Dynamic Polymer Free Volume Monitored by Single-Molecule Spectroscopy of a Dual Fluorescent Flapping Dopant

Article abstract of DOI:10.1021/jacs.1c06428

Single-molecule spectroscopy (SMS) of a dual fluorescent flapping molecular probe (N-FLAP) enabled real-time nanoscale monitoring of local free volume dynamics in polystyrenes. The SMS study was realized by structural improvement of a previously reported

Full text of DOI:10.1021/jacs.1c06428

pubs.acs.org/JACS
Article
Dynamic Polymer Free Volume Monitored by Single-Molecule
Spectroscopy of a Dual Fluorescent Flapping Dopant
Yuma Goto, Shun Omagari, Ryuma Sato, Takuya Yamakado, Ryo Achiwa, Nilanjan Dey, Kensuke Suga,
Martin Vacha,* and Shohei Saito*
Cite This: https://doi.org/10.1021/jacs.1c06428
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Single-molecule spectroscopy (SMS) of a dual
fluorescent flapping molecular probe (N-FLAP) enabled real-
time nanoscale monitoring of local free volume dynamics in
polystyrenes. The SMS study was realized by structural improve-
ment of a previously reported flapping molecule by nitrogen
substitution, leading to increased brightness (22 times) of the
probe. In a polystyrene thin film at the temperature of 5 K above
the glass transition, the spectra of a single N-FLAP molecule
undergo frequent jumps between short- and long-wavelength
forms, the latter one indicating planarization of the molecule in the
excited state. The observed spectral jumps were statistically
analyzed to reveal the dynamics of the molecular environment.
The analysis together with MD and QM/MM calculations show
that the excited-state planarization of the flapping probe occurs only when sufficiently large polymer free volume of more than, at
least, 280 Å3 is available close to the molecule, and that such free volume lasts for an average of 1.2 s.
temperature (Tg),12−14 to characterize diffusion in solutions or
INTRODUCTION
■
melts,15,16 or to monitor polymerization reactions on
Polymer free volume is a practical concept for interpreting
molecular level.17,18 In particular, the SMS method could be
polymer properties connected to the fraction of unoccupied
an attractive alternative to tracking the dynamics of the
volume.1 While arbitrary definition of “free volume” often led
polymer free volume because of its high location specificity,
to controversy, a quantitative evaluation of polymer free
suitable dynamic range, noninvasive nature, and relatively
volume is important for developing membrane materials as
affordable and simple optical instrumentation. Here, we
well as packing materials for chromatography that adsorb,
succeeded in demonstrating such free-volume monitoring by
diffuse, and separate small molecules.2 Local free volume
developing a new flexible fluorophore, nitrogen-embedded
moves from moment to moment due to thermal fluctuation of
flapping molecule (N-FLAP), which shows dual fluorescence
surrounding polymer chains. To evaluate the free volume
(FL) spectrum depending on the bent/planar conformations
directly and quantitatively, several approaches have been
(Figure 1). In comparison with the previously reported
established such as positron annihilation lifetime spectroscopy
flapping anthracene,19 the nitrogen-embedded molecular
(PALS),3 inverse gas chromatography (IGC),4 129Xe NMR
framework led to much improved FL brightness as well as
spectroscopy,5 and molecular probe.6−8 In recent years, some
high photostability, when excited by a visible light. With this
studies have been reported which compare experimental results
unique molecular probe physically doped in polystyrenes, SMS
with molecular dynamics (MD) simulation.9 However, real-
study has been performed to demonstrate that dynamic
time monitoring of the local free-volume dynamics has only
nanoscale changes of polymer free volume can be monitored in
been achieved by single-molecule spectroscopy (SMS) analysis
real time by following local environment-dependent spectral
of conformationally flexible probes.10 In this literature, single-
changes of a single N-FLAP molecule. This is the first
molecule fluorescence lifetime has been analyzed because the
conformational change of the fluorescent probe, twisted
peryleneimide, results in very little spectral difference. This
Received: June 21, 2021
work is an example of the great potential that the SMS
technique holds for spatially resolved dynamic studies of
polymers and soft matter in general.11 Apart from free-volume
probing, SMS has been used, e.g., to analyze heterogeneity of
local relaxation processes of polymers near glass transition
https://doi.org/10.1021/jacs.1c06428
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
© XXXX American Chemical Society
A

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 1. (a) Nitrogen-embedded flapping fluorophore (N-FLAP) as
a bright dual-fluorescent probe and (b) real-time analysis of
polystyrene free volume using the N-FLAP dopant. λF: Fluorescence
wavelength.
application of a dual emissive molecule20 to the free-volume
monitoring in a single-molecule level.
EXPERIMENTAL SECTION
■
Synthesis and Characterization of N-FLAP Probe. A
previously reported flapping anthracene (FLAP in Figure 3a) works
as a fluorescent viscosity probe that shows polarity-independent dual
FL.19 The flapping anthracene undergoes a bent-to-planar conforma-
tional change in the lowest singlet excited state (S1), while it instantly
returns to the most stable bent form in the singlet ground state (S0).
Since the S1 planarization is partially suppressed in a viscous media,
ratiometric FL analysis is available for quantifying local viscosity. The
flapping anthracene has only weak absorption bands in the visible
range, and low photostability has been demonstrated when excited by
a strong UV light.21 This photodegradation behavior is a serious
shortcoming for the application to the single-molecule FL analysis, in
which continuous FL emission is required for counting photon
signals.22 Particularly for the spectroscopic study, bright FL as well as
the high photostability is expected for accumulating photons with
respective energies. Here, we designed a flapping phenazine (N-
FLAP) probe, in which reactive 9 and 10 positions of the original
anthracene moieties were capped by nitrogen atoms to suppress
undesired photoreactions.
