A new fluorescent pH probe for extremely acidic conditions

Analytica Chimica Acta 820 (2014) 146–151

Contents lists available at ScienceDirect

Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca

A new fluorescent pH probe for extremely acidic conditions Yu Xu a,c , Zheng Jiang b,c , Yu Xiao a,c , Fu-Zhen Bi a , Jun-Ying Miao b,∗ , Bao-Xiang Zhao a,∗ a

School of Chemistry and Chemical Engineering, Shandong University, Jinan 250100, PR China School of Life Science, Shandong University, Jinan 250100, PR China c Taishan College, Shandong University, Jinan 250100, PR China b

h i g h l i g h t s

g r a p h i c a l

• A new fluorescence probe for very

A new coumarin-based fluorescent probe can detect highly acidic conditions in both solution and bacteria with high selectivity and sensitivity.

low pH was synthesized and characterized. • The probe can monitor pH in solution and bacteria. • The two-step protonation of N atoms of the probe leads to fluorescence quenching.

a r t i c l e

i n f o

Article history: Received 27 November 2013 Received in revised form 13 February 2014 Accepted 20 February 2014 Available online 22 February 2014 Keywords: Coumarin Fluorescence probe pH Highly acidic condition E. coli

a b s t r a c t A novel turn-off fluorescent probe based on coumarin and imidazole moiety for extremely acidic conditions was designed and developed. The probe with pKa = 2.1 is able to respond to very low pH value (below 3.5) with high sensitivity relying on fluorescence quenching at 460 nm in fluorescence spectra or the ratios of absorbance maximum at 380 nm to that at 450 nm in UV–vis spectra. It can quantitatively detect pH value based on equilibrium equation, pH = pKa − log[(Ix − Ib )/(Ia − Ix )]. It had very short response time that was less than 1 min, good reversibility and nearly no interference from common metal ions. Moreover, using 1 H NMR analysis and theoretical calculation of molecular orbital, we verified that a two-step protonation process of two N atoms of the probe leaded to photoinduced electron transfer (PET), which was actually the mechanism of the fluorescence quenching phenomenon under strongly acidic conditions. Furthermore, the probe was also applied to imaging strong acidity in bacteria, E.coli and had good effect. This work illustrates that the new probe could be a practical and ideal pH indicator for strongly acidic conditions with good biological significance. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Proton, as a familiar cation, plays key roles in biological system. The fluctuation of pH has obvious effect on numerous cellular events, such as cellular metabolism [1–3], cellular growth [4], signal transduction [5], chemotaxis [6], apoptosis

∗ Corresponding authors. Tel.: +86 531 88366425; fax: +86 531 88564464. E-mail addresses: [email protected] (J.-Y. Miao), [email protected], [email protected] (B.-X. Zhao). http://dx.doi.org/10.1016/j.aca.2014.02.029 0003-2670/© 2014 Elsevier B.V. All rights reserved.

a b s t r a c t

[7] and autophagy [8]. Therefore, monitoring pH changes inside living cells is crucial for exploring cellular functions and understanding physiological and pathological processes in organisms. Most of the known pH fluorescent indicators belong to two categories. Some of them respond to neutral pH range from 6 to 8 [9–11]. Others can detect weak acidic pH in the range from 4 to 6 [12–14]. But very few have been applicable for more acidic conditions with pH below 4 [15–17]. Thus, monitoring the very low intracellular pH conditions is still challenging. In spite of the fact that the majority of living organisms could hardly survive in

Y. Xu et al. / Analytica Chimica Acta 820 (2014) 146–151

147

strong acidic environment, there still exists a considerable number of microorganisms, such as helicobacter pylori and “acidophiles”, that particularly favor this harsh living conditions [15,16]. Moreover, in some eukaryotic cells, acidic pH has important effect on organelles along the secretory and endocytic pathways [18,19]. Enteric pathogen is another example, which is able to reach small intestine by passing through the highly acidic mammalian stomach, causing life-threatening infections [20]. Even for mammals, there are some parts with very low pH value, such as gastric juice, the pH level of which can also influence their physiological process remarkably [21]. Because of lacking effective ways to detect such acidic pH in living species, the precise pH values in these cellular compartments remain elusive [17]. Thus, it is necessary and meaningful to develop new fluorescent probe that can be applied in such strong acidic conditions in living systems. Here, based on coumarin, we designed and synthesized a new pH probe with a pKa 2.1 and applies it to bacteria.

