Removal of chromium(VI) from wastewater using weakly and strongly basic magnetic adsorbents: adsorption/desorption property and mechanism comparative studies

Abstract

Two novel strongly basic magnetic adsorbents, quaternary ammonium-modified polystyrene and chitosan magnetic microspheres (Pst–MIMCl and CTS–GTMAC), were prepared using the in situ coprecipitation and emulsion cross-linking methods under mild conditions, with features of strong magnetic responsiveness and high quaternary ammonium group contents. The Cr(VI) adsorption/desorption properties and mechanisms of strongly and weakly basic magnetic adsorbents were compared through simulated wastewater. The strongly basic adsorbent exhibited low pH dependence, and the main adsorption mechanism was ion exchange. The weakly basic adsorbent exhibited high pH dependence, and the major adsorption mechanism was electrostatic attraction. Besides, the strongly basic adsorbent required higher desorption conditions than the weakly basic adsorbent owing to the difference of the desorption mechanisms. Furthermore, the removal selectivity of the strongly and weakly basic magnetic adsorbents was estimated by the chromium plating wastewater. The results demonstrated that the strongly basic magnetic adsorbents exhibited higher selectivity than the weakly basic magnetic adsorbents. In addition, the Pst–MIMCl was selected as the optimal magnetic adsorbent for Cr(VI) recovery from wastewater, with the advantages of strongly magnetic responsiveness, wide pH applicable range, high removal efficiency, high adsorption selectivity and good reusability.

---

1 Introduction

A large amount of Cr(VI)-containing wastewater exceeding the emission standard is a potential hazard to the environment and human health due to its extremely toxic and carcinogenic properties.^1,2 To eliminate this problem, several treatment methods including chemical precipitation, electrodialysis, membrane filtration and adsorption have been developed.^3,4 Among them, adsorption is recognized as one of the most effective methods for the removal of Cr(VI) from wastewater, especially from acidic wastewater (pH 1.0–7.0), with the advantages of simple technological equipment, easy operation, recovery of metal ions and reusability of the adsorbents.⁵ Besides, the recovered Cr(VI) can be reused in industrial processes, which meets the sustainable development strategy.

However, the traditional adsorption processes are not easy for continuous operation, and have many problems such as channeling, which brings on a low efficiency of production.^6–9 Magnetic separation technology has the advantages of convenient of quickly separation, easy for continuous process and high treatment efficiency, and it can effectively solve the defects of the traditional adsorption processes.^10,11 In our previous works, a new technology named gas-assisted magnetic separation was developed, and showed a great potential of continuous recovery of the Cd-containing wastewater and protein solution.^12,13

The foundation of magnetic separation technology is the preparation of magnetic separating materials. However, the reported magnetic adsorbents for Cr(VI) removal mainly focused on weakly basic materials,^14–17 which was modified with the primary, secondary or tertiary amines. The adsorption process was highly pH-dependent, with the optimal pH values of 2.0–3.0. Meanwhile, most of these adsorbents had low adsorption capacities. The ethylenediamine-modified magnetic polymer¹⁵ and chitosan/montmorillonite¹⁶ were used for the removal of Cr(VI), and the maximum adsorption capacities were only 32.15 and 35.71 mg g⁻¹ at the optimal pH values of around 2.0. These results indicated that the majority of the developed weakly basic magnetic adsorbents had a narrow pH applicable range, and were more suitable for Cr(VI) removal at low pH values.

To address this problem, magnetic adsorbents with high adsorption capacities, including polyethylenimine modified poly(glycidyl methacrylate),¹⁸ poly(vinyl alcohol)¹⁹ and chitosan magnetic microspheres,²⁰ and ethylenediamine-modified magnetic cellulose nanocomposite²¹ (PGMA–PEI, PVA–PEI, CTS–PEI and CE–EDA), were developed through the improvement of amino group contents in our previous works, and these adsorbents had high adsorption capacities in a wide pH range. The maximum adsorption capacities of PGMA–PEI magnetic microspheres at pH 2.0 and 7.0 were up to 492.61 and 90.5 mg g⁻¹, respectively.¹⁸ However, these magnetic adsorbents have the disadvantages of complicated modification route and expensive modification reagents requirement, which limited their practical application. An inspiring idea is to develop novel magnetic adsorbents with weak pH-dependence, which can keep stable adsorption capacities in a wide pH range. To realize this aim, the strongly basic magnetic adsorbents would be a good choice.^22–24 Therefore, to explore a strongly basic magnetic adsorbent with weak pH dependence and high adsorption capacity has an important application meaning. Furthermore, the comparative studies on adsorption/desorption properties and mechanisms of weakly and strongly basic magnetic adsorbents have a significant research meaning.

Nowadays, the research on Cr(VI) removal by magnetic adsorbents was mainly carried out in simulated wastewater, and there was still few report on the treatment of real wastewater. Electroplating effluent is one of the main sources of Cr(VI)-containing wastewater,^25,26 which can serve as the model to evaluate the removal efficiency and selectivity of weakly and strongly basic magnetic adsorbents in multiple ions coexisting system. In this study, two novel strongly basic magnetic adsorbents (Pst–MIMCl and CTS–GTMAC) were firstly prepared and characterized. The Cr(VI) comparative studies of the adsorption/desorption properties and mechanisms of these two strongly basic magnetic adsorbents and other four weakly basic magnetic adsorbents in our previous works were carried out through simulated and electroplating wastewaters in a bath mode.

2 Materials and methods

2.1 Materials

Merrifield resin (Pst–Cl, 2% crosslinked, 200–400 mesh) was purchased from Alfa Aesar (Ward Hill, MA, USA). Glycidyl trimethylammonium chloride (GTMAC) and polyethylenimine (PEI) were purchased from Sigma-Aldrich (St. Louis, MO, USA). All other chemicals were the products of Sinopharm Chemical Reagent Co., Ltd (Beijing, China) and were used as obtained. Potassium dichromate (K₂Cr₂O₇) was applied as the source of Cr(VI) simulation wastewater. The chrome plating wastewater (pH: 2.96) was kindly provided by Beijing Guotaihuateng Co., Ltd.

