Asymmetric diethyl- and diphenylzinc additions to aldehydes by using a fluorine-containing chiral amino alcohol: A striking temperature effect on the enantioselectivity, a minimal amino alcohol loading, and an efficient recycling of the amino alcohol

Article abstract of DOI:10.1002/chem.200400703

A chiral pyrrolidinylmethanol derivative containing perfluoro-ponytails (5) was prepared from (S)-proline. The use of this perfluoro-substituted amino alcohol in catalytic asymmetric additions of organozinc reagents to aldehydes affords products with high

Full text of DOI:10.1002/chem.200400703

FULL PAPER
Asymmetric Diethyl- and Diphenylzinc Additions to Aldehydes by Using a
Fluorine-Containing Chiral Amino Alcohol: A Striking Temperature Effect
on the Enantioselectivity, a Minimal Amino Alcohol Loading, and an
Efficient Recycling of the Amino Alcohol
Jin Kyoon Park,[a] Hong Geun Lee,[a] Carsten Bolm,*[b] and B. Moon Kim*[a]
Abstract: A chiral pyrrolidinylmetha-
nol derivative containing perfluoro-po-
nytails (5) was prepared from (S)-pro-
line. The use of this perfluoro-substitut-
ed amino alcohol in catalytic asymmet-
ric additions of organozinc reagents to
aldehydes affords products with high
enantioselectivities in both pure
hexane and a mixture of hexane and
FC-72 (perfluorohexane). Enantiomer-
ic excesses up to 94 and 88% ee have
been achieved in Et2Zn and Ph2Zn ad-
ditions, respectively. For the reactions
in the biphasic solvent system a striking
temperature effect was observed. Thus,
when the temperature was raised from
0 to 408C the ee value of the product
increased from 81 to 92%. Further-
more, the catalyst loading could be re-
markably low, and with only 0.1 mol%
of amino alcohol 5 a product with 90%
ee was obtained in the Et2Zn addition
to benzaldehyde in hexane. The per-
fluoro-ligand was easily recovered by
simple phase separation, and until the
ninth repetition its reuse proceeded
without significant loss of enantioselec-
tivity and reactivity.
Keywords: amino alcohols · asym-
metric catalysis · biphasic catalysis ·
catalyst recycling
Introduction
recycling methods for asymmetric reactions.[2] The FBS
strategy allows one to separate fluorous ligands from the re-
action mixture by simple phase separation at lower tempera-
ture after the reaction has been performed at elevated tem-
perature, where the fluorous ligand becomes soluble to both
phases.
In asymmetric catalysis aimed at large-scale production of
chiral compounds, the cost of the catalyst is often too high.
To resolve this issue, recycling of catalysts has been recog-
nized as one of the most practical approaches. For this pur-
pose immobilization of a homogeneous catalyst onto a het-
erogeneous system has frequently been employed.[1] Even
though various chiral ligands have efficiently been attached
onto solid supports, heterogeneous applications of homoge-
neous chiral catalysts have often been associated with unde-
sirable properties such as a reduced kinetic profile and a
low enantioselectivity.
Successful examples of FBS include asymmetric protona-
tions,[3] epoxidations,[4] diethylzinc and triethylaluminum-
mediated addition reactions,[5] and others.[6] Diorganozinc
addition reactions to carbonyl compounds have extensively
been studied and many excellent catalysts have been devel-
oped to give the resulting secondary alcohols with very high
enantioselectivities.[7] The first application of a diethylzinc
addition in FBS was reported by Kleijn et al.[5a] and since
then several similar approaches have been detailed by Na-
kamura et al.[5b,d,f] and Tian et al.,[5c,e] who employed fluo-
rous chiral binaphthol derivatives and fluorous chiral amino
alcohols, respectively. Herein, we wish to present an effec-
tive perfluoro catalyst for diorganozinc addition reactions
based on the modification of pyrrolidinylmethanol[8,9] deriv-
atives. We chose the perfluoro-prolinol system 5 due to the
high enantioselectivity described for reactions involving
known prolinol derivatives.
