Solution Structures of Pyrophthalones, I Structure and Conformation of l,3-indandionato-2(2-pyridinium)-betai'n a 1 H/ 13 C NMR Approach

Transcription

1 Solution Structures of Pyrophthalones, I Structure and Conformation of l,3-indandionato-2(2-pyridinium)-betai'n a 1 H/ 13 C NMR Approach Erhard T. K. Haupt, Heindirk tom Dieck* Institut für Anorganische und Angewandte Chemie der Universität Hamburg, Martin-Luther-King-Platz 6, D-2000 Hamburg 13, FRG and Panayot R. Bontchev Faculty of Chemistry, University of Sofia, 1 Anton Ivanov Avenue. Sofia 1126, Bulgaria Z. Naturforsch.42b,31-36(1987);receivedJune30/September8,1986 'H/ 13 C NMR, Pyrophthalones, Solvent Dependent Structure The complete analysis of the 'H/ 13 C NMR spectra of a-pyrophthalone and related compounds demonstrates that the earlier static planar description of the molecules is invalid for polar solvents, and here the stability of any intramolecular hydrogen bond is small. Introduction D[5,6] The structure of a-pyrophthalone (a-pph; l,3-indandionato-2(2-pyridinium)-betain) is a matter of controversy since the first synthesis of the compound [1 4], The tautomeric and mesomeric structures shown in Scheme 1 have been used to explain the numerous analytical data from UV [5 9], IR [7, 9, 10], EPR [9, 11], X-ray [12] and NMR spectroscopy [10, 12, 13]. The same problem exists for the dye-stuff quinophthalone (QPH) [14, 15], which is included for comparison. Although charge separation and protonation of the nitrogen has been mentioned sometimes [3, 5, 6, 11, 12, 14, 15], many papers stated a planar structure and inferred its stabilization by hydrogen bonding between O and N. Our contribution demonstrates that their description as rigid planar molecules is incorrect for polar solvents. In fact, the system is very sensitive to the nature of the solvent [9], This is proven by analysing the *H and 13 C NMR spectra of a-pph (I), a-qph (II) and N-methyl-a-PPH (III) in different solvents (Scheme 2). Scheme 1. H 112 * Reprint requests to Prof. H. tom Dieck. Verlag der Zeitschrift für Naturforschung, D-7400 Tübingen /87/ /$ 01.00/0 Scheme 2. R = H (I) R = CH (Hl Dieses Werk wurde im Jahr 2013 vom Verlag Zeitschrift für Naturforschung in Zusammenarbeit mit der Max-Planck-Gesellschaft zur Förderung der Wissenschaften e.v. digitalisiert und unter folgender Lizenz veröffentlicht: Creative Commons Namensnennung 4.0 Lizenz. This work has been digitalized and published in 2013 by Verlag Zeitschrift für Naturforschung in cooperation with the Max Planck Society for the Advancement of Science under a Creative Commons Attribution 4.0 International License.

2 32 E. T. K. Haupt et al. Solution Structures of Pyrophthalones Materials and Methods coupling of the acid proton, which resonates at 14, to H 6 (7 ~6.5 Hz) and H 3, respectively, gives a clear indication that protonation takes place at the nitrogen and yields an undoubtful entrance to the pyridine proton system. In QPH, the 4 7( ) coupling constant allows the differentiation in the AB-system, where the latter also has a long range coupling to H 8 of the annellated benzene ring, which can be analyzed completely with this information. Thus, complete 13C assignments are available via H,C-COSY experiments [16]. The dipolar structure of the molecule, resulting from the protonation of the nitrogen, is confirmed by the chemical shift of the negatively polarized C2' at [20], The most remarkable result is observed in the aromatic part of the spectra in DMSO (or other polar solvents like C D 3 O D, HMPA, Acetone etc.). They yield a symmetrical AA'BB'-pattern (Fig. 1) or three carbon signals, respectively (Table II), which is not consistent with a rigid planar structure of the molecule. However, in CDC13 (and CD2C12) solutions the expected unsymmetrical proton pattern was obtained as well as the corresponding number of 13C signals (Table II). The splitting in the 13C NMR spectrum is exactly observed on those lines which are attributed to the aromatic part of the DMSO spectrum and the Ad values are dependent on the distance from C 2 ', which even allows the aromatic proton assignment via H,C-COSY. Exactly the same behaviour is observed for QPH. This is the first direct evidence for the planarity of these compounds in sol- The synthesis of the compounds I III was performed according to the literature [2, 3]. The NMR spectra are recorded on a Bruker AM-360 spectrometer (8.45 T) at room temperature. Chemical shifts are referred to the solvent signals (chloroform = 7.25/77.0 ; DMSO = 2.6/43.5 ). ] H NMR spectra are recorded with 32 k data points and zerofilling prior to Fourier transformation. The H,HCOSY spectra (COSY-45) [16] are performed using standard Bruker software, typically with 128 increments in tj and 512 W data points in t 2. In ti zerofilling to 256 W data points was performed. The spectral width was carefully adjusted to the region of interest, avoiding solvent signal overlap (DMSO). The 90 -pulse width was adjusted on the actual sample using the fast procedure described elsewhere [17]. All proton NMR data result from calculated spectra, iterated convergently to the experimental data with the manufacturer's program [18]. Carbon multiplicities are determined via A P T spectroscopy [19] optimized for/ C H = 160 Hz and the assignments are based on H,C-COSY experiments [16]. Results and Discussion The *H/13C NMR data of I III in DMSO and CDCI3 are summarized in Tables I and II. A representative spectrum is shown in Fig. 1. The 'H NMR assignments are based on chemical shifts ( H 5 = signal at highest field), coupling constants (J56 = 4-6 Hz; JM = 8 Hz), H,H-COSY spectra [16] and iterative spin simulations [18]. In PPH the meas. "111 calc. / 11: Fig. 1. Representative a-pph (I) in DMSO. spectrum of

