Polyglycerol-supported Co- and Mn-salen complexes as efficient and recyclable homogeneous catalysts for the hydrolytic kinetic resolution of terminal epoxides and asymmetric olefin epoxidation

Article abstract of DOI:10.1002/ejoc.200701230

In this report we demonstrate the suitability of hyperbranched polyglycerol as high-loading support for asymmetric catalysis. A polyglycerol-supported unsymmetrical salen ligand was prepared and purified by means of ultrafiltration. The polymeric ligand was metalated with cobalt acetate to afford the corresponding supported Co-salen complex which was further utilized in the hydrolytic kinetic resolution (HKR) of terminal epoxides. Kinetic studies for HKR of 1,2-epoxy-hexane using polyglycerol-supported Co-salen showed improved catalytic performance compared to the respective non-immobilized catalyst. This observation points to the presence of a positive dendrimeric effect assuming a cooperative bimetallic mechanism. Furthermore, a polymer-supported Mn-salen catalyst was prepared and successfully applied in the asymmetric epoxidation of olefins. The respective dendritic catalyst was recycled by precipitation up to three times in the epoxidation of 6-cyano-2,2-dimethylchromene (ee up to 95%). Experiments with repetitive batches revealed enhanced stability of the catalyst as a result of immobilization whereby the total turnover number increased from 23 in a single batch to around 80 for four repetitive batches. Wiley-VCH Verlag GmbH & Co. KGaA, 2008.

Full text of DOI:10.1002/ejoc.200701230

FULL PAPER
DOI: 10.1002/ejoc.200701230
Polyglycerol-Supported Co- and Mn-salen Complexes as Efficient and
Recyclable Homogeneous Catalysts for the Hydrolytic Kinetic Resolution of
Terminal Epoxides and Asymmetric Olefin Epoxidation
Maryam Beigi,[a][‡] Sebastian Roller,[b] Rainer Haag,[c] and Andreas Liese*[d]
Keywords: Asymmetric catalysis / Epoxidation / Hyperbranched polymers / Kinetic resolution / Polyglycerol
In this report we demonstrate the suitability of hyper- ative bimetallic mechanism. Furthermore, a polymer-sup-
branched polyglycerol as high-loading support for asymmet- ported Mn-salen catalyst was prepared and successfully ap-
ric catalysis. A polyglycerol-supported unsymmetrical salen
ligand was prepared and purified by means of ultrafiltration.
The polymeric ligand was metalated with cobalt acetate to
afford the corresponding supported Co-salen complex which
was further utilized in the hydrolytic kinetic resolution (HKR)
plied in the asymmetric epoxidation of olefins. The respec-
tive dendritic catalyst was recycled by precipitation up to
three times in the epoxidation of 6-cyano-2,2-dimeth-
ylchromene (ee up to 95%). Experiments with repetitive
batches revealed enhanced stability of the catalyst as a result
of terminal epoxides. Kinetic studies for HKR of 1,2-epoxy- of immobilization whereby the total turnover number in-
hexane using polyglycerol-supported Co-salen showed im-
proved catalytic performance compared to the respective
non-immobilized catalyst. This observation points to the
presence of a positive dendrimeric effect assuming a cooper-
creased from 23 in a single batch to around 80 for four repeti-
tive batches.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2008)
Introduction
Chiral derivatives of the tetradentate Schiff base[1] known
as salen [H2salen = bis(salicylidene)ethylene diamine] have
attracted considerable attention in the field of asymmetric
catalysis.[2] The related structure 1, optimized by Jacobsen
and co-workers[3] is ranked as one of the most important
ligands for chiral catalysts.[4] Over the last decade, diverse
metal complexes of salen have been prepared and success-
fully applied for a multitude of reactions including epoxid-
ation,[5–8] ring opening of epoxides,[9–15] Diels–Alder reac-
tion,[16–19] C–H oxidation,[20–22] C–H amination,[23] hydro-
cyanation[24–26] aerobic oxidative coupling[27] and oxi-
dations of ketone silyl enol ethers.[28–30]
Efforts for the immobilization of achiral salen ligands
went on for a long time,[31] but since the introduction of the
efficient chiral types of salen ligands[3,32–33] as well as the
explosion of interest in combinatorial chemistry, this area
has grown significantly. Preparation of the corresponding
immobilized catalyst is expected to simplify the product re-
covery and to increase the catalyst stability. The advantage
of enhanced stability is especially important in the case of
the salen ligand since the degradation tendency is a major
limitation for some low-molecular weight salen species
such as manganese complexes.[34,35] Numerous accounts for
salen immobilization using diverse kinds of inorganic
(e.g. silica,[19] zeolites,[36,37] inorganic polymers) or organic
supports [e.g. polystyrene,[15,38–41] polyamidoamine (PA-
MAM),[14] poly(ethylene) glycol (PEG),[40] or poly(norbor-
nene)[42]] is already outlined in the literature.
