Article pubs.acs.org/Langmuir
Adsorption Behavior of Rare Earth Metal Cations in the Interlayer Space of γ‑ZrP Takahiro Takei,*,† Kiyoaki Iidzuka,† Akira Miura,‡ Sayaka Yanagida,† Nobuhiro Kumada,† Eisuke Magome,§ Chikako Moriyoshi,§ and Yoshihiro Kuroiwa§ †
Center for Crystal Science and Technology, University of Yamanashi, 7-32 Miyamae, Kofu, Yamanashi 400-8511, Japan Faculty of Engineering, Division of Applied Chemistry, Hokkaido University, Kita 13 Nichi 8, Kita-ku, Sapporo 060-8628, Japan § Department of Physical Science, Hiroshima University, 1-3-1 Kagamiyama, Higashi-hiroshima, Hiroshima 739-8526, Japan ‡
S Supporting Information *
ABSTRACT: Adsorption competencies of rare earth metal cations in γ-zirconium phosphate were examined by ICP, synchrotron X-ray diffraction (SXRD), and ab initio simulation. The adsorption amounts are around 0.06−0.10 per zirconium phosphate. From the SXRD patterns of the adsorbed samples, the basal spacing estimated by c sin β increased linearly with an increasing ionic radius of rare earth metal cation, though a and b lattice constants show no change. These SXRD patterns can be classified into four groups that have different super lattices. The four superlattices have multiplicities of x131, x241, and x221 for the xabc axis, and the location of the rare earth metal cation in the original unit cell changes depending on the superlattice cell. In the x131 superlattice, Yb and Er occupied the site near the zirconium phosphate layer, though La and Ce in the x221 superlattice remained in the center position between the phosphate sheet. For the ab initio simulation of γ-ZrP with the typical rare earth metal cations (Tb, Eu, Dy, and La), the results of simulation show a similar tendency of the position estimated by SXRD refinements.
■
INTRODUCTION Rare earth elements are widely used for strong magnets, luminescence, abrasive substances, and a wide range of other applications1,2 because rare earth elements have an f electron as a valence orbital. Most of the rare earth elements become trivalent cations, and consequently, the numbers of 4f electrons in the trivalent cation vary from 0 to 14 depending on the atomic number from La to Lu. Generally, because the 4f electrons are localized, these f electrons in the valence orbital might work as an activator for the magnetic or luminescence substances. Such rare earth elements are widely used for many functionalized materials, and the elements are widely distributed in industrial products; therefore, recovery of rare earth elements from industrial waste is quite important.3−5 The layered inorganic materials can be strong candidates for the recovery of some elements, ions, or molecules because the interlayer space can be used as storage space.6−8 Generally, there are three types of layered inorganic materials: an intrinsically anionic layer with complementary cations, an intrinsically cationic layer with complementary anions, and a layer with no ionic materials. In the first type, there are many compounds such as layered titanate, layered manganate, layered perovskite, layered metal phosphates, etc. In these compounds, complementary cations existing between anionic layers should be proton, alkali, or alkali earth cations for ion exchangeability. © XXXX American Chemical Society
In particular, proton-including layered materials as a complementary cation sometimes show active soft chemical reactivities, such as ion exchange, intercalation, and exfoliation. For ion exchange, the reactivity may be affected by the charge density and electronegativity of the host layered material. For recovery of rare earth cations, inorganic materials should have high chemical durability under acidic conditions. In particular, for recovery from industrial waste, a strong acid tends to be used for dissolution of the rare earth elements from the waste. Then, the dissolved rare earth elements must be extracted from the solution. Therefore, the material that is used as the adsorbent of the rare earth cations must have a strong durability with respect to acid. In this study, some metal phosphate and other inorganic materials will be used for recovery of rare earth elements from an acidic solution. In particular, for γ-zirconium phosphate, the adsorption sites in the crystal structure are also examined by synchrotron XRD and ab initio simulation.
