Thermodynamic Insights on a Bistable Acid–Base Switchable Molecular Shuttle with Strongly Shifted Co-conformational Equilibria

Article abstract of DOI:10.1002/chem.201604783

Bistable [2]rotaxanes in which the affinities of the two stations can be reversed form the basis of molecular shuttles. Gaining quantitative information on such rotaxanes in which the ring distribution between the two stations is largely nonsymmetric has proven to be very challenging. Herein, we report on two independent experimental methodologies, based on luminescence lifetime measurements and acid–base titrations, to determine the relative populations of the two co-conformations of a [2]rotaxane. The assays yield convergent results and are sensitive enough to measure an equilibrium constant (K≈4000) out of reach for NMR spectroscopy. We also estimate the ring distribution constant in the switched (deprotonated) state (K′<10^?4), and report the highest positional efficiency for stimuli-induced shuttling to date (>99.92 %). Finally, our results show that the pK_aof the pH-responsive station depends on the ring affinity of the pH-insensitive station, an observation that paves the way for the design of new artificial allosteric systems.

Full text of DOI:10.1002/chem.201604783

A Journal of
Accepted Article
Title: Thermodynamic insights on a bistable acid-base switchable
molecular shuttle with strongly shifted co-conformational
equilibria
Authors: Giulio Ragazzon, Alberto Credi, and Benoit Colasson
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: Chem. Eur. J. 10.1002/chem.201604783
Link to VoR: http://dx.doi.org/10.1002/chem.201604783
Supported by

10.1002/chem.201604783
Chemistry - A European Journal
FULL PAPER
Thermodynamic insights on a bistable acid-base switchable
molecular shuttle with strongly shifted co-conformational
equilibria
Giulio Ragazzon,[a] Alberto Credi,*[b] and Benoit Colasson*[a,c]
Dedicated to Professor Vincenzo Balzani, pioneer of molecular machines, on his 80th birthday
Abstract: Bistable [2]rotaxanes in which the affinities of the two
stations can be reversed form the basis of molecular shuttles.
Gaining quantitative information on such rotaxanes in which the ring
distribution between the two stations is largely nonsymmetric has
proven to be very challenging. Here we report on two independent
ability of an external stimulus (chemical, electrochemical or
photonic) to reversibly modify the relative affinity of the ring for
the two stations, leading to a drastic structural change in the
molecule. A simplified view of the process pictures the ring
moving from one station to the other one after the application of
the stimulus (molecular shuttling). This view, however, does not
represent the behavior of a statistically significant population of
rotaxanes, because an equilibrium exists between the two
translational isomers (or co-conformations) corresponding to the
positioning of the ring on either site on the axle.[7] In line with
proposed definitions,[8] the most abundant translational isomer is
referred to as the stable co-conformation (SCC) while the
second isomer is called the metastable co-conformation (MCC).
Hence, a full characterization of the starting state of a
bistable rotaxane - which in turn is important for understanding
successive stimuli-induced switching processes - involves the
measurement of the equilibrium constant K = [SCC]eq/[MCC]eq.
This is a nontrivial task because K is an intramolecular quantity
which reflects relative binding strengths within an individual
molecular entity. Indeed, the free energy difference between the
SCC and MCC can be accurately calculated from the value of K,
experimental methodologies
- based on luminescence lifetime
measurement and acid-base titrations - to determine the relative
populations of the two co-conformations of a [2]rotaxane. The
assays yield convergent results and are sensitive enough to
measure an equilibrium constant (K ~ 4000) out of reach for NMR
spectroscopy. We also estimate the ring distribution constant in the
switched (deprotonated) state (K’<10–4), and report the highest
positional efficiency for stimuli-induced shuttling to date (>99.92%).
Finally, our results show that the pKa of the pH-responsive station
depends on the ring affinity of the pH-insensitive station, an
observation that paves the way for the design of new artificial
allosteric systems.
Introduction
thereby
intercomponent interactions.
enabling
correlations
with
the
noncovalent
Rotaxanes and related interlocked molecules are intriguing
chemical architectures with high relevance for the development
of artificial nanoscale machines and switches as well as
functional materials.[1] The tremendous progresses for their
synthesis have broadened the scope of their use.[2] In particular
bistable [2]rotaxanes, consisting of a macrocycle threaded on a
molecular axle with two potential sites of interaction (the
‘stations’) and stoppered at both ends, can offer a wide range of
functions including molecular transport,[3] ion sensing,[4]
information storage[5] and catalysis.[6]
1H NMR Spectroscopy can be used to measure [SCC]eq and
[MCC]eq but, because of its inherent sensitivity, it provides
reliable results only when
K
does not exceed ca. 20.[9]
Rotaxanes characterized by high values of K, however, are
required to make efficient molecular shuttles. Thus, new assays
that can yield information on bistable rotaxanes with highly
shifted co-conformational equilibria need to be developed.
