DOI:
10.1039/C6RA12269H
(Paper)
RSC Adv., 2016,
6, 68695-68704
The removal of aqueous uranium by SBA-15 modified with phosphoramide: a combined experimental and DFT study†
Received
11th May 2016
, Accepted 30th June 2016
First published on 4th July 2016
Abstract
Phosphoramide-modified ordered mesoporous silica (SBA-DEPA) materials were prepared via a two-step process involving: (1) the synthesis of phosphoramide via amidation of phosphoryl chloride with a primary amine and (2) modification of the phosphoramide onto SBA-15. The successful preparation was confirmed by FT-IR and NMR spectroscopy. As indicated by the ICP-AES and XRF analysis results, the phosphoramide group has a grafting ratio as high as 12.5%. The morphological information and N2 adsorption–desorption technique proved the highly ordered structure and large specific area of the material, respectively. An excellent performance for uranium sorption was found with a loading maximum of 311.6 mg g−1, a marked promotion relative to unmodified SBA-15. A strong pH and CO2 dependence suggested that near neutral conditions favored the maximum uranium sorption. To our surprise, the sorption took only 5 min to reach a 90% capacity and 20 min for equilibrium, which is extraordinary when compared to many other sorbent materials. The ionic strength partially influences the sorption, which indicates outer and inner sphere sorption both play parts in the process. A sorption capacity of over 90% of the original indicates the excellent reproducibility of SBA-DEPA. DFT calculations based on cluster models suggested a ‘tri-dentate-like’ structure for the uranyl-surface binding site, which may explain the excellent sorption ability of this material.
1. Introduction
Due to the increasing demand for energy and resources, the nuclear industry is undergoing rapid development. An increasing amount of uranium, as the main component of nuclear fuel, is inevitably released into the environment, which has placed the ecosystem in jeopardy. Therefore, environmental protection and emergency treatment has become an important issue. Several techniques have been applied and studied, including chemical precipitation,1 ion-exchange,2 sorption,3,4 membrane filtration,5 coagulation-flocculation,6 and electrochemical processes.7 Among them, sorption has been proved to be most effective and time-saving, and thus has the potential to replace conventional methods.
Sorption of uranium by mesoporous materials is a recently developed method. SBA-15, as a silica star material,8 has been considered the most promising candidate for applications in sorption and separation due to its high surface area, large pore volume, controllable pore size and physical/chemical stability.9 The application of SBA-15 in uranium removal has been proved to be successful with a fast sorption rate and relatively high sorption capacity (205 mg g−1).10 Moreover, researchers are still searching for ways to further improve the sorption ability. Various functional groups have been grafted onto its inner and outer surfaces, and outstanding uranium-extraction performance has been shown.11–16 These functional groups include dihydroimidazole, amino, phosphonate, phosphonic, polyvinylchloride, and iminodiacetic. When compared to other functional groups, phosphate is the most effective. There is a strong coordination effect between the P
O group and uranyl ions, which has been confirmed by the classical PUREX and TRPO processes (the liquid–liquid extractions of TBP and TRPO for uranium, respectively). It is also validated by numerous crystal and sorption studies. Shi's group17 synthesized a co-condensation mesoporous product, NP10 using diethylphosphatoethyltriethoxysilane (DPTS) and tetraethoxysilane (TEOS), and the results showed that it had a maximum sorption capacity of 303 mg g−1 and achieved a fast equilibrium time of 30 min. Wu and co-workers18 compared the sorption abilities of SBA-15 grafted using both diethylethylphosphonate (DEP) and ethylphosphonic acid (PA). He found that SBA-PA showed a higher sorption capacity than SBA-DEP, which is due to the higher steric hindrance of the DEP group. Moreover, in the reusability experiments of SBA-PA, the sorption ability did not markedly change over six cycles. Chen's group19 developed mesoporous silica sorbents functionalized with alkylphosphine oxide ligands, which had effective sorption towards U(VI) even in strong HNO3 media. Dudarko and Dai20 also synthesized phosphonic-acid-grafted SBA-15 and applied it in uranium extraction under neutral to moderate basic conditions. The nanocomposite had a specific area as large as 533 m2 g−1 and the grafting rate was ranging from 1.0 to 3.0 mmol g−1. The optimal sorption capacity towards uranium reached 55 g kg−1.
Phosphoramide is a class of phosphorous compound that contains P
O groups. Studies have shown that phosphoramides can form strong complexes with uranyl. Shundalau21 studied two complexing structures of hexamethyl phosphoramide (HMPA) with uranyl chloride using quantum chemistry method. He found that two HMPA molecular ligands participated in the coordination as an electron donor. The P
O and U
O bonds are distinctly elongated, accompanied with a long-wave shift of the P
O stretching mode as large as 160 cm−1. A list of crystallography studies have also demonstrated that phosphoramide could form strong interactions with uranyl with a coordination U–O(P) bond as short as ∼2.3 Å22,23. In the extraction studies of actinides with ionic liquids, which have been considered as green chemistry, phosphoramide groups were functionalized onto other molecules, such as imidazolium24 and quaternary ammonium cations,25 to obtain a good liquid/liquid extraction performance, especially for uranium. Polymeric materials bearing phosphoramide groups have been investigated in uranium sorption. Villemin et al.26 prepared a resin (polyethyleniminephenylphosphonamidic acid) containing the phosphoramide group and used it for uranium extraction. The phosphoramide group was found to be inert against oxidizing reagents and strongly acidic medium, and the resin had a sorption capacity of 39.7 mg g−1. Cao et al.27,28 fabricated two phosphoramidic co-polymers by grafting dichlorophenylphosphine oxide onto a polystyrene-type backbone and found them to have an excellent affinity towards uranium (90–107 mg g−1 at room temperature). However, the study on the sorption of phosphoramide-bearing materials for uranium is still rare and the properties of this type of sorbent remain unknown.
