A Convenient Access to a [2]Rotaxane Proton Shuttle by Using a Fluorous Ponytail

Article abstract of DOI:10.1055/s-0034-1380814

The properties of rotaxanes and their constituents, ring and axle, sometimes do not differ much from one another resulting in tedious workup. In the case of a rotaxane designed to shuttle protons across a biological membrane (3-4 nm), molecular weight, shape, and functional groups of axle and rotaxane are similar. But when the macrocyclic ring of the rotaxane carries a fluorous residue, the fluorous effect distinguishes the rotaxane from the axle because the latter carries no fluorine atoms. This concept has been exploited to synthesize a [2]rotaxane in which the macrocyclic ring is protonable and the axle contains a permanent positive charge. Upon protonation/deprotonation of the macrocycle, a shuttling process is induced, which can lead to the transport of protons.

Full text of DOI:10.1055/s-0034-1380814

SYNTHESIS
0
0
3
9
-
7
8
8
1
1
4
3
7
-
2
1
0
X
© Georg Thieme Verlag Stuttgart · New York
2015, 47, 2485–2495
paper
2485
O. Beyer et al.
Paper
Synthesis
A Convenient Access to a [2]Rotaxane Proton Shuttle by Using a
Fluorous Ponytail
Ole Beyer
Britta Hesseler
Ulrich Lüning*
Christian-Albrechts-Universität zu Kiel, Otto-Diels-Institut für
Organische Chemie, Olshausenstraße 40, 24098 Kiel, Germany
luening@oc.uni-kiel.de
Dedicated to Professor Bernd Giese on the occasion of his 75th
birthday
Received: 26.03.2015
Accepted after revision: 18.04.2015
Published online: 17.06.2015
which forms an aci-nitro derivative when irradiated,4 and
several derivatives thereof have been developed.5 By placing
such a photoacid on one side on the membrane, a pH gradi-
ent can be generated by irradiation.
DOI: 10.1055/s-0034-1380814; Art ID: ss-2015-t0202-op
Abstract The properties of rotaxanes and their constituents, ring and
axle, sometimes do not differ much from one another resulting in te-
dious workup. In the case of a rotaxane designed to shuttle protons
across a biological membrane (3–4 nm), molecular weight, shape, and
functional groups of axle and rotaxane are similar. But when the macro-
cyclic ring of the rotaxane carries a fluorous residue, the fluorous effect
distinguishes the rotaxane from the axle because the latter carries no
fluorine atoms. This concept has been exploited to synthesize a [2]ro-
taxane in which the macrocyclic ring is protonable and the axle contains
a permanent positive charge. Upon protonation/deprotonation of the
macrocycle, a shuttling process is induced, which can lead to the trans-
port of protons.
The last element for a proton pump is a proton carrier. It
must have the following features: it must be located in the
membrane spanning it from one water layer to the other, it
must be protonable by the acidity which is generated by the
irradiation; however, it shall not be too basic as the proton
shall be released at the opposite side of the membrane, and
last, the carrier must be able to cross the membrane but
shall not leave it in order to avoid ‘bleeding’.
Mechanically interlocked molecules (MIMs)6 perfectly
allow a restricted mobility. In a rotaxane,7 the ring can
move from one side of the rotaxane to the other but cannot
leave the axle. Accordingly, we have devised a [2]rotaxane
containing a protonable macrocyclic ring8 to be used as a
proton shuttle. The axle of the rotaxane is designed unsym-
metrically with a permanent positive charge and a binding
site for the macrocycle on the side from which the proton
shall be taken up.
In the resting state, the macrocycle is bound to the bind-
ing site. When a proton is delivered from this side [by a low
pH (passive transport), or by use of a photoacid (active
transport)], the macrocycle becomes protonated. The per-
manent positive charge in the axle repels the protonated
macrocycle, which moves to the opposite end of the rotax-
ane. There, the proton can be released. If the rotaxane spans
a membrane this process of uptake of a proton on one side
and release on the opposite side of the membrane has al-
lowed the transport of proton across a lipophilic mem-
brane. When this process was propeled by the use of a pho-
toacid, a pH gradient will be formed.
Key words rotaxane, trapping, macrocycles, proton transport, fluo-
rine
The ‘fuel’ in cellular processes is adenosine triphosphate
(ATP). It is synthesized by the enzyme ATPase from adenos-
ine diphosphate (ADP) and phosphate.1 This endergonic
process is powered by a pH gradient. The pH gradient itself
can be generated in different ways, for instance by metabo-
lism of glucose. In contrast, halobacteria use light to build
up a pH potential across the biological membrane.2
Molecular machines and motors try to mimic biological
processes.3 To mimic a light driven proton pumping pro-
cess, a photoacid must be coupled with a transmembrane
transporter. Attention has to be paid to the minute thick-
ness of the biological membrane. The wavelengths of visible
light (ca. 380–780 nm) are two orders of magnitude longer
than the thickness of a biological membrane (3–4 nm).
Therefore, light will always irradiate both sides of a
biomembrane.
For the first realization of a [2]rotaxane proton shuttle
of this design,8 the following elements were incorporated:
the protonable macrocycle contained a 2,6-disubstituted
pyridine, the binding site for the macrocycle was an amide
Photoacids are compounds which undergo an intramo-
lecular reaction upon irradiation, resulting in an isomer
with an increased acidity. A prototype is 2-nitrotoluene,
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 2485–2495

2486
O. Beyer et al.
Paper
Synthesis
group, and the permanent positive charge was introduced
in form of a pyridinium ion. The rotaxane itself was synthe-
sized using the trapping method. In this case, the copper-
catalyzed 1,3-dipolar cycloaddition of an azide to an alkyne
(‘click’ reaction) was used because copper coordinates to
the pyridine unit of the macrocycle.9
Several [2]rotaxanes were synthesized but the yields
were not satisfactory.10 A major problem was the separa-
tion of the desired [2]rotaxane from the free axle, which is
also formed in the synthesis. Although free axle and rotax-
ane differ by the macrocycle, they are still too similar.
Therefore, we have changed the nature of the macrocycle by
introduction of a fluorinated residue.
When fluorine atoms are present, interactions between
molecules are drastically changed.11 Fluorinated molecules
behave different than other (nonfluorous) molecules, dif-
ferent than polar and also unpolar ones. When the extent of
fluorination is high, even separate phases are formed, and
the fluorous effect has been exploited in multilayer systems
of immiscible liquids.12 But also in compounds with a lower
degree of fluorination, the fluorine content alters the na-
ture of a molecule sufficiently to allow easier workup, for
instance by extraction, or by chromatography with fluori-
nated phases.11 Therefore, a fluorous ‘ponytail’ was at-
tached to the macrocycle of a [2]rotaxane proton shuttle8 as
depicted in Figure 1.
Figure 1 Mode of operation of a transmembrane proton shuttle based
on a [2]rotaxane: a) In the resting state, the macrocycle is bound to a
binding site next to a permanent positive charge. From this side a pro-
ton is delivered. b) The macrocycle becomes protonated and is repelled
by the permanent positive charge in the axle. c) Once the protonated
macrocycle reaches the other end of the rotaxane, the proton can be
released, and the deprotonated macrocycle will diffuse back to the
binding site.
fluorous iodide. Due to the low yields of the Mitsunobu re-
action, the Williamson ether synthesis was preferred. But
also in this reaction, the conditions had to be optimized.
The combination of silver carbonate as base with toluene as
solvent led to the best yield of fluorous ester 3 (93%),
whereas standard conditions for Williamson ether synthe-
sis led to no product at all. Fluorous diester 3 was then re-
duced with sodium borohydride to yield fluorous diol 4 in
90% yield. Next, 4 was brominated with phosphorus tribro-
mide to give 84% of fluorous dibromide 5 (Scheme 1).
Starting from fluorous dibromide 5, macrocycle precur-
sor 7 could be obtained by a double Williamson ether syn-
thesis with literature known benzyl alcohol 6.15 First, sodi-
um hydride was used as base but the ether synthesis only
gave 5% of the desired diene 7. The low yield could be traced
back to elimination reactions at the fluorous ponytail. By
mass spectrometry, elimination of hydrogen fluoride and
removal of the whole fluorous ponytail could be detected.
