Synthesis of Fluorescent 1,7-Dipyridyl-bis-pyrazolo[3,4-b:4',3'-e]pyridines: Design of Reversible Chemosensors for Nanomolar Detection of Cu2.

An efficient access toward novel tridentate ligands based on 1,7-dipyridinyl-substituted bis-pyrazolo[3,4-b:4',3'-e]pyridines (BPs) and their usefulness as fluorescent probes for cation detection is reported. The synthesis proceeds by a three-step sequence starting from 2-chloropyridine (1), all reactions were performed using microwave radiation under solvent-free conditions, and an overall yield of up to 63% was obtained. Photophysical properties of three representative 1,7-dipyridinyl-BPs (PBPs, 6a-6c) substituted at position 4 with different donor (D) or acceptor (A) groups were investigated. Compounds exhibited large Stokes shift in different solvents and strong blue light emission in both solution and solid state, and quantum yields were as high as 88% for some of them; thus, a twisted intramolecular charge transfer (TICT) fluorescence mechanism characteristic of the 1,4,7-triaryl-BPs was confirmed. The 4-phenyl-substituted probe (Ph-PBP, 6b) was used successfully in the detection of some metals (Cu2+, Co2+, Ni2+, and Hg2+) by fluorescence quenching phenomena, which could be reversed in the presence of ethylenediamine. This probe showed a greater sensitivity toward Cu2+ in concentrations as low as 26 nM, and in the process of "on-off-on" for this fluorescent molecular switch, only 1 equiv of the analyte was used.

Introduction

Fused N-heterocycles have attracted considerable attention, owing to their wide biological and physicochemical applications, and hence became an important area of research in synthetic organic chemistry.[1] For instance, pyrazolo[3,4-b]pyridine derivatives are an important system due to their proven utility as bioactive compounds,[2] organic fluorophores,[3] and ligands to coordination complexes.[4] Pyrazolo[3,4-b]pyridines synthesis mainly involves the interaction between N-substituted 5-aminopyrazoles with 1,3-bis-electrophilic reagents (e.g., β-dicarbonyl compounds,[5a] α,β-unsaturated carbonyl compounds,[5b] among others[5c−5f]).[5] Reactions using α,β-unsaturated carbonyl compounds have been successfully applied in forming this precursor in situ from arylaldehydes and active methylene compounds (AMC) by tricomponent reactions.[6,7a] However, sometimes, these reactions proceed without the participation of AMC, obtaining bis-pyrazolo[3,4-b:4′,3′-e]pyridines (BPs) via the pseudo-tricomponent reaction of arylaldehyde with 2 equiv of aminopyrazole,[7] according to the synthetic method reported.[8] BPs have received little biological interest, although they have been attractive for photophysical applications[8c,9] due to their 1,4,7-triaryl-substituted derivatives exhibiting a high fluorescence in both solution and solid state by a typical twisted intramolecular charge transfer (TICT) mechanism (Figure ).[9,10]

Figure 1

(a) Pyrazolo[3,4-b]pyridine and bis-pyrazolo[3,4-b:4′,3′-e]pyridine (BP). (b) Structure of 3,5-dimethyl-1,4,7-triphenyl-BP represented as a molecular rotor.

Fluorescence phenomena that involve charge transfer (CT) have been observed in various N-heterocycles substituted with different electron–donor (D) and electron–acceptor (A) groups;[11] these processes include intramolecular charge transfer (ICT),[11a] metal–ligand charge transfer (MLCT),[11b] and twisted intramolecular charge transfer (TICT).[11c] Phenomena governed by TICT involved molecular conformations and a strong intramolecular CT occurring in the excited state by polar solvent relaxation around the molecule to produce a continuing rotation of electron donor and acceptor around single bonds until it is twisted to about 90°.[12] The equilibration between a relaxed perpendicular conformer and a coplanar conformer regularly results in dual fluorescence, that is, from a high energy band by relaxation of the locally excited (LE) state and from a lower energy band by emission from the TICT state.[12] This phenomenon is very sensitive to D–A efficacy and strength, the microenvironment (e.g., polarity and viscosity of solvent), and/or steric hindrance (e.g., alkyl groups near the D–A junction).[12] In addition, if the structural relaxation of excited states includes more than one bond simultaneously (e.g., polyaryl-substituted molecules), the resulting TICT states may acquire great fluorescence quantum yields. The class of fluorophores based on TICT mechanisms can be called “molecular rotors”, which could vary its fluorescence intensity in sterically restricted environments that limit molecular rotations, for example, viscous media or the presence of metal ions to form complexes.[12,13] Consequently, the design of fluorescent sensors for metal ions based on TICT is an interesting innovative area of research due to their notable structural and photophysical properties of modular design, in addition to advantages offered by all fluorescence-based probes (e.g., simple handling, high sensitivity, and real-time response monitoring).[14]

Fluorescent probes have served as useful tools for selective recognition of metals and thus have been widely exploited to detect biologically or environmentally relevant metal cations, such as Hg2+, Pb2+, Ni2+, Mg2+, Fe2+, Cu2+, Co2+ ions, etc.[15−20] However, the TICT-based fluorescent chemosensors are still very scarce, although it is believed that the well-designed probes should show a very good act.[12,16] Copper is an essential trace element for the activities of enzymes because of its redox-active nature, but at higher concentrations, Cu2+ ions can be highly toxic to the organisms because they can displace other metal ions that act as cofactors in enzyme-catalyzed reactions. The unregulated Cu2+ ions can cause oxidative stress, and their increase in the neuronal cytoplasm may contribute to the etiology of Alzheimer’s or Parkinson’s disease.[17] Cobalt is another vital trace element being part of cobalamin and a few metalloproteins. Co2+ ions can be mildly toxic, and unregulated exposure may cause detrimental effects including heart disease, elevated red blood cells accompanied by increased cells in the bone marrow, vasodilation, and flushing.[18] Nickel is also an essential trace element in biological systems (respiration, biosynthesis, and metabolism) but is an important environmental pollutant; however, a very few fluorescent probes for Ni2+ ions have been reported due to their paramagnetic nature causing fluorescence quenching.[19] On the other hand, mercury is one of the more severe environmental pollutants that is very harmful to humans; specifically, methylmercury, yielded from the microbial biomethylation of Hg2+, is known to cause brain damage and other chronic diseases.[20] Therefore, research studies on rapid and sensitive analysis of the previously described metal ions are much needed.

