SOME ASPECTS OF THE RADIATION CHEMISTRY OF ORGANIC SOLUTES BY GABRIEL STEIN Department of Physical Chemistry, Hebrew University, Jerusalem Received 3 1 st January, 1952 The reactions leading to the eventual formation of free radicals in irradiated aqueous solutions are discussed. It is concluded that, particularly for sparsely ionizing radiations, electron ejection and capture processes are of importance for the understanding of the resultant chemical changes. Reactions of organic substrates, especially some dyes, in aqueous and non-aqueous systems are described, and it is shown that reversible reduction processes occur. Their mechanism is discussed with reference to th& possible biological significance. Other reactions of organic substrates, especially simple aromatic substances, are described which prove the formation and the role of the OH radical.The specific functions of molecular oxygen in the radiation chemistry of organic substrates are shown to depend on the nature of the substrate. Regarding the mode of action of ionizing radiations on dilute aqueous solutions, Weiss’s explanation 1 has been accepted in general. According to this, the solvent molecules are split, to yield in the end neutral fragments : H20 %%%*%+ H + OH (1) The individual processes leading to this final result have been the subject of some discussion. Lea 2 considered ionization to be the main primary process leading to observable chemical action, the chemical effects due to energy absorbed in excitation being negligible. The reason for this is that, in processes involving ionization, according to H2O %-%%%% + H20+ + e, the electron ejected will react at a site remote from the place of the original ioniza- tion, so that the radicals formed by processes like (3) and (4) (44 will have a smaller chance to recombine than those formed by a direct dissociation as in reaction 1, where the solvent cage will facilitate recombination to some ex- tent.394 Recently, it has been suggested5 that most of the chemical effects of alpha particles may be explained on the basis of the emission of ultra-violet quanta, which would cause chemical effects due to reaction 1.It has been assumed that, in this case, the charges produced in the primary reaction 2, by the alpha particles, recombine by virtue of coulombic attraction.On the other hand, attention has been drawn recently 6 to the possible significance of unstable ionic intermediates, other than those appearing in reactions 1-4, for the chemical processes. The suggestion has also been made 7 that, in the electron capture process 4, molecular hydrogen is produced directly according to 227 (2) H20+ + aq -+H& + OH H20- + aq -+ H + OH;, H20 + e -+ H2O- H20 + e -+ H2 + 0- (4b)228 ORGANIC SOLUTES This process was assumed to have a significant probability to account for a number of the chemical reactions observed. In the following, some r a n t experimental results involving organic substrates are discussed which may have a bearing on these problems. processes involved in the radiation chemistry of aqueous systems have often been considered in the light of experience gained from work with gaseous systems.It might be advantageous to consider the unique properties of aqueous solutions and, in other cases, to compare processes with effects in solids, to which many of the properties of aqueous systems are akin. Thus, whenever ionization processes occur in aqueous solution, sclvation will occur, owing to the rearrangement of the solvent dipoles in the field of the freshly produced charge. This rearrangement reduces the free energy of the system to a very great extent, so that reaction 3 will take place and be sufficiently exothermic to cause the formation of a neutral free radical and a hydrated proton. The field due to this will be much smaIler than could be deduced from the value of the original ionization potential of the HzO molecule.Whilst the depth of the freshly formed electron trap is thus decreased, simultaneously the free energy of the liberated electron is also decreased, owing to the polarization it causes in the medium. By the time the electron is slowed down to thermal energies, it may come under the influence of the electron affinity of the aqueous system itself or that of other suitable acceptors present in solution. A quantity related to the energy gained when an electron attaches itself to liquid water has been recently estimated from photochemical data.8 There is a likelihood therefore that, in aqueous solutions, the back reaction THE IONIZATION PROCESS AND THE REACTIONS OF THE CHARGED ENTITIES.