Download PDF
Synthesis 2013; 45(13): 1807-1814
DOI: 10.1055/s-0033-1338876
paper
© Georg Thieme Verlag Stuttgart · New York

Diastereoselective Syntheses of Highly Substituted Methylenecyclopropanes via Copper- or Iron-Catalyzed Reactions of 1,2-Disubstituted 3-(Hydroxy­methyl)cyclopropenes with Grignard Reagents

Xiaocong Xie
Brown Laboratories, Department of Chemistry and Biochemistry, University of Delaware, Newark DE 19716, USA   Fax: +1(302)8316335   Email: jmfox@udel.edu
,
Joseph M. Fox*
Brown Laboratories, Department of Chemistry and Biochemistry, University of Delaware, Newark DE 19716, USA   Fax: +1(302)8316335   Email: jmfox@udel.edu
› Author Affiliations
Further Information

Publication History

Received: 08 April 2013

Accepted after revision: 08 May 2013

Publication Date:
06 June 2013 (online)

 
 PDF Download Permissions and Reprints All articles of this category


Dedicated to Scott Denmark in recognition of his contributions to the chemistry of strained molecules

Abstract

Described are diastereoselective syntheses of highly substituted methylenecyclopropanes from 1,2-disubstituted 3-(hydroxymethyl)cyclopropenes with allylic ether leaving groups, compounds that were constructed via alkylation of cyclopropenecarboxylic acid dianions. Substitution reactions of 1,2-disubstituted 3-(hydroxymethyl)cyclopropenes with Grignard reagents proceed with good yield and high diastereoselectivity. Under copper(I)-catalyzed conditions, the substitution reactions proceed to give syn-addition products, whereas Fe(acac)3 catalysis gives the products of anti-addition.


#

Key words

cyclopropene - diastereoselective - dianion - methylenecyclopropane - substitution

Methylenecyclopropanes are highly strained molecules (39.5 kcal/mol)[1] that serve as useful building blocks for the assembly of complex molecular structures.[2] Work in recent years has established the ability of methylenecyclopropanes to participate in ring-opening,[3] cycloaddition,[4] ring-expansion,[5] and carbon–carbon double-bond addition reactions.[6] Furthermore, a number of the transformations of methylenecyclopropanes have shown to be stereospecific.[3f] [4e] [f] [7]

As a consequence of this rich reactivity, there has been a growing interest in the diastereoselective synthesis of methylenecyclopropanes from cyclopropene precursors. Inspiration for this approach to methylenecyclopropane synthesis was provided by de Meijere, who reported that palladium(0)-catalyzed nucleophilic SN2′ substitution could form an alkylidenecyclopropane in low yield.[8] Contemporaneously, Marek,[9] Corey,[10] Rubin,[11] and our group[12] showed that common cyclopropene precursors can be used to provide access to a range of highly substituted methylenecyclopropanes or alkylidenecyclopropanes. Marek showed that 1-(1-hydroxyalkyl)cycloprop-1-enes can be resolved by Sharpless epoxidation, and then treated with alkyl Grignard reagents to give alkylidenecyclopropanes or methylenecyclopropanes in good yield, diastereoselectivity and E/Z selectivity.[9a] [b] Marek further described copper-catalyzed carbomagnesation and hydrometalation reactions of chiral cyclopropenylcarbinol derivatives, including formal SN2′ reactions to set chiral quaternary centers.[9c] Marek also showed that 2-methylcycloprop-1-enyl(phenyl)methyl acetate can undergo sigmatropic­ rearrangement to give an alkylidenecyclopropane stereospecifically.[9d] Alternatively, Corey and co-workers have shown that 2-(bromomethyl)-1-tosylcycloprop-2-ene could be synthesized in 94% ee and treated with nucleophiles to provide methylenecyclopropanes with excellent diastereoselectivity.[10] Rubin and co-workers have found that methylenecyclopropylphosphine oxides could be synthesized with excellent diastereocontrol from aryl-1-(hydroxymethyl)cycloprop-2-enes.[11] It has also been established that methylenecyclopropanes can be generated by the base-catalyzed isomerization of 1-methylcyclopropene precursors.[13]

In earlier studies, our group had shown that Grignard reagents can convert 1-(alkoxymethyl)cyclopropenes into methylenecyclopropanes, as illustrated in Scheme [1] (a).[12] The regioselectivity of substitution reactions of Grignard reagents with (alkoxymethyl)cyclopropenes contrasted prior studies on carbomagnesation of 1-alkylcyclopropenes, where the normal course of reactivity is for the nucleophile to add to the more substituted side of the double bond.[14] The reaction to form methylenecyclopropanes proceeded with excellent regio- and diastereoselectivity. Methyl-, alkyl- (1° or 2°), allyl-, and benzylmagnesium halides were suitable nucleophiles, and MEM or SEM ethers served as effective leaving groups.[12] Herein, we describe studies that expand the scope of methylenecyclopropane formation via facially selective substitution reactions of 1-(alkoxymethyl)cyclopropenes with Grignard­ reagents.

Zoom Image
Scheme 1 Substitution reactions of cyclopropenes with allylic ether leaving groups

The diastereoselective synthesis of methylenecyclopropanes from cyclopropenes had been limited to examples of ‘terminal’ cyclopropene precursors. We sought to develop diastereoselective substitution reactions of ‘internal’ cyclopropenes[15] with Grignard reagents. Challenges for developing such reactions would include the synthesis of 1,2-disubstituted cyclopropenes with appropriate allylic leaving groups, as well as the avoidance of competing carbometalation products that would result from addition of the Grignard reagent with the opposite regioselectivity [Scheme [1] (b)]. Recently, Davies has reported an enantio­selective method for the cyclopropenation of internal alkynes using gold catalysis.[16] Complementary methods that convert terminal cyclopropenes into internal cyclopropenes have been reported by Eckert-Maksić,[17] Gevorgyan­,[18] and Lam.[19] Our group has developed dianion approaches to convert terminal cyclopropenes into internal cyclopropenes,[13a] [20] in which terminal cyclo­-propenes were treated with two equivalents of base to provide stable dianions that can subsequently combine with electrophiles to give internal cyclopropenes. Building on this dianion strategy, we envisioned the strategy for diastereoselective methylenecyclopropane formation shown in Scheme [2]. Cyclopropenecarboxylic acids of structure A are readily available in racemic or in enantiomerically enriched form through enantioselective cyclopropenation of alkynes.[21] Treatment of A with an organolithium would produce the dianion B, which could be alkylated by MOMCl to give ester C. Reduction to D would be followed by diastereoselective substitution to give E.

