INITIATEDBY NUCLEAR PROCESSES HALOGEN ATOMREACTIONS
July 5 , 1957
kcal. and Coughlin’s free energy values show a probable error due to using an estimated entropy of germanic oxide. Table I1 shows the heat and entropy of formation of germanic oxide, both amorphous and hexagonal, together with the corresponding values of ~ i l i c a . ~For germanium the comparison was made a t 850°K. in order that the error due to the use of the estimated heat capacity may be avoided in the data for the hexagonal modification. TABLE I1 Amorphous Hexagonal
H,kcal. -127.6 - 131.7
S,e.u. -40.6(calcd.)
Amorphous Tridymite
-202.5 -204.8
-42.27 -43.11
State
CrO* (at 850’K.)
Si02 (at 298°K.)
-41.4
Solubility of Oxygen.-In the above discussion it was assumed that germanium metal does not dissolve oxygen. However, if there is any solubility in this case, it would decrease the activity of germanium. Contrary to this apprehension, Candidus and Tuomi’O have reported that the solubility is negligible, and Trumbore and his coworkers11-12 have shown error in the work of Hoch (9) F. D.Rossini, D. D. Wagman. W. H. Evans, S. Levine a n d I. Jaffe, “Selected Values of Chemical Thermodynamic Properties,” Series I, N a t . Bur of Standards, Washington, 1952. (10) E.S. Candidus a n d D. Tuomi, J . Chem. Phys., $3, 588 (1955). (11) F. A. Hassion, C. D. Thurmond and F. A. Trumbore, J . Phys. Chcm., 69, 1076 (1955). (12) F.A. Trumbore, C . D. Thurmond and M. Kowalchik, J. Chcm. Phys., 24, 1112 (1956).
[CONTRIBUTION FROX
THE
3367
and Johnstonls who observed a large solubility of oxygen. It may be concluded, therefore, that germanium does not dissolve oxygen appreciably. Supplement.-After the present experiment had finished, the authors found that Ono and COworkers14 had investigated the same equilibrium by a flow method. They obtained -141.8 kcal./ mole as the standard heat of formation of germanic oxide. They measured only from the reduction side, while the present measurement were carried out from both the reduction and oxidation sides with sufficient agreement. Besides, the present results seem to show that a circulation method is more suitable than a flow method, since hydrogen pressure is comparable to water vapor pressure in equilibrium. Then, the present results may be more accurate. Acknowledgment.-The authors wish t o express their thanks to Prof. Leo Brewer for his friendly criticism and to Prof. Isao Kayama, Hokkaido Teachers College, who persuaded them to participate in the research group studying germanium. The sample of germanic oxide was supplied by Mr. Hideo Baba, Electrical Communications Laboratory. The preliminary testing of this experiment was carried out by Mr. Satoshi Kado, t o whom the authors are also indebted. ‘(13) M.Hoch and H. L. Johnston, ibid., 22, 1376 (1954). (14) K. Ono, Y.Inada and I. Konno, Bull. Research I n s t . of M i n . Dcssing and Mctall.. 11, 159 (1955).
SAPPORO, JAPAN
DEPARTMENT O F CHEMISTRY OF THE UNIVERSITY OF WISCONSIN]
Halogen Atom Reactions Initiated by Nuclear Processes in Hydrocarbon Solutions BY SUKUMAR ADITYA’AND JOHN E. WILLARD RECEIVED FEBRUARY 4,1957
I.T.
The organic yields of It2*and Brm activated as a result of t h e P7(n,7)11*8, BrTs(n,y)Brm and BrSom+Br“ processes are higher in dilute hydrocarbon solutions than in pure organic halides. They increase with increasing chain length of the hydrocarbon. For the (n,?) activations in hydrocarbon solutions they decrease by about one-fifth in going from 0 to 100’. These variations in yield are all rather insensitive t o the addition of free halogen as scavenger and therefore involve “hot” processes. The results, correlated with those of other investigations, suggest that a number of previously unexplained facts about reactions of hydrogenous material with 1’28, Br80 and CIS8 activated by the (n,y) process may be explained by ionic reactions. The organic yields fromithe Braomisomeric transition in liquid hydrocarbons depend on the source molecule containing the Br8Om t o a n extent which is consistent with earlier observations of failure of the transition t o rupture C-Br bonds in liquid bromine as the solvent.
