J. Chem. SOC. Faraday Trans. I, 1982, 78, 1141-1 148 Kinetics of the Gas-phase Addition Reactions of Trichlorosilyl Radicals Part 3.-Additions to 2-Olefins BY TAKAAKI DOHMARU* AND YOSHIO NAGATA Radiation Centre of Osaka Prefecture, Shinke-cho, Sakai, Osaka, Japan Received 1 1 th May, 198 1 The following Arrhenius parameters for the forward and reverse steps of trichlorosilyl radical additions to trans-but-2-ene, cis-but-2-ene, cis-pent-2-ene, 2-methyl-but-2-ene and cyclopentene have been obtained by a competitive method. The relevant elementary reactions are (3) * SiC1, + CH,COCH, + (CH,),cOSiCl, SIC^,+ +< >-$--s~cI, (5, - 5 ) \ I I I and c-y -SiCl, + HSiC1, + H~-~-SiCl, + SiC1, (6) / E5-E3 log(A-,/A, E+-EB olefin log (A,/A,) /kJ mol-l /mol ern-,) /kJ rno1-I temp./K cis-CH,CH=CHCH, 0.91+_0.02 0 4.09 k 0.10 77.0 f 0.8 397-462 trans-CH,CH=CHCH, 0.81+0.01 0 3.99 & 0.17 76.1 _+ 1.3 378-501 cis-CH,CH=CHCH,CH, 0.57+0.03 0 5.39 0.10 85.8 _+ 0.8 406-455 * * cis-CH,CH=CHCH,CH, 0.89+0.01 0 5.39k0.10 85.8k0.8 406-455 * CH,CH=C(CH,), 1.44k0.30 0 5.17+ 1.00 78.7k8.4 406-456 * CH,CH=C(CH,), 0.24f0.32 0 5.174 1.00 78.7k8.4 406-456 cyclopentene -0.80k0.15 - 10.6k 1.3 1.40k0.77 61.1 k7.1 406-489 The rate parameters of reaction ( 5 ) are expressed per reaction site; an asterisk indicates the site of addition in an unsymmetrical olefin.Evaluated values of A_, and A, imply a fairly ‘loose’ transition state in reaction (5). The Si-C bond energy has been estimated. *SiCl, radicals have been revealed to be electrophilic and susceptible to steric hindrance.In Part 2‘ of this series we reported a kinetic study of trichlorosilyl radical additions to 1-olefins. Using the same techniques we have carried out a similar study on cis- bu t-2-ene, zrans- but-2-ene, cis-pen t -2-ene, 2-methyl-bu t-2-ene and cyclopentene. In addition to the rate parameters, the orientation of addition in unsymmetrical olefins has been examined carefully because it may give some information about the nature of the transition state.2 11411142 ADDITION OF TRICHLOROSILYL RADICALS EXPERIMENTAL Trichlorosilane and acetone were purified as described previ~usly.~ Reagent grade cis- but-2-ene, trans-but-2-ene (Takachiho Kagaku) were used as received. cis-Pent-2-ene, 2- methyl-but-2-ene and cyclopentene were vacuum distilled. The purities of these olefins were checked by g.1.c.The sample preparations, photolyses and analyses were performed on a conventional greaseless and mercury-free vacuum line. The light from a 50 W medium-pressure mercury arc was filtered (Toshiba UV-33) and admitted to a cylindrical quartz reaction vessel of 139 cm3. Once photolysed, the reaction mixture was allowed to diffuse to a gas-sampling loop and was immediately analysed by a gas-chromatograph with a Gow-Mac gas-density detector. The adduct from each olefin was identified mainly by g.c.-m.s. analyses, which also made it possible to distinguish the isomeric adducts from an unsymmetrical olefin. In addition, the g.1.c. retention time of an adduct was compared with that of the authentic sample prepared by a radiation-induced reaction of trichlorosilane with the respective 0lefi1-1.