Structure of the Dimer of tetraphenylcyclobutadiene - Journal of the

Justin R. Griffiths , G. Paul Savage , John Tsanaktsidis , and Ronny Priefer ... Hong, Guoping Cai, Peter Djurovich, William P. Weber, and Mark E...
0 downloads 0 Views 279KB Size
Nov.

;J.

COMMUNICATIONS TO THE EDITOR

l!W

drying. This final solution was used for the gas chromatographic measurement (flame detector) and, as indicated above, accounted for a t least 98y0 of the basic products. The materials identified were : phenylpyridines. 0 . 2 1 i g., 0,0014 mole; benzene, 0.343 g . q O,(lO44 mole; diphenylurea, 0.106 g., 0.0025 mole; and aniline, 0.093 g., 0.001 mole. These accounted for t i i c G of the phenyl radicals. X residue, 0.266 g., was also formed which would account for the rest. The results show a definite enhancement of substitution in the 4-position of all the complexes but one, and in the %-positionin most cases. It is not evident whether the amount of substitution a t the 3-position is decreased or remains the same. The substituent effects can be explained by considering a back donation of electrons to the pyridine ring by the metal ion. This d-11 n-electroii conjugation would involve a filled d,, orbital of the metal and its interaction with the 7electron system in the pyridine ring.6 The back donation to the ligand would be especially effective toward the 2- and 4-positions and is similar to that shown to occur in other coordination c o m p o ~ n d s . ~ Acknowledgment.--The authors wish to thank the National Science Foundation for its support through Grant G-l(lOO4 and a National Science Foundation Cooperative Fellowship to A. XV.G . for the summer of 1O M .

4725 n W

I-... 4

! '

I

0

Fig 1.--The structure a n d position in the unit cell of one molecule of octaphenylcyclooctatetraene. T h e origin is a t the lower left corner, the 5 axis a t the molecular center is a t O ' , / ~ I / ~ , t h e -Iaxis 1 is a t s = y = 0, a n d t h e centers of symmetry are a n d 0':?0. Bond distances are shown a s circles a t 000, '/K1/a3!,, 1-2 = 1.51, 1-2' 1.35, 2-2' = 2.18, 1-1' = 2.46, 1-3 = 1.50, a n d 2-9 = 1.50 A . The twelve independent C-C distancesoin 0 01 A the benzene rings have a n average distance of 1.39

