COMMUNICATIONS TO THE EDITOR
Sept. 5 , 1963
2673
TABLE I Ultraviolet spectrad---
7 -
M.P., Compounda
I
c.*
246
I1
220-222
Nucleoside from $BPZe Adenosine*
218-222
9-a-DKibofuranosyladenineh I11 IV
233
(g./lOO ml.
HaO)
-84.9 f 0 . 2 (0.351) 0 (0.397) 0' (0.262) -60.4 (0,7) 24 (0.65)
+
20 1 275-277 216
RrC
7
0 . 1 NaOH
0 . 1 HC1
[U]"D
Xmsx.
ax
Xmax,
cx
ma
10-8
mir
10 - 0
0.19
272
13.6
270
9.8
.15
273
13.3
271
9.9
273
13.6
271
9.8
.20
260
14.2
260
4.3
.17
257
.75 .89
275 300
17.5 28.2
272 231h 333
2.2 7.0 17 9
150
v'
278 11 5 31 278 14 3 28 1 VI1 a Satisfactory elemental analyses (C, H, N ) were obtained for each compound except VII, which was not analyzed * Melting c Whatpoints below 260" were determined on a Kofler Heizbank; those above 260' were determined in a capillary and are uncorrected man No. 1 paper, water-saturated butanol. Determined with a Cary Model 14 spectrophotometer. e Data from ref 3 TemElemental analyses show that V is a dipurinylmercury. perature not specified. Calcd. from Radenoslne. * Data from ref. 14.
'
of crude 3-benzyl-7-P-~-ribofuranosyladenine (VII) was obtained as a brown glass. The benzyl group of this nucleoside VI1 was removed by hydrogenolysis using 5% palladium-on-charcoal catalyst in a mixture of ethanol and water a t 80' and 47 p.s.i. of hydrogen. The resulting nucleoside, obtained in 31% yield, was identified as 7-@-~-ribofuranosyladenine(I) by its ultraviolet spectrum (showing sugar attachment a t N-7) and application of the trans ruleg (indicating the configuration of the glycosyl linkage formed by this coupling must be @lo)). Coupling of the same dipurinyl mercury V with 5-0benzoyl-D-ribofuranosyl bromide 2,3-cyclic carbonate13 followed by deblocking of the resultant nucleosides with methanolic sodium methoxide gave a mixture of 3benzyl-7-B-D-ribofuranosyladenine(VII) and 3-benzyl7-a-~-ribofuranosyladenine(VIII) which could not be separated but could be debenzylated catalytically as described earlier for the p-anomer. Partition chromatography of the resulting mixture, using a cellulose powder column and butanolLwater (86:14), gave first 7-@-D-ribofuranosyladenine( I ) , which crystallized on seeding with a crystal of material obtained from the first coupling described earlier, and then i'-a-~-ribofuranosyladenine ( I I ) , purified through its picrate. Properties of the anomeric nucleosides and some of their precursors are given in Table I , along with those of the nucleoside from pseudovitamin B12, adenosine, and its a-anomer. The samples of the &anomer I prepared by the two different coupling reactions were identical, whereas the properties of the a-anomer I1 are in good agreement with those reported for the nucleoside from pseudovitamin Biz. The optical rotations of the anomers (@, -84.9 0.2'; a , 0") confirm the assignment of a- and @-configurationsand agree well with the relative values for adenosine and its a-anomer (-60.4' and f24').
*
(9) B. R . Baker, Ciba Found S y m o . , Chem. Bioi. Purines, 120 (1957). ( I O ) S o true exceptions t o the trans rule have been reported in the synthesis of purine nucleosides 11 I n rare instances a minor amount of a-anomer has been detected'? b u t the @-anomeris always the principal product of the reaction I n our reaction only one anomer could be detected. (11) J . A. Montgomery and H J . Thomas, Advan. Carbohydrate Chem., 17, 301 ( 1 9 0 2 ) . (12) B R . Baker, R E . Schauh, and H. M . Kissman, J. A m . Chem. Soc., 7 7 , 3911 (195.51; H 51 Kissmanand B R Baker, ibid.,79, ,5534 ( 1 9 5 7 ) . 113) This ribose derivative. which contains n o acyloxy group a t C-2 t o cause stereospecific entry n i a purine a t C-19 giving rise t o a @-ribonucleoside only, has been u i e d t o prepare a mixture of adenosine and its a-anomer ' 4 (14) R S Wright. G M . Tener, and H. G Khorana, J . A m . Chem Soc., 80, 2004 (1958).