Recently, fluorous organic biphase systems (FBS) pio-
neered by Hovꢀrth have been recognized as highly effective
[a] Dr. J. K. Park, H. G. Lee, Prof. Dr. B. M. Kim
School of Chemistry, Seoul National University
Seoul 151-747, (South Korea)
Fax : (+82)2-872-7505
E-mail: kimbm@snu.ac.kr
[b] Prof. Dr. C. Bolm
Institut fꢁr Organische Chemie der RWTH Aachen
Professor-Pirlet-Strasse 1, 52056 Aachen (Germany)
E-mail: carsten.bolm@oc.rwth-aachen.de
Chem. Eur. J. 2005, 11, 945 – 950
DOI: 10.1002/chem.200400703
ꢂ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
945

Results and Discussion
Fluorous amino alcohol 5 was
synthesized from known (S)-
proline derivative 1 as shown in
Scheme 1.[10] For the introduc-
tion of a perfluoro-ponytail at
the 4-position of the aromatic
ring, 4-bromophenylmagnesium
bromide was prepared and cou-
pled in situ with ester 1, fur-
nishing compound 2 in 70%
yield. Treatment of
2 with
NaOH in MeOH provided
cyclic carbamate 3 quantitative-
ly. Introduction of the fluoro-
ponytail was accomplished by
adding perfluoroalkyl iodide
slowly to the DMSO solution of
3 in presence of copper powder.
Scheme 1. Preparation of amino alcohol 5. a) 1,4-Dibromobenzene, Mg, Et2O, 70%; b) MeOH, NaOH, quanti-
tative; c) Cu, C8F17I, DMSO, 1208C, 70% or C10F21I, DMSO, 1208C, trace; d) DIBAL, toluene, 508C, quantita-
tive.
For this coupling reaction, it was critical to maintain the re-
action temperature at 1208C and to add the perfluoroalkyl
iodide slowly over a period of 30 min. In this manner, the
desired product 4a was obtained in 70% yield. Removal of
the carbamate protecting group and introduction of N-
methyl functionality was achieved through reduction of 4a
by using diisobutylaluminum hydride (DIBAL). The desired
N-methylated derivative 5 was obtained in a quantitative
yield. By using the same protocol, the synthesis of com-
pound 4b bearing a longer perfluoro-ponytail was also at-
tempted, however, the reaction from 3 to 4b was extremely
sluggish and only a trace amount of the desired product was
obtained. This result may be due to the lower solubility of
C10F21I in the reaction medium than C8F17I, which was spar-
ingly soluble. For a control experiment, non-perfluoropony-
tail containing ligand 6[10] was prepared.
92% when the reactions were performed at 0, 20 and 408C,
respectively, as clearly indicated in the plot in Figure 1. At
608C, the reaction proceeded very fast (in 10 min), however,
Figure 1. Temperature profile of the enantioselectivity in the diethylzinc
addition reaction to benzaldehyde under monophasic and biphasic reac-
tion conditions.
Investigation of the asymmetric diethylzinc addition to
benzaldehyde was carried out by using 3 mol% of the
amino alcohol 5 or 6, which was deprotonated with n-butyl-
lithium (3.7 mol%) before starting the catalysis. Monophasic
(hexane) and biphasic (hexane/FC-72) solvent systems were
employed, and reactions at various temperatures were ex-
amined. The results are summarized in Table 1. Surprisingly
in the case of fluorous amino alcohol 5 in the biphasic sol-
vent system, the ee of the product increased from 81, 86 to
the enantioselectivity was somewhat reduced (88% ee). This
increase in enantioselectivity is in sharp contrast to the reac-
tions involving non-fluorous amino alcohol 6 (Figure 1),
where the ee of the product gradually decreases from 92 to
89% in the fluorous biphasic system as the reaction temper-
ature was increased from 0 to 608C, respectively. Although
the origin of this peculiar temperature effect is not clear, we
reason that it could be partly due to an increased solubility
of fluorous amino alcohol 5 at elevated temperature.