In the present study, the diethylphosphoramide group (DEPA) was functionalized onto the ordered mesoporous SBA-15 via amidation and a thermally induced reaction. Composition analysis, morphological/size-distribution information and surface/pore data were achieved via ICP-AES, FT-IR, 1H-NMR, XRF, SEM, TEM, DLS and N2-adsorption techniques. The sorption properties, such as the effect of pH, ionic strength, contact time as well as sorption isotherms, were elucidated in detail. To further understand the mechanisms of the sorption processes, DFT calculations based on cluster models were performed. We would like to point out that, to our best of knowledge, this is the first report on the preparation of phosphoramide-modified silica composite materials, as well as their sorption nature towards uranium.
2. Materials and methods
2.1 Materials
Analytical grade chemicals were used throughout the investigation. 3-(2-Aminoethylamino)propyltrimethoxysilane (APTMS), tetraethylorthosilicate (TEOS) and triethylamine (TEA) were purchased from Aladdin Chemical Corporation. SBA-15 was offered by Xianfeng Nano. Corp. (Nanjing, China). Diethyl chlorophosphate (DECP) was provided by Alfa Aesar. Uranyl nitrate was provided by Dingtian Chem. Corp. (Xi'an).
2.2 Synthesis of the SBA-DEPA adsorbents
The synthesis experiment was performed according to the route, as shown in Fig. 1.
 |
| Fig. 1 Schematic of the synthetic route used to prepare SBA-DEPA. | |
2.2.1 Synthesis of APTMS-DEPA. The functionalized silane coupling agent APTMS-DEPA was synthesized from APTMS and DECP using TEA as a catalyst. In brief, equal amounts of APTMS and TEA were dissolved in dry toluene under a flow of nitrogen. The mixture was stirred vigorously and placed in a water bath at around 5 °C. Toluene containing DECP was added dropwise to the solution. The reaction was continued overnight. Then, filtration and evaporation were carried out, and the buff product was obtained in 95% yield. DEPA is the abbreviation used for the functional group di-ethyl-phosphoramide in following sections.
2.2.2 Grafting of APTMS-DEPA on SBA-15. Before preparation, SBA-15 was activated in 6 M HCl aqueous solution at 100 °C for 2 h, followed by rinse and desiccation. In the synthesis processes, an amount of SBA-15 powder was ultrasonically dispersed in dry ethanol, and then an amount of the pre-synthesized APTMS-DEPA was added dropwise to the solution. The reaction was stirred for 2 h at ambient temperature and refluxed for another 5 h. Through filtration, washing and drying, a buff product was obtained in 70–80% yield. Different amounts of APTMS-DEPA were added to the solution to achieve different grafting ratios with 12.5% as the upper limit of the grafting process.
2.3 Material characterization methods
2.3.1 Fourier transform infrared spectroscopy (FT-IR). A Bruker vertex-70 infrared spectrometer was used to qualitatively investigate the vibrational properties of the organic compounds. All samples were tested at a 4 cm−1 resolution and 128 scans in a 4000–400 cm−1 range. Some coupling peaks were manually divided into several single peaks.
2.3.2 Scanning electron microscopy (SEM) and transmission electron microscopy (TEM). SEM (FEI-Sirion200) and TEM (FEI F20) were used to study the morphology of SBA-15 and its derivatives.
2.3.3 ICP-AES. Inductively coupled plasma atomic emission spectrometry (ICP-AES) was used to determine the grafting percentage. In brief, we established a standard curve for phosphorus using Na3PO4 as the source. The SBA-DEPA composite was dissolved in a HNO3/HF mixed solvent and the solution was sealed and heated in a microwave digestion furnace at 80 °C for 10 min forming a homogeneous solution. The phosphorus content was determined via AES measurements and the grafting percentage of the DEPA group was calculated.
2.3.4 NMR spectroscopy. NMR spectroscopy was performed under full automation on an AVANCE III 600 MHz spectrometer (Bruker, Germany). All NMR spectra were phased and baseline-corrected. The measurements included 1H, 13C and 31P NMR spectra.
2.3.5 N2 adsorption–desorption isotherms. To determine the specific area, pore volume and pore sizes, N2 adsorption–desorption experiments were carried out using a Quantachrome Autosorb-1 apparatus. The samples were degassed at 100 °C for 8 h and the isotherms were then studied for another period of time.
2.3.6 Dynamic light scattering and zeta-potential analysis. The DLS technique was employed to investigate the dynamic radii, as well as the IEP (isoelectric point), of the prepared samples on a Brookhaven 90Plus-PALS instrument. Samples were treated ultrasonically prior to determination.
