Electrocatalytic dehalogenation of α-bromoketones in the presence of Cp2TiCl2

Article abstract of DOI:10.1007/s11172-007-0152-7

Processes of direct and electrocatalytic (in the presence of electrochemically reduced Cp₂TiCl₂) reduction of three α-bromoketones containing the C(sp³)-Br or C(sp²)-Br bond, viz., 2-bromo-and 2,6-dibromo-4,4-dimethylcyclohexa-2,5-dien-1-ones and α-bromo-acetophenone, were studied by cyclic voltammetry and preparative electrolysis. In all cases, the dissociative electron transfer proceeds via the concerted mechanism. Preparative electrolysis of these α-bromoketones in the presence of Cp₂TiCl₂ affords the reductive debromination products in 40-80% yield at low cathodic potentials (-0.85 V vs. Ag/AgCl/KCl). In the case of 2,6-dibromo-4,4-dimethylcyclohexa-2,5-dien-1-one in the potentiostatic regime, only one bromine atom can be eliminated selectively.

Full text of DOI:10.1007/s11172-007-0152-7

Russian Chemical Bulletin, International Edition, Vol. 56, No. 5, pp. 1013—1019, May, 2007
1013
Electrocatalytic dehalogenation of αꢀbromoketones
in the presence of Cp2TiCl2
O. M. Nikitin, G. V. Gavrilova, and T. V. Magdesievaꢀ
Department of Chemistry, M. V. Lomonosov Moscow State University,
1 Leninskie Gory, 119992 Moscow, Russian Federation.
Eꢀmail: tvm@org.chem.msu.ru
Processes of direct and electrocatalytic (in the presence of electrochemically reduced
Cp2TiCl2) reduction of three αꢀbromoketones containing the C(sp3)—Br or C(sp2)—Br bond,
viz., 2ꢀbromoꢀ and 2,6ꢀdibromoꢀ4,4ꢀdimethylcyclohexaꢀ2,5ꢀdienꢀ1ꢀones and αꢀbromoꢀ
acetophenone, were studied by cyclic voltammetry and preparative electrolysis. In all cases, the
dissociative electron transfer proceeds via the concerted mechanism. Preparative electrolysis of
these αꢀbromoketones in the presence of Cp2TiCl2 affords the reductive debromination prodꢀ
ucts in 40—80% yield at low cathodic potentials (–0.85 V vs. Ag/AgCl/KCl). In the case of
2,6ꢀdibromoꢀ4,4ꢀdimethylcyclohexaꢀ2,5ꢀdienꢀ1ꢀone in the potentiostatic regime, only one
bromine atom can be eliminated selectively.
Key words: electrochemical reduction, αꢀbromoketones, electrocatalytic reactions, mediaꢀ
tor catalysis, titanocene dichloride.
In most cases, the direct electrochemical reduction of
alkyl and vinyl halides occurs at rather high negative poꢀ
tentials1 and, therefore, the dehalogenation of compliꢀ
cated polyfunctional compounds is usually nonselective.
The use of nonꢀdirect electrocatalytic reduction (mediaꢀ
tor redox catalysis) makes it possible to decrease considꢀ
erably the potential of the reaction, which can be substanꢀ
tial for selectivity of the process.
the reduction mechanism. Therefore, it was interesting to
reveal whether the electrocatalytic dehalogenation of simiꢀ
lar compounds is possible in the presence of the electroꢀ
chemically reduced titanocene dichloride and whether this
process is a model of reactions involving discorhabdines
in the nature or not.
