Expedient Synthesis of Thioether-Functionalized Hydrotris(indazolyl)borate as an Anchoring Platform for Rotary Molecular Machines

Article abstract of DOI:10.1002/ejoc.201800990

Major improvements in the synthesis of surface-mounted rotary molecular machines based on ruthenium(II) complexes are reported. The development of a one-pot indium(III)-mediated “N-deprotection/ester reductive sulfidation” sequence allowed step economy, reproducibility and high efficiency in the synthesis of the thioether-functionalized tripodal ligand. Switching to the thallium salt of hydrotris(indazolyl)borate and to microwave heating further optimized the preparation of the common intermediate in the modular synthesis of symmetric and dissymmetric molecular motors and gears. The penta(4-bromophenyl)cyclopentadienyl ruthenium(II) key precursor is now reproducibly synthesized in 5 steps and 31 % overall yield on the longest linear sequence. Subsequent fivefold Suzuki–Miyaura coupling with ferroceneboronic acid led to a new C₅-symmetric pentaferrocenyl molecular motor.

Full text of DOI:10.1002/ejoc.201800990

A Journal of
Accepted Article
Title: Expedient synthesis of thioether-functionalized
hydrotris(indazolyl)borate as an anchoring platform for rotary
molecular machines
Authors: Guillaume Erbland, Yohan Gisbert, Gwénaël Rapenne, and
Claire Kammerer
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201800990
Link to VoR: http://dx.doi.org/10.1002/ejoc.201800990
Supported by

10.1002/ejoc.201800990
European Journal of Organic Chemistry
Expedient synthesis of thioether-functionalized
hydrotris(indazolyl)borate as an anchoring platform
for rotary molecular machines
Guillaume Erbland,[a] Yohan Gisbert,[a] Gwénaël Rapenne [a,b] and Claire Kammerer*[a]
Abstract: Major improvements in the synthesis of surface-mounted
rotary molecular machines based on ruthenium(II) complexes are
reported. The development of a one-pot indium(III)-mediated “N-
deprotection / ester reductive sulfidation” sequence allowed step
economy, reproducibility and high efficiency in the synthesis of the
thioether-functionalized tripodal ligand. Switching to the thallium salt
of hydrotris(indazolyl)borate and to microwave heating further
optimized the preparation of the common intermediate in the
modular synthesis of symmetric and dissymmetric molecular motors
and gears. The penta(4-bromophenyl)cyclopentadienyl ruthenium(II)
key precursor is now reproducibly synthesized in 5 steps and 31%
overall yield on the longest linear sequence. Subsequent five-fold
Suzuki-Miyaura coupling with ferroceneboronic acid led to a new C5-
symmetric penta-ferrocenyl molecular motor.
a Cu(111) surface as a rotor and used the tip of a Scanning
Tunneling Microscope (STM) to trigger an electrically-driven
rotation.[7] As a result, one of the enantiomeric surface-bound
molecules underwent rotary motion with 5% directionality, which
has been ascribed to the intrinsic chirality of the STM tip. In
2011, Feringa incorporated four overcrowded alkene-based
motors acting as wheels into a molecular car scaffold and
proved that the electrically-driven rotary motion of the motors
can be converted into directional translation of the nanovehicle
on a Cu(111) surface.[8]
In this context, we reported the design and synthesis of
electron-fueled molecular motor 1 displaying a dissymmetrically-
substituted cyclopentadienyl rotor, a scorpionate ligand as the
stator and a ruthenium(II) center as a joint allowing the rotary
motion of the upper part with respect to the lower one (Figure
1).[11] The tripodal stator was functionalized with thioether groups
to ensure a strong anchorage on gold surface via chemisorption.
This molecular motor was indeed studied at the single molecule
scale at low temperature (5 K) under Ultra-High Vacuum (UHV)
by STM.[9] The latter was used not only to image the molecular
motor, but above all to trigger its rotation via the injection of
tunneling electrons. Most importantly, it was discovered that the
Introduction
The field of artificial molecular machines has witnessed a
tremendous development over the last two decades, with an
initial impetus related to the synthesis of mechanically
interlocked systems allowing to control the motion of (at least)
one of their components.[1,2] Among nanomachines, artificial
molecular motors are expected to transform the supplied energy
into a unidirectional translational or rotational motion, ultimately
leading to recovery of the work performed.[3] Pioneering
examples of chemically- and light-fuelled rotary motors were
reported as early as in 1999 by the groups of Kelly[4] and
Feringa,[5] respectively, exhibiting controlled unidirectional
motion in solution. Although many very elegant molecular rotary
motors using chemical or light energy have been developed ever
since[3] and have shown interesting applications,[6] the
exploitation of electronic energy is still rare and only scarce
examples have been reported at the single molecular scale[7-9] or
in self-assembled monolayers.[10] Sykes and co-workers
employed a butyl methyl sulfide single-molecule chemisorbed on
dissymmetric
character
of
the
rotor,
with
four
phenyleneferrocenyl groups and a truncated tolyl arm, is of
paramount importance. On the one hand, it allows a direct
monitoring of the rotation (and of its direction) by following the
position of the shorter arm, which serves as a tag. On the other
hand, it was discovered that this dissymmetry induces a
reversible behaviour, with a direction of rotation which depends
on the location of the tip during the pulse (above one of the
longer ferrocenyl-appended arms, or above the shorter
truncated arm).
This motor belongs to a broader family of motors that have
been reported along the years (Figure 1).[12] This family mostly
involves C5-symmetric rotors, with arms of various lengths and
nature.[11,13] Indeed, while the dissymmetric motor 1 displays a
diameter of about 2 nm with a single para-phenylene separating
the cyclopentadienyl and ferrocenyl units, several motors with
longer arms have been reported. These C5-symmetric rotors
included phenylethynyl linkers[13b] but also insulating groups
such as bisethynyl trans-platinum(II)[13b] or bicyclooctyl[13c]
spacers. Moreover, most of the motors had been originally
designed to be deposited on insulating oxide surfaces and
therefore display ester anchoring groups.
[a]
[b]
CEMES, Université de Toulouse, CNRS, 29, rue Jeanne Marvig,
31055 Toulouse, France
E-mail: kammerer@cemes.fr
http://www.cemes.fr/kammerer
Graduate School of Materials Science, Nara Institute of Science
and Technology, 8916-5 Takayama, Ikoma, Nara, Japan
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800990
European Journal of Organic Chemistry
Subsequent oxidative addition on ruthenium(0) cluster
Ru3(CO)12 leads to complex 4,[14] in accordance with the method
reported by Connelly and Manners.[15] Conversion of this piano-
stool complex into the key intermediate 2 is finally achieved via
ligand exchange in the presence of the potassium salt of
thioether-functionalized
hydrotris(indazolyl)borate
(5.K).
Although this synthetic route suffers from the low yield of the last
step, it allows the formation of the cyclopentadienyl-Ru(II)
complex, which appeared impossible under other tested
conditions due to the strong steric hindrance of the five phenyl
rings attached to the cyclopentadienyl core.