Figure 3. (a) Absorption and FL spectra of N-FLAP, compared with
those of the previously reported flapping anthracene (FLAP) in
CH2Cl2. Relative area of the FL spectra is proportional to the
brightness of the probes. (b) Solvent dependence (εr: relative
permittivity) and (c) environment dependence in the FL spectra of
N-FLAP. Excitation wavelength λex = 365 nm. (d) Calculated energy
profiles of N-FLAP′ in the excited state (S1) and the ground state
(S0). With the fixed COT bending angle θ (see inset), structural
optimizations were performed by (TD-)DFT calculations at the
TPSSh/6-31+G(d) level.
N-FLAP was newly prepared by the Pd-catalyzed phenazine
synthesis.23 The coupling reaction of compounds 124 and 2 followed
by an in situ oxidation provided N-FLAP in 45% yield (Figure 2a).
increased more than 10-fold (5.8 × 104 M−1 cm−1) compared with
that of the flapping anthracene (4.1 × 103 M−1 cm−1 at 398 nm). The
difference in the molar absorption coefficients was supported by time-
dependent density functional theory (TD-DFT) calculations. A model
structure, N-FLAP′, was treated for the calculations, in which the
terminal OC6H13 substituents of N-FLAP were replaced by OCH3
groups. Strong visible absorption of an allowed transition (f ≈ 0.5) is
expected for N-FLAP′, while only a weak visible absorption with small
oscillator strength (f ≈ 0.06) is assigned for FLAP (Figure S17). In
addition, molecular orbital (MO) energy levels originating from the
phenazine wings of N-FLAP′ become lower than those of the
anthracene wings of FLAP by the nitrogen substitution. The lower
LUMO level may lead to the suppression of photo-oxidation in N-
FLAP.
N-FLAP showed a green fluorescence peaked at 518 nm with a
large Stokes shift of 4800 cm−1 in CH2Cl2, indicating a conforma-
tional planarization in the lowest singlet excited state (S1), as seen in
other flapping anthracenes (Figure 3a). Absolute fluorescence
quantum yield of N-FLAP (ΦF = 0.42) was higher than that of
FLAP (ΦF = 0.27).25 Thus, the brightness of N-FLAP, defined as ε ×
ΦF, is 22 times higher than that of FLAP (Figure 3a, inset table),
overcoming the drawbacks of the previous FLAP probe for the SMS
use. Polarity independence in the FL spectra of N-FLAP was
confirmed in various solvents (toluene, CH2Cl2, acetone, DMF, and
DMSO) covering a wide range of relative permittivity (εr) from 2.4
(toluene) to 46.7 (DMSO) at room temperature. The peak
wavelength of the FL band shifted within a small range of 514−523
nm (Figure 3b), suggesting a negligible charge-transfer character in
the excited state. When dispersed in a rigid poly(methyl methacrylate)
(PMMA) matrix, N-FLAP emitted a blue FL (λF = 460 nm, ΦF =
Figure 2. (a) Synthesis of N-FLAP. SPhos Pd G2: A commercially
available second generation SPhos precatalyst. (b) Single crystal X-ray
structure analysis. Thermal ellipsoids were drawn at the 50%
probability level. Hydrogen atoms were omitted for clarity.
Single-crystal X-ray diffraction analysis revealed a V-shaped structure
of N-FLAP with the cyclooctatetraene (COT) bending angles,
defined in Figure 3d, of θ = 36.5° and 39.2° (Figure 2b).
Optical Properties of N-FLAP Probe. Absorption bands of N-
FLAP were red-shifted (415 nm in CH2Cl2) in comparison with those
of the previously reported flapping anthracene (Figure 3a).
Remarkably, the molar absorption coefficient (ε) in the visible region
B
https://doi.org/10.1021/jacs.1c06428
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
0.13) with smaller Stokes shift of 2360 cm−1 (Figure 3c). This drastic
FL chromism was also observed with increasing viscosity of a 2-
methyltetrahydrofuran solution at lower temperatures (Figure S7d).
These behaviors indicate that the conformational planarization of N-
FLAP in S1 is suppressed in the rigid (highly viscous) environments.
To support the explanation of conformational dynamics in
S1, (TD-)DFT calculations were conducted. Although the calculation
results are very sensitive to the selected levels of theory (SI, Section
4.1), here, we chose the TPSSh/6-31+G(d) level for better
interpreting the experimental results. Potential energy profiles of N-
FLAP′ were delineated with changing the COT bending angle θ by 2
degrees in the ground state (S0) and the excited state (S1) (Figure
3d). In S0, the optimized structure of N-FLAP′ takes a V-shaped form
(θ = 38.8°), while the planar structure gives a transition state. In
contrast, the most stable geometry in S1 is the planar conformation (θ
= 0°) at which the S0−S1 transition is allowed (f ≈ 0.3). The
calculation results rationalize the observed green bright FL with a
large Stokes shift (Figure S15e). In addition, an electronic
configuration switch was predicted during the conformational
planarization in S1, as was the case with reported FLAP.19,25
Thermal and Viscoelastic Properties of Host Polymers. High
molecular-weight polystyrene (high MW PS; Mw ≈ 192000) and low
MW PS (Mw ≈ 850) were selected for this study. Glass transition
temperatures (Tg) of high MW PS and low MW PS were determined
to be 102 and 18 °C, respectively, based on inflection points of the
DSC profiles in the cooling process. Dynamic viscoelasticity
measurements showed that the polystyrenes hardened at around Tg,
and the storage modulus (G′) was 0.9 × 109 Pa (65 °C and below) for
high MW PS and 7 × 107 Pa (25 °C) for low MW PS (Figure 4).
for the high and low MW PS can be calculated as 13 and 0.9 nm,
while the length of the bent N-FLAP′ conformation is about 2 nm.