Introduction There are several reasons for studying the chemical consequences of activation of halogen atoms by nuclear processes2 in each of a series of hydrocarbons: (1) the sizes and weights of the molecules with which the energetic halogen atom is surrounded while losing its energy can be varied without varying the types of bonds to be broken (;.e., only C-C and C-H); (2) conservation of momentum requirements make it impossible for a halogen atom to enter organic combination in such a medium by a “billiard ball collision” mechanism ;
(3) preliminary observations3 on such systems have shown changes in organic yields4 dependent on the size of the hydrocarbon molecules; (4) such studies should contribute to a better understanding of the mechanismsof the various hydrogen replacement reactions cited in the next paragraph; ( 5 ) the mechanism of replacement of H1 by Ha in tritium recoil reactions5a (and possibly also in systems activated by the decay of tritiumsb may be similar t o
(1) Fulbright-Smith-Mundt Scholar 1953-1954, Project Assoc. (1954-19551, Univ. of Wisconsin; Ravenshaw College, Cuttack. India. (2) For a discussion of the physical processes involved and a review of earlier work on their chemical consequences see: J. E. Willard, Ann. R69. Nucl. Sci., 3, 193 (1953); Ann. Rev. Phy:. Chem., 6 , 141 (1955).
nuclear transformation which become stabilized in organic rather than inorganic combination. (5) (a) See, forexample, and references: R. Wolfgang, J. Eipner and F.S . Rowland, J . Chem. Phys., 60, 1137 (1956); (b) K. E. Wilzbach, TEXS JOURNAL, 1 9 , 1013 (1957).
(3) (a) S. Goldhaber, R. S. H. Chiang and J. E. Willard, THIS 73, 2271 (1951); (b) S. Goldhaber and J. E. Willard, ibid., 74, 318 (1952). (4) T h e organic yield is the percentage of the atoms produced by JOURNAL,
SUKUMAR ADITYAAND
3368
JOHN
E. WILLARD
Vol. 79
that of replacement of hydrogen by halogens activated by nuclear processes. I t has been shown that both iodine6and bromine’ atoms activated by the ( n j r ) process can replace hydrogen in pentane, forming the amyl halide, that C13aatoms from the C137(n,r)C138 process undergo similar reactions in heptane,8 cyclohexane and ben~ e n eand , ~ that the 1127(n,y)1128 reaction in liquid CHI produces both CH3112S and C Q H J ~Further ~ ~ . ~ evidence indicates that substitution for hydrogen can occur by a unique “hot” non-radical reaction; i.e., atoms or ions freshly formed by the 1127(n,y)1128 process reactla with gaseous CHd in a process of the type IlZ8 CHI --L CH3IIZ8 H , and freshly formed C13s atoms give a similar reaction with gaseous butane or butyl chloride.8
Extraction and Counting.-Following neutron exposure of solutions of the alkyl iodides, a small amount of elemental halogen was added t o the irradiation vessel t o prevent sticking of tracer amounts of inorganic halogen on t h e walls. T h e liquid was then extracted with aqueous sulfite-halide solution. Counting was done by placing the liquid in an annular jacket designed to slide over a type 90NB Amperex Geiger tube. All counts a e r e corrected for the composition of the medium,13 t h e relative counting efficiency of t h e jacket, the volumes of t h e solutions, and for radioactive dereaction, the cay. ~ ~ I n the studies of t h e Br79(n,-y)Br80m aqueous and organic layers were allowed t o stand 2 t o 3 hours following separation, before counting, t o allow the Br@Jni (4.4 hr.) t o re-establish equilibrium with t h e Bra0 (18 min.) daughter which emits t h e radiation which is counted. When BrEomwas to be determined following irradiation of CCloBr, or of alkyl bromides scavenged with 12, adequate time was allowed for the C138 (37 min.) or II28 (25 min.) activity to decay. Two or more experiments were run for each condition tested and the results given in the tables are the averages of the values obtained. T h e organic yields3 of individual deExperimental terminations regularly agreed within &0.5yc of the average Reagents.