I t is now possible to state with certainty t h a t the nucleoside moiety of pseudovitamin BIZis in fact 7 - c ~ ~ ribofuranosyladenine. Synthesis of the nucleosides from other BIZvitamins is in progress. KETTERISG-MEYER LABORATORY
SOUTHERN RESEARCH INSTITUTE
A. MOSTGOMERY H. JEANETTE THOMAS
JOHN
BIRMINGHAM, ALABAMA RECEIVED JULY22, 1963
A Two-Stage, Two-Center Decarboxylation' Sir: The elusive question of simultaneity in multicentertype reactions continues to receive its most critical examination in studies of Diels-Alder reactions.2 Suggestions that a range of transition state structures might be accessibleZb,'are supported by results which clearly demonstrate the polar ambivalence of such states3 and warn that classical elucidative techniques might well distort the very phenomena they seek to m e a s ~ r e . The ~ dilemma can be avoided by isotopic substitution, preferably a t the reaction sites but a t adjacent atoms. Interpretations of these latter, a-hydrogen isotope effects have been rendered ambiguous by the discovery6a of one clear exception to what had been regarded6bas a universal, though necessarily empirical pattern. Less equivocal decisions should follow from determination of primary isotope effects a t both reaction sites and we here report the first such investigation in this area.' Thermal decarboxylation of the a-pyrone-maleic anhydride adduct8 was chosen for study assuming (a) t h a t this might legitimately be regarded as a DielsAlder retrogression and (b) that the structural sim( I ) T h e study was supported by the Air Force Office of Scientific Research, under Contract S o .49(038)-942 and G r a n t No. 142-63 (2) (a) R B Woodward and T . J . Katz, Tetrahedron, 6, 70 (1959), (h) J. A. Berson and A. Remanick, J A m . Chem. SOL.,83, 4947 (1961); (c) C . Walling and H . J Schugar, ibid , 86, 607 (1963); (d) S Seltzer, ibid., 86, 1360 (1963) (3) J . Sauer and H. Wiest, Angew Chem , 74, 353 (1962) (4) This has been clearly recognized in the study of rate dependence on pressure ( C Walling and D . D. Tanner, J A m . Chem. Soc., 86, 612 (1963)). (5) (a) D . E . Van Sickle, Tetrahedron Letters, 687 (19613; ( b ) S. Seltzer, ibid., 457 (1962) (6) (a) R . E. Weston, J r , and S . Seltzer, J . Phyr. Chem , 68, 2192 ( l 9 6 2 ) , (b) S Seltzer, J . A m . Chem Soc , 83,202.5 (1961). ibid., 86, 14 (1963). (7) (a) G A . R o p p , V F Raen, and A. J . Weinherger, ibid , 7 6 , 3697 (19.53); ( b ) K . F. W. Bader and A N.Bourns, Can J Chem , 3 9 , 348 (1961), are the most closely related studies in object (a) and approach ( b ) (8) (a) 0 Diels and K Alder. A n n Chem., 490, 2.57 (1931); (h) B R Landau, P h D . Thesis, Harvard University. 1960
COMMUNICATIONS TO THE EDITOR
2674
plicity of the dienophilic fragment might simplify interpretation of the experimental results. Kinetic analysis, C14 tracer studies, product isolation, and structure proof are in accord with the first assumption but will be described more fully a t another time. C13 and 01* isotope effects were both determined a t levels of natural abundance by consecutive mass spectral analyses of purified C 0 2 obtained (a) after 5-10% reaction, (b) after ten half-lives, and (c) from a commercial cylinder.g The results in, Table I , column 2, have been corrected for instrumental "mass discrimination" and 017contribution t o the m / e = 45 peak. Other corrections were evaluated but then ignored when they proved to be less than the experimental uncertainty. These include : instrumental background, incomplete mass resolution, isotopic fractionation and/or exchange during C 0 2 purification, 0'' fractionation during decarboxylation, and C13 content in the C8H603 fragment. TABLE I --Predicted--IIIA
Observed
1.032 f 0.001 1.015 + 0.002
kiz/kn kie/kiR
vf/v*c P*/V*O "*/V*O'
0
1.0298 1,0155 1.0073 1.0023 1.0290
80 0
+
h
c
IIIB
1.0293 1,0222 1.0068 1 . 0000 1.0455
't
Vol. 85
is conveniently taken to be a common decomposition mode, the separation of rigid OCO' and C8H603along an axis joining their centers of mass. Mass dependence of the corresponding frequency ( v * ) is thereby determinedI2 and we may use the T-method1lband both our observed isotope effects to estimate the relative degrees of bond stretching a t the two sites.13
+
'/z(Aa*o AU*O') Aa *C
(k - A)(;(& - A)(;*=
kid&
-
')
kiz/kii
- 1)