All these aforementioned aspects, together with our interest in developing efficient protocols for the construction of novel pyrazole derivatives[5d,11a,21] and our recent research about the design of fluorescent probes,[22] have encouraged us to propose the synthesis of 1,7-dipyridinyl-BPs (PBPs) 6a–6e. These compounds are novel tridentate ligands, whose synthesis proceeds by a three-step sequence under microwave starting from 2-chloropyridine (1). Ligands 6a–6e contain a fused tricyclic structure that is dialkyl- and triaryl-substituted and also have three pyridine nitrogen atoms (C=N:) suitably situated to achieve the formation of chelates.[14,17−20,22a,23] Thus, we have hypothesized that PBPs 6a–6e have the necessary features for their use in the detection of metal ions through the fluorescence quenching process based on TICT mechanisms (Scheme b).[8−10,12] In this work, compounds 6a–6e showed a high fluorescence emission in different solvents and large Stokes shifts due to the TICT process being favored. In addition, the presence of a 2-pyridyl group at positions 1 and 7 of the bis-pyrazolopyridinic core has not yet been reported. The 1-(2-pyridyl)pyrazole 4 playing an important role in the achievement of probes 6a–6e has been obtained in poor yield via a two-step sequence under reflux that starts from 2-chloropyridine (1).[24] Likewise, the reaction of arylaldehydes 5 with 2 equiv of the 1-phenylpyrazole 4′ to produce 1,4,7-triaryl-BPs in high yields has been reported (Scheme a).[8] However, the presence of the 2-pyridyl group at position 1 of the starting aminopyrazole (i.e., 4) could decrease its reactivity with 1,3-bis-electrophilic reagents, which would complicate the synthesis of 6a–6e (Scheme ).

Scheme 1

Approach for the Synthesis of PBPs 6a–6e; (a) Reported Synthesis and (b) Proposed Synthesis

Results and Discussion

Synthesis

We started our work by synthesizing 2-hydrazinopyridine (2) through an aromatic nucleophilic substitution (SNAr) reaction between 2-chloropyridine (1) and hydrazine monohydrate,[5f,24a] obtaining the precursor 2 as a yellow-orange solid in 88% yield. Then, the synthesis of 5-amino-3-methyl-1-(2-pyridyl)pyrazole (4) was carried by the cyclocondensation of 2 with 3-amino-2-butenenitrile (3),[5f,24b] thus forming the expected 5-aminopyrazole 4 as yellow crystals in 91% yield. Notably, both reactions proceeded efficiently under solvent-free conditions via microwave-assisted organic synthesis (MAOS) using a specialized reactor, which is novel and useful for us due to the synthetic potential of these precursors.[5−8] Results of these two reaction steps are shown in Scheme .

Scheme 2

Synthesis of 5-Amino-3-methyl-1-(2-pyridyl)pyrazole (4)Reaction conditions: (i) 1 (10.6 mmol) and NH2NH2·H2O (42.4 mmol); (ii) 2 (0.7 mmol) and 3 (0.7 mmol). Both reactions proceeded under solvent-free conditions in microwave.

With the 5-amino-1-(2-pyridyl)pyrazole precursor 4 in hand, we envisaged that the microwave-assisted reaction between 4 and arylaldehydes 5 could be used to synthesize 1,7-di(2-pyridinyl)-bis-pyrazolo[3,4-b:4′,3′-e]pyridines 6a–6e (Scheme b). It is important to note that the precursor 4 could have less reactivity than the 5-amino-1-phenylpyrazole 4′ frequently used in the achievement of this poly-heterocyclic system[8] due to the electron–acceptor nature of the 2-pyridyl group on compound 4.[5f,22b,24a] Likewise, reported reactions under solvent-free microwave conditions were carried out in a Pyrex glass open vessel using a domestic oven (Scheme a).[8a] In this context, we proposed to carry out MAOS using our focused microwave reactor to properly control the reaction conditions.[5d,5f,11a,21,25] Initially, we evaluated the synthesis of the fluorophore 6′ via the reaction of benzaldehyde (5b) with 2 equiv of the freshly synthesized amine 4′(26) to establish initial conditions of the synthesized ligand 6b. Compound 6′ could not be used to chelate metals but can work as a reference in our later photophysical studies. Pleasantly, the reaction under solvent-free microwave conditions at 220 °C for 15 min provided the fluorescent bis-pyrazolopyridine 6′ in 79% yield, but when the reaction was carried out with the pyrazole 4 under the same conditions, a novel nonfluorescent product, 6b1, was obtained (Scheme ).

Scheme 3

Synthesis of bis-Pyrazolopyridine 6′ and the PBP 6b via the Intermediate 6b1aaReaction conditions: (i) 4 (1.0 mmol) and 5 (0.5 mmol) under solvent-free conditions in microwave.

NMR spectrum data and HRMS analysis suggested that 6b1 was resulted from the reaction between two molecules of the amine 4 and one benzaldehyde molecule (5b) with the later loss of one water molecule (see the Experimental Section and Supporting Information).[8a,27] Despite the analysis by NMR that confirms the formation of the intermediate 6b1, its corresponding exact mass was not observed by HRMS. The observed ion corresponds to the exact mass of the other intermediate 6b2, evidencing an 1,4-elimination process under those experimental conditions. The microwave irradiation of 6b1 for 5 additional min at 250 °C led to the formation of the desired fluorescent compound 6b in quantitative yield. Likewise, this product was directly obtained during 15 min of reaction at 250 °C. These findings confirm that the reactions proceed via intermediate 6b1 (and 6b2) that later cyclizes with subsequent loss of ammonia and hydrogen molecules (Scheme ), such as what was proposed in a previous work.[8] Results showed that higher temperatures tend to favor the formation of the product 6b without using any additive or catalyst, which confirms our hypothesis about the lower reactivity toward electrophilic reagents of the aminopyrazole 4 versus 4′ by the presence of a 2-pyridyl group on 4. Besides, we tested the reaction using FeCl3 according to the conditions reported by Yin and co-workers,[8c] but there was no conversion in said conditions or even under microwave at different temperatures (130–200 °C), possibly because of the chelating nature of 4.

