Fluorescent polymer films based on photo-induced electron transfer for visualizing water

As fluorescent materials for visualization, detection, and quantification of a trace amount of water, we have designed and developed a PET (photo-induced electron transfer)-type fluorescent monomer SM-2 composed of methyl methacrylate-substituted anthracene fluorophore-(aminomethyl)-4-cyanophenylboronic acid pinacol ester (AminoMeCNPhenylBPin) and achieved preparation of a copolymer poly(SM-2-co-MMA) composed of SM-2 and methyl methacrylate (MMA). Both SM-2 and poly(SM-2-co-MMA) exhibited enhancement of the fluorescence emission with the increase in water content in various solvents (less polar, polar, protic, and aprotic solvents) due to the formation of the PET inactive (fluorescent) species SM-2a and poly(SM-2-co-MMA)a, respectively, by the interaction with water molecules. The detection limit (DL) of poly(SM-2-co-MMA) for water in the low water content region below 1.0 wt% in acetonitrile was 0.066 wt%, indicating that poly(SM-2-co-MMA) can act as a PET-type fluorescent polymeric sensor for a trace amount of water in solvents, although it was inferior to that (0.009 wt%) of SM-2. It was found that spin-coated poly(SM-2-co-MMA) films as well as 15 wt% SM-2-doped polymethyl methacrylate (PMMA) films produced a satisfactory reversible fluorescence off–on switching between the PET active state under a drying process and the PET inactive state upon exposure to moisture, which is demonstrated by the fact that the both the films are similar in hydrophilicity to each other from the measurement of the water contact angles on the polymer film surface. Herein we propose that PET-type fluorescent polymer films based on a fluorescence enhancement system are one of the most promising and convenient functional dye materials for visualizing moisture and water droplets.


Introduction
In recent years, concern has been raised about the development of uorescent sensors and their functional materials such as polymer lms and sensor-immobilized membranes for visualizing water in solutions, solids, and gas or on material surfaces, from the viewpoint of their potential applications to environmental and quality control monitoring systems and industry, as well as fundamental study in photochemistry, analytical chemistry, and photophysics. 1- 23 Several investigations have been conducted on the design and synthesis of organic uorescent sensors and polymers for the detection of water based on ICT (intramolecular charge transfer), [24][25][26][27][28][29][30][31][32][33][34] ESIPT (excited state intramolecular proton transfer), [35][36][37][38] PET (photo-induced electron transfer), [39][40][41][42][43][44][45][46] or solvatochromism [47][48][49][50][51][52] and the elucidation of the optical sensing properties based on changes in wavelength, intensity, and lifetime of uorescence emission depending on the water content. It was demonstrated that most of ICT-and ESIP-type uorescent sensors and uorescent conjugated polymers exhibited attenuation of the uorescence emission, that is, uorescence quenching (turn-off) systems with the increase in water content in solvents, and were suitable for the detection and quantication of a trace amount of water (below 1-10 wt% in almost every case) in solvents. However, one can see that the uorescence quenching (turn-off) systems make it difficult to visually conrm the presence of water in samples and on material surfaces. On the other hand, the PETtype uorescent sensors are based on a uorescence enhancement (turn-on) system and showed the increase in the uorescence intensity with the increase in water content in solvents, so that it allowed us to visually conrm the presence of water in samples and on material surfaces. Thus, we have focused on the design and development of PET-type uorescent sensors for the detection and quantication of water and the preparation of their functional materials for visualizing water. In our continuous work for the improvement of the sensitivity and accuracy of PET-type uorescent sensors for water during the past decade, we have demonstrated that anthracene-(aminomethyl)-4-cyanophenylboronic acid pinacol ester (Amino-MeCNPhenylBPin) OF-2 and its derivative SM-1 having a hydroxymethyl group on the anthracene skeleton were highly sensitive PET-type uorescence sensors for the detection and quantication of a trace amount of water in polar, less polar, protic, and aprotic solvents ( Fig. 1a and b). 23,42 In each sensor, the PET takes place from the nitrogen atom of the amino moiety to the photoexcited anthracene uorophore in the absence of water, leading to quenching of the uorescence. When water was added to the solution of OF-2 or SM-1, the nitrogen atom of the amino moiety was protonated or strongly interacted with water molecules to form the PET inactive (uorescent) species OF-2a or SM-1a, and as a result, a drastic enhancement of the uorescence emission was observed due to the suppression of PET. Indeed, the detection limits (DLs) and quantitation limits (QLs) of OF-2 and SM-1 for water in acetonitrile were, respectively, 0.009 wt% and 0.026 wt% and 0.004 wt% and 0.013 wt%, which were equivalent or superior to those of the uorescence quenching systems (turn-off) based on the reported ICT-type [24][25][26][27][28][29][30][31][32][33][34] and ESIPT-type 35-38 uorescent sensors. Thus, the PET method based on the uorescence enhancement (turn-on) system makes it possible to visualize, detect, and determine a trace amount of water in solvents.