2.4 Sorption experiments
In a batch experiment, 1 mg of the adsorbent was mixed with uranyl nitrate in a 5 mL solution. After the sorption process, the samples were isolated from the supernatant using centrifugation and the supernatant solution was analyzed using a UV-Vis spectrometer (Shimazu, UV-Vis 1800) at 651.5 nm in the presence of Arsenazo III (a complexing agent) to obtain the concentration of uranyl ion. The amount of uranium being adsorbed was calculated according to eqn (1): |
 | (1) |
where Qe denotes the equilibrium sorption capacity, C0 and Ce represent the initial and equilibrium uranium concentrations, respectively, V is the total volume and m is the sorbent mass.
2.5 Density functional theory (DFT) studies
The quantum mechanical study using the density functional theory (DFT) method29,30 was carried out using the Gaussian 09 package suite.31 All calculations were under the B3LYP functional,32–34 a hybrid functional that is mostly used in the actinide field.35–37 Full optimizations were performed from different initial structures both in the gas and solution phase. Gibbs free energies were obtained via frequency calculations. The ECP60MWB-SEG basis set and relative pseudo potential were introduced for uranium,38,39 assuming inner 60 electrons as the core and the outer 32 electrons free for inter-atomic calculations. For light atoms (C, H, O, N, P, Si), the 6-31G(d) was used for geometry optimization, which is most often used. For frequency calculations, 6-311G(d,p) was employed, which describes the atomic orbitals more accurately and considers a greater polarization effect for hydrogen. The influence of the solvent (water) was modeled using the SMD continuum solvation method,40 wherein the solute is immersed in a shape adapted isotropic polarizable continuum, with a dielectric constant ε = 78.3553 for water.
3. Results and discussion
3.1 Properties of the materials
The infrared and 1H-NMR spectra make it clear that the APTMS-DEPA was successfully synthesized via the amidation reaction, as shown in Fig. S1.† Through peak detachment, we could easily observe in the APTMS-DEPA infrared spectrum that υP–N (920 cm−1) was formed and υP–Cl (1295 cm−1) was missing. Moreover, υP
O (1228 cm−1) and δSi–O–C (1088 cm−1) co-exist from the two reactants. In the 1H-NMR spectra, the exact attachments of chemical shifts are assigned on the molecular structures. The primary amine –NH2 (δ = 1.11 ppm) was changed to a secondary amine –NH– (δ = 2.16 ppm). Moreover, the peaks corresponding to the methyl and methylene groups from both APTMS (δ = 3.38, 2.69, 1.41, 0.45 ppm) and DECP (δ = 3.86, 1.13 ppm) co-exist. Moreover, 31P NMR spectroscopy was also carried out and it was clear that the chemical shifts of P experience a large shift after the synthesis of APTMS-DEPA, as shown in Fig. S2.†
The three composite materials comprised SBA-15 grafted with APTMS-DEPA were characterized using ICP-AES and the content of functional groups were 4.6%, 8.5% and 12.5%. Moreover, X-ray fluorescence spectrometry (XRF) was used to determine the phosphorus mass on the surfaces, as shown in Fig. 2-I. It is clearly observed that the main peak Kα and its tailed peak SKα3 are located at 2θ = 141° and 139°, respectively. The tendency of the three spectra, obtained from this semi-quantitative method, is in accordance with the ICP-AES results. 13C NMR spectroscopy was also performed to confirm the target SBA-DEPA product and the spectra, as well as the ascription, are available in Fig. S3.†
 |
| Fig. 2 Sample characterization: (I) XRF spectra of the SBA-15 derivatives; (II) SEM (a, b) and TEM (c, d) images of SBA-12.5%DEPA; (III) the hydration radius distribution obtained via DLS measurements; (IV) the N2-adsorption/desorption isotherms for SBA and its DEPA derivatives. | |
The morphological information for the SBA-12.5DEPA sample, characterized by SEM and TEM, suggests that the particles are rod-like with a diameter of 300–400 nm and a length of 1–2 μm, and have a hexagonal ordering mesoporous structure with an approximately 5 nm pore diameter (Fig. 2-II). From the size distribution obtained via the DLS technique, the hydration sizes of the samples ranged from 4 to 6 μm (Fig. 2-III), a little larger than that found from the SEM results.
The N2 adsorption–desorption isotherms for all the samples are shown in Fig. 2-IV. All the samples display a typical type-IV isotherm with an H-1 type hysteresis loop. This type of adsorption is characteristic of mesoporous materials with a 2D-hexagonal structure.41 The BET specific surface area as well as the pore volume and size based on the BJH method are listed in Table 1. It is clearly observed that the surface specific area as well as the pore volume and size decreases with the extent of modification; however, they are still large enough for uranium sorption.
Table 1 Parameters of the N2 adsorption/desorption for the mesoporous materials
|
SBA-15 |
SBA-4.6%DEPA |
SBA-8.5%DEPA |
SBA-12.5%DEPA |
Specific surface area (m2 g−1) |
499.1 |
473.1 |
383.7 |
365.1 |
Pore volume (cm3 g−1) |
0.894 |
0.752 |
0.685 |
0.665 |
Pore size (nm) |
12.15 |
10.80 |
9.58 |
9.51 |
3.2 Sorption
3.2.1 The effects of pH and CO2. The pH effect on the sorption process was radically attributed to the reliability of uranyl species on acidity. The high valence on its equatorial plane gives it an affinity with Lewis bases, especially hydroxyl and carbonate ions. Studies have shown that there are several uranyl–hydroxide species in aqueous solutions, namely, [(UO2)m(OH)n]2m−n ([m, n] is often used to designate uranyl hydroxide complexes), and some may co-exist under certain conditions (i.e. concentration, pH, and temperature). The pH dependence of uranyl species was simulated using the VMinteq3.0 software suite (calculating the equilibrium state using Kc and Ksp of several possible components) and the fitted data are shown in Fig. 3a. It is observed that the aquo uranyl ion [1, 0] is absolutely dominant in highly acidic conditions. Under weakly acidic conditions, three soluble species [1, 1], [2, 2] and [3, 5] become prominent. Schoepite as a precipitate emerges under near neutral circumstances. In moderate and strongly basic solutions, carbonate ions have a high affinity with uranyl ions and form stable complexes. The amounts of the uranyl hydroxide complexes are reduced significantly.
