Enhanced Gearing Fidelity Achieved through Macrocyclization of a Solvated Molecular Spur Gear

Article abstract of DOI:10.1021/jacs.1c01885

Molecular spur gear dynamics with high gearing fidelity can be achieved through a careful selection of constituent molecular components that favorably position and maintain the two gears in a meshed configuration. Here, we report the synthesis of a new macrocyclic molecular spur gear with a bibenzimidazole stator combined with a second naphthyl bis-gold-phosphine gold complex stator to place two 3-fold symmetric 9,10-diethynyl triptycene cogs at the optimal distance of 8.1 ? for gearing. Micro electron diffraction (μED) analysis confirmed the formation of the macrocyclic structure and the proper alignment of the triptycene cogs. Gearing dynamics in solution are predicted to be extremely fast and, in fact, were too fast to be observed with variable-temperature 1H NMR using CD2Cl2 as the solvent. A combination of molecular dynamics and metadynamics simulations predict that the barriers for gearing and slippage are ca. 4 kcal mol-1 and ca. 9 kcal mol-1, respectively. This system is characterized by enhanced gearing fidelity compared to the acyclic analog. This is achieved by rigidification of the structure, locking the two triptycenes in the preferred gearing distance and orientation.

Full text of DOI:10.1021/jacs.1c01885

pubs.acs.org/JACS
Article
Enhanced Gearing Fidelity Achieved Through Macrocyclization of a
Solvated Molecular Spur Gear
Marcus J. Jellen,$ Ieva Liepuoniute,$ Mingoo Jin, Christopher G. Jones, Song Yang, Xing Jiang,
Hosea M. Nelson,* K. N. Houk,* and Miguel A. Garcia-Garibay*
Cite This: J. Am. Chem. Soc. 2021, 143, 7740−7747
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Molecular spur gear dynamics with high gearing
fidelity can be achieved through a careful selection of constituent
molecular components that favorably position and maintain the two
gears in a meshed configuration. Here, we report the synthesis of a
new macrocyclic molecular spur gear with a bibenzimidazole stator
combined with a second naphthyl bis-gold-phosphine gold complex
stator to place two 3-fold symmetric 9,10-diethynyl triptycene cogs at
the optimal distance of 8.1 Å for gearing. Micro electron diffraction
(μED) analysis confirmed the formation of the macrocyclic structure
and the proper alignment of the triptycene cogs. Gearing dynamics in
solution are predicted to be extremely fast and, in fact, were too fast
1
to be observed with variable-temperature H NMR using CD2Cl2 as the solvent. A combination of molecular dynamics and
metadynamics simulations predict that the barriers for gearing and slippage are ca. 4 kcal mol−1 and ca. 9 kcal mol−1, respectively.
This system is characterized by enhanced gearing fidelity compared to the acyclic analog. This is achieved by rigidification of the
structure, locking the two triptycenes in the preferred gearing distance and orientation.
INTRODUCTION
■
The beginning of nanoscience and nanotechnology can be
traced back to 1959 when Richard Feynman gave his famous
lecture, There’s Plenty of Room at the Bottom, during which he
challenged scientists and engineers to make a (1/64)3 in3
operational electric motor.1 Present day synthetic chemists are
involved in making molecular and supramolecular entities with
both structural and functional analogies to the macroscopic
objects, including gears, pumps, and switches.2 Early examples
of molecular gears were published by Iwamura3 and Mislow4 in
the 1980s, when they reported molecular bevel gear 1 (Figure
1), with two triptycenes covalently linked to a methylene group
or an oxygen atom, respectively. In these landmark studies,
they were able to show that the two triptycenes display fast and
correlated dynamics, resembling those of macroscopic bevel
gears. Since these pioneering studies, many triptycene-based
bevel gears have been developed, including systems with three
linear triptycenes5 or six cyclic triptycenes,6 switchable two-
gear systems,7 and a four-gear system with tunable dynamics.8
The study of molecular spur gears (MSG), with two gearing
cogs arranged in parallel, has seen remarkably less success than
its off-parallel counterparts.
Figure 1. Representative triptycene-based bevel (1), spur (2-6), and
macrocyclic (6-7) gears. All spur gears arrange triptycenes in parallel
while bevel gears position cogs at an off-parallel orientation.
Received: February 22, 2021
Published: May 17, 2021
Similar to the macroscopic objects, molecular gearing occurs
when the two triptycenes undergo concerted, disrotatory
motion, while slippage occurs either by conrotatory motion or
independent rotation. As suggested by Siegel, the primary
figure of merit for determining a gear’s quality is gearing
https://doi.org/10.1021/jacs.1c01885
J. Am. Chem. Soc. 2021, 143, 7740−7747
© 2021 American Chemical Society
7740

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
fidelity, which is defined as the number of geared rotations per
slippage events (Fgear = kgear/kslip).9 Bevel gears with high
gearing fidelity were developed utilizing tightly meshed
systems in which the transition state leading to slippage was
much higher in energy than the one leading to gearing (>30
kcal mol−1 vs 1−2 kcal mol−1). Spur gears utilizing triptycene
have strict geometric requirements to function properly, and
there are no examples of structures shown to displayed high
gearing fidelity. As discussed by Siegel and co-workers, the
optimal distance for triptycene gearing is 8.1 Å, while the
slippage transition state occurs at a distance of 7.2 Å between
the rotational axes of the two gears. From this, the difficulty
with achieving high-functioning spur gears becomes apparent,
the transition state leading to slippage occurs at a shorter
distance than does the one for gearing. This means that in
order to have a properly functioning spur gear, stabilizing,
enthalpic interactions must lower the energy of the gearing
transition state to cause it to be favored over slippage. This
issue is only exacerbated by the entropic penalty that arises as
there are many rotational processes that can lead to slippage,
while there is only one gearing trajectory.
