1 Introduction

Due to the extensive usage of global nuclear explosion test [1], nuclear power plant [2], and nuclear waste treatment [3], the release of radioactive wastewater has attracted worldwide attention. The radioactive wastewater does great harm to human health and the whole biosphere. Among them, cesium is a highly poisonous radionuclide that has a tremendous impact on environment for a long time [4,5,6]. Cs+ in the wastewater can eventually enter the food chain and be ingested by humans, it has become one of the most urgent issues to develop effective materials for highly efficient enrichment of cesium from radioactive wastewater.

An excellent adsorbent material must have some advantages, such as a large specific surface area, a good pore structure, a certain thermal stability, chemical stability, radiation resistance and so on [7]. Among many adsorbent materials, the ion exchange adsorbent can better meet the enrichment demand of marine radioactive cesium. The inorganic ion exchange sorbent has superiorities of good mechanical properties and radiation resistance, easy operation and more uniform ion exchange position. The adsorption selectivity for Cs+ is high, and there are many researches on the treatment of high-level liquid waste [8]. The inorganic ion exchange adsorbents used for the adsorption of Cs+ in seawater mainly include inorganic minerals [9], polyvalent metal phosphates [10], and metal ferrocyanides [11,12,13].

Although inorganic minerals can be used for the adsorption of radioactive Cs+ in water, it mostly takes hours or even days to complete the adsorption equilibrium of inorganic mineral adsorbing materials, so it is difficult to meet the timeliness requirements for rapid enrichment detection of marine radionuclides [14]. Multivalent metal phosphates, such as phosphopeptides [15], zirconium phosphates [16], etc., have excellent physicochemical properties and ion exchange capacity, and have good selective adsorption characteristics for alkali metal ions. However, polyvalent metal phosphate ion exchange adsorption materials also have the problem of being susceptible to coexisting ions. Because seawater contains a large amount of cations such as K+ and Na+, it directly affects the adsorption efficiency and applicable field of Cs+ in seawater [17, 18].

Metal ferrocyanide (Prussian blue compound) has a cubic crystal form and the interior is a cavity structure, the different interstitial ions in the crystal lattice are classified into two types, soluble and insoluble [19,20,21,22,23]. The soluble metal ferrocyanide formula can be expressed as KnM[Fe(CN)6], and its water dispersibility is superior to the insoluble metal ferrocyanide. Among them, the transition metal ferrocyanide has extremely high chemical stability, its lattice and hydrated Cs+ size match, and strong adsorption for Cs+, which is a kind of highly promising radioactive Cs+ adsorbent. Among the ferrocyanides of different transition metals, zinc, copper and nickel ferrocyanide have better adsorption capacity for Cs+ [24,25,26,27,28].

Here, copper ferrocyanide powder was synthesized by coprecipitation. The prepared CuFC powder was used as an adsorbent to adsorb Cs+ ions in radioactive wastewater, by batch sorption technique to assess the adsorbability of CuFC from radioactive wastewater [29,30,31,32]. The sorbent was characterized by X-ray diffractometry XRD, infrared spectroscopy FT-IR, scanning electronmicroscope (SEM), energy dispersive spectrometer (EDS) and thermogravimetric analysis (TGA). In addition, batch adsorption experiments were carried out to evaluate the adsorption behavior under different experimental conditions, such as adsorbent dosage, temperature, contact time, solution pH, initial cesium concentration, and the influence of competitive ions(K+, Na+, Ca2+ and Mg2+). Finally adsorption kinetics and thermodynamic studies were performed to understand the adsorption mechanism.

2 Experimental section

2.1 Preparation of CuFC powder

Synthesis of CuFC: The Cu(NO3)2 (0.0776 mol/L) solution was slowly added to the K4Fe(CN)6 (0.1 mol/L) solution under magnetic stirring at 2000 rpm. A dark red precipitate was immediately produced and stirring was continued for 5 min to form a stable suspension. The suspension was placed in a 50 mL sample tube and then centrifuged at 8500 rpm for 10 min. After removing the supernatant, the resulting reddish brown precipitate was washed with deionized water. This process was repeated 2–3 times until the supernatant was clear.

