Synthesis and properties of photochromic polymer contain spiro-oxazine induced by ultraviolet light

ABSTRACT The special photochromic mechanism of spiro-oxazine (SO) refers to low thermostability and weak fatigue resistance. High light transmittance and biocompatibility are the two advantages of polymethyl methacrylate (PMMA), but it does not perform well in terms of tensile strength. Here, a new type of SO (1-(4'methacryloyloxybutyl)- 3,3-dimethyl-9'-hydroxy- 3H-spiro-naphthyl oxazine) was synthesized and introduced as a photochromic monomer into methyl methacrylate (MMA) system, which is an environment-friendly method. Compared with other photochromic polymers that have been reported, the advantages of this material mainly perform in the following areas: (1) The active group was introduced into the N-alkyl chain of SO and copolymerized with PMMA, and the fading rate of the photochromic polymer was dropped by 76.3% compared to SO. (2) Physical and chemical methods were used to improve the flexibility of PMMA-SO. The tensile strength of polymers has been reduced to 3.92 MPa. It also has a high elongation of 269.84% at a break. (3) PMMA-SO still has a significant fatigue resistance after 25 cycles of UV and visible light irradiation. In the future, it has potential applications in the fields of smart glass, data storage, biomimetic materials, and so on. Graphical abstract


Introduction
Photochromism, [1] as we know, means that compound A which occurs in a specific chemical reaction is isomerized to compound B under a certain wavelength of light, always accompanied by changes in color. According to the reaction mechanism, organic photochromic small molecules are divided into homolysis and heterolysis of bonds, pericyclic reaction, cis-trans isomerization, and so on. Spiro-oxazine has always been a concern due to its excellent photo-responsiveness. Without UV, SO is in a colorless closed-loop structure and the indole ring is orthogonal to the naphthalene ring, as shown in Scheme 1. Under UV, the bond of C-O in SO is broken. SO is isomerized into photo-merocyanine (PMC) and both the two ring systems are coplanar.
The photochromic history of organic small molecules has covered more than a century that has had a significant impact on the application of dynamic materials. [2,3] Researchers are particularly interested in dynamic materials [4] controlled by light since they can be easily adjusted with high precision with no chemical pollutants. [1,5] However, most photochromic units present in dynamic materials have several drawbacks, especially photo-fatigue and poor thermostability, that also limit their application. [6] Some researchers are attempting to solve these issues, but the results are inconclusive. [7] PMMA is a promising material with good mechanical qualities, great transparency, and human body histocompatibility. [8] It has been applied in a wide variety of industries, such as building, [9] automobiles, [10,11] and medicine. [12] But it would be much more attractive for scientists if it had lightresponsive features. Bahareh R. et al. [13] synthesized a few polymer assemblies with micellar geometry based on amphiphilic copolymer chains from PMMA and poly-n-isopropyl acrylamide (PNIPAM) with spiro-pyran chain end groups, where PNIPAM was responsive to temperature variation and PMMA with spiro-pyran was reversibly responsive to ultraviolet and visible light. The polymer assemblies could serve as novel multi-stimuli-responsive smart drug delivery systems for the controlled release of doxorubicin (DOX) upon different stimuli such as temperature, pH, and light.
Physical blending [14] and chemical bonding [15][16][17][18][19] are typically used to integrate photochromic compounds with polymers. Physical mixing has the apparent advantages of being a simple and viable modification procedure, low preparation cost, [20] and a more selective polymer matrix corresponding to the SO. However, this method does not improve photochromic molecule stability, fatigue resistance, or fading rate, and the number of photochromic small molecules added is still limited. [21] Xie et al. [22] studied the photochromic properties of SO when it is physically mixed with various basic materials and discovered that the photochromic rate and fatigue resistance of SO-doped SiO 2 films were better than SO-doped PMMA films. The essential reason is that PMMA is a viscous polymer, and the adhesion between PMMA and SO prevents SO from photoisomerizing to a level. As a result, if SO is chemically attached to the polymer, the flexibility of the chemical bond can expand the photochromic molecule's isomerization region and considerably improve its light response rate. Tam Huu Nguyen et al. [23] synthesized a kind of novel photo-controlled dynamic film named poly(MMA-r-MSP) by atom transfer radical polymerization, which has excellent properties of photoresponsivity. However, compared with PMMA, the thermal stability of poly(MMA-r-MSP) is not significantly improved, and the mechanical properties were not measured and characterized. The strategy of chemical bonding can also be used in biomedicine. Shahnaz Rahimi et al. [24] synthesized a kind of functional nano-particle polymer with SO-COOH and dextran using chemical methods. It is responsive to both UV and pH, which can be applied to control the release of drugs.
Based on current research progress, a newly synthesized spiro-oxazine monomer (SO) containing a functional group of methyl acrylate was introduced into PMMA by ultraviolet-light-induced radical polymerization. Adding a plasticizer and a new monomer to the photo-controlled dynamic material improved its quality even more. This photochromic polymer's thermal fading rate is substantially lower than in previous findings, indicating that the PMC form's thermostability is much higher. In addition, the polymerization efficiency of SO and MMA is also greatly improved. The following are the project's major innovations: (1) The reaction that broke the C=C bonds to form free radicals was triggered with clean energy, which reduced reaction time and simplified the operation. (2) The reactive monomer, MMA, served as both the reactant and the solvent in the reaction, which was environmentally friendly and saved solvents. (3) The film was formed by directly pouring the polymer solution after the polymerization into the mold, which could be adjusted for films of different shapes and sizes without the need for secondary pressing plate forming.

