Synthesis of a self-assembling gold nanoparticle-supported organocatalyst for enamine-based asymmetric aldol reactions

Article abstract of DOI:10.1016/j.tet.2016.02.065

The self-assembling gold nanoparticle (GNP)-supported l-proline derivative, which was readily synthesized in four steps from 4-hydroxy-l-proline and chloroauric acid, was used for enamine-based aldol reactions. The modified Brust-Schiffrin (BS) synthesis

Full text of DOI:10.1016/j.tet.2016.02.065

Tetrahedron 72 (2016) 1984e1990
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Synthesis of a self-assembling gold nanoparticle-supported
organocatalyst for enamine-based asymmetric aldol reactions
P eꢀ ter Lajos S oꢀ ti a,b, Hiroki Yamashita , Kohei Sato , Tetsuo Narumi c, Mitsuo Toda b,
b
b
b,
b,
c
a
b,c,d,
*
Naoharu Watanabe , Gy o€ rgy Marosi , Nobuyuki Mase
Department of Organic Chemistry and Technology, Budapest University of Technology and Economics, Budafoki ut 8, Budapest, H-1111, Hungary
Applied Chemistry and Biochemical Engineering Course, Department of Engineering, Graduate School of Integrated Science and Technology, Shizuoka
a
b
University, 3-5-1 Johoku, Hamamatsu, Shizuoka, 432-8561, Japan
c
Graduate School of Science and Technology, Shizuoka University, 432-8561, Japan
Green Energy Research Division, Research Institute of Green Science and Technology, Shizuoka University, 432-8561, Japan
d
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 6 January 2016
Received in revised form 17 February 2016
Accepted 27 February 2016
Available online 3 March 2016
The self-assembling gold nanoparticle (GNP)-supported L-proline derivative, which was readily syn-
thesized in four steps from 4-hydroxy-L-proline and chloroauric acid, was used for enamine-based aldol
reactions. The modified Brust-Schiffrin (BS) synthesis was favored over the ligand exchange reaction in
order to develop a new and simple synthetic pathway for nanoparticle-supported catalyst. Use of im-
mobilized organocatalyst on nanoscale solid carrier led to excellent selectivities (up to 94:6 dr and 94%
ee) in asymmetric reactions of ketones and benzaldehydes. GNPs operate under pseudo-homogeneous
conditions which are easily recycled and reused at least five times without significant loss of weight,
activity, diastereo- and enantioselectivities.
Keywords:
Organocatalyst
Supported catalysis
Nanoparticle
Ó 2016 Elsevier Ltd. All rights reserved.
Aldol reaction
1. Introduction
Organocatalysts are widely applied for asymmetric reactions in
organic synthesis.1 Proline and proline-based derivatives repre-
sent organocatalysts that show high catalytic activity and stereo-
selectivity in asymmetric CeC bond forming reactions such as the
,2
3
4
5
6
aldol, Mannich, Michael, and DielseAlder reactions via enamine
7
and iminium intermediates. Unfortunately, high catalyst loadings
up to 30 mol %) are required in many cases;
3
e7
(
therefore, recycling
and reuse of the active catalyst is required to ensure the
environment-friendly character of these catalytic processes. In
general, problems related to catalyst recycling and reuse can be
8
e10
solved by immobilization on a solid carrier
progress has been made over recent years in this area;
ever, the catalytic reaction becomes heterogeneous and loses the
and impressive
1
1e13
how-
14
benefits of homogeneous phase conditions. If the catalyst is
immobilized on a nanoscale solid carrier, the benefits of both het-
erogeneous and homogeneous catalytic reactions can be combined
Fig. 1. GNP-supported L-proline derivatives as organocatalysts.
15,16
Recently, magnetic nanoparticle-supported organocatalysts
were reported; however, the use of a magnetic nanocore as
a carrier requires several synthetic steps to immobilize the cat-
(Fig. 1).
17e20
alyst.
Thus, our attention turned to the BrusteSchiffrin (BS)
synthesis, a well-known bottom-up method to prepare ligand-
*
Corresponding author. Tel./fax: þ81 53 478 1196; e-mail address: mase.
21e24
nobuyuki@shizuoka.ac.jp (N. Mase).
protected metal core nanoparticles.
This method allows the
http://dx.doi.org/10.1016/j.tet.2016.02.065
0
040-4020/Ó 2016 Elsevier Ltd. All rights reserved.

P.L. S oꢀ ti et al. / Tetrahedron 72 (2016) 1984e1990
1985
formation of metal nanocore and complete capping of metal
nanocores with a catalytically active component in a single syn-
thetic step. Very recently, Santacruz et al. reported the prepara-
tion and application of a gold nanoparticle (GNP)-supported
cinchonine derivative. This supported catalyst showed moder-
ate selectivity during recycling similar to an earlier study by
Malkov et al.,26 which was explained by the loss of the catalyst
from the support. Khiar et al. also reported the preparation and
application of GNP-supported organocatalyst though the yield of
With key intermediate 3 in hand, we studied the self-assembly
of the GNP-supported catalyst; the results are summarized in Table
1. The catalyst loading of the final product strongly depended on
the modified BS synthesis conditions. We found that the amount of
immobilized catalyst could be improved from 0.50 to
2
5
ꢀ1
0.95 mmol g
borohydride.
by increasing the addition time of sodium
aldol product dropped significantly to 30% after the third recy-
Table 1
cling run.27 To the best of our knowledge, there is no report on
Self-assembly of GNP-supported organocatalysts
Entry Catalyst NaBH4
Addition
time (min) (mmol g
Catalyst loading Particle size (nm)b
complete capping of metal nanocores with an organocatalytically
active component in a single synthetic step. Therefore, we aimed
to synthesize a GNP-supported organocatalyst by a modified one-
phase BS method (single synthetic step) instead of ligand ex-
change to overcome the mentioned disadvantages of nano-
particles supported catalyst.
