Article pubs.acs.org/ac
Cite This: Anal. Chem. 2019, 91, 8926−8932
Organosiloxane and Polyhedral Oligomeric Silsesquioxanes Compounds as Chemiluminescent Molecular Probes for Direct Monitoring Hydroxyl Radicals Mingxia Sun,† Yingying Su,‡ Wenxi Yang,§ Lichun Zhang,§ Jianyu Hu,† and Yi Lv*,‡,§ †
College of Architecture & Environment, ‡Analytical & Testing Center, and §Key Laboratory of Green Chemistry & Technology, Ministry of Education, College of Chemistry, Sichuan University, Chengdu, Sichuan 610064, China
Downloaded via UNIV OF SOUTHERN INDIANA on July 22, 2019 at 09:34:41 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
S Supporting Information *
ABSTRACT: Despite the tremendous progress in the research of luminescent probes for reactive oxygen species (ROS), designing luminescent ROS probes with high sensitivity for the individual ROS is still retarded because of their high reactivity and the rapid and complex interconversion reactions among them. Herein, organosiloxane and polyhedral oligomeric silsesquioxanes (POSS) compounds are designed as a novel class of luminescent molecular probes to produce extraordinary chemiluminescence (CL) based on the specific electrophilic attack of •OH. No CL signal can be obtained by the other ROS and strong oxidants. AEAP-POSS formed by hydrolytical condensation of 3-(2-aminoethylamino)propyl]trimethoxysilane (AEAPTMS) is constructed to covalently link a dye molecular, perylene diimide derivative (PDI), and an intramolecular chemiluminescence resonance energy transfer (CRET) system is obtained to realize the red shift of CL wavelength and enhanced CL intensity. This probe based on CRET is applied to monitor inherent •OH in ambient particular matter (PM2.5 and PM10). Density functional theory (DFT), ion chromatograph, X-ray photoelectron spectroscopy (XPS), particle size analysis, and fluorescence spectrum (FL) are applied to study the CL mechanism. These studies discover that electronically carbonyl CH3CO• is the CL emitter, and the silicon−oxygen skeleton in the organosiloxane and POSS compounds plays the key role in undergoing chemiluminescence (CL) reaction.
R
the molecular CL probe for individual ROS with extensive application has been taking a lot of attention. The most familiar chemiluminescent molecular probes are luminol16,17 and TCPO18−20 for H2O2; lucigenin21 and CLA22−24 toward O2−•; and anthracene25,26 and 1,2-dioxetanes27 for 1O2. As for • OH, the most reactive oxygen radical, indoxyl-β-glucuronide (IBG)28 and phthalhydrazide29 have been employed. Unfortunately, they are not further studied due to the nonspecific identification or indirect detection of •OH. To overcome the limitation, strategies involving nanomaterials have been performed. They are based on the radiative recombination of • OH-injected holes and electron in nanomaterials.30,31 Although the studies open a new door to design luminescent probe for •OH selective detection, it is essential to develop a class of chemiluminescent molecular probes with high biocompatibility for monitoring individual •OH directly. Because of outstanding biocompatibility, specific physical properties, and tunable organic groups at the silicon atoms,
eactive oxygen species (ROS) are products of molecular oxygen by normal and abnormal redox functions, that is, hydrogen peroxide (H2O2), singlet oxygen (1O2), hypochlorite ion (ClO−), peroxynitrite (ONOO−), hydroxyl radical (•OH), hydroperoxyl radical (HOO•), and the superoxide radical (O2−•).1−5 Despite the important role of ROS in chemical and biological processes, the nature of the ROS remains scarcely understood to date because of their complex interconversion reactions among all of the ROS. Therefore, identification, quantification, and kinetics evaluation of the individual species of ROS in complex samples has been a research focus in the environmental, biological, and medical fields.6−9 Luminescent probes with spatiotemporal resolutions and high sensitivity are ideal for directly monitoring of ROS.10 In particular, the development of fluorescent probes enables selective detection of different ROS in biological samples.11−13 Chemiluminescent probes that do not require photoexcitation can avoid the drawback for fluorescent probes such as photobleaching and background light interference as the promising alternative luminescent probes. Recent works about the exploration of nanoparticles in detecting individual ROS exhibit great potential in real-time monitoring ROS of living specimens.14,15 Like the classical luminol−H2O2 system, © 2019 American Chemical Society
Received: February 3, 2019 Accepted: June 20, 2019 Published: June 20, 2019 8926
DOI: 10.1021/acs.analchem.9b00637 Anal. Chem. 2019, 91, 8926−8932
Article
Analytical Chemistry
Chemiluminescent Spectra. The static injection analysis system for determination of •OH was as follows: Typically, 300 μL of the organosiloxane and POSS compounds at near neutral solutions was added into the quartz tube and mixed with 300 μL of 10−3 M FeSO4 before entering the detector. Subsequently, 300 μL of different concentrations H2O2 of solution was injected into the reaction quartz tube by flow pump (BT100-02) with a speed-controller and reacted with FeSO4 to produce •OH.30 The whole reaction processes were recorded via the Ultra Weak Luminescence Analyzer (BPCL-2TGC) with 880 V negative voltage. The CL peak height of the signal recorded was measured relative to CL intensity. As for the PM detection, 300 μL of the AEAP-POSS-PDI was added into the quartz tube, and 600 μL of the PM suspension was injected into the reaction quartz tube by flow pump. The CL peak height of the signal recorded was measured relative to CL intensity.