Zoom Image
Scheme 2 Dianion alkylation/substitution as a route to methylenecyclopropanes

The use of dianion chemistry to prepare (methoxymethyl)cyclopropenes is outlined in Scheme [3]. Cyclopropenecarboxylic acid 1 [20] was sequentially treated with 2.2 equivalents of MeLi and 5 equivalents MOMCl. Without chromatography, the crude reaction mixture was then treated with TMSCHN2 to give the desired methyl ester 2 and the MOM-substituted ester (~1: 1). The crude mixture of esters was then reduced by DIBAL-H to give alcohol 3 in 46% yield over three steps. Similar procedures were applied to 4 and 6 to give 5 and 7 in 38% and 50% yield over three steps, respectively.

Zoom Image
Scheme 3 Synthesis of 1,2-disubstituted 3-(hydroxymethyl)cyclopropenes with allylic leaving groups
Zoom Image
Scheme 4 Copper(I)-catalyzed addition of Grignard reagents to provide methylenecyclopropanes bearing quaternary centers

Substitution reactions of internal cyclopropenes 3, 5, and 7 with Grignard reagents were catalyzed by CuI in 61–89% yield (Scheme [4]). With MeMgBr, compounds 8 and 9 were formed with 12:1 and >95:5 diastereoselectivity, respectively, for the product of Grignard reagent addition syn relative to the hydroxymethyl group. With 2-(1,3-dioxan-2-yl)ethylmagnesium bromide, compound 10 was formed from 5 in 71% yield with a dr of 3:1. Allylmagnesium bromide with 5 gave 11 in 86% yield with dr of 2:1. Methylenecyclopropanes 12 and 13 could be prepared with excellent selectivity for the syn isomers from cyclopropenes 3 and 7, respectively. The higher selectivity with 3 and 7 is attributed to the influence of their C3 substituents, which accentuate the directing ability of the deprotonated hydroxymethyl group by blocking addition to the opposite face of the cyclopropene.

Additions of aryl Grignard reagents were also investigated and CuBr·SMe2 was found to be the catalyst of choice (Scheme [4]). Several aryl Grignard reagents were studied and found to give methylenecyclopropanes with good yields. Cyclopropenes 3 and 7 gave methylenecyclopropanes 14 and 15, respectively, with high yields (75–91%) and diastereoselectivity (>95:5 dr). The substitution reaction of 1-hexyl-3-(hydroxymethyl)cyclopropene 5 with 4-fluorophenylmagnesium bromide gave 16 in 68% yield and 10:1 dr, and with 2-naphthylmagnesium bromide gave 17 in 48% yield and 14:1 dr.

Zoom Image
Scheme 5 Reversal of facial selectivity in Fe(acac)3-catalyzed addition of aryl Grignard reagents to 3-(hydroxymethyl)cyclopropenes

It was also observed that iron catalysts could be used to catalyze the substitution reaction of Grignard reagents to form methylenecyclopropanes. Previously, Nakamura had shown that FeCl3 could be used to catalyze the addition reactions of Grignard reagents to cyclopropenone acetals­.[22] In the present study, it was found that iron(III)-catalyzed substitution reactions to form methylenecyclopropanes can proceed with the opposite sense of diaste­reoselectivity relative to copper(I) catalyzed reactions (Scheme [5]). Fe(acac)3 was found to be the most selective catalyst and aromatic Grignard reagents were the most selective nucleophiles.

The Fe(acac)3 catalyzed reaction of 4-fluorophenylmagnesium bromide with 5 gave 18 in 73% yield with an anti/­syn ratio of 8:1. o-Tolylmagnesium bromide addition to 5 gave the anti diastereomer 19 in 10:1 dr and 56% yield. Also studied were the iron-catalyzed reactions between 4-fluorophenylmagnesium bromide with 7 and methylmagnesium bromide with 3. Both 7 and 3 are disubstituted at C3 of the cyclopropene. For these substrates, where both faces are substituted at C3, the substitutions took place exclusively syn to the hydroxymethyl group to give 15 and 20 (Scheme [6]).

Zoom Image
Scheme 6 Iron-catalyzed carbomagnesation takes place on the syn-face with cyclopropenes that are disubstituted at C3
Zoom Image
Figure 1 Shielding effects in methylenecyclopropanes with aromatic substituents support stereochemical assignments

In support of the stereochemical assignments, there are significant shielding effects that are observed in the 1H NMR spectra of aryl-substituted methylenecyclopropanes. For example, the diastereotopic protons on the methylene of the hydroxymethyl for 18 resonate at δ = 3.94 and 3.77, whereas the analogous protons on isomeric compound 16 resonate at δ = 3.43 and 3.12 (Figure [1]). The upfield chemical shifts for 16 are consistent with the syn relationship between the aryl substituent and the hydroxymethyl group. Likewise, the methylene protons on the hydroxymethyl group of compounds 14 (δ = 3.65, 3.19), 15 (δ = 3.18, 3.06), and 17 (δ = 3.40, 3.13) also resonate at high field. For each of compounds 12, 14, and 20, one of the diastereotopic protons on the butyl group is shielded by the phenyl on the same face of the cyclopropane, resulting in shift to high field (δ = 0.66–0.85). As noted previously, a similar effect is observed for compound 21,[14b] where the analogous proton resonates at δ = 0.65 (Figure [1]).

In conclusion, diastereoselective syntheses of highly substituted methylenecyclopropanes from chiral, 1,2-disubstituted cyclopropenes have been developed. Dianion chemistry was used to construct cyclopropene precursors with allylic ether leaving groups. Substitution reactions 3-(hydroxymethyl)cycloprop-1-enes with Grignard reagents can be catalyzed by copper(I) catalysts to give syn-addition products, or with Fe(acac)3 to give anti-addition products.

All reactions were carried out in round-bottom flasks that were flame-dried and cooled under N2. All commercially available reagents were used as received. THF was distilled from Na/benzophenone. Hexanes and Et2O were dried with columns prepacked with activated neutral alumina. Chromatography was carried out using Silicycle silica gel (40–63D, 60Å). For 13C NMR, multiplicities were distinguished using a ATP pulse sequence: typical methylene and quaternary carbons appear ‘up’; methine and methyl carbons ‘down’.


