New Selective Synthesis of Dithiaboroles as a Viable Pathway to

ACS eBooks; C&EN Global Enterprise. A; Accounts of Chemical .... Publication Date (Web): June 1, 2016. Copyright © 2016 American Chemical Society...
0 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/IC

New Selective Synthesis of Dithiaboroles as a Viable Pathway to Functionalized Benzenedithiolenes and Their Complexes Simon H. Schlindwein,† Katharina Bader,‡ Carlo Sibold,† Wolfgang Frey,§ Petr Neugebauer,‡ Milan Orlita,∥ Joris van Slageren,‡ and Dietrich Gudat*,† †

Institute of Inorganic Chemistry, ‡Institute of Physical Chemistry, and §Institute of Organic Chemistry, University of Stuttgart, 70550 Stuttgart, Germany ∥ Laboratoire National des Champs Magnétiques Intenses, 38042 Grenoble, France S Supporting Information *

ABSTRACT: A synthetic protocol to synthesize 2-bromobenzo-1,3,2-dithiaboroles in one step from easily accessible benzene bis(isopropyl thioether)s has been developed. The reaction is remarkably specific in converting substrates with two adjacent iPrS moieties while leaving isolated thioether functions and other functional groups intact. On the basis of the spectroscopic detection or isolation of reaction intermediates, a mechanistic explanation involving a neighborgroup-assisted dealkylation as a key step is proposed. The resulting products featuring one or two dithiaborole units were isolated in good yields and fully characterized. Subsequent methanolysis, which was carried out either as a separate reaction step or in the manner of a one-pot reaction, gave rise to functionally substituted benzenedithiols. The feasibility of a methylphosphoryl-substituted benzenedithiol to act as a dianionic S,S-chelating ligand was demonstrated with the formation of paramagnetic Ni(III) and Co(III) complexes. Selective reduction of the phosphoryl group afforded a rare example of a phosphino dithiol which was shown to act as a monoanionic P,S-bidentate ligand toward Pd(II). All complexes were characterized by spectral data and X-ray diffraction studies, and the paramagnetic ones also by superconducting quantum interference device magnetometry.

■

INTRODUCTION Since they were first published by Schrauzer1 in 1962, dithiolenes (ligands containing the enedithiolate, benzenedithiolate, or dithione units as shown in Chart 1, or intermediate

themselvesin particular of specimens featuring further donor functionalitieshas received much less attention. Still, throughout the years, several dithiolenes containing additional indole,13 thiophene,14 thioether,15 nitrile,16 or phosphine17 moieties have been prepared, and some of their reactivity toward transition metals has been explored. It has to be noted, however, that many syntheses of these functionalized dithiolenes involve tedious multistep processes or usage of highly toxic reagents such as phosgene, and that some species are only accessible in the coordination sphere of metal complexes, while the free ligands are still unavailable.8,17 Benzenedithiolates are usually generated through deprotonation of dithiols, which are frequently prepared by cleavage of alkylated thioether precursors. This deprotection can be accomplished by several protocols. Reductive dealkylation with alkaline metals requires usually harsh reaction conditions18 which may be incompatible with easily reducible functional groups. An alternative approach is the acidic dealkylation of tert-butyl thioethers, which is driven by the high stability of the tertiary carbenium ion intermediate in an SN1 reaction.19,20 In addition, a Lewis acid-induced deprotection protocol employ-

Chart 1

radical anions featuring an intermediate oxidation state) have attracted interest mainly due to their redox activity and their ability to stabilize transition-metal ions in unusual coordination spheres.2−8 Dithiolenes and their complexes have been further intensely studied in the synthesis of free nitrogen-containing radicals,9 as optical materials, in homogeneous electron transfer catalysis, and generally for their distinct electronic structures.2−8 In addition, dithiolenes have been employed as valuable building blocks in the synthesis of various heterocycles such as dithiaphospholes and dithiaphospholanes,10 dithiaboroles,11 and dithiadiboranes.12 In contrast, the synthesis of the ligands © XXXX American Chemical Society

Received: April 2, 2016

A

DOI: 10.1021/acs.inorgchem.6b00821 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry ing BBr3 has been reported.21 A certain disadvantage of all these protocols is that usually all available thioether functionalities in a molecule are cleaved and selective deprotection is unfeasible. In the course of attempts to extend our previous studies on catechol phosphines and their coordination chemistry22 to analogous phosphine-substituted dithiols that might act as building blocks for unprecedented transition-metal complexes, we discovered a reaction which permits the specific dealkylation of benzene 1,2-dithioethers but leaves isolated thioether groups and other functional groups untouched. The key intermediates in this transformation are benzo-1,3,2-dithiaboroles, which can be readily isolated. Although these heterocycles have been shown to be useful reagents in the synthesis of borylated arenes either via cyclooligomerization reactions23 or via direct electrophilic borylations24 and may thus serve as precursors for the widely used arene boronates, they have not received ample interest, perhaps due to their high sensitivity and tedious synthesis. The possibility to access benzenedithiaboroles in one step from benzene dithioethers offers considerable advantages over the conventional synthesis via deprotection of dithioethers to dithiols and subsequent condensation with BX3 or X2BR (X = Cl, Br, I; Scheme 1), and thus facilitates the access to this potentially useful class of compounds.