Due to the small molecular weight of the low MW PS (which can
be even regarded as a mixture of oligostyrenes), the end groups would
play a significant role. Then, we analyzed the commercially available
low MW PS sample by NMR spectroscopies and high-resolution
APCI (atmospheric pressure chemical ionization) mass spectrometry
(see the details in the final chapter of the SI). In conclusion, we have
reasonably determined the terminal structures as a sec-butyl group and
a hydrogen atom, which are probably attached in the process of anion
polymerization.28 On the basis of the analyses, tacticity of the
oligostyrene was found to be atactic, and the oligomer sample is
mainly composed of the 6−8 mers. As a result, no specific interaction
between the terminal end groups and N-FLAP is expected in the
following study.
Methods of Single-Molecule Fluorescence Spectroscopy.
For single-molecule FL studies, toluene solutions of the host PS
polymers and a trace of N-FLAP were prepared (the concentration of
N-FLAP: 10−9 M) and then spin-coated to form ca. 100 nm thick
sample films on glass substrates. After annealing the PS films by
heating above Tg for approximately 30 min, we carried out the SMS
experiments at 23 °C in a N2 atmosphere on top of an inverted FL
microscope equipped with an imaging spectrograph (Figure 5a).
Figure 5. (a) Scheme of the experimental setup. (b) An example of
microscopic FL image, in which the bright spots correspond to the N-
FLAP single molecules dispersed in a spin-coated polystyrene thin
film. (c) An example of single-molecule FL spectrum of the N-FLAP
probe.
Switching to the spectral mode of the spectrograph enables
wavelength dispersion of FL signal from individual molecules, the
dynamics of which can be then monitored for extended periods of
time. Figure 5b,c show a typical FL image of the dispersed N-FLAP
molecules and an example of a single-molecule FL spectrum,
respectively.
Figure 4. (a) DSC profiles and (b) temperature-dependent dynamic
viscoelasticity of the host polymers, high MW PS (left) and low MW
PS (right). Scan rate: 5 °C/min for both measurements. G′: storage
modulus. G″: loss modulus.
RESULTS AND DISCUSSION
■
Results of Single-Molecule Fluorescence Spectrosco-
py. SMS of N-FLAP in the high MW PS was performed first.
As mentioned above, the high MW PS has its Tg at 102 °C
(Figure 4) and represents rigid environment for the N-FLAP
molecule. An example of time evolution of single-molecule FL
spectrum in the high MW PS is shown in Figure S25. At the
measurement temperature of 23 °C (much lower than Tg), the
molecule showed a spectrally stable emission with a FL peak
(0−0 vibrational band) at 445 nm. The single-molecule
spectrum has a clearly resolved vibronic structure, while
ensembles of N-FLAP molecules in bulk PS films provided
only averaged broad structureless spectrum mainly in a blue
The molecular weight between entanglement points (Me) has been
reported as 18700 for polystyrene melt.26 Therefore, the high MW PS
is expected to have ca. 10 entanglement points per a single polymer
chain, while the low MW PS has no entanglement point. The Me value
of polystyrene corresponds to 180 monomer units and thus 360
covalent bonds in the polymer main chain. According to the freely
rotating chain (FRC)27 model, the expected distance between
entangled points can be roughly estimated as 4.1 nm (
=0.154 nm × 2 × 360). Here, 0.154 nm is a saturated C−C
covalent bond length. In addition, the expected end-to-end distances
C
https://doi.org/10.1021/jacs.1c06428
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
region (420−520 nm) (Figure S25), This behavior is typical
for most of the N-FLAP molecules trapped in the high MW
PS, indicating almost complete suppression of the conforma-
tional planarization of N-FLAP in S1.
Next, SMS in the low MW PS was conducted. Since the
measurement temperature of 23 °C is slightly above its Tg (18
°C), the PS chains start undergoing segmental relaxation in the
form of micro-Brownian motion. The dynamic nature of the
polymer matrix is reflected in the spectroscopic behavior of the
individual N-FLAP molecular probes. A typical example of the
time evolution plot is shown in Figure 6a. This molecule stayed
solution of PS, there is a good correspondence between the
single-molecule and ensemble spectral regions both for the
blue and green FL bands. Regarding the ensemble spectra in
low MW PS matrixes; see Figure S11.
Discussion for Dynamic Single-Molecule Fluores-
cence. The appearance of the green FL band of the flapping
probe in solution is a consequence of the low viscosity
environment which enables the planarization of the chromo-
phore in S1.19 In the solid state, the flapping molecule requires
certain free space for the planarization to happen.29 In
polymers, this free space should be closely related to the
concept of free volume,1 although there is a non-negligible
difference between the “free volume” values estimated by
molecular probes and by PALS.10 (PALS studies generally
report smaller volumes than those reported by molecular probe
studies.) In a polymer film above the glass transition
temperature Tg, segmental relaxation causes dynamical changes
to the free volume which can repeatedly form and disappear
around the N-FLAP probe. This temporary appearance of the
free volume would enable the excited-state planarization of N-
FLAP and emission of the green band, and its disappearance
would cause a return to the blue emission (Figure 7a).
Figure 7. Two possible mechanisms for the single-molecule spectral
dynamics in Figure 6. (a) One possibility is that the excited-state
planarization of N-FLAP is allowed only when the free volume
temporarily appeared around the N-FLAP probe. (b) The other is
that compulsory planarization is mechanically induced by dynamic
segmental motions of the surrounding polymer chains.
Figure 6. (a) Time evolution for the FL spectra of a single N-FLAP
molecule dispersed in the low MW PS matrix. The vertical axis
indicates measurement time (from top to bottom). Each horizontal
line represents a FL spectrum with the signal intensity shown in gray
scale. Numbers (1−6) are assigned for each time range and (b) the
corresponding FL spectra, fitted with a sum of three Gaussian
functions. (c) Histogram of the single-molecule FL peak wavelengths
analyzed from 105 individual spectra. FL peaks at shorter wavelengths
are shown as blue bars, while longer ones counted as green bars.
