Synlett 2018; 29(07): 928-932
DOI: 10.1055/s-0036-1591540
letter
© Georg Thieme Verlag Stuttgart · New York

Palladium(0)-Catalyzed Dearomatization of 2-Nitrobenzofurans through Formal (3+2) Cycloadditions with Vinylcyclopropanes: A Straightforward Access to Cyclopenta[b]benzofurans

PSL Research University, Chimie ParisTech, CNRS, Institut de Recherche de Chimie Paris, 11, rue Pierre et Marie Curie, Paris, 75005, France   Email: maxime.vitale@chimie-paristech.fr
,
Maxime Laugeois
PSL Research University, Chimie ParisTech, CNRS, Institut de Recherche de Chimie Paris, 11, rue Pierre et Marie Curie, Paris, 75005, France   Email: maxime.vitale@chimie-paristech.fr
,
PSL Research University, Chimie ParisTech, CNRS, Institut de Recherche de Chimie Paris, 11, rue Pierre et Marie Curie, Paris, 75005, France   Email: maxime.vitale@chimie-paristech.fr
,
Virginie Ratovelomanana-Vidal
PSL Research University, Chimie ParisTech, CNRS, Institut de Recherche de Chimie Paris, 11, rue Pierre et Marie Curie, Paris, 75005, France   Email: maxime.vitale@chimie-paristech.fr
,
PSL Research University, Chimie ParisTech, CNRS, Institut de Recherche de Chimie Paris, 11, rue Pierre et Marie Curie, Paris, 75005, France   Email: maxime.vitale@chimie-paristech.fr
› Author Affiliations

This work was supported by the Ministère de l’Education Nationale, de l’Enseignement Supérieur et de la Recherche and the Centre ­National de la Recherche Scientifique.
Further Information

Publication History

Received: 08 January 2018

Accepted after revision: 15 January 2018

Publication Date:
05 February 2018 (online)

 


Dedicated to Professor Miguel Yus on the occasion of his 70th birthday.

Abstract

In the context of the palladium-catalyzed dearomatization of electron-poor arenes, we report herein that various 2-nitrobenzofurans efficiently undergo a dearomative (3+2) cycloaddition with vinylcyclopropanes. This new method gives access to a wide variety of cyclo­penta[b]benzofuran derivatives in a straightforward manner.


#

Dearomatization reactions represent powerful methods for building high molecular complexity and diversity starting from readily available feedstock materials such as arenes and heteroarenes.[1] [2] [3] [4] For this reason, much effort has been devoted to their development, notably by means of robust metal-based catalytic processes.[2] In this particular area of research, palladium catalysis has undeniably shown great potential.[3,4] Interestingly though, the vast majority of such catalytic transformations have so far capitalized on the nucleophilic character of electron-rich aromatic rings (phenols, anilines, indoles, etc.),[3] and the complementary palladium-catalyzed dearomatization of electron-deficient systems remains comparatively scarce.[4] In 2014, Trost et al. reached a significant milestone in this field by demonstrating that several nitroarenes, including 5-nitroquinoline, could undergo a formal (3+2) dearomative cycloaddition with a trimethylenemethane equivalent (Scheme [1, a]).[4a] [5] More recently, we and the group of Hyland demonstrated that such an approach was not limited to the use of trimethylenemethane equivalents and that, under palladium(0) catalysis, vinylcyclopropanes (VCPs) (Scheme [1, b])[4b] [6] and vinylaziridines (Scheme [1, c])[4c] were also 1,3-dipole precursors capable of promoting the dearomatization of N-protected 3-nitroindoles.

Zoom Image
Scheme 1 Palladium(0)-catalyzed dearomatizations of nitroarenes and the aim of this work

As a continuation of our interest in the use of VCPs in palladium-catalyzed dearomatization reactions, we questioned if such a cycloaddition strategy could be applied to nitrobenzofurans. If successful, we anticipated that it would enable an efficient and atom-economical access to the ­cyclopenta[b]benzofuran core, a key scaffold prevalent in several bioactive flavaglines, such as (–)-rocaglamide (Scheme [1, d]).[7] [8] Although several synthetic methods to this structural motif have been developed,[8,9] such a straightforward palladium-catalyzed dearomative (3+2) cycloaddition approach is, to the best of our knowledge, unprecedented.[10]

At the outset of this study, the feasibility of a palladium catalyzed dearomative cycloaddition process between ­nitro­benzofurans and VCPs was surveyed in a model reaction between 5-bromo-2-nitrobenzofuran (1a)[11] and dicyano ­vinylcyclopropane 2a. When using a catalytic system composed of Pd2(dba)3·CHCl3 (2.5 mol%) and 1,2-bis(diphenylphosphino)ethane (dppe) (5 mol%) in toluene at room temperature, we were pleased to observe the formation of the desired diastereoisomeric cyclopenta[b]­benzo­furan 3aa in 1:1.2 dr (Scheme [2]).

