Copper-catalysed asymmetric 1,4-addition of organozinc compounds to linear aliphatic enones using 2,2′-dihydroxy 3,3′-dithioether derivatives of 1,1′-binaphthalene

Article abstract of DOI:10.1002/1099-0690(200107)2001:13<2435::AID-EJOC2435>3.0.CO;2-0

Directed ortho dilithiation of bis(diethylcarbamate) or bis(MOM)-protected (S_a)-1,1,′-bi(2-naphthol) followed by treatment with R₂S₂ [R = Me, Ph (X-ray structure)] or Me₂Se₂ cleanly affords the 3,3′ derivatives; the free naphthols are produced on deprotection. In the case of the bis(MOM) series, but not that of the bis(carbamates), some racemisation occurs. The ligand 2,2′-dihydroxy-3,3′-dimethylthio-1,1′-binaphthalene shows optimal performance in the addition of ZnEt₂ to linear aliphatic enones (E)-R¹C(O)CH=CHR². Variation of the steric demands of R¹ and R² generates catalytic results consistent with binding of a zinc-based Lewis acid anti to the ene function and with the reactive conformation being s-cis. With enones containing the functions R² = (CH²)nCH(OAlkyl)₂ (n = 0-2), the ZnEt₂ addition products undergo base-promoted cyclisation.

Full text of DOI:10.1002/1099-0690(200107)2001:13<2435::AID-EJOC2435>3.0.CO;2-0

FULL PAPER
Copper-Catalysed Asymmetric 1,4-Addition of Organozinc Compounds to
Linear Aliphatic Enones Using 2,2Ј-Dihydroxy 3,3Ј-Dithioether Derivatives of
1,1Ј-Binaphthalene
Christoph Börner,[a] Michael R. Dennis,[b] Ekkehard Sinn,[c] and Simon Woodward*[a]
Keywords: Asymmetric catalysis / Copper / Nucleophilic additions / S ligands / Selenium / Zinc
Directed ortho dilithiation of bis(diethylcarbamate) or bis-
(MOM)-protected (Sa)-1,1Ј-bi(2-naphthol) followed by treat-
ment with R2S2 [R = Me, Ph (X-ray structure)] or Me2Se2
ZnEt2 to linear aliphatic enones (E)-R1C(O)CH=CHR2. Vari-
ation of the steric demands of R1 and R2 generates catalytic
results consistent with binding of a zinc-based Lewis acid
cleanly affords the 3,3Ј derivatives; the free naphthols are anti to the ene function and with the reactive conformation
produced on deprotection. In the case of the bis(MOM)
series, but not that of the bis(carbamates), some racemisation
occurs. The ligand 2,2Ј-dihydroxy-3,3Ј-dimethylthio-1,1Ј-bi-
naphthalene shows optimal performance in the addition of
being s-cis. With enones containing the functions R2
=
(CH2)nCH(OAlkyl)2 (n = 0−2), the ZnEt2 addition products
undergo base-promoted cyclisation.
Introduction
featuring binding of both the enone and a Gilman-type
cuprate [Cu(R1)2]Ϫ [derived from the terminal organozinc
species Zn(R1)2] in an ordered transition state. Coordina-
tion of strong σ-donors, such as thioethers, is predicted to
result in a dramatic rate increase in conjugate addition, due
to stabilisation of the putative CuIII transition state through
Despite the enormous progress that has been made re-
cently in asymmetric 1,4-additions of organozinc reagents
to enones in the presence of phosphorus-ligated copper
complexes, significant problems remain to be solved in this
area.[1] For example, many catalysts are evaluated by their
performance in the model reaction between ZnEt2 and 2-
cycloalkenones.[2] The primary literature indicates that out-
side this system, significant drops in catalyst performance
can be experienced. Even the remarkable enantioselectivi-
ties achieved by the phosphoramidite ligands, for example,
can suffer with variation of the organozinc reagent and Ϫ
especially Ϫ the enone structure.[3] In particular, linear en-
ones bearing only aliphatic substituents are challenging sub-
strates for phosphorus-based ligands, although these frag-
ments frequently appear in synthetic targets of biological
interest.
charge
transfer,
affording
[(ArSδϩR2)Cu(R1)2-
(enolateδϪ)]Ϫ.[9] In this paper, full details are given of our
investigations of 2,2Ј-dihydroxy-3,3Ј-diorganothio-1,1Ј-bi-
naphthalene ligands (3, Scheme 1), currently the most ef-
fective additives we have found for this chemistry.
Recently, we[4Ϫ7] and others[8] have become interested in
the use of chiral organosulfur ligands in copper-catalysed
additions to linear aliphatic enones in attempts to find
highly selective reactions. In particular, we have designed
species which contain both a thioether (capable of coordin-
ating organocuprates) and aromatic alkoxide donors (al-
lowing strong interaction with the terminal organozinc spe-
cies). We believe that such ligands should favour organised
transition states, shown diagrammatically in structure A,
Results and Discussion
Ligand Structure Studies
Preliminary investigations had indicated that the di-
methyl species 3a was able to deliver significant enantiose-
School of Chemistry, The University of Nottingham, lectivity in CuI-catalysed 1,4-additions of ZnEt2 to (E)-al-
[a]
University Park,
kyl-3-en-2-ones (up to 73% ee) and 2-cyclohexenone (up to
Nottingham NG7 2RD, United Kingdom
77% ee).[6] One obvious strategy to improve this perform-
Fax: (internat.) ϩ 44-115/951-3564
E-mail: simon.woodward@nottingham.ac.uk
Department of Chemistry, The University of Hull,
Cottingham Road, Hull HU7 6RX, United Kingdom
Fax: (internat.) ϩ 44-1482/466410
ance is systematic variation of the nature of the YR sub-
stituent in structure 3. We have developed directed ortho
metallations of carbamate-protected, enantiopure 1,1Ј-bi-
naphthols[10] using Snieckus-type chemistry[11] and this is
shown in Scheme 1 for the (Sa) series. Treatment of the
bis(carbamate) 1a with sBuLi/TMEDA at Ϫ78 °C cleanly
[b]
[c]
Department of Chemistry, University of Missouri Ϫ Rolla,
Rolla, Missouri MO 65409-0010, USA
Fax: (internat.) ϩ 1-573/341-6033
E-mail: esinn@umr.edu
Eur. J. Org. Chem. 2001, 2435Ϫ2446
WILEY-VCH Verlag GmbH, 69451 Weinheim, 2001
1434Ϫ193X/01/0713Ϫ2435 $ 17.50ϩ.50/0
2435

C. Börner, M. R. Dennis, E. Sinn, S. Woodward
FULL PAPER
affords the dilithio species, which when intercepted with the antipodes of 3a؊d (or the precursors 2a؊b and 4a؊d).
R2S2 (R ϭ Me, Ph) yields the 3,3Ј-disubstituted compounds In the absence of a direct assay, racemisation of species
2. However, while the methylthio compound 2a is easily de- leading to 3a had to be investigated indirectly. Firstly, to be
protected with either LiAlH4 or MeLi, producing 3a, the absolutely sure that racemisation was not a feature of the
equivalent phenyl compound 2b reacts sluggishly and not carbamate route (through 2, Scheme 1), the optical rota-
very cleanly with these reagents. Steric protection of the car- tions of repeatedly fractionally crystallised samples of 2a؊b
bamate may be the origin of these problems. In the search were measured. These optical rotations remained constant,
25
for a general solution to the synthesis of a range of 3 deriv- although that of 2b was very low ([α]D ϭ ϩ3 at c ϭ 5.0).
atives, dilithiation of the bis(MOM) derivative 1b under To check whether this class of molecules has any tendency
standard conditions[11] seemed attractive. Subsequent treat- to crystallise as eutectic mixtures, a crystallographic study
ment of these dilithio species with appropriate disulfides af- of 2b was undertaken. The results found that no (Ra)/(Sa)
fords 4 (or dimethyl diselenide for 4d), which can readily co-crystallisation had taken place and that the atom con-
be directly deprotected with HCl to afford the ligands 3 in nectivity expected from the double lithiation had been ob-
good yield.
tained (Figure 1). A similar study of the optical rotation of
3a upon repeated recrystallisation gave no evidence that the
material obtained from this route is less than enantiopure.
Assuming no nonlinear effect (NLE[13]) is associated with
CuI/3a catalysis, the ee value achieved for the formation of
6a suggests that the ee of 3a derived from the MOM-pro-
tected compound (i.e. 4a) is 88% (0.63/0.72 ϫ 100%). As all
of the derivatives 4 were deprotected under identical condi-
tions (acid concentration and hydrolysis time), an identical
deprotection was carried out on the starting material 1b.
The ee of the 1,1Ј-bi(2-naphthol) produced this way was
93%, as determined by HPLC, confirming induction of ra-
cemisation by HCl during MOM deprotection. While we
cannot rule out dramatically different degrees of racemis-
ation across the series 4a؊d, we believe that this is unlikely
and that the enantiopurity of all the ligands 3 produced by
this route is likely to be Ն 88% ee. Given the large differ-
ences in the ligand-derived enantioselectivities (Table 1), mi-
nor differences in the ligand enantiopurity would not affect
the conclusion that ligand structure 3a is already optimal.
Recently, the MOM-protected starting material 1b was sug-
gested as the most effective method for the synthesis of en-
antiopure 3,3Ј-bis(SiMe3)-substituted binol ligands.[14] No
comment was made regarding racemisation during the acid
deprotection of the intermediates in this synthesis.
The poor performance of seleno ligand (Sa)-3d was sur-
prising, as the ‘‘soft’’ selenium donor should ligate and sta-
bilise organocopper(III) intermediates more effectively than
the analogous thioether. Some attempts were made to vary
the reaction conditions for this ligand, with the aim of im-
proving its enantioselectivity. In dichloromethane, the addi-
tion of ZnEt2 to 5a produces 6a in 53% ee, while in Et2O
Scheme 1
The effect of changing the soft donor site could be as- the other enantiomer of the product is produced in 20% ee.
sayed by carrying out copper-catalysed reactions between Use of AlMe3 in THF produces a 41% ee for the equivalent
ZnEt2 and (E)-non-3-en-2-one 5a in the presence of 3/[Cu(- methyl addition product. Additionally, because the loadings
[12]
MeCN)4]BF4
(Table 1). These studies immediately re- of the thioethers 3 in Table 1 are relatively high (20 mol-
vealed two things: firstly, that the methylthioether 3a seems %) some catalytic reactions were studied at reduced ligand
to give the optimal performance with respect to ee value, loadings of 3a and different Cu/3a ratios (Table 2). Initial
and secondly, that the ee value obtained in the catalytic re- investigations revealed that reduction of the copper and li-
action using 3a is dependent on the route by which the li- gand loadings to 2.5 mol-% [Cu(MeCN)4]BF4 and 3a (5
gand is prepared.
mol-%) resulted in incomplete conversion, but significant
The ee values obtained for the product 6a suggest that amounts of 6a were still formed (71% yield) with slightly
the enantiopurity of 3a is compromised when it is prepared reduced induction (up to 68% ee).[6] At these lower ligand
by the MOM route (through 4a). Thus far, we have not loadings, the purity of the reagents and ligands becomes the
been able to find a chiral HPLC column that will separate critical factor in determining the chemical yield and enanti-
2436
Eur. J. Org. Chem. 2001, 2435Ϫ2446