 |
| Fig. 3 The pH dependence of uranyl species: the simulation (a) and experimental (b) results. | |
The experimental results showing the pH dependence of SBA-12.5%DEPA sorbent under certain initial conditions ([U]initial = 28 mg L−1) are presented in Fig. 3b and show a significant increase followed by a slight decline upon increasing the pH value. Protonation of the sorbent surfaces as a function of pH was determined by zeta-potential measurements, as shown in Fig. 4. The electrical charge at the aqueous/surface describes the protonation/deprotonation of the surficial oxygen and the point wherein the zeta-potential = 0, as shown by the red cross mark, indicates the isoelectric point (IEP), meaning that there is no charge at that site. In our experiments, the IEPs are pH 3.45 and 5.03 for neat SBA-15 and SBA-12.5%-DEPA, respectively. The difference may be attributed to the difference in the pKa between the Si–O⋯H and R2NP
O⋯H groups.
 |
| Fig. 4 The zeta-potential at different pH. | |
In detail, the pH effect on the uranyl sorption by the modified SBA-15 sorbents can be divided into three regions.
(1) pH < 5: the sorption capacity increases rapidly with pH. Protonation is depressed and deprotonation is promoted when pH increases, which makes the formation of uranyl ion complexes with the deprotonated surface oxygen atoms easier.
(2) 5 < pH < 8: the sorption capacity increases gently with pH and reaches a maximum. According to Fig. 3a, schoepite emerges under the present initial concentration and rapidly turns to be dominant. Therefore, it is concluded that the schoepite precipitate accounts for the increase in the sorption capacity at near neutral solutions.
(3) pH > 8: some soluble uranyl–carbonate complexes form, which largely inhibit the binding of uranyl ions on the silica surfaces.
This trend for the pH effect on uranyl sorption has been also found by other researchers.42,43 It should be mentioned that to achieve a high sorption capacity and avoid of the influences from precipitations and CO2, all the isotherm experiments were performed at pH 5.
3.2.2 The effect of ionic strength on sorption. Ionic strength is a crucial factor in the sorption process. The effect of ionic strength on the sorption of U(VI) by SBA-12.5%DEPA was studied in the presence of NaClO4 with concentrations varying from 0.1 to 1.0 mol L−1 and a fixed initial condition ([U]initial = 28 mg L−1, pH = 5). The results are shown in Fig. 5. It can be observed that the sorption capacity decreases rapidly from 85 to 74 mg g−1 as the concentration of NaClO4 increases and remains constant at [NaClO4] > 0.5 mol L−1. The influence of ionic strength on SBA-DEPA sorption of uranium is quite significant and this phenomenon is quite similar to that shown in previous reports.18,44 The reason for this is that a fraction of uranyl is physically adsorbed on the sorbent surface by outer-sphere interactions and this type of electrostatic interaction can be partially screened by counter ions. On the contrary, the inner-sphere complexes are formed chemically and are much stronger, thus the ionic strength has almost no impact on it.
 |
| Fig. 5 The sorption capacity of uranyl ions by SBA-12.5%DEPA at different ionic strengths. | |
3.2.3 The effect of contact time on sorption. To understand the effect of the contact time on uranium sorption onto the SBA-DEPA-12.5% composite, experiments were conducted. An injector, with two layers of 0.22 μm filter membranes on its head, were used to extract 2 mL of a suspension each time and the solid/liquid ratio was kept fixed. The sorption ability variation with contact time is presented in Fig. 6. The composite exhibits a considerably rapid sorption rate. Over 90% aqueous uranium was removed from the solution within 5 min and equilibrium was reached within 20 min.
 |
| Fig. 6 The effect of contact time on sorption. | |
3.2.4 Thermodynamics studies. The effect of temperature for the sorption of U(VI) on SBA-12.5DEPA was investigated from 288 K to 323 K. The data from these sorption experiments were used to estimate the thermodynamic parameters such as enthalpy (ΔH°), Gibbs free energy (ΔG°) and entropy (ΔS°), using the following equations: |
 | (3) |
where Kd (mL g−1) is the distribution coefficient in solid and liquid phase, and R is the gas constant 8.314 J mol−1 K−1. ΔH° and ΔS° were calculated from the slope and intercept, respectively, and ΔG° was then obtained (Fig. 7). The results are listed in Table 2.
 |
| Fig. 7 A plot of ln Kd vs. 1/T for U(VI) sorption by SBA-12.5%DEPA. | |
Table 2 The thermodynamic parameters for U(VI) sorption by SBA-12.5%DEPA
Temperature (K) |
ΔG° (kJ mol−1) |
ΔH° (kJ mol−1) |
ΔS° (J mol−1 K−1) |
288 |
−21.66 |
14.67 |
126.13 |
298 |
−22.92 |
|
|
303 |
−23.55 |
|
|
313 |
−24.81 |
|
|
323 |
−26.07 |
|
|
The negative ΔG° values suggest the spontaneous nature of U(VI) sorption. The positive ΔH° value indicates that the sorption was endothermic and higher temperature lead to a more negative ΔG°, which favors the sorption process. The positive ΔS° value means an irregular increase in the randomness on the sorbent surface, suggesting the sorption reaction was irreversible.
