Deltoid versus Rhomboid: Controlling the Shape of Bis-ferrocene Macrocycles by the Bulkiness of the Substituents

Article abstract of DOI:10.1021/acs.organomet.6b00909

Precise structural control of heteroannularly disubstituted ferrocene (Fc) structures is very challenging as the high rotational mobility of the Fc unit allows a large conformational diversity. Herein we present the syntheses, characterization, and electr

Full text of DOI:10.1021/acs.organomet.6b00909

Article
pubs.acs.org/Organometallics
Deltoid versus Rhomboid: Controlling the Shape of Bis-ferrocene
Macrocycles by the Bulkiness of the Substituents
†
†
†
†
†
Viktor Hoffmann, Loïc le Pleux, Daniel Haussinger, Oliver T. Unke, Alessandro Prescimone,
̈
,†,‡,§
and Marcel Mayor*
†
University of Basel, Department of Chemistry, St. Johanns-Ring 19, CH-4056 Basel, Switzerland
‡
Karlsruhe Institute of Technology (KIT), Institute of Nanotechnology (INT), P.O. Box 3640, D-76021 Karlsruhe, Germany
§Lehn Institute of Functional Materials (LIFM), Sun Yat-Sen University, Guangzhou 510275, China
S Supporting Information
*
ABSTRACT: Precise structural control of heteroannularly
disubstituted ferrocene (Fc) structures is very challenging as
the high rotational mobility of the Fc unit allows a large
conformational diversity. Herein we present the syntheses,
characterization, and electrochemical investigation of two
complementary bis-ferrocene macrocycles, built up via
Sonogashira cross coupling and intramolecular ring-closing
reaction. While the X-ray structure of 1,2-ethynylbenzene
bridged bis-ferrocene complex 1 shows a deltoidal conforma-
tion, a stretched oriented rhomboidal bis-ferrocene metalla-
cycle 2 is formed when the peripheral benzene rings are
decorated with bulky tert-butylsulfanyl groups. VT-NMR
spectroscopy is used to assign the rotation of the embedded Fc units in rhomboid 2. Moreover, cyclic voltammetry (CV) of
deltoid 1 and rhomboid 2 indicate that electronic communication between both ferrocenyl groups can be neglected, while the
electrostatic through space coupling is significant.
INTRODUCTION
Coulomb interactions between both ferrocene subunits, bis-Fc
macrocycles consisting of small and rigid scaffolds moved into
the focus of interest. Due to our experience with macrocycles
■
Over the past few decades, fascinating macrocycles comprising
ferrocenes as functional units have been reported such as a
10−15
1
2
3
consisting of ethynyl interlinked aryl subunits,
we became
molecular scissors, tweezers, and springs. Chiral structures
4
interested in 1,2-diethynylbenzene bridges to combine two Fc
subunits in a compact fashion favoring intramolecular Coulomb
interactions. Such a linker was already reported by Diallo et al.,
where in a series of Fc-terminated compounds, 1,2-bis-
(ferrocenylethynyl)-benzene (3 in Figure 1) exhibited interest-
resembling Escher’s endless staircase or ferrocene (Fc)-
5
terminated helicene are remarkable examples of how the 3D
scaffold of ferrocene is used as structure providing “shaping”
synthon. Moreover, the stable and reversible redox chemistry of
6
7
ferrocene has been exploited in molecular wires, sensors, and
16
8
ing electrostatic interaction characteristics. In their study they
showed that the two ferrocenyl groups turn out of plane in
order to minimize the electrostatic repulsion upon oxidation.
This pioneering study even increased our interest in cyclic
derivatives, as the fixation of the Fc subunits within a
macrocyclic framework should limit their ability to sidestep in
order to avoid electrostatic repulsion. Interestingly, two
macrocyclic structures comprising two ferrocenes heteroan-
nularly interlinked by two ortho-diethynyl benzene linkers are
possible, which are both presented in this work.
switches. Furthermore, substituents on the cyclopentadienyl
(
Cp) rings enable the fine-tuning of the energy states involved.
Owing to its extensive use in numerous areas of applied science,
we became interested in the structural control over the
9
embodied ferrocene units as part of the functional application.
We experienced the precise construction of discrete and shape-
persistent ferrocene macrocycles as challenging, in particular
when ferrocene is integrated as heteroannularly disubstituted
subunit and the Cp−Fe−Cp junction is part of the macrocycle
such that its rotational freedom allows for a large structural
diversity.
Among these considerations, we became interested in
macrocycles comprising two ferrocene units which on the
one side will increase the structural variation problematic but
on the other give access to potentially mixed-valence
compounds with exciting intramolecular redox properties. To
minimize the structural diversity and to maximize the extent of
In particular, we describe the syntheses, full characterization,
and electrochemical investigation of deltoidal shaped bis-
ferrocene 1 with both ortho-diethynyl benzene linkers stacked
on top of each other which we thus entitled “deltoid” and of
Received: December 8, 2016
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.6b00909
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 1. Equilibrium between Two Conformers during the
a
Ring-Closing Reaction
Figure 1. Structures of bis-ferrocene macrocycles deltoid 1 and
rhomboid 2, together with 1,2-bis(ferroceneyl-ethynyl)-benzene 3.
macrocycle 2 with both ortho-diethynyl benzene linkers on
opposed sides of the double Fc junction which we simply call
a
The two conformers are expected to yield each in another well-
defined structure, namely, either the deltoid or the rhomboid. The
bulkiness of the substituent R might serve as trigger to favor one
conformer over the other and consequently also direct the resulting
macrocycle.