Meanwhile, under the Coronavirus Disease 2019 (COVID-19) situation, face shields made of polyester or polycarbonate lms and partitions made of acrylic resin are one of convenience and commercially available protective goods for reducing the risk of droplet infection. Therefore, if we can visually conrm the presence of droplets containing infectious viruses on the face shields and partitions, this allows us to accurately remove the viruses by wiping away the droplets. However, the viruscontaining droplets are generally 5 mm or more which is too small for our eyes to see. Nevertheless, because over 90% of the droplets is composed of water, functional materials as well as techniques and methods capable of visualizing water are undoubtedly useful for detecting the virus-containing droplets. In our previous work, for this purpose, we have achieved the preparation of various types of uorescent polymer lms (polystyrene, poly(4-vinylphenol), polyvinyl alcohol, and polyethylene glycol) doped with the PET-type uorescent sensor OF-2 or SM-1. 22,23 It was found that the OF-2or SM-1-doped polymer lms exhibited a reversible switching of the uorescent color between feeble green excimer emission in the PET active state under a drying process and intense blue monomer emission in the PET inactive state upon exposure to moisture or water droplets. Our previous work is the rst to achieve the preparation of PET-type uorescent sensor-doped polymer lms for water, while ICT-type [32][33][34] or ESIPT-type 37,38 uorescent sensor-doped polymer lms and uorescent conjugated polymers [18][19][20][21]53,54 for water based on a uorescence quenching (turnoff) system have been reported. However, the reversibility of the uorescence intensity of OF-2or SM-1-doped polymer lms between the excimer and monomer emissions in the dry-wet process were not fully satisfactory for the practical use for the visualization and detection of water on materials surfaces, due to destruction of the lms during the dry-wet process.
Thus, in this work, to improve the reversibility of the uorescence intensity of PET-type uorescent polymer lms by giving strong durability during the dry-wet process, we have designed and developed a PET-type uorescent monomer SM-2 having a methyl methacrylate group on the anthracene skeleton as a derivative of SM-1 and achieved preparation of a copolymer poly(SM-2-co-MMA) composed of SM-2 and methyl methacrylate (MMA) (Fig. 1c and d). It was found that spin-coated poly(SM-2co-MMA) lms as well as SM-2-doped polymethyl methacrylate (PMMA) lms produced a satisfactory reversible uorescence off-on switching between the PET active state under a drying process and the PET inactive state upon exposure to moisture, which is demonstrated by the fact that the both the lms are similar in hydrophilicity to each other from the measurement of the water contact angles on the polymer lm surface. Herein we propose that PET-type uorescent polymer lms based on a uorescence enhancement system are one of the most promising and convenient functional dye materials for visualizing moisture and water droplets.