Figure 1. Structure of the unidirectional and reversible tetraferrocenyl
molecular motor
1 and of symmetric Ru(II)-based molecular motors
incorporating ester anchoring groups on the scorpionate ligand, and arms of
various lengths and nature connected to the cyclopentadienyl core.
By contrast, the synthesis of the dissymmetric motor 1
(with R1 = Fc and R2 = Me) involved the introduction of the tolyl
group corresponding to the truncated arm at a very early
stage.[11] In addition to an evident lack of modularity, it was
observed that the oxidative addition of the dissymmetric
cyclopentadienyl bromide on the Ru3(CO)12 cluster was less
efficient (43%) than with the corresponding symmetric precursor
3 (78%). This tendency was confirmed by other attempts with
cyclopentadienyl bromides involving one longer and potentially
coordinating arm.[16] In view of future developments of
dissymmetric motors and gears based on the ruthenium
cyclopentadienyl scaffold, the latter approach is not reliable and
modular enough. It is thus envisioned to exploit penta-bromide
key intermediate 2 and perform a first statistical single cross-
coupling to generate the differentiated arm. The remaining four
activated positions could then undergo cross-couplings reactions
to yield various dissymmetric rotors.
However, the technology is currently not mature enough to
address horizontally an electron-fueled rotary motor located
between two electrodes on an insulating surface (i.e.
a
nanojunction). Our efforts are thus currently focused on the
development of new symmetric and dissymmetric structures of
motors and gears,[12b,c] specifically designed for an optimal
interaction with metallic surfaces such as gold (Scheme 1).
This renewed strategy towards dissymmetric motors and
gears highlights the need for important amounts of key
intermediate 2. It thus appeared necessary not only to increase
the efficiency of the coordination of the tripodal ligand (only 20%,
Scheme 2) but also to optimize the synthesis of potassium
hydrotris[6-((ethylsulfanyl)methyl)indazolyl]borate 5.K by itself.
The latter is a bifunctional platform combining anchoring groups
and coordinating sites pointing in opposite directions. The
anchoring groups are necessary to efficiently restrict the
possible translation, rotation and rocking motions of molecules
on a surface. While a single binding point only prevents
translation motions and a second one additionally blocks the
rotation of the molecule, a minimum of three points of anchoring
are necessary to preclude rocking motion. In our design, the
rigid hydrotris(indazolyl)borate scaffold bears three functional
thioether pendant groups well oriented at the 6-position of the
indazole core to very efficiently immobilize this platform on
metallic surfaces such as gold, silver or copper (Scheme 1). On
the opposite side of the ligand, three nitrogen atoms belonging
to the indazole cores are available to coordinate a metal center
such as ruthenium. This tripodal chelating unit thus allows the
covalent binding of coordination complexes on a surface, as in
Scheme 1. Structure of symmetric (R1 = R2) or dissymmetric (R1 ≠ R2)
molecular motors and gears, and related retrosynthetic analysis leading back
to common key intermediate 2.
From a retrosynthetic point of view, all the structures
including a C5-symmetric rotor stem from key intermediate 2,
displaying aryl bromide functions on each arm, ready for various
cross-coupling reactions. This strategy was successfully applied
to the synthesis of our family of symmetric motors, starting from
pentaphenylcyclopentadiene (Scheme 2).[11] The latter is first
brominated on the phenyl para-positions and on the
cyclopentadiene core using neat bromine giving an
hexabrominated compound (3) in a nearly quantitative yield.
Scheme 2. Synthesis of key intermediate 2 starting from pentaphenylcyclopentadiene.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800990
European Journal of Organic Chemistry
Scheme 3. Synthetic route toward the potassium salt of hydrotris(indazolyl)borate 5.K, starting from 3-amino-4-methylbenzoic acid.
the surface-mounted molecular gears or motors reported
previously.[12] Furthermore, lifting the metallic center away from
the surface allows minimizing interferences caused by metal-
surface interactions. In this context, rotors (in the case of rotary
motors) or cranked wheels (in the case of gears) can be
efficiently deposited, anchored and studied without almost any
interaction with the surface, which is particularly important for
Scanning Probe Microscopy experiments.
presence of a catalytic amount of indium(III) iodide as Lewis acid
and an excess of 1,1,3,3-tetramethyldisiloxane (TMDS) as
reducing agent.[19] After overnight heating, thioethers deriving
from aliphatic and aromatic esters and thiols are obtained in
good to excellent yields.
Herein we present our results towards higher efficiency
and step-economy in the preparation of tripodal ligand 5 and of
the ruthenium penta-bromide derivative 2, which serves as a
central platform in the modular synthesis of various molecular
motors and gears. As an application, the synthesis of a new
symmetric penta-ferrocenyl motor with three thioether pendant
groups (Scheme 1, R1 = R2 = ferrocene) is reported.
Scheme 4. Example of the indium(III)-catalyzed reductive sulfidation of methyl
p-methylbenzoate with 1-octanethiol, as reported by Sakai et al.[18]
This method is thus of high synthetic utility since it avoids the
common sequence involving the reduction of the ester,
subsequent conversion of the alcohol into a (pseudo)halide and
final substitution by a sulfur nucleophile.
Results and Discussion
The efficiency and the scope of this transformation drew our
attention since this would allow the direct conversion of indazolyl
ester 8 to thioether 10 (Scheme 3) in a single step (instead of
three), although Sakai et al. did not mention any test on
heteroaromatic derivatives.
As mentioned above, one of our aims was to optimize the
synthetic route towards hydrotris(indazolyl)borate 5 in terms of
efficiency and reproducibility. Indeed, the preparation of such
scorpionate ligand involved 7 steps starting from 3-amino-4-
methylbenzoic acid (Scheme 3).[17] The latter was almost
quantitatively protected as an ethyl ester and subsequently
converted to the 1H-indazole 8 by a Jacobson reaction followed
by acidic hydrolysis of the intermediate acetamide 7 (64% over
two steps). Functional group manipulation to convert the ethyl
ester into the desired ethyl thioether then involved the reduction
of the ester to the corresponding alcohol 9, bromination and
subsequent nucleophilic substitution with ethanethiol. The
thioether-appended 1H-indazole 10 was finally reacted with
KBH4 under solvent-free conditions at 200 °C to afford
selectively the potassium hydrotris(indazolyl)borate KTp4Bo,6-
In order to prevent the coordination of free 1H-indazole to
the In(III) center, the reaction was first tested with the N-acetyl-
protected indazolyl ester 7. When reproducing the conditions
reported by Sakai et al., i.e. 10 mol% indium(III) iodide in
combination with ethanethiol (1.2 equiv.) and 1,1,3,3-
tetramethyldisiloxane (TMDS, 3 equiv. i.e. 6 Si-H equiv.) in 1,2-
dichloroethane at 60 °C for 20h, a complex mixture was
obtained as observed by 1H NMR spectroscopy of the crude
product (Table 1, entry 1). Careful analysis revealed the
presence of traces amounts of unconverted starting material (7).
