Heteroatom-guided torquoselective olefination of α-oxy and α-amino ketones via ynolates

Article abstract of DOI:10.1002/chem.200500574

Ynolates were found to react with α-alkoxy-, α-siloxy-, and α -aryloxyketones at room temperature to afford tetrasubstituted olefins with high Z selectivity. Since the geometrical selectivity was determined in the ring opening of the β-lactone enolate intermediates, the torquoselectivity was controlled by the ethereal oxygen atoms. From experimental and theoretical studies, the high Z selectivity is in duced by orbital and steric interactions rather than by chelation. In a similar manner, α-dialkylamino ketones provided olefins with excellent Z selectivity. These products can be easily converted into multisubstituted butenolides and γ-butyrolactams in good yield.

Full text of DOI:10.1002/chem.200500574

DOI: 10.1002/chem.200500574
Heteroatom-Guided Torquoselective Olefination of a-Oxy and a-Amino
Ketones via Ynolates
Mitsuru Shindo,*[a] Takashi Yoshikawa,[b] Yasuaki Itou,[c] Seiji Mori,[c]
Takeshi Nishii,[d] and Kozo Shishido[b]
Abstract: Ynolates were found to react
with a-alkoxy-, a-siloxy-, and a-arylox-
yketones at room temperature to
afford tetrasubstituted olefins with high
Z selectivity. Since the geometrical se-
lectivity was determined in the ring
opening of the b-lactone enolate inter-
mediates, the torquoselectivity was
controlled by the ethereal oxygen
atoms. From experimental and theoret-
ical studies, the high Z selectivity is in-
duced by orbital and steric interactions
rather than by chelation. In a similar
manner, a-dialkylamino ketones pro-
vided olefins with excellent Z selectivi-
ty. These products can be easily con-
verted into multisubstituted buteno-
lides and g-butyrolactams in good
yield.
Keywords: density functional calcu-
lations
· ketones · olefination ·
torquoselectivity · ynolates
Introduction
Although the phosphorus ylide methods are generally ef-
fective in providing di- and trisubstituted olefins from alde-
hydes and some ketones, the low reactivity and stereoselec-
tivity of these reagents make them less desirable in the olefi-
nation of ketones to furnish tetrasubstituted olefins.[7] The
other olefination processes, even if they are reactive enough
towards hindered ketones, have not been successfully used
in the stereoselective synthesis of tetrasubstituted olefins,
which serve as important synthetic intermediates for synthe-
ses of complex molecules[8] and useful units in medicinal
chemistry[9] and material science.[10] Considering how diffi-
cult it is to achieve high efficiency in the olefination of ke-
tones with conventional reagents, development of a novel
The synthesis of alkenes by the olefination of carbonyl com-
pounds has received considerable attention due to the effi-
ciency and convenience of this methodology.[1] In addition
to methods based on phosphorus ylides, such as the Wittig[2]
and the Horner–Wadsworth–Emmons reactions,[3] other re-
lated processes involving sulfur[4] and silicon[5] ylides have
been used in synthetic organic chemistry. Furthermore, gem-
bimetallic reagents (or metal carbenoids) have been report-
ed for the olefination of carbonyl compounds.[6]
reaction with
a new mechanism would be extremely
[a] Prof. Dr. M. Shindo
useful.[11]
Institute for Materials Chemistry and Engineering, Kyushu University
6-1, Kasugakoen, Kasuga, Fukuoka 816-8580 (Japan)
Fax : (+81)92-583-7875
We have developed a novel methodology for the genera-
tion of ynolate anions 3 by cleavage of ester dianions 2 de-
rived from a,a-dibromo esters 1[12] (Scheme 1).[13] Since yno-
late anions are ketene anion equivalents, they are expected
to act as multifunctional reactive species.[14]
E-mail: shindo@cm.kyushu-u.ac.jp
[b] T. Yoshikawa, Prof. Dr. K. Shishido
Institute of Health Biosciences
The University of Tokushima Graduate School
Sho-machi 1, Tokushima 770-8505 (Japan)
[c] Y. Itou, Prof. Dr. S. Mori
Faculty of Science, Ibaraki University
Mito, Ibaraki 310-8512 (Japan)
[d] T. Nishii
Faculty of Pharmaceutical Sciences, Tokushima Bunri University
180 Yamashiro-cho, Tokushima 770-8514 (Japan)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author: Experimental
procedures, spectral data of the starting ketones and the products,
and Cartesian coordinates of the optimized TSs (PDF).
Scheme 1. Synthesis of ynolates.
524
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 524 – 536

FULL PAPER
We have reported an olefination of aldehydes and ketones
using the ynolate anions 3,[15] via ring opening of the b-lac-
tone enolates (oxetenoxide ions) derived from the cycload-
dition of ketones and aldehydes with ynolates, with good to
moderate E/Z selectivity (Scheme 2).[16] When aryl alkyl ke-
Scheme 3. Initial working hypothesis of stereoselective olefination.
stituted alkenes, and mechanistic investigations supported
by theoretical calculations, which led to an unexpected con-
clusion.
Scheme 2. Olefination of ketones via ynolates.
Results
tones were used as substrates, the E/Z selectivity was in the
range of 4–8. Furthermore, we found a stereoelectronic
effect on the E/Z selectivity, whereby ketones having an
electron-donating group at the para position of the aryl sub-
stituent exhibited much better E selectivity.[17] The E/Z se-
lectivity is determined in the ring opening of the b-lactone
enolate intermediates, and stereocontrol would be governed
by torquoselectivity, in which the electron-donating substitu-
ent preferentially rotates outward, and the electron-with-
drawing substituent inward.[18] The opposite explanation in-
volving geminal bond participations has also been report-
ed.[19] In our case, successful olefinations were limited to
electron-rich aryl alkyl ketones. Recently, we achieved high
Z selectivity in the olefination of acylsilanes to give vinylsi-
lanes (silyl alkenes).[20] In this case, the silyl group efficiently
Olefination of a-siloxy- and a-alkoxyketones: Based on this
concept, we selected a-alkoxyketones as the substrates, ex-
pecting chelation of the ethereal oxygen atom to the lithium
cation. Olefinations of a-hydroxy or a-siloxyketones giving
tetrasubstituted olefins by the Horner–Emmons reaction[22]
or via ynamines[23] have been reported, for some of which
good selectivity was achieved, but they are less efficient or
lack generality.[24] We first treated a-methoxypropiophenone
(6a) with lithium ynolate 3a at room temperature for
30 min to obtain the desired olefin 7a in good yield with ex-
cellent Z selectivity after methyl esterification (Table 1,
entry 1). The a-methoxymethoxy and a-(p-methoxybenzy-
loxy)propiophenones (6b, 6c) also gave the desired olefins
(7b, 7c) with high Z-selectivities (Table 1, entries 2 and 3),
as did the a-siloxypropiophenones (6d, 6e, 6 f), although
one would expect the siloxy oxygen atom to be less Lewis
basic (Table 1, entries 4–7).[25] Interestingly, even if the lithi-
ꢀ
acts as an “electron-accepting” group due to its Si C s* or-
bitals.[21] In spite of these successful results, the development
of general methods, in which the stereochemistry is predicta-
ble, for the efficient olefination of ketones still remains a
challenge.
As described above, ynolates are potential olefination re-
agents toward ketones. If this method were to show general-
ly high E/Z selectivity, it would become an alternative to the
ylide and carbenoid methods. To achieve high efficiency, we
therefore looked for other efficient, strong stereocontrolling
factors for olefination. In the olefination of aldehydes, be-
cause it is easy to discriminate between the Hand R groups
(RC(=O)H), many successful results have been reported.
Ketones RC(=O)R’, however, have similar substituents R
and R’; thus, olefination reagents cannot easily distinguish
between them. We anticipated that highly general and
stereo-predictable olefination could be achieved by using a
ketone bearing a strongly directing group. The directing
group should be a versatile functional group, widely used
and readily available, which can be easily converted to other
functional groups. Based on mechanistic considerations, and
the fact that the E/Z selectivity is controlled in the thermal
ring opening of the b-lactone enolate intermediate, we envi-
sioned an intermediate having a coordination site which
would chelate to the lithium cation, as in 5, and thereby de-
termine the direction of ring opening (Scheme 3). Here we
describe a highly Z-selective olefination of a-oxy and a-
amino-substituted ketones via ynolates to provide tetrasub-
Table 1. Olefination of a-alkoxy and a-siloxy propiophenones via lithium
ynolates.[a]
Entry
6
X
7
Ratio
Yield
[%]
[Z:E][b]
1
2
3
4
5
6
7
8
9
6a
6b
6c
6d
6e
6e
6 f
6g
6h
6i
OCH3
7a
7b
7c
7d
7e
7e
7 f
7g
7h
7i
>99:1
98:2
>99:1
99:1
98:2
>99:1
98:2
98:2
98:2
–
81
50
76
71
85
40[d]
85
68
70
0
OCH2OCH3
OCH2C6H4OCH3-p
OTES
OTBS
OTBS[c]
OTIPS
OPh
OC6H4OCH3-p
OC6H4NO2-p
CH3
10
11[e]
12[e]
6j
6k
7j
7k
75:25[f]
86:14[f]
86
89
H
[a] TES=triethylsilyl, TBS=tert-butyldimethylsilyl, TIPS=triisopropyl-
silyl. [b] The stereochemistry was determined by NOE experiments. See
Supporting Information. [c] [12]crown-4 (20 equiv) was added to the lithi-
um ynolate solution. [d] The reaction time was 48 h. [e] From ref. [15b].
[f] E:Z ratio.
Chem. Eur. J. 2006, 12, 524 – 536
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
525