■
EXPERIMENTAL SECTION
Adsorption of a Rare Earth Cation to the Inorganic Compounds in the Aqueous Solution in the Presence of 12 Received: July 24, 2016 Revised: August 29, 2016
A
DOI: 10.1021/acs.langmuir.6b02747 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir Rare Earth Cations. Aqueous solutions in which rare earth metals Y, La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Er, and Yb coexist were prepared by dissolving rare earth metal nitrates in distilled water at a concentration of 1000 ppm of each metal (total of 12000 ppm). Inorganic compounds (40 g each), α-zirconium phosphate (α-ZrP, Toagosei Co., Ltd.), γ-zirconium phosphate (γ-ZrP, Daiichi Kigenso Kagaku Kogyo Co., Ltd.), α-titanium phosphate (α-TiP, homemade9), hydroxyapatite (HAp, homemade10), montmorillonite (Kunimine Industries Co., Ltd.), mesoporous silica (MPS, homemade11), porous glass (homemade12), and calcium sulfite (Ogihara Inc., Ltd.), were added to the rare earth metal cation coexisting aqueous solution (200 mL). Then, the solutions were stirred at room temperature. During stirring, the 0.5 mL solution was sampled during each period within 60 h for rare earth metal concentrations. For measurement of the concentrations, the picked solutions were diluted at arbitrary times with distilled water for an increase of the amount of solution. In addition, the degrees of dissolution of the compounds were measured at pH 0, 0.5, 1.5, 2.3, and 3.5 for several inorganic compounds. Adsorption of a Rare Earth Metal Cation to γ-ZrP in an Aqueous Solution Containing a Single Rare Earth Cation. The aqueous solution containing a single rare earth cation was used to examine adsorption sites within the crystal structure of the γ-ZrP. The aqueous solution was prepared with a rare earth metal nitrate and distilled water with an M/γ-ZrP ratio of 4 (M, rare earth metal). Then, 2 g of γ-ZrP was placed in the prepared aqueous solutions, and they were then stirred for 4 days to adsorb with stirring. After the period, the samples were filtered, washed with distilled water three times, and dried. The dried samples were regarded as adsorbed γ-ZrP with rare earth metal. For measurement of amounts of adsorbed rare earth metal, the following selective leaching process was used. The dried samples were placed in a 60% HNO3 aqueous solution to leach the adsorbed rare earth metals by protonation reaction. The filtrate solution was examined for contained rare earth metal. Characterizations. The amount of adsorbed rare earth metal cations was measured by ICP atomic emission spectroscopy (SPS3520UV-DD, Hitachi High-Tech Science Corp.). The adsorbed γ-ZrP in the crystal structure was examined by synchrotron XRD (SXRD) using a Debye−Scherrer camera at BL02B2 in SPring-8. An X-ray wavelength of 0.496071 Å was used. The sample was added to the borosilicate glass capillary having a radius of 0.2 mm. The crystal structures of the sample in γ-ZrP were refined by Rietan-FP.13 The adsorbed structures were also examined by ab initio simulation using the ab initio total energy and molecular dynamics program VASP (Vienna ab initio simulation program) developed at the Institute für Materialphysik of the Universität Wien.14,15
Figure 1. Adsorption tendency depending on the reaction time for Yb cation in the presence of 12 sorts of rare earth cations. Initial concentrations are 1000 ppm for each rare earth metal cation.