Using a combination of molecular dynamics (MD) and
quantum mechanical (QM) calculations, we recently explored
the gearing and slippage trajectories in MSGs 2-5 (Figure 1).10
In this study, we discovered that diyne-linked gears 2 and 411
have calculated barriers of 5 kcal mol−1 for gearing and of 1−2
kcal mol−1 for slippage, respectively. Interestingly, the
monoalkyne-linked gear 3 shows the inverse preference for
gearing over slippage, which we attribute to the reduction in
flexibility of 3 over 4 by having only a single alkynyl linker
rather than two. Our calculations also showed that the
flexibility of 5 might reduce its gearing efficiency, dropping
the gearing barrier to 2.4 kcal mol−1, which is quite close to
that of slippage at 3.5 kcal mol−1 (down from 3.4 and 7.6 kcal
mol−1, respectively, when calculated with DFT). Clearly, when
one considers the many different vibrational and torsional
modes accessible to various MSGs rather than simply the
transition state (TS) energies, a more complicated picture of
gearing emerges. From this vein, the design of MSGs requires
not only careful engineering of gearing and slippage barriers
but also designing highly rigid structures where torsional
degrees of freedom are limited to gearing processes.12
Figure 2. Design of new macrocyclic MSG built with intermeshed
triptycene blades held in place by a stator consisting of a
bibenzimidazole and a gold-phosphine capping group.
groups to form the desired macrocycle by linking 8a/8b with
digold complex 9.17 The use of deuterated gears was
2
envisioned as a tool to simplify the H NMR spectrum by
removing potential overlapping of aromatic signals from the
two components of the stator. Our group’s experience with
gold complexes inspired us to bond the naphthyl diphosphine
9 to 8 via an Au1+ coordination bond.18 Compound 9 was
chosen as the 2,7-naphthyl substitution pattern to place the
two phosphine groups 8.07 Å apart, a close fit for the idealized
bibenzimidazole stator. Complexation with Au(I) conveniently
coordinates alkyne and phosphine ligands at an angle of 180°,
allowing for the construction of MSG 10a/b with the two
triptycenes oriented in parallel and separated by ∼8.1 Å.
RESULTS AND DISCUSSION
■
Synthesis and Electron Diffraction Structure. With the
above design principles in mind, we set out to synthesize MSG
10a (Scheme 1). Synthesis for 10a started with 9,10-
diethynyltriptycene 11a, which was treated with 1 equiv of n-
Scheme 1. Synthesis of Macrocyclic MSGs 10a and 10b
One method to reduce the number of degrees of freedom
present in a linear or branched molecule is to convert it into a
cyclic structure. This is particularly true for macrocyclization
reactions, as evidenced by the need for templating agents to
overcome the high entropic penalties present in their
syntheses.13 While macrocyclic MSGs such as 6 and 7 from
the Bryan14 and Toyota15 laboratories, respectively, have been
developed, there was no systematic study of how the
macrocyclization affects gearing efficiency in either case. We
envisioned that locking the triptycene cogs into the proper
orientation for gearing using a cyclic structure would restrict
high amplitude bending modes that might reduce gearing
efficiency or facilitate independent rotation of either triptycene,
and ultimately enhance gearing fidelity. From this under-
standing, we sought to synthesize a new macrocyclic MSG and
characterize its dynamics using a combination of molecular
dynamics (MD) simulations and density functional theory
(DFT) calculations. Our design utilizes 5, the bibenzimidazole
stator developed by Siegel and co-workers, which places two
triptycenes in parallel, 8 Å apart (Figure 2). We prepared 8a
and its deuterated analogue 8b16 with two parallel ethynyl
7741
https://doi.org/10.1021/jacs.1c01885
J. Am. Chem. Soc. 2021, 143, 7740−7747

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
BuLi followed by 1.15 equiv of triisopropylsilyl chloride at −78
°C. Subsequent overnight reflux followed by chromatographic
separation resulted in the isolation of 12a in 40% yield with the
double protected and unprotected compounds recollected and
used in proceeding reactions. Compound 12a was then reacted
with the boc-protected bibenzimidazole stator 13 using
standard Sonogashira conditions to afford 14a in 60% yield.
Both boc-protecting groups could then be removed by
treatment with a 1:10 mixture of trifluoroacetic acid (TFA)
and dichloromethane (DCM). This crude mixture was reacted
with an excess of 1-bromo-2-chloroethane in the presence of
Cs2CO3 and THF at elevated temperatures. A short silica
column was run on this mixture to remove any excess 1-
bromo-2-chloroethane before reaction with TBAF in THF to
yield macrocyclization precursor 8a.
