Removal of phosphate and chromium(VI) from liquids by an amine-crosslinked nano-Fe₃O₄ biosorbent derived from corn straw

Abstract

A magnetic biocomposite based bio-sorbent (corn straw) was prepared after in situ co-precipitation with Fe²⁺ and Fe³⁺ solutions and amine functionalization. The characteristics of amine cross-linked magnetic corn straw (ACMCS), spent samples after phosphate and Cr(VI) adsorption, and regeneration samples were evaluated by VSM, TEM, SEM, FTIR, XRD and XPS. Adsorption tests of phosphate and Cr(VI) were also conducted. The TEM images in accord with the VSM results showed that a high number of Fe₃O₄ particles had been introduced onto the skeleton of corn straw. The XPS results indicated the existence of interactions between amine groups and phosphate/Cr(VI) and low losses of Fe and N in the regenerated adsorbent. Kinetic models showed the rate-limiting step to be chemisorption involving valence forces through sharing or exchange of electrons between adsorbate and adsorbent. After three adsorption–desorption cycles (HCl and NaOH as eluents), there was only a slight loss in adsorption capacity of Cr(VI) and phosphate; this demonstrated that the magnetic biocomposite was renewable in a continuous adsorption process.

---

1. Introduction

Currently, water pollution is a serious environmental problem that has received considerable attention throughout the world. A large amount of chromate waste water is discharged into the water environment from different industries, such as dye factories, electroplate factories, steelwork, and chemical plants.^1,2 Hexavalent chromium [Cr(VI)] is considered to be a top-priority toxic pollutant by the US Environmental Protection Agency; it has high water solubility and is also 500 times more toxic, mutagenic and carcinogenic than Cr(III).^3,4 Several species of Cr(VI) exist under different pH conditions. At pH > 7, chromate (CrO₄²⁻) will more readily exist in solution; at pH values between 1 and 6, hydrogen chromate (HCrO₄⁻) will preferentially exist. Therefore, in natural solutions, Cr(VI) consists of these expected species, including CrO₄²⁻, HCrO₄⁻, and dichromate Cr₂O₇²⁻, which are quite soluble and movable in water streams.⁵ Phosphate is a normal nutrient for botany; however, when additional phosphate is discharged in nearby static water bodies, an excessive growth of photosynthetic aquatic micro- and macro-organisms will be encouraged in these water bodies. As a result, concentrations of phosphate exceeding the desired limit are a major cause of the eutrophication of natural water bodies.^6–8 Therefore, the removal of Cr(VI) and phosphate from wastewater before discharge into environmental water is required.

Many technologies have been employed for the uptake of Cr(VI) and phosphate from aqueous solutions, such as chemical precipitation, reverse osmosis, electrodialysis, contact filtration, adsorption and advanced biological methods.^9–11 Among these, adsorption has proved to be a convenient and effective method due to its low cost, flexibility of design, easy operation, and insensitivity to biological materials in aqueous environments.⁹ Some interesting studies on the use of Fe₃O₄ nanoparticles in the removal of heavy metals provided our group with new ideas for convenient and time-saving methods to address the issue of the separation of materials from the treated water.^10–12 Moreover, in view of the increasing importance of “green thinking” in process industries as well as environmental protection, low cost and the availability of natural products, the use of agricultural by-products as adsorption carriers in wastewater treatment is a viable option.^13,14 Therefore, to explore a magnetic bio-sorbent derived from agricultural by-products with high phosphate and Cr(VI) adsorption capacities has practical significance. Based on the overview of all magnetic adsorbents, we found that there were still few reports related to the use of agricultural by-products-based magnetic bio-sorbents for anion removal.

In this study, our group used corn straw as a raw biomaterial to synthesize magnetic absorbents. The first stage of preparation was the in situ co-precipitation of Fe₃O₄ on the surface of the corn straw using Fe²⁺ and Fe³⁺ solutions.^15,16 This was then followed by amine cross-linking through reactions with epichlorohydrin, triethylenetetramine and triethylamine, forming amine crosslinked magnetic corn straw (ACMCS).^15,16 The resulting clean material, phosphate/Cr(VI) laden samples and regenerated samples were characterized by vibrating sample magnetometry (VSM), transmission electron microscopy (TEM), scanning electron microscopy (SEM), Fourier transform infrared spectroscopy (FTIR), X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS). Subsequently, the ACMCS was employed for the removal of Cr(VI) and phosphate from solution, and the respective adsorption properties of these species were also determined as a function of pH, concentration, contact time, temperature and regeneration conditions.

2. Methods and materials

2.1. Materials and reagents

The corn straw was obtained from Liao Cheng, Shandong Province, China. It was first washed with water and then dried at 60 °C in an air dry oven. Finally, it was crushed and sieved to 300 to 400 μm before use. All reagents used in the tests, including FeSO₄, FeCl₃, NaH₂PO₄, K₂Cr₂O₇, epichlorohydrin, N,N-dimethylformamide, triethylenetetramine and triethylamine, were bought from Sinopharm Group Co. Ltd and were all analytically pure.

2.2. Preparation of ACMCS

The preparation of amine functionalized biosorbent was conducted by a two-step reaction that was described in our previous work but with some modifications, e.g. corn stalk as the starting material and triethylenetetramine as the curing agent.^17,18 The Fe₃O₄ in situ co-precipitation was first conducted under oxygen-free conditions, followed by the amine functionalization process (Fig. 1).

Fig. 1 Preparation process for amine crosslinked magnetic corn straw.

In the first process, the reaction was conducted in a three-necked flask (1000 mL) at 70 °C. Foremost, the reaction system was maintained under oxygen-free conditions by purging with N₂. First, the iron solution was prepared by dissolving 0.3125 mol of FeSO₄ and 0.0625 mol of FeCl₃ in 250 mL of deionized water. Subsequently, 4 g of corn straw was placed in the iron solution (240 mL), and then 25 mL of NH₃·H₂O (25%) was added to the three-necked flask. The reaction was maintained for 4 h. After the reaction, the Fe₃O₄ loaded corn straw was washed with deionized water and dried in a vacuum oven at 104 °C for 2 h.

