Hydroxyl ionic liquid (HIL)-immobilized quinuclidine for Baylis-Hillman catalysis: Synergistic effect of ionic liquids as organocatalyst supports

Article abstract of DOI:10.1016/j.tet.2005.12.045

Hydroxyl ionic liquid (HIL) has been explored as a novel support for Baylis-Hillman catalyst. The HIL-supported catalyst showed a better catalytic activity compared to other IL-immobilized catalyst that has no hydroxyl group attached to the IL scaffold. T

Full text of DOI:10.1016/j.tet.2005.12.045

Tetrahedron 62 (2006) 2537–2544
Hydroxyl ionic liquid (HIL)-immobilized quinuclidine
for Baylis–Hillman catalysis: synergistic effect of ionic
liquids as organocatalyst supports
Xueling Mi,a Sanzhong Luo,b Hui Xu,a Long Zhanga and Jin-Pei Chenga,b,
*
aDepartment of Chemistry and State Key Laboratory of Elemento-organic Chemistry, Nankai University, Tianjin 300071, China
bInstitute of Chemistry, Chinese Academy of Sciences, Center for Molecular Science, Beijing 100080, China
Received 3 October 2005; revised 16 November 2005; accepted 20 December 2005
Abstract—Hydroxyl ionic liquid (HIL) has been explored as a novel support for Baylis–Hillman catalyst. The HIL-supported catalyst
showed a better catalytic activity compared to other IL-immobilized catalyst that has no hydroxyl group attached to the IL scaffold. The
hydroxyl group linked on IL played an important role in facilitating efficient catalysis under solvent-free conditions. The corresponding
Baylis–Hillman and aza-Baylis–Hillman adducts were obtained in good to excellent yields in all cases examined. The HIL-supported
quinuclidine can be readily recovered and reused for six times without significant loss of catalytic activity.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
than enolate–aldehyde coupling (step A) as the rate-
determining step in the initial stage of a Baylis–Hillman
reaction,5 is clearly in line with this proposal.
The tertiary amine-catalyzed C–C bond-forming reaction of
aldehydes with activated alkenes is widely referred to as
Morita–Baylis–Hillman reaction, which has become one of
the most popular C–C bond-forming reactions due to its
many advantages regarding to atomic economy, non-metal
catalysis, mild conditions, as well as the promising utility of
the multifunctional products.1 Recent efforts in this area
have been focused largely on developing efficient reaction
systems and exploring asymmetric Baylis–Hillman reac-
tions.1b,2 These studies have shown that the use of protic
solvents or protic additives could accelerate the coupling
reaction quite significantly and provide the presently most
efficient Baylis–Hillman reactions.3,4 In addition, a general
inspection of the successful asymmetric catalysts reveals
that the catalysts all bear a common structure motif with a
‘proton’-donating moiety, which may be viewed as a critical
determinant for chiral transduction.2 Altogether, it seems
likely that the involvement of a protic medium or a protic
additive may be the prerequisite for developing efficient
Baylis–Hillman synthesis and asymmetric catalysis. The
recent finding that a protic additive could mediate
intramolecular proton transfer (Scheme 1, step B) rather
O
O
H
OMe
OMe
R3N
R3N
PhCHO
step A
ROH: Protic additive
H
R
R3N
OH
O
O
H
CO2Me
Ph
Ph
CO2Me
step B
R3N
Scheme 1.
Another important area of current research is the develop-
ment of supported Baylis–Hillman reactions and the
recyclable catalysts.6 This became of our recent focus
because excess or stoichiometric small Lewis base catalysts
have to be used under normal homogeneous B–H
conditions. Progresses in literature along this line now
include the exploration of poly-DMAP and the synthesis of
supported phosphines for Baylis–Hillman catalysis.6a,b In
our previous work, we have developed ionic liquid-
supported quinuclidines for Baylis–Hillman catalysis on
Keywords: Ionic liquid; Quinuclidine; Baylis–Hillman reaction;
Immobilization.
*
Corresponding author. Fax: C86 10 62554449;
e-mail addresses: jpcheng@nankai.edu.cn; chengjp@mail.most.gov.cn
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.12.045

2538
X. Mi et al. / Tetrahedron 62 (2006) 2537–2544
the basis of the biphasic strategy.7 An inspection revealed
that optimal result was achieved when using methanol as
reaction medium, which is similar to the accelerating effect
of protic solvent in other Baylis–Hillman reactions as
highlighted above.3,4 We could then speculate that the use of
a hydroxyl containing ionic liquid alone may exert a similar
accelerating effect. If this is the case, it will offer additional
advantages such as elimination of a protic solvent, reduction
of the catalyst loading, and a more efficient catalyst
recycling for the Baylis–Hillman reactions. This deduction
was considered to be reasonable because it was observed
previously that a hydroxyl group in an amine type catalyst
did exert accelerating effect on some coupling reactions.8,9
In the present work, we explored the use of HILs as
organocatalyst support. To our delight, we indeed observed
a significant beneficial effect of the HIL catalyst over its
non-hydroxyl counterpart. The first example of such type of
Baylis–Hillman catalysts and their successful applications
are hereby reported.
First, catalytic capability of the hydroxyl ionic liquid-
immobilized quinuclidines was tested using 1a as a
representative catalyst and the reaction of p-chlorobenzal-
dehyde and methyl acrylate as a model B–H reaction
(Scheme 4); the results were summarized in Table 1. As
therein revealed, the reactions proceeded quite smoothly in
the presence of 20 mol% of 1a and showed a significant
acceleration effect when conducted in protic solvents. For
example, the reaction in methanol afforded 72% yield of the
desired Baylis–Hillman product in 8 h, while the same
reaction in THF provided only 32% of yield (Table 1, entry
1 vs entry 4).
