March 20, 1957
COMMUNICATIONS TO THE EDITOR
1517
nature of pure ozone and the extreme facility with which i t explodes or detonates, has so far prevented combustion studies with pure ozone. C& I Before studies of the behavior of pure ozone with various fuel gases could be started profitably i t was first necessary to achieve the combustion or decomposition flame of pure ozone to oxygen. It could be expected, since the exothermic heat (at constant When a-chlorocyclopentanone is allowed to react pressure of 1 atm. and 291’K.) of the reaction O3+ with ethyl xanthamidate in boiling 1-propanol, 1.5 O? equals - 33,923 f 180 cal./mole, that apthere was obtained the crystalline compound, preciable flame temperatures should be attained 2,3,5,6 - tetrahydro - 4(H) - cyclopentathiazolo - 2- and that a stable ozone-oxygen flame should exist. one. Purification by recrystallization from water Flame temperature calculations can be made with or water-ethyl alcohol (9 :1) mixture gave product great accuracy since the enthalpy and the dissociaof m.p. 144-145’; Anal. Calcd. for C6H7NOS: C, tion constants for the 0 2 e 2 0 reaction are well 51.06; H, 5.02; N, 9.93; S, 22.60. Found: C, known ; these temperatures are, a t initial condi51.02; H, 4.96; N, 9.96; S, 22.90. Ultraviolet tions of 298’K. and 1.0 atm. pressure, for 100, ’ 0s in 0 2 , respectively, 66.7, 40.0 and 18.2 mole % absorption measurements a t PH 7 gave 252 mp, E 4430; ”’?A: 229 mp, E 2500. Meth- 2677O, 2277”, 1687” and 1027°K. By using pure ozone, containing less than a few ylation of this thiazolone with methyl iodide in a parts per million of organic impurities, as debasic medium gave rise to I which, after purification by crystallization from water, melted a t scribed originally by C. E. Thorp,2 we have been 70-71”. Anal. Calcd. for C7HsNOS: N, 9.04; able to burn ozone-oxygen mixtures to oxygen in S, 20.68. Found: N, 8.90; S, 21.00. Absorp- the entire range from 17-100 mole yo 03. The alltion data in the ultraviolet are 253 mp, Pyrex glass apparatus was extremely simple; i t consisted of a narrow glass gasholder, from which 231 mp, E 2720. E 4170; The pharmacodynamic evaluation of this com- any desired mixture of 0 8 - 0 2 or pure 0 3 could be pound revealed a rather surprising activity in re- delivered a t any predetermined rate, by simple disFrom the gasholder the ducing experimental pain. The Wolf-Hardy prin- placement with water. ciple for testing analgesia was employed in experi- gas mixture went to a Pyrex glass, quartz, or alumental animals according to a method described minum tip. Stopcocks greased with C,F?,+ 9 or by Gross.’ A beam of heat was directed against Kel-F, were used. The burning velocities were the tips of the tails of white mice and the reaction determined by the standard schlieren method. time measured from the onset of pain stimulus to In the range of 17 to about 50 mole % 0 3 , the flame the tail flick. Analgetic effects were expressed in cannot be observed visually, but can be seen very terms of prolonged reaction time to the pain stimu- easily on the screen of the schlieren apparatus. lus. With this method the above-mentioned com- The flame is visible above this range. Pure 100% pound produced a marked prolongation of the re- ozone burns with a faint, non-luminous flame, blue action time in the range of one-tenth to one-fifth in color, with a typical pink cast. The experiof the LD50. It was characterized by a rapid on- mental burning velocities, a t 298°K. and 1.0 atm. set of action and a maintenance of the analgetic pressure, are shown in Fig. 1. The ozone flame is of particular theoretical ineffect for several hours following parenteral administration. Compared to aminopyrine the mate- terest since i t is the simplest flame imaginable. rial was more potent, somewhat less toxic and de- Outside of the “fuel” O3 and the “product of combustion” 0 2 , the only possible intermediates are void of antipyretic activity. oxygen atoms. (6) According to Chemical Abstracts, an alternate name for I is NIn recent years Drs. J. 0. Hir~chfelder,~ Theomethylcyclopentano- [d]4-thiazolin-Z-one. dore von K8rm&n4and R. Sandri,5 and their as(7) F . Gross, Helu. Phrsiol. A d a , C31 (1947), w5. RESEARCH DEPARTMENT GEORGE DESTEVENS sociates, have developed the theory of laminar CIBAPHARMACEUTICAL PRODUCTS, INC. HEINOA. LUTS flame propagation. Dr. von K & r m h presented SUMMIT, YEWJERSEY J U R GA. SCHNEIDER his results recently6 and they are in essential agreeRECEIVED JANUARY 30, 1957 ment with Fig. 1. The more extensive data of Dr. Sand$ are given in Table I and are compared with THE PURE OZONE TO OXYGEN FLAME1 our experimental results in Fig. 1. As can be seen,
getics. This new substance is 2,3,5,6-tetrahydro3-methyl-4(H)-cyclopentathiazolo-2-one (I).6
AzzroH
Sir :
We live in a world of molecular oxygen and an overwhelming number of studies on combustion are devoted to oxidation with molecular oxygen. On the other hand, the active modification of the element, or ozone, has not been studied in the pure form. Ozone has been known since 1785, when van Marum observed its formation in the electric spark discharge in oxygen. The highly sensitive (1) This research was supported by the United States Air Force through the Air Force O 5 c e of Scientific Research of the Air Research and Development Command under Contract No. 18(600)-1475.
