Seaborgite, LiNa₆K₂(UO₂)(SO₄)₅(SO₃OH)(H₂O), the First Uranyl Mineral Containing Lithium

The Blue Lizard mine in Red Canyon, Utah, is a remarkable source of new minerals, especially sodium-uranyl sulfates. The astounding diversity and relatively high structural complexity of uranyl-sulfate minerals were recently emphasized by Gurzhiy and Plášil (2019). A large number of stable combinatorial linkages of uranyl and sulfate tetrahedra are possible, with the topological arrangements appearing to be strongly affected by at least three parameters: pH (Plášil et al. 2014), cation content, and water content. In general, sodium-uranyl-sulfate minerals follow the same structural unit topology trends as do other uranyl minerals (Lussier et al. 2016), where uranyl polyhedra preferentially polymerize into extended structures via linkages through their equatorial vertices, most often forming infinite chain or infinite sheet topologies. However, finite cluster topologies are relatively abundant among the sodium-uranyl-sulfate minerals, for reasons that are not completely clear. Understanding the hierarchical arrangements of these structures and how conditions of formation influence the crystallized topologies is important to understanding the crystal-chemical nature of U-S systems, and for uranyl mineralogy as a whole.

The new Blue Lizard mine uranyl sulfate seaborgite, described herein, contains essential sodium; however, it also includes essential potassium and, most significantly, lithium. While sodium and, especially potassium, form relatively weak bonds within such structures, the role of lithium is different. Lithium-oxygen bonds, particularly in LiO₄ tetrahedral coordination, are somewhat stronger and, in the seaborgite structure, serve to further link (or polymerize) the uranyl sulfate clusters.

Seaborgite is named in honor of American chemist Glenn T. Seaborg (1912–1999) who was involved in the synthesis, discovery, and investigation of 10 transuranium elements (including seaborgium), earning him a share of the 1951 Nobel Prize in Chemistry. Seaborg's scientific accomplishments are numerous and changed the course of world history. Perhaps most notably, Seaborg and coworkers discovered plutonium in 1940 and he isolated the first weighable sample of plutonium in 1942. The Manhattan Project produced the first plutonium-fueled nuclear bomb that was detonated in New Mexico at the Trinity test site on July 16, 1945. Seaborg served as Chairman of the United States Atomic Energy Commission from 1961 to 1971 during which time he worked to advance nuclear energy. Seaborg was a strong proponent for arms control and the peaceful use of nuclear energy.

The new mineral and name were approved by the Commission on New Minerals, Nomenclature and Classification of the International Mineralogical Association (IMA 2019-087). One holotype specimen of seaborgite is deposited in the collections of the Natural History Museum of Los Angeles County, Los Angeles, California, U.S.A., catalog number 74163.

Seaborgite was found underground in the Blue Lizard mine (37°33′26″N 110°17′44″W), Red Canyon, White Canyon District, San Juan County, Utah, U.S.A. The mine is about 72 km west of the town of Blanding, Utah, and about 22 km southeast of Good Hope Bay on Lake Powell. Detailed historical and geologic information on the Blue Lizard mine is described elsewhere (cf. Kampf et al. 2015) and is primarily derived from a report by Chenoweth (1993).

Abundant secondary uranium mineralization in Red Canyon is associated with post-mining oxidation of asphaltum-rich sandstone beds laced with uraninite and sulfides in the damp underground environment. Seaborgite was found in an area rich in K-bearing sulfates (e.g., metavoltine, voltaite, zincovoltaite), along with several other potentially new sodium and potassium uranyl sulfate minerals. Potassium enrichment has so far not been observed in secondary uranyl mineralization elsewhere in the Blue Lizard mine nor in any of the nearby U deposits in Red Canyon that we have investigated, and this is the first uranyl mineral found that contains essential Li. It seems likely that K and Li are sourced from Li- and K-bearing clays in the sediments.

