The generation of aryl anions by double electron transfer to aryl iodides from a neutral ground-state organic super-electron donor

Article abstract of DOI:10.1002/anie.200700554

(Chemical Equation Presented) It takes two to cyclize: Aryl halides are reduced to aryl anions by double electron transfer from the neutral ground-state electron donor 1 (see scheme), as shown by the formation of a cyclic ketone (2). The reduced compound (3) is also formed. Calculations show that the loss of two electrons from 1 is both thermodynamically and kinetically viable and generates a more planar resonance-stabilized structure.

Full text of DOI:10.1002/anie.200700554

Communications
DOI: 10.1002/anie.200700554
Electron Transfer
The Generation of Aryl Anions by Double Electron Transfer to Aryl
Iodides from a Neutral Ground-State Organic Super-Electron Donor**
John A. Murphy,* Sheng-ze Zhou, Douglas W. Thomson, Franziska Schoenebeck,
Mohan Mahesh, Stuart R. Park, Tell Tuttle,* and Leonard E. A. Berlouis
Stabilized carbanions can be prepared by
deprotonation of the corresponding
carbon acids, but more reactive carban-
ions, such as aryl anions, are usually made
by metal–halogen exchange or by elec-
[
1]
trochemical routes. Whereas the use of
metals and organometallic compounds to
form aryl anions is well established, the
use of totally organic electron-transfer
agents, in the absence of photochemical
[
2]
activation, is completely unknown.
Herein we report the first neutral organic
molecule to achieve this feat.
[
3]
We recently reported the reactions
of aryl halides with the fully character-
ized electron donor 1 (Scheme 1). Com-
pound 1 is one of a series of compounds
[
4]
that had been prepared previously for
the purpose of studying their electro-
chemical properties; their redox poten-
tials had shown that they are strong
reducing agents and that they can
Scheme 1. Formation of radicals with benzimidazole-derived donor 1. Ms=methanesulfonyl.
[
4]
undergo sequential loss of two electrons
reduction potentials for the resulting
dication 8 and radical cation 7 are E (DMF) = ꢀ0.76 and
(
[
4a]
[5]
transfer of a single electron. Recent investigations have
1
/2
ꢀ
0.82 V, respectively, versus the saturated calomel electrode
SCE)); thus they react spontaneously in air. However, their
reactivity with organic molecules had not been tested
elucidated the standard reduction potential for the conversion
0
(
of aryl radicals into aryl anions (E = + 0.05 V versus SCE for
the parent phenyl case), and so we wondered whether aryl
[
3]
previously. Treatment of donor 1 with aryl iodides 2 resulted
in efficient cyclizations to indolines 4, which is consistent with
the intermediacy of aryl radicals 3, and, therefore, with the
anions 5 could be formed under our conditions. Cyclization of
[
6]
aryl anions onto unactivated alkenes has been reported, but
could anion 5 be contributing to the yields of indolines 4 in
our reactions? The addition of aryl anions to unactivated
alkenes is reported only to occur in systems where no
alternative reactions, such as deprotonation of acidic sites or
attack upon conventional electrophilic groups, are possible. In
our case, aryl anions 5 would be expected to preferentially
deprotonate the radical cation 7 or dication 8 to give arene 6,
or alternatively, undergo nucleophilic attack on these species
or on the solvent (DMF). However, arene 6 was not observed,
and no products from the nucleophilic attack by aryl anions 5
were seen. This finding suggested that anions were not formed
from the reaction of donor 1 with aryl halides.
[
*] Prof. Dr. J. A. Murphy, Dr. S. Z. Zhou, D. W. Thomson,
F. Schoenebeck, Dr. M. Mahesh, S. R. Park, Dr. T. Tuttle,
Dr. L. E. A. Berlouis
WestCHEM
Department of Pure & Applied Chemistry
University of Strathclyde
295 Cathedral Street, Glasgow, G11XL (UK)
Fax: (+44)548-4246
E-mail: john.murphy@strath.ac.uk
tell.tuttle@strath.ac.uk
[**] We thank the EPRSC (S.Z.Z., D.W.T., F.S., and S.R.P.), the CVCP
Nevertheless, to be clear on this point and also prior to
testing for the formation of aryl anions with other electron
donors, a diagnostic test was needed for aryl anions. A
reaction therefore needed to be identified that would
unambiguously distinguish aryl anions from aryl radicals. It
is well known that reactive anions, such as aryl anions, attack
(
Universities UK), and the University of Strathclyde (M.M.) for
funding, the EPRSC National Mass Spectrometry Service Centre,
Swansea for mass spectra, and Dr. K. Daasbjerg for helpful
discussions.
