r
E
N A L
Registered in U S Puterzt Bffice
O F
@ Copyright, 1974, by t h e A m e r i c a n Chemical Society
____l__-____l__l__..____
VOLUME 78, NUMBE
'81 23, 1974
chemistry of the Decay-Induced Gas-Phase Halogen Exchan ,3-Dichlorobutanes1 aniel, Hans J . Ache,* i4epartvent of Chemistry. Virginia Polytechnic /f?stituleand Slate University, B/acksbtirg. Virginia 2406 J
:nslitut fur Mukiearchemic. Kernforschungsanlage Jiiiich. Jtiiich, Wesl Germany (Recewed January 18, 7974)
The stereochemistry of halogen-for-halogen substitution a t asymmetric carbon atoms by decay-produced Immune and iodine species was studied in diastereomeric alkyl halide molecules such as d,E- and rneso2,3-dichlorobutane. Energetic sOBr ions were generated uta the HomBr(IT)80Brnuclear process using r as source and allowed to react with the substrate molecules while still possessing excess kinetic energy or after having become thermalized in collisions with argon atoms. Iz51 ions were produced uza the EC decay of 125Xe. In unmoderated systems a predominance of retention of configuration is generally observed. In the presence of excess argon, however, the degree of stereospecificity of the reaction varies with the type of diastereomer used as substrate. In the case of the thermodynamically less stable d,E systern the stereospecificity decreases from about 2.5 (as expressed by the ratio of retention tQ inversion) in the absence of Ar moderator to 0.4 at high Ar concentrations whereas, under the same conditions, only small changes in the stereospecificity are observed if the meso diastereomer is the substrate The results are explained on the basis of an electrophilic substitution involving a front-side attack of a positive halogen ion. In the case of the d,l system the initially formed halocarbocation can then undergo two competing pipocesses: excitation racemization or Cl+ transfer, which, depending on the environment, can be kinetically or thermodynamically controlled.
Introductioni Previous work by several investigators has suggested that a considerable fraction of the halogen species generated b,y nuclear processes involving inner shell vacancies or 1z5Xe(EC)1251 reacts as thersuch as *OmlBr(IT)HoOE~r mal ionic species with organic substrate molecules in the gas phase.2-1'i The isomeric transition (IT) of aOmBr incorporated in alkyl bromides gives rise to vacancy cascades following internal conversion and eventually to a substantial fragmentation of the molecule uia Coulomb explosion leading to SOBr ions with multiple positive charges.15 Similarly, the electron capture decay (E@)of 125Xe results in the formation of multiply charged positive lz5I ions. If these processes occur in the presence of a large excess of a rare gas having ;an ionization potential intermediate between the first and second ionization potential of Br or I the charge of the RQElror 1251 ions will be rapidly reduced
to unity by charge transfer processes,15 while the excess kinetic energy of some electron volts resulting from the Coulomb explosion or neutrino emission, respectively, is removed from the ions in unreactive collisions. The resulting thermal sOBr+ or Iz5I+ ions can subsequently undergo electrophilic reactions with organic substrate molecules. This is quite different from the hot homolytic substitution reactions of nucleogenic tritium and halogen atoms resulting from neutron-induced nuclear reactions, where the recoil species have very high initial kinetic energies and lose their possible charge very quickly while slowing down to the chemical energy range where they react predominantly as neutral atoms.16 Thus, it seems that the study of the reactions of decay-induced halogen species could significantly contribute to our knowledge of the detailed reaction mechanisms and parameters involved in electrophilic substitution reactions of simple hydrocarbons in the gas phase. It appears that this technique can be of 1043