organosiloxane and polyhedral oligomeric silsesquioxane (POSS) compounds have attracted a great deal of attention and have exhibited considerable potential for selective ion identification,32 catalysis,33 and luminescence.34−36 On the other hand, •OH tends to attack sites with high negative charge density,37,38 while organosiloxane compounds composed of inorganic Si−O fragment and potential reactive R−Si fragment (especially R were aliphatic groups) have high negative charge density,39 and POSS generated from the hydrolytic condensation of organosiloxane compounds exhibits diffusive negative contours at the exterior to enable the attack of •OH.40−42 In view of the above-mentioned properties of • OH, organosiloxane, and POSS, we anticipated organosiloxane and POSS compounds will be an ideal candidate to be designed as a class of chemiluminescent molecular probes for monitoring individual •OH directly. Herein, we first report the particular CL generated from reactions of organosiloxane and POSS compounds appended with aliphatic groups with •OH as shown in Scheme 1.
■
RESULTS AND DISCUSSION CL Spectral Responses of Organosiloxane and POSS Compounds for •OH. Thirteen kinds of organosiloxane compounds and two small molecules were used to analyze the CL signals, and the results were shown in Figure 1a. Ultraweak CL signals were observed when n-propylamine reacted with • OH. However, compounds 3 and 11 that both contained silicon−oxygen skeleton and aminopropyl group generated strong CL signals. In particular, the CL intensity of compound 4 was obviously decreased in which the number of Si−O bonds was reduced; meanwhile, the CL intensity significantly increased 2 times by using compound 13 that had bridged two trimethoxysilane parts at both ends. Furthermore, all trimethoxysilanes containing different R1 (compounds 6−12) enable producing strong CL signals, and the CL intensities of them were −(CH2)2CH3 > −(CH2)3NH2 > −(CH2)3SH > −(CH2)3Cl > −(CH2)2CF3. The intensities of CL decreased with the increase of the electronegativity of heteroatoms and numbers. For organosiloxane compounds containing amino groups (compounds 7, 9, and 11), the CL intensity was NH > 2NH > 3NH. Interestingly, no CL signal can be obtained for compounds 1, 2, and 5 in Figure 1 when the R1 of them were the −CH2CH3, −CH3, and −CH2CH2CH2OOC(CH3)C CH2, respectively. Therefore, silicon−oxygen skeleton and aliphatic groups in organosiloxane compounds played an important role in undergoing the CL reaction. To further study the CL reaction, density functional theory (DFT) calculations were conducted on the atomic polar tensors (APT) charge and MOs for organosiloxane compounds with the B3LYP/6-311++G* method shown in Table S1. As we know, among all ROS, •OH shows the most positive redox potential (2.72 V), and it was easier to lose electrons to the •OH with the higher HOMO level of the substance. According to the MOs analysis, the introduction of halogen atoms reduced the energy of HOMO orbit, which prevented them from being oxidized by •OH, and therefore the CL intensities correspondingly decreased. Because of the more negative charge density of the C atom on the organosiloxane compounds, the more favorable electrophilic •OH would attack it to make the dehydrogenation reaction take place. From the results, we could see that the introduction of a silicon−oxygen skeleton significantly enhanced the negative charge density of organosiloxane compounds as compared to the n-propylamine, and the charge density of C in position-1 (close to silicon) on the organosiloxane compounds was the
Scheme 1. •OH-Induced Chemiluminescence of Organosiloxane and POSS Compounds and the POSS as the Platform To Develop the Intramolecular CRET System
Ammonium-POSS can be able to covalently link dye molecule, perylene diimide derivative (PDI), and so an intramolecular chemiluminescence resonance energy transfer (CRET) system is constructed. This luminescent probe based on CRET is selective and sensitive enough to directly monitor inherent • OH in ambient particular matter (PM2.5 and PM10). The series of CL were shown in Scheme 1. The proposed strategy will open a new avenue to design new chemiluminescent molecular probes for various applications.
