The rate at which dissolved heavy metal ions are removed from aqueous solution by solid sorbents is a significant factor for application in water quality control. Fig. 8 show the effect of adsorption time on the adsorption capacity at different initial solution concentrations. It is evident from these figures that the adsorption of Cr(VI) increases with increasing contact time. The adsorption equilibrium time of 300 mg g⁻¹ initial Cr⁶⁺ solution is slightly longer than the 100 mg L⁻¹ and 200 mg L⁻¹. The contact time was set to 80 min in each experiment to ensure each adsorption process can reach the equilibrium. The removal of Cr(VI) was found to be dependent on the initial concentration. The amount of Cr(VI) adsorbed, q_e (mg g⁻¹), increased with increase in initial concentration. Further, the adsorption was rapid in the early stages and then gradually decreased and became almost constant after the equilibrium point.

Fig. 8 Effect of contact time on chromium ions adsorbed onto PmPD/Fe₃O₄/o-MWCNTs at different initial solution concentrations.

3.6. Adsorption dynamics

It is essential to evaluate the adsorption kinetics as well as adsorption equilibrium using theoretical models in order to further investigate the adsorption of Cr(VI) by PmPD/Fe₃O₄/o-MWCNTs magnetic nanocomposites. The kinetic adsorption data were simulated with the pseudo-first-order equation and the pseudo-second-order rate equation, which are expressed as follows:^39–41where k₁ (min⁻¹), and k₂ (g mg⁻¹ min⁻¹) are the pseudo-first-order and pseudo-second-order rate adsorption constants, respectively. q_e and q_t are the amounts of Cr(VI) adsorbed (mg g⁻¹) at equilibrium and time t (min), respectively.

A straight line of log(q_e − q_t) versus t suggests the applicability of this kinetic model, q_e and k₁ can be determined from the intercept and slope of the plot (Fig. S6†). The plot t/q_t versus t should give a straight line if second-order kinetics are applicable and q_e and k₂ can be determined from the slope and intercept of the plot (Fig. S7†). Initial adsorption rates h (mg g⁻¹ min⁻¹) were also calculated from the data of pseudo-second-order kinetic model as follows:

h = k₂q_e² (8)

The kinetic parameters and correlation coefficients are listed in Table 3. Clearly, the correlation coefficients made by pseudo-first-order is a little smaller than that of pseudo-first-order kinetic mode, indicating that the pseudo-second-order kinetic mode give a good fit to the experimental kinetic data with a much higher correlation coefficient (r² > 0.99). Furthermore, the calculated adsorption capacities of pseudo-first-order differs much from the measured results, while the calculated adsorption capacities of pseudo-second-order kinetic mode are almost identical to the measured adsorption capacities. Thus experiment results supports the assumption behind the model that the rate limiting step in adsorption of heavy metals are chemisorption involving valence forces through the sharing or exchange of electrons between adsorbent and metal ions.

Table 3 Kinetic parameters for Cr⁶⁺ adsorption onto PmPD/Fe₃O₄/o-MWCNTs at various concentrations

C0 (mg L^−1)	Pseudo-first-order			Pseudo-second-order
	k1	qe	r1^2	k2	qe	h	r2^2
100	0.0659	82.6	0.96	0.0032	207	137	0.99
200	0.0706	140.0	0.98	0.0023	267	169	0.99
300	0.0499	150.0	0.96	0.0011	342	131	0.99

---

3.7. Intraparticle diffusion model (Waber–Morris model)

The overall reaction kinetics for the adsorption of Cr(VI) is a pseudo-second-order process. However, this could not highlight on the rate-limiting step. The mechanism of sorption is generally considered to involve three steps, one or any combination of which can be the rate-controlling mechanism: (i) mass transfer across the external boundary layer film of liquid surrounding the outside of the particle; (ii) sorption at a site on the surface (internal or external) and the energy will depend on the binding process (physical or chemical); this step is often assumed to be extremely rapid; (iii) diffusion of the adsorbate molecules to an sorption site either by a pore diffusion process through the liquid filled pores or by a solid surface diffusion mechanism.⁴² The probability of the intra-particle diffusion was explored by using the following equation suggested by Weber and Morries:⁴³

q_t = k_dt^1/2 + C (9)

where q_t is adsorption capacity at any time t and k_d is the intraparticle diffusion rate constant (mg g⁻¹ min^−1/2) and C is the film thickness. If the plot of q_t against the square root of time should be a straight line and pass through the origin, implying that the rate limiting step be the intraparticle diffusion. If the plot is non-linear or linear but does not pass through the origin, then film diffusion or chemical reaction controls the adsorption rate.⁴⁴