Results and discussion
The PET-type uorescent monomer SM-2 was prepared by the reaction of SM-1 23 with methacryloyl chloride (Scheme 1). Then, polymerization was carried out by a ratio of SM-2 and MMA of 1 : 20 using 2,2 0 -azobis(isobutyronitrile) (AIBN) 55 as a free radical initiator to give poly(SM-2-co-MMA) as a white solid (M n ¼ 18 900, M w /M n ¼ 2.08, 17% yield). As the result, the 1 H NMR spectrum indicated that the molar ratio (y/x) of MMA unit (y) and SM-2 unit (x) and the weight percentage (wt%) of SM-2 in the obtained poly(SM-2-co-MMA) were determined to be ca. 40 and ca. 15 wt%, respectively. The optical sensing ability of the PET-type uorescent monomer SM-2 for water in solvents was investigated by photoabsorption and uorescence spectral measurements in 1,4dioxane and THF as less polar solvents, acetonitrile as a polar solvent, and ethanol as a protic solvent containing various concentrations of water (in the water content region below 10 wt%) (Fig. 2). As with the cases of OF-2 42 and SM-1, 23 SM-2 in all the four solvents showed a vibronically-structured photoabsorption band in the range of 300 nm to 400 nm originating from the anthracene skeleton and did not undergo appreciable changes in the absorbance and shape upon the addition of water to the solutions (Fig. 2a, c, e and g). For the corresponding uorescence spectra, SM-2 in the absolute solvents exhibited a feeble and vibronically-structured uorescence band with a uorescence maximum wavelength (l  max ) at around 420 nm in the range of 400 nm to 500 nm, which is attributed to the monomer emission originating from the anthracene uorophore in the PET active state (Fig. 2b, d, f and h). On the other hand, in the low water content region below 1.0 wt%, the uorescence band increased in the intensity with the increase in the water content in the solution, which is attributed to the formation of the PET inactive (uorescent) species SM-2a by the addition of a water molecule, as with the cases of OF-2 42 and SM-1 23 (Fig. 1a-c). As shown in Fig. 3a, the acetonitrile solution of SM-2 without the addition of water did not show visual uorescence emission but exhibited the blue uorescence emission originating from the anthracene uorophore upon the addition of water. In fact, to conrm the formation of the PET inactive species SM-2a by the interaction with a water molecule, we performed 1 H NMR spectral measurements of SM-2 with and without the addition of water in the acetonitrile-d 3 solution (2.0 Â 10 À2 M) (Fig. 4). The 1 H NMR spectrum of the SM-2 solution (water content of 0.49 wt%) without the addition of water showed an obvious signal that can be assigned to a single chemical species with the SM-2 structure. On the other hand, some additional signals appeared in both the aliphatic and aromatic regions in the 1 H NMR spectrum of the SM-2 solution with water content of 2.3 wt%, compared to that of the solution without the addition of water, indicating the existence of other chemical species as well as SM-2. Moreover, for the 1 H NMR spectrum of the SM-2 solution with water content of 13 wt%, the chemical shis of the methyl protons H a of boronic acid pinacol ester, the aminomethyl protons H c , and the aromatic protons H n and H o of the anthracene skeleton showed considerably upeld shis, while those of the methylene protons H e next to the anthracene skeleton and the aromatic protons H i and H k of the phenyl group showed considerably downeld shis. Consequently, the  fact strongly indicates that the PET inactive species SM-2a interacted with water molecules occurred upon the addition of water to the SM-2 solution (Fig. 1c), as with the cases of OF-2 42 and SM-1. 23 The sensitivity and accuracy of SM-2 for the detection of water in solvents were evaluated by the changes in the uorescence peak intensity at around 420 nm and the plots against the water fraction in solvents (Fig. 5). The plots for SM-2 demonstrated that the uorescence peak intensity increased linearly as a function of the water content in the low water content region below 1.0 wt% in all four solvents (Fig. 5a), while the uorescence intensity leveled off when the water content reached 1.0 wt% as with the cases of OF-2 42 and SM-1. 23 The results of the plots for SM-2 are as follows: The correlation coefficient (R 2 ) values for the calibration curves of SM-2 were 0.974-0.988, which indicates good linearity. A linear change in uorescence intensity as a function of water content is one of the factors required for the practical use of a uorescence sensor for water. [39][40][41][42][43] The intercept values (2.1-76.7) demonstrated that the plots for 1,4-dioxane, THF, and acetonitrile t straight lines passing through almost the origin, which also indicates the uorescence enhancement due to the formation of the PET inactive species SM-2a with the increase in the water content. Meanwhile, it is considered that the enhanced uorescence of SM-2 in absolute ethanol is attributed to the suppression of PET by the hydrogen bonding between the hydroxyl group of ethanol and the amino moiety of SM-2, as with the cases of OF-2 42 and SM-1. 23 It is worth mentioning here that there was a little difference in the m s values (288-355) for SM-2 between the four solvents, while the m s values for SM-2 were equivalent to those for OF-2 but smaller than those for SM-1 (Table 1). The large m s values for SM-1 relative to SM-2 and OF-2 can be attributed to the fact that the uorescence emission property was improved by the introduction of a hydroxymethyl group to the anthracene uorophore. Actually, uorescence quantum yields (F  ) of OF-2, SM-1, and SM-2 in absolute acetonitrile were below 2%, but in acetonitrile with 1.0 wt% water content, the F  (20%) of SM-1 was higher than those (13% and 12%, respectively) of OF-2 and SM-2. The DLs and QLs of SM-2 for water in the solvents were determined based on the following equations: DL ¼ 3.3s/m s and QL ¼ 10s/m s , where s is the standard deviation of blank sample and m s is the slope of a calibration curve obtained from the plot of the uorescence  peak intensity at around 420 nm the water fraction in the low water content region below 1.0 wt% (Fig. 5b). The DLs and QLs of SM-2 for water were, respectively, 0.011 and 0.035 wt% in 1,4dioxane, 0.01 and 0.03 wt% in THF, 0.009 and 0.028 wt% in acetonitrile, and 0.01 and 0.032 wt% in ethanol, which were equivalent to those of OF-2 but inferior to those of SM-1. Consequently, it was found that methyl methacrylatesubstituted anthracene-AminoMeCNPhenylBPin SM-2 can act as a PET-type uorescent sensor for the detection and quanti-cation of a trace amount of water in polar, less polar, protic, and aprotic solvents, as with the reported PET-type uorescent sensors for water including OF-2 and SM-1.