It was also show that the HIL-supported quinuclidine 1a had
a better catalytic activity than its non-hydroxyl-bearing
counterpart 3a (or 3b), regardless of the reaction media
(Table 1). Considering the similarity of their structures of 1a
and 3, it was envisioned that the hydroxyl group in the ionic
liquid support 1a must be responsible for its higher activity.
It was conceived that the HIL support itself may serve as a
protic additive to promote the Baylis–Hillman reaction in a
manner similar to the protic additives in conventional cases.
Indeed, further optimization of reaction conditions revealed
that the reaction under solvent-free conditions provided the
best result (90% in 8 h, Table 1, entry 11). In sharp contrast,
the same reaction in the presence of a similar amount of
non-HIL 3 was very sluggish and afforded less than 10% of
the B–H product in 8 h under solvent-free conditions
(Table 1, entry 11). These results indicated that the hydroxyl
group linked on the ionic liquid played a critical role in
promoting the reaction.5
The ionic liquid precursor consists of a hydroxyl group and
an amino group at the two ends of each alkyl chain,
respectively, of which the amino group was linked to a
quinuclidine moiety (Scheme 2). The synthesis of the HIL-
supported quinuclidines was straight forward as depicted in
Scheme 3. Reductive amination of 3-quinuclidinone with
the synthetic bifunctional ionic liquid afforded the desired
HIL-bound quinuclidines (1a–1c, 2a–2c) as colorless or
pale yellow viscous liquid. This new type of ionic liquids
was characterized by 1H, 13C NMR, IR and MS.10
ionic liquid
Previously, ionic liquids have been shown in a few cases to
exert positive effect on the Baylis–Hillman reaction rate
when serving as reaction media.6d–i,11 In our study, the use
of hydroxyl containing ionic liquid as reaction media was
found to give moderate accelerating effect in the presence of
non-HIL-supported catalyst 3a (Scheme 5, compared with
Table 1, entries 7–11). Based on these results, further
exploration led to the identification of hydroxyl containing
OH
H2N
catalyst anchor
proton donor
Scheme 2.
Br
NH2 HBr
N
N
NH2
N
N
HO
HO
O
X-
R
R
H
, rt
N
N
N
NaBH4/H2O, rt
N
HO
N
X-
R
1a: R=H, X=Br
1b: R=H, X=BF4
1c: R=H, X=PF6
2a: R=CH3, X=Br
2b: R=CH3, X=BF4
2c: R=CH3, X=PF6
H
N
3a: X= Br
3b: X= BF4
3c: X= PF6
N
N
X-
C4H9
N
Scheme 3.
OH
O
O
1a (20 mol%) or
3 (30 mol%)
CHO
OCH3
OCH3
+
Solvent (0.5 mL)
Cl
Cl
rt, 8 h
Scheme 4.

X. Mi et al. / Tetrahedron 62 (2006) 2537–2544
2539
Table 1. Reactions of p-chlorobenzaldehyde with methyl acrylate in the
presence of ionic liquid-supported quinuclidine 1a (0.2 equiv) or 3
(0.3 equiv) in 0.5 mL of solvent for 8 ha
The immobilized chiral ionic liquids 2 were synthesized
starting from L-alanine. The diastereoisomers obtained were
inseparable and used directly as the B–H catalyst. As
revealed in Table 2, the HIL-supported quinuclidines 2
maintained a comparable activity as 1 (Table 2, entries 6–8),
but with poor enantioselectivity (!10% ee). The low
selectivity may arise from an improper distance between the
nucleophilic center and the protic moiety,2a,b thus the
structure of catalyst should be further optimized.
Entry
Solvent
Catalyst/yield (%)b
1a
3a
1
2
3
4
5
6
7
8
CH3OH
72
63
43
32
48
29
43
47
69
53
90d
37
36
18
8
5
9
C2H5OH
CH3CN
THF
CH2Cl2
CHCl3
[BMIM]BF4
[BMIM]PF6
[bupy]BF4
[BMMIM]BF4
None
13c
13c
21c
15c
!10c,d
The HIL-supported catalysts can be readily recovered after the
reaction using the biphasic strategy (i.e., homogeneous
reaction and heterogeneous separation). The catalyst can
then be reused directly for the following runs. Itwas found that
1a could be recycled for at least six times without significant
loss of its catalytic activity (Table 2, entries 9–13).
9
10
11
a Carried out on 0.5 mmol scale in the presence of 0.2 equiv of the catalyst
1a at room temperature. Molar ratio of p-chlorobenzaldehyde/methyl
acrylateZ1:1.5.
Under the optimized conditions established above, we then
examined the Baylis–Hillman reactions of acrylates with a
variety of aldehydes under solvent-free conditions
(Scheme 7, Table 3). As shown therewith, both aliphatic
and aromatic aldehydes can undergo very efficient Baylis–
Hillman reactions with acrylates in the presence of 20 mol%
of 1a, giving the corresponding Baylis–Hillman adducts 4a–
4s in good to excellent yields (55–98%, 0.5–24 h). These
results demonstrated that the HIL-supported quinuclidines
were excellent Baylis–Hillman catalysts, with activities
comparable to or better than some of the currently best
catalytic systems in the literature.3h For example, the
present system with 20 mol% of 1a could provide fairly
good isolated B–H adduct yields even for the reactions of
inert substrates p-anisaldehyde (55%, Table 3, entry 11). In
comparison, a similar reaction of p-anisaldehyde under the
optimal conditions of the literature in the presence of
100 mol% quinuclidine gave 67% yield after 48 h.3h In
addition, no apparent side reactions of aldehydes (such as
aldol) were observed.
b Isolated yields of pure product.
c Compound 3b was employed instead of 3a.
d Molar ratio of p-chlorobenzaldehyde/methyl acrylateZ1:2.0.