2.2 Preparation of the adsorbents

The N-methylimidazole chloride-modified polystyrene magnetic microspheres (Pst–MIMCl) were prepared using in situ coprecipitation method. The GTMAC-functionalized chitosan magnetic microspheres (CTS–GTMAC), with the magnetic silica nanoparticles as the magnetic cores, were prepared by the emulsion cross-linking method and modified by the reactions with EDA and GTMAC. The preparation routes of Pst–MIMCl and CTS–GTMAC are shown in Scheme 1, and the detailed procedures of preparation of the adsorbents were provided in the ESI.† The weakly basic magnetic adsorbents (PGMA–PEI, PVA–PEI, CTS–PEI and CE–EDA) were synthesized according to our previous works,^18–21 which were used to analyze the differences of the adsorption/desorption properties and mechanisms between the weakly and strongly basic magnetic adsorbents.

Scheme 1 Preparation routes of Pst–MIMCl (A) and CTS–GTMAC (B) magnetic microspheres.

2.3 Characterization

The morphologies of Pst–MIMCl and CTS–GTMAC magnetic microspheres were characterized by a scanning electron microscopy (SEM, JEOL JSM-6700F, Japan). Fourier transforms infrared spectra of preparation routes of Pst–MIMCl and CTS–GTMAC were recorded on a spectrophotometer (FT-IR, Bruker T27, Germany) between 4000 and 400 cm⁻¹. The magnetic properties of Pst–MIMCl and CTS–GTMAC were determined by a vibrating sample magnetometer (VSM, LakeShore 7307, USA) at room temperature. The MIMCl capacity of Pst–MIMCl was calculated by the N elemental analysis (Elementar Vario Macro cube, Germany). The amino group content of CTS–EDA and quaternary ammonium group content of CTS–GTMAC were estimated using the volumetric method and conductometric titration according to the previous work.²³ The surface zeta potentials of the PGMA–PEI and Pst–MIMCl at different pH values were obtained using a zeta potential analyzer (Zeta PALS, Brookhaven Instruments Co., USA). The XPS spectra of PGMA–PEI and Pst–MIMCl before and after Cr(VI) adsorption were obtained by a X-ray photoelectron spectrometer (XPS, Thermo ESCALAB 250Xi, USA).

2.4 Adsorption experiments

The effects of pH, initial Cr(VI) concentration, contact time on the adsorption of Cr(VI) onto Pst–MIMCl and CTS–GTMAC were tested in batch experiments to evaluate the adsorption properties. Briefly, 25 mg of adsorbents was added into 50 mL of Cr(VI) solution with known concentrations in 100 mL flasks, and the pH values of the system were adjusted by 2 mol L⁻¹ HCl or 2 mol L⁻¹ NaOH solution. The flasks were shook in a water bath shaker at 150 rpm and at 25 °C. In the desorption experiments, the 0.01–1 mol L⁻¹ NaOH solution and the 0.1–0.5 mol L⁻¹ NaOH + 0.1–0.5 mol L⁻¹ NaCl (1 : 1) solution were selected to desorb the Cr(VI)-loaded PGMA–PEI and Pst–MIMCl microspheres. In the adsorption experiments of chrome plating wastewater, 0–10 g L⁻¹ of the adsorbents were added, and the optimal usage amount was determined when the Cr(VI) concentration was reduced to meet the emission standard (0.5 mg L⁻¹). The inductively coupled plasma optimal emission spectrometer (ICP-OES, Perkin Elmer Optima 7000DV, USA) was used to measure the total Cr, Na⁺, Mg²⁺, Ca²⁺, Cu²⁺, Ni²⁺, Zn²⁺, Al³⁺, and Fe³⁺ concentrations in the simulated wastewater and electroplating wastewater. The analysis of Cr(VI) in electroplating wastewater was determined using a UV-vis spectrophotometer (Hitachi U-4100 Japan) at 540 nm⁻¹ after complexation with 1,5-diphenylcarbazide.^27,28 The Cr(III) concentration was estimated using the subtraction method. The adsorption capacity at time t (mg g⁻¹) was determined as follows:where C₀ (mg L⁻¹) and C_t (mg L⁻¹) are the Cr(VI) concentration at initially and time t (min), V (mL) is the volume of the medium and m (mg) is the adsorbent dose.

3 Results and discussion

3.1 Characterization of the adsorbents

The SEM images and magnetic hysteresis loops of Pst–MIMCl and CTS–GTMAC magnetic microspheres are shown in Fig. 1. Both Pst–MIMCl and CTS–GTMAC had standard spherical forms with narrow size distributions, and the average particle diameters were estimated as 70.6 and 223.2 μm, respectively. Besides, it was clear that no remanence, hysteresis and coercivity were observed from the magnetic hysteresis loops of the Pst–MIMCl and CTS–GTMAC, indicating the superparamagnetic properties of the adsorbents. The saturation magnetization values of Pst–MIMCl and CTS–GTMAC were determined as 12.5 and 10.5 emu g⁻¹, respectively. Both the Pst–MIMCl and CTS–GTMAC showed high magnetic responsiveness and could be effectively recovered from aqueous solution by an external magnet.

Fig. 1 SEM images and magnetization curves of Pst–MIMCl (A and C) and CTS–GTMAC (B and D) magnetic microspheres; the inset illustrates the magnetic separation of Cr(VI) on Pst–MIMCl and CTS–GTMAC from aqueous solution under the application of an external magnetic field and the time from (a) to (b) was less than 1 min.

