An effective heterogeneous l-proline catalyst for the direct asymmetric aldol reaction using graphene oxide as support

Article abstract of DOI:10.1016/j.jcat.2012.11.024

Pristine l-proline was non-covalently loaded on the graphene oxide (GO) sheet in a simple route by mixing them in aqueous solution. Technologies of characterization well suggested that l-proline was efficiently loaded on the two sides and edge of the GO sheet through hydrogen-bonding or/and ionic interaction, giving the excellent l-proline/GO hybrid catalyst for the direct asymmetric aldol reaction. The unique multilayered structure of the GO carrier with sufficient interlayer space favored reagents' diffusion toward l-proline chiral moiety and therefore resulted in the high catalytic efficiency of the heterogeneous l-proline. Excellent yield (96%) with high enantiomeric excess (79% ee) was obtained in the direct aldol reaction of 2-nitrobenzaldehyde with acetone catalyzed by l-proline/GO hybrid, which was comparable to that observed in the reactions promoted by l-proline itself. Furthermore, the l-proline/GO hybrid used as a heterogeneous catalyst could be easily recovered and recycled for seven times without significant loss of the reactivity.

Full text of DOI:10.1016/j.jcat.2012.11.024

Journal of Catalysis 298 (2013) 138–147
Contents lists available at SciVerse ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
An effective heterogeneous L-proline catalyst for the direct asymmetric aldol
reaction using graphene oxide as support
Rong Tan a, Chengyong Li a, Jianqing Luo a, Yu Kong a, Weiguo Zheng a, Donghong Yin a,b,
⇑
a Key Laboratory of Chemical Biology and Traditional Chinese Medicine Research (Ministry of Education), Hunan Normal University, Changsha, Hunan 410081, China
b Technology Center, China Tobacco Hunan Industrial Corporation, No. 426 Laodong Road, Changsha, Hunan 410014, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Pristine
mixing them in aqueous solution. Technologies of characterization well suggested that
ciently loaded on the two sides and edge of the GO sheet through hydrogen-bonding or/and ionic inter-
action, giving the excellent -proline/GO hybrid catalyst for the direct asymmetric aldol reaction. The
unique multilayered structure of the GO carrier with sufficient interlayer space favored reagents’
diffusion toward -proline chiral moiety and therefore resulted in the high catalytic efficiency of the het-
erogeneous -proline. Excellent yield (96%) with high enantiomeric excess (79% ee) was obtained in the
direct aldol reaction of 2-nitrobenzaldehyde with acetone catalyzed by -proline/GO hybrid, which
was comparable to that observed in the reactions promoted by -proline itself. Furthermore, the
-proline/GO hybrid used as a heterogeneous catalyst could be easily recovered and recycled for seven
times without significant loss of the reactivity.
L
-proline was non-covalently loaded on the graphene oxide (GO) sheet in a simple route by
Received 1 September 2012
Revised 23 October 2012
Accepted 19 November 2012
L-proline was effi-
L
L
Keywords:
L
L-Proline
L
GO sheet
L
Chiral heterogeneous catalyst
Direct asymmetric aldol reaction
Enantiomeric selectivity
L
Ó 2012 Elsevier Inc. All rights reserved.
1. Introduction
behaviors of PVC-TEPA-supported
Lu et al. [13] synthesized -proline-functionalized polymers as
supported organocatalysts. Bae et al. [14,15] grafted -proline onto
L-proline in the aldol reaction.
L
The direct asymmetric aldol reaction is one of the most impor-
tant CAC bond-forming reactions in organic synthesis and is of
much importance in the pharmaceutical, agrochemical, and fine
chemical industries [1–4]. In 2000, List et al. demonstrated the
use of proline as an efficient catalyst for the direct asymmetric
aldol reaction between unmodified ketone and a variety of alde-
hydes [5]. The proline used as an important chiral small-molecule
organocatalyst has been drawn much attention since it is easily
accessible, environmentally safe, and available in both the
enantiomeric forms [6–8]. However, it suffered from the unavoid-
able drawbacks of homogeneous catalytic processes (e.g., lower
thermal stability and difficulties in catalyst separation and
L
the heterogenized silica for catalyzing the asymmetric aldolization.
Calderón et al. [16,17] catalyzed the asymmetric aldol reaction by
proline on mesoporous materials. Miao and Chan [18] prepared an
IL-anchored proline catalyst, which was efficient and recyclable for
asymmetric aldol reaction. The heterogeneous catalysts indeed can
be recycled for reuse, but some of them are less efficient due to the
low accessibility of substrates. Recently, layered double hydroxides
(LDHs) have emerged as an attractive support to develop the
L
-proline/LDHs hybrid using intercalation of
LDH through ion-exchange method [23]. It is found that the
immobilization of -proline on LDHs has no adverse effect on the
catalytic activity of -proline. Thus, this represents a fascinating
strategy for developing the heterogeneous -proline catalyst with
L-proline in MgAAl
L
recovery). Immobilization and recycling of
L
-proline have received
L
considerable concerns in recent years. Several types of supports,
such as polymer [9–12], silica [13–17], ionic liquid (IL) [18,19],
b-cyclodextrin [20], Merrifield resin [21] and magnetite [22], are
usually considered for the immobilizations of proline and its
derivatives. Gruttadauria and his co-workers [9–11] used polysty-
rene-supported proline-based organic catalysts for the direct
asymmetric aldol reaction. Zou et al. [12] investigated the catalytic
L
high efficiency and easy reusability by supporting the
on layered materials.
L-proline
Recently, a newly layered material-graphene oxide (GO) has at-
tracted much attention of scientists all over the world, since it
exhibits unique surface properties (oxygenated functional groups
on the basal planes and edge), high-specific surface area, and easy
exfoliation into monolayers under water [24–26]. In particular, the
GO consists of intact graphitic regions interspersed with
sp3-hybridized carbons containing carboxyl, hydroxyl, and epoxide
functional groups on the edge, top, and bottom surface of each
sheet and sp2-hybridized carbons on the aromatic network [26].
⇑
Corresponding author at: Key Laboratory of Chemical Biology and Traditional
Chinese Medicine Research (Ministry of Education), Hunan Normal University,
Changsha, Hunan 410081, China. Fax: +86 731 88872531.
E-mail address: yindh@hunnu.edu.cn (D. Yin).
0021-9517/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcat.2012.11.024

R. Tan et al. / Journal of Catalysis 298 (2013) 138–147
139
The presence of the abundant oxygen functional groups provides
GO sheet with large capability of loading organic molecules
through covalent or non-covalent approaches, facilitating develop-
ment of a broad novel class of materials with enhanced properties
and even introducing new functionalities to GO sheet [27,28]. Yang
et al. have reported high-efficiency loading of doxorubicin hydro-
chloride (DXR) on GO sheet (0.91 mg/mg) through hydrogen-
performed on Netsch STA449c. The sample weight was ca. 10 mg
and was heated from room temperature up to 800 °C with 10 °C/
min using alumina sample holders. Analytical high performance
liquid chromatography (HPLC) was carried out on Waters 2695
Separations Module with Waters 2996 photodiode array detector
using Daicel chiralpak OB-H, AD-H or AD columns.
bonding interaction between GO and DXR, as well as the
p–p
2.2. Preparation of L-proline/GO hybrid
stacking interaction [29]. Apart from the abundant oxygen func-
tional groups, the key to understanding the large loading capability
of GO sheet lies in the large theoretical specific surface area (up to
400–1500 m2 gꢀ1) [30], as well as the high efficient utilization of
surface area because both sides of the nanosheet are accessible.