■
EXPERIMENTAL SECTION Chemicals. (3-Aminopropyl)trimethoxysilane (APTMS, 97%), methyltrimethoxysilane (97%), trimethoxy(propyl)silane (98%), 3-chloropropyltrimethoxysilane (98%), (3mercapto-propyl)trimethoxysilane (96%), 3-[2-(2-aminoethylamino)-ethylamino]-propyl-trimethoxysilane (95%), [3-(2amino-ethylamino)propyl]trimethoxysilane (AEAPTMS, 97%), tetraethoxysilane (99%), 3-(diethoxy(methyl)silyl)propan-1-amine (98%), bis[3-(trimethoxysilyl) propyl]amine (BTMSPA, 90%), trimethoxy(3,3,3-trifluoropropyl)silane, and aminopropyl-triethoxysilane were obtained from Sigma-Aldrich. Potassium superoxide was purchased from Aladdin (Milwaukee, WI). Trifluoromethanesulfonic acid (CF3SO3H, 99%) and 1,6,7,12-tetrachloroperylene tetracarboxylic acid dianhydride (PDI, 98%) were bought from Adamas. Other reagents were bought from Chengdu United Institute of Chemical & Reagent. Ultrapure water (18.2 MΩ/cm) was obtained from a Milli-Q ultrapure system. All chemicals were at least of analytical grade. 8927
DOI: 10.1021/acs.analchem.9b00637 Anal. Chem. 2019, 91, 8926−8932
Article
Analytical Chemistry
produced by •OH. Furthermore, we systematically studied the CL behaviors of compound 11 (APTMS) reacted with physically relevant oxidants and some strong oxidants in aqueous solution. Only •OH could generate a strong CL signal as shown in Figure S1. During our study of the advanced oxidation of APTMS, the CL was found to depend on the pH of the buffer shown in Figure S1c. No CL was observed at pH ≥ 8, but the CL increased progressively with pH decreasing in the range of 2−7, which corresponded to the oxidation capacity of •OH (E0 = (2.84−0.06 pH) V). Compounds A (APTMS-POSS) and B (AEPA-POSS) were synthesized from compounds 11 and 9 (AEAPTMS). To confirm the POSS structure, 29Si NMR, XRD, and ESI MS measurements were carried out. Because POSS had a rigid cage-like structure, the XRD measurements of the POSS showed many sharp diffraction peaks in Figure S2, which corresponded to the reported structures. Yet from the results of XRD, the APTMS-POSS indicated a higher-ordered regular structure than AEAP-POSS. This might be due to the difference in crystallinities between APTMS-POSS and AEAP-POSS. As known to us, the cagelike octamer (T8POSS) was often a crystalline compound, while the cagelike decamer (T10-POSS) and the cagelike dodecamer (T12-POSS) were amorphous in ammonium-functionalized POSS (structures shown in Scheme S1). The APTMS-POSS synthesized was almost T8-POSS, and AEAP-POSS was a mixture of T8POSS and T10-POSS, which was evidenced by the 29Si NMR in Figures S16 and S17. The 29Si NMR spectrum of APTMSPOSS in DMSO-d6 at 40 °C showed only a signal at ca. 67 ppm in T3 signal assigned to T8-POSS. As compared to it, the 29 Si NMR spectrum of AEAP-POSS showed two signals in the T3 region at 66.6 ppm (a main signal) and at 68.6 ppm (a minor signal) that belonged to T8-POSS and T10-POSS, respectively. Also, the molar ratio of T8 to T10 was calculated to be 0.9:0.1. Furthermore, the formation of APTMS-POSS and AEAP-POSS was also evidenced by the ESI-MS. The peaks corresponding to the mass of APTMS-POSS and AEAPPOSS were both observed. To confirm the structures of the substituent groups in the two POSS, IR and 1H NMR measurements were performed to successfully prove the formation of ammonium-functionalized POSS in Figures S2 and S16. To discuss the effects of structure on CL response, we synthesized the T8+T10+T12-APTMS-POSS using 1-hexanol as the solvent instead of water and the amorphous net-like APTMS using hydrochloric acid as a catalyst. The CL intensities of different structures from hydrolytic condensation of APTMS were T8-APTMS-POSS > T10+T12-APTMS-POSS > net-like APTMS. Moreover, if AEAP-POSS was combined with compound 13 through a chemical bond to form watersoluble POSS polymer (APTMS-POSS polymer was not watersoluble), the CL intensity was greatly reduced (Figure 1c). These results indicated that the T8-POSS would greatly promote CL generation. The CL response of T8-APTMSPOSS provided a CL intensity approximately 10 times higher than that by APTMS in Figure 1d. As for the solvents used in the process of synthesis POSS, the CL signal of using water as solvent was stronger than that of an organic solvent like 1hexanol. It was because that residual of organic solvents could react with the •OH and then restrained the CL response. After that, the CL behaviors of APTMS-POSS and AEAP-POSS reacted with physically relevant oxidants, and other oxidants were studied (Figures 1e and S3). Indeed, only •OH could
Figure 1. (a) CL intensities of 13 kinds of organosiloxane compounds and two small molecules induced by •OH. (b) The CL wavelength of trimethoxysilanes with different R1 with Fenton reagent and Fentonlike reagent. (c) The CL responses of different structures from hydrolytic condensation of organosiloxane compounds. (d) The CL intensities of APTMS and APTMS-POSS. The inset was the CL wavelength of them. (e) The CL responses of different ROS and some oxidants with APTMS-POSS. The inset was the EPR spectra of APTMS-POSS with Fenton reaction using DMPO as the trapping agent in aqueous solution.