#

Methylenecyclopropane Synthesis; General Procedure A

A soln of 2.0 M cyclopropenecarboxylic acid in THF was added to a round-bottomed flask. The mixture was cooled to –78 °C and stirred for 5 min. MeLi (2.2 equiv) was added and the mixture was stirred for a further 5 min. The acetone/dry ice bath was replaced with an ice bath and the mixture was allowed to warm to 0 °C over 5 min. MOMCl (5 equiv) was added and the mixture was allowed warm to r.t. while stirring for 1 h. The mixture was quenched with H2O (5 mL) and extracted with EtOAc (3 × 10 mL). The combined organic layers were dried (anhyd MgSO4), then filtered and concentrated in vacuo to afford the crude product as a brown oil. The crude product was dissolved in MeOH (~0.2 mmol/mL) and transferred to a round-bottom flask and cooled to 0 °C with an ice bath. TMSCHN2 was added dropwise until N2 evolution had ceased (~5 equiv). H2O (5 mL) was added to the mixture, which was then extracted with EtOAc (3 × 10 mL). The combined organic layers were dried (anhyd MgSO4), filtered, and concentrated in vacuo to afford the crude product as a yellow oil. The crude product was dissolved in anhyd THF (~0.1 mmol/mL) and transferred to an oven-dried round-bottom flask. The mixture was cooled to –78 °C and stirred for 5 min. DIBAL-H (5 equiv) was added via syringe over the course of 10 min and the mixture was stirred at –78 °C for 2 h. EtOAc­ (10 mL) was added followed by 3 M HCl (10 mL), and the mixture was allowed to warm to r.t. The mixture was extracted with EtOAc (3 × 10 mL). The combined organic layers were dried (anhyd MgSO4), then filtered and concentrated in vacuo. The residue was purified by chromatography (silica gel, 50% EtOAc–hexanes) to give the desired 3-(hydroxymethyl)cycloprop-1-ene.


#

Grignard Reagent to 3-(Hydroxymethyl)cycloprop-1-enes Catalyzed by Cu(I) or Fe(acac)3; General Procedure B

To a soln of the (hydroxymethyl)cyclopropene derivative in hexanes/Et2O (amounts specified below) was added the specified Cu(I) or Fe(III) catalyst (10 mol%). The Grignard reagent (5 equiv) was added via syringe and the mixture was allowed to stir at r.t. for 16 h. The reaction was quenched with sat. NH4Cl soln and extracted with EtOAc (3 × 10 mL). The combined organic layers were washed with brine and dried (anhyd MgSO4), then filtered and concentrated in vacuo to afford the crude product as a colorless to yellow oil. In the case of aryl Grignard reagents, the combined organic layers were additionally washed with 10% KOH (3 × 10 mL). The crude product was purified by chromatography (silica gel, 10% EtOAc–hexanes) to give the desired methylenecyclopropane.


#

2-Butyl-1-(hydroxymethyl)-1-phenyl-3-(methoxymethyl)cycloprop-2-ene (3)

General procedure A was followed with 2-butyl-1-phenylcycloprop-2-enecarboxylic acid (1,[20] 251 mg, 1.16 mmol), 1.6 M MeLi in Et2O (1.6 mL, 2.6 mmol), and MOMCl (467 mg, 5.80 mmol) in THF (7.0 mL); 2.0 M TMSCHN2 in hexanes (2.9 mL, 5.8 mmol) in MeOH (10 mL); and finally 1.0 M DIBAL-H in THF (5.8 mL, 5.80 mmol) in THF (6.0 mL) to give 3 (130 mg, 0.528 mmol, 46%) as a colorless oil.

Small peaks attributed to inseparable impurities were found in the 1H NMR at δ = 4.72, 3.90, 3.37, 2.90, and in the 13C NMR at δ = 129.0, 63.7, 59.6.

IR (neat): 3440, 3058, 3024, 2957, 2930, 2872, 1600, 1494, 1465, 1448, 1377, 1192, 1153, 1098, 1006, 769, 700 cm–1.

1H NMR (400 MHz, CDCl3): δ = 7.25–7.33 (m, 4 H), 7.15–7.19 (m, 1 H), 4.49 (d, J = 13.3 Hz, 1 H), 4.40 (d, J = 11.2 Hz, 1 H), 4.32 (dd, J = 13.3, 1.2 Hz, 1 H), 3.69 (d, J = 11.2 Hz, 1 H), 3.44 (s, 3 H), 2.45–2.61 (m, 2 H), 1.89 (s, 1 H), 1.54–1.61 (m, 2 H), 1.36–1.45 (m, 2 H), 0.93 (t, J = 7.3 Hz, 3 H).

13C NMR (100 MHz, CDCl3): δ = 145.4 (C), 128.2 (CH), 126.2 (CH), 125.2 (CH), 118.7 (C), 110.5 (C), 68.3 (CH2), 64.6 (CH2), 58.8 (CH3), 34.5 (C), 29.6 (CH2), 24.0 (CH2), 22.5 (CH2), 13.8 (CH3).

HRMS-CI (NH3): m/z [M – OH] calcd for C16H21O: 229.1587; found: 229.1595.


#

2-Hexyl-1-(hydroxymethyl)-3-(methoxymethyl)cycloprop-2-ene (5)

General procedure A with 2-hexylcycloprop-2-enecarboxylic acid (4,[20] 251 mg, 1.49 mmol), 1.6 M MeLi in Et2O (2.1 mL, 3.4 mmol ), and MOMCl (601 mg, 7.47 mmol) in THF (9.0 mL); 2.0 M TMSCHN2 in hexanes (3.7 mL, 7.5 mmol) in MeOH (15 mL); and finally 1.0 M DIBAL-H in THF (7.5 mL, 7.5 mmol) in THF (7.5 mL) gave 5 (113 mg, 0.570 mmol, 38%) as a colorless oil.

Peaks attributable to inseparable impurities (~6%) were found in the 1H NMR at δ = 4.73, 3.49, 3.36, 1.07.

IR (neat): 3401, 2929, 2858, 1465, 1378, 1190, 1101, 1018 cm–1.