Consequently, more catalysts are available at higher
temperature, culminating in the unprecedented enan-
tioselectivity improvement. When the reaction with
amino alcohol 5 was run in hexane at 08C, the prod-
uct ee was 92% (Figure 1). In hexane, reactions with
both amino alcohols 5 and 6 led to slightly reduced
enantioselectivities when the temperature was in-
creased. Amino alcohol 6 (Figure 1) gave products
with somewhat higher enantioselectivities than 5, but
the trend of decreasing selectivity according to the
temperature rise was the same. When n-butyllithium
Abstract in Korean:
946
ꢂ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 945 – 950

Asymmetric Additions
FULL PAPER
was omitted from the reaction employing 6 at 08C, the ee
dropped to 72%. Moreover, product of only 57% ee was
obtained by switching the solvent to toluene. The reaction
was relatively fast compared with the previously reported
ones.[7] Monitoring the aldehyde conversion in the reaction
using amino alcohol 5 at 408C with a React-IR indicated
that it was complete within 30 min. At 608C the reaction
took only 10 min.
at room temperature. After cooling to 08C, however, two
phases separated. When the reaction was carried out at
408C with the fluorous amino alcohol 5, the reaction
medium still remained separated in two phases. The
mixture was cooled to 08C before phase separation. It was
1
found by H NMR spectroscopy that the peaks of fluorous
amino alcohol 5 were not detected from the organic phase,
and the amino alcohol was recovered from the fluorous
phase almost quantitatively, although the partition coeffi-
cient of the free amino alcohol 5 in FC-72 over hexane is
only 2.3. This effect could be explained by the formation of
aggregates of the ligand 5-Li·Et2Zn complex,[11] which
should enforce the catalyst system to stay in the fluorous
phase.
Table 1. Enantioselectivities of the diethylzinc addition reactions to ben-
zaldehyde in monophasic and biphasic systems depending on the temper-
ature.
Solvent
ee [%][b]
408C
Table 3. Partition coefficients of amino alcohol 5 between representative
organic solvents and FC-72.
(ligand[a]
)
08C
208C
608C
FC-72/hexane (5)
FC-72/hexane (6)
Hexane (5)
81
86
91
90
94
92 (80[c]
89
)
88 (84[d]
)
Solvent
Partition coefficient[a]
(P=Cfluorous phase/Corganic phase
92
92 (72[d], 57[e]
95
87
86
91
)
)
90
93
CH3CN
CHCl3
hexane
THF
47
Hexane (6)
1.6
2.3
1.2
2.5
[a] Reactions were carried out at 0.15m concentration by using 3 mol%
of amino alcohol and 3.7 mol% of nBuLi. [b] Enantioselectivities and
yields were determined by HPLC and/or GC analysis using a Chiralcel
OD and/or a cyclodex b column. In all cases the yields were over 95%,
with 2–3% of benzyl alcohol as a side product. [c] Results with 1 mol%
of amino alcohol. [d] Result without nBuLi addition. [e] Toluene was
used as a solvent.
toluene
[a] Determined as follows: A mixture of 50 mg of amino alcohol 5 in FC-
72 (1 mL) and an organic solvent (1 mL) was stirred at rt for 10 min.
Then the two phases were separated at 08C and solvents in each phase
were evaporated in vacuo. The contents of the fluorous amino alcohol 5
in each phase were determined by measuring the weights of the residue.
In order to examine the scope of the fluorous ligand
system, Et2Zn addition reactions to other aldehydes were
carried out at 408C in fluorous biphasic solvents in the pres-
ence of the amino alcohol 5 (Table 2). While the reaction
with p-chlorobenzaldehyde exhibited similar enantioselectiv-
ity (90% ee) to that of benzaldehyde, reactions with p-me-
thoxybenzaldehdye and cinamaldehyde showed somewhat
decreased selectivities (83 and 78% ee, respectively).
Repetition experiments by using 5 were carried out in
FC-72/hexane, and it was found that the fluorous amino al-
cohol could be recovered by simple phase separation and
used at least nine times without significant loss of enantiose-
lectivity and reactivity. The results of these experiments are
summarized in Table 4. After the ninth reaction a product
of 81% ee was obtained, which is about the value (80% ee)
obtained from the reaction with 1 mol% of catalyst at the
same temperature. This result indicates that after nine runs
the remaining catalyst in the reaction mixture may be ap-
proximately 1 mol%.
Recently, Katagiri et al. reported that fluorinated amino
alcohol promoted unexpectedly higher aggregation of dieth-
ylzinc species, which showed an interesting concentration
effect on the enantioselectivity. Compared to the non-fluo-
rous ligand a higher concentration of the fluorous ligand
was required to give better enantioselectivities.[12] Gratify-
ingly, in our case a small quantity of 5 was sufficient to
ensure high enantioselectivity (Table 5). To our surprise, we
observed almost constant ee values irrespective of the cata-
lyst loading until 0.1 mol%, which may indicate to the
amount of the dissolved catalyst available in this reaction
mixture. This is in sharp contrast to the result obtained with
amino alcohol 6, where a gradual decrease of the enantiose-
lectivity upon decreasing amount of the amino alcohol was
observed. The most noticeable difference in that respect was
observed in reactions involving 0.05 mol% of amino alco-
Table 2. Enantioselectivity of the diethylzinc addition to aldehydes at 408C
in FC-72/hexanes (1/1).