In order to characterize the preparation route of the adsorbents, the FT-IR spectra of products at each stage of Pst–MIMCl and CTS–GTMAC were recorded. As shown in Fig. 2A, in the spectrum of Pst–Cl, the strong band at 665 cm⁻¹ corresponded to C–Cl vibration.²⁴ In contrast, in the Pst–MIMCl spectrum, the band at 665 cm⁻¹ disappeared, and a new band appeared at 1564 cm⁻¹, which was the characteristic of C–H bond on the methylimidazole.²⁴ This suggested that the MIMCl was successfully grafted onto the Pst microspheres. For magnetic Pst–MIMCl, the band of Fe–O vibration appeared at 579 cm⁻¹, indicating that the Fe₃O₄ nanoparticles were successfully fabricated into the pores of the microspheres.²¹ The FT-IR spectra of products at each stage of CTS–GTMAC are given in Fig. 2B. In the spectrum of the Fe₃O₄ nanoparticles, the band at 582 cm⁻¹ was assigned to the vibration of Fe–O bond.²¹ Next, two new bands at 800 and 1090 cm⁻¹ were observed in the Fe₃O₄@SiO₂, which were assigned to the Si–O–Fe and Si–O bonds.²¹ This suggested that the SiO₂ layer was successfully grafted onto the Fe₃O₄ surface through chemical bond. Furthermore, in the CTS spectrum, the N–H bond in chitosan appeared at 1575 cm⁻¹, and new bands of C–H bond appeared at 2923 and 2858 cm⁻¹, implying that the Fe₃O₄@SiO₂ nanoparticles were successfully coated with chitosan.²⁰ In addition, a small peak at 1725 cm⁻¹ indicated the existence of C=O bond on the free aldehydic groups. In the CTS–EDA spectrum, the C=O band at 1725 cm⁻¹ disappeared, suggesting that the free aldehydic groups were transformed to amino groups. Finally, the asymmetric angular bending of methyl groups of quaternary hydrogen appeared at 1480 cm⁻¹ in the spectrum of CTS–GTMAC,²³ indicating that the quaternary ammonium groups were grafted onto the microspheres through the ring-opening reaction of epoxy groups in GTMAC with EDA.

Fig. 2 FT-IR spectra of products at each stage of Pst–MIMCl (A) and CTS–GTMAC (B) magnetic microspheres.

In addition, the MIMCl content of Pst–MIMCl was determined as 2.06 ± 0.2 mmol g⁻¹. The amino group content of CTS–EDA was calculated as 4.6 ± 0.2 mmol g⁻¹, and the quaternary ammonium group content of CTS–GTMAC was 2.4 ± 0.1 mmol g⁻¹, which indicated that not all the amino groups was transformed to the quaternary ammonium groups. Besides, the Pst–MIMCl and CTS–GTMAC had a high acid resistance (ESI†).

3.2 Adsorption properties

The adsorption properties of Pst–MIMCl and CTS–GTMAC magnetic microspheres were analyzed in batch experiments, and the results are shown in Fig. 3.

Fig. 3 Adsorption properties of Pst–MIMCl (A, C and E) and CTS–GTMAC (B, D and F) magnetic microspheres (Pst–MIMCl: m = 0.05 g, V = 50 mL, T = 25 °C; CTS–GTMAC: m = 0.025 g, V = 50 mL, T = 25 °C).

3.2.1 Effect of the pH on Cr(VI) adsorption. Fig. 3A and B shows the effect of pH on the Cr(VI) adsorption onto the adsorbents. It was obvious that the Pst–MIMCl and CTS–GTMAC magnetic microspheres showed weak pH-dependence compared with the weakly basic magnetic adsorbents in the previous works.^18–21 The adsorbents had high adsorption capacities in the wide pH range of 1.0–7.0, with the adsorption capacities of 63.9 and 71.2 mg g⁻¹ at pH 7.0 and at 25 °C. The difference of the pH-dependence of weakly and strongly basic magnetic adsorbents was attributed to the difference of the adsorption mechanism. The optimized pH values of 3.0 and 2.5 for Pst–MIMCl and CTS–GTMAC were observed and selected for the following adsorption experiments.

3.2.2 Adsorption isotherms. Fig. 3C and D shows the adsorption isotherms of the Pst–MIMCl and CTS–GTMAC magnetic microspheres at 25 °C. It was found that the adsorption capacities of Pst–MIMCl and CTS–GTMAC quickly increased as the equilibrium Cr(VI) concentrations increased at the beginning, and reached a maximum at the equilibrium concentrations of about 50 and 150 mg L⁻¹, respectively. The experimental data was simulated using the Langmuir, Freundlich and Temkin models, which were expressed as follows:²⁹

q_e = K_fC_e^1/n (3)

where C_e (mg L⁻¹) and q_e (mg g⁻¹) are the Cr(VI) concentration and adsorption capacity at equilibrium, respectively. q_m (mg g⁻¹) is the maximum adsorption capacity and K is the Langmuir constant corresponded to adsorption energy. n and K_f ((mg g⁻¹)(L mg⁻¹)^1/n) are the Freundlich constants related to the intensity and capacity of the adsorption, respectively. a_T (L g⁻¹) and b_T (kJ mol⁻¹) are the Temkin adsorption constants corresponding to the heat of sorption and the maximum binding energy, respectively. The simulated parameters are summarized in Table 1. The results demonstrated that the data were fitted better with Langmuir model than both the Freundlich and the Temkin models based on the values of correlation coefficient (R²), indicating the monolayer coverage of Cr(VI) adsorbed onto the surface of the adsorbents. The maximum adsorption capacities of the Pst–MIMCl and CTS–GTMAC were calculated to be 104.0 and 233.1 mg g⁻¹, respectively.

Table 1 Simulating parameters for different isotherm models

Adsorbent	Langmuir model			Freundlich model			Temkin model
	qm	K	R^2	Kf	n	R^2	aT	bT	R^2
Pst–MIMCl	104.0	1.180	0.999	45.213	5.218	0.690	94.384	0.214	0.841
CTS–GTMAC	233.1	0.107	0.999	54.140	3.486	0.917	2.715	0.068	0.987

---

3.2.3 Adsorption kinetics. Fig. 3E and F shows the adsorption kinetic curves of the Pst–MIMCl and CTS–GTMAC magnetic adsorbents. Both the curves suggested that the adsorption rate was fast at the first 20 min, and the equilibrium was reached within 30 min and 40–120 min for the Pst–MIMCl and CTS–GTMAC, respectively. The results revealed that the developed strongly basic adsorbents had a quick adsorption rate. The pseudo-first-order and pseudo-second-order equations were used to analyze the kinetics of the adsorption process, which were expressed by eqn (5) and (6).³⁰where q_t and q_e (mg g⁻¹) are adsorption capacity at time t and at equilibrium, respectively. k₁ (min⁻¹) and k₂ (g mg⁻¹ min⁻¹) are rate constants of the pseudo-first-order and pseudo-second-order models, respectively. The simulated kinetic parameters are listed in Table 2. It was clear that the experimental data was fitted better to the pseudo-second-order model than the pseudo-first-order model according to the values of correlation coefficient and the closing degree between q_e,cal and q_e,exp, and this indicated that the adsorption rate of Cr(VI) onto the adsorbents might be controlled by chemical adsorption. The pseudo-second-order model was also applicable to the adsorption of Cr(VI) onto many adsorbents in the previous works.^14–17

