Inkjet-printed paper-based volatile organic compound (VOC) sensor strips imaged with polydiacetylenes (PDAs) are developed. A microemulsion ink containing bisurethane-substituted diacetylene (DA) monomers, 4BCMU, was inkjet printed onto paper using a conventional inkjet office printer. UV irradiation of the printed image allowed fabrication of blue-colored poly-4BCMU on the paper and the polymer was found to display colorimetric responses to VOCs. Interestingly, a blue-to-yellow color change was observed when the strip was exposed to chloroform vapor, which was accompanied by the generation of green fluorescence. The principal component analysis plot of the color and fluorescence images of the VOC-exposed polymers allowed a more precise discrimination of VOC vapors.
Owing to their toxicological effects and hazards to human health, a wide variety of technologies for the detection of volatile organic compounds (VOCs) have been developed in recent years.1 Although various VOC sensor systems have been reported, most of them have limitations associated with the complicated fabrication steps1a–c or additional equipment for detection.1d–f Recently, as an efficient sensor platform, paper substrates have gained much attention because of their flexible, inexpensive, lightweight, and disposable nature.2 In addition, diverse sensor materials can be readily transferred to a paper substrate by employing simple patterning methods such as inkjet printing technology, a low-cost and low-waste technique.3 Furthermore, if VOC-responsive materials could be printed on paper substrates using a common office inkjet printer, it would be possible to develop effective litmus-type sensors for the detection of VOCs without any sophisticated processes and/or devices.
Polydiacetylenes (PDAs) have been extensively investigated as colorimetric and fluorometric sensor matrices owing to their stimulus-responsive nature as well as their intriguing structural and optical properties.4 In most cases, PDAs obtained by UV irradiation of self-assembled diacetylenes (DAs) undergo both color (blue-to-red) and fluorescence (non-to-red) transitions in response to various organic solvents.5 However, due to their rigid backbone structure, most of solvatochromic PDAs are insoluble in common organic solvents and do not display colors of wavelength below about 500 nm. On the other hand, bisurethane-substituted DA monomer, 5,7-dodecadiyne-1,12-diol bis[((butoxycarbonyl)methyl)urethane] (4BCMU) (Figure 1A) can form a soluble PDA in certain organic solvents, such as chloroform and tetrahydrofuran (THF), because its relatively hydrophobic and the flexible side chains are able to interact with solvent molecules.6 Moreover, a blue-to-yellow color transition of the polymer solution occurs depending on the solubility of the polymer as determined by the solvent.7 The soluble nature of 4BCMU-derived polymer affords a wider colorimetric response spectrum as compared with common solvatochromic PDAs. Accordingly, combination of a modern inkjet printing technology and the VOC-responsive poly-4BCMU should allow fabrication of litmus-type polymeric sensors.
2. Experimental Section
2.1. Preparation of 4BCMU
The bisurethane-substituted DA monomer, 4BCMU was prepared by using methods described in the literature.8 To a solution of 5,7-dodecadiyn-1,12-diol (1.0 g, 5.1 mmol) in THF (20 mL) was added dibutyltin dilaurate (9.1 mL, 15.3 mmol) and triethylamine (0.7 mL, 5.1 mmol). The mixture was stirred at 0 °C for 10 min under N2 atmosphere before adding butyl isocyanatoacetate (2.3 mL, 15.3 mmol). The resultant mixture was stirred at ambient temperature for 24 h and then, precipitated into excess hexane. The precipitates were collected and concentrated to give 4BCMU (2.3 g, 88%). m.p.: 74.5-74.9 °C; 1H NMR (300 MHz, CDCl3, δ): 5.17 (s, 2H), 4.16 (t, 4H), 4.10 (t, 4H), 3.95 (d, 4H), 2.30 (t, 4H), 1.79-1.55 (m, 12H), 1.45-1.32 (m, 4H), 0.94 (t, 6H).
2.2. Preparation of Ink Emulsions
The 4BCMU ink emulsions were prepared by using a previously reported method.9 Briefly, 4BCMU (0.12 g) was dissolved in 0.69 g of 1,2,4-trimethylbenzene. Subsequently, the organic solution was added dropwise to a surfactant solution containing 1.11 g of 1-propanol and 0.3 g of sodium dodecylsulfate (SDS) in 0.78 g of water. The resulting ink emulsion was composed of 4 wt% of 4BCMU, 23 wt% of 1,2,4-trimethylbenzene, 10 wt% of SDS, 37 wt% of 1-propanol and 26 wt% of water.
