Pentiptycene-derived light-driven molecular brakes: Substituent effects of the brake component

Article abstract of DOI:10.1002/chem.201000764

Five pentiptycene-derived stilbene systems (1R; R = H, OM, NO, Pr, and Bu) have been prepared and investigated as light-driven molecular brakes that have different-sized brake components (1H<1OM<1NO< 1Pr<1Bu). At room temperature (298 K), rotation of the pentiptycene rotor is fast (k_rot = 10⁸-10⁹ s^-1) with little interaction with the brake component in the trans form ((E)-1R), which corresponds to the brake-off state. When the brake is turned on by photoisomerization to the cis form ((Z)-1R), the pentiptycene rotation can be arrested on the NMR spectroscopic timescale at temperatures that depend on the brake component. In the cases of (Z)-1NO, (Z)-1Pr, and (Z)-1Bu, the rotation is nearly blocked (k_rot = 2-6 s^-1) at 298 K. It is also demonstrated that the rotation is slower in [D₆]DMSO than in CD₂Cl ₂. A linear relationship between the free energies of the rotational barrier and the steric parameter A values is present only for (Z)-1H, (Z)-1OM, and (Z)-1NO, and it levels off on going from (Z)-1NO to (Z)-1Pr and (Z)-1Bu. DFT calculations provide insights into the substituent effects in the rotational ground and transition states. The molar reversibility of the E-Z photoswitching is up to 46%, and both the E and Z isomers are stable under the irradiation conditions.

Full text of DOI:10.1002/chem.201000764

DOI: 10.1002/chem.201000764
Pentiptycene-Derived Light-Driven Molecular Brakes:
Substituent Effects of the Brake Component
Wei-Ting Sun,[a] Yau-Ting Huang,[a] Guan-Jhih Huang,[a] Hsiu-Feng Lu,[b] Ito Chao,*[b]
Shou-Ling Huang,[a] Shing-Jong Huang,[a] Ying-Chih Lin,[a] Jinn-Hsuan Ho,[a, c] and
Jye-Shane Yang*[a]
Abstract: Five pentiptycene-derived
stilbene systems (1R; R=H, OM, NO,
Pr, and Bu) have been prepared and
investigated as light-driven molecular
brakes that have different-sized brake
the pentiptycene rotation can be arrest-
ed on the NMR spectroscopic time-
scale at temperatures that depend on
the brake component. In the cases of
(Z)-1NO, (Z)-1Pr, and (Z)-1Bu, the
rotation is nearly blocked (krot =2–
6 sꢀ1) at 298 K. It is also demonstrated
that the rotation is slower in
[D6]DMSO than in CD2Cl2. A linear
relationship between the free energies
of the rotational barrier and the steric
parameter A values is present only for
(Z)-1H, (Z)-1OM, and (Z)-1NO, and
it levels off on going from (Z)-1NO to
(Z)-1Pr and (Z)-1Bu. DFT calcula-
tions provide insights into the substitu-
ent effects in the rotational ground and
transition states. The molar reversibili-
ty of the E–Z photoswitching is up to
46%, and both the E and Z isomers
are stable under the irradiation condi-
tions.
components
(1H<1OM<1NO<
1Pr<1Bu). At room temperature
(298 K), rotation of the pentiptycene
rotor is fast (krot =108–109 sꢀ1) with
little interaction with the brake compo-
nent in the trans form ((E)-1R), which
corresponds to the brake-off state.
When the brake is turned on by photo-
isomerization to the cis form ((Z)-1R),
Keywords: conformation analysis ·
fused-ring systems · isomerization ·
photochromism · substituent effects
Introduction
gears,[3] shuttles,[4] and motors.[5] One of the most important
issues in operating a molecular machine is the input of
energy (fuel). Whereas biological systems such as adenosine
triphosphate (ATP) synthases and myosins use chemical
energy,[6] artificial molecular machines can adopt photons
and electrons as well as chemicals as their energy sources.[7]
In this context, photochemically and electrochemically
driven systems are particularly intriguing, because they are
free of the problem of accumulation of chemical waste
during the operation.[8]
The concept of the “bottom-up approach toward miniature
devices” has led to the development of many prototypes of
molecule- and supramolecule-based machinery systems[1]
that mimic the macroscopic counterparts such as brakes,[2]
[a] W.-T. Sun, Y.-T. Huang, G.-J. Huang, S.-L. Huang, Dr. S.-J. Huang,
Prof. Y.-C. Lin, Dr. J.-H. Ho, Prof. J.-S. Yang
Department of Chemistry, National Taiwan University
Taipei 10617 (Taiwan)
We recently communicated[9] a light-driven molecular
brake 1NO (Figure 1), which displays an unprecedented
brake performance: namely, the rotation rate (krot) for the
rotor is decreased by nearly 9 orders of magnitude at ambi-
ent temperature (298 K) upon switching on the brake. In
1NO, the paddle-wheel-like pentiptycene group serves as
the rotor, the dinitrophenyl group as the brake component,
and the vinyl linker as the photocontrollable switch through
trans–cis (E–Z) isomerization. Whereas the brake unit has
negligible noncovalent interactions with the rotor in the
trans (E) form, the interactions in the cis (Z) form are so
significant that rotation of the rotor is nearly inhibited.
Thus, the trans ((E)-1NO) and cis ((Z)-1NO) isomers corre-
Fax : (+886)223636359
E-mail: jsyang@ntu.edu.tw
[b] Dr. H.-F. Lu, Prof. I. Chao
Institute of Chemistry, Academia Sinica
Taipei 11529 (Taiwan)
Fax : (+886)227831237
E-mail: ichao@chem.sinica.edu.tw
[c] Dr. J.-H. Ho
Current address: Department of Chemical Engineering
National Taiwan University of Science and Technology
Taipei 10607 (Taiwan)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000764.
11594
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 11594 – 11604

FULL PAPER
that the brake size would play an important role in the tor-
sional barrier of the pentiptycene rotor.
We report herein the synthesis of a series of 1NO ana-
logues, namely, 1H, 1OM, 1Pr, and 1Bu, by replacing the
nitro groups in the brake component with smaller (H and
methoxy) or larger (isopropyl and tert-butyl) substituents. In
conjunction with 1NO, the substituent effects on the brake
performance could be addressed. Our results show that not
only the size but also the electronic nature of the substitu-
ents affects the ground and the transitional states of the
brake-on from ((Z)-1R), thus leading to substituent-depen-
dent rotation kinetics for the rotor and the brake. The sub-
stituents also affect the quantum efficiency and molar rever-
sibility of E--Z photoisomerization. The solvent effect on
the rotation kinetics is also demonstrated.
Results and Discussion
Figure 1. a) Molecular structures and b) schematic illustrations of pentip-
tycene-derived light-driven molecular brakes (1R) that differ in the sub-
stituents in the brake component. The trans ((E)-1R) and cis ((Z)-1R)
isomers correspond to the brake-off and brake-on states. The numerical
labels for protons and carbons are for discussion of the VT NMR spectra.
The white and brown curved arrows denote the internal rotation of the
rotor and the brake groups, respectively.
Synthesis: The stilbene backbone in 1R (R=NO, H, OM,
Pr, or Bu) was constructed by either Wittig reaction be-
tween the pentiptycene carbaldehyde 2 and the correspond-
ing phosphorus ylides or by Heck reaction with iodopentip-
tycene 3 and substituted styrenes. The synthesis of both the
pentiptycene building blocks 2 and 3 has been reported.[10]
Compounds 1NO, 1H, and 1OM were prepared by the
Wittig route, and 1Pr and 1Bu through the Heck route
(Scheme 1), simply depending on the relative ease of synthe-
sis of the desired phosphorus ylides and styrenes. For exam-
ple, the benzyl chlorides for the Wittig route toward 1NO
and 1OM could be obtained from the reaction of thionyl
chloride and commercially available 3,5-dimethoxybenzyl al-
cohol (4) and 3,5-dinitrobenzyl alcohol (5), respectively. The
styrenes for the Heck route toward 1Pr and 1Bu could be
prepared from the corresponding 3,5-dialkyl bromobenzenes
by reacting with vinylmagnesium bromide under the cataly-
spond to the brake-off (krot =108–109 sꢀ1) and the brake-on
(krot =3 sꢀ1) states, respectively. Because of the C2 symmetry
of the pentiptycene rotor, there are two large U-shaped cav-
ities and two shallower V-shaped ones. In the Z form, the
former can accommodate the brake component, but the
latter are inaccessible to the brake component as a result of
severe steric interactions with the bridgehead and phenylene
hydrogen atoms. Consequently, conformers with the brake
component aligned toward the U- and V-shaped cavities cor-
respond to the ground state and the rotational transition
state, respectively (Figure 2). Thus, the brake component en-
counters two isoenergetic ground and transition states
during a full turn of the rotor in (Z)-1NO. We can envisage
sis of [PdACHTUNTGRENNG(U dppf)Cl2] (dppf=1,1’-bis(diphenylphosphino)fer-
rocene). The starting material 1-bromo-3,5-di(tert-butyl)ben-
zene (6) is commercially available, and 1-bromo-3,5-diiso-
propylbenzene (7) can be obtained from 2-amino-1,3-diiso-
Figure 2. Rotational energy diagram for a full turn of the rotor in (Z)-1R
(the brake-on state). The rotation angle is arbitrarily defined.
