Tris[2,2,6,6-tetramethyl-8-(trimethylsilyl)benzo[1,2-d;4,5-d′]bis(1,3-dithiol)-4-yl]methanol diethyl ether monosolvate

The title compound is a precursor of a stable triarylmethyl radical used in EPR-spectroscopy. It’s structure features a propeller-like conformation of the phenyl rings and a sterically crowded geometry at the central carbon.


Chemical context
The reported triarylmethanol 1 is the direct precursor of the corresponding triarylmethyl radical. Such tetrathiarylmethyl radicals, also called trityl radicals, can be used as spin labels for EPR-based distance measurements (Reginsson et al., 2012;Kunjir et al., 2013) and have recently been employed for structure determination in proteins (Jassoy et al., 2017;Yang et al., 2012) as well as nucleic acids (Shevelev et al., 2015). They are also used for dynamic nuclear polarization experiments (Jä hnig et al., 2017). Trityl radicals feature a very narrow linewidth in EPR spectra, slow spin-spin relaxation at room temperature and show line-broadening depending on the oxygen concentration in their surroundings. The latter property also makes them suitable as oxygen probes (Frank et al., 2015). However, most of the trityl radicals reported in the literature feature carboxylic acid derivatives as substituents in the para-position. The title compound 1 is a promising precursor for differently para-substituted trityl alcohols and their corresponding radicals.

Structural commentary
Compound 1 crystallizes (in space group P1 with the unit cell containing two molecules) from diethyl ether as a racemic mixture with respect to the propeller-like conformation of the ISSN 2056-9890 aryl building blocks. The unit cell consists of one P-and one M-configured molecule, as shown in Fig. 1.
The structure of the title compound deviates from C 3 symmetry, since the dihedral angles between the aryl planes are not equivalent (AE73.7, AE73.7, AE70.2 ). Moreover, the structure of 1 exhibits an Si-C ar bond length of 1.909 (3) Å to 1.945 (4) Å , whereas a bond length X 3 Si-C ar of 1.863 (14) Å is typically expected (Allen et al., 1987). This elongation of the Si-C ar bond may be due to the sterical stress at the para-positions caused by vicinal sulfur atoms. Additionally, the bond angles between the tetrathiaryl substituents at C1 are 112.2 (2), 113.5 (2) and 114.0 (2) , exceeding the tetrahedral angle of 109.5 . Therefore, regarding its geometry, C1 is situated between a tetrahedral and a trigonal-planar environment with a deviation of 0.409 (4) Å from the plane through atoms C2, C17 and C32. This coincides with the experimental observation that the title compound forms the corresponding carbocation with low effort, meaning its structure is already similar to the transition state according to Hammond's postulate. However, the C1-O1 bond length of 1.439 (3)  Crystal structure of the title compound, 1. Displacement ellipsoids are at the 50% probability level. Only the major disorder component is shown. Table 1 Hydrogen-bond geometry (Å , ). fits the value expected for tertiary alcohols, which is 1.440 (12) Å (Allen et al., 1987) and does not show any elongation. Regarding the envelope-configured 1,3-dithianes, C-S-C angles between 94.4 (2) and 96.1 (2) and C-C-S-C torsion angles in 1,3-dithianes between 18.7 and 26.9 are observed, with the methylene groups pointing either above or below the aromatic ring plane although without regularity. This is also observed within the crystal structure of the unsubstituted trityl alcohol 2 (Fig. 2). The molecular structure of compound 1 features an O1-H1Á Á ÁS8 hydrogen bond with a donor to acceptor atom distance of 3.031 (2) Å , which falls into the regime of a moderately strong hydrogen bond according to Jeffrey (1997). In addition, the H1Á Á ÁS8 distance of 2.32 Å is significantly shorter than 2.90 Å , the sum of the van der Waals radii (Bondi, 1964). The remaining five intramolecular hydrogen bonds listed in Table 1 belong into the category of weak electrostatic hydrogen bonds, with the shortest having a donor-acceptor atom distance of 3.435 (3) Å and the longest a donor-acceptor distance of 3.926 (5) Å . Other contacts between the molecules were not observed.

Supramolecular features
In the crystal, a number of C-HÁ Á ÁS interactions occur (Table 1).

