Double fragmentation in cation radicals. An example in the NADH analogues series

Article abstract of DOI:10.1002/(SICI)1099-1395(1998110)11:11<774::AID-POC40>3.0.CO;2-T

The cation radical of 9-tert-butyl-N-methylacridan, generated electrochemically or photochemically, offers, in the presence of strong bases, a remarkable example of a double fragmentation. Whereas in acidic or weakly basic media the tert-butyl radical is cleaved with concomitant formation of the methylacridinium cation, the presence of a strong base triggers the cleavage of both the methyl group borne by the nitrogen atom and the tert-butyl group on C-9 leading to acridine, formaldehyde and the tert-butyl anion, even though methylacridinium cation is stable under these conditions. The origin of this unprecedented behavior resides in the prior deprotonation of the methyl group borne by the nitrogen atom which outruns the usual deprotonation at the 9-carbon because this is slowed by the steric hindrance due to the presence of the tert-butyl group.

Full text of DOI:10.1002/(SICI)1099-1395(1998110)11:11<774::AID-POC40>3.0.CO;2-T

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 11, 774–780 (1998)
Double fragmentation in cation radicals. An example in the
NADH analogues series
AgneÁ s Anne, Sylvie Fraoua, Jacques Moiroux* and Jean-Michel SaveÂant*
Laboratoire d'Electrochimie Mole culaire de l'Universite Denis Diderot, Unite Mixte de Recherche UniversiteÂ, CNRS No. 7591, 2 Place
Jussieu, 75251 Paris Cedex 05, France
Received 5 November 1997; revised 8 January 1998; accepted 3 February 1998
ABSTRACT: The cation radical of 9-tert-butyl-N-methylacridan, generated electrochemically or photochemically,
offers, in the presence of strong bases, a remarkable example of a double fragmentation. Whereas in acidic or weakly
basic media the tert-butyl radical is cleaved with concomitant formation of the methylacridinium cation, the presence
of a strong base triggers the cleavage of both the methyl group borne by the nitrogen atom and the tert-butyl group on
C-9 leading to acridine, formaldehyde and the tert-butyl anion, even though methylacridinium cation is stable under
these conditions. The origin of this unprecedented behavior resides in the prior deprotonation of the methyl group
borne by the nitrogen atom which outruns the usual deprotonation at the 9-carbon because this is slowed by the steric
hindrance due to the presence of the tert-butyl group. 1998 John Wiley & Sons, Ltd.
KEYWORDS: double fragmentation; cation radicals; NADH analogues
INTRODUCTION
one carbon–nitrogen bond and of one carbon–carbon (or
carbon–oxygen) bond triggered by the initial deprotona-
tion of the cation radical.
The chemistry of cation radicals attracts sustained
attention in connection with their importance in organic
and biological processes. Anodic oxidations are also
currently analyzed in terms of cation radical chemis-
try.1ab The reactions of ion radicals fall into two
categories, one in which they react as radicals, the other
in which they react as Lewis and/or Brønsted acids
(cation radicals) or as Lewis and/or Brønsted bases (anion
radicals).1c Coupling of the unpaired electrons in the
encounter of two ion radicals leading to a dimer is one of
the most typical reactions of the first type. Concerning
cation radicals, such reactions are of crucial importance
in the dynamics of the first stages of electropolymeriza-
tions leading to conductive polymers.2 Reaction with a
nucleophile, yielding, after deprotonation, anodic sub-
stitution products is one of the reactions of the second
type.3 Deprotonation may also be the first step of cation
radical transformation4–6 leading to a radical that may be
further oxidized or undergo carbon–carbon or carbon–
heteroatom bond fission. Another possibility is that
carbon bond fragmentation occurs in the first stage of
the cation radical evolution as, for example, in the case of
tert-butylated analogues of NADH.7
NADH analogues belong to the general class of
dihydropyridines. An analysis of single and double
fragmentation reactions in cation radicals of dihydropyr-
idines may be important for a better understanding of the
mechanism of metabolisation of these compounds by
cytochrome P450 which opposes their use as calcium
antagonists.8
The compound we have investigated in this respect is
CH3AHt-Bu (Chart I).
RESULTS AND DISCUSSION
With the exception of CH3AHt-Bu, the electrochemically
generated cation radicals of all the tert-butyl derivatives
.
illustrated undergo a C—t-Bu cleavage, yielding the t-Bu
‡
radical and the pyridinium cation, AH , whatever the pH
of the solution.7a With CH3AHt-Bu, the formation of t-
.
