p-Diethynylbenzene-based molecular wires, Fe-C{triple bond, long}C-p-C6H2X2-C{triple bond, long}C-Fe [Fe = Fe(η5-C5Me5)(dppe)]: Synthesis, substituent effects and unexpected formation of be

Article abstract of DOI:10.1016/j.jorganchem.2009.01.019

A series of p-diethynylbenzene-based molecular wires, Fe-C{triple bond, long}C-p-C₆H₂X₂-C{triple bond, long}C-Fe (3) (Fe = FeCp^*(dppe)), is prepared and their wire-like performance is estimated on the basis of t

Full text of DOI:10.1016/j.jorganchem.2009.01.019

Journal of Organometallic Chemistry 694 (2009) 1840–1847
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
p-Diethynylbenzene-based molecular wires, Fe–C„C-p-C6H2X2–C„C–Fe [Fe =
Fe(g
5-C5Me5)(dppe)]: Synthesis, substituent effects and unexpected formation
of benzodifuran complex
*
Yumi Matsuura, Yuya Tanaka, Munetaka Akita
Chemical Resources Laboratory, Tokyo Institute of Technology, R1-27, 4259 Nagatsuta, Midori-ku, Yokohama 226-8503, Japan
a r t i c l e i n f o
a b s t r a c t
A
series of p-diethynylbenzene-based molecular wires, Fe–C„C-p-C6H2X2–C„C–Fe (3) (Fe = FeCp*
Article history:
Received 11 November 2008
Received in revised form 12 January 2009
Accepted 13 January 2009
Available online 20 January 2009
(dppe)), is prepared and their wire-like performance is estimated on the basis of the KC and Vab values.
It has been revealed that electron-donating substituents (X) improve the performance. The benzodifuran
complex 4 unexpectedly formed from the derivative with X = OH shows the performance comparable
to 3.
Ó 2009 Elsevier B.V. All rights reserved.
Keywords:
Iron
Acetylide
Molecular wire
Substituent effects
Redox
X-ray crystallography
1. Introduction
[7,8]. It is notable that some of the aryl-substituted derivatives
show wire-like performance even superior to the polyynediyl com-
Organic
p
-conjugated systems combined with metal fragments
plexes and thus the family of the complexes is regarded as ‘‘tun-
able molecular wires”.
have been studied from the viewpoint of molecular devices [1,2].
In particular, one-dimensional organometallic molecular wires
with redox-active metal attachments have been studied most
extensively [3–6], because the electron-donating metal end caps
can stabilize the cationic radical species resulting from oxidation
and furthermore their wire-like performance is conveniently
evaluated on the basis of their electrochemical properties, KC
(comproportionation constant) for the 1e-oxidized, mixed valence
species.
In order to further study substituent effects our research efforts
were focused on a series of the p-diethynylbenzene derivatives,
Fe–C„C-p-C6H2X2–C„C–Fe. In addition to the major purpose, we
attempted switching of the wire-like performance by introduction
of pH-sensitive substituents (X = OH and NMe2) [9]. Although the
latter attempts were not always successful, a new
p-conjugated
polycyclic system, i.e. benzodifuran complex, resulted from the
precursor with X = OH.
The organic p-conjugated parts can be constructed by a combi-
nation of a limited number of sp-(C„C) and sp2-hybridized hydro-
carbyl components (e.g. C@C and phenylene). Of a variety of
structural motifs of the linkers studied so far, polyynediyl bridges
[(C„C)n] turn out to be superior to other bridges as revealed by
the studies, in particular, using the highly electron-donating
2. Results and discussion
2.1. Synthesis of p-diethynylbenzene-based molecular wires,
Fe–C„C-p-C6H2X2–C„C–Fe (3)
M((g
5-C5R5)L2 fragments [M/L2 = Fe/(PR3)2 [4], Ru/(PR3)2 [5],
Re/(NO)(PR3) [6]] (Scheme 1). The polyynediyl ligands, however,
cannot be modified, because no substituent can be introduced onto
the sp-hybridized carbon atoms. In a previous paper, we reported
synthesis of a series of endiyne–diiron complexes, Fe–C„C–
C(R)@C(R)–C„C–Fe [Fe = FeCp*(dppe)], and the substituent effects
The p-diethynylbenzene derivatives 3 were prepared by the
‘‘vinylidene method” following the procedures reported for the
parent compound 3d by Lapinte (Scheme 2) [4d,g,o,10]. The sol-
vated cationic FeCp*(dppe) species ([Cp*(dppe)Fe(MeOH)]+), which
was formed via ionization of Fe–Cl in MeOH in the presence of
KPF6, captured 1-alkyne to cause spontaneous isomerization to
the vinylidene intermediate 22+(PF6)2. Subsequent deprotonation
afforded the acetylide complexes 3 except for the OH derivative
* Corresponding author.
E-mail address: makita@res.titech.ac.jp (M. Akita).
0022-328X/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2009.01.019

Y. Matsuura et al. / Journal of Organometallic Chemistry 694 (2009) 1840–1847
1841
M
O
Fe
Fe
Fe-Cl
C
1) KPF6 / MeOH
2) 1g (X= OH)
4
O
X
R
C
3) NaOMe / MeOH
M
C
C
C C M
M
C
C nM
C
C
C
R
22+g =
H
O
X
C
HO
OH
Fe
Fe
M
OH
Fe
C C
H
Scheme 1.
H
H2
C
O
O
Fe
Fe
C
O
OH
H2
1g (see below) [11]. The obtained products 3 were sparingly
soluble in common organic solvents, as reported for 3d [4d], but
readily characterized by spectroscopic features, in particular, on
the basis of the m(C„C) vibrations and the simple NMR spectra
containing single sets of signals for the aryl and Fe moieties in ac-
H
H
22+g'
1) BuLi
2) I-FeCp*(CO)2
3) dppe / hν
S
S
Fe
Fe
cord with the symmetrical structures.
5
S
S
For measurements of the IVCT bands (see below), the neutral
compounds 3 were converted into the monocationic species 3+
by treatment with [FeCp2]PF6 in CH2Cl2.
Scheme 3.
