R2 R3 OH O R2 R3 O OH S R2 R3 OH OMe O R2 R3 O O O2N O O NMe2 R1 iv i,ii,iii v,vi,vii viii,ix,x xi (For 1a) 1a–k 4a,i 1l 2a,c,f–h 3a,b,j–k R3 OH 5a,b,d,e CO2Et CH2 COR1 6 xii a bc def NMe2 NEt2 NMe2 NMe2 NEt2 NMe2 NMe2 NMe2 NMe2 HH OMe HHHHH Cl HHH OMe OMe F Cl Br H R1 R2 R3 N[CH2]3CH2 N[CH2]4CH2 NMe2 j k l H H H H NO2 H g hi 414 J. CHEM. RESEARCH (S), 1997 J. Chem. Research (S), 1997, 414–415† Chromone Studies. Part 9.1 Dynamic NMR Analysis of Rotational Isomerism in 2-(N,N-Dialkylamino)chromones† Perry T.Kaye* and Isaiah D. I. Ramaite Department of Chemistry, Rhodes University, P.O. Box 94, Grahamstown, 6140, South Africa Dynamic 1H NMR spectroscopy has been used to explore the influence of substituents on the internal rotation of the amino group in a series of twelve 2-(N,N-dialkylamino)chromones; the rotational barriers (DG‡) in CDCl3 and CD2Cl2 were found to range from 38 to 52 kJ molµ1. In previous dynamic NMR studies of chromone derivatives we have investigated internal rotation in chromone-2-carboxamides2 and chromone-derived acrylamides.1 In these studies it became apparent that A-ring substituents have little effect on the somewhat remote carboxamide and enamine rotors.In the 2-amino derivatives examined here, however, it was anticipated that substituent effects on the electron density at C(2) would be clearly reflected in the relative magnitudes of the C(2)·N rotational barriers. The title compounds 1a–l were synthesised using the established routes outlined in Scheme 1.3–7 The A-ring substituents were chosen to elucidate the influence of electronreleasing and electron-withdrawing substituents on the C(2)·N rotational barriers and, hence, the electron-density at C(2), while symmetrically disubstituted 2-amino groups were selected to facilitate interpretation of the dynamic NMR data.For each of the compounds studied, splitting of the N-alkyl signals was observed (see Fig. 1), coalescence occurring in the temperature range 194–267 K (Table 1).The splitting is attributed to hindered rotation of the dialkylamino moiety about the C(2)·N bond, arising from the delocalisation effects illustrated in Fig. 2(a). Examination of the data in Table 1 reveals several interesting trends. (i) While the variation in the DG‡ values is typically small (DG‡=44.9�1.2 kJ molµ1 for compounds 1a–i,l) relative to the estimated error, the electronic effects of the A-ring substituents are, in fact, discernible.Electron-withdrawing groups enhance nitrogen lone pair delocalisation [Fig. 2(a)] and thus increase the rotational barrier, while electron-releasing groups have the opposite effect [DG‡ for 1l (R2=NO2)a1i (R2=Cl)a1a (R2=H)a1c (R2=OMe) and for 1f (R3=Cl)a1g (R3=F), 1a (R3=H)a1d (R3=OMe).‡ (ii) Changing the N-alkyl substituents from methyl to ethyl is accompanied by an increase in DG‡ (cf. compounds 1a and 1b, and 1d and 1e), reflecting the greater electron-releasing inductive effect of ethyl relative to methyl.