Supporting information for this article (experimental details) is
available on the WWW under http://www.angewandte.org or from
the author.
[
7]
esters, while carbon radicals do not. Hence, the iodoester 9
5
178
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 5178 –5183

Angewandte
Chemie
was selected as a test substrate (Scheme 2). This system was
tested firstly with a known procedure for forming aryl
radicals. Heating 9 in the presence of tris(trimethylsilylsilane)
Substrate 9 was then tested with donor 1 (2.5 equiv) in
DMF (Scheme 3). No indanone 10 was produced, but the
reduced compound 11 was isolated (67%). The absence of
indanone indicated that aryl anions had not been formed.
(12) and the radical initiator azobisisobutyronitrile (AIBN) in
Scheme 2. Different outcomes of aryl anion and aryl radical reactions
derived from iodide 9.
toluene, gave exclusively the reduced product 11 (70% yield).
No trace of indanone 10 was observed; this result is totally
consistent with the expected intermediacy of aryl radicals.
Iodoester 9 was then tested with a known method for the
formation of aryl anions. Reaction of trimethyl(tributylstan-
[
8]
nyl)silane (13) with fluoride ions (CsF) in the presence of 9
led to the formation of the indanone 10 (68%) together with
the reduced compound 11 (14%). The presence of the
indanone clearly indicated that formation of an aryl anion
had occurred. Formation of 11 likely also arose from the aryl
anion, with quenching of the anion resulting from minute
traces of water in the dried reagents, solvents, or glassware.
These two reactions therefore establish the validity of using 9
for illustrating the presence of aryl anions; the formation of
indanone product 10 gives unambiguous evidence of the
presence of aryl anions in any test reaction that we performed.
Interestingly, attempts to access ketone 10 through other
reactions that should afford aryl anions, by using the reagents
tBuLi (acting as a nucleophile on the IꢀAr bond) as well as
Scheme 3. Reactions of iodoaryl esters with donors 1 and 22.
Thus, despite the experimentally determined redox potentials
for 1 (E1/2 = ꢀ0.76, ꢀ0.82 V versus SCE, as stated above)
being much more negative than the standard potential for the
0
reduction of an aryl radical (E = + 0.05 V versus SCE for the
phenyl radical), the reaction does not proceed. This finding
can be attributed to the difference between the standard
potential and the actual potential for the reduction of an aryl
radical to an aryl anion under our experimental conditions;
Andrieux and Pinson sawthe peak attributed to this
reduction at ꢀ0.64 V versus SCE under the conditions used
sodium naphthenide or magnesium metal (both as electron
donors), were all unsuccessful, and afforded complex mix-
tures of products that included, at best, minute traces of
indanone 10. In effect, these reagents are too reactive; indeed
the reaction of tBuLi with indanone has already been
[
5]
in their experiments. Under our conditions, we assume that
the actual potential may be even more negative. To check this,
we selected a more powerful reducing agent from the same
class of organic donors as 1.
[
9]
[4a,b,c]
reported. Furthermore, cyclic voltammetry indicated that
Electrochemical reduction of dication 24 (X = Br)
ketone 10 (E (DMF) = ꢀ2.02 V versus SCE) is more easily
affords 22 (Scheme 4), which is one of the more powerful
electron donors in this class (reduction of 24 (X = Br):
1
/2
reduced than iodoester 9 (E (DMF) = ꢀ2.14 V versus SCE
p
[
4a]
(
see the Supporting information; the Ag/AgCl electrode
E (MeCN) = ꢀ1.18 V and ꢀ1.37 V (ir) versus SCE,
1
/2
potential is 0.04 V negative of SCE), and hence it is not
surprising that the ketone was not stable in the presence of
where “ir” represents an irreversible process; E -
1
/2
[
4a]
[4b]
(DMF) = ꢀ1.20 V versus SCE; and E (MeCN)
=
1
/2
0
such strong electron donors as sodium naphthenide (E =
ꢀ1.12 V and ꢀ1.28 V (ir) versus SCE).
[
10]
ꢀ
2.47 V versus SCE)
and magnesium metal (known to
Since our work would require significant quantities of 22,
[
11]
produce pinacol reduction of ketones).
a direct synthetic route to it was required, and none was
Angew. Chem. Int. Ed. 2007, 46, 5178 –5183ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5179

Communications
electron transfer from donor 22, and indicate that 51% and
5% represent minimum percentages for the formation of the
anion.