As seen from Fig. 9 the plots possess multi-linear portions, which indicated that two or more steps influence the sorption process. Fig. 9 shows that the straight line did not pass through the origin and this further indicated that the intraparticle diffusion was not only the rate controlling step. Meanwhile, the plots for the Weber–Morris kinetic model for PmPD/Fe₃O₄/o-MWCNTs at three different initial concentrations have three distinct regions. The slope of the linear portion indicated the rate of the sorption. The lower slope corresponded to a slower sorption process. At up to 10 min, the Cr⁶⁺ adsorption rate is relatively rapid is due to the diffusion in bulk phase to the exterior surface of adsorbent and the easily accessible sites, as well as the initial high Cr⁶⁺ concentration in the solution. A slow increase of Cr⁶⁺ adsorption is observed between 10 and 60 min attribute to the intraparticle diffusion of Cr⁶⁺ into mesopores. With the extension of time (60–300 min), the third linear portion indicate the adsorption–desorption equilibrium.

Fig. 9 The plot of intraparticle diffusion modelling of Cr⁶⁺ adsorbed onto PmPD/Fe₃O₄/o-MWCNTs with different initial concentrations.

3.8. Thermodynamic studies

Temperature dependence of the adsorption process is associated with several thermodynamic parameters, including Gibbs free energy change ΔG^o, enthalpy change ΔH^o and entropy change ΔS^o.

ΔG^o were used to decide whether the adsorption process is spontaneous or not. ΔG^o were calculated from the following equation:⁴⁵

ΔG^o = −RT ln k_c (10)

where k_c is the equilibrium constant, C_Ae is the solid phase concentration at equilibrium (mg L⁻¹), T is the temperature in Kelvin and R is the gas constant (8.314 J mol⁻¹ K⁻¹).

Relation between ΔG^o, ΔH^o and ΔS^o can be expressed by the following equation;

ΔH^o and ΔS^o were obtained from the slope and intercept of the plot of the Van't Hoff plot of k_c versus 1/T (Fig. 10). The calculated results are listed in Table 4, which indicates that Cr⁶⁺ adsorption on PmPD/Fe₃O₄/o-MWCNTs is a spontaneous process. However, the negative value of ΔG^o decreased with an increase in temperature, indicating that the spontaneous nature of adsorption of Cr⁶⁺ were inversely proportional to the temperature. Enhancement of adsorption capacity of adsorbent at higher temperatures may be attributed to the enlargement of pore size and/or activation of the adsorbent surface.⁴⁶ The positive value of ΔH^o confirmed the endothermic nature of adsorption which was also supported by the increase in value of Cr⁶⁺ uptake of the adsorbent with the rise in temperature. A possible explanation is that the metal ions are well hydrated and will require breaking of the hydration sheath to proceed for adsorption, which consequently requires high energy. Thus, high temperature favors the dehydration process and the adsorption phenomenon.⁴⁷ The positive value of ΔS^o shows the feasibility of adsorption and the increased randomness at the sorbent/solution interface during the adsorption of metal ions onto PmPD/Fe₃O₄/o-MWCNTs.

Fig. 10 Van't Hoff Plot for the adsorption of Cr⁶⁺ ions adsorption by PmPD/Fe₃O₄/o-MWCNTs nanoparticles.

Table 4 Thermodynamic parameters for Cr⁶⁺ adsorption onto PmPD/Fe₃O₄/o-MWCNTs

T(K)	ln kc	ΔG^o (kJ mol^−1)	ΔH^o (kJ mol^−1)	ΔS^o (J mol^−1 k^−1)
273	3.239	−7.350
303	3.842	−9.678	22.096	106.98
333	5.008	−13.864

---

3.9. Desorption studies

Regeneration of adsorbent for repeated reuse is of crucial importance for keeping the process costs down and recovering the metals extracted from the liquid phase. The system pH values has a great influence on the Cr⁶⁺ adsorption, therefore the reusability experiments were conducted at alkaline condition. Fig. 11 shows the recycling of PmPD/Fe₃O₄/o-MWCNTs magnetic nanocomposites for the removal of Cr⁶⁺. It is observed that the magnetic nanocomposites still possessed more than 52% adsorption capacities for Cr⁶⁺ after five cycles of reuse, reflecting a good reusability and high stability of the PmPD/Fe₃O₄/o-MWCNTs magnetic nanocomposites.

