摘要:

1976 977 Indirect 13C-lH Coupling in Asymmetrically Trisubstituted Benzenes : a Carbon4 3 Nuclear Magnetic Resonance Study By Ftiedrich von Massow,' Botanisches lnstitut der Universitat Karlsruhe (TH), Lehrstuhl 1, 75 Karlsruhe, Kaiserstraae 2, W. Germany Martin A. R. Smith,* Bruker-Physik AG, 751 2 Rheinstetten-Fo, Silberstreifen, W. Germany Undecoupled 13C n.m.r. spectra of a series of asymmetrically trisubstituted benzenes have been obtained by proton ' gated '-decoupling, and show invariablya characteristic sharp doublet in the aromatic region. The origin of this phenomenon is discussed in terms of the indirect 13C-lH coupling pathways available within the ring as dictated by the positions of the substituents. It is shown that a doublet without fine structure arises from the carbon atom having no protons meta, that it is a generally predictable facet of these systems and as such avaluable aid to assign- ment.DURINGthe course of l3C n.m.r. investigations into the structure of a previously unknown component (1) of essential oils of PimPirteZZa anisum L1,we noted a proton gated-decoupling spectrum (see Experimental section) OMe doublet in the gated spectrum of (2) except that in this instance it arose from C-5. In order to throw further light on this phenomenon and in the interest of promoting a better understanding OMeHoLHoLCH2 Me OMe ' '' $' '-. 151 OMe C02 H7 ci which contained a strong doublet (J = 162 Hz, from direct 13C-lH coupling) without the fine structure one would expect from indirect coupling.On the basis of a comparison of observed and calculated chemical shifts 2 we were able to assign this resonance to C-3. A similar conclusion applied to the observation of an identical 1 K. H. Kubeczka, F. v. Massow, V. Fomacek, and M. A. R. Smith, 2. Naturforsch. 1976, 316,283. 161 n, of the l3C-lH multiplets to be expected from the un- substituted ring carbons in trisubstituted benzenes of a (a) J. B. Stothers, ' Carbon-13 N.M.R. Spectroscopy,'Academic Press, New York and London, 1972, p. 197; (b) sub-stituent effect data for -CH=CH-CH, taken from an anisole spectrum in CDC1,; C-1 + 11.1 p.p.m.; o = -1.6 p.p.m.; m = +0.8 p.p.m., p = 0.0 p.p.m;. (G) no substituent data for -CH=CH-CO,H or ally1 was available.We based the sub-stituent effects of both groups, therefore, on -CH=CH, (see ref. 2a). 978 this type, we have extended our investigations to include a wider range of similar compounds. These are: eugenol (3), isoeugenol (4), 4-methoxyisoeugenol (5), and ferulic acid (6). Our results show that under normal experimental con- ditions of 6 kHz sweep widths such a sharp doublet (W112 ca. 1 Hz) is to be expected from those carbons having no proton on the meta-carbon atoms. We regard this observation as a valuable aid to assignment. EXPERIMENTAL Compound (1) was obtained as previously described.l Compounds (2) and (5) are specially synthesised for us by Dr. K. H. Kubeczka who also donated (3) and (4). The remaining compounds were obtained commercially from EGA Chemie AG and used as received.13C Fourier trans- form n.m.r. spectra were measured in CDCI, [compounds (1)-(5)] or [2H,]acetone [compounds (6)-(9)] solution using SiMe, as internal reference. Control experiments be- tween the two solvents showed no significant solvent effects. Spectra were obtained on a Bruker WH-90 instrument. Typically pulse widths were 2.5 ps (22"dip angle), sweep widths 6 kHz and each transient comprised 8 k, or if time per- mitted, 16 k data points. Gated decoupling involves the ap- plication of a strong modulated proton decoupling pulse which is switched off immediately prior to the measuring pulse and the beginning of data acquisition. After the pulse the re- ceiver samples the state of the magnetisation in a plane orthogonal to the field direction (in which Overhauser en- hancement has accrued).In consequence the carbon lines retain both nuclear Overhauser enhancement and proton coupling since the latter appears instantaneously. The technique, therefore, enables considerable time saving during the accumulation of proton-coupled 13C spectra. Proton decoupling times were usually 2 s and were repeated 0.2 s after storage of each transient. RESULTS AND DISCUSSION The chemical shifts of all ring carbons, together with