3.2.5 Sorption isotherms. The sorption isotherm is of the most importance and reflects the solute equilibrium relationship between the ones adsorbed onto the solid phase and the counterpart existing in solution under a given condition. The sorption capacities of SBA-15 and its derivative materials as a function of the equilibrium uranium concentration were determined. All these experiments were performed with various initial uranium concentrations and the other factors were kept fixed (sorbent solid/liquid ratio: 0.2 mg mL−1, temperature: 298 K, pH = 5, total volume: 5 mL). Four isotherms were finally obtained for the four sorbents, which are SBA-15, SBA-4.6%DEPA, SBA-8.5%DEPA and SBA-12.5%DEPA, respectively, as shown in Fig. 8.
 |
| Fig. 8 The sorption isotherms for bear and functionalized SBA-15 (left) and the sorption capacity as a function of the grafting rate of the DEPA functional group (right). | |
With the assumption of the Langmuir sorption theory, the monolayer sorbate covers the surface and no subsequent interactions among them occur. In other words, no further adsorption can occur after adsorption saturation on the surface. The linear form of the Langmuir isotherm can be expressed as follows:
|
 | (4) |
where
b is the Langmuir constant and
Qm is the maximum adsorption capacity.
Ce and
Qe are the concentration (mg L
−1) and adsorption capacity at equilibrium, respectively.
Qm can be achieved from the plot of
Qe vs. Ce.
Moreover, the Freundlich expression is an empirical equation based on the sorption on a heterogeneous surface and can be given as follows:
where
KF and
n are the Freundlich constants related to the adsorption capacity and intensity, respectively. It can be obtained from the plot of
Qe against
Ce.
The Langmuir and Freundlich models fitting data are listed in Table 3. It was observed that the Langmuir model fits better than Freundlich model for all the isotherms, thus the monolayer mechanism best describes the sorption processes that are occurring. The maximum sorption capacity Qm can also be obtained from the Langmuir isotherms. As the DEPA grafting rate increases, there is a great promotion of the sorption ability of the materials, as shown in Fig. 8. 12.5% is the maximal grafting rate in our experiments and adding excessive DECP did not result in a larger rate. Therefore, the SBA-DEPA composite sorbents have an utmost Qm value of 311.6 mg g−1.
Table 3 Parameters for the sorption isotherms
Samples |
Langmuir |
Freundlich |
Qm, mg g−1 |
b, L g−1 |
R2 |
KF, mg g−1 (L mg−1)−1/n |
n |
R2 |
SBA-15 |
109.4 |
0.147 |
0.977 |
25.2 |
2.849 |
0.968 |
SBA-4.6%DEPA |
144.8 |
0.099 |
0.983 |
24.9 |
2.492 |
0.950 |
SBA-8.5%DEPA |
191.8 |
0.073 |
0.989 |
27.8 |
2.327 |
0.981 |
SBA-12.5%DEPA |
311.6 |
0.077 |
0.983 |
36.0 |
2.033 |
0.937 |
3.2.6 Reproducibility. To investigate the reproducibility of the composite material, experiments were performed at a certain initial sorption ([U]0 = 28 mg L−1, T = 298 K, pH = 5) and elution ([HNO3] = 1 mol L−1) conditions for SBA-12.5DEPA. As observed from Fig. 9, the sorption capacity of the first cycle is distinctly reduced when comparing to the origin, due to the difficulty in dispersion of uranyl ions inside the sorbent pores. From the second till the last cycle, the sorption capacity remains almost constant, suggesting the excellent reproducibility of the SBA-DEPA composite.