-Eastman white label CzHsBr was purified by for t h e (n,?) studies and 3 ~ 1 %for t h e isomeric transition mechanical stirring with successive portions of concentrated studies. Isomeric Transition Experiments.-For t h e isomeric &SO4 for 12 hours each until further portions showed no coloration. It was then washed with water, dried with transition experiments, the Brz or organic compound conMgSOd and fractionally distilled. Phillips “pure” grade taining the BrXamwas added t o 25 ml. of t h e hydrocarbon t o pentane, hexane, heptane and nonane were used. T h e be tested in a glass-stoppered flask and allowed t o stand for first three of these were treated with HsSO, in the same 2-3 hours. Extraction was carried out as described above. way as t h e ClHbBr following which they were stirred with The time of separation was taken as t h e time when the Brz for 12 hours before dry- color had just disappeared from t h e organic layer. When 0.5 N KMnO4 solution 6 N in ing and distillation. The nonane and decane (Matheson) the source of BrsOm was Brz, the activity in the organic layer were fractionally distilled and then subjected to the H z S 0 4 decayed n i t h a n 18 min. half-life. T h e aqueous layer was treatment, washing and drying. Following purification all allowed t o stand until parent-daughter equilibrium was of the hydrocarbons were tested by adding 10-3 cc. of ele- reached, before counting to determine t h e total 4.4 hr. acmental bromine labeled with Bra* (36 hr.) t o 25 cc. of the tivity. n‘hen the Br*Om was in a n organic compound the 18 liquid. T h e solution was allowed to stand in the dark for min. activity mas counted in the aqueous phase and the 4.4 about two hours, following which the BrPwas extracted with hr. in the organic. I n the latter case the organic 5ield was aqueous sulfite and the ratiionctivity in the aqueous and or- given by one minus the ratio of the aqueous 18 min. activity to the organic 4.4 hr. activity, both activities being corganic layers was determined by counting. The fraction of the bromine which entered organic combination was always rected to the time of extraction. Ip a s Scavenger for (n,?) Reactions on Bromides.-Brp less tlian 0.2’7c. Hydrocarbons purified only by passing has been d e m o n ~ t r a t e d ~to~ be , ‘ ~a ponerful tool as a scaventhrough a column of silica gel shoivcd a high bromine pick-up ger for thermal atoms and radicals produced by the ( n , ~ ) in similar tests. Michigan Cliemical Co. CCI3Br was freed from stabilizer by distillation, illuminated while con- process in alkyl bromides and 1 2 has been shown15t o serve taining dissolved Brz, extracted with aqueous sulfite, washed similarly for alkyl iodides. In t h e present work Iz was uqed as a scavenger for the (n,?) reaction on solutions of C2H5Br with water and dried. in various hydrocarbons in order t o minimize the problems Preparation of Organic Compounds Containing Brwm.-CzHaBrsam and CHBrZBr8Om were prepared by allowing of the reactivity of Brz Kith hydrocarbons in the presence vapors of the non-radioactive compounds t o stand for a few of light and ?-radiation. T h e data of Table I show t h a t its minutes in a flask coated with freshly sublimed illBra, which scavenger effect on bromides is equivalent to t h a t reported had been prepared by adding radioactive Brs1*t o dry alumi- earliePbJ4 for Brz. num, using vacuum system techniques. CCI3Br8Om was TABLE I prepared by illuminating a mixture of CC13Br and radiobromine. ORGANICYIELDS OF THE (n,r) PROCESS IN C2H6Br AND I N Neutron Irradiations .