The actuation of micro- and nanostructures controlled by external stimuli remains one of the exciting challenges in nanotechnology due to the wealth of fundamental questions and potential applications in energy harvesting, robotics, sensing, biomedicine, and tunable metamaterials. Photoactuation utilizes the conversion of light into motion through reversible chemical and physical processes and enables remote and spatiotemporal control of the actuation. Here, we report a fast light-to-motion conversion in few-nanometer thick bare polydopamine (PDA) membranes stimulated by visible light. Light-induced heating of PDA leads to desorption of water molecules and contraction of membranes in less than 140 μs. Switching off the light leads to a spontaneous expansion in less than 20 ms due to heat dissipation and water adsorption. Our findings demonstrate that pristine PDA membranes are multiresponsive materials that can be harnessed as robust building blocks for soft, micro-, and nanoscale actuators stimulated by light, temperature, and moisture level.
The actuation of micro- and nanostructures controlled by external stimuli remains one of the exciting challenges in nanotechnology due to the wealth of fundamental questions and potential applications in energy harvesting, robotics, sensing, biomedicine, and tunable metamaterials. Photoactuation utilizes the conversion of light into motion through reversible chemical and physical processes and enables remote and spatiotemporal control of the actuation. Here, we report a fast light-to-motion conversion in few-nanometer thick bare polydopamine (PDA) membranes stimulated by visible light. Light-induced heating of PDA leads to desorption of water molecules and contraction of membranes in less than 140 μs. Switching off the light leads to a spontaneous expansion in less than 20 ms due to heat dissipation and water adsorption. Our findings demonstrate that pristine PDA membranes are multiresponsive materials that can be harnessed as robust building blocks for soft, micro-, and nanoscale actuators stimulated by light, temperature, and moisture level.
A skeletal muscle tissue,
which has been optimized over hundreds
of millions of years of evolution, represents the role model for the
design and synthesis of soft materials serving as artificial muscles
and actuators. Notably, even the most complex motion results from
simple contractions of antagonistic muscle pairs transferred to the
skeleton via tendons.[1] From a mechanical
point of view, muscle fiber contraction is more favorable as the expansion
can lead to instability and fiber buckling.[2] The actuators efficiency and versatility require discernible displacement
at low energy input[3] and multiresponsive
behavior,[4] respectively. Other crucial
factors include their operational temperature that needs to be minimized
to avoid degradation, as well as their response time and weight-lifting
ability.[5] The external stimuli for actuation
include heat, light, electric fields and currents, changes in humidity,
and exposure to vapors. Light is advantageous over other stimuli for
remote and spatiotemporal control of the actuation.The search
for photoresponsive materials for artificial muscles
mimicking their natural counterparts faces several trade-offs between
light-to-motion conversion efficiency, light wavelength, dynamics,
biocompatibility, operational temperature and environment, flexibility,
multifunctionality, simplicity, and cost, to name a few.[6−9] To date, the vast majority of photoactuators employed photochemical
and photothermal effects.[10] In photochemical
materials, light-activated molecular level transformations (e.g.,
cis–trans isomerization, ring-opening and ring-closing, bond
exchange, cycloadditions) can lead to macroscopic changes of dimensions
or shape. Light-induced heating in photothermal materials can result
in positive or negative thermal expansion, phase transitions, or adsorption/desorption
of molecules, which are converted into mechanical motion. Typically,
photothermal actuators merge light-absorbing heaters (e.g., dyes and
semiconducting or plasmonic nanostructures) with thermoresponsive
materials such as liquid crystal elastomers, hydrogels, shape memory
polymers, and inorganic compounds with volume-changing structural
phase transitions.[9,11−14]In this work, we explore
potential photochemical and photothermal
effects in polydopamine (PDA). PDA is a multifunctional, bioinspired
polymer with diverse applications for biomedical[15,16] and environmental[17] purposes, catalysis[18,19] and photocatalysis,[20,21] sensing,[22] photonics[23] and optoelectronics.[24] Notably, PDA has excellent photothermal properties
over the entire visible spectrum,[25] similar
as the closely related analogues of the melanine family,[26,27] and remains structurally stable up to 400 K.[28] These features were utilized in composite photoactuators
combining PDA heaters (nanoparticles, thin films) with thermoresponsive
polymers.[29−31] Furthermore, the hydrophilic[28] and water-swelling properties of PDA-reduced graphene oxide (PDA-RGO)
composites were harnessed in bilayers or thin films converting moisture
gradients or near-IR irradiation into motion.[32,33] Nevertheless, these composite structures operate due to multistep
processes or involve additional materials serving only as a mechanical
scaffold. Accordingly, they are relatively slow with response times
in the range of seconds to minutes.[29,30,32,33]Here, we demonstrate
for the first time fast light-to-motion conversion
in ultrathin, pristine PDA membranes fabricaded by electropolymerization.