 |
| Fig. 9 The U(VI) sorption performance over six regeneration cycles. | |
3.2.7 Sorption mechanisms through DFT calculations. As discussed above, the introduction of phosphoramide groups significantly improves the uranyl sorption capacity of SBA-15. It is important for us to gain more understanding of the mechanisms of the sorption processes. For minerals, the outer-sphere and inner-sphere binding modes are usually considered as two forms of sorption, resulting from electrostatic and coordination interactions between uranyl and the surface oxygen atoms, respectively. However, for functionally modified materials, the grafted groups become the localized binding sites due to the strong coordination with uranyl ions, which is much different from their pure minerals. Therefore, we have conducted a deep investigation from a quantum chemistry point of view to find out the sorption mechanisms for this type of functionally modified material. Herein, we established some localized surface models of SBA15/DEPA/uranyl to optimize the structures in which SBA-15 was substituted by a Si7O15H12 cluster. This is because we are most concentrated on the local sorption sites instead of the whole periodic surface. Moreover, periodic system calculations are really tough for quantum mechanical calculations using the Gaussian09 suite. In fact, we have previously tried a cluster with only three silicon atoms; however, we found the DEPA functional group tended to bend towards the silica surface and we speculated that the surface Si–OH might display a synergistic effect. Therefore, we extended the cluster surface to seven silicon atoms with four additional Si–OH groups. This model takes the effects of both the DEPA and Si–OH groups into consideration and is clearly reasonable and reliable. In the optimizations, the two bottom atomic layers were fixed. The frequencies were calculated afterwards and no imaginary frequency was found. Though several types of polymeric uranyl species co-exist in solution, only mononuclear uranyl ion was found on the surfaces of different minerals45 and this has been adopted by nearly all the DFT calculations. Therefore, only mononuclear uranyl was considered in our modeling approach.After all the optimizations, we found that the ‘tri-dentate-like’ structure was the most stable one whatever the initial structure. The whole DEPA group bends seriously towards the silica substrate and the phosphoramide oxygen (P
O) forms a ‘tri-dentate-like’ structure together with two adjacent oxygen atoms on the silica surface (see Fig. 10). As there might be water ligands around uranyl and deprotonation of the hydroxyl groups on the silica surface may occur, four possible situations have been taken into account: (1) no water and protonation, (2) no water and deprotonation, (3) two waters and protonation, and (4) two waters and deprotonation. The full optimized structures are shown in Fig. S4a–d.† The bond lengths, reaction equations and relative energies are listed in Table 4. In each structure, the uranyl ion is bound to P
O and two surface oxygen atoms in a ‘tri-dentate’ mode. The U–O(
P) bonds are in the range of 2.29–2.46 Å and the U–Osurf bonds are significantly shorter when deprotonated (∼2.3 Å) when compared to their protonated derivatives (∼2.6 Å). In the EXAFS studies of uranyl sorption on SiO2/Al2O3/montmorillonite surfaces conducted by Sylwester,46 nearly all the U–Osurf bond lengths were 2.3–2.5 Å and they were shorter in the inner-sphere complexes (2.35 Å) when compared to the outer-sphere complexes (2.4 Å). For the theoretical studies in both solution and on the mineral surfaces,47–49 the U–Oyl bond length was ∼1.8 Å as the first coordination shell and in the second shell the U–Oeq bonds have been calculated in the range of 2.1–2.6 Å. Shi's group found the U–O(
P) bond length was ∼2.45 Å in UO22+/CMPO complexes and ∼2.41 Å in UO22+/DEHP− complexes both in the most stable structures.48,49 From Table 4, it is clearly observed that the water ligands and the deprotonation of surficial oxygen both play important roles in sorption. When two water ligands participate, the ΔG values have an increase ranging from 3 to 8 kcal mol−1, and the bond lengths of U–O(
P) and U–Osurf are prolonged by 0.03–0.05 Å and 0.03–0.09 Å, respectively. Moreover, after deprotonation of the surface oxygen groups, the decrease in the ΔG values was ∼50 kcal mol−1 and the U–Osurf bonds were significantly shortened by as much as 0.32 Å. The changes derived from the deprotonation of surficial oxygen are much more significant than the participation of water ligands. Among all the possible geometries, the ‘tri-dentate like’ uranyl complexes of deprotonated surficial oxygen without water ligands is the most stable one, as shown in Fig. S4b.† In this structure, the U–O(
P) bond length (2.41 Å) agrees quite well with those in Shi's reports. Moreover, it is confirmed that the uranyl ion is in an inner-sphere structure according to the short U–Osurf bond as 2.26 Å.
 |
| Fig. 10 The full optimized structures of silica-DEPA (a) and its uranyl–sorption complex (b). U (light blue), O (red), P (purple), N (dark blue), H (white) with other parts represented by sticks. | |
Table 4 The reaction energies and U–O bond lengths of the resultant structures
Model |
Reactions |
ΔG kcal mol−1 |
U–O( P) Å |
U–Osurf Å |
a |
UO2(H2O)52+ + SBA-DEPA ⇒ (UO22+)·(SiOH)2·DEPA + 5H2O |
−127.4 |
2.29 |
2.58 |
b |
UO2(H2O)52+ + SBA-DEPA ⇒ (UO22+)·(SiO−)2·DEPA + 5H2O + 2H+ |
−180.3 |
2.41 |
2.26 |
c |
UO2(H2O)52+ + SBA-DEPA ⇒ (UO22+)·(SiOH)2·DEPA·(H2O)2 + 3H2O |
−124.9 |
2.32 |
2.67 |
d |
UO2(H2O)52+ + SBA-DEPA ⇒ (UO22+)·(SiO−)2·DEPA·(H2O)2 + 3H2O + 2H+ |
−172.3 |
2.46 |
2.29 |
This result describes the state of uranyl in the sorption processes. The grafting of functional groups increases the density of the material, which is unfavorable to sorption. However, sorption is a process of dynamical equilibrium and the sorption sites are greatly increased by functionalization. This enhances the sorption capacity of SBA-15 to a large extent.
Sorption of aqueous U(VI) by the organo-phosphorus group modified materials described in previous reports have shown high practicability and efficacy, and the high affinity of ‘P
O’ with uranium. Zhang et al.50 synthesized a group of mesoporous silica adsorbents functionalized with alkyl phosphine oxide with different chain lengths, through a two-step process. The materials showed great resistance to strong acid (∼10 mg g−1 in 1 M or 2 M HNO3 solutions). Liu et al.51 prepared phosphate-functionalized grapheme oxide via an Arbuzov reaction and found an excellent sorption capacity as high as ∼250 mg g−1. Wei et al.52 synthesized two phosphorus-modified SBA-15 sorbents from azido-grafted silica and found that the positively charged sorbent had a standout sorption performance under neutral conditions because of the existence of negatively charged uranium carbonate species. When compared to these studies, in our present study we selected phosphoramide as the functional group because the electron repulsive forces of the amino group obtain an increase in electron density around the oxygen atom, which enhances its affinity with f-block elements. Moreover, what we aimed to find out was whether the modification of this group has a large effect for sorption and this has been successfully confirmed, as shown in Fig. 8 (an increase in the sorption capacity as high as 200% when compared to unmodified SBA-15).
