Systematic Approach for the Construction of Niobium and Tantalum

Publication Date (Web): April 6, 2016 ... Treatment of the imido complexes [MCl3(NR)py2] (R = tBu, 2,6-Me2C6H3; M = Nb 1, 3; Ta 2, 4) (Xyl = 2,6-Me2C6...
0 downloads 13 Views 2MB Size
Download PDF
Article pubs.acs.org/IC

Systematic Approach for the Construction of Niobium and Tantalum Sulfide Clusters Manuel Gómez, Cristina Hernández-Prieto, Avelino Martín, Miguel Mena, and Cristina Santamaría* Departamento de Química Orgánica y Química Inorgánica, Universidad de Alcalá, Campus Universitario, E-28805 Alcalá de Henares, Madrid, Spain S Supporting Information *

ABSTRACT: Treatment of the imido complexes [MCl3(NR)py2] (R = tBu, 2,6-Me2C6H3; M = Nb 1, 3; Ta 2, 4) (Xyl = 2,6Me2C6H3) with (Me3Si)2S in a 1:1 ratio afforded the new cube-type sulfide clusters [MCl(NR)py(μ3-S)]4 (R = tBu, 2,6Me2C6H3; M = Nb 5, 7; Ta 6, 8) with loss of Me3SiCl. Reactions of 5 and 6 with cyclopentadienyllithium in 1:4 ratio resulted in the rupture of the coordinative M−S bonds and the replacement of a pyridine molecule and a chlorine atom by an η5cyclopentadienyl group in each metal center, affording the compounds [M(η5-C5H5)(NtBu)(μ-S)]4 (M = Nb 9, Ta 10). These processes may develop through formation of the complexes [M4(η5-C5H5)2(μ-Cl)(NtBu)4py2(μ3-S)2(μ-S)2](C5H5) (M = Nb 11, Ta 12), also obtained by reaction of 5 and 6 with cyclopentadienyllithium in 1:3 ratio. As further evidence, 11 and 12 led to complexes 9 and 10 by treatment with one more equivalent of the lithium reagent. The structural study of these metal sulfide clusters has been also performed by X-ray crystallography.

■

INTRODUCTION Transition metal sulfide clusters have been known for a long time. However, most research on synthetic sulfide clusters has been inspired by the architecture present in the nitrogenase system,1 and, therefore, iron and molybdenum are the elements studied more extensively by far, while the progress of group 5 sulfide clusters has been slower and piecemeal.2 Some authors explain this situation by the difficulty to develop such chemistry with a metal having a better affinity for oxygen than for sulfur.2e Focusing our attention on metal sulfide clusters with a cubetype structure,3 these species have been extremely interesting subjects of study due to their implications as reagents in organometallic synthesis,4 bioinorganic electron transfer,5 and industrial catalyst processes.6 These clusters are known for some early (Ti, V, Mo, W) and late (Re, Fe, Ru, Rh, Ni, Pd, Pt, Cu) transition metals and are characterized by a cubic core with four transition metal atoms occupying alternate corners of the cube along with four triply bridging sulfide ligands in the remaining corners. An examination of the structures of both homo- and heterometallic cube-type compounds reveals that in many of the homometallic clusters all of the metal atoms exhibit an identical pseudo-octahedral environment; there are only a few examples in which the metal atoms show two different coordination geometries. Other coordination environments in this kind of clusters may be pseudotetrahedral, square pyramidal, or trigonal bipyramidal. On the other hand, in the © XXXX American Chemical Society

mixed-metal clusters the different metals usually display diverse coordination geometries.3b To provide further knowledge on this challenging area, herein, we report a systematic way to synthesize a series of cube-type tetranuclear niobium and tantalum sulfide clusters by reaction between the haloimido derivatives [MCl3(NR)py2] (R = tBu, Xyl; M = Nb, Ta) and hexamethyldisilathiane (Me3Si)2S. Additionally, the reactivity of these [M4S4] compounds with cyclopentadienyllithium is also reported.

■

EXPERIMENTAL SECTION

General Procedures. All manipulations were carried out under a dry argon atmosphere using Schlenk-tube and cannula techniques or in a conventional argon-filled glovebox. Solvents were carefully refluxed over the appropriate drying agents and distilled prior to use: C6D6 and hexane (Na/K alloy), CDCl3 (CaH2), pyridine (P4O10), diethyl ether and tetrahydrofuran (Na/benzophenone), and toluene (Na). Starting materials [MCl3(NtBu)py2] (M = Nb 1, Ta 2),7 and [Li(η5-C5H5)]8 were synthesized according to published procedures. (Me3Si)2S was purchased from Aldrich and used as received. Microanalyses (C, H, N, S) were performed in a LECO CHNS-932 microanalyzer. Repeated attempts to obtain satisfactory elemental analysis for complexes 3, 5, 6, and 12 were unsuccessful. Received: December 8, 2015

A

DOI: 10.1021/acs.inorgchem.5b02816 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