Entry
Aldehyde
Catalyst nBuLi
loading [mol%]
[mol%]
Reaction Yield
t [h]
ee
[%][a] [%][b]
1
2
p-chlorobenzaldehyde
p-methoxybenzalde-
hyde
3
3
3.6
3.6
1
1
90
80
90[c]
83[d]
3
cinnamaldehyde
3
3.6
1
85
78[e]
[a] After column chromatography. [b] Determined by HPLC analysis by
using chiral columns. [c] Chiralcel OD-H 99:1 heptane/2-propanol,
1.0 mLminÀ1
. .
[d] Chiralcel AD 98:2 heptane/2-propanol, 1.0 mLminÀ1
[e] Chiralcel OD-H 98:2 heptane/2-propanol, 1.0 mLminÀ1
.
The approximate partition coefficients of the fluorous
amino alcohol 5 between representative organic solvents
and FC-72 were determined experimentally, and the results
are summarized in Table 3. Hexane and FC-72 turned into a
homogeneous phase by addition of fluorous amino alcohol 5
Chem. Eur. J. 2005, 11, 945 – 950
www.chemeurj.org
ꢂ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
947

C. Bolm, B. M. Kim et al.
Table 4. Enantioselectivities of diethylzinc addition reactions to benzaldehyde and recycling of amino alcohol
5 by simple phase separation.
simple phase separation. When
nBuLi was employed as addi-
tive, a decrease in the enantio-
selectivity was observed, which
is in contrast to the results ob-
tained in the previously dis-
cussed diethylzinc additions
(Table 6).
In summary, we prepared a
pyrrolidinylmethanol derivative
from (S)-proline containing per-
Run
1st
92
2nd
92
3rd
91
4th
92
5th
91
6th
91
7th
86
8th
86
9th
82
ee (%) of the product
Table 6. Enantioselectivity in the phenyl transfer reaction to p-chloro-
benzaldehyde catalyzed by amino alcohol 5.
hols 5 and 6, which gave 71 and 30% ee, respectively. The
results are summarized in Table 5.
Table 5. Enantioselectivity in the diethylzinc addition reaction to benzal-
dehyde in hexanes depending on the catalyst loading.
Entry
Solvent
T
[8C]
Reaction
time [h]
ee [%]
Config.
1
2
FC-72/hexane
FC-72/hexane
FC-72/toluene
FC-72/toluene
rt
1
77
87
78
88
S
S
S
S
40
40
40
0.5
0.5
0.5
Entry
Amino
alcohol
[mol%]
Reaction
time
[h]
ee [%][a]
Amino alcohol
3[a]
4
5
6
1[b]
2[b]
3[c]
4[c]
5[c]
6[d]
10
3
1
0.5
0.1
0.05
0.5
1
1
2
4
90
91
94
90
90
71
90
93
90
87
83
30
[a] nBuLi was added as additive.
fluoro-ponytails. It was successfully applied in catalyzed
asymmetric additions of organozinc reagents to aldehydes in
monophasic and biphasic solvent systems. At 408C enantio-
meric excesses of up to 94 and 88% were achieved in Et2Zn
and Ph2Zn additions, respectively. Both reactions were com-
plete within one hour. Reactions employing the fluorous
amino alcohol 5 exhibited a striking temperature effect in
the biphasic solvent system. Thus, the ee of the product in-
creased from 81 to 92% when the temperature was raised
from 0 to 408C. Under optimal conditions, use of only
0.1 mol% of the catalyst in hexane provided the ethylated
product of 90% ee. The perfluoro ligand was easily recov-
ered by simple phase separation, and its reuse in nine se-
quential runs proceeded without significant loss of enantio-
selectivity and reactivity.
20
[a] Enantioselectivities and yields were determined by HPLC and GC
analysis by using a Chiralcel OD and a cyclodex-b column, respectively.