was obtained as light yellow crystals in 35% yield (400 mg). mp: 242–244 ◦ C. IR (KBr), : 3436, 3348, 3235, 2975, 2896, 1685, 1563, 1451, 1373, 1306, 1246, 1189, 1127, 1011, 902, 860, 793, 697, 642, 597, 508, 441 cm−1 ; 1 H NMR (DMSO-d6 , 300 MHz), ı (ppm): 1.11 (t, 6H, NCH2 CH3 , J = 6.6 Hz), 3.39–3.41 (m, 4H, NCH2 CH3 ), 3.63 (s, 12H, OCH3 ), 5.32 (s, 1H, tert-CH), 6.27 (s, 4H, NH2 ), 6.51 (s, 1H, coumarin-H), 6.59 (d, 1H, coumarin-H, J = 8.7 Hz), 6.96 (s, 1H, coumarin-H), 7.36 (d, 1H, coumarin-H, J = 8.7 Hz); 13 C NMR (DMSOd6 , 75 MHz), ı (ppm): 168.24 (4C), 161.24, 160.23, 154.71 (2C), 149.21, 136.72, 128.59, 122.94, 108.37 (2C), 96.27, 92.43, 53.21 (4C), 43.87 (2C), 30.16, 14.05, 12.31 (2C); HRMS: calcd. for C26 H32 N7 O6 + [M + H]+ 538.2414, found: 538.2477. Additionally, the single crystal of L was obtained by volatilizing the mixed solvent (ethanol/ethyl acetate = 1:1, v/v) slowly at ambient temperature. The crystal structure was determined by X-ray single crystal diffraction.

2. Materials and methods

In this study, all the calculations were implemented with the Gaussian09 program package [25]. The structure of these molecules in ground state were optimized using the density functional theory (DFT) method, CAM-B3LYP, with the 6-31g** basis set. Vibrational frequency analyses were carried out to ensure the minimums of the ground state were reached on the potential energy surfaces. On the basis of these optimized structures, the absorption spectra were predicted by time-dependent (TD-DFT) method. The solvent effects were modeled with the polarizable continuum model (PCM) model.

2.1. Materials All reagents and solvents were purchased from commercial sources and used without further purification. The solutions of metal ions were prepared from nitrate salts which were dissolved in deionized water. Deionized water was used throughout the process of absorption and fluorescence determination. All samples were prepared at room temperature, shaken for 10 s and rested for 1 h before UV–vis and fluorescence determination. Britton–Robinson (B–R) buffer was prepared with 40 mM acetic acid, boric acid, and phosphoric acid. Dilute hydrochloric acid or sodium hydroxide was used for tuning pH values. 2.2. Instruments Thin-layer chromatography (TLC) involved silica gel 60 F254 plates (Merck KGaA). Melting points were determined on an XD4 digital micro melting point apparatus. 1 H NMR spectra were recorded on a Bruker Avance 300 (300 MHz) spectrometer and 13 C NMR spectra were recorded on a Bruker Avance 300 (75 MHz) spectrometer, using d6 -DMSO as solvent and tetramethylsilane (TMS) as an internal standard. IR spectra were recorded with an IR spectrophotometer VERTEX 70 FT-IR (Bruker Optics). HRMS spectra were recorded on a Q-TOF6510 spectrograph (Agilent). Fluorescent measurements were recorded on an F-7000 (Hitachi) luminescence spectrophotometer and UV–vis spectra were recorded on a U-4100 UV–Vis–NIR Spectrometer (Hitachi). The pH measurements were measured by use of a PHS-3C digital pH-meter (YouKe, Shanghai). Images of E. coli cells were captured with a laser confocal microscope (Carl Zeiss LSM-700, Germany). The single crystal was measured on a Bruker-AXS CCD single-crystal diffractometer with graphite-monochromated Mo K␣ radiation source ( = 0.71073 Å). 2.3. Synthesis of 3-(bis(2-amino-4,6-dimethoxypyrimidin-5-yl) methyl)-7-(diethylamino)-2H-chromen-2-one (L) 7-(Diethylamino)-2-oxo-2H-chromene-3-carbaldehyde 1 and 4,6-dimethoxypyrimidin-2-amine 2 was synthesized as described previously in the literature [22–24]. Compound 1 (0.735 g, 3 mmol) was dissolved in 20 mL ethanol, while 4,6-dimethoxypyrimidin-2-amine 2 (0.515 g, 3.3 mmol) was dissolved in 10 mL ethanol. Mix the two shares of solution and add 4 drops of glacial acetic acid. The mixture was then heated to reflux for 12 h. The solvent was evaporated and the crude product was purified by column chromatography using petroleum ether/ethyl acetate (1:2, v/v) as an eluent. The target compound L