mmol) and (Me3Si)2S (M = Nb, 0.42 g, 0.63 mmol; Ta, 0.34 g, 0.53 mmol) were heated to 80 °C for 24 h to yield dark orange (M = Nb) or pale yellow (M = Ta) microcrystalline solids. 7: Yield 0.12 g (46%). IR (KBr, cm−1): ν̅ = 3060 (w), 1604 (vs), 1484 (s), 1445 (vs), 1304 (s), 1292 (s), 1220 (s), 1096 (m), 1071 (m), 1042 (m), 1012 (s), 769 (s), 756 (s), 691 (s), 634 (w), 409 (m). 1H NMR (500 MHz, C5D5N): δ = 10.49 (m, 2H, Ho py), 7.70−6.50 (overlapping signals, 6H, py and 2,6-Me2C6H3N), 2.23 (s, 6H, 2,6-Me2C6H3N). 13C{1H} NMR (125 MHz, C5D5N): δ = 150.3−117.2 (py, 2,6-Me2C6H3N), 18.1 (2,6-Me 2 C 6 H 3 N). Anal. (%) Calcd for C 52 H 56 Cl 4 N 8 Nb 4 S 4 (1434.77): C, 43.53; H, 3.93; N, 7.81. Found: C, 41.72; H, 3.70; N, 7.31. 8: Yield 0.17 g (73%). IR (KBr, cm−1): ν̅ = 3062 (w), 1634 (m), 1607 (s), 1536 (s), 1482 (vs), 1248 (w), 1165 (m), 1096 (m), 857 (s), 760 (m), 682 (s), 492 (w). 1H NMR (500 MHz, C5D5N): δ = 12.75 (m, 2H, Ho py), 7.60−6.60 (overlapping signals, 6H, py and 2,6Me2C6H3N), 2.22 (s, 6H, 2,6-Me2C6H3N). 13C{1H} NMR (125 MHz, C5D5N): δ = 151.0−116.0 (py, 2,6-Me2C6H3N), 18.0 (2,6Me2C6H3N). Anal. (%) Calcd for C52H56Cl4N8S4Ta4 (1786.94): C, 34.95; H, 3.16; N, 6.27. Found: C, 34.54; H, 3.20; N, 6.14. Repeated attempts to obtain satisfactory elemental analysis (S) for these compounds were unsuccessful. Preparation of [Ta(η5-C5H5)(NtBu)(μ-S)]4 (10). [TaCl(NtBu)py(μ3-S)]4 (6) (0.60 g, 0.38 mmol) was solved in toluene (30−40 mL), and [Li(η5-C5H5)] (0.11 g; 1.52 mmol) was added. The mixture reaction was left heating at 90 °C for 4 days. After filtration the solution was concentrated to a few milliliters and cooled at −20 °C to give yellow crystalline solid identified as 10. Yield: 0.25 g (48%). IR (KBr, cm−1): ν̅ = 3087 (m), 2959 (s), 2914 (s), 1604 (m), 1438 (m), 1352 (s), 1224 (s), 1117 (s), 1011 (s), 806 (s). 1H NMR (500 MHz, C6D6): δ = 6.64 (s, 20H, C5H5), 1.36 (s, 36H, Me3CN). 13C{1H} NMR (125 MHz, C6D6): δ = 109.8 (C5H5), 66.0 (Me3CN), 33.2 (Me3CN). Anal. (%) Calcd for C36H56N4S4Ta4 (1396.91): C, 30.95; H, 4.04; N 4.01; S, 9.18. Found: C, 30.91; H, 4.02; N, 4.44; S, 9.38. Preparation of [M4(η5-C5H5)2(μ-Cl)(NtBu)4py2(μ3-S)2(μ2-S)2](C5H5) (M = Nb 11, Ta 12). The preparations were similar to that for 10, but with [MCl(NtBu)py(μ3-S)]4 (0.30 g; M = Nb 5, 0.24 mmol; M = Ta 6, 0.18 mmol) in toluene (30−40 mL) and [Li(η5C5H5) (M = Nb, 0.052 g; 0.72 mmol; M = Ta, 0.039 g; 0.54 mmol). The reaction mixture was left stirring at room temperature (M = Nb) or heated at 90 °C (M = Ta) for 3 days. After filtration the solution was concentrated to a few milliliters at −20 °C to give green (M = Nb) and red (M = Ta) crystalline solids. 11: Yield 0.17 g (60%). IR (KBr, cm−1): ν̅ = 2962 (s), 2916 (s), 1601 (m), 1441 (m), 1352 (m), 1255 (s), 1013 (m), 811 (s). 1H NMR (500 MHz, C6D6): δ = 9.57 (m, 4H, Ho py), 7.10−6.30 (m, 6H, Hm,Hp py), 6.58 (s, 10H, η5-C5H5), 5.79 (s, 5H, C5H5), 1.82 (s, 18H, Me3CN), 1.18 (s, 18H, Me3CN). 13 C{1H} NMR (125 MHz, C6D6): δ = 154.7 (Co py), 137.2 (Cp py), 123.0 (Cm py), 111.8 (η5-C5H5), 108.6 (C5H5), 68.1 (Me3CN), 66.7 (Me3CN), 32.0 (Me3CN), 31.6 (Me3CN). Anal. (%) Calcd for C41H61ClN6Nb4S4 (1173.31): C, 41.97; H, 5.24; N 7.16; S, 10.93. Found: C, 42.07; H, 5.25; N 7.13; S, 11.07. 12: Yield 0.19 g (69%). IR (KBr, cm−1): ν̅ = 2963 (s), 2916 (s), 1599 (m), 1440 (m), 1353 (m), 1232 (s), 1115 (m), 1013 (m), 802 (s). 1H NMR (500 MHz, C6D6): δ = 9.72 (m, 4H, Ho py), 7.10−6.30 (m, 6H, Hm,Hp py), 6.61 (s, 10H, η5-C5H5), 5.80 (s, 5H, C5H5), 1.79 (s, 18H, Me3CN), 1.20 (s, 18H, Me3CN). 13C{1H} NMR (125 MHz, C6D6): δ = 154.8 (Co py), 137.5 (Cp py), 123.3 (Cm py), 110.7 (η5-C5H5), 108.6 (C5H5), 65.3 (Me3CN), 63.3 (Me3CN), 33.1 (Me3CN), 31.7 (Me3CN). Anal. (%) Calcd for C41H61ClN6S4Ta4 (1525.47): C, 32.28; H, 4.03; N 5.51; S, 8.41. Found: C, 32.83; H, 4.27; N, 6.33; S, 8.95. Crystal Structure Determination of Complexes 4, 5, 5a, 6, 7a, 8, 9, 11, and 12. Crystals were grown by slow evaporation at room temperature of saturated toluene or benzene-d6 (5) solutions. Then crystals were removed from the Schlenks and covered with a layer of a viscous perfluoropolyether (FomblinY). A suitable crystal was selected with the aid of a microscope, mounted on a cryoloop, and immediately placed in the low-temperature nitrogen stream of the diffractometer. The intensity data sets were collected at 200 K (150 K in the case of 12) on a Bruker-Nonius KappaCCD diffractometer equipped with an Oxford Cryostream 700 unit. Crystallographic data