In the subsequent process, the reaction was conducted in a three-necked flask (1000 mL) at the temperature of stage change. In the first stage, 5.4 g of Fe₃O₄ loaded corn straw (magnetic corn straw) was suspended in 15 mL of epichlorohydrin and 10 mL of N,N-dimethylformamide in a flask with stirring for 1 h at 85 °C. In the second stage, triethylenetetramine (6 mL) was added dropwise to the flask and stirred for 30 min at 75 °C. Finally, 10 mL of trimethylamine was added to the mixture, which was stirred for 2 h at 85 °C. The product was separated from the solution by vacuum filtration and dried for 12 h at 80 °C in an air dry oven. About 15.4 g of ACMCS was obtained by using 4 g of corn straw. The adsorbent (ACMCS) was gathered and stored.

The use of N,N-dimethylformamide enhanced the susceptibility of the epoxide ring in epichlorohydrin to attack the hydroxyl groups in cellulose. Quaternary amine groups originating from trimethylamine were then introduced onto the epoxypropyl by-products using triethylenetetramine as a curing agent.

2.3. Characteristics of all magnetic samples

The magnetic properties of the adsorbent (ACMCS), Cr(VI) laden ACMCS, and phosphate laden ACMCS were measured by vibrating sample magnetometry (VSM) using a magnetometer (LDJ9500) at normal atmospheric temperature.

Scanning electron microscopy (SEM) images and transmission electron microscopy (TEM) of the non-adsorption sample and the saturated adsorption sample were obtained with NoVa™ Nano SEM 250 and FEI Tecnai G20 instruments, respectively.

The grafted amine groups were evaluated according to the nitrogen contents in ACMCS and virgin corn stalk. These were measured with an element analyzer (Elementar Vario EL III, Germany).

The functional groups in ACMCS, Cr(VI) laden ACMCS, and phosphate laden ACMCS were evaluated using FTIR (Perkin-Elmer “Spectrum BX” spectrometer) with the spectrum scanned from 400 to 4000 cm⁻¹. The Fe₃O₄ crystalline form of ACMCS was qualitatively analyzed by X-ray diffraction (XRD). In the analysis, the ACMCS diffraction crystal surface was compared to the Fe₃O₄ standard PDF card.

The elemental speciation and surface binding states of ACMCS, Cr(VI) laden ACMCS, phosphate laden ACMCS and regenerated ACMCS were analyzed by XPS. In the XPS analysis, 0.1 g of ACMCS was placed in 50 mL of phosphate/bichromate solution with a concentration of 1 g L⁻¹. The measurements were conducted with a spectrometer (ESCALAB 250) with MgKα irradiation (1486.71 eV of photons) as the X-ray source.

2.4. Adsorption tests

The stock phosphate and Cr(VI) solutions were prepared by dissolving Na₂HPO₄ (AR grade) and K₂Cr₂O₇ (AR grade) in deionized water. The ACMCS (0.1 g) was placed in 25 mL of the stock solutions of different concentrations and conditions in a set of 50 mL Erlenmeyer flasks. The residual phosphate was determined at the wavelengths of UV-maximum (λ_max) at 700 nm by the ammonium molybdate spectrophotometric method with a UV-visible spectrophotometer (model UV754GD, Shanghai). Similar operations were carried out to determine the residual Cr(VI) using the UV-visible spectrophotometer at 540 nm by the 1,5-diphenylcarbohydrazide spectrophotometric method.

In the pH effect experiments, the initial pH values of phosphate and Cr(VI) solution (20 mg P per L, 50 mg Cr per L) were first adjusted to a pH range from 3.09 to 11.84 using 1 mol L⁻¹ of HCl or NaOH solution. Mixing was carried out at room temperature (20 ± 1 °C) on a rotary mixer (120 times per min) for 60 min.

Adsorption kinetics experiments were carried out by mixing a series of ACMCS solutions with concentrations of 10, 20, 40 mg P per L and 20, 50, 80 mg Cr per L (20 ± 1 °C). Each sample was extracted at required time intervals in the range of 1 to 90 min. Then, the residual phosphate and Cr(VI) in the solutions were analyzed. Isotherm experiments were conducted by maintaining the water bath at 25, 35 and 45 °C. The phosphate solutions (pH 6.0 to pH 7.0) had a concentration range of 10 to 300 mg P per L. The Cr(VI) solutions (pH 6.0 to 7.0) had a concentration range of 10 to 1500 mg Cr per L. After mixing for 12 h, the residual phosphate and Cr(VI) was then detected. The saturation absorptive ACMCS was placed in NaOH or HCl solution (0.2 mol L⁻¹) for 3 h. After washing and drying, the regenerated ACMCS was used to conduct the re-adsorption experiments. The adsorption–desorption operations were conducted for 3 cycles and each regeneration efficiency was evaluated.

3. Results and discussion

3.1. Characteristics of all magnetic samples

3.1.1. VSM curves of ACMCS, phosphate laden ACMCS and Cr(VI) laden ACMCS. The magnetic properties of the adsorbent (ACMCS), Cr(VI) laden ACMCS, and phosphate laden ACMCS were measured by VSM at room temperature. Fig. 2A shows the curve of ACMCS, reflecting zero coercivity and reversible hysteresis behavior. Although there was a difference from previous studies, the results still illustrated that this magnetic adsorbent was superparamagnetic, with moderate hysteresis loss. The magnetic biosorbents were reported with a range of 6.33 to 8.89 emu g⁻¹.^19–21 The saturated magnetization value of clean ACMCS was measured to be 9.23 emu g⁻¹, which was slightly higher than the previous data. The ACMCS could be easily recovered from the solution by adding an external magnetic field with high superparamagnetism.