“rhomboid”. In the case of rhomboid 2, we further analyze the
dynamic structural behavior by NMR analyses. Both target
structures differ in their spatial arrangement and consequently
also in the symmetry elements involved. As the R substituents
of deltoid 1 are hydrogen atoms, the structure has a mirror
plane comprising both Fe atoms of the ferrocene junctions, and
the well-developed Sonogashira-type coupling chemistry with
ferrocene building blocks.
selected at an early stage of the stepwise assembly strategy, each
macrocycle has to be synthesized from scratch.
16,18,20−22
As the substituent R is
a C -axis going through the middle of the connection between
2
both Fe atoms and a point in the middle between both parallel
benzene units. In the case of rhomboid 2, both R substituents
are bulky tert-butylsulfanyl groups, and the symmetry element is
an inversion point between both Fe-atoms of the ferrocene
junctions.
To scrutinize the hypothesis presented in Scheme 1, we
focused on two model compounds each favoring one of the two
precursor conformers. In order to favor the deltoid
conformation, the steric interaction between both R groups
has to be minimized and we thus decided to get rid of the
substituent (R = H in 1). To steer the precursor equilibrium
toward the rhomboid conformation, the bulky tert-butylsulfanyl
group has been selected (R = St-Bu in 2). The bulkiness of the
substituent was not the only selection criteria; the tert-
butylsulfanyl substituent can also be considered as a masked
thiol anchor group which might become interesting to integrate
the macrocycle in future single-molecule junction experiments.
Cyclization Strategy. After several unsuccessful attempts
to assemble bis-ferrocene macrocycles by dimerizing suitably
functionalized ferrocene derivatives, we decided for a stepwise
assembly of both bridging linkers. While such a synthetic
strategy guarantees good control over the formed connections,
it comes with a considerable increase in the number of
consecutive reaction steps required, even comprising statistical
steps with limited atom economy. In contrast, the subsequent
formation of the second linker can also be considered as
opportunity to control the spatial arrangement of the
macrocycle. As sketched in Scheme 1, the Fc junctions and
the ethynyl linkers can be considered as rotating joints giving
access to both structures, the deltoid and the rhomboid. In the
precursor of the deltoid (Scheme 1, right side on the top), the
two benzene rings are arranged parallel on top of each other
stabilized by stacking interactions, as already reported in the
RESULTS AND DISCUSSION
■
Synthesis. As already mentioned above, the macrocycles
were synthesized in a stepwise assembling strategy consisting of
repetitive reaction sequences consecutively closing one linker
after the other. As a key building block, the properly masked
1,2-diethynylbenzene derivative was required. To optimize both
chemical control and synthetic flexibility, an orthogonal pair of
ethynyl protection groups was desired. The combination
triisopropylsilyl (TIPS) and 2-hydroxy-propyl (HOP) as
17
solid state structures of 1,1′-bis(phenylethynyl)ferrocene and
9
,18
further decorated derivatives thereof.
However, potential
substituents (R in Scheme 1) come in close proximity in the
deltoid precursor, and the steric repulsion probably disfavors
this conformer when large enough substituents are involved.
Instead, the rotational motions around the ferrocene and
ethynyl joints might favor a conformer enabling the ring-closing
reaction on the opposed side, resulting in the desired
rhomboid. The equilibrium between both precursors should
depend on the balance between attractive stacking interactions
and steric repulsion, which should be adjustable by the choice
of the substituent R.
23
ethynyl masking groups reported by Kukula et al. was ideally
suited for the purpose. The synthesis of deltoid 1 is displayed in
Scheme 2. Monoprotected 1,2-diethynylbenzene derivative 5
was prepared by successive Pd-catalyzed coupling of 1-bromo-
2-iodobenzene with 2-methyl-3-butyn-2-ol and (triisopro-
pylsilyl)acetylene profiting from the difference in reactivity of
the two halogen substituents, followed by removal of the
triisopropylsilyl protection group by treatment with a 1 M
9
TBAF solution. We recently reported the high-yield
The assembly of the macrocycles and their precursors is
based on a combination of acetylene scaffolding strategies and
experience we made in C−C coupling reactions between 1,1′-
19
diiodoferrocene (6) and acetylenes with the palladium/
B
DOI: 10.1021/acs.organomet.6b00909
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
a
Scheme 2. Linear Seven-Step Synthesis of the Deltoid Bis-ferrocene Metallacycle 1
a
Reagents and conditions: (a) (i) Pd(PPh ) Cl , CuI, NEt , 2-methyl-3-butyne-2-ol, 50 °C, 3 h, (ii) TIPSA, 90 °C, 12 h; (b) TBAF, THF, 25 °C; (c)
3
2
2
3
Pd(MeCN) Cl , [(t-Bu) PH]BF , CuI, THF/DIPA, 60 °C, (d) TBAOH, toluene, 70 °C 1 h; (e) FcI , Pd(MeCN) Cl , [(t-Bu) PH]BF , CuI, THF/
2
2
3
4
2
2
2
3
4
DIPA, 60 °C, (f) TBAOH, toluene, 70 °C; (g) Pd(MeCN) Cl , [(t-Bu) PH]BF , CuI, THF/DIPA, 1.5 mM in [11], 60 °C.
2
2
3
4
2
4
21
phosphine couple elaborated by Inkpen et al. For the
assembly of Fc-macrocycles reported here, we used the
70 °C for 3 h, and free acetylene 11 was isolated as orange solid
in 73% yield after FCC. With terminal acetylene and
iodoferrocene functionalized bis-ferrocene derivative 11, the
ideal precursor for the intramolecular Sonogashira reaction as
cyclization step was available. To perform the cyclization step, a
1.5 mM solution of 11 in a THF/DIPA mixture was prepared
and treated with the same reagents and conditions as those
described above for the coupling between 5 and 6. After work
up, we isolated as the only reaction product formed in
reasonable yields deltoid bis-ferrocene metallacycle 1 as pale
orange solid in 29% yield. The remaining material probably
forms larger oligomers and polymers which were not isolated.
