Macrocyclic cyclooctene-supported AlCl-salen catalysts for conjugated addition reactions: Effect of linker and support structure on catalysis

Article abstract of DOI:10.1002/chem.200801611

AlCl-salen (salen = N.N′ bis(salicylidene)ethylenediamine dianion) catalysts supported onto macrocyclic oligomeric cyclooctene through linkers of varying length and flexibility have been developed to demonstrate the importance of support architecture on c

Full text of DOI:10.1002/chem.200801611

DOI: 10.1002/chem.200801611
Macrocyclic Cyclooctene-Supported AlCl–Salen Catalysts for Conjugated
Addition Reactions: Effect of Linker and Support Structure on Catalysis
Nandita Madhavan,[b] Tait Takatani,[b] C. David Sherrill,[b] and Marcus Weck*[a, b]
Abstract: AlCl–salen (salen=N,N’-
bis(salicylidene)ethylenediamine di-
anion) catalysts supported onto macro-
cyclic oligomeric cyclooctene through
enhanced the cyanide addition reac-
tion, most likely by improving salen–
salen interactions in the transition
state, it lowered the reaction rate for
the monometallic indole reaction. For
both reactions, significant increase in
catalytic activity was observed for cata-
lysts with the longest linkers. The effect
of the flexible macrocyclic support on
catalysis was further exemplified by the
enhanced activity of the supported cat-
alyst in comparison with its unsupport-
ed analogue for the conjugate addition
of tetrazoles, which is known to be cat-
alyzed by dimeric m-oxo–salen cata-
lysts. Our studies with the cyclooctene
supported AlCl–salen catalysts pro-
vides significant insights for rationally
designing highly efficient AlCl–salen
catalysts for a diverse set of reactions.
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
linkers of varying length and flexibility
have been developed to demonstrate
the importance of support architecture
on catalyst activity. The role played by
the support and the linkers in dictating
catalyst activity was found to vary for
reactions with contrasting mechanisms,
such as the bimetallic cyanide and the
monometallic indole addition reactions.
While the flexible support significantly
Keywords: aluminum · asymmetric
catalysis
N,O ligands · ROMP (ring-opening
metathesis polymerization)
supported catalysts
· conjugated addition ·
·
Introduction
recovery and reuse of catalysts, whereas the low-molecular-
weight supports are often designed to enhance the catalytic
activity of the supported catalysts in comparison with their
non-supported analogues. While there has been significant
advancement in the development of recoverable cata-
lysts,[1–3] there have been limited efforts correlating the
effect of support architecture to the activity of the
Organometallic salen (salen=N,N’-bis(salicylidene)ethyl-
ACHTUNGTRENNUNGenediamine dianion) catalysts have emerged as powerful
tools in facilitating the synthesis of important materials and
therapeutic agents. Environmental concerns associated with
the disposal of such catalysts, coupled with financial con-
cerns due to their increasing costs, have led to considerable
research focused towards the development of green and eco-
nomically viable alternatives such as supported metal–salen
catalysts.[1–5] Several supports ranging from high-molecular-
weight polymers and inorganic supports to dendrimers and
oligomers have been explored for salen complexes.[1,6–11] The
high-molecular-weight supports typically facilitate the easy
AHCTUNGTRENNUNG
a support material, such as a polymer or a surface, a cata-
lyst, and a linker connecting the two. The nature of the
linker and the support material can have a profound effect
on the activity of the catalyst. Extremely high catalytic activ-
ities have been observed for catalysts with dimeric, oligo-
meric, and dendritic supports for the bimetallic Co–salen-
catalyzed reactions, due to an increased local concentration
of catalysts on the support.[6,10,11,13] In contrast, for the mon-
ometallic Mn–salen-catalyzed epoxidation reaction, supports
with lower local concentration of catalysts have been found
to enhance the reactivity.[7,14–16] Furthermore, longer and
flexible linkers have been demonstrated to improve both
the bimetallic Co–salen hydrolytic kinetic resolution (HKR)
reaction as well as the monometallic Mn–salen epoxidation
reaction.[17–19] Another class of metal–salen catalysts, Al–
salen complexes, are known to catalyze industrially impor-
[a] Prof. Dr. M. Weck
Molecular Design Institute and Department of Chemistry
New York University, New York, NY 10003 (USA)
Fax : (+1)212-995-4895
E-mail: marcus.weck@nyu.edu
[b] Dr. N. Madhavan, T. Takatani, Prof. Dr. C. D. Sherrill,
Prof. Dr. M. Weck
School of Chemistry and Biochemistry
Georgia Institute of Technology, Atlanta, GA 30332 (USA)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200801611.
1186
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1186 – 1194

FULL PAPER
tant conjugate addition reactions, some of which are hy-
pothesized to follow a bimetallic pathway,[20,21] while others
presumably follow a monometallic pathway.[22] Hence, Al–
salen complexes are a perfect system to study the effect of
support structure and linker length on the catalysis of
mono- and bimetallic reactions and to verify the proposed
guidelines in the literature of supported catalyst design.[12]
Despite this fact, there are only two studies reported in the
literature pertaining to the design of supported Al–salen cat-
alysts[23,24] and only one of them is pertinent to conjugate ad-
dition reactions.[23] Herein, we introduce macrocyclic cyclo-
octene-supported, AlCl–salen catalysts to study the impor-
tance of catalyst structure on the conjugate addition of nu-
cleophiles to a,b-unsaturated imides and ketones (Figure 1).
side-chain, catalyst 2 having a medium-sized linker with
moderate flexibility, while catalyst 3 is based on the longest
and most flexible linker.
Results and Discussion
The oligomeric catalysts 1–3 can be synthesized readily in
high yields from commercially available starting materials.
Catalysts 1 and 2 were synthesized from the reaction of
salen ligand 4 with cyclooctene carboxylic acids 5 or 6 via
the intermediates 7–10 (Scheme 1). Catalyst 3 was synthe-
sized from ligand 11[23] (Scheme 2). The salen ligands 4 and
11 were readily obtained from cyclohexane monohydro-
chloride salt by the one-pot synthesis of unsymmetrical
salens developed by our group.[27] The ligands were coupled
with cyclooctene carboxylic acids in the presence of dicyclo-
hexylcarbodiimide (DCC) and 4-(dimethylamino)pyridine
(DMAP) to afford the corresponding unmetalated mono-
mers (7, 8, and 12) in high yields. Solutions of monomers 7,
8, and 12 at concentrations of 0.1m in dichloromethane were
subjected to ROMP using Grubbsꢁ third-generation initiator
to afford the cyclic oligomers 9, 10, and 13, respectively. The
formation of the macrocycles was confirmed by 1H NMR
spectroscopy, gel-permeation chromatography, and mass
1
spectrometry. H NMR spectroscopy indicated the absence
of end-group signals, while GPC data showed the formation
of lower molecular weight oligomeric species. MALDI-TOF
mass spectrometry provided direct evidence for the forma-
tion of the cyclic structures as signals corresponding to mac-
rocycles of different sizes with values of “n” ranging from
two to ten were observed. These results are in close analogy
to our recent reported cyclooctene-based Co–salen macrocy-
cles.[10] The cyclic oligomers were then subjected to metala-
tion by addition of a solution of diethyl aluminum chloride
to afford catalysts 1–3 in 90–99% yields. Complete metala-
tion was confirmed by the disappearance of the phenolic
Figure 1. Supported (salen)AlCl catalysts attached to oligomeric cyclo-
ACHTUNGTRENNUNGoctene macrocycle through varied linkers.
Our investigations are the first to compare and contrast the
effect of support architecture on mono- and bimetallic
AlCl–salen-catalyzed reactions. We demonstrate that a flexi-
ble support significantly enhances the bimetallic reaction
and lowers the reaction rate for the monometallic reaction,
while longer linkers increase the catalytic activity for both
reactions.
