An easy and fast fabrication strategy to obtain Photochromic Films (PFs) for naked-eye detection of oxygen is presented. These PFs are based on the photoreductive activity of TiO2 nanoparticles combined with the redox-driven color switching property of methylene blue, embedded in a photocurable and tunable air-permeable polyethylene glycol diacrylate (PEGDA) matrix. The PF is fabricated by a single-step process: the UVA light exposure initiates the polymerization and simultaneously reduces the blue-color dye in its colorless form. The resulting PF exhibits fast discoloration and modulable recoloration time in the air. The tunability of PFs color-switching can be used for engineering colorimetric sensors with preset oxygen responsive ranges to fulfill specific application requirements.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Smart materials, able to undergo reversible color switching in response to external stimuli, have recently gained particular attention due to their strategic role in practical applications as sensor systems. Real-time oxygen detection is relevant in different fields of science and technology, [1–3] and color switching systems, appeared as a valid and cheaper alternative to traditional oxygen sensing methods (e.g., electrochemical sensing, chromatographic analysis), that instead, require expensive instruments and trained operators [4–6]. For these reasons, optical oxygen detection methods have been exploited in recent years, since they are easy and convenient to use, cost-effective, and non-destructive. Colorimetric sensors refer to a class of smart materials containing molecules (e.g. azobenzenes, spiropyrans, diarylethenes, etc.), [7–11] that can be reversibly switched between two chemical forms through the illumination of light. In particular, colorimetric sensors for detecting oxygen can be activated “on-demand” and remotely by light, undergoing reversible color changes that can be switched back by oxygen during exposure to air. Most colorimetric oxygen sensors are based on a color transition due to a reaction of a reduced dye with oxygen. Methylene blue (MB) is one of the most common redox dyes, due to its easy reduction to the colorless leuco-MB (LMB) phase and its fast re-oxidation in presence of oxygen [12,13]. In absence of oxygen, LMB remains in its colorless state, but upon exposure to air, it is re-oxidized to the blue MB form, thereby acting as an O2 indicator . The MB reduction can occur as a result of an electrochromic reaction using various reducing agents (e.g. citric acid, sodium citrate, methanol), [15,16] or as a result of electron and proton transfer from a UV-absorbing photocatalyst (e.g., TiO2, ZnO, SnO2 nanoparticles) [7,17,18]. Many strategies have been proposed to develop stable solid colorimetric films as oxygen sensors, including redox dye in solid media that restrict their molecular mobility. The color switching system can be embedded into different polymers, e.g., gelatin, hydroxyethylcellulose, polyvinyl alcohol, ethylcellulose, and polypyrrolidone to obtain thin solid photochromic films [7,17,19,20] that exhibit recoloration times depending on the oxygen permeability of matrices. In the majority of cases, the fabrication of PF devices requires a first step of many hours to dry the photochromatic matrix at high temperature to obtain a solid film and a second step of UV exposure to photo-reduce the MB to the colorless LMB [19,21]. Here, an innovative PF, that exploits titania nanoparticles (TiO2-NPs) and MB dye in a UVA-photocrosslinkable polymer, is proposed. This system allows the dye activation and matrix fabrication in one single step that lasts a few minutes. The transparent PEGDA polymer is an ideal matrix to encapsulate the sensing dye molecules and ensuring photoresponsive stability of the solid and flexible film [22,23]. Furthermore, varying the molecular weight (MW) of PEGDA, the mechanical and diffusive properties of the film can be changed affecting the PF recoloration time from colorless to blue. We have performed a study on the response times of PFs with different PEGDA molecular weights after their exposure to ambient air, which could be of interest in real-life applications, such as monitoring of leakages in packaging, as an example. This system responds to the presence of O2 through a clear color change, visible by the naked eye [24–26]. Moreover, its tunability in color recovery response opens up to a wide range of applications from food, pharmaceutical [7–10,27–39], data recording [40–42] industries, to new fields as indicator controllers to prevent baggage door opening. Thus, we have explored a smart method to quickly fabricate flexible photochromic films whose photochromic properties can be tuned according to the desired application.
