Lack of time and material does not permit US to obtain additional evidence regarding these alternate linkages. Logical approaches would be by the use of the Edman method on the aspartyl-glutaminyl-leucine peptide' and by partial acid degradation of the nonapeptide mentioned above. Details of this and the earlier work will be reported shortly. EXPERIMENTAL THERAPEUTICS A Y D KATHERINE S. HOWARD MEDICINAL CHEMICAL SECTIONS ROBERT G. SHEPHERD RESEARCH DIVISION E. ANNEIGNER AMERICASCYANAMID COMPAXY DAVIDS. DAVIES STAMFORD, CONNECTICUT PAVLH. BELL RECEIVED MAY11, 1963
agents such as hydrogen over platinum to the corresponding hydroquinone. I can be converted to a variety of stable substitution products (11-VII) which have some chemical properties analogous to those of substituted benzoand naphthoquinones. X
11,
x = c1
111, X = Br
V, X = -0CH3 VI, X = -NHu YII, x = -OH
0
I reacted with chlorine or bromine in acetic acid to yield the yellow halogen derivatives I1 and 111, respectively. I1 had m.p. 114-115'. Anal. Calcd. PHENYLCY CLOBUTADIENOQUINONE for CloHsOzCl: C, 62.36; H, 2.62. Found: C, Sir: H , 2.54. I11 had m.p. 128-129'. Anal. No authentic cylobutadiene derivatives appear to 62.20; Calcd. for C10H602Br:C, 50.66; H , 2.13. Found: have yet been prepared except for biphenyleneIa and related substanceslb)?.Bwhere the four-niem- C, 50.64; H , 2.13. Reduction of I11 with amalgamated zinc and hydrochloric acid gave phenylcybered rings are part of fused-ring system^.^ The clobutane. 111 reacted rapidly with alcoholic silfailure to detect cyclobutadiene or its non-fused ring ver nitrate and, with sodium iodide in acetone, derivatives among the products of appropriate yielded IV, m.p. 163-165' (dec.). Anal. Calcd. synthetic reactions can be ascribed either to excesCloH50zI: C, 42.26; H, 1.77. Found: c, sive ring strain or unfavorable electronic configura- for 42.38; H , 1.87. Methanolysis of 111 afforded pale t i o n ~ . ~Strong ,~ evidence that ring strain is not the yellow V, m.p. 151-152' (Anal. Calcd. for most important factor is now provided by a syntheCI'Ha03: C, 70.20; H, 4.28. Found: C, 69.51; sis of the quite stable phenylcyclobutadienoquinH , 4.16) while ammonia in dry benzene gave the one (phenylcyclobutendione, I ) which to a first ap- colorless essentially neutral amino derivative (VI), proximation is expected to have the same degree of m.p. 279-281' (dec.). Anal. Calcd. for C U I H ~ O ~ N : ring strain as phenylcyclobutadiene. C, 69.36; H, 4.07; N, 8.09. Found: C, 69.37; H, 4.04; N, 8.08. With dilute acetic acid, 111 CBHb- / =O was hydrolyzed to colorless VII, m.p. 205-208' I1 (dec.). VI1 is soluble in water, is an exceptional I O strong acid for a C, H, 0 compound ( ~ K A 1) and gives a magenta color with ferric chloride soluPhenylacetylene with trifluorochloroethylene a t tion. Anal. Calcd. for CloHaOs: C, 68.96; H, 120' for 24 hours gave 1,1,2-trifluoro-2-chloro-3-3.48; neut. equiv., 176. Found: C, 68.87; H, phenylcyclobutene6; b.p. 52-53' (0.4 mm.), n% 3.61; neut. equiv., 174. VI1 is converted to V 1.,5117. Anal. Calcd. for Cl~H6F3C1:C, 54.92; with diazomethane. H, 2.77. Found: C, 54.83; H, 2.89. Hydrolysis6 The stable existence of I-VI1 suggests the possiof the adduct with 92% sulfuric acid a t 100' af- bility that quinone-type derivatives of unknown forded bright-yellow crystalline I in 73% yield; cyclic polyolefins such as pentalene and heptalene m p . 152-1 53' (dec.) after recrystallization from might be more stable than the parent hydrocarbons. acetone. Anal. Calcd. for C10H602: c, 75.94; Investigation of the chemical and physical propH , 3.52; mol. wt., 155. Found: C, 76.00; H , erties of I-VI1 is being continued. 