3 Table I. 'H NMR chemical shifts () and coupling constants (Hz) for I III in CDCl 3 /DMSO (0 = phenyl; the d-data for H7' in II (CDC1 3 ) are taken from the H,C-COSY correlation, while the couplings are only estimated; because of strong signal overlap, a successful simulation was impossible). a-pph (I), CDCl, H6 H7' a-pph (I), DMSO H6 H7' a-qph (II), CDC1, - H8 H7' 0 H60 H70 H ? H60 H70 a-qph (II), DMSO - H8 H7' 0 H60 H70 H n H60 H70 N-Methyl-PPH (III), CDC1 3 NCH 3 H6 H7' NCH, N-Methyl-PPH (III), DMSO NCH 3 H6 H7' NC u Tit'

4 34 E.T. K. Haupt et al. Solution Structures of Pyrophthalones Compound a-pph a-pph a-qph a-qph Solvent CDC1 3 DMSO CDClj DMSO Carbon 1'3' ' '7' (4') (4') (7') (7') 5'6' (6') (6') (5') (5') 8'9' NMe NMePPH CDClj NMePPH DMSO Table II. 13 C NMR chemical shifts () for I III in CDCfV DMSO (0 = phenyl). ution, since in earlier publications the aromatic part was not analyzed because of its complexity in low field NMR spectra and the 13 C data were not available. As the chemical shift differences of 7H7' and 7, respectively, are rather small, the lower symmetry of the system is best demonstrated in a H,C-COSY spectrum (Fig. 2). In contrast, methylation of the nitrogen yields symmetrical spectra for III independent of the solvent. The different types of spectra obtained show that the average molecular symmetry changes when going from low polar solvents to polar ones, or when a bulky substituent with a large steric hindrance is introduced. It seems reasonable, that the chemical shift of can serve as an indicator for the stereochemical behaviour of the molecule because it is either near the oxygen or above/below the negatively polarized C2'. Starting with III, we assume from steric reasons a perpendicular conformation of the pyridinium ring, and thus the chemical shift of at <8.4 reflects this situation independent of the solvent. In the halogenated hydrocarbon solvents, removal of the 7.0 l.u ft 0 o 4' 3 r 5 -<E 5 lb i Fig. 2. H.C-COSY spectrum of a-pph (I) in CDClj.