In this work, tunable polymeric oxygen-sensitive films based on the color switching of redox dye using titanium oxide-assisted photocatalytic reaction were fabricated. The system contained MB as redox dye, which was photoreduced to the colorless LMB phase by the electrons of titanium dioxide nanoparticles (Fig. 1(a)). TiO2 NPs were synthesized through a high-temperature hydrolysis reaction in presence of a nonionic polymer capping, i.e., poly (ethylene glycol)-block-poly (propylene glycol)-block-poly (ethylene glycol) (P-123), which binds the NPs and acts as sacrificial electron donor (SED). On excitation by UVA illumination (365 nm), the TiO2 NPs electrons are promoted from the valence to the conduction band of the UV-absorbing photocatalyst and create electron-hole pairs. The photogenerated holes by the TiO2 NPs are sequestered by P-123, and the relative photogenerated electrons are free to reduce the redox dye (MB) to its colorless LMB form . Both the TiO2 particles and MB have low toxicity and costs and have already been widely used in cosmetic, medical, and other industries. In addition, the color switching scheme could be extended to other redox dyes . The TiO2 NPs and MB were mixed with PEGDA containing a photoinitiator (PI), to form a blue dispersion. Darocur 1173 is a common PI used to initiate the photopolymerization process, thus, chosen for preliminary tests. PI represents only a minor part of the formulation ranging from 0.025 to 2%v/v, namely in no harmful concentration. Anyway, other photoinitiators, including lithium phenyl-2,4,6-trimethyl-benzoyl phosphinate (LAP), which has low toxicity, can be selected. Several pre-polymeric mixtures were prepared with PEGDA at different MW to obtain PFs with different mechanical properties. The liquid prepolymer was sandwiched between two quartz plates, which also prevented the diffusion of O2 into the system during polymerization, and exposed under UV light obtaining a solid bleached PF. The blue PF presented a maximum absorbance at 665 nm, related to the oxidized-MB. Under UVA irradiation, the absorption peak decreased over time due to the reduction to LMB, which resulted in a colorless PF (Fig. 1(b)). Upon contact with oxygen, the leuco dye oxidized to MB in about three hours that entailed a color transition from colorless to blue. This strategy allows the fabrication of robust and flexible films, as shown in Fig. 1(c).
Introducing TiO2 nanoparticles into the polymeric acrylate matrix can lead to a non-homogenous scattering profile and modify the transparency of materials. The uncontrolled size of the NPs (aggregation phenomenon) can affect the photopolymerization process. To investigate the light scattering impact on PF fabrication, diverse batches composed of multiple syntheses of NPs TiO2 were compared (see Fig. S1 in supporting information). As shown by the transmission electron microscopy (TEM) images in Fig. 1(d), the as-synthesized TiO2 NPs have irregular shapes with an average size of 50 nm. NPs-TiO2 was added in ultrapure water and sonicated for 30 minutes prior to Dynamic Light Scattering (DLS) measurements. In Fig. 1(d), the TiO2 NPs size (50 ± 20 nm) and the absence of aggregates in the suspensions were confirmed by DLS. Further, pre-polymer mixtures (TiO2/MB/PI) were photopolymerized and bleached by UV light. All mixtures of different batches discolored uniformly in 2 minutes and recovered the initial blue color with comparable trendlines, indicating that the formation of possible NP aggregates has a relevant impact neither on the penetration of UV-light exposure nor on the switching phenomenon (see Fig. S1 in supporting info).
Once the UVA light was turned off, the UV-activated oxygen indicator remained bleached indefinitely in absence of oxygen while its color was restored in the air at a rate depending on the PEGDA MW. The PFs were positioned in a sealed optical chamber in the pure N2 atmosphere, to investigate the PF behavior in absence of O2. The PF remained in its colorless state for as long as N2 was fluxed into the chamber as confirmed by absorption spectra at 665 nm recorded over the time and reported in Fig. S2 of the supporting info.
