Enrico La
Cascia
ab,
Xavier
Sanz
ac,
Carles
Bo
c,
Andrew
Whiting
*b and
Elena
Fernandez
*a
aDepartment Química Física i Inorgànica, University Rovira i Virgili C/Marcellí Domingo s/n, Tarragona, Spain. E-mail: mariaelena.fernandez@urv.cat
bCentre for Sustainable Chemical Processes, Dept. of Chemistry, Durham University, South Road, Durham DH1 3LE, UK. E-mail: andy.whiting@durham.ac.uk
cICIQ, Tarragona, Spain
First published on 10th December 2014
The adduct [MeO → Bpin–Bpin]− efficiently mediates the β-boration of α,β-unsaturated imines formed in situ. The use of chiral phosphines as additives, and in particular the chiral phosphine (S)-MeBoPhoz, enables the catalytic asymmetric reaction to proceed with higher enantioselectivity than the analogue copper(I) mediated reaction.
![]() | ||
Scheme 1 (A) Cu catalyzed β-boration of α,β-unsaturated imines (ref. 7a,b,c); (B) Cu catalyzed β-boration of in situ formed α,β-unsaturated imines (ref. 7d,e,f); (C) Fe(II) activation of α,β-unsaturated imines towards the β-boration reaction (ref. 8); (D) organocatalytic β-boration reaction (this work). |
Here, we have developed an asymmetric organocatalytic approach to generate C–B bonds at the β-position of an unsaturated imine, i.e.Scheme 1, pathway D, as an alternative strategy to synthesize γ-aminoalcohols. Towards this end, we focus our efforts on the in situ generation of a model α,β-unsaturated imine, i.e. (E)-1-phenyl-N-(4-phenylbutan-2-ylidene)methanamine, from 4-phenyl-3-buten-2-one (1) and benzylamine in THF with the dehydrating reagent, MK10.7a,dAfter 6 hours, the boron reagent bis(pinacolato)diboron (B2pin2) is added to the intermediate α,β-unsaturated imine; however, even when the reaction was performed at 70 °C, no β-borated product 2a was observed (Table 1, entry 1). The addition of a base and MeOH to activate the diboron, via quaternization, was also insufficient at promoting the β-boration (Table 1, entry 2), unless a small amount of phosphine (10 mol% PCy3) was added to the reaction (see Table 1, entry 3). However, the replacement of the base by the phosphine alone was not enough to activate the diboron (Table 1, entry 4). It seems, therefore, that the base/MeOH combination is essential for the diboron activation and that the role of the phosphine could be related to a similar pre-activation of the substrate as we have previously observed in the analogue metal-free β-boration of α,β-unsaturated carbonyl compounds, which is also assisted by phosphines.9 Isolated yields were obtained for the corresponding γ-amino alcohols by reduction with NaBH4 in methanol and oxidation with H2O2 in NaOH.
Entry | Substrate | Additives | Product | % Conv.b [IY]c | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
a Standard conditions: ketone (0.5 mmol), NH2Bn (0.5 mmol), THF (2 mL), MK-10 (140 mg), B2pin2 (1.1 eq.), Cs2CO3 (15 mol%), MeOH (2.5 eq.), PCy3(10 mol%). b Conversion determined by 1H NMR spectroscopy. c Isolated yield calculated for syn-γ-amino alcohol (see ESI for reaction conditions). d NH2Bu (0.5 mmol). | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 |
![]() |
— |
![]() |
— | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
2 | ′′ | Cs2CO3/MeOH | ′′ | — | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
3 | ′′ | PCy3/Cs2CO3/MeOH | ′′ | 99 [56] | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
4 | ′′ | PCy3 | ′′ | — | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
5d | ′′ | PCy3/Cs2CO3/MeOH |
![]() |
90 [66] | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
6 |
![]() |
PCy3/Cs2CO3/MeOH |
![]() |
99 [47] | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
7 |
![]() |
PCy3/Cs2CO3/MeOH |
![]() |
96 [37] | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
8 |
![]() |
PCy3/Cs2CO3/MeOH |
![]() |
95 [51] | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
9 |
![]() |
PCy3/Cs2CO3/MeOH |
![]() |
99 [68] | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
10 |
![]() |
PCy3/Cs2CO3/MeOH |
![]() |
97 [30] |
With these preliminary results in hand, we extended this observation to other ketone and amine combinations, to develop a general organocatalytic methodology for the β-boration of α,β-unsaturated imines. Interestingly, nBuNH2 was also a versatile amine for the imine formation with 1, and was compatible with the organocatalytic β-boration to produce quantitatively the β-borated imine 2b (Table 1, entry 5). Electron accepting and electron releasing substituents at the para-position of the phenyl group of the ketone substrates 3 and 5, respectively, did not change the reaction outcome (Table 1, entries 6 and 7). Even α,β-unsaturated ketones with alkyl moieties at the β-position were equally susceptible to quantitative β-boration, whether cyclic or acyclic (Table 1, entries 8–10). Hence, it can be seen that the organocatalytic β-boration of in situ formed α,β-unsaturated imines is a general and indeed new methodology for the formation of β-borylated imines in a one-pot reaction.