Once the optimal conditions to obtain the ligand 6b was achieved, we then examined the scope of this pseudo-tricomponent reaction using a variety of arylaldehydes, 5a–5e, substituted with different donor (D) or acceptor (A) groups. Gratifyingly, we found that the corresponding microwave-assisted reaction afforded the desired bis-pyrazolo[3,4-b]pyridines 6a–6e in good yields (75–81%), and all of them showed strong blue light emission in both solution and solid state (Scheme ). The reaction proceeded with operational simplicity, and almost no loss of efficiency was observed for aldehydes tested, which indicated the low electronic influence of the substituents on the reactivity. The fluorescent ligands 6a–6e were efficiently synthesized in three reaction steps under solvent-free microwave conditions starting from 2-chloropyridine (1) and hydrazine monohydrate. The structures of all synthesized compounds (precursors 2 and 3, the intermediate 6b′, and products 6a–6e/6′) were elucidated by HRMS analysis, 1H spectroscopy, and 13C NMR spectroscopy. The whole carbon skeleton was assigned using 13C NMR spectra, combining with DEPT and two-dimensional 1H and 13C shift correlation HSQC and HMBC experiments (see the Experimental Section and Supporting Information for details).

Scheme 4

Synthesis of PBPs 6a–6e under Solvent-Free Microwave ConditionsaaReaction conditions: 4 (1.00 mmol) and 5a–5e (0.50 mmol) under solvent-free conditions. The photograph was taken using 50 μM solutions in ethanol and in solid state. A hand-held UV lamp under a long wavelength (λ = 365 nm) was used.

Photophysical Properties of 6a–6c and 6′

At this stage of our investigation, we selected the compounds 6a–6c to carry out photophysical studies and thus establish the scope of this type of ligand as fluorescent probes for the detection of metallic cations. These three fluorophores were selected due to the fact that they are substituted at position 4 with different donor (D) or acceptor (A) groups (i.e., 6a (4-An), 6b (Ph), and 6c (4-Py)). Equally, the 1,4,7-triphenyl-BP 6′ was subjected to photophysical studies to have a reference in our research toward the design of fluorescent probes (Figure ). The UV–vis and fluorescence emission spectra of compounds 6a–6c and 6′ were done in toluene (PhMe), dichloromethane (DCM), acetonitrile (ACN), dimethylsulfoxide (DMSO), and ethanol (EtOH) as solvents of different polarities (see Figures S1–S4 and Table S1, Supporting Information). Fluorophores 6a–6c and 6′ exhibited large Stokes shifts (9623–13700 cm–1), strong blue light emission (high quantum yields), and poor solvatochromic shift trend in the different solvents evaluated. However, compounds 6c and 6′ displayed a weak fluorescence intensity in apolar and extremely polar solvents (PhMe and DMSO) and greater Stokes shifts (11454–13700 cm–1), possibly by the highest charge symmetry (lower polarity) of 6c (triPy-BP) and 6′ (triPh-BP) versus 6a and 6b. Ethanol also showed a curious result due to its specific hydrogen bonding interactions with these molecules (e.g., 6a–6c versus 6′ and 6a/6c versus 6b), which could be explained by the nature of their aryl groups. Compound 6′ only possesses phenyl groups that should not be affected by ethanol, while the 1,7-dipyridyl groups of 6a–6c would interact with this protic solvent via hydrogen bonding. In fact, 6′ exhibited its strongest fluorescence intensity in ethanol with a quantum yield (ϕF) of >99%, but this value decreased markedly in polar aprotic solvents and much more in nonpolar solvents. Similarly, the 4-aryl group of compounds 6a (4-Py) and 6c (4-An) would be more affected by ethanol versus 6b (Ph). The highest fluorescence quantum yields of ligands 6a–6c were recorded in DMSO because this solvent is able to stabilize charge separation in the excited state by dipolar interactions. Among the compounds studied (6a–6c and 6′), 6b showed the strongest fluorescence intensity with quantum yields (ϕF) in PhMe, DCM, ACN, DMSO, and EtOH of 39, 74, 67, 88, and 50%, respectively. Ligand 6a presented the lowest fluorescence emission by its lower stability in the excited state, caused by the electron-donating 4-methoxyphenyl group (Figure ). These findings suggest that the fluorescence of 6 and 6′ is very sensitive to the nature of the solvent as well as the effect of their aryl groups, that is, microenvironment effects and the twisting of several bonds in structural relaxation processes.[12]

Figure 2

Structure of photophysically studied fluorophores 6a–6c and 6′.

It is important to note that in the planar bis-pyrazolopyridinic moiety, an ICT phenomenon occurs due to the π-excedent character of pyrazolic ring versus pyridine ring, but its triaryl-derivatives (e.g., 6 and 6′) have a conformation, which is not completely planar in the solid state; in fact, the aryl group at position 4 is almost orthogonal to the heterocyclic core, probably due to some type of steric hindrance caused by methyl groups at positions 3 and 5.[28] Thus, part of energy supplied by photons is at expense in the molecular arrangement that leads to a conformation almost planar in the excited state, which allows a total CT from a donor moiety of the molecule to the acceptor 4-aryl group (Figure ). Consequently, these photophysical results confirm that fluorescence properties of bis-pyrazolopyridines are governed by a TICT mechanism sensitive to sterically restricted environments that limit molecular rotations around aryl–BP bonds.[9,10] The TICT phenomenon is responsible for this type of compounds having high fluorescence quantum yields, and also, ligands 6 can be successfully used as fluorescent probes of metal ions.[9−12,23] Compound 6b was used in further studies of the design of fluorescent probes of metal ions because this ligand displayed the strongest fluorescence intensity. Ethanol was selected as a solvent due to the fact that this is a green solvent, is easily accessible, has high miscibility in water, and is easy to handle; also, it was taken into account that the 4-phenyl group of 6b is not affected by ethanol (Figure ).

Figure 3

Normalized UV–vis absorption (gray line, 10 μM) and fluorescence (blue line, 1 μM) spectrum of 6b in absolute ethanol. The sample was excited at 250 nm.

