Diastereo- and enantioselective construction of γ-butenolides through chiral phosphane-catalyzed allylic alkylation of morita-baylis-hillman acetates

Article abstract of DOI:10.1002/ejoc.201100704

A series of multifunctional, chiral amide-phosphane organocatalysts have been designed and synthesized for the allylic substitution of Morita-Baylis-Hillman (MBH) acetate with 2-trimethylsilyloxyfuran for butenolide synthesis. This reaction was achieved in good to excellent yield (42-98%) and high ee (85-99%) with respect to a wide range of substrates in absolute MeOH or CH₃CN, using chiral amide-phosphane organocatalysts with an amide moiety including an active proton. NMR tracing experiments identified the critical phosphonium intermediates involved in the catalytic cycles. Computational studies disclosed the origins of diastereo- and enantioselectivity, in particular, revealing that the active proton of the amide moiety is the critical factor for the catalyst to have high enantiofacial control. The diastereo- and enantioselective construction of γ-butenolides was developed through allylic alkylation of Morita-Baylis-Hillman acetates catalyzed by chiral amide-phosphanes, affording the corresponding products in good yield (42-98%) and high ee (85-99%) with a wide range of substrates. The reaction mechanism was investigated by NMR tracing experiments and computational studies. Copyright

Full text of DOI:10.1002/ejoc.201100704

FULL PAPER
DOI: 10.1002/ejoc.201100704
Diastereo- and Enantioselective Construction of γ-Butenolides through Chiral
Phosphane-Catalyzed Allylic Alkylation of Morita–Baylis–Hillman Acetates
Yin Wei,[a] Guang-Ning Ma,[b] and Min Shi*[a,b]
Keywords: Synthetic methods / Asymmetric catalysis / Organocatalysis / Butenolide synthesis
A series of multifunctional, chiral amide–phosphane organo-
catalysts have been designed and synthesized for the allylic
substitution of Morita–Baylis–Hillman (MBH) acetate with 2-
trimethylsilyloxyfuran for butenolide synthesis. This reaction
was achieved in good to excellent yield (42–98%) and high
ee (85–99%) with respect to a wide range of substrates in
absolute MeOH or CH3CN, using chiral amide–phosphane
organocatalysts with an amide moiety including an active
proton. NMR tracing experiments identified the critical phos-
phonium intermediates involved in the catalytic cycles. Com-
putational studies disclosed the origins of diastereo- and
enantioselectivity, in particular, revealing that the active pro-
ton of the amide moiety is the critical factor for the catalyst
to have high enantiofacial control.
asymmetric methodologies.[1] In this context, using 2-silyl-
oxyfuran as a nucleophilic partner in the Mukaiyama aldol
Introduction
Efficient methods for the construction of γ-butenolide ring reaction,[2] Mukaiyama–Michael reaction,[3,4] and Mukai-
systems have received considerable attention as these ubiq- yama–Mannich-type addition (vinylogous Mannich reac-
uitous structural motifs have been found in numerous natu- tion)[5] has emerged as the effective strategy for butenolide
ral products and the γ-butenolide synthon has become a synthesis.[6] Recently, nucleophilic phosphane organocata-
valuable architectural platform for the development of new lysis has proved to be a powerful strategy to deliver multi-
Scheme 1. Postulated catalytic mechanism for γ-butenolide synthesis through tandem SN2Ј/SN2Ј substitution.
functional compounds.[7,8] Krische first reported the phos-
[a] State Key Laboratory of Organometallic Chemistry, Shanghai
phane-catalyzed allylic alkylation of MBH acetates with 2-
silyloxyfuran.[9,10] Upon exposure of MBH acetates to a
substoichiometric amount of triphenylphosphane (20 mol-
%) in the presence of 2-trimethylsilyloxyfuran (TMSOF),
regiospecific allylic substitution occurs to provide γ-buteno-
lides in good to excellent yields, with high region- and dia-
stereoselectivities. Krische suggested that the phosphane-
Institute of Organic Chemistry, Chinese Academy of Sciences,
354 Fenglin Road, Shanghai 200032, China
[b] Laboratory for Advanced Materials & Institute of Fine Chemicals,
School of Chemistry & Molecular Engineering, East China
University of Science & Technology,
130 MeiLong Road, Shanghai 200237, China
E-mail: mshi@mail.sioc.ac.cn
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201100704.
5146
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5146–5155

Diastereo- and Enantioselective Construction of γ-Butenolides
Scheme 2. Chiral phosphanes catalyzed allylic substitution of MBH acetates with TMSOF.
catalyzed allylic substitution of MBH acetates exhibited an showed low reactivities and enantioselectivities. Chen has
exceptionally high level of regiospecificity through a tan- developed the direct asymmetric γ-allylic alkylation of but-
dem SN2Ј/SN2’ mechanism (Scheme 1).[9,11] The generation enolides with MBH carbonates promoted by modified cin-
of an electrophile–nucleophile ion pair, which suppresses di- chona alkaloids, which provided an alternative method to
rect addition of the nucleophile to the less substituted enone access γ-butenolides.[13]
moiety of the starting MBH acetate, is proposed as a key
In this article, we demonstrate that highly diastereo- and
step in this catalytic transformation. We first developed the enantioselective allylic alkylation of a wide range of MBH
asymmetric version of allylic substitution of MBH acetates acetates with TMSOF can be achieved using a series of
with TMSOF to furnish γ-butenolides using an efficient multifunctional, chiral phosphane organocatalysts afford-
multifunctional chiral phosphane organocatalyst under ing γ-butenolides in good to excellent yields with high ee.
mild conditions (Scheme 2).[12] Our previous investigations Moreover, mechanistic investigations were undertaken to
on asymmetric versions of the allylic substitution of MBH reveal the stereochemical control on this allylic substitution.
acetates revealed that the amide moiety of the catalyst was
the efficient functional group for achieving high enantio-
selectivity. Further studies demonstrated that employing
Results and Discussion
water as an additive substantially increased the enantio-
selectivity for this asymmetric reaction. However, the sub-
strate scope was limited to MBH acetates derived from acti-
vated α,β-unsaturated ketones and aromatic aldehydes. The
Catalyst Screening
To seek more efficient catalysts and further evidence for
substrates derived from aliphatic aldehydes or acrylates the mechanism of stereochemical control of the phosphane-
Figure 1. Chiral phosphane catalysts L1–6.
