Nov., 1935
COMMUNICATIONS TO THE EDITOR
2331
sults for samples of (11) obtained from two differ- complete decomposition the pressure is approxient experiments axe: mately double the initial pressure of azomethane ; (2) the ratio of the pressure increase at room temCr found perature to the pressure of “permanent” gases formed by the reaction however, diminishes 44.507,) Run 44.39 44.47 Run2 14.49 Run2 from a value of a t least two, in the initial stages, to unity a t complete decomposition. A plausible Mean 44.46 Mean 14.47 assumption to explain these facts is that the prodCr calcd., 44.44 NHs calcd., 14.55 ucts include, with nitrogen, other “permanent” This compound has not been reported previously gases consisting largely of methane. This tentaand unlike the parent substance gives no test for tive conclusion is being tested by chemical analythe peroxide group. (4) The presence of small sis. amounts of impurities has a marked influence The above observations hold for both photoupon the rate of decomposition of (I), in some chemical decomposition at 30’ and for thermal cases increasing it to the point of explosive viodecomposition a t 300”. The course of the therlence (danger!). In some of our experiments (I) mal reaction at 300’ was unaffected by mercury was purposely left impure and may have contained ammonium chromate, dichromate or per- vapor at a pressure of 100 mm. approximately, when this was intentionally introduced. chromate. (5) The reaction does not proceed The thermal reaction rates in quartz fit the under the conditions mentioned in (2) if carried unimolecular formula, like Ramsperger’s, but our out in the dark or near-dark. Radiation, in or rate constants are roughly half as great as his. near the visible, is apparently essential to start The decomposition in Pyrex is being investigated. the reaction. (6) If the starting pressure in the The ten-degree temperature coefficient from the tube be approximately 1 cm., (I) decomposes dark reaction rates a t 260, 270, 280 and 290’ is in the same scintillating manner as under ( 2 ) 2.9, in agreement with that of Ramsperger, and but the product gives the peroxide test and its apparently affected but little by the presence of composition is variable (37-38% Cr, 21-22% mercury vapor. NH3), lying between that of (I) and (11). It CHEMICAL LABORATORY probably consists of a mixture of the latter two. MALLINCKRODT HARVARD UNIVER~ITY LAWRENCE J. HEIDT (7) The density of (11) is 2.073 g./cc. at 23.5’. CAMBRIDGE, MASSACHUSETTS GEORGES. FORBES RECEIVED SEPTEMBER 25, 1935 It is practically insoluble in carbon tetrachloride, stable in air under ordinary conditions, and hydrolyzed in water. It detonates in the neighborSEPARATION OF GUANIDINE AND hood of 200” in the atmosphere, leaving Crz03. METHYLGUANIDINE BY MEANS OF BETAIt is interesting to note that (11) may be formuNAPHTHALENESULFONYL CHLORIDE lated as a partial dehydration product of ammo- Sir: nium dichromate : (NH&Crz07 = 2Cr03.NH3 In a study of the relation of simple and substiH20, and that neither ammonium dichromate nor tuted guanidines to health and disease, occasion chromate shows any sign of reaction under the came to separate guanidine from methylguanidine. conditions mentioned in (2). Finding the benzenesulfonyl derivatives, preDEPARTMENT O F CHEMISTRY WILLIAMF. EHRET pared by Ackermann [Z. physiol. Clzm., 47, WASHINGTON SQUARECOLLEGE ARTHURGREENSTONE NEWYo= UNIVERSITY 366 (1906); 48, 382 (1906)] unsuitable for the NEWYORK,N. Y. differential precipitation of these guanidines, reRECEIVED JULY30, 1935 course was had to beta-naphthalenesulfonyl chloride, which was found to give hitherto undeTHE PHOTOLYTIC AND THERMAL DECOMPOSI- scribed derivatives of guanidine and methylTION PRODUCTS OF AZOMETHANE. THERMAL guanidine, suitable for the separation of these REACTION RATES IN QUARTZ, 260 TO 290’ substances from each other and from other closely Sir: related compounds. Experimental. Di-(6-naphthalenesulfony1)Pressure readings at room temperature, - 78.5 and -183’ at various stages of decomposition of guanidine.-To one mol. of guanidine carbonate initially pure azomethane show that (1) upon (1.5 g,) dissolved in 10 cc. of water was added 1.0
+