We show that bare PDA membranes exhibit submillisecond contractions
triggered by light and spontaneous expansion in dark conditions within
milliseconds. These features are essential for building bottom-up,
soft matter, ultrafast, nano-, and microphotoactuators. We have studied
this phenomenon by investigating the effect of light, temperature,
and moisture level with optical microscopy, reflectivity, Brillouin
light scattering, and sum-frequency-generation spectroscopy. Our findings
reveal that the contraction/expansion results from the desorption/adsorption
of water molecules from/to the membranes.
Results and Discussion
Polydopamine
Membranes
PDA films with homogeneous surface
roughness (about 2.8 nm), thickness of about t =
15 nm, and Young modulus of about E = 12 GPa were
prepared by electropolymerization.[34] The
PDA films were polymerized on a gold electrode surface using cyclic
voltammetry, and the potential ranged between ±0.5 V at a scan
rate of 2 mV/s. This relatively slow potential sweep has been set
to promote the formation of the hydroxyl indole-like lammelar structure.[35] The chemical reactions leading to PDA formation
are shown in Figure a. After incubation in carbonate buffer to increase the cross-linking
density, the film was desorbed from the gold surface through an electrochemical
removal cycle[34] and mechanically stripped
from the surface using a sacrificial layer of poly(vinyl alcohol)
(PVA). The film was transferred and suspended over circular holes
of 60 μm diameter in a 1 μm thick Si3N4 membrane. After the film transfer, the PVA layer was removed
by dissolution in water. Such a method allows the fabrication of ultrathin
films in a homogeneous and reproducible fashion. Although ultrathin
PDA films have been already prepared by different approaches,[36,37] the advantage of our method is the preparation of large area membranes
(possible up to mm scale) with outstanding mechanical properties and
the tunability of the film properties by the synthesis parameters,
such as scan rate (V/s) and ending potential. More details about sample
preparation and characterization can be found in the Supporting Information 1 (SI 1) and elsewhere.[34]Figure b shows optical images of the Si3N4 membrane
grid covered with PDA film and the morphology of the PDA membranes.
Figure 1
(a) The
synthesis procedure of PDA with hydroxyl indole-like structures.[34] (b) Optical images of the Si3N4 membrane grid covered with the PDA film (upper panel) and
a free-standing PDA membrane (lower panel). The scale bar is 10 μm.
(a) The
synthesis procedure of PDA with hydroxyl indole-like structures.[34] (b) Optical images of the Si3N4 membrane grid covered with the PDA film (upper panel) and
a free-standing PDA membrane (lower panel). The scale bar is 10 μm.
Photoactuation
The photoactuation
of the PDA membranes
has been triggered and observed with the experimental apparatus shown
in Figure a, which
is based on two continuous-wave (CW) lasers and a white light source.
The deformation of the membrane was periodically stimulated with 660
nm CW laser light with 100 mW maximum power. The spot diameter of
the stimulating red laser light matched the membrane size. At this
wavelength, the measured light absorption of the PDA membrane (15
nm thick) was 4.4%. Taking into account the membrane thickness, the
measured absorption of the PDA membranes is in accordance with the
optical[38] and photothermal[25−27] properties of PDA (SI 2).
