COMMUNICATIONS TO THE EDITOR
March 5 , 1964
927
The Triclinic Crystal Form of a ,P,r ,G-Tetraphenylporphine
Sir : A recent X-ray structure determination of tetraphenylporphine2 does not distinguish between pyrrole rings with and without hydrogen attached to nitrogen, since the tetragonal space group requires 4 molecular symmetry. Since fourfold symmetry is higher than the molecule permits, it may either be inexact or the result of molecular disorder within the crystal. The X-ray analysis of a triclinic t e t r a p h e n y l p ~ r p h i n crystal, e~ space group PI, avoids the above difficulties because of lower, 1, molecular symmetry and thus distinguishes two types of pyrrole rings and the two central hydrogen atoms. Results of this analysis are reported here. Deviations from a least-squares plane determined by the 24 atoms of the porphine nucleus are given in Fig. 1.
I! 1 I
H
Fig. 1.-Electron
spin resonance spectrum of polycrystalline [(n-C1Hs)aNlz[Rh(MST)zI.
Pd(MNT)2-, and P t ( M K T ) 2 - complexes, which have = 1/2.7f8 From the e.s.r. and static susceptibility results, we conclude t h a t the square-planar Rh(MNT)Z-' complex has a spin-doublet ground state. T o our knowledge, this is the first well-characterized mononuclear, paramagnetic rhodium(I1) complex.l5 T h e ground state of R h ( M N T ) 2 - 2 is . . .(4bZg)2 (4ag)l = 2Ag,using the molecular orbitals derived2 for N I ( M N T ) ~ - ~ .The low energy ligand-to-4ag chargetransfer band is located a t 15,800 cm.-' ( E s 4000) in Rh(MNT)2-2. All the planar complexes mentioned above with a 2A, ground state show this type of low energy charge-transfer band.
S
( 7 ) I n frozen DMF-CHC13 solutions, t h e following g values have been obtainedg: PU'i(.MMNT)z-, gl = 1.996, 9% = 2 043, ga = 2.140; Pd(?*ISTiS-, Fig. 1.-Atomic nomenclature gl = 1 9 5 6 , E? = 2.046, g3 = 2.065, P t ( M S T ) z - , gi = 1.825, g? = 2 067, nuclear least-squares plane of Solid [ ( ~ - C I H ~ ] ~ N I [ S ~ ( T D T ) ~ ] ga = 2 221; these complexes have S = I / ? . shows only t w o lines, with 8 1 = 9 04.5 and g , l = 2 193.10 However, t h e in-plane anisotropy is resolved in solid [(C~Ha)As(CH~l1[Ni(TDT)zJ, with gl = 2 020. gz = 2.075, g3 = 2.130 In a frozen DMF~-CHC13solution N i ( T D T ) z - gives gl = 2.011, 82 = 2.043, g3 = 2.181 1% T h e anion N i ( T D T ) z - was t h e first reported example of a nickel complex with a square-planar structure and a spin-doublet ground state.10 (8) I t is interesting t o note t h a t C o ( M N T ) ? - Z . for which both spin-quartet'3 and spin-doubletl4 ground states have been proposed, has a considerably different g3 from t h a t of [(n-CrH,)rN]2fRh(MNT)z],which is a spin-doublet, ~ l [(n-CIHs).iSlzfor a single crystal of 5 % [ ( n - C ~ H s ) a N l z [ C o ( M N T l in [ N i ( M N T ) ? 1 ,t h e values gi = 1 9 7 7 , g~ = 2 025, and g i = 2.798 have been reported ' 4 (9) A . Davison, N Edelstein, R. H . Holm, and A H Maki, J . A m Chem S o c , 86, 2029 (1963). (10) H B. Gray a n d E. Billig, i b i d . , 86, 2019 (1963). (11) I Bernal. unpublished results (12) We thank Dr N Edelstein for measuring this spectrum for us (13) H B. G r a y , K Williams, I . Bernal, and E Billig, J . A m Chem. Soc , 84, 3596 (1962). (14) A . Davison, N. Edelstein. R . H . Holm, and A . H Maki. t b i d . . 86, 3049 (1963). (1.51 There are very few paramagnetic rhodium compounds of a n y structural type Paramagnetic Rh(IV1 fluorides have been reported by R. S Xyholm a n d A . G . Sharpe, J . Chem. Soc., 3579 (1952). (16) ( a ) N I H Postdoctoral Fellow, 1963-1964, (h) N S F Graduate Fellow, 1963-1964.