~ RESULTS AND DISCUSSION Trichlorosilane, acetone and one of the 2-olefins (cis-but-2-ene, trans-but-2-ene, cis-pent-2-ene, 2-methyl-but-2-ene and cyclopentene) were mixed and photolysed at ca.440 K. Reaction products were almost exclusively the adducts of trichlorosilane to acetone and the respective olefin. The adduct to either end of the double bond was produced when an unsymmetrical olefin was used. The photoirradiation time was arranged to keep the reactant consumption < 5%. To suppress the formation of the higher telomers,l the ratio of olefin to trichlorosilane was kept as small as possible. Photolysis of acetone in the presence of HSiCl, and olefin has been shown to proceed via the following free-radical chain reactions3 hv CH,COCH, + 2 CH3 + CO CH, + HSiC1, (TCS) + CH, + SiCl, SiCl, + CH,COCH, (Ac) + (CH,)$OSiCl, \ / \ I SiCI, + C=C, (01) s ,C-<i-sicl, / (CH,),~OSiCl, + HSiCl, + (CH,),CHOSiCl, (A) + SiCl, \ I I I $-~-SiC13 + HSiCl, + HF-7-SiCl, (0) + SiCI,.In the scheme above, two chain cycles occur concurrently and result in the adducts A and 0. Assuming low conversions of the reactants and long reaction chains, a conventional steady-state treatment of the above scheme leads to the following rate equation (7) (RA/RO) ([olI/[Acl) = (k3/k5) {' + (k-5/k6) [TCS1-'l where k, denotes the rate constant of reaction (x). The rates of formation of A(R,) and 0 (R,) were measured as a function of TCS concentrations at various temperatures. They are given in table 1 in the form convenient for testing the validity of eqn (7).Acetone concentrations were (0.27-2.7) x mol cm-, and the ratio of olefin to acetone was from 0.1 to 0.3. From cis-pent-2-ene both 2-trichlorosilylpentane and 3-trichlorosilylpentane were formed. The ratio of the rate of formation of the two isomers was extensively studied and was found to be &/R3 = 0.471 k0.033. The ratio was virtually constant underT. DOHMARU AND Y. NAGATA 1143 TABLE 1 .-DEPENDENCE OF (R,/R,) ([Ol]/[Ac]) (b) ON [HSiCl,]-' (a) FOR trans-sUT-2-E~E, TEMPERATURES (U = 1 O5 cm3 mol-l) CiS-BUT-2-ENE, CiS-PENT-2-ENE, 2-METHYL-BUT-2-ENE AND CY CLOPENTENE AT VARIOUS 0 \*- / -\ \- \=/ \ * / \ -\ T/K alU b T / K a/U b T / K a / U b T / K a / U b T / K a/U b ~~ 378 378 378 378 404 404 404 404 416 416 416 416 423 423 423 423 428 428 428 428 433 433 433 433 438 438 438 438 443 443 443 443 448 448 448 448 452 452 452 452 477 477 477 477 50 1 50 1 50 1 50 1 50 1 ~~ 1.86 0.075 3.72 0.085 5.31 0.083 7.44 0.082 1.86 0.091 3.72 0.132 5.31 0.140 7.44 0.169 1.82 0.106 3.64 0.154 5.21 0.190 7.28 0.229 1.82 0.125 3.64 0.184 5.21 0.228 7.28 0.276 1.86 0.145 3.72 0.221 5.31 0.276 7.44 0.356 1.22 0.141 3.64 0.250 5.21 0.331 7.28 0.412 1.82 0.177 3.64 0.300 5.21 0.419 7.28 0.532 1.82 0.195 3.64 0.387 5.21 0.501 7.28 0.680 1.82 0.233 3.64 0.474 5.21 0.608 7.28 0.820 1.86 0.306 3.72 0.545 5.31 0.689 7.44 1.036 1.86 0.760 3.72 1.43 5.31 1.99 7.44 2.85 1.86 1.52 2.48 1.99 3.72 2.90 5.31 4.27 7.44 5.98 397 1.77 0.075 397 3.48 0.091 397 7.04 0.105 418 1.77 0.0953 418 3.53 0.123 418 5.29 0.146 418 7.07 0.194 443 1.77 0.178 443 3.55 0.266 443 5.32 0.379 443 