2.3. Evaluation of wire-like performance of complexes 3–5
2.2. Formation of benzodifuran complex 4 and preparation of its
S-analogue, benzodithiophene complex 5
2.3.1. Electrochemical evaluation
In order to evaluate the performance of the obtained diiron
complexes 3–5 as molecular wires they were subjected to CV mea-
surements [14]. All monocationic p-diethynylbenzene species 3+
showed two reversible redox waves separated by 250–330 mV
Reaction of the OH derivative 1g with the Fe reagent did not af-
ford the expected acetylide complex 3g but the unexpected doubly
fused benzodifuran complex 4 (Scheme 3). The red product 4
turned out to be an isomer of 3g as judged by the ESI-MS data
(m/z = 1334.8) but its IR spectrum did not contain an O–H vibration
essential for the expected structure. Although 4 could not be char-
acterized by its spectroscopic data alone, a clue to the structure
was obtained from X-ray crystallography of the cationic intermedi-
ate 22+g0 obtained by repeated recrystallization of a reaction mix-
ture of the Fe reagent and 1g [12; see also Section 3.4]. The
cationic intermediate 22+g0 was characterized as the dicationic
Fischer-type cyclic di(O-substituted-carbene) species [13] with
the doubly fused tricyclic structure, which should result from
nuclephilic addition of the OH group at the electrophilic vinylidene
(D
E), as typically exemplified for 3+a ꢀ PF6 (Fig. 1a) [15]. The CV
charts for the other derivatives are included in the Supporting
Information (Fig. S1). From the
DE values the comproportionation
constants (KC) are determined to be 2.2 ꢁ 104–4.3 ꢁ 105 (Table 1).
The benzodifuran complex 4 and its S-analogue 5 show the KC val-
ues comparable to those of 3. The KC values for complexes 3–5 fall
in the range for those of class II Robin-Day compounds [16].
2.3.2. Near IR spectra
Monocationic species of high-performance molecular wires
show an intense and broad IVCT (intervalence charge transfer)
band in the near IR (NIR) region, and the Vab value obtained from
the spectrum is regarded as another measure for the performance
as molecular wire, in case the monocationic species is isolable [17].
Although no theoretical relationship between the KC and Vab values
is established, an efficient molecular wire shows larger KC and Vab
values.
a
-carbon atom in 22+g followed by proton migration (Scheme 3).
Action of a base to 22+g0 should cause deprotonation from the
methylene moieties to form the furan structures in 4. For a com-
parison sake, the benzodithiophene complex 5, an S-analogue of
4, was prepared via a different route, i.e. reaction of lithiated ben-
zodithiophene with I-FeCp(CO)2 followed by photochemical ligand
replacement with dppe (Scheme 3). The neutral compounds 4 and
5 were also converted into the monocationic species 4+ ꢀ PF6 and
5+ ꢀ PF6, respectively, by treatment with [FeCp2]PF6.
All monocationic species 3+–5+ show intense NIR absorptions in
the range of 3000–10000 cmꢂ1. A typical NIR spectrum for 3+a is
Fe
X
X
C
H
H
C
1) KPF6 / MeOH-THF
NaOMe
Fe-Cl
C
C
Fe
Fe
C
C
C
C
C
C Fe
MeOH
X
2)
3a-f
X
X
X
H C
C
C
C H
(PF6
22+a-f·(PF6)2
)
2
Fe
[FeCp2]PF6
(0.9 equiv.)
1a-f
X
X
X= NMe2 (a), OMe (b),
CH3 (c), H (d),
Fe =
Fe
C
C Fe
PF6
Ph2P
PPh2
F (e), CF3 (f).
3+a-f
Scheme 2.

1842
Y. Matsuura et al. / Journal of Organometallic Chemistry 694 (2009) 1840–1847
ε x 10-4
2.5
a
b
2.0
1.5
1.0
0.5
0
0
-200 -400 -600 -800 -1000 -1200 -1400 -1600
mV (vs. [FeCp2] / [FeCp2]+)
3000 4000 5000 6000 7000 8000 9000 10000
wavenumber / cm-1
Fig. 1. (a) A cyclic voltammogram for 3+a ([3+a ꢀ PF6] = 2.0 ꢁ 10ꢂ3 M; [Bu4N ꢀ PF6] = 0.1 M; Pt electrode; observed in CH2Cl2). (b) A NIR spectrum for 3+a ꢀ PF6 (observed in
CH2Cl2; black: observed spectrum; blue: deconvoluted Gaussian curves; red: sum of the deconvoluted curves). (For interpretation of the references in color in this figure
legend, the reader is referred to the web version of this article.)
Table 1
CV data for the diiron complexes 3–5.a
Complex (X/rp
)
E1c (mV)
E1a (mV)
jE1c ꢂ E1a j (mV)
E11=2 (mV)
E2a (mV)
E2c (mV)
jE2c ꢂ E2a j (mV)
E11=2 (mV)
D
E (mV)
ic1/ia1
ic2/ia2
KC
3a (NMe2/ꢂ0.83)
3b (OMe /ꢂ0.27)
3c (Me/ꢂ0.07)
3d (H/0)
3e (F/0.06)
3f (CF3/0.54)
4
ꢂ935
ꢂ869
ꢂ835
ꢂ792
ꢂ739
ꢂ704
ꢂ1020
ꢂ890
ꢂ867
ꢂ802
ꢂ769
ꢂ728
ꢂ673
ꢂ639
ꢂ932
ꢂ777
68
67
66
64
66
65
88
113
ꢂ901
ꢂ836
ꢂ802
ꢂ760
ꢂ706
ꢂ672
ꢂ976
ꢂ834
ꢂ602
ꢂ559
ꢂ554
ꢂ536
ꢂ451
ꢂ409
ꢂ675
ꢂ634
ꢂ535
ꢂ492
ꢂ487
ꢂ472
ꢂ387
ꢂ347
ꢂ597
ꢂ525
67
67
67
64
64
62
78
109
ꢂ569
ꢂ526
ꢂ520
ꢂ504
ꢂ419
ꢂ378
ꢂ636
ꢂ580
333
310
282
256
287
294
340
254
1.0
1.0
0.9
1.1
0.8
1.1
1.0
0.9
1.0
1.0
1.1
1.0
0.8
1.0
1.0
0.9
4.3 ꢁ 105
1.8 ꢁ 105
5.9 ꢁ 104
2.2 ꢁ 104
7.2 ꢁ 104
9.5 ꢁ 104
5.7 ꢁ 105
2.0 ꢁ 104
5
Observed in CH2Cl2; [complex] = ; [NBu4PF6] = 0.1 M; Pt electrode; the values are reported with respect to the FeCp2/[FeCp2]+ couple.
a
1/2
shown in Fig. 1b. (Spectra for the other derivatives are included in
the Supporting Information (Fig. S2 and Table S1) [18]). The
absorptions appear in the lower energy region so that they overlap
with the solvent absorptions (CH2Cl2). Deconvolution analysis re-
vealed that the NIR absorptions can be analyzed as a sum of three
Gaussian curves [17]. Of the NIR absorptions the most intense
bands of the lowest energies were assigned to the IVCT bands
[17], and the Vab values were obtained on the basis of their spectral
parameters (Table 2).
