Article pubs.acs.org/Langmuir
Polymer Vesicle Sensor for Visual and Sensitive Detection of SO2 in Water Tong Huang,† Zhilin Hou,† Qingsong Xu,† Lei Huang,‡ Chuanlong Li,† and Yongfeng Zhou*,† †
School of Chemistry and Chemical Engineering, State Key Laboratory of Metal Matrix Composites, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China ‡ School of Chemical Engineering and Technology, Harbin Institute of Technology, Harbin 150001, PR China S Supporting Information *
ABSTRACT: This study reports the first polymer vesicle sensor for the visual detection of SO2 and its derivatives in water. A strong binding ability between tertiary alkanolamines and SO2 has been used as the driving force for the detection by the graft of tertiary amine alcohol (TAA) groups onto an amphiphilic hyperbranched multiarm polymer, which can selfassemble into vesicles with enriched TAA groups on the surface. The polymer vesicles will undergo proton exchange with cresol red (CR) to produce CR-immobilized vesicles (CR@vesicles). Subsequently, through competitive binding with the TAA groups between CR and SO2 or HSO3−, the CR@vesicles (purple) can quickly change into SO2@vesicles (colorless) with the release of protonated CR (yellow). Such a fast purple to yellow transition in the solution allows the visual detection of SO2 or its derivatives in water by the naked eye. A visual test paper for SO2 gas has also been demonstrated by the adsorption of CR@vesicles onto paper. Meanwhile, the detection limit of CR@vesicles for HSO3− is approximately 25 nM, which is improved by approximately 30 times when compared with that of small molecule-based sensors with a similar structure (0.83 μM). Such an enhanced detection sensitivity should be related to the enrichment of TAA groups as well as the CR in CR@ vesicles. In addition, the CR@vesicle sensors also show selectivity and specificity for the detection of SO2 or HSO3− among anions such as F−, Br−, Cl−, SO42−, NO2−, C2O42−, S2O32−, SCN−, AcO−, SO32−, S2−, and HCO3−.
1. INTRODUCTION In recent years, polymer vesicles have been vastly investigated owing to their great potential in drug delivery, biomimetic materials, and microreactors.1−7 Among them, using polymer vesicles as chemical sensors has aroused great interest because of their unique properties, especially good water solubility, good stability, and facile functionalization.8−12 A series of vesicle-based colorimetric sensors in water have been fabricated by covalently introducing specific functional moieties into the vesicle matrix to detect viruses, toxins, metal ions, DNAs, and proteins.13−18 For example, Jung and co-workers reported the specific colorimetric detection of proteins using bidentate aptamer-conjugated vesicles.15 Park and co-workers developed a multiplex biosensor for pathogen detection based on crosslinked polydiacetylene vesicles.16 Yoon and co-workers reported a new Cu2+ sensor based on azide- and alkynefunctionalized vesicles.17 Despite the great progress in polymer vesicle sensors for the detection of biologically important species, little attention has been paid to the detection of gas. Sulfur dioxide (SO2) is a common air pollutant and a potential health hazard.19−23 The reported chemical sensors for SO2 detection generally work in organic solvents. With growing concern over SO2 for its risk of ecosystem and health, a water-soluble, sensitive, and fast© 2016 American Chemical Society
response colorimetric sensor for SO2 is indispensable. As we know, vesicles are very good carriers for aqueous detection. For example, Yuan and Zhao proposed CO2-sensitive polymer vesicles in water.24,25 Yan reported H2S gasotransmitterresponsive polymer vesicles in water.26 However, to our knowledge, there are few reports on vesicle sensors for the detection of SO2. The difficulty might lie in the lack of an appropriate acceptor and unsatisfactory sensitivity. Recently, the strong binding ability between tertiary alkanolamines and SO2 reported by Heldebrant has provided a new opportunity for SO2 detection.27 Meanwhile, hyperbranched polymer vesicles, defined as branched polymersomes (BPs), have shown intriguing properties such as a large population of terminal functional groups, better stability, and appealing solution behaviors.28 Such intrinsic properties have also made BPs promising candidates for sensing applications. By combining BPs and tertiary alkanolamines, herein, we report a fast, sensitive, and water-soluble vesicle sensor for the visual detection of SO2 derivatives. Received: October 24, 2016 Revised: December 3, 2016 Published: December 6, 2016 340