che, and G.Stocklin special importance in cases where other methods such as ion cyclotron resonance or high-pressure mass spectrometry cannot provide the necessary information, i.e., in cases es to study the stereochemistry of the reaceveral isomers may result from the substitution. An interesting application of this technique is the recent investigation by Cacace and Stocklin7 in which they studied the isomer distribution following the electrophilic attack of halogen in various benzene derivatives. In the present study we tried to assess the stereochemical course of the decay-induced halogen for halogen substitution following 8omBr(IT)soBr and 125Xe(EC)1251in meso- and d , i-2,3-dichlorobutane in the gas phase under various experimental conditions. An attempt will be made to explain the observed results in terms of a mechanism which invohes an electrophilic (front side) attack of the halogen ion a t the asymmetric carbon atom leading to the formation of a relatively long-lived complex which may undergo s u ~ ~ ~ e ~ ~racemization. ently This mechanism is similar to th.2 mechanism recently proposed by Cacace17 for the gas-phase electrophilic attack of gaseous Brmsted acids su,ch as 3HeT+ at asymmetric carbon atoms. ateriais. CF3Br: CFIZ=CH-CH=CH~, Cl2, and 0 2 were obtained from Matheson Chemical Co. with a stated purity level greater than 99.0%. Argon and xenon with a stated purity level OS 99.999% were purchased from Air Reduction GO. and used without further purification. The meso and d,il forms of 2,3-dichlorobutane were prepared by stereospecific addition of 6 1 2 to trans- and cis-2-butene, respectively.18 The products were purified by gas chromatography, using 4-rrx glass columns ( 5 mm i d . ) with 20% DEGS on Chromosorb W, 60--80 mesh a t 80" and 100 cc of He/min. Similarly, the diastereomers of l-bromo-2,3-dichlorobutanes and the diastereomers of 2-bromo-2,3-dichlorobutanes were obtaine'd by stereospecific addition of C12 to cis- and trans-1- and -2-bromo-2-butene (purchased from K & K), respectively. Purification was achieved by gas chromatography on the columns described above at 110". erythro- aad Ihrtio-2-bromo-3-chlorobutane and 2-iodo3-chlorobutar~ewhich were used in small amounts as carriers were propared following procedures described in the literature19-z1 and subsequently gas chromatographically purified .z2 Preparatiol;! of 80BrSource. CF3Br was preferentially r sou.rce because of its ionization potential of 12.3 eV, which lies above that of Bri- and thus would not interfere with th'e Br+ reaction by charge transfer. A few mil1igram.s of CF3Br were sealed in a quartz capillary and i.rrtadiateci in the VPI and SU nuclear reactor a t a neutron flux density of about 1012n cm-2 sec-l for 30 min at 40". After the reactor irradiation the contents of the quartz ampoule were subjected to gas chromatographic purification on E . 2-m Poropak Q column. The GFSBr fraction, containing eomBr-, go&-, and 82Br-labeled CFSBr formed during the irradiation, was trapped and transferred to the reaction The other radioisotopes of bromine prodaced do not impose any problems, since they lead to stable daughters not giving rise to any relevant products. Preparation of I2":'Xe. About 15 ml of Xe at 5 atm were sealed in quartz anipoules and neutron irradiated in the BRJ-X nuclesr reactor of the Nuclear Research Center Juelich a t taeiitron flux density of 8 x 1012 sec-I for The Journai of Physical Chemistry. Voi 78. No. 7 7. 7974
