Thermal and photoinduced reduction of some benzyl bromides by an NAD(P)H model: The effect of electron withdrawing groups on mechanism and reactivity

Article abstract of DOI:10.1016/S0040-4039(02)01825-7

A series of compounds, o-bromomethylbenzylidene-malononitrile (1), dimethyl o-bromomethyl-benzylidenemalonate (2) and methyl α-cyano-o-bromomethylcinnamate (3) were reduced in the dark and under irradiation by an NAD(P)H model, 1-benzyl-1,4-dihydronicotin

Full text of DOI:10.1016/S0040-4039(02)01825-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 7527–7530
Thermal and photoinduced reduction of some benzyl bromides by
an NAD(P)H model: the effect of electron withdrawing groups
on mechanism and reactivity
Hongyi Wang,a,b,* Danmei Dai,a,b Youcheng Liua,b,* and Qingxiang Guoa,b
aDepartment of Chemistry, University of Science and Technology of China, Hefei, 230026, PR China
bNational Laboratory of Applied Organic Chemistry and Department of Chemistry, Lanzhou University,
Lanzhou 730000, PR China
Received 17 July 2002; accepted 23 August 2002
Abstract—A series of compounds, o-bromomethylbenzylidene-malononitrile (1), dimethyl o-bromomethyl-benzylidenemalonate
(2) and methyl a-cyano-o-bromomethylcinnamate (3) were reduced in the dark and under irradiation by an NAD(P)H model,
1-benzyl-1,4-dihydronicotinamide (BNAH). Two different mechanisms were found, i.e. one-step hydride transfer (polar pathway)
and multi-step sequence initiated by single electron transfer (SET pathway). The effect of electron-withdrawing groups on the
reactivity of the substrates were discussed in terms of Hammett substituent constants, 13C NMR chemical shift values and cyclic
voltammetric redox potentials, and their correlations. © 2002 Elsevier Science Ltd. All rights reserved.
As an important redox coenzyme in biological reac-
tions, NAD(P)H has attracted much interest both in its
function and mechanism. Although it is believed that in
real biological systems electron transfer from
NAD(P)H to substrates (aldehyde, ketone, imine or
ester) is rarely seen, NAD(P)H and its models have
been reacted with various unsaturated substrates and
the reaction mechanism has been extensively studied by
chemists.1 An issue of controversy is whether the for-
mal hydride transfer from NAD(P)H to the substrates
occurs in one-step hydride transfer (polar pathway) or
in a multi-step sequence involving electron transfer as
the initial step (SET pathway).2 Numerous substrates
with different features have been used as probes to
react with the NAD(P)H model, 1-benzyl-1,4-dihydro-
nicotinamide (BNAH) to distinguish the mechanisms
and reveal the structure–reactivity relationship.3
Our group has used some probes with carbonꢀcarbon
double bonds activated by electron withdrawing groups
(EWGs) and both pathways were found operating in
the mechanism, depending on the substrates, NAD(P)H
models, and reaction conditions.4 We have also studied
EWG activated benzyl bromides (by one or two nitro
groups in the phenyl ring) and found SET pathway
dominating the reaction.5 Either thermal (dark) or pho-
toinduced conditions were used in these studies.
In order to gain more insight into the effect of EWGs
on the substrate reactivity and reaction mechanism we
have designed a series of substrates combining both
activated double bond and benzylic bromide in the
same molecule. They are o-bromomethylbenzylidene-
malononitrile (1), dimethyl o-bromomethylbenzylidene-
malonate (2) and methyl a-cyano-o-bromomethyl-
cinnamate (3), prepared by NBS bromination of corre-
sponding toluenes, which were prepared from o-methyl-
benzaldehyde and corresponding CꢀH acids by
Knoevenagel condensation. This design would make it
possible to investigate several aspects: first, the selectiv-
ity of BNAH toward the two functional groups, i.e.
Scheme 1.
* Corresponding authors. Present address of H.W.: Chemistry
Department, University of Florida, Gainesville, FL 32611, USA;
e-mail: hwang@chem.ufl.edu
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01825-7

