July 20, 1957
CONXUNICATIONS TO THE EDITOR
3925
sion of the electrolysis. Heat is supplied by an colorless platelets [11.1 g., m.p. 105.5-107.5', XmSx 281 mp ( E = 3,160) and 287 mp ( e = 3,140), electric heating mantle. Mercury (5000 g.) is placed in the cell and chunks 5.82, 6.18, 6.37, and 7.37 p (chloroform): [a]"D +122'; calcd. for C&&: C, 73.71; H, 8.44; of electrolyte are added quickly. The cell is OCH3, 17.32. Found: C, 73.93; H, 8.43; OCH3, closed and the electrolyte is melted down in an 17.051. Compound I1 formed a bright yellow 2,4- atmosphere of argon. A carbon anode is then inserted and the electrolysis commenced :%tabout dinitrophenylhydrazone [m.p. 203.5-207.5°, Am,, 3.04, 5.83, 6.23, 6.33, and 7.61 p ; calcd. for C,S- 5 volts and 6 amperes a t temperatures of 300-350'. Amalgam tends to climb through the salt t o the H34N407:C, 62.44; H, 6.36. Found: C, 62.64; anode. Therefore it is desirable to stir the cathode H, 6.461. To confirm the structure assigned to 11, methyl during electrolysis to avoid premature termination 0-methyl-7-acetonylpodocarpatewas synthesized of the experiment. At the conclusion of the in the following manner: methyl O-methyl-7- electrolysis (50-100 amp.-hr.) the cell is opened (IV) was converted by a Will- and the inner sleeve removed with the electrolyte acetylp~docarpate~ gerodt reaction to the corresponding thiomor- within it. The amalgam, which varies in viscosity pholide [VI m.p. 177-180.5°, Amax 281 mp ( E = from a free flowing liquid to a semi-solid paste, 17,600), 5.81, 6.20, 6.37 and 6.71 fi (chloroform); depending on beryllium concentration, is found [cY]*~D +101" (ethanol); calcd. for Cz5H35N04S: floating on the mercury a t the bottom of the cell. S, 7.20; N, 3.14. Found: S , 7.20; N, 3.171. The current efficiency of the electrolysis is approxiHydrolysis of V with sulfuric acid in acetic acid mately 657,. The formation of amalgam evidently is sensitive gave the corresponding monobasic acid [VI, m.p. 144.5-147', presoftened, X max 281 mp ( E = 2,690) to small traces of oxygen, Rigorous purification and287mp ( E = 2,630), 5.82, 6.18, 6.35, and6.66p of the argon cover is required and it is essential (KBr); calcd. for C21H2805: C, 69.97; H, 7.83; that the cell be leak tight if amalgam is to be OCH3, 17.22; COOH, 12.49. Found: C, 70.23; formed. HI 8.03; OCHa, 16.87; COOH, 12.331. Addition It is evident from this description that it also of an ether solution of VI t o an excess of methyl- might be possible to prepare a beryllium amalgam magnesium bromide gave I1 which was identical by a displacement reaction with sodium amalgam [melting point, no depression on mixing: ultra- and the fused-salt mixture used in electrolysis. violet and infrared spectra] with the compound Properties of the Amalgam.-The amalgam is underived from the Willgerodt reaction on methyl 0- stable in air, decomposing spontaneously .with the methyl-7-propionylpodocarpate. formation of a black powder which is approximately When heated with sulfur and morpholine, methyl 75% Be by weight. Dilute amalgams can be conberyllium by pinholing. 0-methyl-7-acetonylpodocarpate (11) was con- centrated to about verted to 111 (4170 yield). The 27, amalgam is a thick pasty material from which additional mercury can be removed by such (3) W. P. Campbell and D. Todd, THISJOURNAL, 62, 1287 (1940). methods as pressing in a die or heating in vacuo. DIVISIONOF CHEMICAL RESEARCH Surprisingly, the room temperature X-ray difG. D. SEARLE AND COMPANY ROY H. BIBLE, JR. CHICAGO 80, ILLINOIS fraction pattern of the amalgam is the same as solid (-78') mercury. The exact nature of the RECEIVED JUNE 3, 1957 amalgam is not known. By analogy with the Mg-Hg system the compound BeHg2 ( 2 2 % Be) is suspected. It is hoped this information will assist THE PREPARATION OF BERYLLIUM AMALGAM' in further investigation of the Be-Hg systern. Sir: (5) Stanford Research Institute, Menlo P a k , California. The preparation of a beryllium amalgam by electrolysis of concentrated solutions of BeCL SIZVASIA-CORSISG M. C. K E L L ~ CORPORATION R. B . HOLDEN into a mercury cathode has been claimed by Prytz2 XUCLEAR C. I. WHITMAN but this method is not confirmed elsewhere. A new B a Y S I D E , NEW Y O R K RECEIVED JUNE 10, 1957 and novel method, involving electrolysis of beryllium from a NaCI-BeCI2 fused salt mixture into a mercury cathode has been developed in this laboTOTAL SYNTHESIS OF PENTACYCLOSQUALENEI ratory and applied t o the preparation of beryllium metaL3 Sir: Procedure.-Equimolar quantities of NaCl and Despite the probable genesis of the pentacyclic BeClzare melted together (224°)4to form an electro- hydropicene triterpenes2 from squalene by cationlyte. Beryllium chloride can be prepared by hydro- olefin cyclization with subsequent rearrangement chlorination of beryllium metal at 500". of carbon and h y d r ~ g e n , ~the $ ~ Jpentacyclic subThe electrolysis cell consists of a 4-liter resin stance I which would be formed from squalene by reaction flask with a cylindrical inner sleeve to (1) This investigation was supported by a fellowship (AF-6570) from facilitate removal of the electrolyte a t the conclu(1) Presented at Miami meeting of American Chemical Society April 7-12, 1957. Research supported by the Atomic Energy Commisslon. (2) M. Prytz, Z. anorg. allgem. Chem., 199, 113 (1930). (3) M. C. Kells, R. B . Holden, C. 1. Whitman, AEC report SEP-207 (1956). (4) J. M. Schmidt, Bull. SOC. chim.. 39, 1686-1703 (1926).
the National Institute of Arthritis and Metabolic Diseiises, Public Health Service. (2) u-Amyrin, 8-amyrin, taraxerol, taraxasterol, friedelia, alnusenol and their oxygenated derivatives. (3) A. Eschenmoser, L. Ruzicka, 0. Jeger and D. Arigoni, Hclu. Chim. A d a , 38, 1890 (1955). (4) E. J. Corey and J. J. Ursprung, THIS JOURNAL, 76, :.83 (1956). ( 5 ) T. T. Tchen and R. Bloch, i b i d . , 7 8 , 1516 (1956).