The difference in the reactivity of 1 and 22 merits some
4
comment. The simplest observation is that the corresponding
dications 8 and 24, which result from the loss of two electrons,
feature two new heterocyclic aromatic rings (shown in bold in
Scheme 5). The aromatic stabilization energy, which is fully
Scheme 4. Preparation of imidazole-derived donor 22.
available, although a scheme for such a transformation was
[
4c]
[12]
reported by Taton and Chen, while Shi et al. refer to
failed attempts to generate 22 from the dibromide analogue
of 21.
We conveniently prepared diiodide 21 (Scheme 4) from
imidazole and diiodopropane. In this case, the desired donor
2
2 was formed from 21 using NaH as a base with liquid
Scheme 5. The aromaticity of dications 8 and 24.
ammonia as the solvent. Liquid ammonia was selected as it
affords a way to generate pure donor 22. Thus, following the
generation of the donor 22, evaporation of the ammonia
afforded a solid residue that was easily extracted with diethyl
ether, thereby affording donor 22 (98%) as an air-sensitive
pure yellowsolid. Ho we ver, it is not necessary to use this
route to prepare 22; it can alternatively be prepared in situ
from 21 using DMF as the solvent. This procedure was used
for the reactions shown in Scheme 6.
present only in the dications, explains why the loss of the
second electron from 1 and 22 is not much more difficult than
the first. In the case of the dication 8, the newaromatic
heterocyclic rings are fused with existing benzene rings,
whereas for dication 24 the newimidazolium rings are not
fused, and, as such, impart greater aromatic stabilization
energy to that dication.
Reaction of the substrate 9 with donor 22 (Scheme 3)
afforded two products, the indanone 10 (16%) and the
reduced compound 11 (70%). The indanone 10 provides the
required positive proof of the intermediacy of an aryl anion
for the first time in these reactions, and so 22 is the first
neutral organic reagent to achieve such a transformation, the
hallmark of the super-electron donor (SED) sought by us and
others in recent years. The isolation of the reduced compound
We were keen to compare the equilibrium geometries of
the relevant dications at a deeper level (Figure 1). The NMR
spectra of dications 8 and 24 were notably different: the
NCH protons of the trimethylene bridge in dication 8 were
2
diastereotopic (d = 4.52–4.61 (2H, m, CH ), 5.25–5.32 ppm
2
(2H, m, CH )), indicative of a rigid helical twist, whereas the
2
NCH protons for the dication of 22 were a simple triplet (d =
2
4.61 ppm (8H, t, J = 6.0, 4 ” CH )), thus indicating a freely
2
1
1 could reflect the intermediacy of aryl anion 14 and/or the
moving segment and no helicity. As shown below, this is borne
out by computational studies (for full details of the optimi-
zation procedure see the Supporting Information).
corresponding aryl radical. The aryl anion could abstract a
proton from the radical-cation 23 or dication 24 while the
putative aryl radical could abstract a hydrogen atom from the
same sites. The formation of the indanone 10 in 16% yield
thus reflects a minimum amount of aryl anion formed in this
reaction.
Loss of two electrons from 22 results in the more planar
structure 24. The central CꢀC bond is increased in length
from 1.36 to 1.44 Š, which indicates the formation of a
resonance-stabilized structure. The donor molecule 1 also
experiences the formation of a resonance-stabilized structure
with a similar increase in bond length. However, in contrast to
If the aryl anion 14 derived from substrate 9 affords the
reduced product 11, then this protonation reaction is in
competition with the cyclization to form indanone 10. Tw o 22, the overall structure of 1 becomes increasingly less planar
further ester substrates, 15 and 18, were therefore selected
that might lead to faster cyclizations, and hence to a more
accurate picture of the percentage of aryl anion formed in
these reactions with aryl halides. The reaction of 15 and 18
with donor 22 again afforded cyclization in both cases: 15
yielded ketone 16 (51%) as well as reduced product 17 (21%)
as an inseparable mixture, with yields determined by cali-
as electrons are removed. The angle t between the planes of
the benzimidazole rings increases from 168 in 1 to 428 in 8 (the
dication derived from 1) whereas in 22 the angle t between
the planes of the imidazole rings is decreased from 108 to 1.58
on formation of the dication 24 (Figure 1). The larger changes
observed in the structure of 1 indicate that the electronic
structure of the neutral species is unable to easily accom-
modate the formation of positive charge, which in turn
suggests larger reorganization energies for this molecule. To
test this theory, we have calculated the internal reorganization
1
brated H NMR spectroscopy, while iodide 18 led to the
isolation of xanthone (19, 45%) and reduced product 20
(
49%). These results confirm the generation of aryl anions by
5
180
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 5178 –5183

Angewandte
Chemie
Table 1: The component reorganization energies and the half-reaction
[
a]
free energies DG for the donor (D) and anion (A) compounds.
c
ꢀ1
Energies are in kcalmol
.