The film recoloration process is due to the LMB re-oxidation in presence of O2 by the following reaction (Eq. (1)):
The LMB oxidation is connected with the oxygen diffusion capability through the solid film. In general, the permeability of plastic films at a given temperature depends on the morphology, molecular weight, and, in the case of copolymers, also on the composition. For this reason, the recoloration time of PFs, dependent on their oxygen permeability, was investigated by changing the molecular weight of PEGDA polymers. The MW and the crosslinking density affect the glass transition temperature of PFs. To clarify their role in the rate of the recoloration process, PEGDAs with a molecular weight ranging from 250 to 10.000 Da were investigated. However, phase separation was observed in the prepolymer solution containing PEGDA 250 and the aqueous solution of TiO2 NPs, thus it was excluded from further investigation. On the contrary, a good dispersion of the TiO2 NPs was obtained in prepolymer solutions containing PEGDA at higher MWs after a vigorous stirring. Prepolymer solutions containing TiO2/MB/PI mixed with PEGDA 575, 700, and 10.000 Da, respectively, were exposed to UVA light for 2 min obtaining bleached PFs with a thickness of 150 µm. TiO2/MB/PI/PEGDA575 (PF575) and TiO2/MB/PI/PEGDA700 (PF700) were more rigid PFs than TiO2/MB/PI/PEGDA10000 (PF10000), which appeared as a highly flexible film. All PFs maintained their colorless states when placed between glass slides in a nitrogen atmosphere, proving the action of molecular oxygen as the oxidizing agent. In fact, when the polymerized PFs were extracted from the glass coverslips and exposed to the oxygen in the air, LMBs began to return to their original blue color. PFs exhibited different color-switching rates from the colorless state to the blue color depending on the PEGDA MWs used to fabricate films, as observed in Fig. 2(a). The UV-VIS absorption spectra of PF575, PF700, and PF1000 over the time, when exposed to ambient air, are reported in Fig. 2(b); the increase in absorbance over time is attributed to the blue coloration of the PFs. In Fig. 2(c), the kinetic recoloration as a function of PEGDA MW is reported. The dynamic of the recoloration process was highlighted by monitoring the absorption peak at 665 nm from 1 to 150 min after UV-bleaching. The absorbance at 665 nm gave information about the yield of the LMB oxidation (Eq. (1)). On increasing the polymer MW, the LMB recoloration time was faster. In more details, PF10000 exhibited about 67% of recoloration, while PF575 only reached 17% after 15 min in the air. An intermediate value of about 38% was obtained from PF700. The MB absorbance intensity reached a plateau when LMB was completely re-oxidized in air, exhibiting a diffusive regime. The LMB oxidation speed seemed to be influenced by oxygen diffusion in the different PFs. The increase of the PEGDA MW resulted in a larger mesh size (ɛ) and a higher average molecular weight between crosslinks (Mc). Larger ɛ together with higher Mc could explain the different recoloration rates of PFs from 575 to 10000. These factors influenced the mobility of the polymer chain and thus, directly the glass transition temperature (Tg) entirely correlated with the LMB oxidation rate. An estimation of the oxygen diffusion coefficient (D) in PF575, 700, and 10000 was provided at 25 °C by the following equation:2(c) reports the increase of LMB reconversion into MB over time. When the absorbance of MB stopped increasing, the diffusive oxygen reached the most remote point of the acrylic film and it corresponded to the diffusion time (t). The t values extrapolated for PF575, PF700, and PF10000, were 160 min, 120 min, and 90 min, respectively.
As a consequence, oxygen diffusion coefficients in the three PFs had values of 2.9 × 10−13, 3.9 × 10−13, and 5.1 × 10−13 m2s-1, respectively (Fig. 3(a)). The air diffusion coefficient of each PFs was correlated with the Tg of the PEGDA films. The glass transition temperatures of polyacrylate films were determined by DSC and given in Fig. 3(a). As hypothesized, the rate of LMB oxidation decreased with the increase of PF Tg. This result confirms the possibility to tune PFs recoloration times with systematic variation in MWs without changing other properties, such as the chemical composition. Once the action of the molecular weight on the color-switching had been ascertained, we focused on the effect of temperature on PFs. Since PFs have the potential to be used as O2 indicators for food packaging, it was important to monitor their performance at different temperatures. We chose to investigate as proof of concept PF575 at 4 °C, representative of food storage within a fridge, and 40 °C, ascribable to daily summer maximum temperature. In Fig. 3(b), the recovery profiles, after photoactivation, at 4 °C and 40 °C are compared with PF575 at room temperature. The recovery reaction rate is much reduced at the lower temperature and slightly faster at 40 °C in comparison to the recoloration rate at room temperature.
Furthermore, PFs exhibited photoreversible color-switching properties: discoloration in a short period of a few minutes upon UV illumination and recoloration in hours when in ambient air. The maximum absorbance intensity of MB was obtained for all PFs for at least two cycles of reduction-oxidation. PFs irradiated by a third cycle of UV-light in the air did not return completely to their original blue color after oxidation. The surface-bound SEDs on TiO2 NPs could be gradually consumed by photogenerated holes during UV illumination and eventually exhausted. Further, a prolonged period of UV irradiation in the presence of oxygen led to partial decomposition of MB. However, a disposable colorimetric sensor should be preferable for specific applications.