Since no further signal corresponding to an acetamide moiety
was detected, it was concluded that the expected N-acetyl
protected indazole carrying the ethyl thioether function had not
been formed. To our delight, it appeared that the free 1H-
indazole scaffold bearing the thioether function was present in
the mixture, although in traces amounts. This indicates that the
desired reductive sulfidation takes place, although with a
concomitant deprotection of the N-acetyl group which most
probably consumes both the thiol and the silane. Indeed,
increasing the quantity of thiol and silane to 2.2 and 6
equivalents, respectively, the free 1H-indazole 10 was
gratifyingly formed as major product and isolated in 57% yield
(entry 2). To check if the possible coordination of the free 1H-
indazolyl ester 8 or thioether 10 formed during the reaction might
deactivate the indium(III) catalyst and thus hamper the reaction,
indium(III) iodide was introduced in slight excess (1.1 equivalent)
compared to the substrate (entry 3). Under otherwise
unchanged conditions, only a minor improvement of the isolated
CH2SEt
(5.K) which was purified by sublimation of unreacted
indazole 10.
This synthetic sequence has been intensely used in our
group as
a
standard way to prepare indazole-derived
scorpionate ligands carrying either ester (from a direct reaction
between KBH4 and 8) or thioether anchoring groups. Now that
our main interest lies in molecular machines to be studied on
metallic surfaces, it was desirable to renew the synthetic route
towards thioether-appended indazole 10, especially as a lack of
efficiency and reproducibility of the ethanethiol-mediated
nucleophilic substitution was observed throughout the years.
Improved synthetic route to thioether-appended indazoles
Interestingly, Sakai et al. recently reported the direct conversion
of esters into sulfides via an indium-catalyzed reductive
sulfidation reaction (Scheme 4).[18] The ester is directly reacted
with the appropriate thiol in 1,2-dichloroethane (1,2-DCE) in the
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800990
European Journal of Organic Chemistry
yield was obtained (65%). In this reaction, a by-product (13) was
formed in a ratio of 1:4 compared to product 10 according to the
1H NMR spectrum of the crude mixture. This by-product had
already been observed in various amounts under catalytic
conditions and corresponds to the N-ethyl derivative of 10. The
N-ethyl moiety in 13 most probably results from a direct
reduction of the acetamide group, as already reported under
similar conditions.[20] To overcome this unproductive side
reaction, the quantity of nucleophile was increased to 4
equivalents and the desired free 1H-indazole 10 was obtained
as a single product in 86% yield (entry 4). Finally, in view of the
synthesis of scorpionate ligands on a large scale, the same
reaction was run on 1.5 g (compared to 150 mg in entries 1-4) to
give rise to thioether indazole 10 in a comparable 85% isolated
yield (entry 5).
functionalized indazole (10) is now very conveniently prepared in
three steps on the gram scale starting from 3-amino-4-
methylbenzoic acid in a reproducible 70% overall yield (Scheme
5), instead of six steps and a non-reproducible overall yield of 12
to 36% previously.
Following this improvement, the reproducibility of the
synthesis of the hydrotris(indazolyl)borate salt 5.K and the
efficiency of the subsequent ligand exchange to yield key
intermediate 2 were tackled.
Beneficial use of the thallium salt of the tripodal ligand as
intermediate in the synthesis of rotor 2
Hydrotris(pyrazolyl)borates and by extension hydrotris(indazolyl)
borates, are mostly prepared by heating an excess of the
appropriate pyrazole (respectively indazole) with KBH4 or NaBH4
under neat conditions, as described above for the synthesis of
the potassium salt 5.K.[21] However, the thallium(I) salts of such
derivatives have appeared as mild and valuable reagents for the
transfer of the scorpionate ligands onto various transition metals,
as they exhibit a lower reducing ability than the corresponding
alkali metal salts.[22] The crystallinity of the thallium(I) complexes
also renders their isolation more easy, and we thus envisioned
the synthesis of the thallium salt TlTp4Bo,6-CH2SEt 5.Tl. The
thallium salts of scorpionate ligands are usually obtained by
cation exchange, reacting the alkali metal salt with TlNO3. In
2008, two new methods were reported by Kitamura et al.: i)
preparation and isolation of thallium borohydride TlBH4 and
subsequent reaction with pyrazole or indazole derivatives[23] or ii)
a solvent-free one-pot method using KBH4 and Tl2SO4 in a 2:1
Table 1. Optimization of the reaction conditions for the direct conversion of
indazolyl ester 7 to thioether 10,[a] in one single step instead of four.
Entry
InI3
EtSH
(Me2SiH)2O
(equiv.)
Yield (%)[b]
(equiv.)
(equiv.)
1
0.1
0.1
1.1
1.1
1.1
1.2
2.2
2.2
4
3
6
6
6
6
Traces
57[c]
65[c]
86
2
ratio
to
yield
the
thallium
hydrotris(pyrazolyl)-
or
hydrotris(indazolyl)borate via an in-situ cation exchange.[24]
Given the toxicity of thallium(I), the latter one-pot procedure was
selected to avoid the isolation of the thallium borohydride
reagent. Indazole 10 was thus reacted in the presence of
potassium borohydride and thallium sulfate to give rise to
thallium hydrotris(indazolyl)borate 5.Tl in 54% yield (Scheme 5).
Generating the thallium salt instead of the potassium salt leads
to a comparable yield (54% vs 55%, respectively) but avoids the
tedious and non-reproducible sublimation procedure as the
thallium salt is efficiently recrystallized. However, the main
advantage of using the thallium intermediate 5.Tl lies in the
ligand exchange step. Indeed, the reaction of 2 equivalents of
thallium hydrotris(indazolyl)borate 5.Tl with 1 equivalent of the
ruthenium cyclopentadienyl complex 4 in THF at 100 °C for 48h
afforded the target compound 2 in 67% yield, compared to 20%
with the potassium salt 5.K under otherwise similar conditions
(Scheme 2). To our delight, the efficiency of this reaction was
even further increased to 82% replacing classical heating with
3
4
5[d]
4
85
[a] The reactions were run on
a 150 mg (0.65 mmol) scale unless
otherwise stated. [b] Isolated yield unless otherwise stated. [c] By-product
13 resulting from the direct reduction of the N-acetyl group into an N-ethyl
moiety was detected in the crude mixture. [d] The reaction was run on a
1.5 g (6.5 mmol) scale.
Application of the reaction conditions for the reductive
sulfidation of esters developed by Sakai et al. and subsequent
optimization now allow for a direct conversion of N-acetyl
indazolyl ester 7 into the corresponding N-deprotected indazolyl
thioether 10 in high yield and in a single step. This one-pot
deprotection
/ reductive sulfidation not only shortens the
synthetic route towards tripodal ligand 5.K but also avoids the
tedious isolation of 1H-indazolyl ester 8. The thioether-
Scheme 5. Optimized synthetic route yielding the key penta-bromide rotor 2 in 5 steps and 31% overall yield in the longest linear sequence (instead of 8 steps and
maximum 4% yield previously).
