NRUCTURAL WD§E$ OF ALDEHYDES ANID QXEME O-METHYL ETHERS EY NEfiCLEAR MAGNETEC RESONANCE SPECTRQS‘COPY Thesls for H1. Daqru of pk. D. MECHLGAN STATE UNIVERSETY Nelson C. T. Hsi 1966 Q__¥___,A___. ____‘ ”Jun“ .a: Y LIBRARY ‘ meals ‘ Michigan State . '. University r! I b This is to certify that the thesis entitled STRUCTURAL STUDIES OF ALDEHYDES AND OXIME O-METHYL ETHERS BY NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY presented by Nelson C. T. Hsi has been accepted towards fulfillment of the requirements for Ph . D . degree in Chemi Stry @f/(qugj Major professor Date November 1, 1966 0-169 ABSTRACT STRUCTURAL STUDIES OF ALDEHYDES AND OXIME O-METHYL ETHERS BY NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY by Nelson C. T. Hsi Nuclear magnetic resonance Spectroscopy was applied to the study of semiquantitative conformational analysis in aliphatic aldehydes and oxime O-methyl ethers. The time averaged Spin—spin coupling constants between the aldehydic and arprotons of eighteen substituted acetal- dehydes were studied as functions of temperature and solvent. Interpretation of the data in terms of rotamers I and II, whereby a single bond eclipses the carbonyl group, led to the 0 O H\ I R ,x ' \H I, ' H R H I II following conclusions. (1) Monosubstituted acetaldehydes: in the absence of solvent when R is methyl, ethyl, n-propyl, gramyl, isopropyl or phenyl, II is favored over I by AH0 of 800, 700, 600, 500, 400 and about 500 cal./mole, reSpectively. When R is tfbutyl, II is less stable than I by an enthalpy of Nelson C. T. Hsi 250 cal./mole. The ratio I/II increases with increase in solvent polarity, except for phenylacetaldehyde where it decreases. (2) Disubstituted acetaldehydes: in the absence of solvent when both substituents are methyls or only one of the substituents is methyl, the more stable rotamer, enthalpy-wise, has the methyl eclipsing the carbonyl (II); when neither substituent is methyl, I is the more stable rotamer. The ratio I/II increases with increase in solvent polarity. (5) Cycloalkanecarboxaldehydes: when the ring is cyclohexyl, II is more stable; when it is cyclopentyl, I is slightly more stable, when it is cyclobutyl, II is slightly more stable; and when it is cyclopropyl, I is much more favored. Again the ratio I/II increases with increase in solvent polarity. Conformations and configurations were assigned to several aldehyde and ketone oxime O-methyl ethers from coupl- ing constant and chemical shift studies. Interpretation of the data for the syn isomers (methoxy gig to the aldehydic proton) of the aldehyde derivatives in terms of rotamers III and IV led to the conclusion that for both mono- and N-OCH3 -OCH3 H R , H H III IV disubstituted acetaldehyde derivatives, III is energetically Nelson C. T. Hsi favored. Interpretation of the data for the app; isomers led to the conclusion that, whereas V is the only signifi- cant rotamer for the disubstituted acetaldehyde derivatives, both VI and VII are equally important for the monosubstituted acetaldehyde derivatives. H3CO-N H3CO-N H3CO-N V VI VII Conformational analysis of the aldehyde derivatives by means of comparison of chemical shifts further substantiated the conclusions reached from the coupling constant studies. Conformations of the ketone derivatives were also assigned and discussed on a qualitative basis by comparing chemical shifts. The effect of benzene on the chemical shifts of these derivatives was interpreted in terms of Specific association between the solvent benzene and the solute. From the syn/333i isomer ratios, it was concluded that there is no meaningful correlation between group size and isomer stability. STRUCTURAL STUDIES OF ALDEHYDES AND OXIME O-METHYL ETHERS BY NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY BY Nelson C. T. Hsi A THESIS Submitted to Michigan State University K in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1966 '\\ .\\ To My Parents ii ACKNOWLEDGMENT The author wishes to express his sincere appreciation to Dr. G. J. Karabatsos for his patience, guidance and friendship during the course of this investigation. Appreciation is extended to the Atomic Energy Com- mission for financial support from June, 1963 to September, 1964 and to the Lubrizol Oil Company for a fellowship from September, 1964 to September, 1965. iii TABLE OF CONTENTS INTRODUCTION . . . . . . . . . . . . . . . . . . RESULTS. . . . . . . . . . . . . . . . . . . . . A. Aliphatic Aldehydes. . . . . . . . . . . B. Oxime O-Methyl Ethers. . . . . . . . . . I. Chemical Shifts . . . . . . . . . . II. gyannti Isomers. . . . . . . . . . III. Spin-spin Coupling Constants. . . . DISCUSSION . . . . . . . . . . . . . . . . . . . A. Aliphatic Aldehydes. . . . . . .'. . . . I. Monosubstituted Acetaldehydes . . . II. Disubstituted Acetaldehydes . . . . III. Cycloalkanecarboxaldehydes. . . . . (a) Cyclohexanecarboxaldehyde. . (b) Cyclopentanecarboxaldehyde. . (c) Cyclobutanecarboxaldehyde . . (d) Cyclopropanecarboxaldehyde. . IV. Consideration of Other Conformations. B. Oxime O-Methyl Ethers. . . . . . . . . . I. Conformations of the Syn Isomers. . II. Conformations of the Anti Isomers . (a) Monosubstituted acetaldehyde derivatives . . . . . . . . (b) Disubstituted acetaldehyde derivatives . . . . . . . . . III. Chemical Shifts . . . . . . . . . . (a) Solvent effects . . . . . . . (b) Conformations . . . . . . . . IV. gyngnti Isomers. . . . . . . . . . EXPERIMENTAL . . . . . . . . . . . . . . . . . A. Reagents and Compounds . . . . . . . . B. Solvents . . . . . . . . . . . . . . . C. Synthesis. . . . . . . . . . . . . . . . I. Cyclopropanecarboxaldehyde. . . iv Page (NCNN I‘POOUTU'I (II P 45 45 45 47 50 50 51 52 52 54 57 57 59 59 61 61 62 63 65 67 67 67 67 67 TABLE OF CONTENTS - Continued Page II. Cyclopentanecarboxylic Acid. . . . . . . 68 III. N,N-Dimethyl Cyclopentanecarboxamide . . 69 IV. Cyclopentanecarboxaldehyde . . . . . . . 70 V. N,N-Dimethy1 Cyclobutanecarboxamide. . . 71 VI. Cyclobutanecarboxaldehyde. . . . . . . . 71 VII. N,N-Dimethy1-5,3-dimethyl-butyramide . . 71 VIII. thutyl acetaldehyde . . . . . . . . . . 71 IX. Oxime O-methyl Ethers. . . . . . . . . . 72 D. N.M.R. and U.V. Spectra . . . . . . . . . . . 72 REFERENCES......................74 TABLE II. III. IV. VI. VII. VIII. IX. XI. XII. LIST OF TABLES Spin-spin Coupling Constants of Aldehydes Solvent Effects on J Relative Population of Aldehydic Rotamers Solvent Effect on the Relative Populations of Aldehydic Rotamers. H AHO Values for 4%//’fl\‘\H ' R AGge as Function of Solvent Chemical shifts (T-Values) of Oxime O-Methyl Ethers. . . . . . . ocis - 0 (A6). trans O-methyl Ethers . V . in benzene in c.p.s., Syn and Anti Percentages and AGZO Values for CHCHO O of Aldehydes. in p.p.m., of Oxime - V in carbon tetrachloride of Oxime O-Methyl Ethers . Syn-——+hAnti of Oxime O-Methyl Ethers . Ultraviolet Spectra of Some Oxime O-Methyl Ethers in Cyclohexan Spin4spin_Coupling Constants of Neat Liquid H\: //OCH3 e . . vi (Av). Page 11 12 18 19 21 25 28 31 35 55 LIST OF TABLES - Continued TABLE XIII. XIV. XV. XVII. XVIII. Effect of Solvent Polarity on J of HlHQ H OCH3 N/ O O O O O O O O O O R1R2CHQ Spin-spin Coupling Constants of Neat Liquid N// . . . . . . . . . . . . . H Effect of Solvent Polarity on J of HlHa R1R2CHQ' N/OCH3 H H //OCH3 Rotamer population of . . . R1R2CHQ OCH3 OCH3 AHO for N// // H R H H )///”\\\H R/ I H Boiling Points of Oxime O-Methyl Ethers . . . vii Page 36 37 38 42 43 73 INTRODUCTION Nuclear magnetic resonance Spectroscopy has been success- fully applied for the study of quantitative or semiquantita- tive conformational analysis in mobile systems under conforma- tional equilibration (1,2,3,4,5,6). However, the success of this approach relies on the availability of conformational models from which the nuclear spin coupling and/or chemical shift parameters for the various possible rotamers are derived. In the absence of such models, the problem of conformational analysis is reduced to qualitative, or at best, semiquantita- tive level, for the necessary parameters must be estimated for the hypothetical conformational models from theoretical or empirical relationships or be obtained from "frozen out" rotamers at reduced temperatures. Conformational analysis of a compound can be successfully studied by employing either the time averaged coupling con- stants or the chemical shift parameters of appropriate nuclei. The work of Gutowsky and co—workers in substituted ethanes (6) amply illustrates such a case. The energetically favored forms, or rotamers, of substituted ethanes are the staggered configurations Ia, Ib and Ic. The experimentally observed H1 H1 H]. A / ,3 H2 A A B /\ H2 x ~ Y xfix Y x Y H2 B A Ia Ib Ic coupling constant, which is the time averaged coupling con- stant of Ia, Ib and lo is given by equation (1), where Jt Jobsd = xIa Jt + XIb Jg + xIc Jg (1) is the trans coupling (dihedral angle between Hl-C-C' and C-C'-H2 planes is 1800), Jg the gauche coupling (dihedral angle is 600) and the X's are the respective mole fractions. The factors governing the overall appearance of the n.m.r. spectra include the relative energies of the rotational isomers, the potential barriers to internal rotation about the C-C bond, and the chemical shifts and coupling constants characteristic of each rotamer. These quantities can be ob- tained most completely and directly for a compound if the potential barriers are high enough such that the n.m.r. spectrum at low temperatures is a superposition of Spectra for the various rotamers. However, rotational averaging invariably occurs in most cases, which simplifies the spectrum but re- duces its information content. Nonetheless, by careful esti- mation of Jt and J9 from theoretical and/or empirical relation- ships, the relative stabilities of the three rotamers have been calculated from the temperature dependence of the coupling constant. Rotational isomerism about carbon-carbon single bond has been extensively studied for saturated hydrocarbons and sub- stituted ethanes in the liquid phase. For example, AHO trans --+'gauche is about +800 cal./mole for nfbutane (7), +500 cal./mole for n-hexane (7). +730 cal./mole for 1,2-dibromo- ethane (8,9) and -900 cal./mole for 1,1,2,2-tetrabromoethane (10). There is, however, only very limited information on rotational isomerism involving a tetrahedral carbon bonded to a trigonal carbon. Therefore, the purpose of this research deals with investigations directed toward elucidation of the