Paolucci and co-workers used cyclic voltammetric (CV)
experiments to evaluate K.[10] Stoddart and co-workers also
The operation of a bistable rotaxane as a switch relies on the
reported
a method based on CV whose sensitivity was
comparable with that of 1H NMR.[11a] Later, by using variable
scan-rate CV, the sensivity was dramatically increased,[11b,11c]
enabling the measurement of K values up to 104. These
methods, however, can only be applied to rotaxanes having a
well behaved electroactive unit (e.g. a redox-switchable ring
recognition site). For example such a sensitive assay is still
missing for pH-triggered bistable rotaxanes, another important
and widely studied class of switches.[1f]
[a]
[b]
G. Ragazzon, Dr. B. Colasson
Dipartimento di Chimica “G. Ciamician”
Università di Bologna, via Selmi 2, 40126 Bologna (Italy)
Prof. Dr. A. Credi
Dipartimento di Scienze e Tecnologie Agro-alimentari
Università di Bologna, Viale Fanin 44, 40127 Bologna (Italy)
E-mail: alberto.credi@unibo.it
Here we report on an alternative and sensitive methodology
for the quantitative investigation of the co-conformational
equilibria of [2]rotaxanes, and we apply it on an acid-base
switchable molecular shuttle designed to exhibit a largely
nonsymmetric ring distribution between the two stations. As the
only structural requirement of this approach is the presence of a
luminescent moiety in the rotaxane,[12] it can exhibit a general
applicability and is complementary to the strategies discussed
[c]
Dr. B. Colasson
Laboratoire de Chimie et de Biochimie Pharmacologiques et
Toxicologiques (CNRS UMR 8601)
Université Paris Descartes Sorbonne Paris Cité, 45 rue des Saints-
Pères, 75006 Paris (France)
E-mail: benoit.colasson@parisdescartes.fr
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/chem.201604783
Chemistry - A European Journal
FULL PAPER
above.
individual guests representing each station can offer a prediction
of the co-conformational distribution in the rotaxane.[11c] We
therefore measured the intermolecular association constant of
DB24C8 with compounds NI-amm+ and NI-tria+ (Figure 1b).
Since we anticipated a low association constant with the
triazolium axle, the titrations were performed in CH2Cl2, a
solvent expected to enhance the electrostatic and solvophobic
interactions that stabilize the supramolecular complex.[19]
The results are confirmed by another independent
experimental approach based on acid-base titrations of the
rotaxane and appropriate model compounds. This method also
enabled us to estimate the ring distribution constant after the
application of the switching signal (K'), and to determine the
highest efficiency for reversible (forward and backward) stimuli-
induced shuttling reported to date. Another significant outcome
of this investigation is that the acidity of the pH-sensitive site
depends on the affinity of the ring for the pH-insensitive station,
and that the interaction between the two sites is mediated by the
movement of the ring. This kind of behaviour can contribute to
the understanding of allosteric phenomena in biomolecules and
has intriguing implications for the design of artificial molecular
devices.
a)
-
2 PF6
O O
O
N
O
O
N
O
N
H2
N
N
O
O
O
O O
MCC-Rot2s
Results and Discussion
K
-
2 PF6
Design
O O
Recently, the 1,2,3-triazolium unit (tria+) has been identified as a
convenient secondary station for a dibenzo[24]crown-8 ring
(DB24C8) in rotaxanes in which a dialkylammonium unit (amm+)
plays the role of the primary station.[13,14] While the
O
N
O
N
N
O
O
O
N
H2
N
O
O
O O
SCC-Rot2s
intermolecular apparent association constant between
a
b)
dialkylammonium guest and DB24C8[15] is ca. 104–105 M–1 in
CH2Cl2,[16] this value has not yet been reported in the case of a
triazolium axle, presumably because of the very inefficient
complexation. Indeed, it was found that the triazolium moiety
-
PF6
O
O
O
N
O
O
N
O
N
O
O
O
O
N
H2
N
N
O
cannot be used as
a
template for rotaxane formation.[13]
O O
Consequently, the triazolium unit is considered to be a weak
station within a rotaxane. Such a difference in the behavior of
Rot1s
NI
the two guests should ensure
a largely unbalanced co-
-
-
PF6
PF6
conformational distribution, and prompted us to synthesize and
investigate a [2]rotaxane containing ammonium and triazolium
stations in the axle for the development of a sensitive method for
measuring large values of K.
O
N
O
O
N
O
H2
N
N
N
N
Figure 1a shows the bistable rotaxane studied in this work,
Rot2s, and the equilibrium linking its SCC and MCC. A few
model compounds necessary for the experiments are
represented in Figure 1b. The details on the synthesis and
characterization of the compounds are reported in the
Supplementary Information. In particular, Rot1s is structurally
identical to Rot2s except for the absence of the secondary
triazolium station; thus, the only translational co-conformation of
Rot1s is that shown in the figure. The axle in both Rot2s and
NI-amm+
NI-tria+
Figure 1. Structural formulas of (a) Rot2s and the equilibrium between its two
translational isomers, SCC and MCC, and (b) model compounds studied in
this work.
The association between the model axle NI-amm+ and
DB24C8 was monitored by UV-absorption and fluorescence
spectroscopies. Upon addition of an increasing amount of
DB24C8 to a solution of NI-amm+ (ca. 50 µM), the band
centered at 355 nm underwent a hypsochromic shift (Figure
S38). This blue shift is consistent with the loss of an
intramolecular hydrogen bond between the ammonium unit and
the carbonyl groups of NI when the interaction with the ring is
established (vide infra). The addition of DB24C8 also caused a
strong decrease of the fluorescence intensity (Figure 2a). The
quenching of the fluorescence is ascribed to a photoinduced
electron transfer (PET) from the crown ether to the singlet
excited state of the naphthalimide unit.[20] From the analysis of
the spectroscopic titration profiles an apparent association
Rot1s possesses
a fluorescent naphthalimide (NI) stopper
placed in the vicinity of the ammonium station. Encouraged by
the results obtained for many previously reported fluorescent
rotaxanes,[1f,17,18] we anticipated that the fluorescence properties
of the naphthalimide group would be sensitive to the position of
the ring on the axle and could be used to quantitatively report on
its distribution.
Intermolecular association of the ring with individual
stations
As in a rotaxane the two stations compete intramolecularly for
the ring, the stability of the 1:1 complexes of the latter with
This article is protected by copyright. All rights reserved.