(2310mmax
)
(in cmꢂ1), the electronic coupling can be evaluated
III
using the equation V ¼ 1=2mmax (in cmꢂ1) = 6.20 ꢁ 10ꢂ5 mmax (in
ab
eV) [16].
The spectral parameters and the obtained Vab values are sum-
marized in Table 2 (see also Table S1). The V and V values fall
II
ab
III
ab
in the ranges of 0.01–0.08 and 0.25–0.34, respectively. For all com-
pounds, the 1/2 values are smaller than the critical values, suggest-
m
ing that they belong to class III, but this is in contrast to their rather
small KC values mentioned above. Such a class of compounds is re-
garded as ‘‘class IIB” compounds, as discussed by Lapinte [4o].
The Vab value for class II Robin-Day compounds can be calculated
II
according to the Hush formula:
V
¼ 2:55 ꢁ 10ꢂ6
(
mmax
ꢀ ꢀ
e
ab
m1/2
)
1/2rꢂ1 (in eV), where mmax
,
e
,
m
1/2, and r denote the absorption
2.3.3. Substituent effects on p-diethynylbenzene system 3
maximum, absorption coefficient, half-height width, and distance
between the two redox active centers, respectively [16]. On the
other hand, for class III Robin-Day compounds, for which the
The Hammet plot [19], a classical evalution method of the sub-
stituent effect, was applied to the series of the p-diethynylbenzene
system 3. But the situation is not so simple, because the ethynyl
groups are located ortho with respect to one substituent but meta
half-height width
(m1/2) is narrower than the critical value
Table 2
IVCT bands for the diiron complexes 3+–5+.a
II
ab
III
ab
b
R
mmax (cmꢂ1
)
e
(Mꢂ1 cmꢂ1
)
m
1/2(exp) (cmꢂ1
)
m1/2 (cmꢂ1
)
V
(eV)
V
(eV)
3+a
3+b
3+c
3+dc
3+e
3+f
4+
4554
4570
4100
5100
4150
4080
5055
5400
21800
21600
17550
8770
15500
13800
5400
1022
1310
1750
1350
1250
1620
1300
1300
3243
3249
3077
3432
3096
3070
3417
3532
0.070
0.079
0.078
0.054
0.062
0.067
0.045
0.010
0.282
0.283
0.254
0.316
0.257
0.253
0.313
0.335
5+
250
a
PF6 salts analyzed by deconvolution based on three Gaussian curves.
Vab values when assumed as Class III species (see text).
Taken from Ref. [4d].
b
c

Y. Matsuura et al. / Journal of Organometallic Chemistry 694 (2009) 1840–1847
1843
σp
σp
a
b
0.6
0.6
CF3 (3f)
CF3
CF3
0.4
0.2
0.4
0.2
H
F (3e)
Me (3c)
F
F
0
0
Me
OMe
Me
OMe
H
H
(3d)
-0.2
-0.4
-0.6
-0.8
-1.0
-0.2
-0.4
-0.6
-0.8
-1.0
OMe (3b)
1
2
E1/2
E1/2
ΔE
NMe2 (3a)
NMe2
-600
NMe2
250
300
mV
350
-400
-800
mV (vs. [FeCp2] / [FeCp2]+)
σp
σp
σI
c
d
e
0.6
0.6
0.6
CF3 (3f)
CF3
CF3
F
0.4
0.2
0.4
0.2
0.4
0.2
II
III
ΔE
Vab
Vab
F
(3e)
H
(3d)
F
0
0
0
H
Me
Me
-0.2
-0.4
-0.6
-0.8
-1.0
-0.2
-0.4
-0.6
-0.8
-1.0
-0.2
-0.4
-0.6
-0.8
-1.0
(3c)
OMe
OMe
(3b)
NMe2 (3a)
NMe2
0.06
0.08
280
300 0.06
0.24
0.25
0.26
0.27
0.28
0.25
0.26
VabII / eV
VabIII / eV
mV
Vab / eV
II
ab
III
ab
Fig. 2. Plots of (a)
D
E, (b) E11=2 and E12=2, (c) VII , (d) VIII for 3 against
rp, and (e) plots of
D
E, V , and V for 3e,f against rI
.
ab
ab
with respect to the other substituent. Furthermore, because ro is
not defined owing to the steric reason, rp is used instead to see
the resonance effect of the substituents.
from the IVCT bands of the monocationic species 3+. The obtained
II
III
V
and V values for them (Table 2) are plotted against
r
p (Fig. 2c
ab
ab
II
and d). Because (1) the differences are very small
0:024 eV;
mined accurately because of the IVCT bands overlapping with the
solvent absorptions, only a trend can be seen from the plots and
a rough relationship between the Vab and rp values is noted, i.e.
ðDV
¼
ab
III
When the E11=2 values are plotted against
r
r
p and
p; solid line in Fig. 2b),
m)). This is presumably because (1) the electron-donating
substituents employed in this study are the groups with the lone
pair electrons on the -heteroatom (a, b, and e) or the groups only
(or mainly) with the -inductive effect (c and f). Substituent effects
of such groups can be roughly correlated to p. In other words, (2)
electron-withdrawing substituents with the vacant p-orbital at the
-position, which are better described by m, are not included in
the present study. The following discussion, therefore, is based on
r
m, a better fit-
DV
¼ 0:063 eVÞ and (2) the Vab values cannot be deter-
ab
ting is obtained for the former (R2 = 0.919 (
0.775 (
r
a
r
electron-donating substituents (with smaller rp values) give larger
Vab values. In addition, 3e,f do not always show better perfor-
mance, as suggested by the CV measurements. The abnormal
behavior noted for 3e,f (the large KC values), therefore, should be
due to some other reason.