Eur. J. Org. Chem. 2011, 5146–5155
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5147

Y. Wei, G.-N. Ma, M. Shi
FULL PAPER
catalyzed allylic substitution, we modified the amide moiety Table 1. Screening of L1–6 in the allylic substitution of MBH acet-
ate 3a in MeOH.[a]
and synthesized a series of catalysts (Figure 1). L1 and L2
were synthesized from the corresponding benzoyl chloride
and cyclopentanecarbonyl chloride. The phosphane–Schiff
base L3 was synthesized from 3,5-dichloro-2-hydroxybenz-
aldehyde and (R)-(–)-2-(diphenylphosphanyl)-1,1Ј-binaphth-
yl-2Ј-amine. Proline (pro) and its derivatives are proven
catalysts in a wide range of enantioselective reactions,[14]
which inspired us to involve the pro moiety in the structures
of L4 and L5. The chirality of the pro–amide moiety and
the chiral phosphane backbone may simultaneously affect
the stereochemical preference and enantioselectivity of the
catalyst. In order to find out whether there was a significant
effect of match/mismatch between the chirality of the pro
moiety and the chirality of the phosphane backbone, we
synthesized the matches of these N-Boc-pro-amide-phos-
phane catalysts L4 and L5 without the Boc protecting
Entry Catalyst
Time [h] Yield [%][c] ee [%][d] Absolute config.[e]
group. Additionally, the axially chiral phosphane-oxazoline
L6,[15] previously reported as a ligand in a phosphane/sil-
ver(I)-catalyzed asymmetric vinylogous Mannich reaction,
was also included for evaluation.
1
2
3
4
5
6
L1
L2
L3
(aR,S)-L4
(aR,S)-L5
L6
36
36
48
5
91
94
36
94
76
33
89
96
0
98
92
28
(1S,2R)
(1S,2R)
–
(1S,2R)
(1S,2R)
(1R,2S)
Our previous studies revealed that protic methanol can
significantly facilitate the reaction in the absence of water
with high enantioselectivity.[12] Thus, we initially examined
the performance of L1–6 in absolute MeOH. The MBH
acetate 1a (1.0 equiv.) and TMSOF (2, 1.5 equiv.) were em-
ployed as the starting materials for the asymmetric allylic
alkylation in absolute MeOH. The simplified amide–phos-
phane bifunctional L1 and L2 were found to be efficient
and highly enantioselective for this reaction, producing γ-
butenolide syn-(1S,2R)-3a in good yields (L1 91%, and L2
94%) with excellent enantioselectivities (L1 89% ee, and L2
96% ee, Table 1, entries 1 and 2). It should be mentioned
that the syn-diastereomer was generated predominantly (dr
12
36
[a] Reaction conditions: MBH acetate 1a (0.1 mmol, 25.3 mg),
TMSOF (2, 0.15 mmol, 25 μL), catalyst (0.02 mmol, 20 mol-%),
absolute MeOH (0.5 mL, 0.2 m), room temperature for the time
indicated. [b] dr Ͼ 95:5 means that the minor isomer could not be
detected by 1H NMR spectroscopy. [c] Isolated yields. [d] ee values
were determined by HPLC using a Chiralcel AD-H column. [e] The
absolute configurations were determined by comparing the optical
rotation [α]D with known data or HPLC analysis.
Ͼ95:5 indicating that the minor isomer could not be de- ficient and highly enantioselective for this reaction, produc-
tected by 1H NMR spectroscopy). However, the phos- ing γ-butenolide syn-(1S,2R)-3a in good yields (L1 65%, L2
phane–Schiff base L3 could not catalyze this reaction ef- 72%) with excellent enantioselectivities (L1 93% ee, L2
ficiently, furnishing the racemic product syn-3a in low yield 97% ee, Table 2, entries 1, 2). The phosphane–Schiff base
(36% yield, 0% ee, Table 1, entry 3). This result indicated L3 could not catalyze this reaction efficiently, furnishing
that the amide proton in the catalyst plays a critical role syn-(1R,2S)-3a in low yield and enantioselectivity (36%
in enantioselectivity. The catalysts (aR,S)-L4 and (aR,S)-L5 yield, 12% ee, Table 2, entry 3). (aR,S)-L4 and (aR,S)-L5
with pro moieties both facilitated this reaction. In particu- with pro moieties still gave good results. (aR,S)-L4 gave the
lar, (aR,S)-L4, which includes a Boc-protected group, gave best result, affording syn-(1S,2R)-3a in 76% yield with the
the best result with the shortest reaction time (5 h), afford- highest ee of 99% at room temperature (Table 2, entry 4).
ing syn-(1S,2R)-3a in 94% yield with the highest ee of 98% (aR,S)-L6, without an active amino proton, cannot mediate
at room temperature (Table 1, entry 4). Catalyst (aR,S)-L6, this transformation, only yielding trace amounts of product
without an active amino proton, yielded a product with the after 3 d (Table 2, entry 10). We wondered whether the con-
opposite configuration, syn-(1R,2S)-3a, in 33% yield with figuration of the stereocenter in the catalyst had an effect
28% ee (Table 1, entry 6), which again demonstrates that on the enantioselectivity, thus, we also conducted reactions
the amide proton in the catalyst significantly affects the using the N-Boc-pro-amide-phosphane catalysts L4 with
enantioselectivity.
different stereochemical preference (Table 2, entries 4–7).