1H NMR (400 MHz, CDCl3): δ = 4.43 (d, J = 13.6 Hz, 1 H), 4.29 (d, J = 13.6 Hz, 1 H), 3.77 (dd, J = 10.8, 3.8 Hz, 1 H), 3.45 (s, 3 H), 3.23 (dd, J = 10.8, 6.1 Hz, 1 H), 2.43–2.47 (m, 2 H), 1.80–1.82 (m, 2 H), 1.52–1.59 (m, 2 H), 1.28–1.39 (m, 6 H), 0.91 (t, J = 6.8 Hz, 3 H).

13C NMR (100 MHz, CDCl3): δ = 118.6 (C), 110.7 (C), 68.7 (CH2), 66.3 (CH2), 58.6 (CH3), 31.6 (CH2), 29.0 (CH2), 27.5 (CH2), 25.8 (CH2), 23.6 (CH), 22.6 (CH2), 14.1 (CH3).

HRMS-CI (NH3): m/z [M – OH] calcd for C12H21O: 181.1587; found: 181.1593.


#

2-Hexyl-1-(hydroxymethyl)-1-methyl-3-(methoxymethyl)cycloprop-3-ene (7)

General procedure A was followed with 2-hexyl-1-methylcycloprop-2-enecarboxylic acid (6,[20] 563 mg, 3.09 mmol), 1.6 M MeLi in Et2O (4.3 mL, 6.9 mmol), and MOMCl (1.244 g, 15.45 mmol) in THF (16 mL); 2.0 M TMSCHN2 in hexanes (7.7 mL, 15 mmol) in MeOH (5.0 mL); and finally 1.0 M DIBAL-H in THF (15.5 mL, 15.5 mmol) in THF (15 mL) to give 7 (324 mg, 1.53 mmol, 50%) as a colorless oil.

Small peaks attributed to an inseparable impurity (5%) were found in the 1H NMR at δ = 4.71, 3.41, and in the 13C NMR at δ = 95.9, 60.4.

IR (neat): 3430, 2929, 2858, 1455, 1368, 1190, 1101, 1054, 1011, 914 cm–1.

1H NMR (400 MHz, CDCl3): δ = 4.38 (dt, J = 13.5, 0.9 Hz, 1 H), 4.28 (dt, J = 13.5, 1.8 Hz, 1 H), 3.67 (d, J = 10.6 Hz, 1 H), 3.42 (s, 3 H), 3.36 (d, J = 10.6 Hz, 1 H), 2.41–2.45 (m, 2 H), 1.70 (br s, 1 H), 1.50–1.57 (m, 2 H), 1.27–1.40 (m, 6 H), 1.16 (s, 3 H), 0.90 (t, J = 6.8 Hz, 3 H).

13C NMR (100 MHz, CDCl3): δ = 124.4 (C), 115.9 (C), 70.1 (CH2), 65.7 (CH2), 58.5 (CH3), 31.6 (CH2), 29.1 (CH2), 27.6 (CH2), 27.6 (C), 25.0 (CH2), 22.6 (CH2), 20.5 (CH3), 14.1 (CH3).

HRMS-CI (NH3): m/z [M – OH] calcd for C13H23O: 195.1743; found: 195.1748.


#

3β-Hexyl-2α-(hydroxymethyl)-3α-methyl-1-methylenecyclopropane (8)

General procedure B with 5 (41 mg, 0.21 mmol), CuI (4 mg, 0.02 mmol), and 3.0 M MeMgBr in Et2O (0.35 mL, 1.1 mmol) in hexanes (2 mL) and Et2O (1 mL) gave 8 (34 mg, 0.19 mmol, 90%) as a pale yellow oil. A similar experiment with 5 (20 mg) gave 8 (16 mg, 87%).

IR (neat): 3326, 2958, 2928, 2857, 1465, 1378, 1142, 1014, 886 cm–1.

1H NMR (400 MHz, CDCl3): δ = 5.34 (dd, J = 2.4, 0.7 Hz, 1 H), 5.29 (app s, 1 H), 3.80 (dd, J = 11.6, 6.0 Hz, 1 H), 3.58 (dd, J = 11.6, 9.0 Hz, 1 H), 1.52–1.56 (m, 1 H), 1.28–1.45 (m, 11 H), 1.17 (s, 3 H), 0.90 (t, J = 7.0 Hz, 3 H).

13C NMR (100 MHz, CDCl3): δ = 144.3 (C), 101.2 (CH2), 61.5 (CH2), 39.1 (CH2), 31.4 (CH2), 28.9 (CH2), 28.4 (CH), 26.3 (CH2), 23.8 (C), 22.2 (CH2), 15.5 (CH3), 13.7 (CH3).

HRMS-CI (NH3): m/z [M – OH] calcd for C12H21: 165.1638; found: 165.1642.


#

3β-Hexyl-2α-(hydroxymethyl)-2β,3α-dimethyl-1-methylenecyclopropane (9)

General procedure B with 7 (27 mg, 0.13 mmol), CuI (2 mg, 0.01 mmol), and 3.0 M MeMgBr in Et2O (0.22 mL, 0.66 mmol ) in hexanes (1.3 mL) and Et2O (0.6 mL) gave 9 (19 mg, 0.10 mmol, 76%) as a pale yellow oil. A similar experiment with 7 (25 mg) gave 9 (16 mg, 70%). Small peaks attributed to an inseparable impurity were found in the 1H NMR at δ = 1.19–1.22.

IR (neat): 3352, 3060, 2985, 2927, 2858, 1446, 1377, 1015, 937, 886, 724, 603 cm–1.

1H NMR (400 MHz, CDCl3): δ = 5.23 (s, 1 H), 5.20 (s, 1 H), 3.65 (d, J = 11.7 Hz, 1 H), 3.59 (d, J = 11.7 Hz, 1 H), 1.56 (s, 1 H), 1.25–1.47 (m, 13 H), 1.22 (s, 3 H), 0.90 (t, J = 7.2 Hz, 3 H).

13C NMR (100 MHz, CDCl3): δ = 150.4 (C), 99.8 (CH2), 67.3 (CH2), 34.9 (CH2), 31.9 (CH2), 29.6 (CH2), 28.5 (C), 27.0 (CH2), 26.4 (C), 22.7 (CH2), 17.6 (CH3), 16.2 (CH3), 14.1 (CH3).

HRMS-CI (NH3): m/z [M – OH] calcd for C13H23: 179.1794; found: 179.1802.