relative stabilities of rotamers IIa and IIb as func- tions of X, Y and R by n.m.r. IIa IIb Several investigations have shown that the stable con- formation of a tetrahedral carbon bonded to a trigonal carbon is IIc, whereby a single bond (C-R) eclipses the C=X IIc double bond. These include Raman and infrared studies on chloroacetone (11), haloacetyl halides (12,13) and N—methyl- chloroacetamide (14); microwave studies on acetaldehyde (15), propionaldehyde (16), acetyl chloride (17) and propene (18); electron diffraction studies on aliphatic ketones (19) and aldehydes (20,21); and nuclear magnetic resonance studies on propionaldehyde (22) and olefins (23,24,25,26,27). Furthermore, the coupling constant of propionaldehyde has been found to be temperature dependent. On the basis of these evidences, rotational isomerism in aliphatic aldehydes (X=O) and oxime O-methyl ethers (X=NOCH3) were studied, using the same general approach as in substituted ethanes. The oxime O-methyl ethers represent an interesting case where both rotational and configurational isomerism can be studied simultaneously. They are suitable for the elucidation of the relative stabilities of IIa and 11b as functions of X, Y and R. Furthermore, because of configurational isomerism about the carbon-nitrogen double bond, they are also suitable models for studying the relative stabilities of 11a and 11b not only when the tetrahedral carbon is gi§_to the lone pair of electrons, but also when it is gig to the methoxy group. In the latter case, IIa and IIb may be sufficiently destabil— ized to make 111a and IIIb competitive in stability with IIa and IIb. Cyclopropanecarboxaldehyde (21) and ethyl a,ardi- fluoro- and d,drdichloroacetates (28) have been reported to have two-fold rather than three—fold barriers to rotation. $H3 N O \ IIIa IIIb RESULTS A. Aliphatic Aldehydes Table I summarizes the coupling constants at various temperatures between the aldehydic proton and the drprotons of all the aliphatic aldehydes investigated. All coupling constant values were measured at 50 c.p.s. sweep width and were averages of several measurements with an accuracy of .i0.03 c.p.s. To ensure internal consistency values were always checked against the coupling constant of acetaldehyde, O and -300, respectively 2.85, 2.88 and 2.90 c.p.s. at 38°, 0 (22,29). The coupling constants of these aliphatic aldehydes except those of tfbutylacetaldehyde, di-tfbutylacetaldehyde and cyclopropanecarboxaldehyde are smaller than that of acetaldehyde. The coupling constants of monosubstituted acetaldehydes increase with increase in temperature, except that of Efbutylacetaldehyde, which decreases with increase in temperature and that of phenylacetaldehyde which is temperature independent. The coupling constants of disubStituted acetal- dehydes vary as follows: (1) when one of the substituents is methyl, the coupling constants increase with increase in temperature, and (2) when neither substituent is methyl, the coupling constants decrease with increase in temperature. Table I. Spin-Spin Coupling Constantsa of Aldehydes J , Aldehyde -500 0O CHCHg8° 70O CH3CHO 2.90 2.88 2.85 MeCHZCHO 1.08 1.22 1.51 EtCH2CHO 1.42 1.55 1.89 1.80 '2—PrCH2CHO 1.51 1.60 1.75 1.80 ngmCH2CHO 1.48b 1.58 1.75 1.78 .g-prcngcno 1.81 1.88 1.92 2.05 .E-BuCH2CHO 2.95 2.94 2.92 2.84 censcnacno 2.18 2.20 2.24C 2.40b 2.40b 2.40b 2.45b (Me)2CHCHO 1.01 1.12 1.17 1.55d (Et)2CHCHO 2.52 2.55 2.58 2.25 (g-Bu)2CHCH0 8.20b 8.00b 5.75b Me(Et)CHCHO 1.58 1.80 1.87 1.70 Me(g-Pr)CHCHO 1.45 1.59 1.78 1.75 Me(C6H5)CHCHO 1.07 1.25 1.51 1.45 Et(g-Bu)CHCHO 2.70b 2.55 2.52 2.55 DPCHO 8.14 5.95 5.75 5.55 <>»CHo 1.72 1.82 [3»080 2.11 2.12 2.12 2.05 {)scno 0.92 1.05 1.14 1.15 aUnless otherwise denoted all coupling constants are those of neat solutions; values in c.p.s. bAbout 10% solution in carbon tetrachloride. CValue at 900. dValue at 80°. The coupling constants of cycloalkanecarboxaldehydes vary as follows: (1) those of cyclobutyl and cyclohexyl increase with increase in temperature, (2) that of cycloprOpyl de- creases appreciably with increase in temperature and (3) that of cyclopentyl is almost temperature independent. Table II summarizes the effect of solvents on the coupl- ing constants of several aldehydes. The coupling constants increase with increase in solvent polarity, except that of acetaldehyde, which shows only small variations and that of phenylacetaldehyde, which decreases with increase in solvent polarity. The vicinal proton-proton coupling constant depends on several parameters and a qualitative estimate of the trends to be expected has been made by Karplus (30,31). According to the predications of valence bond calculations (30,32,33), the vicinal coupling constant depends on the dihedral angle 0, the hybridization, the HCC bond angles and the C-C bond length. It is certainly reasonable to assume that the changes of coupling constants observed for the aliphatic aldehydes result mainly, if not exclusively, from the changes of the dihedral angle. The temperature dependence of the coupling constant indi- cates that the relative stabilities of the various rotamers of a substituted acetaldehyde can be at least qualitatively assessed. Assuming Jt > Jg, where Jt iS the traps coupling constant (dihedral angle of vicinal protons 1800) and J9 the Table II. Solvent Effects on JCHCHO of Aldehydes f- JHH,C'p°S'a j Aldehyde Cyclohexane Nitrobenzene Acetonitrile CH3CHO 2.79 2.85 2.87 MeCH2CHO 1.25 1.50 1.55 thuCH2CHO 2.80 2.95 5.05 (Et)2CHCHO 2.25 2.40 2.55 Me(Et)CHCHO 1.85 1.70 1.78 Et(QfBu)CHCHO 2.40 2.80 2.70 DPCHO 5.05 5.80 5.80 [>»CHo 1.97 2.15 2.50 (:fcno 1.00 1.15 1.20 C5H5CH2CHO 2.40 2.18 2.00 aAll values are at 360. gauche coupling constant (dihedral angle 600). the observed coupling constant which is an average of the contributions from all the rotamers should be independent of temperature if IVa, IVb and V are energetically equivalent. If V is H1 H2 R H H H // z/ [I H; R’ H’l R H1 H2 IVa IVb V more stable than IVa, the coupling constant Should increase with increase in temperature; and if less stable, it should decrease. Similarly, for disubstituted acetaldehydes and cycloalkanecarboxaldehydes, the observed coupling constant Should be independent of temperature if VI, VIIa and VIIb (also, VIII, IXa and IXb) are energetically equivalent. If VIIa is more stable than VI (also IXa more stable than VIII), the coupling constant should increase with increase in temperature and if less stable, it Should decrease. H R1 R2 VI VIIa VIIb 10 H / CH2 /CH2 (CH2)n H (CH2)n H // H \ // // CHé ’ CH2 0 m m\ CH2 H O m N :3 VIII IXa IXb Table III summarizes per cent populations of the various rotamers of substituted acetaldehydes. Table IV Shows their dependence on solvent. Values for per cent populations of the various rotamers for monosubstituted acetaldehydes were calculated from equation 2, where x is the per cent population of IV and Jobsd = X(Jt + Jg)/2 + (1 - x)J (2) g (1 - x) that of V. Values for disubstituted acetaldehydes and cycloalkanecarboxaldehydes were calculated from equation 3, where y is the percent population of VI (also VIII) and = yJ + (1 - y)J (5) Jobsd t g (1 - y) that of VII (also IX). Solutions for these values require prior knowledge of the value of both Jt and J9. Evaluation of Jt and J9 could be achieved by the following approach. Equation 4 expresses the coupling constant of acetaldehyde at all temperatures and Jobsd = 1/5 (Jt + 2 J9) (4) 11 Table III. Relative Population of Aldehydic Rotamersa H b /‘ ,>c’fi‘\H % \ o o R o Aldehyde —30 0 36 70 MeCHECHO 23 31 34 EtCH2CHO 57 40 45 48 grPrCH2CHO 39 42 46 48 QfAmCH2CHO 59C 41 48 47 ifPrCH2CHO 48 50 51 55 thuCH2CHO 80 79 79 77 csnscngcno 85C 85C 64C(58) 660(60) (Me)2CHCHO 19 20 21 25d (Et)2CHCHO 40 57 57 58 (trBu)2CHCHO 92C 89C 85C Me(Et)CHCHO 28 27 28 Me(anr)CHCHO 25 27 29 Me(C5H5)CHCHO 19 22 25 25 Et(QfBu)CHCHO 42C 40 40 57 D~CHO 91 88 85 80 (yam 28 50 [DrCHO 54 54 54 55 CDrCHO 17 19 20 21 aUnless otherwise indicated these values are those of neat solutions. bThe remaining per cent correSpondS to the rotamer having the R group eclipsing the carbonyl CAbout 10% solution in carbon tetrachloride. dValue from 600. 12 Table IV. Solvent Effect on the Relative Populations of Aldehydic Rotamers H % ’ R Cyclohexane Acetonitrile Aldehyde 58° 58° MeCH2CHO 33 35 EfBuCH2CHO 78 85 (Et)2CHCHO 58 40 Me(Et)CHCHO 27 29 Et(QrBu)CHCHO 58 42 D»CH0 78 88 [>PCHO 52 57 (:7 GHQ 19 21 CSHSCH2CHO 85 55 13 also of substituted acetaldehydes at very high temperatures (approaching free rotation) or at all temperatures if the various rotamers should happen to be energetically equivalent. Assuming tfbutylacetaldehyde exists exclusively in IV, equation 5 expresses its coupling constant. Solution of t If these values are correct, then the maximum value of the equation 4 and equation 5 gives J = 3.1 and J9 = 2.7 c.p.s. Jobsd = 1/2 (Jt + Jg) (5) coupling constant for any substituted acetaldehyde would be equal to 3.1 c.p.s. It is obvious, therefore, that these values are incorrect, Since the coupling constant for diet- butylacetaldehyde is 6.2 c.p.s. Assuming di-tfbutylacetal- dehyde exists exclusively in VI, Jt should have a value of 6.2 c.p.s. and Jg a value of 1.2 c.p.s. These values repre- sent the lower and upper limits reSpectively. It is incorrect to assume that Jt and J9 are the same for acetaldehyde, monosubstituted acetaldehydes and disub- stituted acetaldehydes, as substitution of an alkyl group for a proton decreases the coupling constant. For example, while the coupling constant of ethane (34) is 8.0 c.p.s., those of propane (35) and isobutane (36) are only 7.3 and 6.8 c.p.s. Considerations based on electronegativity have Shown that the substitution of an alkyl group for a proton Should decrease the coupling constant by 0.3 to 0.5 c.p.s. (37,38). It can be shown that in aliphatic aldehydes each 14 alkyl (also phenyl) substituent decreases the average coupl- ing constant by about 0.4-0.5 c.p.s. For example, while the coupling constant of acetaldehyde (temperature independent) is 2.85 c.p.s., that of phenylacetaldehyde (monosubstituted and temperature independent) is only 2.40 c.p.s., and that of cyclopentanecarboxaldehyde (disubstituted and temperature independent) is about 2.1 c.p.s. When the coupling constants of various disubstituted acetaldehydes are plotted against temperature, as shown in Figure 1, the lines converge on extrapolation at high temperatures (approaching free rotation) around about 2.0 