Abnormal trends as observed for 3e,f by CV measurements are
often ascribed to either destabilization or solvation of the dication-
ic species 32+ formed by the second oxidation (Scheme 4) [20]. Be-
r
a
r
the plots against r
p. The E11=2 values for 3a–f also show a good linear
relationship (R2 = 0.929) with those for the related mononuclear
species, Fe–C„C–C6H4-X-p, reported by Paul and Lapinte [4i,k,n,p].
cause, in the present case, (1) only derivatives with the highly
r-
The linear E11=2
ꢂ
rp correlation (Fig. 2b) indicates that electron-
donating substituents facilitate the first oxidation of 3, as expected.
electron-withdrawing groups are deviated from the trend and (2)
the second oxidation processes of them ðE21=2Þ are shifted toward
the anodic side, the destabilization would be the main reason.
In contrast to the good fitting of the E11=2
linear relationship is noted between p and
ꢂ
rp plots no apparent
E (Fig. 2a). When the
r
D
When the r
I and r0R values for the F-containg groups are compared
trend of the second redox potentials E11=2 (Fig. 2b) is inspected, dis-
continuity is found between 3d and 3e, suggesting that the series of
compounds 3a–d and 3e,f should be discussed separately. The
shifts of E21=2 for 3e,f to the anodic side compared to 3a–d result
(F:
r
I = 0.50, r0R ¼ ꢂ0:35; CF3:
r
I = 0.45, r0R ¼ 0:01), the inductive
effects (rI) of the two groups are comparable in contrast to the res-
onance effects ðr0RÞ being significantly different owing to the pres-
ence/absence of lone pair electrons on the atoms adjacent to the
aromatic ring. When rI is taken as the measure, it is understand-
able that 3e,f with the substituents of the similar rI values (0.50
in increase of the
values.
DE values eventually leading to the large KC
In order to see the discontinuity issue from another side the
performance was evaluated on the basis of the Vab values obtained
(F)/0.45 (CF3)) give the very similar E2 values (ꢂ419 mV (F)/
1=2
ꢂ378 mV (CF3); Fig. 2e). The VII and VIII plots against
rI also bring
ab
ab

1844
Y. Matsuura et al. / Journal of Organometallic Chemistry 694 (2009) 1840–1847
X
were reported for the endiyne system [7] described in Introduction
and other systems [20].
3
Fe
C
C
C
C
C
Fe
Combination of the results of the electrochemical and IVCT
measurements leads to the conclusion that, for the p-diethynyl-
benzene system 3, electron-donating substituents improve the
wire-like performance. On the other hand, the large KC values ob-
X
- e+
X
X
etc.
Fe
C
C
C
Fe
Fe
Fe
C
C
C
C
Fe
Fe
Fe
served for the derivatives with highly
groups such as F and CF3 (3e,f) do not reflect the real perfor-
mance owing to the -inductive effects and, therefore, the esti-
r-electron-withdrawing
3+
A
B
X
X
r
X
mation should be confirmed by taking into account the results
of other estimation methods including the IVCT measurement
(Vab).
C
C
C C
C
X
- e+
X
X
2.4. Other comments
Fe
C
C
C
C
Fe
Fe
C
C
C C
2.4.1. Switching performance of the pH-sensitive derivatives
32+
D
E
In addition to the study on molecular wires, more sophisticated
functions such as switch are essential for development of molecu-
lar circuits [1]. Recently, we reported photoswitchable molecular
wires containing the photochromic dithienylethene units [9]. pH-
control is another promising switching mechanism and, on the ba-
sis of this concept, we planed preparation of compounds 3a
(X = NMe2) and 3g (X = OH). The NMe2 derivative 3a was obtained
successfully, whereas the attempted preparation of 3g via the
vinylidene method resulted in the formation of the cyclized benzo-
difuran complex 4, as described above (Scheme 3). Then proton-
ation behavior of 3a was monitored by CV and IVCT
measurements. The CV trace of 3a in the presence of 1 equivalent
of CH3COOH was similar to that of 3a with the exception of disap-
pearance of the oxidation wave of the first redox process (Fig. S4),
but the IVCT band disappeared upon addition of CF3COOH
(1 equiv.) (Fig. S5), suggesting occurrence of some switching
behavior. Subsequent deprotonation with NEt3 gave the CV trace
identical to that of the neutral species 3a. But these changes did
not result from protonation at the NMe2 moiety but from proton-
ation at the C„C moiety leading to the cationic vinylidene species,
[Fe„C„C(H)–C6H2(NMe2)2–C„C–Fe]+, i.e. reversion of the forma-
tion process of 3a (Scheme 2), because similar behavior was ob-
served for the other derivatives (e.g. 3d) without an acid-
sensitive functional group (X). Although the mechanism was differ-
ent from what we anticipated, the acetylide molecular wire 3
turned out to exhibit switching function by the protonation-depro-
tonation procedures at the Fe–C„C moiety.
X
X
e
e
FeIII FeIII
FeII FeIII
FeII FeII
2
1
E1/2
E1/2
32+
3+
3
destabilization
stabilization
destabilization
electron-donating X
electron-donating X
large ΔE large KC
highly electron-withdrawing X
and / or
Scheme 4.
the plots for 3e to the proximity of those of 3f (Fig. 2e) [18]. The
small deviation noted for E11=2 of 3e (Fig. 2b) may be also ascribed
to this effect, and the situation may be better described by the dot-
ted lines separated into the two series 3a–d and 3e,f in a manner
similar to the analysis of E12=2 described above. These consider-
ations suggest that the large KC values for 3e,f do not always indi-
cate better performance but result from the inductive effects of the
highly
r-electron-withdrawing substituents.
The substituent effects observed by the CV measurements can
be interpreted in terms of Scheme 4. As to the shift of E11=2, elec-
tron-donating substituents should increase the electron density
at the metal center in the neutral species 3 to make the oxida-
tion easier, i.e. cathodic shift of E11=2. When attention is turned
to the monocationic species 3+ resulting from the 1e-oxidation,
the electron-donating substituents in the normal 3a–d series
should stabilize one of the resonance structures B with the cat-
ion radical center on the aromatic carbon atom adjacent to the
2.4.2. Benzodifuran (4) and -thiophene complexes (5)
Because molecular wires with fused heteroaromatic bridges are
rare [2], performance of the benzodifuran complex 4 as molecular
wire was examined as compared with the S-analogue 5, the benzo-
dithiophene complex. The electrochemical and NIR data for them
are summarized in Tables 1 and 2, and the original data are in-
cluded in Figs. S1 and S2. Complexes 4 and 5 showed the KC values
comparable to those of 3 and 4, in which the two iron centers are
separated by the carbon linkages consisting of eight carbon atoms.