2.4. Calculation methods

2.5. Bacteria culture and imaging E. coli (Trans 5a) was incubated at 37 ◦ C in Luria-Bertani (LB) culture (Trptone 10 g/L, yeast extract 5 g/L, NaCl 10 g/L) for 12 h in a table concentrator (ZHI CHENG ZHWY-2112B, China) at 180 rpm. Then the culture was centrifuged (Heal Force Neofuge-18R, China) in 10 mL Eppendorf tubes at 5000 rpm for 5 min to collect E. coli cells. The sediment was resuspended with hydrochloric acid at different pH (0.61, 2.07, 3.62), respectively. 5 min after resuspension, the pH probe dissolved in DMSO was added into every tube to make the final probe concentration to be 10 ␮M. E. coli cells with the probe were incubated in a table concentrator as mentioned above for 2 h, then smeared on slides and observed by laser confocal microscopy (Carl Zeiss LSM-700, Germany) at the wavelength of 405 nm. 3. Results and discussion 3.1. Synthesis of probe (L) The general synthetic route of probe L is given in Scheme 1. The structure of probe L was characterized by IR, 1 H NMR, 13 C NMR, HRMS spectra and X-ray single crystal diffraction (CCDC No. 962732, Fig. 1). 3.2. Spectroscopic properties and optical responses to pH Spectroscopic properties of probe L were studied. From Fig. 2, we can find that the fluorescence intensity was high and stable when the pH value of the buffer is above 3.98, while it decreased drastically when the buffer pH went down from 3.98 to 0.65. Meanwhile, the quantum yield () decreased from 0.65 to 0.06 as calculated by the following formula, according to the literature [26]. ˚u =

(˚s )(Fu )(As )(u )2 (Fs )(Au )(s )2

148

Y. Xu et al. / Analytica Chimica Acta 820 (2014) 146–151

Scheme 1. The synthetic route of probe L.

quantitative relationship can be described by the equilibrium function as follow [27]: pH = pKa − log[

Fig. 1. The crystal structure of probe L.

where  is fluorescence quantum yield, F is the integrated area under the corrected emission spectrum (excitation wavelength 385 nm), A is the absorbance at the excitation wavelength,  is the refractive index of the solution, and subscripts u and s are the unknown and standard, respectively. Using quinine sulphate dehydrate (99.0%) in 0.1 N H2 SO4 as the main standard. Moreover, in Fig. 2(b), the plot of emission based intensity versus pH exhibited “S” shaped calibration graph and its

Ix − Ib ] Ia − Ix

where Ia = 5540 and Ib = 0, which are the fluorescent intensity of the probe in its acid and conjugate base form, respectively. The pKa value was calculated to be 2.1 according to the equation in literature [28]. In Fig S1, we described the linear regression relationship between “log[(Ia − Ix )/(Ix − Ib )]” and pH value with the function of Y = −1.16 pH + 2.17 and R2 = 0.9909. From this plot, we can calculate the pH value of any sample with pH ranged from 0.5 to 3.5 based on their fluorescent intensity. Moreover, the fluorescence intensity of probe L is reversible between pH 1 and pH 7 (Fig. S2), which allows it to monitor a system with a shifty pH value and report the real time acidity. The time course analysis (Fig. S3) revealed its fast response property (<1 min) and the interference of metal ions to the probe is negligible (Figs. S4 and S5), which guarantees the effective use of the probe in complex environment. As shown in absorption spectra (Fig. 3(a)), in neutral or weakly acidic solution (pH > 3.5), the absorption maximum peaked at 450 nm, however it decreased and a new absorption maximum at 380 nm appeared when pH of the buffer decreased from 3.5, resulting in an isosbestic point at 401 nm. Moreover, the ratios of the absorbance maximum at 380 nm to that at 450 nm (A380 /A450 ) varied with buffer’s pH as shown in Fig. 3(b), which provided us with another way to monitor pH quantitively.

