A recyclable dendritic osmium catalyst for homogeneous dihydroxylation of olefins

Article abstract of DOI:10.1016/j.tet.2010.08.062

A series of osmate (OsO₄^2-) core dendrimers was prepared by an ion-exchange technique through the mixing of K ₂OsO₄ and a bis(quaternary ammonium bromide) core dendrimer, which consisted of poly(benzyl ether) dendron. By employing an osmate core dendrimer as a homogeneous catalyst, dihydroxylation reactions of olefins proceeded rapidly, and the dendritic osmium catalyst was recovered by reprecipitation and then reused. Furthermore, a dendritic effect on the recyclability of a catalyst was observed. In the case of asymmetric dihydroxylation reactions, the corresponding diol was obtained in a high chemical yield with a fair enantiomeric excess (ee). In this case, not only the dendritic osmium catalyst but also the chiral ligand could be recovered by reprecipitation and reused efficiently up to five times.

Full text of DOI:10.1016/j.tet.2010.08.062

Tetrahedron 66 (2010) 8536e8543
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
A recyclable dendritic osmium catalyst for homogeneous dihydroxylation
of olefins
,
a
b
b
a
Ken-ichi Fujita a , Manabu Yamazaki , Taku Ainoya , Teruhisa Tsuchimoto , Hiroyuki Yasuda
*
a National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba Central 5, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan
b Department of Applied Chemistry, School of Science and Technology, Meiji University, Higashimita, Tama-ku, Kawasaki, Kanagawa 214-8571, Japan
a r t i c l e i n f o
a b s t r a c t
A series of osmate (OsO42ꢀ) core dendrimers was prepared by an ion-exchange technique through the
mixing of K2OsO4 and a bis(quaternary ammonium bromide) core dendrimer, which consisted of poly
(benzyl ether) dendron. By employing an osmate core dendrimer as a homogeneous catalyst, dihy-
droxylation reactions of olefins proceeded rapidly, and the dendritic osmium catalyst was recovered by
reprecipitation and then reused. Furthermore, a dendritic effect on the recyclability of a catalyst was
observed. In the case of asymmetric dihydroxylation reactions, the corresponding diol was obtained in
a high chemical yield with a fair enantiomeric excess (ee). In this case, not only the dendritic osmium
catalyst but also the chiral ligand could be recovered by reprecipitation and reused efficiently up to five
times.
Article history:
Received 5 August 2010
Received in revised form 26 August 2010
Accepted 27 August 2010
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
catalytic sites immobilized within the dendrimer, are particularly
useful as synthetic catalysts not only because of their unique re-
Osmium-catalyzed dihydroxylation and asymmetric dihydrox-
ylation with chiral ligands are some of the most useful trans-
formations for the preparation of vicinal diols from olefins.1
Although these reactions have had widespread application in the
synthesis of chiral drugs, natural products, fine chemicals, and so on,
few fruitful industrial applications have been identified due to their
high cost, toxicity, and possible contaminationwith toxic osmium in
the final product.2 One of the most promising solutions to these
problems is the immobilization of the catalytic osmium to an in-
soluble matrix. Kobayashi et al. have used the microencapsulation of
OsO4 in polymer to achieve the recyclability of the catalyst.3 Chou-
dary et al. have heterogenized the osmium catalyst by an ion-ex-
change technique.4 In another fascinating design, OsO4 was
immobilized onto a silica-anchored tetrasubstituted olefin by Jacobs
et al.5 Efforts have thus been directed at heterogenizing the catalytic
osmiumon a polymeror silica gel, orevenvia encapsulation, toavoid
contamination of the final product and also to recycle the osmium
species.6 But the lower activity of the heterogeneous catalysts re-
mains severe problems. There are few examples of recyclable ho-
mogeneous osmium catalysts that have higher levels of activity than
those of heterogeneous catalysts.7
activity and fair solubility but also because of their recyclability by
reprecipitation.9 Presently, metallodendrimers with a catalytic site
at the core have received considerable attention due to their unique
selectivity and reactivity caused by specific reaction fields con-
structed by the dendron.10 Some of them show a positive dendritic
effect on regio- or stereoselectivity, meaning that their selectivity is
enhanced by an increase in the generation number of the den-
drimer.11,12 Recently, several examples of positive dendritic effects
on chemical yields have been reported by us13 and byother groups.14
A few examples of these unique phenomena are thought to be
caused by the suppression of the decomposition of catalytic metal
within the dendrimer.13b,14d,15 Thus, by the introduction of catalytic
osmium at the core site of the dendrimer, the decomposition of os-
mium catalyst was expected to be suppressed. We have expected
that recyclable homogeneous osmium catalysts, with higher levels
of activity than heterogeneous catalysts, would be produced by the
use of the core site of high-generation dendrimers. We report herein
the simple synthesis of a series of novel dendritic supports with a bis
(quaternary ammonium bromide) core, the preparation of the cor-
responding osmate (OsO42ꢀ) core dendrimer by an ion-exchange
technique and its application to homogeneous dihydroxylations of
olefins and their asymmetric version. We found that the third-
generation dendritic osmium catalyst could be recovered most ef-
ficiently by reprecipitation after the dihydroxylation reaction and
reused. We also observed the dendritic effect on the recyclability of
a catalyst.16
Dendrimers are fascinating macromolecules due to their unique
physical and chemical properties, which are caused by their well-
defined hyperbranched framework.8 Metallodendrimers, with
* Corresponding author. E-mail address: k.fujita@aist.go.jp (K. Fujita).
0040-4020/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2010.08.062

K.-i. Fujita et al. / Tetrahedron 66 (2010) 8536e8543
8537
2. Results and discussion
2.1. Preparation of dendritic osmium catalyst 2(Gn)
For the catalytic dihydroxylation reaction of olefins, OsO4, and
K2OsO4 were each used as a catalyst. Recently, K2OsO4 was widely
used due to its low volatility. By carrying out the dihydroxylation
reaction in aqueous acetonitrile by the use of 1 mol % of K2OsO4
according to Scheme 1, a brown solid appeared on the glassware
surface, as shown in Fig. 1. It is thought to be osmium oxide, which
is probably produced by the aggregation of catalytic osmium
species before the reaction is complete. From this result, in the
case of the preparation of a recyclable homogeneous osmium
catalyst, it is thought that not only we immobilize the catalytic
osmium on soluble polymer-supports, but also we must suppress
the aggregation of the introduced catalytic osmium. Then, it is
expected that the introduction of catalytic osmium at the core
site of a high-generation dendrimer would suppress the de-
composition of the catalyst, which is caused by the aggregation of
osmium.
