The structures of two

Research Article Received: 16 April 2013 Revised: 20 May 2013 Accepted: 3 June 2013 Published online in Wiley Online Library: 8 July 2013 (wileyonlinelibrary.com) DOI 10.1002/mrc.3983 The structures of two aldazines: [1,1′-(1E,1′E)hydrazine-1,2-diylidenebis(methan-1-yl-1ylidene)dinaphthalen-2-ol] (Lumogen) and 2,2′(1E,1′E)-hydrazine-1,2-diylidenebis(methan-1yl-1-ylidene)diphenol (salicylaldazine) in the solid state and in solution Artur M. S. Silva,a Vera L. M. Silva,a Rosa M. Claramunt,b Dolores Santa María,b* Marta B. Ferraro,c Felipe Reviriego,d Ibon Alkorta,d** and José Elguerod A combination of NMR spectroscopy and theoretical methods Density functional theory including dispersion corrections (DFT-D) was used to study the structures of Lumogen and salicylaldazine. In the solid state, Lumogen exists as the dihydroxy tautomer 1a (an azine, C¼N–N¼C) as was already known from an X-ray determination. In a deuterated dimethyl sulfoxide solution, another tautomer is observed besides 1a; its structure corresponds to the hydroxy-oxo tautomer 1b (a hydrazone, C¼N–NH–Csp2). In what concerns salicylaldazine, we have observed only the dihydroxy tautomer 2a. Copyright © 2013 John Wiley & Sons, Ltd. Keywords: Lumogen; salicylaldazine; NMR; CPMAS; DFT-D calculations; temperature effects Introduction 530 We have been interested in the structure and configuration of azines of aldehydes (aldazines) and ketones (ketazines), i.e. compounds presenting the >C¼N–N¼C< motive.[1–10] After, we reported the mesogenic properties of some mixtures of azines.[11–14] Recently, we have devoted some papers to study theoretically their properties, particularly the mechanism of isomerization about the C¼N bonds.[15–18] Convinced that azines deserve more attention, we decided to study the cases of Lumogen (1) and salicylaldazine (2). These two compounds have closely related structures (Fig. 1). Lumogen [often written as Liumogen, Lumogen Yellow S, Liumogen LT or Liumogen bright yellow (CAS 2387-03-3)] (1) is a commercial pigment material with fluorescent properties that facilitate its use for wavelength-converting optical coatings. A common application is in UV downconversion, as it absorbs radiation in the UV and re-emits at visible wavelengths. Lumogen has been used to build a white LED.[19] Because of its very high conversion efficiency, Lumogen films are used to increase the quantum efficiency of silicon-based photon detectors such as charge coupled devices (CCD's).[20–22] Lumogen coatings are also routinely applied to commercial CCD's; see, for example, Jobin Yvon Inc data sheets for front illuminated UV sensitive (FI UV) CCD's at http://www.jobinyvon.co.uk/ukdivisions/OSD/ccd_detector.htm or Princeton Instruments information on back illuminated CCD's at http://www.mso.anu.edu.au/observing/detlab/ccdlab/ccd/ ccdchar/coating/unichrm.pdf. To further illustrate this application, Lumogen Yellow coatings have been used in the UV imaging Magn. Reson. Chem. 2013, 51, 530–540 systems of the Hubble Space Telescope[23] and the Cassini–Huygens Spacecraft that reached the Saturnian's system in 2004. The NMR study of Lumogen was never reported. Its crystal structure has been determined thrice (UJOTES, UJOTES01 and UJOTES02)[24] and corresponds to 1a (E,E configuration) (Figs 2 and 3). Lumogen (1) was prepared by reacting hydrazine with 2hydroxynaphthalene-1-carbaldehyde, a commercial compound. Lumogen can exist in three tautomeric forms: the dihydroxy * Correspondence to: Dolores Santa María, Departamento de Química Orgánica y Bio-Orgánica, Facultad de Ciencias, Universidad Nacional de Educación a Distancia (UNED), Senda del Rey 9, E-28040 Madrid, Spain. E-mail: dsanta@ccia. uned.es ** Correspondence to: Ibon Alkorta, Instituto de Química Médica, Centro de Química Orgánica ‘Manuel Lora-Tamayo’, IQM-CSIC, Juan de la Cierva, 3, E-28006 Madrid, Spain. E-mail: [email protected] a Chemistry Department and QOPNA, University of Aveiro, 3810-193 Aveiro, Portugal b Departamento de Química Orgánica y Bio-Orgánica, Facultad de Ciencias, Universidad Nacional de Educación a Distancia (UNED), Senda del Rey 9, E-28040 Madrid, Spain c Departamento de Física, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires and IFIBA, CONICET, Ciudad Universitaria, Pabellón 1, 1428 Buenos Aires, Argentina d Instituto de Química Médica, Centro de Química Orgánica ‘Manuel Lora-Tamayo’, IQM-CSIC, Juan de la Cierva, 3, E-28006 Madrid, Spain Copyright © 2013 John Wiley & Sons, Ltd. The structures of two aldazines: Lumogen and salicylaldazine in the solid state and in solution azine 1a [1,1′-(1E,1′E)-hydrazine-1,2-diylidenebis(methan-1-yl-1ylidene)dinaphthalen-2-ol], the hydroxy-oxo hydrazone 1b [(Z)-1(((E)-2-((2-hydroxynaphthalen-1-yl)methylene)hydrazinyl)methylene) naphthalen-2(1H)-one] and the dioxo hydrazine 1c [(1Z,1′Z)-1,1′(hydrazine-1,2-diylbis(methan-1-yl-1-ylidene))dinaphthalen-2(1H)one] (Fig. 3). Salicylaldazine (salicylaldehyde azine) (2), which exists in the dihydroxy tautomeric form (Figs 4 and 5), has been much studied in photophysics. Its excited-state intramolecular proton transfer from the phenolic hydroxy group to the nitrogen of the methine bond is analogous to the fast enol → keto tautomerization of other 2-hydroxybenzenes.[25] Other salicylaldehyde azines (substituted on the phenyl ring) were studied as fluorophores,[26] for their piezochromism.