Single electron transfer (SET) reaction studies. A possible new role for nitrone spin trap

Article abstract of DOI:10.1016/S1386-1425(97)00269-2

A new role for nitrone (PBN:N-tert. Butyl α-phenyl nitrone) spin-trap in single electron transfer (SET) reactions is proposed.

Full text of DOI:10.1016/S1386-1425(97)00269-2

Spectrochimica Acta Part A 54 (1998) 779–784
Single electron transfer (SET) reaction studies. A possible new
role for nitrone spin trap
Md. N. Khan a, Suchandra Bhattacharjee a, Harish Chandra a,*,
Martyn C.R. Symons b
a Chemistry Department, North Eastern Hill Uni6ersity, Bijni Complex, Shillong 793 003, India
b Department of Applied Chemistry, DeMontfort Uni6ersity, Leicester, LE1 9BH, UK
Received 28 October 1997; accepted 31 October 1997
Abstract
A new role for nitrone (PBN:N-tert. Butyl h-phenyl nitrone) spin-trap in single electron transfer (SET) reactions
is proposed. © 1998 Elsevier Science B.V. All rights reserved.
Keywords: Single electron transfer; Nitrone; Tert. butylsuccinimidylmethyl nitroxide; DMSO; DMSO-d6
1. Introduction
these complexes oxidise a variety of organic com-
pounds and abstract hydrogen atoms from weak
C–H bonds. Janzen et al. [6] have reported that
benzyl halides react quantitatively with com-
pounds such as DMSO and we feel that species
The dual role of nitrones as an oxidant/reduc-
tant, as well as a spin trap is currently under
investigation. During the course of studies using a
range of solvents we have discovered some inter-
esting results for dimethyl sulphoxide (DMSO)
which we describe herein.
DMSO is an important source of methyl radi-
cals, both thermally [1], and photochemically [2].
The reaction of DMSO with OH radicals is a well
’
like PhCH Cl would certainly be more reactive.
We now report a new reaction in which N-
chlorosuccinimide reacts with PBN and DMSO in
dark to give the stable succinimidylmethyl adduct
of 2-methyl-2-nitrosopropane (MNP). The pro-
posed mechanism is shown in Scheme 1.
’
’
known source of CH3 radicals [3]. Attack by
other oxygen centred radicals is also thought to
’
generate CH3 radicals, via initial addition to sul-
phur [4]. Neta et al. [5] have reported that Br and
Cl atoms form complexes with DMSO and that
2. Experimental
N-chlorosuccinimide (Aldrich) was recrystal-
lized from dichloromethane prior to use. Dried,
spectrograde quality solvents and PBN (Aldrich)
were used as supplied. Spin trapping experiments
* Corresponding author. Tel.: +91 0364 226593; fax: +91
0364 760076; e-mail: harish@nehus.ren.nic.in
1386-1425/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved.
PII S1386-1425(97)00269-2

780
M.N. Khan et al. / Spectrochimica Acta Part A 54 (1998) 779–784
Scheme 1.
were performed using ca. 10−3 M solutions and
the samples were thoroughly degassed directly
prior to use. A Varian E-109, X band spectrome-
ter with 100 kHz field modulation was used for
ESR measurements. Solutions of N-chlorosuccin-
imide in 10% DMSO in benzene were mixed with
degassed solutions of PBN in a specially built
ESR cell in the dark. The reaction was monitored
continuously in the cavity of ESR spectrometer.
All experiments were carried out at room temper-
ature (2092°C).
solved feature for a second radical centre (vii)
(Fig. 1a) with hyperfine parameters aN=15.2 G,
aiH=11.4 G and aN%=1.4 G (from succinimidyl
nitrogen). This is well simulated using the above
hfsc and line width of 0.6 G (Fig. 1b) and agrees
well with the experimental one. These hyperfine
parameters are in good agreement with those
reported by Perkins et al. [8] and assigned to
radical tert.-butylsuccinimidyl methyl nitroxide.
These were formed by thermal or photochemical
decomposition of tert.butyleuccinimide-peroxy-
carboxylate using MNP as a spin trap.
3. Results and discussion
4. Mechanism
The first radical species (iii) detected had aN=
35
37
12.18 G, aiH=0.81 G, a =6.20 G, a =5.12 G.
We suggest that on the basis of their redox
properties [9], the reaction is initiated in all prob-
ability by a single electron transfer (SET) from
PBN to N-chlorosuccinimide (step 1) resulting in
the formation of nitrone radical cations (PBN’+),
Cl
Cl
These values agree well with those reported for
the chloro adduct of PBN [7]. This adduct is
unstable and after ca. 30 min the spectrum had
been totally replaced by an intense and well re-

M.N. Khan et al. / Spectrochimica Acta Part A 54 (1998) 779–784
781
Fig. 1. (a) ESR spectra of tert.butylsuccinimidylmethyl nitroxide (vii); (b) Computer simulated ESR spectra for adduct (vii)
[aN=15.2 G, abH=11.4 G, aN%=1.4 G, line width=0.6 G].
’
succinimidyl radicals (S ) and chloride ions (Cl−).
This leads to the ready formation of the chloro
adduct of PBN, via ion-pair formation and subse-
quent collapse [9]. We postulate that this adduct
breaks down by i-scission, to give MNP and the
MNP yields the stable radical di-tert.butyl nitrox-
ide [{(CH3)3C}2NO ], and we confirmed that these
radicals are generated using high concentrations
of PBN and low concentrations of DMSO.
In order to explain the dominance of the final
adduct (vii) we suggest that radicals can add to
’
’
’
PhCH Cl radical (R ). It is well established that