Based on the fascinating layered structure of GO material, as
well as the large capability of loading organic molecules, we reason
In the experiments, GO was prepared by the oxidation of high-
purity graphite powder (99.9999%, 200 mesh) with H2SO4/KMnO4
according to the method of Hummers and Offeman [32]. After re-
peated washing of the resulting yellowish-brown cake with hot
water, the powder of GO was dried at room temperature under
vacuum overnight. FT-IR (KBr): 3390, 3132, 1735, 1621, 1224,
1050, 581 cmꢀ1
The mixture of the dried GO (0.1 g) and
.
that the GO can be used as an efficient support for
lows for desirable catalytic properties. It is well known that the
carboxyl and secondary amine groups of the -proline are capable
of participating in hydrogen bond [31], and we thus decide to
introduce the -proline into the GO sheet through the hydrogen-
bonding interaction between the -proline and the GO sheet to
prepare the -proline/GO hybrid. We envision that the GO sheet
can efficiently load -proline molecular on its two sides and edge,
L-proline and al-
L
-proline (0.2 g) was
L
sonicated in deionized water for 0.5 h and further stirred at room
temperature for 24 h. After centrifugal separation by using an
Eppendorf 5804 centrifuge operated at 10,000 rpm for 15 min,
the precipitate was collected and dried overnight under vacuum
L
L
L
to give brown flake solid of
3369, 2976, 2741, 2301, 1617, 1398, 1326, 1163, 1036, 837, 775,
673, 620 cmꢀ1. The
-proline content in the -proline/GO hybrid
is 4.06 mmol/g, which was determined according to the content
L-proline/GO hybrid. FT-IR (KBr):
L
resulting in a sandwich-like hybrid (
of GO sheet alternate with layers of
L
-proline/GO) where layers
-proline. The intercalated
L
L
L
L-proline component serves as the catalytic sites, and the GO sheet
of nitrogen element in
analyses.
L-proline/GO hybrid analyzed by elemental
as support is expected to favor accessibility of reagents, which in
turn increases the catalytic efficiency of the heterogeneous
For comparison, we also prepared the graphite-supported
-proline catalyst -proline/graphite) and the active carbon-
supported -proline catalyst ( -proline/AC) according to the above
similar preparation procedure. -proline content in the prepared
L-proline catalyst. Furthermore, the non-covalent method, which
L
(L
does not need to tune the structures of the -proline and the GO
L
L
L
sheet, represents an interesting strategy for the attachment of
active sites on the GO sheet for use in catalytic reactions. Herein,
L
catalysts was also measured in terms of nitrogen element percent-
age as obtained from nitrogen elemental analyses results. The
results are listed in Table 1.
L
-proline was efficiently loaded on GO sheet simply by using the
non-covalent approach. It is interesting that the layered
-proline/GO hybrid, in which reagents can free access interlamel-
L
lar space, functions as the heterogeneous catalyst, but behaves as
the homogeneous catalyst. Thus, it presents comparable catalytic
2.3. General procedure for the direct asymmetric aldol reaction
activity and enantioselectivity relative to the pristine
and can be easily separated for reuse by centrifugation.
L-proline
2.3.1. Typical procedure for the asymmetric aldol reactions of acetone
with various aryl aldehydes
A mixture of catalyst (0.035 g, 0.14 mmol
L-proline content in
the -proline/GO hybrid) and acetone (4 ml) was stirred at 30 °C
L
2. Experimental
for 10 min. Subsequently, the corresponding aromatic aldehyde
(0.5 mmol) was added. The resulting mixture was stirred at room
temperature until the reaction was judged to be complete based
on TLC analysis. The catalyst was then separated by centrifuga-
tion. The upper organic phase with the product was concentrated
under reduced pressure. The crude products were purified by
2.1. Materials and methods
4-nitrobenzaldehyde, 2-chlorobenzaldehyde, and 4-acetamido-
benzaldehyde were obtained by TCI. 2-nitrobenzaldehyde, 2-
naphthaldehyde and 4-bromobenzaldehyde were bought from
Acros. 3-pentanone, cyclohexanone, and cyclopentanone were
purchased from Aldrich. Other commercially available chemicals
were laboratory grade reagents from local suppliers. They were
used without further purification, except for the aromatic alde-
hyde, which was purified by distillation.
column chromatography on silica gel (Acros, 40–60 lm, 60 Å,
eluent hexane/ethyl acetate 5:1). Enantiomeric excess (ee value)
was determined by HPLC on Daicel chiralpak OB-H or AD
columns.
1H NMR spectra of samples were recorded at a Varian-500 spec-
trometer. Tapping mode atomic force microscopy (AFM) measure-
ments were performed using a multimode SPM from Digital
Instruments with a Nanoscope IIIa Controller. X-ray diffraction
(XRD) patterns were recorded on a Philips X’PERT-Pro-MPD diffrac-
2.3.1.1. (4R)-Hydroxy-4-(2-nitrophenyl)-butan-2-one 1. Enantiomeric
excess was determined by HPLC with a Daicel chiralpak OB-H col-
umn (i-PrOH/hexane = 15:85), 25 °C, 254 nm, 1 ml/min; major
antienantiomer tR 8.4 = min and minor antienantiomer tR = 7.3 min.
1H NMR (CDCl3, 500 MHz): d = 7.95–7.42 (m, 4H), 5.68 (d,
J = 10.7 Hz, 1H), 3.73 (s, 1H), 3.15–2.72 (m, 2H), 2.24 (s, 3H).
tometer using Cu K radiation (k = 1.542 Å). A continuous scan
a
mode was used to collect 2h from 5° to 40°. Fourier transform
infrared (FT-IR) spectra were obtained as potassium bromide
pellets with a resolution of 4 cmꢀ1 and 32 scans in the range
400–4000 cmꢀ1 using an AVATAR 370 Thermo Nicolet spectropho-
tometer. Elemental analyses of N were carried out on Vario EL III
Elemental analyses made in Germany. The thermogravimetric
and differential thermogravimetric (TG–DTG) analysis was
2.3.1.2. (4R)-Hydroxy-4-(4-nitrophenyl)-butan-2-one 2. Enantio-
meric excess was determined by HPLC with a Daicel chiralpak
OB-H column (i-PrOH/hexane = 15:85), 25 °C, 254 nm, 1 ml/min;
major antienantiomer tR = 14.0 min and minor enantiomer
tR = 16.1 min. 1H NMR (CDCl3, 500 MHz): d = 8.22–7.28 (m, 4H),
5.27 (s, 1H), 3.67 (s, 1H), 2.87 (d, J = 6.7 Hz, 2H), 2.23 (s, 3H).

140
R. Tan et al. / Journal of Catalysis 298 (2013) 138–147
Table 1
2.3.2.2. (2R,10S)-2-(Hydroxy-(2-nitrophenyl)methyl)cyclohexan-1-one
Chemical composition of different L-proline hybrids.