their calculated positions l32 are summarised in Table 1. With the exception of the C-4 of ferulic acid the observed values are all in very good agreement with the theory.This applies especially to those atoms (C-3 or C-5) * for which we observe a sharp doublet in the gated-decoupling spectrum. The isolated, unsubstituted carbon (C-6 or C-2) shows a coupled spectrum which consists of a doublet (from direct coupling) with fine structure ap- proximating doublets or triplets. With the exception of the ether (2) the apparent coupling constants lay between 5.5-6 Hz. The third unsubstituted carbon (C4 or C-6) showed variable fine structure of the basic doublet and an apparent indirect coupling constant of from 5.5 to 7.4 Hz. Real and apparent coupling constant data are summar- ised in Table 1. The observations are illustrated in the case of compound (5)in the Figure. In the case of ferulic acid (6) we found coincident chemical shifts for C-5 and the olefinic side-chain * Throughout this section the carbons referred to are in com- pound (1) or (2)-(6) respectively.t I.U.P.A.C.convention is used throughout. Thus the carb- oxy-carbon is designated a. J.C.S. Perkin I1 H MM4w 1 -!--!-(a) Broad band, (b) off-resonance, and (c) gated-decoupling spectra of 4-methoxyisoeugenol from 150 to 100 p.p.m. Vertical pecked-lines below each spectrum denote the ring carbons. The inserts shown in the gated spectrum is 260 Hz 8-carb0n.t Furthermore, the combined signal gave only a sharp doublet in the gated decoupling spectrum. This 1976 979 observation seemed to warrant further investigation and origin of a sharp doublet in undecoupled l3C n.m.r.we accordingly examined a series of similar acids in spectra of substituted benzenes to (a) direct 13C-1H which the interfering ring carbon (a) would not be coupling and (b) a combination of coupling to ortho-and expected to show a sharp doublet (vide infra) and (b) para-protons in the absence of meta-coupling, i.e. when would be expected to have a different chemical shift the remaining indirect couplings are less than (8 k inter- compared to ferulic acid. Thus, in synapic acid (7) and ferogram) or of the order of (16 k interferogram) the p-cumaric acid (8) as well as in ferulic acid, a strong theoretical resolution. Such an observation is generally doublet without fine structure was always found at ca. predictable and in our examples this applies successfully TABLE1 Observed and calculated 132 ring carbon chemical shifts a and coupling constants Compound (1)b*c Compound (2) b~~*~ Compound (3) * I Carbon 6f J(l3CH) JFins ' 6 J(13CH) JFlne 6 J("CH)c-1 131.4 (133.4) 138.5 (137.3) 131.9 (130.3) c-2 141.6 (141.2) 108.9 (112.9) 157 4 111.5 (113.0) 166 c-3 123.3 (123.1) 162 0 151.7 (153.3) 146.8 (146.6) c-4 113.3 (114.5) 162 6.4 137.3 (136.2) 144.1 (139.6) c-5 157.3 (157.4) 122.8 (123.1) 162 0 114.7 (116.3) 157 0 C-6 111.1(112.9) 169 h 118.1 (119.5) 171 h 121.3 (119.6) 159 7.3 Compound (4) Compound (5) Compound (6) @ &f \ r -4 r 7 6 J(13CH) JFlne 6 ~(13~~3)c-1 130.3 (131.9) 130.7 (130.9) Jpine' s J(13CH) JFine 126.8 (130.3) c-2 108.0 (113.4) 155 5.9 108.7 (111.6) 145 5.6 111.3 (113.0) 155 6.9 c-3 146.3 (147.1) 148.7 (144.2) 148.0 (146.5) c-4 144.4 (140.1) 148.0 (143.4) 149.2 (139.6) C-6 114.3 (116.9) 158 0 111.2 (114.0) 157 0 115.6 (116.3) 159 0 C-6 118.9 (120.0) 167 5.9 118.4 (118.2) 167 5.8 122.7 (119.6) 156 h P.p.m.downfield of internal SiMe,. i, Ref. 1. Ref. 2a and b. d Chemical shift taken from mean of broad-band-, off-resonance-, @and gated-decoupling spectra. Ref. 2a and c. Calculated values in parentheses. Apparent indirect 13C--lH coupling, see text. h Not resolved or too complex. TABLE2 13C Chemical shift a and coupling-constant data for the side-chain carbons of compounds (6)-(9) C-u(CO,H) c-P C-Y A# 3 C s C -7 Compound 6 J(13C02HJ JWH) 6 J(l*CH) JPine 6 J(13CH) JFtne (6) 167.9 7.2 3.2 115.6 159 0 144.6 153 4.6 (7) 169.7 6.3 2.8 116.3 159 0 146.6 154 4.5 (8) 168.8 7.1 2.8 116.7 169 0 146.9 162 4 (9) 167.7 7.2 3.1 119.1 161 0 145.1 163 4 P.p.m.downfield of internal SiMe,. 5 Footnote g Table 1. 