ABSTRACT PART I THE CRYSTAL AND MOLECULAR STRUCTURE OF DIBASKETENE PART II THE REACTION OF ALLYL LITHIUM WITH STILBENE BY Neil Evan Jones Photolysis of basketene, pentacyclo— [5.4.0.02’5.05’8.04’Z7 dec-9-ene, yields a dimer which crystallizes in the space group PI with Z a 1. It was necessary to use a combination of symbolic addition and Patterson methods to solve the structure of this com- pound. Dibasketene is a simple 2+2 cycloadduct which has the 223; stereochemistry about the central four- membered ring. PART II In an attempt to carry out a- £2+§7 anionic carbo- cycloaddition reaction, allyl lithium was reacted with stilbene. The product was l-methyl-2,3,4,5-tetraphenyl- cyclohexane. Similarly, B-methallyl lithium gave 1,1—dimethyl-2,3,4,S-tetraphenylcyclohexane; and a-meth— allyl lithium gave a small amount of gig-1,2-dimethyl- 3,4,5,6-tetraphenylcyclohexane. Neil Evan Jones The structures of these compounds were assigned on the basis of their spectral properties; and, conversely, their structural similarity made possible a very detailed interpretation of their spectra. A mechanism for the formation of these compounds, involving a series of 1,2 carbanion-olefin additions, is discussed. PART I THE CRYSTAL AND MOLECULAR STRUCTURE OF DIBASKETENE PART II THE REACTION OF ALLYL LITHIUM WITH STILBENE By Neil Evan Jones A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY' Department of Chemistry 1971 To Frances 11 ACKNOWLEDGMENTS The author wishes to thank Dr. Eugene LeGoff, Dr. Alexander Tulinsky, Dr. Donald Farnum, Dr. William Deadman, Dr. Alaistair Macdonald, and ur. Richard Vandlen for their advice and assistance during his research program. 111 TABLE OF CONTENTS Page PART I INTRODUCTION........................................2 EXPERIMENTAL........................................4 SOLUTION OF THE STRUCTURE...........................6 REFINEMENT OF THE STRUCTURE........................14 RESULTS AND DISCUSSION ..... ........................17 PART II INTRODUCTION.......................................29 RESULTS AND DISCUSSION.............................52 EXPERIMENTAL.......................................44 General Procedures............................44 1-methyl-2,5,4,S-tetraphenylcyclohexane (2)...45 Reaction of allyl lithium with stilbene under very mild conditions...0.0.0.0000000000000000046 trans-1,2-dipheny1cyc10pentane................#6 Attempted dehydrogenation of compound.2.......47 % l-dimethyl-2,3,4,S-tetraphenylcyclohexane 0..OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOSS 9 £5. cis-l,2-dimethyl-§,4,5,6-tetrapheny1cyclo- hexane (§)oooooooooooooooooooooooooooooooooooosg Attempted preparation of 1-benzy1-2,5,4,5- tetraphenylcyclohexane (9) or l-methyl- 2,3,4,5,6-pentaphenylcyclohexane (l2).........65 Attempted preparation of 2,5-diphenyl-3- methylcyclohexene (ll)........................65 iv Attempted reaction of cyclooctene with tri- phenylcyc10pr0pylidine phosphorane...........64 BIBLIOGRAPHYOOOOOOOOOOOOOOOOOOOOOOOCOO0.00.00.00.0065 LIST OF TABLES Table Page 1. Distribution of IEI's..........................6 2. Starting signs and symbols.....................7 3. Symbol and sign relationships..................8 4 . Interatomic distances and bond angles and their average 68d.8000000OOOOOOOOOOOOOOO0.0.0.20 5. Atomic parameters and their average esd's.....23 6. Calculated and observed structure factors.....24 7. Least-squares planes..........................25 8 . Statistical comparisons of bond lengths and angleSOOOOOOOOOOOOOOOOOOOOOOOOOOOOO...0.0.0.0026 LIST OF FIGURES Figure Page 1. A molecular fragment to be compared with C(3)’C(8)-C(9)-C(10) in Figure 30.0000000000009 2. The (hkO) Patterson projection...............12 5. The carbon skeleton of dibasketene...........19 4. The infrared spectrum of compound 2..........49 5. The ultraviolet spectrum of compound 2.......48 6 . (a) The 60 mo NMR spectrum of compound 2.....50 (b) The 100 mc NMR spectrum of compound 5....50 7. Irradiated frequency-sweep NMR spectrum of compound S.000......OOOOOOOOOOOOOOOOO00.0.0005]- 8. The NMR spectra of cis and trans l—methyl- 4-pheny1cyclohexane..........................52 9. The infrared spectrum of compound g..........56 10. The mass spectrum of compound 2..............55 11. The mass spectrum of compound Q..............57 12. The 60 mo NMR spectrum of compound é.........58 13. The calculated and observed NMR spectra of compound £000.00...OOOOOOOOOOOOOOI00.0.0.0...37 14. The infrared spectrum of compound g..........60 15. The mass spectrum of compound g..............61 16. The 60 mo NMR Spectrum of compound g.........62 17. A prOposed mechanism for the formation of compound 2000......OCOOOOCOOOOOCO0.0.0.00000042 18. The gas chromatogram of crude compound 5.....54' vi "...there is still some art left to crystal structure analysis..." G.H. Stout and L.H. Jensen vii Part I The Crystal and Molecular Structure of Dibasketene INTRODUCTION The hydrocarbon basketene, l, pentacyclo- £§.4.O.02’5.03’8.04’Z7 dec-9-ene, can be made from cyclooctatetraene in an overall yield of 12% (1). CO [>0 CO / PbI= (Aer/‘55 2 N where IUH’I = IFS" ’exp (is—:51)Zfi 1:1 and where f1 is the scattering factor of the i-th atom in the unit cell. SOLUTION OF THE STRUCTURE Symbolic Addition The distribution of the lEl's, shown in Table 1, established the centrosymmetry of the unit cell and, hence, of the molecule (4). Table 1. Distribution of [El's. Experimental Theoretical Centric Noncentric Ave. E2 1.000 1.000 1.000 Ave. [Ea-1| 0.954 0.968 0.736 Ave. [El 0.804 0.798 0.886 96 IE] 21 31.4 32.0 36.8 % |E|222 4.7 5.0 1.8 % IE|223 0.2 0.3 0.01 The symbolic addition method for direct sign deter- mination (5) was applied to the 202 reflections whose IEI's z 1.0. The origin of the unit cell was fixed by assigning positive signs to three linearly independent reflections; and the signs of four additional reflections were represented by symbols (Table 2). _Table 2. Starting signs and symbols. h k /' IE] Sign 6 -2 l 3.70 + -8 1 2 2.47 + 7 -1 l 2.42 + 6 -l O 2.33 a -7 l 2 2.49 b 3 3 0 2.25 c 5 -l 1 2.37 d The signs of the remaining 195 reflections were expressed in terms of these signs and symbols by means of the 22 formula sign of EEO asign onEbe-vlib Equation 1. each with a high certainty, C, of correctness, c =II