It is known2 that in the nature many dehalogenation
processes proceed via the reduction mechanism involving
the catalysts containing the compounds of Co,3,4 Ni,5—7
Cu,8 and other metals. We have previously shown9,10 that
the electrochemically reduced titanocene dichloride
(Cp2TiCl2) can serve as a catalyst for the reductive
dehalogenation of benzylic halides, whereas aryl halides
usually undergo no dehalogenation. The present work conꢀ
tinues these studies and is devoted to the elucidation of
the applicability of this approach to cleavage of the C—Hal
bond of other types. αꢀBromoketones containing the
C(sp3)—Br or C(sp2)—Br bond, which is activated by
the adjacent carbonyl group, namely, αꢀbromoacetoꢀ
phenone (1), 2,6ꢀdibromoꢀ (2) and 2ꢀbromoꢀ4,4ꢀdiꢀ
methylcyclohexaꢀ2,5ꢀdienꢀ1ꢀone (3), were chosen as obꢀ
jects for the study. The latter two compounds are of speꢀ
cial interest, because they represent the structural fragꢀ
ment of discorhabdine molecules C and E (4 and 5, reꢀ
spectively), which are present in the Lafrunculia sea
sponge. It is known11 that the antimicrobial properties of
natural discorhabdines are due to the ability of the C—Hal
bond to cleave. This cleavage proceeds presumably via
Experimental
Cyclic voltammograms (CV curves) were recorded using an
IPCꢀWin potentiostat at the stationary Pt electrode (d = 5 mm)
against Bu4NBF4 in anhydrous organic solvents (THF, MeCN)
at 20 °С in a special electrochemical cell switched on according
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 5, pp. 977—983, May, 2007.
1066ꢀ5285/07/5605ꢀ1013 © 2007 Springer Science+Business Media, Inc.

1014
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 5, May, 2007
Nikitin et al.
to the threeꢀelectrode scheme. The potential sweep (υ) was varied
from 50 to 500 mV s–1. Electrolyses were carried out using a
Pꢀ5827M potentiostat in a divided 10ꢀmL electrolytic cell; the
working electrode was a graphite tissue 1×1 cm2 in size mounted
on a Pt wire. In all cases, the counter Pt electrode and the
Ag/AgCl/KCl(sat.) reference electrode (potential –0.47 V (in
MeCN) vs. Fc/Fc+) were used. The potential peak values were
recalculated taking into account ohmic losses.
LCꢀMS analysis was carried out on a Finnigan MAT SSQ
7000 instrument (ionization energy 70 eV) equipped with an
HPꢀ5 (30 m) capillary silicone column.
Acetonitrile (pure grade) was stirred for 12 h over CaH2,
distilled, then refluxed for 2 h over P2O5, and distilled again
collecting the fraction with b.p. 81—82 °C (760 Torr); THF
(pure grade) was stirred over KOH, then distilled from LiAlH4,
and stored over sodium benzophenoneketyl. The freshly disꢀ
tilled portion of the solvent was used in each entry.
of the starting dibromoketone) passed through the solution, the
yield of 2ꢀbromoꢀ4,4ꢀdimethylcyclohexaꢀ2,5ꢀdienꢀ1ꢀone was
82% and the efficiency was 71%.
Debromination of αꢀbromoacetophenone (1). Titanocene
dichloride (7.5 mg, 3•10–5 mol) was dissolved in MeCN (10 mL),
and Bu4NBF4 (190 mg, 5•10–5 mol) and αꢀbromoacetophenone
(1) (19.8 mg, 10–4 mol) were added. Preparative electrolysis was
carried out according to the general procedure at a potential of
–0.85 V (vs. Ag/AgCl/KCl) and stopped after 21.4 C of electricꢀ
ity (2.2 F per 1 mole of the initial bromoketone) passed through
the solution. The yield of acetophenone was 54%, and the effiꢀ
ciency was 49%. The fraction of the starting bromoketone 1
was 5%. The reaction mixture also contained the transꢀhalogeꢀ
nation product: αꢀchloroacetophenone (35%).
Results and Discussion
Preparative electrolysis was performed in an appropriate solꢀ
vent in the potentiostatic regime at the reduction potential of
Cp2TiCl2 (–0.85 V) and the concentration of Cp2TiCl2 and
organic halide equal to 3 and 7—10 mmol L–1, respectively. The
electrolysis course was monitored using a digital amperometer
from a change in the current. After the end of electrolysis
(Q ≈ 2 F per 1 mole of organic halide), the solvent was evapoꢀ
rated in vacuo (~100 Torr) and the reaction products were exꢀ
tracted with ether and examined using GLC and LCꢀMS
analyses. The yields and efficiencies were determined as based
on the starting halide.