Figure 2
(a) Scheme of the experimental
setup for the observation of light,
temperature, or pressure-induced membrane actuation. The sample is
mounted in a temperature- and pressure-controlled microscope stage
and becomes illuminated with a red (660 nm) laser light that triggers
the photoactuation. The laser light spot size approximately matches
the membrane area. The change of the membrane morphology is visualized
by a CCD camera. Reflectivity measurements using a low power green
(532 nm) laser light are used to study fast morphological changes
of the membrane. Symbols: L1, L2, L3, L4, lenses; F, optical filter;
BS, beamsplitter; A, aperture; MO, microscope objective; CW, continuous
wave. (b) Optical images of a PDA membrane subjected to light-induced
contraction at ambient conditions. The power of the incident red laser
light is 30 mW. The red laser light in the ON state is not visible
due to the optical filter F. (c) The photoactuated state (laser ON)
of the PDA membrane at 50 and 60 mW laser power (upper and lower panels,
respectively).
(a) Scheme of the experimental
setup for the observation of light,
temperature, or pressure-induced membrane actuation. The sample is
mounted in a temperature- and pressure-controlled microscope stage
and becomes illuminated with a red (660 nm) laser light that triggers
the photoactuation. The laser light spot size approximately matches
the membrane area. The change of the membrane morphology is visualized
by a CCD camera. Reflectivity measurements using a low power green
(532 nm) laser light are used to study fast morphological changes
of the membrane. Symbols: L1, L2, L3, L4, lenses; F, optical filter;
BS, beamsplitter; A, aperture; MO, microscope objective; CW, continuous
wave. (b) Optical images of a PDA membrane subjected to light-induced
contraction at ambient conditions. The power of the incident red laser
light is 30 mW. The red laser light in the ON state is not visible
due to the optical filter F. (c) The photoactuated state (laser ON)
of the PDA membrane at 50 and 60 mW laser power (upper and lower panels,
respectively).The light power of the red laser
was modulated using a square waveform
signal (on–off keying) with varied frequency (1–100
Hz) and amplitude (10–100 mW). The deformation of the membrane
was directly observed with optical microscopy using low intensity
white light illumination. Simultaneously, the mechanical response
of the membrane was probed with a low-power (<1 mW) green laser
light (532 nm) reflected from the membrane and detected by a photodiode.Figure b shows
the response of the PDA membrane to the illumination with the red
light as observed with optical microscopy. In the absence of light
(OFF), the surface of the membrane was buckled with dense wrinkles
(upper panel in Figure b). Irradiation (ON) at the center of the membrane [lower panel in Figure b and Supplementary Movie 1 (SM 1) with 30 mW power
caused an immediate flattening of the surface. The same phenomenon
was observed for different light powers (SM 2). The buckling and wrinkles rapidly re-emerged once the red laser
light was switched off and the membrane returned to the initial state. Figure c shows the flat
(laser ON) state of the PDA membrane at 50 mW (upper panel) and 60
mW (lower panel). From these images, it becomes evident that the surface
morphology of the irradiated PDA remains unchanged. By further increasing
the red laser power the light-driven motion stayed reversible up to
90 mW (SI 3).
Dynamics of Photoactuation
To elucidate PDA contraction
and expansion dynamics, we utilized the reflectivity experiment depicted
in Figure a. Figure a illustrates how
the reflectivity measurements can capture fast morphological changes
of the PDA membranes and reveal the dynamics of photoactuation. Initially
(red laser OFF, upper panel of Figure a), the membranes were buckled with wrinkles resulting
in diffuse reflection of the green probing light and suppressed photodiode
signal. The membrane contracts and flattens when exposed to high-power
laser irradiation (red laser ON, central panel of Figure a). This behavior results in
specular reflection of the green laser light captured by the photodiode.