DOI: 10.1021/acs.langmuir.6b03869 Langmuir 2017, 33, 340−346
Article
Langmuir Scheme 1. Self-Assembly, Surface Functionalization, and SO2-Detection Behavior of the HSP-TAA Vesiclesa
a
In HSP-TAA, the hydrophobic HBPO core is blue, the hydrophilic PEO arms are red, and the TAA groups are green. In the vesicles, the deprotonated CR is purple, and the protonated CR is yellow. B microscope. The aqueous solution was dropped onto a glass slide and then observed directly under the microscope. The scanning electron microscopy (SEM) results were obtained on a JEOL 7401F field-emission scanning electron microscope at an accelerating voltage of 5 kV. The samples were prepared by placing a small drop of the polymer solution onto a silicon wafer and then drying at room temperature for 24 h. The samples were coated with a thin film of gold before being measured. The transmission electron microscopy (TEM) results were obtained on a JEM-2010/INCA instrument (Oxford Instruments) operating at an accelerating voltage of 200 kV. For TEM measurements, a small drop of the polymer solution was placed onto 400 mesh copper grids coated with a parlodion film stabilized with vacuum-evaporated carbon and then was dried at room temperature for 24 h. The UV−vis measurements were recorded under a Shimadzu UV-2600 spectrometer at room temperature, using deionized water as the reference. 1H NMR and 13C NMR spectra were recorded on a Varian Mercury Plus 400 MHz spectrometer using deuterated chloroform or deuterated dimethyl sulfoxide (DMSO-d6) as solvents at 20 °C, and tetramethylsilane was used as the internal reference. The molecular weights of the products were measured using GPC on a HLC-8320GPC system (Tosoh, EcoSEC GPC System) at 40 °C with N,N-dimethylformamide or tetrahydrofuran (THF) as the mobile phase at a flow rate of 0.6 mL/min. The inductively coupled plasma (ICP) emission spectrometer results were obtained on an ICAP6000 (Thermo) system with wavelengths from 847 to 166 nm. DLS measurements were recorded on aqueous solutions at 25 °C using a Malvern Zetasizer Nano S instrument (Malvern Instruments, Ltd.) equipped with a 4 mW He−Ne laser light operating at λ = 633 nm and a scattering angle of 90°. The zeta potentials were determined using a zeta potential analyzer (Malvern Zetasizer 2000). 2.3. Synthesis of HSP-TAA. The objective polymer HSP-TAAs were synthesized through a two-step reaction (Figures S1 and S2). In step 1, HBPO-star-PEOs were reacted with epichlorohydrin to obtain the polymers of HSP-EPs. In the 1H NMR spectrum of HSP-EPs (Figure S3), new peaks at δ (ppm) 2.83 (−CH2−epoxy) and 2.63 (−CH2−epoxy) emerged, indicating that the epoxy groups (EP) were grafted onto HSP successfully. The grafting ratio of EP in HSP-EPs was approximately 21%. In step 2, a nucleophilic ring-opening reaction
The main concept of our polymer vesicle sensor is illustrated in Scheme 1. The vesicles are generated through the selfassembly of an amphiphilic hyperbranched multiarm copolymer HBPO-star-PEO-TAA (HSP-TAA) with a hydrophobic hyperbranched poly(3-ethyl-3-oxetanemethanol) (HBPO) core and many linear poly(ethylene oxide) (PEO) arms end-capped with tertiary amine alcohol (TAA) groups. Then, with the addition of the aqueous cresol red (CR) solution, the TAA groups on the surface of vesicles undergo proton exchange with CR, which produces CR@vesicles in purple, with the deprotonated CR spreading all over the surface.29,30 Then, if the TAA groups in CR@vesicles encounter SO2 or its derivatives, they will react with them to form colorless SO2@vesicles, followed by the release of protonated CR in yellow.31−34 Such a purple−yellow color transformation in solution can be readily used for the visual detection of SO2 derivatives. In addition, the CR@ vesicles are sensitive and show a low detection limit of 25 nM.