7.09 0.495 452 1.77 0.235 452 3.53 0.403 452 5.32 0.579 452 7.09 0.754 462 1.77 0.326 462 3.55 0.577 462 4.16 0.663 462 5.32 0.810 462 5.92 0.959 462 7.09 1.129 406 1.76 0.185 406 3.51 0.231 406 5.01 0.289 406 7.02 0.342 423 1.76 0.281 423 3.51 0.419 423 5.01 0.554 423 7.02 0.728 439 1.76 0.464 439 3.51 0.881 439 5.01 1.168 439 7.02 1.449 455 1.76 0.773 455 3.51 1.671 455 5.01 2.373 455 7.02 3.128 406 1.89 0.089 406 1.76 0.148 406 3.78 0.167 406 3.51 0.144 406 5.40 0.206 406 5.01 0.166 406 7.56 0.363 406 7.02 0.164 423 1.89 0.203 423 1.76 0.155 423 3.78 0.366 423 3.51 0.198 423 5.40 0.476 423 5.01 0.198 423 7.56 0.650 423 7.02 0.192 440 1.89 0.465 439 1.76 0.179 440 3.78 0.902 439 3.51 0.232 440 5.40 1.34 439 5.01 0.285 440 7.56 1.88 439 7.02 0.294 448 1.26 0.423 456 1.76 0.265 448 1.51 0.579 456 3.51 0.319 448 1.89 0.746 456 5.01 0.361 448 2.52 0.948 456 7.02 0.474 448 3.78 1.42 472 1.76 0.323 448 4.45 1.79 472 3.51 0.523 448 5.40 2.11 472 5.01 0.653 448 7.56 3.08 472 7.02 0.803 456 1.89 1.13 489 1.76 0.550 456 3.78 2.21 489 3.51 0.895 456 5.40 3.02 489 5.01 1.15 456 7.56 4.06 489 7.02 1.51 * Indicates the site of addition.An adduct to the other site is not shown here; see text.1144 ADDITION OF TRICHLOROSILYL RADICALS the present experimental conditions. The values of b for the 2-isomer are easily calculable from those for the 3-isomer using RJR, and are omitted from table 1. Speir and Webster, studied peroxide-initiated reactions of trichlorosilane with pent-2-ene between 353 and 368 K in the liquid phase and suggested the formation of the two isomers with R,/R, = 2.3.It is of interest that the value in the gas phase is much different from that in liquid phase, but the difference cannot be explained at the present time. A detailed study of the dependence of the ratio on phase and temperature is continuing in our laboratory. 2-Methyl-but-2-ene gives 94.1 _+ 1.6% 2-methyl-3-trichlorosilylbutane and 5.9 f 1.6% 2-methyl-2-trichlorosilylbutane; the ratio of the two isomers is also invariant under these experimental conditions. No experimental value is available for this ratio. Values (RJR,) ([Ol]/[Ac]) and ITCS1-l in table 1 were plotted according to eqn (7); the linearity of the plots at each temperature was excellent for all the olefins studied.Eqn (7) shows that the intercept of the linear plot gives k,/k, and the slope gives (k3/k5) ( k 5 / k 6 ) . Accordingly, k-,/k, is obtained from the slope divided by its intercept. Arrhenius parameters of k,/k, and k-,/k, for each substrate were calculated by a weighted linear least-squares method.,* The results are given in tabular form in the abstract; error limits are the standard deviations from least-mean-squares plots. EVALUATION OF ELEMENTARY RATE PARAMETERS Hydrogen abstraction reactions by alkyl radicals have been extensively studied so that A factors of this type of reaction are generalized to be 1011.5*0.5 cm3 mol-l ~ - 1 . ~ Recently, Arthur et aL8 evaluated the kinetic data for hydrogen abstraction from silanes.From their compilation we remove the A factor for hydrogen abstraction from HSiCl, and obtain as an average value 