-All neutron irradiations for studies CCllBr AS A FUNCTION OF CONCENTRATION OF IODINE of the (n,r) process were made with a n Sb-Be photoneutron < CpHiBr-CChBr source1*which contained about 10 curies of Sblz7 and emitted 12, Av org yield, 12, Av. orir yield, mole fraction 76 mole fraction “D about 107 neutrons/sec. when received. T h e liquids to be irradiated were contained either in soft glass test-tubes set 39.0 0.000 32.0 0.000 in a paraffin block, or in the annular space between the walls 37.5 .001 31.3 ,001 of a vessel shaped like a wide mouthed dewar, which was 33.5 .002 28.4 .002 surrounded with water. During irradiations the source was positioned in the center of the paraffin block or of t h e dewar31.5 .005 27.1 ,004 like vessel. T h e 7-ray dosage received by t h e samples was 29.5 ,0066 ,010 26.7 in t h e range of about los to 10‘ r./hr. Irradiations for the ,015 2 4 . 5 production of Brmm (4.4 hr.) were of 8-12 hours duration .020 23.0 and those for the production of 1 1 2 8 (25 min.) lasted 1-2 hr. T h e samples irradiated a t temperatures above 25’ were conResults and Discussion tained in sealed tubes in a bath of thermostated paraffin oil.
+
+
I
(6) A. F. Reid, P h y s . Rep., 69, 530 (1946). (7) L. Friedman and W. F. Libby, J . Chcm. P h y s . , 17, 647 (1949). ( 8 ) J. C. W. Chien a n d J. E. Willard, THISJOCRXAI,, 75, 61GO (1953). (9) J. M. hliller and R . W. Dodson, J . Chenz. P h y s . , 16, 865 (1050). (10) (a) J. 17. Hornig, G. Levey and J. E. Tt’illnrtl, ibid , 20, 1556 (1952); (b) C. Levey and J . E. Willard, ibid., 2 5 , 90-1- (1956). (11) Made by irradiation of ampoules of liquid bromine in t h e C P 5 nuclear reactor of the Argonne S a t i o n a l Laboratory. (12) T h e source was obtained from the Oak Ridge National Laboratory (Radioisotopes Catalogue page 141 (1952)) and irradiated a t the Brookhaven National Laboratory.
Effect of Dilution of Alkyl Halide with Hydrocarbon.-The data of Table I1 show that, consistent with results reDorted earlier for CIH6Br and C2H61in hexane,3bthe organic yield for theBr79(n,y)BrRam process on bromine in the form of C2H5Br is increased by dilution with either pentane or heptane, a limiting value characteristic of the pure hy(13) R. S. H. Chiang a n d J. E. Willard, Science, 112, 81 (1950). (14) J. F. Hornig and J. E. Willard, TEISJOURNAL, 75, 461 (1953). (15) G. Levcy and J. E. Willard, ibid.. 74, 6161 (1852).
HALOGEN ATOMREACTIONS INITIATED
July 5 , 1957
BY
NUCLEAR PROCESSES
3360
drocarbon apparently being reached a t concentrations between 0.1 and 0.01 mole fraction. The increase over the value in pure C2HSBrtoward this limiting value is qualitatively proportional to the mole fraction of hydrocarbon. The increase is predominantly, and possibly wholly, the result of a type of chemical event which occurs to the tagged halogen atoms before they have had an opportunity to diffuse as thermal atoms in the system; that is it is the result of a “ h ~ t ” ~ process. ,~b This is illustrated by the fact that the difference in organic yields between pure C2H61and hydrocarbon s o h tions containing 0.05 mole fraction C2HJ is, within the experimental error, independent of the concentration of I2scavenger present (Fig. 1). TABLE I1 ORGANICYIELDSOF THE Br7g(n,r)Br*@’ PROCESS IN SOLUTIONS OF CzHSBrI N %-PENTANE AND HEPTANE AT DIFFEREST CONCESTRATIOSS CzHaBr, mole fraction
Av. organic yield, % I n pentane I n heptane
1.00 0.80 .75 .66 .50 .20 -10 .02 .01
32.0
32.0 37.3 37.7 40.0 42.0 44.8 46.0 47.1 47.4
.. ..
