Polysiloxane-supported NAD(P)H model 1-benzyl-1,4-dihydronicotinamide: Synthesis and application in the reduction of activated olefins

Article abstract of DOI:10.1021/jo020319x

A new polysiloxane-supported NAD(P)H model, 1-benzyl-1,4-dihydronicotinamide, was designed and synthesized, which can efficiently reduce many activated olefins under mild conditions. The most advantageous features of this new polysiloxane-supported reductant are (i) easy workup and separation of the reaction products and (ii) good potential for recycling use of the reductant, which makes this new polysiloxane-supported NAD(P)H model a promising alternative both in research laboratories and in industrial processes.

Full text of DOI:10.1021/jo020319x

used.7 In fact, these NAD(P)H model compounds them-
selves generally possess excellent reducibility, so that
they may be effectively applied in the reduction of many
unsaturated substrates.8
P olysiloxa n e-Su p p or ted NAD(P )H Mod el
1-Ben zyl-1,4-d ih yd r on icotin a m id e:
Syn th esis a n d Ap p lica tion in th e Red u ction
of Activa ted Olefin s
As green chemistry has become a major concern to
organic chemists in recent years,9 solid or solid-supported
catalysts have received much more attention, because
they may offer several advantages in preparative proce-
dures, e.g., simplifying workup and separation and
recycling of the catalyst. These features may lead to
economical automation and may effectively reduce pol-
lution of hazardous compounds, advancing to an envi-
ronmentally benign process.10 In this connection, immo-
bilizing NAD(P)H models not only provides a novel design
for a new reductant, but may also lead to an efficient
green methodology for organic synthesis. In a previous
paper,11 a polymer-bound NAD(P)H model was reported,
which consisted of NAD(P)H model 1-benzyl-1,4-dihy-
dronicotinamide and a copolymer of styrene and divinyl-
benzene (Merrifield-type resin). However, Merrifield
resin is largely affected by solvents, i.e., swelling in an
organic medium and contracting in an aqueous solvent,
which evidently retards the reduction of the NAD(P)+
model moiety inside the polymer in a basic aqueous medi-
um in the preparation of the polymer-bound NAD(P)H
model. SiO2 is an inorganic support and is little affected
by solvents. Recently, SiO2 has received much attention
in catalyst immobilization and solid-phase synthesis,12
and a few papers concerning the immobilization of a
NAD(P)H model on silica have appeared, which was
performed by grafting silica onto the nitrogen atom at
the 1-position on the pyridine ring.13 As well-known, the
Baolian Zhang,† Xiao-Qing Zhu,*,† J in-Yong Lu,†
J iaqi He,† Peng G. Wang,‡ and J in-Pei Cheng*,†
State Key Laboratory of Elemento-Organic Chemistry,
Department of Chemistry, Nankai University,
Tianjin 300071, China, and Department of Chemistry,
Wayne State University, Detroit, Michigan 48202
xqzhu@nankai.edu.cn
Received May 6, 2002
Abstr a ct: A new polysiloxane-supported NAD(P)H model,
1-benzyl-1,4-dihydronicotinamide, was designed and syn-
thesized, which can efficiently reduce many activated olefins
under mild conditions. The most advantageous features of
this new polysiloxane-supported reductant are (i) easy
workup and separation of the reaction products and (ii) good
potential for recycling use of the reductant, which makes
this new polysiloxane-supported NAD(P)H model a promis-
ing alternative both in research laboratories and in indus-
trial processes.
Nicotinamide adenine dinucleotide (NADH) and its
phosphate derivative (NADPH) have long been known
to act as coenzymes in biological redox reactions. It was
established that in the reduced form of the coenzyme the
active part is 1,4-dihydropyridine.1 Thus, 1-benzyl-1,4-dihy-
dronicotinamide (BNAH), Hantzsch 1,4-dihydropyridine
(HEH), 10-methyl-9,10-dihydroacridine (AcrH2), and many
other 1,4-dihydropyridine derivatives are widely used as
models to mimic the function of NAD(P)H in biological
reductions of various unsaturated compounds.2-6 Tre-
mendous activities have been carried out to focus on the
mechanistic details of these systems. To our knowledge,
however, little attention has been paid to these models
acting as reducing agents in organic synthesis except for
a few cases where some chiral NAD(P)H models were
(6) (a) Zhu, X.-Q.; Liu, Y.-C.; Cheng, J .-P. J . Org. Chem. 1999, 64,
8980-1. (b) Zhu, X.-Q.; Liu, Y.-C.; Wang, H.-Y.; Wang W. J . Org. Chem.
