2. PHYSICAL METHODS OF STRUCTURE DETERMINATION Part (i) Nuclear Magnetic Resonance Spectroscopy By J. Feeney (Varian Research Laboratory Walton-on- Thames Surrey) NUCLEAR magnetic resonance spectroscopy continues to have widespread use as a powerful means of determining molecular structures. In fact almost all papers concerned with organic structural and stereochemical work contain some reference to the technique. However this report does not attempt to reflect the routine analytical usage of the n.m.r. technique but rather to present the more specialised aspects of its application. Some of the more noteworthy contributions to the field have been con- cerned with critical re-examination of several standard procedures to assess their validity. This will be evident in the sections of the report dealing with kinetic conformational and double-resonance studies.For example there is now firm experimental confirmation of the validity of the method of full line- shape analysis for studying kinetic processes. Special attention has been devoted to two of the more recent developments namely n.m.r. studies in liquid-crystal media and the application of intra- molecular nuclear Overhauser effects in making spectral assignments. Chemical Shifts.-At present one cannot carry out the necessary detailed quantum mechanical calculations to explain chemical shifts found for the various nuclei in different molecular environments. To explain differences in observed chemical shifts one often resorts to specific physical models (such as those based on anisotropic inductive or electric-field effects).Despite the limitations of such an approach to the problem it is often possible to calculate chemical-shift differences which are of useful practical consequence. For example Whalley and his co-workers'. have modified the McConnell equation used for estimating shielding contributions from anisotropic effects to enable it to be used for cases where the distance between the shielding group and the observed proton is small compared with the length of the induced dipole. They found that when estimating the chemical-shift change accom- panying,a change of substituent one must consider the shielding effects of all bonds displaced as well as those of all the bonds introduced.For the case of carbon-carbon double bonds deshielding is predicted for protons in the J. W. Apsimon W. G. Craig P. V. Demarco D. W. Mathieson L. Saunders and W. B. Whalley Tetrahedron 1967,23,2339. J. W. Apsimon. W. G. Craig P. V. Demarco D. W. Mathieson L. Saunders and W. B. Whalley Tetrahedron 1967,23,2357. J. Feeney regions at the ends of the double bonds but a proton outside this region will be shielded whether it is in the plane of the double bond or above it which is not as had been predicted previously. Following on this work Karabatsos (1) and his co-workers3 suggested that anisotropic effects for carbonyl groups are not as predicted previously. Thus a proton HA in (1) in the plane of the carbonyl group might be expected to be shielded rather than deshielded.The observed temperature-dependence of chemical shifts in some substituted ketones and aldehydes can be best interpreted in terms of rotameric population changes and the above shield’ng considerations. Anisotropic shielding effects in puckered cyclobutane rings & and heterocyclic nitrosamine~~ have also been examined. Aromatic ring-current chemical-shift contributions have been calculated and compared favourably with available experimental data for even alternant pentacyclic hydrocarbons6 and cyclic polyene~.~ Discussion of the validity of the ring-current approach has continued.8* Adcock and Dewar” have examined an extensive series of 1- and 2-fluoro- naphthalenes the substituent effects on the chemical shifts are shown to differ from substituent effects on physical and chemical properties of other side-chains.In an examination of hydrocarbons and their ions linear correlations of ‘H and I3C chemical shifts with the corresponding calculated partial charges on the atoms were found.” Chemical shifts of methyl groups in ortho-toluenes cis-vinyl hydrogens in olefinic systems and the H-3 aromatic protons in 1,2-disubstituted benzenes have been found to give linear correlations with an empirical parameter Q = P/Ir3 where P is the C-X bond polarisability I is the first ionisation potential and I is the C-X bond length.” A linear correlation between chemical-shift changes of a proton and its distance from a neighbouring methyl group has been pointed out for derivatives of bicyclo[2,2,l]heptane and bicyclo[2,2,2]octane.l3 G. J. Karabatsos G. C. Sonnichsen N. Hsi and D. J. Fenoglio J. her. Chem. Soc. 1967 89 5067. N. Nakagawa S. Saito A. Suzuki and M. Itoh Tetrahedron Letters 1967 1003. Y. L. Chow Angew. Chem. 1967,79,51. T. B. Cobb and J. D. Memory J. Chem. Phys. 1967,47,2020. F. Baer H. Kuhn and W. Regel 2.Naturforsch. 1967,22a 103. * J. I. Musher J. Chem. Phys. 1967,46,1219. J. M. Gaidis and R. West J. Ctrem. Phys. 1967,46 1218. lo W. Adcock and M. J. S. Dewar J. Amer. Chem. SOC.,1967,89,379. N. C. Ba$d and M. A. Whitehead Theor. Chim. Acta 1966,6 167. l2 W. B. Smith and J. L. Roark J. Amer. Chem. SOC. 1967,89,5018. l3 E. Pretsch H. Immer C. Pascual K. Schaffner and W. Simon Helv.Chim. Acta 1967,50 105. Physical Methods of Structure Determination 7 Gusten and his co-~orkers'~ have studied the influence of ring substituents on the spectra of substituted cis-and truns-stilbenes. Coupling Constants.-Coupling constants especially proton-proton and carbon-proton coupling constants are more amenable to theoretical calcu- lation than are chemical shifts and considerable success in calculating coupling constants has been achieved in the past few years. Using a valence-bond variational calculation for cis and frans H-H coupling constants in ethylene Chandra and Narasimhan lS have produced values in excellent agreement with observed values (calculated JHH trllm 18-73 JHHcis13-35 c./sec.; observed 19.1 and 11.5 respectively).They treated the problem as an eight-electron eight-orbital (six 0-and two n-orbitals) problem. Gunther16 has discussed the effects of delocalisation on H-H spin coupling in unsaturated systems. H-H geminal coupling constants have been calculated using the Dirac-Van Vleck vector model.17 Dixon18 has reported a MO theory for proton spin-spin coupling constants which is simplified such that one requires only two MO parameters one for the n-interactions of carbon hybrid orbitals and one for the interaction of orbitals on the same atom to give a reasonable prediction of observed trends in coupling constants. Attempts to calculate the H-F coupling constant in hydrogen fluoride using a refined perturbational calculation indicate that the calculated values are extremely sensitive to the wave functions used and it was concluded that presently available wave functions do not have the required accuracy.' The Pople and Santry20$21 MO approach has been used successfully to calculate 13C-H and 29Si-H coupling constants.22.23 From a theoretical interpretation of substituent effects on such coupling constants in methanes and silanes it was predicted that the electronegativity of the substituents will be of paramount importance. Dixon has used a valence-bond treatment to rationalise the proportionality between 13C-H coupling constants and the carbon s-~haracter.~~ Several interesting correlations between coupling constants and other molecular parameters have been pointed out. The most interesting of these is an empirical relationship found between vicinal proton-proton coupling constants and reaction rates for elimination and substitution type processes in aliphatic system^.^ Karplus type correlations between some H-F vicinal l4 H.Gusten and M. Salzwedel. Tetrahedron 1967,23 173. 187. Is P. Chandra and P. T. Narasimhan,Mol. Phys. 1967,12,523. l6 H. Gunther Tetrahedron Letters 1967,7%7. H. G. Hecht J. Phys. Chem. 1967,71,1761. W.T. Dixon J. Chem SOC.