[27] 2′-Hydroxyacetophenone azine (a methyl group in place of the CH) has been studied by IR and density functional theory (DFT) calculations.[28] The X-ray structure of 2 has been determined (SALIAZ, Fig. 4)[24]; the compound shows polymorphism with one form SALIAZ01, 02, 03 and 06 crystallizing in the P21/n space group with one independent molecule and the other SALIAZ04 and 07 in the P21/c space group with two independent molecules. We have studied the P21/n structure (refer to the Experimental Section). The compound can exist in three tautomeric forms similarly to Lumogen (Fig. 5). For this compound, we will discuss only the solid-state NMR results, where the tautomer present is 2a. On the basis of our previous experience, the E/Z isomerism about the C¼N bond in aldazines derived from aromatic aldehydes such as 1 and 2 can be ruled out because the aryl group always adopts the E configuration. Besides, the barriers are very high (90–100 kJ mol 1), and the populations do not depend on the temperature (vide infra) but are very sensitive to the phase.[29,30] Experimental Section Chemistry Synthesis of 1,1′-[hydrazine-1,2-diylidenebis(methanylylidene)]bis (naphthalen-2-ol) (1) An ethanol solution (10 ml) of 2-hydroxy-1-naphthaldehyde (500 mg, 2.9 mmol) was added to a solution of hydrazinum sulfate (171.5 mg, Figure 1. Structure of Lumogen (1) and salicylaldazine (2). Figure 2. X-Ray crystal structure of UJOTES 1. Figure 5. Tautomers of salicylaldazine (2). Figure 3. Tautomers (proton transfer, PT) of Lumogen (1). 531 Figure 4. Left, X-Ray crystal structure of SALIAZ03 2; right, calculated geometry of 2a. Magn. Reson. Chem. 2013, 51, 530–540 Copyright © 2013 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/mrc A. M. S. Silva et al. 1.32 mmol) in ammonia (171.6 μl). The resulting reaction mixture was stirred at room temperature under nitrogen atmosphere for a few minutes, and then it was heated at 50 °C for 4 h. After half an hour of reaction, we observed the formation of a shiny yellow precipitate. After 4 h at 50 °C, the mixture was cooled to room temperature, and the obtained solid was removed by filtration, washed with water and a small portion of ethanol and then recrystallized in ethanol. The azine was obtained as a shiny yellow solid in 60% yield, mp 309–310 °C (ethanol). Lit. 308 °C (decomp.).[31] Salicylaldazine is a commercial compound. (glycine) = 176.1 ppm] and 15N spectra to 15NH4Cl and then converted to nitromethane scale using the following relationship: δ15N(nitromethane) = δ15N(ammonium chloride) 338.1 ppm. Typical acquisition parameters for 13C CPMAS were as follows: spectral width, 40 kHz; recycle delay, 50 s; acquisition time, 30 ms; contact time, 2 ms; and spin rate, 12 kHz. And for 15N CPMAS, these were as follows: spectral width, 40 kHz; recycle delay, 50 s; acquisition time, 35 ms; contact time, 7 ms; and spin rate, 6 kHz. NMR experiments To check the polymorphism of salicylaldazine, we have compared the power diffraction data generated from the single-crystal structures (SALIAZ02, P21/n, vs SALIAZ07, P21/c) that are very different in the 10° to 40° 2θ region with the experimentally determined powder X-ray diffraction with our sample (Fig. 6). There is no doubt that our sample corresponds to SALIAZ02. 1 H and 13C NMR spectra were recorded on a Bruker 300 (Alfa-Aesar, referencia A10527, A Johnson Mattey Company) [300.13 MHz (1H) and 75.47 MHz (13C)] spectrometer with TMS as internal reference. Unequivocal 1H and 13C assignments were made on the basis of 2D HSQC (1H/13C) and HMBC (delays for one bond and long-range J C/H couplings were optimized for 145 and 7 Hz, respectively) experiments. 13 C (100.73 MHz) and 15N (40.60 MHz) CPMAS NMR spectra have been obtained on a Bruker WB 400 spectrometer at 300 K using a 4-mm DVT probehead and a 4-mm diameter cylindrical zirconia rotor with Kel-F endcaps. Operating conditions involved 2.9 μs 90° 1H pulses and decoupling field strength of 86.2 kHz by two-pulse phase modulation sequence. The non-quaternary suppression (NQS) technique to observe only the quaternary carbon atoms was employed; before the acquisition, the decoupler is switched off for a very short time of 25 μs.[32] 13C spectra were originally referenced to a glycine sample, and then the chemical shifts were recalculated to the Me4Si [for the carbonyl atom δ X-ray powder diffraction Computational details Geometries of the different structures of compound 1 were fully optimized at the B3LYP theoretical level,[33,34] with the 6-311++G (d,p) basis set[35] as implemented in the Gaussian 03 program.[36] Harmonic frequency calculations[37] verified the nature of the stationary points as minima (all real frequencies). 13C and 15N absolute shieldings of compounds 1a–1c have been calculated over the fully optimized geometries within the gauge-including atomic orbitals (GIAO) approximation.[38,39] Polarizable continuum model (PCM) calculations[40] were used as implemented in Gaussian 03. 