Samples for IR spectroscopy were prepared as KBr pellets and recorded on FT-IR spectrophotometers: PerkinElmer Spectrum 2000 and PerkinElmer Frontier (4000−400 cm−1). NMR spectra were recorded on Mercury-300, UnityPlus-300, and VNMRS-500 spectrometers. The 1H and 13C{1H} chemical shifts were referenced to the solvent signals, and chemical shifts were reported relative to tetramethylsilane. Preparation of [MCl3(NXyl)py2] (M = Nb 3, Ta 4) (Xyl = 2,6Me2C6H3). To a stirred suspension of MCl5 (M = Nb, 1.27 g, 4.70 mmol; Ta, 4.15 g, 11.6 mmol) in 20 mL of toluene was slowly added diethyl ether (M = Nb, 4 mL; Ta, 8 mL), immediately producing a clear orange (M = Nb) or pale yellow (M = Ta) solution. After a pyridine solution (M = Nb 2.5 mL; Ta 5 mL) of NH(2,6Me2C6H3)(SiMe3) (M = Nb, 1.83 g, 9.41 mmol; Ta, 5.77 g, 23.10 mmol) was added, the color of the mixtures quickly changed to red. The resulting suspensions were stirred at room temperature for 3 h (M = Nb) or heated to 60 °C for 1 day (M = Ta). Over this time, the solid was filtered off, washed with hexane (2 × 5 mL), and dried under vacuum, giving 3 and 4 as red microcrystalline solids. 3: Yield 1.64 g (73%). IR (KBr, cm−1): ν̅ = 1605 (vs), 1486 (s), 1444 (vs), 1306 (s), 1222 (vs), 1096 (m), 1069 (s), 1040 (s), 982 (m), 757 (vs), 701 (vs), 638 (s), 597 (m). 1H NMR (300 MHz, CDCl3): δ = 9.08 (m, 2H, Ho pyax), 8.82 (br m, 2H, Ho pyec), 8.0−6.7 (overlapping signals, 9H, py and 2,6-Me2C6H3N), 2.66 (s, 6H, 2,6-Me2C6H3N). 13C{1H} NMR (75 MHz, CDCl3): δ = 152.3 (broad signal, Co pyax), 151.6 (Co py ec ), 140.0−123.0 (py and 2,6-Me 2 C 6 H 3 N), 19.0 (2,6Me2C6H3N). Anal. (%) Calcd for C18H19Cl3N3Nb (476.63): C, 45.35; H, 4.02; N 8.82. Found: C, 45.80; H, 4.36; N, 7.61. 4: Yield: 2.58 g (82%). IR (KBr, cm−1): ν̅ = 1610 (vs), 1487 (s), 1459 (s), 1447 (vs), 1328 (s), 1223 (vs), 1099 (m), 1071 (s), 1047 (s), 1016 (s), 869 (m), 758 (vs), 736 (s), 689 (vs), 644 (s), 582 (m). 1H NMR (300 MHz, CDCl3): δ = 9.06 (m, 2H, Ho pyax), 8.86 (br m, 2H, Ho pyec), 8.0−6.4 (overlapping signals, 9H, py and 2,6-Me2C6H3N), 2.64 (s, 6H, 2,6-Me2C6H3N). 13C{1H} NMR (75 MHz, CDCl3): δ = 152.6 (Co pyax), 151.9 (broad signal, Co pyec), 145.0−124.0 (py and 2,6Me2C6H3N), 18.7 (2,6-Me2C6H3N). Anal. (%) Calcd for C18H19Cl3N3Ta (564.67): C, 38.28; H, 3.39; N, 7.44. Found: C, 37.97; H, 3.74; N, 7.40. General Procedure for the Synthesis of [MCl(NR)py(μ3-S)]4 (R = tBu, Xyl; M = Nb 5, 7; Ta 6, 8). A toluene (20 mL) solution of [MCl3(NR)py2] was placed into a Carius tube fitted with a Young’s valve, and under rigorously anhydrous conditions, a toluene (10 mL) solution of (Me3Si)2S was added. The argon pressure was reduced, and the reaction mixture was stirred and heated. The solid formed was filtered off, washed with hexane (2 × 5 mL), and dried under vacuum. Preparation of [MCl(NtBu)py(μ3-S)]4 (M = Nb 5, Ta 6). [MCl3(NtBu)py2] (1.00 g; M = Nb 1, 2.33 mmol; Ta 2, 1.93 mmol) and (Me3Si)2S (M = Nb, 0.41 g, 2.33 mmol; Ta, 0.34 g, 1.93 mmol) were heated to 80 °C (M = Nb) or 110 °C (M = Ta) for 24 (M = Nb) or 48 h (M = Ta) to yield ochre (M = Nb) or dark yellow (M = Ta) solids. 5: Yield 0.69 g (90%). IR (KBr, cm−1): ν̅ = 2968 (s), 2916 (m), 1603 (s), 1486 (m), 1445 (vs), 1357 (m), 1239 (vs), 1220 (vs), 1071 (m), 1040 (m), 1010 (m), 756 (m), 694 (vs), 632 (m), 556 (m), 430 (w). 1H NMR (300 MHz, C6D6): δ = 8.89 (m, 2H, Ho py), 6.33 (t, 1H, Hp py), 5.94 (m, 2H, Hm py), 1.75 (s, 9H, Me3CN). 13 C{1H} NMR (75 MHz, C6D6): δ = 153.7 (Co py), 137.2 (Cp py), 122.7 (Cm py), 69.5 (Me3CN), 30.6 (Me3CN). Anal. (%) Calcd for C36H56Cl4N8Nb4S4 (1242.59): C, 34.80; H, 4.54; N, 9.02; S, 10.32. Found: C, 35.60; H, 4.72; N, 9.20; S, 10.17. 6: Yield 0.52 g (70%). IR (KBr, cm−1): ν̅ = 2966 (s), 2915 (m), 1605 (s), 1486 (m), 1445 (vs), 1355 (m), 1267 (vs), 1219 (vs), 1155 (m), 1071 (m), 1042 (m), 1012 (m), 757 (s), 693 (vs), 635 (m), 554 (m), 432 (m). 1H NMR (300 MHz, C6D6): δ = 9.03 (m, 2H, Ho py), 6.30 (t, 1H, Hp py), 5.94 (m, 2H, Hm py), 1.74 (s, 9H, Me3CN). 13C{1H} NMR (75 MHz, C6D6): δ = 154.2 (Co py), 137.6 (Cp py), 123.0 (Cm py), 66.1 (Me3CN), 32.5 (Me3CN). Anal. (%) Calcd for C36H56Cl4N8S4Ta4 (1594.75): C, 27.11; H, 3.54; N, 7.02; S, 8.04. Found: C, 27.35; H, 3.78; N, 7.74; S, 9.04. Preparation of [MCl(NXyl)py(μ3-S)]4 (M = Nb 7, Ta 8). [MCl3(NXyl)py2] (0.30 g; M = Nb 3, 0.63 mmol; Ta 4, 0.53 B