c.p.s. ratherfthan 2.8 c.p.s. The difference of about 0.8 c.p.s. thus represents the combined effect of both alkyl substituents on the average coupling constant. In other words, each alkyl substituent decreases the average coupling constant by about 0.4 c.p.s. A more rigorous and direct approach involves the Simul- taneous evaluation of Jt’ Jg and AH0 for each substituted acetaldehyde. For monosubstituted acetaldehydes, these quantities could be evaluated from eq. 7 and for disubstituted acetaldehydes from equation 9. Keq (monosubstituted) = 2 (1ex)/x (6) AH0 = -RT ln(Jt-Jg- 2 Jobsd)/(Jobsd- Jg) (7) Keq (disubstituted) = (1 - y)/2y (8) AH0 = -RT ln 1/2(Jt - Jobsd)/(Jobsd - Jg) (9) This approach requires A80 = 0 for the equilibrium between the various rotamers. This assumption may be correct if the .omomoummzy!.m ecu homomofimmmovmz .m Nomomofimvmz .8 Nomodmno .m h.omomodfirfi .m uomomoflsmnevum .8 .e mesmem 15 “DOV B 05a oma 0m 0m ow omu a O. mu m G .u mu 0 w I“ 1N fim.m.ov (llhu mmh .0 an an an mu nu AV mv an a .v .n 16 substituent is a relatively small group, such as a methyl; it could hardly be true when the substituent is a larger alkyl group, e,g,, tfbutyl group, as rotation of this group would be much more hindered in V than in IV. The difficulties involved with the exact solution of equations 7 and 9 for each substituted acetaldehyde have necessitated the use of a much Simpler, though less rigorous, approach to this problem. If Jt and Jg for acetaldehyde could be ascertained, it is possible to use these values for all substituted acetaldehydes by correcting JObsd for the effect of an alkyl substituent; namely, 0.4 c.p.s. The values that give the most consistent results are Jt = 7.6 and Jg = 0.5 c.p.s. (calculated from equation 4). A J of t 7.6 c.p.s. for acetaldehyde is certainly reasonable. This value is chosen for the following reasons: (1) Jt should have a lower limit of 7.0 c.p.s. (6.2 + 0.8), as the highest coupling constant value for di-tfbutylacetaldehyde is 6.2 c.p.s. and (2) a 7.7 c.p.s. coupling constant is observed with 0,8-unsaturated aldehydes (39), which supposedly exist in the sjtgang conformation. The data in Tables III and IV were calculated from equation 10 for monosubstituted acetal- dehydes and equation 11 for disubstituted acetaldehydes using 0.4 c.p.s. as the correction factor for each alkyl or aryl substituent. + 0.4 = x (Jt + Jg)/2 + (1 - x)Jg (10) + 0.8 = th + (1 - y)Jg (11) Jobsd Jobsd 17 Table V summarizes the enthalpy differences, calculated from plots of log K.y§ 1/T, between individual rotamers: e,g,, V.y§ IVa, VI yg VIIa and VIII ys IXa. Table VI sum- marizes the effect of solvent on the free energy difference at 360 between such individual rotamers. For disubstituted acetaldehydes where R1 # R2, AH0 and AGO values were calcu- lated assuming VIIa and VIIb were equivalent. These values are therefore only meaningful for serving as a bSSiS for comparison. The accuracy of AH0 and AGO values depends on the values chosen for Jt' J9 and the correction factor for substituent effect. To obtain an estimate on this accuracy, these values were calculated as functions of J , J9 and the substituent t effect. The results were as follows: (1) with the substituent corrections set at 0.3 and 0.5 c.p.s., these values increased and decreased by about 5%, and (2) by changing Jt from 7.2 to 8.0 c.p.s., they varied by approximately 10%. Other factors that might affect the accuracy of these values are experi- mental errors, changes in the dielectric constants of liquids with temperature and contributions from excited states and torsional oscillations. Therefore an error of.i30% seems to be a reasonable upper estimate of the accuracy of these values. H ///fl\\\ R Table V. AHO Values for ‘> £1 -—4> H /,, ee——-’/, R H Aldehyde AHO cal./mole a MeCH2CHO -800 EtCHZCHO —700 .n-PrCHQCHO -600 anmCH2CHO -500 .i-PrCH2CHO -400 .E-BuCHZCHO +250 CSHSCH2CHO -500 (0b) (Me)2CHCHO —500 (Et)2CHCHO +250 (EfBu)2CHCHO +1100b Me(Et)CHCHO —200C Me(QéPr)CHCHO -200C Me(C6H5)CHCHO -4OOC Et(anu)CHCHO +500C [>»CH0 +1500 <>»CH0 -150d [3»CH0 rvo {:fcno -400 Unless otherwise stated these are values of neat solutions. From about 10% solution in carbon tetrachloride. These values were calculated as if R; = R2. QJOU'OJ Calculated from only two temperatures. 19 Table VI. A636 as Function of Solvent , H‘s— IR H 4883 cal./mole Aldehyde Cyclohexane Acetonitrile MeCHZCHO —880 -820 _t-BuCH2CHO +550 +550 (Et)2CHCHO +70 +180 Me(Et)CHCHO -180 -120 Et(§rBu)CHCHO +150 +250 [>-CH0 +1100 +1500 [>ecno -50 +90 {ijH0 —480 -380 C6H5CH2CHO -50 -540 20 B. Oxime O-Methyl Ethers I. Chemical Shifts Table VII summarizes the chemical shifts of the oxime O-methyl ethers when in neat liquid, in benzene, or in carbon tetrachloride. The notation used to distinguish various protons is shown in Xa and Xb. Each proton is referred to as H1 R1 NonvOCHg er~0CH3 CH -CH CH 6 a B Xa Xb -CH C1. .gig or trans with respect to the methoxy group. For simplicity, the following convention is used throughout the text: the EYE isomer has the methoxy group gi§_to the smaller R group, e,g,, the syn isomer of propionaldehyde oxime O-methyl ether has structure XI and that of 2-butanone oxime O-methyl ether 0CH3 OCH3 / / N f CHs-CH2’////fl\\\\‘H CHs-CH2’//L\\CH3 XI XII structure XII. Similarly the any; isomer has the methoxy group trans to the smaller R group. Assignments of protons as gig or trans are based on the unequal isomer distribution for unsymmetrical oxime O-methyl ethers. Using the accepted concepts of steric effects, the more intense of the two Signals was assigned to the syn isomer. Umscflucoo 88.8 88.8 88.8 88.8 8800 mwv 88.8 88.8 88.8 88.8 8882 fiwv 08.8 80.8 88.8 8880 m_8xmmovoimo 88.8 88.8 88.8 8.888 mfimxmmovodmo 08.8 08.8 8882 mfimxmmovogmo 88.8 88.8 88.8 88.8 88.8 88.8 8880 mxemommovmo 88.8 08.8 08.8 08.8 88.8 8800 mxmmommovmo 88.8 88.8 88.8 88.8 88.8 88.8 8882 NAmmommovmo 88.8 88.8 88.8 88.8 88.8 88.8 88.8 8880 mxmmovmo 88.8 88.8 88.8 88.8 88.8 88.8 88.8 88.8 8800 81880880 88.8 88.8 88.8 88.8 88.8 88.8 . 88.8 88.8 8882 81880880 88.8 88.8 88.8 88.8 88.8 8880 8880880 88.8 88.8 88.8 88.8 88.8 88.8 8800 8880880 88.8 88.8 88.8 88.8 88.8 88.8 8882 8880880 88.8 88.8 80.8 88.8 88.8 8880 mxmmovommo 88.8 88.8 88.8 88.8. 88.8 88.8 «800 8188080880 88.8 88.8 88.8 88.8 88.8 88.8 8882 818080880 88.8 88.8 88.8 88.8 88.8 88.8 8880 880 88.8 88.8 88.8 88.8 88.8 88.8 8.888 880 88.8 88.8 88.8 88.8 88.8 88.8 8882 880 mcmuu 880 mcmuu M80 mamuu mac mcmuu 880 mcmuu 880 mcmuu 880 ucm>aom mm w mmoo 8880888 Amwwcm 8880888 8880888 88 dmoozuomm8 muozum Hmnuozlo mEAxO mo AmoSHm>I.8V muMHnm amoeamnu .HH> 8898 22 Umscflucoo 88.8 88.8 88.8 80.8 88.8 88.8 88.8 88.8 umdz 81880080 88 88.8 80.8 88.8 88.8 88.8 88.8 8880 8880880 88 88.8 88.8 88.8 88.8 08.8 88.8 8800 8880880. 88 88.8 88.8 88.8 88.8 88.8 88.8 8882 8880880 88 88.8 88.8 88.8 88.8 88.8 08.8 8880 8888000880 88 88.8 88.8 88.8 88.8 88.8 88.8 8800 8888000880 88 88.8 88.8 88.8 88.8 88.8 88.8 ummz 8888000880 88 88.8 88.8 80.8 88.8 88.8 88.8 88.8 8880 880880 88 08.8 88.8 88.8 88.8 88.8 88.8 88.8 8800 880880 88 88.8 88.8 88.8 00.8 88.8 88.8 88.8 8882 880880 88 88.8 88.8 88.8 8880 880 88 88.8 88.8 88.8 8800 880 88 88.8 08.8 88.8 8882 880 88 88.8 88.8 88.8 8880 AMHHWV 88.8 88.8 88.8 88.8 8800 AMHHWV 88.8 88.8 88.8 88.8 8882 AHHHV 88.8 88.8 08.8 88.8 88.8 08.8 8880 flHva 88.8 88.8 08.8 88.8 88.8 88.8 8882 WHHV 88.8 88.8 88.8 88.8 8880 .va mcmuu mHU mcmuu mHU mommy mHU mcmuu mHU mcmnu 880 mcmuu 880 ucm>Hom mm 8 8800 Ammovmm Amowdm Ammoqdm 8880888 88002wommd 880088000 u 88> 8888 23 88.8 80.8 88.8 88.8 88.8 88.8 8880 8880 8888008 mm.8 88.8 88.8 08.8 88.8 mm.8 8800 mmmo mammovm 88.8 80.8 88.8 88.8 88.8 88.8 8882 8880 8888008 88.8 88.8 88.8 88.8 88.8 08.8 08.8 88.8 8880 8880880 8888008 88.8 88.8 88.8 08.8 08.8 08.8 88.8 8800 8880880 8188008 88.8 m8.m 88.8 88.8 mm-8 88.8 8m.8 om.8 umoz mm808mo 8883008 88.8 88.8 88.8 88.8 88.8 8880 88880080 mxmmove 08.8 mm.m 88.8 mm.8 No.8 8800 mAmmova mfimmovm 88.8 88.8 88.8 88.8 88.8 8882 8A8movmo 8188008 88.8 88.8 80.8 88.8 88.8 8880 880880 88088 88.8 88.8 88.8 88.8 88.8 8800 880880 88088 88.8 88.8 88.8 88.8 88.8 8882 880880 88088 88.8 80.8 88.8 8880 8880 88 88.8 80.8 88.8 8800 8880 88 88.8 80.8 88.8 8882 8880 88 88.8 88.8 80.8 88.8 88.8 88.8 88.8 88.8 8880 88880080 88 om.m 88.0 88.8 No.8 mm.8 88.8 - mw.m om.m 8800 mAmmovmo 88 mcmuu mHU mcmuu Lmflo mgmuu mHU msmuu mHU mcmuu mHU mean» 880 ucm>aom mm 1 8800 8880888 Aroma: 8880888 8880888 88 dmooznomml 880088800 u 88> 8888 24 The assignments for symmetrical compounds were made to conform to those of the unsymmetrical ones. For more detailed dis- cussion, refer to reference (40) and previous papers in that series. The chemical shifts, calculated from first order Spectral analysis, are accurate to i0.03 p.p.m., except those of ethyl, isopropyl, cyclopentyl and diethyl carbinyl groups, whose accuracy is less. Table VIII summarizes the differences in chemical Shifts of gig_and trans protons, A0, which are accurate to.i0.001 p.p.m. A positive A0 means the gig protons resonate at higher fields than the trans protons; a negative the reverse. The pertinent points are: (1) In neat liquid H1 resonates at lower fields when gig than when trans to the methoxy, iig., A0 is negative (”/-0.08 p.p.m.); in benzene solution, A0'S become more negative except that of acetaldoxime O-methyl ether. (2) In neat liquid, demethyl protons resonate at slightly higher fields when gig than when trans, i,g,, A0 is positive, in benzene solution, the signals cross over, iig., A0 is negative. (3) In neat liquid, ormethylene and or methine protons resonate at appreciably lower fields when gig than when trans, i,g,, A0 is negative. A0 values for kaethine protons are comparable in magnitude to those of H1, whereas those for oemethylene protons are smaller. In benzene solution, A0 values are about the same for aemethylene protons but become more negative for armethine protons. (4) In neat 25 085c8ucoo 80.01 80.01 80.01 80.0+ 880880880 880 00.0 80.01 80.0+ 80.0+ 88.01 80.01 80.0+ mmono 800 80.01 80.0+ 880 880 80.0+ 80.0+ 80.0- AMva 0 80.0+ 80.0+ 88.0- 88.0- 88.0- 88.0- mmv 8 80.0+ 80.0+ 88.8- 80.8- flwv 8 00.0 80.0+ 88.81 88.01 88.0- 8008880008800800000 m 00.0 00.0+ 00.81 08.01 00.01 00.01 888mOmmOVmu m 00.0 80.0+ 08.0- 88.01 88.0-80088088800008880000 m 00.0 80.0+ 80.0+ 80.0+ 88.01 88.0- 88.0- 88.0- 8008808880000 8 80.0+ 80.0+ 80.0+ 80.0+ 88.0- 88.0- 88.0- 88880000 8 80.0+ 80.0+ 08.0- 88.0- 08.0- 8880800 8 80.0+ 80.0+ 88.01 88.01 88.01 8880000800 8 80.0+ 80.0+ 88.0- 08.01 88.01 88880080880 8 80.0+ 80.0+ 88.0- 88.0- 800888800 8 80.0+ 80.0+ 88.0- 880880880 0 00.0+ 00.01 mmommo m 00.01 00.0 00.01 80.0+ 80.01 00.01 8mu m 8880 8882 8880 8882 8880 8882 8880 8882 8880 8882 8880 8882 mm 88 088800080 8880-0080 88018080 888018080 8880-8088 888088 88002108888 m 888088 880882-0 888x0 88 ..E.Q.m :8 mcmuu .1880 8 . HHH> GHQMB 26 .8m 00 880 mmoo mc8>mn 005008 0:0 mo 08030 000 @9000 hxonume 0:0 00m 80580> 880 .88800210 \1 'l. Q .mmum>mn 0:0 m>808mom “00000 030 c050 @8088 0800cmmfi 00308 0 0m 8000commu c00oum hxocumE 0:0 00 080 030 0030 c005 mws8m> m>8000020 08.0- 08.01 00.0- 00.0+ 08.0- 88.01 0800. 