[a] NRW Graduate School of Chemistry,
48149 Münster, Germany
[b] BASF AG (Polymerforschung),
67056 Ludwigshafen, Germany
[c] Institute of Chemistry and Biochemistry, Freie Universität
Berlin,
Takustrasse 3, 14195 Berlin, Germany
[d] Institute of Technical Biocatalysis, Hamburg University of
Technology (TUHH),
Denickestrasse 15, 21073 Hamburg, Germany
Fax: +49-40-42878-2127
E-mail: liese@tuhh.de
[‡] Current address: Department of Organic Chemistry, University
of Dortmund
Among different classes of supports, star-shaped and hy-
perbranched polymers are remarkable candidates. Dendritic
Otto-Hahn Strasse 6, 44221 Dortmund, Germany.
Eur. J. Org. Chem. 2008, 2135–2141
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2135

M. Beigi, S. Roller, R. Haag, A. Liese
FULL PAPER
catalyst[43] preparations are expected to lead to high-load- resolution (HKR) of terminal epoxides and olefin epoxid-
ing catalysts with good solubility due to globular shapes of ation is investigated. Moreover, the catalyst retainability has
the respective polymers which offer the reward of carrying been inspected.
out the reactions in one-phase systems. The main advan-
tages of homogeneously soluble catalysts comprise normal
reaction kinetics, facile compound characterization and ab-
sence of diffusion limitations (which is very often the case
Results and Discussion
with heterogenized catalysts). Furthermore, preparation of
the corresponding homogeneous polymeric catalyst features
the possibility of catalyst recovery via appropriate pro-
Preparation of Co- and Mn-salen Complexes Supported on
Polyglycerol via Unsymmetrical salen Synthesis
cedure (e.g. precipitation, ultrafiltration, etc.) and in con-
clusion, leads to higher Total Turnover Numbers (TTN).
Symmetrical salen preparation can be simply ac-
complished following its facile and general synthesis pro-
Hyperbranched polyglycerol (PG)[44,45] is the polymeric cedure, comprising the condensation of a diamine with two
support that is utilized in the present work. This soluble equivalents salicylaldehyde. In contrast, synthesis of unsym-
polymer is an aliphatic polyether polyol and is easily access- metrical salen bearing different substituents on the two aro-
ible via anionic ring-opening polymerization of glycidol in matic rings turned out to be problematic and therefore, it is
kilogram quantities.[46] PG possess a chemically stable not surprising that there are few publications mainly in the
backbone against a multitude of reagents which affords the recent years dealing with this approach.[14,29,41,42,47,48] Re-
possibility of performing reactions under a wide variety of cently, we have reported a simple and straightforward pro-
conditions.
cedure for preparation of high-loading polyglycerol salen
In the past, we have reported[47] the preparation of a po- analogues.[47,49] The respective method using initial hyper-
lyglycerol-supported Cr-salen complex and its application branched polyglycerol 2 (Mn = 8000 gmol–1) was applied
for an asymmetric Diels–Alder reaction as well as its reten- (Scheme 1) and the final polymeric ligand 3 could be reco-
tion in a membrane reactor. In this contribution, the prepa- vered in 65% yield. The NMR analysis indicated the corre-
ration of the hyperbranched polyglycerol-supported Co- sponding loading of 1.6 mmol ligand g–1 which already re-
and Mn-salen complexes via unsymmetrical salen synthesis veals the advantage of using dendritic polymer as high-
is reported. Application of the corresponding polymeric loading support. The polymeric ligand was efficiently puri-
catalysts in asymmetric catalysis including hydrolytic kinetic fied by means of ultrafiltration using membrane technology
Scheme 1. Preparation of polyglycerol-supported Co- and Mn-salen complexes. a) 1-Co(OAc)2, MeOH, toluene, 2-glacial acetic acid,
toluene, b) 1-Mn(OAc)2·4H2O, O2, DMF, 2-LiCl.
2136
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 2135–2141

Polyglycerol-Supported Co- and Mn-salen Complexes
(see Experimental Section) as an alternative for preparative
GPC.[47]
Polymeric salen was loaded with Co by introducing
Co(OAc)2 under inert atmosphere to form red PG-salen-
CoII. Prior to use, the CoII complex is catalytically inactive
and thus has to be converted into the active CoIII complex
4 via one-electron oxidation with acetic acid. In accord with
literature reports,[11,41] the oxidation is accompanied by a
the colour change from red to dark brown. Elemental
analysis indicated a loading of 0.76 mmolg–1 for cobalt
which means 54% of salen centers are occupied by cobalt.
The final PG-salen-Co catalyst had excellent solubility
properties in most of the common organic solvent such as
THF, dichloromethane or toluene.