1999, 64, 8982-3. (c) Zhu, X.-Q.; Liu, Y.-C. J . Org. Chem. 1998, 63,
2786. (d) Zhu, X.-Q.; Zou, H.-L.; Yang, P.-W.; Liu, Y.; Cao, L.; Cheng,
J .-P. J . Chem. Soc., Perkin Trans 2 2000, 1857-61. (e) Li, B.; Liu,
Y.-C.; Guo, Q.-X. J . Photochem. Photobiol., A 1997, 103, 101. (f) Wang,
H.-Y.; Liu, Y.-C.; Zhu, X.-Q.; Guo, Q.-X. Chin. J . Chem. 1999, 17, 1884.
(7) (a) Ohno, A.; Tsutsumi, A.; Kawai, Y.; Yamazaki, N.; Kikata,
Y.; Okamura, M. J . Am. Chem. Soc. 1994, 116, 8133. (b) Skog, K.;
Wennerstrom, O. Tetrahedron Lett. 1992, 33, 1751. (c) Kanomata, N.;
Nakata, T. Angew. Chem., Int. Ed. Engl. 1997, 36, 1207-11. (d)
Combret, Y.; Duflos, J .; Dupas, G.; Bourguignon, J .; Queguiner, G.
Tetrahedron: Asymmetry 1993, 4, 1635. (e) Bedat, J .; Levacher, V.;
Dupas, G.; Queguiner, G.; Bourguignon, J . Chem. Lett. 1995, 327. (f)
De Kok, P, M. T.; Bastiaansen, L. A. M.; Van Lier, P. M.; Vekemans,
J . A. J . M.; Buck, H. M. J . Org. Chem. 1989, 54, 1313.
(8) Zhu, X.-Q.; Wang, H.-Y.; Wang, J .-S.; and Liu, Y.-C. J . Org.
Chem. 2001, 66, 344-7.
(9) (a) Anastas, P. T.; Warner, J . C. Green Chemistry; Oxford
University Press: Oxford, 2000; pp 14-99. (b) Sauer, N. N. Green
Chemistry: Frontiers in Benign Chemical Syntheses and Processes;
Oxford University Press: New York, 1999.
(10) (a) Seeberger, P. H.; Danishefsky, S. J . Acc. Chem. Res. 1998,
31 (10), 685-95. (b) Sammelson, R. E.; Kurth, M. J . Chem. Rev. 2001,
101 (1), 137-202. (c) Lorsbach, B. A.; Kurth, M. J . Chem. Rev. 1999,
99 (6), 1549-82.
(11) (a) Dupas, G.; Bourguignon, J .; Ruffin, C.; Queguiner, G..
Tetrahedron Lett. 1982, 23, 5141-5144. (a) Vitry, C.; Vasse, J .-L.;
Dupas, G.; Levacher, V.; Queguiner, G.; Bourguignon, J . Tetrahedron
2001, 57, 3098.
† Nankai University.
‡ Wayne State University.
(1) Murakami, Y.; Kikuchi, J .; Hisaeda, Y. and Hayashida, O. Chem.
Rev. 1996, 96, 721.
(2) (a) Fukuzumi, S.; Yorsisue, T. Bull. Chem. Soc. J pn. 1992, 65,
715-9. (b) Coleman, C, A.; Rose, J . G.; Murray, C. J . J . Am. Chem.
Soc. 1992, 114, 9755-62. (c) Fukuzumi, S.; Nishizawa, N.; Tanaka, T.
J . Org. Chem. 1984, 49, 3571-8.
(3) (a) Beijer, N. A.; Vekemans, J . A. J . M.; Buck, H. Recl. Trav.
Chim. Pays-Bas 1990, 109, 434-6. (b) Fukuzumi, S.; Mochizuki, S.
Tanaka, T. J . Am. Chem. Soc. 1989, 111, 1497-9. (c) Tanner, D. D.;
Singh, H. K.; Kharrat, A.; Stain, A. R. J . Org. Chem. 1987, 52, 2141.
(d) Tanner, D. D.; Stain, A. R. J . Org. Chem. 1988, 53, 1642.
(4) (a) Kanomata, N.; Suzuki, M.; Yoshida, M.; Nakata, T. Angew.
Chem., Int. Ed. 1998, 37, 1410-2. (b) Fukuzumi, S.; Ishikama, M.;
Tanaka, T. Tetrahedron 1984, 42, 1021-34. (c) Singh, S.; Sharma, V.
