摘要:

J. CHEM. SOC. PERKIN TRANS. 2 1990 lmines and Derivatives. Part 23." Anomalous IH NMR Spectrum of N-[I-(I -Naphthyl)ethylidene]-1-phenyl-2-propylamine Conformation in Solution Atropisomerism and an X-Ray Crystal Structure t Thomas A. Hamor," W. Brian Jennings," Lee D. Proctor and Malcolm S. Tolley School of Chemistry University of Birmingham PO Box 363 Birmingham B 152TT Derek R. Boyd* and Teresa Mullan Department of Chemistry The Queens University of Belfast Belfast B 19 5AG 'H NMR spectra of the title imine in solution indicate the presence of three stereoisomers due to €/Z isomerism about the imino bond and atropisomerism about the 1 -naphthyl imino bond in the Z- isomer. The most abundant Z-isomer which was isolated in crystalline form exhibits a highly unusual NMR signal at 6 5.9 assigned to a naphthyl proton cc to the imino group.The X-ray crystal structure reveals that this hydrogen atom is situated only 2.7 a above the face of the phenyl ring of the N-I -phenyl-2-propyl moiety and hence experiences a large diamagnetic ring current shielding effect. An attractive edge-to-face ring interaction involving this naphthyl proton and the n-electrons of the proximate phenyl group could account for this observation. The marked temperature dependence of this NMR signal is quantitatively analysed in terms of a fast equilibrium in solution between the conformation favoured in the solid state and a second conformation which lacks the ring current effect. In the course of our investigations into imines and derivatives,' the title imine (1) was prepared from amphetamine (l-phenyl-2- propylamine) and 1-naphthylethanone and was subjected to routine chemical and spectroscopic analysis.While microanalytical data and the IR [vmax 1 645 cm-' (C=N) in Nujol] and mass (M' at m/z 288,11%) spectra were in accord with the expected structure the 'H NMR spectrum (Figure 1) appeared to be anomalous. An unexpected doublet signal was evident at 6 5.91 in deuteriochloroform solution a region usually associated with vinylic protons. Additionally two sets of alkyl resonances were observed in freshly prepared solutions uiz. methyl doublet signals at 6 1.11 (component A 62%) and 1.05 (component B 38%) and imino methyl singlet signals at 6 2.30 (A 62%) and 6 2.33 (B 38%).Furthermore when the solution was allowed to stand for a few hours a third set (C) of signals developed at 6 1.38 (d) 2.00 (s) and 4.1 1 (m). After ca. 3 days at ambient temperature (20 "C) an equilibrium was attained at a component distribution A B:C = 49 :31 :20% in deuteriochloroform solution. 13C NMR spectra of (l) see the Experimental section corroborated the above findings. A freshly prepared solution showed doubling of most carbon resonances and on standing a third set of signals slowly developed. In particular the equilibrated deuteriochloroform solution showed three imino I3C=N signals at 6 167.1 (component B) 166.7 (A) and 166.3 (C). This observation together with the absence of I3C NMR signals attributable to a =CH2 moiety establishes that all three components of (1) are imines.Hence imine-enamine tautomer- ism is not responsible for the multiplicity of signals. More definitive structural information was obtained from an experiment in which a few crystals of imine (1) were dissolved in deuteriochloroform pre-cooled to -50 "C and then a 270 MHz 'H NMR spectrum was quickly recorded at -50°C. The resulting spectrum was essentially that of component A with only a trace contribution from component B.1 The proportion of B increased to its equilibrium value when the sample tube was removed from the probe for a few minutes allowed to warm to ambient temperature and the spectrum