DOI: 10.1021/acs.inorgchem.5b02816 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry for all complexes are presented in Table S1 in the Supporting Information. The structures were solved, by using the WINGX package,9 by direct methods (SHELXS-2013 for complexes 4, 5a, 7a, 8, 9, and 11)10 or intrinsic phasing (SHELXT-2014 for 5, 6, and 12)10 and refined by least-squares against F2 (SHELXL-2014).10 All the hydrogen atoms were positioned geometrically and refined by using a riding model. All crystals diffracted weakly, and only data collections up to θ = 25° could be performed for crystals of 6, 7a, 8, and 12. Complex 4 crystallized with one disordered molecule of toluene. By using the corresponding Shelxl’s PART commands10 and FVAR variables, two positions were refined with 50.8% and 49.2% occupancy, respectively. On the other hand, two and a half molecules of benzene crystallized with each cube-type unit of 5. In both cases, all the nonhydrogen atoms were refined anisotropically. Crystals of 5a, 6, 7a, and 9 crystallized with a huge number of toluene solvent molecules, but it was not possible to get sensible chemical models for them. The Squeeze11 procedure of the PLATON package was employed to remove the contribution of that electronic density to the structure factors, obtaining solvent-accessible volumes in the range 34.7−38.32% of the unit cell volume. All the non-hydrogen atoms were refined anisotropically. However, it is important to highlight the sensitivity of complex 9 to loss of solvent; it decomposes at room temperature in a few seconds from marvelous octahedral transparent orange crystals to yellow powder even if they are perfectly covered with the perfluoropolyether. X-ray diffraction study of 8 enabled modeling up to four toluene solvent molecules, but the presence of more solvent still forced us to run the Squeeze11 procedure to remove the contribution of the nonmodeled solvent molecules to the structure factors. All the nonhydrogen atoms were refined anisotropically. The crystal structure of complex 11 presented a disordered pyridine N1−C15 unit, linked to Nb1, and it was refined in two positions; the PART commands and a second FVAR parameter led us to assign occupancies of 61.9% and 38.1%, respectively. Moreover, the cyclopentadienide anion presented severe disorder too; two positions were found for the ring, but only an isotropic model with 25% occupancy could be used for the carbon atoms. Similarly to 11, the crystal structure of 12 showed disorder in the N3−C19 pyridine, with 64.43% and 35.57% occupancy for the two positions found, respectively. The above-mentioned treatment for the cyclopentadienide anion in 11 was also used for 12. Unfortunately, we could not model the toluene solvent molecules for 11 and 12, and we had to use the Squeeze11 procedure. All non-hydrogen atoms were refined anisotropically, except those of the cyclopentadienide rings for 11 and 12.

Scheme 1. Syntheses of Cube-Type Clusters 5−8

[MCl3(NR)py2] (R = tBu, Xyl; M = Nb 1, 3; Ta 2, 4) with an equimolar amount of (Me3Si)2S afforded the new cube-type sulfide clusters [MCl(NR)py(μ3-S)]4 (R = tBu, Xyl; M = Nb 5, 7; Ta 6, 8) along with Me3SiCl, as outlined in Scheme 1. Compounds 5 and 6 are discretely soluble in the usual solvents such as benzene, toluene, or tetrahydrofuran and scarcely soluble in hexane, while otherwise compounds 7 and 8 are hardly soluble in the mentioned solvents but soluble in pyridine and chloroform. Fortunately, we succeeded in elucidating the crystal structures of complexes 5, 6, and 8 by X-ray diffraction studies. The NMR spectra reveal the resonances for the tertbutylimido ligands and the corresponding distinctive signals of the pyridine molecules, consistent with a symmetric structure in solution. By comparing the 1H and 13C{1H} NMR chemical shifts for complexes 5 and 6 with the pseudooctahedral imido [MCl3(NtBu)py2] (M = Nb 1, Ta 2),7b the resonances found for the quaternary carbon of the tert-butylimido moiety in the cube-type species (M = Nb 5, 69.5 ppm; M = Ta 6, 66.1 ppm) and in the starting materials (M = Nb 1, 72.6 ppm; M = Ta 2, 67.0 ppm) are in agreement with the electronegativity values for sulfur, nitrogen, and chlorine. The IR spectra of all complexes show the characteristic absorption corresponding to the MN− stretching vibration (ν̅ 1304−1383 cm−1).7b,13 The solid-state structure of complex 5 is depicted in Figure 1, while the analogous molecular structures of 6 and 8, together