Copper-Catalysed Asymmetric 1,4-Addition of Organozinc Compounds to Linear Aliphatic Enones
Table 1. Treatment of (E)-non-3-en-2-one 5 with ZnEt2 in the presence of [Cu(MeCN)4]BF4 (10 mol-%) and ligands 3aϪd (20 mol-%)
FULL PAPER
Ligand
Chemical yield [%][a]
(c.y.)
Product ee [%][b]
(from MOM series ligands)
Product ee [%][b]
(from carbamate series ligands)
3a
3b
3c
3d
85
58
75
68
63 (ϩ)-(R)[c]
25 (Ϫ)-(S)[c]
12 (ϩ)-(R)[c]
23 (ϩ)-(R)[c]
73 (ϩ)-(R)[c]
Ϫ
Ϫ
Ϫ
[a]
[b]
Carried out as THF solutions at Ϫ20 °C with [CuI]initial ϭ 23 m; [3]initial ϭ 46 m. Ϫ
Determined by G.C. on an oktakis(6-O-
[c]
methyl-2,3-di-O-pentyl)-γ-cyclodextrin column. Ϫ
Stereochemistry assigned by comparison with analogous (ϩ)-(R)-methyl addition
compound (see later, also ref.[6]).
MϪOϭC ഠ 120°[15,16]). The relative populations of the four
species 5c,a, 5c,s, 5t,a, and 5t,s in the ‘‘loaded’’ state of the
catalyst therefore directly affect the Re/Si ratio and hence
the ee value provided by the catalyst. Considerable informa-
tion is available both on carbonyl group Lewis acid bind-
ing[15] and on the mechanism of organocuprate additions[9]
to enones. To the best of our knowledge, however, few at-
tempts had been made to apply this information to asym-
metric copper-catalysed additions of organozinc com-
pounds.
To obtain information on the nature of the asymmetric
transition state attained with ligand 3a, a series of enones
(5aϪl) were prepared by either crossed-aldol or Wittig
methods. The structures of these compounds were expected
to favour one of the species 5c,a, 5c,s, 5t,a, and 5t,s signific-
Figure 1. Molecular structure of 2b shown with 30% probability
antly. Although many of these species are known, great care
was taken to obtain pure (E) samples rather than the (E)/
(Z) mixtures occasionally produced by the literature routes.
These species were treated with ZnEt2 in the presence of
[Cu(MeCN)4]BF4 and stereochemically pure (Sa)-3a under
identical conditions (Scheme 3, Table 3). To ensure com-
plete reproducibility in these mechanistic studies, the load-
ing of 3a was kept high. Additionally, because of the ra-
cemisation associated with the MOM route, all 3a used in
subsequent studies was prepared from the bis(carbamate)
1a. The preference for zinc-derived Lewis acids to bind syn
(5c,s or 5t,s) or anti (5c,a or 5t,a) to the ene function was
investigated first. We reasoned that as the size of R1 is in-
creased any tendency for a ZnX2 fragment to bind anti
should be suppressed.
ellipsoids; hydrogen atoms have been omitted for clarity; selected
˚
bond lengths [A] and angles [°]: SϪC(3) 1.774(6), SϪC(16)
1.767(6), O(1)ϪC(2), O(1)ϪC(11), N(1)ϪC(11), 1.337(8),
N(1)ϪC(12) 1.463(8), N(1)ϪC(14) 1.474(8); C(3)ϪSϪC(16)
106.8(3), C(2)ϪO(1)ϪC(11) 116.2(4), O(2)ϪC(11)ϪN(1) 127.2(6)
oselectivity. Typically, a range of chemical yield (c.y.) and
ee values are obtained, and these are shown in Table 2 to-
gether with the effects of temperature and solvent change
on the reaction.
Enone Structure Studies
For linear enones, free rotation produces s-cis (5c) and s-
trans (5t) (Scheme 2). Exchange between these two con-
formers is of key importance in asymmetric catalysis, as it
exchanges the Re/Si faces of the enone (the stereochemical
The behaviour of enones 5a؊c is consistent with binding
descriptors of the front-side faces at the β carbon atom are of the zinc Lewis acid anti to the ene functions, as in struc-
shown in Scheme 2). The presence of zinc-derived Lewis ac- tures 5c,a and 5t,a (Scheme 2). Reduction in the enone react-
ids ZnX2 (X ϭ Et, alkoxide) in the asymmetric conjugate ivity and selectivity through an increase in the steric de-
reaction complicates the situation, as this may bind the car- mand of R1 would not be expected if syn binding is a major
bonyl lone pair either syn or anti to the ene function (while contribution. Unsaturated esters normally present ground
ionic Lewis acids, such as Liϩ, bind carbonyl groups with states possessing a (Z) configuration about the ϭCϪOR
little spatial preference, ‘‘softer’’ Lewis acids bind the car- ester bond and are therefore expected to be poor substrates
bonyl group while preserving its sp2 hybridisation; that is, if anti carbonyl group binding is a requirement. This proved
Eur. J. Org. Chem. 2001, 2435Ϫ2446
2437

C. Börner, M. R. Dennis, E. Sinn, S. Woodward
Table 2. Treatment of (E)-non-3-en-2-one 5 with ZnR2 in the presence of [Cu(MeCN)4]BF4 and ligand 3a
FULL PAPER
Ligand loading
[mol-%] (conc.[a]
CuI loading
[mol-%] (conc.[a]
Conditions
Product c.y. (ee)
[from MOM series ligands] [%][b]
Product c.y. (ee)
[from carbamate series ligands] [%][b]
)
5 (46 m)
8 (36 m)
8 (36 m)
8 (19 m)
8 (65 m)
20 (46 m)
20 (46 m)
20 (46 m)
20 (46 m)
20 (46 m)
20 (46 m)
2.5 (23 m)
4 (18 m)
8 (36 m)
4 (9 m)
4 (33 m)
10 (23 m)
10 (23 m)
10 (23 m)
10 (23 m)
10 (23 m)
10 (23 m)
R ϭ Et, THF, Ϫ20 °C
R ϭ Et, THF, Ϫ20 °C
R ϭ Et, THF, Ϫ20 °C
R ϭ Et, THF, Ϫ20 °C
R ϭ Et, THF, Ϫ20 °C
R ϭ Et, THF, 0 °C
Ϫ
Ϫ
71 (68)
54 (65)
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
58 (62)
44 (61)
52 (61)
71 (64)
32 (43)
20 (12[c])
32 (43)
Ͻ 16 (58)
n.r.[d]
R ϭ Et, THF, Ϫ50 °C
R ϭ Et, toluene, Ϫ20 °C
R ϭ Et, CH2Cl2, Ϫ20 °C
R ϭ Me, THF, Ϫ20 °C
R ϭ CH2TMS, THF, Ϫ20 °C
Ϫ
Ϫ
[a]
[b]
nitial concentrations. Ϫ Determined by GC on an oktakis(6-O-methyl-2,3-di-O-pentyl)-γ-cyclodextrin column. Unless stated other-
wise, the (ϩ)-(R) isomer was predominant by comparison with analogous (ϩ)-(R)-methyl addition compound (see later, also ref.[6]). Ϫ
(Ϫ)-(S) isomer. Ϫ No reaction.
[c]
[d]
to be the case: 5d is completely unreactive, although this
might simply be due to the known low reactivity of unsatur-
ated esters in copper-catalysed conjugate addition. Com-
pound 5e is expected to show complete anti binding, due to
the presence of the chelate. The result is somewhat ambigu-
ous. While the presence of a clean catalytic reaction rein-
forces the idea of anti coordination in this case, as the (Ϫ)-
6e antipode is isolated, in all other cases (ϩ)-6 is formed
when (Sa)-3a is used. We have assigned the (ϩ)-6 stereoiso-
mers the (R) configuration on the basis of the facts that the
(ϩ) enantiomer also corresponds to (R) in the analogous
methyl addition compound and that the (ϩ/Ϫ) elution be-
haviour for the methyl and ethyl compounds is the same
under a range of different chiral GC conditions. Thus, for
enones other than 5e, the ligand (Sa)-3a affords the (R)-6
conjugate addition product. For 5e, either the presence of
the heteroatom alters the sign of the optical rotation, or
more probably the presence of a chelating substrate changes
the asymmetric transition state, reversing the selectivity.
The propensity for the linear enones to react in an s-cis
vs. s-trans conformation could be examined using enones
5f؊g. We supposed that, because of our design strategy (cf.
Scheme 2
structure A), that transition states featuring s-cis conforma-
tions would be much more susceptible to steric clashes
caused by increasing the size of substituent R2 than those
proceeding from the alternative s-trans conformer (clearly
this is a simplification, as conformers 5c,s and 5t,s will also
be affected to some degree, but in view of the lack of literat-
ure data this simple model was pursued to see if the results
were consistent with structure A). Additionally, because of
the proposed close proximity of the zinc and copper centre
substrates, processing of ether-like oxygen atoms may be
able to adopt binding motifs not available to simple enones.
Comparing runs using 5a, 5f, and 5g, it can clearly be seen
that, as the steric imposition on the transition state in-
creases, the chemical yield falls (nC5H11, 85%; iPr, 59%;
CH2iPr, 43%) but the ee value rises (nC5H11, 73%; iPr, 77%;
CH2iPr, 79%), indicating increased congestion in the trans-
ition state. Enone 5h fails to react, indicating that the chiral
Scheme 3
2438
cleft in the catalyst structure cannot accommodate R2 ϭ
Eur. J. Org. Chem. 2001, 2435Ϫ2446

Copper-Catalysed Asymmetric 1,4-Addition of Organozinc Compounds to Linear Aliphatic Enones
FULL PAPER
Table 3. Asymmetric conjugate addition of ZnEt2 to enones 5a؊l and 8؊9 catalysed by [Cu(MeCN)4]BF4 (10 mol-%) and (Sa)-2b (20
mol-%) in THF at Ϫ20 °C
Enone
R1[a]
R2
c.y. 6 [%][b]
ee 6 [%][b][c]
5a
5b
5c
5d
5e
5f
5g
5h
5i
Me
iPr
tBu
nC5H11
nC5H11
nC5H11
nC5H11
nC5H11
iPr
CH2-iPr
tBu
85
61
0
72 (ϩ)-(R)
39 (ϩ)-(R)
Ϫ
OMe
CH2OMe
Me
0
Ϫ
65
59
43
0
52
58
59
24 (Ϫ)[d]
77 (ϩ)-(R)
79 (ϩ)-(R)
Ϫ
Me
Me
Me
Me
CH(OMe)2
CH2CH(OEt)2
18 (ϩ)-(R)[e]
70 (ϩ)-(R)[e]
52 (ϩ)-(R)[e]
5j
5k
Me
5l
7
Me
Ϫ
Ϫ
Ϫ
Ϫ
CH2CH2CH(OEt)2
65
78
0
54
55
69 (ϩ)-(R)[e]
77 (Ϫ)-(S)[f]
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
8
9a
9b
Ϫ[g]
72Ϫ86[h]
[a]
[b]
Carried out as THF solutions at Ϫ20 °C with [CuI]initial ϭ 23 m; [3]initial ϭ 46 m. Ϫ
Determined by GC on an oktakis(6-O-
[c]
methyl-2,3-di-O-pentyl)-γ-cyclodextrin column. Ϫ
Stereochemistry assigned by comparison with analogous (ϩ)-(R)-methyl addition
Absolute stereochemistry not clear. Ϫ Determined on derived aldehyde. Ϫ Stereochemistry
compound (see later, also ref.[6]). Ϫ
[d]
[e]
[f]
[g]
[h]
based on known optical rotation. Ϫ 1:1 mixture of diastereomers at C(2); no stereocentre generated by conjugate addition. Ϫ
1:1
mixture of diastereomers at C(2); ee determined by 13C NMR of the alcohol 11 by CBS reduction.
tBu. These results suggest that an s-cis conformation is fa- four separate stereoisomers. The conjugate addition prod-
voured, as originally suggested by Feringa,[1] and further uct of 10b also proved to be rather resistant to ee deter-
experiments support this idea. The addition of ZnEt2 to mination by chemical derivatisation. Attempts to prepare
cyclohexenone (7) catalysed by CuI/(Sa)-3a affords the (S) acetals or imines from 10b using enantiopure diols or
addition product (77% ee).[17] As 7 can only react from s- amines also proved ineffective. In a final attempt to deter-
trans transition states, this reversal of enantiofacial selectiv- mine the enantioselectivity for the addition of ZnEt2 to 9b
ity strongly points to an s-cis transition state for the linear (Table 3), 10b was reduced with BH3·THF in the presence
enones 5. The s-trans and s-cis conformers of enone 8 are of (R)-CBS catalyst. Recent evidence indicates that sub-
expected to be almost isoenergetic. It fails to react, indicat- strates closely related to 10b are reduced with very high ee
ing that certain s-trans conformers cannot be accessed in values and that these give resolved NMR spectra.[19] In the
the mechanism of action of this catalysts. Finally, enones case of 10b, CBS reduction was expected to generate the
9a؊b could be used as enforced s-cis enones. As the literat- stereoisomers 11. Indeed, the 13C NMR spectrum showed
ure preparations of 9 predate NMR instrumentation, it is a clear splitting of peaks belonging to the major and minor
vital to ensure that the proposed (E) configuration is correct isomer of the asymmetric conjugate addition product. To
if it is to be used for mechanistic studies. Fortunately, the corroborate this result, the same asymmetric addition was
presence in compound 9b of a set of small (1.95Ϫ2.72%) performed using (Ra)-3a instead of (Sa)-3a and the addition
but conclusive nuclear Overhauser effects between the ring product reduced under identical conditions as 10b. Accord-
3-CH2 group and the allylic and homoallylic methylene ing to the 13C NMR spectra, ethyl addition gave an ee in
groups of the nC5H11 substituent confirmed the (E) assign- the range of 72Ϫ86%, depending on which pair of peaks
ment presumed in the original literature.[18] Compounds 9 was measured, assuming that the CBS reduction occurs
react smoothly with ZnEt2 to produce 10, reinforcing the with almost complete enantioselectivity.
suggestion that catalysts derived from 3a act on the acyclic
substrates 5 through an s-cis configuration.
The possibility of tridentate (CuϪalkene, ZnϪcarbonyl,
and ZnϪadditional donor) binding of enones could be ex-
After quenching with HCl(aq), the product enolates de- amined using substrates 5i؊l (Table 3). As the structure of
rived from 9 afford ketones with a stereogenic centre at C(2) the active catalyst in this system is not known, there may
(Scheme 4). As this centre arises through unselective pro- be more than one zinc Lewis acid site present if the catalyst
tonation, it is attained in a stereorandom manner. Com- is aggregated or if both naphtholate units in ligand 3 are
pound 9a thus affords a 1:1 mixture of enantiomers 10a, capped by a ZnEt unit. Additional binding of the enone
even though no chiral centre is generated by the conjugate through an acetal oxygen atom is expected to change the
addition; they can be separated by chiral GC. While we nature of the transition state. The shortest tethered acetal
could also separate the equivalent diastereomers of 10b on 5i clearly experiences such a fate, but the outcome is highly
a variety of chiral GC columns, we could not resolve the detrimental to the enantioselectivity and the addition pro-
Eur. J. Org. Chem. 2001, 2435Ϫ2446
2439