The photoreductive ability to switch the original blue color (MB) to colorless (LMB) dye of PFs was due to the synergic action of both TiO2 NPs and PI. To investigate the action of each component, colorimetric films were fabricated in three different combinations: PF without TiO2 NPs (PEGDA/MB/PI), PF without PI (PEGDA/MB/TiO2), and PF system (PEGDA/MB/PI/TiO2) in Fig. 4(a). In the absence of TiO2 (PEGDA/MB/PI), MB molecules were reduced into LMB only by radical species generated from PI. Darocur molecules underwent photocleavage upon UV exposure and produced radicals, which initiated the radical polymerization and reduced the oxidized dye, contemporary. Furthermore, PEGDA/MB/PI did not come back to its original color in the air after a second cycle of UV-light (2 min) exposure, showing a non-reversible behavior. In the case of PEGDA/MB/TiO2, TiO2 generated free electrons, which did not just trigger the MB-bleaching but also initiated the radical polymerization of PEGDA. In Fig. 4(b), the absorbance of PF575 and PF700 in comparison respectively with PEGDA575/MB/TiO2 and PEGDA700/MB/TiO2, at 5 min, 30 min, and 90 min are shown. In Fig. 4(c), kinetic studies of PF and PEGDA/MB/TiO2, fabricated with the same polymerization parameters but in absence of PI, are reported. Differently from PFs (max. absorbance around 1.0), PEGDA/MB/TiO2 did not return to their original blue color after oxidation in the air (max. absorbance around 0.2). After 5 min, the absorbance spectrum of PF575 showed an intensity comparable to that of PEGDA575/MB/TiO2. After 30 min, PF575 exhibited 3 times higher absorbance intensity concerning PEGDA575/MB/TiO2, and, finally, after 90 min PF575 showed a 10 times higher absorption than PEGDA575/MB/TiO2. Analogous behavior was observed for PF700 and PEGDA700/MB/TiO2. This could be explained by the inner filter effect of Darocur 1173 . The PI molecules and their photolysis products, decreased the penetration of the incident light, preserving the dye degradation. In absence of the Darocur filter effect, the intense UV-irradiation could strongly affect the capability of MB re-oxidation. Lower exposure times could reduce the MB degradation, however, the complete photopolymerization of acrylate monomers was not guaranteed in these conditions. The comparison between PF and PEGDA/MB/TiO2 showed different rates of LMB re-oxidation in the air for systems containing PEGDA575 and PEGDA700. The same result was not achievable with PEGDA10000. In this last case, PEGDA10000/MB/TiO2 prepolymer appeared still liquid at the same UV-irradiation time and no-polymerization occurred. Longer exposure times were not enough to guarantee the polymerization of PEGDA10000/MB/TiO2 prepolymer, which did not occur in absence of PI for PEGDA10000. To summarize, PEGDA/MB/TiO2 exhibited poor recoloration capability.
Moreover, the possibility to introduce further polymers layers on the top and/or on the bottom of PFs to tune the recoloration rates and time intervals was, herein, studied. Polymers such as polyethylene terephthalate, polyethylene, polypropylene, are generally used as oxygen barrier films overlaying the colorimetric oxygen indicator [26,45]. In this work, the recovery of the color of PFs in ambient air was slowed down by the PEGDA cover . A prepolymer solution containing PEGDA/PI was dropped on top of PFs, covered by a glass slide, and cured by UVA directly on blue film. The same procedure was carried out for the bottom part of PFs. The final result was a multilayer photochromic film (m-PF), having PF completely embedded in additional PEGDA layers. The sandwich structure, obtained by PF enclosed in two transparent layers, was employed to reduce the oxygen permeability. In Fig. 5(a), the absorbance spectra at different times of PF575 and m-PF575 after bleaching, were reported. As shown in Fig. 5(b), the PF showed a full coloration after 160 min, while the m-PF reached a complete re-oxidation of LMB after about 2 days. The slow recoloration rate of the m-PF was clearly due to the adding of the cover layers that affected the oxygen diffusion, and thus, the LMB oxidation.
