Literature DB >> 35535567

Sigmatropic [1,5] Carbon Shift of Transient C3 Ammonium Enolates.

Garrit Wicker1, Rundong Zhou1, Roland Schoch1, Jan Paradies1.   

Abstract

The stereospecific sigmatropic [1,5] carbon shift of C3 ammonium enolates is discovered. According to mechanistic, kinetic and computational experiments, this new rearrangement proceeds via the catalytic generation of a transient C3 ammonium enolate by intramolecular aza-Michael addition. This intermediate rapidly undergoes [1,5] sigmatropic carbon migration to furnish the respective tetrahydroquinoline-4-ones with excellent diastereoselectivities of d.r. >99 : 1 and in 61-98 % yield.
© 2022 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH.

Entities:  

Keywords:  Ammonium Enolate; Asymmetric; Density Functional Calculations; Sigmatropic Rearrangement; Tetrahydroquinoline

Year:  2022        PMID: 35535567      PMCID: PMC9401041          DOI: 10.1002/anie.202204378

Source DB:  PubMed          Journal:  Angew Chem Int Ed Engl        ISSN: 1433-7851            Impact factor:   16.823


Rearrangement reactions are one of the most useful tools for construction of organic scaffolds. In fact, [1,2] and [3,3] sigmatropic rearrangements are typical reactions used to achieve molecular complexity with high stereocontrol. Observations of sigmatropic shifts of hydrogen, silicon or acyl groups across extended π‐systems, such as cyclopentadiene, indenyl or quinone[ , ] structures, have provided a basis for in‐depth understanding of organic chemistry in addition to the synthetic benefits. Electrophilic nitrogen‐ylide rearrangements, like the [1,2] and [2,3] Stevens and Sommelet–Hauser rearrangements, offer stereoselective access to synthetically useful amino building blocks through nitrogen to carbon chirality transfer. The [1,2] Stevens progresses by homolytic C−N bond dissociation with formation of a caged radical pair, while the [2,3] Stevens and the Sommelet–Hauser rearrangements proceed by a concerted mechanism (Scheme 1a and b).
Scheme 1

a) [1,2] and b) [2,3] Stevens rearrangement; c) sigmatropic [1,5] carbon shift of C3 ammonium enolates.

a) [1,2] and b) [2,3] Stevens rearrangement; c) sigmatropic [1,5] carbon shift of C3 ammonium enolates. Both rearrangements require, first, quaternization of the nitrogen atom and, as a second step, formation of the ylide by deprotonation. There have been only a few reported examples of catalytic creation of the ylide intermediate. And although C3 ammonium enolates are well‐known intermediates in aza‐Michael and Morita–Baylis–Hillman reactions, cyclic C3 ammonium enolates have yet to be considerd as precursors for sigmatropic [1,5] rearrangements (Scheme 1c). We had observed during our recent investigations of redox isomerization an unsusal migration of a benzylic fragment (Scheme 2, top).
Scheme 2

Lewis acid‐induced benzyl migration.

Lewis acid‐induced benzyl migration. Unexpectedly, the reaction of the amino chalcone d‐rac‐1 a with B(C6F5)3 did not furnish the corresponding deuteride shift product, but rather the carbon migration product d1‐rac‐2 a as a single diastereomer in 90 % yield. Remarkably, when the reaction was performed with the enantiopure (S)‐phenylethyl derivative (S)‐1 a, only the trans‐2 a was obtained as a single enantiomer (e.r. 99 : 1) in 83 % yield (Scheme 2, bottom). Crystal structure analysis of the reaction product permitted indirect assignment of the absolute configuration to that of trans‐(2S,3R)‐(S)‐phenylethyltetrahydroquinolone 2 a, which indicates that the reaction proceeds under excellent stereocontrol with complete chirality transfer and retention of configuration of the migrating carbon center. Careful inspection of the reaction by 1H NMR spectroscopy did not provide deeper insight into the reaction mechanism, because only the starting materials and trans‐2 a were detectable. However, the 11B NMR featured resonances at δ=−1.8 ppm and −5.8 ppm, accounting for tetragonal B−O adducts. We anticipated that the reaction might proceed by a Lewis acid‐induced intramolecular aza‐Michael reaction setting the stage for a so far unprecedented [1,5] carbon shift of a transiently generated C3 ammonium enolate. To support that such a process was at work, we isolated and characterized the C3 ammonium enolate 4 as the reaction product of 4‐N,N‐(dimethylamino)chalcone (3) with 1.0 equiv B(C6F5)3 (Scheme 3).
Scheme 3

Reaction of 3 with B(C6F5)3; molecular structure of 4; selected bond lengths: N(1)−C(116) 1.5080(12) Å; N(1)−C(117) 1.5138(12) Å; C(107)−C(108) 1.3435(13) Å.

