Introduction

Due to the excessive amounts of heavy metal produced by anthropogenic activities discharged into the environment, the pollution of heavy metals has aroused wide concern. Among the major categories of heavy metals, radionuclides, known for its chemical toxicity and radiotoxicity, has gained wide public attention. Uranium, as the important source materials of nuclear energy, can be easily detected in the environment due to effluents from the smelting of uranium ore and normal discharge of radioactive waste from nuclear power plants (Ogar et al. 2014; Gudkov et al. 2016).Unfortunately, uranium is inhaled and ingested by human beings eventually through the food chain. Once consuming more than tolerated levels of uranium, it can lead to an increase risk of lots of diseases and physical deformities (ATSDR 2013; Soltani et al. 2019). Meanwhile, the non-renewable characteristics of uranium may prevent the sustainable utilize of nuclear energy (Li et al. 2016). Therefore, it is crucial to remove and recover of uranium from polluted water efficiently.

Traditional methods of uranium removal and recovery include solvent extraction (Cheira 2020; Mathuthu et al. 2019), chemical precipitation (Djedidi et al. 2009), reverse osmosis (Abdel-Khalek et al. 2011; Schulte-Herbrueggen et al. 2016), micellar ultrafiltration (Cojocaru et al. 2009), and adsorption. These methods mentioned above have been demonstrated effective in the field of uranium-contaminated wastewater treated. However, limitations of these methods, such as high energy consumption, secondary pollution, and incomplete removal, are unfavorable to the widespread popularization and promotion of these methods. Among these methods, adsorption gains an advantage over other methods owing to its high efficiency and environmentally friendly.

Biosorption is an adsorption process that utilizes biological materials as adsorbents (Ghasemi et al. 2011). Over the past decades, biosorption has attracted great attentions owe to its unparalleled advantage, such as simplicity, repeatability, eco-friendly, and so on (Yi et al. 2016). In recent years, many biosorbents are being widely used in uranium wastewater treatment, for instance, Citrus limon peels (Šabanović et al. 2019), Saccharomyces cerevisiae (Zheng et al. 2018), and Lemna sp. (Vieira et al. 2019). Among the biosorbents mentioned above, microalgae possess incomparable technical advantages. Firstly, the cultivation of algae is relatively simple (Hu et al. 2011). Secondly, it is easier be harvested because of widely distributed, locally abundant, and much larger size than bacteria (Yu et al. 2014). Finally, and most importantly, it has been confirmed that microalgae have a strong affinity for metals. The functional groups on the cell wall such as carboxyl and amino are important sites for the microorganism to bind metals (Vilar et al. 2008; He and Chen 2014). Recently, freshwater or marine algae, such as Spirulina platensis (Mohammed 2019), Gracilaria corticata (Dabbagh et al. 2018), Cystoseira sp. (Cem et al. 2017), Cladophora hutchinsiae (Bagda et al. 2017), and brown algae (Moghaddam et al. 2013), attracted a great deal of attention for uranium removal. However, according to our literature survey, rarely investigation on the biosorption of uranium by Anabaena flos-aquae has been reported.

Anabaena flos-aquae, a species of filamentous cyanobacteria, is known for its nitrogen-fixing abilities and is one of cyanobacterial species that produce toxins. If it could be harvested and played a vital role in the uranium-containing wastewater treatment, then we can realize a “win-win situation.” The influences of various experimental parameters on the biosorption of uranium were investigated. The characterization of raw and U-loaded Anabaena flos-aquae was performed by Fourier transform infrared spectroscopy (FTIR), scanning electron microscopy (SEM), and energy dispersive X-ray fluorescence spectrometry (EDS).

Materials and methods

Preparation of biosorbent

Anabaena flos-aquae strain was obtained from the Freshwater Algae Culture Collection at the Institute of Hydrobiology, Chinese Academy of Sciences. The cells of Anabaena flos-aquae were cultivated in BG-11 medium in a fermenter until the end of the exponential growth phase. The biomass was then harvested by centrifugation and washed twice with distilled water. Afterwards, the precipitate was freeze-dried, passed through a 100-size mesh, and finally preserved in desiccator for following experimental use.