532 Figure 6. The two polymorphs of salicylaldazine: top left, SALIAZ02, P21/n, one independent molecule; top right, SALIAZ07, P21/c, two independent molecules; bottom, experimental X-ray powder diffraction. wileyonlinelibrary.com/journal/mrc Copyright © 2013 John Wiley & Sons, Ltd. Magn. Reson. Chem. 2013, 51, 530–540 The structures of two aldazines: Lumogen and salicylaldazine in the solid state and in solution The absolute shieldings (σ, ppm) were transformed into chemical shifts (δ, ppm) by means of the following empirical equations. δ1 H ¼ 31:0–0:97σ1 H: ½17Š δ13 C ¼ 175:7–0:963σ13 C: According to the calculations, in gas phase, only tautomer 1a should be present, whereas in dimethyl sulfoxide (DMSO) at 20 °C, 1.8% of tautomer 1b should accompany the major one. Taking into account that there are two identical tautomers, 1b and 1b′, the proportion of the minor isomer should be 3.6%. In neither case, the very unstable tautomer 1c should be observed. Because the single N–N bond isolated both moieties, Lumogen should, in a first approximation, behave as two independent parts, i.e. tautomer 1b should be intermediate in energy relative to tautomers 1a and 1c, 27.0 kJ mol 1 in the gas phase and 20.8 kJ mol 1 in DMSO. Actually, it is more stable by 11.1 mol 1, showing that the tautomerism of one half influence the tautomerism of the other half. The structures corresponding to 1a, 1b and 1c are represented in Fig. 7. The calculated geometry of tautomer 1a is almost identical to that determined by X-ray crystallography (UJOTES 1, Fig. 1). Tautomer 1b is in the way between 1a and 1c, with two identical transition states. (1) ½41Š (2) δ15 N ¼ –152:0–0:946σ15 N: ½41Š (3) These equations were established using a large collection of data and relate the calculated values for the gas phase with experimental values determined in solution and in the solid state [only (2) and (3)].[42] For this reason, GIAO calculations over PCM solvated molecules do not improve considerably the results. However, as we will show further on, in the present case, GIAO/ PCM/MSO explains better certain 13C chemical shifts. Quantum ESPRESSO (QE)[43] was employed to optimize the crystal and the isolated molecule structures, using DFT-D.[44–46] The DFT gauge-including projector augmented wave (GIPAW) method with pseudopotentials to approximate the core electron wavefunction, as implemented in the program Quantum ESPRESSO, is used to predict the complete 13C chemical shift tensors for all carbons and, for both configurations, crystal and isolated molecule, at the same level of theory, PBE.[47,48] The parameters employed to make QE calculations were set to achieve convergence in the Self consistent field (SCF) energy. The details of their selection are the following. (i) The DFT-D pseudopotentials from www.quantum-espresso.org are the normconserving pbe-tm-gipaw and the ultrasoft pbe-rrjkus-gipaw-dc, and both of them were tested for every calculation. (ii) The convergence of the SCF calculations, conv, was varied from 10 7 to 10 12 and set in 10 10. (iii) The energy cutoff for the wavefunction, ecutwfc, was varied between 35 and 95 and set in 65, corresponding to the minimum number of plane waves to achieve SCF energy convergence. (iv) The k points were varied between 1 and 4 in each dimension and set in k = 2, employing again the criteria of reaching the minimum SCF energy. All calculations were performed using version 5.0.1 of QE. Calculations were performed on four-core nodes (Intel I5 processors, 3.0 GHz) with 16 GB RAM. Theoretical part: salicylaldazine – geometries The structure corresponding to 2a is represented in Fig. 4 right; the near identity of both structures is apparent. NMR part: Lumogen – solution and solid-state results In solution, the existence of two structures is observed in 1H as well in 13C NMR in the form of narrow signals. This means that the oxo/hydroxy tautomerization barrier is high in the NMR time scale and that, if tautomer 1b is present, the signals from both halves, the hydroxy and oxo, should be observed. The 1H NMR spectrum of Lumogen 1 at room temperature (20 °C) shows the signals of two structures, in a proportion of 60.9 : 39.1. To obtain some information about these structures and to fully characterize each of them, we acquired 1H NMR spectra at different temperatures. At 80 °C, we observed almost only one structure (81.6%), and we run all the necessary spectra for their characterization. The protonated carbon assignment were mainly based on the HSQC spectrum and confirmed by some long-range connectivities found in the HMBC spectrum. This spectrum allowed the unequivocal assignment of the non-protonated carbons (Fig. 8A). At room temperature (20 °C), the signals of structures 1a and 1b can be seen. The percentage of each of the structures are based on the integrals of the N¼CH (s, 10.00 ppm) and OH (s br, 12.88 ppm) of 1a and of the NHCH (s, 10.82 ppm) and NHCH (brs, 12.01 ppm) of 1b. In the spectrum at 10 °C, the proportion changed to 56.2 : 43.8. At this temperature, we have also unequivocally assigned all the proton and carbon resonances through HSQC and HMBC spectra. We have also carried out a NOESY Results and Discussion Theoretical part: Lumogen – energies and geometries We have reported in Table 1 the free energy results concerning the three tautomers of Lumogen. 