most negative among all C atoms, which meant that the oxidation reaction preferred to start at this site. To further study the CL properties of the system, we measured the CL wavelength of some organosiloxane compounds in Figure 1b. Compounds 10, 11, and 12 all showed the same CL wavelength, and this indicated that these different organosiloxane compounds had the same CL luminophore; the CL emission spectrum was characterized by a broad band (450−600 nm), which was centered at 490 nm measured by a filter in the BPCL system (the accurate CL wavelength was 505 nm measured on the fluorescence instrument with xenon lamp turned off as shown in Figure 1d). After that, we found that CL could be produced not only with the classic Fe(II)-mediated Fenton systems, but also with other redox-active metal-mediated Fenton-like systems including Cr3+−H2O2 and Ti3+−H2O2, catalytic system Fe3O4− H2O2, and the well-known ozonation systems, including O3 and O3/H2O2 shown in Figure S1. •OH was produced by all of these systems. All of them confirmed that the CL was indeed 8928
DOI: 10.1021/acs.analchem.9b00637 Anal. Chem. 2019, 91, 8926−8932
Article
Analytical Chemistry
have added the AEAP-POSS/•OH and n-propyltrimethoxysilane (compound 12)/•OH systems to study. The solidphase microextraction method in GC−MS was used to analyze the products. As shown in Figure S6, GC−MS analysis permitted identification of the following products in the AEAPPOSS/•OH system, ethenyl ethanoate, and in the npropyltrimethoxysilane/·OH system, methoxyethene and 1,4pentadiene. Furthermore, the products of n-propylamine/•OH and n-propanol/•OH systems (without siloxane structure) were studied. As shown in Figure S7, the major product in both of them was identified as propionaldehyde. Obviously, as compared to n-propylamine and n-propanol, organosiloxane preferred to produce vinyl alcohol compounds. According to the work of Hu,43 the calculated transition energy of CH3CO• and wavelength from the first excited state to the ground state were 2.45 eV and 506 nm, which was able to further form these vinyl alcohol compounds. So we preliminarily proposed that electronically excited CH3CO• is the CL emitter. In ion chromatograph analysis (shown in Figure S9), the concentration of formic acid was AEAP-POSS/•OH system > npropyltrimethoxysilane/•OH system > n-propylamine/•OH system, which was consistent with CL intensity. CO2 can also be generated by all of them. Therefore, as compared to npropylamine, organosiloxane preferred to be completely oxidized to generate a stronger CL signal, which indicated that the introduction of the siloxane part could contribute to the dehydrogenation reaction. Except for these small molecules and ion, it was interesting to find out that the products had fluorescence properties and presented ex-dependent FL behaviors in Figure 2b. Size analysis was carried out, and two peaks located in the range of 0.8−1.7 and 15.6−43.8 nm were observed from the APTMSPOSS/•OH system. Moreover, the nanoparticles produced in the APTMS-POSS/•OH system were directly observed by TEM in Figure S5. In the XPS spectrum, the nanoparticles appeared as a new peak at 288.6 eV attributed to the CO bond in C 1s spectra, and in the spectrum of N 1s, a new peak at 400.8 eV assigned to the NO2 bond (oxidation product of amino group) was also found (Figure S10). All of these revealed that nanoparticles were the carbonyl products during CL process assigned to the carbonyls. 43,44 Was this luminescent nanoparticle generation in situ by •OH responsible partly for CL emission? All the products of eight kinds of organosiloxane compounds in Table S2 were found have the FL signals except that of n-propylamine, which meant that containing a silicon−oxygen skeleton also contributed to form luminescent nanoparticles. As compared to the CL, which had the same emission wavelength centered at 505 nm, the maximum emission wavelengths of FL among eight organosiloxane compounds were different and changed from 345 to 435 nm. With the enhancement of CL for the eight organosiloxane, the fluorescence intensity did not change regularly. Especially, the AEAP-POSS/•OH system also had almost the same FL properties as the AEAPTMS/•OH system shown in Figure S4, indicating that the luminescent nanoparticles might be produced as the products by the condensation of large quantities of siloxane through forming a Si−O−Si bond and further agglomeration by positive and negative charge attraction and hydrogen bonding. On the basis of the above results, a molecular mechanism for • OH-dependent CL production by radical was proposed (Figure S11): •OH first attacked organosiloxane or POSS compounds in position-1 via dehydrogenation reaction to form
generate a strong CL signal, while other ROS and oxidants did not. Furthermore, the pH had the same influence on both APTMS/•OH and APTMS-POSS/•OH CL system in the range from 4 to 7. Yet as shown in Figure S3, the CL intensity reduced when pH was below 4 (AEAP-POSS was below 3), which was supposed to be the partial hydrolyzation of POSS that happened at strong acidity condition. Next, the CL wavelength of APTMS-POSS was detected and found to be in the range of 450−600 nm and centered at 505 nm, which was the same as that of APTMS (Figure 1d). As can be seen from the results of DFT (Figure 2a), as compared to APTMS and
Figure 2. (a) MOs analysis of APTMS-POSS and AEAP-POSS determined by density functional theory calculations. Parts (b) and (c) were the FL emissions of APTMS-POSS/•OH and AEAPPOSS/•OH. The inset of (b) on the left was the APTMS-POSS and on the right was the APTMS-POSS/•OH under the UV lamp. The inset of (c) on the left was the AEAP-POSS and AEAP-POSS/•OH.