We discovered that removal of the boc- groups before TBAF
was critical for maintaining a yield of 40% for these reactions.
Presumably, the bibenzimidazole stator was decomposed by
the presence of ammonium hydroxide in our TBAF solution if
we attempted to deprotect the alkynes first. Finally, 8a was
dissolved in CH2Cl2 with a single drop of triethyl amine (TEA)
and treated with 2 equiv of (dimethylthiol)gold(I) chloride to
forge a covalently linked organogold template for macro-
cyclization.19 Dropwise addition of 9 dissolved in CH2Cl2
produced the macrocyclic MSG 10a in 60% yield after
subsequent workup and trituration of the crude product.
Molecular spur gear 10b was synthesized using the same
procedure with deuterated samples of 11b, which proceeded
with similar yields.
Figure 3. Crystal structure of 10a (A) face on with key distances
measured, (B) side on showing the effects of utilizing tetrahedral
phosphines as stators, and (C) space-filling model with the phosphine
ligand and alkynes hidden to show alignment of the triptycenes. (D)
Crystal packing of 10a with phenylenes on four intermeshed
triptycenes highlighted.
Once the solvent had been removed, the solubility of 10a or
10b became severely limited in all explored solvents, even at
elevated temperatures. Despite their very low solubility,
spectral characterization was possible by 1H NMR, FTIR,
arrange to intermesh their triptycene cogs (Figure 3D). This
finding would likely preclude any solid-state dynamics and
prompted us to focus solely on the solution-phased properties
of 10a.
2
and HRMS (SI section), but we were not able to obtain H
and 13C NMR data within reasonable acquisition times as a
result of sensitivity differences of 9.65 × 10−3 (assuming 100%
labeling) and 1.76 × 10−4 (at natural abundance),
respectively.20 Although large enough single crystals for X-ray
diffraction could not be obtained, we were able to grow
microcrystals suitable for electron diffraction by slow
evaporation of 10a from bromobenzene. Using continuous-
rotation electron diffraction, the structure was solved at 1.10 Å
in space group P21/n.21 Analysis of the crystal structure
(Figure 3A) reveals that the bibenzimidazole stator takes on a
somewhat puckered and twisted shape that forces two
connecting points for each alkyne to be ca. 7.74 Å apart
rather than the expected ca. 8.10 Å for the planar structure.
Gratifyingly, the distance between each triptycene bridgehead
carbons are ca. 7.90 and 7.73 Å apart for the bibenzimidazole
and phosphine side of the molecular, respectively. The two
phosphines on the coordinated stator are approximately ca.
7.58 Å apart, and 0.15 Å closer together than on the
bibenzimidazole stator. This short distance may account for
the puckering observed on the bibenzimidazole. Interestingly,
the tetrahedral geometry of the phosphines causes substantial
stator twisting on the gold-coordinated side of the molecule
(Figure 3B), which brings the two triptycenes slightly out of
parallel alignment. Closer inspection of the triptycene cogs
(Figure 3C) shows that one blade of the first triptycene neatly
packs into the fold of the second one in a Cs symmetric
arrangement, indicating that the two triptycenes can engage in
configurations conducive to gearing. Additionally, analysis of
the crystal packing revealed that multiple molecules of 10a
Variable-Temperature NMR. The labeling of both
triptycenes of the molecular spur gear with a suitable
substituent, as indicated by a dot in two of the blades in
Figure 4, results in desymmetrized rotators that lead to the
Figure 4. Distinct sets of rotational gearing cycles (I, II, and III) of
labeled triptycenes following the disrotatory interconversion of
isomers by a gearing mechanism. Slippage requires conrotation and
interconverts among isomers of the three sets. Three possible slippage
transitions are shown. Enantiomeric conformers are denoted with
prime symbols. Figure adapted from ref 10.
7742
https://doi.org/10.1021/jacs.1c01885
J. Am. Chem. Soc. 2021, 143, 7740−7747

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
formation of 18 different conformers (a-i’) distributed in three
distinct, isomeric cycles (I, II, and III).22 Close analysis of
these cycles reveals that gearing cycles I and II are related by a
mirror plane, and thus can be classified as dl isomers, while
gearing cycle III has an internal mirror plane, leading to its
classification as the meso cycle. Conformers within a single
cycle can interconvert via a concerted, disrotatory (gearing)
motion of each triptycene, while conrotatory rotation or
independent rotation leads to slippage, interconverting the
cycles. If gearing occurs with high fidelity, the time average
solution NMR spectrum of an ideally labeled gear will reveal
two sets of signals for each of the four triptycene protons in a
2:1 ratio that corresponds to isomers a-f and a’-f’, and
conformers g-i’ in the meso cycle. In a situation where gearing
fidelity is low, these isomeric cycles will interconvert, leading to
a single set of four proton signals that represent the time-
averaged spectrum of all the individual conformers a-i’. The
first experimental test of our design was aimed at exploring the
chemical shift changes of both the triptycene α- and β-protons
are easy to follow. The β-protons are highly shielded due to
interactions with the neighboring triptycene blades and are
easily identified by their coupling pattern as two sets of
apparent triplets between 6.9 and 6.5 ppm. By comparison, the
α-protons pointing toward the two different stator components
occur as apparent doublets between 8.1 and 7.75 ppm. A
number of changes were observed for the four triptycene blade
signals. The β-protons observed at 6.60 ppm at 295 K shifted
upfield by 32 Hz, while the signal at 6.87 ppm shifted upfield
by only 8.7 Hz. In contrast to this asymmetric change, both α-
protons shifted upfield by approximately 40 Hz as the
temperature was lowered within the same range. As the
temperature is lowered, the downfield signal experiences a 40
Hz upfield shift, similar to the two α-protons and the upfield β-
proton. Notably, the phosphine phenyl groups that are marked
with a black arrow start as two sets of complex signals, one at
approximately 7.55 ppm and another around 7.65 ppm, and
they both shift upfield and merge as the temperature reaches
187 K. A similar broadening for all signals suggests that there
1
dynamics of the blades by variable-temperature H NMR,
knowing that without proper labeling we can only ascertain
whether rotational exchange between the three nonequivalent
blade site occurs faster or slower than the NMR time scale. We
expected to gain an additional insight into the energetics of the
gearing and slippage processes by taking advantage of
computational methods, which could eventually be used to
decide the type(s) of substituents that could make the two
cycles experimentally observable.