Table 2 Kinetic parameters for the adsorption of Cr(VI) onto the PGMA–PEI₆₀₀ microspheres at different initial concentrations

Adsorbent	C0 (mg L^−1)	qe,exp (mg g^−1)	Pseudo-first-order			Pseudo-second-order
			k1	qe,cal (mg g^−1)	R^2	k2	qe,cal (mg g^−1)	R^2
Pst–MIMCl	50	40.2	0.027	2.66	0.521	0.017	40.7	0.999
	100	80.2	0.039	5.61	0.476	0.009	81.2	0.999
	150	99.1	0.041	5.04	0.554	0.012	99.9	0.999
CTS–GTMAC	50	90.8	0.012	5.6	0.126	0.0053	90.7	0.999
	150	182.2	0.037	76.1	0.857	0.0011	187.3	0.999
	250	220.8	0.022	38.9	0.683	0.0013	224.2	0.999

---

3.2.4 Comparative study of adsorption properties. The adsorption properties of the strongly basic magnetic adsorbents, including adsorption capacity, adsorption rate and suitable pH range, were compared with other magnetic adsorbents reported in the literatures. As shown in Table 3, it was found that the developed strongly basic magnetic adsorbents simultaneously had the advantages of high adsorption capacity, rapid adsorption rate and wide pH range. In addition, among the weakly basic magnetic adsorbents in our previous works, only the PGMA–PEI could achieve the pH range of 1.0–7.0. In contrast, although the adsorption capacities of Pst–MIMCl and CTS–GTMAC were much lower than that of PGMA–PEI, a wild pH range of 1.0–7.0 was achieved.

Table 3 Adsorption property comparison of various magnetic adsorbents for Cr(VI) removal

Adsorbents	qm (mg g^−1)	E.T. (min)	pH range^a	Ref.
a Adsorption capacity above 50 mg g^−1.
Poly(MMA–DVB–GMA)–EDA	61.35	60	2.0–3.0	15
Chitosan/montmorillonite	35.71	100	1.0–3.0	16
Magnetic chitosan beads	69.4	>60	3.0–5.0	17
Magnetic chitosan–EDA	51.813	10	2.0	29
γ-Fe2O3@δ-FeOOH	25.8	—	2.5	31
Fe3O4@PAA–DETA	11.24	—	5.0	32
NiFe2O4	30.0	—	5.0	33
Fe3O4–PEI–MMT	8.8	—	5.0	34
Fe3O4–Cyanex-301	30.8	120	2.0	35
ZnFe2O4–Ce^3+	57.24	4320	2.0	36
Fe3O4@polypyrrole	169.49	180	1.0–6.0	37
Poly(GMA–EGDMA)–PEI	137.7	120	1.0–6.0	38
Poly(MA–DVB)–EDA	∼36	60	3.0	39
Magnetic chitosan–CAGS	58.48	110	2.0	40
Magnetic chitosan nanoparticles	55.80	150	2.0–3.0	41
Cyclodextrin–chitosan/GO	67.66	—	1.0–3.0	42
MnFe2O4/chitosan nanoparticles	15.4	360	6.0	43
MnFe2O4/chitosan composites	51.79	480	5.0	44
Fe3O4-fungus@alginate-PAA	6.97	720	1.0	45
PGMA–PEI	492.6	10	1.0–7.0	18/previous work
PVA–PEI	88.4	8	1.0–3.0	19/previous work
CTS–PEI	236.4	60–120	1.0–6.0	20/previous work
CE–EDA	171.5	10	1.0–6.0	21/previous work
Pst–MIMCl	104.0	30	1.0–7.0	This work
CTS–GTMAC	233.1	40–120	1.0–7.0	This work

---

3.3 Comparative study of adsorption mechanisms

As mentioned above, there are great differences between the weakly and strongly basic magnetic adsorbents. The weakly basic magnetic adsorbents exhibited strong pH-dependence, while the strongly basic magnetic adsorbents showed weak pH-dependence. In order to clarify these reasons, the adsorption mechanisms of weakly and strongly basic magnetic adsorbents were studied by the analysis of the characterization results of zeta-potential, XPS and FT-IR. The PGMA–PEI and Pst–MIMCl served as the model adsorbents of the weakly and strongly basic magnetic adsorbents, respectively.

The pH value is an important factor that affects not only the ionic forms of the chromium species, but also the surface charge of the adsorbent. Cr(VI) can exist in five forms in an aqueous solution,⁴⁶ including H₂CrO₄, HCrO₄⁻, CrO₄²⁻, HCr₂O₇⁻ and Cr₂O₇²⁻. Among them, CrO₄²⁻ is the main species at pH above 6.0, while HCrO₄⁻ and Cr₂O₇²⁻ are the major components at pH values of 2.0–6.0, and the H₂CrO₄ is dominant at pH below 1.0. Fig. 4 shows the zeta potentials of PGMA–PEI and Pst–MIMCl in the pH range of 4–12. It was found that the zeta potential increased as the pH decreased. The isoelectric points (pI) of PGMA–PEI and Pst–MIMCl were determined as 9.9 and 9.0, respectively. This indicated that the PGMA–PEI and Pst–MIMCl was positively charged when pH < pI, and could adsorb the Cr(VI) ions through electrostatic attraction. For PGMA–PEI, the adsorption capacity increased as the pH decreased from 7.0 to 2.0, and then decreased as the pH decreased from 2.0 to 1.0 (Fig. S1†). It could be well explained by the fact that the protonation degree of the amino groups increased as the pH decreased from 7.0 to 2.0, which significantly improved the adsorption capacity of the PGMA–PEI. The decreasing adsorption capacity at pH < 2.0 was attributed to the conversion of chromium species, in which the neutral H₂CrO₄ was formed at pH < 2.0. These results revealed that the electrostatic attraction was the major adsorption mechanism of the PGMA–PEI. However, the pH-dependence of Pst–MIMCl was much lower than that of the PGMA–PEI, and the decreasing adsorption capacity of Pst–MIMCl at pH 1.0 and 7.0 can be explained by the conversion of chromium species. It could be concluded the electrostatic attraction played a less important role in the adsorption of Cr(VI) onto Pst–MIMCl.