8. Enantiomeric excess for anti HPLC was determined by HPLC with
a Chiralpak AD column (i-PrOH/hexane = 15:85), 25 °C, 254 nm,
1.0 ml/min, major antienantiomer tR = 13.4 min, and minor antien-
antiomer tR = 14.2 min. 1H NMR (500 MHz, CDCl3): d: anti: 7.86–
7.44 (m, 4H), 5.46–5.44 (d, J = 7.0 Hz, 1H), 2.47–2.44 (m, 1H),
2.49–2.34 (m, 2H), 2.09–2.08 (m, 1H), 1.87–1.66 (m, 6H).
Entry Samples
N content
(wt.%)
L
-proline content
L-proline
content (wt.%)
(mmol/g)
1
2
3
5.68
2.09
0.83
4.06
1.49
0.59
46.7
17.1
6.78
L
L
L
-proline/GO
-proline/AC
-proline/Graphite
2.3.2.3. (2R,10S)-2-(Hydroxy-(2-nitrophenyl)methyl)cyclopentan-1-one
9. Enantiomeric excess for anti HPLC was determined by HPLC
2.3.1.3. (4R)-Hydroxyl-4-(2-chlorophenyl)butan-2-one 3. Enantio-
meric excess was determined by HPLC with a Daicel chiralpak
AD column (i-PrOH/hexane = 7.5:92.5), 25 °C, 254 nm, 0.8 ml/
min; major antienantiomer tR = 10.9 min and minor antienantio-
mer tR = 12.3 min. 1H NMR (500 MHz, CDCl3): d = 7.61–7.18 (m,
4H), 5.52–5.49 (m, 1H), 3.63 (s, 1H), 3.00–2.65 (m, 2H), 2.21(s, 3H).
with
a Chiralpak AD column (i-PrOH/hexane = 10:90), 25 °C,
220 nm, 1.0 ml/min; major antienantiomer tR = 15.5 min and
minor antienantiomer tR = 16.3 min. 1H NMR (500 MHz, CDCl3):
d: anti: 7.92–7.35 (m, 4H), 5.38–5.36 (d, J = 9.0 Hz, 1H), 2.79–2.74
(m, 1H), 2.34–2.28 (m, 1H), 2.13–2.05 (m, 2H), 1.99–1.95
(m, 1H), 1.69–1.62 (m, 4H).
2.3.1.4. (4R)-Hydroxyl-4-(4-bromophenyl)butan-2-one 4. Enantio-
meric excess was determined by HPLC with a Daicel chiralpak
AD column (i-PrOH/hexane = 7.5:92.5), 25 °C, 254 nm, 0.8 ml/
min; major antienantiomer tR = 15.2 min and minor antienantio-
mer tR = 16.1 min. 1H NMR(500 MHz, CDCl3): d = 7.47–7.21 (m,
4H), 5.10 (t, J = 8.7 Hz, 1H), 3.43 (s, 1H), 2.86–2.76 (m, 2H), 2.19
(s, 3H).
3. Results and discussion
3.1. Preparation of the L-proline/GO hybrid
GO has a layered morphology with hydroxyl and epoxy groups
functionality disrupting the hexagonal carbon basal planes on the
interior of multilayered stacks of graphene oxide, and carboxyl
groups decorating the periphery of the planes. The structural char-
acteristics, together with its high-specific surface area, provide GO
sheet with large capability of loading L-proline simply through
non-covalent method. In addition, the mean interlayer distance be-
tween the GO sheet is reported to be in the range of 0.6–1.1 nm
2.3.1.5. (4R)-Hydroxyl-4-(4-acetamidophenyl)butan-2-one 5. Enan-
tiomeric excess was determined by HPLC with a Daicel chiralpak
AD column (i-PrOH/hexane = 10:90), 25 °C, 254 nm, 0.8 ml/min;
major antienantiomer tR = 50.0 min and minor antienantiomer
tR = 55.7 min. 1H NMR(500 MHz, CDCl3): d = 7.47–7.26 (m, 4H,),
5.12–5.11 (m, 1H), 3.34 (s, 1H), 2.89–2.77 (m, 2H), 2.20 (s, 3H),
2.17 (s, 3H).
[33], which is far broader than the thickness of
(ca. 0.29 nm, reported in [23]), suggesting a flexible orientation of
interlayer -proline molecule in interlayer voids and the high
accessibility of substrates. Base on the specific physicochemical
properties of GO, we try to prepare the heterogeneous -proline/
GO hybrid catalyst by intercalating the -proline into the GO inter-
layer galleries. A strategy that we have designed here is to main-
tain the pristine -proline backbone, since both the carboxylic
acid and the pyrrolidine functionalities are essential for effective
asymmetric induction. The -proline/GO hybrid was prepared sim-
ply by mixing the GO sheet with pristine -proline in aqueous solu-
L-proline molecule
L
2.3.1.6. (4R)-Hydroxy-4-(2-naphthyl)butan-2-one 6. Enantiomeric
excess was determined by HPLC with a Daicel chiralpak AD column
(i-PrOH/hexane = 7.5:92.5), 25 °C, 254 nm, 0.8 ml/min; major anti-
enantiomer tR = 22.6 min and minor antienantiomer tR = 27.0 min.
1H NMR (500 MHz, CDCl3): d = 7.80–7.43 (m, 7H), 5.31–5.29
(m, 1H), 3.50 (s, 1H), 2.96–2.84 (m, 2H), 2.18 (s, 3H).
L
L
L
L
2.3.2. Typical procedure for the asymmetric aldol reactions of ketone
with 2-nitrobenzaldehyde
L
tion, where the hydrogen-bonding interaction between hydroxyl
and epoxyl groups on the GO sheet with carboxyl and secondary
A mixture of catalyst (0.035 g, 0.14 mmol L-proline content in
the L -PROLINE/GO HYBRID) corresponding ketone (5 mmol) and DMSO
(4 ml) was stirred at room temperature for 10 min. Subsequently,
the 2-nitrobenzaldehyde (0.5 mmol) was added. The resulting mix-
ture was stirred at room temperature until the reaction was judged
to be complete based on TLC analysis. The reaction was quenched
by adding saturated NH4Cl solution, and then, the catalyst was
separated by centrifugation. The products were extracted with
Et2O (3 ꢁ 5 ml), the combined extracts were dried over anhydrous
Na2SO4, the solvent was evaporated in vacuo, and the residue was
amine groups in
L-proline could be formed during the preparation
procedure. Therefore, the driving force for
L
-proline binding to the
GO sheet should be primarily attributed to hydrogen-bonding
interaction, which occurs on the basal planes existing in the inter-
layer of the GO minerals. Also, the ionic interaction may also occur
at the carboxylic group sites on the edge of layered mineral, which
was proposed in Scheme 1. In order to investigate the interaction
between
less oxygen functional groups, such as active carbon and graphite,
were employed for the -proline loading. It is found that the load-
ing of -proline on GO (as high as 4.06 mmol/g) is much higher than
L-proline and GO sheet, other carbonaceous supports with
purified by column chromatography on silica gel (Acros, 40–60 lm,
L
60 Å, eluent n-hexane/EtOAc 3:1). The diastereoselectivity (dr) of
aldols’ products were measured by 1H NMR of the crude reaction
mixture. The enantioselectivity of the anti-isomers was deter-
mined by HPLC on Daicel chiralpak AD-H or AD column. The abso-
lute configurations of the products were assigned by analogy with
the previously reported results [28].