2. EXPERIMENTAL SECTION 2.1. Materials. Epichlorohydrin (99%) and 2-methylaminoethanol (99%) were purchased from Aldrich and used without further purification. Sodium hydride (60%, dispersion in mineral oil) was purchased from Adamas and used after washing out the mineral oil. Dimethylformamide and CH2Cl2 (AR grade, Shanghai Chemical Reagent Co.) were refluxed with CaH2 and then distilled before use. All other chemical reagents (such as DMSO) were purchased from Shanghai Chemical Reagent Co. and used as received. HBPO-starPEOs were synthesized and characterized according to our previous study,35 which had a number-average molecular weight (Mn) of 7400 measured using gel permeation chromatography and a polydispersity index (Mw/Mn) of 2.5. The degree of branching of the HBPO core was 39%, and the molar fraction of PEO segments ( f EO) was 73%. The number-average degree of polymerization of the PEO arms was 2.7. 2.2. Instrumentation and Measurements. The morphologies of the vesicles were observed by optical microscopy on a Leica DM4500 341
DOI: 10.1021/acs.langmuir.6b03869 Langmuir 2017, 33, 340−346
Article
Langmuir of the epoxy groups was carried out between the HSP-EPs and 2methylaminoethanol.36 The detailed synthetic procedure is shown in the Supporting Information. 2.4. Preparation of CR@Vesicles. Ten milligrams of HSP-TAAs was dissolved in 5 mL of the THF solvent for 10 min. Then, the solvent was evaporated to dryness. Ten milliliters of distilled water was added with vigorous stirring at room temperature. The appearance of turbidity in the solution indicated the formation of HSP-TAA vesicles. The final polymer concentration was 1 mg/mL, and the TAA group concentration was approximately 0.9 mM. Then, 100 μL of the CR aqueous solution (52 mg/mL) was added into 10 mL of the HSP-TAA vesicle solution. After stirring for 25 min, the TAA groups completed proton exchange with the CR, leading to the deprotonated anionic CR species (purple). The excess CR molecules in solution were subsequently removed using a dialysis method.
hydrophobicity of alkanolamine increased, leading to the increase in hydrophobicity of the vesicles.37−39 Thus, the vesicle would aggregate together driven by the intervesicular hydrophobic interactions, and an aggregation-induced fusion among the vesicles was observed. Figure 1c shows some vesicle intermediates in the fusion process (red circles), which lead to the formation of many big vesicles of approximately 15 μm in diameter (Figure 1d). 3.2. Surface Functionalization of HSP-TAA Vesicles. By proton exchange, the TAA-coated vesicles were capable of immobilizing CR on the surface. For the experiment, the suspension of vesicles was mixed with the aqueous CR solution, followed by vigorous stirring. The basic TAA groups were protonated and would attract anionic CRs around the vesicle surface by electrostatic complexation. There are three lines of evidence to support this. First, the TEM and SEM measurements (Figure 2a,b) indicate that the vesicles after mixing with
3. RESULTS AND DISCUSSION 3.1. Characterization and Self-Assembly of HSP-TAAs. The objective polymers of HSP-TAAs have a number-average molecular weight of 11 000 according to GPC (Figure S4). The percentage grafting of TAA groups in HSP-TAAs is approximately 18% according to the NMR characterization (Figure S5). The obtained HSP-TAA polymers could selfassemble into vesicles approximately 1−5 μm in diameter under the optical microscope through a direct hydration process by adding polymers into water with a constant concentration of 1 mg/mL (Figure 1a). The thin unilamellar
Figure 2. Characterizations of the CR@vesicles. TEM (a) and SEM (b) images. The inset in (a) shows a digital image of the purple CR@ vesicle solution; (c) EDX of CR@vesicles; (d) TEM image of CR@ vesicles without staining treatment after bubbling of SO2 (inset, digital image of the related yellow vesicle solution); (e) and (f) SO2 test paper fabricated by dipping filter papers into the CR@vesicle solution and showing the color change before (e) and after (f) exposure of gaseous SO2.