1011.6*o.8 cm3 mol-l s-l. Therefore it is reasonable to assume that the A factor for reaction (6) is The attacking radicals in reaction (6) are p-substituted alkyl radicals, and may be considered to have small polarities, since the substituent effect is attenuated by the -CH2- groupO9 In this narrow series of very similar reactions, the Polanyi relationlo may be used to estimate the activation energies. Activation energies for the reaction of -CH, (18 kJ mo1-l)l1 and -CH,CH, (22.2 kJ mol-l)ll with HSiCl, were adopted as standards. The enthalpy changes were calculated from D(H-SiCl,) = 382 kJ mol-l obtained by Walsh12 and the well-established values of D(R-H).13 Using the k , values thus obtained, k-, values may be calculated as shown in table 2.The values for 1-olefins previously obtained, are also presented in table 2 for the purpose of the following discussion. The values of log A _ , for 2-olefins other than cyclopentene are much greater than 13.4, the value giving ASS = 0 at 440 K. It follows that SiCl, detachment and consequently SiCl, addition proceed through fairly ' loose' transition states in the case of 2-olefins as well as 1-olefins. Cyclopentene behaves in a completely different way. More work on cyclic olefins would be needed to undertake a detailed discussion. (8) cm3 mol-l s-l. A , is related to A _ , by eqn (8): log (AJA-,) = AS5O/19.15 +log,, (e x RT) where AS: is a standard entropy change accompanying reaction ( 5 ) and log,, (e x RT) is a conversion factor for the change in standard state from 1 atm at s.t.p. to 1 mol cm-,.Entropies of the respective adduct radical may be calculated by the * In runs for the 2-olefins other than cyclopentene, the error limits of the intercept values were much larger than those of the slope values at higher temperatures. Thus, the intercept values were averaged first and the k-,/k, ratios were calculated using the averaged values.T. DOHMARU AND Y. NAGATA 1145 TABLE 2.-EVALUATED VALUES OF ELEMENTARY RATE PARAMETERS 15.0- 117.2 kJ mol-I/RT In 10 14.8- 108.4 kJ mol-l/RT In l o RTln 10 15.3- 113.0 kJ mol-l/RTIn 10 14.8- 107.1 kJ mol-l/RT In 10 16.3- 116.3 kJ mol-l/RT In 10 15.6- 102.1 kJ mol-l/RT In 10 15.5- 101.3 kJ mol-'/RTln 10 16.9- 110.9 kJ mol-l/RT In 10 16.9- 110.9 kJ mol-l/RT In 10 16.7- 105.9 kJ mol-l/RT In 10 16.7- 103.8 kJ rnol-l/RT In 10 12.9-86.2 kJ mol-I/RT In 10 ~ 12.7 12.3 12.7 12.1 13.7 13.2 13.1 14.8 14.8 14.0 13.9 10.1 a Revised value.SiCl, additions to ethylene have been reinvestigated recently14 because the Taken Statistically adjusted for the previous study3 may have been inaccurate due to small amount of a certain impurity. from ref (1). number of identical sites in the olefins. The starred site shows the reaction centre. method described by O'Neal and Benson.15 As a first step, So[CH,CH2CH(CH,)SiCl,] is calculated as 464 J K-l moI-l employing the bond-additivity relations for silicon compounds.ls* Then corrections are made to obtain So[CH,cHCH(CH,)SiCl,] : the electronic degeneracy was included as ( + R In 2) and the variation in the barriers of the two hindered rotations was taken as 2.1 x 2 J K-l mol-l (a two-thirds