Application of a mild hydrothermal method to the synthesis of mixed transition-metal(II)/uranium(IV) fluorides

Justin Felder , Jeongho Yeon , Mark Smith and Hans-Conrad zur Loye *
Department of Chemistry and Biochemistry, University of South Carolina, Columbia, SC 29208, USA. E-mail:

Received 8th November 2016 , Accepted 22nd December 2016

First published on 9th January 2017

Single crystals of five transition metal uranium fluorides were obtained via the use of a mild hydrothermal route. Uranyl acetate was used as both the uranium source and the reducing agent for an in situ reduction of U(VI) to U(IV). The synthesized materials are present as both two- and three-dimensional structures and contain uranium in 9-fold coordination environments. Magnetic susceptibility measurements indicate that the reported materials remain paramagnetic down to 2 K, with no evidence for the existence of long-range magnetic ordering. Thermogravimetric analysis studies of the reported materials are also presented.


Research in the field of uranium crystal chemistry has become more widespread due to the continued interest by the scientific community in the reaction and crystal chemistry of uranium, especially reduced uranium. Uranium bearing materials have become prevalent due to the large-scale use of uranium in nuclear fuel rods and the resulting unused or unwanted materials remaining within the spent fuel rods. Consequently, research continues to explore new uranium containing materials to better understand the chemistry of such phases and to develop advanced materials for use in the construction of new efficient fuel rod assemblies. Finally, the realization that we have to sequester these materials for long time periods in a safe form inside repositories has driven the study of safe and stable wasteform materials to store existing as well as the continually newly generated radioactive waste that is currently stored at various locations across the United States, as well as in other nuclear power utilizing countries.1–5

Over the past decade an ever increasing number of oxides and fluorides containing uranium in reduced oxidation states have been reported,6–10 in part due to the development of convenient synthetic techniques that provide ready access to oxidation states other than U(VI). Overall, however, the majority of existing and reported new materials still contain uranium in its most oxidized state of U(VI).11–18 We have been interested in exploring the U(IV) chemistry of fluorides and in developing facile synthetic methods to help us reach our goal. We have previously reported a mild hydrothermal method for synthesizing U(IV) fluorides that employs the acetate ion as an organic reducing agent along with a dilute hydrofluoric acid solution that acts as solvent and fluorinating agent. Although other organic reducing agents are available and effective, the use of acetate allows the convenient use of UO2(CH3COO)2 as a starting material, combining the uranium source and the reducing agent.19

Our previous reports have included the discovery of various binary, ternary, and quaternary U(IV) fluorides, including alkali- and transition-metal containing materials. These reports represent an extensive increase in the number of known U(IV) materials and have enabled the exploration of the intriguing optical and magnetic properties of the U(IV) ion.20–24 Herein we continue to expand the library of U(IV) materials and report on the synthesis, thermal, and magnetic properties of five new compositions belonging to two distinct families of uranium fluorides: MUF6·3H2O, and M(H2O)6U2F10·2H2O.


Materials and methods

UO2(CH3COO)2·H2O (International Bio-Analytical Industries, ACS grade), Mn(CH3COO)2·4H2O (Alfa Aesar), Co(CH3COO)2·4H2O (Alfa Aesar, 98%), Ni(CH3COO)2·4H2O (Aldrich, 98%), Zn(CH3COO)2·2H2O (Fisher Scientific) and HF (EMD, 48%) were used as received.

Caution: Although the uranyl acetate used in this experiment contains depleted uranium, standard precautions for handling radioactive materials should be observed. All uranium-containing materials were handled in labs specially designated for the study of radioactive materials.

Caution: Hydrofluoric acid is acutely toxic and corrosive, and must be handled with extreme caution while using appropriate protective gear. If contact with the liquid or vapor occurs, proper treatment procedures should immediately be followed and medical attention promptly sought.

All reported materials were synthesized using a mild hydrothermal technique. MUF6·3H2O, where M = Mn (1) and Zn (2), were prepared by mixing 2 mmol of uranyl acetate and 2 mmol of manganese acetate or zinc acetate with 1 mL of distilled water in a 23 mL PTFE crucible. Likewise, M(H2O)6U2F10·2H2O, where M = Co (3), Ni (4) and Zn (5), were prepared by mixing 2 mmol of uranyl acetate and 2 mmol of cobalt acetate, nickel acetate, or zinc acetate with 1 mL of distilled water in a 23 mL PTFE crucible. Additionally, 1 mL of aqueous HF was added slowly to each of the five reaction mixtures. Each crucible was then sealed in a stainless-steel autoclave and placed in a programmable oven. The oven was ramped to 200 °C and held at that temperature isothermally for 24 hours. After dwelling, the oven was cooled at a rate of 0.1 °C per minute to 40 °C, at which point it was allowed to cool naturally to room temperature.