Fig. 2 Magnetization curves of ACMCS (A), SEM images of ACMCS at different magnifications: (B) ×200, (C) ×2000, and SEM images of spent ACMCS at different magnifications: (D) ×2000, (E) ×5000.

The magnetic properties of laden ACMCS are also indicated in Fig. 2A. The saturated magnetization value of Cr(VI) laden ACMCS was measured to be 4.52 emu g⁻¹, and that of phosphate laden ACMCS was measured to be 5.86 emu g⁻¹. The saturated magnetization value of spent ACMCS showed hysteresis loss in both the Cr(VI) and phosphate adsorption processes. However, the relatively weak magnetic properties could be enough to separate the biosorbent from the treated solution by an external magnetic field, as the spent ACMCS still had a highly saturated magnetization value. As a result, ACMCS could be recovered from the aqueous solution after the adsorption process.

3.1.2. The SEM images of ACMCS. The SEM images of different magnifications of the non-adsorption sample and the saturated adsorption samples are shown in Fig. 2B–E. The SEM result indicated that ACMCS could provide a powerful contact area and adequate active sites for Cr(VI) and phosphate removal. First, the lignocellulosic structures of ACMCS were in evidence, which provided a smooth surface with high homogeneity (B). In addition, the structures increased the BET surface area of ACMCS, which is based on intricate and complex pores and channels (C). Also, it was obvious that the principal component of adsorption covered the surface of ACMCS (C).

After the Cr(VI) and phosphate adsorption, there was almost no change on surface of the spent ACMCS; however, in some cases, some milky particles were attached to or covered the surface of the spent samples (D and E). This phenomenon, which may originate from the accumulated adsorbates, was not commonly detected in all the SEM images of the spent samples. However, these milky particles would block the channels and cover the magnetic particles on the surface of ACMCS, which partially explained the decreased magnetization of spent ACMCS.

3.1.3. TEM images of ACMCS. The TEM images, in accord with the VSM results, showed that a large amount of Fe₃O₄ had been introduced onto the skeleton of corn straw (Fig. 3A and B). The particle diameters of Fe₃O₄ were in the range of 6.9 to 14.5 nm (with an average diameter of approximately 10.1 nm). This data was compared with previous reports, and they were found to be consistent.^15,16,22,23 As a result, the TEM results indicated that ACMCS could be separated from aqueous solution by adding a magnetic field.

Fig. 3 TEM images of ACMCS: (A) 100 nm, (B) 20 nm; FT-IR spectra of clean, phosphate/Cr(VI) laden ACMCS and magnetized adsorbent without amine (C); XRD of clean ACMCS, Fe₃O₄ and bare corn straw (D).

3.1.4. FT-IR spectra of virgin corn stalk, ACMCS, phosphate laden ACMCS and Cr(VI) laden ACMCS. The infrared spectra of virgin corn stalk, clean ACMCS, phosphate laden ACMCS and Cr(VI) laden ACMCS are shown in Fig. 3C. The adsorption peak at 3386 cm⁻¹ indicated the possible existence of hydroxyl groups on the surface of ACMCS. Fe–O vibrations located at 559, 617, and 786 cm⁻¹ are observed in the FT-IR spectrum of ACMCS after co-precipitation.^16,24,25 The adsorption peak at around 1640 cm⁻¹ was caused by the vibration of aromatic cyclic groups, and the bands at 1330 cm⁻¹ corresponded to C–N bending vibrations in ACMCS, which were absent in the FTIR spectrum of virgin corn stalk. These spectra obviously demonstrated that Fe₃O₄ nanoparticles had been embedded on the virgin straw and that amine groups had been introduced onto the surface of the magnetic corn straw after co-precipitation and amine cross-linking reactions.

It was obvious that the specific bands (1052 cm⁻¹ and 1640 cm⁻¹) in phosphate loaded ACMCS were greatly enhanced. The specific band at 624 cm⁻¹ and the adsorption peak at 890 cm⁻¹ in Cr(VI) laden ACMCS are shown. In addition, the C–N bond (1330 cm⁻¹) in ACMCS shifted after the adsorption of Cr(VI) or phosphate. These observations indicated that Cr(VI) or phosphate was successfully adsorbed onto the surface of ACMCS.

3.1.5. Nitrogen content and XRD analysis. The nitrogen content in ACMCS was determined to be 9.89%, while it was only 0.46% in virgin corn stalk. The significant increase in the nitrogen content of ACMCS indicated that the amine reactions proceeded efficiently and that a large number of amine groups were grafted onto the framework of the magnetic samples. Based on the chemical composition of the quaternary amine group in –CH₂CHOHCH₂NH(CH₂CH₂NH)₃CH₂OHCHCH₂N(CH₂CH₂)₃⁺, the content of the quaternary amine charge was about 1.35 eq(+) g⁻¹.

The XRD diffraction spectra of ACMCS, virgin corn stalk and Fe₃O₄ are shown in Fig. 3D. The special peaks at 30.0, 35.7, 43.3, 57.5 and 62.8 were parallel to the different crystal planes of pure Fe₃O₄ with a spinal structure. In addition, it was obvious that the pattern of Fe₃O₄ crystal corresponded to the spectrum of ACMCS.

Based on the TEM, FT-IR and XRD analysis, it was obvious that the Fe₃O₄ nanoparticles had been successfully introduced onto the template of corn straw.

3.1.6. XPS of clean ACMCS, phosphate/Cr(VI) laden ACMCS and regenerated ACMCS. Clean ACMCS, phosphate laden ACMCS, Cr(VI) laden ACMCS and regenerated ACMCS were analyzed by XPS. The results of elemental speciation and the surface binding state are shown in Fig. 4. The basic elements of all ACMCS were C and O, which corresponded to the peak BE (banding energies) at around 284 eV and 531 eV, respectively (Fig. 4A). In addition, their atomic contents were evaluated to be 65.0 to 69.3% for C 1s and 17.6 to 22.8% for O 1s, which depended on the raw corn straw. The Cl 2p (196.4 eV) in clean ACMCS disappeared in phosphate laden ACMCS and Cr(VI) laden ACMCS. The peak BE of P 2p or Cr 2p was observed at 132.2 or 577.4 eV after the adsorption, and it indicated the existence of phosphate or Cr(VI) on the surface of ACMCS.

