Hypervalent Bismuthides La3MBi5 (M = Ti, Zr, Hf) and Related

Apr 10, 2017 - In Zintl–Klemm compounds, main-group elements form various types of polyanionic networks such as (Sb2)4– dumbbells in Ba5Sb4(1) and...
1 downloads 10 Views 2MB Size
Download PDF
Article pubs.acs.org/IC

Hypervalent Bismuthides La3MBi5 (M = Ti, Zr, Hf) and Related Antimonides: Absence of Superconductivity Taito Murakami,† Takafumi Yamamoto,† Fumitaka Takeiri,† Kousuke Nakano,† and Hiroshi Kageyama*,†,‡ †

Department of Energy and Hydrocarbon Chemistry, Graduate School of Engineering, Kyoto University, Nishikyo-ku, Kyoto 615-8510, Japan ‡ CREST, Japan Science and Technology Agency (JST), Kawaguchi, Saitama 332-0012, Japan S Supporting Information *

ABSTRACT: We successfully synthesized the ternary bismuthides La3MBi5 (M = Ti, Zr, Hf). These compounds crystallize in the hexagonal Hf5Sn3Cu-anti type structure (space group: P63/mcm) consisting of face-sharing MBi6 octahedral chains and hypervalent Bi linear chains, both separated by La atoms. Magnetic susceptibility and electrical resistivity measurements revealed that all of the compounds, including the solid solution La3Ti(Bi1−xSbx)5, exhibit a Pauli paramagnetic behavior without any trace of superconductivity down to 1.85 K, as opposed to a recently reported 4 K superconductivity in La3TiSb5. The absence of superconductivity is supported by first-principles band calculations of La3TiBi5 and La3TiSb5 that demonstrate similar electronic structures with threedimensional Fermi surfaces.

1. INTRODUCTION In Zintl−Klemm compounds, main-group elements form various types of polyanionic networks such as (Sb2 ) 4− dumbbells in Ba5Sb41 and Sb− zigzag chains in CaSb2.2 These polyanionic networks can be rationalized by Zintl−Klemm electron-counting rules that assume a complete charge transfer from the most electropositive metals to main-group elements. However, there exist several networks of heavy p-group elements that do not follow the rules and are instead explained in terms of hypervalent concepts proposed by Hoffmann.3 For example, an infinite square net in SmSb2-, ZrCuSi2-, and SrZnBi2-type structures has a formal electron count of 6 (i.e., Sb−). Other hypervalent networks include two-dimensional (2D) square honeycomb and zigzag square lattices and onedimensional (1D) linear chains.3 These compounds show metallic conductivity derived from holes in the p band, and intriguing transport properties such as charge density waves (CDW), superconductivity, and quantum Hall effects have been reported.4−6 Recently, new ternary antimonides RE3MSb5 (RE = rare earth; M = Ti, Zr, Hf, Nb) and U3MSb5 (M = Sc, Ti, Zr, Hf, V, Nb, Cr, Mn) with a hexagonal Hf5Sn3Cu-anti type structure (315 structure) have been reported by Mar and co-workers.7−11 As shown in Figure 1, the 315 structure consists of two types of 1D chains running along the hexagonal c axis, face-sharing MSb6 octahedral chains and hypervalent linear Sb chains, being separated by RE or U atoms. The Sb−Sb length within the linear chain (3.1−3.2 Å) is slightly longer than classical Zintl− Klemm bonds (2.8−2.9 Å).3 Several interesting physical properties associated with their quasi-1D crystal structure are © 2017 American Chemical Society

Figure 1. La3MBi5 (M = Ti, Zr, Hf) viewed along the hexagonal c axis. Green, blue, and brown spheres represent La, M, and Bi atoms, respectively.

known.9 The most remarkable is superconductivity in RE3TiSb5 (RE = La, Ce, Nd) with a critical temperature Tc of 4.1 K for La, 4.0 K for Ce, and 3.7 K for Nd, as revealed by magnetic susceptibility and resistivity measurements.12 From firstprinciples calculations considering the [TiSb5]9− substructure, it was concluded that these compounds possess a highly 1D electronic structure derived from Ti 3d bands, which is responsible for the occurrence of superconductivity. Received: January 20, 2017 Published: April 10, 2017 5041