recorded again at -50 "C. Subtraction of initial and equilibrated spectra with L1 ,.I.I*,& ,, I ,,,I I /,,I I ., I ,, I ,, I ,, I,,,I I I,< I I, I ,, I IIII 1 I,.. I ,,II L 98765432 1 WPPm) Figure 1.90 MHz 'H NMR spectrum of imine (1) recorded at 25 "C in deuteriochloroform after standing for 2 days to equilibrate. appropriate weightings provided individual spectra of com- ponents A and B (Figures 2 and 3). Both components A and B clearly have the same skeletal structure with CHMe doublet signals at 6 1.11 ('J 6.0 Hz) and t Contribution from the Crystallography Unit Universities of Aston and Birmingham. $The crystallization of isomer A from a pentane solution of imine (1) which consists of a three-component equilibrium mixture involves a second-order diastereoisomeric transformation. Thus as isomer A crystallizes out of the warm solution the equilibrium will be driven in favour of this component. \ ,CH2Ph8Hyph @ c,'\ -c/ C=N CH3 C=N CH /\ \ CH3 *' CH3 ** (la) (lb) C=N / \ p 3 CH3 ,c H 'CH2Ph I ' ' 6I ' ' 4 ' ' I ' I ' i8 7 5 3 2 WPPm) Figure 2.270 MHz 'H NMR spectrum of imine (1) recorded in deuteriochloroform at -50 "C following dissolution of the crystals at -50 "C. The signals of atropisomer (lb) have been removed by spectral subtraction. * The observation that the benzylic protons in components A-C give NMR signals appropriate to the AB part of an ABX system and that the CH resonance is a complex multiplet eliminates the possibility (suggested by a referee) that one of these components might be the structurally isomeric imine 1-C,,H,CH(Me)N=C(Me)CH,Ph.In any case isomerization of this type usually requires a strong base to remove the NCH proton.7 One of the AB subspectra in this benzylic ABX system was a single line hence the individual values of 3JAxand 3JBxcould not be ascertained. J. CHEM. SOC. PERKIN TRANS. 2 1990 1'1'1'1'l'l'l'l 8 7 6 5 4 3 2 1 G(PPm) Figure 3. 270 MHz 'H NMR difference spectrum obtained in deuteriochloroform at -50 "C showing the spectrum of atropisomer (lb); peak labelled 'X'is a subtraction spike. 1.05 ( 3J6.2 Hz) respectively; complex CHMe multiplet signals at 6 3.11 and 3.14; singlet MeC= signals at 6 2.30 and 2.33; and complex naphthyl ring signals. Significantly the benzylic resonance of each component consisted of two AB subspectra centred at 6 2.71 and 2.79 respectively.Manual analysis of the AB section of these ABX spectra gave 'JAB13.0 Hz 3JAxca. 8.9 Hz and 3JBx 13.1 Hz JAxca. 4.1 Hz for component A and JAB ca. 7.2 Hz and 3JBxca. 6.3 Hz for component B. This confirms the integrity of the amphetamino PhCH,CH moiety where the diastereotopic benzylic protons exhibit an AB part of an ABX system." The CH (X) signal of both components is a complex multiplet owing to further coupling to the methyl group. The 'H NMR signals of components A and B are assigned to 1-naphthyl atropisomers (la) and (1b) respectively of the imine in the 2-configuration. The 1-naphthyl moiety is twisted out of the imino plane and rotation is restricted by the proximate cis N-alkyl group.This leads to a chiral axis which in combina- tion with the chiral N-1-phenyl-2-propyl centre gives rise to diastereoisomers A and B.Hindered rotation about 1-naphthyl imino bonds has been observed previously in related com-pounds '-'where a 1-naphthyl group is located cis to an N-alkyl group. Subtraction of the 270 MHz 'H NMR spectrum (recorded at ambient temperature in deuteriochloroform solution) of the rapidly equilibrated A/B component mixture from that of the fully equilibrated three component mixture afforded a spectrum of the minor component (C). This difference spectrum confirmed the intact skeletal amphetamine imine structure of component C with signals at 6 1.38 (d 3J6.2 Hz CHMe) 2.00 (s MeC=) 2.97 (AB part of ABX 'JAB 13.1 Hz 3JAx+ JBx13.1 Hz CH,Ph),Y 4.11 (X multiplet which is split further by Me CH) and 7.2-7.8 (complex aromatic).Hence component C which slowly develops on standing in solution at ambient temperature is assigned to the E-imine (lc).