(A) 1967 1879. l9 Y. Kato and A. Saika J. Chem. Phys. 1967,46 1975. 2o J. A. Pople and D. P. Santry Mol. Phys. 1964,8,1. 21 J. A. Pople and D. P. Santry Mol. Phys. 1965,9 311. 22 R. Ditchfield M. A. Jensen and J. N. Murrell J. Chem. SOC.(A) 1967 1674. 23 J. N. Murrell P. E. Stevenson and G. T.Jones Mol. Phys. 1967 12,265. 24 W. T. Dixon Theor. Chim. Actu 1966,4,359. 25 W. T. Dixon Tetrahedron Letters 1967,2531. J. Feeney coupling constants and dihedral angles have been observed.26 In a series of asymmetrical benzylic compounds the JHH gem values between the non-equivalent protons of the methylene group show a linear correlation with Hammett o-c~nstants.~~* 28 From an examination of an extensive series of perfluorovinyl compounds CF,=CFX Rest2' has reported several linear relationships between JFFgem values and the mean of the "F chemical shifts of the terminal ethylenic CF2 fluorine nuclei a different linear relationship is observed depending on wehther X is organic a main-group metal a metalloid or a transition metal.In a series of substituted fluorobenzenes the 13C-H and 13C-F coupling constants were found to depend on both the substituent and its position of sub~titution.~~ 13C-H coupling at a particular position usually increases with increase in electron-withdrawing power of the sub- stituent. Long-range coupling constants. Calculations of the long-range (5 bond) coupling constants in buta-1,3-dienes indicate that in the s-trans-isomer the n-mechanism dominates the co~pling.~ Long-range couplings have been reported for carboxylic acid estersJ2 (CH3COOCH, JCH3-CH3 0.21 c./sec.) alkyl formatesJ3 (HACOOCHXCHY JAX -0432 to -1.00 JAY+0.42 to +0.63 c./sec.) and orcinol derivativesJ4 (2) (JCH3-H-ortho 0*6&-0.80,ScH3-H-poro 0.57 c./sec.) and such couplings in 1,3-dioxans have been explored further.35 26 K.L. Williamson Yuan-Fang Li F. H. Hall and S. Swager J. Amer. Chem. SOC. 1966,88,5678. 27 R. R. Fraser P. Hanbury and C. Reyes-Zamora Canad. J. Chem. 1967,452481. R. W. Franck and J. Auerback Ccnad. J. Chem. 1967,452489. 29 A. J. Rest J. Chem. Phys. 1967,47 1559. 30 S. Mohanty and P. Venkateswarlu,MoI. Phys. 1967,12 277. 31 A. V. Cunliffeand R. K. Harris MoI. Phys. 1967 13 269. '* K. Hayamizu and 0.Yamamoto J. MoI. Spectroscopy 1967,22 119. '' K. Hayamizu and 0.Yamamoto J. Mol. Spectroscopy 1967,23 121. 34 D. T. Witiak D. B. Patel and Y. Lin J. Amer. Chem. SOC. 1967,89 1908. '' J. E. Anderson J. Chem. SOC. (B) 1967,712. Physical Methods of Structure Determination 9 In 1-methylpyrene (3) the CH absorption appears as a doublet (JHae3 1.3c./sec.) this splitting is much larger than normally observed for interaction between a methyl group and an adjacent aromatic proton and indicates that there is a substantial amount of double-bond character in the intervening bond.36 A similar effect is observed in 9-methylphenanthrene (4).A study of long-range stereospecific H-F coupling in the t-butyl derivative (5) has revealed the presence of hindered rotation.37 The room-temperature spectrum of the methyl groups shows a doublet (JHF2.9 c./sec. 6H) and a singlet (JHF-0,3H) which is evidence for two different types of methyl group existing in the hindered form of the molecule shown in (5). At 200" a single doublet absorption (JHF2.2 c./sec.) is observed.Pascual and Simon3' have examined several compounds containing geminal methyl groups; long-range CH,-CH coupling ( >O-3 c./sec.) is observed only in cases where the carbon atom bonded to the two methyl groups is either substituted by one or two oxygen functions or is sp2-hybridised. Observation of line-broadening resulting from stereospe,cific long-range coupling in epoxidation products has been used to determine the stereochemical course of the reaction.,' Signs of coupling constants. To measure the success of theoretical attempts to predict coupling constants it is necessary that we should know not only the magnitude of a coupling constant but also its absolute sign. It is not easy to measure directly absolute signs of coupling constants (it can be done by studying spectra recorded for molecules dissolved in the nematic phase of a liquid crystal) but fortunately they can often be obtained indirectly with greater ease by measuring relative signs of coupling constants in systems where one of the coupling constants is known to have a particular absolute sign (e.g.J13C-H for directly bonded nuclei is always positive). Relative signs of coupling constants can often be determined from a complete spectral analysis in cases where the nuclei are strongly coupled. However easier and more direct methods are available by using one of a variety of double-resonance techniques (such as selective decoupling;' spin-tickling:' nuclear Overhauser effects,42 transient selective irradiati~n?~ and transient nut at ion^.^^).For example using heteronuclear spin-tickling on the spectrum of dimethylphenylphos- phine (CH,),PhP JlsGSlp(-14 & 1 c./sec.) is shown to be opposite in sign to J13(-~ (+130-3 & @2) which proves that J13~31pis absolutely nega- tive in sign.45 In the protonated cation of this phosphine (CH3),PhPH+Br- the sign of J13C-31~ (+56 1 c./sec.) is shown to be absolutely positive 36 E. Clar B. A. McAndrew and M. Zander Tetrahedron 1967,23,985. 37 J. P. N. Brewer H. Heaney and B. A. Marples Chem. Comm. 1967,27. 38 C. Pascual and W. Simon Helv. Chim. Acta 1967,50 94. 39 R. G. Carlson and N. S. Behn J. Org. Chem. 1967,32 1363. 40 R. Freeman and D. H. Whiffen Mol. Phys. 1961,4,321. 41 R. Freeman and W. A. Anderson J. Chem. Phys. 1962,37,2053.42 R. Kaiser J. Chem. Phys. 1963,39,2435. 43 R. A. Hoffman B. Gestblom. and S. Forsen. J. Chem. Phys. 1964,40,3734. 44 J. A. Ferretti and R. Freeman J. Chem. Phys. 1966,44 2054. 45 W. McFarlane Chem. Comm. 1967,58. 10 J. Feeney by the same method. The absolute signs of several other coupling constants have been measured in this way for example 13C-119Sn (-) 'H-"'Sn (-),46 I3C-l4N (+) between directly bonded nuclei in methyl i~ocyanide,~~ 29Si-H (+) in tetramethyli~lane,~~ P-H and P-F coupling constants P-P (+) for directly bonded phosphorus nuclei,4i and C-H ( +)and P-H (+) for directly bonded nuclei in dimethyl phosphite" (CH,O),P(O)H. Manatt and his co-~orkers~~ have determined the relative signs of coupling constants for the systems F-C-P F-C-P-H P-H F-C-F F-C-P-F P-F F-C-C-P F-C-C-F F-C-C-C-P F-C-C-C-F F-C-P-P P-P H-C-P H-C-P-P and F-C-P-P-C-H in the series of molecules CF,PH2 CF3PF2 (CF3)2PH (CF3)2 PF F[CF2] 2pc12 F[ CF2] ,Pcl2 (CF3)2PP(CH3)2 and CF3P-+ P(CH3)2.Multiple-resonance techniques were used to determine the relative signs of the six proton-proton coupling constants in isopropenylacetylene (6).' By irradiating at the CH frequency in a double-resonance experiment one obtains a simple decoupled spectrum expected for an ABX stin system on which tickling experiments can be carried out to give signs of J,AB JAX and JBx. An interesting alternation in signs is observed for JpH (+25 c./sec.) and JpHe ( T6 c./sec.) in the phospholene (7).52 Double Resonance.-Even the more sophisticated double-resonance ex-perimenrs are now being used routinely not only by the specialist but also by those interested mainly in the molecular structural information available Me HA from n.m.r.studies. This results from the widespread availability of instru- mentation capable of making such measurements with relative ease. However the need for caution in analysing decoupled spectra when large decoupling 46 W. McFarlane J. Chem. SOC.(A),1967,528. 47 W. McFarlane J. Chem. SOC.(A),1967 1660. 48 R. R. Dean and W. McFarlane Mol. Phys. 1967 12,289. 49 S. L. Manatt D. D. Elleman A. H. Cowley and A.,B. Burg J. Amer. Chem. SOC.,1967,89,4544. W. McFarlane J. Chem. SOC.(A) 1967,1148. 51 H. M. Hutton and T. Schaefer Cunad. J. Chem.1967,45 1165. 52 D. Gagnaire J. B. Robert and J. Verrier Chem. Comm. 1967 819. Physical Methods of Structure Determination 11 fields are used has been stressed by Connor and his co-~orkers.