Supporting multiple organometallic catalysts on poly(norbornene) for cyanide addition to α,β-unsaturated imides

Article abstract of DOI:10.1016/j.molcata.2010.10.023

A two-catalyst system comprising of salen(AlCl) and pybox(ErCl₃) complexes supported onto poly(norbornene) has been developed. As a proof of concept the activity of the two-catalyst system was gauged for the addition of cyanide to α,β-unsaturated imides which has been shown to follow a bimetallic mechanism. The activity of the supported two-catalyst system was significantly higher than catalytic systems derived from mixtures of the two catalysts. In addition, the co-polymer could be readily recovered and reused up to 3 cycles without significant loss in conversions and yields. Nevertheless, a decrease in enantioselectivity of the cyanide adduct was observed with each subsequent cycle indicating some loss in catalytic selectivity. The reported strategy opens up avenues for supporting multiple catalysts on the same polymer backbone for catalyst-based one-pot cascade reactions.

Full text of DOI:10.1016/j.molcata.2010.10.023

Journal of Molecular Catalysis A: Chemical 334 (2011) 1–7
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Editor’s Choice paper
Supporting multiple organometallic catalysts on poly(norbornene) for cyanide
addition to ␣,␤-unsaturated imides
Nandita Madhavana, William Sommera, Marcus Wecka,b,∗
a School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, GA 30332, United States
b Molecular Design Institute and Department of Chemistry, New York University, New York, NY 10003, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A
two-catalyst system comprising of salen(AlCl) and pybox(ErCl3) complexes supported onto
Received 27 August 2010
Accepted 26 October 2010
Available online 4 November 2010
poly(norbornene) has been developed. As a proof of concept the activity of the two-catalyst system
was gauged for the addition of cyanide to ␣,␤-unsaturated imides which has been shown to follow a
bimetallic mechanism. The activity of the supported two-catalyst system was significantly higher than
catalytic systems derived from mixtures of the two catalysts. In addition, the co-polymer could be readily
recovered and reused up to 3 cycles without significant loss in conversions and yields. Nevertheless, a
decrease in enantioselectivity of the cyanide adduct was observed with each subsequent cycle indicat-
ing some loss in catalytic selectivity. The reported strategy opens up avenues for supporting multiple
catalysts on the same polymer backbone for catalyst-based one-pot cascade reactions.
© 2010 Elsevier B.V. All rights reserved.
Keywords:
Supported catalysis
Salen
ROMP
Aluminum
Bimetallic mechanism
1. Introduction
An ideal reaction to evaluate our multiple supported catalyst
system is the conjugate addition of cyanide to ␣,␤-unsaturated
A rapidly emerging area in salen catalysis involves the immobi-
lization of salen catalysts onto polymers [1–4]. Supported catalysts
can be removed readily from reaction mixtures and often recycled
resulting in higher turn over numbers as well as lower metallic
waste disposal for these catalysts as compared to their unsupported
analogs. We have demonstrated that a flexible polymer backbone
enhances the activity of salen catalysts for reactions involving a
bimetallic transition state such as the hydrolytic kinetic resolu-
tion of epoxides or the cyanide addition to ␣,␤-unsaturated imides
[1,5–9]. We hypothesize that the flexible backbone enhances the
local catalyst concentration thereby improving the catalytic activ-
ity of transformations that have bimetallic transition state. There
are, however, no examples where multiple catalysts that work
together in a bimetallic transition state have been supported
onto the same polymer backbone. In this contribution, we real-
ize such a goal by extending our expertise in supported salen
catalysis and norbornene chemistry by immobilizing two catalysts
onto poly(norbornene) using highly functional group tolerant ring-
opening metathesis polymerization (ROMP) [10,11]. Furthermore,
we investigate the catalytic activity of the resulting catalyst system
and compare it to its non-supported analogues.
imides since it has been reported that a combination of salen(AlCl)
and pybox(ErCl3) complexes works superior to each individ-
ual complex for the reaction catalysis [12,13]. Poly(norbornene)
was chosen as the support since norbornene monomers can
be polymerized in a highly controlled and often living fashion
using functional group tolerant ring-opening metathesis poly-
merization (ROMP). Furthermore, we have demonstrated that
poly(norbornene) enhances the catalytic activity of salen(AlCl) for
the cyanide addition reaction making poly(norbornene)-supported
salen (AlCl) catalysts ideal candidates for our study [8]. The basic
target copolymer 1 consisting of poly(norbornene) that supports
salen(AlCl) and pybox(ErCl3) complexes is shown in Fig. 1.
2. Experimental
2.1. 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 solu-
tions were transferred via syringe or cannula. Dichloromethane
and acetonitrile were distilled from calcium hydride, acetone from
potassium carbonate, toluene from sodium, and 2-propanol from
calcium sulfate. Thionyl chloride was distilled prior to use. Cau-
tion: Trimethylsilyl cyanide and hydrogen cyanide are highly toxic
∗
Corresponding author at: Molecular Design Institute and Department of Chem-
istry, New York University, New York, NY 10003, United States.
Tel.: +1 212 992 7968; fax: +1 212 995 4895.
E-mail addresses: marcus.weck@nyu.edu, mw125@nyu.edu (M. Weck).