In order to ignore the effect of the protic solvent on the All four catalysts facilitate this substitution reaction
enantioselectivity, we optimized the catalysts in the presence smoothly, giving γ-butenolide 3a over the syn-diastereomer.
of an aprotic solvent. Thus, we conducted the reactions (aR,S)-L4 and (aR,R)-L4, which have the same axial chiral-
using L1–6 (20 mol-%) in acetonitrile. The results are sum- ity but the opposite stereochemical configuration of the pro
marized in Table 2. The performance of L1–6 in acetonitrile stereocenter, have the same stereochemical preference, lead-
was similar to that in absolute methanol. The simplified ing to syn-(1S,2R)-3a in 76% yield with 99% ee and 54%
amide–phosphane bifunctional catalysts L1 and L2 are ef- yield with 96% ee, respectively (Table 2, entries 4 and 5). It
5148
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5146–5155

Diastereo- and Enantioselective Construction of γ-Butenolides
was found that replacing the l-pro moiety with d-pro some- Optimization of Reaction Conditions and Scope
how diminished the catalytic reactivity. On the other hand,
Using (aR,S)-L4, the reaction conditions were optimized
(aR,S)-L4 and (aS,S)-L4, which have the same chirality at
in terms of the solvent, catalyst loading, and temperature,
and the results are summarized in Table 3. Initially, using
the pro moiety but opposite axial chirality, led to products
with opposite configuration with similar ee values: (aR,S)-
L4 99% ee (1S,2R) vs. (aS,S)-L4 98% ee (1R,2S), (Table 2,
entry 4 vs. 6); (aR,R)-L4 96% ee (1S,2R) vs. (aS,R)-L4 98%
ee (1R,2S), (Table 2, entry 5 vs. 7). These results reveal that
the axial chirality of the chiral phosphane backbone deter-
mines the stereochemical preference of products. Removing
the N-Boc protecting group from (aR,S)-L4 and (aR,R)-L4
the reaction conditions considered optimal in our previous
study[12] (20 mol-% catalyst loading, toluene as the solvent,
6.0 equiv. H2O as the additive, at room temperature), the
reaction proceeded smoothly and 3a was isolated in 90%
yield with 96% ee (Table 3, entry 1). However, as mentioned
above, with absolute methanol as the solvent, the reaction
was complete within 5 h at room temperature, giving the
desired γ-butenolide 3a in 94% yield with 98% ee (Table 1,
entry 4). Presumably, a protic additive or solvent donates
resulted in (aR,S)-L5 and (aR,R)-L5, which have already
been used by our group as very efficient catalysts in an all-
ylic amination process,[16] gave the desired γ-butenolides in
protons, which can be involved in the formation of hydro-
moderate to good yields (45–70%), with slightly diminished
gen bonding in the key transition state (TS), leading to high
ee values (93–95%). It should be noted that the active
enantioselectivity. Decreasing the reaction temperature to
amino proton of the pro moiety did not influence the asym-
10 °C did not significantly affect the yield or enantio-
metric induction. (aR,S)-L4 has been identified as the most
selectivity but increased the reaction time (Table 3, entry 2).
efficient catalyst and was used for the further optimization
Decreasing the catalyst loading to 10 mol-% and 5 mol-%
of reaction conditions and the investigation of the scope of
diminished the yield without affecting the enantioselectivity
the reaction.
(Table 3, entries 3 and 4). We showed above that the reac-
tion catalyzed by (aR,S)-L4 proceeded smoothly in the
presence of CH3CN (Table 2, entry 4). We therefore investi-
gated the temperature effect again in the presence of
CH3CN. Increasing the temperature to 50 °C or decreasing
the temperature to –20 °C led to a decrease in yield and
enantioselectivity (Table 3, entries 5 and 6). Subsequently,
Table 2. Screening L1–6 in the allylic substitution of MBH acetate
3a in the presence of CH3CN.[a]
Table 3. Optimization of reaction conditions for the allylic substitu-
tion of the MBH acetate 3a.[a]
Entry
1
x
Solvent
Temp. Time Yield[c] ee[d]
[mol-%]
[°C]
[h]
[%]
[%]
Entry Catalyst
Time [h] Yield [%][c] ee [%][d] Absolute config.[e]
20
toluene/
r.t.
24
90
96
1
2
3
4
5
6
7
8
9
10
L1
L2
L3
(aR,S)-L4
(aR,R)-L4
(aS,S)-L4
(aS,R)-L4
(aR,S)-L5
(aR,R)-L5
(aR,S)-L6
48
48
72
24
72
72
48
48
72
72
65
72
36
76
54
47
83
70
45
93
97
12
99
96
98
98
95
(1S,2R)
(1S,2R)
(1R,2S)
(1S,2R)
(1S,2R)
(1R,2S)
(1R,2S)
(1S,2R)
(1S,2R)
n.d.
6.0 equiv. H2O
MeOH (absolute)
MeOH (absolute)
MeOH (absolute)
CH3CN
CH3CN
THF
THF/4 Å MS
CH2Cl2
Et2O
2
3
4
5
6
7
8
9
20
10
5
10
r.t.
r.t.
50
–20
r.t.
r.t.
r.t.
r.t.
r.t.
r.t.