Scheme 1. Retrosynthesis of molecular brakes 1.
Chem. Eur. J. 2010, 16, 11594 – 11604
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11595

J.-S. Yang, I. Chao et al.
propylbenzene by following the literature procedures.[11] It
should be noted that the Wittig route gave a mixture of E
and Z isomers, which could be separated by column chroma-
tography (1NO) or by preparative high-performance liquid
chromatography (HPLC; 1H and 1OM). In contrast, the
Heck route led to the E isomers only. Under the irradiation
of 340 nm light, the E isomers were converted to the corre-
sponding Z isomers with a nearly quantitative yield. See the
Experimental Section for detailed synthetic procedures and
compound characterization data.
Variable-temperature (VT) NMR spectroscopy and rotation
kinetics: The two aryl groups (i.e., the rotor and the brake
units) of the stilbene backbone in 1R can undergo inde-
pendent internal rotation about the aryl–vinyl single bonds
(Figure 1b). In the E form, the rotational barriers are ex-
pected to be low for both the rotor and the brake due to
their negligible steric interactions. However, their close
proximity in the Z form significantly raises their rotational
barriers. In principle, the bulkier is the brake unit, the larger
is the steric hindrance, and thus a larger rotational barrier
for the rotor is expected. On the basis of the steric parame-
ter A values (in kcalmolꢀ1),[12] the relative size of the brake
unit in 1R increases in the order 1H (A=0)<1OM (A=
0.6)<1NO (A=1.1)<1Pr (A=2.2)<1Bu (A=4.8).
To investigate the rotation kinetics (barriers and rates) of
(Z)-1R, we have carried out variable-temperature 1H and
13C NMR spectroscopic studies (500 and 125 MHz, respec-
tively) and spectral simulations. The spectra were recorded
in CD2Cl2 for 1H and 1OM to observe decoalescence of the
signals below room temperature and in [D6]DMSO for
1NO, 1Pr, and 1Bu to reach coalescence of the split signals
above room temperature (vide infra). To compare the sol-
vent effect on the rotation kinetics, 1Pr was also studied in
CD2Cl2. Spectral simulations for retrieving the rotation rate
were exerted based on VT 13C NMR spectra rather than the
1
corresponding H NMR spectra, because the latter is com-
plicated by the multiplicity of proton signals and the pres-
ence of more than one type of spin system in the pentipty-
cene rotor. In the following, the experimental and simulated
NMR spectra will be shown for the two extremes in terms
of the brake size, 1H and 1Bu, and those for 1OM and 1Pr
are supplied as Supporting Information. The detailed
VT NMR spectra for 1NO have recently been communicat-
ed.[9]
Figure 3. Aromatic region of a) 1H and b) 13C NMR spectra of (E)-1H
and (Z)-1H in CD2Cl2, and that of c) 1H and d) 13C NMR spectra of (E)-
1Bu and (Z)-1Bu in [D6]DMSO at 298 K. The solid circles denote peaks
due to the styryl group (the brake and the vinyl switch components), and
the Arabic numerals denote the corresponding protons and carbon atoms
labeled in Figure 1.
1
Figure 3 shows the aromatic region of H and 13C NMR
spectra of (E)-1H, (Z)-1H, (E)-1Bu, and (Z)-1Bu at 298 K.
Except for some quaternary carbon atoms, the signals can
be unambiguously assigned on the basis of a series of 2D
NMR spectroscopic experiments, including COSY, HSQC,
NOESY, and ROESY (see the Supporting Information).
The numerical labels in Figure 3 correspond to the labeled
protons and carbon atoms in Figure 1. The presence of only
one set of signals for both the E isomers is, as expected, an
indication of fast rotation of both the rotor and the brake
mation). Our previous DFT calculations on (E)-1NO sug-
gest a rotational barrier as low as 4.45 kcalmolꢀ1.[9] Regard-
ing the Z isomers, the pentiptycene methine nuclei signals
are also of one set for (Z)-1H but become two sets for (Z)-
1Bu. This suggests that rotation of the pentiptycene rotor in
(Z)-1H and (Z)-1Bu at 298 K is at a rate that is faster and
slower, respectively, than the NMR spectroscopic timescale.
The situation with (Z)-1NO resembles the case of (Z)-
1Bu,[9] but (Z)-1OM and (Z)-1Pr lie in between the cases
of (Z)-1H and (Z)-1Bu: namely, the signals for (Z)-1OM
ꢀ
units about the Cvinyl Caryl single bonds. This is also true for
(E)-1OM, (E)-1NO, and (E)-1Pr (see the Supporting Infor-
11596
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 11594 – 11604

Light-Driven Molecular Brakes
FULL PAPER
and (Z)-1Pr are broad and un-
resolved (see the Supporting
Information). This arises from
the fact that at 298 K the rota-
tion rate for the rotor in (Z)-
1OM and (Z)-1Pr is approach-
ing the NMR spectroscopic
timescale. For the nuclei in the
brake unit, there is only one set
of signals for all five com-
pounds at 298 K, thereby indi-
cating that the U-shaped cavi-
ties of the pentiptycene group
provide sufficient space for fast
rotation of the brake compo-
nents. Nevertheless, the rota-
tion rate of the brake unit
should also depend on the sub-
stituents. We will show below
that the rotation of the brake
unit could not be arrested on
the NMR spectroscopic time-
scale for all cases, except for
the one with the bulky tBu sub-
stituents ((Z)-1Bu), even at
temperatures as low as 178 K.
Figure 4 shows the pentipty-
cene region of 1H and 13C NMR
spectra for (Z)-1H at different
temperatures. The pentiptycene
methine signals (those with nu-
merical labels) split into two
sets upon cooling from 298 to 223 K, thereby indicating an
arrest of the rotation of the pentiptycene rotor on the NMR
spectroscopic timescale. The coalescence temperature (Tc)
for C4 and C4’ is near 273 K, which corresponds to an energy
barrier of DG°(273K) =13.7 kcalmolꢀ1. On the basis of line-
shape simulations (Figure 4c), the rotation is nearly blocked
at 223 K, and the rate constant (k) for interconversion be-
tween the two isoenergetic conformers of (Z)-1H is only
2 sꢀ1. In other words, it takes around 1 s for a 3608 rotation
of the rotor (i.e., krot =k/2). The activation energy (Ea) and
pre-exponential factor (log A) and enthalpic (DH°) and en-
tropic (DS°) contributions to the free energy of activation
(DG°) obtained by Arrhenius and Eyring plots (see the Sup-
porting Information) are reported in Table 1.
Figure 4. Pentiptycene peripheral phenylene (blade of the rotor) region of the a) experimental proton,
b) carbon, and c) simulated carbon VT NMR spectra of (Z)-1H (CD2Cl2). Values of temperature (T in K) and
interconversion rate (k in sꢀ1) between the two isoenergetic states are also given for every trace.
Figure 5 shows the pentiptycene region of 1H and
13C NMR spectra for (Z)-1Bu at different temperatures.
Since the rotation of the pentiptycene rotor is slower than
the NMR spectroscopic timescale at 298 K, as indicated by
the presence of two set of signals for the pentiptycene me-
thine nuclei (Figure 3), it requires a higher temperature to
observe coalescence of the signals. Indeed, the peaks of C3
and C3’ coalesce upon raising the temperature to 353 K.
Like the case of (Z)-1H, spectral simulations on the VT
13C NMR spectra (Figure 5c) allow one to retrieve the rate
constants k and thus krot for (Z)-1Bu. The Ea, log A, DH°,
DS°, and DG° values are reported in Table 1.