Scheme 2. Synthesis of dendrimer 1(Gn).
Scheme 1. Dihydroxylation catalyzed by K2OsO4.
Fig. 2. Structural formulas of Gn dendrons (n¼1e3).
2ꢀ
Next, novel OsO4 core dendrimers were prepared by an ion-
exchange procedure, as follows (Table 1). In the case of G1, 1(G1),
which has low solubility in water and in most organic solvents, and
K2OsO4 were vigorously stirred in water to afford 2(G1), which was
Fig. 1. Dihydroxylation catalyzed by K2OsO4.
Table 1
Preparation of OsO4 core dendrimer 2(Gn)a
2ꢀ
In this study, we have introduced OsO42ꢀ at the core site of the
(quaternary ammonium bromide) core dendrimer by an ion-
exchange technique. To suppress the leaching of osmium off the
soluble polymer-backbone during the catalytic dihydroxylation
reaction, we synthesized bis(quaternary ammonium bromide) core
dendrimers 1(Gn), which immobilized catalytic osmium with the
bidentate ligation, according to Scheme 2.17 After the addition of
N,N,N0,N0-tetramethylethylenediamine to an N,N-dimethylform-
amide (DMF) solution of poly(benzyl ether) dendritic bromide
GneBr18 (structural formulas of Gn are shown in Fig. 2), the re-
action mixture was stirred at 70 ꢁC for 15 h under an argon at-
mosphere. In all generations, the corresponding bis(quaternary
ammonium bromide) core dendrimers 1(Gn) were obtained in
excellent chemical yields.
Gn
Solvent
Os contentb of 2(Gn) (mmol/g)
Found
Calcd
G1
G2
G3
H2O
H2O/CH2Cl2
H2O/Toluene
0.36
0.76
0.22
1.49
0.82
0.43
a
Reaction conditions: 1(Gn), an equimolar amount of K2OsO4, carried out at room
temperature for 5 h.
b
Determined by ICP-AES.

8538
K.-i. Fujita et al. / Tetrahedron 66 (2010) 8536e8543
collected by filtration of the reaction mixture. In the cases of G2 and
G3, 1(Gn) and K2OsO4 were vigorously stirred in water and den-
drimer-soluble organic solvent as a two-phase condition, followed
by extraction. The Os content of 2(Gn) was determined by ICP-AES
after microwave digestion. The Os content of 2(G1) was much lower
than the calculated value, likely due to the low solubility of 1(G1) in
water in the preparation of 2(G1). In all generations, it was found
that the ion-exchange for OsO42ꢀ was incomplete.
Furthermore, in the case of G3, even the third round of dihy-
droxylation was completed within 0.7 h with an excellent chemical
yield. In contrast, in the case of G1, the third round of dihydrox-
ylation proceeded very slowly to afford the corresponding diol in
only a 43% yield, even when allowed to proceed for 14 h. A com-
parison of increases in the reaction time by increasing the recycling
number shows that the increase in the reaction time for G3
(0.5 h/0.7 h) is the smallest among all generations. The relation-
ship between the generation number of 2(Gn) and increases in the
reaction time by increasing the recycling number is one of the
positive dendritic effects.11e14 To our knowledge, these results are
the first to be reported of the dendritic effect on the recyclability of
a catalyst.
2.2. Homogeneous dihydroxylation of olefins with dendritic
osmium catalyst
We then examined the utility of the dendrimer 2(Gn) as an
osmium catalyst by performing cis-dihydroxylation of olefins
(Table 2). By employing 1 mol % of various generations of dendritic
osmium catalysts 2(Gn), cis-dihydroxylation reactions of trans-
Next, we carried out the dihydroxylation reaction of trans-
b-methylstyrene by the use of 1 mol % of the third-generation 2(G3)
under various aqueous conditions (Table 3). In aqueous tert-butyl
alcohol, the dihydroxylation proceeded slowly to afford the corre-
sponding diol in a low chemical yield, probably because of the low
solubility of 2(G3) (Table 3, 29% for entry 1). In aqueous acetone, on
the other hand, the corresponding diol was obtained in a good
chemical yield (Table 3, 87% for entry 2) although the reaction time
in this case exceeded that in aqueous acetonitrile. As a result, it was
found that aqueous acetonitrile was the most suitable solvent for
the dihydroxylation reaction catalyzed by the dendritic osmium
catalyst 2(G3).
b-methylstyrene were carried out by the use of N-methylmorpho-
line N-oxide (NMO) as a re-oxidant in aqueous acetonitrile
(CH3CN/H2O¼4:1 (v/v)) at room temperature. After the dis-
appearance of trans-b-methylstyrene, which was monitored by
thin-layer chromatography (TLC), the reaction mixture was evap-
orated and dissolved in a small amount of acetone. The acetone
solution was poured into water and the precipitated dendritic
osmium catalysts were recovered and subsequently reused.
Table 2
Dihydroxylation of trans-
b
-methylstyrene catalyzed by 2(Gn)a
Table 3
Dihydroxylation of trans-b
-methylstyrene in various solventsa
Gn
Gn
Yieldb (%) (Time (h))
First
Entry
Solvent
Time (h)
Yieldb (%)
Second
Third
1
2
3
tBuOH/H2O
Acetone/H2O
CH3CN/H2O
5
1
0.5
29
87
93
G1
G2
G3
90 (0.5)
97 (0.5)
93 (0.5)
97 (2)
97 (1)
97 (0.5)
43 (14)
93 (2)
96 (0.7)
a
a
Reaction conditions: 2(G3) (1 mol %), olefin (1 equiv), NMO (1.2 equiv), organic
solvent/H2O (4:1, v/v, 0.25 M based on olefin), carried out at room temperature for
indicated time.