37.0 37.5 38.0 40.4 41.0 42.2
Effects of Chain Length of Hydrocarbon and of Temperature.-From Table I11 and Fig. 1 i t may be seen that the organic yields of Ilz8from the 112’ (n,y)1128 process in solutions of 0.05 mole fraction TABLE I11 ORGAXIC YIELDSOF THE (n,y) PROCESS ON ETHYL HALIDES I X HYDROCARBOXS IN THE PRESENCE OF IODINE SCAVENGER 12,
mole fraction
0.000 .001 .00 13 ,0019 .002 ,004 ,005
Pentane
Organic yield in Heptane
Yields from C&I,” % 45.0 47.0 .. 44.2 39.0 .. 37.0 37.8 41.6 35.0 .. 38.0
.. ..
Yields from CzHsBrb 0.000 40.4 46.0 .. 34.1 ,007 .. 33.3 ,0015 The mole fraction of CzHJ was 0.05. tion of CIHLBrwas 0.10.
Decane
52.5 48.2
.. ..
45.6
..
..
49.0 39.0 37.8 The mole frac-
( I*),
mole frocrlon.
Fig. 1.-Effect of iodine scavenger on the organic yields of the Br79(n,r)Bra and I1W(n,y)I1% processes in hydrocarbon solutions as a function of solvent and temperature. Concentration of CzHaBr for upper two curves 0.10 mole fraction; concentration of C&&I for lower three groups of curves 0.05 mole fraction. Each division on the scales a t the right represents 2% in organic yield. The scale for each group of curves is labeled with a single percentage value opposite the center of the curves to which it applies.
of C2HJ in n-pentane, n-heptane and n-decane, and atom. The most probable cause of such an effect of BrS0from the Br79(n,y)Br8a process on CzH6Br seems to be a difference in the effectiveness of the a t 0.1 mole fraction in n-heptane and n-decane, in- solvent cages for retaining the tagged atom in concrease with increasing chain length of the hydrocarbon. Because the difference persists even in the presence of sufficient scavenger to nearly eliminate encounters of the thermalized tagged atom with radicals by diffusion, it must be concluded that the chain length of the hydrocarbon molecules affects the probability that the tagged atom will enter organic combination before diffusion as a thermalized
tact with an organic radical it has just formed long enough for a stable compound to be formed. Because the effectiveness of the solvent cages might be expected to decrease with increasing temperature, and also because the higher solubility of iodine a t higher temperatures makes it possible to use higher concentrations of iodine scavenger, determinations were made of the organic yields of the
3370
SUKUMAR ADITYAAND JOHN E. WILLARD
VOl. 79
(n,y) reaction on ethyl iodide in decane and in heptane a t five temperatures and a t a series of iodine concentrations. The results, given in Table IV and Fig. 1 indicate a decrease by about one fifth in organic yield for a 100' increase in teniperature a t all iodine concentrations in both solvents. The fact that the temperature coefficient is observed in the presence of 0.01 mole fraction 1 2 scavenger indicates that the change in temperature alters the probability of some process by which the enters organic combination before it has undergone many collisions subsequent to being moderated to thermal energies. As in the case of the effect of chain length it seems most plausible to ascribe this to a change in the effectiveness of solvent cages for preventing the I'28atom from escaping from a radical which it has just formed.