Target structure 1 displayed very limited solubility features in
all organic solvents investigated, and we were initially afraid of
facing challenging boundary conditions for its characterization.
However, we were pleased to realize that its poor solubility is
reflected in its pronounced tendency to crystallize, and single
crystals suitable for X-ray analysis were obtained by slow
evaporation of a solution of 1 in dichloromethane. X-ray
crystallography unambiguously corroborated the identity of
compound 1 and displayed its conformation (Figure 2). Bis-
2
5
corresponding air-stable trialkylphosphonium salt, since
preliminary studies showed that the advantages in handling
and storage outbalanced the moderate loss in yields. We thus
treated 1,1′-diiodoferrocene 6 with 3 equiv of acetylene 5 in the
presence of catalytic amounts of Pd(MeCN) Cl , [(t-Bu) PH]-
2
2
3
BF , and CuI in a THF/DIPA mixture at 60 °C for 12 h. After
aqueous workup, the reaction products were isolated by flash
column chromatography (FCC) on silica gel. As main product,
,1′-diethynyl-substituted ferrocene derivative 8 was isolated in
7% yield, together with 23% of singly ethynyl-substituted
compound 7. We were particularly interested in doubly HOP
protected diacetylene 8, as considerable differences in the Rf
values of the compounds as a function of the number of polar
protection groups were expected. This enables the chromato-
graphic separation of the differently protected compounds, an
important feature when statistical deprotection reactions are
considered. Indeed, treatment of doubly HOP protected
4
1
4
diacetylene 8 with a n-Bu NOH solution in degassed toluene
4
for 1 h at 70 °C provided monoprotected derivative 9 in very
good 60% isolated yield after FCC. In order to favor the
decoration of the free ethynyl group of 9 with an additional 1-
iodo-1′-yl ferrocene unit, a 5-fold excess of the 1,1′-
diiodoferrocene was used for the coupling between 9 and 6.
Apart from the ratio of the coupled reagents, reaction
conditions similar to those described above for the coupling
between 5 and 6 were applied. After workup, desired open-
chain bis-ferrocene derivative 10 was isolated in 39% yield by
FCC. The major side product we observed was the oxidative
acetylene coupling product providing diethynyl interlinked Fc-
dimer 12 in 12% yield. Despite extensive precautions to exclude
oxygen and air from the reaction mixture, the formation of 12
could not be suppressed and might originate from the
activation cycle forming the catalytically active Pd(0) species.
The remaining HOP protection group in 10 was removed again
ferrocene 1 has C symmetry and resembles the expected
2
deltoidal shaped geometry, with small deviations from the
Figure 2. Solid-state structure of the deltoid 1. ORTEP plots with
ellipsoids plotted at the 50% probability level (hydrogen atoms are
omitted for clarity).
by treatment with a n-Bu NOH solution in degassed toluene at
4
C
DOI: 10.1021/acs.organomet.6b00909
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
a
Scheme 3. Synthesis of Rhomboidal Bis-ferrocene Metallacycle 2 Consisting of 11 Consecutive Steps
a
Reagents and conditions: (a) NH SCN, Oxone, MeOH, 25 °C; (b) CuBr , t-BuONO, MeCN/DCM, 65 °C; (c) (i) LAH, THF, 25 °C (ii) 2-
4
2
chloro-2-methylpropane, AlCl , 25 °C; (d) (i) Pd(PPh ) Cl , CuI, 2-methyl-3-butyn-2-ol, THF/NEt , 25 °C., 5 h, (ii) TIPSA, 80 °C, 16 h; (e)
3
3
2
2
3
TBAF, THF, 25 °C; (f) TBAOH, toluene, 70 °C, 3 h; (g) Pd(MeCN) Cl , [(t-Bu) PH]BF , CuI, THF/DIPA, 60 °C, (h) Pd(dppf)Cl , CuI, 18,
2
2
3
4
2
THF/DIPA, 70 °C; (i) TBAF, THF, 25 °C; (j) 5 equiv of 6, Pd(dppf)Cl , CuI, THF/DIPA, 50 °C, 2 h; (k) TBAOH, toluene, 70 °C; (l) Pd(PPh ) ,
2
3 4
DIPA, toluene 0.1 mM in [24], 90 °C.
geometrical ideal arrangement of an isosceles triangle, with
enclosed angles of 61.01 and 66.13°, respectively. The deviation
is caused by one ferrocene unit, which is inclined with respect
to the aryl plane by ∼7°, while the other ferrocene is almost in
plane. The iron centers are almost coplanar, based on Cp
centroid−Fe1−Fe2 enclosed angle of 89.32°, and separated by
only 7.116 Å. The interplanar phenyl distance is with 3.577 Å
slightly longer than the Cp−Cp distance (3.311 Å), and both
phenyl rings are sitting on top of each other. In addition, the
two Fe atoms and the center between the phenyl rings share a
plane of symmetry due to their parallel arrangement and
fixation in the bis-metallacyclic framework. Deltoid 1 was
arrangement and steer the ring-closing reaction toward the
rhomboidal anti conformation. As already discussed in the
“Cyclization Strategy” section above, we chose tert-butylsulfanyl
groups as R substituents combining the required bulkiness with
promising features as potential anchor groups. It is noteworthy
that our initial approach considering even bulkier triphenylme-
thylsufanyl groups failed, due to severe deactivation of the
triphenylmethylsufanyl decorated aryl system in the subsequent
C−C coupling reactions.