Table 1 also summarizes the activation parameters for the
rotation of the pentiptycene rotor in (Z)-1OM, (Z)-1NO,
and (Z)-1Pr. Since the five compounds of (Z)-1R differ
Table 1. VT 13C NMR spectroscopic data and activation parameters for the rotation of the pentiptycene rotor in (Z)-1R.
Compound
Solvent
Tc[a] [K]
DG°
Ea
Log A
DH°
DS°
DG°
k(298K)
[sꢀ1
c
(298K)
[kcalmolꢀ1
[kcalmolꢀ1
]
[kcalmolꢀ1
]
[kcalmolꢀ1
]
[calmolꢀ1 Kꢀ1
]
]
N
]
(Z)-1H
CD2Cl2
CD2Cl2
[D6]DMSO
CD2Cl2
[D6]DMSO
[D6]DMSO
273
308(4,4’)
348(3,3’)
G
13.7
14.9
16.9
–
16.9
17.8
10.5ꢁ0.1
11.7ꢁ0.1
14.8ꢁ0.5
13.2ꢁ0.1
14.1ꢁ0.3
14.0ꢁ0.8
10.5ꢁ0.1
10.5ꢁ0.1
11.5ꢁ0.2
10.9ꢁ0.1
11.3ꢁ0.2
10.8ꢁ0.5
10.0ꢁ0.1
11.1ꢁ0.1
14.1ꢁ0.5
12.6ꢁ0.1
13.5ꢁ0.3
13.3ꢁ0.7
ꢀ12.1ꢁ0.4
ꢀ12.3ꢁ0.5
ꢀ7.6ꢁ1.4
ꢀ10.6ꢁ0.3
ꢀ9.0ꢁ0.8
ꢀ11.5ꢁ2.3
13.6ꢁ0.1
14.8ꢁ0.2
16.4ꢁ0.5
15.8ꢁ0.1
16.2ꢁ0.3
16.8ꢁ0.7
686
89
6
17
11
4
(Z)-1OM
(Z)-1NO
(Z)-1Pr
–
343
N
(Z)-1Bu
353(3,3’)
ACHTUNGTRENNUNG
[a] Coalescence temperature for protons with labels (see Figure 1) shown in parentheses.
Chem. Eur. J. 2010, 16, 11594 – 11604
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11597

J.-S. Yang, I. Chao et al.
extent of increase in DG°
(298K)
for (Z)-1H and (Z)-1OM on
changing the solvent from
CD2Cl2 to [D6]DMSO, their
DG°
values in [D6]DMSO
(298K)
would be 14.0 and 15.2 kcal
molꢀ1, respectively. The absence
of
a
linear relationship in
Figure 6 reveals that the sub-
stituent effects on the ground
versus the transition states of
(Z)-1R are different from those
on the equatorial versus axial
conformation of cyclohexane.
Regarding the p-conjugated
nature of (Z)-1R, the electronic
effect of the substituents should
also be important in determin-
ing the energy of (Z)-1R in the
rotational ground and transition
states.
Figure 5. Pentiptycene peripheral phenylene (blade of the rotor) region of the a) experimental proton,
b) carbon, and c) simulated carbon VT NMR spectra of (Z)-1Bu ([D6]DMSO). Values of temperature (T in
K) and interconversion rate (k in sꢀ1) between the two isoenergetic states are also given for every trace.
It is interesting to point out
that among the five (Z)-1R
only in the substituents in the brake component (Figure 1a),
their differences and similarities in rotation kinetics will re-
flect the size and electronic effects of the substituents. In
this context, we adopt the steric parameter A values,[12] de-
fined based on the energy difference (in kcalmolꢀ1) between
substituted cyclohexane in equatorial versus axial positions,
for the discussion of the substituent size effect.
species (Z)-1NO possesses the largest DH° value and the
smallest DS° value (Table 1). One possible explanation is
the presence of specific attractive interactions for (Z)-1NO
in the ground state. In view of the polar nature of the nitro
groups, it could be attributed to electrostatic interactions be-
tween the dinitrophenyl group (the brake component) and
the peripheral phenylene rings of pentiptycene (the blades
of the rotor). However, such interactions (favorable in en-
thalpy) require more specific alignments between the phen-
ylene rings (unfavorable in entropy). This phenomenon is
reminiscent of the concept of enthalpy–entropy compensa-
tion.[14]
The plot of DG°
in [D6]DMSO against the substitu-
(298K)
ent A values for (Z)-1R is shown in Figure 6. Since the ex-
perimental DG°
values were not determined entirely in
(298K)
the same solvent but either in [D6]DMSO or in CD2Cl2, we
need to consider the influence of solvent on DG°
(298K)
before addressing the substituent effects in (Z)-1R. The de-
The relatively broad 1H NMR spectroscopic signals for
the tBu groups in (Z)-1Bu at 298 K prompted us to investi-
gate whether the rotation of the brake unit (denoted by the
brown curved arrows in Figure 1b) could be arrested on the
NMR spectroscopic timescale at low temperatures. Indeed,
termined DG°
value for (Z)-1Pr in [D6]DMSO is
(298K)
0.4 kcalmolꢀ1 higher than in CD2Cl2 (Table 1), which is con-
sistent with its higher viscosity (CH2Cl2: 0.45 MPas,
DMSO: 2.20 MPas).[13] By the assumption of a similar
1
on the basis of the VT H NMR spectra in CD2Cl2 and the
simulated spectra (Figure 7), the interconversion rate con-
stant (k) between the two isoenergetic states is as low as
12 sꢀ1 at 178 K. The coalescence temperature (Tc) for the
two tBu groups is near 238 K, which corresponds to a DG°
c
value of 10.7 kcalmolꢀ1. The Ea and DH° values are (7.0ꢁ
0.1) and (6.7ꢁ0.1) kcalmolꢀ1, respectively, both of which
are only one-half the size for the pentiptycene rotation (14.0
and 13.3 kcalmolꢀ1, Table 1). However, the entropic term
(DS° =(ꢀ15.3ꢁ0.3) calKꢀ1 molꢀ1) that contributes to the ro-
tational barrier is considerably larger than the cases of the
pentiptycene rotation (Table 1). This corresponds to a low
log A value (9.7ꢁ0.1). Consequently, the free energy of acti-
vation at 298 K (DG°(298K)) is significant (11.2 kcalmolꢀ1).
Figure 6. The relationship between the size of substituents (A value) on
the brake component and the rotational free energy of activation (DG°)
for the rotor of (Z)-1R in [D6]DMSO at 298 K.
Computational study: To gain further insight into the sub-
stituent effects on the ground and the transition states of
11598
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 11594 – 11604

Light-Driven Molecular Brakes
FULL PAPER
for (Z)-1H and smaller for the other (Z)-1R species. It is
noted that the Boltzmann distribution treatment narrows
the difference (from 0.35 to 0.19 kcalmolꢀ1) between the
calculated and experimental DG°
(298K)A
1NO. Although the substituent dependence of the calculat-
ed DG°
(298K)A(rotor) is less prominent than that of the experi-
CTHUNGTRENNUNG
mental values, the order (Z)-1H<(Z)-1OM<(Z)-1Pr<
(Z)-1Bu is consistent with the observed trend. The calculat-
ed DG°
(298K)ACHTUNGERTN(NUNG rotor) of (Z)-1NO is close to that of (Z)-1Bu,
similar to what is observed experimentally, but the order is
reversed. Overall, both experimental and DFT results show
that DG°
(298K)ACHTUNGERTN(NUNG rotor) does not correlate linearly with the
substituent A value.
Figure 8 shows the DFT-calculated ground- and transi-
tion-state lowest-energy conformations of (Z)-1NO and
(Z)-1Bu along the rotor rotation coordinate (Figure 2). In
Figure 7. The upfield region d=ꢀ2–3 ppm of the a) experimental and
1
b) simulated VT H NMR spectra of (Z)-1Bu (CD2Cl2, 500 MHz). Values
of temperature (T in K) and interconversion rate (k in sꢀ1) between the
two isoenergetic states are also given for every trace.
