A quantitative demonstration of reversible stiffness upon appropriate light stimulus in a spiropyran-polymeric composite is presented. The polymeric films containing 3% wt. of the photochromic spiropyran were irradiated with alternating ultraviolet and visible light and the storage modulus was measured. A reversible change in modulus of about 7% was observed. The modulus change was attributed to an interaction of the polar merocyanine with the polymeric chains and/or to a variation of effective free volume induced by merocyanine aggregates formed in the polymer upon ultraviolet irradiation. The effect is fully reversed when the merocyanine isomers turn back to the spiropyran state after visible irradiation.
Mechanical deformation and light-matter interaction can be closely interconnected in some specific materials. In the case of optical properties modified by mechanical action (e.g., color change), the phenomenon involved is called mechanochromic. In the reversed case, the phenomenon is known as photomechanic and occurs to materials that undergo a macroscopic and reversible deformation when exposed to a light source. Photomechanical actuators made of polymers/photochromic composite materials can find interesting applications where the tuning of the mechanical properties of a smart material is required. Photochromic molecules (PMs) are promising polymer fillers that suit this purpose, due to their ability to switch between two different isomeric conformations upon light irradiation. The majority of the demonstrations of photomechanical effect are obtained in materials prepared by a covalent bonding between polymers and PMs, and they are detected as reversible volume changes in the used systems upon the photoisomerization process.1–7 However, actual mechanical parameters of the composite systems were not calculated. Among the many PMs, spiropyran (SP) was often used due to its large structure rearrangement upon reversible photoconversion with light in the UV/visible (VIS) part of the spectrum. Such structure rearrangement results from isomerization between transparent spiropyran and colored merocyanine (MC) upon cleavage/formation of the C-O spiro bond and ring opening/closing. On the other side, replacing the complex chemical synthesis by simple blending of the polymer and PMs may provide a simple solution for the fabrication of photoactuators. A proper choice of the polymer and PMs is necessary in order to tailor effectively the mechanical properties of the material, due to the usually quite low efficiency of the photomechanical effect. So far the photochromic blends of SP and polymers were proposed for applications in rather few papers, e.g., the works of Athanassiou et al.8,9 and Fragouli et al.,10 where it was shown that SP-polymeric films (5 to 10% wt.) can change reversibly their volume due to the isomeric transformations of the embedded photochromic molecules, acting as light-controlled mechanical actuators or as optical gratings with reversible diffraction efficiency.
In this work, following a different approach, we attempt to translate the structural changes of the photochromic fillers into quantitative variations of the storage modulus of the composite materials, creating thus devices with tunable compliance.11 The measurements were performed in situ using a dynamic mechanical thermal analyzer (DMTA), following UV or VIS illumination on samples made of photochromic SP, shown in the inset of Fig. 1, blended with poly(ethyl methacrylate-co-methyl acrylate) (PEMMA). The composite was prepared by mixing pellets of PEMMA polymer (PARALOIDTM B-72, Rohmihaas, MW = 88 000 (Ref. 12) and powder of spiropyran (1,3,3-Trimethylindolino-6′-nitrobenzopyrylospiran, TCI Europe) with concentration of the latter 3% by weight, and then dissolved in chloroform. The polymer-photochromic solution was then poured on a Teflon disc and dried at room temperature, first in air and then in a desiccator for >72 h, to ensure solvent evaporation. Samples were cut from the obtained disc, with dimensions of approximately 7 × 30 × 0.080 ± 0.015 mm3. For irradiation, a commercial broadband light source (LightningCure LC8, Hamamatsu, Japan) was used selecting the UV and VIS parts of the spectrum by two glass filters (UG11 for UV and GG435 for VIS). The intensity of the light, measured at the sample position, was 1 mW/cm2 for UV and 26 mW/cm2 for VIS.
(Color online) Absorption spectra of the PEMMA/SP samples, for the different irradiation times used in the tests. The black line shows the absorbance of the sample before irradiation and corresponds to 0% SP to MC conversion. The irradiation time of 1 min UV corresponds to 50% conversion, 4 min UV to 90% conversion. The inset shows the molecular structures of the SP (left) and MC (right) isomers.
(Color online) Absorption spectra of the PEMMA/SP samples, for the different irradiation times used in the tests. The black line shows the absorbance of the sample before irradiation and corresponds to 0% SP to MC conversion. The irradiation time of 1 min UV corresponds to 50% conversion, 4 min UV to 90% conversion. The inset shows the molecular structures of the SP (left) and MC (right) isomers.
The chemical structures of spiropyran and colored merocyanine are shown in Fig. 1 together with the absorption spectra measured with a spectrophotometer (Varian, Cary 6000i, Agilent Technologies, USA). The transparent SP polymeric film exhibits absorption only in the UV spectral range below 440 nm (black line). After 1 min of UV irradiation an absorption band centered at 556 nm was observed indicating formation of the open-ring MC (light blue line). When UV irradiation time is increased to 4 min, the band maximum exhibited a small blue shift by 4 nm (dark blue line). Irradiation with the visible light (λ > 435 nm) caused reversed photoconversion to the SP (indicated with the green arrows). Prior to the mechanical test, the spectra corresponding to different UV irradiation times were measured in order to find the irradiation conditions for two different levels of conversion of the SP molecules to their isomeric MC form. The percentage of photoconversion was evaluated by the relative intensity of the 556/552 nm peak. An irradiation time of 4 min was needed to obtain the formation of MC at a level of approximately 90% of the saturation value. From such conversion degree, the time necessary for complete recovery to the SP state upon VIS irradiation was evaluated as 7 min. Similarly, 1 min of UV and 3 min of VIS were needed to convert approximately 50% of the molecules and to revert them back to the original state, respectively. The samples' thickness reassures almost homogeneous photoconversion of the photochromic molecules throughout the samples.