DOI: 10.1021/acs.inorgchem.7b00031 Inorg. Chem. 2017, 56, 5041−5045

Article

Inorganic Chemistry

stable LaBi phase, rather than La3TiBi5. The excess Ti may be included in unknown impurity phases. Similar results were obtained for La−Zr−Bi and La−Hf−Bi systems. All of the samples prepared were highly air and moisture sensitive. Figure 2a−c displays SXRD patterns taken at room temperature. The diffraction peaks could also be indexed on

In comparison with the 315 antimonides, bismuthides of this structural type have been much less explored, and only La3MBi5 (M = Mg, Sc) and RE3MnBi5 have been reported.13−16 It is thus interesting to explore La3MBi5, since it then allows one to examine how the physical properties of the 315 phase vary by substituting Bi for Sb. In this paper, we report on the synthesis and structural characterization of the new bismuthides La3MBi5 (M = Ti, Zr, Hf) using a standard solid-state reaction. Magnetic susceptibility and electrical resistivity measurements revealed that La3MBi5 compounds are Pauli paramagnetic metals with no superconducting transition observed down to 1.85 K, in contrast to La3TiSb5 with Tc = 4.1 K.12 Unexpectedly, we observed no superconductivity for the whole solid solution La3Ti(Bi1−xSbx)5 even for x = 1 (La3TiSb5). In combination with first-principles calculations, we discuss the origin of the controversial observations.

2. EXPERIMENTAL SECTION Polycrystalline samples of La3MBi5 (M = Ti, Zr, Hf) and the solid solution La3Ti(Bi1−xSbx)5 (x = 0, 0.2, 0.4, 0.6, 0.8, 1) were synthesized via high-temperature solid-state reactions. La chips (99%, Wako) and Ti (99.9%, Kojund Chemical), Zr (99.9%, Kojund Chemical), Hf (98%, Kojund Chemical), Sb (99.9%, Kojundo Chemical), and Bi (99.9%, Kojundo Chemical) powder were mixed, weighed, and pelletized in a nitrogen-filled drybox with oxygen and moisture levels below 0.1 ppm. Each pellet was wrapped with a Ta foil, sealed in an evacuated quartz tube, heated to 1000 °C at a rate of 100 °C/h, maintained at that temperature for 48 h, and cooled to room temperature at a rate of 100 °C/h. The powder X-ray diffraction (XRD) experiments were performed with a Bruker AXS D8 Advance instrument using Cu Kα radiation. Since the samples obtained are highly air sensitive, they were covered with Kapton tape during the XRD measurements. Synchrotron XRD (SXRD) experiments were carried out using a large Debye−Scherrer camera installed at SPring-8 BL02B2 (λ = 0.41987 Å). The finely ground powder samples were filtered by a strainer with 32 μm square holes and put into a Pyrex capillary with inner 0.1 mm diameter. The samples were diluted with LiH powder due to the high X-ray absorptivity of Bi. The obtained SXRD data were analyzed by Rietveld refinements using the JANA2006 program.17 Magnetization measurements as a function of temperature were carried out using a magnetic property measuring system (Quantum Design, MPMS) down to 1.85 K. The temperature dependence of electrical resistivity was measured by a standard dc four-terminal method using a physical property measuring system (Quantum Design, PPMS) down to 2 K. We calculated band dispersions for La3TiSb5 and La3TiBi5 using the full potential linearized augmented plane wave (FPLAPW) method implemented in the WIEN2k code.18 The Perdew−Burke−Ernzerhof (PBE) parametrization19 of the generalized gradient approximation (GGA) was applied as an exchange-correlation term with and without spin−orbit coupling (SOC) and 8000 k points were used. The cutoff of the plane was determined by RKmax = 7.0, where R = muffin tin radius was set to be 2.5 au for all atoms. Spin polarization was not considered in these calculations.