IIBk->”Eh—>_—k>|) and N is the number of nonhydrogen atoms in the unit cell. Relationships among the signs and symbols, given by individual terms on the right side of Equation 1, are shown in Table 3. Table 3. Symbol and sign relationships. Implication loglo C a = + 120.5 b = + 4.8 c = + 25.0 d = + 51.1 a = b 47.0 a = c 2.1 a = d 40.5 b = c 0.0 b = d 37.3 c = d 4.0 abc = + 0.5 abd = + 44.5 acd a + 3.4 bed a + 0.0 abcd = + 0.4 The data in Table 3 clearly imply that a = b . d a +, c i a or b or d and is therefore -; and this combina- tion gives a reasonable distribution of positive and negative signs to the 202 IEI's. Accordingly, an E map was computed with these signs. Wrong E map The E map contained a large number of peaks, and the following approach was used to try to separate the real from the Spurious: (a) The 24 strongest peaks in the asymmetric unit, of which a maximum of 10 could be real, were contoured. (b) For each of the 24 xi,yi,zi coordinates the values of the sharpened Patterson function at 2xi,2yi,2zi and at 1:1-1cj,y1-y:j,zi-z;j etc. were computed. A total of five peaks were thus found to be consistent with the Patterson function in both inter- and all igtggmolecular vectors. (c) Of these five peaks, four appeared to form a part of the assumed molecular structure (Figure 1); and one was within a bonding distance of its mate at -x,-y,-z. 1.58 R 104° 97° 1.55 R 1.53 2 Figure l. A molecular fragment to be compared with C(3)-C(8)-C(9)-C(10) in Figure 3. 10 (d) A structure factor calculation with the five peaks gave an R, R =- )3 llFol-IFCII/ 2m x 100%. of 60% but strongly contradicted the assumed sign of the symbol d. (e) A Fourier synthesis based on the five peaks resur- rected many of the doubtful peaks in the original E map and did not reveal a molecular structure. (f) Fearing that one or more of the five peaks was Spurious deSpite the rigorous selection process, three- dimensional atomic Patterson superpositions (6) were carried out at each using both the sum and minimum functions as a measure of agreement; but, again, no molecular structure was revealed. At this point the original choice of signs was called seriously into question. 11 Right E map In order to obtain phase information independent of the symbolic addition method, the (hkO) Patterson projection, Figure 2, was examined. The overall appear- ance of this projection suggests that the molecule strad- dles the (110) plane. the largest |F|, should lFllOI’ therefore be positive; and this is what Equation 1 indicated. Furthermore, the large peak at x = 0.04, y = 0.22 is the projection of the largest peak in the three-dimensional Patterson which occurs at z a 0.06 and is ca. 1.5 R from the origin. If this peak is the coin— cidence of the vectors between the eight pairs of atoms which form zero-atom bridges Spanning the mirror plane of dibasketene (see Figure 3), then the (110) plane is a good approximation to that mirror plane. To test this hypothesis, a model of dibasketene was built on the same scale as the projection map, the assumption was made that the center of symmetry of the molecule lies at fi,%,%; and approximate x,y atomic co- ordinates were obtained by placing the model on top of the map. A calculation of the (hkO) structure factors gave an R of only 63%; 222, of the 27 reflections whose [E] 's .>_ 1.0, the 19 for which [Fe] 2 2.0 were all con- sistent only with the sign combination a a +, b = c a d = -. An E map computed with this set of signs, which had many contradictions on the right side of Equation 1, con- tained only ten strong, sharp peaks and revealed the molecular structure on inSpection. .popmnm ohm hpwmnop ponmswmnmospooao swan mo mSOHwom .qoapoononm nomsoppmm onnv one .m onswwm 13 Afterthoughts There are two likely explanations for the fact that the correct sign combination was not the most self-consis- tent in terms of Equation 1. (a) Equation 1 was derived on the assumption of a ran- dom distribution of scattering matter in the unit cell (7). This is clearly not the case with dibasketene: The molecule is very compact and has a mirror plane that is not part of the Space group symmetry. (b) In solving the structure of the alkaloid jamine (8), which crystallizes in the Space group PI, Karle and Karle also found a beguilingly self-consistent set of signs which synthesized recognizable molecular fragments in the wrong places in the unit cell. They had to compute and examine several E maps to find the correct one. They ascribed their difficulties to the fact that, with a triclinic crystal, the only relationship among the F's that can be utilized in applying Equation 1 is F(hk/) . F(fiE/—); 1.8., there are no direct indications that a given symbol is negative. REFINEMENT OF THE STRUCTURE A Fourier synthesis with the observed structure factor amplitudes and the signs calculated from the atomic positions in the E map gave improved atomic coordinates. Full-matrix least squares refinement of the structure (9), beginning at R a 24.3%, was carried out in five stages. (a) One cycle of unit-weight refinement of the coordi- nates, isotropic thermal parameters, and scale factor reduced R to 18.5%. The average B was reduced from 1.79 to 1.47 and the scale, from 1.0 to 0.9. To determine the reason for the direction and magnitude of these changes, a difference map was computed; and the positions of nine of the ten hydrogen atoms were revealed. (b) The scale factor and the individual BJS were returned to their original values, and a structure factor calcu- lation including the hydrogen atoms gave an R of 15.7%. At this point a weighting scheme was decided upon. wc IFOI> = 1/02( |F0l> where N (’(‘Fol) ‘ IFOI £§LIIFOI1 - chlil/'Foli Equation 2L N 14 15 and where N is the number of [Foli's between [Fol and IF0|+ 1.0 and Fc = F calculated at R a 15.7%. Zero weight was given to Six strong, low-angle reflections which showed severe secondary extinction and to re- flections whose |F0l:51.0. Three cycles of refinement of the scale and the carbon-atom parameters, in the presence of the hydrogen atoms, reduced R to 12.7%. (c) The six reflections showing extinction were cor- rected by means of 8 Darwin plot; and two more cycles of refinement reduced R to 10.3%. (d) A second difference map was computed and the remaining hydrogen atom found. A final cycle of iso- trOpic refinement reduced R to 10.1%. (e) Two cycles of refinement with aniostrOpic thermal parameters for carbon reduced R to a final value of 9.1%. At this point parameter Shifts were less than 10% of their esd's. The R factor, omitting the weak, zero-weight reflections, is 8.6%. An R factor of 5-8% is considered reasonable for structures determined by comparable methods. In the case of dibasketene the rather high value of the final R is probably due largely to poor alignment and centering of the crystal during the data collection. The anisotrOpic refinement reduced the weighted R factor from 13.7% to 12.3%. Given the increase in the number of parameters, this improvement is significant, according to Hamilton's R-ratio test (10), at the 0.005 16 level assuming that all errors are random. A correlation coefficient is a measure of the inter— dependence of two parameters. If it is large, simultaneous least-squares