Construction of di- and tetra-ferrocenyl spiroborate complexes from catechol building blocks and their redox behaviors

Article abstract of DOI:10.1039/c3dt52503a

In this paper, we report a method for the construction of multi-ferrocenyl spiroborate assemblies from discrete mono- and di-ferrocenyl complexes. To prepare the building blocks for the spiroborate assemblies, 4- ferrocenylveratrole (1) and 4,5-diferrocenylveratrole (3) were first synthesized by Negishi cross-coupling between ferrocene and the corresponding aryl halides. Compounds 1 and 3 were characterized by NMR and UV-vis spectroscopies and electrochemical methods, namely cyclic voltammetry and differential pulse voltammetry. The potential splittings of 167 and 263 mV for the oxidations of the two ferrocenyl groups of 3 in CH₂Cl₂-n-Bu ₄NPF₆ and n-Bu₄NB(C₆F ₅)₄, together with the presence of an intervalence charge-transfer transition of 3⁺ in the NIR region, indicate a strong inter-ferrocenyl interaction through the ortho connection in the veratrole bridge. The treatment of 1 and 3 with BBr₃, followed by complexation with B(OH)₃ in the presence of KOH, afforded potassium salts of di- and tetra-ferrocenyl spiroborate assemblies, K2 and K4, respectively. The spiroborate assemblies K2 and K4 were characterized by NMR and UV-vis spectroscopies, ESI-MS, and electrochemical methods. K2 exhibited a potential splitting value of 302 mV in DMF-n-Bu₄NPF₆, indicating electronic interactions over the boron center. K4 exhibited four one-electron-oxidation waves in DMF-n-Bu₄NPF₆ and n-Bu₄NB(C₆F₅)₄; the one-electron-oxidized species 4 is thermodynamically more stable than 4 ⁺ and 4²⁺. The overall E_1/2¹-E _1/2⁴ spread for K4 is 201 mV larger than the E _1/2¹-E_1/2² splitting for 3 in DMF-n-Bu₄NPF₆, indicating that the spiroborate link between diferrocenyl units led to electronic interactions over the boron center. The Royal Society of Chemistry 2014.

Full text of DOI:10.1039/c3dt52503a

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Construction of di- and tetra-ferrocenyl
spiroborate complexes from catechol building
blocks and their redox behaviors†
Cite this: Dalton Trans., 2014, 43,
1368
Keishiro Tahara,* Tetsuhiro Akita, Shohei Katao and Jun-ichi Kikuchi*
In this paper, we report a method for the construction of multi-ferrocenyl spiroborate assemblies from
discrete mono- and di-ferrocenyl complexes. To prepare the building blocks for the spiroborate assem-
blies, 4-ferrocenylveratrole (1) and 4,5-diferrocenylveratrole (3) were first synthesized by Negishi cross-
coupling between ferrocene and the corresponding aryl halides. Compounds 1 and 3 were characterized
by NMR and UV-vis spectroscopies and electrochemical methods, namely cyclic voltammetry and differen-
tial pulse voltammetry. The potential splittings of 167 and 263 mV for the oxidations of the two ferro-
cenyl groups of 3 in CH2Cl2–n-Bu4NPF6 and n-Bu4NB(C6F5)4, together with the presence of an
intervalence charge-transfer transition of 3+ in the NIR region, indicate a strong inter-ferrocenyl inter-
action through the ortho connection in the veratrole bridge. The treatment of 1 and 3 with BBr3, fol-
lowed by complexation with B(OH)3 in the presence of KOH, afforded potassium salts of di- and tetra-
ferrocenyl spiroborate assemblies, K2 and K4, respectively. The spiroborate assemblies K2 and K4 were
characterized by NMR and UV-vis spectroscopies, ESI-MS, and electrochemical methods. K2 exhibited a
potential splitting value of 302 mV in DMF–n-Bu4NPF6, indicating electronic interactions over the boron
center. K4 exhibited four one-electron-oxidation waves in DMF–n-Bu4NPF6 and n-Bu4NB(C6F5)4; the one-
Received 12th September 2013,
Accepted 18th October 2013
electron-oxidized species 4 is thermodynamically more stable than 4+ and 42+. The overall E1/21–E1/2
4
2
spread for K4 is 201 mV larger than the E1/21–E1/2 splitting for 3 in DMF–n-Bu4NPF6, indicating that the
DOI: 10.1039/c3dt52503a
www.rsc.org/dalton
spiroborate link between diferrocenyl units led to electronic interactions over the boron center.
mixed-valence monomeric complexes such as oxo-centered
triruthenium clusters and diiron–sulfur complexes have been
Introduction
The construction of covalent or non-covalent assemblies from shown to dimerize through effective bidentate linkers. Versa-
preformed coordination compounds as building blocks is a tile methodologies for building up assemblies from mixed-
promising method for the realization of molecular elec- valence subunits are still not fully established, so research to
tronics.1 In particular, the construction of assemblies from find effective linkers for the synthesis of new structural motifs
discrete mixed-valence complexes has attracted increasing is important.
attention because of their potential applications in quantum
Diferrocenyl compounds are attractive candidates for the
cellular automata (QCA).2 For the molecular expression of QCA study of intramolecular electron transfer in mixed-valence
devices as wires and circuits, one of the first steps is the complexes.4 The benefits of the study of diferrocenyl com-
construction of a dimer of a discrete mixed-valence complex. pounds come primarily from various advantages of ferrocene
In this context, the pioneering work on supramolecular dimer such as the high reversibility and stability of the Fe(III)/Fe(II)
structures offers insights into the spatial arrangements of dis- couple and the low oxidation potential, allowing effective elec-
crete mixed-valence complexes as building blocks.3 Some tronic coupling with organic π-conjugated components.4
A
wide range of mixed-valence diferrocenyl compounds, includ-
ing biferrocene,5 biferrocenylene,6 and compounds with
different heteroatoms and π-conjugated bridges,7–12 have been
synthesized using the versatile aromatic reactivity of ferrocene.
Although there have been a number of reports on mixed-
valence diferrocenyl compounds, few attempts to construct
dimeric assemblies using them as building blocks have been
reported.13–15 A key aspect of this task is the design of a
Graduate School of Materials Science, Nara Institute of Science and Technology
(NAIST), 8916-5, Takayama, Ikoma, Nara 630-0192, Japan.
E-mail: taharak@ms.naist.jp
†Electronic supplementary information (ESI) available: Additional experimental
details and NMR and UV-vis-NIR spectroscopic data. CCDC 955166. For ESI
and crystallographic data in CIF or other electronic format see DOI:
10.1039/c3dt52503a
1368 | Dalton Trans., 2014, 43, 1368–1379
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Scheme 1 Syntheses of ferrocenylveratrole (1), diferrocenyl spiroborate (K2), diferrocenylveratrole (3) and tetraferrocenyl spiroborate (K4), and
atom numbering used for 1H-NMR assignments.
diferrocenyl monomer that can be dimerized using effective the presence of a palladium catalyst under Negishi coupling
linkages. In their pioneering work on developing such a strat- conditions, followed by workup procedures, afforded 3 as an
egy, Heinze et al. used an elaborate combination of biferrocene orange crystalline solid in 53% yield. In contrast, the yield of
amino acids and ureylene amide linkages.13
1 obtained using the same procedure with 4-bromoveratrole
To explore new synthetic methods for assembly from diferro- was only 5%. Suzuki coupling of ferroceneboronic acid and
cenyl building blocks, we focused on spiroborate linkages 4-bromoveratrole afforded 1 in 15% yield. This improved yield
formed via complexation of B(OH)3 with diols or polyols.16 of 1 is comparable to the yield obtained by a previous method
Spiroborate linkages have been shown to be useful motifs for using a coupling reaction between ferrocene and the diazo-
the dimerization of functional diol subunits as a result of the nium salt of 1,2-dimethoxyaniline.19 Compounds 1 and 3 were
1
ease of their reversible formation and the resulting robust identified by H- and 13C{1H}-spectroscopies (Fig. S1 and S2†)
structures, especially in the field of molecular recognition.17,18 and 1D 1H-NMR differential NOE experiments (Fig. S7 and
In this study, we report a method for the construction of a S8†). Single crystals of the new compound 3, suitable for X-ray
tetraferrocenyl spiroborate assembly from a discrete diferroce- diffraction (XRD) analysis, were obtained from a chloroform–
nyl complex, as shown in Scheme 1. To design a diferrocenyl hexane solution of 3 at room temperature. The solid-state
building block, we chose a catechol backbone, not only as an molecular structure of 3 is shown in Fig. 1. As is common for
electron-rich π-conjugated connector of the two ferrocenyl o-diferrocenylbenzene derivatives,9b in 3, one ferrocenyl group
groups, but also as a chelating ligand for a boron center. In is above the benzene plane and the other is below it, whereas
this manner, we synthesized 4,5-diferrocenylveratrole (FcVFc, m-diferrocenylbenzene was reported to adopt a conformation
3) and the monoferrocenyl counterpart (FcV, 1), and clarified with the two ferrocenyl groups at the same side with respect to
the effect of inter-ferrocenyl interactions through the ortho
connection in the veratrole bridge on the electronic structure
and electrochemical behavior. We also report the electrochemi-
cal behaviors of di- and tetra-ferrocenyl spiroborate complexes
K2 and K4, and compare them with those of the corresponding
building blocks 1 and 3, to investigate the effect of spiroborate
linking on the redox properties.
