NOTES
1075
2.5
2.0
i
u
2 1.5
+
Downloaded by NEW YORK UNIV on September 13, 2015 | http://pubs.acs.org Publication Date: March 1, 1965 | doi: 10.1021/j100887a511
10
1 .o
0.5 2.2
2.4
2 3
2.5
10*/T.
Figure 1 . Temperature dependence of the first-order rate constant: a (left-hand scale), experiments near 1500 mm.; b (right-hand scale), experiments near 100 mm.; 0-, packed vessel, 100 mm. ; Q, packed vessel, 1500 mm.; 0,experiments referred to in footnote 7.
ular energy transfer in activated molecules. The possibility of a decrease in rate at higher pressures has been pointed out by Wilson5 and has been discussed el~ewhere.~In this connection it has also been found that the first-order rate constants observed for the thermal decomposition of cyclobutane up to 1500 mm. appear to show a "conventional" high pressure limit.* Flowers and Frey'O have reported that a change in rate constant does not occur with 1,l-dimethylcyclopropane over the pressure range 16 to 1596 mm. From the presently available information the value of the activation energy obtained from the 1500-mm. experiments ( E , = 32.9 =k 0.7 kcal./mole) may be regarded as approximating the activation energy for the isomerization of cyclobutene a t the high pressure limit. This value is in good accord with that previously determined by extrapolation methods ( E , = 32.7 f 0.8 kcal./mole).a The difference between the values of the activation energy from the 100-mm. and the 1500-mm. data probably represents an experimental variation and not the decrease in activation energy expected when a quasi-unimolecular reaction is well into the falloff region. The isomerization of cyclobutene has an activation energy markedly lower than the values obtained for the decomposition of cyclobutane and its alkyl derivatives
(61-63 kcal./mole) and proceeds in a much lower temperature range. Various factors which might contribute to decreasing the stability of the ring in the case of cyclobutene have been discussed earlier,2,11 and there are several indications of considerable strain in the cyclobutene ring.12 According to Nangia and Benson13 the difference in the strain energy between cyclobutene and cyclobutane needed to account for the lower E. for cyclobutene is rather unreasonable. A suggestion has been made recently that the low activation energy& due to the ionic character of the transition state.14 The activation energies for the isomerizations of several substituted cyclobutenes (31.536.0 kcal./mole) are in the same low range as that observed for cyclobutene and the differences in the rates have been explained in terms of changes in (a) the strain energy, (b) steric repulsion, and (c) resonance stabilization of the activated complex.15 The entropy of activation at 150" calculated for the 1500-mm. experiments from the expression A = Ke(kBT/h) exp(AS*/R) is +0.1 e.u. for K = 1 and +1.5 e.u. for K = 0.5 compared to an over-all entropy change of +4.5 e.u. for the isomerization a t 127" (which is indicated from the entropy values for cyclobutene'6 and 1,3-butadiene1').
Acknowledgment. The authors wish to thank Mr. Carl Whiteman for his assistance and Professor W. R. Moore of the Massachusetts Institute of Technology for a sample of cyclobutene. (9) R. W. Carr, Jr., Ph.D. Thesis, University of Rochester, 1962. (10) M. C. Flowers and H. M. Frey, J . Phgs. Chem., 65, 373 (1961). (11) E. Vogel, Angew. Chem., 66, 640 (1954); 68, 189 (1956). (12) E. Goldish, K. Hedberg, and V. Schomaker, J . Am. Chem. SOC., 78, 2714 (1956); R. C. Lord and D. G. Rea, ibid., 79, 2401 (1957). (13) P. Nangia and S.W. Benson, ibid., 84, 3411 (1962). (14) S.W. Benson, "Advances in Photochemistry," Vol. 2, W. A. Noyes, Jr., G. S. Hammond, and J. N. Pitts, Jr., Ed., Interscience Publishers, Inc., New York, N. Y., 1964, p. 14. (15) H. M. Frey, Trans. Faraday SOC.,60, 83 (1964); E. Gil-Av and J. Shabtai, J . Org. Chem., 2 9 , 257 (1964). (16) A. Danti, J. Chem. Phye., 2 7 , 1227 (1957). (17) J. G. Aston, G. Szasz, H. W. Woolley, and F. G. Brickwedde, ibid., 14, 67 (1946).