10
48
48
24
96
12
20
12
24
12
48
93
65
54
40
50
85
65
80
83
63
59
98
98
97
85
98
92
73
95
92
96
99
20
20
20
20
20
20
20
20
93
trace
n.d.[f]
10
11
12
DCE
DMF
[a] Reaction conditions: MBH acetate 1a (0.1 mmol, 25.3 mg), 2
(0.15 mmol, 25 μL), catalyst (0.02 mmol, 20 mol-%), CH3CN
(0.5 mL, 0.2 m) at room temperature for the indicated time. [b] dr
[a] Reaction conditions: MBH acetate 1a (0.1 mmol, 25.3 mg), 2
(0.15 mmol, 25 μL), catalyst (x mol-%), solvent (0.5 mL, 0.2 m) for
the indicated time. [b] dr Ͼ 95:5 means that the minor isomer could
not be detected by 1H NMR spectroscopy. [c] Isolated yields. [d]
The ee values were determined by HPLC using a Chiralcel AD-H
column.
1
Ͼ 95:5 means that the minor isomer could not be detected by H
NMR spectroscopy. [c] Isolated yields. [d] The ee values were deter-
mined by HPLC using a Chiralcel AD-H column. [e] The absolute
configurations were determined by comparing the optical rotation
[α]D with known data or HPLC analysis. [f] n.d.: not determined.
Eur. J. Org. Chem. 2011, 5146–5155
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5149

Y. Wei, G.-N. Ma, M. Shi
FULL PAPER
we focused on the investigation of polar aprotic solvent ef- reaction. Thus, a series of MBH acetates derived from
fects on the enantioselectivities as protic species (e.g., H2O methyl vinyl ketone (MVK, 1b–p), ethyl vinyl ketone (EVK,
or MeOH) may be involved in the formation of hydrogen 1q and 1r), and methyl acrylate (1s) was prepared and its
bonds with the substrate and catalyst, which makes it diffi- reactivity explored in absolute MeOH (Table 4). Aromatic
cult to identify whether there are hydrogen bonds formed aldehyde-derived MBH adducts 1b–l with various substitu-
directly between the substrate and catalyst. Other polar tion patterns on the benzene ring underwent the reactions
aprotic solvents, such as CH2Cl2, DCE, Et2O, THF, and smoothly, affording the corresponding allylic alkylation
DMF (dried by standard methods), were also suitable for products 3b–l in good to excellent yields (75–97%) along
this reaction, affording γ-butenolide 3a in good to excellent with excellent diastereoselectivities (dr Ͼ 95:5) and enantio-
enantioselectivities (73–99% ee, Table 3, entries 7–12). In selectivities (94–98% ee) (Table 4, entries 1–11). The sub-
general, the allylic substitution of MBH acetates with strates prepared from aliphatic aldehydes, 1m and 1n, led to
TMSOF catalyzed by (aR,S)-L4 (20 mol-%) can proceed the corresponding γ-butenolides 3m and 3n in 60% yield
either in the presence of a protic solvent (MeOH) or aprotic with 85% ee and in 42% yield with 88% ee, respectively
solvent (CH3CN) at room temperature, affording the prod- (Table 4, entries 12 and 13). Similarly, the heteroatom-con-
uct in high yield with excellent enantioselectivity.
taining substrates, 1o and 1p, provided the corresponding
With these optimized conditions in hand, our next aim γ-butenolides in good yields and excellent enantioselectivi-
was to examine the substrate scope and limitations of the ties (3o: 74% yield, 97% ee; 3p: 73% yield, 96% ee, Table 4,
entries 14 and 15). As for the EVK-derived MBH acetates
1q and 1r, the alkylation products 3q and 3r were obtained
Table 4. (aR,S)-L4 catalyzes the allylic substitution of various in 92% yield with 97% ee and in 61% yield with 94% ee,
MBH acetates 1b–s with TMSOF in absolute MeOH.[a]
respectively (Table 4, entries 16 and 17). However, the acry-
late-derived substrate 1s, which was less electrophilic,
showed less reactivity, affording the corresponding product
3s in low yield (50%) but with high enantioselectivity (91%
ee) (Table 4, entry 18).
Subsequently, we investigated the reactions 1b–s in
CH3CN (Table 5). Aromatic aldehyde-derived MBH ad-
ducts 1b–j with electron withdrawing substituents or no
substituent on the benzene ring also underwent the reac-
Table 5. (aR,S)-L4 catalyzes the allylic substitution of various
MBH acetates 1b–s with TMSOF in CH3CN.[a]
Entry Substrate R1
R2
Time [h] Yield [%][c] ee [%][d]
1
2
3
4
5
6
7
8
1b
1c
1d
1e
1f
1g
1h
1i
4-BrC6H4
4-FC6H4
3-ClC6H4
2-ClC6H4
4-CNC6H4 Me
4-CF3C6H4 Me
4-NO2C6H4 Me
3-NO2C6H4 Me
C6H5
4-NO2C6H4 Et
C6H5 Et
4-NO2C6H4 OMe
Me
Me
Me
Me
12
12
24
24
20
12
24
36
72
36
48
96
3b: 88
3c: 70
3d: 87
3e: 94
3f: 98
3g: 98
3h: 95
3i: 95
3j: 75
3q: 83
3r: 66
3s: 50
99
96
97
99
97
98
98
98
97
95
92
93
9
1j
Me
10
11
12
1q
1r
1s
[a] Reaction conditions: MBH acetate (1, 0.1 mmol), 2 (0.15 mmol, [a] Reaction conditions: MBH acetate (1, 0.1 mmol), 2 (0.15 mmol,
25 μL), (aR,S)-L4 (0.02 mmol, 20 mol-%, 13.0 mg), CH3CN
(0.5 mL, 0.2 m) at room temperature for the indicated time. [b] dr
Ͼ 95:5 means that the minor isomer could not be detected by H
NMR spectroscopy. [c] Isolated yields. [d] The ee values were deter-
mined by HPLC using a Chiralcel AD-H or OD-H column.