-The H20f + e -+ H20 will occur to a much smaller extent than either in the gas phase or in most other solvents.The situation is somewhat modified with densely ionizing radiations, where the sites of subsequent ionizations along the track are near enough to each other to permit combination between an electron originating from a preceding site and a positive trap formed farther along the track. Thus, whilst the recombination between the or&inaZ partners seems unlikely, there is an increased likelihood, with densely ionizing particles only, of ionic recombination along the track. The fact that, even in gases, the rate of decrease of ionization due to recombination is proportional to n2, where n is the number of ion pairs present per unit volume, indicates that recombination is mainly between positive ions and electrons which do not originate from the same event.Thus, ionic recombination in aqueous systems irradiated with gamma or hard X-rays will not play a large role. With alpha particles such processes, as well as others due to the high ionization density, e.g. interaction of radicals,9 may play a role. They may account for the ultra- violet emission in this case, as well as for the reduced chemical efficiency of alpha particles, compared with gamma and X-rays, which has been observed in some cases. 10 It is thus to be expected that, in aqueous systems irradiated with sparsely ionizing radiations, reactions 3 and 4 will predominate. Other positive ions, which may be formed instead of H20+, e.g. OH+, may be expected to undergo a similar fate. Owing to the very high energy of hydration they will yield the hydrated proton and the corresponding neutral entity before other chemical reactions can occur, unlike the situation in the gas phase.In the following, the action of sparsely ionizing radiations, e.g. gamma and X-rays, will be considered only. ejected electron in the gas phase have been discussed by Burton and Magee.3 In aqueous solution, the electron may interact with the solvent which is in great excess, or with other electron acceptors present which have electron affinities (5) CHEMICAL ACTIONS DUE TO THE EJECTED ELECTRONS.-The reactions due to theGABRIEL STEIN 229 exceeding that of water. In addition to some inorganic positive ions which may possibly play such a role, in acid solutions the hydrated proton may interact either according to ~ 3 6 (6) in an exothermic process or, if reactions 4 and 4a have aIready taken place, accord- ing to Weiss’s suggestion 11 (7) This latter ion will then presumably serve as an electron acceptor, yielding molec- ular hydrogen.In both cases it will be expected that the evolution of hydrogen and the decomposition of water will be dependent on the pH, whilst in the case of direct formation of H2 according to 4b it would presumably be independent of it. In fact it has been found that, whilst neutral, pure water yields only a small quantity of hydrogen gas,l2 the decomposition of water and formation of hydrogen gas are greatly increased by the presence of Hf ions originating from acids, which themselves are not attacked, and which do not enter chemical reactions.13 This dependence on the pH seems to indicate that the importance of reactions of the type of 6 and 7 outweighs that of 46.The experiments of Dainton and James 6 support this view. It is possible that 4b is responsible for the constant, relatively small amount of H2 found in the decomposition of dilute aqueous solutions, even in that of H2O2.14 If this view is correct, it would be expected that, under the influence of sparsely ionizing radiations, under suitable experimental conditions, reduction processes will proceed readily, caused either by the ejected electron, or the H atom resulting from it. Recently it was found that, when methylene blue is irradiated in aqueous systems,l~~ 16 different phenomena may occur, according to the experimental conditions.If the irradiation is carried out whilst molecular oxygen is present in the system, irreversible reactions may take place. These involve the destructive oxidation of the dye with a relatively low yield, depending on the concentration of the dye.17 If, however, the irradiation is carried out in the absence of molec- ular oxygen, a reversible decoloration can take place with high efficiency. The decoloration can be reversed by the admission of molecular oxygen.16 The methylene blue may be incorporated into a gel, containing 1 to 10 % of gelatine or agar and only 0.003 % of the dye. Under these conditions, the re- versible decoloration will proceed readily in the absence of 0 2 and the reduction yield can be further increased by the incorporation of substances