+
ꢀ
+
ꢀ
2+
2ꢀ
DA!D C+A C
D C+A C!D +A
DGc
li
DG
li
c
A=IB
D=1
D=22
66.1
17.4
23.6
ꢀ66.9
79.0
69.1
52.7
11.8
12.5
ꢀ85.5
93.8
83.0
+
+
2+
ꢀC
ꢀC
2ꢀ
[
a] The half-reactions are D!D C; D C!D ; A!A ; A !A
.
Perhaps surprisingly, the sum of the component internal
reorganization energies for the reactions involving 22 are
greater than those involving 1 (that is, l [Eq. (1)] = 44.8 kcal
i
ꢀ
1
ꢀ1
mol ; l [Eq. (2)] = 32.6 kcalmol ; li[Eq. (3)] = 41.7 kcal
mol ; and l [Eq. (4)] = 32.3 kcalmol ). However, the for-
i
ꢀ
1
ꢀ1
i
mation of the positive charge in 1 is approximately 10 kcal
ꢀ
1
mol less favorable than in 22, and as such the reactions
involving 1 are more endergonic (Table 2).
Table 2: Activation and reaction free energies for the reactions described
Figure 1. Optimized structures of donor molecules 22, 23, 24, 1, 7,
and 8. The central CꢀC bond length is given in Š. t is the angle
ꢀ1
in Equations (1)–(4). All energies are in kcalmol
.
*
between the planes formed by the imidazole/benzimidazole rings in
degrees.
Equation
DG
DG
R
(
(
(
(
1)
2)
3)
4)
12.3
6.9
17.4
12.8
2.2
ꢀ2.5
12.1
8.3
energies of 1 and 22 for the radical cation and dication of each
compound.
The internal reorganization energy was calculated using
Equations (1)–(4), where iodobenzene (IB) was used as a
model acceptor compound:
[
16]
As a first approximation, the activation energy can be
calculated from Marcus theory [Eq. (7)] using the internal
reorganization energies and the reaction free energies.
2
2
1
7
2 þ IB ! 23þC þ IBꢀC
ð1Þ
ð2Þ
ð3Þ
ð4Þ
þC
þ IBꢀC ! 242þ þ IB2ꢀ
ꢀ
ꢁ
3
2
l
i
DGR
li
DG* ¼ 4 1 þ
ð7Þ
þ IB ! 7þC þ IBꢀC
The activation energies for Equations (1) and (2)
Table 2) are both sufficiently lowto explain the experimen-
þC
þ IBꢀC ! 82þ þ IB2ꢀ
(
tally observed activity. In the overall reaction of 22 + IB!
The internal reorganization energy is calculated as, for
example, in Equation (1), with Equations (5) and (6):
2
+
2ꢀ
2
4
+ IB , the first electron transfer is the rate-limiting step
and the reaction is mildly exergonic. Thus, with 22 the
electron transfer is both thermodynamically and kinetically
viable.
l
i
ð22Þ þ l
i
ðIBÞ
l
i
¼
ð5Þ
2
The electron-transfer reactions involving 1 are less
favorable. The initial electron transfer encounters a larger
energy barrier, relative to that in 22; however, under our
experimental conditions this transfer could occur. A compar-
ison of Equations (1) and (2) relating to 22 with Equations (3)
and (4) relating to 1, shows that both the energy barriers to
reaction and the free energy changes are more favorable with
donor 22, which is in line with our experimental findings.
The maintenance of planarity during electron loss from 22
may be a factor assisting its ability to act as a very strong two-
electron donor. The flat structure would be able to interact
closely with the p system of the acceptor arene and assist with
stabilization through formation of an intimate complex.