Debromination of 2ꢀbromoꢀ4,4ꢀdimethylcyclohexaꢀ2,5ꢀdienꢀ
1ꢀone (3). A. Titanocene dichloride (7.5 mg, 3•10–5 mol) was
dissolved in THF (10 mL), and Bu4NBF4 (380 mg, 10–4 mol)
and bromoketone 3 (13.8 mg, 7•10–5 mol) were added. Preꢀ
parative electrolysis was carried out using the general procedure
at a potential of –0.85 V (vs. Ag/AgCl/KCl) and stopped after
15.9 C electricity (2.35 F per 1 mole of the starting bromoketone)
passed through the solution. The yield of 4,4ꢀdimethylcyclohexaꢀ
2,5ꢀdienꢀ1ꢀone was 33%, and the efficiency was 28%. MS, m/z:
122 [M]+, 107 [M+ – Me], 92 [M+ – 2 Me], 64 [M+ – 2 Me –
CO]. The fraction of the unreacted bromide 3 was ~65%.
B. The experiment was carried out analogously but a soluꢀ
tion of Bu4NBF4 (190 mg, 5•10–5 mol) in MeCN was used
as the supporting electrolyte. For an electric charge of 14.2 C
(2.1 F per 1 mole of the initial bromoketone), the yield of
4,4ꢀdimethylcyclohexaꢀ2,5ꢀdienꢀ1ꢀone was 38% and the effiꢀ
ciency was 36%.
Debromination of 2,6ꢀdibromoꢀ4,4ꢀdimethylcyclohexaꢀ2,5ꢀ
dienꢀ1ꢀone (2). A. Titanocene dichloride (7.5 mg, 3•10–5 mmol)
was dissolved in THF (10 mL), and Bu4NBF4 (380 mg, 10–4 mol)
and dibromoketone 2 (22.5 mg, 8•10–5 mol) were added. Preꢀ
parative electrolysis was carried out according to the general
procedure at a potential of –0.85 V (vs. Ag/AgCl/KCl) and
stopped after 15.5 C of electricity (2.05 F per 1 mole of the initial
dibromoketone) passed through the solution. The yield of
2ꢀbromoꢀ4,4ꢀdimethylcyclohexaꢀ2,5ꢀdienꢀ1ꢀone was 52%, and
the efficiency was 51%. MS, m/z: 200 [M]+, 121 [M+ – Br], 106
[M+ – Br – Me], 91 [M+ – Br – 2 Me], 63 [M+ – Br –
2 Me – CO]. The fraction of the unreacted dibromide 2
was ~40%.
Direct electrochemical reduction (ER) of organic haꢀ
lides is a classical example of the dissociative electron
transfer (DET) accompanied by the σꢀbond cleavage. The
DET is a convenient and chemically pure method for the
generation of a reactive radical species due to the oxidaꢀ
tion or reduction process. The DET reactions have first
been considered theoretically.12 To the present time, nuꢀ
merous electronꢀtransfer reactions accompanied by the
cleavage of the C—C, C—O, O—O, C—Hal, and C—S
σꢀbonds, as well as σꢀbonds of other types (see, e.g.,
Refs 13 and 14 and literature cited therein). The dissociaꢀ
tive electron transfer from the donor to acceptor can be
interꢀ or intramolecular and proceed via the concerted or
twoꢀstep mechanism.