*This assignment is supported by the 10 ppm upfield shift of the vinylic methyl carbon signal of component C in the 13C NMR spectrum (Experimental section) relative to its position in components A and B. Assignment of the vinylic methyl 13C signals was confirmed by observing that they essentially disappeared (due to deuteriation of the vinylic methyl protons ') when ca. 0.15 cm3 of deuteriomethanol were added to the deuteriochloroform solution of the imine. Previous work on related imines has established that an a-carbon situated syn to an N-alkyl group resonates 9-15 ppm upfield from an a-carbon situated syn to the imino nitrogen lone-pair electron^.^,^ The relatively slow equilibration of components A and B with component C at ambient tem- perature also confirms that C rather than B is the E-imine (lc).Previous studies of related N-alkyl ketimines have established that 2 -E isomerization occurs with a half-life of several hours at ambient temperature.2,6 On the other hand rotation J. CHEM. SOC. PERKIN TRANS. 2 1990 Figure 4. Stereoscopic view of the solid-state structure of imine (1) showing the crystallographic atom numbering. Table 1. Fractional atomic co-ordinates ( x lo4)with esds in parentheses.X Y Z 536(5) 5 003(3) 2 949(3) 2 910(6) 4 731(4) 4 040(4) 3 521(6) 3 829(4) 4 077(3) 3 291(7) 3 196(4) 3 310(4) 3 893(8) 2 309(5) 3 364(5) 4 753( 10) 2 004(6) 4 136(5) 5 002(8) 2 579(5) 4 861(6) 4 383(7) 3 470(4) 4 819(4) 4 621(7) 4 089(4) 5 587(4) 3 995(7) 4 962(4) 5 534(3) 3 151(7) 5 277(4) 4 765(3) 2 085(7) 5 077(4) 3 181(3) 3 260(8) 5 674(5) 2 633(4) -681(6) 4 494(4) 3 496(4) -1 476(9) 3 731(5) 2 948(5) -2 025(6) 5 187(4) 3 825(4) -1 306(7) 5 919(4) 4 424(5) -605(8) 6 726(4) 4 083(6) 120( 10) 7 382(6) 4622(8) 150(10) 7 257(7) 5 508(8) -506( 10) 6 459(7) 5 884(6) -1 254(8) 5 791(5) 5 345(5) 2 681 5 982 4 749 about 1-naphthyl-imine bonds takes place very rapidly at ambient temperature (though it is slow at -50 "C).Thus the barrier to interconversion of atropisomers A and B measured from the dynamic coalescence of the amphetamino methyl doublet signals at 107 "C (270 MHz) in 1,2,4-trichlorobenzene- [2H,,]-p-xylene (7:1) was determined to be AGtbA 20.5 & 0.3 kcal mol-'. This barrier is similar to barriers for rotation about 1-naphthyl-imine bonds in other Z-imines and is signifi- cantly less than the barriers to E -+ 2 is~merization.~,~,~ The E-isomer (lc) does not show evidence of atropisomerism owing to fast rotation on the NMR time-scale of the much less sterically restricted 1 -naphthyl moiety which is located trans to the bulky N-alkyl group and cis to the nitrogen lone-pair electrons. Our principal interest in the 'H NMR spectrum of imine (1) concerned the anomalous doublet signal at 6 5.91.Integration established that this signal was due to one proton of the major component A. It is far upfield of the region 6 7-8 normally associated with naphthyl ring protons. The direction of the roofing (tenting) of this doublet on a low-field spectrometer indicated that it was coupled to a signal within the aromatic envelope. This was confirmed by observing that the doublet signal at 6 5.91 collapsed to a singlet on irradiation at 6 7.18. The 'H NMR spectrum of component A established that the unusual doublet signal which moved upfield to 6 5.66 at -50 "C (see below) is coupled to the triplet at 6 