Recyclable copper catalysts based on imidazolium-tagged C 2-symmetric bis(oxazoline) and their application in D-A reactions in ionic liquids

Article abstract of DOI:10.1039/c1gc15788d

Functional imidazolium ionic liquids have been developed as a new class of versatile catalysts. C₂-symmetric and unsymmetric imidazolium-tagged bis(oxazoline) ligands were prepared, and the anions of the ligands were altered by ion-exchange reactions. The catalysts based on the new ligands and Cu(OTf)₂ were applied in asymmetric Diels-Alder reactions between N-acryloyl/N-crotonoyloxazolidinones 15 and 1,3-cyclohexandiene/cyclopentadiene 16 in different ionic liquids and in dichloromethane (DCM). The catalysts achieved a high level of activity and enantioselectivity, as well as good recyclability: cycloadduct (S)-17ab was attained at 98% conversion and 97% ee in [Bmim]NTf₂. Moreover, the catalyst could be recycled 20 times without an obvious loss of activity or enantioselectivity. By comparison, we deduced that the C₂ symmetry of the new ligands was crucial for obtaining high ee values. Toxicity studies of the ligands were performed for the first time.

Full text of DOI:10.1039/c1gc15788d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2011, 13, 2963
www.rsc.org/greenchem
PAPER
Recyclable copper catalysts based on imidazolium-tagged C2-symmetric
bis(oxazoline) and their application in D–A reactions in ionic liquids†
Zhi-Ming Zhou,*a,b Zhi-Huai Li,a Xiao-Yan Hao,a Xiao Dong,a Xin Li,a Li Dai,a Ying-Qiang Liu,a
Jun Zhang,a Hai-feng Huang,a Xia Lia and Jin-liang Wanga
Received 4th July 2011, Accepted 9th August 2011
DOI: 10.1039/c1gc15788d
Functional imidazolium ionic liquids have been developed as a new class of versatile catalysts.
C2-symmetric and unsymmetric imidazolium-tagged bis(oxazoline) ligands were prepared, and
the anions of the ligands were altered by ion-exchange reactions. The catalysts based on the new
ligands and Cu(OTf)2 were applied in asymmetric Diels–Alder reactions between N-acryloyl/
N-crotonoyloxazolidinones 15 and 1,3-cyclohexandiene/cyclopentadiene 16 in different ionic
liquids and in dichloromethane (DCM). The catalysts achieved a high level of activity and
enantioselectivity, as well as good recyclability: cycloadduct (S)-17ab was attained at 98%
conversion and 97% ee in [Bmim]NTf2. Moreover, the catalyst could be recycled 20 times without
an obvious loss of activity or enantioselectivity. By comparison, we deduced that the C2 symmetry
of the new ligands was crucial for obtaining high ee values. Toxicity studies of the ligands were
performed for the first time.
often dramatically decreases after several runs. Such decreases
Introduction
are usually caused by the catalysts leaching during product
During the past several decades, chiral bis(oxazolines) (box),
extraction from the reaction system or by the catalysts abating
such as 1, have proven to be very effective ligands that afford
high levels of activity and enantioselectivity in many reactions.1–8
Because most catalysts are expensive and because the metal
as well as the ligands are highly toxic to the environment
and humans, the recycling of catalysts now assumes an even
greater importance. A number of approaches for preventing
the expensive chiral metal complex from running off during
the recovery and/or recycling process have been investigated.9
Among them, the use of ionic liquids as a biphasic system is a
very promising approach.10 The properties of ionic liquids, such
as low volatility, immiscibility with non-polar organic solvents
and stability, make them excellent candidates for solvents in
green processes.11–14 In addition, ionic liquids can improve
catalyst stability, and some unexpected transformations that
would be impossible under ordinary reaction conditions have
been observed in ionic liquids.15,16 However, even if the catalyst
dissolves well in an ionic liquid, the activity of the catalyst
during the purification of the ionic phase before recycling.17
The limitations of a catalyst or ligand in ionic liquids may
be minimised by the presence of an ion tag on the frame of
the ligand. The ionic nature could ideally render the catalyst
or ligand insoluble in non-polar solvents so that the product
could be extracted to leave a recyclable phase that contains
the catalyst.17,18 The ion-tagging strategy has attracted more
attention in recent years.
The asymmetric Diels–Alder (D–A) reaction is one of the
representative reactions that are catalysed by chiral box–metal
complexes, and several previous attempts have been made to
immobilise catalysts with ionic liquids in this reaction. Meracz
and Oh19 have reported that ionic liquids can enhance the
enantioselectivity and the endo : exo ratio of the D–A reac-
tion between N-crotonoyloxazolidinone and cyclopentadiene.
Compared to the ee obtained in dichloromethane (DCM),
the ee obtained in 1,3-dibutylimidazolium tetrafluoroborate
[DiBuIm]BF4 increased from 52% to 92%, and the endo : exo
ratio increased from 79 : 21 to 97 : 3. Chang-Eun Yeom and
Hye Won Kim16 have reported that the reactivity and selectivity
were highly dependent upon the properties of the ionic liquids.