Ensemble FL spectra of N-FLAP in a drop-cast high MW PS film
(blue solid line) and in a PS-containing toluene solution (green solid
line) were also shown.
Alternatively, the segmental micro-Brownian motion of the
surrounding polymer chains itself could also cause mechanical
planarization of N-FLAP in the ground state, which would also
result in the green emission from the planarized N-FLAP in a
confined space (Figure 7b). To distinguish between these two
possible mechanisms, i.e., existence of a free volume vs
segmental motion-induced planarization, we carried out
control experiments in both high and low MW PS using
another flapping dye, which has been reported as a flexible
mechanophore that does not undergo spontaneous excited-
state planarization even in solution.30 The flexible mechano-
phore (A-FLAP) has bulky triisopropylsilyl (TIPS) ethynyl
groups introduced on the anthracene moieties (Figure S27),
and when mechanically compressed in a crystalline phase it
attains the planar conformation and emits red-shifted
fluorescence. Calculated energy barrier for the mechanical
planarization of A-FLAP in S0 is practically same as that of N-
FLAP. The SMS results of A-FLAP in both high and low MW
PS showed that, irrespective of the matrix, the single-molecule
spectra are of the short-wavelength form, namely, the bent
conformation of A-FLAP is preserved (Figure S27). This result
in the blue form most of the time with the peak position at 464
nm, but occasionally undergoes a spectral jump to the green
form, reminiscent of the solution-phase FL spectra with a peak
top around 520 nm (Figure 3b). In other N-FLAP molecules
(Figure S26), the spectral jumps occurred with varying
frequency. There is a wide distribution of times for which
the molecule stays in the green form. The observed FL peak
wavelengths also showed a large variance in the green range
(490−540 nm). Such variety of the green FL spectra is evident
even for the same N-FLAP molecule in different time ranges
(see the spectra 1, 3, and 6 in Figure 6b). The spectral
distribution is characterized by plotting a histogram of the FL
peak positions based on the analysis of 105 spectra (Figure 6c).
When overlaid with ensemble FL spectra of N-FLAP molecules
dispersed in a rigid high MW PS matrix and in a toluene
D
https://doi.org/10.1021/jacs.1c06428
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
means that even in the rubbery-state PS film above Tg the
segmental motion does not cause mechanical planarization of
the A-FLAP or N-FLAP dyes, and that the observed green FL
of N-FLAP is emitted via the excited-state planarization during
the temporal appearance of the local free volume in the low
MW PS film (Figure 7a).
To get an insight into the dynamics of free volume
formation and disappearance, we further analyzed the spectral
time traces of single N-FLAP molecules in the low MW PS.
Specifically, we evaluated the time intervals for which the
molecule stayed in the blue form before jumping to the green
form and, separately, the intervals for which the molecule
stayed in the green form before reversing back to the blue one.
Histograms of both intervals (Figure 8) show exponential
distributions which, when fitted with single-exponential
functions, provide the characteristic duration times of the
blue form (τb) of 10.4 0.5 s and of the green form (τg) of 1.2
0.1 s. In the picture of polymer free volume, these results
show that it takes an average of 10.4 s for a sufficiently large
free volume to form around the N-FLAP molecule, and that
this volume lasts for an average of 1.2 s before it disappears. It
is only during the presence of the sufficient free volume that
the N-FLAP molecule can undergo repeated planarization after
each excitation and emit the green fluorescence continuously
for several seconds. We have estimated that the minimum free
volume required for the N-FLAP planarization is ca. 280 Å3
(green part in Figure 8e; see also Figure S32). Here, the
volume swept out during planarization has been estimated.
The green part is composed of the terminal part swept out by
the flapping wings and the central part swept out by the
planarized COT. From the superimposed model of the bent
and planar structures with the center of gravity fixed, the
terminal and central parts were roughly estimated as 170 and
110 Å3, respectively. The sum of these two parts gave 280 Å3.
This estimate is comparable with upper limits of distributions
of free volume in PS measured by positron annihilation31 or
fluorescence photochromic probes,6 although the existence of
the N-FLAP probe itself could have an influence for the
estimation of polymer free volume and therefore the result
should be carefully compared with other measurement
methods. The large difference between the two characteristic
times τb and τg reflects the formation and disappearance
mechanisms of the large free volume. The formation is
associated with cooperative motion of several segments of
chains surrounding the molecule. However, once the free
volume is formed, a motion of a single segment is sufficient for
its destruction. The τg should therefore reflect the segmental
relaxation time (α-relaxation) whereas the τb is longer due to
the cooperative nature of the formation process. The τg of 1.2 s
is well within the range of α-relaxation times measured by
dielectric spectroscopy for PS with a varying range of
molecular weight around the Tg temperature.32
It is worth noting that the time scale of the photophysical
excitation events is much shorter than the segmental motion of
the polymer chain. Considering the photon density of the laser,
the average period between these excitation events per
molecule has been estimated to be ca. 0.2 ms. Further, it
takes 300−500 ms to accumulate enough fluorescence signal to
obtain fluorescence spectra (such as the individual frames in
Figure 6). Free volume change occurs on the time scale of the
segmental motion (several seconds). On the basis of the above,
we propose that the segmental motion determines the local
free volume around the probe, and the fast structural dynamics
of the excited N-FLAP is dependent on the degree of the local
free volume. During the periods of sufficiently large free
volume, N-FLAP repeats the fast photophysical cycle of (1) S0
to S1 excitation at the bent form, (2) spontaneous planarization
in S1, (3) fluorescence emission (or nonradiative decay) giving
the planar form in S0, and (4) spontaneous bending in S0.
Thus, the long-wavelength spectrum is observed. By contrast,
during the period of small free volume, excited N-FLAP cannot
planarize in S1 and therefore the accumulated fluorescence
signal of the bent conformation gives the short-wavelength
spectrum.