‡
Bu and of the 10-methylacridinium cation, CH3AH ,
was also observed, albeit only in acidic media, namely
between pH 9.4 and 14.7. In the presence of bases of
higher pKa starting with collidine (2,4,6-trimethylpyr-
idine, pKa = 15.6), a different behavior was observed. It
has been shown earlier that the cation radicals of 9-
substituted acridans, CH3AHR (R = Ph, CH3, CH2Ph6b,c
and CN6d) do not undergo any C—C cleavage but rather
In the same series of compounds we have found still
another type of transformation, namely the cleavage of
*Correspondence to: J. Moiroux or J.-M. Save´ant, Laboratoire
d’Electrochimie Mole´culaire de l’Universite´ Denis Diderot, Unite´
Mixte de Recherche Universite´, CNRS No. 7591, 2 Place Jussieu,
75251 Paris Cedex 05, France.
deprotonate in the presence of a base to yield the
.
corresponding radicals, CH3AR , and, after a second
1998 John Wiley & Sons, Ltd.
CCC 0894–3230/98/110774–07 $17.50

DOUBLE FRAGMENTATION IN CATION RADICALS
775
‡
1
electron transfer, the CH3AR cation. The cation radicals
goes up to 500 V s with the strongest base). The
.
‡
of all the AH2 derivatives illustrated also deprotonate at
standard potential of the CH3AHt-Bu/CH3AHt-Bu
couple derived from these experiments as the mid-point
between the anodic and cathodic peak potentials is 0.910
V vs SCE. These observations prove that the cation
radical is indeed the initial intermediate of the sequence
of reactions leading to the formation of acridine.
C-9 in the presence of a base.6b,c One would have
.
‡
therefore expected that CH3AHt-Bu would deprotonate,
yielding CH3At-Bu and eventually CH3At-Bu , when
the pH becomes high enough for this reaction to outrun
the extrusion of the t-Bu radical. In fact, a different
behavior was observed, namely the formation of acridine
and formaldehyde corresponding to the following overall
reaction:
.
‡
.
At low scan rates (v < 1 V s 1), in the presence of a
strong base (pKa ꢀ 15.6), a two-electron irreversible
anodic wave is observed (Fig. 2, full line) which shifts in
.
‡
CH3AHt-Bu can be generated electrochemically or
photochemically. Its UV–visible spectrum obtained by
laser pulse photolysis in the presence of an oxidative
quencher (CCl4) is as shown in Fig. 1. The initial
formation of the cation radical in electrochemical
experiments can be detected by means of cyclic
voltammetry where a one-electron reversible wave is
observed upon raising the scan rate even in the most basic
media (the scan rate required to observe the reversibility
the negative direction upon increasing the concentration
of base. The quasi-reversible wave appearing upon scan
reversal (re-reduction followed by re-oxidation) is the
same as the wave obtained with a solution containing
only protonated acridine besides the supporting electro-
lyte. These observations contrast with the behavior that is
found in the absence of base (Fig. 2, dotted line). The
irreversible anodic wave is significantly smaller. The re-
reduction wave is irreversible and has a different location
1998 John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 11, 774–780 (1998)

776
A. ANNE ET AL.
.
‡
Figure 1. Absorption spectrum of CH3AHt-Bu in acetoni-
trile recorded ms after the pulse. CH3AHt-Bu
Figure 2. Cyclic voltammograms of CH3AHt-Bu (2 mM) in the
absence (dotted line) and presence (full line) of a strong base
(0.17 M 2,4,6-trimethylpyridine) in acetonitrile ‡ 0.1 M
Bu4NBF4 at 20°C. Successive potential scans from 0 to
1.2 V, from 1.2 to 1.2 V and from 1.2 to 0 V at v = 0.2 V
1
(0.16 mM) ‡ 0.1% CCl4; 120 mJ laser pulse at 308 nm
1
s
‡
and shape. It corresponds to the reduction of the CH3AH
.
cation formed upon extrusion of t-Bu radical as
discussed earlier:7a
electrolyzed solution is reported in Table 1 (under the
heading ‘CV assay’). The species that could be assayed
through their anodic or cathodic peaks under such
‡ꢁ
‡
CH3AHt-Bu !CH3AH ‡ t-Buꢁ
Protons are produced upon further oxidation of the
products. Since there is no base present in the solution
other than the starting molecule, CH3AHt-Bu, this is
partly protonated and thus partly inactivated toward
oxidation, thus explaining why the anodic wave is
smaller in the absence of base than in the presence of a
strong base. In the presence of a weak base, the same
cleavage of the cation radical occurs but the peak height
corresponds to two electrons per molecule and the peak
potential does not depend on the concentration of base.