Once cool, the autoclaves were opened to reveal the product present in the form of single crystals sitting within the reaction liquid. The mother liquor was decanted and the product single crystals were collected by vacuum filtration. The product was washed thoroughly with distilled water, and acetone respectively, and allowed to dry under vacuum. Although each reaction yield was relatively low (∼35% based on uranium), the product was present as a single pure phase except in the case of the zinc materials, which crystallize as approximately 50% ZnUF6·3H2O and 50% Zn(H2O)6U2F10·2H2O, making isolating individual products difficult.

Single-crystal X-ray diffraction

X-ray diffraction intensity data were collected using a Bruker SMART APEX diffractometer (Mo Kα radiation, λ = 0.71073 Å) or a Bruker D8 QUEST diffractometer equipped with a PHOTON 100 CMOS area detector and an Incoatec microfocus source (Mo Kα radiation, λ = 0.71073 Å).25 The raw area detector data frames were processed with SAINT+.26 An absorption correction based on the redundancy of equivalent reflections was applied to the data with SADABS.26 The reported unit cell parameters were determined by least-squares refinement of a large array of reflections taken from each data set. Initial structural models were obtained with SHELXS using direct methods.27 Subsequent difference Fourier calculations and full-matrix least-squares refinement against F2 were performed with SHELXTL28 or SHELXL-201427 using the ShelXle interface.29

The compounds MnUF6·3H2O and ZnUF6·3H2O crystallize in the monoclinic system. The space group C2/c was consistent with the pattern of systematic absences in the intensity data, and was confirmed by structure solution. The compounds are isostructural with the NiUF6·3H2O phase.30 The asymmetric unit consists of a manganese or zinc atom located on a crystallographic inversion center (Wyckoff site 4c), one uranium atom and one water oxygen atom located on a two-fold axis of rotation (site 4e), and three fluorine atoms and one oxygen atom located on general positions (site 8f). All atoms were refined with anisotropic displacement parameters. All hydrogen atoms were located in difference maps and refined free with isotropic displacement parameters. No deviation from full occupancy was observed for either of the metal atoms.

The compounds Co(H2O)6U2F10·2H2O, Ni(H2O)6U2F10·2H2O, and Zn(H2O)6U2F10·2H2O crystallize in the monoclinic system. The space group P21/c was consistent with the pattern of systematic absences in the intensity data, and was confirmed by structure solution. The compounds are isostructural with the cobalt/neptunium analog CoNp2F10(H2O)8.31 The asymmetric unit consists of a M(II) atom located on a crystallographic inversion center (Wyckoff site 2d), one uranium, five fluorine, four oxygen and eight hydrogen atoms, all of which are located on positions of general crystallographic symmetry (site 4e). All atoms were refined with anisotropic displacement parameters. All hydrogen atoms were located in difference maps and refined isotropically with their O–H distances restrained to be similar (SHELX SADI instruction). The H–H distance in water O4 of Zn(H2O)6U2F10·2H2O was further restrained to 1.25(4) Angstroms using a SHELX DANG instruction. No deviation from full occupancy was observed for either of the metal atoms. Crystallographic data and selected interatomic distances for all reported materials can be found in Tables 1 and 2 respectively.

Table 1 Single crystal structure refinement information
  1 2 3 4 5
Empirical formula MnUF6(H2O)3 ZnUF6(H2O)3 CoU2F10(H2O)8 NiU2F10(H2O)8 ZnU2F10(H2O)8
Color Green Green Green Green Green
Crystal size (mm) 0.06 × 0.08 × 0.12 0.04 × 0.07 × 0.10 0.02 × 0.16 × 0.20 0.04 × 0.06 × 0.08 0.04 × 0.06 × 0.08
F.W. (g mol−1 F.U.) 461.02 471.45 869.12 868.9 875.56
Temperature (K) 294(2) 100(2) 294(2) 100(2) 100(2)
Wavelength (Å) 0.71073 0.71073 0.71073 0.71073 0.71073
Crystal system Monoclinic Monoclinic Monoclinic Monoclinic Monoclinic
Space group C 2/c C 2/c P21/c P21/c P21/c
Unit cell parameters
a (Å) 12.366(4) 12.1137(7) 11.0745(4) 10.9605(7) 11.0116(5)
b (Å) 6.975(2) 6.9186(4) 7.0989(3) 7.0441(4) 7.0571(3)
c (Å) 8.081(3) 7.9813(4) 8.8499(3) 8.8514(6) 8.8476(4)
β (°) 93.201(6) 92.9113(15) 94.1040(10) 94.532(2) 94.2424(13)
Volume (Å3) 695.90(4) 668.05(6) 693.87(5) 681.25(7) 685.66(5)
Z 4 4 2 2 2
Density (Mg m−3) 4.4 4.687 4.16 4.236 4.241
Absorption coefficient (mm−1) 25.134 27.884 24.614 25.234 25.448
Number of reflections collected 4427 14[thin space (1/6-em)]520 8989 40[thin space (1/6-em)]599 28[thin space (1/6-em)]820
Number of independent reflections 872 1221 1728 3655 2015
Data/constraints/parameters 872/0/65 1221/0/66 1728/6/129 3655/28/130 2015/29/130
Goodness-of-fit on F2 1.077 1.13 1.102 1.125 1.104
Final R indices R 1 = 0.0152 R 1 = 0.01 R 1 = 0.0210 R 1 = 0.0150 R 1 = 0.0118
wR2 = 0.0372 wR2 = 0.0235 wR2 = 0.0524 wR2 = 0.0272 wR2 = 0.0272
Largest diff. peak and hole (e Å−3) 0.817 and −1.245 1.281 and −0.84 1.314 and −1.410 1.382 and −1.005 1.099 and −1.106