Figure 8. (a) Time evolution of single-molecule FL spectra of N-
FLAP in the low MW PS; the shaded areas indicate the short (blue)
and long (green) wavelength forms. (b) Histogram of time duration
of the blue spectral form before switching to the green one,
constructed from 90 data points of different 21 molecules and fitted
with an exponential. (c) Histogram of time duration of the green
spectral form before switching to the blue one, constructed from 84
data points of different 21 molecules and fitted with an exponential.
(d) Histogram of reorientation times of single PDI molecules in the
low MW PS constructed from 955 events and fitted with an
exponential. Inset shows an example of defocused FL image. Overall,
73 out of 130 PDI molecules measured in the low MW PS showed the
rotational motion. PDI: N,N′-di-n-octyl-3,4,9,10-perylenetetracarbox-
ylic diimide. (e) Estimated minimum volume required for the excited-
state planarization of N-FLAP′. In the space-filling models, terminal
alkyl chains were not considered because conformationally flexible
moieties generally do not disturb dynamic motions of rigid
chromophores.33
There is also a way of directly measuring the local relaxation
times by monitoring reorientation of single-molecule fluo-
rescent probes doped in the film.13,14,34 As a reference, we have
carried out such measurement using a perylene diimide (PDI)
dye and the technique of defocused imaging35,36 (Figure 8d).
However, rather than analyzing the reorientation process by
plotting the autocorrelation function and fitting it with the
E
https://doi.org/10.1021/jacs.1c06428
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
stretched exponential (KWW) function,10,13 we chose to plot a
distribution of reorientation times for a statistical sample of
single molecules. For this purpose, reorientation time is
defined as the time for which the molecule stays in a certain
orientation before reorientation, by more than 30°, occurs.
Since such reorientation supposedly reflects a segmental
motion of a nearby PS chain, this analysis enables direct
comparison with the distribution of the τg times obtained for
the N-FLAP molecules. The results (presented also in Figure
S28) show that the distribution of reorientation times is also
exponential, and fitting with a single exponential function
provides a characteristic reorientation time τr of 0.75 s. This
value is generally in very good agreement with the 1.2 s of τg,
and indicates that the lifetime of the free volume is indeed
determined by the segmental motion.
While both τb and τg show overall exponential distributions,
there is an interesting trend observed from traces of different
individual molecules in Figure S26. At least within the
observation time of 160 s, there appears to be a kind of local
memory effect in spectral traces, i.e., a molecule which tends to
have short τg and long τb shows this behavior throughout the
observation time, and the same is true for long τg and moderate
τb (Figure S26b,d, respectively). This observation points to
spatial heterogeneity of the α-relaxation process that has been
observed before in poly(n-butyl methacrylate) (PnBMA) near
Tg10 as well as in supercooled liquids37 and that persists for
minutes of the measurement times. The effect of glass/polymer
interface and polymer film surface with air are the most
probable origins of the observed memory effect. For example, a
thin immobile layer at the interface and thin liquid-like layer at
the polymer film surface has been reported,14 which would be
consistent with the observations of very short or very long
green emitting forms, respectively. Similarly, surface mobile
layer was used to explain the observed diffusion.38
We note that, in principle, the observed spectral dynamics of
the N-FLAP dye could be alternatively explained by assuming
that the dye-surrounding polymer becomes temporarily liquid-
like, enabling the excited-state planarization in the same way as
in a solution. However, such local liquid-like state is not very
probable at or near Tg. In addition, it should also be observable
in the PDI reorientation experiment where it would lead to
periods of fast rotation, none of which has been observed. This
mechanism of green-emission formation would be more
relevant at higher temperatures close to the melting point.
MD Simulations and QM/MM Calculations. To validate
the proposed mechanism, we also performed calculations
considering the host polymer (500 entangled chains of the low
MW PS). First, classical MD simulations of the low MW PS
with a single N-FLAP′ molecule were conducted to obtain a
PS geometry that contains the sufficient void space for the
flapping motion of N-FLAP′. While such the geometry was not
obtained in the time range of 1000 ns when simulated at 25
°C, those geometries were generated by the simulation at high
temperature (see SI for details). Then, the N-FLAP′ molecule
with the fixed COT bending angle was placed in the void
space, and additional MD calculations were conducted only for
the PS chains that were spatially overlapped with N-FLAP′.
After that, constraint geometry optimization was performed
with the QM/MM39,40 (quantum mechanics/molecular
mechanics) calculations (Figure 9a). The PS geometry was
fixed and treated with the MM model, while the electronic
structure of N-FLAP′ was considered with the QM model.
After several trials at different calculation levels, it was found
Figure 9. (a) Structural model for the QM/MM calculations, in
which N-FLAP′ is placed at the center of the low MW PS (500
polymer chains). Green void (calculated by POVME 2.041)
corresponds to the free volume around the N-FLAP′ probe. (b)
Relative energy of N-FLAP′(S1)@Low MW PS in the ONIOM (TD-
CAM-B3LYP/6-31+G(d):MM) calculations.
that the ONIOM (TD-CAM-B3LYP/6-31+G(d):MM) calcu-
lation provided a reasonable interpretation of the observed
single-molecule FL behavior of N-FLAP in the low MW PS.
When the total energy of N-FLAP′(S1)@Low MW PS was
plotted by changing the COT bending angle θ of N-FLAP′
every 5 degree, local energy minima were obtained at which N-
FLAP′ takes an almost planar geometry (θ = 5°) and a bent
geometry (θ = 30°) (Figure 9b). Due to the confined space in
the host polymer, completely planar N-FLAP′(S1) geometry
(most relaxed in the gas phase) did not give an energy
minimum of N-FLAP′(S1)@Low MW PS. This result
suggested that the relaxed N-FLAP′(S1) geometry depends
on the specific shape of the temporarily appeared void space,
and therefore the spectral form of the green FL band have a
structural variety (Figure 6 and Figure S24). The predicted
energy barrier between bent and planar conformations of N-
FLAP′(S1) in the polymer is also consistent with the clear FL
switch between the blue and green forms, rather than a gradual
shift of the FL peak.