00080080 80.0+ 80.0+ 88.0+ 88.0+ 80.0- 88.0- 08.0- 08.0- 0800080 00080080 80.0+ 80.0+ 80.0- 88.01 00080080 00080080 88.0- 88.0- 0800 080080080 08.0- 88.0- 0800 080080 00.0+ 00.0+ 08.01 08.01 080000 080080 08.01 08.0- 0800 080 00.0 00.0 08.0+ 80.0+ 80.81 08.01 80.01 00.0 08030080 000 80.01 80.01 08.01 88.01 80.01 80.01 8300080 080 80.01 80.01080.01080.01 88.01 88.01 88.01 80.0+ mfimmovommo 080 80.0- 80.0- 00.01 80.0+ 00080080080 080 0800 0002 0800 0002 0800 0002 0800 0002 0800 0002 0800 0002 08 0m 00800000 0080-8080 08080008 0080-0000 8080-0004 088084 08002u00848 J 000880000 - HHH> 00008 27 liquid and in benzene solution, B-methyl protons resonate at higher fields when gig than when trans, 1,3,, A6 is positive. Table IX summarizes the Av values (Av = v in benzene - v in carbon tetrachloride). A positive Av means that the proton resonates at a higher field in benzene than in carbon tetra— chloride; a negative, the reverse. The most striking feature of the data is the lower field absorption of so many protons E? in benzene, which generally causes upfield shifts, than in carbon tetrachloride. The pertinent points for subsequent dis- cussions are: (1) benzene shifts both gig and trans H1 down- 5 g field, except those of acetaldehyde and cyclopropanecarboxalde- hyde derivatives; gi§_protons are shifted more than trans. (2) Benzene shifts both gi§_and ££§n§_ormethyl protons upfield, except the gig protons of the methyl Efbutyl ketone derivative. (3) Benzene shifts gig ormethylene and Q8methine protons down- field, whereas it shifts the trans protons upfield, with the exception of the trans ofimethine proton of the difEbetYl‘ acetaldehyde derivative. (4) Benzene shifts both gig and trans B-methyl protons upfield, except the gi§_protons of the ethyl group of the ethyl tfbutyl ketone derivative. (5) Benzene shifts methoxy protons downfield. The half widths, 0.6—0.9 c.p.s., of Sléle differ notic- ably from those of EgaggeHl, 1.2-1.7 c.p.s., in the tempera- ture range -500 to 900. Similar broadening observed in formal- doxime O-methyl ether was attributed to incomplete quadrupole washout of JHCN (41). 28 085880800 0.00 8.08+ 8.8+ 080 088 8.8- 8.08- 0.0- AHHV 8 0.81 0.8- 8.8+ 8.8+ flvv m 0.8- 0.01 0.88- 0.0008000080 8 8.81 0.0- O801858.88- 8.8- 0800008008080080080 8 8.8- 8.8- 8.8+08.88- 8.8- 8.81 08080080080 8 0.01 0.01 0.01 08008000800808080080 8 8.8- 8.8- 08.8+00.8- 8.81 8.8- 0800800080080 8 8.8- 8.8- 8.88+ 8.08+ 8.8+ 8.881 0.81 00080080 8 0.0- 8.8- 0.0+ 8.8- 8.8- 0800080 8 8.81 8.81 0.8+ 8.8- 8.81 0008000080 8 8.8- 8.8- 00.0+ 8.81 8.81 00080080080 8 8.81 0.01 00.88+ 8.8- 8.8- 080000800 8 0.01 8.8- 08.8+ 0.8- 080080080 8 8.81 080080 8 0.8- 8.8- 8.08+ 8.88+ 0.0+ 0.88+ 080 8 88000 880 88000 880 88800 880 88800 880 88000 880 88800 880 mm 88 008000>< 0080-80>< A80l00>< A08olo0>< 0080800>< 008V>< 08002n00888 |. '- 8080083080000 £00080 c8 0888880 88 808£0m 8%308210 mE8XO 00 ..8.m.0 c8 .880QV >1 .3 .NH 0808H 29 .0080 > 8008803 88 >0 .8808>00 0:0 m>800mmc “08880 00:08: 00 88 0888880 88 808080880 00:0 808E 88580> 8>808mom0 ~08000>< 0.8- 8.8- 8.08+ 8.88+ 8.81 0.0- 0800 00080080 0.0- 8.8+ 8.8+ 8.8+ 0.0 8.81 0800080 08080080 8.8- 0.0+ 0.8- 00080080 00080080 8.81 0.81 0.01 0808000 080080 0.8- 8.8+ 8.8+ 0.8+ 8.81 080080 080080 0.81 0.01 8.8+ 0800 080 0.01 8.8+ 0.01 0008000 080 8.81 0.01 0.0+ 8.8+ 0.8+ 8.81 8.8+ 8.8+ 08080080 080 0.8- 8.81 8.8+ 0.0- 8.8+ 8.8+ 0800080 080 0.81 0.81 0.01 0.81 8.88+ 8.8+ 0808000080 080 8.08- 0.81 8.8+ 0.8+ 8.8+ 0.0+ 8.8+ 080080 080 mmmmw 880 88000 880 80000 880 88000 880 88000 880 88000 880 mm 88 0080-80>< A80io0>< 0080100>< 0080l00>< 0880>< 080o2u00888 005080800 - 80 08008 50 II. Syn-anti Isomers Table X summarizes syn and Egg; percentages (accurate to 15%) and the free energy differences between these isomers at 400. Assuming the stabilities of these geometric isomers are mainly controlled by the steric factors of the sub- stituents, it seemed desirable to establish a relative scale 7 12:1! of "effective" group size that could be applied to configura- tional isomerism about carbon-nitrogen double bonds. Unfortu- nately, the data failed to give any meaningful correlation between group size and isomer stability. For example, from the syn/gag; ratio for acetaldoxime O-methyl ether methyl would have to be effectively smaller than hydrogen. From the aldehyde series ethyl would be smaller than any other alkyl group except methyl, benzyl, and cyclopropyl, yet from the ketone series it would have to be larger than all other ‘n-alkyl groups including ne0pentyl. Similarly, from the methyl ketone series, phenyl would be larger than isopropyl, yet from direct competition between the two groups, the reverse would have to be true. Table XI summarizes the ultraviolet spectra data of several oxime O-methyl ethers. Alkyl phenyl oxime O-methyl ethers were chosen, since the derivatives of the aliphatic carbonyl compounds show no strong absorption above 220 mu. The results show that as the alkyl group was varied from methyl to ethyl to isopropyl, and the percentage of the gigrphenyl isomer increased from 2% to 16% to 61%, both A Xand 6 ma decreased. 51 . 0 Table X. Syn and anti Percentages and AG4O Values for syn-—V anti of Oxime O-Methyl Ethersa R1R2C=NOCH3 Percent Percent AGED R1 R2 gyg_b yang; (Kcal/mole) H CH3 48 52 -0.06 H CH2CH3 54 46 +0.10 H CH2CH2CH3 61 39 0.28 H (CH2)5CH3 58 42 0.20 H CH2CH(CH3)2 58 42 0.20 H CH2C(CH3)3 64 56 0.56 H CH2C8H5 51 49 0.02 H CH(CH3)2 76 24 0.71 H CH(CH3)CH2CH3 71 29 0.55 H CH(CH3)CH(CH3)CH2CH3 69 51 0.48 H CH(CH2CH3)2 71 29 0.55 H CH(CH2CH3)(CH2)3CH3 65 55 0.58 H CH[C(CH3)3]2 100 o <:] 54 46 0.10 <::] 68 52 0.46 H <::>> 74 26 0.65 m CH3 CH2CH3 81 19 0.90 CH3 CHECH2CH3 72 28 0.58 CH3 CH2CH(CH3)2 74 26 0.64 CH3 CH2C(CH3)3 76 24 0.71 CH3 CH2C6H5 71 29 0.56 CH3 CH(CH3)2 86 14 1.1 CH3 C(CH3)3 100 CH2CH3 CH(CH3)2 65 57 0.55 CH2CH3 C(CH3)3 100 CH3 CeHs 98 2 2.4 09 "_L 7....— *v—v—trr ‘. ‘30‘1 o 1“ .' Continued 32 Table X - Continued R1R2C=NOCH3 Percgnt Percent AGED R1 R2 gyg_ anti (Kcal/mole) CH2CH3 CeHs 84 16 1.0 CH2CH2CH3 C6H5 80 20 0.85 CH(CH3)2 06H5 59 61 -0.28 CH2C6H5 CH(CH3)2 60 40 0.25 aData from neat liquids. bSyn is the isomer having the methoxy group cis to R1. 35 Table XI. Ultraviolet Spectra of Some Oxime O-Methyl Ethers in Cyclohexane R1R2C=NOCH3 xmax’ mu 8 x 105 R1 R2 H C6H5 265 13.8 CH3 C5H5 252 10.8 CH2CH3 C6H5 248a 8.5 CH2CH2CH3 C5H5 248a 7.7 CH(CH3)2 ceH5 256b 5.0 aWeak shoulder at about 262 mu. bWeak shoulder at about 247 mu. 34 III. Spin-Spin Coupling Constants Table XII summarizes the coupling constants between proton H1 and the drprotons for the syn isomer of the oxime O-methyl ethers at various temperatures. All values were averages of several measurements with an accuracy of $0.03 c.p.s. To ensure internal consistency, values were always checked with the coupling constant of acetaldehyde. All coupling constants decreased with increase in temperature, except that of acetaldoxime O-methyl ether which remained constant. The coupling constant of cyclopropanecarboxaldehyde oxime O-methyl ether experienced the largest decrease, about 15% in the range -300 to 900. Table XIII summarizes the effect of solvent on the coupl- ing constants for the syn isomers. Increase in solvent polarity increased the coupling constants, except that of the phenylacetaldehyde derivative, which decreased with increase in solvent polarity and that of the acetaldehyde derivative, which showed only small variations. Table XIV summarizes the coupling constants between pro— ton H1 and the afiprotons for the app; isomers of the oxime O-methyl ethers at various temperatures. Table XV summarizes the effect of solvent on the coupling constants of these isomers. Several features of the data are worth noting and comparing with those of the data for the syn_isomers. (1) There is an abrupt increase in the coupling constant in changing from the monosubstituted acetaldehyde derivatives .80008008 8000000000080 800000 R00 0 60080 35 88.8 08.8 88.8 80.8 80.8 80.8 AHHV 00.8 88.8 80.8 88.8 00.0 08.0 AND 000;. 080.0 008.0 080.0 080.8 A 00.0 88;. 00.0 88.0 08.8 00.8 A 00.00 80.00 80.00 80.00 88.00 000.00 0008000 0008000 00.0 08.0 00.0 88.0 08.0 00.8 080000800 080080 00.8 80.0 08.0 88.0 80.0 88.0 080080 080080 80.8 08.8 08.8 88.8 00.0 80.0 0800800080080 080 08.8 80.8 80.8 88.8 08.8 00.0 080080 080 08.8 08.8 00.8 80.8 80.8 080 080 00.8 08.8 88.8 88.8 08.8 8800 8 08.8 08.8 08.8 88.8 88.8 88.8 0008000 8 08.8 00.8 00.8 08.8 88.8 00.8 00080080 8 08.8 88.8 08.8 88.8 00.8 00.8 080000800 8 08.8 08.8 00.8 80.8 080080 8 80.8 08.8 08.8 08.8 080 8 08.8 08.8 08.8 8 8 I .n 008 000 008 000 00 008 08 8 0800808 6 H A.8.Q.UV m Eh L\ \\ZHHHAHW 0800 08 0800808 I\ z 0800\\ I|l|//// 008000 0082 00 808008800 08000800 8008:8008 .HHN 8000B H8 36 Table XIII. Effect of Solvent Polarity on JHlHQ of H1 //OCH3 N R1R2CHQ H1 OCH3 ::>z:::N// R R CH /——— JHlHa(c.p.s.)——~\ 1 2 0 R1 R2 Cyclohexanea Acetonitrilea H H 5.88 5.90 H CH3 5.70 5.90 H CH2CH3 5.95 6.15 H (CH2)4CH3 6.00 6.15 H CH(CH3)2 6.50 6.50 H C(CH3)3 6.90 6.90 H C6H5 6.65 6.55 CH3 CH3 5.80 6.05 CH3 CH2CH3 6.60 6.95 CH3 CH(CH3)CH2CH3 7.00 7.55 CH2CH3 CH2CH3 7.55 7.80 CHZCHg (CH2)3CH3 7.65 8.05 C(CH3)3 C(CH3)3 10.55 10.45 [:> 7.50 8.50 [:>> 6.50 7.05 <:::> 5.80 6.15 a10% solutions at 400. 57 00.0 80.0 00.0 0N.> 8N.0 08.0 .AHHHV 80.8 08.8 88.8 00.0 80.0 08.0 .AHHQ 88.8 00.8 80.8 00.0 00.0 00.0 .AHQ 80.8 00.8 80.8 80.8 88.8 88.8 080008800 080030 88.0 00.8 00.8 80.8 8N.8 88.8 mmommo 080880 88.0 00.0 00.0 00.8 00.8 80.8 0800000000000 000 80.0 08.0 08.0 88.0 80.0 80.0 mmommU 800 80.0 0N.0 88.0 88.0 08.0 000 080 80.8 00.8 08.8 08.8 00.8 08.8 8880 m 80.8 80.8 00.8 80.8 00.8 88.8 8008000 E 00.8 08.8 88.8 00.8 08.8 80.8 00080080 8 08.8 88.8 00.8 08.8 80.8 00.8 800000000 m 80.8 80.8 08.8 80.8 00.8 000880 8 80.8 88.8 80.8 88.8 080 m 08.8 08.8 08.8 m m 000 000 008 000 00 008i mm 08 H8 0.8.0.000m0mh .