a period of 75 min. Aliquots of about 2 ml (STP) were subsequently transferred into the reaction vessel. The other radioisotopes of Xe produced during neutron irradiation do not interfere since the EC decay of Iz7Xe leads to the stable (nondetectable) 12?I and the neutron rich isotopes of Xe decay by p- emission to Cs isotopes. Preparation of Reaction Mixture. The reaction was carried out in a spherical, specifically designed, Pyrex glass vessel with a total volume of 500 ml. In- and outlet valves were Kontes' greaseless high vacuum values. The vessel was filled by standard vacuum line technique with the desired amounts of reactants and additives. In the case of the Br+ investigation the reaction mixture was allowed to stand (in the dark, at room temperature) for 100 min to permit the "Br to attain equilibrium with the 8OmBr while in the reaction vessel. In the case of the 1251+experiments the exposure time was 70 min at room temperature. Sample Analysis. At the end of the reaction the *OBr or 1251labeled products were collected in traps at liquid nitrogen temperature and appropriate amount of carriers dissolved in CH2C12 added. The resulting solution was washed first with aqueous (dilute) N a 2 S 0 3 solution then with distilled water and dried. The analysis of the so& or 1251labeled products was accomplished by a discontinuous radio-gas chromatographic technique.23524 Columns used were either a 4-rn glass column (5 rnm i.d.) with 20% DEGS on Chromosorb W 60-80 mesh, temperature programmed, or a 8-m glass column (5 mm i d . ) 15% SF-96 on Chromosorb W, 60-80 mesh operated at 80". The latter column was used for the separation of the diastereomeric 2-iodo-3-chlorobutanes. Radioactivity Assay. The separated radioactive products were directly trapped from the effluent gas stream by bubbling the effluent gas from the gas chromatograph through toluene solutions containing liquid scintillation fluors. More than 99% of all the reaction products identified in this investigation were retained in the solution, with the exception of CF&, for which calibration runs were carried out and appropriate corrections made.23 The 8oBr-countingwas done by liquid scintillation spectrometry applying appropriate energy discrimination. Alternatively the 1251labeled fractions were adsorbed on charcoal tubes24 and counted in a well type scint,illat,ion counter. Radionuelidic purity was checked by y spectrometry in the case of W r by using the 665-keV The radiochemical yields of the products, i. e., the ratio of the soBr activity present a t the end of the reaction in each individual product to the total 80Br activity (at end of the reaction) were computed by using t,he well-known equations for radioactive decay and growth. In the case of lZ5I ( t l l 2 = 60 days) decay corrections were not necessary due to the long halr-life and the radiochemical yields were obtained by comparing the activity of the individual fractions with that of the total activity directly measured on an ajiquot,.
Results and Discussion Several series of experiments were carried out to identify the products resulting from the reaction of and 1251species generated in the nuclear decay of *OrnBr or lz5Xe with meso- and d, 1-2,3-dichlorobutane as substrate in the gas phase and to determine their relative radiochemical yields. Of particular interest for this study were the $*Br or l z 5 T substitution products derived from the two diastereomers
Gas-Phase Halogen Exchange in 2,3-Dichlorobutanes
1045
of the 2,3-dichlorobutane substrate molecules, which include in the case of rneso-2,3-dichlorobutane the following products as shown schematically
I
I
I
I
60
80
ARGON MODERATOR EFFECT 806r FOR CI EXCHANGE IN maso MU0 6 , 1 - 2 , 3 DICHLOROBUTANE FOLLOWING a o m m L U Q B ~
H H
I
H
H
l
l
E CH,Br-C--C--CHJ
I / c1 c1
I /
*
I I
CH,-C-C-CH,
dli
I I I* I Br Cl
CH,-C-C-CH,
3
(7070) €rom the front side resulting in a threecentered bond s t r ~ c t u r e ~ 7 ~ ~ 9 ~ ~ ~
I
,