~~ When yH2/2x becomes comparable in magnitude with the chemical shifts (vA -vx) between irradiated and observed groups the spectrum can become more complicated rather than simplified. Bloch Siegert type shifts also complicate the analysis of such spectra. Because of these problems it is not possible to carry out clean spin-decoupling experiments in the AMX spin system of vinyl acetate. When one is decoupling to achieve a simplified but still complex spectrum from an even more complex spectrum one is faced with difficulties unless there is some feature of the decoupled spectrum to indicate when optimum de- coupling conditions have been achieved.An excellent example of where such experiments are possible is provided by the application of double resonance to the n.m.r. spectrum of triptycene (8).54When the methine proton frequency is irradiated the resulting decoupled spectrum is that expected from an AA’BB’ spin system which is still a complex spectrum. Because one knows that the decoupled spectrum must show the symmetry one expects for an AA’BB’ type spectrum the optimum decoupling conditions can be achieved with some confidence. Spin-tickling has been used to obtain a complete analysis of AA’BB’ spin system from the ring protons in dimethyl cis-and trans-1,2-dibromo-cyclobutane-l,2-dicarbo~ylate.~~ The effects of relaxation in n.m.r.double- resonance spectra of weakly coupled spin systems have been ~onsidered.’~ Several computer programmes for the calculation of n.m.r. double-resonance spectra have been developed.57 Cyc1obutanes.-It is only recently that the n.m.r. parameters of cyclobutanes have become well documented. Fleming and Williams58 have now provided a comprehensive summary of all the literature proton-proton coupling constants for cyclobutanes cyclobutanones cyclobutenes and cyclobutenones. For cyclobutanes Jujccjs4.6-1 1.5 Juierrans2.@-10.7 JBem-10.9 to -14.4 c./sec. The Juiccisand Juictrans values are often similar in magnitude in the same molecule but Juiccisis usually larger than Juictrons; this was found to be the case in twenty cyclobutanes studied by Weitkamp and K~rte.~~ When the cyclo- ’’ T.M. Connor D. H. Whiffen and K. A. McLauchlan Mol. Phys. 1967,13,221. 54 K. G. Kidd G. Kotowycz and T. Schaefer Canad. J. Chem. 1967,45,2155. 55 E.Lustig E. P. Ragelis N. Duy and J. A. Ferretti J. Amer. Chem. SOC.,1967,89 3953. ” B.D.Nageswara Rao and L. Lessinger Mol. Phys. 1967 12 221. 57 G.Govil and D. H. Whiffen Mol. Phys. 1967,12,449. 58 I. Fleming and D. H. Williams Tetrahedron 1967,23,2747. 59 H.Weitkamp and F. Korte Tetrahedron 1966,Supplement No. 7,75. 12 J. Feeney butane ring is incorporated into a strained bicyclo[2,l,l]hexane system very different coupling constants from the above are observed (Jgemvalues in the range -5.4 to -8-4 c./sec.). In cyclobutenes coupling constants between olefinic protons are in the range 25-40c./sec.while vicinal coupling between protons attached to adjacent sp3-and sp2-carbons is small (0-1 c./sec.). The spectrum of cyclobutanone (9) has recently been reassigned after analysing as an AA’BB’B”B’” spin system :60 the coupling constants are JAA -11-1; JABcis +10.0;JBBp -17.5;JAW +6.4 ;JBWIcis +4.6 ;JBB,**trpns -2*8c./sec. Weitkamp and Korte” found that in the twenty cyclobutanes they studied the shielding values for the ring protons were always in the range z 6-2-8.2 and the chemical-shift contributions were found to be additive. Studies of Nuclei other than Hydrogen.-Carbon-1 3. Considering the enormous potential in this field comparatively few publications on ”C resonance appeared in 1967.Weigert and Roberts6’ have used spectrum- accumulation techniques to record the ’ spectrum of benzene. The molecule constitutes a seven-spin system with all nuclei being magnetically nm-equiva- lent and the spectrum was fully analysed using a computer to give JCCH+1.0 JmCH+7*4 and Jcccm -1.1 c./sec. Grant62 has likewise examined 13C- enriched acetic acid (such that 0.2%of molecules had two carbon-13 nuclei) to observe the spectrum of the species ‘3CH313COOH. By heteronuclear decoupling he showed that Jcc (57.6 c./sec.) has the same sign as JCH(positive). ’ resonance has been used successfully to characterise aromatic petroleum fraction^.^ The various types of aromatic and aliphatic carbons give clearly distinguishable ‘ absorption signals in the accumulated spectra.A theoretical interpretation of ’ shifts in aromatic compounds (biphenyl naphthalene phenanthrene and pyrene) which includes the effects of G-and n-electron densities on a local-charge parameter has been put forward.64 It has been noted that carbon atoms in a spatially crowded environment have chemical shifts at higher fields than e~pected.~’ A comparison66 of “B chemical shifts in some boron-nitrogen compounds (H,BNH3) with the 13C chemical shifts in the analogous alkanes (H,CCH,) indicates a linear relationship given by 613C(p.p.m.) (from benzene) =1-716”B (p.p.m.) (from BF,.Et,O) +81 ‘O L. H. Sutcliffe and S. M. Walker J. Phys. Chem. 1967,71 1555. 61 F. J. Weigert and J. D. Roberts J. Amer. Chem. SOC.1967,89,2967. 62 D. M. Grant J. Amer. Chem. SOC. 1967,89 2228. ’’S. A. Knight Chem. and Ind. 1967 1921. 64 T. D. Alger D. M. Grant and E. G. Paul J. Amer. Chem. SOC.,1966,88 5397. ”D. M. Grant and B. V. Cheney J. Amer. Chem. SOC. 1967,89,5315. 66 B. F. Spielvogel and J. M. Purser J. Amer. Chem. SOC. 1967,119 5294. Physical Methods of Structure Determination 13 13C studies of substituted methyl ben~oates,~~ cyclopropyl ketones,68 and substituted an is ole^^^ have been reported by Stothers and his co-workers. When the 3C spectra of 4-substituted pyridines70 are examined the observed substituent effects are very similar to those observed in benzene (to 3 p.p.m.). 3C chemical shift data from fifteen methylcyclohexanes have been rep~rted.~' Litchman and Grant72 have measured 13C-13C coupling-constant variations with changes in polar substituents X in the series XC(CH,) ;charge polarisa- tion effects due to electronegative groups do not play a major role in controlling the coupling constants.The implication of this is that 13C-13C values provide a better criterion for bond hybridisation than do 13C-H values. Weigert and Roberts7 have characterised the hybridisation in three-membered ring systems by studying such effects in halogenated cyclopropanes. 13CH Satellite spectra. In 'H spectra of organic molecules one can often detect multiplets arising from the molecules containing 13C nuclei present in 1.108 % natural abundance. When the 13C is directly bonded to protons one observes a large doublet splitting from 13C-'H spin-spin interaction with components almost symmetrically disposed about the much more intense resonance band for the analogous protons bonded to "C atoms (for example in methane JcH is 125 c./sec.).In some molecules the introduction of a 13C atom results in nuclei which are magnetically equivalent in the non-13C- containing molecule becoming magnetically non-equivalent. For such mole- cules the 13CH satellite spectra often feature coupling constants which do not appear in the spectrum of the 12C-containing molecules. For example benzene shows a single absorption for the normal spectrum obtained for the "C-containing molecules. The molecules containing one 13C atom have nuclei which are magnetically non-equivalent such that we now have a seven-spin system.74 From a detailed analysis of the 13CH satellite spectra the ring proton-proton coupling constants can be deduced (Jortho7.54 J,,, 1.37 JpWa0.69 JcH 158.34 c./sec.).Hill and Roberts75 have examined the 13CH satellites of cyclobutene and their findings in conjunction with information from the 'H spectrum of deuteriated cyclobutene derivatives provide a more accurate set of coupling constants. It is of paramount importance to include non-bonded C-H coupling constants in the analysis of the 13CH satellite spectra ; a re-examination of the 13CH spectra of p-benzoquinone including such non-bonded coupling constants gave values for the H-H coupling constants very different from those reported previou~ly.