7.18 and forms part of a doublet-triplet-doublet system (Figure 2) assigned to the protons at the 2- 3- and 4-positions on the 1-naphthyl ring.An unambiguous assignment was obtained from selectively de- coupled 67.8 MHz 13C NMR spectra recorded at ambient temperature. Selective irradiation of the proton doublet at 6 5.91 was observed to remove a 4.0 Hz coupling from the '3C=N multiplet (a doublet of quintets) of the major component A leaving a quintet signal (J 6.5 Hz). The residual quintet splitting arises from a two-bond 13C,H coupling to the imino methyl protons together with a three-bond coupling of a similar size (6.5 Hz) to the NCH proton. The removal of the 4.0 Hz I3C,H coupling by irradiation of the proton doublet at 6 5.91 clearly identifies the latter resonance as belonging to the proton on the 2-position of the 1-naphthyl ring since only this proton would be expected to show a three-bond coupling of ca.4 Hz to the imino carbon. Clearly this proton signal is shifted far upfield of the normal resonance position of naphthalene ring P-protons 6 7.5. The upfield shift ca. 1.6 ppm is so large that the only viable explanation that can be advanced is a strong shielding effect from an adjacent n-electron circulation (aromatic ring current). The N-1-phenyl-2-propyl moiety is the only possible origin of such an effect though there does not appear to be any evident reason as to why the phenyl moiety should lie close to the naphthyl proton at the 2-position. Since the low temperature dissolution experiment (see above) established that the crystals (isolated from a pentane solution) were composed of the major isomer A which was associated with the anomalous naphthyl proton resonance it was considered worthwhile to investigate the crystalline material by X-ray crystallography.The resulting structure and the crystallographic numbering scheme used is shown in Figure 4. The atomic co-ordinates are given in Table 1 and the bond lengths bond angles and selected torsion angles are listed in Table 2. The absolute configuration at C(13) is (R),established by the synthesis of the material from R-( +)-amphetamine. The X-ray analysis establishes that the major isomer (A) is the Z-atropisomer (la). The torsion angles about C( 1)-C( 1l) C( 1 1)-N( l) N( 1)-C( 13) C( 13)-C( 13 and C( 15)-C( 16) which define the overall conformation of the molecule are listed in Table 2.The torsion angle at the imino double bond C( 1)-C( 1 1 )=N( 1)-C( 13) which is -4.0" establishes the Z-configuration. The naphthalene residue is oriented nearly perpendicular to the essentially planar C( 1)-C(1l)[C( 12)]=N( 1)-C( 13) moiety with the naphthyl B ring directed anti to the amphetamine benzyl group (Figure 4). The two aromatic residues are each planar to within 0.01 A and the angle between them is 48.2". Bond lengths are generally normal and mostly agree with expected values to within the limits of experimental error. In the naphthalene residue however the C( 1)-C(2) and C(6)-C(7) bond lengths are ca. 0.05 8 shorter than those found in the crystal structure of naphthalene while the adjacent C(Z)-C(3) and C(7)-C(8) lengths are ca.0.05 8 longer than in naphthalene. The naphthalene residue therefore while maintaining approximate 2/m (C,,) symmetry has lost the higher mmm (D,,,)symmetry 8,9 of naphthalene itself. MNDO molecular orbital calculations lo on the title compound in which starting from the X-ray structure all the parameters of the naphthalene residue were optimized resulted in a geometry virtually identical with that of the naphthalene molecule with approximatemmm symmetry. The statistically highly significant deviation of these bond lengths from the naphthalene values may therefore arise from some subtle electronic or steric effect to which the MNDO approximation is not sensitive.However the possibility that these anomalous bond lengths are merely the result of random errors in the structure analysis cannot be entirely ruled out (despite their apparent statistical significance) although in view of the otherwise normal bonding geometry this seems unlikely. Table 2. Bond lengths A and bond angles O with esds in parentheses and selected torsion angles esds ca. 0.7'. 