1 Table 1. Absolute free energies (hartree), relative free energies (kJ mol ), equilibrium constants at 20 °C and dipole moments (D) for the gas phase and DMSO solution (PCM); B3LYP/6-311++G(d,p) calculations Phase ΔG ΔGrel 1a 1b 1c 1a 1b 1c Gas Gas Gas DMSO DMSO DMSO 1107.94350 1107.93744 1107.92298 1107.95457 1107.95086 1107.93871 0.0 15.9 53.9 0.0 9.7 41.6 Magn. Reson. Chem. 2013, 51, 530–540 Copyright © 2013 John Wiley & Sons, Ltd. K 1 1.46 × 10 — 1 1.84 × 10 — μ 3 2 0.00 1.44 0.46 0.00 2.50 0.65 wileyonlinelibrary.com/journal/mrc 533 Compound A. M. S. Silva et al. 1a 1b 1c TS Figure 7. A view of the optimized geometries of tautomers 1a, 1b and 1c and the Transition state (TS) between 1a and 1b. Table 2. Variation of the populations of both isomers with the temperature; we have defined KT = [1b]/[1a] Figure 8. (A) Main connectivities observed in the HMBC spectrum (80 °C) of the structure 1a; (B) NOE cross peaks observed in the NOESY spectrum of 1a and 1b at 10 °C. 534 spectra to confirm the proton assignments of both structures and also to establish their stereospatial relationships (Fig. 8B). Lumogen is very insoluble at the concentrations required for NMR even for high-field spectrometers. Among the usual deuterated solvents, the only useful one is deuterated DMSO (DMSO-d6). In 1H NMR at room temperature (20 °C), there are the signals corresponding to two tautomers. Then, when the solution was heated, the presence of the minor tautomer almost disappears. On the other hand, decreasing the temperature increases the population of the minor tautomer. The percentages reported in Table 2 correspond to the 1H NMR integrals. Note that the population of the minor isomer decreases when the temperature increases, from 43.8% at 10 °C to 18.4% at 80 °C. When we insert these values into the ΔGT = RT Ln KT equation (R = 8.3145 J mol 1 K 1), the relationship is linear (R2 = 0.999), and using the ΔGT values to obtain ΔHT and ΔST (ΔGT = ΔHT TΔST) leads to ΔHT = (14.8 ± 0.2) kJ mol 1 and ΔST = (54.3 ± 0.8) J K 1. Note that in Table 1, ΔGrel = 9.7 kJ mol 1, a larger difference than the results reported in Table 2. wileyonlinelibrary.com/journal/mrc Temperature Major Minor °C K 1a 1b/1b′ 80 60 40 20 10 353.15 333.15 313.15 293.15 283.15 81.6 76.5 69.3 60.9 56.2 18.4 23.5 30.7 39.1 43.8 KT (minor/ major) 0.225 0.307 0.443 0.642 0.779 Ln KT 1.489 1.180 0.814 0.443 0.249 ΔGT 4.374 3.269 2.120 1.080 0.587 The 1H and 13C chemical shifts vary slightly with the temperature (Tables 3 and 4). In the solid state, the structure of Lumogen as determined by diffraction is 1a (Fig. 1). Its 13C and 15N CPMAS spectra are reported in Figs. 9 and 10. The accidental coincidence of the C2 (q) and the CH at 161.9 ppm was verified recording the NQS: the quaternary carbons appear at 108.3 (C1), 128.6 (C4a), 133.7 (C8a) and 161.9 ppm (C2). Another CPMAS experiment was carried out with contact times of 200 μs instead of 2 ms resulting in a spectrum where the quaternary carbons disappear or are very small. An NQS sequence was used to ascertain that the quaternary carbons are C1 (117.9 ppm) and C2 (160.2 ppm) (Fig. 11). The difference in 15N chemical shifts, 9.4 ppm ( 58.3 ppm, Fig. 10, and 48.9 ppm, Fig. 12), corresponds to calculated values of 63.4 1a and 55.2 ppm 1, i.e. 8.2 ppm. Comparison of calculated and experimental chemical shifts: identification of the tautomers of Lumogen 1 in solution We have carried out 13C GIAO/B3LYP/6-311++G(d,p) calculations of the three tautomers of 1 in the gas phase and in DMSO (PCM) (Table 5). Copyright © 2013 John Wiley & Sons, Ltd. Magn. Reson. Chem. 2013, 51, 530–540 The structures of two aldazines: Lumogen and salicylaldazine in the solid state and in solution Table 3. 1 1 1 H chemical shifts (ppm) and some H– H SSCC (Hz) in DMSO-d6 Temperature (°C) 80° H3 H5 H6 H7 H8 CH XH 10° Major 1a Major 1a Minor 1b Major 1a Minor 1b (81.6%) (60.9%) (39.1%) (56.2%) (43.8%) 7.27, d J = 9.0 8.01, d J = 9.0 7.90, dd J = 8.0, 1.4 7.44, ddd, J = 1.0, 7.0, 8.0 7.62, ddd, J = 1.4. 7.0, 8.5 8.58, d, J = 8.5 9.86, s 12.65, brs, OH H4 20° 7.30, d J = 9.0 8.05, d J = 9.0 7.93, d J = 7.9, 1.2 7.45, dd J = 6.9, 7.9 7.63, ddd J = 1.2. 6.9, 8.5 8.65, d J = 8.5 10.00, s 12.88, brs, OH 7.25, d J = 9.1 8.15, d J = 9.1 7.90, d J = 8.6 7.45 under the major 7.63 under the major 8.93, d J = 8.8 10.82, s 12.01, brs, NH/OH 7.36, d J = 9.1 8.12, d J = 9.1 7.93, d J = 8.0 7.52, m 7.31, J = 9.0 8.21, J = 9.0 7.96, J = 8.2 7.50, 7.69, m d d d m 7.69, m 8.71, d J = 8.9 10.08, s 13.00, brs, OH 9.01, d J = 8.7 10.87, s 12.15, brs, NH/OH 13 Table 4. C chemical shifts (ppm) in DMSO-d6; in the last column, 13 the C CPMAS results are reported Major 1a 80 °C C1 C2 C3 C4 C4a C5 C6 C7 C8 C8a CH C1′ C2′ C3′ C4′ C4a′ C5′ C6′ C7′ C8′ C8a′ CH′ 108.3 159.7 118.5 134.3 128.5 128.5 123.5 127.6 121.3 132.1 160.4 — — — — — — — — — — — Minor 1b CPMAS 10 °C r.t. 108.3 159.7 118.5 134.3 128.5 128.5 123.5 127.6 121.3 132.1 160.4 112.6 193.1 129.1 138.8 129.7 129.1 124.6 127.8 122.5 131.9 164.3 108.3 161.9 120.9 133.7 128.6 128.6 124.6 128.6 120.9 133.7 161.9 correlation coefficient (R2 values, which also is reflected in the error of the parameters), it is important to have an intercept close to 0 and a slope close to 1. We have reported in Table 6 the results we have obtained. i. There is no doubt that the major tautomer is 1a because the agreement is very good; refer to Eqns (4a) and (4b), the GIAODMSO values being a little better (0.99 vs 0.98 and a smaller intercept). The worse point corresponds to the CH, which appears experimentally at 160.4 ppm, whereas the fitted value according to Eqn (4a) is 156.0 ppm and according to Eqn (4b) 157.1 ppm. ii. On the basis of energy calculations and the simple considerations we discussed previously, we consider that the minor isomer is 1b. We have assumed that the