Even though the membrane presented some folded parts (Figure S5), the obtained signal was not influenced
and a clear and reproducible response signal was obtained. The lower
panel of Figure a
illustrates a schematic of the duty cycle of the pump laser, the sample
response, and the time-evolving intensity of the reflected light from
the membrane. Figure b displays the membrane reflectivity as a function of time in response
to three exemplary modulated (on–off keying) frequencies of
the red laser light (SM 3).
Figure 3
(a) Principle for measuring
the dynamics of wrinkles with time-resolved
reflectivity of the membranes. Red laser light ON triggers the flattening
of the membrane and specular reflection of the green light (upper
panel), whereas the red laser OFF state corresponds to diffuse reflection
of the green light (middle panel). In the lower panel, “incident”
and “reflected” intensities refer to the green light
used to probe the membrane motions, and the red-shaded areas mark
the temporal intervals when the red pump laser is on. (b) Periodic
contraction and flattening of an irradiated PDA membrane. The membrane
becomes flat for 0.5 s, 50 ms, and 6 ms with a repetition rate of
1, 10, and 100 Hz, respectively. (c) Left panel: time-trace of laser-driven
contraction and flattening during the first 1 ms (blue points) and
fitting with an exponential rise (black line). Right panel: Relaxation
via swelling and wrinkling/buckling from 250 to 280 ms (blue points)
and fitting with a biexponential decay (black line). The red vertical
lines mark the start and end of the laser irradiation, respectively.
(d) Fatigue testing of PDA membranes subjected to laser-induced contraction.
The sample was periodically irradiated for 50 ms every 100 ms with
10 mW incident power of laser light.
(a) Principle for measuring
the dynamics of wrinkles with time-resolved
reflectivity of the membranes. Red laser light ON triggers the flattening
of the membrane and specular reflection of the green light (upper
panel), whereas the red laser OFF state corresponds to diffuse reflection
of the green light (middle panel). In the lower panel, “incident”
and “reflected” intensities refer to the green light
used to probe the membrane motions, and the red-shaded areas mark
the temporal intervals when the red pump laser is on. (b) Periodic
contraction and flattening of an irradiated PDA membrane. The membrane
becomes flat for 0.5 s, 50 ms, and 6 ms with a repetition rate of
1, 10, and 100 Hz, respectively. (c) Left panel: time-trace of laser-driven
contraction and flattening during the first 1 ms (blue points) and
fitting with an exponential rise (black line). Right panel: Relaxation
via swelling and wrinkling/buckling from 250 to 280 ms (blue points)
and fitting with a biexponential decay (black line). The red vertical
lines mark the start and end of the laser irradiation, respectively.
(d) Fatigue testing of PDA membranes subjected to laser-induced contraction.
The sample was periodically irradiated for 50 ms every 100 ms with
10 mW incident power of laser light.The contraction and flattening of the membrane happened in submillisecond
time scales after the laser is switched-on (Figure c, left panel). The reflectivity data have
been represented with exponential decay functions to extract the response
time of photoactuation and relaxation (SI 4). On the basis of this procedure, the time-constant for contraction
and flattening is found to be τc = (140 ± 10)
μs, which approaches the instrumental time-resolution. Notably,
the observed response to the light stimulus is substantially faster
than in the prior studies on, for example, NTE polymers,[39] photochemical actuators,[10] PDA-coated liquid crystal elastomers,[31] and PEDOT-Tos bilayer structures[40] (100–1000 s, 1–10 s, 0.1 s, and 3 ms, respectively)
and comparably fast with VO2-based actuators (milliseconds
to 160 μs).[9,41]The membrane remained flat
as long as the driving laser light was
on. Once the red light was switched off, the sample relaxed back to
the buckled state with slower relaxation lasting about 20 ms, as shown
in the right panel of Figure c. The relaxation dynamics can be represented with a biexponential
function with time-constants τ1 = (1.1 ± 0.1)
ms and τ2 = (15 ± 2) ms. The biexponential dynamics
are compatible with a two-stage mechanism where a fast water adsorption
is followed by slower structural relaxation. Such two-stage dynamics
has been observed during polymer-swelling, albeit the typically reported
time scales are slower. For instance, the water adsorption and structural
relaxation of zwitterionic polymer films (61 nm thick when dry) lasted
few seconds and 20 min, respectively.[42] The aqueous swelling of PDA coatings (79–85 nm thick) is
completed in 120 min or more depending on the cross-linking density.[28]One of the key parameters of actuating
materials is fatigue due
to cyclic loading. Therefore, we subjected the membranes to loading
over a certain number of light-induced cycles with 10 Hz frequency. Figure d shows that periodic
irradiation with 10 mW of the incident red laser light could keep
the PDA membrane in motion even after 45 600 and 90 000
cycles. The time-constant for the contraction remains around 0.1 ms
throughout cycling. Nevertheless, the relaxation of the membranes
revealed some fatigue after prolonged actuation. Possibly, a long
irradiation cycle caused additional cross-linking of PDA[28] and reduced water adsorption.