Consideration of Mechanisms.-The "hot" reactions observed as described above must involve one of the following mechanisms: (a) combination of a neutral recoil atom with an organic radical it had just formed, and with which it combines either before escaping from their common solvent cage or after only a few diffusive encounters with solvent molecules; (b) attack of an energetic neutral recoil atom on a solvent molecule in a displacement reaction in which it replaces a radical or atom; (c) formation of a molecule-ion by reaction of a recoil halogen ion with a solvent molecule, followed by neutralization or reaction of the molecule-ion to yield a stable organic halide molecule. I t is impossible, a t present, to distinguish between these mechanisms for liquid systems. Gas phase studies have shown, howwhen activated as a reever, that IlZ8,BrsOand sult of formation by the (n,y) process can all enter TABLE IV organic combination by displacement reactions beORGANICYIELDSOF THE (n,r ) PROCESS ON CSHbI' IN tween the halogen atom or ion and hydrocarbon HEPTANE AND DECANEAS A FUNCTION OF TEMPERATURE molecules, under conditions where the possibility of IS, reaction of the tagged species with free radicals is mole Organic yield in % a t fraction 00 350 550 750 1000 eliminated.8~10~22 There is evidence that the atom Heptane 0.000 5 0 . 0 4 6 . 5 4 5 . 0 41.0 must have a charge for this type of reaction to oc.001 46.0 .. . . 37.5 cur.10b I t is probable, therefore, that reactions of 44.2 40.2 3 7 . 5 3 5 . 5 ,002 ions contribute to the "hot" portions of the reac,005 . . 3 8 . 0 34.0 33.0 tions discussed in this paper and to many other ,010 . . 3 5 . 0 3 3 . 4 31.0 "hot" reactions initiated by nuclear transformations Decane .000 5 5 . 0 51.5 5 0 . 0 46.5 45.5 in both the gas and condensed phases. It is reason.OOl . . 42.6 . . 5 1 . 4 48.2 able to assume that these ionic reactions are related * 002 50.0 4 5 . 4 44.8 40.5 39.0 in type to the highly efficient ion-molecule reac.005 .. 42.0 4 0 . 5 3 7 . 8 3 6 . 5 t i o n ~currently ~~ under intensive study by mass .010 .. 40.0 38.5 36.0 3 4 . 4 spectrometric techniques. It seems probable that the formation of carbon chains by recoil tritium The mole fraction of C2HJ was 0.05. atoms in methane,6a and the other highly efficient The caging effect conceptlBused above has been reactions of tritium atoms activated by recoil and quantitatively demonstrated17-19 for the photo- by tritium decay5b are due to ionic reactions of the chemical dissociation of I 2 in solution. Lampe and type revealed by the mass spectrometric studies. The organic yield of 5oy0 observed for the gas Noyesla have shown that the primary quantum yield for this dissociation changes with solvent from phase reactions of methane with iodine activated 66% for hexane to 14y0 for carbon tetrachloride to by the Il2'(n,y)IlZ8 process is drastically lowered by 4% for hexachlorobutadiene a t 25O, the trend being a few mole per cent. of C2HJ and the yield in pure o the ionizafor the dissociation yield to decrease with increase C2HJ gas is only about l ~ o . lBecause in molecular weight of the solvent. Both the pri- tion potential of the C2HsI (9.5 e.v.) is lower than mary quantum yield,'* and the rate constant for that of the I atom (10.4 e.v.) the effect of ClHJ in the atom recombination reactionz0have been shown lowering the yield with CH4 has been ascribed to to increase with increase in temperature. neutralization of the ions by C2HJ before The only report in the literature on the effect of their kinetic energy has been sufficiently reduced to temperature on the organic yields of the ( n , r ) proc- allow them to form a stable combination. It is ess in pure liquid ethyl halides,21of which we are possible that similar factors may play some role in aware gives values of 37.0 and 3S.t3yO for CzHJ a t the reactions of mixtures of liquid alkyl halides and -78 and 20' and of 34.5 and 31.8% for C2HbBra t hydrocarbons with halogens activated by the (n,r) - 78 and 0" indicating less temperature sensitivity processes. Major differences from the gas phase than that shown in the present work for hydrocar- results might be expected, however, due to caging bon solutions a t higher temperatures. Limited effects, differences in charge transfer potentials, and data3" indicate a small negative temperature coef- differences in the nature and stability of the moleficient for the organic yield of Brso from the (n,y) cule ions formed. Among the experimentally oband isomeric transition processes in CC&Br and served differences are the low organic yields of little or no temperature effect for CIS*in CC14. in gaseous CsH51(lye) lo as compared to liquid Ca(16) J. Franck and E. Rabinowitch, Trans. Faraday Soc., 30, 120 (1934). (17) R. Marshal and N. Davidson, J . Chcm. P h y s . , 21; 2086 (1953). (IS) F. W. Lampe and R. M. Noyes, THIS JOURNAL, 76, 2140 (1954). (19) R. L. Strong and J. E. Willard, ibid., 79, 2098 (1957). ( 2 0 ) S. Aditya and J. E. Willard, ibid., 79, 2680 (1957). (21) J. C . Roy, R. R. Williams, Jr., and W. H. Hamill, ibid., 7 6 ,
3278
(1954).