Fig. 4 Zeta potentials of the PGMA–PEI (A) and Pst–MIMCl (B) at various pH values.

Fig. 5 shows the XPS spectra of fresh and Cr(VI)-loaded PGMA–PEI and Pst–MIMCl. From Fig. 5A and C, it was clear that the Cr 2p peaks appeared after the adsorption of Cr(VI), indicating that the Cr(VI) ions were successfully adsorbed by the adsorbents. From the XPS spectrum of Cr 2p (Fig. 5B and D), the main state of adsorbed Cr was determined as Cr(VI) (PGMA–PEI: Cr 2p_1/2, 587.80 eV, Cr 2p_3/2, 578.30 eV; Pst–MIMCl: Cr 2p_1/2, 588.40 eV, Cr 2p_3/2, 579.34 eV),^47,48 and this implied that the Cr(VI) was not remarkably reduced to Cr(III) owing to the rapid adsorption rate. For PGMA–PEI, in the XPS spectra of C 1s, N 1s and O 1s before and after Cr(VI) adsorption (Fig. S2†), the binding energies of –NH–, –NH₂, C=O, C–OH and C–O–C shifted to higher positions, which might be attributed to the protonation of these groups in acidic environment. For Pst–MIMCl, in the XPS spectrum of Cl 2p (Fig. 5E), it was obvious that the Cl content was significant decreased after Cr(VI) adsorption, suggesting that ion exchange interaction existed in the adsorption of Cr(VI). Meanwhile, in the XPS spectra of C 1s and N 1s (Fig. S3†), no shifts of the binding energy were observed. Based on the property of weak pH-dependence, the ion exchange should be the main adsorption mechanism of Pst–MIMCl.

Fig. 5 XPS spectra of PGMA–PEI (A and B) and Pst–MIMCl (C–E) magnetic microspheres before and after Cr(VI) adsorption.

Furthermore, in order to identify the direct interaction between the adsorbents and the Cr(VI) ions, the FT-IR spectra of PGMA–PEI and Pst–MIMCl before and after Cr(VI) adsorption were recorded (Fig. 6). For PGMA–PEI (Fig. 6A), in the spectrum of PGMA–PEI–Cr, the adsorption bands at 554, 761, 890 and 950 cm⁻¹ appeared, which were assigned to the Cr=O and Cr–O–Cr bonds.⁵ Meanwhile, the adsorption band of N–H at 1570 cm⁻¹ shifted to 1622 cm⁻¹ after adsorption, and then shifted back to 1570 cm⁻¹ when regenerated using alkali solution. The same blue shift (from 1575.7 to 1635.8 cm⁻¹) appeared after Cr(VI) adsorption onto tetraethylenepentamine modified PGMA magnetic microspheres in the previous work,⁴⁸ which was attributed to the formation of N → Cr(III) coordination bonds. The shifted adsorption peak disappeared after desorption by acid solution rather than alkali solution, which was explained by the fact that the N → Cr(III) coordination bonds could be only destroyed by acid solution. However, according to the XPS characterization results of PGMA–PEI and Pst–MIMCl, Cr(VI) was not significantly reduced to Cr(III), so the coordination interaction between Cr(III) and amino groups could be ignored. Therefore, it was conjectured that the hydrogen bonding interaction might exist in the adsorption of Cr(VI), which could be easily destroyed by alkali solution. In contrast, in the spectra of Pst–MIMCl before and after adsorption, no remarkable shift was found except the appearance of the bands of Cr=O and Cr–O–Cr. This result implied that there was no existence of hydrogen bonding interaction in the adsorption of Cr(VI), which could be attributed to the fact that the N-containing groups existed in the forms of tertiary amine group (–N<) and quaternary ammonium group (>N⁺<) on the adsorbent.

Fig. 6 FT-IR spectra of PGMA–PEI (A) and Pst–MIMCl (B) before and after Cr(VI) adsorption and desorbed by eluents.

As discussed above, the adsorption mechanism of weakly basic adsorbents was governed by electrostatic attraction coupled with hydrogen bonding interaction, and the electrostatic attraction was the main adsorption mechanism (Fig. S4†). The adsorption mechanism of strongly basic adsorbents should be electrostatic attraction-assisted ion exchange interaction, and the ion exchange was the major adsorption mechanism. Besides, because the CTS–GTMAC simultaneously had the features of weakly and strongly basic adsorbents, the main adsorption mechanisms could be determined as electrostatic attraction coupled with ion exchange (Fig. 7).

Fig. 7 Effect of the eluent concentration on the desorption of Cr(VI)-adsorbed PGMA–PEI (A) and Pst–MIMCl (B) microspheres.

3.4 Comparative study of desorption properties

In this study, the 0.01–1 mol L⁻¹ NaOH solution and the 0.1–0.5 mol L⁻¹ NaOH + 0.1–0.5 mol L⁻¹ NaCl (1 : 1) solution were selected to desorb the Cr(VI)-loaded PGMA–PEI and Pst–MIMCl microspheres. Fig. 8 shows the effects of eluent concentrations on desorption of the adsorbents. For PGMA–PEI, the desorption efficiency significantly improved with the increasing concentration of NaOH solution from 0 to 0.1 mol L⁻¹, and then retained constant in the NaOH concentration range of 0.1–1 mol L⁻¹, with the desorption efficiency of 99.1% by 0.1 mol L⁻¹ NaOH solution. For Pst–MIMCl, the eluting effect by the mixture of NaOH + NaCl was better than those by the NaOH solution or the NaCl solution, which was consistence with the previous work.²⁴ The desorption efficiency of 99.2% was achieved using the mixture of 0.3 mol L⁻¹ NaOH + 0.3 mol L⁻¹ NaCl. Therefore, it could be concluded that the strongly basic adsorbent required higher desorption conditions than the weakly basic adsorbent. The difference of desorption conditions of weakly and strongly basic magnetic adsorbents might be explained by the difference of desorption mechanisms. For PGMA–PEI, the NaOH addition was favorable for the deprotonation of the amino groups and conversion of chromium species (eqn (7)). For Pst–MIMCl, the NaOH addition was used for the conversion of chromium species, and the NaCl addition was used for the ion exchange of CrO₄²⁻ (eqn (8) and (9)). In addition, the Cr(VI)-loaded CTS–GTMAC could also be effectively eluted by 0.3 mol L⁻¹ NaOH + 0.3 mol L⁻¹ NaCl, with the desorption efficiency of 95.6%.