M. Shindo et al.
Table 2. Z-Selective olefination of a-siloxy and a-alkoxy acyclic ketones via ynolates.[a]
um cation was trapped by
crown ether, the high selectivity
still remained, albeit with mod-
erate yield (Table 1, entry 6).
Furthermore, the a-phenoxy
Entry
1
3 (R)
Ketone
Major product
Yield [%]
Z/E
and
a-(p-methoxy)phenoxy
groups, the ethereal oxygen
atoms of which are much less
Lewis basic, afforded excellent
Z selectivity (Table 1, entries 8
and 9). a-(p-Nitrophenoxy)pro-
piophenone (6i) did not pro-
vide the desired product, and a
complex mixture was obtained.
By contrast, the reactions of
propiophenone (6k) and iso-
propyl phenyl ketone (6j) gave
olefins with poorer selectivity
(Table 1, entries 11 and 12).
The dramatic difference in se-
lectivity can be explained by an
a-oxygen effect, although it is
uncertain if this is the chelation
effect shown in Scheme 3.
Bu
59
66
>99:1
2
3
Me
Me
>99:1
>99
98
95:5
4
5
Me
Me
94:6
92:8
60
To examine the generality of
this process, various kinds of a-
siloxy- and a-alkoxyketones (8)
were subjected to the reaction.
As shown in Table 2, most of
the acyclic a-oxyalkyl alkyl and
a-oxyalkyl aryl ketones afford-
ed olefins with high Z selectivi-
ty (Table 2, entries 1–12). Even
sterically hindered ketones
bearing a-quaternary carbon
centers can afford the desired
olefins in good yield with high
Z selectivity (Table 2, entries 4,
8, and 10). The optically pure
chiral ketones 8i and 8j, pre-
pared from d-mannitol, furnish-
ed the Z-olefins without loss of
optical purity (Table 2, en-
tries 11 and 12). It is notewor-
thy that the reaction of tert-
butyl tert-butyldimethylsiloxy-
methyl ketone (8h) afforded
only the Z-olefin in which the
tert-butyl group rotated out-
ward in the ring opening of the
b-lactone enolate (Table 2,
entry 10), while olefination
with pinacolone showed the op-
posite selectivity (Scheme 4),
that is, the tert-butyl group
preferentially rotated inward.
This remarkable contrast sug-
6
iPr
76
98:2
7
8
Ph
75
91
93
94
>99:1
>99:1
>99:1
>99:1
>99:1
Me
Me
Me
Me
9
10
11
26
(>99% ee)
66
(>99% ee)
(>99% ee)
12
Me
>99:1
(>99% ee)
[a] TBS=tert-butyldimethylsilyl, TBDPS=tert-butyldiphenylsilyl.
526
www.chemeurj.org
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 524 – 536

Torquoselective Olefination of a-Oxy and a-Amino Ketones
FULL PAPER
Table 3. Z-Selective olefination of a-siloxy cyclic ketones via ynolates.
Entry 3 (R) Ketone
Major product Yield [%] Z:E
1
2
Me
Me
60
40
83:17
Scheme 4. Olefination of pinacolone.
50:50
94:6
gested that the presence of an a-oxy group is critical for
stereocontrol.
Next, a-oxy cyclic ketones were examined (Table 3). Al-
though 2-(tert-butyldimethylsiloxy)cyclopentanone did not
give the desired olefin but rather a complex mixture, 2-silox-
ycyclohexanone 8k afforded the desired Z olefin with mod-
erate selectivity (Table 3, entry 1). To investigate the confor-
mational effect of the a-siloxy group, cis- and trans-4-
methyl-2-(tert-butyldimethylsiloxy)cyclohexanone (8l, 8m)
were subjected to olefination. As shown in entries 2 and 3 in
Table 3, syn form 8l gave no selectivity, while anti form 8m
showed high Z selectivity, albeit in modest yield. This re-
3
4
5
Me
Me
Me
33
83 >99:1
markable contrast suggested that the conformation of the
siloxy group, which affects the O C C=O dihedral angle of
the substrate, is related to the torquoselectivity. The chemi-
cal shifts and coupling constants in the HNMR spectra in-
a
ꢀ
ꢀ
90
97:3
1
dicated equatorial orientation of the siloxy group in the syn
form 8l (Scheme 5) and axial orientation in the anti form
8m (Scheme 6).
These results suggest that the conformationally fixed
equatorial position of the a-siloxy group is not suitable for
stereocontrol in ring opening of the b-lactone enolate inter-
mediates, whereas the axial position is. For mechanistic con-
siderations, we tried to determine the stereochemistry of the
corresponding b-lactone intermediate to see whether the
ynolate attacked the a-siloxycyclohexanones axially or equa-
torially. Ynolate 3a reacted with cis-2-siloxy-4-methylcyclo-
hexanone (8l) at ꢀ788C, followed by treatment with iodo-
methane, to give b-lactone 28 with 20:1 diastereoselectivity.
The resulting compound was deprotected with TBAF to
6
7
Me
Me
70
0
86:14
none
–
afford g-lactone 29 by translactonization in good yield.[26]
The X-ray crystal structure analysis of 29 disclosed the
trans-fused bicyclic structure
(Figure 1),
which
indicates
equatorial attack of ynolate 3a
on 8l (Scheme 5).
Similarly,
trans-2-siloxy-4-
methylcyclohexanone (8m) was
treated with ynolate 3a at
ꢀ788C, followed by methyla-
tion, to afford a 3:1 diastereo-
meric mixture of b-lactone 30.
Only the major isomer could be
desilylated with one equivalent
of TBAF to afford hydroxy b-
lactone 31 as a single isomer,
which could not be converted
into the g-lactone, due to the
axial orientation of the hydroxy
group. Thus, this hydroxyl
Scheme 5. Equatorial attack of the ynolate on 8l. TBS=tert-butyldimethylsilyl, TBAF=tetrabutylammonium
fluoride.
Chem. Eur. J. 2006, 12, 524 – 536
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
527