Turning back to the case of a strong base, strong
indications exist at this stage that acridine is the main
reaction product and that the sequence of reactions that
eventually lead to this product starts with the initial
formation of the cation radical, immediately followed by
a step in which the added base is involved. It is also worth
conditions
are
CH3AHt-Bu,
N-methylacridane
‡
(CH3AH2), acridine, N-methylacridone, CH3AH and
the dimer (CH3AH)2 resulting from the one-electron
reduction of CH3AH .
High-performance liquid chromatography (HPLC)
was also used to analyze the electrolyzed solution (see
Experimental section). The yields thus found are reported
in Table 1 (under the heading ‘HPLC assay’). The
agreement between the CV and the HPLC assays is
satisfactory.
‡ 10
The main product is acridine. However, a small
amount of CH3AH is still found, indicating that the t-
‡
Bu extrusion reaction that occurs in the absence of base,
or presence of weak bases, still competes, albeit to a
small extent.
The HPLC analysis of a solution of 0.16 mM CH3AHt-
Bu and 0.2 M 2,4,6-trimethylpyridine in acetonitrile after
one laser pulse irradiation was also carried out. As can be
‡
noting that the CH3AH cation is stable in the presence
of the strong bases used here.9 Therefore, the reaction
triggered by the latter is not merely the conversion into
‡
acridine of the CH3AH cation that would be formed
initially.
Table 1. Distribution of products in the electrochemical and
photochemical oxidation of CH3AHt-Bu in the presence of
2,4,6-trimethylpyridine
More information on product distribution was gained
at the preparative scale. A potential-controlled electro-
lysis of a solution containing 2.18 mM CH3AHt-Bu, 0.2 M
2,4,6-trimethylpyridine and 0.1 M Bu4NBF4 in acetoni-
trile was carried out at 1.0 V vs SCE at a platinum grid
working electrode. Aliquots were assayed by means of
cyclic voltammetry during the course of the electrolysis.
Electrolysis was stopped when the height of the anodic
peak due to the oxidation of the remaining CH3AHt-Bu
was ca 10% of its original value. The number of electrons
exchanged was found to be 2.1, in agreement with cyclic
voltammetry (Fig. 2). The resulting composition of the
Controlled-potential
Laser flash
electrolysis
photolysis:
CV assay HPLC assay HPLC assay
Compound
(%)
(%)
(%)
CH3AHt-Bu
CH3AH
(CH3AH)2
CH3AH2
11
13
0
8
11
0
93
2
0
0
‡
0
0
Acridine
N-Methylacridone
75
0
73
0
3.5
0
1998 John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 11, 774–780 (1998)

DOUBLE FRAGMENTATION IN CATION RADICALS
777
.
‡
Table 2. Rate constants of deprotonation of CH3AHt-Bu by
strong bases
a
Log
c
Baseb
pKa,BH‡
Logkd Logke (kH/kD)f
2,4,6-Trimethylpyridine
Benzylamine
tert-Butylamine
Piperidine
15.6
16.8
18.1
18.9
2.7 <4.5
4.6
4.7
5.0
5.1
6.8
0.6
0.6
>5.4
a
In acetonitrile at 20°C
b
15 < [B] < 150 mM, in buffered or unbuffered medium.
c
From Refs7b and11
.
d
Cyclic voltammetric determination, uncertainty = Æ0.2.
e Laser flash photolysis determination ([CH3AHt-Bu] = 0.16 mM, 0.1%
CCl4), uncertainty = Æ0.2.
Figure 3. IR spectrum of a solution containing initially
1.92 mM CH3AHt-Bu, 0.1 M Bu4NBF4 and 0.35 M 2,4,6-
trimethylpyridine in acetonitrile electrolyzed at a controlled
potential of 1.0 V vs SCE until the concentration of the
acridine produced is 1.1 mM. The arrows indicate two of the
bands that reveal the presence of formaldehyde and whose
heights can be used to determine the formaldehyde
concentration (see text)
f Cyclic voltammetric determination, uncertainty = Æ0.3.