The C-C Bond Dissociation Energy in C*F6l by E. Tschuikow-Roux Jet Propulsion Laboratory. California Institute of Technology, Pasadena, California (Received November 9,1964)
Recently, Bibby and Carter2 reported a mass spectrometric study of the dissociation and ion production Volume 69,Number 9 March, 1966
KOTES
Downloaded by NEW YORK UNIV on September 13, 2015 | http://pubs.acs.org Publication Date: March 1, 1965 | doi: 10.1021/j100887a511
1076
in C2F6 and other Quorocarbon gases. From their measured appearance potentials and a quoted value of I(CF3) = 9.35 e.v., which in turn was deduced from measured A(CF3+) in CF3Br3* and D(CF3Br)3band represents an upper limit, Bibby and Carter obtain a surprisingly high C-C bond energy in C2Fs, D(CF3-CF3) = 139.5 kcal./mole. This value is to be compared with earlier mass spectrometric studies as well as other data obtained by various indirect methods. Thus Dibeler, et id.,* deduced D(CF3-CF3) = 124 kcal. /mole from appearance potential measurements in CF3CH3 and C2F6 using a derived value of the ionization potential, I(CF3) = 8.9 e.v. based on their A(CF3+) and A(CHs+), and the known I(CH3). This value of the ionization potential for CF3 is much lower than the measured value of Farmer, et ~ l . who , ~ determined I(CF3) = 10.10 e.v. and also the appearance potentials for CF3+ from CF, and CF3H which lead to D(CF3-F) and D(CF,-H) values consistent with the heats of formation of these compounds. From these data they obtained AHf(CF3)= -117 =k 2 kcal./mole. Using I(CF3) = 10.10 e.v. together with the A(CF3+) from C2F6 reported by Dibeler, et al., leads to D(CF3-CF3) = 97 kcal./nzole, while Bibby and Carter's value of A(CF3+) = 15.4 e.v. gives a bond energy of 122 kcal./mole. On the other hand a value of 69 kcal./mole is obtained if the above AHf(CF3) is used with AHf(C2F6)= -303 kcal./mole.6 Farmer, et al., believed this bond energy to be too low and suggested that AHf(CZF6) may be in error, however subsequent work showed the value to be indeed lower. Thus, D (CF3-CF3)values calculated from kinetic plus thernzocheniical data tend to be lower. Pritchard, et aLI7determined the C-H bond energy in fluoroform from a study of the reversible reaction between CF3 radicals and methane, and the known value for D(CH3-H). Based on this work they deduced the values: AHf(CF3) = -119 and D(CF3-CF,) = 65 kcal./mole. At the time of their work the heat of formation of Quoroform was not yet known and they used an estimated value of - 169 kcal./mole together with AHf(CzF6) = -303 kcal./mole. Using the calorimetric value for A H ~ ( C F ~ = H )- 162.6 kcal./moles leads to a revision of the above quantities: AHf(CF3) = - 112.7 and D(CF3-CF,) = 77.6 kcal./mole. A higher value for the bond dissociation energy in C2F6 IS supported by the recent work of Corbett. Tarr, and Whittlegon the vapor phase bromination of fluoroform. These authors determined L)(CF3-H) = 109.5 kcal./mole (a rather high value) which leads to AHf(cF3) = -105 and D(CF,-CF3) = 93 kcal./mole. However, Pritchard and Thommarsonlo redetermined the C-H bond energy in Quoroforni by the classical kinetic method used The Journal of Physical Chemiatry
previously7 and their result of D(CF3-H) = 102 kcal./ mole corroborates earlier work. Rabinovitch and used the diffusion flame method to study the reaction between sodium atoms and the series of compounds: CH3C1, CH2FCl, CHF2C1, and CF3Cl. From the activation energies of the primary step in the series they evaluated the C-Cl bond energies in these compounds which in turn enabled them to calculate AHr(CF3) = - 119.5 and D(CFs-CF3) = 64 kcal./mole! The above divergency of results of the C-C bond dissociation energy in C2F6 and the heat of formation of the CF3 radical clearly indicates that further work is necessary to resolve this question. Especially desirable would be a direct pyrolytic value for the dissociation energy. Mercer and PritchardI3 have made an attempt to obtain this quantity by pyrolyzing C2Fe in a flow system with toluene as carrier and found the rate constant for the decomposition (eq. 1) to be: k1= CzF6 +2CF3
(1)
108 exp(-50,000/RT) set.-'. In view of the abnormally low frequency factor the authors do not regard the observed activation energy as a measure of the dissociation energy and suggest that the reaction was probably heterogeneous. Thus no significance can be attached to this pyrolysis work. We have studied the kinetics of the thermal decomposition of C2F6 in the presence of hydrogen in a single pulse shock tube under homogeneous reaction conditions in the temperature range 1300 to 1600°K. and de(1) This paper presents results of one phase of research carried out at the Jet Propulsion Laboratory, California Institute of Technology, under Contract No. NSA 7-100, sponsored by the National Aeronautics and Space Administration. (2) M. M. Bibby and G. Carter, Trans. Faraday Soc., 59,2455 (1963). (3) (a) J. Marriott and J. D. Craggs, J . Electron., 1, 405 (1955); (b) A. H. Sehon and M. Sawarc. Proc. Roy. SOC.(London), A209, 110 (1951). (4) V. H. Dibeler, R. 31. Reese, and F. L. Mohler, J . Chem. Phys., 20, 761 (1952). (5) J. B. Farmer, I. H. S. Henderson, F. P. Lossing, and D. G. H. Marsden, ibid.,24, 348 (1956). (6) F. W. Kirkbride and F. G. Davidson, A'ature, 174, 79 (1954). (7) G. 0. Pritchard, H. 0. Pritchard, H. 1. Schiff, and A. F. TrotmanDickenson, Chem. I d . (London), 896 (1955): Trans. Faraday Soc.,
52, 849 (1956). (8) C. A. Neugebauer and J. L. Margrave, J Phys. Chem., 62, 1043 (1958). (9) P. Corbett, A. M. Tarr. and E. Whittle. Trans. Faraday Soc.: 59, 1609 (1963). (10) G. 0. Pritchard and R. L. Thommarson. J . Phys. Chem.. 68, 568 (1964). (11) B. S. Rabinovitch and J. F. Reed, J . Chem. Phys., 22, 2092 (1954). (12) J. F. Reed and B. S. Rabinovitch, J . Phys. Chem.. 61, 598 (1957). (13) P. D. Mercer and H. 0. Pritchard, J . Chem. Soc., 2843 (1957).