Materials
Anhydrous ethanol and methanol were dried with MgSO for 24 h and distilled at 77 ~ 78°C and 63 ~ 64°C, respectively. Other reagents were of AR grade and used without further purification. Meth acryloyl chloride and 3-methyl-2-butanone were purchased from Shanghai Aladdin Bio-Chem Technology Co., Ltd. (China). 2,7-Dihydroxy naphthalene was purchased from Shanghai Di Bai Biotechnology Co., Ltd. (China). 1,4-Dibromobutane was purchased from Shanghai Meiruil Chemical Technology Co., Ltd. (China). Other reagents were supplied by Chengdu Kelon Chemical Reagent Co., Ltd. (China). 1 HNMR (400 MHz) spectra were recorded on a Bruker AVANCE (Switzerland) with TMS as the internal standard. The solvent of the 1 HNMR spectrometry is CDCl 3 , and the solvent of MS is ethyl alcohol. Mass spectra were recorded on a VG Autospec 3000 spectrometer (UK). IR spectroscopy was performed on a Nicolet 170SX (USA) at room temperature and 40% humidity. Ultraviolet-visible spectroscopy was performed on a Mapada ultraviolet-1800 ultravioletvisible spectrophotometer (Shanghai, China) at room temperature. Mechanical properties were tested on an Instron 5567 electronic universal testing machine loaded with a 1 kN load cell at a rate of 50 mm/ min. At least three dumbbell-shaped specimens were prepared for each test. TG was carried out on SDT-Q600 thermo-analyzer instrument from 30°C to 600°C under a nitrogen atmosphere with a linear heating rate of 10°C/min.

Synthesis
The synthetic routes are shown in Scheme 2.
Tetraethylammonium bromide (0.5 g) was added serving as the phase transfer catalyst and the mixture was heated. The sodium methacrylate solution was slowly added, and the reaction continued for 48 h. The organic layer was separated, and the chloroform was removed by concentration. The residue was purified by column chromatography (petroleum ether:ethyl acetate = 4:1) and washed and recrystallized with methanol to obtain a white solid of 0.24 g (17.9%). MS (-C ESI) m/z: 470.22.