4. Conclusions
In summary, a series of SBA-DEPA mesoporous silica with different grafting rates were synthesized via an amidation and surface grafting reaction. Due to both a large surface area and high affinity, SBA-DEPA exhibited a quite satisfying sorption performance for uranium. pH and CO2 have a significant impact on sorption with pH 5 selected as the acidic condition to retain a high sorption ability and avoid precipitation of schoepite. The sorption experiments proved that SBA-DEPA can effectively adsorb uranium with a saturated capacity of 311.6 mg g−1. The Langmuir isotherm fits better than the Freundlich isotherm, suggesting a monolayer sorption process. The sorption mechanism of DEPA-grafted silica was investigated using DFT calculations based on cluster models and the SMD model with ‘tri-dentate like’ surficial coordination being demonstrated. The findings explain the excellent uranium sorption ability of SBA-DEPA.
Acknowledgements
This study was supported by the National Natural Science Foundation of China (21507118 & 21303068). We really appreciate the contributions of Prof. Dongqi Wang (Institute of High Energy Physics Academy of Sciences) and Prof. Yongfan Zhang (Fuzhou University) to this study.
References
- Z. Shi, C. Liu, J. M. Zachara, Z. Wang and B. Deng, Environ. Sci. Technol., 2009, 43, 8344–8349 CrossRef CAS PubMed.
- V. Stucker, J. Ranville, M. Newman, A. Peacock, J. Cho and K. Hatfield, Water Res., 2011, 45, 4866–4876 CrossRef CAS PubMed.
- L. Zhou, C. Shang, Z. Liu, G. Huang and A. A. Adesina, J. Colloid Interface Sci., 2012, 366, 165–172 CrossRef CAS PubMed.
- J. L. Vivero Escoto, M. Carboni, C. W. Abney, K. E. deKrafft and W. Lin, Microporous Mesoporous Mater., 2013, 180, 22–31 CrossRef CAS.
- S. K. Singh, S. K. Misra, M. Sudersanan and A. Dakshinamoorthy, Sep. Sci. Technol., 2009, 44, 169–189 CrossRef CAS.
- X. Chen, D. Ding, X. Gong, N. Hu, G. Li, Y. Liu, X. Nie and Y. Wang, CN101597113-A, 2009.
- C. S. Wang, Y. Liu, H. He, F. X. Gao, L. S. Liu, S. W. Chang, J. H. Guo, L. Chang, R. X. Li and Y. G. Ouyang, J. Radioanal. Nucl. Chem., 2013, 298, 581–586 CrossRef CAS.
- J. Ma, Z. Wang, Y. Shi and Q. Li, RSC Adv., 2014, 4, 41597–41604 RSC.
- M. Karimi, A. Badiei and G. Mohammadi Ziarani, RSC Adv., 2015, 5, 36530–36539 RSC.
- X. Wang, G. Zhu and F. Guo, Ann. Nucl. Energy, 2013, 56, 151–157 CrossRef CAS.
- L. Yuan, Y. Liu, W. Shi, Z. Li, J. Lan, Y. Feng, Y. Zhao, Y. Yuan and Z. Chai, J. Mater. Chem., 2012, 22, 17019–17026 RSC.
- X. Wang, L. Yuan, Y. Wang, Z. Li, J. Lan, Y. Liu, Y. Feng, Y. Zhao, Z. Chai and W. Shi, Sci. China: Chem., 2012, 55, 1705–1711 CrossRef CAS.
- J. Gao, L. a. Hou, G. Zhang and P. Gu, J. Hazard. Mater., 2015, 286, 325–333 CrossRef CAS PubMed.
- Y. Liu, L. Yuan, Y. Yuan, J. Lan, Z. Li, Y. Feng, Y. Zhao, Z. Chai and W. Shi, J. Radioanal. Nucl. Chem., 2012, 292, 803–810 CrossRef CAS.
- M. Gao, G. Zhu, X. Wang, P. Wang and C. Gao, J. Radioanal. Nucl. Chem., 2015, 304, 675–682 CrossRef CAS.
- Y. Wang, L. Song, L. Zhu, B. Guo, S. Chen and W. Wu, Dalton Trans., 2014, 43, 3739–3749 RSC.
- L. Yuan, Y. Liu, W. Shi, Y. Lv, J. Lan, Y. Zhao and Z. Chai, Dalton Trans., 2011, 40, 7446–7453 RSC.
- Y. Wang, L. Zhu, B. Guo, S. Chen and W. Wu, New J. Chem., 2014, 38, 3853–3861 RSC.
- W. Zhang, G. Ye and J. Chen, RSC Adv., 2016, 6, 1210–1217 RSC.
- O. A. Dudarko, C. Gunathilake, N. P. Wickramaratne, V. V. Sliesarenko, Y. L. Zub, J. Górka, S. Dai and M. Jaroniec, Colloids Surf., A, 2015, 482, 1–8 CrossRef CAS.