ꢀ
1 a
)
1
2
3
4a
4b
4c
Solid
Solution 30
Solution 60
<1
0.50
0.71
0.95
d
w3
w3
a
Determined by elemental analysis based on the nitrogen atom wt % (nitrogen as
an indicator atom for the proline moiety on the Au core).
b
Determined based on TEM analysis.
2
. Results and discussion
Using the one-phase BS method, Au(III) ions are first reduced to
Transmission electron microscopy (TEM) showed that the par-
ticles have a spherical shape with a diameter of about 3 nm, and
their physical properties are independent of the preparation pro-
cess (Fig. 2). Aggregation of the particles was observed in the case of
2
8
Au(I) by thiols, which leads to the formation of [Au(I)-SR]
n
.
To
avoid the formation of undesired metal thiolate polymer that
cannot be reduced completely to the thiolate monolayer-protected
2
9
gold core by sodium borohydride, disulfide derivate 3 was used as
the self-assembling component instead of the thiol. Therefore, the
first step for the preparation of GNPs was protection of the thiol
4b with a lower catalyst loading (Fig. 2, left image). Moreover, 4a
could not be dispersed in the DMSO/cyclohexanone medium used
to prepare the colloids for TEM investigations. The dispersion of 4c
showed randomly distributed and well-dispersed single particles
3
0
functional group by oxidation, which led to the formation of
a disulfide bond between two molecules of thiol 1 (Scheme 1). The
resulting bifunctional carboxylic acid 2 was converted to 11,11ʹ-
disulfanediyldiundecanoyl chloride with oxalyl chloride, and sub-
without any aggregates (see Fig.
Supplementary data).
2 right image and the
The colloidal properties of ligand-protected Au clusters were
examined by UV/VIS spectroscopy to gain further information
sequently used for chemoselective O-acylation of 4-hydroxy-L-
proline (5) in an acidic medium without isolation to prevent un-
32
about the properties of 4b and 4c. A surface plasmon (SP) band
wanted N-acylation.3
1
appeared at 512 nm in 4c, with the highest catalyst loading, in
contrast to 4b, where an SP band was undetectable (Fig. 3). This
result is consistent with our prior assumption, i.e., the proline
content promotes the dispersion of the supported catalyst in
a medium.
IR spectrometry was used to confirm the chemical structure of
the supported catalyst. According to the IR spectra, the functional
group peaks of the GNPs 4c correspond to O-lauroyl-trans-4-
3
3
hydroxy-
L
-proline (native catalyst, 6), where typical bands for
ꢀ1
ꢀ1
the free amine (1611 cm ), carbonyl (1732 cm ), and alkyl groups
2850 and 2917 cm ) were observed (Fig. 4).
ꢀ
1
(
To demonstrate the catalytic activity of the GNP-supported
catalyst, 4c was investigated in asymmetric aldol reactions
Table 2). GNP 4c was easily and finely dispersed in the reaction
(
mixture (Fig. 3-b), and showed high activity. In most cases, the
reactions afforded the anti-aldol products in high yields with high
to excellent diastereo- and enantioselectivities (Table 2). Com-
pounds 9a-h were obtained with good results up to 99% yield and
8
9% ee. As shown in Table 2, the substituents on the benzaldehyde
can greatly influence the reactivities. A comparison of entries 1 and
indicates that the reaction time can be reduced by electron-
7
withdrawing groups, with an associated increase in enantiose-
lectivity from 77% to 88% ee. Using p-methoxy benzaldehyde as an
acceptor, the immobilized organocatalyst afforded aldol product 9f
with 85:15 dr and 78% ee. Therefore, the supported catalyst toler-
ates both electron-withdrawing and electron-donating sub-
stituents in the direct asymmetric aldol reaction, which is
3
consistent with the unsupported catalyst.
The recyclability of the supported catalyst was examined for
1e5 cycles in the direct aldol reaction between 7a and 8a (Table 3).
After 24 h, a fine dispersion of nanoparticles was obtained (Fig. 3-
b), and the GNPs were recovered by changing the dielectric
Scheme 1. Synthesis of GNP-supported and O-lauroyl-trans-4-hydroxy-L-proline
catalysts.

1986
P.L. S oꢀ ti et al. / Tetrahedron 72 (2016) 1984e1990
Fig. 2. TEM images of GNPs 4b (left) and 4c (right).
Table 2
GNP-supported proline-catalyzed asymmetric aldol reactions
Entry
n
X
Time (h)
Yield (%)a
drb
ee (%)c
Product
1
2
3
4
5
6
7
8
1
1
1
1
1
1
1
0
NO
Cl
Br
2
24
72
72
48
72
166
120
5
97
99
90
99
99
23
65
98
90:10
91:9
88
89
87
83
86
78
77
78
9a
9b
9c
9d
9e
9f
90:10
90:10
90:10
85:15
91:9
CN
CO
2
Me
d
OMe
H
9g
9h
NO
2
35:65
a
Isolated yield.
Determined by 1H NMR (anti:syn).
Enantiomeric excesses of the anti-products were determined by chiral HPLC
b
c
analysis.
d
Conversion of the aldehyde was 31% based on 1H NMR.
Fig. 3. Colloidal behavior of supported catalyst: UV/VIS spectra of 4b and 4c; Images of
reaction mixtures: (a) at 0 h, (b) after 24 h, (c) after addition of EtOAc and (d) following
2
min of storage.
Table 3
Recycling of GNP-supported proline catalyst 4c
Cycles
Conversion (%)a
Yield (%)b
drc
ee (%)d
Recovery (%)
1
2
3
4
5
98
99
99
99
99
97
98
98
98
98
90:10
91:9
91:9
92:8
92:8
88
88
88
88
88
ꢁ99
ꢁ99
ꢁ99
ꢁ99
ꢁ99
a
Determined by 1H NMR.
Isolated yield.
b
c
Determined by 1H NMR (anti:syn).
d
Enantiomeric excesses of the anti-products were determined by chiral HPLC
Fig. 4. IR spectra of 4c and 6.
analysis.