K. Tetrahedron Lett. 1979, 29, 2733-4. (d) Merjer, H. P.; Van Niel, J .
C. G.; Pandit, U. K. Tetrahedron 1984, 40, 5185. (e) De Nie-Sarink,
M. J .; Pandit, U. K. Tetrahedron Lett. 1979, 26, 2449-52.
(5) (a) Lu, Y.; Liu, B.; Cheng, J .-P. Chem. J . Chin. Univ. 1997, 18,
391. (b) Merjer, H. P.; Van Niel, J . C. G.; Pandit, U. K. Tetrahedron
1984, 40, 5185. (c) De Nie-Sarink, M. J .; Pandit, U. K. Tetrahedron
Lett. 1979, 26, 2449-2452. (d) Singh, S.; Sharma, V. K. Tetrahedron
Lett. 1979, 29, 2733-4.
(12) (a) Nakamura, K.; Fujii, M.; Ohno, A.; Oka, S. Tetrahedron Lett.
1984, 25 (36), 3983. (b) Nakamura, K.; Fujii, M.; Oka, S.; Ohno, A.
Chem. Lett. 1985, 523. (c) Yasui, S.; Fujii, M.; Nakamura, K.; Ohno,
A. Bull. Chem. Soc. J pn. 1987, 60, 963-8. (d) Fujii, M.; Nakamura,
K.; Yasui, S.; Ohno, A. Bull. Chem. Soc. J pn. 1987, 60, 2423-7. (e)
Yasui, S.; Fujii, M.; Ohno, A. Bull. Chem. Soc. J pn. 1987, 60, 4019-
26.
10.1021/jo020319x CCC: $25.00 © 2003 American Chemical Society
Published on Web 03/25/2003
J . Org. Chem. 2003, 68, 3295-3298
3295

SCHEME 1
reactivity of NAD(P)H models is strongly dependent on
the size of the substituent at the 1-position on the
pyridine ring; silica grafted onto the nitrogen atom at
the 1-position, of course, must make the reactivity of
NAD(P)H models quite low. In addition, polysiloxane has
been confirmed to be a much better inorganic support
than silica and has a higher content of functional
groups,14 which inspired us to graft polysiloxane onto the
nitrogen atom in the CONH2 group of the nicotinamide.
We report here the synthesis of polysiloxane-immobilized
1-benzyl-1,4-dihydronicotinamide (polysiloxane-BNAH,
1) and its successful application in the reduction of some
activated olefins.
The synthetic route of polysiloxane-BNAH is shown
in Scheme 1. Aminopolysiloxane14 was treated with
nicotinic anhydride to give the corresponding polysilox-
ane-grafted nicotinamide, which was then refluxed with
benzyl bromide in acetonitrile to give polysiloxane-grafted
1-benzylnicotinamide cation. The formed intermediate
cation was reduced by sodium dithionite in basic aqueous
solvent to offer polysiloxane-BNAH, with 0.75 mmol of
1,4-dihydropyridine/g of polysiloxane-BNAH.15
To find the applicability of this new polysiloxane-
supported NAD(P)H model as a reductant in organic
synthesis, many activated olefins (2) and some allylic and
benzylic bromides (6 and 7) were examined. When the
olefins 2a -2n were treated with 1 in dry acetonitrile at
room temperature, the reaction results showed that the
double bond in 2 was efficiently reduced. If the substrate
6 was treated with 1 in dry acetonitrile at room temper-
ature, cyclopropane derivative 8 and its open-chain
compound 9 as the reduction products were obtained, but
if the substrate 7 was used instead of 6 to react with 1,
only indan derivative 9 was obtained. The detailed results
are summarized in Tables 1 and 2.
activated by electron-withdrawing groups such as CN and
CO2Et. When the olefin is benzylidenemalononitrile and
ethyl R-cyanocinnamate and their derivatives 2a -2n , the
reactions give the corresponding double-bond-reduced
products 4 in excellent yield, suggesting a potential
synthetic application of the new reductant for activated
olefins under mild conditions. According to the reaction
time and product yields, it is clear that benzylidene-
malononitrile and its derivatives are more efficient than
the corresponding ethyl R-cyanocinnamate and its de-
rivatives in the reductions, which indicates the develop-
ing negative charge on the active center of the substrate
during the reduction processes, since the power of CN to
attract electrons (Hammett substituent parameter σ )
0.66)16 is larger than that of CO2Et (σ ) 0.45),16 which
makes the kinetic barrier to attack the double bond easier
to overcome. For the same reason, it is quite understand-
able that no reaction was observed with olefin 2o,
because NMe2 is a very strong electron-donating group
(σ ) -0.83).16 A tentative reaction mechanism of 2 with
1 may be proposed as shown in Scheme 2: a hydride
departed from the dihydropyridines first attached to the
â-carbon of 2 to form a carbanion intermediate, which
then attracted a proton to form the final reduction
product 4. It is worth noting that when 9-fluorenylidene-
malononitrile and ethyl R-cyano-9-fluorenylideneacetate
(2p and 2q) are used as substrates, two products are
formed; one is the double-bond-reduced products 4 and
the other is hydrolytic decomposition product ketones 5.