Polyglycerol-supported Mn-salen was prepared by re-
fluxing polymeric salen with Mn(OAc)2·4H2O in DMF,
while air was bubbled through the reaction mixture. The
resultant complex was treated with LiCl for ideal counter-
ion substitution by stirring at room temperature for ad-
ditional 12 h to afford the final catalyst 5 as brown powder.
Elemental analysis showed corresponding loading of
1 mmolg–1 for manganese indicating that 71% of salen cen-
ters are loaded with manganese.
Figure 1. Time course for the HKR of 1,2-epoxyhexane 7 using
0.05 mol-% catalyst and 0.55 equiv. water in THF.
Table 1. Hydrolytic kinetic resolution of rac-1,2-epoxyhexane (7) at
low catalyst loading.
Catalyst
Loading
[mol-%]
Diol
Conversion
%
% ee[a]
salen-Co-OAc (6)
PG-salen-Co-OAc (4)
0.05
0.05
98.0
98.8
28.5
32.0
[a] Determined by chiral GC (Chiraldex G-TA).
Hydrolytic Kinetic Resolution of Terminal Epoxides
Breinbauer and Jacobsen[14] reported on the application
of polyamidoamine (PAMAM) as dendrimeric support for
salen. Application of the dendrimeric bound catalysts (PA-
MAM) in the HKR of terminal epoxides showed enhanced
reactivity of corresponding catalysts compared to its low
molecular weight analogue. This positive dendrimeric effect
is probably due to a bimetallic catalysis mechanism which
is believed to be followed in HKR of epoxides using salen
type catalyst.[50,51] This led the authors to the assumption
that dendrimeric analogues might reinforce cooperative
catalytic activity between catalytic units.
Figure 2. Schematic representation of bimetallic cooperative
mechanism[14] with a) monomeric catalyst and b) dendritic polygly-
cerol.
These results prompted us to investigate whether any
dendrimeric effect is present in case of polyglycerol-sup-
ported salen catalyst in HKR of terminal epoxides. There-
fore, 1,2-epoxyhexane 7 was chosen as the model epoxide
for detailed studies. Kinetic studies for the HKR of 7 cata-
lyzed by monomeric 6 and dendritic 4 cobalt salen catalysts
were performed. As shown in Figure 1 and Table 1, a posi-
tive dendritic effect was also observed in case of dendritic
polyglycerol-supported catalyst since HKR of 1,2-epoxy-
hexane showed enhanced reactivity of PG bound catalyst 4,
As shown in Figure 3, the enantiomeric excess is increased
in the course of the reaction and the respective epoxides
could be recovered in high ee indicating the efficiency of
polyglycerol-supported salen cobalt catalyst for HKR of ra-
cemic terminal epoxides (Table 2).
Asymmetric Epoxidation of Olefins
As outlined in literature, different forms of salen ligand
compared to low molecular weight analogue, 6. This obser- bearing manganese are competent catalysts for asymmetric
vation also demonstrates on cooperative bimetallic cataly- epoxidation (AE) of olefins[7] firstly reported by Jacob-
sis, conjecturing that the dendritic backbone increases the sen[52] and Katsuki.[53] Therefore, we were keen to check
possibility of intramolecular cooperation between cobalt whether the polyglycerol-supported Mn-salen catalyst gives
active sites by bringing the groups adjacent to each other rise to any change in epoxidation reaction in solution com-
(Figure 2).
pared to its non-immobilized analogues as it was the case
For practical HKR, higher catalyst loadings (0.5 mol-%) for HKR. However, AE proceeds over a significantly dif-
were applied. HKR of different terminal epoxides was car- ferent mechanism than HKR. The proposed mechanism of
ried out using 0.55 equiv. of water in THF. Reactions were the Jacobsen–Katsuki epoxidation is based on Kochi in-
initiated at 0 °C and then warmed up to room temperature. vestigations[54] where the MnV-oxido complexes are pos-
Eur. J. Org. Chem. 2008, 2135–2141
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2137

M. Beigi, S. Roller, R. Haag, A. Liese
FULL PAPER
Figure 3. Time course for the enantiomeric excess of different terminal epoxides using PG-supported catalyst 4 (5 mmol epoxide, 0.55
equiv. water and 0.5 mol-% catalyst).
Table 2. Hydrolytic kinetic resolution of different terminal epoxides
using PG-supported salen catalyst 4.
Different kinds of conjugated olefins were chosen as sub-
strates (Table 3). Following published procedures,[57] the re-
actions were carried out by solving the polymeric catalyst
5, olefin and N-methylmorpholine N-oxide (NMO) in
dichloromethane followed by subsequent cooling of the
mixture to the desired temperature. The epoxidation pro-
ceeds by adding m-chloroperbenzoic acid (mCPBA) to the
cooled mixture in two equal portions over a period of two
minutes. Initial screening and kinetic studies for epoxid-
ation of 6-cyano-2,2-dimethylchromene already revealed
significant catalytic activity of polyglycerol-supported Mn-
salen since nearly quantitative conversion was reached using
4 mol-% catalyst in less than four minutes (Figure 4). De-
spite HKR, the presence of a racemic, non-catalytic back-
ground epoxidation was observed. Fortunately, this reaction
was much slower than the catalytic one, where after 1 h less
than 1% conversion is achieved.