Because of the increased activity of their catalyst; the amount of
ligand–metal complex could be reduced to 0.6 mol% without
a significant compromise in the selectivity and recycled 18
times, with a decrease of ee in the 6th run. Imidazolium-tagged
bis(oxazoline) 2 was prepared and used as a chiral ligand in the
aR&D Center for Pharmaceuticals, School of Chemical Engineering and
the Environment, Beijing Institute of Technology, Beijing, PR China,
100081. E-mail: zzm@bit.edu.cn; Fax: +86 010 68912664; Tel: +86 010
68912664
bState Key Laboratory of Explosion Science and Technology, Beijing
Institute of Technology, Beijing, PR China, 100081
† Electronic supplementary information (ESI) available: Preparation of
ligands and theoretical calculations results of 10b–Cu–15a. See DOI:
10.1039/c1gc15788d
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2963–2971 | 2963
©

View Article Online
copper-catalysed D–A reaction between N-acryloyl/N-
crotonoyloxazolidinones and 1,3-cyclohexandiene in the
ionic liquid 1-ethyl-3-methylimidazolium bis[(trifluoromethyl)-
sulfonyl]-imide [Emim]NTf2.20 Complete conversion and ena-
tioselectivities of up to 86% ee were obtained for the reaction
between N-acryloyloxazolidinone and 1,3-cyclohexandiene in
[Emim]NTf2. In this case, the catalyst could be recycled 10 times
without loss of activity or enantioselectivity. An ee of up to 95%
was obtained for the reaction between N-acryloyloxazolidinone
and cyclopentadiene, although the recyclability of the catalyst
in this case was not discussed. This report revealed that
the introduction of an imidazolium tag into bis(oxazoline)
significantly improved the capability to recover and reuse the
catalyst for reactions performed in ionic liquids.
Results and Discussion
Preparation of C2-symmetric/unsymmetric imidazolium-tagged
bis(oxazoline)
Deprotonation of 3 in THF with 1.2 eq of BuLi was performed
at 0 ◦C. The addition of the resultant solution to an excess of 4-
iodobutoxy-(tert-butyl)dimethylsilane 4 in THF afforded the de-
sired (tert-butyl)dimethylsilane-butyl-substituted bis(oxazoline)
5. Hydrolysis of 5 with TBAF in THF gave the hydroxyl-
substituted box 6, which was converted to tosyl-substituted box
7 by TsCl in CH2Cl2. The imidazolium tag was introduced in
the final step by the reaction of 7 with 1,2-dimethylimidazole in
DMF at 70 ◦C, which afforded the tosyl salts 10. C2-symmetric
box 9 was obtained when 7a was reacted with imidazole in
DMF catalysed by Cs2CO3. Box 9 reacted with methyl iodide
in Et2O to give iodide salt 10d. The hydroxyl-substituted box 6a
was chloridised using PPh3 in CCl4 as the reagent. Chloride
salt 10g was obtained by treating the product 8 with 1,2-
dimethylimidazole in Et2O. The hexafluorophosphoric salts 10e
and f were synthesised from the tosyl salts through ion exchange
of 10a and b (see Scheme 1). The preparation of C2-unsymmetric
imidazolium-tagged chiral bis(oxazoline) ligand 14 was similar
to that of the C2-symmetric box, except the amount of BuLi and
(4-iodobutoxy)(tert-butyl)dimethylsilane 4 in the deprotonation
step wa◦s changed to 0.5 eq, and the temperature was maintained
at -78 C (see Scheme 2). 3b was converted to 18 by (CH2O)n
and Et3N. 19 was afforded once 18 reacted with Tf2O in the
presence of 1,2-dimethylimidazole as a base at -78 ◦C, after that
the imidazolium tag was introduced by the reaction of 19 with
1,2-dimethylimidazole in CH3CN. 20b was synthesised through
ion exchange of 20a (see Scheme 3).
Among the different immobilisation approaches, the preser-
vation of the C2 symmetry of the ligand has been regarded as
a desirable feature, which was thought to be crucial for high
ee values.21–24 However, Cozzi25 and Mandoli’s work26 clearly
demonstrated that boxes with C2 symmetry are not necessary
for high enantioselectivity in the cyclopropanation of differ-
ent olefin substrates and other copper-catalysed asymmetric
transformations with insoluble polymer-bound-approach box
ligands. The enantioselectivity actually matches the variations
in the steric hindrance at the bridge position but not those
in ligand symmetry. Whether C2-symmetric features affect the
ionic-tagged box ligands has not been investigated until now.
Our team has synthesized a series of novel quarternary ammo-
nium salt-modified chiral ferrocenylphosphine-imine ligands,
and their applicability in asymmetric C–C and C–N bond for-
mation was demonstrated. High enantioselectivity was obtained
in the Pd-catalyzed asymmetric substitution of 1,3-diphenyl-
2-propenyl acetate, with dimethyl malonate (up to 94.6% ee)
and benzylamine (up to 92.6% ee).27 Herein, we designed and
prepared the imidazolium-tagged bis(oxazoline); for compar-
ison, both C2-symmetric and C2-unsymmetric ligands were
included. Their performance in the copper-catalysed asymmetric
D–A reaction between N-acryloyl/N-crotonoyloxazolidinones
and 1,3-cyclohexandiene/cyclopentadiene in different sol-
vents were evaluated and compared. When C2-symmetric
10b was used as a ligand, the D–A reaction between N-
acryloyloxazolidinones and 1,3-cyclohexandiene in the ionic liq-
uid 1-butyl-3-methylimidazolium bis[(trifluoromethyl)sulfonyl]-
imide [Bmim]NTf2 yielded a 98% conversion with a 90 : 10
endo:exo ratio and 97% ee. More importantly, the catalyst in the
reaction system could be recycled as many as 20 times without
obvious loss of activity and enantioselectivity.