reduction in barrier height was assumed). Mass differences, changes in vibration and moment- of-inertia differences were negligibly small and there was no symmetry change.? It follOws that SO(CH,tHCHCH,) = 474 (171) J K-l mol-l.I SiCl, The value in parenthesis is C:, obtained in the same way. Similarly, So(CH,CH2cHCHCH,) = 5 14 (1 94) J K-l mol-1 I SiCl, SO(CH,CH,CHtHCH,) = 5 13 (194) J K-l mol-I I SiC1, * S0[CH3CH,CH(CH3)SiCl3] = 9 x So(C-H) + 3 x So(C-C) + S"(C-Si) + 3 x So(Si-C1) - 3R In 3 t The radical centre was assumed to retain its planar configuration, so that there is no change in the (symmetry) + R In 2 (optical isomer).number of optical isomers.1146 ADDITION OF TRICHLOROSILYL RADICALS and SO (4. -) = 459 (159) J ~ - 1 mol-1 SiCl, were obtained. As the entropies and heat capacities of eSiC1, and the olefins are available in the literature,' we can get AS: at 500 K for each reaction site, and accordingly determine the values of A, using eqn (8). The A, values are also listed in table 2, together with the values previously obtained for 1-olefins. A, can now be calculated from A,. The average value of log(A3/cm3 mob1 s-l) obtained from all the olefins in table 2 is 12.6f 1.0. It is of interest to compare this value with log A, (isobutene) since acetone and isobutene have similar structures.The difference between the two A factors is small but tends to imply that the transition state of isobutene is a little looser than that of acetone. si-c B O N D ENERGY Combining our experimental activation energies with the z-bond energies7 of olefins, the bond energy of Si-C may be obtained as follows: D(Si-C) = g - A z (9) where is the n-bond energy of an individual olefin defined as in eqn (10) : q(CRlR2=CR3R4) = D(H--CR,R,CHR,R,) - D(H-CRlR2CR,R4) = Afl(CR,R,CHR,R,) + Aq(CHR,R,tR,R,) - Afl(CR1R2=CR3R4) - Aq(CHRlR2CHR3R4). (10) The heats of formation in eqn (10) are found in the literature7$ l7 or easily calculated using the group additivity rule.8 The absolute value of E3 is not known at present, but we may assume tentatively that E3 = 12.6 kJ mol-l (3.0 kcal mol-l).Then E, is provided from the table in the abstract. A% is given as E, - E-, + AnRT where AnRT is a conversion factor for the change in standard state; in this case An = - 1. The value of D(Si-C) thus obtained may be classified according to the number of substituents at the carbon to which the trichlorosilyl group attaches: primary, secondary or tertiary. Data from 1-olefins give D(Si-Cpri,) = 352+5 kJ mol-', data from (CH,),C=CHCH, give D(Si-C,,,,) = 325 kJ mol-1 and data from the rest of the 2-olefins give D(Si-C,,,) = 332 & 4 kJ mol-l. D(Si-CPri,) is in good agreement with D(Si-C) = 356f 17 kJ mol-l obtained by Potzinger et aZ.18 from electron-impact experiments combined with thermochemical calculations. D(Si-C,,,) and D(Si-C,,,,) seem a little too small compared with D(Si-Cpri,).All these values may be the upper limit of our data because E3 = 12.6 kJ mol-l, which this bond-energy calculation is based upon, may probably be the lower limit. * REACTIVITY OF *SiC13 RADICALS The rate constants for the addition of SiCl, radicals to olefins are compared with those of some other radicals in table 3. The O(3P) atoms and *NF2 radicals displayT. DOHMARU AND Y. NAGATA 1147 TABLE 3.