Reaction of 3 with B(C6F5)3; molecular structure of 4; selected bond lengths: N(1)−C(116) 1.5080(12) Å; N(1)−C(117) 1.5138(12) Å; C(107)−C(108) 1.3435(13) Å. The bond length between C(107)−C(108) of 1.3435(13) Å is in agreement with a benzylic enolate double bond (1.3565(11) Å). Notably, the nitrogen atom is quaternized, with one methyl group in equatorial position and the second axial and antiperiplanar to the phenyl substituent. The axial N(1)−C(117) bond is slightly elongated compared to the equatorial N(1)−C(116) bond, suggesting electronic interaction with the aromatic system. Since the conjugate addition is of utmost importance for the overall reaction, we investigated substituent effects at the Michael acceptor using 4‐methoxybenzyl (PMB) as migrating group (Scheme 4). To our delight, electron‐rich and electron‐deficient substituents at the olefin were very well tolerated. However, we found that the reaction rate is sensitive to electronic modifications. The Hammett analysis of 1 b–d displays a positive slope of ρ=1.889±0.13, which implies that the negative charge arising from nucleophilic attack at the Michael position in the rate‐determining step needs to be stabilized. The tetrahydroquinolinone bearing a quaternary stereocenter 2 e was obtained in 88 % yield as a single diastereomer. Also, double substitution at the Michael position was tolerated such that 2 f was produced in 65 % yield. Bulky tert‐butyl substituents (2 g) as well as heterocyclic ones, such as furanyl (2 h), thiophenyl (2 i) or pyridyl (2 j), were well tolerated. The molecular structure of 2 j justifies our assignment of it to the trans diastereomer. Next, we studied the impact of the nitrogen substituents on the rearrangement reaction (Scheme 5).
Scheme 4

Modifications at the olefin in the sigmatropic [1,5] carbon rearrangement; B(2,4,6‐F3‐C6H2)3 (10 mol %) was used as catalyst in the Hammett analysis.

Scheme 5

Scope of the nitrogen substituent in the sigmatropic [1,5] carbon rearrangement; [a] performed at 60 °C.

Modifications at the olefin in the sigmatropic [1,5] carbon rearrangement; B(2,4,6‐F3‐C6H2)3 (10 mol %) was used as catalyst in the Hammett analysis. Scope of the nitrogen substituent in the sigmatropic [1,5] carbon rearrangement; [a] performed at 60 °C. Larger substituents on the nitrogen atom, like ethyl (2 k) or isobutyl (2 l), did not suppress the reaction, but slightly diminished yields were obtained, with preservation of the excellent diastereoselectivity of >99 : 1. Thereafter, we investigated the scope of the migrating group attached to the nitrogen atom. The phenylethyl derivatives 2 m–o were obtained as single diastereomers in 78–89 % yield. The optical purity of (S)‐1 n (96 % ee) was quantitatively transferred to the three stereocenters of (2S,3R)‐2 n. The fluoro derivatives 1 p and 1 q required elevated temperatures of 60 °C, probably due to the reduced nucleophilicity of the nitrogen atom. However, both products 2 p and 2 q were obtained without depletion of the perfect diastereoselectivity at 87 % and 89 % yield, respectively. We successfully expanded the scope of the migrating groups to the cinnamyl and propargyl derivatives 1 r and 1 s, which cleanly underwent migration. The products 2 r and 2 s were obtained in 86 % and 75 % yield as single diastereomers (d.r. >99 : 1). The impact of various substituents on the aromatic backbone was probed (Scheme 6).
Scheme 6

Modification of the aromatic ring; [a] the reaction was performed for 18 h.