population (>97:3). One of the phenyl rings in 1 and the cation in [Et4N]10 showed disorder over two positions which was solved by PART and freely refined. Further crystallographic data and details on the structure solution are given in the Supporting Information. CCDC-1455613 (1), CCDC-1455614 (5), CCDC-1455615 ([Et3NH]11), CCDC-1455616 (12), CCDC-1455617 (4), CCDC1455618 (2), CCDC-1455619 ([Et4N]10), CCDC-1455620 (7b), CCDC-1455621 (7d′), CCDC-1455622 (7c), and CCDC-1455623 (9d) contain the supplementary crystallographic data for this paper. These data can be obtained free of charge from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/ cif. Superconducting quantum interference device (SQUID) magnetometric measurements were performed on a Quantum Design MPMS3135 magnetometer. Weighed amounts of the finely ground powder samples were wrapped in Teflon tape and mounted on the sample holder. The experimental data were corrected for diamagnetism by the means of Pascal’s constants, and simulations were carried out with the Matlab toolbox “EasySpin”.28 X-band electron paramagnetic resonance (EPR) spectra were recorded on evacuated powder samples with a Bruker EMX EPR spectrometer where cryogenic temperatures were reached by using an Oxford Instruments continuous flow cryostat. A home-built high-frequency EPR (HFEPR) spectrometer was used for recording HFEPR spectra from pressed pellets of the powdered samples. Experimental EPR data were simulated with “EasySpin”,28 and all g values were determined with an accuracy of Δg = ± 0.002. Far-infrared transmission spectra were recorded at 4.2 K on a Bruker IFS 66v/s FTIR spectrometer. Eicosane pellets containing the studied compound were measured in an 11 T superconducting solenoid. A mercury lamp was used as the radiation source and a composite bolometer as the detector. 3-[(Diphenylphosphoryl)methyl]-1,2-dichlorobenzene (1). 2,3-Dichlorobenzaldehyde (5.0 g, 29 mmol) was dissolved in a degassed mixture of acetic acid (20 mL) and concentrated hydrochloric acid (10 mL). Diphenylphosphine (5.0 mL, 29 mmol) was added. The reaction mixture was heated to 100 °C for 12 h, cooled to rt, and then quenched with cold water (200 mL). The white precipitate was suction filtered and recrystallized from hot EtOH (yield 9.38 g, 91%). 31P{1H} NMR (CDCl3): δ = 29.2 (s). 1H NMR (CDCl3): δ = 3.84 (d, 3JPH = 13.8 Hz, 2 H), 7.03 (dd, 3JHH = 7.9 Hz, 1 H), 7.23 (d, 3JHH = 8.1 Hz, 1 H), 7.32−7.50 (m, 7 H), 7.59−7.69 (m, 4 H). 13C{1H} NMR (CDCl3): δ = 35.6 (d, 1JPC = 66.1 Hz), 127.0 (d, 5 JPC = 2.7 Hz), 128.6 (d, 2JPC = 11.8 Hz), 129.1 (d, 4JPC = 3.0 Hz), 130.1 (d, 3JPC = 4.5 Hz), 131.1 (d, 3JPC = 9.3 Hz), 131.9 (d, 1JPC = 100.1 Hz), 132.1 (d, 4JPC = 2.9 Hz), 132.2 (d, 2JPC = 7.5 Hz), 132.7 (d, 3 JPC = 5.6 Hz), 133.11 (d, 4JPC = 2.7 Hz). Anal. Calcd for C19H15Cl2OP (361.2): C, 63.18; H, 4.19. Found: C, 62.79; H, 4.18. 3-[(Diphenylphosphoryl)methyl]-1,2-bis(isopropylthio)benzene (2). KOtBu (7.76 g, 69 mmol) was dissolved in degassed dimethylacetamide (DMAA) (25 mL). 2-Propanethiol (6.5 mL, 69 mmol) and, after 10 min, 1 (5.00 g, 14 mmol) were added, and the reaction mixture was then heated to 105 °C for 4 days. The mixture was allowed to cool to rt and quenched with cold water (200 mL). The aqueous phase was extracted with EtOAc (3 × 50 mL), and the combined organic phases were washed with first aq Na2CO3, then water, and then brine (each 3 × 50 mL) and finally dried over Na2SO4. All volatiles were evaporated in vacuum, and the residue was recrystallized from hot acetone (yield 5.18 g, 85%). 31P{1H} NMR (CDCl3): δ = 30.8 (s). 1H NMR (CDCl3): δ = 1.09 (d, 3JHH = 6.7 Hz, 6 H), 1.22 (d, 3JHH = 6.6 Hz, 6 H), 3.20 (sept, 3JHH = 6.7 Hz, 1 H), 3.30 (sept, 3JHH = 6.6 Hz, 1 H), 4.17 (d, 3JPH = 14.3 Hz, 2 H), 7.04 (d. 3 JHH = 7.9 Hz, 1 H), 7.13 (dd, 3JHH = 7.6 Hz, 3JHH = 7.8 Hz, 1 H), 7.29−7.49 (m, 7 H), 7.56−7.68 (m, 4 H). 13C{1H} NMR (CDCl3): δ = 22.7 (s), 23.2 (s), 36.3 (s), 36.4 (d, 1JPC = 66.1 Hz), 39.8 (s), 126.2 (d, 4JPC = 2.8 Hz), 127.4 (d, 3JPC = 4.5 Hz), 128.7 (d, 5JPC = 2.5 Hz), 128.4 (d, 3JPC = 11.7 Hz), 131.2 (d, 2JPC = 9.8 Hz), 131.7 (d, 4JPC = 2.8 Hz), 132.3 (d, 1JPC = 99.4 Hz), 133.1 (d, 2JPC = 7.3 Hz), 137.9 (d, 3JPC = 7.1 Hz), 144.9 (d, 4JPC = 2.5 Hz). EI-MS: 440.1 (calcd: 440.14). Anal. Calcd for C25H29OPS2 (440.60): C, 68.15; H, 6.63; S, 14.55. Found: C, 67.96; H, 6.69; S, 14.60.

Scheme 1. Generic Two-Step Synthesis of 2-Halogeno-1,3,2dithiaboroles from Benzene 1,2-Dialkyl Thioethers

Here, we report on the results of our studies on the synthesis of 2-bromobenzo-1,3,2-dithiaboroles and donor-functionalized dithiolenes, and demonstrate the use of the latter as interesting complex ligands.

■

EXPERIMENTAL SECTION

All manipulations were carried out under an atmosphere of dry argon using standard Schlenk line techniques. Solvents were dried by standard procedures and degassed using three freeze−pump−thaw cycles where necessary. Diphenylphosphine25 and isopropyl aryl thioethers26 were prepared by known procedures. All other chemicals were commercially available and used as received. Solution NMR spectra were recorded on a Bruker Avance 250 spectrometer (1H, 250.1 MHz; 13C, 62.8 MHz; 31P, 101.2 MHz; 11B, 80.2 MHz) at 295 K. Elemental analyses (C, H, N, S) were determined on an Elementar Micro Cube analyzer. UV−vis spectra of solutions or solid samples (in diffuse reflectance) were recorded on a J&M model TIDAS spectrometer. Melting points were determined in sealed capillaries on a Büchi B-545 melting point apparatus. Single-crystal X-ray diffraction data were collected on a Bruker AXS Nanostar C diffractometer equipped with a Kappa APEX II Duo charge-coupled device (CCD) detector and a KRYO-FLEX cooling device at 100(2) K for 2, 4, 5, 7b, 7d′, 12, 110(2) K for 1, and 150(2) K for 9d, respectively, using Cu Kα radiation (λ = 1.54178 Å) for 12 and Mo Kα radiation (λ = 0.71073 Å) for the other samples. Crystals were selected under Paratone-N oil, mounted on nylon loops, and immediately placed in a cold stream of N2. The structures were solved by direct methods (SHELXS-97)27 and refined with a full-matrix leastsquares scheme on F2 (SHELXL-2014 and SHELXL-97).27 Semiempirical absorption corrections from equivalents were applied for all structures. Non-hydrogen atoms were refined anisotropically. Complex 2 crystallized as a rotational twin, and the low quality of the data set led to low C−C bond precision. A disorder of the bromine atom in 7c over two positions was noted but not refined due to the highly unequal B