Whether excess hydride anions or the sodium alcoholates
attack the slightly acidic methylene group next to the fluo-
In the previous proton shuttles,8,10 the pyridine ring of
the macrocycle was 4-methoxy substituted. In order not to
influence the pyridine ring and its basicity too much, we
decided to use a perfluoroalkylethoxy substituent, in par-
ticular the perfluorohexylethoxy residue because the re-
spective iodide and the respective alcohol are commercially
available. The chosen ponytail possesses two methylene
groups as spacer, which isolates any influence of the fluo-
rous residue from the pyridine and for instance its basicity.
The fluorous ponytail was introduced into the 29-mem-
bered pyridine macrocycle 9 in a nine-step synthesis. Cheli-
damic acid (1) and its ethyl ester 2 were prepared according
to literature procedures.13,14 To connect the fluorous pony-
tail to the hydroxyl group of chelidamic ester 2, two ether
forming reactions were tested: a Mitsunobu reaction with a
fluorous alcohol and a Williamson ether synthesis with a
EtO2C
N
CO2Et
EtO2C
N
CO2Et
HO2C
N
CO2H
ICH2CH2C6F13
Ag2CO3
H2SO4
EtOH
toluene
OCH2CH2C6F13
OH
OH
1
2
3
54%
93%
N
N
NaBH4
THF
HO
OH
PBr3
Br
Br
CHCl3
OCH2CH2C6F13
OCH2CH2C6F13
4
5
90%
84%
Scheme 1 Synthesis of fluorous dibromide 5
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 2485–2495

2487
O. Beyer et al.
Paper
Synthesis
rous residue has not been investigated, instead sterically
demanding lithium diisopropylamide was chosen. Applying
this base, the double Williamson ether synthesis led to 34%
of isolated diene 7.
Ring-closing metathesis of macrocycle precursor 7 using
Grubbs’ catalyst (1st generation) yielded unsaturated mac-
rocycle 8 as a mixture of the E- and Z-isomers in 59% yield.
Hydrogenation of this mixture using platinum(IV) oxide as
catalyst gave the saturated fluorous macrocycle 9 in 75%
yield (Scheme 2). Overall, 130 mg of the fluorous ring 9 was
obtained in a total yield of 5%.
Although this yield of 14% is the highest yield so far for
the synthesis of a rotaxane using the half axles 10 and 11,
the yield is comparably low with respect to related but
smaller [2]rotaxanes, which have also been synthesized
from simpler azide and alkyne half axles using the Cu(I)-
catalyzed 1,3-cycloaddition. Leigh et al. were able to im-
prove the yield of [2]rotaxane formation by using five
equivalents of each half axle instead of only one equiva-
lent.16 But due to the longer and more time consuming syn-
theses of the half axles 10 and 11, this approach was not
chosen here because it would also produce more than 80%
of free axle.
OCH2CH2C6F13
The structure of the fluorous rotaxane 12 was elucidat-
ed by different mass spectrometry and NMR spectroscopy
experiments. MALDI-TOF spectrometry supplied a peak at
m/z = 2520, which corresponds to the mass of the rotaxane
12 without its hexafluorophosphate counter ion. The ele-
mental composition was confirmed by high resolution ion
cyclotron mass spectrometry, which showed a signal for
m/z = 1260.698. This corresponds to a protonated, doubly
charged rotaxane 12 without the hexafluorophosphate
counter ion.
OH
N
N
O
O
Br
Br
LDA
THF
+ 2
OCH2CH2C6F13
O
4
O
O
4
4
5
6
7
A comparison of NMR spectra of rotaxane 12, free axle
13, and macrocycle 9 proved the structure of rotaxane 12
and furthermore determined the position of the macrocycle
on the axle (Figure 2). The chemically induced upfield shifts
of the pyridinium ion signals (3, 4, 5, 6) and signals 1, 2, and
7 of rotaxane 12 argue that the preferred position of the
macrocycle is close to the pyridinium ion. Due to the aro-
matic rings of the macrocycle, the encapsulated part of the
axle is shielded and the respective signals are shifted up-
field. The position of macrocycle could also be confirmed by
2 D NOESY experiments, which showed cross signals be-
tween the aromatic protons c of macrocycle and signals 2,
3, 4, 5, and 6 of the axle. Two interactions may contribute to
a preferred positioning of the macrocycle at this end of the
axle: (i) formation of a hydrogen bond between the amide
proton of the axle and the pyridine nitrogen atom of the
macrocycle, and (ii) π–π-interactions between the aromatic
rings of the macrocycle and the electron poor pyridinium
ion of the axle.
34%
Grubbs I
CH2Cl2
OCH2CH2C6F13
OCH2CH2C6F13
N
N
O
O
O
O
H2, PtO2
CHCl3
O
O
O
O
4
4
4
4
9
75%
8
59%
Scheme 2 Synthesis of fluorous macrocycle 9
The synthesis of the envisaged fluorinated rotaxane 12
was carried out as developed for the nonfluorous analogue.8
Cu(I)-catalyzed 1,3-dipolar cycloaddition (‘click’ reaction)
of an alkyne half axle 108 with an azide half axle 118 was
chosen to generate a triazole containing axle. Due to the co-
ordination of the copper ions to the nitrogen atom of mac-
rocycle 9, the click reaction may take place inside the mac-
rocycle thus producing the desired rotaxane 12 besides the
free axle 13. The copper ion simultaneously acts as catalyst
for the 1,3-cycloaddition and as template to form the me-
chanically interlocked structure 12. Therefore, copper had
to be used in equimolar and not catalytic amounts. The de-
sired fluorous rotaxane 12 was isolated in 14% yield
(Scheme 3).
But also other signals of the macrocyclic ring 9 support
1
the rotaxane structure. The H NMR spectrum of the sym-
metric macrocycle 9 shows one singlet each for the CH2
groups ArCH2 and PyCH2. But in rotaxane 12, these signals
become more complex. Due to the unsymmetric axle, the
macrocycle in rotaxane 12 is less symmetric. It possesses a
front and a back side, and consequently the CH2 singlets be-
come pairs of doublets in rotaxane 12.
The mechanically interlocked structure of rotaxane 12
was also proven by DOSY NMR experiments. The diffusion
constant of a molecule is correlated with its solvodynamic
radius, which depends on the size and shape of the mole-
cule. Due to the smaller size of macrocycle 9 in comparison
to axle 13 and rotaxane 12, macrocycle 9 diffuses fastest
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 2485–2495

2488
O. Beyer et al.
Paper
Synthesis
PF6
O
O
6
H
N
O
O
3
N
O
Cu(I)(MeCN)4PF6
CH2Cl2
10
OCH2CH2C6F13
O
N
N3
O
O
+
+
O
5
O
O
11
4
4
9
OCH2CH2C6F13
PF6
N
O
O
6
O
H
O
N
O
O
O
N
3
N
4
N
O
N
O
O
4
O
4
12
14%
+
PF6
O
O
6
H
N
O
3
O
O
N
N
N
4
O
N
O
13
48%
Scheme 3 Synthesis of fluorous rotaxane 12
and therefore possesses the largest diffusion constant (Ta-
ble 1). For rotaxane 12, the diffusion constant derived from
signals of ring protons is identical to the one obtained from
protons of the axle giving additional evidence for the me-
chanical interlocking of axle and macrocyclic ring in rotax-
ane 12. In a separate experiment, axle 13 and macrocycle 9
were mixed. Remarkably, the diffusion constant deter-
mined from ring protons in this mixture was different from
free macrocycle 9 but also rotaxane 12. This argues for su-
pramolecular interactions between macrocycle 9 and axle
13 when they are mixed, probably the same interactions,
which are responsible for the preferred position of the mac-
rocycle close to the pyridinium unit of the axle in rotaxane
12. A dynamic equilibrium between free axle 13 and free
macrocycle 9 and such a supramolecular aggregate could
explain the slightly smaller diffusion constant for macrocy-
cle 9 in the mixture when compared to the free macrocycle.
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 2485–2495

2489
O. Beyer et al.
Paper
Synthesis
Figure 2 1H NMR spectra of free axle 13 (bottom), rotaxane 12 (middle), and macrocycle 9 (top) in CD2Cl2. Protons of the axle are numbered, protons
of the macrocyclic ring are labeled with letters. Chemically induced shifts are indicated by solid lines.