Given that the reduction potential of aryl iodides is about
where:
l ð22Þ ¼ ðE ðR ÞꢀE ðR ÞÞ þ ðE ðR ÞꢀE ðR ÞÞ
ð6Þ
i
N
C
N
N
C
N
C
C
E (R ) is the energy of the neutral species in the geometry
N
C
of the radical cation; E (R ) is the energy of the neutral
species in its optimized geometry; E (R ) is the energy of the
cationic species in the geometry of the neutral compound; and
N
N
C
N
E (R ) is the energy of the cationic species in its optimized
C
C
geometry. The analogous process is used to calculate l (IB)
Table 1). This procedure, with similar methodology to that
employed here, has previously been shown to provide reliable
results for the calculation of reorganization energies.
i
(
[
13–15]
Angew. Chem. Int. Ed. 2007, 46, 5178 –5183ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5181

Communications
ꢀ
2.2 V versus SCE, the stabilization achieved through such
complexes is expected to be crucial.
slightly easier (E1/2 = ꢀ2.02 V versus SCE), and yet is not
observed. This phenomenon is commonly encountered in
[
10]
The above differences between donors 1 and 22 should
also be evident in other aspects of reactivity, and we were
keen to compare the donors in their reactions with a range of
challenging substrates (Scheme 6). Reaction of 1 with 9-
mediated electrochemistry, but is perhaps not well recog-
nized in other areas of synthetic chemistry. Organic and
inorganic mediators can facilitate redox events such as the
reduction of aryl halides in solution; this arises through
formation of intimate complexes and through ion pairing,
which is not readily available at electrodes. Thus, for example,
indirect electrochemical reduction of aryl halides such as
0
chlorobenzene (E = ꢀ2.78 V) can be achieved in benzonitrile
0
(
E = ꢀ2.44 V) as the solvent by use of appropriate media-
[
18]
tors. A key feature of such reactions is that they feature an
easily and irreversibly cleavable bond (for example, CꢀX,
where X = halogen).
0
Unlike sodium naphthenide (E = ꢀ2.46 V versus SCE)
and magnesium, donor 22, (E1 = ꢀ1.20 V versus SCE) does
/2
not reduce ketones, and this selectivity may later be used to
advantage in synthetic chemistry. The easy availability and
tunability of donors such as 22 opens the door for the design
of highly selective but powerful electron-transfer reagents.
We are currently exploring the possibility of related donors
partaking in a catalytic cycle.
Experimental Section
22: 1,1’:3,3’-Bis(trimethylene)bis(imidazolium) diiodide (21; 10 g,
Scheme 6. Reactions of aryl bromides and chlorides with donor 22.
21.184 mmol, 1.0 equiv) was placed in a Schlenk flask under argon
and dried in vacuo at 1008C for 2 h. After cooling the mixture to RT,
it was purged with argon gas, and then sodium hydride (6.779 g of
bromophenanthrene (25) in DMF at 1008C gave very poor
conversion. Even when 5 equivalents of donor were used,
only 7% of phenanthrene (26) was produced. However,
phenanthrene (26) was formed in 96% yield when 1.5 equiv-
alents of donor 22 were used in DMF at 1008C. Similarly, 1
was completely ineffective in reducing 1-bromonaphthalene
60% dispersion of NaH in mineral oil, 169.47 mmol, 8.0 equiv) added.
The mixture was washed with dry hexane (3 ” 80 mL) in an argon
atmosphere, and a dry-ice condenser was attached to the flask. The
residual hexane in the reaction mixture was removed under vacuum
and the system was back-filled with argon gas. Ammonia (150 mL)
was condensed into the flask while a steady flow of argon gas was
maintained at all times during the course of the reaction. The stirred
suspension turned yellowas the ammonia was refluxed at RT for 2 h;
the ammonia was then evaporated slowly, and the flask transported
into a glove box under nitrogen. The yellowsolid residue was then
extracted with deoxygenated dry diethyl ether (3 ” 80 mL). The
yellow suspension was filtered and the filtrate was evaporated to
afford a yellowsolid. This wa s dried in vacuo to obtain 22 (4.47 g,
(27) to naphthalene (28; 0% yield). In contrast, 22 gave an
8
9
6% yield of naphthalene from this substrate. The chlorides
-chloroanthracene (29) and 2-chloroanthracene (31) were
also reduced to anthracene (30; 99% and 97%, respectively)
by 22 under these conditions.
In summary, donor 22 shows unprecedented reactivity for
a neutral, ground-state organic electron donor, which reduces
aryl halides to aryl anions. The formation of cyclic ketones 10,
9
8%) as a yellow solid which was stored at RT under nitrogen.