RX + e
RX + e
R• + X– (concerted)
RX•– R• + X– (stepwise)
This or another mechanism occurs due to the sum of
many factors,15 the most important of which are the drivꢀ
≠
ing force and the internal activation barrier (∆G0 ). The
latter includes such parameters as the energies of bond
reorganization and dissociation. The driving force is esꢀ
tablished by the comparison of the standard potentials for
the concerted and stepwise mechanisms. The direction of
the process depends on both the substrate nature and
external factors, e.g., temperature or potential.13 The existꢀ
ence of the radical anion, whose lifetime is longer than
the period of bond vibration (~10–13 s), is a necessary but
not sufficient preꢀrequisite for the stepwise mechanism to
occur. In addition to two opposite mechanisms, there are
boundary situations due to which distinctions between
the concerted and stepwise mechanisms become not so
unambiguous. A problem arises how to determine the
criterion for the occurrence of the process via this or
another mechanism. For instance, it was shown13 that a
possible criterion can be the electrochemical transfer coꢀ
efficient α calculated from the experimental data. The
both mechanisms are characterized by the linear depenꢀ
B. The experiment was carried out similarly, but a solution
of Bu4NBF4 (190 mg, 5•10–5 mol) in MeCN was used as the
supporting electrolyte. For 17.9 C of electricity (2.3 F per 1 mole

Electrocatalytic reduction of αꢀbromoketones
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 5, May, 2007
1015
dence of the α value on the potential and, therefore, the
nonlinear region in the plot indicates that the mechanism
has changed. As a rule, for the concerted mechanism
controlled by the slow step of heterogeneous electron
transfer α < 0.5 (Ep is more negative than E0), whereas for
the stepwise mechanism α > 0.5 (Ep is less negative
than E0).15
reduction of 1—3 containing the C(sp3)—Br and
C(sp2)—Br bonds activated by the adjacent carbonyl
group. These reactions are classified as processes
with DET.
Direct electrochemical reduction of bromoketones 1—3.
Two oneꢀelectron redox transitions can be observed in
the CV curves of the direct ER of αꢀbromoacetopheꢀ
none (1). The first oneꢀelectron peak (E1 = –1.18 V) is
related to the fast elimination of the bromide ion, which
is oxidized during the reverse scan at a potential of 0.9 V.
The second peak is attributed to the reduction of the
carbonyl group (E2 = –1.75 V). The data obtained agree
well with published data.17 The direct electrochemical
reduction of compound 1 can be presented by Scheme 1.17
For substituted benzylic halides, the stepwise mechaꢀ
nism of ER occurs only in the case of the 4ꢀnitroꢀ
substituted derivative (α > 0.5); for other compounds of
this series (CNꢀ, OMeꢀsubstituted), the mechanism is
most likely concerted.15 The special properties of the
4ꢀnitrosubstituted substrates were also found16 when the
ER of a series of substituted peroxy benzoates proceeding
with the O—O bond cleavage were studied. It turned out
that the stepwise mechanism was observed only for
4ꢀNO2C6H4CO3But, whereas the unsubstituted peroxides
exhibit the purely concerted DET process; for the
4ꢀCOMeꢀ, 3ꢀNO2ꢀ, and 4ꢀCNꢀsubstituted derivatives, the
transition from the stepwise to concerted mechanism was
found with the shift of the applied potential to the region
of less negative values. It is shown17 in the case of substiꢀ
tuted αꢀhalogenꢀ and αꢀbenzoyloxyacetophenones that
for the good leaving group (Br– anion) the mechanism is
determined by the reactivity of the radical anion formed.
In the case of the electronꢀwithdrawing substituents in
the benzene ring (NO2), the process proceeds via the
stepwise mechanism (α > 0.5), whereas the electronꢀdoꢀ
nating substituents (OMe) are characterized by the conꢀ
certed mechanism of the reaction (α < 0.5). When the
substrate contains the poorly leaving group (F, PhCOO),
the process proceeds via the stepwise mechanism regardꢀ
less of the nature of the substituent in the aromatic ring.
The DET mechanism in electrocatalytic processes is
studied much more poorly. It has recently been shown18
that the electrocatalytic reduction of several benzylic
halides in the presence of [NiI(salen)]– (salen is
N,N´ꢀbis(salicylidene)ꢀ1,2ꢀdiaminoethane) proceeds via
the mechanism of innerꢀsphere DET. It has been menꢀ
tioned earlier9,10,18 that the coordination of TiIII with the
substrate plays an important role in the electrocatalytic
reduction of benzylic halides in the presence of titanocene
dichloride. For nitrobenzyl halides, the process begins
from the coordination of TiIII at the NO2 group of organic
halide, is accompanied by the primary electron transfer to
its π*ꢀorbital (localized on the NO2 group and aromatic
ring), and is followed by the intramolecular electron transꢀ
fer to the σ*ꢀorbital of the C—Hal bond resulting in its
cleavage.10 In the case of unsubstituted benzyl halides,
the coordination of TiIII to the halogen atom (Cl, Br) is
accompanied by the concerted electron transfer to the
σ*ꢀorbital of the C—Hal bond and cleavage of the
latter.10,19
Scheme 1
The presence of a sufficiently acidic αꢀproton in the
molecule of compound 1 induces autoprotolysis, due to
which the number of consumed electrons becomes equal
to two per two substrate molecules. Thus, formally oneꢀ
electron reduction is observed, although the eceꢀprocess
occurs.
