Proton Transfer Reactions Occurring in Gas-PhaseRadioly sisBY P. AUSLOOS AND (MRs.) S . G. LIASNational Bureau of Standards, Washington, D.C. 20234Received 4th January, 1965From the products formed in the gas-phase radiolysis of various reaction mixtures containingdeuterium labelled compounds, it is inferred that: (a) H:, ArH+, KrH+, and probably XeH+transfer a proton to n-pentane. In all cases, the protonated pentane ion decomposes to methane,ethane, propane, and the corresponding butyl, propyl, and ethyl ions. The relative probabilityof these three fragmentation processes does not vary with the nature of the proton donor, withinexperimental error. (b) H i , ArHf, KrH+ and CHS effectively transfer a proton to cyclopropane.On the basis of isotopic labelling experiments, it is deduced that the protonated cyclopropane re-arranges to the sec-propyl ion structure prior to or during reaction.Similarly, the protonatedcyclobutane rearranges to the sec-butyl ion structure. (c) H:, ArH+ and KrH+ transfer their protonto ethylene, propylene and butene, to form mainly ethyl, sec-propyl, sec-butyl ions, respectively.Carbonium ions such as C2HS can also transfer a proton to olefins. In the latter case, however,alternative modes of reaction such as addition and hydride transfer reactions occur as well.Gas-phase proton transfer reactions have been reported to occur in the massspectrometer, but they have seldom been observed at atmospheric pressures underthe action of ionizing radiation. The major reason for this has been that thepresence of ionic reactants or intermediates in such systems can only be inferredfrom a product analysis after irradiation.However, in recent radiolytic studieswhich were carried out in the presence and in the absence of an electrical field,and in which extensive use was made of deuterium-labelled compounds, a distinctioncould be made between products of ion-molecule reactions and those of neutralexcited molecule decompositions.1 It has also been demonstrated that it is feasibleto trace the course of ionic reactions in a radiolytic system in much the same wayas free radical reactions are studied in photolytic or pyrolytic systems. Becauseone sees only the neutral product of the ion-molecule reaction in radiolysis, theinformation obtained from the radiolytic system is complementary to that gainedfrom mass spectrometry where only the charged species are observed.Conventionalanalysis of the neutral products formed in the radiolysis of suitable deuterium-labelled compounds has the advantage that it is possible to establish, from thepositioning of the deuterium atoms in any particular product, the structure of thereacting ion or of the intermediate complex. These techniques were applied instudies 29 3 on proton-transfer reactions occurring in the gas-phase radiolysis ofhydrocarbons. The results presented here are an extension of that work.EXPERIMENTALMATERIALSThe deuterated compounds were obtained from Merck, Sharp and Dohme of Canada,Limited. All compounds were purified by means of gas chromatography.Mass-spectro-metric analysis indicated that n-pentane-dl;? contained 10 % CSDllH ; CyClOprOpa.ne-d63P . AUSLOOS AND S . G . LIAS 37contained 7-4 % C~DSH; propylene-d6 contained 4.3 % C3DsH; ethylene44 contained3.4 % C2D3H; propane-d8 contained 3 % C3D7H; and isobutaned10 contained 8 %C4DgH. Phillips reagent-grade CH4 was purified by repeated slow distillation from - 195to -220°C. Assayed reagent-grade hydrogen, xenon, krypton, and argon were obtainedfrom the Air Reduction Company. The deuterium gas, obtained from General DynamicsCorporation, contained 0.5 % HD.IRRADIATION AND ANALYSISPyrex reaction vessels of 500 ml provided with breakseals were used in all experiments.Prior to being filled, the cells were heated under vacuum close to the m.p.of Pyrex. Thecells were irradiated at 40f5"C in the National Bureau of Standards 50,000 Curie 6OCosource. Dosimetry was based on the measurement of the saturation current in a speciallyconstructed reaction vessel.1 Taking a W value for H2 of 36.3 eV, the energy absorbed byhydrogen is 1 . 