1381-1169/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2010.10.023

2
N. Madhavan et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 1–7
Fig. 1. Representation of the poly(norbornene) supported salen(AlCl) and pybox(ErCl3) catalyst.
and should be handled extremely carefully in a fume hood as per
the experimental protocol mentioned below. Analytical thin layer
chromatography (TLC) was performed using Silica XHL pre-coated
(250 ␮m thickness) glass backed TLC plates from Sorbent Tech-
nologies. Eluting solvents are reported as volume ratios or volume
percents. Compounds were visualized using UV light, potassium
permanganate, p-anisaldehyde or iodine stains. Flash column chro-
filtered and concentrated in vacuo to afford an oily residue. The
residue was recrystallized with ethyl acetate (8 mL) to afford 3.54 g
(90%) of ether 5 as a white solid. 1H NMR (300 MHz, CDCl3) ı = 7.85
(s, 2H, HAr(pyr)), 7.50–7.30 (m, 5H, HAr(Bn)), 5.20 (s, 2H, CH2(Bn)), 4.45
(q, 4H, J = 7.2 Hz, OCH2CH3); 1.44 (t, 6H, J = 7.2 Hz, OCH2CH3); 13C
NMR (75 MHz, CDCl3) ı = 166.8, 164.9, 150.5, 134.9, 129.1, 129.0,
128.0, 114.8, 71.0, 62.6, 14.4; HRMS (FAB) calcd. for C18H20NO5
(MH+) 330.1342, found 330.1352.
˚
matography was performed using silica gel 60 A (230–400 mesh).
All 1H and 13C NMR spectra were recorded on Varian Mercury
Vx 300 or Varian Mercury Vx 400 spectrometers using CDCl3 as
the solvent. Chemical shifts are expressed in parts per million (ı),
coupling constants (J) are reported in Hertz (Hz), and splitting pat-
terns are reported as singlet (s), doublet (d), triplet (t), quartet (q),
unresolved multiplet (m), and apparent (app). All NMR spectra are
referenced to residual solvent peaks as the standard with ı values
of 7.26 ppm for CDCl3. High-resolution mass spectra were obtained
from the Georgia Institute of Technology mass spectrometry lab.
Chiral HPLC analyses were performed on a Shimadzu-10A system,
using Pirkle-l-Leucine column from Regis Technologies, Inc.
2.3. Diacid chloride 6 [14]
Pyridyl benzyl ether 5 (2.35 g, 0.007 mol, 1 equiv.), ethanol
(56 mL) and potassium hydroxide (1.40 g, 0.03 mol, 3.5 equiv.) were
added to a 100 mL round-bottomed flask equipped with a stir bar.
The reaction mixture was heated to 50 ◦C and allowed to stir for
45 min. Subsequently, the reaction mixture was cooled to room
temperature and filtered. The precipitate was washed with cold
ethanol to afford 2.40 g (98%) of the pyridyl dicarboxylic dipotas-
sium salt as a white solid. The dipotassium salt (2.50 g, 7.1 mmol,
1 equiv.) was then added in portions over 1 h to a solution of thionyl
chloride (7.80 mL, 105 mmol, 14.8 equiv.) in toluene (16 mL) in a
100 mL Schlenk flask. After the addition of the salt was complete,
the reaction mixture was heated under reflux for 4.5 h under an
atmosphere of Ar, following which the reaction was concentrated
under vacuum to afford a white residue. To the residue CH2Cl2
(300 mL) was added and the resultant solution was filtered. The
filtrate was concentrated in vacuo to afford a white residue which
was recrystallized from hexane (30 mL) to afford 1.60 g (73%) of the
diacid chloride 6 as a white solid. 1H NMR (300 MHz, CDCl3) ı = 7.87
2.2. Pyridyl benzyl ether 5 [14]
Pyridyl diester 4 [15] (2.62 g, 0.012 mol, 1 equiv.) and acetone
(50 mL) were added to a flame dried 100 mL Schlenk flask equipped
with a stir bar and a reflux condenser under an atmosphere of Ar.
Subsequently, benzyl bromide (4.28 g, 0.025 mol, 2.1 equiv.) and
potassium carbonate (3.40 g, 0.025 mol, 2.1 equiv.) were added to
the flask. The reaction mixture was heated under reflux for 19 h
under an atmosphere of Ar, following which the reaction mixture
was cooled and filtered. The filtrate was concentrated in vacuo to
afford a yellow oil, which was dissolved in CH2Cl2 (200 mL). The
resultant solution was washed with water (2 × 100 mL), saturated
NaCl solution (100 mL), dried over anhydrous magnesium sulfate,
(s, 2H, HAr(pyr)), 7.50–7.30 (m, 5H, HAr(Bn)), 5.26 (s, 2H, CH2(Bn)); 13
NMR (75 MHz, CDCl3) ı = 169.8, 167.4, 150.8, 134.1, 129.4, 129.3,
128.01, 115.9, 71.7.
C

N. Madhavan et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 1–7
3
2.4. Pyridyl diamide 7
46.6, 43.4, 41.9, 33.9, 32.8, 30.5, 28.7, 28.0, 25.4; HRMS (FAB) calcd.
for C14H22O2 Br (MH+) 301.0803, found 301.0825.
