Removal of uranium(VI) from aqueous solution by Camellia oleifera shell-based activated carbon: adsorption equilibrium, kinetics, and thermodynamics

Along with the increase in anthropogenic production and living activities, various heavy metals and organic pollutants are discharged continuously into water bodies and soils each year (Djehaf et al. 2017; Igbanoi et al. 2019). For instance, leachates derived from municipal solid wastes often contain toxic chemicals and lead to the pollution of surface and groundwater (Iwuoha & Akinseye 2019). Notably, considerable attention is being paid to the pollution from heavy metals given that they are non-biodegradable, prone to accumulate in the environment, and likely to have deleterious effects on human beings and living organisms at low concentrations. In this respect, recent studies have focused on monitoring and evaluation of their potential ecotoxicity by means of bioassays using higher plants and natural bioluminescent bacteria as indicators (Iqbal 2016; Abbas et al. 2018; Iqbal et al. 2019). Dysfunction tests on the propagation of insects were applied to evaluate the degree of poisonous metal contamination in polluted areas (Sun et al. 2016). In summary, metal contamination poses a serious threat to human survival and has become one of the most pivotal environmental problems for global concern.

For instance, U(VI) is considered one of the typical heavy metal ions with chemical and radioactive toxicity. Because of its accumulation effect, the enrichment of U(VI) in biological tissues can result in severe diseases and damages to living organisms, including toxic hepatitis, skin corrosion, kidney damage, leukemia, genetic aberration, and cancer (Asic et al. 2017; Bjorklund et al. 2017). The World Health Organization has set the maximum allowable level of uranium in domestic water supplies at 15 μg/L (Banning & Benfer 2017). Given that the concentration of U(VI) is usually higher than the permissible levels, we must search for economic and effective methods for the removal and recycling of uranium from wastewater.

Various physicochemical methods have been developed to remove uranium from wastewater; these methods include solvent extraction, electrochemical treatment, oxidation–reduction method, reverse osmosis, ion exchange, chemical precipitation/flocculation, and membrane nano-filtration (Bhalara et al. 2014). However, these methods often have shortcomings, such as high operation cost, large energy consumption, incomplete metal removal, and generation of secondary contaminants. These shortcomings restrict the further application of these methods in large-scale in-situ operations. Adsorption stands out as one of the most effective methods in wastewater treatment and is commonly used in the removal of heavy metals worldwide (Abbas et al. 2017; Jennifer & Ifedi 2019; Alaqarbeh et al. 2020; Alkherraz et al. 2020; Awwad et al. 2020). Such wide usage of adsorption can be attributed to its advantages, such as easily obtainable sources, low operation cost, environmental friendliness, and minimization of sludge generation.

In recent years, the adsorption of U(VI) by metal oxides (Wang et al. 2013), layered double hydroxide composites using various intercalants (Zou et al. 2017; Chen et al. 2018; Reda et al. 2019), and amidoxime-based materials (Lu et al. 2017; Qin et al. 2017; Zeng et al. 2017), have been investigated. These materials exhibit relatively high uranium adsorption capacity. In addition, other investigators have demonstrated that a variety of agricultural and forestry wastes, such as citrus peel (Bhatti et al. 2016), Musa paradisiaca peels (Ibisi & Asoluka 2018), sweet potato peels (Chidi & Kelvin 2018), rice husk (Kausar et al. 2017), groundnut shell (Babarinde & Onyiaocha 2016), and modified Vigna radiata (Naeem et al. 2017) can be used as low-cost biosorbents to remove toxic heavy metal ions effectively.