increases, reaching approximately 0.6 or 0.7. For γ-ZrP and HAp, the amount of Yb cation steeply decreases by ion exchange or adsorption and the remainder seems to be zero. From these results, γ-ZrP and HAp are apparently better than other adsorbents. Figure 2 shows the degree of dissolution versus the pH of some adsorbents. These plots indicate the chemical stability of the materials. The amounts of dissolution of HAp and calcium sulfite are apparently larger than those of other samples. α-ZrP, γ-ZrP, and montmorillonite show very low solubility in the strongly acidic region. For the recovery of the rare earth metals
■
RESULTS AND DISCUSSION Comparison of Inorganic Adsorbents. Figure S1 shows the dependence of the time of adsorption of rare earth metal to α-ZrP, γ-ZrP, HAp, and montmorillonite on reaction period in an aqueous solution that includes 12 rare earth metals. From these plots, the amounts of adsorption seem to be constant for longer than 10 h in all samples. Especially, γ-ZrP and HAp can adsorb a lot of rare earth metals simultaneously. Montmorillonite and α-ZrP have relatively poor adsorption competence. Focusing on a variety of rare earth cations in the same adsorbent, we find little difference between the rare earth cations (Figure S1). Figure 1 shows the adsorption tendency depending on the reaction time for Yb cation in the presence of 12 kinds of rare earth cations. These dependencies show that the amount adsorbed is much different by sample species. As shown in Figure S1, only γ-ZrP and HAp show amounts of adsorption definitely larger than those of other samples. For MPS, montmorillonite, and α-TiP, the amounts adsorbed are small, and the consequent amount will be around 0.4. For α-ZrP, calcium sulfite, and porous glass, the adsorbed amount
Figure 2. Degree of dissolution of some adsorbents, α-ZrP, γ-ZrP, HAp, montmorillonite, and calcium sulfite, vs pH. B
DOI: 10.1021/acs.langmuir.6b02747 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir from an urban mine, the waste from the devices should be placed in a strong acid solution to leach included rare earth metals selectively. Because rare earth metal cations show stable states under the acidic conditions, γ-ZrP seems to be better than HAp and the other materials because of its high adsorbing competency and low solubility. Table S1 shows theoretical adsorption capacity, adsorption periods, and degree of dissolution at pH 0 for typical adsorbents. The theoretical adsorption capacities for HAp and calcium sulfite were estimated by replacement of all Ca with Yb. Therefore, these values may be overestimated. The adsorption periods, t0.7, t0.4, and t 0.1, show necessary periods for decreases in Yb concentration to 0.7, 0.4, or 0.1, respectively, in C/C0. Adsorption Behavior of γ-ZrP. Figure 3 shows the reaction time and concentration of rare earth metal cations on a
Figure 4. Amounts of adsorption rare earth metal cation in the singlerare earth cation solution estimated by the acid leaching process. Bars and plots with line show molar ratios and mass fractions, respectively.
light rare earth metals, respectively. For these amounts of adsorption in the presence of a single cation, the difference in adsorption between heavy and light rare earth metals becomes smaller than the adsorption in the presence of multiple cations. From molar ratio x of the adsorbed cation, the degree of ion exchange can be estimated to be around 10% or slightly lower. Figure S2 shows XPS spectra of the Eu-adsorbed sample. From these curves, the surface states did not change as determined by ion exchange because of the similar P 2p and Zr 3d spectra. From Eu 3d spectra, the Eu cation is surely adsorbed to the sample because of the different spectra of Eu 3d; and most of the Eu cation is trivalent, and approximately one-forth seems to be in the divalent state. Figure S3 shows the SXRD patterns of adsorbed γ-ZrP in the presence of a single rare earth cation. At a glance, there are no differences in the patterns over a wide range. Upon careful observation, there is some diffraction in the sample adsorbed with rare earth metal cations in the range from 2.4° to 4.0°. These low levels of diffraction possibly result from the existence of a superlattice. In addition, for the 101, 012, and 102 diffractions in the range from 6° to 6.6°, intensity ratio changes and peaks coalesce around each other in La and Ce samples. The 7−8° diffractions also overlap in these samples. These coalescences and overlaps of the peaks result from changes in lattice parameters. From such peak emergence and overlap, these patterns can be divided into five groups: none and Y; La and Ce; Pr, Nd, Sm, Eu, and Gd; Tb and Dy; and Er and Yb. These groups are designated x111, x221, x241a, x241b, and x131, respectively. These abbreviations, xabc, indicate the multiplicity of each axis for superlattice per original lattice. Table S2 shows these groups and their superlattice candidates. The table also includes the possible combination of diffraction for a superlattice. From Table S2 and Figure 4, the superlattice candidate seems to be no relation to the amount of adsorbed rare earth metal cations. Structural Considerations. For refinement of the superlattice structure, the space groups were presumed to be triclinic. The refined data show no change in lattice parameters α and γ from 90° during coarse refinement. Therefore, α and γ were fixed to 90° for further fine refinements. Figure 5 shows typical refined SXRD data of these superlattice groups, γ-ZrP, La, Eu, Dy, and Yb for x111, x221, x241a, x241b, and x131, respectively. In each superlattice, there are huge amounts of atomic sites as follows: 152, 304, 304, and 114 atomic sites in
Figure 3. Relationship between reaction time and concentration of rare earth metal cation on a logarithmic axis measured in the presence of rare earth metal cations in a solution that includes γ-ZrP. Initial concentrations are 1000 ppm for each rare earth metal cation.