A new strategy to obtain PFs with tunable oxygen-responsive time was engineered. PFs were composed of TiO2 NPs, MB, PI in a PEGDA matrix at different MWs. PFs pre-polymers were illuminated by UVA, obtaining solid and colorless PFs. Their recoloration rate in the air depends on oxygen diffusion through the thick, solid, and flexible film. Oxygen permeability could be modulated by varying the physical properties of PFs, in particular the MW of the polymer matrix. PF575 was fully colored again in about 160 min, slower than PF700 (120 min) and PF10000 (90 min). Nevertheless, a slight recoloration is already appreciable at 5 min for all PFs. In addition, qualitative information about the elapsed time after the first exposure of PFs to oxygen can be provided to the user. Further, the recovery reaction rate has been demonstrated to be dependent on temperature, thus for practical applications, a control standard i.e., completely oxidized PF, could be provided in parallel with the reduced PF. To slow down further the recoloration rate of PF575, it was covered by both slides with two PEGDA layers, obtaining multilayer PFs. The re-oxidation of LMB in m-PF575 lasted 2 days, in comparison to PF575 which returned to the original blue color in the order of tens of minutes. To conclude, the time of color transition in PFs could be modulated by choosing the appropriate polymer molecular weight. Suitable physical properties have been investigated to tune the performance of the PFs to fulfill the requirements of different applications.
4. Experimental section
Titanium(IV) chloride (TiCl4), poly(ethylene glycol)-block-poly(propylene glycol)-block-poly(ethylene glycol) (P-123), methylene blue (MB), diethylene glycol (DEG), Poly(ethylene glycol) diacrylate average Mn 250, 575, 700, 10.000 (PEGDA), Darocur 1173, were purchased from Sigma Aldrich.
4.2 Synthesis of TiO2 NPs
Titanium dioxide was synthesized following the protocol . A mixture of poly(ethylene glycol)-block-poly(propylene glycol)-block-poly(ethylene glycol) (P-123, 300 mg), titanium tetrachloride (TiCl4, 0.25 mL), NaOH (0.25 mL), and diethylene glycol (DEG, 10 mL) was heated to about 250° C in a 50 mL flask with vigorous stirring. The resulting mixture was heated at 250 °C for 3h. Then, the solution was cooled to room temperature, and a white precipitate was obtained by adding acetone and centrifuging at 11000 rpm for 10 min. The product was washed several times with acetone and water to remove any residues.
The absorption spectrum of TiO2 nanoparticles in H2O was analyzed by UV/Vis (UV-VIS, Cary 100 spectrometer, VARIAN). The absorption spectrum of TiO2 NPs exhibited a strong absorption from 250 to 375 nm. The NPs size of 50 nm was evaluated with a Dynamic Light Scattering (DLS, Zetasizer Nano ZS, Malvern Instruments equipped with a HeNe laser 633 nm, fixed scattering angle of 173°, 25 °C). The refractive index (n) and absorption (k) parameters to obtain the volume percent intensity (%) of TiO2 NPs were set at n = 2.4 and k = 0.002, respectively.
4.3 Transmission electron microscopy
The morphology of TiO2 NPs was investigated via a transmission electron microscope (TEM, Jeol JEM-1400, Jeol Ltd, Japan). Samples were dropped on a carbon-coated copper TEM grid before air-drying overnight at room temperature. Imaging was performed at 120 kV at a magnification of 340000✕.
4.4 Photochromic film fabrication
The composition consists of a solution of MB (0.1M, 10 µl), PEGDA (500 µL), Darocur 1173 (2% v/v) and a water solution of TiO2 (100 mg/mL, 50 µL). The mixture of TiO2, acrylates, MB, and photoinitiator TiO2/PEGDA/MB/PI defined pre-polymer PF was stirred for about 10 min to obtain a homogeneous blue solution. Different mixtures were obtained with PEGDA at different MWs, i.e. 250, 575, 700, 10.000. Furthermore, mixtures without PI or TiO2 were prepared following the same protocol. Then, a pre-polymer PF mixture (100 µL) was placed on a glass substrate 15 × 15 mm and covered upside with another glass. The mixtures containing PEGDA 575 and 700 were irradiated by UVA-exposure box (UV-Belichtungsgerät 2) for 2 minutes, while ones with PEGDA10000 for 5 minutes. After UV illumination, the pre-polymer resulted in a solid, flexible, colorless PF. The PF575, PF700, and PF10000 were fabricated in triplication by three different TiO2-NPs synthesis. The kinetics of recoloration are reported as the average with standard error of the mean (SEM).