Fig. 11 The reusability of the PmPD/Fe₃O₄/o-MWCNTs nanocomposites for chromium ion adsorption.

3.10. Cr(VI) adsorption mechanisms by PmPD/Fe₃O₄/o-MWCNTs

As stated above, the initial pH of Cr(VI) solution greatly affects the Cr(VI) adsorption performance of PmPD/Fe₃O₄/o-MWCNTs. This indicates that the H⁺ played an important role during the adsorption. One of the reasons is that the Cr(VI) has powerful oxidizability, which makes it easy to be reduced to Cr(III) at low pH values.⁴³ The reaction is shown as follows:³

HCrO₄⁻ + 7H⁺ + 3e⁻ ↔ Cr³⁺ + 4H₂O (13)

Cr₂O₇²⁻ + 6e⁻ + 3H⁻ ↔ 2Cr³⁺ + 7H₂O (14)

According the previous literature, Fe₃O₄ was also a kind of adsorption group for Cr⁶⁺. The surface of Fe₃O₄ particle exhibit either positive or negative charge depends on pH values. Hence, Fe₃O₄ would attract negative Cr⁶⁺ species at low pH by electrostatic interactions.

The most important adsorption group is PmPD, which can greatly improve the adsorption capacity of PmPD/Fe₃O₄/o-MWCNTs. According to previous literature,⁴⁸ the benzenoid amine can be oxidized to quinoid imine by Cr(VI), which proves the redox reaction between PmPDs and Cr(VI). The –NH– of PmPD turns into –N = after Cr(VI) adsorption and Cr(III) may be chelated with imino groups of the adsorbents. Meanwhile, the appearance of –N⁺ = demonstrates the doping of Cr(VI) and chelation of Cr(III) during the adsorption.

3.11. Comparison of adsorbent performance with literature data

A comparison of the chromium adsorption capacity of PmPD/Fe₃O₄/o-MWCNTs with various adsorbents in Table 5. It shows that chromium adsorption capacity varied in wide range. The chromium removal capacities of other adsorbents were lower than that of PmPD/Fe₃O₄/o-MWCNTs.^2,36,49–56 Thus, the comparison suggests that PmPD/Fe₃O₄/o-MWCNTs have great potential as heavy metal ion adsorbents in wastewater treatment.

Table 5 Comparison of adsorption capacity of various adsorbents for the removal of Cr(VI)

Adsorbents	Adsorption capacity (mg g^−1)	Ref.
Magnetic particles	16.0	49
Graphene/MgAl-layered double hydroxide	177.6	50
Modified magnetic chitosan chelating resin	58.4	51
Haruna zeolite (HZ) modified with surfactant hexadecylpyridinium bromide	14.3	52
Sawdust	41.5	53
Nanocrystalline akaganeite	80.0	54
Polyacrylamide grafted sawdust	12.4	55
Activated carbon derived from acrylonitrile-divinylbenzene	81.5	36
Fe3O4 hollow microspheres/graphene oxide	32.3	2
Amine impregnated graphene oxide	232.5	56
PmPD/Fe3O4/o-MWCNTs	346.0	This study

---

4. Conclusions

In summary, the PmPD/Fe₃O₄/o-MWCNTs were successfully prepared by a straightforward method. The FTIR and XRD results of the samples demonstrated that PmPD and Fe₃O₄ had been successfully attached onto the MWCNTs.

The TEM showed the morphology of the PmPD/Fe₃O₄/o-MWCNTs. As far as we know, this is the first time that PmPD/Fe₃O₄/o-MWCNTs magnetic nanocomposites have been synthesized and used as an adsorbent for the removal of Cr⁶⁺. The Langmuir, Freundlich and Temkin adsorption models were applied to describe the equilibrium isotherms. The maximum adsorption capacity calculated from the Langmuir isotherm model of PmPD/Fe₃O₄/o-MWCNTs magnetic nanocomposites is 346 mg g⁻¹ at pH 2.0. Kinetic study showed that the pseudo-second-order model provided a very good fitting (r² > 0.99) for the kinetic data of the adsorption under all investigated concentrations. The thermodynamic parameters ΔH^o, ΔG^o and ΔS^o were calculated with temperature and pressure. The results show a feasible, spontaneous and exothermic adsorption process. The mechanism of adsorption includes both physical and chemical adsorptions. The adsorption–desorption cycle results showed that the adsorption capacity can remain up to 52% after five times of usage. Therefore, the PmPD/Fe₃O₄/o-MWCNTs have great potential to be used as efficient adsorbents to adsorb heavy metal from wastewater.