(iii) Compared to the ‘parent’ system 1a, the pyrrolidinyl derivative 1j exhibits a significantly higher rotational barrier, while the piperidinyl analogue 1k has a correspondingly lower value for DG‡. In our investigation of N,N-disubstituted chromone-2-carboxamides,2 the N·CO rotational barrier for the pyrrolidinyl derivative was also significantly higher than for the piperidinyl analogue · an observation attributed to the greater ease with which the pyrrolidine nitrogen assumes the planar sp2 arrangement necessary for effective lone-pair delocalisation.The insolubility of the piperidinyl analogue (1k) in CDCl3 necessitated a solvent change (to CD2Cl2); however, this is considered unlikely to affect the magnitude of DG‡ significantly, given the similarity of the data obtained for the pyrrolidinyl analogue 1j in both solvents (DG‡ 52.3�0.1 kJ molµ1). (iv) The rotational barriers measured for the 2-(dimethylamino)chro- *To receive any correspondence (e-mail: chpk@hippo.ru.ac.za).†This is a Short Paper as defined in the Instructions for Authors, Section 5.0 [see J. Chem. Research (S), 1997, Issue 1]; there is therefore no corresponding material in J. Chem. Research (M). ‡Both 6- and 7-substituents (i.e. R2 and R3) are capable of mesomeric interaction with C-2 (see ref. 8). Scheme 1 Reagents and conditions: i, BF3.OEt2, Et2O; ii, Cl2C=N+Me2Clµ, Cl[CH2]2Cl2; iii, MeOH, 50 °C; iv, ButOK, CS2, C6H6; v, K2CO3, EtI, (MeCO)2O; vi, MCPBA, Cl[CH2]2Cl; vii, R2NH (R=Me, Et); viii, LDA, THF; ix, MeCONMe2; x, Tf2O, CH2Cl2; xi, conc.H2SO4–conc. HNO3; xii, EtO2CCH2CONR2 (R=Me, Et), POCl3 Fig. 1 Partial 1H NMR spectra of compound 1i in CDCl3 at selected temperaturesO O N R¢ R¢ R •• O O– N R¢ R¢ R + ( a) O H O R O O N R¢ R¢ R ( b) NMe2 CONMe2 •• ( c) •• J. CHEM. RESEARCH (S), 1997 415 mones (DG‡ ca. 44 kJ molµ1) lie between those determined for 1-(dimethylamino)cyclohexene (24.9 kJ molµ1)9 and for the conjugated enamine, 4-(dimethylamino)but-3-en-2-one (55.9 kJ molµ1).10 The influence of conjugation is particularly evident in the chromone-derived enamines examined previously [see Fig. 2(b)],1 which exhibit unusually high rotational barriers (ca. 67 kJ molµ1). Although the enamine moiety in the title compounds 1 is conjugated with the carbonyl group, nitrogen lone-pair delocalisation is presumably inhibited by competitive delocalisation involving the chromone ether oxygen [see Fig. 2(c)], thus accounting for the somewhat lower rotational barriers observed for these compounds. Experimental The 2-aminochromones required for this study were obtained following the literature methods3–7 outlined in Scheme 1. Compounds 1a, 1b, 1d, 1e, 1h and 1k are known; analytical data for the remaining compounds, which are new and which gave satisfactory NMR (1H and 13C) and MS data, are summarised in Table 2.Variable-temperature 1H NMR spectra were recorded for solutions of the 2-aminochromones 1a–l in CDCl3 or CD2Cl2 on a Bruker AMX 400 NMR spectrometer, equipped with a variable temperature unit, which has been calibrated using 80% ethylene glycol in (CD3)2SO. Temperature stability was judged to be �0.1 K and the overall error in coalescence temperatures (Tc) estimated to be �2 K. Frequency separations at coalescence (Dvc) were obtained by extrapolation as described by Lai and Chen.12 We thank the Deutscher Akademischer Austauschdienst (DAAD) and the Foundation for Research Development (FRD) for bursaries (to I.D. I. R.), and Rhodes University and the FRD for generous financial support. Received, 24th April 1997; Accepted, 23rd July 1997 Paper E/7/02801F References 1 Part 8, P. T. Kaye and I. D. I. Ramaite, J. Chem. Res. (S), 1995, 78. 