L
that on the active carbon (1.49 mmol/g) and on the graphite
(0.59 mmol/g) (Table 1, entry 1 vs. 2, 3). It is suggested that the
presence of the abundant oxygen functional groups on the planar
surface or edge of GO sheet, as well as the high efficient utilization
of the surface area of the GO sheet, is responsible for the higher
L
-proline loading on GO sheet.
2.3.2.1. (3R,4S)-3-Methyl-4-hydroxy-4-(2-nitrophenyl)-butan-2-one
7. Enantiomeric excess for anti HPLC was determined by HPLC
with a Daicel chiralpak AD-H column (i-PrOH/hexane = 15:85),
25 °C, 254 nm, 1.0 ml/min; major antienantiomer tR = 7.3 min and
minor antienantiomer tR = 8.4 min. 1H NMR (CDCl3, 500 MHz): d:
anti: 7.95–7.42 (m, 4H), 5.40 (d, J = 9.8 Hz, 1H), 3.61 (s, 1H),
3.47–3.25 (m, 1H), 2.13 (s, 3H), 1.42–1.30 (m, 3H).
3.2. Samples characterization
3.2.1. AFM
We employed AFM to establish the thickness and surface
roughness of the L-proline/GO hybrid since AFM characterization

R. Tan et al. / Journal of Catalysis 298 (2013) 138–147
141
OH
H
N
C
H
O
O
H
O
O
O
O
N
N
OH
COOH
H
H
H
H
COO
H
COOH
N
OOC
HOOC
H2O, RT, 24 h
HO
O
O
O
O
H
O
H
H
O
O
H
C
N
C
HO
N
HO
Scheme 1. Illustration of the preparation of L-proline/GO hybrid. The red arrows demonstrate hydrogen bonds. (For interpretation of the references to color in this figure
legend, the reader is referred to the web version of this article.)
has been one of the most direct methods of probing the loading of
organic molecular on GO sheets by quantifying the thickness of the
sheets after they dispersed in a solvent. Fig. 1 shows the AFM im-
age of L-proline/GO hybrid. The sample used for AFM study was
prepared by depositing the hybrid on new cleaved mica surfaces
and dried at room temperature. It is found that sufficiently dilute
suspensions of the sample prepared with the aid of ultrasound re-
sults in the exfoliated
L-proline/GO hybrid. The average thickness
of the as-prepared -proline/GO hybrid measured from the height
L
profile of the AFM image is ca. 2.0 nm, being somewhat larger than
the typical thickness of the single-layer GO (ca. 1.4 nm) [24]. An in-
crease in the thickness (ca. 0.6 nm) of
arise from a certain content of -proline molecules loaded onto
the GO surface. If we assume that monolayered -proline molecules
cover both sides of GO sheet, the estimated thickness of the
monolayered -proline molecules is ca. 0.30 nm. This value is in
agreement with to the reported thickness of -proline molecule
(ca. 0.29 nm) [23]. We thus propose that the monolayered -proline
molecules are loaded on both sides of GO sheet without aggrega-
tion, although the distinct -proline molecules are not observed
L-proline/GO hybrid may
L
L
L
L
L
L
on the plain GO sheets in the AFM image. On the other hand, the
cross-sectional view of the sheets displays the slight height varia-
tions of about 0.2–0.3 nm, which further confirms the load of the
monolayered L-proline on the GO sheet. Furthermore, the height
of the sample shows another obvious stage with the increment
of 3.0 nm, corresponding to the labeled distinct light ‘‘bulge’’ in
the AFM image of the
ered structure of the
L
-proline/GO hybrid. Based on the multilay-
-proline/GO hybrid, it is believed that the
L
light ‘‘bulge’’ should be the accumulated hybrid sheet. Accordingly,
the interlayer spacing of the -proline/GO hybrid can be derived to
L
be ca. 1.0 nm. The layered structure of the L-proline/GO hybrid is
also confirmed by XRD pattern.
3.2.2. XRD
XRD patterns are used to study the change of GO sheet in struc-
ture during the loading process. Fig. 2 shows powder XRD results of
GO material, L-proline, and L-proline/GO hybrid. XRD pattern of GO
shows a very strong (002) peak (denoted as #, 2h) centered at 9.4°,
corresponding to an average interlayer distances of ca. 0.95 nm
(Fig. 2a). Notably, as for L-proline/GO hybrid, the C (002) peak (de-
noted as #, 2h) shifts to 8.9° (Fig. 2b), corresponding to a larger
interlayer distance of ca. 0.99 nm. The value is approach to that
determined by AFM (1.0 nm). The larger interlayer spacing sug-
Fig. 1. AFM image of
concentration of 0.01 mg/ml in water.
L-proline/GO hybrid on freshly cleaved mica, with a
gests the intercalation of L-proline into the interlayer spacing of
multilayered GO material during the loading process as opposed
to simple absorption on the external surface. In addition to the C
(002) peak (2h) at 8.9°, the main intense diffraction peaks of pris-
intercalation of
a sandwich-like hybrid material where layers of GO sheet alternate
with layers of -proline. The substrates thus can easily access to the
active sites in the gallery regions, which allows the heterogeneous
L-proline into the GO interlayer galleries provides
tine
also observed in the XRD pattern of
the presence of the -proline on the GO sheet (Fig. 2b vs. c). The
L
-proline (denoted as ꢂ) based on the standard spectrum are
L
-proline/GO hybrid due to
L
L

142
R. Tan et al. / Journal of Catalysis 298 (2013) 138–147
*
c
*
#
*
b
226
*
*
*
*
*
a
b
a
220
#
10
20
30
40
2-theat, degrees
Fig. 2. XRD patterns of the samples of GO (a),
L-proline (b) and
L
-proline/GO (c).
a
1735
1050
200
250
300
350
400
450
500
Wavelength (nm)
b
1224
1036
1163
Fig. 4. UV–vis spectra of GO (a) and L-proline/GO (b) in aqueous solution.
1617
3390
3369
c
(ANHA) of
(Fig. 3c vs. b). Furthermore, the characteristic band at 1735 cmꢀ1
relating to the (C@O) of carboxyl group in the spectrum of GO
L-proline and the oxygen functional groups of GO sheet
2741
2976
m
sheet disappeared after the L-proline is loaded (Fig. 3b vs. a), pro-
3423
viding the existence of ionic interaction between the ANHþ2 A
group and the ACOOꢀ group originated from the reaction between
1622
2000
2776
2500
3058
3500
4000
3000
1500
1000
500
the secondary amine group (ANHA) of
group (ACOOH) on the edge of GO sheet [36]. All the results
encourage us to anticipate that the pristine -proline bind to the
L-proline and the carboxylic
Wavenumbers (cm-1)
L
Fig. 3. FT-IR spectra of GO (a), L-proline/GO (b) and pristine L-proline (c).
planar surface or edge of GO sheet through hydrogen-bonding or/
and ionic interaction, respectively.