3.85; mol. wt. (Rast), 151. I had Amax = 286 mp, CONTRIBUTION No. 2000 E 2.3 X lo5and strong infrared absorption a t S.6 p . GATESAND CRELLINLABORATORIES EDGAR J. SMUTNY -1chloroform solution of I with 30%. hydrogen per- CALIFORNIA INSTITUTE OF TECHNOI,OGY J O H N D. ROBERTS oxide yielded phenylmaleic anhydride, m.p. 119- PASADENA 4. CALIF. 120', which did not depress the m.p. of an authenRECEIVED MAY20, 1955 tic sample.' Reduction of I with amalgamated zinc and hydrochloric acid yielded phenylcyclobuNEW REAGENT FOR RESOLUTION BY COMPLEX tane which was identified by comparison of its in- A FORMATION ; THE RESOLUTION OF PHENANTHROfrared spectrum with material previously prepared.6 13.4-cl PHENANTHRENE' I is much more stable to water and oxygen than o- Sir: benzoquinone. It is apparently not reduced by We wish to report the synthesis of 2-(2,4,5,7-tet(1) (a) W. C. Lothrop, THISJ O U R N A L , 63, 1187 (1941); (h) 64, ranitro-9-fluorenylideneaminooxy)-propionic acid, 1098 (1942). 111, a useful reagent for the optical resolution of cer(2) R . F. Curtis and G. Viswanath, Chem. ond Ind., 1174 (1954). tain aromatic compounds which do not have the (3) M. P. Cava and J. F. Stucker, ibid., 440 (1955). functional groups usually needed to effect resolution (4) An excellent survey of attempts t o prepare non-fused ring cyclobutadiene derivatives was recently presented by E. R . Buchman, Ahst. by conventional reagents. of A.C.S. meeting, Sept. 13. 1954, p. 9-0. The dextro and levo forms of 111 were made by ( 5 ) For references and discussion see J . D. Roberts, A. Streitwieser, condensing 2,4,5,7-tetranitrofluorenone,I, with Jr.. and Clare hl. Regan, THISJ O U R N A L , 7 4 , 4579 (1952).
-
.
(6) Cf.J . D. Roberts, 0 . R . Kline and H. E. Simmons, J r . , ibid., 76, 4765 (1953). ( 7 ) I , . E. Ililler, H. R . Staler and n. J . hlann, Ibid., 71, 374 (1949).
I
I
(1) A p a r t of this material was presented before t h e Division of Organic Chemistry a t t h e 127th meeting of t h e American Chemical Society, Cinclnnatl, Ohio, >larch, 19n5.
June 20, 19,3,i
COMMUNICATIONS TO THE EDITOR
-3k2 1
( - ) and ( +)-(isopropylideneamino6xy)-propionic lution of a variety of racemic aromatic compounds. acid, 11, respectively, in the presence of p-toluene- The extensions and limitations in the use of I11 and analogous compounds is under study as well as the sulfonic acid exchange reaction represented by equation (1). 0 The reagent I1 was made by treating the sodium salt of acetoxime with methyl 2-brom0propionate.~ 2-(Isopropylideneamino6xy)-propionicacid, 11, m.p. 58-61', (Found: C, 49.5;N, 9.7), was resolved by treating two equivalents of racemic IT with 1.5 equivalents of (-)-ephedrine in 1: 1 benzene-SkelI1 I lysolve B. The resulting salt, m.p. 116-llRo, was CH? decomposed with hydrochloric acid. The acid SOCHCOOH thus produced was recrystallized from 5 : 1 Skellysolve B-acetone to give (+)-11, m.p. 83.5-85.3', [ c Y ] ~ $30.8' ~D (c, 7.6,water, 2 dm.). ( -) -2- (2,4,5,7-Tetranitro-9-fluorenylideneaminooxy)-propionic acid, m.p. 198-200° (Found: NO2 SO:! C, 43.0, H, 2.0, N, 15.8)) [ c ~ ] ~-97.1' ~ D ( 6 , 1.5, di111 oxane), was made by gently refluxing a mixture of When a solution of two equivalents of l-naph- one equivalent of 2,4,5,7-tetranitrofluorenone,1.5 thyl-sec-butyl ether and one equivalent of (-1-111, equivalents of (+) -2- (isopropylideneaminooxy) [a]*'D -91.4' (c. 1.6, dioxane), in acetic acid was propionic acid and 1.7 equivalents of p-toluenesulcooled a purple complex separated. The ethers re- fonic