1
are no internal dynamic processes in the H NMR time scale
indicative of signal coalescence resulting from a slow site
exchange. In fact, chemical shift changes might arise due to
changes in the macrocycle structure, potentially mediated by
hydrogen-bonding interaction between the bibenzidimazole
stator and the α-protons in the triptycene gear with broadening
caused by an increase in the viscosity of CD2Cl2 as it nears its
freezing point (Figure SI27).
1
As shown in Figure 5, VT H NMR measurements with
MSG 10b were carried out in CD2Cl2 solutions with
One may conclude from the above experimental observa-
tions that the rotational dynamics of MSG 10, whether gearing
or slippage, are determined by free energy barriers that are easy
to surmount even at 187 K. However, it is possible that
differences in chemical shifts in the three nonequivalent blade
position of each triptycene gear are not enough to elicit a
noticeable effect in the NMR spectra. A simple estimate of the
conditions that are needed to make rotational dynamics
observable in solution by VT NMR can be obtained by
analyzing the combined effect of the activation energy (ΔG‡)
and the difference in chemical shift between different exchange
sites (Δν), on the observed coalescence temperature (TC), as
determined by the following equation23
ΔG≠ = 4.55*10−3*T*(9.97*log T − log(500*Δν))
c
c
For coalescence to be experimentally observable near the
temperature limit of ca. Tc ≈ 180 K of CD2Cl2, gearing
dynamics would require activation free energy values (ΔG‡) in
the range of ca. 8.5 to 5.5 kcal mol−1 for chemical shift
differences between exchange sites (Δν) that differ from ca.
200 to 1500 Hz, respectively. This suggests that in addition to
high barriers, different sites would require substantially
different shielding environments (Δν) for gearing cycles to
be unambiguously detected by VT NMR. With this
information at hand, we turned to density functional theory
(DFT) and performed NMR calculations to obtain the
isotropic shift values for hydrogen atoms of the three different
triptycene blades. Calculated isotropic shielding values of the
most stable gear structure revealed frequency variations in the
range of 500 to 1500 Hz at a spectrometer frequency of 500
MHz, as the one used in our experiments (SI Table SI1). On
the basis of these values, we can conclude that an
experimentally measurable rotational barrier at or above 187
K would need activation free energies greater than ca. 7−8 kcal
mol−1. However, it should be noted that this result is not an
Figure 5. Variable-temperature 1H spectra of MSG 10b at
temperatures ranging from 295 to 187 K. The α- and β-protons are
marked with blue and red arrows, respectively. Signals indicated by
the black arrow correspond to the phosphine ligand.
temperatures ranging from 295 to 187 K, which is close to
the freezing point of the solvent and the limit for our
spectrometer. Similar results obtained with 10a between 298
and 212 K are shown included Figure SI26. Triptycene protons
parallel to the axis of rotation and ortho- to the bridgehead
carbon (α-protons) are labeled with blue arrows while protons
meta- to the bridgehead carbon (β-protons) are indicated with
red arrows. While the spectra are relatively complicated,
7743
https://doi.org/10.1021/jacs.1c01885
J. Am. Chem. Soc. 2021, 143, 7740−7747

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
indication of gearing and/or slippage, as both can occur with
high exchange rates over low barriers. In fact, for distinct
gearing cycles to be experimentally observed in solution by VT
1H NMR, it is not enough to have optimal gearing barriers, but
it would be also essential that slippage has a barrier that is
substantially higher than 7−8 kcal mol−1 as lower values will
interconvert structures of the two types of cycles with
corresponding NMR time scale.