Asymmetric Diels–Alder reactions
We began our study by evaluating the newly synthe-
sised ligands in an asymmetric D–A reaction between N-
acryloyloxazolidinone 15a and cyclohexa-1,3-diene 16b. The
catalysts were typically prepared in MeOH by stirring Cu(OTf)2
(10 mol%) and the appropriate bis(oxazoline) (11 mol%) at room
temperature for 1 h. After the metal–ligand complex formed,
MeOH was removed under vacuum. The solvents (ionic liquids
or dried DCM) were added to the system, followed by the
addition of the substrates. The resultant mixture was stirred
for the required time at room temperature.
Table 1 summarises the results of a comparative study
of imidazolium-tagged bis(oxazoline) 10a–g, 20a–b with dif-
ferent anions, C2-unsymmetric imidazolium-tagged chiral
bis(oxazoline) ligand 14 and unfunctionalised (S,S)-t-Bu-box
3b, 7b as ligands for the Diels–Alder reaction between 15a and
16b in [Bmim]NTf2. The tert-butyl-substituted bis(oxazoline)
was the optimum ligand and yielded significantly higher ees than
its phenyl-substituted and diphenyl-substituted counterparts in
most cases (Table 1, entries 1 and 2, entries 4, 7 and 10). The
catalyst derived from 10b yielded cycloadduct (2S)-17ab at 98%
conversion and 97% ee in [Bmim]NTf2 (Table 1, entry 4), whereas
the catalyst derived from 10a yielded only 80% ee (Table 1, entry
4). The same reaction catalysed by the complex based on 2 could
achieve 77% ee in [Emim]NTf2.20 The ideal performance of the
catalyst based on 10b was also expected because of the longer
2964 | Green Chem., 2011, 13, 2963–2971
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Scheme 1 The synthesis of C2 symmetric imidazolium-tagged bis(oxazoline).
alkyl chains on the carbon bridge between the two oxazoline
moieties. For ligands with shorter alkyl chains, the results of the
D–A reaction were decreased (Table 1, entries 4 and 8, entries
11 and 12).
in [BMIm]BF4 and [BMIm]PF6, results in these two ionic liquids
obviously contrast, this was proposed to be caused by differences
in the electronic properties of the anion parts. We presume the
anion parts of the ligand affect the results through steric bulk
and the guess could be demonstrated below in a theoretical
mechanistic study.
A reaction without catalyst in [Bmim]SbF6 has been carried
out by Yeom and Kim,16 the reaction yielded a 91 : 9 endo : exo
mixture of the racemic adduct in 15% yield. We presume that
the high ee values were caused by the preservation of the C2-
symmetry feature of the ligands, which was consistent with
Burguete’s report.21
The anions of the ligands affect the enantioselectivity
and activity. Tosyl-anion ligands 10a and 10b yielded better
enantioselectivities and activities than hexafluorophosphoric,
bromo, chloride, or iodide salts (Table 1, entry 3–9). The C2-
symmetric copper complex of bis(oxazoline) 10b yielded better
enantioselectivity and activity (entry 4, 98% conversion and
97% ee) than the C2-unsymmetric catalyst based on 14 (entry
10, 82% conversion and 83% ee). The endo : exo ratio did not
vary significantly (Table 1, entry 3–9). Qian31 has reported a
chiral anion modified ionic liquid derived from L-proline as an
asymmetric organocatalyst for direct asymmetric aldol reactions
The performance of the copper complexes based on the
[OTs]- salt 10b and unfunctionalised (S,S)-t-Bu-box 3b as
Lewis-acid catalysts for the Diels–Alder reaction between
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2963–2971 | 2965
©

View Article Online
Scheme
3
The synthesis of C2-symmetric imidazolium-tagged
bis(oxazoline) (short line).
Table 1 Screening of the ligands in an asymmetric Diels–Alder
reactiona
Scheme
2 The synthesis of C2-unsymmetric imidazolium-tagged
bis(oxazoline).
N-acryloyloxazolidinone 15a and cyclohexa-1,3-diene 16b in
[Bmim]NTf2 or DCM were studied; the results are shown in
Table 2. The ionic-tagged ligand showed almost no reactivity
in DCM. The reaction only occurred in hydrophobic ionic
liquids, such as [Bmim]PF6 and [Bmim]NTf2, whereas no
reaction was observed in hydrophilic ionic liquids, such as
[Bmim]BF4, and polar solvents, such as MeOH. This result
was consistent with previous reports in the literature.20 An
enhancement in both the rate and enantioselectivity was also
obtained with a catalyst based on (S,S)-t-Bu-box in [Bmim]NTf2
compared with the same catalyst in DCM. The enhancement
in both the rate and the enantioselectivity may result from
an improvement of the stability of the catalyst by ionic liq-
uids as well as from unexpected transformations that occur
in the ionic liquid–catalyst system that would be impossible
in DCM.
Entry
Ligand
Conversion [%]
endo:exob
ee [%]b
1
2
3
4
5
6
7
8
3b
7b
82
74
97
98
62
54
96
98
63
82
74
61
82 : 18
88 : 12
85 : 15
90 : 10
80 : 20
85 : 15
85 : 15
90 : 10
85 : 15
83 : 17
88 : 12
90 : 10
84
91
80
97
32
40
47
59
43
83
43
18
10a
10b
10c
10d
10e
10f
10g
14
9
10
11
12
20a
20b
a The ratio of 15a/16b/ligand/Cu(OTf)2 is 1.0/3.0/0.11/0.1, and the
reactions were carried out for 48 h. b Enantiomeric excess and the
endo : exo ratio were determined by HPLC (chiralpak AD-H column,
hexane/iPrOH: 98 : 2, 1.0 mL min-1).