Results and discussion
Synthesis and characterization of mono- and
di-ferrocenylveratroles (1 and 3)
First, we synthesized 4-ferrocenylveratrole (1) and 4,5-diferro-
cenylveratrole (3) using the reported ferrocenyl-coupling proce-
dures9a,10 from the corresponding aryl halides, as shown in
Scheme 1. The treatment of ferrocenylzinc chloride (FcZnCl)
with 4,5-diiodoveratrole (1,2-diiodo-4,5-dimethoxybenzene) in ellipsoids set at the 50% probability level.
Fig. 1 ORTEP drawing of molecular structure of 3, with thermal
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1368–1379 | 1369

View Article Online
Paper
Dalton Transactions
the benzene plane in the solid state.9c In 3, steric constrains
The DFT calculation for 1 indicates contributions from the
between the two ferrocenyl groups, particularly between H23 iron-centered dxy and dx2–y2 orbitals in the HOMO, HOMO−1,
and C5 (2.73 Å), and H23 and C6 (2.76 Å), increase the twist and HOMO−2. The HOMOs of ferrocene and veratrole are well
angles between the benzene and cyclopentadienyl (Cp) rings hybridized, constructing the HOMO and HOMO−2 in 1 as
(32.06° for Fe1 and 52.65° for Fe2). This steric repulsion has anti-bonding and bonding combinations, as shown in
almost no effect on the planarity of the benzene ring, with a Fig. S12.† The HOMO is localized at the ferrocene moiety with
torsion angle C5–C4–C18–C19 of −2.76° (Table S2†). Both the an extension on to the benzene ring and the methoxy groups.
ferrocenyl groups showed an almost eclipsed conformation The contributions from the ferrocene and veratrole moieties to
(2.81° for Fe1 and 11.03° for Fe2). The Fe–Fe distance was 6.72 Å. the HOMO are 52% and 48%, respectively (Table S4†). In con-
The DFT optimized geometry of 3, using the B3LYP/ trast to that for 1, the DFT calculation for ferrocenylbenzene as
LanL2DZ (Fe atom) and 6-31G(d) (all other atoms) levels of a reference compound revealed that the HOMO is more exten-
theory, agrees well with the geometry determined by XRD ana- sively localized at the ferrocene moiety (93%), with a greatly
lysis; the Fe–Fe distance was 6.86 Å and the two ferrocenyl decreased contribution from the benzene moiety (7%). The
groups were above and below the benzene plane without interaction of ferrocene with conjugated substituents is more
taking account the solvent effects (Fig. S11 and Table S3†). complex than that of purely organic compounds because both
A similar optimized geometry of 3 was obtained using the the metal-based and the ligand-based orbitals of ferrocene
polarized continuum model (PCM) approach (CH2Cl2) (Fe–Fe interact with the orbitals of the π-system.20 It should be noted
distance 6.89 Å). These results imply that the two ferrocenyl that the proximity of the energy levels of the HOMOs of ferro-
groups take an up–down arrangement with respect to the cene and veratrole is effective for the construction of hybrid
benzene ring in solution. In the DFT optimized structure in orbitals (HOMO and HOMO−2) in 1, whereas the HOMO of
the gas phase, the bimetallic complex 3 showed a more sym- benzene interacts with the lower-lying ligand-based orbitals of
metric geometry than that in the crystal structure because of ferrocene in ferrocenylbenzene. In the case of 3, contributions
the absence of crystal-packing effects; both the ferrocenyl from the iron-centered dxy and dx2–y2 orbitals are observed in
groups showed an eclipsed conformation and the twist angles the HOMO to HOMO−4. The HOMOs of ferrocene and vera-
between the benzene and Cp rings were comparable (42.1° for trole are well hybridized, constructing the HOMO, HOMO−1,
both ferrocenyl groups). In the DFT optimized structure of 1 in and HOMO−4 in 3, as shown in Fig. S12.† The HOMO is de-
the gas phase the twist angle between the benzene and the Cp localized across the ferrocenyl groups through the benzene
ring (26.9°) was lower than that of 3, reflecting the lack of ring, and the methoxy groups also contribute to the HOMO.
steric constraints between the two ferrocenyl groups in 1.
The contributions from the two ferrocenyl groups and the vera-
To investigate their electronic structures, the UV-vis spectra trole moiety to the HOMO are 55% and 45%, respectively. The
of the mono- and di-ferrocenyl compounds 1 and 3 were veratrole moiety therefore serves as an electron-rich bridge,
obtained in CH2Cl2. In agreement with a previous report,19 which can mediate electronic interactions between the ferroce-
compound 1 exhibited absorption bands at 286 and 449 nm, nyl groups in 3. The optical properties of 1 and 3 agree well
which were attributed to π → π* and symmetry-forbidden d–d the transition energies and intensities of 1 and 3 obtained
transitions, as shown in Fig. 2. These bands are typical for aryl- from TD-DFT calculations, as shown in Fig. S13 and Table S5.†
ferrocene derivatives.20 The molar extinction coefficients of the
The electrochemical behavior of the diferrocenyl compound
d–d transitions for 1 and 3 were, respectively, 3.4 and 5.5 times 3 was investigated using cyclic voltammetry (CV) and differen-
higher than that for ferrocene itself. In contrast, no discernible tial pulse voltammetry (DPV) in CH2Cl2 containing n-Bu4NPF6
peak shifts of the π → π* and d–d transitions for 1 and 3 rela- and n-Bu4NB(C6F5)4, respectively, at 298 K. The data for the
tive to ferrocene were observed.
latter are shown in Fig. 3. In both CH2Cl2–n-Bu4NPF6 and
CH2Cl2–n-Bu4NB(C6F5)4 solutions, the diferrocenyl complex 3
Fig. 2 UV-vis spectra of ferrocene derivatives 1 and 3 in CH2Cl2. The
molar extinction coefficients in the 389–600 nm range were multiplied
by a factor of 10.
Fig. 3 Cyclic voltammograms (left) and differential pulse voltammograms
(right) of 1 and 3 in CH2Cl2 containing n-Bu4NB(C6F5)4 (0.1 M). Scan
rate: 100 mV s−1; [1] = [3] = 1.0 mM.
1370 | Dalton Trans., 2014, 43, 1368–1379
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Table 1 Electrochemical data for 1 and 2a
ortho connection in the veratrole bridge. This interpretation is
consistent with the delocalization of the HOMO of 3 across the
ferrocenyl groups through the veratrole bridge, mentioned
above.
Counter anion of
electrolyte
1
2
b
c
Cmpd
E1/2
E1/2
ΔE1/2
Kc
−
1
1
3
3
PF6
568
569
486
492
—
—
653
755
—
—
167
263
—
—
To investigate its optical properties and electron transfer
characteristics, compound 3 was chemically oxidized to the
mixed-valence monocation 3+ using AgSbF6. The mixed-
valence cation 3+ in CH2Cl2 displayed three new absorption
bands, as shown in Fig. S14.† The first one is a shoulder at
around 600 nm in the visible region, the second one is a band
at 946 nm in the near infrared (NIR) region, and the third one
is a broad band at 1685 nm in the NIR region. The first and
second bands of 3+ are typical of ferrocenium–polyaromatic
dyads with electron-rich aryl groups.23 In the case of the one-
electron-oxidized species from 4-ferrocenylveratrole 1, it was
reported that an NIR band at 920 nm in acetonitrile could be
assigned to an intramolecular charge-transfer (ICT) transition
from the aryl group to the ferrocenium center.19 We also gener-
ated the monocation 1+ using AgSbF6 in CH2Cl2, and observed
similar shoulder bands at around 600 nm and a band at
992 nm, as shown in Fig. S14.† The third band of 3+ in the
NIR region is typical of weakly coupled class II mixed-valence
complexes of diferrocenylbenzene.9b In the case of the mixed-
valence cation of o-diferrocenylbenzene, it has been reported
that an NIR band at 1529 nm in acetonitrile could be assigned
to intervalence charge-transfer (IVCT) between the metal
centers. A Gaussian deconvolution of the IVCT bands of 3+
gave the spectroscopic parameters of energy, intensity, and
−
B(C6F5)4
PF6
6.66 × 102
2.79 × 104
−
−
B(C6F5)4
a Potentials in mV vs. Fc*+/Fc*. Use of tetrabutylammonium salts as
electrolytes. b ΔE1/2 = potential difference between two redox processes.