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800990
European Journal of Organic Chemistry
microwave heating (100 °C, 3 x 10 min) in acetonitrile as solvent
(Scheme 5).
order to assess the role of each reagent. As expected, omission
of the Lewis acidic indium(III) species totally prevented the
reaction (entry 6). In the absence of thiol, the major product was
N-ethylindazolyl ester (14), resulting from the complete reduction
of the carbonyl moiety to the corresponding methylene (entry
7).[20] Finally, the reducing agent is also crucial in this procedure
as no reaction occurs in the absence of hydrosilane (entry 8). It
is thus assumed that this deprotection reaction is formally a
reductive sulfidation of the acetamide moiety, with the indazole
core playing the role of the leaving group.
The indium(III)-mediated one pot deprotection / reductive
sulfidation sequence followed by the preparation of the thallium
salt of the scorpionate ligand has significantly shortened the
synthetic route to the penta-bromide rotor 2 and increased its
efficiency. Indeed, this key building block towards molecular
motors and gears is now prepared in 5 steps in 31% overall yield
in the longest linear sequence, compared to 8 steps and a
maximum 4% overall yield formerly.
Extension of the indium(III)-mediated methodology to the
synthesis of ester-appended scorpionate ligands
Table 2. Screening of conditions for the N-deprotection of N-acetylindazole 7
and corresponding blank experiments.[a]
During our work towards the reductive sulfidation of indazolyl
ester 7, it was observed that the deprotection of the N-
acetylindazole to give the corresponding 1H-indazole 8 occurs
prior to the effective transformation of the ester. Deprotection of
the N-acetyl protecting group is most commonly performed
under drastic basic (hydroxide, alkoxide) or acidic (HCl)
conditions,[25,26] provided that these conditions are compatible
with additional functional groups. For instance, in our synthetic
route towards hydrotris(indazolyl)borates, the crude N-
acetylindazole obtained via the Jacobson reaction was
deprotected using concentrated HCl at 60 °C (Scheme 3). The
1H-indazole 8 was obtained in a moderate 64% yield over two
steps, and it was observed that a significant amount of product
was lost during the acidic deprotection step, possibly via
hydrolysis of the ester.
Entry
1
InX3
t (h)
16
4
Conv. (%)[b]
Yield (%)[c]
InI3
100
0
28[d]
-
2
In(OAc)3
In(OTf)3
InCl3
InCl3
-
3
4
5
nd[e]
nd
96
-
4
20
48
48
48
48
91
100
0
5
Since the amide function appears more reactive than the
ester towards In(III)/TMDS/EtSH reaction conditions, the
possible selective deprotection of the indazole core under mild
conditions was investigated.
6
7[f]
8[h]
InCl3
InCl3
96[g]
5
In a first test, the slight excess of indium(III) iodide (1.1
equiv.) was maintained while the amounts of ethanethiol and
TMDS were reduced to 1.2 and 3 equiv. respectively, in order to
favor the single reaction of the amide function (Table 2, entry 1).
After 16h at room temperature, the starting material was fully
converted but a mixture of products was obtained, including 10
0
-
[a] The reactions were run on a 150 mg scale (entries 1-5) or on a 75 mg
scale (entries 6-8). [b] Conversion of the starting material 7 as determined
by 1H NMR spectroscopy. [c] Isolated yield. [d] The desired product 8 was
formed along with reductive sulfidation product 10 (13% isolated yield) and
other unidentified products, as observed by 1H NMR spectroscopy of the
crude product. [e] Not determined. [f] Ethanethiol was omitted. [g] N-
ethylindazolyl ester 14 and the desired product 8 were obtained in the
crude mixture, in a 2:1 ratio. [h] 1,1,3,3-Tetramethyldisiloxane was omitted.
resulting from the N-deprotection
/
reductive sulfidation
sequence. The desired deprotected indazolyl ester 8 was
isolated in 28% yield. Less reactive indium(III) species were
tested next in order to avoid the conversion of the ester moiety.
In the presence of indium(III) acetate and triflate, the conversion
was very low after 4h at room temperature (entries 2 and 3).
Conversely, indium(III) chloride promoted a smooth reaction with
91% conversion of the starting material after 20h (entry 4).
Increasing the reaction time to 48h led to full conversion and the
desired deprotected 1H-indazole 8 was isolated in a satisfactory
96% yield (entry 5). Blank experiments were next carried out in
With these new deprotection conditions in hand, indazolyl
ester 8 is now efficiently synthesized in 79% overall yield starting
from 3-amino-4-methylbenzoic acid (Scheme 6), as compared to
the 63% obtained previously (Scheme 3). The preparation of
ester-appended hydrotris(indazolyl)borate 11.K, used as stator
in studies on insulating surfaces, will benefit from this
improvement implying an increase from 43% to 54% of the
overall yield on this four-step sequence.
Scheme 6. Optimized synthetic route towards the potassium salt of ester-functionalized hydrotris(indazolyl)borate 11.K.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800990
European Journal of Organic Chemistry
Conclusions
Test-reaction: Synthesis of a new molecular motor and last
major improvement using microwave activation
In this work, high efficiency and step-economy was achieved in
the preparation of thioether-functionalized hydrotris(indazolyl)
borates to be used as anchoring platforms in surface-mounted
rotary molecular machines, but also as a general platform to
anchor various metallic centers on gold, silver or copper
surfaces. Indeed, the development of a one-pot “N-deprotection/
ester reductive sulfidation” sequence allowed the direct
conversion of the N-acetylindazolyl ester precursor into the
corresponding 1H-indazolyl thioether in a single high-yielding
step mediated by indium(III) iodide. The thioether-functionalized
indazole is now very conveniently prepared in three steps on the
gram scale starting from 3-amino-4-methylbenzoic acid in a
reproducible 70% overall yield, instead of six steps and a non-
reproducible overall yield of 12 to 36% previously. Furthermore,
milder conditions involving indium(III) chloride and the same
TMDS/EtSH system were identified for the selective cleavage of
the N-acetyl protecting group leading to the free 1H-indazolyl
ester in excellent yield, which is of interest for the synthesis of
ester-appended scorpionate ligands.
So as to probe the efficiency of a complete synthetic sequence
up to molecular machines, the preparation of the new molecular
motor 12 incorporating a symmetric rotating subunit was tackled
(Scheme 7).
The penta-bromide rotor 2, displaying a thioether-functionalized
tripodal ligand, was submitted to Suzuki-Miyaura coupling
conditions in the presence of a large excess of ferroceneboronic
acid, using palladium diacetate / SPhos as catalytic system,
potassium phosphate as base in anhydrous toluene. Penta-aryl
bromide 2 thus underwent five successive Suzuki-Miyaura
couplings to give rise to the desired penta-ferrocenyl molecular
motor 12 in 17% yield after 48h at 100 °C (Scheme 7, classical
heating).
Considering the future preparation of motors and gears that may
involve more complex and expensive boronic acid derivatives,
decreasing the amount of such partner was highly desirable.
