I. INTRODUCTION
[1,2,3]Triazolo[1,5-a]pyridines 1 (Jones and Abarca, Reference Jones and Abarca2010) are heterocyclic aromatic scaffolds that have been shown to be extremely efficient for the preparation of 6,6′-disubstituted-2,2′-bipyridines (Jones et al., Reference Jones, Pitman, Lunt, Lythgoe, Abarca, Ballesteros and Elmasnaouy1997), terpyridine-like compounds (Ballesteros-Garrido et al., Reference Ballesteros-Garrido, Abarca, Ballesteros, de Arellano, Leroux, Colobert and García-España2009), and a large amount of 2,6-disubstituted pyridines (Jones and Sliskovic, Reference Jones and Sliskovic1980; Jones et al., Reference Jones, Mouat and Tonkinson1985). Aryl- or heteroaryl-substituted triazolopyridines present intense fluorescence (Abarca et al., Reference Abarca, Aucejo, Ballesteros, Blanco and García-España2006) and coordination properties (Battaglia et al., Reference Battaglia, Carcelli, Ferraro, Mavilla, Pelizzi and Pelizzi1994; Ballesteros et al., Reference Ballesteros, Abarca, Samadi, Server-Carrió and Escrivà1999; Niel et al., Reference Niel, Gaspar, Muñoz, Abarca, Ballesteros and Real2003; Arcís-Castillo et al., Reference Arcís-Castillo, Piñeiro-López, Muñoz, Ballesteros, Abarca and Real2013). These properties have been used in the last years for the preparation of sensors of biologically relevant cations as Zn2+ and anions as cyanide or nitrite (Chadlaoui et al., Reference Chadlaoui, Abarca, Ballesteros, Ramírez de Arellano, Aguilar, Aucejo and García-España2006; Ballesteros-Garrido et al., Reference Ballesteros-Garrido, Delgado-Pinar, Abarca, Ballesteros, Leroux, Colobert, Zaragozá and García-España2012). We are interested in the development of new triazolopyridinic systems containing diazine rings in order to access to fluorescent tridentate ligands. With this idea we have designed the new compounds 3-phenyl-7-(pyrazin-2-yl)-[1,2,3]triazolo[1,5-a]pyridine (2) and 7-(pyridazin-3-yl)-3-pyridin-2-yl-[1,2,3]triazolo[1,5-a]pyridine (3) (Figure 1).
Figure 1. (Color online) Chemical structures and synthesis attempts of compounds 2, 3, and 4.
On the particular behavior of compounds 1, regio-selective lithiation at the seven-position results as the most powerful strategy to access multi-dentate compounds (Jones and Sliskovic, Reference Jones and Sliskovic1982; Abarca et al., Reference Abarca, Ballesteros and Chadlaoui2004). The preparation of compounds 2 and 3 was tried by means of direct metalation with BuLi at C7, followed by trapping with pyrazine or pyridazine. Electron-deficient aromatic nitrogen-based compounds (pyridines, pyridazines, etc.) trend to accept nucleophilic attack on the alpha position to the nitrogen leading to adducts that can be rearomatized by simple oxidation with KMnO4 (Kress, Reference Kress1979).
It is important to remark that in compound 1b (and thus in 3) a ring chain isomerization takes place (Abarca et al., Reference Abarca, Alkorta, Ballesteros, Blanco, Chadlaoui, Elguero and Mojarrad2005). This particular feature allows the preparation of a tridentate compound 4 instead of 3. Once the substituent has been introduced at position 7 (structure 3), the triazole ring opens and recloses back by the nitrogen less electron poor, the one from the pyridine ring. This affords the tridentate structure 3-[6-(pyridazin-3-yl)-pyridin-2-yl]-[1,2,3]triazolo[1,5-a]pyridine (4).