Fig. 4 XPS of clean, phosphate/Cr laden and regenerated ACMCS: (A) wide scan; (B) Fe 2p; (C) Cl 2p; (D) N 1s of phosphate laden and regenerated ACMCS. (E) N 1s of Cr laden and regenerated ACMCS; (F) P 2p; (G) Cr 2p.

3.1.6.1. Fe 2p of clean ACMCS and regenerated ACMCS. The BE of Fe 2p were analyzed at 710.0 eV and 713.3 eV, which correspond to Fe(II) and Fe(III) ((Fig. 4B)). Both Fe(III) and Fe(II) were obviously detected in all the ACMCS samples. The atomic ratio of Fe in the clean ACMCS was about 0.98%. After adsorption of phosphate/Cr(VI), the atomic ratio of Fe in phosphate laden regenerated ACMCS was about 0.72%, and that in Cr(VI) laden regenerated ACMCS was 0.61%. In addition, the ratio of Fe(II) and Fe(III) in regenerated ACMCS was slightly lower than that in clean samples (Fig. 4B). This indicated that a small amount of Fe(II) was reduced to Fe(III) during the adsorption process. The Fe concentrations in the aqueous solutions were also detected. The Fe(II) ions in the phosphate adsorption system were found to be about 0.09 to 0.7 mg L⁻¹. In contrast, very limited amounts of Fe(II) ions were detected in the Cr(VI) adsorption system; this may be due to the rapid reaction between Fe(II) and Cr(VI). The concentrations of Fe(III) ions in both adsorption systems were detected, and they were in the range of 1.2 to 3.2 mg L⁻¹.
3.1.6.2. Cl 2p of clean ACMCS and regenerated ACMCS. Cl 2p (196.4 eV) in clean ACMCS and regenerated ACMCS are shown in Fig. 4C. The Cl atomic ratio decreased to 3.37% after the regeneration process with HCl solution (0.1 mol L⁻¹). Moreover, the slight decrease of the Cl atomic ratio from 4.12% proved that ACMCS could be reused with low loss in an adsorption capacity.
3.1.6.3. N 1s of clean ACMCS, phosphate/Cr(VI) laden ACMCS and regenerated ACMCS. The N 1s BE (393.2 eV to 406.5 eV) in all the ACMCS samples is shown in Fig. 4D, which indicated the existence of the N⁺ in quaternary nitrogen N 1s (406.5 eV) and N in the amide (∼398.1 eV).²⁶ The peak of N⁺ was shifted after the phosphate and Cr(VI) adsorption process, whereas the band of N in amide was constant (Fig. 4D). As a result, N⁺ was one of the main functional groups during the adsorption process. In addition, the atomic ratio of N 1s decreased slightly after the desorption process, which proved the stability of the amine groups during the adsorption–desorption process.
3.1.6.4. Cr 2p and P 2p of saturated ACMCS and regenerated ACMCS. Cr 2p and P 2p of saturated ACMCS and regenerated ACMCS are shown in Fig. 4F and G. After the regeneration, the atomic ratio of phosphate decreased from 1.61% to 0.26% while the Cl atomic ratio increased. At the same time, the atomic ratio of Cr(VI) decreased from 1.76% to 0.73%. This result, along with the Cl 2p of clean ACMCS and regenerated ACMCS, demonstrated the good renewability and the adsorption mechanism of the system.

3.2. Effect of pH on Cr(VI) and phosphate adsorption

The effect of pH on the adsorption of Cr(VI) and phosphate by ACMCS is shown in Fig. 5. In the pH effect experiments, the pH values of phosphate and Cr(VI) solutions were determined after the adsorption process. The equilibrium pH showed an increasing trend with original pH ranges of less than 7.56 (phosphate) and 7.1 (Cr(VI)), whereas the equilibrium pH decreased when the original pH was greater than 7. In all cases, equilibrium pH was achieved within 10 to 25 min (ESI Fig. S1†). This difference between the original pH and the equilibrium pH was found in both adsorption processes; therefore, this change could be partially due to the special structure of ACMCS. ACMCS contains inherent weakly acidic hydroxyl and alkali amidogen groups, which would function as a buffer solution in the system.

Fig. 5 Effect of pH on Cr(VI) and phosphate adsorption (concentration: phosphate 20 mg P per L, Cr(VI) 50 mg Cr per L, contact time: 60 min, ACMCS dosage: 4 g L⁻¹, temperatures: 20 °C).

Fig. 5A shows the adsorption curve of phosphate. The results indicated that extreme pH weakened the adsorption effect significantly; the adsorbance increased from 56.6% to 97% with increasing pH value from 3.24 to 7.56, and the adsorbance decreased to 65.5% with increasing pH value from 7.56 to 11.84. When the pH was lower than 6.0, the HPO₄²⁻ and hydrogen ions formed H₂PO₄⁻; when the pH was lower than 4.0, H₃PO₄ and H₂PO₄⁻ were the major ions in the phosphate system. These reactions would interfere with the adsorption of phosphate by ACMCS. When the pH was greater than 10.0, the major ions in the phosphate were HPO₄²⁻ and PO₄³⁻. However, the significant increase in the number of OH⁻ ions, which compete with the phosphate ions, would hinder the phosphate adsorption. These factors weakened the bond strength between the negatively charged phosphate ions and the positively charged ACMCS surface. As a result, phosphate uptake by ACMCS declined under strong acid and strong base pH conditions.¹⁷

Fig. 5B shows the adsorption curve of Cr(VI), which indicates that the Cr(VI) adsorption capacities decreased with increasing pH value. The Cr(VI) uptake declined from 96.2% to 58% with increasing pH value from 2.92 to 11.80. At low pH, there were a great number of hydrogen ions in the solution, which augmented the positive charge on the ACMCS surface and enhanced the bond strength between anionic Cr(VI) ions and ACMCS.²⁷ Although Cr(VI) would change to Cr(III) at low pH, previous studies reported that the amounts of total Cr and Cr(VI) at low pH are more or less the same, which indicates that the presence of Cr(III) in the final solution is insignificant.²⁸ Also, the ACMCS had a larger specific area which can replace Cr(III) ions with positively charged groups.²⁹ As a result, the Cr(III) ions cannot cause the effect of pH on Cr(VI) adsorption. In addition, in Fig. 5B, the curve is flatter than the curve in Fig. 5A at pH values from 2.0 to 8.0. This indicates that ACMCS has a better acid tolerance of Cr(VI) adsorption.