(22) A. Gordus and J . E. Willard, ibid., in press. (23) (a) H. Eyring, J. 0. Hirschfelder and H. S. Taylor, J . Chcm. Phys., 4, 479 (1936); (b) V. L. Tal'roze and A. K.Lgubimova, Dohlady A k a d . Nauk S . S . S . R . , 66, 909 (1952) (C. A , . 47, 2590 (1953)); (c) D. P. Stevenson and D. 0. Schissler, J . Chcm. Phyr., 23, 1353 (1955); (d) D. 0. Schissler and D. P. Stevenson. ibid., 24, 926 (1958); (e) 0.G. Melsels, W. H. Hamill and R. R. Williams, Jr., ibid., 26, 790 (1956).
July 5, 1957
HALOGEN ATOMREACTIONS INITIATED BY NUCLEAR PROCESSES
Hd(42%),24 and in gaseous C2He(3%>, C~H.4473 and C1Hlo(5%)~' as compared to the liquid hydrocarbons tested in the work of this paper (Table 111). Likewise the organic yield of reacting with CH4 in the gas phase is much more sensitive to added CzHbI (ie., 5 mole % CzHSI reduces the yield in gaseous CHq by more than 50%)lDbthan are the yields from the hydrocarbons of this paper in the liquid phase. I n all probability these differences are due in part t o the fact that non-ionic reactions in which the recoil atom can combine with a radical it has formed are made possible by the caging effect and greater probability of diffusive recombination in the liquid. Correlation with Reactions of C18a.-Unlike the reactions of and B P , the reactions of Clas activated by the (n, y ) process in liquid hydrogenous media give a constant organic yield (215%) which is independent of whether the solvent is an alkyl chloride or a hydrocarbon or a mixture of the two, and is independent of the chain length.a This unexpected independence of the yield on environment has had no satisfactory explanation. Consistent with the discussion in the preceding paragraphs i t may now be explained tentatively by the hypothesis that only ionic reactions lead to organically bound Cls*in liquid hydrogenous media, that 21% of the C137(n,y)C138 events occur with internal conversion which produces ions,26and that all such ions result in organically bound chlorine. Chlorine atoms, unlike bromine and iodine atoms abstract H from C-H bonds with very low activation energy. I n media where such bonds are available, every neutral chlorine atom would be expected to be fixed in inorganic combination as HCI. BF0 and IlZ8atoms, which do not abstract hydrogen as readily, have a greater opportunity to react with radicals which they form in losing the energy from the (n,-y) process. Consequently they may be expected to show organic yields which vary with the varying fragmentation and caging characteristics of different media. Similar variations are to be expected for CI3*in non-hydrogenous solvents where the activation energy for reaction of chlorine atoms with the solvent is high. This is illustrated by the relatively high organic yield (44% total, ca. 32% "hot") and scavenger effect observed* with carbon tetrachloride in contrast to the 21% total organic yield with no scavenger effect for hydrogenous solvents. Reaction of Hydrocarbons with Bromine Activated by the Brm Isomeric Transition.-The data of Table V show that the organic yields of bromine activated by the isomeric transition process in hydrocarbons increase with increasing chain length of hydrocarbon regardless of scavenger concentration, as in the case of (n,y) activation. Although the two nuclear processes are very different both can contribute both charge and recoil energy to the atoms involved. These data add further to the growing that the chemical events (24) G. Levey and J. E. Willard, THIS JOURNAL, 74, 6161 (1952). (25) S. Wexler and T. H. Davies, J . Ckcm. Pkys., 20, 1688 (1952). have shown that at least 12% of B r m , 18% of Brm, 25% of Brs: and 50% of 11% formed by the (n,r) process are positively charged as a result of internal conversion during stabilization of the compound nucleus following neutron capture. (26) G. Levey and J. E. Willard, THISJOURNAL, 78, 2351 (1956). (27) J . B. Evans and J. B. Willard, ibid., 78, 2908 (1956).