Apart from the above agricultural and forestry residues, various carbonaceous materials were extensively used to remove radionuclides from aqueous solution. Sun et al. (2017) reported that the sorption of U(VI) by sulfonated graphene oxide was independent of ionic strength, and the maximum sorption capacity calculated from Langmuir model was 45.05 mg/g. Xu et al. (2020) investigated the adsorption of U(VI) by hyacinth-derived biochar and observed that the adsorption process can follow pseudo-second-order kinetics with a corresponding U(VI) adsorption capacity of 55.32 mg/g. Activated carbon is also one of the most important carbonaceous materials that can be widely applied in the biosorptive removal of heavy metals and other contaminants because it has a porous structure with a high surface area ranging from 600 m²/g to 2,000 m²/g (Lu et al. 2015; Legrouri et al. 2017). The removal of organomercurial compounds from wastewater by using activated carbons was examined and compared by using an ion exchange resin – Amberlite GT73 (Velicu et al. 2007). The adsorption capacity of activated carbons was greater than that of the ion exchange resin. Activated carbon-enhanced activated sludge process is also used to treat petroleum wastewater (Jafarinejad 2017). Many industrial, agricultural, and forestry residues, such as lignocellulosic waste (Seyyedeh et al. 2020), barley straw (Pallares et al. 2018), date press cake (Norouzi et al. 2018), walnut shell (Teixeira et al. 2019), guava seeds (Pezoti et al. 2016), loofah (Shih et al. 2020), and coal pitch (Zhong et al. 2016) can be used to prepare various activated carbons (Fazal-ur-Rehman 2018). Nonetheless, their adsorption performance toward various heavy metals is limited. Researchers should continue searching for cheap and easily available biomass that can be used to prepare activated carbon with high metal adsorption capacity.

Camellia oleifera is a woody oil-bearing tree species distributed specifically in South China, with the total cultivation area of up to 3.5 × 10⁶ hm² and annual yield around 560 million tons (Fan et al. 2017). The seeds of Camellia oleifera are mainly used to produce high-grade tea oil, but most of Camellia oleifera shells are annually thrown away as agricultural wastes and used as fuels or fertilizers in the countryside. As one of the underused agricultural byproducts of Camellia oil, the shells are not only easily available but can also be converted into activated carbon because they contain abundant lignin, hemicellulose, and cellulose. Furthermore, given that Camellia oleifera shells are inexpensive and nontoxic to the environment, we can turn such waste into a profitable material if the shells are processed into a high-performance adsorbent for use in the treatment of uranium-bearing wastewater. In this work, we used Camellia oleifera shells to prepare porous activated carbon with H₃PO₄ activation method. Afterward, the Camellia oleifera shell-based activated carbon (COSAC) was employed to remove U(VI) ions from an aqueous solution. The effects of various parameters affecting the removal of uranium ions by adsorption onto COSAC powder were investigated, and the results were analyzed in detail. The findings were followed up by the investigation of the proper isotherms and kinetics models. The present article aimed to evaluate the capability of COSAC to remove uranium ions from aqueous solutions.

U₃O₈ was procured from the School of Nuclear Resources and Nuclear Fuel Engineering, University of South China. The other main chemicals, including Arsenazo III, H₃PO₄, HCl, H₂O₂, and HNO₃ were purchased from Sinopharm Chemical Reagent Company, Shanghai, China. All chemicals used in this research were of analytical pure grade.

Camellia oleifera shells were obtained from Xiayuan village, Xinxu town, Lanshan county, Hunan Province, Central China, during the picking season of Camellia fruit (Figure S1). The shells were rinsed with distilled water to eliminate the water-soluble impurities and dried in an oven at 105 °C for 6 h. The dried shells were crushed and milled to between 0.85 and 2.00 mm (10 and 20 mesh, respectively). Then, the milled Camellia oleifera shells were impregnated with concentrated phosphoric acid solution (shell:H₃PO₄ with 3:1 mass ratio) for 6 h and then dried at 105 °C for 24 h to remove water. Then, the dried samples were carbonized in a muffle furnace at 550 °C for 2 h to modify their porous structures. The resultant activated carbon was thoroughly rinsed with distilled water to remove the excess phosphorous compounds and then dried in a vacuum drying oven at 105 °C overnight. Finally, the activated carbon was cooled, crushed, and filtered. Only the activated carbon with the particle size distribution from 0.2 mm to 0.5 mm was collected and stored in a desiccator ready for use.