Table 1 Diffusion Constants D at 298 K for Rotaxane 12, its Compo-
nents 13 and 9 and a Mixture of the Two Determined in CD2Cl2
tected, indicating similar solvodynamic radii of rotaxane
12, free axle 13, and supramolecular aggregates of axle 13
and macrocycle 9.
D [10–10 m2s–1] signals for
Finally, the ability of rotaxane 12 to transport a proton
along the axle was investigated by 1H NMR experiments. Di-
fluoroacetic acid was chosen as proton source for two rea-
sons. Its acidity is quite strong but in contrast to trifluoro-
acetic acid, its concentration can be determined in 1H NMR
spectra by integration. After addition of this acid to a solu-
tion of rotaxane 12, the pyridine nitrogen atom of the mac-
rocyclic ring became protonated. The repulsive interaction
between the protonated, therefore positively charged mac-
rocycle and the pyridinium cation of the axle led to a move-
ment of the macrocycle along the axle (Scheme 4).
Axle
Macrocycle
4.64
Rotaxane 12
Axle 13
4.67
4.88
Macrocycle 9
9.35
7.38
Axle 13 + macrocycle 9 4.59
When the diffusion constants (see Table 1) that were
obtained from protons of the axle are compared for free
axle 13 and rotaxane 12, only small differences can be de-
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 2485–2495

2490
O. Beyer et al.
Paper
Synthesis
OCH2CH2C6F13
a
PF6
1
N
O
O
2
O
6
O
13
15
16
H
N
7
3
17
11
10
O
O
N
N
12
14
8
19
9
b
c
18
N
O
N
4
6
O
20
5
O
O
d
O
4
Et3N
CHF2COOH
OCH2CH2C6F13
a
CHF2COO
PF6
N
H
1
O
O
2
O
6
O
H
N
15
7
16
17
3
14
11
10
13
O
O
N
N
12
9
8
19
b
N
18
O
N
4
O
20
6
c
5
O
O
d
O
4
Scheme 4
The movement of the macrocyclic ring in rotaxane 12 as
well as its new position on the axle could be identified by
1H NMR spectra and 2D NOESY experiments. A downfield
shift of the signals 1, 2, 4, 5, and 6 of the axle indicated that
the protonated macrocycle is no longer next to the pyridin-
ium ion. Instead, an upfield shift of the signals 11 to 15 was
observed (Figure 3). This suggests that the macrocycle is
predominantly located between the triazole ring and an ad-
jacent aromatic ring (11, 12) (Scheme 4). This position may
be preferred due to π–π-stacking of the aromatic rings or a
hydrogen bond between the proton at the pyridine nitro-
gen atom of the macrocycle and a triazole nitrogen atom.
Neutralization with triethylamine led to deprotonation
of the macrocycle and consequently, the macrocycle moved
back to its original position next to the pyridinium ion
(Scheme 4). The respective NMR spectrum showed that
nearly all signals shifted back to their original position (Fig-
ure 3). Slight differences of some signals may be caused by
an exchange of the counter ion in vicinity to the pyridinium
unit. Besides the original hexafluorophosphate ion, now an
excess of difluoracetate ions is present.
that the yield could be doubled for the fluorous rotaxane
12. It must be mentioned that the same workup and clean-
ing procedures were used. This leads to the conclusion that
the fluorous ponytail has a huge impact on the chemical be-
havior of the rotaxane and thus on its separation and yield.
Especially the difficult separation of the rotaxane from the
free axle was simplified due to the fluorous ponytail. The
fluorous rotaxane 12 shall show a faster mobility in chro-
matographies on silica gel. Indeed, this could be seen in TLC
experiments. While the nonfluorous rotaxane possesses an
Rf value of 0.13 on silica gel with dichloromethane–metha-
nol (40:1) as eluent, the fluorous rotaxane 12 has an Rf val-
ue of 0.40. As a result, the separation from the free axle be-
comes easier.
To improve the separation even more, fluorous solid-
phase extraction (FSPE) or column chromatography with
fluorous silica gel might be helpful. In TLC experiments
with fluorous silica gel, we observed a totally different be-
havior for the fluorous rotaxane 12 and the free axle but
only if the hexafluorophosphate anion was exchanged
against a nonfluorous anion like bromide. Due to strong in-
teractions of the hexafluorophosphate anion with the fluo-
rous silica gel, no retention was observed for the fluorous
rotaxane 12 as well as for the free axle. The TLC experi-
ments with bromide as counter ion are shown in Table 2.
The main interest of this work was to optimize the
workup and yield of the rotaxane synthesis from Hesseler.8
A comparison of the yields of the nonfluorous rotaxane (7%)
by Hesseler8 with the fluorous rotaxane 12 (14%) showed
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 2485–2495

2491
O. Beyer et al.
Paper
Synthesis
Figure 3 1H NMR spectra in CD2Cl2 of rotaxane 12 (bottom), rotaxane 12 after addition of 11 equiv of difluoroacetic acid (middle) and rotaxane 12
after neutralization with 11 equiv of triethylamine (top). Protons of the axle are numbered, protons of the ring are labeled with letters (see (Scheme 4).
Chemically induced shifts are indicated by solid lines.
Table 2 TLC Experiments on Fluorous Silica Gel
cycloaddition (‘click’ reaction). Due to the fluorous ponytail,
the separation of macrocyclic ring 9 and rotaxane 12 was
possible by chromatography, and TLC studies showed that
the use of fluorous stationary phases can simplify the
workup even more. The properties of the isolated [2]rotax-
ane 12 were investigated by NMR spectroscopy and a shut-
tling process of the macrocyclic ring along the axle upon
protonation/deprotonation could be verified.
Rf
Solvent
9
13
12
THF
1
1
1
MeOH
0.61
0
0.76
0.63
0.06
MeOH–H2O (4:1)
0
A fluorophobic solvent mixture like methanol–water
elutes the nonfluorous axle 13 only. By using more fluoro-
philic solvents like pure methanol or tetrahydrofuran, the
fluorous macrocycle 9 as well as the fluorous rotaxane 12
could be eluted. This shows that the free axle might easily
be separated by column chromatography with fluorous sili-
ca gel or even with fluorous solid-phase extraction. Never-
theless these experiments have shown the big potential of
fluorous ponytails even for very large molecules like rotax-
ane 12.
All commercially available reagents were used without further purifi-
cation. Anhydrous solvents were obtained with suitable desiccants.
Chelidamic acid (1),13,14 4-hydroxypyridine-2,6-dicarboxylic acid di-
ethyl ester (2),17 4-(hex-5-enyloxy)phenylmethanol (6),15 (1-{2-oxo-
2-[6-(4-{6-[4-(pent-4-ynyloxy)phenyloxy]hexyloxy}phenyloxy)hex-
yl]amino}ethyl)-3-{4-[tris(4-tert-butylphenyl)methyl]phenyl}pyridi-
nium hexafluorophosphate (10),8 and 6-azidohexanoic acid [3,3,3-
tris(4-tert-butylphenyl)]propyl ester (11)8 were prepared according
to literature procedures. The behavior of rotaxane 12 on fluorous sili-
ca gel was checked by analytical TLC using FluoroFlash® F-254 pre-
coated glass plates.
1H and 13C NMR spectra were recorded with Bruker DRX 500 and
Bruker Avance 600 NMR spectrometers. For the numbering used to
assign chemical shifts for the aromatic and pyridine rings of rotaxane
12 and axle 13, please refer to the Supporting Information.
A protonable [2]rotaxane 12 was synthesized by the
trapping method from a fluorous pyridine macrocycle 9, an
alkyne terminated half axle 10, and a second azide termi-
nated half axle 11 using the copper(I) catalyzed 1,3-dipolar
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 2485–2495

2492
O. Beyer et al.
Paper
Synthesis
Mass spectra were recorded with Bruker-Daltronics Biflex III
(MALDI), Bruker-Daltronics APEX 3 FT-ICR 7.05 T (ICR), and Jeol
AccuTOF GCV 4G mass spectrometers. The matrix used for MALDI-
TOF MS analysis is 4-chloro-α-cyanocinnamic acid (Cl-CCA).