+

either in the 110 type of plane of the isolated cube or perpendicular to this type of plane. This model, when placed a t the four equivalent sites of symmetry in the crystal, has only the one degree of freedom con6 H C . A . K i n g , k;. K o r o s , a n d S . >Nelson, I. .Valure, 196,,572 (1962). i i i J. C h a t t a n d I-, A . D u n c a n s u n , J . C h r m . SOL., 2939 ( 1 9 , 3 ) , see also sisting of arbitrary orientation about the i axis 13 1' C r a i g a n d G I l o g g e t t . i b r d . 1189 (1963). X very good approximation to the molecular orienta1 8 ) I B L I R e s e a r c h L a b o r a t o r y . San Jose. Calif.. t o w h o m inquiries should h e 5ent tion finally found was obtained by comparison of values R O YJ . G R I T T E R ~ of I< = ZliFo, - , F CZ L>EPARTMEST O F CHEMISTRY ~ Fo , computed for various ARTHURW'. GODFREY orientations from about 1 Oy0 of the observed diffracI-SIYERSITY O F COSSECTICUT S T O R R SCOSSECTICUT , tion maxima. The minimum value of li yielded a RECEIVED X U G ~ S 5, T 1961 structure which later comparison showed was very similar to the (PhAlN P h ) 4 structure, kindly communicated to us by XV. s. LIcDonald. However, the C Structure of the Dimer of Tetraphenylcyclobutadiene atoms of the previously assumed C, cube moved considerably, during the course of least-squares refinement, Sir: to the final positions shown in Fig. 1. Clearly, A three-dimensional X-ray diffraction study of Cs- this refinement has yielded the octaphenylcyclooctatetP h u , " ?where Ph is C6H5,has yielded the octaphenylraene structure. The present value of /i is (1.12 for cyclooctatetraene structure. These results exclude the the 1627 observed reflections, 1.4 5 F l r k i 5 3336, and octaphenylcubane structure, as well as the tetrabithe thermal motion is highly anisotropic. Hence we phenylcyclooctane formula4 which had been rendered considered the possibility that octaphenylcubane might improbable earlier on spectroscopic e ~ i d e n c e . ~ be transformed within the single crystul to octaphenylIi'e have confirmed the previously reported3 crystal cyclooctatetraene upon exposure to X-rays. Theresynimetry of I& a , the four molecules of CsPhqper unit fore, a sample previously unexposed to X-rays was cell, and the unit cell diniensions of a = 19.49 and c examined first by infrared and then by Kaman tech= 1 0 . ( 5 -4. The compound (PhXl?JPh)4,in which the niques, exposed to X-rays for a day. and then re-exA1 and X atoms are arranged a t the corners of a c ~ b e , ~ , ~ amined by infrared and Raman methods. N o evidence has the same space group and four molecules per unit of transformation of the molecule by X-ray irradiation cell; furthermore, its lattice dimensions of a = 20.0 could be detected. Hence, we conclude that the origiand r = 10.9 are increased over those for CbPhR nal sample as prepared is indeed octaphenylcycloocby an amount which seemed consistent with an initial tatetraene. The intermolecular contacts are, however, assumpt.ion of a slightly smaller cube for the Cn unit. primarily determined by the phenyl groups, which are Consequently, we chose as a trial molecular structure a oriented within the molecule in such a way t h a t an a-H cube of eight C atoms to each of which was attached a atom of one phenyl group points toward the center of phenyl group having the C-Ph bond along a body the closest neighboring benzene ring. This suggestion diagonal of the cube, and having the C6H6plane oriented of a "dimple" toward the center of a benzene ring was , 1) H H F r e e d m a n , J A n i C h r m Coc , 83, 214.5 (1961). first noted in the intermolecular packing of the biphen2 ) fl H B r a y e . \T' H u h e l . a n d I C a p l i e r . tbid , 83, 1106 (1961) ylene crystal.* ~:i, Ii H Freeilnian and 1) R P e t r r s e n i b r d , 84, 2837 (1962). 1, R C C0uk50n a n d 1) IT' Jones, Z'roc ChPm. Soc , 11.5 (1963) The internal angles of the tub-shaped cyclooctatetH H F r e e d m a n a n d R S G o h l k e , tbiii , 249 (1963) raene (COT) ring are 1%0°, in agreement with pre-


I c l ) o n a l d . , b i d . . 366 (1962) K R X l c l ~ ~ ~ ann ~d U' l d S h1cl)onald. ibrd , 3 8 2 (1R6,'O

( 8 ) J \Vaqer a n d C S

1.11,

J A m ( - ' h i m ';or

6 6 , XO:{.i ( l ! r l l )

4'726

COJIRIUKICATIONS TO T E ~ E EDITOR

sumed sp'-hybridization within nearly planar ethylenelike residues. The angle between the rr-system about atom C1 and its adjoining ring is 50.:2', and that between Cn and its adjoining ring is 46.2'. These angles would permit soin: conjugation, but the double bond distance of 1.33 A. and the single bond distance of 1.51 within the COT ring as well as the normal sp' to sp' bond length of 1.50 8.for both CICa and CsCy give no indication of abno5nial bond shortenings. Indeed, the I . X % and 1.462-A. distances of the electron diffraction studyg and the 1 . 3 - and 1.4Ci-x. distances of the X-ray diffraction study"' of COT itself suggest that perhaps intramolecular strain is more important than conjugation in C,Pha. Certainly thc. ring has not assumed one of the highly distorted fpr:i;s found" in the iron tricarbonyl complexes. Acknowledgment.---\lTe wish to thank the Office of Naval Research, the Kational Institutes of Health, and the -1dvanced Research Projects Agency for support, \Ye are also indebted to P. D. Hartlett for valuable discussions, to D. R . Petersen for the study of possible transformation by X-rays, and to T. R . R . LlcDonald, 11.. S. hlcDonald, and J . I . Jones for cornmunicating to us their results on the (Ph-llSPh!, structure.