Figure 1. Characterizations of the HSP-TAA vesicles. (a) Optical micrograph of the vesicles in neutral water; (b) plots of zeta potentials vs pH; (c) and (d) vesicle intermediates in fusion at pH > 12. The vesicles labeled with red circles in (c) and (d) are the same vesicles in and after fusion.
wall structure and the hollow lumen of the vesicles were also proved by the TEM measurements (Figure S6) after the vesicles were negatively stained with a 2% aqueous phosphotungstic acid solution. The nanovesicles according to the TEM measurements were due to the polydisperse nature of the vesicle size. The dynamic light scattering (DLS) measurements also show a broad size distribution for the HSP-TAA vesicles (Figure S7). The zeta potential measurements show that the HSP-TAA vesicles were positively charged below the isoelectric point of pH 9 (Figure 1b), which is attributed to the protonation of TAA groups on the surface of the vesicles. With an increase in the solution pH far above the isoelectric point (pH > 12), the
CR become more rigid; meanwhile, the vesicle walls are fully covered with black nanoparticles (inset in Figure 2a,b) when compared with the HSP-TAA vesicles (Figure S6). These nanoparticles should be attributed to the formation of salts between TAA and CR on the surface of vesicles, which certainly enhances the rigidity of the vesicles. Second, the energydispersive X-ray (EDX) spectroscopy proves the existence of elemental sulfur attributed to CR on the surface of CR@ vesicles (Figure 2c). Third, the CR@vesicle aqueous solution turned purple (inset in Figure 2a), which corresponds to the deprotonated CR. The loading of CR can be facilely regulated 342
DOI: 10.1021/acs.langmuir.6b03869 Langmuir 2017, 33, 340−346
Article
Langmuir
Figure 3. (a) UV−vis absorbance spectral changes in CR@vesicles upon the addition of an increasing amount of HSO3− (0−80 μM); (b) effect of HSO3− concentrations on the absorbance at 573 nm and their relationship fitted with the linear curve (inset); (c) UV−vis spectral of the CR− MDEA complexes with the continuous addition of an increasing amount of HSO3− (0−6 mM); (d) influence of HSO3− content upon the colorimetric response of CR@vesicles (black) and CR−MDEA complexes (red).
3.3. Capturing Ability of HSP-TAA Vesicles for Bisulfite. SO2 can dissolve in water to produce sulfurous acid and subsequently form its derivatives, that is, bisulfite (HSO3−) and sulfite (SO32−). The above-mentioned experiments as shown in Figure 2 have proven the capturing ability of HSP-TAA vesicles for SO2 qualitatively. However, it is not easy to directly detect SO2 quantitatively. Thus, instead of SO2, SO2 derivatives such as bisulfite have been commonly used.40,41 The adsorption behavior of the HSP-TAA vesicles for bisulfite was evaluated using the ICP emission spectrometer. For the experiments, 20 mL of the HSP-TAA vesicle solution (1 mg/ mL) was loaded inside of the dialysis bag (molecular-weight cutoff 0.5 kDa) and then dialyzed against the sodium bisulfite solution (30 μg/mL, 100 mL) (Figure S10). The results of the ICP analysis indicated that the sulfur content inside of the dialysis bag (42.18 μg/mL) was nearly double the sulfur content outside (21.99 μg/mL). As a control, for HSP vesicles, the sulfur contents inside and outside of the dialysis bag were basically the same (26.24 and 25.61 μg/mL, respectively). Evidently, the HSP-TAA vesicles are able to capture bisulfite. The capturing mechanism should be the same as that in Scheme 1, and the SO2@vesicles were generated through the formation of zwitterionic compounds, which was further proved by the TEM (Figure S11) and EDX (Figure S12) measurements. 3.4. Sensing Ability of the CR@Vesicles for HSO3−. Subsequently, the sensing ability of the CR@vesicles for HSO3− was investigated through UV−vis absorption spectra. With the continuous addition of HSO3−, the maximum absorption band around 573 nm attributed to the CR anchored on the vesicles decreased and almost disappeared, whereas a new absorption band around 434 nm attributed to the protonated CR appeared and increased accordingly (Figure 3a). Meanwhile, the color of the vesicle solution changed from purple to bright yellow. Such a result further supports the SO2-detection mechanism as proposed in Scheme 1, in which the anchored CR molecules in