New reaction conditions were thus developed, involving 7.5
equiv. of ferroceneboronic acid (i.e. 1.5 equiv. per aryl bromide),
sodium tert-butylate (22.5 equiv.) as base and [1,1’-
bis(diphenylphosphino)ferrocene]dichloropalladium PdCl2(dppf)
(20 mol% per aryl bromide) as catalyst in anhydrous toluene.
Microwave heating at 135 °C for 1 h gave rise to the desired
penta-ferrocenyl product in a highly satisfactory 45% yield
(Scheme 7, microwave heating). This step leads to the formation
of five new C-C bonds in a single synthetic operation, with a
yield of 85% per Suzuki-Miyaura cross-coupling in spite of the
steric hindrance. These new coupling conditions now allow an
expedient synthesis of this symmetric molecular motor,
displaying short ferrocenylphenylene arms, in 6 steps in the
longest linear sequence with an overall yield of 14%.
In view of the synthesis of families of ruthenium(II) complexes
based on a thioether-functionalized stator and a penta-aryl
cyclopentadienyl rotor, the preparation of the penta(4-
bromophenyl)cyclopentadienyl common intermediate was
improved to reach 31% overall yield in the longest linear
sequence (5 steps), instead of a maximum of 4% (over 8 steps)
previously. A five-fold Suzuki-Miyaura cross-coupling of this key
precursor gave rise to a new C5-symmetric molecular motor with
five ferrocenylphenylene arms, proving that molecular machines
can now be obtained efficiently in 6 steps via our renewed
synthetic route. The modular synthesis of various symmetric and
dissymmetric molecular gears, exploiting the same synthetic
strategy, is now underway.
Experimental Section
General methods: All commercially available chemicals were of reagent
grade and were used without further purification. Isoamyl nitrite,
anhydrous 1,2-dichloroethane, anhydrous THF, anhydrous acetonitrile,
anhydrous toluene, ethanethiol, potassium phosphate and sodium tert-
butoxide were purchased from Aldrich. 3-Amino-4-methylbenzoic acid,
potassium
borohydride,
thallium(I)
sulfate
and
1,1’-
bis(diphenylphosphino)ferrocenepalladium(II)
dichloride
(dichloro-
methane adduct) Pd(dppf)Cl2 CH2Cl2 were purchased from Acros.
Trichloroindigane, triiodoindigane and ferroceneboronic acid were
purchased from Alfa Aesar. 1,1,3,3-Tetramethyldisiloxane and
palladium(II) acetate were purchased from ABCR. Potassium acetate
was purchased from Lancaster, acetic anhydride from Fluka and 2-
dicyclohexylphosphino-2’-6’-dimethoxybiphenyl (SPhos) from TCI.
Compounds 3,[14] 4,[14] 6,[17] and 11.K[17] were prepared according to the
corresponding published procedures. All reactions were carried out using
standard Schlenk techniques under an argon atmosphere. Thin layer
chromatography (TLC) was performed on pre-coated aluminum-backed
silica gel 60 UV254 plates (Macherey-Nagel) with visualization effected
Scheme 7. Synthesis of the new molecular motor 12 via a five-fold Suzuki-
Miyaura cross-coupling under classical or microwave heating.
The new molecular motor 12 was fully characterized and
exhibits free rotation at room temperature as shown by NMR
spectroscopy. Its properties will be investigated by Scanning
Probe Microscopy techniques in the near future.
using ultraviolet irradiation (λ
= 254, 366 nm). Flash column
chromatography was carried out on 230-400 mesh silica gel (Aldrich)
unless otherwise stated. Microwave reactions were carried out using
CEM Discover LabMate. NMR, IR and mass spectra were recorded by
the appropriate services of the Toulouse Institute of Chemistry (ICT -
FR2599). NMR spectra were recorded with a Bruker Avance 300 or
Avance 500 spectrometer and full assignments were made with the
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800990
European Journal of Organic Chemistry
assistance of COSY, HMBC and HSQC spectra when necessary. 1H and
13C NMR chemical shifts (δ) are reported in ppm relative to the signal of
tetramethylsilane (TMS). Residual solvent signals were used as an
internal reference. Coupling constants (J) are given in Hz and the
following abbreviations have been used to describe the signals: singlet
(s); doublet (d); triplet (t); quadruplet (q); quintuplet (quint); multiplet (m).
The numbering system used for the assignment of signals in new
compounds 5.Tl, 7 and 12 is provided in the supporting information
document, along with the corresponding spectra. IR spectra were
recorded with a Nicolet 6700 FTIR-ATR. Only selected characteristic
peaks are recorded. High-resolution mass spectra (HRMS) were
performed with a Waters GCT Premier spectrometer for desorption
Ethyl 1-acetyl-1H-indazole-6-carboxylate (7): A dry three-necked
round bottom flask was successively charged with ethyl 3-amino-4-
methylbenzoate 6 (2.0 g, 11.2 mmol, 1.0 equiv.), anhydrous toluene (50
mL), potassium acetate (1.1 g, 12.3 mmol, 1.1 equiv.) and acetic
anhydride (3.8 mL, 40.5 mmol, 3.6 equiv.). Isoamyl nitrite (3.0 mL, 22.3
mmol, 2.0 equiv.) was then added dropwise over 15 min. The resulting
gelatinous mixture was stirred and heated at reflux for 16h with a 15%
NaOH trap. The completion of the reaction was monitored by TLC. The
solution was then evaporated to dryness to give an orange-brownish
solid. The crude product was dissolved in 50 mL of CH2Cl2, filtered on a
pad of silica gel and concentrated under reduced pressure. The residue
was purified by column chromatography (CH2Cl2/pentane gradient from
30:70 up to 70:30) to give compound 7 (2.1 g, 9.2 mmol) as an orange
solid in 82% yield. Rf=0.6 (ethyl acetate/cyclohexane 30:70); m.p. 95 °C;
1H NMR (300 MHz, CDCl3, 25°C): δ=9.07 (m, 1H, Hd), 8.15 (d, 4J = 0.9
Hz, 1H, Ha), 8.02 (dd, 3J = 8.3 Hz, 4J = 1.4 Hz, 1H, Hc), 7.75 (dd, 3J = 8.3
Hz, 4J = 0.8 Hz, 1H, Hb), 4.43 (q, 3J = 7.1 Hz, 2H, Hf), 2.80 (s, 3H, He),
1.43 (t, 3J = 7.1 Hz, 3H, Hg); 13C NMR (75 MHz, CDCl3, 25°C): δ=171.0
(C8), 166.3 (C10), 139.4 (C1), 138.8 (C2), 131.5 (C5), 129.0 (C7), 125.5
(C3), 120.7 (C4), 117.2 (C6), 61.6 (C11), 23.1 (C9), 14.5 (C12); IR (ATR):
ṽ=1715 cm−1 (C=O); HRMS (DCI/CH4): calcd for C12H13N2O3
[MH]+:233.0926, found: 233.0935.
chemical ionization (DCI/CH4) and with
a Waters Xevo G2 QTof
spectrometer for electrospray ionization (ESI). Melting points were
measured with a Krüss M5000 melting-point apparatus or with a Kofler
hot bench and are uncorrected. UV/Vis spectra were recorded with a
Shimadzu UV-26000 spectrometer (sh=shoulder, ε [mol−1dm3cm−1] is
reported in parentheses).