C. Börner, M. R. Dennis, E. Sinn, S. Woodward
FULL PAPER
dynamic control. For example, the cyclisation of 13c results
only in the unusual unsaturated aldehyde (S)-15: No 7-
membered ring product is isolated. (The change in the ste-
reochemical descriptor is a consequence of the CIP priority
rules, and not of the chemical transformation.)
Conclusion
Ligand optimisation studies have indicated that the 3,3Ј-
bis(SMe) derivative 3a derived from the carbamate 1a is the
most effective with respect to enantioselectivity. Variation
of the enone structure has revealed that the catalyst derived
from (Sa)-3a causes linear enones to adopt an s-cis con-
formation with a zinc-derived Lewis acid bound to the car-
bonyl lone pair anti to the ene function. While precise fea-
tures of the asymmetric transition state are not yet com-
pletely clear, a working mnemonic is given by structure B.
The nature of the steric block that is the apparent cause of
the enantioselectivity is not yet clear; a naphthyl ring is one
likely candidate.
Scheme 4
ceeds with very low induction (18% ee). The higher homo-
logues do not apparently bind the zinc Lewis acid site(s)
through their acetal functions and the chemical yields and
enantioselectivities achieved (Table 3) are comparable to
those of their hydrocarbon analogues. In all cases 5i؊l, the
initial product acetals undergo facile hydrolysis to the alde-
hydes 12.
We reasoned that aldehydes 12 might have some potential
as intermediates in prostenoid and homoprostenoid syn-
thesis. Treatment of ethereal solutions of 12 with 6  NaOH
resulted in ring closure but dehydration was facile. For 12a,
none of the 3-ethyl-4-hydroxycyclopentanone aldol product
was isolated, only the enone 13a being obtained. Few direct
routes to such enones are available,[20] but the route is not
practical in this case, due to the low ee value of the starting
material 6i and potential stereochemical lability of the
chiral centre. However, GC analysis of 13a indicated that
no further racemisation had occurred during the base-pro-
moted cyclisation. Carrying out the ring closure of 12b at 0
°C, it was possible to isolate a small amount of 14 (5%),
along with the dehydrated 13b as the major product (87%).
Isolated (3R,5S)-14 proved resistant to dehydration on
standing in CDCl3 for at least 3 months. This behaviour,
together with its 1H NMR spectrum, is consistent with
isolation of the all-equatorial isomer. The ratio of 13b/14
from this reaction may reflect the diastereoselectivity in the
cyclisation (Scheme 5); the derived axial alcohol spontan-
eously eliminates, producing 13b. The products isolated in
this base-promoted chemistry are formed under thermo-
Experimental Section
General: Infrared spectra were recorded using a Nicolet Avatar 360
FT-IR infrared spectrophotometer. Ϫ 1H and 13C NMR spectra
were recorded either with Jeol (GX 270) or with Bruker (AM 400,
AV400, DRX500) spectrometers at ambient temperature, using
tetramethylsilane as standard; J values are given in Hz. Ϫ Mass
spectra were obtained with AIMS902 (electron impact, EI, or
chemical ionisation CI), VG-ZAB (EPSRC service, Swansea), or
70E VG (fast atom bombardment, FAB) machines. Ϫ Elemental
analyses were performed using a CE-440 elemental analyser. Ϫ Op-
tical rotations were measured with Jasco, DIP370 Digital or AA-
10 Polarimeter instruments in units of 10Ϫ1 °·cm2·gϪ1 (c in g/100
cm3). Ϫ Chemical yield (c.y.) and enantiomeric excess (ee) analysis
of the catalysis were carried out with a Varian 3380 gas chromato-
graph, using either LIPODEX A (ex. MachereyϪNagel[21]), oc-
takis(6-O-methyl-2,3-di-O-pentyl)-γ-cyclodextrin (6-Me-2,3-pe-γ-
CD),[22] or oktakis(2,6-di-O-methyl-3-O-pentyl)-γ-cyclodextrin
(2,6-Me-3-pe-γ-CD)[23] columns using undecane as an internal
standard. Details of the chromatographic separations are given in
Table 4. Ϫ Tetrahydrofuran (THF) was distilled from Na/benzo-
phenone under argon. Diethyl ether and hexane were dried with
sodium wire. Catalytic reactions were carried out under argon, us-
ing standard Schlenk techniques. Column chromatography and
ˆ
TLC analyses were performed on silica gel, Rhone Poulenc Sorbsil
and Merck Kieselgel 60 F254ϩ366, respectively. Light petroleum
Scheme 5
2440
ether refers to the fraction with b.p. 40Ϫ60 °C.
Eur. J. Org. Chem. 2001, 2435Ϫ2446

Copper-Catalysed Asymmetric 1,4-Addition of Organozinc Compounds to Linear Aliphatic Enones
Table 4. Assay of the enantiomeric excesses (ee) of the conjugate addition products resulting from ZnEt2 addition
FULL PAPER
Enone
5a
Product
6a
Column
Programme
Elution order [min.] (hand)
6-Me-2,3-pe-γ-CD[a]
2,6-Me-3-pe-γ-CD[c]
2,6-Me-3-pe-γ-CD[c]
6-Me-2,3-pe-γ-CD
2,6-Me-3-pe-γ-CD
6-Me-2,3-pe-γ-CD
6-Me-2,3-pe-γ-CD
6-Me-2,3-pe-γ-CD
75 °C isothermal
95 °C isothermal
95 °C isothermal
65 °C isothermal
65 °C isothermal
75 °C isothermal
75 °C isothermal
75 °C isothermal
29.1 (Ϫ)-(S)[b]
30.2 (ϩ)-(R)[b]
16.2 (Ϫ)-(S)[b]
17.7 (ϩ)-(R)[b]
28.7 (Ϫ)
5b
6b
5e
6e
31.0 (ϩ)[d]
5f
6f
18.3 (Ϫ)-(S)[b]
19.0 (ϩ)-(R)[b]
25.4 (Ϫ)-(S)[b]
26.2 (ϩ)-(R)[b]
17.2 (Ϫ)-(S)[b]
21.1 (ϩ)-(R)[b]
27.1 (Ϫ)-(S)[b]
28.3 (ϩ)-(R)[b]
64.2 (Ϫ)-(S)[b]
67.4 (ϩ)-(R)[b]
5g
6f
5i[e]
5j, 5k[e]
5l[e]
12a
12b
12c
[a]
[b]
Oktakis(6-O-methyl-2,3-di-O-pentyl)-γ-cyclodextrin (ref.[22]). Ϫ
Based on comparison of the optical rotation sign with that of the
analogous methyl addition compound, the absolute stereochemistry of which has been determined (ref.[6]). Ϫ [c] Oktakis(2,6-di-O-methyl-
3-O-pentyl)-γ-cyclodextrin (ref.[23]). Ϫ Absolute stereochemistry not clear. Ϫ Determined on aldehyde after hydrolysis.
[d]
[e]
Ligand Preparations
allowed to warm to room temperature (4 h) and was then quenched
with NH4Cl(aq). The volatiles were removed under vacuum and the
residue extracted with dichloromethane in the usual way. The re-
sulting solution was dried (MgSO4) and concentrated to yield 4 as
oils. Ϫ Crude 4 was dissolved in the minimum amount of dichloro-
methane and treated with a solution of methanol (20 mL), to which
concentrated HCl (3 mL, 37% w/w) had been added. The mixture
was stirred (16 h), the solvent removed and the residue extracted
into dichloromethane. After washing with brine and drying of the
organic layer, the solvent was removed. Crude 3a was isolated by
recrystallisation from ethanol; 3b was obtained as a powder after
column chromatography (diethyl ether/light petroleum ether, 1:1);
3c was isolated as an oil after chromatography (diethyl ether/light
petroleum ether/dichloromethane, 1:1:2).
(Sa )-2,2Ј-Bis(N,N-diethylcarbamoyloxy)-3,3Ј-diphenyl-
thio-1,1Ј-binaphthalene [(Sa)-2b] via Carbamate 1a: A solution of
sBuLi in hexanes (1.3 ; 8.7 mL, 11.3 mmol) was added dropwise
over 9 min to a stirred solution of (Sa)-1a (2.5 g, 5.16 mmol) and
TMEDA (1.5 mL, 10.3 mmol) in dry THF (30 mL) at Ϫ78 °C
under an inert gas. The reaction was stirred for a further 5 min,
diphenyl disulfide (2.49 g, 11.4 mmol) was then added, and the re-
action mixture was stirred for 1 h at Ϫ78 °C. The mixture was
then brought to ambient temperature and quenched with saturated
aqueous NH4Cl solution, the volatiles were removed under high
vacuum, the residue was extracted into dichloromethane, the layers
were separated and the organic fraction was dried with MgSO4.
Column chromatography (EtOH/dichloromethane, 1:49) followed
by crystallisation from dichloromethane/light petroleum ether gave
(Ra )- 3 , 3 Ј-D i ph e ny l th i o-1, 1Ј-b in ap ht ha lene -2, 2Ј-di ol
[(Ra)-3b]: 54%, nominally Ͼ 88% ee; m.p. 58Ϫ60 °C; [α]D21 ϭ ϩ143
(c ϭ 1.01 in CHCl3). Ϫ δH (400 MHz, CDCl3): 6.32 (s, 2 H, OH),
7.16Ϫ7.38 (m, 16 H, Ar), 7.81 (d, J ϭ 8.9 Hz, 2 H, 5-H), 8.15 (s,
2 H, 4-H). Ϫ δC (67.8 MHz, CDCl3): 114.7, 121.1, 124.2, 124.7,
126.8, 127.8, 128.1, 128.5, 129.2, 129.4, 134.4, 134.8, 136.1, 150.9.
29
colourless crystals (38%); m.p. 145Ϫ146 °C; [α]D ϭ ϩ3 (c ϭ 5.0
in CHCl3). Ϫ C42H40N2S2O4 (700.9): calcd. C 72.0, H 5.75, N 4.0,
S 9.15; found C 72.1, H 5.8, N 4.1, S 9.3. Ϫ δH (400 MHz, 50 °C,
CDCl3) ϭ 0.5Ϫ0.9 (12 H, br m, CH2CH3), 2.87 (4 H, br s,
CH2CH3), 3.04 (4 H, br s, CH2CH3), 7.21Ϫ7.37 (12 H, br m, Ar),
7.47 (4 H, br m, Ar), 7.66 (2 H, br d, J 8.0, 5-H), 7.79 (2 H, br s,
4-H). Ϫ δC (100.4 MHz, CDCl3, 55 °C) ϭ 12.8, 13.4, 41.8, 42.0,
125.9, 126.0, 126.3, 127.0, 127.3, 129.2, 130.6 br, 131.9 br, 132.8,
135.4 br, 146.4, 152.5. Ϫ ν˜ (KBr disc) [cmϪ1] ϭ 3060w (Ar CϪH),
2980w, 2940w, 1720vs (CϭO), 1280s, 1220s, 1160s, 750s. Ϫ MS
(EI); m/z (%): 700 (2) [Mϩ], 591 (4), 100 (100), 72 (72) {found
(HRMS, EI) for [Mϩ] 700.2430, C42H40N2O4S2 requires 700.2430}.
Ϫ Attempted deprotection of 2b (1.50 g, 2.15 mmol) either with
MeLi/LiBr complex [21.5 mmol; on 2b as a diethyl ether solution
(11 mL), 16 h reflux] or with LiAlH4 [8.86 mmol; on 2b as a THF
solution, 16 h reflux] failed to give clean formation of 3b.
Ϫ ν (KBr disc) [cmϪ1] ϭ 3398s br (OH), 3052w (Ar CϪH),1617m,
˜
1581m, 1420m, 1267m, 755m, 742s. Ϫ MS (EI); m/z (%): 502 (100)
[Mϩ] {found (HRMS, EI) for [Mϩ] 502.1077, C32H22O2S2 re-
quires 502.1061}.
(Ra)-3,3Ј-Diethylthio-1,1Ј-binaphthalene-2,2Ј-diol [(Ra)-3c]: 64%,
21
nominally Ͼ 88% ee; oil; [α]D ϭ ϩ99 (c ϭ 1.00 in CHCl3). Ϫ δH
(400 MHz, CDCl3) ϭ 1.34 (t, J ϭ 7.4 Hz, 6 H, CH2Me), 2.93, (q,
J ϭ 7.4 Hz, 4 H, CH2Me), 6.55 (s, 2 H, OH), 7.12 (d, 2 H, J ϭ
6.8 Hz plus small unresolved long-range couplings, 8-H), 7.23 (ddd,
2 H, J ϭ 1.4, 6.8, 8.2 Hz, 6-H or 7-H), 7.31 (ddd, 2 H, J ϭ 1.2,
6.8, 8.2 Hz, 6-H or 7-H), 7.81 (d, 2 H, J ϭ 8.2 Hz plus small
unresolved long-range couplings, 5-H), 8.12 (s, 2 H, 4-H). Ϫ δC
(67.8 MHz, CDCl3) ϭ 14.6, 30.0, 113.9, 123.0, 124.0, 124.6, 127.2,
General Procedure for Lithiation of 2,2Ј-Bis(methoxymethoxy)-1,1Ј-
binaphthalene (1b), Treatment with R2Y2 (RY ؍ MeS, EtS, PhS,
and MeSe) Followed by in situ Deprotection: nBuLi (1.75 mL of a
2.5  hexane solution, 4.38 mmol) was added at room temperature
under nitrogen to a solution of 1b (0.54 g, 1.43 mmol) in anhydrous
diethyl ether. The solution was stirred (3 h) at ambient temperature.
The dilithio species was cooled (0 °C) and a solution of R2Y2
(5.00 mmol) in THF (20.0 mL) added. The resulting solution was
127.8, 129.0, 133.8, 134.3, 150.9. Ϫ ν (KBr disc) [cmϪ1] ϭ 3517m
˜
br, 3398s br (2 ϫ OH), 3052m, 2973s, 2926s, 2873m (4 ϫ CϪH),
1620m, 1496s, 1450s, 1418s, 1270s, 1146s, 749s. Ϫ MS (EI); m/z
(%): 502 (100) [Mϩ] {found (HRMS, EI) for [Mϩ] 406.1059,
C24H22O2S2 requires 406.1061}.
Eur. J. Org. Chem. 2001, 2435Ϫ2446
2441