acid monohydrate, in glacial acetic acid. The covered from the complex and the mother liquor product was then precipitated with water and the were optically active, [ a I 3 l D -7.8' (c, 6.4 ethyl yellow solid recrystallized from a mixture of acetic acetate) and [ c ~ ] ~ ' D4-6.4' (c, 11, ethyl acetate), re- acid and butyl chloride. The (+)-I11 was made spectively. Similarly, methyl 2-(1-anthry1)-propio- similarly by starting with I and (-)-11. nate was partly resolved into the corresponding (4) Compare E. Borek and H. T. Clark, THISJ O U R N A L , 68, 2020 (+)- and (-)-forms, [a]"D + l o l o and [a]''D (1936) ( 5 ) T h e substance also melts and resolidifies or sinters between 125O -66.0' by means of ( + ) - I I I , [ a ] " D +84' (c, 1.3, dioxane). The above rotations are undoubtedly not and 139". maximal but represent the result of a single separa- MCPHERSON CHEMICAL LABORATORY MELVINS. NEWMAN UNIVERSITY WILSONB. LUTZ OHIOSTATE tion of complex. COLUMBUS 10, OHIO DANIELLEDNICER RECEIVED APRIL21, 1955
BISCYCLOPENTADIENYLRHENIUM HYDRIDE-A NEW TYPE OF HYDRIDE
Sir:
When rhenium pentachloride (0.03 mole) is allowed to react with sodium cyclopentadienide (0.2 When a solution of phenanthro [3,4-c]phenan- mole) in tetrahydrofuran solution and the evapothrene,2 IV, m.p. 331-231.5° (uncor.), in benzene rated purple reaction mixture is heated a t 120-1 70' was treated with one-half equivalent of the complex- in vacuum, a yellow crystalline sublimate, m.p., based on the ing agent the solution became dark red. Upon the 101-l~i2', is obtained in yields -207' addition of ethanol, the hydrocarbon separated as halide. The molecular weight of the compound by hexagonal plates. Three further applications of this the isothermal distillation method in benzene and treatment followed by nine conventional recrystal- ether is 312 A 3. The compound cannot, however, ~ be biscyclopentadienylrhenium, (CSHS)ZRe,since it lizations afforded IV, m.p. 263-267') [ a I z 4-3640' (c, 0.0988, chloroform). The optical stability of this is diagmagnetic in the crystal and the melt from 77 compound is indicated by the fact that i t remelts a t to .510'K., and in organic solvents. The only rea234-250', showing that even a t its map.racemiza- sonable explanation of this fact is that the comtion is not complete. This is the first example of pound is a hydride. Chemical analysis does not, the resolution of a hydrocarbon which owes its however, provide an unequivocal answer [ A n d . asymmetry to intramolecular overcrowding. Found: C, 37.7; H, 3.6; Re, 38.'iT,. Required for N-sec-Butylpicramide, a compound which might (CSH5)2ReH; C, 3 7 . 8 5 ; H, 3.39; Re, 58.6(i; mol. have been used for resolution by means of complex wt., 317.51. Proof of the hydridic nature of the compound formation has been prepared.J Racemic N-sec-butylpicramide was partially resolved by complex comes from studies with a Varian Associates high formation with (+)-P-naphthylcamphylamine but resolution nuclear maqnetic resonance spectromethe partially resolved reagent was never used to re- ter. In carbon disulfide solutions, a resonance solve any other compound. peak attributable to the protons of the cyclopentaBecause of the considerable complexing ability of dienyl rings is observed with a shift relative t o the 111, this reagent should prove valuable for the reso- reference, water, of - 1 . 1 i p.p.m. (H-Hr/H,); a second resonance peak of approximately one-tenth (2) The synthesis of this hydrocarbon will be described later. the intensity is observed with a displacement of (3) R . Weiss and A. Aheles. Monntsh., 19, 238 (1932).