10.1002/chem.201604783
Chemistry - A European Journal
FULL PAPER
constant[16] of log(Kamm+) = 5.61 ± 0.15 was determined for the
[NI-amm⊂DB24C8]+ complex.[21] The 1H NMR spectrum of a 1:1
mixture of NI-amm+ and DB24C8 (Figure S39) attests the
formation of the pseudorotaxane complex. The value of the
association constant is in line with those found for similar
ammonium/DB24C8 complexes under similar conditions.[16,22]
The association constant between a triazolium axle and
DB24C8 has never been reported, most likely because of the
low stability of the corresponding complex. Indeed, 1H NMR
experiments performed on ring/axle mixtures at mM
concentrations did not show any appreciable complexation. It is
likely that the formation of ion pairs between the positively
the great difference of affinity for the crown ether between the
ammonium and triazolium stations acknowledged in the
literature.[13]
Determination of the ring distribution from luminescence
measurements
The UV-visible absorption and emission spectra of NI-amm+,
Rot1s and Rot2s in CH3CN are similar and are dominated by the
bands of the NI chromophore (λmax = 354 nm, ε = 14000 M–1 cm–
1). In the case of Rot1s and Rot2s the absorption band of the
DB24C8 component (λmax = 275 nm) is also present. All
compounds show a luminescence band centered at ca. 390 nm,
assigned to the NI fluorescence (Figures S49-S50). The relevant
fluorescence data are given in Table 1.
–
charged axle and its PF6 counteranions competes with
threading and, as for the ammonium guest, significantly lowers
the
apparent
binding
constant
at
relatively
high
concentrations.[16]
Table 1. Fluorescence data (quantum yields Φ, fluorescence lifetimes τi and
pre-exponential factors βi) for NI, NI-amm+, Rot1s and Rot2s
.
τi / ps (βi / a.u.) [a]
Φ / % [a]
18.4 ± 1
20.2 ± 1.2
1.0 ± 0.1
1.8 ± 0.2
τ1
1530
τ2
–
τ3
NI
–
NI-amm+
Rot1s
Rot2s
2660
–
–
–
621 (5)
763 (6)
186 (20)
220 (12.3)
2620 (0.11)
[a] In CH3CN at room temperature. All the fluorescence intensity decays are
shown in the SI.
The emission quantum yield of NI-amm+ is very close to that
of the bare naphthalimide model NI. The presence of the
ammonium is nevertheless testified by a longer lifetime (2.66 ns
vs 1.53 ns), a 2 nm bathochromic shift of the absorption band,
and a fluorescence maximum shifted to longer wavelengths
(Figure S49). These observations are consistent with the
presence of an intramolecular hydrogen bond between the
ammonium unit and a carbonyl group of the naphthalimide both
in the ground and excited states.[23] The emission quantum
yields in the two rotaxanes are significantly smaller because of
PET quenching.[20]
Figure 2. Fluorescence spectral changes (left) and corresponding titration
curves (right) in CH2Cl2 at room temperature of (a) NI-amm+ (54 µM, λexc
=
342 nm) and (b) NI-tria+ (53 µM, λexc = 340 nm) by DB24C8. The white
triangles are the experimental points and the dashed line is the curve fitting
obtained according to a 1:1 association model.
The design of the axle NI-tria+ takes advantage of the
efficient quenching of the naphthalimide fluorescence by
DB24C8 mentioned above. Fluorescence titrations performed at
low concentration (ca. 50 µM) in CH2Cl2 enabled us to measure
for the first time the intermolecular association constant between
a triazolium axle and DB24C8. No absorption spectral changes
were noticed upon adding the ring to a solution of NI-tria+. On
the contrary, the NI fluorescence was substantially quenched
(Figure 2b). Fluorescence lifetime measurements on a mixture
of NI-tria+ (100 µM) and DB24C8 (10 mM) ruled out a dynamic
quenching. Two different lifetimes (τ1 = 1.78 ns and τ2 = 637 ps)
were obtained, the longer one being close to that of NI-tria+
alone (τ = 1.64 ns) (Figures S41-S42). The data could be fitted
according to the formation of a 1:1 complex with an apparent
association constant of log(Ktria+) = 1.38 ± 0.03. From the ratio
of the two intermolecular association constants, a value of logK
= 4.23 ± 0.18 can be predicted. This high figure reflects indeed
Insightful
information
about
the
co-conformational
distribution of Rot2s can be obtained from a careful analysis of
the time-correlated fluorescence decays. While NI-amm+
exhibits
a mono-exponential decay (τ = 2.66 ns), the
fluorescence decay of Rot1s is biexponential with τ2 = 621 ps
and τ3 = 186 ps. In analogy with previous studies on the time-
resolved luminescence properties of crown-ether ligands,[24]
these two short lifetimes are tentatively ascribed to the decay of
the NI excited singlet in two different conformations of the ring
placed on the ammonium station.[25] Interestingly, in Rot2s a third
longer lifetime (τ1 = 2.62 ns) similar to the lifetime determined for
the ammonium model NI-amm+ is present, while the two shorter
lifetimes τ2 = 763 ps and τ3 = 220 ps) have values comparable to
those found in Rot1s (Table 1 and Figure 3).[26]
Since the emission decay (~ns) is much faster than the ring
shuttling (~µs-ms),[27,28] the NI fluorophores in the two co-
This article is protected by copyright. All rights reserved.

10.1002/chem.201604783
Chemistry - A European Journal
FULL PAPER
conformations can be treated as two isolated fluorophores
placed in different environments. By comparison with the
behavior of the models, the lifetime τ1 for Rot2s was therefore
attributed to the MCC, in which the ring is far enough from the
fluorophore to avoid any interaction with the NI excited state.