M. Shindo et al.
Scheme 6. Equatorial attack on 8m preferred to axial attack. TPAP=tetrapropylammonium perruthenate; NMO=N-methylmorpholine-N-oxide.
This two-step conversion to the butenolides can be carried
out in one pot. Thus, after completion of the olefination,
3% HCl/EtOH was added to the resulting reaction mixture,
which was then concentrated in vacuo (conditions A), or
heated under reflux (conditions B) to produce the buteno-
lides 33 in good yield (Table 5). Therefore, this is an effi-
cient method for the construction of synthetically important
butenolides starting from a-siloxyketones.
Olefination of a-amino ketones: Following the above con-
cept, we examined the olefination of a-aminoketones-
(Table 6). The reaction of ynolate 3a with the N-methylben-
zylaminoketone 46 proceeded smoothly to give a g-amino
unsaturated carboxylate. Since isolation of the resulting
amino acid could be potentially difficult, it was treated in
situ with thionyl chloride (Method A) to provide N-methyl-
1,5-dihydropyrrol-2-one (g-lactam 54) in 49% yield (Table 6,
entry 1). The intermediate acyl chloride probably was imme-
diately trapped by the amino group intramolecularly to gen-
erate the lactam by releasing benzyl chloride. Attempted
isolation of the amino esters by addition of MeOSOCl, pre-
pared by mixing thionyl chloride and methanol (Method B),
resulted in formation of the same g-lactam 54 quantitatively
(Table 6, entry 2). No (E)-g-amino esters could be detected;
therefore, we presumed the Z selectivity to be extremely
high. In the cyclization to the lactams, the leaving groups on
nitrogen should be activated by alkyl groups such as benzyl
and allyl, since ketone 47 having a dimethylamino group did
not furnish the lactam (Table 6, entry 3). In the case of N,N-
dibenzylamine (Table 6, entries 4 and 5), method A provid-
ed the lactam 55 in better yield (Table 6, entry 4). The ap-
propriate cyclization conditions seem to be dependent on
the substrate. If the N-R1 group were easily removable, this
process would be synthetically useful. We therefore tried to
use carbamoyl (Boc and Cbz) and p-methoxyphenyl groups
as R1, but instead of the desired lactams, only complex mix-
tures were generated (Table 6, entries 6–8). Reaction of the
unprotected compound (R1 =H) was likewise unsuccessful
(Table 6, entry 9). Finally, the N,N-diallylamine 53 provided
the desired g-lactam 60 in good yield (Table 6, entry 10),
Figure 1. X-ray crystal structure of 29. ORTEP plot with 50% probability
thermal ellipsoids.
group was oxidized with TPAP, and the resulting ketone 32
was reduced by NaBH4 to provide g-lactone 29 along with
the b-lactone 31 (Scheme 6). Judging from the chemical
1
shifts and the coupling constants in the HNMR spectrum
of 8m, the siloxy group in 8m is in the axial position. From
this result, it was found that the equatorial attack of the
ynolate 3a on the trans-2-siloxy-4-methylcyclohexanone 8m
was predominant.
Seven-, eight-, and twelve-membered cyclic ketones bear-
ing a-tert-butyldimethylsiloxy substituents afforded the de-
sired olefins with good to excellent Z selectivity (Table 3,
entries 4–6). The extremely hindered and conformationally
fixed ketone did not react with the ynolate (Table 3,
entry 7).
From these results, it was found that acyclic siloxyketones
generally provide the desired olefins in good yields with
high Z selectivity, and in the olefination of cyclic siloxyke-
tones, the conformational flexibility, or certain conforma-
tions, might be required to achieve high selectivity and good
yields.
Synthesis ofbutenolides : The (Z)-g-siloxy-a,b-unsaturated
esters 9 were treated with acid to provide the butenolides 33
(Table 4), which are frequently found in natural products.[27]
Except for the volatile butenolides 34–36 (Table 4, en-
tries 1–3), the desired butenolides were obtained in good to
excellent yields (Table 4, entries 4–9).
528
www.chemeurj.org
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 524 – 536

Torquoselective Olefination of a-Oxy and a-Amino Ketones
FULL PAPER
Table 4. Conversion to butenolides.
Table 5. One-pot synthesis of butenolides.
Entry
1
R
Ketone
Conditions[a] Butenolide Yield [%]
Entry
1
g-Siloxyacrylate
Butenolide
Yield [%]
24
Bu
B
B
62
14
2
3
Me
Me
2
68
A
43
3
4
5
65
>99
>99
4
5
6
Me
Me
Me
A
A
B
96
75
98
6
7
84
83
7
8
Me
Me
A
A
86
80
8
9
>99
9
Me
A
68
65
[a] Conditions A: After addition of 3% HCl/EtOH to the carboxylate,
the reaction mixture was concentrated in vacuo. Conditions B: After ad-
dition of 3% HCl/EtOH to the carboxylate, the reaction mixture was re-
fluxed for 3 h.
and the N-allyl group was efficiently removed by treatment
with Pd/C in the presence of BF3·OEt2 (Scheme 7).[28]
Based on these results, olefination reactions using with
several combinations of ynolates and a-aminoketones were
examined (Table 7). The less substituted a-aminoacetophe-
none 61 and the sterically hindered a-aminopinacolone 62
gave g-lactams 64 and 65, respectively (Table 7, entries 1
and 2). The phenyl-, isopropyl-, and butyl-substituted yno-
lates also furnished the desired lactams in good yield
(Table 7, entries 3–5). The dialkyl a-amino ketone 63 gave
Chem. Eur. J. 2006, 12, 524 – 536
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
529

M. Shindo et al.
Table 6. Olefination of a-aminopropiophenone derivatives providing
g-butyrolactams.
Scheme 7. Removal of the N-allyl moiety.
Entry
Amino ketone
R2
Condi-
tions[a]
g-Lactam
R1
Yield [%]
69 under conditions B in good yield. Moreover, the E iso-
mers could not be detected in any of these cases.
1
2
3
4
5
6
7
8
9
46 Me
46 Me
47 Me
48 benzyl
48 benzyl
49 Boc
50 Cbz
51 p-methoxyphenyl benzyl
52 Hbenzyl
benzyl
benzyl
Me
benzyl
benzyl
benzyl
benzyl
A
B
B
A
B
B
B
B
54 49
54 >99
54 complex mixture
55 68
55 36
56 complex mixture
57 complex mixture
58 complex mixture
59 complex mixture
60 87
Chelation or not?: The E/Z stereochemistry of the olefina-
tion products is determined in the thermally conrotatory
ring opening of the b-lactone enolate. Since this ring open-
ing would be an irreversible, exothermic reaction, the rela-
tive thermodynamic stability of the products might contrib-
ute very little to the selectivity, compared with the relative
energy of the transition states of the ring-opening process,
which must therefore be responsible for the E/Z selectivity.
Originally, we proposed chelation control as a working hy-
pothesis, in which the ethereal oxygen atom chelates the
lithium cation to produce the Z olefin (Scheme 3). However,
the following experimental observations are not consistent
with the chelation mechanism:
B
10 53 allyl
11 53 allyl
allyl
allyl
A
B
60 53
[a] Method A: A THF solution of SOCl2 (3 equiv) was added at ꢀ108C
and the solution was stirred at room temperature for 12–17 h. Method B:
A THF solution of MeOSOCl, prepared from SOCl2 and MeOH, was
added at ꢀ108C and the solution was stirred at room temperature for
12–17 h.
1) The sterically hindered
siloxy and phenoxyl groups,
Table 7. Olefination of a-aminoketones affording g-butyrolactams.
Entry
Ynolate (R)
Aminoketones
Conditions[a]
Product
Yield [%]
52
the ethereal oxygen atoms
of which are supposed to be
much poorer Lewis bases,[25]
are also effective for the in-
duction of high Z selectivity
(Table 1, entries 5 and 7;
Table 2, entry 2).
1
Me
A
2) Even when olefination of
6e was carried out in the
presence of an excess of
[12]crown-4, the selectivity
still remained high (Z:E
>99:1) (Table 1, entry 6).
This suggests that the metal
cation is not always re-
quired for stereoselection,
and indeed in this case, che-
lation is not a stereocontrol-
ling factor.
2
Me
A
53
3
4
5
Ph
iPr
Bu
A
A
A
90
60
79
3) In the olefination of 2-silox-
ycyclohexanones, an axially
oriented siloxy group in-
duced high
Z selectivity
(Table 3, entry 3), while an
equatorial group afforded
6
Me
B
73
no
selectivity
(Table 3,
entry 2). Ynolates predomi-
nantly undergo equatorial
attack in both cases to fur-
nish b-lactone enolates, as
[a] Conditions A: MeOSOCl (ca. 3 equiv) in methanol was added a room temperature. Conditions B: SOCl2
(ca. 3 equiv) was added at ꢀ108C.
530
www.chemeurj.org
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 524 – 536