Fluorescence and UV–Vis Response of Ligand 6b to Metal Ions

Fluorescence spectra of ligand 6b were taken with an excitation wavelength (λexc) of 250 nm in ethanol–water solution (99:1, v/v at pH = 7.14) to observe the response of blue fluorescence emission toward Cu2+ and other metal ions (Na+, K+, Mg2+, Ca2+, Ba2+, Co2+, Cr3+, Fe3+, Ni2+, Zn2+, Al3+, Cd2+, Hg2+, and Pb2+) dissolved in distilled water. Sensing studies were performed at pH = 7.14 because it is close to the physiological pH (about 7–7.4), but since ligands 6a–6e have several pyridine-type nitrogen atoms (C=N), we would expect notorious changes in their fluorescence intensities under pH acid with enhanced blue-shifted emission due to the formation of pyridinium ions.[3b]

Preliminary studies were carried out by adding 10 and 100 equiv of each metal to solutions of 6b at 1 μM (fluorescence) and 10 μM (UV–vis), respectively. Absorption spectra showed no significant changes with any of species evaluated, but fluorescence emission studies exhibited very interesting results since moderate-to-total quenching of blue fluorescence was observed later 1 min upon addition of several of the metal ions added (Figure and Figure S5). Probe 6b showed good selectivity toward some ions of biological and environmental importance, such as Cu2+, Co2+, Ni2+, and Hg2+, although this ligand was more sensitive and selective for Cu2+, and the rank was given as follows: Cu2+ > Co2+ > Ni2+ > Hg2+. In fact, the photograph shown in Figure c was taken using only 1.5 equiv of the respective cation, and in this one, total quenching of blue fluorescence is observed solely with Cu2+. Interestingly, compound 6′ showed no changes when Cu2+ was added due to the fact that this ligand is not tridentate, which proved the usefulness of this 1,4,7-triphenyl-substituted blue fluorophore as a reference in our studies (Figure ). Therefore, we have verified our hypothesis that ligands 6a–6e have the necessary structural features for their use in metal ion detection by the fluorescence quenching process based on the TICT process; however, these ions had almost no effect on the absorption spectra. These ligands could be used in future research studies based on coordination complexes with biological, catalytic, and/or technological applications as well as in studies of acidochromism due to their important electronic properties.

Figure 4

(a) Fluorescence spectra of 6b (1 μM) in absolute ethanol–water solution (99:1, v/v) in the presence of 10 equiv of various metal ions, λexc = 250 nm. (b) Fluorescence intensity response of 6b to Cu2+ and other metal ions. The emission intensity was measured at λem = 412 nm. (c) The photograph was taken using 20 μM solutions of 6b in ethanol and 1.5 equiv of the respective cation. A hand-held UV lamp under a long wavelength (λ = 365 nm) was used.

Given our interesting preliminary results regarding the fluorescence response of compound 6b toward Cu2+, Co2+, Ni2+, and Hg2+, we further studied the reaction using these four metal ions to determine their needed amount to achieve total fluorescence quenching. In this way, we carried out this study by adding 0.2 to 100 equiv of each metal to 1 μM solutions of the fluorescent ligand 6b, and fluorescence emission spectra (λexc = 250 nm) were recorded later 1 min upon addition of the metal ions (Figure and Figure S6). Results of fluorescence intensity of 6b with different concentrations of Cu2+ are shown in Figure . Likewise, sensitivity data of 6b toward each of the four metals, that is, detection and quantitation limits (LOD and LOQ), are summarized in Table (also see Figure S7, Supporting Information). It is important to clarify that sensitivity is different to LOD; the first is only the relationship with the calibration curve (straight) slope, indicating the change in the property versus analyte,[29] while LOD indicates the lowest amount of analyte in a sample that can be detected, but not necessarily quantified, under the stated conditions of the study.[30] These results confirm that ligand 6b is highly sensitive for Cu2+ since, only with addition of 0.2 equiv of metal, the fluorescence intensity decreases to >50% (Figure a). Likewise, detection and quantification limits of 6b for Cu2+ are very low (Table , entry 1), which are 0.026 and 0.086 μM (R2 = 0.998, Figure S7a), respectively, meaning that the chemosensor is able to detect just around 1.6 μg of copper in 1 L of dissolution. Accordingly, this method offers high sensitivity for determination of low concentration of Cu2+, which is an ion of high biological impact.[17] On the other hand, values of LOD and LOQ for Co2+, Ni2+, and Hg2+ also are somewhat low (Table , entries 2–4), which means that, in general, the designed ligands 6a–6e are very efficient for these metals and they would form very stable chelates in each case. The measure was also obtained for the other photophysically studied ligands 6a and 6c, but the differences are negligible and the order of magnitude is the same in each experiment.

Figure 5

(a) Fluorescence spectra of 6b (1 μM) in absolute ethanol–water solution (99:1, v/v) upon addition of increasing equivalents of Cu2+ (0.2–1 equiv, λexc = 250 nm). (b) Reversibility of the fluorescence signal “on–off–on” via the complex 6b–Cu using ethylenediamine (Diam, 0.2–1.0 equiv).

Table 1

Sensitivity, LOD, and LOQ of 6b for M2+ (M: Cu, Co, Ni, and Hg)a

entry	M2+	sensitivity (k)	LOD (μM)	LOQ (μM)
1	Cu2+	29.480	0.026	0.086
2	Co2+	24.092	0.124	0.415
3	Ni2+	12.886	0.157	0.522
4	Hg2+	6.347	0.837	2.792

Detection and quantitation limits (LOD and LOQ, respectively) were obtained via plot fluorescence intensity I versus [M2+], where k is the respective slope. See the Experimental Section and Supporting Information for details.

Notably, the chemosensor 6b was found to bind Cu2+ ions reversibly as tested by reacting with ethylenediamine. The progressive addition of a dissolution of diamine (in ethanol) to a solution of 6b–Cu complex resulted in an enhancement of the fluorescence intensity (Figure b). The enhancement of the fluorescence is a result of the high affinity of ethylenediamine (en) toward Cu2+ (Ka = 4.0 × 1010 L mol–1 at 25 °C),[31] which resulted in the decomplexation of the 6a–Cu chelate and later release of the fluorescent ligand 6b. As expected, the reversibility results using the other active ions (i.e., Co2+, Ni2+, and Hg2+) were also satisfactory (Figure S8).