MS (EI): m/z = 502 [M + H]+, 501 [M]+•, 500 [M – H]+, 483 [M – H2O]+,
454 [M – CH4O2]+.
HRMS (EI): m/z [M]+• calcd for C15H12F13NO5: 501.0609; found:
501.0598.
4-[2-(Perfluorohexyl)ethyloxy]pyridine-2,6-dicarboxylic Acid Di-
ethyl Ester (3)
2,6-Bis(bromomethyl)-4-[2-(perfluorohexyl)ethyloxy]pyridine (5)
PBr3 (3.75 mL, 39.9 mmol) was added dropwise to a solution of alco-
hol 4 (2.00 g, 3.99 mmol) in CHCl3 (150 mL). The mixture was stirred
6 h under reflux and afterwards 12 h at r.t. Sat. aq NaHCO3 (150 mL)
was added slowly and the mixture was stirred for 2 h at r.t. The aque-
ous phase was extracted with CHCl3 (3 × 50 mL), the combined organ-
ic layer was washed with brine (100 mL), and dried (MgSO4). The sol-
vent was removed under reduced pressure and the crude product was
purified by column chromatography (silica gel, n-heptane–EtOAc,
3:1; Rf = 0.30) to give 5 as a colorless solid; yield: 2.10 g (84%); mp
97 °C.
A mixture of 4-hydroxypyridine-2,6-dicarboxylic acid diethyl ester
(2; 5.03 g, 21.0 mmol) and Ag2CO3 (6.56 g, 23.8 mmol) in anhydrous
toluene (350 mL) was stirred at 120 °C under N2 for 2 h. Then, 1-iodo-
2-(perfluorohexyl)ethane (7.68 mL, 31.4 mmol) was added dropwise
and the reaction mixture was stirred for 3 d at 120 °C. Insoluble mate-
rial was removed by filtration and washed with EtOAc (200 mL). The
solvent was removed under reduced pressure and the crude product
was purified by column chromatography (silica gel, CH2Cl2–EtOAc
15:1; Rf = 0.33) to give 3 as a colorless solid; yield: 11.4 g (93%); mp
70 °C.
IR (ATR): 1597, 1571 (arom C=C), 1236, 1189 (C–O), 1165, 1141, 1122
(C–F), 1043 (C–O), 659 cm–1 (C–Br).
IR (ATR): 2916, 2849 (aliph C–H), 1716 (C=O), 1600, 1584 (arom C=C),
1232, 1188 (C–O), 1143, 1118 (C–F), 1039 cm–1 (C–O).
1H NMR (500 MHz, CDCl3): δ = 6.93 (s, 2 H, PyH), 4.52 (s, 4 H,
1H NMR (500 MHz, CDCl3): δ = 7.79 (s, 2 H, PyH), 4.48 (q, 3J = 7.1 Hz, 4
3
3
PyCH2Br), 4.37 (t, J = 6.6 Hz, 2 H, OCH2CH2), 2.68 (tFt, JH,F = 18.2 Hz,
3
H, OCH2CH3), 4.45 (t, 3J = 6.5 Hz, 2 H, OCH2CH2), 2.71 (tFt, JH,F = 18.0
3J = 6.6 Hz, 2 H, CH2CH2CF2).
Hz, 3J = 6.5 Hz, 2 H, OCH2CH2CF2), 1.45 (t, 3J = 7.1 Hz, 6 H, OCH2CH3).
13C NMR (125 MHz, CDCl3): δ = 166.0 (s, 4-PyC), 158.3 (s, 2,6-PyC),
117.1 (tF, CH2CF2CF2)*, 109.3 (d, 3,5-PyC), 60.4 (t, OCH2CH2CF2), 32.7 (t,
13C NMR (125 MHz, CDCl3): δ = 165.9 (s, 4-PyC), 164.5 (s, PyCO2Et),
150.5 (s, 2,6-PyC), 117.3 (tF, 2JC,F = 32.4 Hz, CH2CF2CF2)*, 114.1 (d, 3,5-
2
*
CH2Br) 31.0 [t(tF), JC,F = 21.9 Hz, CH2CH2CF2]. Only visible in the
2
PyC), 62.5 (t, OCH2CH3), 60.9 (t, OCH2CH2CF2), 31.0 [t(tF), JC,F = 21.8
HMBC spectrum.
*
Hz, CH2CH2CF2], 14.2 (q, OCH2CH3). Only visible in the HMBC spec-
19F NMR (470 MHz, CDCl3): δ = –80.8 (s, 3 F, CF3), –113.2 (s, 2 F,
CH2CF2), –121.8 (s, 2 F, CH2CF2CF2), –122.8 (s, 2 F, CF2CF3), –123.5 (s, 2
F, CF2CF2CF3), –126.1 (s, 2 F, CH2CF2CF2CF2).
trum.
19F NMR (470 MHz, CDCl3): δ = –80.8 (s, 3 F, CF3), –113.2 (s, 2 F,
CH2CF2), –121.8 (s, 2 F, CH2CF2CF2), –122.8 (s, 2 F, CF2CF3), –123.5 (s, 2
F, CF2CF2CF3), –126.1 (s, 2 F, CH2CF2CF2CF2).
MS (EI): m/z = 627 [M + H]+, 626 [M]+•, 546 [M – Br]+, 468 [M + H –
Br2]+.
MS (EI): m/z = 586 [M + H]+, 585 [M]+•, 584 [M – H]+, 512 [M –
HRMS (EI): m/z [M]+ calcd for C15H10Br2F13NO: 624.8921; found:
624.8934.
CO2CH2CH3]+, 439 [M – C6H10O4]+.
HRMS (EI): m/z [M]+̇• calcd for C19H16F13NO5: 585.0821; found:
585.0820.
2,6-Bis[4-(hex-5-enyloxy)phenylmethyloxymethyl]-4-[2-(perfluo-
rohexyl)ethyloxy)pyridine (7)
2,6-Bis(hydroxymethyl)-4-[2-(perfluorohexyl)ethyloxy]pyridine
(4)
A
solution of 4-(hex-5-enyloxy)phenylmethanol (6; 840 mg,
4.07 mmol) in anhydrous THF (30 mL) under N2 was cooled to –78 °C.
i-Pr2NLi (2 M solution in THF, heptane, ethylbenzene, 2.04 mL,
4.07 mmol) was added dropwise and the mixture was stirred for 1 h
at r.t. Afterwards, a solution of bromide 5 (730 mg, 1.16 mmol) in an-
hydrous THF (20 mL) was added, the mixture was heated slowly up to
50 °C, and then stirred for 3 d at 50 °C. After the addition of H2O
(50 mL), THF was removed under reduced pressure. The aqueous
phase was extracted with CH2Cl2 (3 × 50 mL), the combined organic
layer was washed with brine (100 mL), and dried (MgSO4). The sol-
vent was removed under reduced pressure and the crude product was
purified by column chromatography (silica gel, n-heptane–EtOAc,
3:1; Rf = 0.14) to give 7 as a yellow oil; yield: 350 mg (34%).
NaBH4 (675 mg, 17.8 mmol) was added to a solution of ethyl ester 3
(3.48 g, 5.95 mmol) in anhydrous THF (150 mL). The mixture was
kept at 80 °C for 6 h. Afterwards, EtOH (50 mL) was added and the
mixture was stirred for 12 h at r.t. The solvent was removed under re-
duced pressure and the residue was partitioned between EtOAc
(100 mL) and H2O (100 mL). The aqueous layer was extracted with
EtOAc (3 × 150 mL), the combined organic layer was washed with
brine (100 mL), and dried (MgSO4). The solvent was removed under
reduced pressure and 4 was obtained as a colorless solid; yield: 2.67 g
(90%); mp 155 °C.
IR (ATR): 3300 (O–H), 1604, 1575 (arom C=C), 1237, 1191 (C–O),
1161, 1141 (C–F), 1090 cm–1 (C–O).
1H NMR (500 MHz, CDCl3): δ = 6.75 (s, 2 H, PyH), 4.72 (s, 4 H,
PyCH2OH), 4.35 (t, 3J = 6.5 Hz, 2 H, OCH2CH2), 3.34 (br s, 2 H, OH), 2.66
(tFt, 3JH,F = 18.2 Hz, 3J = 6.5 Hz, 2 H, CH2CH2CF2).