Table 2 Selected interatomic distances
M = Mn, Zn 1 2 M = Co, Ni, Zn 3 4 5
U(1)–F(1) 2.238(2) 2.2947(11) U(1)–F(1) 2.215(3) 2.2075(11) 2.2092(15)
U(1)–F(2) 2.284(2) 2.2461(11) U(1)–F(2) 2.301(2) 2.3052(11) 2.3055(14)
U(1)–F(3) 2.318(2) 2.3143(11) U(1)–F(2) 2.423(2) 2.4136(10) 2.4168(13)
U(1)–F(3) 2.431(2) 2.4040(12) U(1)–F(3) 2.340(2) 2.3341(10) 2.3371(13)
U(1)–O(2) 2.585(5) 2.536(2) U(1)–F(3) 2.363(2) 2.3497(10) 2.3437(13)
U(1)–F(4) 2.343(2) 2.3424(11) 2.3412(13)
M(1)–F(1) 2.093(2) 2.0278(11) U(1)–F(4) 2.347(2) 2.3549(11) 2.3553(13)
M(1)–F(2) 2.103(2) 2.0079(11) U(1)–F(5) 2.351(2) 2.3489(10) 2.3498(13)
M(1)–O(1) 2.199(3) 2.1090(15) U(1)–F(5) 2.353(2) 2.3511(10) 2.3506(13)
M(1)–O(1) 2.075(3) 2.0498(13) 2.0796(18)
M(1)–O(2) 2.088(3) 2.0442(14) 2.0767(18)
M(1)–O(3) 2.120(3) 2.0722(14) 2.1085(18)

Powder X-ray diffraction

Powder X-ray diffraction (PXRD) data were collected on polycrystalline samples that were ground from the product single crystals. Powder data were collected on a Rigaku Ultima IV diffractometer using Cu Kα radiation. Data were collected over the two-theta range 10° to 65°, with a 0.02° step size.

Energy dispersive spectroscopy (EDS)

Energy dispersive spectroscopy was performed directly on product single crystals mounted on an SEM stud with carbon tape. EDS was performed with a Tescan Vega-3 SEM equipped with a Thermo EDS attachment. The SEM was operated in low vacuum mode and utilized a 30 kV accelerating voltage and 20 second accumulating time.

Magnetic properties

Magnetic measurements were carried out using a Quantum Design MPMS 3 SQUID magnetometer. Field cooled (FC) and zero-field cooled (ZFC) measurements were performed under an applied magnetic field of 0.1 T in the temperature range of 2 K–300 K. Magnetization measurements were collected at 2 K by sweeping the applied magnetic fields between −5 T and 5 T. All magnetic data were collected on polycrystalline powders obtained by grinding the single crystal products. The raw data were corrected for radial offset and shape effects according to the method described by Morrison.32

Thermal properties

Thermal property measurements were performed on both single crystals and polycrystalline powder samples on a TA SDT Q600 TGA. Samples were heated in a nitrogen flow from room temperature to 600 °C. Low temperature, isothermal measurements were performed in an attempt to dehydrate the samples by heating single crystals at 60 °C for 6 hours.

Results and discussion

Synthetic considerations

Mild hydrothermal synthesis provides many advantages over existing solid state reactions, including the ability to access reduced oxidation states of many species. This is especially important for uranium chemistry where the reduced U(IV) ion is much larger and thus significantly less soluble than the more oxidized U(VI) ion. This makes the synthesis of U(IV) containing materials difficult; however, by utilizing an in situ reduction step that uses the aqueous acetate ions activated by HF, soluble U(IV) precursors are created that are incorporated into reduced U(IV) containing products, especially fluorides. Although other organic reducing agents are effective, including tartrates and oxalates,33,34 the commercial availability of uranyl acetate makes the co-availability of the uranium source and reducing agent convenient. In addition to uranyl acetate, transition metal (M2+) acetates may be used to provide soluble metal species while simultaneously increasing the concentration of the reducing acetate ion.

In addition to providing the stated advantages over conventional solid-state synthetic techniques, the mild hydrothermal synthesis is often preferred to high temperature (∼350–700 °C) and high pressure hydrothermal methods because of its much lower operating temperature (∼150–250 °C) and low autogenous pressures enabling these reactions to be run in inexpensive off the shelf PTFE-lined steel autoclaves. Our definition of mild hydrothermal reactions are those run below the critical point of water, while regular hydrothermal reactions are run above the critical point of water. Overall, the lower cost, temperature, and pressure requirements of mild hydrothermal syntheses make it an attractive option for materials discovery.