Fig. 2. (a) Fluorescence spectra of L (10 ␮M) in solution (1:1, B-R–EtOH, v/v, 0.1 M NaCl) with different pH, ex = 385 nm; (b) Fluorescence intensity at 460 nm versus pH, ex = 385 nm.

Y. Xu et al. / Analytica Chimica Acta 820 (2014) 146–151

149

Fig. 3. (a) UV–vis absorption spectra of L (10 ␮M) in solution (B-R:ethanol = 1:1, v:v) at different pH. (b) The ratio of the absorbance at 380 nm to that at 450 nm varied with pH.

3.3. The mechanism of detecting the strong acidity 3.3.1. 1 H NMR Fig. 4 shows the 1 H NMR comparison of probe L in neutral and strong acidic conditions respectively. The downfield shift of H(1) 0.208 ppm indicated the protonation of the N atom in diethylamino group, and the downfield shift of H(2) 0.228 ppm was possibly contributed to the protonation of the N atom in pyrimidine ring. The chemical shift of H(3), H(4), H(5) and H(6) were due to the intense influence of the protonated N atoms on the distribution of electron density on coumarin ring. Besides, the peak of H(7) disappeared because of the proton exchange between H and D on amino-group. There is the same result in the 1 H NMR titration with trifluoroacetic acid (TFA) (Fig. S6). With the addition of TFA, the pH of the system went down gradually, which leaded to the peaks of H(1–6) shifted to the downfield step by step. According to these two figures, we know that it is the protonation process of some N atoms of the probe, rather than chemical reaction that leads to the fluorescence

quenching. Chances are high that the N atom in diethylamino group and the N atom in pyrimidine ring were protonized in acidic conditions. But based on these experimental data, we still cannot sure about the exact protonation positions, the protonation order and which step of protonation leads to the changes of optical properties of the probe. Therefore, we continued to do some theoretical calculation to address these problems. 3.3.2. Theoretical calculation To understand the mechanism of fluorescence quenching caused by the protonation, we analyzed the protonation process by theoretical calculation. Comparing the energy of every possible optimized structure with one proton bonded to a heteroatom of the probe using the density functional theory (DFT) method, we found that the N atom of pyrimidine is the easiest point to be protonated. After the first-step protonation, using the same method, we got that the second proton should be added to the N atom of diethylamino group (Scheme 2). The optimized structures of L, one-step

Fig. 4. (a) The NMR spectroscopy of L in DMSO-d6 after adding a drop of DCl; (b) The NMR spectroscopy of L in DMSO-d6 .

150

Y. Xu et al. / Analytica Chimica Acta 820 (2014) 146–151

Scheme 2. The protonation process of probe L in extreme acidic condition.

protonated L (LH) and two-step protonated L (LH2) are shown in Fig. S7. It is worth noting that probe L consist of a coumarin ring as a fluorophore linked to two pyrimidine moieties as electron donors via a tertiary carbon as a spacer. This is a typical structure that PET which takes place from the pyrimidine moiety to the coumarin may cause fluorescence quenching of the fluorophore [29]. Thus, we calculated the molecular orbital of probe L, LH and LH2 and explained the fluorescence quenching mechanism through PET, according to the method reported in [30]. On the basis of optimized ground-state structures of the studied molecules, we drew the electron distribution and picked out the orbitals mainly located in fluorophore part to form the left column; similarly, orbitals mainly located in the donor part are picked to form the right column in Fig. S8. As for probe L and LH, when an electron of the HOMO of fluorophore is excited to its LUMO by radiation, the PET process between fluorophore and donor is impossible to occur, because the HOMO of the donor is lower in energy than that of the fluorophore. Consequently, after a very short time, the electron has to jump back to the HOMO of the fluorophore. Meanwhile, energy is released in the form of fluorescence emission and that is the reason why L and LH exhibit strong fluorescence property. However, upon the secondstep protonation, the HOMO of the donor of LH2 is higher in energy than that of the fluorophore. Thus, after an electron is promoted from the HOMO of the fluorophore to its LUMO, an empty position forms in the HOMO of the fluorophore. Meanwhile, another electron from the HOMO of the donor transfers to this empty position, leaving another empty position in the HOMO of the donor. Then, the excited electron will transfer to this new empty position on HOMO of the donor, rather than jump back to the HOMO of the fluorophore. Therefore, the energy is released via PET process without radiation and this process causes fluorescence quenching of LH2. Moreover, the absorption maximum of LH2 predicted by theoretical calculation (Fig. S9) had an obvious blue-shift, compared to that of L and LH. Although there was some deviation, the tendency of the variation of the absorption maximum was corresponded to the experimental results. Therefore, we can conclude that a two-step protonation process occurred when the probe was in extreme acidic condition. The changes of the spectroscopic properties were due to the secondstep protonation via PET. 3.4. Fluorescence imaging in bacteria To verify the potential biological application of probe L, we detected the highly acidic condition in bacteria. To simulate the strongly acidic environment in several types of bacteria [15,16], we used buffer with pH 3.51, 2.07 and 0.65, respectively, to incubate E. coli. Then, we added probe L to measure the pH value and imaging it. From the image captured with the fluorescence microscope (Fig. 5(a)), it is clear that the bacteria in highly acidic medium with the pH 0.65 had no fluorescence, while those in the environment with pH 2.07 exhibited visible fluorescence and the fluorescence

