Magnetically Frustrated Quaternary Chalcogenides with

Jun 28, 2017 - Synopsis. Interpenetrating diamond lattices constructed by cross-linking of the adamantane supertetrahedra of transition-metal-containi...
2 downloads 13 Views 4MB Size
Download PDF
Article pubs.acs.org/IC

Magnetically Frustrated Quaternary Chalcogenides with Interpenetrating Diamond Lattices Sudip Mohapatra,† Amit Adhikary,† Kartik Ghosh,‡ and Amitava Choudhury*,† †

Department of Chemistry, Missouri University of Science and Technology, Rolla, Missouri 65409, United States Department of Physics, Astronomy and Materials Science and Center for Applied Science and Engineering, Missouri State University, Springfield, Missouri 65897, United States

‡

S Supporting Information *

ABSTRACT: A series of quaternary sulfides of the composition Na3MGaS4 (M = Mn (1), Fe (2), and Co (3)) have been synthesized in sealed quartz ampules. In these compounds, divalent transition metal and Ga occupy the same crystallographic site in the Ga−S network, forming a supertetrahedral, T2 (adamantane) unit, through the cornersharing of four M/GaS4 tetrahedra. The corner sulfur atoms of the T2 clusters are further connected to similar T2 units to form an open continuous three-dimensional (3D) anionic framework of composition {[Ga2M2S8]n}6−. The framework resembles a zinc blende structure type if each T2 cluster is considered as a single tetrahedron and two such frameworks are intertwined to generate channels wherein reside the extra-framework Na+ ions. Placement of transition metals (Mn or Fe or Co) in the corner of a perfect supertetrahedron, adamantane building unit, generates an ideal lattice for geometrical magnetic frustration, which, on dilution with nonmagnetic metal (Ga), creates an ideal case for random frustration. Preliminary magnetic measurements indicate high negative values of the Weiss constant (−200 to −400 K) and the absence of any magnetic ordering, reinforcing the presence of magnetic frustration in all of these compounds.

■

INTRODUCTION Multinary chalcogenides are an attractive class of compounds that not only display structural diversity but also exhibit a variety of properties. The structural variety is often a consequence of the ability of main group elements to form a large number of building blocks through M−Q (M = Ga, In, Ge, Sn, Sb, etc; Q = S, Se, Te) bonds.1 Due to the covalency of the metal−chalcogen bonds, often these materials are semiconducting.2 These building blocks when combined with transition metals (unpaired d or f electrons) form fascinating structures with properties that combine both the spin of the unpaired electron and the semiconducting nature of the framework. Recently discovered materials, SbPS4, Na2EuSiSe4, and Na1.515EuGeS4, represent unique structures with a threedimensional (3-D) assembly of nanotubules;3−5 on the other hand, FexPb4−xSb4Se10 and FeSb2Se4 represent examples of magnetic semiconductors.6,7 The propensity to form noncentrosymmetric crystal structures in the chalcogenide family has also resulted in many complex chalcogenides showing strong second harmonic generation responses.8−11 All of these attributes have made multinary chalcogenides a rich playground for solid state chemists, who constantly employ chemical intuition-driven exploratory synthesis routes to create new compositions and structure types for potential applications in diverse directions.12 The elements of group 13, especially Ga and In, have a strong tendency to form a wide range of anionic polychalcogenide networks. For example, an interesting structural aspect of the gallium and indium chalcogenide is the formation of different supertetrahedral secondary building © 2017 American Chemical Society

units (clusters with tetrahedral shaped fragments) under different experimental conditions.13−16 In such compounds, supertetrahedra are further cross-linked via T−Q−T (T = supertetrahedra, Q = chalcogen) linkages to form flexible frameworks with interesting structural features. Most of these compounds with open-framework structures are synthesized under solvo/hydrothermal conditions at relatively low temperatures.13−18 However, synthesis of multinary gallium-chalcogenides incorporating transition metals in the presence of alkali chalcogenide flux under solid state conditions are less explored. We, therefore, wanted to investigate how transition metals influence the Ga−chalcogenide networks structurally at relatively high temperatures. Toward this goal, we have synthesized a new series of transition-metal-containing quaternary Ga−S compounds, Na3MGaS4 (M = Mn (1); Fe (2); Co (3)) employing solid state synthetic routes. The structure consists of two interpenetrating anioinic frameworks built up of supertetrahedral, T2 (adamantane),19 building units with charge-balancing Na ions occupying the channels. Introduction of transition metals into the Ga site of an adamantane building unit makes it an exceptional compound where mixed transition metals/Ga form a near perfect tetrahedral lattice ideal for magnetically diluted geometrical magnetic frustration. Furthermore, interpenetration is a structural feature which is not very rare in chalcogenide frameworks;13,16−20 however, interpenetration of two magnetiReceived: January 15, 2017 Published: June 28, 2017 7650