Batch experiments

All adsorption experiments were duplicated at least to make reliable and repeatable results. The experiments were performed by setting the reaction mixture of Anabaena flos-aquae (10 mg) and 30 mL uranium solution of certain concentration into a conical flask, so as to characterize specific sorption parameters: initial pH, contact time, initial uranium concentration, and environmental temperature. The initial pH (2–8) was adjusted with trace amount of NaOH and HNO3. The effects of different gradient of Anabaena flos-aquae dose (10, 20, 30, 40, 50, 60, 80, 100 mg) on uranium biosorption were studied. Then, the mixture was rotated at 120 rpm in a rotary shaker. The uranium concentration in the supernatant was determined at 650 nm with UV-Vis spectrophotometer after centrifugation. The removal efficiency (R) and sorption amount (Q) of uranium were calculated by Eq. (1) and Eq. (2).

$$ R\left(\%\right)=\frac{C_0-{C}_e}{C_0}\times 100\% $$
(1)
$$ Q=\frac{\left({C}_0-{C}_e\right)\times {V}_{,}}{m} $$
(2)

where C0 and Ce are the initial uranium concentration and equilibrium uranium concentration (mg/L), respectively; V is the volume of the solution (L); and m is the mass of the absorbent (g).

Characterization

The surface morphology of Anabaena flos-aquae was examined by SEM (GeminiSEM 300) coupled with EDS (operating conditions: probe current 45 nA, accelerating voltage 20 kV and counting time 60 s). The samples were gold-coated before observation to enhance the electrical conductivity. Functional groups on the surface of Anabaena flos-aquae were observed using FTIR (Bruker Vector 22). To perform FTIR analysis, sample disks were prepared by mixing 0.15 g potassium bromide (KBr) with 0.005 g dry Anabaena flos-aquae and pressed into tablet form. Then, the spectra were recorded over the 400–4000 cm−1 region with a resolution of 0.2 cm.

Results and discussion

Effect of biomass dose

As presented in Fig. 1, the adsorption capacity decreased with the increase of biomass dose. The maximum biosorption capacity was 173.06 mg/g with 10 mg biomass dose. Similar results have been reported regarding the effect of biomass dose on biosorption capacity (Li et al. 2016). The negative relationship between biosorption capacity and biomass dose may result from the aggregate effect of sorbents (Deng et al. 2011). A “screening effect” that increased electrostatic interactions between cells, limited availability of binding sites, and reduced mixing generated by the higher sorbent dose results in a lower uranium sorption per unit of sorbent. In conclusion, 10 mg of Anabaena flos-aquae is appropriate for the following uranium biosorption experiments.

Fig. 1
figure 1

The effect of sorbent dose on the uranium adsorption by Anabaena flos-aquae (C0 = 70 mg/L, t = 1 h, V = 30 mL, T = 298.2 K, pH = 5.0)

Kinetic studies

The influence of contact time on uranium biosorption was investigated. As shown in Fig. 2, the adsorption capacity increased sharply during the first 20 min and then reached equilibrium of maximum uranium sorption amount 196.4 mg/g at 50 min. Therefore, the reaction time of 1 h was employed in the following experiments.

Fig. 2
figure 2

Kinetics of uranium adsorption onto Anabaena flos-aquae (C0 = 70 mg/L, m = 10 mg, V = 30 mL, T = 298.2 K, pH = 5.0)

The pseudo-first-order model and the pseudo-second-order model were employed to describe the adsorption kinetic characteristic. The two models can be expressed by the following Eq. (3) and Eq. (4).

$$ \mathrm{Pseudo}-\mathrm{first}-\mathrm{order}:{q}_t={q}_e\ \left(1-\exp \left(-{k}_1t\right)\right) $$
(3)
$$ \mathrm{Pseudo}-\mathrm{second}-\mathrm{order}:{q}_t=\frac{t}{\left(1/{k}_2{q}_{\mathrm{e}}^2\right)+\left(t/{q}_e\right)} $$
(4)

where qe is the uranium uptake amount at equilibrium, and qt refers to the uranium adsorption capacity at any time “t”; k1 and k2 refer to the rate constant of the pseudo-first-order and pseudo-second-order sorption, respectively. The parameters obtained by the two models are presented in Table 1.

Table 1 The parameters of two models for uranium adsorption onto Anabaena flos-aquae

Table 1 shows the correlation coefficient (R2) of the pseudo-second-order model is higher than the value of the pseudo-first-order model. Meanwhile, the calculated qe value of pseudo-second-order model (197.71 mg/g) was approximate to the value measured by the experiment (Qe,exp = 196.4 mg/g). It indicated that the adsorption process is more favor of the pseudo-second-order model, which implying that adsorption process could be chemisorption (Idris et al. 2013; Humelnicu et al. 2011).