Consequently we associated τ2 and τ3 with the SCC, in which
the ring surrounds the amm+ station. The remarkable
correspondence between the lifetimes of the studied models and
those found in Rot2s confirms that the two stations can be
considered as spectroscopically independent and are
appropriately described by the reference compounds. With this
requirement satisfied, the pre-exponential factors βi represent
the fractions of SCC (αSCC) and MCC (αMCC), weighted for a
factor F which takes into account the inherent photophysical
differences (e.g. fast static quenching, radiative rates, absorption
and emission spectra) of the NI fluorophore in the two
situations.[29] The F factor can be measured with time-correlated
fluorescence decay experiments by comparing the initial photon
count accumulated during the same time under identical
instrumental conditions for two solutions of Rot1s and NI-amm+
having the same concentration (Figure S47). From the ratio
between the initial photon counts for NI-amm+ and Rot1s a value
F = 9.82 ± 0.04 was determined. Hence, K can be obtained from
equation (1) where βSCC = β2 + β3 and βMCC = β1:
intermolecular association constants. This discrepancy can arise
from the fact that NI-tria+ is not a perfectly accurate model for
the triazolium station in Rot2s. Another explanation is related to
the different solvents employed in the two methods. The choice
of CH2Cl2 for the intermolecular association experiments was
dictated by the need of enhancing the stability of the [NI-
tria⊂DB24C8]+ complex. On the other hand the luminescence
behavior observed in CH3CN was quite rich and, as discussed
above, contained interesting structural information.
Time-resolved fluorescent measurements were also
performed in CH2Cl2, where the emission of Rot2s exhibited a
biexponential decay with τ2 = 781 ps and τ3 = 219 ps (Figure
S48). In line with the previous discussion, these two components
are assigned to two different arrangements of the ring in the
SCC. The absence of a third decay component with longer
lifetime (τ1), attributable to the metastable translational isomer,
indicates that in this particular case the fluorescence-based
method is not sensitive enough to detect the MCC in CH2Cl2.
Such an observation points to a significantly higher value for K,
in agreement with the prediction based on intermolecular
association data obtained in the same solvent. Presumably, in
CH3CN the interaction of the ring with the amm+ station is
affected to a higher extent than with the tria+ one.
Acid-base titrations
K = αSCC/αMCC = F×(βSCC/βMCC
)
(1)
As discussed above, the quantification of the co-conformational
equilibrium with the luminescence-based method requires the
knowledge of the fluorophore photophysics and
a careful
analysis of its luminescence decay. We therefore set up a more
practical assay, which takes advantage of the acid-base
properties of Rot2s and is based on two titration experiments.
The operation of Rot2s as a pH-controlled molecular shuttle
is shown in Scheme 1, wherein co-conformational and acid-base
equilibria are represented as the horizontal and vertical
processes, respectively. The reversible stimuli-induced shuttling
is based on the fact that the SCC and MCC for Rot2s and its
deprotonated counterpart Rot'2s correspond to opposite
translational isomers. In fact, for Rot'2s only the isomer in which
the ring encircles the tria+ station was observed by 1H NMR
spectroscopy, in keeping with the fact that the amm+ site is
switched off upon deprotonation (Figure S40).[13,14]
Considering that deprotonation of SCC-Rot2s affords
immediately and quantitatively SCC-Rot'2s, the titration of Rot2s
with a base enables the determination of an apparent acidity
constant Ka,Rot linking these two forms (Scheme 1). On the other
hand, it is reasonable to assume that NI-amm+ is an appropriate
model for measuring the acidity constant of the ammonium
station of Rot2s when it is not surrounded by the ring (Ka,Mod). A
thermodynamic cycle can thus be established between the two
co-conformations of the protonated rotaxane (Rot2s) and the
deprotonated rotaxane (Rot'2s), as shown by the full arrows in
Scheme 1. The translational equilibrium constant K is then
straightforwardly calculated from equation (2):
Figure 3. Fluorescence intensity decays in CH3CN at room temperature for
Rot1s (red) and Rot2s (blue) and their respective residuals. The green curves
are the two best fit for both decays and correspond to the parameters listed in
Table 1. Conditions: λexc = 340 nm, λem = 415 nm, c = 10 µM.
By introducing the experimental values in equation (1) an
equilibrium constant logK = 3.2 ± 0.2 can be determined. This
value is independent on the wavelength of observation.
The K value measured with this method is one order of
magnitude lower than that calculated from the ratio of the
K = Ka,Mod/Ka,Rot
(2)
This article is protected by copyright. All rights reserved.

10.1002/chem.201604783
Chemistry - A European Journal
FULL PAPER
-
2 PF6
-
O
O
2 PF6
O
O
O
O
O
N
O
N
N
O
H2
N
O
N
O
O
O
N
N
K
N
H2
N
O
N
O
+ B
+ B
O
O
O
O O
O O
MCC-Rot2s
SCC-Rot2s
Ka,Rot
Ka,Mod
-
-
PF6
PF6
O
O
O
O
K'
O
O
O
N
O
N
O
N
O
O
N
N
O
H
H
N
O
N
N
N
O
N
O
O
O
O O
O
O O
SCC-Rot'2s
MCC-Rot'2s
-
-
+ (BH+)(PF6 )
+ (BH+)(PF6 )
Scheme 1. Square scheme describing the switching between different co-conformations (horizontal processes) and protonated/deprotonated forms (vertical
processes) for rotaxane Rot2s. The base employed in the titration and its conjugated acid are labelled respectively as B and BH+. The thermodynamic cycle
linking K and the acidity constants Ka,Rot and Ka,Mod is identified by the full arrows (valid for B = CH3CN). The translational isomer shown in grey is not observed
experimentally. See the text for more details.
