Polyaromatic Hydrocarbon (PAH)-Based Aza-POPOPs: Synthesis, Photophysical Studies, and Nitroanalyte Sensing Abilities

1,4-Bis(5-phenyl-2-oxazolyl)benzene (POPOP) is a common scintillation fluorescent laser dye. In this manuscript, the synthesis of 2-Ar-5-(4-(4-Ar’-1H-1,2,3-triazol-1-yl)phenyl)-1,3,4-oxadiazoles (Ar, Ar’ = Ph, naphtalenyl-2, pyrenyl-1, triphenilenyl-2), as PAH-based aza-analogues of POPOP, by means of Cu-catalyzed click reaction between 2-(4-azidophenyl)-5-Ar-1,3,4-oxadiazole and terminal ethynyl-substituted PAHs is reported. An investigation of the photophysical properties of the obtained products was carried out, and their sensory response to nitroanalytes was evaluated. In the case of pyrenyl-1-substituted aza-POPOP, dramatic fluorescence quenching by nitroanalytes was observed.


Introduction
One of the most important tasks of modern synthetic organic chemistry is obtaining new compounds that will find wide application in various industrial areas and medicine. One of the most representative examples of a novel type of compounds is azaheterocyclic fluorophores, which can also act as drug candidates and contain cyclic azole or azine core as a common pharmacophore or ligand unit, as well as fused or conjugated (poly)aromatic moieties as fluorogenes and/or receptor units. 1,4-Bis(5-phenyl-2-oxazolyl)benzene (POPOP) is a well-known organic fluorophore (Figure 1), which is commonly used as a spectrum shifter, including in multilayer scintillation screen to visualize radiation that is invisible to the human eye [1,2]. Due to its excellent photophysical properties, especially its high quantum yield (up to 97.5% in cyclohexane or 91% in ethanol [3]), POPOP is successfully utilized in dye vapor lasers [4,5]. On the other hand, aza-analogues of POPOP, such as 1,3,4oxadiazoles, are of wide interest due to the promising biological activity of oxadizoles [6][7][8] and their intriguing photophysical properties [9][10][11][12].
The term "click chemistry" was introduced for the first time in 1998 by K. Barry Sharpless, the 2001 and 2022 Nobel Prize Laureate in Chemistry, and it was fully described by B.K. Sharpless, H. C. Kolb, and M.G. Finn in 2001 [13]. Since then, this approach has gained worldwide acknowledgment for its simple reaction technique leading to single product formation without by-products. The copper(I)-catalyzed azide/alkyne "click" reaction (also termed Sharpless "click" reaction) occurs through the interaction between a terminal alkyne and an azide in the presence of Cu(I) catalysis, and results in a cycloaddition product-1,4 disubstituted 1,2,3-triazole [14][15][16][17]. This reaction is also known as Cu(I)-catalyzed Huisgen 1,3-dipolar cycloaddition between alkynes and azides (CuAAC) [18,19]. Additionally, 1,3-dipolar cycloaddition of organic azides with alkynes as dipolarophiles is the most straightforward way to obtain useful 1,2,3-triazoles [5]. Applications of the Cu-catalyzed (cyclo)addition reactions have already contributed to many areas of modern chemistry, including asymmetric synthesis [20][21][22][23][24], and Cu(I)catalyzed azide-alkyne cycloaddition (CuAAC) is a convenient tool for bioconjugation reactions, peptidomimetic chemistry, polymer and materials sciences, and supramolecular chemistry [25,26]. The chemistry of 1,2,3-triazoles has gained much attention since its discovery, and various synthetic protocols have been developed for the synthesis of this moiety [27,28]. In addition, 1,2,3-triazoles are one of the most important connective linkers and functional aromatic heterocycles in modern chemistry [29]. In addition, it is well known that 1,2,3-triazoles, as highly valuable N-heterocyclic compounds, are ubiquitous in many pharmaceuticals and bioactive molecules [30,31]. In this manuscript, we report the synthesis of novel POPOP aza-analogues (Figure 1) via the click reaction between 2-(4-azidophenyl)-5-(aryl)-oxadiazole-1,3,4 and ethynylsubstituted (poly)arenes, as well as an investigation of the photophysical and sensing properties of the obtained products.