2.2 Characterization of CuFC powder

The surface morphology and structure of CuFC powder were characterized systematically using the scanning electron microscope (SEM–EDS, JSMIT300, Japan Electronics Corporation). Fourier transform infrared spectrometer (FT-IR, Nicolet 6700, America) was used to recorded from 400 to 4000 cm−1. X-ray diffractometry patterns (XRD) of copper ferrocyanide powder before and after adsorption were acquired using a X-ray diffractometer (RU-200B/D/MAX-RB, Japan). Thermogravimetric Analysis (TGA, STA449F3) measurements were conducted by using a TGA thermogravimetric analyzer from 38 to 600 °C at a heating rate of 10 °C.

2.3 Sorption experiments

Cs+ sorption on CuFC was performed by batch sorption technique at 298 K. The sorbents and adsorbates solution were added in polyethylene tubes to achieve required concentrations of various constituents. The suspension was adjusted to required pH values by adding trace amounts of 0.1 or 0.01 mol/L HCl or NaOH, and then shaken for 3 h to get fully sorption equilibrium. After that, sorbents were separated from liquid phases by a centrifugal separation method. The concentration of Cs+ was determined by atomic absorption spectrophotometric method. Sorbed amounts of Cs+ onto CuFC were calculated from the difference between the starting concentration (C0, mg/L) and the balance concentration (Ce, mg/L) (qe = (C0V0− CeVe)/m and η = (C0V0 − CeVe) × 100%/C0V0, where qe is the adsorption capacity and η is removal rate, m (g) is the mass of CuFC, V0 is the initial volume of solution, Ve is the diluted solution volume. All the experimental data were the average of duplicate determinations, and the average uncertainties were < 5%.

3 Results and discussion

3.1 Effect of sorbent content

Cs+ sorption onto CuFC at various solid contents is shown in Fig. 1. The maximum adsorption amount appeared at 0.4 g/L. At this time, the adsorption rate of CuFC to Cs+ was 58.12%, and the adsorption amount was 15.2 mg/g, cesium distribution coefficient was lgKd = 3.5 ± 1.0 ml/g. Compared with other types of ferrocyanide in Table 1, copper ferrocyanide can effectively enrich cesium ions from low concentration wastewater (the initial concentration 10 mg/L) especially when the dosage is 2.0 g/L, the adsorption rate reaches 96.74%。 The reason for the above phenomenon may be that as the CuFC powder in the system increases, the active adsorption sites increase, and more Cs+ is adsorbed on the surface of the adsorbent, so the adsorption rate increases. However, since the total amount of Cs+ is constant and the adsorbents collide and aggregate after increasing, the active site of the unit mass adsorbent decreases, so the unit adsorption amount also decreases [36]. For practical application in radioactive wastewater treatment, the optimal dosage for CuFC under similar condition is 2.0 g/L.

Fig. 1
figure 1

Effect of CuFC dosage on Cs+ adsorption (v = 25 mL, T = 298 K, pH = 7)

Full size image
Table 1 Comparison of different types of ferrocyanide
Full size table

3.2 Effect of initial pH on Cs+ sorption

The sorption of Cs+ on CuFC was greatly influenced by the solution pH (Fig. 2). The sorption increased from about 76 to 99.3% when the pH value increased from 1.0 to 7.0, and maintained the high level at pH 7.0–9.0, then decreased steeply from 99.3 to 84.5% in the pH range of 9.0–13.0. The reason for the low adsorption performance under acidic conditions(pH = 1–5) was mainly due to the competitive adsorption between H+ and Cs+ in the solution, occupying the adsorption site, which greatly affected the adsorption process. Under weak alkaline or neutral conditions (pH = 7–9), the H+ competitive adsorption was weakened, and the adsorption capacity of CuFC to Cs+ was higher. With the increase of alkalinity(pH > 9), the OH− in the solution increases continuously, and there was a strong force between Fe3+ and OH−, which caused the Fe–CN–Cu bond in the CuFC to break and the material structure was destroyed, so the adsorption effect was lowered.