Seaborgite is a very rare mineral in the secondary mineral assemblages of the Blue Lizard mine. It occurs on a thick crust of gypsum overlaying matrix comprised mostly of subhedral to euhedral, equant quartz crystals that are recrystallized counterparts of the original grains of the sandstone. Other secondary phases found in close association with seaborgite are copiapite, ferrinatrite, ivsite, metavoltine, römerite, and other potentially new uranyl sulfate minerals.

Crystals of seaborgite are long flattened prisms or blades, up to about 0.2 mm in length, typically in radiating sprays (Fig. 1). Crystals are elongated on [100], flattened on {010}, and exhibit the forms {100}, {010}, {001}, and {101} (Fig. 2). Twinning was observed optically under crossed polars and is either by reflection on {001} or by rotation around [001].

The mineral is light green-yellow and transparent with vitreous luster and very pale-yellow streak. Seaborgite exhibits bright lime-green fluorescence under a 405 nm laser. It has a Mohs hardness of about 2½ based on scratch tests. The mineral has brittle tenacity, curved or conchoidal fracture, and one good cleavage on {100}. The density measured by flotation in a mixture of methylene iodide and toluene is 2.97(2) g/cm³. The calculated density is 3.015 g/cm³ for the empirical formula (using Li, Na, and K based on the structure refinement) and single-crystal cell; 3.004 g/cm³ for the ideal formula. The mineral is immediately soluble in H₂O at room temperature.

Seaborgite is optically biaxial (–) with α = 1.505(2), β = 1.522(2), γ = 1.536(2) measured in white light. The 2V measured using extinction data analyzed with EXCALIBRW (Gunter et al. 2004) is 85(1)°; the calculated 2V is 83.6°. The dispersion is moderate, r < ν. The partially determined optical orientation is X ^ a ≈ 10°. Crystals are pleochroic with X colorless, Y and Z light green-yellow; X < Y ≈ Z. The Gladstone–Dale compatibility, 1 – (K_P/K_C), (Mandarino 2007) is –0.009 (superior) based on the empirical formula (using Li, Na, and K based on the structure refinement) using k(UO₃) = 0.118, as provided by Mandarino (1976).

Raman spectroscopy was conducted on a Horiba XploRA PLUS using a 532 nm diode laser, 50 μm slit, 2400 gr/mm diffraction grating, and a 100× (0.9 NA) objective. The spectrum, recorded from 4000 to 100 cm^–1, is shown in Figure 3.

Two weak bands with centers at 3570 and 3475 cm^–1 are assigned to ν(OH) stretching vibrations. Using the empirically derived equation of Libowitzky (1999), the calculated O···O distances of the corresponding hydrogen bonds are between ~3.0 and ~2.8 Å, in reasonable agreement with the hydrogen bond lengths determined from the structure refinement. Several very broad low intensity bands centered at ~2600 and ~1800 cm^–1 are probably overtones or combination bands. No apparent band related to the ν₂(δ) bending vibrations of H₂O is present at approximately 1600 cm^–1, which is not surprising considering the low sensitivity of Raman for the non-symmetrical vibrations.

There has been no reliable computational/theoretical research focused on differentiating SO₄ and SO₃OH in Raman spectra; therefore, our assignments of the vibrations connected with the sulfate tetrahedra in seaborgite are tentative. The ν₃(SO₄/SO₃OH) antisymmetric stretching vibrations occur as weak bands at 1203, 1194, 1173, 1139, and 1091 cm^–1. Several weak to strong bands at 1045, 1026, 1015, 1002, and 979 cm^–1 are assignable to the ν₁ symmetric stretching vibrations of SO₄ and SO₃OH groups. The presence of six symmetrically unique SO₄ tetrahedra in the seaborgite structure lead to the multiple split bands in this region. The weak band at 917 cm^–1 is related to the ν₃(UO₂)²⁺ antisymmetric stretching vibration, while the band at 885 cm^–1 is assigned to the ν(S–OH) mode (Plášil et al. 2013). The ν₁(UO₂)²⁺ symmetric stretching vibration is present as a very strong band at 850 cm^–1. Bartlett and Cooney (1989) provided an empirical relationship to derive the approximate U-O_Ur bond lengths from the band position assigned to the UO₂²⁺ stretching vibrations, which gives 1.76 Å (ν₁) and 1.77 Å (ν₃), in excellent agreement with the average U1-O_Ur bond length from the X-ray data: 1.757 Å. At least seven overlapping weak bands between 657 and 586 cm^–1 are attributable to the ν₄(δ)(SO₄/SO₃OH) bending vibrations, with centers at 657, 647, 641, 634, 621, 605, and 586 cm^–1. Those at 479, 463, 444, and 425 cm^–1 belong to the ν₂(δ)(SO₄/SO₃OH) bending vibrations. A band at 250 cm^–1 is attributable to the ν₂(δ)(UO₂)²⁺ bending vibrations and/or possibly to ν(U–O_eq) bending modes. The remaining bands arise due to unassigned phonon modes.