1
protons in the H NMR spectrum of the catalysts and the
Our catalyst design utilizes the oligomeric cyclooctene
macrocycle as the backbone support, which is readily ob-
tained by the ring-opening metathesis polymerization
(ROMP) of cyclooctene-derived monomers under dilute
conditions.[25,26] The dramatic enhancements shown by the
macrocyclic cyclooctene backbone with Co–salen[10] makes it
an ideal candidate to study the catalysis of the bimetallic
AlCl–salen-catalyzed cyanide addition to a,b-unsaturated
imides. The flexible cyclooctene support is also a good ex-
ample to investigate whether the increased local concentra-
tion of AlCl–salen complexes on the support has a detri-
mental effect on the catalysis of a reaction that involves a
monometallic transition state including the addition of
indole to a,b-unsaturated ketones. To investigate the effect
of linker length and flexibility on the catalytic activity, the
AlCl–salen complexes were attached to the macrocyclic
backbone by using three different linkers as illustrated in
Figure 1. The linkers vary in length and flexibility with cata-
lyst 1 having the shortest linker resulting in the least flexible
aluminum ICP elemental analyses of the catalysts.[28]
The activity of the catalysts was first assessed for the
asymmetric addition of cyanide to a,b-unsaturated imide 14,
which has been reported to follow second-order kinetics
with respect to the catalyst (Table 1).[20,21] The cyanide was
generated in situ from isopropanol and trimethylsilyl cya-
nide in the presence of the imide 14 and the catalyst in tolu-
ene at 458C. The reactions were sealed to minimize cyanide
leakage, which prevented the evaluation of the reaction
progress by thin-layer chromatography. Therefore, the reac-
tion with each catalyst was carried out for a time period of
18 h, following which the yield of the isolated product 15
was determined to compare the efficiency of the catalyst.[29]
Table 1 illustrates the yields and enantioselectivities for 15
using the catalysts 1–3.
The nature of the linker played an important role in dic-
tating the yields and enantioselectivities of 15. Lower yields
(70%) were obtained with 1, which had the shortest and
most rigid linker, with respect to 2 and 3 with the medium
Chem. Eur. J. 2009, 15, 1186 – 1194
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1187

M. Weck et al.
Scheme 1. Synthetic pathway to obtain catalysts 1 and 2.
Scheme 2. Synthetic route for catalyst 3.
and long linkers, respectively. The improved yields of 15
with an increase in the length and flexibility of the linker
suggests that a more flexible linker makes the interaction
between the two catalytic AlCl–salen centers that form the
bimetallic transition state more facile. The enantioselectivi-
ties of the products also depended on the structure of the
catalyst. Catalyst 3 in which the linker was attached to the
salen through a tertiary carbon atom gave the best enantio-
meric excesses (eeꢁs) in comparison to 2 and 1. Surprisingly,
the ee of 15 obtained with 2 was slightly lower than that ob-
tained with 1. In all cases, the yields obtained for adduct 15
using the cyclooctene catalysts were superior to the yields
obtained with the Jacobsen AlCl–salen catalyst (<20%)
under the same reaction conditions.[23] The higher yields sug-
1188
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1186 – 1194

Asymmetric Catalysis
FULL PAPER
Table 1. Effect of linker on yield and selectivity of the bimetallic cyanide
addition.
ed in Table 2 indicate that the atomic charges on the atoms
do not vary with the nature of the linker.
Table 2. B3LYP/6-31G* atomic charges on the atoms in the catalytic
center.
Linker
Al
N
N
O
O
Cl
Catalyst
Yield[a] [%]
ee [%][b]
short
medium
long
0.53
0.52
0.53
ꢀ0.30
ꢀ0.30
ꢀ0.29
ꢀ0.21
ꢀ0.21
ꢀ0.23
ꢀ0.25
ꢀ0.24
ꢀ0.26
ꢀ0.33
ꢀ0.33
ꢀ0.33
ꢀ0.39
ꢀ0.39
ꢀ0.39
1
2
3
1 (short)
2 (medium)
3 (long)
70
94
98
93
82
98
[a] Isolated yield. [b] Enantiomeric excess determined by HPLC analysis
After establishing the positive effect of the macrocyclic
support on the bimetallic cyanide addition reaction, we
wished to investigate if the increased local concentration of
the catalyst on the support would lead to lower activities for
a monometallic reaction. Therefore, the activity of 1 and
3[31] were studied for the addition of indole 17 to the a,b-un-
saturated ketone 16,[32] which was reported to follow a mon-
ometallic pathway in the literature.[22,33] The reactions were
carried out in toluene for 18 h and the yields for the product
were determined after purification by using flash column
chromatography.
using a chiral Pirkle-l-leucine column.
gest that the macrocyclic support does indeed play a signifi-
cant role in enhancing the reaction. The most notable fact
was that a 50% lower catalyst loading (5 mol%) than that
reported for the unsupported AlCl–salen catalyst was used,
indicating the importance of the nature of the support on
the catalysis.[20] The initial rates of the most active cyclooc-
tene catalyst 3 and the unsupported catalyst were compared
1
by using H NMR spectroscopy. The kinetics were studied
by setting up the cyanide addition reaction in five NMR
tubes and quenching each reaction at different time periods.
Conversions were determined by comparing the areas ob-
tained upon the integration of the product peaks with that
for 1,1,2,2-tetrachloroethane, which was used as an internal
standard.
As illustrated in Table 3, lower yields and eeꢁs were ob-
served for the product obtained when catalyst 1 was used.
The lower yields and eeꢁs indicated that the increased
Table 3. Effect of linker on yield and enantioselectivity of monometallic
indole addition reaction.
For 3, the reaction was 80% complete within 8 h, while
the Jacobsen catalyst showed less than 20% conversion
after 6 h. The kinetics data indicate that the activity of 3 is
significantly higher than that for the unsupported Jacobsen
(AlCl) catalyst, most likely due to the bimetallic nature of
the catalysis and the enhanced proximity of two metal sites
due to the support structure as well as the flexibility of the
support and the linker (Figure 2). To ensure that the differ-
ence in catalytic activities between 1–3 were not due to
changes in the electronic properties of the catalytic center,
B3LYP/6–31G* computational studies were carried out to
determine the atomic charges on the ligand and the alumi-
num in the presence of the linkers.[30] The results as illustrat-
Catalyst
Yield[a] [%]
ee [%][b]
1
2
3
Jacobsen Catalyst
1 (short)
3 (long)
81
65
92
37
26
24
[a] Isolated yield. [b] Enantiomeric excess determined by HPLC analysis
using a chiralcel OD column.
crowding of catalysts on the cyclooctene support did indeed
lower the catalytic activity. A similar effect for the selectivi-
ties and conversions has been observed with Mn–salen sys-
tems for the monometallic asymmetric epoxidation reac-
tion.[7] The lower selectivities might be attributed to the dif-
ficulty of the reagents in accessing the catalytic site due to
excessive crowding. While the selectivities remained low for
adduct 18 obtained with 3, the yields were significantly im-
proved and even higher than that obtained with the unsup-
ported catalyst. We believe this intriguing increase in yield
is a result of the greater solubility of 3 in comparison to 1
and the unsupported catalyst in the reaction medium due to
the presence of the long hydrophobic alkyl chains. To prove
this hypothesis, the reactions were carried out in a reaction
medium, such as methylene chloride, in which all the cata-
Figure 2. Effect of support on catalytic activity: comparison of initial
&
^
rates of 3 ( ) with the non-supported Jacobsen catalyst ( ).