Different methods are known for the estimation of
the electrochemical transfer coefficient from the voltꢀ
ammetric data
α = 1.85RT/[F(Ep/2 – Ep)],
dEp/d(logυ) = 1.15RT/αF,
where Ep/2 is the potential at the half of the peak current
value, and F is the Faraday constant (96 500 C mol–1).
The electrochemical transfer coefficient α calculated
from the difference Ep – Ep/2 (see Ref. 20) for comꢀ
pound 1 is equal to 0.17. The estimate based on the
Ep—logυ plot gives a similar value (α = 0.19). This indiꢀ
cates that halogen is eliminated via the concerted mechaꢀ
nism, which agrees with published data.19
The electrochemical reduction of bromocyclohexaꢀ
dienes has not been studied earlier. In the case of
2,6ꢀdibromoꢀ4,4ꢀdimethylcyclohexaꢀ2,5ꢀdienꢀ1ꢀone (2),
In the present work, we compared the processes of
direct and electrocatalytic (in the presence of Cp2TiCl2)

1016
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 5, May, 2007
Nikitin et al.
i/µA
tains no acidic protons, the twoꢀelectron reduction ocꢀ
curs, unlike that for αꢀbromoacetophenone (1). Traces of
water present in the solvent can be a source of protons.
The CV curve for 2ꢀbromoꢀ4,4ꢀdimethylcyclohexaꢀ
2,5ꢀdienꢀ1ꢀone (3) contains by one peak less; however, in
other respects it resembles the curve for dibromocycloꢀ
hexadienone 2. The observed values of potentials of the
reduction peaks are E1 = –1.65 V and E2 = –2.02 V.
As it is known, the C(sp2)—Br bond is substantially
stronger than the C(sp3)—Br bond. However, even with
an increase in the sweep to 500 mV s–1 we could not
detect the peak of radical anion reꢀoxidation in the
CV curve, which indicates its low stability. The bromide
ion is a good leaving group and, therefore, even the presꢀ
ence of the strong electronꢀwithdrawing carbonyl group
does not result in the formation of a stable radical anion.
The α coefficient for dibromocyclohexadienone 2 calꢀ
culated from the data presented in Fig. 2 is equal to 0.25.
The estimate by the Ep – Ep/2 difference (at the sweep
500 mV s–1) gives α = 0.29. This implies that the conꢀ
certed DET mechanism also occurs in this case.
–20
–60
–100
–140
–2000 –1500 –1100 –500
0
E/V
Fig. 1. CV curve of solution of 2,6ꢀdibromoꢀ4,4ꢀdiꢀ
a
methylcyclohexaꢀ2,5ꢀdienꢀ1ꢀone (2) in MeCN (1 mmol L–1
)
,
against 0.05 M Bu4NBF4 (Pt electrode, υ = 100 mV s–1
vs. Ag/AgCl/KCl).
the former two peaks (E1 = –1.48 V, E2 = –1.72 V) are
irreversible and correspond to the consecutive eliminaꢀ
tion of the bromine atoms, and the most negative peak
(E3 = –2.04 V) is reversible to the reduction of 4,4ꢀdiꢀ
methylcyclohexadienone (Fig. 1). The obtained value is
well consistent with the data in Ref. 21, where the DET
reactions of spiro[2.5]octaꢀ4,7ꢀdienꢀ6ꢀone accompanied
by the C—C bond cleavage were studied.
Electrocatalytic reduction of compounds 1—3 in the
presence of titanocene dichloride. The electrocatalytic reꢀ
duction of organic halides by transition metal complexes
was studied22 much more poorly than the direct ER, beꢀ
cause the presence of a catalyst affects strongly the charꢀ
acter of DET. The ER processes using the complexes
of Ni (see Ref. 17 and literature cited therein), Co,23,24
The sequence of the ER steps of 2,6ꢀdibromoꢀ4,4ꢀ
dimethylcyclohexaꢀ2,5ꢀdienꢀ1ꢀone (2) is shown in
Scheme 2. Since the initial molecule of compound 2 conꢀ
TiIII 9,10 Cu,8 Ag,25 and other metals. Unfortunately, the
,
mechanisms of reactions that occur are not always studꢀ
ied in detail.