Preparation of polymer MMA-SO
The composition route is shown in Scheme 3 and 4. MMA monomer 1 (10 g, 100 mmol), SO monomer 2 (10 mg, 0.018 mmol), and photo-initiator 1-hydroxycyclohexyl phenyl ketone (0.1 g, 0.4 mmol) were mixed in a round-bottom flask. The mixture was then shaken at room temperature for 5 min and irradiated under a 400 W high-pressure ultraviolet lamp placed at 35 cm. After 30 min of irradiation, the resulting mixture turned sticky before solidifying into a rigid pale-yellow solid. The resulting solids were washed with ethanol to remove the unreacted monomers. The elution was repeated until the ethanol supernatant did not turn blue after being exposed to ultraviolet light, indicating that the unbonded SO molecules were completely removed. The undried precipitate was placed under ultraviolet light and turned from pale yellow to light blue, indicating that the SO monomer was successfully inserted into the polymer, giving the polymer photochromic properties.

Preparation of mixed PMMA+SO
To verify whether SO was introduced into PMMA by chemical bonds, PMMA powder and SO were dissolved in 5 mL dichloromethane, stirred evenly, and poured into the polytetrafluoroethylene mold. After the solution was completely volatilized, the FTIR spectra of the blend and the copolymer were measured, respectively.

FTIR
FTIR of the copolymer (the red curve in Fig.1) and blend polymer (the black curve in Fig.1) of SO and MMA were shown that the vibration peaks of the aromatic ring skeleton were 1480 cm −1 , 1557 cm −1 , and 1608 cm −1 , respectively, and the characteristic absorption peaks of the C=O bond in methyl acyloxy were located at 1730 cm −1 . The existence of these absorption peaks indicated that both of them contained SO groups. In comparison, the C=C stretching vibration peak of olefin in the blend curve appears at 1630 cm −1 , but not in the red curve, indicating that SO in the red curve is covalently connected with MMA, while the unpolymerized monomer of PMMA-SO in the black curve has been eluting in the form of physical blending. Therefore, SO is successfully introduced into PMMA through covalent bonding.

Discussion on Photochromic Properties of SO in Ethanol Solution
Firstly, we researched the photo-responsibility of a single organic molecule in ethanol solution. It can be seen from Fig. 2 that before UV excitation, the absorption of the solution in the visible region is close to 0 (curve A), and after UV excitation, strong and wide absorption peaks appear around 61 nm. The reason is that SO was in a colorless closed-loop structure and the indole ring was orthogonal to the naphthalene ring without UV. Under UV, the bond of C-O in SO was broken. SO was isomerized into photo-merocyanine (PMC) and both the two ring systems are coplanar. At the same time, the PMC of spiro-oxazine has many different isomers, so the absorption band is wide.  The ethanol solution of SO was successively irradiated under 125 W UV light for 5 s, 10 s, 15 s, and 20 s, and its visible absorption spectrum (500-700 nm) was quickly tested. As shown in Fig. 3, the intensity of spirooxazine at λ max increased as the extent of the UV radiation time. The phenomenon explains that more closedloop structures transform into PMC as the extent of the UV radiation time.

Polymerization Efficiency of PMMA-SO
To check the polymerization efficiency of SO and MMA, a standard curve was established to intuitively express the changes in polymerization efficiency with a different mass of added SO. Firstly, ultraviolet-visible spectroscopy of SO ethanol standard solutions of different concentrations (5 × 10 −5 mol/L, 4 × 10 −5 mol/L, 3 × 10 −5 mol/L, 2 × 10 −5 mol/L, 1 × 10 −5 mol/L, 5 × 10 −6 mol/L as 1 ~ 6) was performed. Secondly, the standard curve was established with the absorbance of SO ethanol standard solutions at the wavelength of 270 nm as shown in Fig. 4a. Regression of the solution concentration and absorbance was performed with fitting equation y = 0.181x -0.0014 as shown in Fig. 4b. Thirdly, after the photopolymerization of SO (M1) and MMA, the unpolymerized monomers were thoroughly eluted with anhydrous ethanol. All eluent was collected and diluted to 600 mL before ultraviolet spectroscopy. The mass (M 2 ) of unpolymerized monomers was calculated with the absorbance at 270 nm according to the standard equation. Then, the polymerization efficiency of SO was calculated through (M 1 -M 2 )/M 1 . According  to Table 1, the polymerization efficiency peaked at 92.65% when the proportion of SO was 0.4 wt%, which was much higher than the 78% that we reported before. [28] This may be attributed to the weaker steric hindrance of the active group methyl acrylate at the N position so that the polymerization efficiency is higher.