1H and 13C NMR spectroscopic data of compound 2 are consistent with the structure of a 3,7-disubstituted triazolopyridine. In contrast, compound 4 has a structure of a 2,6-disubtituted pyridine. Structure 3 (containing a monosubstituted pyridine) should afford one signal of the hydrogen atom in C6′ position. The common chemical shift and coupling constant for this kind of signals are around 8.5–9.0 ppm and 4–5 Hz. In the spectrum, this signal is not observed. However, signals from a three-substituted triazolopyridine, like a hydrogen in C7 position, are observed. This clearly indicates the presence of rearranged compound 4 (Abarca et al., Reference Abarca, Alkorta, Ballesteros, Blanco, Chadlaoui, Elguero and Mojarrad2005). In order to add more evidences of the proposed structures, X-ray diffraction (XRPD) reveals as the most suitable and powerful technique. In this case, we used powder X-ray diffraction. This technique has provided good results with similar molecules (Adam et al., Reference Adam, Ballesteros-Garrido, Vallcorba, Abarca, Ballesteros, Leroux, Colobert, Amigó and Rius2013).
The crystal structures of 2 and 4 have been solved from X-ray powder diffraction data using a recently developed direct-space (DS) methodology based on local least-squares (LS) minimizations and implemented in the program TALP (Vallcorba et al., Reference Vallcorba, Rius, Frontera and Miravitlles2012b). DS methods (Cerný and Favre-Nicolin, Reference Cerný and Favre-Nicolin2007) are very powerful for solving organic compounds but require their chemical structure to be previously determined or confirmed by other techniques (e.g. NMR). In addition, both compounds contain three aromatic rings of well-known geometry. This information can be introduced to increase the precision of the initial molecular model for DS methods. As a counterpart, these systems usually give rise to powders with marked preferred orientation (PO) affecting the measured powder diffraction data. Even with moderate PO, DS methods can solve crystal structures. However, PO effect should be minimized for a proper refinement of the structures, e.g. by measuring in transmission geometry (Figure 2).
Figure 2. Laboratory powder diffraction data of 4 collected in reflection Bragg–Brentano (gray pattern) and transmission Debye–Scherrer (black pattern) geometries (CuKα 1,2 radiation) with the respective refined crystal structures (top). The Rietveld refinement from reflection geometry data results in slightly distorted atomic rings and a higher residual value (R wp = 0.102; χ = 2.64) than the structure obtained from transmission geometry data (R wp = 0.055; χ = 1.41).
II. EXPERIMENTAL
A. General methods
Melting points were determined on a Kofler heated stage and are uncorrected. NMR spectra were recorded on a Bruker AC300 MHz in CDCl3 as solvent. The system used was high-resolution mass spectrometry (HRMS) electron impact (EI) quadrupole time of flight (QqTOF) 5600 system (Applied Biosystems-MDS Sciex). Mode positive. Conditions: Gas1 35 psi, GS2: 35, CUR: 25, temperature: 450 °C, ion spray voltage: 5500 V and collision energy: 25–35 V. IR spectra were recorded using a Thermoscientific Nicolet FT IR iS10. All reactives used are from commercial sources (Aldrich). 3-Phenyl-[1,2,3]triazolo[1,5-a]pyridine 1a (Boyer and Goebel, Reference Boyer and Goebel1960) and 3-(2-pyridyl)-[1,2,3]triazolo-[1,5-a]pyridine 1b (Battaglia et al., Reference Battaglia, Carcelli, Ferraro, Mavilla, Pelizzi and Pelizzi1994; Abarca et al., Reference Abarca, Ballesteros and Elmasnaouy1998) were prepared as described elsewhere.