Reaction conditions: 2(Gn) (1mol %), olefin (1 equiv), NMO (1.2 equiv),
CH3CN/H2O (4:1, v/v, 0.25 M based on olefin), carried out at room temperature for
indicated time.
b
Determined by integration of 1H NMR absorptions referring to an internal
b
Determined by integration of 1H NMR absorptions referring to an internal
standard.
standard.
We then prepared various third-generation dendritic catalysts 5
and 6, which differed in the length between two nitrogens. Bis
(quaternary ammonium bromide) core dendrimers 3, 4, and the
corresponding osmate core dendrimers 5, 6 were prepared accord-
ing to the procedures similar to those used for 1(G3) and 2(G3),
respectively (Scheme 3).
In these catalytic conversions, 2(Gn) was completely dissolved
in the reaction mixture, allowing the dihydroxylation reactions to
proceed homogeneously to complete in 0.5 h in all generations
(‘First’ column in Table 2). Next, by carrying out reprecipitation for
the catalyst recycling, the recovery of 2(Gn) was low in the cases of
G1 and G2 (ca. 50% and 80% for G1 and G2, respectively). This is
likely due to the dendritic catalyst 2(Gn) having a lower molecular
weight, as it was somewhat dissolved in the water solvent. In
contrast, in the case of G3, catalyst 2(G3) was recovered nearly
quantitatively.
We carried out the dihydroxylation reaction of trans-b-
methylstyrene by the use of 1 mol % of the third-generation den-
dritic catalysts 2, 5, 6. The reaction conditions and the catalyst
recycling procedure were the same as in Table 2. The chemical yield,
Scheme 3. Synthesis of dendrimer 3, 4 and preparation of 5, 6.

K.-i. Fujita et al. / Tetrahedron 66 (2010) 8536e8543
Table 5 (continued )
8539
Table 4
Dihydroxylation of trans-
b-methylstyrene catalyzed by the third-generation den-
Entry
Olefin
Product
Time (h) Yieldb (%)
dritic catalystsa
6
0.5
81
7
8
0.5
2
87
90
Entry
Catalyst
m
Yield (%)b
(Reaction time, Os leaching)
First
Second
Third
1
2
3
2
5
6
2
3
4
93
97
96
(0.5 h, 2.2%)
88
(0.7 h, 2.2%)
86
(0.5h, 1.5%)
91
(0.7 h, 2.7%)
95
(0.7 h, 0.9%)
89
(1 h, 3.5%)
93
9
3
89
94
(0.5 h, 3.6%)
(0.7 h, 2.4%)
(1.3 h, 2.8%)
a
Reaction conditions: Catalyst (1 mol %), olefin (1 equiv), NMO (1.2 equiv),
CH3CN/H2O (4:1, v/v, 0.25 M based on olefin), carried out at room temperature for
indicated time.
10
1.5
b
Determined by integration of 1H NMR absorptions referring to an internal
a
The reaction conditions were the same as in Table 2.
Isolated yield.
standard.
b
We subsequently performed cis-dihydroxylation of various ole-
the reaction time, and the leaching of osmium off each dendrimer
backbone are shown in Table 4. The dihydroxylation reaction pro-
ceeded smoothly to afford the corresponding diol in a good
chemical yield in all cases. On the other hand, the leaching of
osmium was larger in the cases of 5 and 6 than in the case of 2.
Therefore, the increase in the reaction time by increasing the
recycling number in the case of 2 (0.5 h/0.7 h) was smallest
among all catalysts. As a result, it was found that 2 could be recycled
most efficiently in the catalytic dihydroxylation reaction.
fins by employing 1 mol % of the third-generation dendritic osmium
catalyst 2(G3) according to Table 5. In all cases, the dihydroxylation
reactions proceeded smoothly. Especially, in the case in which
substituted styrene derivatives were used, the dihydroxylations
were completed within 0.5 h (Table 5, entries 1e4). The dihydrox-
ylations catalyzed by the dendritic osmium catalyst 2(G3) were
more rapid than those reported for the immobilized osmium cata-
lysts,3e6 due to the homogeneity of 2(G3).
The reusability of catalyst 2(G3) was examined again using
various olefins (Table 6). In these experiments, it was found that
catalyst 2(G3) could be efficiently recycled up to five times by
reprecipitation, and the corresponding diols were consistently
obtained in good chemical yields in 0.5e2 h in all cases. The
reaction mixture was homogeneous and the decomposition of
catalyst 2(G3) was not observed in the reaction vessel, unlike the
case when K2OsO4 was used as a catalyst (Fig. 1), as shown in
Fig. 3. Also considering the dendritic effect on the recyclability of
a catalyst (Table 2), the efficient recycling of 2(G3) was probably
due to the aggregative decomposition of the catalytic osmium
being suppressed by the steric hindrance of the dendron owing
Table 5
Dihydroxylation of various olefins catalyzed by 2(G3)a
Entry
1
Olefin
Product
Time (h) Yieldb (%)
2ꢀ
0.5
86
to the introduction of OsO4 at the core site of the dendrimer
(Fig. 4).
Table 6
Catalyst recycling in dihydroxylations by use of 2(G3)a
2
0.5
90
Entry
Olefin
Yield (%)b
Third
First
86
Second
90
Fourth
88
Fifth
91
1
2
3
87
92
86
3
4
0.5
0.5
92
86
90
92
91
93
88
88
83
93
4
86
81
92
90
89
86
87
80
95
89
5
5
1
91
a
The reaction conditions were the same as in Table 2. Carried out for 0.5e2 h.
Isolated yield.
b
(continued on next page)

8540
K.-i. Fujita et al. / Tetrahedron 66 (2010) 8536e8543
then attempted the asymmetric dihydroxylation of olefins accord-
ing to the Sharpless procedure.