H H thus yielding a su‘bstitution product which retains the original con,figuration. To acconnt for the considerable amount of “inversion” which occurred in the case of the d,l form upon addition of moderator, one would have to postulate that the resulting complex sur’vimes immediate decomposition, i e . , its energy contents is not exceedingly high, however, the complex has sufficient energy for racemization to occur during the lifetime of the complex leading to the observed ratio of erythro to tlireo product which a t high Ar concentrations is close to the thermodynamic equilibrium concentration of the two diastereomers. On the other hand, only a slight increase in the stereospecificity of the substitution reaction occurs when rneso-2,3-dichlorobutaneis the substrate. The initial attack by the electrophile predominantly leads to the erythro form (retention), but the threo product is also formed (possibly by back-side attack). Excitation racemization, if it occurs, will not lead to drastic changes since the two diastereomers are present in yields close to their thermodynamic equilibrium. The thermodynamically less stable threo form will be predominantly affected +JY excitation racemization giving rise to the erythro form. Indeed, the ratio erythro to threo is seen to slightly increase with increasing moderator concentration. In order to further test this excitation racemization mechanism one would have to add very efficient energy sinks, which would quickly remove any excess energy from the complex, or strong (gaseous) B r ~ n s t e dbases to initiate a quick Cl+ transfer, i.e., to reduce the possibility for racemization either by deactivation of the complex or by shortening the lifetime of the complex. A preliminary experiment in this direction was done by adding m a l l amounts of methanol (vapor), which is known to be a good energy sink and a relatively strong rlansted base, to t h e system. The results, listed in Table I, show a drastic increase in retention of the original con-
figuration if d, 1-2,3-dichlorobutane 1s the substrate and Br+ the electrophile, whereas only a slight change in the stereospecificity is observed when m~so-2,3-dichlorobutane is the substrate. The fact that only the d,l system is affected by the addition of methanol is in the agreement with the assumptions made above and the findings demonstrated in Figures 1, 3, and 4. Obviously, the postulated “excitation racemization” occurs to a significant extent only in the thermodynamically less stable d, 1 system. In the meso form racemization plays a minor role since a front-side attack of the halogen ion leads preferentially to the thermodynamically stable erythro form whereas in the d,E system mostly the thermodynamically less stable erythro diastereomer is initially formed. In this latter case racemization is either kinetically or thermodynamically controlled, depending on the environment.31 The assumption of two competing processes occurring after the formation of the halocarbocation in the d, 1 system, namely, racemization and C1+ transfer (see eq 4) would then explain the trends observed in Figures 1, 3, and 4. As the Ar concentration increases, racemization also increases due to the lacking possibility for 61+ transfer to substrate molecules. In the presence of methanol the stereospecificity increases, since Clf transfer to this molecule is more effective than to substrate molecules. CI’ front-side attack
d,E form
.
* [halocarbocation]* + acceptor
I
racemization
quick CI+ transfer
[halocarbocation]* + RX CI+ transfer
[eryihro product] (4)
1
[erythro
-k
threo products]
Additional studies using different Br$nsted bases and other good energy sinks which would change the rate of energy and CX+ or proton transfer and thus affect the lifetime of the postulated complex are presently carried out to test the proposed mechanism.
TABLE I: “OBr Found in Each of the Two Diastereomers of 2-Bromo-3-chlorobutane (BCB)aFollowin aOmBp(IT)Wrin meso- or d,Z-2,3-Dichlorobutanes (DCB) Under Various Experimental Conditions ~
_ _ _ l _ l
Subs;tra.te (12 ‘For?) _ _ _ I _
70 8OBr in BCRC ___-_______
Additives, Torrb -. 0 2
Ar
CH?=CH-CH=CH?
20 20 20 20 201 20
760 760 760
20
760 760 760
20
CHsOH
Erythro
. _ I _ _
meso-DCB d,I-DCB
10 10
78.2 75.4 80.2 70.0 62.1 8.5
Threo -
21.8 24.6 19.8
30.0 37.9 91.5
+