logarithmic axis measured in the presence of rare earth metal cations. For most of the cations, amounts of ≳90% can be adsorbed. From these plots, the amounts of adsorption are apparently different for heavy and light rare earth metal cations. In the case of heavy rare earth metal cations, 99% of the cations can be adsorbed. Such a different adsorption competency might be applicable for adsorption selectivity. However, each cation might prefer a different adsorption site. In the coexistence situation, the adsorption site seems to be complex. Therefore, the interaction of each cation and skeleton of γ-ZrP should be examined by adsorption of a rare earth cation in the singlecation circumstance, and structures of the adsorbed γ-ZrP will be refined by SXRD patterns. Figure 4 shows the amount of adsorption estimated by the acid leaching process. The amounts are expressed as the molar ratio (x) in Zr(PO4)Lnx(H2−3xPO4)·nH2O and as the mass ratio (mass percent) by ion exchange in a single-cation aqueous solution. Filled and empty bars represent data for heavy and C
DOI: 10.1021/acs.langmuir.6b02747 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
superlattice) as empty plots, which are not multiplied. For the a and b axes, the lattice constants seem to be constant or decrease slightly depending on the ionic radius of the rare earth metal cation. For the c axis, the lattice constant steeply increases with the radius of the rare earth cation. The tendency to increase shows a linear relationship, and the slope apparently changes at around 113 pm. For lattice constant β, the slope can be divided into two regions similar to the c axis. Above 113 pm, β tends to increase steeply with ionic radius. These changes in tendency possibly result from a somewhat weak attractive interaction due to the larger rare earth metal cation size; a large cation provides a small electric field. The interlayer spacing calculated from lattice constants was also included in this figure. From the plots, the interlayer spacing increases linearly depending on the ionic radius with a slope of 2.4 nm/nm. Because the slope exceeds the equivalent value of 2, such a large value indicates the existence of a plural site in the direction of the c axis, or a complex adsorption mechanism to the complex γ-ZrP structure that has two sorts of phosphate, H2PO4 and PO4 tetrahedra. Figure 7 shows projected crystal structures of the adsorbed γZrP with La, Eu, Dy, and Yb, which have typical superlattices of x221, x241a, x241b, and x131, respectively. The adsorbed rare earth metal cations seem to be distributed inhomogeneously. One of the reasons for superlattice formation may be the entropy of mixing. When various original cells of γ-ZrP with different amounts adsorption and different locations of rare earth metal cations exist, cells with a similar adsorption circumstance might not be near each other because of the increase in mixing entropy. Such an increase of entropy probably results in superlattice formation. For La, the x221 superlattice structure was used and most of the La atoms exist between H2PO4 tetrahedra on opposite sides of phosphate nanosheets. For Eu and Dy, x241a and x241b superlattices can be used. In these structures, some of the Eu and Dy cations were displaced toward the PO4 tetrahedron. For the Yb cation, the 131 superlattice was used to refine the crystal structure. From these refined models, a small rare earth element seems to occupy the site closest to the phosphate nanosheet. Generally, a rare earth