It should be noted that the calculated volume (green void)
in Figure 9a includes the so-called excess free volume
(Vfree:exs)1 in addition to the volume swept out during the
planarization (280 Å3; green part in Figure 8e). Since the
excess free volume is spatially continuous, we cut it off by
setting the radius of 10 Å from the center of N-FLAP′ (see the
calculation method in the SI). We also note that the hard-core
van der Waals volume of N-FLAP′ has been calculated as 455
Å3 and it was almost constant (454−456 Å3) regardless of the
COT bending angle (θ = 0−50°). The hard-core volume is not
included in the green void.
CONCLUSIONS
■
A nitrogen-embedded flapping molecule has been developed as
a superior environment-sensitive dual fluorescent probe for
single-molecule fluorescence spectroscopy. Nitrogen substitu-
tion on the flapping wings enabled efficient visible excitation,
leading to high brightness (22 times) of the probe. A
suppression of the photo-oxidation process also results in
long-term observation in the single molecule study. The probe
in the low molecular weight polystyrene (Tg = 18 °C) provided
dynamic single-molecule fluorescence spectrum at 23 °C that
frequently jumps to the long-wavelength form, originating from
F
https://doi.org/10.1021/jacs.1c06428
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
the temporarily permitted excited-state planarization dynamics
of the flapping probe in sufficiently large polymer free volume.
In this way, the probe is capable of visualizing local free volume
changes on the order of 280 Å3 on subsecond time scales for
extended periods of time. Statistical analysis of the duration
time in the short- and long-wavelength spectra demonstrated
that it takes 10.4 0.5 s on average for the large free volume to
Nilanjan Dey − Graduate School of Science, Kyoto University,
Sakyo-ku, Kyoto 606-8502, Japan; Present
Address: Department of Chemistry, Birla Institute of
Technology and Sciences-Pilani, Hyderabad, Telangana
500078, India.; orcid.org/0000-0002-3988-1509
Kensuke Suga − Graduate School of Science, Kyoto University,
Sakyo-ku, Kyoto 606-8502, Japan
generate, while it lasts for 1.2
0.1 s on average. MD
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.1c06428
simulations and the following QM/MM calculations, in which
the probe molecule was treated as the QM model while
polystyrene chains were treated as the MM model, supported
the dynamic spectral behaviors and the rough estimation of the
free volume. The use of single-molecule imaging techniques
can, in principle, also provide 3-dimensional localization of the
probe and enable position-dependent studies of the free
volume dynamics. The parallelization of wide-field and spectral
imaging would be a next challenge, as it has been only partially
realized.42
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
JST PRESTO (FRONTIER) and JST FOREST Grant
Numbers JPMJPR16P6 (S.S) and JPMJFR201L (S.S.), JSPS
KAKENHI Grant Numbers JP21H01917 (S.S), JP19H02684
(M.V), and JSPS Fellowships Grant Numbers JP19J22034
(T.Y) and JP18F18335 (N.D), and Toray Science Foundation
(S.S). We thank Prof. Hideki Yorimitsu (Kyoto University) for
his help with the high-resolution mass spectrometry measure-
ment. R.S. was supported by a special postdoctoral researcher
program at RIKEN. The computations were performed at the
Research Center for Computational Science (RCCS) in the
Institute of Molecular Science (Okazaki, Japan) and the
HOKUSAI (RIKEN).
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.1c06428.
Experimental and theoretical details (PDF)
Accession Codes
CCDC 1987186 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
via www.ccdc.cam.ac.uk/data_request/cif, or by emailing
data_request@ccdc.cam.ac.uk, or by contacting The Cam-
bridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: + 44 1223 336033.
REFERENCES
■
(1) White, R. P.; Lipson, J. E. G. Polymer Free Volume and Its
Connection to the Glass Transition. Macromolecules 2016, 49, 3987−
4007.
(2) Low, Z.-X.; Budd, P. M.; McKeown, N. B.; Patterson, D. A. Gas
Permeation Properties, Physical Aging, and Its Mitigation in High
Free Volume Glassy Polymers. Chem. Rev. 2018, 118, 5871−5911.
(3) Dlubek, G.; Pointeck, J.; Bondarenko, V.; Pompe, G.; Taesler,
C.; Petters, K.; Krause-Rehberg, R. Positron Annihilation Lifetime
Spectroscopy (PALS) for Interdiffusion Studies in Disperse Blends of
Compatible Polymers: A Quantitative Analysis. Macromolecules 2002,
35, 6313−6323.
AUTHOR INFORMATION
■
Corresponding Authors
Shohei Saito − Graduate School of Science, Kyoto University,
Sakyo-ku, Kyoto 606-8502, Japan; orcid.org/0000-0003-
1863-9956; Email: s_saito@kuchem.kyoto-u.ac.jp
Martin Vacha − Department of Materials Science and
Engineering, School of Materials and Chemical Engineering,
Tokyo Institute of Technology, Meguro-ku, Tokyo 152-8552,
Japan; orcid.org/0000-0002-5729-9774;
(4) Yampolskii, Y. P.; Kaliuzhnyi, N. E.; Durgarjan, S. G.
Thermodynamics of Sorption in Glassy Poly(vinyltrimethylsilane).
Macromolecules 1986, 19, 846−850.
(5) Golemme, G.; Nagy, J. B.; Fonseca, A.; Algieri, C.; Yampolskii, Y.
129Xe-NMR study of free volume in amorphous perfluorinated
polymers: comparsion with other methods. Polymer 2003, 44,
5039−5045.