C. Börner, M. R. Dennis, E. Sinn, S. Woodward
FULL PAPER
(Ra )-3,3Ј-Dimethylseleno-1,1Ј-binaphthalene-2,2Ј-diol
[(Ra)-3d]: 61%, nominally Ͼ 88% ee; m.p. 169 °C; [α]D ϭ ϩ114
Methyl (E)-Oct-2-enoate (5d): Yield 65%.
400 MHz) ϭ 0.87 (t, J ϭ 7.1 Hz, 3 H, CH2Me), 1.24Ϫ1.37 (m, 4
Ϫ δH (CDCl3,
21
(c ϭ 0.56 in CHCl3). Ϫ C22H18O2Se2 (472.3): calcd. C, 55.9, H 3.8; H, 2 ϫ CH2 of nC5H11), 1.41Ϫ1.49 (m, 2 H, CH2 of nC5H11), 2.19
found C 55.6, H 4.1. Ϫ δH (400 MHz, CDCl3) ϭ 2.41 (s, 6 H, (dq, 2 H, J ϭ 1.6, 7.0 Hz, ϭCHCH2), 3.71 (s, 3 H, OMe), 5.81 (dt,
SeMe), 5.98 (s, 2 H, OH), 7.10 (d, 2 H, J ϭ 6.8 Hz plus small J ϭ 15.6, 1.6 Hz, 1 H, COCHϭ), 6.97 (dt, J ϭ 15.6, 7.0 Hz, 1 H,
unresolved long-range couplings, 8-H), 7.25 (ddd, 2 H, J ϭ 1.3,
6.8, 8.2 Hz, 6-H or 7-H), 7.33 (ddd, 2 H, J ϭ 1.3, 6.8, 8.2 Hz, 6-
H or 7-H), 7.81 (d, 2 H, J ϭ 8.2 Hz plus small unresolved long-
range couplings, 5-H), 8.05 (s, 2 H, 4-H). Ϫ δC (67.8 MHz,
CH2CHϭ). Ϫ δC (CDCl3, 67.8 MHz) ϭ 13.8 (CH2Me), 22.3, 27.5,
31.1, 32.0 (4 ϫ CH2 of nC5H11), 51.1 (OMe), 120.7 (COCHϭ),
149.6 (CH2CHϭ), 167.0 (CO). Ϫ ν (thin film) [cmϪ1] ϭ 2960s,
˜
2926s, 2858s (3 ϫ CϪH), 1717s (CϭO), 1659s (CϭC), 1458m,
CDCl3) ϭ 7.6, 112.3, 121.6, 124.2, 124.6, 127.0, 127.4, 129.7, 132.3, 1436m, 1379m, 1272s, 1205s, 1175s, 1129m, 1045m, 988. Ϫ MS
132.9, 150.6. Ϫ ν (KBr disc) [cmϪ1] ϭ 3508s, 3410s br, 3339s (3 (EI); m/z (%): 156 (3) [Mϩ], 125 (29), 87 (100). Ϫ These proper-
˜
ϫ OH), 3053w (Ar CϪH), 2919w, 1570m, 1495m, 1419m, 1384m, ties[27] and others (b.p.[28]) are consistent with literature data for 5d.
1203m, 1198m, 1174m, 1136s, 750s. Ϫ MS (EI); m/z (%): 473 (42)
(E)-6-Methylhept-3-en-2-one (5g): Yield 73%.
Ϫ δH (CDCl3,
[Mϩ, 80Se], 471 (100) [found (HRMS, EI) for [Mϩ] 473.9658,
C22H18O2Se2 requires 473.9637].
400 MHz) ϭ 0.93 (d, J ϭ 6.7 Hz, 6 H, CHMe2), 1.78 (sept, 1 H,
J ϭ 7.1, 6.7 Hz, CHMe2), 2.11 (2 H, apparent dt, J ϭ 1.4, 7.1 Hz,
7.4, CH2CHMe2), 2.24 (s, 3 H, COMe), 6.07 (dt, J ϭ 15.9, 1.4 Hz,
1 H, COCHϭ), 6.78 (dt, J ϭ 15.9, 7.4 Hz, 1 H, CH2CHϭ). Ϫ
δC (CDCl3, 67.8 MHz) ϭ 22.2 (2 C, CHMe2), 26.6 (COMe), 27.7
(CHMe2), 41.5 (CH2), 132.1 (COCHϭ), 147.4 (CH2CHϭ), 198.7
(CO). Ϫ ν˜ (thin film) [cmϪ1] ϭ 2958s. 2930s, 2871s (3 ϫ CϪH),
1720sh, 1697sh, 1674s (CϭO), 1627s (CϭC), 1466m, 1362m,
1252m, 982m. Ϫ MS (FAB); m/z: 126 (8) [Mϩ], 111 (31), 84 (36), 69
(100) {found (HRMS, FAB) for [Mϩ] 126.1042, C8H14O requires
126.1045}. Ϫ These properties and others (b.p.) are consistent with
literature data for 5g.[29]
Enone Preparations and Characterisation of the Conjugate Addition
Products: Enones 5a, 5f, and 7؊8 are commercially available. The
remaining substrates were prepared by either aldol or
WittigϪHorner/WadsworthϪEmmons techniques. Except for 5i,
the formation of only the (E) isomer was confirmed by the presence
of a large vicinal coupling. For 5i, an approximate 3:1 (E)/(Z) ratio
was obtained; the (E) component was separated by column chro-
matography.
Enone Preparation by Aldol Condensation (Substrates 5b؊d, 5g؊h):
R1COMe (R1 ϭ iPr, tBu, OMe, Me; 10.0 mmol) was added at Ϫ80
°C to a solution of LDA [prepared from iPr2NH (1.5 mL,
(E)-5,5-Dimethylhex-3-en-2-one (5h): Yield 75%. Ϫ δH (CDCl3,
400 MHz) ϭ 1.08 (s, 9 H, tBu), 3.23 (s, 3 H, COMe), 5.98 (d, 1 H,
J ϭ 16.2, COCHϭ), 6.77 (d, 1 H, J ϭ 16.2, tBuCHϭ). Ϫ δC
(CDCl3, 67.8 MHz) ϭ 26.9 (COMe), 28.6 (CMe3), 33.6 (CMe3),
126.3 (COCHϭ), 157.9 (CH2CHϭ), 199.8 (CO). Ϫ ν˜ (thin film)
[cmϪ1] ϭ 2964s, 2869s (2 ϫ CϪH), 1699s, 1678s (2 ϫ CϭO), 1623s
(CϭC), 1478m, 1464m, 1361s, 1255s, 984m. Ϫ MS (EI); m/z (%):
126 (16) [Mϩ], 111 (100) {found (HRMS, EI) for [Mϩ] 126.1039,
C8H14O requires 126.1045}. Ϫ These properties and others (b.p.)
are consistent with literature data for 5h.[29]
10.5 mmol) and nBuLi (4.4 mL of 2.5
 hexane solution,
11.0 mmol), 30 min, 0 °C] in THF (20 mL) and the mixture was
stirred for 20 min. The resulting enolate was treated with R2CHO
(nC5H11, iPr, CH2iPr, tBu) and the mixture was stirred for another
hour at Ϫ80 °C. After standard workup and removal of the volat-
iles, the crude β-hydroxycarbonyl compound was dissolved in ether
(20 mL) and treated with conc. HCl (2 mL). The resultant enone
was isolated by flash chromatography (eluent: diethyl ether/light
petroleum ether, 1:5). For 5d, the initial aldol product was treated
sequentially with MeSO2Cl and DBU to effect dehydration.
Enone Preparation by Wittig؊Horner/Wadsworth؊Emmons Meth-
odology (Substrates 5e, 5i؊l): Appropriate aldehydes (4.00 mmol)
were refluxed in THF (10 mL) in the presence of either Ph3Pϭ
CH(COMe)[30] (4.00 mmol) or the ylide derived from (MeO)2-
P(O)CH2COCH2OMe[31] and K2CO3 in toluene for 5e. The result-
ant enones were isolated by flash chromatography (eluent: diethyl
ether/light petroleum ether, 1:1).
(E)-2-Methyldec-4-en-3-one (5b): Yield 82%.
Ϫ δH (CDCl3,
400 MHz) ϭ 0.89 (t, J ϭ 7.1 Hz, 3 H, CH2Me), 1.10 (d, J ϭ 6.9
Hz, 3 H, CHMe2), 1.23Ϫ1.35 (m, 4 H, 2 ϫ CH2 of nC5H11),
1.35Ϫ1.50 (m, 2 H, CH2 of nC5H11), 2.20 (dq, 2 H, J ϭ 1.5, 7.5
Hz, ϭCHCH2), 2.81 (sept, 1 H, J ϭ 6.9, CHMe2), 6.15 (dt, J ϭ
15.7, 1.5 Hz, 1 H, COCHϭ), 6.87 (dt, J ϭ 15.7, 7.0 Hz, 1 H,
CH2CHϭ). Ϫ δC (CDCl3, 67.8 MHz) ϭ 13.9 (CH2Me), 18.5
(CHMe2), 22.4, 27.8, 31.3, 32.4 (4 ϫ CH2 of n-C5H11), 38.4
(CHMe2), 128.3 (COCHϭ), 147.3 (CH2CHϭ), 204.1 (CO). Ϫ MS
(EI); m/z (%): 168 (7) [Mϩ], 140 (8), 125 (100). Ϫ These properties
and others (b.p., IR spectrum) are consistent with literature data
for 5b.[24]
(E)-1-Methoxynon-3-en-2-one (5e): Yield 78%. Ϫ δH (CDCl3,
400 MHz) ϭ 0.89 (t, J ϭ 7.1 Hz, 3 H, CH2Me), 1.24Ϫ1.38 (m, 4
H, 2 ϫ CH2 of nC5H11), 1.42Ϫ1.51 (m, 2 H, CH2 of nC5H11), 2.21
(dq, 2 H, J ϭ 1.5, 6.9 Hz, ϭCHCH2), 3.42 (s, 3 H, OMe), 4.16 (s,
2 H, COCH2). Ϫ 6.25 (dt, J ϭ 15.9, 1.5 Hz, 1 H, COCHϭ), 6.97
(dt, J ϭ 15.9, 6.9 Hz, 1 H, CH2CHϭ). Ϫ δC (CDCl3, 67.8 MHz) ϭ
13.7 (CH2Me), 22.2, 27.4, 31.1, 32.4 (4 ϫ CH2 of nC5H11), 59.0
(OMe), 76.4 (CH2O), 125.8 (COCHϭ), 148.7 (CH2CHϭ), 196.7
(E)-2,2-Dimethyldec-4-en-3-one (5c): Yield 71%. Ϫ δH (CDCl3,
400 MHz) ϭ 0.85 (t, J ϭ 7.1 Hz, 3 H, CH2Me), 1.11 (s, 9 H, tBu),
1.24Ϫ1.34 (m, 4 H, 2 ϫ CH2 of nC5H11), 1.40Ϫ1.48 (m, 2 H, CH2
of nC5H11), 2.16 (dq, 2 H, J ϭ 1.5, 7.1 Hz, ϭCHCH2), 6.46 (dt,
J ϭ 15.3, 1.5 Hz, 1 H, COCHϭ), 6.90 (dt, J ϭ 13.3, 7.0 Hz, 1 H,
CHϭCH2). Ϫ δC (CDCl3, 67.8 MHz) ϭ 13.9 (CH2Me), 22.3 (CH2),
26.1 (tBu), 27.8, 31.3, 32.4 (3 ϫ CH2 of nC5H11), 38.0 (CMe3),
(CO). Ϫ ν (thin film) [cmϪ1] ϭ 2970m, 2932s, 2874s (3 ϫ CϪH),
˜
1701s (CϭO), 1654s (CϭC), 1458m, 1420m, 1384m, 1200m,
1126m. Ϫ MS (EI); m/z (%): 170 (4) [Mϩ], 147 (27), 125 (36) [Mϩ
Ϫ CH2OMe] {found (HRMS, EI) for [Mϩ] 170.1260, C10H18O2
requires 170.1307; for [Mϩ Ϫ CH2OMe] 125.0963, C8H14O re-
quires 125.0966}.
˜
124.0 (COCHϭ), 147.5 (CH2CHϭ), 204.1 (CO). Ϫ ν (thin film)
[cmϪ1] ϭ 2992m, 2959s, 2861s (3 ϫ CϪH), 1711s, 1691s (2 ϫ Cϭ
O), 1625m (CϭC), 1477m, 1467m, 1366m, 1084m, 990. Ϫ MS (EI);
m/z (%): 165 (100) [Mϩ Ϫ tBu] [found (HRMS, FAB) for [Mϩ]
182.1665, C12H22O requires 182.1671]. Ϫ Compound 5c has been
described in the literature[25,26] but no spectroscopic or physical
data were reported.[27]
(E)-5,5-Dimethoxypent-3-en-2-one (5i): Yield 82%. Ϫ δH (CDCl3,
400 MHz) ϭ 2.28 (s, 3 H, COMe), 3.35 (s, 6 H, OMe), 4.95 [1 H,
dd, J ϭ 1.3, 4.1 Hz, CH(OMe)2] 6.32 (dd, J ϭ 16.2, 1.3 Hz, 1 H,
CHCHϭ), 6.56 (dd, J ϭ 16.2, 4.1 Hz, 1 H, COCHϭ). Ϫ δC
(CDCl3, 67.8 MHz) ϭ 27.1 (COMe), 52.8 (2 C, OMe), 100.9
2442
Eur. J. Org. Chem. 2001, 2435Ϫ2446