7528
H. Wang et al. / Tetrahedron Letters 43 (2002) 7527–7530
activated double bond and benzylic bromide; second,
the reaction product would indicate the reaction mech-
anism, i.e. double bond reduction–intramolecular
cyclization would indicate polar pathway while
debromination would indicate SET pathway;5,6 third,
the combination patterns of EWGs (2CN, CN–CO2Me
and CO2Me) would serve as a good model to study the
effect of EWGs on reactivity. With this in mind, We
studied their reactions with BNAH under both thermal
and photoinduced conditions (Scheme 1).
give the dimer 9, strongly supporting this mechanism
(Scheme 3).
From the results described above it is clearly seen that
the dark reactions took ‘polar’ pathway and the pho-
toinduced reactions took the ‘SET’ pathway. This actu-
ally reflects the selectivity of the two functional groups,
i.e. activated double bond and benzylic bromide. Also
the results showed the successful distinguishing of the
reaction pathways as the design. Compound 1 favors
the polar pathway even under photoinduced conditions,
indicating its reaction rate with ground state BNAH is
much faster than that with excited BNAH*. Compound
3 however, is inert to ground state BNAH and readily
reacts with excited BNAH*. Compound 2 lies between
these two cases. It undergoes BNAH reaction in the
dark. Under photoinduced conditions, it competes with
ground state BNAH and excited BNAH* in two differ-
ent pathways at comparable rates. In conclusion the
three compounds have the reactivity sequence as 1>2>
3. These results are apparently due to the different
combinations of EWGs the compounds carry (2CN,
CN–CO2Me and 2CO2Me for 1, 2 and 3 respectively).
The fact that the other portion of these substrates is
exactly the same makes it possible to ‘separately’ study
this ‘net’ EWG effect.
Compound 1, 2, or 3 (0.5 mmol) and BNAH (0.6
mmol) were reacted in 15 mL of acetonitrile at room
temperature in the dark under argon. After 4 h the
reaction was quenched and worked up. The reaction of
1 and 2 gave cyclized compound 4 and 5 in yields of 90
and 80%, respectively.7 Reaction of 3 did not give any
product at all. Even heating at 60°C for 4 h or stirring
at room temperature for a week did not bring about
any reaction. It appears likely that the BNAH reaction
takes place via hydride transfer from BNAH to 1 or 2
(polar pathway) and the resulting carbanion undergoes
intramolecular nucleophilic substitution (SNi) to give
the cyclic product (Scheme 2).
Compound 1, 2, or 3 (0.5 mmol) and BNAH (0.6
mmol) were also reacted in 15 mL of acetonitrile at
room temperature upon irradiation with a high-pres-
sure mercury lamp under argon. A piece of Pyrex glass
was placed between the reaction vessel and the lamp to
filter off UV light and give light of u>320 nm. After 6
h the reaction was worked up and products were sepa-
rated and identified. For 1, the major product was still
4 (80%), with small amount of debrominated product 6
(<10%). In the case of 2, however, cyclic product 5 and
debrominated product 7 were obtained in comparable
yields of 40 and 30%, respectively. The reaction of 3
gave debrominated product 8 and dimer 9 in yields of
30 and 40%, respectively. In the photoinduced reaction,
excited BNAH*8 (umax=354 nm) transferred an elec-
tron to the substrate, the resulting radical anion liber-
ated the bromide anion to form a benzyl radical, the
benzyl radical then abstracted a hydrogen from BNAH
to give debrominated product5,6 and the BNA. radical
transferred an electron to another molecule of substrate
to continue the chain reaction. In the case of 3 the
benzyl radical also underwent bimolecular coupling to
In linear free energy correlations the substituent con-
stants (|p) represent the electron withdrawing or donat-
Scheme 3. Photoinduced BNAH reductions of 1, 2 and 3.
Scheme 2. Thermal BNAH reductions of 1, 2 and 3.