Chem. Eur. J. 2009, 15, 1186 – 1194
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1189

M. Weck et al.
lysts are highly soluble; comparable rates were obtained for
all catalysts. Based on our observations with the indole addi-
tion reaction, we hypothesize that the detrimental effects of
the support are not very pronounced and could be offset by
an increased solubility of the catalyst as observed with 3.
The rates for the catalysis of the indole addition reaction
by 1, 3, and the unsupported catalyst were determined by
lytic activity of 3 and its unsupported AlCl–salen analogue
(Table 4). The yields for the formation of the adduct 20
were found to be higher when 3 was employed, with the dif-
Table 4. Conjugate addition of 5-phenyltetrazole to a,b-unsaturated
AHCTUNGTREGkNNNU etones and imides.
1
using H NMR spectroscopy and their activities were com-
pared (Figure 3). The kinetics were studied by setting up the
R
Catalyst
Loading [mol%] Yield [%][a]
1
2
3
4
5
16: Ph
16: Ph
Jacobsen catalyst
3 (long)
5
5
5
5
20a: 73
20a: 91
20b: 74
20b: 98
20b: 85
14: -NHCOPh Jacobsen catalyst
14: -NHCOPh 3 (long)
14: -NHCOPh 3 (long)
2.5
[a] Isolated yield.
ference being more pronounced for the reaction with imide
14 than with ketone 16. It was notable that high yields were
obtained with only 5 mol% of the catalyst, which again is a
50% lower loading than that of the dimeric m-oxo–salen cat-
alyst. In the case of imide 14, the loading could be lowered
even further by 50% (2.5 mol%) without significantly com-
promising the yield of the adduct 20b.
Figure 3. Comparison of initial rates for the indole addition reaction. Ja-
!
&
^
cobsen catalyst: ; catalyst 1: , catalyst 3:
.
reactions in [D8]toluene in NMR tubes in the presence of an
internal standard, such as 1,1,2,2-tetrachloroethane. NMR
spectra were obtained at different time intervals in order to
determine the conversions. The initial rate for the reaction
catalyzed by 3 is significantly higher than that with 1 and
the Jacobsen catalyst as evidenced by greater than 50% con-
versions within 2.5 h. The pronounced difference in the ac-
tivities of 1 and 3 illustrates that for this monometallic reac-
tion, the hydrophobicity of the linker plays a more impor-
tant role than its flexibility in improving catalysis.
The studies with the indole and cyanide addition reaction
indicated that 3 was the most active catalyst among the oli-
gomeric-cyclooctene-supported Al–salen systems. After in-
vestigating the support and linker effects, we wished to
study if the close proximity of the AlCl–salen on our cyclo-
octene backbone in 3 could be exploited for reactions such
as the addition of tetrazole 19 to a,b-unsaturated ketones
and imides that have been catalyzed by the dimeric m-oxo–
Al–salen catalyst.[34] Since, the mechanism of the tetrazole
addition has not been reported, we were uncertain if the
steric and electronic differences between AlCl–salen and
the dimeric oxo–salen catalysts would affect the yields and
selectivities of the product. The reaction with the dimeric
salen catalyst has been reported at room temperature with
5 mol% of catalyst (10 mol% with respect to Al). With the
Jacobsen AlCl–salen catalyst as well as 3, we observed that
elevated temperatures of 558C were necessary to facilitate
the reaction and little or no selectivities were observed, indi-
cating that an AlCl–salen catalyst might not be the ideal cat-
alyst for this transformation. Despite the low eeꢁs, we were
keen on studying the effect of the macrocyclic support on
the catalysis and set out to compare the difference in cata-
To understand the magnitude of difference in the activi-
ties of 3 and the unsupported catalyst, we monitored the re-
1
action kinetics by using H NMR spectroscopy in the pres-
ence of 1,1,2,2-tetrachloroethane as an internal standard
(Figure 4). In all cases the reaction progressed significantly
faster with 3 when compared to Jacobsen catalyst. For the
addition to imide 14, the differences in rates were more pro-
nounced; the conversion to the product 20b was observed in
less than 10 h with 5 mol% of 3. The activity with 2.5 mol%
catalyst, although lower than that with 5 mol% 3, was still
higher than that for 5 mol% of the unsupported Jacobsen
catalyst. The enhancement of catalytic activity by virtue of
the flexibility of the macrocyclic backbone exemplifies the
crucial role of flexible supports on the reaction catalysis.
Studies are currently ongoing to incorporate the lessons
learnt from the macrocyclic backbone into the development
other supported catalysts that show not only enhanced reac-
tivities, but also good selectivities for the addition of nucleo-
philes such as tetrazoles.
Conclusion
We have used AlCl–salen catalysts attached to oligomeric
macrocyclic cyclooctene supports by varied linkers as model
systems to study the effect of linker and support on bi- and
monometallic conjugate addition reactions. The flexibility
and ability of the cyclooctene support to enhance interaction
between neighboring AlCl–salen units makes the catalysts
superior to their unsupported analogues for the bimetallic
1190
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1186 – 1194

Asymmetric Catalysis
FULL PAPER
CAUTION: Trimethylsilyl cyanide and hydrogen cyanide are highly toxic
and should be handled extremely carefully in a fume hood as per the ex-
perimental protocol mentioned below.
Analytical thin-layer chromatography (TLC) was performed on Silica
XHL pre-coated (250 mm thickness) glass-backed TLC plates from Sorb-
ent Technologies. Eluting solvents are reported as volume ratios or
volume percents. Compounds were visualized by using UV light or potas-
sium permanganate stains. Flash column chromatography was performed
with silica gel 60 ꢂ (230–400 mesh). All 1H and 13C NMR spectra were
recorded on Varian Mercury Vx 300 or Varian Mercury Vx 400 spec-
trometers with CDCl3 as solvent. Chemical shifts are expressed in parts
per million (d), coupling constants (J) are reported in Hertz (Hz), and
splitting patterns are reported as singlet (s), doublet (d), triplet (t), quar-
tet (q), unresolved multiplet (m), and apparent (app). All NMR spectra
are referenced using residual solvent peaks as the standard with d values
of 7.26 ppm for CDCl3. High-resolution mass spectra were obtained from
the Georgia Institute of Technology mass spectrometry lab. Gel-permea-
tion chromatography (GPC) analyses were performed on a Waters GPC
system with a Waters 1515 binary pump coupled with a Waters 2414 re-
fractive index detector, and referenced to poly
ACHTUNGTREN(NUNG styrene) standards.
G
1.0 mLminꢀ1. Chiral HPLC analyses were performed on a Shimadzu-
10 A system, using Pirkle-l-Leucine column from Regis Technologies or
a Chiralcel OD column from Daicel Chemical Industries.