metal cation forms an aqua complex with several H2O molecules in not only the aqueous solution but also the crystal with zeolitic water. In these structures, the distances from rare earth cations to oxygen are around 0.23−0.27 nm.16−19 For coordination of oxygen to the rare earth metal cation, the mean coordination number is 4.4−5.0 when the maximal M−O separation is set to 0.3 nm. Generally, the coordination number of the rare earth aqua complex may be 8− 9. The coordination number is apparently smaller than the general value because of the limited interlayer space. Figure 8 shows the mean relative separation of rare earth metals from the center of the c direction in the unit cell that can be expressed as |z − 0.5|. For the crystal structure of γ-ZrP, the interlayer space exists at the center of the c direction. Thus, this mean relative separation may be affected by interaction between the rare earth metal and inorganic layer (zirconium phosphate block). In this figure, larger values mean a location closer to the ZrP layer. From Figure 8, each superlattice apparently has particular constant values. Except for the x241b lattice, the value decreases depending on the increase in the cation radius of the rare earth metal. For the x241b supercell structure of Dy and Tb, the rare earth metal cation exists exceptionally close to the center location between the ZrP nanosheets. The reason for this phenomenon is difficult to understand. Perhaps the reason is provided by a kind of tetrad
Figure 5. Typical refined SXRD data of superlattice groups γ-ZrP, La, Eu, Dy, and Yb for x111, x221, x241a, x241b, and x131, respectively. The wavelength is 0.496071 Å.
x221, x241a, x241b, and x131 unit cells, respectively. Such huge numbers of sites tend to result in diffusion of any factor for the refinement. Therefore, the isotropic atomic displacement factors of all Zr, P, and O atoms within the phosphate layer were fixed to 1.0. On the other hand, coordinates of the rare earth metals were refined. In all refined data, values of Rwp are around 1.3−2.3% and values of Rp 1.0−1.7. These values indicate that the refinements are sufficiently accurate for consideration of the structural adsorption site. Table 1 shows space group and refined lattice parameters of samples with γZrP, Y, La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Er, and Yb adsorbed. From this table, the standard deviation of the c axis seems to be somewhat larger for La and Ce samples than for others. The β values for La and Ce of >100° are also larger. Figure 6 shows the lattice constants of the original unit cell (not D
DOI: 10.1021/acs.langmuir.6b02747 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir Table 1. SXRD Refinement Results for Each Rare Earth Metal Ion Adsorbed on γ-ZrP superlattice space group a b c β V Rwp Rp superlattice space group a b c β V Rwp Rp
none
Y
La
Ce
Pr
Nd
x111 P21 5.3760(4) 6.6339(3) 12.448(2) 98.699(7) 438.84(8) 1.843 1.343 Sm
x111 P21 5.3560(6) 6.6329(2) 12.543(3) 97.756(9) 441.5(1) 2.046 1.404
x221 P1 10.666(2) 13.2482(2) 13.011(6) 101.50(3) 1802(1) 2.066 1.471
x221 P1 10.667(2) 13.2593(3) 12.922(8) 100.84(3) 1795.(1) 2.243 1.572
x241a P1 10.695(2) 26.5229(6) 12.823(3) 98.73(4) 3595.(1) 2.244 1.666 Er
x241a P1 10.706(1) 26.5274(5) 12.714(4) 98.01(1) 3576.(1) 1.869 1.392 Yb
x241a P1 10.705(1) 26.5280(5) 12.680(3) 97.83(1) 3567(1) 1.819 1.332
Eu
Gd
Tb
Dy
x241a P1 10.6992(8) 26.5100(5) 12.645(1) 97.778(9) 3553.5(5) 1.797 1.329
x241a P1 10.6984(6) 26.5110(4) 12.635(1) 97.750(6) 3550.8(3) 1.514 1.124
