Regiodivergent 1,4 versus 1,6 asymmetric copper-catalyzed conjugate addition

Article abstract of DOI:10.1002/anie.200803735

(Chemical Equation Presented) Metal matters: A highly regiodivergent copper-catalyzed asymmetric conjugate addition to α,β and γ,δ Michael acceptors is described. Zinc and aluminum reagents afford the 1,6 adduct with good to moderate enantioselectivity in the presence of ligand 1 (see scheme). In contrast, Grignard reagents used with hydroxy imidazolium ligand 2 afforded the 1,4 adduct with excellent ee values.

Full text of DOI:10.1002/anie.200803735

Communications
DOI: 10.1002/anie.200803735
Asymmetric Catalysis
Regiodivergent 1,4 versus 1,6 Asymmetric Copper-Catalyzed
Conjugate Addition**
HØl›ne HØnon, Marc Mauduit, and Alexandre Alexakis*
The conjugate addition of carbon nucleophiles to
a,b-unsaturated carbonyl groups[1] is a powerful carbon-
carbon bond-forming process, and catalytic asymmetric
versions have become useful synthetic tools for the gener-
ation of tertiary[2] and quaternary carbon stereocenters.[3]
Despite a large body of literature on 1,4-conjugate additions,
analogous 1,6-addition methods are underdeveloped. In fact,
the presence of three electrophilic sites and the difficulties in
controlling the regioselectivity have limited the investigation
of this reaction. Most often, copper reagents exclusively
provided the 1,6-addition product,[4] and when we performed
a reaction with diethylzinc and L1 we observed only
1,6-addition compounds in 35% ee (Scheme 1).[5]
binaphthyl) catalysis, Hayashi et al. described the 1,6-con-
jugate addition of aryl zinc reagents to 3-alkenyl cyclohexen-
2-ones such as 1, with up to 98% ee.[8] This class of substrates
attracted our attention and we report herein our strange
results on the regiodivergent 1,4 or 1,6 copper-catalyzed,
asymmetric conjugate addition (ACA) of different alkyl
metal sources (RMgX, R2Zn, R3Al) to a,b and g,d Michael
acceptors.
We first examined the addition of diethylzinc to 3-(1-
propenyl) 2-cyclohexen-1-one (1) using different ligands
(Figure 1). Unsurprisingly, only the 1,6-addition compound
was obtained as deconjugated isomer 2a’ (Scheme 2). To
Scheme 1. 1,6 addition of diethyl zinc.
Recently Fillion et al. reported the asymmetric synthesis
of benzylic tertiary and quaternary stereogenic centers in
good yields and selectivities by using a 1,6-conjugate addition
of dialkylzinc reagents to Meldrumꢀs acid acceptors in the
presence of L1.[6] Also, Feringaꢀs group demonstrated that
ferrocene-based diphosphine ligands, such as (R)-1-[(S)-2-
diphenylphosphino)ferrocenyl]ethyldicyclohexylphosphine
((R,S)-josiphos), led to 1,6-conjugate addition to linear
dienoates using Grignard reagents.[7] They obtained high
enantioselectivities (up to 97% ee) and regioselectivities. By
using Rh/binap (binap = 2,2’-bis(diphenylphosphanyl)-1,1’-
Figure 1. Selected ligands used for this study.
prevent the formation of oxidative byproducts,[9] hydrochloric
acid that was degassed with argon was used for quenching the
reaction. The isomerization of 2a’ using 1 equivalent of
1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) under argon led to
the totally reconjugated adduct 2a. The best enantioselectiv-
ity (89% ee) was achieved with ligand L2. Triethylaluminum
also gave 1,6-adduct 2a (after reconjugation) but in a lower
yield. Also, the best ligand in this case, L1, gave a lower
enantioselectivity (68% ee).
To examine the scope of the 1,6 ACA, we next studied the
addition of Grignard reagents to 1. Using chiral ligands
L1–L3, as well as binap[10] or josiphos,[7] only led to the
1,6 adduct (after reconjugation) in low enantioselectivity.