\L\ .ZHHRAW. 0000\\ .0800808 008 ZIIJ\\. 005000 0082 00 800008000 00000300 0008:0008 .>HX 80008 0800\\ IlL// 0800808 58 Table XV. Effect of Solvent Polarity on J of HlHa R1R2CHQ' //OCH3 >2:N . H1 RlR2CHCI /OCH3 /\:N r JHlHa(C.p.S.)—_—'\ H1 R1 R2 Cyclohexane Acetonitrile (400)a (40°)a H H 5.70 5.60 H CH3 5.50 5.50 H CH2CH3 5.60 5.50 H (CH2)4CH3 5.60 5.60 H CH(CH3)2 5.65 5.55 H C(CH3)3 6.10 6.00 H C6H5 5.50 5.70 CH3 CH3 7.50 7.40 CH3 CH2CH3 7.70 7.80 CH3 CH(CH3)CH2CH3 8.10 8.25 CH2CH3 CH2CH3 8 . 50 8 . 4O CH2CH3 (CH2)3CH3 8.20 8.40 <::] 8.80 9.25 /r__ \ 7.00 7.50 <:::> 7.20 7.40 a10% solutions. 39 to the disubstituted acetaldehyde derivatives. (2) Whereas the coupling constant of the disubstituted acetaldehyde derivatives decreased with increase in temperature, those of the monosubstituted acetaldehyde derivatives behaved irregularly. (3) The coupling constants of the disubstituted acetaldehyde derivatives increased with increase in solvent polarity. The coupling constants of the monosubstituted acetaldehyde derivatives, however, decreased slightly or re- mained unchanged, except that of the phenylacetaldehyde derivative, which increased with increase in solvent polarity. The fact that the coupling constants for the anti isomers of the disubstituted acetaldehyde derivatives behaved similarly to those of the §y2_isomers, whereas those of monosubstituted acetaldehyde derivatives did not, is important in subsequent discussions of the conformations of the §y2_and ag£i_isomers. 0n the basis of the same considerations applied to the aldehydes, the relative stabilities of the various rotamers for the gyg_isomer of a substituted acetaldehyde derivative can be qualitatively assessed from the temperature dependence of the coupling constant. Equation 12 expresses the per cent Jobsd = p(Jt + Jg)/2 + (1 — p)Jg (12) populations of the various rotamers for the syn isomer of the monosubstituted acetaldehyde derivatives, where p is the per cent population of XIII and (l-p) that of XIV. Similarly, equation 13 expresses the per cent populations of the 4O N-OCH3 N-OCH3 N-OCH3 H1 H2 R / H , H /, H H2 ” R” Hl/ R H1 H2 XIIIa XIIIb XIV disubstituted acetaldehyde derivatives, where q is the per cent Jobsd th + (1-q)Jg (13) population of XV and (1—q) that of XVI. N-OCHg N-OCH3 N‘OCH3 H R1 R2 H H H I I / R1” R2” H” R2 H R1 xv XVIa XVIb Applying the same type of arguments used for the alde- hydes, these values are calculated by evaluating Jt and J9 for acetaldoxime O-methyl ether. This could be done as follows. Assuming di-tfbutylacetaldehyde oxime O-methyl ether exists exclusively in XV, then its J is J . J could obsd t Q then be calculated from equation 14 which expresses the coupl- ing constant of acetaldoxime O-methyl ether. These values can be checked by assuming tfbutylacetaldehyde oxime O-methyl Jobsd = 1/5 (Jt + 2J9) (14) ether also exists exclusively in XIII. From equation 15 Jobsd = 1/2 (Jt + Jg) (15) 41 which expresses its coupling constant and equation 15, Jt and J9 can be calculated. However, this evaluation involves the incorrect assumption that Jt and J9 are the same for acetal- dehyde, monosubstituted acetaldehyde and disubstituted acetaldehyde derivatives. The error thus introduced could be reduced by applying a correction factor for each alkyl or aryl substituent, just as in the case of the aldehydes. Using a correction factor of 0.4 c.p.s. for each substituent, the values Jt = 11.5 and J9 = 5.2 c.p.s. were obtained from both tfbutyl and di-tfbutyl acetaldehyde oxime O-methyl ethers. The small variation of the coupling constants of these two compounds at low temperatures and the insensitivity of the coupling constants to solvent polarity supports the assumption that they exist mainly in XIII and XV (hydrogen eclipsing the carbonyl). Table XVI summarizes per cent populations of the various rotamers for the syn isomer of substituted acetaldehyde derivatives. These values were calculated from equation 16 for monosubstituted acetaldehyde derivatives and equation 17 for disubstituted acetaldehyde derivatives. + 0.4 p(Jt + Jg)/2 + (1—p)Jg (16) Jobsd + 0.8 = th + (1-q)J (17) Jobsd 9 Table XVII summarizes the enthalpy differences that were calculated from plots of log K.y§ 1/T. These values are prob— ably accurate to.i50%. In addition to errors of about 5-10% 42 .0000000000000 000000 00 00000000 000 000 mm mm 00 m0 m0 00 A“ v 00 m0 08 08 08 00 _ v mmm mmm 000 mm mam va mm 00 mm mm 00 00 flwv 0m mm 0m 0m mm mm mxmmovo mxmmovo mm om mm mm 00 00 mmomxmmov 000000 00 mm mm 00 00 mm mmommo mmommo mm «m «m 08 mm 00 mmommoxmmovmo «mo 00 om om 00 mm 00 000000 «00 N0 00 00 00 00 mmo mmo m0 mm mm mm mm mmmo 0 0m 00 mm 000 000 mxmmovo m 00 0m 00 mm mm mm mxmmovmo 0 00 m0 m0 m0 00 00 mmovxmmov m 00 00 m0 om 000000 0 80 «0 00 00 000 m I N 0. om 000 com 00¢ 00 com 0 m m\\ m 0 m - 00 m m R . m \ZHA m000:0 m”000 0 000000 00 0000000000 00E00om .0>X 0000B /0CH3 /0CH3 N N H Table XVII. AHO for Ei///fl\\0H -—-—9 H / H RCHgCH=NOCH3 AHO(cal/mole) R CH3 +580 CH2CH3 +570 (CH2)4CH3 +590 0v500a) CH(CH3)2 +650 C(CH3)3 +4,500 C6H5 +1.200 R1R2CHCH=NOCH3 R1 R2 CH3‘ CH3 +500 CH2CH3 CH2CH3 +700 C(CH3)3 C(CH3)3 +4,400 4 O O +1,200 ovaooa) +780 +450 aFrom 5% solution in carbon tetrachloride. 44 that are introduced by experimental uncertainties in Jobsd and temperature control, appreciable and presently unde- terminable errors may be introduced by disregarding the contri- butions to Jobsd from torsional oscillations and excited t all monosubstituted and disubstituted acetaldehyde derivatives. vibrational states and by using only one set of J and J9 for DISCUSSION A. Aliphatic Aldehydes I. Monosubstituted Acetaldehydes When R is methyl, ethyl, gfpropyl, gfamyl or isopropyl, V (alkyl eclipsing the carbonyl) is more stable than IVa or IVb (hydrogen eclipsing the carbonyl). When R is tfbutyl, V is less stable than IVa or IVb. The 800 cal./mole enthalpy difference between IVa and V when R is methyl is comparable to the 900 cal./mole difference obtained by microwave (16) and 1000 cal./mole difference obtained by nuclear magnetic resonance spectroscopy (22). Wilson and Butcher (16) proposed XVII as the structure for the most stable rotamer for XVII propionaldehyde. Therefore it is quite reasonable to assign structure XVIII when R is ethyl, n—propyl and Q-amyl and structure XIX when R is isopropyl as the most stable rotamer for the monosubstituted acetaldehydes. 45 46 XVIII XIX The relative populations of IV and V are solvent de- pendent. For all monosubstituted acetaldehydes except when R is phenyl, the population of V (alkyl eclipsing the carbonyl) decreases as the solvent polarity increases; §,g,, when R is tfbutyl the population of V is 25% in cyclohexane and 17% in acetonitrile. This means the free energy difference AGO IV —a>v becomes more positive in going from cyclohexane to acetonitrile. Such changes with solvent polarity are certainly reasonable, in view of the expected higher dipole moment of IV over V, as shown in XX and XXI. However, it is pertinent O H :1: R \ R H XX XXI to point out that the increase in J cannot be due solely obsd to changes in the relative populations of IV and V, since the coupling constant of acetaldehyde also increases (by only about 2 to 5%) in going from cyclohexane to acetonitrile. 47 Phenylacetaldehyde represents an interesting case. In non-polar solvents, such as carbon tetrachloride, IVa and V are energetically equivalent. In polar solvents, V becomes more stable than IVa; gag., V is more stable than IVa by about 500 cal./mole in acetonitrile. This is in sharp con- trast to the other monosubstituted acetaldehydes, and can be readily explained on the reasonable assumption that V has a higher dipole moment than IV (Sp2 carbon more electronegative than Sp3 carbon), as shown in XXII and XXIII. The greater XXII XXIII effect of solvent polarity on the ratio IV/V when R is phenyl than alkyl agrees well with the fact that the phenyl group contributes more to the dipole moment of the individual rotamers than the alkyl group. II. Disubstituted Acetaldehydes When R1=R2, the data afford the following conclusions: (1) If the alkyl groups are methyl, VIIa (alkyl eclipsing the carbonyl) is more stable than VI (hydrogen eclipsing the carbonyl) by 500 cal./mole and (2) if the alkyl groups are ethyl or tfbutyl, VI is more stable than VIIa by 250 cal./mole and 1100 cal./mole, respectively. 48 The apparent inconsistency of VI (hydrogen eclipsing) being the more stable rotamer of diethylacetaldehyde and V (ethyl eclipsing) of ethylacetaldehyde needs some explana- tion. In order to understand this difference, it is necessary to examine in detail the conformation of each rotamer. The most stable conformation of the ethyl group when the hydrogen eclipses the carbonyl is XXIV, whereby the alkyl chain is all trans and completely staggered. If the alkyl chain were to be kept all trans staggered in VII, a 1.5-eclipsing methyl- H .