R-phenyl glycinol (1.10 g, 7.8 mmol, 2.2 equiv.), CH2Cl2 (10 mL)
and triethylamine (2.90 mL, 21 mmol, 6 equiv.) were added to a
100 mL Schlenk flask equipped with a magnetic stir bar under
an atmosphere of Ar. The solution was cooled to 0 ◦C and a solu-
tion of the diacid chloride 6 (1.10 g, 3.5 mmol, 1 equiv.) in CH2Cl2
(13 mL) was slowly added to it over a period of 30 min. After the
addition was complete, the reaction mixture was warmed to room
temperature and allowed to stir for 24 h, following which thionyl
chloride (2.60 mL, 35 mmol, 10 equiv.) was added to the solution.
The reaction mixture was heated under reflux for 3 h and sub-
sequently poured into ice water (100 mL). The organic layer was
separated and washed with saturated NaCl solution (100 mL), 0.1 M
aqueous potassium carbonate (300 mL), dried over anhydrous mag-
nesium sulfate, filtered and concentrated in vacuo to afford an oily
residue. The residue was subjected to flash column chromatog-
raphy (CH2Cl2) to afford 980 mg of the hydrochloride salt of the
product as an off-white solid. The solid was stirred with 5% aqueous
NaOH (9 mL) in methanol (27 mL) for 24 h, following which CH2Cl2
was added. The organic layer was separated, dried over magne-
sium sulfate, filtered and concentrated to afford 883 mg (46%) of
the neutralized product 7. 1H NMR (300 MHz, CDCl3) ı = 8.64 (d, 2H,
2.7. Non-metallated pybox monomer 10
Norbornene ester 9 (477 mg, 1.58 mmol, 2.5 equiv.), potassium
carbonate (218 mg, 1.58 mmol, 2.5 equiv.) and anhydrous DMF
(5 mL) were added to a Schlenk flask (25 mL) equipped with a
magnetic stir bar and an addition funnel. A solution of 8 (290 mg,
0.633 mmol, 1 equiv.) in anhydrous DMF (3 mL) was added slowly
to the reaction mixture at room temperature via the addition funnel
over 1 h. Upon completion of the addition, the reaction mixture was
heated to 60 ◦C and allowed to stir for 48 h, following which DMF
was removed in vacuo to give a yellow oily residue. CH2Cl2 (20 mL)
and water (20 mL) were added to the residue and the organic layer
was separated. The aqueous layer was subjected to extractions
with CH2Cl2 (2 × 50 mL) and the combined organic extracts were
dried over anhydrous magnesium sulfate, filtered and concentrated
in vacuo to afford an oily residue. The crude mixture was purified
by flash column chromatography (Gradient elution: CH2Cl2 → 3:1
CH2Cl2/EtOAc) to afford 247 mg (64%) of the monomer 10 as an oil.
1H NMR (300 MHz, CDCl3) ı = 7.99 (s, 2H, HAr(pyr)), 7.50–7.30 (m,
10H, HAr(Ph)), 6.15–6.05 (m, 2H, CH = CH(nb)), 5.44 (dd, 2H, J = 10.5,
2 Hz, CHPh), 4.92 (t, 2H, J = 8.7 Hz, CH2(linker)), 4.41 (t, 2H, J = 8.8 Hz,
CH2(linker)), 4.15–4.04 (m, 6H, CH2(linker), CH2(pybox)), 3.00 (br s, 1H,
CH(nb)), 2.91 (br s, 1H, CH(nb)), 2.24–2.17 (m, 1H, CH(nb)), 1.90–1.70
(m, 3H, CH2(linker), CH(nb)), 1.72–1.60 (m, 2H, CH2(linker)), 1.54–1.30
(m, 7H, CH2(linker), CH(nb)); 13C NMR (75 MHz, CDCl3) ı = 176.7,
166.4, 163.8, 148.4, 142.0, 138.3, 136.1, 129.1, 128.2, 126.8, 113.1,
75.7, 70.5, 68.8, 64.7, 62.9, 47.0, 46.8, 43.6, 42.1, 33.1, 30.7, 29.3,
25.9; HRMS (ESI+) calcd. for C37H40N3O5 (MH+) 606.2962, found
606.2982.
J = 8.2 Hz, CONH), 7.92 (s, 2H, HAr(pyr)), 7.50–7.30 (m, 15H, HAr(Bn)
,
HAr(Ph)), 5.55–5.45 (m, 2H, NHCHPh), 5.15 (d, 2H, J = 4 Hz, CH2(Bn)),
4.05–3.88 (m, 4H, CH2Cl); 13C NMR (75 MHz, CDCl3) ı = 168.0, 163.1,
150.6, 138.5, 135.0, 129.1, 129.0, 128.9, 128.5, 127.9, 126.8, 112.2,
70.9, 53.8, 48.6; HRMS (FAB) calcd. for C30H28N3O3 Cl2 (MH+)
548.1508, found 548.1467.