P.L. S oꢀ ti et al. / Tetrahedron 72 (2016) 1984e1990
1987
constant of the reaction mixture by the addition of EtOAc. Based on
70
the colloidal behavior, the GNPs aggregated immediately (Fig. 3-c)
and the reaction mixtures became heterogeneous, thus facilitating
the separation of immobilized organocatalyst and aldol product.
The nanoparticles could be recovered by decantation after centri-
fugation. Consequently, while 4c has the physical properties of
a heterogeneous catalyst, during the enamine cycle it operates in
a pseudo-homogenous phase due to the formation of more soluble
enamine intermediates on the gold core in organic solvents (see
Graphical abstract).
As shown in Table 3, the reaction times, yields, conversions, and
stereoselectivities were unchanged during the recycling process,
which indicates the high efficiency of our supported catalyst 4c. The
unchanged reaction times (exactly 24 h) point out the unchanged
catalytic activities which presumably rule out the possibility of any
aggregation of nanoparticles in the reaction mixture.
6
0
0
5
40
5
30
20
10
0
6
4c
0
5
10
15
20
25
Time (h)
Fig. 5. Comparison of unsupported catalysts (5 and 6) and GNP catalyst (4c) in the
The recovery of the GNP 4c was determined by weight mea-
surement and residual metal content of the reaction mixture.
According to inductively coupled plasma optical emission spec-
troscopy (ICP-OES) measurements, the residual gold was less than
aldol reaction (see the reaction conditions in Table 4 entry 6, 7, and 8).
Next, the substrate concentrations of aldol reaction were de-
creased from 1 M to 0.05 M in order to investigate its effect for the
asymmetric induction (Table 4, entries 6e8). Surprisingly, the
enantioselectivities of catalysts 5 and 6 were lower (74% and 76% ee,
respectively) than that of the GNP (89% ee) at 0.05M, namely the
stereoselectivities are influenced by the substrate concentration in
the case of the unsupported catalysts (Table 4, entry 1, 2, 6, and 7),
but the stereoselectivities of the immobilized catalyst 4c were
unchanged (Table 4, entry 4 and 8).
2
mg (<0.03%) (see Supplementary data).
To investigate the stereoselectivity of the supported-catalyst, we
compared it with native catalyst 5 (4-hydroxyl-L-proline) and 6 (O-
lauroyl-trans-4-hydroxy- -proline), in the direct aldol reaction be-
L
tween 7a and 8a (Table 4). In the case of catalysts 5 and 6, the
enantioselectivities were 94% and 95%, respectively (entries 1 and
2
). With supported catalyst 4c, increasing the loading to 20 mol %
improved the dr and ee of aldol product 9a from 90:10 to 94:6 and
from 88% to 94%, respectively, when the reactions were performed
These results could be explained by the neighbor interactions of
nearby catalyst moieties on the nanoscale carrier. Considering the
one-proline-mechanism of proline-catalyzed aldolization, it is
ꢂ
at 0 C (entries 4 and 5). Based on these results, similar stereo-
selectivities can be achieved by varying the reaction conditions of
the immobilized catalyst 4c. On the other hand, GNP-supported
catalyst 4b with less catalyst loading showed lower diastereo-
and enantioselectivities (entry 3).
34
due to the neighboring long alkyl chain of catalyst species which
35,36
could provide ‘reactive environment’
for the substrates and for
the intermediates on the gold nanocore hence the molecular en-
vironment of substrates does not change by dilution. While in the
case of unsupported catalysts, the molecular environment of in-
termediates and catalyst molecules changed by dilution resulting
decreased enantioselectivities.
Table 4
Comparison of unsupported catalysts (5 and 6) and GNP catalyst in the aldol reaction
Our ‘reactive environment’ is also supported by the results of 4b
ꢀ1
(
0.71 mmol g catalyst loading of GNP) catalyzed aldol reaction. As
shown in Table 4 (entry 3 and 4), decreasing the number of catalyst
ꢀ
1
species from 0.95 to 0.71 mmol g on the gold nanocore (in other
words: change the reactive environment and size of reactive
pocket) led to lower enantioselectivity (from 88% to 73% ee). This
effect is similar to the catalytic mechanism of enzymatic trans-
formations in the biological systems, i.e., our GNP can be inter-
preted as an artificial enzyme where one of the catalytically active
molecules on the surface of nanocore seems to be an active center
of the enzyme (see Fig. 1).
Entry Catalyst Concentration (M) Time (h) Yield (%)a drb
ee (%)c
86:14 94
92:8 95
1
2
3
4
5
6
7
8
5
6
4b
4c
4c
5
1
1
1
1
16
16
24
24
40
72
72
90
97
98
98
97
97
97
98
98
83:17 73
90:10 88
94:6
95:5
97:3
92:8
d
1
94
74
76
89
0.05
0.05
0.05
6
4c
3. Conclusion
a
b
c
Isolated yield.
Determined by 1H NMR (anti:syn).
Enantiomeric excesses of the anti-products were determined by chiral HPLC
Proline-functioned GNPs were synthesized and applied in the
asymmetric aldol reaction of benzaldehyde derivatives and cyclic
donors, resulting in aldol products with high to excellent stereo-
selectivities. The recycling efficiency of our nanoparticle-supported
catalyst was excellent, without significant loss of weight, activity,
and selectivity. These results pointed out that the preparation
processes strongly influence it. In addition, the supported catalyst’s
activity is comparable with that of the native counterparts, and the
enantioselectivity of the immobilized catalyst is independent on
the substrate concentration due to the unchanged molecular en-
vironment of substrates. Although the gold as a carrier material is
slightly expensive; the preparation of GNP-supported proline re-
quired only four easy and simple synthetic steps while magnetic
analysis.
d
ꢂ
The reaction was carried out at 0 C with 20 mol % catalyst.
