COMMUNICATIONS TO THE EDITOR
1062
TABLEI THEEMISSIONOR POSITRONS AND ELECTRONS AMF
AMExP~I."
+HI = +He* = H 3 +B9 = Be9 fC11 = Bll +NlS = e13
Bond
0.0021 -0.0002 0.0023 +e+ .0011 .0011 .0022 +e+ - .0003 - .0027 .0024 e+ - .0013 .0033 .0020 e+ .0015 .0037 .0022 = "6 +e+ - .0021 - .0043 .0022 +Fll = 0 1 1 + e + .0025 - .0047 .0022 Li8 = +Be8 e- ,0112 .0031 - .0143 B i 2 = +C12 + e - ,0117 ,0043 - .0160 N16 = +Ole e- .0070 .0052 - .0122 P o = +Ne20 e- .0064 .0059 - .0123 n1 = +HI e- .0010 ,0012 .0022 a Fowler, Delsasso and Lauritsen, Phys. Rev., 49, 561 (1936). The AM for the non-radioactive reactions are from Ref. 1. The masses of e + and e- have been included in AMP. +e+
-
++
-
-
+ + + +
-
TABLEI1 THREETYPES OR NUCLEAR REACTIONS
+ +
H8 HI = LiT H1 = BI1 + H1 = NU + H1 = Fl9 + HI = He* n1 = Cl1 + n1 =
+
AM~rptl.
AMP
Bond
0.0006
0.0219
016
-0.0213 .0183 .0173 - .0129
Ne20 He4
-
He4 Be*
-
e 1 2
-
+ HI "8 016 + H 1 = FIT c1% + n1 = C1* C12
011
H* Lie B1O N14
- .0021 - .0008
3
+
- .0054 - .0045
= 017
+ H3 = He4 + H2= Be' + HZ
+ H2
=
.0140
.0223 .0198
-
-
C12
-
= 016
.0255 .0236 .0275 .0220
.0016 .0025 .0036
-
-
2He' = 3He4 = 4He4 = 6He4 =
Be' C12 0" Ne'O
7He4 =
Si28
AMF
-0.000 +0.0024 .0068 - .OS1 .0121 - .I56
-
,021 .041
.0191
.0198 .0165
.0043
.om
.0002 .0004
.0221 .0194
.0031
.0052
.0040 .0003 .OOM .OOO7 .0015 .0026 ,0034
TABLE111 OF 4n NUCLEI FROM He4 THEFORMATION AMErpti.
.ow9
-BondCalcd.
0.00%
.0146 .0277 .0402 ,078
VOl. 55
defect for a number of the heavier nuclei using Latimer's nuclear model, but the present accepted mass of He4necessitates a revision of those calculations and, in fact, requires the use of the experimental values of the radii instead of the somewhat smaller values which they used. A summary of the calculated bond energies and the comparison with the values predicted by the model for the lighter elements is given in Table '111. The bond energies used are 0.005 for the bond between two mass particles and 0.01'7 between three particles. It should be emphasized that these values in terms of the model are not entirely arbitrary, the first is taken as 1/6 of the total experimental bond energy of He4, 0.031M units, since in a tetrahedron there are six pairs of interactions, and the second is approximately '/zthe He4 value. DEPARTMENT OR CHEMISTRY WENDELL M. LATIMER UNIVERSITY OF CALIFORNIA BERKELEY, CALIF. RECEIVED MAY14, 1936
TRANS- AND CIS-AS-OCTAHYDROPHENANTHRENE
Sir:
The hydrocarbon skeleton of morphine (or more correctly that of dihydromorphine) consists prin,0048 cipally of a 1,2,3,4,9,10,11,12-octahydrophenan.0051 threne (as-octahydrophenanthrene) nucleus, and .m1 therefore the preparation of derivatives of this .0262 hydrocarbon carrying functional groups of the .0251 morphine molecule has been undertaken. Amino .0301 .0254 alcohols derived from symmetric octahydrophenanthrene were described in a recent communication [van de Kamp and Mosettig, THISJOURNAL,
57, 1107 (1935)l. Model
In the preparation of as-octahydrophenan-
0.005 threne according to the synthetic methods known .015 so far, either the cis- or the trans-form, or more ,030 likely a mixture of the two, can be expected. -042
As far as we know no statements concerning the configuration of this hydrocarbon have been made Three types of reactions are given in Table 11: previously. We prepared the as-octahydrophen(1) the addition of a mass particle, proton or anthrene by effecting the dehydration and the neutron, to a 4n 3 nucleus, ( 2 ) the addition of a isomerization of 1-P-phenylethylcyclohexanol mass particle to a 4n nucleus and (3) the addition [prepared according to Cook and Hewett ( J . 2 nucleus. With few exceptions Chem. SOC.1098 (1933)l with phosphorus pentof H 2 t o a 4n the bond energies for each group are constant oxide in one step. The yield in this 1 s t step within the experimental error. Many other calculated on the carbinol, was 90%. Phosphorus examples may be given showing the same general pentoxide has already been employed by Bardhan and Sengupta [J.Chem. Soc., 2520 (1932)l in the constancy. Latimer and Libby (Ref. 1) calculated the mass ring closure of 2-~-phenylethylcyclohexanol.