In order to perform an accurate assay, the pKa of the bases
used for the titration of the rotaxane and the axle model should
be as close as possible to the pKa of the molecules under
investigation.
A
search for bases compatible with our
experiments (that is, operating in CH3CN and spectroscopically
silent) and known pKa was thus performed. The analysis of the
absorption spectral changes observed upon titration of Rot2s by
Et3N (pKa = 18.82)[30] yielded to a value of pKa,Rot = 19.9 ± 0.1
(Figure 4a). When the titration was performed on NI-amm+ using
a weaker base (PhCH2NH2, pKa = 16.91)[30], a value of pKa,Mod
=
16.3 ± 0.1 was obtained (Figure 4b). By introducing these values
in equation (2), it was calculated that logK = 3.6 ± 0.2. Within
errors, this number is the same as that determined with the
luminescence-based method.
The titration of Rot1s with a base can provide information on
the acid-base properties of the amm+ site when it is surrounded
by DB24C8 (vertical process in the right hand part of Scheme 1).
Indeed, no spectroscopic modification could be observed when
Rot1s was treated with
a
huge excess of 1,4-
18.2).[31] This
Figure 4. UV-absorption spectral changes (left) and corresponding titration
curves (right) in CH3CN at room temperature of (a) Rot2s (29 µM) by Et3N and
(b) NI-amm+ (56 µM) by PhCH2NH2. The white triangles are the experimental
points and the dashed line is the curve fitting according to a proton exchange
model.
diazabicyclo[2.2.2]octane (DABCO, pKa
=
observation indicates that the pKa of the amm+ unit encircled by
DB24C8 is larger than 23.9.[32]
This article is protected by copyright. All rights reserved.

10.1002/chem.201604783
Chemistry - A European Journal
FULL PAPER
The free energy values associated with the processes
base or acid in an appropriate excess), the overall yield of the
device in a full shuttling cycle (base-induced forward and acid-
induced backward movements) amounts to 99.92% (see the SI).
This is indeed the highest yield reported to date for a stimuli-
controlled molecular shuttle. It is worthwhile to note that a
rotaxane capable of performing efficient shuttling cycles requires
that the co-conformational equilibria in both states (in the
present case, the protonated and deprotonated ones) are
strongly shifted towards opposite co-conformations (Scheme 1);
in other words, the system should be designed such that K is
very large while K' is very small.
displayed in Scheme
1
can be calculated from the
corresponding equilibrium constants (ΔG° = –RT lnK) and an
energy-level diagram describing the acid-base induced co-
conformational switching of Rot2s can be constructed (Figure 5).
The levels shown in Figure 5 correspond to absolute (SCC) or
relative (MCC) energy minima; the energy maxima (i.e., the
barriers) that separate the two minima in each curve are not
known in the present case. However, these values could be
measured from kinetic experiments performed by, e.g., stopped-
flow spectrometry, as it was demonstrated for a related pH-
responsive bistable rotaxane.[27] Indeed, an accurate knowledge
of such an energy diagram implies that the switching properties
of the device are fully characterized in terms of both
thermodynamics and kinetics.
The energy-level diagram shown in Figure
5 is also
interesting because it clearly shows that the acid strength of the
pH-switchable site is affected by the position of the ring.
Previous experiments indicate that deprotonation of the
ammonium unit is extremely difficult to achieve in the absence of
an alternative station for the crown ether macrocycle.[9b, 14l, 33-35]
The data collected on the present rotaxane show that the acidity
of its ammonium unit changes by at least 7.6 pKa units upon
switching the ring position.
As a matter of fact, by adjusting the relative stabilities of the
SCC and MCC in the protonated and deprotonated forms, one
can modulate the acid-base properties of the pH-sensitive
station. In principle, such a result could even be achieved by
tuning the ring affinity of the sole pH-insensitive station. This
observation is conceptually very important because it enables
the possibility of designing acid or basic sites with made-to-order
pKa using the very same functional group (e.g. a secondary
amine as in the present case). In other words, the acid-base
properties of the site are determined by its dynamic molecular
environment – a behavior which resembles that exhibited by
many enzymes.[36]
Conclusions
The distribution between the two translational isomers (co-
conformations) in a bistable [2]rotaxane containing DB24C8 as
the ring, and ammonium and triazolium recognition sites in its
axle, has been quantified with two novel and sensitive
methodologies. The first one relies on time-correlated
fluorescence measurements and the second one is based on
appropriately designed acid-base titrations. The great difference
of affinity of the ring for the two stations could be measured in
the interlocked compound (logK = 3.6 in CH3CN). So far, K could
be determined by NMR spectroscopy experiments in case of low
values (up to ca. 20) or with more sensitive CV-based methods
but only for redox active switches.[10,11]
Figure 5. Simplified energy-level diagram as a function of the relative ring-
axle position for the rotaxane in its protonated (bottom) and deprotonated (top)
forms in CH3CN (the energies are given in kJ/mol for B = CH3CN).
On the basis of the energy levels shown in Figure 5 we can
evaluate the distribution constant in the deprotonated state to be
K' = [MCC-Rot'2s]eq/[SCC-Rot'2s]eq < 10–4. The knowledge of the
The K values measured from the acid-base and the time-
correlated fluorescence assays are the same within errors.
While the lifetime-based assay requires a fluorescent rotaxane
in which the emissive excited state of the fluorophore is
influenced by to the position of the ring, the acid-base approach
seems more general because in principle it can be applied to
any pH-switchable rotaxane. The sensitivity of the fluorescence
assay strongly depends on the spectroscopic differences of the
fluorophore in the two different co-conformations. For instance,
values of
K and K' enables us to make quantitative
considerations about the shuttling efficiency of the rotaxane. A
constant K = 4000 means that 99.97% of the rings in a
population of rotaxanes are located on the primary station
(amm+). On the other hand, with a value of K' lower than 10–4,
more than 99.99% of the rings reside on the tria+ station after
deprotonation of the ammonium site. Hence, in the hypothesis
that the acid-base switching reactions are performed
quantitatively (which can be done by adding a sufficiently strong
This article is protected by copyright. All rights reserved.

