Radiation Resistance and Adsorption Behavior of Aluminum Hexacyanoferrate for Pd

Irradiation resistance is important for adsorbents used in radioactive environments such as high-level liquid waste. In this work, a silica-based composite adsorbent (KAlFe(CN)6/SiO2) was synthesized and γ-irradiated from 10 to 1000 kGy. The angles of the main X-ray diffraction peaks slightly decreased with the increase in irradiation dose, and a minor decomposition of CN− occurred after irradiation to 1000 kGy, indicating that the KAlFe(CN)6/SiO2 adsorbent could preserve structural integrity with a dose below 100 kGy. In 1 to 7 M HNO3, the adsorption ability of the irradiated KAlFe(CN)6/SiO2 remained performant, with a higher Kd than 1625 cm3 g−1. The adsorption equilibrium of Pd(II) in 3 M HNO3 was attained within 45 min before and after irradiation. The maximal adsorption capacity Qe of the irradiated KAlFe(CN)6/SiO2 on Pd(II) ranged from 45.1 to 48.1 mg g−1. A 1.2% relative drop in Qe was observed after 100 kGy irradiation, showing that γ-irradiation lower than 100 kGy insignificantly affected the adsorption capacity of KAlFe(CN)6/SiO2. Calculating and comparing the structures and free energies of different adsorption products via the density functional theory (DFT) method showed that KAlFe(CN)6/SiO2 was more inclined to completely adsorb Pd(II) and spontaneously generate Pd[AlFe(CN)6]2.


Introduction
The development of nuclear energy has brought evident benefits in reducing carbon emissions and promoting human health [1,2], although the disposal of spent highly radioactive fuel discharged from nuclear power plants is complex and challenging [3][4][5] partly because of the limited miscibility of platinum group metals (PGMs) [6]. The plutonium uranium extraction (PUREX) process is commonly used to reprocess spent fuel and produces approximately 500 L of radioactive high-level liquid waste (HLLW) per ton of spent fuel. HLLW is dangerous, and has a complex chemical composition and radiation levels of up to 1000 Ci/L [7]. If HLLW is released into the biosphere, radioactive and heavy-metal elements may migrate to the soil and plants, destroying the genetic material of organisms and human organs [8,9]. Therefore, the proper disposal of HLLW, such as by separating high-heat radionuclides, is necessary to protect humans and the biosphere [10][11][12].
Glass solidification is widely used to form a solidified product for HLLW that can envelop radioactive elements and reduce waste volume. This is a type of technology that dissolves the ions in HLLW into amorphous borosilicate glass or phosphate glass to form a new, stable structure. Some studies indicated that the separation of Pd is beneficial for the safety of glass solidification [13][14][15]. Palladium has a low solubility of 0.3 mg g −1 in glass, facilitating precipitation during glass solidification [16]. The precipitated substance has high density, viscosity, and electric conductivity, leading to problems such as blocking the outlet and short-circuiting the electrodes in the melter, threatening production safety. In addition, palladium is a precious strategic resource with distinctive properties that is widely where C 0 is the concentration of Pd(II) in the preadsorption solution, C e represents the equilibrium concentration, V is the volume of the Pd(II) solution, and m is the mass of the KAlFe(CN) 6 /SiO 2 adsorbent. The experimental adsorption capacity of Pd(II) (Q e ) is defined as follows:

DFT Calculations
In exploring the adsorption mechanism of Pd(II) by KAlFe(CN) 6 , the optimized structures and free energies of the adsorbent and adsorption products were calculated on the basis of DFT. Calculations were performed via VASP 6.2. The modeling and visualization of the hexacyanoferrates were conducted in VESTA [33]. Considering that the adsorption capacity may have been different, the Pd(II) products with adsorption rates of 100% (Pd[AlFe(CN) 6 ] 2 ) and 50% (Pd 0.5 K[AlFe(CN) 6 ] 2 ) were calculated. The crystal parameters and free-energy changes caused by the difference in Pd(II) adsorption sites in Pd 0.5 K[AlFe(CN) 6 ] 2 were compared. The generalized gradient approximation (GGA) method was selected as the wavefunction optimization potential. Calculations were performed in the inverted space by using the Perdew-Burke-Ernzerhof (PBE) functional [34].