3 5 ~ 1018 eV/mole sec.After irradiation, quantitative analysis of the products was carried out by expandingan aliquot of the irradiated material into an F and M fractometer provided with a silicagel or an alumina column, a flame ionization detector, and temperature programming.Subsequently, hydrogen was removed through a spiral trap immersed in liquid hydrogenas a refrigerant. Methane was then distilled and the remainder of the sample was intro-duced on to a Perkin-Elmer vapour fractometer (silica gel column) from which the productcompounds were collected separately from the helium stream at the exit of the instrument.All fractions were introduced into a Consolidated 21-101 mass spectrometer in order todetermine their isotopic compositions.RESULTS AND DISCUSSIONOnly a summary of the experimental observations is presented in this paper.Many of the results quoted are derived from detailed product analyses not givenin their entirety here.Tables containing the isotopic compositions of the differentproducts and the derivations of the ion pair yields and relative reaction rates dis-cussed can be obtained from the authors upon request.Nitric oxide was added as a free radical scavenger in most radiolysis experiments.Although removal of free radicals by NO usually simplifies the interpretation ofthe radiolysis mechanism, NO may also, to some extent, interact with some of theions present in the system.In several cases, therefore, the concentration of NOwas varied over a certain range in order to assess the effect of NO on the ionicreaction mechanism.PROTON TRANSFER TO n-PENTANEWhen hydrogen is irradiated in the presence of a small amount of alkane, RH,the following reaction mechanism will occur2 :Hz+H2+Hz+HH3f +RH-+H,+RHzThe protonated hydrocarbon thus formed will eliminate H2 or a smaller alkane,and the corresponding alkyl ion. For example, it was inferred that protonatedn-pentane-dlz decomposes as follows ;C5D12H'-+CD,H+sec-C4D: (3)C5D, 2H+-+C2D5H + sec-C,D; (4)C5Dl,H+-+C3D7H + C,Df ( 5 38 GAS-PHASE RADIOLYSISThe fragment carbonium ions (CnDin+1) formed in reactions (3), (4) and (5) reactwith n-pentane in a hydride transfer reaction :forming the corresponding fully-deuterated hydrocarbons as products (table 1).TABLE 1.-PROTON TRANSFER TO n-PENTANE-& ; MOLECULES FORMED PER ION PAIRsystemn-C~D12+ NO(1 : 0.05)H2+n-CsD12+NO(300 : 1 : 0.5)Ar+ H2+ n-C5D12+ NO(1100:400: 1:0.6)Kr+H2+n-C5D12+NO(500 : 300: 1 : 0.5)Xe+ H2 + n-C5D12 + NO(300 : 300 : 1 : 0.5)H2+ Xe+ n-C5D12+ NO(300 : 1 : 1 : 0.5) *H2+Xe+n-C5D12+NO(300 : 10 : 1 : 0.5) *0.56 0.440.34 0-29n.d.0.29n.d. 0.0990.60 0.480.77 0.67(0.63)0.14 0.140.050 0.0580.053 0.058- 0.390.1 1 0.130.12 0.40(0.13)0-069 0.0650.029 0.0240.029 0.026- 0.120.054 0.0650.065 0.15(0.062)* Values given are molecules formed per H i initially produced.Values in parentheses are valueswhich may be attributed to proton tansfer reactions after correction for contributions from charge-transfer radiolysis. n.d., not determined; -, not formed.No protonated alkanes except CH3 and C2Hj have beed observed in the massspectrometer, indicating that the higher homologues have a relatively short lifetime.This is corroborated with the protonated pentane by the fact that, even at a pressureof 25 cm, the ion pair yields ascribed to processes (3), (4) and (5) (0.56, 0.14, and0.07, respectively) account for 0.77 of the HZ ions formed.This is reasonably closeto unity if one considers that other modes of decomposition may still be unaccountedfor. For example, the decomposition of CsD12H+ to give hydrogen and a pentylion would be difficult to detect by the analytical methods used in this study.It is of interest to determine whether protonated pentane could be producedusing other proton donors such as ArH+, KrH+ or XeH+, and whether the relativeprobabilities of the modes of decomposition would vary with the