~~ Finer and Harris77 67 K.S. Dhami and J. B. Stothers Canad. J. Chem. 1967,45,233. 68 D. H. Marr and J. B. Stothers Canad. J. Chem. 1967,45 225. 69 K. S. Dhami and J. B. Stothers Canad. J. Chem. 1966,44,2855. 70 H. L. Retcofsky and R. A. Friedel J. Phys. Chem. 1967,71,3592. 71 D. K. Dalling and D. M. Grant J. Amer. Chem. SOC.,1967,89,6612. 'I2 W. M. Litchman and D. M. Grant J. Amer. Chem. SOC. 1967,89,6775. 73 F. J. Wiegert and J. D. Roberts J. Amer. Chem. SOC. 1967,89,5962. 74 J. M. Read R. E. Mayo and J. H. Goldstein J. Mol. Spectroscopy 1967,22,419. 75 E. A. Will and J. D. Roberts J. Amer. Chem. SOC.,1967,89 2047. " G. Govil J. Chem. SOC.(A),1967 1416. 77 E. G. Finer and R. K. Harris Mol. Phys. 1967 13 65. 14 J. Feeney have analysed the 13CH satellites in systems which normally have high sym-metry (X,AA'X' systems) where the reduced symmetry allows one to extract coupling constants which are not featured in the normal spectrum.A unique set of spectral parameters has been obtained from analysis of the 13CH spectra of acetaldehyde diethyl acetal The 13C-H coupling con- stants for the two non-equivalent protons (a b) in the methylene group are not equal (JcH 141.01 and 139-64 c./sec.) which provides another criterion of non-equivalence in such structures. 3CH satellites of homodimers of thimine and dimethylthymine have provided coupling constants essential to the stereochemical characterisation of the molecules.79 From the 13CH spectrum of cyclopropane,gOit is possible to obtain Jgem-4.34 & 0.03; JCk8.97 & 0.01 ; J,,,, 5.58 & 0.01 c./sec.Oxygen-17. A review on the chemical applications of 170nuclear resonance has appeared recently.'' By using 170labelling one can study the reversible hydration and dehydration of acetaldehydeg2 and other carbonyl-containing' molecules. Thus in aqueous solution we obtain R'R2C(OH) + R1R2C=0 + H20 and by enriching the aliphatic carbonyls in 170and observing the intensities of the two different 170signals one can obtain the equilibrium constant for the above equilibrium. The carbonyl 170absorption is in the range -520 to -560 p.p.m. from H21 70,while the gem-diol has a similar chemical shift to that of water.83 170-enriched asymmetric fl-diketones give a broad 170band for the keto-tautomer while two bands are observed for the two non-equivalent oxygen nuclei of the enolic f01-m.~~ By measuring the chemical shifts of the enol peaks one can estimate the equilibrium constant for the equilibrium involving the two possible enolic tautomers PhCO-CH :C(OH)Me + PhC(0H):CH-CO-Me Nitrogen.Mathias and his co-workersa5 have used the heteronuclear double-resonance method to measure 14N chemical shifts in some thioamides " L. S. Rattet L. Mandell and J. H. Goldstein J.Amer. Chem. SOC.,1967,89 2253. 79 D. P. Hollis and S. Yi Wang J. Org. Chem. 1967,32 1620. V. S. Watts and J. H. Goldstein J. Chem. Phys. 1967,46,4165. 81 B. L. Silver and Z. Luz. Quart. Rev. 1967 458. P. Greenzaid Z. Luz and D. Samuel J. Amer. Chem. SOC.,1967,89 756. " P. Greenzaid Z.Luz and D. Samuel .I.Amer. Chem. SOC.,1967.89. 749. 84 M. Gorodetsky Z. Luz and Y.Mazur J. Amer. Chem. SOC., 1967,89 1183. 8s P. Hampson and A. Mathias Mol. Phys.. 1967 13 361. Physical Methods of Structure Determination and thia~oles.~~ 2- and 8-hydro~yquinolines,~~ In this method the 14N frequencies are measured by observing the effect on the NH proton spectrum of a second irradiating field operating at the frequency of the I4Nnucleus under investigation. Mathia~~~ showed that 2-amino- and 2-methylamino-benzo- thiazole exist in the amino-form (11) rather than the imino-form (12) and that for the 2-hydroxyquinoline studied the 0x0-form (13) rather than the hydroxy- form (14) exists.86 Witanowski has measured 14N chemical shifts by direct H (1 1) (1 2) H (13) observation ofthe nuclei in aromaticnitro-compounds,88 nit rile^,^' is on it rile^,^^ and nitro alkane^.^' In nitroalkanes the 14Nshifts are observed over a range of 70 p.p.m.;it was found possible to predict chemical shifts to 1 p.p.m. accuracy by using simple additivity rules for substituents CH, RCH, and C1 in the series R'R2R3C*N02.Witanowski has proposed two primary reference standards for I4N studies nitromethane for organic systems and the nitrate ion for aqueous systems.88 The chemical-shift difference between the two standards is zero. An extensive collection of '5N-H coupling constants for directly bonded nuclei and also for three- and four-bond coupling has been rep~rted.~' In the syn-isomers of oximes geminal JISN-Hvalues in the range 2-6-4.2 c./sec.have been observed while in the anti-isomers the range of values is from 14.2to 16.3 and is much larger because of the effects of the nitrogen l~ne-pair.~' 14N-l H coupling constants have been measured in alkylpyridinium halides" and in enammonium saltsg3 [for example in (CH,),N+CH* CH, JNHrrMs +5-6 JNHgem +3.6 JNHck +2*6 c./sec.]. The relative signs and solvent- 86 P. Hampson and A. Mathias Chem. Comm. 1967,371. '' A. Mathias Mol. Phys. 1967 12 381. M. Witanowski L. Stefaniak and G. A. Webb J. Chem. SOC. (B),1967 1065. 89 M. Witanowski Tetrahedron 1967,23,4299. 90 M. Witanowski and L. Stefaniak J. Chem. SOC. (B),1967 1061. 91 A. K. Bose and I. Kugajevsky Tetrahedron 1967,23 1489; J. P.Kintzinger and J. M. Lehn Chem. Cornrn..1967. 660. 92 J. F. Biellmann and H. Callot Bull. SOC. chim. France 1967,2 397. " J. M. Lehn and R. Seher Chem. Comm.. 1966. 847. 16 J. Feeney dependencies of "N-'H spin-coupling constants in ["N]quinoline its ethiodide and ["N]quinoline oxide have been mea~ured.'~ Solvent Effects.-Solvent effects on chemical shifts. Benzene-induced solvent effects have been extensively used in investigations of the 'H spectra of pyri- dines,g5 q~inolines,~~ pyrroles," indoles," ortho-and meta-substituted methoxybenzene~,'~ polymethoxy-substituted aromatic^,'^ three-membered ring carbonyl epoxides," ketonic derivatives of cyclopr~pane,~~ amines,' O0 and maleic anhydride. ' ' Further discussion of the mechanism of the benzene-induced solvent effects has been reported.lo' On the basis of evidence from freezing-point diagrams and the effects of aromatic solvents on chemical shifts of t-butyl and 1-adamantyl halides Fort and Lindstrom'03 have suggested that a specific 1 :1 complex is not formed but rather that there is a slight geometrical ordering of the solvent around the solute molecules with a rapid exchange between ordered and non-ordered solvent molecules; such an arrangement would give the n.m.r. spectrum expected for a long-lived species without being thermodynamic- ally equivalent to such a species. The usefulness of dimethyl sulphoxide as a solvent for hydroxy-compounds has been further exploited;'04 not only are exchange processes suppressed in this solvent but the observed hydroxy group chemical shifts become independ- ent of solvent at low concentrations of the alcohol.Solvent effects on halogenopr~penes'~~ and substituted pyridinesio6 have also been studied. Abraham and Cooper'07 have extended their electrostatic theory of medium effects to account for the effect of the medium on the energy differences between rotational isomers of any solute molecule; in the case of 1,1,2-trichloroethane the theory gives good agreement with the experimental values of the energy difference between rotational isomers in the liquid and vapour states. Homer'" has described an experimental method for determining the contribution to the shielding from the magnetic anisotropy of the solvent. Solvent effects on coupling constants.Because coupling constants are in- sensitive to anisotropic effects it is possible that studying solvent effects of coupling constants (such as JcH)might provide a better way of investigating 94 K. Tori M. Ohtsuru K. Aono Y. Kawazoe and M. Ohnishi J. Amer. Chem. SOC.,1967,89,2765. 