CH3AHt-Bu „ CH3AHt-Buꢁ‡ ‡ e
k
CH3AHt-Buꢁ‡ ‡ B ! ꢁCH2AHt-Bu ‡ BH
‡
ꢁCH2AHt-Bu ‡ CH3AHt-Buꢁ‡
!
seen in Table 1, acridine is again produced, even if the
conversion to products is small in these experiments.
The formation of acridine implies the oxidative
extrusion of the methyl group from the nitrogen atom.
We therefore looked for formaldehyde among the
reaction products. The IR spectrum of the electrolyzed
‡
CH2AHt-Bu ‡ CH3AHt-Bu
Scheme 1
*
‡
solution is shown in Fig. 3. The two bands at 3548 and
The kinetics of the reaction of CH3AHt-Bu with
strong bases may also be followed by the decrease of its
UV–visible spectrum with time after generation by a
laser pulse in the presence of an oxidative quencher
1
3616 cm
are absent in the spectrum of a solution
‡
containing only CH3AHt-Bu, CH3AH , acridine and
2,4,6-trimethylpyridine, introduced in acetonitrile at the
same concentrations as those given by the CV or HPLC
*
‡
(CCl4). At 660 nm, CH3AHt-Bu absorbs appreciably
right after the pulse (see Fig. 1) and the absorbance
decreases to zero with increasing time. First-order
kinetics are obeyed, the pseudo first-order rate constant
1
assays. The absorbance at 1610 cm is also markedly
smaller. Addition of formaldehyde brings about the
appearance of the two bands at 3548 and 3616 cm 1 and
1
a significant increase in the absorbance at 1610 cm
.
Obviously H2CO is produced in the electrochemical
oxidation of CH3AHt-Bu in the presence of 2,4,6-
trimethylpyridine. Plots of the peak heights at 3548 and
1
3616 cm vs H2CO concentration in the 0–2 mM range
give two calibration curves showing that both peak
heights are roughly proportional to the H2CO concentra-
tion over this range. Using the calibration curves, the
H2CO: acridine ratio is 0.85 Æ 0.05, a result proving that
the observed demethylation eventually produces H2CO.
The CV oxidation wave of CH3AHt-Bu in the presence
of a strong base passes from a two-electron irreversible
behavior to a one-electron reversible behavior upon
raising the scan rate. We may use this variation of the
peak height to investigate the kinetics of the reaction of
the cation radical with the base. The variations of the
peak current with the scan rate and the base concentration
are displayed in Fig. 4. They fit well with a ‘DISP1’
mechanism1c such as that depicted in Scheme 1 in which
it is assumed that the reaction of the base with the cation
radical involves a deprotonation of the methyl group
borne by the nitrogen. The values of the deprotonation
rate constants thus obtained are summarized in Table 2.
Figure 4. Variation of the rate constant of the reaction
between CH3AHt-Bu* obtained from the oxidation of
‡
CH3AHt-Bu (3 mM) and tert-butylamine (concentration
in mM) in acetonitrile ‡ 0.5 M Et4NBF4. Scan rate, v, in V
s
1. [B] = 12.4 mM (*), 81.6 mM (~), 150 mM (&). The solid
line represents the theoretical variation of ip/ip0 (ip is the peak
current and ip0 the peak current for a one-electron reversible
transfer) for a DISP1 mechanism with k = 104.7 l mol
(see text)
s
1
1
1998 John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 11, 774–780 (1998)

778
A. ANNE ET AL.
Scheme 2
being proportional to the concentration of the strong base
as already observed for the deprotonation of cation
radicals of other NADH analogues, using the same
technique.6 The resulting second-order rate constants are
reported in Table 2. The agreement between the rate
constants derived from cyclic voltammetry and laser flash
photolysis is satisfactory in the range where each of the
two techniques could be used.
In order to prove that the base does deprotonate the
cation radical at the methyl borne by the nitrogen atom,
as depicted in Scheme 1, we investigated the kinetic
isotope effect resulting from the replacement of N-CH3
by N-CD3. As can be seen in Table 2, there is indeed a
significant kinetic isotope effect, confirming that the
deprotonation of the cation radical at the methyl borne by
the nitrogen atom is the rate-determining step of the
reaction eventually leading to acridine and formaldehyde.
