Enantioselective catalysis in water: Mukaiyama-aldol condensation promoted by copper complexes of bisoxazolines supported on poly(ethylene glycol)

Article abstract of DOI:10.1039/b409971k

(S)-3-Phenyl-2-aminopropanol-derived bisoxazolines supported on a modified poly(ethylene glycol) were shown to be effective Cu(II) ligands for the enantioselective Mukaiyama-aldol condensation of various aldehydes with the trimethylsilyl keteneacetal of m

Full text of DOI:10.1039/b409971k

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
Enantioselective catalysis in water: Mukaiyama-aldol condensation
promoted by copper complexes of bisoxazolines supported on
poly(ethylene glycol)
Maurizio Benaglia,*a Mauro Cinquini,a,b Franco Cozzia,b and Giuseppe Celentanoc
a
Dipartimento di Chimica Organica e Industriale, Centro di Eccellenza CISI,
Universita’ degli Studi di Milano, Via Golgi 19, I-20133 Milano, Italy.
E-mail: maurizio.benaglia@unimi.it; Fax: +390250314159; Tel: +390250314171
Dipartimento di Chimica Organica e Industriale, CNR-ISTM,
b
Universita’ degli Studi di Milano, Via Golgi 19, I-20133 Milano, Italy
Istituto di Chimica Organica, Facolta’ di Farmacia, Universita’ degli Studi di Milano,
c
Via Venezian 21, 20133 Milano, Italy
Received 2nd July 2004, Accepted 14th September 2004
First published as an Advance Article on the web 15th October 2004
(S)-3-Phenyl-2-aminopropanol-derived bisoxazolines supported on a modified poly(ethylene glycol) were shown
to be effective Cu(II) ligands for the enantioselective Mukaiyama-aldol condensation of various aldehydes with the
trimethylsilyl keteneacetal of methyl isobutyrate carried out in water. Enantiomeric excesses comparable to those
obtained with nonsupported ligands in the same solvent were observed. The solubility of the ligand in water, ensured
by the presence of the polymeric support, allowed a very convenient catalyst-recycling procedure involving simple
removal of the reaction product by extraction in Et2O and addition of fresh reagents to the catalyst-containing
aqueous solution. The chemical and stereochemical efficiency of the catalyst was only marginally eroded over its use
in three reaction cycles.
Introduction
The several advantages offered by the use of water as a solvent in
terms of cost, safety, and environmental impact, prompted the
development of suitable methods and conditions to carry out
some organic reactions in this solvent.1 However, the number
of these processes is drastically reduced when enantioselective
catalytic reactions are considered.2 Solubility reasons are largely
responsible for this fact and a variety of expedients have been
adopted to circumvent this problem. These mainly include
the use of organic cosolvents, surfactants, and hydrophilic
auxiliaries.
The presence of an organic cosolvent (typically EtOH or THF
as the largely dominant component of mixtures with water)
subtracts most of its appeal to operating in what, even if is
defined an aqueous medium, really is a very wet organic solvent.
The use of surfactants, that in some cases led to very interesting
results, suffers from a certain degree of unpredictability, due to
modes of action largely depending on the type of reaction.2 In
Chart 1 PEG-supported bisoxazolines 1a–d.
contrast to this, the method of grafting hydrophilic residues onto
water insoluble reactants relies on a simple concept (increased
fresh substrate and reagent to the catalyst-containing aqueous
solubility = increased reactivity) and has found extensive
solution. Here, we wish to report the results of this study.
applications. For example, in medicinal chemistry chains of
poly(ethylene glycols) (PEGs) have been widely employed to
Results and discussion
increase the bioavailability of certain drugs.3
Over the last few years, we have been interested in the
immobilization of chiral ligands and catalysts on modified
PEGs.4 Among the supported catalytic systems developed,
bisoxazolines 1a and 1b (PEG-Box, Chart 1) were found to be
efficientandrecyclableCu-ligandsforcyclopropanationandene-
type reactions, affording the products in enantiomeric excesses
(ee) comparable to those obtained with the nonsupported
ligands.4a,5 The presence of the hydrophilic PEG chain (average
Mr 5000 daltons) in these derivatives led us to investigate the
use of PEG-Box to perform enantioselective Mukaiyama-aldol
condensations in water.6,7 The solubility of PEG-Box in water
could also allow a very simple catalyst recycling, based on
extraction of the reaction product in an organic solvent in which
the catalyst is not soluble (Et2O, t-BuOMe) and addition of
Inspection of the literature data showed that only two examples
of enantioselective catalytic Mukaiyama-aldol reactions in
water have been reported. Kobayashi et al. described the
reaction shown in equation 1 (Scheme 1) that afforded product
(S)-2 in 86% yield and 53% ee when carried out in the presence of
3 mol% of Triton®X-100 as a surfactant.6b Wang et al. reported
that in the presence of the gallium (III) complex of a chiral
half-crown ether the reaction shown in equation 2 (Scheme 1)
proceeded in 41% yield to afford a 90:10 syn:anti mixture of
aldol products, the syn isomer (S,S)-3 having an 84% ee.6e
The use of the Cu(II) complexes of PEG-Box 1a, 1b in these
reactions led to the products in low yield and moderate ee. For
example, when the condensation in eq. 2 was carried out with
Cu(SbF6)2/1a as the catalyst (20 mol%, water, 0 to 23 °C, 24 h)
T h i s j o u r n a l i s
©
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 4 0 1 – 3 4 0 7
3 4 0 1

View Article Online
Scheme 1 Examples of enantioselective catalytic Mukaiyama-aldol
reactions carried out in water.
the aldol adduct was isolated in 17% yield as a 75:25 mixture
of syn:anti isomers, compound 3 having 38% ee. Based on the
enhanced performances offered by bisoxazolines featuring a
benzyl or an iso-propyl group at the stereogenic center when
the reaction in eq. 1 was carried out in EtOH:water, 9:1,6b
PEG-Box 1c, 1d were prepared following the reaction sequence
reported in Scheme 2.