L
-proline/GO catalyst to perform as the homogeneous catalyst in
the direct asymmetric aldol reaction.
3.2.3. UV–vis
UV–visible absorption spectra of pure GO and
hybrid in aqueous solution furthermore verified the interaction be-
tween GO and -proline, as described in Fig. 4. It was obvious that
the GO exhibits a maximum absorption at 220 nm, which corre-
sponds to
pꢂ transitions of aromatic C@C bonds. The charac-
teristic peak shows a redshift to 226 nm in the UV–vis spectrum
of -proline/GO hybrid. The slight redshift relative to GO suggests
that the presence of the ground-state electron donor–acceptor
L-proline/GO
3.2.2. FT-IR
L
The binding behavior of pristine L-proline toward GO can be
confirmed by characterizing the GO sheet, hybrid material of
-proline/GO, and pristine -proline using FT-IR spectroscopy, as
p
?
L
L
shown in Fig. 3. The FT-IR spectrum of GO shows characteristic
vibration bands of oxygen-containing groups at round 3390,
1735, 1224, and 1050 cmꢀ1, which are associated with the stretch-
ing vibration modes of COOAH/OAH, carboxylic C@O, CAOH, and
epoxy CAO groups present on the surface of GO, respectively
(Fig. 3a) [34,35]. While, the wavenumber of the characteristic band
of the oxygen functional groups decreases from 3390, 1224, and
1050 cmꢀ1 to 3369, 1163, and 1036 cmꢀ1, respectively, after form-
L
interaction between the GO and
brid [29,37].
L-proline in the L-proline/GO hy-
3.2.4. TG–DTG
Thermogravimetric analysis (TGA) associated with the decom-
position profiles of the GO, -proline, and -proline/GO hybrid un-
der a nitrogen atmosphere provides further evidence for the
intercalation of -proline into the GO interlayer galleries, as well
as the presence of hydrogen-bonding interaction between -proline
ing
(C@O) in the spectrum of GO sheet. The observation indicates that
-proline is successfully loaded onto GO, and the shift of character-
L-proline/GO hybrid, except for the band corresponding to
L
L
m
L
L
istic bands may be due to the hydrogen-bonding interaction
between these two species (Fig. 3b vs. a). The deduction also can
be drawn from the evidence that the absorption band assigning
to the ACOOH group in pristine
1617 cmꢀ1, compared with that of
vs. b). In addition, the bands at 2776 cmꢀ1 corresponding to
L
molecular and GO sheet, as shown in Fig. 5. Pure GO sheet shows
the main weight loss centered at 160 °C, which is presumably
due to pyrolysis of the labile oxygen-containing functional groups
(Fig. 5a) [38]. Fig. 5b displays the TG–DTG curves of L-proline/GO
hybrid. Obviously, the decomposition temperature of the oxygen-
L
-proline, shifts from 1622 to
L
-proline/GO hybrid (Fig. 3c
the asymmetric stretching vibration of the NAH group in
L
-proline
containing functional groups decreased to 140 °C after the
[23] are also found to shift to a lower position at 2741 cmꢀ1, due to
L-proline loading. It is probably attributed to the extended inter-
the formation of hydrogen bond between secondary amine group
layer distance originated from the intercalation of -proline, which
L

R. Tan et al. / Journal of Catalysis 298 (2013) 138–147
143
c
A
B
b
140
a
c
b
160
a
100 200 300 400 500 600 700 800
Temperature (oC)
100 200 300 400 500 600 700 800
Temperature (oC)
Fig. 5. Thermogravimetric (A) and differential thermogravimetric (B) results of the GO (a), L-proline/GO (b) and pristine L-proline (c) under a nitrogen atmosphere.
reduces the interlayer interaction forces between GO nanosheets
and further reduces thermal stability of the oxygen-containing
functional groups [39]. Moreover, the temperature of the succes-
decomposition profiles in the corresponding TG–DTG curves
(Fig. 6C and D), corresponding to the removal of water (first step
in the range of 46–100 °C), pyrolysis of oxygen-containing groups
(second step in the range of 112–237 °C), and the decomposition
sive pyrolysis of the
gets increased comparing with that of the pristine
an indirect proof for the presence of hydrogen-bonding interaction
between -proline molecular and the oxygen-containing functional
groups of GO sheet in the -proline/GO hybrid (Fig. 5b vs. c).
To the detailed study, the relationship between the oxygen-con-
taining functional groups GO sheet and -proline loading, TGA was
further performed to monitor the decomposition profile of the pris-
tine -proline and -proline/GO hybrid under air atmosphere. For
comparison, the control catalysts of -proline/AC and -proline/
L
-proline moiety in the
L-proline/GO hybrid
L-proline. It is
of
the mass percentage of the oxygen-containing groups (ca.
7.0 wt.%) and the -proline moiety (ca. 16 wt.%) in the -proline/
L-proline moiety (third step in the range of 254–533 °C). Thus,
L
L
L
L
AC hybrid can be estimated according to the weight loss of corre-
sponding decomposition range (Fig. 6C). The contents of the oxy-
L
gen-containing groups and the
graphite hybrid are estimated to be 1 and 7 wt.%, respectively
(Fig. 6D). The -proline content estimated according to the TG data
L-proline moiety in the L-proline/
L
L
L
L
L
is close to those calculated from the nitrogen elemental analysis
(see Table 1). Notably, higher content of the oxygen-containing
graphite are also characterized by means of the TG-DTG under
air atmosphere, and the results obtained are depicted in Fig. 6.
The pristine L-proline shows two distinct steps of weight loss in
the combined TG–DTG curves (Fig. 6A). The first weight loss cen-
ters at 68 °C, which is due to a loss of water. The second large
groups gives rise to higher loading of
catalysts. The results provide supporting evidence that the high-
efficiency loading of -proline depends on the amount of the oxy-
genated groups in the support. The high loading of -proline on
the GO sheet, together with the unique binding behaviors between
-proline and GO sheet, makes the -proline/GO hybrid efficient
L-proline for the supported
L
L
weight loss assigned to the successive cleavage of the
pears at 215 °C. The weight loss extends up to ca. 585 °C until the
-proline is completely decomposed under air flow. Fig. 6B displays
the TG–DTG curves of -proline/GO hybrid, in which three major
L-proline ap-
L
L
L
and stable in the reaction system.
L
steps are observed. The first step (weight loss = ca. 10 wt.%) before
100 °C corresponds to removal of the surface-adsorbed water and
interlayer water molecules. The second step involves the weight
loss centered at 131 °C and the followed weight loss at 220 °C.