Figure 2. Refined synchrotron powder XRD patterns of (a) La3TiBi5, (b) La3ZrBi5, and (c) La3HfBi5. Red crosses and green and blue lines represent observed, calculated, and difference profiles, respectively. First, second, and third green ticks from top to bottom represent the positions of Bragg peaks of La3MBi5, Bi, and LaBi, respectively. There are several unindexed peaks (not assigned as any superreflections), which were excluded from the refinements. (d) Lattice constants as a function of x for La3Ti(Bi1−xSbx)5.

the basis of the hexagonal cell, along with Bi (and LaBi for M = Zr) impurity phases. LiH used for dilution was not detected. The lattice parameters were determined from Le Bail refinements as given in Table 1. The obtained values, a = 9.70−9.76 Å and c = 6.46−6.52 Å, are slightly longer than those of La3MSb5 (a = 9.53−9.59 Å and c = 6.28−6.36 Å),7,8 reflecting a larger covalent radius of Bi in comparison with Sb. Rietveld refinements were performed assuming the hexagonal Hf5Sn3Cu-anti type structure (space group P63/mcm) with La at 6g (x, 0, 1/4), M at 2b (0, 0, 0), Bi(1) at 6g (x, 0, 1/4), and Bi(2) at 4d (1/3, 2/3, 0). Bi and LaBi (for M = Zr) were included as secondary and ternary phases. The refinements converged successfully, yielding reasonable crystallographic

3. RESULTS AND DISCUSSION When a stoichiometric ratio of La, Ti, and Bi was reacted, we only observed LaBi and Ti. However, when an excess Ti condition was applied, a new hexagonal phase with a = 9.70 Å and c = 6.46 Å appeared in the XRD profile, along with Bi. The hexagonal pattern is similar to that of La3TiSb5, suggesting a successful formation of La3TiBi5. The best quality (i.e., least impurity phase) was achieved from a nominal composition of La3Ti2Bi5. The amount of LaBi increases when it is prepared in a lesser Ti/La ratio, probably due to faster formation of the 5042

DOI: 10.1021/acs.inorgchem.7b00031 Inorg. Chem. 2017, 56, 5041−5045

Article

Inorganic Chemistry

(La3+)3M4+(Bi3−)3(Bi2−)2, where Bi2− represents the hypervalent Bi chain. Given the absence of a guest-free compound (i.e., La3Bi5), “La3Bi5” may become stable only when transition metals (M = Ti, Zr, Hf) are incorporated, in contrast to Zr5MSb3, which also forms Zr5Sb3.25 The magnetic susceptibilities of La3TiBi5, La3ZrBi5, and La3HfBi5 in an applied field of 1000 Oe are shown in Figure 3a.

Table 1. Crystallographic Data for La3TiBi5, La3ZrBi5, and La3HfBi5a a (Å) c (Å) GOF Rp (%) Rwp (%) a

La3TiBi5

La3ZrBi5

La3HfBi5

9.69876(6) 6.46447(6) 2.09 2.52 4.77

9.76130(6) 6.51972(7) 3.03 4.45 6.30

9.75092(8) 6.50865(8) 3.44 4.41 7.37

Space group P63/mcm; Z = 2.

parameters and reliability factors (Rwp = 4.77% and GOF = 2.09 for Ti, Rwp = 6.30% and GOF = 3.03 for Zr, and Rwp = 7.37% and GOF = 3.44 for Hf). When the occupation factor g of each atom was allowed to vary, it converged to unity, verifying the stoichiometric composition. The refined parameters are summarized in Table 2. A large thermal parameter is observed Table 2. Atomic Coordinates and Equivalent Isotropic Displacement Parameters (100 Å2) for La3TiBi5, La3ZrBi5, and La3HfBi5

a

atom

site

ga

La Ti Bi(1) Bi(2)

6g 2b 6g 4d

1 1 1 1

La Zr Bi(1) Bi(2)

6g 2b 6g 4d

1 1 1 1

La Hf Bi(1) Bi(2)

6g 2b 6g 4d

1 1 1 1

x La3TiBi5 0.61791(16) 0 0.2527(1) 1/3 La3ZrBi5 0.6187(2) 0 0.26135(13) 1/3 La3HfBi5 0.6175(2) 0 0.26009(15) 1/3

y

z

Uiso/100 Å2

0 0 0 2/3

1/4 0 1/4 0

0.94(4) 2.5(2) 0.777(18) 0.72(2)

0 0 0 2/3

1/4 0 1/4 0

0.78(5) 0.56(10) 0.71(2) 0.77(3)

0 0 0 2/3

1/4 0 1/4 0

0.72(5) 0.66(6) 0.76(3) 0.78(3)

Figure 3. (a) Temperature dependence of magnetic susceptibility for La3TiBi5, La3ZrBi5, and La3HfBi5 under an applied field of 1000 Oe. (b) Temperature dependence of electrical resistivity ρ for La3TiBi5 and La3TiBi5.

