Photoredox-Switchable Resorcin[4]arene Cavitands: Radical Control of Molecular Gripping Machinery via Hydrogen Bonding

Article abstract of DOI:10.1002/chem.201704788

Semiquinones (SQ) are generated in photosynthetic organisms upon photoinduced electron transfer to quinones (Q). They are stabilized by hydrogen bonding (HB) with the neighboring residues, which alters the properties of the reaction center. We designed, synthesized, and investigated resorcin[4]arene cavitands inspired by this function of SQ in natural photosynthesis. Cavitands were equipped with alternating quinone and quinoxaline walls bearing hydrogen bond donor groups (HBD). Different HBD were analyzed that mimic natural amino acids, such as imidazole and indole, along with their analogues, pyrrole and pyrazole. Pyrroles were identified as the most promising candidates that enabled the cavitands to remain open in the Q state until strengthening of HB upon reduction to the paramagnetic SQ radical anion provided stabilization of the closed form. The SQ state was generated electrochemically and photochemically, whereas properties were studied by UV/Vis spectroelectrochemistry, transient absorption, and EPR spectroscopy. This study demonstrates a photoredox-controlled conformational switch towards a new generation of molecular grippers.

Full text of DOI:10.1002/chem.201704788

DOI: 10.1002/chem.201704788
Full Paper
&
Supramolecular Chemistry
Photoredox-Switchable Resorcin[4]arene Cavitands: Radical
Control of Molecular Gripping Machinery via Hydrogen Bonding
[a]
[b, c]
Darius Talaat,[a] Julia Nomrowski,[d] Nils Trapp,[a]
´
Jovana Milic, Michal Zalibera,
Laurent Ruhlmann,[e] Corinne Boudon,[e] Oliver S. Wenger,*[d] Anton Savitsky,*[c]
Wolfgang Lubitz,*[c] and FranÅois Diederich*[a]
Dedicated to the memory of Donald J. Cram
Abstract: Semiquinones (SQ) are generated in photosyn-
thetic organisms upon photoinduced electron transfer to
quinones (Q). They are stabilized by hydrogen bonding (HB)
with the neighboring residues, which alters the properties of
the reaction center. We designed, synthesized, and investi-
gated resorcin[4]arene cavitands inspired by this function of
SQ in natural photosynthesis. Cavitands were equipped with
alternating quinone and quinoxaline walls bearing hydrogen
bond donor groups (HBD). Different HBD were analyzed that
mimic natural amino acids, such as imidazole and indole,
along with their analogues, pyrrole and pyrazole. Pyrroles
were identified as the most promising candidates that ena-
bled the cavitands to remain open in the Q state until
strengthening of HB upon reduction to the paramagnetic
SQ radical anion provided stabilization of the closed form.
The SQ state was generated electrochemically and photo-
chemically, whereas properties were studied by UV/Vis spec-
troelectrochemistry, transient absorption, and EPR spectros-
copy. This study demonstrates a photoredox-controlled con-
formational switch towards a new generation of molecular
grippers.
are molecular grippers,[7–10] which can be effectively based on
resorcin[4]arene cavitand platforms that feature the ability to
controllably open and close in response to external stimuli,[11]
such as temperature,[7,12] solvent identity,[11,13] or chemical
agents.[7–10,14,15] Such actuation enables these systems to rever-
sibly encapsulate smaller guest molecules and thereby func-
tion as molecular sensors, actuators, or elements in nanorobot-
ics.[10,16–18] Moreover, their capacity to act as elements of multi-
state switches renders them promising components of molecu-
lar-level memory devices.[10,19,20] However, a key requirement
for the application of nanoscale grippers is the usage of physi-
cal stimuli, such as voltage or light, to control their behav-
ior.[6,10,17,21]
Introduction
Miniaturization of technology stimulates engineering of func-
tional nanosystems that integrate molecular elements into
electrical circuits as switches, sensors, actuators, or logic
gates.[1–6] The systems of particular interest to nanotechnology
´
[a] Dr. J. Milic, D. Talaat, Dr. N. Trapp, Prof. F. Diederich
Laboratory of Organic Chemistry, ETH Zurich
Vladimir-Prelog-Weg 3, 8093 Zurich (Switzerland)
E-mail: diederich@org.chem.ethz.ch
[b] Dr. M. Zalibera
Institute of Physical Chemistry and Chemical Physics
Slovak University of Technology, Radlinskꢀho 9, 81237 Bratislava (Slovakia)
E-mail: michal.zalibera@stuba.sk
Exploiting light to operate a molecular machinery is most ef-
fectively demonstrated in natural photosynthesis.[23,24] This
complex biomolecular process involves light harvesting by the
photosynthetic reaction centers, such as the bacterial reaction
center (RC) and photosystems I and II, which enable photoin-
duced electron transfer to take place, while redox-active qui-
nones act as electron acceptors.[22,23,25] The sophistication of
the natural approach lies within its aptitude to tune the molec-
ular machinery by altering noncovalent interactions as a func-
tion of the state of the underlying redox-active components
(Figure 1).[26–30] Accordingly, quinones (Q) are able to engage in
weak noncovalent interactions, whereas their reduction to the
semiquinone (SQ) state increases this ability by several orders
of magnitude, enabling them to act as strong hydrogen bond
acceptors that interact with the neighboring protein residues
[c] Dr. M. Zalibera, Dr. A. Savitsky, Prof. W. Lubitz
Max Planck Institute for Chemical Energy Conversion
Stiftstrasse 34–36, 45470 Mꢁlheim an der Ruhr (Germany)
E-mail: wolfgang.lubitz@cec.mpg.de
anton.savitsky@cec.mpg.de
[d] J. Nomrowski, Prof. O. S. Wenger
Department of Chemistry, University of Basel
St. Johanns-Ring 19, 4056 Basel (Switzerland)
E-mail: oliver.wenger@unibas.ch
[e] Prof. L. Ruhlmann, Prof. C. Boudon
Laboratoire d’ꢂlectrochimie et Chimie Physique du Corps Solide
Institut de Chimie de Strasbourg, Universitꢀ de Strasbourg
4 rue Blaise Pascal, CS 90032, 67081 Strasbourg (France)
E-mail: lruhlmann@unistra.fr
Supporting information and the ORCID identification number for the
author of this article can be found under https://doi.org/10.1002/
chem.201704788.
Chem. Eur. J. 2017, 23, 1 – 11
1
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
mations (open or closed) in three redox states (Q, SQ, and
HQ).[10]
Herein, we present a development of molecular grippers
that can perform as binary conformational switches in aprotic
environments in response to voltage or light inspired by the
functionality of semiquinones in natural photosynthesis. The
grippers are based on resorcin[4]arene cavitand platforms
equipped by two quinone (Q) units analogous to the ubiqui-
nones (QA and QB) in the photosynthetic active site. The plat-
form was designed to comprise two quinoxaline (Qx) walls
bearing hydrogen bond donor groups (HBD) that mimic cer-
tain amino acid residues in the reaction center, such as His and
its analogues of comparable HBD strength.[22,26,43,44] The hydro-
gen bonding sites were devised to be insufficiently strong for
stabilizing the closed form of the gripper in the oxidized Q
state, yet attain this ability in the SQ state to provide the driv-
ing force for molecular gripping (Figure 2). The SQ state was
generated electrochemically, by cyclic voltammetry and photo-
chemically via photoreduction by [Ru(bpy)3]2+ [10]
Pyrrole-based
.