like benzoic acid in the system.The role of this would be to take up the OH radicals formed, which otherwise would reoxidize the dye.16 If the irradiation is carried out in the presence of 0 2 , there will be no decolora- tion up to a certain dose, beyond which decoloration will proceed readily. If the irradiation is carried out in the presence of carbon dioxide, a partial inhibition of the decoloration is observed. These results can be explained on the assumption that the dye interacts with either the electrons or with the H atoms formed from the electrons, according to H& + e -+ H ~ , + g + H -+ H L ~ dye + e -+ dye- dye + H --+ dyeH. In aqueous solutions we have been unable to differentiate between these two pro- cesses. In the presence of 0 2 , by virtue of its greater electron affinity, the reaction 0 2 + e --f 0 2 - , (9) or the corresponding reaction with the H atom 0 2 + H -+ HO2 ( 9 4 will take place in preference to reactions 8, 8a.to take place, and 0 2 admitted afterwards, the reaction If reactions 8, 8a are permitted (10) dye- + 0 2 --f dye + 0-2230 ORGANIC SOLUTES will take place involving a complete or partial electron transfer. C02 competes with the dye 16a but less efficiently than 0 2 and is itself reduced according to c02 + e' --f c02-- (1 1) CO2 + H -+ HC02. (1 l a ) That the original assumption of the role of C02 is correct6 is supported by the recent important experiments of Garrison and co-workers 18 which show the reduction of C02 in the absence of 0 2 and the formation of formic acid under the influence of ionizing radiations.The gel system described above manifests several interesting properties. In its composition it is not unlike biological systems and, in it, one particular substance, present in very small quantities only, is preferentially affected through a series of reversible oxidation-reduction processes which are initiated by the reducing entity formed by the radiations. In such a fairly complex system, the molecule with the greatest electron a f i i t y will retain in the end all those electrons or H atoms which have not undergone other, irreversible processes. The system may therefore serve as a schematic example of highly organized biological structures, and its behaviour may be used in the interpretation of the great radiation sensitivity of some living organisms in which one important component, say an enzyme, may be preferen- tially affected, although present in minute quantities only.These effects of the reducing entity may also be compared with the experiments of Forssberg.19 The enzyme catalase, in aqueous solution, in the absence of 0 2 , is inactivated on ir- radiation. This has been attributed to the action of the reducing entity, and this interpretation was supported by the fact that only those added substances provided a protection against inactivation which themselves were potential acceptors of the reducing entity. A reinterpretation of the experiments of Zimmer 1% may serve as further support for the conclusions reached here regarding the importance of electron (or H atom) trapping centres.He found that methylene blue is decolorized not only in aqueous, but also in glycerol solution. When this glycerol solution of the dye was under- cooled to - 70" C, forming a rigid glass containing the dye in solution, and was then irradiated, decoloration did not take place. If, however, this irradiated glass was allowed to warm up to room temperature without further irradiation, de- coloration occurred. Zimmer interpreted his results in terms of the splitting of the solvent into unspecified active radicals, which cannot diffuse away at - 70" C, but do so when the temperature is raised. Were the situation such, it would be more reasonable to expect recombination of the rigidly held, bulky radicals as soon as the temperature is raised.It is more likely that ionization takes place as the primary process, the electron which has been ejected travelling in the solid. It may then be trapped by (i) the solvent, (ii) the dye or (iii) positive ions created through irradiation according to I = Cl[S11 + C2[S21 + C3[S31 +- . . . (1 2) where [Sj] denotes the concentration of electron acceptors Sj, ci is a proportionality factor related to its electron affinity and I is the total number of electrons liberated by the radiation. Even though the electron affinity of the solvent is comparatively small, at these low temperatures the velocity of the subsequent reaction involving the release of a captured electron from a trap - d[Si-]/dt = Aj exp (- Ei/RT) [Si-] (1 2 4 will be greatly decreased.Consequently, electrons once trapped will