2 D. N. Davidson and P. T. Kaye, J. Chem. Soc., Perkin Trans. 2, 1991, 927. 3 J. Morris, D. G. Wishkia and Y.Fang, J. Org. Chem., 1992, 57, 6502. 4 J. R. Bantick and J. L. Suschitzky, J. Heterocycl. Chem., 1981, 18, 679. 5 J. A. Morris, G. Donn and Y. Fang, Synth. Commun., 1994, 24, 849. 6 A. Balbi, G. Roma, M. Mazzei and A. Ermili, Farmaco, Ed. Sci., 1983, 38, 784; A. Ermili, A. Balbi, M. Di Braccio and G. Roma, Farmaco, Ed. Sci., 1977, 32, 713. 7 A. Balbi, G. Roma, M. Mazzei and A. Ermili, Farmaco, Ed. Sci., 1983, 38, 784. 8 D. N. Davidson, P. T. Kaye and I. D. I. Ramaite, J. Chem.Res. (S), 1993, 462. 9 J. E. Anderson, D. Casarini and L. Lunazzi, Tetrahedron Lett., 1988, 29, 3141 (Chem. Abstr., 1989, 10, 134396g). 10 J. Dabrowski and L. Kozerski, Org. Magn. Reson., 1973, 5, 469. 11 R. J. Smith, D. H. Williams and K. James, J. Chem. Soc., Chem. Commun., 1989, 682. 12 Y. H. Lai and P. Chen, J. Chem. Soc., Perkin Trans. 2, 1989, 1665. Table 1 Dynamic 1H NMR data for internal rotation of the N,N-dialkylamino moiety in compounds 1a–l Tc/ DvC/ DG‡c/ Compd.R1 R2 R3 Ka Hzb kJ molµ1 1a 1b 1c 1d 1e 1f 1g 1h 1i 1j 1k 1l NMe 2 NEt2 NMe2 NMe2 NEt2 NMe2 NMe2 NMe2 NMe2 l l N[CH2]3CH2 l l N[CH2]4CH2 NMe2 HH OMe H H NO2 HHH OMe OMe Cl F Br HH HH 222 226 219 213 219 223 221 222 225 267 264e 194e 233 81.8 78.3 65.6 43.2 72.0 72.4 73.5 77.7 83.0 142.4d 115.9d 120.3d 95.0 44.2 45.1 44.0 43.5 43.8 44.7 44.2 44.3 44.8 52.4 52.2 37.8 46.1 aCoalescence temperature (�2 K) in CDCl3. bFrequency separation of bands at coalescence; estimated errors R�2.0 Hz.cFree energy of activation for rotation, DG‡=RTc (22.96+ln Tc/Dvc) (see ref. 11); estimated errors �0.5 kJ molµ1. dFor NCH2 signals. eVariable temperature spectra recorded in CD2Cl2. Table 2 Analytical data for new 2-aminochromones 1 Compd. R1 R2 R3 Mp/°C Found Mol. formula Required 1c 1f 1g 1h 1i 1j 1l NMe2 NMe2 NMe2 NMe2 NMe2 l l N[CH2]3CH2 NMe2 OMe HHH Cl NO2 H Cl F Br HHH 150–152a 185a 138–140a 203b 158–160a 148–150c 216–218d C, 65.2; H, 6.05; N, 6.3% M+, 219.0886 C, 58.5; H, 4.5; N, 6.1% M+, 223.0408 M+, 207.0679 M+, 266.9884 C, 58.6; H, 4.5; N, 6.3% M+, 223.0398 C, 72.6; H, 6.2; N, 6.3% M+, 234.0632 C12H13NO3 C11H10ClNO2 C11H10FNO2 C11H10BrNO2 C11H10ClNO2 C13H13NO2 C11H10N2O4 C, 65.7; H, 5.9; N, 6.4% M, 219.0895 C, 59.2; H, 4.5; N, 6.3% M, 223.0400 M, 207.0696 M, 266.9895 C, 59.1; H, 4.5; N, 6.3% M, 223.0400 C, 72.6; H, 6.05; N, 6.5% M, 234.0641 aFrom EtOAc. bFrom MeOH–CH2Cl2. cFrom ligroin. dFrom EtOH. Fig. 2 (a) Nitrogen lone-pair delocalisation inhibiting rotation about the N·C(O) bond in 2-aminochromones; (b) conjugative effects in chromone-derived acrylamides; and (c) competitive delocalisation involving the chromone ether