Adsorption of vanadium with amorphous hydrated chromium oxide

The continued expansion of application fields and market demands is leading to rapid development of the vanadium industry (Khalid et al. 2017). Meanwhile, large amounts of vanadium-containing wastewater have been produced. Vanadium is toxic and can exist in three valence states – +3, +4 and +5 – in aqueous solution, of which +5 is the most stable (Kong et al. 2020). The +5 oxidation state is also the most toxic of the three (Yu & Yang 2019). Vanadium is recognized as a potentially dangerous pollutant following closely behind lead, mercury and arsenic (Naeem et al. 2007) so it is essential to remove vanadium ions from wastewater prior to discharge.

Various methods have been investigated for vanadium removal from vanadium-containing wastewater, etc, including chemical precipitation (Xiong et al. 2018), ion exchange (Hu et al. 2009), solvent extraction (Wang et al. 2021), adsorption (Li et al. 2014; Padilla-Rodríguez et al. 2015; Peng et al. 2017), and so on. Comparatively, adsorption has been used widely as it is highly effective, inexpensive, simple to operate and has high adsorption efficiency (Padilla-Rodríguez et al. 2015; Peng et al. 2017). Adsorption performance is closely related to the adsorbent's physicochemical characteristics (Hong et al. 2019). An adsorbent should have high adsorption capacity and depth, so that the vanadium concentration in the treated wastewater meets the vanadium discharge standard (e.g., <1.0 mg-V·L⁻¹, State Administration for Quality Supervision, Inspection and Quarantine, Ministry of Environmental Protection, People's Republic of China 2011).

Vanadium and chromium are in ores such as vanadium titano-magnetite, chromite, etc (Yang et al. 2020), so chromium could also be recovered in vanadium industry wastewater. As a result, use of a chromium compound for vanadium removal might be feasible.

Hydroxides have been reported for the adsorption of vanadium, chromium, phosphate, etc. (Naeem et al. 2007; Cheng et al. 2011; Li et al. 2014). A sample with enhanced crystallinity was reported as having a significantly smaller specific surface and poor adsorption capacity (Cheng et al. 2011), indicating the advantage of using amorphous adsorbents.

In this study, an amorphous hydrated chromium oxide was prepared via a hydrogen reduction method (Bai et al. 2006; Liang et al. 2020), its structure, specific surfaces and porosities characterized, and its vanadium adsorption properties, as prepared, studied. Variations in pH, adsorbent dosage, temperature, time, common co-existing ions and desorption efficiency were investigated. Appropriate kinetic and equilibrium models were applied to gain insight into vanadium adsorption characteristics.

Analytical grade potassium chromate (K₂CrO₄), sodium metavanadate (NaVO₃·2H₂O), hydrochloric acid (HCl), sodium hydroxide (NaOH), sodium nitrate (NaNO₃), sodium sulfate (anhydrous, Na₂SO₄), sodium chloride (NaCl), potassium chloride (KCl), magnesium chloride (MgCl) and sodium chloride (anhydrous, CaCl₂) were purchased from Sinopharm Chemical Reagent Co., Ltd (China). The hydrogen gas, purchased from Praxair Inc, was 99.99% pure by volume. All solutions used were prepared with deionized water.

Amorphous hydrated chromium oxide was prepared by hydrogen reduction. The heating apparatus, a tube stove with programmable temperature controller, was used as the reactor for the gas-solid reductive reaction. A nichrome boat loaded with potassium chromate was put into the sealed tube stove and hydrogen was introduced to form a reducing atmosphere. The stove was heated to 450 °C and kept at that temperature for 1.0 hour, before cooling naturally to ambient temperature. The resulting product was washed several times until all soluble components were removed completely. Finally, the washed product was dried at 80 °C for 24.0 hours.

X-ray diffraction (XRD) and Fourier transform infrared spectrometry (FT-IR) were used to determine the phases and structure of the sample. The XRD analysis was carried out with a Rigaku diffractometer in 2θ range 5 to 90° using CuKα radiation. The FT-IR spectra were recorded with a Spectra GXFT-IR Spectrometer with scan number 8 and 4 cm⁻¹ resolution. Field emission scanning electron microscope (SEM) (JSM-7610F) image analysis was carried out with15.0 kV accelerating voltage. BET specific surface and pore size distribution were tested using nitrogen adsorption and desorption in a physical and chemical adsorption instrument (NOVA3200e).

Batch adsorption studies were conducted to determine the equilibrium time to reach saturation. Typically, 20 mL of 300 mg-V·L⁻¹ vanadium ion solution was mixed with a known dose of adsorbent in 50 mL vials (oscillation 180 r·min⁻¹). All experiments were performed in triplicate, and the parameters varied in the study were initial pH (1.0 to 13.0), adsorbent dose (⁠⁠) (6:1 to 90:1), temperature (25 to 60 °C) and contact time (0.17 to 24.0 hr). The solution pH was adjusted by adding HCl or NaOH.