AH of the reaction.Therefore, several H2 +inert gas + C5D12 + NO mixtures were irradiated, in whichthe ratio inert gaslhydrogen was kept sufficiently large so that more than 95 %of the energy was absorbed by the inert gas. With argon and krypton,Kr++Hz+KrH++H (7)Ar++Hz-+ArH++H (8)are the major modes of reaction of Ar+ and Kr+ in such a system.4 Charge transferfrom krypton to hydrogen cannot occur because the krypton ionization potentialis lower than that of hydrogen, while mass spectrometric studies5 have shown thatcharge transfer from argon to hydrogen should be of minor importance comparedto reaction (8).If KrH+ or ArH+ formed in these mixtures does transfer its proton to n-CsD12,the major lower hydrocarbon products which should be observed would be theproducts resulting from decomposition processes (3), (4) and (3, followed by thP.AUSLOOS AND S . G . LIAS 39hydride transfer reaction (6). The fact (table 1) that yields of CD3H, C2DsH andC3DjH are comparable to those of C4Dl0, C3Ds and CzD6, respectively, showsthat processes (3), (4) and (5) occur in these systems to the exclusion of chargetransfer from the inert gas to pentane, which would lead to the formation of a largeyield of C3Ds and smaller amounts of C2D6 and C4D10, all in the absence of partiallydeuterated products.6 If the product yields are calculated relative to the energyabsorbed by the inert gas (as in table l), the sum of processes (3), (4) and (5) accountfor about 50 % of the Ar+ and Kr+ ions.On the other hand, the product yieldsaccount for more than 10 times the maximum number of H i ions which could beformed in the system by reaction (l), thus demonstrating conclusively that the protondonors were ArH+ and KrH+.In contrast to the results obtained on the krypton and argon mixtures when aXe+H2+CsD12+NO mixture is irradiated with more than 95 % of the energyabsorbed by Xe, the observed product distribution is closely similar to that ob-served in the xenon-sensitized radiolysis of pentanes in the absence of hydrogen,clearly indicating that charge transfer from xenon to pentane is the major processtaking place.On the basis of the product yields, at least 80 % of the Xe+ ions canthus be accounted for.These observations are not unexpected since there is no definite proof in theliterature for the formation of XeH by reaction (9)Xe++H2-+XeH++H. (9)Stevenson and Schissler 7 point out that this reaction has not been observed in themass spectrum of xenon + hydrogen mixtures and hence the cross-section of thisreaction must be less than 50 times that for the analogous formation of ArHf.In an attempt to form XeH+, a mixture of hydrogen and xenon was irradiatedin which the relative concentrations were adjusted so that at least 86 % of theenergy was absorbed by hydrogen.When pentane was added to such a mixturein concentration equal to that of xenon (table 1, 6th experiment), the product dis-tribution and yields were nearly identical to those observed in the radiolysis ofH2 + C5D12 +NO mixtures in the absence of xenon. Because, according to Thompsonand Schaeffer,ss 9 reaction (1 0),H l +Xe+XeH++H, (10)occurs with a rate comparable to the collision frequency, one would expect that aproton is transferred to xenon from H'j at least as rapidly as to C5D12, thus, leadingto diminution in product yields if XeH+ does not, in turn, transfer its proton topentane, but rather undergoes neutralization.10 Such a drop in yield is not ob-served even when the concentration of xenon is increased to 10 times that of pentane,thus indicating that proton transfer from XeH+ to pentane may be efficient.The relative probabilities of the different modes of decomposition of the pro-tonated pentane ion formed by proton transfer from H;, ArH+, KrH+, XeH+ and,as recently reported,ll CHS, do not change although the AH of the reaction differswith the different proton donors. Although a variation in the ion distribution withthe AH of the proton transfer process has been observed in the mass