3.3. Effect of ionic strength on Cr(VI) and phosphate adsorption

The effect of ionic strength on ACMCS adsorption of Cr(VI) and phosphate is shown in ESI Fig. S2.† In this experiment, NaCl was considered as the added electrolyte, which shares ions with hydrochloric acid and sodium hydroxide. The capacities for phosphate and Cr(VI) adsorption both decreased, from 93.2% to 56.0% and 98.2% to 88.6%, respectively, with a series of NaCl solutions with different densities, from 0 to 2000 mg Cl per L. The addition of the electrolyte decreased the thickness of the electric double layers and weakened the electrostatic interaction between the adsorbate and the adsorbent.³⁰ In addition, the adsorption sites were covered by ions with a charge balance and high ionic strength; thus, the electrostatic interaction between the adsorbate and the adsorbent was weakened.³¹ Also, the ion-exchange between the adsorbate and adsorbent was disrupted by the added electrostatic ion.³² All of these factors weakened the ACMCS uptake capacity of Cr(VI) and phosphate. In this experiment, a concentration of 355 mg Cl per L was employed as a special concentration in the series of NaCl solution densities; its ionic strength was the same at the pH values of 2 and 12. Under these conditions, the uptake capacities of phosphate and Cr(VI) were 86.8% and 95.9%. Compared with the pH experiment, the result showed a relatively slight effect on the Cr(VI) and phosphate adsorption. Therefore, the conclusion and analysis of the pH studies were persuasive, even without control of the ionic strength.

Adsorption experiments for bare corn straw and the magnetized corn straw without amine were also conducted. The uptake capacities of the magnetized and bare corn straw for phosphate were both lower than 3% (data not shown). In contrast, the adsorption capacities of magnetized and bare corn straw for Cr(VI) were about 10.3% and 18.9%. These results indicated that the magnetic property of the adsorbent had a strong effect on the adsorption of Cr(VI). Thus, these results indicated that the adsorption of Cr(VI) and phosphate was mainly based on the amine groups.

3.4. Regeneration studies

3.4.1. Regeneration efficiencies of ACMCS. Laden phosphate ACMCS and laden Cr(VI) ACMCS were desorbed by using 0.2 mol L⁻¹ HCl and NaOH solutions. The adsorption–desorption processes were conducted for three cycles, and the result is shown in Fig. 6. The adsorption quantity of phosphate by ACMCS decreased from 40.05 (mg g⁻¹) to 35.86 (mg g⁻¹) after elution by NaOH and from 40.05 (mg g⁻¹) to 37.25 (mg g⁻¹) after elution by HCl. The adsorption quantity of Cr(VI) by ACMCS decreased from 224.17 (mg g⁻¹) to 216.69 (mg g⁻¹) after elution by NaOH; however, a significant loss in adsorption capacity occurred after three adsorption–desorption cycles by HCl. As a result, HCl seemed to be more effective to regenerate the phosphate laden ACMCS than NaOH; contrastingly, NaOH was effective to regenerate Cr(VI) laden ACMCS, while HCl had no effect.

Fig. 6 Regeneration results using 0.2 mol L⁻¹ HCl and NaOH as eluents.

In phosphate adsorption–desorption processes, HCl may be more effective than NaOH as an eluent due to competitive adsorption. A previous study³³ indicated that chloride is a competitive ion in the phosphate adsorption process. In addition, the major ions in the phosphate were HPO₄²⁻ and PO₄³⁻ under strongly acidic conditions, which significantly weakened the bond strength between the phosphate ions and the ACMCS surface. These factors enhanced the eluting power of HCl; however, NaOH was only a competitive factor in phosphate adsorption. The loss of adsorbed Cr(VI) and phosphate after the chemical desorption processes was due to the destruction of quaternary ammonium functional groups on the surface of ACMCS.¹⁷ Also, it was validated by the XPS results, which indicated a decrease in N 1s after regeneration.

After three adsorption–desorption cycles (HCl and NaOH as eluents), the slight losses in Cr(VI) and phosphate adsorption capacity demonstrated that the magnetic biocomposite was renewable during a continuous adsorption process.

3.4.2. Properties of regenerated ACMCS. All phosphate/Cr(VI) laden ACMCS samples after eluting with HCl and NaOH solutions were detected by XPS, and their atomic ratios of Fe are shown in Fig. 7. There was a gradual decrease in the atomic ratios of Fe for all regenerated ACMCS samples over three adsorption–desorption cycles. The atomic ratio of Fe in clean ACMCS was about 0.98%. It decreased to 0.74% to 0.75%, 0.65% to 0.61% and 0.61% in the sequential 1^st, 2^nd, and 3^rd cycles of regeneration, respectively (Cr(VI) laden ACMCS, Fig. 7A). In contrast, the atomic ratio of Fe decreased less (0.68% to 0.72%) in phosphate laden ACMCS after the three cycles. This result corresponded well to the saturated magnetization of phosphate/Cr(VI) laden ACMCS samples.

Fig. 7 Atomic ratio of Fe (%) change in regenerated ACMCS during adsorption–desorption cycles: (A) Cr(VI) laden ACMCS; (B) phosphate laden ACMCS.