782
M.N. Khan et al. / Spectrochimica Acta Part A 54 (1998) 779–784
Fig. 2. (a) ESR spectra of deuterated analogue of adduct (vii); (b) Computer simulated ESR spectra of deuterated analogue of
adduct (vii) [aN=15.25 G, aDi =1.65 G, aN%=1.4 G, line width=0.5 G].
lated with aN=15.25 G, aiD=1.75 G (aHi /6.51=
’
DMSO, with consequent formation of CH3 radi-
aiD), and aN%=1.4 G (from succinimidyl nitrogen)
cals (step 2). This radical is then trapped primarily
and line width 0.5 G matches very well with the
experimental spectra.
by MNP, which is known to have a much higher
reactivity than PBN towards CH3 radicals [10]. In
’
In the light of these observations, the spectra
shown in Fig. 2a can be assigned as due to the
deuterated analogue of (vii). These experimental
order to check these postulates, low concentra-
tions of MNP were added during the course of the
reaction. This resulted in the rapid growth of an
intense spectrum from adduct (vii). This result
strongly supports the suggestion that MNP is an
important intermediate. Also traces of N-methyl-
succinimide were added. This had no affect on the
spectra for adduct (vii). This rules out the possi-
bility that N-methyleuccinimide is formed from
’
results confirm that CH3 originates from DMSO
alone. However, when similar experiments were
’
done in DMSO alone, we did not see C1-PBN
adduct (iii) (due to its high instability in highly
polar solvents) but adduct (vii) was observed after
ca. 15 min.
’
’
CH3 and S followed by H-abstraction therefrom
and subsequent trapping by MNP.
On the basis of these experimental observa-
tions, we propose that PBN is a good single
electron transferring agent. The detection of
chloro adduct of PBN is in accord with the ‘in-
verted’ spin trapping mechanism proposed by
Eberson et al. [9]. It is appreciated that this
sequence of reactions involves many stages, some
of relatively low probability. However, the clear
’
’
We propose that R and/or S abstract hydro-
gen from adduct (v) leading to methylene nitrone
(vi) as observed similarly by Perkins et al. [8]. The
product (vi) thus formed, is itself a highly reactive
’
spin trap and reacts with S yielding (vii). In order
to confirm the source of methyl radicals, we re-
peated the work with DMSO-d6 and found (3×7)
line spectra, Fig. 2a. The spectra, Fig. 2b simu-
’
detection of the Cl–PBN seems to constitute
good evidence for the proposed SET reaction.

M.N. Khan et al. / Spectrochimica Acta Part A 54 (1998) 779–784
783
Fig. 2. (Continued)
Also, it seems clear that radical (vii) is correctly
identified, and, as yet, the suggested reaction
sequence in scheme is the only route that
we have been able to formulate. According to
Kochi et al. [11], the charge transfer complexes
are precurssor to the single electron transfer
processes. In our system, we did observe an
isosbestic point by UV spectroscopy (in n-hex-
ane) Fig. 3, wherein the relative concentration
of PBN and N- chlorosuccinimide was varied.
This proof and the significant difference in
their redox properties clearly suggests the feasi-
bility of single electron transfer process in this
system.

784
M.N. Khan et al. / Spectrochimica Acta Part A 54 (1998) 779–784
Fig. 3. UV spectra of PBN and N-chlorosuccinimide in n-hexane showing Isosbestic point. – · – · –, UV spectra of PBN alone,
6.1×10−5 M; ——, UV spectra of N-chlorosuccinimide alone, 6.1×10−5 M; – – –, UV spectra of NCS and PBN in I:I; · · ·,
UV spectra of NCS and PBN in I:II; – × – × –, UV spectra of NCS and PBN in I:III.
3625; (b) W. Damerau, Z. Chem. 9 (1969) 343; (c) H.
Hatano, J. Phys. Chem. 76 (1972) 135.
Acknowledgements
[4] (a) M.L. Poutsma, J. Am. Chem. Soc. 87 (1965) 2172; (b)
Mr. Md. Nadeem Khan and Suchandra Bhat-
D.E. Applequist, John A. Landgrebe, J. Am. Chem.
tacharjee acknowledge the Financial support from
the Department of Atomic Energy, Government
of India and State Government of Meghalaya,
respectively.
(1964) 1543.
[5] Lian C.T. Shoute, P. Neta, J. Phys. Chem. 94 (1990)
2447.
[6] E.G. Janzen, Guoman Chen, Tammy M. Bray, Lester A.
Reinke, J. Lee Poyer, Paul B. McCay, J. Chem. Soc.
Perkin Trans. 2 (11) (1993) 1983.
[7] L. Eberson, J. Chem. Soc. Perkin Trans. 2 (1992) 1807.
[8] G.R. Chalfont, M.J. Perkins, J. Chem. Soc. (B) (1970)
401.
References
[9] L. Eberson, John Lind, Gabor Merenyi, J. Chem. Soc.
Perkin Trans. 2 (1994) 1181.
[1] J.W. Hudgens, T.G. DiGiuseppe, M.C. Lin, J. Chem.
Phys. 79 (1983) 571.
[10] M.J. Perkins, Adv. Phys. Org. Chem. 17 (1980) 33.
[11] J.K. Kochi, Angew. Chem. Inter. Ed. (English) 27 (1988)
1227.
[2] K. Gollnick, H.U. Strauss, Pure Appl. Chem. 33 (1973)
217.
[3] (a) W.T. Dixon, R.O.C. Norman, J. Chem. Soc. (1964)
.