DOI: 10.1021/acs.inorgchem.7b00121 Inorg. Chem. 2017, 56, 7650−7656

Article

Inorganic Chemistry Table 1. Crystal Data and Refinement Parameters for 1−3 empirical formula formula weight T/K wavelength/Å crystal system space group a/Å b/Å c/Å volume/Å3 Z ρcalc/Mg m−3 μ/mm−1 goodness-of-fit (S) R [I > 2σ(I)]a wR (F2) (all data)b δF/e Å−3 a

1

2

3

Na3MnGaS4 321.87 298 0.71073 tetragonal I41/acd 12.970(4) 12.970(4) 18.796(6) 3162(2) 16 2.705 6.121 1.174 R1 = 0.0337 wR2 = 0.0871 0.99 and −0.99

Na3FeGaS4 322.78 298 0.71073 tetragonal I41/acd 12.832(4) 12.832(4) 18.632(6) 3068(2) 16 2.795 6.553 1.134 R1 = 0.0423 wR2 = 0.1247 1.29 and −1.10

Na3CoGaS4 325.86 298 0.71073 tetragonal I41/acd 12.891(4) 12.891(4) 18.476(5) 3070(2) 16 2.820 6.819 1.070 R1 = 0.0322 wR2 = 0.0885 0.54 and −0.65

R1= ∑||Fo| − |Fc|/∑|Fo|. bwR2 = {∑[w(Fo2 − Fc2)2]/∑[w(F02)2]}1/2, w = 1/[σ2(Fo)2 + (aP)2 + bP], where P = [Fo2 + 2Fc2]/3. electron densities (q-peaks) remaining in the difference Fourier maps. The weighted R-factor, wR2, did not converge to a low value as expected for such heavy elements with reasonably good thermal parameters, and also a charged-balanced formula (Na3Ga2S4) could not be attained assuming common oxidation states of the respective elements. Besides the high value of the weighted R-factor, an unusual high value for the second parameter (b-parameter) in the suggested weighting scheme clearly indicated uncorrected systematic errors due to improper absorption correction or unresolved disorder. This led us to believe that there may be some unaccounted for partial occupancy in Na2 as revealed by its high thermal parameter or constitutional disorder in the Ga site due to the mixed occupancy of M (Mn, Fe, and Co) and Ga, since transition metals were used in the syntheses. The occupancy refinement of Na2 site did not show the presence of any vacancy. Therefore, in the next cycle of refinement, the Ga site was refined as a mixed occupancy site by placing both Ga and M, keeping their coordinates the same while freely refining their occupancies and constraining the sum of the occupancy of Ga and M to be 100%. Such refinement yielded a satisfactory weighted R-factor, a dramatic improvement in weighting parameters, and yielded an occupancy of 50% (0.49(6)) for both Ga and M, which also led to a charge-balanced formula, Na3MGaS4, assuming M in a +2 oxidation state. Magnetic measurements confirmed the presence of M in a +2 oxidation state. We did not observe any superstructure reflection using reciprocal lattice viewer, RLATT,22 indicating random disorder of Ga and M in the same crystallographic site. The last cycles of refinements included an anisotropic thermal parameter refinement for all the atoms and implementation of the suggested weighting scheme. Crystal data and final structure refinement parameters are given in Table 1, and selected bond lengths for each compound are provided in Table 2. The atomic coordinates along with their thermal displacement parameters are supplied as Supporting Information in the form of Tables S1 and S2. Efforts to synthesize the pure monometallic form of Na3MGaS4 either with transition metals (Na3M2S4; M = Mn, Fe, and Co) or Ga

cally frustrated lattices is unprecedented. In this article, we report the synthesis, structure, Raman spectroscopy, thermal stability, and preliminary magnetic properties of 1−3 displaying an excellent model system to study magnetic frustration.