DEPARTMEST O F CHEMISTRY COLUMBIA USIVERSITT YEW YORK,SEW YORK10027 RECEIVED JANUARY
and deviations in A. from t h e triclinic tetraphenylporphine
Of note are (1) the tilt (6.6') of the N-1 pyrrole ring, which makes the close central hydrogen-hydrogen contact -0.2 A. longer than a planar configuration allows; ( 2 ) the tilt (9.1') of the C-C~-(C~H~),-! line with respect to the plane, for which a close (3.40 A.) intermolecular contact may account; and (3) the rotation of the phenylgroups ((C6H6), 61.0', ( C ~ H ~ ) . B 63.1'). The phenyl and pyrrole rings are planar within the error of their determination. Figure 2 shows the intramolecular bond distances and angles. The standard deviations in bond lengths and angles involving only c a r b p and nitrogen atoms are of respectively. the order of 0.005-0.010 A. and 0.5-0.8°j Those for bonds to hydrogen atoms are -0.05 A. The accuracy is best for the porphine nucleus and poorest for (C6H5)~. Of significance are (1) single bonds isolating the phenyl rings from the nucleus; (2) approximate double bonds C-142-2 and C-3-C-4, which apparently remove these atoms from the interior resonant system ; and (3) flat bonding about the interior carbon atoms.
-
-
(1) This research has been supported by t h e S a t i o n a l Institutes of H e a l t h ,
E. BILLIG U S Public Health Service, and Lederle 1,ahoratories Division. American S.I. S H U P A C K ' ~ ~Cyanamid Company. JAMES H. WATERS (2) J I, Hoard, M J H a m o r , and T . 4 H a m o r . J A m C h r m SOL, 86, RAYMOND WILLIAMS^^^ 2334 (1963). HARRYB. GRAY (3) A highly purified sample was kindly provided by Dr. Henry E Rosenberg. 2, 1964
COMMUNICATIONS TO THE EDITOR
928
I
/
I I*
loo
Vol. 86
I
1
,,I
0
-
THF Ethyl Ether
A Hexane o Dioxane
9 -L
w
F 60
s
Y
Lu
20 Fig. 2.-Bond angles and distances of triclinic tetraphenylporphine; bonds t o hydrogen atoms are broken
The molecular configuration of triclinic tetraphenylporphine is markedly different from that of the tetragonal form, showing that the porphine system is flexible. Their bond lengths and angles, however, are consistent if the one crystallographically independent pyrrole ring in the tetragonal case is compared t o an average of the two independent ones in the triclinic case. Such averaging, however, ignores significant differences between the two types of pyrrole ring. Preliminary results? from this laboratory on silver tetraphenylporphine indicate the possibility of obtaining triclinic crystals of metalloporphinesj and thereby observing intramolecular differences among their pyrrole rings also. The crystallographic details of this structure determination will appear elsewhere.