Torquoselective Olefination of a-Oxy and a-Amino Ketones
FULL PAPER
Computational reaction pathway: To assess the possibility of
chelation by the 4-alkoxymethyl group in the ring-opening
reaction, we optimized the transition-state structures in the
reaction of 70 (Scheme 8). We found four TSs that lead to Z
Figure 2. Hypothetical chelation of the lithium cation by equatorial (left)
and axial (right) ethereal oxygen atom.
shown in Schemes 5 and 6. If the torquoselectivity is
controlled by chelation, the equatorial siloxy group
should have been more potent than the axial siloxy
group (Figure 2), although organolithium compounds
form complex aggregation networks in solution.[29] The
experimental results, however, would indicate the oppo-
site trend, as described above.
Scheme 8. Reaction for theoretical calculations.
alkenes (TSZ1–TSZ4) and three that lead to E alkenes
(TSE1–TSE3). The structures are shown in Figure 3, and the
relative energies are listed in Table 8. Transition states TSZ4
and TSE3 are chelated structures, wherein both the methox-
yl and the carboxylate groups are coordinated to the lithium
atom, and are much higher in electronic energy, enthalpy,
and Gibbs free energy than the other, nonchelated TSs.
Hence chelation pathways can be ruled out.[37] The higher
energies of TSZ4 and TSE3 are due to the deformations of
the substrate structure. This is consistent with the high selec-
tivity found experimentally even when a crown ether is
added or a lithium ynolate reacts with a bulky siloxy ketone.
Among the TSs investigated, TSZ1 and TSE1 were the
most stable giving Z and E alkenes, respectively; TSZ1 is
more stable than TSE1 by 11.0 kJmolꢀ1 in enthalpy and
13.8 kJmolꢀ1 in Gibbs energy. These results are in good
agreement with the high Z selectivity obtained experimen-
tally. In TSZ1, the methoxyl group at the inwardly rotated
4-methoxymethyl group is nearly perpendicular to the
Theoretical studies: The following questions still remain
regarding the stereoselectivity: 1) What is the reaction
mechanism of the ring opening of the b-lactone enolate?
2) Is the proposed chelation control a genuinely important
factor? 3) If not, then what is the role of the a-oxy group in
the stereoselectivity? To help us answer these questions, we
employed B3LYP hybrid DFT calculations. Natural bond or-
bital (NBO) theory,[30] which is particularly helpful in deter-
mining the stereoelectronic effects of conformations and or-
ganic reactions,[31] was used for analyses of the stereoselec-
tivity. The importance of hyperconjugative effects on the
stereoselectivities in ring opening of two b-lactone enolates,
namely, 4-tert-butyl-4-methyloxet-2-enoxide and 4-trimethyl-
silyl-4-methyloxet-2-enoxide, are already known on the basis
of NBO theory.[21]
Computational methods: We employed the B3LYP hybrid
functional[32] with the 6-31G(d) basis set[33] for a b-lactone
enolate intermediate, namely, lithium 4-methoxymethyl-3,4-
dimethyloxet-2-enoxide (70), in which the lithium atom is
solvated by two Me2O molecules; the transition states of the
ring opening process; and the products.[34,35] Normal coordi-
nate analyses were performed for stationary points. One
imaginary frequency was confirmed at each optimized TS
structure.
The origin of the stereoelectronic effects of the ring open-
ing of the b-lactone enolate derivative was examined with
the aid of NBO analysis.[36] The transition states of the ring
opening are reactantlike rather than productlike according
the optimal Lewis structure search. Second-order perturba-
tion analysis of bonding NBOs and antibonding NBOs was
carried out for these transition states. The second order in-
teraction energy is expressed as Equation (1), where s/s*
and F are the filled/vacant NBO and Fock matrices, respec-
tively, es and es are the NBO energies of the bonding/lone
pair and those of antibonding/Rydberg, respectively, and
NBOs are mutually orthogonal.
4
1
5
a
ꢀ
breaking C ··O bond, and the C H bond is antiperiplanar
to the breaking C4···O1 bond (O1-C4-C5-Ha dihedral angle
ꢀ168.28; labeling of the C, O, and Hatoms is shown in
6
5
ꢀ
Figure 3). Transition state TSZ2, in which the O C bond
and the breaking C4···O1 bond are mutually antiperiplanar,
is slightly higher in energy than TSZ1. To clarify the hyper-
conjugative interactions in the transition states which
govern the torquoselectivity, an NBO analysis was attempt-
ed. As in the transition states described in our previous
report,[21] several hyperconjugative interactions were ob-
Table 8. Relative activation energies (DDE°), activation enthalpies
(DDH°), and Gibbs free energies of activation (DDG°) at 298.15 K of
TSs relative to TSZ1 in kJmolꢀ1 at the B3LYP/6-31G(d) level. The
values in bold are the lowest energies of the TSE and TSZ.
DDE°
DDH°
DDG°
TSZ1
TSZ2
TSZ3
TSZ4
TSE1
TSE2
TSE3
0.0
4.0
0.0
4.5
0.0
3.0
10.9
20.2
13.8
7.4
14.8
16.3
10.7
12.1
31.7
14.3
16.3
11.0
11.6
32.2
F2ij
De
2
*
hsjFjs i
ð2Þ
ð1Þ
42.6
E
¼ ꢀ2
¼ ꢀ2
ss
*
es ꢀes
*
Chem. Eur. J. 2006, 12, 524 – 536
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
531

M. Shindo et al.
higher energy of TSZ2 is due to
its larger steric repulsion. The
larger dipole moment of TSZ2
(6.33 D) compared with that of
TSZ1 (5.26 D) suggests that the
former could be detected in
polar media.
Mechanistic considerations: Tor-
quoselectivity in the ring open-
ing of the b-lactone enolates
should be considered in eluci-
dating the stereoselectivity.
Houk et al. found that torquo-
selectivity in ring opening of cy-
clobutenes is subject to stereo-
electronic and steric effects. Ac-
cording to their theoretical and
experimental studies, the elec-
tron-donating substituent pref-
erentially rotates outward, and
the electron-withdrawing sub-
stituent inward.[18] In their early
articles,[39] the stereoelectronic
factors of the substituents
mainly involved unsaturated
groups with p orbitals and halo
substituents with lone pairs.
The p orbitals and the lone
pairs are electron-donating,
while the p* orbitals are elec-
tron-withdrawing.
Recently,
Murakami et al.[40] and Houk
et al.[41] indicated the significant
role of s* orbitals in the ring-
opening transition states of cy-
clobutenes (Figure 4). They
pointed out that, in the transi-
tion states of the ring opening
of cyclobutenes, trialkylsilyl and
fluoromethyl groups with low-
lying s* orbitals prefer inward
rotation, because the overlap
between the breaking s orbital
and the s* orbital of the sub-
stituents stabilizes the transition
Figure 3. Structures of the transition states of ring opening of 2Me2O-solvated lithium 4-methoxymethyl-4-
methyloxet-2-enoxide with bond lengths [Š].
states of the inward rotation.
Recent NBO studies by Mori
4
1
ꢀ
served. Although the interactions between s*(C O ) and
and Shindo also support the hypotheses of Houk and Mura-
kami on the basis of the ring opening of the b-lactone eno-
late, especially the importance of the interaction of the elec-
7
c
5
a
ꢀ
ꢀ
antiperiplanar s(C H and C H ) lead to high stabilization
energy in TSZ1, TSZ3, TSE1, and TSE2, no definitive dif-
ferences were observed among them. In TSZ2, the interac-
ꢀ
tron-rich oxygen atom with the C Y bond antiperiplanar to
4
1
5
6
ꢀ
ꢀ
tion of s(C O ) with antiperiplanar s*(C O ) of
ꢀ15.8 kJmolꢀ1 is one of the most important secondary NBO
interactions, and is larger than that of ꢀ3.6 kJmolꢀ1 in
TSZ1. The total steric exchange energy[38] of TSZ2 is larger
than that of TSZ1 by 16.7 kJmolꢀ1. This suggests that the
the breaking C···O bond. Inagaki et al. ascribed a different
ꢀ
explanation, namely, participation of the geminal C X
ꢀ
bond. However, such C X bond participation is not impor-
tant in ring opening of alkylsilyloxetenoxides and dialkylox-
ꢀ
etenoxides compared with the X Y bond participation, es-
532
www.chemeurj.org
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 524 – 536