Figure 1. a) Hydrogen bonding of the redox-active ubiquinone units QA and
QB in the photosynthetic reaction center of Rb. sphaeroides, (PDB 1DV3)[22]
with neighboring His residues (N···O distances are 2.5 ꢁ and 2.8 ꢁ, respec-
tively). b) Schematic of the photoredox-switchable molecular gripper whose
closed form is stabilized by the hydrogen bonding (dashed lines) in the sem-
iquinone (SQ) state.
cavitands 1–4 were identified as the most promising candi-
dates for photoredox-controlled gripping (Figure 3a), which
was demonstrated by using naphthoquinone-based cavitands
1 and 2. Further optimization of the gripping properties led to
the development of triptycene-based cavitands 3 and 4. More-
over, in order to identify the design principles and provide
carrying hydrogen bond donors (HBD), such as His (Fig- better understanding of the electronic and conformational
ure 1a).[26,31–36] These redox-controlled noncovalent interactions properties in the SQ state, we also studied cavitands 5–6 that
can trigger conformational changes in the reaction center that did not contain any hydrogen bond donors (Figure 3b), as well
modify the redox environment and further play a role in the as model systems 7–19 that represent single cavitand walls
formation of the chemiosmotic gradient required for the oper- bearing alternative HBD groups (Figure 3c).
ation of the ATP synthase, which ultimately converts the elec-
tromagnetic inputs into chemical energy.[25,37–39]
The utility of this approach to control the photoredox-pow-
ered molecular machinery remains underexploited in the
design of artificial nanodevices.[40–42] As a step towards apply-
ing the redox-dependence of noncovalent interactions in a
functional molecular system, we reported the first redox-
switchable resorcin[4]arene cavitand by installing naphthoqui-
none (NQ) units into its backbone together with alternating
quinoxaline (Qx) walls equipped with carboxamide hydrogen
bond acceptor groups.[18] Upon two-electron chemical reduc-
tion in protic media, the hydroquinone (HQ) was formed
which acted as a hydrogen bond donating site that interacts
Figure 2. Photoredox switching concept of resorcin[4]arene cavitands com-
with the neighboring hydrogen bond acceptors to favor the
closed conformation. However, dependence of this reaction on
the proton source and poor reversibility in the HQ state pre-
vented control of the redox process electrochemically and
thereby limited the further development of this system to-
wards a prospective molecular device.[17] For this purpose, es-
tablishing a system that would be able to perform the binary
switch between an open and a closed form within aprotic en-
prising quinone (Q) and quinoxaline (Qx) walls equipped with hydrogen
bond donating (XH) groups. Such cavitands can act as molecular grippers
by adopting an open conformation in the quinone (Q) state until strength-
ening of hydrogen bonding (dashed lines) provides stabilization for the
closed form in the reduced semiquinone (SQ) state, which can be achieved
both electrochemically and photochemically. D=electron donor; bpy=2,2’-
bipyridine.
vironments in response to electrical or electromagnetic stimuli Results and Discussion
would be of furthermost importance. This objective could be
Molecular design considerations
addressed by relying on our recently developed bis-semiqui-
none-based resorcin[4]arenes, which were shown to exist pre- The target gripper scaffold was constructed based on three
dominantly as stable triplet dianions that can act as paramag- design principles. First, we relied on resorcin[4]arene platforms
netic elements of six-state switches able to adopt two confor- that were suitable for studying molecular gripping due to their
&
&
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
2
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Figure 4. Molecular design considerations for photoredox-switchable molec-
ular grippers: a) Optimized geometries (DFT B3LYP/6-31G(d))[10, 18] of 2 in the
open form of the oxidized Q and closed form of the reduced SQ state,
which is stabilized by hydrogen bonding (dashed line) between the HBD (in
this case, pyrrole) and the carbonyl groups. Formation of the closed form re-
quires rotation of the HBD group away from the competitive HB with the
Qx wall (black arrows) towards the Q wall by a dihedral angle q. b) Confor-
mational analysis (DFT B3LYP/6-31G(d))[18] of the quinoxaline walls equipped
with HBD groups (Im, Pyr, Pyz, and Ind) of different propensity to form hy-
drogen bonds with the neighboring walls. This propensity is reflected in the
height of the energy barrier for changing the conformational minima (at
q=08) to the one with feasible hydrogen bonding (area marked in gray
with N···O distances between 2.5–3.5 ꢁ at q=35–658).[18] HBD=hydrogen
bond donor, Im=imidazole, Pyr=pyrrole, Pyz=pyrazole, Ind=indole.
Figure 3. a) Target cavitands 1–4, b) reference systems 5–6, and c) model
compounds 7–19 employed in this study.
ability to adopt two conformations, an open and a closed
one.[11,45] The platform was equipped with quinone (Q) and qui-
noxaline (Qx) walls carrying HBD groups (XH, Figure 2). While
naphthoquinone walls (NQ) were studied in order to provide
stabilization of the SQ state through charge delocalization
within the bicyclic p-system, the triptycene-quinones (TQ)
were employed with the aim of enhancing the binding proper-
ties (Figure 3).[17,18] Furthermore, in order to ensure that grip-
ping occurs only in the SQ state, as a secondary design princi-
ple we considered HBD groups of moderate strength that are
not interacting strongly with the Q walls, yet rather engage in
competitive HB with the Qx walls (Figure 4a). We focused our
attention on donors with comparable strength to those of the
natural amino acid residues that engage in hydrogen bonding
within the photosynthetic reaction centers (His or Trp),[22,26,43,44]
such as imidazole (Im) and indole (Ind). We also considered
their analogues of comparable HB strength to Im (pKa =18.6)
and Ind (pKa =21), such as pyrazole (Pyz, pKa =19.8) and pyr-
role (Pyr, pKa =23) (Figure 4b).[46] The phenolic OH group of Tyr
(pKa =10), as well as other stronger amide HBD groups present
in the reaction centers,[46] could contribute to stabilizing the
SQ state in the photosynthetic protein complexes while also
engaging in HB with the oxidized Q state, which was not desir-
able for photoredox-switchable grippers.
tive HB with the Qx wall, in order to be able to stabilize the
closed form of the gripper (Figure 4a). This conformational
change by a dihedral angle q occurs with a certain energy bar-
rier that depends on the nature of the HBD group. Lowering
this barrier leads to preorganized systems that can more favor-
ably adopt the closed conformation. DFT calculations of the
conformational preferences for the Qx walls at the B3LYP/6-
31G(d) level of theory[10,18] suggested that this barrier is higher
in case of Im-containing systems, moderate for those of Pyr
and Pyz, and the lowest in case of Ind (Figure 4b). This was in
line with the previous studies of Im-based cavitands that were
shown to favor aggregation due to propensity towards inter-
molecular HB, which prevents formation of the closed confor-
mation.[18] This aggregation tendency renders Im analogues un-
suitable in the design of resorcin[4]arene-based molecular grip-
pers. These studies demonstrated that the impact of the Im
group on the geometry in solution was significant despite
minor differences in the calculated energy barriers compared
to other wall substituents.[18] Similarly, greater steric demand
for Ind-containing walls (for more details, refer to Section S8 in
the Supporting Information) imposes a potential obstacle for
their application in redox-switching. We therefore identified
Pyr- and Pyz-derived systems as more promising candidates
for photoredox-switchable grippers.
Finally, in addition to the HBD strength, an important third
criterion that was considered in the design was the conforma-
tional compatibility of the HBD-carrying Qx wall with the
closed geometry of the gripper.[17,18] This criterion arises from
the requirement of the HBD group to incline towards the car-
bonyl group of the redox-active wall, away from the competi-
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
3
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Synthesis
mediate 25,[47] which was accessed through selective iodina-
tion of 2,3-dichloroquinoxaline (Scheme 1A). This strategy pro-
vided a versatile synthetic approach for accessing a variety of
quinoxaline walls by installing HBD groups in the final synthet-
ic step (for more details, see Section S1 in the Supporting In-
formation), which significantly improved the method previous-
ly undertaken in functionalizing quinoxaline walls that relied
on benzothiadiazole intermediates,[8,18] such as 26. The report-
ed approach[8,18] only offered traces of the target precursor
owing to the sensitivity of the HBD groups to these conditions,
proving the strategy developed in the course of this study
more effective. Quinone-based resorcin[4]arene cavitand plat-
forms 27–31 (Scheme 1B)[10,18,48] were reacted with the corre-
sponding walls under basic conditions via nucleophilic substi-
tution at ambient temperature to offer the target cavitands 1
and 3, as well as reference systems 5–6 (Scheme 1B; for de-
tails, see Section S1 in the Supporting Information). Acid-medi-
ated deprotection yielded cavitands 2 and 4, whose properties
in the SQ state were further investigated.