The weight loss of spent ACMCS after each regeneration was also evaluated, and the results are shown in Fig. 8. The loss of weight during the adsorption–desorption cycles was about 3.0% to 3.8%. Three aspects contributed to this loss: (i) the destruction of cellulose and hemicellulose in the structure of ACMCS; (ii) the loss of nano-Fe₃O₄ particles; (iii) the destruction of functional groups (amine groups) during the brine desorption cycles. XPS analysis demonstrated the decrease in the atomic ratio of N and Fe. This may be a drawback of this magnetic biosorbent. Thus, future research should be carried out to prevent the weight loss of ACMCS, especially the loss of nano-Fe₃O₄ particles and amine groups, during the regeneration process.

Fig. 8 Weight ratio change of regenerated ACMCS during adsorption–desorption cycles: (A) Cr(VI) laden ACMCS; (B) phosphate laden ACMCS.

3.5. Adsorption kinetics

The adsorption kinetics study provided significant information on the reaction pathways and demonstrated the uptake rate of the solution. The adsorption rate of phosphate and Cr(VI) onto ACMCS was analyzed by three important kinetic models.^34–36

Pseudo first-order model:

ln(q_e − q_t) = ln q_e − k₁t (1)

where q_e and q_t are the amounts of phosphate adsorbed per gram of ACMCS at equilibrium and time t (mg g⁻¹); k₁ is the pseudo first-order rate constant (min⁻¹).

Pseudo second-order model:

where k₂ is the equilibrium rate constant of pseudo second-order (g mg⁻¹ min⁻¹).

Elovich equation:

where α is the initial adsorption rate of the Elovich equation (mg g⁻¹ min⁻¹) and β is the desorption constant related to the extent of surface coverage and activation energy for chemisorption (g mg⁻¹).

The parameters of three kinetic models were shown in ESI Table S1 and ESI Fig. S3.† It was obvious that the R² values from the pseudo second-order kinetic model (0.999) were higher than that from the pseudo first-order model. Also, the calculated q_e² values were in good agreement with the experimental q_e values for all phosphate and Cr(VI) concentrations. This indicated that the adsorption of phosphate and Cr(VI) onto ACMCS followed the pseudo second-order model. This model represented the rate-limiting step as chemisorption involving valence forces through sharing or exchange of electrons between phosphate/Cr(VI) and ACMCS.³⁷

3.6. Adsorption isotherms

The equilibrium data of phosphate and Cr(VI) are shown in Fig. 9. The result indicated that the phosphate adsorption by ACMCS was an exothermic process and that the Cr(VI) adsorption was an endothermic process.

Fig. 9 (a) Adsorption isotherm of phosphate onto ACMCS; (b) Langmuir (phosphate); (c) Freundlich (phosphate); (d) Temkin (phosphate). (e) Adsorption isotherm of Cr(VI) onto ACMCS; (f) Langmuir (Cr(VI)); (g) Freundlich (Cr(VI)); (h) Temkin (Cr(VI)).

The equilibrium data were fit with three important isotherms.^38–41

Langmuir model:

where Q_max is the maximum adsorption capacity (mg g⁻¹); b is the Langmuir constant (mg⁻¹); C_e is the equilibrium concentration of adsorbate (mg L⁻¹); and q_e is the adsorption capacity (mg g⁻¹).

Freundlich model:

where K_F is the Freundlich constant (mg g⁻¹ mg⁻¹ L⁻¹) and n is a dimensionless exponent between 0 and 1 relating to the degree of surface heterogeneity.

Temkin equation:

q_e = B ln(AC_e) (6)

where A and B are Temkin isotherm constants.

The parameters of the three isotherms are shown in Table 1. The Langmuir model fitted well with the data of phosphate and Cr(VI) adsorption onto ACMCS (R² > 0.990). This indicated that the adsorption of phosphate and Cr(VI) by ACMCS were based on the homogenous distribution of adsorption sites on the surface. In addition, the Q_max obtained from the Langmuir model was close to the equilibrium data. As a result, the adsorbent showed a relatively high adsorption capacity among reported work. However, the R² values obtained from the Freundlich model were lower than those from the Langmuir model. It was obvious that Cr(VI) adsorption onto ACMCS was more suitably fitted with the Freundlich model than the phosphate adsorption.

Table 1 Isotherm parameters for adsorption capacity expressions

Adsorbate	T (°C)	Langmuir			Freundlich			Temkin
		Qmax (mg g^−1)	b	R^2	KF	n	R^2	A	B	R^2
P	25	40.1	0.0271	0.997	4.59	2.61	0.937	0.33	7.97	0.991
	35	37.5	0.0241	0.997	3.47	2.35	0.916	0.27	7.79	0.985
	45	34.3	0.0232	0.998	2.87	2.26	0.913	0.25	7.21	0.979
Cr(VI)	25	285.7	0.0052	0.993	7.56	1.96	0.974	0.08	53.3	0.948
	35	322.6	0.0051	0.992	7.83	1.91	0.968	0.07	59.3	0.956
	45	344.8	0.0051	0.995	7.99	1.87	0.963	0.07	64.9	0.961

---

The Temkin equation considers that the adsorption isotherms were based on adsorbent–adsorbate interactions, which caused the decrease in the adsorption heat of molecules in the layer with coverage. The R² values obtained from the Temkin model (0.979 to 0.991) of phosphate adsorption were higher than those of the Freundlich model. As a result, the phosphate adsorption onto ACMCS was based on an exothermic process and potential interactions between the phosphate and the homogenous surface. In contrast, the Cr(VI) adsorption onto ACMCS was based on an endothermic reaction and the homogenous distribution of adsorption sites on the surface.