Torquoselective Olefination of a-Oxy and a-Amino Ketones
FULL PAPER
ꢀ
lower lying s*(C O) orbitals, would have been thought to
afford better selectivity, but gave instead a complex mixture
that included a small amount of the desired products.
Conclusion
We have developed a highly Z-selective olefination of a-oxy
and a-amino ketones via ynolate anions under mild condi-
tions. The stereocontrol mechanism can be explained by or-
ꢀ
Figure 4. Orbital interaction of the transition states for cyclobutene ring
opening, as depicted by Murakami and Houk.
bital interactions between the s orbital of the breaking C O
bond or p orbital of the enolate and the s* orbital of the
ꢀ
ꢀ
C O or C N bonds of the substituent in the ring opening of
the b-lactone enolate intermediates, and/or the chelation to
lithium. This is therefore the first general method for the
olefination of ketones, and the stereochemistry can be theo-
retically predicted. In addition, the products are easily con-
verted to synthetically useful multisubstituted butenolides
and g-lactams. These results demonstrate that ynolates are
not only efficient olefination reagents but also can be used
as multifunctional carbanions.
pecially in ring opening of 4-methoxymethyl-4-methyloxet-
2-enoxide.
Alabugin and Zeidan published theoretical studies on the
hyperconjugative acceptor ability of s bonds.[42] According
ꢀ
ꢀ
to the NBO analysis, s*(C OH) and s*(C NH2) are much
ꢀ
ꢀ
better acceptors than s*(C H) and s*(C CH3). Based on
these reports and on our NBO second-order perturbation
analyses, we propose the following stereocontrol mechanism
of the olefination of a-oxy and a-amino ketones. In the ring
4
1
ꢀ
opening of b-lactone enolates, the breaking s(C O ) orbital
of the substituent overlaps with the s*(C OR) orbitals effi-
ciently to stabilize the inward transition state (Figure 5).
Experimental Section
5
ꢀ
General methods: 1HNMR spectra were measured in CDCl 3 solution
and referenced to TMS (0.00 ppm) on AL400 (400 MHz) spectrometers,
unless otherwise noted. 13C NMR spectra were measured in CDCl3 solu-
tion and referenced to CDCl3 (77.0 ppm) on AL400 spectrometers
(100 MHz). IR spectra were recorded on a JASCO FT/IR-410 spectrome-
ter. Mass spectra were obtained on a JEOL GX303 and GCMS (JMS-
AM SUN 200). Column chromatography was performed on silica gel
(Kanto Chemical Co.). Thin-layer chromatography was performed on
precoated plates (0.25 mm, silica gel Merck Kieselgel 60 F245). Tetrahy-
drofuran was freshly distilled from sodium benzophenone ketyl. tert-Bu-
tyllithium was titrated with diphenylacetic acid. All reactions were per-
formed in oven-dried glassware under positive pressure of argon, unless
otherwise noted. The stereochemistry was determined by NOE experi-
ments (Supporting Information), unless otherwise noted.
Figure 5. Transition state of inward rotation is stabilized by orbital inter-
4
1
5
ꢀ
ꢀ
action between s(C O ) and s*(C OR).
Representative procedure for olefination of a-oxyketones: (Z)-methyl 4-
methoxy-2-methyl-3-phenyl-2-pentenoate (7a): A solution of tert-butyl-
lithium (3.58 mL, 4.8 mmol, 1.34m in pentane) was added dropwise to a
solution of ethyl 2,2-dibromopropionate (312 mg, 1.2 mmol) in THF
(6 mL) at ꢀ788C under argon. The yellow solution was stirred for 3 h at
ꢀ788C and allowed to warm to 08C. After 30 min, the resulting colorless
reaction mixture was warmed to room temperature, and then a solution
of 2-methoxy-1-phenylpropan-1-one (6a, 164 mg, 1.0 mmol) in THF
(2 mL) was added. After 0.5 h, methyl iodide (0.62 mL, 10 mmol) and
hexamethylphosphoramide (HMPA) (1.7 mL, 10 mmol) were added.
After 18 h, saturated NH4Cl solution (10 mL) was added and the result-
ing mixture was extracted with ethyl acetate. The organic phase was
washed with water, saturated NaHCO3 solution, and brine, dried over
MgSO4, filtered, and concentrated to afford a yellow oil, which was puri-
fied by HPLC to yield 190 mg (81%) of the ester as a pale yellow oil.
1HNMR (400 MHz, CDCl 3): d=1.18 (d, J=6.4 Hz, 3H), 1.71 (s, 3H),
3.36 (s, 3H), 3.81 (s, 3H), 4.40 (q, J=6.4 Hz, 1H), 7.11 (dd, J=1.2 Hz,
8.0 Hz, 2H), 7.28–7.38 ppm (m, 3H); 13C NMR (100 MHz, CDCl3) d=
17.7, 19.9, 51.7, 56.7, 77.6, 127.2, 128.0, 128.1, 128.8, 137.3, 145.4,
The high Z selectivity of the ketones, even with less Lewis
basic or sterically hindered ethereal a-oxygen atoms, can be
explained by this mechanism, and lithium cations are not re-
quired for the stereoselectivity. In the conformationally
5
ꢀ
fixed enolates, the s* orbital of the axially oriented C OR
4
1
ꢀ
bond can efficiently overlap with that of the breaking C
O
5
ꢀ
bond, while the s* orbital of the equatorially oriented C
OR bond can not (Figure 6). This would explain why the
former transition states led to good Z selectivity and the
latter to none. The a-acyloxy ketones, which should have
170.7 ppm; IR (neat): n˜ =1730 cmꢀ1
;
MS (EI): m/z: 234 [M+], 59
(100%); HRMS (EI) calcd for C14H18O3: 234.1256; found: 234.1248.
3a-Hydroxy-3,3,6-trimethylhexahydrobenzofuran-2-one (29): A solution
of tert-butyllithium (3.18 mL, 4.8 mmol, 1.51m in pentane) was added
Figure 6. Orbital interactions of the conformationally fixed system.
dropwise to a solution of ethyl 2,2-dibromopropionate (312 mg,
Chem. Eur. J. 2006, 12, 524 – 536
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
533