These experiments were also used to determine equivalents of the necessary analyte (M2+ or ethylenediamine) that should be used in the whole on–off–on process (Table ), where only 1 equiv of the analyte was used in the switch obtained through the complex 6b–Cu (Table , entry 1, and Figure b). Reversibility of the fluorescence signal (“turn on” or “turn off”) is an important feature of metal ion-responsive probes, confirming in this study that the observed fluorescence quenching is due to active ions M2+ and not the result of artifacts.

Table 2

Equivalents of Analyte Used in the “Molecular Switch” Based on 6b–Ma

entry	complex	metal ionb	diaminec
1	6b–Cu2+	1	1
2	6b–Co2+	2	3
3	6b–Ni2+	10	4
4	6b–Hg2+	80	>4

The on–off–on process data are displayed.

Necessary equivalents of M2+ for fluorescence quenching.

Necessary equivalents of ethylenediamine for release of fluorescent ligand 6b (turn on).

Afterward, we examined the scope of obtained 6b–M complexes in ethanolic dissolution toward the design of a novel fluorescent turn-on chemosensor for cyanide based on the most sensitive complex 6b–Cu, according to similar methods for this detection.[32] The cyanide ion (CN) is an excellent ligand for coordination compounds, but it is also an extremely toxic species in the human blood stream that causes cytotoxic hypoxia and cellular asphyxiation due to the formation of a stable complex with cytochrome c oxidase, which is the most significant target of cyanide exposure since its inhibition prevents tissues from using oxygen.[22b,22c,32,33] Thus, the design of fluorescent probes based on metal complexes is of great interest for the sensitive and selective detection of CN.[32] This preliminary study was carried out by adding 10 to 32 equiv of NaCN dissolved in ethanol–water (90:10) to an ethanolic dissolution of 6b–Cu (Figure a), which allowed us to determine that 32 equiv is the needed amount of CN to achieve a total fluorescence turn on. In addition, detection and quantification limits of 6b–Cu for CN were calculated as 0.324 and 1.079 μM (R2 = 0.992, Figure b), respectively. This LOD is well below the level permitted in drinking water according to the U.S. EPA (0.2 ppm),[22b]c and the World Health Organization (1.9 μM).[34] However, it is important to remember that these are preliminary results and, in future works, the selectivity and competition with other anions should be evaluated, especially taking into account that this type of complex (6–Cu) in solution can be easily dissociated in the presence of different anions.

Figure 6

(a) Fluorescence spectra of 6b and 6b–Cu (1 μM) in ethanol–water solution upon addition of increasing concentration of CN (10–32 equiv, λexc = 250 nm). (b) Fluorescence integral area of 6b–Cu (1 μM) with different concentrations of CN (1000–3200 nM).

HRMS-ESI Analysis of 6a–Cu Complex

To evidence the formation process of type 6–Cu complexes, we carried out a high-resolution mass spectrometry (HRMS) analysis using an equimolar mixture between probably the most reactive fluorescent ligand 6a and Cu(NO3)2 in acetonitrile–water (10:1, v/v), which leads to quick fluorescence quenching. The electrospray ionization mass spectrum (HRMS-ESI) was obtained by dilution in acetonitrile of an aliquot of the previously prepared mixture containing the complex 6a–Cu (Figure ). The ion peak at m/z 255.0558, with an intensity of 100%, corresponds to the dication species [6a + Cu2+]2+ (calcd, 255.0546), and the isotopic peak pattern supports the presence of copper.[35] This experiment also allowed us to detect an ion peak at m/z 572.0994, with an intensity of 78%, which is assigned to the monocation species [6a + CuNO3+]+ (calcd, 572.0976); see Figures S9–S11 for details (Supporting Information).

Figure 7

HRMS (ESI+) spectrum for the complex 6a with Cu2+ in acetonitrile along with the isotopic peak pattern (a) observed and (b) calculated.

Plausible Sensing Mechanism of 6b toward Cu2+

In general, the sensing mechanism and binding mode of ligands 6a–6e with the active metal ions M2+ were proposed based on our experimental results (Scheme ). As mentioned above, these ligands exhibited strong blue light emission by a TICT fluorescence mechanism, typical of compounds containing the bis-pyrazolopyridinic core (BP), which is sensitive to environments that limit molecular rotations around aryl–BP bonds of 6a–6e.[8−10] Accordingly, upon addition of some metal ions M2+ to a dissolution of these tridentate ligands, fluorescence quenching is observed as a result of the coordination 6–M, which results in the increase of their structural rigidity and inhibition of the TICT processes. The sensing mechanism of 6b toward Cu2+, a possible ligand–metal charge transfer, and its reversibility process are shown in Figure . It is important to note that the ligand–metal charge transfer is favored by the tridentate nature of 6b also preventing the intramolecular charge transfer in 6b, thus producing fluorescence quenching by a TICT mechanism (Figure ).

Scheme 5

Proposed Sensing Mechanism and Binding Mode of 6a with Cu2+

Conclusions

To sum up, we have developed an efficient and expeditious protocol to achieve a novel series of 1,7-dipyridinylsubstituted bis-pyrazolo[3,4-b:4′,3′-e]pyridines, 6a–6e, in high overall yield (58–63%). Products were obtained in three reaction steps (SNAr and two cyclocondensation processes) under microwave irradiation starting from 2-chloropyridine (1) and in high isolated yields at each step. All synthesized compounds were adequately characterized by spectroscopic and HRMS analysis. Besides the synthetic interest of the novel triaryl-substituted tridentate ligands 6a–6e, their photophysical importance is highlighted due to the fact that they exhibit strong blue light emission and proved to be key fluorophores based on TICT processes by studying photophysical properties of three representative ligands. These ligands were successfully used as turn-off fluorescent probes for metal ion detection since they have three pyridine nitrogen atoms suitably situated to achieve the formation of chelates, where the probe 6b showed a good binding ability to Cu2+ with low detection and quantification limits (26 and 86 nM, respectively, R2 = 0.998). Ligand 6b was found to bind Cu2+ reversibly (on–off–on process) using 1 equiv of ethylenediamine as a decomplexing agent of 6b–Cu. Notably, an ethanolic dissolution of the complex 6b–Cu was preliminary studied as a turn-on fluorescent probe for CN– detection, obtaining detection and quantification limits of 324 and 1079 nM (R2 = 0.992), respectively. Therefore, ligands 6a–6e or complexes 6–M could be used to design novel fluorescent probes in the detection of different relevant species, so we expect to extend our studies in this fascinating field of research.