The two weight loss peaks are well distinguished in the corre-
sponding DTG curve, which are logically assigned to the successive
pyrolysis of the labile oxygen-containing functional groups (weight
loss = ca. 23 wt.%). It is the presence of the abundant oxygenated
3.3. Catalytic performance
Apart from the ultrahigh loading capacity of GO sheet toward
-proline, the other salient feature of the -proline/GO hybrid
L
L
material is the efficient utilization of active sites because all the ac-
tive sites located on the surface of the layer are accessible, which
will render the
metric aldol reaction. The catalytic efficiency of the resultant
-proline/GO hybrid catalyst was investigated using the direct
L-proline/GO hybrid efficient for the direct asym-
species responsible for the highly efficient loading of
The third step, in the temperature range of 317–780 °C, is logical
to assign to the successive cleavage of the -proline moiety [23],
since the weight loss in this range (ca. 46 wt.%) is approximate to
the content of -proline moiety calculated from the elemental anal-
ysis (46.7 wt.%). The increased decomposition temperature of the
-proline suggests that the guest/host interaction was done
through the oxygen-containing functional groups of the GO sheet
(Scheme 1). Therefore, we can deduce that the -proline is success-
L
-proline.
L
asymmetric aldol reaction of 2-nitrobenzaldehyde and acetone at
room temperature. The results are presented in Table 2. The GO
L
material and pristine
purposes.
L-proline were also examined for comparison
L
GO material itself was found to be inactive for the direct asym-
metric aldol reaction, and practically, no aldol product was ac-
quired even if 0.035 g GO sheet was used (Table 2, entry 1).
L
L
When pristine L-proline was employed in 30 mol% catalyst loading,
fully loaded on the GO sheet by forming the non-covalent bond
rather than physical adsorption. The non-removable residue is
probably belonged to the remaining carbon. The control catalysts
the aldol product was isolated in a chemical yield (95%) and an ee
value (79%) (Table 2, entry 2). In general, the immobilization of the
homogeneous catalyst causes a decrease in the catalytic activity.
of
L-proline/AC and
L-proline/graphite also show the three-step
However, the immobilization of L-proline on GO sheet via

144
R. Tan et al. / Journal of Catalysis 298 (2013) 138–147
110
100
90
80
70
60
50
40
30
20
100
80
60
40
20
0
A
B
0
100 200 300 400 500 600 700 800
0
100 200 300 400 500 600 700 800
102
100
98
D
100
95
90
85
80
75
70
C
96
94
92
90
88
0
100 200 300 400 500 600 700 800
0
100 200 300 400 500 600 700 800
Temperature (oC)
Fig. 6. TG–DTG results of the L-proline (A), L-proline/GO (B), L-proline/AC (C) and L-proline/graphite (D) under an air atmosphere.
Table 2
The results of the direct asymmetric aldol reaction between acetone and 2-nitrobenzaldehyde over different L-proline catalysts.a
CHO
NO2
O
NO2
OH
O
Catalyst
30 oC
H3C
CH3
Entry
Catalyst
Catalyst amount (g)
Time (h)
Yieldb (%)
Eec (%)
1
2
3
GO
Pristine
0.035
24
6
6
Trace
95
96
n.d.d
79
79
L
-proline
0.017e
0.035e
L
L
L
L
L
-proline/GO
4
5
6
7
0.025
0.012
0.035
0.035
24
48
24
24
95
79
-proline/GO
-proline/GO
-proline/AC
95
78
10
n.d.d
n.d.d
trace
-proline/Graphite
a
b
c
Reaction conditions: 2-nitrobenzaldehyde (0.5 mmol), acetone (4 ml), 30 °C.
Isolated yield after column chromatography.
Ee was determined by HPLC (Daicel chiralpak OB-H column) on pure products.
Not determined.
d
e
The amount of catalyst is 30 mol% of 2-nitrobenzaldehyde.
hydrogen-bonding or/and ionic interaction has no adverse effect
on the catalytic activity of -proline in the asymmetric aldol reac-
tion. In particular, the hydrogen-bonding interaction between the
carboxyl group (ACOOH) of -proline and the hydroxyl (AOH)
dispersion of L-proline/GO in reaction system, allowed the hetero-
geneous catalyst to behave as a homogeneous catalyst. A reduction
in the amount of catalyst to 0.025 g (which amounts to 20 mol%,
L
L
regarding catalytic
L-proline sites), or further to 0.012 g (which
group of GO sheet makes the proton of the carboxyl (ACOOH)
group more acidic, which is favorable for the direct asymmetric
aldol reaction present here [5,40–44]. The isolated yield and ee va-
amounts to 10 mol%, regarding catalytic
in a relatively slow reaction but still gave quantitative yield and
79% ee within 48 h (Table 2, entries 4 and 5).
L
-proline sites), resulted
lue for the
served by using pristine
(30 mol%) under the identical conditions (yield of 96% and ee of
79%) (Table 2, entry 3 vs. 2). The high efficiency of the heteroge-
neous catalyst might be due to the unique multilayered structure
of the GO carrier, which provided both stable anchoring site for ac-
L
-proline/GO hybrid could be comparable to those ob-
-proline at the same catalyst loading
The catalytic activities are also correlated in detail with the
L
amount of loaded
line/graphite hybrid with low
of aldol product in control experiments, even though their dosages
(0.035 g) were identical with that of the -proline/GO (Table 2,
L
-proline. Both
L-proline/AC hybrid and L-pro-
L-proline loading gave small amount
L
entries 6 and 7). The results suggested that the layered GO sheet
with the unique basal plane structure, together with the abundant
tive sites and accessible void for reagents. Furthermore,
intercalated in the GO material, could be used as stabilizer for dis-
persing -proline/GO sheet in the reaction system [45]. The high
accessibility of substrate to active sites, as well as the well
L-proline,
oxygen functional groups, was the optimal carrier for
L-proline,
L
which provided the
high efficiency.
L-proline/GO with large L-proline loading and

R. Tan et al. / Journal of Catalysis 298 (2013) 138–147
145
100
90
80
70
60
50
40
30
a
a
b
a'
1036
1617
2000
3369
2976
1163
1500 1000
2741
4000
3500
3000
2500
500
1
2
3
4
5
6
7
Wavenumbers (cm-1)
Run times
Fig. 8. FT-IR spectra of the fresh
L-proline/GO catalyst (a) and the recovered L-
Fig. 7. The reuse of L-proline/GO in the direct asymmetric aldol reaction between
proline/GO after the 7th reuse (a’).
acetone and 2-nitrobenzaldehyde at 30 °C (a: yield; b: ee value).
3.4. Recycling
acetone and aromatic aldehydes bearing electron-withdrawing
groups on the aromatic ring furnished the desired b-hydroxy car-
bonyl aldol products in excellent yields (in the range of 66–95%)
within 6 h (Table 3, entries 1–2). The halogenated aromatic alde-
hydes could be employed even though longer reaction times were
generally required (Table 3, entries 3–4). The 4-acetamidobenzal-
dehyde used as an electron-rich aromatic aldehyde showed the
moderate reactivity with the presence of 30 mol% of catalyst
(Table 3, entry 5). The 2-naphthaldehyde was also found to be less
reactive due to the 2-naphthyl group, which was unfavorable for
nucleophilic attack due to steric hindrance (Table 3, entry 6).
Obviously, the electronic nature of substituents did not show much
influence on the enantioselectivity, and the stereochemical
outcome of the condensation was regulated by the proline residue.
While the nitro group located at ortho position seemed to play an
important role in helping the stereocontrol of the condensation.