XRD Analysis before and after Irradiation
In studying whether irradiation changes the crystal structure of KAlFe(CN) 6 , XRD characterization was performed on the KAlFe(CN) 6 powder sample after irradiation ( Figure 1). Table 1 shows the angle of incidence and lattice parameters of the KAlFe(CN) 6 powder before and after irradiation. As shown in Figure 1 and Table 1, the lattice of KAlFe(CN)6 expanded slightly after irradiation for 10, 50, 100, or 1000 kGy. The angles of the three main diffraction peaks slightly decreased with the increase in irradiation dose from 17.672°, 25.028°, and 36.051° to 17.482°, 24.840°, and 35.746°, respectively. Nevertheless, the strength and shape of the peaks in the spectrum remained similar throughout irradiation, indicating that irradiation did not cause a notably change in the substance, and the structure of the crystal remained stable. The results can be explained by the action mechanism of irradiation. The effect of γ rays is identical with that of β rays except for the thermal effect [35]. Under γ-ray irradiation, free electrons were formed in the KAlFe(CN)6 structure because of the photoelectric and Compton effects. The electrons interacted with the adsorbent and caused atomic displacements, forming defects such as dislocation loops [36,37].
Further analysis showed that, after irradiation at 1000 kGy, the lattice parameter of the KAlFe(CN)6 adsorbent increased slightly from 10.0155 to 10.0464 Å, and the unit cell remained face-centered cubic. Thus, the irradiation dose from 10 to 1000 kGy hardly affects the lattice structure and irregularity of KAlFe(CN)6, and the radiation resistance of the adsorbent was excellent.

SEM and FT-IR Analysis before and after Irradiation
The SEM images of KAlFe(CN)6/SiO2 before and after irradiation are shown in Figure  2. The adsorbent particle was spherical with a diameter of about 100 µm before irradiation. It maintained that shape and surface morphology with some minor cracks forming after 50 and 1000 kGy of γ irradiation. The adsorbents did not collapse or disintegrate after irradiation, indicating that KAlFe(CN)6/SiO2 had good radiation resistance.  As shown in Figure 1 and Table 1, the lattice of KAlFe(CN) 6 expanded slightly after irradiation for 10, 50, 100, or 1000 kGy. The angles of the three main diffraction peaks slightly decreased with the increase in irradiation dose from 17.672 • , 25.028 • , and 36.051 • to 17.482 • , 24.840 • , and 35.746 • , respectively. Nevertheless, the strength and shape of the peaks in the spectrum remained similar throughout irradiation, indicating that irradiation did not cause a notably change in the substance, and the structure of the crystal remained stable. The results can be explained by the action mechanism of irradiation. The effect of γ rays is identical with that of β rays except for the thermal effect [35]. Under γray irradiation, free electrons were formed in the KAlFe(CN) 6 structure because of the photoelectric and Compton effects. The electrons interacted with the adsorbent and caused atomic displacements, forming defects such as dislocation loops [36,37].
Further analysis showed that, after irradiation at 1000 kGy, the lattice parameter of the KAlFe(CN) 6 adsorbent increased slightly from 10.0155 to 10.0464 Å, and the unit cell remained face-centered cubic. Thus, the irradiation dose from 10 to 1000 kGy hardly affects the lattice structure and irregularity of KAlFe(CN) 6 , and the radiation resistance of the adsorbent was excellent.

SEM and FT-IR Analysis before and after Irradiation
The SEM images of KAlFe(CN) 6 /SiO 2 before and after irradiation are shown in Figure 2. The adsorbent particle was spherical with a diameter of about 100 µm before irradiation. It maintained that shape and surface morphology with some minor cracks forming after 50 and 1000 kGy of γ irradiation. The adsorbents did not collapse or disintegrate after irradiation, indicating that KAlFe(CN) 6 /SiO 2 had good radiation resistance.  [38]. The peak at 554 cm −1 was assigned to the C-Fe stretching vibration [39]. The peaks at 2126 and 2078 cm −1 correspond to the C≡N stretching vibration [38,40], indicating the presence of hexacyanoferrate. The peak of CN − weakened with the increase in irradiation dose, which indicates that the γ ray destroyed a minor portion of CN − [41,42]. The above analysis shows that KAlFe(CN)6/SiO2 could preserve structural integrity under up to 1000 kGy of irradiation.