Vol. S(j

x.

OH

I

data. we used gas--liquid chromatography combined with continuous combustion and flow counting, carrying out multiple analysis on single mature plants or a few grams of seedlings. The column temperature was programmed from lSO-%GO', and the total effluent was passed through a triode- argon detector followed by combustion and counting with a flow-through proportional counter. A4dual pen recorder simultaneously plotted the output of a counting ratemeter and the ion current from the triode-argon mass detector, giving mass and radioactivity in differential-type curves. The useful inass range was 0.1 to %(I p g , , and activities of about 100 c.p.m. were detectable in any one peak. A mixture of codeine, thebaine, laudanosine, papaverine. cryptopine, and narcotine was completely re(9) 0.B a s t i e n s e n . I. H e d h e r g , a n d K H e d h e r g , f C h ~ mP h y s , 27, K(11 solved ( a 16 in. X 6 ft. column, nonpolar silicone gum (1957). on silanized Celite), as were the four phenolic alkaloids (10) J. B r e g m a n . p r i v a t e c o m m u n i c a t i o n , 14.58 morphine, laudanosoline, r e t i d i n e , ; and laudanine. (11) B Ilickens a n d \V S 1,ipscomb. f Chcm P h y s , 3 7 , 208-1 i l R 6 2 ) G. S. PATVLEY as their trimethylsilyl ethers6 (cyano-silicone polymer DEPARTMENT OF CHEMISTRY HARVARD ~SIVERSITY IT'. S.LIPSCOMB as liquid phase). As confirmatory analytical adjuncts CAMBRIDGE, MASSACHCSETTS 02138 two-dimensional Chromatography on alumina irnpregnated paper, paper chromatography using two EASTERK RESEARCH LABORATORY H.H.FREEDMAN different systems,' and thin layer chromatography8 THEDOlV CHEMICAL COMPANY were employed. I7R.4MIXGH,4M, hl.4SSACHUSETTS Alfter the natural occurrence of reticulineg in very RECEIVEDSEPTEMBER 21, 1964 low concentrations in 120-day poppies and %day seedlings was shown by g.1.c. (Table I ) , several hundred 5-day seedlings were exposed to I4CO2for 2.5 h r . , and Reticuline a s the Benzyltetrahydroisoquinoline the alkaloids were extracted immediately (after addiPrecursor of Thebaine in Biosynthesis with tiop of 1 mg. of reticuline carrier). Gas, paper, and Carbon-14 Dioxide' thin layer chromatography showed reticuline to Sir: account for more than 80% of the activity in the Earlier work2 established the rapid d e novo synthesis phenolic alkaloids, while thebaine accounted for more of thebaine (I) from 14C02and its primacy in the hythan SOTc of the nonphenolic alkaloid activity. Similar drophenanthrene alkaloid series. Recent evidence. results were obtained using a 120-day plant (1 day prior based on feeding experiments, showed that reticuline to blossom). Comparison of the specific activities of (11) can serve as a precursor of thebaine.3 \Ve now reticuline and thebaine shows reticuline to have a much present evidence, derived from short exposures to higher specific activity than thebaine ; the reticuline : I4CO2 of both seedlings and budding plants of P. (,4! 1: C a r a c e . .%. Gual-ine, a n d I n a m - u l - H a q A n n [ h i m : I < o m e l , SO, O I J somnijbriun L . , which confirms these findings and 'l!iIjO). K \V G o p i n a t h , T R C h r i n d a c h a r i , a n d N. V i s w a n a t h a n , Be? 92, establishes beyond question the role of reticuline as the l i i d i , l