In all cases the yields were over 95% and benzyl alcohol was formed in
2–5%. (The production of benzyl alcohol increased when the catalyst
loading was lowered). [b] Reaction concentrations were 0.15m. [c] Re-
action concentrations were 0.3m. [d] Reaction concentration was 1m.
Next, we examined the performance of the fluorous
amino alcohol 5 in phenyl transfer reactions to an aromatic
aldehyde.[13] Recently, Zhao et al. reported enantioselective
phenylations using pyrrolidinylmethanol derivatives.[9] Vari-
ous (S)-proline-derived amino alcohols were examined and
the best enantioselectivity (92.6% ee, at À308C) was ach-
ieved with 15 mol% of 6, pure Ph2Zn and a slow addition of
the substrate (p-chlorobenzaldehyde). By using only
10 mol% of 6 at À308C, a product with 89.1% ee was ob-
tained. In our study we followed the method developed by
one of the authors employing mixtures of Ph2Zn and Et2Zn
as aryl source.[13] With 10 mol% of amino alcohol 5 at 408C
the desired diarylmethanol having an ee of 88% was ob-
tained almost quantitatively. No ethylation product was ob-
served. Thus, the enantioselectivity was close to the one ob-
tained with 6 at much lower temperature. After the catalysis
amino alcohol 5 was recovered almost quantitatively by
Experimental Section
Compound 2: Magnesium turnings (484 mg, 19.9 mmol) were added to a
two-necked flask charged with Et2O (5 mL). The mixture was sonicated
for 1 h. A solution of 1,4-dibromobenzene (4.69 g, 19.9 mmol) in Et2O
(15 mL) was slowly transferred into the reaction mixture, and stirring
was continued for 2 h. Compound 1 was then added at 08C and the mix-
ture stirred for additional 2 h. After the addition of a saturated aqueous
solution of NH4Cl (100 mL) at 08C, the reaction mixture was extracted
with ethyl acetate (3ꢃ100 mL). Combined organic phases were dried
over anhydrous MgSO4 and concentrated under reduced pressure to give
948
ꢂ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 945 – 950

Asymmetric Additions
FULL PAPER
the crude product. Column chromatography (hexane/ethyl acetate 5:1)
gave the product (6.72 g, 13.9 mmol, 70%). 1H NMR (300 MHz, CDCl3,
1.8 mmol) was added again. Stirring was continued for additional 30 min
at 408C, and benzaldehyde (61 mL, 0.60 mmol) was injected at 408C. The
reactions were repeated 9ꢃ. The monophasic reaction conditions are the
same as those of the biphasic one except for the addition of FC-72.
258C, TMS):
d =7.48–7.42 (m, 4H), 7.28–7.26 (m, 4H), 4.85 (dd,
2J(H,H)=8.9, 4.0 Hz, 1H), 4.20–4.11 (m, 2H), 3.47–3.41 (m, 1H), 3.00–
2.98 (m, 1H), 2.14–2.07 (m, 1H), 1.89–1.86 (m, 1H), 1.59–1.52 (m, 1H),
1.27 (t, 3J(H,H)=7.1 Hz, 3H), 0.92–0.88 (m, 1H); 13C NMR (75 MHz,
CDCl3, 258C, TMS): d =160.0, 142.0, 139.0, 132.1, 131.9, 127.9, 127.5,
85.3, 69.2, 46.5, 29.4, 25.2; HRMS (FAB+): m/z: calcd for C35H30N2O4:
481.9961; found: 481.9989 [M+H]+.
Typical reaction procedure for the diphenylzinc addition: In a glove box
a dried Schlenk flask was charged with diphenylzinc (20 mg, 0.09 mmol).
The flask was sealed and removed from the glove box. Freshly distilled
toluene (500 mL) and FC-72 (500 mL) were added followed by diethylzinc
(24 mL, 0.225 mmol). After stirring the mixture for 30 min at room tem-
perature, compound 5 (10 mg, 0.009 mmol) was added. Stirring was con-
tinued for additional 10 min at 408C, and p-chlorobenzaldehyde (13 mg,
0.09 mmol) was then added directly in one portion. The Schlenk flask
was sealed, and the reaction mixture was stirred at 408C for 30 min. The
mixture was diluted with toluene (2 mL) and FC-72 (2 mL). The upper
toluene phase was decanted carefully under an Ar atmosphere. The or-
ganic phase was washed with water (2 mL). Drying with anhydrous
MgSO4 and evaporation of the solvent from the filtrate under reduced
pressure gave the product in almost quantitative yields. The purity of
product was checked by 1H NMR spectroscopy, and the enantiomeric
excess of the product determined through HPLC analysis.