1.266(6) C(8)-C(9) 1.3 5 1 (7) 1.458(6) C( 9)-C( 10) 1.41 l(7) 1.387(7) C(l1)-C(12) 1.506(7) 1.361 (7) C( 13)-C( 14) 1.511(7) 1.533(7) C( 13)-C( 15) 1.532(7) 1.486(7) C( 15)-C( 16) 1.498(8) 1.406(7) C( 16)-C( 17) 1.385(8) 1.364(8) C( 16)-C(21) 1.405(9) 1.415(9) C( 17)-C( 18) 1.370(10) 1.387( 10) C( 18)-C(19) 1.351( 11) 1.374(7) C( 19)-C(20) 1.381(11) 1.476(7) C(2O)-C(2 1) 1.389(9) Bond angles C( 1 1)-N-C( 13) 120.7(5) C(1)-C( 10)-C(9) 122.7(6) C(2)-C(l)-C(10) 117.5(5) N-C( 1 1)-C( 1) 127.5(5) C(2)-C( 1)-C( 1 1) 1 l8.9(5) N-C( 1 1)-C( 12) 118.7(5) c(lo)-c(l)-c(ll) 1 23.5( 5) C( 1)-C( 1 1)-C( 12) 1 13.2(4) C( l)-C(2)-C(3) 120.2(5) N-C( 13)-C( 14) 108.9(5) C(l)-C(2)-C(7) 122.8(5) N-C( 13)-C( 15) 1 07.7( 4) C(3)-C(2)-C(7) 117.0(5) C( 14)-C( 13)-C( 15) 11 1.7(5) C(2)-C(3)-C(4) 119.1(6) C( 13)-C( 15)-C( 16) 11 3.4(4) C( 3)-C(4)-C( 5) 1 20.3 (7) C( 15)-C( 16)-C( 17) 12 1.0(7) C(4)-C(5)-C(6) 122.2( 7) C( 15)-C(16)-C(21) 12 1.1(6) C(5)-C(6)-C(7) 118.1(8) C( 17)-C( 16)-C(21) 117.8(7) C(2)-C( 7)-C( 6) 123.4(7) C( 16)-C( 17)-C( 18) 121.5 (9) C(2)-C( 7)-C( 8) 117.7(5) C(17)-C(18)-C(19) 120.2( 10) C( 6)-C( 7)-C( 8) 119.0(6) C( 18)-C( 19)-C(20) 120.8( 10) C( 7)-C(8)-C(9) 118.0(5) C( 19)-C(2O)-C(21) 119.5(9) C(S)-C(S)-C( 10) 121.2(5) C( 16)-C(21)-C(20) 120.1 (8) Torsion angles C( 10)-C( 1)-C( 1 1)-N -95.8 C(2)-C( 1)-C( 1 1)-N 87.3 c(lo)-c(l)-c(l1)-c(12) 75.6 C( 1)-C( 1 l)-N-C( 13) -4.0 C( 12)-C( 11)-N-C( 1 3) -175.0 C( 1 1)-N-C( 13)-C( 15) 115.8 C( 1 1 )-N-C( 13)-C( 14) -122.9 N-C( 13)-C( 15)-C( 16) -64.4 C( 14)-C( 13)-C( 1 5)-C( 16) 176.0 C(13)-C( 1 5)-C( 16)-C( 17) 86.5 C( 13)-C( 15)-C( 16)-C(2 1) -90.8 TIK Figure 5.Plot of observed 6 values (solid circles) in iso-octane and calculated 6 values (solid line) for C(10)-H us. temperature/K. J. CHEM. SOC. PERKIN TRANS. 2 1990 The most interesting feature of the crystal structure is that the aromatic hydrogen atom attached to the 2-position of the l-naphthyl ring [C(lO)-H using the crystallographic numbering system in Figure 41 is situated very close to the face of the amphetamine phenyl ring.Specifically this roton in its calculated position assuming C-H = 1.08 fi lies 2.68 8,above the plane of the amphetamine phenyl ring and the normal to this plane is displaced only 0.37 8 from the centre of the phenyl ring. Insertion of these measurements into the Johnson-Bovey ring-current equations gives a predicted upfield shift for C(lO)-H of 2.31 ppm due to the diamagnetic n-electron circulation. Subtraction of this shift from the normal resonance position of naphthalene P-protons (6 7.5) results in a predicted 6 value of 5.2 for C(lO)-H.This compares with the observed 6 value of 5.91 in deuteriochloroform at ambient temperature. However the chemical shift of this resonance was found to be unusually temperature dependent moving upfield on lowering the temperature (6 5.66 at -50 "C in deuteriochloroform compared with 6 5.91 at 20 "C). Iso-octane was selected as the solvent for a more detailed variable temperature study owing to the wide gap between its freezing and boiling points and its non-polar weakly solvating nature. The plot of the C( 10)-H signal position us. temperature (Figure 