signals corresponding to the hydroxy part of tautomer 1b are identical to those of 1a (again, the GIAODMSO values are a little better). An examination of the 13C calculated values shows that this is indeed the case save for the CHs (156.9 and 146.2 ppm) [the same happens for the CHs of the oxo part between 1b and 1c Copyright © 2013 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/mrc 535 To decide between different options concerning the tautomers, we have compared the experimental 13C chemical shifts of Tables 3 and 4 (both DMSO-d6 and CPMAS) with the calculated values of Table 5 (both gas and DMSO-d6) for the three possible tautomers 1a, 1b and 1c. To do this, we have calculated different regression equations to select the better ones. Besides the Magn. Reson. Chem. 2013, 51, 530–540 13 Figure 9. C CPMAS NMR spectrum of Lumogen (spectral width, 40 kHz; recycle delay, 50 s; acquisition time, 30 ms; contact time, 2 ms; scans, 365; and spin rate, 12 kHz). A. M. S. Silva et al. Table 5. 13 C calculated chemical shifts (ppm) Atom 15 Figure 10. N CPMAS NMR spectrum of Lumogen (spectral width, 40 kHz; recycle delay, 50 s; acquisition time, 35 ms; contact time, 7 ms; scans, 7086; and spin rate, 6 kHz). C1 C2 C3 C4 C4a C5 C6 C7 C8 C8a CH C1′ C2′ C3′ C4′ C4a′ C5′ C6′ C7′ C8′ C8a′ CH′ Gas DMSO 1a 1b 1c 1a 1b 1c 108.2 163.7 118.9 134.5 128.0 128.9 123.1 128.8 118.2 134.4 156.9 — — — — — — — — — — — 108.3 162.2 118.9 134.2 128.3 129.3 123.2 128.6 117.4 132.5 146.2 107.4 182.4 127.4 139.6 126.3 129.3 122.8 128.1 116.2 134.5 144.2 — — — — — — — — — — — 106.7 180.6 125.8 138.9 126.6 129.3 122.4 128.7 115.9 133.8 140.2 109.1 163.3 118.8 136.7 128.4 130.2 124.1 129.3 119.0 133.4 158.6 — — — — — — — — — — — 108.6 161.6 118.5 135.4 128.7 129.5 124.2 129.7 118.3 133.3 149.0 107.4 182.5 126.5 140.6 126.8 129.6 123.2 130.0 116.7 135.3 147.2 — — — — — — — — — — — 106.5 181.4 125.4 141.0 127.2 130.0 123.3 129.4 117.3 134.7 142.2 13 Figure 11. C CPMAS NMR spectrum of salicylaldazine (spectral width, 40 kHz; recycle delay, 50 s; acquisition time, 30 ms; contact time, 2 ms; scans, 365; and spin rate, 12 kHz). 15 Figure 12. N CPMAS NMR spectrum of salicylaldazine (spectral width, 40 kHz; recycle delay, 50 s; acquisition time, 35 ms; contact time, 7 ms; scans, 1329; and spin rate, 6 kHz). (144.2 and 140.2 ppm)]. A first regression with all the values of 1a and 1b leads to Eqns (5a) and (5b). 536 An examination of the residuals shows that the largest deviations correspond to the CHs of the bridge of 1b. Introducing a dummy variable for these two 13C signals leads to Eqns (6a) and (6b). The dummies are very important (16.2 and 13.8 ppm, depending on the calculations being gas phase or DMSO, respectively). On the other hand, if we suppose that the signals of the oxo part of 1b belong to 1c, Eqns (7a) and (7b) without dummies and Eqns (8a) and (8b) with dummies are obtained. Considering that the number of points is larger [for instance, there are two CHs in Eqns (6a) and (6b) and only one in Eqns (8a) and (8b)] and that the correlation coefficients are slightly lower, we conclude that the minor isomer is 1b. wileyonlinelibrary.com/journal/mrc iii. It appears that the structure of 1a is as represented with both its O–H···N HBs intact in DMSO. On the other hand, the structure of 1b has probably the weaker N–H···O HBs partially broken in DMSO, explaining the anomalies of the Ar-CH¼N carbon atoms. iv. The data of Fig. 9 (solid state) agree with the values in DMSO-d6 for tautomer 1a and with the calculated values [Eqns (9a) and (9b)]. The worse point corresponds to the CH. In this case, the use of DMSO calculated values does not improve the correlations. v. The calculated 15N chemical shifts are 62.0 ppm for 1a, 201.3 and 93.7 ppm for 1b and 219.4 ppm for 1c. As expected for tautomerism involving N atoms, 15N NMR is the method of choice, but in solution for insoluble compounds, 15N labeling is compulsory. The anomaly of carbons CH and C2 of Lumogen and salicylaldazine We have already encountered the problem of the CH that needs an empirical correction (dummy). The second worse points in the equations of Table 6 correspond to the C2. We will examine in this section the problem of these signals in compounds 1a and 2a (dihydroxy tautomers, the only present in the solid state). The use of PCM–DMSO calculated 13C chemical shifts of 1 partly corrects the anomalies found in the gas phase both in solution and in the solid state between 25% and 30%, but still the agreement is not satisfying. We then checked if the anomalies are found in salicylaldazine (Table 7). We have used the equations devised for Lumogen [Eqns (4a), (4b), (9a) and (9b)] to predict the experimental values and to calculate the residuals. Copyright © 2013 John Wiley & Sons, Ltd. Magn. Reson. Chem. 2013, 51, 530–540 The structures of two aldazines: Lumogen and salicylaldazine in the solid state and in solution Table 6. Different regression equations for Eqn Conditions 4a 4b 5a 5b 6a DMSOmajor DMSOmajor DMSOmaj+min DMSOmaj+min DMSOmaj+min 6b DMSOmaj+min 7a 7b 8a DMSOminor DMSOminor DMSOmaj+min 8b DMSOmaj+min 9a 9b CPMAS CPMAS 13 C NMR chemical shifts (δ, ppm) Method Intercept GIAOgas 1a GIAODMSO 1a GIAOgas 1a + 1b GIAODMSO 1a + 1b GIAOgas 1a + 1b Dummy GIAODMSO 1a + 1b Dummy GIAOgas 1a GIAODMSO 1a GIAOgas 1a + 1b Dummy GIAODMSO 1a + 1b Dummy GIAOgas 1a GIAODMSO 1a (4.2 ± 5.6) (2.9 ± 4.5) (6.4 ± 7.0) (7.4 ± 7.0) (0.5 ± 4.4) (1.4 ± 4.5) (13.8 ± 2.3) (13.8 ± 16.7) (13.2 ± 16.2) (9.3 ± 9.4) (8.3 ± 9.4) (18.9 ± 4.8) (1.2 ± 5.2) (0.3 ± 5.0) 2 Slope Number of points R (0.97 ± 0.04) (0.97 ± 0.03) (1.06 ± 0.05) (1.06 ± 0.05) (1.01 ± 0.03) (16.2 ± 2.3) (1.01 ± 0.03) 11 11 33 33 0.98 0.99 0.93 0.94 33 0.98 33 11 11 0.98 0.90 0.91 11 0.96 11 11 11 0.96 0.99 0.99 (1.1 ± 0.1) (1.1 ± 0.1) (1.09 ± 0.07) (20.2 ± 4.4) (1.08 ± 0.08) (1.00 ± 0.04) (1.00 ± 0.04) Table 7. The problem of the signals of the carbon atoms involved in the HB Mol. Exp. DMSO Exp. CPMAS Calc. gas Calc. DMSO Eqn Residual Fitted 2.9 +4.4 2.0 157.1 2.4 +4.4 163.3 158.6 Difference 2.3 +3.1 1.0 +1.9 3.5 +0.9 2.1 0.0 162.6 156.0 161.7 1a 1a 1a 1a 1a 1a 1a 1a C2 CH C2 CH C2 CH C2 CH 159.7 160.4 159.7 160.4 — — — — — — — — 161.9 161.9 161.9 161.9 163.7 156.9 — 158.6 163.7 156.9 163.3 158.6 — — 163.3 4b — — 9b 9b 4a 4a 4b +3.3 9a 9a 1.4 +3.3 2a 2a 2a 2a 2a 2a 2a 2a C2 CH C2 CH C2 CH C2 CH 159.7 164.6 159.7 164.6 — — — — — — — — 160.2 164.1 160.2 164.1 163.1 162.6 — — 163.1 162.6 — — — — 162.3 164.3 — — 162.3 164.3 4a′ 4a′ 4b′ 4b′ 9a′ 9a′ 9b′ 9b′ Comparison of GIAO/DFT with GIPAW calculations Magn. Reson. Chem. 2013, 51, 530–540 Predicted 162.0 161.5 160.7 162.7 163.7 163.2 162.3 164.3 The statistical treatments of these data are reported in Table 9. Several conclusions can be drawn from Table 9. 1. When the data of 1a and 2a are put together, the anomaly of C2 disappears, other atoms showing larger deviations. 2. The errors concerning the CH of 2a are much lower and are not significantly different from other atoms. 3. In what concerns the CPMAS experimental results, according to the R2 values, GIPAW does not improve the gas-phase calculations. 4. The anomaly concerning the CH of 1a is not corrected by the GIPAW calculations. 5. The DMSO cavity simulates the solid state. This is not surprising because an isolated molecule in a DMSO cavity shows some similarity with a molecule surrounded by other identical molecules. Copyright © 2013 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/mrc 537 To this point, all our calculations were made on an isolated molecule, in the gas phase or surrounded by a continuum with the dielectric properties of DMSO. We want in this part of our paper to compare the CPMAS data of 1a and 2a with the σ calculated values using the GIPAW DFT-D method (refer to the Computational Details Section). The calculations for the crystal structures employing pbe-rrjkus-gipaw-dc are indicated as PBE-crystal-rrjkus and PBE-mol-rrjkus for the configurations of crystal and isolated molecule, respectively. Correspondingly, PBE-crystal-tm and PBE-mol-tm indicate calculations using the norm-conserving pbe-tm-gipaw. As it is indicated in the Computational Details Section, all the systems have been optimized with QE at the same level of theory, pbe-rrjkus-gipaw-dc and pbe-tm-gipaw. The data are reported in Table 8. 164.3 157.5 A. M. S. Silva et al. Table 8. Comparison of different approaches to calculated solid-state absolute shieldings (ppm); experimental values in δ (ppm) Atom Exp. GIAO gas GIAO DMSO PBE-crystal-rrjkus PBE-crystal-tm PBE-mol-rrjkus PBE-mol-tm C1 L C2 L C3 L C4 L C4a L C5 L C6 L C7 L C8 L C8a L CH L C1 S C2 S C3 S C4 S C5 S C6 S CH S 108.3 161.9 120.9 133.7 128.6 128.6 124.6 128.6 120.9 133.7 161.9 117.9 160.2 116.2 132.8 119.4 132.8 164.1 70.08 12.44 59.01 42.80 49.50 48.61 54.60 48.70 59.73 42.94 19.48 60.24 13.04 60.72 43.86 59.70 44.64 13.58 69.20 12.92 59.12 40.49 49.13 47.28 53.60 48.18 58.82 43.91 17.77 59.67 13.91 61.47 42.90 58.42 43.17 11.90 57.89 0.75 46.91 33.24 40.02 38.12 45.52 41.65 49.45 35.16 11.34 49.25 5.87 51.15 36.38 52.05 34.57 2.14 58.18 0.66 47.47 33.39 40.07 38.43 45.76 41.79 49.39 35.12 11.36 44.89 0.52 46.71 32.66 48.32 30.60 3.25 60.13 1.76 51.33 34.35 41.26 41.53 48.19 41.98 50.16 36.38 13.34 49.75 4.20 51.22 36.16 52.32 35.05 1.97 55.67 4.13 47.73 30.87 36.91 38.11 44.30 38.20 46.56 32.32 8.18 45.43 1.21 46.92 32.30 48.58 31.38 3.59 L, Lumogen; S, salicylaldazine. Table 9. Regression analysis (the experimental value of CH L is 161.9 ppm); the data for 17 points correspond to a set where the CH of Lumogen has been removed 2 R Worse atom CH L error 0.990 0.992 0.983 0.965 0.978 0.975 CH L CH L CH L CH L CH L CH L +4.9 +3.9 +6.1 +8.6 +7.7 +7.6 157.0 158.0 155.8 153.3 154.2 154.3 0.994 0.995 0.989 0.978 0.989 0.986 CH S C3 L C1 L C3 S C1 L C1 L +2.7 +2.4 3.3 4.7 2.9 3.8 (173.9 ± 0.9) (173.7 ± 0.8) (165.1 ± 1.0) (161.8 ± 1.3) (164.8 ± 1.0) (160.1 ± 1.0) Method 18 points GIAO gas GIAO DMSO PBE-crystal-rrjkus PBE-crystal-tm PBE-mol-rrjkus PBE-mol-tm 17 points GIAO gas GIAO DMSO PBE-crystal-rrjkus PBE-crystal-tm PBE-mol-rrjkus PBE-mol-tm CH L fitted Intercept (175.4 ± 1.2) (174.9 ± 1.0) (166.6 ± 1.2) (163.7 ± 1.7) (166.6 ± 1.4) (161.7 ± 1.3) Slope (0.95 ± 0.02) (0.95 ± 0.02) (0.96 ± 0.03) (0.92 ± 0.04) (0.93 ± 0.04) (0.90 ± 0.04) (0.92 ± 0.02) (0.93 ± 0.02) (0.92 ± 0.02) (0.88 ± 0.03) (0.89 ± 0.02) (0.86 ± 0.03) 538 Figure 13. Related systems. wileyonlinelibrary.com/journal/mrc Copyright © 2013 John Wiley & Sons, Ltd. Magn. Reson. Chem. 2013, 51, 530–540 The structures of two aldazines: Lumogen and salicylaldazine in the solid state and in solution 6. PCM calculations using water instead of DMSO do not improve the results (all the values of Table 9 for DMSO remains almost unchanged). Conclusions The azines studied in this paper belong to a series of compounds related to the Bertolasi and Gilli's resonance-assisted hydrogen bond concept, i.e. π-bond cooperativity (or synergic interplay between π-delocalization and H-bond strengthening).