Adsorption Properties in Nitric Acid before and after Irradiation
The adsorption performance of the irradiated KAlFe(CN)6/SiO2 was investigated. Figure 4 shows the adsorption effects of different concentrations of HNO3 on Pd(II) after receiving 10 to 1000 kGy of irradiation. Distribution coefficient Kd was slightly reduced The FT-IR spectra of KAlFe(CN) 6 /SiO 2 before and after irradiation are shown in Figure 3. The peak at 3430 cm −1 was attributed to the O-H stretching vibration peak, and 1621 cm −1 was the H-O-H bending vibration peak [38]; both came from crystal water. The peaks at 1110 and 801 cm −1 were assigned to the Si-O stretching vibration in SiO 2 , and 473 cm −1 corresponds to the inplane bending vibration of O-Si-O [38]. The peak at 554 cm −1 was assigned to the C-Fe stretching vibration [39]. The peaks at 2126 and 2078 cm −1 correspond to the C≡N stretching vibration [38,40], indicating the presence of hexacyanoferrate. The peak of CN − weakened with the increase in irradiation dose, which indicates that the γ ray destroyed a minor portion of CN − [41,42]. The above analysis shows that KAlFe(CN) 6   The FT-IR spectra of KAlFe(CN)6/SiO2 before and after irradiation are show ure 3. The peak at 3430 cm −1 was attributed to the O-H stretching vibration peak, a cm −1 was the H-O-H bending vibration peak [38]; both came from crystal water. Th at 1110 and 801 cm −1 were assigned to the Si-O stretching vibration in SiO2, and corresponds to the inplane bending vibration of O-Si-O [38]. The peak at 554 c assigned to the C-Fe stretching vibration [39]. The peaks at 2126 and 2078 cm −1 cor to the C≡N stretching vibration [38,40], indicating the presence of hexacyanoferr peak of CN − weakened with the increase in irradiation dose, which indicates that t destroyed a minor portion of CN − [41,42]. The above analysis shows that KAlFe(C could preserve structural integrity under up to 1000 kGy of irradiation.

Adsorption Properties in Nitric Acid before and after Irradiation
The adsorption performance of the irradiated KAlFe(CN)6/SiO2 was inves Figure 4 shows the adsorption effects of different concentrations of HNO3 on Pd receiving 10 to 1000 kGy of irradiation. Distribution coefficient Kd was slightly