95 J. Ronayne and D. H. Williams J. Chem. SOC.(B) 1967 805. 96 J. H. Bowie J. Ronayne and D. H. Williams J. Chem. SOC.(B) 1967,535. 9' H. M. Fales and K. S. Warren J. Org. Chem. 1967,32 501. 98 D. W. Boykin A. B. Turner and R. E. Lutz Tetrahedron Letters 1967,817. 99 J. Seyden-Penne D. Arnaud J. L. Pierre and M. Plat Tetrahedron Letters 1967,3719. loo D. J. Barraclough P. W. Hickmott and 0.Meth-Cohn Tetrahedron Letters 1967,4289. lol C. Ganter L. G.Newman and J. D. Roberts Tetrahedron 1966 Supplement No. 8 507. lot J. Ronayne and D. H. Williams J. Chem. SOC.(B) 1967,540. lo3 R. C. Fort and T. R. Lindstrom Tetrahedron 1967,23,3227. lo4 R. J. Ouellette D. L. Marks and D. Miller J. Amer. Chem. SOC. 1967,89,913. lo' F. Hruska D. W. McBride and T. Schaefer Canad. J. Chem. 1967,45 1081. loci R. J. Chuck and E. W. Randall J. Chem. SOC.(B),1967,261. lo' R. J. Abraham and M. A. Cooper J. Chem. SOC.(B) 1967,202. lo* J. Homer Tetrahedron 1967,23,4065. 17 Physical Methods of Structure Determination solute-solvent interactions than by studying chemical-shift changes with solvent.'0g J in bromoform has been measured in 30 different solvents (204.31 c./sec. in cyclohexane 211.60 in dimethylformamide) and the solvent effects on J values for thirteen other substituted methanes have been also reported.The results can be explained in terms of specific interactions such as hydrogen-bonding.'Og Weak hydrogen-bonding effects (self-association) have also been invoked to explain the changes in JcHin chloroform with changes in temperature [observed changes from 20943-209.0 c/sec. from -61" to +58" (vapour phase)]. '' The solvent dependence of H-F coupling constants in trifluoroethylene and vinyl chloride has been measured :I1' the orientation of the solute dipole strongly affects the solvent-dependencies of the JHFgem values. Electric-field effects from the solvent appear to be the main factor controlling the changes in coupling constants but the strong molecular associations will alsocontribute.Hutton and Schaefer1l2 have found that solvent effects on JHFck and JHHgem in 1-chloro-1-fluoroethylenedepend primarily on the dielectric constant of the medium. JHForrho and (JHFortho + JHFnreta) in substituted fluorobenzenes increase algebraically as the dielectric constant of the medium increases.' ' Small changes of the coupling constants hetween ring protons in nitro- aromatics have been measured on changing solvents (largest change is +0.14 & 0.1 c./sec.).114 Kinetic Studies.-The most notable contributions to this field have been contained in papers dealing with the validity of the various methods of studying kinetic processes using n.m.r. studies. Anet and Bourn"' have made a careful study of the kinetic parameters for changes in the conformation of C2H11]-cyclohexane by two different methods namely full line-shape analysis and the double-resonance method of Hoffman and Forsen.' ' The line-shape analysis was carried out at different temperatures with the deuterium nuclei decoupled ; thus the system is a simple problem involving two equally populated sites undergoing exchange between axial and equatorial positions.Both methods gave values of AF* AH* and AS* for the chair-to-boat process in good agreement with each other. However the AS* result is different from that obtained by a third method involving spin-echo measurements.' ' Sheppard and Harris'I8 have drawn attention to the difficulties encountered in deter- mining meaningful values of entropy changes from n.m.r.studies of ring- inversion in cyclohexane. log V. S. Watts and J. H. Goldstein J. Phys. Chem. 1966,70,2887. 'lo A. W. Douglas and D. Dietz J. Chem. Phys. 1967,445 1214. S. L. Smith and A. M. Ihrig J. Chem. Phys. 1967,445 1181. 'I2 H. M.Hutton and T. Schaefer Canad. J. Chem. 1967,45,1111. 'I3 H. M. Hutton B. Richardson and T. Schaefer Canad. J. Chem. 1967,45 1795. S. L. Smith and A. M. Ihrig J. Mol. Spectroscopy 1967,22 243. 'I5 F. A. L. Anet and A. J. R. Bourn J. Amer. Chem. SOC. 1967,89 760. R. A. Hoffman and S. Forsen J. Chern. Phys. 1963,39,2892. 'I7 A. Allerhand F. Chen and H. S. Gutowsky J. Chem. Phys. 1965,42,3040. N. Sheppard and R. K.Harris J. Mol. Spectroscopy 1967,23,23. J. Feeney Mannschreck and his co-workers' '' have measured the rates of internal rotation around C-N amide bonds in the rotamers (15) and (16) of N-benzyl- N,2,4,6-tetramethylbenzamideby two different methods namely line-shape analysis and by an equilibration method.The equilibration method involves isolating form (15) as a pure rotamer and then observing the formation of form (16) with time from its n.m.r. spectrum. The activation parameters obtained by the two methods are in good agreement. Gutowsky and his co-workers'20 showed concurrently that there was good agreement between the parameters obtained from complete line-shape analysis and an equilibration method for the rate of internal rotation about the C-N amide bond in N-benzyl-N-methyl formamide. Thus we now have experimental confirmation of the validity of the method of line-shape analysis for studying kinetic processes.There has been an interesting use of line-shape analysis on the 19'Hg-H proton satellite spectra of dimethylmercury to study intermolecular exchange of methyl 122 In the presence of aluminium chloride catalyst the exchange is very rapid and the satellite bands disappear completely which indicates the exchange of methyl groups to be an intermolecular process.'21 Inability to detect 33S-'9F satellites in the "F spectrum of sulphur tetra- fluoride has been advanced as evidence for the existence of an intermolecular rather than intramolecular fluorine exchange process. 123 The large number of n.m.r. studies of intramolecular rate processes have included the following.Ring-inversion in 1,1-difluorocyclohexanes,' 24 ~ycloheptane,'~' cyclohep- tene,'25 and heterocyclic analogues of metacyclophane (17).'26 Internal rotation about the following bonds :phenyl-oxygen bond in phenolic esters,127 the ==C-N bond in enamines,'28~'29 the =N-N bond in hydra- A. Mannschreck A. Matthews and G. Rissmann J. Mol. Spectroscopy 1967,23 15. 120 H. S. Gutowsky J. Jonas and T. H. Siddall J. Amer. Chem. Soc. 1967,89,4300. N. S. Ham E. A. Jeffery T. Mole and S. N. Stuart Chem. Comm. 1967,254. D. N. Ford P. R. Wells and P. C. Lauterbur Chem. Comm. 1967,616. 123 E. L. Muetterties and W. D. Phillips J. Chem. Phys. 1967,46 2861. 124 S. L. Spassov D. L. Griftith E. S. Glazier K. Nagarajan and J. D. Roberts J. her. Chem. SOC.1967,89 88. R. Knorr C. Ganter and J. D. Roberts Angew. Chem. Internat. Edn. 1967,6 556. 126 I. Gault B. J. Price and I. 0.Sutherland Chem. Comm. 1967 540. 12' T. H. Siddall W. E. Stewart and M. L. Good Canad. J. Chem. 1967,45 1290. lZ8 A. Mannschreck and U. Koelle Tetrahedron Letters 1967,863. Y.Shvo E. C. Taylor and J. Bartulin Tetrahedron Letters 1967 3259. Physical A4 ethods of Structure Determination 19 zones,12s the N-0 bond in ON-diacylhydroxylamines (18),I3O the N-N bond in NN'-diacylhydrazines (19),' ' the aryl-nitrogen bonds in substituted and the N-N bond in heterocyclic nitro~amines.'~~ amide~,'~~ Studies of hindered rotation have also been carried out on NN-dimethylcarbamates,' 34 amides,' 35 thioamides,' 35 and nitroaromatic amines.' 36 Dimethylnitrosamine (20) has been studied in the vapour state at various temperatures to provide R ,co.,N-0 R"C0 separate estimates of the intra- and inter-molecular contributions to the barrier to internal rotation about the N-N bond.'37 The vapour-state results agree well with those observed in dilute solutions in carbon tetrachloride. Nitrogen-atom inversion in 1,2,6-trimethylpiperidine where the nitrogen inversion is found to be three orders of magnitude slower than that in tertiary acyclic arnine~.'