refinement of the parameters may be Slow or may give incorrect results. Correlation coefficients between atomic positional parameters are generally less than 0.2 but will be greater than this when unit cell angles are Significantly different from 900 and when the atoms are related by crystallographic symmetry or pseudo- symmetry (11). Because the angle a in dibasketene is 115°, the correlation coefficients between the y and z coord- inates of individual carbon atoms were disturbingly high, ca. 0.4; but this did not prove to be an obstable to smooth refinement of the structure. The variance of a unit-weight reflection is given by the formula 2 V a JZwfl‘o—Fc) D.F. where D.F., the number of degrees of freedom, is equal to the number of reflections minus the number of param- eters describing the structure. If V is significantly greater than 1.0, the weighting scheme underestimates the average error in the structure factors. For dibasketene V = 1.00. This is reasonable since the weighting scheme, Equation 2, is based upon the average differences in the calculated and observed structure factors. RESULTS AND DISCUSSION The molecular geometry of dibasketene is Shown in Figure 3. The important interatomic distances and bond angles are given in Table 4. Table 5 gives the final atomic parameters; and Table 6 lists the calculated and observed structure factors. The average carbon-carbon bond length in dibasketene is 1.55 R; and the average carbon-hydrogen length is 1.07 R, exactly the values found for cubane (12). The information in Table 4 relating to 0(6) and 0(7) is con- sistently anomalous and suggests that there is decreased p-character in the Sigma bonds formed by these two atoms. It is an eXperimental fact (1) that when the double bond in basketene itself is hydrogenated, the reaction is complicated by the hydrogenolysis of the bond corres- ponding to C(6)-C(7). [H 7 The accuracy of the value of 1.52 R for the C(6)-C(7) bond length is confirmed by the reported structure of basketsne-9-ol p-bromobenzoate,ft(l3). 17 18 ,QOCC6H4Br This value is closer to the carbon-carbon bond length in cyclOpropane, 1.524 1 0.014 R (14), than to that in cyclo- butane, 1.548 t 0.003 R (15). In Table 7 the (110) plane is compared with the mirror plane of the molecule, taken as the least-squares plane of the midpoints between the mirrored atoms. The deviations of the midpoints from this plane are insig- nificant at the 95% level (16). Moreover, as shown in Table 8, there is no Significant difference in the arrangements of the crystallographically nonequivalent atoms on Opposite Sides of the mirror plane of the molecule. The negligible correlation coefficients found between the corresponding coordinates of the mirrored atoms during refinement of the structure prove that there is no true crystallographic mirror plane in the unit cell and that the Space group is indeed PI. Finally, the interatomic distances and bond angles in dibasketene are compared in Table 4 with the corres- ponding values reported for;fi. There are no significant differences between the two structures (see Table 8). l9 mmH.oHo.om.mHo ea.sHo m oaomowoopSSI .oammoofio H.mHo.m.oo.oH.mo.s.so.m.mo.Ha.mo.o.m.oHV Spam .onopoxmmpflp Ho Sopoaoxm Sopnmo one .wm onswam 2O Table 4. Interatomic distances and bond angles and their average esd's. Distances Angles 0(1)-0<2> 1.563 0(2) -0(1) -c<12> 89.90 C(1)-C(12) 1.54 (1.58)8 0(2) -0(1) -0(20) 115.3 0(1)—0(20) 1.55 (1.51) 0(12)-0(1) -0(20) 111.7 (112)8 C(2)-C(5) 1.55 (1.51) 0(1) -C(2) -C(3) 115.6 0(3)-0(4) 1.54 (1.52) 0(1) -0(2) -0(11) 90.1 0(3)-0(8) 1.55 (1.52) 0(3) -0(2) -C(ll) 112.1 (112) 0(4)-0(5) 1.55 (1.57) 0(2) -0(3) -C(4) 116.9 (117) 0(4)-C(7) 1.57 (1.55) C(2) -C(5) -C(8) 116.5 (117) 0(5)-0(6) 1.56 (1.55) 0(4) -0(3) -C(8) 86.8 (86) 0(5)-C(10) 1.55 (1.52) 0(3) -0(4) —0(5) 111.6 (112) 0(6)-0(7) 1.52 (1.52) 0(5) -C(4) -C(7) 90.2 (90) 0(6)-0(9) 1.57 (1.55) 0(5) -C(4) -C(7) 89.4 (89) C(7)-C(8) 1.57 (1.55) 0(4) -C(5) -C(6) 89.5 (89) 0(8)-0(9) 1.55 (1.57) 0(4) —0(5) —0(10) 111.1 (112) 0(9)-0(10) 1.54 (1.52) 0(6) -C(5) -0(10) 90.2 (90) 0(5)-C(10) 2.68 (2.65) 0(5) -C(6) -C(7) 90.9 (91) esd 0.018 (0.02) 0(5) -C(6) —0(9) 85.1 (85) 0(7) —C(6) -C(9) 90.5 (91) 0(4) -C(7) -C(6) 90.2 (91) 0(4) -C(7) -C(8) 85.5 (85) C(6) -C(7) —C(8) 90.6 (91) 0(5) —C(8) -C(7) 89.9 (90) 0(5) -0(8) -0(9) 110.6 (112) 0(7) -C(8) -C(9) 89.4 (89) 0(6) -C(9) -C(8) 89.5 (89) 0(6) -0(9) -0(10) 90.5 (90) 0(8) -C(9) -C(10) 112.1 (112) 0(5) -C(10)-C(9) 86.4 (86) 0(5) -0(10)-0(ll) 117.2 (117) 0(9) -c(10)-0(11) 116.2 (117) esd 0.5 (1) Table 4 (cont'd.) 21 0(1) -H(l) 1.158 0(2) -0(1) -H(l) 114° 0(2) -H(2) 1.09 0(12)-0(1) -H(l) 121 0(3) -H(3) 1.04 0(20)-0(1) -H(l) 105 0(4) -H(4) 1.07 0(1) -0(2) -H(2) 109 0(5) -H(5) 1.11 0(3) -0(2) —H(2) 111 0(6) -H(6) 0.95 C(11)-C(2) -H(2) 118 0(7) -H(7) 0.97 0(2) -C(5) -H(3) 111 0(8) -H(8) 1.08 0(4) -0(3) -H(3) 116 0(9) -H(9) 1.19 0(8) -C(5) -H(5) 107 C(lO)-H(10) 1.05 0(5) -0(4) -H(4) 122 esd 0.098 0(5) -0(4) -H(4) 116 C(7) -C(4) -H(4) 121 0(4) -C(5) -H(5) 111 C(6) -C(5) -H(5) 126 C(lO)-C(5) -H(5) 123 0(5) -C(6) -H(6) 113 0(7) -C(6) -H(6) 155 0(9) -C(6) -H(6) 127 0(4) -C(7) -H(7) 126 0(6) -C(7) -H(7) 155 0(8) -C(7) -H(7) 115 0(5) -C(8) -H(8) 124 0(7) -0(8) -H(8) 125 0(9) —C(8) -H(8) 114 0(6) -C(9) -H(9) 124 C(8) -C(9) -H(9) 109 C(10)-C(9) -H(9) 126 0(5) -C(10)—H(10) 110 0(9) -C(10)-H(10) 119 C(ll)-C(10)-H(10) 107 esd 5° h a flflie value of the correSponding distance or angle in 4. 22 .oasooaoa on» no mamas Mommas on» moonow powmso>w moonmop SA moawgd paon on» can maonpmwad SH mnpwmoa pnop one .pm onsmam mg: Tom m.om 0.06 m.mHH m.om . . N.m® ff...‘ 0 BHH f. A mm H nunu A. q . «H A v) a. .r t 98 V 7 \ a as 4.x g4 z a... mom ‘ a. e 23 OH0.0 666.6 466.6 -- 66.6 66 66 6 66 66 6 6666.6 6666.6 6666.6 666 666.6 666.6 666.6 66.6 46.6 66 66 66- 666 666 64 6666.6 6666.6 6666.6 66666 466.6 464.6 666.6 66.6 66.6 666 66 66- 666 466 66 6466.6 6666.6 6646.6 6666 666.6 666.6 666.6 66.6 64.6 66 4 66- 666 666 66 6646.6 6666.6 6666.6 6666 666.6 666.6- 666.6 66.6 66.6 666 66 66 666 666 66 6666.6 6666.6 6666.6 6666 666.6 666.6 666.6 66.6 66.6 666 66 66- 666 466 66 6466.6 6666.6 6666.6 6666 666.6 664.6 666.6 66.6 66.6 666 46 66- 466 666 66 6666.6 6666.6 6646.6 6666 646.6- 666.6 646.6 64.6 64.6 66 66 6- 666 666 66 6666.6 6446.6 6666.6 6466 666.6 666.6- 666.6 66.6 66.6 666 6- 66- 666 666 66 4664.6 6646.6 6666.6 6666 666.6 666.6 666.6 64.6 66.6 66 6 66- 666 666 66 6666.6 6666.6 6666.6 6666 646.6 666.6 666.6 66.6 66.6 666 66 66- 666 666 66 6666.6 6664.6 6664.6 6666 as New {on :66 36 6w 6w 6w 6m 666 6w o: E {a so: .mM\6qonpooHo mo 6:65 666 0.66 .m.Q .mpanon 066066 on» 6666 "mm mo means S6 666 m.m oaQoSpomfi one .44QH hp poHHawuasa comp o>mn m.m one .An\xmwmm + \anQN + MSNHQN +m\me +Nxmmm +mnHHQVIVQKo mmB coma cowmmohnxo popowm 065pmsogaop one .pmgfimos 0663 mumposwnmm Bout-Somoupmn on 999 ”popcop 666 mean £6633 op msouw nonnwo opp no omonp on 66066 new 6603 meopw Sowonpmn on» mo m.m map 606 .mopmSprooo Sopwuaopnmo wSHUSoamonnoo on» mo omonp 6086» m6 op 66:60 pom 6663 mopwcwpnooo Sopmlaowospmn on» go 6.660 on» pSoSoqamos wqwnsn .m.pwo mumsobm 660:» can whoposwnmn oaaopd .m manna 24 Calculated and observed structure factors. Table 6. The structure factors are on an absolute scale in units of electrons; F000 a 140 electrons. frAL l .n, -n ch 7 -I -I M -‘o‘ S. 