Thermal Stimulus
To elucidate whether the contraction
of the membrane was due to photothermal or photochemical processes,
we subjected PDA membranes to uniform heating. The temperature was
gradually increased from 295 to 367 K (10 K/min) at ambient pressure.
The optical images of the membranes captured at selected temperatures
are displayed in Figure a (SM 4). Here, we can notice that the
wrinkles start vanishing at about 333 K and the membrane is fully
flattened at 343 K. This process was fully reversible, that is, cooling
to room temperature returned the membrane surface to the initial buckled
state. Hence, we concluded that the origin of the contraction was
not photochemical as the same effect was achieved with a temperature
rise of about 40 K above room temperature. Conclusively, the red laser
light absorbed by the membrane serves as a local heat source.
Figure 4
(a) Optical
images of PDA membrane at exemplary temperatures during
the heating–cooling cycle. (b) The BLS spectra of PDA membranes
with the 660 nm laser light either off or on (30 mW). (c,d) The dependence
of the A0 wave frequency on the absorbed laser power and the temperature
of PDA, respectively. Experimental data and polynomial fittings are
shown with blue points and red lines, respectively. (e) The calculated
laser-induced temperature rise (ΔT) as a function
of the absorbed laser power.
(a) Optical
images of PDA membrane at exemplary temperatures during
the heating–cooling cycle. (b) The BLS spectra of PDA membranes
with the 660 nm laser light either off or on (30 mW). (c,d) The dependence
of the A0 wave frequency on the absorbed laser power and the temperature
of PDA, respectively. Experimental data and polynomial fittings are
shown with blue points and red lines, respectively. (e) The calculated
laser-induced temperature rise (ΔT) as a function
of the absorbed laser power.To determine the laser-induced temperature rise in the membrane,
we employed Brillouin light scattering (BLS). BLS is an inelastic
light scattering technique that can be used to study hypersonic acoustic
waves/phonons, elastic properties,[43] and
photoinduced phenomena[44] of semiconducting
nanomembranes. The BLS measurements were carried out with a CW laser
with wavelength λ = 532 nm, low power (0.5 mW), in backscattering
geometry and with an angle of incidence θ = 22° corresponding
to the acoustic wavenumber q = 4π sin θ/λ
= 8.85 μm–1. Figure b shows the BLS spectra obtained for the
PDA membrane at room temperature with the red laser off and during
red laser light irradiation. The illumination of the membrane with
the intense red light causes a red-shift of the peak corresponding
to the flexural (A0) acoustic wave.[34,43] The spectral
position of the A0 wave was measured as a function of the absorbed
power of the red laser light. The measured frequency f and the phonon wavenumber q are then used to extract
the phase velocity v = 2πf/q of the A0 wave. The results plotted in Figure c reveal systematic
reduction of v while increasing the absorbed light
power. This effect can be attributed to elastic softening of the membrane
and changes of thickness due to heating.Figure d displays
the temperature dependence of v measured under uniform
heating and the red laser light switched off. Noticeably, the global
temperature rise results again in reduced velocity of the A0 wave,
similar to the laser irradiation. These two experiments can be used
to calculate the local temperature rise in the irradiated membranes
(Figure e). For the
incident laser power of 30 mW (as in Figure b) the local temperature rise is about 75
K. Conclusively, the laser-induced heating is sufficiently high (>40
K, Figure a) to eliminate
surface wrinkles. Thus, the laser-induced PDA contraction can be solely
explained by the photothermal effect.Furthermore, the results
of Figure c–e
are compatible with reduction of the PDA
thickness upon laser irradiation or temperature increase. The velocity
of the A0 wave is proportional to both the Young modulus E and the thickness of the membrane[43]t as v ∝ E1/2t3/2. The results in Figure c show that with
a incident laser power of 30 mW (1.3 mW absorbed) v decreased by 18%. This effect can result either from a 33% decrease
of E or a 12% decrease of t. However,
prior studies showed that the Young modulus of PDA significantly increases
with heating.[45] Thus, the observed reduction
of v in Figure c,d can be explained by membrane contraction, which
corroborates with the results of optical microscopy.