1
H NMR (400 MHz, C D ): d = 1.38–1.43 (4H, m, 2 ” CH ), 2.43–2.45
6
6
2
1
3
(
8H, m, 4 ” CH ), 5.48 ppm (4H, s, 4 ” C=CH); C NMR (100 MHz,
2
1
6, and 19 demonstrates that aryl iodides are reduced to aryl
[10]
C D ): d = 31.49 (CH ), 54.37 (CH ), 120.37 (C), 127.62 ppm (CH).
6
6
2
2
anions by 22 at room temperature. Aryl bromides and
chlorides are reduced by 22 at higher temperature. The very
different behavior of 22 and 1 demonstrates a sharp demar-
cation between the reductive capabilities of these reagents
with aryl halides.
Atomic absorption (AAS) analysis: 0.005 mg of Na in 100.0 mg of
tetraazafulvalene 22 (0.047 mol%).
(0.1632 V) versus Ag/AgCl/KCl(sat.), averaged over 5 runs; condi-
E1/2(DMF) = ꢀ1.1072 V
tions: 22 (10.463 mm), Bu NPF (49.917 mm), Pt working and counter
4
6
ꢀ
1
electrodes, 50 mVs
.
Cyclization with 22: A solution of methyl 2-(2-iodophenoxy)-
benzoate (18; 0.0992 g, 0.28 mmol, 1.0 equiv) dissolved in dry DMF
Interestingly, our finding that potentials more negative
than ꢀ1.0 V are needed for the formation of aryl anions is
(15 mL, deoxygenated) was treated with tetraazafulvalene 22
[
17]
mirrored in a recent report by Otero et al. These reductions
do not of course reflect the conditions that lead to the
estimation of standard potentials. However, this illustrates
that the use of reduction potentials, determined from electro-
chemical procedures, to determine the feasibility of chemical
reactions, can be misleading. Thus, 22 should be too weak a
(0.097 g, 0.448 mmol, 1.6 equiv) at RT for 18 h inside a glove box in
a nitrogen atmosphere. The reaction was stopped by the addition of
hydrochloric acid (20 mL, 2m) and the mixture was extracted with
diethyl ether (3 ” 20 mL). The combined organic phases were further
washed with hydrochloric acid (4 ” 20 mL, 2m), followed by a mixture
of brine (20 mL) and hydrochloric acid (10 mL, 2m) solution. The
ethereal layer was separated, dried over anhydrous sodium sulfate,
filtered, and evaporated to dryness. The residue was purified by
column chromatography on silica gel (eluant: toluene/dichlorome-
donor, by about 1 V, to reduce iodoarenes (E = ꢀ2.2 V), and
p
yet reduces these substrates very easily. On the other hand,
[
19]
the reduction of ketones, such as indanone 10, should be
thane/petroleum ether 1:1:3) to give methyl 2-phenoxybenzoate
5
182
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 5178 –5183

Angewandte
Chemie
[
20]
(
20) as a colorless liquid (32 mg, 49%) and xanthone (19) as a white
[7] For exceptions, triggered by the presence of organotitanium
+
solid (25 mg, 45%). 20: HRMS (ESI): m/z 229.0859 [M+H] ;
C H O requires: [M+H] 229.0859; IR (neat film, NaCl): n˜ =
H NMR (400 MHz,
CDCl ): d = 3.89 (3H, s, CH ), 7.04–7.09 (3H, m, ArH), 7.09–7.19
reagents, see a) A. Fernandez-Mateos, P. H. Teijon, R. R.
Clemente, R. R. Gonzalez, Tetrahedron Lett. 2006, 47, 7755 –
7758; b) Y. Yamamoto, D. Matsumi, R. Hattori, K. Itoh, J. Org.
Chem. 1999, 64, 3224 – 3229.
+
1
4
12
3
ꢀ
1
1
3
068, 3039, 2951, 1732, 1603, 1483 cm
;
3
3
(
(
1
1H, m, ArH), 7.27 (1H, ddd, J = 7.6, 7.6, 1.1 Hz, ArH), 7.39–7.43
[8] a) K. R. Wursthorn, H. G. Kuivila, G. F. Smith, J. Am. Chem.
Soc. 1978, 100, 2779 – 2789; b) M. Mori, N. Isono, N. Kaneta, M.
Shibasaki, J. Org. Chem. 1993, 58, 2972 – 2976; c) M. Mori, N.