2.5. Pyridyl diamide OH 8
A solution of benzylated diamide 7 (320 mg, 0.583 mmol,
1 equiv.) in ethyl acetate (45 mL) and methanol (15 mL) and 10%
palladium on carbon (ca. 100 mg) were added to a 100 mL round-
bottomed flask equipped with a magnetic stir bar. A hydrogen
balloon was added to the flask and the reaction mixture was stirred
for 18 h in an atmosphere of hydrogen. Subsequently, the reaction
mixture was filtered through a pad of celite. The celite was rinsed
with ethyl acetate (50 mL) and the combined filtrates were con-
centrated in vacuo to afford 265 mg (99%) of the pure product as a
white solid. 1H NMR (300 MHz, CDCl3) ı = 9.34 (br s, 1H, OH), 8.64
(d, 2H, J = 8.2 Hz, CONH), 7.99 (s, 2H, HAr(pyr)), 7.50–7.30 (m, 10H,
2.8. Metallated pybox monomer 3
Monomer 10 (200 mg, 0.330 mmol, 1 equiv.), erbium chloride
(90 mg, 0.33 mmol, 1 equiv.) and acetonitrile (15 mL) were added
to a Schlenk flask (25 mL) equipped with a magnetic stir-bar. The
reaction mixture was stirred at room temperature under an atmo-
sphere of Ar for 2.5 h. The solvent was removed in vacuo to afford
286 mg (99%) of the metallated monomer 3 as an off-white solid.
Since Er is paramagnetic the complex could not be characterized
by NMR spectroscopy. Anal. calcd. for C37H39N3O5ErCl3: C, 50.54;
H, 4.47; N, 4.78; Er, 19.02; found: C, 46.48; H, 4.83; N, 4.67; Er, 18.9.
HAr(Ph)), 5.55–5.45 (m, 2H, NHCHPh), 4.05–3.88 (m, 4 H, CH2Cl); 13
C
NMR (75 MHz, CDCl3) ı = 167.2, 163.5, 150.2, 138.2, 129.2, 128.6,
126.8, 113.7, 54.0, 48.7; HRMS (ESI+) calcd. for C23H22N3O3 Cl2
(MH+) 458.1033, found 458.1075.
2.9. Catalyst 1
A solution of pybox monomer 3 (40 mg, 0.045 mmol, 10 equiv.)
and salen monomer 2 [8] (37 mg, 0.045 mmol, 10 equiv.) in deoxy-
genated CH2Cl2 (2 mL) was added to a scintillation vial (20 mL).
Subsequently, a solution of Grubbs 3rd generation initiator (4 mg,
0.005 mmol, 1 equiv.) in deoxygenated CH2Cl2 (1.5 mL) was added
to the vial. The reaction mixture was stirred for 1 h at room tem-
perature, following which 5 drops of ethyl vinyl ether was added
to quench the polymerization. The reaction mixture was concen-
trated to 2 mL, the polymer was precipitated using diethyl ether
and isolated via centrifugation. The process of precipitation and
centrifugation was repeated three times to afford 71 mg (92%) of
catalyst 1 as an off-white solid. Anal. calcd. for C86H109N5O9ErAlCl4:
C, 61.02; H, 6.49; N, 4.14; Er, 9.88; Al, 1.59; found: C, 56.43; H, 6.37;
N, 4.91; Er, 9.93; Al, 1.9.
2.6. Norbornene ester 9 [11]
Norbornene exo-acid (500 mg, 3.62 mmol, 1 equiv.) and CH2Cl2
(30 mL) were added to a 100 mL round-bottomed flask equipped
with a magnetic stir bar and a reflux condenser. To the solu-
tion, DCC (747 mg, 3.62 mmol, 1 equiv.), 6-bromo hexanol (474 ␮L,
3.62 mmol, 1 equiv.), and DMAP (catalytic) were added. The reac-
tion mixture was heated under reflux for 16 h under an atmosphere
of Ar, following which the reaction mixture was diluted with CH2Cl2
and filtered. The filtrate was dried over magnesium sulfate, filtered
and concentrated in vacuo to afford an oil. The crude mixture was
purified by flash column chromatography (20:1, hexane/EtOAc)
to afford 880 mg (81%) of the ester 9 as a colorless oil. 1H NMR
(300 MHz, CDCl3) ı = 6.15–6.05 (m, 2H, CH = CH(nb)), 4.08 (t, 2H,
J = 6.6 Hz, OCOCH2), 3.40 (t, 2H, J = 7.1 Hz, CH2Br), 3.10 (br s, 1H,
CH(nb)), 2.91 (br s, 1H, CH(nb)), 2.21 (m, 1H, CH(nb)), 1.95–1.80 (m, 3H,
2.10. Catalyst precursor 11
CH2(alk), CH(nb)), 1.65 (m, 2 H, CH2(alk)), 1.54–1.33 (m, 7H, CH2(alk)
,
A solution of pybox monomer 10 (70 mg, 0.116 mmol, 25 equiv.)
and salen monomer 2 [8] (94 mg, 0.116 mmol, 25 equiv.) in deoxy-
CH(nb)); 13C NMR (75 MHz, CDCl3) ı = 176.5, 138.2, 136.0, 64.5, 46.8,

4
N. Madhavan et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 1–7
genated CH2Cl2 (6 mL) was added to a scintillation vial (20 mL).
Subsequently, a solution of Grubbs 3rd generation initiator (4 mg,
0.005 mmol, 1 equiv.) in deoxygenated CH2Cl2 (3 mL) was added
to the vial. The reaction mixture was stirred for 1 h at room tem-
perature, following which 5 drops of ethyl vinyl ether was added
to quench the polymerization. The reaction mixture was concen-
trated to 3 mL, the polymer was precipitated using diethyl ether
and isolated via centrifugation. The process of precipitation and
centrifugation was repeated three times to afford 114 mg (70%) of
catalyst precursor 11 as a pale yellow solid. The precursor is insol-
uble in most organic solvents and could not be characterized by
NMR spectroscopy.