However, when N-heterocyclic carbene (NHC) ligand L4[11]
was employed, the 1,4 adduct was obtained as the slightly
major regioisomer! Surprisingly, this result corresponds to a
conjugate addition at the most hindered position, thus
generating an all-carbon quaternary center with high enan-
tioselectivity (97% ee). After optimization of the reaction
[*] Dr. H. HØnon, Prof. Dr. A. Alexakis
Department of Organic Chemistry
University of Geneva
30 quai Ernest Ansermet, Geneva 4, Switzerland 1211
Fax: (+41)22-3793215
E-mail: alexandre.alexakis@chiorg.unige.ch
Homepage: http://www.unige.ch/sciences/chiorg/alexakis/
Dr. M. Mauduit
UMR CNRS 6226 “Sciences Chimiques de Rennes”
Ecole Nationale SupØrieure de Chimie de Rennes
Av. duGØnØral Leclerc, 35700 Rennes, France
[**] The authors thank the Swiss National Research Foundation (grant
No. 200020-113332) and COST action D40 (SER contract No.
C07.0097) for financial support.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200803735.
9122
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 9122 –9124

Angewandte
Chemie
DBU failed. In contrast, triethylaluminum was a
more reactive reagent and compound 11 was
obtained with moderate enantioselectivity
(Table 2, entry 2), with L2 being the best ligand.
We also examined the addition of trimethylalu-
minum to 10; the best copper salt was CuTC
(copper thiophene carboxylate) and the best
ligand was SimplePhos (L3; Table 2, entry 3),[12]
giving compound 11 with a moderate enantio-
Scheme 2. Enantioselective 1,6-conugate addition of diethylzinc and trialkylaluminum
reagents to compound 1.
conditions, particularly the use of dichloro-
methane as the solvent, the 1,4 adduct
became almost the exclusive product
(Table 1).
nAlkyl Grignard reagents provided the
Scheme 3. Enantioselective 1,4-conjugate addition of methyl Grignard.
1,4 adducts with greater than 95% selectiv-
ity and ee values as high as 99% (Table 1,
entries 1, 3, and 4). iso-Propyl and iso-butyl
Grignard reagents afforded a separable
mixture of both regioisomers, whereas
methyl Grignard gave only the 1,6 adduct.
It seems that the natural trend for 1,6 addi-
tion,[4] as well as the preference for the least
substituted position,[2b] are difficult to over- Scheme 4. Enantioselective 1,4-conjugate addition to other substrates. Cy=cyclohexyl.
come. A solution to this problem was to
have a substrate with equally substituted
positions, such as 4 (Scheme 3). Thus, methyl Grignard was
now able to deliver 1,4 adduct 5 with excellent regioselectivity
(100%) and enantioselectivity (92%). The hydrogenation of
5 provided the saturated analogue, whose absolute config-
uration is already known.[3c]
This regioselective 1,4 ACA was extended to other similar
substrates having different substitution patterns (Scheme 4).
Substrates 6 and 8 reacted in exactly the same way, affording
the 1,4 adducts in high regio- and enantioselectivities.
Another interesting dienic substrate was bicyclic com-
pound 10, to which we applied our best conditions for 1,6 or
1,4 ACA (Table 2). Diethylzinc was not very reactive, and
only the deconjugated 1,6 addition product 11 was observed
with poor enantioselectivity. Attempts to reconjugate 11 using
meric excess of 56%. Ethyl Grignard underwent a regiose-
lective 1,4 ACA with 10 in the presence of L4 to give 2a with
excellent enantioselectivity (Table 2, entry 4).
This regiodivergent ACA is quite intriguing. Experiments
with simpler NHCꢀs (Arduengoꢀs carbene[13]) and Grignard
reagents gave exclusively the 1,6 adduct. Only when an OH
group was present on the NCH was the 1,4 adduct present.
Other carbenes, similar to L4,[11] gave lower enantioselectiv-
ity, but good 1,4 selectivity. It is well known that the only
observable p complex on such polyethylenic ketones is the
one on the a,b position.[4] Although the w adduct is usually
obtained, it may be speculated that if the reductive elimi-
nation step is fast, the copper(III) intermediate may collapse
readily to afford the 1,4 product. This is, for example, the case
with Yamamotoꢀs reagents (RCu/BF3).[14]
Of synthetic interest is the olefinic appendage of the
1,4 adduct. For example, compound 3c was cyclized by ring
Table 1: Enantioselective 1,4-conjugate addition of several Grignard
reagents to compound 1.
Table 2: 1,6- or 1,4-conjugate addition to substrate 10.
Entry
R
Product 2/3[b]
Conv. [%][b] Yield [%][c] ee [%][d]
1
2
3
4
5
6
Et
Me
Bu
3a
2b
3b
<1:99 100
62
n.d.