All of the susceptibility data are positive and almost temperature-independent down to 1.85 K, a behavior characteristic of a Pauli paramagnetic metal. As shown in Figure 3b, the resistivity for La3TiBi5 decreases gradually as the temperature is decreased down to 2 K, and no sign of superconductivity was found, as opposed to La3TiSb5 showing superconductivity at 4 K.12 In order to obtain more insight into the physical properties of La3TiBi5 and La3TiSb5, we performed first-principles calculations based on the refined structures. We note that Moore et al. conducted extended Hückel calculations on La3TiSb5 but only considered the [TiSb5]9− substructure and observed disperse bands crossing the Fermi level only along the Γ−A direction (parallel to the c* axis).12 Furthermore, the density of states (DOS) curve has a large maximum located at the Fermi level mainly derived from a Ti 3d orbital. From these observations, the authors claimed that La3TiSb5 is a highly anisotropic metal with the main electron conduction coming from Ti 3d orbitals. Our calculation is the complemental study that can identify the influence of La atoms, as well as Bi, on the electronic structures. We found that the band dispersions of La3TiSb5 (Figure S3a in the Supporting Information) exhibit no gap at the Fermi level, in agreement with the previous study. However, while there are three disperse bands cross along the Γ−A direction, there are several other bands that cut the Fermi level in the Γ−M−K plane (perpendicular to the c* axis), which makes the electronic structure of La3TiSb5 threedimensional (3D) like. This means that the addition of a La atom to the calculations substantially changes the electronic structure. We also calculated band structures with SOC, as shown in Figure S5a in the Supporting Information. It is seen that the inclusion of SOC induces splitting of the bands around the Fermi level in the Γ−M−K plane, but several bands still remain showing a 3D character of the electronic structures even in the presence of SOC. In addition, the DOS of La3TiSb5

g represents the occupancy factor of each site.

for Ti, but it is likely to be extrinsic given the presence of impurities, the low scattering factor of Ti (vs La and Bi), and the observation of normal values for M = Zr, Hf. Figure 1 represents the crystal structure of La3MBi5. It is isostructural with RE3MSb5 (RE = rare earth, M = Ti, Zr, Hf, Nb), U3MSb5 (M = Sc, Ti, Zr, Hf, V, Nb, Cr, Mn), and RE3MBi5 (M = Mg, Sc, Mn).7−16 The cell parameters of title compounds are comparable to those of known bismuthides (a = 9.71−9.79 Å and c = 6.48−6.56 Å). The La−Bi length within the LaBi9 tricapped trigonal prism ranges from 3.266 to 3.539 Å, which is comparable to that observed in LaBi (3.289 Å).20 The M−Sb length increases in the order M = Ti, Hf, Zr (2.937 Å for Ti, 3.027 Å for Zr, and 3.013 Å for Hf), which is consistent with the covalent radii of the transition metals and is comparable to those observed in binary bismuthides such as Ti2Bi (2.864−2.937 Å),21 ZrBi2 (3.045−3.143 Å),22 and HfBi2 (3.045−3.189 Å).23 The M−M distance in La3MBi5 is longer than those observed in elemental metal, suggesting that metal− metal bonding is not obvious. The Bi−Bi distance within the linear chain (3.232−3.260 Å) is slightly longer than that of the zigzag Bi chains in EuBi2 (3.225 Å)24 classified as classical Zintl−Klemm systems. The hypervalent concept to the title compounds gives the formal charges of 5043