Fig. 2
figure 2

Effect of pH of reaction system on adsorption of Cs+ (v = 25 mL, m = 0.05 g, T = 298 K)

Full size image

3.3 Effect of sorbent contact time

Figure 3 showed the effect of contact time on Cs+ adsorption. During the whole adsorption process, the adsorption rate was the fastest in the first 50 min, and the adsorption capacity increased significantly. After 50 min, the adsorption gradually became equilibrium. At this time, the adsorption amount and adsorption rate of Cs+ were about 4.9 mg/g and 94.70%, respectively.

Fig. 3
figure 3

Effect of contact time on adsorption of Cs+ (v = 25 mL, m = 0.05 g, T = 298 K, pH = 7)

Full size image

This was because in the initial stage of adsorption, since the adsorbent material had more active sites, Cs+ was rapidly adsorbed to the active site. In the middle of adsorption, as the active site was continuously occupied, the Cs+ adsorbed on the material had a resistance to the diffusion of Cs+ in the solution, which make it difficult for Cs+ to diffuse into the interior of the particle, so the adsorption process was slow. In the late stage of adsorption, when the active sites were all occupied by Cs+, the adsorption reached equilibrium.

3.4 Effect of initial concentration of Cs+

It can be seen apparently from Fig. 4 that the adsorption rate of CuFC to Cs+ was increased with the Cs+ initial concentration and then reached equilibrium. This was because the adsorption of Cs+ by CuFC was not saturated at low concentration, the contact probability of adsorbent with Cs+ was small, and under the effect of increasing concentration driving force, when the concentration increased, the contact between adsorbent and Cs+ were increased. The amount of adsorption of Cs+ on the material was increasing. Thereafter, the initial concentration of the solution was further increased, and since the adsorbent material contained a certain number of active sites, the adsorption capacity for Cs+ was also limited. At this time, the initial concentration of the solution was increased, and the active sites of the adsorbent material were all occupied, so the adsorption amount hardly changed, and the adsorption tended to be saturated.

Fig. 4
figure 4

Effect of initial concentration on adsorption of Cs+ (v = 25 mL, m = 0.05 g, T = 298 K, pH = 7)

Full size image

3.5 Effect of coexisting ions

Interfering ions can affect the presence and activity of metal ions. The effect of common cations K+, Na+, Ca2+ and Mg2+ on the adsorption of Cs+ by CuFC was shown in Fig. 5. Among them, K+ and Na+ showed a slight inhibitory effect on Cs+ adsorption. With the increased of K+ and Na+ ionic strength, the adsorption capacity and adsorption rate of Cs+ were decreased slightly, and the inhibitory effect of K+ was slightly stronger than Na+. When the concentration was increased from 0 to 400 mg/L, the adsorption amount of Cs+ decreased by 0.16 mg/g, and the adsorption rate decreased by 4.6%. Under the same conditions of Na+, the adsorption amount of Cs+ decreased by only 0.11 mg/g, and the adsorption rate decreased by 3.5%. The presence of Ca2+ and Mg2+ hardly inhibited the adsorption of Cs +. This may be caused by the difference in the hydration radius of the ions. closer to the Cs+ hydration radius, the stronger the competitive adsorption capacity. (Cs+ hydration radius is 3.25 Å, K+ and Na+ are close to 3.3 Å, 3.6 Å, Ca2+ is 4.1 Å, and Mg2+ is 4.25 Å [37]). K+ and Cs+ had similar ionic radii and hydration energy. During the adsorption process, K+ will compete with Cs+ for a limited adsorption site, resulting in a decrease of adsorption capacity, but this inhibition was very limited. CuFC had good anti-interference in salty solutions.