Modification of the aromatic ring; [a] the reaction was performed for 18 h. The products 2 t–2 v were obtained in 60 % to 81 % yields. The ortho‐CF3 substituted derivative 2 w was obtained in 21 % yield, probably as a result of steric crowding of the carbonyl group. Electronic effects seem to play a minor role because the meta‐CF3 substituted 1 x was converted into 2 x in 90 % yield, although the N‐atom should be significantly less nucleophilic. Furthermore, bromo, methyl, nitro and methoxy groups were well tolerated (2 y–2 ab). All products were produced with an excellent diastereoselectivity of >99 : 1. Finally, we investigated the reaction of (S)‐1 a in detail by density functional theory (DFT) on the PW6B95/def2‐QZVPP//PBEh‐3c/def2‐mSVP level including dispersion correction D3BJ and the CPCM(CHCl3) solvent model as implemented in the ORCA package after preoptimizations with GFN2‐xTB. Transition states were located by the growing string method in combination with GFN2‐xTB and then further optimized by DFT‐level calculations. The binding of B(C6F5)3 to (S)‐1 a is endergonic by 6.0 kcal mol−1 and induces the change from the s‐cis to the s‐trans conformer. Both conformations provide a perfect setup for the aza‐Michael addition (Figure 1, see Supporting Information for details). The addition of the nitrogen atom to the α,β‐unsaturated carbonyl fragment is the rate‐ and stereo‐discriminating step. The barrier of the Si‐face attack (TS1) is 2.5 kcal mol−1 lower in free energy than the Re‐face attack (TS1′).
Figure 1

Calculated free reaction energies in kcal mol−1 and transition state geometries ([B]=B(C6F5)3).

Calculated free reaction energies in kcal mol−1 and transition state geometries ([B]=B(C6F5)3). The diastereomeric C3 ammonium enolates INT and INT′ are nearly equal in free energy. For each of the diastereomeric transition states of the [1,5] carbon shift, two low‐energy conformers were identified, which differ by the orientation of the phenylethyl group (Figure 1, bottom: TS2, TS2′ and stacked‐TS2, stacked‐TS2′). For both C3 ammonium enolate intermediates (INT and INT′), the lowest barriers for the [1,5] carbon shift via stacked‐TS2 and stacked‐TS2′ are lower than those for aza‐Michael addition, so it is reasonable that these intermediates were not detectable by NMR spectroscopy. The stacked‐TS2 benefits energetically from both the stabilizing C6H5‐C6F5 and quinoline‐migrating group π‐π interactions (see NCI plots in Supporting Information, Figure S5 and S6). The higher barrier of TS2 by 6.2 kcal mol−1 may be attributed to the reduced π‐π interaction of the migrating group with the tetrahydroquinoline. A comparable ΔΔG* of 6.4 kcal mol−1 was computed for the [1,5] shift of 1 b (Figure S7 and S8).] The stabilizing C6H5−C6F5 interaction is absent in both TS2′ transition states, which may account for the higher barrier of approx. 15–17 kcal mol−1. However, both of these interactions are not crucial for the sigmatropic shift, because the aliphatic derivatives 1 f and 1 g (see Scheme 4) and the substituted benzoids 1 t–1 x (see Scheme 6) were cleanly converted into the corresponding tetrahydroquinolinones. The formation of (2S,3R)‐2 a⋅B(C6F5)3 is exergonic by 22.4 kcal mol−1, and it is more stable than the corresponding free species. This is in agreement with the 11B NMR experiments, which confirmed that only tetragonal B−O adducts were present. The transition state stacked‐TS2 was studied in more detail by reoptimization with the range‐separated hybrid functional ωB97X‐D3BJ/def2‐TZVP. The electronic structure of the transition state in the [1,2] Stevens rearrangement has been the object of some debate and can be best described as singlet biradical. Unrestricted DFT calculations converged to the closed shell determinant (⟨S2⟩=0.000, ΔE UKS‐RKS=0.75 kcal mol−1), so stacked‐TS2 can be classified as a closed‐shell species. This is supported by a complete active space self‐consistent field (CASSCF) (2,2)/def2‐TZVP calculation which shows only minor LUMO population of 0.073. This result is corroborated by the observation that the reaction rate was not influenced when the reaction was performed in the presence of 1.0 equiv of TEMPO (see Supporting Information). The localized orbital analysis reveals double bond character for both the N‐aryl (aniline) and for the enolate fragment (Figure 2, top).
Figure 2

Localized orbitals of the benzo‐N and of the enolate π‐bond (UKS‐ ωB97X‐D3BJ/def2‐TZVP) of the transition state structure stacked‐TS2 (top); orbital analysis of Woodward–Hoffmann symmetry‐allowed processes (bottom).