4. Conclusions

ACMCS was prepared from corn straw as a raw material after in situ co-precipitation with Fe₃O₄ solution and amine functionalization with epichlorohydrin, N,N-dimethylformamide, triethylenetetramine and trimethylamine. The characteristics of the ACMCS, phosphate and Cr(VI) laden samples and regeneration samples were evaluated by VSM, TEM, SEM, FTIR, XRD and XPS. The FTIR and XRD results indicated the existence of interactions between amine groups and phosphate or Cr(VI). The phosphate adsorption onto ACMCS was inefficient at very high pH; however, ACMCS adsorbed Cr(VI) adsorption very well under strongly acidic conditions. The Freundlich model and the pseudo second-order model best fitted the data of Cr(VI) adsorption and indicated that the adsorption process was endothermic; both demonstrated physical adsorption and chemical adsorption. However, the Langmuir and pseudo second-order model best fitted the data of phosphate adsorption and indicated the potential interactions between the phosphate and the homogenous surface of ACMCS. In addition, the ACMCS had good regeneration performance after three adsorption–desorption cycles with high adsorption capacity. This study demonstrated that the magnetic biocomposite was an efficient water treatment material with high capacity for regeneration and recovery.

Acknowledgements

The research was supported by the National Natural Science Foundation of China (51178252, 51508307) and a China Postdoctoral Science Foundation funded project (2014M560556, 2015T80721). This work was also supported by grants from the Tai Shan Scholar Foundation.