The quality test indexes of COSAC were determined in terms of the Chinese National Standard for Nut Shell Activated Carbon Test (No. GB/T 7702-1997). Scanning electron microscopy (SEM, Model S-4800 Hitachi, Tokyo, Japan) images were used to investigate the morphology and structure of the COSAC surface. The samples were fixed to a SEM stub using a conductive tape and sputter coated with gold.

The stock solution of U(VI) (1,000 mg/L) was prepared by dissolving 1.1792 g U₃O₈ in a mixture of HCl, H₂O₂, and HNO₃ (Li et al. 2018). All working solutions were obtained by diluting the above stock solution with distilled water to achieve the desired concentrations. The initial pH was adjusted to the required values with dilute solutions of NaOH and HCl. The pH of the solutions was determined using a pHS-3C digital pH meter (Shanghai Leici Device Works, Shanghai, China) with a combined glass-calomel electrode.

The adsorption experiments were performed as a function of the initial pH, contact time, temperature, and initial U(VI) concentration in 250 mL conical flask containing 100 mL U(VI) solution in a thermostatic shaker. The initial solution pH (2.0–7.0), contact time (0–180 min), temperature (298–318 K), and initial U(VI) concentration (20–240 mg/L) were evaluated. Briefly, 0.2 g COSAC powder was added to a series of flasks, each containing 100 mL U(VI) solution. Then, the flasks were agitated on a reciprocal rotary shaker at 140 r/min. The solution was periodically monitored to analyze the concentration of U(VI) remaining in the solution using a standard Arsenazo-III method (Khan et al. 2006). All measurements were performed thrice, and average values were presented.

The quality testing indexes of COSAC were analyzed as follows: granularity diameter, 10 − 20 mesh; specific surface area, 1,300 m²/g; total pore volume, 0.9 cm³/g; pH, 8.4; ash content, ≤9%; strength, ≥91%; moisture content, ≤4%; iodine value, 960 mg/g; methylene blue value, 135 mg/g; residual chlorine adsorption rate, ≥85%. The surface morphology of raw COSAC before U(VI) adsorption and one reaction after U(VI) adsorption was observed by SEM (Figure 1). A rough, irregular, and porous texture appeared on the surface of raw COSAC. This surface assumed a fractured and fragmented pattern, which is a typical feature for carbonization of plant tissues (Figure 1(a) and 1(b)). Notably, this material had an unordered macroporous structure with pore diameters of 10–20 μm (pores with widths exceeding about 50 nm are called macropores). After U(VI) adsorption, an evident change in the surface morphology of COSAC was the disappearance of original macropores (Figure 1(c) and 1(d)). The U(VI) sequestration by numerous active groups on the COSAC surface might have caused the clogging of macropores. Moreover, COSAC featured a high Brunauer, Emmett, and Teller specific surface area of 1,300 m²/g and a high total pore volume of 0.9 cm³/g. In general, high surface area and appropriate porous structure are often pivotal features of high-performance adsorbents. Therefore, COSAC is likely to be a promising carbonaceous material for the adsorptive removal of U(VI) contaminant.

Contact time is usually considered an important factor affecting the adsorption process. The adsorption of U(VI) proceeded relatively fast at the preliminary stage. Thus, the longer the adsorption process, the higher the U(VI) removal efficiency (Figure S2a). Then, the adsorption reached equilibrium at about 60 min, after which the U(VI) removal efficiency remained constant with the further increase in time. This phenomenon can be interpreted as follows. A high number of adsorption sites on the adsorbent surface were still empty at the initial stage. However, as the time elapsed, these empty sites were gradually occupied by adsorbate molecules, leading to the increase in U(VI) removal. When considerable active sites were occupied by adsorbate molecules, the adsorption experienced difficulty to proceed due to the repulsion between the solute molecules and the adsorbed layer. Hence, metal adsorption at the late stage was relatively slow before reaching equilibrium. Fasfous & Dawoud (2012) investigated the binding of U(VI) by multiwalled carbon nanotubes and confirmed that the time required to reach equilibrium was also approximately 60 min. By contrast, Caccin et al. (2013) reported a longer equilibrium time of about 48 h for the uptake of U(VI) on activated carbon derived from coconut shells.