In the next step we considered the possibility of inducing asymmetry into the formation of the new C–B bond using this organocatalytic approach. Hence, we proposed that chiral phosphine additives might interact with the substrate and provide an asymmetric environment for the β-boration with the Lewis acid–base adduct [i.e. MeO− → Bpin–Bpin]. This concept had already been successfully demonstrated in the β-boration of α,β-unsaturated ketones with B2pin22a,d or BpinBdan (dan = 1,8-diaminonaphthalene) (Scheme 2),2e and the hypothesis of the role of the phosphine in the asymmetric induction has also been postulated from both an experimental and theoretical point of view.9
![]() | ||
Scheme 2 β-Boration of α,β-unsaturated ketones with B2pin2 and BpinBdan (dan = 1,8-diaminonaphthalene), assisted by chiral phosphines. |
However, since imine functionality is more sterically hindered and less polarized than the carbonyl group, we were interested to ascertain whether asymmetric induction would be more or less efficient. Hence, we initiated our studies with substrate 1 and conducted the imine formation with benzylamine, followed by β-boration with the Lewis acid–base [MeO− → Bpin–Bpin] adduct in the presence of a series of chiral diphosphines. Preliminary results using chiral Josiphos-type of diphosphines did not provide any significant asymmetric induction, which contrasts with the efficient trends observed with the corresponding ketones.2a Remarkably, however, when the [MeO− → Bpin–Bpin] adduct was used with the diphosphine (S)-MeBoPhoz (P1), total conversion was observed together with moderate enantioselectivity of the β-borated product (54% ee, Table 2, entry 1). When subtle changes were made to the reaction conditions, such as a lower base loading or a different reaction temperature, conversions and enantioselectivities remained essentially unchanged. However, when the β-boration was carried out in the presence of CuCl (3 mol%), conversions from 1 to 2a were high but lower ees were observed (32% ee, Table 2, entry 2). Note that the isolated yields of the product are given for the final syn-γ-aminoalcohol after a highly stereoselective reduction protocol with NaBH4 in MeOH, as reported previously,7b followed by oxidation with H2O2 in NaOH.
Entry | β-Borated imine | Method | % Conv.b | % eec | % I.Y.d | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
a Conditions for method A: ketone or aldehyde (0.5 mmol), amine (0.5 mmol), THF (2mL), MK-10 (140 mg), B2pin2 (1.1 eq.), Cs2CO3 (15 mol%), MeOH (2.5 eq.), (S)-MeBoPhoz (10 mol%), 70 °C; for method B: same as method A + CuCl (3 mol%), 25 °C. b Conversion determined by 1H NMR spectroscopy. c Enantioselectivity determined from HPLC-MS. d Isolated yield for the corresponding syn γ-aminoalcohol (see ESI for reaction conditions). e ee calculated on the 4-(N-benzhydrylacetamido)butan-2-yl acetate derivative. | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 |
![]() |
A | 90 | 54 | 59 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
2 | B | 99 | 32 | 40 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
3 |
![]() |
A | 94 | 53 | 47 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
4 | B | 80 | 32 | 40 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
5 |
![]() |
A | 98 | 50 | 49 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
6 | B | 95 | 45 | 43 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
7 |
![]() |
A | 96 | 70 | 61 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
8 | B | 88 | 61 | 57 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
9 |
![]() |
A | 99 | 51 | 48 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
10 | B | 92 | 33 | 57 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
11 |
![]() |
A | 99 | 57e | 73 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
12 | B | 95 | 29e | 52 |
Since (S)-MeBoPhoz has been shown to be the most active and enantioselective additive for accessing β-boryl imines, in this metal free context, we extended this study to other similar chiral phosphines, i.e.P2–P4. We concluded that (R)-PhEt-(R)-BoPhoz (P4) provides comparable asymmetric induction than the close phosphine P1, and higher than the enantioselectivities provided by the other analogues, i.e.P2 and P3, in which the amine is either mono- or di-substituted (Fig. 1).