+
+
.037
.OS3
COMMUNICATIONS TO THE EDITOR
June, 1936
The product thus obtained, after refluxing with sodium and distilling off, was fractionated, using a Widmer fractionating column. By repeated fractionation (until constancy of physical data was reached) the as-octahydrophenanthrene was separated into two main fractions, A and B, and a relatively small mixed fraction. Fraction "C. B' "* Mm. Refr. index d'6r
{
TABLE I A (20%) 135.5-135.7 10.5-10.8 ~ P1.5460 D 0.9828
B (70%) 142.6-142.8 9.2 n " J . 81.5592 ~ 1.0053
From a comparison of the above data with those of trans- and cis-decahydronaphthalene [Hiickel, Ann., 441, 42 (1925)], we assign by analogy to the hydrocarbon A the trans-configuration and to the hydrocarbon B the cis-configuration.
n
A
,.:u
1063
Zelinsky and Turowa-Pollak, Ber., 65, 1299 (1932)]. COBBCHEMICAL LABORATORY JACOBVAN DE KAMP UNIVERSITY, VIRGINIA ERICHMOSETTIC RECEIVED MAY13, 1936
STRUCTURE OF VITAMIN BI Sir: Certain provisional features of the structure previously proposed [THIS JOURNAL, 57, 229 (1935)] require revision. We now feel justified in proposing Structure I for the vitamin. CH. N=C-NHyHCl
I I HnC-C C-CH-N., II II N-CH
I--"
/C=C-CHKHzOH
I
I \CH-S
c1 I
N=COH
I I CHsC C-CHeSOaH I/ II N-CH
I1
N=C-OH
I
I
HC C-CH2OC2Hs.HNOa N-CH ll I/
111
We obtained by liquid ammonia cleavage of the vitamin a free base, CeHloNr, which gives a b double banded absorption quite different from A. d,l-TransB . d ,1- Cisthe single bands of 2,6, 4,6 or 5,6-diaminopyrimias-Octahydrophenanthrenes dines but closely akin to those of 5-alkyl 6-amino I t is possible that on repeating this separation pyrimidines. (Alkyl groups in position 5 have a on a larger scale, we will obtain slightly different profound influence on absorption of 6-amino and better physical data on the isomeric as-octa- pyrimidines; alkyls in other positions have minor hydrophenanthrenes. Furthermore, the possi- effects.) The ultraviolet absorption of an exbility that one of the fractions may be a constant tended series of pyrimidines provided convincing boiling mixture, still remains. We have prepared evidence that the second amino group of the base, from each of the isomers a methyl ketone, its C~HIONI, is in a side chain. This base forms a semicarbazone, and the carboxylic acid by oxida- dipicrate, m. p. 225', presumably identical with tion of the methyl ketone. the picrate of Windaus [Z. physiol. Chenz., 237, 100 (1935)j. TABLE I1 My associate, Dr. J. K. Cline, was also able to DERIVATIVES OF obtain from the amino sulfonic acid [THISJOURtrans- AND Cis-US-OCTAHYDROPHENANTURENE NAL, 57, 1093 (1935)] by the action of sodium in Trans-. m. p., OC Cis-. m. p., OC. liquid ammonia a small yield of a base, C6HeN3, -COCHs 94-94.5 oily -- sernicarbazone . 230-231.5 211-213 which was identified by mixed melting points of mixed m. p. 192-203 the picrates, 221°, as 2,5-dimethyl-ti-amino-COOH 226-228 230-232 pyrimidine which was synthesized for absorption mixed m. p. 180-190 studies. This is the first identified pyrimidine We wish to mention the remote possibility that to be obtained from the vitamin. Its significance the acetyl group and consequently the carboxyl: was greatly enhanced by subsequent synthesis of group are attached in different positions in the two I1 which is undistinguishable by any known hydrocarbons. Furthermore, it has to be taken means from the oxy sulfonic acid of natural origin into account that, in the Friedel-Crafts reac- [Ref. 31. tion, partial isomerization may have taken place We have synthesized five new ethoxy derivatlirough the action of the aluminurn chloride [cf. tives of G-oxypyrimicline; others are described