Salen ligand 4: (R,R)-1,2-Diaminocyclohexane monohydrochloride[27]
(642 mg, 4.27 mmol, 1 equiv), methanol (22 mL), and some 4 ꢂ molecular
sieves were added to a flame dried 100 mL Schlenk flask equipped with a
stir bar under an atmosphere of argon. Subsequently, 3,5-di-tert-butyl-2-
hydroxybenzaldehyde (1.00 g, 4.27 mmol, 1 equiv) was added. The solu-
tion was allowed to stir for 3 h followed by the addition of 3-tert-butyl-
2,5-dihydroxybenzaldehyde[10] (829 mg, 4.27 mmol, 1 equiv) as a solution
in CH2Cl2 (22 mL) and NEt3 (1.2 mL, 8.5 mmol, 2 equiv). The reaction
mixture was stirred for an additional 3 h and then concentrated in vacuo
to afford a brown residue. CH2Cl2 (200 mL) and water (200 mL) were
added to the residue. The organic layer was separated and the aqueous
layer was extracted with CH2Cl2 (2ꢃ100 mL). The combined organic
layers were washed with water and saturated NaCl solution, dried over
anhydrous magnesium sulfate, filtered, and concentrated in vacuo to
afford a brown solid. The crude product was subjected to flash column
chromatography (gradient elution: 10:1!5:1 hexane/EtOAc) to afford
1.6 g (74%) of the salen 4 as a bright yellow solid. TLC Rf =0.35 (5:1
hexane/EtOAc); 1H NMR (300 MHz, CDCl3): d=13.4 (brs, 2H; OH),
8.28 (s, 1H; N=CH), 8.18 (s, 1H; N=CH), 7.30 (d, J=2.5 Hz, 1H; HAr),
6.96 (d, J=2.5 Hz, 1H; HAr), 6.80 (d, J=3.0 Hz, 1H; HAr), 6.46 (d, J=
3.0 Hz,1H; HAr), 3.29 (m, 2H; 2NCHCH2), 1.98–1.86 (m, 4H; 2 CH2),
Figure 4. Comparison of initial rates for the addition of tetrazole 17 to
a,b-unsaturated ketones (top) and imides (bottom) between catalyst 3
&
and unsupported Jacobsen AlCl–salen catalyst. Top: Catalyst 3: ; Jacob-
~
&
^
sen: ; Bottom: Catalyst 3 (5mol%) ; catalyst 3 (2.5mol%) ; Jacobsen
^
.
catalyst (5 mol%)
cyanide addition to a,b-unsaturated imides. In contrast, the
increased crowding of catalytic units on the support lowers
the selectivity and conversion (in the case of catalyst 1) for
the monometallic indole addition to a,b-unsaturated ke-
tones. Catalyst 3 with the longest linker was found to en-
hance the bi- and monometallic reaction pathway. For the
bimetallic pathway the longer linker facilitated salen–salen
interaction in the transition state, while for the monometal-
lic pathway, it improved solubility of the catalyst in the reac-
tion medium leading to improved activities. Lastly the effect
of the support flexibility was observed for the conjugate ad-
dition of tetrazoles for which significant enhancement of
catalytic activity was observed with the optimal catalyst 3
relative to its non-supported analogue. Such a study that
compares the effect of the support and linker on the catalyt-
ic activity of reactions which follow different mechanistic
pathways paves the way for the rational design of highly
active and selective supported catalysts.
1.74 (m, 2H; CH2), 1.45 (m, 2H; CH2), 1.41(s, 9H; CCAHTUNGRTNE(NUNG CH3)3), 1.38 (s,
9H; C(CH3)3), 1.24 (s, 9H; C(CH3)3), 1.19 (s, 3H; CH3), 1.19–1.24 (m,
A
ACHTUNGTRENNUNG
2H; CH2), 1.01–1.1 ppm (m, 2H; CH2); 13C NMR (75 MHz, CDCl3): d=
165.9, 164.9, 158.1, 154.4, 146.7, 140.2, 139.9, 138.5, 136.4, 126.9, 126.0,
118.3, 117.9, 117.7, 114.5, 72.4, 72.2, 60.5, 34.9, 34.8, 34.0, 33.15, 33.1, 31.4,
29.4, 29.2, 24.2, 14.2 ppm; HRMS (ESI+) calcd for C32H47N2O3 [M+H]+:
507.3581; found: 507.3547.
General procedure for the synthesis of salen cyclooctene esters: A solu-
tion of the salen ligand (1 equiv) in CH2Cl2 (0.13m) was added to a flame
dried round-bottom flask equipped with a magnetic stir bar and a reflux
condenser. DCC (1.1 equiv), the appropriate cyclooctene derived alcohol
(1 equiv), and DMAP (catalytic) were added to this solution. The reac-
tion mixture was heated under reflux for 12 h under an atmosphere of
Ar, following which the reaction mixture was diluted with CH2Cl2 and fil-
tered. The filtrate was dried over magnesium sulfate, filtered, and con-
centrated in vacuo to afford a yellow solid. The crude mixture was sub-
jected to flash column chromatography to afford the target product.
Salen cyclooctene ester 7: Salen ligand 4 (500 mg, 0.987 mmol), acid 5[10]
(152 mg, 0.987 mmol), and DCC (204 mg, 0.987 mmol) were used. Flash
chromatography (20:1 hexane/EtOAc) afforded 338 mg (58%) of the
product as a bright yellow solid. TLC Rf =0.43 (10:1 hexane/EtOAc);
1H NMR (300 MHz, CDCl3): d=13.84 (s, 1H; OH), 13.63 (brs, 1H;
OH), 8.30 (s, 1H; N=CH), 8.22 (s, 1H; N=CH), 7.31 (d, J=2.2 Hz, 1H;
Experimental Section
General: All starting materials were obtained from commercial suppliers
and used without further purification unless otherwise stated. All air- or
moisture-sensitive reactions were performed using oven-dried or flame-
dried glassware under an inert atmosphere of dry argon or nitrogen. Air-
or moisture-sensitive liquids and solutions were transferred by means of
a syringe or cannula. Dichloromethane was distilled from calcium hy-
dride, benzene from sodium, and 2-propanol from calcium sulfate.
Chem. Eur. J. 2009, 15, 1186 – 1194
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1191

M. Weck et al.
HAr), 6.97 (d, J=2.4 Hz, 1H; HAr), 6.89 (d, J=2.6 Hz, 1H; HAr), 6.72 (d,
J=2.0 Hz, 1H; HAr), 5.72 (m, 2H; CH=CH), 3.31 (m, 2H; 2NCHCH2),
2.73–2.36 (m, 2H; CHcyclooct), 2.27–2.08 (m, 4H; CH(cyclooct)), 2.06–1.63 (m,
11H; CH, CH2(cyclooct,cyclohex)), 1.45 (m, 2H; CH2(cyclohex)), 1.41 (s, 9H; C-
MS (MALDI-TOF) calcd for (C41H58N2O4)n: m/z (%): 1285.9 (100) [M]+
(n=2), 1928.4 (68) [M]+ (n=3), 2570.9 (22) [M]+ (n=4), 3213.4 (7)
[M]+ (n=5); GPC Mn =1300, Mw =1800, PDI=1.35.
Salen oligomer 10: Salen cyclooctene ester 8 (150 mg, 0.210 mmol) and
Grubbsꢁ third-generation initiator (7.4 mg, 0.008 mmol) were used. Purifi-
cation by flash column chromatography (gradient elution: 20:1!3:1
hexane/EtOAc) afforded 133 mg of the oligomer 10 (89%) as a bright
yellow solid. TLC Rf =0.23 (5:1 hexane/EtOAc); 1H NMR (300 MHz,
CDCl3): d=13.8 (brs, 2H; OH), 8.30 (s, 1H; N=CH), 8.22 (s, 1H; N=
CH), 7.32 (d, J=2.6 Hz, 1H; HAr), 6.99 (d, J=2.4 Hz, 1H; HAr), 6.93 (d,
J=2.7 Hz, 1H; HAr), 6.77 (d, J=2.9 Hz, 1H; HAr), 5.5–5.2 (brm, 2H;
CH=CH), 4.91 (m, 1H; CHOCO), 3.33 (m, 2H; 2NCHCH2), 2.82 (m,
2H; CH2(linker)), 2.70 (m, 2H; CH2(linker)), 2.2–1.3 (m, 18H; CH, CH2(cyclooct,
(CH3)3); 13C NMR
ACHTUNGTRENNUNG(CH3)3), 1.38 (s, 9H; CAHCUTNTRGEG(NNUN CH3)3), 1.24 ppm (s, 9H; CACTHNUGTRENNGUN
(75 MHz, CDCl3): d=176.5, 165.9, 164.7, 158.0, 157.9, 141.7, 139.9, 138.4,
136.3, 130.7, 129.5, 126.9, 125.9, 122.6, 121.3, 118.2, 117.7, 72.4, 72.2, 43.2,
34.9, 34.8, 34.0, 33.1, 31.6, 31.5, 31.4, 29.5, 29.4, 29.1, 27.8, 25.9, 25.4, 24.7,
24.2 ppm; HRMS (ESI+) calcd for C41H59N2O4 [M+H]+: 643.4469;
found: 643.451.