Chemical analyses for all elements except Li (8 points on 2 crystals) were performed on a JEOL JXA-8230 electron microprobe using Probe for EPMA software. The analytical conditions used were 10 kV accelerating voltage, 10 nA beam current, and a beam diameter of 10 μm. Raw X-ray intensities were corrected for matrix effects with a φρ(z) algorithm (Pouchou and Pichoir 1991). Time-dependent intensity corrections were applied to data for Na and K. No other elements were detected by EDS and wavescans at multiple currents and beam sizes showed no N above background. Crystals of seaborgite experienced considerable damage under the electron beam. The amounts of Na, and to a lesser extent K, reported in the EPMA are significantly lower than those based on the structure refinement; this is attributed to the failure of the time-dependent intensity corrections to fully account for the volatility of Na and K.

Li, Na, and U were measured using laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS). The ion signals for Li, Na, and U from 2 crystal aggregates were measured using an Element 2 sector field high-resolution inductively coupled plasma mass spectrometer (Thermo-Fisher Scientific) in low mass resolution mode coupled with an UP-213 (New Wave Research) Nd:YAG deep UV (213 nm) laser ablation system. Prior to the ablation of samples, the Element 2 was tuned using a multi-element solution containing 1 ng·g^–1 of each Li, In, and U to obtain maximum ion sensitivity. The laser ablation analyses involved acquiring background ion signals for 60 s with the laser on and shuttered, and this was followed by 60 s of data acquisition. The laser was operated using a 30 μm spot size, repetition rate of 5 Hz, and 65% power output, which corresponded to a fluence of ~8.4 J/cm². Two areas on two crystals were examined using single spot analyses. The background corrected ion signals (counts per second) obtained for Li, Na, and U are reported as an atomic ratio relative to that recorded for U, which was used to calculate a corresponding wt% oxide value, as absolute abundances could not be determined due to a lack of an appropriate matrix-matched external standard. The analytical value obtained for Na, while higher than that obtained by EPMA, is also significantly lower than that based on the structure refinement, as is the value obtained for Li; the “undermeasurements” are probably due to the fact that we cannot adequately account for the ionization efficiency differences.

Because insufficient material is available for a direct determination of H₂O, it has been calculated based upon the structure determination (U+S = 7 apfu, O = 27 apfu). Analytical data are given in Table 1. The empirical formula using Na measured via EPMA is Li_0.79Na_5.02K_2.02(UO₂)(SO₄)₅(SO₃OH)(H₂O), which has a charge deficiency of 1.17 due to undermeasurements of Li, K, and Na. The empirical formula using Na measured via LA-ICP-MS is Li_0.79Na_5.19K_2.02(UO₂)(SO₄)₅(SO₃OH)(H₂O), which has a charge deficiency of 1.00 due to undermeasurements of Li, K, and Na. The empirical formula using Li, Na, and K based on the structure refinement is Li_1.00Na_5.81K_2.19(UO₂)(SO₄)₅(SO₃OH)(H₂O). The simplified formula is LiNa₅(Na,K)K₂(UO₂)(SO₄)₅(SO₃OH)(H₂O) and the ideal formula is LiNa₆K₂(UO₂)(SO₄)₅(SO₃OH)(H₂O), which requires Li₂O 1.37, Na₂O 17.08, K₂O 8.65, UO₃ 26.28, SO₃ 44.13, H₂O 2.48, total 100 wt%.