IL-immobilized quinuclidine 1 as highly efficient Baylis–
Hillman catalyst. The observations that the hydroxyl-
bearing ionic liquids as catalyst supports could significantly
promote the B–H reactions may be interpreted by
suggesting a key role of the hydroxyl group on the IL
support in activating aldehyde carbonyl or promoting an
intramolecular proton transfer (or both) in a similar way to
that suggested by Aggarwal.5 However, the detailed
mechanism as to how exactly the hydroxyl group is
functioning in the B–H transition state is still unclear at
the present stage and will need further investigation.
Nevertheless, the finding that the HILs could serve as a
synergistic reagent and support as well as a recoverable
medium for Baylis–Hillman reactions provides a useful clue
for designing highly efficient and recyclable catalysts for
Baylis–Hillman reactions.
Under solvent-free conditions, several other hydroxyl ionic
liquid-supported quinuclidines were also examined
(Scheme 6, Table 2). The HILs with different counter ions
showed little effect on the reactions (Table 2, entries 1–5).
Among all the HILs examined, 1a provided slightly better
results in terms of reaction yields, and thus was selected as
the catalyst for subsequent experiments. Increasing the
loading of the catalyst did not lead to an increase in the
yields (Table 2, entries 2 and 3), probably due to the
formation of byproducts.
The analogous reactions of acrylonitrile with various
aromatic aldehydes were also investigated (Scheme 8,
Table 4). High yields were achieved for all the reactions
in less reaction time. As shown in Table 4, catalyst 1a
was much more effective than 3b in the cases tested.
For example, for the commonly known inert substrate
p-anisaldehyde, the yield of 83% was achieved in 8 h in the
presence of 20 mol% of 1a (Table 4, entry 4). In contrast,
only 69% yield was obtained after 24 h when using 3b
(30 mol%) as the catalyst. For more active aldehydes such
OH
O
O
3a (30 mol%)
CHO
OCH3
OCH3
+
Cl
Cl
N
N
Br-
HO
8 h, 31%
(0.5 mL)
Scheme 5.
OH
O
O
CHO
Catalyst (0.2 equiv)
solvent-free, rt, 8 h
OCH3
OCH3
+
Cl
Cl
Scheme 6.

2540
X. Mi et al. / Tetrahedron 62 (2006) 2537–2544
Table 4. Baylis–Hillman reactions of aldehydes (1.0 equiv) and acryloni-
trile (2.0 equiv) in the presence of 1a (0.2 equiv) at room temperaturea
Table 2. Reactions of p-chlorobenzaldehyde with methyl acrylate in the
presence of various HIL-supported quinuclidinesa
Yield (%)b
Entry
R1
Time (h)
Product
Yield (%)b
Entry
Catalyst
Time (h)
1
2
3
4
5
6
7
8
Ph
p-ClPh
1.5
1.5
4
5a
5b
5c
5d
5e
5f
96
97
91
1
2
3
4
5
6
7
8
1a
1ac
8
8
4
8
8
8
8
8
8
8
8
12
12
90
88
74
86
86
79
84
86
88
88
82
86
84
1ad
1b
p-CH3Ph
p-CH3OPh
p-NO2Ph
2-Pyridyl
3-Pyridyl
4-Pyridyl
8 (24)c
0.5
83 (69)c
85
98
1c
2a
2b
2c
1a (2nd)
1a (3rd)
1a (4th)
1a (5th)
1a (6th)
10 min
10 min
10 min
5g
5h
98
98
9e
10e
11e
12e
13e
a Carried out on 0.5 mmol scale.
b Isolated yields.
c Results using 30 mol% of 3b as catalyst in 2.0 equiv of methanol.
as pyridinecarboxaldehydes, almost all substrates were
quantitatively converted to the desired Baylis–Hillman
products in only 10 min (Table 4, entries 6–8).
a Carried out on 0.5 mmol scale in the presence of 0.2 equiv of the catalyst
(except those specified) at room temperature. Molar ratio of p-
chlorobenzaldehyde/methyl acrylateZ1:2.
b Isolated yields of pure product.
c 0.3 equiv of 1a was used.
The Baylis–Hillman reactions involving the poor Michael
acceptors, cyclic enones, were usually quite sluggish under
normal conditions. Many attempts have been made to
explore efficient catalysts to accelerate this kind of
reaction.7,12 In our previous work, we found that the
reaction of 2-cyclohexenone could undergo quite effective
(fair to high yields) B–H reactions in methanol using ionic
liquid-supported quinuclidine 3b as the catalyst.7 In the
present work, comparable or better results were obtained in
the presence of 20 mol% of 1a (Scheme 9, Table 5). No
apparent aldol byproducts, except for the case of piperanol
(Table 5, entry 5), were observed. In contrast, there were
substantial aldol adducts in most cases when using 3b as the
catalyst.7 Furthermore, the inert p-anisaldehyde could also
undergo quite efficient Baylis–Hillman reaction with the
inert Michael acceptor 2-cyclohexenone, giving 51% yield
in quite a short reaction time (Table 5, entry 4). These
results indicated that the HIL-bound catalyst 1a of the
present work was a very efficient catalyst for the Baylis–
Hillman reactions involving 2-cyclohexenone.
d 0.5 equiv of 1a was used.
e Reuse of catalyst 1a from 1 to 6 run.
OH
R2
Catalyst 1a (0.2 equiv)
R2
R1 CHO +
R1
4a 4r (55 98%)
Scheme 7.