■

EXPERIMENTAL SECTION

Synthesis. For the synthesis of compounds 1−3, 1 mmol of M (M = Fe, Mn, Co), 1 mmol of Ga, 2.5 mmol of S, and 1.5 mmol of Na2S were loaded into quartz ampules inside a nitrogen-filled glovebox (O2 < 1 ppm), which were flame-sealed under vacuum and placed in a furnace. The temperature of the furnace was increased from 25 to 700 °C at a rate of 10 °C/h, kept at 750 °C for 120 h, and then cooled to room temperature at 10 °C/h. The product in each case was washed with N,N-dimethylformamide to remove any excess flux and byproducts. The dark red, dark green, and brown colored slightly moisture-sensitive crystals of Mn, Co, and Fe containing products, respectively, were obtained and subsequently used for single-crystal Xray diffraction study and other characterizations. Synthesis targeting the same compositions as in 1−3 with pure transition metals (Na3M2S4; M = Mn, Fe, and Co) or pure Ga has been attempted following an identical protocol and characterized employing powder X-ray diffraction. X-ray Crystallography. Suitable single-crystal for each phase was dipped into paratone oil and mounted on a thin glass fiber using commercially available super glue. X-ray single crystal diffraction data were collected on a Bruker Smart−Apex diffractometer equipped with a sealed tube X-ray source with graphite monochromated Mo−Kα radiation (λ = 0.71073 Å) operating at 45 kV and 35 mA. The SMART software21 was used to collect data at room temperature (298 K) employing a scan of 0.3° in ω with an exposure time of 20 s/frame. The program SAINT22 was used for integration of diffraction profiles, and absorption correction was made employing the program, SADABS.21 The structures were solved by direct methods using SHELXS-9722 and difference Fourier syntheses. Full-matrix leastsquares refinement against |F2| was carried out using the SHELXTLPLUS23 suite of programs. Finally, the crystal structures for 1−3 were refined using SHELX-2014 incorporated in WinGX software.24 The space group for the compounds was unambiguously assigned to I41/ acd (No. 142) based on the systematic absences. The positions of Ga and three S were easily assigned from the difference Fourier maps. The isotropic refinement indicated good thermal parameters for Ga and S. After the first cycle of refinement, two more positions were assigned to Na atoms (Na2 and Na1). Again, isotropic thermal parameters for all the atoms except Na2 were very good. Na2 had slightly higher thermal parameters compared to Na1, and there were no significant residual

Table 2. Selected Bond Lengths [Å] for Na3MGaS4 [M = Mn (1), Fe (2), and Co (3)]

a

7651

Na3MGaS4

M = Mn

M = Fe

M = Co

Ga1/M1−S1 Ga1/M1−S1a Ga1/M1−S2 Ga1/M1−S3

2.3051(14) 2.3208(14) 2.3367(10) 2.3721(10)

2.2736(17) 2.2876(17) 2.3061(11) 2.3347(13)

2.2716(14) 2.2914(14) 2.3045(11) 2.3378(12)

−y + 3/4, x − 1/4, −z + 1/4. DOI: 10.1021/acs.inorgchem.7b00121 Inorg. Chem. 2017, 56, 7650−7656

Article

Inorganic Chemistry

Ga/M−S bonds in the corner-shared (Ga/M)S4 tetrahedral unit.25 The peaks at lower frequency range are due to Na−S bonds.25 Differential Scanning Calorimetry (DSC). Thermal stability of the compounds 1−3 was characterized by DSC study. Compounds Na3FeGaS4 and Na3CoGaS4 have a similar melting point at ∼727 °C, whereas Na3MnGaS4 melts at a slightly higher temperature, 801 °C (Figure S3). Additional small peaks are observed for 2 (Fe) and 3 (Co), which could be due the presence of a minor impurity in the case of 2 and due to some phase transition in the case of 3. Magnetic Measurements. The molar magnetic susceptibility measurements of 1−3 were carried out with a Quantum Design MPMS SQUID magnetometer while warming in a 2 T applied field from 2 to 300 K after zero-field and field-cooled (ZFC and FC) conditions. The isothermal magnetization was subsequently measured at 5 K in an applied field of −5 to +5 T.

(Na3Ga2S4) under identical synthetic conditions fail to produce similar structure types, as revealed from their powder X-ray diffraction patterns (Supporting Information, Figure S1), indicating that mixed occupancy of transition metal and Ga may be required to stabilize these phases (1−3). The powder X-ray diffraction patterns of 1−3 collected on coarse ground samples match very well with simulated diffraction patterns generated from the single-crystal atomic coordinates, indicating phase purity of the synthesized materials (Figure 1). However, there is evidence of some impurity phase in the case of compound 2, Na3FeGaS4.

■

RESULT AND DISCUSSION Structural Description. The asymmetric unit of the compounds 1−3 consists of one 1:1 mixed occupied M/Ga (M = Mn, Fe, Co), two Na, and three sulfur sites with a composition Na1.5Ga0.5M0.5S2 (Figure 2a). Each of the unit cells comprises 16 asymmetric units if the formula is written as Na3MGaS4. Na1, S2, and S3 atoms are located in the special position with twofold symmetry, all remaining atoms are in the general positions. The Ga/M site adopts distorted tetrahedral coordination geometry. Bond distances of compounds 1−3 are given in Table 2. All Ga/M−Q distances are in-between the Ga−S and M−S distances, respectively. For example, Ga/Mn−S bond distances are in the range of 2.3051(14)−2.3721(10) Å (Table 2), which falls in the intermediate range of pure Mn−S and Ga−S distances. Pure Mn−S bond distances in Na6MnS4 are in the range 2.434(2)−2.440(3) Å, whereas the pure Ga−S bond distance in Na6Ga2S6 is 2.237(3) Å.26,27 Ga/M−S bond distances decrease from M = Mn to M = Co, consistent with decreasing ionic radii of +2 transition-metal ions in high-spin tetrahedral coordination (Table 2). S−M/Ga−S bond angles for compounds 1−3 are in the range 99.88(2)−119.53(2)°, reflecting the degree of distortion from ideal tetrahedral coordination geometry. Ga/MS 4