This synthetic technique works well for the reported compositions, providing phase-pure single crystal samples for facile property measurements, purity analysis (by PXRD), and structure determination. The one exception is the two zinc containing compositions that crystallize as an approximate 50% mixture. The similarity in size and crystal morphology between the two phases made separating them exceedingly difficult and, while crystals suitable for structure determination are present for both families, the difficulty in separating the two phases prevents certain bulk property measurements, such as magnetic susceptibility, from being conducted. Attempts to optimize the synthesis to favor one phase over the other were unsuccessful.

Crystal structure

MUF6·3H2O. MnUF6·3H2O and ZnUF6·3H2O crystallize in the monoclinic space group C2/c, and are isostructural. The structure is built of two unique structural units: 9-coordinate U(IV) polyhedra and 6-coordinate M(II) octahedra, which are both shown in Fig. 1. The metal octahedra are distorted, consisting of four short equatorial fluoride ions, and two long axial aqua ligands. The uranium polyhedron consists of 8 fluoride ions capped by an aqua ligand. The uranium polyhedra each share four fluoride ions with each other to form infinite UF4F4/2(H2O) chains running along the c crystallographic direction. The polyhedra within the chains alternate orientation by 180° so that the capping aqua ligand is alternately up (+b direction) or down (−b direction). This is shown in Fig. 2, which depicts the one-dimensional uranium fluoride chains. The metal octahedra are isolated from one another, but are connected to the uranium fluoride chains. Each of the four equatorial fluoride ions of the metal octahedron is shared by a different uranium ion, so that each octahedron can be said to be corner sharing with four different uranium polyhedra. In addition, two of the uranium polyhedra belong to one uranium fluoride chain, and the other two to another, such that the metal octahedra function to bridge the one-dimensional uranium chains. Fig. 3 shows a representation of the metal octahedra bridging two uranium fluoride chains. Fig. 4 shows the overall structure that emphasizes the three-dimensional nature of the crystal structure.
image file: c6qi00491a-f1.tif
Fig. 1 The local coordination environments of both the U4+ and M2+ ions in structures 1 and 2 (MUF6·3H2O). The uranium (left) is coordinated by eight fluoride ligands, and one aqua ligand. The metal (right) is coordinated by four fluoride ligands and two aqua ligands. Uranium is shown in dark green, the M2+ in purple, fluorine in light green, oxygen in red, and hydrogen in black.

image file: c6qi00491a-f2.tif
Fig. 2 The connectivity of uranium polyhedra in (MUF6·3H2O). The UF4F4/2(H2O) polyhedra share an edge via two bridging fluoride ligands, and form infinite chains along the [c] direction. The polyhedra alternate orientation by 180° so that the aqua ligands remain as far apart as possible. Uranium is shown in dark green, fluorine in light green, oxygen in red, and hydrogen in black.

image file: c6qi00491a-f3.tif
Fig. 3 The infinite UF4F4/2(H2O) chains are bridged by the metal octahedra. Each metal octahedron is connected to four uranium polyhedra, two per chain. Each chain connects to four other chains in this manner, although only one of these connections is shown here. Metal polyhedra are shown in purple, uranium in dark green, fluorine in light green, oxygen in red, and hydrogen in black.

image file: c6qi00491a-f4.tif
Fig. 4 An overall depiction of materials 1 and 2 as seen down the [c] direction. The structure is built up of one dimensional chains of UF8(H2O) polyhedra which connect to four other chains via bridging metal octahedra. Alternatively, this three-dimensional structure can be thought of as being constructed from layers (in the [b,c] plane) of uranium chains separated and connected by layers of metal polyhedra. The metal octahedra are shown in purple, uranium in dark green, fluorine in light green, oxygen in red, and hydrogen in black.
M(H2O)6U2F10·2H2O. Co(H2O)6U2F10·2H2O, Ni(H2O)6U2F10·2H2O, and Zn(H2O)6U2F10·2H2O crystallize in a two-dimensional structure that is more typical of a mild hydrothermal synthesis. Again, the structure is built of two different units, a UF9 polyhedron and a M(H2O)6 octahedron. Fig. 5 shows both local coordination environments. The uranium centers share fluoride ions via either corner or edge sharing with other uranium centers thus creating a two-dimensional layer. These layers are separated by layers of isolated metal hexa-aqua octahedra and interstitial waters, which are held in place between the aqua and fluoride ligands solely by hydrogen bonding. Fig. 6 shows these layers as they relate to the overall structure. The uranium-containing layer is present as a two-dimensional U2F8F2/22− sheet formed by corner and edge sharing UF9 polyhedra. These polyhedra corner share along the c axis, and edge share along the b-axis, which can be seen in Fig. 7.
image file: c6qi00491a-f5.tif
Fig. 5 The local coordination polyhedra of both uranium and the metal(II) ion in structures 3–5 (MU2F10·8H2O). Uranium (left) is present as UF9 polyhedra. The high coordination number is characteristic of U(IV). The M(II) ion is present as hexa-aqua octahedral complexes. The M(II) polyhedra are shown in blue to differentiate them from structures 1 and 2. As before, uranium is dark green, fluorine is light green, oxygen is red and hydrogen is black.