Salen cyclooctene ester 8:[10] Salen ligand 4 (500 mg, 0.987 mmol), acid
6[35] (223 mg, 0.987 mmol), and DCC (204 mg, 0.987 mmol) were used.
Flash chromatography (gradient elution: 20:1!5:1 hexane/EtOAc) af-
forded 650 mg (92%) of the product as a bright yellow solid. TLC Rf =
cyclohex)), 1.45 (m, 2H; CH2(cyclohex)), 1.41 (s, 9H; C
ACHTUGNTRENUN(GN CH3)3), 1.38 (s, 9H; C-
1
R
ACHTUNGTRENNUNG
0.22 (10:1 hexane/EtOAc); H NMR (300 MHz, CDCl3): d=13.89 (s, 1H;
171.98, 171.6, 166.1, 164.8, 158.4, 158.2, 141.7, 140.2, 138.8, 136.6, 132–128
(2C, multiple signals HC=CH), 127.1, 126.2, 122.9, 121.5, 118.4, 117.98,
72.7, 72.4, 35.2, 35.1, 34.3, 33.5, 33.3, 31.6, 29.7, 29.5 (overlapping signals),
29.1, 28.6, 24.5, 22.5 ppm; MS (MALDI-TOF) calcd for (C44H62N2O6)n:
m/z (%): 1429.9 (100) [M]+ (n=2), 2144.4 (75) [M]+ (n=3), 2858.9 (18)
[M]+ (n=4), 3574.5 (5) [M]+ (n=5); GPC Mn =1900, Mw =2500, PDI=
1.33.
OH), 13.62 (brs, 1H; OH), 8.30 (s, 1H; N=CH), 8.22 (s, 1H; N=CH),
7.32 (d, J=2.5 Hz, 1H; HAr), 6.99 (d, J=2.5 Hz, 1H; HAr), 6.93 (d, J=
2.9 Hz, 1H; HAr), 6.78 (d, J=3.0 Hz, 1H; HAr), 5.7–5.6 (m, 2H; CH=
CH), 4.87 (m, 1H; CHOCO) 3.33 (m, 2H; 2NCHCH2), 2.81 (m, 2H;
CH2(linker)), 2.67 (m, 2H; CH2(linker)), 2.4–2.05 (m, 2H; CHcyclooct), 2.0–1.55
(m, 14H; CH, CH2(cyclooct,cyclohex)), 1.45 (m, 2H; CH2(cyclohex)), 1.42 (s, 9H;
(CH3)3); 13C NMR
CACHTUNGTRENNUNG(CH3)3), 1.39 (s, 9H; CAHCTNUTRGEG(NNUN CH3)3), 1.25 ppm (s, 9H; CACTHNUGTRENNGUN
(75 MHz, CDCl3): d=171.6, 171.59, 166.1, 164.8, 158.4, 158.1, 141.7,
140.2, 138.8, 136.6, 130.1, 129.8, 127.1, 126.2, 122.9, 121.6, 118.4, 117.98,
76.4, 72.7, 72.4, 35.2, 35.1, 34.3, 33.9, 33.8, 33.5, 33.3, 31.6, 29.7, 29.67
(overlapping signals), 29.5, 29.4, 25.7, 25.0, 24.5, 22.5 ppm; HRMS (ESI+)
calcd for C44H63N2O6 [M+H]+: 715.4681; found: 715.4639.
Salen cyclooctene ester 12: Salen ligand 11[23] (241 mg, 0.381 mmol), acid
5[10] (74 mg, 0.38 mmol), and DCC (79 mg, 0.38 mmol) were used. Flash
chromatography (20:1 hexane/EtOAc) afforded 180 mg (61%) of the
product as a bright yellow solid. TLC Rf =0.48 (10:1 hexane/EtOAc);
1H NMR (300 MHz, CDCl3): d=13.73 (s, 1H; OH), 13.72 (brs, 1H;
OH), 8.32 (s, 1H; N=CH), 8.30 (s, 1H; N=CH), 7.31 (d, J=2.4 Hz, 1H;
HAr), 7.23 (d, J=2.3 Hz, 1H; HAr), 7.00 (d, J=2.4 Hz, 1H; HAr), 6.92 (d,
J=2.2 Hz, 1H; HAr), 5.8–5.59 (m, 2H; CH=CH), 3.98 (t, J=6.6 Hz, 2H;
CH2(linker)), 3.33 (m, 2H; 2NCHCH2), 2.48–2.29 (m, 3H; CHcyclooct), 2.2–
Salen oligomer 13: Salen cyclooctene ester 12 (161 mg, 0.209 mmol) and
Grubbsꢁ third-generation initiator (7.4 mg, 0.008 mmol) were used. Purifi-
cation by flash column chromatography (gradient elution: 25:1!10:1
hexane/EtOAc) afforded 125 mg of the oligomer 13 (78%) as a bright
yellow solid. TLC Rf =0.5 (5:1 hexane/EtOAc); 1H NMR (300 MHz,
CDCl3): d=13.7 (brs, 2H; OH), 8.31 (s, 1H; N=CH), 8.30 (s, 1H; N=
CH), 7.30 (d, J=2.5 Hz, 1H; HAr), 7.22 (d, J=2.4 Hz,1H; HAr), 7.00 (d,
J=2.5 Hz, 1H; HAr), 6.91 (d, J=2.5 Hz, 1H; HAr), 5.4–5.2 (brm, 2H;
CH=CH), 4.00 (brm, 2H; CH2(linker)), 3.32 (brm, 2H; 2NCHCH2), 2.1–1.3
(m, 27H; CH, CH2(cyclooct,cyclohex,linker)), 1.40 (s, 9H; C
(CH3)3), 1.23 (s, 9H; C(CH3)3), 1.19 (s, 3H; C(CH3)2), 1.18 (s, 3H; C-
(CH3)2), 1.05 ppm (m, 2H; CH2(linker)); 13C NMR (75 MHz, CDCl3): d=
ACHTUGNTRENUN(NG CH3)3), 1.396 (s, 9H;
C
N
N
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
176.6, 166.0, 165.98, 158.2, 158.16, 140.1, 138.7, 136.6, 136.5, 130–129.5
(2C, multiple signals HC=CH), 127.3, 126.94, 126.9, 126.3, 118.1, 118.07,
72.6, 72.58, 64.6, 44.5, 45.3–43.2 (1C, multiple signals CHCO2), 37.2 (over-
lapping signals), 35.2, 35.16, 34.2, 33.6, 33.55, 31.6 (overlapping signals),
30.1, 29.7, 29.65, 29.2, 29.1, 28.9, 28.1, 26.0, 25.97, 24.8 (overlapping sig-
nals), 24.6 ppm (overlapping signals); MS (MALDI-TOF) calcd for
(C50H76N2O4)n: m/z (%): 1538.3 (100) [M]+ (n=2), 2306.9 (67) [M]+
(n=3), 3076.6 (15) [M]+ (n=4), 3844.3 (3) [M]+ (n=5); GPC Mn=1850,
Mw =2600, PDI=1.39.