To investigate the optical sensing ability of the copolymer poly(SM-2-co-MMA) for water, photoabsorption and uorescence spectra of poly(SM-2-co-MMA) were measured in acetonitrile containing various concentrations of water (Fig. 6). As with the case of SM-2, poly(SM-2-co-MMA) in absolute acetonitrile exhibited a vibronically-structured photoabsorption band in the range of 300 nm to 400 nm and a feeble and vibronically-structured uorescence band (l  max ¼ ca. 420 nm) in the range of 400 nm to 500 nm originating from the anthracene skeleton in the PET active state. The photoabsorption spectra showed unnoticeable changes with the increase in the water content in the acetonitrile solutions. In contrast, the uorescence intensity of the monomer emission band originating from the anthracene uorophore increased almost linearly with the increase in the water content in the low water content region below ca. 1.0 wt% in the acetonitrile solutions due to the formation of the PET inactive species poly(SM-2-co-MMA)a by the addition of water molecules (Fig. 1d), while the uorescence intensity leveled off when the water content reached 1.0 wt% as with the cases of SM-2 (Fig. 5c). One can see that the acetonitrile solution of poly(SM-2-co-MMA) without the addition of water did not show any visual uorescence emission but exhibited the blue uorescence emission originating from the anthracene uorophore upon the addition of water (Fig. 3b). Indeed, the plot of the uorescence peak intensity at around 420 nm versus the water fraction in the low water content region below 1.0 wt% showed that the calibration curve had a good linearity with the m s value of 50 (Fig. 5d). However, the m s value (50) for poly(SM-2-co-MMA) was much smaller than that (355) of SM-2 (Table 1). Based on the calibration curve, the DLs and QLs of poly(SM-2-co-MMA) for water in acetonitrile were estimated to be 0.066 and 0.2 wt%, respectively, which were inferior to those (0.009 and 0.028 wt%) of SM-2. The deterioration of the DL and QL values of poly(SM-2co-MMA) compared with SM-2 may be attributed to the dynamic motion of the main chain, to which the anthracene skeleton was directly attached, and hydrophobic environment of the polymer chain, which can inhibit the interaction of the SM-2 moiety with water molecules, leading to the non-radiative decay of the photoexcited anthracene uorophore. In fact, in acetonitrile with 1.0 wt% water content, the F  (5%) of poly(SM-2-co-MMA was lower than that (12%) of SM-2. Nevertheless, it was found that the copolymer poly(SM-2-co-MMA) composed of SM-2 and MMA can act as a PET-type uorescent polymeric sensor for the detection and quantication of a trace amount of water in solvents.