■

RESULTS AND DISCUSSION In an effort to provide a rational methodology for the preparation of sulfur cluster complexes bearing niobium and tantalum we decided to use the mononuclear species [MCl3(NR)py2] (M = Nb, Ta) as starting materials.7 Their synthetic procedure comprises the reaction of MCl5 with the corresponding amines R(Me3Si)NH and pyridine in hydrocarbon solvents. Thus, we have synthesized the new complexes [MCl3(NXyl)py2] (Xyl = 2,6-Me2C6H3; M = Nb 3, Ta 4) following the procedures published for the tert-butylimido derivatives.7b The proposed structures for 3 and 4 are summarized in Scheme 1, while the syntheses and characterizing data are collected in the Experimental Section. Additionally, a single-crystal X-ray diffraction analysis of 4 is shown in the Supporting Information. On the other hand, it has already been established that the use of silylated chalcogenides offers easy access to transition metal clusters.2b,f,12 Thus, we used the distorted octahedral imido compounds [MCl3(NR)py2] as building block complexes and the hexamethyldisilathiane reagent, (Me3Si)2S, as sulfur source. Therefore, treatment of the imido complexes

Figure 1. Molecular structure of 5. Thermal ellipsoids are at 50% probability. Hydrogen atoms have been omitted for clarity. The full listing of the core bond angles and distances can be found in the Supporting Information. C

DOI: 10.1021/acs.inorgchem.5b02816 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry with the full listing of the core bond angles and distances found for 5, can be found in the Supporting Information. All of them reveal homometallic M4S4 cube-type clusters where four metal atoms and four triply bridging sulfur atoms alternatively occupy the vertices of the cube. Nb−S−Nb and S−Nb−S angles are in the range 86.7(1)−99.1(1)° and 80.5(1)−92.1(1)°, respectively. In the case of 5 and 6, each metal is also linked to a chlorine atom, a pyridine molecule, and a tert-butylimido ligand in a pseudooctahedral environment; in that way the whole symmetry of the molecules corresponds to an S4 point group, with the axis running perpendicular to the face formed by Nb1, S1, Nb2, and S2 in 5 (see Figure 1 and Figure S3 for 6). However, the higher steric hindrance of the 2,6-dimethylphenylimide fragments prevents adopting the same symmetry of 5 or 6 in the case of complex 8. Additionally, the three cube-type structures show remarkable differences between the lengths of metal−sulfur bonds; for example, in complex 5 the Nb−S bonds located in trans position with respect to the imide groups present an average length of 2.82(2) Å, while those located in cis position show shorter average distances of 2.41(1) and 2.510(6) Å, respectively. On the other hand, the bond distance from the niobium metal centers to the imido groups are in the range found for the precursor monomeric species (see Figure S1). As expected for these metal centers, it is not possible to find relevant differences in the geometrical parameters within the core when comparing the structures of 5, 6, and 8 (see Supporting Information). To our knowledge, only the crystal structures of two homometallic M4O4 (M = Nb, Ta)14 and a few heterometallic M2M′2E4 (M = Nb, Ta; M′ = Ni, Cu, Ag, E = S, Se) cube-type clusters have been reported to date,2f,g,15 complexes 5−8 then being the first group 5 homometallocubane M4S4 complexes. During the examination under the microscope of a fraction of crystals of 5 we were able to identify a new species whose solidstate structure revealed an opened-cube cluster, 5a (Figure 2), with the same empirical formula as that of 5. Analogously, the existence of complex 7a was also discovered by an X-ray study (Figure S5 in the Supporting Information). Unfortunately, attempts to synthesize or detect them in solution were unsuccessful. Their molecular structures can be described as double-opened cubes, where the two halves of the molecules are related by a center of symmetry. Both figures evidence that one of the three M−S bonds of each metal center found in 5 or 6 has disappeared, leading to the existence of only μ-S bridging ligands with slightly shorter M−S distances (∼0.1 Å) and the created vacancies occupied by two μ3-chlorine atoms in the analogous complexes 5a and 7a, respectively. Geometrical parameters around the niobium metal centers still allow assigning them pseudooctahedral environments as those found for 5. As can be observed in Figure 2, all ligands located in trans position with respect to the imide groups again show longer distances than those found in cis position. It is noticeable that CSD16 contains only a few homometallic molecular structures with a geometry similar to those of 5a and 7a, and all of them comprise [M4(μ3-S)2(μ-S)4] (M = Mo, W)17 cores where the group 6 elements are not bridged by halogen atoms. In order to move forward with our study on these cube-type clusters having metal−sulfur cores, along with a chloride ligand bound to the metal atom, we next investigated the reactivity of the soluble complexes 5 and 6 with cyclopentadienyllithium in the corresponding metathesis reaction. Thus, the treatment of the heterocubane species with four equivalents of the lithium

Figure 2. Molecular structure of 5a. Thermal ellipsoids are at 50% probability. Hydrogen atoms have been omitted for clarity. Selected bond lengths (Å) and angles (deg): Cl1−Nb2 2.438(2), Cl3a-Nb2 2.407(1), Cl3a-Nb1 2.507(1), Cl3a-Nb1a 2.789(1), Nb1−N2 1.753(4), Nb1−N1 2.323(5), Nb1−S5 2.342(1), Nb1−S4 2.558(1), Nb2−N22 1.747(5), Nb2−N21 2.362(5), Nb2−S5a 2.478(2), Nb2− S4 2.725(2), Nb2−Cl3a−Nb1 102.2(1), Nb2−Cl3a−Nb1a 87.6(1), Nb1−Cl3−Nb1a 92.9(1), N2−Nb1−N1 93.1(2), S5−Nb1−Cl3a 100.8(1), S5−Nb1−S4 165.1(1), Cl3a−Nb1−S4 83.0(5), S5−Nb1− Cl3 84.8(1), Cl3−Nb1−Cl3a 87.1(1), S4−Nb1−Cl3 80.9(1), N22− Nb2−N21 96.7(2), Cl3a−Nb2−S5a 90.7(1), Cl3a−Nb2−S4 81.4(1), S5−Nb2−S4 85.2(1), Nb1−S4−Nb2 92.7(1), Nb1−S5−Nb2a 96.8(1). Symmetry transformation used to generate equivalent atoms: (a) 1−x, 2−y, 2−z.