Table 3. Baylis–Hillman reactions of aldehydes (1.0 equiv) and acrylates
(2.0 equiv) catalyzed by ionic liquid-supported quinuclidine 1a (0.2 equiv)
at room temperaturea
Entry
R1
R2
Time (h) Product
Yield
(%)b
1
2
3
4
5
6
7
8
n-C3H7
i-C4H9
n-C6H13
Ph
p-ClPh
COOCH3
COOCH3
COOCH3
COOCH3
COOC2H5
COOC4H9
COOCH3
COOCH3
COOCH3
COOCH3
COOCH3
COOCH3
COOCH3
COOCH3
COOCH3
COOCH3
COOCH3
12
12
12
8
8
8
8
8
8
10
24
1
1
0.5
0.5
0.5
1
4b
4c
4d
4e
4f
4g
4h
4i
4j
4k
4l
4m
4n
4o
4p
4q
4r
79
74
80
94
88
86
91
84
72
70
55
97
94
95
98
97
93
The aza-Baylis–Hillman reaction was also examined
(Scheme 10, Table 6). With 20 mol% of 1a, the aza-
p-ClPh
o-ClPh
m-ClPh
p-CH3Ph
p-iPrPh
p-CH3OPh
o-NO2Ph
m-NO2Ph
2-Pyridyl
3-Pyridyl
4-Pyridyl
2-Furyl
O
OH
O
9
10
11
12
13
14
15
16
17
Catalyst 1a (0.2 equiv)
R1
R1 CHO
+
6a–6e (51–84%)
Scheme 9.
a The reaction was carried out in 0.5 mmol scale.
b Isolated yields.
Table 5. Baylis–Hillman reactions of aldehydes (1.0 equiv) and 2-
cyclohexenone (2.0 equiv) in the presence of 1a (0.2 equiv) at room
temperaturea
Entry
R1
Time (h)
Product
Yield (%)b
1
2
3
4
5
Ph
p-ClPh
p-CH3Ph
p-CH3OPh
Piperonal
8
8
12
16
16
6a
6b
6c
6d
6e
77
84
OH
CN
65
51
Catalyst 1a (0.2 equiv)
CN
R1 CHO
+
R1
73 (9)c
a Carried out on 0.5 mmol scale.
b Isolated yields.
5a–5h (83–97%)
c The number in parentheses is the yield of aldol byproduct.
Scheme 8.

X. Mi et al. / Tetrahedron 62 (2006) 2537–2544
2541
O
O
O
S
O
S
CH3
CH3
N
HN
R2
R2
Catalyst 1a (0.2 equiv)
+
R1
R1
7a–7j (76–97%)
Scheme 10.
Baylis–Hillman reactions of N-sulfonated imines (i.e.,
N-arylmethylidene-4-methylbenzenesulfonamides) with
various B–H substrates such as methyl acrylate and
acrylonitrile were promoted to give exclusively the desired
aza-Baylis–Hillman adducts (76–97%). Even the relatively
inert N-4-methyoxyphenylmethylidene-4-methylbenzene-
sulfonamide could undergo efficient aza-B–H reaction
with methyl acrylate and acrylonitrile in good to excellent
yields (Table 6, entries 5 and 10).
the solution was extracted with diethyl ether. The ether
extract was rotary-evaporated and the crude product was
purified by flash chromatography on silica gel to afford the
desired product. The remaining layer was further vacuumed
to dryness and the resulting catalyst was reused directly
for next run. The reactions using the recycled catalyst were
conducted in the same manner.
2.1.1. Synthesis of HIL-bound quinuclidine 1a (RZH,
XZBr). The synthesis of hydroxyl containing amino ionic
liquid (RZH) followed our previous procedure:7 to a stirred
solution of 2-imidazol-1-yl ethanol (5.6 g, 50 mmol) in
50 mL of dry ethanol was added 3-bromopropylamine
hydrobromide (10.95 g, 50 mmol). The solution was heated
to reflux for 24 h. The ethanol was then removed under
vacuo and the solid residue was dissolved in a minimal
quantity of water and was brought to about pHZ10 by the
addition, in small portions, of solid KOH. The water
solution was then concentrated to dryness and the residue
was extracted with ethanol–THF. The combined extracts
were concentrated to give the desired product (11.33 g,
Table 6. Aza-Baylis–Hillman reactions of aldehydes (1.0 equiv) and
electrophiles (5.0 equiv) in the presence of 1a (0.2 equiv) at room
temperaturea
Entry R1
R2
Time (h) Product
Yield (%)b
1
2
3
4
5
6
7
8
9
10
4-Cl
H
2-Cl
4-CH3
4-CH3O
4-Cl
H
2-Cl
COOCH3
COOCH3
COOCH3
COOCH3
COOCH3
CN
CN
CN
CN
CN
4
2
4
4
8
2
1
2
2
4
7a
7b
7c
7d
7e
7f
7g
7h
7i
91
76
84
88
96
94
90
92
83
97
4-CH3
4-CH3O
1
91%) as a pale yellow viscous liquid. H NMR (CD3OD,
7j
300 MHz): 7.20 (s, 1H), 6.98 (s, 1H), 4.15 (t, 2H, JZ
4.8 Hz), 3.91 (t, 2H, JZ5.1 Hz), 3.83 (t, 2H, JZ5.1 Hz),
2.73–2.69 (m, 2H), 2.08 (t, 2H, JZ7.2 Hz); 13C NMR
(CD3OD, 75 MHz): d 139.2, 124.2, 123.6, 62.5, 60.8, 53.5,
39.1, 33.9; MS for C8H16N3OC (MC), calcd 170.13, found
170.47; BrK (MK), calcd 78.92, found 78.93.
a Carried out on 0.2 mmol scale.
b Isolated yields.
In summary, we have disclosed HILs as novel supports
for Baylis–Hillman catalyst. The HIL-bound quinuclidine
demonstrated better catalytic activity than its non-
hydroxyl ionic liquid analogues and was well applied
to a range of Baylis–Hillman substrates under solvent-
free conditions. This beneficial matrix effect of HIL was
ascribed to the presence of hydroxyl group in the
proximity of active sites, which accelerated the reaction
via hydrogen-bonding activation and/or assisting intra-
molecular proton transfer. The HIL-bound catalysts
possess all the advantages of the non-HIL-bound
catalysts and can be readily recovered from the reaction
mixture and reused for at least six times without
significant loss of catalytic activity. The solvent-free
reaction system provided a highly efficient Baylis–
Hillman synthesis under environmentally friendly
conditions.