'b 7. -L b ‘o‘ h 7 b b 0 P I 9 q g I b 7 I 2 I ‘ 'I 0. "oh 0 on I. 1 I Q g 7 7 I ? A § I I I I 17.5 . I o a o o o . SJOIII.I2.°)3'.‘0590.02??°5I23s.9,.1IQ“,I01.9?fiIQOOHQ‘A‘5.‘II‘°1I.I‘737351‘20o.¢Z.JIg.ssn“flfi o o o u I ,- ' 0.0.0.0... I... I. 00.00.000.00... OI DOOOIDDOOOICOOO.00.000.00.00...IOOOIOOOIOOIOOOOOOOOOC ,5 .I I.7I 59$? .3738'6 sI77II‘bI ”QII67QI? 9I1IQflhII??7§I°°III QOI§01 O°II5§ 25 “?6§I".' ' I' ' ' | ' I | LIOI?OIOI2I6I?I§I,IQIPIQIPIbIIIIZIflI20IIFIOI7IfillIbIPIh|91I2IIIGII2II2IbI?3hI?1?I/III‘I?IIII) II???JJOIIII?2273IJJOAQ£5559000|II22711IIII77771111b£66‘SRflflOII73IIIP277JJ1JBQQSOOIIII???1)1£ .................... ................... ............... .......... "6666600666666h66666656666657?777177777777717771777777777!HRAGIOIOR88088880850089900999909999 .. .-.o.....o.................-..o..-.................-.o.............-.-.............-.. LIOJQJIQIQ¢~0100l22716-U-9-m75010.hafla\mmb.1.21.006alb>W1mPWm1mnlm02032076Ao7lh1l777|59262509697505°? ..OOOOOIIOOOOOOIICIOOIO0.0-IIOOOCO.no.0.0.0...Ilo0000.00.00IOIOOIOOOIOOOOIOOOOOIOOIOOOOOOOICU C671I2§6I.77? 211355.!Ihflb?|31431$lfln.APQ7 l?.?9§bfl\ 0%!216660261?J?\§?77fl3155116 350|I6§b|67 'I7II . . . I. . .I...I .. I . . .. . . . I . . . II.. . .. . I. I . I . . . . . . . 5022\793.17066IIIQQOIOIIIQIHIbIQfiIAl96193u765l?HQ0!???4‘10SlIIfl7§I9IIIh§397RZ7A§I77666%?I69I6 .00. 0.. ....U...l....0.0.0....OOOOICIOOIIIIOOIIIOU...OOI.OI..0..........I.......C.OO0..I... OTIb3255 I3I277Q37A 9I5?5‘I‘6§11I7QI SPOb 7|10§SAI 7Al)\hh60?‘615?hh79¢9l3753lb QSI )765Ifi3 'I2II I I I I I I I II I I I L23.0I?3§0I2)0I20II?J§5I7I68I23h5I?169I?I6I?1|)|hI7Ihfll/IOI?0II?1OIZJQSI?JQSI?3b§|23§?IIlIhl? KnflOIIIII7777‘1‘66IIIII7)???‘IIIIbheht‘KfiShhhflnnnIIII?)?)IIIthIII77727111116~66£§§§§AhflnnnII o........-................ .oo....-................. thkhbbfihbtbbbfibbh‘bfibfihbtbbbfibhbbhhhbhfih«bk«HSSKSMHH55%SH555555559538555555555585855956hhhhh ............................................................................................ LobGOIhZAllbboS ZOYQQIIIr-uHCIh-klukl41.7wSun/unusanrvalslnwosnHOOIHHM-WsbIIaQSIIIISOs‘RQMIN-MUSIPA) .IOIIOIOOOIIOIOIOOOOOOOOOIOOOOIOIOIIona-Ioowlaooutwit...-IIIOOOOOOOIOOODODIIOOOOCODOOD3'00... (7?.IA?.16076I9I6599Ih|5|96785fifihiIlIIIP/IHHHQ?h.Infllfi«HI/IIINIIIOShZQRO EQQHIIQISI 1.69 .II 'I . . . . . .III .I.. 7 . .... . .. . 7 .. III . .. . .. .IIII .I .I. . I . . . .I .. . . . . . . . SQURVIRA?)?¢£UnYOIGIOIIUIIIOQRhnInIPIIn/IHI/OlsRQIIHIIIfluIINIHIBIIHRNHSKDPII$¢3II4610IKnnQIAO BIICOIIDOOOOIOIOOODOOOO9.0.0.000...oa-Ioa...oeoocuottoco-IIonIIUOOOOIQODOICOOIOIOIOOIIO..0... 07? 3?? ~60”?IOIIJVM3626666NOHAKIIIZIuhIRIblwlnh SIG/(athIRIthIJIISIS?I #06742896I IIYO I25 'I II I I III III-II I a I I I LI?)hflI’IOI’IOI70I’IhHI71b“I?I§HI)IbHI/I~I)1I)IHOIPIEOI)IOIIIOI)OI?I£HI2355I21§9I7365I9IhI’II KIIII2722IIIIQQQSIIIII7?2273IIIIbka655555h560000l||ll222fiII|InthfiIIIII?????JIJIJb§~6655§§6500 c.......................... ........................... H7??277?77777??7?72797??7???9??7?777?277222211111IIIIIIIIIIIIII‘IIII‘III‘IIIIII‘IIIIII1111116 ...............................................-......—..................................... L 109171non7OAAAGS)ISAOZhaldo-I?I 6260n2194514u«uHQQuIn255nnIQGOA1-wo «.s-WmMIISOSobsbSm-Mnandmmulzni ‘I....I..........I....II............Ol.0I..........l...l.0.I........OI.....I...............'. CI68I61973II56hJI?6§I5IIbI 6|I?IIIII2?.I6I Ifll.IH[HHIQHHIIAH5A22152hl555 III II IIQI «IIIhOH F . . .II II.. . . . .. . ... .I I7I ?II II. . . . .3 I . I .. . . . I. . . III . . . .. .. . . . . .. SQQQI‘IPIJOVYFQ620IIQI \thhIk'IIEI7hIflo‘9fiIHhukIfibhflONhMIKb2IQQI)QII69370066?IhIfi70I0Q6 ARII HI...0.00.0.0.00IOOIOIOOOOIIOODOIOSIn.III...II-soell-00-tooIIOOOOOOIOOOOIOOOOOOOCOCO-CO0.0... 0I9325.0I)?15051I‘hbnfiIIs HAIIVII‘HQ’) ’69 SARIIQIQAbKHHIIII'QIIRSI575b7 QJI QJIIIBI OQI‘Vnfi ' III II I I’ll ’II II JI I I III I?0I)flI?°I°0I?I7IOI/UI7flIHIflnI/IHIOIHHflIlIhHI/I«OI/ICI/IHI)“II/IBHI’IbHI?IQHI?‘65I?J~I)II/‘bfl KOIIIIZZIJQIII23‘000l||22I1hIII200IIPJIOOOfilllll)?))llIIkbbSkIIIII?????3I111hhhhtSfififihthflnfll ...-. .oo. . .o...........-............. H7]?7”'1'7I7711I”HNNU““HaflflflflnquoqWQQWIIIIIIIIIIII IIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII?I??? ...................................................... 113001....-7~692?7n..-?50674hull)unnurrWQISZObIIIIIA2685...).-.-Ia!4190.925428-689~u5|55ll$7|n2065I-M,fl¢h..w11- ‘0 0.0.00.0...0.0.0.0..COO-IIOOOOOOOOO0..0....lIIOOOOIIOI.000......0.00.00.00.00...0......000 (39,957II) 676 IQ.65627,)I49I.I1I07I.5“IZ.IH~7II INHIBHI.Ifl756?.l‘bflI)‘°fifibfiI7 .OIIIJC7.‘ .hI Fl . . .. ... II .I ... . ..I . . .. . . .. . I .. . . .. ... I . . I...7I . . . . . . . . 50222593203537ZOIQIZHOIIIIICNSQIOVOINHQSGCIIh66I463IJII5I.0QP¥QI$597006I3067H28332I0§0H33I83h H C.....I.l.....l..l...00......IICDOOO0..CD...ICOCO-OI.OI...’......O...OO...I.III...IO....... OSOB96???‘ lelh” QOQINIIIhOI JHIIQI In]? IQIIII INIRIH UOIhQ! I669I53I6566I85 99/«MSI 7| N? ' | | - I I I ll ' ' L0I)JUI83°IZ°I°I?IQIZIAIII6I93IIOI)IflIaIOIEIOIZOI00I/IIIII?II/I0I?IOI’IOI/OI)0I0°I23I3II3II°I IIIII2222333§OSIIII????JIIIa665%0000IIII?P)?IIJhtSIIII/(f‘II«6COOHOIIII???JIIthIIII??ZJJI§00 ................. ............. ........... Nbbtfih‘t‘thb‘bbbflfibbhbb6664666hasSHHQSSSSSSSSSQSHQHHHSHSH9%HSHShhhbbbhhhhhbhhflhhhflhhh65666577 L9fl59563I0700917b J‘IbQflblIIJIQS)IhlI?b“RIP?IHHSIIRI|9A[IQOIIQOIOQIhBSOQPIObII JI909317I7W0IHHH ‘ .0........OO.I.I..OOOICOOOIUDOOOOOOIODIOCOOOIDODUODOIDOOO...l............ .0......l.....l Cth I“9III6.72I5IIIII96609 SIII95670ZIJII27336II76.RIIQPI.IIHIIIJ??I37I626II§~ “556.5QII9I73III 5.. 2 . . P .I. . I. .. I . 2. .. . . .I . . . .. .. )oI II II?.. . ..I.. . a . . . c . .. . Shbfl§9217~QI790°IIIIIIh9690200Ihfl97.05IHIU)N§U§QSOAIHQQ)IIM51hI79§I575396§7557.66I86IAI60036I BOOIOOOOOOOOOIOOOIOOOOOIOUOOOOOOOCOOOOOIIODOOOOOOIIOOOOOOIIIOOOOOOIOOIII .00.........00.0...I O°I6Z30III3I7?I?.I 2I0‘60905 HAQAICHIHQIH??7J77I,6 ?I75?II7IHI IIJJIJn372SNQ§5576I55§?0I..?I3 fl , I I I I 2 I II III II L)SIPOIZJhCI?)00I210I230I20I|?Ihl?)6l)IGIIJQI230I3160I9I6OIIJOIIOIPOIIIJbIIJ6I236I236I730I?36 ISSOOOOOOOIIIII222233II.‘hSSIIII?27PIJIIh6‘h555°00fl0IIIII?’??I‘I.§QSSIIII77223331§“655$.oooo o... o.................. .......o...-....o.. NIIII22222222?2222??2222??