~' An examination of the 'H n.m.r. spectrum of l-alkyl- aziridine (21) at various temperatures confirms that steric factors accelerate the inversion but by a much smaller amount than previously th0~ght.I~' Synchronous inversion at two bonded nitrogen atoms in ring compounds (22) has been further in~estigated.'~' Valency isomerism in 1,2-divinylaziridine (23) can be conveniently studied using n.rn.r.l4' 13' B.J. Price and 1. 0.Sutherland Chrm. Comm.. 1967 1070. 131 G. J Bishop B. J. Price and I. 0.Sutherland. Chrm. Comm.. 1967 672. 13' Y Shvo. E. C. Taylor. K Mislow. and M. Rahan. .I Amer rlicm SOC. 1967.89.4910 I" Y. L. Chow Angew. Chem. internat. Edn. 1967,6,75. 134 E. Lustig W. R. Benson and N. Duy J. Org. Chem. 1967,32 851. 13' R. C. Neuman D. N. Roark and V. Jonas J. Amer. Chem. SOC.,1967,89,3412. 136 J. A. Weil A. Blum A. H. Heiss and J. K. Kinnaird J. Chem. Phys. 1967,46 3132. 13' R. K. Harris and R. A. Spragg Chem. Comm. 1967,362. J. J. Delpuech and M. N. Deschamps Chem.Comm. 1967 1188. S. J. Brois J. Amer. Chem. SOC. 1967,89,4242. J. E. Anderson and J. M. Lehn 1.Amer. Chem. SOC. 1967,89,81. E. L. Stogryn and S. J. Brois J. Amer. Chem. SOC.,1967,89 605. J. Feeney Conformational Studies.-Conformational studies using n.m.r. have often relied on the fact that t-butylcyclohexyl systems when used as model com- pounds are assumed to give chemical shifts for the remote ring-protons which are the same as those in a related system without the t-butyl group. However this assumption has been shown to be in~0rrect.l~~ Thus attempts to obtain conformational parameters by relating averaged chemical shifts of a proton in a mobile cyclohexane system to those observed in the related fixed t-butyl- cyclohexyl derivatives may not always be justified.To study interconverting systems under conformational equilibrium one needs to estimate the parameters in the individual conformers. Some of the problems involved in this process can be circumvented by a method which combines dipole-moment data with coupling-constant data to obtain informa- tion about mobile ring systems.'43 This is based on the existence of a linear correlation between the squares of the dipole moments and the sum of the vicinal coupling constants JAx+ J, in a series of conformationally inverting ring compounds with similar geometry and similar polar substituents [for example a series of trans-1,2-dihalogenocyclohexanes(24)l. These problems do not arise when one can measure the parameters in the spectra of the indi- vidual conformers by freezing out the separate conformers.For example the conformational equilibria found in trans- 1,4-dibromo- trans-1,4-dichloro- and trans-1 -bromo-4-chloro-cyclohexane can be determined with confidence by integrating the separate axial and equatorial proton signals observed in the low-temperature spectra of the molecules. 144 The conformational preference of the trifluoromethyl group in cyclohexanes has been measured. 14' At -73" cis-4-methyl-1-trifluoromethylcyclohexane shows two CF fluorine absorption bands corresponding to CF,-axial ($cp3 66.63 p.p.m. from CFCl,) and CF,-equatorial (+cF3 74.5 p.p.m.) positions in the two frozen-out conformers. It is suggested that the difference between the two CF groups could be exploited in assigning the configuration of carboxy- substituents at centres of unknown stereochemistry in rigid molecules ; the suggested method consists of converting the carboxy-group into a trifluoro- methyl group using SF, which is an easy and selective process.The CF group thus formed would reveal its configuration through its "F chemical shift.',' In six-membered rings of type (25) the ratio of the average vicinal coupling S. Wolfe and J. R. Campbell Chem. Comm.,1967,872. 143 C. Altona H. R. Buys H. J. Hageman and E. Havinga Tetrahedron 1967,23,2265. 144 G. Wood and E. P. Woo Canad. J. Chem. 1967,45,2477. 14' E. W. Della J. Amer. Chem. Soc. 1967,89 5221. Physical Methods of Structure Determination J,,,, to the average value of J, is independent of the electronegativity of X and Y.146In molecules which adopt a perfect chair conformation this ratio has a value of ca.2.0. Deviations from this value have been used to indicate devia- tions to non-chair and distorted-chair conformations in a qualitative fashion. By examining cyclohexanols and pyranose derivatives in dimethyl sulphoxide solution stereospecific long-range coupling constants involving hydroxy- protons can be observed and used in conformational analysis.'47* 148 For trans-and cis-4-methylcyclohexanol,the respective values for JCHoHare 4.64 and 3.60 c./s~c.'~~ The small differences in the values render this method more difficult to use for conformational studies than some of the other methods. Conformational studies have been carried out on 1,1,4,4-tetramethyl-cyclohexanes,' 49 protonated methylcyclohexanones,' 50 tetrahydropyran,' 1,3-dioxan,'52 1,3-0xathiolans,'~~ cyclobutanes,' 54 cycl~butanols,'~~ and cyclo but ylamines ' ' and on hexahydro-1,3,5- trimethyl- 1,3,5-triazine-S- tri -oxan.Carbonium Ions.-A review article on n.m.r. studies of carbonium ions has ap~eared.'~' Olah and his co-~orkers'~~-'~~ have published a most extensive systematic study of protonated organic molecules all examined in the same protonating media. By using the strong acid system FSO,H-SbF dissolved in sulphur dioxide solution at -60° OlahlS8 and Wein~tein'~' with their co-workers have protonated ethers,' aliphatic alcohols 160 aldehydes,lS9 ketones,16' carboxylic acids,' 58 thiols and sulphides,162 aldimines and keti- mines,'63 and other organic molecules.When methyl alcohol is treated in 146 J. B. Lambert J. Amer. Chem. SOC. 1967,89 1836. 147 J. J. Uebel and H. W. Goodwin J. Org. Chem. 1966,31,2040. 14' J. C. Jochims G. Taigel A. Seeliger P. Lutz and H. E. Driesen Tetrahedron Letters 1967,4363. 149 R. W. Murray and M. L. Kaplan Tetrahedron 1967,23 1575. 150 T. D. J. D'SiIva and H. J. Ringold Tetrahedron Letters 1967 1505. G. Gatti A. L. Segre and C. Morandi J. Chem. SOC. (B),1967 1203. 15' J. E. Anderson F. G. Riddell and M. J. T. Robinson Tetrahedron Letters 1967,2017. 153 D. J. Pasto F. M. Klein and T. W. Doyle J. Amer. Chem. SOC.,1967,89 4368. 15' G. M. Whitesides J. P. Sevenair and R. W. Goetz J. Amer. Chem.SOC. 1967,89 1135. Is' I. Lillien and R. A. Doughty J. Amer. Chem. SOC. 1967,89 156. 15' H. S. Gutowsky and P. A. Temussi J. Amer. Chem. SOC. 1967,89,4358. 157 H. Cheradame and G. Mavel Ann. Chim. France 1966,1,449. Is' G. A. Olah and J. M. Bollinger J. Amer. Chem. SOC. 1967 89 2993; G. A. Olah and D. H. O'Brien ibid. p. 1725. M. Brookhart G. C. Levy and S. Winstein J. Amer. Chem. SOC. 1967,89 1735; G. A. Olah D. H. O'Brien and M. Calin ibid. p. 3582. 160 G. A. Olah J. Sommer and E. Namanworth J. Amer. Chem. SOC. 1967,89 3576. 16' G. A. Olah M. Calin and D. H. O'Brien J. Amer. Chem. SOC.,1967,89,3586. 162 G. A. Olah D. H. O'Brien and C. U. Pittman J. Amer. Chem. SOC. 1967,89,2996. 163 G. A. Olah and A. M. White J. Amer. Chem. SOC. 1967,89 3591 ;G. A.Olah and P.Kreien Buhl ibid. p. 4756. 22 J. Feeney this way the species CH,OH,+ (FOH2' -9.4 p.p.m.) is obtained because exchange rates are negligible at -60" in these systems one can measure the coupling JHH 3.6 c./sec. At temperatures above +60° the protonated alcohols break down to form carbonium ions (R++ H30+),and the kinetics of the cleavage have been studied by n.m.r. for several alcohols.160 For aliphatic aldehydes tiOH+ values for the protonated species are in the range -15 to -17.p.pm. In protonated acetaldehyde at -60° one can detect two forms of the species which differ in that the added proton can be either cis or trans to the hydrogen on the carbonyl carbon atom.'" For protonated ketones16' tiOH+ values in the range -13 to -15 p.p.m.are observed while for protonated acetone the values are tiOH+ -14.93 p.p.m. and J 1.0 c./sec. The parameters for protonated thiols'62 and alkoxy carbonium ions'58 are CH3SH2+,ti,, 6.45 p.p.m. JHH 8.0 c./sec.; CH30CH2+,tiCHz+ 9.94 p.p.m. J 1.0 c./sec. The long-range coupling J 1.0 c./sec. observed for CH30CH2+is much larger than the values found normally in ethers which suggests the formation of an sp2-centre as indicated in structure (26). a+,Me 8 Liquid-crystal Studies.