Malonate 44a was transformed into bisamides 6a, 6b by a
three step procedure involving ester hydrolysis, acid chloride
formation, and reaction with aminoalcohols 5a, 5b. The crude
products, obtained in 89 and 92% yields, respectively, were
used as such owing to the observation that chromatographic
purification resulted in some decomposition. Oxazoline-ring
closure was performed following the procedure of Denmark
et al.8 to give 7a, 7b (78% yield for both products). Deallylation4a
gave the crude phenols 8a,b that were supported on a modified
PEG by reaction with the readily available mesylate 9a9 in the
presence of Cs2CO3 (DMF, 55 °C, 60 h).4a The yields (two steps)
for 1c and 1d were 78 and 85%, respectively. By the same reaction
sequence, bisoxazoline 10 supported on Triton® X-405 (reduced)
was prepared in similar yield by reaction of phenol 8b with
ad hoc synthesized mesylate 9b.10
Ligands 1c, d were then tested in the synthesis of adduct 2, that
was chosen as the model reaction (20 mol% of ligand, 19 mol%
of copper salt, 1.5 mol equiv. of silyl keteneacetal, water, 0 to
23 °C, 24 h). This relatively large amount of catalyst was selected
in agreement with that used in similar reactions promoted by
polymer-supported bisoxazolines.5 From the data collected in
Table 1 (entries 1–4) it can be seen that the use of the benzyl-
substituted PEG-Box 1d led to the production of (S)-2 in ee
(determined by HPLC on a chiral stationary phase) higher than
those observed with 1c, and comparable to or slightly superior
than those observed by Kobayashi et al. (53%, see above).6b
However, the reaction occurred in poor yield independently
of either the ligand employed, the polymeric support, or the
copper counterion. We believe that the poor solubility of the
aldehyde substrate in water is likely to be responsible for this
result. Triton®-supported ligand 10 performed slightly better
than 1d in terms of stereocontrol but was worse in terms of
chemical yield.
Scheme 2 Synthesis of PEG- and Triton®-supported bisoxazolines
1c, 1d and 10.
those obtained for adducts 11 (R = 4-NO2Ph, up to 58% yield)
and 12 (R = 3-NO2Ph, 55% yield). The yield could be futher
increased by using a larger amount of catalyst. For example,
with 30 mol% of catalyst 1d/Cu(OTf)2 adduct 11 was isolated in
83% yield (entry 6). As in the case of the reaction carried out on
benzaldehyde, the condensation involving 4-NO2 benzaldehyde
proceeded with slightly higher ee in the presence of Cu(SbF6)2
(entry 7) and Triton®-supported ligand 10 (entries 8 and 9). In
contrast to PEG-Box 1d however, ligand 10 was partially soluble
in Et2O, and could be completely separated from the reaction
product only by column chromatography.11 For this reason, the
recycling experiments described below were carried out using the
catalyst derived from ligand 1d.
Heteroaromatic aldehydes (R = 2-furyl and 2-thiazolyl) could
also be employed in this reaction (entries 11 and 12). The lower
ee observed in the case of adduct 14 (R = 2-thiazolyl, ee 31%)
was tentatively explained as interference by the heterocyclic
nitrogen upon the ligand–copper complexation, resulting in
lower stereocontrol. Support of this explanation was found
The reaction was then extended to other aldehydes to afford
the products 11–14, which are presented in Chart 2. As can be
seen from the data reported in Table 1, the use of electron-poor
aldehydes resulted in improved condensation yields, such as
3 4 0 2
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 4 0 1 – 3 4 0 7

View Article Online
Table 1 Enantioselective synthesis of adduct 2 catalyzed by Cu(II) complexes of PEG-box
Entry
Catalyst
RCHO
Product
Yield (%)a
Ee (%)b
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
1c/Cu(OTf)2
1d/Cu(OTf)2
1d/Cu(SbF6)2
10/Cu(OTf)2
1d/Cu(OTf)2
1d/Cu(OTf)2c
1d/Cu(SbF6)2
10/Cu(OTf)2
10/Cu(SbF6)2
1d/Cu(OTf)2
1d/Cu(OTf)2
1d/Cu(OTf)2
1d/Cu(OTf)2d
1d/Cu(OTf)2e
1d/Cu(OTf)2f
1d/Cu(OTf)2g
Ph
Ph
Ph
Ph
2
2
34
26
13
12
40
83
48
58
41
55
51
75
38
38
50
51
40
55
62
56
50
54
60
58
63
51
45
31
45
43
45
38
2
2
4-NO2Ph
4-NO2Ph
4-NO2Ph
4-NO2Ph
4-NO2Ph
3-NO2Ph
2-Furyl
2-Thiazolyl
4-NO2Ph
4-NO2Ph
3-NO2Ph
3-NO2Ph
11
11
11
11
11
12
13
14
11
11
12
12
a Isolated yields after flash chromatography. b As determined by HPLC on a chiral stationary phase (see Experimental section). c Reaction carried out
in the presence of 0.3 mol equiv. of catalyst. d With a catalyst sample recycled from the reaction of entry 5. e With a catalyst sample recycled from the
reaction of entries 5 and 13. f With a catalyst sample recycled from the reactions of entry 5. g With a catalyst sample recycled from the reactions of
entries 5 and 15.
Conclusions
In conclusion, two new PEG-supported enantiomerically
pure bisoxazolines have been prepared and employed in
combination with Cu(II) salts as catalysts for the Mukaiyama-
aldol condensation between the trimethylsilyl keteneacetal of
methyl isobutyrate and various aldehydes, carried out in water
as the only reaction solvent. Enantiomeric excesses similar to
those obtained under the same conditions with nonsupported
ligands were observed. The reaction proceeded in higher yields
when electron-poor aldehydes were used as the substrates. The
high solubility of the ligand in water allowed a very convenient
Chart 2 Structures of aldol products 11–14.
catalyst-recycling procedure involving simple removal of the
reaction product by extraction in Et2O and addition of fresh
when the condensation was extended to 2-pyridylcarbaldehyde
to afford a racemic product in 77% yield. The existence of a
negative stereochemical effect exerted by a chelating substituent
on the aldehyde was further demonstrated by the fact that the
condensation involving ethyl glyoxalate as the aldehyde afforded
the product in 70% yield and only 13% ee.
reagents to the catalyst-containing aqueous solution. The
chemical and stereochemical efficiency of the catalyst was only
marginally eroded over its use in three reaction cycles. Work is in
progress to find new applications of PEG-Box as chiral ligands
for catalytic enantioselective reactions in water.
An interesting opportunity offered by the use of water as
reaction solvent for these Cu(II)–PEG-Box catalyzed aldol
reactions is represented by the possibility of recycling the
catalyst simply by removing the product via extraction with a
solvent in which the catalyst is not soluble, and adding fresh
reagents to the catalyst-containing aqueous phase.12 This
procedure would represent an obvious advantage over the use
of Cu–PEG-Box in organic solvent, the recycling of which
requires: i) ligand decomplexation with cyanide ions; ii) ligand
precipitation with Et2O; and iii) metal replenishment to obtain a
species as catalytically active as a freshly prepared catalyst.4a,13
To demonstrate the feasibility of this simple recycling
procedure, the synthesis of adduct 11 was performed three times
under the conditions of entry 5 (entries 5, 13, and 14 in Table 1).
In these experiments the product was extracted with diethylether
after each cycle and the reagents were added to the aqueous
phase containing the catalyst to run the subsequent cycle. We
were pleased to find that only a small decrease in chemical yield
(from 40%, 1st cycle, to 38%, 2nd and 3rd cycles) and in ee (from
50%, 1st cycle, to 45 and 43%, 2nd and 3rd cycles, respectively)
was observed in these experiments, hence allowing one to avoid
the tedious procedures often associated with the recovery,
reactivation, and recycling of polymer-supported metal-based
catalysts.12 Catalyst recycling could also be performed using the
same catalyst in different reactions. Thus, the catalyst employed
for the synthesis of adduct 11 (entry 5, Table 1) was successfully
recycled twice for the preparation of adduct 12 (entries 15 and
16, Table 1).