DOI: 10.1021/acs.inorgchem.6b00821 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry 3-[(Diphenylphosphoryl)methyl]benzene-1,2-dithiol (4). An ice-cooled solution of 2 (2.0 g, 4.5 mmol) was dissolved in fluorobenzene (PhF) (20 mL) and treated with BBr3 (1.3 mL, 13.5 mmol). The stirred mixture was then heated to 70 °C for 12 h, allowed to cool to rt, and stored at −20 °C. An off-white precipitate formed which was filtered off and washed with cold PhF (2 × 5 mL) and hexane (3 × 15 mL). The residue was dissolved in MeOH, and all volatiles were removed in vacuum. This procedure was repeated until the oily residue solidified (yield 1.61 g, 72%). 31P{1H} NMR (CDCl3): δ = 48.5 (s). 1H NMR (CDCl3): δ = 3.46 (s, 1 H), 3.89 (s, 1 H), 4.43 (d, 3JPH = 14.1 Hz, 2 H), 7.05 (dd, 3JHH = 7.7 Hz, 8.0 Hz, 1 H), 7.29 (d, 3JHH = 8.0 Hz, 1 H), 7.39 (d, 3JHH = 7.7 Hz, 1 H), 7.49−7.56 (m, 4 H), 7.63−7.72 (m, 6 H). 13C{1H} NMR (CDCl3): δ = 37.7 (d, 1JPC = 67.1 Hz), 126.7 (d, 5JPC = 2.9 Hz), 128.3 (d, 4JPC = 3.3 Hz), 128.4 (d, 3 JPC = 4.7 Hz), 128.6 (d, 3JPC = 11.8 Hz), 129.8 (d, 3JPC = 6.2 Hz), 131.2 (d, 2JPC = 9.3 Hz), 131.8 (d, 1JPC = 99.3 Hz), 132.1 (d, 4JPC = 2.7 Hz), 134.4 (d, 2JPC = 8.6 Hz), 136.5 (d, 4JPC = 2.8 Hz). Anal. Calcd for C19H17OPS2 (356.44): C, 64.02; H, 4.81; S, 17.99. Found: C, 63.42; H, 4.86; S, 17.52. 3-[(Diphenylphosphino)methyl]-1,2-mercaptobenzene (5). A stirred solution of 3 (1.0 g, 2.8 mmol) in methyl tert-butyl ether (MTBE) (80 mL) was treated with LiAlH4 (320 mg, 8.4 mmol). The mixture was refluxed for 1 day, quenched with 1 M degassed hydrochloric acid, and extracted with MTBE (2 × 50 mL) under strict exclusion of air. The combined organic phases were washed with water and brine (2 × 30 mL each) and the volatiles evaporated without further purification. The product was obtained as slightly pink crystals (yield 770 mg, 81%). 31P{1H} NMR (C6D6): δ = −16.2 (s). 1H NMR ((C6D6): δ = 3.26 (s, 2 H), 3.70 (br s, 2 H), 6.17−6.29 (m, 2 H), 6.58−6.65 (m, 1 H), 6.76−6.85 (m, 6 H), 7.05−7.14 (m, 4 H). 13 C{1H} (C6D6): δ = 36.4 (d, 1JPC = 16.3 Hz), 125.8 (d, 3JPC = 1.8 Hz) 127.8 (d, 4JPC = 1.3 Hz), 128.0 (s), 128.3 (d, 3JPC = 6.5 Hz), 128.6 (s), 133.1 (d, 2JPC = 18.1 Hz), 133.1 (s) 137.8 (d, 1JPC = 16.2 Hz), 138.8 (s), 139.0 (s). Anal. Calcd for C19H17PS2 (340.45): C, 67.03; H, 5.03; S, 18.83. Found: C, 66.83; H, 5.34; S, 18.38. Generic Procedure for the Synthesis of Benzo-1,3,2dithiaboroles. The appropriate thioether (1 mmol) was dissolved in PhF (15 mL) and BBr3 added (1 equiv per functional group, 1−4 mmol) under ice cooling. The mixture was then stirred for 4−12 h at 70 °C, allowed to cool to rt, and stored overnight at −20 °C. (2-Bromo-4-(isopropylthio)benzo[d]-1,3,2-dithiaborole Tribromoborane Adduct (7b). The reaction mixture was evaporated to dryness and the residue suspended in hexane. The insoluble fraction was filtered off and the filtrate stored at −20 °C. Colorless blocks precipitated which were collected by careful decantation, washed with hexane, and dried under vacuum (yield 195 mg, 35%). 11B{1H} NMR (CDCl3): δ = 2.2 (s), 55.4 (s). 1H NMR (CDCl3): δ = 1.40 (d, 3JHH = 6.7 Hz, 6 H), 3.72 (sept, 3JHH = 6.7 Hz, 1 H), 7.36 (t, 3JHH = 7.8 Hz, 1 H), 7.71 (d, 3JHH = 7.9 Hz, 1 H), 7.72 (d, 3JHH = 7.8 Hz, 1 H). 13 C{1H} (CDCl3): δ = 22.3 (s), 43.4 (s), 126.5 (s), 126.8 (s), 129.5 (s), 141.8 (s), 146.5 (s). Anal. Calcd for C9H10B2Br4S3 (555.61): C, 19.46; H, 1.81; S, 17.31. Found: C, 19.34; H, 2.31; S, 17.41. 2,6-Dibromobenzo[1,2-d:4,5-d′]bis(1,3,2-dithiaborole) (7c). Thin colorless needles precipitated which were filtered off, washed with hexane, and recrystallized from warm PhF (yield 215 mg, 56%). 11 1 B{ H} NMR (CDCl3): δ = 52.6 (s). 1H NMR (CDCl3): δ = 7.97 (s, 2 H). 13C{1H} (CDCl33): δ = 123.0 (s), 139.3 (s). Anal. Calcd for C6H2B2Br2S4 (383.77): C, 18.78; H, 0.53; S, 33.42. Found: C, 18.89; H, 0.95; S, 33.63. 2,7-Dibromobenzo[1,2-d:3,4-d′]bis([1,3,2]dithiaborole) (7d). Thin colorless needles precipitated which were filtered off and washed with hexane. The mother liquor was treated with hexane and again stored at −20 °C to give a second crop of crystalline material which was again collected by filtration and washed with hexane. The combined precipitates were washed once more with hexane and dried under vacuum (yield 184 mg, 48%). 11B{1H} NMR (CDCl3): δ = 52.3 (s). 1H NMR (CDCl3): δ = 7.58 (s, 2 H). 13C{1H} (CDCl33): δ = 123.7 (s), 139.2 (s). Anal. Calcd for C6H2B2Br2S4 (383.77): C, 18.78; H, 0.53; S, 33.42. Found: C, 18.82; H, 0.70; S, 33.15.