#

3α-[2-(1,3-Dioxan-2-yl)ethyl]-3β-hexyl-2α-(hydroxymethyl)-1-methylenecyclopropane (10)

General procedure B with 5 (30 mg, 0.15 mmol), CuI (4 mg, 0.02 mmol), and 0.5 M (1,3-dioxan-2-ylethyl)magnesium bromide in THF (1.5 mL, 0.75 mmol) in hexanes (1.4 mL) gave 10 (32 mg, 0.11 mmol, 75%) as a pale yellow oil. A similar experiment with 5 (55 mg) gave 10 (52 mg, 66%).

Peaks attributable to the diastereomer of 10 (dr 3:1) were detected in the 1H NMR at δ = 5.34, 5.29, 3.60, and in the 13C NMR at δ = 143.5, 101.8, 61.0, 34.7, 31.7, 29.5, 29.2, 28.8, 27.3, 26.5, 25.4. Small peaks attributed to inseparable impurities were found in the 1H NMR at δ = 3.69, 3.30, 0.91.

IR (neat): 3433, 3062, 2926, 2853, 2731, 2658, 1467, 1431, 1403, 1379, 1286, 1241, 1145, 1082, 1004, 944, 926, 887, 853, 728, 641, 583, 529 cm–1.

1H NMR (400 MHz, CDCl3): δ = 5.32 (d, J = 1.8 Hz, 1 H), 5.26 (app s, 1 H), 4.51–4.56 (m, 1 H), 4.09–4.13 (m, 2 H), 3.73–3.82 (m, 3 H), 3.53–3.59 (m, 1 H), 2.03–2.07 (m, 1 H), 1.81–1.84 (m, 1 H), 1.26–1.80 (m, 16 H), 0.88 (t, J = 7.0 Hz, 3 H).

13C NMR (100 MHz, CDCl3): δ = 143.6 (C), 101.6 (CH), 101.5 (CH2), 66.5 (CH2), 61.0 (CH2), 35.3 (CH2), 32.3 (CH2), 31.4 (CH2), 29.1 (CH), 28.9 (CH2), 27.8 (C), 25.9 (CH2), 25.3 (CH2), 22.6 (CH2), 22.2 (CH2), 13.7 (CH3).

HRMS-CI (NH3): m/z [M + H] calcd for C17H31O3: 283.2267; found: 283.2263.


#

3α-Allyl-3β-hexyl-2α-(hydroxymethyl)-1-methylenecyclopropane (11)

General procedure B with 5 (41 mg, 0.21 mmol), CuI (4 mg, 0.02 mmol), and 1.0 M allylmagnesium bromide in Et2O (1.1 mL, 1.1 mmol) in hexanes (3.0 mL) and Et2O (1.5 mL) gave 11 (35 mg, 0.17 mmol, 81%) as a pale yellow oil. A similar experiment with 5 (19 mg) gave 11 (18 mg, 90%).

Peaks attributable to the diastereomer of 11 (dr 2:1) were found in the 1H NMR at δ = 5.74–5.85, 5.39, 5.33, 4.98–5.03, 3.74–3.79, 3.62–3.66, 2.23–2.29, 2.08–2.12, and in the 13C NMR at δ = 143.2, 135.7, 116.7, 102.5, 61.5, 40.1, 31.8, 29.7, 29.5, 28.6, 26.8. Peaks attributable to an inseparable impurity were found in the 1H NMR at δ = 5.00, 0.65, 0.50, and in the 13C NMR at δ = 32.8, 29.0.

IR (neat): 3328, 3076, 2958, 2928, 2857, 1640, 1466, 1439, 1140, 1015, 912, 888 cm–1.

1H NMR (400 MHz, CDCl3): δ = 5.86–5.96 (m, 1 H), 5.38 (m, 1 H), 5.31 (m, 1 H), 5.05–5.16 (m, 2 H), 3.81 (dd, J = 11.8, 5.8 Hz, 1 H), 3.58 (dd, J = 11.8, 9.2 Hz, 1 H), 2.29–2.35 (m, 1 H), 2.14–2.20 (m, 1 H), 1.53–1.65 (m, 2 H), 1.28–1.42 (m, 10 H), 0.90 (t, J = 7.0 Hz, 3 H).

13C NMR (100 MHz, CDCl3): δ = 143.3 (C), 136.7 (CH), 116.4 (CH2), 102.3 (CH2), 61.6 (CH2), 35.9 (CH2), 34.2 (CH2), 31.8 (CH2), 29.3 (CH2), 29.2 (CH), 27.9 (C), 26.3 (CH2), 22.6 (CH2), 14.1 (CH3).

HRMS-CI (NH3): m/z [M – OH] calcd for C14H23: 191.1794; found: 191.1803.


#

3α-Allyl-3β-butyl-2α-(hydroxymethyl)-1-methylene-2β-phenylcyclopropane (12)

General procedure B with 3 (34 mg, 0.14 mmol), CuI (2 mg, 0.01 mmol), and 1 M allylmagnesium bromide in Et2O (0.7 mL, 0.7 mmol) in hexanes (1.4 mL) and Et2O (0.7 mL) gave 12 (21 mg, 0.082 mmol, 59%) as a pale yellow oil. A similar experiment with 3 (34 mg) gave 12 (22 mg, 62%).

IR (neat): 3413, 3060, 3024, 2956, 2929, 2860, 1639, 1600, 1493, 1446, 1379, 1146, 1026, 912, 890, 763, 701, 620, 555, 527 cm–1.

1H NMR (400 MHz, CDCl3): δ = 7.25–7.36 (m, 5 H), 5.98–6.05 (m, 1 H), 5.69 (s, 1 H), 5.54 (s, 1 H), 5.15–5.22 (m, 2 H), 4.08 (d, J = 11.6 Hz, 1 H), 3.81 (d, J = 11.6 Hz, 1 H), 2.62 (dd, J = 15.2, 7.0 Hz, 1 H), 2.31 (dd, J = 15.2, 6.6 Hz, 1 H), 1.03–1.44 (m, 6 H), 0.75 (t, J = 7.3 Hz, 3 H), 0.62–0.69 (m, 1 H).