The ee value of the 4-hydroxy-4-(2-nitrophenyl)-butan-2-one
was slightly higher than other aldol products even though all
of the aromatic aldehydes afforded the same major enantiomer
of the b-hydroxy ketone (Table 3, entry 2 vs. 1). Therefore,
2-nitrobenzaldehyde was chosen as the model to check the
feasibility of using other ketones as aldol donors in the direct
asymmetric aldol reaction.
Binding of
the -proline/GO hybrid stable in the reaction system. After the
reaction, the heterogeneous -proline/GO catalyst could be facilely
separated from the reaction mixture by centrifugation. The upper
organic phase with the product was separated from the lower cat-
alyst by simple decantation. The recovered catalyst was washed
with ethyl acetate and dried at 40 °C overnight for reused.
Fig. 7 showed the results of the recovery and reusability of the
L-proline to the edge and plane of GO sheet makes
L
L
L
-proline/GO hybrid. To our delight, the L-proline/GO hybrid could
be reused for seven times with no appreciable decrease in yield
and enantioselectivity of aldol product, which demonstrated the
prepared
L-proline/GO catalyst possess excellent stability and
reusability. The reaction was completely stopped by the removal
of the catalyst. Elementary analysis of the recovered catalyst was
used for determining the L-proline leaching in terms of nitrogen
element percentage, and there were no significant changes in the
nitrogen element percentage compared with the fresh catalyst.
The inactivity of supernatants, as well as the maintaining of nitro-
gen element percentage in the L-proline/GO catalyst, indicated that
any leaching of active species into solution was insignificant.
Further evidence of the stability of the heterogeneous catalyst
was provided by the FT-IR spectra of the L-proline/GO hybrid with
fresh and reused for seven times (Fig. 8a vs. a’). No significant
changes of the catalyst took place even after reuse for seven times.
Given the scope of ketone, butanone, cyclohexanone, and cyclo-
pentanone as aldol donors were also investigated into the direct
asymmetric reaction with 2-nitrobenzaldehyde. The results are
summarized in Table 4. The butanone exhibited the similar activity
as the acetone and provided almost quantitative yield of aldol
product (up to 92%), as shown in Table 4 (Table 4, entry 2 vs. 1).
While, different from the acetone, butanone as the aldol donor
produced two different aldol products due to the presence of two
These observations suggested that the efficient L-proline/GO cata-
lyst was perfectly stable during the direct asymmetric aldol reac-
tion presented here and was readily recyclable from the reaction
system. Notably, the recycle of the pristine L-proline resulted in a
significant decrease in activity (from 95% to 65% yield in the second
cycle) with no change of the ee value (79%) due to the problem of
quantitative recovery.
different
a-carbon centers in the butanone. The desired product
of (3R,4S)-3-methyl-4-hydroxy-4-(2-nitrophenyl)-butan-2-one
3.5. The asymmetric aldol reactions of ketone with various aryl
aldehydes
was obtained in only 35% yield with the lower diastereoselectivity
of 59:41 (anti/syn), and the enantioselectivity of the anti-isomer
was 86% (Table 4, entry 2). We also examined the feasibility of
using cyclic ketones, such as cyclohexanone and cyclopentanone,
as aldol donors. Although a longer reaction time was required in
comparison with acyclic ketones, satisfactory results were
obtained. Cyclohexanone reacted with 2-nitrobenzaldehyde to
generate aldol adduct in a high yield (up to 96%) with an excellent
dr of 93:7 (anti/syn). The ee value for the anti-isomer was 92%
(Table 4, entry 3). In the case of cyclopentanone, similarly, high
yield of 93% was isolated within 8 h, while the dr and ee values
of aldol were somewhat lower (Table 4, entry 4 vs. 3).
Application scope of the catalytic system was then examined
under the optimized conditions, and the results are summarized
in Table 3. A wide range of aromatic aldehydes bearing either elec-
tron-rich or electron-deficient substitutes underwent reaction
with acetone to produce the corresponding b-hydroxy ketone in
moderate to high yield and enantioselectivity at room tempera-
ture. The electron nature and steric hindrance of the substituents
on the aromatic aldehydes affected the yield of the corresponding
aldol products dramatically. In general, the reaction between

146
R. Tan et al. / Journal of Catalysis 298 (2013) 138–147
Table 3
The direct asymmetric aldol reaction between acetone and aromatic aldehydes catalyzed by L-proline/GO.a
O
O
OH
Ar
O
GO/L-proline
30 oC
H3C
CH3
H
Ar
Entry
1
Ar
Product
Reaction time (h)
6
Yield (%)b
96
Ee (%)
2-NO2C6H4
79
NO2
OH
O
c
(1) c
2
4-NO2C6H4
6
95
68
O
OH
c
c
NO2
(2)
3
4
2-ClC6H4
4-BrC6H4
12
24
91
66
66
63
Cl
O
O
OH
OH
d
(3) d
d
d
Br
(4)
5
6
4-AcNHC6H4
2-Naphthyl
24
24
43
56
55
55
O
O
OH
OH
d
O
N C CH3
H
(5) d
d
(6) d
a
b
c
Reaction conditions:
The same as Table 2.
Ee was determined by HPLC (Daicel chiralpak OB-H column) on pure products.
Ee was determined by HPLC (Daicel chiralpak AD column) on pure products.
L-proline/GO (0.035 g, 30 mol% of aromatic aldehydes), aromatic aldehydes (0.5 mmol), acetone (4 ml), 30 °C.
d
Table 4
The direct asymmetric aldol reaction between ketones and 2-nitrobenzaldehyde catalyzed by L-proline/GO.a
NO2
O
C
O
OH
R2
O
NO2
GO/L-proline
R1
H
R1
DMF, 30 oC
R2
Entry
1
Product
Reaction time (h)
6
Yield (%)b
96
dr(anti/syn)c
Ee (%)
/
79
NO2
NO2
O
OH
d e
,
(1) d, e
2
3
5
35 (92)g
59:41
93:7
86
92
O
O
O
OH
OH
OH
f
(7) f
(8) h
24
96
NO2
NO2
h
4
8
93
83:17
89
h
(9) h
a
b
c
Reaction conditions:
The same as Table 2.
L-proline/GO (0.035 g, 30 mol% of 2-nitrobenzaldehyde), 2-nitrobenzaldehyde (0.5 mmol), ketone (5 mmol), DMSO (4 ml), 30 °C.
Determined by 1H NMR analysis, major product is anti.
d
e
f
Reaction conditions: L-proline/GO (0.035 g, 30 mol% of aromatic aldehydes), aromatic aldehydes (0.5 mmol), acetone (4 ml), 30 °C.
Ee was determined by HPLC (Daicel chiralpak OB-H column) on pure products.
Ee was determined by HPLC (Daicel chiralpak AD-H column) on the anti-product.
Only the regionisomer as shown was obtained. The total yield of isolated aldol products is given in parentheses.
Ee was determined by HPLC (Daicel chiralpak AD column) on the anti-product.
g
h

R. Tan et al. / Journal of Catalysis 298 (2013) 138–147
147
[12] J. Zou, W. Zhao, R. Li, H. Zhang, Y. Cui, J. Appl. Polym. Sci. 118 (2010) 1020.
[13] A. Lu, T.P. Smart, T.H. Epps, D.A. Longbottom, R.K. ÓReilly, Macromolecules 44
(2011) 7233.