Next, to evaluate the possibility for the PET-type uorescent sensor to function in polymer matrices for visualization and detection of water, we prepared spin-coated poly(SM-2-co-MMA) lms on glass substrates, and photoabsorption and uorescence spectra of the spin-coated poly(SM-2-co-MMA) lms before and aer exposure to moisture were repeatedly measured several times. In addition, PMMA lms doped with OF-2 or SM-2 at 15 wt% as well as 50 wt% were prepared on glass substrates by spin-coating process for comparison with the spin-coated poly(SM-2-co-MMA) lms, which contained ca. 15 wt% SM-2 unit ( Fig. 7 and 8). The as-prepared 15 wt% and 50 wt% OF-2or SM-2-doped PMMA lms as well as the poly(SM-2-co-MMA) lms (in dry process) showed a vibronically-structured photoabsorption band in the range of 300 nm to 400 nm originating from the anthracene skeleton (Fig. 7a, c, e and 8a, c). For the corresponding uorescence spectra in dry process, the 15 wt% OF-2or SM-2-doped PMMA lms and the poly(SM-2-co-MMA) a Slope of calibration curve. b Detection limit (DL) and quantitation limit (QL) of sensor for water. c Detection limit (DL) and quantitation limit (QL) of sensor for water. lms exhibited a feeble and vibronically-structured uorescence band with a l  max at around 415-430 nm in the range of 400 nm to 500 nm, which is attributed to the monomer emission originating from the anthracene uorophore in the PET active state (Fig. 7b, d and f), but the 50 wt% OF-2or SM-2doped PMMA lms showed a broad and feeble uorescence band with a l  max at around 450 nm in the range of 400 nm to 600 nm attributable to the excimer emission originating from the anthracene uorophore in the PET active aggregate state ( Fig. 8b and d). When all the OF-2or SM-2-doped PMMA lms and the poly(SM-2-co-MMA) lms were exposed to moisture (in wet process), the photoabsorption spectral shape did not undergo appreciable changes, although a slight change in the absorbance was observed due to the disturbance of the baselines in the photoabsorption spectra (Fig. 7a, c, e and 8a, c). For the 15 wt% OF-2or SM-2-doped PMMA lms and the poly(SM-2-co-MMA) lms, the corresponding uorescence spectra in wet process showed the enhancement of the vibronically-structured monomer emission band originating from the anthracene uorophore in the PET inactive state (Fig. 7b, d and f). On the other hand, the 50 wt% OF-2or SM-2-doped PMMA lms showed an appearance of the monomer emission band with a l  max at around 415-430 nm and the enhancement of the excimer emission band with a l  max at around 450 nm, that is, the enhancement of the broad uorescence band originating from the anthracene uorophore in the range of 400 nm to 600 nm arising from the PET inactive state upon exposure to moisture ( Fig. 8b and d). It is worth noting here that when all the OF-2or SM-2-doped PMMA lms and the poly(SM-2-co-MMA) lms aer exposure to moisture were dried in the atmosphere, the photoabsorption and uorescence spectra recovered the original spectral shapes before exposure to moisture. Actually, one can see that the 15 wt% SM-2-doped PMMA lms and the poly(SM-2-co-MMA) lms initially exhibited visually imperceptible blue emission in the PET active state but the visual blue monomer emission in the PET inactive state upon exposure to moisture (Fig. 3c and d). Meanwhile, the 50 wt% SM-2-doped PMMA lms showed feeble green excimer emission in the PET active state before exposure to moisture but the bluish green monomer and excimer emissions in the PET inactive state upon exposure to moisture (Fig. 9f). Therefore, for all the OF-2or SM-2-doped PMMA lms and the poly(SM-2-co-MMA) lms, the reversibility of the uorescence intensity at the l  max in the dry-wet process was investigated (Fig. 9a-e). It was found that the dry-wet cycles of the 15 wt% OF-2or SM-2-doped PMMA lms and the poly(SM-2-co-MMA) lms showed a good reversible switching of the uorescent intensity even in the ve times dry-wet process (Fig. 9a-c). However, the 50 wt% OF-2or (e) Photoabsorption and (f) fluorescence spectra (l ex ¼ 375 nm) of poly(SM-2-co-MMA) film before (in dry process) and after (in wet process) exposure to moisture. For all the photoabsorption spectra, baseline-correction was made to be the same absorbance at 420 nm. SM-2-doped PMMA lms showed that the uorescence intensity in the wet process was attenuated from the third time onward (Fig. 9d and e). The poor reversibility of the uorescence intensity of the 50 wt% OF-2or SM-2-doped PMMA lms may be attributed to destruction of the lms during the dry-wet process and/or promotion of aggregate formation of OF-2 and SM-2 aer dry process. We also measured the water contact angles on the polymer lm surfaces to investigate the hydrophilicity of the 15 wt% SM-2-doped PMMA lms and the poly(SM-2-co-MMA) lms (Fig. 10). The water contact angles on the polymer lm surfaces were 68.3 and 68.2 for the 15 wt% SM-2doped PMMA lms and the poly(SM-2-co-MMA) lms, respectively, clearly indicating that the hydrophilicity was similar to each other. Thus, the fact provides the evidence that the both lms shows similar reversible switching in the uorescent intensity in the dry-wet process. Consequently, this work demonstrated that PET-type uorescent polymer lms based on a uorescence enhancement system produce a satisfactory reversible uorescence off-on switching between the PET active state and the PET inactive state during the dry-wet process, and thus are one of the most promising and convenient functional dye materials to enable the visualization and detection of moisture and water droplets.

