Structure of a push–pull olefin prepared by ynamine hydroboration with a borandiol ester

The title compound is a demonstration of hydroboration of ynamines with borandiol ester. The bond lengths of the resulting push–pull olefin are discussed.


Chemical context
Boronic esters are frequently used to transfer organic groups to transition metals, for example in the transmetallation step of the Suzuki-Miyaura reaction. Hydroboration of ynamines with borandiol esters produces amino-functionalized boronic esters in one step and perfect atom economy.
For true ynamines, to the best of our knowledge, only two attempts of such reactions have been reported so far. These either failed (Witulski et al., 2000) or were reported without reaction details and characterization data (Zhuo et al., 2001). More recently it was found that the exceptionally active Pier's borane, HB(C 6 F 5 ) 2 , can readily hydroborate l-propynyl-2,2,6,6-tetramethylpiperidine (Wang et al., 2018). Borandiol esters are expected to be less reactive because the electron deficiency at the boron is reduced by -donation from the oxygen atoms.
Given the limited precedent for ynamine hydroboration, the more comprehensive literature for enamine hydroboration was consulted (Goralski & Singaram, 2012;Dembitsky et al., 2002), as their reactivity is expected to be controlled by similar effects. Compared to simple olefin substrates, conjugation of the C C bond with nitrogen dictates the regioselectivity and increases the reactivity of enamines. However, the presence of a nitrogen atom in the reactant and product enables the formation of unreactive Lewis acid-base adducts with the ISSN 2056-9890 hydroborating reagent. Building on the vast knowledge of the reactivity of different borane-amine adducts in hydroboration (Brown & Murray, 1984;Brown et al., 1999), a bulky iso-propyl and an phenyl group were selected as substituents for the ynamine nitrogen. The former should weaken adducts for steric reasons, whereas the phenyl group is expected to reduce the nucleophilicity of the nitrogen by conjugation.
Ynamine hydroboration using a borandiol ester was reinvestigated and succeeded for a substrate that follows the developed design principles. The product of such a reaction contains a C C double bond flanked by both an electrondonating group (EDG), the amine, and an electron-withdrawing group (EWG), the boronate. Therefore it belongs to the class of push-pull (captodative) olefins, which are known to have unusual properties such as weak -bonds or biradical reactivity (Viehe et al., 1985).

Structural commentary
The asymmetric unit ( Fig. 1) contains two almost identical (r.m.d.s = 0.11 Å ) independent molecules. As judged by the B1-C11-C10-N1 torsion angles of 171.6 (2) and 175.5 (2) , the central C-C bond is only slightly twisted from planarity. The two phenyl groups in the molecule are rotated, by 43 and 49 , with respect to that plane. The centroids of two phenyl groups in one molecule are on average 3.747 Å apart, which suggests intramolecular -stacking. The mean distances are 1.521 Å for the B1-C11 bond, 1.365 Å for the C10-N1 bond and 1.369 Å for the central C10-C11 bond.

Database survey
Contributions from the zwitterionic resonance structure À B C-C N + are expected to increase with donor and acceptor group strength. This should be observable as a shortening of the B-C and C-N bonds and an elongation of the C-C bond. Following this idea, bond lengths of 1 were compared to those in the structurally related compounds 2-5

Figure 1
The molecular structures of the two independent molecules of the title compound 1 with displacement ellipsoids drawn at the 50% probability level. Table 1 Comparison of bond lengths (in Å ) in 1 with those in the similar compounds 2-5.
Average distances and standard deviations are given whenever there is more than one molecule in the asymmetric unit. Typical bond lengths for vinylboranes and conjugated enamines were obtained from statistical analysis.

Compound
B-C C-C C-N CCDC  -Woźniak et al., 2011). Following this, the zwitterionic resonance structure is most important in 4, which has a strong donor and a strong acceptor. On the other end of the scale lies 2, which has a weak donor and a weak acceptor. The other molecules, including 1, lie between these two extremes.
In order to compare with olefins that either have a donor or an acceptor group, the Cambridge Structural Database (CSD, Version 5.41, update of November 2019; Groom et al., 2016) was searched for vinyl boronates and enamines. Bond-length distributions and the exact query structures are shown in Fig. 3 and Fig. 4. The data set for vinyl boronates consists of about 90% of pinacol boronates and contains only a few catechol boronates. Compared with typical bond lengths in this data set, the B-C bond is shorter and the C C bond is longer in 1-5, which indicates stronger conjugation. The only exception is 2, whose C C bond is shorter.
For enamines the C-C bond length has an average of 1.341 Å , which is about 0.025 Å longer than the value of 1.316 (15) Å for regular internal olefins (Allen et al., 2006). In 1, 3, 4 and 5, the C-C bonds are on average 1.378 Å , and thereby longer than in enamines. C-N bond lengths for enamines are distributed more uniformly than C-C lengths. Inspection of the structures in which C-N distances are longer than 1.39 Å revealed that these structures typically either have a nitrogen whose lone pair is not coplanar with the C C bond, or a nitrogen that is part of a carbazole or morpholine. As conjugation with the formal double bond between C10 and C11 is absent or reduced in these, only structures with C-N bond lengths below 1.39 Å were used for comparison. The average C-N bond length of about 1.36 Å for that subset is similar to the C-N bond lengths in 1, 3, 4 and 5. Overall, comparison with enamines reveals that C-C bonds are longer in push-pull olefins, whereas C-N bond lengths are unaffected. This suggests that conjugation with the boron affects the C-C bond length more than the C-N bond length.

Synthesis and crystallization
The title compound was prepared by the multi-step sequence shown in Fig. 5.
Under a counterflow of argon, 3.08 g of N-isopropyl-N-(phenylethynyl)aniline (13.1 mmol, 1.0 eq.) were placed in an oven-dried 20 ml Schlenk flask with a Young valve. The flask and its contents were purged three times by applying high vacuum followed by flushing with argon. 6.5 ml dry tert-butyl methyl ether were added and the mixture stirred vigorously to ensure mixing of the two liquids. 2.1 ml of catecholborane (19.5 mmol, 1.5 eq.) were added, the flask closed, and the reaction mixture heated to 323 K for 16 h. Cooling to room temperature led to the precipitation of the product. The supernatant was removed and the precipitate recrystallized from 40 ml of tert-butyl methyl ether. X-ray quality crystals were obtained in a yield of 1.88 g (40%). Notes: (a) Schlenk techniques are necessary because ynamines and catecholborane are moisture-sensitive; (b) the reaction also works well in diethyl ether, 1,4-dioxane or THF.

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 2. H atoms were positioned geometrically and refined as riding: C-H = 0.95-0.98 Å and U iso (H) = 1.2U eq (C) or 1.5U eq (C-methyl). The absolute structure was not determined because of unreliable Flack and Hooft parameters. Reaction sequence used for the synthesis of the title compound.

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. 8 reflections were omitted (some are equivalents). These were checked visually and are all results of high background around the beamstop (beginning ice formation or crystalline powder covering the sample). 0 1 0 is clearly shadowed by the beamstop. Absolute structure is not claimed due to unreliable Flack and Hooft parameters.