To a solution of the synthetic hydroxyl containing amino
imidazolium bromide (2.5 g, 10 mmol) in 20 mL of dry
methanol was added 3-quinuclidinone (1.88 g, 15 mmol).
After stirring for 12 h at ambient temperature, the reaction
was cooled to 0 8C and sodium borohydride (0.855 g,
22.5 mmol) was added in small portions. The solution was
stirred for 4 h, followed by adding small quantity of water.
The resulting mixture was then stirred for additional 30 min
and concentrated to dryness. The residue was extracted with
chloroform and the extracts was concentrated, followed by
extensive washing with diethyl ether to provide the expected
product 1a (3.09 g, 86%) as a pale yellow viscous liquid. IR
(KBr, cmK1): 3397, 2945, 1581, 1457, 1399, 1164, 1075. 1H
NMR (CD3CN, 300 MHz): 7.51 (s, 1H), 7.06 (s, 1H), 6.89
(s, 1H), 4.27 (t, 2H, JZ5.4 Hz), 3.90 (t, 2H, JZ4.5 Hz),
3.17–3.09 (m, 2H), 2.86–2.73 (m, 4H), 2.68–2.60 (m, 3H),
2.45–2.39 (m, 2H), 2.13–2.08 (m, 3H), 1.90–1.85 (m, 2H),
1.76–1.71 (m, 1H), 1.61–1.53 (m, 2H), 1.48–1.40 (m, 2H);
13C NMR (CD3CN, 75 MHz): d 138.8, 124.1, 123.6, 68.1,
62.5, 61.0, 58.1, 56.7, 55.8, 53.3, 44.6, 31.2, 26.7, 25.7,
20.3; MS for C15H27N4OC (MC), calcd 279.22, found
279.61; BrK (MK), calcd 78.92, found 78.98; HRMS for
C15H27N4OC (MC), calcd 279.2185, found 279.2179.
2. Experimental
2.1. General procedure
In a 5 mL vial, 1a (36 mg, 0.1 mmol), aldehyde (0.5 mmol)
and activated alkene (1.0 mmol) were mixed. The resulted
homogeneous solution was stirred at ambient temperature
and monitored by TLC. After the indicated reaction time,

2542
X. Mi et al. / Tetrahedron 62 (2006) 2537–2544
2.1.2. Synthesis of hydroxyl based ionic liquid-bound
quinuclidine 1b (RZH, XZBF4). The sodium tetrafluoro-
borate (2.47 g, 22.5 mmol) was added to a solution of the
synthetic hydroxyl containing amino imidazolium bromide
(RZH, 3.75 g, 15 mmol) in ethanol/THF and stirred for
3 days. The suspension was filtered to remove the
precipitated bromide salt and the organic phase was
concentrated. The residue was then re-dissolved in a small
amount of chloroform, and filtered to remove the inorganic
salt. The solvent was removed under vacuo to afford a pale
yellow liquid (3.12 g, 57%). 1H NMR (CD3OD, 300 MHz):
7.19 (s, 1H), 6.98 (s, 1H), 4.13 (t, 2H, JZ5.1 Hz), 3.91 (t,
2H, JZ4.8 Hz), 3.82 (t, 2H, JZ5.4 Hz), 2.91–2.83 (m, 2H),
2.18 (t, 2H, JZ7.2 Hz); 13C NMR (CD3OD, 75 MHz): d
139.2, 124.3, 123.5, 62.5, 61.0, 53.4, 38.3, 31.6; MS for
C8H16N3OC (MC), calcd 170.13, found 170.40; BF4K
(MK), calcd 87.00, found 87.03.
279.2185, found 279.2179; MS for C15H27N4OC (MC),
calcd 279.22, found 279.55; BFK4 (MK), calcd 144.96,
found 145.15.
2.1.4. Synthesis of hydroxyl based ionic liquid-bound
quinuclidine 2a (RZCH3, XZBr). The synthesis of
hydroxyl containing amino ionic liquid (RZCH3) still
followed our previous procedure: to a stirred solution of 2-
(1-imidazolyl)propanol (8.64 g, 68 mmol) in 68 mL of dry
ethanol was added 3-bromopropylamine hydrobromide
(14.88 g, 68 mmol). The solution was heated to reflux for
24 h. The ethanol was then removed in vacuo and the solid
residue was dissolved in a minimal quantity of water and
brought to about pHZ10 by the addition, in small portions,
of solid KOH. The water solution was then concentrated to
dryness and the residue was extracted with ethanol–THF.
The combined extracts were concentrated to give the
desired product (RZCH3, 17.29 g, 96%) as a pale red
viscous liquid. 1H NMR (CD3OD, 300 MHz): 7.24 (s, 1H),
6.98 (s, 1H), 4.69–4.58 (m, 1H), 4.43–4.38 (m, 2H), 3.78–
3.71 (m, 2H), 3.66–3.59 (m, 2H), 2.26–2.13 (m, 2H), 1.60
(d, 3H, JZ6.9 Hz); 13C NMR (CD3OD, 75 MHz): d 137.7,
123.7, 126.6, 65.6, 60.4, 56.8, 38.5, 31.8, 18.6; MS for
C9H18N3OC (MC), calcd 184.14, found 184.42; BrK (MK),
calcd 78.92, found 78.97.
The synthesis of 1b followed similar procedure as that of
1a. With 3.12 g of imidazolium tetrafluoroborate
(8.5 mmol), 2.57 g of 1b (82%) was obtained as a colorless
viscous liquid. IR (KBr, cmK1): 3396, 2947, 1501, 1396,
1
1164, 1091. H NMR (CD3OD, 300 MHz): 7.19 (s, 1H),
6.98 (s, 1H), 4.35 (t, 2H, JZ4.8 Hz), 3.92 (t, 2H, JZ
4.8 Hz), 3.17–3.10 (m, 2H), 2.82–2.77 (m, 4H), 2.68–2.59
(m, 3H), 2.46–2.44 (m, 2H), 2.14–1.09 (m, 3H), 1.93–1.89
(m, 2H), 1.77–1.72 (m, 1H), 1.61–1.55 (m, 2H), 1.48–1.40
(m, 2H); 13C NMR (CD3OD, 75 MHz): d 138.8, 124.1,
123.6, 68.3, 62.5, 61.0, 58.1, 56.7, 55.9, 53.4, 44.6, 31.2,
26.7, 25.6, 20.3; MS for C15H27N4OC (MC), calcd 279.22,
found 279.57; BFK4 (MK), calcd 87.00, found 87.03; HRMS
for C15H27N4OC (MC), calcd 279.2185, found 279.2177.