Figure 1. Comparison of simulated and experimental powder X-ray diffraction patterns of compounds 1−3. The arrows indicate the presence of impurity phase in 2, Na3FeGaS4. Raman Spectral Studies. To verify the local coordination, Raman spectra were collected for compounds 1−3 (Figure S2, Supporting Information). Coarse ground crystalline samples were loaded on a glass slide for Raman spectroscopic analysis. The spectra were collected on a Lab RAM Aramis Horiba Jobin Yvom spectrometer using a HeNe laser source with a 0.03 filter with a 20 s exposure time. For compounds 1−3, sharp peaks have been found in the range of 300−400 cm−1 which can be assigned as different stretching modes of

Figure 2. (a) Asymmetric unit of 1−3, shown for compound 1. Thermal ellipsoids are given at 40% probability. Only atoms appearing in the asymmetric unit are labeled. (b) T2 supertetrahedral secondary building unit, shown for compound 1 (Ga2Mn2S10) of [GaMnS4]3−. (c) T2 supertetrahedral building units (Ga2Mn2S10) are connected to each other through vertex sharing and form a 3D lattice, {Ga2Mn2S8}∞. The cyan balls (dummy atoms) are the centers of gravity of each T2 supertetrahedra. 7652

DOI: 10.1021/acs.inorgchem.7b00121 Inorg. Chem. 2017, 56, 7650−7656

Article

Inorganic Chemistry

Figure 3. (a) Interpenetrated 3D structure of Na3GaMnS4 with Na+ in cavities (Na−S bonds are deleted for clarity). (b) A simplified view of the 3D interpenetrated structure along a-direction (blue and red balls represent the center of supertetrahedra in two different lattices).

type) with T2 clusters has been reported in UCR-21 and ICF21 framework topologies.15,16 Another important aspect that distinguishes the current compounds with other reported interpenetrated chalcogenide frameworks originates in their synthetic routes. A large number of interpenetrated chalcogenide frameworks were synthesized by hydrothermal or solvothermal methods where alkali metal ions or organoammonium ions acted as charge-balancing cations or templates.13−20 When alkali metals were used instead of organoammonium ions as charge-balancing cations in the hydrothermal synthesis, water molecules invariably incorporated into the structure residing in the channels along with the alkali ions.16 It was also noticed by Feng’s group that, in solution-based synthesis, the presence of M2+ ions (M2+ = Mn2+, Fe2+, Co2+, Zn2+) in M3+−S systems (M3+ = Ga3+ and In3+) generally led to an increase in cluster size from T3 to T4 or T5, whereas incorporation of M4+ cations in the M3+−S system (M3+ = Ga3+ and In3+) led to a decrease in the cluster size from T3 to T2.20 However, under solid state conditions and at high temperatures, we are able to stabilize the T2 cluster incorporating M2+ in the M3+−S system (M3+ = Ga3+) though with a mixed occupancy or doping in the metal site. Also note that T2 clusters found in the UCR-21 and ICF-21 are noninterpenetrating lattices presumably due to the presence of organic amine and water molecules, respectively, taking care of the voids through weak interactions.15,16 In the absence of solvent or organoammonium cations, T2 clusters form interpenetrating diamond lattices to fill up some of the void space. Structures with interpenetrated lattices made by a purely solid state high temperature route are rare.31 It can be anticipated that the anionic framework of chalcogenides with extra-framework cations synthesized under superdry solid state condition can exhibit fast ionic conductivity because of the facile migration of alkali ions through the channel constructed by softer chalcogenide anions. Moreover, dry method of preparation of this framework is expected to be advantageous over other wet methods because the presence of water or organoammonium cations in the structure in hydro/solvother-