Zoom Image
Scheme 2 A preliminary study of the Pd(0)-catalyzed dearomatization of 5-bromo-2-nitrobenzofuran (1a) with VCP 2a

Although the proof of concept had been rapidly achieved, despite full consumption of 2a, we were not able to reach full conversion of 1a as a probable result of the known propensity of VCPs to undergo competitive oligomerization in the presence of a palladium(0) complex.[12] In this context, we decided to optimize the reaction conditions to improve the efficiency of this new dearomative process (Table [1]).

Employing the Pd2(dba)3·CHCl3/dppe catalytic system and fixing the reaction time at 2 hours, the influence of the solvent was examined at room temperature. Switching from toluene to THF or dioxane did not lead to any significant improvement (Table [1], entries 2 and 3). In acetonitrile, full conversion of 1a was obtained but the cyclopenta[b]benzofuran 3aa (1:1 dr) was isolated in 68% yield (entry 4). Indeed, in this case, the concurrent formation of several by-products could be witnessed in the 1H NMR spectrum of the crude reaction mixture. Probing the use of chlorinated solvents (entries 5–7), excellent conversions were achieved in 1,2-dichloroethane and dichloromethane. In the latter case, the cyclopentannelation product 3aa (1:1 dr) was obtained in 88% yield and, as such, dichloromethane was selected for the rest of our study (entry 7). Next, we decided to evaluate whether other phosphorus-based ligands could improve the diastereoselectivity of this cycloaddition reaction. Analogs of dppe such as 1,3-bis(diphenylphosphino)propane (dppp) and bis(diphenylphosphino)methane (dppm) induced a significant decrease of reactivity without improving the stereoselectivity (entries 8 and 9). In the same way, bidentate 1,1′-bis(diphenylphosphino)ferrocene (dppf) and Xantphos, or monodentate triphenyl- and tri-(ortho-tolyl)phosphine led to poor results (entries 10–13). For this reason, dppe was identified as the optimum ligand for this transformation.

Table 1 Optimization of the Reaction Conditions

Entry

Solvent

Ligand (mol%)

Conversion (%)a,b

drc

1

toluene

dppe (5)

77

1:1.2

2

THF

dppe (5)

83

1:1.1

3

dioxane

dppe (5)

74

1:1.1

4

MeCN

dppe (5)

>95 (68)

1:1

5

CHCl3

dppe (5)

89

1:1

6

DCEd

dppe (5)

>95 (73)

1:1

7

CH2Cl2

dppe (5)

>95 (88)

1:1

8

CH2Cl2

dppp (5)

87

1:1

9

CH2Cl2

dppm (5)

27

1:1

10

CH2Cl2

dppf (5)

72

1:1

11

CH2Cl2

Xantphos (5)

<5

12

CH2Cl2

PPh3 (10)

<5

13

CH2Cl2

P(o-Tol)3 (10)

27

1:1.2

a Determined by 1H NMR analysis of the crude reaction mixture using 2-bromo-1,3,5-trimethoxybenzene as an internal standard.

b Yield of isolated product in parentheses.

c Determined by 1H NMR analysis of the crude reaction mixture.

d DCE = 1,2-dichloroethane.

With this optimized catalytic system in hand, we then studied the scope and limitations of this new dearomatization reaction. Employing vinylcyclopropane 2a, the influence of the 2-nitrobenzofuran partner was first surveyed (Scheme [3]).