13C NMR (100 MHz, CDCl3): δ = 145.7 (C), 138.7 (C), 136.1 (CH), 129.4 (CH), 127.8 (CH), 126.3 (CH), 116.2 (CH2), 102.9 (CH2), 66.6 (CH2), 38.6 (C), 34.5 (CH2), 32.5 (C), 31.6 (CH2), 28.1 (CH2), 22.2 (CH2), 13.5 (CH3).

HRMS-CI (NH3): m/z [M – OH] calcd for C18H23: 239.1794; found: 239.1789.


#

3α-Allyl-3β-hexyl-2α-(hydroxymethyl)-2β-methyl-1-methyl­ene­cyclopropane (13)

General procedure B with 7 (38 mg, 0.18 mmol), CuI (4 mg, 0.02 mmol), and 1.0 M MeMgBr in Et2O (0.90 mL, 0.90 mmol ) in hexanes (2 mL) and Et2O (1 mL) gave 13 (27 mg, 0.12 mmol, 68%) as a pale yellow oil. A similar experiment with 7 (12 mg) gave 13 (9 mg, 67%).

IR (neat): 3366, 3075, 2956, 2927, 2858, 1639, 1466, 1378, 1015, 912, 888 cm–1.

1H NMR (400 MHz, CDCl3): δ = 5.88–5.98 (m, 1 H), 5.28 (s, 1 H), 5.25 (s, 1 H), 5.09–5.14 (m, 2 H), 3.66 (d, J = 11.8 Hz, 1 H), 3.62 (d, J = 11.8 Hz, 1 H), 2.41 (dd, J = 15.4, 5.7 Hz, 1 H), 2.23 (dd, J = 15.4, 8.0 Hz, 1 H), 1.28–1.58 (m, 14 H), 0.91 (t, J = 6.6 Hz, 3 H).

13C NMR (100 MHz, CDCl3): δ = 148.9 (C), 137.2 (CH), 116.1 (CH2), 100.6 (CH2), 66.8 (CH2), 35.2 (CH2), 31.8 (CH2), 31.0 (CH2), 30.1 (C), 29.5 (CH2), 29.2 (C), 26.8 (CH2), 22.7 (CH2), 16.6 (CH3), 14.1 (CH3).

HRMS-CI (NH3): m/z [M – OH] calcd for C15H25: 205.1951; found: 205.1951.


#

3β-Butyl-2α-(hydroxymethyl)-1-methylene-2β,3α-diphenylcyclopropane (14)

General procedure B with 3 (27 mg, 0.11 mmol), CuBr·SMe2 (3 mg, 0.01 mmol) and 3.0 M PhMgBr in Et2O (0.18 mL, 0.55 mmol) in hexanes (1.0 mL) gave 14 (27 mg, 0.092 mmol, 84%) as a pale yellow oil. A similar experiment with 3 (6 mg) gave 14 (7 mg, 98%).

IR (neat): 3384, 3059, 3025, 2957, 2930, 2958, 1600, 1494, 1447, 1379, 1027, 896, 762, 702 cm–1.

1H NMR (400 MHz, CDCl3): δ = 7.26–7.52 (m, 10 H), 5.95 (s, 1 H), 5.90 (s, 1 H), 3.65 (d, J = 11.7 Hz, 1 H), 3.19 (d, J = 11.7 Hz, 1 H), 1.81–1.88 (m, 1 H), 0.99–1.16 (m, 5 H), 0.80–0.90 (m, 1 H), 0.69 (t, J = 7.0 Hz, 3 H).

13C NMR (100 MHz, CDCl3): δ = 143.4 (C), 139.4 (C), 138.2 (C), 129.3 (CH), 128.4 (CH), 127.9 (CH), 127.9 (CH), 126.5 (CH), 126.2 (CH), 105.3 (CH2), 67.4 (CH2), 39.9 (C), 38.1 (C), 35.9 (CH2), 28.8 (CH2), 22.0 (CH2), 13.4 (CH3).

HRMS-CI (NH3): m/z [M – OH] calcd for C21H23: 275.1794; found: 275.1791.


#

3α-(4-Fluorophenyl)-3β-hexyl-2α-(hydroxymethyl)-2β-methyl-1-methylenecyclopropane (15)

General procedure B with 7 (34 mg, 0.16 mmol), CuBr·SMe2 (7 mg, 0.03 mmol), and 1.0 M 4-fluorophenylmagnesium bromide in THF (0.8 mL, 0.80 mmol) in hexanes (1.6 mL) gave 15 (32 mg, 0.12 mmol, 72%) as a pale yellow oil. A similar experiment with 7 (16 mg) gave 15 (16 mg, 77%).

Alternatively, general procedure B with 7 (16 mg, 0.075 mmol), Fe(acac)3 (3 mg, 0.01 mmol), and 1.0 M 4-fluorophenylmagnesium bromide in THF (0.38 mL, 0.38 mmol) in hexanes (1 mL) and Et2O (0.5 mL) gave 15 (14 mg, 0.051 mmol 67%) as a pale yellow oil.

IR (neat): 3369, 3064, 2924, 2857, 1604, 1507, 1465, 1379, 1222, 1157, 1107, 1094, 1016, 893, 841, 815, 733, 608, 534 cm–1.

1H NMR (400 MHz, CDCl3): δ = 7.22–7.27 (m, 2 H), 6.94–6.99 (m, 2 H), 5.61 (s, 1 H), 5.49 (s, 1 H), 3.18 (dd, J = 11.5, 6.0 Hz, 1 H), 3.06 (dd, J = 11.5, 5.0 Hz, 1 H), 1.99–2.05 (m, 1 H), 1.50–1.56 (m, 1 H), 1.43 (s, 3 H), 1.04–1.26 (m, 9 H), 0.83 (t, J = 6.8 Hz, 3 H).

13C NMR (100 MHz, CDCl3): δ = 161.5 (C) [d, 1 J(CF) = 244 Hz], 146.6 (C), 136.5 (C) [d, 4 J(CF) = 3 Hz], 130.4 (CH) [d, 3 J(CF) = 8 Hz], 115.0 (CH) [d, 2 J(CF) = 21 Hz], 103.4 (CH2), 67.2 (CH2), 35.5 (C), 35.0 (CH2), 31.7 (CH2), 31.2 (C), 29.2 (CH2), 27.3 (CH2), 22.6 (CH2), 15.6 (CH3), 14.0 (CH3).

HRMS-CI (NH3): m/z [M – OH] calcd for C18H24F: 259.1857; found: 259.1860.