From the yield ratio of 4 to 5, it is clear that reducing
the electron-withdrawing power of X favors the genera-
tion of 5. Following the same trend, it should be expected
that 4 should be further disfavored if X is a hydrogen
atom or one R group is an electron-donating group. In-
deed, this is exactly what we observed with olefins
2r -2w . These reactions went very slowly, and the
product was only ketones 5. To probe the formation route
of 5, free BNAH was used instead of 1 to react with
2r -2w ; no 5 was observed for 3 days, which ruled out
the possibility that 5 formed from the direct reaction of
olefin with 1. But if the free polysiloxane was used
instead of 1 to react with olefin 2s, 5s was obtained also,
which indicates that 5 must come from a slow reaction
of olefin with the trace of water contained in polysilox-
ane.17 This hydrolytic reaction, of course, should be
As shown in Table 1, the new polysiloxane-supported
NAD(P)H model can effectively reduce olefins which are
(13) (a) Georges, D.; Patrick, T.; Thierry, T.; J acques, C.; Denis, L.;
J ean, B.; Guy, Q. New J . Chem. 1989, 13, 255-8. (b) J ean, B.; Patrick,
T.; Thierry, T.; Denis, L.; Georges, D.; Guy, Q. Bull. Soc. Chim. Fr.
1989, 2, 264-6.
(14) (a) EI-Nahhal, I. M.; Parish, R. V. J . Organomet. Chem. 1993,
452. 19-22. (b) Khatib, I. S.; Parish, R. V. J . Organomet. Chem. 1989,
369. 9-16.
(15) The content of the 1,4-dihydropyridine units was estimated by
the reaction with N,N,N′,N′-tetramethyl-p-phenylenediamine radical
cation perchlorate in anhydrous acetonitrile. 1,4-Dihydropyridine can
be rapidly oxidized by N,N,N′,N′-tetramethyl-p-phenylenediamine
radical cation perchlorate, so that the content of the 1,4-dihydropyri-
dine can be determined by the decrease in the absorption at λ )
564 nm.
(16) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165-95.
(17) Rechavi, D.; Lemaire, M. Org. Lett. 2001, 3, 2493-6.
3296 J . Org. Chem., Vol. 68, No. 8, 2003

TABLE 1. Rea ction s of 1 w ith Activa ted Olefin s 2 in Dr y Aceton itr ilea
activated olefins
compd
R′
R′′
X
T (°C)
time (h)
products
yieldb (%)
2a
2b
2c
2d
2e
2f
2g
2h
2i
2j
2k
2l
2m
2n
2o
2p
2q
2r
Ph
m-BrPh
m-FPh
m-ClPh
p-CF3Ph
p-CH3Ph
p-OCH3Ph
Ph
p-CF3Ph
m-FPh
p-CH3Ph
p-ClPh
m-BrPh
p-OCH3Ph
p-N(CH3)2Ph
H
CN
CN
CN
CN
CN
CN
CN
CO2Et
CO2Et
CO2Et
CO2Et
CO2Et
CO2Et
CO2Et
CO2Et
CN
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
rt
1
1
1
1
1
3
3
5
5
5
5
3
5
4a
4b
4c
4d
4e
4f
4g
4h
4i
4j
4k
4l
4m
4n
92
95
95
93
96
93
65
80
90
95
90
95
93
90
H
H
H
H
H
H
H
H
H
H
H
H
H
H
4
48
12
12
50
24
24
24
24
24
no reaction
9-fluorenylidene
9-fluorenylidene
9-fluorenylidene
4p (65) + 5p (5)c
CO2Et
H
CN
4q (40) + 5q (19)c
5r
5s
5t
5u
5v
5w
17
17
15
11
14
15
2s
2t
2u
2v
2w
Ph
Ph
Ph
CH3
Ph
CH3
c-Pr
CH3
Ph
CN
CO2Et
CO2Et
CONH2
CH3
a
b
All reactions were carried out as described in the Experimental Section; olefin:1 ) 1:1.5 (mol/mol). Yields of isolated products after
chromatographic purification. c Determined by capillary GC analysis of the crude reaction mixture after filtration.