c Comproportionation constants obtained from eqn (3).
exhibited two well-defined reversible waves, indicating sequen-
tial oxidation of the two ferrocene groups and the formation of
one- and two-electron-oxidized species, 3+ (Fc+VFc) and
32+ (Fc+VFc+) (eqn (1)). The first and the second redox poten-
1
tials (E1/2 and E1/22) are listed in Table 1. The first oxidation
potential peak of veratrole (1,2-dimethoxybenzene) appeared at
a more positive potential (≈ 1.68 and 1.67 V vs. Fc*+/Fc* with
n-Bu4NPF6 and n-Bu4NB(C6F5)4, respectively; Fc* denotes deca-
methylferrocene) than that of 3. In the CH2Cl2–n-Bu4NB(C6F5)4
1
2
solution, the E1/2 value of 3 is 77 mV lower, and the E1/2
value of 3 is 186 mV higher, than the redox potential of the
monoferrocenyl compound 1, giving a 263 mV potential split-
ting (Fig. 3). These results clearly indicate inter-ferrocenyl
interactions through the ortho connection in the veratrole
bridge. The potential splitting between the two chemically
2
equivalent redox sites, ΔE (= E1/2 − E1/21), can be used as an
indicator of electronic communication between the redox
sites, although some electrostatic repulsion also contributes to
the potential splitting. When the mixed-valence species 3+ is
in equilibrium with the homovalent species 3 and 32+
(eqn (2)), the comproportionation constant Kc can be obtained
from ΔE (eqn (3)).
bandwidth at the half-height (vmax = 5934 cm−1, εmax
=
158 M−1 cm−1, Δv1/2 = 4308 cm−1). Hush analysis using these
parameters and the Fe–Fe distance from the XRD date of 3
gave an electronic coupling of HAB = 195 cm−1. The HAB value
for 3+ is comparable to that for the mixed-valence complexes
of o-diferrocenylbenzene (HAB = 202 cm−1, vmax = 6540 cm−1
,
εmax = 91 M−1 cm−1, Δv1/2 = 6750 cm−1 in acetonitrile). The
monocation 3+, with a veratrole backbone, therefore retains the
mixed-valence characteristics of diferrocenylbenzene derivatives.
1
2
FcVFc Eꢀ1!=2 FcþVFc Eꢀ1!=2 FcþVFcþ
ð1Þ
ð2Þ
ꢀeꢀ
ꢀeꢀ
Kc
FcVFc þ FcþVFcþ ꢀ! 2FcþVFc
Synthesis and characterization of di- and tetra-ferrocenyl
spiroborate assemblies K2 and K4
2
Kc ¼ ½FcþVFcꢁ =½FcVFcꢁ½FcþVFcþꢁ ¼ expðΔE ꢂ F=RTÞ ð3Þ
It has recently been demonstrated that the combination of To prepare a spiroborate assembly from compound 1, depro-
CH2Cl2–n-Bu4NB(C6F5)4 reduces solvent–analyte interactions tection of the methoxy group of 1 was performed using BBr3 in
−
as a result of the low ion-paring capability between B(C6F5)4
CH2Cl2. Although the deprotection was successful, the product
and positively charged analytes.21 This helps electrochemical solution turned from orange to green during workup in air,
interpretations of complex multi-redox processes with posi- indicating the formation of a typical ferrocenium species.
tively charged analytes. The obtained Kc value for 3 in CH2Cl2– Similar instability of 4-ferrocenylcatechol in air was reported
n-Bu4NB(C6F5)4 was 42 times higher than that in CH2Cl2–n- by Colbran et al.19 We therefore performed the deprotection of
Bu4NPF6. This large difference in the thermodynamic stabi- the methoxy group of 1 using degassed solutions in a N2
lities of 1+ implies some electrostatic repulsion between the atmosphere, and used the obtained crude product, 4-ferro-
ferrocenyl units of 32+. It has been reported that when CH2Cl2– cenylcatechol, for the next step. The obtained catechol was
n-Bu4NPF6 is used as a medium, the potential splitting value treated with B(OH)3 and potassium hydroxide in degassed
corresponds roughly to the electronic communication through acetone. Recrystallization afforded the potassium salt of the
a conjugated bridge because the strong ion-pairing results in a diferrocenyl spiroborate (K[FcCBCFc], K2). In contrast to 4-fer-
very little electrostatic effect.22 The relatively large potential rocenylcatechol, K2 is stable in air as a result of coordination
splitting of 3 observed with n-Bu4NPF6 (167 mV) therefore of the catecholate ligands to the boron center. Compound K2
implies inter-ferrocenyl electronic communication through the was identified by 1H-, 13C{1H}-, 11B-NMR spectroscopies and
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1368–1379 | 1371

View Article Online
Paper
Dalton Transactions
Fig. 4 1H-NMR spectra of 1, K2, 3, and K4 in acetone-d6.
differential NOE experiments (Fig. S3, S4 and S9†). Except for
the disappearance of signals derived from methoxy groups, all
the proton signals of K2 were shifted upfield compared with
those of 1, as shown in Fig. 4, indicating the effects of the
negative charge of the boron center. A characteristic peak from
spiroborate salts was also observed in the 11B-NMR spectrum
of K2, indicating successful formation of the spiroborate struc-
ture. The ESI-MS measurements confirmed the successful
complexation of B(OH)3 and 4-ferrocenylcatechol in a 1 : 2
ratio, as shown in Fig. 5. Compound K2 exhibited two absorp-
tion bands at 310 and 443 nm in DMF, as shown in Fig. 6. The
former is attributed to π → π* transitions of the K[B(O2C6H4)2]
moiety, and the latter to symmetry-forbidden d–d transitions
of the ferrocenyl groups. The compound K2 showed good solu-
bility in polar solvents such as DMF and acetone, but poor
Fig. 6 UV-vis spectra of K2, K4, and K[B(O2C6H4)2] in DMF.
solubility to more apolar solvents such as CH2Cl2 and CHCl3,
reflecting the borate anion structure of K2. Recently, it has
been demonstrated that boron-containing linkages, including
boroxine,24 boronic ester,25 a double-boron bridge,26 a B–O–B
linkage formed via condensation of bis(boronic acid),27 and
three- or four-coordinated boron atom spacers,7a,b are useful
for the synthesis of multi-ferrocenyl compounds. However, to
the best of our knowledge, ferrocene or metallocene com-
pounds containing spiroborate linkages have not been
reported to date. Although an aryl spiroborate, [B(O2C6H4)2]−,
has been used as a ligand for transition-metal catalysts,28
a
new derivative containing two ferrocenyl groups (K2) gives
accessibility to electrochemistry, as discussed below.
The potassium salt of the diferrocenyl spiroborate
K[FcCFcBFcCFc] (K4) was successfully obtained from 3 using
the same procedure as for K2. Compound K4 was identified by
ESI-MS, and 1H-, 13C{1H}-, and 11B-NMR spectroscopies
(Fig. S5 and S6†). Although 4,5-diferrocenylcatechol decom-
posed rapidly in air, compound K4 was stable in air as a result
of the coordination of the catecholate ligands with the boron
1
center. The H-NMR spectrum of K4 in acetone-d6 exhibited a
Fig. 5 ESI-MS of acetone solutions of K2 and K4. Inset: experimental
(bottom) and theoretical (top) isotopic distributions for 2− and 4−.
similar signal pattern to that of 3, except for the disappearance
1372 | Dalton Trans., 2014, 43, 1368–1379
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
of signals derived from methoxy groups, as shown in Fig. 4. in a polar medium. In both DMF–n-Bu4NPF6 and DMF–n-
This confirms the highly symmetric structure of compound Bu4NB(C6F5)4 solutions, compound K2 displayed two oxidation
K4, and the solution does not contain boron-containing processes. The reference borate compound K[B(O2C6H4)2]
species except for the 1 : 2 complex K4. DOSY-NMR measure- exhibited the first oxidation peak at a more positive potential
ments revealed that K4 has a diffusion constant of 7.3 × 10−10 (≈ 1.00 and 1.03 V vs. Fc*+/Fc* with n-Bu4NPF6 and n-Bu4NB-
cm2 s−1 in DMF-d7 solution at 298 K, which is different from (C6F5)4, respectively), so the two oxidation peaks of K2 indicate
that of the precursor 3 (8.8 × 10−10 cm2 s−1). This result also sequential oxidation of two ferrocenyl groups and the for-
confirms that the solution of K4 contains only a 1 : 2 complex. mation of 2 (Fc+CB−CFc) and 2+ (Fc+CB−CFc+) (eqn (4)). The
1
The ESI-MS also confirms the successful complexation of first and second redox potentials (E1/2 and E1/22) are listed in
1
B(OH)3 and 4,5-diferrocenylcatechol in a 1 : 2 ratio, as shown Table 2. In DMF–n-Bu4NPF6, the E1/2 value of K2 is 78 mV
2
in Fig. 5. Compound K4 showed similar solubility in organic lower, and the E1/2 value of K2 is 224 mV higher, than the
solvents as that of K2. As in the case of K2, compound K4 redox potential of compound 1, giving a potential splitting of
−
exhibited two absorption bands at 295 and 431 nm in DMF, as 302 mV. The use of B(C6F5)4 as the counter anion of the sup-
shown in Fig. 6. The molar extinction coefficient of the latter porting electrolyte increased the potential splitting by 10 mV.