The kinetics parameters in Equations (4)–(6) can be obtained from the curve fittings in Figure S2(b)–S2(d), and the fitted results are listed in Table 1. The R² values of pseudo-first-order and intraparticle diffusion models were distant from 1.0. Thus, neither of the two models can appropriately describe the U(VI) adsorption kinetics due to the evident void of linear correlation. However, the situation differed for pseudo-second-order model. Not only the R² value of the model (0.9946) was very close to 1.0, but the theoretical value of q_e calculated from the model (72.46 mg/g) was close to the observed experimental value (q_exp = 71.81 mg/g). Therefore, the adsorption of U(VI) onto COSAC follows the pseudo-second-order model considerably better than the other two models. Thus, pseudo-second-order model can be appropriately used to describe the adsorption kinetics. Although our result is similar to that of previous research on the U(VI) adsorption by olive stone-based activated carbon and magnetic Fe/Zn layered double oxide@carbon nanotube, which also followed pseudo-second-order kinetics (Kutahyali & Eral 2010; Chen et al. 2018), it differed from that of an earlier report on the U(VI) adsorption by activated carbon from coconut shell, which followed pseudo-first-order kinetics (Caccin et al. 2013).

The pseudo-second-order kinetics model is based on the assumption that the rate-limiting step may be chemical adsorption. Based on the above analysis, chemical adsorption is the dominant mechanism for U(VI) removal by COSAC. The key factor determining the adsorption rate is the chemical reactions occurring between the active sites on the COSAC surface and U(VI) in solution.

Wastewater treatment using traditional or untraditional adsorbents is highly dependent on the acidity or alkalinity of effluent because it influences the aquatic chemistry of heavy metals and the surface charge of adsorbents. The adsorption of U(VI) onto COSAC was carried out over the wide pH range from 2.0 to 7.0 (Figure 2). R and q increased with the increase in pH until the maximum value at pH 5.5, at which the maximum R and q were 89.1% and 71.28 mg/g, respectively, was reached. Afterward, the values decreased when the pH value exceeded 5.5. These results can be interpreted as follows. The activated carbon surface was coated with numerous active functional groups (hydroxyl, carboxyl, phosphate, etc.) with electrostatic adsorption capability. The solution pH was relatively low, uranium existed as the uncoordinated form of uranyl (UO₂²⁺) ions in the aqueous solution, and the aforementioned functional groups on the surface of COSAC were protonized and positively charged. Thus, COSAC displayed a weak U(VI) adsorption performance owing to the strong repelling force between uranyl ions and protonized functional groups. Accompanied with the increase in pH, these functional groups on the COSAC surface were progressively deprotonized and negatively charged. The electrostatic affinity between the functional groups with negative charges and uranyl was augmented, thus leading to increased U(VI) removal. As the solution pH was further increased beyond 5.5, U(VI) was prone to exist as stable forms of uranyl-carbonato and uranyl-hydroxyl-carbonato complexes, e.g. [UO₂(CO₃)₂]²⁻, [UO₂(CO₃)₃]⁴⁻, and [(UO₂)₂(CO₃)(OH)₃]⁻, in the solution (Fox et al. 2019). The shift in uranium speciation can hinder its adsorption process because of the electrostatic repulsion between the negatively-charged uranyl complexes and functional groups with negative charges. In addition, the volume of uranyl complex molecules is extremely large, which may be unbeneficial to their adsorption. Our results differ from those of previous research on the adsorption of U(VI) by graphene oxide-supported uniform magnetite nanoparticle composites, in which the U(VI) removal reached the maximum at the neutral pH range (6.0–8.0) and dropped slightly beyond pH 8.0 (Ding et al. 2017). According to the above analysis, a pH 5.5 was selected as the equilibrium time for other experiments.