spectrometer,l*these changes are apparently not due to a change in the relative probability of theinitial fragmentations, but rather to a lower or higher degree of decomposition ofthe fragment carbonium ion.No evidence for such fragmentation is seen at themuch higher pressures at which the radiolytic experiments were carried out,indicating that these ions are apparently stabilized because of the higher collisionfrequency40 GAS-PHASE RADIOLYSISPROTON TRANSFER TO CYCLOALKANESWhen a proton is transferred to a cycloalkane, the resulting protonated entity,C,H;,+ 1, is isomeric to a carbonium ion and, therefore, in contrast to the protonatedalkanes such as C5D12H+ discussed above, can be expected to have some stability.That is, the protonated cycloalkane may subsist for a sufficiently long time causingit to react as a carbonium ion instead of decomposing, in which case the structureof the ion can be determined.Proton transfer to cyclopropane was studied indetail. However, exploratory experiments indicate that analogous processes occurwith other cycloalkanes such as methylcyclopropane and cyclobutane.That fragmentation of the protonated cyclopropane formed in reaction (1 1)is not an important process is shown in the radiolysis of H ~ + c - C ~ H ~ mixtureswhere the ion pair yields of the products up to C3 account for only about 12 %of the Hf in the system. The major products up to C3 are methane and ethylenewhich are formed in equal amounts with ion pair yields of about 0.12 at an H2/C-CfH6ratio of 630.These yields vary little with the amount of NO added to the reactionmixture or a change of the ratio H~/c-C~H~.A plausible mechanism for the formation of methane would be reaction (11)followed byThe vinyl ion, which would be formed in this decomposition, may react with cyclo-C3H; -+CH,+ C2H:. (12)propane,C2Hz + c-C~H~+C~H: + CzH4,to form ethylene. This would account for the fact that, in the scavenged experi-ments, the ethylene yield is approximately equal to that of methane. The factthat the ethylene fraction observed in the radiolysis of H2 + c - C ~ H ~ + c-CJD~ mixturesconsists entirely of C2H4 and C2D4 is consistent with the suggestion that theethylene is formed as a result of a reaction such as (13).Because only a fraction of the C3H; ions formed in reaction (11) apparentlydecompose, it is of interest to determine the modes of reaction, as well as the structureof the stabilized entity, C3H3.A derivation of this structure is of particular interestin view of the suggestion of Meyerson et aZ.139 14 that C3H3 ions, originating fromthe fragmentation of alkane ions in the mass spectrometer, have a protonated cyclo-propane ring structure. On the other hand, Stevenson 15 has postulated that theseions acquire the sec-propyl ion structure.Since propane is a minor product (ion pair yield <0.007) in the scavenged H2-cyclopropane radiolysis, it is evident that the protonated cyclopropane producedin reaction (1 1) does not undergo a hydride transfer reaction,C3H; +C-C3H6-*C3Hg+C3H: (14)with cyclopropane.If the protonated cyclopropane does acquire the sec-propylion structure, this observation would be expected since it was demonstrated in oneexperiment that sec-propyl ions, produced in the radiolysis of isobutane-dlo,l6 didnot react with additive cyclopropane to produce propane-&,but reacted entirely with the isobutane to form propane-dg,sec-C3DT +iso-C4D10+C3D8 + C4D,f. (16P. AUSLOOS AND S. G. LIAS 41However, from these results, it could be inferred that sec-propyl ions react withc-C& at a rate which is 0.64 that of reaction (16), to form a more complex ion whichis probably removed from the system by NO.Thus sec-propyl ions, if produced, cannot be detected in the hydrogen-cyclo-propane system unless some compound is added to the system with which thepropyl ion undergoes a hydride-transfer reaction.Cyclohexane was chosen forthis purpose, not only because sec-propyl ions react readily with cyclohexane 17but also because an experiment