25 μL), (aR,S)-L4 (0.02 mmol, 20 mol-%, 13.0 mg), CH3CN
(0.5 mL, 0.2 m) at room temp. for the indicated time. [b] dr Ͼ 95:5
means that the minor isomer could not be detected by 1H NMR
spectroscopy. [c] Isolated yields. [d] The ee values were determined
by HPLC using a Chiralcel AD-H or OD-H column.
1
5150
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5146–5155

Diastereo- and Enantioselective Construction of γ-Butenolides
tions smoothly, affording the corresponding products 3b–3j NMR Tracing Experiments
in high yields with excellent ee (Table 5, entries 1–9). These
Having established a series of new efficient multifunc-
results are similar to their performance in the presence of
absolute MeOH. However, we observed that the reactions
in the presence of CH3CN need longer reaction times than
in the presence of absolute MeOH. The reactions of aro-
matic aldehyde-derived MBH adducts with electron donat-
ing substituents on the benzene ring (1k, 1l), aliphatic alde-
hyde-derived MBH adducts (1m, 1n), and heteroatom-con-
taining substrates (1o, 1p) were extremely sluggish and
could not achieve significant yields in a week. Thus we
stopped these reactions and did not analyze the outcomes.
As for the EVK-derived MBH acetates 1q and 1r, the corre-
sponding alkylation products 3q and 3r were obtained in
83% yield with 95% ee and in 66% yield with 92% ee,
respectively (Table 5, entries 10 and 11). However, the acry-
late-derived substrate 1s also showed less reactivity in
CH3CN, affording 3s in low yield (50%), but with high
enantioselectivity (93% ee) (Table 5, entry 12).
As shown in Scheme 1, the catalytic mechanism for γ-
butenolide synthesis was proposed to be a tandem SN2Ј/
SN2Ј substitution. The reaction is initiated by conjugate ad-
dition of the catalyst (phosphane) and MBH adducts to
furnish an electrophilic leaving group ion pair, which reacts
with the pronucleophile (TMSOF) to form hypervalent
anions or “ate” complexes, and the requisite electrophile–
nucleophile ion pair intermediates are generated. In order
to examine the leaving group effect for the phosphane-cata-
lyzed allylic alkylation of MBH adducts, a series of acyl
derivatives based on MBH adducts were prepared and ex-
posed to 20 mol-% (aR,S)-L4 at room temperature in the
presence of CH3CN. Under the typical reaction conditions,
the allylic substitution proceeded smoothly for all the MBH
adducts bearing different leaving groups, producing γ-but-
enolide 3h in moderate to good yields (95% R = Ac, 85%
R = Boc, 60% R = Bz, 36% R = p-NO2-Bz) with high
enantioselectivities (94–98% ee) (Table 6, entries 1–4).
tional amide–phosphane organocatalysts for allylic substi-
tutions of MBH acetates with TMSOF and the reaction
scope and limitations, we investigated the mechanistic de-
tails by NMR tracing experiments (see Supporting Infor-
mation for details). We identified critical phosphonium in-
termediates involved in the catalytic cycles.
Theoretical Investigation on the Origins of Diastereo- and
Enantioselectivity
Krische has proposed that the high diastereoselectivity
attained in these substitutions arises as a consequence of a
mechanism involving endo-selective Diels–Alder cycload-
dition of the siloxyfuranate complex with the enone ob-
tained by the addition of the phosphane followed by subse-
quent Grob-type fragmentation (Scheme 3).[9] On the basis
of the TSs proposed by Krische and TS geometries pro-
posed for various nucleophilic additions to carbonyl groups
in vinylogous Mannich reactions,[17] a set of four limiting
TSs are depicted in Scheme 4 and their structures optimized
at the HF/3-21G* level are shown in Figure 2. The energies
of these TSs were calculated at the MP2/6-31G(d)//HF/3-
21G* level of theory and their relative energies are shown
in Scheme 4. In terms of energy, Diels–Alder-like TS A and
B are much more stable than open TS C and D by 80 kJ/
mol and 64.5 kJ/mol or so, respectively, presumably due to
the π–π stacking interaction between the furan ring and
C=C double bond in TS A and B. The repulsion between
the carbonyl group and OTMS group in exo TS B may lead
to an increase in energy. Thus, the energy of endo TS A is
lower than that of exo TS B by 15.4 kJ/mol, leading to the
syn-diastereomer as the major product, consistent with ex-
Table 6. Further investigation on the leaving group effect of the
MBH acetate 1.[a]
Entry
R
Time [h]
Yield [%][c]
ee [%][d]
1
2
3
4
Ac
Boc
Bz
p-NO2-Bz
24
24
48
48
95
85
60
36
98
98
94
97
[a] Reaction conditions: MBH acetate (1, 0.1 mmol, 32.1 mg), 2
(0.15 mmol, 25 μL), (aR,S)-L4 (0.02 mmol, 20 mol-%), CH3CN
(0.5 mL, 0.2 m) at room temperature for the indicated time. [b] dr
1
Ͼ 95:5 means that the minor isomer could not be detected by H
NMR spectroscopy. [c] Isolated yields. [d] The ee values were deter-
mined by HPLC using a Chiralcel AD-H column.
Scheme 3. Krische’s proposed transition states for diastereoselectiv-
ity.
Eur. J. Org. Chem. 2011, 5146–5155
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5151

Y. Wei, G.-N. Ma, M. Shi
FULL PAPER
Scheme 4. The transition states corresponding to the study of the origin of diastereoselectivity.