image file: c6qi00491a-f6.tif
Fig. 6 An overall depiction of structures 3–5 (MU2F10·8H2O) along with a break down of the layered structure. The uranium fluoride layer forms a two-dimensional U2F102− sheet with polyhedra that both corner- and edge share. These uranium fluoride layers are interspersed by metal layers. The metal polyhedra are isolated and separated by interstitial waters. This represents a true layered structure where the layers are connected only by intermolecular forces, in this case hydrogen bonding interactions. Uranium is shown in green, the metal ions in blue, fluorine in light green, oxygen in red, and hydrogen in black.

image file: c6qi00491a-f7.tif
Fig. 7 A close look at the uranium fluoride layer. Uranium centers are bridged by two fluoride ligands in the [b] direction (edge sharing), and in the [c] direction by one fluoride ligand (corner sharing). Red circles highlight shared edges, and shared corners are highlighted by blue circles for clarity. Uranium is shown in dark green, and fluorine in light green.

The U(IV) ion commonly exists in high coordination environments with 8 or more ligands surrounding the U(IV) center. As the library of U(IV) materials expands, it becomes possible to observed recurring structural motifs across series of U(IV) containing materials. One such structural motif is the capped trigonal prismatic coordination environment that U(IV) often adopts. This trigonal prismatic environment can be found in numerous previously reported U(IV) containing materials,8,22,24,35,37 and also all five compositions reported herein. The trigonal prism may be capped by two or three ligands on the rectangular faces of the prism depending on the particular compound. The high coordination of uranium dictates that any given U center will be surrounded by at least 8 ligands. In the case of fluoride ligands, each polyhedron would have an overall negative charge, favoring the polyhedra to condense into chains, layers, or 3D frameworks, rather than remaining isolated molecular species. In the case of MnUF6·3H2O and ZnUF6·3H2O the uranium centers are present in the aforementioned tricapped trigonal prismatic environment. In this particular case, the trigonal prism is highly distorted due to the presence of a capping aqua ligand.

In the case of Co(H2O)6U2F10·2H2O, Ni(H2O)6U2F10·2H2O, and Zn(H2O)6U2F10·2H2O, the uranium centers form two-dimensional layers that are separated by the charge balancing M(II) octahedra. These two-dimensional sheets are identical to layers found in AU2F9 (A = K, Rb).22 Unlike AU2F9, where the layers are connected vertically, the layers in Co(H2O)6U2F10·2H2O, Ni(H2O)6U2F10·2H2O, and Zn(H2O)6U2F10·2H2O are separated, a consequence of the large hexa-aqua complexes that reside between the layers.

Powder X-ray diffraction and EDS

Powder diffraction data, which were collected over the range 10°–65° 2θ, are displayed for all materials in Fig. S1–S4. ZnUF6·3H2O and Zn(H2O)6U2F10·2H2O could not physically be separated into phase-pure samples and, consequently, the powder diffraction data for the mixed-phase sample are shown. A Whole Pattern Fit was performed on the powder diffraction data for all other samples to demonstrate phase purity.

Energy dispersive spectroscopy confirmed the presence of U and the respective transition metal in all of the reported materials. EDS is a semi-quantitative measurement that cannot be used to deduce any reliable quantitative information about lighter elements such as oxygen or fluorine. Nonetheless, EDS was able to confirm the presence of both oxygen and fluorine in the crystals as a qualitative measure. EDS measurements were performed on several crystals in a batch to confirm that results were representative of the entire set.

Magnetic properties

Each material contains two magnetic ions: U4+ and a M2+ species (M = Mn, Co, and Ni). Accurate magnetic measurements on the Zn phases were unable to be obtained due to difficulties in separating the pure phases from the reaction mixture, and are thus not reported here.
MUF6·3H2O. The structure of MnUF6·3H2O consists of infinite chains of UF4F4/2(H2O) polyhedra that are edge shared through two fluoride ligands. Additionally, the infinite chains are connected by MnF4(H2O)2 octahedra via fluoride ligands, which also serve to bridge two uranium polyhedra on each chain. Magnetic coupling via the superexchange mechanism has been well documented to occur in fluorides,36 and the greater orbital extent of the of the 5f uranium orbitals compared to 4f lanthanide orbitals open up the potential for interesting fd magnetic coupling that is not typically observed. The full magnetic susceptibility data are shown in Fig. S8.