In the present work, we chose Cp2TiCl2 as the cataꢀ
lyst. Its electrochemical behavior has been studied in rather
detail.19,26—32 Titanocene dichloride is reduced to TiIII at
a potential of –0.85 V (vs. Ag/AgCl/KCl) at the Pt elecꢀ
trode in THF, while in MeCN it is reduced at a potential
Scheme 2
of –0.8 V to form species of three types: [Cp2TiCl2]•–
,
Ep/mV
1675
1650
1625
1600
1575
–1
–0.5
logυ
Fig. 2. Reduction potential of dibromoketone 2 (Ep) as a funcꢀ
tion of the potential sweep (logυ) (correlation coefficient 0.9828).
HSolv is solvent

Electrocatalytic reduction of αꢀbromoketones
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 5, May, 2007
1017
i/µA
Table 1. Results of potentiostatic electrolysis of αꢀbromoketones
1—3 (C = 10 mmol L–1) in the presence of Cp2TiCl2 (C0
3 mmol L–1, E = –0.85 V vs. Ag/AgCl/KCl)
=
10
0
–10
–20
αꢀBromoꢀ Solꢀ
Debromination product (%)
ketone
vent
yield
efficiency
1
2
2
3
3
MeCN
MeCN
THF
MeCN
THF
54
49
–30
82*
52*
38
71*
51*
36
1
–40
2
33
28
–50
–60
3
* For the monodehalogenation product.
–800
–400
0
E/mV
Based on the CV and preparative electrolysis data, the
presumable mechanism of electrocatalytic conversion can
be presented by Scheme 3.
Fig. 3. CV curves for Cp2TiCl2 (1) and Cp2TiCl2 with an addiꢀ
tive of 0.5 (2) and 1 mmol L–1 (3) bromoketone 3 (THF, 0.05 M
Bu4NBF4, υ = 100 mV s–1, Pt electrode, vs. Ag/AgCl/KCl).
Cp2TiCl, and [Cp2TiCl]2, whose relative yield depends
on the concentration of the starting Cp2TiCl2.30 The inꢀ
troduction of Cp2TiCl2 into the reaction mixture makes it
possible to decrease considerably the reduction potential
of haloketones 1—3. The change in the character of
Cp2TiCl2 reduction and TiIII reꢀoxidation in the presence
of 2ꢀbromoꢀ4,4ꢀdimethylcyclohexaꢀ2,5ꢀdienꢀ1ꢀone (3) is
shown in Fig. 3. As it is seen, the addition of bromide 3
results in the successive increase in the cathodic curꢀ
rent of Cp2TiCl2 reduction and the disappearance of
the peak of TiIII reꢀoxidation, indicating the catalytic
process.
Scheme 3
An analogous situation is observed in the case of comꢀ
pound 2: the successive increase in the cathodic current
of Cp2TiCl2 reduction and the disappearance of the peak
of TiIII reꢀoxidation.
It can be seen from the comparison of the reduction
potentials of compounds 2 and 3 that dibromosubstituted
derivative 2 is reduced by 0.17 V more positively than
monobromide 3. The difference of potentials between the
reduction peaks of these compounds and Cp2TiCl2 is 0.48
and 0.65 V, respectively. Since the electrochemical gap
for the dibromosubstituted derivative is smaller, it can be
expected that the electron transfer would be faster in this
case. The preparative electrolysis data for compounds 2
and 3 in the presence of Cp2TiCl2 confirm this assumpꢀ
tion. The yield of the debromination product is much
higher in the case of the dibromosubstituted derivative
(Table 1).
As indicated by the preparative electrolysis data, the
mediator ER process allows one to eliminate selectively
only one bromine atom in ketone 2. The halogen atom
in product 3 begins to eliminate to a noticeable extent
only after the complete consumption of dibromide 2,
because the reaction rate is substantially higher in the
latter case.