(2-Bromo-4,7-bis(isopropylthio)benzo[d]-1,3,2-dithiaborole Bis(tribromoborane) Adduct (7d′). The mother liquor obtained after separation of 7d was once more stored at −20 °C. After 1 day colorless blocks precipitated which were collected by careful decantation, washed with hexane, and dried under vacuum (yield 5 mg, 0.5%). 11B{1H} NMR (CDCl3): δ = −6.2 (s), 52.3 (s). 1H NMR (CDCl3): δ = 1.31 (d, 3JHH = 6.8 Hz, 12 H), 3.55 (sept, 3JHH = 6.8 Hz, 2 H), 7.45 (s, 2 H). 13C{1H} (CDCl3): δ = 22.8, 41.1, 129.4, 131.7, 145.6. The amount of product was too small to obtain an elemental analysis. 2-Bromo-4-[(diphenylphosphoryl)methyl]benzo[d]-1,3,2-dithiaborole Tribromoborane Adduct (9d). The off-white precipitate was collected by filtration, washed with cold PhF and hexane, and dried under vacuum (yield 473 mg, 68%). 31P{1H} NMR (CDCl3): δ = 53.9 (q). 11B{1H} NMR (CDCl3): δ = −12.2 (s), 51.6 (s). 1H NMR (CDCl3): δ = 4.60 (d, 3JPH = 14.4 Hz, 2 H), 6.99 (m, 1 H), 7.07 (m, 1 H) 7.31 (m, 1 H), 7.40−7.52 (m, 4 H), 7.58−7.72 (m, 6 H). 13C{1H} (CDCl3): δ = 33.8 (d, 2JPC = 67.5 Hz), 124.7 (s), 129.36 (s), 133.38 (s), 135.4 (s), 142.0 (s). Anal. Calcd for C19H15B2Br4OPS2 (695.66): C, 32.80; H, 2.17; S, 9.22. Found: C, 33.76; H, 2.29; S, 9.00. Tetraethylammonium Bis(3-[(diphenylphosphoryl)methyl]benzene-1,2-dithiolato)cobaltate(III) ([Et4N]10). A solution of 4 (200 mg, 0.56 mmol) in degassed EtOH (20 mL) was treated with NEt3 (0.16 mL, 1.12 mmol) and NEt4Cl (1.86 g, 11.2 mmol). Addition of [(Co(H2O)6]Cl2 produced a yellow-brown precipitate. The mixture was stirred for 4 h, after which it was exposed to air. The yellow precipitate immediately turned deep blue. After the mixture was stirred overnight, the precipitate was collected by filtration, washed with EtOH, water, and MTBE (3 × 15 mL each), and dried in vacuum (yield 186 mg, 74% based on Co). 31P{1H} NMR (CD2Cl2, 295 K): δ = 99.7 (br s). 1H NMR (CD2Cl2, 295 K): δ = −33.7 (2 H), −25.7 (2 H), −21.6 (2 H), −4.5 (8 H), 1.7 (12 H), 7.0 (12 H), 9.7 (8 H), 21.3 (4 H). The paramagnetic nature of this compound prevented 13C NMR data from being obtained. Anal. Calcd for C46H50CoNO2P2S4 (898.03): C, 61.52; H, 5.61; N, 1.56; S, 14.28. Found: C, 61.09; H, 5.67; N, 1.68; S, 14.29. Triethylammonium Bis(3-[(diphenylphosphoryl)methyl]benzene-1,2-dithiolato)niccolate ([Et3NH]11). A solution of 4 (200 mg, 0.56 mmol) in degassed EtOH (20 mL) was treated with NEt3 (0.16 mL, 1.12 mmol) and HNEt3Cl (1.52 g, 11.2 mmol). Addition of [(Ni(H2O)6]Cl2 (133 mg, 0.56 mmol) produced a deep purple solution. The mixture was stirred for 4 h, after which it was exposed to air. The deep purple solution immediately turned deep green, and a precipitate of the same color appeared. The slurry was stirred overnight. The precipitate was then collected by filtration, washed with EtOH, water, and MTBE (each 3 × 15 mL), and dried in vacuum (yield 204 mg, 78% based on Ni). Due to the paramagnetic nature of this product, no interpretable NMR data could be obtained. (−)-ESI-MS: 765.96 [C 38 H 30 NiNO 2 P 2 S 4 − ]. Anal. Calcd for C44H46NiNO2P2S4 (869.74): C, 60.76; H, 5.33; N, 1.61; S, 14.46. Found: C, 60.68; H, 5.42; N, 1.55; S, 14.71. Bis(2-[(diphenylphosphino)methyl]-6mercaptobenzenethiolato)palladium(II) (12). A mixture of Pd(acac)2 (45 mg, 0.15 mmol) and 5 (100 mg, 0.30 mmol) in THF (10 mL) was placed in an ultrasound bath. A yellow precipitate began to form. The mixture was then stirred for 12 h and filtered. The precipitate was washed with THF (3 × 10 mL) and dried in vacuum (yield 104 mg, 89% based on Pd). 31P{1H} NMR (DMF-d7): δ = 65.4 (s). − 1H NMR (DMF-d7): δ = 3.65 (m, 4 H), 5.54 (s, 2 H), 6.89 (t, 3 JHH = 7.6 Hz, 2 H), 7.07 (d, 3JHH = 7.6 Hz, 2 H), 7.34 (d, 3JHH = 7.6 Hz, 2 H), 7.67−7.80 (m, 6 H), 8.07−8.18 (m, 4 H). − 13C{1H} NMR (DMF-d7): δ = 35.0, 125.9, 128.8, 131.2, 134.0. (+)-ESI-MS: 806.98 [M + Na]. Anal. Calcd for C9H10S3B2Br4 (785.28): C, 58.12; H, 4.11; S, 16.33. Found: C, 58.13; H, 4.16; S, 15.98. Methanolysis of the Dithiaboroles 7a−d. An excess of degassed MeOH was added to the corresponding isolated dithiaborole at 0 °C with stirring.29 After the violent reaction stopped, the mixture was allowed to warm to rt and stirred for 30 min, after which all volatiles were removed in vacuum. This procedure was repeated three times, and the resulting crude product was then dried in vacuum. C

DOI: 10.1021/acs.inorgchem.6b00821 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

■

RESULTS AND DISCUSSION Synthesis and Characterization of Functionalized Benzene-1,2-dithiols and Benzo-1,3,2-dithiaboroles. The synthesis of our original target, (phosphinomethyl)benzenedithiol 5, was accomplished as outlined in Scheme 2.

The smooth dealkylation of 2 with BBr3 is somewhat unexpected since it has been reported that thioethers featuring alkyl substituents other than tBu or benzyl show much inferior reactivities.19,20 To establish a reaction mechanism and clarify whether the reactivity of 2 is possibly enhanced by a neighbor group effect, we repeated the reaction with simpler starting materials (Scheme 3).

Scheme 2. Synthesis of (Phosphinomethyl)benzenedithiol 5

Scheme 3. Deprotection of Benzene Thioethers with BBr3a

The key intermediate, bisthioether 2, was prepared in two steps starting from 2,3-dichlorobenzaldehyde. Hydrophosphination of the aldehyde was carried out using a slight modification of our previously reported protocol,30 and the resulting phosphine oxide 1 was converted into 2 by nucleophilic aromatic substitution with KSiPr. Both 1 and 2 were readily isolated after workup and recrystallization and characterized by NMR, elemental analysis, and X-ray diffraction (XRD). As anticipated, the deprotection of the thioether functions in 2 turned out to be the most critical step. Reactions with sodium or strong Lewis acids, which are commonly employed for this purpose,18,31 were unspecific. NMR spectroscopic reaction monitoring gave evidence for the occurrence of P−C cleavage and PO reduction processes, which confirmed that the phosphine oxide moiety is indeed not sufficiently stable to survive the harsh conditions required for the S-dealkylation. Searching for a milder deprotection agent, we found eventually that selective cleavage of the SiPr groups is feasible with BBr3. Using this reagent, dithiol 4 was obtained in good yield after methanolysis and workup, and fully characterized by NMR, elemental analysis, and XRD. Reaction monitoring by 11 B{1H} NMR allowed us to detect the signal of an intermediate with a chemical shift of 51.6 ppm that is characteristic for a bromodithiaborole.23 The product precipitated upon cooling of the reaction mixture to −20 °C and was isolated as an off-white powder. The 11B NMR spectrum of this product shows, besides the signal of the dithiaborole unit, an additional signal with a chemical shift (δ(11B) −12.2 ppm) that is typical for a BBr3 adduct. These findings led to a structural assignment as 3, which was confirmed by further spectral data and an elemental analysis. Finally, we reduced the phosphine oxide 4 with LiAlH4 in refluxing MTBE. Workup and recrystallization from hexane afforded the desired phosphine 5 in acceptable to good overall yield (four steps, 45%). The product was fully characterized and represents the first example of a “free” phosphine-substituted dithiolene precursor.

a Reagents and conditions: 1 equiv of BBr3 per SiPr group, PhF, 12 h, 70 °C.