(Z)-1R, we have carried out DFT calculations. In the previ-
ous communication,[9] we reported that at the BMK/6-
311+G**//B3LYP/6-31G* theory level[15] the calculated
DG°
value for the rotor (DG°
(298K)ACHTUNGRETNNG(UN rotor)=16.75 kcal
(298K)
molꢀ1) for (Z)-1NO agrees well with the experimental value
(16.4 kcalmolꢀ1 in [D6]DMSO) even though solvation was
not considered in the calculations. Thus, the same theory
level was adopted for the calculation of the other (Z)-1R
compounds. As in the previous study, the octyloxy group
para to the brake unit was replaced by a methoxy group to
expedite the calculations. What is different from the previ-
ous study is that the rotation barriers are estimated on the
basis of free energies of multiple conformers with the con-
sideration of Boltzmann distribution, rather than the lowest-
energy structures of the ground and transition states.[16] The
Figure 8. DFT-derived lowest-energy structures for the ground (G) and
transition (T) states of a) (Z)-1NO and b) (Z)-1Bu along the pentipty-
cene rotation coordinate. Select distances [ꢁ] and angles [8] are shown.
the ground states, the brake component is located in the U-
shaped cavities with the brake phenylene group tilted
toward one of the rotor phenylene rings. As the tBu group
is bulkier than the NO2 group, the extent of tilting is less in
(Z)-1Bu than (Z)-1NO in the lowest-energy structures. In
the transition states, the brake phenylene ring is perpendicu-
lar to the rotor central phenylene ring and oriented above
the V-shaped cavities of the rotor. The steric interactions be-
tween the brake phenylene ring and the bridgehead hydro-
gen atoms of the rotor are significant, as indicated by the
short distances between the ring center and the Hs and by
the increased aryl–vinyl bond angles on going from the
ground state (ca. 1308) to the transition state (ca. 1408). The
substituent effect is also manifested in the transition state by
the comparison of (Z)-1NO to (Z)-1Bu, in which the atom-
difference between the DFT-derived DG°
(298K)ACHTUNGTRNE(NUNG rotor) values
(Table 2) and the observed data (Table 1) is 1.32 kcalmolꢀ1
Table 2. DFT-derived rotational barriers and homodesmotic reaction
free energy[a] for (Z)-1R in the ground and transition states along the
rotor and the brake rotation coordinates at 298 K.
[e]
Compound DGhom(G)[b]
DGhom(T)[c]
DDG°[d]
DG°
(298K)
rotor brake rotor brake rotor brake
(Z)-1H
0.00
1.02
0.29
2.55
2.46
0.00
1.25
1.96
2.89
3.64
0.00
2.89
0.33
3.31
6.26
0.00
0.23
1.67
0.34
1.18
0.00
1.87
0.04
0.76
3.80
14.92
15.15
16.59
15.26
6.77
8.63
6.81
7.53
(Z)-1OM
(Z)-1NO
(Z)-1Pr
(Z)-1Bu
16.10 10.57
[a] In kcalmolꢀ1 calculated at the BMK/6-311+G**//B3LYP/6-31G*
theory level. [b] The (Z)-1H and (Z)-1R in Scheme 2 are in the ground
state. [c] The (Z)-1H and (Z)-1R in Scheme 2 are in the transition state.
[d] DDG° =DGhom(T)ꢀDGhom(G). [e] The free energy of the rotational
barrier at 298 K.
ꢀ
to-ring (i.e., Hbridgehead Phbrake) distance is somewhat larger in
(Z)-1Bu (2.80 ꢁ) than (Z)-1NO (2.73 ꢁ).
The relative substituent effect on the ground versus the
transition states can be evaluated by homodesmotic reac-
Chem. Eur. J. 2010, 16, 11594 – 11604
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11599

J.-S. Yang, I. Chao et al.
tions shown in Scheme 2, in which DGhom is the free energy
of (Z)-1R (disubstituted (Z)-1H) and benzene relative to
(Z)-1H and meta-disubstituted benzene (m-R). When both
Scheme 2. The homodesmotic reaction employed to estimate the influ-
ence of the brake substituents on the stability of the ground and the tran-
sition states of (Z)-1R.
(Z)-1H and (Z)-1R are in the ground state, the DGhom
values, denoted as DGhom(G), allow one to compare the rela-
tive stability of (Z)-1R in their ground states. Likewise, the
corresponding DGhom values in the transition states of the
Figure 9. DFT-derived transition-state structures (top and side views) for
a) (Z)-1NO and b) (Z)-1Bu along the brake rotation coordinate. Select
distances [ꢁ] are shown.
rotor rotation, denoted as DGhom(T)ACHTUNTRGNEUNG(rotor), reveal the rela-
tive stability of (Z)-1R in their transition states along the
rotor rotation coordinate (Figure 2). As summarized in
Table 2, both the DGhom(G) and DGhom(T) rotor values are
positive. Evidently, the substituents destabilize not only the
transition state but also the ground state. It is interesting to
note that the DGhom(T) rotor value increases monotonically
with increasing size of the substituents in the brake compo-
nent, but no explicit order is present for the DGhom(G)
values. The DGhom(G) value for (Z)-1NO is relatively small.
It is possible that the strong dipole of the nitro substituents
lowers the energy of (Z)-1NO by interacting with the pe-
ripheral phenylene rings of pentiptycene,[17] which compen-
sates for the steric repulsion. This is consistent with the ar-
gument based on the DH° and DS° values determined by
NMR spectroscopy (vide supra, Table 1). As a result of the
comparable substituent effects on the ground and the transi-
phenylene rings on the side of the U-shaped cavities, at
ꢀ
which C H···O interactions are present.
Therefore, despite the larger size of NO2 versus H, these
NO2-related attractive interactions counterbalance the in-
creased steric congestion in (Z)-1NO versus (Z)-1H in both
the ground and transition states, which corresponds to the
relatively small DGhom(G) and DGhom(T) brake values (0.29
and 0.33 kcalmolꢀ1, respectively) for (Z)-1NO. In contrast,
the large DGhom(T) brake value (2.89 kcalmolꢀ1) for (Z)-
1OM indicates that the polar methoxy groups do not induce
any explicit attractive interactions with the rotor blades in
the transition state. This is also true for the nonpolar sub-
stituents iPr and tBu (Figure 9b). It is interesting to note
that the DGhom(T) brake values for those of nonpolar sub-
stituents (i.e., (Z)-1H, (Z)-1Pr, and (Z)-1Bu) correlate well
(slope=1.34, R2 =0.99) with the substituent A values (see
the Supporting Information).
tion
A
states
(i.e.,
°A
DDG CTHUNGETNRN(UNG rotor)=DGhom(T)-
(rotor)ꢀDGhom(G)=0.23–1.67 kcalmolꢀ1), the substituent ef-
fects on the rotational barriers of (Z)-1R are smaller than
predicted on the basis of the A values.
Despite the independent internal rotation coordinates for
the rotor and the brake units, the strong steric and/or elec-
tronic interactions between the two units in (Z)-1R suggest
that their rotations are far from completely independent.
For example, formation of the transition states from the
ground states along the rotor rotation coordinate (e.g., (Z)-
1NO in Figure 8a) requires the rotation of not only the
rotor but also the brake unit (Figure 2). Thus, the rotation
dynamics of the brake would affect that of the rotor. An
animated view of these processes has been supplied as Sup-
porting Information.
We have also calculated the rotational barriers and homo-
desmotic reaction free energy for the brake rotation of (Z)-
1R at 298 K (Table 2), since the experimental value of
DG°
good agreement between the DG°
(brake) for (Z)-1Bu is available for comparison. A
brake data derived
(298K)
by DFT (10.57 kcalmolꢀ1) and determined by NMR spec-
troscopy (11.2 kcalmolꢀ1) is again observed. The difference
in the brake rotational barrier between (Z)-1H, (Z)-1OM,
(Z)-1Pr, and (Z)-1Bu suggests that the brake rotation is
also dependent of the substituents. The similar rotational
barriers calculated for (Z)-1NO and (Z)-1H show that the
nitro substituents affect the ground and the transition states
Photoisomerization: The photochemical switching between
the brake-off ((E)-1R) and brake-on ((Z)-1R) states was
conducted in CH2Cl2. The molar ratio of (E)-1R to (Z)-1R
([E]/[Z]) in photostationary states (PSS) depends on the
quantum yields for (E)-1R!(Z)-1R (FEZ) and (Z)-1R!