DOI: 10.1021/acs.inorgchem.7b00031 Inorg. Chem. 2017, 56, 5041−5045

Article

Inorganic Chemistry

4.0 K for Ce, and 3.7 K for Nd), despite distinct differences in the unit cell parameters. From these results, we suspect that the reported superconductivity originates from elemental Sn (Tc ≈ 4 K) or Sn-related alloy used as a flux during synthesis. To support this, the critical field of Hc = 370 Oe at 2 K is close to the reported value for Sn.26

(Figure S3b) reveal a pseudogap like strong suppression at the Fermi level, a prominent feature that has not been seen in the former calculations. Most La d states are located above the Fermi energy, but there are non-negligible contributions to the DOS at the Fermi level, indicating that La is not perfectly ionic and causes the 3D electronic structure. We obtained very similar band dispersions and DOS curves for the isostructural bismuthide La3MBi5 (both with and without SOC), as shown in Figure 4 and Figures S5 and S6 in the Supporting Information, which is rather surprising given the remarkable difference in physical properties between La3TiBi5 and La3TiSb5.12

4. CONCLUSION We successfully synthesized the new bismuthides La3MBi5 (M = Ti, Zr, Hf) crystallizing in the Hf5Sn3Cu-anti type structure. Magnetic susceptibility and electronic resistivity measurements indicated that all of the compounds, including the solid solution La3Ti(Bi1−xSbx)5, are Pauli paramagnetic metals, without any sign of superconductivity down to 1.85 K as opposed to the previous report on La3TiSb5. First-principles calculations for La3TiBi5 and La3TiSb5 revealed 3D electronic structures, which may hamper the appearance of novel physical properties derived from 1D chains. However, we wish to address the fact that hypervalent bismuthides with 1D Bi−Bi linear chains are exclusively confined to RE3MBi5 systems obtained by Pan et al.,14,15 Zelinska et al.,16 and the present study. Thus, the situation is quite different from hypervalent antimonides with 1D Sb−Sb linear chains, where a number of examples have been reported such as Ti1−xZrxSb, Zr11−xVxSb8, Zr13−xVxSb10, Ti11Sb8, and Li2Sb27−32 in addition to RE3MSb5 and U3MSb5. We believe that further expansion of hypervalent bismuthides with 1D Bi−Bi linear chains is possible, which may eventually lead to the discovery of some exotic properties.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.7b00031. Supporting graphics (PDF) Crystallographic data (CIF)

Figure 4. (a) Calculated band structure for La3TiBi5 where Γ = (0,0,0), K = (1/3,1/3,0), M = (1/2,0,0), A = (0,0,1/2), L = (1/2,0,1/ 2), and H = (1/3,1/3,1/2) in the Brillouin zone. (b) Total and projected densities of states for La3TiBi5. Fermi level EF = 0 eV.

■

These results led us to prepare a solid solution of La3Ti(Bi1−xSbx)5 to examine how the superconductivity emerges as a function of x. XRD patterns of additionally prepared samples of La3Ti(Bi1−xSbx)5 (x = 0.2, 0.4, 0.6, 0.8, 1) are shown in Figure S1 in the Supporting Information. Although some of the specimens contain impurity phases such as LaSb, both a and c axes increase almost linearly with increasing x (Figure 2d), and the lattice parameters of La3TiSb5 agree well with those previously reported.12 This assures the successful preparation of the whole solid solution. Unexpectedly, the susceptibility measurements at low temperatures under a magnetic field of 20 Oe (Figure S2 in the Supporting Information) revealed that superconductivity is absent in the entire x range, including x = 1 (La3TiSb5). For La3TiSb5, the absence of superconductivity down to 2 K was checked also from the resistivity measurements, as shown in Figure 3b. Now let us discuss the origin of the contradictory observations regarding La3TiSb5. The former report12 shows that the superconducting volume fraction for La3TiSb5 is as low as 8 × 10−4 emu/fu at 2 K, indicating that superconductivity in La3TiSb5 is not a bulk nature. Moreover, the appearance or disappearance of superconductivity by rare-earth substitution is not systematic: RE3TiSb5 (RE = La, Ce, Nd) shows superconductivity while RE = Pr, Sm does not. In addition, Tc hardly changes on lanthanide substitution (Tc = 4.1 K for La,