Epoxide
Loading
[mol-%]
Epoxide
Conversion
% ee[a]
1,2-Epoxyhexane (7)
Epichlorohydrin (8)
Phenyl glycidyl ether (9)
0.5
0.5
0.5
92.0
90.5
50.0
52.5
51.5
85.3[b]
[a] Determined by chiral GC (Chiraldex G-TA). [b] Determined by
chiral HPLC (Chiral Cel OD).
tulated to be the catalytically active species. The MnIII-salen
complex is expected to be converted into a high-valent oxo-
manganese(V) species through an oxygen transfer from the
oxidant. Subsequently, catalytically active oxomanganese-
(V) species transfer the oxygen further to the olefin double
bond and is afterwards eventually converted into the MnIII
-
salen complex. Nevertheless, beside this redox pathway, di-
merization of oxidomanganese(V) with MnIII can also oc-
The results for the epoxidation of different olefins cata-
cur, which forms catalytically inactive µ-oxidomanga- lyzed by the homogeneously soluble polyglycerol-supported
nese(IV) species. As a consequence, the catalyst is deacti- Mn salen complex are shown in Table 3. All these reactions
vated after a few turnovers. For this reason, immobilization were very fast and completed in less than 10 min at 0 °C.
has been considered as a proper pathway to enhance the Decreasing the temperature to –78 °C improved the
catalyst stability, since it is proposed that anchoring the enantioselectivity, but longer reaction times were needed,
MnIII-salen moieties to a carrier leads to lower local con- probably as a consequence of deficient solubility and lower
centration of the catalyst and also restricted complex mo- activity of the polymeric catalyst at low temperatures. The
bility (site isolation).[34,55,56] As a result, the possibility of values in parentheses represent reported enantiomeric ex-
dimerization is expected to be suppressed.
cess using the free Jacobsen catalyst at –78 °C.[55] Obviously,
2138
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 2135–2141

Polyglycerol-Supported Co- and Mn-salen Complexes
Table 3. Asymmetric epoxidation of different olefins using PG-sup- the polyglycerol-supported catalyst 5 showed very high ac-
ported catalyst.
tivity, most often comparable to its low molecular weight
analogue (Table 3, entry 6,7), in epoxidation reaction.
Repetitive Batch Operation
One of the main advantages of catalyst immobilization is
offering the possibility of catalyst recycling that is impor-
tant in view of economy and applicability. Moreover, as al-
ready mentioned, a major limitation for low-molecular
weight species of salen manganese complexes is their in-
herent degradation tendency.[34,35] Therefore, we tried to
find out whether the dendritic catalysts can be recycled and
whether the polymeric catalyst preparation leads to en-
hanced catalyst stability.
Recycling of the dendritic catalyst in repetitive batch ex-
periments was realized by precipitation with hexane. Poly-
glycerol-supported salen-Mn-Cl, 5, used in the epoxidation
of 6-cyano-2,2-dimethylchromene, could be recycled via this
procedure up to three times (see Table 4). The maximum
conversion was decreased with each catalyst recycling step,
but nevertheless the second run showed nearly identical re-
sults to the first run in terms of conversions and 75% con-
version could be still achieved in the forth run after 10 min.
The ee values of each run showed a decrease after each
recycling step from 95% for the first run to 88% in the forth
run. Most probably metal leaching is the reason for loss of
catalyst efficiency in the current experiment since the dark
brown colour of the polymeric catalyst faded by each
recycling step as similar observation is documented in the
[a] Conversion calculated on the basis of GC (HP-5) data related
literature.[40] As a consequence the racemic side reaction be-
to an internal standard. [b] Determined by chiral GC (Cyclodex-
came more prominent.
B). [c] Data in parantheses represent enantiomeric excess reported
using the free Jacobsen catalyst at –78 °C. [d] Determined by meas-
uring optical rotation.
Table 4. Repetetive batch operation of catalyst 5 in epoxidation of
6-cyano-2,2-dimethylchromene 12 at 0 °C.
Cycle
Time [min]
% Conversion[a]
% ee[b]
1
2
3
4
Ͻ 5
Ͻ 5
Ͻ 5
Ͻ 5
98
96
90
75
95
92
91
88
[a] Conversion calculated on the basis of GC (HP-5) data related
to 2-naphthonitrile as internal standard. [b] Determined by chiral
GC (Cyclodex-B).