[49] Some examples are reported in Fig. 13, most of them concerning Schiff bases, which can be viewed as ‘half-azines’. The Schiff base 3 was studied by NMR concluding that only the OH tautomer a was present.[50–52] More recently, Potrzebowski et al. reported the case of two very similar Schiff bases one existing in the conventional phenol-imine tautomer a and the other in the less usual keto-amine tautomer c (there is an OH group ortho to the carbonyl).[53] Compound 4 was also studied by NMR concluding that both tautomers a and c were present.[54] The same conclusion was reported for derivative 5.[55] A series of papers deal with naphthalene derivatives 6 (a Schiff base) and 7 (an azo compound); by using NMR, the proportion of both tautomers was established[51,56] (in the solid state only, tautomer a was present)[57]; electronic spectra were used to determine the thermodynamic parameters. In the case of 5, they were ΔHT = 2.5 kJ mol 1, ΔHT = 9.1 J mol 1 K 1 (ethanol) and ΔHT = 0.8 kJ mol 1, ΔHT = 22.7 J mol 1 K 1 (methylcyclohexane/toluene).[58,59] Note that their ΔHT values have the same sign but lower absolute values than ours ( 14.8 kJ mol 1). On the other hand, our ΔST is much larger (+54.4 J mol 1 K 1) and of the opposite sign to those reported in Fig. 13. The dipolar resonance forms of Fig. 13 could explain the anomalies of the signals of carbon atoms involved in the pseudo six-membered ring (CH, C1 and C2) modifying the position of the proton involved in the HB. This has been demonstrated for Schiff bases by Potrzebowski et al.[53] These authors carried out a series of calculations moving the H along the intramolecular HB finding that its positions is determinant to explain the experimental results. Previous results in this line were reported by Limbach et al.[60,61] All the studies based on NMR (chemical shifts, coupling constants, deuterium-induced chemical shifts differences and others) are based on average signals and interpolation methods that need values for both tautomers usually corresponding to model compounds.[62] Here, for the first time, the signals of both tautomers were observed in solution, and the tautomeric equilibrium constant directly measured. Our results may be useful to help understanding the case of more complex situations such as of the azo-hydrazone tautomerism of C3-symmetric trisazo dyes[63] and other dyes,[64] as well as the effect of protonation,[65,66] complexation with metals[67,68] and excited states,[69,70] over the equilibria discussed in the present paper. Acknowledgements Magn. Reson. Chem. 2013, 51, 530–540 [1] E. Arnal, J. Elguero, R. Jacquier, C. Marzin, J. Wilde. Bull. Soc. Chim. Fr. 1966, 877. [2] J. Elguero, R. Jacquier, C. Marzin. Bull. Soc. Chim. Fr. 1966, 2132. [3] J. Elguero, R. Jacquier, C. Marzin. Bull. Soc. Chim. Fr. 1966, 3005. [4] J. Berthou, J. Elguero, R. Jacquier, C. Marzin, C. Rérat. Compt. Rend. Acad. Sci. (Paris) 1967, 265C, 513. [5] J. Elguero, R. Jacquier, C. Marzin. Bull. Soc. Chim. Fr. 1968, 713. [6] J. Elguero, R. Jacquier, C. Marzin. Bull. Soc. Chim. Fr. 1969, 1367. [7] J. Elguero, R. Jacquier, C. Marzin. Bull. Soc. Chim. Fr. 1969, 1374. [8] V. Tabacik, V. Pellegrin, J. Elguero, R. Jacquier, C. Marzin. J. Mol. Struct. 1971, 8, 173. [9] J. Elguero, C. Marzin, J. Berthou. Bull. Soc. Chim. Fr. 1973, 3303. [10] D. Sanz, M. A. Ponce, R. M. Claramunt, C. Fernández-Castaño, C. Foces-Foces, J. Elguero. J. Phys. Org. Chem. 1999, 12, 455. [11] M. Marcos, E. Meléndez, J. L. Serrano, J. Elguero, R. Phan Tan Luu, D. Mathieu. An. Quim. 1983, 79B, 664. [12] M. Marcos, E. Meléndez, J. L. Serrano, D. Mathieu, R. Phan Tan Luu, J. Elguero. Bull. Soc. Chim. Belg. 1983, 92, 429. [13] J. L. Serrano, M. Marcos, E. Meléndez, C. Albano, S. Wold, J. Elguero. Acta Chem. Scand. 1985, B39, 329. [14] J. Elguero, C. Jaime, M. Marcos, E. Meléndez, F. Sánchez-Ferrando, J. L. Serrano. J. Mol. Struct. (THEOCHEM) 1987, 150, 1. [15] F. Blanco, I. Alkorta, J. Elguero. J. Mol. Struct. (THEOCHEM) 2007, 847, 25. [16] I. Alkorta, F. Blanco, J. Elguero. Arkivoc 2008, vii, 48. [17] A. M. S. Silva, R. M. S. Sousa, M. L. Jimeno, F. Blanco, I. Alkorta, J. Elguero. Magn. Reson. Chem. 2008, 46, 859. [18] F. Blanco, I. Alkorta, J. Elguero. Croat. Chem. Acta 2009, 82, 173. [19] S. C. Allen, A. J. Steckl. Appl. Phys. Lett. 2008, 92, 143309. [20] J. H. MacDonald, A. R. Mahon, L. A. Allers, J. F. Hassard, A. Mainwood, R. J. Ott. Nucl. Instr. Methods Phys. Res. A 1997, 392, 227. [21] P. F. Morrissey, S. R. McCandliss, P. D. Feldman, S. D. Friedman. Appl. Opt. 1994, 33, 2534. [22] K. R. McIntosh, G. Lau, J. N. Cotsell, K. Hanton, D. L. Bätzner, F. Bettiol, B. S. Richards. Prog. Photovolt: Res. Appl. 2009, 17, 191. [23] J. Janesick, Scientific Charge Coupled Devices, SPIE, Bellingham, WA, USA, 2001. [24] (a) Cambridge Structural Database, F. H. Allen. Acta Crystallogr. Sect. B, 2002, 58, 380; (b) F. H. Allen, W. D. S. Motherwell, Acta Crystallogr. Sect. B 2002, 58, 407. (c) CSD version 5.32, updated Feb. 2011. http://www.ccdc.cam.ac.uk [25] H. Görner, S. Khanra, T. Weyhermüller, P. Chaudhuri. J. Phys. Chem. A 2006, 110, 2587. [26] W. Tang, Y. Xiang, A. Tong. J. Org. Chem. 2009, 74, 2163. [27] X. Chen, R. Wei, X. Xiang, Z. Zhou, K. Li, P. Song, A. Tong. J. Phys. Chem. C 2011, 115, 14353. [28] J. Grzegorzek, Z. Mielke, A. Filarowski. J. Mol. Struct. 