Various substrates were subsequently tested using the C2-
symmetric imidazolium-tagged chiral bis(oxazoline) 10b as the
ligand. The results are summarised in Table 3. The oxazolidin-2-
one substitutes were more active than those based on pyrrolidin-
2-one (Table 3, entry 1–5, entry 6–10). Cyclopentadiene was the
most reactive diene: 94% conversion and 90% ee were obtained
within 5 min of the addition of N-acryloyloxazolidinone and
cyclopentadiene, whereas the same result after the addition of
N-acryloyloxazolidinone and cyclohexa-1,3-diene required 48 h
(Table 3, entry 2). Almost no reaction occurred after the addition
of N-acryloyloxazolidinone and isoprene.
Recyclability of the catalysts
To evaluate the recyclability of the catalysts generated from our
new ligand 10b, we performed the asymmetric D–A reaction
between N-acryloyloxazolidinone 15a and cyclohexa-1,3-diene
2966 | Green Chem., 2011, 13, 2963–2971
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 2 Screening of solvents in the asymmetric D–A reaction of N-
acryloyzl-l,3-oxazolidine-2-one 15a and cyclohexa-1,3-diene 16ba
Entry Ligand Solvent
Conversion [%] endo : exob ee [%]b
Fig. 1 Variation in percentage conversion (hatched), percentage ee
(reticle) and percentage endo (blank) upon recycling the Diels–Alder
reactions between 15a and cyclohexa-1,3-diene 16b in [Bmim]NTf2 using
catalysts generated from 10b and Cu(OTf)2.
1
2
3
4
5
6
7
8
10b
10b
10b
10b
10b
10b
3b
CH2Cl2
0
95
0
—
88 : 12
—
—
90 : 10
—
—
80
—
—
97
—
80
84
[Bmim]PF6
[Bmim]BF4
[Bmim]BF4
[Bmim]NTf2 98
MeOH
CH2Cl2
0
0
70
79 : 21
82 : 18
3b
[Bmim]NTf2 82
a The ratio of 15a/16b/ligand/Cu(OTf)2 is 1.0/3.0/0.11/0.1.
b Enantiomeric excess and the endo : exo ratio were determined by HPLC
(chiralpak AD-H column, hexane/iPrOH: 98 : 2, 1.0 mL min-1).
Table 3 Asymmetric D–A reaction of various substrates using the
catalyst derived from ligand 10ba
Entry Product Time
Conversion [%] endo : exob
ee [%]b
1
2
3
4
5
6
7
8
9
17aa
17ab
17ac
17ba
17bb
17ca
17cb
17cc
17da
17db
5 min
48 h
24 h
60 min
48 h
48 h
48 h
48 h
48 h
48 h
94
98
0
99
73
23
9
8
0
0
88 : 12
90 : 10
—
93 : 7
93 : 7
89 : 11
89 : 11
89 : 11
—
90
97
—
96
77
17
10
9
Fig. 2 Variation in percentage conversion (1), percentage ee (2) and
percentage endo (3) upon recycling the Diels–Alder reactions between
15a and cyclohexa-1,3-diene 16b or cyclopentadiene 16a in [Bmim]NTf2
using catalysts generated from Cu(OTf)2 and 10b (a), 14 (b), and 3b (c),
respectively.
—
—
10
—
a The ratio of 15a/16b/ligand/Cu(OTf)2 is 1.0/3.0/0.11/0.1.
b Enantiomeric excess and endo : exo ratio were determined by HPLC
(chiralpak AD-H, Chiralcel OD-H, Chiralcel OJ-H, hexane/iPrOH).
the enantioselectivity or conversion. The catalyst based on
the C2-unsymmetric imidazolium tagged box 14 could also be
subjected to 20 cycles with little loss of reaction activity and
enantioselectivity, although the catalyst produced a relatively
lower ee value (from 83% to 78%) and conversion (82% to
66%). In contrast, the catalyst generated from the neutral ligand
3b could be reused only 7 times; the reaction activity and
enantioselectivity decreased sharply during the 8th reaction
using the ionic liquid. This result was consistent with the results
of Chang-Eun Yeom.16 it seems that the endo : exo ratio could
not be significantly affected by the cycle time. The loss of
activity and enantioselectivity during recycling with the catalyst
based on 3b and the maintenance activity and enantioselectivity
for catalysts derived from 10b and 14 demonstrated that the
introduction of an ionic tag to the ligand assures that the catalyst
is insoluble in non-polar solvents. Consequently, the product
16b in [Bmim]NTf2. After extraction with diethyl ether, the ionic-
liquid-containing catalyst was subjected to vacuum to remove
traces of diethyl ether, then flushed with inert gas and charged
with further portions of 15a and 16b at room temperature. The
activity and enantioselectivity were maintained even after the
catalyst was reused 20 times (Fig. 1, conversion 83%, ee 93% on
the 20th cycle).
The recyclability of the catalyst based on the C2-symmetric
imidazolium-tagged box 10b, C2-unsymmetric box 14 and
traditional ligand 3b were studied and compared; the results
are summarised in Fig. 2. The imidazolium-tagged catalyst
derived from C2-symmetric imidazolium-tagged box 10b could
be used at least 20 times without a significant decrease in
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2963–2971 | 2967
©

View Article Online
Table 4 Toxicity studies of the ligands
could be extracted, which left the catalyst in the recyclable phase.
To prove that the catalyst containing imidazolium-tagged box
could maintain better activity in [Bmim]NTf2, we decreased the
amount of catalyst to 5% and even to 1 mol%. The conversion of
the Diels–Alder reactions between 15a and cyclopentadiene 16a
exhibited a slight decrease (91% and 82%, respectively) after 48 h.