Although the TiIII complexes are strong reducing
agents, a possibility of the parallel process of interaction
of paramagnetic Cp2TiIIICl with organic radicals to

1018
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 5, May, 2007
Nikitin et al.
form an organotitanium compound cannot be excluded
(Scheme 4). However, when the reaction products are
isolated from the reaction mixture, the complex formed
can decompose to yield the same dehalogenation product.
tential for αꢀbromoacetophenone is E0(RBr/RBr•–) =
1.76—2.01 V and, hence, the outerꢀsphere electron transꢀ
fer from the coordinately saturated radical anion to orꢀ
ganic halide (Scheme 6) is lowly probable.
Scheme 6
Scheme 4
It can be assumed that the innerꢀsphere reaction rate
is higher, especially if taking into account the high affinity
of titanium to oxygen (Scheme 7).
Scheme 7
Electrocatalytic debromination can be carried out for
compound 1 as well. When the latter is added to a soluꢀ
tion containing Cp2TiCl2, the CV curve exhibits an inꢀ
crease in the peak corresponding to the reduction of TiIV
to TiIII and a decrease in the peak of TiIII reꢀoxidation,
which is characteristic of the catalytic process. The preꢀ
parative electrolysis under these conditions gave acetoꢀ
phenone in 54% yield (see Table 1). The assumed mechaꢀ
nism of redox catalysis can be presented by Scheme 5.
Principally, a possibility of the formation of organoꢀ
titanium intermediates (see Scheme 4) of the type
PhC(O)CH2TiCp2Cl or PhC(=CH2)OTiCp2Cl cannot be
excluded.
Thus, the data obtained indicate that the
Cp2TiCl2/Cp2TiCl redox system can serve as a catalyst
for the reductive dehalogenation of αꢀhaloketones. The
reaction occurs at low cathodic potential (–0.85 V
vs. Ag/AgCl/KCl(sat.)) and affords the target debrominaꢀ
tion products in a preparative yield (40—80%).
Scheme 5
It should be mentioned in conclusion that dissociative
electron transfer often occurs in the natural catalytic proꢀ
cesses.2,13,33 Intramolecular DET from the donor to acꢀ
ceptor in the natural systems occurs at rather long disꢀ
tances (>10 Å) and very rapidly (105 s–1), although the
driving force of such reactions is low, as a rule (only
~0.1 eV). This is probably achieved due to the geometry
optimization of the reacting particles and a great role of
the intramolecular interactions (for instance, hydrogen
bonds) in a protein molecule. Our data indicate that
bromoketones 2 and 3 containing the C(sp2)—Br bond,
which is activated by the adjacent electronꢀwithdrawꢀ
ing group, easily undergo reductive dehalogenation at
rather low (close to "biological") potential values. Perꢀ
haps, this process models the biological effect of natural
discorhabdines C and E containing the fragments of the
indicated ketones.
Since the solution of electrochemically reduced
Cp2TiCl2 contains both [Cp2TiCl2]•– and Cp2TiCl or its
dimer,30 different forms of TiIII can act as reducing agents.
However, according to the estimate,17 the standard poꢀ
This work was financially supported by the Russian
Foundation for Basic Research (Project No. 05ꢀ03ꢀ
32759).

Electrocatalytic reduction of αꢀbromoketones
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 5, May, 2007
1019
17. A. Gennaro, A. A. Isse, and F. Maran, J. Electroanal. Chem.,
2001, 507, 124.
References
18. R. J. Enemerke, J. Larsen, T. Skrydstrup, and K. Daasbjerg,
J. Am. Chem. Soc., 2004, 126, 7853.
19. C. Andreux, J.ꢀM. Saveant, A. Tallec, R. Tardivel, and
C. Tardy, J. Am. Chem. Soc., 1997, 119, 2420.
20. L. Nadjo and J.ꢀM. Saveant, J. Electroanal. Chem., 1973,
48, 113.
21. J. P. Phillips, J. G. Gillmore, P. Schwartz, L. E. Brammer,
J. D. Berger, and J. M. Tanko, J. Am. Chem. Soc., 1998,
120, 195.