x241b P1 10.7229(8) 26.5738(4) 12.625(3) 97.751(8) 3564.7(8) 1.554 1.150
x241b P1 10.6948(8) 26.5099(4) 12.563(3) 97.495(8) 3531.3(9) 1.550 1.181
x131 P1 5.3611(4) 19.8997(9) 12.538(2) 97.807(6) 1325.2(3) 1.385 1.044
x131 P1 5.3630(4) 19.8911(3) 12.397(2) 98.756(6) 1307.1(2) 1.690 1.258
apparently distributed between 110 planes for x241a and along 121̅ planes for x241b. These different distributions can be observed in the SXRD patterns in Figure 5. For x241a, a 110 diffraction line can be observed. However, a 121̅ diffcation line emerged rather than a 110 line for x241b. Therefore, the existence of Dy or Tb near the center location for x241b may be one of the reasons for the different distributions in the superlattices. Regardless, this tendency may result from the strength of the cationic electric field provided by the rare earth cation. That is, a large electric field produces a strong interaction with the ZrP layer, which provides a close separation. In addition, coordination of x and y will be limited at locations close to the ZrP layer because of steric hindrance from H2PO4 tetrahedra. Therefore, the freedom of x and y coordination should depend on the size of the rare earth metal. In other words, the adsorbed small rare earth metal cation will be localized. Such different localization of the adsorption sites may affect strongly formation of the superlattice. Consequently, we think that the cation size of the rare earth metal possibly determines the type of superlattice. Figure 9 shows simulated atomic positions of typical rare earth metals La, Eu, Dy, and Yb, as determined by VASP. These simulations were calculated using the basic unit cell (not superlattice) because superlattices include huge numbers of atoms to make the calculation difficult and time-consuming. Therefore, the original cell was used to ensure equilibria were reached. During MD calculation, the rare earth atoms are displaced to their optimal positions and become stable in the original unit cell. In this figure, the values of |z − 0.5| from the simulation are added. The simulated coordination z seems to be shifted similarly to the refined structure from SXRD. However, the values are around 0.4 times as large as those determined by SXRD refinement, which can be regarded as actual values. Such a discrepancy can be caused by the existence of the superlattice and/or choice of potential. Unfortunately, site occupancy cannot be calculated in the MD simulation by VASP. In these simulations, only a single site of a rare earth metal was calculated unavoidably. Therefore, the cation may not be in a stable location.
Figure 6. Lattice constants of the original unit cell (not superlattice) and basal spacing of adsorbed γ-ZrP in the presence of a single rare earth cation. Squares, triangles, circles, and crosses represent lattice constants a, b, c, and β, respectively.
effect.20 Figure S4 shows three projected continuous superlattice cells along the [1̅11] direction, which is the intersection vector between 110 and 121̅ planes. From these projected structures, rare earth metal cations (colored) seem to be E
DOI: 10.1021/acs.langmuir.6b02747 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Figure 8. Mean relative separation, |z − 0.5|, of rare earth metals from the center of the c direction in the unit cell calculated by Rietveld refinement.
Figure 9. Schematic illustrations of the simulated original unit cell with atomic positions of typical rare earth metals La, Eu, Dy, and Yb determined by VASP.