Compound 2: In a dry tube designed for microwave irradiation and
under argon, compound 5.Tl (160 mg, 0.20 mmol, 2.0 equiv.),
ruthenium(II) complex 4 (110 mg, 0.10 mmol, 1.0 equiv.) were introduced
and degassed anhydrous acetonitrile (4 mL) was added. The tube was
sealed and the reaction mixture was heated under microwave irradiation
at 100 °C for 3 x 10 min. A pressure of 5 bar was achieved due to the CO
evolution and this pressure was released between heating cycles. The
completion of the reaction was monitored by TLC. The resulting mixture
was diluted with CH2Cl2 and filtered through a pad of silica gel. The
solvents were removed under reduced pressure and the residue was
purified by column chromatography (CH2Cl2/cyclohexane gradient from
0:100 up to 30:70) to afford compound 2 (128 mg, 0.08 mmol, 82%) as
an orange solid. Rf=0.3 (CH2Cl2/cyclohexane 30:70); 1H NMR (300 MHz,
CD2Cl2, 25°C): δ=7.88 (bs, 3H), 7.80 (d, 4J = 0.5 Hz, 3H), 7.35 (dd, 3J =
8.4 Hz, 4J = 0.5 Hz, 3H), 7.21 (m, 20H), 7.04 (dd, 3J = 8.4 Hz, 4J = 1.4 Hz,
3H), 3.90 (s, 6H), 2.46 (q, 3J = 7.4 Hz, 6H), 1.27 (t, 3J = 7.4 Hz, 9H);
13C NMR (125 MHz, CD2Cl2, 25°C): δ=143.6, 140.2, 137. 8, 135.1, 132.1,
130.6, 122.3, 122.0, 121.9, 120.0, 110.9, 87.1, 37.4, 25.3, 14.3; UV/Vis
(CH2Cl2): λmax (ε)= 298 (26800), 312 nm (25600 mol−1dm3cm−1); HRMS
(ESI+): calcd for C65H54BBr5N6RuS3 [MH]+:1528.8674, found: 1528.8652.
The data match those reported in the literature.[11]
Ethyl 1H-indazole-6-carboxylate (8): In a dry Schlenk tube under argon
were successively added ethyl 1-acetyl-1H-indazole-6-carboxylate 7 (150
mg, 0.65 mmol, 1.0 equiv.), trichloroindigane InCl3 (157 mg, 0.71 mmol,
1.1 equiv.) and anhydrous 1,2-dichloroethane (2 mL). Ethanethiol (60 µL,
0.78 mmol, 1.2 equiv.) and 1,1,3,3-tetramethyldisiloxane (TMDS) (353 µL,
1.94 mmol, 3.0 equiv.) were then successively added and the reaction
mixture was stirred at room temperature for 16h. The completion of the
reaction was monitored by TLC and the reaction medium was then
evaporated to dryness. The crude product was dissolved in CH2Cl2 (10
mL), transferred to a separatory funnel and water (3 mL) was added. The
layers were separated and the aqueous phase was then extracted with
CH2Cl2 (3×5 mL). The combined organic layers were dried over
anhydrous MgSO4, filtered, and concentrated under reduced pressure.
The crude material was purified by column chromatography (ethyl
acetate/cyclohexane 1:1). to afford compound 8 (118 mg, 0.62 mmol) as
a white solid in 96% yield. Rf=0.32 (ethyl acetate/cyclohexane 30:70);
m.p. 125°C; 1H NMR (300 MHz, CDCl3, 25°C): δ=10.25 (br s, 1H), 8.28
4
(m, 1H), 8.15 (d, J = 0.8 Hz, 1H), 7.88 (dd, 3J = 8.5 Hz, 4J = 1.3 Hz, AB
system, 1H), 7.81 (dd, 3J = 8.5 Hz, 4J = 0.8 Hz, AB system, 1H), 4.44 (q,
3J = 7.1 Hz, 2H), 1.44 (t, 3J = 7.1 Hz, 3H).; 13C NMR (63 MHz, CDCl3,
25°C): δ=166.8, 139.5, 135.0, 129.0, 125.7, 121.6, 120.7, 112.1, 61.3,
14.4; IR (ATR): ṽ=3317 (N-H), 1688 cm−1 (C=O). The data match those
reported in the literature.[17]
Thallium
hydrotris{6-[(ethylsulfanyl)methyl]indazol-1-yl}
borate
(5.Tl): 6-[(Ethylsulfanyl)methyl]-1H-indazole 10 (530 mg, 2.76 mmol, 3.0
equiv.), KBH4 (61 mg, 1.11 mmol, 1.2 equiv.) and Tl2SO4 (285 mg, 0.565
mmol, 0.6 equiv.) were successively placed in a dry Young-type Schlenk
tube. The mixture was stirred at 140 °C for 1h under an argon stream in
an open system, and the Schlenk tube was then sealed and heated for
2h at 180 °C. The mixture was allowed to cool to room temperature. The
Schlenk tube was connected to an argon line, and the internal pressure,
which had been raised by the evolution of hydrogen gas, was carefully
released. Subsequent heating of the closed system during 3h at 180 °C
6-[(Ethylsulfanyl)methyl]-1H-indazole (10): In
a dry Schlenk tube
under argon were successively added ethyl 1-acetyl-1H-indazole-6-
carboxylate 7 (150 mg, 0.65 mmol, 1.0 equiv.), triiodoindigane InI3 (352
mg, 0.71 mmol, 1.1 equiv.) and anhydrous 1,2-dichloroethane (2 mL).
Ethanethiol (0.2 mL, 2.6 mmol, 4.0 equiv.) and 1,1,3,3-
tetramethyldisiloxane (TMDS) (0.7 mL, 3.9 mmol, 6.0 equiv.) were then
added and the reaction mixture was heated at 60 °C during 16h. The
completion of the reaction was monitored by TLC and the reaction
medium was evaporated to dryness. The crude product was dissolved in
CH2Cl2 (10 mL), transferred to a separatory funnel and water (3 mL) was
added. The layers were separated and the aqueous phase was then
extracted with CH2Cl2 (3×5 mL). The combined organic layers were dried
over MgSO4, filtered and concentrated. The product was purified by
column chromatography (ethyl acetate/cyclohexane 20:80) to yield
compound 10 (107 mg, 0.56 mmol, 86%) as a white solid. Rf=0.3 (ethyl
acetate/cyclohexane 30:70); m.p. 66-67°C; 1H NMR (300 MHz, CDCl3,
25°C): δ=10.09 (bs, 1H), 8.05 (d, 4J = 0.9 Hz, 1H), 7.71 (dd, 3J = 8.3 Hz,
4J = 0.9 Hz, 1H), 7.44 (m, 1H), 7.18 (dd, 3J = 8.3 Hz, 4J = 1.4 Hz, 1H),
3.87 (s, 2H), 2.46 (q, 3J = 7.4 Hz, 2H), 1.25 (t, 3J = 7.4 Hz, 3H); 13C NMR
(63 MHz, CDCl3, 25°C): δ=140.4, 138.0, 134.4, 122.7, 122.4, 120.8,
followed by cooling to room temperature yielded
a white solid.