–NH₃⁺⋯HCrO₄⁻ + OH⁻ → –NH₂ + CrO₄²⁻ (7)

2(–N⁺(CH₃)₃⋯HCrO₄⁻) + 2OH⁻ → –2N⁺(CH₃)₃⋯CrO₄²⁻ + CrO₄²⁻ (8)

–2N⁺(CH₃)₃⋯CrO₄²⁻ + 2Cl⁻ → 2 (–N⁺(CH₃)₃⋯Cl⁻) + CrO₄²⁻ (9)

Fig. 8 Removal properties of the developed magnetic materials for chrome plating wastewater.

3.5 Adsorption of electroplating wastewater

In this study, the removal property and selectivity of the strongly and weakly basic magnetic adsorbents were estimated by the electroplating wastewater, and the optimal magnetic adsorbent was selected based on the removal efficiency, adsorbent dose, adsorption times, synthesis cost and ion selectivity.

3.5.1 Removal properties. First of all, the effect of adsorbent dose on Cr(VI) removal from electroplating wastewater was performed at 25 °C. As shown in Fig. 8, it was clear that the Cr(VI) concentration decreased as the adsorbent dose improved, and the higher adsorption capacity of the adsorbent had, the less adsorbent dose of the first time adsorption needed. The relative parameters are summarized in Table 4. The strongly basic magnetic adsorbents including Pst–MIMCl and CTS–GTMAC required only once adsorption to reduce the Cr(VI) concentration to the emission standard (0.5 mg L⁻¹). However, the weakly basic magnetic adsorbents including PGMA–PEI, PVA–PEI, CTS–PEI and CE–EDA needed 3, 2, 2, 4 times adsorption, respectively. These results might be attributed to the high hydrophilicity of the amino groups (–NH₂, –NH– and –N<) on the weakly basic adsorbents. When the Cr(VI) concentration was reduced to the minimum, the continuous adding of adsorbent could not continuously reduce the Cr(VI) concentration, owing to the fact that the adsorbent adsorbed more water (protonation of the amino groups) than Cr(VI) ions. In addition, based on the properties of removal efficiency, adsorbent dose, adsorption times and synthesis cost, the Pst–MIMCl exhibited the highest adsorption properties. When the adsorbent dose was 7.0 g L⁻¹, the removal efficiency of the first time adsorption was up to 99.7%, and the Cr(VI) concentration was reduced to 0.26 mg L⁻¹ (<0.5 mg L⁻¹).

Table 4 Adsorption comparison of the developed magnetic materials

Adsorbent		Removal efficiency (%)	Adsorbent dose^a (g L^−1)	Adsorption times	Total adsorbent dose^b (g L^−1)	Cost (RMB/t)
a Adsorbent dose of the first time adsorption.b Adsorbent dose when the Cr(VI) concentration was reduced to the emission standard (0.5 mg L^−1).
Weakly basic adsorbents	PGMA–PEI	91.3	2	3	6	∼6.3
	PVA–PEI	97.5	8	2	12	∼1.7
	CTS–PEI	98.6	6	2	8	∼5.1
	CE–EDA	82.9	8	4	18	∼2.4
Strongly basic adsorbents	Pst–MIMCl	99.7	7	1	7	∼1.6
	CTS–GTMAC	99.5	8	1	8	∼4.9

---

3.5.2 Selectivity adsorption and recovery of Cr(VI). Table 5 lists the component contents of the first time adsorption solution. It was obvious that the strongly basic magnetic adsorbents exhibited higher selectivity than those of the weakly basic magnetic adsorbents. Especially, the Pst–MIMCl exhibited the highest ion selectivity, and only a small amount of Fe³⁺ could be adsorbed on the adsorbent by the coordination interaction. However, for the weakly basic magnetic adsorbents, the complex adsorption of Cu²⁺, Fe³⁺, Ni²⁺ and Cr³⁺ was significant due to the high content of primary, secondary and tertiary amines. In addition, the 0.2 mol L⁻¹ NaOH solution and 0.3 mol L⁻¹ NaOH + 0.3 mol L⁻¹ NaCl solution were used to desorb the Cr(VI)-loaded weakly and strongly basic magnetic adsorbents, and the component contents of the regenerated solution are provided in Table 6. It was found that the adsorbed Cu²⁺, Fe³⁺, Ni²⁺ and Cr³⁺ ions on the weakly magnetic adsorbents and CTS–GTMAC could not be desorbed by the eluents and precipitated on the adsorbents. Therefore, the Pst–MIMCl was still the optimal selection for the removal of Cr(VI) according to the ion selectivity. Besides, when the eluent volume was one tenth of the wastewater volume, the enrichment factor for Pst–MIMCl was up to 9.40, indicating that the enriching recovery of Cr(VI) was achieved through the regeneration of the adsorbent.

Table 5 Component contents of the first time adsorption solution

Adsorbent	Cr(VI) (mg L^−1)	Na^+ (mg L^−1)	Mg^2+ (mg L^−1)	Ca^2+ (mg L^−1)	Cu^2+ (mg L^−1)	Ni^2+ (mg L^−1)	Cr^3+ (mg L^−1)	Fe^3+ (mg L^−1)
Stock	89.30 ± 1.66	78.27 ± 1.03	19.79 ± 0.83	58.15 ± 0.21	33.14 ± 0.35	11.23 ± 0.18	81.60 ± 1.66	23.12 ± 0.22
PGMA–PEI	7.77 ± 0.44	75.28 ± 1.27	18.55 ± 0.20	56.59 ± 0.17	0.64 ± 0.06	0.96 ± 0.05	2.04 ± 0.19	0.59 ± 0.04
PVA–PEI	2.21 ± 0.32	75.69 ± 1.42	18.78 ± 0.25	56.80 ± 0.24	2.64 ± 0.43	6.86 ± 0.23	7.28 ± 0.09	0.59 ± 0.02
CTS–PEI	1.22 ± 0.18	73.68 ± 1.07	18.93 ± 0.38	56.48 ± 0.46	1.34 ± 0.39	6.54 ± 0.19	3.61 ± 0.30	0.59 ± 0.03
CE–EDA	15.29 ± 0.46	78.11 ± 1.59	18.61 ± 0.32	56.82 ± 0.31	6.55 ± 0.15	1.29 ± 0.07	4.92 ± 0.44	0.60 ± 0.02
Pst–MIMCl	0.26 ± 0.05	76.64 ± 1.21	19.46 ± 0.43	58.06 ± 0.53	32.93 ± 0.19	11.02 ± 0.10	78.54 ± 0.91	7.74 ± 0.14
CTS–GTMAC	0.41 ± 0.08	74.92 ± 0.99	19.51 ± 0.51	57.41 ± 0.37	10.27 ± 0.68	11.30 ± 0.08	28.36 ± 0.81	3.35 ± 0.04