Experimental
General
1H NMR spectra were recorded at 300 MHz in chloroform-
d (CDCl3) unless otherwise stated, and were referenced to
tetramethylsilane (TMS) at 0.00 ppm; peak assignments were
based on direct and long-range C–H correlations as well as on
two-dimensional experiments. 13C NMR spectra were recorded
at 75 MHz and were referenced to 77.0 ppm in CDCl3. Optical
rotations were measured at the Na–D line in a 1 dm cell at 22 °C.
IR spectra were recorded on thin film or as solutions in DCM.
After reactions, PEG-supported product purification involved
evaporation of the reaction solvent in a vacuum and addition
of the residue dissolved in 2 cm3 of CH2Cl2 to Et2O (50 cm3 g−1
of polymer), which was stirred and cooled at 0 °C. After 30 min
stirring at 0 °C, the obtained suspension was filtered through a
sintered glass filter and the solid repeatedly washed on the filter
with Et2O (up to 100 cm3 g−1 of polymer, overall).
Yield and purity determination of PEG-supported compounds
The yield of the PEG-supported compounds were determined
by weight with the assumption that Mr is 5000 Da for the PEG
fragment. The Mr actuallyrangedfrom4500–5500. Theindicated
yields were for pure compounds. The purity of these compounds
was determined by H NMR analysis in CDCl3 at 300 MHz
with pre-saturation of the methylene signals of the polymer at
1
d = 3.63. When recording the NMR spectra, a relaxation time of
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 4 0 1 – 3 4 0 7
3 4 0 3

View Article Online
6 s and an acquisition time of 4 s were used to ensure complete
relaxation and accuracy of the integration. The relaxation delay
was selected after T1 measurements. The integration of the
signals of the PEG-OCH3 fragment at d = 3.36 were used as an
internal standard. The estimated integration error was ± 5%.
The Triton®-supported compounds were handled similarly,
although purity determination was more difficult because
Triton®-405 X (reduced) is a mixture of several species.
used in subsequent reactions. An analytically pure sample was
obtained by flash chromatography, with a 2:8 CH2Cl2 :AcOEt
mixture as eluant, as a pale yellow thick oil that solidified on
standing in the freezer (the solid melted when warmed-up to
room temperature). It had [a]2D2 −13.1 (c 0.4 in CH2Cl2). Found:
C, 72.4; H, 7.2; N, 5.3; C32H38N2O5 requires: C, 72.4; H, 7.2;
N, 5.3%. IR: mmax/cm−1 3340, 1661. 1H NMR: d 7.31–7.10 (10H,
m, aromatic H of two benzyl groups), 6.93 (1H, d, J = 8.2 Hz,
NH), 6.84 (2H, B part of an AB system, J = 8.6 Hz, aromatic
H meta to allyloxy group), 6.72 (2H, A part of an AB system,
J = 8.7 Hz, aromatic H ortho to allyloxy group), 6.54 (1H, d,
J = 8.0 Hz, NH), 6.12–5.95 (1H, m, CHCH2), 5.40 (1H, d,
J = 17.2 Hz, CCHH), 5.28 (1H, d, J = 10.2 Hz, CCHH),
4.48 (2H, d, J = 5.4 Hz, ArOCH2), 4.25–4.15 (2H, m, two
CHN), 3.75–3.65 (2H, m, CH2OH), 3.56–3.43 (2H, m, CH2OH),
3.35 (1H, bs, OH), 3.20 (1H, bs, OH), 3.00–2.70 (6H, m, three
CH2Ar), 1.12 (3H, s, Me on quaternary aliphatic C). 13C NMR:
d 173.6 (CO of one amide group), 172.8 (CO of one amide
group), 157.0 (C–allyloxy), 137.5 (aromatic C bound to CH2 in
one aminoalcohol residue), 137.1 (aromatic C bound to CH2 in
one aminoalcohol residue), 133.2 (CHCH2), 130.9 (2 × C meta
to allyloxy), 129.1 (2 × C ortho to CH2 in the phenyl ring of one
aminoalcohol residue), 129.0 (2 × C ortho to CH2 in the phenyl
ring of one aminoalcohol residue), 128.6 (2 × C meta to CH2 in
the phenyl ring of one aminoalcohol residue), 128.5 (2 × C meta
to CH2 in the phenyl ring of one aminoalcohol residue), 128.3
(C para to allyloxy), 126.7 (C para to CH2 in the phenyl ring of
one aminoalcohol residue), 126.6 (C para to CH2 in the phenyl
ring of one aminoalcohol residue), 117.5 (CHCH2), 114.3
(2 × C ortho to allyloxy), 68.7 (O–CH2CH), 64.2 (CH2OH of
one aminoalcohol residue), 63.9 (CH2OH of one aminoalcohol
residue), 54.6 (quaternary aliphatic C), 53.1 (NHCH of one
aminoalcohol residue), 52.8 (NHCH of one aminoalcohol
residue), 43.2 (CH2 bound to the quaternary aliphatic carbon)
36.9 (CH2 of benzyl group of one aminoalcohol residue), 36.7
(CH2 of benzyl group of one aminoalcohol residue), 18.2 (Me
bound to quaternary aliphatic C).