22. J.ꢀM. Saveant, Adv. Phys. Org. Chem., 1990, 26.
23. A. Isse, A. Gennaro, and E. Vianello, J. Electroanal. Chem.,
1998, 444, 241.
24. A. Isse, A. Gennaro, and E. Vianello, J. Chem. Soc., Dalton
Trans., 1996, 1613.
25. A. A. Isse, L. Falciola, P. R. Mussini, and A. Gennaro,
Chem. Commun., 2006, 344.
26. Y. Mugnier, C. Moise, and E. Laviron, J. Organomet. Chem.,
1981, 204, 61.
27. N. El Murr and A. Chaloyard, J. Organomet. Chem., 1981,
212, 39.
1. M. Baizer, Organic Electrochemistry. An Introduction and
a Guide, Marcel Dekker, New York, 1973.
2. T. V. Magdesieva, Ross. Khim. Zh., 2005, XLIX, № 5, 56
[Mendeleev Chem. J., 2005, XLIX, No. 5 (Engl. Transl.)].
3. U. E. Krone, R. K. Thauer, and H. P. C. Hogenkamp,
Biochemistry, 1989, 28, 4908.
4. G. Hollinger, G. Wohlfarth, and G. Dukert, FEMS Microbiol.
Res., 1998, 22, 383.
5. S. P. J. Albracht, D. AnkelꢀFuchs, J. W. Van der Zwaan,
R. D. Fontijn, and R. K. Thauer, Biochem. Biophys. Acta,
1986, 870, 50.
6. G. K. Lahirl, L. J. Schussel, and A. M. Stiozenberg, Inorg.
Chem., 1992, 31, 4991.
7. M. W. Renner, L. R. Furenlid, K. M. Barkigia, A. Forman,
H.ꢀK. Shim, D. J. Simpson, K. M. Smith, and J. Fajer,
J. Am. Chem. Soc., 1991, 113, 6891.
8. T. Osako, K. D. Karlin, and S. Itoh, Inorg. Chem., 2005,
29, 381.
9. T. V. Magdesieva, M. Grachik, A. Vallat, O. M. Nikitin,
P. I. Dem´yanov, and M. A. Vorotyntsev, Izv. Akad. Nauk,
Ser. Khim., 2005, 196 [Russ. Chem. Bull., Int. Ed., 2005,
54, 201].
28. S. P. Gubin and S. A. Smirnova, J. Organomet. Chem., 1969,
20, 229.
29. S. V. Kukharenko, G. L. Soloveichik, and V. V. Strelets,
Metalloorg. Khim., 1989, 2, 395 [Organomet. Chem. USSR,
1989, 2 (Engl. Transl.)].
30. E. Samuel and J. Vedel, Organometallics, 1989, 8, 237.
31. R. J. Enemaerke, J. Larsen, T. Skrydstrup, and K. Daasbjerg,
Organometallics, 2004, 23, 1866.
10. T. V. Magdesieva, M. Graczyk, A. Vallat, O. M. Nikitin,
P. I. Demyanov, K. P. Butin, and M. A. Vorotyntsev,
Electrochim. Acta, 2006, 52, 1265.
11. J. W. Blunt, V. L. Calder, and G. D. Fenwick, J. Nat. Prod.,
1987, 50, 290.
12. J. M. Saveant, J. Am. Chem. Soc., 1987, 109, 6788.
13. S. Antonello and F. Maran, Chem. Soc. Rev., 2005, 34, 418.
14. F. Maran and M. S. Wokrentin, Interface, 2002, 11, 44.
15. C. Andreux, A. Gorande, and J.ꢀM. Saveant, J. Am. Chem.
Soc., 1992, 114, 6892.
32. R. J. Enemaerke, J. Larsen, T. Skrydstrup, and G. Hjolland,
Organometallics, 2005, 24, 1252.
33. A. G. Griesbeck, T. Heinrich, M. Oelgemoller, J. Lex, and
A. Molis, J. Am. Chem. Soc., 2002, 124, 10972.
16. S. Antonello and F. Maran, J. Am. Chem. Soc., 1999,
121, 9668.
Received December 11, 2006;
in revised form February 26, 2007