(2) C. E. Thorp, U. S. Patent 2,700,648, Jan. 25, 1965; see also “Bibliography of Ozone Technology,” Vol. 11’ “Physical and Pharmacological Properties,” 1955, Armour Research Foundation of Illinois Institute of Technology, Chicago, Ill. (3) J. 0.Hirschfelder, C. F. Curtiss and D. E. Campbell, J . P k y s . Chem., 67, 403 (1953); Proc. IVth International Symposium on Combustion, 1953, p. 197. (4) T . von Kdrmdn and S. S. Penner, “Selected Combustion Problems,’’ (AGARD) : Combustion Colloquium, Cambridge University, England, pp. 5-41 (1953), and C. A . , 48, 10409e (1954). (5) R . Sandri, Canadian J . Chcm., 3 4 , 313, 324, 331 (1956). (6) T.von Kbrm&n, Proc. VIth International Symposium on Combustion, held at Yale University, Aug. 19-24, 1956. (7) An extension of Sandri’s theory to ozone-rich mixtures will be published soon in Canadian J . Chcm. (1957).
COMMUNICATIONS TO THE EDITOR
1,ilS
0 LU
______ eo
I 40
60
MOLE
Y' o
03 ---
Fig 1.-Burning velocities of ozone and ozone A oxygen mixtures (initial gas temperature = 298"K., pressure = 1 atm. abs).
pure ozone a t 1 atm. pressure and a t an initial teniperature of 298°K. (25OC.) has a burning velocity of 472 ( * 12) cni./sec.; a t 195'K. i t is 270 (f7) cm./sec. Pure ozone can be burned to oxygen for substantial periods of time without explosion and detonation, like a regular combustible gas mixture. TABLE I VELOCITIESFOR INITIALT = 298°K 1. PLAVE Mole '101 25 40 50 60 TO SO 90 Burning velocity, Vo (cm./s 41 139 214 290 349 386 411
100 430
Pure ozone was combusted with various fuel gases. The diffusion flame of pure ozone and cyanogen is extremely bright and produces a ternperature of 5ZOOOK. a t 1.0 atrn. A detailed description of our experiments will be published elsewhere.8
Vol. 79
dine, which is then nitrosated, reduced, formylated and subsequently ring-closed.2 The preparation of ?-alkyl- and 2-aryladenines is more cumbersome, since direct condensation of arnidines with malononitrile proceeds anomalously* and the desired 2substituted-4,6-diaminopyrimidinesmust be prepared by other methods4-? The synthesis of derivatives such as isoguanine (2-hydroxy-6-aminopurine) is also circuitous, since urea does not condense satisfactorily with malononitrile and the 2hydroxy group must be introduced i n d i r e ~ t l y . ~ . ~ , ~ We now wish to describe a facile synthesis of 2-substituted adenines which would appear to be generally applicable. Thermal cyclization of amidine salts of isonitrosonialononitrile (I) (R = -CH3; m.p. 141-143'. Anal. Calcd. for C5H;N5Q; C, 39.2: 11, 4,R; N, 45.7. Found: C, 39.4; H, 4.4; N, 45.4. R = -Ce,H5; m p . 