10.1002/chem.201604783
Chemistry - A European Journal
FULL PAPER
Crowley, S. M. Goldup, A. L. Lee, D. A. Leigh, R. T. McBurney, Chem.
Soc. Rev. 2009, 38, 1530−1541; c) D. Thibeault, J. F. Morin, Molecules
2010, 15, 3709−3730; d) K. D. Hanni, D. A. Leigh, Chem. Soc. Rev.
2010, 39, 1240−1251; e) J. E. Beves, B. A. Blight, C. J. Campbell, D. A.
Leigh, R. T. McBurney, Angew. Chem. Int. Ed. 2011, 50, 9260−9327; f)
R. J. Bordoli, S. M. Goldup, J. Am. Chem. Soc. 2014, 136, 4817−4820.
a) J. Li, Y. Li, Y. Guo, J. Xu, J. Lv, Y. Li, H. Liu, S. Wang, D. Zhu, Chem.
Asian J. 2008, 3, 2091−2096; b) C. Schäfer, G. Ragazzon, B. Colasson,
M. La Rosa, S. Silvi, A. Credi, Chemistry Open 2016, 5, 120−124.
a) S. S. Zhu, P. J. Carroll, T. M. Zwager, J. Am. Chem. Soc. 1996, 118,
8713−8714; b) N. H. Evans, P. D. Beer, Angew. Chem. Int. Ed. 2014,
53, 11716-11754; c) M. J. Langton, P. D. Beer, Acc. Chem. Res. 2014,
47, 1935−1949.
the higher the quantum yield ratio in favor of the minor
(metastable) co-conformation, the more sensitive the
measurement. Also, the deconvolution of the emission decay is
facilitated if the difference between the lifetimes attributed to the
two co-conformations is significant. For the case reported here,
K values of ~103 with an upper limit estimated at ~104 could be
measured. Last but not least, to gain information on the co-
conformational equilibrium distribution the emission decay
should be faster than the ring shuttling. This requirement is
easily fulfilled with organic fluorophores, whose fluorescence
lifetimes (~ns) are much shorter than typical ring shuttling times
(~µs-ms).[27,28]
On the contrary, the acid-base assay does not rely on the
photophysical properties of the system (as long as there is an
analytical signal to be monitored in the titrations) and its
sensitivity is limited only by experimental considerations – for
instance, the availability of the suitable bases whose pKa values
in the solvent of choice are reported. The results of these
experiments, summarized in Figure 5, not only show that the
investigated rotaxane is a highly efficient molecular shuttle, but
also indicate that the affinity of the (pH-insensitive) tria+ station
for the DB24C8 ring influences the acid-base behavior of the
amm+ station. This observation is highly interesting because, in
principle, it enables the use of programmed allosteric effects to
bring about predetermined acid-base properties in artificial
multicomponent systems.
[3]
[4]
[5]
[6]
J. E. Green, J. W. Choi, A. Boukai, Y. Bunimovich, E. Johnston-
Halperin, E. Delonno, Y. Luo, B. A. Sheriff, K. Xu, Y. S. Shin, H. R.
Tseng, J. F. Stoddart, J. R. Heat, Nature 2007, 445, 414−417.
a) D. A. Leigh, V. Marcos, M. R. Wilson, ACS Catalysis 2014, 4,
4490−4497; b) E. A. Neal, S. M. Goldup, Chem. Commun. 2014, 50,
5128−5142; c) V. Blanco, D. A. Leigh, V. Marcos, Chem. Soc. Rev.
2015, 44, 5341−5370.
[7]
[8]
M. N. Chatterjee, E. R. Kay, D. A. Leigh, J. Am. Chem. Soc. 2006, 128,
4058−4073.
A. H. Flood, A. J. Peters, S. A. Vignon, D. W. Steuerman, H. R. Tseng,
S. Kang, J. R. Heath, J. F. Stoddart, Chem. Eur. J. 2004, 10,
6558−6564.
[9]
a) E. Busseron, C. Romuald, F. Coutrot, Chem. Eur. J. 2010, 16,
10062-10073; b) C. Romuald, E. Busseron, F. Coutrot, J. Org. Chem.
2010, 75, 6516-6531.
[10] a) A. Altieri, F. G. Gatti, E. R. Kay, D. A. Leigh, D. Martel, F. Paolucci, A.
M. Z. Slawin, J. K. Y. Wong, J. Am. Chem. Soc. 2003, 125, 8644-8654;
b) F. Scarel, G. Valenti, S. Gaikwad, M. Marcaccio, F. Paolucci, A.
Mateo-Alonso, Chem. Eur. J. 2012, 18, 14063-14068.
The two methods presented in this work expand the toolbox
for the characterization of the thermodynamic behavior of
bistable interlocked compounds and related species exhibiting a
pH-dependent behavior. The information that can be accessed
with these experimental approaches is essential for a detailed
understanding of the switching mechanism and for assisting the
design of new molecular devices and machines.
[11] a) J. W. Choi, A. H. Flood, D. W. Steuerman, S. Nygaard, A. B.
Braunschweig, N. N. P. Moonen, B. W. Laursen, Y. Luo, E. Delonno, A.
J. Peters, J. O. Jeppesen, K. Xe, J. F. Stoddart, J. R. Heath, Chem. Eur.
J. 2006, 12, 261−279. b) A. C. Fahrenbach, J. C. Barnes, H. Li, D.
Benitez, A. N. Basuray, L. Fang, C. H. Sue, G. Barin, S. K. Dey, W. A.