derivative in toluene at room temperature (M = Nb) or 90 °C (M = Ta) proceeds to give the compounds [M4(η5-C5H5)4(NtBu)4(μ-S)4] (M = Nb 9, Ta 10), as orange or yellow solids, respectively, in moderate yields, in the case of 10. This reaction is straightforward with substitution of the chloride and pyridine ligands in complexes 5 and 6 by the η5-cyclopentadienyl rings, as outlined in Scheme 2 (top). Compound 10 was characterized by elemental analysis and spectroscopic data, and the structure of 9 was determined by single-crystal X-ray diffraction. The structure of this complex is presented in Figure 3, along with a selection of bonds and angles, and shows a cyclooctane-like arrangement of niobium and sulfur atoms in a boat conformation, the two halves of the molecule being related by a C2 axis. A staggered disposition of the tert-butylimido and cyclopentadienyl ligands is also observed for neighboring metal centers. The coordination sphere around the niobium atoms is the usual three-legged piano stool geometry. The niobium−sulfide bond lengths, in the range 2.368(3)− 2.435(3) Å, are shorter than those found in complexes 5 and 7a, probably due to the higher strain in the cube-type structures and the absence of ligands in trans position with respect to the imide ligands in 9. To our knowledge, this is the first M4S4 crystal structure with all the sulfur atoms acting as μ-S bridges, which constitutes a very opened structure, extremely sensitive to moisture and loss of solvent. In fact, single crystals of 10 were also obtained, but only a few frames could be registered before their decomposition, although enough to calculate the unit cell and see a boat Ta4S4 D

DOI: 10.1021/acs.inorgchem.5b02816 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

of compound 10. However, all attempts to isolate 9 in pure form for the spectroscopic characterization were unsuccessful. Upon monitoring the metathesis reaction of the cube-type derivative 5 with [Li(η5-C5H5)] by 1H NMR in benzene-d6 solutions at room temperature, the presence of another set of signals along with that found for compound 9 was revealed. In an attempt to isolate and characterize new species, we decided to carry out the reactions of 5 and 6 with [Li(η5-C5H5)] in different stoichiometric ratios. Thus, when three equivalents of cyclopentadienyllithium were added to toluene solutions of 5 or 6, the compounds [M4(NtBu)4(η5-C5H5)2(μ3-S)2(μ2-S)2(μ2Cl)py2][C5H5] (M = Nb 11, Ta 12) were isolated, as green and red microcrystalline solids, respectively, in moderate yields (Scheme 2). Furthermore, we could demonstrate that the addition of one equivalent of cyclopentadienyllithium to benzene-d6 solutions of complexes 11 and 12 in NMR tube essays afforded complexes 9 and 10, respectively, in a quantitative way as outlined in Scheme 2. Complexes 11 and 12 were fully characterized by NMR and IR spectroscopy, elemental analysis, and single-crystal X-ray diffraction. The 1H NMR spectra evidenced the existence of two different tert-butylimido and cyclopentadienyl ligands, together with the presence of pyridine (see Scheme 2). The characteristic bands corresponding to the ν(MN−) stretching vibration were observed.7b,13 Crystals of 11 and 12 derivatives proved to be useful for an X-ray structure analysis and revealed the ionic structure of these complexes depicted in Scheme 2. The solid-state structure of complex 11 is shown in Figure 4 along with a selection of

Scheme 2. Syntheses of Sulfur Clusters 9−12

Figure 3. Molecular structure of 9. Thermal ellipsoids are at 50% probability. Hydrogen atoms have been omitted for clarity. Selected average bond lengths (Å) and angles (deg): Nb1−N1 1.759(8), Nb1− S2 2.386(3), Nb1−S1 2.435(3), Nb2−N2 1.757(9), Nb2−S1a 2.368(3), Nb2−S2 2.427(3), N1−Nb1−S2 96.6(3), N1−Nb1−S1 105.6(3), S2−Nb1−S1 110.9(1), N2−Nb2−S1a 100.0(3), N2−Nb2− S2 103.6(3), S1a−Nb2−S2 109.3(1), Nb2a−S1−Nb1 117.9(1), Nb1−S2−Nb2 116.4(1), C16−N1−Nb1 166.7(9), C26−N2−Nb2 168.5(9). Symmetry transformation used to generate equivalent atoms: (a) 1−x, y, −z+3/2.

Figure 4. Molecular structure of 11. Thermal ellipsoids are at 50% probability. Hydrogen atoms have been omitted for clarity. Selected bond lengths (Å) and angles (deg): N2−Nb1 1.760(5), N4−Nb2 1.772(6), Nb1−N1 2.367(6), Nb1−S4 2.410(2), Nb1−Cl1 2.417(2), Nb1−S1a 2.601(2), Nb1−S1 2.829(2), Nb2−S1 2.365(2), Nb2−S4 2.394(2), N2−Nb1−N1 89.6(3), N2−Nb1−S4 93.2(2), N1−Nb1−S4 92.6(2), N2−Nb1−Cl1 100.6(2), N1−Nb1−Cl1 162.1(2), S4−Nb1− Cl1 101.4(1), N2−Nb1−S1a 108.6(2), N1−Nb1−S1a 74.9(2), S4− Nb1−S1a 154.5(1), Cl1−Nb1−S1a 87.9(1), N2−Nb1−S1 176.6(2), N1−Nb1−S1 87.0(2), S4−Nb1−S1 86.6(1), Cl1−Nb1−S1 82.8(1), S1−Nb1−S1a 70.9(1), N4−Nb2−S1 100.1(2), N4−Nb2−S4 103.7(2), S1−Nb2−S4 98.5(1), Nb2−S1−Nb1a 123.6(1), Nb2− S1−Nb1 82.5(1), Nb1−S1−Nb1a 78.1(1), Nb1−Cl1−Nb1a 90.2(1), Nb2−S4−Nb1 91.5(1). Symmetry transformation used to generate equivalent atoms: 1−x, y, 0.5−z.

molecular core, identical to that of 9, in the first steps of the crystal structure determination. Two singlets assignable to equivalent tert-butylimido ligands and cyclopentadienyl groups, appear in the 1H NMR spectrum E