(E)-1R (FZE) and on the molar absorptivity of (E)-1R (eE)
°A
to a similar extent (DDG CTHNUGTENR(NUG brake)=0.04 for (Z)-1NO in
Table 2). Indeed, as shown in Figure 9a, the transition state
of (Z)-1NO corresponds to a perpendicular orientation of
the brake phenylene ring with respect to the pentiptycene
11600
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 11594 – 11604

Light-Driven Molecular Brakes
FULL PAPER
and (Z)-1R (eZ) at the excitation wavelengths (lex) accord-
molar reversibility (R) upon irradiation at 250 nm is 46%
for 1Pr and 1Bu and slightly lower for 1H and 1OM. The
absorption spectra of 1NO differ from the others by shifting
the peak maxima toward the shorter wavelengths and by ex-
tending the long-wavelength shoulders toward longer wave-
lengths. This could be attributed to the strong electron-with-
drawing nitro groups that lead to a stronger donor–acceptor
(charge-transfer) electronic character. As a result of the re-
duced ratio eE/eZ even at the long-wavelength tail (400 nm),
the efficiency for (E)-1NO!(Z)-1NO switching is signifi-
cantly lower than the other species. This in turn leads to a
low switching capacity of R=20%. The isomerization quan-
tum yields and the [E]/[Z] values in PSS are reported in
Table 3. These values are consistent with Equation (1)
within experimental errors (ca. 10%).
ing to Equation (1):[18]
½Eꢂ=½Zꢂ ¼ ðeZFZEÞ=ðeEFEZ
Þ
ð1Þ
Generally, the values of FEZ and FZE are independent of lex
based on the Kasha–Valivov rule.[19] Thus, control of the
isomer distribution [E]/[Z] mainly relies on the eE/eZ ratio.
To achieve the maximum forward and reversed conversion
of (E)-1R!(Z)-1R and (Z)-1R!(E)-1R, the solutions
were excited at wavelengths of the maximum and the mini-
mum ratio of eE/eZ, respectively. The absorption spectra of
both isomers of 1H are shown in Figure 10 and those of
1NO are in Figure 11. The absorption spectra of 1OM, 1Pr,
and 1Bu resemble those of 1H, with the maximum and min-
imum eE/eZ values near 340 and 250 nm, respectively. In
these cases, the molar absorptivity of the Z isomers is low or
negligible at 340 nm, and thus the Z isomer is predominant
(>98%) in PSS with 340 nm irradiation (Table 3). The
Table 3. Photochemical data for molecular brakes 1R.[a]
[b]
[c]
[d]
[e]
Compound
FEZ
FZE
[E]/[Z]340
[E]/[Z]250
R[f] [%]
1H
0.48
0.39
0.17
0.37
0.28
0.21
1:99
43:57
41:59
45:55
46:54
46:54
43
39
20
46
46
1OM
1NO
1Pr
2:98
25:75[g]
1:99
n.d.[h]
0.49
n.d.[h]
0.38
1Bu
1:99
[a] In CH2Cl2. [b] Quantum yields for (E)-1R!(Z)-1R photoisomeriza-
tion. [c] Quantum yields for (Z)-1R!(E)-1R photoisomerization.
[d] Isomer ratio ([E]/[Z]) in PSS at lex =340 nm. [e] Isomer ratio ([E]/
[Z]) in PSS at lex =250 nm. [f] Molar reversibility (DE or DZ) in PSS
with lex =250 versus 340 nm. [g] lex =400 nm. [h] n.d.=not determined.
The quantum yields of 1Pr are expected to be similar to that for 1Bu in
view of their similar absorption spectra and isomer distribution in PSS.
To test the photostability of both the E and Z isomers of
1R, multiple photoswitching between the two PSS of maxi-
mum and minimum [E]/[Z] ratios have been carried out for
all five compounds. The results for 1H and 1NO are shown
in Figures 10 and 11, respectively, and the corresponding
data for the others are supplied as Supporting Information.
No apparent decomposition was detected after 5–7 switching
cycles for all cases, thereby showing that molecular brakes
1R are quite robust under the operation conditions.
Figure 10. Absorption spectra of (E)-1H (curve a) and (Z)-1H (curve c)
and their photostationary states irradiated with alternating 340 (curves c)
and 250 nm (curves b) UV-light irradiation in dichloromethane. Inset
shows the changes in absorbance at 298 nm starting from (E)-1H (19 mm)
for 6 switching cycles.
Conclusion
The pentiptycene-derived stilbenes 1H, 1OM, 1NO, 1Pr,
and 1Bu display distinct internal rotation rate for the pen-
tiptycene rotor in the trans ((E)-1R) versus the cis ((Z)-1R)
forms, which can be reversibly switched by light with an effi-
ciency of 39–46% in molar ratio. At 298 K, the rotation rate
in (E)-1 is at a magnitude of krot =108–109 sꢀ1 for all five
compounds, but it varies from 2 to 686 sꢀ1 for (Z)-1R, de-
pending on the nature of the substituents in the brake com-
ponent. The substituents exert steric repulsion, electronic in-
duction, and specific attractive electrostatic interactions with
the pentiptycene rotor in the ground state as well as the
transition state of (Z)-1R along the rotation coordinate. It is
remarkable that DFT calculations provide a nice prediction
Figure 11. Absorption spectra of (E)-1NO (curve a) and (Z)-1NO (curve
d) and their photostationary states irradiated with alternating 250 (curves
b) and 400 nm (curves c) UV-light irradiation in dichloromethane. Inset
shows the changes in absorbance at 322 nm starting from (E)-1NO
(19 mm) for 7 switching cycles.
Chem. Eur. J. 2010, 16, 11594 – 11604
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11601

J.-S. Yang, I. Chao et al.
General procedure for the synthesis of 3,5-dialkyl styrenes from 1-
bromo-3,5-dialkylbenzene: Vinylmagnesium bromide (6 mL, 6 mmol)
was slowly added to a mixture of 1-bromo-3,5-diialkylbenzene (2 mmol),
of the substituents effect on the rotation kinetics. Our re-
sults demonstrate not only an efficient and tunable control
of the Brownian rotary motion of molecular rotors by pho-
tons and small structural variations[20,21] but also the poten-
tial utility of the rigid pentiptycene framework[10,22] in con-
structing molecular machines.
[Pd
N
argon. The mixture was kept stirring for 10 min and then heated to reflux
for 12 h. The mixture was cooled, 5% HCl(aq) (1 mL) was added, and
then extracted with CH2Cl2. The organic layer was dried over anhydrous
MgSO4, and the filtrate was concentrated under reduced pressure. Flash
column chromatography with hexane as the eluent afforded the styrene
product. 3,5-diisopropylstyrene: yield=75%; 1H NMR (400 MHz,
CDCl3): d=1.26 (d, J=6.9 Hz, 12H), 2.89 (septet, J=6.9 Hz, 2H), 5.21
(dd, J=10.9, 1.0 Hz, 1H), 5.74 (dd, J=17.6, 1.0 Hz, 1H), 6.72 (dd, J=
17.6, 10.9 Hz, 1H), 6.99 (t, J=1.6 Hz, 1H), 7.10 ppm (d, J=1.6 Hz, 2H);
13C NMR (125 MHz, CD2Cl2): d=24.0, 34.2, 113.2, 121.9, 124.5, 137.4,
137.5, 149.1 ppm; HRMS (FAB): m/z: calcd for C14H20: 188.1565 [M+];
found: 188.1559. 3,5-Di(tert-butyl)styrene: yield=84% yield; 1H NMR
(400 MHz, CD2Cl2): d=1.35 (s, 18H), 5.22 (dd, J=11.2 Hz, 1.2 Hz, 1H),
5.74 (dd, J=17.6 Hz, 1.2 Hz, 1H), 6.75 (dd, J=17.6 Hz, 11.2 Hz, 1H),
7.28 (d, J=1.2 Hz, 2H), 7.36 ppm (t, J=1.2 Hz, 1H).
Experimental Section
General: 1H (500 MHz) and 13C (125 MHz) NMR spectra were acquired
using a Bruker DMX 500 spectrometer with 5 mm gradient triple-reso-
nance broadband inverse (TBI) and triple-resonance broadband observe
(TBO) probes, respectively. The chemical shifts for 1H and 13C spectra
were referenced to the signals of tetramethylsilane (d
(1 H)=0 and d-
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
General Wittig route for the synthesis of 1R (1H, 1OM, 1NO): A mix-
1
suitable delay time (2 and 6 s for H and 13C, respectively). Other spectra
(COSY, NOESY, ROESY, and HSQC) were measured using pulse se-
quences in the Bruker software package. In the case of variable-tempera-
ture measurements, the actual sample temperature was well calibrated by
1H signals of ethylene glycol and methanol so that the temperature error
was assured to be within ꢁ1 K. Signal acquisition was begun after a suffi-
cient temperature equilibration time (10–15 min). Fitting of dynamic
NMR spectroscopic line shapes at different temperatures was performed
with the Topspin 2.0 program, Bruker BioSpin Group. Infrared spectra
were recorded using a Nicolet Magna-IR 550 Spectrometer Series II.