First, we examined a re-oxidant of the asymmetric dihydrox-
ylation (Table 7). After the stirring of the mixture of the dendritic
osmate 2(G3) (1 mol %) and the chiral ligand (DHQ)2-PHAL (1 mol
%) in aqueous acetonitrile (CH3CN/H2O¼4:1 (v/v)), a re-oxidant,
such as NMO and K3[Fe(CN)6], and trans-b-methylstyrene were
added. The stereochemistry of the major isomer, which was de-
termined by HPLC in the use of a chiral phase column, is shown in
Table 7. When NMO was used as a re-oxidant, the asymmetric
dihydroxylation proceeded smoothly to afford the corresponding
diol with an 87% enantiomeric excess (ee) (Table 7, entry 1). On the
other hand, when K3[Fe(CN)6] was used, the asymmetric dihy-
droxylation was rather slow and the chemical yield was low, al-
though the ee was slightly higher than that when NMO was used
(Table 7, entry 2). In this case, as the reaction mixture was muddy
because of the low solubility of K3[Fe(CN)6] in aqueous acetonitrile,
the dendritic catalyst could not be recovered by reprecipitation
after the dihydroxylation reaction. Next, by employing NMO as a re-
oxidant, we carried out asymmetric dihydroxylation by the slow
Fig. 3. Dihydroxylation catalyzed by 2(G3).
addition of trans-b-methylstyrene for the indicated times (Table 7,
0.5 h and 4 h in entries 3 and 4, respectively). The dihydroxylation
reaction proceeded completely as soon as we added an olefin. Al-
though the ratio of catalysts to olefins in entries 3 and 4 was much
larger than that in entry 1 due to the slow addition of an olefin, ee’s
in entries 3 and 4 did not exceed that in entry 1, contrary to our
expectation.
We subsequently performed the asymmetric dihydroxylation of
various olefins by employing 1 mol % of the chiral dendritic osmium
catalyst and NMO as a re-oxidant in aqueous acetonitrile, as shown
in Table 8. In all cases, the asymmetric dihydroxylation reactions
proceeded smoothly to afford the corresponding diol in good
chemical yields. In the case of aromatic olefins, the asymmetric
dihydroxylation reactions were completed in 0.5 h and the ee’s
were moderate to excellent. In contrast, in the case of aliphatic
olefins such as 1-decene and trans-5-decene, the asymmetric re-
actions were slightly slow with completion in 1e1.5 h, and the ee’s
were somewhat low (Table 8, entries 9 and 10).
Finally, the reusability of the chiral osmium catalyst was ex-
amined (Table 9). The reaction conditions were the same as in Table
8, and the chiral osmium catalyst was recycled according to the
same procedure as shown in Table 2. The asymmetric dihydrox-
ylation reactions proceeded smoothly to complete within 0.5e2 h
in each round, and the ee’s were hardly decreased even in the fifth
Fig. 4. The aggregation of catalytic osmium is suppressed by the steric hindrance of
the dendron.
2.3. Asymmetric dihydroxylation of olefins with dendritic
osmium catalyst and chiral ligand
The catalytic asymmetric dihydroxylation of olefins, using a cat-
alytic amount of osmium tetroxide in the presence of a chiral ligand,
allows access to a wide variety of enantiomerically pure vicinal di-
ols.19 In 1992, Sharpless et al. reported a catalytic system based on
biscinchona alkaloids such as 1,4-bis(9-O-dihydroquininyl)phtha-
lazine ((DHQ)2-PHAL), which has received a great deal of interest
due to the broad scope of substances available and the high enan-
tioselectivities attained.20 Encouraged by our previous results, we
Table 7
Asymmetric dihydroxylation in various conditions
Entry
Re-oxidant
Time (h)
Yielda (%)
ee (%)
1
2
3
4
NMO
K3[Fe(CN)6]
NMO
0.5
2
85
66
87
99
87
90
86
86
0.5b
4b
NMO
a
Isolated yield.
An olefin was added slowly for the indicated time.
b

K.-i. Fujita et al. / Tetrahedron 66 (2010) 8536e8543
8541
Table 8
3. Conclusion
Asymmetric dihydroxylation of various olefins
We prepared novel osmate core dendrimers to be used as ho-
mogeneous catalysts for the dihydroxylation of olefins. These
dendritic osmium catalysts were recovered by reprecipitation after
the dihydroxylation reaction and then reused. Furthermore, a pos-
itive dendritic effect on the recyclability of a catalyst was observed.
The third-generation dendritic osmium catalyst was recycled up to
five times. It is supposed that the efficient catalyst recycling was
probably due to the aggregative decomposition of the catalytic
Entry
Olefin
Time (h) Yielda (%) ee (%)
osmium being suppressed by the steric hindrance of the dendron
2ꢀ
owing to the introduction of OsO4
dendrimer.
at the core site of the
1
2
3
0.5
0.5
0.5
85
95
94
87
70
63
Furthermore, by applying these dendritic osmium catalysts to
the asymmetric version, the corresponding diol was obtained in
a high chemical yield with a fair ee. In this case, by reprecipitation
after the asymmetric dihydroxylation, both the dendritic osmium
catalyst and the chiral ligand were recovered simultaneously and
reused.
It was found that the immobilization of catalytic metals on
dendrimers at the core position is suitable for the design of ho-
mogeneous recyclable catalysts.
4
5
0.5
0.5
88
91
68
69
4. Experimental section
4.1. Materials
6
7
0.5
0.5
81
87
90
94
G3-Br was prepared by the method in the literature.12c Degassed
solvent, which was bubbled by argon for more than 1 h was used in
the preparation of 2, 5, and 6. Other reagents and dry solvents were
commercially available and were used as received. Kieselgel 60 F254
(Merck) was used for TLC, and Wakogel C-300 (Wako) was used for
silica gel column chromatography.
8
0.5
92
96
9
1
96
96
45
65
4.2. Measurement
10
1.5
Microanalyses were performed with a CE Instruments EA1110
elemental analyzer. IR spectra were recorded on a Shimadzu IR
Prestige-21 FT-IR spectrophotometer. 1H and 13C NMR spectra were
measured with a Bruker Avance III 400 spectrometer (1H: 400 MHz,
13C: 100 MHz). Chemical shifts were reported as parts per million
a
Isolated yield.