Using 2-nitrobenzofuran (1b), the corresponding dearomatized cycloadduct 3ba was obtained in 90% yield and 1:1 dr. The reactivity of various 5-substituted 2-nitrobenzofurans was then evaluated. While, in all cases, no significant increase in the diastereoselectivity was observed, the stereoelectronic properties of the substituent present at position 5 did not seem to have a key impact on the overall efficiency of the dearomative cycloaddition process. Accordingly, cycloadducts bearing methyl (3ca), tert-butyl (3da), fluoro (3ea), chloro (3fa), methoxy (3ga) or trifluoromethoxy (3ha) groups were all obtained in satisfactory yields (68–88%). This trend was confirmed when subjecting the 6- and 7-substituted 2-nitrobenzofurans 1il to the reaction conditions. Indeed, the corresponding methyl-, bromo- and methoxy-substituted cyclopenta[b]benzofurans 3iala were isolated in yields ranging from 59–83% and approximately 1:1 dr values. On the other hand, 2-nitrobenzo­furans 1m,n bearing a substituent at position 4 proved to be more sluggish in their reactions with 2a. The corresponding cycloadducts 3ma and 3na were both obtained in only 51% yield, suggesting that an increase of steric hindrance next to the electrophilic 3-position may, to some ­extent, impede the desired cycloaddition process. Whereas the diastereoselectivity was not significantly improved for 3ma (1:1.3 dr), the naphthyl-derived cyclopenta[b]benzofuran 3na was formed in a higher 1:2.3 dr.

Next, the influence of the vinylcyclopropane partner was evaluated by submitting 5-bromo-2-nitrobenzofuran (1a) to Pd(0)-catalyzed reactions with VCPs 2bf (Scheme [4]).

Zoom Image
Scheme 3 Evaluation of various 2-nitrobenzofurans
Zoom Image
Scheme 4 Evaluation of other vinylcyclopropanes

Pleasingly, this novel palladium-catalyzed dearomatization reaction proved to be compatible with the use of VCP diesters 2bd. Accordingly, the corresponding cyclo­penta[b]benzofurans 3abad were obtained in good to ­excellent yields (87–97%) and slightly better diastereoselectivities than in the previous series. Switching to the indan-1,3-dione-derived vinylcyclopropane 2e resulted in a ­significant improvement of the cycloaddition stereo­selectivity (1:4.3 dr), albeit with a low 33% yield under the standard reaction conditions. When employing 2 equivalents of 5-bromo-2-nitrobenzofuran (1a) and a prolonged reaction time of 72 hours, the dearomative cycloaddition process could compete more efficiently with the polymerization of 2e, such that the desired cyclopentannulated product 3ae was obtained in a superior 81% yield (1:4.4 dr).[13] On the other hand, the use of Meldrum’s acid or barbituric acid derived VCPs 2f and 2g did not result in the desired dearomatization reaction. Whereas in the first case no reaction occurred, the degradation of 2g was preferentially observed over the formation of the desired cycloadduct 3ag.

Finally, we decided to test if this palladium-catalyzed dearomative cycloaddition reaction would be compatible with the use of isomeric 3-nitrobenzofurans. For this purpose, the 5-acetoxy derivative 4 was prepared and reacted with VCP 2a.[14] Satisfyingly, we obtained the desired cyclo­addition product 5 in 94% yield (1.6:1 dr), suggesting that this new dearomatization method is not limited to 2-nitrobenzofurans (Scheme [5]).

Zoom Image
Scheme 5 Evaluation of a 3-nitrobenzofuran derivative

In summary, we have herein demonstrated that a variety of 2-nitrobenzofurans undergo a palladium-catalyzed dearomative (3+2) cycloaddition with vinylcyclopropanes. This novel methodology offers an efficient and atom-economical access to a wide range of cyclopenta[b]benzofuran derivatives with average to excellent yields and can also be applied to the use of a 3-nitrobenzofuran.[15] Further studies concerning the development of an enantioselective variant of this dearomatization reaction are currently underway and will be reported in due course.


#

Acknowledgment

We warmly thank Sandra Segondy for her contribution to the synthesis of several 2-nitrobenzofurans.

Supporting Information



Zoom Image
Scheme 1 Palladium(0)-catalyzed dearomatizations of nitroarenes and the aim of this work
Zoom Image
Scheme 2 A preliminary study of the Pd(0)-catalyzed dearomatization of 5-bromo-2-nitrobenzofuran (1a) with VCP 2a
Zoom Image
Scheme 3 Evaluation of various 2-nitrobenzofurans
Zoom Image
Scheme 4 Evaluation of other vinylcyclopropanes
Zoom Image
Scheme 5 Evaluation of a 3-nitrobenzofuran derivative