Chloroform (10 mL) was added and the resulting suspension was
transferred to a conical centrifuge tube. After centrifugation at 3000 rpm
for 30 min, the supernatant (8 mL) was separated and 8 mL of chloroform
were added. The operation was repeated three times. The combined
supernatants were evaporated to dryness. The residue was solubilized in
a minimum amount of CH2Cl2, and MeOH was then added (v/v = 1:1).
The thallium salt 5.Tl crystallized by slow evaporation to give white
crystals (392 mg, 0.496 mmol) in 54% yield. 1H NMR (300 MHz, CD2Cl2,
25°C): δ=8.05 (d, 4J = 0.8 Hz, 3H, Ha), 8.01 (m, 3H, Hd), 7.62 (dd, 3J =
8.3 Hz, 4J = 0.8 Hz, 3H, Hb), 7.11 (dd, 3J = 8.3 Hz, 4J = 1.4 Hz, 3H, Hc),
3.93 (s, 6H, He), 2.46 (q, 3J = 7.4 Hz, 6H, Hf), 1.25 (t, 3J = 7.4 Hz, 9H,
Hg); 13C NMR (75 MHz, CD2Cl2, 25°C): δ=144.9 (C2), 137.6 (C1), 133.5
(C5), 122.7 (C7), 122.5 (C4), 121.0 (C3), 112.4 (C6), 37.0 (C8), 25.7 (C9),
14.7 (C10); HRMS (ESI-): calcd for C30H34BN6S3 [M-Tl]-:584.2136, found:
584.2136.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800990
European Journal of Organic Chemistry
109.6, 36.3, 25.4, 14.3; IR (ATR): ṽ=3183 cm−1 (N-H); HRMS (DCI/CH4):
calcd for C10H13N2S [MH]+:193.0799, found: 193.0799. The data match
those reported in the literature.[17]
2nd edn, 2008; b) J.-P. Sauvage, Angew. Chem. 2017, 129, 11228-
11242; Angew. Chem. Int. Ed. 2017, 56, 11080-11093; c) C. J. Bruns, J.
F. Stoddart, The Nature of the Mechanical Bond: From Molecules to
Machines, John Wiley and Sons, Hoboken, 2016.
[2]
[3]
a) S. Erbas-Cakmak, D. A. Leigh, C. T. McTernan, A. L. Nussbaumer,
Chem. Rev. 2015, 115, 10081-10206; b) C. Pezzato, C. Cheng, J. F.
Stoddart and R. D. Astumian, Chem. Soc. Rev. 2017, 46, 5491-5507.
a) J. C. Chambron, C. Dietrich-Buchecker, G. Rapenne, J.-P. Sauvage,
Chirality 1998, 10, 125-133; b) G. S. Kottas, L. I. Clarke, D. Horinek, J.
Michl, Chem. Rev. 2005, 105, 1281-1376; c) E. R. Kay, D. A. Leigh, F.
Zerbetto, Angew. Chem. 2007, 119, 72-196; Angew. Chem. Int. Ed.
2007, 46, 72-191; d) S. Kassem, T. van Leeuwen, A. S. Lubbe, M. R.
Wilson, B. L. Feringa, D. A Leigh, Chem. Soc. Rev. 2017, 46, 2592-
2621 and references therein.
Compound 12:
Conditions A (classical heating): In a dry Schlenk tube under argon,
penta-arylbromide 2 (26 mg, 17 µmol, 1.0 equiv.), palladium(II) acetate
(3.9 mg, 17 µmol, 1.0 equiv.), K3PO4 (72 mg, 340 µmol, 20 equiv.),
ferroceneboronic acid (156 mg, 680 µmol, 40 equiv.) and 2-
dicyclohexylphosphino-2'-6'-dimethoxybiphenyl (SPhos) (14 mg, 34 µmol,
2.0 equiv.) were successively introduced and degassed anhydrous
toluene (1.5 mL) was added. The resulting suspension was stirred at
100 °C for 48 hours and the completion of the reaction was monitored by
TLC. The reaction mixture was allowed to cool to room temperature,
filtered on a neutral alumina pad (using CH2Cl2) and evaporated in vacuo.
The residue was purified by column chromatography (neutral alumina,
CH2Cl2/cyclohexane 30:70) followed (if required) by a recrystallization in
a heptane/MeOH mixture (1:1). Complex 12 was obtained as an orange-
red solid (6 mg, 2.9 µmol, 17%).
[4]
[5]
T. R. Kelly, H. De Silva, R. A. Silva, Nature 1999, 401, 150-152.
N. Koumura, R. W. J. Zijlstra, R. A. van Delden, N. Harada, B. L.
Feringa, Nature 1999, 401, 152-155.
[6]
a) R. Eelkema, M. M. Pollard, J. Vicario, N. Katsonis, B. S. Ramon, C.
W. M. Bastiaansen, D. J. Broer, B. L. Feringa, Nature 2006, 440, 163;
b) K.-Y. Chen, O. Ivashenko, G. T. Carroll, J. Robertus, J. C. M.
Kistemaker, G. London, W. R. Browne, P. Rudolf, B. L. Feringa, J. Am.
Chem. Soc. 2014, 136, 3219-3224; c) Q. Li, G. Fuks, E. Moulin, M.
Maaloum, M. Rawiso, I. Kulic, J. T. Foy, N. Giuseppone, Nat.
Nanotechnol. 2015, 10, 161-165; d) A. Saywell, A. Bakker, J. Mielke, T.
Kumagai, M. Wolf, V. Garcia-Lopez, P.-T. Chiang, J. M. Tour, L. Grill,
ACS Nano 2016, 10, 10945-10952.