M. Shindo et al.
1.2 mmol) in THF (6 mL) at ꢀ788C under argon. The yellow solution
was stirred for 3 h at ꢀ788C and allowed to warm to 08C. After 30 min,
the resulting pale yellow reaction mixture was cooled to ꢀ788C, and a so-
lution of a-oxyketone 8l (242 mg, 1.0 mmol) in THF (2 mL) was added.
After 0.5 h, methyl iodide (0.31 mL, 5 mmol) was added. After 1.5 h, sa-
turated NaHCO3 solution (10 mL) was added, and the mixture was al-
lowed to warm to room temperature. The resulting mixture was extracted
with ethyl acetate. The organic phase was washed with water, saturated
NaHCO3 solution, and brine, dried over MgSO4, filtered, and concentrat-
ed to give a 20:1 diastereomeric mixture of the b-lactone 28 (346 mg) as
4H), 2.50–2.56 ppm (m, 1H); 13C NMR (100 MHz, CDCl3): d=17.3 (q),
18.8 (q), 19.9 (q), 28.9 (t), 30.0 (t), 31.4 (d), 48.6 (t), 56.4 (s), 86.7 (s),
172.9 (s), 205.4 ppm (s); IR (neat): n˜ =1724, 1828 cmꢀ1; MS (EI): m/z:
196 [M+], 70 (100%); HRMS (EI) calcd for C11H16O3: 196.1099; found:
196.1098.
Synthesis of29 and 31 rfom 32 : NaBH4 (2.2 mg, 0.058 mmol) was added
to a solution of b-lactone 32 (11.4 mg) in MeOH(0.5 mL) at 0 8C, and
the mixture was stirred for 15 min at 08C. The reaction mixture was then
concentrated in vacuo. Water was added to the residue, which was ex-
tracted with AcOEt, dried over MgSO4, filtered, and concentrated to
give a 3:2 mixture of 31 and 29 (11.4 mg, 99%) as a colorless turbid oil.
a
pale yellow oil. Tetrabutylammonium fluoride (TBAF, 1.0 mL,
1.0 mmol, 1.0m in THF) was added to a solution of the b-lactone 28
(200 mg) in THF (10 mL) at 08C, and then the mixture was stirred for
1.5 h at room temperature. Water was added, and the reaction mixture
was extracted with Et2O, dried over MgSO4, filtered, and concentrated to
give a residue, which was purified by column chromatography (silica gel,
CHCl3/AcOEt 90/10) to afford g-lactone 29 (71.4 mg, 71% in two steps)
as a colorless solid. Lactone 29 was recrystallized from ethyl acetate/
hexane: m.p. 144–1458C; 1HNMR (400 MHz, CDCl 3): d=1.05 (d, J=
5.6 Hz, 3H), 1.17 (s, 3H), 1.23 (s, 3H), 1.43 (dt, J=4.4 Hz, 12.8 Hz, 1H),
1.48–1.71 (m, 5H), 1.99–2.04 (m, 1H), 2.17 (s, 1H), 4.20 ppm (dd, J=
4.0 Hz, 12.0 Hz, 1H); 13C NMR (100 MHz, CDCl3): d=17.3 (q), 19.9 (q),
21.8 (q), 27.8 (t), 30.8 (d), 31.5 (t), 47.8 (s), 78.0 (s), 80.8 (d), 181.6 ppm
3,4-Dimethyl-2(5H)-furanone (34)[43] (Table 4, entry 1): 6m HCl solution
was added to a solution of methyl ester 11 (30 mg, 0.078 mmol) in EtOH
(1 mL), and then the mixture was refluxed for 2.5 h. Water was added to
the reaction mixture and it was extracted with CH2Cl2. The organic layer
was washed with a saturated NaHCO3 solution (pH>7) and brine, dried
over MgSO4, and concentrated to give a residue, which was purified by
column chromatography (silica gel, hexane/AcOEt 80/20), followed by
preparative TLC to afford butenolide 34 (2.1 mg, 24%) as a pale yellow
oil. 1HNMR (400 MHz, CDCl 3): d=1.83 (s, 3H), 2.02 (s, 3H), 4.62 ppm
(s, 2H); 13C NMR (100 MHz, CDCl3): d=8.4 (q), 12.3 (q), 72.5 (t), 123.1
(s), 156.1 (s), 175.2 ppm (s); IR (neat): n˜ =1748 cmꢀ1; MS (EI): m/z: 112
[M+], 55 (100%); HRMS (EI) calcd for C6H8O2: 112.0524; found:
112.0530.
(s); IR (CHCl3): n˜ =1777, 3610 cmꢀ1 MS (EI): m/z: 198 [M+], 70
;
(100%); elemental analysis calcd (%) for C11H18O3: C 66.64, H9.15;
found: C 66.39, H9.13.
Representative procedure for one-pot synthesis of butenolides (Table 5,
conditions B): A solution of tert-butyllithium (3.48 mL, 4.8 mmol, 1.38m
in pentane) was added dropwise to a solution of ethyl 2,2-dibromopropi-
onate (312 mg, 1.2 mmol) in THF (6 mL) at ꢀ788C under argon. The
yellow solution was stirred for 3 h at ꢀ788C and allowed to warm to 08C.
After 30 min, the resulting colorless or pale yellow reaction mixture was
warmed to room temperature, and a solution of a-oxyketone 8a (188 mg,
1.0 mmol) in THF (2 mL) was added. After 0.5 h, 3% HCl/EtOH was
added to the mixture, which was refluxed for 3 h and concentrated in
vacuo. The residue was quenched with saturated NaHCO3 solution
(10 mL) and extracted with ethyl acetate. The organic phase was washed
with brine, dried over MgSO4, filtered, and concentrated. The residue
was purified by column chromatography (silica gel, hexane/AcOEt 90/10
to 80/20) to afford butenolide 43 (95.0 mg, 62%) as a yellow oil. 3-Butyl-
4-methyl-2(5H)-furanone (43)[44] (Table 5, entry 1): 1HNMR (400 MHz,
CDCl3): d=0.92 (t, J=7.2 Hz, 3H), 1.32 (sext, J=7.2 Hz, 2H), 1.48
(quint, J=7.2 Hz, 2H), 2.02 (s, 3H), 2.26 (t, J=7.6 Hz, 2H), 4.62 ppm (s,
2H); 13C NMR (100 MHz, CDCl3) d=12.2 (q), 13.7 (q), 22.4 (t), 23.0 (t),
29.9 (t), 72.3 (t), 127.1 (s), 156.4 (s), 174.9 ppm (s); IR (neat): n˜ =
1747 cmꢀ1; MS (EI): m/z: 154 [M+], 112 (100%).
CCDC-261697 contains the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
5-Hydroxy-3,3,7-trimethyl-1-oxaspiro[3.5]nonan-2-one (31): A solution of
tert-butyllithium (3.22 mL, 4.8 mmol, 1.49m in pentane) was added drop-
wise to a solution of ethyl 2,2-dibromopropionate (312 mg, 1.2 mmol) in
THF (6 mL) under argon at ꢀ788C. The yellow solution was stirred for
3 h at ꢀ788C and allowed to warm to 08C. After 30 min, the resulting
pale yellow solution was cooled to ꢀ788C and then a solution of a-oxy-
ketone 8m (242 mg, 1.0 mmol) in THF (2 mL) was added. After 0.5 h,
methyl iodide (0.31 mL, 5 mmol) was added and the solution was stirred
for 1.5 h. Saturated NaHCO3 solution (10 mL) was added, and the mix-
ture was allowed to warm to room temperature. The resulting mixture
was extracted with ethyl acetate. The organic phase was washed with
water, saturated NaHCO3 solution, and brine, dried over MgSO4, filtered,
and concentrated to give a 3:1 diastereomeric mixture of b-lactone 30 as
a
pale yellow oil. Tetrabutylammonium fluoride (TBAF, 1.0 mL,
1.0 mmol, 1.0m in THF) was added to a solution of b-lactone 30 in THF
(20 mL) at 08C, and the mixture was stirred for 1.0 h at room tempera-
ture. Water was added and the mixture was extracted with Et2O, dried
over MgSO4, filtered, and concentrated. The residue was purified by
column chromatography (silica gel, hexane/AcOEt 70/30) to afford b-lac-
General procedure for one-pot synthesis of the butenolides (Table 5, con-
ditions A): A solution of tert-butyllithium (4.8 equiv) was added dropwise
to
a solution of ethyl 2,2-dibromopropionate (1.2 equiv) in THF at
ꢀ788C under argon. The yellow solution was stirred for 3 h at ꢀ788C
and allowed to warm to 08C. After 30 min, the resulting colorless or pale
yellow reaction mixture was warmed to room temperature, and a solution
of the a-oxyketone (1.0 equiv) in THF was added. After 0.5 h, 3% HCl/
EtOHwas added, and the resulting mixture was concentrated in vacuo.
The residue was neutralized with saturated NaHCO3 solution and ex-
tracted with ethyl acetate. The organic phase was washed with brine,
dried over MgSO4, filtered, and concentrated. The residue was purified
by column chromatography to afford the butenolide.
tone 31 (107 mg, 54% in
2 steps) as a
pale yellow oil: 1HNMR
(400 MHz, CDCl3): d=0.93 (d, J=6.4 Hz, 3H), 1.19–1.30 (m, 1H), 1.30
(s, 3H), 1.39–1.41 (m, 1H), 1.45 (s, 3H), 1.47–1.51 (m, 1H), 1.61–1.69 (m,
1H), 1.73–1.83 (m, 1H), 1.86–2.01 (m, 2H), 4.13–4.16 ppm (m, 1H);
13C NMR (100 MHz, CDCl3): d=18.0 (q), 18.2 (q), 21.7 (q), 24.5 (d), 26.8
(t), 29.6 (t), 38.4 (t), 54.7 (s), 68.7 (d), 83.5 (s), 176.0 ppm (s); IR (neat):
n˜ =1805, 3492 cmꢀ1; MS (EI): m/z: 198 [M+], 70 (100%); HRMS (EI):
calcd for C11H18O3: 198.1256; found: 198.1254.
3,3,7-Trimethyl-1-oxaspiro[3,5] nonane-2,5-dione (32): N-Methylmorpho-
line N-oxide (NMO, 23.4 mg, 0.2 mmol) and powdered 4 Š molecular
General procedure for one-pot synthesis of g-lactam from a-amino
ketone (Table 6, method B): A solution of tert-butyllithium (4.8 equiv)
sieves (50 mg) were added to
a solution of b-lactone 31 (19.8 mg,
was added dropwise to
a solution of ethyl 2,2-dibromopropionate
0.1 mmol) in dry CH2Cl2 (1.0 mL) at room temperature. After 10 min,
tetrapropylammonium perruthenate (3.5 mg, 0.01 mmol) was added to
the mixture, which was stirred for 1 h at room temperature, then filtered
through Celite. The filtrate was washed with saturated Na2SO3, saturated
CuSO4, and brine. The organic layer was dried over MgSO4, filtered, and
concentrated. The residue was purified by column chromatography (silica
gel, hexane/AcOEt 70/30) to afford ketone 32 (17.1 mg, 87%) as a color-
less oil: 1HNMR (400 MHz, CDCl 3): d=1.05 (d, J=7.6 Hz, 3H), 1.34 (s,
3H), 1.48 (s, 3H), 1.74–1.81 (m, 1H), 1.83–1.91 (m, 1H), 2.22–2.42 (m,
(1.2 equiv) in THF at ꢀ788C under argon. The yellow solution was stir-
red for 3 h at ꢀ788C and allowed to warm to 08C. After 30 min, the re-
sulting colorless or pale yellow reaction mixture was warmed to room
temperature, and a solution of the a-amino ketone (1.0 equiv) in THF
was added. After 0.5 h, MeOSOCl (ca. 3 equiv), prepared from MeOH
and SOCl2 at ꢀ108C, was added at room temperature. After 15–20 h, the
reaction mixture was concentrated in vacuo, and MeOHwas added. The
residue was concentrated in vacuo, diluted with Et2O, acidified with 5%
HCl, and extracted with Et2O. The water phase was basified with 28%
534
www.chemeurj.org
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 524 – 536