2.3. Preparation of Paper-Based Volatile Organic Compound Sensors
After removal of black ink from a conventional inkjet office printer cartridge (HP C9351A), the cartridge was thoroughly washed with ethanol, water, and dried by blowing with N2. The 4BCMU ink emulsion (0.2 mL) was loaded in the cartridge and then printed on unmodified paper by using a computer-controlled thermal inkjet office printer (HP Deskjet D2360). The printed image, after irradiating with 254 nm UV light (1 mW cm−2) for 3 min, was exposed to the organic solvent vapor for 3 s at room temperature. The color and fluorescent images were obtained by using a digital camera (Panasonic DMC-FX68) and a fluorescence microscope (Olympus BX51 W/DP70), respectively. The red and green fluorescence intensity distributions across the center of the VOC-exposed images were plotted by OriginPro 8.5 software.
3. Results and Discussion
In order to fabricate a paper-based sensor strip for differentiation of VOCs, the DA monomer, 4BCMU, shown in Figure 1A was applied to an inkjet-printable microemulsion system.9 Preparation of a stable microemulsion was carried out by adding 4BCMU dissolved in 1,2,4-trimethylbenzene to an aqueous solution containing 1-propanol and SDS (See Section 2). The organic-soluble nature of 4BCMU allows generation of highly concentrated organic droplets in an aqueous solution with assistance of surfactants and co-surfactants. The 4BCMU microemulsion was printed on unmodified paper using a conventional computer- controlled thermal inkjet printer (HP Deskjet D2360). Since monomeric 4BCMU supramolecules do not absorb visible light, the printed image is not visible after printing. However, UV irradiation (254 nm, 1 mW cm−2, 3 min) of the image can induce the polymerization of 4BCMU monomers on the paper, which results in the generation of the blue-colored poly-4BCMU dot image (Figure 1B, top). A strip of paper printed and photopolymerized with 4BCMU along with the structure of poly-4BCMU are displayed in Figure 1B (bottom). In addition, the inkjet printing technique enables generation of not only a simple dot image but also a diverse range of images designed by end-users. For instance, a molecular structure of 4BCMU and a university logo can be readily imaged on a paper substrate (Figure 1C).
The poly-4BCMU is known to be soluble in particular organic solvents, such as chloroform and THF.6, 7 Since the color of the polymer is related to its conjugated backbone structure,10 the degree of solubility is an important factor in chromatic transition of the poly-4BCMU. The solvatochromic properties of the poly-4BCMU powder were observed in common organic solvents (Figure 2A). The blue-colored poly-4BCMU powder is stable in hexane, a nonsolvent of the polymer, and maintains its original blue color. In contrast, owing to the partially dissolved monomeric and/or oligomeric DAs that can cause distortion of arrayed π-orbitals and shorten the effective conjugation length of the polymers, red-colored aggregates are formed in methanol, ethanol, acetone, and toluene as well as in ethyl acetate (EA) and ether. Moreover, the poly-4BCMU powder is highly soluble in THF, methylene chloride (MC) and chloroform with generation of a yellow color due to the severely disordered polymer backbone structure. The solvatochromism of the paper sensor strip printed with poly-4BCMU after dipping in the corresponding solvents shows similar results to those obtained with the polymer powder (Figure 2B). This observation supports that the colorimetric responses of the paper sensor strip upon exposure to the organic solvents result from different solubility of the poly-4BCMU. Upon removal of the solvent, while the blue-to-red color transition is irreversible, the yellow dot images, that had reacted with THF, MC, and chloroform, reverted back to the red-colored state after drying (Figure S1, Supporting Information). In these cases, because the printed poly-4BCMU was significantly dissolved in THF, MC, and chloroform, only small amount of polymers remained on the dried paper. Therefore, the color of the obtained images turned out to be a pale-red in comparison with the other red dot images shown in Figure S1 (Supporting Information).
In the next phase of the investigation, we examined the colorimetric response of the poly-4BCMU sensor strip in the presence of an organic solvent vapor. The poly-4BCMU immobilized strip was exposed to each solvent vapor for 3 s at room temperature. As can be seen in Figure 2C, hexane does not induce any observable color change to the polymeric sensor. Though the blue-colored image was changed to purple upon exposure to methanol, and a blue-to-red color transition was observed when the paper sensor strip was exposed to ethanol, acetone, toluene, EA, or ether. However, unlike the results observed from the incubated sensor strips, only red or dark orange-colored image was obtained upon exposure to vaporized THF or MC, respectively. Interestingly, among the tested solvent vapors, only chloroform vapor was found to promote a color transition from blue to red to yellow. This observation may come from the exceptional solubility of the poly-4BCMU in chloroform. Removal of the poly-4BCMU immobilized strip from the chloroform vapor resulted in the instantaneous color change from yellow to red and the yellow-to-red colorimetric transition was observed to be reversible without losing the color intensity. The movie clip provided in the Supporting Information demonstrates the irreversible blue-to-red and reversible red-to-yellow color transitions upon exposure to the chloroform vapor.