Conditions
microwave irradiation and under argon, penta-arylbromide 2 (50 mg, 33
µmol, 1.0 equiv.), 1,1'-bis(diphenylphosphino)ferrocene-
B (microwave irradiation): In a dry tube designed for
palladium(II)dichloride dichloromethane PdCl2(dppf) CH2Cl2 (27 mg, 33
µmol, 1.0 equiv.), sodium tert-butoxide (71 mg, 740 µmol, 22.5 equiv.),
ferroceneboronic acid (57 mg, 246 µmol, 7.5 equiv.) were successively
introduced and degassed anhydrous toluene (3.6 mL) was added. The
tube was sealed and the reaction mixture was heated under microwave
irradiation at 135 °C for 1h (a pressure of 5 bar was achieved). The
reaction mixture was allowed to cool to room temperature, filtered on a
neutral alumina pad (using CH2Cl2) and evaporated in vacuo. The
residue was purified by column chromatography (neutral alumina,
CH2Cl2/cyclohexane 30:70) followed (if required) by a recrystallization in
a heptane/MeOH mixture (1:1). Complex 12 was obtained as an orange-
red solid (30 mg, 15 µmol, 45%). Rf=0.2 (CH2Cl2/cyclohexane 30:70,
SiO2); 1H NMR (500 MHz, CD2Cl2, 25°C): δ=8.11 (br s, ~3H, Ha), 7.92 (bs,
3H, Hd), 7.36 (m, 13H, Hb and Hh), 7.20 (d, 3J = 8.5 Hz, 10H, Hi), 6.99 (dd,
3J = 8.4 Hz, 4J = 1.3 Hz, 3H, Hc), 4.55 (dd, 3J = 1.9 Hz, 4J = 1.8 Hz, 10H,
Hj), 4.24 (dd, 3J = 1.9 Hz, 4J = 1.8 Hz, 10H, Hk), 3.94 (s, 25H, Hl), 3.90 (s,
6H, He), 2.48 (q, 3J = 7.3 Hz, 6H, Hf), 1.28 (t, 3J = 7.3 Hz, 9H, Hg);
13C NMR (125 MHz, CD2Cl2, 25°C): δ=144.1 (C2), 140.4 (C1), 138.6 (C15),
137.6 (C5), 133.9 (C13), 132.4 (C12), 125.1 (C14), 122.6 (C7), 122.4 (C4),
120.4 (C3), 111.4 (C6), 87.8 (C11), 84.7 (C16), 70.2 (C19), 69.6 (C17), 66.7
(C18), 36.9 (C8), 25.7 (C9), 14.8 (C10); UV/Vis (CH2Cl2): λmax (ε)= 292
(30300), 340 nm (16000 mol−1dm3cm−1); HRMS (ESI+): calcd for
C115H99BFe5N6RuS3 [MH]+:2053.3108, found: 2053.3103.
[7]
[8]
[9]
H. L. Tierney, C. J. Murphy, A. D. Jewell, A. E. Baber, E. V. Iski, H. Y.
Khodaverdian, A. F. McGuire, N. Klebanov, E. C. H. Sykes, Nat.
Nanotechnol. 2011, 6, 625-629.
T. Kudernac, N. Ruangsupapichat, M. Parschau, B. Macia, N. Katsonis,
S. R. Harutyunyan, K.-H. Ernst, B. L. Feringa, Nature 2011, 479, 208-
211.
U. G. E. Perera, F. Ample, H. Kersell, Y. Zhang, G. Vives, J. Echeverria,
M. Grisolia, G. Rapenne, C. Joachim, S.-W. Hla, Nature Nanotechnol.
2013, 8, 46-51.
[10] Y. Zhang, H. Kersell, R. Stefak, J. Echeverria, V. Iancu, U. G. E. Perera,
Y. Li, A. Deshpande, K.-F. Braun, C. Joachim,G. Rapenne, S.-W. Hla,
Nature Nanotech. 2016, 11, 706-712.
[11] G. Vives, G. Rapenne, Tetrahedron 2008, 64, 11462-11468.
[12] a) C. Joachim, G. Rapenne, Top. Curr. Chem. 2014, 354, 253-277; b)
R. Stefak, A. M. Sirven, S. Fukumoto, H. Nakagawa, G. Rapenne,
Coord. Chem. Rev. 2015, 287, 79-88; c) C. Kammerer, G. Rapenne,
Eur. J. Inorg. Chem. 2016, 2214-2226.
[13] a) G. Vives, G. Rapenne, Tetrahedron Lett. 2006, 47, 8741-8744; b) G.
Vives, A. Carella, S. Sistach, J.-P. Launay, G. Rapenne, New. J. Chem.
2006, 30, 1429-1438; c) G. Vives, A. Gonzalez, J. Jaud, J.-P. Launay,
G. Rapenne, Chem. Eur. J. 2007, 13, 5622-5631.
[14] A. Carella, R. Poteau, G. Rapenne, J.-P. Launay, Chem. Eur. J. 2008,
14, 8147-8156.
[15] N. G. Connelly, I. Manners, J. Chem. Soc., Dalton Trans., 1989, 30,
283-288.
Acknowledgements
[16] Unpublished results.
[17] A. Carella, G. Vives, T. Cox, J. Jaud, G. Rapenne, J.-P. Launay, Eur. J.
Inorg. Chem. 2006, 980-987.
This work was supported by the University Paul Sabatier
(Toulouse, France) and the Centre National de la Recherche
Scientifique (CNRS). It has received funding from the Agence
Nationale de la Recherche (ANR) (ACTION project ANR-15-
CE29-0005) and from the European Union’s Horizon 2020
research and innovation programme under the project MEMO,
grant agreement No 766864.
[18] T. Miyazaki, S. Kasai, Y. Ogiwara, N. Sakai, Eur. J. Org. Chem. 2016,
1043-1049.
[19] For a recent review on the use of TMDS in organic synthesis, see: J.
Pesti, G. L.Larson, Org. Process Res. Dev. 2016, 20, 1164-1181.
[20] N. Sakai, K. Fujii, T. Konakahara, Tetrahedron Lett. 2008, 49, 6873-
6875.
[21] S. Trofimenko, Chem. Rev. 1993, 93, 943-980.
[22] C. Santini, M. Pellei, G. Gioia Lobbia, G. Papini, Mini-Rev. Org. Chem.
2010, 7, 84-124.
Keywords: scorpionate ligand • indazole • reductive sulfidation •
[23] M. Kitamura, Y. Takenaka, T. Okuno, R. Holl, B. Wünsch, Eur. J. Inorg.
Chem. 2008, 1188-1192.
indium • molecular motor
[24] K. Tsuda, K. Miyata, T. Okuno, M. Yoshimura, S. Tanaka, M. Kitamura,
Tetrahedron Lett. 2008, 2990-2993.
[1]
a) V. Balzani, A. Credi, M. Venturi, Molecular Devices and Machines:
Concepts and Perspectives for the Nanoworld, Wiley-VCH, Weinheim,
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800990
European Journal of Organic Chemistry
[25] T. W. Greene, P. G. M. Wuts, Protective Groups in Organic Synthesis,
Third Edition, John Wiley and Sons, New York, 1999, pp. 552-554.
[26] H.-P. Jacquot de Rouville, D. Villenave, G. Rapenne, Tetrahedron 2010,
66, 1885-1891.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201800990
European Journal of Organic Chemistry
Entry for the Table of Contents
FULL PAPER
The development of a one-pot
Molecular machines
indium(III)-mediated “N-deprotection /
ester reductive sulfidation” sequence
led to major improvement in the
synthesis of thioether-functionalized
hydrotris(indazolyl)borate ligands,
used as anchoring platforms in a
family of surface-mounted rotary
molecular motors and gears.
Guillaume Erbland, Yohan Gisbert,
Gwénaël Rapenne, Claire Kammerer*
Page No. – Page No.
Expedient synthesis of thioether-
functionalized
hydrotris(indazolyl)borate as an
anchoring platform for rotary
molecular machines
This article is protected by copyright. All rights reserved.