Moisture Level
Stimulus
The photothermal effect that
drives contraction of PDA can be related with negative thermal expansion
(NTE)[39] or moisture desorption/adsorption.[46] To distinguish between these two possibilities,
we investigated the behavior of the membrane under low air pressure,
which reduces moisture (Figure a). Figure a (left panels) displays optical images captured at ambient pressure
of 1 bar and 1.5 mbar at 295 K, corresponding to a relative humidity
of about 60% and 0%, respectively. The decrease of the air pressure
resulted in membrane flattening already at room temperature. The full
reversibility of the process indicates that desorption/adsorption
of water causes PDA contraction/expansion. Notably, heating of the
(dehydrated) membrane in vacuum from 295 to 353 K (rightmost panel
of Figure a) did not
lead to any visible wrinkles. Thus, the intrinsic thermal expansion
of PDA was negligible in our study.
Figure 5
Optical images of a PDA membrane at (from
left to right) room temperature
and ambient atmosphere, room temperature and 1.5 mbar pressure, and
temperature of 353 K and pressure of 1 bar and 1.5 mbar, respectively.
(b) Summary of the individual and average fitted OH band area during
two temperature cycles. The insert displays SFG spectra of PDA deposited
on Si3N4 at 293 K (blue) and 345 K (red) and
their corresponding fit curves. Both spectra show a CH stretch and
OH stretch band, with the area of the OH band decreasing upon heating.
(c) A scheme of PDA membrane contraction/expansion due to light-,
heat- or moisture-induced desorption/adsorption of water molecules.
The insets show the proposed microscopic lamellar-like structure of
PDA[35] that can adsorb (left) or desorb
(right) water molecules from the atmosphere.
Optical images of a PDA membrane at (from
left to right) room temperature
and ambient atmosphere, room temperature and 1.5 mbar pressure, and
temperature of 353 K and pressure of 1 bar and 1.5 mbar, respectively.
(b) Summary of the individual and average fitted OH band area during
two temperature cycles. The insert displays SFG spectra of PDA deposited
on Si3N4 at 293 K (blue) and 345 K (red) and
their corresponding fit curves. Both spectra show a CH stretch and
OH stretch band, with the area of the OH band decreasing upon heating.
(c) A scheme of PDA membrane contraction/expansion due to light-,
heat- or moisture-induced desorption/adsorption of water molecules.
The insets show the proposed microscopic lamellar-like structure of
PDA[35] that can adsorb (left) or desorb
(right) water molecules from the atmosphere.In order to investigate the molecular changes that could occur
in the PDA membrane upon thermal switching, sum-frequency-generation
(SFG) vibrational spectra of PDA deposited on Si3N4 were measured. This technique allows the investigation of
the polymer in a highly surface sensitive and chemically specific
manner.[47,48] SFG spectra (inset in Figure b) taken at 293 and 345 K show bands at 2950
and 3400 cm–1 attributed to polymer CH stretch modes
and to OH stretch modes of the polymer and/or of water, respectively.