Database survey
The Cambridge Structural Database (CSD, Version 5.38; Groom et al., 2016) contained two structures of para-substituted trityl radicals [ESECUB (Decroos et al., 2011) and TIXCEJ (Liu et al., 2008)] and one structure determination for compound 2 (REGBUG; Driesschaert et al., 2012). As found here for compound 1, the reported structure of 2 also deviates from C 3 symmetry, with dihedral angles for the aryl planes of AE75.3, AE70.7, AE69.9 . However, in contrast to the crystal structure reported here, Driesschaert et al. (2012) do not report on any hydrogen bonding within the structure of 2 but the C-HÁ Á ÁS distances are very similar than those in Table 1.
Tris-(2,2,6,6-tetramethylbenzo [1,2-d;4,5-d]bis [1,3]dithiol-4yl)methanol 2 (4.00 g, 4.52 mmol) was dissolved in 200 mL of dry diethyl ether under argon. Dry tetramethylethylendiamine (6.80 mL, 5.24 g, 45.1 mmol, 10 eq.) was added and the solution was cooled to 273 K. Subsequently, n-butyl lithium (2.5 M in hexanes, 18.08 mL, 45.2 mmol, 10 eq.) was added dropwise. The reaction mixture was allowed to warm up to room temperature while stirring for 3 h. Afterwards, the reaction mixture was cooled down to 195 K and trimethylsilyl chloride (6.30 mL, 5.40 g, 49.7 mmol, 11.0 eq.) was added dropwise. Then, the cooling bath was removed and the mixture was stirred for 16 h at room temperature. The reaction was then quenched with 10 mL 1 M NaOH and the organic solvents were removed under reduced pressure. The dark-greenish residue was taken up in methylene chloride (200 mL) and washed with water (200 mL) twice. The organic phase was separated and dried over sodium sulfate. After removal of the solvents under reduced pressure, the crude product was purified by washing with acetone. For that, the residue was suspended in acetone (50 mL) and treated with ultrasound for 3 min. Then, the mixture was centrifuged at 3200 g (Eppendorf Centrifuge 5810 R) for 5 min, whereupon a colorless solid separated. This procedure was repeated with the precipitated solid three times, until the supernatant was clear and almost colorless. The pure product was obtained as a colorless solid after drying the precipitate under vacuum with a yield of 3.32g (3.01 mmol, 67%). The pure product was then crystallized in the following way: compound 1 was dissolved in diethyl ether, the clear solution placed in an open tube at 278 K and the solvent was slowly evaporated over three days. This yielded light-yellow plates of 1 suitable for X-ray diffraction. 1 H NMR (500 MHz, CD 2 Cl 2 , 298 K, in ppm): 6.50 (s, 1H), 1.77 (s, 18H), 1.65 (s, 9H), 1.61 (s, 9H), 0.46 (s, 27H). 13 C NMR (126 MHz, CD 2 Cl 2 , 298 K, in ppm): 144.92, 144.53, 140.83, 138.79, 133.56, 130.66, 85.11, 62.13, 61.86, 34.92, 32.24, 29.33, 27.20, 2.66. The assignment of NMR signals for trityl alcohols has been discussed by Tormyshev et al. (2012)

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 2. H atoms were positioned geometrically and refined using a riding model as idealized hydroxy and methyl groups (SHELXL AFIX codes 147 and 137), thus including free rotation around the respective C-O and C-C bonds. U iso (H) was set to 1.5 times U eq (C,O). At a first attempt, a diethyl ether solvent molecule was modeled over three partially occupied positions summing up to one molecule. This model still contained a residual of approximately Synthesis of the title compound 1. two electrons, which could not be further incorporated into an appropriate model of a fourth orientation of the ether. Therefore, we decided to use the PLATON SQUEEZE (Spek, 2015) solvent masking procedure as implemented in OLEX2 (Dolomanov et al., 2009). The calculated solvent void in the unit cell has a volume of 580 Å 3 and 127 e have been recovered. The previous model of the refined parts of the diethyl ether molecules without the use of solvent masking is added as a part of a SHELXL res file to the refine_spe-cial_details section of the CIF file. The C5-bonded trimethylsilyl group shows a half-to-half disorder over two positions slightly above and below the plane of the respective phenyl ring. This disorder could be resolved by individual refinement of the respective parts with occupancy factors linked together via a free variable [occupancy ratio 0.504 (4):0.496 (4)]. Additionally two Si-C distance restraints to 1.80 (1) Å were applied for two Si-C bonds, and some U iso and U aniso restraints were used. Atom S2 is disordered over two positions in a 0.509 (7):0.491 (7) ratio. The two disordered S atoms were treated with SIMU/ISOR restraints; the bond lengths to neighbouring atoms C4 and C8 were subjected to a SADI restraint.   Data collection: SMART (Bruker, 2015); cell refinement: SAINT (Bruker, 2015); data reduction: SAINT (Bruker, 2015); program(s) used to solve structure: SHELXS (Sheldrick, 2008); program(s) used to refine structure: SHELXL2014/7 (Sheldrick, 2015); molecular graphics: OLEX2 (Dolomanov et al., 2009); software used to prepare material for publication: OLEX2 (Dolomanov et al., 2009).

Funding information
where P = (F o 2 + 2F c 2 )/3 (Δ/σ) max = 0.001 Δρ max = 0.89 e Å −3 Δρ min = −1.34 e Å −3 Extinction correction: SHELXL2014/7 (Sheldrick 2015), Fc * =kFc[1+0.001xFc 2 λ 3 /sin(2θ)] -1/4 Extinction coefficient: 0.0018 (2) Special details Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes. Refinement. H atoms were positioned geometrically and refined using a riding model as idealised hydroxy-and methyl groups (AFIX codes 147 and 137), thus including free rotation around the respective C-O and C-C bonds. The U iso (H) was set to 1.5 times U eq (C/O). At a first attempt a diethyl ether solvent molecule was modeled over three partially occupied positions summing up to one molecule. This model contained still Q-peaks of approx. 2 electrons, which could no be further incorporated into an appropriate model of a forth orientation of the ether. Therefore, we decided to use the solvent masking procedure -as implemented in Olex2 (Dolomanov et al., 2009)). The previous model of the refined parts of diethyl ether molecules is added as a part of a Shelx-RES-file to this section. The C5-bonded trimethylsilyl group shows a half-to-half disorder over two positions slightly above and below the plane of the respective phenyl ring. This disorder could be resolved by individual refinement of the respective parts with occupancy factors linked together via a free variable. Additionally two Si-C distance restraints to 180 (1) pm has been applied for two Si-C bonds, and some U iso and U aniso restraints were used.