(S,S)-Bis-N-1-[1-(1-methylethyl)-2-hydroxyethyl]-2-methyl-2-[4-
(1-propenyl-3-oxy) phenylmethyl] 1,3-propandiamide 6a
To a stirred solution of aminoalcohol 5a (1.52 g, 14.69 mmol)
and triethylamine (5.10 cm3, 36.72 mmol) in dry CH2Cl2 (25 cm3)
stirred under nitrogen and cooled at 0 °C, a solution of the acid
dichloride obtained from diester 44a (2.30 g, 7.64 mmol) in
CH2Cl2 (5 cm3) was added dropwise. The mixture was stirred
for 3 h while the reaction temperature was allowed to rise to
23 °C and the reaction was quenched by addition of a saturated
aqueous solution of ammonium chloride (30 cm3). The aqueous
phase was extracted three times with CH2Cl2 and the combined
organic phases were dried and concentrated under vacuum to
give the crude product (2.96 g, 6.80 mmol, 89%). This was shown
by 1H NMR analysis to be pure enough to be used in subsequent
reactions. An analytically pure sample was obtained by flash
chromatography, with a 2:8 CH2Cl2 :AcOEt mixture as eluant,
as a pale yellow thick oil that solidified on standing in the freezer
(the solid melted when warmed-up to room temperature). It had
[a]2D2 −18.8 (c 0.46 in CH2Cl2). Found: C, 66.5; H, 8.7; N, 6.3;
C24H38N2O5 requires: C, 66.3; H, 8.8; N, 6.5%. IR: mmax/cm−1
1
3368, 1661. H NMR: d 7.08 (2H, B part of an AB system,
J = 8.5 Hz, aromatic H meta to allyloxy group), 7.01 (1H, d,
J = 8.8 Hz, NH), 6.79 (2H, A part of an AB system, J = 8.5 Hz,
aromatic H ortho to allyloxy group), 6.63 (1H, d, J = 8.6 Hz,
NH), 6.07–5.95 (1H, m, CHCH2), 5.39 (1H, d, J = 17.2 Hz,
CCHH), 5.27 (1H, d, J = 10.2 Hz, CCHH), 4.48 (2H, d,
J = 5.2 Hz, ArOCH2), 3.80–3.45 (8H, m, two CH2OH and two
CHN), 3.35 (1H, B part of an AB system, J = 13.3 Hz, CHHAr),
2.98 (1 H, part A of an AB system, J = 13.3 Hz, CHHAr), 1.85–
1.70 (2H, m, two CHMe2), 1.29 (3H, s, Me bound to quaternary
aliphatic C), 0.94 (3H, d, J = 6.9 Hz, one Me of i-Pr groups),
0.92 (3H, d, J = 6.8 Hz, one Me of i-Pr groups), 0.87 (3H, d,
J = 6.7 Hz, one Me of i-Pr groups), 0.78 (3H, d, J = 6.8 Hz,
one Me of i-Pr groups). 13C NMR: d 174.0 (CO of one amide
group), 173.4 (CO of one amide group), 157.6 (C–allyloxy),
133.2 (CHCH2), 131.0 (2 × C meta to allyloxy), 128.6 (C para
to allyloxy), 117.5 (CHCH2), 114.4 (2 × C ortho to allyloxy),
68.7 (ArOCH2), 63.6 (HOCH2 of one aminoalcohol residue),
63.5 (HOCH2 of one aminoalcohol residue), 57.4 (NHCH of
one aminoalcohol residue), 57.2 (NHCH of one aminoalcohol
residue), 55.0 (quaternary aliphatic C), 43.5 (ArCH2), 29.0
(CHMe2 of one aminoalcohol residue), 28.8(CHMe2 of one
aminoalcohol residue), 19.5 (one Me of i-Pr), 18.7 (one Me of
i-Pr), 18.5 (one Me of i-Pr), 18.4 (one Me of i-Pr), 18.3 (Me
bound to quaternary aliphatic C).
Synthesis of Box 7a
To a stirred solution of bis-amide 6a (0.30 g, 0.68 mmol) and
triethylamine (0.42 cm3, 3.00 mmol) in CH2Cl2 (5 cm3) kept
under nitrogen and cooled at 0 °C, mesyl chloride (0.12 cm3,
1.50 mmol) dissolved in CH2Cl2 (1 cm3) was added dropwise.
The mixture was stirred at RT for 0.5 h. A saturated aqueous
solution of ammonium chloride (6 cm3) was then added, and the
aqueous phase was extracted three times with CH2Cl2 (15 cm3).
The combined organic phases were dried and concentrated
under vacuum to give the crude product. This was added to a
0.5 M solution of NaOH in methanol:water 1:1 (4 cm3) and the
resulting mixture was stirred at reflux for 3 h. The mixture was
then cooled, the organic solvent was evaporated under vacuum,
and the aqueous phase was extracted three times with CH2Cl2
(15 cm3). The combined organic phases were washed with a
saturated aqueous solution of NaCl, dried and concentrated
under vacuum to give the crude product that was purified
by flash chromatography with a 6:4 CH2Cl2 :AcOEt mixture
as eluant. The product (0.21 g, 0.53 mmol, 78% yield) was a
pale yellow thick oil that solidified on standing in the freezer
(the solid melted when warmed-up to room temperature).
It had [a]2D2 −66.4 (c 0.45 in CH2Cl2). Found: C, 72.0; H, 8.6;
N, 7.2; C24H34N2O3 requires: C, 72.3; H, 8.6; N, 7.0%. IR:
mmax/cm−1 1656. 1H NMR: d 7.09 (2H, B part of an AB system,
J = 8.7 Hz, aromatic H meta to allyloxy group), 6.80 (2H, A
part of an AB system, J = 8.6 Hz, aromatic H ortho to allyloxy
group, 6.07–5.95 (1H, m, CHCH2), 5.40 (1H, d, J = 17.3 Hz,
CCHH), 5.27 (1H, d, J = 10.8 Hz, CCHH), 4.51 (2H, d,
J = 5.3 Hz, ArOCH2), 4.30–4.20 (2H, m, CH2O of oxazoline),
4.12–3.92 (2H, m, CH2O of oxazoline), 3.28 (1H, B part of an
AB system, J = 13.5 Hz, CHHAr), 3.23 (1H, A part of an AB
system, J = 13.5 Hz, CHHAr), 1.85–1.70 (2H, m, two CHMe2),
(S,S)-Bis-N-1-(1-phenylmethyl-2-hydroxyethyl)-2-methyl-2-[4-
(1-propenyl-3-oxy) phenylmethyl] 1,3-propandiamide 6b
To a stirred solution of aminoalcohol 5a (1.10 g, 7.29 mmol)
and triethylamine (2.50 cm3, 18.22 mmol) in dry CH2Cl2 (20 cm3)
stirred under nitrogen and cooled at 0 °C, a solution of the acid
dichloride obtained from diester 44a (1.14 g, 3.79 mmol) in
CH2Cl2 (3 cm3) was added dropwise. The mixture was stirred
3 h while the reaction temperature was allowed to rise to 23 °C
and the reaction was quenched by addition of a saturated
aqueous solution of ammonium chloride (30 cm3). The
aqueous phase was extracted three times with CH2Cl2 and the
combined organic phases were dried and concentrated under
vacuum to give the crude product (1.85 g, 3.48 mmol, 92%).