Fig. 5. (a) Imaging acidity in E. coli cells with probe L (10 ␮M). (a–c) pH 0.61; (d–f) pH 2.07; (g–i) pH 3.62. a, d, g are the fluorescence images; b, e, h are the white light images; c, f, i are the overlaps of the two (scale bar: 5 ␮m). (b) The fluorescence intensity quantitation was analyzed by the ImageJ. Results are presented as means ± SE with replicates n = 3.

Y. Xu et al. / Analytica Chimica Acta 820 (2014) 146–151

intensity of those in the buffer with pH 3.51 is much higher. Furthermore, we quantified the fluorescence intensity using ImageJ and depicted the fluorescence intensities by color histogram at the blue channel (Fig. 5(b)). Based on the fluorescence intensity extracted from the images, we believe that we could calculate a specific pH to an image of bacteria according to the equilibrium equation, pH = pKa − log[(I − Imin )/(Imax − I)] where Imax and Imin are the fluorescent intensity of the probe in their acid and conjugate base form respectively and I is the fluorescence intensity observed from images. These results indicate the probe is able to imaging very low pH condition in biological system and we believe that it can work well in other real biological systems with highly acidic environment. 4. Conclusions In summary, linking coumarin fluorophore to imidazole derivative, a new “turn off” fluorescent probe was designed and developed to detect extremely low pH. The fluorescence intensity peaked at 460 nm remained high and stable in neutral, basic and weak acidic environment. When pH decreased from 3.5, the fluorescence decreased drastically. The “S” shaped plot allowed us to quantitatively detect pH value according to equilibrium equation, pH = pKa − log[(Ix − Ib )/(Ia − Ix )] or the linear regression based log[(Ia − Ix )/(Ix − Ib )] versus pH. Also, the absorption maximum at 450 nm in UV–vis spectra decreased and a new absorption peak at 380 nm appeared and increased simultaneously. The probe had good reversibility, good selectivity and very short response time that less than 1 min. Thus, it had big potential to be used in complicated conditions and monitor real time pH. More importantly, the cell assay proved that the probe had good effect in imaging strong acidity in bacteria and could be able to detect the pH value via analyzing the fluorescence intensity extracted from the images. Furthermore, relying on the changes of peak positions in 1 H NMR spectra and theoretical calculations of the energy of molecular orbital, the detection mechanism had been verified to PET, caused by a two-step protonation process of two N atoms of the probe molecule. All in all, the probe had good ability to detect and imaging low pH conditions in both solution and bacteria. We believe it will be beneficial to study in chemical and biological systems. Acknowledgements This study was supported by 973 Program (2010CB933504) and Foundation of Talent Training of Fundamental Subject of China (Grant No: J1103314). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.aca.2013.12.001.