The magnetic susceptibility versus temperature data of MnUF6·3H2O are shown in Fig. 8. The inverse susceptibility data were fit to the Curie–Weiss law over the region of 200 K–300 K where the sample exhibits paramagnetic behavior. The material deviates slightly from Curie–Weiss behavior below 50 K. The observed magnetic moment of 6.92μB agrees with the calculated value of 6.92μB. For the calculation the moment for manganese was assumed to be the spin-only value, while a uranium moment of 3.58μB, based on full Russel–Saunders, coupling was used.

image file: c6qi00491a-f8.tif
Fig. 8 The magnetic susceptibility and inverse magnetic susceptibility of material MnUF6·3H2O. The measurement was taken from 2 K to 300 K in an applied field of 0.1 T. The zero-field cooled data is shown. The material is paramagnetic down to 2 K, with only slight deviation from Curie–Weiss behavior at low temperature. The expected transition to a non-magnetic singlet ground state is not observed.

The slight ferromagnetic deviation seen in the susceptibility data is difficult to attribute to any one cause. It is unlikely that the manganese ions order magnetically since such a large moment would contribute greatly to the magnetic susceptibility. The uranium chains present the possibility of magnetic order; however, no long-range spin ordering is observed in the magnetic susceptibility data. Interestingly, we do not observe the transition from a magnetic triplet state to a low temperature nonmagnetic singlet state that we have previously observed in U(IV) fluoride systems.37–39 It is possible that the strong moment of manganese overshadows the nonmagnetic transition, however the moment of uranium is strong enough that its loss should be observed, indicating that it remains paramagnetic through 2 K.

M(H2O)6U2F10·2H2O. Compounds Co(H2O)6U2F10·2H2O and Ni(H2O)6U2F10·2H2O also have two different magnetic ions, however the transition metal is present in isolated octahedra, connected only by hydrogen bonding interactions and is thus not expected to order. The uranium ions are present as UF9 polyhedra which both corner and edge share to form a two-dimensional sheet. The complex interconnectedness of uranium polyhedra allows for several possible exchange pathways. Fig. 9 shows the magnetic susceptibility for Co(H2O)6U2F10·2H2O and Ni(H2O)6U2F10·2H2O. Both compounds show a broad deflection from Curie–Weiss behavior at approximately 25 K. The Curie–Weiss fit for these materials was performed over the temperature range of 150 K–300 K.
image file: c6qi00491a-f9.tif
Fig. 9 The magnetic susceptibility and inverse magnetic susceptibility of materials 3 and 4 (MU2F10·8H2O M = Co, Ni). The measurement was taken from 2 K to 300 K in an applied field of 0.1 T. The zero-field cooled data is shown. The materials are paramagnetic except for a slight positive deviation from Curie–Weiss behavior below 50 K. The expected transition to a non-magnetic singlet state is not observed.

The use of spin-only magnetic contributions for the transition metal and full Russel–Saunders coupling for uranium generated effective magnetic moments of 7.26μB calculated for Co(H2O)6U2F10·2H2O (7.02μB observed) and 5.65μB calculated (5.64μB observed) for Ni(H2O)6U2F10·2H2O, both of which are in good agreement with the observed values. Unlike material 1 which has a complex three-dimensional structure, the layered structure and isolated M2+ cations suggest that any magnetic ordering is likely to come only from U–F–U interactions. As with MnUF6·3H2O, the slight deviation from Curie–Weiss behavior is too ambiguous to suggest the presence of magnetic ordering.

Unexpectedly, both materials in this family also do not exhibit the expected transition to a low temperature nonmagnetic ground state, remaining paramagnetic down to 2 K. In fact, where a decreased moment is expected for a nonmagnetic transition, we observe a slightly increased moment (compared to Curie–Weiss behavior) at low temperatures.

Thermal properties

Both reported material families contain crystalline water, both interstitial and bound water (in the case of M(H2O)6U2F10·2H2O) or only bound water in the case of MUF6·3H2O. We were motivated to perform thermal experiments to probe the structural transformations observed by the movement of water in and out of the structure. The powder diffraction data for the thermal decomposition products are found in Fig. S5–S7.
MUF6·3H2O. Fig. 10 shows the TGA plot for MnUF6·3H2O. Two distinct dehydration events can be observed, the first beginning at 80 °C, and the second at 128 °C. The material decomposes into the binary fluorides UF4 and MnF2 as a direct result of the second dehydration. Attempts to dehydrate single crystals of the compound at 90 °C to force a single crystal to single crystal transition almost succeeded. Preliminary structural studies revealed that the UF8(H2O) polyhedron is dehydrated, leaving infinite chains of UF8 polyhedra with only minor changes to the lattice parameters. Fig. 11 shows the transformation of the infinite uranium fluoride chains. Attempts to obtain full structure solutions on the dehydrated crystals, however, were unsuccessful due to the deteriorated crystal quality during heating and the accompanying water loss. Table S1 gives atomic positions, and Table S2 gives lattice parameters and R factors from the preliminary single crystal study.
image file: c6qi00491a-f10.tif
Fig. 10 Thermogravimetric analysis results for 1 (MnUF6·3H2O). Shown is the plot of weight loss as a function of temperature. Results show that one water is lost by 130 °C, with the final two waters lost by 331 °C, accompanied by the decomposition of the material.