NOTES
1077
duced the rate constant for the unimolecular decomposition (eq. 1) to be: kl = 1.7 X 10l8 exp(-94,400/ R T ) sec.-l with an estimated uncertainty of =k4 kcal./mole.'* If we accept an activation energy of 1 to 2 kcal./mole for the recombination of CF3 radica1%15'16 then D(CF3-CF3) 93 kcal./mole and with AHr(C2Fe) = -303 kcal./mole We find AHf(CF3) = - 105 f 2 kcal./mole. These results, together with all previous investigations, are summarized in Table I. Our results are in agreement with the value deduced by Corbett, Tarr, and Whittle, which was based, however, on their unusually high C-H bond energy in CFZH. The above value for the heat of formation of CF3,when used with the now relatively well established D(CF3-H) = 102 kcal./mole, leads to AHf(CF3H) = - 155 kcal./mole, which is in disagreement with the experimental value of -162.6 kca1.l mole reported by Neugebauer and Margrave. It is obvious that the results could be reconciled if either AHf(CF3H) were indeed higher as indicated above, or if AHf(C2F6)were correspondingly lower (-318 kcal./ mole) then Kirkbride and Davidson reported. Thus a redetermination of these quantities may be desirable.
Downloaded by NEW YORK UNIV on September 13, 2015 | http://pubs.acs.org Publication Date: March 1, 1965 | doi: 10.1021/j100887a511
*
Table I : Bond Dissociation Energies and Heats of Formation"
or excitation energy of the ion. However, the value of Farmer, et al., although still high, is in fair agreement with our result. (14)E. Tschuikow-Roux, to be published. (15) P. B. Ayscough, J . Chem. Phys., 24, 944 (1956). (16) G . 0.Pritchard and J. R. Dacey, Can. J. Chem., 38, 182 (1960).
Radiation Yields of Carbon Monoxide and Dioxide for Some Aromatic Carbonyl Compounds by A. A. Miller General Electric Research Laboratory, Schenectady, Sew York (Received October 99,1964)
An aromatic group in an organic molecule can produce opposing effects on its radiation chemistry: (a) increased stability, and (b) activation of specific bonds. These effects are illustrated in this paper by measurements of carbon monoxide and dioxide yields and ratios in the radiolysis of selected carbonyl compounds.
Experimental D(CFaCFs)b
139.5 122d 124 97f 69 65 77.6 93 79 64 93 4
+
D(CF8-H)
AHt(CFdb
..
-82
..
-90.5 -8 9 . 5
..
.,.
102 + 2 102 f 2
..
109.5 + 102.6 ..
..
+
Method
Ref.
Electron impact
C
Electron impact Electron impact
e g
-117 2 - 119* H abstraction i -112.7 1 . 5 - 105 A 2 Bromination of CFIH j -112.1 Calculatedk j,1 - 119.5 Na diffusion flame m -105 2 Pyrolysis n
Unless otherwise stated the following heats of formation were used in these calculations: AHdCzF,) = -303 kcal./mole and AHf(CF3H) = -162.6 kcal./mole. Values are in kcal./mole. Based on Z(CFp) = 10.10 e.v. from ref. 5. e Ref. 4. Ref. 1. Based Based on A(CF3+) = 14.3 e.v. from ref. 4. 0 Ref. 5. on AHf(CF3H) = -169 kcal./mole. ' Ref. 7. Ref. 9. Calculated from D(CF3-Br) = 64.5s and AHdCFaBr) = -149.8 kcal./mole.g Ref. 3b. Ref. 11. This work. a
Unless specified otherwise, Eastman White Label materials were used. Samples were outgassed and irradiated in a gas-measurement cell' with 1.5-Jlev. electrons from a G.E. resonant transfornier unit at an intensity of 20 Mr./niin. and near room temperature. The evolved gas was analyzed in a mass spectrometer for Hz,CO, COz, and CH,. The dosimetry was by an air-ionization chamber' and the radiation yields, G(molecules)/100 e.v., are based 011 an energy absorption of 1 N r . = 6 X l O I 9 e.v./g.
Results and Discussion Table I, the first eight esters are grouped as fo]lows : methyl, phenyl, and benzyl acetates (1-3), b~nZoates(4-6), and phellylacetates (7, 8). The total gas yield, G(gas), was predoluiriantly co and C o n ,except for the acetates, where apprec.iable yields of H2 and CH, were also produced. G'(C0 COY) represents the combined yields of carbon inonoxide and dioxide,
+
I