Effect of pH

pH was demonstrated one of the most important parameters influencing the metal adsorption (Ghorbani et al. 2008; Gok and Aytas 2009). The uranium removal efficiency (R) and uranium sorption amount (Q) increased dramatically with an increasing pH from 2 to 5 as can been seen in Fig. 3. The removal efficiency of uranium at pH 2 was nearly 3.8%, and corresponding uranium sorption amount was 8.06 mg/g. When pH ascent from 2 to 5, the maximum uranium removal efficiency (R) of 84.6% and uranium sorption amount of 177.6 mg/g were observed at pH 5. The increasing positivity of the adsorbents, the electrostatic repulsion between positively charged uranyl cations, and high concentrations of H+ in the reaction mixture result in the low adsorption capacity at lower pH (Ai et al. 2013; Cao et al. 2013). With increasing pH, an increase of negative charges owes to the deprotonated on the Anabaena flos-aquae, finally enhanced the biosorption of positively charge uranium on the cell surface of Anabaena flos-aquae. However, when the pH was above 5, the biosorption of uranium decreased gradually. The decline of the uranium sorption efficiency may result from the formation of precipitate (4UO3·9H2O) when the pH is higher than 5 (Ghorbani et al. 2008; Khani 2011). Therefore, the optimum pH (pH = 5) for subsequent uranium adsorption on Anabaena flos-aquae experiments was selected.

Fig. 3
figure 3

The effect of pH on the uranium adsorption by Anabaena flos-aquae (C0 = 70 mg/L, m = 10 mg, V = 30 mL, T = 298.2 K, t = 60 min)

Adsorption isotherm

To illuminate the mechanism of the adsorption process, the Langmuir and Freundlich model analyses were conducted. Langmuir isotherm is commonly applied to describe monolayer adsorption (Foo and Hameed 2010; Ding et al. 2014). In contrast, the Freundlich model assumes multilayer sorption on the heterogeneous surface (Foo and Hameed 2010; Xiong et al. 2013). The two models can be expressed by Eq. (5) and Eq. (6).

$$ \mathrm{Langmuir}:{q}_e=\frac{q_m{k}_L{C}_e}{1+{k}_L{C}_e} $$
(5)
$$ \mathrm{Freundlich}:{q}_e={k}_F{C}_e^{1/{n}_F} $$
(6)

where Ce refers to the equilibrium concentration (mg/L) and qe refers to the equilibrium adsorption capacity(mg/g), qm is the maximum adsorption capacity(mg/g), and KL is the constant related to the energy of adsorption; KF and nF denote the Freundlich constant and Freundlich exponent, respectively.

The value of qe increased firstly and then reached equilibrium state later with the concentration increased, which revealed that the initial concentration provides a driving force for uranium adsorbed onto Anabaena flos-aquae (Han et al. 2018; Gok and Aytas 2009). The parameters presented in Table 2 were calculated from the linear curves presented in Fig. 4. The higher R2 value of the Langmuir isotherm model and the sorption capacity calculated from the Langmuir model (qm = 190.1 mg/g) which was the approach to the value measured by the experiment (168.5 mg/g) both indicate that the biosorption of uranium onto Anabaena flos-aquae can be well described by the Langmuir model. Based on the isotherm analysis, it was confirmed that the uranium adsorption process onto Anabaena flos-aquae was likely monolayer coverage.

Table 2 Adsorption isotherm parameters calculated with Langmuir and Freundlich models
Fig. 4
figure 4

Isotherm models of uranium biosorption by Anabaena flos-aquae. The solid lines are Freundlich model simulation, and the dashed lines are Langmuir model simulation

Notably, the biosorption capacity of adsorbents was always evaluated by the value of the maximum adsorption capacity (qm). For example, the qm for Dictyopteris polypodioides was 62.5 mg/g (Bampaiti et al. 2016), 113.5 mg/g for Saccharomyces cerevisiae (Faghihian and Peyvandi 2012), 152 mg/g for Cladophora hutchinsiae (Bagda et al. 2017), and 97.15 mg/g for Chlorella vulgaris (Amini et al. 2012). Evidently, a comparison of the maximum capacities among these adsorbents proved the Anabaena flos-aquae has a considerable high biosorption capacity for uranium. Besides, the low operational cost and maximization of benefits give the Anabaena flos-aquae more advantages over lots of adsorbents.