B. Synthesis of 3-phenyl-7-(pyrazin-2-yl)-[1,2,3]triazolo[1,5-a]pyridine (2)
At −40 °C, butyllithium (2.2 ml, 1.1 eq) in hexane (1.3 M) was added dropwise to a stirred solution of 3-phenyl-[1,2,3]triazolo[1,5-a]pyridine 1a (0.5 g, 2.56 mmol) in toluene (50 ml). After 30 min a solution of pyrazine (0.62 g, 7.7 mmol, 3 eq) in toluene (5 ml) was added dropwise to the reaction mixture, and was allowed to react for 2 h. Then the solution was allowed to reach room temperature and a solution of KMnO4 (0.4 g, 2.81 mmol, 1.1 eq) in 50 ml of water was added. After 30 min the organic layer was decanted, filtered with celite, and separated. The aqueous layer extracted with dichloromethane (3 × 50 ml). The combined organic layers were dried over sodium sulfate, filtered, and evaporated. The crude was purified by chromatotron (ethyl acetate/hexane from 1:5 to 1:1) affording 3-phenyl-7-(pyrazin-2-yl)-[1,2,3]triazolo[1,5-a]pyridine 2 as a colorless solid (0.42 mg, 61%). Mp: 141–142 °C. 1H NMR (300 MHz, CDCl3): δ = 10.32 (d, J = 0.9 Hz, 1H), 8.78–8.72 (m, 2H), 8.15 (dd, J = 8.8, 0.9 Hz, 1H), 8.02–7.99 (m, 3H), 7.58–7.43 (m, 4H). 13C NMR (75 MHz, CDCl3): δ = 146.3 (CH), 145.4 (C), 145.3 (CH), 144.5 (CH), 138.8 (C), 134.6 (C), 131.9 (C), 131.3 (C), 129.2 (CH), 128.4 (CH), 127.2 (CH), 126.0 (CH), 119.5 (CH), 117.8 (CH). MS (EI): m/z(%) = 245 (54), 218 (13), 191 (13), 167 (100), 139 (3). HRMS for C16H12N5(M+ + 1): 274.1087, found: 274.1086. IR (neat, cm−1): 3108, 3031, 1540, 1458, 1394, 1264, 1127, 1058, 998, 918, 853, 722, 689.
C. Synthesis of 3-[6-(pyridazin-3-yl)-pyridin-2-yl]-[1,2,3]triazolo[1,5-a]pyridine (4)
At −40 °C, butyllithium (2.2 ml, 1.1 eq) in hexane (1.3 M) was added dropwise to a stirred solution of 3-(2-pyridyl)-[1,2,3]triazolo[1,5-a]pyridine 1b (0.5 g, 2.55 mmol) in toluene (50 ml). After 30 min a solution of pyridazine (0.62 g, 7.7 mmol, 3 eq) in toluene (5 ml) was added dropwise to the reaction mixture, and was allowed to react for 2 h. Then the solution was allowed to reach room temperature and a solution of KMnO4 (0.4 g, 2.81 mmol, 1.1 eq) in 50 ml of water was added. After 30 min the organic layer was decanted, filtered with celite, and separated. The aqueous layer was extracted with dichloromethane (3 × 50 ml). The combined organic layers were dried over sodium sulfate, filtered, and evaporated. The crude was purified by chromatotron (ethyl acetate/hexane from 1:5 to 1:1) affording 3-[6-(pyridazin-3-yl)pyridin-2-yl]-[1,2,3]triazolo[1,5-a]pyridine 4 (0.140 mg, 20%). Mp: 232–233 °C (AcOEt). 1H NMR (300 MHz, CDCl3): δ = 9.26 (dd, J = 4.9, 1.7 Hz, 1H), 8.81 (ddd, J = 7.0, 1.0, 1.0 Hz, 1H), 8.69 (ddd, J = 8.9, 1.2, 1.2 Hz, 1H), 8.63 (dd, J = 7.8, 1.0 Hz, 1H), 8.60 (dd, J = 8.5, 1.7 Hz, 1H), 8.49 (dd, J = 7.9, 1.0 Hz, 1H), 8.01 (dd, J = 7.9, 7.9 Hz, 1H), 7.69 (dd, J = 8.6, 4.9 Hz, 1H), 7.45 (ddd, J = 8.9, 6.7, 1.0 Hz, 1H), 7.10 (ddd, J = 6.9, 6.9, 1.3 Hz, 1H). 13C NMR (75 MHz, CDCl3): δ = 159.0 (C), 153.2 (C), 152.0 (C), 151.3 (CH), 138.2 (CH), 137.3 (C), 132.1 (C), 127.2 (CH), 126.8 (CH), 125.7 (CH), 124.4 (CH), 121.9 (CH), 120.8 (CH), 120.3 (CH), 116.0 (CH). MS (EI): m/z (%) = 247 (100), 230 (5), 220 (22), 193 (55), 167 (56), 140 (9). HRMS for C15H11N6 (M+ + 1): 275.1040, found: 275.1037. IR (neat, cm−1): 3086, 3025, 1529, 1438, 1402, 1368, 1113, 1035, 741, 691.