We then compared the reactivities of immobilized 4c and un-
supported catalysts (5, 6) in the aldol reaction progress of 7a and
a. Conversions were determined by monitoring the reaction
mixtures by H NMR. The conversion versus reaction time diagram
shows similar reaction progress (at 12 h w30% conversion), al-
though the catalytic activity of 4c is slightly lower than that of 5 and
8
1
6
after 12 h (Fig. 5).

1988
P.L. S oꢀ ti et al. / Tetrahedron 72 (2016) 1984e1990
nanoparticle- or polymer-supported organocatalysts require many
synthetic steps and longer synthetic process which tend to raise
their cost. Furthermore, the recycling study showed that the re-
covery of GNPs is excellent and not required any special processes.
Considering all these advantages of GNPs, our new synthetic
pathway for the immobilization of catalyst can promote an alter-
native route in the field of supported catalysis.
53%). Elemental analysis: C: 20.81%, H: 2.84%, N: 1.33%, S: 3.71%; IR
(neat, cm ): 633, 1163, 1410, 1617, 1709, 2847, 2917.
ꢀ
1
Above mentioned process was used for preparation of 4a (yield:
206 mg, 48%) and 4b (yield: 210 mg, 49%) but in case of 4a, solid
sodium borohydride was added to the reaction mixture in one
portion and in case of 4b, the dropping time of sodium borohydride
solution was 30 min.
Elemental analysis (4a): C: 10.43%, H: 1.42%, N: 0.70%, S: 1.95%;
Elemental analysis (4b): C: 20.19%, H: 2.83%, N: 0.99%, S: 4.07%.
4
4
4
. Experimental
4
.1.4. Preparation and characterization of O-lauroyl-trans-4-
31
.1. Synthesis of GNP-supported catalyst
hydroxy- -proline (6). Trans-4-Hydroxy-L-proline (5, 500 mg,
3
argon atmosphere. To the solution, lauroyl chloride (1.7 mL,
7.63 mmol) was added in one portion and the solution was stirred
L
.81 mmol) was dissolved in TFA (2.1 mL) at room temperature in
0
30,37
.1.1. 11,11 -Dithiobis(undecanoic acid) (2).
To a suspension of
O (540 mL)
11-mercaptoundecanoic acid (1.15 g, 5.00 mmol) in H
2
was added sodium hydroxide (2.80 g, 67.9 mmol) and hydrogen
peroxide solution (30 wt %, 11 mL) was filled to a well-stirred clear
solution. After being stirred for 90 min, concentrated HCl (37 wt %,
at room temperature for 2 h Et
with an ice/water bath to give a fine white precipitate which was
vacuum filtered, washed with Et O (5 mL) and dried in vacuum at
2
O (7.5 mL) was added under cooling
2
1
(
(
1 mL) was added and the mixture was extracted with EtOAc
3ꢃ150 mL). The organic solution was washed with distilled water
2ꢃ50 mL), saturated aqueous NaCl (50 mL) and dried over Na SO
room temperature. The hydrochloride salt of 6 was dissolved in
boiling MeOH (2 mL) and hot MTBE (3 mL) was added to the boiling
solution which was left to crystallize for 5 h at room temperature.
The crystals were vacuum filtered and dried in vacuum at room
temperature. The recrystallized hydrochloride salt (100 mg,
2
4
.
After removing the solvent, disulfide 2 was obtained as a white
ꢂ
1
solid (945 mg, 87%). Mp 90.8e92.2 C; H NMR (300 MHz, DMSO-
1.18e1.34 (m, 24H), 1.43e1.52 (m, 4H), 1.54e1.66 (m, 4H), 2.18
d
6
)
d
2 3
0.29 mmol) was dissolved in 10% K CO solution (2 mL) stirred for
10 min and AcOH was added to adjust pH to 7. The solid precipitate
13
(
d
t, J¼7.3 Hz, 4H), 2.69 (t, J¼7.2 Hz, 4H); C NMR (75 MHz, DMSO-d
6
)
24.5, 27.8, 28.6, 28.6, 28.8, 28.9, 29.9, 33.7, 38.0, 174.6.
was filtered and recrystallized from MeOH (1 mL) to give 6 as white
ꢂ
19
needle shape crystals (56 mg, 62%). Mp 185.4e187.4 C (dec); [
a]
D
0
0
0
0
ꢀ1
4
.1.2. (2S,2 S,4R,4 R)-4,4 -((11,11 -Disulfanediylbis(undecanoyl))bi-
ꢀ17.9 (c 0.09, MeOH); IR (neat, cm ): 633, 718, 1161, 1581, 1612,
1
s(oxy))bis(2-carboxypyrrolidin-1-ium) chloride (3). The disulfide 2
3
1733, 2850, 2917 (see Fig. 4); H NMR (300 MHz, CF COOD) d 0.76 (t,
(
(
2
300 mg, 0.69 mmol) was suspended in dichloromethane
3.1 mL) in argon atmosphere and oxalyl chloride (0.2 mL,
.44 mmol) was dropped to a well-stirred reaction mixture at
J¼6.5 Hz, 3H), 1.12e1.28 (m, 16H), 1.52e1.69 (m, 2H), 2.42 (t,
J¼7.7 Hz, 2H), 2.52e2.65 (m, 1H), 2.74e2.88 (m, 1H), 3.81e3.96 (m,
13
2H) 4.85e4.95 (m, 1H), 5.78e5.60 (m, 1H); C NMR (75 MHz,
room temperature. A resulted peal yellow but clear solution was
evaporated by vacuum pump after 3 h reaction time, keeping in
3
CF COOD) d 14.7, 24.4, 26.7, 31.0, 31.0, 31.3, 31.3, 31.5, 31.5, 33.9,
36.9, 54.9, 61.6, 75.8, 177.7, 180.5.
argon atmosphere. Previously recrystallized trans-4-hydroxy-L-
proline (181 mg, 1.38 mmol) was dissolved in TFA (3.1 mL) by
stirring at room temperature in argon atmosphere. The solution
was cooled just below room temperature with an ice/water bath
and it was added by a syringe in one portion for acid chloride.