![]() | ||
Fig. 1 Comparison of the chiral phosphine additives P2–P4 for asymmetric β-boration of α,β-unsaturated imines 4, 6 and 8. |
To gain a deeper insight into the reaction mechanism and compare with other substrates that we reported previously,1b we conducted DFT-based theoretical studies (Scheme 3).
Initially, we postulated that the methoxide ion can quaternize a boron atom of the B2pin2 molecule forming the activated adduct [MeO → Bpin–Bpin]− (chosen as the origin of the energies). This adduct can then react with the model α,β-unsaturated imine through a transition state TS, which corresponds to the nucleophilic attack of the sp2 boron atom on the β-carbon of the α,β-unsaturated imine. The structural features of the TS show the cleavage of the B–B bond (ΔdB–B = 0.257 Å) and the formation of the new B–C bond (dB–C = 2.078 Å). After this transition state (TS), a negatively charged intermediate I is formed. Also in this step, a molecule of (pin)B–OMe is released as the by-product. The anionic intermediate I is then protonated in the presence of an excess of B2pin2 and MeOH, regenerating again the active species [MeO → Bpin–Bpin]− and hence the β-borated product. At this point, it is interesting to compare energy values computed herein, with those obtained for the metal-free β-boration of ketones, esters and aldehydes.1b For the model imine (E)-1-phenyl-N-(4-phenylbutan-2-ylidene)methanamine, (2a), the transition state TS is higher (ΔG≠ = 32.3 kcal mol−1) than that found for acrolein (ΔG≠ = 16.7 kcal mol−1), 3-buten-2-one (ΔG≠ = 18.7 kcal mol−1), methyl acrylate (ΔG≠ = 21.5 kcal mol−1) and styrene (ΔG≠ = 25.1 kcal mol−1), but lower in energy than propylene (ΔG≠ = 35.9 kcal mol−1). This fact can be explained by the lower electrophilicity of the Cβ of the α,β-unsaturated imine which makes it less reactive towards the nucleophilic attack. Moreover, the intermediate I for the imine (ΔG = –17.2 kcal mol−1) is energetically more stable than the reactants, as expected, but less stable than the corresponding analogues for the activated alkenes.1b This can be also rationalized by the fact that the negative charge that is generated is more stabilized by the oxygen atom than the nitrogen due to their different electronegative characters. It is worth mentioning that the reaction energies computed for this model α,β-unsaturated imine substrate are in a similar range to those previously computed for ketones, aldehydes and esters, thus justifying the similarity in the reaction conditions (T = 70 °C) as described above.
Finally, we addressed the role of the chiral phosphine in not only mediating the catalytic reaction but importantly, guiding the asymmetric C–B bond formation. A possible interaction between a model phosphine of reduced steric congestion PMe3, and the α,β-unsaturated imine 2a, is to form a phosphonium enolate intermediate (Fig. 2).10–12 We compared this with the corresponding α,β-unsaturated ketone-derived enolate species (Fig. 2). Interestingly, the imine-derived phosphonium enamide formed from PMe3 and 2a is higher in energy than the corresponding ketone-derived phosphonium enolate intermediate, which explains why that reaction has to be carried out at 70 °C, and does not proceed readily at lower temperature. Hence, the origin of the asymmetric induction when using (S)-MeBoPhoz may result from the protonation of the zwitterionic phosphonium enamide with MeOH, and formation of a tight ion-pair between the resulting [B2pin2·MeO]− adduct and the chiral phosphonium imine, i.e. as in 15 (Scheme 4), as we have postulated before.9
![]() | ||
Fig. 2 Reaction energy profile for the formation of phosphonium enolates. Electronic and Gibbs free energies (in parentheses) are given in kcal mol−1. |
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ob02478h |
This journal is © The Royal Society of Chemistry 2015 |