UV-Light Responsiveness
At room temperature (25°C), the initial polymerization product of SO and MMA was rigid with no photochromism under ultraviolet light. As the temperature gradually decreased, photochromism appeared and UV-light responsiveness enhanced because the polymethyl methacrylate polymer chain became more flexible, resulting in a higher ring-opening rate. Two methods could be adopted to improve the flexibility of the polymer to verify the effect of film flexibility on the photochromic properties: (1) adding a certain mass fraction of plasticizer to the polymer and (2) embedding a softer polymer segment in the monomer.
Six circular films were prepared by adding plasticizer dibutyl phthalate (DBP) of different qualities to the copolymerization product with unreacted monomer SO eluted. The proportion of the plasticizer DBP to the total mass was 0 wt%, 5 wt%, 10 wt%, 15 wt%, 20 wt%, and 25 wt%, named d 1~ d 6 , respectively. As for the other method, the mass ratio of SO was set to 0.1%, and different ratios of MMA and butyl acrylate (BA) were initiated, eluted, and dried to prepare circular films b 1 ~ b 6 . The BA ratio of b 1~ b 6 to the total mass was 0 wt%, 10 wt%, 20 wt%, 30 wt%, 40 wt%, and 50 wt%, respectively.
The films d 1~ d 6 and b 1~ b 6 all turned blue under ultraviolet light of a 125 W high-pressure mercury lamp (365 nm) for 20 s and slowly faded in the dark. Fig. 5 shows the ultraviolet light stimulated coloration of the films d 1~ d 6 and b 1~ b 6 at ordinary temperature. As the content of BA and DBP increases, the coloration of the film deepens. The ultraviolet absorbance of PMMA-SO films with different mass ratios of DBP and BA at the maximum absorption wavelengths (λ max ) is shown in Fig. 7. The absorbance in λ max of polymer films peaked when the BA content was 50 wt% (b 6 ) or the DBP content was 20 wt% (b 5 ). It is indicated that the polymer films manifest varied absorbance under ultraviolet radiation depending on the flexibility of the molecular chain. When the polymer chain is softer, the ringopening resistance of SO is weaker, which results in the UV-light responsiveness of the polymer is greater. From Fig. 5, we found that the photochromic effect of d 5 and b 6 was the best. We selected d 5 and b 6 as experimental subjects for the next research.     At a temperature of 0°C, the films d 5 and b 6 were irradiated under a 125 W UV lamp for 20 s, and their UV-visible absorption spectra in the wavelength range of 500-700 nm were measured. As shown in Fig. 6, the maximum absorption wavelengths λ max of film d 5 and b 6 after UV excitation were 610 nm and 614 nm, respectively.

Thermal Fading Rate
To explore the influence of polymer flexibility on the thermal fading rate, the absorbance (A t ) of SO ethanol solution of 5 × 10 −5 mol/L at λ max was measured immediately after a 20s exposure to a 125 W highpressure mercury lamp (365 nm). The absorbance of the SO ethanol solution after UV stimulated coloration was recorded as A ∞ and the absorbance after placing it in the darkroom for 24 h was recorded as A 0 . The absorbance values of d 5 and b 6 were measured and recorded in the same way. According to Fig. 8, the absorbance of the three samples at λ max gradually decreased with time before reaching equilibrium, indicating that ring-closing had occurred. The relationship between absorbance and time (0 ~ 60 s) of the three samples at λ max was simulated by first-order kinetic curves as shown in Fig. 9 (a-c). The fitted kinetic curves were all straight lines with correlation coefficients (R) all above 98%. The fading rates, and the slope of the fitted lines, of the two films, were significantly lower than those of the SO ethanol solution, and film b 6 had the lowest. We analyzed this result and concluded that the ringclosing process of SO molecules in film 6 was slow due to steric hindrance between polymer segments. Especially, when SO was covalently bonded with the polymer chain, the space hindrance of the ringclosing process was significantly greater than that with the physically added plasticizer DBP. According to Table 2, we found that compared with SO in the ethanol solution, the fading rate of d 5 and b 6 dropped by 73.5% and 76.3%, respectively.