The reaction mixture was stirred at room temperature for 24 h.
4.2. General method for aldol reaction using GNP supported
proline catalyst
Benzaldehyde derivatives 8 (0.06 mmol), GNPs, aldol donor 7
Under cooling with an ice/water bath, Et
carefully to give a russet sticky solid which was decanted,
washed with Et
O (3ꢃ10 mL) and dried in vacuum for 48 h to
2
O (20 mL) was added
(10 equiv), DMSO:water (94:6, 65 mL) solvent mixture and magnetic
stir bar were added to a 2 mL capped glass vial and it was stirred at
ꢂ
2
25 or 0 C. Conversion was monitored by TLC. During the working
give proline derivative 3 as a russet solid (372 mg, 73%). Mp
up, the reaction mixture was diluted with EtOAc (1 mL) and the
liquid phase was removed by decantation after 2 min of centrifu-
gation at 4000 rpm. The solid residue was suspended in EtOAc
(1 mL) and the previous procedure was used for separation of the
liquid phase after 10 min of stirring, and this procedure was re-
peated twice. The combined organic solutions were filtered
through a sort silica pad, washed with distilled water (3 mL), dried
ꢂ
19
ꢀ1
7
7
d
4.4e77.7 C (dec); [
a
]
D
ꢀ10.2 (c 0.1, MeOH); IR (neat, cm ):
1
20, 1154, 1210, 1745, 2850, 2917; H NMR (300 MHz, CD
3
OD)
1.26e1.41 (m, 24H), 1.57e1.73 (m, 8H), 2.34e2.48 (m, 6H),
.53e2.64 (m, 2H), 2.68 (t, J¼7.5 Hz, 4H), 3.42e3.55 (m, 2H), 3.68
2
(
(
dd, J¼13.5, 4.5 Hz, 2H), 4.57 (dd, J¼10.5, 7.5 Hz, 2H), 5.39e5.49
m, 2H); 13C NMR (75 MHz, DMSO-d
8.7, 28.8, 28.9, 33.4, 34.2, 37.9, 50.2, 53.1, 57.7, 68.6, 72.1, 169.6,
6
) d 24.1, 27.7, 28.4, 28.5, 28.6,
2
2 4
over Na SO , and the solvent was evaporated. The diastereomeric
1
1
70.4, 172.6, 174.6; HRMS (ESI, m/z): calcd for C32
H
56
N
2
O
8
S
2
ratios of the crude products (9a-h) were determined by H NMR
after evaporation. The enantiomeric excesses of anti-products 9b-h
were determined by chiral HPLC analysis after purification of the
crude products by preparative TLC. For preparative TLC, hex-
aneeethyl acetate solvent mixtures were used as eluent. In the case
of 9a, the crude product was used directly for chiral HPLC analysis.
The absolute configuration of aldol products 9 were determined by
þ
(
MþH ) 661.35563, found 661.35649.
4.1.3. Preparation of GNP supported catalyst 4c. A 500 mL eggplant-
shaped flask, which was flushed argon gas previously, was charged
with gold(III) chloride$4H O (338 mg, 0.82 mmol), disulfide de-
rivative of proline 3 (300 mg, 0.41 mmol) and dehydrated DMSO
410 mL) was added, keeping in argon atmosphere. To the clear
bright yellow solution, sodium borohydride solution (169 mg,
.10 mmol in MeOH (14.4 mL)) was dropped over 1 h and the
2
3
8
(
comparison of the HPLC-data with our previous results.
4
resulted black suspension was stirred at room temperature for 24 h.
During the working up, the black solid component was decanted,
washed with EtOH (2ꢃ15 mL), AcOH (2ꢃ10 mL), EtOH (6ꢃ15 mL),
4.3. Recycling of GNPs
General method was used during the recycling. The supported
catalyst 4c was reused after overnight drying in vacuum at room
temperature without further treatment.
Et
2
O (1ꢃ15 mL), respectively, and dried in vacuum at room tem-
perature to give GNP supported catalyst 4c as a black solid (228 mg,

P.L. S oꢀ ti et al. / Tetrahedron 72 (2016) 1984e1990
1989
4
.4. Characterization of aldol products
55.4, 57.6, 74.4, 114.0, 128.4, 133.4,159.5, 216.0. Enantiomeric excess
of anti-product: 78%, determined by HPLC (Daicel Chiralpak AD-H,
1
4
.4.1. (S)-2-((R)-Hydroxy(4-nitrophenyl)methyl)cyclohexanone
Hexane/i-PrOH¼90/10), UV
l
¼254 nm, flow rate 0.5 mL minꢀ , t
R
3
8
0
0
(
9a). Yield: 14.5 mg, 97%; diastereomer ratio 94:6, determined by
H NMR of crude product. Colorless solid; mp 129e130 C; [
(2S, 1 R) 39.7 min, t (2R,1 S) 41.0 min.
R
1
ꢂ
19
a
]
D
1
þ6.3 (c 0.1, CHCl
3
); H NMR (300 MHz, CDCl
3
)
d
1.18e1.91 (m, 5H),
4.4.7. (S)-2-((R)-Hydroxy(phenyl)methyl)cyclohexanone
(9g). Yield: 8.0 mg, 65%; diastereomer ratio 91:9, determined by
3
8
1
.97e2.23 (m, 1H), 2.25e2.69 (m, 3H), 4.08 (s, 1H), 4.90 (d, J¼8.4 Hz,
13
1
ꢂ
19
1
H), 7.51 (d, J¼8.8 Hz, 2H), 8.22 (d, J¼8.8 Hz, 2H); C NMR (75 MHz,
24.8, 27.7, 30.8, 42.8, 57.3, 74.2, 123.8, 128.1, 148.6, 215.1.