Adsorption isotherm can be used to delineate the relationship between the amount of a solute adsorbed and its concentration in the equilibrium solution at a fixed temperature. In addition, studies should focus on gaining insights into the adsorbent–adsorbate interactions, estimate the adsorption capacity of adsorbents, and optimize adsorbent applications. In this work, adsorption experiments were carried out at the initial U(VI) concentration range of 40 mg/L to 240 mg/L. Evidently, the U(VI) adsorption equilibrium capacity increased with increase in the concentration of U(VI) (Figure 3). The equilibrium data for the U(VI) − COSAC adsorption system were further fitted using the Langmuir, Freundlich, D-R, and Temkin isotherms (Figure S3). Table 2 presents the corresponding parameters and R² values for the four isotherm models. The R² values of Freundlich, D-R, and Temkin isotherms were far from 1.0, suggesting that the adsorption onto COSAC cannot be described by these three isotherms. By contrast, the R² value of Langmuir isotherm was notably closer to 1.0. Hence, the adsorption of U(VI) onto COSAC can be best characterized by the Langmuir model, and the monolayer maximum capacity was 78.93 mg/g. Additionally, the R_L values obtained were in the range of 0–1 for U(VI) concentrations of 40 mg/L–240 mg/L (Table 2), indicating the favorable adsorption of U(VI) onto COSAC.

q_max is an important index for characterizing the adsorption performance of adsorbents and considered a pivotal parameter for engineering design. Previous studies have reported the q_max values of other carbonaceous materials for U(VI) adsorption, e.g. 20.76 mg/g for montmorillonite@carbon composite (Zhang et al. 2015), 39.1 mg/g for multiwalled carbon nanotubes (Fasfous & Dawoud 2012), 46.2 mg/g for Arsenazo-functionalized magnetic carbon composite (Li et al. 2018), 67.9 mg/g for amidoxime-modified multiwalled carbon nanotubes (Wu et al. 2018), and 163 mg/g for carbonaceous spheres from glucose (Cai et al. 2016). The q_max of COSAC (78.93 mg/g) in our work is notably greater than most of the above-mentioned carbonaceous materials. Thus, COSAC may be regarded as a potential high-performance adsorbent for U(VI) removal.

To evaluate the effect of temperature on the U(VI) adsorption by COSAC, the U(VI) removal capability was examined at different temperatures (298, 308, and 318 K). U(VI) adsorption decreased from 71.28 mg/g to 74.16 mg/g with the increase in temperature from 298 to 318 K (Figure S4). The thermodynamic parameters (ΔG⁰, ΔH⁰, and ΔS⁰) were calculated (Table 3). The negative value of ΔG⁰ indicates the spontaneous nature of adsorption of U(VI) by COSAC for all the tested temperatures. The ΔG⁰ value became more negative with the increase in temperature, indicating that the adsorption process was favorable at high temperatures. The positive value of ΔH⁰ confirmed the endothermic nature of removal process, whereas the positive value of ΔS⁰ suggests that this adsorption will result in increased randomness. Similar results have been reported for U(VI) adsorption onto other carbonaceous adsorbents (Zhou et al. 2015; Sun et al. 2017). In short, the adsorption of U(VI) by COSAC may be an enthalpically unfavorable (endothermic) and entropy-driven spontaneous process in nature.

In this work, characterization, uranium adsorption tests, adsorption equilibrium, and kinetics studies of COSAC were conducted to examine its U(VI) adsorption capability and to understand the corresponding adsorption mechanism. The adsorption of U(VI) on COSAC was investigated as a function of contact time, solution pH, and adsorbate concentration. The contact time of 60 min was sufficient to attain equilibrium. Kinetics data fit very well the pseudo-second-order model, suggesting that chemical adsorption is the rate-limiting step. The optimum pH for the maximum removal of U(VI) was observed at pH 5.5. The isotherm adsorption data were modeled best by the linear Langmuir equation with the monolayer adsorption capacity of 78.93 mg/g, which was comparable to those reported in literature. Thermodynamic calculations suggest that the adsorption process was endothermic, spontaneous, and accompanied by the increase in entropy. Overall, COSAC can be a highly attractive candidate for the effective removal of uranium from wastewater. Regeneration of the adsorbent and column experiments were subject to further research.

This research was sponsored by National Natural Science Foundation of China (No. 41773133), the Science and Technology Development Project of Hengyang City (No. 2019yj011171), and the Center for the Functionalization of Metal-Organic Frameworks Materials Studies.

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