Copper-Catalysed Asymmetric 1,4-Addition of Organozinc Compounds to Linear Aliphatic Enones
FULL PAPER
[CH(OMe)2], 132.7 (COCHϭ or CHCHϭ), 140.9 (COCHϭ or ring), 2.15 (2 H, apparent tquint, J ϭ 7.6, 1.5 Hz, CH2CHϭ), 2.32
CHCHϭ), 198.2 (CO). Ϫ ν (thin film) [cmϪ1] ϭ 2994m, 2938s, (t, J ϭ 7.5 Hz, 2 H, COCH2 in ring), 2.58 (2 H, tdt, J ϭ 7.3, 2.5
˜
2832s (3 ϫ CϪH), 1701s, 1682s (2 ϫ CϭO), 1642m (CϭC), 1360s, Hz, 1.3), 6.51 (1 H, J ϭ 7.6, 2.5 Hz, ϭCH). Ϫ (δC (CDCl3,
1257s, 1131s (CϪO), 1057s, 982. Ϫ MS (EI); m/z (%): 144 (1) [Mϩ],
67.8 MHz): 11.8 (CH2Me), 18.8 (CH2 of ring), 22.0 (ϭCHCH2),
129 (35), 113 (100). Ϫ C7H12O3 (144.2): calcd. C 58.32, H 8.39; 25.6 (CH2 of ring), 27.9, 29.5, 31.4 (3 ϫ CH2 of nC5H11), 37.6
found C 58.13, H 8.39. Ϫ Compound 5i has been described in the
(COCH2 of ring), 135.7 (COCϭ), 136.5(CH2CHϭ), 206.3 (CO). Ϫ
ν˜ (thin film) [cmϪ1] ϭ 2966s, 2877s (2 ϫ CϪH), 1717s (CϭO),
1651m (CϭC), 1462m, 1410m, 1286m, 1225m, 1202m, 1142m,
993s, 732m. Ϫ MS (CI); m/z (%): 124 (27) [Mϩ], 95 (22) {found
(HRMS, CI) for [Mϩ] 124.0885, C8H12O requires 112.0888}.
literature but no spectroscopic or physical data were reported.[32]
(E)-7,7-Diethoxyhex-3-en-2-one (5j): Yield 55% (two steps, starting
from alcohol). Ϫ δH (400 MHz, CDCl3) ϭ 1.19 (t, J ϭ 7.1 Hz, 6
H, OCH2Me), 2.26 (s, 3 H, COMe), 2.54 (ddd, 2 H, J ϭ 7.1, 5.5
Hz, 1.5, CH2CHϭ), 3.49 (dq, 2 H, J ϭ 9.4, 7.1 Hz, OCH2α), 3.63
(dq, 2 H, J ϭ 9.4, 7.1 Hz, OCH2β), 4.58 [1 H, t, J ϭ 5.5, CH(OEt)2],
6.10 (dt, J ϭ 16.1, 1.4 Hz, 1 H, COCHϭ), 6.73 (dt, J ϭ 16.1, 7.1
Hz, 1 H, CH2CHϭ). Ϫ δC (CDCl3, 67.8 MHz) ϭ 15.1 (2 C,
CH2Me), 26.7 (COMe), 37.1 (CH2CHϭ), 61.5 (2 C, OCH2Me),
101.2 [CH(OEt)2], 133.4 (COCHϭ), 142.6 (CH2CHϭ), 198.4 (CO).
Ϫ ν˜ (thin film) [cmϪ1] ϭ 2977s, 2930m, 2882m (3 ϫ CϪH), 1700m,
1677s (2 ϫ CϭO), 1630m (CϭC), 1372m, 1360m, 1255m, 1126s
(CϪO), 1062s, 978m. Ϫ MS (EI); m/z (%): 142 (1) [Mϩ Ϫ EtOH],
113 (29), 103 (100) {found (HRMS, EI) for [Mϩ Ϫ EtOH]
141.0916, C8H13O2 requires 141.0916}.
(E)-2-Hexylidenecyclopentanone (9b): Synthesis by literature
methods.[36,37] Yield 50%. Ϫ δH (400 MHz, CDCl3) ϭ 0.88 (t, J ϭ
6.9 Hz, 3 H, CH2Me), 1.23Ϫ1.35 (m, 4 H, 2 ϫ CH2 of nC5H11),
1.40Ϫ1.49 (m, 2 H, CH2 of nC5H11), 1.92 (2 H, quint, J ϭ 7.5,
central CH2 of ring), 2.13 (2 H, apparent qt, J ϭ 7.5, 1.5 Hz,
CH2CHϭ), 2.43 (t, J ϭ 7.5 Hz, 2 H, COCH2 in ring), 2.57 (2 H,
tdt, J ϭ 7.5, 2.7 Hz, 1.5), 6.54 (1 H, J ϭ 7.5, 2.7 Hz, ϭCH). Ϫ δC
(CDCl3, 67.8 MHz) ϭ 13.9 (CH2Me), 19.7 (CH2 of ring), 22.4
(CH2 of nC5H11), 26.6 (CH2 of ring), 27.9, 29.5, 31.4 (3 ϫ CH2 of
nC5H11), 38.5 (COCH2 of ring), 136.3 (CH2CHϭ), 137.1 (COCHϭ),
207.1 (CO). Ϫ ν (thin film) [cmϪ1] ϭ 2950s, 2853s (2 ϫ CϪH),
˜
1723s (CϭO), 1654m (CϭC), 1460m, 1231m, 1185m, 974s. Ϫ MS
(FAB); m/z: 167 (84) [Mϩ ϩ H], 112 (100) {found (HRMS, FAB)
for [Mϩ ϩ H] 167.1439, C11H19O requires 167.1436}. Ϫ These
properties and others (b.p.) are consistent with literature data for
9b. The results of an NOE study between the ring 3-CH2 and the
methylene functions of the C5H11 chain are consistent with isola-
tion of only the (E) double-bond isomer.
(E)-5-(1,3-Dioxolan-2-yl)pent-3-en-2-one (5k): Yield 63% (two steps,
starting from alcohol). Ϫ δH (400 MHz, CDCl3) ϭ 2.26 (s, 3 H,
COMe), 2.60 (ddd, 2 H, J ϭ 1.4, 4.6 Hz, 6.8, ϭCHCH2), 3.84Ϫ4.03
(m, 4 H, OCH2CH2O), 5.00 (t, J ϭ 4.6 Hz, 1 H, CHCH2), 6.17
(dt, J ϭ 16.0, 1.4 Hz, 1 H, CHCO), 6.78 (dt, J ϭ 10.0, 6.8 Hz,
1 H, ϭCHCH2). Ϫ δC (CDCl3, 67.8 MHz) ϭ 26.7 (COMe), 37.0
(CH2CHϭ), 64.9 (2 C, OCH2), 102.3 [CH(OCH2)2], 133.9
(COCHϭ), 141.1 (CH2CHϭ), 198.1 (CO). Ϫ ν (thin film) [cmϪ1] ϭ
Structural Confirmation of the Isolated 1,4-Addition Products: The
conjugate addition products were isolated directly from the cata-
lytic reactions by flash chromatography. Because of the small scale
of the reactions and the presence of undecane, internal standard
accurate analytical or [α]D values could not be obtained in all cases.
˜
2968m, 2889s (2 ϫ CϪH), 1715m, 1676s 1677s (2 ϫ CϭO), 1631
(CϭC), 1363m, 1257m, 1137s (CϪO), 1036m, 977m. Ϫ MS (CI);
m/z (%): 155 (33) [Mϩ Ϫ H], 87 (100) {found (HRMS, FAB) for
[Mϩ ϩ H] 157.0865, C8H13O3 requires 157.0865}.
(؉)-4-Ethylnonan-2-one (6a): δH (400 MHz, CDCl3) ϭ 0.82 (t, J ϭ
7.4 Hz, 3 H, Me of C5-Et), 0.87 (t, J ϭ 6.8 Hz, 3 H, Me of nC5H11),
1.18Ϫ1.39 (m, 10 H, 4 ϫ CH2 of nC5H11 and CH2 of C5-Et), 1.85
(apparent sept, 1 H, J ϭ 6.3, CHEt), 2.12 (s, 3 H, COMe), 2.33 (2
H, J ϭ 7.8, COCH2). Ϫ δC (CDCl3, 67.8 MHz) ϭ 12.8 (Me of C5-
Et), 16.0 (Me of n-C5H11), 24.6, 28.2, 28.3 (2 ϫ CH2 of nC5H11
and C5-Et), 32.3 (CHEt), 34.1, 35.4 (2 ϫ CH2 of nC5H11), 37.3
(E)-7,7-Diethoxyhept-3-en-2-one (5l): Yield 58% (two steps, starting
from alcohol). Ϫ δH (400 MHz, CDCl3) ϭ 1.13 (t, J ϭ 7.1 Hz, 6
H, OCH2Me), 1.68Ϫ1.73 (m, 2 H, ϭCHCH2CH2), 2.16 (s, 3 H,
COMe), 2.21Ϫ2.27 (m, 2 H, H2CHϭ), 3.38Ϫ3.46 (m, 2 H,
OCH2α), 3.53Ϫ3.62 (m, 2 H, OCH2β), 4.42 (t, J ϭ 5.6 Hz, 1 H,
CH(OEt)2], 6.01 (dt, J ϭ 16.0, 1.5 Hz, 1 H, COCHϭ), 6.76 (dt,
J ϭ 16.0, 6.8 Hz, 1 H, CH2CHϭ). Ϫ δC (CDCl3, 67.8 MHz) ϭ
15.1 (2 C, CH2Me), 26.7 (COMe), 27.5 (CH2CH2CHϭ), 37.1
(CH2CHϭ), 61.2 (2 C, OCH2Me), 101.9 [CH(OEt)2], 131.2
(COMe), 50.5 (COCH2), 211.4 (CO). Ϫ ν (thin film) [cmϪ1] ϭ
˜
2959s, 2927s, 2873m, 2858m (4 ϫ CϪH), 1717s (CϭO), 1462m,
1355m, 1165m. Ϫ MS (EI); m/z (%): 171 (100) [Mϩ ϩ H], 141 (18)
[Mϩ Ϫ Et], 112 (19), 99 (18) {found (HRMS, EI) for [Mϩ]
170.1761, C11H22O requires 170.1761}. Ϫ These properties are con-
sistent with literature structure and data.[6]
(COCHϭ), 147.5 (CH2CHϭ), 198.4 (CO). Ϫ ν (thin film) [cmϪ1] ϭ
˜
2975s, 2930m, 2879m (3 ϫ CϪH), 1698m, 1676s (2 ϫ CϭO),
1628m (CϭC), 1444m, 1361m, 1254m, 1128s (CϪO), 1062s, 979m.
Ϫ MS (EI); m/z (%): 155 (12) [Mϩ Ϫ OEt], 123 (16), 111 (23), 109
(30), 95 (46) {found (HRMS, FAB) for [Mϩ Ϫ EtOH]155.1062,
C9H15O2 requires 155.1072}.
(؉)-5-Ethyl-2-methyldecan-3-one (6b): δH (400 MHz, CDCl3) ϭ
0.83 (t, J ϭ 7.4 Hz, 3 H, Me of C5-Et), 0.86 (t, J ϭ 6.9 Hz, 3 H,
Me of nC5H11), 1.06 (d, J ϭ 6.9 Hz, 6 H, CHMe2), 1.12Ϫ1.35 (m,
10 H, 4 ϫ CH2 of nC5H11 and CH2 of C5-Et), 1.87 (apparent sept,
1 H, J ϭ 6.3, CHEt), 2.34 (2 H, ABX, J ϭ 7.7, 6.3 Hz, COCH2),
2.57 (sept, 1 H, J ϭ 6.9, CHMe2). Ϫ δC (CDCl3, 67.8 MHz) ϭ 11.1
(Me of C5-Et), 14.3 (Me of n-C5H11), 18.5 (CHMe2), 22.9, 26.6,
26.65, 32.4, 33.8 (5 ϫ CH2 of nC5H11 and C5-Et), 35.2 (CHEt),
Preparation of the Precursor Aldehydes: For enones 5jϪl, the re-
quired aldehydes were obtained from (EtO)2CH(CH2)nCHϭCH2
(n ϭ 0, 1) or 2-vinyl-[1,3]dioxolane (5.00 mmol) by hydroboration
under literature conditions to give the known alcohols.[32Ϫ34] The
alcohols (1.16 mmol) were oxidised under Swern conditions
[DMSO (181 mg, 2.32 mmol), oxalyl chloride (0.15 mL,
1.73 mmol) and NEt3 (0.56 mL, 4.06 mmol)]. The derived alde-
hydes,[32,33,35] proving rather reactive, were used directly as crude
products in the organophosphorus couplings. The aldehyde
(MeO)2CHCHO is commercially available.
41.4 (CHMe2), 45.4 (COCH2), 215.6 (CO). Ϫ ν (thin film) [cmϪ1] ϭ
˜
2961s, 2928s, 2874m, 2855m (4 ϫ CϪH), 1713s (CϭO), 1460m,
1383m, 1045m. Ϫ MS (FAB); m/z (%): 198 (10) [Mϩ], 155 (55), 86
(68), 71 (100) {found (HRMS, FAB) for [Mϩ] 198.1984, C13H26O
requires 198.1987}.
(E)-2-Propylidenecyclopentanone (9a): Synthesis by literature (؊)-4-Ethyl-1-methoxynonan-2-one (6e): δH (400 MHz, CDCl3) ϭ
methods.[36,37] Yield 50%. Ϫ δH (400 MHz, CDCl3) ϭ 1.05 (t, J ϭ 0.84 (t, J ϭ 7.5 Hz, 3 H, Me of C5-Et), 0.86 (t, J ϭ 6.8 Hz, 3 H,
7.6 Hz, 3 H, CH2Me), 1.92 (2 H, quint, J ϭ 7.5, central CH2 of
Eur. J. Org. Chem. 2001, 2435Ϫ2446
Me of nC5H11), 1.15Ϫ1.38 (m, 10 H, 4 ϫ CH2 of nC5H11 and CH2
2443