by decreasing the amount of TAA groups on the vesicles through the co-assembly of HSP-TAAs and HSPs (Figure S8). When SO2 gas was bubbled into the CR@vesicle solution, a clear color change from purple to bright yellow was observed (inset in Figure 2d) within 1 min. The TEM image (Figure 2d) shows that the vesicular structure was retained after the adsorption of SO2. The DLS measurements show that there is no clear change in the vesicle size by the addition of SO2 into the CR@vesicles (Figure S7). In other words, the CR@vesicles will change into SO2@vesicles after the adsorption of SO2. So, what is the mechanism? To address it, the interactions of SO2 with HSP-TAAs were investigated using 1H NMR spectroscopy in DMSO-d6. As shown in Figure S9, after bubbling of SO2, the peak corresponding to the OH of TAA groups at 4.3 ppm completely disappeared. Meanwhile, all peaks associated with the CH3 and CH2 groups bonded to the nitrogen and oxygen atoms were found to shift downfield by approximately 0.5 ppm. Evidently, SO2 interacts chemically with TAA groups in HSPTAAs to form a thermally stable zwitterionic compound with a covalently bound −OSO2− as shown in Scheme 1, which has also been reported by Heldebrant and co-workers.27 The black dots on the surface of SO2@vesicles in the TEM image (Figure 2d) should be attributed to the as-formed zwitterionic compounds. Thus, when SO2 interacts with CR@vesicles, it will react with the TAA groups in the vesicles to form SO2@ vesicles; meanwhile, the protonated CRs were released from the vesicles to the solution, which turned the solution from purple to yellow. The mechanism is summarized in Scheme 1. More interestingly, like the commonly used pH paper, the CR@vesicles can also be used to prepare the visual test paper for SO2 gas. As shown in Figure 2e, a purple test paper was obtained by dipping it into the CR@vesicle solution for a moment and then drying. When the test paper remained for 30 s in a sealed vial filled with SO2 gas, it changed from purple to yellow quickly (Figure 2f). 343
DOI: 10.1021/acs.langmuir.6b03869 Langmuir 2017, 33, 340−346
Article
Langmuir
CR@vesicles, we studied the response of CR@vesicles to other anions, such as F−, Br−, Cl−, SO42−, H2PO4−, NO2−, C2O42−, S2O32−, SCN−, AcO−, SO32−, S2−, and HCO3− (0.1 M each). Figure S14 shows the photograph of CR@vesicle solutions incubated with the above-mentioned ions in water. Only the solution containing H2PO4− showed a purple to yellow change, whereas solutions with other anions had negligible color changes. However, the detection limit of CR@vesicles for H2PO4− was determined to be 0.53 μM according to Figure S15, which was much larger than that of HSO3− (25 nM). Furthermore, the CR@vesicles exhibited a faster response to HSO3− than to H2PO4−. The time-dependent UV−vis absorbance spectra of CR@vesicles (Figure S16a) suggested that in the presence of H2PO4− (0.667 mM), the amount of CR anchored on the vesicles (573 nm) decreased less than 50% in 60 s and remained unchanged after 120 s. By contrast, all anchored CR molecules on the surface of CR@vesicles released into the aqueous solution within 30 s in the presence of HSO3− (0.667 mM) (Figure S16b). In addition, the CR@vesicles also showed no response to NO and CO gases and to biological thiols, such as glutathione, cysteine, and glycine, but showed a slow response to H2S gas. These results indicate that the CR@ vesicles have specificity for the detection of SO2 or HSO3− in a faster and more sensitive way among many anions; however, the acidic analytes such as H2PO4− and H2S might cause interference to some degree. In addition, it should be noted that the sensing tests of the CR@vesicles designed in this article were all done in neutral deionized water. The most appropriate working pH range for the CR@vesicles is approximately 7−10. Under acidic conditions, the acid will interfere with the detection. Under the more alkaline condition, the vesicles are not stable in water and will fuse with each other to form precipitates.
purple on the surface of CR@vesicles will be released into the aqueous solution in a protonated state in yellow with the introduction of SO2 or HSO3−. The decrease in the peak at 573 nm showed a linear relationship (R2 = 0.995) with the increased concentration of HSO3− from 0 to 40 μM (Figure 3b), from which a detection limit of 25 nM was calculated based on the signal-to-noise ratio (S/N = 3). It is much lower than the threshold levels of HSO3− in medicines and food (