---

Table 6 Component contents of the regenerated solution

Adsorbent	Cr(VI) (mg L^−1)	Cu^2+ (mg L^−1)	Ni^2+ (mg L^−1)	Cr^3+ (mg L^−1)	Fe^3+ (mg L^−1)	Enrichment factor
PGMA–PEI	783.50 ± 8.11	7.46 ± 0.79	2.43 ± 0.17	142.91 ± 1.56	2.43 ± 0.04	9.61 ± 0.15
PVA–PEI	831.74 ± 8.32	5.23 ± 0.45	4.94 ± 0.23	128.76 ± 1.88	2.65 ± 0.07	9.31 ± 0.09
CTS–PEI	801.57 ± 7.85	10.25 ± 0.96	5.92 ± 0.16	103.23 ± 1.05	2.75 ± 0.09	9.00 ± 0.08
CE–EDA	701.61 ± 7.23	9.87 ± 0.65	3.48 ± 0.31	115.37 ± 1.35	2.89 ± 0.10	9.48 ± 0.07
PSt–MIMCl	839.70 ± 9.74	2.95 ± 0.33	2.21 ± 0.25	20.70 ± 1.96	2.92 ± 0.13	9.40 ± 0.11
CTS–GTMAC	833.84 ± 5.55	15.35 ± 0.79	9.85 ± 0.41	95.86 ± 3.75	6.95 ± 0.05	9.34 ± 0.06

---

3.5.3 Reusability. The reusability of the Pst–MIMCl was also considered in the application of Cr(VI) removal from chrome plating wastewater. In this study, the Cr(VI)-loaded adsorbent was desorbed using a 0.3 mol L⁻¹ NaOH + 0.3 mol L⁻¹ NaCl solution, and then washed with distilled water to the pH of 7.0 for reuse in the next run. As shown in Fig. 9, no remarkable removal efficiency loss (<4%) was observed after ten adsorption–desorption cycles, and the residual Cr(VI) concentration increased slightly (<3.4 mg L⁻¹). This result indicated that the Pst–MIMCl could be easily desorbed and reused in the removal of Cr(VI) from electroplating wastewater.

Fig. 9 Removal efficiency of Pst–MIMCl and residual Cr(VI) concentrations during ten adsorption–desorption cycles (adsorption: m = 7 g L⁻¹; V = 50 mL; t = 30 min and T = 25 °C. Desorption: eluent, 0.3 mol L⁻¹ NaOH + 0.3 mol L⁻¹ NaCl solution; V = 5 mL; t = 30 min and T = 25 °C).

4 Conclusions

In this study, two novel strongly basic magnetic adsorbents (Pst–MIMCl and CTS–GTMAC) were firstly prepared and well characterized. The Cr(VI) adsorption/desorption properties of these two strongly basic magnetic adsorbents and other four weakly basic magnetic adsorbents were evaluated through simulated wastewater. The weakly basic magnetic adsorbent had high pH-dependence, and its main adsorption mechanism was electrostatic attraction. The strongly basic magnetic adsorbent exhibited weak pH-dependence, and ion exchange was the dominant adsorption mechanism. Furthermore, the strongly basic magnetic adsorbents exhibited higher selectivity than the weakly basic magnetic adsorbent. In addition, the Pst–MIMCl was selected as the optimal magnetic adsorbent for Cr(VI) recovery from wastewater, with the advantages of strongly magnetic responsiveness, wide pH applicable range, high removal efficiency, high adsorption selectivity and good reusability.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 21106162), the National Key Natural Science Foundation of China (No. 21136009), the Major Project of National Natural Science Foundation of China (No. 51090382), the General Research Project of Liaoning Education Department (L2015045), and the Youth foundation of Dalian Polytechnic University (67007908).