Both powder and single-crystal X-ray studies were carried out using a Rigaku R-Axis Rapid II curved imaging plate microdiffractometer with monochromatized MoKα radiation. For the powder study, a Gandolfi-like motion on the φ and ω axes were used to randomize the sample, which consisted of several crystals. Observed d values and intensities were derived by profile fitting using JADE 2010 software (Materials Data, Inc., Livermore, California). Data are given in Online Material¹ Table S1. The observed powder diffraction pattern compares very well with the pattern calculated from the crystal structure (Fig. 4)

The relatively small crystal size only allowed structure data to be collected to 40 °2θ; consequently, the data to parameter ratio (5.45) was less than optimal. The Rigaku CrystalClear software package was used for processing the structure data, including the application of an empirical absorption correction using the multi-scan method with ABSCOR (Higashi 2001). The structure was solved by the charge-flipping method using SHELXT (Sheldrick 2015a). SHELXL-2016 (Sheldrick 2015b) was used for the refinement of the structure.

A cation site with scattering power, coordination, bond lengths, and bond valence appropriate for Li was located. (The presence of Li was independently confirmed by LA-ICP-MS.) Nine cation sites other than S, H, and U were located. Two fully occupied by K (K1 and K2), five fully occupied by Na (Na1 thru Na5), one split Na site (Na6a and Na6b), and one occupied jointly by Na and K (Na/K), which refined to Na_0.62K_0.38. All non-hydrogen atoms were successfully refined with anisotropic displacement parameters, but several O sites exhibited strongly oblate and/or prolate ellipsoids. This may indicate some local disorder (or local “flexibility”) in the structure, but the splitting of the sites did not appear warranted. At least some of the ellipsoid anisotropy may be due to inadequacies in the empirical absorption correction, although a shape-based absorption correction yielded a higher R_int and did not lessen the ellipsoid anisotropies.

Difference-Fourier syntheses located all H atom positions associated with the H₂O groups, which were then refined with soft restraints of 0.82(3) Å on the O-H distances and 1.30(3) Å on the H-H distances and with the U_eq of the OH H atom set to 1.5 times the OH O atom and that for each H₂O H atom set to 1.2 times that of the H₂O O atom. The crystallographic data can be found in the original CIF (as Online Material¹). Selected bond distances are given in Table 2 and a bond valence analysis in Table 3.

The U site in the structure of seaborgite is surrounded by seven O atom sites forming a squat pentagonal bipyramid. This is typical coordination for U⁶⁺ in which the two short apical bonds of the bipyramid constitute the uranyl group (Burns 2005). The two apical O atoms of the bipyramids (O_Ur) form short bonds with the U, and this unit comprises the UO₂²⁺ uranyl group. Five equatorial O atoms (O_eq) complete the U coordination. All O_eq atoms also participate in SO₄ groups. The UO₇ bipyramid is surrounded by five SO₄ tetrahedra centered by S1(×2), S2, S3, and S4, each of which shares one O_eq corner of the UO₇ bipyramid. One additional SO₄ tetrahedron (centered by S5) and one SO₃OH tetrahedron (centered by S6) are not linked to the UO₇ bipyramid.