References

1. C. Raji and T. S. Anirudhan, Batch Cr(VI) removal by polyacrylamide-grafted sawdust: Kinetics and thermodynamics, Water Res., 1998, 32, 3772–3780.
2. G. Bayramoğlu and M. Y. Arica, Adsorption of Cr(VI) onto PEI immobilized acrylate-based magnetic beads: Isotherms, kinetics and thermodynamics study, Chem. Eng. J., 2008, 139, 20–28.
3. R. Zhang, W. Bo and H. Ma, Studies on Chromium(VI) adsorption on sulfonated lignite, Desalination, 2010, 255, 61–66.
4. Y. Zeng, H. Woo, G. Lee and J. Park, Removal of chromate from water using surfactant modified Pohang clinoptilolite and Haruna chabazite, Desalination, 2010, 257, 102–109.
5. N. Ahmed El, Potential of pomegranate husk carbon for Cr(VI) removal from wastewater: Kinetic and isotherm studies, J. Hazard. Mater., 2009, 1, 132–141.
6. X. Xu, B. Y. Gao, X. Tan, Q. Y. Yue, Q. Q. Zhong and Q. Li, Characteristics of amine-crosslinked wheat straw and its adsorption mechanisms for phosphate and chromium(VI) removal from aqueous solution, Carbohydr. Polym., 2011, 84, 1054–1060.
7. X. Xing, B. Gao, W. Wang, Q. Yue, W. Yu and S. Ni, Adsorption of phosphate from aqueous solutions onto modified wheat residue: Characteristics, kinetic and column studies, Colloids Surf., B, 2009, 70, 46–52.
8. Z. Ren, X. Xu, B. Gao, Q. Yue and W. Song, Integration of adsorption and direct bio-reduction of perchlorate on surface of cotton straw based resin, J. Colloid Interface Sci., 2015, 459, 127–135.
9. X. Lv, J. Xu, G. Jiang, J. Tang and X. Xu, Highly active nanoscale zero-valent iron (nZVI)-Fe₃O₄ nanocomposites for the removal of chromium(VI) from aqueous solutions, J. Colloid Interface Sci., 2012, 369, 460–469.
10. Y. Liu, R. Fu, Y. Sun, X. Zhou, S. A. Baig and X. Xu, Multifunctional nanocomposites Fe₃O₄@SiO₂-EDTA for Pb(II) and Cu(II) removal from aqueous solutions, Appl. Surf. Sci., 2016, 369, 267–276.
11. X. Lv, X. Xue, G. Jiang, D. Wu, T. Sheng, H. Zhou and X. Xu, Nanoscale zero-valent iron (nZVI) assembled on magnetic Fe₃O₄/graphene for chromium(VI) removal from aqueous solution, J. Colloid Interface Sci., 2014, 417, 51–59.
12. L. Tan, J. Xu, X. Xue, Z. Lou, J. Zhu, S. A. Baig and X. Xu, Multifunctional nanocomposite Fe₃O₄@SiO₂–mPD/SP for selective removal of Pb(II) and Cr(VI) from aqueous solutions, RSC Adv., 2014, 4, 45920–45929.
13. T. A. H. Nguyen, H. H. Ngo, W. S. Guo, T. Q. Pham, F. M. Li, T. V. Nguyen and X. T. Bui, Adsorption of phosphate from aqueous solutions and sewage using zirconium loaded okara (ZLO): Fixed-bed column study, Sci. Total Environ., 2015, 523, 40–49.
14. T. A. H. Nguyen, H. H. Ngo, W. S. Guo, J. Zhang, S. Liang, D. J. Lee, P. D. Nguyen and X. T. Bui, Modification of agricultural waste/by-products for enhanced phosphate removal and recovery: Potential and obstacles, Bioresour. Technol., 2014, 169, 750–762.
15. W. Song, B. Gao, X. Xu, F. Wang, N. Xue, S. Sun, W. Song and R. Jia, Adsorption of nitrate from aqueous solution by magnetic amine-crosslinked biopolymer based corn straw and its chemical regeneration property, J. Hazard. Mater., 2016, 304, 280–290.
16. W. Song, B. Gao, T. Zhang, X. Xu, X. Huang, H. Yu and Q. Yue, High-capacity adsorption of dissolved hexavalent chromium using amine-functionalized magnetic corn straw composites, Bioresour. Technol., 2015, 190, 550–557.
17. T. Wang, X. Xu, Z. Ren, B. Gao and H. Wang, Adsorption of phosphate on surface of magnetic reed: characteristics, kinetic, isotherm, desorption, competitive and mechanistic studies, RSC Adv., 2016, 6, 5089–5099.
18. W. Song, B. Gao, T. Zhang, X. Xu, X. Huang, H. Yu and Q. Yue, High-capacity adsorption of dissolved hexavalent chromium using amine-functionalized magnetic corn straw composites, Bioresour. Technol., 2015, 190, 550–557.
19. H. Ma, J.-B. Li, W.-W. Liu, M. Miao, B.-J. Cheng and S.-W. Zhu, Novel synthesis of a versatile magnetic adsorbent derived from corncob for dye removal, Bioresour. Technol., 2015, 190, 13–20.
20. D. Mehta, S. Mazumdar and S. K. Singh, Magnetic adsorbents for the treatment of water/wastewater—A review, J. Water Process Eng., 2015, 7, 244–265.
21. X.-F. Sun, B. Liu, Z. Jing and H. Wang, Preparation and adsorption property of xylan/poly(acrylic acid) magnetic nanocomposite hydrogel adsorbent, Carbohydr. Polym., 2015, 118, 16–23.
22. N. Farzin Nejad, E. Shams and M. K. Amini, Synthesis of magnetic ordered mesoporous carbon (Fe-OMC) adsorbent and its evaluation for fuel desulfurization, J. Magn. Magn. Mater., 2015, 390, 1–7.
23. C. Zhang, L. Chen, T.-J. Wang, C.-L. Su and Y. Jin, Synthesis and properties of a magnetic core–shell composite nano-adsorbent for fluoride removal from drinking water, Appl. Surf. Sci., 2014, 317, 552–559.
24. N. M. Mahmoodi, Synthesis of core–shell magnetic adsorbent nanoparticle and selectivity analysis for binary system dye removal, J. Ind. Eng. Chem., 2014, 20, 2050–2058.
25. F.-L. Mi, S.-J. Wu and Y.-C. Chen, Combination of carboxymethyl chitosan-coated magnetic nanoparticles and chitosan-citrate complex gel beads as a novel magnetic adsorbent, Carbohydr. Polym., 2015, 131, 255–263.
26. R. Pietrzak, XPS study and physico-chemical properties of nitrogen-enriched microporous activated carbon from high volatile bituminous coal, Fuel, 2009, 88, 1871–1877.
27. S. S. Baral, N. Das, G. R. Chaudhury and S. N. Das, A preliminary study on the adsorptive removal of Cr(VI) using seaweed, Hydrilla verticillata, J. Hazard. Mater., 2009, 171, 358–369.
28. S. Gupta and B. V. Babu, Removal of Toxic Metal Cr(VI) from Aqueous Solutions Using Sawdust as Adsorbent: Equilibrium, Kinetics and Regeneration Studies, Chem. Eng. J., 2009, 150, 352–365.
29. S. Chen, Q. Yue, B. Gao and X. Xu, Equilibrium and kinetic adsorption study of the adsorptive removal of Cr(VI) using modified wheat residue, J. Colloid Interface Sci., 2010, 349, 256–264.
30. J. D. Filius, D. G. Lumsdon, J. C. L. Meeussen, T. Hiemstra and W. H. V. Riemsdijk, Adsorption of fulvic acid on goethite, Geochim. Cosmochim. Acta, 2000, 64, 51–60.
31. V. J. P. Vilar, C. M. S. Botelho and A. R. B. Rui, Influence of pH, ionic strength and temperature on lead biosorption by Gelidium and agar extraction algal waste, Process Biochem., 2005, 40, 3267–3275.
32. J. Lützenkirchen, Ionic Strength Effects on Cation Sorption to Oxides: Macroscopic Observations and Their Significance in Microscopic Interpretation, J. Colloid Interface Sci., 1997, 195, 149–155.
33. S. Zhong, D. Chen, Q. Xu and T. Chen, Adsorption of nitrate and Cr(VI) by cationic polymer-modified granular activated carbon, Pattern Recognition, 2011, 175, 298–305.
34. X. Sun, L. Yang, Q. Li, J. Zhao, X. Li, X. Wang and H. Liu, Amino-functionalized magnetic cellulose nanocomposite as adsorbent for removal of Cr(VI): Synthesis and adsorption studies, Chem. Eng. J., 2014, 241, 175–183.
35. J. Liu, R. Zhu, T. Xu, Y. Xu, F. Ge, Y. Xi, J. Zhu and H. He, Co-adsorption of phosphate and zinc(II) on the surface of ferrihydrite, Chemosphere, 2015, 144, 1148–1155.
36. K. Klusácek, R. R. Hudgins and P. L. Silveston, Multiple steady states of an isothermal catalytic reaction with Elovich adsorption, Chem. Eng. Sci., 1989, 44, 2377–2381.
37. P. Wu, W. Wu, S. Li, X. Ning, N. Zhu, L. Ping, J. Wu, Y. Chen and D. Zhi, Removal of Cd²⁺ from aqueous solution by adsorption using Fe-montmorillonite, J. Hazard. Mater., 2009, 169, 824–830.
38. A. Mittal, D. Jhare and J. Mittal, Adsorption of hazardous dye Eosin Yellow from aqueous solution onto waste material De-oiled Soya: Isotherm, kinetics and bulk removal, J. Mol. Liq., 2013, 179, 133–140.
39. A. A. El-Bindary, M. A. Hussien, M. A. Diab and A. M. Eessa, Adsorption of Acid Yellow 99 by polyacrylonitrile/activated carbon composite: Kinetics, thermodynamics and isotherm studies, J. Mol. Liq., 2014, 197, 236–242.
40. A. S. Cukrowski, Equilibrium uptake, isotherm and kinetic studies of Cd(II) adsorption onto iron oxide activated red mud from aqueous solution, J. Mol. Liq., 2015, 202, 165–175.
41. L. Cui, L. Hu, X. Guo, Y. Zhang, Y. Wang, Q. Wei and B. Du, Kinetic, isotherm and thermodynamic investigations of Cu²⁺ adsorption onto magnesium hydroxyapatite/ferroferric oxide nano-composites with easy magnetic separation assistance, J. Mol. Liq., 2014, 198, 157–163.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra06801d

---