[14] S.J. Bae, S.W. Kim, T. Hyeon, B.M. Kim, Chem. Commun. (2000) 31.
[15] A. Zamboulis, N.J. Rahier, M. Gehringer, X. Cattoën, G. Niel, C. Bied, J.J.E.
Moreau, M.W.C. Man, Tetrahedron: Asymmetry 20 (2009) 2880.
[16] F. Calderón, R. Fernndez, F. Snchez, A. Fernndez-Mayoralas, Adv. Synth. Catal.
347 (2005) 1395.
[17] E.G. Doyagüez, F. Calderon, F. Sanchez, A. Fernndez-Mayoralas, J. Org. Chem. 72
(2007) 9353.
[18] W. Miao, T.H. Chan, Adv. Synth. Catal. 348 (2006) 1711.
[19] S. Luo, X. Mi, L. Zhang, S. Liu, H. Xu, J. Cheng, Angew. Chem. Int. Ed. 45 (2006)
3093.
[20] J. Huang, X. Zhang, D.W. Armstrong, Angew. Chem. Int. Ed. 46 (2007) 9073.
[21] J. Li, G. Yang, Y. Qin, X. Yang, Y. Cui, Tetrahedron: Asymmetry 22 (2011) 613.
[22] H. Yang, S. Li, X. Wang, F. Zhang, X. Zhong, Z. Dong, J. Ma, J. Mol. Catal. A 363–
364 (2012) 404.
[23] Z. An, W. Zhang, H. Shi, J. He, J. Catal. 241 (2006) 319.
[24] Y. Xu, H. Bai, G. Lu, C. Li, G. Shi, J. Am. Chem. Soc. 130 (2008) 5856.
[25] C. Petit, T.J. Bandosz, Adv. Funct. Mater. 20 (2010) 111.
[26] K. Nikolaos, P.E. Solon, S. Evangelia, T. Nikos, Carbon 48 (2010) 854.
[27] X. Hu, L. Mu, J. Wen, Q. Zhou, Carbon 50 (2012) 2772.
[28] A.R. Siamaki, A.E.R.S. Khder, V. Abdelsayed, M.S. El-Shall, B.F. Gupton, J. Catal.
279 (2011) 1.
4. Conclusions
The unique layered GO material was shown to be a suitable sup-
port for the highly efficient loading of
non-covalent method to afford a layered
catalyst for the direct asymmetric aldol reaction. Technologies of
characterization of AFM, XRD, FT-IR, UV–vis, and TGA results con-
L
-proline through a simple
L
-proline/GO hybrid
firmed the loading of
sheet through hydrogen-bonding or/and ionic interaction. The
obtained -proline/GO hybrid provided relevant aldol products in
yield and enantioselectivity comparable to those observed in the
reactions promoted by -proline itself. Furthermore, the heteroge-
L-proline on the two sides and edge of the GO
L
L
neous catalyst was stable in the reaction system and could be eas-
ily recovered from the reaction mixture.
Acknowledgments
The project was financial supported by the National Natural
Science Foundation of China (Grant Nos. 21003044, 20973057),
the Natural Science Foundation of Hunan Province (10JJ6028)
and the Program for Science and Technology Innovative Research
Team in Higher Educational Institutions of Hunan Province.
[29] X. Yang, X. Zhang, Z. Liu, Y. Ma, Y. Huang, Y. Chen, J. Phys. Chem. C 112 (2008)
17554.
[30] R. Nie, J. Wang, L. Wang, Y. Qin, P. Chen, Z. Hou, Carbon 50 (2012) 586.
[31] Y. Chi, S.T. Scroggins, E. Boz, J.M.J. Fréchet, J. Am. Chem. Soc. 130 (2008) 13287.
[32] W.S. Hummers Jr., R.E. Offeman, J. Am. Chem. Soc. 80 (1958) 1339.
[33] N.V. Medhekar, A. Ramasubramaniam, R.S. Ruoff, V.B. Shenoy, ACS Nano. 4
(2010) 2300.
References
[34] Y.C. Si, E.T. Samulski, Nano. Lett. 8 (2008) 1679.
[35] A.J. Patil, J.L. Vickery, T.B. Scott, S. Mann, Adv. Mater. 21 (2009) 3159.
[36] K. Su, T. Pa, Y. Zhang, The Analytic Method of Spectrum, East China University
of Science & Technology Press, Shanghai, 2002. 100.
[37] D. Depan, J. Shah, R.D.K. Misra, Mater. Sci. Eng. C 31 (2011) 1305.
[38] X. Tang, W. Li, Z. Yu, M.A. Rafiee, J. Rafiee, F. Yavari, N. Koratkar, Carbon 49
(2011) 1258.
[39] Q. Zhang, W. Li, T. Kong, R. Su, N. Li, Q. Song, M. Tang, L. Liu, G. Cheng, Carbon,
2012. http://dx.doi.org/10.1016/j.carbon.2012.08.025.
[40] Ö. Reis, S. Eymur, B. Reis, A.S. Demir, Chem. Commun. (2009) 1088.
[41] J.N. Moorthy, S. Saha, Eur. J. Org. Chem. 3864 (2006) 3864.
[42] Z. Tang, Z. Yang, X. Chen, L. Cun, A. Mi, Y. Jiang, L. Gong, J. Am. Chem. Soc. 127
(2005) 9285.
[1] E.R. Jarvo, S.J. Miller, Tetrahedron 58 (2002) 2481.
[2] K. Nakamura, R. Yamanka, T. Matsuda, T. Harada, Tetrahedron: Asymmetry 14
(2003) 2659.
[3] W. Hotz, F. Tanaka, C.F. Barbas III, Acc. Chem. Res. 37 (2004) 580.
[4] S. Mukherjee, J.W. Yang, S. Hoffmann, B. List, Chem. Rev. 107 (2007) 5471.
[5] B. List, R.A. Lerner, C.F. Barbas, J. Am. Chem. Soc. 122 (2000) 2395.
[6] S. Luo, X. Mi, S. Liu, H. Xua, J. Cheng, Chem. Commun. (2006) 3687.
[7] N. Zotova, A. Franzke, A. Armstrong, D.G. Blackmond, J. Am. Chem. Soc. 129
(2007) 15100.
[8] G. Chen, X. Fu, C. Li, C. Wu, Q. Miao, J. Organomet. Chem. 702 (2012) 19.
[9] M. Gruttadauria, F. Giacalone, A.M. Marculescu, R. Notoa, Adv. Synth. Catal. 350
(2008) 1397.
[10] M. Gruttadauria, A.M.P. Salvo, F. Giacalone, P. Agrigento, R. Noto, Eur. J. Org.
Chem. 2009 (2009) 5437.
[43] D. Gryko, R. Lipi’nski, Adv. Synth. Catal. 347 (2005) 1948.
[44] Y. Zhou, Z. Shan, J. Org. Chem. 71 (2006) 9510.
[45] H. Bai, C. Li, G. Shi, Adv. Mater. 23 (2011) 1089.
[11] S. Calogero, D. Lanari, M. Orrù, O. Piermatti, F. Pizzo, L. Vaccaro, J. Catal. 282
(2011) 112.