4.5 Thermal analysis
Differential scanning calorimetry (DSC) was performed with TA Discovery DSC, in the presence of nitrogen purge at 10 °C/min. Approximately 5 mg of the sample was sealed into a Tzero pan with a hermetic Tzero lid; TA TRIOS software was used to analyze the DSC thermograms.
4.6 Experimental conditions
The experiments of PF recoloration were performed under laboratory conditions at ambient air at room temperature. All UV/vis spectra and digital photographs were recorded in the same conditions. The PFs recoloration kinetic was also conducted in a sealed optical chamber with nitrogen with a flux of 1000 sccm. The PF remained in its colorless state in the pure N2 atmosphere, for the total duration of flux (3 h), while it changed slowly to blue color when re-exposed to the atmosphere (see support information). The measurements were also conducted at different temperatures i.e., at 4 °C storing PF575 in a fridge, and at 40 °C by placing PF575 in the oven. The absorption spectra were acquired after 15, 30, 60, and 90 min.
The absorbance spectra of all the fabricated PFs were measured through a transmission mode customized setup by exploiting the high transparency of the films. The measures were performed at normal incidence. Briefly, the optical setup consists of a halogen light source with a high spectral resolution in the visible and near-infrared spectrum (400-1600 nm). The light is conveyed to the sample by a Thorlabs optic fiber equipped with a collimator. The transmitted light was collected by an optical spectrum analyzer (ANDO AQ6315B). The spectra were collected at suitable time intervals to achieve a time-dependent recoloration profile in the range 500-800 nm and the MB absorbance peak time-variation was monitored by performing a peak analysis at 665nm on the collected spectra. For each PF, the measurements were performed on three different points along with the 15×15 mm patch at each time interval and the resulting data were obtained from their average in post-processing. The absorbance of the oxygen indicator at time 0 (corresponding to the fully transparent sample in which the recoloration was not started yet) was used as a reference standard.
The authors are grateful to Mr. Vitaliano Tufano for the support of experiments in an inert atmosphere and to Dr. E. Amendola and A. Martone of the Institute of Polymers, Composites, and Biomaterials (IPCB) of CNR for helpful discussions.
The authors declare no conflicts of interest.
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
See Supplement 1 for supporting content.
1. W. Boonsong, S. Chimparos, O. Adeleke, and W. Ismail, “Real-time air monitoring using oxygen sensor embedded RFID with wireless mesh sensor network platform, ICEIC 2017 International Conference on Electronics, Information, and Communication, 472–477 (2017).
2. C. Chue, C. J. Leo, W. P. Chan, M. R. B. Damalerio, M. Cheng, J. H. Cheong, and Y. Gao, “Characterization of CMOS electrochemical oxygen sensor for biomedical applications,” in 2015 IEEE International Conference on Electron Devices and Solid-State Circuits (EDSSC) (2015), pp. 325–328.
3. D. Lazik and H. Geistlinger, “Oxygen sensor for real-time and in situ long-term monitoring,” Groundwater 2000, 429–430 (2020).