In order to assure the absence of catalyst cleavage from
the polymeric backbone, a control experiment was carried
out: after separation of the polymeric catalyst by filtration,
the remnant of the reaction mixture was reused by addition
of substrate without appending any fresh catalyst. Since no
further formation of epoxide could be detected, the hypoth-
esis of salen catalyst release from the polyglycerol backbone
can be excluded. The Total Turnover Number (TTN) is cal-
culated to be increased from 23 for a single batch to around
80 for four repetitive batches which already represent in-
creased catalyst stability as a result of recycling possibility.
Conclusions
Figure 4. Asymmetric epoxidation of 6-cyano-2,2-dimeth-
ylchromene 12 with 5 (100 m olefin, 50 m IS, 200 m mCPBA,
500 m NMO, 4 m catalyst 5, 3 mL DCM, 0 °C).
In this contribution, the suitability of hyperbranched po-
lyglycerol as high-loading support is evaluated. An unsym-
Eur. J. Org. Chem. 2008, 2135–2141
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2139

M. Beigi, S. Roller, R. Haag, A. Liese
FULL PAPER
the reactor and volume fractions were collected at the reactor out-
let. Solvent was flushed through the reactor until 35 residence times
were collected. Then the reactor outlet was connected to the second
inlet of the reactor and the retained polymer was collected for up
to 5 residence times. Evaporation of the solvent delivered (ca. 0.3 g)
of pure 3; yield 60%.
metrical salen ligand was prepared through a simple and
straightforward procedure and subsequently bound to the
preformed polyglycerol. The high-loading polyglycerol salen
analogue was successfully purified by means of ultrafiltra-
tion. The corresponding cobalt- and manganese-loaded cat-
alysts showed very high activity and selectivity in HKR of
racemic epoxides and olefin epoxidation. Interestingly, ki-
netic studies revealed the presence of a positive dendrimeric
effect for dendritic polyglycerol, since HKR of 1,2-epoxy-
hexane showed enhanced reactivity for the PG-bound cata-
lyst, compared to the low molecular weight analogue. This
observation supports a cooperative bimetallic mechanism.
The polymer-supported Mn-salen catalyst used in epox-
idation of 6-cyano-2,2-dimethylchromene could be recycled
by precipitation up to three times. The results of repetitive
batches operation confirmed the fact that immobilization
simplifies product recovery and led to increased catalyst sta-
bility (TTN). The advantage of enhanced stability is of ut-
most importance in the case of the salen manganese com-
plex since the degradation tendency is a major limitation
for its low-molecular weight analogues.
Complex 4: In a one-necked Schlenk flask, purple solution of
Co(OAc)2 (0.085 g, 0.48 mmol, 1.5 equiv.) in dry MeOH (5 mL)
was dropped into the yellow solution of 3 (0.2 g, 1.64 mmolg–1,
0.32 mmol) in dry and degassed toluene (15 mL) under Ar where
the solution turned dark red over orange. The mixture was stirred
for 24 h. Then the mixture was concentrated and the residue was
dissolved in p.a. DCM and was filtered in order to remove surplus
cobalt acetate. Concentration of the filtrate gave the cobalt(II)
complex as a dark red solid. IR (neat): ν = 2949, 2920, 2857, 1597,
˜
1530, 1438, 1383, 1318, 1256, 1203, 1165, 1085, 1042, 940, 897,
787, 730, 694, 641, 568 cm–1.
The CoII complex was dissolved in toluene (10 mL) and treated
with glacial acetic acid (37 µL, 0.6 mmol). The mixture was stirred
without exclusion of air for 2 h during which time the colour turned
to brown. Concentration of the solution under vacuum afforded
4 as a brown solid. According to AAS measurements, loading of
0.76 mmolg–1 for cobalt was found; conversion: 54%. 1H NMR
(CDCl3, 600 MHz): δ = 7.3–7.15 (br. s, 2 H, CH=N), 7.1–6.9 (br.
m), 3.0 (br., 2 H), 2.6–2 (br. m, PG backbone), 1.9–0.8 (br. m, 38
Experimental Section
General: Starting materials were purchased from Aldrich, Fluka
or Acros and used without further purifications unless mentioned
otherwise. The solvents used in the reactions were of p.a. quality
or purified and dried according to standard methods. 1H NMR
Spectra: Varian Unity Plus 600 MHz. Chemical shifts are reported
in ppm downfield from tetramethylsilane with the solvent reso-
nance as the internal standard (deuteriochloroform; 1H: 7.24 ppm).
IR spectra: Bruker IFS 28-Spectrometer or a Digilab FTS 4000.