tetrahedra are corner-shared to form a T2 supertetrahedral cluster, Ga2M2S10 (Figure 2b). The T2 clusters are further connected through their vertices (S2) (Figure 2c) to form an open continuous three-dimensional (3D) framework of composition [{GaMS4}n]3−. Furthermore, two identical 3D [{GaMS4}n]3− anionic lattices are assembled together through interpenetration and generate two types of cavities wherein reside the extra-framework Na+ ions (Figure 3a). Two sodium ions (Na1 and Na2) are present in distorted octahedral coordination geometry. Na−S bond distances are in the range 2.639(3)−3.273(5) Å for all the compounds. The longer axial Na2−S distances give rise to a highly tetragonally distorted octahedral coordination geometry for Na2. The Na−S bond lengths are comparable to those reported in the literature.26,27 Each noninterpenetrating lattice when analyzed separately revealed that they form six-membered ring channels along the [1 1 1] direction. The number of bonds in the ring is counted by joining the nodes (center of gravity) of each T2 cluster. Such connectivity between the nodes also revealed that each of these lattices belongs to a cubic ZnS structure type or more appropriately an interpenetrating diamond lattice since all clusters are equivalent (Figure 3b). It is to be noted here that, despite interpenetration of the two diamond lattices, there is a considerable amount of void space to host the alkali ions. The interpenetrated diamond lattices are not uncommon in chalcogenide chemistry mainly with Ga and In; however, the differences exist in the cluster type, which can vary from T2 to T5. For example, T3 clusters in DMA-InSSB1,13 UCR-7,28 and [C4NH12]6[Ga10S18],29 T4 clusters in Cd4In16S33,19 UCR-5,28 ICF-5,16 and OCF-5,18 and T5 clusters in UCR-1730 and ICF-1716 form interpenetrating diamond lattices. Interpenetrating diamond lattices can also be formed by two individual lattices made of a mixture of cluster types. For example, in [C4NH12]16[Ga10S18M4Ga16S33] (M = Co, Zn)29 and UCR-19ZnGaS-TETA,28 interpenetrating diamond lattices are formed by individual lattices where T3 and T4 clusters alternate. To our knowledge, interpenetrating diamond lattices with T2 clusters were not reported previously; however, a noninterpenetrating diamond lattice (or cubic ZnS structure 7653

DOI: 10.1021/acs.inorgchem.7b00121 Inorg. Chem. 2017, 56, 7650−7656

Article

Inorganic Chemistry

from the presence of a minor impurity. For compound 3, experimental μeff of 4.21 μB is higher than the spin-only value of 3.89 μB due to effective spin−orbit coupling, leading to the mixing of the ground state term 4A2 with the excited state term 4 T2(F).32 The high negative values of the Weiss constant for the compounds (see Table 3), 1−3 indicate the presence of predominant antiferromagnetic interactions between neighboring metal centers and the occurrence of magnetic frustration. Further, the isothermal magnetization (M) values measured at 5 K for 1−3 exhibit a very sluggish increase of magnetization with increasing applied field (Figure 5). Moreover, magnet-

mally prepared materials may not be good for electrochemical applications.16 Magnetic Properties. The temperature dependent zerofield-cooled dc magnetic measurements of powdered polycrystalline samples of 1−3 were performed in SQUID MPMS. The plots of temperature dependent molar and inverse molar magnetic susceptibility χ M and χ M −1 , respectively, for compound 1−3 are given in Figure 4. The susceptibility plots

Figure 4. Plot of temperature dependent molar magnetic susceptibility (χM) and inverse molar magnetic susceptibility (χM−1) at an applied magnetic field of 2 T for 1−3.

display that there is a continuous increase of the χM value asymptotically with decreasing temperature and reach a maximum value of 0.102, 0.046, and 0.0318 emu mol−1, respectively, for 1−3 at the lowest measured temperature. There is no divergence of the FC and ZFC data of susceptibility measurements at 2 T applied field. In the case of Na3CoGaS4 (3), at 133 K, there is a jump in χM upon cooling from room temperature to low temperatures due to some sort of transition. The exact nature of this transition is currently not clear to us. Due to the absence of any net magnetization, it cannot be assigned as a ferrimagnetic transition. The plots of χM−1 vs T are linear at higher temperatures of 150−300 K (Figure 4), and below 150 K, the χM−1 vs T plots tend to deviate from linearity, which is especially pronounced for 2 and of course in 3 due to the transition. The linear fitting of the data in the temperature range of 150−300 K for 1 and 200−300 K for 2 and 3 using Curie−Weiss law χM−1 = C/(T − θP) yields the Weiss constant, θP, and Curie constant, C, as listed in Table 3. The effective paramagnetic moments (μeff) calculated from the Curie constants are consistent with the theoretical spin-only magnetic moment (Table 3) except for Fe(II) in 2, which shows a rather lower value. For compound 1, the experimental μeff value of 5.86 μB is close to the theoretical spin-only value of 5.9−6.0 μB for five unpaired electrons of Mn2+ (μeff (theo) = 5.91 μB), and for 2, the experimental μeff value of 3.89 μB is significantly lower than theoretical μeff = 4.89 μB for Fe2+. The significantly lower experimental μeff value for compound 2 most probably arises

Figure 5. Field dependence of magnetization for compounds 1−3 at 5 K.