AUTHOR INFORMATION

Corresponding Author

*E-mail for H.K.: [email protected]. ORCID

Hiroshi Kageyama: 0000-0002-3911-9864 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by a CREST project (JPMJCR1421) and Grants-in-Aid for Scientific Research on Innovative Areas “Mixed anion” (JP16H06439, JP16H06440) from the MEXT. We thank W. Yoshimune for his help in electrical resistivity measurements. SXRD experiments were performed at SPring-8 with the approval of the Japan Synchrotron Radiation Research Institute (JASRI) (Proposal No. 2014B1104).

■

REFERENCES

(1) Brechtel, E.; Cordier, G.; Schäfer, H. Z. Ba5Sb4 − the first alkaline earth pnictide with the Gd5Si4 structure type. Z. Naturforsch. 1981, 36, 1341. (2) Deller, K.; Eisenmann, B. Preparation and crystalline structure of CaSb2. Z. Anorg. Allg. Chem. 1976, 425, 104. (3) Papoian, G. A.; Hoffmann, R. Hypervalent bonding in one, two, and three dimensions: extending the Zintl-Klemm concept to 5044

DOI: 10.1021/acs.inorgchem.7b00031 Inorg. Chem. 2017, 56, 5041−5045

Article

Inorganic Chemistry nonclassical electron-rich networks. Angew. Chem., Int. Ed. 2000, 39, 2408. (4) Song, C.; Park, J.; Koo, J.; Lee, K. B.; Rhee, J. Y.; Budg, S. L.; Canfield, P. C.; Harmon, B. N.; Goldman, A. I. Charge-density-wave orderings in LaAgSb2: An x-ray scattering study. Phys. Rev. B: Condens. Matter Mater. Phys. 2003, 68, 035113. (5) Han, F.; Malliakas, C. D.; Stoumpos, C. C.; Sturza, M.; Claus, H.; Chung, D. Y.; Kanatzidis, M. G. Superconductivity and strong intrinsic defects in LaPd1−xBi2. Phys. Rev. B: Condens. Matter Mater. Phys. 2013, 88, 144511. (6) Masuda, H.; Sakai, H.; Tokunaga, M.; Yamasaki, Y.; Miyake, A.; Shiogai, J.; Nakamura, S.; Awaji, S.; Tsukazaki, A.; Nakao, H.; Murakami, Y.; Arima, T.; Tokura, Y.; Ishiwata, S. Quantum Hall effect in a bulk antiferromagnet EuMnBi2 with magnetically confined twodimensional Dirac fermions. Science Advances 2016, 2, e1501117. (7) Bolloré, G.; Ferguson, M. J.; Hushagen, R. W.; Mar, A. New Ternary Rare-Earth Transition-Metal Antimonides RE3MSb5 (RE = La, Ce, Pr, Nd, Sm; M = Ti, Zr, Hf, Nb). Chem. Mater. 1995, 7, 2229. (8) Ferguson, M. J.; Hushagen, R. W.; Mar, A. Crystal structures of La3ZrSb5, La3HfSb5, and LaCrSb3. Structural relationships in ternary rare-earth antimonides. J. Alloys Compd. 1997, 249, 191. (9) Mar, A.; Tougait, O.; Potel, M.; Noël, H.; Lopes, E. B. Anisotropic transport and magnetic properties of ternary uranium antimonides U3ScSb5 and U3TiSb5. Chem. Mater. 2006, 18, 4533. (10) Tkachuk, A. V.; Muirhead, C. P. T.; Mar, A. Structure and physical properties of ternary uranium transition-metal antimonides U3MSb5 (M = Zr, Hf, Nb). J. Alloys Compd. 2006, 418, 39. (11) Brylak, M.; Jeitschko, W. U3TiSb5, U3VSb5, U3CrSb5, and U3MnSb5 with ″Anti″-Hf5Sn3Cu type structure. Z. Naturforsch. B: Chem. Sci. 1994, 49, 747. (12) Moore, S. H. D.; Deakin, L.; Ferguson, M. J.; Mar, A. Physical properties and bonding in RE3TiSb5 (RE = La, Ce, Pr, Nd, Sm). Chem. Mater. 2002, 14, 4867. (13) Pecharsky, A. O.; Gschneidner, K. A., Jr. Crystal structure and magnetic properties of the new ternary compound Ce3MnBi5. J. Alloys Compd. 1999, 287, 67. (14) Pan, D. C.; Sun, Z. M.; Mao, J. G. Synthesis and crystal structures of La3MgBi5 and LaLiBi2. J. Solid State Chem. 2006, 179, 1016. (15) Pan, D. C.; Sun, Z. M.; Lei, X. W.; Mao, J. G. Synthesis and Crystal Structure of La3ScBi5. Chin. J. Inorg. Chem. 2006, 22, 1449. (16) Zelinska, O. Y.; Mar, A. Ternary Rare-Earth Manganese Bismuthides: Structures and Physical Properties of RE3MnBi5 (RE = La−Nd) and Sm2Mn3Bi6. Inorg. Chem. 2008, 47, 297. (17) Petríček, V.; Dušek, M.; Palatinus, L. Crystallographic computing system JANA2006: general features. Z. Kristallogr. - Cryst. Mater. 2014, 229, 345. (18) Blaha, K. S. P.; Madsen, G. K. H.; Kvasnicka, D.; Luitz, J. WIEN2K: An Augmented Plane Wave+ Local Orbitals Program for Calculating Crystal Properties; Technische Universitat: Vienna, 2001. (19) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 1996, 77, 3865. (20) Nomura, K.; Hayakawa, H.; Ono, S. The lanthanum-bismuth alloy system. J. Less-Common Met. 1977, 52, 259. (21) Auer-Welsbach, H.; Nowotny, H. N.; Kohl, A. Untersuchung reibungspyrophorer Ti-Legierungen; Ti2Bi, ein neuer Strukturtyp. Monats. Chem. 1958, 89, 154. (22) Eberle, D.; Schubert, K. Strukturuntersuchungen im System Zirkon-Wismut und einigen quasihomologen Legierungen. Zeitschrift für Metallkunde 1968, 59, 306. (23) Hulliger, F. TiAs2-type Phases. Nature 1964, 204, 991. (24) Sun, Z. M.; Mao, J. G. Synthesis and crystal structure of EuBi2. J. Solid State Chem. 2004, 177, 3752. (25) Garcia, E.; Corbett, J. D. Chemistry in the Polar Intermetallic Host Zr5Sb3. Fifteen Interstitial Compounds. Inorg. Chem. 1990, 29, 3274. (26) Shaw, R. W.; Mapother, D. E.; Hopkins, D. C. Critical Fields of Superconducting Tin, Indium, and Tantalum. Phys. Rev. 1960, 120, 88.