in which cyclohexane was added to hydrogen in theabsence of any other compound showed that proton transfer to cyclohexane producesessentially no product compounds below c.6.It is seen (table 2) that, when c-C6D12 is added in various concentrations toH2 + C-C3H6 + NO mixtures, propane becomes the major product. More than90 % of the propanes formed consist of CH3CHDCH3, thus demonstrating thatreaction (1 7) occurs and that protonated cyclopropane rearranges to the sec-propylion structure prior to or during reaction.If C3H3, formed in a process such as(ll), retained its ring structure, then, on statistical grounds, it would be expectedthat CH2DCH2CH3 should be a major product. The above observations are alsosupported by the fact that, in the radiolysis of D2 + ~C3D6-t- C-C~H~Z. mixtures,propane consists entirely of CD3CDHCD3.From the observation that the propane-& formed in the radiolysis of H2+C-C3D6 + C-CgD12 mixtures contained approximately 80 % CH2HCD2CD3, it may beconcluded that the protonated c - C ~ D ~ isomerizes mainly to the structure CD2HCDC3which is statistically favoured over the alternate configuration CD3CHCD;.From earlier results 17 and those cited in this discussion, it is calculated thatthe rate of propyl ion addition to cyclopropane is 0.42 times as fast as reaction(17).Knowing this, it is possible to calculate, from the ion pair yields ofsec-C,H; +c-C,D,,+CH,CHDCH, + C,DTl, (17)TABLE 2.-PROTON TRANSFER TO CYCLOPROPANE; MOLECULES FORMED PER ION PAIRCH4 GH4 CHSCHDCH, CH2DCHzCHzD calc. C3H:H2f C-C3H6+ C-CgD12f NO 0.084 0.076 0.22 <0.002 0.65H2+ C-C3H6+ C-CgD12+ NO n.d. 0.047 0.26 <0*002 0.36Ar+ H ~ + c - C ~ H ~ + c-CgD12+ NO 0-032 0.04 0-25 <0*002 0-53KrS H2-t c-C&+ c-C&2+ NO n.d. 0.07 0.17 0.005 0.37Xe+ H2+ c-C3H6+ c-C&2+ NO 0-021 0.17 0.040 0.086 0.086(176 : 1 : 0.2 : 0.1)(195 : 1 : 1 : 0.5)(169 : 69 : 1 : 0.37 : 0.15)(99 : 67 : 1 : 0-37 : 0.15)(65 : 72 : 1 : 0-36 : 0.15)CH3CHDCH3 given in table 2, the total ion pair yield which can be ascribed to propylions.In addition, if one assumes that a proton is transferred with the sameprobability to C-C3H6 as to c-CgD12, noting that process (12) occurs in this systemwith an ion pair yield of 0-08, an ion pair yield of 0.91 is obtained for H;. Thus,it may be concluded that the proton-transfer reaction is highly efficient and that themajority of the protonated cyclopropane ions arrange to the stable sec-propylion structure and react further with either cyclopropane or the additive cyclohexane.A few H2 f C-C3H6 4- C-CsD12 + inert gas mixtures were irradiated for which theratio inert gas/hydrogen was kept sufficiently large so that more than 95 % of thetotal dose was absorbed by the inert gas. Processes (7) and (8) should again b42 GAS-PHASE RADIOLYSISthe major modes of reaction of the Krf or Art ions.It can be seen (table 2) thatCH3CHDCH3 is the major product formed in these systems, indicating that processes(7) and (8) are followed by the proton-transfer reactions :ArH+ 4- C-C,H,+Ar+ C,HT (18)KrH' +c-C3H6+Kr + C3HT (19)C3H6D2 is not a product, clearly showing that charge transfer does not occur inthese systems and direct radiolysis is unimportant, as it has been demonstrated 18that the cyclopropane parent ion, if formed, would undergo the D ;-transfer reactionThe yields of the observed products in these experiments calculated relative to theenergy absorbed by the inert gases account for about 80 % of the Ar+ and 50 %of the Kr+ ions. The proton-transfer reaction is, therefore, a major process under-gone by ArH+ and KrH+ in these mixtures.This finding, as well as the resultsof the pentane experiments discussed above, contradicts the assumption 1 9 s 20 thatthe only fate of ArH+ in the presence of hydrocarbons is the neutralization processArH++e-+Ar+H. (21)Since processes other than (21) occur, the rate constants derived in those studieswill be in error.When xenon is added to the H2 + c - C ~ H ~ + c-C6D12 +NO system, CH3CHDCH3is only a minor product, and CH2DCHDCH3 is formed in appreciable yield,demonstrating that parent cyclopropane ions are formed by the charge-transfermechanism :followed by reaction of C3H: with c-CgD12 (reaction (20)). Thus, again the H+-transfer reaction (9) occurs at a much lower rate than the analogous processes (7)and (8).It may be expected that CHS, which is formed in the radiolysis of methane 11by the processwould effectively transfer a proton to c - C ~ H ~ ,Xe++C-C3H6+Xe+C3Hl (22)CH: + CH4+CHS + CH3 (23)CHf $C-C,H6+CH,+C3HT.