Capillary gas chromatography (GC) for determination of enantio-
meric excess or conversion calculation (related to an internal stan-
dard): Hewlett–Packard (Agilent) GC, HP 6890, column: achiral
HP-5 (30 mϫ0.32 mmϫ0.25 µm), detector: FID, carrier gas: H2;
Hewlett–Packard (Agilent) GC, HP 5890, column: Supelco® ß-
DEX 120 (30 mϫ0.25ϫ0.25 µm) from Whisky and Cognaclacton
or Chiraldex G-TA (30 mϫ0.25 mm) from Advanced Separation
Technologies Inc., detector: FID, carrier gas: H2. High Perform-
ance Liquid Chromatography (HPLC): Hewlett–Packard (Agilent)
HPLC, HP 1100, binary pump, column: Chiralcel® OD
(250 mmϫ4.6 mm) from DAICEL Chemical Industries, LTD, de-
tector: UV. Atomic Absorption Spectroscopy (AAS): Video 22,
AA/AE spectrometer from Instrumentation Laboratory.
H) ppm. IR (neat): ν = 2959, 2864, 1608, 1528, 1436, 1385, 1320,
˜
1258, 1167, 1086, 1016, 867, 796, 693, 568 cm–1.
Complex 5: A solution of the polymer-bound ligand 3 (0.3 g, 1.6
4 mmolg–1, 0.48 mmol) in DMF (150 mL) was heated to 90 °C.
Upon addition of Mn(OAc)2·4H2O (0.58 g, 2.4 mmol, 5 equiv.) the
colour of the mixture changed from yellow to dark brown. Heating
was continued for another 3 h during which time air was bubbled
through the solution my means of a glass tube pipe. Afterwards,
LiCl (0.2 g, 4.8 mmol, 10 equiv.) was added and the mixture was
cooled down. After stirring the solution for 1 h at room tempera-
ture, the mixture was filtered and the solvent removed under re-
duced pressure. Then DCM (150 mL) was added and the brown
organic layer was washed with brine and H2O, respectively. The
organic layer was dried with Na2SO4, filtered and evaporated to
afforded 5 as a brown powder. According to AAS measurements
the metal loading was 1 mmolg–1 for Mn; conversion: 71%. IR
(KBr): ν = 2961 (m), 2870, 1655, 1616 (s, C=N), 1542, 1437, 1261
˜
(s), 1096 (m), 1026 (s, PG backbone), 864, 802 (m), 699, 568 cm–1.
General Procedure for Hydrolytic Kinetic Resolution of Racemic
Terminal Epoxides: The exact amount of the catalyst was dissolved
in THF (2 mL). After the addition of epoxide (5 mmol, 1 equiv.),
the mixture was cooled to 0 °C. After 1 h, deionized H2O
(3.5 mmol, 63 µL, 0.7 equiv.) was added and the mixture was left
to warm up to room temperature. Progress of reaction was moni-
tored via GC (for 7 and 8, Cyclodex γ-TA) or HPLC (for 9, Chi-
ralCel OD, hexane/iPrOH used as eluent).
Polyglycerol-Supported salen Ligand 3: Preparation from hyper-
branched polyglycerol (Mn = 8000 gmol–1, OH group loading:
13.5 mmolg–1) according to ref.[47] Purification of the crude prod-
uct via ultrafiltration (as an alternative for preparative GPC) de-
scribed below led to pure 3 with 1.6 mmolg–1 ligand loading as
determined by NMR spectroscopy.
General Procedure for Purification of Crude Polymeric Ligand 3 via
Ultrafiltration: A stainless steel ultrafiltration cell, with a reaction
volume of 10 mL was applied. Toluene was used as the solvent and
was pumped through the reactor using a piston pump (Pharmacia
P 500). The membranes [Starmem 120 (MWCO = 200 gmol–1)]
from Membrane Extraction Technology[58] were preconditioned in
the solvent overnight before being installed in the reactor. Prior to
addition of crude polymeric ligand 3 the membrane was thoroughly
rinsed with solvent (flow rate: 50 mL h–1) until a constant pressure
of 1 MPa was reached. Then polymer (ca. 0.5 g) was charged to
General Procedure for Kinetic Experiments: The proper catalyst
(0.012 mmol based on cobalt), rac-1,2-epoxyhexane 7 (3 mL,
25 mmol) and 1-(2-methylbutyl)benzene (200 µL; as internal stan-
dard) were dissolved in 4 mL of THF. After the reaction mixture
was stirred for 1 h at 0–4 °C, H2O (250 µL, 14 mmol, 0.55 equiv.)
was added. 10 µL samples were withdrawn at certain times. In or-
der to remove H2O and the polymeric catalyst, samples were passed
through a thin layer of silica using Et2O as eluent. Then the sam-
ples were analyzed by GC. [HP-5, inj. temp. 300 °C, det. temp.:
300 °C, flow rate: 25 mLmin–1, 50 °C/0 min/10 °C min–1/300 °C, tR
2140
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 2135–2141

Polyglycerol-Supported Co- and Mn-salen Complexes
[21] Miyafuji, T. Katsuki, Synlett 1997, 836–838.
[22] N. Komiya, S. Noji, S. I. Murahashi, Tetrahedron Lett. 1998,
39, 7921–7924.