Synthesis of Target Fluorophores
As one of the azole fragments, the aza-analog of 1,3-oxazole was used as it is easily derived synthetically [32,33] from 1,3,4-oxadiazol. As mentioned above, 1,3,4-oxadiazole fragment is widely presented in many compounds with various promising biological activities and possesses intriguing photophysical properties. 1,2,3-triazole, an azadeoxa analogue of 1,3-oxazole, was introduced as second azole by using Cu(I)-catalyzed click reaction. Based on these click reactions, we successfully synthesized 4-azidophenyloxadiazoles 2a,b as precursors of the azido components. These 4-azidophenyloxadiazoles were prepared for the first time by means of modified Sandmeyer reaction starting from 4-(5-phenyl-1,3,4oxadiazol-2-yl)anilines [34] (Scheme 1). As a second step, we used Cu(I)-promoted azido-alkyne coupling (CuAAC) (Scheme 2) to construct the 1,2,3-triazole ring by using two different approaches. The first approach involved the use of cuprous sulphate derived in situ via the reaction of sodium ascorbate with copper(II) sulphate pentahydrate in aqueous THF (H2O:THF 1:9 v/v) to obtain the target products 3 in 73-96% yields. The second approach involved modified reaction conditions, namely using cuprous bromide in dry DMF. The main reason for using dry DMF is the poor solubility of the starting PAH-ethynyl derivatives in aqueous THF. The As a second step, we used Cu(I)-promoted azido-alkyne coupling (CuAAC) (Scheme 2) to construct the 1,2,3-triazole ring by using two different approaches. The first approach involved the use of cuprous sulphate derived in situ via the reaction of sodium ascorbate with copper(II) sulphate pentahydrate in aqueous THF (H 2 O:THF 1:9 v/v) to obtain the target products 3 in 73-96% yields. The second approach involved modified reaction conditions, namely using cuprous bromide in dry DMF. The main reason for using dry DMF is the poor solubility of the starting PAH-ethynyl derivatives in aqueous THF. The target products 3 were obtained in 50-99% yields. It is worth mentioning that the main advantage of the reported copper(I)-catalized azido-alkyne coupling over the classical Huisgen 1,3-dipolar cycloaddition [35] is that the high regioselectivity of the former reaction results in only 1,4-isomer, whereas non-catalyzed 1,3-dipolar cycloaddition gives a mixture of 1,4-and 1,5-isomers [36]. As a second step, we used Cu(I)-promoted azido-alkyne coupling (CuAAC) (Scheme 2) to construct the 1,2,3-triazole ring by using two different approaches. The first approach involved the use of cuprous sulphate derived in situ via the reaction of sodium ascorbate with copper(II) sulphate pentahydrate in aqueous THF (H2O:THF 1:9 v/v) to obtain the target products 3 in 73-96% yields. The second approach involved modified reaction conditions, namely using cuprous bromide in dry DMF. The main reason for using dry DMF is the poor solubility of the starting PAH-ethynyl derivatives in aqueous THF. The target products 3 were obtained in 50-99% yields. It is worth mentioning that the main advantage of the reported copper(I)-catalized azido-alkyne coupling over the classical Huisgen 1,3-dipolar cycloaddition [35] is that the high regioselectivity of the former reaction results in only 1,4-isomer, whereas non-catalyzed 1,3-dipolar cycloaddition gives a mixture of 1,4-and 1,5-isomers [36].
The structures of all the obtained compounds were confirmed by means of 1 H and 13 C NMR spectroscopy, mass spectrometry, and elemental analysis (See ESI for details).

Photophysical and Sensing Properties of the Obtained Compounds
The above-mentioned azaheterocycles 3 could be considered as POPOP aza-analogs. However, initial photophysical studies showed that the photophysical properties of the obtained compounds 3 are not quite similar to those of POPOP, which are, probably, due to the contribution of both peripheral (poly)aromatic substituents and azole moieties. Thus, in the spectra of pyrene-bearing fluorophores 3c and 3g, both absorption and emission are red-shifted in comparison with POPOP, while the emission spectra of other fluorophores are blue-shifted compared to that of POPOP. Among all the fluorophores, a pronounced blue shift is observed for fluorophore 3a. Additionally, the emission spectrum The structures of all the obtained compounds were confirmed by means of 1 H and 13 C NMR spectroscopy, mass spectrometry, and elemental analysis (See ESI for details).