d–d transition band is 1.98 times larger than that for K2; this Such large separations of the oxidation potentials in polar
corresponds to the number of ferrocene groups.
media indicate strong electronic interactions between two
To date, a number of tetraferrocenyl compounds displaying ferrocenylcatecholate (FcC) units over the boron center in K2.
inter-ferrocenyl interactions have been reported, including An electrostatic contribution to splitting of the redox waves
those with various π-conjugated bridges,10a–d,11b,12a,29 amide seems unlikely because of the large separation among the fer-
and ureylene bridges,13 heteroatom spacers,7b–e,30 and other rocenyl sites within the K2 molecule. Similar electrochemical
bridges.31 In contrast to the variety of synthetic routes to tetra- behavior was reported for a 9-oxidophenalenone complex of
ferrocenyl compounds, few have been reported for dimeriza- boron containing two phenalenyl units in the same molecule
tions using diferrocenyl building blocks.13–15 Notably, Mueller- by the spiro-linkage.32 It was reported that in the spiroborate
Westerhoff used the coordination of a diferrocenyl dithiolate complex, the spin density was delocalized over the two units as
to a Ni center to afford a tetraferrocenyl complex.14a In the a result of spiro-conjugative interactions among oxygen func-
present study, we used the coordination of a diferrocenyl cate- tionalities linked by the boron center.32a Similarly, in the
cholate to a boron center to afford K4, whose structural feature present study, the two FcC units do not act independently but
is a tetracoordinated boron anionic bridge.
are coupled through a spiroborate linkage insofar the HOMO
of K2 is concerned. This interpretation might be reasonable
when one considers the large contributions from the veratrole
moiety to the HOMO in the parent 1 system, as discussed
above, enabling overlap between the π-orbitals of the two FcC
units. Indeed, the effects of spiro-conjugation of boron-
centered tetrahedral geometry have been reported elsewhere.16b,33
Electrochemical behaviors of 1 and K2 in DMF
The electrochemical behaviors of the monoferrocenyl complex
1 and the diferrocenyl spiroborate complex K2 were investi-
gated in DMF by CV and DPV; the results are shown in Fig. 7.
Because of the poor solubility of K2 in CH2Cl2, the comparison
of the electrochemical properties of 1 and K2 was carried out
1
2
FcCBꢀCFc Eꢀ1!=2 FcþCBꢀCFc Eꢀ1!=2 FcþCBꢀCFcþ
ð4Þ
ð5Þ
ꢀeꢀ
ꢀeꢀ
Kc
FcCBꢀCFc þ FcþCBꢀCFcþ ꢀ! 2FcþCBꢀCFc
To investigate its optical properties and electron transfer
characteristics, compound K2 was chemically oxidized to the
mixed-valence neutral radical 2, using AgSbF6 in DMF at room
temperature. The one-electron-oxidized species 2 exhibited a
new absorption band at 681 nm, characteristic of ferrocenium
species, as shown in Fig. S15.† The neutral radical 2 also
exhibited a weak NIR absorption band (vmax = 7207 cm−1
,
εmax = 73 M−1 cm−1, Δv1/2 = 2201 cm−1), which is a character-
istic feature of 2. Such an absorption band was not observed
for compound 1 upon one-electron-oxidation using AgSbF6 in
DMF. The reference compound K[B(O2C6H4)2] did not exhibit
any absorption band in the NIR region in the presence of
AgSbF6. The NIR absorption of 2 is therefore assigned to the
IVCT transition between the FcC units. The presence of
the IVCT absorption of 2, together with the splitting seen in
the first and second oxidation potentials of K2, indicates elec-
tronic interactions between the FcC units over the boron
Fig. 7 Cyclic voltammograms (left) and differential pulse voltammograms
(right) of 1, K2, 3, and K4 in DMF containing n-Bu4NPF6 (0.1 M). Scan
rate: 100 mV s−1. [1] = [K2] = [3] = [K4] = 1.0 mM.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1368–1379 | 1373

View Article Online
Paper
Dalton Transactions
Table 2 Electrochemical data for 1, K2, 3, and K4 in DMFa
1
2
3
4
b
b
b
# of Fc
Compound
Counter anion of electrolyte
E1/2
E1/2
E1/2
E1/2
Kc1
Kc2
Kc3
−
Mono
Mono
Di
Di
Di
Di
Tetra
Tetra
1
1
K2
K2
3
3
K4
K4
PF6
454
452
376
364
436
436
363
357
—
—
—
—
—
—
—
—
587
581
—
—
—
—
—
—
680
689
—
—
—
—
—
—
—
—
—
—
—
—
—
—
−
B(C6F5)4
PF6
678
676
552
562
506
506
1.27 × 105
1.88 × 105
91.4
−
−
B(C6F5)4
−
PF6
−
B(C6F5)4
135
261
330
−
PF6
23.4
18.5
37.3
67.0
−
B(C6F5)4
a Potentials in mV vs. Fc*+/Fc*. Use of tetrabutylammonium salts as electrolytes. b Comproportionation constants.
Kc1
center. The disappearance of absorptions over a wide region
was observed within hours for a solution of the neutral radical
2 at room temperature, suggesting decomposition of com-
pound 2. Compound 2 is also expected to be unstable, based
on the low reversibility of the second redox wave in the CV
curve of K2. Further investigation of the two-electron-oxidized
species of K2 was therefore not performed in this study.
4
ꢀ þ 4þ ꢀ! 24
ð7Þ
Kc1
4
þ þ 43þ ꢀ! 242þ
ð8Þ
ð9Þ
Kc3
4
2þ þ 44þ ꢀ! 243þ
1
In DMF–n-Bu4NPF6, the E1/2 value for K4 is 73 mV smaller
than that for 3. The negative charge on the boron center and
the two additional ferrocenyl moieties as electron-donating
groups are advantageous for removing an electron in the first
Electrochemical behaviors of 3 and K4 in DMF
The electrochemical behaviors of the diferrocenyl complex 3
and the tetraferrocenyl spiroborate complex K4 were investi-
gated in DMF by CV and DPV, as shown in Fig. 7. The diferro-
cenyl complex 3 exhibited two reversible waves, indicating
sequential oxidations of the two ferrocene groups and for-
mation of 3+ (Fc+VFc) and 32+ (Fc+VFc+) (eqn (1)). The first and
2
oxidation process of K4. In contrast, the E1/2 value for K4 is
only 46 mV smaller than that for 3. As a result, the potential
1
2
splitting between E1/2 and E1/2 for K4 is 27 mV larger than
that for 3, giving a 2.9 times higher comproportionation con-
3
4
stant for 4 than that for 3+. However, the E1/2 and E1/2 values
for K4 are 35 mV and 128 mV, respectively, larger than the
E1/22 values for 3. The resulting overall E1/21–E1/24 spread for K4
1
second redox potentials (E1/2 and E1/22) are listed in Table 2.
1
In DMF–n-Bu4NPF6, the E1/2 value of 3 is 18 mV lower, and
the E1/22 value of 3 is 98 mV higher, than the redox potential of
the monoferrocenyl compound 1, giving a 116 mV potential
splitting. The potential splitting of 3 in DMF–n-Bu4NPF6 was
51 mV smaller than that in CH2Cl2–n-Bu4NPF6. The strong sol-
vation effect of the mixed-valence species in DMF–n-Bu4NPF6
decreased the thermodynamic stability of 3+.