Torquoselective Olefination of a-Oxy and a-Amino Ketones
FULL PAPER
aqueous NH3 and extracted with Et2O. The combined organic layer was
dried over MgSO4, filtered, and concentrated to give a residue, which
was purified by column chromatography to afford the lactam.
Am. Chem. Soc. 2002, 124, 14832–14833; d) N. Harada, A. Saito, N.
Koumura, H. Uda, B. de Lange, W. F. Jager, H. Wynberg, B. L. Fer-
inga, J. Am. Chem. Soc. 1997, 119, 7241–7248.
3,5-dimethyl-4-phenyl-1,5-dihydro-2-pyrrolone (59):[45] BF3·Et2O (0.063 mL,
[11] For recent examples of stereoselective construction of tetrasubstitut-
ed olefins via carbometalation of alkynes, see: a) K. Itami, T.
Kamei, J. Yoshida, J. Am. Chem. Soc. 2003, 125, 14670–14671; b) C.
Zhou, D. E. Emrich, R. C. Larock, Org. Lett. 2003, 5, 1579–1582;
c) N. Zhu, D. G. Hall, J. Org. Chem. 2003, 68, 6066–6069; d) P. E.
Tessier, A. J. Penwell, F. E. S. Souza, A. G. Fallis, Org. Lett. 2003, 5,
2989–2992; e) L. F. Tietze, K. Kahle, T. Raschke, Chem. Eur. J.
2002, 8, 401–407; f) J. P. Marino, H. N. Nguyen, J. Org. Chem. 2002,
67, 6291–6296; g) M. Hojo, Y. Murakami, H. Aihara, R. Sakuragi,
Y. Baba, A. Hosomi, Angew. Chem. 2001, 113, 641–643; Angew.
Chem. Int. Ed. 2001, 40, 621–623. For other approaches, see h) I.
Maciagiewicz, P. Dybowski, A. Skowronska, Tetrahedron 2003, 59,
6057–6066; i) M. Shimizu, T. Fujimoto, H. Minezaki, T. Hata, T.
Hiyama, J. Am. Chem. Soc. 2001, 123, 6947–6948; j) H. Kakiya, H.
Shinokubo, K. Oshima, Tetrahedron 2001, 57, 10063–10069; k) M.
Ide, M. Nakata, Synlett Synlett 2001, 1511–1515; l) X. Liu, M. Shimi-
zu, T. Hiyama, Angew. Chem. 2004, 116, 897–900; Angew. Chem.
Int. Ed. 2004, 43, 879–882.
[12] M. Shindo, Y. Sato, R. Koretsune, T. Yoshikawa, K. Matsumoto, K.
Itoh, K. Shishido, Chem. Pharm. Bull. 2003, 51, 477–478.
[13] a) M. Shindo, Tetrahedron Lett. 1997, 38, 4433–4436; b) M. Shindo,
Y. Sato, K. Shishido, Tetrahedron 1998, 54, 2411–2422; c) M.
Shindo, R. Koretsune, W. Yokota, K. Itoh, K. Shishido, Tetrahedron
Lett. 2001, 42, 8357–8360.
[14] For reviews, see a) M. Shindo, Synthesis 2003, 2275–2288; b) M.
Shindo, J. Syn. Org. Chem. Jpn. 2000, 58, 1155–1166; c) M. Shindo,
Yakugaku Zasshi 2000, 120, 1233–1246; d) M. Shindo, Chem. Soc.
Rev. 1998, 27, 367–374.
0.5 mmol) was added to
a solution of N-allyl lactam 60 (113 mg,
0.5 mmol) and 10% Pd/C (113 mg) in absolute EtOH(5 mL) at room
temperature, and then the mixture was refluxed. After 11 h, the reaction
mixture was filtered through Celite and concentrated. The residue was
purified by column chromatography (silica gel, hexane/AcOEt 40/60) to
afford g-lactam 59 (80.1 mg, 86%) as
a
yellow powder: 1HNMR
(400 MHz, CDCl3): d =1.21 (d, J=6.4 Hz, 3H), 2.02 (d, J=1.6 Hz, 3H),
4.59 (q, J=7.2 Hz, 1H), 7.33 (dd, J=1.2 Hz, 7.2 Hz, 2H), 7.39 (t, J=
7.6 Hz, 1H), 7.46 ppm (t, J=7.2 Hz, 2H); 13C NMR (100 MHz, CDCl3):
d=9.8 (q), 19.0 (q), 54.3 (d), 127.9 (d), 128.4 (d), 128.5 (d), 128.5 (s),
133.0 (s), 155.1 (s), 174.6 ppm (s); IR (neat): n˜ =3232, 1682 cmꢀ1; MS
(EI): m/z: 187 [M]+ (100%).
Acknowledgement
This work was supported in part by a Grant-in-Aid for Scientific Re-
search in the Priority Area (A) “Exploitation of Multi-Element Cyclic
Molecules” from the Ministry of Education, Culture, Sports, Science and
Technology, Japan, and PRESTO, JST. We thank Prof. R. L. Danheiser
(MIT) for helpful discussions. The generous allotment of computational
time from the Research Center for Computational Science, the National
Institutes of Natural Sciences, Japan, is gratefully acknowledged.
[15] a) M. Shindo, Y. Sato, K. Shishido, Tetrahedron Lett. 1998, 39, 4857–
4860; b) M. Shindo, Y. Sato, T. Yoshikawa, R. Koretsune, K. Shishi-
do, J. Org. Chem. 2004, 69, 3912–3916.
[1] S. E. Kelly in Comprehensive Organic Synthesis, Vol. 1 (Eds.: B. M.
Trost, I. Fleming, S. L. Schreiber), Pergamon, Oxford, 1991,
pp. 729–817.
[16] For pioneering work, see: a) C. J. Kowalski, K. W. Fields, J. Am.
Chem. Soc. 1982, 104, 321–323. For Lewis acid mediated olefination
with silyl ynol ethers, see: b) C. J. Kowalski, S. Sakdarat, J. Org.
Chem. 1990, 55, 1977–1979.
[2] For a representative review of Wittig olefination, see: B. E. Maryan-
off, A. B. Reitz, Chem. Rev. 1989, 89, 863–927.
[3] For a representative review of the Horner–Wadsworth–Emmons re-
action, see: W. S. Wadsworth, Org. React. 1977, 25, 73–253.
[4] For a review, see: P. R. Blakemore, J. Chem. Soc. Perkin Trans. 1
2002, 2563–2585.
[5] For reviews, see: a) D. J. Ager, Org. React. 1990, 38, 1; b) A. G. M.
Barrett, J. M. Hill, E. M. Wallace, J. A. Flygare, Synlett 1991, 764–
770; c) L. F. van Staden, D. Gravestock, D. Ager, Chem. Soc. Rev.
2002, 31, 195–200.
[6] For selected examples, see: a) F. N. Tebbe, G. W. Parshall, G. S.
Reddy, J. Am. Chem. Soc. 1978, 100, 3611–3613; b) N. A. Petasis,
E. I. Bzowej, J. Am. Chem. Soc. 1990, 112, 6392–6394. For reviews,
see c) O. G. Kulinkovich, A. de Meijere, Chem. Rev. 2000, 100,
2789–2834; d) R. C. Hartley, G. J. McKierman, J. Chem. Soc. Perkin
Trans. 1 2000, 2763–2793.
[7] For recent examples of olefination of ketones leading to tetrasubsti-
tuted olefins, see: a) S. Sano, K. Yokoyama, R. Teranishi, M. Shiro,
Y. Nagao, Tetrahedron Lett. 2002, 43, 281–284; b) S. Sano, T. Take-
hisa, S. Ogawa, K. Yokoyama, Y. Nagao, Chem. Pharm. Bull. 2002,
50, 1300–1302; c) T. Mantani, K. Shiomi, T. Konno, T. Ishihara, H.
Yamanaka, J. Org. Chem. 2001, 66, 3442–3448.
[8] For example, the Sharpless asymmetric epoxidation of trisubstituted
allylic alcohols (tetrasubstituted olefins) provides the corresponding
epoxides with high enantioselectivity. Since the products have con-
tiguous asymmetric carbon atoms, they would be useful for synthesis
of natural products: a) T. J. Erickson, J. Org. Chem. 1986, 51, 934–
935; b) J. A. Marshall, T. M Jenson, J. Org. Chem. 1984, 49, 1707–
1712.
[17] M. Shindo, Y. Sato, K. Shishido, J. Org. Chem. 2000, 65, 5443–5445.
[18] a) W. R. Dolbier, Jr., H. Koroniak, D. J. Burton, A. R. Bailey, G. S.
Shaw, S. W. Hansen, J. Am. Chem. Soc. 1984, 106, 1871; b) W.
Kirmse, N. G. Rondan, K. N. Houk, J. Am. Chem. Soc. 1984, 106,
7989–7991; c) N. G. Rondan, K. N. Houk, J. Am. Chem. Soc. 1985,
107, 2099–2111; d) E. A. Kallel, Y. Wang, D. C. Spellmeyer, K. N.
Houk, J. Am. Chem. Soc. 1990, 112, 6759–6763; e) S. Niwayama,
E. A. Kallel, C. Sheu, K. N. Houk, J. Org. Chem. 1996, 61, 2517–
2522; f) for a review, see: g) W. R. Dolbier, Jr., H. Koroniak, K. N.
Houk, C. Sheu, Acc. Chem. Res. 1996, 29, 471–477.
[19] a) H. Ikeda, T. Kato, S. Inagaki, Chem. Lett. 2001, 270–271; b) M.
Yasui, Y. Naruse, S. Inagaki, J. Org. Chem. 2004, 69. 7246–7249.
[20] M. Shindo, K. Matsumoto, S. Mori, K. Shishido, J. Am. Chem. Soc.
2002, 124, 6840–6841.
[21] S. Mori, M. Shindo, Org. Lett. 2004, 6, 3945–3949.
[22] F. Bonadies, A. Cardilli, A. Lattanzi, S. Pesci, A. Scettri, Tetrahe-
dron Lett. 1995, 36, 2839–2840.
[23] a) S. I. Pennanen, Tetrahedron Lett. 1977, 18, 2631–2632; b) S. I.
Pennanen, Tetrahedron Lett. 1980, 21, 657–658.
[24] For other olefinations of a-oxy ketones and aldehydes, see: a) M.
Koreeda, P. D. Patel, L. Brown, J. Org. Chem. 1985, 50, 5910–5912;
b) P. Garner, S. Ramakanth, J. Org. Chem. 1987, 52, 2629–2631.
c) C. Harcken, S. F. Martin, Org. Lett. 2001, 3, 3591–3593; d) K.
Ukai, D. Arioka, H. Yoshino, H. Fushimi, K. Oshima, K. Utimoto,
S. Matsubara, Synlett 2001, 513–514.
[25] In Cramꢁs chelation model, siloxy groups did not give efficient che-
lation. a) X. Chen, E. R. Hortelano, E. L. Eliel, S. V. Frye, J. Am.
Chem. Soc. 1992, 114, 1778–1784; b) S. Mori, M. Nakamura, E. Na-
kamura, N. Koga, K. Morokuma, J. Am. Chem. Soc. 1995, 117,
5055–5065; for a review, see: c) A. Mengel, O. Reiser, Chem. Rev.
1999, 99, 1191–1223.
[9] For an example, see: A. S. Levenson, V. C. Jordan, Eur. J. Cancer
1999, 35, 1628–1639.
[10] For recent examples, see: a) T. Fukaminato, T. Sakaki, T. Kawai, T.
Tamai, M. Irie, J. Am. Chem. Soc. 2004, 126, 14843–14849; b) J. F.
Valliant, P. Schaffer, K. A. Stephenson, J. F. Britten, J. Org. Chem.
2002, 67, 383–387; c) R. Rathore, M. I. Deselnicu, C. L. Burns, J.
Chem. Eur. J. 2006, 12, 524 – 536
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
535