Compound 3: Compound 2 (2.42 g, 5.00 mmol) was dissolved in MeOH
(15 mL), and after the addition of NaOH (400 mg, 10 mmol) the mixture
was stirred for 4 h at rt. Solvent was evaporated under reduced pressure,
and then water (30 mL) was added. The reaction mixture was extracted
with dichloromethane (3ꢃ50 mL). Combined organic extracts were dried
over anhydrous MgSO4, and the solvent was evaporated to afford the
product (2.19 g, 5.00 mmol) in a quantitative yield. 1H NMR (300 MHz,
CDCl3, 258C, TMS): d = 7.51–7.47 (m, 4H), 7.35 (d, 2J=8.3 Hz, 2H),
7.23 (d, 2J=8.0 Hz, 2H), 4.45 (dd, 2J=10.5, 5.4 Hz, 1H), 3.77–3.68 (m,
1H), 3.29–3.21 (m, 1H), 2.03–1.86 (m, 2H), 1.75–1.68 (m, 1H), 1.14–1.06
(m, 1H); 13C NMR (75 MHz, CDCl3, 258C, TMS): d =159.0, 145.3,
142.8, 131.5, 131.1, 130.3, 129.7, 122.1, 122.0, 81.4, 77.6, 66.4, 62.6, 48.2,
30.1, 23.5, 15.0; HRMS (FAB+): m/z: calcd for C35H30N2O4: 435.9543;
found: 435.9547 [M+H]+.
Acknowledgement
Compound 4: Freshly activated copper (2.75 mmol, 200 mg) and com-
pound 3 (200 mg, 0.46 mmol) were added into a two-neck flask charged
with anhydrous DMSO (1.15 mL). The mixture was heated to 1208C and
perfluorooctyl iodide (291 mL, 1.10 mmol) was slowly added to the reac-
tion mixture over 30 min. Stirring was continued for 24 h at 1208C and
then the mixture was poured into ethyl acetate (10 mL). The precipitated
brownish solid was removed by filtration. The filtrate was concentrated
under reduced pressure to afford a yellowish solid. Recrystallization of
this crude solid from pentane gave the desired product (357 mg, 70%,
0.32 mmol). [a]2D0 =À56.2 (c=0.01 in EtOAc); 1H NMR (300 MHz,
CDCl3, 258C, TMS): d =7.69–7.54 (m, 8H), 4.55 (dd, 2J=10.6, 5.4 Hz,
1H), 3.78–3.72 (m, 1H), 3.32–3.24 (m, 1H), 2.07–1.90 (m, 2H), 1.79–1.75
(m, 1H), 1.14–1.07 (m, 1H); 13C NMR (75 MHz, CDCl3, 258C, TMS): d
=159.8. 146.7, 143.7, 127.7, 127.4, 126.4, 126.0, 85.0, 69.4, 46.4, 29.3, 25.1;
elemental analysis calcd (%) for C34H15F34NO2: C 36.61, H 1.36, N 1.26;
found C 36.71, H 1.40, N 1.21.
Generous financial support to BMK from the SBS Foundation in the
form of Foreign Research Aid Scholarship is acknowledged and J.K.P.
and H.G.L. thank BK21 fellowships from the Ministry of Education, Re-
public of Korea. C.B. is grateful to the Fonds der Chemischen Industrie
and to the Deutsche Forschungsgemeinschaft (DFG) within the SFB 380
“Asymmetric Synthesis by Chemical and Biological Methods” for finan-
cial support.
[1] a) Chiral Catalyst Immobilization and Recycling (Eds.: D. E. DeVos,
I. F. J. Vankelecom, P. A. Jacobs), Wiley-VCH, Weinheim, 2000;
b) S. Brꢄse, F. Lauterwasser, R. E. Ziegert, Adv. Synth. Catal. 2003,
345, 862–929; c) special issue on recoverable catalysts and reagents:
J. A. Gladysz (Ed. ), Chem. Rev. 2002, 102, 3215–3892.
[2] Reviews: a) I. T. Horvꢀth, Acc. Chem. Res. 1998, 31, 641–650;
b) D. P. Curran, Angew. Chem. 1998, 110, 1230–1255; Angew. Chem.