Figure 2. The optimized structures of TS A–D at the HF/3-21G* level.
5152
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5146–5155

Diastereo- and Enantioselective Construction of γ-Butenolides
perimental observations. On the other hand, the energy dif- face or si face attack. We investigated the relative energies
ference of open TS C and D is small (only 0.3 kJ/mol), of four key TSs involving (aR,S)-L4 shown in Scheme 5,
hence it cannot account for high the diastereoselectivity ob- and their optimized structures are depicted in Figure 3,
served experimentally.
which may account for the stereochemistry observed experi-
Having rationalized the origin of diastereoselectivity, we mentally. Similar to the achiral phosphane, the energy of
further investigated the origin of enantioselectivity and the endo TS I involving (aR,S)-L4 is lower than that of exo TS
absolute stereochemistry of the substitution adducts in the III by 21.6 kJ/mol, leading to high diastereoselectivity. If
catalysis of chiral phosphanes. Enantiofacial control is pre- TMSOF approached the enone at the re face, a strong hy-
sumably governed by the energy differentiation between two drogen bond could form between the amide N–H and the
modes of attack of TMSOF towards the enone, namely, re substrate carbonyl group, resulting in the more stable TS I
Scheme 5. The transition states of (aR,S)-L4 corresponding to the study of origin of enantioselectivity.
Figure 3. The optimized structures of TS I–IV at the HF/3-21G* level.
Eur. J. Org. Chem. 2011, 5146–5155
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5153

Y. Wei, G.-N. Ma, M. Shi
FULL PAPER
with a Yanagimoto micro melting apparatus and uncorrected.
NMR spectra were recorded with a Bruker spectrometer 400 MHz
or 300 MHz (1H NMR), 75 MHz or 100 MHz (13C NMR) in
CDCl3, respectively. Chemical shifts were reported in ppm down-
field from internal TMS. 31P NMR spectra were recorded at
162 MHz or 121 MHz in CDCl3 with 85% H3PO4 as the external
reference. J values are in Hertz. Optical rotations were determined
at 589 nm (sodium D line) using a Perkin–Elmer 341 MC Polarime-
(NH···O=C 1.831 Å) and TS III (NH···O=C 1.880 Å).
However, the amide N–H and the substrate carbonyl group
are orientated parallel to each other in TS II and TS IV,
which cannot lead to a hydrogen-bonding interaction. In
addition, a strong hydrogen bond could not be formed in
TS II, and the repulsion between the phosphane and OTMS
moieties may be another reason for the relatively high en-
ergy of TS II. Thus, the large energy difference between
TS I and TS II may account for the high enantioselectivity
ter and [α] values are given with units of 10 cm2 deg–1 g–1. Mass
D
spectra were recorded with a HP-5989 instrument using EI/ESI
observed experimentally. The amide active proton stabilizes methods. IR spectra were recorded with a Perkin–Elmer PE-983
spectrometer with absorption in cm–1. Chiral HPLC was performed
one of the TS diastereoisomers, leading to high enantio-
using a SHIMADZU SPD-10A vp series instrument with chiral
selectivity. Therefore, L3 and L6 without an amide active
columns (Chiralpak AD-H, OD-H columns, φ 4.6ϫ250 mm,
proton lose enantiofacial control. In a similar manner, we
further investigated four TSs V–VIII for chiral (aR,S)-L5
without the Boc group and obtained similar results (see
Supporting Information for details).
Daicel Chemical Co. Ltd.). Organic solvents were dried by stan-
dard methods where necessary. Commercially available reagents
were used without further purification. All reactions were moni-
tored by TLC with Huanghai GF254 silica gel coated plates. Flash
We have previously reported that the addition of H2O
column chromatography was carried out using 300–400 mesh silica
improved enantioselectivity.[12] We investigated the key TSs
gel at increased pressure. MBH acetates were prepared according
to a literature method.[18]
using the catalyst shown in Scheme 2 with one molecule of
H2O to explore its effect on the enantioselectivity. Prelimi-
nary results revealed that a hydrogen-bonding interaction
between the C=O group in the amide terminus of the cata-
lyst and water could be formed in the endo TSs. In the re
General Procedure for the Asymmetric Allylic Alkylation of MBH
Acetates: To a flame-dried Schlenk tube charged with MBH acetate
(0.1 mmol, 25.3 mg), (aR,S)-L4 or (aR,S)-L5 (0.02 mmol), and an-
hydrous CH2Cl2 or CH3CN (0.5 mL) was added 2 (0.15 mmol,
face approaching mode, the oxygen atom in water has an 25 μL) under argon. The reaction was allowed to stir at 25 °C under
argon. The reaction was monitored by TLC analysis. After the
starting substrates were consumed, the solvent was removed under
reduced pressure, and the residue was purified by silica gel column
chromatography (eluent: EtOAc/petroleum ether = 1:6 to 1:4) to
give the corresponding alkylation adducts 3.
interaction with the Si atom in the OTMS group, which
directs the OTMS group to leave. However, this kind of
interaction cannot be formed in the si face approaching
mode (see Supporting Information for details). This may
explain why the addition of water can increase the enantio-
selectivity.