Enantiomeric excess of anti-product: 94%, determined by HPLC
Daicel Chiralpak IB, Hexane/i-PrOH¼90/10), UV ¼254 nm, flow
H NMR of crude product. Colorless solid; mp 99e101 C; [
a
]
D
1
CDCl
3
)
d
þ11.7 (c 0.1, CHCl
3
); H NMR (300 MHz, CDCl
3
)
d
1.08e1.90 (m, 5H),
1.98e2.15 (m, 1H), 2.26e2.71 (m, 3H), 3.95 (s, 1H),4.01 (s, 1H), 4.79
13
(
l
(d, J¼8.9 Hz, 1H), 7.19e7.70 (m, 5H); C NMR (75 MHz, CDCl
24.8, 27.9, 30.9, 42.8, 57.6, 74.9, 127.3, 128.1, 128.6, 141.2, 215.9.
Enantiomeric excess of anti-product: 77%, determined by HPLC
(Daicel Chiralpak AS-H, Hexane/i-PrOH¼90/10), UV ¼254 nm,
0
R R
flow rate 0.5 mL min , t (2S, 1 R) 26.0 min, t (2R,1 S) 28.2 min.
3
)
ꢀ
1
0
0
rate 0.5 mL min , t
R
(2S, 1 R) 28.5 min, t
R
(2R, 1 S) 33.5 min.
d
4.4.2. (S)-2-((R)-(4-Chlorophenyl) (hydroxy)methyl)cyclohexanone
l
3
8
ꢀ1
0
(
9b). Yield: 14.2 mg, 99%; diastereomer ratio 91:9, determined by
H NMR of crude product. Colorless solid; mp 95e96 C; [
1
ꢂ
19
a
]
D
þ22.5
1
(c 0.1, CHCl
3
); H NMR (300 MHz, CDCl
3
)
d
1.20e1.80 (m, 5H),
4.4.8. (S)-2-((R)-Hydroxy(4-nitrophenyl)methyl)cyclopentanone
(9h). Yield: 13.8 mg, 98%; diastereomer ratio 35:65, determined
by H NMR of crude product. Anti-product: colorless solid; mp
3
8
2
1
4
.00e2.12 (m, 1H), 2.24e2.60 (m, 3H), 3.99 (s, 1H), 4.74 (d, J¼8.7 Hz,
13
1
H), 7.20e7.33 (m, 4H); C NMR (75 MHz, CDCl
3
)
d
24.8, 27.8, 30.8,
ꢂ
19
1
2.8, 57.5, 74.3, 128.6, 128.8, 133.8, 139.7, 215.7. Enantiomeric excess
87e89 C; [
1.45e1.85 (m, 3H), 1.95e2.09 (m, 1H), 2.19e2.55 (m, 3H), 4.75 (br
s,1H), 4.85 (d, J¼9.1 Hz,1H), 7.54 (d, J¼8.9 Hz, 2H), 8.22 (d, J¼8.9 Hz,
a
]
D
ꢀ21.3 (c 0.1, CHCl
3 3
); H NMR (300 MHz, CDCl )
of anti-product: 89%, determined by HPLC (Daicel Chiralpak AD-H,
d
¼254 nm, flow rate 0.5 mL minꢀ , t
1
Hexane/i-PrOH¼90/10), UV
l
R
0
0
13
(
2S, 1 R) 27.0 min, t
R
(2R,1 S) 31.3 min.
2H); C NMR (75 MHz, CDCl
3
) d 20.4, 26.9, 38.7, 55.2, 74.6, 123.9,
127.6, 147.9, 148.9, 222.5. Enantiomeric excess of anti-product: 78%,
4
.4.3. (S)-2-((R)-(4-Bromophenyl) (hydroxy)methyl)cyclohexanone
determined by HPLC (Daicel Chiralpak AD-H, Hexane/i-PrOH¼95/
3
8
ꢀ1
0
R R
(2S, 1 R) 90.5 min, t
(9c). Yield: 15.3 mg, 90%; diastereomer ratio 90:10, determined
5), UV
l
¼254 nm, flow rate 0.5 mL min , t
1
ꢂ
19
0
by H NMR of crude product. Colorless solid; mp 82e83 C; [
a
]
D
(2R,1 S) 94.3 min.
1
þ17.1 (c 0.1, CHCl
.03e2.13 (m, 1H), 2.31e2.60 (m, 3H), 3.97 (d, J¼2.7 Hz, 1H), 4.75
dd, J¼8.7, 2.3 Hz, 1H), 7.20 (d, J¼8.3 Hz, 2H), 7.48 (d, J¼8.3 Hz, 2H);
3 3
); H NMR (300 MHz, CDCl ) d 1.22e1.77 (m, 5H),
2
4.5. Residual gold determination
(
1
3
C NMR (75 MHz, CDCl
3
)
d
24.8, 27.8, 30.8, 42.8, 57.4, 74.4, 121.9,
4-Nitrobenzaldehyde (8a, 10.0 mg, 0.06 mmol) and GNPs 4c
(6.8 mg, 10 mol % catalyst related to benzaldehyde) or O-lauroyl-
129.0, 131.7, 140.3, 215.6. Enantiomeric excess of anti-product: 87%,
determined by HPLC (Daicel Chiralpak AD-H, Hexane/i-PrOH¼90/
trans-4-hydroxy-
a 2 mL capped glass vial. The cyclohexanone 7a (69 m
DMSO/water (94/6) solvent mixture (65
L
-proline 6 (2.0 mg, 0.006 mmol) were filled in
¼254 nm, flow rate 0.5 mL minꢀ , t
1
(2S, 1 R) 29.2 min, t
0
L, 0.60 mmol),
10), UV
l
R
R
0
(
2R,1 S) 34.3 min.