Thus, the C2-symmetric imidazolium-tagged box could be well
maintained andreusedefficientlyin the ionic liquid [Bmim]NTf2.
More importantly, the recyclability of the catalyst based on
ligand 10b was the best catalyst in asymmetric D–A reactions
reported thus far; both the activity and the enantioselectivity
were well maintained even after being recycled 20 times.
Toxicity studies of the ligands
The luminescent-bacteria toxicity test (LBT) was developed as
an alternative to animal testing28 and has become a basic test for
ecotoxicological testing of chemicals, waste water and eluates
from soil and sediment.29 Herein, we employed this method
to test the toxicity of ligands. The LC50 of the ligands was
studied, and the results are compared in Table 4. We presume
that the variation in the toxicity is caused mainly by variations
of the ligands’ anion component. The ligands were more toxic
than traditional organic ligand 1 when the anion was [OTs]-
(Table 4, entries 1,7,9 and 10), whereas they were less toxic
than (or comparable with) 1 when the anion was [PF6]- or
[I]- (Table 4, entries 1,7,11–13). However, considering the good
recyclability of the catalyst based on 10b, our ligands were more
environmentally friendly.
Theoretical mechanistic study
A theoretical mechanistic study was conducted to explain
the origin of the enantioselectivity at a molecular level. The
enantioselectivity-determining step of the Diels–Alder reaction
catalysed by bis(oxazoline)–Cu(II) 10b–Cu–15a was calculated
using the B3LYP/6-31G(d) scheme in the Gaussian03 software
package (Fig. 3). The calculation of the presumed catalyst–
substrate complex suggested a degree of distortion for the bound
dienophile, which was similar to the result calculated by Evans.30
A more marked difference in the steric environment about the
two prochiral faces, as compared to the stereochemical model
calculated by Johnson and Evans,30 was noted. The results
are consistent with the high level of observed enantioselection.
Because of the imidazolium salt tagged to the long alkyl chains
on the carbon bridge between the two oxazoline moieties and
especially because of the bulky [OTs]- anion section, the steric
environment was enhanced once the diene attacked from the a-
Si face. (Since one of the long alkyl chains on the carbon bridge
was more curved than the other one, the imidazolium salt tagged
to it and the bulky [OTs]- anion section were close to the Cu(II),
which increased the steric block and made it more difficult for
the diene to attack from the a-Si face.) This result provided
additional evidence for the square-planar model. In addition, we
conjectured that after the number of alkyl-chain carbon atoms
on the carbon bridge changed from 4 to 2, the imidazolium salt
and the [OTf]- anion would not enhance the steric environment
in the transition state because the cation/anion section could
not approach the Cu(II) but would hinder the diene attacking
from the a-Re face and result in a decrease in the ee value.
Entry
Ligand
LC50 [mg mL-1]
1
2
3
4
5
6
7
8
1
45
68
76
3a
3b
a
3c
—
7b
55
—
33
—
11
11.1
75
100
70
7c
9
10a
10b
10c
10d
10e
10f
14
9
10
11
12
13
14
—
a the LC50 could not be detected.
Conclusions
In conclusion, C2-symmetric and unsymmetric imidazolium-
tagged bis(oxazoline) ligands were successfully and conveniently
prepared, and the anions of the ligands could be altered by
ion exchange. The catalysts based on the new ligands and
Cu(OTf)2 were applied to asymmetric Diels–Alder reactions
2968 | Green Chem., 2011, 13, 2963–2971
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Fig. 3 B3LYP/6-31G(d) optimised geometry of the 10b–Cu–15a complex model for enantioselective cycloadditions. Hydrogen atoms were omitted
for clarity.
between N-acryloyl/N-crotonoyloxazolidinones 15 and 1,3-
cyclohexandiene/cyclopentadiene 16 in different ionic liquids
and DCM. The catalyst derived from 10b yielded cycloadduct
(S)-17ab at 98% conversion and 97% ee in [Bmim]NTf2.
Furthermore, the catalyst based on 10b could be recycled at least
20 times without an obvious loss of activity or enantioselectivity.
Compared to previously reported reaction systems, the perfor-
mance of the new catalyst in hydrophobic ionic liquids, especially
in [Bmim]NTf2, achieved our initial design goal: both a high
level of enantioselectivity and good recyclability were attained.
At the same time, our study showed that the C2-symmetry of
the new ligands was crucial for achieving high ee values. Our
process, which was simple and easy to operate, possesses vast
potential for environmentally friendly processes in the chemical
industry. Further research on C2-symmetric imidazolium-tagged
bis(oxazoline) ligands and their performance in asymmetric
reactions is ongoing in our laboratory.
and 13C NMR spectra were recorded on a Bruker AMX 300
instrument. Analytical high-performance liquid chromatogra-
phy (HPLC) was performed on an Knauer Smartline series
HPLC equipped with a variable wavelength detector and Daicel
Chiralpak AD-H, Chiralcel OD-H and Chiralcel OJ-H columns.
Enantiomeric excess was calculated from the HPLC profile.