(24)This process is exothermic by about 65 kcal if one accepts 21 a value of 234 kcalof AHf(CHf), and a sec-propyl ion structure for C3H3. The formation ofCH3CHDCH3 in the radiolysis of a CH4 + C-C& + c-CgD12 mixture (table 2)confirms the occurrence of process (24). Also, the fact that the value of 0.69, whichcan be calculated for the ion pair yield of sec-C3Hi, agrees closely with the estimatedvalue of 0.68 for the ion pair yield of CHj,ll indicating that reaction (24) occurswith a high efficiency.In the radiolysis of H2+C4H8 (600: 1) mixtures, fragmentation of the C4Hi,which may be expected to be formed by the proton-transfer reactionH: + c - C ~ H ~ + C ~ H ~ +Hz (25)is of minor importance.Ethane and ethylene are the major lower hydrocarbonproducts, but they account for only 10 % of the H3 ions. On the other hand,CH3CH2CHDCH3 is a product in the radiolysis of a H2 + C-C4Hg + c-CgD12 + NP. AUSLOOS AND S. G . LIAS 43(400 : 1 : 0.4 : 0.2) mixture, indicating that a fraction of the C4Hi ions formed inprocess (25) are stabilized and rearrange to the sec-butyl ion structure prior to,or during reaction.PROTON TRANSFER TO OLEFINSProton transfer to an olefin gives a protonated entity which, as with the proton-ated cycloalkanes, is isomeric to a carbonium ion and, in view of the results presentedabove, may be expected to react as such.Unlike the protonated cycloalkane,however, the protonated olefin can assume a carbonium ion structure withoutring opening or rearrangement.Hydrogen or deuterium was irradiated in the presence of small amounts ofethylene, propylene, 1 -butene, cis-2-butene, and trans-2- butene. Only thoseexperiments with propylene additives will be discussed in detail.When hydrogen is irradiated in the presence of a small amount of added propylene,the formation of H3 by reaction (1) shouId be foIlowed byThe only products up to C4 which are observed in such an experiment are methaneand ethylene (ion pair yield = 0.023), thus demonstrating that fragmentation ofthe protonated propylene occurs only to a minor extent.Because propane is notobserved, the C3H'; produced in reaction (25) must undergo a hydride-ion transferreaction with propylene to a limited extent, if at all. On the other hand, in anindependent experiment, propyl ions formed in the radiolysis of isobutane wereobserved to add to propylene, probably to form a heavier ion which can be removedfrom the system by NO. Therefore, in order to measure the yield of the propylions, varying amounts of cyclohexane were added as interceptor to hydrogen+propylene mixtures in a series of experiments. As with the analogous cyclopropaneexperiments discussed above, CH3CHDCH3 is the major product produced in theirradiation of a H2 + CH3CHCH2 + C-CgH12 +NO mixture, thus showing thatreaction (26) is followed by reaction (17) in this system and, therefore, that the pro-tonated propylene, C3H';, has the sec-propyl ion structure prior to, or during re-action.That some rearrangement occurs in the protonated intermediate is shownby the irradiation of H2 + CD3CDCD2 + C - C ~ D I ~ mixtures, where the productpropanes consist of CD~CDZCD~H and CD3CDHCD3 in a ratio of about 3.5 to 1.Knowing that the rate of propyl ion addition to propylene is 1.53 times fasterthan hydride transfer with cyclohexane 17 (reaction (17)), we may deduce from theyields of propane attributed to reaction (17) a total propyl ion yield of about 0.44.Further, assuming that a proton is transferred from H i to