This was shown by 1H NMR analysis to be pure enough to be
3 4 0 4
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 4 0 1 – 3 4 0 7

View Article Online
1.43 (3H, s, Me bound to quaternary aliphatic C), 0.94 (3H, d,
J = 6.9 Hz, one Me of i-Pr groups), 0.90 (3H, d, J = 7.3 Hz, one
Me of i-Pr groups), 0.87 (3H, d, J = 7.0 Hz, one Me of i-Pr
groups), 0.83 (3H, d, J = 6.7 Hz, one Me of i-Pr groups). 13C
NMR: d 168.0 (CN of one oxazoline), 167.9 (CN of one
oxazoline), 157.5 (C–allyloxy), 131.8 (2 × C meta to allyloxy),
130.8 (CHCH2), 129.4 (C para to allyloxy), 117.9 (CHCH2),
114.6 (2 × C ortho to allyloxy), 72.4 (CH of one oxazoline), 72.0
(CH of one oxazoline), 70.4 (CH2 of one oxazoline), 70.1(CH2
of one oxazoline), 69.2 (ArOCH2), 43.9 (quaternary aliphatic
C), 41.7 (ArCH2), 32.9 (CHMe2), 32.6(CHMe2), 21.6 (Me on
quaternary aliphatic C), 19.2 (one Me of i-Pr), 19.0 (one Me of
i-Pr), 18.2 (one Me of i-Pr), 17.8 (one Me of i-Pr).
and PPh3 (0.024 g, 0.09 mmol) was refluxed for 60 min. The
resulting mixture was cooled at RT and SiO2 (1 g) was added in
one portion. After 15 min stirring at RT the mixture was filtered
through a Celite cake, the solvent was evaporated under vacuum
to give the crude product (0.069 g, 0.19 mmol, 95% yield) as a
pale yellow thick oil that was used as such for the subsequent
transformation.
Synthesis of Box 8b
A solution of O-allyl protected Box 7b (0.45 g, 0.90 mmol) in
ethanol (14.0 cm3) containing Pd(OAc)2 (0.020 g, 0.09 mmol)
and PPh3 (0.105 g, 0.40 mmol) was refluxed for 60 min. The
resulting mixture was cooled at RT and SiO2 (2 g) was added in
one portion. After 15 min stirring at RT the mixture was filtered
through a Celite cake, the solvent was evaporated under vacuum
to give the crude product (0.39 g, 0.86 mmol, 95% yield) as a
pale yellow thick oil that was used as such for the subsequent
transformation.
Synthesis of Box 7b
To a stirred solution of bis-amide 6b (1.48 g, 2.80 mmol) and
triethylamine (1.71 cm3, 12.32 mmol) in CH2Cl2 (25 cm3) kept
under nitrogen and cooled at 0 °C, mesyl chloride (0.48 cm3,
6.16 mmol) dissolved in CH2Cl2 (5 cm3) was added dropwise.
The mixture was stirred at RT for 0.5 h. A saturated aqueous
solution of ammonium chloride (30 cm3) was then added and the
aqueous phase was extracted three times with CH2Cl2 (30 cm3).
The combined organic phases were dried and concentrated under
vacuum to give the crude product. This was added to a 0.5 M
solution of NaOH in methanol:water 1:1 (17 cm3) and the
resulting mixture was stirred at reflux for 3 h. The mixture was
then cooled, the organic solvent was evaporated under vacuum,
and the aqueous phase was extracted three times with CH2Cl2
(35 cm3). The combined organic phases were washed with a
saturated aqueous solution of NaCl, dried and concentrated
under vacuum to give the crude product, that was purified by
flash chromatography with a 6:4 CH2Cl2 :AcOEt mixture as
eluant. The product (1.08 g, 2.18 mmol, 78% yield) was a pale
yellow thick oil that solidified on standing in the freezer (the
solid melted on standing at room temperature). It had [a]2D2 −30.7
(c 0.45 in CH2Cl2). Found: C, 77.4; H, 6.7; N, 5.8; C32H34N2O3
requires: C, 77.7; H, 6.9; N, 5.7%. IR: mmax/cm−1 1656. 1H NMR: d
7.34–7.10(12H,aromaticHof twobenzylgroupsandtwo metaH
of ArO), 6.85 (2H, A part of an AB system, J = 8.6 Hz, aromatic
H ortho to allyloxy group), 6.13–5.99 (1H, m, CHCH2), 5.42
(1H, d, J = 17.0 Hz, CCHH), 5.28 (1H, d, J = 11.0 Hz,
CCHH), 4.53 (2H, d, J = 5.3 Hz, ArOCH2), 4.45–4.35 (2H,
m, two CHN), 4.23 (2H, dt, J = 8.0, 6.5 Hz, CH2O of oxazoline),
4.02 (2H, dt, J = 8.0, 7.6, CH2O of oxazoline), 3.27 (1H, B part
of AB system, J = 13.7 Hz, CHHArO), 3.20 (1H, A part of AB
system, J = 13.7 Hz, CHHArO), 3.18–3.01 (2H, m, CH2Ar),
2.70–2.46 (2H, m, CH2Ar), 1.43 (3H, s, Me bound to quaternary
aliphatic C). 13C NMR: d 168.0 (CN of one oxazoline), 167.9
(CN of one oxazoline), 158.0 (C–allyloxy), 137.8 (aromatic
C bound to CH2 in one oxazoline), 137.7 (aromatic C bound
to CH2 in one oxazoline), 133.3 (CHCH2), 131.4 (2xC meta
to allyloxy), 129.3 (2 × C ortho to CH2 in the phenyl ring of
one oxazoline), 129.2 (2 × C ortho to CH2 in the phenyl ring
of one oxazoline), 128.5 (2 × C meta to CH2 in the phenyl
ring of one oxazoline), 128.4 (2 × C meta to CH2 in the phenyl
ring of one oxazoline), 128.0 (C para to allyloxy), 126.4 (C para
to CH2 in the phenyl ring of one oxazoline), 126.3 (C para to
CH2 in the phenyl ring of one oxazoline), 117.4 (CHCH2),
114.2 (2 × C ortho to allyloxy), 72.1 (CH2 in one oxazoline),
71.8 (CH2 in one oxazoline), 68.8 (O–CH2CH), 67.2 (CH in
one oxazoline), 67.0 (CH in one oxazoline), 43.3 (quaternary
aliphatic C), 41.6 (CH2 of benzyl group of one oxazoline), 41.3
(CH2 of benzyl group of one oxazoline), 41.2 (CH2 bound to
quaternary aliphatic carbon), 21.1 (Me bound to quaternary
aliphatic C).
Synthesis of PEG-supported Box 1c
To a solution of mesylate 9a8 (0.48 g, 0.09 mmol) and Box 8a
(0.047 g, 0.13 mmol) in DMF (4.0 cm3) stirred under nitrogen,
Cs2CO3 (0.14 g, 0.39 mmol) was added. The mixture was
warmed up to 55 °C and stirred at that temperature for 60 h.