C. Börner, M. R. Dennis, E. Sinn, S. Woodward
FULL PAPER
of C5-Et), 1.88 (apparent sept, 1 H, J ϭ 6.4, CHEt), 4.69 (2 H, ent sept, J ϭ 6.3, CHEt), 2.11 (s, 3 H, COMe), 2.40 (2 H, ABX,
ABX, J ϭ 7.3, 6.4 Hz, COCH2), 3.40 (s, 3 H, OMe), 3.98 (s, 2 H,
16.4, 6.6, COCH2), 3.40Ϫ3.50 (m, 2 H, OCH2Me), 3.56Ϫ3.66 (m,
CH2OMe). Ϫ δC (CDCl3, 67.8 MHz) ϭ 10.7 (Me of C4-Et), 14.0 2 H, OCH2Me), 4.52 (t, J ϭ 6.0 Hz, 1 H, CH[OEt]2). Ϫ Deprotec-
(Me of nC5H11), 22.6, 26.2, 26.3, 32.0, 33.4 (5 ϫ CH2 of nC5H11
tion with dil. HCl (2.0 ) provided the aldehyde. Ϫ [α]D22ϭ ϩ70
and C5-Et), 34.9 (CHEt), 43.2 (COCH2CH), 59.2 (OMe), 78.0 (c ϭ 0.57, CHCl3). Ϫ δH (400 MHz, CDCl3) ϭ 0.91 (t, J ϭ 7.4 Hz,
(COCH2OMe), 208.7 (CO). Ϫ ν (thin film) [cmϪ1] ϭ 2959s, 2929s, 3 H, Me of C4-Et), 1.32Ϫ1.46 (m, 2 H, CH2Me), 2.14 (s, 3 H,
˜
2852m (3 ϫ CϪH), 1708s (CϭO), 1462m, 1406m, 1285m, 1192m, COMe), 2.26Ϫ2.38 (m, 5 H, CHOCH2CHCH2CO), 9.74 (s, 1 H,
935m. Ϫ MS (FAB); m/z (%): 200 (3) [Mϩ], 155 (79) [Mϩ
Ϫ
CHO). Ϫ MS (FAB); m/z (%): 142 (1) [Mϩ], 113 (14), 99 (26), 83
(20), 59 (23), 58 (100) {found (HRMS, FAB) for [Mϩ]142.0996,
C8H14O2 requires 142.0994}. Ϫ Compound 12b was also obtained
from catalytic reactions using 5k.
CH2OMe], 85 (37), 71 (100) {found (HRMS, FAB) for [Mϩ]
200.1775, C12H24O2 requires 200.1776}.
(؉)-4-Ethyl-5-methylhexan-2-one (6f): δH (400 MHz, CDCl3) ϭ
22
0.82 (d, J ϭ 6.7 Hz, 3 H, CHMe2α), 085 (d, J ϭ 6.7 Hz, 3 H, (؉)-4-Ethyl-6-oxoheptanal (12c): [α]D ϭ ϩ9 (c ϭ 0.33, CHCl3).
CHMe2β), 0.86 (t, J ϭ 7.5 Hz, 3 H, Me of C4-Et), 1.15Ϫ1.26 (m, Ϫ δH (400 MHz, CDCl3) ϭ 0.82 (t, J ϭ 7.4 Hz, 3 H, Me of C4-
1 H, CH2αMe), 1.29Ϫ1.40 (m, 1 H, CH2βMe), 1.67Ϫ1.81 (m, 2 H,
CHMe2 and CHEt), 2.14 (s, 3 H, COMe), 2.25 (dd, J ϭ 16.2, 7.3 CHOCH2CH2), 1.86 (1 H, apparent sept, J ϭ 6.4, CHEt), 2.10 (s,
Hz, 1 H, COCH2α), 2.38 (dd, J ϭ 16.2, 5.4 Hz, 1 H, COCH2β). Ϫ 3 H, COMe), 2.28 (dd, 1 H, J ϭ 6.9, 16.7 Hz, COCH2α), 2.35Ϫ2.42
δC (CDCl3, 67.8 MHz) ϭ 12.0 (Me of C4-Et), 18.7 (CHMe2α), 19.8 (m, 3 H, CHOCH2 and COCH2β), 9.72 (t, J ϭ 1.7 Hz, 1 H, CHO).
Et), 1.18Ϫ1.37 (m, 2 H, CH2Me), 1.49Ϫ1.63 (m, 2 H,
(CHMe2β), 24.2 (CH2Me), 29.5, 30.5 (CHMe2 and CHEt), 41.4
Ϫ δC (CDCl3, 67.8 MHz) ϭ 10.6 (Me of C4-Et), 25.4, 26.0 (2 ϫ
(COMe), 45.3 (COCH2), 209.7 (CO). Ϫ ν (thin film) [cmϪ1] ϭ CH2), 30.4 (CHEt), 34.3 (COMe), 41.2, 47.7 (2 ϫ CH2), 202.2
2961s, 2933s, 2894m, 2875m (4 ϫ CϪH), 1719s (CϭO), 1466m, (CHO), 208.4 (CO). Ϫ ν˜ (thin film) [cmϪ1] ϭ 2963m, 2931m, 2876w
1369m, 1355m, 1166m. Ϫ MS (ES); m/z (%): 142 (1) [Mϩ], 99 (19), (3 ϫ CϪH), 1717s (2 ϫ CϭO), 1412m, 1357m, 1163m. Ϫ MS
˜
84 (64), 69 (52), 58 (55), 43 (100) {found (HRMS, EI) for [Mϩ
ϩ
(FAB); m/z (%): 156 (2) [Mϩ], 113 (21), 98 (31), 81 (26), 58 (100)
H] 143.1436, C9H19O requires 143.1436}. Ϫ This compound has {found (HRMS, FAB) for [Mϩ] 156.1152, C9H16O2 requires
been described in the literature but no spectroscopic or physical
156.1150}. Ϫ The precursor acetal (6l) was not isolated or charac-
terised.
data have been reported.[4,38,39]
(؉)-4-Ethyl-6-methylheptan-2-one (6g): δH (400 MHz, CDCl3) ϭ 2-(1-Ethylpropyl)cyclopentanone (10a): Isolated as a 1:1 mixture of
0.85 (t, J ϭ 7.5 Hz, 3 H, Me of C4-Et), 0.86 (d, J ϭ 6.0 Hz, 3 H,
enantiomers at C-2. Ϫ δH (400 MHz, CDCl3) ϭ 0.84, (t, J ϭ 7.4
CHMe2α), 088 (d, J ϭ 6.3 Hz, 3 H, CHMe2β), 1.01 (dt, J ϭ 13.6, Hz, 3 H, Me), 0.89 (t, J ϭ 7.4 Hz, 3 H, Me), 1.19 (m, 3 H),
6.8 Hz, 1 H, CH2α-iPr), 1.16 (dt, J ϭ 13.6, 6.8 Hz, 1 H, CH2β 1.50Ϫ1.79 (m, 4 H), 1.98Ϫ2.11 (m, 3 H), 2.15Ϫ2.22 (m, 1 H),
-
iPr), 1.22Ϫ1.38 (m, 2 H, CH2Me), 1.59 (1 H, J ϭ 6.8, CHCH2Me) 2.28Ϫ2.37 (m, 1 H). Ϫ δC (CDCl3, 67.8 MHz) ϭ 13.7, 14.2 (2 ϫ
1.88Ϫ1.98 (m, 1 H, CHMe2), 2.13 (s, 3 H, COMe), 2.31 (apparent CH2Me), 22.6, 25.8, 26.2, 26.5 (4 ϫ CH2, 2 ϫ CH2Me, C-3, C-4),
˜
sept, 2 H, ABX, J ϭ 7.1, 6.8 Hz, COCH2). Ϫ δC (CDCl3, 41.2 (C-5), 42.5 (exo-CH), 53.7 (ring-CH), 223.0 (CO). Ϫ ν (thin
67.8 MHz) ϭ 10.5 (Me of C4-Et), 22.6 (CHMe2α), 22.8 (CHMe2β), film) [cmϪ1] ϭ 2963s, 2935s, 2876m (3 ϫ CϪH), 1734s (CϭO),
25.1 (CHMe2 or CHEt), 26.4 (CH2Me), 29.5, 30.5 30.3 (COMe), 1462m, 1407w, 1380w, 1272w, 1150m, 919w, 734m. Ϫ MS (CI); m/
32.9 (CHMe2 or CHEt), 43.2 (CH2-iPr), 48.4 (COCH2), 209.1 z (%): 309 (100) [Mϩ Ϫ dimer], 154 (11) [Mϩ], 137 (35), 125 (10),
(CO). Ϫ ν˜ (thin film) [cmϪ1] ϭ 2960s, 2915s, 2896m, 2871m (4 ϫ
84 (18) {found (HRMS, CI) for [Mϩ ϩ H] 155.1416, C10H19O re-
CϪH), 1710s (CϭO), 1459m, 1362m, 1162m. Ϫ MS (ES); m/z (%): quires 155.1436}.
156 (1) [Mϩ], 98 (38), 69 (58), 58 (100) {found (HRMS, FAB) for
(؊)-2-(1-Ethylhexyl)cyclopentanone (10b): Isolated as a 1:1 mixture
[Mϩ] 156.5111, C10H20O requires 156.1514}.
of diastereomers at C-2, ee of the addition determined by 13C
25
4-Ethyl-5,5-dimethoxypentan-2-one (6i) and (؉)-2-Ethyl-4-oxopen-
NMR of CBS-reduced 10b (see below). Ϫ [α]D ϭ Ϫ0.3 (c ϭ 1.6,
tanal (12a): The initial kinetic acetal (6i) was observed if the reac- CHCl3). Ϫ δH (400 MHz, CDCl3) ϭ 0.83, 0.86, 0.875, 0.88 (4
tion mixture was quenched with buffer (pH ϭ7). Ϫ δH (400 MHz,
closely overlapping methyl ssignals each 3 H, t, J ϭ 7.4, Me of C4-
CDCl3) ϭ 0.88 (t, J ϭ 7.5 Hz, 3 H, Me of C4-Et), 1.27 (1 H, Et and Me of nC5H11); the rest of the spectrum is not informative
dquint, J ϭ 13.8, 7.5 Hz, CH2αMe), 1.49 (ddq, 1 H, J ϭ 13.8, 5.2 and composes an envelope of signals in the range 1.15Ϫ2.40. Ϫ δC
Hz, 7.5, CH2βMe), 2.14 (s, 3 H, COMe), 2.20Ϫ2.31 (m, 2 H, (CDCl3, 67.8 MHz) ϭ 12.2, 12.6, 14.3 (2 C) [4 ϫ CH2Me], 21.0,
COCH2α and CHEt), 2.53Ϫ2.59 (m, 1 H, COCH2β), 3.32 (s, 3 H, 22.8, 22.9, 24.5, 24.6, 24.65, 25.4, 27.3, 27.8, 31.6, 32.2, 32.3, 32.4
OMeα), 3.35 (s, 3 H, OMeβ), 4.16 [d, J ϭ 5.2 Hz, 1 H, CH(OMe)2]. (14 ϫ CH2, including overlapping signals), 39.1, 39.2 (2 ϫ CH),
˜
Ϫ After deprotection with dil. HCl (2.0 ), the aldehyde (7a) dis-
39.6 (2 C, 2 ϫ CH2), 52.2, 50.6 (2 ϫ CH), 222.2 (2 C, CO). Ϫ ν
played a 1H NMR spectrum consistent with that in the literat-
(thin film) [cmϪ1] ϭ 2959s, 2927s, 2874m, 2855m, (4 ϫ CϪH),
ure:[40] δH (400 MHz, CDCl3) ϭ 0.95 (t, J ϭ 7.5 Hz, 3 H, Me of 1735s (CϭO), 1465m, 1406w, 1378w, 1271w, 1150m. Ϫ MS (EI);
C4-Et), 1.51Ϫ1.61 (m, 1 H, CH2αMe), 1.70Ϫ1.81 (m, 1 H, m/z (%): 196 (1) [Mϩ], 167 (1) [Mϩ Ϫ Et], 125 (4) [Mϩ Ϫ pentyl], 84
CH2βMe), 2.19 (s, 3 H, COMe), 2.40Ϫ2.48 (m, 1 H, COCH2α), (100) [pentanone] {found (HRMS, EI) for [Mϩ] 196.1831, C13H24O
2.82Ϫ2.92 (m, 2 H, COCH2β and CHEt), 9.71 (s, 1 H, CHO). Ϫ requires 196.1827}.
The compound could not be isolated analytically pure and was
Reduction of 10b by BH3·THF with (R)-CBS Catalysis and ee De-
cyclised directly (see later).
termination: BH3·THF (1  in THF, 0.44 mL, 0.44 mmol) was ad-
4-Ethyl-5,5-diethoxyhexan-2-one (6j) and (؉)-3-Ethyl-5-oxohexanal
ded at room temperature to a solution of (R)-methyloxazaborolid-
(12b): The initial kinetic acetal (6j) was observed if the reaction ine (1  in toluene, 0.44 mL, 0.44 mmol) in dry dichloromethane.
mixture was quenched with buffer (pH ϭ 7). Ϫ δH (400 MHz, After the mixture had been stirred for 1 h, a solution of 10b
CDCl3) ϭ 0.85 (t, J ϭ 7.4 Hz, 3 H, Me of C4-Et), 1.16 (t, J ϭ 7.1 (0.44 mmol) in dry dichloromethane (0.5 mL) was added over a
Hz, 3 H, OCH2Me) overlapped by 1.16 (t, J ϭ 7.1 Hz, 3 H,
period of 20 min by syringe pump. The solution was stirred over-
OCH2Me), 1.28Ϫ1.39 (m, 2 H, CH2 of C4-Et), 1.46Ϫ1.53 (m, 1 H, night, quenched with methanol and stirred for a further hour. The
CHCH2αCH), 1.58Ϫ1.64 (m, 1 H, CHCH2βCH), 2.01 (1 H, appar- volatiles were removed under vacuum and the crude product puri-
2444
Eur. J. Org. Chem. 2001, 2435Ϫ2446