1.43 (m, 24H; CH, CH2(cyclooct,cyclohex,linker)), 1.41 (s, 9H; C
9H; C(CH3)3), 1.24 (s, 9H; C(CH3)3), 1.2 (s, 3H; C(CH3)2), 1.19 (s, 3H;
(CH3)2), 1.05 ppm (m, 2H; CH2(linker)); 13C NMR (75 MHz, CDCl3): d=
ACHTUGNTREN(UNNG CH3)3), 1.40 (s,
A
R
ACHTUNGTRENNUNG
CACHTUNGTRENNUNG
178.0, 166.0, 165.99, 158.2, 158.16, 140.1, 138.7, 136.6, 136.5, 130.7, 129.8,
127.3, 126.95, 126.91, 126.3, 118.1, 118.08, 72.6, 72.59, 64.6, 44.5, 43.7,
37.3, 37.2, 35.2, 35.1, 34.3, 33.6, 33.5, 31.9, 31.6, 30.1, 29.7, 29.6, 29.2,
29.17, 28.9, 28.1, 26.1, 25.97, 24.7, 24.6 (overlapping signals), 24.4 ppm;
HRMS (ESI+) calcd for C50H77N2O4 [M+H]+: 769.5878; found: 769.5882.
General procedure for the metalation of oligomeric salen: A solution of
the oligomer (1 equiv) in CH2Cl2 (0.16m) were added to a flame dried
Schlenk flask (10 mL) equipped with a magnetic stirrer bar in an atmos-
phere of argon. A solution of diethyl aluminum chloride in toluene
(1.8m, 1 equiv) was added slowly to the reaction flask. The reaction mix-
ture was allowed to stir for 3 h, following which the solvents were re-
moved in vacuo to afford a yellow solid. The resultant solid was rinsed
with hexane (3ꢃ1.5 mL) and dried to afford the metalated oligomeric
catalyst.
General procedure for the synthesis of macrocyclic oligomeric salen: In a
scintillation vial equipped with a magnetic stir bar, a solution of Grubbsꢁ
third-generation initiator (0.04 equiv) in deoxygenated CH2Cl2 was added
to a solution of monomer (1 equiv) in deoxygenated CH2Cl2 (final solu-
tion concentration 0.1m with respect to monomer). The reaction mixture
was stirred at RT for 30 min, followed by the addition of a few drops of
ethyl vinyl ether to quench the reaction. The reaction mixture was con-
centrated in vacuo and the resultant residue was purified by flash column
chromatography to obtain the pure oligomeric product.
Salen oligomer 9:[10] Salen cyclooctene ester 7 (161 mg, 0.25 mmol) and
Grubbsꢁ third generation initiator (8.8 mg, 0.01 mmol) were used. Purifi-
cation by flash column chromatography (gradient elution: 25:1!10:1
hexane/EtOAc) afforded 117 mg of the oligomer 9 (73%) as a bright
yellow solid. TLC Rf =0.46 (5:1 hexane/EtOAc); 1H NMR (300 MHz,
CDCl3): d=13.89 (s, 1H; OH), 13.63 (brs, 1H; OH), 8.30 (s, 1H; N=
CH), 8.24 (s, 1H; N=CH), 7.31 (d, J=1.0 Hz, 1H; HAr), 6.98 (d, J=
2.1 Hz, 1H; HAr), 6.89 (s, 1H; HAr), 6.74 (s, 1H; HAr), 5.6–5.3 (brm, 2H;
CH=CH), 3.31 (brm, 2H; 2NCHCH2), 2.53 (brm, 1H; CHCO2), 2.4–1.4
Oligomeric catalyst 1: Oligomer 9 (100 mg, 0.156 mmol) and a solution
of diethyl aluminum chloride in toluene (86.0 mL, 0.155 mmol) were
used to afford 108 mg of catalyst 1 (99%) as a bright yellow solid.
1H NMR (300 MHz, CDCl3): d=8.33 (brs, 1H; N=CH), 8.30 (brs, 1H;
N=CH), 7.56 (s, 1H; HAr), 7.13 (s, 1H; HAr), 7.09 (s, 1H; HAr), 6.97 (s,
1H; HAr), 5.6–5.3 (brm, 2H; CH=CH), 3.74 (brm, 1H; NCHCH2), 3.09
(brm, 1H; NCHCH2), 2.53 (brm, 1H; CHCO2), 2.8–1.3 (m, 18H; CH,
CH2(cyclooct, cyclohex)), 1.53 (brs, 18H; 2CACHTNUGTRENN(UG CH3)3), 1.31 ppm (s, 9H; CACHTUNGTRENNUNG(CH3)3);
13C NMR (75 MHz, CDCl3): d=175.3, 167.1, 164.7, 158.8, 158.5, 141.1,
140.7, 138.8 (overlapping signals), 132.5–129.7 (2C, multiple signals HC=
CH) , 128.3, 126.6, 123.1 (overlapping signals), 118.6 (overlapping sig-
nals), 72.7 (overlapping signals), 45.3–43.2 (1C, multiple signals CHCO2),
35.8, 35.4, 34.2, 33.5, 32.6 (overlapping signals), 31.6, 30.8–30.2 (m), 29.9,
29.7, 27.6–26.8 (m), 25.5, 23.8 ppm; ICP calcd (%): Al 3.8; found: Al 4.7.
(m, 18H; CH, CH2(cyclooct,cyclohex)), 1.40 (s, 9H; CACHTUNGTRNEN(UNG CH3)3), 1.39 (s, 9H; C-
A
ACHTUNGTRENNUNG
175.2, 166.1, 164.8, 158.3, 158.1, 141.7, 140.1, 138.7, 136.6, 132.5–128.7
(2C, multiple signals HC=CH), 127.1, 126.2, 122.9, 121.5, 118.4, 117.96,
72.7, 72.3, 45.3–43.2 (1C, multiple signals CHCO2), 35.2, 35.1, 34.2, 33.5,
33.4, 32.1, 31.6, 30.8–30.2 (m), 29.7, 29.3, 27.6–26.8 (m), 24.5, 24.2 ppm;
Oligomeric catalyst 2: Oligomer 10 (110 mg, 0.154 mmol) and a solution
of diethyl aluminum chloride in toluene (86.0 mL, 0.155 mmol) were used
1192
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1186 – 1194

Asymmetric Catalysis
FULL PAPER
to afford 110 mg of catalyst 2 (92%) as a bright yellow solid. 1H NMR
(300 MHz, CDCl3): d=8.4–8.0 (brm, 2H; 2N=CH), 7.54 (s, 1H; HAr),
7.13 (d, 1H; HAr), 7.08 (s, 1H; HAr), 6.96 (s, 1H; HAr), 5.5–5.2 (brm, 2H;
CH=CH), 4.9 (brm, 1H; CHOCO), 4.0–3.0 (brm, 2H; 2NCHCH2), 2.82
(m, 2H; CH2(linker)), 2.70 (m, 2H; CH2(linker)), 2.2–1.2 (m, 20H; CH,
1H;CHCHHCO), 2.38 (s, 3H; CH3), 1.5 ppm (d, J=7 Hz, 3H; CH3);
13C NMR (75 MHz, CDCl3): d=200.3, 137.5, 135.7, 133.0, 130.6, 128.7,
128.4, 125.3, 120.9, 119.3, 119.2, 115.7, 110.7, 45.9, 27.5, 21.3, 12.3 ppm.
General procedure for the kinetic studies for the indole addition reac-
tion: The appropriate catalyst (0.0137 mmol, 0.1 equiv), ketone 16
(20.0 mg, 0.137 mmol, 1 equiv), [D8]toluene (650 mL), indole 17 (27.0 mg,
0.206 mmol, 1.5 equiv), and 1,1,2,2-tetrachloroethane (8 mL) were added
to an NMR tube. NMR spectra were obtained at different time points
and comparison of the integration of the product peaks with the tetra-
chloroethane peak was used to determine conversions.