References

1. S. Basha, Z. Murthy and B. Jha, Chem. Eng. J., 2008, 137, 480–488.
2. M. C. Liu, T. Wen, X. L. Wu, C. L. Chen, J. Hu, J. Li and X. K. Wang, Dalton Trans., 2013, 42, 14710–14717.
3. D. Mohan and C. U. Pittman Jr, J. Hazard. Mater., 2006, 137, 762–811.
4. T. A. Kurniawan, G. Y. S. Chan, W.-H. Lo and S. Babel, Chem. Eng. J., 2006, 118, 83–98.
5. D. Zhou, L. Zhang, J. Zhou and S. Guo, Water Res., 2004, 38, 2643–2650.
6. S. W. Zhang, M. Y. Zeng, W. Q. Xu, J. X. Li, J. Li, J. Z. Xu and X. K. Wang, Dalton Trans., 2013, 42, 7854–7858.
7. G. X. Zhao, T. Wen, C. L. Chen and X. K. Wang, RSC Adv., 2012, 2, 9286–9303.
8. L. Fan, C. Luo, M. Sun and H. Qiu, J. Mater. Chem., 2012, 22, 24577.
9. S. Iijima, Nature, 1991, 354, 56–58.
10. C. Lu, H. Chiu and C. Liu, Ind. Eng. Chem. Res., 2006, 45, 2850–2855.
11. H. H. Cho, K. Wepasnick, B. A. Smith, F. K. Bangash, D. H. Fairbrother and W. P. Ball, Langmuir, 2010, 26, 967–981.
12. X. Ren, C. Chen, M. Nagatsu and X. Wang, Chem. Eng. J., 2011, 170, 395–410.
13. X. Tan, M. Fang, C. Chen, S. Yu and X. Wang, Carbon, 2008, 46, 1741–1750.
14. H. Chen, J. Li, D. Shao, X. M. Ren and X. K. Wang, Chem. Eng. J., 2012, 210, 475–481.
15. Q. Wang, X. K. Wang, Z. F. Chai and W. P. Hu, Chem. Soc. Rev., 2013, 42, 8821–8834.
16. S. Agnihotri, M. J. Rood and M. Rostam-Abadi, Carbon, 2005, 43, 2379.
17. P. Zong, S. Wang, Y. Zhao, H. Wang, H. Pan and C. He, Chem. Eng. J., 2013, 220, 45–52.
18. M. A. Salam, G. Al-Zhrani and S. A. Kosa, C. R. Chim., 2012, 15, 398–408.
19. E. Salehi, S. S. Madaeni, L. Rajabi, A. A. Derakhshan, S. Daraei and V. Vatanpour, Chem. Eng. J., 2013, 791–801.
20. M. Zhu and G. Diao, Nanoscale, 2011, 3, 2748–2767.
21. C. Cunha, S. Panseri, D. Iannazzo, A. Piperno, A. Pistone, M. Fazio, A. Russo, M. Marcacci and S. Galvagno, Nanotechnology, 2012, 23, 465102.
22. L. Ji, L. Zhou, X. Bai, Y. Shao, G. Zhao, Y. Qu, C. Wang and Y. Li, J. Mater. Chem., 2012, 22, 15853.
23. P. Xu, G. M. Zeng, D. L. Huang, C. L. Feng, S. Hu, M. H. Zhao, C. Lai, Z. Wei, C. Huang, G. X. Xie and Z. F. Liu, Sci. Total Environ., 2012, 424, 1–10.
24. V. K. Gupta, R. Kumar, A. Nayak, T. A. Saleh and M. A. Barakat, Adv. Colloid Interface Sci., 2013, 193–194, 24–34.
25. X. G. Li, M. R. Huang, W. Duan and Y. L. Yang, Chem. Rev., 2002, 102, 2925–3030.
26. L. M. Bronstein, X. Huang, J. Retrum, A. Schmucker, M. Pink, B. D. Stein and B. Dragnea, Chem. Mater., 2007, 19, 3624–3632.
27. J. Wan, W. Cai, J. Feng, X. Meng and E. Liu, J. Mater. Chem., 2007, 17, 1188.
28. L. Stobinski, B. Lesiak, L. Kövér, J. Tóth, S. Biniak, G. Trykowski and J. Judek, J. Alloys Compd., 2010, 501, 77–84.