To a solution of the synthetic hydroxyl containing amino
imidazolium bromide (RZH, 2.64 g, 10 mmol) in 20 mL of
dry methanol was added 3-quinuclidinone (1.88 g,
15 mmol). After stirring for 12 h at ambient temperature,
the reaction was cooled to 0 8C and sodium borohydride
(0.855 g, 22.5 mmol) was added in small portions. The
solution was stirred for 4 h, followed by adding a small
quantity of water. The resulting mixture was then stirred
for an additional 30 min and concentrated to dryness. The
residue was extracted with chloroform and the extract was
concentrated, followed by extensive washing with diethyl
ether to provide the expected product 2a (3.48 g, 93%) as a
pale yellow viscous liquid. IR (KBr, cmK1): 3137, 2944,
2874, 2312, 1558, 1457, 1400, 1167, 1046. 1H NMR
(CD3OD, 300 MHz): 7.90 (s, 1H), 7.01 (s, 1H), 6.97 (s, 1H),
4.61–4.56 (m, 1H), 4.38–4.33 (m, 2H), 3.89–3.84 (m, 1H),
3.75–3.58 (m, 2H), 3.33 (br, 1H), 3.18–3.11 (m, 1H), 2.80
(br, 3H), 2.66–2.59 (m, 3H), 2.45–2.41 (m, 1H), 2.10 (t, 2H,
JZ6.9 Hz), 1.90 (br, 2H), 1.74 (br, 2H), 1.59 (d, 3H, JZ
6.9 Hz); 13C NMR (CD3OD, 75 MHz): d 136.2, 122.0,
120.8, 67.3, 64.5, 58.9, 58.9, 56.3, 54.9, 48.0, 47.4, 46.4,
28.3, 24.8, 19.0; HRMS for C16H29N4OC (MC), calcd
293.2336, found 293.2339; MS for C16H29N4OC (MC),
calcd 293.23, found 293.49; BrK (MK), calcd 78.92, found
78.93.
2.1.3. Synthesis of hydroxyl based ionic liquid-bound
quinuclidine 1c (RZH, XZPF6). The potassium hexa-
fluorophosphate (2.76 g, 15 mmol) was added to a solution
of the synthetic hydroxyl containing amino imidazolium
bromide (RZH, 2.5 g, 10 mmol) in ethanol/THF and stirred
for 3 days. The suspension was filtered to remove the
precipitated bromide salt and the organic phase was
concentrated. The residue was then re-dissolved in a small
amount of chloroform, and filtered to remove the inorganic
salt. The solvent was removed under vacuo to afford a pale
yellow liquid (2.15 g, 51%). 1H NMR (CD3OD, 300 MHz):
7.19 (s, 1H), 6.98 (s, 1H), 4.12 (t, 2H, JZ5.4 Hz), 3.89 (t,
2H, JZ4.8 Hz), 3.82 (t, 2H, JZ4.8 Hz), 2.91–2.86 (m, 2H),
2.19 (t, 2H, JZ7.2 Hz); 13C NMR (CD3OD, 75 MHz): d
137.7, 124.2, 123.5, 65.6, 60.4, 56.8, 38.5, 31.8, 18.6; MS
for C8H16N3OC (MC), calcd 170.13, found 170.42; PF6K
(MK), calcd 144.96, found 145.13.
The synthesis of 1c followed the similar procedure as that of
1a. With 2.15 g of imidazolium hexafluorophosphate
(5.1 mmol), 2.67 g of 1c (92%) was obtained as a colorless
viscous liquid. IR (KBr, cmK1): 3158, 2946, 1561, 1457,
1398, 1164, 1078, 847. 1H NMR (CD3OD, 300 MHz): 7.18
(s, 1H), 6.98 (s, 1H), 4.33 (t, 2H, JZ5.1 Hz), 3.90 (t, 2H,
JZ5.1 Hz), 3.16–3.10 (m, 2H), 2.85–2.73 (m, 4H), 2.70–
2.60 (m, 2H), 2.45–2.40 (m 2H), 2.12–2.07 (m, 3H),
1.93–1.88 (m, 2H), 1.77–1.73 (m, 1H), 1.61–1.55 (m, 2H),
1.48–1.40 (m, 2H); 13C NMR (CD3OD, 75 MHz): d 138.8,
124.1, 123.6, 68.2, 62.5, 61.0, 58.1, 56.6, 55.8, 53.3, 44.6,
31.2, 26.7, 25.7, 20.3; HRMS for C15H27N4OC (MC), calcd
2.1.5. Synthesis of hydroxyl based ionic liquid-bound
quinuclidine 2b (RZCH3, XZBF4). The sodium tetra-
fluoroborate (3.29 g, 30 mmol) was added to a solution of
the synthetic hydroxyl based amino imidazolium bromide
(RZCH3, 5.28 g, 20 mmol) in ethanol/THF and stirred for
3 days. The suspension was filtered to remove the
precipitated bromide salt and the organic phase was
concentrated. The residue was then re-dissolved in a small
amount of chloroform, and filtered to remove the inorganic
salt. The solvent was removed under vacuo to afford a pale
yellow liquid (2.51 g, 46%). 1H NMR (CD3OD, 300 MHz):

X. Mi et al. / Tetrahedron 62 (2006) 2537–2544
2543
7.79 (s, 1H), 7.61 (s, 1H), 4.69–4.59 (m, 1H), 4.45–4.40 (m,
2H), 3.89–3.85 (m, 2H), 3.66–3.59 (m, 2H), 2.37–2.27 (m,
2H), 1.59 (d, 3H, JZ8.4 Hz); 13C NMR (CD3OD, 75 MHz):
d 137.0, 124.1, 122.7, 65.8, 60.3, 58.4, 37.8, 29.2, 18.5; MS
for C9H18N3OC (MC), calcd 184.14, found 184.37; BF4K
(MK), calcd 87.00, found 87.05.