Conclusions
We have designed and developed the PET-type uorescent monomer SM-2 composed of methyl methacrylate-substituted anthracene uorophore-AminoMeCNPhenylBPin and the copolymer poly(SM-2-co-MMA) composed of SM-2 and MMA, as uorescent sensors for visualization, detection, and quantication of a trace amount of water. It was found that both SM-2 and poly(SM-2-co-MMA) exhibited enhancement of the uorescence emission with the increase in water content in various solvents (less polar, polar, protic, and aprotic solvents) due to the formation of the PET inactive (uorescent) species SM-2a and poly(SM-2-co-MMA)a, respectively, by the interaction with water molecules. The detection limit of poly(SM-2-co-MMA) for water in acetonitrile was 0.066 wt%, indicating that poly(SM-2co-MMA) can act as a PET-type uorescent polymeric sensor for a trace amount of water in solvents, although it was inferior to that (0.009 wt%) of SM-2. Moreover, we have achieved the preparation of the spin-coated poly(SM-2-co-MMA) lms as well as 15 wt% SM-2-doped PMMA lms and demonstrated that both the polymer lms produced a satisfactory reversible uorescence off-on switching between the PET active state under a drying process and the PET inactive state upon exposure to moisture. Consequently, this work proposes that PET-type uorescent polymer lms are one of the most promising and convenient functional dye materials to enable the visualization and detection of moisture and water droplets.

General
Melting points were measured with an AS ONE ATM-02. IR spectra were recorded on a SHIMADZU IRTracer-100 spectrometer by the ATR method. 1 H and 13 C NMR spectra were recorded on a Varian-500 FT NMR spectrometer. Highresolution mass spectral data were acquired by APCI on a Thermo Fisher Scientic LTQ Orbitrap XL. Photoabsorption spectra were observed with a SHIMADZU UV-3600 plus. Fluorescence spectra were measured with a Hitachi F-4500 spectrophotometer. The uorescence quantum yields were determined by a Hamamatsu C9920-01 equipped with a CCD using a calibrated integrating sphere system. The addition of water to 1,4-dioxane, THF, acetonitrile, or ethanol solutions containing SM-2 or acetonitrile solutions containing poly(SM-2co-MMA) was made in terms of weight percent (wt%). The determination of water in solvents was done with MKC-610 and MKA-610 Karl Fischer moisture titrators (Kyoto Electronics Manufacturing Co., Ltd) based on Karl Fischer coulometric titration for below 1.0 wt% and volumetric titration for 1.0-10 wt%. Polymer number-average molecular weights (M n ) and  molecular weight distributions (M w /M n ) were determined by size exclusion chromatography (SEC) at 40 C using a SHI-MADZU Prominence-i LC-2030 plus with a guard column (LF-G, Shodex), two series-connected columns (LF-804, Shodex), a UV detector, and a differential refractive index detector (RID-20A). THF was used as the eluent, and poly(methyl methacrylate) (PMMA) standards were used to calibrate the SEC system. Static water contact angles were measured at ve different positions on a substrate by the sessile drop technique using a Kyowa Interface Science DMo-602 contact angle meter.
Preparation of poly(SM-2-co-MMA) lm A solution of poly(SM-2-co-MMA) (8.0 mg) in toluene (0.4 mL) was stirred for 3 h at room temperature, while poly(SM-2-co-MMA) has dissolved quickly. To prepare a polymer lm, 150 mL of a poly(SM-2-co-MMA) solution was spin-coated (1000 rpm for 30 s) on a glass substrate (MIKASA MS-A-100 Opticoat Spincoater). The spin-coated lms were dried in air. The resulting poly(SM-2-co-MMA) lms were exposed to moisture for 60 s using a humidier.
Preparation of 15 wt% and 50 wt% OF-2-or SM-2-doped PMMA lms A solution of PMMA (8.5 mg and 5.0 mg for 15 wt% and 50 wt%, respectively) in toluene (0.5 mL) was stirred for several hours at 60-70 C until PMMA has dissolved, and then, OF-2 or SM-2 (1.5 mg and 5.0 mg for 15 wt% and 50 wt%, respectively) was added to the solution. To prepare a polymer lm, 150 mL of a OF-2-PMMA solution or a SM-2-PMMA solution was spin-coated (1000 rpm for 30 s) on a glass substrate (MIKASA MS-A-100 Opticoat Spincoater). The spin-coated lms were dried in air. The resulting 15 wt% and 50 wt% OF-2or SM-2-doped PMMA lms were exposed to moisture for 60 s using a humidier.

Conflicts of interest
There are no conicts to declare.