Computational Studies. To analyze gearing dynamics by
computational methods,24 we sought to assess how well the
two triptycenes stay intermeshed in the macrocyclic
architecture as compared to the half-open gear with a
bibenzimidazole stator previously reported by Siegel et al.9
We performed molecular dynamics (MD) simulations at 300 K
for 1 μs for the two molecular gear systems immersed in a
DCM solvent box and tracked the distances between the
bridgehead carbons in intermeshed triptycenes. MD simu-
lations were performed using the GPU code (pmemd.cuda)25
of the AMBER 16 package.26 Parameters for all nonmetal
atoms were generated within the antechamber module using
the general AMBER force field (gaff).27 Parameters for the
Au1+ ions were generated using metal center parameter builder
(MCPB.py).28 Further computational details are described in
the SI section and in a recent publication from our group.24
Figure 6 depicts the range of distances that the intermeshed
Having gained insight into the positive impact of macro-
cyclization on the gearing process, we performed MD
simulations in the temperature range used in the VT-NMR
experiments in dichloromethane, from 207 to 300 K. The
simulations were run for 1 μs to allow the system to visit all the
energetically relevant configurations. Rotational trajectories
were analyzed by tracing the motion of each triptycene gear via
dihedral angle changes that vary from −180° to 180°. The
dihedral angle for each triptycene is defined as the angle
between the plane of the bibenzimidazole stator and the plane
of one of the triptycene blades. For a meshed triptycene system
undergoing correlated rotation, gearing is observed when
dihedral angles associated with different triptycenes turn in
opposite directions by 60°. The MD trajectories performed at
207 K for 1 μs showed multiple gearing events (Figure 7A).
Plotting the dihedral angle evolution in time shows that as one
triptycene gear is changing its dihedral angle in one direction
by 60°, the other simultaneously responds in the opposite
direction, as required for geared rotation. As shown in the
bottom frames in Figure 7, the extent of correlated behavior
between the two gears can be appreciated by plotting one
dihedral as a function of the other. At 207 K the occupied sites
are consistent with the three symmetry-related ground state
configurations achieved via gearing (Figure 7A, bottom frame).
As the MD simulations temperature increases, more gearing
events are observed. At 247 K (Figure 7B) and 300 K (Figure
7C), multiple symmetry-related gearing trajectories each offset
by a ca. 60° are populated. Configurations outside the diagonal
gearing trajectories indicate slippage events.
In order to evaluate the activation barrier to gearing and
map the energy landscape, we utilized an accelerated sampling
technique via well-tempered metadynamics simulations.
Metadynamics simulation were performed using the NAMD
3.0 program.29 The free energy surface was reconstructed as a
function of the dihedral angles describing each triptycene
(Figure 8A). By introducing a bias potential (or force) that
acts on the selected variables, the potential mean force (PMF)
plot along the chosen coordinates was generated (Figure 8B).
The lowest energy trajectories match the ones obtained with
MD simulations (Figure 7C) and indicate efficient gearing.
The energy barriers for gearing and slippage were calculated to
be ca. 4 kcal mol−1 and ca. 9 kcal mol−1, respectively. Notably,
the energy barrier for gearing is smaller than the 7.3 kcal mol−1
barrier calculated by Siegel et al. for the half-open gear with
bibenzimidazole stator in the gas phase.9 The slight asymmetry
in the PMF plot is attributed to the asymmetric structure of the
molecular spur gear and its twisted framework.
Similar metadynamics studies were previously done by our
group on Siegel’s half-open molecular spur gear with an
ethenyl bridged bibenzimidazole stator. For that system, we
confirmed low rotational barriers for gearing and slippage (2.4
and 3.5 kcal mol−1, respectively) and showed perfectly
symmetric PMF plot.10 We believe that the locking of
triptycenes in a meshed configuration, at suitable interaxle
distance for gearing, stabilizes the gearing transition state and
allows the system to efficiently undergo gearing dynamics.
These computational observations suggest that gearing
dynamics are outside the experimental NMR time scales, and
the calculated barriers for gearing and slippage are too low for
the motion to be observed experimentally via phase isomer
formation.
Figure 6. Interaxle distances between bridgehead carbons in
intermeshed triptycenes for half-open gear (green) and locked
macrocyclic gear (blue) as simulated in DCM at 300 K. The
distribution of interaxle distances is wider for the half-open gear
systems.
triptycenes are able to explore while undergoing their
correlated motion. Triptycenes in the half-open gear (colored
in green) spanned the range of ca. 7.0 to 9.5 Å with two broad
peaks observed at ca. 7.4 Å and ca. 8.2 Å consistent with
distances that are favorable for the slippage and gearing
transition states, respectively. Notably, more conformations
were found at distances suitable for the slippage transition state
than for gearing. The distance range between triptycenes in the
macrocyclic gear (colored in blue) was smaller varying from ca.
7.0 Å to ca. 8.6 Å with the highest probability density observed
at ca. 8.1 Å, which suggests that triptycenes undergo their
dynamics at a distance that is expected to favor the transition
state for gearing. These theoretical observations support the
use of macrocyclization in order to keep triptycene gears
tightly intermeshed as a strategy to enhance the gearing
fidelity.
7744
https://doi.org/10.1021/jacs.1c01885
J. Am. Chem. Soc. 2021, 143, 7740−7747

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 7. Dynamic simulations of triptycene gearing carried out for 1 μs at (A) 207 K, (B) 247 K, and (C) at 300 K in a locked macrocyclic gear
immersed in DCM solvent box. Different colors in each plot represent dihedral angle values for the two intermeshed triptycene gears. A gearing
event is observed when a dihedral angle for each gear is changing by 60° in opposite directions. Correlated maps show how the two dihedral angles
are simultaneously changing in opposite directions; as one dihedral is increasing, the other is simultaneously decreasing. As the simulation
temperature increases, more correlated motion events are observed and more symmetry-related gearing trajectories are generated.