Kaneta, N. Isono, M. Shibasaki, J. Organomet. Chem. 1993, 455,
255 – 260; d) M. Mori, N. Kaneta, M. Shibasaki, J. Organomet.
Chem. 1994, 464, 35 – 40.
[9] A. Orliac-Le Moing, J. Delaunay, A. Lebouc, J. Simonet,
Tetrahedron 1985, 41, 4483 – 4493.
[10] J. Simonet, J. F. Pilard in Organic Electrochemistry (Eds.: H.
Lund, O. Hammerich), Marcel Dekker, NewYork, 4th ed., 2001,
p. 1171.
2H, m, ArH), 7.52–7.57 (1H, m, ArH), 8.01 ppm (1H, dd, J = 7.8,
1
3
.8 Hz, ArH); C NMR (100 MHz, CDCl ): d = 52.3 (CH ), 118.4
3
3
(CH), 121.1 (CH), 123.3 (CH), 123.4 (C), 123.7 (CH), 129.9 (CH),
1
32.0 (CH), 133.8 (CH), 156.4 (C), 157.8 (C), 166.4 ppm (C); EIMS:
+
m/z 228 (M , 76%), 197 (100), 168 (28), 108 (57), 77 (78). 19:
m.p. 168–1708C (lit.
1
[
20]
m.p. 172–1738C); HRMS (ESI): found:
+
+
97.0596 [M+H] , C H O requires: 197.0597 [M+H] ; IR (KBr):
13 8 2
ꢀ
1
1
n˜ = 3054, 2987, 1655, 1609 cm
;
H NMR (400 MHz, CDCl ): d =
3
7
.38–7.42 (2H, m, ArH), 7.51 (2H, d, J = 8.3 Hz, ArH), 7.72–7.76
1
3
(
(
(
2H, m, ArH), 8.36 ppm (2H, dd, J = 7.9, 1.6 Hz, ArH); C NMR
100 MHz, CDCl ): d = 118.2 (CH), 122.1 (C), 124.1 (CH), 127.0
CH), 135.0 (CH), 156.5 (C), 177.5 (C); EIMS: m/z 196 (M , 100%),
[11] J. March in Advanced Organic Chemistry Reactions, Mechanisms
and Structure, 3rd ed., Wiley-Interscience, NewYork, 1985,
p. 1110 – 1112.
3
+
1
68 (59), 139 (53), 92 (11), 74 (17), 63 (25).
Computational methods: Structures were optimized with the
[12] Z. Shi, V. Goulle, R. P. Thummel, Tetrahedron Lett. 1996, 37,
2357 – 2360.
[13] S. V. Rosokha, S. M. Dibrov, T. Y. Rosokha, J. K. Kochi, Photo-
chem. Photobiol. Sci. 2006, 5, 914 – 924.
[14] S. V. Rosokha, M. D. Newton, M. Head-Gordon, J. K. Kochi,
Chem. Phys. 2006, 324, 117 – 128.
[15] S. V. Rosokha, J. M. Lu, M. D. Newton, J. K. Kochi, J. Am.
Chem. Soc. 2005, 127, 7411 – 7420.
[16] The Marcus equation allows the activation energy for the
electron transfer to be calculated from the total reorganization
energy (that is, the total of the internal and solvent contribu-
tions) and the reaction free energy. However, we use in this case
only the internal reorganization energies, as studies on similar
systems have shown the solvent reorganization energy contri-
bution to be minimal (for example, Refs. [13]–[15]). Nonethe-
less, the calculated activation energies should be considered as
lower limits of the real value.
[
21]
gradient-corrected BP86 functional. Systems involving the iodine
atom employed a large-core, quasi-relativistic, effective core poten-
[
22]
tial with the associated (16s11p6d)/[9s6p1d] valence basis set; all
[
23]
other atoms were described with the def2-TZVP basis set. Single-
point solvent-phase calculations of the gas-phase-optimized struc-
[
24]
tures were performed using the polarizable continuum model with
a dielectric constant of 38.3 for DMF. For the single-point calculations
[
21a–b,25]
the B3LYP functional
G(d,p) basis set.
was used in conjunction with the 6-311 +
[
26]
+
Received: February 7, 2007
Revised: April 25, 2007
Published online: June 4, 2007
Keywords: aryl anions · electron transfer · reduction ·
synthetic methods
.
[
[
[
[
[
17] M. D. Otero, B. Batanero, F. Barba, Tetrahedron Lett. 2006, 47,
8215 – 8216.