Although it might have been anticipated that the ratio (of mean changes in force constants a t oxygen to those a t carbon) would be somewhat less than one,I4 the calculated value is -0.10 =k 0.09. Whether truly negative or zero, the result is sufficiently irreconcilable with the decomposition mode used t o generate it as to preclude further realistic discussion of any such transition state. Left only with 111, defined by 4a*0 = 4a*o' = 0, we need but choose an appropriate decomposition mode to predict k I 6 / k l 8and so consider two possibilities: A, in which rigid OCO' and C8Hs03 unfold about a stationary bridgehead pivot, during C-C cleavage, and B, in which 0 serves as pivot.15 Prediction of k12/k13 is inherently less reliable in this case since it requires the further ad hoc assignment of Aac* and net frequency change. The traditiona1lzbvalues of -4.50 mdynes/A. and 900 cm.-' were chosen both to avoid prejudice and to permit comparison with the evaluation of similar assumptions elsewhere. Table I clearly reveals IIIA to be the simplest description of the transition state consistent w i t h both isotope efects. Obvious extensions of this technique to related systems are being pursued. (11) ( a ) J . Bigeleisen a n d M. Wolfsberg, A d v a n . Chem. Phys., 1, 68 (1968); (b) i b i d . , 1, 18-26, 67 (1958). (12) hl. Wolfsberg, J . Chem. Phys., 93,21 (1960). (13) A a *c, A a *o, and Aa *o' are t h e activation changes in diagonal cartesian force constants, a t t h e carbonyl carbon, alkyl oxygen, and carbonyl oxygen, along any one of t h e three axes, and t h e sum is taken over all three axes. refer t o t h e molecule bearing the heavier isotope a t Subscripts following Y t h e position indicated. Zero time reaction and isotopic homogeneity of 0 and 0', are implicitly assumed and can he justified under our conditions (14) Thermochemical d a t a require t h a t t h e CHz-C bond in methyl acetate be weaker t h a n t h e CHa-0 b y some 2-14 kcal./mole. (1.5) These provide the P * ratios b y analogy with mechanical models. Cf, J . Bigeleisen and M. Wolfsberg, J . Chem. Phys., 21, 1972 (1953); i b i d . , 22, 1264 (1954); P Yankwich a n d R . M . I k e d a , J . A m . Chem. Soc., 81, 1532 (1959). (16) M . J Stern and M . Wolfsberg, J . Chem. P h y s . , in press (17) Union Carbide Corporation Research Fellow, 1962-1963.
*
or
0
0
I
0
0
II
\
m 1
o=c=o
These results are most simply analyzed in terms of the idealized transition states I , 11, and 111. Thus, the observed k12/k13 is sufficiently similar to those obtained in reactions where C-C cleavage is unambiguous'" as to exclude I from further consideration. 11, however, requires more quantitative evaluation and, since we would have i t encompass a wider range of possible structures, a more careful definition. This (9) A Consolidated Engineering Corporation 21-401 mass spectrometer was used in its isotope ratio recording mode. Individual sample analyses were reproducible t o ztO,02-0.05% for C a n d 10.03-0.08% f o r 0. Calculated isotope effects derive from five independent experiments for those of carbon a n d f r o m three for those of oxygen. T h e solvent was dimethyl phthalate a n d the temperature was 130.1'. (IO) Malonic acid, 138', 1037; mesitoic acid 92", 1.032; ttichloroacetate ion, 70". 1 034,11nB y contrast: xanthate pyrolysis, looo, l.000.7b
DEPARTMENT O F CHEMISTRY M. J. GOLDSTEIN JR." G. L. THAYER, CORNELL UNIVERSITY ITHACA, KEW YORK RECEIVED JUSE 1, 1963
A New Boron Hydride Ion B9HL4Sir: Decaborane is degraded in aqueous base to boric acid, hydrogen, and a new, stable boron hydride ion of the composition B9HI4-. The possibility that such an ion might exist has been discussed' and, in two instances, ultraviolet absorption maxima were observed2which we now suspect were due to its presence. Salts containing the anion have now been isolated and the anion has been characterized by analyses, by conversion to known compounds, and by determination of the stoichiometry of the preparative reaction. (1) F E. Wang, P G. Simpson, and W N . Lipscomb, J . Chem. Phys , 36, 133.5 (1961); W. h'.Lipscomb, Pvoc .Vatl. A c a d . Sci. c'. S , 47, 1791 (1961) (2) G. W . Schaeffer, J J . Burns, T. J. K!ingen, L A . Martincheck, and R . W. Rozett, Abstracts of Papers, 135th Kational Meeting of t h e American Chemical Society, April, 19,59. p . 4 4 M ; M .F. Hawthorne, A R . Pitochelli, R. D . S t r a h m , and J . J . Miller, J . A m Chem. Soc , 82, 1825 (1960).