Thermostability Analysis of the Polymers
The thermal decomposition temperature of the polymer was measured with a thermogravimetric analyzer on samples of 1 mm × 1 mm under the protection of nitrogen. The heating range was 20 ~ 200°C at 10°C/ min. The thermogravimetric analysis of polymers PMMA, d 1, d 5 , and b 6 was performed, showing that the initial thermal decomposition temperature of b 6 was much higher than that of the other three samples (Fig. 10a) and phase separation did not occur in the whole heating process (Fig. 10b). It is indicated that the polymerization of PMMA-SO with soft segment BA by covalent bond not only increases the softness of the polymer segment but also has a stronger rigid structure and stronger thermal stability than the method of physical doping plasticizer.

Mechanical Properties Test the Polymer
The weight-average molecular weight and average molecular weight distribution of PMMA, d 1 , d 5 , and b 6 were measured by polymer gel chromatography as shown in Table 3. The average molecular weights of the polymers b 6 were slightly lower than those of the PMMA blank sample. This may be due to the inhibitory effect of the hydroxyl group on polymerization. As for the mechanical properties compared with PMMA blank sample, the tensile strength of d 5 and Figure 8. The absorbance at λmax of SO in ethanol and copolymers after irradiation . b 6 decreased significantly, while the elongation at break of it increased significantly. The result suggests that the polymer segment of b 6 has higher flexibility on the premise of guaranteeing a certain degree of polymerization.

Fatigue Resistance Test
To test the effect of introducing SO on the fatigue resistance of the polymer, the absorbance at λ max of an SO ethanol solution (5 × 10 −5 mol/L) and films d 5 and b 6 was measured immediately after exposure to a 125 W high-pressure mercury lamp for 20 s. The test subjects were then transferred to a a dark room to let the coloration fade completely before being irradiated again for 20 s and measuring the absorbance at λ max . The process was repeated 25 times to see the changes in absorbance with the increase in cycles as shown in Fig. 11.
After UV radiation over 15 cycles, the absorbance of SO in ethanol at λ max decreased slightly (Fig. 11a). But according to Figs. 11(b-c), the absorbance of film b 5 and d 6 showed no significant trend of decrease, indicating good fatigue resistance. This was mainly because the smaller free space between spiro-oxazine molecules in     the polymer chain segments hindered the photooxidation of the ring-opening body. Therefore, introducing SO into the polymer system could significantly improve its antifatigue properties.

Conclusion
SO was successfully designed and synthesized before being introduced into PMMA by chemical bonding to improve its thermostability and fatigue resistance. According to the experimental results, the polymerization efficiency was as much as 92.65% when the proportion of SO was 0.4 wt%, which was much higher than we previously reported. Adding plasticizer or polymer soft segment to the polymer system would improve the polymer's flexibility. With BA content of 50 wt% or DBP content of 20 wt%, the polymer films showed apparent and stable photochromic phenomenon, good thermostability, fatigue resistance, and mechanical properties. Moreover, the polymerization of SO with MMA was triggered by ultraviolet light, with MMA serving as the benign monomer so that no solvent was needed. All monomers were polymerized into a viscous gel that only needed to be imprinted once. The polymerization processes conformed to the concept of green chemistry and were very easy to operate. All things considered, this study provided a new idea for improving the thermostability and fatigue resistance of photochromic materials. The novel photochromic material could be modified for applications in the field of optical information storage, building decoration, protection, and anticounterfeiting.