(317 mV) is therefore 201 mV larger than the overall E1/21–E1/2
2
spread for 3 (116 mV). Similar properties of the potential split-
ting for K4 were observed in DMF–n-Bu4NB(C6F5)4. Heinze
et al. reported dimers of mixed-valence biferrocene amino
acids linked by ureylene and amide bridges, and the effects of
the bridges on the potential splitting.15 In their pioneering
work, they showed that the amide bridge affects the potential
splitting of the resulting tetraferrocenyl dimer (1e−/1e−/2e−
process), but the ureylene bridge has not largely effected (2e−/
2e− process). A comparison of 3 and K4 in the present study
revealed that linking the diferrocenyl units (FcCFc) with the
spiroborate motif increased the potential spread. This implies
electronic interactions between the FcCFc units over the boron
center. In contrast, the four one-electron-oxidation processes
for K4 occurred almost within the potential range between
The tetraferrocenyl complex K4 in DMF–n-Bu4NPF6 exhibi-
ted four one-electron redox waves, as shown in Fig. 7, indicat-
ing sequential oxidations of four ferrocene groups and
formation of the oxidized species (eqn (6)). The redox waves of
the DPV curve were fitted using Gaussian peaks to determine
the accurate oxidation potential. The potential splitting
1
2
between E1/2 and E1/2 (143 mV) is larger than that between
3
1
2
E1/22 and E1/2 (81 mV) or that between E1/2 and E1/2 (93 mV).
The comproportionation constants Kc1, Kc2, and Kc3 (eqn (7)–
(9)) were determined from these splitting values, and are listed
in Table 2. The larger Kc1 value for the mixed-valence neutral
radical 4 indicates thermodynamic stability of the one-elec-
tron-oxidized species, whereas the smaller Kc2 and Kc3 values
for 4+ and 42+ imply potential difficulties in generating these
species. There are two possible oxidation isomers for 4+,
denoted by Fc+CFcB−Fc+CFc and Fc+CFc+B−FcCFc, but the
former geometry is likely to be favored electrostatically.
1
2
E1/2 and E1/2 for K2. This means that the introduction of
third and fourth ferrocenyl groups into K2 does not increase
the electronic interactions over the boron center, but increases
the tendency to form two more degenerate orbitals on both
sides of the boron center. However, because K4 still main-
tained a wide E1/21–E1/2 potential spread in polar media, the
spiroborate-mediated electronic interaction is effective to some
extent in the tetranuclear complex.
4
1
2
3
4
Compound K4 was chemically oxidized to the mixed-
valence neutral radical 4, using 1 equiv. of AgSbF6 in DMF at
E1=2
E1=2
E1=2
4ꢀ ꢀ! 4 Eꢀ1!=2 4þ ꢀ! 42þ ꢀ! 43þ
ð6Þ
ꢀeꢀ
ꢀeꢀ
ꢀeꢀ
ꢀeꢀ
1374 | Dalton Trans., 2014, 43, 1368–1379
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
room temperature. The one-electron-oxidized species 4 exhibi-
The UV-vis-NIR absorption spectra were measured using a
ted a new absorption band at 750 nm, as shown in Fig. S15;† JASCO V-670 spectrometer at room temperature. At least two
this was attributed to the ferrocenium center. The neutral solution concentrations were prepared ([1] = 0.125, 1.25 mM;
radical 4 also exhibited a very weak NIR absorption band [3] = 0.25, 2.5 mM, [K2] = 0.25, 2.5 mM; [K4] = 0.50, 5.0 mM)
(vmax = 6422 cm−1, εmax = 134 M−1 cm−1, Δv1/2 = 1758 cm−1). and 1 equiv. of AgSbF6 was added to each solution to deter-
These spectroscopic properties of 4 resemble those of 2, which mine the molar absorption coefficients of the oxidized species.
suggests that compound 4 also retains electronic interactions The 1H-, 13C{1H}-, and 11B-NMR spectra were recorded using
over the boron center between the diferrocenyl units (FcCFc). JEOL JNM-ECP400 and JNM-ECA600 spectrometers installed at
4
This interpretation is consistent with the E1/21–E1/2 potential the Nara Institute of Science and Technology; tetramethyl-
spread of K4 in the electrochemical study. We could not silane (TMS) was used as an internal standard (0 ppm) for
observe another IVCT transition within the FcCFc unit, poss- 1H-NMR analysis and B(OCH3)3 was used as an external stan-
ibly because of masking by the ferrocenium band. Further dard (0 ppm) for 11B-NMR analysis. 1D 1H-NMR differential
investigation of the chemical oxidation of K4 was not per- NOE spectra were recorded using JEOL JNM-ECA600 spectro-
formed in this study because of the potential difficulties in meters (Fig. S7–10†). The ESI-MS was obtained using a JEOL
generating 4+ and 42+, as judged by their small comproportio- JMS-T100CS spectrometer. Elemental analyses were performed
nation constants.
using a Perkin Elmer 2400 II CHNS/O elemental analyzer.
All voltammetric experiments were carried out using a BAS
It was demonstrated that anionic four-coordinated boron
bridges mediate significant ferrocene–ferrocene interactions, electrochemical analyzer. All experiments were performed
measured by electrochemical methods, but little electron using a conventional three-electrode system at 298 K. A plati-
transfer, measured using NIR spectroscopy, in the di- and num wire (1.6 mm in diameter) was used as the counter elec-
tetra-ferrocenyl compounds.7b,26a,30a Such behavior is very trode, a glassy-carbon electrode (3.0 mm in diameter) as the
similar to our findings regarding catecholate-coordinated working electrode, and an Ag/AgCl (3.0 M NaCl) electrode as
spiroborates K2 and K4, possibly implying a common feature the reference electrode. Non-aqueous organic solutions of
of anionic four-coordinated boron bridges in compounds with ferrocenyl compounds containing n-Bu4NPF6 or n-Bu4NB
multi-redox active moieties.
(C6F5)4 (0.1 M) were deaerated prior to each measurement, and
an argon atmosphere was maintained inside the cell through-
out each measurement. Each experiment was first performed
in the absence of any internal standard and then repeated in
the presence of decamethylferrocene (Fc*). A separate exper-
iment using only ferrocene and decamethylferrocene was also
performed. The potentials are quoted relative to the Fc*+/Fc*
couple. In this setup, the Fc+/Fc couple was observed at 560,
616, 484, and 484 mV vs. Fc*+/Fc*, and the Fc+/Fc couple was
observed at 664, 686, 634, and 640 mV vs. Ag/AgCl in CH2Cl2–
n-Bu4NPF6, CH2Cl2–n-Bu4NB(C6F5)4, DMF–n-Bu4NPF6, and
DMF–n-Bu4NB(C6F5)4, respectively. The potential data quoted
relative to the Fc+/Fc couple are also shown in Tables S6 and
S7.†
Conclusions
In this study, we reported a method for the construction of
multi-ferrocenyl spiroborate assemblies from ferrocenyl and
diferrocenyl subunits, and their redox behaviors. The ferroce-
nyl and diferrocenyl subunits 1 and 3 were synthesized by
Negishi cross-coupling between ferrocene and the corres-
ponding aryl halide derivative of veratrole. The potential split-
tings of 167 and 263 mV in CH2Cl2–n-Bu4NPF6 and n-Bu4NB-
(C6F5)4, respectively, for oxidation of the ferrocenyl groups in
3, together with the presence of an IVCT transition of 3+, indi-
cate strong inter-ferrocenyl interactions through the ortho con-
nection in the veratrole bridge. Treatment of 1 and 3 with
BBr3, followed by complexation with B(OH)3 in the presence of
KOH, afforded the potassium salts of di- and tetra-ferrocenyl
spiroborate assemblies K2 and K4. Electrochemical investi-
gations showed that the spiroborate linkage between the
mono- and di-ferrocenyl units led to electronic interactions
over the boron center.
Synthesis of 4-ferrocenylveratrole, FcV (1)
Negishi coupling. A pentane solution of tert-butyllithium
(1.6 M, 7.0 mL) was added to a mixture of ferrocene (1.35 g,
6.90 mmol) and potassium tert-butoxide (78.3 mg,
0.698 mmol) in dry THF (30 mL) at −78 °C. After stirring for
1 h at −78 °C, a zinc chloride–THF complex (3.26 g,
11.3 mmol) was added to the reaction mixture. After stirring
for 1 h at −78 °C and then 1 h at room temperature, tetrakis-
(triphenylphosphine)palladium(0) (Pd(PPh3)4) (97.8 mg, 8.46 ×
10−5 mol) and 4-bromoveratrole (1-bromo-4,5-dimethoxy-
benzene) (1.08 g, 4.74 mmol) were added to the reaction
mixture. The reaction mixture was heated at 60 °C for 48 h and
evaporated to dryness. After the addition of CH2Cl2 (200 mL)
Experimental
Materials and methods
All solvents and chemicals used in the syntheses were of to the residue and filtration, the CH2Cl2 solution was washed
reagent grade and used without further purification. n-Bu4NB- with water (100 mL × 3). After drying over anhydrous mag-
(C6F5)4 was synthesized according to a previously reported nesium sulfate (MgSO4), the organic phase was concentrated to
procedure.34
dryness. The target product 1 was purified by silica-gel column
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1368–1379 | 1375

View Article Online
Paper
Dalton Transactions
chromatography using toluene as the eluent. Yield: 80 mg acetone-d6, ppm): δ = 4.01 (s, 10H, C5H5), 4.17 (pt, 4H,
(5%). Mp: 106.1–107.3 °C. Anal. Calc. for C18H18FeO2: C, 67.10; C5H4(2b)), 4.54 (pt, 4H, C5H4(2a)), 6.41 (d, J = 7.5 Hz, 2H, C(6)–
H, 5.63. Found: C, 67.12; H, 5.60%. UV-vis (DMF): [λmax/nm] H), 6.71 (dd, J = 7.5, 1.8 Hz 2H, C(5)–H), 6.77 (d, J = 1.8 Hz, 2H,
(εmax), 286 (1.37 × 104), 448 (342). The NMR data were consist- C(3)–H). 11B-NMR (DMSO-d6, ppm): δ = −3.50. 13C{1H}-NMR
ent with those previously reported.19 Suzuki coupling.