5) departs only slightly from linearity over the range studied (-100 to +95 "C). At -100 "C this signal had moved further upfield to 6 5.0 which corresponds with the predicted chemical shift given by the Johnson-Bovey ring current calculations based on the X-ray conformation.However it is clear from the plot in Figure 4 that even lower 6 values would be obtained at lower temperatures. Attempts to obtain 6 values for C(lO)-H above 100°C (in decalin) were thwarted by signal collapse due to the onset of fast site-exchange between this signal and another signal located in the aromatic envelope. Above 100 "C the atropisomers (la) and (lb) interconvert rapidly on the NMR time-scale by rotation about the naphthyl-imine bond. The marked temperature dependence of the chemical shift of C(lO)-H in isomer (la) can be rationalized in terms of an equilibrium in solution between two molecular conformations of atropisomer (la); one approximating to the conformation (I) existing in the solid state (Figure 4),and another (11) where the amphetamine phenyl ring is much further away from C( 10)-H which therefore has no ring current shielding (Scheme).Con- formations (I) and (11) interconvert rapidly on the NMR time-scale at all accessible temperatures. Conformation (I) Conformation (11) C( 10)-H shift 61 611 Fractional population PI (1 -PI) K = (1 -PJP1 = exp[(TAS* -AH")/RU (1) Computer fitting (Figure 5) of the observed 6 us. T plot (in iso-octane) gave AS* = 2.77 cal mol-' K-' * AH+' = 1 305 cal mol-' where aI = 4.8 ppm 611= 7.5 ppm. Scheme. * 1cal = 4.184 J. J. CHEM. SOC. PERKIN TRANS. 2 1990 An interactive computer program was used to generate 6 us.temperature curves according to equations (1) and (2) in the Scheme from various values of AS" AH" 61 and 611. The optimum agreement between calculated and experimental plots depicted in Figure 5 was obtained with the values given in the Scheme. The optimized value of (4.8) is very close to the chemical shift of C(lO)-H (6 5.2) given by the Johnson-Bovey ring current calculations on the solid state conformation. The chemical shift of C(lO)-H in the second proposed conformer gave a satisfactory fit at till 7.5 which is the normal signal position for a naphthalene P-hydrogen. It is clear from the thermodynamic parameters in the Scheme that the solid state conformation (I) is energetically favoured in solution and hence becomes more populated at lower temper- ature causing the observed upfield movement of C(lO)-H with decreasing temperature.The entropy term on the other hand favours conformation (11) which therefore becomes in- creasingly populated at high temperatures. It is reassuring that the estimated value of AS+ (2.8 cal mol-' K-') is of a reasonable magnitude for a conformational equilibrium. Furthermore the lower entropy of the solid-state conformer (I) is in line with expectations as this conformer is quite sterically hindered due to the close proximity of the naphthyl ring and the benzyl group. Although the precise geometry of the proposed second solution conformation (11) is unknown it is almost certainly less sterically congested than conformation (1).It is interesting to consider why isomer (la) favours on an enthalpy basis a hindered conformation (I) where the naphthyl ring is in close proximity to the amphetamine phenyl moiety and almost orthogonal to the imino plane.It seems unlikely that this conformation is forced on the system by even greater steric repulsions in all alternative conformations. Indeed an inspec- tion of molecular models indicates several other less congested conformations available to (la) resulting from rotation about the N-CH and CH-CH,Ph bonds. The proposed