Fig. 5
figure 5

Effect of common cations on adsorption of Cs+ (v = 25 mL, m = 0.05 g)

Full size image

3.6 Adsorption kinetics

In order to clarify the sorption process of cesium in CuFC, the pseudo-first-order kinetic model and the pseudo-second-order kinetic model were applied to analysis the experimentally observed kinetic data, their linear form can be expressed as:

First order kinetic model:

$${ln(q}_{e}-{q}_{t})=ln{q}_{e}-{k}_{1}t$$
(1)

Second order kinetic model:

$$\frac{t}{{q}_{t}}=\frac{t}{{q}_{e}}+\frac{1}{{k}_{2}{q}_{e}^{2}}$$
(2)

where qe is the sorption amount at equilibrium (mg/g), qt is the sorption amount at t (mim), k1 is the rate constant (min−1), and k2 is the rate constant of second-order Kinetic model (g mg−1 min−1).

Figure 6 showed that the first-order kinetic model (R2 = 0.54) and the pseudo-secondary kinetic model (R2 = 0.999) were used to research the reaction kinetics of cesium on CuFC. In the adsorption process, the fitting effect of the first-order kinetic model was not good, and the linear fitting correlation coefficient R2 was not high; However, the pseudo-secondary kinetic model has a good fitting effect (R2 = 0.999), and the relative error between the theoretical adsorption value and the experimental value of Cs+ was small. It showed that the adsorption of Cs+ by CuFC was a heterogeneous diffusion process, and there were many reaction mechanisms, not a simple first-order reaction.

Fig. 6
figure 6

Adsorption kinetic model fitting for adsorption of Cs+ on CuFC (v = 25 mL, m = 0.05 g)

Full size image

3.7 Adsorption isotherm

Adsorption isotherm are mathematical models that describe how adsorbates interact with adsorbents. In this study, all the experimental date were analyzed according to the two common nonlinear: Freundlich and Langmuir. The Langmuir model in Fig. 7a shown that the maximum adsorption capacity of CuFC for cesium was 11.12 mg/g, and the RL was 0.610 (0 < RL < 1) when the initial concentration was 10 mg/L, suggesting that CuFC had higher affinity for cesium and also it was beneficial to promote the reaction. The Freundlich model in Fig. 7b shown that the characteristic constant n was 0.72 (1 < n < 10), indicating that the adsorption of Cs+ by CuFC does not conform to the Freundlich model.

Fig. 7
figure 7

Adsorption isotherm model of CuFC for Cs+ at 25 °C (v = 25 mL, m = 0.05 g)

Full size image

Table 2 listed the separation factor RL of Cs+ adsorbed by CuFC at different concentrations at 298 K, and the values were all between 0 and 1, indicating that the adsorption of Cs+ by CuFC belonged to preferential adsorption (favorable adsorption).

Table 2 Separation factor of Cs+ (RL)
Full size table

3.8 Characterization of CuFC powder

3.8.1 FTIR spectra of CuFC

Figure 8 showed the FTIR spectrum of EPS before and after adsorption. At 2090 cm−1, there was a sharp peak of –C≡N, which showed a blue shift compared to the characteristic peak of free cyano (The free cyano characteristic peak usually appears at 2600 cm−1), This was because the cyano group was combined with metal ions, which provide electrons to the metal ions [38]. The sharp characteristic peak appearing at around 592 cm−1 was the deformation vibration absorption peak of –Fe(II)–C≡N, and the characteristic peak appearing at around 484 cm−1 was the deformation vibration absorption peak of –C≡N–Cu(II). The characteristic peaks appearing in the above three points showed that the synthesized powder had obvious characteristic peaks of CuCF, and the deformation vibration in the plane of Fe(II)–C≡N–Cu(II) was formed, and the CuCF powder was successfully synthesized. In addition, at 3425 cm−1, a broad and strong peak was the stretching vibration peak of O–H, and the absorption peak at 1611 cm−1 was the stretching vibration peak of O–H. These two characteristic peaks may be attributed to the presence of water molecules in the material gap.

Fig. 8
figure 8

FT-IR spectras of CuFC before and after adsorption of Cs+

Full size image

3.8.2 XRD spectra of CuFC

Figure 9 showed the XRD spectrum of CuFC before and after adsorption. The synthesized CuFC powder had a sharp diffraction peak at 2θ of 10°–60°, indicating that the compound had good crystallinity. Both the two samples displayed six characteristic peaks at 2θ = 17.7°, 25.2°, 35.7°, 40.4°, 44.3°, 51.9°, corresponding to 200, 220, 400, 420, 424, 440 diffractive surfaces. The pattern was consistent with the standard pattern of copper ferrocyanide (JCPDS card no. 02–0381), and it can be determined that the synthesized powder was CuFC. In addition, the diffraction peak appeared at 55.7° of the XRD pattern may be an impurity peak remaining during the synthesis of copper ferrocyanide.