DOI: 10.1021/acs.inorgchem.5b02816 Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

■

ACKNOWLEDGMENTS Financial support for this work was provided by the Ministerio de Economiá y Competitividad (CTQ2013-44625-R) and the Universidad de Alcalá (CCG2014/EXP-022). C.H.-P. thanks the Spanish Ministerio de Educación, Ciencia y Deporte (MECD) for a Beca de Colaboración.

interatomic distances and angles, while the analogous molecular structure of 12 can be found in Figure S6 of the Supporting Information. As Figure 4 illustrates, a distorted ladder-type tricyclic structure similar to that published by Coucouvanis and Kalyvas for (PPN)2[Fe4S4(NO)6]18 is found for the cationic part of compound 11 and consists of two [Nb(η5-C5H5)(NtBu)] and two [Nb(NtBu)py] building units linked by four bridging sulfur and one chlorine atom, showing a C2 symmetry with the axis running through Cl1, perpendicular to the mean plane of the ladder. The pyridine-coordinated niobium atoms exhibit a distorted pseudo-octahedral geometry with the equatorial positions being occupied by two sulfur atoms (S4 and S1a), one chlorine atom (Cl1), and a pyridine molecule (N1) and the axial positions by a tert-butylimido ligand and the third sulfur atom. On the other hand, the environment of the adjacent niobiums is again a three-legged piano-stool with the cyclopentadienyl rings located on the apical positions and two sulfur atoms and one tert-butylimido ligand occupying the basal apexes. A distorted pyramidal geometry is adopted by the μ3sulfur atom (S1), while the angle formed around S4 is close to 90°. In order to complete the study on the reaction of 5 and 6 with cyclopentadienyllithium, we performed the treatment in a lower ratio to check the possibility to isolate new species. Unfortunately, all attempts were unsuccessful, and the reactions always led to complexes 11 and 12. However, it is reasonable to propose the formation of intermediates, such as those depicted in Scheme 2, by substitution of the chloride and pyridine ligands of the external metal atoms in the derivative 5a (see Figure 2).

■

CONCLUSIONS We have developed a systematic route to a family of homometallic sulfide cube-type clusters bearing M4S4 (M = Nb, Ta) cores taking as starting point the trichloroimido compounds [MCl3(NR)py2] and (Me3Si)2S. The closed geometry of these new complexes can be opened by treatment with cyclopentadienyllithium in different ratios to afford interesting tetranuclear niobium or tantalum derivatives. The possibility to incorporate other inorganic fragments through interaction with the sulfide ligands in the reported compounds is under study. ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.5b02816. Crystallogrphic data (CIF) Experimental data for the X-ray diffraction studies on 4· C7H8, 5·2.5C6D6, 5a, 6, 7a, 8·4C7H8, 9, 11, and 12. ORTEP drawings and selected bond distances and angles for the molecular structures of 4·C7H8 (Figure S1), 5 (Figure S2), 6 (Figure S3), 8 (Figure S4), 7a (Figure S5), and 12 (Figure S6) (PDF)