IR (ATR): 2934, 2861 (aliph C-H), 1631 (C=C), 1600, 1582, 1512 (arom
C=C), 1237, 1205 (C–O), 1162, 1144 (C–F), 1104 (C–O), 911 cm–1
(C=C–H).
1H NMR (500 MHz, CDCl3): δ = 7.30 (d, 3J = 8.5 Hz, 4 H, 3,5-ArH), 6.94
(s, 2 H, 3,5-PyH), 6.88 (d, 3J = 8.5 Hz, 4 H, 2,6-ArH), 5.83 (ddt, 3J = 17.0
Hz, 3J = 10.1 Hz, 3J = 6.7 Hz, 2 H, CH2CH=CH2), 5.04 (dd, 3J = 17.0 Hz,
2J = 1.4 Hz, 2 H, CH=CHHtrans), 4.98 (d, 3J = 10.1 Hz, 2 H, CH=CHHcis),
13C NMR (125 MHz, CDCl3): δ = 165.7 (s, 4-PyC), 160.6 (s, 2,6-PyC),
117.1 (tF, CH2CF2CF2)*, 105.5 (d, 3,5-PyC), 64.3 (t, CH2OH), 60.2 (t,
2
*
OCH2CH2CF2), 31.0 [t(tF), JC,F = 21.9 Hz, CH2CH2CF2]. Only visible in
3
the HMBC spectrum.
4.62 (s, 4 H, Py-CH2O), 4.58 (s, 4 H, ArCH2O), 4.34 (t, J = 6.5 Hz, 2 H,
PyOCH2CH2CF2), 3.96 (t, 3J = 6.5 Hz, 4 H, ArOCH2), 2.65 (mc, 2 H,
CH2CH2CF2), 2.13 (mc, 4 H, CH2CH=CH2), 1.80 (mc, 4 H, OCH2CH2), 1.57
(mc, 4 H, OCH2CH2CH2).
19F NMR (470 MHz, CDCl3): δ = –80.8 (s, 3 F, CF3), –113.2 (s, 2 F,
CH2CF2), –121.8 (s, 2 F, CH2CF2CF2), –122.8 (s, 2 F, CF2CF3), –123.5 (s, 2
F, CF2CF2CF3), –126.1 (s, 2 F, CH2CF2CF2CF2).
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 2485–2495

2493
O. Beyer et al.
Paper
Synthesis
13C NMR (125 MHz, CDCl3): δ = 166.0 (s, 4-PyC)*, 159.7 (s, 2,6-PyC)
158.8 (s, 1-ArC), 138.5 (d, CH=CH2), 129.7 (s, 4-ArC), 129.6 (d, 3,5-
ArC), 117.3 (tF, CH2CF2CF2)*, 114.7 (t, CH=CH2), 114.5 (d, 2,6-ArC),
106.2 (d, 3,5-PyC), 72.8 (t, ArCH2O), 72.1 (t, PyCH2O), 67.8 (t, ArOCH2),
19F NMR (470 MHz, CDCl3): δ = –80.8 (s, 3 F, CF3), –113.2 (s, 2 F,
CH2CF2), –121.8 (s, 2 F, CH2CF2CF2), –122.8 (s, 2 F, CF2CF3), –123.5 (s, 2
F, CF2CF2CF3), –126.1 (s, 2 F, CH2CF2CF2CF2).
MS (EI) : m/z = 850 [M + H]+, 849 [M]+•, 848 [M – H]+, 483 [M –
2
60.0 (t, PyOCH2), 33.4 (t, CH2CH=CH2) 31.0 [t(tF), JC,F = 21.5 Hz,
C
18H17O3]+, 469 [M + H – C18H17O4]+.
CH2CH2CF2], 28.7 (t, ArOCH2CH2), 25.3 (t, ArOCH2CH2CH2). * Only visi-
MS (MALDI-TOF, Cl-CCA): m/z = 872 [M + Na]+, 850 [M + H]+.
ble in the HMBC spectrum.
Anal. Calcd for C39H40F13NO5: C, 55.13; H, 4.74; N, 1.65. Found: C,
55.30; H, 4.25; N, 1.88.
19F NMR (470 MHz, CDCl3): δ = –80.8 (s, 3 F, CF3), –113.2 (s, 2 F,
CH2CF2), –121.8 (s, 2 F, CH2CF2CF2), –122.8 (s, 2 F, CF2CF3), –123.5 (s, 2
F, CF2CF2CF3), –126.1 (s, 2 F, CH2CF2CF2CF2).
54-[2-(Perfluorohexyl)ethyloxy]-3,7,10,21-tetraoxa-1,9-(1,4)-
dibenzena-5-(2,6)-pyridinaheneicosaphane (9)
MS (EI): m/z = 878 [M+H]+, 877 [M]+•, 876 [M – H]+, 673 [M + H –
C
C
13H17O2]+, 616 [M + H – C17H24O2]+, 590 [M + H – C19H28O2]+, 483 [M –
26H34O3]+, 468 [M + H – C26H34O4]+.
H2 was bubbled through a suspension of PtO2 (700 μg, 2.96 μmol) in
acid-free CHCl3 (3 mL) for 30 min. A solution of macrocycle 8 (28 mg,
33 μmol) in acid-free CHCl3 (2 mL) was added, and the mixture was
flushed with H2 for further 2 h, followed by stirring for 3 d under H2.
The solvent was removed under reduced pressure and the crude
product was purified by column chromatography (silica gel, CH2Cl2–
EtOAc, 5:1, Rf = 0.54) to give 9 as a colorless solid; yield: 21 mg (75%);
mp 85 °C.
MS (MALDI-TOF, Cl-CCA): m/z = 900 [M + Na]+, 878 [M + H]+.
Anal. Calcd for C41H44F13NO5: C, 56.10; H, 5.05; N, 1.60. Found: C,
56.17; H, 4.75; N, 1.84.
54-[2-(Perfluorohexyl)ethyloxy]-3,7,10,21-tetraoxa-1,9(1,4)diben-
zena-5(2,6)pyridinaheneicosaphan-15-ene (8)
A mixture of the macrocycle precursor 7 (300 mg, 342 μmol) and
benzylidene-bis(tricyclohexylphosphine)dichlororuthenium
(28.1 mg, 34.2 μmol) in anhydrous CH2Cl2 (350 mL) was stirred at r.t.
under N2 for 4 d. Afterwards, ethyl vinyl ether (1 mL) was added and
the mixture was stirred for 1 h at r.t. The solvent was removed under
reduced pressure and the crude product was purified by column chro-
matography (silica gel, CH2Cl2–EtOAc, 5:1, Rf = 0.20) to give 8 as a col-
orless solid; yield: 172 mg (59%). Product 8 is a mixture of cis- and
trans-isomers (ratio 1:2).
IR (ATR): 2926, 2861 (aliph C–H), 1599, 1576, 1514 (arom C=C), 1240,
1195 (C–O), 1166, 1143, 1120 (C–F), 1061 cm–1 (C–O).
1H NMR (500 MHz, CDCl3): δ = 7.22 (d, 3J = 8.6 Hz, 4 H, 3,5-ArH), 6.90
(s, 2 H, 3,5-PyH), 6.81 (d, 3J = 8.6 Hz, 4 H, 2,6-ArH), 4.60 (s, 4 H,
ArCH2O), 4.37 (br s,
4 = 6.5 Hz, 2 H,
H, PyCH2O), 4.37 (t, 3J
PyOCH2CH2CF2), 3.97 (t, 3J = 6.4 Hz, 4 H, ArOCH2), 2.66 (mc, 2 H,
CH2CH2CF2), 1.72 (mc, 4 H, OCH2CH2), 1.45–1.37 (m, 4 H,
OCH2CH2CH2), 1.30–1.25 [m, 8 H, O(CH2)3CH2CH2].