The synthesis of rhomboid 2 is displayed in Scheme 3. The
assembly of the ortho-diethynylbenzene building block started
with the regioselective oxidative thiocyanation of 2-iodoaniline
in para position, using Oxone and ammonium thiocyanate,
which provided 2-iodo-4-thiocyanato-aniline 13 in excellent
1
further characterized by H and HMBC NMR spectra and
MALDI-ToF mass spectrometry. Due to the poor solubility of
1
3
27
1
, we were not able to record meaningful C NMR spectra.
yield. Aniline 13 was readily sublimed in batch scales of up to
The exclusive formation of the syn conformation supports
15 g and was converted to bromo derivate 14, using Cu(II)
28
the first part of our working hypothesis, namely, that in the
absence of steric restriction the deltoid bis-metallacycle is
favored. Bulky substituents (tert-butyl and tert-pentyl) are
known to restrict the revolving motion of the Cp rings in
bromide and tert-butyl nitrite in a 5:1 ratio of acetonitrile/
methylene chloride. Reduction of thiocyanate using LAH to
provide the free thiol which was alkylated with 2-chloro-2-
methylpropane in the presence of catalytic amounts of
aluminum chloride to yield 15 in 76% over both steps. With
bromo-iodo-aryl derivative 15 in hands, the different reactivity
of both halides in Pd-catalyzed Sonogashira coupling allows for
the subsequent coupling with two differently protected
2
6
ferrocene model compounds. The comparable spacing
between both phenyl rings and the Cp rings in the solid-state
structure of 1 suggested that substituents of comparable
dimensions should be bulky enough to handicap the syn
D
DOI: 10.1021/acs.organomet.6b00909
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 3. (a) Equilibrium between the different conformers of rhomboid 2 accessible via rotational motion of a ferrocene subunit within the
macrocycle’s framework. Exclusively the α and β ferrocenyl protons are labeled for visibility. The helical chirality descriptors P and M refer
1
exclusively to the arrangement of the macrocycle and ignore the peripheral t-BuS substituent. (b) Region of the ferrocene signals of the recorded H
NMR spectra of 2 in CD Cl in a temperature range from 298 to 173 K. The signals of the β ferrocene protons were used for LSA and are
2
2
highlighted in light blue.
acetylene derivatives in a single reaction step. Thus, 15 together
DIPA/THF. After keeping the reaction mixture at 70 °C for 4
h, desired doubly substituted Fc derivative 20 was isolated also
in moderate yields of 35% by FCC. Ferrocene 20 is again an
orthogonally HOP and TIPS protected diethynyl, and the TIPS
group was efficiently removed by treatment with TBAF in wet
THF at 25 °C giving monodeprotected diethynylferrocene
moiety 21 almost quantitatively. To favor the formation of
open-chain bis-ferrocene 22, the Sonogashira reaction was
performed with a 5-fold excess of diiodoferrocene 6. The excess
with both catalysts (Pd(PPh ) Cl and CuI were dissolved in a
3
2
2
degassed 3:1 THF/NEt mixture and 2-methyl-3-butyn-2-ol
3
was added at 25 °C. After 3 h, the monitoring of the reaction
mixture by thin layer chromatography (TLC) displayed the
complete consumption of the acetylene and (triisopropylsilyl)-
acetylene was added, and the reaction mixture was heated at 80
°C overnight (16 h). FCC of the concentrated crude reaction
mixture provided differently protected diethynyl 16 in 81%
yield. The orthogonal protection groups enable either the
removal of triisopropylsilane or the hydroxypropyl and thus the
perfect control over the demasking of the ethynyl groups in
para and meta position, respectively. Doubly protected
diethynyl 16 was either treated with fluoride ions in wet
THF to give singly HOP protected diethynyl 17 in good yield,
of 6 and the catalysts (Pd(dppf)Cl and CuI were dissolved in a
2
degassed THF/DIPA (2:1) mixture to which acetylene 21
dissolved in degassed THF was added. After keeping the
reaction mixture for 2 h at 50 °C, the reaction products were
isolated by FCC. Besides desired open-chain bis-ferrocene 22,
which was isolated in 38% yield, again oxidative acetylene
coupling product 23 was obtained in 23% despite of all
attempts to avoid the presence of oxygen. For the final
deprotection, HOP-ethynyl derivative 22 was treated with n-
or it was exposed to n-Bu NOH in degassed toluene at 70 °C
4
for 3 h to provide singly TIPS protected diethynyl 18 in good
yields. Either one of the iodine atoms of 1,1′-diiodoferrocene
had to be substituted consecutively by one of two
monoprotected diethynyls 17 and 18. This is particularly
important to control the symmetry of the targeted rhomboid
bis-ferrocene metallacycle, as the substitution of both iodines of
Bu NOH in degassed toluene at 70 °C for 3 h to provide open-
4
chain precursor 24 with an acetylene at one end and an
iodoferrocene subunit at the other end.
With open-chain precursor 24 in hand, we were striving for
exploring suitable cyclization conditions. To our surprise, all
attempts using typical Sonogashira reaction conditions like all
the variations used during the assembly of the linkers also
failed. Independent of the initial concentration of 24, the
desired intramolecular cyclization product was not detected,
but instead, exclusively intermolecular reaction products were
found by mass-spectrometric analysis of the reaction mixtures,
pointing at oxidative acetylene coupling as main competing
reaction. After numerous unsuccessful attempts, we investigated
copper-free reaction conditions and were delighted to observe
the mass signal pointing at the formation of the macrocycle by
an intramolecular C−C coupling reaction. By far the best result
6
with the same monoprotected diethynyl would yield after ring
closure in a rhomboid metallacycle with C symmetry instead of
the here desired C symmetry. To favor monosubstitution,
HOP-masked diethynyl 17 was treated with a 3-fold excess of
2
i
1
,1′-diiodoferrocene using the already described Sonogashira
coupling conditions, providing monosubstituted iodoferrocene
1
9 in moderate isolated yields of 37%.