■

REFERENCES

(1) (a) Karlin, K. D. Science 1993, 261, 701−708. (b) Eady, R. R. Chem. Rev. 1996, 96, 3013−3030. (c) Beinert, H.; Holm, R. H.; Münck, E. Science 1997, 277, 653−659. (d) Stiefel, E. I. Pure Appl. Chem. 1998, 70, 889−896. (e) Malinak, S. M.; Coucouvanis, D. In Progress in Inorganic Chemistry; Karlin, K. D., Ed.; John Wiley & Sons, Inc.: New York, 2001; Vol. 49, pp 599−662. (f) Lee, S. C.; Holm, R. H. Chem. Rev. 2004, 104, 1135−1157. (g) Ohki, Y.; Tatsumi, K. Z. Anorg. Allg. Chem. 2013, 639, 1340−1349. (h) Hoffman, B. M.; Lukoyanov, D.; Yang, Z.-Y.; Dean, D. R.; Seefeld, L. C. Chem. Rev. 2014, 114, 4041−4062. (i) Ribbe, M. W.; Hu, Y.; Hodgson, K. O.; Hedman, B. Chem. Rev. 2014, 114, 4063−4080. (2) (a) Sola, J.; Do, Y.; Berg, J. M.; Holm, R. H. Inorg. Chem. 1985, 24, 1706−1713. (b) Fenske, D.; Ohmer, J.; Hachgenei, J.; Merzweiler, K. Angew. Chem., Int. Ed. Engl. 1988, 27, 1277−1296. (c) Leblanc, J.C.; Moise, C.; Volpato, F.; Brunner, H.; Gehart, G.; Wachter, J.; Nuber, B. J. Organomet. Chem. 1995, 485, 237−242. (d) Kawaguchi, H.; Tatsumi, K. Organometallics 1997, 16, 307−309. (e) SimonnetJégat, C.; Sécheresse, F. Chem. Rev. 2001, 101, 2601−2611. (f) Pätow, R.; Fenske, D. Z. Z. Anorg. Allg. Chem. 2002, 628, 1279−1288. (g) Pätow, R.; Fenske, D. Z. Z. Anorg. Allg. Chem. 2002, 628, 2790− 2794. (h) Clérac, R.; Fenske, D.; Issac, I.; Rothenberger, A. J. Cluster Sci. 2004, 15, 189−198. (i) Sokolov, M. N.; Fedin, V. P. Coord. Chem. Rev. 2004, 248, 925−944. (j) Sokolov, M. N.; Gushchin, A. L.; Abramov, P. A.; Virovets, A. V.; Peresypkina, E. V.; Kozlova, S. G.; Kolesov, B. A.; Vicent, C.; Fedin, V. P. Inorg. Chem. 2005, 44, 8756− 8761. (k) Taniyama, N.; Ohki, Y.; Tatsumi, K. Inorg. Chem. 2014, 53, 5438−5446. (3) (a) Bandy, J. A.; Davies, C. E.; Green, J. C.; Green, M. L. H.; Prout, K.; Rodgers, D. P. S. J. Chem. Soc., Chem. Commun. 1983, 1395−1397. (b) Harris, S. Polyhedron 1989, 8, 2843−2882. (c) Shibahara, T. Coord. Chem. Rev. 1993, 123, 73−147. (d) Hernández-Molina, R.; Sykes, A. G. Coord. Chem. Rev. 1999, 187, 291−302. (e) Llusar, R.; Uriel, S. Eur. J. Inorg. Chem. 2003, 2003, 1271−1290. (f) Seino, H.; Kaneko, T.; Fujii, S.; Hidai, M.; Mizobe, Y. Inorg. Chem. 2003, 42, 4585−4596. (g) Sorribes, I.; Lloret, F.; Waerenborgh, J. C.; Polo, V.; Llusar, R.; Vicent, C. Inorg. Chem. 2012, 51, 10512−10521. (4) (a) Shibahara, T. Adv. Inorg. Chem. 1991, 37, 143−173. (b) Dance, I.; Fisher, K. In Progress in Inorganic Chemistry; Karlin, K. D., Ed.; John Wiley & Sons, 1994; Vol. 41, pp 637−803. (c) Kabashima, S.; Kuwata, S.; Hidai, M. J. Am. Chem. Soc. 1999, 121, 7837−7845. (d) Hidai, M.; Kuwata, S.; Mizobe, Y. Acc. Chem. Res. 2000, 33, 46−52. (5) (a) Holm, R. H.; Ciurli, S.; Weigel, J. A. In Progress Inorganic Chemistry; Lippard, S. J., Ed.; John Wiley & Sons, 1990; Vol. 38, pp 1− 74. (b) Holm, R. H. Adv. Inorg. Chem. 1992, 38, 1−71. (c) HernándezMolina, R.; Sokolov, M. N.; Sykes, A. G. Acc. Chem. Res. 2001, 34, 223−230. (d) Münck, E.; Bominaar, E. L. Science 2008, 321, 1452− 1453. (6) (a) Herbst, K.; Monari, M.; Brorson, M. Inorg. Chem. 2002, 41, 1336−1338. (b) Seino, H.; Masumori, T.; Hidai, M.; Mizobe, Y. Organometallics 2003, 22, 3424−3431. (c) Takei, I.; Dohki, K.; Kobayashi, K.; Suzuki, T.; Hidai, M. Inorg. Chem. 2005, 44, 3768− 3770. (d) Seino, H.; Hidai, M. Chem. Sci. 2011, 2, 847−857. (e) Shim, Y.; Yuhas, B. D.; Dyar, S. M.; Smeigh, A. L.; Douvalis, A. P.; Wasielewski, M. R.; Kanatzidis, M. G. J. Am. Chem. Soc. 2013, 135, 2330−2337. (f) Buchwalter, P.; Rosé, J.; Braunstein, P. Chem. Rev. 2015, 115, 28−126. and references therein.

■

■

Article

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest. F

DOI: 10.1021/acs.inorgchem.5b02816 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry (7) (a) Chiu, H.-T.; Chuang, S.-H.; Tsai, C.-E.; Lee, G.-H.; Peng, S.M. Polyhedron 1998, 17, 2187−2190. (b) Galajov, M.; García, C.; Gómez, M. Dalton Trans. 2011, 40, 413−420. (8) Dinnebier, R. E.; Behrens, U.; Olbrich, F. Organometallics 1997, 16, 3855−3858. (9) Farrugia, L. J. J. Appl. Crystallogr. 2012, 45, 849−854. (10) (a) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr. 2008, A64, 112−122. (b) Sheldrick, G. M. Acta Crystallogr. 2015, C71, 3−8. (c) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Adv. 2015, A71, 3−8. (11) Spek, A. L. Acta Crystallogr. 2015, C71, 9−19. (12) (a) Dehnen, S.; Eichhöfer, A.; Fenske, D. Eur. J. Inorg. Chem. 2002, 2002, 279−317. (b) Corrigan, J. F.; Fuhr, O.; Fenske, D. Adv. Mater. 2009, 21, 1867−1871. (c) MacDonald, D. G.; Corrigan, J. F. Phil. Trans. R. Soc., A 2010, 368, 1455−1472 and references herein.. (13) Heinselman, K. S.; Miskowski, V. M.; Geib, S. J.; Wang, L. C.; Hopkins, M. D. Inorg. Chem. 1997, 36, 5530−5538. (14) Chen, S.-J.; Abbott, J. K. C.; Steren, C. A.; Xue, Z.-L. J. Cluster Sci. 2010, 21, 325−337. (15) (a) Lorenz, A.; Fenske, D. Z. Z. Anorg. Allg. Chem. 2001, 627, 2232−2248. (b) Pätow, R.; Issac, I.; Fenske, D. Z. Z. Anorg. Allg. Chem. 2003, 629, 1437−1440. (c) Bechlars, B.; Issac, I.; Feuerhake, R.; Clérac, R.; Fuhr, O.; Fenske, D. Eur. J. Inorg. Chem. 2008, 2008, 1632− 1644. (16) Allen, F. H. Acta Crystallogr., Sect. B: Struct. Sci. 2002, B58, 380− 388. (17) (a) Tsuge, K.; Mita, S.; Fujita, H.; Imoto, H.; Saito, T. J. Cluster Sci. 1996, 7, 407−421. (b) Kuwata, S.; Mizobe, Y.; Hidai, M. J. Chem. Soc., Dalton Trans. 1997, 1753−1758. (c) Schwarz, D. E.; Rauchfuss, T. B.; Wilson, S. R. Inorg. Chem. 2003, 42, 2410−2417. (d) Sattler, A.; Parkin, G. Polyhedron 2014, 84, 74−86. (18) Kalyvas, H.; Coucouvanis, D. Polyhedron 2007, 26, 4765−4778.

G

DOI: 10.1021/acs.inorgchem.5b02816 Inorg. Chem. XXXX, XXX, XXX−XXX