The UO₇ bipyramids are linked to one another by pairs of S1O₄ tetrahedra to form a [(UO₂)₂(SO₄)₈]^4– uranyl-sulfate cluster, which is topologically identical to the cluster in the structure of bluelizardite, Na₇(UO₂)(SO₄)₄Cl(H₂O)₂ (Plášil et al. 2014); the two clusters differ in the relative rotation of 1- and 2-connected tetrahedra only, so they can be transformed one into another without the breaking of chemical bonds (Fig. 5). The Li is in regular tetrahedral coordination, typical for Li. Each of the vertices of the LiO₄ tetrahedron is shared with an SO₄ tetrahedron (2× S2O₄ and 2× S5O₄). Two LiO₄ tetrahedra and two S2O₄ tetrahedra form a four-member corner-sharing (LiO₂)₂(S2O₄)₂ ring in the {100} plane; the [(UO₂)₂(SO₄)₈]^4– uranyl-sulfate clusters and the (LiO₂)₂(S2O₄)₂ rings link through the S2O₄ tetrahedra to form a band lying in the {100} plane and extending along [010] (Fig. 6). The S5O₄ tetrahedra form links in the [100] direction between LiO₄ tetrahedra in adjacent bands. The UO₇ pentagonal bipyramids, LiO₄ tetrahedra, and SO₄ tetrahedra (centered by S1 through S5) thereby form a thick heteropolyhedral layer parallel to {001} (Fig. 7). The S6O₃OH tetrahedron does not participate in this layer linkage.

The two K sites (K1 and K2) are both eightfold coordinated, as is the mixed Na/K site. The Na1, Na2, Na4, and Na5 sites are sixfold coordinated, the Na3 site is sevenfold coordinated and the split Na6 sites (Na6a and Na6b) are each fivefold coordinated. All bond valence sums (Table 3) for these large monovalent cation sites are reasonable. The K1 site is located at the center of channels that run through the center of the heteropolyhedral layer. The other large cation sites K2, NaK, Na1, Na2, Na3, Na4, and Na5, as well as the S6O₃OH tetrahedron and the OW27 H₂O group occupy the space between and around the periphery of the heteropolyhedral layers with bonding between them resulting in a framework (Fig. 8). Among all structures containing U⁶⁺, that of seaborgite is unique.

Nevertheless, in spite of the structural uniqueness of seaborgite, it is noteworthy that its structural complexity, I_G,total = 510.17 bits/cell (after Krivovichev 2012, 2013, 2014, 2018), falls within the most frequent range of complexities observed for uranyl sulfates, 500 to 600 bits/cell (Gurzhiy and Plášil 2019).

Seaborgite is the first uranyl mineral that contains structurally essential lithium, although many synthetic inorganic compounds contain both lithium and uranium. Only two synthetic uranyl sul-fates contain lithium, and these are exotic nanoscale cage cluster compounds (Qiu et al. 2017). In seaborgite, the lithium cations are in tetrahedral coordination with the four oxygen atoms contributed by monodentate sulfate tetrahedra. Whereas the fundamental building blocks consisting of uranyl dimers connected to eight sulfate tetrahedra in seaborgite have been observed in other minerals and synthetic compounds, the presence of the lithium-centered tetrahedra stitches these together with additional sulfate tetrahedra to form highly unique uranyl sulfate layers. Within these layers are infinite rods consisting of lithium and sulfate tetrahedra that are made possible by the small size of the lithium cation. The large hydrated radius of lithium that consists of two hydration spheres and its high enthalpy of hydration indicates it is unlikely that extended uranyl sulfate units containing lithium polyhedra exist in the aqueous solution from which seaborgite crystallized. Incorporation of lithium tetrahedra in the structure of seaborgite occurred during crystallization caused by evaporation likely close to dryness, and the uncommon coexistence of sufficient uranyl ions and lithium cations in the same natural aqueous solution, combined to produce this unusual mineral and its corresponding structure.

Funding to the University of Notre Dame was provided by the Chemical Sciences, Geosciences and Biosciences Division, Office of Basic Energy Sciences, Office of Science, U.S. Department of Energy, Grant No. DE-FG02-07ER15880. Funding to J.P. was provided by the Czech Science Foundation (20-11949S). This study was also funded by the John Jago Trelawney Endowment to the Mineral Sciences Department of the Natural History Museum of Los Angeles County.

S. Krivovichev, an anonymous reviewer, and the Technical Editor are thanked for constructive comments, which improved the manuscript. G. Diego Gatta is thanked for shepherding the manuscript through the review process.