0
1
C I A
0 1
I
OA
I
Fig 1 -The influence of solvent and temperature upon the formation of stilbene from n-butyllithium and benzyl chloride
reaction provides the basis for a widely used analytical method for the quantitative determination of butyllithium, and inasmuch as the presence of stilbene could not be readily accommodated in the proposed mechanism above, we decided to explore how the reaction conditions, such as solvent and temperature, influenced the course of this r e a ~ t i o n . ~Figure 1
( 4 ) K Shapiro and A . Tulinsky, in preparation (.ij l h e structures of tetragonal forms of nickel etioporphyrin-I1 ( M B
Crute, Acta C ' r y ~ t . ,l a , 24 (lQg59)j,nickel etioporphyrin-I ( E . B Fleischer, J . A m Chrm S o c , 85, 146 (1868)\, and copper tetraphenylporphine i R . B Fleischer. ibrii , 8 6 , 13.53 ( 1 9 6 3 ) ) , which is isomorphous t o tetragonal tetraphenylporphine. have been reported
I11 STERLISGCHEMISTRY LABORATORY STUART SILVERS YALECSIVERSITY A . TULINSKY shows how the yield of trans-stilbene was affected by XEXVHAVES,CONKECTICUT these reaction conditions. The unique influence of RECEIVED J A N U A R4, Y 1964 The Influence of Tetrahydrofuran on the Reaction of n-Butyllithium with Benzyl Chloride : Halogen-Metal Interconversion us. a-Hydrogen-Metal Interconversion Sir: We wish to report an extraordinary influence of tetrahydrofuran upon the mechanism of the reaction of benzyl chloride with n-butyllithiuni. Earlier studies had indicated that this reaction proceeded via halogenmetal interconversion (I), leading to the formation of benzyllithium ( I ) as a transient intermediate which was rapidly consumed in subsequent coupling reactions to form the observed products, bibenzyl (11) and naniylbenzene ( 1 1 1 ) , ' , 2 \Vhen we carried out this reaction a t room temperature in tetrahydrofuran, however, there n-as obtained, in addition to I1 and 111, approximately a 207; yield of trans-stilbene. As this ( I ) H Gilman and A H Hauhein. J A m C h r m . Sor 66, 1,515 (1944). I ? ) I< G Jones and H Chlman. Din Rrociions. 6, 339 (1951)
tetrahydrofuran as a solvent for this reaction, particularly a t low temperatures, is evident from this f i g ~ r e . Additionally, ~ a small amount of a yellow oil was isolated in the chromatographic separation of the products from the tetrahydrofuran reactions (eluted
(3) T h e procedure adopted was a s follows: At the prescribed temperature, 30 mi of a solution of 1%-butyllithiumin heptane (1 mmole ml ! was added slowly (5 min.! t o a solution of 3.8 R (30 mmolesi of benzyl chloride in 100 ml of the appropriate d v e n t under argon. After 2 hr , the reaction was poured into an equal volume o f a 50-.50 water-petroleum ether ( b p 3 7 - 5 1 O mixture T h e organic phase was separated. dried over Na,SOa. and vacuum stripped A representative portion of the residue was placed on an alumina column (24 in X 1 in ! for chromatographic separation o f the products In a typical separation. the n-amylbenzene wa4 the first product eluted with petroleum ether (infrared spectrum identical with authentic sample), followed closely by bibenzyl ( m p. 52'. lit 5 2 . 5 ' . no depre-sion in mixture melting point, identical infrared spectrum as authentic sample) lians-Stilbene ( m p. 1 2 4 - 1 2 i 0 , lit. 124". identical infrared and ultraviolet spectra with authentic sample) A n a l Calcd for C I I H ~ ~C ., 93 :3, H , A 7 F o u n d : C. 93 3 . H . 6 9 , was eluted with a 2 5 voI To solution of benzene in petroleum ether (4) Even very small amounts of tetrahydrofuran exert a discernihle effect T h u s , little or no stilbene was detected when the reaction was run in In contrast. when a 1- 1 molar ratio ot tetrahexane a t rimm temperature h y d r o f u r a n ~ - h u t y l l i t h i u m(30 mmoles of each) in hexane (100 ml ) was used approximately a 107' yield of 1 1 ans-atilbene was isolated after 2 hr reaction