14S150'. Anal. Calcd. for Clo€19NSO: C, 55.8; H, 4.3; N, 32.5. Found: C, 55.7; H, 4.0; N, 32.6. R = -NH2; m.p. 157-158'. A d . Calcd. for C4H6N60: C, 31.2; H, 3.9; N, 54.5. Found: C, 31.3; H, 3.9; N, 55.0) in 2-methyl-5-ethylpyridine yielded 2-substituted 4.6-diamino-5-nitrosopyrimidines (11) (R = -CI&5; Anal. Found: C, 39.2; H, 4.6; N, 46.1: K = -CsH,; Anal. Found: C? 55.9; H. 3.9; N, 32.6: R = --NHp;" Ami. Found: C, 30.7: H, 3.5; N, 55.0). The salts (T) were prepared readily in almost quantitative yield by mixing an aniidine hydrochloride in ethanol solution with the silver salt of isonitrosonialononitrile,I1 removing silver chloride by filtration and concentrating the ethanol. In some instances, isolation of I wa? not necessary prior to cyclization; for example, heating guanidine carbonate with potassium isonitrosomalononitrile (I1I) (m.p. 209211'; Anal. Calcd. for C3NaOK: N, 31.6. Found: N, 31.6) in dirnethylformamide yielded 2,4,6triamino-5-nitrosopyrimidine ('I1: R = -NH2) in SS% yield, and condensation of IT1 with urea in sodium ethoxide solution yielded 2-hydroxy-4,6-
diamino-5-nitrosopyrimidine. 9,12 Heating these 2-substituted 4,6-diamino-5-nitrosopyriniidines with a mixture of formamide, formic acid and sodium hydros~lfite'~ yielded 3substituted adenines in high yield. I n this manner, 2-methyladenine5*14t1h was prepared in 74% over-
(2) For a recent review of purine chemistrv, see A. Bendich in "The Nucleic Acids, Chemistry and Biology," ed. by E. Chargaff and J. N. Davidson, Vol. 1, Academic Press, New York, N. Y.,1955, p. 81. (3) G. W. Kenner, B. L?tbgoe, A. R. Todd and A. Topham, J . Chern. Soc., 388 (1943). (8) See (a) Proc. VIth International Symposium on Combustion. (4) G. W. Kenner, B. Lythgoe, A. R. Todd and A. Topham, i b i d . . Aug. 19-24, 1956, (b) Proc International Ozone Conference Chicago, 574 (1943). I l l , Nov. 28-30. 195G ( 5 ) J. Baddiley, B. Lythgoe, D. McNeil and A. R. Todd, ibid., 383 THERESEARCHINSTITUTE OF (1943). TEMPLE UNIVERSITY A. G. STRENG ( 6 ) H. R. Heme. W. J. Clegg and C. W. Smart, J . Org. Chent., 11, PHILADELPHIA 44, PENNSYLVANIA A. V. GROSSE 1320 (1952). RECEIVEDFEBRUARY 8, 1957 (7) G. A . Howard, B. Lythgoe and A. R. Todd. J . Chem. Soc., 476 (1944). ( 8 ) H. L. Wheeler and G. S. Jamieson, A m . Chem. J . , 32,342 (1904). A FACILE SYNTHESIS OF 2-SUBSTITUTED (9) A. Bendich, J. P. Tinker and G . B. Brown. THISJOURNAL, 70, ADENINES' 3109 (1948). (10) a '. Traube, Ber.. 37, 4A-L-I (1904). (11) G. Ponzio, Gnzz. chim. itol., 61, 561 (1981). The conventional synthetic route to adenines in(12) H. Plielaiid a n d K . Liebig, A n x . , 5 5 5 , 146 ( l n i i i . volves the condensation of guanidine or thiourea (13) H. Hredereck and A . Erlenhofer. Brr , 8 8 , li30Ci (1953). with malononitrile to give a 4,6-diarninopyrimi(14) IT. W. Dion, I).G. C d k r n s and J . J I'tiffner, T H I S JOliKNAl, 1 6 , 948 (1954). (1) This investigation was supported by a research grant (C-2.551) (15) .f. Baddiley, B. 1,ythyoe aiirl A I