(epoxide) = 3.2 min, tR (internal standard) = 5.8 min, tR (diol) =
6.1 min].
General Procedure for Epoxidation of Unfunctionalized Olefins: To
a solution of salen-Mn-Cl type catalyst (0.0123 mmol, 4 mol-%) in
DCM (3 mL), olefin (0.31 mmol), internal standard (0.16 mmol,
0.5 equiv.) and NMO (0.18 g, 1.54 mmol, 5 equiv.) were added.
When the reaction mixture was cooled down to the desired tem-
perature, mCPBA (0.15 g, 0.6 mmol, 2 equiv.) was added. After the
reaction was complete 1  KOH and DCM were added. Drying the
organic layers with MgSO4 and removal of the solvent yielded
crude epoxide.
[23] Y. Kohmura, T. Katsuki, Tetrahedron Lett. 2001, 42, 3339–
3342.
[24] M. S. Sigman, E. N. Jacobsen, J. Am. Chem. Soc. 1998, 120,
5315–5316.
[25] P. Vachal, E. N. Jacobsen, Org. Lett. 2000, 2, 867–870.
[26] Y. N. Belokon, M. North, T. Parsons, Org. Lett. 2000, 2, 1617–
1619.
[27] R. Irie, K. Masutani, T. Katsuki, Synlett 2000, 1433–1436.
[28] D. R. Reddy, E. R. Thornton, J. Chem. Soc., Chem. Commun.
1992, 172–173.
General Procedure for Repetitive Batch Experiments: To a solution
of PG-supported Mn-salen (0.022 g, 0.55 mmol Mn g–1,
0.0123 mmol metal, 0.04 mol-%), 2,2-dimethyl-2H-1-benzopyran-
6-carbonitrile (0.057 g, 0.31 mmol) and 2-naphthonitrile as internal
standard (0.022 g, 0.15 mmol) in DCM, NMO (0.18 g, 1.54 mmol,
5 equiv.) was added. After cooling the reaction mixture to the de-
sired temperature, mCPBA (0.15 g, 0.6 mmol, 2 equiv.) was added.
After 5 min, the reaction was completed and H2O as well as DCM
were added. After drying the organic layers with MgSO4, the sol-
vent was removed and the catalyst was precipitated from hexane.
The recovered catalyst was reused in the next batch under the same
condition as previously for the first step.
[29] S. Laue, L. Greiner, J. Wöltinger, A. Liese, Adv. Synth. Catal.
2001, 343, 711–772.
[30] L. Greiner, S. Laue, A. Liese, C. Wandrey, Chem. Eur. J. 2006,
12, 1818–1823.
[31] C. Bied-Charreton, M. Frostin-Rio, D. Pujol, A. Gaudemer, J.
Mol. Catal. 1982, 16, 335–348.
[32] R. Irie, K. Noda, Y. Ito, T. Katsuki, Tetrahedron Lett. 1991,
32, 1055–1058.
[33] E. N. Jacobsen, in Comprehensive Organometallic Chemistry
(Ed.: E. W. Abel, F. G. A. Stone, G. Wilkinson), Pergamon,
New York, 1995, pp. 1097–1135.
[34] D. C. Sherrington, Catal. Today. 2000, 57, 87–104.
[35] L. Canali, E. Cowan, H. Deleuze, C. L. Gibson, D. C. Sher-
rington, Chem. Commun. 1998, 2561–2562.
[36] S. B. Ogunwumi, T. Bein, Chem. Commun. 1997, 901–902.
[37] M. J. Sabater, A. Corma, A. Domenech, V. Forn, H. Garcia,
Chem. Commun. 1997, 1285–1286.
Acknowledgments
M. B. acknowledges NRW Graduate School of chemisry for a doc-
toral scholarship and additional financial support. Deutsche For-
schungsgemeinschaft (DFG) is acknowledged for a joint research
grant (CERC-3 Project).
[38] N. E. Leadbeater, M. Marco, Chem. Rev. 2002, 102, 3217–3274.
[39]
S. Peukert, E. N. Jacobsen, Org. Lett. 1999, 1, 1245–1248.
[40] T. S. Reger, K. D. Janda, J. Am. Chem. Soc. 2000, 122, 6929–
6934.
[41] X. Zheng, C. W. Jones, M. Weck, Chem. Eur. J. 2006, 12, 576–
583.
[42] M. Holbach, M. Weck, J. Org. Chem. 2006, 71, 1825–1836.
[43] R. V. Heerbeek, P. C. J. Kamer, P. W. N. M. Leeuwen, J. N. H.
Reek, Chem. Rev. 2002, 102, 3717–3756.
[44] a) S. R. Sandler, F. R. Berg, J. Polym. Sci. Polym. Chem. Ed.
1966, 4, 1253–1259; b) T. Tsuruta, S. Inoue, H. Koinuma,
Macromol. Chem. 1968, 112, 58–65.