second conformation (11) which becomes increasingly populated in solution at higher temperatures is evidently one of these possibilities. Accordingly it seems likely that the congested conformation (I) is energetically favoured due to a weak attractive interaction between the naphthyl and phenyl rings.This view is supported by the fact that conformation (I) is energetically favoured in the 2-form atropisomer (la) which is itself favoured over both the other atropisomer (lb) and the E-isomer (lc). Regarding the nature of the possible attractive interaction in (la) it is clearly not a classical face-to-face charge transfer n-complex as the naphthyl and phenyl rings are mutually inclined at an angle of 48". The X-ray and NMR results suggest that there may be a weak attractive interaction of some sort between the naphthyl C(lO)-H and the n-electrons of the neighbouring phenyl ring. Attractive 'edge-to-face' interactions between two aromatic rings have recently been proposed in other systems.' 2-16 Both intramolecular '2,13 and intermolecular l4 attractive interactions have been discussed. Usually the aro- matic hydrogen is found to lie 2.8-2.9 8 above the centroid of a second aryl residue (c$ 2.7 8 in the present case).ab initio Molecular orbital calculations '5,16 on model interactions involving aromatic acid side chains have led to the suggestion that the edge-to-face effect lowers the enthalpy by 1-2.5 kcal mol-'. The variable temperature n.m.r. results (Scheme) suggest that the attractive edge-to-face interaction in (la) may be worth ca. 1.3 kcal mol-' in enthalpy terms i.e. within the range estimated by theoretical calculations.* For details of the CCDC deposition scheme see 'Instructions for Authors,' (1990) J. Chem. SOC.,Perkin Trans. 2,1990,issue 1. Experimental NMR spectra were recorded on JEOL GX-270 and FX-90Q spectrometers. Variable temperature n.m.r. studies were per- formed on the latter instrument and probe temperatures were measured using a digital copper-constantan thermocouple with the lead inserted into a sample tube at the level of the coil. Theoretical 6 us. temperature curves were generated from equations (1) and (2) (Scheme) using a BASIC program running on an Amstrad PCW9512 micro computer equipped with a Radio Shack graphic printer. N-[ 1-(l'-Naphthyl)ethylidene]-l-phenyl-2-propylamine(l).-This imine was obtained in 83% yield by heating under reflux in a Dean-Stark apparatus an equimolar mixture of 1-(1'-naphthy1)ethanone and R-( +)-1 -phenyl-2-propylamine [( +)-amphetamine] in toluene containing a trace of toluene-p- sulphonic acid. Recrystallization from pentane afforded colourless crystals m.p.97-98 "C [a]D -34.5" (CHCI,) (Found C 87.6; H 7.3; N 5.0. C21H21N requires C 87.8; H 7.3; N 4.9%); v, 1 640 cm-' (C=N); 6,(90 MHz; CDCl,) 1.05 [d CHMe (lb)] 1.11 [d CHMe (la)] 1.38 [d CHMe (lc)] 2.00 [s =Me (lc)] 2.30 [s %Me (la)] 2.33 [s,=CMe (lb)] 2.6-3.0 [m CH (la-)] 3.17 [m NCH (la) and (lb)] 4.11 [m NCH (lc)] 5.91 [d C(lO)-H (la)] 6.8-7.9 [m ArH (la+)]. L(67.8 MHz; CDCl,) 19.85 [=CMe (lc)] 21.67 [CHMe (lb)] 21.94 [CHMe (la) and (lc)] 29.18 [=CMe (la)] 29.34 [=CMe (lb)] 44.50 [CH (lc)] 44.69 [CH (lb)] 44.73 [CH (la)] 58.16 [CH (lc)] 59.65 [CH (lb)] 59.73 [CH (la)] 122.7-141.7 (numerous aromatic signals) 166.31 [C=N (lc)] 166.66 [GN (la)] 167.08 [C=N (lb)].X-Ray Structural Analysis and Rejinement.-A crystal (0.4 x 0.4 x 0.6 mm) of imine (1) was mounted on an Enraf- Nonius CAD-4 diffractometer; cell dimensions and intensi- ties were measured by w/28 scans with graphite-mono-chromated Mo-K radiation and scan range o = (1.0 + 0.35 tan 8)'. Scan speed varied from 1.8 to 5.0" min-' depending on the intensity. 