Copper-Catalysed Asymmetric 1,4-Addition of Organozinc Compounds to Linear Aliphatic Enones
FULL PAPER
fied by flash chromatography (diethyl ether/pentane, 1:3) to give (3R,5S)-3-Ethyl-5-hydroxycyclohexanone (3R,5S)-(14): Yield 5%
alcohol 11 (74%). Isolated as a 1:1 mixture of diastereomers at C-
(70% ee). Ϫ δH (400 MHz, CDCl3) ϭ 0.93 (t, J ϭ 7.4 Hz, 3 H,
2, ee value of the ethyl addition determined by 13C NMR: 72Ϫ86%, CH2Me), 1.39Ϫ1.49 (m, 2 H, CH2Me), 1.51Ϫ1.64 (m, 2 H, C4H2),
depending on which pair of peaks was measured. Ϫ δH (400 MHz,
CDCl3) ϭ 0.81, 0.85, 0.87, 0.90 (4 closely overlapping methyl sig-
1.73 (d, J ϭ 3.9 Hz, 1 H, OH), 1.94 (dt, 1 H, J ϭ 1.1, 14.0 Hz,
C6Hax), 2.22 (m, 1 H, C3Hax), 2.33 (ddd, 1 H, J ϭ 1.1, 11.4 Hz,
nals each 3 H, t, J ϭ 7.3, Me of C4-Et and Me of nC5H11), 3.95 13.5, C2Hax), 2.38 (ddt, 1 H, J ϭ 14.1, 3.9 Hz, 2.0, C6Heq), 2.73
and 4.23 (1 H, 2 ϫ m, C1-H); the rest of the spectrum was not (ddt, 1 H, J ϭ 13.5, 4.8 Hz, 2.0, C2Heq), 3.93 (m, 1 H, C5Hax).
informative, comprising an envelope of signals in the range
1.15Ϫ1.90. Ϫ δC (CDCl3, 67.8 MHz) ϭ 9.8, 11.0 (CH3, minor iso-
mer: 9.4, 11.5), 14.2 (CH3), 21.6, 22.8, 23.0, 23.1, (minor isomer:
24.5), 25.2 (minor isomer: 25.5), (minor isomer: 26.6) 26.9, (minor
isomer: 27.1) 27.3, (minor isomer: 27.9) 28.1, 30.2, 30.3, 31.3, 32.4,
32.5, 34.8, 36.0 (8 ϫ CH2, including overlapping signals), (minor
isomer: 37.7) 37.8, 41.4 (CH), 49.1, (minor isomer: 51.3) 51.4 (CH),
73.8, (minor isomer: 76.2) 76.9 (C-1). Ϫ MS (FAB); m/z (%): 181
(5) [Mϩ Ϫ OH], 85 (13) [cyclopentanol] {found (HRMS, FAB) for
[Mϩ Ϫ OH] 181.1927, C13H25 requires 181.1956}.
X-ray Crystallography of (Sa)-2b: Colourless prisms were grown
from dichloromethane/light petroleum ether, C42H40O4N2S2, M ϭ
700.9, orthorhombic, space group P21212, a ϭ 12.154(8), b ϭ
3
˚
˚
16.732(3), c ϭ 9.286(6) A, V ϭ 1888(3) A , Z ϭ 2, Dc ϭ 1.23 g
cmϪ3, µ(Mo-Kα) ϭ 1.75 cmϪ1, F(000) ϭ 740, T ϭ 296 K, prism
0.50 ϫ 0.25 ϫ 0.25 mm. Measurements were made as previously
described,[43] using a Rigaku AFC6S diffractometer with graphite-
˚
monochromated Mo-Kα radiation (λ ϭ 0.71069 A) using ω-2θ
scans for 2152 unique reflections. A direct-method solution was
applied using MITHRIL.[44] Full-matrix, least-squares anisotropic
refinement on F2 was applied to all non-hydrogen atoms to give
R ϭ 0.067, wR ϭ 0.059 for 1416 independently observed reflections
[F2o Ͼ 1σ(F2o), 2θmax ϭ 52.0°] and 226 variables. Goodness of fit
1.11, max. shift/error in final cycle 0.01, max./min. peak in final
Details of Chromatographic Separations of Enantiomers: The ee as-
says on compounds 6a؊l and 12a؊b were carried out by GC with
a Varian 3380 machine, using helium carrier gas and the columns
and conditions given in Table 4. The injector and detector port
temperatures were 150 and 200 °C respectively.
3
˚
difference map 0.33, Ϫ0.29 e·A . Crystallographic data (excluding
structure factors) for (Sa)-2b have been deposited with the Cam-
bridge Crystallographic Data Centre as supplementary publication
no. CCDC-152542. Copies of the data can be obtained free
of charge on application to CCDC, 12 Union Road, Cambridge
Base-Promoted Cyclisations of 12a؊c Affording (13a؊b, 14, and
15): Diethyl ether solutions of (R)-12a؊c (0.26 mmol, 0.26 ) were
stirred overnight with 6  NaOH (1.0 mL) at ambient temperature.
Normal extractive workup followed by flash chromatography af-
forded 10؊11 in 80Ϫ85% yield. Treatment of (R)-12b at 0 °C (12 h)
afforded (R)-13b (87%), together with (3R,5S)-14 (5%).
CB2 1EZ, UK [Fax: (internat.)
deposit@ccdc.cam.ac.uk].
ϩ 44-1223/336-033; E-mail:
Acknowledgments
4-Ethylcyclopent-2-enone (13a): Yield 80% (18% ee).
Ϫ δH
We thank the EPSRC for support of this project through grants
GR/M75341, GR/M84909 and GR/N37339 and for access to their
Mass Spectrometry Service (University of Swansea). S. W. is grate-
ful to the EU for support through the COST (working groups D12/
0009/98 and D12/0022/99). We thank Dr. A. J. Blake for crystallo-
graphic advice.
(400 MHz, CDCl3) ϭ 0.98 (t, J ϭ 7.4 Hz, 3 H, CH2Me), 1.46 (ddq,
1 H, J ϭ 13.5, 5.2 Hz, 7.4, CH2αMe), 1.61 (ddq, 1 H, J ϭ 13.5, 6.1
Hz, 7.5, CH2βMe), 1.97 (dd, J ϭ 18.8, 2.1 Hz, 1 H, COCH2α), 1.97
(dd, J ϭ 18.8, 6.3 Hz, 1 H, COCH2β), 2.88 (m, 1 H, CHEt), 6.16
(dd, J ϭ 5.7, 2.1 Hz, 1 H, ϭCH), 7.63 (dd, J ϭ 5.7, 2.5 Hz, 1 H, ϭ
CH). ϪMS (CI); m/z (%): 100 (49) [Mϩ], 82 (100) {found (HRMS,
CI) for [Mϩ] 110.0732, C7H10O requires 110.0732}. Ϫ These values
are consistent with 12a generated by a different route.[41]
[1]
[1a]
Reviews:
N. Krause, Angew. Chem. Int. Ed. 1998, 37,
[1b]
283Ϫ285. Ϫ
36, 187Ϫ204. Ϫ
346Ϫ353.
For recent examples see: M. Dieguez, S. Deerenberg, O. Pam-
ies, C. Claver, P. W. N. M. van Leeuwen, P. Kamer, Tetrahedron:
Asymmetry 2000, 11, 3161Ϫ3166 and references in there.
A. Alexakis, C. Benha¨ım, X. Fournioux, A. van der Heuvel, J.-
M. Leveque, S. March, S. Rosset, Synlett 1999, 1811Ϫ1813.
S. M. W. Bennett, S. M. Brown, J. P. Muxworthy, S. Woodward,
Tetrahedron Lett. 1999, 40, 1767Ϫ1770.
S. M. W. Bennett, S. M. Brown, G. Conole, M. R. Dennis,
P. K. Fraser, S. Radojevic, M. McPartlin, C. M. Topping, S.
Woodward, J. Chem. Soc., Perkin Trans. 1 1999, 3127Ϫ3132.
S. M. W. Bennett, S. M. Brown, A. Cunningham, M. R.
Dennis, J. P. Muxworthy, M. A. Oakley, S. Woodward, Tetra-
hedron 2000, 56, 2847Ϫ2855.
C. Börner, W. A. König, S. Woodward, Tetrahedron Lett. 2001,
42, 327Ϫ329.
O. Pamies, G. Net, A. Ruiz, C. Claver, S. Woodward, Tetrahed-
ron Asym. 2000, 11, 871Ϫ877.
N. Krause, Angew. Chem. Int. Ed. Engl. 1997,
(؉)-(R)-5-Ethylcyclohex-2-enone (13b): Yield 87% (70% ee). Ϫ δH
(400 MHz, CDCl3) ϭ 0.93 (t, J ϭ 7.4 Hz, 3 H, CH2Me), 1.38Ϫ1.45
(m, 2 H, CH2Me), 1.95Ϫ2.17 (3 H, CHCH2CHϭ), 2.40Ϫ2.48 (m,
1 H, COCH2α), 2.50Ϫ2.56 (m, 1 H, COCH2β), 6.01 (1 H, apparent
dt, J ϭ 10.0, 1.5 Hz, COCHϭ), 6.97 (ddd, 1 H, J ϭ 2.1, 5.5 Hz,
10.0). Ϫ δC (CDCl3, 67.8 MHz) ϭ 11.0 (CH2Me), 28.5 (CH2 ring
or CH2Me), 31.9 (CH2 ring or CH2Me), 36.8 (CHEt), 44.1 (CH2
ring or CH2Me), 129.7 (ϭCH), 150.0 (ϭCH), 199.2 (CO). Ϫ MS
(CI); m/z (%): 124 (53) [Mϩ], 96 (67), 82 (100) {found (HRMS, CI)
for [Mϩ] 124.0889, C8H12O requires 124.0888}. Ϫ The chiroptical
properties are as described in the literature.[42]
[1c]
B. L. Feringa, Acc. Chem. Res. 2000, 33,
[2]
´
`
[3]
[4]
[5]
ˆ
[6]
(؉)-(S)-4-Ethyl-2-methylcyclopent-1-enecarbaldehyde
(S)-(15): Yield 85% (69% ee); [α]D25ϭ ϩ5 (c ϭ 0.435, CHCl3). Ϫ
δH (400 MHz, CDCl3) ϭ 0.89 (t, J ϭ 7.3 Hz, 3 H, CH2Me), 1.39
(2 H, apparent quint, J ϭ 7.3, CH2Me), 2.11 (s, broadened by long-
range coupling, 3 H, ϭCMe), 2.14Ϫ2.26 (m, 2 H, ring CH2),
2.63Ϫ2.76 (m, 2 H, ring CH2), 9.96 (s, 1 H, CHO). Ϫ δC (CDCl3,
67.8 MHz) ϭ 12.3 (CH2Me), 14.3 (ϭCMe), 28.9, 35.9 (2 ϫ CH2),
[7]
[8]
[9]
S. Woodward, Chem. Soc. Rev. 2000, 29, 393Ϫ401; E. Naka-
mura, S. Mori, Angew. Chem. Int. Ed. Engl. 2000, 39,
3750Ϫ3771
37.4 (CHEt), 46.8 (CH2), 188.2 (CHO). Ϫ ν (thin film) [cmϪ1] ϭ
˜
2959m, 2923m, 2878w, 2853w (4 ϫ CϪH), 1666s (CϭO), 1383m,
1218m. Ϫ MS (FAB); m/z (%): 138 (63) [Mϩ], 109 (92), 68 (100)
{found (HRMS, FAB) for [Mϩ] 138.1046, C9H14O requires
138.1045}.
[10]
[11]
M. R. Dennis, S. Woodward, J. Chem. Soc., Perkin Trans. 1
1998, 1081Ϫ1085.
P. J. Cox, W. Wang, V. Snieckus, Tetrahedron Lett. 1992, 17,
2253Ϫ2256.
Eur. J. Org. Chem. 2001, 2435Ϫ2446
2445