Fig. 9
figure 9

XRD spectras of CuFC before and after adsorption of Cs+

Full size image

3.8.3 FE-SEM and EDS image of CuFC

Figure 10 showed the SEM spectra of CuFC before and after adsorption. The shape of CuFC before and after adsorption was irregular, the size was about 50–200 nm and there was serious agglomeration. The surface of the CuFC before adsorption was rough, which was favorable for the adsorption point exposure to promote adsorption. After adsorption of Cs+, the surface porosity of the material decreased, and the overall tended to be regular. In the scanning electron micrographs before and after adsorption, cubic CuFC with sharp edges and corners can be distinguished, and the particle size was about several tens of nanometers.

Fig. 10
figure 10

CuFC powders with different magnification before and after adsorption (a, c before adsorption; b, d after adsorption)

Full size image

As shown in Fig. 11a that the surface of CuFC was detected to contain elements such as N, K, Fe, and Cu and so on before adsorption, but Cs was not detected. C was not detected because the conductive paste used contains C element, and the scanning signal of C in CuFC coincided with the signal of C in the conductive paste. After adsorption, it can be clearly seen from Fig. 11b that the orange-red Cs signal appeared on the surface of CuFC. In Fig. 12 showed that the content of the K decreased as 2.75%, and the content of Cs was increased by 2.61%. The content of other elements before and after adsorption had not changed significantly. The results were consistent with previous studies [39]. It can be speculated that the adsorption of Cs+ by CuFC may be mainly caused by ion exchange between K+ and Cs+ inside the structure. The main adsorption mechanism is shown in Fig. 13.

Fig. 11
figure 11

EDS layer diagram of CuFC before and after Cs+ adsorption

Full size image
Fig. 12
figure 12

EDS total spectrum of CuFC before and after Cs+ adsorption

Full size image
Fig. 13
figure 13

Main adsorption mechanism of Cs+ by CuFC

Full size image

3.8.4 TGA-DTG curvers of CuFC

The TG curve of copper ferrocyanide showed a gradient change (Fig. 14), which can be roughly divided into four stages: I (38–155 °C), II (155–220 °C), III (220–320 °C), IV (320–600 °C). The copper ferrocyanide structure containe two kinds of water: one is the structurally adsorbed zeolite water, and the other is the coordination water in the structure [40]. In stage I, the adsorbed water, zeolite water, etc. on the surface of the material were evaporated and the mass loss was about 20%; the mass loss in stage II was 11%, which may be attributed to the release of the coordination water in the material structure. The above two-stage temperature rise did not cause a change in the crystal structure of the material, and the mass loss was actually a water loss process of the material. The temperature in stage III, the crystal structure of copper ferrocyanide was broken, and C≡N was broken and decomposed [41], causing about 2% mass loss. The temperature in the IV stage continued to rise, and intermediate products due to structural damage of the material are produced. About 5% of the mass loss may be caused by the escape of gases such as CO2 and NO2 [42].

Fig. 14
figure 14

TGA-DTG curves of CuFC

Full size image

4 Conclusions

CuFC was prepared by co-precipitation method. The result of FTIR of the obtained product showed obvious absorption peak of CuFC. The XRD pattern was consistent with the standard spectrum, indicating that CuFC was successfully synthesized. The results obtained from XRD, FTIR, SEM and EDS showed that the appearance, chemical composition and atomic bonding mode of CuFC before and after adsorption did not change significantly. The mass loss of K element was basically consistent with the increase of mass of Cs element, while the mass fraction of other elements such as N, Fe and Cu did not change significantly. CuFC adsorption on Cs+ was caused by ion exchange between K+ and Cs+ in its structure. TGA analysis showed that CuFC had good thermal stability.