67
65
25
39
97
–
97
>99
95
99
Entry RM
L* Solvent T [8C] 11/12[a] Conv. [%][a,b] ee [%][c]
100:0 100
4:96 100
5:95 100
35:65 100
44:56 100
1
2
3
5
Et2Zn
L2 Et2O
L2 Et2O
À10 100:0
À10 100:0
À10 100:0
11
100 (45)
100 (54)
11
69
56
96
But-3-enyl 3c
iPr
iBu
Et3Al
3d
3e
Me3Al[d] L3 Et2O
EtMgBr L4 CH2Cl2 À10
2:98 100 (73)
[a] All reactions were performed with 1 (0.5 mmol), RMgX (2 equiv),
Cu(OTf)2/L4 6/9 mol%, in CH2Cl2 at À108C. [b] Determined by GC-MS
methods. [c] Yield of isolated 1,4 adduct. [d] Determined by GC methods
using a chiral stationary phase. n.d.=not determined.
[a] The product ratio 11/12 and the conversion were determined by GC-
MS methods. [b] Yield of isolated products in parentheses. [c] Deter-
mined by GC analysis using a chiral phase. [d] Run with 2 equivalents of
RM and with CuTC. RM=alkylmetal reagent.
Angew. Chem. Int. Ed. 2008, 47, 9122 –9124
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9123

Communications
closing metathesis (Scheme 5)[15] to afford spiro compound
13. Alternatively, adduct 9 was oxidatively cleaved to afford
ketoester 14. Besides its synthetic versatility, this transforma-
tion allowed us to determine the ee value of adduct 9.
NH4Cl-solution and 5 mL of CH2Cl2 were added, after which the
layers were separated. After extraction with CH2Cl2 (2 ” 5 mL), the
combined organic extracts were dried and evaporated. Flash chro-
matography (pentane/diethyl ether 90:10) afforded desired com-
pound 3a.
1H NMR (100 MHz, CDCl3): d = 0.78 (t, J = 7.5 Hz, 3H), 1.37 (q,
J = 7.5, 2H), 1.61–1.69 (m, 6H), 1.76–1.84 (m, 1H), 2.12 (d, J =
14.0 Hz, 1H), 2.16–2.33 (m, 2H), 2.46 (d, J = 14.0 Hz, 1H), 5.15 (d,
J = 16 Hz, 1H), 5.34 ppm (dq, J1 = 6.0 Hz et J2 = 16.0 Hz, 1H).
13C NMR (100 MHz, CDCl3): d = 7.9 (CH3), 18.3(CH ), 21.8 (CH2),
3
34.2 (CH2), 35.2 (CH2), 41.2 (CH2), 44.2 (C), 49.8 (CH2), 125.3(CH),
136.6 (CH), 212.0 ppm (CO). HRMS (EI): [M]+ found 166.1360, calcd
for C11H18O
:
166. 1357. [a]2D0 = + 72.24 degcm3 gÀ1 dmÀ1 (c =
1.4 gcmÀ3, CHCl3), 97% ee. The enantiomeric excess was determined
on the hydrogenated compound by GC analysis employing LIP-
ODEX-E (75-40-1-100): Rt1 = 36.88 min (minor), Rt2 = 39.09
(major). The corresponding racemic saturated compound was
obtained by copper-catalyzed conjugate addition of nPrMgBr to
3-ethyl-2-cyclohexenone.
Scheme 5. Synthetic transformations of the 1,4 adducts.
In addition, the resulting enolate from the 1,4 ACA could
be trapped with Ac2O (Scheme 6).[16] Enol acetate 15 was
used to regenerate the lithium enolate,[17] which upon
allylation gave a 3:1 ratio of monoallylated adduct 16 (as a
cis/trans mixture) and bisallylated 17. Both 16 and 17 under-
went a facile ring closing metathesis to provide products 18
and 19, respectively. Although 16 was a mixture of isomers, a
single product,19, was obtained; presumably the one with the
cis ring junction.
In summary, we have disclosed an unusual regiodivergent
1,4- or 1,6-asymmetric conjugate addition. Although the
1,6 adducts had moderate to good enantioselectivity, the
1,4 adducts had excellent ee values for an all-carbon quater-
nary stereocenter. Additional work is in progress for a better
understanding of the mechanistic insights.
Received: July 30, 2008
Published online: October 16, 2008
Keywords: conjugate addition · copper · Grignard reagent ·
.
regioselectivity · synthetic methods
[1] P. Perlmutter, in Conjugate Addition Reactions in Organic
Synthesis, Pergamon Press, Oxford, 1992.