(6) Victor, J. G.; Torkelson, J. M. On Measuring the Distribution of
Local Free Volume in Glassy Polymers by Photochromic and
Fluorescence Techniques. Macromolecules 1987, 20, 2241−2250.
(7) Itagaki, H.; Horie, K.; Mita, I. Luminescent probe studies of the
microstructure and mobility of solid polymers. Prog. Polym. Sci. 1990,
15, 361−424.
Email: vacha.m.aa@m.titech.ac.jp
Authors
Yuma Goto − Department of Materials Science and
Engineering, School of Materials and Chemical Engineering,
Tokyo Institute of Technology, Meguro-ku, Tokyo 152-8552,
Japan
Shun Omagari − Department of Materials Science and
Engineering, School of Materials and Chemical Engineering,
Tokyo Institute of Technology, Meguro-ku, Tokyo 152-8552,
Japan
Ryuma Sato − Cellular and Molecular Biotechnology Research
Institute, National Institute of Advanced Industrial Science
and Technology (AIST), Koto-ku, Tokyo 135-0064, Japan;
orcid.org/0000-0003-2523-8966
Takuya Yamakado − Graduate School of Science, Kyoto
University, Sakyo-ku, Kyoto 606-8502, Japan; orcid.org/
0000-0002-0730-2904
(8) Strehmel, B.; Strehmel, V.; Younes, M. Fluorescence Probes for
Investigation of Epoxy Systems and Monitoring of Crosslinking
Processes. J. Polym. Sci., Part B: Polym. Phys. 1999, 37, 1367−1386.
(9) Jansen, J. C.; Macchione, M.; Tocci, E.; De Lorenzo, L.;
Yampolskii, Y. P.; Sanfirova, O.; Shantarovich, V. P.; Heuchel, M.;
Hofmann, D.; Drioli, E. Comparative Study of Different Probing
Techniques for the Analysis of the Free Volume Distribution in
Amorphous Glassy Perfluoropolymers. Macromolecules 2009, 42,
7589−7604.
(10) Vallée, R. A. L.; Cotlet, M.; Van Der Auweraer, M.; Hofkens, J.;
̈
Mullen, K.; De Schryver, F. C. Single-Molecule Conformations Probe
Free Volume in Polymers. J. Am. Chem. Soc. 2004, 126, 2296−2297.
Ryo Achiwa − Graduate School of Science, Kyoto University,
Sakyo-ku, Kyoto 606-8502, Japan
G
https://doi.org/10.1021/jacs.1c06428
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(11) Wöll, D.; Braeken, E.; Deres, A.; De Schryver, F. C.; Uji-I, H.;
Hofkens, J. Polymers and Single Molecule Fluorescence Spectroscopy,
What Can We Learn? Chem. Soc. Rev. 2009, 38, 313−328.
(31) Liu, J.; Deng, Q.; Jean, Y. C. Free-Volume Distributions of
Polystyrene Probed by Positron Annihilation: Comparison with Free-
Volume Theories. Macromolecules 1993, 26, 7149−7155.
(32) Roland, C. M.; Casalini, R. Temperature Dependence of Local
Segmental Motion in Polystyrene and Its Variation with Molecular
Weight. J. Chem. Phys. 2003, 119, 1838−1842.
(12) Deres, A.; Floudas, G. A.; Mullen, K.; Van der Auweraer, M.;
̈
De Schryver, F. C.; Enderlein, J.; Uji-i, H.; Hofkens, J. The Origin of
Heterogeneity of Polymer Dynamics near the Glass Temperature as
Probed by Defocused Imaging. Macromolecules 2011, 44, 9703−9709.
(13) Paeng, K.; Kaufman, L. J. Single Molecule Experiments Reveal
the Dynamic Heterogeneity and Exchange Time Scales of Polystyrene
near the Glass Transition. Macromolecules 2016, 49, 2876−2885.
(14) Oba, T.; Vacha, M. Relaxation in Thin Polymer Films Mapped
across the Film Thickness by Astigmatic Single-Molecule Imaging.
ACS Macro Lett. 2012, 1, 784−788.
(15) Habuchi, S.; Sato, N.; Yamamoto, T.; Tezuka, Y.; Vacha, M.
Multimode Diffusion of Ring Polymer Molecules Revealed by a
Single-Molecule Study. Angew. Chem., Int. Ed. 2010, 49, 1418−1421.
(16) Ito, S.; Itoh, K.; Pramanik, S.; Takatsugu, K.; Takei, S.;
Miyasaka, H. Evaluation of Diffusion Coefficient in a Dextrin-Based
Photo-Curable Material by Single Molecule Tracking. Appl. Phys.
Express 2009, 2, 075004.
(33) Chen, J.; Kistemaker, J. C. M.; Robertus, J.; Feringa, B. L.
Molecular Stirrers in Action. J. Am. Chem. Soc. 2014, 136, 14924−
14932.
(34) Schroeyers, W.; Vallée, R.; Patra, D.; Hofkens, J.; Habuchi, S.;
̈
Vosch, T.; Cotlet, M.; Mullen, K.; Enderlein, J.; De Schryver, F. C.
Fluorescence Lifetimes and Emission Patterns Probe the 3D
Orientation of the Emitting Chromophore in a Multichromophoric
System. J. Am. Chem. Soc. 2004, 126, 14310−14311.
(35) Böhmer, M.; Enderlein, J. Orientation Imaging of Single
Molecules by Wide-field Epifluorescence Microscopy. J. Opt. Soc. Am.
B 2003, 20, 554−559.
(36) Patra, D.; Gregor, I.; Enderlein, J. Image Analysis of Defocused
Single-Molecule Images for Three-Dimensional Molecule Orientation
Studies. J. Phys. Chem. A 2004, 108, 6836−6841.