be preserved in this state for longer periods than at room temperature. Therefore, since the number of solvent electron traps greatly exceeds the number of other traps which compete with it for the electrons, decoloration will not occur. This picture will be correct, if the probability of electron capture by one particular type of trap is proportional to the number of such traps and if the factor of proportionality CiGABRIEL STEIN 23 I is not a sensitive function of the temperature. On warming, reaction (12a) sets in, involving electron release from the solvent traps and retrapping by the dye. Alternatively, reactions analogous to reaction 4a can take place and normal decoloration can be observed. If this interpretation is correct, electron trapping should occur at very low temperatures even in solid water, and ice should show e.g. paramagnetic resonance after irradiation, owing to the presence of odd ions. Whilst this experiment has not been carried out yet owing to experimental complications, recently somewhat similar experiments were performed supporting the views detailed above and bringing the study of these processes more in line with previous knowledge derived from sirmlar processes in solids.The bulk of such information has hitherto been derived from the study of irradiated ionic crystals, where trapping was mainly by irregularities, the state and distribution of which was not always precisely known or controllable. How- ever, certain dyes, e.g. Sudan 111, when incorporated in plastics, form a solid system in which the nature and numbers of at least some of the trapping centres are known. On irradiation with X- or gamma-rays,20 these coloured plastics exhibit a deepening of colour, quite unlike their behaviour on irradiation with ultra-violet light, which results in bleaching.The deepening of the colour is due to the development of a well-defined absorption band in the red portion of the spectrum. Similar to the methylene-blue gels, the formation of this band can be inhibited by the presence of 0 2 in the plastic. If once formed by the irradiation of the plastic in the absence of 0 2 , it can be made partially reversible by the diffusion of 0 2 into the plastic. The irradiated plastics, coloured and uncoloured, show paramagnetic resonance after irradiation 21 which, on detailed investigation, indicates trapping of the ejected electrons partly by the plastic “ lattice ” itself, and partly by the dye.Electron ejection and trapping processes in plastics have previously been considered by Winogradoff.22 Preliminary experiments on the effect of the temperature at which irradiation has been carried out 21b indicate that, at lower temperatures, the relative trapping efficiency of the shallower trap increases greatly and that, on raising the tempera- ture, electron release and re-trapping processes do occur. Thus, when the plastics are irradiated in liquid nitrogen, the pattern of paramagnetic resonance observed at these low temperatures is different from that obtained by irradiating the same plastics at room temperature, and changes into the latter, if the specimen irradiated at low temperatures is allowed to warm up. Related thermal or photochemical processes have been observed previously in other solids 23 indicating release of trapped electrons and retrapping.In these cases, electrons ejected by more energetic radiations and later trapped may be released from the traps by, e.g. infra-red radiation or thermally, and retrapped in other deeper, or shallower, traps. If such reversible electron trapping processes have really some significance in biological systems, then the experiments reported here may bear some relatior, to the recently observed fact that, if irradiation with ionizing radiations of some biological structures is preceded or followed by irradiation in the visible or infra- red, the biological effects may be altered in a manner which cannot be obtained by infra-red irradiation alone.24 REACTIONS OF THE OH RmICALS.