5 35 E.T. K. Haupt et al. Solution Structures of Pyrophthalones a-pph 0 tcdcy N-Me-PPH U 6 A a-pph [DMSO] N-Me-PPH Fig. 3. "H chemical shift correlation for the pyridinyl protons for I and III in CDC1, and DMSO. methyl-group shifts the proton to >8.4 (Fig. 3). Following from the dissymmetry in the spectrum we must assume a planar or nearly planar conformation so that this chemical shift indicates the alternative situation. The location of near the oxygen is expressed by a relatively small Ad value. This may have its reason in molecular deformations already found by X-ray analysis [12], which all tend to separate from the neighboured oxygen atom. Back to DMSO as representative of polar solvents, the symmetry of the spectra indicates "perpendicularity", while the chemical shift of (>8.4 ) indicates planarity. Thus, the only explanation for this phenomenon is a rapid exchange with respect to the NMR time scale: This yields the symmetrical spectra, but the chemical shift of is still the same as in the planar situation due to the exchange between the two equivalent positions, so that no "mid-weighted" chemical shift can be observed. Any attempt to freeze a planar conformation in a polar solvent (acetone, 80 C) failed leading to the assumption that the barrier for this process is very low. It may be of interest that the same behaviour has been overlooked in the interpretation of the 13 C NMR data of 2-propionyl-l,3-indandione [21]. Conclusion Highfield 1 H/ 13 C NMR spectroscopy demonstrates that a-pph, a-qph and similar compounds exhibit spectra of reduced symmetry in low polar solvents in accordance with a planar conformation. The proton is situated at the nitrogen, which has been also derived recently from 15 N NMR results [22], and the dipolar resonance form is dominating in the ground state. In contrast, the N-methyl compounds adopt a perpendicular conformation and enable us to use 6() as an indicator for this situation. The analysis of the DMSO spectra, representative of polar solvents, indicates a rapid equilibrium between two equivalent planar arrangements. In polar solvents where they are best soluble, the conformation of these compounds is therefore not additionally stabilized by intramolecular hydrogen bonding. By other physical methods mentioned above this solvent dependence has not been worked out. The question whether the planarity of these molecules is due to weak conjugation effects or weak intramolecular hydrogen bonding is still open. The structural information given in this paper, however, has been derived without any use of the concept of hydrogen bonding. This work was supported as part of an official cooperation program between the Universities of Hamburg (FRG) and Sofia (Bulgaria).

6 36 E.T. K. Haupt et al. Solution Structures of Pyrophthalones [1] E. Jacobsen and C. L. Reimer, Ber. Dtsch. Chem. Ges. 16, 513, 1082, 2602 (1883). [2] H. von Huber, Ber. Dtsch. Chem. Ges. 36, 1653 (1903). [3] R. Kuhn and F. Bär, Liebigs Ann. Chem. 516, 155 (1935). [4] D. G. Manly, A. Richardson (Jr.), A. M. Stock, C. H. Tilford, and E. D. Amstutz, J. Org. Chem. 23, 373 (1958). [5] J. Amiel, J. Ploquin, L. Sparfei, G. LeBaut, and R. Floc'h, Bull. Soc. Chim. Fr. 1974, [6] J. Kacens, O. Neilands, and J. Linabergs, Latv. PSR Zinat. Akad. Vestis, Khim. Ser. 1972, 576. [7] J. R. Cook and D. F. Martin, J. Inorg. Nucl. Chem. 26, 571 (1964). [8] R. C. Floc'h, D. G. Leblois, G. Y. LeBaut, P. C. Cadiot, J. Ploquin, and C. J. Claire, J. Chem. Res. (M) 1981, [9] P. R. Bontchev, M. Mitewa, S. Minchev, M. Kashchieva, and D. Mechandjiev, J. Prakt. Chem. 325, 803 (1983). [10] J. Ploquin, L. Sparfei, G. LeBaut, R. Floc'h, and Y. Letourneux, J. Heterocycl. Chem. 17, 961 (1980). [11] V. Kadis, J. Fridmanis, and E. Lavrinovics, Nov. Polyarogr., Tezisy Dokl. Vses. Soveshch. Polyarogr., 6th 1975, 79. [12] A. Kemme, M. Bundule, J. Bleidelis, E. Liepins, E. Lavrinovics, and J. Fridmanis, Khim. Geterotsikl. Soedin. 1978, (Scheme 1: CA: 89, c). [13] J. J. Kacens, O. J. Neilands, and A. V. Burkevitza, Izv. Akad. Nauk SSR, Ser. Khim. 1972, 576. [14] F. Kehrer, P. Niklaus, and B. K. Manukian, Helv. Chim. Acta 50, 2200 (1967). [15] B. K. Manukian, P. Niklaus, and H. Ehrsam, Helv. Chim. Acta 52, 1259 (1969). [16] R. Benn and H. Günther, Angew. Chem. 85, 381 (1983); Angew. Chem., Int. Ed. Engl. 22, 390 (1983). [17] E. Haupt, J. Magn. Reson. 49, 358 (1982). [18] PANIC-Program, Bruker-Software Aspect [19] S. L. Patt and J. N. Shoolery, J. Magn. Reson. 46, 535 (1982). [20] H. O. Kalinowski, S. Berger, and S. Braun, 13 C-NMR- Spektroskopie, Thieme-Verlag, Stuttgart [21] H. Y. Aboul-Enein, A. J. Jado, H. M. El-Fatatry, and E. A. Lotfi, Spectr. Lett. 13, 577 (1980). [22] H. Fritz, Bull. Soc. Chim. Belg. 93, 559 (1984).