Table 9
Chiral catalyst recycling in asymmetric dihydroxylations by use of 2(G3)a
downfield from TMS as an internal standard in
d units. Coupling
constants (J) are given in hertz (Hz). Melting points were de-
termined on a Mettler FP90 microscopic plate, and were un-
corrected. High performance liquid chromatography (HPLC) was
performed with a Shimadzu 10ADVP LCsolution system using
a Daicel chiral column. ICP-AES was recorded on a Shimadzu ICPS-
8100 analyzer.
Olefin
First
Second
Third
Fourth
Fifth
Yieldb (%)
ee (%)
85
87
92
86
95
86
94
85
92
85
Yieldb (%)
ee (%)
95
70
98
67
99
67
97
68
97
67
4.3. Synthesis of dendritic quaternary ammonium bromide
1(Gn), 3 and 4: typical procedure
a
The reaction conditions were the same as in Table 8. Carried out for 0.5e2 h.
Isolated yield.
b
A dry DMF solution (8 mL) of N,N,N0,N0-tetramethylethylenedi-
amine (74.6 mg, 0.642 mmol) and 3,5-bis[3,5-bis(3,5-dime-
thoxybenzyloxy)benzyloxy]benzyl bromide (1.408 g, 1.343 mmol)
was stirred at 70 ꢁC for 15 h under an argon atmosphere. The re-
action mixture was evaporated to dryness, and the residue was
purified with silica gel column chromatography (ethyl acetate/
methanol¼3/1 as an eluent) to obtain 1(G3) (1.398 g, 0.632 mmol)
in a 98% yield.
Compounds 1(G2), 3 and 4 were synthesized according to the
above procedure. In the case of the synthesis of 1(G1), a white
powder was formed during the reaction. The reaction mixture was
evaporated to dryness and was filtered to obtain 1(G1), followed by
washing with acetone.
dihydroxylation reaction of either olefin.21 As a result, it was sup-
posed that not only the dendritic osmium catalyst 2(G3) but also
the chiral ligand (DHQ)2-PHAL could be recovered completely by
reprecipitation so that the subsequent asymmetric dihydroxylation
reaction proceeded efficiently. In the case of the commercially
available microencapsulated OsO4 catalyst, the chiral ligand was
recovered by acidebase extraction after the filtration of the
immobilized OsO4 catalyst.3c In our catalyst recycling system, both
the immobilized osmium catalyst and the chiral ligand were re-
covered simultaneously to promote the subsequent asymmetric
dihydroxylation efficiently.

8542
K.-i. Fujita et al. / Tetrahedron 66 (2010) 8536e8543
4.3.1. N,N0-Ethane-1,2-diyl N,N0-bis(3,5-dimethoxybenzyloxy) N,N,N0,
N0-tetramethylammonium dibromide (1(G1)). White powder; mp
201.5e202 ꢁC; IR (KBr) 3483, 2976, 2952, 1598, 1350, 1208, 1175,
1154, 1066, 1044, 934, 843 cmꢀ1; 1H NMR (400 MHz; CDCl3/CD3OD)
temperature under an argon atmosphere. The resulting two-phase
mixture was vigorously stirred for 5 h. The reaction mixture was
extracted with toluene (80 mLꢂ3). After removal of the solvent
under reduced pressure, 2(G3) was obtained as a black solid
(1.667 g). The Os content of 2(G3) was determined by ICP-AES after
microwave digestion by the addition of HNO3.
d
¼6.81 (d, J¼2.0 Hz, 4H), 6.60 (t, J¼2.0 Hz, 2H), 4.78 (s, 4H), 4.55 (s,
4H), 3.84 (s, 12H), 3.28 (s, 12H); 13C NMR (100 MHz; CDCl3/CD3OD)
d
¼162.0, 128.9, 111.9, 103.1, 69.1, 57.9, 56.0, 50.7. Anal. Calcd for
C24H38Br2N2O4$0.2H2O: C, 49.53; H, 6.65; N, 4.81; Br, 27.46%.
Found: C, 49.64; H, 6.68; N, 4.83; Br, 27.16%.
4.4.1. Compound (2(G1)). Black powder; mp 203e205 ꢁC; Os con-
tent: 0.36 mmol/g; IR (KBr) 3418, 3000, 2966, 2943, 2841, 1595,
1328, 1207,1157, 1065, 853, 833 cmꢀ1. Anal. Found: C, 41.46; H, 5.71;
N, 3.96%.
4.3.2. N,N0-Ethane-1,2-diyl
N,N0-[3,5-bis(3,5-dimethoxybenzyloxy)-
benzyloxy]N,N,N0,N0-tetramethylammoniumdibromide(1(G2)). White
powder; mp 183e184.5 ꢁC; IR (KBr) 3386, 3002, 2962, 2838, 1597,
4.4.2. Compound (2(G2)). Black powder; mp 79e81 ꢁC; Os content:
0.76 mmol/g; IR (KBr) 3446, 2955, 2938, 2837, 1598, 1205, 1155,
1052, 833 cmꢀ1. Anal. Found: C, 54.39; H, 5.85; N, 2.41%.
1465, 1430, 1298, 1205, 1159, 1052, 834 cmꢀ1 1H NMR (400 MHz;
;
CDCl3)
8H), 6.37 (t, J¼2.2 Hz, 4H), 4.97 (s, 4H), 4.94 (s, 8H), 4.82 (s, 4H), 3.75
(s, 24H), 3.20 (s, 12H); 13C NMR (100 MHz, CDCl3)
138.6, 127.9, 112.1, 105.5, 104.7, 99.9, 70.1, 67.6, 57.9, 55.5, 50.0. Anal.
Calcd for C56H70Br2N2O12: C, 59.89; H, 6.28; N, 2.49; Br, 14.23%.
Found: C, 59.88; H, 6.12; N, 2.41; Br, 13.96%.
d¼6.77 (d, J¼1.8 Hz, 4H), 6.62e6.59 (m, 2H), 6.55 (d, J¼2.2 Hz,
d¼161.0, 160.0
4.4.3. Compound (2(G3)). Black powder; mp 49e51 ꢁC; Os content:
0.22 mmol/g; IR (KBr) 3447, 2961, 2936, 2838, 1457, 1205, 1155,
1052, 832 cmꢀ1. Anal. Found: C, 63.29; H, 5.95; N, 1.21%.