image file: c6qi00491a-f11.tif
Fig. 11 The transformation of the uranium fluoride chain induced by the loss of a crystalline water. Results obtained from preliminary single crystal diffraction data. Uranium is shown in dark green, fluorine in light green, oxygen in red, and hydrogen in black.
M(H2O)6U2F10·2H2O. Co(H2O)6U2F10·2H2O and Ni(H2O)6U2F10·2H2O undergo four distinct dehydration events. Fig. 12 shows the TGA plot for Co(H2O)6U2F10·2H2O, which is representative of both Co(H2O)6U2F10·2H2O and Ni(H2O)6U2F10·2H2O. The initial dehydration at 53 °C yields a polycrystalline powder, which was determined via PXRD to be CoU2F10·4H2O. This represents a total loss of four waters, two of which are interstitial. The other waters are lost from the cobalt hexa-aqua octahedron, leaving square planar Co(H2O)4. Fig. 13 shows the change in the powder diffraction data due to the dehydration. Further heating destroys the crystal structure and leaves UF4 and CoF2 as the major decomposition products.
image file: c6qi00491a-f12.tif
Fig. 12 Thermogravimetric analysis for material 3 (CoU2F10·8H2O). Shown is the plot of weight loss versus temperature. Results show that the first dehydration is accompanied by four waters lost, rapidly followed by the remaining four waters and decomposition by 301 °C.

image file: c6qi00491a-f13.tif
Fig. 13 Powder X-ray diffraction data before and after TGA of material 3 (CoU2F10·8H2O). Blue is the calculated powder pattern of material 3 obtained from the single crystal structure solution. Red is the observed pattern before TGA. Black shows the change in the pattern caused by the thermal dehydration at 60 °C.


The mild hydrothermal technique coupled with a convenient in situ reduction by the acetate ions was shown again to be a versatile tool for synthesizing new U(IV) containing materials. The materials presented herein showcase the structural versatility of the U(IV) ion by adopting both three dimensional and layered two dimensional structures under nearly identical synthetic conditions. In addition, the magnetic properties demonstrate that, although there is a lack of apparent magnetic order, the 5f unpaired electrons can remain viable for potential long-range order despite the fact that they commonly undergo a transition to a low temperature nonmagnetic singlet ground state due to thermal depopulation of excited f orbital states.


Research supported by the US Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering under award DE-SC0008664.