-The oxidizing action of the radiations in aqueous solutions may possibly have been explained by the assumption of active entities other than the OH radical, which has been assumed above to result from the ready hydration of the charged positive fragment, as in reaction (3).Recently, experimental evidence has been obtained 253 26 which strongly indicates that the active species is in fact a free radical and, most likely, the OH radical. It has been known that OH radicals formed by other means, e.g. by Fenton’s reagent (H202 + Fez+ salt) according to the Haber-Weiss mechanisms27 are capable of hydroxylating aromatic compounds, e.g.benzene to phenol.28 Reinvestigation of this process using ionizing radiations instead of Fenton’s reagent has shown 26232 ORGANIC SOLUTES that similar processes occur in this case as well. The elucidation of the reaction mechanism has shown that the reaction occurs in two distinct steps. In the first of these a free organic radical is formed from the aromatic molecule, which is dehydrogenated. This indicates a reaction according to The phenyl radical can undergo secondary reactions which differ according to the experimental conditions. In the absence of 0 2 it can react, e.g. (i) with another OH radical forming phenol ; (ii) with another similar organic radical, yielding diphenyl ; (iii) with an H atom re-forming benzene; or (iv) it may attack another benzene molecule, yielding diphenyl that way.In the presence of 0 2 the organic radical will react preferentially with this or with the HO2 radical formed from it. In this manner, phenol formation will be enhanced and diphenyl formation sup- pressed, in the presence of 02.26 These results show that OH radicals are produced in irradiated water. The interaction of these with organic acceptors may lead to the formation of organic free radicals. Further proof for this has been provided in a series of experi- ments25-31 in which the analogous reaction was carried out using mono- substituted benzene derivatives as the substrates. In these, the existing sub- stituent may exert a directive action with respect to a second incoming substituent.If so, the ratios of the three possible di-substituted derivatives will not be in statistical proportions (2 ortho : 2 meta : 1 para). The quantitative study of the formation of the three hydroxylated derivatives obtained from these substrates has shown that, in all these cases, the directive effect of the existing substituent has operated in a manner which is, for such groups as the -N02,296 -COOH,29c -Cl,31 not their usual mode of operation. Namely, in all these cases, the substitution was mainly in the para position, with some ortho and meta formed. Together with the study of the influence of an already existing -OH group 30 where no meta-sub- stituted derivative was formed at all, the results were in agreement with the assump- tion of a free radical mechanism, and with the experimental results of other workers, using free radicals of a different nature and of different origin in similar aromatic substitution reactions.32~ 33 It has also been shown,30 that when the reacting entity is not a free radical, but a charged entity, e.g.the (FeOH)Z+ ion, different substitu- tion ratios may be obtained. On the one hand, this study of the aromatic substitution ratios serves as a confirmation of the role of OH radicals in radiation chemistry; on the other hand, the action of radiations on aqueous solutions provides a suitable method for the study of organic reaction mechanisms involving free radicals. THE ROLE OF MOLECULAR OXYGEN.-It has already been shown that the 0 2 molecule may play an important part in the reactions occurring in irradiated aqueous solutions, owing to its electron affinity 347 6 and to the fact that reaction 10a, the attachment of an H atom to the 0 2 molecule, is both strongly exothermic and does not require any appreciable activation energy.35 Thus in irradiated systems, whenever 0 2 is present or being formed during the irradiation process, the impression may be created, when the behaviour of other solutes is being investigated, that reduction processes are suppressed. In reality these do occur, but it is 0 2 that is being reduced preferentially. In such systems, 0 2 may exert a protective action as with methylene blue,l6 with certain metal-ion solutions 33 and presumably also catalase.19 This will be also the case whenever 0 2 is present in a solution containing an organic substrate which reacts more readily with H atoms than with either OH or HO;! radicals.Thus, in the deamina- tion of certain amino acids,360 it was shown that this process proceeded more readily in the presence of H2 in the solution than either in vacuo or in the presence of 0 2 . The role of H2 is to react according to H2 + OH -+ HOH + H. (14)GABRIEL STEIN 233 The presence of 0 2 results in the removal of H atoms and a protection effect. In other cases, as shown in the preceding section, the presence of 0 2 may lead to the increased formation of one particular product, either through its addition to an organic free radical, which has been formed primarily, or through the intermediary formation of the HO;! radical, which thus removes H atoms from the solution, and can then add to the organic