151

References [1] M.T. Montgomery, T.J. Boyd, C.L. Osburn, R.E. Plummer, S.M. Masutani, R.B. Coffin, Desalination 249 (2009) 861–864. [2] T. Otomo, K. Higaki, E. Nanba, K. Ozono, N. Sakai, Journal of Biological Chemistry 286 (2011) 35283–35290. [3] X.Y. Wang, Z.F. Pi, W.L. Liu, Y.F. Zhao, S.Y. Liu, Chinese Journal of Chemistry 28 (2010) 2494–2500. [4] Y.H. Oh, I.J. Oh, C. Jung, S.Y. Lee, J. Lee, Journal of Microbiology and Biotechnology 20 (2010) 1677–1680. [5] J. Sjöholm, K.G.V. Havelius, F. Mamedov, S. Styring, Biochemistry 49 (2010) 9800–9808. [6] W.S. Zhang, B. Tang, X. Liu, Y.Y. Liu, K.H. Xu, J.P. Ma, L.L. Tong, G.W. Yang, Analyst 367 (2009) 367–371. [7] C. Kawai, F.S. Pessoto, C.V. Graves, A.M. Carmona-Ribeiro, I.L. Nantes, Journal of Bioenergetics and Biomembranes 45 (2013) 421–430. [8] J.W. Wojtkowiak, R.J. Gillies, Autophagy 8 (2012) 1688–1689. [9] B.K. McMahon, R. Pal, D. Parker, Chemical Communications 49 (2013) 5363–5365. [10] U.C. Saha, K. Dhara, B. Chattopadhyay, S.K. Mandal, S. Mondal, S. Sen, M. Mukherjee, S.V. Smaalen, P. Chattopadhyay, Organic Letters 13 (2011) 4510–4513. [11] R. Gui, X. An, W. Huang, Analytica Chimica Acta 767 (2013) 134–140. [12] H.S. Lv, S.Y. Huang, B.X. Zhao, J.Y. Miao, Analytica Chimica Acta 788 (2013) 177–182. [13] M. Liu, Z. Ye, C. Xin, J. Yuan, Analytica Chimica Acta 761 (2013) 149–156. [14] X. Du, N. Lei, P. Hu, Z. Lei, D.H. Ong, X. Ge, Z. Zhang, M.H. Lam, Analytica Chimica Acta 787 (2013) 193–202. [15] H.L. Li, H. Guan, X.R. Duan, J. Hu, G.R. Wang, Q. Wang, Organic and Biomolecular Chemistry 11 (2013) 1805–1809. [16] M.Z. Tian, X.J. Peng, J.L. Fan, J.Y. Wang, S.G. Sun, Dyes and Pigments 95 (2012) 112–115. [17] M.Y. Yang, Y.Q. Song, M. Zhang, S.X. Lin, Z.Y. Hao, Y. Liang, Angewandte ChemieInternational Edition 51 (2012) 7674–7679. [18] T.A. Krulwich, G. Sachs, E. Padan, Nature Reviews Microbiology 9 (2011) 330–343. [19] G. Loving, B. Imperiali, Journal of the American Chemical Society 130 (2008) 13630–13638. [20] D.S. Merrell, A. Camilli, Current Opinion in Microbiology 5 (2002) 51–55. [21] H.Y. Fan, J. Wang, G.C. Yan, X.H. Huo, L.J. Mu, J.K. Chu, W.W. Niu, Z.Y. Duan, J.C. Ma, J. Wang, Z.Y. Wang, Experimental and Therapeutic Medicine 5 (2013) 912–916. [22] Y.T. Yang, F.J. Huo, J.J. Zhang, Z.H. Xie, J.B. Chao, C.X. Yin, H.B. Tong, D.S. Liu, S. Jin, F.Q. Cheng, X.X. Yan, Sensors and Actuators B 166 (2012) 665–670. [23] J.S. Wu, W.M. Liu, X.Q. Zhuang, F. Wang, P.F. Wang, S.L. Tao, X.H. Zhang, S.K. Wu, S.T. Lee, Organic Letters 9 (2007) 33–36. [24] K.S. Parikh, S.P. Vyas, Journal of Chemical and Pharmaceutical Research 4 (2012) 2109–2111. [25] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery, J.E. Peralta Jr, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, O. Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, in: Gaussian 09, Revision B.01, Gaussian Inc, Wallingford, CT, 2010. [26] Z.L. Gong, L.W. Zheng, B.X. Zhao, Journal of Luminescence 132 (2012) 318–324. [27] S. Derinkuyu, K. Ertekin, O. Oter, S. Denizalti, E. Cetinkaya, Analytica Chimica Acta 588 (2007) 42–49. [28] A. Lobnik, O.I. Murkovic, O.S. Wolfbeis, Analytica Chimica Acta 367 (1998) 159–165. [29] B. Valeur, Molecular Fluorescence Principles and Applications, Wiley-VCH, Weinheim, 2002, p. 287. [30] Z. Xu, A.M. Ren, J.F. Guo, X.T. Liu, S. Huang, J.K. Feng, Photochemistry and Photobiology 89 (2013) 300–309.