UV/Vis absorption spectra were measured using a Varian Cary300 Bio-
type at room temperature. Photoswitching experiments were conducted
on N2-bubbled solutions (10ꢀ5 m) at selected wavelengths using a 75 W
Xe arc lamp and monochromator. Quantum yields of photoisomerization
were measured on optically dense N2-bubbled solutions (10ꢀ3 m) at
350 nm using a 75 W Xe arc lamp and monochromator. trans-4-(Phenyl-
ture of
2 (0.20 g, 0.34 mmol) and benzylphosphonium chloride salts
(0.69 mmol), K2CO3 (0.94 g, 6.81 mmol), and CH2Cl2 (25 mL) was heated
to reflux for 24 h. The solution was cooled and concentrated under re-
duced pressure, and the residue was dissolved with CH2Cl2 and H2O. The
organic layer was washed with H2O and then dried over anhydrous
MgSO4, and the filtrate was concentrated under reduced pressure. Prepa-
rative HPLC with EA/hexane (v/v=5:95) as the eluent afforded the
product. (E)-1H: yield=55%; m.p. 250–2518C; 1H NMR (500 MHz,
CD2Cl2): d=0.95–0.98 (m, 3H), 1.39–1.42 (m, 4H), 1.46–1.56 (m, 4H),
1.71 (m, 2H), 2.05 (tt, J=6.7, 7.6 Hz, 2H), 3.96 (t, J=6.7 Hz, 2H), 5.73
(s, 2H), 5.78 (s, 2H), 6.74 (d, J=16.4 Hz, 1H), 6.92–6.97 (m, 8H), 7.30–
7.37 (m, 8H), 7.41–7.44 (m, 1H), 7.52–7.55 (m, 2H), 7.57 (d, J=16.4 Hz,
1H), 7.76–7.77 ppm (m, 1H); 13C NMR (125 MHz, CD2Cl2): d=14.5,
23.3, 27.0, 30.0, 30.2, 31.1, 32.6, 48.7, 51.6, 76.8, 124.0, 124.2, 124.4, 125.7,
125.7, 126.9, 127.3, 128.7, 129.4, 135.8, 137.0, 137.9, 143.1, 145.9, 146.2,
149.3 ppm; HRMS (FAB): m/z: calcd for C50H44O: 660.3392 [M+];
found: 660.3394. (Z)-1H: yield=28%; m.p. 98–1018C; 1H NMR
(500 MHz, CD2Cl2): d=0.96–0.98 (m, 3H), 1.39–1.48 (m, 4H), 1.48–1.54
(m, 2H), 1.55–1.56 (m, 2H), 1.70–1.73 (m, 2H), 2.03–2.06 (m, 2H), 3.98
(t, J=6.8 Hz, 2H), 5.46 (s, 2H), 5.69 (s, 2H), 6.80 (d, J=12.1 Hz, 1H),
6.83–6,91 (m, 17H), 7.11 (d, J=12.1 Hz, 1H), 7.31 ppm (d, J=7.5 Hz,
4H); 13C NMR (125 MHz, CD2Cl2): d=14.5, 23.3, 27.0, 30.0, 31.1, 32.2,
32.5, 48.8, 51.9, 76.7, 123.7, 124.1, 125.4, 125.7, 125.9, 134.4, 135.9, 137.3,
142.4, 145.8, 146.0, 149.3 ppm. (E)-1OM: yield=53%; m.p. 254–2558C;
1H NMR (500 MHz, CD2Cl2): d=0.93–0.97 (m, 3H), 1.22–1.32 (m, 2H),
1.36–1.42 (m, 4H), 1.46–1.56 (m, 4H), 1.67–1.74 (m, 2H), 2.01–2.07 (m,
2H), 3.94–3.96 (m, 8H), 5.73 (s, 2H), 5.76 (s, 2H), 6.55 (t, J=2.2 Hz,
1H), 6.64 (d, J=16.4 Hz, 1H), 6.89(t, J=2.2 Hz, 2H), 6.92–6.96 (m, 8H),
7.30–7.35 (m, 8H), 7.55 ppm (d, J=16.4 Hz, 1H); 13C NMR (125 MHz,
CD2Cl2): d=14.5, 23.3, 27.0, 30.0, 30.2, 31.1, 32.2, 32.5, 48.7, 51.6, 56.1,
76.7, 100.3, 105.6, 124.0, 124.1, 124.9, 125.7, 125.7, 126.7, 135.8, 136.9,
139.8, 143.1, 145.9, 146.1, 149.4, 161.9 ppm; HRMS (FAB): m/z: calcd for
C52H48O3: 720.3603 [M+]; found: 720.3602. (Z)-1OM: yield=26%; m.p.
ACHTUNGTRENNUNGamino)stilbene was used as a reference standard (F=0.34 in dichlorome-
thane).[23] The extent of photoisomerization (<10%) was determined
using HPLC analysis (Waters 600 Controller and 996 photodiode array
detector) without back-reaction corrections. The reproducibility error
was <10% of the average. DFT calculations were performed at the
BMK/6-311+G**//B3LYP/6-31G* theory level14] for (Z)-1 in the gas
phase with the consideration of Boltzmann distribution of the possible
conformations of the substituents. To expedite the DFT calculations, the
octyl group in (Z)-1 was replaced by a methyl group for all cases. All the
calculations were performed with the Gaussian 03 package.[24]
Materials: THF and CH2Cl2 were dried with sodium metal and CaH2, re-
spectively, and distilled before use. All the other solvents for spectra and
isomerization quantum-yield measurements were HPLC grade and used
as received. Compounds 2 and 3 were prepared according to the litera-
ture procedures.[10] 3,5-Dialkyl styrenes were prepared by palladium-cata-
lyzed cross-coupling reaction of Grignard reagents with dialkylbromo-
benzene.[25] Synthetic details for compounds (E)-1NO and (Z)-1NO have
been reported.[9]
1
186–1888C; H NMR (500 MHz, CD2Cl2): d=0.97 (m, 3H), 1.40–1.45 (m,
4H), 1.48–1.50 (m, 4H), 1.69–1.73 (m, 2H), 2.02 (quin, J=6.8 Hz, 2H),
2.73 (s, 6H), 3.90 (t, J=6.8 Hz, 2H), 5.50 (s, 2H), 5.69 (s, 2H), 5.86 (d,
J=2.3 Hz, 2H), 5.97 (t, J=2,3 Hz, 1H), 6.68 (br, 4H), 6.89 (d, J=
11.8 Hz, 1H), 6.93 (br, 6H), 7.04 (d, J=11.8 Hz, 1H), 7.30 ppm (br, 6H);
13C NMR (125 MHz, CD2Cl2): d=14.5, 23.3, 27.0, 30.0, 30.2, 31.1, 32.5,
48.8, 51.9, 53.6, 76.6, 101.4, 106.8, 123.3, 123.9, 124.2, 125.4, 126.3, 126.6,
134.7, 136.2, 138.7, 142.5, 145.8, 146.1, 148.9, 160.9 ppm.
Synthesis of 1-(chloromethyl)-3,5-dimethoxybenzene: SOCl2 (0.6 mL,
8.3 mmol) was added to a mixture of 3,5-dimethoxybenzyl alcohol (1.0 g,
5.9 mmol), triethylamine (2 mL), and CH2Cl2 (25 mL) at 08C. The mix-
ture was then heated to reflux for 3 h. The solution was cooled and con-
centrated under reduced pressure, and the residue was extracted with
CH2Cl2 and H2O. The organic layer was dried over anhydrous MgSO4,
and the filtrate was concentrated under reduced pressure. Flash column
chromatography with EA/hexane (v/v=50:50) as the eluent afforded the
product with a yield of 86%. M.p. 46–478C (ref. [26]=468C); 1H NMR
(400 MHz, CDCl3): d=3.80 (s, 6H), 4.52 (t, J=2.8 Hz, 2H), 6.41 (d, J=
2.3 Hz, 1H), 6.54 ppm (s, 2H).
General Heck route for the synthesis of (E)-1R ((E)-1Pr and (E)-1Bu):
A mixture of 3 (50 mg, 0.08 mmol), PdACHTUNTRGENNUG(OAc)2 (6 mg, 0.02 mmol), PACHTUNGTRENNUNG(o-
tolyl)3 (12 mg, 0.04 mmol), 3,5-dialkylstyrene (0.24 mmol), anhydrous
DMF (1.5 mL), and triethylamine (1.5 mL) was heated to 908C for 16 h
under argon. The mixture was concentrated under reduced pressure, and
the residue was dissolved in CH2Cl2 and washed wish brine. The organic
layer was dried over anhydrous MgSO4 and the filtrate was concentrated
under reduced pressure. Column chromatography with CH2Cl2/hexane
General procedure for the synthesis of phosphonium halide salts: A mix-
ture of benzyl chloride (0.5 g, 4.9 mmol), PPh3 (1.4 g, 5.4 mmol), and tolu-
ene (5 mL) was heated at reflux for 24 h. The precipitate was filtered off
and afforded the salt with a yield over 95%.