A
(acetone-d6, ppm): δ = 66.31 (–CHvCH(Fc)–C(Ph)), 68.30
mixture of ferroceneboronic acid (2.53 g, 11.0 mmol), 4-bromo- (–CHvCH(Fc)–C(Ph)), 69.94 (C5H5), 89.84 [C(Fc)–C(4)], 107.11
veratrole (2.27 g, 9.92 mmol), Pd(PPh3)4 (0.580 g, 0.502 mmol), (C(3)–H), 107.92 (C(6)–H), 115.82 (C(5)–H), 127.95 (C(4)–C5H4),
and K3PO4 (4.62 g, 20.0 mmol) in dry dioxane (100 mL) was 152.48 (C(1)–O), 153.52 (C(1)–O). HR-ESI-MS (m/z). Calc. for
refluxed under N2 for 50 h. The reaction mixture was evapor- C32H24BFe2O4 ([M − K]−): 590.05962. Found: 590.05975. UV-vis
ated to dryness. After the addition of CH2Cl2 (100 mL) to the (DMF): [λmax/nm] (εmax), 310 (1.93 × 104), 443 (901).
residue and filtration, the CH2Cl2 solution was washed with
Synthesis of K[FcCFcBFcCFc] (K4)
water (100 mL × 3). After drying over anhydrous MgSO4, the
organic phase was concentrated to dryness. The target product Compound 3 (0.200 g, 0.395 mmol) was dissolved in dry
1 was purified by silica-gel column chromatography using degassed CH2Cl2 (20 mL) and cooled to −78 °C. A CH2Cl2 solu-
toluene as the eluent. Yield: 0.513 g (15%).
tion of BBr3 (1.0 M, 1.2 mL) was then added to the solution.
After stirring for 30 min at −78 °C and then 30 min at room
temperature, the starting material was completely converted,
Synthesis of 4,5-diferrocenylveratrole, FcVFc (3)
A pentane solution of tert-butyllithium (1.6 M, 10 mL) was as confirmed by TLC. Degassed H2O (20 mL) was then added
added to a mixture of ferrocene (1.87 g, 9.55 mmol) and pot- to the reaction mixture. After stirring vigorously for 30 min,
assium tert-butoxide (0.114 g, 1.02 mmol) in dry THF (30 mL). the organic phase was obtained using a phase separator paper
After stirring for 1 h at −78 °C, a zinc chloride–THF complex and concentrated to dryness. The resulting crude product
(4.55 g, 15.7 mmol) was added to the reaction mixture. After (185 mg), potassium hydroxide (0.130 g, 2.31 mmol), and
stirring for 1 h at −78 °C and then 1 h at room temperature, B(OH)3 (12.0 mg, 0.193 mmol) were dried and dissolved in
Pd(PPh3)4 (67 mg, 5.8 × 10−5 mol) and 4,5-diiodoveratrole degassed acetone (30 mL). After refluxing for 21 h, the reaction
(1.24 g, 3.19 mmol) were added to the reaction mixture. The solution was filtered and the filtrate was concentrated to
reaction mixture was heated at 60 °C for 48 h and evaporated dryness. The target product 3 was recrystallized from acetone–
to dryness. After addition of CH2Cl2 (200 mL) to the residue hexane. Yield: 130 mg (50%). Mp: >250 °C. 1H-NMR (TMS,
and filtration, the CH2Cl2 solution was washed with water acetone-d6, ppm): δ = 3.99 (pt, 2H, C5H4), 4.01–4.04 (7H, C5H4,
(100 mL × 3). After drying over anhydrous MgSO4, the organic C5H5), 6.97 (s, 1H, C6H2). 11B-NMR (DMSO-d6, ppm): δ = −3.28.
phase was concentrated to dryness. The target product 3 was 13C{1H}-NMR (acetone-d6, ppm): δ = 66.96 (C5H5), 69.84, 71.10
recrystallized from CH2Cl2–hexane. Yield: 0.847 g (53%). Mp: (C5H4), 91.49 [C(Fc)–C(4)], 111.00 (C(3)–H), 126.06 (C(4)–C5H4),
1
223.2–223.5 °C. H-NMR (TMS, CDCl3, ppm): δ = 4.00 (s, 3H, 152.13 (C–O). HR-ESI-MS (m/z). Calc. for C52H40BFe4O4 ([M −
OCH3), 4.04 (s, 5H, C5H5), 4.09 (pt, J = 1.9 Hz, 2H, C5H4), 4.11 K]−): 963.04173. Found: 964.04348. UV-vis (DMF): [λmax/nm]
(pt, J = 1.9 Hz, 2H, C5H4), 7.30 (s, 1H, C6H2). 13C{1H}-NMR (εmax), 295 (2.42 × 104), 431 (1.78 × 103).
(CDCl3, ppm): δ = 55.9 (OCH3), 67.2 (C5H4), 69.3 (C5H5), 70.5
Single crystal X-ray diffraction (XRD) analysis
(C5H4), 83.4 [C(Fc)–C(4)], 114.3 (C(3)–H), 129.5 (C(4)–C5H4),
146.7 (C(1)–O). Anal. Calc. for C28H26Fe2O2: C, 66.44; H, 5.18. Single crystals of 1 suitable for XRD analysis were obtained by
Found: C, 66.55; H, 4.88%. UV-vis (DMF): [λmax/nm] (εmax), 289 diffusion of hexane into a chloroform solution containing 1 at
(1.72 × 104), 450 (561).
ambient temperature. Data were collected with a Rigaku
ValiMax RAPID RA-Micro7HFM using Mo Kα radiation at
−150 °C. The diffraction data were processed with RAPID
Synthesis of K[FcCBCFc] (K2)
Compound 1 (0.255 g, 0.791 mmol) was dissolved in dry AUTO on a Rigaku program, and the structures were solved by
degassed CH2Cl2 (20 mL) and cooled to −78 °C. A CH2Cl2 solu- direct methods and refined on F2 by full-matrix least-squares
tion of BBr3 (1.0 M, 2.5 mL) was then added to the solution. using CrystalStructure and SHELXL-97 (Table S1†). CCDC
After stirring for 30 min at −78 °C and then 30 min at room 955166 contains the supplementary crystallographic data for 3
temperature, the starting material was completely converted, in this paper.
as confirmed by TLC. Degassed H2O (20 mL) was then added
to the reaction mixture. After stirring vigorously for 30 min,
DFT calculations
the organic phase was obtained using a phase separator paper The calculations were carried out using the Gaussian09
and concentrated to dryness. The resulting crude product program package.35 The geometries of 1 and 3 were optimized
(233 mg), potassium hydroxide (0.300 g, 5.36 mmol), and using DFT methods without symmetry constraints. The three-
B(OH)3 (24.5 mg, 0.396 mmol) were dried and dissolved in parameterized Becke–Lee–Yang–Parr (B3LYP) hybrid exchange-
degassed acetone (30 mL). After refluxing for 16 h, the reaction correlation functional36 was used with the Lanl2DZ (Hay–Wadt
solution was filtered and the filtrate was concentrated to ECP) basis set37 for the Fe atom and the 6-31G(d) basis set38
dryness. The target product 3 was recrystallized from acetone– for the other atoms, with or without taking account of solvent
hexane. Yield: 130 mg (50%). Mp: >250 °C. 1H-NMR (TMS, effects. The stability of the optimized structure was confirmed
1376 | Dalton Trans., 2014, 43, 1368–1379
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
by calculating the molecular vibrational frequencies, in which
no imaginary frequencies were observed. The solvation effects
of CH2Cl2 were modeled using the PCM approach.39 The
single-point calculations were carried out using the same basis
set as used for the geometry optimizations, without taking
account of the solvent effects. Molecular composition analysis
was conducted using the GaussSum Program. The TD-DFT
method was used to calculate the transition energies and the
oscillator strengths relevant to the absorption spectra, without
taking account of solvent effects.