4.4.4. Compound (5). Black powder; mp 76e78 ꢁC; Os content:
0.33 mmol/g; IR (KBr) 3383, 3001, 2936, 2837, 1597, 1456,1373,1323,
1206, 1155, 1051, 835 cmꢀ1. Anal. Found: C, 60.78; H, 5.66; N, 1.09%.
4.3.3. N,N0-Ethane-1,2-diyl N,N0-[3,5-bis[3,5-bis(3,5-dimethoxybenzyl-
oxy)benzyloxy]benzyloxy] N,N,N0,N0-tetramethylammonium dibromide
(1(G3)). White powder; mp 72.5e74.5 ꢁC; IR (KBr) 3356, 3000,
2937, 2838, 1590, 1429, 1300, 1270, 1205, 1150, 1052, 1009,
4.4.5. Compound (6). Black powder; mp 70.5e72.5 ꢁC; Os content:
0.20 mmol/g; IR (KBr) 3399, 2999, 2938, 2839,1597,1458,1342,1323,
1206, 1155, 1051, 837 cmꢀ1. Anal. Found: C, 62.56; H, 5.99; N, 1.08%.
834 cmꢀ1; 1H NMR (400 MHz; CDCl3)
d¼6.71 (d, J¼1.6 Hz, 4H), 6.67
(d, J¼2.0 Hz, 8H), 6.61e6.56 (m, 2H), 6.52 (d, J¼2.0 Hz, 16H), 6.51 (t,
J¼2.0 Hz, 4H), 6.35 (t, J¼2.2 Hz, 8H), 4.90 (s, 24H), 4.86 (s, 4H), 4.67
(s, 4H), 3.72 (s, 48H), 3.10 (s, 12H); 13C NMR (100 MHz; CDCl3)
4.5. Dihydroxylation reaction: typical procedure
d
¼160.9, 160.0, 159.9, 139.1, 138.8, 127.8, 111.9, 106.6, 105.3, 104.7,
101.7, 99.8, 70.0, 67.7, 57.9, 55.4, 49.9. Anal. Calcd for
C120H134Br2N2O28: C, 65.15; H, 6.11; N, 1.27; Br, 7.22%. Found: C,
64.99; H, 5.93; N, 1.22; Br, 7.08%.
To a CH3CN/H2O (4:1, v/v) solution (3.75 mL) of 2(G3) (48.5 mg,
0.01 mmol) were added trans-b-methylstyrene (120.2 mg,
1.017 mmol) and N-methylmorpholine N-oxide (NMO; 141.6 mg,
1.209 mmol) successively at room temperature under an argon
atmosphere. After stirring the resulting mixture for 0.5 h, the
dihydroxylation reaction was completed (monitored by TLC). The
reaction mixture was evaporated and dissolved in a small amount
of acetone. The acetone solution was poured into 6 mL of water to
precipitate 2(G3) as a grayish powder. After centrifugal separation,
the aqueous solution was decanted, and the recovered 2(G3) was
reused for subsequent dihydroxylation reactions.
The decanted aqueous solution was concentrated under reduced
pressure, and was purified with silica gel column chromatography
(hexane/ethyl acetate¼1/1 as an eluent) to obtain 1-phenyl-1,2-
propanediol (133.3 mg, 0.876 mmol) in an 86% yield.
4.3.4. N,N0-Propane-1,2-diyl
N,N0-[3,5-bis[3,5-bis(3,5-dimethoxy-
N,N,N0,N0-tetramethylammonium
benzyloxy)benzyloxy]benzyloxy]
dibromide (3). White powder; mp 71e73 ꢁC; IR (KBr) 3399, 2999,
2936, 2837, 1595, 1458, 1373, 1298, 1206, 1153, 1049, 831, 681 cmꢀ1
1H NMR (400 MHz; CDCl3)
;
d
¼6.78 (d, J¼1.7 Hz, 4H), 6.67 (d,
J¼2.1 Hz, 8H), 6.64e6.62 (m, 2H), 6.54 (d, J¼2.2 Hz, 16H), 6.52
(t, J¼2.0 Hz, 4H), 6.37 (t, J¼2.2 Hz, 8H), 4.95 (s, 8H), 4.92 (s, 16H),
4.64 (s, 4H), 3.74 (s, 52H), 3.05 (s, 12H), 2.46 (s, 2H); 13C NMR
(100 MHz; CDCl3)
d
¼160.9, 160.0, 159.9, 139.1, 138.8, 128.6, 112.0,
106.6, 105.3, 104.5, 101.7, 99.8, 70.1, 70.0, 67.8, 62.4, 55.4, 50.2, 19.1.
Anal. Calcd for C121H136Br2N2O28: C, 65.28; H, 6.16; N,1.26; Br, 7.18%.
Found: C, 65.00; H, 6.18; N, 1.22; Br, 7.08%.
In Tables 2e4, the chemical yields were determined by in-
tegrating 1H NMR absorptions by referring to an internal standard,
which was added to the decanted aqueous solution. The leaching of
osmium in Table 4 was determined by ICP-AES measurement of the
decanted aqueous solution.