Reaction of 1,2-bis(isopropylthio)benzene (6a) with 1 equiv of BBr3 in fluorobenzene at rt proceeded slowly under release of isopropyl bromide, which was identified by 1H NMR spectroscopy. Further NMR spectroscopic monitoring indicated that the reaction stopped eventually after consumption of one thioether group (and thus release of 1 equiv of iPrBr; see the Supporting Information). Careful analysis of 1H and 11B NMR spectra indicated the presence of a further boroncontaining product featuring a single iPrS unit with two diasterotopic CH3 groups and an unsymmetrically substituted aromatic ring, which we assign as an intramolecularly donorstabilized boranylthiolate, 9a (Scheme 4). Conversion into the 1,3,2-dithiabromoborole 7a commenced at rt but needed heating to 70 °C to go to completion. The product 7a was not isolated but unambiguously identified by comparison of the observed NMR data with previously published literature values.32 A notable acceleration of the reaction rate was observed when an excess of BBr3 was employed or the reaction mixture heated from the beginning. The signal of unreacted (or excess) BBr3 is not stationary under these conditions but migrates from an initial chemical shift of 1 ppm to a final value of ∼34 ppm expected for pure BBr3 (this signal is only visible when an excess of BBr3 is employed; for spectra see the Supporting Information). These findings are readily explained if one assumes that BBr3 and 6a exist in rapid dynamic exchange with a labile Lewis adduct, 8a (Scheme 4). Having established that the reaction proceeded successfully with the simplest benzene dithioether, we were interested to learn what happens if the substrate contains an uneven number of iPrS groups. Monitoring the reaction of 1,2,3-tris(isopropylthio)benzene (6b) with 3 equiv of BBr3 (1 equiv D

DOI: 10.1021/acs.inorgchem.6b00821 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

benzene and 1,3-bis(isopropylthio)benzene, which exhibit only isolated thioether groups, do not undergo any BBr3-induced dealkylation under similar reaction conditions. The involvement of intramolecularly donor-stabilized thiolatodibromoboranes was positively proven by the serendipitous isolation of a small amount of a crystalline intermediate, 9d (Chart 2), in the reaction of 6d with BBr3. The colorless plates separated when an unquenched reaction mixture was stored at −5 °C and were characterized by XRD (Figure 1).

Scheme 4. Postulated Reaction Mechanism for the Deprotection of Bis(isopropylthio)benzenes Using 5a as a Generic Example

Chart 2

for each thioether) in fluorobenzene at 70 °C disclosed again the presence of the characteristic 11B NMR signal of a dithiaborole. After isolating the product, we were surprised to find that the 11B NMR spectrum shows the signal of another boron atom supposedly belonging to a donor-bound BBr3 unit. Since the 1H NMR spectrum suggested that one thioether group had remained intact, we assigned the constitution of the product as 7b. This assignment was further confirmed by elemental analysis and an XRD study of the product. Even after extended reaction times (72 h instead of 12 h), no deprotection of the remaining thioether group was observed. 1,2,4,5-Tetrakis(isopropylthio)benzene (6c) reacted analogously to 6a and, after workup, gave rise to crystalline bis(bromodithiaborole) 7c, which was fully characterized. 1,2,3,4-Tetrakis(isopropylthio)benzene (6d) afforded, according to the 1H and 11B NMR spectra of the reaction solution, a mixture of two products which were subsequently identified as 7d (major product) and 7d′ (minor product). Both species were isolated by fractional crystallization and were characterized by NMR data and elemental analysis (7d) or NMR data and XRD (7d′). On the basis of the experimental observations made during the individual reactions, we propose a tentative reaction mechanism (Scheme 4). Addition of BBr 3 to 1,2-bis(isopropylthio)benzenes induces immediate formation of a dynamically equilibrating mixture of the starting materials and a Lewis adduct, e.g., 8a. Cleavage of the first S−C bond is envisaged to proceed via a transient donor-stabilized borenium intermediate, A, which is formed by intramolecular nucleophilic attack of the second SiPr group at the boron atom and extrusion of a bromide, and decays further through attack of the anion at the carbon atom of one of the SiPr moieties to give 9a. Finally, the dithiaborole is formed by thermally induced cleavage of the second equivalent of iPrBr. The accelerating effect of excess BBr3 is attributable to a Lewis acid-induced catalysis of the final reaction step. Furthermore, the failure of 7b,d′ to undergo cleavage of the remaining isolated thioether group indicates that the cooperative action of a neighboring Lewis donor (which is considered to facilitate formation of the ionic intermediate A) is presumably a crucial prerequisite for a smooth reaction. This hypothesis is corroborated by the finding that (isopropylthio)-

Figure 1. Molecular structure of 9d. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atoms were omitted for clarity.

The crystallographic data of 7b,c,d′ (see Figure 2 and Table 1) extend the sparse data basis on structures of sulfanylboranes or halogenodithiaboroles found in the literature.33 The fused ring systems in all compounds show the expected planar arrangements and the boron and sulfur atoms in the S-bound BBr3 units the expected distorted tetrahedral and pyramidal geometries, respectively. The Br−B and S−B bond lengths are within the expected values for dithiaboroles (Table 1). The planar benzodithiaborole rings in all compounds are arranged in layers and exhibit significant intermolecular S−Br or Br−Br interactions between layers (see the Supporting Information). The intermolecular Br−S contacts in 7c (3.649 Å) and 7d′ (3.625 Å) are slightly shorter and the Br−Br contact in 7b (3.588 Å) is considerably shorter than the sums of van der Waals radii according to Rowland (Br···S, 3.68 Å; Br···Br, 3.74 Å)34 and fit well into the description of halogenbonding interactions.35 The structural parameters of the ligands 2, 3, and 5 (see the Supporting Information) show no peculiarities and match those of the oxygen analogues prepared by Chikkali.30 Hydrolysis of the isolated dithiaboroles 3 and 7a−d with MeOH and evaporation of the volatiles in vacuum afford very pure thiols. In the case of 7a−d, which lack a phosphine oxide moiety, the reaction mixture can also be directly quenched with MeOH in a violent reaction to give a product that can be used for further reactions without purification. E

DOI: 10.1021/acs.inorgchem.6b00821 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 2. Molecular structures of 7b, 7c, and 7d′ (from left to right). Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atoms were omitted for clarity. Selected bond distances are given in Table 1.