C. Börner, M. R. Dennis, E. Sinn, S. Woodward
FULL PAPER
[12]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
G. J. Kubas, Inorg. Synth. 1990, 28, 68Ϫ70.
C. Girard, H. B. Kagan, Angew. Chem. Int. Ed. 1998, 37,
2922Ϫ2959.
L.-Z. Gonj, L. Pu, Tetrahedron Lett. 2000, 41, 2327Ϫ2331.
Review on Lewis acid binding of carbonyl groups: S. Shambay-
T. Tsuda, T. Yoshida, T. Kawamoto, T. Saegusa, J. Org. Chem.
1987, 52, 1624Ϫ1627.
C. Canevet, T. Röder, O. Vostrowsky, H. J. Bestman, Chem.
Ber. 1980, 113, 1115Ϫ1120.
R. Heilmann, G. de Gaudemaris, P. Arnaud, G. Scheutbrandt,
Bull. Chem. Soc. Chim. Fr. 1957, 112Ϫ118.
Ph3PϭCH(COMe) is commercially available from the Lancas-
ter Chemical Company.
(MeO)2P(O)CH2COCH2OMe: H. Feisthauer, R. Neidlein,
Helv. Chim. Acta 1996, 79, 895Ϫ912.
M. Benechie, B. Delpech, Q. Khuong-Huu, F. Khuong-Huu,
Tetrahedron 1992, 48, 1895Ϫ1910.
J. Uenishi, M. Motoyama, K. Takahashi, Tetrahedron Asym.
1994, 5, 101Ϫ110.
G. Traverso, D. Pirillo, G. Rescia, Farmaco Ed. Sci. 1979, 34,
229Ϫ233; Chem. Abstr. 1979, 90, 186301k.
G. D. Cuny, S. L. Buchwald, J. Am. Chem. Soc. 1993, 115,
2066Ϫ2068.
E. Lee-Ruff, P. G. Khazanie, Can. J. Chem. 1978, 56, 803Ϫ807.
G. Lardelli, V. Lamberti, W. T. Weller, A. P. de Jonge, Recl.
Trav. Chim. Pays-Bas 1967, 86, 481Ϫ503.
A. Alexakis, J. Vastra, P. Mangeney, Tetrahedron Lett. 1997,
38, 7745Ϫ7748.
[13]
[14]
[15]
ati, S. L. Schreiber, in: Comprehensive Organic Synthesis (Eds.:
B. M. Trost, I. Fleming), Pergamon, Oxford, 1991, vol 1, chap-
ter 1.10, pp. 283Ϫ353.
[16]
N. Bresciani Pahor, L. Randaccio, Inorg. Chim. Acta 1980, 45,
L11ϪL12; D. M. L. Goodgame, S. P. W. Hill, A. M. Smith, D.
J. Williams, J. Chem. Soc., Dalton Trans. 1994, 859Ϫ870.
[17]
We quoted the reverse of this selectivity in error in an earlier
paper (ref.[6]). Although the error was typographical in nature
[the (Ra), (Sa) descriptors of two ligands were inadvertently in-
terchanged], all stereochemical assignments in this paper were
subjected to meticulous rechecking.
[18]
R. Arnecke, U. Groth, T. Köhler, Liebigs Ann. Chem. 1994,
9, 891Ϫ894.
A. F. Simpson, C. D. Bodkin, C. P. Butts, M. A. Armitage, T.
[36]
[37]
[19]
Gallagher, J. Chem. Soc., Perkin Trans. 1 2000, 3047Ϫ3054.
R. Imbos, M. H. G. Brilman, M. Pineschi, B. L. Feringa, Org.
[38]
[39]
[20]
Lett. 1999, 1, 623Ϫ625.
http://www.macherey-nagel.ch/
W. A. König, D. Icheln, T. Runge, I. Pforr, A. Krebs, J. High
[21]
X. Hu, H. Chen, X. Zhang, Angew. Chem. Int. Ed. 1999, 38,
3518Ϫ3521.
[22]
[40]
[41]
H. Ahibrecht, A. von Daacke, Synthesis 1987, 24Ϫ28.
K. B. Kingsbury, J. D. Carter, A. Wilde, H. Park, F. Takusa-
gawa, L. McElwee-White, J. Am. Chem. Soc. 1993, 115,
10056Ϫ10065.
G. H. Posner, L. L. Frye, Isr. J. Chem. 1984, 24, 88Ϫ90.
J. R. Backhouse, H. M. Lowe, E. Sinn, S. Suzuki, S. Wood-
ward, J. Chem. Soc., Dalton Trans. 1995, 1489Ϫ1495.
C. J. Gilmore, J. Appl. Crystallogr. 1984, 17, 42Ϫ46.
Received January 11, 2001
Res. Chromatogr. 1990, 13, 702Ϫ707.
W. A. König, B. Gehrcke, D. Icheln, P. Evers, J. Doennecke, W.
[23]
Wang, J. High Res. Chromatogr. 1992, 15, 367Ϫ372.
Y.-Z. Huang, C. Chen, Y. Shen, Synth. Commun. 1989, 19,
501Ϫ510.
T. Kawasaki, T. Ichige, T. Kitazume, J. Org. Chem. 1998, 63,
7525Ϫ7528.
A. Yamagisawa, S. Habaue, K. Yasue, H. Yamamoto, J. Am.
[24]
[42]
[43]
[25]
[44]
[26]
Chem. Soc. 1994, 116, 6130Ϫ6141.
[O01011]
2446
Eur. J. Org. Chem. 2001, 2435Ϫ2446