The area of the OH-stretch band (Figure b; SI 5) decreases
upon heating and recovers upon subsequent cooling. Some dampening
of the effect due to polymer aging during the prolonged exposure to
laser radiation is observed. While spectral contributions of the OH
groups of the polymer and the water adsorbed in the film cannot be
decomposed, the observed changes are consistent with a reversible
temperature-induced change to the structure of the polymer and with
the hypothesis of water removal.Figure c illustrates
the proposed microscopic mechanism manifested as the macroscopic contraction
of PDA in response to external stimuli. The relatively slow potential
sweep that was used during the electro-polymerization of PDA favors
the formation of a hydroxyl indole-like lamellar structure.[35] Thus, in the present work, ultrathin PDA is
expected to contain short semicrystalline molecular segments of 5–6
hydroxyindole (DHI) units held together by weaker intermolecular bonds
and water molecules trapped in the free volume that remains after
cross-linking. In this work, the structural characterization with
HR-TEM and FTIR is not feasible due to the small thickness of the
membranes. However, quasi-crystalline graphenelike molecular segments
of PDA have been observed in thicker specimens.[49,50] We propose that the adsorption (desorption) of water can strengthen
(weaken) the intermolecular interactions in PDA, allowing in this
way the rearrangement of the DHI crystalline unit and leading to the
observed light-driven contraction. In ambient conditions, the membrane
adsorbs moisture leading to increased distances between the PDA segments
and expansion. Desorption of water from PDA, driven by light, temperature
changes, or vacuum, causes collapsing of PDA segments and macroscopic
contraction of the membrane. Interestingly, the temperature at which
the membranes become flat (i.e., 370 K in Figure b and 333–343 K in Figure a) match the temperature for
desorption of surface-bound water from PDA (340–350 K).[28,51]
Conclusions
Herein, we demonstrated for the first time
fast photoactuation
of bare polydopamine membranes fabricated with state-of-the-art electropolymerization
and PVA-assisted transfer onto holey substrates. With this method,
PDA exhibits photoactuation without requiring additional mechanical
or photothermal components as in prior studies.[29−33,40] Visible light drives
ultrafast contraction (<140 μs) of the membranes, while switching
off light leads to spontaneous expansion in milliseconds time scale.
The ultrafast response time is attributed to the small inertia of
the membrane, fast water desorption due to the small thickness, and
rapid heat transfer to the surroundings (see also SI 6). The stimulated contraction mimics natural muscle fibers’
behavior as opposed to the most frequently used photoactuators based
on thermal expansion and bending.[41,52] Contraction/expansion
of the membranes can also be driven by heat and moisture. The observed
phenomenon is attributed to the desorption/absorption of water molecules
by PDA to/from the atmosphere. The proposed explanation is in line
with the effect of molecular penetrants known for swelling of polymers[42] including PDA.[28]Bare PDA membranes made by electropolymerization can be used as
multistimuli building blocks for soft micro- and nanodevices.[41,52] PDA offers broad-spectrum photothermal features, relatively high
Young modulus,[34] and strong adhesion on
multifarious surfaces,[17,53] making it useful for nanoscale
remote actuation, artificial muscles, moisture or light sensing, energy
harvesting, and adaptative optics or metamaterials (photonic and phononic).
Moreover, the actuation of wrinkles on membranes can be used in pressure-
and strain-sensitive, moisture-erasable, vibrotactile displays, and
data-recording devices.[40,54] Finally, PDA can be
patterned and decorated with metals using lithographic techniques,[53] which can be useful for the mass production
of nanorobots or other nanodevices.[55] We
envisage that this work will stimulate further fundamental studies
on actuation by nonmonochromatic light, optimization of the preparation
conditions, and the detailed microscopic picture of the water dynamics
in PDA.
Authors: B Graczykowski; M Sledzinska; M Placidi; D Saleta Reig; M Kasprzak; F Alzina; C M Sotomayor Torres Journal: Nano Lett Date: 2017-11-17 Impact factor: 11.189