The effect of the hetero atom, however, is also significant as can be
seen from comparison with the S-analogue 5, which showed the KC
value smaller than that of 4 by a factor of ca. 30. The different per-
substituent (X) through
p-interaction with the lone pair elec-
trons on X (C) or hyperconjugation with the methyl group
(Scheme 4). This effect should cause anodic shift of the second
oxidation process ðE21=2Þ. Synergism of these two effects brings
about an increase of the
sults observed.
DE value to be consistent with the re-
On the other hand, the highly
r
-electron-withdrawing F and
formance can be explained by taking into account r. When the rp
CF3 groups should destabilize the electron-deficient dicationic spe-
cies 32+ formed by the second oxidation, in particular, through the
resonance structure E, where the electron-withdrawing X groups
directly influence the electronic structures of the Fe moieties to
cause lowering the electron densities at the Fe centers. Although
similar destabilization is applicable to the monocationic species
3+, the effect should be more serious for the more electron-defi-
values for the OH and SH groups are compared, the OH group
(ꢂ0.37) is more electron-donating than the SH group (0.15), and
the difference mainly comes from the resonance effect as is evident
from the considerably different r0R values (ꢂ0.40 (OH) versus
ꢂ0.15 (SH); cf.
rI: 0.25 (OH), 0.25 (SH)). The lone pair electrons
on the O atom, the atomic size of which is comparable to that of
the carbon atom, should stabilize the C-centered cationic radical
cient species 32+ than for 3+. As a result, the
ger compared to the other derivatives (Scheme 4). Similar effects
D
E values become lar-
effectively through
p-donation, as discussed for the 3-series com-
pounds (Scheme 4).

Y. Matsuura et al. / Journal of Organometallic Chemistry 694 (2009) 1840–1847
1845
2.93, 3.20 (4H ꢁ 2, br ꢁ 2, CH2 in dppe), 3.45 (4H, m, CH2), 5.99
(2H, s, C6H2), 7.51–6.96 (40H, m, Ph); dH (CD3CN) 105.1), which
was washed with ether and pentane. To the brown intermediate
dissolved in THF (30 mL) was added NaOMe (1.2 mmol; prepared
from Na and MeOH), and the mixture was stirred for 20 min. After
removal of the volatiles under reduced pressure, the product was
extracted with toluene and passed through an alumina column. Re-
moval of the solvent under reduced pressure and washing with
acetonitrile and pentane gave 4 as red powder (620 mg, 87% yield).
4: dH (C6D6) 1.53 (30H, s, Cp*), 1.85, 2.48 (4H ꢁ 2, br ꢁ 2, CH2), 5.86,
6.76 (2H ꢁ 2, s ꢁ 2, Ar), 7.55–7.08 (40H, m, Ph); dP (C6D6) 100.3;
ESI-MS: 1335 (M+1) [24].
3. Experimental
3.1. General methods
All manipulations were carried out under an inert atmosphere
by using standard Schlenk tube techniques. THF, toluene, pentane
(Na–K alloy), CH2Cl2, MeCN (P2O5), and MeOH (Mg(OMe)2) were
treated with appropriate drying agents, distilled, and stored under
argon. For the analytical instruments used in the present study, see
our previous papers [7]. Fe–Cl [21], benzodithiophene [22], and the
precursors 1b–e,g [23] were prepared according to the reported
methods, and the synthetic procedures for 1a,f are included in
the Supporting Information. Other chemicals were purchased and
used as received.
3.4. Formation of the Fischer carbene species 22+g0 ꢀ [(
l-O)(FeCl3)2]
and di-Cp*-substituted benzodifuran A and [Fe(
j2-dppe)2
3.2. Preparation of 3
(NCMe)2]2+[( -dppe){FeCl3}2]2ꢂ B [12]
l
As a typical example, the experimental procedures for 3a are
described [4d]. The other derivatives were prepared following
essentially the same procedures.
Repeated recrystallization of a reaction mixture of 1g/Fe–Cl/
KPF6 from MeCN gave single crystals of 22+g0 having [Cl3Fe–O–
FeCl3]2ꢂ as the counter anion and two more products, the
Fe–Cl (580 mg, 0.98 mmol), KPF6 (167 mg, 0.91 mmol), and 1a
(100 mg, 0.47 mmol) were dissolved in a 9:1 mixture of MeOH–
THF (12 mL), and the resultant mixture was stirred for 12 h at
room temperature. Removal of the volatiles under reduced pres-
sure, extraction with ether, filtration through a Celite plug, and re-
moval of the solvent under reduced pressure left brown powder
(vinylidene intermediate). To the brown intermediate dissolved
in THF (35 mL) was added NaOMe (1.4 mmol; prepared from Na
and MeOH), and the mixture was stirred for 20 min. After removal
of the volatiles under reduced pressure, the product was extracted
with toluene and passed through an alumina column. Removal of
the solvent under reduced pressure and washing with ether and
pentane gave 1a as red powder (173 mg, 26% yield). 3a: dH
(C6D6) 1.63 (30H, s, Cp*), 1.96, 2.91(4H ꢁ 2, br ꢁ 2, CH2), 2.60
(12H, s, NMe2) 6.83 (2H, s, C6H2), 8.12–7.10 (40H, m, Ph); dP
di-Cp*-substituted benzodifuran A and [Fe( 2-dppe)2(NCMe)2]2+
j -
[(l
-dppe){FeCl3}2]2ꢂ B, which were characterized by X-ray crystal-
lography (see Supporting Information). The organic component A
could be isolated by TLC separation and further characterized by
1H NMR [dH (CD3CN) 1.35 (6H, s, Me), 0.68, 1.84 (12H ꢁ 2, s ꢁ 2,
Me), 6.39, 7.37 (2H ꢁ 2, s ꢁ 2, Ar)].
P
O
P
Fe
Fe
P
P
O
22+g'·[(μ-O)(FeCl3)2]
(Ph and [(μ-O)(FeCl3)2]2-: omitted for clarity)
(C6D6) 94.2; IR (KBr/ cmꢂ1 (C„C) 2041 cmꢂ1; ESI-MS: 1389
) m
(M+1); Anal. Calc. for C86H92N2P4Fe2: C, 77.35; H, 6.67; N, 2.02.