In contrast to our prior experience with this catalytic system,
the reaction progress was slower, and we observed notable
quantities of dehalogenated ferrocene derivatives. We thus
altered the catalytic system to Pd(dppf)Cl with copper(I)
2
iodide for the next coupling and added TIPS-masked diethynyl
1
8 in a slight excess to the preheated mixture of 19 dissolved in
E
DOI: 10.1021/acs.organomet.6b00909
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
was achieved with Pd(PPh ) and DIPA in toluene without an
additional copper salt.
used to determine the rate constant of interconversion keTc ≈
3
4
−1
⧧
7
1.0 s with an activation energy of ΔG = 38.5 ± 2.2 kJ
eT
⧧
c
A 0.1 mM solution of 24 in degassed toluene together with
the Pd catalyst and DIPA was kept at 70 °C for 4 h. FCC of the
concentrated crude reaction mixture provided desired rhom-
boid bis-ferrocenemetallacycle 2 as orange solid in 54% isolated
yield. Rhomboid 2 was fully characterized by high-resolution
−1
⧧
−1
−1
mol , ΔH = 44.8 ± 2.0 kJ mol , and ΔS = 34.0 ± 1.6 J K
e
e
−1
mol (see Figure S77). The positive entropy term suggests a
transition state with reduced symmetry (see Figure S10).
It is noteworthy that similar dangling motions of ferrocene
subunits heteroannularly integrated in macrocycles have already
been reported (e.g., the helical ferrocene porphyrinoids
1
13
mass spectrometry and H and C NMR spectra, supple-
mented by advanced NMR experiments (NOESY, HMBC,
HMQC, and VT NMR experiments) to determine the spatial
arrangement of the subunits.
29
̨
́
reported by Stepien et al.).
DFT Calculations. To get a better insight into the energies
involved in the revolving motion of the Fc subunits analyzed by
the VT NMR experiment discussed above, the rotational barrier
was estimated by calculating the energy content of three
extreme conformers (see Figure 4), namely, the starting point
Spectroscopic Evidence of the Rhomboid Macrocycle.
The two peripheral tert-butylsulfanyl groups of bis-ferrocene 2
considerably increased its solubility in polar aprotic solvents
(e.g., acetone, DCM, THF, EtOAc, and DMF) but also reduced
its tendency to crystallize, and so far, all attempts to grow single
crystals suitable for X-ray analysis failed. It thus remained a
spectroscopic challenge to prove its identity, in particular the
rhomboid macrocyclic structure of bis-ferrocene 2. Molecular
model analyses displayed an interesting feature of the
rhomboidal arrangement which might be observed by variable
temperature NMR (VT NMR) experiments. While the
rhomboid macrocyclic arrangement of 2 is shape-persistent
and cannot interconvert to the deltoid arrangement observed
for 1, an additional motion of the macrocyclic framework was
expected for 2.
The freely suspended ferrocene subunits between both
phenyl rings mounted with revolving ethynyl joints should
enable a rotary motion of each Fc subunit within the
macrocyclic framework as sketched in Figure 3a. Thereby, the
rotation of a single ferrocene unit by 180° should result in a
helical arrangement of the macrocycle with a pseudo C2
symmetry. It is noteworthy that the term “pseudo” for the
symmetry description is only correct if exclusively the
macrocyclic structure is considered and the two peripheral
tert-butylsulfanyl substituents are ignored. Subsequent rotation
of the second ferrocene unit renders the initial conformer. Such
a rotation is only possible in the rhomboid anti arrangement
and should allow its differentiation from the more rigid stacked
deltoid arrangement, which does not allow for such motions.
Also, the frequency of the rotation is expected to depend on the
sample’s temperature and might alter the differentiation of the
NMR signals of the involved Fc subunits. We thus recorded the
Figure 4. Relative energy diagram of the calculated conformers of the
macrocycle 2 together with the Gibbs free energy of the transition
state obtained by the VT NMR experiment in red.
of coplanar rhomboid 2 with C symmetry, the “helical”
i
macrocycle with pseudo-C symmetry as end point of a single
2
Fc subunit 180° rotation, and the transition state with a
rotation of the moving Fc subunit of about 90°, which was
considered as proxy for the highest point in the energetic
landscape between both conformers. The geometries of the
coplanar and the helical conformer of rhomboid 2 were
optimized. Calculations were carried out using the Gaussian09
1
H NMR spectra of 2 at 25 °C and successively cooled the
sample to −100 °C (Figure 3b). At 25 °C, we observe two
overlapping high-order multiplets, one for the 4 Cp α-protons
30
suite of codes, on the B3LYP level of theory using a mixed
(
Cp-ring in para-position) and the 4 Cp α′-protons (Cp-ring in
basis set. C, H, and S atoms were treated with 6-31G**,
31,32
meta-position) that resonate at δ = 4.59 ppm and a further
high-order multiplet for 8 Cp β-protons at δ = 4.39 ppm. The
observed spectrum reflects the averaged conformation in fast
exchange at the NMR time scale.
Gradual lowering of the temperature forces the revolving
motion of the ferrocene joints to match the NMR time scale
and hence differentiate the protons from interior and exterior
position, in which all α-protons become anisochronous δ (αint
whereas Fe was treated with the LANL2DZ basis set.