The suspension was stirred at room temperature in an atmosphere
of Ar for 1.5 h, following which the solvent was removed in vacuo to
afford the pybox/ErCl3 catalyst. An appropriate amount of the salen
catalyst (if required) was added to the Schlenk tube. The catalyst
mixture was dried azeotropically with toluene (2 × 50 ␮L). Subse-
quently, the imide 13 (50 mg, 0.26 mmol, 1 equiv.), toluene (80 ␮L)
and TMSCN (132 ␮L, 1.06 mmol, 4 equiv.) were added to the tube.
The reaction mixture was heated gently with a heat gun, immersed
in an oil bath at 45 ◦C and isopropanol was added (81 ␮L, 1.06 mmol,
4 equiv.). The flask was sealed and allowed to stir for 18 h, follow-
ing which the reaction mixture was vented into a FeSO4 solution to
quench any un-reacted hydrogen cyanide. After allowing the HCN
to bubble out for 10–15 min, the solvent was removed in vacuo to
afford the crude cyanide adduct 14 which was purified by flash
column chromatography.
2.11. Small molecule pybox 16
Tetrahydrofuran (10 mL) and 60% sodium hydride in oil (187 mg,
4.68 mmol, 2.7 equiv.) were added to a Schlenk flask (50 mL)
immersed in an ice bath at 0 ◦C. A solution of benzylated dimide
7 (950 mg, 1.73 mmol, 1 equiv.) in THF (12 mL) was added slowly to
the reaction mixture. The reaction mixture was stirred at 0 ◦C for
1.5 h, following which the insoluble residue was removed by filtra-
tion. The filtrate was concentrated in vacuo to afford a residue to
which diethyl ether (9 × 100 mL) was added. The suspension was
triturated and filtered. The combined filtrates were concentrated
in vacuo to afford a mixture of pybox 16 and diimide 7 (1:0.2).
Since the crude mixture decomposed in the presence of silica or
alumina, the mixture was resubjected to the reaction conditions to
ensure complete conversion of diimide 7 to pybox 16. After reaction
work-up as described above 600 mg (73%) of pybox 16 was obtained
as an off-white solid. 1H NMR (300 MHz, CDCl3) ı = 7.95 (s, 2H,
HAr(pyr)), 7.50–7.30 (m, 15H, HAr(Bn), HAr(Ph)), 5.45 (dd, 2H, J = 10.5,
1.7 Hz CHPh), 5.15 (s, 2H, J = 4 Hz, CH2(Bn)), 4.92 (dd, 2H, J = 10.6,
1.7 Hz, CHH(pybox)), 4.41 (app t, 2H, J = 8.7 Hz, CHH(pybox)); 13C NMR
(75 MHz, CDCl3) ı = 166.0, 163.8, 148.5, 141.9, 135.3, 129.1, 129.0,
128.7, 128.1, 127.8, 127.1, 113.3, 75.8, 70.8, 70.5; HRMS (ESI+) calcd.
for C30H26N3O3 (MH+) 476.1969, found 476.1924.
2.14. General procedure for cyanide addition reaction using
catalyst 1 (Table 2) and recycling studies
Catalyst 1 (22.4 mg, 0.013 mmol, 0.05 equiv.) was added to a
Schlenk tube (50 mL) equipped with a magnetic stir bar. The
catalyst was dried azeotropically with toluene (2 × 50 ␮L). Subse-
quently, the imide 13 (50 mg, 0.26 mmol, 1 equiv.), toluene (80 ␮L)
and TMSCN (132 ␮L, 1.06 mmol, 4 equiv.) were added to the tube.
The reaction mixture was heated gently with a heat gun, immersed
in an oil bath at 45 ◦C and isopropanol was added (81 ␮L, 1.06 mmol,
4 equiv.). The flask was sealed and allowed to stir for 18 h, follow-
ing which the reaction mixture was vented into a FeSO4 solution to
quench any un-reacted hydrogen cyanide. After allowing the HCN
to bubble out for 10–15 min, the solvent was removed in vacuo.
Ethyl acetate (4 × 20 mL) was added to extract the crude cyanide
adduct 14, while the catalyst 1 precipitated from the solution. The
combined supernatant solutions were concentrated to afford the
crude cyanide adduct 14 which was purified by flash column chro-
matography. The residue comprised catalyst 1 which was dried
under vacuum and used for subsequent catalytic cycles.
2.12. Homopolymer pybox 12
2.15. N-[3-(S)-cyano-butyryl]-benzamide 14
A solution of pybox monomer 10 (70 mg, 0.116 mmol, 50 equiv.)