\ 04 H CH3 . XXVI XXVII proton interaction and a less severe methyl-carbonyl inter- action (XXV) would result. Rotation of one of the ethyl groups to avoid these interactions leads to conformations XXVI and XXVII, which suffer similar interactions. Consequently, VI becomes more stable than VII. In ethylacetaldehyde, however, 49 the conformation having the ethyl group eclipsing the carbonyl does not suffer from such interactions, as shown in XXVIII. XXVIII XXIX 0f the two interactions shown in XXV, the 1,5-eclipsing methyl-proton interaction is probably the more severe one and hence the one responsible for VI being more stable than VII. That the methyl-carbonyl interaction cannot be too significant is attested by the fact that when R is isopropyl, V is more stable than IV, even though in V, as shown in XXIX, such an interaction does exist. Apparently two such interactions, as in t—butylacetaldehyde (XXX), are sufficient to reverse the relative stability of the rotamers. When R1#R2, the data afford the following conclusions. If R1=methyl and Ra=a1kyl, VIIa (methyl eclipsing) is the more stable rotamer. VI (hydrogen eclipsing) and VIIb (alkyl 50 eclipsing) are practidally energetically equivalent. The latter conclusion is drawn from the observation that AHO for these compounds, if R; is treated as equivalent to R2, is about half that of dimethylacetaldehyde. (If neither R1 or R2 are methyl, then the most stable rotamer is VI, apparently for the same reasons given for diethylacetaldehyde. The solvent effects for disubstituted acetaldehydes parallel that of monosubstituted acetaldehydes and are in accord with the proposed conformations. Since VI has a higher dipole moment than VII, the ratio VI/VII increases with increase in solvent polarity. III. Cycloalkanecarboxaldehydes The relative stabilities of VIII and IX depend very much on the ring size. When n=5 (cyclohexyl), IX is more stable. When n=2 (cyclopentyl), IX is slightly less stable. When n=1 (cyclobutyl), IX is slightly more stable. When n=O (cyclo- propyl), IX is much less stable than VIII. (a) Cyclohexanecarboxaldehyde. The fact that IX (alkyl eclipsing the carbonyl) is more stable than VIII by about 400 cal./mole is as expected. In either conformation XXXI or XXXII, the alkyl chain is all gauche and staggered, and the Lg T H \r/H H 2 H2 XXXI XXXII 51 1.5-eclipsing methyl-proton and the methyl-carbonyl inter- actions shown in XXV are absent in these conformations. Furthermore, since in conformation.XXXII the carbonyl bisects the H1CH2 angle, it is not at all surprising that cyclohexane- carboxaldehyde shows exactly the same behavior as dimethyl- acetaldehyde. (lfi Cyclopentanecarboxaldehyde. The fact that IXa (alkyl eclipsing) is very slightly less stable than VIII was not anticipated. However, this interesting observation can be readily explained if one compares XXXIII with XXII. The cyclopentyl ring is certainly less puckered than the cyclohexyl. Using the envelope form (42) for cyclopentanecarboxaldehyde, it can be seen that in conformation XXXIII (alkyl eclipsing the carbonyl), the carbonyl is closer to H2 than H1. H H1 H2 XXXIII The ideal situation where the carbonyl bisects the H1CH2 angle, such as in the case in cyclohexanecarboxaldehyde (XXXII) no longer exists. Apparently the proximity between the carbonyl and H2 is sufficient to destabilize IXa to the extent that it becomes slightly less stable than VIII. 52 (c) Cyclobutanecarboxaldehyde. It is certainly reason- able to assume that cyclobutanecarboxaldehyde is also puckered. There is ample evidence in the literature (45) that has established the puckering of the cyclobutyl ring. The same argument used for cyclopentanecarboxaldehyde can be applied to cyclobutanecarboxaldehyde, XXXIV. H r———H 0 H2 XXXIV (d) Cyclopropanecarboxaldehyde. The complete reversal in cyclopropanecarboxaldehyde, £23., VIII (hydrogen eclipsing) being more stable than IXa (alkyl eclipsing) by about 1.5 kcal./mole, can be explained by extending the arguments used for cyclopentane- and cyclobutanecarboxaldehyde. It can be seen that in IX the carbonyl eclipses H2 (XXXV). This Hl H H A 2 O/// interaction apparently destabilizes IX to such an extent that VIII becomes energetically favored. In addition, whatever factorSforce a,B—unsaturated aldehydes to assume the s-trans conformation (59) may be responsible for the greater stability 55 of VIII over IXa. It is pertinent to point out that in calcu- lating AHO, Jt and J9 for cyclopropanecarboxaldehyde were assumed to be the same as those of other substituted acetal- dehydes. This assumption is probably incorrect, because of the changes in angles and in carbon hybridization in the cyclo- propane ring. If Jt is larger than the value that was used, then a more reasonable value for AHO may be 1 kcal./mole rather than 1.5 kcal./mole. .Bartell (20,21) has suggested that in the gas phase cyclo- propanecarboxaldehyde exists 50% in VIII (hydrogen eclipsing the carbonyl) and 50% in XXXVI (carbonyl bisecting the cyclo- propane ring) rather than VIII and IX. The present data do XXXVI not permit one to make an unequivocal choice between the two possibilities. However, on the basis‘of the following argu- ments, VIII and IX rather than VIII and XXXVI seem more reasonable in the liquid phase. Assuming VIII and XXXVI as the only rotamers, the ob- served coupling constant is expressed by equation 18, where y is the fractional p0pulation of VIII, (1-y) that of XXXVI and JC is the cis coupling constant (dihedral angle 0). Jobsd = th + (1-y)JC (18) 54 The strong dependence of J on temperature clearly indicates that VIII andlMVI are not energetically equivalent. On the basis of Karplus's calculations on the relative magnitude of Jt and Jc’ it can be concluded that VIII is more stable than XXXVI. The difference of the coupling constants at two dif— ferent temperatures can be expressed by equation 19, where p and q are the population of VIII at T1 and T2, respectively. (T1) - J (T2) = (p-q)(Jt-Jc) (19) Jobsd obsd To account for the large variation of the coupling constant with temperature, it can be seen from equation 19 that Bartell's interpretation requires either p to be much larger than q or JC to be much smaller than Jt' Neither possibility is very likely to be true. The variation of the coupling with solvent polarity is again in agreement with the higher dipole moment of VIII over IX. IV. Consideration of Other Conformations The preceding discussion has shown that the data can be well interpreted in terms of eclipsed conformations. It is, however, pertinent and necessary to consider the bisecting conformations, XXXVI}: and XXXVIII for monosubstituted acetal- dehydes and XXXIX and XL for disubstituted acetaldehydes. Equations 2, 5 and 4 become 20, 21, 22 respectively, where JC is the cis coupling and Jlgo is the coupling constant 55 R H2 H1 H1 3. H3 \\ \ \ / H H ‘ H 1 H2 H R XXXVIIa XXXVIIb XXXVIII O 0 R1 H Ra Ra R} H\ \ H \\ H \\ \ H H R2 R1 XXXIX XLa XLb when the dihedral angle is 1200. Since JC should be comparable 5 (monosubstituted) = X(JC+J120)/2 + (l-X)J120 (20) Jobsd Jobsd (disubstituted) = yJC + (1—y)J120 (21) JObSd = 1/3 (JC + 2J120) (22) in magnitude to J and Jlgo comparable to Jg, the data on the t temperature studies could also be interpreted in terms of bisecting conformations. However, these conformations can be excluded on the basis of the following arguments. (1) As discussed previously, microwave and electron dif- fraction studies have established that in the gas phase the minimum energy conformations are eclipsing rather than bisect- ing. There is no reason to expect the reverse to occur in solution. 56 (2) Since XXXVII and XXXIX should have higher coupling constants than XXXVIII and XL, one is forced to conclude that increase in the size of R shifts the equilibrium in favor of XXXVII and XXXIX (in general Jobsd increases as R increases in size). In terms of steric factors such a con- clusion is highly improbable. (3) Since XXXVIII and XL should have higher dipole moments than XXXVII and XXXIX, the observed coupling constants should decrease with increase in solvent polarity. Experi- mentally, however, the observed coupling constants increase with increase in solvent polarity. Although the data have been successfully interpreted in terms of eclipsing conformations, i,g., with the assumption that the dihedral angle ¢ is zero (XLI), it is necessary to H XLI emphasize that small variations in ¢ would not in any reSpect alter the interpretation of the results. In fact the results should not be viewed as proof that the dihedral angle is zero. The causes responsible for the greater stability of V (alkyl eclipsing the carbonyl) over IV (hydrogen eclipsing the carbonyl), even when R is isopropyl, are not well understood. 57 The possibility of hydrogen bonding in V is a plausible explanation. However, it cannot be the sole factor responsi- ble for V being more stable, as attested by the phenyl- acetaldehyde case. An alternate and more attractive explana- tion involves the more favorable dipole-dipole and dipole- induced dipole interactions in V over IV, resulting from two interacting groups whose distance is in the attractive portion of the van der Waals curve. B. Oxime O-Methyl Ethers I. Conformations of theggyp Isomers The data on the coupling constants are in good accord with eclipsed conformations. In fact, the behavior of the .EXE isomers of the oxime O-methyl ethers parallels closely that of the aldehydes. For example, for monosubstituted acetalde- hyde derivatives AHO becomes more positive in changing R from methyl to t-butyl, g._g., R = methyl, AHO= +380 caL/mole; R = tfbutyl, AHO = + 4500 cal./mole. For disubstituted acetaldehyde derivatives, AHO becomes more positive as the alkyl substituents get larger. For cycloalkanecarboxaldehyde deriva- tives, AHO of the cyclopropyl is more positive than that of cyclopentyl, which in turn is more positive than that of the cyclohexyl. AHO of the diethylacetaldehyde derivative is more positive than that of the monoethyl; that of the cyclohexyl is similar to that of the dimethyl rather than that of the diethyl. 58 The interpretation of these results is the same as that used to interpret those of the aldehydes. The effect of solvent polarity on rotamer population further supports the proposed conformations. Because of the higher dipole moment of XIII over XIV and XV over XVI, the ratio XIII/XIV and XV/XVI should increase with increase of solvent polarity. Jobsd does indeed increase as the solvent is changed from cyclohexane to acetaonitrile. As expected, phenylacetaldehyde oxime O-methyl ether behaves exactly the opposite. The arguments previously applied to aldehydes against bisecting conformations (XLII and XLIII) can also be applied here. The most important evidence against XLII and XLIII is OCH3 OCH3 // // XLII XLIII the increase of Jobsd with increase of solvent polarity. Since XLII should have a higher dipole moment than XLIII, increase of solvent polarity should decrease Jobsd (assuming JC > J120)- It is necessary to point out that there is one significant difference between the syn isomers of the oxime O-methyl ethers and the aldehydes; namely, the rotamer populations of 59 the oxime O-methyl ethers parallel closely those of olefins (26) rather than those of aldehydes, although from structural considerations (XLIV-XLV-XLVI), the reverse might have been expected. Whereas the alkyl-carbonyl eclipsing conformation - OCH3 kr/ H H , H H g // [/I H XLV ’ XLIV XLVI is more stable for the aldehydes, the reverse is true for olefins and oxime O-methyl ethers. Although the causes responsible for this difference are not clear, the availability of electrons for possible dipole—dipole (or dipole-induced dipole) interactions might be the controlling factor. II. Conformations of the Anti Isomers Since a quantitative interpretation of the spin-Spin coupling constants of the anti isomers is not possible, several qualitative interpretations will be presented. (a) Monosubstituted agetaldehyde geriyatives. The data on the coupling constants contain two noticeable features: (1) J varies irregularly with temperature and the variation obsd is small. (2) Jobsd decreases or remains constant as the sol- vent polarity increases. A reasonable interpretation in terms of conformations will have to accommodate both of these features. 