#

3α-(4-Fluorophenyl)-3β-hexyl-2α-(hydroxymethyl)-1-methyl­enecyclopropane (16)

General procedure B with 5 (21 mg, 0.11 mmol), CuBr·SMe2 (2 mg, 0.01 mmol), and 1.0 M 4-fluorophenylmagnesium bromide in THF (0.55 mL, 0.55 mmol) in hexanes (1.0 mL) and Et2O (0.5 mL) gave 16 (18 mg, 0.069 mmol, 65%) as a pale yellow oil. A similar experiment with 5 (27 mg) gave 16 (25 mg, 70%).

Small peaks attributed to the minor diastereomer of 16 were found in the 1H NMR at δ = 7.44–7.51, 5.68, 5.51, 3.94, 3.76, 2.02, 1.85, 0.90 and in the 13C NMR at δ = 129.5, 115.1, 104.4, 61.5, 32.9, 31.9, 29.4, 27.4. Small peaks attributed to an inseparable impurity were found in the 1H NMR at δ = 5.85, and in the 13C NMR at δ = 128.2, 126.2, 114.8, 32.2, 29.8.

IR (neat): 3357, 2956, 2929, 2857, 1604, 1509, 1465, 1222, 1157, 1016, 894, 839 cm–1.

1H NMR (400 MHz, CDCl3): δ = 7.31–7.36 (m, 2 H), 6.98–7.05 (m, 2 H), 5.73 (d, J = 2.3 Hz, 1 H), 5.59 (d, J = 1.6 Hz, 1 H), 3.43 (dd, J = 11.6, 5.8 Hz, 1 H), 3.12 (dd, J = 11.6, 8.9 Hz, 1 H), 2.20–2.27 (m, 1 H), 1.90–1.95 (m, 1 H), 1.60 (s, 1 H), 1.20–1.44 (m, 9 H), 0.85 (t, J = 7.1 Hz, 3 H).

13C NMR (100 MHz, CDCl3): δ = 161.6 (C) [d, 1 J(CF) = 243 Hz], 140.7 (C), 134.9 (C) [d, 4 J(CF) = 3 Hz], 130.6 (CH) [d, 3 J(CF) = 8 Hz], 115.1 (CH) [d, 2 J(CF) = 21 Hz], 105.3 (CH2), 61.9 (CH2), 40.1 (CH2), 33.0 (C), 31.7 (CH2), 31.0 (CH), 29.1 (CH2), 27.1 (CH2), 22.6 (CH2), 14.1 (CH3).

HRMS-CI (NH3): m/z [M – OH] calcd for C17H22F: 245.1700; found: 245.1701.


#

3β-Hexyl-2α-(hydroxymethyl)-3α-(naphthalen-2-yl)-1-methyl­enecyclopropane (17)

General procedure B with 5 (43 mg, 0.22 mmol), CuBr·SMe2 (9 mg, 0.04 mmol), and 0.25 M 2-naphthylmagnesium bromide in THF (4.0 mL, 1.0 mmol) in hexanes (2 mL) gave 17 (33 mg, 0.11 mmol, 50%) as a pale yellow oil. A similar experiment with 5 (21 mg) gave 17 (14 mg, 45%).

Small peaks attributed to the diastereomer of 17 were found in the 1H NMR at δ = 5.76, 5.55, 3.96, 3.80, and in the 13C NMR at δ = 128.0, 127.6, 126.5, 126.2, 126.0, 125.5, 104.3, 61.5, 32.7, 32.4, 29.4, 27.5.

IR (neat): 3341, 3063, 2955, 2929, 2856, 1594, 1507, 1463, 1395, 1378, 1341, 1131, 1073, 1020, 891, 803, 780, 736 cm–1.

1H NMR (400 MHz, CDCl3): δ = 7.77–7.82 (m, 4 H), 7.42–7.55 (m, 3 H), 5.83 (d, J = 2.0 Hz, 1 H), 5.64 (d, J = 1.7 Hz, 1 H), 3.40 (dd, J = 11.6, 5.8 Hz, 1 H), 3.13 (dd, J = 11.6, 9.0 Hz, 1 H), 2.36–2.43 (m, 1 H), 1.95–2.01 (m, 1 H), 1.15–1.45 (m, 10 H), 0.81 (t, J = 6.8 Hz, 3 H).

13C NMR (100 MHz, CDCl3): δ = 140.9 (C), 136.7 (C), 133.3 (C), 132.3 (C), 128.1 (CH), 127.9 (CH), 127.7 (CH), 127.6 (CH), 126.8 (CH), 126.1 (CH), 125.7 (CH), 105.4 (CH2), 61.9 (CH2), 39.8 (CH2), 33.9 (C), 31.7 (CH2), 31.5 (CH), 29.1 (CH2), 27.3 (CH2), 22.6 (CH2), 14.0 (CH3).

HRMS-CI (NH3): m/z [M+] calcd for C21H26O: 294.1984; found: 294.1990.


#

3β-(4-Fluorophenyl)-3α-hexyl-2α-(hydroxymethyl)-1-methyl­enecyclopropane (18)

General procedure B with 5 (18 mg, 0.091 mmol), Fe(acac)3 (6 mg, 0.02 mmol), and 1.0 M 4-fluorophenylmagnesium bromide in THF (0.46 mL, 0.46 mmol) in hexanes (1 mL) and Et2O (0.5 mL) gave 18 (17 mg, 0.065 mmol, 71%) as a pale yellow oil. A similar experiment with 5 (31 mg) gave 18 (31 mg, 75%).

Small peaks attributed to the diastereomer of 18 were found in the 1H NMR at δ = 5.73, 5.59, 3.42, 3.14, 2.23, 1.91, and in the 13C NMR at δ = 140.8, 130.6, 115.0, 105.2, 61.8, 40.0, 31.0, 29.1, 27.1. Small peaks attributed to an inseparable impurity were found in the 1H NMR at δ = 6.26, 4.33.

IR (neat): 3331, 3068, 2929, 2857, 1604, 1509, 1466, 1378, 1296, 1222, 1157, 1138, 1113, 1093, 1015, 893, 838, 815, 727, 531 cm–1.