/?2????22222222I??Z??JJ3)JIJIIJJI1|1333333‘33333333333333333JJb065‘ L62326653§I7OU§200250269b9°°66607297597???696§952996I60561NQIQI§ZCIGPK§I06559I8692§‘0586.I°Bb ‘0... 0.0.00000000000 .IOIICOOOOOIIIOIOOIOIOOOIIOIIOOCI. 00...... C...l¢...............l.. C7I3I957?IZ§252375?65 Iflbflfis? b.25lb I32675b37I1I76J050Ik5294llI89IPII0§0653I9379.053 96IJII6 '6...... . . o .. I . .J.. . . . . . I .. I? I 5.. . .I ..II.I II. .. . . I .. . o . . o . . . o 5123.0639400I053700950. QQOIIJII60012506555I25770734!?IQ)flIIbflIhIIbOJJI 0:3596563066 090627769269 8.0.00.0...OIIIIOOOOOO.I .OII...IOOIOIOOO'OIOIOOOOQOOIICOOIIOIIUOOOOOOIOOOOC..... ........ OMZQIOSIZ 0.252.75I660I766652 6 2586 3324756I7I660n5HI63II019595I303?HI060563I°§.9 TOJ7QTJJII. ' . I I II ' I LIZJQOI?‘§.I230I,30IZOII?‘§SI?3.§I2165I?Ifi‘I73bI?II2‘60I?Ib0I)IOI?I°I7°I0IIJ.I23.‘I23.5I7I.I, IOOOOIIIIIZZZZJJJJQQQSSIIIII222223JJ336.6b655556660000IIIII22?23333§bbSSOIIIIZZZIZSJJJJQd 655 ...... ...... ......... .o-.-..... co. H000.°°°°°°°°.°°°oooooooooooooooooo000000000000OOOIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 25 Table 7. Least—squares planes. The equation of the plane is Ax + By' + Cz‘ = D where x, y‘, and z* are reSpectively orthogonalized axes, and D is set equal to the perpendicular distance from the origin to the plane. Plane A B 0 D (110) 0.5343 0.7198 0.4432 5.3901 8 ma 0.5052 0.8291 0.2397 4.9759 (4531.)b 0.4835 0.8436 0.2335 4.8780 aDeviations of the interatomic midpoints from this plane: 0(1)-0(2) 0.0012 R 0(3)-0(10)-0.0014 0(4)-0(5) -0.0005 0(6)-0(7) 0.0013 C(8)-0(9) -0.0005 b This plane is determined by the points (1,0,0), (0,.8,0), and (-4,0,0) and contains the point (320606). 26 Table 8. Statistical comparisons of bond lengths and angles. The largest difference in C-C—C bond angles related by the mirror plane of dibasketene is 1.50. The esd of this difference is 0.850. The probability of observing so large a relative difference is 0.08. Since there are 15 such pairs of mirrored angles, this difference is not significant. Similarly: Compared Largest Esd of Probability No.of quantities difference difference pairs mirrored 13° 7.4° 0.08 15 C-C-H angles mirrored 0.013 0.0148 0.48 5 0-0 lengths mirrored 0.118 0.128 0.36 5 C—H lengths 0-0 lengths 0.048 0.0228 0.07 15 in g and 4 0-0—0 angles l.4° 1.12° 0.42 26 in g and 4 "Chemistry remains an experimental science." R.B. Woodward 27 PART II The Reaction of Allyl Lithium with Stilbene 28 INTRODUCTION The Woodward-Hoffmann orbital symmetry rules (17) predict that the following thermal [E’+ 27'cycloaddition reactions will occur, or not occur, concertedly: / ‘\\ + Ci) -——————4> lllllll’ic) (a) H . 6} __+_.. (>6 6) //' \\_ + (i) -———*———’- lllIII’IC) (C) II 1 Ci) ———————+~ [:::>>C) (d) A reaction of the type (a) has been carried out (18); but its concertedness has not been demonstrated. 29 3O Iiil’I + ‘::i>‘CH5 ' ' 0H3 Many cases of concerted heterocycloadditions of the type (d) are known. The largest class of these con- sists of the 1,3-dipolar additions, for example 6 G C6H50QEN//N06H5 + 0685\CH_CH,0685 ——-———6> C6H5 C H C6H5 N/’ C H which have been extensively studied and reviewed (19) by Huisgen. An anionic cycloaddition of the type (d) has also been carried out (20); but, again, its con— certedness has not been demonstrated. H C) CGHSCCEN/’CHC6H5 * 0635‘01160114.06115 "‘“” 06H5.\ 0 H 6 5 6 5 6 5 /Z:Ti:g\ /2:if:g\ C H N C H C H N C H 31 This reaction probably succeeds, whether it is concerted or not, because the negative charge in the product is localized on a more electronegative atom than carbon; and it is probably because of the thermodynamically unfavor- able charge localization that no pure carbocycloadditions of the type (d) are known (21). Allyl anions have been observed to react with certain terminal olefins, but only by 1,2 addition (22) L// and to initiate the polymerization of styrene (23). However, it was h0ped that a nonterminal, polymerization- resistant olefin could be found whose reaction with an allyl anion would, under the right experimental conditions, be of the type (d). RESULTS AND DISCUSSION Accordingly, the following reaction was tried: An ethereal solution of allyl lithium was refluxed with an equimolar amount of Egggg-stilbene for an hour. The usual workup procedures produced a yellow oil, contain- ing no stilbene, which solidified after standing for several days. Recrystallization of the crude solid gave a white powder, 2, with a melting point of 145-47°. The two expected products, 22 and 2b, melt at 660 (24) and (25° (25) reSpectively. / 06115 06115 6 The characterization of the actual product required a thorough spectroS00pic investigation. Elemental analysis and mass Spectrometry established the empirical formula of’2 as 031H3o--a molecule con- structed from £32 stilbene molecules, an allyl group, and a hydrogen atom. The infrared spectrum of 5, Figure 4, Shows both 32 33 saturated aliphatic and aromatic C-H stretching bands and a typical monoalkyl substituted phenyl absorption pattern; but it does not Show, as does 2b, a vinyl group band at 1640 cm'l. 20 was therefore immediately excluded as a possible structure. 06H5 06H5 /// 20 CSHS 06HS -.-n1-....._l .—..._...._1 6. _ . I The ultraviolet Spectrum of 2, Figure 5, has Amax = 260 nm with fine structure and 6 :- 891 l/cm-mole. The four phenyl rings in 2 are therefore unconjugated, and the remaining double bond equivalent required by the empirical formula may well be a ring. The 60 mc NMR Spectrum of 2, Figure 6a, Shows the 20 aryl hydrogens as a pair of pseudo singlets at 87.00 and 56.76 and the 10 aliphatic hydrogens as a complex multiplet between 33.8 and 51.0. On the basis of this information, structure 2d, which could have arisen from a ZZ’+ 27'0ycloaddition followed by a simple 1,2 addition to stilbene, was postulated. 0 H 6 5‘. C635 50- 0 H 0685 6 5 34 However, structure Se is also consistent with the above Spectral data; and it alone can explain the 100 m0 NMR Spectrum of 2, Figure 6b. CH 0635 3 C6H5 (a) + 06HSCHSCH06H5 ———. 2e C685 06H5 At 100 mc the 4 benzylic hydrogens in 2e appear as a multiplet at 63.79-2.76, the 3 homobenzylic hydrogens as a multiplet at 62.51-1.86, and the methyl group as a doublet, J . 7.0 06, at 51.09. Furthermore, irradiation of the Spectrum at 52.33 causes the methyl group doublet to collapse to a broad singlet (see Figure 7). Nothing can be said about the stereochemistry of 2. Its methyl resonance does resemble that of giggl-methyl- 4-phenylcyclohexane, in which the methyl group is con- fined almost exclusively to an axial position, and is quite unlike that of the correSponding Egggg isomer in which the methyl group is necessarily equatorial (26) (see Figure 8); but the structures are too dissimilar to allow any reliable stereochemical inferences. 