Generation of metalloenamines by carbon-carbon bond formation: Ring opening reactions of 2-methyleneaziridines with organometallic reagents

Article abstract of DOI:10.1039/b005623p

Ring opening of 2-methyleneaziridines with Grignard reagents in the presence of CuI yields metalloenamines in a regiospecific fashion which can be further reacted with electrophiles to produce functionalised ketones via a one-pot process.

Full text of DOI:10.1039/b005623p

View Article Online / Journal Homepage / Table of Contents for this issue
Generation of metalloenamines by carbon–carbon bond formation: ring
opening reactions of 2-methyleneaziridines with organometallic reagents
Jerome F. Hayes,a Michael Shipman*b and Heather Twinb
a SmithKline Beecham Pharmaceuticals, Old Powder Mills, Tonbridge, Kent, UK TN11 9AN
b School of Chemistry, University of Exeter, Exeter, Devon, UK EX4 4QD. E-mail: M.Shipman@exeter.ac.uk
Received (in Liverpool, UK) 7th July 2000, Accepted 31st July 2000
First published as an Advance Article on the web 4th September 2000
Ring opening of 2-methyleneaziridines with Grignard rea-
gents in the presence of CuI yields metalloenamines in a
regiospecific fashion which can be further reacted with
electrophiles to produce functionalised ketones via a one-pot
process.
methyleneaziridine 1 (R = CHMePh) with Bu2CuLi (1.5
equiv.) and BF3·OEt2 (1.5 equiv.) in THF (278 ? 0 °C) and
subsequent hydrolysis with aqueous acid furnished heptan-
2-one in 40% isolated yield after distillation. Further studies
determined that this ring opening reaction can be accomplished
more conveniently using BuMgCl in the presence of BF3·OEt2
and a catalytic amount of CuI (Table 1, entry 1). The modest
isolated yield observed for the formation of heptan-2-one is due
to its volatility, as the chemical conversion is excellent as
judged by GC analysis using an internal standard. The copper
catalysed ring opening of methyleneaziridines with Grignard
reagents appears general and we have used it to make a variety
of methyl ketones 2 in moderate to good yields (Table 1).†
Importantly, in control experiments, we have determined that at
low temperatures, both CuI and the Lewis acid are essential for
rapid ring opening to occur.
Our initial attempts to alkylate the presumed metalloenamine
intermediate produced by ring opening of methyleneaziridine 1
(R = CHMePh) with BuMgCl, CuI and BF3·OEt2 were entirely
unsuccessful. Introduction of an electrophile such as BnCl into
the vessel prior to aqueous hydrolysis did not result in the
formation of any further alkylated products. Control experi-
ments using a metalloenamine made by the traditional method
of imine deprotonation using EtMgCl revealed that the presence
of BF3·OEt2 in the reaction mixture completely suppresses
alkylation reactions of the metalloenamine. To overcome this
problem, we sought conditions to effect the ring opening of
2-methyleneaziridines with organometallic reagents in the
absence of an added Lewis acid. After further experimentation,
we determined that the ring opening could be achieved using