Cs2CO3 was then removed by filtration of the cooled mixture
and the solvent was evaporated under vacuum. The residue was
taken up into CH2Cl2 (1 cm3) and the product was precipitated
with Et2O. The product (0.41 g, 0.075 mmol, 82% yield) was
1
isolated by filtration as a pale brown solid. H NMR: d 7.11
(2H, B part of an AB system, J = 8.6 Hz, aromatic H ortho
to methylene group bound to quaternary aliphatic C), 7.07
(2H, B part of an AB system, J = 8.6 Hz, aromatic H meta
to MeOPEGO residue), 6.85 (2H, A part of an AB system,
J = 8.6 Hz, aromatic H meta to methylene group bound to
quaternary aliphatic C), 6.77 (2H, A part of an AB system,
J = 8.6 Hz, aromatic H ortho to MeOPEGO residue), 4.23–3.95
(8H, m, CH2O and CHN of Box, and PEGOCH2CH2OPh), 3.90
(2H, t, J = 6.7 Hz, CH2CH2CH2OPh), 3.84 (2H, t, J = 6.5 Hz,
PEGOCH2CH2OPh), 3.42 (2H, t, J = 6.7 Hz, MeOCH2CH2O),
3.36 (3H, s, MeOPEG), 3.23 (1H, B part of an AB system,
J = 13.5 Hz, one H of methylene group bound to quaternary
aliphatic C), 3.19 (1H, A part of an AB system, J = 13.5 Hz, one
H of methylene group bound to quaternary aliphatic C), 2.73
(2H, t, J = 6.5 Hz, PEGOPhCH2CH2CH2), 2.00–2.10 (2H, m,
PEGOPhCH2CH2CH2), 190–1.65 (2H, m, CHMe2), 1.42 (3H, s,
Me bound to quaternary aliphatic C), 0.93 (3H, d, J = 6.7 Hz,
one Me of i-Pr groups), 0.88 (3H, d, J = 7.2 Hz, one Me of i-Pr
groups), 0.86 (3H, d, J = 7.2 Hz, one Me of i-Pr groups), 0.83
(3H, d, J = 6.7 Hz, one Me of i-Pr groups).
Synthesis of PEG-supported Box 1d
To a solution of mesylate 9a (3.65 g, 0.7 mmol) and Box 8b
(0.512 g, 1.13 mmol) in DMF (12.0 cm3) stirred under nitrogen,
Cs2CO3 (1.10 g, 3.38 mmol) was added. The mixture was warmed
up to 55 °C and stirred at that temperature for 60 h. Cs2CO3
was then removed by filtration of the cooled mixture and the
solvent was evaporated under vacuum. The residue was taken
up into CH2Cl2 (4 cm3) and the product was precipitated with
Et2O. The product (3.49 g, 0.63 mmol, 89% yield) was isolated
by filtration as a pale brown solid. 1H NMR: d 7.35–7.05 (14H,
m, ten hydrogens of benzyl groups and four hydrogens meta to
oxygen atoms in the other two aromatic rings), 6.87–6.75 (4H,
m, four remaining aromatic hydrogens), 4.50–4.40 (2H, m,
CHN), 4.24–3.99 (6H, PEGCH2OPh and CH2O of oxazolines),
3.95–3.83 (4H, m, PEGOCH2CH2OPh and CH2CH2CH2OPh),
3.42 (2H, t, J = 6.7 Hz, MeOCH2CH2O), 3.36 (3H, s, MeOPEG),
3.25 (1H, B part of an AB system, J = 13.5 Hz, one H of
methylene group bound to quaternary aliphatic C), 3.19 (1H,
Synthesis of Box 8a
A solution of O-allyl protected Box 7a (0.081 g, 0.20 mmol)
in ethanol (3 cm3) containing Pd(OAc)2 (0.005 g, 0.02 mmol)
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 4 0 1 – 3 4 0 7
3 4 0 5

View Article Online
A part of an AB system, J = 13.5 Hz, one H of methylene
group bound to quaternary aliphatic C), 3.15–3.00 (2H, m, two
hydrogens of methylenes of oxazoline benzyl group), 2.74 (2H,
t, J = 6.8 Hz, PEGOPhCH2CH2CH2), 2.65–2.50 (2H, m, two
hydrogens of methylenes of oxazoline benzyl group), 2.00–2.10
(2H, m, PEGOPhCH2CH2CH2), 1.42 (3H, s, Me bound to
quaternary aliphatic C).
water (1 cm3), the resulting solution was cooled at 0 °C, and
benzaldehyde (0.027 cm3, 0.27 mmol) and the silyl keteneacetal
(0.081 cm3, 0.40 mmol) were added in this order. The mixture
was stirred for 24 h while the temperature was allowed to rise
to 23 °C. The reaction mixture was then extracted with Et2O
(3 × 5 cm3) and the combined organic phases were dried and
concentrated under vacuum. The residue was purified by flash
chromatography with a hexanes:AcOEt 8:2 mixture as eluant
to afford the product (S)-2 (0.014 g, 0.07 mmol, 26% yield)
as a white solid; mp 70–71 °C (lit.:14 71–72 °C); [a]2D2 +17.0
(c 0.5 in CHCl3) (lit.,14 [a]22 +30.6 (c 1.05 in CHCl3) for a sample
having ee > 99%). The eeDwas established by HPLC analysis on
a Chiralpak AD column, flow rate 0.8 cm3 min−1, k = 210 nm;
hexane:ethanol 95:5; tR 14.3 min (minor) and 20.4 min (major).
The product had 1H NMR data identical to those reported.14
Cu(SbF6)2 waspreparedaccordingtothereportedprocedure.15
The reactions performed using the Cu(SbF6)2–PEG-Box
complexes were carried out following the same procedure
employed in the Cu(OTf)2 catalyzed ones.
Synthesis of mesylate 9b
Mesylation of Triton® X-405 (reduced): To a stirred solution
of Triton® X-405 (reduced) (1.01 g, 0.51 mmol) in CH2Cl2 (8
cm3) kept under nitrogen, mesyl chloride (0.12 cm3, 1.52 mmol)
and trioctylamine (0.89 cm3, 2.03 mmol) were added in this
order. The mixture was stirred at RT for 24 h, and the volatile
materials were removed under vacuum. The resulting residue
was dissolved in CH2Cl2 (1 cm3) and the product was precipitated
with diethylether (70 cm3). The solid was filtered, washed on the
filter with pentane (10 cm3) and dried under high vacuum, to
give 0.780 g of product (75% yield) that was used as such. The 1H
NMR spectrum (see ref. 10) showed the presence of the MeSO2
group at d = 3.10.
Using the procedure described for the synthesis of compound
2, the following products were obtained. The absolute
configurations of these products were tentatively assigned on
the basis of a common eluting behavior on the same chiral
stationary phase.
Insertion of the spacer. To a solution of Triton® X-405
(reduced) mesylate (3.25 g, 1.58 mmol) in dry DMF (11.0 cm3)
kept under nitrogen at 55 °C, 3-(4-hydroxyphenyl)propan-1-ol
(0.482 g, 3.17 mmol) and Cs2CO3 (1.03 g, 3.17 mmol) were
added. The mixture was stirred at that temperature for 46 h.