The
transition state with one Fc subunit perpendicular to the
second one was located using the Synchronous Transit Guided
33
Quasi-Newton (STQN) method. As already expected from
the VT NMR experiment, very comparable energy contents
were obtained for the two conformers, with the relaxed helical
−1
ground state conformer being 3.6 kJ mol above the coplanar
⧧
−1
ground state structure. With ΔE = 70.1 kJ mol , the
≠
αext ≠ α′ ≠ α′ ). Likewise, β-protons diverge into two
calculated activation energy of the Fc revolving motion is of
int
ext
⧧
signals with different chemical shifts δ(β = β′ ) ≠ δ(β
=
comparable dimension as the Gibbs free energy value ΔG =
int
int
ext
38.5 kJ mol− obtained by the VT NMR experiment. The lower
value obtained experimentally might be explained by the
solvation of the transition state in the NMR experiment, which
is not considered in the DFT calculation.
1
β′ ). This splitting can be explained by the magnetic
ext
inequivalence of the two Cp rings due to the different
substitution at the unsymmetric phenyl ring and the spatial
variance of protons facing the exterior or interior of the
macrocycle. The coalescence temperature (T = 186.7 ± 0.5 K)
In conclusion, the analysis of the VT NMR spectra and their
comparison with calculated values fully corroborated not only
c
was identified by line shape analysis (LSA) of β-protons and
F
DOI: 10.1021/acs.organomet.6b00909
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
the rhomboidal arrangement of macrocycle 2 but also the
second part of our starting hypothesis, namely, that the
bulkiness of the peripheral substituents steers the conforma-
tional equilibrium and thereby determines the product formed
in the ring closing C−C coupling reaction (see Scheme 1).
Electrochemical Analysis. Both bis-ferrocene metalla-
cycles, deltoid 1 and rhomboid 2, consist of two ferrocene
subunits which are each interlinked with both Cp units by
ortho-diethynylbenzene connections. While the separation of
both Fc subunits is comparable in both structures, the mobility
of each Fc subunit within the framework of the metallacycle
alters considerably, as already discussed in the section before.
As the Fc joints are at the same time the redox-active subunits
of both structures, their electrochemical investigation was of
particular interest.
between two redox centers; hence, low electronic coupling
between those redox centers leads to independent response.
Consequently, if a single redox wave is observed, then low
electronic through-bond communication between the redox
centers can be assumed. This is in accordance with the single
redox wave recorded for fully conjugated 1,2-bis-
(ferrocenylethynyl)-benzene, when n-Bu NPF in CH Cl is
4
6
2
2
16
used.
Barrier
̀
e and Geiger reported that using a perfluorinated
electrolytes such as BARF (tetrakis[3,5-bis(trifluoromethyl)-
phenyl]-borate) weakens ion-pairing effects and favors the
observation of electrostatic through-space communication.
42
Consequently, we recorded CV and SWV voltammograms of
bis-ferrocene 1 and 2 with NaBARF as supporting electrolyte in
CH Cl (Figure 6). In the case of compound 2, a well-resolved
2
2
The redox properties of 1 and 2 were first studied using
cyclic voltammetry (CV) and square wave voltammetry (SWV)
in CH Cl using n-Bu NPF as the supporting electrolyte. The
2
2
4
6
results are displayed in Figure 5 and summarized in Table 1.
Figure 6. CV and SWV (dashed lines) curves of 1 (a) and 2 (b) with
NaBARF as supporting electrolyte in CH Cl vs SCE at 25 °C.
2
2
Figure 5. CV and SWV (dashed lines) curves of 1 (a) and 2 (c) with
n-Bu NPF as supporting electrolyte in DCM at 25 °C. The curve in
two one-electron redox curve was recorded with half-wave
potentials at E1/2 = 0.13 V and E1/2 = 0.41 V vs SCE (Figure
6b). Each redox event exhibits reversible behavior (i /i ≈ 1, i
4
6
panel b further displays the irreversible behavior of 2 when a wide
potential window is exploited.
pa pc
p
1/2
∝
v
, see Figures S5 and S6, respectively), and voltammo-
s
Table 1. Half-Wave Potentials (E1/2) = (Epa − E )/2, SWV
grams of bis-ferreocene 1 displays a broad wave, with a lowered
half-wave potential at E1/2 = 0.53 V vs SCE (Figure 6a). This
seemingly complementary characteristic can be explained with
the structure of the compounds. In the neutral form, the
ferrocenes in 1 are relaxed and have no motivation to obtain
opposing sides.
pc
Potentials in Parentheses for Complexes 1 and 2 in CH Cl
2
2
a
+
n-Bu NPF 0.1 M or NaBARF 0.005 M
4
6
complex
E1/2 n-Bu NPF
E1/2 NaBARF
E1/2 NaBARF
4
6
1
2
0.63 (0.67)
0.63 (0.66)
0.54 (0.56)
0.13 (0.18)
0.41 (0.46)
When the redox centers are oxidized, the Coulomb repulsion
a
−1
III
Glassy carbon WE Ø = 2 mm, ν = 0.1 V s , E vs SCE in CH Cl at
between the two Fe atoms forces them to turn out of plane.
2
2
2
5 °C with the respective electrolyte.