Grignard reagents in the presence of CuI (20 mol%) by warming
the mixture to room temperature and stirring for an extended
period (24 h) prior to introduction of the electrophile. Using
these conditions, treatment of methyleneaziridine 1 (R =
CHMePh) with BuMgCl then BnCl yielded alkylated ketone 4
in 78% isolated yield after imine hydrolysis and silica gel
Following their introduction independently by Stork1 and by
Wittig,2 metalloenamines (metallated imines) have become
established as one of the best methods of forming carbon–
carbon bonds at the a-carbon atom of aldehydes and ketones.3
Metalloenamines are known to be extremely reactive and
undergo clean C-alkylation reactions with a wide range of
electrophiles including primary and secondary alkyl halides,
aldehydes, ketones, nitriles, epoxides and oxetanes. Conse-
quently, alkylation reactions of these carbanions are widely
used in organic synthesis. Metalloenamines are traditionally
made by deprotonation of the corresponding imine using a
strong base such as EtMgBr or LDA. One limitation with this
method stems from the fact that regiochemical control in the
enolisation process is not always possible when imines derived
from unsymmetrical ketones are used, resulting in mixtures of
alkylated products being formed.4 Few alternative methods for
the generation of metalloenamines have been developed, with
the only notable contributions being made by Wender.5 We
imagined that ring opening of 2-methyleneaziridines with
organometallic reagents might provide an alternative entry into
metalloenamines, as depicted in Scheme 1. We felt that this
conceptually new approach to metalloenamines might offer two
significant advantages over the traditional route to such species
by way of imine enolisation. Firstly, the overall transformation
would result in the formation of two new carbon–carbon bonds
and thus could be considered as an efficient and potentially
highly flexible three-component coupling reaction. Secondly,
the ring opening of the aziridine might be expected to provide
the metalloenamine as a single regioisomer, circumventing
problems associated with regioselectivity in the deprotonation
of unsymmetrical imines. Here, we describe our preliminary
results regarding the ring opening reactions of 2-methyleneazir-
idines with organometallic reagents and demonstrate their
potential as metalloenamine precursors.
Table 1
At the outset of this study, no ring opening reactions of
methyleneaziridines using carbon based nucleophiles had been
reported.6,7 Initially, we chose to investigate the ring opening
reactions of methyleneaziridine (±)-1 (R = CHMePh) as it is
very easy to prepare on a multigram scale.8 Since Ganem has
shown that simple N-alkylated aziridines can be ring opened
with Gilman-type cuprates in the presence of boron trifluoride
ethereate,9 we anticipated that these conditions might effect ring
opening of 2-methyleneaziridines. Gratifyingly, treatment of
Scheme 1
DOI: 10.1039/b005623p
Chem. Commun., 2000, 1791–1792
This journal is © The Royal Society of Chemistry 2000
1791