AEAPTMS, the HOMO level of APTMS-POSS and AEAPPOSS, respectively, increased a lot, making it more easily lose an electron. This was further proved by that the AEAP-POSS could quickly turn potassium permanganate into the color of wine red attributed to the generation of low valence manganese, but AEAPTMS could not (shown in Figure S3). Meanwhile, potassium permanganate still did not produce any CL signal to reveal the specific reaction induced by •OH. The CL QY values of these POSS were determined by using the luminol−H2O2 system as a reference on a CL analyzer (seen in the Supporting Information). CL QY values of APTMS-POSS and AEAP-POSS that were calculated on the basis of the molar concentration were 1.08 × 10−4 and 8.9 × 10−5 E/mol, respectively. Mechanism of CL Response. To further understand the intrinsic CL of organosiloxanes and POSS compounds induced by •OH, electron paramagnetic resonance (EPR) spectrum (shown in Figures 1e and S3) was applied to verify that only • OH was reduced after Fenton reagent reacted with APTMSPOSS. Specifically, the major products were identified as formic acid and nitrate in the APTMS-POSS/•OH system by ion chromatography shown in Figure S4 and CO2 in Figure S5 (about 43% APTMS-POSS can be mineralized to inorganic carbon after 30 min). To further explore the mechanism, we 8929
DOI: 10.1021/acs.analchem.9b00637 Anal. Chem. 2019, 91, 8926−8932
Article
Analytical Chemistry
performed to further characterize the structure of AEAPPOSS-PDI (shown in Figure S18). In addition, the photoluminescence of AEAP-POSS-PDI was also studied in the FL spectrum (shown in Figure S13). As for the PDI, the fluorescence emission peak in THF solution was about 570 nm. However, in the aqueous solution of AEAP-POSS-PDI, its emission peak shifted to 580 nm, which may be caused by the polarity of the solvent and the introduction of POSS to inhibit the self-aggregation of PDI. The reproducibility of the probe was examined by 12 successive measurements of 1 × 10−3 M FeSO4 and 1 × 10−3 M H2O2, and the relative standard deviation (RSD) of the CL response was only 2.0%, demonstrating a good reproducibility. The stability of the probe was investigated. After 10 days of storage at room temperature, the CL response only decreased 6.7% from its initial response intensity. The CL responses of CRET systems toward •OH were measured. Significantly, the CL intensity of AEAP-POSS-PDI dramatically increased as compared to that of AEAP-POSS, while the CL intensity of APTMS-POSS-PDI decreased a lot as compared to that of APTMS-POSS (shown in Figures 3a and S13). Considering that APTMS-POSS could not be oxidized by KMnO4 (Figure S3), the influence of PDI chemically anchored on the POSSs was investigated on AEAP-POSS. That PDI might make POSSs harder to oxidize was proved by the phenomenon that AEAP-POSS-PDI was oxidized hardly by KMnO4 (Figure S13). Theoretically, as compared to APTMS-POSS, APTMS-POSS-PDI would also become more difficult to be attacked by •OH and a CRET process produced hardly, and thus the CL intensity of APTMS-POSS-PDI decreased. However, the AEAP-POSSPDI could still be oxidized by •OH because of the higher HOMO level of AEAP-POSS than APTMS-POSS (Figure 2) and CRET produced; the CL intensity of AEAP-POSS-PDI significantly increased as compared to AEAP-POSS. So AEAPPOSS was a suitable platform to construct CRET system. Also, the CRET ratio was as high as near 96% measured by the area ratio.49 After that, we studied the CL behaviors of AEAPPOSS-PDI reacted with other ROS (Figure S13b) and found that the system still maintained high selectivity to •OH. To verify the CRET system, we measured the CL wavelength of AEAP-POSS-PDI in luminescence pattern on an FL instrument. As shown in Figure 3b, its CL wavelength red-shifted to 580 nm as compared to that of AEAP-POSS centered at 505 nm and was consistent with the FL emission of aqueous AEAPPOSS-PDI to prove that the CL energy transfer was achieved. Furthermore, we applied the AEAP-POSS-PDI for •OH detection under the condition of pH 7.5. As shown