2 568 reflections were scanned within 2 < 0 < 26". Of these 2 370 were unique (merging Rin = 0.018) (Friedel pairs not merged). Two standard reflections measured every 2 h showed no significant variation in intensity.1474 structure amplitudes with F > 50(q were considered observed and used in the analysis. Crystal Data.-CzlH ,N M = 287.4 orthorhombic P2,2'2' a = 7.821(1) b = 14.408(4) c = 15.104(11) A V = 1702.0 8 2 = 4 D = 1.122 g cm- F(OO0) = 616 p(Mo- K,) = 0.060mm-',h = 0.710698,. The structure was determined by direct methods and refined by least squares using anisotropic thermal parameters for the heavier atoms. Hydrogen atoms were placed in calculated positions riding on their respective bonded atoms. Isotropic temperature factors were refined for the hydrogen atoms. Unit weights which resulted in a satisfactory weighting analysis were used in the least-squares refinement. The refinement converged to R = 0.059 with a maximum shift/error ratio of 0.09.The residual electron density in a final difference map was within the range -0.19 to 0.37 e A-3.The inverse structure gave identical R. Complex neutral-atom scattering factors were employed. Computations were carried out on the University of Birmingham Honeywell computer and at the University of Manchester Regional Computer Centre with the SHELXS~,' SHELX76,' * and PLUTO ' programs. Lists of anisotropic thermal parameters and hydrogen atom co-ordinates for (1) have been deposited at the Cambridge Crystallographic Data Centre.* References 1 Part 22 A. J. Boyd D. R. Boyd N. D. Sharma K. Dev W. B. Jennings and 0.W. Howarth J. Chem. Soc. Perkin Trans. 1 1989 537. 2 D. R. Boyd W.B. Jennings and L. C. Waring J. Org. Chem. 1986 51,992. 3 D. R. Boyd S. Al-Showiman and W. B. Jennings J. Org. Chem. 1978,43,3335. 4 S. S. Al-Showiman J. Chern. Res. (S),1985,147. 5 W. B. Jennings V. E. Wilson D. R. Boyd and P. B. Coulter Org. Mugn. Reson. 1983 21,279. 6 W. B. Jennings S. Al-Showiman D. R. Boyd and R. M. Campbell J. Chem. Soc. Perkin Trans. 2,1976,1501. 7 D. Kost E. H. Carlson and M. Raban J. Chem. Soc. Chem. Commun. 1971,656. 8 C. P. Brock and J. D. Dunitz Actu Crystullogr. Sect. B 1982 38 2218 and references therein. 9 I. Natkaniec A. V. Belushkin W. Dyck H. Fuess and C. M. E. Zeyen Z. Kristallogr. 1983 163,285. 10 M. J. S. Dewar and W. Thiel J. Am. Chem. SOC.,1977,99,4899. The version of MNDO used was that contained in the program AMPAC written by the Dewar Research Group The University ofTexas USA. 11 C.E. Johnson and F. A. Bovey J. Chem. Phys. 1958,29 1012. J. CHEM. SOC. PERKIN TRANS. 2 1990 12 P. R. Ashton E. J. T. Chrystal J. P. Mathias K. P. Parry A. M. Z. Slawin N. Spencer J. F. Stoddart and D. J. Williams Tetrahedron Lett. 1987 28,6367 13 A. M. Z. Slawin N. Spencer J. F. Stoddart and D. J. Williams J. Chem. Soc. Chem. Commun. 1987,1070. 14 F. H. Kohnke J. F. Stoddart A. M. Z. Slawin and D. J. Williams Acta Crystallogr. Sect. C 1988,44 742 and references cited therein. 15 S. K. Burley and G. A. Petsko J. Am. Chem. Soc. 1986,108,7995. 16 G. Karlstrom P. Linse A. Wallqvist and B. Jonsson J. Am. Chem. SOC.,1983 105 3777. 17 G. M. Sheldrick SHELX~~,Program for Crystal Structure Solution University of Gottingen Federal Republic of Germany 1986. 18 G. M. Sheldrick sHELx76 Program for Crystal Structure Deter- mination University of Cambridge 1976. 19 W. D. S. Motherwell PLUTO Program for Plotting Molecular and Crystal Structures. Report ‘Cambridge Crystallographic Files,’ University of Manchester Regional Computer Centre Manchester 1981. Paper 9jO1767D Received 26th April 1989 Accepted 16th August 1989

ISSN:1472-779X    

DOI:10.1039/P29900000025

出版商:RSC    

年代:1990

数据来源: RSC