Localized orbitals of the benzo‐N and of the enolate π‐bond (UKS‐ ωB97X‐D3BJ/def2‐TZVP) of the transition state structure stacked‐TS2 (top); orbital analysis of Woodward–Hoffmann symmetry‐allowed processes (bottom). The distances of the N atom and enolate C atom to the benzyl C atom deviate only marginally, by 0.08 Å (r N‐benzyl=2.87 Å, r enolate‐benzyl=2.95 Å). This indicates a concerted migration of the benzyl group in a slightly asynchronous transition state. The reaction can therefore be classified as a sigmatropic [1,5] rearrangement of a transiently formed C3 ammonium enolate. This is in accordance with Woodward–Hoffmann rules, which prescribe that it proceed via a π4s+σ2s or π4a+σ2a process (Figure 2, bottom). In summary, we discovered a new stereospecific [1,5] sigmatropic rearrangement of transiently generated C3 ammonium enolates. In contrast to other electrophilic rearrangements, the reactive species is catalytically generated through an intramolecular aza‐Michael reaction. Kinetic, mechanistic and computational experiments all indicate that the rate‐determining step is the initial formation of the zwitterionic C3 ammonium enolate. The symmetry‐allowed [1,5] shift proceeds with retention of configuration of the migrating carbon atom. The tetrahydroquinoline‐4‐ones were obtained in high yields and with excellent diastereoselectivity (d.r. >99 : 1).

Conflict of interest

The authors declare no conflict of interest. As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors. Supporting Information Click here for additional data file.
  47 in total

1.  Control of kinetics and thermodynamics of [1,5]-shifts by aromaticity: a view through the prism of Marcus theory.

Authors:  Igor V Alabugin; Mariappan Manoharan; Boris Breiner; Frederick D Lewis
Journal:  J Am Chem Soc       Date:  2003-08-06       Impact factor: 15.419

2.  Balanced basis sets of split valence, triple zeta valence and quadruple zeta valence quality for H to Rn: Design and assessment of accuracy.

Authors:  Florian Weigend; Reinhart Ahlrichs
Journal:  Phys Chem Chem Phys       Date:  2005-08-04       Impact factor: 3.676

3.  Accurate Coulomb-fitting basis sets for H to Rn.

Authors:  Florian Weigend
Journal:  Phys Chem Chem Phys       Date:  2006-01-03       Impact factor: 3.676

4.  Rhodanine-based tau aggregation inhibitors in cell models of tauopathy.

Authors:  Bruno Bulic; Marcus Pickhardt; Inna Khlistunova; Jacek Biernat; Eva-Maria Mandelkow; Eckhard Mandelkow; Herbert Waldmann
Journal:  Angew Chem Int Ed Engl       Date:  2007       Impact factor: 15.336

5.  Sigmatropic rearrangements of 'onium' ylids.

Authors:  J B Sweeney
Journal:  Chem Soc Rev       Date:  2009-03-05       Impact factor: 54.564

6.  Tandem catalytic allylic amination and [2,3]-Stevens rearrangement of tertiary amines.

Authors:  Arash Soheili; Uttam K Tambar
Journal:  J Am Chem Soc       Date:  2011-08-03       Impact factor: 15.419

7.  Synthesis of Bridged Tetrahydrobenzo[b]azepines and Derivatives through an Aza-Piancatelli Cyclization/Michael Addition Sequence.

Authors:  Shengdong Wang; Régis Guillot; Jean-François Carpentier; Yann Sarazin; Christophe Bour; Vincent Gandon; David Lebœuf
Journal:  Angew Chem Int Ed Engl       Date:  2019-11-27       Impact factor: 15.336

8.  Frustrated Lewis Pair Catalyzed Dehydrogenative Oxidation of Indolines and Other Heterocycles.

Authors:  Alexander F G Maier; Sebastian Tussing; Tobias Schneider; Ulrich Flörke; Zheng-Wang Qu; Stefan Grimme; Jan Paradies
Journal:  Angew Chem Int Ed Engl       Date:  2016-09-05       Impact factor: 15.336

9.  Sigmatropic [1,5] Carbon Shift of Transient C3 Ammonium Enolates.

Authors:  Garrit Wicker; Rundong Zhou; Roland Schoch; Jan Paradies
Journal:  Angew Chem Int Ed Engl       Date:  2022-06-10       Impact factor: 16.823

View more
  1 in total

1.  Sigmatropic [1,5] Carbon Shift of Transient C3 Ammonium Enolates.

Authors:  Garrit Wicker; Rundong Zhou; Roland Schoch; Jan Paradies
Journal:  Angew Chem Int Ed Engl       Date:  2022-06-10       Impact factor: 16.823

  1 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.