The effects of common co-existing ions were studied in mixed solutions with an initial vanadium concentration of 500 mg-V·L⁻¹, and the co-existing ions varied from 100 to 500 mg·L⁻¹. The experimental conditions adopted with the co-existing ions were initial pH 2.0, adsorbent dosage (⁠⁠) 60:1, 30 °C and 10.0 hours.

The FT-IR spectrum of the as-prepared sample – range 400 to 4,000 cm⁻¹, resolution 1 cm⁻¹ – is presented in Figure 1, and shows major peaks at 3,375, 2,017, 1,635, 835 and 505 cm⁻¹. The strong adsorption peak at 505 cm⁻¹ relates to Cr(III)-O antisymmetric stretching vibration, and that at 835 cm⁻¹ to Cr(III)-O-H bending vibration (Yang et al. 2011). The peak at 1,635 cm⁻¹ is caused by bending vibrations in free water molecules. The band corresponding to 1,700 to 2,100 cm⁻¹ is the H-O-H stretching vibration (Ratnasamy & Leonard 1972). The other peaks, at 2,800 to 3,500 cm⁻¹, arise from OH stretching on the water surface, which originate in the chemical adsorption of water dissociation (Ratnasamy & Leonard 1972). These characteristic peaks indicate that the sample is γ-CrOOH (Bai et al. 2006; Liang et al. 2020).

Figure 2 shows the XRD pattern and SEM image of the sample. Only weak diffraction peaks can be seen in the XRD pattern (Figure 2(a)), indicating that the sample is amorphous. In the SEM image (Figure 2(b)), the sample exhibits some lamination and is stacked into irregular agglomerates.

To determine the sample's specific surface and porosity, N₂ adsorption and desorption analysis was carried out. Figure 3(a) and (b) present the N₂ adsorption/desorption isotherms and pore size distribution of the sample, respectively. The hysteresis loop in the isotherms (Figure 3(a)) is typical type H4 (IUPAC classification), indicating the presence of meso- and micro-pores associated with capillary condensation (Thommes et al. 2015). In Figure 3(b), the pore size distribution curve shows a broad peak between about 10 and 50 nm with a maximum at 30 nm. The BET analysis thus indicates that the as-prepared γ-CrOOH is a typical mesoporous material with 99.2 m²·g⁻¹ specific surface and 0.2 cm³·g⁻¹ pore volume.

Metal ion adsorption on the adsorbent can be affected by the solution's initial pH. Twelve vanadium species can coexist in solution (Sharififard & Rezvanpanah 2021). The cationic species VO₂⁺ exists mainly at pH ≤2.0, the anionic species, such as decavanadates and other mono- or poly-vanadates, mainly at higher pHs (Li et al. 2014). The initial solution pH was adjusted between 1.0 and 13.0 with HCl and NaOH solutions. The adsorption test results are shown in Figure 4.

The solution's initial pH affects the extent of vanadium adsorption onto γ-CrOOH strongly, in behavior typical of metal cation adsorption. Adsorption efficiency increases from pH 1.0 to 2.0 – maximum efficiency is about pH 2.0 – and decreases as the pH increases from 3.0 to 13.0. This is likely associated with competition between H⁺ and VO₂⁺ for available surface sites (Li et al. 2014). Extreme acidity (pH <2.0) makes vanadium adsorption unfavorable, with negligible adsorption at pH 1.0. The sharp decrease in adsorption at pH ≥ 3.0 may arise because vanadium ions exist mainly as anionic species in that pH range rather than cations, which are the optimum for adsorption (Sharififard & Rezvanpanah 2021). Thus maximum vanadium adsorption occurs at pH 2.0.

A series of molar proportions were selected as adsorbent reference doses on the basis of the vanadium concentration in solution, and the vanadium adsorption efficiency was determined at pH 2.0. As shown in Figure 5, vanadium adsorption efficiency depends strongly on the adsorbent dosage, increasing from 19 to 98% as the molar ratio increases from 6:1 to 90:1. When the ratio is below 60:1, adsorption efficiency increases linearly to 88.14%; above that, it increases more slowly. This is thought to result from the increasing adsorbent dosage, providing more adsorption sites and making the efficiency improve gradually. It can be confirmed that the appropriate adsorbent dosage ratio is 60:1.