radical.This process may result in the specific forma- tion of one kind of organic product at the expense of another. Thus, the forma- tion of phenol occurs at the expense of the diphenyl formation in the presence of 02.26 In such cases, when one of the products is biologically active, whilst the other is possibly inactive, the presence of 0 2 may effect in this manner the bio- logical action of the radiation. This aspect has been emphasized recently when it was shown that, in the presence of 0 2 , organic products may be formed which are biologically significant and which are not formed at all in the absence of 0 2 . Thus from phenol, o-quinone is formed,30 whilst from amino acids, the cor- responding keto acid is obtained.37 These products resemble those obtained in biological processes.Similar reactions may influence the oxidation yields in inorganic systems. In all such cases, the presence of 0 2 will manifest itself in an increased reaction yield. Finally, it is also possible that the presence of 0 2 will make no difference to the apparent chemical yield, measured on one particular substrate. This may occur whenever H atoms and H02 radicals have similar reactivities with the substrate concerned. Thus it was observed that, with some amino acids, deamination occurred to the same extent under both sets of conditions.36a Similarly, with Ce4+ ions in acid solution, reduction processes will proceed to the same extent in de-aerated or 0 2 containing solutions.38 The influence of 0 2 will therefore be quite different according to the nature of the substrate, the experimental conditions and the particular process being observed.It may manifest itself, according to circumstances, in a protection effect, a poten- tiation, or no effect at all. 1 Weiss, (a) Nature 1944, 153, 748 ; (b) Trans. Faradzy SOC., 1947, 43, 314 ; (c) Brit. J . Radiol., Suppl., 1947, 1, 56. 2 Lea, (a) Actions of Radiations on Living Cells (Cambridge University Press, 1946) ; (6) Brit. J. Radiol., suppl., 1947, 1, 59. 3 cf. Burton, Ann. Rev. Physic. Chem., 1950, 1, 113. 4 Dainton, Brit. J. Radiol., 1951, 24,428. 5 Dee and Richards, Nature, 1951,168,736. 6 cf. Dainton and Collinson, Ann. Rev. Physic. Chem., 1951, 2, 99. 7 Haissinsky and Magat, Compt.rend., 1951, 233, 954. 8 Farkas and Klein, J. Chem. Physics, 1948, 16, 886. 9 Gray, J. Chim. Phys., 1951,48, 172. 10 Dale, Gray and Meredith, Phil. Trans. Roy. Soc. A, 1949,242, 33. 11 Weiss, Nature, 1950, 165, 728. 12 cf. e.g. Allen, J. Physic. Chem., 1948, 52, 479. 13 Milling, Stein and Weiss, to be published. 14 Johnson, J. Chem. Physics, 1951, 19, 1204; Lefort, Compt. rend., 1951, 233, 1194. 15 (a) Gallic0 and Camerino, Experientia, 1948, 4, 110; (b) Zimmer, Naturwiss. 1944, 16 Day and Stein, (a) Nature, 1950, 166, 146 ; (b) Nucleonics, 1951, 8, 11, 34. 17 Shekhtman, Krasnovskii and Vereshchinskii, Dokludy A N . SSSR, 1950, 74, 767 ; cf. also Vereshchinskii, Uspekhi Khim., 1951, 5, 737. Regarding related photo- chemical experiments cf. Schenck and Kinkel, Naturwiss., 1951, 38, 355, 503. 18 Garrison, Morrison, Hamilton, Benson and Calvin, Science, 1951, 114, 416. 19 Forssberg, Nature, 1947, 159, 308. 20 Day and Stein, Nature, 1951, 168, 644. 21 (a) Schneider, Day and Stein, Nature, 1951, 168, 645 ; (b) Day, Schneider and Stein, 22 Winogradoff, Nature, 1950, 165, 123. 32, 375 ; (c) for early work cf. ref. (2a). unpublished results. H234 INACTIVATION OF BACTERIOPHAGE 23 Garlick, Luminescent Materials (Oxford University Press, 1949), p. 35, 163 ; Curie, Nature, 1950, 166, 70. Regarding the role of organic dyes as electron acceptors, cf. e.g. Putseiko and Terenin, Doklady A.N. SSSR, 1950, 70,401 ; Chem. Abstr., 1951,45, 5018. 24 cf. e.g., Yost, Genetics, 1951, 36, 176. 25 Stein and Weiss, Nature, 1950, 166, 1104. 26 Stein and Weiss, (a) Nature, 1948, 161, 650 ; (6) J. Chem. Soc., 1949, 3245. 27 Haber and Weiss, Proc. Roy. SOC. A, 1934, 147, 332. 28 Cross, Bevan and Heiberg, Ber., 1900, 33, 2015. 29 Loebl, Stein and Weiss, (a) J. Chem. SOC., 1949, 2074 ; (6) J. Chem. Soc., 1950, 2704 ; 30 Stein and Weiss, J . Chem. SOC., 1951, 3265. 31 Johnson, Stein and Weiss, J. Chem. SOC., 1951, 3275. 32 Hey, Nechvatal and Robinson, J . Chem. SOC., 1951, 2892. 33 DeTar and Scheifele, J . Arner. Chem. SOC., 1951, 73, 1442. 34 Massey, Negative Ions (Cambridge University Press, 1950), p. 28. 35 Farkas and Sachsse, 2. physik. Chem. B, 1934, 27, 111. 36 Stein and Weiss, J. Chem. Soc., 1949, 3256 ; Stein, Watt and Weiss, to be published. 37 Johnson, Scholes and Weiss, Science, 1951, 114, 412. 38 Milling, Stein and Weiss, to be published. (cl J. Chem. Soc., 1951, 405.