M. Shindo et al.
[26] For translactonization of siloxy b-lactones to g-lactones, see: a) R.
Zemribo, M. S. Champ, D. Romo, Synlett 1996, 278–280; b) L. Four-
nier, A. Gaudel-Siri, P. J. Kocienski, J.-M. Pons, Synlett 2003, 107–
111.
[27] For a recent review, see: N. B. Carter, A. E. Nadany, J. B. Sweeney,
J. Chem. Soc. Perkin Trans. 1 2002, 2324–2342.
Cui, K. Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefa-
nov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts,
R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A.
Nanayakkara, C. Gonzalez, M. Challacombe, P. M. W. Gill, B. John-
son, W. Chen, M. W. Wong, J. L. Andres, C. Gonzalez, M. Head-
Gordon, E. S. Replogle, J. A. Pople, Gaussian, Inc., Pittsburgh, PA,
1998.
[28] S. Jaime-Figueroa, Y. Liu, J. M. Muchowski, D. G. Putman, Tetrahe-
dron Lett. 1998, 39, 1313–1316.
[29] For a representative review, see: D. Seebach, Angew. Chem. 1988,
100, 1685; Angew. Chem. Int. Ed. Engl. 1988, 27, 1624–1654.
[30] A. E. Reed, L. A. Curtiss, F. Weinhold, Chem. Rev. 1988, 88, 899–
926, and references therein.
[36] NBO Version 5.0 implemented by E. D. Glendening, A. E. Reed,
J. E. Carpenter, F. Weinhold.
[37] If one assumes unsolvated pathways, the chelated TSs are more
stable than the nonchelated ones. These results suggest that we must
take into account solvent coordination in the current studies.
[38] a) J. K. Badenhoop, F. Weinhold, J. Chem. Phys. 1997, 107, 5406–
5421; b) J. K. Badenhoop, F. Weinhold, J. Chem. Phys. 1997, 107,
5422–5432; c) J. K. Badenhoop, F. Weinhold, Int. J. Quantum Chem.
1999, 72, 269–280.
[39] a) S. Niwayama, E. A. Kallel, C. Sheu, K. N. Kouk, J. Org. Chem.
1996, 61, 2517–2522; b) S. Niwayama, E. A. Kallel, D. C. Spellmey-
er, C. Sheu, K. N. Houk, J. Org. Chem. 1996, 61, 2813–2825, and ref-
erences therein.
[40] a) M. Murakami, Y. Miyamoto, Y. Ito, Angew. Chem. 2001, 113,
182–184; Angew. Chem. Int. Ed. 2001, 40, 189–190; b) M. Muraka-
mi, M. Hasegawa, H. Igawa, J. Org. Chem. 2004, 69, 587–590; c) M.
Murakami, M. Hasegawa, Angew. Chem. 2004, 116, 4982–4984;
Angew. Chem. Int. Ed. 2004, 43, 4874–4876; d) M. Murakami, I.
Usui, M. Hasegawa, T. Matsuda, J. Am. Chem. Soc. 2005, 127,
1366–1367.
[41] P. S. Lee, X. Zhang, K. N. Houk, J. Am. Chem. Soc. 2003, 125, 5072–
5079.
[42] I. V. Alabugin, T. A. Zeidan, J. Am. Chem. Soc. 2002, 124, 3175–
3185.
[43] E. P. Woo, F. C. W. Cheng, J. Org. Chem. 1986, 51, 3706–3707.
[44] R. L. Edwards, A. J. S. Whalley, J. Chem. Soc. Perkin Trans. 1 1979,
803–806.
[45] B. Kryczka, A. Laurent, B. Marguet, Tetrahedron 1978, 34, 3291–
3298.
[31] a) V. Pophristic, L. Goodman, Nature 2001, 411, 565–568; b) F.
Weinhold, Nature 2001, 411, 539–541; c) F. M. Bickelhaupt, E. Baer-
ends, Angew. Chem. 2003, 115, 4315–4320; Angew. Chem. Int. Ed.
2003, 42, 4183–4188; d) F. Weinhold, Angew. Chem. 2003, 115,
4320–4326; Angew. Chem. Int. Ed. 2003, 42, 4188–4194; e) Y. Mo,
W. Mu, L. Song, M. Ling, Q. Zhang, J. Gao, Angew. Chem. 2004,
116, 2020–2024; Angew. Chem. Int. Ed. 2004, 43, 1986–1990;
f) Recent experimental results show that the equilibrium conforma-
tion of 3-hexyne is syn-eclipsed. R. K. Bohn, Phys. Chem. A J. Phys.
Chem. B 2004, 108, 6814–6816.
[32] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652; b) C. Lee, W.
Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
[33] W. J. Hehre, L. Radom, P. von R. Schleyer, J. A. Pople, Ab Initio
Molecular Orbital Theory, John Wiley, New York, 1986, and refer-
ences therein.
[34] For previous theoretical studies on ring opening of oxetenes, in addi-
tion to ref. [5], see: a) K. van der Meer, J. J. C. Mulder, Theor. Chim.
Acta. 1975, 37, 159–169; b) H. Yu, W-T. Chan, J. D. Goddard, J.
Am. Chem. Soc. 1990, 112, 7529–7537; c) R. P. Hsung, C. A. Zific-
sak, L.-L. Wei, C. J. Douglas, H. Xiong, J. A. Mulder, Org. Lett.
1999, 1, 1237–1240; d) M. Oblin, M. Rajzmann, J.-M. Pons, Tetrahe-
dron 2001, 57, 3099–3104.
[35] Gaussian98, RevisionA.11.3, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski,
J. A. Montgomery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich,
J. M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J.
Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli,
C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q.
Received: May 24, 2005
Published online: September 28, 2005
536
www.chemeurj.org
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 524 – 536