29. L. Chai, L. Zhang, H. Wang, W. Yu and P. Sang, Mater. Lett., 2010, 64, 2302–2305.
30. W. Jiang, M. Pelaez, D. D. Dionysiou, M. H. Entezari, D. Tsoutsou and K. O'Shea, Chem. Eng. J., 2013, 222, 527–533.
31. S. K. Sahu, V. K. Verma, D. Bagchi, V. Kumar and B. D. Pandey, Indian J. Chem. Technol., 2008, 15, 397–402.
32. H. Qian, Y. Hu, Y. Liu, M. Zhou and C. Guo, Mater. Lett., 2012, 68, 174–177.
33. Y. Sun, Q. Yue, B. Gao, Y. Gao, Q. Li and Y. Wang, Chem. Eng. J., 2013, 217, 240–247.
34. J. Febrianto, A. N. Kosasih, J. Sunarso, Y. H. Ju, N. Indraswati and S. Ismadji, J. Hazard. Mater., 2009, 162, 616–645.
35. B. E. Reed and M. R. Matsumoto, Sep. Sci. Technol., 1993, 28, 2179–2195.
36. D. Duranoğlu, A. W. Trochimczuk and U. Beker, Chem. Eng. J., 2012, 187, 193–202.
37. X.-Z. Chu, Y.-J. Zhao, Y.-H. Kan, W.-G. Zhang, S.-Y. Zhou, Y.-P. Zhou and L. Zhou, Chem. Eng. J., 2009, 152, 428–433.
38. Y. Zhang, Y. Li, L. Q. Yang, X. J. Ma, L. Y. Wang and Z. F. Ye, J. Hazard. Mater., 2010, 178, 1046–1054.
39. J. Hu, C. Chen, X. Zhu and X. Wang, J. Hazard. Mater., 2009, 162, 1542–1550.
40. J. Wang, K. Pan, Q. He and B. Cao, J. Hazard. Mater., 2013, 244–245, 121–129.
41. Y. S. Ho and G. McKay, Chem. Eng. J., 1998, 70, 115–124.
42. W. H. Cheung, Y. S. Szeto and G. McKay, Bioresour. Technol., 2007, 98, 2897–2904.
43. C. S. Sundaram, N. Viswanathan and S. Meenakshi, J. Hazard. Mater., 2008, 155, 206–215.
44. N. Li, Z. Mei and X. Wei, Chem. Eng. J., 2012, 192, 138–145.
45. A. Sari, M. Tuzen, D. Citak and M. Soylak, J. Hazard. Mater., 2007, 149, 283–291.
46. R. Han, Z. Lu, W. Zou, W. Daotong, J. Shi and Y. Jiujun, J. Hazard. Mater., 2006, 137, 480–488.
47. V. K. Gupta, D. Pathania, S. Sharma and P. Singh, J. Colloid Interface Sci., 2013, 401, 125–132.
48. W. Yu, L. Zhang, H. Wang and L. Chai, J. Hazard. Mater., 2013, 260, 789–795.
49. Y. Li, S. Zhu, Q. Liu, Z. Chen, J. Gu, C. Zhu, T. Lu, D. Zhang and J. Ma, Water Res., 2013, 47, 4188–4197.
50. X. Yuan, Y. Wang, J. Wang, C. Zhou, Q. Tang and X. Rao, Chem. Eng. J., 2013, 221, 204–213.
51. Y. G. Abou El-Reash, M. Otto, I. M. Kenawy and A. M. Ouf, Int. J. Biol. Macromol., 2011, 49, 513–522.
52. Y. Zeng, H. Woo, G. Lee and J. Park, Microporous Mesoporous Mater., 2010, 130, 83–91.
53. S. Gupta and B. V. Babu, Chem. Eng. J., 2009, 150, 352–365.
54. N. K. Lazaridis, D. N. Bakoyannakis and E. A. Deliyanni, Chemosphere, 2005, 58, 65–73.
55. S. Deng and R. Bai, Water Res., 2004, 38, 2423–2431.
56. A. Santhana Krishna Kumar, S. Singh Kakan and N. Rajesh, Chem. Eng. J., 2013, 230, 328–337.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra10282g

---