Synthesis of 5,5-bis[(R)-4,5-dihydro-4-phenyloxazol-2-
yl]nonane-1,9-di(tert-butyl)dimethylsilane 5a
A solution of 3a (1.000 g, 3.25 mmol) in 20 mL of dry THF
was cooled to 0 ◦C under a N2 atmosphere. A solution of n-
BuLi (4.48 mL, 2.9 M in hexane, 6.50 mmol) was added to
the solution dropwise, the resulting mixture was stirred at 0 ◦C
for 1 h, and then 4 (3.680 g, 11.70 mmol) was added slowly to
the mixture. After the addition was complete, the ice bath was
removed, and the solution was left to warm to room temperature
and stirred for an additional 12 h. After this time, the reaction
mixture was quenched by the addition of water (15 mL) and
extracted with DCM (3 ¥ 15 mL). The organic fractions were
combined, washed with brine (1 ¥ 40 mL) and H2O (1 ¥
40 mL), and dried over Na2SO4. The solvent was removed
under vacuum to afford a thick, dark, oily residue, which was
purified by column chromatography (SiO2, PE/EtOAc, 10/1)
Experimental section
General Procedures
All manipulations involving air-sensitive materials were per-
formed using standard Schlenk-line techniques under an atmo-
sphere of nitrogen or argon in oven-dried glassware. DCM was
distilled from calcium hydride, and methanol was distilled from
Mg. Copper(II) triflate and the ionic liquid were purchased from
commercial suppliers and used without further purification. 1H
to afford 5a as a pale yellow oil; yield: 1.130 g (70%). [a]D20
+90.5 (c = 0.50, CH2Cl2). H NMR(400 MHz, CDCl3): d =
7.34–7.25 (m, 10H, oxazoline-ph), 5.25–5.20 (dd, J = 8.4 Hz and
1.6 Hz, 2H, oxazoline-CHaHbO), 4.67–4.62 (dd, J = 8.4 Hz and
=
1
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2963–2971 | 2969
©

View Article Online
1.6 Hz, 2H, oxazoline-CHaHbO), 4.12–4.08 (t, J = 8.4 Hz, 2H,
oxazoline-CHN), 3.63–3.60 (m, 4H, CH2CH2O), 2.13–2.11 (m,
4H, CCH2CH2), 1.58–1.54 (m, 4H, CH2CH2CH2), 1.39–1.35
(m, 4H, CH2CH2CH2), 0.89 (s, 18H, OSiC(CH3)3), 0.02 (s,12H,
OSiCH3). 13C NMR (100 MHz, CDCl3): d = 169.3, 142.5, 128.9,
127.7, 126.9, 75.3, 69.8, 63.1, 46.5, 33.3, 32.8, 26.2, 20.6, -5.0.
MS (ESI): m : z = 679.5 M+1+. Anal. calc. for C39H62N2O4Si2:
C, 68.98; H, 9.20; N, 4.13; found: C, 69.26; H, 8.58;
N, 4.61.
Synthesis of 1,1¢-{5,5-bis[(R)-4,5-dihydro-4-phenyloxazol-2-
yl]nonane-1,9} bis-(1,2-dimethyl-1H-imidazole) diOTs 10a
7a (0.270 g, 0.36 mmol) and 1,2-dimethyl-1H-imidazole were
dissolved in 2 mL of DMF, and the solution was heated to
70 ◦C for 24 h under N2 atmosphere. DMF was removed under
high vacuum. The residue was washed with Et2O (3 ¥ 10 mL)
until it cha◦nged to a light-yellow solid; yield: 0.290 g (85%).
1
mp: 82–84 C [a]20= +55.4 (c = 0.08, CH3OH), H NMR (400
MHz, CDCl3): d D= 7.71–7.66 (m, 6H, imidazole-CH + OTs-
ph-CH), 7.49–7.18 (m, 10H, OTs-ph-CH +oxazoline-ph-CH),
7.10–7.08 (m, 6H, OTs-ph-CH +oxazoline-ph-CH), 5.23–5.21
(dd, 2H, J = 8.0 Hz and 2.0 Hz, 2H, oxazoline-CHaHbO), 4.68–
4.64 (dd, 2H, J = 8.0 Hz and 2.0 Hz, 2H, oxazoline-CHaHbO),
4.17–4.07 (m, 4H, oxazoline-CHN + CH2CH2mim), 3.82 (s,
6H, imidazole-CH3), 3.48–3.45 (m, 4H, CCH2CH2), 2.65(s,
6H, imidazole-CH3), 2.31(s, 6H, OTs-CH3), 1.84–1.18(m, 8H,
CH2CH2CH2). 13C NMR (100 MHz, CDCl3): d = 168.6, 144.3,
144.1, 129.3, 128.9, 128.8, 127.8, 127.4, 126.9, 126.1, 122.9,
121.6, 76.9, 75.3, 69.6, 48.4, 35.5, 32.5, 29.8, 29.5, 21.5, 21.0.