because rare earth metals dissolve in strongly acidic aqueous solutions. Therefore, γ-ZrP seems to be a very strong candidate for the recovery of rare earth metals. In fact, the amounts of adsorption of rare earth metal on α-ZrP and γ-ZrP were compared, and the amount with γ-ZrP is much larger than that with α-ZrP. The larger amount of adsorption on γ-ZrP probably results from many sites within the interlayer space due to two dissociable protons on the PO4 tetrahedron and to the bumpy surface with H2PO4 tetrahedral grafts. Generally, a rare earth metal cation generally can be adsorbed to a lone pair of the oxygen. Therefore, the rare earth metal cation may tend to adsorb into negatively charged space, including some lone pairs. In addition, smaller rare earth cations show stronger interactions with the anionic ZrP nanosheet because of its strong electric field and can make inroads against the ZrP by Coulombic adsorption. In this paper, ab initio simulations were performed by VASP. The location of adsorption of rare earth cations La, Eu, Dy, and Yb in the simulation relatively fairly agrees with SXRD refinement results. However, the shift from the center position seems to be apparently smaller than that determined by the refinement. On the other hand, the coordination number seems to be in good agreement with the refined structure. In addition, these results show that a smaller cation size produces a larger coordination number. A sample exhibits such a larger coordination number with a smaller rare earth cation. This inclination corresponds to a general tendency because a small cation has a large electric field. In the simulation, PBE
Figure 7. Projected crystal structures of superlattices x221, x241a, x241b, and x131 including La, Eu, Dy, and Yb, respectively, calculated by Rietveld refinement.
For the coordination structure with oxygen, the coordination numbers of La, Eu, Dy, and Yb are 4, 5, 5, and 6, respectively. These coordination numbers are similar to the mean coordination number of the actual structure determined by SXRD refinement. From these considerations, the simulation shows similar behavior for displacement along the direction of the c axis to the actual position of the rare earth metals. Such relatively good agreement suggests that the ab initio simulations may be used to assess the ion exchange behavior of rare earth cations.
■
CONCLUSION Rare earth metals are very useful especially for luminescence and magnetic materials. However, rare earth is relatively expensive, and separation of rare earth metals is difficult F
DOI: 10.1021/acs.langmuir.6b02747 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
(11) Takei, T.; Dobashi, T.; Yonesaki, Y.; Kumada, N.; Kinomura, N. Formation Kinetics of Chemically Vapor-Deposited Carbon on Mesoporous Silica. J. Ceram. Soc. Jpn. 2007, 115, 541−545. (12) Takei, T.; Ota, H.; Dong, Q.; Miura, A.; Yonesaki, Y.; Kumada, N.; Takahashi, H. Preparation of Porous Material from Waste Bottle Glass by Hydrothermal Treatment. Ceram. Int. 2012, 38, 2153−2157. (13) Izumi, F.; Momma, K. Three-dimensional visualization in powder diffraction. Solid State Phenom. 2007, 130, 15−20. (14) Kresse, G.; Furthmüller, J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B: Condens. Matter Mater. Phys. 1996, 54, 11169−11186. (15) Kresse, G.; Joubert, D. From ultrasoft pseudopotentials to the projector augmented-wave method. Phys. Rev. B: Condens. Matter Mater. Phys. 1999, 59, 1758−1775. (16) Giester, G.; Ž ák, Z.; Unfried, P. Syntheses and crystal structures of rare earth basic nitrates hydrates: Part III. [Ln6(μ6−O)(μ3− OH)8(H2O)12(η2−NO3)6](NO3)2·xH2O, Ln = Y, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu; x = 3, 4, 5, 6. J. Alloys Compd. 2009, 481, 116− 128. (17) Luo, Q.-H.; Howell, R. C.; Dankova, M.; Bartis, J.; Williams, C. W.; Horrocks, W. D., Jr.; Young, V. G., Jr; Rheingold, A. L.; Francesconi, L. C.; Antonio, M. R. Coordination of Rare-Earth Elements in Complexes with Monovacant Wells−Dawson Polyoxoanions. Inorg. Chem. 2001, 40, 1894−1901. (18) Morss, L. R.; Rogers, R. D. Syntheses and crystal structures of [M(NO3)2(tpen)][NO3]·3H2O (M= La, Tb), rare earth complexes with strong M−N bonds. Inorg. Chim. Acta 1997, 255, 193−197. (19) Giester, G.; Unfried, P.; Ž ák, Z. Syntheses and crystal structures of some new rare earth basic nitrates II: [Ln6O(OH)8(H2O)12(NO3)6](NO3)2·xH2O, Ln = Sm, Dy, Er; x(Sm) = 6, x(Dy) = 5, x(Er) = 4. J. Alloys Compd. 1997, 257, 175−181. (20) Kawabe, I. Lanthanide tetrad effect in the Ln3+ ionic radii and refined spin-pairing energy theory. Geochem. J. 1992, 26, 309−335.