m
L) and magnetic stir bar
ꢂ
were added to the solid materials and stirred at 25 C for 24 h. In case
of GNP catalyst, the reaction mixture was diluted with EtOAc (1 mL)
in order to aggregate GNPs and liquid phase was removed by de-
cantation after 2 min of centrifugation at 4000 rpm. The solid resi-
due was suspended in EtOAc (1 mL) and the previous procedure was
used for separation of the liquid phase after 10 min of stirring, and
this procedure was repeated twice. The combined organic solution
or the reaction mixture (in case of using O-lauroyl-trans-4-hydroxy-
4
.4.4. 4-((R)-Hydroxy((S)-2-oxocyclohexyl)methyl)benzonitrile
3
8
(
9d). Yield: 13.6 mg, 99%; diastereomer ratio 90:10, determined by
H NMR of crude product. Colorless solid; mp 82e83 C; [
1
ꢂ
19
a]
D
þ19.5 (c
1
0
.1, CHCl
3
); H NMR (300 MHz, CDCl
3
)
d
1.18e1.93 (m, 5H), 1.99e2.19
(
m,1H), 2.25e2.69 (m, 3H), 4.04 (s,1H), 4.84 (d, J¼8.7 Hz,1H), 7.45 (d,
13
J¼8.1 Hz, 2H), 7.65 (d, J¼8.1 Hz, 2H); C NMR (75 MHz, CDCl
7.7, 30.8, 42.8, 57.3, 74.4, 111.9, 118.9, 128.0, 132.4, 146.6, 215.2. En-
antiomeric excess of anti-product: 86%, determined by HPLC (Daicel
3
) d 24.8,
2
L-proline 6 as a catalyst) was evaporated, solid residue was treated
Chiralpak AD-H, Hexane/i-PrOH¼90/10), UV
l
¼254 nm, flow rate
0
R R
.5 mL min , t (2S, 1 R) 46.7 min, t (2R,1 S) 58.9 min.
with 0.5 mL aqua regia, diluted with methylene chloride (2 mL) and
it was washed with distilled water (10ꢃ1.5 mL). The combined water
phase was transferred to a volumetric flask and made up to 20.0 mL
with distilled water which was used for ICP-OES measurement. For
the ICP-OES calibration, 0.001, 0.01, 0.1, 1.0 and 5.0 mg L gold(III)
chloride standard solutions were used and emission intensity
measured at 267.595 nm (see Supplementary data).
ꢀ
1
0
0
4
.4.5. Methyl 4-((R)-hydroxy((S)-2-oxocyclohexyl)methyl)benzoate
3
8
ꢀ1
(9e). Yield: 15.6 mg, 99%; diastereomer ratio 90:10, determined
1
ꢂ
19
by H NMR of crude product. Colorless solid; mp 62e64 C; [
a
]
D
1
þ6.7 (c 0.1, CHCl
3 3
); H NMR (300 MHz, CDCl ) d 1.16e1.87 (m, 5H),
1.99e2.21 (m, 1H), 2.29e2.69 (m, 3H), 3.92 (s, 1H),4.01 (s, 1H), 4.85
13
(
d, J¼8.4 Hz, 1H), 7.40 (d, J¼8.4 Hz, 2H), 8.03 (d, J¼8.4 Hz, 2H);
NMR (75 MHz, CDCl 24.8, 27.8, 30.8, 42.8, 57.4, 74.4, 121.9, 129.0,
31.7, 140.3, 215.6. Enantiomeric excess of anti-product: 86%, de-
C
4.6. Monitoring reaction progress of aldol reaction
3
) d
1
4-Nitrobenzaldehyde (8a, 4.0 mg, 0.03 mmol) and 10 mol %
catalyst (4c, 5, 6; related to 8a) were added to a NMR tube. Cyclo-
hexanone (7a, 27.4 mL, 0.27 mmol) and DMSO-d /D O (94/6) solvent
6 2
termined by HPLC (Daicel Chiralpak AS-H, Hexane/i-PrOH¼80/20),
ꢀ
1
0
UV
l
¼254 nm, flow rate 0.5 mL min , t
R
(2S, 1 R) 30.3 min, t
R
0
(
2R,1 S) 43.4 min.
mixture (580
m
L) were filled to the solid materials and the reaction
ꢂ
mixture was shaked at 25 C. Conversions were determined by
monitoring the reaction mixtures by H NMR (see Supplementary
1
4
.4.6. (S)-2-((R)-(4-Methoxyphenyl)
(hydroxy)methyl)cyclohexa-
38
none (9f). Yield: 3.2 mg, 23%; diastereomer ratio 85:15, de-
termined by H NMR of crude product. Colorless solid; mp
data).
1
ꢂ
19
1
118e120 C; [
a]
D
þ13.2 (c 0.1, CHCl
3
); H NMR (300 MHz, CDCl
3
)
Acknowledgements
d
1.19e1.84 (m, 5H), 2.02e2.13 (m, 1H), 2.29e2.66 (m, 3H), 3.80 (s,
H), 3.92 (s, 1H), 4.75 (d, J¼8.9 Hz, 1H), 6.88 (d, J¼8.6 Hz, 2H), 7.24
3
This study was supported in part by a Grant-in-Aid for Young
Scientists (A) (No. 23685035) for scientific research from the Japan
13
(
d, J¼8.6 Hz, 2H); C NMR (75 MHz, CDCl
3
)
d
24.8, 27.9, 30.9, 42.8,

1990
P.L. S oꢀ ti et al. / Tetrahedron 72 (2016) 1984e1990
17. Riente, P.; Mendoza, C.; Peric
8. Wang, B. G.; Ma, B. C.; Wang, Q.; Wang, W. Adv. Synth. Catal. 2010, 352,
923e2928.
9. Yacob, Z.; Nan, A.; Liebscher, J. Adv. Synth. Catal. 2012, 354, 3259e3264.
20. Kong, Y.; Tan, R.; Zhao, L.; Yin, D. Green Chem. 2013, 15, 2422e2433.
1. Brust, M.; Walker, M.; Bethell, D.; Schiffrin, D. J.; Whyman, R. J. Chem. Soc., Chem.
aꢀ s, M. A. J. Mater. Chem. 2011, 21, 7350e7355.
Society for the Promotion of Science. The authors would like to
express their special thanks to Suzuki Motor Corporation for fi-
nancial support of P eꢀ ter Lajos S oꢀ ti’s Fellowship. We also thank
Professor N. Sakamoto for the helpful discussion, Mr. S. Takeuchi
and Mr. H. Kusanagi for technical support with ICP-OES and ele-
mental analysis measurements.