3371
induced by the two types of activation are largely identical. TABLE V ORGANIC YIELDSOF THE Brsom + B F ISOMERIC TRANSITION IN SOLUTIONS OF BROMINEIN HYDROCARBONS Brz, Organic yield, 70,in mole fraction
Pentane
Hexane
Heptane
9 x 10-6 1 X 10-6 9 x 10-6 0.0001
39.5
..
..
.0009 ,001 .005
33.0
..
.01
.02
.. 39.0
.. .. .. ..
Nonane
..
..
Decane
..
..
.. ..
41.0
42.3
45.5
..
..
..
37.4
39.2 35.0
.. ..
40.5
..
..
42.0 40.5
27.4
30.8
42.5
.. ..
..
..
..
.. .. 49.1
..
Table VI shows further that the organic yields from the isomeric transition of bromine in hydrocarbons vary with the compound which is used as the source of bromine even when its mole fraction is The differences between difas low as 1 X ferent source compounds are, to a first approximaTABLE VI TRANSITION ORGANIC YIELDSOF THE Braom + B+' ISOMERIC IN HYDROCARBONS WITH DIFFERENTCOMPOUNDS A S THE SOURCE OF BROMINE Organic yield, in %, with CHBrP
Pentane Heptane
43.0 51.0 40.5b 54.0
CClaBra
..
55.0 46.Ob 60.0
C2HsBrm
44.5 52.0
*.
Brra
39.5 42.5 35.0b 49.0
Decane 55.0 0 The concentration of the bromine containing compound was 1 X 10-6 mole fraction in each case except for the second Reaction solution contained row of values for heptane. mole fraction of Brl. 5X
tion, independent of the hydrocarbon solvent and of the concentration of bromine scavenger. It appears that they are the result of partial failure of the CBr bond to rupture following isomeric transition, as observed earlier2*for CC13Br,CRHbBr and CH3Br in liquid bromine as the solvent. The differences between the organic yields from CClaBr and from Br2 observed for the solvents of Table VI are 12.5, 11.0 and l l . O % , in striking agreement with the observed for bond rupture failure of 10-13%z2~28 CC13Br in the previous work; the differences between C2H5Br and Brz of 5, 9.5 and 6% are similar to the failure value of 6.5% for CH3BrZ8;and the differences between CHzBrz and Brz of 3.4, 8.5, 5.5 and 5.0% correspond to the failure value of 4.3%.22 If the interpretation here is correct, it is of particular interest that the retention, in the parent molecule, of bromine atoms which have undergone isomeric transition, is so similar in hydrocarbon solvents and in liquid bromine. Acknowledgment.-This work was supported in part by the United States Atomic Energy Commission and in part by the University Research Committee with funds made available by the Wisconsin Alumni Research Foundation. MADISON,WISCONSIN (28) R. S. H. Chiang and J. E. Willard, ibid., 74, 6213 (1952).