Adsorption Properties in Nitric Acid before and after Irradiation
The adsorption performance of the irradiated KAlFe(CN) 6 /SiO 2 was investigated. Figure 4 shows the adsorption effects of different concentrations of HNO 3 on Pd(II) after receiving 10 to 1000 kGy of irradiation. Distribution coefficient K d was slightly reduced after irradiation, and the magnitude of the drop decreased with the increase in HNO 3 concentration. In 5 and 7 M HNO 3 concentration, K d remained almost the same from 0 to 100 kGy. In 1 to 3 M HNO 3 concentration, K d was highly sensitive to the adsorption rate because the rate was high. In addition, K d remained high after 1000 kGy of irradiation, ranging from 1625 to 38,445 cm 3 g −1 , indicating that the adsorption ability of the irradiated KAlFe(CN) 6 /SiO 2 remained performant in 1 to 7 M HNO 3 .
oxics 2023, 11, x FOR PEER REVIEW after irradiation, and the magnitude of the drop decreased with the increase centration. In 5 and 7 M HNO3 concentration, Kd remained almost the same kGy. In 1 to 3 M HNO3 concentration, Kd was highly sensitive to the adso cause the rate was high. In addition, Kd remained high after 1000 kGy of irra ing from 1625 to 38,445 cm 3 g −1 , indicating that the adsorption ability of KAlFe(CN)6/SiO2 remained performant in 1 to 7 M HNO3. The adsorption rate is an important factor in judging adsorption perfor 5 shows the effects of contact time on Pd(II) adsorption. The adsorption KAlFe(CN)6/SiO2 adsorbent for Pd(II) rapidly increased with time. When th reached 15 min, the adsorption ratios of Pd(II) were over 87%, 86%, and 81% 1000 kGy irradiated adsorbents, respectively, indicating excellent adsorp Subsequently, the adsorption rate of Pd(II) slowed down because KAlFe(CN)6/SiO2 tended to be saturated. After 45 min, the adsorption KAlFe(CN)6/SiO2 reached equilibrium, and the equilibrium adsorption rate 99.8%, and 99.8%, respectively.
Further, the pseudo-first-order kinetic equation (Equation (3)) and ps order kinetic equation (Equation (4)) were used for fitting to explore the a control steps of KAlFe(CN)6/SiO2 on Pd(II) [43]. The adsorption rate is an important factor in judging adsorption performance. Figure 5 shows the effects of contact time on Pd(II) adsorption. The adsorption ratios of the KAlFe(CN) 6 /SiO 2 adsorbent for Pd(II) rapidly increased with time. When the contact time reached 15 min, the adsorption ratios of Pd(II) were over 87%, 86%, and 81% for 0, 50, and 1000 kGy irradiated adsorbents, respectively, indicating excellent adsorption kinetics. Subsequently, the adsorption rate of Pd(II) slowed down because the sites on KAlFe(CN) 6 /SiO 2 tended to be saturated. After 45 min, the adsorption of Pd(II) by KAlFe(CN) 6 /SiO 2 reached equilibrium, and the equilibrium adsorption rates were 99.9%, 99.8%, and 99.8%, respectively.
where k 1 and k 2 are fitted parameters related to the adsorption rate. The pseudo-second-order equation fitted better than the pseudo-first-order equation ( Figure 5). For the pseudo-first-order equation, the correlation coefficients were lower than 0.95; the values of the pseudo-second-order equation, on the other hand, were close to 1, indicating good linearity (Table 2). Moreover, the Q e values of KAlFe(CN) 6 /SiO 2 for Pd(II) obtained with the pseudo-second-order equation were 20.95, 20.82, and 20.61 mg g −1 for the irradiated adsorbents with 0, 50, and 1000 kGy, respectively, which values were close to the experimental data. k 2 shows that the adsorption rate remained approximately equal before and after 50 kGy irradiation. These results indicate that the pseudo-second-order equation could describe the adsorption kinetics of KAlFe(CN) 6 /SiO 2 well at different nitric acid concentrations, and the adsorption type of Pd(II) through the KAlFe(CN) 6 /SiO 2 adsorbent was controlled via chemical adsorption.
where Q max is the maximal adsorption capacity predicted by the model; K L , 1⁄n, K R , K P , and g are the fitting parameters. As shown in Figure 6 and Table 3, the coefficients of determination R 2 of the Langmuir and Redlich-Peterson models were higher than 0.96, and the fitted parameter g in the Redlich-Peterson model was between 0.96 and 0.995, indicating that the Redlich-Peterson model could be converted into the Langmuir model in these cases. With a smaller irradiation dose than 1000 kGy, the adsorption of KAlFe(CN) 6 /SiO 2 on Pd(II) corresponded to the Langmuir model, suggesting that the adsorption sites on KAlFe(CN) 6 /SiO 2 were on a single layer, and the adsorbed Pd(II) had no internal interaction [44]. As shown in Figure 6 and Table 3, the coefficients of determination R of the Langmuir and Redlich-Peterson models were higher than 0.96, and the fitted parameter g in the Redlich-Peterson model was between 0.96 and 0.995, indicating that the Redlich-Peterson model could be converted into the Langmuir model in these cases. With a smaller irradiation dose than 1000 kGy, the adsorption of KAlFe(CN)6/SiO2 on Pd(II) corresponded to the Langmuir model, suggesting that the adsorption sites on KAlFe(CN)6/SiO2 were on a single layer, and the adsorbed Pd(II) had no internal interaction [44]. The fitted results of Equation (5) show that, after irradiation with 10, 50, 100, and 1000 kGy, the maximal adsorption capacities Q e of KAlFe(CN) 6 /SiO 2 on Pd(II) were 48.1, 47.8, 47.7, and 45.1 mg g −1 , respectively. The relative decrease in the unirradiated sample was only 0.4%, 1.0%, 1.2%, and 6.2%, respectively. The results indicate that the adsorption capacity of KAlFe(CN) 6 /SiO 2 was insignificantly influenced by the irradiation dose and remained virtually unchanged after no more than 100 kGy of irradiation. Thus, KAlFe(CN) 6 /SiO 2 could be used for the separation and recovery of Pd(II) under radioactive conditions.
As shown in Table 4, compared with other adsorbents, the adsorption capacity of KAlFe(CN) 6 /SiO 2 was higher with shorter equilibrium time.