-&LATIVE RATE CONSTANTSa OF ADDITION OF RADICALS TO OLEFINS - SiC13b 9 NFZC 0 (3P)d CH3e substrate (460 K) (373 K) (298 K) (453 K) CH,COCH, = f =/ -/\ - -/\/ - -/ -\ \=/ \- -\ \=/\ \-/ -\ Q 0.3 1 3.1 3.4 4.1 13.4 4.7 3.7 6.5 16.7 1.4 1 4.4 4.3 19.6 10.3 10.8 33.6 7.9 1 5.8 5.8 1 0.7 0.7 25 1.1 24 28 0.2 0.4 79 0.4 27 a Rate constants per double bond (not per reaction site) were employed for comparison.This work and ref (1). A. J. Dijstra, J. A. Kerr and A. F. Trotman-Dickenson, J. Chem. Soc. R. J. Cvetanovic and A, 1967, 105. R. J. Cvetanovic, Adu. Photochem., 1963, 1, 161. R. S. Irwin, J. Chem. Phys., 1967, 46, 1694. f Reinvestigated value. typical electrophilic trends and the CH, radicals display much less se1e~tivity.l~ SiCl, radicals behave similarly to O(,P) and *NF, for all the 1-olefins. In view of the trend in the 2-olefins, however, the increment in reactivity due to one alkyl group substitution becomes 2 or 3 times smaller than in the 1-olefins for the case of *SiCl, radicals. No such trend is observed in the case of O(,P) atoms. These trends may be interpreted in terms of the steric effect of the bulky *SiCl, radical.The orientations of addition to the unsymmetrical 2-olefins were not affected on changing the reaction temperature from 406 to 456 K; however, this fact alone does not provide any decisive information on the configuration of the transition state. T. Dohmaru and Y. Nagata, J. Chem. SOC., Faraday Trans. 1, 1979, 75, 2617. T. Dohmaru, Y. Nagata and J. Tsurugi, Chem. Lett., 1973, 1031. A. M. El-Abbady and L. C. Anderson, J. Am. Chem. SOC., 1958,80, 1737. J. L. Speier and J. A. Webster, J. Org. Chem., 1956, 21, 1044. R. J. Cvetanovic, R. P. Overend and G. Paraskevopoulos, Znr. J. Chem. Kinet., 1975, S1, 249. S. W. Benson, Thermochemical Kinetics (Wiley, New York, 2nd edn, 1976). N. L. Arthur and T. N. Bell, Rev. Chem. Intermediates, 1978, 2, 37. S. H. Marcus, W. F. Reynolds and S. I. Miller, J. Org. Chem., 1966, 31, 1872. lo J. A. Kerr, in Free Radicals, ed. J. K. Kochi (Wiley, New York, 1973), chap. 1. l1 J. A. Kerr, A. Stephens and J. C. Young, Znt. J. Chem. Kinet., 1969, 1, 371. l 2 R. Walsh and M. J. Wells, J. Chem. Soc., Faraday Trans. I , 1976, 72, 1212. l3 G. A. Russel, in Free Radicals, ed. J. K. Kochi, (Wiley, New York, 1973), chap. 7. l4 T. Dohmaru and Y. Nagata, to be published. l5 H. E. O’Neal and S. W. Benson, in Free Radicals, ed. J. K. Kochi (Wiley, New York, 1973), chap. * J. M. Tedder and J. C. Walton, Acc. Chem. Res., 1976, 9, 183. 17. H. E. O’Neal and M. A. Ring, Znorg. Chem., 1966, 5, 435.1148 ADDITION OF TRICHLOROSILYL RADICALS *' S. W. Benson, F. R. Cruickshank, D. M. Golden, G. R. Hauger, H. E. O'Neal, A. S. Rodgers, R. Shaw and R. Walsh, Chem. Rev., 1969, 69, 279. P. Potzinger, A. Ritter and J. K. Krause, 2. Naturforsch., Teil A , 1975, 30, 347. (Butterworths, London, 1972). 19 J. A. Kerr and M. J. Parsonage, Eualuated Kinetic Data on Gas Phase Addition Reactions (PAPER 1 /753)