The preceding observations thus lead to the reaction
mechanism depicted in Scheme 2. Instead of the last step
of Scheme 2, one could envisage that the 9-tert-butyl-N-
hydroxymethylacridan would decompose to CH2O and
acridan and that the latter would be oxidized to acridine.
This pathway would, however, consume two additional
electrons, leading to a total electron stoichiometry of 4.
The very fact that the electron stoichiometry is only 2
implies that the tert-butyl group is expelled as an anion. t-
Bu may react with any proton donor present to give 2-
methylpropane or with the acetonitrile solvent to give 2-
imino-3,3-dimethylbutane or a polymer. As reported in
the Experimental section, all attempts to identify 2-
methylpropane or 2-imino-3,3-dimethylbutane (or 2-
keto-3,3-dimethylbutane after addition of water) by
means of gas-phase chromatography (GPC) and IR
absorption spectroscopy, respectively, were unsuccess-
ful. The cleavage of t-Bu probably initiates the anionic
oligomerization of solvent molecules, explaining why the
solution turns purple during the electrolysis in the
presence of 2,4,6-trimethylpyridine (unbuffered medium)
or orange in a buffered medium (2,4,6-trimethylpyridine
plus HClO4).
The next question to be addressed is why deprotona-
*
‡
tion of CH3AHt-Bu involves the methyl borne by the
nitrogen atom rather than the hydrogen borne by C-9 as it
*
*
*
‡
‡
‡
does with CH3AH2 , CH3AHPh , CH3AH CH3 and
*
‡
CH3AH CH2Ph . It has already been noted that the
intrinsic barrier for the deprotonation of these cation
radicals at the C-9 position increases in the indicated
order as a consequence of steric hindrance according to
the representation given in Scheme 36e (for clarity, the
lateral benzene rings of CH3AHR and the substituents on
pyridine have been omitted). The steric effect arises at the
level of the successor complex and causes the intrinsic
barrier to increase while the precursor complex and the
transition state remain approximately unchanged. One
1998 John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 11, 774–780 (1998)

DOUBLE FRAGMENTATION IN CATION RADICALS
779
Scheme 3
may therefore suggest that, with a tert-butyl group in the
9-position, steric hindrance is large enough for the
competing N-CH3 deprotonation to become the preferred
pathway, thus triggering the sequence of reactions
depicted in Scheme 2. It is interesting in this connection
that deprotonation at the N-CH3 group has previously
been observed in the case of CH3AH2 in photochemical
experiments. The sentitizer being a ketone or a quinone,
the photogenerated basic center, viz. the anionic oxygen
of the ketyl radical, is thus located much closer to the N-
CH3 group and than to C-9 in the radical ion pair resulting
from the photoelectron transfer reaction.12
HPLC assays were performed as follows. Aliquots
(20 ml) of the reaction mixture were injected on to a
reversed-phase Hypersil C18; ODS2 column. The mobile
phases were (A) 80:20 methanol–water ‡ 10 mM Na2H-
PO4 and (B) methanol. The gradient consisted of a 5 min
isocratic step with 100% A, followed by a 7 min isocratic
step with 100% B and a 5 min linear step returning to the
initial conditions. The flow-rate was 1.0 ml min 1. The
chromatographic peaks were identified by UV spectro-
photometry at 290 and/or 342 nm by comparison with
authentic samples. Quantitation of acridine (tr = 5.4 min),
‡
AHCH3 (tr = 6.4 min) and unreacted t-BuAHCH3
(tr = 14.9 min) was derived from integrated peak vs
concentration calibration curves.
In IR absorption measurements the cell thickness,
between CaF2 windows, was 0.5 mm. Calibration curves
showed that the peak heights at 3548 and 3616 cm
CONCLUSIONS
1
The cation radical of CH3AHt-Bu can be generated
electrochemically or photochemically. In acidic or
weakly basic media it undergoes the elimination of the
were proportional to the formaldehyde concentration in
the range 0–2 mM. Searching evidence for the possible
production of 2-imino-3,3-dimethylbutane, we pro-
ceeded as follows. Water (2 mM) was added to the
electrolyzed solution to hydrolyze the imine to 2-keto-
3,3-dimethylbutane, which exhibits characteristic IR
absorption bands at 3000, 1700, 1450, 1350 and
1140 cm 1. We did not detect the appearance of any of
those bands.