CH2(cyclooct,cyclohex)), 1.5 (s, 18H; 2CACTHNUTGRNENUG(CH3)3), 1.3 ppm (s, 9H; CACHUTGNTREN(NUNG CH3)3);
13C NMR (75 MHz, CDCl3): d=172.1, 171.6, 165.4, 161.8, 143.6, 141.3,
140.6, 138.9, 138.7, 132–128 (2C, multiple signals HC=CH), 129.3, 128.5,
128.1, 125.5, 123.1, 118.6, 118.4, 74.5, 74.4, 35.8, 35.2, 34.3, 34.2 (overlap-
ping signals), 31.6, 29.95, 29.7 (overlapping signals), 29.6, 28.7, 25.5,
23.97 ppm; ICP calcd (%): Al 3.5; found: Al 4.2.
General procedure for the tetrazole addition reaction: The appropriate
catalyst, tetrazole 19 (24.0 mg, 0.167 mmol, 1.2 equiv), and toluene
(700 mL) were added to a scintillation vial (3 mL) equipped with a mag-
netic stir bar. The appropriate a,b-unsaturated compound (0.137 mmol,
1 equiv) was added to the solution and the reaction was allowed to stir
for 26 h. Subsequently, the reaction mixture was concentrated in vacuo
and purified by flash column chromatography to afford the adduct 20.
Oligomeric catalyst 3: Oligomer 13 (123 mg, 0.160 mmol) and a solution
of diethyl aluminum chloride in toluene (89 mL, 0.16 mmol) were used to
afford 120 mg of catalyst 3 (90%) as a bright yellow solid. 1H NMR
(300 MHz, CDCl3): d=8.4–8.2 (brm, 2H; 2N=CH), 7.54 (s, 1H; HAr),
7.47 (s, 1H; HAr), 7.1 (s, 1H; HAr), 7.9 (s, 1H; HAr), 5.5–5.3 (brm, 2H;
CH=CH), 4.02 (brm, 2H; CH2(linker)), 3.6–3.3 (brm, 2H; 2NCHCH2), 2.6–
Tetrazole adduct 20a: TLC Rf =0.46 (3:1 hexane/EtOAc); 1H NMR
(300 MHz, CDCl3): d=8.17–8.1 (m, 2H; HAr), 8.0–7.95 (m, 2H; HAr),
7.65–7.55 (m, 1H; HAr), 7.52–7.4 (m, 5H; HAr), 5,71 (m, 1H; CHCH2),
3.98 (dd, J=17.9, 6.4 Hz, 1H; CHCHH), 3.6 (dd, J=17.8, 6.9 Hz, 1H;
CHCHH), 1.78 ppm (d, J=6.5 Hz, 3H; CH3); 13C NMR (75 MHz,
CDCl3): d=196, 165.1, 136.4, 133.96, 130.5, 129.1, 129.0, 128.3, 127.7,
127.1, 56.6, 44.3, 21.3 ppm.
Tetrazole adduct 20b:[34] TLC Rf =0.21 (3:1 hexane/EtOAc); 1H NMR
(300 MHz, CDCl3): d=9.17 (brs, 1H; NH), 8.18–8.1 (m, 2H; HAr), 7.9–
7.82 (m, 2H; HAr), 7.63–7.55 (m, 1H; HAr), 7.53–7.4 (m, 5H; HAr), 5,61
(m, 1H; CHCH2), 4.07 (dd, J=18.4, 9.7 Hz, 1H; CHCHH), 3.64 (dd, J=
18.4, 5.2 Hz, 1H; CHCHH), 1.77 ppm (d, J=7 Hz, 3H; CH3); 13C NMR
(75 MHz, CDCl3): d=172.7, 166.1, 165.1, 133.8, 132.3, 130.4, 129.3, 129.0,
128.1, 127.7, 127.1, 56.3, 43.6, 21.3 ppm.
1.0 (m, 27H; CH, CH2(cyclooct,cyclohex,linker)), 1.52 (s, 18H; 2C
ACHTUGNTREN(UNNG CH3)3), 1.3 (s,
9H; C(CH3)3), 1.26–1.23 (s, 6H; 2C(CH3)2), 0.85 ppm (m, 2H; CH2(linker));
A
ACHTUNGTRENNUNG
13C NMR (75 MHz, CDCl3): d=176.6, 166.7, 166.1, 165.0, 162.1, 158.2,
141.4, 137.8, 131.4, 131–129.5 (2C, multiple signals HC=CH), 128.8,
127.97, 126.9, 126.2, 118.4 (overlapping signals), 72.1 (overlapping sig-
nals), 64.6, 44.5, 45.3–43.2 (1C, multiple signals CHCO2), 37.2 (overlap-
ping signals), 35.8, 35.2, 34.3, 33.6, 33.2, 31.6 (overlapping signals), 30.1,
29.9, 29.65, 29.4, 29.2, 28.9 (overlapping signals), 26.0, 25.5, 24.9 (overlap-
ping signals), 24.5 ppm (overlapping signals); ICP calcd (%): Al 3.3;
found: Al 4.1.
General procedure for the cyanide addition reaction: The appropriate
catalyst (0.013 mmol, 0.05 equiv) was added to a Schlenk tube (25 mL)
and dried azeotropically with toluene (2ꢃ50 mL). Subsequently, imide 14
(50 mg, 0.26 mmol, 1 equiv), toluene (80 mL), and TMSCN (132 mL,
1.06 mmol, 4 equiv) were added to the flask. The reaction mixture was
heated gently with a heat gun and immersed in an oil-bath at 458C, fol-
lowing which 2-propanol (81 mL, 1.06 mmol, 4 equiv) was added. The
flask was sealed and allowed to stir for 18 h, following which the flask
was vented into an aqueous solution of FeSO4 to quench unreacted HCN
gas. After allowing the HCN to bubble out for 5–10 min, the solvent was
removed under vacuum. The crude reaction mixture was subjected to
flash column chromatography with 3:1 (hexane/EtOAc) to afford the
adduct 15 as a white solid.[20] Enantiomeric excess determined by chiral
HPLC (Pirkle-l-Leucine, 5% ethanol/hexanes, 0.7 mLminꢀ1, 254 nm);
General procedure for the kinetic studies for tetrazole addition reaction:
The appropriate catalyst, tetrazole 19 (24.0 mg, 0.167 mmol, 1.2 equiv),
[D8]toluene (700 mL), 1,1,2,2-tetrachloroethane (10 mL), and the a,b-unsa-
turated compound (0.137 mmol, 1 equiv) was added. Aliquots (20 mL)
were removed from the reaction mixture at different time intervals and
diluted with CDCl3 (400 mL). NMR spectra were obtained at different
time points and comparison of the integration of the product peaks with
the tetrachloroethane peak was used to determine conversions.
Theoretical methods: Density functional theory (DFT), executed with
the Jaguar suite of programs,[36] was used to compute the optimized sin-
glet state structures of the AlCl–salen catalysts with the B3LYP function-
al[37–39] and the 6-31G* basis set. Frequency computations at the con-
verged geometries were performed to ensure the structures corresponded
to potential energy minima. Atomic charges were computed by fitting to
the DFT electrostatic potential.[40–42]
1
TLC Rf =0.14 (3:1 hexane/EtOAc); H NMR (300 MHz, CDCl3): d=8.88
(s, 1H; NH), 7.89 (dd, J=8.0, 1.5 Hz, 2H; HAr), 7.64 (t, J=7 Hz, 1H;
HAr), 7.54 (t, J=6.3 Hz, 2H; HAr), 3.47 (dd, J=17.2 , 7 Hz, 1H; CHCN),
3.31 (m, 1H CNCHCHH); 3.2 (m, 1H; NCCHCHH), 1.45 ppm (d, 3H;
J=7.3 Hz, CH3).