Model systems comprising single cavitand walls 7–11 and 18–
19[10] (Figure 3c) were prepared via nucleophilic substitution of
2,3-dichloroquinone intermediates 20–21 (for structures and
details, refer to Supporting Information, Section S1). Synthesis
of models 12–17 was based on Suzuki cross-coupling with qui-
noxaline intermediates 22–23 (see the Supporting Information,
Section S1) and the corresponding N-protected HBD group
precursors. Deprotection of the N-Boc, N-THP and OCH3
groups in the final synthetic step offered model systems for
studying hydrogen bonding in the SQ state. The possibility to
isolate both HBD-protected models and their unprotected de-
rivatives bearing HBD groups was key to identifying signatures
of HB in the SQ states. This requirement also rendered systems
based on Pyr groups more suitable as compared to their Pyz
or Ind analogues, since the deprotection of both Pyz and Ind
groups occurred more easily, already under the cross-coupling
reaction conditions.
Wall precursor 24 was prepared by Suzuki cross-coupling
with the functionalized HBD groups and quinoxaline inter-
Scheme 1. Synthesis of cavitands 1–6: a) 1. LiTMP, THF, ꢀ788C, 30 min; 2. TMSCl, THF, ꢀ788Cꢀ258C, 1 h; 3. ICl, THF, 08C, 1 h, 25 (75%); b) [Pd(PPh3)4], K2CO3,
PhMe, THF, H2O, 958C, 48 h; 24 (79%); c) [Pd(PPh3)4], K2CO3, PhMe, THF, H2O, 958C, 72 h; 26 (79%); structures of 24 and 26 in the crystal. Solvent molecules
are omitted for clarity (for details, refer to Section S3 in the Supporting Information); d) HCl, EtOH, 0–708C, 16 h, 27 (63%); e) 20, K2CO3, DMSO, 508C, 16 h, 28
(75%); f) 21, K2CO3, DMSO, 508C, 16 h, 29 (54%); g) catechol, CsF, DMF, 708C, 40 min, 30 (43%); h) 2-hydroxypyridine, Cs2CO3, DMF, 708C, 40 min, 31 (29%);
i) Cs2CO3, DMF, 508C, 12 h, 1 (21%); j) 1. Cs2CO3, DMF, 508C, 12 h, 3 (39%); k) Cs2CO3, THF, 708C, 12 h, 5 (33%); l) Cs2CO3, THF, 708C, 12 h, 6 (89%; synthetic ap-
proach detailed in Section S1 in the Supporting Information); m) TFA, CH2Cl2, 0–258C, 12 h, 2 (91%); n) TMSOTf, 2,6-lutidine, CH2Cl2, 258C, 2 h, 4 (59%). Boc=
tert-butyloxycarbonyl; TMP=2,2,6,6-tetramethylpiperidide; THF=tetrahydrofuran; Ph=phenyl; Me=methyl; DMSO=dimethyl sulfoxide; DMF=dimethylfor-
mamide; TFA=trifluoroacetic acid; TMSOTf=trimethylsilyl trifluoromethanesulfonate.
&
&
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
4
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Semiquinone radical anion formation
shown by rotating disc voltammetry (RDV). The second quasi-
reversible reduction wave was attributed to formation of the
quinone-dianion Q2ꢀ (Figure 5a; for details see Section S4,
Supporting Information).[10,51] UV/Vis spectroelectrochemistry
upon electrochemical reduction in the domain of the first re-
duction wave revealed absorption in the range between 450
and 550 nm, which was an indication of the SQ state formation
(Figure 5b).[10,52,53] This spectroscopic signature was also identi-
fied by UV/Vis transient absorption spectroscopy upon photo-
chemical reduction (Figure 5c). Finally, the paramagnetic SQ
radical anions generated by chemical and photochemical re-
duction featured characteristic septet CW EPR spectra originat-
ing from the interaction of the unpaired electrons with nuclear
spins of two pairs of equivalent protons. The observed hyper-
fine coupling constants (hfcs) aH(2H)=0.069 mT and aH(2H)=
0.038 mT shown in the example of model 7 (Figure 5d) result
from the delocalization of the spin density within the naphtho-
quinone (NQ) moiety. The hfc values were in agreement with
Cꢀ[54–57]
The reduction of the Q to the SQ states was performed:
1) electrochemically, via cyclic voltammetry, or 2) photochemi-
cally by using [Ru(bpy)3]2+ as a photocatalyst in presence of
Et3N as sacrificial electron donor (for more details, see Sec-
tion S5 in the Supporting Information). Chemical reduction by
using CoCp2 was also performed for comparison, with the aim
of identifying the spectroscopic signatures in the SQ state.[10]
The formation of the SQ states was monitored by UV/Vis spec-
troelectrochemistry, transient absorption, as well as EPR spec-
troscopy of model systems and prospective grippers (Figure 5;
for more details, see the Supporting Information).[10,49,50]
Cyclic voltammograms showed SQ formation after the first
reversible reduction step that occurred in the domain between
ꢀ1.0 and ꢀ1.2 V (vs. Fc+/Fc as an internal standard) in CH2Cl2.
This step corresponds to a one-electron reduction for model
systems, and two-electron reduction of cavitand systems, as
the previous studies of NQ
or its biological ana-
logues.[54,57,58] The EPR spectra of radical anions generated
both chemically and photochemically were superimposable,
confirming the formation of the desired species under all re-
duction conditions (Figure 5c; for more details see the Sup-
porting Information).
Hydrogen bonding with the semiquinone radicals
Control of the photoredox gripping machinery required iden-
tification of the hydrogen bonding (HB) signatures upon re-
duction to the SQ state and evaluation of its impact on the
properties of the system. Studies of the reduced ubiquinones
in the photosynthetic reaction center[22,32,35,43,59,60] as well as
systems that mimic their photochemistry[39,61] attracted consid-
erable attention over the past decade, and HB in the SQ state
1
was analyzed by cyclic voltammetry and H ENDOR spectrosco-
py.[22,59,62] We therefore investigated model compounds 7–19
that represent single gripper walls (Figure 3) by means of
these methods in order to categorize the indicators of HB and
analyze their influence on redox, magnetic, and photophysical
properties in the SQ state.