Table 1. Selected Distances (Å) for 7b, 7c, 7d′, and 9d S(1)−B(1) B(1)−Br(1) S(3)−B(2) B(2)−Br(2) S(4)−B(2)

7b

7c

7d′

9d

1.776(9) 1.911(8) 1.972(6) 1.984(6)

1.788(9) 1.905(8)

1.787(4) 1.902(4) 1.984(5) 1.988(4)

1.787(3) 1.910(3) 1.895(3) 2.015(3) 1.969(3)

Metal Complexation Studies. Reactions of 4 and 5 with selected transition-metal salts were studied to demonstrate the complexation behavior of these donor-functionalized dithiols. Co and Ni Complexes of 4. Reaction of 4 with [M(H2O)6]Cl2 (M = Co, Ni) followed by subsequent air oxidation gave complexes [Et4N]10 and [Et3NH]11 (Chart 3), Chart 3 Figure 3. Molecular structure of the complex anion 10. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atoms and the counterion (NEt4+) were omitted for clarity. Selected bond distances (Å) and angles (deg): Co(1)−S(1), 2.1583(8); Co(1)−S(2), 2.1625(8); S(2)−C(5), 1.757(3); S(1)−C(4), 1.762(3); P(1)−O(1), 1.491(2); S(1)−Co(1)−S(2), 92.14(3); S(2)−Co(1)−S(1)*, 87.86(3).

The UV−vis spectra of [Et4N]10 and [Et3NH]11 (Figure S5) show the characteristic low-energy absorption bands that had previously been observed for other monoanions of the type [M(dithiolene)2]− (M = Co, Ni) and had been assigned as a combination of metal-to-ligand charge transfer (MLCT) and d−d transitions (M = Co) or an interligand charge transfer (LLCT) transition (M = Ni), respectively.36,37,39 The 1H and 31 1 P{ H} NMR spectra (Figures S7 and S8) of [Et4N]10 are characterized by the presence of strong, temperature-dependent paramagnetically induced chemical shifts (which are largest for the protons in the thiolene ring and the adjacent CH2 groups featuring the shortest distances to the metal atom and vary linearly with 1/T) which had previously been observed for similar Co(III) complexes.36 Apart from the temperaturedependent variation of the chemical shifts, below 273 K a broadening and eventually decoalescence of individual signals are observed. A similar behavior has previously been reported for [Co(tdt)2]− 36 (tdt = toluenedithiol) and reflects presumably the presence of a reversible interconversion between isomers with cis- and trans-alignment of the pendant phosphane oxide units.

which were isolated as blue or green single crystals, respectively, after recrystallization from EtOH. Characterization by analytical and spectral data and single-crystal XRD studies (see Figure 3 and Table S3) revealed that the spectral data and structural features are closely similar to those of previously reported paramagnetic bis(benzodithiolato) complexes with the same metal ions.36,37 The metal atoms in both complexes feature the expected36,37 square planar coordination spheres with a trans-arrangement of the peripheral phosphoryl groups. The PO units point away from each other and display neither intra- nor intermolecular interactions with the metal ions, but one phosphine oxide unit in 11 exhibits a moderately strong38 hydrogen bond to the proton of the triethylammonium cation (O···N, 269.4(6) Å; O···H, 1.80(5) Å; N−H···O, 155(5)°). F

DOI: 10.1021/acs.inorgchem.6b00821 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry To obtain deeper insight into the magnetic behavior, both complexes were studied in detail with SQUID magnetometry and X-band EPR, HFEPR, and far-infrared spectroscopy. The field- and temperature dependent magnetic behavior of Ni complex [Et3NH]11 was investigated by direct current SQUID (DC-SQUID) magnetometry. The magnetization as a function of the applied magnetic field (Figure S11) can be fitted nicely with Brillouin functions (S ≈ 1/2, giso = 2.077, T corresponding to measurement). The χT(T) curve (Figure S12) converges at high temperatures to a value of χT = 0.480 cm3 mol−1 K−1, while at low temperatures a decrease of χT to approximately 0.320 cm3 mol−1 K−1 is observed. The high-temperature limit corresponds to the expected Curie value for S = 1/2 and giso = 2.26. This g value is substantially higher than that found from the low-temperature magnetization curves. In addition, the temperature dependence of χT is also surprising. It is unlikely that it is due to antiferromagnetic exchange coupling, because no obvious exchange paths are identifiable in the crystal structure. Furthermore, the low-temperature χT value does not tend toward zero as would be expected for an antiferromagnetic coupling. However, the presence of a temperature dependent equilibrium between two species with different g values could be responsible for the decrease in χT. To substantiate the presence of multiple species, X-band EPR spectra of [NEt3H]11 were recorded at 9.47 GHz and 5− 150 K (Figures S13 and S14). The spectrum recorded at 5 K contains three features with comparable intensities and an additional fourth peak with lower intensity. This observation substantiates the assumption that more than one paramagnetic species is present in the sample. The data can be fit as superposition of the lines of two S = 1/2 systems featuring rhombic g tensors with similar values of the principal components and relative abundances of 1:0.5 (“system 1” and “system 2”; for the parameters see Table 2). The relative

Figure 4. HFEPR spectra recorded from a pressed powder pellet of [Et3NH]11 at 5 K and various frequencies (solid lines) and results of simulations using the parameters listed in Table 2 (dashed lines).

frequencies. The spectra can be fitted with two rhombic S = 1/2 spin systems (Table 2), confirming the parameter set determined by X-band EPR spectroscopy. Temperature-dependent HFEPR measurements at 300 GHz were recorded from 2 up to 300 K (Figure S15). While the resonance lines of the two spin systems remain visible over the whole temperature range, their relative intensities change, with the lines of spin system 2 gaining in relative intensity at higher temperatures. This temperature dependence confirms that a dynamic isomerization process connects the two species. The failure to observe a similar temperature dependence in the Xband EPR spectra (vide supra) is attributed to the inferior g resolution at lower frequencies, which causes the resonance lines of different species to overlap. In summary, the results of SQUID magnetometry and Xband EPR and HFEPR spectroscopy all indicate the presence of two paramagnetic species in [Et3NH]11. The average g values obtained from the EPR measurements are consistent (X-band EPR, 2.078; HFEPR, 2.080), while SQUID magnetometry suggests a species with gav significantly larger than 2.077 being present at higher temperatures. At the moment, we cannot explain this discrepancy. DC-SQUID magnetometry was used to investigate the fieldand temperature-dependent behavior of [Et4N]10. The reduced magnetization plot (Figure 5, left panel) shows that the data curves obtained at different temperatures are nonsuperimposable, which indicates the presence of a zerofield splitting. The magnetization versus magnetic field measurements at 20 K can be perfectly simulated with S = 1, g = 2.27, and D = 44 cm−1 (Figure S16), while toward lower temperatures an increasing mismatch between experimental and simulated data is observed. A subtraction of the simulated from the measured data results in a Brillouin-shaped curve (Figure S17) which indicates a paramagnetic impurity is present in the sample. The room temperature χT value of 1.38 cm3 K−1 mol−1 decreases to 0.14 cm3 K−1 mol−1 at 1.8 K. The nonzero slope of the high-temperature region suggests the presence of a temperature-independent paramagnetic contribution. The

Table 2. Simulation Parameters for EPR Spectra of [Et3NH] 11 system 1

system 2

9.47 GHz

100−410 GHz

weight 1 gx = 2.005 gy = 2.046 gz = 2.187 weight 0.5 gx = 2.005 gy = 2.081 gz = 2.187

weight 1 gx = 2.005 gy = 2.044 gz = 2.185 weight 0.1 gx = 2.014 gy = 2.041 gz = 2.185

intensities of the two spin systems do not change significantly between 5 and 150 K (but see the comment further below). The extracted g values are comparable to literature data for square planar [Ni(benzenedithiolene)2]− complexes37,40 which have been described as containing delocalized class III mixedvalence ligand radicals bound to low-spin d8 Ni(II) central ions. The two spin systems in 11 could originate from the presence of stereoisomers with cis/trans-orientation of the unsymmetrical ligands or different conformational alignment of the dipolar phosphoryl moieties. The observed temperature dependence would favor the latter. For further investigation of the sample composition, HFEPR spectra were measured from pressed powder pellets of [Et3NH] 11 at 100, 300, and 410 GHz at 5 K (Figure 4). All spectra show two spin systems, with increasing resolution for higher G

DOI: 10.1021/acs.inorgchem.6b00821 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 5. DC-SQUID magnetometry results for [Et4N]10 as a Teflonwrapped powder pellet, showing the field dependence of the magnetization at different temperatures (left) and the temperature dependence of χT (right; open squares denote experimental data, and the solid line denotes the result of simulations using the parameters given in the text).