-A liquid-crystal solvent in the nematic phase has many of its molecules aligned with respect to themselves in domains; in a magnetic field these domains are homogeneously ordered to give an even more ordered system. In 1963 Saupe and Englert'64 demonstrated that molecules dissolved in the nematic phase of a liquid-crystal solvent are themselves aligned to a high degree in the magnetic field.The n.m.r. spectra of molecules examined in this way are substantially different from normal n.m.r. spectra and it is possible to obtain information which is not available in normal solvents where we do not see any anisotropic magnetic interactions (those dependent on orientation of the molecules with the magnetic field). The observed absorption bands are well resolved because although the molecules are ordered they do not lose their translational motion which averages out intermolecular dipole- dipole interactions. Absorption bands extending over a range of several thousand cycles per second are observed and these can be explained in terms of direct dipole-dipole intramolecular interactions which we do not see normally in a high-resolution n.m.r.spectrum. Molecules such as cyclopropane which have magnetically equivalent nuclei and show only a single absorption when examined in an isotropic solvent give rise to complex spectra when examined in the nematic phase of a solvent such as 4,4'-di-n-hexyloxyazoxy-benzene.'65 From a detailed analysis of the spectrum'65 one can extract the absolute signs and magnitudes of JHHck +9.5 & 1 and JHHtrans +55 & 1.0 164 A. Saupe and G. Englert Phys. Rev. Letters 1963 11 462. 165 L. C. Snyder and S. Meiboom J. Chem. Phys. 1967,47 1480. Physical Methods of Structure Determination c./sec. The 13CHsatellite spectra in the nematic-phase spectrum have also been analysed.By assuming the symmetry of cyclopropane to be D, and that the C-C bond length is 1.510A one can calculate the C-H bond length 1.123 and the HCH angle 114.4".In addition to obtaining information concerning relative bond lengths bond angles absolute signs and magnitudes of indirect spin-spin coupling constants dipole-dipole interactions and chemical-shift anisotropies one can also use the technique for studying the nature of the nematic phase of liquid-crystal solvents. Excellent review articles dealing with the theory and application of liquid-crystal solvents in n.m.r. spectroscopy have been written.'66-'68 Here we will be content to examine the requirements for obtaining such spectra and the extent of the new information available therein.A typical liquid-crystal material which has been used is p-azoxyanisole which has a temperature range for the nematic phase of 118-135"c. To obtain high-quality spectra it is necessary to eliminate temperature gradients in the sample and to maintain the temperature constant to +0.1" if possible. It is also desirable to ensure that the sample is homogeneous with regard to its composition. A significant advance in the experimental technique was provided by the discovery that eutetic mixtures of various liquid-crystals can be chosen such that the mixture is in the nematic phase at the normal probe operating temperat~re.'~'~ For example a 4:1 mixture of (27) and (28) has a melting point of 29"and a clearing point of 78".16' (28) It is usual to examine the samples in the non-spinning condition because rapid rotation would destroy the molecular alignment with the magnetic field.However it has been shown recently"' that slow rotation (5-10 c./sec.) appears to improve resolution without interfering with the other properties of the spectrum. A new kind of nematic phase has been described recently. This is formed from a mixture of c8 or Cl0 alkyl sulphates the corresponding alcohol sodium sulphate and water in the proportions 40 :5 :5 :50 respec-ti~e1y.l~~ The nematic-phase temperature range is from 10 to 75"c and it is 166 A. D. Buckingham and K. A. McLauchlan Progr. N.M.R. Spectroscopy 1967,2 16' G.R. Luckhurst Quart. Rev. 1968 in the press. 16* G. R. Luckhurst &err.Chem.-Ztg 1967,68 113. 169 H. Spiesecke and J. Bellion-Jourdan Angew. Chem. Internat. Edn. 1967,6,450. 170 D. Demus 2.Naturforsch. 1967,22a 285. E. E. Burnell and C. A. de Lange personal communication. K. D. Lawson and T. J. Flautt J. Amer. Chem. SOC.,1967,89 5489. J. Feeney possible to spin the samples to improve resolution. Some of the systems which have been examined recently by the liquid-crystal technique are cyclopro- pane,165 cy~lobutane,'~~ (JHHet + 10.4,JHHtrnns +4.9c./sec.) methyl fluoride'74 (J13cHis shown to be absolutely positive J13cFnegative JHFyem positive) and ethyl iodide' " (the spectrum was analysed using a modified composite- particle technique). Saupe and his co-workers have examined 3C-labelled a~etonitrile,'~~ methyl iodide,176 methan01,'~~ acetylene,177 and several acetylenic compounds1 77 by this method.Buckingham and Burnell' 78 have studied chemical-shift anisotropies in oriented molecules. They recommend using molecules of high symmetry (CF, CH, SF,) as internal reference materials and advocate caution in interpreting small shifts observed on passing from nematic to isotropic phase of liquid-crystal solution. Bernheim and Krugh' 79 found that they could determine "F chemical-shift anisotropies with reasonable accuracy but it was more difficult in the case of protons. Spectral Assignments from Nuclear Overhauser Effects.-In 1965 Anet and Bourn'" illustrated how nuclear Overhauser experiments can be used to determine which nuclei in a molecule are in close spatial proximity to each other by virtue of their being connected through an intramolecular spin-lattice relaxation mechanism.For this experiment to be successful it is necessary that the intramolecular effects should dominate the relaxation ;thus intermolecular relaxation effects must be minimised by efficient degassing of the system and by using solvents which do not contain high-magnetic-moment nuclei. For two protons A and B in close spatial proximity (the intramolecular relaxation is inversely dependent on the sixth power of the internuclear distance) if this relaxation mechanism is the dominant one then when one saturates protons A in a double-resonance experiment there is an increase of as much as 50 % in the integrated intensity of the band for the B nuclei.Thus if the assignment of absorption A is certain then it is possible to infer the assignment for the B nuclei. This technique provides an additional powerful aid in spectral assignment and is becoming used more widely. For example Nouls and his co-workers18 ' have examined the n.m.r. proton spectrum of a mixture of the two isomers (29) and (30)by this method. Irradiation of (29)at the CH frequency causes a 30 % increase in intensity of HA which identifies the absorptions for both CH and HA protons in (29) where the nuclei are in close spatial proximity. This experi- ment was successful even though trifluoroacetic acid was used as the solvent. 173 S. Meiboom and L. C. Snyder J. Amer. Chem. SOC.,1967,89 1038. 17* R.A. Bernheim and B. J. Lavery J. Amer. Chem. SOC.,1967,89 1279. C. M. Woodman Mol. Phys. 1967,13,365. 17' A. Saupe G. Englert and A. Povh 'Advances in Chemistry Series' 1967 vol. 63 p. 51. G. Englert A. Saupe and T. P. Weber 2.Naturforsch. 1968,22a 152. A. D. Buckingham and E. E. Burnell J. Amer. Chem. SOC.,1967,89,3341. 179 R. A. Bernheim and T. R. Krugh J. Amer. Chem. SOC., 1967,89,6785. 