Figure 5. Spectroscopic signatures of SQ radical anion state formation in the
example of model 7 (0.5–1 mm): a) Cyclic voltammogram (CV) recorded in
CH2Cl2 at ambient temperature using scanning rate of 100 mVsꢀ1. The first
reversible reduction step occurring between ꢀ1.0 V and ꢀ1.2 V (vs. Fc+/Fc
as an internal standard) corresponds to the SQ formation. b) UV/Vis spectra
upon electrochemical reduction in CH2Cl2 at ambient temperature. Inset
shows the corresponding CV at 3 mVsꢀ1, and the potentials where the color-
highlighted spectra were taken are indicated by equally colored circles. The
UV/Vis absorption in the 450–550 nm range was indicative of the SQ radical
anion formation. c) UV/Vis transient absorption spectra obtained from [Ru(b-
py)3](PF6)2 (35 mm) solution in degassed CH2Cl2 under inert Ar atmosphere in
the presence of Et3N (0.1m) after a time delay of 2 ms upon excitation at
532 nm with laser pulses of ꢁ10 ns width. The absorption in the 450–
550 nm domain corresponds to the same spectroscopic indicators of the SQ
state as observed by spectroelectrochemistry, with slightly reduced intensity
around 450 nm owing to the bleaching of the photocatalyst (for more de-
tails, refer to the Supporting Information). The crystal structure of model 7 is
presented in the inset. d) CW EPR spectra of the SQ radical anions obtained
by reduction (top) with CoCp2 in CH2Cl2 at ambient temperature, and
(bottom) photoreduction with [Ru(bpy)3](PF6)2 and Et3N at 420 nm during
15 min irradiation in the RPR200 Rayonet photoreactor with 16 lamps of
8 W. Black lines show the experimental spectra and red lines the simulations
using spin Hamiltonian parameters giso =2.0048 and hfcs aH(2H)=0.069 mT,
aH (2H)=0.038 mT, assigned to protons of the NQ moiety. The spectra gen-
erated chemically and photochemically are superimposable, which indicates
the formation of identical paramagnetic species (for more details for other
systems, refer to the Supporting Information).
The impact of HB on the redox properties in the SQ state
was evident by cyclic voltammetry (CV). This effect was reflect-
ed in the shifts of the reduction potentials to more positive
values to an extent that was dependent on the HB
strength.[31,63] The observed potential shifts indicative of stabili-
zation by HB were even more evident by squarewave voltam-
metry (SWV), as a more sensitive electrochemical method.[64,65]
CV and SWV of models 7–11 (Figure 6a) revealed two distinct
indicators of HB in the SQ state by comparing models compris-
ing protected HBD groups (10–11) to their unprotected ana-
logues bearing hydroxyl (OH) groups (8–9). The key indicator
of HB was the shift of the first reduction potential to more
positive values that suggested stabilization of the SQ in pres-
ence of interacting HBD groups. Another indicator was the loss
of reversibility of the second reduction wave that was likely to
occur due to protonation of the Q2ꢀ in presence of HBD
groups.[31,63] The potential shift was shown to be more pro-
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
5
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
reversibility even for the first reduction step, followed by sec-
ondary reactions, rendered the Ind group unsuitable in the
design of switchable molecular grippers (for more details, see
Section S4 in the Supporting Information). Furthermore, slight-
ly more pronounced shifts of the first reduction potential for
the Pyr-containing model 13 as compared to the Pyz-based
system 15 suggested Pyr moieties to be better suited, presum-
ably owing to the existence of the pyrazole moiety in two tau-
tomeric forms, only one of which features an NH group that
can participate in the HB (for details, refer to Sections S1 and
S4 in the Supporting Information). This finding further empha-
sized the fitness of Pyr-based systems as compared to their
Pyz analogues. Shortening of N···O distances in the SQ state of
Pyr-based model 13 (from 4.9 ꢁ in the Q to 3.8 ꢁ in the SQ
state, Figure 7) suggested by DFT calculations revealed its pro-
pensity for weak HB that can be accounted for changes of the
properties in the SQ state.
Figure 6. a) Cyclic voltammograms of models 7–11 (~0.5 mm) recorded in a
standard three-electrode electrochemical cell with Pt working electrode in
CH2Cl2 at ambient temperatures using scanning rates of 100 mVsꢀ1. Shifts of
the first redox potential to more positive values in presence of proximal
HBD (dashed red line) were indicative of stabilization via HB in the SQ
state.[31,63] Loss of reversibility of the second reduction waves (red arrows)
compared to semi-reversible reduction for protected HBD groups (blue
arrows) indicated presence of HBD groups. b) Squarewave voltammograms
(SWV) of models 12–15 indicate shifts of the first redox potential to more
positive values (dashed red line) and loss of reversibility of the second re-
duction potential (red arrows) due to stabilization via intramolecular hydro-
gen bonding in the SQ state. SWV was recorded in a standard three-elec-
trode electrochemical cell with GC working electrode in CH2Cl2 at ambient
temperatures using 25 mV pulse amplitude, 30 Hz pulse frequency, and
5 mV step increments (for details, see Section S4 in the Supporting Informa-
tion).
In addition to changes of the redox properties owing to HB,
a direct indicator of noncovalent interactions upon redox inter-
conversion from the Q to the SQ radical anion state can be
found in the alteration of their magnetic properties. The inter-
actions of the unpaired electrons in the SQ states with the
1
neighboring HBD groups can be probed by EPR and H ENDOR
spectroscopy. Extensive EPR studies of biologically relevant qui-
nones in the past[32,66–69] revealed that HB affects the spin den-
sity distribution in the SQ radical and results in a decrease of
nounced in case of proximal HBD groups (as for model 8) than
those that were distant from the carbonyl group and could
thus not engage in intramolecular interactions with the SQ
(such as for model 9). The loss of reversibility of the second re-
duction wave suggested presence of the HBD groups in solu-
tion even in cases where it was not participating in intramolec-
ular interactions with the SQ. An example of such case is
model 9, which contains distant OH groups that could not par-
ticipate in HB for geometrical reasons. CV and SWV of 9 fea-
tured loss of reversibility of the second reduction wave that
was not accompanied by shifts of the first reduction potential,
unlike in case of model 8. As presence of HBD groups did not
induce any shifts of the first reduction potential due to inabili-
ty to engage in intramolecular HB, this experiment excluded
the influence of intermolecular HB on the shifts of the redox
potential for the range of concentrations employed in the
study. The key indicators of HB and their effect on the redox
properties were dependent on the HB strength, hence they
were more pronounced for systems bearing stronger OH
groups that were proximal to the carbonyl group in the SQ
state. These indicators were nevertheless apparent even for
model systems 12–17 equipped with moderate HBD groups of
interest in the design of molecular grippers (Figure 6b). CV and
SWV of models 13 and 14 revealed stabilization of their corre-
sponding SQ states through interaction with the HBD groups,
which were found to be reversible and thus suitable for further
application (Figure 6b). In case of Ind-based system 17, loss of
Figure 7. Optimized geometries of model 13 in the: a) oxidized Q, and b) re-
duced SQ state obtained by DFT calculations at B3LYP 6-31G(d) level of
theory, highlighting conformational changes driven by hydrogen bonding,
as reflected in the N···O distances and NꢀH···O angles in the SQ state (N···O
distances alter from 4.9 ꢁ in the Q to 3.8 ꢁ in the SQ state). The geometry
reveals absence of HB in the Q, and presence of very weak HB in the SQ
1
state. The field dependence of the H Davies Q-band (34 GHz) ENDOR spec-
tra of: c) 12·ꢀ, and d) 13·ꢀ (~0.5 mm) in CD2Cl2/CD3CN (5:1, v/v) by using the
tinv-tRF-T-tp-t-2tp-t-echo sequence (tinv =200 ns, tRF =20 ms, tp =32 ns, T=5 ms,
t=500 ns at 80 K). The blue dotted arrow indicates the signals due to hy-
perfine coupling with the hydrogens involved in the intramolecular hydro-
gen bonding (HB). The top panels in (c) and (d) show the corresponding
first derivative of echo-detected EPR spectra. The singularities corresponding
to g-tensor principal values are marked. The black arrow in the top panel of
(d) indicates the shift of the gx value characteristic for the HB (for details, see
Section S5 in the Supporting Information).