[45] E. J. Vandenberg, J. Polym. Sci. A 1985, 23, 915–949.
[46] a) A. Sunder, R. Mülhaupt, R. Haag, H. Frey, Adv. Mater.
2000, 12, 235–239; b) A. Sunder, R. Hesselmann, H. Frey, R.
Mühlhaupt, Macromolecules 1999, 32, 4240–4246.
[47] C. Hajji, S. Roller, M. Beigi, A. Liese, R. Haag, Adv. Synth.
Catal. 2006, 348, 1760.
[48] M. McGarrigle, C. T. Dalton, A. M. Daly, D. G. Gilheany, J.
Mol. Catal. A Chem. 2005, 231, 205–220.
[49] S. Roller, Ph. D. Thesis, University of Dortmund, 2005.
[50] E. N. Jacobsen, Acc. Chem. Res. 2000, 33, 421–431.
[51] L. B. Hansen, J. L. Leighton, E. N. Jacobsen, J. Am. Chem.
Soc. 1996, 118, 10924–10925.
[1] H. Schiff, Ann. Suppl. 1864, 3, 343.
[2] L. Canali, D. C. Sherington, Chem. Soc. Rev. 1999, 28, 85–93.
[3] E. N. Jacobsen, W. Zhang, A. R. Muci, J. R. Ecker, L. Deng,
J. Am. Chem. Soc. 1991, 113, 7063–7064.
[4] J. F. Larrow, E. N. Jacobsen, Top. Organomet. Chem. 2004, 6,
123–152.
[5] E. N. Jacobsen, in Catalytic Asymmetric Synthesis (Ed.: I.
Ojima), VCH Publishers, New York, 1993, pp. 159–202.
[6] T. Katsuki, Coord. Chem. Rev. 1995, 140, 189–214.
[7] E. M. McGarrigle, D. G. Gilheany, Chem. Rev. 2005, 105,
1563–1602.
[8] T. Katsuki, Adv. Synth. Catal. 2002, 344, 131–147.
[9] J. F. Larrow, S. E. Schaus, E. N. Jacobsen, J. Am. Chem. Soc.
1996, 118, 7420–7421.
[10] J. L. Leigton, E. N. Jacobsen, J. Org. Chem. 1996, 61, 389–390.
[11] M. Tokunaga, J. F. Larrow, F. Kakiuchi, E. N. Jacobsen, Sci-
ence 1997, 277, 936–938.
[12] M. H. Wu, K. B. Hansen, E. N. Jacobsen, Angew. Chem. 1999,
111, 2167–2170; Angew. Chem. Int. Ed. 1999, 38, 2012–2014.
[13] J. M. Ready, E. N. Jacobsen, J. Am. Chem. Soc. 1999, 121,
6086–6087.
[52] W. Zhang, J. L. Loebach, S. R. Wilson, E. N. Jacobsen, J. Am.
Chem. Soc. 1990, 112, 2801–2803.
[14] R. Breinbauer, E. N. Jacobsen, Angew. Chem. 2000, 112, 3750–
3753; Angew. Chem. Int. Ed. 2000, 39, 3604–3607.
[15] D. A. Annis, E. N. Jacobsen, J. Am. Chem. Soc. 1999, 121,
4147–4154.
[53] R. Irie, K. Noda, Y. Ito, N. Matsumoto, T. Katsuki, Tetrahe-
dron Lett. 1990, 31, 7345–7348.
[54] K. Srinivasan, P. Michaud, J. K. Kochi, J. Am. Chem. Soc.
1986, 108, 2309–2320.
[55] B. B. De, B. B. Lohray, S. Sivaram, P. K. Dhal, J. Polym. Sci.,
Part A Polym. Chem. 1997, 35, 1809–1818.
[16] S. E. Schaus, J. Brånalt, E. N. Jacobsen, J. Org. Chem. 1998,
63, 403–405.
[17] Y. Huang, T. Iwama, V. H. Rawal, J. Am. Chem. Soc. 2000,
122, 7843–7844.
[18] J. Mihara, T. Hamada, T. Takeda, R. Irie, T. Katsuki, Synlett
1999, 1160–1162.
[56] L. Canali, E. Cowan, H. Deleuze, C. L. Gibson, D. C. Sher-
rington, J. Chem. Soc. Perkin Trans. 1 2000, 2055–2066.
[57] M. Palucki, G. J. McCormick, E. N. Jacobsen, Tetrahedron
Lett. 1995, 36, 5457–5460.
[19] Heckel, D. Seebach, Helv. Chim. Acta 2002, 85, 913–926.
[20] J. F. Larrow, E. N. Jacobsen, J. Am. Chem. Soc. 1994, 116,
12129–12130.
[58] www.membrane-extraction-technology.com
Received: December 29, 2007
Published Online: March 3, 2008
Eur. J. Org. Chem. 2008, 2135–2141
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2141