1
2
1
2
Commun. 1994, 801e802.
2
2. Brust, M.; Fink, J.; Bethell, D.; Schiffrin, D. J.; Kiely, C. J. Chem. Soc., Chem.
Commun. 1995, 1655e1656.
2
2
3. Chen, S.; Kimura, K. Langmuir 1999, 15, 1075e1082.
Supplementary data
4. Hostetler, M. J.; Wingate, J. E.; Zhong, C.-J.; Harris, J. E.; Vachet, R. W.; Clark, M.
R.; Londono, D. J.; Green, J. S.; Stokes, J. J.; Wignall, D. G.; Glish, L. G.; Porter, D.
M.; Evans, D. N.; Murray, R. W. Langmuir 1998, 14, 17e30.
5. Santacruz, L.; Niembro, S.; Santillana, A.; Shafir, A.; Vallri-bera, A. New J. Chem.
Supplementary data associated with this article can be found in
the online version, at http://dx.doi.org/10.1016/j.tet.2016.02.065.
2
2
2
2014, 38, 636e640.
6. Malkov, A. V.; Figlus, M.; Cooke, G.; Caldwell, S. T.; Rabani, G.; Prestly, M. R.;
References and notes
Ko ꢂc ovsk yꢀ , P. Org. Biomol. Chem. 2009, 7, 1878e1883.
7. Khiar, N.; Navas, R.; Elhalem, E.; Valdivia, V.; Fern aꢀ ndez, I. RSC Adv. 2013, 3,
1. MacMillan, D. W. C. Nature 2008, 455, 304e308.
3861e3864.
2
. Stereoselective Organocatalysis: Bond Formation Methodologies and Activation
Modes; Ríos Torres, R., Ed.; John Wiley & Sons: Hoboken, New Jersey, 2013.
. List, B.; Lerner, R. A.; Barbas, C. F., III. J. Am. Chem. Soc. 2000, 122, 2395e2396.
. List, B. J. Am. Chem. Soc. 2000, 122, 9336e9337.
28. Azcarate, J. C.; Corthey, G.; Pensa, E.; Vericat, C.; Fonticelli, M. H.; Salvarezza, R.
C.; Carro, P. J. Phys. Chem. Lett. 2013, 4, 3127e3138.
29. Corthey, G.; Giovanetti, L. J.; Ramallo-L
oꢀ pez, J. M.; Zelaya, E.; Rubert, A. A.;
Benitez, G. A.; Requejo, F. G.; Fonticelli, M. H.; Sal-varezza, R. C. ACS Nano 2010,
4, 3413e3421.
30. Naka, K.; Itoh, H.; Chujo, Y. Bull. Chem. Soc. Jpn. 2005, 78, 501e505.
31. Kristensen, E. T.; Hansen, K. F.; Hansen, T. Eur. J. Org. Chem. 2009, 387e395.
32. Raghuwanshi, V. S.; Ochmann, M.; Hoell, A.; Polzer, F.; Rademann, K. Langmuir
2014, 30, 6038e6046.
3
4
5
6
. List, B.; Pojarlier, P.; Martin, H. J. Org. Lett. 2001, 3, 2423e2425.
. Ahrendt, K. A.; Borths, C. J.; MacMillan, D. W. C. J. Am. Chem. Soc. 2000, 122,
4243e4244.
7. List, B. Tetrahedron 2002, 58, 5573e5587.
8
. Itsuno, S.; Haraguchi, N. Supported organocatalysts In Science of Synthesis:
Asymmetric Organocatalysis; List, B., Mauroka, K., Eds.; Thieme: Stuttgart, 2012;
pp 673e695.
33. For the preparation of 6, the method of Kristensen, et al. was used with
modifications (see reference 31).
9
1
. Gruttadauria, M.; Giacalone, F.; Noto, R. Chem. Soc. Rev. 2008, 37, 1666e1688.
34. Hoang, L.; Bahmanyar, S.; Houk, K. N.; List, B. J. Am. Chem. Soc. 2003, 125, 16e17.
35. Gruttadauria, M.; Giacalone, F.; Mossuto Marculescu, A.; Lo Meo, P.; Riela, S.;
Noto, R. Eur. J. Org. Chem. 2007, 4688e4698.
0. Ayats, C.; Henseler, A. H.; Peric aꢁ s, M. A. ChemSusChem 2012, 5, 320e325.
1
1. Lee, J.-W.; Mayer-Gall, T.; Opwis, K.; Song, C. E.; Gutmann, J. S.; List, B. Science
013, 341, 1225e1229.
2. Arakawa, Y.; Wennemers, H. ChemSusChem 2013, 6, 242e245.
3. Hern aꢀ ndez, J. G.; Juaristi, E. Chem. Commun. 2012, 5396e5409.
4. Fadhel, A. Z.; Pollet, P.; Liotta, C. L.; Eckert, C. A. Molecules 2010, 15, 8400e8424.
5. Stevens, P. D.; Fan, J.; Gardimalla, H. M. R.; Yen, M.; Gao, Y. Org. Lett. 2005, 7,
085e2088.
6. Belser, T.; St o€ hr, M.; Pfaltz, A. J. Am. Chem. Soc. 2005, 127, 8720e8731.
2
36. Li, J.; Yang, G.; Qin, Y.; Yang, X.; Cui, Y. Tetrahedron Asymmetry 2011, 22,
1
1
1
1
613e618.
37. Locatelli, E.; Ori, G.; Fournelle, M.; Lemor, R.; Montorsi, M.; Franchini, M. C.
Chem.dEur. J. 2011, 17, 9052e9056.
38. Mase, N.; Nakai, Y.; Ohara, N.; Yoda, H.; Takabe, K.; Tanaka, F.; Barbas, C. F., III. J.
Am. Chem. Soc. 2006, 128, 734e735.
2
1
́