&
&
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
the anisotropy of the g tensor, when compared to the non-
bonded state. This is mainly reflected in the shift of the gx prin-
cipal value closer to the free electron ge (2.0023), since the g
tensor along x-axes coincides with the direction of the CꢀO
bonds participating in the formation of the hydrogen bond
(for more details, see Section S5 in the Supporting Informa-
tion).[70] This is apparent from the EPR spectra of reduced
models 12 and 13 (top panel, Figure 7c–d; for more informa-
tion, see Section S5 in the Supporting Information) where a
shift of the gx value for the Pyr-containing model 13·ꢀ indicates
HB formation in the SQ state.
of model 15 with free HBD). The photoreduction rate of the
cavitands was shown to increase as compared to model sys-
tems up to 109 sꢀ1 mꢀ1 for reference cavitands 5–6. These rates
are comparable to those at which electron transfer inside
active sites of photosynthetic biological systems takes place,
which was estimated to occur within the 107ꢀ109 sꢀ1 mꢀ1
range.[22,75] Interestingly, similar to the biological systems, the
distances between the two paramagnetic SQ sites inside 5–6
(10 ꢁ in the closed and 13 ꢁ in the open form)[10] were found
to correspond to the orders observed between the QA and QB
ubiquinone units in the reaction center (e.g. 17 ꢁ for Rhodo-
bacter sph.).[22,35,36,76] This geometry could facilitate dipole-
dipole interactions between the two radical anions and further
influence the electronic properties of the system. In accord-
ance with the impact of noncovalent interactions in biological
systems, multiple sites of weak noncovalent interactions can
lead to a more profound effect on the molecular machinery.[77]
A more direct evidence for HB can be obtained from
1H ENDOR spectra due to appearance of an additional hyper-
fine coupling (hfc) between the unpaired electrons and the
corresponding hydrogen nuclei.[22,35,43,44,60] This through-space
interaction of a predominantly dipolar nature is accessible only
1
in the solid state spectra. The Davies H ENDOR spectra of the
SQ radical anions of HBD-protected models 12 and 14 in
frozen solutions (Figure 7c; for more details, see Section S5 in
the Supporting Information) comprise signals arising from the
hfc of the skeletal protons of the backbone, which fall into the
jAj <4 MHz range (Figure 7c shows spectral singularities < ꢂ
2 MHz). However, the spectra of the models comprising free
Photoredox-controlled molecular actuation and gripping
The conformational switch upon redox interconversion of cavi-
tands containing HBD groups was suggested by DFT calcula-
tions (Figure 8, inset; for more details, refer to Section S8 in
the Supporting Information). The optimized geometry (DFT
B3LYP/6-31G(d))[10] of 2 in the oxidized Q state showed that
Pyr groups engage in competitive HB with the nitrogen atoms
of the proximal Qx walls in both conformations of the cavi-
tand. Reduction to the SQ state, however, altered the orienta-
tion of the Pyr groups towards the carbonyl groups as stronger
hydrogen bond acceptors (HBA), which ensured stabilization of
the closed conformation via HB. Conformational changes of re-
sorcin[4]arene cavitands are experimentally typically studied by
1H NMR spectroscopy.[12,17] The signals that are most indicative
of the conformational changes are the methine protons posi-
tioned underneath the walls, as their downfield shift from the
area between 3.5–4.5 ppm towards 5.5–6.5 ppm indicates for-
mation of the closed conformation.[12,17] 1H NMR spectra of Pyr-
containing cavitands 1 and 2 in CDCl3 at ambient temperatures
revealed the preference for the open form in the Q state for
both analogues, in accordance with their design (for more de-
tails, see Section S2 in the Supporting Information). NMR spec-
troscopy however could not be applied to studying the geom-
etry of the paramagnetic SQ intermediates, since the presence
of unpaired electrons leads to broadening and disappearance
of proton resonances.[10]
Cꢀ
Cꢀ
HBD groups, such as 13 (Figure 7d) and 15 (see Figure S62,
Section S5, Supporting Information) reveal additional contribu-
tions from the hfc with the HBD protons at jAj >4 MHz (Fig-
ure 7d shows spectral singularities in the ꢂ2–3 MHz range; for
more details, see Section S5 in the Supporting Information),
which confirm stabilization by weak intramolecular HB in the
Cꢀ
paramagnetic SQ state. Moreover, the ENDOR spectrum of 13
is orientation-selective and suggests that the geometry of the
HB is distorted when compared to the intermolecular HB com-
plexes of SQ with protic solvent molecules (e.g. water or alco-
hols),[66,70,71] as featuring the bridged proton out of the qui-
none ring plane (for details, see Section S5 in the Supporting
Information). This is in line with the DFT calculated model (Fig-
ure 7b) and similar to the ubiquinone in the bacterial RC.[72]
Finally, stabilization of the SQ states via HB also affected
photophysical properties of model compounds and the corre-
sponding cavitands.[73,74] This influence was reflected in the in-
crease of the corresponding rates of photoreduction as re-
vealed by Stern–Volmer analysis (for more details, see Sec-
tion S6.3 in the Supporting Information). Models carrying
strong HBD groups, such as 9, demonstrated rates as high as
the diffusion-controlled limit of the order of 1010 sꢀ1 mꢀ1.[73,74]
This rate was two orders of magnitude higher than the one
observed for model 7, where in absence of HBD groups photo-
reduction was found to take place at the rate of 6.7ꢂ
108 sꢀ1 mꢀ1. Other systems equipped with moderate HBD
groups were found to be photoreduced with the rates in the
order of 106 sꢀ1 mꢀ1, owing to the lowering of the reduction
potential upon substitution with heteroatom-containing aro-
matic rings or oxygen, which decreased the driving force for
photoreduction. Nevertheless, even the presence of weak HB,
as for Pyr or Pyz, increased this rate as compared to their un-
protected analogues approximately two times (e.g. from 1.3ꢂ
106 sꢀ1 mꢀ1 for N-protected model 14 to 3ꢂ106 sꢀ1 mꢀ1 in case
Alternatively, UV/Vis spectroelectrochemistry was demon-
strated to be a powerful method to assess the conformational
properties of resorcin[4]arene cavitands in the paramagnetic
SQ state.[10] This approach revealed that the conformational
changes are indicated by the Qx wall absorption between
300–350 nm, which was verified by the analysis of model
system 18 and reference systems 5 and 6. It was shown that a
~15–20 nm hypso-hypochromic shift of the absorbance in the
area between 300–350 nm accompanies the gripping motion,
1
which was confirmed by both H NMR (in the Q state) and EPR
spectroscopy (in the SQ state).[10] We therefore monitored the
conformational changes of Pyr-containing cavitands 1 and 2
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
7
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
compared to its N-Boc protected analogue 1, suggesting stabi-
lization of the SQ via intramolecular HB. This was also indicat-
1
ed by H ENDOR spectroscopy, which revealed an additional
hyperfine coupling between the unpaired electron of cavitand
2 in the reduced SQ radical anion state and the proton nuclei
of the Pyr HBD group (for details see Section S5, Supporting In-
formation), which supported the conformational switch to the
closed form (Figure 8c). The stabilization was further reflected
in the increase of the rates of the photoreduction for the Pyr-
containing cavitand 2 (4.0ꢂ106 mꢀ1 sꢀ1) as compared to the N-
s
Boc-protected analogue 1 (2.0ꢂ106 mꢀ1 ꢀ1; for more details,
see Section S6 in the Supporting Information), which was in ac-
cordance with the studies of the model systems 12–13 that ex-
hibit weak HB interactions in the SQ state.