4. M. L. Hitchman, Measurement of Dissolved Oxygen (Wiley, 1978).
5. S. J. Valenty, “Gas chromatographic determination of dissolved hydrogen and oxygen in the photolysis of water,” Anal. Chem. 50(4), 669–670 (1978). [CrossRef]
6. X.-D. Wang and O. S. Wolfbeis, “Optical methods for sensing and imaging oxygen: materials, spectroscopies, and applications,” Chem. Soc. Rev. 43(10), 3666–3761 (2014). [CrossRef]
7. Z. Gao, L. Liu, Z. Tian, Z. Feng, B. Jiang, and W. Wang, “Fast-response flexible photochromic gels for self-erasing rewritable media and colorimetric oxygen indicator applications,” ACS Appl. Mater. Interfaces 10(39), 33423–33433 (2018). [CrossRef]
8. S. Y. Xie, X. H. Liu, H. B. Li, and C. Huang, “The application of oxygen indicator in food packaging,” Adv. Mater. Res. 945-949, 2037–2042 (2014). [CrossRef]
9. C. Shillingford, C. W. Russell, I. B. Burgess, and J. Aizenberg, “Bioinspired artificial melanosomes as colorimetric indicators of oxygen exposure,” ACS Appl. Mater. Interfaces 8(7), 4314–4317 (2016). [CrossRef]
10. M. Ghaani, C. A. Cozzolino, G. Castelli, and S. Farris, “An overview of the intelligent packaging technologies in the food sector,” Trends Food Sci. Technol. 51, 1–11 (2016). [CrossRef]
11. S. De Martino and P. A. Netti, “Dynamic azopolymeric interfaces for photoactive cell instruction,” Biophys. Rev. 1(1), 011302 (2020). [CrossRef]
12. S.-K. Lee and A. Mills, “Novel photochemistry of leuco-methylene blue,” Chem. Commun. 18(18), 2366–2367 (2003). [CrossRef]
13. Y. Ohko, T. Tatsuma, T. Fujii, K. Naoi, C. Niwa, Y. Kubota, and A. Fujishima, “Multicolour photochromism of TiO2 films loaded with silver nanoparticles,” Nat. Mater. 2(1), 29–31 (2003). [CrossRef]
14. A. Mills, “Oxygen indicators and intelligent inks for packaging food,” Chem. Soc. Rev. 34(12), 1003–1011 (2005). [CrossRef]
15. A. Mills and J. Wang, “Photobleaching of methylene blue sensitised by TiO2: an ambiguous system?” J. Photochem. Photobiol., A 127(1-3), 123–134 (1999). [CrossRef]
16. N. R. de Tacconi, J. Carmona, and K. Rajeshwar, “Reversibility of photoelectrochromism at the TiO2/methylene blue interface,” J. Electrochem. Soc. 144(7), 2486–2490 (1997). [CrossRef]
17. E. J. Son, J. S. Lee, M. Lee, C. H. T. Vu, H. Lee, K. Won, and C. B. Park, “Self-adhesive graphene oxide-wrapped TiO2 nanoparticles for UV-activated colorimetric oxygen detection,” Sens. Actuators, B 213, 322–328 (2015). [CrossRef]
18. A. Mills, K. Lawrie, J. Bardin, A. Apedaile, G. A. Skinner, and C. O’Rourke, “An O2 smart plastic film for packaging,” Analyst 137(1), 106–112 (2012). [CrossRef]
19. W. Wang, N. Xie, L. He, and Y. Yin, “Photocatalytic colour switching of redox dyes for ink-free light-printable rewritable paper,” Nat. Commun. 5(1), 5459 (2014). [CrossRef]
20. A. Peter, A. Mihaly-Cozmuta, L. Mihaly-Cozmuta, and C. Nicula, “Nanosensor based on TiO2 for detection of oxygen in damaged vacuum packages,” Carpathian J. Food Sci. and Technol. 5(1-2), 9–12 (2013).
21. W. Wang, L. Liu, J. Feng, and Y. Yin, “Photocatalytic reversible color switching based on Titania nanoparticles,” Small Methods 2(2), 1700273 (2018). [CrossRef]
22. P. Dardano, A. Caliò, V. Di Palma, M. F. Bevilacqua, A. Di Matteo, and L. De Stefano, “A photolithographic approach to polymeric microneedles array fabrication,” Materials 8(12), 8661–8673 (2015). [CrossRef]
23. T. B. C. Terencio, V. Bavastrello, and C. Nicolini, “Calcium oxide matrices and carbon dioxide sensors,” Sensors 12(5), 5896–5905 (2012). [CrossRef]
24. B. Serban, C. Cobianu, and O. Buiu, “Oxygen sensing: A review Part 1: Materials and methods for optical and galvanic lead-free oxygen detection,” Ann. Acad. Rom. Sci. Ser. Math. Appl. 6(2), 1 (2013).