11602
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 11594 – 11604

Light-Driven Molecular Brakes
FULL PAPER
(v/v=1:4) as the eluent afforded the product. (E)-1Pr: yield=73%;
m.p.>3008C; 1H NMR (500 MHz, CD2Cl2): d=0.95–0.98 (m, 3H), 1.39–
1.41 (m, 14H), 1.45–1.57 (m, 6H), 1.46–1.56 (m, 4H), 1.71 (m, 2H), 2.05
(m, 2H), 3.06 (septet, J=6.9 Hz, 2H), 3.95 (t, J=6.8 Hz, 2H), 5.73 (s,
2H), 5.80 (s, 2H), 6.71 (d, J=16.5 Hz, 1H), 6.93–6.97 (m, 8H), 7.19 (m,
1H), 7.30–7.36 (m, 8H), 7.42 (m, 2H), 7.52–7.55 (m, 2H), 7.54 ppm (d,
J=16.5 Hz, 1H); 13C NMR (125 MHz, CD2Cl2): d=14.5, 23.3, 24.5, 27.0,
30.0, 30.2, 31.1, 32.6, 35.0, 48.7, 51.6, 76.7, 123.1, 123.7, 124.0, 124.1, 125.1,
125.6, 125.7, 127.1, 134.7, 137.6, 137.7, 143.1, 145.9, 146.2, 149.2,
150.3 ppm; HRMS (FAB): m/z: calcd for C56H56O: 744.4331 [M+];
found: 744.4332. (E)-1Bu: yield=76%; m.p.>3008C; 1H NMR
(500 MHz, [D6]DMSO): d=0.88–0.91 (m, 3H), 1.31–1.32 (m, 4H), 1.34–
1.40 (m, 4H), 1.42 (s, 18H), 1.57–1.60 (m, 2H), 1.92–1.95 (m, 2H), 3.92
(t, J=6.7 Hz, 2H), 5.76 (s, 2H), 5.87 (s, 2H), 6.61 (d, J=16.4 Hz, 1H),
6.91–6.95 (m, 8H), 7.37–7.39 (m, 8H), 7.45 (s, 1H), 7.59 (d, J=1.4 Hz,
2H), 7.70 ppm (d, J=16.4 Hz, 1H); 13C NMR (125 MHz, [D6]DMSO):
d=14.0, 22.1, 25.6, 28.7, 28.9, 29.7, 31.4, 34.6, 47.3, 50.0, 75.5, 121.0, 123.4,
123.7, 125.0, 126.3, 134.9, 136.0, 137.1, 141.6, 142.1, 145.1, 145.3, 148.1,
150.8 ppm; HRMS (FAB): m/z: calcd for C58H60O: 772.4644 [M+];
found: 772.4651.
chey, Chem. Eur. J. 2009, 15, 1836–1843; f) A. Comotti, S. Bracco, P.
Valsesia, M. Beretta, P. Sozzani, Angew. Chem. Int. Ed. 2010, 49,
1760–1764.
[3] a) H. Iwamura, K. Mislow, Acc. Chem. Res. 1988, 21, 175–182; b) G.
Yamamoto, Tetrahedron 1990, 46, 2761–2772; c) A. M. Stevens, C. J.
Richards, Tetrahedron Lett. 1997, 38, 7805–7808; d) J. Clayden, J. H.
Pink, Angew. Chem. 1998, 110, 2040–2043; Angew. Chem. Int. Ed.
1998, 37, 1937–1939; e) L. E. Harrington, L. S. Cahill, M. J. McGlin-
chey, Organometallics 2004, 23, 2884–2891; f) I. Eryazici, P. Wang,
C. N. Moorefield, M. Panzer, S. Durmus, C. D. Shreiner, G. R. New-
kome, Dalton Trans. 2007, 626–628; g) W. Setaka, T. Nirengi, C.
Kabuto, M. Kira, J. Am. Chem. Soc. 2008, 130, 15762–15763.
[4] For light-driven molecular shuttles, see: a) H. Murakami, A. Kawa-
buchi, R. Matsumoto, T. Ido, N. Nakashima, J. Am. Chem. Soc.
2005, 127, 15891–15899, and references therein; b) W. Zhou, D.
Chen, J. Li, J. Xu, J. Lv, H. Liu, Y. Li, Org. Lett. 2007, 9, 3929–3932;
c) K. Hirose, Y. Shiba, K. Ishibashi, Y. Doi, Y. Tobe, Chem. Eur. J.
2008, 14, 3427–3433; d) A. Coskun, D. C. Friedman, H. Li, K. Patel,
H. A. Khatib, J. F. Stoddart, J. Am. Chem. Soc. 2009, 131, 2493–
2495.
[5] a) J. P. Sestelo, T. R. Kelly, Appl. Phys.
A 2002, 75, 337–343;
General procedure for the synthesis of (Z)-1Pr and (Z)-1Bu: An N2-
purged solution of (E)-1R in dichloromethane (1 mm) was irradiated
with 352 nm in a Rayonet photochemical reactor for 30 min. The solvent
was removed under reduced pressure. Preparative HPLC with EA/
hexane (v/v=5:95) as the eluent afforded the product. (Z)-1Pr: yield=
98%; m.p.>86–888C; 1H NMR (500 MHz, [D6]DMSO): d=0.45 (d, J=
6.9 Hz, 12H), 0.89–0.92 (m, 3H), 1.33–1.41 (m, 8H), 1.55 (m, 2H), 1.91–
1.94 (m, 2H), 2.24 (septet, J=6.9 Hz, 2H), 3.89 (t, J=7.0 Hz, 2H), 5.44
(s, 2H), 5.70 (s, 2H), 6.36 (s, 2H), 6.51–6.52 (m, 3H), 6.56 (m, 2H), 6.79
(m, 2H), 6.85 (d, J=11.7 Hz, 1H), 6.90–6.91 (m, 4H), 7.13 (d, J=
11.7 Hz, 1H), 7.27–7.28 (m, 2H), 7.33 ppm (m, 4H); 13C NMR (125 MHz,
[D6]DMSO): d=14.0, 22.1, 23.0, 25.7, 28.8, 28.9, 29.7, 31.4, 32.8, 47.4,
50.6, 75.5, 122.7, 123.2, 123.3, 123.6, 123.7, 124.4, 124.4, 124.6, 124.8,
125.7, 134.4, 134.8, 135.8, 141.3, 144.7, 145.1, 145.2, 145.4, 147.7,
147.9 ppm; HRMS (FAB): m/z: calcd for C56H56O: 744.4331 [M+];
found: 744.4334. (Z)-1Bu: yield=99%; m.p. 116–1208C; 1H NMR
(500 MHz, [D6]DMSO): d=0.63 (br, 18H), 0.89–0.92 (m, 3H), 1.33–1.45
(m, 8H), 1.54–1.57 (m, 2H), 1.90–1.96 (m, 2H), 3.88 (t, J=6.9 Hz, 2H),
5.44 (s, 2H), 5.69 (s, 2H), 6.49–6.56 (m, 5H), 6.79 (t, J=7.2 Hz, 2H),
6.84–6.93 (m, 5H), 7.16 (d, J=11.8 Hz, 1H), 7.28 (d, J=7.2 Hz, 2H),
7.31 ppm (m, 3H); 13C NMR (125 MHz, [D6]DMSO): d=14.0, 22.1, 25.6,
28.7, 28.9, 29.7, 30.5, 31.4, 33.6, 75.4, 120.7, 122.8, 122.9, 123.1, 123.3,
123.6, 124.4, 124.4, 124.4, 124.8, 124.8, 125.7, 134.8, 135.1, 141.3, 144.7,
145.1, 145.2, 145.5, 147.9, 149.5 ppm; HRMS (FAB): m/z: calcd for
C58H60O: 772.4644 [M+]; found: 772.4638.
b) M. F. Hawthorne, J. I. Zink, J. M. Skelton, M. J. Bayer, C. Liu, E.