60 If XLNII and XLVIII (eclipsing conformations) are the only rotamers of the anti isomers, the coupling constants should increase with increase of solvent polarity when R is alkyl and decrease when R is phenyl. The observed trend is H CO 3 \N H3CO\ N \H \ XLVII XLVIII quite the opposite. Furthermore, judging from the small variation of the coupling constants with temperature, XLVII and XLVIII will have to be almost energetically equivalent, _iag., AHO between them is zero, except when R is tfbutyl. Such an interpretation is obviously unreasonable. One is there- fore forced to consider XLIX and L (bisecting conformations). H3CO H3CO \N \ N H R I H\ H\\\ ‘ H H R H XLIX L Since XLIX should have a higher dipole moment than L, the coupling constant should decrease as the solvent polarity in- creases, except that of the phenylacetaldehyde derivative. 61 The data agree reasonably well with this interpretation, although the variation of Jobsd with solvent polarity is too small. However, judging from the temperature dependence of Jobsd’ AHO between XLIX and L should be almost zero, except when R is tfbutyl. Again this interpretation seems unreason- able. The data are best interpreted in terms of XLVII and XLIX as the important rotamers. Since XLIX should have a slightly higher dipole moment than XLVII, increase of solvent polarity should decrease JO only slightly. The fact that XLNII and bsd XLIX are almost energetcially equivalent except when R is .t—butyl is also understandable. (b) Disubstituted acetaldehyde derivatives. The data are in good accord with eclipsed conformations. Since the coupling constants parallel closely those of the syn isomers and alde- hydes, a priori considerations lead to the conclusion that, regardless of the size of R, the most stable rotamer should be LI. H3CO \\N LI III. Chemical Shifts Elucidation of conformations by means of chemical shifts can be quite useful. Accurate knowledge of the anisotropy of 62 the N-OCH3 group would simplify the problem and permit the assignment of reliable conformations. In the absence of such information, however, the simplest solution is intelligent guessing of the anisotropic effect of N-OCHa by comparing it with other groups. As in so many other compounds of the general structure LII, the region in the C=NZ plane (E and F) LII is probably deshielded with respect to the region above and below the plane. In addition, region E is deshielded with respect to F, gag.,.gi§eH1 resonates at lower fields than trans-H1 for compounds of structure LII. Thus, subsequent discussion will be based on the assumption that this is also true for the N-OCH3 group. (a) Solvent effects. The striking feature of the chemical shifts is the effect of benzene on them; namely, whereas some resonances are shifted upfield, several others are shifted downfield (Table IX). A reasonable interpretation of this effect requires stereOSpecific association between the benzene and the oxime O-methyl ether. The data are adequately interpretable in terms of LIII and LIV, whereby the benzene 65 LIII LIV is attracted by the positive charge on the spa-hybridized carbon and is closer to the group that is traps to the methoxy. The s-trans conformation of the C=NOCH3 fragment is chosen in accordance with formaldoxime (44) and with p-p lone pair electron repulsions (45). Models LIII and LIV require that the methoxy be deshielded in benzene, as is indeed the case. Positions A and A‘, and to a lesser extent B", would be deshielded, whereas B, B' and A" would be shielded. (b) Conformations 1) Syn-isomers of substituted acetaldehyde derivatives. From coupling constant studies it has been concluded that XIII (hydrogen eclipsing C=N) is the more stable rotamer for both mono- and disubstituted acetaldehyde derivatives. As the alkyl substituent increases in size, both ormethylene and armethine protons should Spend progressively more time in.. region F (LII). Consequently one would expect progressive shifting of the chemical shifts to lower fields, 222-: the ormethylene protons of the tfbutylacetaldehyde derive should resonate at lower fields than those of all other monosubstituted 64 acetaldehyde derivatives and the osmethine proton of the di-Efbutylacetaldehyde derivative at lower fields than those of any other disubstituted derivatives. The data do not agree with this deduction. This failure could be the result of several factors; such as, differences in the inductive effects of the alkyl groups, change of the orcarbon hybridization, and availability of orproton(s) for hyperconjugation. The only Egang armethine proton that is shifted downfield by benzene, as compared to the neat sample, is that of the di-tfbutylacetaldehyde derivative. This finding implies that the armethine proton lies mainly in region A, as shown in LIII, and further supports the conclusion that this compound exists solely in conformation XIII (hydrogen eclipsing). 2) Anti-isomers of monosubstituted acetaldehyde deriva- tives. The small shift of gigeokmethylene protons in benzene solution as compared to the neat liquid is consistent with the conclusion drawn from coupling constant studies that XLVII and XLIX are significantly populated. However, it is pertinent to point out that the results of the chemical shifts alone do not permit one to draw any definite conclusions. 3) Anti-isomers of disubstituted acetaldehyde derivatives. The observation that both gig and trans armethyl protons resonate at about the same field in the neat liquid, whereas .gig-drmethine protons resonate at appreciably lower fields than trans clearly indicates that the ormethine proton spends most of its time in E (E is deshielded with respect to F) as 65 shown in LII. This agrees well with the conclusion drawn from coupling constant studies that LI is the most stable rotamer of the anti isomers of disubstituted acetaldehyde derivatives. 4) Cis groups of ketone derivatives. Of the cis groups of ketone oxime Ofmethyl ethers, the data afford reasonably accurate conformational assignments only for the isopropyl group. Of the two conformations LV and LVI, only LV is OCH3 OCH3 N// N// H /l H \\ CH3 R R / / CH3 CH3 CH3 LV LVI consistent with the results. It explains the fact that ormethine protons, in contrast to d+methyl and osmethylene protons, resonate at appreciably lower fields when gig than when trans to the oxygen, §,g,, for the di-isopropyl ketone derivative, A6 = -O.56 p.p.m. (neat). IV. Syn-Anti Isomers One interesting feature of the results of syn and anti isomer percentages (Table X) is the greater stability of LVII (anti isomer) over LVIII (syn isomer). This represents yet H3CO OCH3 \\N N// LVII LVIII 66 another case demonstrating the importance of attractive forces between two groups, when at least one group has available polarizable electrons. The greater stability of IJX over LX when x = Y = c1 (46), Br (47), F (48); x = CH3, Y = Cl (49). Br (50), CN (51) has been established. LIX LX The large dependence of the ratio LXI/LXII on R is best interpreted in terms of methoxy-phenyl interactions in LXI, which force the phenyl out of conjugation with the C=N, and in terms of phenyl-R interactions in LXII. When R is methyl, H3co\N N//OCH3 /OCH3 LXI LXII LXIII ethyl or Q-propyl, LXII is much more favored. However, when R is isopropyl, the phenyl-R interactions become severe enough to force the phenyl ring out of conjugation (LXIII) with C=N, and causes the equilibrium to be shifted in favor of LXI. The ultraviolet spectral data amply justify this explanation. EXPERIMENTAL A. Reagents and Compounds Except for tfbutylacetaldehyde, difiE—butylacetaldehyde, cyclopropane-, cyclobutane- and cyclopentanecarboxaldehyde, all aldehydes used were freshly distilled samples of com- mercially available materials. Cyclopentyl bromide, cyclo- pentanecarboxylic acid and tfbutyl acetic acid were obtained from Aldrich Chemical Company, Inc. Cyclopentyl nitrile was obtained from Columbia Organic Chemical Company, Inc.,3and methoxylamine hydrochloride from Eastern Organic Chemicals. B. Solvents Benzene, carbon tetrachloride, acetonitrile and cyclo- hexane were purified from commercially available material by standard methods (52). Benzene-d8 was purchased from Merck, Sharp and Dohme of Canada, Limited. C. Synthesis I. Cyclopropanecarboxaldehyde Cyclopropanecarboxaldehyde was prepared according to the procedure of Brown and Garg (55). In a 1-£., three-necked, round-bottomed flask equipped with a condenser, a dropping funnel and a stirrer, was placed 11.4 g (0.5 mole) of lithium 67 68 aluminum hydride in 500 ml of ether. A nitrogen atmosphere was maintained throughout the reaction. To this stirred solu- tion, 39.65 g (0.45 mole) of ethyl acetate was added over a period of 75 min, the temperature being maintained at 5-70. The reaction mixture was stirred for an additional 30 min. To this solution was added in 5 min 20.1 g (0.3 mole) of cyclopropyl nitrile. The reaction mixture was stirred for 1 hr at 00, and then decomposed with 500 ml of 5N sulfuric acid. The ether layer was separated and the aqueous layer extracted three times with 50-ml portions of ether. The combined ether extracts were washed with saturated sodium bicarbonate solu- tion and water and then dried over anhydrous magnesium sulfate. Cyclopropanecarboxaldehyde, b.p. 99-1040, lit. (54) 97-1000, was obtained in a yield of 4.2 g (20.0%). II. Cyclopentanecarboxylic Acid In a 1-£., three-necked, round-bottomed flask, equipped with a stirrer, a condenser with a drying tube and a dropping funnel, was placed 12.15 g (0.5 mole) of activated magnesium turnings in 200 ml of ether. A solution of 74.45 g (0.5 mole) of cyclopentylbromide in 300 ml of ether was placed in the funnel. About 25 ml of the halide solution was added. Once the reaction started, stirring was commenced and the rest of the halide solution was added over a period of 1 hr to the vigorously refluxing mixture. After completion of the addition, refluxing was maintained by external heating for 50 min. The reaction mixture was then cooled and the solution of the alkyl 69 magnesium bromide was poured slowly onto about 50 g of