4.3.5. N,N0-Butane-1,2-diyl N,N0-[3,5-bis[3,5-bis(3,5-dimethoxybenzyl-
oxy)benzyloxy]benzyloxy] N,N,N0,N0-tetramethylammonium dibromide
(4). White powder; mp 68.5e70.5 ꢁC; IR (KBr) 2995, 2936, 2837,
1595, 1458, 1319, 1206, 1153, 1049, 831, 681 cmꢀ1
;
1H NMR
(400 MHz; CDCl3)
d
¼6.78 (d, J¼1.7 Hz, 4H), 6.66 (d, J¼2.1 Hz, 8H),
6.63e6.60 (m, 2H), 6.53 (d, J¼2.2 Hz,16H), 6.50 (t, J¼2.0 Hz, 4H), 6.35
(t, J¼2.2 Hz, 8H), 4.95 (s, 8H), 4.92 (s, 16H), 4.59 (s, 4H), 3.87 (br s,
4H), 3.72 (s, 48H), 3.00 (s, 12H), 2.07 (s, 4H); 13C NMR (100 MHz;
4.6. Asymmetric dihydroxylation reaction: typical procedure
To a CH3CN/H2O (4:1, v/v) solution (2.5 mL) of 2(G3) (46.6 mg,
0.01 mmol) was added 1,4-bis(9-O-dihydroquininyl)phthalazine
((DHQ)2-PHAL) (8.2 mg, 0.011 mmol). After the reaction mixture
CDCl3)
d
¼161.0, 160.03, 159.95, 139.2, 138.8, 128.8, 112.1, 106.6, 105.3,
104.4,101.7, 99.9, 70.1, 70.0, 67.6, 64.5, 55.4, 49.8, 20.2. Anal. Calcd for
C122H138Br2N2O28$2H2O: C, 64.37; H, 6.29; N, 1.23; Br, 7.02%. Found:
C, 64.33; H, 6.13; N, 1.16; Br, 7.04%.
was stirred for 10 min, trans-b-methylstyrene (118.3 mg,1.00 mmol)
and N-methylmorpholine N-oxide (NMO; 140.5 mg, 1.20 mmol)
were added successively at room temperature under an argon at-
mosphere. After the resulting mixture was stirred for 0.5 h, the
dihydroxylation reaction was completed (monitored by TLC). The
reaction mixture was evaporated and dissolved in a small amount of
acetone. The acetone solution was poured into 6 mL of water to
precipitate 2(G3) and (DHQ)2-PHAL as a grayish powder. After
centrifugal separation, the aqueous solution was decanted, and the
4.4. Preparation of dendritic osmate 2(Gn), 5, and 6: typical
procedure
To a toluene solution (degassed; 80 mL) of 1(G3) (1.547 g,
0.698 mmol), an aqueous solution (degassed; 50 mL) of
K2OsO4$2H2O (256 mg, 0.695 mmol) was added at room

K.-i. Fujita et al. / Tetrahedron 66 (2010) 8536e8543
8543
9. (a) Dendrimer Catalyst; Gade, L. H., Ed.; Springer: Berlin: Heidelberg, 2006; (b)
Méry, D.; Astruc, D. Coord. Chem. Rev. 2006, 250, 1965e1979; (c) Grayson, S. M.;
Fréchet, J. M. J. Chem. Rev. 2001, 101, 3819e3867; (d) Astruc, D.; Chardac, F.
Chem. Rev. 2001, 101, 2991e3023.
10. Twyman, L. J.; King, A. S. H.; Martin, I. K. Chem. Soc. Rev. 2002, 31, 69e82.
11. Review of the dendritic effect: (a) Helms, B.; Fréchet, J. M. J. Adv. Synth. Catal.
2006, 348, 1125e1148; (b) Chow, H.-F.; Leung, C.-F.; Wang, G.-X.; Yang, Y.-Y. C.R.
Chem. 2003, 6, 735e745.
12. Examples of dendritic effect on selectivity: (a) Benito, J. M.; de Jesús, E.; de la
Mata, F. J.; Flores, J. C.; Gómez, R. Chem. Commun. 2005, 5217e5219; (b) Fujita,
K.; Muraki, T.; Sakurai, T.; Oishi, A.; Taguchi, Y. Chem. Lett. 2005, 34, 1180e1181;
(c) Muraki, T.; Fujita, K.; Oishi, A.; Taguchi, Y. Polym. J. 2005, 37, 847e853; (d)
Ribourdouille, Y.; Engel, G. D.; Richard-Plouet, M.; Gade, L. H. Chem. Commun.
2003, 1228e1229; (e) Mizugaki, T.; Murata, M.; Ooe, M.; Ebitani, K.; Kaneda, K.
Chem. Commun. 2002, 52e53; (f) Bhyrappa, P.; Young, J. K.; Moore, J. S.; Suslick,
K. S. J. Am. Chem. Soc. 1996, 118, 5708e5711.
13. (a) Muraki, T.; Fujita, K.; Kujime, M. J. Org. Chem. 2007, 72, 7863e7870; (b)
Hattori, H.; Fujita, K.; Muraki, T.; Sakaba, A. Tetrahedron Lett. 2007, 48,
6817e6820; (c) Fujita, K.; Muraki, T.; Hattori, H.; Sakakura, T. Tetrahedron Lett.
2006, 47, 4831e4843; (d) Muraki, T.; Fujita, K.; Terakado, D. Synlett 2006,
2646e2648.
recovered 2(G3) and (DHQ)2-PHAL were reused for subsequent
asymmetric dihydroxylation reactions.
The decanted aqueous solution was concentrated under reduced
pressure and was purified with silica gel column chromatography
(hexane/ethyl acetate¼1/1 as eluent) to obtain 1-phenyl-1,2-pro-
panediol (128.7 mg, 0.846 mmol) in an 85% chemical yield. Enan-
tiomeric excess (ee) was determined by HPLC analysis using Daicel
Chiralcel OD (87% ee).
Supplementary data
Supplementary data associated with this article can be found in
the online version at doi:10.1016/j.tet.2010.08.062.
References and notes
14. (a) Snelders, D. J. M.; van Koten, G.; Gebbink, R. J. M. K. J. Am. Chem. Soc. 2009,
131, 11407e11416; (b) Wang, Z.-J.; Deng, G.-J.; Li, Y.; He, Y.-M.; Tang, W.-J.; Fan,
Q.-H. Org. Lett. 2007, 9, 1243e1246; (c) Helms, B.; Liang, C. O.; Hawker, C. J.;
Fréchet, J. M. J. Macromolecules 2005, 38, 5411e5415; (d) Fujihara, T.; Obora, Y.;
Tokunaga, M.; Sato, H.; Tsuji, Y. Chem. Commun. 2005, 4526e4528; (e) Zhang,
X.; Xu, H.; Dong, Z.; Wang, Y.; Liu, J.; Shen, J. J. Am. Chem. Soc. 2004, 126,
10556e10557; (f) Fan, Q.-H.; Chen, Y.-M.; Chen, X.-M.; Jiang, D.-Z.; Xi, F.; Chan,
A. S. C. Chem. Commun. 2000, 789e790.
15. Iwasawa, T.; Tokunaga, M.; Obora, Y.; Tsuji, Y. J. Am. Chem. Soc. 2004, 126,
6554e6555.