General procedure for the kinetic studies for the cyanide addition reac-
tion: The addition reaction was divided into 5–6 NMR tubes. In each
tube the appropriate catalyst, imide substrate (7.6 mg. 0.04 mmol,
1 equiv), and toluene (4 mL) were added. Each NMR tube was sealed
with a septum and TMSCN (20 mL, 0.16 mmol, 4 equiv) and 2-propanol
(12 mL, 0.16 mmol, 4 equiv) were added. The reaction was heated to
458C and quenched at the appropriate time by adding CDCl3 (400 mL).
The amount of product formation was calculated by adding a known
amount of 1,1,2,2-tetrachloroethane (0.0075 mmol) in CDCl3 (100 mL) to
the NMR tube as a standard.
Acknowledgements
This research is supported by the U.S. DOE Office of Basic Energy Sci-
ences through the Catalysis Science Contract No. DE-FG02-03ER15459.
M.W. gratefully acknowledges a Camille Dreyfus Teacher/Scholar Award
and an Alfred P. Sloan Fellowship.
General procedure for indole addition reaction: The appropriate catalyst
(0.0137 mmol, 0.1 equiv), ketone 16[32] (20.0 mg, 0.137 mmol, 1 equiv),
and toluene (650 mL) were added to a scintillation vial (3 mL) equipped
with a magnetic stir bar. Indole 17 (27.0 mg, 0.206 mmol, 1.5 equiv) was
added to the solution and the reaction was allowed to stir for 18 h. Subse-
quently, the reaction mixture was concentrated in vacuo and purified by
flash column chromatography to afford the adduct 18.[22] Enantiomeric
excess determined by chiral HPLC (Chiralcel OD, 15% isopropanol/hex-
[1] C. Baleiz¼o, H. Garcia, Chem. Rev. 2006, 106, 3987–4043.
[2] L. Canali, D. C. Sherrington, Chem. Soc. Rev. 1999, 28, 85–93.
[3] N. E. Leadbeater, M. Marco, Chem. Rev. 2002, 102, 3217–3274.
[4] A. Corma, H. Garcia, Adv. Synth. Catal. 2006, 348, 1391–1412.
[5] Q.-H. Fan, Y.-M. Li, A. S. C. Chan, Chem. Rev. 2002, 102, 3385–
3466.
[6] R. Breinbauer, E. N. Jacobsen, Angew. Chem. 2000, 112, 3750–3753;
Angew. Chem. Int. Ed. 2000, 39, 3604–3607.
[7] P. K. Dhal, B. B. De, S. Sivaram, J. Mol. Catal. A 2001, 177, 71–87.
[8] J. M. Notestein, A. Katz, Chem. Eur. J. 2006, 12, 3954–3965.
anes, 0.5 mLminꢀ1
, 254 nm); TLC Rf =0.46 (3:1 hexane/EtOAc);
1H NMR (300 MHz, CDCl3): d=7.92–7.85 (m, 1H; HAr), 7.75–7.65 (m,
1H; HAr), 7.6–7.2 (m, 5H; HAr), 7.15–7.05 (m, 2H; HAr), 3.75 (m, 1H;
CH), 3.54 (dd, J=16.3, 6.9 Hz, 1H; CHHCO), 3.38 (dd, , J=16.2, 7.3 Hz,
Chem. Eur. J. 2009, 15, 1186 – 1194
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1193

M. Weck et al.
[9] H. Sellner, J. K. Karjalainen, D. Seebach, Chem. Eur. J. 2001, 7,
2873–2887.
[10] X. Zheng, C. W. Jones, M. Weck, J. Am. Chem. Soc. 2007, 129,
1105–1112.
[11] D. E. White, E. N. Jacobsen, Tetrahedron: Asymmetry 2003, 14,
3633–3638.
[26] E. Thorn-Csꢅnyi, K. Ruhland, Macromol. Symp. 2000, 153, 145–150.
[27] M. Holbach, X. Zheng, C. Burd, C. W. Jones, M. Weck, J. Org.
Chem. 2006, 71, 2903–2906.
[28] See Experimental Section and Supporting Information for details.
[29] The data was highly reproducible provided the catalysts were stored
in an atmosphere of argon at low temperatures.
[12] N. Madhavan, C. W. Jones, M. Weck, Acc. Chem. Res. 2008, 41,
1153–1165.
[13] R. G. Konsler, J. Karl, E. N. Jacobsen, J. Am. Chem. Soc. 1998, 120,
10780–10781.
[14] L. Canali, E. Cowan, H. Deleuze, C. L. Gibson, D. C. Sherrington,
Chem. Commun. 1998, 2561–2562.
[15] T. S. Reger, K. D. Janda, J. Am. Chem. Soc. 2000, 122, 6929–6934.
[16] M. Holbach, M. Weck, J. Org. Chem. 2006, 71, 1825–1836.
[17] X. Zheng, C. W. Jones, M. Weck, Adv. Synth. Catal. 2008, 350, 255–
261.
[18] B. B. De, B. B. Lohray, S. Sivaram, P. K. Dhal, J. Polym. Sci. Part A
1997, 35, 1809–1818.
[19] F. Minutolo, D. Pini, A. Petri, P. Salvadori, Tetrahedron: Asymmetry
1996, 7, 2293–2302.
[20] G. M. Sammis, E. N. Jacobsen, J. Am. Chem. Soc. 2003, 125, 4442–
4443.
[30] See theoretical methods and Supporting Information for details.
[31] Catalyst 2 with the medium linker was not tested as it led to adducts
with lower enantioselectivities for the cyanide addition reaction.
[32] M. R. Pitts, J. R. Harrison, C. J. Moody, J. Chem. Soc. Perkin Trans.
1 2001, 955–977.
[33] M. Bandini, M. Fagioli, P. Melchiorre, A. Melloni, A. Umani-
Ronchi, Tetrahedron Lett. 2003, 44, 5843–5846.
[34] M. Gandelman, E. N. Jacobsen, Angew. Chem. 2005, 117, 2445–
2449; Angew. Chem. Int. Ed. 2005, 44, 2393–2397.
[35] K. Breitenkamp, J. Simeone, E. Jin, T. Emrick, Macromolecules
2002, 35, 9249–9252.
[36] Jaguar 5.5, L. L. C. Schrçdinger, Portland, OR, 1991–2003.
[37] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
[38] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
[39] P. J. Stephens, F. J. Devlin, C. F. Chabalowski, M. J. Frisch, J. Phys.
Chem. 1994, 98, 11623–11627.
[21] G. M. Sammis, H. Danjo, E. N. Jacobsen, J. Am. Chem. Soc. 2004,
126, 9928–9929.
[40] L. E. Chirlian, M. M. Francl, J. Comput. Chem. 1987, 8, 894–905.
[41] R. J. Woods, M. Khalil, W. Pell, S. H. Moffat, V. H. Smith, Jr., J.
Comput. Chem. 1990, 11, 297–310.
[42] C. M. Breneman, K. B. Wiberg, J. Comput. Chem. 1990, 11, 361–
373.
[22] M. Bandini, M. Fagioli, M. Garavelli, A. Melloni, V. Trigari, A.
Umani-Ronchi, J. Org. Chem. 2004, 69, 7511–7518.
[23] N. Madhavan, M. Weck, Adv. Synth. Catal. 2008, 350, 419–425.
[24] M. Alvaro, C. Baleizao, E. Carbonell, M. El Ghoul, H. Garcꢄa, B.
Gigante, Tetrahedron 2005, 61, 12131–12139.
[25] E. Thorn-Csꢅnyi, K. Ruhland, Macromol. Chem. Phys. 1999, 200,
1662–1671.
Received: August 5, 2008
Published online: December 19, 2008
1194
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1186 – 1194