Found: C, 73.61; H, 6.51; 1.98%. 3b (4% yield; red powder): dH
(C6D6) 1.62 (30H, s, Cp*), 1.98, 2.98 (4H ꢁ 2, br ꢁ 2, CH2), 3.46
(6H, s, OMe), 6.70 (2H, s, C6H2), 8.16–7.04 (40H, m, Ph); dP (C6D6)
O
A
O
[Fe(κ2-dppe)2(NCMe)2][(μ-dppe)(FeCl3)2] B
94.0;
m
(C„C) 2049 cmꢂ1; ESI-MS: 1364 (M+1); Anal. Calc. for
C85.5H89P4Cl3Fe2 (3b ꢀ (CH2Cl2)1.5): C, 68.89; H, 6.02. Found: C,
68.84; H, 6.25%. 3c (4% yield; red powder): dH (C6D6) 1.59 (30H,
s, Cp*), 1.89, 2.73 (4H ꢁ 2, br ꢁ 2, CH2), 2.31 (6H, s, Me), 6.70 (2H,
3.5. Preparation of benzodithiophene complex 5
s, C6H2), 8.00–7.12 (40H, m, Ph); dP (C6D6) 92.9;
m(C„C)
2039 cmꢂ1; ESI-MS: 1331 (M+1); Anal. Calc. for C85H88P4Cl2Fe2
(3c ꢀ CH2Cl2): C, 72.09; H, 6.26. Found: C, 71.87; H, 6.52%. 3e (9%
yield; red powder): dH (C6D6) 1.54 (30H, s, Cp*Me), 1.88, 2.76
(4H ꢁ 2, br ꢁ 2, CH2), 6.88 (2H, t, J = 8.6 Hz, C6H2), 8.07–7.03
LDA was prepared by addition of t-BuLi (1.57 M in hexane,
1.7 mL, 2.7 mmol) to HNPri2 (1.6 mL) dissolved in THF (10 mL) at
ꢂ78 °C [25]. After stirring for 15 min at the same temperature ben-
zodithiophene [22] (50 mg, 0.26 mmol) dissolved in THF (5 mL)
was added dropwise. After further stirring for 30 min I-FeCp*(CO)2
(1.01 g, 2.7 mmol) was added, and the resultant mixture was stir-
red for 20 min at ꢂ78 °C and then for 2 h at room temperature.
The residue obtained by removal of the volatiles under reduced
pressure was washed with hexane, extracted with toluene, and fil-
tered through a Celite plug. Removal of the solvent under reduced
pressure gave the carbonyl complex, (CO)2Cp*Fe-benzodithioph-
ene-FeCp*(CO)2, as yellow powder (36 mg, 20%; dH (CDCl3) 1.78
(40H, m, Ph); dP (C6D6) 93.1; m
(C„C) 2049 cmꢂ1; ESI-MS: 1339
(M+1); Anal. Calc. for C83H82P4Cl2Fe2 (3e ꢀ CH2Cl2): C, 70.00; H,
5.80. Found: C, 69.54; H, 5.97%. 3f (6% yield; red powder): dH
(C6D6) 1.55 (30H, s, Cp*), 1.90, 2.70 (4H ꢁ 2, br ꢁ 2, CH2), 6.88
(2H, s, C6H2), 7.94–7.05 (40H, m, Ph); dP (C6D6) 93.0;
m(C„C)
2032 cmꢂ1; ESI-MS: 1439 (M+1) [24].
3.3. Preparation of benzodifuran complex 4
(s, 30H, Cp*), 6.95, 7.81(2H ꢁ 2, s ꢁ 2, Ar–H); IR (KBr)
m(CO) 1993,
1939 cmꢂ1). The carbonyl intermediate and dppe (48 mg,
0.12 mmol) were dissolved in toluene–acetonitrile mixture (95:5;
50 mL) and irradiated for 1.5 h. The solution color changed from
yellow to orange. Removal of the solvent, extraction with toluene,
and filtration through an alumina pad followed by removal of the
solvent left red solid, which was washed with pentane and aceto-
nitrile. The product 5 was obtained as red powder (60 mg, 93%
Fe–Cl (672 mg, 1.1 mmol), KPF6 (198 mg, 1.1 mmol), and 1 g
(85 mg, 0.54 mmol) were dissolved in a 5:4 mixture of MeOH–
THF (18 mL), and the resultant mixture was stirred for 14 h at
room temperature. Removal of the volatiles under reduced pres-
sure, extraction with CH2Cl2, filtration through a Celite plug, and
removal of the solvent under reduced pressure left brown powder
(the cyclized intermediate 2 g02+: (dH (CD3CN) 1.34 (30H, s, Cp*),

1846
Y. Matsuura et al. / Journal of Organometallic Chemistry 694 (2009) 1840–1847
based on the carbonyl intermediate). 5: dH (C6D6) 1.56 (30H, s, Cp*),
1.88, 2.40 (4H ꢁ 2, br ꢁ 2, CH2), 6.08 (2H, s, Ar), 7.55–6.98 (42H, m,
Ph and Ar); dH (C6D6) 94.9; ESI-MS: 1366 (M+1) [24].
(j) F. Coat, F. Paul, C. Lapinte, L. Toupet, L. Costuas, J.-F. Halet, J. Organomet.
Chem. 683 (2003) 368;
(k) K. Costuas, F. Paul, L. Toupet, J.-F. Halet, C. Lapinte, Organometallics 24
(2005) 2053;
(l) S. Roue, C. Lapinte, T. Bataille, Organometallics 23 (2004) 2558;
(m) F. de Montigny, G. Argouarch, K. Costuas, J.F. Halet, T. Roisnel, L. Toupet, C.
Lapinte, Organometallics 24 (2005) 4558;
3.6. Electrochemical measurements
(n) F. Paul, L. Toupet, J.-Y. Thépot, K. Costuas, J.-F. Halet, C. Lapinte,
Organometallics 24 (2005) 5464;
(o) G.S. Ibn, F. Paul, L. Toupet, C. Lapinte, J. Am. Chem. Soc. 128 (2006)
2463;
(p) F. Paul, B.G. Ellis, M.I. Bruce, L. Toupet, T. Roisnel, K. Costuas, J.-F. Halet, C.
Lapinte, Organometallics 25 (2006) 649;
(q) F. Paul, S. Goeb, F. Justaud, G. Argouarch, L. Toupet, R.F. Ziessel, C. Lapinte,
Inorg. Chem. 46 (2007) 9036;
(r) S. Ibn Ghazala, N. Gauthier, F. Paul, L. Toupet, C. Lapinte, Organometallics
26 (2007) 2308;
(s) S. Szafert, F. Paul, W.E. Meyer, J.A. Gladysz, C. Lapinte, Compt. Rend. Chem.