Goddard III, J. F. Stoddart, Proc. Natl. Acad. Sci. U. S. A. 2011, 108,
20416−20421; c) A. C. Fahrenbach, C. J. Burns, D. Cao, J. F. Stoddart,
Acc. Chem. Res. 2012, 45, 1581−1592.
Acknowledgements
[12] For examples of the application of time-resolved luminescence to the
study of supramolecular dynamics, see e.g.: a) C. Bohne, Chem. Soc.
Rev. 2014, 43, 4037-4050; b) V. Balzani, P. Ceroni, A. Juris,
Photochemistry and Photophysics: Concepts, Research, Applications,
Wiley-VCH, Weinheim, 2014.
B.C. thanks the Centre National de la Recherche Scientifique
(CNRS) and the Université Paris Descartes, Sorbonne Paris
Cité for a “délégation”. This work was supported by the Ministero
dell’Istruzione, dell’Università e della Ricerca (PRIN project
“InfoChem”) and the University of Bologna (FARB project
“SLaMM”).
[13] F. Coutrot, ChemistryOpen 2015, 4, 556−576 and references cited
therein.
[14] a) F. Coutrot, E. Busseron, Chem. Eur. J. 2008, 14, 4784−4787; b) F.
Coutrot, C. Romuald, E. Busseron, Org. Lett. 2008, 10, 3741−3744; c)
C. Romuald, A. Arda, C. Clavel, J. Jimenez-Barbero, F. Coutrot, Chem.
Sci. 2012, 3, 1851−1857; d) V. Blanco, A. Carlone, K. D. Hänni, D. A.
Leigh, B. Lewandowski, Angew. Chem., Int. Ed. 2012, 51, 5166−5169;
e) E. Busseron, F. Coutrot, J. Org. Chem. 2013, 78, 4099−4106; f) Z. J.
Zhang, M. Han, H. Y. Zhang, Y. Liu, Org. Lett. 2013, 15, 1698–1701; g)
S. Chao, C. Romuald, K. Fournel-Marotte, C. Clavel, F. Coutrot, Angew.
Chem. Int. Ed. 2014, 53, 6914−6919; h) V. Blanco, D. A. Leigh, U.
Lewandowska, B. Lewandowski, V. Marcos, J. Am. Chem. Soc. 2014,
136, 15775−15780; i) V. Blanco, D. A. Leigh, V. Marcos, J. A. Morales-
Serna, A. L. Nussbaumer, J. Am. Chem. Soc. 2014, 136, 4905−4908; j)
Z. Meng, C. F. Chen, Chem. Commun. 2015, 51, 8241–8244; k) Z.
Meng, Y. Han, L. N. Wang, J. F. Xiang, S. G. He, C. F. Chen, J. Am.
Chem. Soc. 2015, 137, 9739–9745; l) P. Waelès, B. Riss-Yaw, F.
Coutrot, Chem. Eur. J. 2016, 22, 6837−6845.
Keywords: molecular device • molecular machine • rotaxane •
fluorescence • acidity • supramolecular chemistry • nanoscience
[1]
a) W. R. Browne, B. L. Feringa, Nat. Nanotechnol. 2006, 1, 25-35; b) E.
R. Kay, D. A. Leigh, F. Zerbetto, Angew. Chem. Int. Ed. 2007, 46,
72−191; c) V. Balzani, A. Credi, M. Venturi, Molecular Devices and
Machines – Concepts and Perspectives for the Nanoworld, Wiley-VCH,
Weinheim, 2008; d) A. Coskun, J. M. Spruell, G. Barin, W. R. Dichtel, A.
H. Flood, Y. Y. Botros, J. F. Stoddart, Chem. Soc. Rev. 2012, 41,
4827−4859; e) S. Erbas-Cakmak, D. A. Leigh, C. T. McTernan, A. L.
Nussbaumer, Chem. Rev. 2015, 115, 10081−10206. f) M. Xue, Y. Yang,
X. Chi, X. Yan, F. Huang, Chem. Rev. 2015, 115, 7398−7501.
[2]
a) F. Aricò, J. D. Badjić, S. J. Cantrill, A. H. Flood, K. C. F. Leung, Y.
Liu, J. F. Stoddart, Top. Curr. Chem. 2005, 249, 203−259; b) J. D.
[15] A. G. Kolchinski, D. H. Busch, N. W. Alcock, J. Chem. Soc., Chem.
Commun. 1995, 1289–1291.
This article is protected by copyright. All rights reserved.

10.1002/chem.201604783
Chemistry - A European Journal
FULL PAPER
[16] This value is highly dependent on the concentration because the ionic
strength has a strong effect on the ion pairing in the uncomplexed
thread. We refer here to association constants measured at low
concentrations (ca. 50 µM). For a complete discussion on this effect,
see: a) J. W. Jones, H. W. Gibson, J. Am. Chem. Soc. 2003, 125,
7001−7004; b) H. W. Gibson, J. W. Jones, L. N. Zakharov, A. L.
Rheingold, C. Slebodnick, Chem. Eur. J. 2011, 17, 3192−3206.
[17] G. W. H. Wurpel, A. M. Brouwer, I. H. M. van Stokkum, A. Farran, D. A.
Leigh, J. Am. Chem. Soc. 2001, 123, 11327−11328.
[36] a) T. K. Harris, G. J. Turner, Life 2002, 53, 85−98; b) C. N. Pace, G. R.
Grimsley, J. M. Scholtz, J. Biol. Chem. 2009, 284, 13285−13289; c) H.