ization (M) at 5 T is far below the theoretical saturation moment of spin-only free ions (Ms = gSNβ). The values of the maximum experimental M and theoretical saturation moment at 5 T are given in Table 3. The maximum experimental M values of 1−3 are only ∼10% of the theoretical saturation moment for fully ordered spins of free ions. These observations along with the absence of hysteresis on reversing the field corroborate well with the presence of strong antiferromagnetic interactions and magnetic frustration, as indicated by high frustration parameters, f = |Weiss constant|/T* > 10, where T* denotes a transition temperature (TC, TN, or Tf).33 In the absence of any ordering temperature, we have used the lowest temperature of measurements to characterize frustration index (see Table 3). It is to be noted here that, for compound 3, due to the presence of an unknown transition at 133 K, we did not report a frustration parameter. Further inspection of the magnetic behaviors, such as high frustration parameters and low magnetization, of 1−3 is worthwhile, considering the structural feature of the compound. A tetrahedral arrangement of the four antiferromagnetically coupled spins is inherently geometrically frustrated (Figure 6a), but, however, in this case, magnetic ions in the corners of the tetrahedral unit are diluted 50% by nonmagnetic Ga3+ ions. The

Table 3. Curie−Weiss Fitting Parameters, Theoretical and Experimental μeff, Frustration Parameter, Theoretical Saturated Magnetization, and Maximum Experimental Magnetization Achieved at 5 T for Compounds 1−3 Na3MGaS4

C (emu K/mol)

θP (K)

μeff (expt.)

μeff (theo. spin-only)

f = |θP|/2

M(Nβ)a at H = 5 T

Ms(Nβ)max theoretical

M = Mn (1) M = Fe (2) M = Co (3)

4.29 1.73 2.21

−228.52 −301.71 −393.60

5.86 μB 3.89 μB 4.21 μB

5.92 μB 4.89 μB 3.89 μB

114.2 150.8 NAb

0.67 0.28 0.16

5 4 3

1 Nβ = 5585 cm3 G/mol. bSince there is an upturn at 133 K in the χM(T) plot of 3, which potentially could be a magnetic transition, and therefore, f = |θP|/2 was not used to define the frustration index. a

7654

DOI: 10.1021/acs.inorgchem.7b00121 Inorg. Chem. 2017, 56, 7650−7656

Article

Inorganic Chemistry

Figure 6. (a) Figure depicts the magnetically frustrated tetrahedral lattice formed by the four transition-metal centers. A dummy atom has been placed in the center for the easy visualization of the tetrahedron. (b) Figure shows the interconnectivity of the magnetically diluted tetrahedral units. (c) Figure depicts the A-site magnetic sublattice of the pyrochlore (A2B2O7) structure specifically in the case of Ho2Ti2O7.

■

CONCLUSIONS We have synthesized and characterized a novel class of chalcogenide frameworks where divalent transition metals and trivalent group 13 elements occupy the same crystallographic site to produce interpenetrating anionic frameworks with extraframework alkali ions filling the channels. These compositions and structure type represent a new class of compounds where their unique structure and the presence of magnetic ions in the geometrically frustrated building units makes them a rare class of magnetically frustrated solids. This work also emphasizes the fact that solid state high temperature synthetic conditions can result in structures involving divalent transition metals and group 13 elements, which are not accessible through the solution route synthesis of chalcogenides.

tetrahedra are slightly distorted, as evident by their bond lengths (Figure 6a) and angles. This tetrahedral lattice formed by diluted magnetic ions is similar to tetrahedral sublattices formed by the A- and B-site ions as in pyrochlores (A2B2O7) or B-site ions in spinels (AB2X4).34 However, there is a difference in how these tetrahedral units are connected to each other. Both A- and B-site sublattices in pyrochlore and the B-site sublattice in spinels have the same kind of arrangement of corner-shared tetrahedra, while, in the present case, the tetrahedral units are connected through a bridging ligand effectively separating the tetrahedral lattices (compare (b) and (c) in Figure 6). This interconnectivity of the frustrated tetrahedral lattices and the interpenetration of two such lattices are unique in these compounds. Frustrated tetrahedral lattices separated by diamagnetic spacer ligands have been observed previously in an iron-selenite, [C4N2H12]0.5[Fe2F3(SeO3)2].35 One can also interpret the A-site and B-site sublattices in pyrochlore as interpenetrated lattices, but such interpenetration in turn leads to the edge-sharing of the A-site and B-site tetrahedral units; on the other hand, in the present compounds, the interpenetrated lattices do not have any covalent connectivity between themselves. Magnetic dilution near the percolation threshold is often a deliberate approach to introduce disorder in the frustrated network to study interesting physics at low temperatures. The present compounds could be considered as classic examples of transforming an otherwise disorder free geometrically frustrated system to a random frustrated system by substituting with nonmagnetic ions, which we obtained by default.36−38 Such dilution allows one to study the nature of the magnetic ground state (antiferromagnetic, spin glass, etc.), and in this case, a high dilution should ideally lead to a spin-glass phase.36,37 The absence of any long-range ordered state or spin-glass phase up to 2 K in the present compounds may indicate that the spinglass transition is below 2 K or a high applied field in measuring the susceptibility may have wiped out the spin-glass transition. Further DC magnetic measurements with low applied fields along with AC susceptibility measurements are essential to understand the nature of magnetic frustration in these compounds. Synthesis of an ordered phase with pure magnetic ions and the calculation of percolation threshold will be another aspect of future research on these compounds.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.7b00121. Tables of atomic coordinates, isotropic and anisotropic thermal parameters, PXRD, Raman spectra, and DSC curves for compounds 1−3 (PDF) Accession Codes