Experimental Section

General Information

All reagents were purchased from commercial sources and used without further purification, unless otherwise noted. All starting materials were weighed and handled in air at room temperature. The reactions were monitored by TLC visualized by a UV lamp (254 or 365 nm) and/or with ninhydrin in EtOH. Flash chromatography was performed on silica gel (230–400 meshes). All reactions under microwave irradiation were performed using a sealed reaction vessel (10 mL, maximum pressure = 300 psi) containing a Teflon-coated stirring bar (obtained from CEM). Microwave-assisted reactions were performed in a CEM Discover SP focused microwave (ν = 2.45 GHz) reactor, equipped with a built-in pressure measurement sensor and a vertically focused IR temperature sensor; controlled temperature, power, and time settings were used for all reactions. NMR spectra were recorded at 400 MHz (1H) and 100 MHz (13C) at 298 K using tetramethylsilane (0 ppm) as the internal reference. NMR spectroscopic data were recorded in CDCl3 using the residual nondeuteriated signal for 1H NMR and the deuteriated solvent signal for 13C NMR spectroscopy as internal standards. DEPT spectra were used for the assignment of carbon signals. Chemical shifts (δ) are given in parts per million, and coupling constants (J) are given in hertz. The following abbreviations are used for multiplicities: s = singlet, d = doublet, t = triplet, and m = multiplet. Melting points were collected using a capillary melting point apparatus and are uncorrected. High-resolution mass spectra (HRMS) were recorded using a Q-TOF spectrometer via electrospray ionization (ESI). The electronic absorption and fluorescence emission spectra were recorded in quartz cuvettes having a path length of 1 cm. UV–vis and fluorescence measurements were performed at room temperature (20 °C). For fluorescence measurements, both the excitation and emission slit widths were 5 nm. The 3-methyl-1-phenylpyrazol-5-amine (4′)[26] and 4-propoxybenzaldehyde (5d)[36] syntheses were carried out using a known procedure.

Synthesis and Characterization

Synthesis of 2-Hydrazinopyridine 2

A mixture of 2-chloropyridine (1, 1.20 g, 10.57 mmol) and an excess of hydrazine monohydrate (N2H4·H2O, 2.12 g, 42.35 mmol) was irradiated with microwaves at 150 °C (160 W, monitored by an IR temperature sensor) and maintained at this temperature for 15 min in a sealed tube containing a Teflon-coated magnetic stirring bar. After the completion of the reaction, the mixture was cooled to 50 °C by airflow, poured into an aqueous saturated K2CO3 dissolution (6 mL), and extracted with diethyl ether (3 × 15 mL). The organic layer was washed with H2O, dried over anhydrous Na2SO4, and concentrated under reduced pressure to afford the pure product 2 as a yellow-orange solid (1.02 g, 88%). Mp 40–42 °C (lit. 43–45 °C).[24a]1H NMR (400 MHz, CDCl3): δ = 3.76 (br s, 2H), 6.18 (br s, 1H), 6.65 (m, 2H), 7.44 (m, 1H), 8.10 (d, J = 4.0 Hz, 1H) ppm. 13C{1H} NMR (100 MHz, CDCl3): δ = 106.9 (CH), 114.3 (CH), 137.4 (CH), 147.5 (CH), 161.2 (C) ppm. MS (EI) m/z: 109 (M+, 100%), 79 (M-30, 71), 52 (52). These NMR data matched previously reported data.[24a]

Synthesis of 3-Methyl-1-(pyridin-2-yl)-1H-pyrazol-5-amine 4

An equimolar mixture (6.7 mmol) of freshly synthesized 2-hydrazinopyridine (2, 0.73 g) and 3-aminoacrotonitrile (3, 0.55 g) was irradiated with microwaves under solvent-free conditions at 150 °C (160 W, monitored by an IR temperature sensor) and maintained at this temperature for 10 min in a sealed tube containing a Teflon-coated magnetic stirring bar. The resulting reaction mixture was cooled to 50 °C by airflow and directly purified by flash chromatography on silica gel (eluent: CH2Cl2) to give the 5-aminopyrazole 4 as yellow crystals (1.06 g, 91%). Mp 78–79 °C (lit. 90 °C).[24b]1H NMR (400 MHz, CDCl3): δ = 2.21 (s, 3H), 5.32 (s, 1H), 5.92 (br s, 2H), 7.02 (t, J = 6.5 Hz, 1H), 7.73 (t, J = 6.7 Hz, 1H), 7.91 (d, J = 8.4 Hz, 1H), 8.28 (d, J = 4.0 Hz, 1H) ppm. 13C{1H} NMR (100 MHz, CDCl3): δ = 14.0 (CH3), 89.9 (CH), 113.2 (CH), 119.1 (CH), 138.5 (CH), 146.4 (CH), 149.1 (C), 150.9 (C), 154.5 ppm. HRMS (ESI+): calcd for C9H10N4+, 175.0984 [M + H]+; found, 175.0984.

General Procedure for the Synthesis of 1,7-Di(pyridin-2-yl)-bis-pyrazolo[3,4-b:4′,3′-e]pyridines 6a–6e

A mixture of the 5-aminopyrazole 4 (174.1 mg, 1.00 mmol) and the appropriate aryl-aldehyde 2 (0.50 mmol) was subjected to microwave irradiation under solvent-free conditions at 250 °C (260 W, monitored by an IR temperature sensor) and maintained at this temperature for 15 min in a sealed tube containing a Teflon-coated magnetic stirring bar. The reaction mixture was cooled to 50 °C by airflow and directly purified by flash chromatography on silica gel (eluent: CH2Cl2) to give the expected bis-pyrazolopyridines 6a–6e in good yields.