DFT Calculations
The DFT method was applied to analyze the adsorption products of Pd(II) by KAlFe(CN) 6 . Stable structures with different adsorption rates and positions of adsorption sites were calculated. Figure 7a illustrates the optimized structure of the Pd[AlFe(CN) 6 ] 2 (S1) product of saturated adsorption. Figure 7b-f show the optimized structures of the Pd 0.5 K[AlFe(CN) 6 ] 2 product with an adsorption rate of 50%, namely, S2-1 to S2-5, for different adsorption sites. Table 5 lists the lattice parameters of these optimized structures, proving that all six types of products had FCC cells with similar volumes, where the cell of the saturated adsorption of product S1 was a cube, whereas unsaturated product S2 was a cuboid because of its asymmetry in the position of K and Pd atoms. of the saturated adsorption of product S1 was a cube, whereas unsaturated product S2 was a cuboid because of its asymmetry in the position of K and Pd atoms.  On the basis of the optimized stable structures before and after adsorption, the Helmholtz free energy (F) values were calculated. Thermodynamic calibration was performed via Phonopy [51] and VASPKIT [52], and the change in free energy (ΔF) was calculated as  On the basis of the optimized stable structures before and after adsorption, the Helmholtz free energy (F) values were calculated. Thermodynamic calibration was performed via Phonopy [51] and VASPKIT [52], and the change in free energy (∆F) was calculated as where F represents the free energy, m is the number of Pd atoms, and n is the number of K atoms. The results are illustrated in Figure 8, where the free-energy changes of S1 are in the line chart and those of S2s are in the column chart. The ∆F of S1 was lower than that of S2, from 273.15 to 373.15 K, showing that S1 was spontaneously generated. The KAlFe(CN) 6 adsorbent tended to completely absorb Pd(II) and become Pd[AlFe(CN) 6 ] 2 .

Conclusions
The crystal structure and adsorption ability of gamma-irradiated KAlFe(C on Pd(II) were investigated. SEM, XRD, and FT-IR characterizations showed that positions and peak intensities of the irradiated adsorbent changed slightly, but it structure was stable and not significantly damaged. After irradiation, the Kd dropped slightly, and the magnitude of the drop decreased with the increase i concentration. For the irradiation dose of 100 kGy, Kd on the adsorbent was hig 5846 cm 3 g −1 , and Qe was 48.1 mg g −1 , which was only 1.2% lower than that of the diated samples. The adsorption equilibrium was attained within 45 min, and the tion of Pd(II) followed the pseudo-second-order kinetic equation, which corresp chemical adsorption. The adsorption results indicated that KAlFe(CN)6/SiO2 achi cellent adsorption performance on Pd(II) with 100 kGy irradiation. The DFT calc showed that the stablest theoretical Pd(II) adsorption product of KAlFe(C Pd[AlFe(CN)6]2, and this adsorption process was spontaneous. When the tempera lower than 323.15 K, the most dominant intermediate product was Pd0.5K[AlF with a cell volume of 1013.53 Å 3 , followed by Pd0.5K[AlFe(CN)6]2 with a cell vo 1013.09 Å 3 . Therefore, KAlFe(CN)6/SiO2 could be used to separate and recover P S2 structures, on the other hand, were intermediate products of adsorption. The system free-energy changes indicate that S2-2 and S2-3 were likely generated more spontaneously than the others because of the low ∆F. In addition, the free-energy changes varied with temperature, that is, S2-3 was slightly superior, below 323.15 K, whereas S2-2 changed to the dominant structure of above 323.15 K.

Conclusions
The crystal structure and adsorption ability of gamma-irradiated KAlFe(CN) 6 /SiO 2 on Pd(II) were investigated. SEM, XRD, and FT-IR characterizations showed that the peak positions and peak intensities of the irradiated adsorbent changed slightly, but its crystal structure was stable and not significantly damaged. After irradiation, the K d of Pd(II) dropped slightly, and the magnitude of the drop decreased with the increase in HNO 3 concentration. For the irradiation dose of 100 kGy, K d on the adsorbent was higher than 5846 cm 3 g −1 , and Q e was 48.1 mg g −1 , which was only 1.2% lower than that of the unirradiated samples. The adsorption equilibrium was attained within 45 min, and the adsorption of Pd(II) followed the pseudo-second-order kinetic equation, which corresponds to chemical adsorption. The adsorption results indicated that KAlFe(CN) 6 /SiO 2 achieved excellent adsorption performance on Pd(II) with 100 kGy irradiation. The DFT calculations showed that the stablest theoretical Pd(II) adsorption product of KAlFe(CN) 6 was Pd[AlFe(CN) 6 ] 2 , and this adsorption process was spontaneous. When the temperature was lower than 323.15 K, the most dominant intermediate product was Pd 0.5 K[AlFe(CN) 6 ] 2 with a cell volume of 1013.53 Å 3 , followed by Pd 0.5 K[AlFe(CN) 6 ] 2 with a cell volume of 1013.09 Å 3 . Therefore, KAlFe(CN) 6 /SiO 2 could be used to separate and recover Pd(II) under radioactive conditions for use in other industries. Because of the low beta decay energy, the recovered Pd is essentially safe for industrial applications. Further study into recovering Pd from adsorbents is required, and there are several ongoing works on this topic.