The effect of temperature was investigated in the range 25 to 60 °C (Figure 6). It is noted that vanadium adsorption increases with increasing temperature, although, at 30 °C and above, the increases are slight and the capacity remains around 9.60 mg·g⁻¹. The vanadium adsorption efficiencies are all close to 100%, however, and the concentration is reduced to <1.0 mg-V·L⁻¹. This indicates that higher temperatures favor the process and that it is endothermic (Salehi et al. 2020), and may be because the temperature rise increases the pore size of γ-CrOOH and more surface active sites are available. It may also be that the rates of molecular motion and mass transfer are speeded up by the temperature rise. When the temperature reaches about 30 °C, however, the number of γ-CrOOH surface active sites has reached its maximum and/or the molecular movement rate is at its maximum, making further adsorption capacity increases occur slowly (Li et al. 2014).

The equilibration time between vanadium and the γ-CrOOH adsorbent was determined by varying the contact time between 0.17 and 24.0 hours, keeping the initial pH and adsorbent dosage ratio constant at 2.0 and 60:1, respectively. Vanadium adsorption rates at different contact times are shown in Figure 7. The rate of vanadium adsorption onto γ-CrOOH increases quickly at first but slightly after 4.0 hours. More than 99% vanadium removal is achieved after 3.0 hours. After 10.0 hours contact, the vanadium concentration is 0.75 mg·L⁻¹; this means no further adsorption occurred. So 10.0 hours can be considered the equilibrium time.

The effects of common ions in wastewater, whether anions or cations, such as NO₃⁻, SO₄²⁻, Cl⁻, Na⁺, K⁺, Mg²⁺ and Ca²⁺ were studied (Salehi et al. 2020). The concentration of the co-existing ions was varied between 100 and 500 mg·L⁻¹, but the initial vanadium concentration was always 500 mg·L⁻¹. The other experimental conditions were initial pH 2.0, adsorbent dosage (⁠⁠) 60:1, 30 °C and 10.0 hours – see Figure 8. There is no obvious effect from the presence of the other ions on vanadium adsorption efficiency. In all of this set of trials the adsorption efficiencies exceeded 97%, showing that γ-CrOOH can remove vanadium from real effluents.

To determine the dynamics of vanadium adsorption onto γ-CrOOH, adsorption kinetic experiments were carried out. Two kinetic models – pseudo-first-order and pseudo-second-order – were used to analyze the data (Ho et al. 2000; Li et al. 2014). The fitting data from the two plots are shown in Table 1, which shows that the pseudo-second-order equation is the more appropriate with R² values between 0.986 and 0.994. The pseudo-second-order equation can thus be used to describe the adsorption process, which indicates that the limiting factor for vanadium adsorption onto γ-CrOOH is the chemisorption process (Xie et al. 2020).

The empirical equations of the Langmuir and Freundlich models are the most widely employed in analyzing adsorption processes (Naeem et al. 2007). The two parameter sets are listed in Table 2.

The Langmuir equation indicates that the adsorption energy of each molecule is the same, the surface sites of the adsorbent are independent, adsorption only takes place on some surface sites and there is no interaction between molecules. The maximal vanadium adsorption capacity calculated from the Langmuir model is 9.60 mg·g⁻¹, and the Langmuir constant, R², is 0.995. However, the Freundlich constant, R², is just 0.790. Thus the experimental data are better fitted to the Langmuir than the Freundlich model, which means that the active sites over the entire outer γ-CrOOH surface could be homogenous and vanadium adsorption might form a monolayer on the γ-CrOOH outer surface.

In this work, a mesoporous amorphous γ-CrOOH with a specific surface and pore volume of 99.2 m²·g⁻¹ and 0.2 cm³·g⁻¹ was prepared by hydrogen reduction of K₂CrO₄. It was used as an adsorbent for vanadium from wastewater. In batch mode studies, the adsorption of vanadium onto γ-CrOOH was found to be highly dependent on the solution pH, adsorbent dosage, temperature, and adsorption time.

When an adsorbent dosage ratio of 60:1 was applied to a solution containing 300 mg-V·L⁻¹, under initial conditions pH = 2 at T = 30 °C, more than 99% of the vanadium was removed after 3.0 hours. The residual vanadium concentration in the effluent after 10.0 hours was below 0.75 mg-V·L⁻¹, which meets the vanadium discharge standard (<1.0 mg·L⁻¹, China).

Studies on the effect of common co-existing ions showed that γ-CrOOH can remove vanadium from real effluents efficiently.

The adsorption kinetics data fit well with the pseudo-second-order equation, indicating that the limiting factor for vanadium adsorption ions onto γ-CrOOH was chemisorption. The Langmuir model – with R² = 0.995 – fitted the experimental data better than the Freundlich model, indicating that the active sites over the γ-CrOOH outer surface could be homogenous, with vanadium adsorption occurring on surface sites to form a monolayer and no interaction between the molecules.

This work was financially supported by the National Natural Science Foundation of China (U1903131).

All relevant data are included in the paper or its Supplementary Information.