View Article Online
some of the mass spectral measurements and the EPSRC
Chemical Database Service at Daresbury.10
Notes and references
† All new compounds were fully characterised using standard spectroscopic
and analytical methods.
Scheme 2 Reagents and conditions: (i) BuMgCl (3.0 equiv.), CuI (20
mol%), THF, 230 °C ? rt, 24 h; (ii) BnCl (1.5 equiv.), 40 °C; (iii) 10%
HCl, 50 °C, 2 h.
‡ Typical procedure: CuI (48 mg, 0.252 mmol) in a round-bottomed flask
was heated under vacuum then purged with nitrogen (3 cycles performed).
Freshly distilled THF (4 ml) was added and the mixture cooled to 230 °C
whereupon BuMgCl (2.0 M in THF, 1.89 ml, 3.78 mmol) was added. After
stirring for 10 min, N-(1-phenylethyl)-2-methyleneaziridine (200 mg, 1.26
mmol) in THF (2 ml) was added. The reaction mixture was allowed to warm
up to room temperature and stirred for 24 h. The flask was then cooled to 0
°C and BnCl (218 ml, 1.89 mmol) added dropwise. A reflux condenser was
fitted and the reaction mixture heated at 40 °C for 2 h then allowed to cool
to room temperature and stirred overnight. Finally, 10% aq. HCl (3 ml) was
added, and the mixture heated at 50 °C for 2 h (for all other entries in Table
2, imine hydrolysis was effected using 1 M AcOH–hexane at rt). Upon
cooling, solid NaCl was added and the mixture extracted with Et2O. The
combined organic layers were washed with 0.5 M aq. HCl (2 3 15 ml),
saturated NH4Cl (2 3 20 ml), saturated NaHCO3 (2 3 20 ml) and brine (2
3 20 ml). The organic layer was dried (MgSO4) and the solvent removed in
vacuo. Column chromatography on silica gel (2.5% EtOAc in hexane) gave
1-phenyloctan-3-one (201 mg, 78%) as a yellow oil. nmax (thin film) 2947,
2931, 2856, 1713, 1450 cm21; dH (400 MHz, CDCl3) 7.30–7.17 (5H, m,
Ph), 2.89 (2H, m), 2.72 (2H, m), 2.37 (2H, t, J 7.6), 1.56 (2H, m), 1.30–1.21
(4H, m), 0.88 (3H, t, J 7.2); dC (100.9 MHz, CDCl3) 210.5 (s), 141.2 (s),
128.5 (d), 128.3 (d), 126.1 (d), 44.3 (t), 43.0 (t), 31.4 (t), 29.8 (t), 23.5 (t),
22.5 (t), 13.9 (q); Observed 204.1519; C14H20O requires 204.1514.
§ By comparison with an authentic sample of 3-benzylheptan-2-one using
gas chromatography and 1H NMR spectroscopy.
chromatography (Scheme 2).‡ Significantly, the formation of
metalloenamine 3 is highly regiospecific, and no trace of
3-benzylheptan-2-one could be detected in the crude reaction
mixture.§
We have extended the scope of this method of metal-
loenamine generation to other Grignard reagents and electro-
philes (Table 2).† Moderate to good yields of products are
obtained in all cases and the method provides a flexible and
efficient approach to functionalised ketones. The only notable
limitation with this chemistry is the requirement to use excess
Grignard reagent (3 equiv.) to drive the reaction to completion.
Currently, we are searching for more active copper catalysts in
an effort to address this problem. Future work will also be aimed
at extending the scope of this method and to applying it in
natural product synthesis.
Table 2
1 G. Stork and S. R. Dowd, J. Am. Chem. Soc., 1963, 85, 2178.
2 G. Wittig, H. D. Frommeld and P. Suchanek, Angew. Chem., Int. Ed.
Engl., 1963, 2, 683.
3 For reviews, see (a) P. W. Hickmott, Tetrahedron, 1982, 38, 3363; (b)
J. K. Whitesell and M. A. Whitesell, Synthesis, 1983, 517.
4 For a detailed discussion on the factors that effect regiochemical control
in the deprotonation of unsymmetrical imines, see A. Hosomi, Y. Araki
and H. Sakurai, J. Am. Chem. Soc., 1982, 104, 2081.
5 (a) P. A. Wender and J. M. Schaus, J. Org. Chem., 1978, 43, 782; (b)
P. A. Wender and M. A. Eissenstat, J. Am. Chem. Soc., 1978, 100,
292.
6 For a leading reference on ring opening reactions of methyleneazir-
idines, see D. S. Ennis, J. Ince, S. Rahman and M. Shipman, J. Chem.
Soc., Perkin Trans. 1, 2000, 2047.
7 It has been postulated that the decomposition of 3-lithio-1-tert-butyl-
2-methyleneaziridine involves ring opening of 1-tert-butyl-2-methyle-
neaziridine by this organolithium species, see H. Quast and C. A. Weise
Vélez, Angew. Chem., Int. Ed. Engl., 1974, 13, 342.
8 J. Ince, T. M. Ross, M. Shipman, A. M. Z. Slawin and D. S. Ennis,
Tetrahedron, 1996, 52, 7037.
9 M. J. Eis and B. Ganem, Tetrahedron Lett., 1985, 26, 1153.
10 D. A. Fletcher, R. F. McMeeking and D. Parkin, J. Chem. Inf. Comput.
Sci., 1996, 36, 746.
We are grateful to the EPSRC and SmithKline Beecham
Pharmaceuticals for their generous financial support of this
work. We are indebted to Julie Ince and David Ennis for their
assistance at the early stages of this project. We thank the
EPSRC National Mass Spectrometry Centre for performing
1792
Chem. Commun., 2000, 1791–1792