Int. Ed. Engl. 1998, 37, 1174–1196; c) E. de Wolf, G. van Koten, B.-
J. Deelman, Chem. Soc. Rev. 1998, 31, 641–650; d) M. Cavazzini, F.
Montanari, G. Pozzi, S. Quici, J. Fluorine Chem. 1999, 94, 183–193;
e) R. H. Fish, Chem. Eur. J. 1999, 5, 1677–1680; f) T. Kitazume, J.
Fluorine Chem. 2000, 105, 265–278; g) D. P. Curran, Synlett 2001,
1488–1496.
[3] a) S. Takeuchi, Y. Nakamura, Y. Ohgo, D. P. Curran, Tetrahedron
Lett. 1998, 39, 8691–8694; b) Y. Nakamura, S. Takeuchi, Y. Ohgo,
D. P. Curran, Tetrahedron 2000, 56, 351–355.
[4] a) G. Pozzi, F. Cinato, F. Montanari, S. Quici, Chem. Commun. 1998,
877–878; b) G. Pozzi, M. Cavazzini, F. Cinato, F. Montanari, S.
Quici, Eur. J. Org. Chem. 1999, 1947–1955; c) M. Cavazzini, A.
Manfre, F. Montanari, S. Quici, F. Pozzi, Chem. Commun. 2000,
2171–2172.
[5] a) H. Kleijn, E. Rijnberg, J. T. B. H. Jastrzebski, G. van Koten, Org.
Lett. 1999, 1, 853–855; b) Y. Nakamura, S. Takeuchi, Y. Ohgo, D. P.
Curran, Tetrahedron Lett. 2000, 41, 57–61; c) Y. Tian, K. S. Chan,
Tetrahedron Lett. 2000, 41, 8813–8816; d) Y. Nakamura, S. Takeuchi,
K. Okumura, Y. Ohgo, Tetrahedron 2001, 57, 5565–5571; e) Y. Tian,
Q. C. Yang, T. C. W. Mak, K. S. Chan, Tetrahedron 2002, 58, 3951–
3961; f) Y. Nakamura, S. Takeuchi, K. Okumura, Y. Ohgo, D. P.
Curran, Tetrahedron 2002, 58, 3963–3969.
[6] a) For an asymmetric DA reaction: F. Fache, O. Piva, Tetrahedron
Lett. 2001, 42, 5655–5657; b) for an asymmetric Heck reaction: Y.
Nakamura, S. Takeuchi, S. Zhang, K. Okumura, Y. Ohgo, Tetrahe-
dron Lett. 2002, 43, 3053–3056; c) for an asymmetric allylic alkyla-
tion: M. Cavazzini, G. Pozzi, S. Quici, D. Maillard, D. Sinou, Chem.
Commun. 2001, 1220–1221.
Compound 5: DIBAL (0.915 mmol, 763 mL, 1.2m in toluene) was added
to the suspension of compound
4 (204 mg, 0.183 mmol) in toluene
(1 mL). The mixture was heated to 508C for 18 h and then water (5 mL)
was added. The mixture was extracted with ethyl acetate (3ꢃ5 mL).
Combined organic extracts were dried over anhydrous MgSO4, and sol-
vents were removed under reduced pressure to afford the resulting prod-
uct (202 mg, 0.183 mmol) in a quantitative yield. [a]2D0 =+3.25 (c=0.02
in EtOAc); 1H NMR (300 MHz, CDCl3, 258C, TMS): d =7.80 (d, 2J=
8.2 Hz, 2H), 7.68 (d, 2J=8.2 Hz, 2H), 7.51 (d, 2J=8.2 Hz, 4H), 5.2–4.8
(broad, 1H; OH), 3.8–3.5 (m, 1H), 3.2–3.0 (m, 1H), 2.6–2.4 (m, 1H),
2.0–1.5 (m, 7H); 13C NMR (75 MHz, CDCl3, 258C, TMS): d =126.9
(2C), 125.9, 125.8, 77.3, 59.0, 42.8, 29.9, 23.9, 1.1; elemental analysis calcd
(%) for C34H19F34NO: C 37.01, H 1.74, N 1.27; found C 37.45, H 1.80, N
1.26.