Photophysical and Sensing Properties of the Obtained Compounds
The above-mentioned azaheterocycles 3 could be considered as POPOP aza-analogs. However, initial photophysical studies showed that the photophysical properties of the obtained compounds 3 are not quite similar to those of POPOP, which are, probably, due to the contribution of both peripheral (poly)aromatic substituents and azole moieties. Thus, in the spectra of pyrene-bearing fluorophores 3c and 3g, both absorption and emission are red-shifted in comparison with POPOP, while the emission spectra of other fluorophores are blue-shifted compared to that of POPOP. Among all the fluorophores, a pronounced blue shift is observed for fluorophore 3a. Additionally, the emission spectrum of 3a is the most blue-shifted in comparison to the emission spectra of all other POPOP analogues. Whereas the absorption and emission spectra of POPOP have a weakly expressed vibronic structure, the spectra of all fluorophores 3, including pyrene-containing ones, are blurred. Interestingly, while S 0 →S 1 electronic transition has a higher intensity than S 0 →S 2 transition in POPOP and its analogues 3c and 3e, it is the other way around in the other analogues.
The quantum yield values of the obtained fluorophores vary from 17 to 98%. The introduction of a methoxy group into the aromatic substituent in the 1,3,4-oxazole core results in an increase in the quantum yield. It is especially clearly observed in the values of the quantum yields of pyrenyl-containing analogues 3c (23%) and 3g (84%)-the introduc-tion of a methoxy group results in a dramatical increase in their quantum yields. (Figure 2, Table 1).
containing ones, are blurred. Interestingly, while S0→S1 electronic transition has a higher intensity than S0→S2 transition in POPOP and its analogues 3c and 3e, it is the other way around in the other analogues.
The quantum yield values of the obtained fluorophores vary from 17 to 98%. The introduction of a methoxy group into the aromatic substituent in the 1,3,4-oxazole core results in an increase in the quantum yield. It is especially clearly observed in the values of the quantum yields of pyrenyl-containing analogues 3c (23%) and 3g (84%)-the introduction of a methoxy group results in a dramatical increase in their quantum yields. (Figure 2, Table 1). In addition, the introduction of a pyrene moiety results in an increase in the fluorescence lifetime values to 4.52 ns (3c) and 4.34 ns (3g), which are similar to the data reported for other pyrene derivatives in the literature [37][38][39]. For the other fluorophores 3, the lifetime values vary from 0.49 to 1.22 ns (Table 1).   In addition, the introduction of a pyrene moiety results in an increase in the fluorescence lifetime values to 4.52 ns (3c) and 4.34 ns (3g), which are similar to the data reported for other pyrene derivatives in the literature [37][38][39]. For the other fluorophores 3, the lifetime values vary from 0.49 to 1.22 ns (Table 1).
Next, DFT studies [67][68][69][70][71] were carried out to explain the efficient detection of PETN by the compound 3g. Thus, one can assume that the high sensitivity of 3g to PETN lies in the possibility of an efficient photon-induced electron transfer (PET) from the LUMO of 3g to the LUMO of PETN, which leads to the non-radiative decay of the exited state of the sensor. This process becomes possible if the LUMO of the sensor is much higher in energy. For the evaluation of the PETN quenching mechanism, quantum chemical calculations were carried out based on the B3LYP/def2-TZVP//PM6 level of theory with the help of the Gaussian-09 [72].
In the case of the chemosensor/fluorophore 3g, one can suggest a PET-emission quenching mechanism in the presence of PETN (as a quencher) [73][74][75]. Additionally, this suggestion is strongly supported by the fact that the LUMO energy of the sensor 3g (−2.34 eV) is higher compared to the LUMO energy of the PETN quencher (−2.84 eV). Additionally, the calculated energy difference of 0.5 eV is a driving force of the quenching process (Table 2, Figure 4). Next, DFT studies [67][68][69][70][71] were carried out to explain the efficient detection of PETN by the compound 3g. Thus, one can assume that the high sensitivity of 3g to PETN lies in the possibility of an efficient photon-induced electron transfer (PET) from the LUMO of 3g to the LUMO of PETN, which leads to the non-radiative decay of the exited state of the sensor. This process becomes possible if the LUMO of the sensor is much higher in energy. For the evaluation of the PETN quenching mechanism, quantum chemical calculations were carried out based on the B3LYP/def2-TZVP//PM6 level of theory with the help of the Gaussian-09 [72].
In the case of the chemosensor/fluorophore 3g, one can suggest a PET-emission quenching mechanism in the presence of PETN (as a quencher) [73][74][75]. Additionally, this suggestion is strongly supported by the fact that the LUMO energy of the sensor 3g (−2.34 eV) is higher compared to the LUMO energy of the PETN quencher (−2.84 eV). Additionally, the calculated energy difference of 0.5 eV is a driving force of the quenching process (Table 2, Figure 4).
For the NMR spectra of the synthesized fluorophores (3a-g) please see the Supplementary material ( Figures S2-S10).

Photophysical Investigations Materials and Equipment
Acetonitrile and methylene chloride were used to prepare a solution of POPOP azaanalogues in order to study the photophysical properties, purity levels "for HPLC, UV, IR, GPC". Absorption spectra were recorded on a Shimadzu UV-1800 spectrophotometer (Kyoto, Japan). The emission and excitation spectra were recorded on a Horiba-FluoroMax-4 spectrofluorometer (Irvine, CA 92618, USA). Graphical processing of the absorption, emission and excitation spectra was performed using OriginPro 2015 (64-bit) b9.2.196 software. The absolute quantum yields of the photoluminescence of the compounds were obtained using the integrating sphere of the Horiba-Fluoromax-4 instrument (Irvine, CA 92618, USA). Graphical processing of absorption, emission and excitation spectra was carried out using OriginPro 2015 (64-bit) b9.2.196 software; normalization of all electronic spectra was carried out in the Overlay mode automatically using the "Normalize columns" option using the same software. The absolute quantum yields of the photoluminescence of the compounds were obtained using the integrating sphere of the Horiba-Fluoromax-4 instrument (Irvine, CA 92618, USA). The fluorescence lifetime of the compounds was measured on a Horiba FluoroMax-4 instrument (USA) using the TCSPC (Time Correlated Single-Photon Counting) method.
We chose the maxima closest to 350 nm, because this wavelength is the most preferred excitation wavelength when recording the emission spectrum, according to the operating instructions for the Horiba FluoroMax-4 spectrofluorometer ( Figure 5).