116 p.p.m. in the gated spectrum (Table 2). The ap- to C-3 [compound (l)] or C-5 [compounds (2)-(6)]. propriate signal for trans-cinnamic acid (9) similarly is Further application of these rules to the other unsub- not significantly shifted at 119.1 p.p.m. The origin of stituted carbons predicts the observation of fine structure this signal is, therefore, the olefinic @-carbon. The since meta-protons are now present and the coupling is additional ring, olefinic, and carboxylic carbons of these much greater than the theoretical resolution.It was not compounds could be assigned by off-resonance and gated our purpose to undertake a rigorous analysis of the split- decoupling experiments and, in the case of the ring ting patterns observed for the latter, although it is evi- caxbons, the chemical shifts are in agreement with the dent from the apparent coupling constants observed theory. Table 2 summarises the data for the side-chain (Table 1) that indirect coupling is dominated by the carbons. meta-proton now present. If the rationale above is Examination of indirect 13C--lH couplings in benzene 3 extended further to the couplings observed in the olefinic has shown the following inter-ring couplings (Hz) : side-chain of the acids (6)-(9), a very small indirect J(WH) = 157.5, J(I3CCH) = +l.O, J(13CCCH) = coupling between each carbon and the vicinal proton +7.4, and J(l3CCCCCH) = -1.1.Values for 1,3,5-tri- J(l3CCH) is predicted. Additionally however, C-y is chlorobenzene are not substantially different.4 It is im- placed in the favourable position 3 bonds away from the portant to note the relatively large 3-bond coupling. F. J. Weigert and J. D. Roberts, J. Amer. Chem. SOG.,1967,Given the situation in which differences in carbon 89, 2967.chemical-shifts are substantial we, therefore, ascribe the .a R. Freeman, J. Chew. Phys., 1965, 43, 3087. 980 protons on C-2 and C-6 and always shows a doublet (J = 152-154 Hz) of apparent triplets (J = 4-5 Hz). We consider this treatment justified, particularly since proton-carbon couplings in olefinic systems have been used as models for benzene.3~6 We should also add that the carboxy-carbon (C-a)shows always a clear first-order doublet of doublets (Table 2),the larger coupling varying from 6.3 to 7.2 Hz.This is in excellent agreement with the expected three-bond coupling of ca. 7 Hz to the proton on C-y. Since C-p shows no indirect coupling the smaller coupling of 2.8-3.2 Hz probably derives from the carboxy-proton, although we note that such values appear to be somewhat hard to find in the literat~re.~.' We find that the basic doublet structure in the coupled spectra of C-p in compounds (l),(2), (4), and (5) unlike that in compounds (6)-(9) is supplemented by fine structure. We tentatively ascribe this interaction to the methyl protons.In this respect our observations above, as well as those from earlier work,3,* applied to indirect 13C-lH coupling transmitted between exclusively ~$2 carbons. If, as expected, the magnitude of such coup- 6 G. J. Karabatsos, J. D. Graham, and F. M. Vane, J. Anzer. Chem. SOC.,1962, 84, 37. G. A. Olah and A. M. White, J. Anzev. Chewz. SOC.,1969, 91, 6801. J.C.S. Perkin I1 lings is dominated by the Fermi contact term, particip- ation of an sp3 carbon should predict an even smaller two-bond coupling than expected for vicinal I3C-lH coupling in the corresponding acid compounds. How-ever, the presence of both couplings might be expected to provide for the effects observed. Further work on this problem is in progress. Conclusion.-We have shown, in accordance with earlier observations and theory, that in undecoupled 13C n.m.r. spectra of asymmetrically trisubstituted benz-enes, the C-5 carbon will show only direct proton coupl- ing. In addition, though in a less precise fashion, the splitting patterns for the other unsubstituted carbons can be predicted. It is our opinion that, given the rules we propose, routine gated decoupling experiments such as those we have described, provide a convenient and far less amgibuous means of assignment compared, in parti- cular, to off-resonance decoupling. We offer our sincere thanks to Dr. K. H. Kubeczka for the gift of many of the compounds used (see Experimental section), and to Mr. V. Formacek for valuable discussions. [5/1599 Received, 7th August, 19751 G. A. Olak and A. M. White, J. Amer. Chem. SOC.,1967, 89, 7072.

ISSN:1472-779X    

DOI:10.1039/P29760000977

出版商:RSC    

年代:1976

数据来源: RSC