1H NMR (400 MHz, CDCl3): δ = 7.27–7.32 (m, 2 H), 6.95–7.07 (m, 2 H), 5.68 (d, J = 2.6 Hz, 1 H), 5.51 (d, J = 2.0 Hz, 1 H), 3.94 (dd, J = 11.6, 5.8 Hz, 1 H), 3.77 (dd, J = 11.6, 9.0 Hz, 1 H), 1.92–2.04 (m, 1 H), 1.83–1.88 (m, 1 H), 1.54–1.62 (m, 1 H), 1.44 (s, 1 H), 1.22–1.28 (m, 8 H), 0.86 (t, J = 6.7 Hz, 3 H).

13C NMR (100 MHz, CDCl3): 161.4 (C) [d, 1 J(CF) = 243 Hz], 141.5 (C), 139.6 (C) [ d, 4 J(CF) = 3 Hz], 129.5 (CH) [d, 3 J(CF) = 8 Hz], 115.0 (CH) [d, 2 J(CF) = 21 Hz], 104.2 (CH2), 61.4 (CH2), 32.9 (CH2), 32.6 (C), 32.2 (CH), 31.7 (CH2), 29.4 (CH2), 27.3 (CH2), 22.6 (CH2), 14.0 (CH3).

HRMS-CI (NH3): m/z [M – OH] calcd for C17H22F: 245.1706; found: 245.1694.


#

3α-Hexyl-2α-(hydroxymethyl)-1-methylene-3β-(o-tolyl)cyclopropane (19)

General procedure B with 5 (18 mg, 0.091 mmol), Fe(acac)3 (6 mg, 0.02 mmol), and 1.6 M o-tolylmagnesium bromide in THF (0.29 mL, 0.46 mmol) in hexanes (1 mL) and Et2O (0.5 mL) gave 19 (12 mg, 0.047 mmol, 51%) as a pale yellow oil. A similar experiment with 5 (15 mg) gave 19 (12 mg, 60%).

IR (neat): 3352, 3064, 3019, 2928, 2857, 1602, 1488, 1461, 1378, 1129, 1101, 1080, 1012, 890, 758, 731 cm–1.

1H NMR (400 MHz, CDCl3): δ = 7.23–7.25 (m, 1 H), 7.15–7.17 (m, 2 H), 7.09–7.12 (m, 1 H), 5.74 (d, J = 2.6 Hz, 1 H), 5.50 (d, J = 2.0 Hz, 1 H), 3.96 (dd, J = 11.6, 6.2 Hz, 1 H), 3.88 (dd, J = 11.6, 8.5 Hz, 1 H), 2.51 (s, 3 H), 1.82–1.95 (m, 2 H), 1.49–1.61 (m, 2 H), 1.21–1.31 (m, 8 H), 0.85 (t, J = 6.7 Hz, 3 H).

13C NMR (90 MHz, CDCl3): δ = 142.5 (C), 140.9 (C), 137.3 (C), 130.6 (CH), 130.1 (CH), 126.8 (CH), 125.2 (CH), 103.8 (CH2), 61.6 (CH2), 32.9 (C), 32.1 (C), 31.7 (CH2), 30.5 (CH), 29.5 (CH2), 27.5 (CH2), 22.6 (CH2), 19.4 (CH3), 14.0 (CH3).

HRMS-ESI: m/z [M + Na] calcd for C18H26ONa: 281.1881; found: 281.1877.


#

3β-Butyl-2α-(hydroxymethyl)-3α-methyl-2-methylene-2β-phenylcyclopropane (20)

General procedure B with 3 (19 mg, 0.077 mmol), Fe(acac)3 (5 mg, 0.02 mmol), and 3.0 M MeMgBr in Et2O (0.13 mL, 0.39 mmol ) in hexanes (1 mL) and Et2O (0.5 mL) gave 20 (9 mg, 0.039 mmol, 51%) as a pale yellow oil. A similar experiment with 3 (34 mg) gave 20 (17 mg, 53%).

IR (neat): 3405, 3059, 3024, 2956, 2929, 2871, 1601, 1493, 1446, 1378, 1149, 1012, 889, 701 cm–1.

1H NMR (400 MHz, CDCl3): δ = 7.24–7.38 (m, 5 H), 5.66 (s, 1 H), 5.52 (s, 1 H), 4.08 (d, J = 11.5 Hz, 1 H), 3.82 (d, J = 11.5 Hz, 1 H), 1.55 (s, 1 H), 1.39 (s, 3 H), 1.04–1.38 (m, 5 H), 0.77–0.85 (m, 1 H), 0.74 (t, J = 7.3 Hz, 3 H).

13C NMR (100 MHz, CDCl3): δ = 147.4 (C), 139.5 (C), 129.7 (CH), 128.2 (CH), 126.7 (CH), 102.8 (CH2), 67.1 (CH2), 38.3 (C), 35.6 (CH2), 29.5 (C), 28.8 (CH2), 22.7 (CH2), 17.5 (CH3), 14.0 (CH3).

HRMS-CI (NH3): m/z [M – OH] calcd for C16H21: 213.1643; found: 213.1641.


#
#

Acknowledgment

For financial support we thank NIGMS (NIH R01 GM068650). Data were obtained with instrumentation supported by NSF CRIF:MU, CHE 0840401 and CHE 1048367; NIH S10 RR026962; NIH COBRE 2P20RR017716.

Supporting Information

    for this article is available online at http://www.thieme-connect.com.accesdistant.sorbonne-universite.fr/ejournals/toc/synthesis.
  • Supporting Information


   Permissions and Reprints All articles of this category
Zoom Image
Scheme 1 Substitution reactions of cyclopropenes with allylic ether leaving groups
Zoom Image
Scheme 2 Dianion alkylation/substitution as a route to methylenecyclopropanes
Zoom Image
Scheme 3 Synthesis of 1,2-disubstituted 3-(hydroxymethyl)cyclopropenes with allylic leaving groups
Zoom Image
Scheme 4 Copper(I)-catalyzed addition of Grignard reagents to provide methylenecyclopropanes bearing quaternary centers
Zoom Image
Scheme 5 Reversal of facial selectivity in Fe(acac)3-catalyzed addition of aryl Grignard reagents to 3-(hydroxymethyl)cyclopropenes
Zoom Image
Scheme 6 Iron-catalyzed carbomagnesation takes place on the syn-face with cyclopropenes that are disubstituted at C3
Zoom Image
Figure 1 Shielding effects in methylenecyclopropanes with aromatic substituents support stereochemical assignments