Chem. Soc., 1986, 108, 5857–5865; (h) S. Nakashima,
T. Oda, T. Okuda and M. Watanabe, Inorg. Chem., 1999, 38,
4005–4010; (i) T. Oda, S. Nakashima and T. Okuda, Inorg.
Chem., 2003, 42, 5376–5383; ( j) D. R. Talham and
D. O. Cowan, Organometallics, 1987, 6, 932–937;
(k) W. Zhang, S. R. Wilson and D. N. Hendrickson, Inorg.
Chem., 1989, 28, 4160–4165; (l) N. J. Long, A. J. Martin,
R. Vilar, A. J. P. White, D. J. Williams and M. Younus,
Organometallics, 1999, 18, 4261–4269; (m) L. Xiao,
W. Weissensteiner, K. Mereiter and M. Widhalm, J. Org.
Chem., 2002, 67, 2206–2214; (n) N. Nagahora, S. Ogawa,
Y. Kawai and R. Sato, Tetrahedron Lett., 2005, 46, 4157–
4160; (o) M. Lohan, P. Ecorchard, T. Rüffer, F. Justaud,
C. Lapinte and H. Lang, Organometallics, 2009, 28, 1878–
1890; (p) G. Espino, L. Xiao, M. Puchberger, K. Mereiter,
F. Spindler, B. R. Manzano, F. A. Jalóne and
W. Weissensteiner, Dalton Trans., 2009, 2751–2763;
(q) M. Lohan, F. Justaud, T. Roisnel, P. Ecorchard, H. Lang
and C. Lapinte, Organometallics, 2010, 29, 4804–4817;
(r) M. Lohan, F. Justaud, H. Lang and C. Lapinte, Organo-
metallics, 2012, 31, 3565–3574; (s) A. Zirakzadeh,
M. A. Groß, Y. Wang, K. Mereiter, F. Spindler and
W. Weissensteiner, Organometallics, 2013, 32, 1075–1084.
6 For selected examples, see: (a) M. D. Rausch, R. F. Kovar
and C. S. Kraihanzel, J. Am. Chem. Soc., 1969, 91, 1259–
1261.F. L. Hedberg and H. Rosenberg, J. Am. Chem. Soc.,
1969, 91, 1258–1259.
Acknowledgements
This work was partially supported by a Grant-in-Aid for Young
Scientists B (no. 25790016) and a research grant from the
Murata Science Foundation. The authors would like to thank
Mr Fumio Asanoma for performing the NMR measurements
(NAIST) and Ms Yoshiko Nishikawa (NAIST) for performing the
ESI-MS measurements.
Notes and references
1 (a) C. Joachim, J. K. Gimzewski and A. Aviram, Nature,
2000, 408, 541–548; (b) H. Nishihara, K. Kanaizuka,
Y. Nishimori and Y. Yamanoi, Coord. Chem. Rev., 2007, 251,
2674–2687.
7 For selected examples of heteroatom bridges, see:
(a) J. B. Heilmann, M. Scheibitz, Y. Qin, A. Sundararaman,
F. Jäkle, T. Kretz, M. Bolte, H.-W. Lerner, M. C. Holthausen
and M. Wagner, Angew. Chem., Int. Ed., 2006, 45, 920–925;
(b) L. Kaufmann, J.-M. Breunig, H. Vitze, F. Schödel,
I. Nowik, M. Pichlmaier, M. Bolte, H.-W. Lerner,
R. F. Winter, R. H. Herberb and M. Wagner, Dalton Trans.,
2009, 2940–2950; (c) M. J. MacLachlan, A. J. Lough,
W. E. Geiger and I. Manners, Organometallics, 1998, 17,
1873–1883; (d) D. E. Herbert, J. B. Gilroy, W. Y. Chan,
L. Chabanne, A. Staubitz, A. J. Lough and I. Manners,
J. Am. Chem. Soc., 2009, 131, 14958–14968; (e) S. Bruña,
A. M. González-Vadillo, D. Nieto, C. Pastor and
I. Cuadrado, Organometallics, 2010, 29, 2796–2807;
(f) S. C. Jones, S. Barlow and D. O’Hare, Chem.–Eur. J.,
2005, 11, 4473–4481; (g) B. Bagh, N. C. Breit, K. Harms,
G. Schatte, I. J. Burgess, H. Braunschweig and J. Müller,
Inorg. Chem., 2012, 51, 11155–11167; (h) R. Rulkens,
D. P. Gates, D. Balaishis, J. K. Pudelski, D. F. McIntosh,
A. J. Lough and I. Manners, J. Am. Chem. Soc., 1997, 119,
10976–10986.
2 (a) I. AmLani, A. O. Orlov, G. Toth, G. H. Bernstein,
C. S. Lent and G. L. Snider, Science, 1999, 284, 289–291;
(b) C. S. Lent, QUANTUM CELLULAR AUTOMATA Theory,
Experimentation and Prospects, ed. M. Macucci, Imperial
College Press, London, 2006, pp. 269–276.
3 (a) C. P. Kubiak, Inorg. Chem., 2013, 52, 5663–5676;
(b) P.-C. Lin, H.-Y. Chen, P.-Y. Chen, M.-H. Chiang,
M. Y. Chiang, T.-S. Kuo and S. C. N. Hsu, Inorg. Chem.,
2011, 50, 10825–10834.
4 For selected reviews, see: (a) D. O. Cowan, C. LeVanda,
J. Park and F. Kaufman, Acc. Chem. Res., 1973, 6, 1–7;
(b) D. Astruc, Acc. Chem. Res., 1997, 30, 383–391;
(c) H. Nishihara, Adv. Inorg. Chem., 2002, 53, 41–86;
(d) H. Nishihara, Bull. Chem. Soc. Jpn., 2001, 74, 19–29.
5 For selected examples, see: (a) D. O. Cowan and
F. Kaufman, J. Am. Chem. Soc., 1970, 92, 6198–6204;
(b) C. LeVanda, K. Bechgaard, D. O. Cowan and
M. D. Rausch, J. Am. Chem. Soc., 1977, 99, 2964–2968;
(c) T. Y. Dong, D. N. Hendrickson, K. Iwai, M. J. Cohn,
S. J. Geib, A. L. Rheingold, H. Sano, I. Motoyama and
S. Nakashima, J. Am. Chem. Soc., 1985, 107, 7996–8008;
(d) R. J. Webb, T. Y. Dong, C. G. Pierpont, S. R. Boone,
R. K. Chadha and D. N. Hendrickson, J. Am. Chem. Soc.,
1991, 113, 4806–4812; (e) S. Nakashima, Y. Ueki, H. Sakai
and Y. Maeda, J. Chem. Soc., Dalton Trans., 1996, 139–143;
(f) T.-Y. Dong, M.-C. Lin, L. Lee, C.-H. Cheng, S.-M. Peng
and G.-H. Lee, J. Organomet. Chem., 2003, 679, 181–193;
(g) T. Y. Dong, T. Kambara and D. N. Hendrickson, J. Am.
8 For selected examples of ethynylene bridges, see:
(a) Y. J. Chen, D.-S. Pan, C.-F. Chiu, J.-X. Su, S. J. Lin and
K. S. Kwan, Inorg. Chem., 2000, 39, 953–958;
(b) P. V. Solntsev, S. V. Dudkin, J. R. Sabin and
V. N. Nemykin, Organometallics, 2011, 30, 3037–3046.
9 For selected examples of arene bridges, see: (a) M. Iyoda,
T. Kondo, T. Okabe, H. Matsuyama, S. Sasaki and
Y. Kuwatani, Chem. Lett., 1997, 35–36; (b) C. Patoux,
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1368–1379 | 1377

View Article Online
Paper
Dalton Transactions
C. Coudret, J.-P. Launay, C. Joachim and A. Gourdon, Inorg.
Chem., 1997, 36, 5037–5049; (c) M. C. P. Wang, Y. Li,
N. Merbouh and H.-Z. Yu, Electrochim. Acta, 2008, 53,
Westerhoff and R. Sanders, J. Am. Chem. Soc., 2002, 124,
7262–7263; (c) F. Barrière and W. E. Geiger, J. Am. Chem.
Soc., 2006, 128, 3980–3989.
7720–7725; (d) R.-J. Xie, L.-M. Han, N. Zhu, H.-L. Hong, 22 K. Diallo, C. Absalon, J. Ruiz and D. Astruc, J. Am. Chem.
Q.-L. Suo and P. Fu, Polyhedron, 2012, 38, 7–14. Soc., 2011, 133, 629–641.
10 For selected examples of heterocycle bridges, see: 23 L. Cuffe, R. D. A. Hudson, J. F. Gallagher, S. Jennings,
(a) A. Hildebrandt, T. Rüffer, E. Erasmus, J. C. Swarts and
H. Lang, Organometallics, 2010, 29, 4900–4905;
(b) A. Hildebrandt, D. Schaarschmidt and H. Lang,
C. J. McAdam, R. B. T. Connelly, A. R. Manning,
B. H. Robinson and J. Simpson, Organometallics, 2005, 24,
2051–2060.