- M. B. Shundalau, A. I. Komiak, A. P. Zajogin and D. S. Umreiko, J. Spectrosc. Dyn., 2013, 3, 4–10 Search PubMed.
- M. J. Crawford, A. Ellern, K. Karaghiosoff, F. Martin and P. Mayer, Inorg. Chem., 2010, 49, 2674–2683 CrossRef CAS PubMed.
- C. Villiers, P. Thuéry and M. Ephritikhine, Inorg. Chem. Commun., 2007, 10, 891–893 CrossRef CAS.
- I. M. AlNashef, M. A. Matthews and J. W. Weidner, Abstracts of Papers, American Chemical Society, 2002, vol. 223, 74-IEC 74 Search PubMed.
- A. Ouadi, O. Klimchuk, C. Gaillard and I. Billard, Green Chem., 2007, 9, 1160–1162 RSC.
- O. Abderrahim, M. A. Didi and D. Villemin, J. Radioanal. Nucl. Chem., 2009, 279, 237–244 CrossRef CAS.
- Q. Cao, Y. Liu, C. Wang and J. Cheng, J. Hazard. Mater., 2013, 263(Part 2), 311–321 CrossRef CAS PubMed.
- Q. Cao, Y. Liu, X. Kong, L. Zhou and H. Guo, J. Radioanal. Nucl. Chem., 2013, 298, 1137–1147 CrossRef CAS.
- R. G. Parr and W. Yang, Density-Functional Theory of Atoms and Molecules, Oxford University Press, New York, Oxford, 1989 Search PubMed.
- T. Ziegler, Chem. Rev., 1991, 91, 651–667 CrossRef CAS.
- M. Frisch, G. Trucks, H. Schlegel, G. Scuseria, M. Robb, J. Cheeseman, G. Scalmani, V. Barone, B. Mennucci and G. Petersson, Gaussian Inc., Wallingford, CT, 2009, 270, 271.
- A. D. Becke, Phys. Rev. A: At., Mol., Opt. Phys., 1988, 38, 3098–3100 CrossRef CAS.
- C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter Mater. Phys., 1988, 37, 785–789 CrossRef CAS.
- A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652 CrossRef CAS.
- Y. Ren, H. Tang, L. Shao, J. Zhong, M. Chu, R. Yang and C. Kong, RSC Adv., 2016, 6, 46467–46474 RSC.
- D. Shen, C. Kong, R. Jia, P. Fu and H. Zhang, J. Phys. Chem. A, 2015, 119, 3636–3643 CrossRef CAS PubMed.
- J. Zeng, X. Yang, J. Liao, N. Liu, Y. Yang, Z. Chai and D. Wang, Phys. Chem. Chem. Phys., 2014, 16, 16536–16546 RSC.
- W. Küchle, M. Dolg, H. Stoll and H. Preuss, J. Chem. Phys, 1994, 100, 7535–7542 CrossRef.
- X. Cao and M. Dolg, J. Mol. Struct.: THEOCHEM, 2004, 673, 203–209 CrossRef CAS.
- A. V. Marenich, C. J. Cramer and D. G. Truhlar, J. Phys. Chem. B, 2009, 113, 4538–4543 CrossRef CAS PubMed.
- D. Zhao, Q. Huo, J. Feng, B. F. Chmelka and G. D. Stucky, J. Am. Chem. Soc., 1998, 120, 6024–6036 CrossRef CAS.
- S. Sadeghi and E. Sheikhzadeh, J. Hazard. Mater., 2009, 163, 861–868 CrossRef CAS PubMed.
- H. Zhang, Q. Liu, J. Wang, J. Liu, H. Yan, X. Jing and B. Zhang, RSC Adv., 2015, 5, 5904–5912 RSC.
- H. Wang, L. Ma, K. Cao, J. Geng, J. Liu, Q. Song, X. Yang and S. Li, J. Hazard. Mater., 2012, 229–230, 321–330 CrossRef CAS PubMed.
- E. R. Sylwester, E. A. Hudson and P. G. Allen, Geochim. Cosmochim. Acta, 2000, 64, 2431–2438 CrossRef CAS.
- E. Sylwester, E. Hudson and P. Allen, Geochim. Cosmochim. Acta, 2000, 64, 2431–2438 CrossRef CAS.
- H. Wang, Z. Chai and D. Wang, Dalton Trans., 2015, 44, 1646–1654 RSC.
- C. Wang, J. Lan, Y. Zhao, Z. Chai, Y. Wei and W. Shi, Inorg. Chem., 2013, 52, 196–203 CrossRef CAS PubMed.
- J. Luo, C. Wang, J. Lan, Q. Wu, Y. Zhao, Z. Chai, C. Nie and W. Shi, Dalton Trans., 2015, 44, 3227–3236 RSC.
- W. Zhang, G. Ye and J. Chen, J. Mater. Chem. A, 2013, 1, 12706–12709 CAS.
- X. Liu, J. Li, X. Wang, C. Chen and X. Wang, J. Nucl. Mater., 2015, 466, 56–64 CrossRef CAS.
- Y. Wei, L. Zhang, L. Shen and D. Hua, J. Mol. Liq., 2015 DOI:10.1016/j.molliq.2015.04.056.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra12269h |
|
This journal is © The Royal Society of Chemistry 2016 |
Click here to see how this site uses Cookies. View our privacy policy here.