in deoxygenated CH2Cl2 (2.5 mL) was added to a scintillation vial
(20 mL). Subsequently, a solution of Grubbs 3rd generation initia-
tor (2 mg, 0.002 mmol, 1 equiv.) in deoxygenated CH2Cl2 (2 mL)
was added to the vial. The reaction mixture was stirred for 1.5 h
at room temperature, following which 5 drops of ethyl vinyl ether
was added to quench the polymerization. The reaction mixture was
concentrated to 1 mL, the polymer was precipitated using diethyl
ether and isolated via centrifugation. The process of precipitation
and centrifugation was repeated three times to afford 64 mg (91%)
of pybox homopolymer 12 as an off-white solid. 1H NMR (300 MHz,
CDCl3) ı = 7.8 (br s, 2H, HAr(pyr)), 7.50–7.30 (br m, 10H, HAr(Ph)), 5.41
(br t, 2H, J = 9.52 Hz, CHPh), 5.40–5.10 (br m, 2H, CH = CH(pnb)), 4.88
(br t, 2H, J = 8.9 Hz, CH2(linker)), 4.37 (br t, 2H, J = 8.2 Hz, CH2(linker)),
4.15–4.04 (br m, 6H, CH2(linker), CH2(pybox)), 3.10–2.90 (overlapping
br s, 2H, CH(pnb)), 2.7–1.7 (br m, 6H, CH2(linker), CH(pnb)), 1.60–1.20
(br m, 7H, CH2(linker), CH(nb)); 13C NMR (75 MHz, CDCl3) ı = 176.7,
166.4, 163.8, 148.3, 141.9, 129.1, 128.7, 128.0, 127.1, 126.6, 113.0,
75.8, 70.4, 68.9, 66.1, 64.5, 42 (overlapping signals), 37 (overlapping
signals), 32.4, 28.9, 28.5, 25.9, 25.8, 25.7.
[1H] NMR (300 MHz, CDCl3) ı = 8.88 (s, 1H, NH), 7.89 (dd, 2H,
J = 8.0 Hz, J = 1.5 Hz, HAr), 7.64 (t, 1H, J = 7 Hz, HAr), 7.54 (t, 2H,
J = 6.3 Hz, HAr), 3.47 (dd, 1H, J = 17.2 Hz, J = 7 Hz, CHCN), 3.31 (m, 1H
CNCHCHH); 3.2 (m, 1H, NCCHCHH), 1.45 (d, 3H, J = 7.3 Hz, CH3).
Enantioselectivities were determined using HPLC with a chiral
Pirkle-l-Leucine column with 5% ethanol in hexane as the eluant
(0.7 mL min−1) and the detector set at 254 nm.
2.16. Procedure for the kinetics studies with 5 mol% loading of
catalyst 1
The addition reaction was divided into four NMR tubes. In each
tube the catalyst (3 mg, 0.002 mmol, 0.05 equiv.), imide substrate
13 (7.6 mg. 0.04 mmol, 1 equiv.) and toluene (4 ␮L) were added. The
NMR tube was sealed with a septum and TMSCN (20 ␮L, 0.16 mmol,
4 equiv.) and 2-propanol (12 ␮L, 0.16 mmol, 4 equiv.) were added.
The reaction was heated to 45 ◦C and quenched at the appro-
priate time by adding CDCl3 (200 ␮L). 1,1,2,2-Tetrachloroethane
(0.0075 mmol) in CDCl3 (100 ␮L) was added to the NMR tube as
an internal standard. The NMR tube was agitated in an ultrasound
bath and filtered through a pad of silica to remove the catalyst. The
silica was rinsed with (400 ␮L) of CDCl3 and the combined filtrates
were analyzed by 1H NMR to determine the amount of product 13
found (data shown in Fig. S1 in supporting information).
2.13. General procedure for catalytic studies with ErCl3 complex
generated in situ (Table 1)
The appropriate quantity of pybox precursor (16, 12 or 11), ErCl3
(1 equiv. with respect to pybox) and acetonitrile (250 ␮L) were
added to a Schlenk tube (50 mL) equipped with a magnetic stir-bar.

N. Madhavan et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 1–7
5
Table 1
Comparison of the activity of catalyst 11-ErCl3 with other salen(AlCl) and pybox(ErCl3) mixtures for cyanide addition.
TMSCN, iPrOH
cat (5 mol%)
O
O
CN
O
O
toluene, 45 o
18 h
C
H3C
N
Ph
H3C
N
Ph
H
H
12
13
.
No.
Cat.
Conversiona (yield)b
eec
1
2
3
1
98 (88)
98 (86)
95 (80)
80
57
51
Cycle 2
Cycle 3
a
Complex generated in situ.
Isolated yield.
Determined by HPLC analyses using a chiral Pirkle-l-leucine column.
b
c
Table 2
Catalysis and reusability studies of catalyst 1.
TMSCN, iPrOH
a
CN
O
O
O
O
cat.ErCl3
toluene, 45 oC
H3C
N
H
Ph
H3C
N
Ph
18 h
H
12
13
.
No.
Cat. (conc.)
Yieldb (ee)c
1
2
3
14 (2 mol%) + 15 (3 mol%)
14 (2 mol%) + 16 (3 mol%)
11 (5 mol%)
25 (59)
7 (72)
70 (84)
a
Determined by NMR.
Isolated.
Determined by HPLC analysis using a Chiral Pirkle-l-Leucine column.
b
c
OBn
N
1) KOH, EtOH (98%)
2) SOCl2, toluene
reflux (73%)
OH
OBn
OBn
H2 , Pd/C
EtOAc, MeOH
1) NEt3, phenyl glycinol
2) SOCl2, reflux
BnBr, K2CO3
acetone, reflux
O
O
(99%)
(90%)
(46%)
EtOOC
N
4
COOEt
EtOOC
N
5
COOEt
ClOC
N
6
COCl
NH
HN
Cl
Cl
Ph
7
Ph
O
, [Ru], CH2Cl2
O
O
OR
m
O
n
2
OH
Br
O
O
O
O
6
(70%)
6
9
RO
R''O
K2CO3, DMF, 60 o
(64%)
C
pathway a
pathway b
O
11: m + n = 50; m:n = 1
O
O
N
O
O
OR
N
2
ErCl3
CH3CN
NH
HN
Cl
Cl
N
N
[Ru], CH2Cl2
3
m
n
Ph
8
Ph
10
O
(99%)
(92%)
Ph
Ph
O
O
6
RO
R'O
R =
[Ru] =
R'' =
O
R' = R''.ErCl3
Mes N N Mes
Cl
1: m + n = 20; m:n = 1
N
N
O
N
Cl
Ru
N
O
Ph
Al
Cl
N
O
Br
N
N
6
Ph
Ph
Br
Scheme 1. Synthesis of poly(norbornene)-supported salen(AlCl) and pybox(ErCl3) catalysts.