4-(4-Methoxyphenyl)-3,5-dimethyl-1,7-di(pyridin-2-yl)-1,7-dihydrodipyrazolo[3,4-b:4′,3′-e]pyridine 6a

By following the general procedure at 250 °C and maintaining this temperature for 15 min in the reaction with 4-methoxybenzaldehyde (5a, 68.2 mg, 0.50 mmol), the bis-pyrazolopyridine 6a was obtained as a pale yellow solid (179.0 mg, 80%). Mp 215–216 °C. 1H NMR (400 MHz, CDCl3): δ = 2.20 (s, 6H), 3.94 (s, 3H), 7.10 (d, J = 8.4 Hz, 2H), 7.25 (m, 2H), 7.40 (d, J = 8.4 Hz, 2H), 7.91 (t, J = 7.5 Hz, 2H), 8.68 (d, J = 7.8 Hz, 4H) ppm. 13C{1H} NMR (100 MHz, CDCl3): δ = 15.2 (CH3), 55.4 (CH3), 113.6 (CH), 114.5 (C), 115.0 (CH), 120.8 (CH), 125.7 (C), 130.2 (CH), 138.0 (CH), 142.4 (C), 146.4 (C), 149.1 (CH), 150.8 (C), 151.0 (C), 160.3 (C) ppm. HRMS (ESI+): calcd for C26H22N7O+, 448.1886 [M + H]+; found, 448.1874.

3,5-Dimethyl-4-phenyl-1,7-di(pyridin-2-yl)-1,7-dihydrodipyrazolo[3,4-b:4′,3′-e]pyridine 6b

By following the general procedure at 250 °C and maintaining this temperature for 15 min in the reaction with benzaldehyde (5b, 54.2 mg, 0.51 mmol), the bis-pyrazolopyridine 6b was obtained as a pale yellow solid (156.6 mg, 75%). Mp 230–232 °C. 1H NMR (400 MHz, CDCl3): δ = 2.16 (s, 6H), 7.26 (m, 2H), 7.50 (m, 2H), 7.60 (m, 3H), 7.93 (t, J = 7.5 Hz, 2H), 8.70 (d, J = 7.5 Hz, 4H) ppm. 13C{1H} NMR (100 MHz, CDCl3): δ = 15.0 (CH3), 114.2 (C), 115.1 (CH), 120.8 (CH), 128.2 (CH), 128.8 (C), 129.3 (CH), 133.8 (C), 138.1 (CH), 142.2 (C), 146.3 (C), 149.1 (CH), 150.8 (C), 151.1 (C) ppm. HRMS (ESI+): calcd for C25H20N7+, 418.1780 [M + H]+; found, 418.1787.

3,5-Dimethyl-1,7-di(pyridin-2-yl)-4-(pyridin-4-yl)-1,7-dihydrodipyrazolo[3,4-b:4′,3′-e]pyridine 6c

By following the general procedure at 250 °C and maintaining this temperature for 15 min in the reaction with isonicotinaldehyde (5c, 54.0 mg, 0.50 mmol), the bis-pyrazolopyridine 6c was obtained as a white solid (163.2 mg, 78%). Mp 273–275 °C. 1H NMR (400 MHz, CDCl3): δ = 2.17 (s, 6H), 7.28 (m, 2H), 7.49 (d, J = 5.8 Hz, 2H), 7.93 (t, J = 7.2 Hz, 2H), 8.64 (d, J = 8.3 Hz, 2H), 8.70 (d, J = 3.6 Hz, 2H), 8.89 (d, J = 5.9 Hz, 2H) ppm. 13C{1H} NMR (100 MHz, CDCl3): δ = 15.1 (CH3), 113.2 (C), 115.1 (CH), 121.1 (CH), 123.6 (CH), 138.1 (C), 138.2 (CH), 142.3 (C), 145.4 (C), 149.1 (CH), 149.8 (CH), 150.6 (C), 150.8 (C) ppm. HRMS (ESI+): calcd for C24H19N8+, 419.1733 [M + H]+; found, 419.1744.

3,5-Dimethyl-4-(4-propoxyphenyl)-1,7-di(pyridin-2-yl)-1,7-dihydrodipyrazolo[3,4-b:4′,3′-e]pyridine 6d

By following the general procedure at 250 °C and maintaining this temperature for 15 min in the reaction with 4-propoxybenzaldehyde (5d, 82.2 mg, 0.50 mmol), the bis-pyrazolopyridine 6d was obtained as a pale yellow solid (180.7 mg, 76%). Mp 255–257 °C. 1H NMR (400 MHz, CDCl3): δ = 1.11 (t, J = 7.3 Hz, 3H), 1.90 (m, 2H), 2.20 (s, 6H), 4.05 (t, J = 6.5 Hz, 2H), 7.08, J = 8.3 Hz, 2H), 7.24 (m, 2H), 7.38 (d, J = 8.2 Hz, 2H), 7.91 (t, J = 7.6 Hz, 2H), 8.67 (d, J = 7.3 Hz, 4H) ppm. 13C{1H} NMR (100 MHz, CDCl3): δ = 10.5 (CH3), 15.2 (CH3), 22.6 (CH2), 69.7 (CH2), 114.1 (CH), 114.5 (C), 115.0 (CH), 120.7 (CH), 125.5 (C), 130.1 (CH), 138.0 (CH), 142.5 (C), 146.4 (C), 149.0 (CH), 150.8 (C), 151.0 (C), 160.0 (C) ppm. HRMS (ESI+): calcd for C28H26N7O+, 476.2199 [M + H]+; found, 476.2207.

4-(4-Chlorophenyl)-3,5-dimethyl-1,7-di(pyridin-2-yl)-1,7-dihydrodipyrazolo[3,4-b:4′,3′-e]pyridine 6e

By following the general procedure at 250 °C and maintaining this temperature for 15 min in the reaction with 4-chlorobenzaldehyde (5e, 70.3 mg, 0.50 mmol), the bis-pyrazolopyridine 6e was obtained as a white solid (183.1 mg, 81%). Mp 212–214 °C. 1H NMR (400 MHz, CDCl3): δ = 2.19 (s, 6H), 7.26 (m, 2H), 7.45 (d, J = 8.2 Hz, 2H), 7.59 (d, J = 8.2 Hz, 2H), 7.93 (t, J = 7.3 Hz, 2H), 8.69 (m, 4H) ppm. 13C{1H} NMR (100 MHz, CDCl3): δ = 15.2 (CH3), 114.1 (C), 115.1 (CH), 121.0 (CH), 128.6 (CH), 130.2 (CH), 132.3 (C), 135.7 (C), 138.1 (CH), 140.6 (C), 145.9 (C), 149.1 (CH), 150.8 (C), 151.0 (C) ppm. HRMS (ESI+): calcd for C25H19ClN7+, 452.1390 [M + H]+; found, 452.1387.