Obviously, the rigid structure of 1 does not facilitate free
rotation along the Cp-sp single bonds. Twisting comes along
with high conformational stress, which slows down the electron
transfer rate and consequently broadens the redox wave. In
contrast, bis-ferrocene 2 was shown to have highly flexible
ferrocene joints that can freely twist, thereby reducing
electrostatic repulsion which permits fast and sequential
oxidation. These results allow to tentatively conclude the
following: (i) The electronic through-bond communication
between the two ferrocenes in 1 and 2 is insignificant, which
consequently provides a framework capable of supportin−g
charge localization, evident by absent wave splitting in a [PF6]
electrolyte media. (ii) The large half-wave potential split in
For compound 1, a single, reversible oxidation wave (i /i ≈ 1,
pa pc
1/2
i ∝ v
see Figure S4) at E = 0.63 V vs SCE is observed
p
s
1/2
(Figure 5a) and formally attributed to the overlapping but
II
III
sequential oxidation of both Fe centers to Fe . In the case of
bis-ferrocene 2, two irreversible oxidation waves were observed
at E1 Pa = 0.63 V and E2 Pa = 1.68 V vs SCE when scanning up
to 1.8 V (Figure 5b).
Such reversibility challenges are well-known from studies on
multiferrocenyl compounds when n-Bu NPF in CH Cl is
4
6
2
2
3
4,35
used.
When scanning up to 1.4 V vs SCE, only a single,
1
/2
reversible (i /i ≈ 1, i ∝ v , see Figure S5) two-electron
oxidation wave was detected (Figure 5c). In literature, single-
step multiple-electron transfers are common for multiferrocene
pa pc
p
s
compound 2 of ΔE1 = 280 mV and the broadened redox wave
/2
in bis-ferrocene 1 suggest considerable through-space commu-
nication when BARF is employed as electrolyte. These
characteristics deliver the essentials for field-coupled applica-
−
36−41
complexes when a [PF ] salt is employed.
Diallo et al.
6
deduced thereof that strong ion pairing as facilitated by
nucleophilic [PF6]− ions, shield the electrostatic interaction
tions in material science.
37,43,44
G
DOI: 10.1021/acs.organomet.6b00909
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(
(
5) Katz, T. J.; Pesti, J. J. Am. Chem. Soc. 1982, 104 (1), 346−347.
6) Lu, Q.; Yao, C.; Wang, X.; Wang, F. J. Phys. Chem. C 2012, 116
CONCLUSION
Two new bis-ferrocene macrocycles are presented, namely,
■
(
33), 17853−17861.
7) Lim, J. Y. C.; Cunningham, M. J.; Davis, J. J.; Beer, P. D. Chem.
Commun. 2015, 51 (78), 14640−14643.
8) Xiao, X.; Brune, D.; He, J.; Lindsay, S.; Gorman, C. B.; Tao, N.
Chem. Phys. 2006, 326 (1), 138−143.
(9) Hoffmann, V.; Jenny, N.; Haussinger, D.; Neuburger, M.; Mayor,
rigid C symmetric deltoid molecule 1 and rhomboidal shaped
2
(
bis-ferrocene 2. Both are assembled by very similar
Sonogashira-type coupling chemistry with tuned steric
interactions in the ring closing step as decider whether the
deltoid or the rhomboid structure is formed. While the identity
of 1 is corroborated by its solid-state structure, only rhomboid
(
̈
M. Eur. J. Org. Chem. 2016, 2016 (12), 2187−2199.
(10) Weibel, N.; Mishchenko, A.; Wandlowski, T.; Neuburger, M.;
Leroux, Y.; Mayor, M. Eur. J. Org. Chem. 2009, 2009 (35), 6140−6150.
2
is able to allow for revolving motions of the ferrocene-
subunits observed by VT NMR studies. Electrochemical
investigations of 1 and 2 display through-space electrostatic
interaction between both ferrocene subunits for both macro-
cycles. We are currently investigating the potential of such
rhomboid bis-ferrocene macrocycles as electrochemically
triggered single molecule mechanical transformers.
(
2
(
11) Shu, L.; Mu
009, 7 (6), 1081.
̈
12) Błaszczyk, A.; Chadim, M.; von Hanisch, C.; Mayor, M. Eur. J.
̈
ri, M.; Krupke, R.; Mayor, M. Org. Biomol. Chem.
Org. Chem. 2006, 2006 (17), 3809−3825.
(
13) Shu, L.; Mayor, M. Chem. Commun. 2006, No. 39, 4134.
14) Mayor, M.; Didschies, C. Angew. Chem., Int. Ed. 2003, 42 (27),
(
3
176−3179.
ASSOCIATED CONTENT
■
(15) Mayor, M.; Lehn, J.-M. J. Am. Chem. Soc. 1999, 121 (48),
*
S
Supporting Information
11231−11232.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.6b00909.
(16) Diallo, A. K.; Absalon, C.; Ruiz, J.; Astruc, D. J. Am. Chem. Soc.
2011, 133 (3), 629−641.
(17) Ingham, S. L.; Khan, M. S.; Lewis, J.; Long, N. J.; Raithby, P. R.
J. Organomet. Chem. 1994, 470 (1−2), 153−159.
18) Inkpen, M. S.; White, A. J. P.; Albrecht, T.; Long, N. J. Dalton
Trans. 2014, 43 (41), 15287−15290.
19) Jenny, N. M.; Mayor, M.; Eaton, T. R. Eur. J. Org. Chem. 2011,
011 (26), 4965−4983.
(20) Baumgardt, I.; Butenscho
(6), 1076−1087.
General details, synthesis and spectroscopic data and
(
1
13
copies of H and C NMR spectra of new compounds.
Experimental details on electrochemical measurements.
Further details on theoretical calculations. Copies of VT
NMR, 2D NMR and HRMS spectra of new compounds.
(
2
̈
n, H. Eur. J. Org. Chem. 2010, 2010
(
PDF)
Cartesian coordinates of computed structure of rhom-
boid coplanar 2 (XYZ)
Cartesian coordinates of computed structure of rhom-
boid helical 2 (XYZ)
Cartesian coordinates of computed transition state
(21) Inkpen, M. S.; White, A. J. P.; Albrecht, T.; Long, N. J. Chem.