OX
16: X =
O
N
N
O
O
Al
Cl
O
O
6
N
O
O
N
N
O
6
Ph
Ph
50
15: X = Bn
14
50
Fig. 2. Catalysts used for the cyanide addition.

6
N. Madhavan et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 1–7
3. Results and discussion
ing its interaction with the Al-center in the transition. The effect
worsened when the pybox is also supported on a polymer and is
alleviated in entry 3 where both catalysts are on the same polymer
backbone. This result was intriguing from a mechanistic point of
view despite the fact that the catalytic system derived from poly-
mer 11 was less active than our previously reported salen(AlCl)
homopolymer [8]. Therefore, we wished to optimize the perfor-
mance of the dual catalyst system by synthesizing a more soluble
variant of polymer 11.
Since we attributed the lower solubility of polymer 11 to
the presence of free pybox, we metallated the ligand prior to
polymerization (Scheme 1, pathway b). The metallated monomer
3 was copolymerized with 2 in a 9:9:1 ratio (3:2:initiator) to
afford polymer 1 which was soluble in organic solvents such as
dichloromethane. Characterization by GPC was not possible since
the highly charged, metal complex-containing polymers did not
elute from the GPC columns. Unfortunately, the paramagnetism of
Er prevented characterization by NMR spectroscopy. Therefore, ele-
mental analysis was used to characterize polymer 1. As before the
polymer composition was approximately calculated based on the
monomer/initiator ratio used for the polymerization.
The catalytic activity of polymer 1 was studied for the above
described cyanide addition using the reaction conditions described
above and the results are shown in Table 2. A significant improve-
ment in activity was observed with 98% conversion (88% isolated
yield) for the formation of 14 (entry 1) with comparable enan-
tioselectivities to 11. Furthermore, the product could be isolated
by extraction into ethyl acetate and decantation. The catalyst
remained in the reaction flask and could be used for 3 cycles with no
significant losses in conversion (entries 2 and 3). A drop in ee was
observed indicating a loss in catalytic selectivity with each subse-
quent cycle. Since this loss in enantioselectivity was not observed
with the salen(AlCl) homopolymer reported earlier by our group,
the loss in enantioselectivity might be attributed to the degradation
of the chiral pybox ligand [8]. Despite the loss in enantioselectivity
the fact that we were able to synthesize a reusable complex two-
catalyst system supported onto poly(norbornene) which efficiently
catalyzed the cyanide addition reaction is extremely encouraging
[19].
The synthesis of catalyst 1 can be envisioned by copolymerizing
the R-salen monomer 2 and the R-pybox monomer 3. Linkers com-
prised of seven carbon atoms were chosen for both monomers as
we have demonstrated before that this is the preferred length for
facilitating bimetallic reactions using poly(norbornene)-supported
salen catalysts [5,9]. The R-salen monomer 2 was synthesized as
described earlier [8]. The pybox monomer 3 was synthesized in
several steps from the diester 4 as described in Scheme 1 [15].
The hydroxyl group of 4 is reacted with benzyl bromide and potas-
sium carbonate to afford the ether 5 in 90% yield. The diesters in
5 are then hydrolyzed using potassium hydroxide and ethanol and
subsequently treated under reflux with thionyl chloride to form
the diacid chloride 6 in 73% yield [14]. Acid chloride 6 is then
treated with R-phenyl glycinol and triethyl amine at room tem-
perature for 24 h, followed by the reaction with thionyl chloride
to afford the open form 7 of pybox in 46% yield [16]. The benzyl
group is removed via hydrogenation using palladium on carbon
to afford 8 in 99% yield. Treating 8 with norbornene bromide 9
and potassium carbonate leads to etherification of the hydroxyl
position as well as the ring-closing of the pybox in one-pot to
yield the non-metallated monomer 10 in 64% yield. Since, the Er-
pybox complex was generated in situ during the catalytic reaction
in an earlier report [12], the non-metallated pybox monomer 10
was polymerized with salen monomer 2 using Grubbs’ 3rd gen-
eration initiator to provide random co-polymer 11 in 70% yield
(Scheme 1, pathway a). The resulting co-polymer was not fully
soluble in organic solvents presumably due to the co-ordination
of Al from the salen to some of the free pybox ligands resulting
potentially in crosslinks [17]. Crosslinking of the polymer through
co-ordination of Ru from the initiator to multiple pybox ligands
was ruled out since the homopolymerization of 10 gave the highly
soluble pybox homopolymer 12 which could be readily character-
ized by NMR spectroscopy. The insolubility of the co-polymer 11
prevented its characterization by NMR spectroscopy as well as gel-
permeation chromatography. Therefore, the polymer composition
(m + n and m:n) is based approximately on the monomer/initiator
ratios used during polymerization. Despite the limited solubility
of the copolymer 11 and the incapability to completely charac-
terize 11 we were interested in examining its catalytic activity
before optimizing the synthesis of the catalytic system. Hence, the
catalytic activity of 11 was studied for the cyanide addition to ␣,␤-
unsaturated imide 13 for the synthesis of cyanide adduct 14. The
erbium complex was generated in the reaction flask by stirring 11
with ErCl3 in acetonitrile and removing the solvents prior to addi-
tion of the reagents [12]. The catalytic reaction was carried out in
toluene at 45 ◦C with 5 mol% of the catalyst and the cyanide was
generated in situ from trimethyl silyl cyanide and isopropanol [18].