Obtaining the Intermediate 4,4′-(Phenylmethylene)-bis-(3-methyl-1-(pyridin-2-yl)-1H-pyrazol-5-amine) 6b1

By following the general procedure (see the synthesis of 6a–6e) at 220 °C and maintaining this temperature for 15 min in the reaction with the 5-aminopyrazole 4 (174.2 mg, 1.00 mmol) and benzaldehyde (5b, 53.1 mg, 0.50 mmol), the intermediate 6b1 was obtained as a yellow solid (142.0 mg, 65%). Mp 200–202 °C. 1H NMR (400 MHz, CDCl3): δ = 2.17 (s, 6H), 5.17 (s, 1H), 5.81 (br s, 4H), 7.02 (t, J = 7.4 Hz, 2H), 7.24–7.36 (m, 5H), 7.74 (t, J = 7.5 Hz, 2H), 7.94 (d, J = 8.4 Hz, 2H), 8.22 (d, J = 4.2 Hz, 2H) ppm. 13C{1H} NMR (100 MHz, CDCl3): δ = 12.3 (CH3), 35.4 (CH), 101.1 (C), 113.3 (CH), 119.1 (CH), 126.7 (CH), 128.2 (CH), 128.9 (CH), 138.5 (CH), 140.7 (C), 146.2 (CH), 146.3 (C), 149.9 (C), 154.9 (C) ppm. HRMS (ESI+): calcd for C25H25N8+, 437.2197 [M + H]+ and for C16H15N4+, 263.1291 [M–pyrazole 4 + H]+; found, 263.1295.

Synthesis of 3,5-Dimethyl-1,4,7-triphenyl-1,7-dihydrodipyrazolo[3,4-b:4′,3′-e]pyridine 6′

By following the general procedure at 220 °C and maintaining this temperature for 15 min in the reaction with 3-methyl-1-phenyl-1H-pyrazol-5-amine (4′, 173.3 mg, 1.00 mmol) and benzaldehyde (5b, 53.0 mg, 0.50 mmol), the bis-pyrazolopyridine 6e was obtained as a white solid (164.2 mg, 79%). Mp 239–240 °C (lit. 241–243 °C).[8a]1H NMR (400 MHz, CDCl3): δ = 2.10 (s, 6H), 7.29 (t, J = 7.4 Hz, 2H), 7.48–7.56 (m, 9H), 8.42 (d, J = 7.5 Hz, 4H) ppm. 13C{1H} NMR (100 MHz, CDCl3): δ = 14.8 (CH3), 113.5 (C), 120.3 (CH), 125.1 (CH), 128.1 (CH), 128.8 (CH), 128.9 (CH), 129.0 (CH), 134.3 (C), 139.7 (CH), 141.5 (C), 144.5 (C), 150.5 (C) ppm. HRMS (ESI+): calcd for C27H22N5+, 416.1875 [M + H]+; found, 416.1885. These NMR data matched previously reported data.[8a]

Chemosensor Design

UV–Vis Absorption and Fluorescence Studies

The solvatochromic studies of the compounds 6a–6c and 6′ were carried out from 50 μM stock solutions in PhMe, DCM, ACN, DMSO, and EtOH. UV–vis spectra were recorded at 1 μM, and fluorescence spectra were recorded at 1.0 μM, with λexc = 250 or 280 nm according to maximum absorption wavelength.

Determination of the Relative Quantum Yields

The relative quantum yields of 6a–6c and 6′ were obtained by the comparative method using phenanthrene (ϕF = 0.125 in ethanol at 254 nm) as the reference and calculated according to the following equation[37]where x and st indicate the sample and standard solution, respectively, ϕ is the quantum yield, m is the gradient derived from the linear regression analysis when plotting integrated fluorescence intensity against absorbance, and η is the index of refraction of the solvents.

Response of Ligand 6b to Metal Ions

The 50 μM stock solution of the chemosensor 6b was prepared in ethanol–water (99:1, v/v at pH = 7.14). The salts used in stock solutions were nitrates of each metal ions (K+, Mg2+, Ca2+, Ba2+, Co2+, Cr3+, Fe3+, Ni2+, Cu2+, Zn2+, Al3+, Cd2+, Hg2+, and Pb2+). Inorganic salts were dissolved in distilled water to afford 1 mM aqueous solution. Aliquots of stock solution of 6b were diluted to 5 mL to make the final concentrations of 10 μM for UV–vis and 1 μM for fluorescence. In the selectivity preliminary experiments of 6b to metal ions, the absorption and fluorescence emission spectra (λexc = 250 nm) were recorded upon addition of 10 and 100 equiv of various metal ions. The fluorescence intensities were measured at λem ≈ 412 nm. The fluorescence response in photographs was excitation at 365 nm using a hand-held UV lamp, 6b (20 μM), and 1.5 equiv of the respective cation. The further study of fluorescence response of 6b toward the active ions (Cu2+, Co2+, Ni2+, and Hg2+) was carried out by adding 0.2 to 100 equiv of each metal to 1 μM solutions of 6b.

Determination of Detection and Quantitation Limits (LOD and LOQ)

Detection and quantitation limits of 6b for M2+ (M: Cu, Co, Ni, and Hg) were obtained by 3Sb/k and 10Sb/k, respectively, where k is the slope from the plot fluorescence intensity I versus [M2+], and Sb is the standard deviation of the I intercepts of regression lines.[22,29,30,38]

Reversible Study with Ethylenediamine and Cyanide Ion

Once quenching each dissolution of 6b with the corresponding equivalents of each active metal M2+, a dissolution of ethylenediamine in ethanol–water (99:1) was added, and fluorescence emission spectra (λexc = 250 nm) were recorded upon addition of 0.1 to 10 equiv of diamine. The study with the cyanide ion was carried out using the complex 6b–Cu (λexc = 250 nm) and by adding 1 to 32 equiv of NaCN dissolved in ethanol–water (90:10).