TABLE 2. Cycliza tion s of Som e Allylic a n d Ben zylic Br om id es by 1 in CH 3CN u n d er a n Ar gon Atm osp h er ea
a
b
All reactions were carried out as described in the Experimental Section. Molar ratio of 1 to the substrate. c Yields of isolated products.
largely depressed when a more activated olefin is used,
because the reduction reaction runs much faster.
substrate is o-bromomethylbenzylidenemalononitrile (7a )
and its analogue 7b, only indan structure compounds 9
without open-chain compounds are obtained. Concerning
the cyclization mechanisms of the substrates 6 and 7, it
is reasonable to propose that the reactions take place via
a hydride-transfer mechanism, similar to the proposal
in our previous paper.8 Thus, a hydride ion from 1 added
to the benzyl position of the substrates and the resulting
In Table 2, it is interesting to find that when the
substrates are 2-bromo-1-phenylethylidenemalononitrile
(6a ) and its analogue 6b, the reduction products are
cyclopropane derivatives 8a and 8b along with their
open-chain compounds 4a and 4h , and the open-chain
compounds are major products (see Table 2), but if the
J . Org. Chem, Vol. 68, No. 8, 2003 3297

SCHEME 2
SCHEME 3
SCHEME 4
SCHEME 5. Cyclic P a th w a y for th e Sod iu m
Dith ion ite-Med ia ted Red u ction of Olefin s w ith
P olysiloxa n e-BNAH
carbanions undergo (i) in the cases of 6a and 6b,
intramolecular displacement of the â-bromomethyl group
to produce cyclopropane derivatives 8a and 8b, which
then are further reduced by 1 to become the final open-
chain product 9 as shown in Scheme 3 and (ii) in the
cases of 7a and 7b, intramolecular nucleophilic substitu-
tion of the o-bromomethyl group to produce indan deriva-
tives 9 after initial hydride transfer. It is worth noting
here that since the stability of five-membered ring
compounds 9 is larger that that of three-membered ring
compounds 8, the indan derivatives 9 cannot be further
reduced by 1 to become the open-chain product.
The solvent effect on the reactions was also examined.
As well-known, for the reactions using free NAD(P)H
models such as BNAH, a polar solvent such as acetoni-
trile is quite suitable and a small amount of protic solvent
such as methanol is also required. The reason is that
proton can neutralize the carbanion intermediate formed
from the initial formal hydride transfer to make the
reaction go to the right and to prevent the reverse
reaction to the left (Scheme 2). But for the reactions with
polysiloxane-grafted NAD(P)H model 1, a change of the
solvent system to, e.g., CH2Cl2, 1,4-dioxane, THF, ben-
zene, etc. did not cause a remarkable difference in the
product yields. It is worth noting that, unlike the reac-
tions with regular models, in the reactions with polysi-
loxane-BNAH, no protic medium was needed. The main
reason could be that the free silanol groups on silica can
offer protons to neutralize the formed carbanion and thus
drive the reduction to completion (Scheme 4).17
which retains the good reducing ability of NAD(P)H
model compounds and also possesses several advantages
of inorganic solid-phase synthesis so that it can be
effectively used to reduce many activated olefins. The
easy workup and separation of the products and the good
potential for recycling use of the reagent make this new
reductant a promising alternative both in research
laboratories and in industrial processes.
The most appealing advantage of polysiloxane-BNAH
as a reductant is its recycling use. To test this point, the
polysiloxane-BNA+ formed from the reaction of polysi-
loxane-BNAH with olefins 2 was reduced with sodium
dithionite in a basic medium, and then was treated with
the substrate again. It showed no obvious decrease in
reactivity after the reductant was reused three times. The
recycling operation of polysiloxane-BNAH is demon-
strated in Scheme 5.
Ack n ow led gm en t. These studies were supported by
the National Outstanding Youth Fund (Grant Nos.
20125206 and 29928004) and the Natural Science
Foundation of China (Grant Nos. 20272027 and
29972028), which are gratefully acknowledged.
Su p p or tin g In for m a tion Ava ila ble: Experimental details
for the preparation of Polysiloxane-BNAH and characterization
data of the representative products. This material is available
free of charge via the Internet at http://pubs.acs.org.
In summary, we present here the first covalent het-
erogenization of the NAD(P)H model on polysiloxane,
J O020319X
3298 J . Org. Chem., Vol. 68, No. 8, 2003