Livshits, R. Baer, D. Neuhauser, Science 2004, 303, 1849–1851;
c) B. L. Feringa, J. Org. Chem. 2007, 72, 6635–6652; d) M. M. Pol-
lard, M. Lubomska, P. Rudolf, B. L. Feringa, Angew. Chem. 2007,
119, 1300–1302; Angew. Chem. Int. Ed. 2007, 46, 1278–1280; e) G.
London, G. T. Carroll, T. F. Landaluce, M. M. Pollard, P. Rudolfb,
B. L. Feringa, Chem. Commun. 2009, 1712–1714; f) E. M. Geertse-
ma, S. J. van der Molen, M. Martens, B. L. Feringa, Proc. Natl. Acad.
Sci. USA 2009, 106, 16919–16924.
[6] a) J. E. Walker, Angew. Chem. 1998, 110, 2438–2450; Angew. Chem.
Int. Ed. 1998, 37, 2308–2319; b) R. D. Vale, R. A. Milligan, Science
2000, 288, 88–95; c) K. Kinbara, T. Aida, Chem. Rev. 2005, 105,
1377–1400.
[7] B. Champin, P. Mobian, J.-P. Sauvage, Chem. Soc. Rev. 2007, 36,
358–366.
[8] a) F. M. Raymo, Angew. Chem. 2006, 118, 5375–5377; Angew.
Chem. Int. Ed. 2006, 45, 5249–5251; b) A. Credi, Aust. J. Chem.
2006, 59, 157–169; c) S. Saha, J. F. Stoddart, Chem. Soc. Rev. 2007,
36, 77–92.
[9] J.-S. Yang, Y.-T. Huang, J.-H. Ho, W.-T. Sun, H.-H. Huang, Y.-C.
Lin, S.-J. Huang, S.-L. Huang, H.-F. Lu, I. Chao, Org. Lett. 2008, 10,
2279–2282.
[10] J.-S. Yang, C.-W. Ko, J. Org. Chem. 2006, 71, 844–847.
[11] V. Diemer, H. Chaumeil, A. Defoin, A. Fort, A. Boeglin, C. Carrꢃ,
Eur. J. Org. Chem. 2006, 2727–2738.
[12] E. L. Eliel, S. H. Wilen, L. N. Mander, Stereochemistry of Organic
Compounds, Wiley, New York, 1994.
[13] Handbook of Chemistry and Physics (Ed.: D. R. Lide), CRC Press,
Baoca Raton, 1997.
Acknowledgements
[14] L. Liu, Q-X. Guo, Chem. Rev. 2001, 101, 673–695.
[15] Y. Zhao, D. G. Truhlar, Acc. Chem. Res. 2008, 41, 157–167.
[16] To search for conformers, the following bonds are rotated: the Csp2
Financial support for this research was provided by the National Science
Council and Academia Sinica of Taiwan, ROC. The computing time
granted by the National Center for High-Performance Computing and
the Computing Center of Academia Sinica is acknowledged.
ꢀ
Osp3 bond of the methoxy group para to the brake in all (Z)-1R, the
ꢀ
Csp2 Csp3 bonds of the substituents (tert-butyl and isopropyl) on the
ꢀ
brake unit of (Z)-1Pr and (Z)-1Bu, and the Csp2 Osp3 bond of the
brake unit of (Z)-1OM. There are two ground-state conformers for
(Z)-1H and (Z)-1NO, but it increases to eight unique conformers
for (Z)-1OM, (Z)-1Pr, and (Z)-1Bu. For the transition state of pen-
tiptycene rotation, there is one conformer for (Z)-1H and (Z)-1NO
and four conformers for (Z)-1OM, (Z)-1Pr, and (Z)-1Bu. For the
brake rotation, there are two transition-state conformers for (Z)-
1H, (Z)-1Bu and (Z)-1NO, and eight for (Z)-1OM and (Z)-1Pr. In
the homodesmotic calculations, there are three unique conformers
for meta-dimethoxy (m-OM), meta-diisopropyl (m-Pr), and meta-di-
tert-butyl (m-Bu) benzenes (with one conformer with a degeneracy
of 2), and one conformer for benzene and meta-dinitrobenzene (m-
NO). More detailed results and discussion of the DFT calculations
will be reported elsewhere.
[1] a) V. Balzani, A. Credi, F. M. Raymo, J. F. Stoddart, Angew. Chem.
2000, 112, 3484–3530; Angew. Chem. Int. Ed. 2000, 39, 3348–3391;
b) G. S. Kottas, L. I. Clarke, D. Horinek, J. Michl, Chem. Rev. 2005,
105, 1281–1376; c) E. R. Kay, D. A. Leigh, F. Zerbetto, Angew.
Chem. 2007, 119, 72–196; Angew. Chem. Int. Ed. 2007, 46, 72–191.
[2] a) T. R. Kelly, M. C. Bowyer, K. V. Bhaskar, D. Bebbington, A.
Garcia, F. Lang, M. H. Kim, M. P. Jette, J. Am. Chem. Soc. 1994,
116, 3657–3658; b) P. V. Jog, R. E. Brown, D. K. Bates, J. Org.
Chem. 2003, 68, 8240–8243; c) R. Annunziata, M. Benzglia, M. Cin-
quini, L. Raimondi, F. Cozzi, J. Phys. Org. Chem. 2004, 17, 749–751;
d) I. Alfonso, M. I. Burguete, S. V. Luis, J. Org. Chem. 2006, 71,
2242–2250; e) K. Nikitin, H. Mꢂller-Bunz, Y. Ortin, M. J. McMlin-
Chem. Eur. J. 2010, 16, 11594 – 11604
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11603

J.-S. Yang, I. Chao et al.
[17] The DFT-optimized structure of (Z)-1NO and the Mulliken charges
obtained at the BMK/6-311+G** level were used for single-point
energy evaluation with the Drieding force field. In the ground state,
one of the nitro groups was pointing toward the rotor (termed in-
nitro) and the other (termed out-nitro) was not. When the atomic
charges within the in-nitro group were assigned as zero, the calculat-
[24] Gaussian 03, Revision D.02, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
ery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.
Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cio-
slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-
mi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wall-
ingford CT, 2004.
ed energy was raised by 3.7 kcalmolꢀ1
.
[18] H. Gçrner, H. J. Kuhn, Adv. Photochem. 1995, 21, 1–117.
[19] a) H. Meier, Angew. Chem. 1992, 104, 1425–1446; Angew. Chem.
Int. Ed. Engl. 1992, 31, 1399–1420; b) P. Klꢄn, J. Wirz, Photochemis-
try of Organic Compounds: From Concepts to Practice, Wiley, New
York, 2009.
[20] a) A. M. Schoevaars, W. Kruizinga, R. W. J. Zijlstra, N. Veldman,
A. L. Spek, B. L. Feringa, J. Org. Chem. 1997, 62, 4943–4948;
b) M. K. J. ter Wiel, R. A. van Delden, A. Meetsma, B. L. Feringa,
Org. Biomol. Chem. 2005, 3, 4071–4076; c) M. K. J. ter Wiel, B. L.
Feringa, Tetrahedron 2009, 65, 4332–4339.
[21] M. C. Basheer, Y. Oka, M. Mathews, N. Tamaoki, Chem. Eur. J.
2010, 16, 3489–3496.
[22] a) J.-S. Yang, J.-L. Yan, Chem. Commun. 2008, 1501–1512; b) J.-S.
Yang, J.-L. Yan, Y.-X. Jin, W.-T. Sun, M.-T. Yang, Org. Lett. 2009,
11, 1429–1432.
[25] N. A. Bumagin, E. V. Luzikova, J. Organomet. Chem. 1997, 532,
271–273.
[26] R. Adams, Jr., S. Mackenzie, S. Loewe, J. Am. Chem. Soc. 1948, 70,
664–668.
[23] J.-S. Yang, K.-L. Liau, C.-M. Wang, C.-Y. Hwang, J. Am. Chem. Soc.
2004, 126, 12325–12335.
Received: March 26, 2010
Published online: September 8, 2010
11604
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 11594 – 11604