Cs2CO3 was then removed by filtration of the cooled mixture
and the solvent was evaporated under vacuum. The residue was
taken up into CH2Cl2 (3 cm3) and the product was precipitated
with Et2O. The product (2.36 g, 1.11 mmol, 70% yield) was
isolated by filtration as a pale brown solid. The 1H NMR
spectrum showed the presence of the typical resonances of the
C6H4CH2CH2CH2OH spacer bound to a poly(ethylene glycol)
chain.9
(S)-Methyl 2,2-dimethyl-3-hydroxy-3-(4-nitrophenyl)propanoate
11
The product was a pale yellow solid isolated by flash chromato-
graphy with a hexanes:AcOEt 8:2 mixture as eluant; mp
113–114 °C (lit.,16 114.5 °C); [a]2D2 +2.0 (c 0.62 in CH2Cl2) for
a sample having ee = 50%. The ee was established by HPLC
analysis on a Chiralpak AD column, flow rate 0.8 cm3 min−1,
k = 210 nm; hexane:ethanol 90:10; tR 18.0 min (minor) and
1
20.5 min (major). The product had H NMR data identical to
those reported.16
Mesylation of the Triton® X-405 (reduced)/spacer adduct. To a
stirred solution of the alcohol (1.30 g, 0.61 mmol) in CH2Cl2 (9.6
cm3) kept under nitrogen, mesyl chloride (0.14 cm3, 1.84 mmol)
and trioctylamine (1.08 cm3, 2.45 mmol) were added in this
order. The mixture was stirred at RT for 24 h, and the volatile
materials were removed under vacuum. The resulting residue was
dissolved in CH2Cl2 (1.5 cm3) and the product was precipitated
with Et2O (90 cm3). The solid was filtered, washed on the filter
with pentane (10 cm3), and dried under high vacuum, to give
1.10 g of product (82% yield) that was used as such for the
synthesis of 10. The 1H NMR spectrum showed the presence of
the MeSO2 group at d = 3.07.9
(S)-Methyl 2,2-dimethyl-3-hydroxy-3-(3-nitrophenyl)propanoate
12
The product was a pale yellow solid isolated by flash chromato-
graphy with a hexanes:AcOEt 8:2 mixture as eluant; mp
73–75 °C. Found: C, 57.1; H, 5.85; N, 5.5; C12H15NO5 requires:
C, 56.9; H, 6.0; N, 5.5%. IR: mmax/cm−1 3450, 1720. [a]2D2 +1.8
(c 0.25 in CH2Cl2) for a sample having ee = 51%. The ee was
established by HPLC analysis on a Chiralcel AD column,
flow rate 0.8 cm3 min−1, k = 210 nm; hexane:ethanol 90:10; tR
1
17.8 min (minor) and 20.5 min (major). H NMR: d 8.16–8.30
(2H, m, aromatic H), 7.51–7.80 (2H, m, aromatic H), 5.06 (1H,
s, CHOH), 3.75 (3H, s, MeO), 2.10 (bs, 1H, OH), 1.18 (6H, s,
two Me’s). 13C NMR: d 170.5 (CO), 149.6 (C–NO2), 137.1
(aromatic C–CHOH), 133.9 (C para to NO2), 129.51 (C meta to
NO2), 123.8 (aromatic C between NO2 and alkyl chain), 119.9
(C ortho to NO2), 72.0 (OMe), 51.5 (CHOH), 31.9 (quaternary
aliphatic C), 19.2 (2 × Me).
Synthesis of Triton®-supported Box 10
To a solution of mesylate 9b (1.10 g, 0.50 mmol) and Box 8b
(0.41 g, 0.90 mmol) in DMF (4.5 cm3) stirred under nitrogen,
Cs2CO3 (0.88 g, 2.71 mmol) was added. The mixture was
warmed up to 55 °C and stirred at that temperature for 72 h.
Cs2CO3 was then removed by filtration of the cooled mixture
and the solvent was evaporated under vacuum. The residue was
taken up into CH2Cl2 (1.5 cm3) and the product was precipitated
with Et2O. The product (0.93 g, 72% yield) was isolated by
filtration as a pale brown solid. The 1H NMR spectrum showed
the presence of the benzyl substituted bisoxazoline residue.
(R)-Methyl 2,2-dimethyl-3-hydroxy-3-(2-furyl)propanoate 13
The product was a colorless oil isolated by flash chromatography
with a hexanes:AcOEt 8:2 mixture as eluant, and had 1H NMR
data identical to those reported for the racemic compound.17
A sample of 45% ee had [a]2D2 +7.3 (c 0.1 in CH2Cl2). The ee
was established by HPLC analysis on a Chiralpak AD column,
flow rate 0.8 cm3 min−1, k = 210 nm; hexane:ethanol 95:5; tR
15.1 min (minor) and 22.8 min (major).
Aldol condensations (eq. 1)
The synthesis of (S)-methyl 2,2-dimethyl-3-hydroxy-3-phenyl-
propanoate 2 is illustrative of the general procedure: To a
stirred solution of PEG-Box 1d (0.30 g, 0.053 mmol) in dry
CH2Cl2 (2.0 cm3), Cu(OTf)2 (0.019 g, 0.05 mmol) was added
and the mixture stirred at 23 °C for 3 h. The solvent was then
evaporated under vacuum and the residue was kept under high
vacuum for 5 min. The complex thus obtained was dissolved in
(R)-Methyl 2,2-dimethyl-3-hydroxy-3-(2-thiazolyl)propanoate
14
The product was a pale yellow solid isolated by flash chromato-
graphy with a hexanes:AcOEt 8:2 mixture as eluant; mp
63–65 °C. Found: C, 50.3; H, 5.95; N, 6.6; C9H13NO3S requires:
3 4 0 6
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 4 0 1 – 3 4 0 7

View Article Online
C, 50.2; H, 6.1; N, 6.5%. IR: mmax/cm−1 3520, 1710, 1455.
A sample of 31% ee had [a]2D2 +0.7 (c 0.41 in CH2Cl2). The ee
was established by HPLC analysis on a Chiralcel AD column,
flow rate 0.8 cm3 min−1, k = 210 nm; hexane:ethanol 90:10; tR
on PEG-supported catalytic systems see: T. J. Dickerson, N. N. Reed
and K. D. Janda, Chem. Rev., 2002, 102, 3325–3344; . For a review
on the use of PEG as polymeric support see: D. J. Gravert and
K. D. Janda, Chem. Rev., 1997, 97, 489–509.
5 For an excellent review on the use of supported bisoxazolines in
enantioselective catalysis see: D. Rechavi and M. Lemaire, Chem.
Rev., 2003, 103, 3467–3494; . For recent work not included in this
review see: E. Diez-Barra, J. M. Fraile, I. J. Garcia, E. Garcia-
Verdugo, C. I. Herrerias, S. V. Luis, J. A. Mayoral, P. Sanchez-
Verdù and J. Tolosa, Tetrahedron: Asymmetry, 2003, 14, 773–780;
A. Weissberg and M. Portnoy, Chem. Commun., 2003, 1538–1539;
A. Mandoli, S. Orlandi, D. Pini and P. Salvadori, Chem. Commun.,
2003, 2466–2467.
