4434
COMMUNICATIONS TO THE EDITOR
with sodium yielded the cyclic acyloin (11). Anal. Calcd. for C X H ~ ~ O C,~80.56; : H,13.12: 0, 6.31. Found: C, 80.66; H , 13.00, 0, 6.44. Clemmensen reduction of I1 with deuterated hydrochloric acid produced the cyclic hydrocarbon (111),in 50y0y.ield from I, m.p. 67-68, O6 with ca. 5 atoms of deuterium per molecule. The transformation of I to I1 was then carried out with a 1: 1mixture of 111and xylene as solvent.
I
VOl. 52
topological isomers exist,ge.g., models indicate that acycloalkane of more than 50 carbon atoms can take the form of a simple or a knotted ring. In addition, this latter species can exist in two optical isomers. Further work with larger quantities of material to permit isolation of V is in progress. (9) I1 Frisch and E. Wasserman, Division of Polymer Chemi?try, 138th Meeting A C.S , New York, Sept., 1900.
BELL TELEPHONE LABORATORIES EDELL ~ A S S E R M A N MURRAY HILL,NEWJERSEY RECEIVED JULY 5 , 1960
MECHANISM O F BASE HYDROLYSIS AND I7 THE BASE CATALYZED REACTIONS OF SOME COBALT(II1) - DC1 2.H+ AMMINE COMPLEXES HOOC(CH&;COOH CaaHa3D6 .Si?.: IV I11 Two mechanisms have been proposed for the
2. HOAc
&
t
Unchanged I11 was separated cleanly from the other products by chromatography using pentane as eluent. The fraction containing 11, which was removed with ether-methanol, had infrared bands a t 2105, 2160 and 2200 cm.-' (C-D stretch) which were also observed in the spectrum of 111. Rechromatography of this material demonstrated that the deuterium was associated with the acyloin and was not due to contamination by unaltered 111. This acyloin fraction was oxidized with alkaline hydrogen peroxide giving the diacid (IV). The presence of 111 among the products was demonstrated by its chromatographic behavior, infrared spectrum, m.p. and mixed m.p. The yicld of 11, formed in the presence of 111, varied between 5 and 20%. The yield of I11 in a given oxidation was roughly proportional to that of I1 and constituted about 1% of the a ~ y l o i n . ~ It is proposed that the acyloin fraction contains V, a species comprised of two interlocking but otherwise non-bonded rings. Oxidation of V cleaves the acyloin containing ring and this is followed by unthreading to give 111. I t is unlikely that some other deuterated system would give rise to I11 under oxidative conditions. Chromatography of the material obtained by heating 111, xylene and sodium with IT, previously formed in the absence of 111, gave no deuterium in the acyloin fraction. Oxidation of that fraction produced no hydrocarbon.
reaction of hydroxide ion with complexes of cobalt (111) containing a replaceable ligand. One is the direct displacement, or S N mechanism' ~ and the other the conjugate base dissociation, or S N 1 CB mechanism.2 In water solvent the two mechanisms give rise to the same product, since the intermediate in the conjugate base mechanism would react with the solvent to yield a hydroxo product. However in a non-hydroxylic solvent different products are predicted for the two mechanisms in the event that an excess of some nucleophilic reagent other than hydroxide ion or water is present. We wish to report some results which are easily explained by the S N 1 CB mechanism and cannot be ~ The solvent explained by the S N mechanism. used is dry dimethylsulfoxide. As examples of a large number of similar experiments with various cobalt(II1) complexes and nucleophilic reagents, these reactions illustrate the chief points. Spectral data were used to follow the reactions and identify the reaction products.
+
tranS-Co(t11)2NOyCl+ NO%- ---+ 0.001 M 0.01-0.10M Co(en)2(N02)2+ C1frens-Co(en)2~OnCli OEI- + 0.001 AI 0,001 ilI or less Co(en)pPu'020H+ C1trens-Co(en)2r';02C1+ S O 2 OH- --+ 0,001 11I 0.10 AI 0.001 ill Co(en)2(N02)2+ C1OHCo(en)2(n'Ol)(OFI) SO2- --+ 0.001 nf 0,lOAI indefinite product, some Co(en)2(SO2)2+
+
+
f
We note that V is the first demonstrated example of a compound in which the topology of the system must be considered in describing its structure. For example, Vis a topological "isomer" of the non-interlocked system I1 plus HI.* Other possibilities for (6) L. Ruzicka, M . Hiirbin and AI. Furter, I l e l v . Chim. A c t a , 17, 78 (1934) found m.p. GGD-670 for non-deuterated 111. (7) Calculations by H. Frisch have given a probability of 10-1 for this statistical threading. This small value renders the hypothesis of extended chains of references 2 and 3 somewhat less likely. I t also reduces the probability t h a t more t h a n 1'7? of 111, from t h e oxidatim reaction, arose from higher interlocked systems. (8) T h u s two separate molecules are t o be viewed as isomeric with V, which resembles a single molecule i n many respects. I t should be noted t h a t topological isomers necessarily have identical chemical bonding and arrangements of atoms about rigid centers.
+
+
+
+
(1)
+
(2)
+
(3)
(4)
Reaction (1) shows first order kinetics and has a half-life of 5-6 hr., nearly independent of the nitritc ion concentration. Reaction (2) is very fast, being complete in less than 2 min. when the first reading was made. Reaction (3) is also complete in less than 2 min. yet the product is cleanly the dinitro complex. Reaction (-1) is very much slower than ( 2 ) or (3) and gives a spectrum lacking sharp bands which steadily changes over several days. The following points are also significant: Only catalytic quantities ( l O ~ o ) of hydroxide ion are required to cause a rapid reaction of the cobalt(TI1) (1) D. D. Brown, C. K . Ingold and I