CCDC 1548429−1548431 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing [email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Amitava Choudhury: 0000-0001-5496-7346 Notes

The authors declare no competing financial interest. 7655

DOI: 10.1021/acs.inorgchem.7b00121 Inorg. Chem. 2017, 56, 7650−7656

Article

Inorganic Chemistry

■

Supertetrahedral Zinc Gallium Selenides. J. Am. Chem. Soc. 2011, 133, 9616−9625. (19) Li, H.; Kim, J.; Groy, T. L.; O’Keeffe, M.; Yaghi, O. M. 20 Å Cd4In16S3514‑ Supertetrahedral T4 Clusters as Building Units in Decorated Cristobalite Frameworks. J. Am. Chem. Soc. 2001, 123, 4867−4868. (20) Feng, P.; Bu, X.; Zheng, N. The Interface Chemistry between Chalcogenide Clusters and Open Framework Chalcogenides. Acc. Chem. Res. 2005, 38, 293−303. (21) SMART; Bruker AXS Inc.: Madison, WI. 2002. (22) SAINT and SADABS; Bruker AXS Inc.: Madison, WI, 2008. (23) Sheldrick, G. M. A short history of SHELX. Acta Crystallogr., Sect. A: Found. Crystallogr. 2008, A64, 112−122. (24) Farrugia, L. J. WinGX - A Windows Program for Crystal Structure Analysis. J. Appl. Crystallogr. 2012, 45, 849. (25) Barnier, S.; Guittard, M.; Palazzi, M.; Massot, M.; Julien, C. Raman and infrared studies of the structure of gallium sulphide based glasses. Mater. Sci. Eng., B 1992, B14, 413−417. (26) Bronger, W.; Balk-Hardtdegen, H. Darstellung, Struktur und magnetisches Verhalten von Alkalimetallmanganchalkogeniden A6Mn X4 mit A = Na oder K und X = S, Se oder Te. Z. Anorg. Allg. Chem. 1989, 574, 89−98. (27) Eisenmann, B.; Hofmann, A. Crystal structure of hexasodium dimu-thiobis(dithiogallate) − I. Z. Kristallogr. 1991, 197, 143−144. (28) Zheng, N.; Bu, X.; Feng, P. Nonaqueous Synthesis and Selective Crystallization of Gallium Sulfide Clusters into Three-Dimensional Photoluminescent Superlattices. J. Am. Chem. Soc. 2003, 125, 1138− 1139. (29) Vaqueiro, P.; Romero, M. L. Three-dimensional gallium sulphide open frameworks. J. Phys. Chem. Solids 2007, 68, 1239−1243. (30) Bu, X.; Zheng, N.; Li, Y.; Feng, P. Pushing Up the Size Limit of Chalcogenide Supertetrahedral Clusters: Two- and Three-Dimensional Photoluminescent Open Frameworks from (Cu5In30S54)13‑ Clusters. J. Am. Chem. Soc. 2002, 124, 12646−12647. (31) Aitken, J. A.; Kanatzidis, M. G. Interwoven Pair of Open Frameworks in the Thiosphosphate K6Yb3(PS4)5. J. Am. Chem. Soc. 2004, 126, 11780−11781. (32) Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements, 2nd ed.; Elsevier: Amsterdam, 1997; p 1132. (33) Ramirez, A. P. Strongly geometrically frustrated magnets. Annu. Rev. Mater. Sci. 1994, 24, 453−480. (34) Greedan, J. E. Geometrically frustrated magnetic materials. J. Mater. Chem. 2001, 11, 37−53. (35) Choudhury, A.; Kumar, D. U.; Rao, C. N. R. Three-dimensional organically templated open-framework transition metal selenites. Angew. Chem., Int. Ed. 2002, 41, 158−161. (36) Villain, J. Insulating spin glasses. Z. Phys. B: Condens. Matter Quanta 1979, 33, 31−42. (37) Binder, K.; Young, A. P. Spin glasses: Experimental facts, theoretical concepts, and open questions. Rev. Mod. Phys. 1986, 58, 801−976. (38) Gardner, J. S.; Gingras, M. J. P.; Greedan, J. E. Magnetic pyrochlore oxides. Rev. Mod. Phys. 2010, 82, 53−107.

ACKNOWLEDGMENTS The authors acknowledge the funding from ERDC (Missouri S&T) and University of Missouri Research Board (UMRB). The authors also acknowledge Dr. Gerasimchuk, Missouri State University, Springfield, and Dr. Brow, Missouri S&T, for their help with the DSC and Raman spectroscopic measurements, respectively.