References

1. B. Chen, X. Zhao, Y. Liu, B. Xu and X. Pan, RSC Adv., 2015, 5, 1398–1405.
2. R. V. Kumar, A. K. Basumatary, A. K. Ghoshal and G. Pugazhenthi, RSC Adv., 2015, 5, 6246–6254.
3. Z. Song, W. Li, W. Liu, Y. Yang, N. Wang, H. Wang and H. Gao, RSC Adv., 2015, 5, 13028–13035.
4. L. Deng, Z. Shi and X. Peng, RSC Adv., 2015, 5, 49791–49801.
5. X. Sun, L. Yang, H. Xing, J. Zhao, X. Li, Y. Huang and H. Liu, Colloids Surf., A, 2014, 457, 160–168.
6. Q. Li, X. Tang, Y. Sun, Y. Wang, Y. Long, J. Jiang and H. Xu, RSC Adv., 2015, 5, 25337–25347.
7. H. Eskandarloo, A. Badiei, M. A. Behnajady and G. M. Ziarani, RSC Adv., 2014, 4, 28587–28596.
8. F. E. Soetaredjo, A. Kurniawan, L. K. Ong, D. R. Widagdyo and S. Ismadji, RSC Adv., 2014, 4, 52856–52870.
9. S. Muthusamy and S. Venkatachalam, RSC Adv., 2015, 5, 45817–45826.
10. H. Gao, S. Lv, J. Dou, M. Kong, D. Dai, C. Si and G. Liu, RSC Adv., 2015, 5, 60033–60040.
11. A. M. Donia, A. A. Atia and F. I. Abouzayed, Chem. Eng. J., 2012, 191, 22–30.
12. T. Dong, L. Yang, M. Zhu, Z. Liu, X. Sun, J. Yu and H. Liu, Chem. Eng. J., 2015, 280, 426–432.
13. W. Li, L. Yang, F. Wang, H. Zhou, H. Xing, X. Li and H. Liu, Ind. Eng. Chem. Res., 2013, 52, 4290–4296.
14. Y. Pang, G. Zeng, L. Tang, Y. Zhang, Y. Liu, X. Lei, Z. Li, J. Zhang, Z. Liu and Y. Xiong, Chem. Eng. J., 2011, 175, 222–227.
15. Y. G. Zhao, H. Y. Shen, S. D. Pan and M. Q. Hu, J. Hazard. Mater., 2010, 182, 295–302.
16. D. Chen, W. Li, Y. Wu, Q. Zhu, Z. Lu and G. Du, Chem. Eng. J., 2013, 221, 8–15.
17. G. Huang, H. Zhang, J. X. Shi and T. A. G. Langrish, Ind. Eng. Chem. Res., 2009, 48, 2646–2651.
18. X. Sun, L. R. Yang, H. Xing, J. Zhao, X. Li, Y. Huang and H. Liu, Chem. Eng. J., 2013, 234, 338–345.
19. X. Sun, L. Yang, Q. Li, Z. Liu, T. Dong and H. Liu, Chem. Eng. J., 2015, 262, 101–108.
20. X. Sun, L. Yang, T. Dong, Z. Liu and H. Liu, J. Appl. Polym. Sci., 2016, 3 DOI:10.1002/app.43078.
21. X. Sun, L. Yang, Q. Li, J. Zhao, X. Li, X. Wang and H. Liu, Chem. Eng. J., 2014, 241, 175–183.
22. S. Chen, Q. Yue, B. Gao, Q. Li and X. Xu, Chem. Eng. J., 2011, 168, 909–917.
23. K. Z. Elwakeel, Desalination, 2010, 250, 105–112.
24. L. Zhu, Y. Liu and J. Chen, Ind. Eng. Chem. Res., 2009, 48, 3261–3267.
25. V. Sarin, T. S. Singh and K. K. Pant, Bioresour. Technol., 2006, 97, 1986–1993.
26. Y. Xing, X. Chen and D. Wang, Environ. Sci. Technol., 2007, 41, 1439–1443.
27. L. Xu, M. Luo, L. Yang, X. Wei, X. Lin and H. Liu, J. Ind. Microbiol. Biotechnol., 2011, 38, 1709–1718.
28. P. Karthik, R. Vinoth, S. G. Babu, M. Wen, T. Kamegawa, H. Yamashita and B. Neppolian, RSC Adv., 2015, 5, 39752–39759.
29. X. J. Hu, J. S. Wang, Y. G. Liu, X. Li, G. M. Zeng, Z. L. Bao, X. X. Zeng, A. W. Chen and F. Long, J. Hazard. Mater., 2011, 185, 306–314.
30. H. Y. Zhu, Y. Q. Fu, R. Jiang, J. Yao, L. Xiao and G. M. Zeng, Bioresour. Technol., 2012, 105, 24–30.
31. J. Hu, I. M. C. Lo and G. H. Chen, Sep. Purif. Technol., 2007, 58, 76–82.
32. S. H. Huang and D. H. Chen, J. Hazard. Mater., 2009, 163, 174–179.
33. L. Wei, G. Yang, R. Wang and W. Ma, J. Hazard. Mater., 2009, 164, 1159–1163.
34. I. Larraza, M. López-Gónzalez, T. Corrales and G. Marcelo, J. Colloid Interface Sci., 2012, 385, 24–33.
35. T. Burksa, A. Uheidaa, M. Saleemia, M. Eitaa, M. S. Topraka and M. Muhammeda, Sep. Sci. Technol., 2013, 48, 1243–1251.
36. S. Kuai, Z. Zhang and Z. Nan, J. Hazard. Mater., 2013, 250–251, 229–237.
37. M. Bhaumik, A. Maity, V. V. Srinivasu and M. S. Onyango, J. Hazard. Mater., 2011, 190, 381–390.
38. G. Bayramoglu and M. Y. Arica, Chem. Eng. J., 2008, 139, 20–28.
39. Q. Wang, Y. Guan, X. Liu, M. Yang and X. Ren, Chin. J. Chem. Eng., 2012, 20, 105–110.
40. Y. G. Abou El-Reash, M. Otto, I. M. Kenawy and A. M. Ouf, Int. J. Biol. Macromol., 2011, 49, 513–522.
41. N. N. Thinh, P. T. Hanh, T. T. Ha le, N. Anh le, T. V. Hoang, V. D. Hoang, H. Dang le, N. V. Khoi and T. D. Lam, Mater. Sci. Eng., C, 2013, 33, 1214–1218.
42. L. Li, L. Fan, M. Sun, H. Qiu, X. Li, H. Duan and C. Luo, Colloids Surf., B, 2013, 107, 76–83.
43. Y. Xiao, H. Liang, W. Chen and Z. Wang, Appl. Surf. Sci., 2013, 285, 498–504.
44. Y. Xiao, H. Liang and Z. Wang, Mater. Res. Bull., 2013, 48, 3910–3915.
45. H. Li, Z. Li, T. Liu, X. Xiao, Z. Peng and L. Deng, Bioresour. Technol., 2008, 99, 6271–6279.
46. B. Saha and C. Orvig, Coord. Chem. Rev., 2010, 254, 2959–2972.
47. S. Deng and Y. P. Ting, Environ. Sci. Technol., 2005, 39, 8490–8496.
48. H. Shen, J. Chen, H. Dai, L. Wang, M. Hu and Q. Xia, Ind. Eng. Chem. Res., 2013, 52, 12723–12732.

---

Footnotes

† Electronic supplementary information (ESI) available: Detailed experimental procedures for the preparation of the adsorbents, the effect of pH on Cr(VI) adsorption onto PGMA–PEI, the XPS characterization and the schematic diagram of adsorption mechanisms. See DOI: 10.1039/c5ra27028f

‡ Xitong Sun and Qian Li contributed equally to this work.

---