Conformational features of molecular grippers are known to
affect their binding properties.[10,17] Since binding occurs in the
closed conformation, systems that favor this form are preor-
ganized for encapsulating smaller guest molecules and expect-
ed to present stronger binding affinities. In order to illustrate
this effect, we probed binding of cavitand 4, which features
triptycene-quinone walls that contribute to the enhancement
of the complexation properties.[10,18] Moreover, the size of the
cavity matches well small six-membered-ring molecules, as
demonstrated by the crystal structure of the reference system
6 encapsulating a 1,4-dithiane guest molecule.[10,18] We thereby
explored the binding affinity of 4 towards this guest (for more
details see Section S7, Supporting Information) in order to illus-
trate the gripping abilities. The binding of 1,4-dithiane by 4 in
CD2Cl2 as a competitive solvent at ambient temperature
(298 K) was found to be very weak (Ka <1mꢀ1) in the oxidized
Q state. However, the binding constant of 4 for 1,4-dithiane in
mesitylene-d12, as a commonly employed noncompetitive sol-
vent,[10,18] was higher in the Q state (Ka =2.10ꢂ103 mꢀ1) by an
order of magnitude compared to the reference system 6[10,18]
without any HBD group (Ka =1.27ꢂ102 mꢀ1). The binding was
further enhanced in the reduced SQ state (from Ka =3.99ꢂ
102 mꢀ1 for 6[10,18] to Ka >2.10ꢂ103 mꢀ1 for 4) even in CH2Cl2 as
a competitive solvent, which was reflected in a positive reduc-
tion potential shifts in presence of the guest molecule (for de-
tails, refer to Section S7 in the Supporting Information). This
was in line with the preorganization of a closed conformation
in the SQ state stabilized by HB, which illustrates the ability of
these systems to act as switchable molecular grippers. With
the proof-of-concept in hand, further optimizations by employ-
ing stronger, more versatile and robust HBD groups promise
attractive perspectives for photoredox-switchable grippers, as
well as other functional nanosystems based on this bioinspired
concept.
Figure 8. UV/Vis spectra of: a) N-Boc protected cavitand 1 (0.5–1 mm), and
b) its Pyr-containing analogue 2 (0.5–1 mm) recorded upon electrochemical
reduction at ꢀ1.1 V (vs. Fc+/Fc) with schematic representations in the open
(top) and the closed (bottom) form. Absorbance in the area between 300–
350 nm corresponds to quinoxaline (Qx) absorption[10] and reveals conforma-
tional changes: hypso-hypochromic shift of the Qx absorption for 2 is in ac-
cordance with the conformational change from the open to the closed
form, while absence of this shift in case of 1 results from its inability to
adopt the closed conformation. Insets represent optimized geometries (DFT
B3LYP/6-31G(d)) of 1 in the Q (open form, a) state, and 2 in the SQ (closed
form, b) state. c) Squarewave voltammograms of 1 (top) and 2 (bottom) re-
corded in CH2Cl2 (~0.5 mm) using 25 mV pulse amplitude, 30 Hz pulse fre-
quency, and 5 mV step increments (for more details, see Section S4 in the
Supporting Information). Shift of the first reduction potential to more posi-
tive values (dashed red line) indicates stabilization by HB in the SQ state, as
1
confirmed by H ENDOR spectroscopy (for details, see Section S5 in the Sup-
porting Information). Electrochemical reduction was performed in a standard
three-electrode cell with GC working electrode in CH2Cl2 at ambient temper-
ature (for more details, see Section S4 in the Supporting Information). Inset
represents optimized geometries (DFT B3LYP/6-31G(d)) of 2 in the Q (open
form, top) and SQ (closed form, bottom) states.
upon redox interconversion to the SQ state by spectroelectro-
chemistry. UV/Vis spectra upon electrochemical reduction of 1
and 2 in CH2Cl2 at ambient temperature revealed the SQ for-
mation through appearance of the absorption with a maxi-
mum between 400–500 nm. Moreover, Pyr-containing cavitand
2 exhibited a hypo-hypsochromic shift of the Qx absorption in
the 300–350 nm domain upon reduction to the SQ state,
which indicated that the conformational change took place
(Figure 8b). These changes were not evident for the N-protect-
ed analogue 1 (Figure 8a) in absence of HBD groups. This in-
ability to adopt a closed conformation was also complemented
by steric reasons imposed by the bulky Boc groups, as sug-
1
gested by H NMR spectroscopy in the Q state and the crystal
structures of N-Boc-protected precursors 24 and 26 (Scheme 1;
for more details, refer to Sections S2–S3 in the Supporting In-
formation).
Conclusions
Furthermore, cavitands equipped with HBD groups could
undergo stabilization of the redox-active sites via HB through
interaction with the neighboring walls only in the closed form,
thereby squarewave voltammetry could serve as an indicator
of the “conformational lock” by HB upon redox interconver-
sion. Voltammograms of Pyr-containing cavitand 2 displayed
shifts of the first reduction potential to more positive values as
This study presents a systematic approach in the design, syn-
thesis, and investigation of photoredox-switchable resorci-
n[4]arene cavitands that can act as molecular grippers con-
trolled by voltage or light, which were inspired by the utility of
semiquinones in natural photosynthesis. We designed resorci-
n[4]arene cavitands equipped with quinone walls that mimic
&
&
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
8
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
the ubiquinone environment of the photosynthetic reaction
center, where the enhanced hydrogen bonding to the neigh-
boring residues in the reduced semiquinone state mediates
the molecular machinery. Model systems that represent cavi-
tand walls were equipped with different HBD groups that
mimic certain amino acid residues in the reaction center and
their analogues, such as indole, pyrazole, and pyrrole. Their
study ascertained the signatures of hydrogen bonding and its
impact on electronic, magnetic, and photophysical properties
in the SQ radical anion state generated electrochemically (by
cyclic voltammetry) and photochemically (upon photoreduc-
tion by [Ru(bpy)3]2+). This analysis identified pyrrole-based sys-
tems as the most promising candidates for demonstrating
photoredox-switchable molecular gripping. The redox inter-
conversion of pyrrole-containing cavitands to the SQ radical
anions was evidenced by UV/Vis spectroelectrochemistry, tran-
sient absorption, and EPR spectroscopy, which resulted in the
conformational change accompanied by the enhancement of
the binding properties. Our findings provide a novel approach
in the design of photoredox-switchable resorcin[4]arene cavi-
tands that sets the stage for the development of molecular
gripping devices in the future.
Acknowledgements
This work was supported by the Swiss National Science Foun-
dation (SNF) Grant 200020-159802, the Slovak Grant Agency
VEGA (Contract No. 1/0416/17), the Slovak Research and Devel-
opment Agency (Contract No. APVV-15-0053), the Cluster of
Excellence RESOLV (EXC 1069) funded by the Deutsche For-
schungsgemeinschaft, and Max Planck Society. The authors are
grateful to Christoph Laurich and Petra Hçfer (Max Planck Insti-
tute for Chemical Energy Conversion) for technical assistance,
Dr. Martin Kuss-Petermann (University of Basel) for helpful dis-
cussions, Dr. Michael Solar (ETH Zurich) for performing the X-
ray crystallographic analysis, Dr. Mark Olivier Ebert, Renꢃ
Arnold, and Stephen Burkhardt (ETH Zurich) for help in NMR
spectroscopy, Thomas Schneeberger (ETH Zurich) for recording
the IR spectra, and Dr. Bruno Bernet and Cꢃdric Schaack (both
at ETH Zurich) for proofreading the manuscript. J.M. acknowl-
edges the fellowship from the Dositeja Fund for Young Talents.
Conflict of interest
The authors declare no conflict of interest.
Keywords: host–guest chemistry
· hydrogen bonding ·
Experimental Section
molecular grippers · photoredox switch · supramolecular
chemistry
General information
[1] V. Balzani, M. Gꢄmez-Lꢄpez, J. F. Stoddart, Acc. Chem. Res. 1998, 31,
405–414.
[2] H. J. Shepherd, I. A. Gural’skiy, C. M. Quintero, S. Tricard, L. Salmon, G.
Molnꢅr, A. Bousseksou, Nat. Commun. 2013, 4, 2607–2615.
[3] S. Erbas-Cakmak, D. A. Leigh, C. T. McTernan, A. L. Nussbaumer, Chem.
Rev. 2015, 115, 10081–10206.
[4] J. M. Abendroth, O. S. Bushuyev, P. S. Weiss, C. J. Barrett, ACS Nano
2015, 9, 7746–7768.