The solvatochromic transitions of the poly-4BCMU upon exposure to VOCs were also monitored by absorption spectroscopy. In Figure 3A are shown UV–vis spectra of the poly-4BCMU sensor strips immersed in the organic solvents. The absorption peak at 625 nm for the blue-colored PDA is seen to shift to lower wavelengths of approximately 540 nm for the red-phase PDA upon exposure to methanol, ethanol, acetone, toluene, EA, and ether. Even though the sensor strip in THF solution displays yellow color to the naked eye, the broad absorption peak indicates that the unchanged red-phase PDAs are still remained. The absorption maxima of the PDAs after interaction with MC and chloroform appear at around 470 nm. However, due to the dissolution of PDAs from the sensor strip in the good solvent, as mentioned above, the peak intensity of the image immersed in chloroform is relatively low compared with that of MC. One of the attractive advantages of the vapor sensor system over the solution-based process is that no loss of the sensory polymer is monitored when an inkjet-printed PDA sensor is in contact with the VOC vapors. As expected from the observations in Figure 2C, chloroform vapor affords the most significant spectral shift of the poly-4BCMU, having an absorption maximum at about 470 nm (Figure 3B).
The fluorescent images of the PDAs after exposure to the vapor of the solvents were further recorded with a fluorescence microscope (excitation at 546 or 488 nm) (Figure 4). It is well known that the blue-state PDAs are non-fluorescent while the red fluorescence is generated when a blue-to-red colorimetric transition occurs with the PDAs.4 Accordingly, the blue-colored PDA strip exposed to hexane shows no fluorescence when excited at both wavelengths. The red-phase poly-4BCMU, resulting from interaction with vaporized methanol, ethanol, acetone, toluene, EA, ether, and THF, displays a red fluorescence upon excitation at 546 nm. Interestingly, the dark orange and yellow-colored images, produced by exposure to MC and chloroform, respectively, emit a green fluorescence with simultaneous red fluorescence quenching. After removal of the MC and chloroform vapors, the dark orange and yellow-colored images were reverted back to the red-colored state and disappearance of the green fluorescence was observed. These solvatochromic color/fluorescence transitions are fully reversible and can be repeated more than 300 times. The red and green fluorescence intensity distributions across the center of the VOC-exposed images are shown in Figure S2 (Supporting Information).
In order to extract meaningful information from color and fluorescence data, we employed a principal component analysis (PCA) plot11 for the differentiation of VOCs (Figure 5 and Figure S3, Supporting Information). Values for color (red, green, and blue) and fluorescence (red and green) of the poly-4BCMU obtained after VOC exposure were employed in the PCA method using an open-source statistics program, R software v220.127.116.11 Each value is extracted from nine grid points in the corresponding image by utilizing Adobe Photoshop, and then averaged to obtain the mean values (Figure S4, Supporting Information). As seen in Figure 5, it is noticeable that chloroform can be clearly distinguished from the other VOCs. In addition, MC and THF can be sufficiently distinguished from the PCA plot.
In summary, this investigation has led to the development of a colorimetric and fluorometric sensor system for the detection of VOCs based on inkjet-printed PDAs on paper. The dot image on the paper-based sensor strip printed using 4BCMU microemulsion ink yields blue-colored PDAs by UV-induced photopolymerization. The solubility differences of poly-4BCMU in common organic solvents result in distinguishable color and fluorescence changes upon exposure to VOC vapors. Unlike other vaporized solvents, the chloroform vapor-exposed image on the paper sensor strip undergoes a blue-to-red-to-yellow color transition accompanied by the generation of green fluorescence. In addition, the PCA plot of the color and fluorescence data enables the classification of THF, MC, and chloroform without confusion.
Supporting Information is available from the Wiley Online Library or from the author.
The authors gratefully thank National Research Foundation of Korea (NRF) for financial support through Basic Science Research Program (20120006251 and 2012R1A6A1029029), Nano Material Technology Development Program (2012035286) and Center for Next Generation Dye-Sensitized Solar Cells (20120000593).