2010, 976, 371. [29] M. Oki, Applications of Dynamic NMR Spectroscopy to Organic Chemistry, VCH Publishers, Deerfield Beach, USA, 1985, pp. 364–365. [30] S.-I. Liu. J. Comput. Chem. 2009, 30, 2176. [31] T. Hökelek, S. Bílge, Z. Kilic. Anal. Sci. 2006, 22, x115. [32] S. Berger, S. Braun, 200 and More NMR Experiments, Wiley-VCH, Weinheim, 2004. [33] R. G. Parr, W. Yang, Density-Functional Theory of Atoms and Molecules, Oxford Science Publications: Oxford, New York, 1989. [34] L. J. Bartolotti, K. Fluchick Reviews in Computational Chemistry, vol. 7 (eds: K. B. Lipkowitz, D. B. Boyd), VCH Publishers, NewYork, 1996; p. 187. [35] M. J. Frisch, J. A. Pople, R. Krishnam, J. S. Binkley. J. Chem. Phys. 1984, 80, 3265. [36] Gaussian 09, Revision A.1, 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, Jr., J. E. Peralta, 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, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox, Gaussian, Inc., Wallingford CT, 2009. Copyright © 2013 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/mrc 539 Thanks are given to the Ministerio de Economía y Competitividad of Spain (Projects CTQ2012-35513-C02-02 and CTQ2010-16122) and the Comunidad Autónoma de Madrid (Project MADRISOLAR2, ref. S2009/PPQ-1533). Financial support from CONICET and Universidad de Buenos Aires are gratefully acknowledged. We thank Dr Ulises Acuña (Instituto de Química Física ‘Rocasolano’, CSIC, for suggesting the study of Lumogen. References A. M. S. Silva et al. [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] J. W. McIver, A. K. Komornicki. J. Am. Chem. Soc. 1972, 94, 2625. R. Ditchfield. Mol. Phys. 1974, 27, 789. F. London. J. Phys. Radium 1937, 8, 397. J. Tomasi, B. Mennucci, R. Cammi. Chem. Rev. 2005, 105, 2999. F. Blanco, I. Alkorta, J. Elguero. Magn. Reson. Chem. 2007, 45, 797. J. Pinto, V. L. M. Silva, A. M. S. Silva, R. M. Claramunt, D. Sanz, M. C. Torralba, M. R. Torres, F. Reviriego, I. Alkorta, J. Elguero. Magn. Reson. Chem. 2013, 51, 203. P. Giannozzi, S. Baroni, N. Bonini, M. Calandra, R. Car, C. Cavazzoni, D. Ceresoli, G. L. Chiarotti, M. Cococcioni, I. Dabo, A. Dal Corso, S. Fabris, G. Fratesi, S. de Gironcoli, R. Gebauer, U. Gerstmann, C. Gougoussis, A. Kokalj, M. Lazzeri, L. Martin-Samos, N. Marzari, F. Mauri, R. Mazzarello, S. Paolini, A. Pasquarello, L. Paulatto, C. Sbraccia, S. Scandolo, G. Sclauzero, A. P. Seitsonen, A. Smogunov, P. Umari, R. M. Wentzcovitch, J. Phys. Condens. Matter 2009, 21, 395502-19, www-quantum-espresso.org S. Grimme. J. Comput. Chem. 2004, 25, 1463–1473. S. Grimme. J. Comput. Chem. 2006, 27, 1787–1799. S. Grimme, J. Antony, S. Ehrlich, H. Krieg. J. Chem. Phys. 2010, 132, 154104–154123. J. P. Perdew, K. Burke, M. Ernzerhof. Phys. Rev. Lett. 1996, 77, 3865–3868. J. P. Perdew, K. Burke, Y. Wang. Phys. Rev. B 1996, 54, 16533–16539. V. Bertolasi, P. Gilli, V. Ferretti, G. Gilli, K. Vaughan. New J. Chem. 1999, 23, 1261. A. C. Olivieri, R. B. Wilson, I. C. Paul, D. Y. Curtin. J. Am. Chem. Soc. 1989, 111, 5525. S. H. Alarcón, A. C. Olivieri, M. González-Sierra. J. Chem. Soc. Perkin Trans. 1994, 2, 1067. A. R. Katritzky, I. Ghiviriga, P. Leeming, F. Soti. Magn. Reson. Chem. 1996, 34, 518. M. Jaworska, P. B. Hrynczyszyn, M. Welniak, A. Wojtzcak, K. Nowicka, G. Krasinski, W. Ciesielski, M. J. Potrzebowski. J. Phys. Chem. A 2010, 114, 12522. [54] Z. Rozwadowski, E. Majewski, T. Dziembowska, P. E. Hansen. J. Chem. Soc. Perkin Trans. 1999, 2, 2809. [55] L. Racané, Z. Mihalic, H. Ceric, J. Popovic, V. Tralic-Kulenovic. Dyes Pigm. 2013, 96, 672. [56] S. H. Alarcón, A. C. Olivieri, D. Sanz, R. M. Claramunt, J. Elguero. J. Mol. Struct. 2004, 705, 1. [57] S. H. Alarcón, A. C. Olivieri, P. Jonsen. J. Chem. Soc. Perkin Trans. 1993, 2, 1783. [58] H. Joshi, F. S. Kamounah, G. v. d. Zwan, C. Gooijer, L. Antonov. J. Chem. Soc. Perkin Trans. 2001, 2, 2303. [59] W. M. F. Fabian, L. Antonov, D. Nedeltcheva, F. S. Kamounah, P. J. Taylor. J. Phys. Chem. A 2004, 108, 7603. [60] S. Sharif, G. S. Denisov, M. D. Toney, H. H. Limbach. J. Am. Chem. Soc. 2006, 128, 3375. [61] N. S. Golubev, S. N. Smirnov, P. M. Tolstoy, S. Sharif, M. D. Toney, G. S. Denisov, H. H. Limbach. J. Mol. Struct. 2007, 844-845, 319. [62] J. Elguero, C. Marzin, A. R. Katritzky, P. Linda, The Tautomerism of Heterocycles, Academic Press, New York, 1976, pp. 29. [63] H. Y. Lee, X. Song, H. Park, M.-H. Baik, D. Lee. J. Am. Chem. Soc. 2010, 132, 12133. [64] A. S. Özen, P. Doruker, V. Aviyente. J. Phys. Chem. A 2007, 111, 13506. [65] A. Golcu, M. Tumer, H. Demirelli, R. A. Wheatley. Inorg. Chim. Acta 2005, 358, 1785. [66] N. Galic, Z. Cimerman, V. Tomisic. Spectrochim. Acta A 2008, 71, 1274. [67] D. Ruiz-Molina, J. Veciana, K. Wurst, D. N. Hendrickson, C. Rovira. Inorg. Chem. 2000, 39, 617. [68] W. You, H.-Y. Zhu, W. Huang, B. Hu, Y. Fan, X.-Z. You. Dalton Trans. 2010, 39, 7876. [69] P. Fita, E. Luzina, T. Dziembowska, D. Kopec, P. Piatkowski, C. Radzewicz, A. Grabowska. Chem. Phys. Lett. 2005, 416, 305. [70] M. Ziólek, J. Kubicki, A. Maciejewski, P. Naskrecki, A. Grabowska. J. Chem. Phys. 2006, 124, 124518. 540 wileyonlinelibrary.com/journal/mrc Copyright © 2013 John Wiley & Sons, Ltd. Magn. Reson. Chem. 2013, 51, 530–540