[2] Reviews on asymmetric conjugate additions, see: a) A. Alexakis,
C. Benhaim, Eur. J. Org. Chem. 2002, 3221 – 3226; b) N. Krause,
A. Hoffmann-Röder, Synthesis 2001, 171 – 196; c) T. Hayashi,
Acc. Chem. Res. 2000, 33, 354 – 362.
[3] a) B. M. Trost, C. Jiang, Synthesis 2006, 369 – 396; b) J. Chris-
toffers, A. Baro, Adv. Synth. Catal. 2005, 347, 1473– 1482; c) M.
Vuagnoux-dꢀAugustin, A. Alexakis, Chem. Eur. J. 2007, 13,
9647 – 9662; d) K.-S. Lee, M. K. Brown, A. W. Hird, A. H.
Hoveyda, J. Am. Chem. Soc. 2006, 128, 7182 – 7184.
[4] N. Krause, S. Thorand, Inorg. Chim. Acta 1999, 296, 1 – 11.
[5] K. Allarcon, A. Alexakis, Research Report: Enantioselective 1,6-
Conjugate Additions, Geneva, 2001.
[6] E. Fillion, A. Wilsily, E.-T. Liao, Tetrahedron: Asymmetry 2006,
17, 2957 – 2959.
[7] T. den Hartog, S. R. Harutyunyan, D. Font, A. J. Minnaard, B. L.
Feringa, Angew. Chem. 2008, 120, 404 – 407; Angew. Chem. Int.
Ed. 2008, 47, 398 – 401.
Experimental Section
Synthesis of 3a: Cu(OTf)2 (10.8 mg, 6 mol%) and L4 (14.6 mg,
9 mol%) were dissolved in dry CH2Cl2 (1.5 mL) in a dried Schlenk
tube equipped with septum and stirring bar under nitrogen. The
mixture was cooled to À108C and EtMgBr (2 equiv in Et2O) was
added. The reaction mixture was stirred for an additional 5 min and
then a solution of dienone 1 (0.5 mmol) in dry CH2Cl2 (5 mL) was
added by syringe pump over 15 min. The reaction mixture was stirred
for 1 h at À108C and then an NH4Cl solution (1m, 0.5 mL) was added.
The mixture was warmed to room temperature and then 5 mL of the
[8] T. Hayashi, S. Yamamoto, N. Tokunaga, Angew. Chem. 2005,
117, 4296 – 4299; Angew. Chem. Int. Ed. 2005, 44, 4224 – 4227.
[9] Z. Jing, P. L. Fuchs, J. Am. Chem. Soc. 1994, 116, 5995 – 5996.
[10] S.-Y. Wang, T.-K. Lum, S.-J. Ji, T. P. Loh, Adv. Synth. Catal. 2008,
350, 673– 677.
[11] D. Martin, S. Kehrli, M. dꢀAugustin, H. Clavier, M. Mauduit, A.
Alexakis, J. Am. Chem. Soc. 2006, 128, 8416 – 8417.
[12] L. Palais, I. Mikhel, C. Bournaud, L. Micouin, C. Falciola, M.
Vuagnoux-d’Augustin, S. Rosset, G. Bernardinelli, A. Alexakis,
Angew. Chem. 2007, 119, 7606 – 7609; Angew. Chem. Int. Ed.
2007, 46, 7462 – 7465.
[13] A. J. Arduengo III, M. K. Harlow, J. Am. Chem. Soc. 1991, 113,
361 – 363.
[14] Y. Yamamoto, Angew. Chem. 1986, 98, 945 – 957; Angew. Chem.
Int. Ed. Engl. 1986, 25, 947 – 959; for an interesting isolated
example of 1,4 selectivity, see: R. R. Cesati, J. de Armas, A. H.
Hoveyda J. Am. Chem. Soc. 2004, 126, 96–101.
[15] T. M. Trnka, R. H. Grubbs, Acc. Chem. Res. 2001, 34, 18 – 29.
[16] M. Vuagnoux-dꢀAugustin, A. Alexakis, Tetrahedron Lett. 2007,
48, 7408 – 7412.
[17] H. O. House, V. J. Kramar, J. Org. Chem. 1963, 28, 3362 – 3379.
Scheme 6. Synthetic transformations of enol acetate 15.
9124
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 9122 –9124