1
12.5 min (minor) and 31.2 min (major). H NMR: d 7.80 (1H,
d, J = 3.0 Hz, NCH), 7.39 (1H, d, J = 3.0 Hz, SCH), 5.10 (1H,
s, CHOH), 3.72 (3H, s, MeO), 1.15 (6H, s, two Me’s). 13C
NMR: d 171.5 (CO), 161.5 (CN), 131.8 (C–N), 127.7(C–S),
69.7 (OMe), 51.1 (CHOH), 29.7 (quaternary aliphatic C), 19.1
(2 × Me).
6 Recent examples of enantioselective Mukaiyama-aldol condensa-
tions promoted by chiral catalysts in aqueous media (see solvent
in parentheses): (a) S. Kobayashi, S. Nagayama and T. Busujima,
Chem. Lett., 1999, 71–72 (EtOH:water 9:1); (b) S. Kobayashi,
S. Nagayama and T. Busujima, Tetrahedron, 1999, 55, 8739–8746
(EtOH:water 9:1 or only water in the presence of a surfactant);
(c) S. Nagayama and S. Kobayashi, J. Am. Chem. Soc., 2000, 122,
11531–11532(EtOH:water 9:1); (d) S. Kobayashi, T. Hamada,
S. Nagayama and K. Manabe, Org. Lett., 2001, 3, 165–167
(EtOH:water 9:1); (e) H.-J. Li, H.-Y. Tian, Y.-J. Chen, D. Wang
and C.-J. Li, Chem. Commun., 2002, 2994–2995 (EtOH:water
9 : 1 and 1 : 9, or only water); (f) K. Manabe, S. Ishikawa,
T. Hamada and S. Kobayashi, Tetrahedron, 2003, 59, 10439–10444
(THF:water 9:1).
Ligand recovery
The ligands were recovered from the aqueous phase after
product extraction with Et2O using the following procedure;
to the aqueous solution, water (5 cm3), CH2Cl2 (10 cm3),
and solid KCN (two fold molar excess) were added, and the
mixture vigorously stirred at RT for 10 min. The organic phase
was separated and the aqueous phase was extracted twice
with CH2Cl2 (2 × 5 cm3). The combined organic phases were
dried and concentrated under vacuum to a 1 cm3 volume. The
polymer-supported ligand was recovered by precipitation with
Et2O followed by filtration. 1H NMR analysis of the recovered
material did not show any decomposition. Typical recovery
yields ranged from 70 to 85% for the PEG-Box 1c, 1d and from
50 to 65% for Triton®-supported 10, that precipitated less readily
than 1c, 1d from Et2O.
7 Very recently the use of Cu(OTf)2 complexes of dendrimeric
bisoxazolines in
a Mukaiyama-aldol reaction carried out in
aqueous media (organic solvent : water  9 : 1) was reported:
B.-Y. Yang, X.-M. Chen, G.-J. Deng, Y.-L. Zhang and Q.-H. Fan,
Tetrahedron Lett., 2003, 44, 3535–3538.
8 S. E. Denmark, N. Nakajima and J.-C. Nicaise, J. Org. Chem., 1995,
60, 4884–4892.
Catalyst recycling
9 R. Annunziata, M. Benaglia, M. Cinquini and F. Cozzi, Chem. –Eur.
J., 2000, 6, 133–138.
This was simply achieved by removal of the reaction product
and unreacted starting materials from the aqueous reaction
mixture by extraction with Et2O, followed by cooling the
aqueous solution at 0 °C and addition of fresh aldehyde and
silyl keteacetal.
10 To the best of our knowledge, this represents the first example of
ligand supported on Triton®. 1H NMR analysis showed Triton®
X-405 (reduced) to be a mixture of several species; the most abundant
one contains a 1,4 disubstituted cyclohexane ring having a (Me)3C–
CH2–C(Me)2– residue at C-4 and a –O–(CH2CH2O)39CH2CH2OH
chain at C-1; its average molecular weight is 2000 daltons.
11 When the synthesis of adduct 2 was carried out under the
conditions of entry 4 (Table 1), a single washing of the reaction
mixture with Et2O (see Experimental section) extracted about 15%
of the catalyst.
12 For an extensive discussion of the problems associated with chiral
catalyst recovery and recycling see: Chiral Catalyst Immobilization
and Recycling, ed. D. E. De Vos, I. F. J. Vankelecom and P. A. Jacobs,
Wiley-VCH, Weinheim, 2000.
Acknowledgements
This work was supported by MIUR (Progetto Nazionale Stereo-
selezione in Sintesi Organica. Metodologie ed Applicazioni),
Centro di Eccellenza CISI, and CNR.
References and notes
13 The problem of metal leaching from a supported bisoxazoline–metal
catalyst is a general one. For example, also Cu(II) complexes of
polystyrene-supported bisoxazolines required metal replenishment
before reycling when employed in Mukaiyama-aldol reactions:
S. Orlandi, A. Mandoli, D. Pini and P. Salvadori, Angew. Chem.,
Int. Ed., 2001, 40, 2519–2521.
14 F. Fringuelli, O. Piermatti and F. Pizzo, J. Org. Chem., 1995, 60,
7006–7009.
15 D. A. Evans, G. S. Peterson, J. S. Johnson, D. M. Barnes, K. R.
Campos and K. A. Woerpel, J. Org. Chem., 1998, 63, 4541–4544.
16 T. Mukaiyama, H. Fujisawa and T. Nagakawa, Helv. Chim. Acta,
2002, 85, 4518–4531.
1 (a) Organic Synthesis in Water, ed. P. A. Grieco, Blackie Academic
& Professional, London, 1998; (b) S. Ribe and P. Wipf, Chem.
Commun., 2001, 299–307.
2 U. M. Lindström, Chem. Rev., 2002, 102, 2751–2772.
3 C. G. Wermuth, in The Practice of Medicinal Chemistry,
ed. C. G.Wermuth, Academic Press, London, 1996.
4 (a) R. Annunziata, M. Benaglia, M. Cinquini, F. Cozzi and
M. Pitillo, J. Org. Chem., 2001, 66, 3160–3166; (b) M. Benaglia,
M. Cinquini, F. Cozzi, A. Puglisi and G. Celentano, Adv. Synth.
Catal., 2002, 344, 533–542, and references therein; (c) T. Danelli,
R. Annunziata, M. Benaglia, M. Cinquini, F. Cozzi and G. Tocco,
Tetrahedron: Asymmetry, 2003, 14, 461–467; (d) A. Puglisi,
M. Benaglia, M. Cinquini, F. Cozzi and G. Celentano, Eur. J. Org.
Chem., 2004, 567–573; and references therein. For a recent review
17 C. Le Roux, H. Gaspard-Iloughmane and J. Dubac, Bull. Soc. Chim.
Fr., 1993, 130, 832–842.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2 , 3 4 0 1 – 3 4 0 7
3 4 0 7