cyclohexane at the samerate as to propylene, it is calculated that the average yield of HS accounted for bythe observed products is 0.80, indicating that the proton-transfer reaction to propyleneoccurs with a high efficiency.In similar experiments with ethylene as the added olefin, CH3CH2D appearedas the major product with an ion pair yield of 0.19 when the C ~ H ~ / C - C ~ D I ~ ratiowas 2-0.From relative reaction rates determined in separate experiments (ethylion addition to ethylene occurs with a rate which is 1.2 times as fast as hydridetransfer with cyclohexane), it is calculated that about 98 % of the H+ ions have beenaccounted for, thus demonstrating that proton transfer from HS to ethylene ishighly efficient.Again, in similar experiments with 1 -butene, cis-2-butene, and trans-2-buteneas the olefins added to H2 with c-C6D12 as the butyl ion interceptor, n-butane-&having the structure CH3CHDCH2CH3 appeared as a product in all three cases,CH,CHCH2 + H l -+C,H; + H2 (2644 GAS-PHASE RADIOLYSISthus demonstrating that the protonated butene reacts as a sec-butyl ion.Isobutanewas not a product, indicating that C4HZ does not always rearrange to the thermo-dynamically more stable t-butyl ion.Proton transfer reactions from carbonium ions to larger olefinsCmH,+,+ 1 + CnH2n+CmH2m + CnH,+,+ 1 (27)are usually exothermic, but may occur with low probability because addition tothe double bond, as well as hydride-transfer reactions, may constitute alternativemodes of reaction of the C,Hl,+l ion.From the radiolysis of C3D~fC3H6-NOmixtures, it could be deduced that all three reactions occur. In these mixtures,C3Dg is the source of the C2Dg ions 22 which can either react with C3D8C2D: +C3H6-,C2D5H+C3H5f (30)C2Di + C3H6+C5D5Hz. (31)On the basis of the above mechanism, approximate values of the relative rates ofthe different processes can be obtained from the yields of C2D6, C~DSH, and C2D4and their variations with C3H6 concentration. However, because a fraction ofC2D6 and C2D4 is also formed by reactions 23 other than (28) and (29), more accuratedetermination was based on the isotopic analysis of the ethane and ethylene formedin the radiolysis of C3Dg + C3H8 (1 : 1) mixtures in the presence of various amountsof C3H36 From these results it is calculated that the relative probabilities of reac-tions (28), (29), (30) and (31) are 1.00, 1.07, 0.16 and 0.93, respectively. Radiolyticstudies 3, 17 have shown that CZHf and n-C3Hf can also transfer a proton to otherorganic compounds such as CH30H and CH3COCH3.This research was supported by the U.S. Atomic Energy Commission.1 Ausloos and Gorden, J. Chem. Physics, 1964,41, 1278.2 Ausloos and Lias, J. Chem. Physics, 1964, 40,3599.3 Sandoval and Ausloos, J. Chem. Physics, 1963,38,2454.4 Stevenson and Schissler, J. Chem. Physics, 1955,23, 1353.5 Giese and Maier, J. Chem. Physics, 1961, 35, 1913.6 Ausloos and Lias, J. Chem. Physics, 1964,41,3962.7 Stevenson and Schissler, The ChernicaZ and Biological Action of Radiations (Academic Press,8 Thompson and Schaeffer, J. Amer. Chem. Soc., 1958,80,553.9 Schaeffer and Thompson, Rad. Res., 1959,10,671.10 Maschke and Lampe, J. Amer. Chem. Soc., 1964,86,569.11 Ausloos, Lias and Gorden, J. Chem. Physics, 1963,39, 3341.12 Chupka and Lindholm, Arkiv. Fysik., 1963, 25, 349.13 Rylander and Meyerson, J. Amer. Chem. SOC., 1956,78, 5799.14 Grub and Meyerson, Mass Spectrometry of Organic Ions (Academic Press, New York, 1963),15 Stevenson, Trans. Faraday SOC., 1953,49,867.16 Borkowski and Ausloos, J . Chem. Physics, 1963, 38, 36.17 Borkowski and Ausloos, J. Chem. Physics, 1964, 40, 1128.18 Ausloos and Lias, J. Chem. Physics, in press.19 Smith, Corman and Lampe, J. Amer. Chem. Soc., 1961,83,3559.20 Futrell and Tiernan, J. Chem. Physics, 1963, 38, 150.21 Lampe and Field, J. Amer. Chem. Soc., 1959,81, 3242.22 Ausloos and Lias, J. Chem. Physics, 1962, 36, 3163.23 Ausloos, Lias and Sandoval, Disc. Faraday Soc., 1963, 36, 66.London, 1961), vol. V, pp. 249-254.p. 518