11 (2008) 693;
Electrochemical measurements were made with a BAS 100B/W
analyzer. CV measurements were performed with a Pt electrode for
CH2Cl2 solutions of the samples (ꢃ2 ꢁ 10ꢂ3 M) in the presence of
an electrolyte ([Bu4N ꢀ PF6] = 0.1 M) at room temperature under
an inert atmosphere. The scan rates were 100 mV/s. After the mea-
surement, ferrocene (Fc) was added to the mixture and the poten-
tials were calibrated with respect to the Fc/Fc+ redox couple
(+0.65 V versus Ag/Ag+ (in CH2Cl2)) [14]. Simulation of the electro-
chemical data was performed with Origin 6.1.
(t) F. de Montigny, G. Argouarch, T. Roisnel, L. Toupet, C. Lapinte, S.C.F. Lam,
C.H. Tao, V.W.W. Yam, Organometallics 27 (2008) 1912;
(u) C. Lapinte, J. Organomet. Chem. 693 (2008) 793.
[5] M.I. Bruce, P.J. Low, K. Costuas, J.-F. Halet, S.P. Best, G.A. Heath, J. Am. Chem.
Soc. 122 (2000) 1949;
3.7. Preparation of monocationic species
A mixture of 3a (30 mg, 0.022 mmol) and [FeCp2]PF6 (6.5 mg,
0.20 mmol) in CH2Cl2 (5 mL) was stirred for 15 min at ꢂ78 °C
and then for 3 h at room temperature. The orange suspended mix-
ture turned into a homogeneous purple solution. After removal of
the volatiles under reduced pressure the residue was washed with
ether and pentane, and recrystallized from CH2Cl2–ether to give
3+a ꢀ PF6 as deep purple powder. The other monocationic species
3+–5+ were prepared analogously.
M.I. Bruce, B.D. Kelly, B.W. Skelton, A.H. White, J. Organomet. Chem. 604 (2000)
150;
M.I. Bruce, B.G. Ellis, P.J. Low, B.W. Skelton, A.H. White, Organometallics 22
(2003) 3184;
M.I. Bruce, M.E. Smith, B.W. Skelton, A.H. White, J. Organomet. Chem. 637–639
(2001) 484;
See also recent papers,M.I. Bruce, K. Costuas, T. Davin, J.F. Halet, K.A.
Kramarczuk, P.J. Low, B.K. Nicholson, G.J. Perkins, R.L. Roberts, B.W. Skelton,
M.E. Smith, A.H. White, Dalton Trans. (2007) 5387;
M.I. Bruce, K. Costuas, B.G. Ellis, J.F. Halet, P.J. Low, B. Moubaraki, K.S. Murray,
N. Ouddai, G.J. Perkins, B.W. Skelton, A.H. White, Organometallics 26 (2007)
3735;
Acknowledgments
M.I. Bruce, M.L. Cole, C.R. Parker, B.W. Skelton, A.H. White, Organometallics 27
(2008) 3352;
M.I. Bruce, M. Jevric, C.R. Parker, W. Patalinghug, B.W. Skelton, A.H. White, N.N.
Zaitseva, J. Organomet. Chem. 693 (2008) 2915;
M.I. Bruce, N.N. Zaitseva, B.K. Nicholson, B.W. Skelton, A.H. White, J.
Organomet. Chem. 693 (2008) 2887;
M.I. Bruce, M. Gaudio, G. Melino, N.N. Zaitseva, B.K. Nicholson, B.W. Skelton,
A.H. White, J. Cluster Sci. 19 (2008) 147.
We are grateful to the Ministry of Education, Culture, Sports,
Science and Technology of the Japanese Government and the Japan
Society for Promotion of Science and Technology for financial sup-
port of this research.
[6] M. Brady, W. Weng, Y. Zhou, J.W. Seyler, A.J. Amoroso, A.M. Arif, M. Bohme, G.
Frenking, J.A. Gladysz, J. Am. Chem. Soc. 119 (1997) 775;
T. Bartik, W. Weng, J.A. Ramsden, S. Szafert, S.B. Falloon, A.M. Arif, J.A. Gladysz,
J. Am. Chem. Soc. 120 (1998) 11071;
Appendix A. Supplementary material
CCDC 708620, 708621, 708622 contain the supplementary crys-
tallographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif. Supplementary data associ-
ated with this article can be found, in the online version, at
doi:10.1016/j.jorganchem.2009.01.019.
R. Dembinski, T. Bartik, B. Bartik, M. Jaeger, J.A. Gladysz, J. Am. Chem. Soc. 122
(2000) 810;
See also recent publications and references cited therein.Q.L. Zheng, J.A.
Gladysz, J. Am. Chem. Soc. 127 (2005) 10508;
S. Szafert, J.A. Gladysz, Chem. Rev. 106 (2006);
PR1.Q.L. Zheng, J.C. Bohling, T.B. Peters, A.C. Frisch, F. Hampel, J.A. Gladysz,
Chem. Eur. J. 12 (2006) 6486;
L. de Quadras, F. Hampel, J.A. Gladysz, Dalton Trans. (2006) 2929;
L. de Quadras, E.B. Bauer, J. Stahl, F. Zhuravlev, F. Hampel, J.A. Gladysz, New J.
Chem. 31 (2007) 1594;
References
J. Stahl, W. Mohr, L. de Quadras, T.B. Peters, J.C. Bohling, J.M. Martin-Alvarez,
G.R. Owen, F. Hampel, J.A. Gladysz, J. Am. Chem. Soc. 129 (2007) 8282;
L. de Quadras, E.B. Bauer, W. Mohr, J.C. Bohling, T.B. Peters, J.M. Martin-Alvarez,
F. Hampel, J.A. Gladysz, J. Am. Chem. Soc. 129 (2007) 8296;
L. de Quadras, J. Stahl, F. Zhuravlev, J.A. Gladysz, J. Organomet. Chem. 692
(2007) 1859;
[1] J. Jortner, M.A. Ratner, Molecular Electronics, Blackwell Science, Oxford, 1997;
A. Aviram, M. Ratner, Ann. NY Acad. Sci. (1998) 852;
M. Ratner, Nature 404 (2000) 137;
J.M. Tour, Acc. Chem. Res. 33 (2000) 791;
K.W. Hipps, Science 284 (2001) 536;
R.T. Farley, Q.L. Zheng, J.A. Gladysz, K.S. Schanze, Inorg. Chem. 47 (2008)
2955;
J. Stahl, J.C. Bohling, T.B. Peters, L. de Quadrasi, J.A. Gladysz, Pure Appl. Chem.
80 (2008) 459.