Adsorption thermodynamics

In order to figure out the influence of temperature on the adsorption reaction, the adsorption experiments at different temperatures (288.2 K, 298.2 K, and 308.2 K) were conducted. As presented in Fig. 5a, the amount of uranium adsorbed onto Anabaena flos-aquae increased gradually with the increase of temperature, which demonstrating the adsorption process may be endothermic. Thermodynamic parameters were calculated using the equations given below and listed in Table 3.

$$ \mathrm{Ln}{K}_d=\frac{\mathit{\Delta S}}{R}-\frac{\mathit{\Delta H}}{\mathrm{RT}} $$
(7)
$$ \Delta G=\Delta \mathrm{H}- T\mathit{\Delta S} $$
(8)
$$ {K}_d={q}_e/{C}_e $$
(9)

where Kd refers to the equilibrium constant and R represents gas constant (8.314 J mol−1 K−1); T refers to the absolute temperature in Kelvin (K). The enthalpy and entropy were calculated from the plot of lnKd against 1/T in Fig. 5b. The positive values of ΔH and ΔS proved the adsorption process was endothermic and the increased randomness at the solid-solution interface (Bozkurt et al. 2011; Saleem and Bhatti 2011; Donat 2009). Meanwhile, the negative value of ΔG suggests the adsorption process was spontaneous. In conclusion, the thermodynamic study revealed that the biosorption of uranium onto Anabaena flos-aquae was endothermic and spontaneous.

Fig. 5
figure 5

The effect of temperature on the uranium adsorption by Anabaena flos-aquae (a) and plot of lnKd against 1/T (b)

Table 3 Thermodynamic data of uranium adsorption onto Anabaena flos-aquae

Characterization of Anabaena flos-aquae

FTIR spectra of Anabaena flos-aquae

To characterize specific functional groups involved in biosorption, the FTIR patterns of Anabaena flos-aquae were recorded before and after sorption of uranium. For the Anabaena flos-aquae, the bands at approximately 3305.39 cm−1 corresponded to the O-H and N-H stretching vibrations. The peaks at 1654.62 cm−1 and 1544.7 cm−1 were ascribed to the N-H stretching of amino groups. The band at 1054.87 cm−1 were assigned to the C-OH stretching (Yi et al. 2016).

For the U-loaded Anabaena flos-aquae, the peaks at 1054.87, 1544.7, and 3305.39 cm−1 correspond to -COOH, -NH2, and -OH shifted to 1031.73, 1546.63, and 3313.11 cm−1, respectively (Fig. 6). It implied that the -COOH, -NH2, and -OH performed an important role in uranium biosorption. Furthermore, two new peaks at 1454.06 cm−1 and 916.02 cm−1 were observed. The two peaks represented the stretching frequency of the structure that formed by U atoms combined with oxygen-containing functional groups (Xie and Gao 2007; Zhang et al. 2014; Zhao et al. 2019).

Fig. 6
figure 6

The FTIR spectra of Anabaena flos-aquae before and after uranium adsorption

SEM-EDS of Anabaena flos-aquae

The SEM images of Anabaena flos-aquae before and after uranium loaded were shown in Fig. 7. As can be seen from the micrographs of raw Anabaena flos-aquae with magnification of × 2000 and × 3000 in Fig. 7a, b, the morphology of pure Anabaena flos-aquae is smooth and flat. As presented in Fig. 7c, d, the surface became rough because of the formation of many deep grooves after exposure to uranium. These grooves may give the Anabaena flos-aquae more chance to accommodate considerable uranyl ions.

Fig. 7
figure 7

SEM images of Anabaena flos-aquae before (a magnification of × 2000, b magnification of × 3000) and after adsorption(c magnification of × 2000, d magnification of × 3000); mapping of Anabaena flos-aquae after adsorption (e carbon, f oxygen, g uranium, h potassium, i calcium)

The element content of the sample before and after uranium sorption was analyzed by EDS. According to EDS analysis in Fig. 8a, b, the algae was mainly consisted of carbon (49%), oxygen (36%), and nitrogen (11%) as well as small quantities of sodium, magnesium, phosphorus, and calcium. After exposure to uranium, the uranium peak was detected (1.64%). Besides, the atomic percentage of nitrogen (3.87%) and oxygen (9.11%) decreased. The combination of SEM-EDS and FTIR data indicates that uranium has been successfully adsorbed by Anabaena flos-aquae.

Fig. 8
figure 8

EDS analysis of Anabaena flos-aquae before (a) and after (b) uranium adsorption

Conclusion

As a common species of cyanobacteria bloom, Anabaena flos-aquae was utilized for the biosorption of uranium in this study. The maximum uranium sorption amount was found to reach the value of 196.4 mg/g. The biosorption of uranium onto Anabaena flos-aquae was well fitted to the Langmuir isotherm model and pseudo-second-order model, which proved the adsorption process was monolayer sorption and chemisorption. The results of thermodynamic analysis indicated that uranium uptake was endothermic and spontaneous. The FTIR results demonstrated that the -OH, -NH2, and -COOH performed an important role in uranium biosorption. In conclusion, the effective biosorption of uranium from aqueous solution by cyanobacterium Anabaena flos-aquae makes it a promising, environmentally friendly, and biocompatible biosorbent for the treatment of uranium contaminants.