D. Powder X-ray diffraction
Diffraction data of 2 and 4 were collected at room temperature on a PANalytical X'Pert PRO MPD diffractometer (45 kV/40 mA, CuKα 1,2 radiation, PIXcel multistrip detector with an active detection length of 3.347°) in transmission geometry using a convergent beam with a focalizing mirror. The specimens were loaded into a 0.7 mm Lindemann glass capillary and the patterns were measured from 2.0° to 70.0° (2θ) in steps of 0.013° at 700 s per step (six consecutive scans were done).
The diffraction patterns were indexed using DICVOL04 (Boultif and Louër, Reference Boultif and Louër2004) with figures of merit of M 20 = 49.5 and F 20 = 103.8 (0.0044, 44) for 2 and M 20 = 31.0 and F 20 = 71.9 (0.0043, 65) for 4. The whole-pattern matching and intensity extraction were performed with DAjust software (Vallcorba et al., Reference Vallcorba, Rius, Frontera, Peral and Miravitlles2012a) and the intensities were introduced in TALP (Vallcorba et al., Reference Vallcorba, Rius, Frontera and Miravitlles2012b) for structure solution. Finally, the candidate solution underwent a final restrained Rietveld refinement with RIBOLS (Rius, Reference Rius2013) using all the profile points and with C–H distances fixed at 0.93 Å and refined as rigid bodies (CH) in the final refinement cycles. Crystallographic data and refinement details for both compounds are summarized in Table I.
Table I. Crystallographic data and refinement details for 2 and 4.
III. RESULTS AND DISCUSSION
The crystal structure of 4 (Figure 3) is planar, with the highest dihedral angle, 7.0(3)°, between the triazolopyridine (TAP) and the pyridine (PY) rings. This planar arrangement leads to a molecular stacking along a (Figure 4) controlled by two moderate π–π interactions. The first one (A, Figure 4) is between two adjacent triazolo rings (symm. op. −1/2 + x, 1/2 − y, z), showing a distance between the five-membered-ring centroids of 3.573(4) Å, a dihedral angle of 3° between ring planes and angles of 17.8° and 16.2° between the centroid–centroid vector and the normal vector to the ring planes. The second π-stacking system (B, Figure 4) is between the PD rings (symm. op. −1/2 + x, 1/2 − y, z) with a distance between the six-membered-ring centroids of 3.587(4) Å, a dihedral angle of 8° between ring planes and angles of 20.8° and 16.7° between the centroid–centroid vector and the normal vector to the ring planes. Besides these π–π interactions, the N atoms also interact weakly with the neighboring molecules via C–H···N electrostatic interactions (Table II).
Figure 3. (Color online) Crystal structure of 2 and 4 with the atom numbering and ring acronyms: triazolopyridine (TAP), pyridine (PY), pyradizine (PD), pyrazine (PZ), and phenyl (PH) rings.
Figure 4. (Color online) Crystal structure packing of 4 projected along a direction showing the two systems of π–π interactions (A and B) and the three weak C–H···N interactions.
Table II. Summary of C–H···N interactions in crystal structures 2 and 4.