Fluorometric Titration
Compound 3g was studied as a chemosensor for "turn-off" fluorescence detection of explosives. The chemosensor fluorescence response to nitroanalytes was quantified using the Stern-Volmer static quenching model. The Stern-Volmer fluorescence quenching constants (Ksv) were calculated according to the static quenching equation as the slope of the intensity graph ((I0/I) −1 ) depending on the concentration of the quencher ([Q]) (Equation (1)): Electron-deficient neutral molecules were chosen as quenchers-one of the most common explosives and their decay products 2,4,6-trinitrotoluene (TNT) and 2,4-

Limit of Detection Calculation
The limit of detection (LOD) was calculated on the basis of the data of fluorome titration experiments according to the method published previously [76]. A calibrat curve was plotted between the fluorescence intensity and the quencher concentration obtain a regression curve equation. The LOD was determined using the Equation (2): where σ is the standard deviation of the fluorophore intensity in the absence of an anal obtained via STEYX function in Excel and k is the slope of the calibration curve.

Time-resolved Fluorescence Measurement
We have also measured the time-resolved fluorescence of all the fluorophores which are summarized in Table 3 and Figures 15-21.

Limit of Detection Calculation
The limit of detection (LOD) was calculated on the basis of the data of fluorometric titration experiments according to the method published previously [76]. A calibration curve was plotted between the fluorescence intensity and the quencher concentration to obtain a regression curve equation. The LOD was determined using the Equation (2): where σ is the standard deviation of the fluorophore intensity in the absence of an analyte obtained via STEYX function in Excel and k is the slope of the calibration curve.

Time-Resolved Fluorescence Measurement
We have also measured the time-resolved fluorescence of all the fluorophores 3, which are summarized in Table 3 and Figures 15-21.

Excitation Spectra
For all compounds, the excitation spectra were additionally measured in CH2Cl2 [10 −5 M] (Figures 22-28), at the emission wavelength. The resulting excitation spectra resemble the absorption spectra of the corresponding compounds. A significant decrease in the intensity of short-wavelength excitation bands is observed, which also correlates with the previously reported data [18].

Excitation Spectra
For all compounds, the excitation spectra were additionally measured in CH 2 Cl 2 [10 −5 M] (Figures 22-28), at the emission wavelength. The resulting excitation spectra resemble the absorption spectra of the corresponding compounds. A significant decrease in the intensity of short-wavelength excitation bands is observed, which also correlates with the previously reported data [18].

Excitation Spectra
For all compounds, the excitation spectra were additionally measured in CH2Cl2 [10 −5 M] (Figures 22-28), at the emission wavelength. The resulting excitation spectra resemble the absorption spectra of the corresponding compounds. A significant decrease in the intensity of short-wavelength excitation bands is observed, which also correlates with the previously reported data [18].

Conclusions
In summary, PAH-based POPOP aza-analogues were successfully prepared by means of Cu(I)-catalyzed click reaction between 2-(4-azidophenyl)-5-Ar-1,3,4-oxadiazole and terminal ethynyl-substituted PAHs. Among all the obtained compounds, pyrenyl-substituted fluorophores, such as 3c,g, exhibited the most promising photophysical properties, such as emission up to 441 nm and quantum yield up to 84%, which were closest to the ones reported for POPOP. In most cases, the introduction of an electron-donating methoxy group in the aromatic moiety of these aza-analogues of POPOP improved their photophysical properties. Among the obtained compounds, only pyrene-substituted fluorophore 3g exhibited a well-pronounced fluorescence "turn-off" response toward several common nitroaromatic explosive components, such as DNT and TNT, with 0.8-1.2 × 10 4 M −1 Stern-Volmer (quenching) constants and an LOD of 182 ppb for TNT. In addition, this compound exhibited an excellent response to the hard-to-detect nitro-aliphatic explosive, PETN, with a 1.4 × 10 4 M −1 Stern-Volmer constant and an LOD of 183 ppb. Possible quenching via the PET mechanism for 3g was suggested, and this was further supported by means of DFT quantum chemical calculations.