Far-red triplet sensitized Z-to-E photoswitching of azobenzene in bioplastics

We report the first example of direct far-red triplet sensitized molecular photoswitching in a condensed phase wherein a liquid azobenzene derivative (Azo1) co-assembled within a liquid surfactant–protein film undergoes triplet sensitized Z-to-E photoswitching upon far-red/red light excitation in air. The role of triplet sensitization in photoswitching has been confirmed by quenching of sensitizer phosphorescence by Z-Azo1 and temperature-dependent photoswitching experiments. Herein, we demonstrate new biosustainable fabrication designs to address key challenges in solid-state photoswitching, effectively mitigating chromophore aggregation and requirement of high energy excitations by dispersing the photoswitch in the trapped liquid inside the solid framework and by shifting the action spectrum from blue-green light (450–560 nm) to the far-red/red light (740/640 nm) region.

Preparation of G-TXr-Azo1-PdTPBP film. 11.83 μl of Azo1 (42.258 mM in toluene), and 38 μl of PdTPBP (656 μM in toluene ) were taken in a glass vial and evaporated under reduced pressure. To this added 0.026 g of TX100-reduced (TXr), followed by stirring until Azo1, and PdTPBP gets dissolved in TXr. To the resulting solution added 1 ml of MQ water, followed by stirring. Into the Azo1-TXr-H 2 O solution added 0.22 g of gelatin type A (G), followed by stirring at 80 °C for 10 min. The hot sol was allowed to rest at room temperature for 2 minutes, followed by drop-casting of the 260 μl of the sol on a 3 x 1 cm glass plate and air drying for 48 h. The air-dried G-TXr-Azo1 film contain TX-r = 10.5%, gelatin = 89.4%, PdTPBP = 82 µmol.kg -1 and Azo1 = 1.64 mmol.kg -1 . The greenish semi-transparent film obtained after air drying for 48 h is shown in Fig. 5a of the main manuscript.
Preparation of G-TXr-Azo1-ZnPc film. 11.83 μl of Azo1 (42.258 mM in toluene), and 125 μl of ZnPc (200 μM in toluene ) were taken in a glass vial and evaporated under reduced pressure. To this added 0.026 g of TX100-reduced (TXr), followed by stirring until Azo1 and ZnPc gets dissolved in TXr. To the resulting solution added 1 ml of MQ water, followed by stirring. Into the Azo1-TXr-H 2 O solution added 0.22 g of gelatin type A (G), followed by stirring at 80 °C for 10 min. The hot sol was allowed to rest at room temperature for 2 minutes, followed by drop-casting of the 260 μl of the sol on a 3 x 1 cm glass plate and air drying for 48 h. The air-dried G-TXr-Azo1 film contain TX-r = 10.5%, gelatin = 89.4%, ZnPc = 82 µmol.kg -1 and Azo1 = 1.64 mmol.kg -1 . The light green semi-transparent film obtained after air drying for 48 h is shown in Fig. 6a of the main manuscript.
Optical Measurements. All UV−vis absorption spectra were recorded on a Varian Cary 50 spectrophotometers. Steadystate emission spectra were recorded on a Varian Eclipse spectrophotometer and FLS1000 Photoluminescence Spectrometer, Edinburgh instrument. The UV-Vis and emission spectra in solution were recorded using 2 mm pathlength quartz cuvettes. The phosphorescence spectra of PdTPBP in the presence and absence of Azo1 in the G-TXr film to measure triplet energy transfer were obtained using a home-built setup consisting of Coherent OBIS LS 633 nm diode laser as the excitation source, a 1681 SPEX monochromator, and a photomultiplier tube (PMT) detector. A 633 nm notch filter was used in front of the monochromator to reduce the scattered excitation light reaching the detector. The phosphorescence lifetimes of PdTPBP in the G-TXr-PdTPBP and G-TXr-Azo1-PdTPBP films were measured using Varian Eclipse 1 fluorescence spectrophotometer at λ ex = 640 nm, and λ em = 800 nm. The 340 nm and 440 nm LED's from THORLABS, whereas 640 nm and 740 nm LED's from LEDENGIN were used for photoswitching experiments. The LED powers were used were varied for different photoswitching experiments. For 90 ° excitation of different films the excitation power density (P ex ) of different LED's ( P ex = 2.8 mW cm -2 of 340 nm, 27.6 mW cm -2 of 440 nm, 14.1 mW cm -2 of 640 nm and 17.1 mW cm -2 of 740 nm) were used. For kinetics experiments at 75° excitation of different film samples three different LED powers of each LED mentioned in the figures text were used. For photo-kinetics experiments both sample and LED's were covered with a black cloth in a dark room to avoid the interference of room light. For thermal isomerization measurements the films were cut in way so as they fit in the diameter of the quartz cuvette to experience uniform temperature. The phosphorescence spectrum of ZnPc was measured at 100 K, on Spex Fluorolog 3 spectrofluorometer (Horiba Jobin Yvon) equipped with a Hamamatsu H10330-45 NIR PMT detector at excitation wavelength of 700 nm.
Other Measurements. The X-ray powder diffraction of G-TXr-Azo1-ZnPc film was recorded using D8 Discover, Bruker Bragg-Brentano with Cu source. The differential scanning calorimetry (DSC) traces were obtained by using a TGA / DSC 3+ STARe system (METTLER TOLEDO) under N 2 atmosphere. Before performing measurements, samples were completely dried under vacuum for at least 24 h. The scanning rate was 10 °C min -1 . Reproducible thermogram of 3rd and 4 th cycles were considered for presentation. The cross-section Scanning electron microscopy (SEM) images of the G-TXr-Azo1-ZnPc film were obtained using JEOL JSM-7800F Prime FEG SEM. The acceleration voltage was set to 10 kV and a secondary electron detector was used. Before SEM observation, the sample was dried under vacuum for two days and coated with gold using Edwards S150B gold sputter. The cross-section was obtained by breaking the film in half with hands.
Calculation of the Triplet Energy of Azo1. Over singlet optimized geometries using the B3LYP/cc-pVTZ method, the triplet energy was computed using the restricted open formalism to improve the comparability between singlet and triplet energies as described in Gagliardi, L., Orlandi, G., Bernardi, F. et al. 3 Additionally, the effect of the solvent environment was estimated using the iefpcm model for toluene, without big changes in the relative energies.
No significant differences were found between the Z-isomer of non-derivatized azobenzene and Z-Azo1 as can be seen in Table S1, for that we can assume that the triplet energy of Z-Azo1 is slightly lower (almost equivalent) than the one found for azobenzene, even being so close to the estimated method uncertainty. In contrast a more noticeable difference can be seen for E-isomer which is slightly higher than free azobenzene. Also, some functional benchmark was done to ensure a proper reproduction of the system by B3LYP observing, as described in the literature, a good description of the considered energies. Observing the typical shifts when changing the HF exchange percentage over the different functionals. Supporting Figures   Fig. S1. DSC thermograms of vacuum dried a) TX-100-reduced, and b) Azo1 liquid. Scanning rate 10 o C min -1 .