potentials were used. The PBE potentials are some of the better candidates for ab initio simulation, though they might not be sufficient for calculation of the displacement behavior of the atoms that include f electrons. For more exact simulations, the more suitable potential might be to perform an ab initio simulation.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.6b02747. Additional observations (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The experiments at SPring-8 were performed with the approval of the Japan Synchrotron Radiation Research Institute (JASRI) (Proposals 2013A1299 and 2014A1008). This research was partially supported by the Environment Research and Technology Development Fund (K112031) of the Ministry of the Environment, Japan. Some of the chemicals were provided by Daiichi Kigenso Kagaku Kogyo Co., Ltd., Toagosei Co., Ltd., Kunimine Industries Co., Ltd., and Ogihara Inc., Ltd.
■
REFERENCES
(1) Dutta, T.; Kim, K.-H.; Uchimiya, M.; Kwon, E. E.; Jeon, B.-H.; Deep, A.; Yun, S.-T. Global demand for rare earth resources and strategies for green mining. Environ. Res. 2016, 150, 182−190. (2) Gai, S.; Li, C.; Yang, P.; Lin, J. Recent progress in rare earth micro/nanocrystals: soft chemical synthesis, luminescent properties, and biomedical applications. Chem. Rev. 2014, 114, 2343−2389. (3) Tunsu, C.; Petranikova, M.; Gergorić, M.; Ekberg, C.; Retegan, T. Reclaiming rare earth elements from end-of-life products: A review of the perspectives for urban mining using hydrometallurgical unit operations. Hydrometallurgy 2015, 156, 239−258. (4) Barteková, E.; Kemp, R. National strategies for securing a stable supply of rare earths in different world regions. Resour. Policy 2016, 49, 153−164. (5) Binnemans, K.; Jones, P. T. Perspectives for the recovery of rare earths from end-of-life fluorescent lamps. J. Rare Earths 2014, 32, 195−200. (6) Pérez-Carvajal, J.; Aranda, P.; Berenguer-Murcia, A.; CazorlaAmorós, D.; Coronas, J.; Ruiz-Hitzky, E. Nanoarchitectures based on layered titanosilicates supported on glass fibers: Application to hydrogen storage. Langmuir 2013, 29, 7449−7455. (7) Yu, J. J.; Wang, X. P.; Tao, Y. X.; Hao, Z. P.; Xu, Z. P. Effective NOx decomposition and storage/reduction over mixed oxides derived from layered double hydroxides. Ind. Eng. Chem. Res. 2007, 46, 5794− 5797. (8) Shin, Y.; Ding, H.; Persson, K. A. Revealing the Intrinsic Li Mobility in the Li2MnO3 Lithium-Excess Material. Chem. Mater. 2016, 28, 2081−2088. (9) Takei, T.; Yonesaki, Y.; Kumada, N.; Kinomura, N. Preparation of Oriented Titanium Phosphate and Tin Phosphate/Polyaniline Hybrid Films by Electrochemical Deposition. Langmuir 2008, 24, 8554−8560. (10) Takei, T.; Imazawa, N.; Miura, A.; Kumada, N.; Ogihara, K. Conversion of calcium sulfite waste to hydroxyapatite. Powder Technol. 2013, 237, 400−405. G
DOI: 10.1021/acs.langmuir.6b02747 Langmuir XXXX, XXX, XXX−XXX