Science and Technology (MoST) of China for financial
support.
References and notes
The synthesis of 2b followed a similar procedure to that of 2a.
With 2.22 g of imidazolium tetrafluoroborate (8.2 mmol),
1.92 g of 2b (86%) was obtained as a colorless viscous liquid.
IR (KBr, cmK1): 3132, 2942, 2871, 1558, 1457, 1400, 1115,
1045. 1HNMR(CD3OD, 300 MHz):7.22(s, 1H), 6.98(s, 1H),
4.63–4.58 (m, 1H), 4.39–4.34 (m, 2H), 3.89–3.85 (m, 1H),
3.76–3.70 (m, 2H), 3.37 (br, 1H), 3.18–3.10 (m, 1H), 2.79 (br,
3H), 2.68–2.56 (m, 3H), 2.46–2.40 (m, 1H), 2.12 (t, 2H, JZ
6.9 Hz), 1.90 (br, 2H), 1.74 (br, 2H), 1.59 (d, 3H, JZ6.9 Hz);
13C NMR (CD3OD, 75 MHz): d 137.6, 123.6, 122.3, 68.1,
66.8, 60.3, 58.1, 56.7, 55.8, 48.1, 47.5, 47.1, 29.1, 25.3, 19.6;
HRMS for C16H29N4OC (MC), calcd 293.2336, found
293.2336; MS for C16H29N4OC (MC), calcd 293.23, found
293.56; BFK4 (MK), calcd 87.00, found 87.06.
1. For reviews, see: (a) Ciganek, E. Org. React. 1997, 51, 201.
(b) Basavaiah, D.; Rao, A. J.; Satyanarayana, T. Chem. Rev.
2003, 103, 811. (c) Langer, P. Angew. Chem., Int. Ed. 2000,
39, 3049. (d) Basavaiah, D.; Rao, P. D.; Hyma, R. S.
Tetrahedron 1996, 52, 8001. (e) Marko, I. E.; Giles, P. G.;
Hindley, N. J. Tetrahedron 1997, 53, 1015. (f) Drewes, S. E.;
Roos, G. H. P. Tetrahedron 1988, 44, 4653.
2. For recent examples, see: (a) Matsui, K.; Takizawa, S.; Sasai,
H. J. Am. Chem. Soc. 2005, 127, 3680. (b) Shi, M.; Chen,
L. H.; Li, C. Q. J. Am. Chem. Soc. 2005, 127, 3790. (c) Shi, M.;
Xu, Y. Angew. Chem., Int. Ed. 2002, 41, 4507. (d) Iwabuchi,
Y.; Nakatani, M.; Yokoyama, N.; Hatakeyama, S. J. Am.
Chem. Soc. 1999, 121, 10219. (e) McDougal, N. T.; Schaus,
S. E. J. Am. Chem. Soc. 2003, 125, 12094.
2.1.6. Synthesis of hydroxyl based ionic liquid-bound
quinuclidine 2c (RZCH3, XZPF6). The potassium
hexafluorophosphate (4.14 g, 22.5 mmol) was added to a
solution of the synthetic hydroxyl based amino imidazolium
bromide (RZCH3, 3.96 g, 15 mmol) in ethanol/THF and
stirred for 3 days. The suspension was filtered to remove the
precipitated bromide salt and the organic phase was
concentrated. The residue was then re-dissolved in a small
amount of chloroform, and filtered to remove the inorganic
salt. The solvent was removed under vacuo to afford a pale
yellow liquid (2.96 g, 60%). 1H NMR (CD3OD, 300 MHz):
7.24 (s, 1H), 6.98 (s, 1H), 4.60–4.54 (m, 1H), 4.39–4.34 (m,
2H), 3.88–3.84 (m, 2H), 3.62–3.53 (m, 2H), 2.31–2.21 (m,
2H), 1.58 (d, 3H, JZ6.9 Hz); 13C NMR (CD3OD, 75 MHz):
d 137.6, 123.5, 122.5, 65.5, 60.3, 56.8, 38.4, 31.6, 17.9; MS
for C9H18N3OC (MC), calcd 184.14, found 184.34; PF6K
(MK), calcd 144.96, found 145.14.
3. For examples, see: (a) Byun, H. S.; Reddy, K. C.; Bittman, R.
Tetrahedron Lett. 1994, 35, 1371. (b) Auge, J.; Lubin, N.;
Lubineau, A. Tetrahedron Lett. 1994, 35, 7947. (c) Rezgui, F.;
El Gaied, M. M. Tetrahedron Lett. 1998, 39, 5965. (d) Yu, C.;
Liu, B.; Hu, L. J. Org. Chem. 2001, 66, 5413. (e) Yu, C.; Hu,
L. J. Org. Chem. 2002, 67, 219. (f) Cai, J.; Zhou, Z.; Zhao, G.;
Tang, C. Org. Lett. 2002, 4, 4723. (g) Aggarwal, V. K.; Dean,
D. K.; Mereu, A.; Williams, R. J. Org. Chem. 2002, 67, 510.
(h) Aggarwal, V. K.; Emme, I.; Fulford, S. Y. J. Org. Chem.
2003, 68, 692.