In % of 8oBr incarporeked into BCB erythro threo = 100%. Each sample contains 20 Torr of CFa 80”’Br. Experiments were carried out in a 500-m Pyrex vessel.; reaction time was 100 min at room temperature. Quoted values are the average of four to five experiments. The Journalof Physical Chemistry. Vol 78. No. 11. 1974
S. H. Daniel, H.J~ Ache, and G. Stocklin
(6) E. Tachikawa. Bull. Chem. SOC.Jap.. 42, 1504 (1969). (7) F. CacaceandG. Stocklin,J. Amer. Chem. Soc.. 94, 2518 (1972). (8) M . Yagi, K. Kondo, and T. Kobayashi, Radiochern. Radioanal. Lett.. T 9, 123 (1973). T i (9) M. D. Loberg and M , T. Welch, J. Amer. Chem. SOC.. 95, 1075 (1973). (IO) J. Okamoto and E. Tachikawa, Buil. Chem. Soc Jap., 42, 1504 (1969) (11) R. R, Pettijohn and E. P. Rack,J. Phys. Chem.. 76, 3342 (1972). (12) D. W. Oates. R. L. Ayres, R. W. Helton, M. S. Schwartz, and E. P. Rack, Radiochem. Radioanai. Lett.. 4 , 123 (1970) (13) F. Schroth and J. P. Adloff,J. Chim. Phys.. 61, 1373 (1964). (14) M . $1.Welch, J. Amer. Chem. SOC.,95, 1075 (1973) (15) For review, cf. S. Wexier, Actions Chim. Biol. Radial., 8 , 107 (1965). (16) For a review of hot atom reactions, c f . (a) A. P. Wolf, Advan. Phys. Org. Chem., 2, 202 (1964); (b) R. \Wolfgang, Progr. React, Kinet., 3, 124 (1965); (c) G . Stocklin in "Chemie heisser Atome," Verlag Chemie, Weinheirn. Germany, 1969. (17) F. Cacace and M . Speranza, J. Amer Chern Soc.. 94, 4447 (1 972). (18) M . C. Lucas and C. W. Gould, Jr., J. Amer. Chern. SOC.. G3, 2541 (1941) . (19) G . A. Olah and J. M . Bollinger, J . Amer. Chem. SOC..90,947 (1968). (20) M. Deiephine and L. Voile, Buli. SOC.Chim. F r . , 23, 673 (1920) 0 L ---~___.---____-LL(21) S. Winstein and E, Grunwaid. J. Arner. Chem. Soc., 70, 836 3 20 40 60 80 (1948). MOLE o/' ARGON (22) L. Vasaros, H. J. Machulia, and W . Tornau. J. Chromatogr . 62, L- _ _._ _ ~. _ _ _ 1 4.58 (1971). !50 IO0 1000 (23) S. H. Daniel and H. J. Ache, Radiochim Acta. 19, 132 (1973). TOTAL PRESSURE (TORR) (24) G . Stocklin and W. Tornau, Radiochirn. Acta. 6, 86 (1966); 8 , 95, (1968). Figure 4. Ratio of EoE3rlabeled erythro- to threo-2-bromo-3-chloE. J. Knust, Report Juei-940-NC (1973) robutane vs. mole per cent argon present in reaction mixture The reason for this behavior may be seen in the fact that at these following 80mBrjlT)E05rin meso- and d,l-2,3-dichlorobutane. high moderator concentrations the reaction products are formed at their equilibrium concentrations. Otherwise. one might expect to Experimental condition same as described in Figure 1 . see a change in the product ratio due to a shift in the relative probabilities for racemization and X c transfer as a function of presAcknowledgment. The authors are indebted to Mr. M. sure (vide infra), Schiiller, Mr. Jeff !Vu, and Dr. L. Bartal for valuable ex- (27) F. S. Rowland, C. M. Wai, C. T. Tiny, and G . Miller, "Chemical Effects of Nuclear Transformations." Voi. 1, International Atomic Enperimental. assistance. ergy Agency, Vienna, 1965, p 333. (28) (a) C. M. Wai and F. S. Rowland, J. Phys, Chem.. 74, 434 (1970); (b) ibid., 71, 2752 (1967); (c) H. J. Machuila, Report Jul-873-NC eferenes and, Nates (1972). (1) Work supported by tl?e U S , Atomic Energy Commission, (29) G. A. Olah, G. Klopman, and R. H. Schlosberg, '1. Amer. Chem. (2) J. B. Niclioias and E P. Rack. J. Chem. Phys., 48, 4085 (1968). SOC.,91, 3261 (1969). (3) J. B. Nichoias, J A. Merrigan, and F. P. Rack, J. Chem. Phys., 46, (30) G. A. Qlah and J, A. Olah,J. Amer. Chern. Soc., 03, 1251 (1971). 1996 (1967) (31) This model is very similar to that postulated for the isomerization of substituted aromatic compounds following the electrophilic attack (4) E. Tachikawa. Bull. Chem. SOC.Jap., 43,63 (1970). ( 5 ) E. Tachikawa and T . Kanara, B u l . Chem. Soc. Jap.. 43, 1293 of 805r+.(H. J. Knust, A. Halpern. and G. Stocklin, J. Amer. (1970). Chern. SOC.,in press.)
,
4RGON UODERATOR EFFECT s'Dr t m e w AN0 a , P - 2 , 3 DICHLOROBUTANE
~
The Journal of Physical Chemistry. Voi 78. N o . 1 7 . 1974