Figure 6. Molecular structure of 12. Thermal ellipsoids are drawn at the 50% probability level. C-bound hydrogen atoms and a solvent molecule (DMF) were omitted for clarity. Selected bond distances (Å) and angles (deg): Pd(1)−P(1), 2.2930(11); Pd(1)−S(4), 2.3255(11); S(4)−C(4), 1.768(5); S(5)−C(5), 1.773(5); P(1)−Pd(1)−S(4), 91.73(4); P(1)−Pd(1)−S(4)*, 88.27(4).

curve could be fitted satisfactorily by using the same parameters as for the magnetization curve and adding the 5 × 10−4 cm3 K−1 mol−1 temperature-independent paramagnetic contribution. HFEPR spectroscopy of complex [Et4N]10 was unsuccessful. No signal was detected at 5 K for frequencies between 380 and 1100 GHz, indicating either a negative D value or D > 37 cm−1. Far-infrared spectra (Figure S18) which were recorded to further study the zero-field splitting displayed three signals at 33, 43, and 57 cm−1. As supposed for [NEt3H]11, here also isomerism could lead to multiple species, giving rise to different zero-field splitting values as observed. A refined simulation of the χT(T) curve including the three species with D = 33, 43, and 57 cm−1 found in the far-IR gave no significant improvement over a simulation using a single component with D = 44 cm−1 (Figure S19). Pd Complex of 5. Even if the phosphane oxide moiety in 3 can in principle serve as an additional donor, it behaved in the reaction with cobalt and nickel salts as a pure spectator group. Anticipating that a phosphane unit might perform differently, we turned to study the complexation behavior of 5. Reaction with (cod)PdCl2 in the presence of NEt3 gave spectroscopic evidence for the P-complexation but was unselective, which is not surprising if one considers that phosphanes and thiolates are both known as excellent ligands toward Pd(II). However, we were able to solve the selectivity problem by careful selection of the transition-metal precursor, and succeeded to obtain complex 12 in a selective reaction between Pd(acac)2 and 2 equiv of 5. The product was isolated in good yield as a yellow microcrystalline powder which exhibits low solubility in all common solvents but could be recrystallized from hot DMF. The results of an XRD study revealed the presence of centrosymmetric complex 12 with square planar coordination at palladium and a trans-arrangement of the two bidentate P,Sbound ligands (Figure 6). The stereochemistry deviates from that of the isosteric complex 13 (Chart 3) where the P,O-bound catechol phosphane ligands exhibit a cis-orientation.22 The Pd−P distance in 12 (2.293(1) Å) is slightly longer than in 10 (2.2441(5) Å),22 while the remaining metric parameters show no peculiarities. We interpret the fact that in solution only a single stereoisomer is detectable by 31P{1H} NMR spectrosco-

py as confirmation that, as in 13,22 the geometry observed in the solid state is also conserved in solution.

■

CONCLUSIONS

A synthetic protocol giving access to dithiabromoboroles in a simple one-step reaction from benzene 1,2-diisopropyl thioethers with BBr3 has been established. The reaction avoids many of the problems (unselective and lengthy reactions) associated with the formation and cleavage of tert-butyl thioethers, and displays a remarkable regioselectivity as it leaves isolated thioether functionalities untouched and can thus be used to access functional target molecules featuring additional sulfanyl or phosphoryl substituents, respectively. It was shown that the reaction proceeds via a disulfanylborate intermediate and is significantly accelerated by employing an excess of BBr3 and elevated temperatures. The heterocycles solvolyze easily to benzenedithiols which can be further postfunctionalized. This was demonstrated by reduction of a pendant phosphane oxide unit to give a phosphanyl-decorated benzenedithiol. Undertaking first steps into the exploration of the coordination properties of this ligand, we found that the phosphanyl unit can indeed act as an additional donor site in supporting a P,S-bidentate coordination mode of the ligand. It is envisaged that the ligand may also support other binding modes, including a tridentate P,S,S-coordination to two metals, which mirrors the known behavior of the analogous catechol phosphane ligands. On the contrary, no participation of the additional potential donor site was observed so far for the corresponding phosphanoylbenzenedithiol which was found to act as an S,S-bidentate ligand to form complexes [M(dithiolene)2]− (M = Co, Ni). The square planar metal coordination geometries and the optical properties match those of previously known complexes of the same type, but EPR and magnetometry studies indicate that the solid phases contain mixtures of paramagnetic species which can be possibly addressed as isomers. H

DOI: 10.1021/acs.inorgchem.6b00821 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