180 F. A. L. Anet and A. J. R. Bourn J. Amer. Chem. SOC., 1965,87,5250. J. C. Nouls G. Van Binst and R. H. Martin Tetrahedron Letters 1967,4065. Physical Methods of Structure Determination Similar experiments have been used to distinguish between not only stereo- isomers but also individual conformers of a single compound. For example 7-methoxy-7,12-dihydropleiadeneexists as a 2:1 mixture of axial and equa- torial conformers at -2W (31)and (32).Ig2When the C-12hydrogen (axial) in (32) is irradiated the C-7axial proton signal increases in intensity by 27%.Similar experiments on (31)give no increase in intensity. H OMe Hx: \ Woods and his co-worker~'~~ have also reported work on the ginkgolides (33)where irradiation at the t-butyl proton frequency causes the absorptions of the I J E and F protons in close spatial proximity to increase their intensity and thus confirms the assignments. This experiment was successful even though the solvent was trifluoroacetic acid; it is suggested that the bulky nature of the t-butyl group isolates the protons I J E and F from close approach of the solvent molecules.Novel Techniques.-The problem of resolution enhancement has received considerable attention during the last year. Four methods of measuring a small long-range coupling constant in 3-bromothiophen-2-aldehyde have been c~nsidered.'~~~ 18' (i) Trial and error fitting of the observed multiplet profile by a set of component lines each having the line-shape taken from a known single in the spectrum. (ii) A linear transformation (referred to as convolution) of the observed curve using a function which has such a form that it enhances resolution at the expense of sensitivity. (iii) Examination of Fourier transform lS2 J. G. Colson P. T. Lansbury and F. D. Saeva J. Amer. Chem. SOC.,1967,89,4987. "' M. C. Woods I. Miura Y. Nakadaira A. Terahara M.Maruyama and K. Nakanishi Tetra-hedron Letters 1967,321. R. R. Ernst personal communication. R. R. Ernst R. Freeman B. Gestblom and T. R. Lusebrink Mol. Phys. 1967,13,283. J. Feeney spectra. (iv) An analogue transformation taking place within the spectrometer subtracting a suitable portion of the second derivative from the slow-passage absorption signal. Freeman and Gestblom'86 have discussed anothu method of measuring very small coupling constants using a double-resonance technique ; this is based on the fact that two nuclear sites within the same molecule will experience almost exactly correlated local fields due to field G-ihomogeneity. Double- resonance effects which result from interactions on the molecular scale can be used to take advantage of this fact.Freeman and Gestblom' 87 have shown that fine-structure can be transferred from one resonance absorption to another using double-resonance techniques. If one of the X-lines in an AX-type spectrum is irradiated with a weak field (yxH2/27c 4I JAxI) then each of the A-lines is split into a doublet (spin- tickling); if there is any fine-structure on the X resonance due to coupling to a third group of nuclei Z and if yxH,/2x + I Jxz(,the same fine-structure is observed on the two components of the A-doublet except that the splitting is reduced by a factor of 2. This method has been used to detect very small long-range coupling constants ( +0.05 c./sec.). Reports have appeared concerning further extensions to the method of Hoffman and Forsen' l6 for measuring kinetic processes using double-resonance techniques.When this was used to study the intramolecular reversible slow process in p-nitrosodimethylaniline (34) on irradiation at the H-3 Me M,Me absorption frequency not only the signal from the H-5 proton collapses but also those from H-2 and H-6. This is because H-5 and H-6 have almost identical chemical shifts and being strongly coupled lose their identity as separate nuclei; H-2 and H-6 are interchanged by rotation about the N-aryl bond which explains why H-2 also disappears.'88 Comparatively few papers dealing with high-field (220 Mc./sec.) proton spectra have appeared. '89-1 ' The conformational differences between native and denatured forms of ribonuclease lysozyme and cytochrome C have been investigated.' 89 The authors detected high-field resonances (6 0.7 p.p.m.) in the folded native forms which are absent in the denatured forms because of absence of ring-current effects on the shielding of a certain CH proton in the la6 R.Freeman and B. Gestblom J. Chem. Phys. 1967,47,2744. la' R. Freeman and B. Gestblom J. Chem. Phys. 1967,47 1472. I. C. Calder P. J. Garratt and F. Sondheimer Chem. Comm. 1967,41. lag C. C. McDonald and W. D. Phillips J. Amer. Chem. SOC., 1967,89 1967. C. C. McDonald W. D. Phillips and J. Lazar. J. Anzer. Chem. SOC.,1967,89 4166. 191 N. S. Bhacca and D. Horton Chem. Comrn. 1967 867. Physical Methods of Structure Determination denatured form where the nucleus is no longer in close proximity to an aro- matic ring.In the 220 Mc./sec. 'H spectrum of single-stranded deoxyribo- nucleic acid (DNA),there are two separate thymine methyl absorptions (one for those near to a purine and the other for those near to a pyrimidine neighbour in the 5'-po~ition).'~' By examining enantiometers such as alkylarylcarbinols in optically active solvents the absolute configurations of several enantiomers have been assigned.Ig2 Deuterium resonance has been used to study the mechanism of nucleophilic substitution reactions. 193 35Cl ions have been used as chemical probes in molecules of biological interest by observing 35Cl line-width vari-ations under different condition^."^ This latter method should prove to be a useful technique for investigating biological systems.Burgen and his co-workers have studied line-width variations in 'H spectra to investigate hapten- antibody complex formation. lg5 Miscellaneous Studies.-The limitations of the sub-spectral method of analysis have been discussed in some detail.'96 '97 Hydroxychromenes (39-437) have been characterised by observing the effects of acetylation on the chemical shifts of neighbouring proton^."^ Thus only in structures (35) and (37) does acetylation cause deshielding of the peri-Ha by 0-3-0-4 p.p.m. From examination of an extensive series of 3-aminoacrylic esters (38) J,, proton-proton coupling constants ranging from 7-5 to 15.0 c./sec. have been observed.' 99 Some other compounds examined by n.m.r. were cyclopropanes200-202 192 W.H. Pirkle and S. D. Beare J. Amer. Chem. SOC.,1967,89 5486. 193 L. K. Montgomery A. 0.Clouse A. M. Crelier and L. E. Applegate J. Amer. Chem. SOC. 1967,89,3453. 19' R. L. Ward and J. A. Happe Biochem. Biophys. Res. Comm. 1967,28,785. 195 A. S. U. Burgen 0.Jardetzky J. C. Metcalfe and N. Wade-Jardetzky Proc. Nar. Acad. Sci. U.S.A. 1967,58,447. 196 P. Diehl and D. Trautmann MoZ. Phys. 1966,11,531. 19' P. Diehl and P. Weissenhorn Helv. Chim. Acta 1967,50 143. 198 A. Arnone G. Cardillo L. Merlini and R. Mondelli Tetrahedron Letters 1967,4201. W. Bottomley J. N. Phillips and J. G. Wilson Tetrahedron Letters 1967 2957. 'O0 D. T. Longone and A. H. Miller Chem. Comm. 1967,447. '01 K. L. Williamson and B. A. Braman J. Amer. Chem. Soc. 1967,89,6183.'O' J. Lee C. Parkinson P. J. Robinson and J. G. Speight J. Chem. SOC. (B) 1967,1125. 28 J. F eeney RHN ,R' H/C=C;R" (38) (the shielding effects of alkyl substituents have been summarised) dia~oles,~'~ tetrazole~,''~ pyra~oles,~'~ triaz~les,~'~*''~ methyl silane derivatives,206 nucleo~ides,~~~ substituted styrenes,2 polyurethanes,208 quin~xalines,~'~ helix-coil transformations in polypeptides,' and histidine residues in proteim2l2 203 G. B. Barlin and T. J. Batterham J. Chem. SOC.(B) 1967 516. '04 W. Freiberg C. F. Kroger and R. Radeglia Tetrahedron Letters 1967 2109. '05 C. L. Habraken H. J. Munter and J. C. P. Westgeest Rec.Trao.chim. 1967,86 56. E. A. V.Ebsworth and S. G. Frankiss Trans. Faraday SOC. 1967,63 1574. 207 R.J. Cushley. K. A. Watanabe and J. J. Fox J. Amer. Chem. SOC. 1967,89,394. '08 E. G. Brame R. C. Ferguson and G. J. Thomas Analyt. Chem. 1967,39,517. 209 P. J. Brignell A. R. Katritzky R. E. Reavill G. W. H. Cheeseman and A. A. Sarsfield J. Chem. SOC.(B) 1967 1241. 0.Gurudata J. B. Stothers and T. D. Talman. Canad. J. Chem. 1967,45 731. 2'1 J. A. Ferretti Chem. Comm. 1967 1030. J. H. Bradbury and P. Wilairat Biochem. Biophys. Res. Comm. 1967,29 84.