H. Wang et al. / Tetrahedron Letters 43 (2002) 7527–7530
7529
ing properties of various substituents.9 The |p values
are 0.70 and 0.44 for CN and CO2Me, respectively.
Since the effect of the two EWGs in the same molecule
work in the same direction (actually the effect would be
a little different for the two EWGs because one is cis-
to phenyl group while the other is trans-) their combi-
native electron-withdrawing abilities could be approxi-
mately estimated as 2CN (1.40)>CN–CO2Me
(1.14)>2CO2Me (0.88), which is consistent with the
reactivity sequence of them.
voltammograms were measured on a CV-27 Volta-
mograph using Ag/AgCl electrode as reference with
tetrabutylammonium perchlorate as supporting elec-
trolyte at a scan rate of 250 mV/s. The reductive
potentials (Ep) of 1, 2 and 3 are −0.40, −0.44 and −0.48
V, respectively, again in the sequence of 2CN>CN–
CO2Me>2CO2Me, which parallels the reactivity
sequence.
The three sets of data, lC, |p and Ep were correlated by
plotting them in Excel, giving curves close to straight
lines (Fig. 1).
The 13C NMR chemical shift values of the electro-posi-
tive b-ethylenic carbons of the substrates are a good
measure of the electron deficiency nature of them
caused by the EWGs.10 The determined values (lC) are
156.98, 151.95, and 140.61 ppm for 1, 2 and 3, respec-
tively, holding the sequence 2CN>CN–CO2Me>
2CO2Me, which again is in good accordance with the
reactivity sequence.
There have been some reports on the correlations of
UV, IR, NMR, kinetic data, redox potential, etc., with
Hammett constants to reveal the relative intrinsic reac-
tivity of related compounds.11 However, most of them
had the substituents in the aromatic ring (or on the
alkyl groups in the alkylnicotinamide models11b). The
present study extends the substituents further to outside
of the phenyl ring along conjugate system and it was
proven successful. This correlation method will find
more applications in the similar systems and be of value
in better understanding and controlling of chemical
reactivity.
In order to obtain insight into the EWG effect on the
relative redox reactivity of 1, 2 and 3 their cyclic
Acknowledgements
We would like to thank the financial support by the
National Science Foundation of China (Grant No.
20072036 and 29832040).
References
1. Murakami, Y.; Kikuchi, K; Hisada, Y.; Hayashida, O.
Chem. Rev. 1996, 96, 721.
2. Stout, D. M. Chem. Rev. 1982, 223.
3. Yasui, S.; Ohno, A. Bioorg. Chem. 1986, 14, 70.
4. (a) Liu, Y.; Zhu, X.; Wu, L.; Guo Q.; Li, B.; Wang, D.;
Wu, Y. Book of Abstracts for 5th Eurasia Conference
on Chemical Science; Zhongshan University Press:
Guangzhou, China, Dec. 10–14, 1996; p. 150; (b) Zhu,
X.; Liu, Y. J. Org. Chem. 1998, 63, 2786; (c) Zhu, X.;
Liu, Y.; Wang, H.; Wang, W. J. Org. Chem. 1999, 64,
8982; (d) Liu, Y.; Wang, H.; Yang, Q.; Mak, T. C. J.
Chem. Soc., Perkin Trans. 2 2000, 649; (e) Jiang, H.; Liu,
Y.; Li, J.; Wang, G.; Wu, Y.; Wang, Q.; Mak, T. C.
Chem. Commun. 2002, 882.
5. Presented at Gomberg 2000: a century of organic free
radical chemistry by the International Society of Organic
free radicals, University of Michigan, Ann Arbor June
25–29, 2000. Mono- and dinitrobenzyl bromides were
reduced by BNAH upon irradiation and in the dark,
respectively. Spin-trapping and cyclic voltammetry results
support SET mechanism. Mononitrobenzyl bromides are
inert in the dark.
6. Fukuzumi, S.; Hironaka, K.; Tanaka, T. J. Am. Chem.
Soc. 1987, 105, 4722 and references cited therein.
7. Selected physical and spectroscopic data: 4: white solid,
mp 130–131°C. 1H NMR (400 MHz, CDCl3) l (ppm):
Figure 1.

7530
H. Wang et al. / Tetrahedron Letters 43 (2002) 7527–7530
3.70 (s, 4H), 7.20 (m, 4H). EI-MS m/z: 168 (100), 141 (91),
114 (7). Anal. calcd for C11H8N2: C, 78.55; H, 4.79; N,
10. (a) Abraham, R. J.; Loftus, P. Proton and Carbon-13
NMR Spectroscopy, An Integrated Approach; John &
Wiley Sons: Chichester, 1983; p. 24; (b) Barchiesi, E.;
Bradamante, S.; Ferraccioli, R.; Pagani, G. A. J. Chem.
Soc., Perkin Trans. 2 1990, 375.
11. (a) Coene, B.; Schanck, A.; Dereppe, J.-M.; Meerssche, M.
V. J. Med. Chem. 1984, 27, 694; (b) Burke, J. R.; Frey, P.
A. J. Org. Chem. 1996, 61, 530; (c) Neuvonen, H.;
Neuvonen, K. J. Chem. Soc., Perkin Trans. 2 1999, 1497;
(d) Bender, T. P.; Graham, J. F.; Duff, J. M. Chem. Mater.
2001, 13, 4105; (e) Pihlaja, K.; Ovcharenko, V.; Koleh-
mainen, E.; Laihia, K.; Fabian, W. M. F.; Dehne, H.;
Perjessy, A.; Kleist, M.; Teller, J.; Sustekova, Z. J. Chem.
Soc., Perkin Trans. 2 2002, 329.
16.66. Found: C, 78.60; H, 4.80; N, 16.61. 9: white solid,
1
mp 85–86°C. H NMR (400 MHz, CDCl3) l (ppm): 2.95
(s, 4H), 3.71 (s, 6H), 3.86 (s, 6H), 7.26 (m, 8H), 7.99 (s,
2H). EI-MS m/z: 466 (0.1), 434 (1), 311 (9), 274 (16), 255
(9), 233 (11), 143 (35), 115 (98), 59 (100). Anal. calcd for
C26H26NO8: C, 66.94; H, 5.62. Found: C, 67.02, H, 5.68.
8. Only BNAH is excited under this condition, see:
Fukuzumi, S.; Hironaka, K.; Tanaka, T. Chem. Lett. 1982,
1583.
9. Carey, F. A.; Sundberg, R. J. Advanced Organic Chem-
istry, 2nd ed.; Plenum Press: New York and London, 1984;
p. 183.