■

(15) Larsen, J.; Bechgaard, K. J. Org. Chem. 1987, 52, 3285−3288. (16) Simaõ, D.; Alves, H.; Belo, D.; Rabac, S.; Branco Lopes, E.; Cordeiro Santos, I.; Gama, V.; Duarte, M. T.; Henriques, R. T.; Novais, H.; Almeida, M. Eur. J. Org. Chem. 2001, 3119−3126. (17) Adams, H.; Morris, M. J.; Riddiough, A. E.; Yellowlees, L. J.; Lever, A. B. P. Inorg. Chem. 2007, 46, 9790−9807. (18) Hahn, F. E.; Kreickmann, T.; Pape, T. Eur. J. Inorg. Chem. 2006, 2006, 535−539. (19) Schmidt, M.; Rittig, F. R. Z. Anorg. Allg. Chem. 1972, 394, 152− 156. (20) Lappert, M. F. J. Chem. Soc. 1953, 0, 2784−2788. (21) Rowlands, G. J.; Seacome, R. J. Beilstein J. Org. Chem. 2009, 5, 9. (22) (a) Bauer, G.; Nieger, M.; Gudat, D. Dalton Trans. 2014, 43, 8911−8920. (b) Bauer, G.; Englert, C.; Nieger, M.; Gudat, D. Inorg. Chim. Acta 2011, 374, 240−246. (c) Chikkali, S. H.; Nieger, M.; Gudat, D. New J. Chem. 2010, 34, 1348−1354. (23) Goswami, A.; Pritzkow, H.; Rominger, F.; Siebert, W. Eur. J. Inorg. Chem. 2004, 2004, 4223−4231. (24) Solomon, S. A.; Del Grosso, A.; Clark, E. R.; Bagutski, V.; McDouall, J. J. W.; Ingleson, M. J. Organometallics 2012, 31, 1908− 1916. (25) Wittenberg, D.; Gilman, H. J. Org. Chem. 1958, 23, 1063−1065. (26) Testaferri, L.; Tiecco, M.; Tingoli, M.; Chianelli, D.; Montanucci, M. Synthesis 1983, 1983 (9), 751−755. (27) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr. 2008, 64, 112−122. (28) Stoll, S.; Schweiger, A. J. Magn. Reson. 2006, 178, 42−55. (29) Alternatively, also the crude products obtained after evaporation of the reaction mixtures could be used, allowing significantly improved yields of dithiols to be reached while causing only very minor (if any) spectroscopically detectable increases in the impurity level. (30) Chikkali, S.; Gudat, D. Eur. J. Inorg. Chem. 2006, 2006, 3005− 3009. (31) Umemoto, T.; Garrick, L. M.; Saito, N. Beilstein J. Org. Chem. 2012, 8, 461−471. (32) Spectral data were obtained from Wiley Subscription Services, Inc., Hoboken, NJ. (33) A search of the Cambridge Crystallographic Database for structures of the types R2SBX3 and (RS)2BX on March 11, 2015, revealed nine and seven entries, respectively (X = any halogen). (34) Rowland, R. S.; Taylor, R. J. Phys. Chem. 1996, 100, 7384−7391. (35) (a) Metrangolo, P.; Resnati, G. Chem. - Eur. J. 2001, 7, 2511− 2519. (b) Politzer, P.; Lane, P.; Concha, M. C.; Ma, Y.; Murray, J. S. J. Mol. Model. 2007, 13, 305−311. (c) Metrangolo, P.; Meyer, F.; Pilati, T.; Resnati, G.; Terraneo, G. Angew. Chem., Int. Ed. 2008, 47, 6114− 6127. (36) (a) van der Put, P. J.; Schilperoord, A. A. Inorg. Chem. 1974, 13, 2476−2481. (b) Ray, K.; Begum, A.; Weyhermüller, T.; Piligkos, S.; van Slageren, J.; Neese, F.; Wieghardt, K. J. Am. Chem. Soc. 2005, 127, 4403−4415. (c) Benedito, F. L.; Petrenko, T.; Bill, E.; Weyhermüller, T.; Wieghardt, K. Inorg. Chem. 2009, 48, 10913−10925. (37) (a) Sellmann, D.; Binder, H.; Häußinger, D.; Heinemann, F. W.; Sutter, J. Inorg. Chim. Acta 2000, 300−302, 829−836. (b) Ray, K.; Weyhermüller, T.; Neese, F.; Wieghardt, K. Inorg. Chem. 2005, 44, 5345−5360. (c) Ray, K.; DeBeer George, S.; Solomon, E. I.; Wieghardt, K.; Neese, F. Chem. - Eur. J. 2007, 13, 2783−2797. (d) Machata, P.; Herich, P.; Lušpai, K.; Bucinsky, L.; Šoralová, S.; Breza, M.; Kozisek, J.; Rapta, P. Organometallics 2014, 33, 4846−4859. (38) Jeffrey, G. A. An Introduction to Hydrogen Bonding; Oxford University Press: Oxford, U.K., 1997. (39) Benedito, F. L.; Petrenko, T.; Bill, E.; Weyhermüller, T.; Wieghardt, K. Inorg. Chem. 2009, 48, 10913−10925. (40) Kirmse, R.; Stach, J.; Dietzsch, W.; Steimecke, G.; Hoyer, E. Inorg. Chem. 1980, 19, 2679−2685.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.6b00821. Depiction of intermolecular interactions in crystalline dithioboroles, 1H and 11B NMR spectra of the thioether deprotection reactions, variable-temperature 1H and 31P NMR spectra of [NEt4]10, UV−vis spectra of [NEt4]10 and [NEt3H]11, magnetic measurements of [NEt4]10, EPR spectra of [NEt3H]11, and far-infrared spectra of [NEt4]10 (PDF) Full crystallographic data for 1, 2, 4, 5, 7b, 7c, 7d′, 9d, and 10−12 as CIFs (ZIP)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Funding

Financial support by the Institut für Anorganische Chemie, the Fonds der Chemischen Industrie (Chemiefonds-Stipendium for K.B.), and the Deutsche Forschungsgemeinschaft (DFG) (Grants INST 41/863-1 and INST 41/887-1, and Grant SPP 1601 for J.v.S.) is gratefully acknowledged. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Dipl. Chem. K. Beyer and Dr. S. Strobel for the measurement of the UV−vis spectra, Dr. B. Blaschkowski for help with the SQUID magnetometer, and B. Förtsch for elemental analyses.

■

REFERENCES

(1) Schrauzer, G. N.; Mayweg, V. J. Am. Chem. Soc. 1962, 84, 3221− 3221. (2) McCleverty, J. A. Prog. Inorg. Chem. 1968, 10, 49−221. (3) Schrauzer, G. N. Acc. Chem. Res. 1969, 2, 72−80. (4) Burns, R. P.; McAuliffe, C. A. Adv. Inorg. Chem. Radiochem. 1979, 22, 303−348. (5) Hoyer, E.; Dietzsch, W.; Schroth, W. Z. Chem. 1971, 11, 41−53. (6) Müller-Westerhoff, U. T.; Vance, B.; Yoon, D. I. Tetrahedron 1991, 47, 909−932. (7) Cassoux, P.; Valade, L.; Kobayashi, H.; Kobayashi, A.; Clark, R. A.; Underhill, A. E. Coord. Chem. Rev. 1991, 110, 115−160. (8) Stiefel, E. I., Ed. Progress in Inorganic Chemistry Vol. 52: Dithiolene Chemistry; Wiley & Sons, Inc.: Hoboken, NJ, 2004. (9) Harrison, S. R.; Pilkington, R. S.; Sutcliffe, L. H. J. Chem. Soc., Faraday Trans. 1 1984, 80, 669−689. (10) Baudler, M.; Moog, A.; Glinka, K.; Kelsch, U. Z. Naturforsch., B: J. Chem. Sci. 1973, 28B, 363−365. (11) Goetze, R.; Nöth, H. Z. Naturforsch. 1980, 35B, 1212−1221. (12) (a) Lawlor, F. J.; Norman, N. C.; Pickett, N. L.; Robins, E. G.; Nguyen, P.; Lesley, G.; Marder, T. B.; Ashmore, J. A.; Green, J. C. Inorg. Chem. 1998, 37, 5282−5288. (b) Clegg, W.; Elsegood, M. R. J.; Lawlor, F. J.; Norman, N. C.; Pickett, N. L.; Robins, E. G.; Scott, A. J.; Nguyen, P.; Taylor, N. J.; Marder, T. B. Inorg. Chem. 1998, 37, 5289− 5293. (c) Dai, C.; Johnson, S. M.; Lawlor, F. J.; Lightfoot, P.; Marder, T. B.; Norman, N. C.; Orpen, A. G.; Pickett, N. L.; Quayle, M. J.; Rice, C. R. Polyhedron 1998, 17, 4139−4143. (13) Tian, Z. Q.; Donahue, J. P.; Holm, R. H. Inorg. Chem. 1995, 34, 5567−5572. (14) Kean, C. L.; Miller, D. O.; Pickup, P. G. J. Mater. Chem. 2002, 12, 2949−2956. I

DOI: 10.1021/acs.inorgchem.6b00821 Inorg. Chem. XXXX, XXX, XXX−XXX