13C NMR (125 MHz, CDCl3): δ = 165.6 (s, 4-PyC)*, 159.9 (s, 2,6-PyC)
158.8 (s, 1-ArC), 130.1 (d, 2,6-ArC), 129.1 (s, 4-ArC), 117.3 (tF,
trans-8
CH2CF2CF2)*, 114.6 (d, 3,5-ArC), 106.0 (d, 3,5-PyC), 72.4 (t, ArCH2O),
1H NMR (500 MHz, CDCl3): δ = 7.21 (d, 3J = 8.5 Hz, 4 H, 3,5-ArH), 6.88
(s, 2 H, 3,5-PyH), 6.78 (d, 3J = 8.5 Hz, 4 H, 2,6-ArH), 5.41 (mc, 2 H,
CH=CH), 4.58 (s, 4 H, ArCH2O), 4.39 (br s, 4 H, PyCH2O), 4.36 (t, 3J =
6.4 Hz, 2 H, PyOCH2CH2CF2), 3.93 (t, 3J = 6.7 Hz, 4 H, ArOCH2), 2.66 (mc,
2 H, CH2CH2CF2), 2.02 (mc, 4 H, CH2CH=CHCH2), 1.78–1.68 (m, 4 H,
OCH2CH2), 1.54–1.44 (m, 4 H, OCH2CH2CH2).
13C NMR (125 MHz, CDCl3): δ = 165.8 (s, 4-PyC)*, 159.9 (s, 2,6-PyC),
158.9 (s, 1-ArC), 130.4 (d, CH=CH), 130.1 (d, 3,5-ArC), 129.2 (s, 4-ArC),
117.3 (tF, CH2CF2CF2)*, 114.5 (d, 2,6-ArC), 106.3 (d, 3,5-PyC), 72.4 (t,
ArCH2O), 71.1 (t, PyCH2O), 67.6 (t, ArOCH2), 60.0 (t, PyOCH2), 31.9 (t,
2
70.8 (t, PyCH2O), 67.4 (t, ArOCH2), 59.9 (t, PyOCH2), 31.1 [t(tF), JC,F
=
22.0 Hz, CH2CH2CF2], 29.5 [t, O(CH2)4CH2], 28.7 [t, O(CH2)3CH2], 28.6
*
(t, OCH2CH2), 25.7 (t, OCH2CH2CH2). Only visible in the HMBC spec-
trum.
19F NMR (470 MHz, CDCl3): δ = –80.8 (s, 3 F, CF3), –113.2 (s, 2 F,
CH2CF2), –121.8 (s, 2 F, CH2CF2CF2), –122.8 (s, 2 F, CF2CF3), –123.5 (s, 2
F, CF2CF2CF3), –126.1 (s, 2 F, CH2CF2CF2CF2).
MS (EI): m/z = 852 [M + H]+, 851 [M]+•., 850 [M – H]+, 483 [MH –
C
24H32O3]+, 469 [M – C24H30O4]+.
HRMS (EI): m/z [M]+ calcd for C39H42F13NO5: 851.2855; found:
851.2832.
2
CH2CH=CHCH2), 31.1 [t(tF), JC,F = 22.2 Hz, CH2CH2CF2], 28.4 (t,
OCH2CH2), 25.9 (t, OCH2CH2CH2). * Only visible in the HMBC spectrum.
[2]-{[1-(2-Oxo-2-{[6-(4-{6-[4-(3-{1-[6-(3,3,3-tris{4-tert-butylphe-
nyl}propyloxy)-6-oxohexyl]-1,2,3-triazole-4-yl}propyloxy)phenyl-
oxy]hexyloxy}phenyloxy)hexyl]amino}ethyl)-3-{4-[tris(4-tert-
butylphenyl)methyl]phenyl}pyridiniumhexafluorophosphate]-
rotaxa-[54-(2-{perfluorohexyl}ethyloxy)-3,7,10,21-tetraoxa-
1,9(1,4)dibenzena-5(2,6)pyridinaheneicosaphane]} (12)
cis-8
IR (ATR): 2934, 2869 (aliph C–H), 1599, 1581, 1513 (arom C=C), 1237,
1193 (C–O), 1164, 1143, 1121 (C–F), 1061 cm–1 (C–O).
1H NMR (500 MHz, CDCl3): δ = 7.22 (d, 3J = 8.5 Hz, 4 H, 3,5-ArH), 6.88
(s, 2 H, 3,5-PyH), 6.79 (d, 3J = 8.5 Hz, 4 H, 2,6-ArH), 5.38 (mc, 2 H,
CH=CH), 4.57 (s, 4 H, ArCH2O), 4.44 (br s, 4 H, PyCH2O), 4.36 (t, 3J =
6.4 Hz, 2 H, PyOCH2CH2CF2), 3.95 (t, 3J = 6.7 Hz, 4 H, ArOCH2), 2.66 (mc,
2 H, CH2CH2CF2), 2.06 (mc, 4 H, CH2CH=CHCH2), 1.78–1.68 (m, 4 H,
OCH2CH2), 1.54–1.44 (m, 4 H, OCH2CH2CH2).
13C NMR (125 MHz, CDCl3): δ = 165.8 (s, 4-PyC)*, 159.9 (s, 2,6-PyC),
158.7 (s, 1-ArC), 130.1 (d, 3,5-ArC), 129.9 (d, CH=CH), 129.2 (s, 4-ArC),
117.3 (tF, CH2CF2CF2)*, 114.7 (d, 2,6-ArC), 106.3 (d, 3,5-PyC), 72.4 (t,
ArCH2O), 71.1 (t, PyCH2O), 67.5 (t, ArOCH2), 60.0 (t, PyOCH2), 31.1
[t(tF), JC,F
CH2CH=CHCH2), 25.8 (t, OCH2CH2CH2). Only visible in the HMBC
spectra.
To a mixture of macrocycle 9 (30.0 mg, 35.2 μmol), alkyne half axle 10
(42.9 mg, 35.2 μmol) and azide half axle 11 (21.0 mg, 35.2 μmol) in
anhydrous CH2Cl2 (2 mL) under N2 was added tetrakis(acetoni-
trile)copper(I) hexafluorophosphate (13.1 mg, 35.2 μmol). The mix-
ture was stirred for 2 d at r.t. Afterwards, CH2Cl2 (4 mL) and MeOH
(2 mL) were added to dilute the mixture. KCN (11.5 mg, 176 μmol) in
MeOH (2 mL) was added and the mixture was stirred for 1 h. The sol-
vent was removed under reduced pressure and the crude product was
partitioned between CH2Cl2 (10 mL) and H2O (10 mL). The aqueous
phase was extracted with CH2Cl2 (3 × 10 mL), the combined organic
layer was washed with brine (20 mL), and dried (MgSO4). The solvent
2
=
22.2 Hz, CH2CH2CF2], 28.3 (t, OCH2CH2), 26.6 (t,
*
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 2485–2495

2494
O. Beyer et al.
Paper
Synthesis
was removed under reduced pressure and the crude product was pu-
rified by column chromatography (silica gel, CH2Cl2–MeOH, 40:1; Rf =
0.40) to give 12 as yellow solid; yield: 13 mg (14%). Additionally, the
free axle 13 was isolated; yield: 31 mg (48%).
Axle 13
1H NMR (600 MHz, CD2Cl2): δ = 9.49 (s, 1 H, 2-PyH), 9.01 (br s, 1 H,
CONH), 8.99 (d, 3J = 6.1 Hz, 1 H, 6-PyH), 8.57 (d, 3J = 8.2 Hz, 1 H, 4-
PyH), 7.98 (dd, 3J = 8.2 Hz, 3J = 6.1 Hz, 1 H, 5-PyH), 7.63 (d, 3J = 8.6 Hz,
2 H, 2,6-Ar2H), 7.53 (d, 3J = 8.6 Hz, 2 H, 3,5-Ar2H), 7.36–7.24 (m, 13 H,
3,5-Ar5H, 3,5-Ar1H, 5-triazole-H), 7.24–7.17 (m, 12 H, 2,6-Ar5H, 2,6-
Ar1H), 6.80, 6.78 (2 s, 8 H, 2,3,5,6-Ar3H, 2,3,5,6-Ar4H), 5.68 (s, 2 H,
Py1CH2CONH), 4.28 [t, 3J = 7.2 Hz, 2 H, triazole-CH2(CH2)4CO2], 3.94 [t,
Rotaxane 12
1H NMR (600 MHz, CD2Cl2): δ = 8.49 (s, 1 H, 2-Py1H), 8.04 (d, 3J =
8.1 Hz, 1 H, 4-Py1H), 7.65 (d, 3J = 6.2 Hz, 1 H, 6-Py1H), 7.55 (dd, 3J =
8.1 Hz, 3J = 6.2 Hz, 1 H, 5-Py1H), 7.40 (d, 3J = 8.5 Hz, 2 H, 3,5-Ar2H),
7.33 (d, 3J = 8.7 Hz, 6 H, 3,5-Ar1H), 7.31–7.28 (m, 7 H, 5-triazole-H,
3,5-Ar5H), 7.25 (d, 3J = 8.7 Hz, 6 H, 2,6-Ar1H), 7.23–7.20 (m, 7 H,
CONH, 2,6-Ar5H), 7.16 (d, 3J = 8.5 Hz, 2 H, 2,6-Ar2H), 6.86 (d, 3J =
8.6 Hz, 4 H, 3,5-Ar6H), 6.86 (s, 2 H, 3,5-Py2H), 6.79 (d, 3J = 6.3 Hz, 8 H,
2,3,5,6-Ar3H, 2,3,5,6-Ar4H), 6.31 (d, 3J = 8.6 Hz, 4 H, 2,6-Ar6H), 4.92 (s,
2 H, Py1CH2CONH), 4.43–4.32 (m, 10 H, Py2CH2O, Ar6CH2O, Py2OCH2),
4.27 [t, 3J = 7.2 Hz, 2 H, triazole-CH2(CH2)4CO2], 3.93 [t, 3J = 6.2 Hz, 2 H,
Ar4OCH2(CH2)2-triazole], 3.91–3.85 [m, 6 H, CH2OAr3OCH2(CH2)4CH2OAr4],
3.80 (mc, 2 H, CO2CH2CH2), 3.72–3.64 (mc, 4 H, Ar6OCH2), 3.20–3.14
(m, 2 H, NHCH2), 2.91–2.81 [m, 4 H, Ar4O(CH2)2CH2-triazole,
3J
=
6.2 Hz, 2 H, Ar4OCH2(CH2)2-triazole], 3.91–3.88 [m, 4 H,
Ar3OCH2(CH2)4CH2OAr4], 3.86 [t, 3J = 6.5 Hz, 2 H, NH(CH2)5CH2OAr3],
3.81 (mc, 2 H, CO2CH2CH2), 3.25 (mc, 2 H, NHCH2CH2), 2.91–2.83 [m,
4 H, Ar4O(CH2)2CH2-triazole, CO2CH2CH2], 2.21 [t, J = 7.5 Hz, 2 H, tri-
3
azole-(CH2)4CH2CO2], 2.10 (mc, 2 H, Ar4OCH2CH2CH2-triazole), 1.87
[mc,
2 H,
triazole-CH2CH2(CH2)3CO2],
1.82–1.68
[m,
6 H,
CH2CH2OAr3OCH2CH2(CH2)2CH2CH2OAr4], 1.67–1.55 [m, 4 H, triazole-
(CH2)3CH2CH2CO2,
NHCH2CH2],
1.54–1.47
[m,
8 H,
NH(CH2)2CH2CH2(CH2)2OAr3O(CH2)2CH2CH2(CH2)2OAr4], 1.46–1.37 [m,
2 H, triazole-(CH2)2CH2(CH2)2CO2], 1.31, 1.30 [2 s, 54 H, 6 × C(CH3)3].
13C NMR (150 MHz, CD2Cl2): δ = 173.8 (s, CO2CH2), 164.2 (s, CONH),
153.9, 153.8, 153.7, 153.6 (4 s, 1,4-Ar3C, 1,4-Ar4C), 151.4 (s, 4-Ar2C)*,
149.5 (s, 4-Ar5C), 149.4 (s, 4-Ar1C), 147.7 (s, 4-triazole-C), 144.5 (s, 1-
Ar5C), 144.1 (s, 1-Ar1C), 143.9 (d, 2-PyC), 143.5 (d, 6-PyC), 143.2 (d, 4-
PyC), 141.5 (s, 3-PyC)*, 133.0 (d, 3,5-Ar2C), 131.0 (d, 2,6-Ar1C), 130.3
(s, 1-Ar2C), 129.0 (d, 2,6-Ar5C), 128.2 (d, 5-PyC), 127.0 (d, 2,6-Ar2C),
125.4 (d, 3,5-Ar1C), 125.2 (d, 3,5-Ar5C), 121.3 (d, 5-triazole-C), 115.9
(d, 2,3,5,6-Ar3C, 2,3,5,6-Ar4C), 69.1 [t, CH2OAr3OCH2(CH2)4CH2OAr4],
68.1 (t, Ar4OCH2CH2CH2-triazole), 64.6 [s, C(Ar1)3], 63.1 (t, PyCH2),
63.0 (t, CO2CH2CH2), 54.1 [s, C(Ar5)3], 50.4 [t, triazole-CH2(CH2)4CO2],
40.7 (t, NHCH2), 38.8 (t, CO2CH2CH2), 34.8, 34.8 [2 s, Ar1C(CH3)3,
Ar5C(CH3)3], 34.4 (t, CH2CO2), 31.6 [q, Ar1C(CH3)3, Ar5C(CH3)3],
30.6 [t, triazole-CH2CH2(CH2)3CO2], 29.9, 29.8, 29.7 [3 t,
CH2CH2OAr3OCH2CH2(CH2)2CH2CH2OAr4], 29.5 (t, Ar4OCH2CH2CH2-triazole),
29.2 (t, NHCH2CH2), 26.6 [t, NH(CH2)3CH2(CH2)2OAr3], 26.4 [t,
Ar3O(CH2)2CH2CH2(CH2)2OAr4], 26.3 [t, triazole-(CH2)2CH2(CH2)2CO2],
26.2 [t, NH(CH2)2CH2(CH2)3OAr3], 24.8 [t, triazole-(CH2)3CH2CH2CO2],
22.7 (t, Ar4OCH2CH2CH2-triazole). * Only visible in the HMBC spectra.
3
CO2CH2CH2], 2.65 (mc, 2 H, OCH2CH2CF2), 2.21 [t, J = 7.6 Hz, 2 H, tri-
azole-(CH2)4CH2CO2], 2.10 (mc, 2 H, Ar4OCH2CH2CH2-triazole), 1.85
[mc,
2 H,
triazole-CH2CH2(CH2)3CO2],
1.80–1.69
1.67–1.61
[m,
(m,
6 H,
4 H,
CH2CH2OAr3OCH2CH2(CH2)2CH2CH2OAr4],
Ar6OCH2CH2), 1.61–1.56 [m, 2 H, triazole-(CH2)3CH2CH2CO2], 1.52–
1.48 [m, 6 H, NHCH2CH2, Ar3O(CH2)2CH2CH2(CH2)2OAr4], 1.47–1.44
[m, 2 H, NH(CH2)3CH2(CH2)2OAr3], 1.41–1.38 [m, 4 H, Ar6O(CH2)2CH2],
1.37–1.34 [m, 2 H, NH(CH2)2CH2(CH2)3OAr3], 1.34–1.22 [m, 10 H,
Ar6O(CH2)3CH2CH2, triazole-(CH2)2CH2(CH2)2CO2], 1.30, 1.29 [2 s, 54 H,
6 × C(CH3)3].