The crystal structure of 2 (Figure 3) is also planar with the exception of the phenyl (PH) ring that is slightly rotated by 32(1)°. The crystal packing can be described as zigzag molecular layers propagating along a (Figure 5) with the rotated PH ring in the twist positions. These layers are stacked along c but in this case there is only a weak π–π interaction between the triazolo ring of the TAP and a PZ ring of a neighbor molecule (symm. op. x, 1/2 − y, 1/2 + z), with a distance between ring centroids of 3.791(3) Å, a dihedral angle of 9.4(3)° between ring planes and angles of 23.6° and 23.5° between the centroid–centroid vector and the normal vector to the ring planes. Weak C–H···N intermolecular electrostatic interactions are also present, all of them occurring between molecules of the same layer and along the b direction (Table II).
Figure 5. (Color online) Crystal structure packing of 2 projected along b showing the zigzag layers of molecules and (bottom) the projection along c of one of the layers showing the weak C–H···N intralayer interactions.
The C–H···N intermolecular electrostatic interactions observed in 2 and 4 are weak but important for the crystal packing of these organic compounds (Thalladi et al., Reference Thalladi, Gehrke and Boese2000). The H···N distances range from 2.48 to 2.73 Å (sum of Van der Waals radius = 2.75 Å) and the C–H···N angles are comprised between 120° and 180° (Table II) as expected for these types of interactions (Mascal, Reference Mascal1998).
In both structures, the six-membered rings that contain N atoms (PD, PZ) can be rotated by 180° keeping the overall geometry of the structure. However, this rotation affects the relative position of two N atoms and two H atoms in each ring. The small difference between the C and N X-ray scattering powers and the small contribution of the two H atoms complicates the determination of the ring conformation from powder diffraction data. In the case of 4, the rotation of the PD ring results in two close H atoms (H13 and H5′) and the ring would have to be slightly rotated to avoid this short contact. A refinement of the structure starting from the 180° rotated PD ring leads to a χ = 1.48, which is slightly higher than the previous one (1.41). In spite of the small difference between these two possible ring orientations, powder diffraction data clearly show a better agreement with the chemically less hindered one. The case of 2 is similar, because a rotation of the PZ ring could be also possible (despite the short distance between H3′ and H6); however, this alternative refinement is also worse compared with the previous one (Δχ = 0.1).
IV. CONCLUSION
The crystal structures of two novel substituted triazolopyridines have been completely determined from laboratory powder X-ray diffraction data. The crystal packing is clearly characterized by π–π interactions in 4 and weak intermolecular C–H···N interactions play an important role in the crystal structure packing of both the compounds. The conformations of the six-membered rings determined from powder data correspond to the ones that are less sterically hindered. However, when coordinated to a metal these conformations change in order to face the N atoms to the metal atom resulting in a small planar distortion of the rings (Ramírez de Arellano et al., Reference Ramírez de Arellano, Escrivà, Gómez-García, Mínguez Espallargas, Ballesteros and Abarca2013).
The DS strategy implemented in TALP can solve crystal structures of organic compounds from powders. TALP can use powder data affected by a certain degree of PO of the sample but, for later refinements, it is better to minimize these orientation effects during the powder data collection (i.e. using transmission geometry).
ACKNOWLEDGEMENTS
The authors thank the Ministerio de Economia y Competitividad (Project Nos. MAT2009-07967, MAT2012-35-35247, Consolider NANOSELECT CSD2007-00041, CONSOLIDER-INGENIO SUPRAMED CSD 2010-00065), the Generalitat de Catalunya (SGR2009), and Generalitat Valenciana (Valencia, Spain) (Project PROMETEO 2011/008) for the financal support. O.V. also acknowledges NANOSELECT for a contract. R.A. thanks Generalitat Valenciana for a doctoral fellowship. The transmission XRPD patterns were collected at the CCiT of University of Barcelona and the realization of HRMS spectra was carried out in SCSIE (University of Valencia).
SUPPLEMENTARY MATERIAL
The supplementary material for this article can be found at http://www.journals.cambridge.org/PDJ