(S)-5-[(R)-1-(4-Chlorophenyl)-2-methylene-3-oxobutyl]furan-2(5H)-
one (3a): Yield: 99%; this is a known compound. [α]2D5 = –46.0 (c
= 0.74, CHCl3). 1H NMR (400 MHz, CDCl3, TMS): δ = 7.31–7.27
(m, 5 H), 6.36 (s, 1 H), 6.08 (dd, J = 2.0, 5.6 Hz, 1 H), 6.03 (s, 1
H), 5.52 (ddd, J = 1.6, 3.2, 6.4 Hz, 1 H), 4.41 (d, J = 6.4 Hz, 1 H),
2.31 (s, 3 H) ppm. 13C NMR (100 MHz, CDCl3, TMS): δ = 198.4,
172.5, 155.6, 145.5, 137.6, 133.4, 129.8, 129.6, 129.0, 121.9, 83.0,
46.5, 25.7 ppm. HPLC condition: Chiralcel AD-H column, λ =
230 nm, eluent: hexane/2-propanol = 90:10, flow rate: 0.8 mL/min,
tRmajor = 24.03 min, tRminor = 22.12 min; ee = 98%.
Conclusions
We have synthesized a series of multifunctional chiral
amide–phosphane organocatalysts L1–6 and demonstrated
that a highly enantioselective allylic substitution of MBH
acetate with TMSOF could be achieved by multifunctional
phosphane organocatalysts L1–5 with an active proton on
the amide moiety. The enantioselective allylic substitution
reaction was achieved for butenolide synthesis in good
yields with high ee with respect to a wide range of MBH
acetates. NMR tracing experiments provided some evidence
for the existence of phosphonium intermediates involved in
the catalytic cycle, which supported the proposed mecha-
nism. The computational studies disclosed that Diels–
Alder-like transition states could account for the high dia-
stereoselectivity attained in these phosphane-catalyzed al-
lylic substitution reactions. Theoretical investigations also
Supporting Information (see footnote on the first page of this arti-
cle): Spectroscopic and analytical data for all catalysts, 3, and
NMR tracing experiments; detailed descriptions of experimental
procedures and computational details are also available.
Acknowledgments
Financial support from the Shanghai Municipal Committee of Sci-
ence and Technology (06XD14005, 08dj1400100–2), the National
Basic Research Program of China (973)-2010CB833302, and the
revealed that the active proton of the amide moiety is the National Natural Science Foundation of China (21072206,
20472096, 20902019, 20872162, 20672127, 20821002, 20732008,
and 20702059) is gratefully acknowledged.
critical factor for the catalyst to have high enantiofacial
control. Studies directed towards more efficient chiral cata-
lysts for enantioselective allylic substitution reactions and
their variants are underway.
[1] Reviews on biologically active natural compounds with buteno-
lide moieties. a) For acetogenins, see: F. W. Alali, X.-X. Liu,
J. L. McLaughlin, J. Nat. Prod. 1999, 62, 504–540; b) for car-
diac steroids or cardenolides, see: J. R. Hanson, Nat. Prod. Rep.
2002, 19, 381–389, and references cited therein; c) for furan-
ocembranoid diterpenes, see: A. D. Rodriguez, Tetrahedron
1995, 51, 4571–4618.
Experimental Section
General Remarks: All reactions and manipulations were performed
using standard Schlenk techniques. Melting points were measured
5154
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 5146–5155

Diastereo- and Enantioselective Construction of γ-Butenolides
[2] For a review with regard to the butenolide synthesis from vi-
nylogous aldol reaction, see: G. Casiraghi, F. Zandari, G. Ap-
pendino, G. Rassu, Chem. Rev. 2000, 100, 1929–1972.
[3] For a review on butenolide synthesis from vinylogous Michael
reactions, see: J. Christoffers, Synlett 2001, 723–732, and refer-
ences cited therein.
[4] For recent examples on enantioselective vinylogous Michael re-
actions related to butenolide synthesis, see: a) S. P. Brown,
N. C. Goodwin, D. W. C. MacMillan, J. Am. Chem. Soc. 2003,
125, 1192–1194; b) H. Kitajima, K. Ito, T. Katsuki, Tetrahe-
dron 1997, 53, 17015–17028.
[5] For reviews with respect to the butenolide synthesis from the
corresponding vinylogous Mannich reaction, see: a) S. K. Bur,
S. F. Martin, Tetrahedron 2001, 57, 3221–3242, and references
cited therein; b) S. F. Martin, Acc. Chem. Res. 2002, 35, 895–
904.
[6] Reviews on other methods for the synthesis of γ-butenolides:
a) P. Langer, Synlett 2006, 3369–3381; b) G. Romeo, D. Ian-
nazzo, A. Piperno, R. Romeo, A. Corsaro, A. Rescifina, U.
Chiacchio, Mini-Rev. Org. Chem. 2005, 2, 59–77; c) M. Ito,
Pure Appl. Chem. 1991, 63, 13–22; d) R. Brückner, Curr. Org.
Chem. 2001, 5, 679–718; e) P. A. Jacobi, Adv. Heterocycl. Nat.
Prod. Synth. 1992, 2, 251–298.
6264; n) J.-M. Garnier, C. Anstiss, F. Liu, Adv. Synth. Catal.
2009, 351, 331–338; o) J.-M. Garnier, F. Liu, Org. Biomol.
Chem. 2009, 7, 1272–1275; p) H. Xiao, Z. Chai, C.-W. Zheng,
Y.-Q. Yang, W. Liu, J.-K. Zhang, G. Zhao, Angew. Chem. 2010,
122, 4569–4572; Angew. Chem. Int. Ed. 2010, 49, 4467–4470;
q) J.-J. Gong, T.-Z. Li, K. Pan, X.-Y. Wu, Chem. Commun.
2011, 47, 1491–1493; r) X. Y. Han, Y. Q. Wang, F. R. Zhong,
Y. X. Lu, J. Am. Chem. Soc. 2011, 133, 1726–1729.
[9]
C.-W. Cho, M. J. Krische, Angew. Chem. 2004, 116, 6857–6859;
Angew. Chem. Int. Ed. 2004, 43, 6689–6691.