Figure 9. Ground state and key transition states for 10a, and other
triptycene-based spur gears, as well as key intermolecular interactions
for tuning energetic barriers.
Figure 8. (A) Dihedral angles (atoms circled in red and blue) used to
define triptycene rotation in the macrocyclic gear. (B) Potential mean
force plot for triptycene rotation generated using well-tempered
metadynamics simulations. The top-left to bottom-right diagonals
represent the lowest energy trajectory for gearing, with dihedral angles
changing in opposite directions. Energy scale is in kcal mol−1.
compared to those of the half-open gear 5. In the macrocyclic
gear, the barriers for gearing and slippage were ca. 4 kcal mol−1
and ca. 9 kcal mol−1, respectively, while in the half-open gear 5,
as previously studied by Siegel et al., they were ca. 7.3 kcal
mol−1 and ca. 11.6 kcal mol−1, respectively. While bevel gears
only required tight meshing to create experimentally visible
dynamics, spur gears also require finely tuned energy barriers
that bring the barrier to slippage above 11 kcal mol−1 such that
the gearing dynamics will be observable in solution using
variable-temperature NMR.
These findings suggest that future iterations of triptycene-
based spur gears should utilize macrocyclic stators that
position triptycenes ca. 8.1 Å apart in a rigid structure to
prevent slippage, and then alter the slippage barrier using
asymmetric triptycenes containing a combination of electron
poor and electron rich phenylenes. In this system, favorable
orbital interactions would stabilize the ground state and
gearing transition state, but would be lost in the slippage
CONCLUSIONS
■
Our results suggest that macrocyclization is a valuable strategy
for enhancing gearing fidelity in molecular spur gear systems in
solution by locking the triptycene cogs into the correct
position and orientation for gearing. Using a convenient
convergent synthesis, we were able to create a rigid organogold
complex with triptycenes aligned in parallel and at a proper
gearing distance. The MD simulations showed that macro-
cyclization prevents 10a from accessing slippage transition
states and reverses the preference for slippage observed with
MSG 5 by limiting the bending modes that would otherwise be
accessible to the gear. Metadynamics simulations of the
energetic profile for gearing indicated that macrocyclization
did not change the relative barriers of 10a significantly as
7745
https://doi.org/10.1021/jacs.1c01885
J. Am. Chem. Soc. 2021, 143, 7740−7747

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
transition state (Figure 9), widening the energetic gap between
gearing and slipping.
for JSPS fellows JP17J01104. C.G.J. acknowledges the National
Science Foundation Graduate Research Fellowship Program
(DGE-1650604), the Christopher S. Foote Fellowship, and the
Pat Tillman Foundation for funding. H.M.N. acknowledges the
Packard Foundation and Bristol Myers Squibb for generous
funding. The authors thank Duilio Cascio, Genesis Falcon and
Michael R. Sawaya (UCLA-DOE Institute) for assistance in
crystallography with data processing, and refinement. This
research used resources at the UCLA-DOE Institute’s X-ray
Crystallography Core Facility which is supported by the U.S.
Department of Energy (DE-FC02-02ER63421).
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.1c01885.
■
sı
Spectroscopic characterization, diffraction data, and
computational methods (PDF)
AUTHOR INFORMATION
Corresponding Authors
■
REFERENCES
■
Hosea M. Nelson − Department of Chemistry and
Biochemistry, University of California, Los Angeles,
California 90095-1569, United States; orcid.org/0000-
0002-4666-2793; Email: hosea@chem.ucla.edu
K. N. Houk − Department of Chemistry and Biochemistry,
University of California, Los Angeles, California 90095-1569,
United States; orcid.org/0000-0002-8387-5261;
Email: houk@chem.ucla.edu
(1) Feynman, R. P.; Leighton, R. B.; Sands, M. The Feynman
Lectures on Physics; Addison-Wesley: Reading, MA, 1963; Vol. 1,
Chapter 46.
(2) (a) Balzani, V.; Credi, A.; Raymo, F.; Stoddart, J. F. Artificial
molecular machines. Angew. Chem., Int. Ed. 2000, 39, 3348−3391.
(b) Gattler, D.; Fuks, G.; Heiser, J.; Moulin, E.; Perrot, A.; Yao, X.;
Giuseppone, N. Design of Collective Motions from Synthetic
Molecular Switches, Rotors, and Motors. Chem. Rev. 2020, 120,
310−433. (c) Krause, S.; Feringa, B. L. Towards artificial molecular
factories from framework-embedded molecular machines. Nat. Rev.
Chem. 2020, 4, 550−562.
(3) Kawada, Y.; Iwamura, H. Unconventional Synthesis and
Conformational Flexibility of Bis(l-triptycyl) Ether. J. Org. Chem.
1980, 45, 2547−2548.
(4) Hounshell, W. D.; Johnson, C. A.; Guenzi, A.; Cozzi, F.; Mislow,
K. Stereochemical consequences of dynamic gearing in substituted
bis(9-triptycyl) methanes and related molecules. Proc. Natl. Acad. Sci.