  1. P. C. Burns, R. A. Olson, R. J. Finch, J. M. Hanchar and Y. Thibault, J. Nucl. Mater., 2000, 278, 290–300 CrossRef CAS.
  2. J. M. Jackson and P. C. Burns, Can. Mineral., 2001, 39, 187–195 CrossRef CAS.
  3. K.-T. Kim, J. Nucl. Mater., 2010, 404, 128–137 CrossRef CAS.
  4. J. Ling, J. M. Morrison, M. Ward, K. Poinsatte-Jones and P. C. Burns, Inorg. Chem., 2010, 49, 7123–7128 CrossRef CAS PubMed.
  5. Y. Akifumi, N. Yoshihiro, U. Sadao and O. Tsutomu, Nucl. Technol., 2012, 179, 309–322 Search PubMed.
  6. G. B. Jin and L. Soderholm, J. Solid State Chem., 2015, 221, 405–410 CrossRef CAS.
  7. M. Keskar, S. K. Sali, K. Krishnan and S. Kannan, J. Nucl. Mater., 2016, 478, 245–255 CrossRef CAS.
  8. M. D. Ward, G. N. Oh, A. Mesbah, M. Lee, E. Sang Choi and J. A. Ibers, J. Solid State Chem., 2015, 228, 14–19 CrossRef CAS.
  9. M. D. Ward, E. A. Pozzi, M. Lee, R. P. Van Duyne, E. S. Choi and J. A. Ibers, Inorg. Chem., 2015, 54, 3055–3060 CrossRef CAS PubMed.
  10. N. Yu, V. V. Klepov, S. Neumeier, W. Depmeier, D. Bosbach, E. V. Suleimanov and E. V. Alekseev, Eur. J. Inorg. Chem., 2015, 2015, 1562–1568 CrossRef CAS.
  11. U. Betke and M. S. Wickleder, Eur. J. Inorg. Chem., 2012, 2012, 306–317 CrossRef CAS.
  12. A. Mer, S. Obbade, M. Rivenet, C. Renard and F. Abraham, J. Solid State Chem., 2012, 185, 180–186 CrossRef CAS.
  13. J. Plášil, A. R. Kampf, A. V. Kasatkin, J. Marty, R. Škoda, S. Silva and J. Čeka, Mineral. Mag., 2013, 77, 2975–2988 CrossRef.
  14. C. M. Read, G. Morrison, J. Yeon, M. D. Smith and H.-C. zur Loye, Inorg. Chem., 2015, 54, 6993–6999 CrossRef CAS PubMed.
  15. E. Reynolds, B. J. Kennedy, G. J. Thorogood, D. J. Gregg and J. A. Kimpton, J. Nucl. Mater., 2013, 433, 37–40 CrossRef CAS.
  16. L. B. Serezhkina, M. S. Grigor'ev, A. S. Makarov and V. N. Serezhkin, Radiochemistry, 2015, 57, 20–25 CrossRef CAS.
  17. S. Wu, P. M. Kowalski, N. Yu, T. Malcherek, W. Depmeier, D. Bosbach, S. Wang, E. V. Suleimanov, T. E. Albrecht-Schmitt and E. V. Alekseev, Inorg. Chem., 2014, 53, 7650–7660 CrossRef CAS PubMed.
  18. A. Collomb, M. Gondrand, M. S. Lehmann, J. J. Capponi and J. C. Joubert, J. Solid State Chem., 1976, 16, 41–48 CrossRef CAS.
  19. J. Yeon, M. D. Smith, A. S. Sefat, T. T. Tran, P. S. Halasyamani and H.-C. zur Loye, Inorg. Chem., 2013, 52, 8303–8305 CrossRef CAS PubMed.
  20. J. Yeon, M. D. Smith, A. S. Sefat and H.-C. zur Loye, Inorg. Chem., 2013, 52, 2199–2207 CrossRef CAS PubMed.
  21. J. Yeon, M. D. Smith, J. Tapp, A. Möller and H.-C. zur Loye, J. Am. Chem. Soc., 2014, 136, 3955–3963 CrossRef CAS PubMed.
  22. J. Yeon, M. D. Smith, J. Tapp, A. Möller and H.-C. zur Loye, Inorg. Chem., 2014, 53, 6289–6298 CrossRef CAS PubMed.
  23. J. Yeon, M. D. Smith, G. Morrison and H.-C. zur Loye, Inorg. Chem., 2015, 54, 2058–2066 CrossRef CAS PubMed.
  24. J. Yeon, M. D. Smith, J. Tapp, A. Möller and H.-C. zur Loye, J. Solid State Chem., 2016, 236, 83–88 CrossRef CAS.
  25. APEX2 Version 2014.9-0, SAINT+ Version 8.34A and SADABS Version 2014/4, Bruker Analytical X-ray Systems, Inc., Madison, Wisconsin, USA, 2014 Search PubMed.
  26. SMART Version 5.625, SAINT+ Version 6.45 and SADABS Version 2.05, Bruker Analytical X-ray Systems, Inc., Madison, Wisconsin, USA, 2001 Search PubMed.
  27. G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystallogr., 2008, 64, 112–122 CrossRef CAS PubMed.
  28. APEX2 Version 2014.9-0, SAINT+ Version 8.34A and SADABS Version 2014/4, Bruker Analytical X-ray Systems, Inc., Madison, Wisconsin, USA, 2014 Search PubMed.
  29. ShelXle: a Qt graphical user interface for SHELXL CrossRef PubMed; C. B. Hubschle, G. M. Sheldrick and B. J. Bittrich, J. Appl. Crystallogr., 2011, 44, 1281–1284 CrossRef PubMed.
  30. A. C. Bean, T. A. Sullens, W. Runde and T. E. Albrecht-Schmitt, Inorg. Chem., 2003, 42, 2628–2633 CrossRef CAS PubMed.
  31. A. Cousson, H. Abazli and J. Jove, J. Less-Common Met., 1985, 109, 155–168 CrossRef CAS.
  32. G. Morrison and H.-C. zur Loye, J. Solid State Chem., 2015, 221, 334–337 CrossRef CAS.
  33. A. J. Cortese, B. Wilkins, M. D. Smith, J. Yeon, G. Morrison, T. T. Tran, P. S. Halasyamani and H.-C. zur Loye, Inorg. Chem., 2015, 54, 4011 CrossRef CAS PubMed.
  34. D. Abeysinghe, M. D. Smith, J. Yeon, G. Morrison and H.-C. zur Loye, Cryst. Growth Des., 2014, 14, 4749–4758 CAS.
  35. H. Abazli, A. Cousson, A. Tabuteau, M. Pages and M. Gasperin, Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem., 1980, 36, 2765–2766 CrossRef.
  36. A. Tressaud and J. M. Dance, Adv. Inorg. Chem. Radiochem., 1977, 20, 133–188 CrossRef CAS.
  37. P. M. Almond, L. Deakin, A. Mar and T. E. Albrecht-Schmitt, J. Solid State Chem., 2001, 158, 87–93 CrossRef CAS.
  38. Y.-L. Lai, R.-K. Chiang, K.-H. Lii and S.-L. Wang, Chem. Mater., 2008, 20, 523–530 CrossRef CAS.
  39. C.-M. Wang, C.-H. Liao, P.-L. Chen and K.-H. Lii, Inorg. Chem., 2006, 45, 1436–1438 CrossRef CAS PubMed.


Electronic supplementary information (ESI) available. CCDC 1515454–1515458. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6qi00491a

This journal is © the Partner Organisations 2017