16. (a) Our preliminary report: Fujita, K.; Ainoya, T.; Tsuchimoto, T.; Yasuda, H.
Tetrahedron Lett. 2010, 51, 808e810 our report was presented in Synfacts, 2010,
495.
€
1. Schroder, M. Chem. Rev. 1980, 80, 187e213.
2. (a) Hammond, C. R. CRC Handbook of Chemistry and Physics, 81st ed., Lide, D.R.
Ed.; CRC: Boca Raton, FL, 2000; p 4. (b) Leadbeater, N. E.; Marco, M. Chem. Rev.
2002, 102, 3217e3274.
3. (a) Ishida, T.; Akiyama, R.; Kobayashi, S. Adv. Synth. Catal. 2005, 347, 1189e1192;
(b) Kobayashi, S.; Ishida, T.; Akiyama, R. Org. Lett. 2001, 3, 2649e2652; (c)
Kobayashi, S.; Endo, M.; Nagayama, S. J. Am. Chem. Soc. 1999, 121, 11229e11230;
(d) Nagayama, S.; Endo, M.; Kobayashi, S. J. Org. Chem. 1998, 63, 6094e6095.
4. (a) Choudary, B. M.; Jyothi, K.; Kantam, M. L.; Sreedhar, B. Adv. Synth. Catal.
2004, 346, 45e48; (b) Choudary, B. M.; Chowdari, N. S.; Jyothi, K.; Kantam, M. L.
J. Am. Chem. Soc. 2002, 124, 5341e5349.
5. Severeyns, A.; De Vos, D. E.; Fiermans, L.; Verpoort, F.; Grobet, P. J.; Jacobs, P. A.
Angew. Chem., Int. Ed. 2001, 40, 586e589.
17. However osmium core dendrimers have been reported previously, their
catalytic activity was as low as in the case of heterogeneity: Tang, W.-J.;
Yang, N.-F.; Yi, B.; Deng, G.-J.; Huang, Y.-Y.; Fan, Q.-H. Chem. Commun. 2004,
1378e1379.
18. (a) Muraki, T.; Fujita, K.; Oishi, A.; Taguchi, Y. Polym. J. 2005, 37, 847e853; (b)
Tomoyose, Y.; Jiang, D.-L.; Jin, R.-H.; Aida, T.; Yamashita, T.; Horie, K.; Yashima,
E.; Okamoto, Y. Macromolecules 1996, 29, 5236e5238; (c) Hawker, C. J.; Fréchet,
J. M. J. J. Am. Chem. Soc. 1990, 112, 7638e7647.
19. (a) Kolb, H. C.; Van Nieuwenhze, M. S.; Sharpless, K. B. Chem. Rev. 1994, 94,
2483e2547; (b) Johnson, R. A.; Sharpless, K. B. In Catalytic Asymmetric
Synthesis, 2nd ed.; Ojima, I., Ed.; VCH: Weinheim, 2000; pp 357e398.
20. Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino, G. A.; Hartung, J.; Jeong,
K.-S.; Kwong, H.-L.; Morikawa, K.; Wang, Z.-M.; Xu, D.; Zhang, X.-L. J. Org. Chem.
1992, 57, 2768e2771.
6. (a) Jun, B.-H.; Kim, J.-H.; Park, J.; Kang, H.; Lee, S.-H.; Lee, Y.-S. Synlett 2008,
2313e2316; (b) Jiang, R.; Kuang, Y.; Sun, X.; Zhang, S. Acta Chim. Slov. 2005, 52,
467e470; (c) Kim, K. J.; Choi, H. Y.; Hwang, S. H.; Park, Y. S.; Kwueon, E. K.; Choi,
D. S.; Song, C. E. Chem. Commun. 2005, 3337e3339; (d) Ley, S. V.; Ramarao, C.;
Lee, A.-L.; Østergaard, N.; Smith, S. C.; Shirley, I. M. Org. Lett. 2003, 5, 185e187;
(e) Lee, B. S.; Mahajan, S.; Janda, K. D. Tetrahedron Lett. 2005, 46, 4491e4493; (f)
Miao, J. H.; Yang, J.-H.; Chen, L.-Y.; Tu, B.-X.; Huang, M. Y.; Jiang, Y.-Y. Polym. Adv.
Technol. 2004, 15, 221e224; (g) Yang, J. W.; Han, H.; Roh, E. J.; Lee, S.-g.; Song, C.
E. Org. Lett. 2002, 4, 4685e4688.
7. (a) Jiang, R.; Kuang, Y.; Sun, X.; Zhang, S. Tetrahedron: Asymmetry 2004, 15,
743e746; (b) Chandrasekhar, S.; Narsihmulu, C.; Sultana, S. S.; Reddy, N. R.
Chem. Commun. 2003, 1716e1717; (c) Yao, Q. Org. Lett. 2002, 4, 2197e2199.
8. (a) Dendrimers and Other Dendritic Polymers; Fréchet, J. M. J., Tomalia, D. A., Eds.;
John Wiley: New York, NY, 2001; (b) Newkome, G. R.; Moorefield, C. N.; Vogtle,
F. Dendrimers and Dendrons; Wiley-VCH: Weinheim, 2001; (c) Chase, P. A.;
Gebbink, R. J. M. K.; van Koten, G. J. Organomet. Chem. 2004, 689, 4016e4054;
(d) Fréchet, J. M. J. J. Polym. Sci., Part A: Polym. Chem. 2003, 41, 3713e3725.
21. In the case of the asymmetric dihydroxylation of trans-b-methylstyrene, the
leaching of osmium were 2.3% (first), 1.6% (second), 3.6% (third), 2.5% (fourth)
and 1.5% (fifth)