D. Cahen, G. Hodes, Adv. Mater. 14 (2002) 789;
R.L. Caroll, C.B. Gorman, Angew. Chem., Int. Ed. Engl. 41 (2002) 4378;
N. Robertson, G.A.Mc. Gowan, Chem. Soc. Rev. 32 (2003) 96;
M.A. Reed, T. Lee (Eds.), Molecular Nanoelectronics, American Scientific
Publishers, Stevenson Ranch, CA, 2003;
A.H. Flood, J.F. Stoddart, D.W. Steuerman, J.R. Heath, Science 306 (2004) 2055;
K. Nørgaard, T. Bjørnholm, Chem. Commun. (2005) 1812;
M.C. Petty, Molecular Electronics, From Principles to Practice, Wiley
Interscience, New York, 2008.
[7] M. Akita, Y. Tanaka, C. Naitoh, T. Ozawa, N. Hayashi, M. Takeshita, A. Inagaki,
M.-C. Chung, Organometallics 25 (2006) 5261;
Y. Tanaka, T. Ozawa, A. Inagaki, M. Akita, Dalton Trans. (2007) 928;
T. Ozawa, M. Akita, Chem. Lett. (2004) 1180.
[8] M. Akita, T. Koike, Dalton Trans. (2008) 3523.
[9] Y. Tanaka, A. Inagaki, M. Akita, Chem. Commun. (2007) 1169;
K. Motoyama, T. Koike, M. Akita, Chem. Commun. (2008) 5812.
[10] Attempted preparation of, for example, 3a via Sonogashira coupling between
[2] (b) P.F.H. Schwab, J.R. Smith, J. Michl, Chem. Rev. 105 (2005) 1197;
(a) P.F.H. Schwab, M.D. Levin, J. Michl, Chem. Rev. 99 (1999) 1863.
[3] M.I. Bruce, P.J. Low, Adv. Organomet. Chem. 50 (2004) 231.
[4] (a) F. Paul, C. Lapinte, Coord. Chem. Rev. 178 (1998) 431;
(b) N. Le Narvor, L. Toupet, C. Lapinte, Chem. Commun. (1993) 357;
(c) N. Le Narvor, L. Toupet, C. Lapinte, J. Am. Chem. Soc. 117 (1995) 7129;
(d) N. Le Narvor, C. Lapinte, Organometallics 14 (1995) 634;
(e) N. Le Narvor, C. Lapinte, Compt. Rend. Ser. II Chim. 1 (1998) 745;
(f) M. Guillemot, L. Toupet, C. Lapinte, Organometallics 17 (1998) 1928;
(g) T. Weyland, K. Costuas, A. Mari, J.F. Halet, C. Lapinte, Organometallics 17
(1998) 5569;
1,4-dimethylamino-3,5-diiodobenzene
and
Fe–C„C–H
resulted
in
homocoupling of the latter to give the butadiynediyl complex, Fe–C„C–
C„C–Fe [4b,c].
Use of KOBut in place of NaOMe lowered the yield of 3. The low yields of 3
should be due to their low solubility in organic solvents, which should hinder
effective extraction.
[11]
[12] Formation of A and B suggests occurrence of coupling of the benzodifuran
moiety and the Cp* ligand. The organic product A might result from migratory
insertion of the Cp* ligand to the carbene moiety in 22+g0 followed by
deprotonation and demetalation, and the Fe(0) species formed should be
(h) S. Le Stang, F. Paul, C. Lapinte, Organometallics 19 (2000) 1035;
(i) R. Denis, L. Toupet, F. Paul, C. Lapinte, Organometallics 19 (2000) 4240;

Y. Matsuura et al. / Journal of Organometallic Chemistry 694 (2009) 1840–1847
1847
oxidized during the work up and trapped by dppe, Clꢂ and MeCN present in
the mixture to be converted into B.
[21] C. Roger, P. Hamon, L. Toupet, L. Rabaa, J.-Y. Saillard, J.-R. Hamon, C. Lapinte,
Organometallics 10 (1991) 1045.
[13] W. Petz, Iron–Carbene Complexes, Springer, Berlin, 1992.
[14] N.G. Connely, W.E. Geiger, Chem. Rev. 96 (1996) 877;
F. Barriére, W.E. Geiger, J. Am. Chem. Soc. 128 (2006) 3980.
[15] Because of the low solubility of the neutral species 3 in common organic
solvents CV measurements were examined for the monocationic species 3+.
[16] N.S. Hush, Prog. Inorg. Chem. 8 (1967) 357;
[22] K. Tamiya, Y. Konda, H. Ebata, N. Niihara, T. Otsubo, J. Org. Chem. 70 (2005)
10569.
[23] 1b T.B. Marder, New J. Chem. 27 (2003) 140;
1cS. Takahashi, Y. Kuroyama, K. Sonogashira, N. Hagihara, Synthesis (1980)
627;
diiodo precursorJ.R. Perry, D.B. Wilson, Macromolecules 28 (1995) 3509;
1dM.J. Plater, J.P. Sinclair, S. Aiken, T. Gelbrich, M.B. Hursthouse, Tetrahedron
60 (2004) 6385;
N.S. Hush, Prog. Inorg. Chem. 8 (1967) 391;
M.B. Robin, P. Day, Adv. Inorg. Chem. Radiochem. (1967) 247.
[17] R.C. Rocha, F.N. Rein, H. Jude, A.P. Shreve, J.J. Concepcion, T.J. Meyer, Angew.
Chem., Int. Ed. 47 (2008) 503.
D.L. Mattern, J. Org. Chem. 49 (1984) 3051;
1eT.X. Neenan, G.M. Whitesides, J. Org. Chem. 53 (1988) 2489;
1gG.T. Crisp, T.P. Bubner, Tetrahedron 53 (1997) 11899.
[24] Despite several attempts analytically pure samples could not be obtained.
[25] LDA prepared from n-BuLi gave a low yield mixture of products.
[18] Plots of
D
E, E11=2, and E12=2 data against rI and r0R are shown in Fig. S3.
[19] C. Hansch, S. Leo, R.W. Taft, Chem. Rev. 91 (1991) 165.
[20] M.-C. Chung, X. Gi, B.A. Etzenhouser, A.M. Spuches, P.T. Rye, S.K. Seetharaman,
D.J. Rose, J. Zubieta, M.B. Sponsler, Organometallics 22 (2003) 3485.