2 could not be dehydrogenated, by any of a variety of methods, to l-methyl-2,3,4,5-tetraphenylbenzene. Confirmation of the structural assignment 2e was there- fore sought and found in the comparative Spectral prOp- erties of g, obtained by reacting B-methallyl lithium with stilbene. 35 CH5 06356 H 9) CH3 + ceases.caceas——> CH3 . 6 5 6 06115 ~ 0635 Elemental analysis and mass spectrometry established the ‘ ‘4‘ I! empirical formula of Q as c52352. L- iu-A- —A The infrared spectrum of Q, Figure 9, closely re- sembles that of S but has a strong doublet between 1 1550 and 1400 cm" characteristic of the gem-dimethyl group. The strong band in the infrared Spectrum of 2 at 1375 cm"1 can therefore be assigned to the scissoring vibration of the methyl hydrogens. The mass Spectral fragmentation pattern of g aids in the understanding of that of 2. Peaks at m/e - 180 (stilbene) and m/e a 91 (tropylium ion) are prominent in both Spectra, Figures 10 and 11; but the base peak in the spectrum of Q occurs at m/e = 145 and undoubtedly arises in the following way: CH3 C6H5 CD CH5 035 c H . on o 65 i» y 3 <:>H 0635 H 06H5 06H5 56 By analogy, the moderately abundant fragment in the Spec- trum of 2 at m/e a 131 can be accounted for in two ways. C H' 6 5 69 cm3 H on Cefls C H 3 6 5 -H' H C H -——-» @H .. j ® 6 5 H H H C635 The 60 mo NMR Spectrum of 6, Figure 12, shows the CH C6HS 20 aryl hydrogens as a pair of pseudo singlets at 57.01 and 56.80, the four benzylic hydrogens as a multiplet at 55.8-2.7, the two homobenzylic hydrogens as a multi- plet at 52.4-1.7, and the methyl groups as a pair of singlets at 51.25 and 50.88. The benzylic and homobenzylic hydrogens of 6 con- stitute a Six-Spin ABCDMN system. An analysis of the IN portion of the spectrum was carried out using the LAOCN} computer program (27). Initial input data were taken from compound 7 (28); v1 - V2 ' 709 09 J1,2 . -1204 ” J1.) . 1204 w . 4.} ob J2,3 57 and it was assumed that all the phenyl rings occupy equatorial positions, hence the Jvic‘s - 12.4 as and that V“ a vs a 53.1 and V} =- V6 . 53.4. The chemical shifts and coupling constants which best fit the observed Spectrum of §_are listed below; and the calculated and observed Spectra are shown in Figure 13. 1..-,“ 116.1 (:0.5) as from mus 110.7 (:0.3) as from mus -1u.0 ($0.1) as J1’3 12.6 (10.6) 06 .12,3 = 3.2 (10.6) as ‘E ‘E N H a u The structural similarity of Q and.Z is confirmed by these data. ‘ A A A A - Figure 13. The calculated and observed NMR Spectra'of compound 6. 38 To explore this reaction further, a-methallyl lithium was reacted with stilbene, and the only product that could be isolated was 8. 0635 (a) + c6HSCHsCHCGHS—‘i’é—> 0333:106H5 s H; CH5 C6HS “3 06115 Elemental analysis and mass spectrometry established the _-.-_-_- .r ‘. ._.:. empirical formula of 8 as 052H32. The infrared Spectrum of 8, Figure 14, very closely resembles that of 2, but the band at 1375 cm-1, the scissoring vibration of the methyl hydrogens, is relatively more intense. The mass Spectrum of 8, Figure 15, is also quite Similar to that of 2, but, as in the case of 6, the base peak occurs at m/e a 145. CH C635 CH H 0 H -:-;——4> (D H C6H5 C H 65 The 60 mc NHR spectrum of 8, Figure 16, Shows the 20 aryl hydrogens as a multiplet at 57.1 and a pseudo singlet at 56.87, the 4 benzylic and 2 homobenzylic hydrogens as a very complex multiplet between 54.0 and 39 51.9, and the methyl groups as a pair of doublets at 51.00, J a 7.6 as, and 60.75, J . 6.2 as. If an all equatorial, all 1,2 Eggng, arrangement of the phenyl rings can be assumed, the methyl groups must be gig by virtue of their magnetic nonequivalence. While there is precedent for the reaction of the a-methallyl carbanion with olefins at 0-1 rather than at C-3 (22), it must be emphasized that 8 is merely the only product that has been isolated from the reaction mixture. Other isomers of 8 may account for the remaining 96% of the yield. 9) CH5 2 C6HSCH-CHC6Hi///<: ‘:\\\‘? C6HSCH-CH06HS CH30H2 6H5 3 CsHs C H 6 5 5 5 no A pr0posed mechanism for the reaction of allyl lithium with stilbene is presented in Figure 17. Several experimental observations are relevant to this mechanism. (a) Both gig and trans stilbene react with allyl lithium to give only 2° This requires intermediates like the 4,5-dipheny1-l-pentene-5-y1 anion in which free rotation about the bond between the benzylic carbons is possible. (b) When the reaction is worked up with D20 instead of H20 no deuterium is incorporated into 2. The solvent is the most likely internal proton source. (c) Although the reaction produces 2 in a nearly quanti- tative yield, a small amount of trans-1,2-diphenylcyclo- pentane is in fact formed and can be detected and iden- tified by gas-liquid chromatography (see Figure 18). (d) The cleavage of allyl phenyl ether with lithium gives allyl lithium in only about 60% yield (29); but no unreacted ether contaminates the crude product 2. Lithium phenoxide is a byproduct in the cleavage reaction. Furthermore, when the blood—red solution of allyl lithium is added to the stilbene the mixture turns black; but the black color is completely discharged on quenching with water. The influence, if any, of the source of the allyl lithium on the course of the reaction could be ascer- tained by preparing the allyl lithium solution by cleav- age of allyltriphenyltin with phenyllithium (30). (e) When attempts were made to extend the scope of this reaction still further by preparing 2 or 19 by reaction 41 of cinnamyl lithium with stilbene and by preparing 1; by reaction of allyl lithium with diphenylbutadiene, only intractable gums resulted. C6HS C6HS C H CH CH CH CH C6H50H2 6 5 Cefis Ce 5 6 5 06H5 C H c 2 10 a 42 .m onaonaoo no Soapwahou on» you Smanwsooa pooonoum 4 mausgom assess-Bumps mmoo mmwo Lilli .AIIIII mmwo mmo ®§\< mmoo mmmo mmoo _ .SH magmas mmmo av mfl m mo mmoo mxu mmmm mums AVm mmoo / m.mmo / \ mmmom mmo «amok Immoo \. ® 45 Another approach to the [2+27 anionic carbocyclo- addition reaction was also tried: 616 (+39 (06H5)3p—<] + . —-——-> (06%)}? C. This reaction has greater potential synthetic usefulness than a reaction between an allyl anion and an olefin; but it failed to occur. EXPERIMENTAL General Procedures Infrared Spectra were recorded on a Perkin-Elmer Model 137 Spectrophotometer using KBr discs. NIB Spectra were taken on a varian T—60, A-56-60, or HA-100 spec- trometer with samples in CD013 solution. Chemical Shifts are reported in ppm from tetramethylsilane. Ultraviolet Spectra were recorded on a Uhicam lodel SP-800 Spectro- photometer using 1 cm quartz cells. lass Spectra were obtained with a Hitachi Perkin-Elmer EMU-6 mass Spec- trometer. Gas chromatographic analyses were done on a Varian Aerograph 1200 instrument. Melting points were determined on a Thomas-Hoover capillary melting point apparatus and are uncorrected. Computer calculations of NIH Spectra were done on a CDC-3600 computer. Microanalyses were performed by the Spang Micro- analytical Laboratory, Ann Arbor, Michigan. 