18] J. Simonet, M. Chaquiq El Badre, G. Mabon, J. Electroanal.
Chem. 1990, 281, 289.
19] N. C. Yang, P. Kumler, S. S. Yang, J. Org. Chem. 1972, 37, 4022 –
[
1] a) L. Pause, M. Robert, J.-M. Savꢀant, J. Am. Chem. Soc. 1999,
21, 7158 – 7159; b) R. Munusamy, K. S. Dhathareyan, K. K.
1
Balasubramanian, C. S. Venkatachalam, J. Chem. Soc. Perkin
Trans. 2 2001, 1154 – 1166; c) R. J. Enemaerke, T. B. Christensen,
H. Jensen, K. Daasbjerg, J. Chem. Soc. Perkin Trans. 2 2001,
4
026.
20] A. Shaabani, P. Mirzaei, S. Naderi, D. G. Lee, Tetrahedron 2004,
0, 11415 – 11420.
6
1620 – 1630.
21] a) A. D. Becke, Phys. Rev. A 1988, 38, 3098 – 3100; b) J. P.
Perdew, Phys. Rev. B 1986, 33, 8822 – 8824; c) S. H. Vosko, L.
Wilk, M. Nusair, Can. J. Phys. 1980, 58, 1200 – 1211.
[
2] Much more stabilized trifluoromethyl and benzylic anions can be
formed using the commercially available reagent tetrakisdime-
thylaminoethene (TDAE), but we have shown that that reagent
does not react with aryl halides (M. Mohan, PhD thesis,
University of Strathclyde, 2005); a) S. Ait-Mohand, N. Takechi,
M. Medebielle, W. R. Dolbier, Jr., Org. Lett. 2001, 3, 4271 –
[
[
[
22] P. Schwerdtfeger, M. Dolg, W. H. E. Schwarz, G. A. Bowmaker,
P. D. W. Boyd, J. Chem. Phys. 1989, 91, 1762 – 1774.
23] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297 –
3305.
4273; b) C. R. Burkholder, W. R. Dolbier, Jr., M. Medebielle,
24] J. Tomasi, B. Mennucci in The Encyclopedia of Computational
Chemistry, Vol. 1 (Ed.: P. von R. Schleyer), Wiley, Athens, 1998,
p. 2547.
J. Fluorine Chem. 2001, 109, 39 – 48.
[
3] J. A. Murphy, T. A. Khan, S.-Z. Zhou, D. W. Thomson, M.
Mahesh, Angew. Chem. 2005, 117, 1380 – 1384; Angew. Chem.
Int. Ed. 2005, 44, 1356 – 1360.
4] a) J. R. Ames, M. A. Houghtaling, D. L. Terrian, T. A. Mitchell,
Can. J. Chem. 1997, 75, 28 – 36; b) R. P. Thummel, V. Goulle, B.
Chen, J. Org. Chem. 1989, 54, 3057 – 3061; c) T. A. Taton, P.
Chen, Angew. Chem. 1996, 108, 1098 – 1100; Angew. Chem. Int.
Ed. Engl. 1996, 35, 1011 – 1013; d) S. Hꢁnig, D. Sheutzov, H.
Schlaf, Justus Liebigs Ann. Chem. 1972, 765, 126.
[
25] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648 – 5652; b) R. H.
Hertwig, W. Koch, Chem. Phys. Lett. 1997, 268, 345 – 351; c) C. T.
Lee, W. T. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785 – 789;
d) P. J. Stephens, F. J. Devlin, C. F. Chabalowski, M. J. Frisch, J.
Phys. Chem. 1994, 98, 11623 – 11627.
[
[
26] a) M. J. Frisch, J. A. Pople, J. S. Binkley, J. Chem. Phys. 1984, 80,
3265 – 3269; b) T. Clark, J. Chandrasekhar, G. W. Spitznagel,
P. V. Schleyer, J. Comput. Chem. 1983, 4, 294 – 301; c) R.
Krishnan, J. S. Binkley, R. Seeger, J. A. Pople, J. Chem. Phys.
[
5] C. P. Andrieux, J. Pinson, J. Am. Chem. Soc. 2003, 125, 14801 –
14806.
1980, 72, 650 – 654.
[
6] G. A. Ross, M. D. Koppang, D. E. Bartak, N. F. Woolsey, J. Am.
Chem. Soc. 1985, 107, 6742 – 6743.
Angew. Chem. Int. Ed. 2007, 46, 5178 –5183ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5183