in Figure 3c and d, a linear increase in CL intensities was observed with increasing concentrations of H2O2 in the Fenton system, and the detection limit was 71 nM (S/N = 3). To clarify the direct monitoring of •OH, ambient particular matter (PM2.5 and PM10) samples were determined. The •OH of PM2.5 and PM10 suspension was affirmed by EPR spectrum (shown in Figure S14).50,51 CL curves of AEAP-POSS-PDI with •OH from PM2.5 and PM10 suspensions were obtained by static injection CL analysis. From the results of CL in Figure S15, after mixing, the CL intensities decayed slowly within 100 s, indicating the persistent generation of •OH from PM2.5 and PM10.50 The amount of •OH generated from 1.6 mg mL−1 of PM2.5 suspension was estimated to be 9.5 × 10−7 M and from 1.0 mg mL−1 of PM10 suspension was 1.3 × 10−6 M. There were some reports about the influence of sunlight on enlarging •OH
carbon radical. After that, siloxane-1,2-dioxetanes intermediate was generated by the electrophilic addition and dehydrogenation reaction, and then further decomposed to form the electronically carbonyl CH3CO•. The CL centered at 505 nm was proposed to be emitted from the electronically excited state to its ground state. Finally, the siloxane species spontaneously condensed with itself and residual organosiloxane compounds by hydrolytic condensation and then further agglomerated through positive and negative charge attraction and hydrogen bonding to produce the luminous nanoparticles. However, more studies are needed to better understand the nature of the observed CL emission. Construction of CRET System and •OH Detection in Ambient Particular Matter. CL energy transformations with higher efficiency in general happened when the energy transfer was intramolecular rather than intermolecular.45−47 Ammonium-functionalized POSS could be altered or tailored to special applications through changing the organic functionalities. Therefore, the intramolecular chemiluminescence resonance energy transfer system based on ammonium-POSS was constructed. Perylene diimide derivatives (PDIs) as model dye were chosen to be a suitable light-emitting receptor due to outstanding chemical and photochemical stability.48 Figure 3a
Figure 3. (a) Schematic illustration of synthesis POSSs-PDI. (b) CL wavelength of AEAP-POSS and AEAP-POSS-PDI with •OH measured in F7000 with xenon lamp turned off. (c) CL signal with different concentrations of •OH. (d) The standard curve for the determination of •OH.
illustrated the synthesized process of water-soluble POSS-PDI. 1 H NMR, FTIR, UV−vis absorption, and FL spectrometers were applied for the characterization of compound C (AEAPPOSS-PDI). As the results of UV−vis absorption spectra show (Figure S12), as compared to the PDI and AEAP-POSS, a new absorption peak at 312 nm appeared, which was due to the n−π* transition of amide in the AEAP-POSS-PDI to ensure PDI dye was attached to AEAP-POSS. Furthermore, this was distinctly affirmed by the FTIR spectrum shown in Figure S12. The IR spectrum of the PDI dye showed two typical absorption peaks at 1789 and 1742 cm−1 derived from the OC−O−CO bond. Yet in the spectra of the AEAPPOSS-PDI, these two peaks completely disappeared, and two new strong peaks at 1698 and 1653 cm−1 belonged to the stretching vibration of OC−N−CO bond appeared, which showed that the dye was actually attached to the AEAP-POSS. Next, 1H NMR, 13C NMR, and ESI−MS were 8930
DOI: 10.1021/acs.analchem.9b00637 Anal. Chem. 2019, 91, 8926−8932
Article
Analytical Chemistry
Sichuan University, are gratefully thanked for technical assistance.
generation in atmospheric particle matter. Finally, we conducted the determination with the xenon lamp irradiation of PM2.5 and PM10. We found that within 20 min, the CL intensities of •OH continuously increased, but it did not change much after 40 min in both PM 2.5 and PM 10 suspensions. Therefore, this probe can be used for monitoring • OH in complex samples. This proved that the probe had the potential for extensive applications in analytical and environmental fields.