The substrate was added and the reaction flask was sealed to pre-
vent cyanide leakage which prevented monitoring of the reaction
by thin layer chromatography. Hence, the reaction was carried out
for 18 h to ensure completion. In order to gauge the effect of hav-
ing the two catalysts on the same polymer backbone on catalysis
a comparison was made with dual catalyst systems obtained by
combining salen homopolymer 15 either with the small molecule
pybox 16 or the pybox homopolymer 12, respectively (Fig. 2).
The catalyst studies (Table 1) indicated that the dual catalyst
systems in entries 1 and 2, afforded lower yields and enantiose-
lectivities for the cyanide adduct 14. The best yield of 70% with an
enantioselectivity of 84% was obtained when both catalysts were
supported on the same polymer backbone (entry 3). The transition
state for the reaction is hypothesized to be bimetallic with the imide
co-ordinated to the salen(AlCl) and the cyanide co-ordinated to the
pybox [12]. Hence, having the salen(AlCl) on a polymer backbone
(entry 1) might lower the accessibility of pybox 16 thereby lower-
4. Conclusions
In summary, we have supported salen(AlCl) and pybox(ErCl3)
onto a single poly(norbornene) backbone in a random fashion.
The resulting catalyst system was able to efficiently catalyze the
addition of cyanide to ␣,␤-unsaturated imides. Supporting both
catalysts on the same polymer backbone enhanced the formation of
the cyanide adduct as compared to catalyst systems derived from a
mixture of the two catalysts. Furthermore, the random co-polymer
1 could be used for 3 cycles without significant loss in conver-
sions and yields. A loss in enantioselectivity of the cyanide adduct
was observed indicating a decrease in the catalytic selectivity over
multiple cycles. The fact that we were able to readily support two
catalysts onto the same polymer backbone and demonstrate its util-
ity for catalysis paves the way for developing more efficient two
catalyst systems or multiple catalyst systems that can be utilized
for catalyzing multiple reactions or tandem reactions in a one-pot
setting.
Acknowledgement
This research is supported by the U.S. DOE Office of Basic
Energy Sciences through the Catalysis Science Contract No. DE-
FG02-03ER15459.

N. Madhavan et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 1–7
7
Appendix A. Supplementary data
[8] N. Madhavan, M. Weck, Adv. Synth. Catal. 350 (2008) 419.
[9] N. Madhavan, T. Takatani, C.D. Sherrill, M. Weck, Chem. Eur. J. 15 (2009) 1186.
[10] T.L. Choi, R.H. Grubbs, Angew. Chem. Int. Ed. 42 (2003) 1743.
[11] J.M. Pollino, L.P. Stubbs, M. Weck, Macromolecules 36 (2003) 2230.
[12] G.M. Sammis, H. Danjo, E.N. Jacobsen, J. Am. Chem. Soc. 126 (2004) 9928.
[13] G.M. Sammis, E.N. Jacobsen, J. Am. Chem. Soc. 125 (2003) 4442.
[14] J. Recker, W.M. Müller, U. Müller, T. Kubota, Y. Okamoto, M. Nieger, F. Vögtle,
Chem. Eur. J. 8 (2002) 4434.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molcata.2010.10.023.
References
[15] P. Froidevaux, J.M. Harrowfield, A.N. Sobolev, Inorg. Chem. 39 (2000) 4678.
[16] H. Nishiyama, S. Yamaguchi, M. Kondo, K. Itoh, J. Org. Chem. 57 (1992) 4306.
[17] The insolubility of the catalyst made it extremely difficult to characterize it by
NMR spectroscopy or Mass spectrometry.
[18] Trimethyl silyl cyanide and hydrogen cyanide are extremely toxic and should
be handled carefully in a well-ventilated fume hood.
[1] N. Madhavan, C.W. Jones, M. Weck, Acc. Chem. Res. 41 (2008) 1153.
[2] N.E. Leadbeater, M. Marco, Chem. Rev. 102 (2002) 3217.
[3] C. Baleizão, H. Garcia, Chem. Rev. 106 (2006) 3987.
[4] L. Canali, D.C. Sherrington, Chem. Soc. Rev. 28 (1999) 85.
[5] X. Zheng, C.W. Jones, M. Weck, Adv. Synth. Catal. 350 (2008) 255.
[6] X. Zheng, C.W. Jones, M. Weck, J. Am. Chem. Soc. 129 (2007) 1105.
[7] M. Holbach, M. Weck, J. Org. Chem. 71 (2006) 1825.
[19] The kinetics for the cyanide addition catalyzed by 1 was studied using NMR
spectroscopy. Data shown in the supplemental information.