[5] J. L. Zhang, J. Q. Zhong, J. D. Lin, W. P. Hu, K. Wu, G. Q. Xu, A. T. S. Wee,
W. Chen, Chem. Soc. Rev. 2015, 44, 2998–3022.
Reference compounds 5–6, model systems 7 and 18–19, as well as
precursors 25 and 27–32 were prepared according to the literature
procedures.[8,10,18,47,48] Synthesis of wall precursors 24 and 26, 33–
34, cavitands 1–4, and model compounds 8–17 was established by
employing procedures reported in the Supporting information,
which further provides the following information: characterization
of the compounds, NMR spectroscopy, crystal structure data, elec-
trochemistry, EPR spectroscopy, UV/Vis spectroscopy, spectroelec-
trochemistry, time-resolved luminescence measurements, transient
absorption spectroscopy, binding studies, and DFT calculations.
[6] D. Xiang, X. Wang, C. Jia, T. Lee, X. Guo, Chem. Rev. 2016, 116, 4318–
4440.
[7] Y. Yamakoshi, R. R. Schlittler, J. K. Gimzewski, F. Diederich, J. Mater.
Chem. 2001, 11, 2895–2897.
[8] I. Pochorovski, M.-O. Ebert, J.-P. Gisselbrecht, C. Boudon, W. B. Schweizer,
F. Diederich, J. Am. Chem. Soc. 2012, 134, 14702–14705.
[9] F. Bertani, N. Riboni, F. Bianchi, G. Brancatelli, E. S. Sterner, R. Pinalli, S.
Geremia, T. M. Swager, E. Dalcanale, Chem. Eur. J. 2016, 22, 3312–3319.
Preparation of the SQ species
The preparation of the SQ state was performed electrochemically,
photochemically, and by a chemical reduction using submillimolar
concentrations of the quinone substrates, as indicated in the cap-
tions of the corresponding figures. Electrochemical reduction was
achieved using either cyclic voltammetry (CV) or square wave vol-
tammetry (SWV) in CH2Cl2 containing 0.1m nBu4NBF4 within a clas-
sical three-electrode cell equipped with a working electrode (WE),
the counter electrode (CE), and the reference electrode (RE). The
glassy carbon (GC) disc was used as the WE, the CE was a Pt wire,
and an Ag wire served as a pseudoreference electrode. Photore-
duction was performed by using 0.1 mol% [Ru(bipy)3](PF6)2 in the
presence of 0.1m Et3N at 420 nm for 15 min in either a Rayonet
RPR200 photoreactor with 16 lamps (8 W) or using a Quantel Bril-
liant Nd3+:YAG laser for excitation at 532 nm with a pulse duration
of ~10 ns. Chemical reduction involved addition of CoCp2 to the
degassed CH2Cl2 solutions. The samples were transferred into the
corresponding spectrometers immediately after the reduction. Fur-
ther details are provided in the Supporting Information.
´
[10] J. Milic, M. Zalibera, I. Pochorovski, N. Trapp, J. Nomrowski, D. Neshcha-
din, L. Ruhlmann, C. Boudon, O. S. Wenger, A. Savitsky, W. Lubitz, G. Ge-
scheidt, F. Diederich, J. Phys. Chem. Lett. 2016, 7, 2470–2477.
[11] J. R. Moran, S. Karbach, D. J. Cram, J. Am. Chem. Soc. 1982, 104, 5826–
5828.
[12] V. A. Azov, B. Jaun, F. Diederich, Helv. Chim. Acta 2004, 87, 449–462.
[13] P. Roncucci, L. Pirondini, G. Paderni, C. Massera, E. Dalcanale, V. A. Azov,
F. Diederich, Chem. Eur. J. 2006, 12, 4775–4784.
[14] P. J. Skinner, A. G. Cheetham, A. Beeby, V. Gramlich, F. Diederich, Helv.
Chim. Acta 2001, 84, 2146–2153.
[15] M. Frei, F. Marotti, F. O. Diederich, Chem. Commun. 2004, 1362–1363.
[16] R. J. Hooley, J. Rebek, Jr., Chem. Biol. 2009, 16, 255–264.
[17] I. Pochorovski, F. Diederich, Acc. Chem. Res. 2014, 47, 2096–2105.
[18] I. Pochorovski, J. Milic, D. Kolarski, C. Gropp, W. B. Schweizer, F. Dieder-
ich, J. Am. Chem. Soc. 2014, 136, 3852–3858.
[19] G. de Ruiter, L. Motiei, J. Choudhury, N. Oded, M. E. Van der Boom,
Angew. Chem. Int. Ed. 2010, 49, 4780–4783; Angew. Chem. 2010, 122,
4890–4893.
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
9
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
[20] S. Pookpanratana, H. Zhu, E. G. Bittle, S. N. Natoli, T. Ren, C. A. Richter, Q.
Li, C. A. Hacker, J. Phys. Condens. Matter 2016, 28, 094009.
[21] T. A. Su, M. Neupane, M. L. Steigerwald, L. Venkataraman, C. Nuckolls,
Nat. Rev. Mater. 2016, 1, 16002.
[51] M. T. Huynh, C. W. Anson, A. C. Cavell, S. S. Stahl, S. Hammes-Schiffer, J.
Am. Chem. Soc. 2016, 138, 15903–15910.
[52] N. Vꢆn Anh, R. M. Williams, Photochem. Photobiol. Sci. 2012, 11, 957–
961.
[22] R. Calvo, E. C. Abresch, R. Bittl, G. Feher, W. Hofbauer, R. A. Isaacson, W.
Lubitz, M. Y. Okamura, M. L. Paddock, J. Am. Chem. Soc. 2000, 122,
7327–7341.
[53] T. N. V. Karsili, D. Tuna, J. Ehrmaier, W. Domcke, Phys. Chem. Chem. Phys.
2015, 17, 32183–32193.
[54] M. R. Das, H. D. Connor, D. S. Leniart, J. H. Freed, J. Am. Chem. Soc.
1970, 92, 2258–2268.
[23] J. Barber, Q. Rev. Biophys. 2003, 36, 71–89.
[24] S. Eberhard, G. Finazzi, F.-A. Wollman, Annu. Rev. Genet. 2008, 42, 463–
515.
[55] N. M. Atherton, A. J. Blackhurst, J. Chem. Soc. Faraday Trans. 2 1972, 68,
470–475.
[25] O. Yehezkeli, R. Tel-Vered, J. Wasserman, A. Trifonov, D. Michaeli, R. Ne-
chushtai, I. Willner, Nat. Commun. 2012, 3, 742.
[26] Y. N. Pushkar, J. H. Golbeck, D. Stehlik, H. Zimmermann, J. Phys. Chem. B
2004, 108, 9439–9448.
[27] N. Srinivasan, R. Chatterjee, S. Milikisiyants, J. H. Golbeck, K. V. Lakshmi,
Biochemistry 2011, 50, 3495–3501.
[56] P. Ashworth, W. T. Dixon, J. Chem. Soc. Perkin Trans. 2 1974, 739–744.
[57] B. Epel, J. Niklas, S. Sinnecker, H. Zimmermann, W. Lubitz, J. Phys. Chem.
B 2006, 110, 11549–11560.
[58] C. Teutloff, R. Bittl, W. Lubitz, Appl. Magn. Reson. 2004, 26, 5–21.
[59] W. O. Feikema, P. Gast, I. B. Klenina, I. I. Proskuryakov, Biochim. Biophys.
Acta Bioenerg. 2005, 1709, 105–112.
[28] A. K. Turek, D. J. Hardee, A. M. Ullman, D. G. Nocera, E. N. Jacobsen,
Angew. Chem. Int. Ed. 2016, 55, 539–544; Angew. Chem. 2016, 128,
549–554.