Typical procedure for the repetition reactions: nBuLi (14 mL,
0.022 mmol, 1.6m solution in hexanes) was added to a solution of amino
alcohol 5 (20 mg, 0.018 mmol) in FC-72 (1.5 mL) and hexane (1.5 mL).
After stirring for 30 min, Et2Zn (188 mL, 1.83 mmol) was added. Stirring
was continued for additional 30 min at 408C, and benzaldehyde (61 mL,
0.60 mmol) was injected at 408C. After the reaction was complete, the
mixture was cooled to 08C and the upper organic phase was decanted
carefully under an Ar atmosphere. The organic solution was diluted with
hexanes (5 mL) and quenched with a saturated aqueous solution of
NH4Cl (5 mL). The mixture was extracted with hexanes (2ꢃ5 mL). The
combined organic layer was dried over anhydrous MgSO4 and filtered.
Enantioselectivities and yields were determined through HPLC and GC
analyses. In the repetition reactions, additional nBuLi (14 mL,
0.022 mmol, 1.6m solution in hexanes) and hexane (1.5 mL) were added
to the remaining fluorous phase. After stirring for 30 min, Et2Zn (188 mL,
Chem. Eur. J. 2005, 11, 945 – 950
www.chemeurj.org
ꢂ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
949

C. Bolm, B. M. Kim et al.
[7] a) K. Soai, S. Niwa, Chem. Rev. 1992, 92, 833–856; b) L. Pu, H.-B.
Yu, Chem. Rev. 2001, 101, 757–824.
[8] K. Soai, A. Ookawa, T. Kaba, K. Ogawa, J. Am. Chem. Soc. 1987,
109, 7111–7115.
[9] G. Zhao, X.-G. Li, X.-R. Wang, Tetrahedron: Asymmetry 2001, 12,
399–403.
[10] a) G. Liu, J. A. Ellman, J. Org. Chem. 1995, 60, 7712–7713; b) C.
Bolm, C. Tanyeli, A. Grenz, C. L. Dinter, Adv. Synth. Catal. 2002,
344, 649–656.
[11] a) M. Kitamura, S. Suga, M. Niwa, R. Noyori, J. Am. Chem. Soc.
1995, 117, 4832–4842; b) M. Kitamura, S. Suga, H. Oka, R. Noyori,
J. Am. Chem. Soc. 1998, 120, 9800–9809.
[12] T. Katagiri, Y. Fujiwara, S. Takahashi, N. Ozaki, K. Uneyama,
Chem. Commun. 2002, 986–987.
40, 3284–3308; b) C. Bolm, M. Kesselgruber, A. Grenz, N. Her-
manns, J. P. Hildebrand, New J. Chem. 2001, 25, 13–15; c) C. Bolm,
N. Hermanns, M. Kesselgruber, J. P. Hildebrand, J. Organomet.
Chem. 2001, 624, 157–161; d) C. Bolm, N. Hermanns, A. Claßen, K.
MuÇiz, Bioorg. Med. Chem. Lett. 2002, 12, 1795–1798; e) C. Bolm,
M. Kesselgruber, N. Hermanns, J. P. Hildebrand, G. Raabe, Angew.
Chem. 2001, 113, 1536–1538; Angew. Chem. Int. Ed. 2001, 40, 1488–
1490; f) C. Bolm, J. Rudolph, J. Am. Chem. Soc. 2002, 124, 14850–
14851; g) J. Rudolph, T. Rasmussen, C. Bolm, P.-O. Norrby, Angew.
Chem. 2003, 115, 3110–3113; Angew. Chem. Int. Ed. 2003, 42, 3002–
3005; h) N. Hermanns, S. Dahmen, C. Bolm, S. Brꢄse, Angew. Chem.
2002, 114, 3844–3846; Angew. Chem. Int. Ed. 2002, 41, 3692–3694;
i) J. Rudolph, N. Hermanns, C. Bolm, J. Org. Chem. 2004, 69, 3997–
4000.
[13] a) Review: C. Bolm, J. P. Hildebrand, K. MuÇiz, N. Hermanns,
Angew. Chem. 2001, 113, 3382–3407; Angew. Chem. Int. Ed. 2001,
Received: July 10, 2004
Published online: December 20, 2004
950
ꢂ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 945 – 950