45 l-methyl-2,3J4,5-tetraphenylcyclohexane (5). A solution of allyl lithium in 75 m1 of 2:1 (v/v) tetrahydrofuran—ether, prepared in 33. 60% yield by cleavage of 0.05 mole of allyl phenyl ether with a l2—fold excess of cut lithium wire (29), was poured through a porcelan Buchner funnel-sanS-filter paper into a 200 m1 : 2-neck flask containing 9.0 gm (0.05 mole) of trans- r ‘ stilbene (Note 1) and equipped with a reflux condenser and a nitrogen inlet and outlet system. The mixture was refluxed under nitrogen for an hour and then poured onto ice; and the product was taken up in methylene chloride, 5; dried, and thoroughly freed of solvent under vacuum. The resulting yellow oil was dissolved in a boiling mix— ture of 50 m1 of 95% ethanol and 10 m1 of carbon tetra- chloride from which the product crystallized after stand- ing for 12 hr under refrigeration. A second recrystal- lization from 25 m1 of heptane afforded 3.9 gm (39% yield based on stilbene) (Note 2) of a white powder, mp 145-470. 5221. calcd for C31H30: C, 92.49; H, 7.51. Found: C, 92.39; H, 7.66. Note 1: gig-stilbene may be used in place of £3555- stilbene in the above procedure; but the allyl lithium solution must then be added slowly because of the greater initial vigor of the reaction. Note 2: The IR and NIH Spectra of the crude oil and the purified solid are nearly identical. The actual 46 yield of 5 is therefore almost quantitative. The crude oil was separated into its components by gas-liquid chromatography'(5% SE—30/ChromosorbA' column at 210- 250°). The resulting chromatogram is shown in Figure 18. The peak corresponding to Bragg-l,2-diphenylcyclopentane was determined by addition of an authentic sample to the oil. Reaction of allyl lithium with stilbene under very mild conditions. The above procedure was modified in the following way: The solution of allyl lithium was added dr0pwise over a period of 20 min to only 4.5 gm (0.025 mole) of trggg— stilbene dissolved in 50 ml of tetrahydrofuran which was kept at 00 by means of an ice-Salt bath. The mixture was stirred for an additional 10 min and was then worked up to the crude oil stage. Gas chromatographic analysis of the oil Showed that the reaction had gone to completion. trans-142-diphenylcyclopentane. This compound was prepared by Clemmensen reduction of Egggg-3,4—diphenylcyclopentanone (31) according to Wiedlich's procedure (24). The product had the correct melting point, 66°; and its infrared Spectrum showed complete reduction of the carbonyl group. 47 Attempted dehydrogenation of compound 5. Compound 2 was recovered unchanged after attempted dehydrogenation with palladium on charcoal in refluxing cymene (32) and with sulfur at 2200 (53) and was destroyed by attempted dehydrogenation with selenium at 3200 (33). as noaoxoflo .m> oquowu as I meaoo.ov .m onsonaoo no asnpoono poaoabmhpas one .m osswam O . . , 2... new. 03. (o. :0 WE O“ 1 (C '5 J u - ll 1 Q . . . fl. C 0 . W ......... I" t . . . .1 . a. A I. H A w n . . . :50 -.L A J. .l .1 I 0 ‘ I .l I - Q.I.O a I I l 1| l.'|l‘ 0....-- Ii ‘ . . . . . . .._ ..a.._ . . . . . . ..u a n . .L . . . , . 0.. .. a .. JtiIIth‘ I Dil- .I.I.Il.- .ml III-5.! b It vi | . “1.115..-. A . . . . m . .. .11 e . .u L .. -.. .. . cut I. o l . l .1 I - .. 0 J i . 1 .N . .. -. n l _ . . . i 1.0 . r t IIIILA :0- I. 1 I. P .u. 5‘ .lAluA _. o . . A . u... . n - A . l .. o _ - .. - 1 *. ~ 0 ' . V. IQ,‘ # \l ..w1 1.5..19... til-1-..-.II'III I .0-..“ ' .. u C. I! .‘i A . . .. .c A . . . - . . s . - l ‘ .un.. e 3. . o . It: rl > ‘llidllll III -115 I 5 l... ‘r‘ll I III I t .| ID. ........ 5....A u ....--»--- . -o . .......... . v-01 . ... .h. l . it‘lllvl . . .I t .. . 05.491 I h -5 O OIOII.‘ 85...-.50'1 51'“ ‘5 in. .1 9| t I .I I l Ill-1.55 m o . .. -: : .,; .. -.:. l n .A -0 , |. A. I 5 l .‘Ituk . . ..... A o e . toll! n 1... 3--.}--3L- 1.3.11-3..-112 112.»... u . a ..... I. O. ..... a I O- In m_o .-. -e-: I. .0. . t 1-5.41.0A ll - o, . 5.3!--1 » U..- Vllitlll. -55115 9.4 I... .l O: n V l W . . .0l09A 2 m o m .-: mnu - - z - --- - --. a.-- :.- . . ., -.. _ m o _ 1, -- . .-. ..... . _ . . _ m . . 1 K . -w LIL. « -¢ 5... « m p . . .A 1 .’ V v] - ' .71 .I .II . T. fl ....... II n . ....... A . . . . . . 01 — ' I -~ .lll’» . . . l.- 15.. ,..:| II. w-‘ t 6.5 .. wlc.cvto.0§L . . , . I. IL 1 . . . . ..... ”:17--- to ell V I 111‘ liIll't Ill: V - I T -|¢U.§:..Isrltlvll.llut.trll!l' 10:30-51". 1055 . .u ..... . . ., .. ..... ....ri .c a . . .»-. Littl H s * e ...... ...hr.o.ubu“.l.004 y . ...... . I ... | cuoilf'ol I!!! I‘ll. k .r L [V . ?' O saueqlosqs 49 z .m 6589.80 no 5.30on ponmnmqa one :a onsmfim $205.3; 15253;.) m. : 9 a m a 6 n 4 m 3 mmwo .HH 4mm,. ,M ammo , ammo mmo ammo $36 9.1... 1...... fir: ;. 2.... (6.” Eco 000m 000». 50 .N 656980 no 5.30on S o... 00H one .m ensonaoo so asupooam man as om one QM u 4‘4H41 444444 F(-1tfi ‘4 4151‘ 4 d #1 mmoo mmoo mmoo ixk to. §-8-§*3—3 51 V V I Y t I I Y Y I I V I 1' Y I V Y W I Y Y Ifi I fl 1' I v I I v I v 1 I00 O0. )0. 100 I. O h HID A l . . 1 I n . l . . 1 . I . . 1 I . . . . I l - I - . I . I 1 . I . . . - I . - - n I . n - n I u 7.0 0.0 0.0 "g . o 1 e e 1.0 In L. e r' Y 7' I Y Y Y T I Y' Y Y T I V V V Y I fi' Y Y L T V W I V ‘I' Y J V V I f V Ifi‘ I ' 3 I ' I ’ T l ‘ I T no no no see It. 0 lb rl’ 1 . I 0‘ l’ .1 L A A l A A ‘ A A l A A l A l g L 4 I J L I A A J A m L A I A A 4 I L A A A I A A J A T A A 4‘ I J to 7 o so I. "I ‘ a 4.0 to to 1.0 o Figure 7. Irradiated of compound 2. frequency-sweep NIR spectrum 52 f"~*—=“"1 N d W Figure 8. The NIH spectra of cis and trans l-methyl- a-phenylcyclohexane. z .__ _a_a 5 5 mmwo mmoo mum m a W (- (v ___5 .L ) (I5 ”I: I» 0 C C II a m. 1 .m vqsomaoo Ho aaupooam mama was .oH anamwm m L A." 0 ..L C 9 .. V m 0 Nu O m _ _ z: Ilsa! . 03/» on 65 210° ———.I [ 250° _..___, Figure 18. The gas chromatogram of crude compound 5. ~ 55 lil-dimethyl-2,5,4,S-tetraphegylcyclohexanegLé). A solution of B-methallyl lithium, prepared by cleavage of B-methallyl phenyl ether (34) with lithium, was reacted, as in the preparation of 2, with 9.0 gm (0.05 mole) of Egggg-stilbene. The crude oil was re- crystallized from ethanol-carbon tetrachloride only to give 4.8 gm (46% yield based on stilbene) of a white powder, mp 155-550. Aggl. calcd for 032H32: C, 92.26; H, 7.74. Found: C, 92.15; H, 7.80. .M 25258 no Epoomm v0.3.3.5 one .w 93mg . ...wZOw..:<<_. IhOZmAm><>> m.— E m: m_ : 9 y o m N o m. v m OE com _ 8° 89 .90 82 88 8.8 89. 57 092 __E=-qt 093 .m cadoaaoo mo aanuooam mama one 7.. 