dry ice with stirring. When the dry ice had evaporated, 400 ml of 20% hydrochloric acid and enough ice to keep the mixture cold were added with stirring. After all the solid had dis- solved, the ether layer was separated, washed with three portions of water and dried over anhydrous magnesium sulfate. The ether was stripped and the residual liquid distilled under reduced pressure to yield 25.4 g (45.0%) of cyclopentane- carboxylic acid, b.p. 880 (4.0 mm), lit. (55) 1180 (25 mm). III. NLN-Dimethyl Cyclopentanecarbdxamide In a 150 ml, two-necked, round-bottomed flask, equipped with a condenser and a drying tube, a magnetic stirring bar and a dropping funnel, was placed 35.0 g (0.276 mole) of freshly distilled thionyl chloride. To this solution was added 21.0 g (0.184 mole) of cyclopentanecarboxylic acid over a period of 1 hr with stirring. The reaction mixture was heated by a steam bath for 2 hr and was then allowed to cool. The excess thionyl chloride was removed by distillation. The residual liquid was quickly transferred to another dropping funnel. In a 1-£., three-necked flask, equipped with a condenser, a stirrer and a dropping funnel was placed 100 ml (0.555 mole) of 25% aqueous dimethylamine solution. The solution was cooled in an ice-salt mixture. To this solution the crude acid chloride was added dropwise over a period of 3 hrs. After completion of addition, the reaction mixture was stirred at 70 room temperature for 1 hr. The aqueous solution was satur- ated with sodium chloride and then extracted three times with 100-ml portions of ether. The ether layer was washed with saturated sodium chloride solution and dried over anhydrous magnesium sulfate. The ether was distilled off and the residual brownish liquid was distilled under reduced pressure to give 19.5 g (75.0%) of N,N-dimethylcyclopentane- O carboxamide, b.p. 92-5 (5.0 mm). IV. Cyclopentanecarboxaldehyde The procedure of Brown and Tsukamoto (56) was followed. In a 500 ml, three-necked flask, equipped with a condenser and a drying tube, a stirrer and a dropping funnel, was placed 4.85 g (0.128 mole) of lithium aluminum hydride in 160 ml of ether (1.25 M solution). The flask was cooled by an ice bath. To the stirred solution of the hydride was added over a period of 1 hr 16.9 g (0.192 mole) of ethyl acetate. The reaction mixture was stirred for 50 min at 00. To the stirred slurry of the hydride reagent thus prepared, cooled with an ice bath, was added 18.0 g (0.128 mole) of N,N-dimethylcyclopentane-' carboxamide as rapidly as possible while avoiding too vigorous refluxing of the ether. The reaction mixture was stirred for 1 hr at the same temperature and then decomposed with 5N sulfuric acid. The ether layer was separated and the aqueous layer was extracted twice with 100-ml portions of ether. The combined ether solution was washed with water, shaken with solid sodium bicarbonate, washed again with water and dried 71 over anhydrous magnesium sulfate. After evaporation of the ether, the residual liquid distilled at 154-1580, lit. (57) 1350, yield 9.0 g (72%). V. N,N-Dimethyl Cyclobutanecarboxamide N,N-Dimethyl cyclobutanecarboxamide was prepared by the same procedure used for the preparation of N,N-dimethyl cyclo- pentanecarboxamide. It was obtained in 60% yield, b.p. 77-780 (4 mm), lit. (58) 105° (20 mm). VI. Cyclobutanecarboxaldehyde Cyclobutanecarboxaldehyde was prepared from N,N-dimethyl- cyclobutanecarboxamide according to the procedure of Brown and Tsukamoto (56). It was obtained in 40% yield, b.p. 115-1170, lit. (59) 113—1150. VII. N,N-Dimethyl-S,S-dimethyl-butyramide N,N-Dimethyl-S,S-dimethyl-butyramide was prepared from Efbutyl acetic acid by the same procedure used for the prepara— tion of N,N-dimethylcyclopentanecarboxamide. It was obtained in 72% yield, b.p. 74-750 (6-7 mm). VIII. thutylacetaldehyde lt-Butylacetaldehyde was prepared from the corresponding N,N-dimethyl amide according to the procedure of Brown and Tsukamoto (56). It was obtained in 60% yield, b.p. 107—1080. lit. (60), 102-1040. 72 I__1g(___.~ Oxime O-Methyl Ethers _To an aqueous solution of 0.1 mole aldehyde or ketone, 0.11 mole methoxylamine hydrochloride and 0.11 mole sodium acetate trihydrate was added 95% ethanol until the solution was clear (except for water soluble compounds such as acetal- dehyde). After 20 hrs reflux the solution was extracted three times with 50-ml portions of ether. The ether layer was washed three times with 5% sodium bicarbonate solution, once with water and dried over anhydrous magnesium sulfate. After removal of the ether by slow distillation, the residual liquid was distilled through a fractionating column to give the oxime O-methyl ether in about 50-70% yield. All oxime ethers were clear, sweet smelling liquids. D. N.M.R. and U.V. Spectra All n.m.r. spectra were taken at 60 NC on a Model A-60 Spectrometer (Varian Associates, Palo Alto, Calif.). Undegassed samples were used with tetramethylsilane (TMS) as the internal reference standard (T = 10.00). Chemical shifts were measured with sweep widths of 1000, 500 and 250 c.p.s. Spin-spin coupling constants were measured with sweep width of 50 c.p.s. Ultraviolet Spectra were taken with a Cary 14 recording spectrometer. 73 Table XVIII. Boiling Points of Oxime O-Methyl Ethers ”M R1R2C=NOCH3 R1 R2 5.2. (0c) H CH3 47 H CH2CH3 73-75 H CH2CH2CH3 85-86 H CH2CH(CH3)2 114-115 H CH2C(CH3)3 115-117 H CH2C5H5 77(5-6 mm) H CH(CH3)2 87-90 H CH(CH3)CH2CH3 110-113 H CH(CH3)CH(CH3)CH2CH3 141-145 H CH(CH2CH3)2 128 H CH(CH2CH3)(CH2)3CH3 109 (60 mm) H CH[C(CH3)3]2 88-90 (35 mm) H ‘<:j 110 H <:jL 59 (5 mm) H <:::> 78 (19-20 mm) CH3 CH3 72-75 CH3 CHZCHg 92-95 CH3 CH2CH(CH3)2 126-128 CH3 CH2C(CH3)3 64-65 (35 mm) CH3 CH2C5H5 88-89 (3-4 mm) CH3 CH(CH3)2 109—110 CH3 C(CH3)3 79-80 (154 mm) CH2CH3 CHéCHg 114-116 CH2CH3 CH(CH3)2 128-129 CH2CH3 C(CH3)3 127-129 CH(CH3)2 CH(CH3)2 125-126 CH3 CeHs 97-99 (14-15 mm) CH2CH3 C6H5 74-80 (2—4 mm) CH2CH2CH3 C7H5 88-90 (2-5 mm) CH(CH3)2 CSHS 46-48 (0.5 mm) (1) (2) (5) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) REFERENCES E. L. Eliel and M. H. Gianni, Tetrahedron Letters, 97 (1962). W. C. Neikam and B. P. Dailey, J. Chem. Phys., 58, 445 (1965). _— A. H. Lewin and S. Winstein, J. Am. Chem. Soc.,_§4, 2464 (1962). —‘ A. A. Bothner—By and C. Naar-Colin, ibid., 84, 745 (1962). F. A. L. Anet, ibid., 84, 747 (1962). H. S. Gutowsky, G. G. Belford and P. E. McMahon, J. Chem. Phys., §__§. 3555 (1962). N. Sheppard and G. J. Szasz, ibid., 11, 86 (1949). K. Kuratani, T. Miyazawa and S. Mizuhima, ibid., 21, 1411 (1955). ‘—_ w. D. Gwinn and K. s. Pitzer, ibid., _g, 503 (1948). R. E. Kagarise and D. H. Rank, Trans Faraday Soc., 48, 594 (1952). _— S. Mizushima, T. Shimanouchi, T. Miyazawa, I. Ichishima, K. Kuratani, I. Nakagawa, and N. Shido, J. Chem. Phys., 21. 815 (1955). I. Nakagawa, I. Ichishima, K. Kuratani, I. Miyazawa, T. Shimanouchi, and S. Mizushima, ibid., 20, 1720 (1952). A. Miyake, I. Nakagawa, T. Miyzzawa, I. Ichishima, T. Shimanouchi, and S. Mizushima, Spectrochim. Acta, 15 161 (1958). "" S. Mizushima, T. Shimanouchi, I. Ichishima, T. Miyazawa, I. Nakagawa, and T. Araki, J. Am. Chem. Soc., 18, 2058 (1956). _— R. W. Kolb, C. C. Lin, and E. B. Wilson, Jr., J. Chem. Phys., g2. 1695 (1957). 74 75 (16) S. S. Butcher and E. B. Wilson, Jr., ibid., 40, 1671 (1964). ———— (17) K. M. Sinnott, ibid., 2:, 851 (1961). (18) D. R. Herschbach and L. C. Krishner, ibid., 2_, 728 (1958). _' (19) C. Romers and J. E. G. Creutzberg, Rec. trav. Chim., ii) 551 (1956). (20) L. S. Bartell, B. L. Carroll, and J. P. Guillory, Tetrahedron Letters, 15, 705 (1964). (21) L. s. Bartell and J. P. Guillory, J. Chem. Phys., 45 647 (1965). _— (22) R. J. Abraham and J. A. Pople, Mol. Phys., 5, 609 (1960). (25) E. B. Whipple, J. H. Goldstein and G. R. McClure, J. Am. Chem. Soc., 2, 5811 (1960). (24) E. B. Whipple, J. Chem. Phys., 55, 1059 (1961). (25) A. A. Bothner-By and C. Naar-Colin, J. Am. Chem. Soc., fig. 251 (1961). (26) A. A. Bothner-By, C. Naar-Colin, and H. Gunther, ibid., éi’ 2748 (1962). (27) A. A. Bothner-By and H. Gunther, Discussions Faraday Soc., §4, 127 (1982). (28) T. L. Brown, Spectrochim. Acta, 18, 1615 (1962). (29) J. G. Powels and J. H. Strange, Mol. Phys., 5, 529 (1962). (50) M. Karplus, J. Am. Chem. Soc., £2, 2870 (1965). (51) H. S. Gutowsky, M. Karplus, and D. M. Grant, J. Chem. Phys., g1. 1278 (1959). (52) E. Hiroike, J. Phys. Soc. Japan, 15, 270 (1960). (55) J. Ranft, Ann. d. Physik [71, 2, 522 (1961). (54) R. M. Lynden—Bell and N. Sheppard, Proc. Roy. Soc. (London), A269, 585 (1962). (55) D. R. Whitman, L. Onsager, M. Saunders, and H. E. Dubb, J. Chem. Phys., 2g, 67 (1960). (56) (57) (58) (59) (40) (41) (42) (45) (44) (45) (48) (47) (48) (49) (50) (51) (52) (55) (54) 76 J. s. Waugh and F. w. Dobbs, ibid., 55, 1255 (1959). R. J. Abraham and K. G. R. Pachler, Mol. Phys., 165 (1965-1964). ||\) I R. Glick and A. A. Bothner-By, J. Chem. Phys., 55, 582 (1958). _— J. A. Pople and T. Schaefer, Mol. Phys., 5, 547 (1960). G. J. Karabatsos and R. A. Taller, J. Am. Chem. Soc., :55. 4575 (1964). B. L. Shapiro, S. J. Ebersole, and R. M. Kopehik, J. Mol. Spectr., 11(4), 526 (1965). F. V. Brutcher, Jr, T. Roberts, S. J. Barr, and N. Pearson, J. Am. Chem. Soc., 5;, 4915 (1959). J. B. Lambert and J. D. Roberts, ibid., §§, 5710 (1965). I. N. Levine, J. Chem. Phys., 55, 2526 (1985). N. L. Owen and N. Sheppard, Proc. Chem. Soc., 264 (1965). K. S. Pitzer and J. L. Hollenberg, J. Am. Chem. Soc., :;5, 1495 (1954). J. M. Dowling, P. G. Puranik, A. Meister, and S. I. Miller, J. Chem. Phys., 25, 255 (1957). N. C. Craig and E. A. Entemann, J. Am. Chem. Soc., 55, 5047 (1961). _— R. A. Beaudet, J. Chem. Phys., 55, 2705 (1964). P. S. Skell and R. G. Allen, J. Am. Chem. Soc.,_55, 5997 (1958). '_' J. N. Butler and R. D. McAlpine, Can. J. Chem., 55, 2487 (1985). _— L. F. Fieser, Experimengs in Organic Chemistry, D. C. Heath and Co., 5rd ed., Boston, 1957, p. 281. H. C. Brown and C. P. Garg, J. Am. Chem. Soc., 55, 1085 (1964). '_— H. C. Brown and A. Tsukamoto, ibid., 5, 4549 (1961). 77 (55) H. Adkins and R. W. Rosenthal, ibid., 15, 4550 (1950). (56) H. C. Brown and A. Tsukamoto, ibid., _5, 1089 (1964). (57) G. R. Clemo and J. Ormston, J. Chem. Soc., 562 (1955). (58) J. M. Conia and J. Gore, Compt. Rend., 254, 5708 (1962). V (59) Ya. M. Slobodin and M. V. Blinova, Zhur. Obschei Khim., 25. 1994 (1955). (60) F. C. Whitmore and A. H. Homeyer, J. Am. Chem. Soc., g) 4555 (1955) . RI A" R“ 3” ll L" V" “Iv!” R” 2 9190 mm); 930 (Tillmmmluuu 312