(27) Kleinke, H. Zr1−xTixSb: a novel antimonide on the quasibinary section ZrSb−TiSb with a complex crystal structure exhibiting linear Sb chains and fragments of the TiSb structure. J. Am. Chem. Soc. 2000, 122, 853. (28) Elder, I.; Lee, C.; Kleinke, H. Zr11Sb18: a new binary antimonide exhibiting an unusual Sb stom network with nonclassical Sb−Sb nonding. Inorg. Chem. 2002, 41, 538. (29) Kleinke, H. Metal-rich polyantimonides: internal competition between M−M and Sb−Sb and heteroatomic M−Sb interactions in (Zr,V)13Sb10 and (Zr,V)11Sb8 (M = Zr,V). J. Mater. Chem. 1999, 9, 2703. (30) Bobev, S.; Kleinke, H. Instabilities in the Linear Sb Atom Chain of the new binary antimonide Ti11−xSb8−y. Chem. Mater. 2003, 15, 3523. (31) Müller, W. Preparation and crystal structure of Li2Sb. Z. Naturforsch. 1977, 32, 357. (32) Kleinke, H. From molecular Sb units to infinite chains, layers, and networks: Sb−Sb interactions in metal-rich antimonides. Chem. Soc. Rev. 2000, 29, 411.

5045

DOI: 10.1021/acs.inorgchem.7b00031 Inorg. Chem. 2017, 56, 5041−5045