0 U —— U". ;m m a: E m moo m m m mo co m 0 UHI (u I V _ U w w .u 0 _ 3.. ___l_ .w .HH ohnwwh n 1... l .o c. W C. U U U U 0 L C lb 0 nlv 58 .w. 6559.80 no 3390on ml: on 00 95 .NH enamam O O.— O.“ 0.0 0.0 . a .111: 0‘ 0.. OK 0.. .,_ 4 m .4 _ . . loll; . . . Ax+ld . q . q . + Hy. . or. _ I 1 4 J q ‘ - ‘ d mmmo ammo m \ “mo ammo mac fl mums A...A I: O OO— SN 80 80 . _ . t P . c . L t . . . a. . _ .x_ _ . _ t a»- p - . _\+ . . _ . . . a. he 59 cis-1,2-dimethyl-3,4,5,6-tetraphenzlgyclohexanegfia). A solution of a-methallyl lithium, prepared by cleavage of a-methallyl phenyl ether (55) with lithium, was reacted, as in the preparation of 2, with 9.0 gm (0.05 mole) of trans-stilbene. The crude oil was triturated with 20 ml of pentane; and the solid which separated was filtered off, washed with pentane, and recrystallized from 5 m1 of heptane to give 370 mg (3.5% yield based on stilbene) of a white powder, mp 17a-ao°. éflfll- calcd for 032332: C, 92.26; H, 7.74. Found: C, 92.07; H, 7.76. 60 w— V. CON m— .m announce mo adhuoonm command“ was @205; $053335 0. o m a oom coo 89 .26 82 .3 9:6: Doom open coo: 61 - on ~a- (A! _— OLC ‘ o-az .m onsonaoo no sapwooam name one 0% 0‘2 022 OIZ (:6! mmmo moo ammo m m moo w m IL :U ‘ ' O “I Oil (all 00! .mH unawam 62 ...m.. 0:509:00 Mo 29003 a 05 ow one. .ma 0.5..qu o o- on 0.0 0.. ”0.: an 0.0 Os 0. * 1 I — ‘ 1 ‘ THJ I ‘ Ill \— ‘ ll ‘ I. - 1 4‘ q ‘1 I ‘ ‘ 1 — ‘ 1 I d _ d ‘ T i \A—[ q j J ‘l T - d 1 - 1 ‘ 4 d 1 4 <1 I - d l 1 1- m m woo $00 $8 \ mmoo n no ammo AxA £0 8— 88 80 8' 8n . _. ».FL. . .r +P F...»—t.»L_»..._.»..—L... ..p»—....H.L.L~... 65 Attempted preparation of l-bengyl-2,§,4,§-tetraphenyl- cyclohexane g9) or l-methyl-2,§,4,§,65pentaphenylcyclo- hexane (10)! A solution of cinnamyl lithium, prepared by cleavage of cinnamyl phenyl ether (56) with lithium, was reacted, as in the preparation of’2. with 9.0 gm (0.05 mole) of Egggg- stilbene. The resulting viscous oil would not give a solid product on standing or after trituration with meth— anol or with pentane and was eluted with chloroform from a silicic acid column as a single broad band. Attempted preparation of 2,S-diphenyl-fi-methylcyclohexene (11). A solution of allyl lithium was reacted, as in the prepa- ration of 2. with 5.15 gm (0.025 mole) of gagg§,§£gg§-l,4- diphenylbutadiene. The resulting viscous oil would not give a solid product on standing or after trituration with methanol or with pentane and was eluted with chloroform from a silicic acid column as a single broad band. 64 Attempted reaction of cyclooctene with triphenylczclogro- pylidine phosphorane (57). 2.5 gm (0.005 mole) of triphenyl-B-bromOprOpyl phOSphonium bromide (58) were placed in a 100 ml 5-neck flask equipped with a serum cap, a reflux condenser, and a nitrogen inlet and outlet system; and 1 gm of a 54% sodium hydride suSpension, washed free of mineral oil with pentane, was slurried with 25 ml of dry glyme and poured in. The flask was swept with nitrogen, and the contents were heated to reflux under a static nitrogen pressure. One drop of ethanol was injected; and when, after a few minutes, a bright yellow color indicated the presence of ylid, 0.55 gm (0.005 mole) of cyclooctene was injected. Refluxing was continued for three hours during which time the reaction mixture turned a deep mahogany brown. It was then poured into a beaker of ice containing 5 gm of 48% HBr and extracted with ether. (The expected product would, like the starting phosphonium salt, be insoluble in both ether and water; but no third, solid phase separated at this point.) Evaporation of the aqueous layer left a solid residue consisting solely of sodium bromide. The ether layer was evaporated to a wet oil; and this was taken up in methylene chloride, dried, and concentrated. Gas chromatographic analysis of the concentrate showed its only high—boiling, volatile component to be cyclooctene. Thin-layer chromatographic analysis showed the presence of triphenylphosphine and triphenylphosphine oxide. BIBLIOGRAPHY 6. 7. 8. 9. 10. ll. 12. 13. 14. BIBLIOGRAPHY S. Masamune, H. Cuts, and M.G. Hogben, Tet. Letters, 1017 (1966). E. LeGoff and W. Deadman, Michigan State University, personal communication. D.J. Wehe, W.R. Busing, and H.A. Levy, Program OR ABS, 0RNL—TM-229, Oak Ridge National Laboratory, Oak Ridge, Tennessee (1962). G.H. Stout and L.H. Jensen, "X-ray Structure Deter— mination, a Practical Guide," Macmillan, New York, 1968, p. 521. J. Karle and I.L. Karle, Acta Crystallog.,'gl, 849 (1966) . "' Stout and Jensen, p. 548f. D. Sayre, Acta Crystallog.,_2, 60 (1952). I.L. Karle and J. Karle, Acta Crystallog., 17, 1556 (1964). '— W.R. Busing, K. Martin, and H.A. Levy, Program ORFLS, ORNL-TM-505, Oak Ridge National Laboratory, Oak Ridge, Tennessee (1962). W.C. Hamilton, Acta Crystallog., 18, 502 (1965). Stout and Jensen, p. 395. E.B. Fleischer, J. Amer. Chem. Soc., 89, 5889 (1964). W.G. Dauben, C.H. Schallhorn, and D.L. Wahlen, J. Amer. Chem. Soc., 95, 1446 (1971). H. Gunthard, R.C. Lord, and T.K. McCubbin, Jr., J. Chem. Phys., 25, 768 (1956). 65 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 50. 31. 52. 66 A. Almenningen, 0. Bastiansen, and P.N. Skancke, Acta Chem. Scand., 15, 711 (1961). Stout and Jensen, p. 425. R.B. Woodward and R. Hoffmann, "The Conservation of Orbital Symmetry," Academic Press, New York, 1970, p. 86. H.M.R. Hoffmann, D.R. Joy, and A.K. Suter, J. Chem. Soc., B57 (1968). R. Huisgen, Proc. Chem. Soc., 557 (1961). T. Kauffmann, H. Berg, and E. KOppelmann, Angew. Chem. Internat. Edit., 9, 580 (1970). Woodward and Hoffmann, p. 87. von H. Lehmkuhl and D. Reinehr, J. Organometallic Chem., 2_5_, C47 (1970). R. Waack and M.A. Doran, Polymer,_a, 565 (1961). H.A. Weidlich, Chem. Ber., 21, 1601 (1958). S. Sugasawa and S. Ushioda, Tetrahedron,_2, 48 (1959). E.W..Garbisch and D.B. Patterson, J. Amer. Chem. Soc., eg, 5228 (1963). A.A. Bothner-By, Carnegie-Mellon University, personal communication. J.I. Musher, J. Chem. Phys., 54,594 (1961). J.J. Eisch and A.K. Jacobs, J. Org. Chem., 28, 2145 (1963)- '_' D. Seyferth and M.A. Weiner, J. Org. Chem., 24, 1595 (1959). "' E.L. Totton, R.C. Freeman, H. Powell, and T.L. Yarboro, J. Org. Chem., 26, 545 (1961). D. Rosenthal, P. Grabowich, E. Subo, and J. Fried, J. Amer. Chem. Soc., Q2. 3971 (1965). 33. 54. 55. 56c 37. 58. 67 W.T. House and M. Orchin, J. Amer. Chem. Soc., 83, 639 (1960). "" W.T. Olson, H.F. Hipsher, C.M. Buess, I.A. Goodman, I. Hart, J.H. Lamneck, and L.C. Gibbons, J. Amer. Chem. Soc., 62, 2451 (1947). H.L. Georing and R.R. Jacobson, J; Amer. Chem. Soc., §9. 3277 (1958). L. Claisen and E. Tietze, Chem. Ber., 28, 275 (1925). K. Sisido and K. Utimoto, Tet. Letters, 5267 (1966). K. Friedlich and H.G. Henning, Chem. Ber., 22, 2756 (1959). "' .V‘v ' \Ai ax" mm: 11111111?!“ 1293 "TIWITlfliflum'u Will 3085