In order to obtain evidence for the possible production
of 2-methylpropane, the electrolysis was carried out in a
cell sealed with septa. Both the electrolyzed solution and
the gas phase above it were analyzed by means of GPC. A
Porapak Q column 80–100 mesh of 2 m length (Alltech),
was used in an oven at 70°C with helium under a pressure
of 4 bar as the carrier gas and a thermal conductivity
detector. The retention times of 2-methylpropane, n-
butane and acetonitrile were 12.5, 15 and 21 min,
respectively. Injections of 10 ml of the solution and
500 ml of the gas phase gave no peaks at 12.5 or 15 min.
‡
tert-butyl radical, forming the CH3AH cation. In the
presence of strong bases, it offers a remarkable example
of a double fragmentation involving the extrusion of both
the methyl group borne by the nitrogen atom and the tert-
butyl group on C-9 leading to acridine, formaldehyde and
the tert-butyl anion. The origin of this unprecedented
behavior resides in the prior deprotonation of the methyl
group borne by the nitrogen atom which outruns the usual
deprotonation at C-9 which is slowed by the steric
hindrance caused by the presence of the tert-butyl group.
EXPERIMENTAL
Chemicals. The various methylacridane derivatives were
prepared according to previously described procedures,
CH3AHt-Bu,7cCD3AHt-Bu,6d (CH3AH)2,10 CH3AH
‡
and CH3AH2.13 All other chemicals were obtained from
Aldrich and were of the highest purity available. They
were used as received.
Instruments and procedures. They were the same as
described previously for cyclic voltammetry, electro-
lysis10 and laser flash photolysis.6d The optical pathlength
for UV–visible spectrophotometric detection was 2 mm.
Acknowledgment
We are indebted to Alain Adenier for his help with FTIR
spectroscopy.
1998 John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 11, 774–780 (1998)

780
A. ANNE ET AL.
Anne, P. Hapiot, J. Moiroux, P. Neta and J. M. Save´ant, J. Phys.
REFERENCES
Chem. 95, 2370–2377 (1991); (c) A. Anne, P. Hapiot, J. Moiroux,
P. Neta and J. M. Save´ant, J. Am. Chem. Soc. 114, 4694–4701
(1992); (d) A. Anne, S. Fraoua, P. Hapiot, J. Moiroux and J. M.
Save´ant, J. Am. Chem. Soc. 117, 7412–7421 (1995); (e) A. Anne,
S. Fraoua, V. Grass, J. Moiroux and J. M. Save´ant, J. Am. Chem.
Soc. 120, 2951–2958 (1998).
1. (a) S. D. Ross, M. Finkelstein and E. J. Rudd, Organic Chemistry,
Vol. 32, pp. 211–218. Academic Press, New York (1975); (b) H.
Lund and M. M. Baizer, Organic Electrochemistry, Marcel
Dekker, New York (1991); (c) C. P. Andrieux and J.-M. Save´ant,
in Investigations of Rates and Mechanisms of Reactions;
Techniques of Chemistry, edited by C.F. Bernasconi, Vol. 6, 4th
Edition, Part 2, pp. 305–390. Wiles, New York (1986).
2. (a) G. B. Street, in Handbook of Conducting Polymers, edited by
T.A. Skotheim. Vol. 1, pp. 265–291. Marcel Dekker, New York
(1986); (b) for kinetic studies on the early stages of the
electropolymerization of pyrroles, see C. P. Andrieux, P. Audebert,
P. Hapiot and J.-M. Save´ant, J. Phys. Chem., 95, 10158–10164
(1991).
3. (a) L. Eberson, Adv. Phys. Org. Chem. 18, 79–185 (1982); (b) V.
D. Parker, Acc. Chem. Res. 17, 243–250 (1984); (c) U. C. Yoon
and P. S. Mariano, Acc. Chem. Res. 25, 233–240 (1992); (d) for
more recent work, see M. S. Workentin, L. J. Johnston, D. D. M.
Wayner and V. D. Parker, J. Am. Chem. Soc. 116, 8279–8287
(1994), and references cited therein.
4. (a) For a review on fragmentation of photochemically generated
cation radicals, see A. Albini, M. Mella and M. Freccero,
Tetrahedron 50, 575–607 (1994).
7. (a) S. Fukuzumi, Y. Tokuda, T. Kitano, T. Okamoto and J. Otera, J.