13C NMR (150 MHz, CD2Cl2): δ = 173.8 (s, CO2CH2), 166.3 (s, 4-Py2C),
163.4 (s, CONH)*, 159.7 (s, 2,6-Py2C), 159.2 (s, 1-Ar6C), 153.9, 153.8,
153.7, 153.6 (4 s, 1,4-Ar3C, 1,4-Ar4C), 151.1 (s, 4-Ar2C)*, 149.5 (s, 4-
Ar1C), 149.4 (s, 4-Ar5C), 147.7 (s, 4-triazole-C), 144.7 (d, 2-Py1C),144.5
(s, 1-Ar5C), 144.3 (s, 1-Ar1C), 142.8 (d, 6-Py1C), 141.6 (d, 4-Py1C),
140.3 (s, 3-Py1C), 132.2 (d, 3,5-Ar2C), 130.9 (d, 2,6-Ar1C), 130.7 (d, 3,5-
Ar6C), 129.5 (s, 1-Ar2C), 129.3 (s, 4-Ar6C)*, 129.0 (d, 2,6-Ar5C), 126.9
(d, 2,6-Ar2C), 126.6 (s, 5-Py1C), 125.4 (d, 3,5-Ar5C), 125.3 (d, 3,5-Ar1C),
121.3 (d, 5-triazole-C), 118.3 (tF, Py2OCH2CH2CF2), 115.8 (d, 2,3,5,6-
Ar3C, 2,3,5,6-Ar4C), 114.6 (d, 2,6-Ar6C), 108.7 (d, 3,5-Py2C), 73.8 (t,
Ar6CH2O), 73.4 (t, Py2CH2O), 69.0, 68.9 [t, CH2OAr3OCH2(CH2)4CH2OAr4],
68.1 (t, Ar4OCH2CH2CH2-triazole), 67.8 (t, Ar6OCH2), 64.5 [s, C(Ar1)3],
63.0 (t, CO2CH2CH2), 62.2 (t, Py1CH2), 60.9 (t, Py2OCH2CH2CF2), 54.1 [s,
C(Ar5)3], 50.4 [t, triazole-CH2(CH2)4CO2], 40.7 (t, NHCH2), 38.8 (t,
CO2CH2CH2), 34.8, 34.8 [2 s, Ar1C(CH3)3, Ar5C(CH3)3], 34.4 (t, CH2CO2),
31.6 [q, Ar1C(CH3)3, Ar5C(CH3)3], 31.5 (t, Py2OCH2CH2CF2), 30.6 [t, tri-
azole-CH2CH2(CH2)3CO2], 29.9, 29.8 [2 t, OAr3OCH2CH2(CH2)2CH2CH2OAr4],
29.7 (t, Ar4OCH2CH2CH2-triazole), 29.6 [t, NH(CH2)4CH2CH2OAr3],
29.5 (t, NHCH2CH2), 29.2 [t(tF), Ar6OCH2CH2], 27.2 [t,
NH(CH2)2CH2(CH2)3OAr3], 26.5 [t, Ar6O(CH2)3CH2CH2], 26.4, 26.4 [2 t,
NH(CH2)3CH2(CH2)2OAr3, Ar3O(CH2)2CH2CH2(CH2)2OAr4], 26.3 [t,
Ar6O(CH2)2CH2], 24.8 [t, triazole-(CH2)3CH2CH2CO2], 23.3 [t, triazole-
MS (MALDI-TOF, Cl-CCA): m/z = 1670 [M – PF6]+.
Acknowledgment
The support of the Deutsche Forschungsgemeinschaft (SFB 677) is
greatfully acknowledged.
Supporting Information
Supporting information for this article is available online at
http://dx.doi.org/10.1055/s-0034-1380814.
S
u
p
p
o
nrtIo
g
f
rmoaitn
S
u
p
p
ortiInfogrmoaitn
*
(CH2)2CH2(CH2)2CO2], 22.7 (t, Ar4OCH2CH2CH2triazole). Only visible
References
in the HMBC spectra.
(1) Boyer, P. D. Biochim. Biophys. Acta (Bioenergetics) 1993, 1140,
215.
(2) Baudry, J.; Tajkhorshid, E.; Molnar, F.; Phillips, J.; Schulten, J.
J. Phys. Chem. B 2001, 105, 905.
(3) (a) Molecular Machines and Motors, In Topics in Current Chemis-
try; Vol. 354; Credi, A.; Silvi, S.; Venturi, M., Eds.; Springer-
Verlag: Berlin, 2014. (b) Balzani, V.; Credi, A.; Venturi, M. Molec-
ular Devices and Machines - Concepts and Perspectives for the
Nanoworld, 2nd ed.; Wiley-VCH: Weinheim, 2008. (c) Molecular
19F NMR (470 MHz, CD2Cl2): δ = –72.6 (d, 1JP = 711 Hz, 6 F, PF6), –81.1
(s, 3 F, CF3), –113.5 (s, 2 F, CH2CF2), –122.0 (s, 2 F, CH2CF2CF2), –123.0
(s, 2 F, CF2CF3), –123.7 (s, 2 F, CF2CF2CF3), –126.3 (s, 2 F, CH2CF2CF2CF2).
MS (MALDI-TOF, Cl-CCA): m/z = 2522 [M – PF6]+.
HRMS (FT-ICR, CH2Cl2–MeOH): m/z [M
151H184F13N6O122+: 1260.691; found: 1260.698.
– PF6 +
H]2+ calcd for
C
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 2485–2495

2495
O. Beyer et al.
Paper
Synthesis
Switches, 2nd ed.; Feringa, B. L.; Browne, W. R., Eds.; Wiley-VCH:
Weinheim, 2011. (d) Molecular Machines, In Topics in Current
Chemistry; Vol. 262; Kelly, T., Ed.; Springer-Verlag: Berlin, 2005.
(4) Schwörer, M.; Wirz, J. Helv. Chim. Acta 2001, 84, 1441; and ref-
erences cited therein.
(9) Aucagne, V.; Hänni, K. D.; Leigh, D. A.; Lusby, P. J.; Walker, D. B. J.
Am. Chem. Soc. 2006, 128, 2186.
(10) (a) Hesseler, B. Ph.D. Thesis; Christian-Albrechts-Universität zu
Kiel: Germany, 2014. (b) Zindler, M. Ph. D. Thesis; Christian-
Albrechts-Universität zu Kiel: Germany, 2012.
(5) Frej, A.; Goeschen, C.; Näther, C.; Lüning, U.; Herges, R. J. Phys.
Org. Chem. 2010, 23, 1093.
(11) Handbook of Fluorous Chemistry; Gladysz, J. A.; Curran, D. P.;
Horváth, I. T., Eds.; Wiley-VCH: Weinheim, 2004.
(6) Some reviews on MIMs: (a) Molecular Catenanes, Rotaxanes and
Knots – A Journey Through the World of Molecular Topology;
Sauvage, J.-P.; Dietrich-Buchecker, C., Eds.; Wiley-VCH: Wein-
heim, 1999. (b) Stoddart, J. F. Angew. Chem. Int. Ed. 2014, 53,
11102; Angew. Chem. 2014, 126, 11282. (c) Ayme, J.-F.; Beves, J.
E.; Campbell, C. J.; Leigh, D. A. Chem. Soc. Rev. 2013, 42, 1700;
and references cited therein.
(7) (a) Vetter, W.; Schill, G. Tetrahedron 1967, 23, 3079. (b) Schill, G.
Catenanes, Rotaxanes, and Knots; Academic Press: New York,
1971. (c) Schalley, C. A.; Beizai, K.; Vögtle, F. Acc. Chem. Res.
2001, 34, 465. (d) Bruns, C. J.; Stoddart, J. F. Acc. Chem. Res. 2014,
47, 2186.
(12) (a) Eckelmann, J.; Lüning, U. Chem. Unserer Zeit 2009, 43, 210.
(b) Eckelmann, J.; Lüning, U. J. Chem. Educ. 2013, 90, 224.
(13) Horvath, G.; Rusa, C.; Köntös, Z.; Gerencser, J.; Huszthy, P. Synth.
Commun. 1999, 29, 3719.
(14) Chauvin, A.-S.; Comby, S.; Song, B.; Vandevyver, C. D. B.; Bünzli,
J.-C. G. Chem. Eur. J. 2008, 14, 1726.
(15) Mak, E.; Zindler, M.; Hartkopf, B.; Herges, R.; Lüning, U. Eur. J.
Org. Chem. 2010, 4932.
(16) Aucagne, V.; Berná, J.; Crowley, J. D.; Goldup, S. M.; Hänni, K. D.;
Lusby, P. J.; Ronaldson, V. E.; Slawin, A. M.; Viterisi, A.; Walker,
D. B.; Leigh, D. A. J. Am. Chem. Soc. 2007, 129, 11950.
(17) Lamture, J. B.; Zhou, Z. H.; Kumar, A. S.; Wensel, T. G. Inorg.
Chem. 1995, 34, 864.
(8) Hesseler, B.; Zindler, M.; Herges, R.; Lüning, U. Eur. J. Org. Chem.
2014, 3885.
© Georg Thieme Verlag Stuttgart · New York — Synthesis 2015, 47, 2485–2495