■

REFERENCES

(1) Sheldrick, W. S.; Wachhold, M. Chalcogenidometalates of the heavier Group 14 and 15 elements. Coord. Chem. Rev. 1998, 176, 211− 322. (2) Bensch, W.; Kanatzidis, M. Chalcogenides and chalcogenidometalates. Z. Anorg. Allg. Chem. 2012, 638, 2384−2385. (3) Malliakas, C. D.; Kanatzidis, M. G. Inorganic Single Wall Nanotubes of SbPS4−xSex(0≤x≤3) with Tunable Band Gap. J. Am. Chem. Soc. 2006, 128, 6538−6539. (4) Choudhury, A.; Dorhout, P. K. An Ordered Assembly of Filled Nanoscale Tubules of Europium Seleno-silicate in the Crystal Structure of a Quaternary Compound. J. Am. Chem. Soc. 2007, 129, 9270−9271. (5) Choudhury, A.; Grandjean, F.; Long, G. J.; Dorhout, P. K. Na1.515EuGeS4, A Three-Dimensional Crystalline Assembly of Empty Nanotubules Constructed with Europium(II/III) Mixed Valence Ions. Inorg. Chem. 2012, 51, 11779−11786. (6) Poudeu, P. F. P.; Takas, N.; Anglin, C.; Eastwood, J.; Rivera, A. FexPb4−xSb4Se10: A New Class of Ferromagnetic Semiconductors with Quasi 1D {Fe2Se10} Ladders. J. Am. Chem. Soc. 2010, 132, 5751−5760. (7) Djieutedjeu, H.; Poudeu, P. F. P.; Takas, N. J.; Makongo, J. P. A.; Rotaru, A.; Ranmohotti, K. G. S.; Anglin, C. J.; Spinu, L.; Wiley, J. B. Structural-Distortion-Driven Cooperative Magnetic and Semiconductor-to-Insulator Transitions in Ferromagnetic FeSb2Se4. Angew. Chem., Int. Ed. 2010, 49, 9977−9981. (8) Chung, I.; Kanatzidis, M. G. Metal chalcogenides: a rich source of nonlinear optical materials. Chem. Mater. 2014, 26, 849−869. (9) Zhao, H.-J.; Zhang, Y.-F.; Chen, L. Strong Kleinman-Forbidden Second Harmonic Generation in Chiral Sulfide: La4InSbS9. J. Am. Chem. Soc. 2012, 134, 1993−1995. (10) Bera, T. K.; Jang, J. I.; Song, J.-H.; Malliakas, C. D.; Freeman, A. J.; Ketterson, J. B.; Kanatzidis, M. G. Soluble semiconductors AAsSe2 (A= Li, Na) with a direct-band-gap and strong second harmonic generation: A combined experimental and theoretical study. J. Am. Chem. Soc. 2010, 132, 3484−3495. (11) Chung, I.; Jang, J. I.; Gave, M. A.; Weliky, D. P.; Kanatzidis, M. G. Low valent phosphorus in the molecular anions [P5Se12]5‑ and β[P6Se12]4‑: phase change behavior and near infrared second harmonic generation. Chem. Commun. 2007, 4998−5000. (12) Kanatzidis, M. G.; Sutorik, A. C. The Application of Polychalcogenide Salts to the Exploratory Synthesis of Solid State Multinary Chalcogenides at Intermediate Temperatures. Prog. Inorg. Chem. 1995, 43, 151−265. (13) Cahill, C. H.; Ko, Y.; Parise, J. B. A Novel 3-Dimensional Open Framework Sulfide Based upon the [In10S20]10‑ Supertetrahedron: DMA-InS-SB1. Chem. Mater. 1998, 10, 19−21. (14) Li, H.; Laine, A.; O’Keeffe, M.; Yaghi, O. M. Supertetrahedral Sulfide Crystals with Giant Cavities and Channels. Science 1999, 283, 1145−1147. (15) Zheng, N.; Bu, X.; Wang, B.; Feng, P. Microporous and Photoluminescent Chalcogenide Zeolite Analogs. Science 2002, 298, 2366−2369. (16) Zheng, N.; Bu, X.; Feng, P. Synthetic design of crystalline inorganic chalcogenides exhibiting fast-ion conductivity. Nature 2003, 426, 428−432. (17) Bu, X.; Zheng, N.; Wang, X.; Wang, B.; Feng, P. ThreeDimensional Frameworks of Gallium Selenide Supertetrahedral Clusters. Angew. Chem., Int. Ed. 2004, 43, 1502−1505. (18) Wu, T.; Bu, X.; Zhao, X.; Khazhakyan, R.; Feng, P. Phase Selection and Site-Selective Distribution by Tin and Sulfur in 7656

DOI: 10.1021/acs.inorgchem.7b00121 Inorg. Chem. 2017, 56, 7650−7656