Commun. 2013, 49 (50), 5663−5665.
̈
(22) Krauße, N.; Kielmann, M.; Ma, J.; Butenschon, H. Eur. J. Org.
Chem. 2015, 2015 (12), 2622−2631.
23) Kukula, H.; Veit, S.; Godt, A. Eur. J. Org. Chem. 1999, 1999 (1),
77−286.
24) Hundertmark, T.; Littke, A. F.; Buchwald, S. L.; Fu, G. C. Org.
Lett. 2000, 2 (12), 1729−1731.
(
2
(
structure of rhomboid 2 (XYZ)
Crystallographic information file of compound 1 (CIF)
(
25) Netherton, M. R.; Fu, G. C. Org. Lett. 2001, 3 (26), 4295−4298.
AUTHOR INFORMATION
Corresponding Author
E-mail: marcel.mayor@unibas.ch.
(26) Abel, E. W.; Long, N. J.; Orrell, K. G.; Osborne, A. G.; Sik, V. J.
̌
■
Organomet. Chem. 1991, 403 (1−2), 195−208.
(27) Wu, G.; Liu, Q.; Shen, Y.; Wu, W.; Wu, L. Tetrahedron Lett.
*
2
(
(
(
2
(
005, 46 (35), 5831−5834.
28) Doyle, M. P.; Siegfried, B.; Dellaria, J. F. J. Org. Chem. 1977, 42
14), 2426−2431.
29) Stępien, M.; Simkowa, I.; Latos-Grazy
008, 2008 (15), 2601−2611.
30) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
ORCID
Marcel Mayor: 0000-0002-8094-7813
̇
́
nski, L. Eur. J. Org. Chem.
Notes
́
The authors declare no competing financial interest.
Crystal data of compound 1 (CCDC 1515120) are also
deposited with the Cambridge Crystallographic Data Centre
www.ccdc.cam.ac.uk, e-mail: structures@ccdc.cam.ac.uk, fax:
44 (0)1223 336033).
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
revision E.01; Gaussian, Inc.: Wallingford, CT, 2009.
(
+
ACKNOWLEDGMENTS
The authors thank the Swiss National Science Foundation
SNF) (grant number 200020-159730) for continuous and
■
(
generous financial support.
REFERENCES
■
(
1) Muraoka, T.; Kinbara, K.; Kobayashi, Y.; Aida, T. J. Am. Chem.
Soc. 2003, 125 (19), 5612−5613.
(
2) Lima, C. F. R. A. C.; Fernandes, A. M.; Melo, A.; Gonca
M.; Silva, A. M. S.; Santos, L. M. N. B. F. Phys. Chem. Chem. Phys.
015, 17 (37), 23917−23923.
3) Ikeda, T.; Shinkai, S.; Sada, K.; Takeuchi, M. Tetrahedron Lett.
009, 50 (17), 2006−2009.
4) Fukino, T.; Fujita, N.; Aida, T. Org. Lett. 2010, 12 (13), 3074−
077.
̧ lves, L.
2
(
2
(
(31) Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971, 54
(2), 724−728.
(32) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82 (1), 299−310.
(33) Peng, C.; Bernhard Schlegel, H. Isr. J. Chem. 1993, 33 (4), 449−
454.
3
H
DOI: 10.1021/acs.organomet.6b00909
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(
34) Inkpen, M. S.; Scheerer, S.; Linseis, M.; White, A. J. P.; Winter,
R. F.; Albrecht, T.; Long, N. J. Nat. Chem. 2016, 8, 825.
35) Geiger, W. E.; Barriere, F. Acc. Chem. Res. 2010, 43 (7), 1030−
039.
36) Ou-Yang, J.-K.; Chen, L.-J.; Xu, L.; Wang, C.-H.; Yang, H.-B.
Chin. Chem. Lett. 2013, 24 (6), 471−474.
37) Arima, V.; Iurlo, M.; Zoli, L.; Kumar, S.; Piacenza, M.; Della
(
̀
1
(
(
Sala, F.; Matino, F.; Maruccio, G.; Rinaldi, R.; Paolucci, F.; Marcaccio,
M.; Cozzi, P. G.; Bramanti, A. P. Nanoscale 2012, 4 (3), 813−823.
(
38) Das, N.; Arif, A. M.; Stang, P. J.; Sieger, M.; Sarkar, B.; Kaim,
W.; Fiedler, J. Inorg. Chem. 2005, 44 (16), 5798−5804.
39) Fillaut, J.-L.; Astruc, D.; Linares, J. Angew. Chem., Int. Ed. Engl.
995, 33 (23−24), 2460−2462.
40) Chebny, V. J.; Dhar, D.; Lindeman, S. V.; Rathore, R. Org. Lett.
006, 8 (22), 5041−5044.
41) Diallo, A. K.; Daran, J.-C.; Varret, F.; Ruiz, J.; Astruc, D. Angew.
Chem., Int. Ed. 2009, 48 (17), 3141−3145.
42) Barriere, F.; Geiger, W. E. J. Am. Chem. Soc. 2006, 128 (12),
980−3989.
43) Lent, C. S.; Tougaw, P. D.; Porod, W.; Bernstein, G. H.
Nanotechnology 1993, 4 (1), 49.
44) Schneider, B.; Demeshko, S.; Neudeck, S.; Dechert, S.; Meyer,
F. Inorg. Chem. 2013, 52 (22), 13230−13237.
(
1
(
2
(
(
̀
3
(
(
I
DOI: 10.1021/acs.organomet.6b00909
Organometallics XXXX, XXX, XXX−XXX