N. Motlagh, J. O. Wrabl, J. Li, V. J. Hilser, Nature 2014, 508, 331−339.
[18] a) H. Zhang, J. Hu, D. H. Qu, Org. Lett. 2012, 14, 2334−2337; b) H.
Zhang, B. Zhou, H. Li, D. H. Qu, H. Tian, J. Org. Chem. 2013, 78,
2091−2098.
[19] P. R. Ashton, P. J. Campbell, P. T. Glink, D. Philp, N. Spencer, J. F.
Stoddart, E. J. T. Chrystal, S. Menzer, D. J. Williams, P. A. Tasker,
Angew. Chem. Int. Ed. Engl. 1995, 34, 1865−1869.
[20] Transient absorption spectroscopy experiments performed on Rot1s
and Rot2s, supported by energetic considerations, confirmed that PET
takes place from the macrocyclic ring to the NI singlet excited state.
The full photophysical study of a family of naphthalimide-containing
rotaxanes that includes Rot1s and Rot2s will be reported in due course.
See also: H. Li, J. N. Zhang, W. Zhou, H. Zhang, Q. Zhang, D. H. Qu,
H. Tian, Org. Lett. 2013, 15, 3070–3073.
[21] Data fitting was made with the software SPECFIT, Spectrum Software
Associates, R. A. Binstead, Chapel Hill, 1996.
[22] a) M. Montalti, R. Ballardini, L. Prodi, V. Balzani, J. Chem. Soc. Chem.
Commun. 1996, 2011−2012; b) G. Ragazzon, M. Baroncini, S. Silvi, M.
Venturi, A. Credi, Nat. Nanotechnol. 2015, 10, 70−75.
[23] a) V. Wintgens, P. Valat, J. Kossanyi, L. Biczok, A. Demeter, T. Bérces,
J. Chem. Soc. Faraday Trans. 1994, 90, 411−421; b) D. W. Cho, M.
Fujitsuka, U. C. Yoon, T. Majima, Phys. Chem. Chem. Phys. 2008, 10,
4393−4399.
[24] M. T. Morgan, S. Sumalekshmy, M. Sarwar, H. Beck, S. Crooke, C. J.
Fahrni, J. Phys. Chem. B 2014, 118, 14196–14202.
[25] a) P. R. Ashton, I. Baxter, M. C. T. Fyfe, F. M. Raymo, N. Spencer, J. F.
Stoddart, A. J. P. White, D. J. Williams, J. Am. Chem. Soc. 1998, 120,
2297−2307; b) V. Bleve, C. Schäfer, P. Franchi, S. Silvi, E. Mezzina, A.
Credi, M. Lucarini, ChemistryOpen 2015, 4, 18–21; c) P. Franchi, V.
Bleve, E. Mezzina, C. Schäfer, G. Ragazzon, M. Albertini, D.
Carbonera, A. Credi, M. Di Valentin, M. Lucarini, Chem. Eur. J. 2016,
22, 8745–8750.
[26] A third lifetime component must be added to the exponential function to
afford a reasonable fit of the fluorescence intensity decay. The fitting
with a biexponential function clearly neglects the tail of the decay and
the residuals are not randomly distributed.
[27] S. Garaudée, S. Silvi, M. Venturi, A. Credi, A. H. Flood, J. F. Stoddart,
ChemPhysChem. 2005, 6, 2145−2152.
[28] A. M. Brouwer, C. Frochot, F. G. Gatti, D. A. Leigh, L. Mottier, F.
Paolucci, S. Roffia, G. W. H. Wurpel, Science 2001, 291, 2124−2128.
[29] J. R. Lakowicz in Principles of Fluorescence Spectroscopy, 3rd Ed.,
Springer, Berlin, 2006.
[30] I. Kaljurand, A. Kütt, L. Sooväli, T. Rodima, V. Mäemets, I. Leito, I. A.
Koppel, J. Org. Chem. 2005, 70, 1019−1028.
[31] J. M. Darmon, N. Kumar, E. B. Hulley, C. J. Weiss, S. Raugei, R. M.
Bullock, M. L. Helm, Chem. Sci. 2015, 6, 2737−2745.
[32] The lower limiting pKa value of 23.9 is obtained considering that in
presence of 6000 equivalents of DABCO less than 10% of Rot1s is
deprotonated.
[33] a) N. Kihara, Y. Tachibana, H. Kawasaki, T. Takata, Chem. Lett. 2000,
506−507; b) K. Nakazono, T. Takata, Chem. Eur. J. 2010, 16, 13783–
13794.
[34] D. A. Leigh, A. R. Thomson, Tetrahedron 2008, 64, 8411-8416.
[35] a) J. Cao, M. C. T. Fyfe, J. F. Stoddart, G. R. L. Cousins, P. T. Glink, J.
Org. Chem. 2000, 65, 1937–1946; b) J. D. Badjić, V. Balzani, A. Credi,
J. N. Lowe, S. Silvi, J. F. Stoddart, Chem. Eur. J. 2004, 10, 1926–1935.
This article is protected by copyright. All rights reserved.

10.1002/chem.201604783
Chemistry - A European Journal
FULL PAPER
.
Entry for the Table of Contents
FULL PAPER
In a lifetime: The co-conformational
equilibria of a pH-switchable bistable
[2]rotaxane have been investigated
with two independent and sensitive
assays based on its time-resolved
fluorescence and acid-base
properties. The results show that this
rotaxane exhibits the highest
positional efficiency reported so far for
a molecular shuttle.
Giulio Ragazzon, Alberto Credi,* Benoit
Colasson*
Page No. – Page No.
Thermodynamic insights on a
bistable acid-base switchable
molecular shuttle with strongly
shifted co-conformational equilibria
This article is protected by copyright. All rights reserved.