■
(1) Bartosz, G. Clin. Chim. Acta 2006, 368, 53−76. (2) Schmidt, R. Photochem. Photobiol. 2006, 82, 1161−1177. (3) Burns, J. M.; Cooper, W. J.; Ferry, J. L.; King, D.; DiMento, B. P.; McNeill, K.; Miller, C. J.; Miller, W. L.; Peake, B. M.; Rusak, S. A.; Rose, A. L.; Wait, T. D. Aquat. Sci. 2012, 74, 683−734. (4) Zhang, L. J.; Zhang, Z. M.; Lu, C.; Lin, J. M. J. Phys. Chem. C 2012, 116, 14711−14716. (5) Su, Y. Y.; Deng, D. Y.; Zhang, L. C.; Song, H. J.; Lv, Y. TrAC, Trends Anal. Chem. 2016, 82, 394−411. (6) Finkel, T.; Holbrook, N. J. Nature 2000, 408, 239−247. (7) Sameenoi, Y.; Koehler, K.; Shapiro, J.; Boonsong Sun, K. Y.; Collett, J., Jr.; Volckens, J.; Henry, C. S. J. Am. Chem. Soc. 2012, 134, 10562−10568. (8) Wang, D.; Zhao, L. X.; Ma, H. Y.; Zhang, H.; Guo, L. H. Environ. Sci. Technol. 2017, 51, 10137−10145. (9) Tang, Y. R.; Su, Y. Y.; Yang, N.; Zhang, L. C.; Lv, Y. Anal. Chem. 2014, 86, 4528−4535. (10) You, Y.; Nam, W. Chem. Sci. 2014, 5, 4123−4135. (11) Umezawa, N.; Tanaka, K.; Urano, Y.; Kikuchi, K.; Higuchi, T.; Nagano, T. Angew. Chem., Int. Ed. 1999, 38, 2899−2901. (12) Li, Z.; Tao, L.; Lv, S. W.; Zhuang, Q. G.; Liu, Z. H. J. Am. Chem. Soc. 2015, 137, 11179−11185. (13) Kundu, K.; Knight, S. F.; Willett, N.; Lee, S.; Taylor, W. R.; Murthy, N. Angew. Chem., Int. Ed. 2009, 48, 299−303. (14) Zhou, W. J.; Cao, Y. Q.; Sui, D. D.; Lu, C. Anal. Chem. 2016, 88, 2659−2665. (15) Herman, J.; Zhang, Y. N.; Castranova, V.; Neal, S. L. Anal. Bioanal. Chem. 2018, 410, 6079−6095. (16) Lee, E. S.; Deepagan, V. G.; You, D. G.; Jeon, J.; Yi, G. R.; Lee, J. Y.; Lee, D. S.; Suh, Y. D.; Park, J. H. Chem. Commun. 2016, 52, 4132−4135. (17) Zhao, C. X.; Cui, H. B.; Duan, J.; Zhang, S. H.; Lv, J. G. Anal. Chem. 2018, 90, 2201−2209. (18) Nakashima, K.; Maki, K.; Kawaguchi, S.; Akiyama, S.; Tsukamoto, Y.; Imai, K. Anal. Sci. 1991, 7, 709−713. (19) Singh, A.; Seo, Y. H.; Lim, C. K.; Koh, J.; Jang, W. D.; Kwon, I. C.; Kim, S. ACS Nano 2015, 9, 9906−9911. (20) Cho, S. J.; Hwang, O.; Lee, I. J.; Lee, G. Y.; Yoo, D.; Khang, G.; Kang, P. M.; Lee, D. W. Adv. Funct. Mater. 2012, 22, 4038−4043. (21) Afanas’ev, I. B.; Ostrakhovitch, E. A.; Mikhal’chik, E. V.; Korkina, L. G. Luminescence 2001, 16, 305−307. (22) Li, P.; Liu, L.; Xiao, H.; Zhang, W.; Wang, L.; Tang, B. J. Am. Chem. Soc. 2016, 138, 2893−2896. (23) Niu, J.; Fan, J.; Wang, X.; Xiao, Y.; Xie, X.; Jiao, X.; Sung, C.; Tang, B. Anal. Chem. 2017, 89, 7210−7215. (24) Faulkner, K.; Fridovich, I. Free Radical Biol. Med. 1993, 15, 447−451. (25) Li, X. H.; Zhang, G. X.; Ma, H. M.; Zhang, D. Q.; Li, J.; Zhu, D. B. J. Am. Chem. Soc. 2004, 126, 11543−11548. (26) Zhang, G. X.; Li, X. H.; Ma, H. M.; Zhang, D. Q.; Li, J.; Zhu, D. B. Chem. Commun. 2004, 2072−2073. (27) Hananya, N.; Green, O.; Blau, R.; Satchi-Fainaro, R.; Shaba, D. Angew. Chem., Int. Ed. 2017, 56, 11793−11796. (28) Tsai, C. H.; Stern, A.; Chiou, J. F.; Chern, C. L.; Liu, T. Z. J. J. Agric. Food Chem. 2001, 49, 2137−2141. (29) Miller, C. J.; Rose, A. L.; Waite, T. D. Anal. Chem. 2011, 83, 261−268. (30) Zhou, W. J.; Cao, Y. Q.; Sui, D. D.; Lu, C. Angew. Chem., Int. Ed. 2016, 55, 4236−4241. (31) Sun, M. X.; Deng, D. Y.; Zhang, K. X.; Lu, T.; Su, Y. Y.; Lv, Y. Chem. Commun. 2016, 52, 11259−11262. (32) Sangtrirutnugul, P.; Chaiprasert, T.; Hunsiri, W.; Jitjaroendee, T.; Songkhum, P.; Laohhasurayotin, K.; Osotchan, T.; Ervithayasuporn, V. ACS Appl. Mater. Interfaces 2017, 9, 12812− 12822.