U. S. A. 1980, 77, 6961−6964.
Miguel A. Garcia-Garibay − Department of Chemistry and
Biochemistry, University of California, Los Angeles,
California 90095-1569, United States; orcid.org/0000-
0002-6268-1943; Email: mgg@chem.ucla.edu
Authors
Marcus J. Jellen − Department of Chemistry and Biochemistry,
University of California, Los Angeles, California 90095-1569,
United States
Ieva Liepuoniute − Department of Chemistry and
Biochemistry, University of California, Los Angeles,
California 90095-1569, United States; orcid.org/0000-
0003-2965-8642
Mingoo Jin − Department of Chemistry and Biochemistry,
University of California, Los Angeles, California 90095-
1569, United States; orcid.org/0000-0001-6199-8802
Christopher G. Jones − Department of Chemistry and
Biochemistry, University of California, Los Angeles,
California 90095-1569, United States
Song Yang − Department of Chemistry and Biochemistry,
University of California, Los Angeles, California 90095-
1569, United States
Xing Jiang − Department of Chemistry and Biochemistry,
University of California, Los Angeles, California 90095-
1569, United States; orcid.org/0000-0001-8259-1948
(5) Koga, N.; Kawada, Y.; and, H. I. Recognition of the Phase
Relationship between Remote Substituents in 9.10- Bis(3-chloro-9-
triptycyloxy)triptycene Molecules Undergoing Rapid Internal Rota-
tion Cooperatively. J. Am. Chem. Soc. 1983, 5, 5498−5499.
(6) Ube, H.; Yamada, R.; Ishida, J. I.; Sato, H.; Shiro, M.; Shionoya,
M. A Circularly Arranged Sextuple Triptycene Gear Molecule. J. Am.
Chem. Soc. 2017, 139, 16470−16473.
(7) (a) Setaka, W.; Nirengi, T.; Kabuto, C.; Kira, M.; Ch, X. X.
Introduction of Clutch Function into a Molecular Gear System by
Silane - Silicate Interconversion. J. Am. Chem. Soc. 2008, 2, 15762−
15763. (b) Ube, H.; Yasuda, Y.; Sato, H.; Shionoya, M. Metal-
Centred Azaphosphatriptycene Gear with a Photo- and Thermally
Driven Mechanical Switching Function Based on Coordination
Isomerism. Nat. Commun. 2017, 8, 1−6.
(8) Sanada, K.; Ube, H.; Shionoya, M. Rotational Control of a
Dirhodium-Centered Supramolecular Four-Gear System by Ligand
Exchange. J. Am. Chem. Soc. 2016, 138, 2945−2948.
(9) Frantz, D. K.; Linden, A.; Baldridge, K. K.; Siegel, J. S. Molecular
Spur Gears Comprising Triptycene Rotators and Bibenzimidazole-
Based Stators. J. Am. Chem. Soc. 2012, 134, 1528−1535.
(10) Jiang, X.; Yang, S.; Jellen, M. J.; Houk, K. N.; Garcia-Garibay,
M. A. Molecular Spur Gears with Triptycene Rotators and a
Norbornane-Based Stator. Org. Lett. 2020, 22, 4049−4052.
(11) Toyota, S.; Shimizu, T.; Iwanaga, T.; Wakamatsu, K. Structures
and Conformational Analysis of 1,8-Bis(9-Triptycylethynyl) Anthra-
cene and Its Derivatives as Prototypes of Molecular Spur Gears.
Chem. Lett. 2011, 40, 312−314.
(12) Liepuoniute, I.; Jellen, M. J.; Garcia-Garibay, M. A. Correlated
motion and mechanical gearing in amphidynamic crystalline
molecular machines. Chem. Sci. 2020, 11, 12994−13007.
(13) Marti-Centelles, V.; Pandey, M. D.; Burguete, I.; Luis, S. V.
Macrocyclization Reactions: The Importance of conformational,
Configurational, and Template-Induced Preorgonization. Chem. Rev.
2015, 115, 8736−8834.
(14) Bryan, J. C.; Sachleben, R. A.; Gakh, A. A.; Bunick, G. J.
Molecular Gears: Structures of (9, 10-Triptyceno) Crown Ethers. J.
Chem. Crystallogr. 1999, 29, 513−521.
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.1c01885
Author Contributions
$These authors contributed equally. The manuscript was
written through contributions of all authors. All authors have
given approval to the final version of the manuscript.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank National Science Foundation Grants grants CHE-
1764328, DMR-1700471, and MRI-1532232 for supporting
this work. Computations were performed on the Hoffman2
cluster at UCLA and the Extreme Science and Engineering
Discovery Environment (XSEDE), which is supported by the
NSF (OCI-1053575). M. Jin was supported by the grant-in-aid
7746
https://doi.org/10.1021/jacs.1c01885
J. Am. Chem. Soc. 2021, 143, 7740−7747

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(15) Toyota, S.; Kawahata, K.; Sugahara, K.; Wakamatsu, K.;
Iwanaga, T. Triple and Quadruple Triptycene Gears in Rigid
Macrocyclic Frameworks. Eur. J. Org. Chem. 2017, 37, 5696−5707.