[29] S. Khan, J. S. Sun, G. W. Brudvig, J. Phys. Chem. B 2015, 119, 7722–7728.
[30] S. Saha, G. N. Sastry, J. Phys. Chem. B 2015, 119, 11121–11135.
[31] N. Gupta, H. Linschitz, J. Am. Chem. Soc. 1997, 119, 6384–6391.
[32] M. Flores, R. Isaacson, E. Abresch, R. Calvo, W. Lubitz, G. Feher, Biophys.
J. 2006, 90, 3356–3362.
[60] E. Martin, A. Baldansuren, T.-J. Lin, R. I. Samoilova, C. A. Wraight, S. A. Di-
kanov, P. J. O’Malley, Biochemistry 2012, 51, 9086–9093.
[61] J. D. Megiatto, Jr., D. D. Mꢃndez-Hernꢅndez, M. E. Tejeda-Ferrari, A.-L.
Teillout, M. J. Llansola-Portolꢃs, G. Kodis, O. G. Poluektov, T. Rajh, V.
Mujica, T. L. Groy, D. Gust, T. A. Moore, A. L. Moore, Nat. Chem. 2014, 6,
423–428.
[62] M. Shanmugam, G. Xue, L. Que, Jr., B. M. Hoffman, Inorg. Chem. 2012,
51, 10080–10082.
[33] K. S. Feldman, D. K. Hester, II, J. H. Golbeck, Bioorg. Med. Chem. Lett.
2007, 17, 4891–4894.
[63] Y. Ge, L. Miller, T. Ouimet, D. K. Smith, J. Org. Chem. 2000, 65, 8831–
8838.
[34] J. Hankache, D. Hanss, O. S. Wenger, J. Phys. Chem. A 2012, 116, 3347–
3358.
[35] A. T. Taguchi, P. J. O’Malley, C. A. Wraight, S. A. Dikanov, J. Phys. Chem. B
2015, 119, 5805–5814.
[36] Y. Kato, R. Nagao, T. Noguchi, Proc. Natl. Acad. Sci. USA 2016, 113, 620–
625.
[37] P. D. Boyer, Annu. Rev. Biochem. 1997, 66, 717–749.
[38] M. A. Watson, S. L. Cockroft, Chem. Soc. Rev. 2016, 45, 6118–6129.
[39] L. Favereau, A. Makhal, Y. Pellegrin, E. Blart, J. Petersson, E. Gçransson,
L. Hammarstrçm, F. Odobel, J. Am. Chem. Soc. 2016, 138, 3752–3760.
[40] V. Goulle, A. Harriman, J.-M. Lehn, J. Chem. Soc. Chem. Commun. 1993,
1034–1308.
[64] A. H. Flood, A. E. Kaifer in Supramolecular Chemistry: From Molecules to
Nanomaterials, Vol 2. (Eds.: P. A. Gale, J. W. Steed), Wiley, Chichester,
2012, pp. 451–472.
[65] P. L. Boulas, M. Gꢄmez-Kaifer, L. Echegoyen, Angew. Chem. Int. Ed. 1998,
37, 216–247; Angew. Chem. 1998, 110, 226–258.
[66] O. Burghaus, M. Plato, M. Rohrer, K. Mçbius, F. MacMillan, W. Lubitz, J.
Phys. Chem. 1993, 97, 7639–7647.
[67] O. Nimz, F. Lendzian, C. Boullais, W. Lubitz, Appl. Magn. Reson. 1998, 14,
255–274.
[68] P. J. Van Dam, J.-P. Willems, P. P. Schmidt, S. Pçtsch, A.-L. Barra, W. R.
Hagen, B. M. Hoffman, K. K. Andersson, A. Grꢇslund, J. Am. Chem. Soc.
1998, 120, 5080–5085.
[41] 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.
[42] 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.
[43] E. Martin, R. I. Samoilova, K. V. Narasimhulu, T.-J. Lin, P. J. O’Malley, C. A.
Wraight, S. A. Dikanov, J. Am. Chem. Soc. 2011, 133, 5525–5537.
[44] N. Cox, M. Retegan, F. Neese, D. A. Pantazis, A. Boussac, W. Lubitz, Sci-
ence 2014, 345, 804–808.
[69] C. Teutloff, W. Hofbauer, S. G. Zech, M. Stein, R. Bittl, W. Lubitz, Appl.
Magn. Reson. 2001, 21, 363–379.
[70] S. Sinnecker, E. Reijerse, F. Neese, W. Lubitz, J. Am. Chem. Soc. 2004, 126,
3280–3290.
[71] M. Flores, R. A. Isaacson, R. Calvo, G. Feher, W. Lubitz, Chem. Phys. 2003,
294, 401–413.
[72] M. Flores, R. Isaacson, E. Abresch, R. Calvo, W. Lubitz, G. Feher, Biophys.
J. 2007, 92, 671–682.
[45] V. A. Azov, A. Beeby, M. Cacciarini, A. G. Cheetham, F. Diederich, M. Frei,
J. K. Gimzewski, V. Gramlich, B. Hecht, B. Jaun, T. Latychevskaia, A. Lieb,
Y. Lill, F. Marotti, A. Schlegel, R. R. Schlittler, P. J. Skinner, P. Seiler, Y. Ya-
makoshi, Adv. Funct. Mater. 2006, 16, 147–156.
[46] C. Laurence, M. Berthelot, Perspect. Drug Discovery Des. 2000, 18, 39–
60.
[73] O. S. Wenger, Acc. Chem. Res. 2011, 44, 25–35.
[74] J. Hankache, O. S. Wenger, Phys. Chem. Chem. Phys. 2012, 14, 2685–
2692.
[75] C. C. Page, C. C. Moser, X. Chen, P. L. Dutton, Nature 1999, 402, 47–52.
[76] A. T. Taguchi, P. J. O’Malley, C. A. Wraight, S. A. Dikanov, J. Phys. Chem. B
2014, 118, 9225–9237.
[47] J. Nafe, S. Herbert, F. Auras, K. Karaghiosoff, T. Bein, P. Knochel, Chem.
Eur. J. 2015, 21, 1102–1107.
[77] E. Persch, O. Dumele, F. Diederich, Angew. Chem. Int. Ed. 2015, 54,
3290–3327; Angew. Chem. 2015, 127, 3341–3382.
[48] I. Pochorovski, C. Boudon, J.-P. Gisselbrecht, M.-O. Ebert, W. B. Schweizer,
F. Diederich, Angew. Chem. Int. Ed. 2012, 51, 262–266; Angew. Chem.
2012, 124, 269–273.
Manuscript received: October 10, 2017
[49] W. Kaim, J. Fiedler, Chem. Soc. Rev. 2009, 38, 3373–3782.
[50] J. Bonin, M. Routier, Artif. Photosynth. 2013, 1, 6–15.
Version of record online: && &&, 0000
&
&
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
10
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
FULL PAPER
&
Supramolecular Chemistry
´
J. Milic, M. Zalibera, D. Talaat,
J. Nomrowski, N. Trapp, L. Ruhlmann,
C. Boudon, O. S. Wenger,* A. Savitsky,*
W. Lubitz,* F. Diederich*
&& – &&
Quintessential: Quinone-based resorci-
n[4]arene cavitands were prepared as
bioinspired photoredox switches. Upon
reduction to the paramagnetic bis(semi-
quinone) radical anions, they undergo a
conformational change from an open to
a closed form stabilized by hydrogen
bonding. The study reveals a novel
strategy for the design of molecular
grippers controlled by electrical charge
or light, of interest to the future devel-
opment of artificial molecular machines.
Photoredox-Switchable
Resorcin[4]arene Cavitands: Radical
Control of Molecular Gripping
Machinery via Hydrogen Bonding
Chem. Eur. J. 2017, 23, 1 – 11
www.chemeurj.org
11
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