Organometallics, 2011, 30, 556–563; (c) A. Hildebrandt, 24 (a) J. W. Bats, K. Ma and M. Wagner, Acta Crystallogr., Sect.
D. Schaarschmidt, R. Claus and H. Lang, Inorg. Chem.,
2011, 50, 10623–10632; (d) A. Hildebrandt and H. Lang,
Dalton Trans., 2011, 40, 11831–11837; (e) A. Hildebrandt
and H. Lang, Organometallics, 2013, 32, 5640–5653.
C: Cryst. Struct. Commun., 2002, 58, m129–m132;
(b) D. H. Hua, J. W. McGill, P. D. Robinson and
C. Y. Meyers, Acta Crystallogr., Sect. E: Struct. Rep. Online,
2001, 57, m413–m415; (c) M. V. Makarov, V. P. Dyadchenko
and M. Y. Antipin, Russ. Chem. Bull., 2004, 53, 2768–2773;
(d) P. Thilagar, J. Chen, R. A. Lalancette and F. Jäkle, Orga-
nometallics, 2011, 30, 6734–6741; (e) T.-H. Chen,
W. Kaveevivitchai, N. Bui and O. Š. Miljanic, Chem.
Commun., 2012, 48, 2855–2857.
11 For selected examples of porphyrin bridges, see:
(a) C. Bucher, C. H. Devillers, J.-C. Moutet, G. Royal and
E. Saint-Aman, Coord. Chem. Rev., 2009, 253, 21–36;
(b) V. N. Nemykin, C. D. Barrett, R. G. Hadt, R. I. Subbotin,
A. Y. Maximov, E. V. Polshinb and A. Y. Koposov, Dalton
Trans., 2007, 3378–3389.
25 J. K. Day, C. Bresner, I. A. Fallis, L.-L. Ooi, D. J. Watkin,
S. J. Coles, L. Male, M. B. Hursthousec and S. Aldridge,
Dalton Trans., 2007, 3486–3488.
12 For selected examples of other bridges, see: (a) M. Iyoda,
T. Okabe, M. Katada and Y. Kuwatani, J. Organomet. Chem.,
1998, 569, 225–233; (b) M. Kurosawa, T. Nankawa, 26 (a) M. Scheibitz, R. F. Winter, M. Bolte, H.-W. Lerner and
T. Matsuda, K. Kubo, M. Kurihara and H. Nishihara, Inorg.
Chem., 1999, 38, 5113–5123; (c) C. G. Ferguson,
C. Glidewell, G. Opromolla, C. M. Zakaria and P. Zanello,
J. Organomet. Chem., 1996, 517, 183–190; Y. Yamada,
J. Mizutani, M. Kurihara and H. Nishihara, J. Organomet.
Chem., 2001, 637–639, 80–83; (d) S. Tanaka and
K. Mashima, Inorg. Chem., 2011, 50, 11384–11393.
13 D. Siebler, C. Förster, T. Gasi and K. Heinze, Organometal-
lics, 2011, 30, 313–327.
14 (a) U. T. Mueller-Westerhoff, D. I. Yoon and K. Plourde,
Mol. Cryst. Liq. Cryst., 1990, 183, 291–302; (b) U. T. Mueller-
Westerhoff, B. Vance and D. I. Yoon, Tetrahedron, 1991, 47,
909–932.
M. Wagner, Angew. Chem., Int. Ed., 2003, 42, 924–927;
(b) K. Venkatasubbaiah, L. N. Zakharov, W. S. Kassel,
A. L. Rheingold and F. Jäkle, Angew. Chem., Int. Ed., 2005,
44, 5428–5433; (c) K. Venkatasubbaiah, A. Doshi, I. Nowik,
R. H. Herber, A. L. Rheingold and F. Jäkle, Chem.–Eur. J.,
2008, 14, 444–458.
27 P. Thilagar, D. Murillo, J. Chen and F. Jäkle, Dalton Trans.,
2013, 42, 665–670.
28 (a) R. Taube and J.-P. Gehrke, J. Organomet. Chem., 1987,
327, 419–427; (b) X. He and J. F. Hartwig, J. Am. Chem. Soc.,
1996, 118, 1696–1702; (c) S. A. Westcott, H. P. Blom,
T. B. Marder and R. T. Baker, J. Am. Chem. Soc., 1992, 114,
8863–8869; (d) S. A. Westcott, N. J. Taylor, T. B. Marder,
R. T. Baker, N. J. Jones and J. C. Calabrese, J. Chem. Soc.,
Chem. Commun., 1991, 304–305.
15 C. Villena, J. Losada, P. García-Armada, C. M. Casado and
B. Alonso, Organometallics, 2012, 31, 3284–3291.
16 (a) R. Nishiyabu, Y. Kubo, T. D. James and J. S. Fossey, 29 (a) J. Jiao, G. J. Long, F. Grandjean, A. M. Beatty and
Chem. Commun., 2011, 47, 1124–1150; (b) C. M. Vogels
and S. A. Westcott, Chem. Soc. Rev., 2011, 40, 1446–
1458.
T. P. Fehlner, J. Am. Chem. Soc., 2003, 125, 7522–7523;
(b) V. N. Nemykin, G. T. Rohde, C. D. Barrett, R. G. Hadt,
C. Bizzarri, P. Galloni, B. Floris, I. Nowik, R. H. Herber,
A. G. Marrani, R. Zanoni and N. M. Loim, J. Am. Chem.
Soc., 2009, 131, 14969–14978; (c) V. N. Nemykin,
G. T. Rohde, C. D. Barrett, R. G. Hadt, J. R. Sabin, G. Reina,
P. Galloni and B. Floris, Inorg. Chem., 2010, 49, 7497–7509;
(d) G. T. Rohde, J. R. Sabin, C. D. Barrett and
V. N. Nemykin, New J. Chem., 2011, 35, 1440–1448.
17 (a) N. Kameta and K. Hiratani, Tetrahedron Lett., 2006, 47,
4947–4950; (b) N. Kameta and K. Hiratani, Chem. Commun.,
2005, 725–727.
18 (a) K. Miwa, Y. Furusho and E. Yashima, Nat. Chem., 2010,
2, 444–449; (b) H. Katagiri, T. Miyagawa, Y. Furusho and
E. Yashima, Angew. Chem., Int. Ed., 2006, 45, 1741–1744.
19 S. B. Colbran, S. T. Lee, D. G. Lonnon, F. J. D. Maharaj, 30 (a) D. O. Cowan, P. Shu, F. L. Hedberg, M. Rossi and
A. M. McDonagh, K. A. Walker and R. D. Young, Organo-
metallics, 2006, 25, 2216–2224.
20 S. Barlow and S. R. Marder, Chem. Commun., 2000, 1555–
1562.
T. J. Kistenmacher, J. Am. Chem. Soc., 1979, 101, 1304–1306;
(b) D. Siebler, C. Förster, T. Gasi and K. Heinze, Chem.
Commun., 2010, 46, 4490–4492.
31 A. Ikegami, M. Abe, A. Inatomi and Y. Hisaeda, Chem.–Eur.
J., 2010, 16, 4438–4441.
21 (a) N. Camire, U. T. Mueller-Westerhoff and W. E. Geiger,
J.
Organomet.
Chem.,
2001,
637–639,
823–826; 32 (a) R. C. Haddon, S. V. Chichester and J. H. Marshall, Tetra-
(b) F. Barrière, N. Camire, W. E. Geiger, U. T. Mueller-
hedron, 1986, 42, 6293–6300; (b) S. K. Mandal, S. Samanta,
1378 | Dalton Trans., 2014, 43, 1368–1379
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
M. E. Itkis, D. W. Jensen, R. W. Reed, R. T. Oakley,
F. S. Tham, B. Donnadieu and R. C. Haddon, J. Am. Chem.
Soc., 2006, 128, 1982–1994.
K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas,
J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaus-
sian 09, Gaussian, Inc., Wallingford, CT, 2009.
33 H. Braunschweig, C. Hörl, F. Hupp, K. Radacki and
J. Wahler, Organometallics, 2012, 31, 8463–8466.
34 R. J. LeSuer, C. Buttolph and W. E. Geiger, Anal. Chem.,
2004, 76, 6395–6401.
35 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, 36 D. Becke, J. Chem. Phys., 1993, 98, 5648–5652.
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, 37 P. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 270–
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, 38 P. C. Hariharan and J. A. Pople, Theor. Chim. Acta, 1973, 28,
H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, 213–222.
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, 39 J. Tomasi, B. Mennucci and R. Cammi, Chem. Rev., 2005,
273.
K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand,
105, 2999–3093.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1368–1379 | 1379