Mechanical measurements were conducted on a Q800 DMTA machine (TA Instruments, USA). The samples were fastened on the tensile clamps and irradiated with UV and consequently with the VIS light. Previous to each measurement, the DMTA furnace was closed and each sample was brought to 28 °C where it was kept for 3 min in order to thermally equilibrate it. A pre-stress of 0.05 N was applied to each sample, and the dynamic modulus was measured at 1 Hz with 5 μm sinusoidal amplitude. Each full testing cycle consisted of two mechanical measurements, one after UV and another after VIS irradiation.
Two sets of experiments were performed to quantify the modulus variation for ≈90% (full conversion experiment) and ≈50% (partial conversion experiment) SP to MC conversion, respectively, and to evaluate the tunability of the system. Each set consists of experiments performed on three different samples. On each sample a total number of ten UV/VIS irradiation cycles was performed.
One representative curve from the full conversion experiment is shown in Fig. 2. For each irradiation cycle, a clear difference in the stiffness of the sample incorporating the two different isomeric states can be seen. UV-induced ring opening and formation of the colored MC resulted in the sample’s stiffening, whereas the inversed photochromic reaction was accompanied by sample’s softening. The quantification of the modulus (E′) variation achieved was characterized through a relative variation parameter, defined as , where i is the measurement number (one half of the testing cycle as defined above). The two obtained curves for both full and partial conversion are shown in Fig. 3, where the difference in the modulation can be seen.
(Color online) Storage modulus as a function of the measurement number for the full conversion experiment. Symbols' colors correspond to the spectra in Fig. 1.
(Color online) Storage modulus as a function of the measurement number for the full conversion experiment. Symbols' colors correspond to the spectra in Fig. 1.
(Color online) Relative storage modulus variation as a function of the measurement number for representative tests of the full and partial conversion. Symbols' colors correspond to the spectra in Fig. 1, line colors were chosen for clarity.
(Color online) Relative storage modulus variation as a function of the measurement number for representative tests of the full and partial conversion. Symbols' colors correspond to the spectra in Fig. 1, line colors were chosen for clarity.
The averaged relative variation parameter (calculated from 60 absolute values) was 6.5% ± 1.1% for full conversion and 2.4% ± 1.0% for partial conversion. This is a remarkable outcome, taking into account that it has been achieved with the physical addition of just 3% wt. of PMs and that it is reversible. Due to the novelty of the approach, the results presented here are rather difficult to directly compare to any literature data.
The mechanism behind the measured stiffening is still under debate, with two proposed scenarios fitting best our findings. The first of them considers the interaction of highly polar MC isomers with the polymeric chains. Lin showed that the absorption spectrum of MC embedded in three different polymers is blue-shifted in the cases where there is interaction with the polymer matrix. The interaction increases with increasing polarity of the matrix. Interestingly, such shift takes place also in low-polar polymers but characterized by a low glass transition temperature.13 In that case, interaction is enhanced by the higher chains mobility. The absorption maximum reported in such a polymer (poly(n-butyl methacrylate)) was 560 nm, which is very similar to the value reported here (556/552 nm). The dipole-dipole interaction between the MC and polymeric chains may restrict the chain mobility causing the observed stiffening. VIS irradiation transforms highly polar MC into much less polar SP accompanied by a large structural rearrangement; thus the interaction is weakened and the polymer recovers its mobility and softness.
The second scenario involves aggregates formation as proposed by Athanassiou et al.8 The colored MC, due to its high polarity, forms dimers or aggregates of zwitterionic character upon UV irradiation in solutions14 and in polymeric matrices.8 The latter case was proven by measuring fluorescence spectra of the metastable MC photoisomers responsible for aggregate formation. The MC aggregates, compared to the initial SP stereoisomers, have less effective partial molar volume. For this reason the polymer chains move closer to each other,8 forming stronger inter-chain bonds and causing thus an increased stiffness of the sample. Upon VIS light and SP recovery there are no more MC isomers to form aggregates and the composite reverts to the original stiffness.
In conclusion, we have herein shown a tunable light-induced variation in stiffness of samples made of a blend of a soft polymer, such as PEMMA, with a small amount of photochromic molecules. The main advantage of the system studied is its simplicity, in that it is produced as a physical blend. The stiffness can be controlled exclusively by compact light sources, which do not need wires and connections like in the case of other external stimuli, e.g., electrical or thermal. The perspective of tailoring the mechanical response of such blends by acting on the molecule state/degree of conversion potentially opens the way to various applications such as variable sensitivity detectors (e.g., tactile sensors) or actively compliant components.
We would like to thank Dr. Fernando Brandi for helpful discussions.