MS (ESI): m : z = 304.4 M2+, m : z = 171.0 M-. HRMS (ESI):
calc. for C37H48N6O22+: 304.1919, found: 304.1914
Synthesis of 5,5-bis[(R)-4,5-dihydro-4-phenyloxazol-2-
yl]nonane-1,9-diol 6a
A solution of 5a (0.966 g, 1.40 mmol) in 32 mL of dry THF
was cooled to 0 ◦C under a N2 atmosphere. TBAF (2.100 g)
in 9 mL of THF was added to the solution dropwise, and the
resulting mixture was warmed to room temperature and stirred
for 8 h. The reaction mixture was subsequently quenched by
the addition of water (30 mL), washed with EtOAc (30 mL)
and then with brine (30 mL), and extracted with EtOAc (2 ¥
30 mL). The organic fractions were dried over Na2SO4, and the
solvent was removed under vacuum to afford the crude product,
which was purified by column chromatography (SiO2, EtOAc)
to afford 6a as a yellow oil; yield: 0.434 g (69%). [a]2D0 = +113.6
Synthesis of 1,1¢-{5,5-bis[(R)-4,5-dihydro-4-phenyloxazol-2-
yl]nonane-1,9} bis-(1,2-dimethyl-1H-imidazole) dihexafluoro-
phosphate 10e
1
(c = 0.50, CH2Cl2), H NMR (400 MHz, CDCl3): d = 7.31–
7.20 (m, 10H, oxazoline-ph-CH), 5.22–5.18 (dd, J = 8.0 Hz
and 2.0 Hz, 2H, oxazoline-CHaHbO), 4.65–4.61 (dd, J = 8.0
Hz and 2.0 Hz, 2H, oxazoline-CHaHbO), 4.12–4.08 (t, J = 8.0
Hz, 2H, oxazoline-CHN), 3.53–3.49 (m, 4H, CH2CH2O), 2.07–
2.00 (m, 4H, CCH2CH2), 1.41–1.38 (m, 4H, CH2CH2CH2),1.24–
1.22 (m, 4H, CH2CH2CH2). 13C NMR (100 MHz, CDCl3):
d = 169.4, 142.3, 128.9, 127.9, 126.9, 75.3, 69.5, 61.7, 46.7,
32.5, 20.5. MS (ESI): m : z = 451.3 M+1+. Anal. calc. for
C27H34N2O4: C, 71.97; H, 7.61; N, 6.22; found: C, 71.53; H, 7.08;
N, 5.79.
10a (0.290 g, 0.30 mmol) was dissolved in 2 mL of H2O, and
KPF6 (0.220 g, 1.10 mmol) was added to the solution and stirred
for 6 h at room temperature. The resultant white solid was filtered
and dried under vacuum to afford the product; yield: 0.250 g
(86%). mp: 108–111 ◦C [a]2D0 = +49.3 (c = 0.08, MeOH), 1HNMR
(400 MHz, DMSO): d = 7.64–7.58 (m, 4H, imidazole-CH), 7.57–
7.26 (m, 10H, oxazoline-ph-CH), 5.21–5.20 (dd, 2H, J = 8.0 Hz
and 2.0 Hz, 2H, oxazoline-CHaHbO), 4.67–4.66 (dd, 2H, J =
8.0 Hz and 2.0 Hz, 2H, oxazoline-CHaHbO), 4.11–4.07 (m, 6H,
oxazoline-CHN + CH2CH2mim), 3.74 (m, 6H, imidazole-CH3),
2.57(s, 6H, imidazole-CH3), 1.98–1.24(m, 12H, CH2CH2CH2).
13C NMR (100 MHz, CDCl3): d = 169.4, 142.3, 139.3, 128.9,
128.6, 127.8, 127.4, 126.9, 126.1, 122.8, 121.6, 69.6, 48.4, 35.5,
32.5, 29.8, 29.5, 21.0, 10.2. MS (ESI): m:z = 304.6 M2+, m:z
= 144.9 M-. HRMS (ESI): calc. for C37H48N6O22+: 304.1919,
Synthesis of 5,5-bis[(R)-4,5-dihydro-4-phenyloxazol-2-
yl]nonane-1,9-ditosylate 7a
To a solution of 6a (0.434 g, 0.96 mmol) in 20 mL of DCM, TsCl
(0.736 g, 3.86 mmol) and Et3N (0.390 g, 3.86 mmol) were added
under a N2 atmosphere, and the mixture was stirred for 8 h. The
reaction mixture was subsequently quenched by the addition of
brine (20 mL). The organic layer was washed with water (3 ¥
20 mL) and dried over Na2SO4, and the solvent was removed
under vacuum to afford the crude product, which was purified by
column chromatography (SiO2, EtOAc/PE60-90 1/1) to afford
7a; yield: 0.540 g (70%). mp: 116–118 ◦C [a]D20 = +162.5 (c = 2.00,
-
found: 304.1914, calc. for PF6 : 144.9807, found: 144.9809.
Acknowledgements
The National Technological Project of the Manufacture and
Innovation of Key New Drugs (2009ZX09103-143) and the
Major Projects Cultivating Special Program in Technology
Innovation Program (Grant NO. 2011CX01008) of the Beijing
Institute of Technology are gratefully acknowledged for their
financial support.
1
CH2Cl2), H NMR (CDCl3): d = 7.76 (m, 4H, OTs-ph-CH),
7.35–7.23 (m, 10H, OTs-ph-CH + oxazoline-ph-CH), 5.23–5.19
(dd, J = 8.0 Hz and 2.0 Hz, 2H, oxazoline-CHaHbO), 4.66 (dd,
J = 8.0 Hz and 2.0 Hz, 2H, oxazoline-CHaHbO), 4.11 (t, J = 8.0
Hz, 2H, oxazoline-CHN), 4.02 (t, J = 8.0 Hz, 4H, CH2CH2O),
2.42 (s, 6H, OTs-CH3), 2.04–1.95 (m, 4H, CCH2CH2), 1.70–
1.65 (m, 4H, CH2CH2CH2), 1.33–1.36 (m, 4H, CH2CH2CH2).
13C NMR (100 MHz, CDCl3): d = 168.4, 144.8, 142.2, 129.9,
128.9, 127.9, 126.8, 75.2, 70.3, 69.7, 46.1, 32.3, 29.1, 21.7, 20.1.
MS (ESI): m : z = 759.4 M+1+, Anal. calc. for C41H42N2O8S2:
C, 64.88; H, 6.11; N, 3.69; found: C, 64.54; H, 6.53;
N, 3.48.
References
1 D. A. Evans, K. A. Woerpel, M. M. Hinman and M. M. Faul, J. Am.
Chem. Soc., 1991, 113, 726–728.