Am. Chem. Soc. 115, 8960–8968 (1993); (b) A. Anne, J. Moiroux
and J. M. Save´ant, J. Am. Chem. Soc. 115, 10224–10230 (1993);
(c) A. Anne, S. Fraoua, J. Moiroux and J. M. Save´ant, J. Am. Chem.
Soc. 118, 3938–3945 (1996); (d) For carbon–carbon fragmentation
in other cation radicals, see (e)–(j) references therein; the
sequencing of deprotonation C—C cleavage is not always
obvious;7j (e) P. Maslak, Top. Curr. Chem. 168, 1-46 (1993); (f)
P. Maslak, T. M. Vallombroso, W. H. Chapman and J. N. Narvaez,
Angew. Chem., Int. Ed. Engl. 33, 73–75 (1994); (g) L. Chen, M. S.
Farahat, H. Gan, S. Farid and D. G. Whitten, J. Am. Chem. Soc.
117, 6398–6399 (1995); (h) H. Gan, U. Lenhos, I. R. Gould and D.
G. Whitten, J. Phys. Chem. 99, 3566–3573 (1995); (i) J. H. Horner,
F. N. Martinez, O. M. Musa, M. Newcomb and H. Shahin, J. Am.
Chem. Soc. 117, 11124–11135 (1995); (j) R. D. Burton, M. D.
Bartberger, Y. Zhang, J. R. Eyler and K. S. Schanze, J. Am. Chem.
Soc. 118, 5655–5664 (1996).
5. (a) See Refs 4(a) and (b)–(d) and references cited therein; (b) E.
Bacciochi, T. Del Giaco and F. Elisei, J. Am. Chem. Soc. 115,
12280–12289 (1993); (c) B. Karki, J. P. Dinnocenzo, J. P. Jones
and K. R. Korzekwa, J. Am. Chem. Soc. 117, 3657–3664 (1995);
(d) F. D. Lewis, D. M. Bassani, E. L. Burch, B. E. Cohen, J. A.
Engleman, G. D. Reddy, S. Schneider, W. Jaeger, P. Gedeck and
M. Gahr, J. Am. Chem. Soc. 117, 660–669 (1995); (e) X. Zhang, S.
R. Yeh, Hong, M. Freccero, A. Albini, D. E. Falvey and P. S.
Mariano, J. Am. Chem. Soc. 116, 4211–4220 (1994); (f) V. D.
Parker and M. Tilset, J. Am. Chem. Soc. 113, 8778–8781 (1991);
(g) L. M. Tolbert, R. K. Khanna, E. Popp, L. Gelbaum and L. A.
Bottomley, J. Am. Chem. Soc. 112, 2373–2378 (1990).
6. (a) With the cation radicals of NADH analogs, Brønsted plots have
been obtained using a wide range of opposing bases. The pKas of
the cation radicals could be derived from experimental data, hence
allowing the determination of the intrinsic barriers,6b–e which is
seldom the case in cation radical deprotonation studies; (b) A.
8. (a) J. S. Lee, N. E. Jacobsen and P. R. Ortiz de Montellano,
Biochemistry 27, 7703–7710 (1988); (b) F. P. Guengerich, W. R.
Brian, M. Iwasakei, M. A. Sari, C. Ba¨a¨rnhielm and P. Berntsson, J.
Med. Chem. 34, 1838–1844 (1991).
9. A. Anne and J. Moiroux, J. Org. Chem. 55, 4608–4614 (1990).
10. A. Anne, P. Hapiot, J. Moiroux and J. M. Save´ant, J. Electroanal.
Chem. 331, 959–970 (1992).
11. (a) G. Cauquis, A. Deronzier, D. Serve and E. Vieil, J. Electroanal.
Chem. 60, 205–215 (1975); (b) I. M. Kolthoff, M. K. Chantoni and
S. Bhowmik, J. Am. Chem. Soc. 90, 23–28 (1968); (c) J. F.
Coetzee, Prog. Phys. Org. Chem. 4, 76–77 (1967).
12. (a) K. S. Peters, E. Pang and J. Rudzki, J. Am. Chem. Soc. 104,
5535–5537 (1982); (b) L. E. Manring and K. S. Peters, J. Am.
Chem. Soc. 107, 6452–6458 (1985).
13. R. M. G. Roberts, D. Ostovic and M. M. Kreevoy, Faraday
Discuss. Chem. Soc. 74, 257–265 (1982).
1998 John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 11, 774–780 (1998)