■
CONCLUSIONS In summary, we designed and evaluated a new class of luminescent molecular probes for •OH selective detection. Thirteen kinds of organosiloxane compounds with aliphatic groups were observed to produce CL in the presence of •OH. The combination of low HOMO energies, the presence of Si− O bond, and high negative charge density of aliphatic groups played decisive roles in CL generation. By anchoring a dye molecule on cage-like T8-AEAP-POSS, an intramolecular CRET system was constructed. The proposed probe was successfully used to directly monitor inherent •OH in ambient particular matter. DFT, ion chromatograph, XPS, particle size analysis, and FL studies were carried out to study the CL mechanism. •OH preferred to attack position-1 of the C atom on organosiloxane compounds or POSS to generate electronically carbonyl CH3CO• as the CL emitter. Luminescent nanoparticles were generated in situ during the CL process as the products. This study provides a class of the most effective known luminescent molecular probes for selective detection of • OH. More importantly, further enhanced CL signal and the red shift of wavelength made it have great potential for various applications, such as bioanalysis, biosensors, and biological imaging.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.9b00637. Description of the extensive method, general selectivity study, CL quantum yield of POSS, ion chromatograph analysis, structure of different cage-like POSS, DFT results of MOs and atomic polar tensors analysis, XRD, FTIR, 29Si NMR, ESI−MS, 1H NMR, EPR, XPS, UV/ vis absorption, FL emissions, size analysis, and TEM results (PDF)
■
REFERENCES
AUTHOR INFORMATION
Corresponding Author
*Phone/fax: +86-28-8541-2798. E-mail:
[email protected]. ORCID
Yi Lv: 0000-0002-7104-2414 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We gratefully acknowledge financial support from the National Natural Science Foundation of China (nos. 21575093 and 21675113) and the Science & Technology Department of Sichuan Province of China (19YYJC2423). Dr. Pengchi Deng and Dr. Shuguang Yan from the Analytical & Testing Center, 8931
DOI: 10.1021/acs.analchem.9b00637 Anal. Chem. 2019, 91, 8926−8932
Article
Analytical Chemistry (33) Apel, K.; Hirt, H. Annu. Rev. Plant Biol. 2004, 55, 373−399. (34) Page, S. E.; Wilke, K. T.; Pierre, V. C. Chem. Commun. 2010, 46, 2423−2425. (35) Zhuang, M.; Ding, C.; Zhu, A.; Tian, Y. Anal. Chem. 2014, 86, 1829−1836. (36) Li, Z.; Liang, T.; Lv, S.; Zhuang, Q.; Liu, Z. J. Am. Chem. Soc. 2015, 137, 11179−11185. (37) Gao, H.-Y.; Mao, L.; Li, F.; Xie, L.-N.; Huang, C.-H.; Shao, J.; Shao, B.; Kalyanaraman, B.; Zhu, B.-Z. Environ. Sci. Technol. 2017, 51, 2934−2943. (38) Mao, L.; Liu, Y.-X.; Huang, C.-H.; Gao, H.-Y.; Kalyanaraman, B.; Zhu, B.-Z. Environ. Sci. Technol. 2015, 49, 7940−7947. (39) Anderson, S. E.; Bodzin, D. J.; Haddad, T. S.; Boatz, J. A.; Mabry, J. M.; Mitchell, C.; Bowers, M. T. Chem. Mater. 2008, 20, 4299−4309. (40) Janeta, M.; John, L.; Ejfler, J.; Lis, T.; Szafert, S. Dalton Trans. 2016, 45, 12312−12321. (41) Hanprasit, S.; Tungkijanansin, N.; Prompawilai, A.; Eangpayung, S.; Ervithayasuporn, V. Dalton Trans. 2016, 45, 16117−16120. (42) El Aziz, Y.; Bassindale, A. R.; Taylor, P. G.; Horton, P. N.; Stephenson, R. A.; Hursthouse, M. B. Organometallics 2012, 31, 6032−6040. (43) Hu, J.; Xu, K. L.; Jia, Y. Z.; Lv, Y.; Li, Y. B.; Hou, X. D. Anal. Chem. 2008, 80, 7964−7969. (44) Kellogg, R. E. J. Am. Chem. Soc. 1969, 91, 5433−5436. (45) Huang, X. Y.; Li, L.; Qian, H. F.; Dong, C. Q.; Ren, J. C. Angew. Chem. 2006, 118, 5264−5267. (46) So, M. K.; Xu, C. J.; Loening, A. M.; Gambhir, S. S.; Rao, J. H. Nat. Biotechnol. 2006, 24, 339−343. (47) Bag, S.; Tseng, J. C.; Rochford, J. Org. Biomol. Chem. 2015, 13, 1763−1767. (48) Ben, H.-J.; Ren, X.-K.; Song, B.; Li, X.; Feng, Y.; Jiang, W.; Chen, E.-Q.; Wang, Z.; Jiang, S. J. Mater. Chem. C 2017, 5, 2566− 2576. (49) Zheng, L.; Wang, Y. X.; Zhang, G. X.; Xu, W. B.; Han, Y. J. J. Lumin. 2010, 130, 995−999. (50) Gehling, W.; Khachatryan, L.; Dellinger, B. Environ. Sci. Technol. 2014, 48, 4266−4272. (51) Khachatryan, L.; Dellinger, B. Environ. Sci. Technol. 2011, 45, 9232−9239.
8932
DOI: 10.1021/acs.analchem.9b00637 Anal. Chem. 2019, 91, 8926−8932