25. A. Mills and K. Lawrie, “Novel photocatalyst-based colourimetric indicator for oxygen: Use of a platinum catalyst for controlling response times,” Sens. Actuators, B 157(2), 600–605 (2011). [CrossRef]
26. S.-K. Lee, M. Sheridan, and A. Mills, “Novel UV-activated colorimetric oxygen indicator,” Chem. Mater. 17(10), 2744–2751 (2005). [CrossRef]
27. K. Won, N. Y. Jang, and J. Jeon, “A natural component-based oxygen indicator with in-pack activation for intelligent food packaging,” J. Agric. Food Chem. 64(51), 9675–9679 (2016). [CrossRef]
28. E. Mohammadian, M. Alizadeh-Sani, and S. M. Jafari, “Smart monitoring of gas/temperature changes within food packaging based on natural colorants,” Compr. Rev. Food Sci. Food Saf. 19(6), 2885–2931 (2020). [CrossRef]
29. J. Kerry and P. Butler, Smart Packaging Technologies for Fast Moving Consumer Goods (John Wiley & Sons, 2008).
30. Y. Salinas, J. V. Ros-Lis, J.-L. Vivancos, R. Martínez-Máñez, M. D. Marcos, S. Aucejo, N. Herranz, I. Lorente, and E. Garcia, “A novel colorimetric sensor array for monitoring fresh pork sausages spoilage,” Food Control 35(1), 166–176 (2014). [CrossRef]
31. Y. Chen, G. Fu, Y. Zilberman, W. Ruan, S. K. Ameri, Y. S. Zhang, E. Miller, and S. R. Sonkusale, “Low cost smartphone diagnostics for food using paper-based colorimetric sensor arrays,” Food Control 82, 227–232 (2017). [CrossRef]
32. R. M. S. da Cruz, Food Packaging: Innovations and Shelf-Life (CRC Press, 2019).
33. V. Orlien and T. Bolumar, “Biochemical and nutritional changes during food processing and storage,” Foods 8(10), 494 (2019). [CrossRef]
34. T. Bolumar, M. L. Andersen, and V. Orlien, “Mechanisms of radical formation in beef and chicken meat during high pressure processing evaluated by electron spin resonance detection and the addition of antioxidants,” Food Chem. 150, 422–428 (2014). [CrossRef]
35. C. E. Reimers, K. M. Fischer, R. Merewether, K. L. Smith Jr, and R. A. Jahnke, “Oxygen microprofiles measured in situ in deep ocean sediments,” Nature 320(6064), 741–744 (1986). [CrossRef]
36. J. Ji, N. Rosenzweig, I. Jones, and Z. Rosenzweig, “Novel fluorescent oxygen indicator for intracellular oxygen measurements,” J. Biomed. Opt. 7(3), 404–409 (2002). [CrossRef]
37. B. P. F. Day, “Modified Atmosphere Packaging (MAP),” Food Biodeterioration and Preservation 165–192 (n.d.).
38. S. Arashisar, O. Hisar, M. Kaya, and T. Yanik, “Effects of modified atmosphere and vacuum packaging on microbiological and chemical properties of rainbow trout (Oncorynchus mykiss) fillets,” Int. J. Food Microbiol. 97(2), 209–214 (2004). [CrossRef]
39. R. Coles, D. McDowell, and M. J. Kirwan, Food Packaging Technology (CRC Press, 2003).
40. E. M. Breitung, E. M. S. van Hamersveld, D. R. Olson, and M. B. Wisnudel, “Limited play data storage media and method for limiting access to data thereon,” USPTO patent 6733950 (May 11, 2004).
41. M. Ushamani, K. Sreekumar, C. S. Kartha, and R. Joseph, “Complex methylene-blue-sensitized polyvinyl chloride: a polymer matrix for hologram recording,” Appl. Opt. 41(10), 1984–1988 (2002). [CrossRef]
42. M. Ushamani, K. Sreekumar, C. S. Kartha, and R. Joseph, “Fabrication and characterization of methylene-blue-doped polyvinyl alcohol-polyacrylic acid blend for holographic recording,” Appl. Opt. 43(18), 3697–3703 (2004). [CrossRef]
43. W. Wang, M. Ye, L. He, and Y. Yin, “Nanocrystalline TiO2-catalyzed photoreversible color switching,” Nano Lett. 14(3), 1681–1686 (2014). [CrossRef]
44. L. X. Hien, D. M. Thanh, and N. H. Tai, “Influence of Darocur 2018 photoinitiator on the photocrosslinking and properties of UV-cured coatings based on an epoxidiacrylate oligomer and hexanediol diacrylate monomer,” Vietnam J. Chem. 56(6), 761–766 (2018). [CrossRef]
45. Y. Chen, Y. Ye, and Z.-R. Chen, “Vapor-based synthesis of bilayer anti-corrosion polymer coatings with excellent barrier property and superhydrophobicity,” J. Mater. Sci. 54(7), 5907–5917 (2019). [CrossRef]