(16) Jiang, X.; Rodriguez-Molina, B.; Nazarian, N.; Garcia-Garibay,
M. A. Rotation of a Bulky Triptycene in the Solid State: Towards
Engineered Nanoscale Artificial Molecular Machines. J. Am. Chem.
Soc. 2014, 136, 8871−8874.
(17) For comparison, we also attempted to synthesize the symmetric
macrocycle replacing 9 with another bibenzimidazole stator. However,
after multiple attempts to close the macrocycle with Sonogashira
coupling, no amount of the symmetric product could be obtained. We
believe that this method suffered from competing pathways such as
polymer formation and created insoluble mixtures that were not
amenable to complete characterization.
(18) (a) Jin, M.; Chung, T. S.; Seki, T.; Ito, H.; Garcia-Garibay, M.
A. Phosphorescence Control Meditated by Molecular Rotation and
Aurophilic Interactions in Amphidynamic Crystals of 1,4-Bis[tri-(p-
fluorophenyl)phosphane-gold(I)-ethynyl]benzene. J. Am. Chem. Soc.
2017, 139, 18115−18121. (b) Jin, M.; Yamamoto, S.; Seki, T.; Ito,
H.; Garcia-Garibay, M. A. Anisotropic Thermal Expansion as the
Source of Macroscopic and Molecular Scale Motion in Phosphor-
escent Amphidynamic Crystals. Angew. Chem. 2019, 131, 18171−
18178.
(19) Vicente, J.; Chicote, M.-T.; Alvarez-Falcon, M. M. Gold(I) and
Silver(I)/Gold(I) Complexes Derived from C6Me4(C⋮CH)2-1,2.
Organometallics 2005, 24, 4666−4675.
(20) Bruker Almanac, 2011, ISBN-13 978−3−929432−25−4.
(21) Jones, C. G.; Asay, M.; Jim, L. J.; Kleinsasser, J. F.; Saha, F.;
Fulton, T. J.; Berkly, K. R.; Cascio, D.; Malyutin, A. G.; Conley, M. P.;
Stoltz, B. M.; Lavallo, V.; Rodriguez, J. A.; Nelson, H. M.
Characterization of reactive organometallic species via MicroED.
ACS Cent. Sci. 2019, 5, 1507−1513.
(22) (a) Kawada, Y.; Iwamura, H. Phase Isomerism in Gear-Shaped
Molecules. Tetrahedron Lett. 1981, 22, 1533−1536. (b) Iwamura, H.;
Mislow, K. Stereochemical Consequences of Dynamic Gearing. Acc.
Chem. Res. 1988, 21, 175−182.
(23) Hesse, M.; Maier, H.; Zeeh, B. Spectroscopic Methods in
Organic Chemistry; Thieme Publishing Group: Stuttgart, 2007, 74−
242.
(24) Liepuoniute, I.; Sanders, J. N.; Garcia-Garibay, M. A.; Houk, K.
N. Computational Investigation into Ligand Effects on Correlated
Geared Dynamics in Dirhodium Supramolecular Gears − Insights
Beyond the NMR Experimental Window. J. Org. Chem. 2020, 85,
8695−8701.
(25) Salomon-Ferrer, R.; Götz, A. W.; Poole, D.; Le Grand, S.;
Walker, R. C. Routine Microsecond Molecular Dynamics Simulations
with AMBER on GPUs. 2. Explicit Solvent Particle Mesh Ewald. J.
Chem. Theory Comput. 2013, 9, 3878−3888.
(26) Case, D. A.; Cerutti, D. S.; Cheatham, T. E., III; Darden, T. A.;
Duke, R. E.; Giese, T. J.; Gohlke, H.; Goetz, A. W.; Greene, D.;
Homeyer, N.; Izadi, S.; Kovalenko, A.; Lee, T. S.; LeGrand, S.; Li, P.;
Lin, C.; Liu, J.; Luchko, T.; Luo, R.; Mermelstein, D.; Merz, K. M.;
Monard, G.; Nguyen, H.; Omelyan, I.; Onufriev, A.; Pan, F.; Qi, R.;
Roe, D. R.; Roitberg, A.; Sagui, C.; Simmerling, C. L.; Botello-Smith,
W. M.; Swails, J.; Walker, R. C.; Wang, J.; Wolf, R. M.; Wu, X.; Xiao,
L.; York, D. M.; Kollman, P. A. AMBER 2017; University of
California: San Francisco, 2017.
(27) Wang, J.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case, D.
A. Development and Testing of a General Amber Force Field. J.
Comput. Chem. 2004, 25, 1157−1174.
(28) Li, P.; Merz, K. M. MCPB.py: A Python Based Metal Center
Parameter Builder. J. Chem. Inf. Model. 2016, 56, 599−604.
(29) Phillips, J. C.; Braun, R.; Wang, W.; Gumbart, J.; Tajkhorshid,
E.; Villa, E.; Chipot, C.; Skeel, R. D.; Kale, L.; Schulten, K. Scalable
Molecular Dynamics with NAMD. J. Comput. Chem. 2005, 26, 1781−
1802.
7747
https://doi.org/10.1021/jacs.1c01885
J. Am. Chem. Soc. 2021, 143, 7740−7747