[10]
For other reactions of allylic substitution of MBH adducts and
nucleophilic catalysis based on phosphane conjugate addition,
see: a) C.-W. Cho, J.-R. Kong, M. J. Krische, Org. Lett. 2004,
6, 1337–1339; b) L.-C. Wang, A. L. Luis, K. Agapiou, H.-Y.
Jang, M. J. Krische, J. Am. Chem. Soc. 2002, 124, 2402–2403;
c) P. K. Koech, M. J. Krische, J. Am. Chem. Soc. 2004, 126,
5350–5351; d) Y. Du, X. Han, X. Lu, Tetrahedron Lett. 2004,
45, 4967–4971; e) T.-Z. Zhang, L.-X. Dai, X.-L. Hou, Tetrahe-
dron: Asymmetry 2007, 18, 1990–1994; f) D. J. V. C. van Steenis,
T. Marcelli, M. Lutz, A. L. Spek, J. H. van Maarseveen, H.
Hiemstra, Adv. Synth. Catal. 2007, 349, 281–286; g) H. Park,
C.-W. Cho, M. J. Krische, J. Org. Chem. 2006, 71, 7892–7894.
For recent mechanistic studies on amine-catalyzed nucleophilic
substitutions of Baylis–Hillman acetates, see: M. Baidya, G. Y.
Remennikov, P. Mayer, H. Mayr, Chem. Eur. J. 2010, 16, 1365–
1371.
a) Y.-Q. Jiang, Y.-L. Shi, M. Shi, J. Am. Chem. Soc. 2008, 130,
7202–7203; b) H.-P. Deng, Y. Wei, M. Shi, Eur. J. Org. Chem.
2011, 1956–1960; c) Y.-L. Yang, C.-K. Pei, M. Shi, Org. Bi-
omol. Chem. 2011, 9, 3349–3358.
[11]
[12]
[13]
[7] For reviews on phosphane organocatalysis, see: a) D. S.
Glueck, Chem. Eur. J. 2008, 14, 7108–7117; b) G. C. Fu, Acc.
Chem. Res. 2006, 39, 853–860; c) J. L. Methot, W. R. Roush,
Adv. Synth. Catal. 2004, 346, 1035–1050; d) X. Lu, C. Zhang,
Z. Xu, Acc. Chem. Res. 2001, 34, 535–544; e) Y. Wei, M. Shi,
Acc. Chem. Res. 2010, 43, 1005–1018; f) A. Marinetti, A. Voitu-
riez, Synlett 2010, 174–194.
[8] For recent examples of phosphane catalyzed reactions, see: a)
J. E. Wilson, G. C. Fu, Angew. Chem. 2006, 118, 1454–1457;
Angew. Chem. Int. Ed. 2006, 45, 1426–1429; b) Z. Lu, S. Zheng,
X. Lu, Org. Lett. 2008, 10, 3267–3270; c) X.-F. Zhu, J. Lan, O.
Kwon, J. Am. Chem. Soc. 2003, 125, 4716–4717; d) V. Srira-
murthy, G. A. Barcan, O. Kwon, J. Am. Chem. Soc. 2007, 129,
12928–12929; e) Y. S. Tran, O. Kwon, J. Am. Chem. Soc. 2007,
129, 12632–12633. For recent reports of chiral bifunctional
phosphane organocatalyzed asymmetric reactions, see: f) K.
Matsui, S. Takizawa, H. Sasai, Synlett 2006, 761–763; g) B. J.
Cowen, S. J. Miller, J. Am. Chem. Soc. 2007, 129, 10988–10989;
h) M.-J. Qi, T. Ai, M. Shi, G. Li, Tetrahedron 2008, 64, 1181–
1186; i) T. Kano, Y. Yamaguchi, O. Tokuda, K. Maruoka, J.
Am. Chem. Soc. 2005, 127, 16408–16409; j) Y.-Q. Fang, E. N.
Jacobsen, J. Am. Chem. Soc. 2008, 130, 5660–5661; k) Z.-Y.
Lei, X.-G. Liu, M. Shi, M. Zhao, Tetrahedron: Asymmetry
2008, 19, 2058–2062; l) X. T. Meng, Y. Huang, R. Y. Chen,
Chem. Eur. J. 2008, 14, 6852–6856; m) K. Yuan, L. Zhang, H.-
L. Song, Y. Hu, X.-Y. Wu, Tetrahedron Lett. 2008, 49, 6262–
H.-L. Cui, J.-R. Huang, J. Lei, Z.-F. Wang, S. Chen, L. Wu,
Y.-C. Chen, Org. Lett. 2010, 12, 720–723.
[14] For selected reviews on proline-catalyzed reactions, see: a) B.
List, Tetrahedron 2002, 58, 5573–5590; b) M. Movassaghi,
E. N. Jacobsen, Science 2002, 298, 1904–1905; c) W. Notz, F.
Tanaka, C. F. Barbas III, Acc. Chem. Res. 2004, 37, 580–591.
[15] H.-P. Deng, Y. Wei, M. Shi, Adv. Synth. Catal. 2009, 351, 2897–
2902.
[16] G.-N. Ma, S.-H. Cao, M. Shi, Tetrahedron: Asymmetry 2009,
20, 1086–1092.
[17] The four limiting TSs described here are similar to those
studied in the vinylogous Mannich reaction, see: S. K. Bur,
S. F. Martin, Org. Lett. 2000, 2, 3445–3447.
[18] a) M. Shi, J.-K. Jiang, C.-Q. Li, Tetrahedron Lett. 2002, 43,
127–129; b) M. Shi, C.-Q. Li, J.-K. Jiang, Chem. Commun.
2001, 833–834.
Received: May 19, 2011
Published Online: August 5, 2011
Eur. J. Org. Chem. 2011, 5146–5155
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5155