2 R. Rasappan, D. Laventine and O. Reiser, Coord. Chem. Rev., 2008,
252, 702–714.
3 E. J. Corey, N. Imai and H. Y. Zhang, J. Am. Chem. Soc., 1991, 113,
728–729.
2970 | Green Chem., 2011, 13, 2963–2971
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
ˇ
ˇ
4 J. Bayardon, O. Holczknecht, G. Pozzi and D. Sinou, Tetrahedron:
Asymmetry, 2006, 17, 1568–1572.
5 S. Doherty, P. Goodrich, C. Hardacre, V. Paˆrvulescu and C. Paun,
Adv. Synth. Catal., 2008, 350, 295–302.
6 C. T. Qian and L. C. Wang, Tetrahedron: Asymmetry, 2000, 11, 2347–
2357.
7 P. O’Leary, N. P. Krosveld, K. P. De Jong, G. V. Koten and R. J. M.
K. Gebbink, Tetrahedron Lett., 2004, 45, 3177–3180.
8 D. Frain, F. Kirby, P. McArdle and P. O’Leary, Tetrahedron Lett.,
2010, 51, 4103–4106.
9 (a) D. Rechavi and M. Lemaire, Chem. Rev., 2002, 102, 3467–3493;
(b) C. Jonsson, K. Hallman, H. Anderson, G. Stemme, M. Malkoch,
E. Malmstrom, A. Hult and C. Moberg, Bioorg. Med. Chem. Lett.,
2002, 12, 1857–1861.
10 (a) N. Hiroto, N. Yasuhiro, T Kouichi, F Reiko, U Koji and T
Mitsuhiro, Heterocycles, 2008, 76, 381–390; (b) J M. Fraile, J. I.
Garc´ıa, C. I. Herrer´ıas, J. A. Mayoral, S. Gmough and M. Vaultier,
Green Chem., 2004, 6, 93–98.
17 R. Sebesta, I. Kmentova´ and S. Toma, Green Chem., 2008, 10, 484–
496.
18 M. Lombardo, M. Chiarucci and C. Trombini, Chem.–Eur. J., 2008,
14, 11288–11291.
19 I. Meracz and T. Oh, Tetrahedron Lett., 2003, 44, 6465–6468.
20 S. Doherty, P. Goodrich, C. Hardacre, J. G. Knight, M. T. Nguyen,
V. I . P aˆrvulescu and C. Paun, Adv. Synth. Catal., 2007, 349, 951–963.
21 M. I. Burguete, J. M. Fraile, J. I. Garc´ıa, E. Garc´ıa-Verdugo, S. V.
Luis and J. A. Mayoral, Org. Lett., 2000, 2, 2908.
22 M. J. Fernandez, J. M. Fraile, J. I. Garc´ıa, J. A. Mayoral, M. I.
Burguete, E. Garc´ıa-Verdugo, S. V. Luis and M. A. Harmer, Top.
Catal., 2000, 13, 303–309.
23 M. I. Burguete, J. M. Fraile, J. I. Garc´ıa, E. Garc´ıa-Verdugo, C. I.
Herrer´ıas, S. V. Luis and J. A. Mayoral, J. Org. Chem., 2001, 66,
8893–8901.
24 M. I. Burguete, J. M. Fraile, E. Garc´ıa-Verdugo, S. V. Luis, V.
Mart´ınez-Merino and J. A. Mayoral, Ind. Eng. Chem. Res., 2005,
44, 8580–8587.
11 J. Dupont, R. F. de Souza and P. A. Z. Suarez, Chem. Rev., 2002,
102, 3667–3692.
25 R. Annunziata, M. Benaglia, M. Cinquini, F. Cozzi and M. Pitillo,
J. Org. Chem., 2001, 66, 3160–3166.
12 V. I. Paˆrvulescu and C. Hardacre, Chem. Rev., 2007, 107, 2615–2665.
13 S. F. Liu and J. L. Xiao, J. Mol. Catal. A: Chem., 2007, 270, 1–43.
14 P. Wasserscheid and P. S. Schulz, in Ionic Liquids in Synthesis, ed.
P. Wasserscheid, T. Welton, 2nd edn, Wiley-VCH, Weinheim, 2007,
369–463.
15 (a) C. E. Song, Chem. Commun., 2004, 1033–1034; (b) J. Ding, V.
Desikan, X. X. Han, T. L. Xiao, R. F. Ding, W. S. Jenks and D. W.
Armstrong, Org. Lett., 2005, 7, 335–337; (c) L. C. Branco and C. A.
M. Afonso, J. Org. Chem., 2004, 69, 4381–4389.
26 A. Mandoli, R. Garzelli, S. Orlandi, D. Pini, M. Lessi and P.
Salvadori, Catal. Today, 2009, 140, 51–57.
27 H. Yuan, Z. Zhou, J. Xiao, L. Liang and L. Dai, Tetrahedron:
Asymmetry, 2010, 21, 1874–1884.
28 A. A. Bulich, K. K. Tung and G. Scheibner, J. Biolumin. Chemilumin.,
1990, 5, 71–77.
29 M. Pattard and H. Moser, Ecotoxicological Characterization of
Waste, Springer, New York, 2009. pp. 105–115.
30 J. S. Johnson and D. A. Evans, Acc. Chem. Res., 2000, 33, 325–335.
31 Y. Qian, X. Zheng and Y. Wang, Eur. J. Org. Chem., 2010, 19, 3672–
3677.
16 C. E. Yeom, H. W. Kim, Y. J. Shin and B. M. Kim, Tetrahedron Lett.,
2007, 48, 9035–9039.
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2963–2971 | 2971
©