4. (a) Luo, S. Z.; Peng, Y. Y.; Zhang, B. L.; Wang, P. G.; Cheng,
J.-P. Curr. Org. Synth. 2004, 1, 405. (b) Luo, S. Z.; Wang,
P. G.; Cheng, J.-P. J. Org. Chem. 2004, 69, 555. (c) Luo, S. Z.;
Mi, X. L.; Xu, H.; Wang, P. G.; Cheng, J.-P. J. Org. Chem.
2004, 69, 8413. (d) Luo, S. Z.; Zhang, B. L.; He, J. Q.;
Janczuk, A.; Wang, P. G.; Cheng, J.-P. Tetrahedron Lett. 2002,
43, 7369. (e) Luo, S. Z.; Mi, X. L.; Wang, P. G.; Cheng, J.-P.
Tetrahedron Lett. 2004, 45, 5171.
The synthesis of 2c followed a similar procedure to that of
2a. With 2.8 g of imidazolium hexafluorophosphate
(8.5 mmol), 2.58 g of 2c (92%) was obtained as a colorless
viscous liquid. IR (KBr, cmK1): 3162, 2946, 2875, 2310,
5. Aggarwal, V. K.; Fulford, S. Y.; Lloyd-Jones, G. C. Angew.
Chem., Int. Ed. 2005, 44, 1706.
6. (a) Huang, J.; Shi, M. Adv. Synth. Catal. 2003, 345, 953. (b)
´
Corma, A.; Garcıa, H.; Leyva, A. Chem. Commun. 2003, 2806.
1
1558, 1457, 1396, 1166, 1052, 845. H NMR (CD3OD,
(c) Zhao, L. J.; He, H. S.; Shi, M.; Toy, P. H. J. Comb. Chem.
2004, 6, 680. (d) Rosa, J. N.; Afonso, C. A. M.; Santos, A. G.
Tetrahedron Lett. 2001, 57, 4189. (e) Aggarwal, V. K.; Emme,
I.; Mereu, A. Chem. Commun. 2002, 1612. (f) Hsu, J.-C.; Yen,
Y. H.; Chu, Y. H. Tetrahedron Lett. 2004, 45, 4673. (g) Kim,
E. J.; Ko, S. Y.; Song, C. E. Helv. Chim. Acta 2003, 86, 894.
(h) Johnson, C. L.; Donkor, R. E.; Nawaz, W.; Karodia, N.
Tetrahedron Lett. 2004, 45, 7359. (i) Kumar, A.; Pawar, S. S.
J. Mol. Catal. A: Chem. 2004, 211, 43.
300 MHz): 7.22 (s, 1H), 6.97 (s, 1H), 4.60–4.54 (m, 1H),
4.36–4.31 (m, 2H), 3.87–3.83 (m, 1H), 3.75–3.69 (m, 2H),
3.37 (br, 1H), 3.17–3.13 (m, 1H), 2.78 (br, 3H), 2.67–2.54
(m, 3H), 2.44–2.40 (m, 1H), 2.10 (t, 2H, JZ6.9 Hz), 1.89
(br, 2H), 1.76 (br, 2H), 1.58 (d, 3H, JZ6.9 Hz); 13C NMR
(CD3OD, 75 MHz): d 137.6, 123.6, 122.3, 68.2, 66.5, 60.3,
58.1, 56.6, 55.7, 47.5, 47.0, 44.6, 29.0, 26.7, 20.3; MS for
C16H29N4OC (MC), calcd 293.23, found 293.58; BrK
(MK), calcd 144.96, found 145.13.
7. Mi, X. L.; Luo, S. Z.; Cheng, J.-P. J. Org. Chem. 2005, 70,
2338.
All the Baylis–Hillman products are known
compounds.4,7,13
8. Drewes, S. E.; Freese, S. D.; Emslie, N. D.; Roos, G. H. P.
Synth. Commun. 1988, 18, 1565.
´
9. (a) Marko, I. E.; Giles, P. R.; Hindley, N. J. Tetrahedron 1997,
´
53, 1015. (b) Bailey, M.; Marko, I. E.; Ollis, D.; Rasmussen,
P. R. Tetrahedron Lett. 1990, 31, 4509.
Acknowledgements
10. Due to the high polarity and basic character of the hydroxyl ionic
liquids, unidentified peaks were still observed in the NMR
spectra after the general work-up procedures. The obtained
We are grateful to the Natural Science Foundation of China
(NSFC 20452001 and 20421202) and the Ministry of

2544
X. Mi et al. / Tetrahedron 62 (2006) 2537–2544
catalysts from different batches gave similar characterization
data and consistent experimental results.
Tetrahedron Lett. 2000, 41, 2165. (e) Pei, W.; Wei, H.-X.;
Li, G. G. Chem. Commun. 2002, 2412. (f) Basavaiah, D.;
Sreenivasulu, B.; Rao, A. J. J. Org. Chem. 2003, 68, 5983.
(g) Patra, A.; Batra, S.; Joshi, B. S.; Roy, R.; Kundu, B.;
Bhaduri, A. P. J. Org. Chem. 2002, 67, 5783.
´
11. Pegot, B.; Vo-Thanh, G.; Gori, D.; Loupy, A. Tetrahedron
Lett. 2004, 45, 6425.
12. (a) Li, G.-G.; Wei, H.-X.; Gao, J.-J.; Caputo, T. D.
Tetrahedron Lett. 2000, 41, 1. (b) Sugahara, T.;
Ogasawara, K. Synlett 1999, 419. (c) Kataoka, T.; Iwama,
T.; Tsujiyama, S.; Iwama, T.; Watanabe, S. Tetrahedron
1998, 54, 11813. (d) Yamada, Y. M. A.; Ikegami, S.
13. (a) Shi, M.; Xu, Y.-M.; Shi, Y.-L. Chem. Eur. J. 2005, 11,
1794. (b) Balan, D.; Adolfsson, H. J. Org. Chem. 2002, 67,
2329. (c) Balan, D.; Adolfsson, H. J. Org. Chem. 2001,
66, 6498.