(37) Zondervan, R.; Kulzer, F.; Berkhout, G. C. G.; Orrit, M. Local
Viscosity of Supercooled Glycerol near Tg Probed by Rotational
Diffusion of Ensembles and Single Dye Molecules. Proc. Natl. Acad.
Sci. U. S. A. 2007, 104, 12628−12633.
(17) Wöll, D.; Uji-i, H.; Schnitzler, T.; Hotta, J.; Dedecker, P.;
Herrmann, A.; De Schryver, F. C.; Mullen, K.; Hofkens, J. Radical
̈
Polymerization Tracked by Single Molecule Spectroscopy. Angew.
Chem., Int. Ed. 2008, 47, 783−787.
(18) Ueda, K.; Aizawa, M.; Shishido, A.; Vacha, M. Real-Time
Molecular-Level Visualization of Mass Flow during Patterned
Photopolymerization of Liquid-Crystalline Monomers. NPG Asia
Mater. 2021, 13, 25.
(19) Kotani, R.; Sotome, H.; Okajima, H.; Yokoyama, S.; Nakaike,
Y.; Kashiwagi, A.; Mori, C.; Nakada, Y.; Yamaguchi, S.; Osuka, A.;
Sakamoto, A.; Miyasaka, H.; Saito, S. Flapping Viscosity Probe That
Shows Polarity-Independent Ratiometric Fluorescence. J. Mater.
Chem. C 2017, 5, 5248−5256.
(20) Behera, S. K.; Park, S. Y.; Gierschner, J. Dual Emission: Classes,
Mechanisms and Conditions. Angew. Chem., Int. Ed. 2021, 9789.
(21) Kimura, R.; Kitakado, H.; Osuka, A.; Saito, S. Flapping
Peryleneimide as a Fluorescent Viscosity Probe: Comparison with
BODIPY and DCVJ Molecular Rotors. Bull. Chem. Soc. Jpn. 2020, 93,
1102−1106.
(38) Flier, B. M. I.; Baier, M. C.; Huber, J.; Mullen, K.; Mecking, S.;
̈
Zumbusch, A.; Wöll, D. Heterogeneous Diffusion in Thin Polymer
Films as Observed by High-Temperature Single-Molecule Fluores-
cence Microscopy. J. Am. Chem. Soc. 2012, 134, 480−488.
(39) Lin, H.; Truhlar, D. G. QM/MM: what have we learned, where
are we, and where do we go from here? Theor. Chem. Acc. 2007, 117,
185−199.
(40) Senn, H.; Thiel, W. QM/MM methods for biomolecular
systems. Angew. Chem., Int. Ed. 2009, 48, 1198−1229.
(41) Durrant, J. D.; Votapka, L.; Sørensen, J.; Amaro, R. E. POVME
2.0: An Enhanced Tool for Determining Pocket Shape and Volume
Characteristics. J. Chem. Theory Comput. 2014, 10, 5047−5056.
(42) Zhang, Z.; Kenny, S. J.; Hauser, M.; Li, W.; Xu, K. Ultrahigh-
Throughput Single-Molecule Spectroscopy and Spectrally Resolved
Super-resolution Microscopy. Nat. Methods 2015, 12, 935−938.
(22) Yang, S. K.; Shi, X.; Park, S.; Ha, T.; Zimmerman, S. C. A
Dendritic Single-Molecule Fluorescent Probe That Is Monovalent,
Photostable and Minimally Blinking. Nat. Chem. 2013, 5, 692−697.
(23) Laha, J. K.; Tummalapalli, K. S. S.; Gupta, A. Palladium-
Catalyzed Domino Double N-Arylations (Inter- and Intramolecular)
of 1,2-Diamino(hetero)arenes with o,o>-Dihalo(hetero)arenes for the
Synthesis of Phenazines and Pyridoquinoxalines. Eur. J. Org. Chem.
2013, 2013, 8330−8335.
(24) Kimura, R.; Kuramochi, H.; Liu, P.; Yamakado, T.; Osuka, A.;
Tahara, T.; Saito, S. Flapping Peryleneimide as a Fluorogenic Dye
with High Photostability and Strong Visible-Light Absorption. Angew.
Chem., Int. Ed. 2020, 59, 16430−16435.
(25) Yamakado, T.; Takahashi, S.; Watanabe, K.; Matsumoto, Y.;
Osuka, A.; Saito, S. Conformational Planarization versus Singlet
Fission: Distinct Excited-State Dynamics of Cyclooctatetraene-Fused
Acene Dimers. Angew. Chem., Int. Ed. 2018, 57, 5438−5443.
(26) Wu, S. Chain Structure and Entanglement. J. Polym. Sci., Part B:
Polym. Phys. 1989, 27, 723−741.
(27) Livadaru, L.; Netz, R. R.; Kreuzer, H. J. Stretching Response of
Discrete Semiflexible Polymers. Macromolecules 2003, 36, 3732−3744.
(28) Quirk, R. Anionic Polymerization. In Handbook of Polymer
Synthesis, Characterization, and Processing; Saldívar-Guerra, E.,
Vivaldo-Lima, E., Eds.; John Wiley & Sons: 2013; pp 127−162.
(29) Kotani, R.; Liu, L.; Kumar, P.; Kuramochi, H.; Tahara, T.; Liu,
P.; Osuka, A.; Karadakov, P. B.; Saito, S. Controlling the S1 Energy
Profile by Tuning Excited-State Aromaticity. J. Am. Chem. Soc. 2020,
142, 14985−14992.
(30) Yamakado, T.; Otsubo, K.; Osuka, A.; Saito, S. Compression of
a Flapping Mechanophore Accompanied by Thermal Void Collapse
in a Crystalline Phase. J. Am. Chem. Soc. 2018, 140, 6245−6248.
H
https://doi.org/10.1021/jacs.1c06428
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX