Gas-phase Kinetics of the Difluoroamino-radical By A. J. White UNIVERSITY OF WALES INSTITUTE OF SCIENCE AND TECHNOLOGY, CARDIFF CF1 3NU 1 Introduction Reviews of the fluorides of nitrogen,lP2 in general, and the difluoroamino- radical: in particular, appeared in the mid-1960s. At that time the gas-phase kinetic study of the reactions of the difluoroamino-radical was in its infancy -the only reactions studied being those of the hydrogen-abstraction reactions from alkanes4 and acetonea5 Since then various kinetic studies involving the difluoroamino-radical have been investigated by Trotman-Dickenson and his research school at the Edward Davies Chemical Laboratory, Aberystwyth. It is mainly the results of that research which form the basis of this review.The difluoroamino-radical (*NF,) is a versatile radical which undergoes all the usual radical reactions of addition, abstraction, disproportionation, and combination. Furthermore, it acts as a radical trap, and because the difluoro- amino-compounds so formed do not undergo further reaction, this often affords a simple method of studying organic radical decomposition reactions: the decomposition reactions of the propoxycarbonyl radical and the butoxyl radical and a whole range of acyl radical decomposition reactions have been studied in this way, although the results of these decomposition reactions are not con- sidered in this review. The difluoroamino-radical reactions have activation energies which are similar to those found for iodine reactions. The A factors, however, are usually on the low side -the only exception appears to be combination reactions with acyl and alkoxy-carbonyl radicals. This whole question is best viewed in the light of other nitrogen-containing radical kinetics and indeed the results seem to fit in quite well with these.2 Tetrafhorohydrazine and the Difluoroamino-radical A. General Properties.-Tetrafluorohydrazine (N,F,) is a colourless gas, but the liquid phase is often coloured purple, blue, or pink owing to nitric oxide im-purity.l It melts at about 111 K and boils at about 200 K. The vapour pressure can be expressed* by the Clausius-Clapeyron equation (1) : C. B. Colburn, Endeavour, 1965, 24, 138. J. K. Ruff, Chem. Rev., 1967, 67, 665. C.B. Colburn, Chem. in Britain,1966, 2, 336. J. Grzechowiak, J. A. Kerr, and A. F. Trotman-Dickenson,Chem. Comm., 1965, 109. * J. Grzechowiak, J. A. Kerr, and A. F. Trotman-Dickenson, J. Chem. SOC.,1965,5080.* C. B. Colburn and A. Kennedy, J. Amer. Chem. Soc., 1958,80,5004. Gas-phase Kinetics of the DipUoroamino-radical lOgP/mmHg = 6.33 -692 T-' (1) The gas is toxic' and has an odour resembling that of fluorine. Light has been reporteda to induce it to explode. Tetrafluorohydrazinewas first synthesized in 1958 by Colburn and Kennedy& by allowing nitrogen trifluoride to react with a fluorine acceptor, e.g. stainless steel, copper, arsenic, antimony, or bismuth, at 623-673 K. Although it was not realized at the time, this synthesis was actually the synthesis of the difluoro- amino-radical (*NF2) which, on cooling, dimerized to form tetrafluorohydrazine.S The reaction proceeds according to equations (2) and (3) where M = Cu, As, Sb, Bi, or stainless steel: B.Tetrduorohydrazine-Difiuoroamino-radical Equilibrium.-Even though the original synthesis of tetrafluorohydrazine was via the difluoroamino-radical, it was not until two years later that Colburn, Johnson, and their co-workers showeds-l1 the existence of the tetrafluorohydrazine-difluoroamino-radical equilibrium (4): NzF4+ 2 *NF2 (4) Table 1 summarizes their results, together with the mass spectral data given by Herron and Dibeler12 and more recent measurements of pressure variation with temperature at constant volume.18 Table 1 Tetrafluorohydrazine-difluoroamino-radicalequilibrium Temperature AH/ AS1 Method range/K kJ mol-l J mol-l K-' Ref.373-423 83.3 f2.1 167+8 10($) 423-523 85.8f0.8 -13 U.V. -90.8 + 8.4 188k8 10 E.p.r. 34-35 80.8 + 4.2 -11 Mass Spectrometry 32-73 90.0 + 6.7 -12 The existence of this equilibrium is readily appreciated from thermochemical considerations. From the heats of formation of NF3 (-132.8 kJ mol-l),l4 T. R. Carson and F. T. Wilinski, Toxicol. Appl. Pharmacol., 1964, 6,447.* A. P. Modica and D. F. Hornig, Princeton University, Report No. 357-275, October 1963. C. B. Colburn and F. A. Johnson,J. Chem. Phys., 1960,33, 1869. F. A. Johnson and C. B. Colburn, J. Amer. Chem. SOC., 1961,83, 3043. l1 L.H. Piette, F. A. Johnson, K. A. Booman, and C. B. Colburn, J. Chem. Phys., 1961,35, 1481. l* J. T. Herron and V. H. Dibeler, J. Chem. Phys., 1961, 35, 747. l3 G. von Ellenreider, E. Castellano, and H. J. Schumacher, 2.phys. Chem., 1967, 55, 144. l4 L. C. Walker, J. Phys. Chem., 1967, 71, 361. White *NF, (35.6 kJ ~OI-~),~~ the strength of the first N-F and OF(78.9 kJ m~l-~),'~ bond in nitrogen trifluoride can be calculated to be 247.3 kJ mol-l. This value can be compared with the average N-F bond energy of 280.7 kJ mob1in nitrogen trifluoride.16 This indicates that the strength of the two remaining N-F bonds in the difluoroamino-radical must average 297.4 kJ mol-I. Thus it is energetically unfavourable for the difluoroamino-radical to abstract a fluorine atom from tetrafluorohydrazine to form nitrogen trifluoride.This is in contrast to ammonia, where the first N-H bond is the ~trongest.~ C. Structure of Tetrafluorohydrazine and the Difluoroamino-radic1.-The micro-wave~pectrurn~~of tetrafluorohydrazine shows that there are two 'most probable' configurations of tetrafluorohydrazine: one, the planar form, in which the angle between the two NF, groups is 180" and the other, the non-planar form, in which the angle of rotation of one NF2 group with respect to the other is 65". Electron-diffractionlS and infraredlD studies, however, indicate that the tetra- fluorohydrazine molecule exists in the twisted configuration, as predicted by molecular orbital calculations.20 Electron diffraction18 yielded a value of 139 pm for the N-F bond distance and 104" for the F-N-F angle in the NF2 group in tetrafluorohydrazine, and 136 pm and 103" in the difluoroamino-radical.3 Addition Reactions The kinetics of the addition of tetrafluorohydrazine to olefins have been exten- sively st~died.~l-~~ The overall reaction may be written as: NF2 NF2 II R1-C=C-R2 + NzF4 +R1-C -G-R2 II I1 and the proposed mechanism as in equations (5)-(10): NzF4 S 2*NF2 01 + *NF2+*OINF,* *OINF,* -+ 01 + *NF2 *OINF,* + M +*OlNF2+ M l5 H. A. Skinner and G. Pilcher, Quurr. Rev.,1963, 17, 264. l6 Y. N. Inel, Ph.D. Thesis, University of Wales, 1968. l7 D. R. Lide and D. E. Mann, J. Chem. Phys., 1959, 31, 1129. I* R. K. Bohn and S. H. Bauer, Znorg.Chem., 1967,6, 304. I9 M. D. Harmony, R. J. Myers, L. J. Schoen, D. R. Lide, jun., and D. E. Mann, 3. Chem. Phys., 1961, 35, 1129. ao J. P. Simons, J. Chem. SOC.,1965, 5406. a1 A. J. Dijkstra, J. A. Kerr, and A. F. Trotman-Dickenson, J. Chem. SOC.(A), 1966, 582. ** A. J. Dijkstra, J. A. Kerr, and A. F. Trotman-Dickenson, J. Chem. SOC.(A), 1967, 105. A. J. Dijkstra, J. A. Ken, and A. F. Trotman-Dickenson, J. Chenz. SOC.(A), 1967, 864. Gas-phase Kinetics of the Difuoroamino-radical where 01 represents the olefin, *OINF2* is a vibrationally excited radical, and M is any molecule in the system capable of removing excess energy. From steady-state considerations, the rate equation may be written as equation (11) : 1---t[Ol][-NF2] kobs [Product] where t = time.The Arrhenius parameters of reaction (6) are given in Table 2. Table 2 Addition of difiuoroamino-radicals to olefins Olefin Temperature range/K E6/ kJ mol-1 log ml mol-l s-l Ethylene 351428 64.9 10.6 Propene 334-391 57.3 10.2 But-1-ene 334-391 56.9 10.1 trans-But-2-ene 334-391 49.8 9.5 cis-But-2-ene 334-39 1 49.8 9.5 Isobutene 314-373 49.8 9.8 2-Methyl but-Zene 314-373 42.3 9.0 2,3-Dime t hyl bu t -2-ene 314-373 34.7 8.3 Cyclopen t ene 334-391 46.0 8.9 Vinyl bromide 351-405 55.2 9.6 Vinyl chloride 351405 54.0 9.4 There was found to be a linear relationship between Es, the activation energy of the first step, and the ionization potential of the olefin, showing that the olefin which produces the more stable free radical reacts faster.E6 Values also show that substitution of a hydrogen atom by a methyl group in ethylene lowers the activation energy (E6) by 7.5 kJ mol-1 per methyl group. This indicates that the difluoroamino-radical is an electropilic species. Butadiene was found to be more reactive than straight-chain olefins but halogen substitution reduces reactivity. Pankratov et al.24 studied the addition of tetrafluorohydrazine to isobutene and found the overall rate constant k was given by k = 3.2 x lo9exp(-52 300/RT) ml mol-1 s-l Dijkstra et a1.22 found the rate constant of the initial step (6) to be k6 = 9.0 x lo9exp(-49 800/RT)ml mol-l s-l 14 A. V. Pankratov, L.A. Akhanshchikova, and Yu. A. Adamova, Russ. J. Inorg. Chem., 1968, 13, 1513. 20 White Although these rate constants agree fairly well with one another, Pankratov and his workers did not study the effect of constant [MI or constant I.NF2]on the rate constant as Dijkstra et al. had done, and their investigations into the mechanism of the addition were not as thorough as those of Dijkstra et aZ. The values of k,,can be compared with the rate constants for attack on olefhs by other radicals given in refs. 23 and 24. The difluoroamino-radical behaves differently from both methyl and trifluoromethyl radicals. This is because the reactions of methyl and trifluoromethyl radicals with alkenes are largely deter- mined by steric effects. The reactivity of hydrogen atoms with alkenes has been shown by CVetanovic26 to be correlated with atom-localization energies -these also follow a different pattern from that of the diiluoroamino-radical.There is, however, quite good correspondence between the rate constant for the difluoro- amino-radical addition to alkenes and the rate constants for the addition of oxygenas and bromine The reactions of bromine atoms depend upon the establishment of a pre-equilibrium and special factors may determine the rates. Oxygen atoms are electrophilic species that directly at tack the n-electrons of the double bond, as do probably the electrophilic difluoroamino-radicals. The issue regarding 0-or n-complexes between difluoroamino-radicals and olefins is not, however, so clear-cut.Additional important evidence concerns the addition of difluoroamino-radicals to cis- and trans-but-2-ene. The cis-isomer is found in the products of the addition of difluoroamino-radicals to trans-but-Zene and vice versa. This is in accord with the proposed mechanism, involving the re- versible formation of the adduct radical, and is good evidence in support of the mechanism. At the same time it would seem to indicate the formation of a o-complex from addition to the butenes since the isomerization would be less likely from a x-complex. The Arrhenius parameters for the radical decomposition reaction (9) can be calculated from the values of (E, -Elo)and log (A9/Alo) on the assumption that k,, = 13.4 ml mo1-1 s-l. The values are given in Table 3, where it can be seen that in general both log A, and E, decrease by the introduction of sub- stituents into the olefin.4 Hydrogen-abstraction Reactions A. Alkanes.-Grzechowiak et aZ.* found that when a mixture of tetrafluoro-hydrazine and an alkane was heated, the rate of disappearance of the alkane is consistent with the abstraction of a hydrogen atom from the alkane followed by the coupling of the alkyl radical thus formed with a difluoroamino-radical to form an alkyl-difluoroamine, as shown in equations (12)-(14) : N2F4 + 2*NFz (12) RH + *NF2-R*+ HNF, (13) Re + *NF2-+ RNF2 (14) Oa K. R. Jennings and R. J. Cvetanovic, J. Chem. Phys., 1961, 35, 1233. *I R. J. Cvetanovic, Canud.J. Chem., 1960,38, 1678. P.I. Abell, Trans.Furaday SOC.,1964, 60, 2214. Gas-phase Kinetics of the Difluoroamino-radical Table 3 Radical decomposition reactions Temperature log A,/ EOl range/K s-l kJ mol-l C2H4 373-428 12.9 57.3 C3H6 334-391 13.4 56.4 CHa=CHCHgCH3 334-391 12.5 43.1 CH3CH=CHCH3 334-391 14.0 56.8 CH3CH=CHCH3 334-391 14.0 56.8 *C,HioNF, +*NF2 + (CH,),C=CHCH, 314-373 13.3 40.5 C,H,2NF2 7oNF2 + (CHs)2C=C(CHs)2 314-373 12.1 34.7 CyClO-C gH sNF2 ---t .NFa + CyClO-C 5H 8 336391 11.4 38.9 *CaH,BrNFS +*NF2+ C2H3Br 351405 12.3 47.7 CaHSCINF2 +oNF2 + CaHSCl 351-405 10.9 38.5 Arrhenius parameters for this attack were obtained by following the consump- tion of the alkane with time. Recently,28 the rate of hydrogen abstraction from alkanes has been found directly by measurement of the rate of formation of the corresponding alkyl-difluoroamine.The results are given in Table 4. Table 4 Rates of hydrogen abstraction from alkanes by difluoroamino-radicals Temperature log A/ El log k (400 K) range/K ml mol-1 s-l kJ mol-1 (per H atom) Ref. Primary H CSHS 352463 11.80 1 09 -3.18 28 Neopentane 453-555 13.22 112 -2.48 4 Secondary H C3H8 352463 10.39 94.4 -2.25 28 n-C,H 10 352-463 12.29 103 -1.78 28 CyClO-C,H,o 352463 11.36 92.3 -1.64 28 n-C4H1 0 453-555 11.83 92.9 -0.88 4 CyClO-C,H1o 453-555 10.93 83.3 -0.97 4 Tertiary H Isobutane 352463 11.04 86.15 -0.24 28 428-555 10.49 77.29 0.37 4 The results in Table 4 show that the ease of hydrogen abstraction increases in the sequence: primary C-H < secondary C-H < tertiary C-H, as expected.The abstraction of primary hydrogen from neopentane at 400 K is a factor of five faster than from propane when compared on a per hydrogen atom basis. a8 P. Cadman, C. Dodwell, A. F. Trotman-Dickenson, and A. J. White, J. Chem. Soc. (A), 1971,2967. 22 White Grzechowiak's rates6 are on the whole faster than the recent28 ones, although the results for secondary hydrogen abstraction are very similar, especially when allowance is made for the differences in secondary carbon-hydrogen bond energies in propane, n-butane, and cyclopentane. Grzechowiak's method of analysis did not allow for the concurrent attack on the primary hydrogen atoms of n-butane. If this allowance is made by using his primary hydrogen abstrac- tion results, his n-butane results are nearly a factor of ten larger than the more recent results.This factor is also present when the two results for cyclopentane are compared, where no correction is necessary. The difference in the two sets of results may be associated with impurities released by the attack of difluoro- amhe (HNF3 and/or difluoroamino-radicals on the Pyrex glass at the higher temperatures used in the alkane-consumption method. Comparison of the rates (or activation energies) of hydrogen abstraction by difluoroamino-radicals with those of methyl, trifluoromethyl, iodine, and shows that difluoroamino-radicals are less reactive than methyl, trifluoromethyl, and bromine but more reactive than iodine; the Figure shows this.The activation energy of hydrogen abstraction from alkanes has been found previously to be related to the strengths of the carbon-hydrogen bond broken by the semi-empirical Evans-Polanyi equation (15) : E = a[D(R-H) + /3] (1 5) This relationship is also shown in the Figure. A value of a = 0.90 is obtained for the difluoroamino-radical compared with a(.I) = 0.97, a(.Br) = 0.86, and a(.Me) = a(CFJ = 0.49. It seems likely that the same factors which govern the activation energies of hydrogen abstractions by the other species shown in the Figure also govern those of the difluoroamino-radical. The Arrhenius parameters for the reverse reactions, i.e. the attack of alkyl radicals on difluoroamine, can be calculated from the activation energies of the forward reactions together with the enthalpysntropy changes of the reactions by use of equations (16) and (17): E-n= En -AH," -AnRT (1 6) log A-,JAn = (-AS/2.3R) + An log RT + An/2.3 (1 7) The results calculated from the Arrhenius parameters quoted in ref.28 are summarized in Table 5. AHf"(.NFJ was taken13 as 35.6 kJ mol-l, S"(.NFJ as*O D. M. Golden, R. Walsh, and S. W. Benson, J. Amer. Chem. SOC.,1965,87,4053. D. B. Hartley and S. W. Benson, J. Chem. Phys., 1963, 39, 132. 31 P. S. Nangia and S. W. Benson, J. Amer. Chem. SOC.,1964, 86,2773, E. E. Chekhov, A. C. Isailingols, and I. I. Ioffe, Nefrekhimiya, 1967, 7, 717. H. Teranishi and S. W. Benson, J. Amer. Chem. SOC.,1963,85,2887. 34 G. C. Fettis, J.H. Knox, and A. F. Trotman-Dickenson, J. Chem. SOC.,1960, 4177. 36 W. M. Jackson, J. R. McNesby, and B. deB. Darwent, J. Chem. Phys., 1962,37, 1610. A. S. Gordon and S. R. Smith, J. Phys. Chem., 1962,66,521. P. B. Ayscough, J. C. Polanyi, and E. W. R. Steacie, Canad.J. Chem., 1955, 33, 743. 38 P. B. Ayscough and E. W. R. Steacie, Canad.J. Chem., 1956,34, 103. soG.0. Pritchard, H. 0. Pritchard, H. I. Schiff, and A. F. Trotman-Dickenson, Trans. Faraday Soc., 1956,52,849. 40 S. W. Benson, 'Thermochemical Kinetics', Wiley, New York,1968. 23 Gas-phase Kinetics of the Di'uoroamino-radical I I I I 60 400 440 D(R-H)/kJ xnol-' Figure Polanyiplot for X + RH where X = A, *I;By*NF2;C, -Br; D, *Me; E, CF8.The values for *Iand *Brhave been displaced upwards by 10 kJ mol-1 249.8 J mol-1 K-l, AHf"(HNF9 as4' -65.3 kJ mol-l, and So(HNF2)as42 253.1 J mol-1 K-l.The thermochemical values for the alkanes and alkyl radicals were taken from refs. 40 and 43. The only Arrhenius parameters published for a comparable reaction are those for the attack of methyl radicals on ammonia, which has been to have a considerably higher activation energy of abstraction, probably owing to a stronger N-H bond being broken. B. Alkenes.-The Arrhenius parameters for the hydrogen-abstraction reaction from alkenes22~46~46 by difluoroamino-radicals are given in Table 6. But-1-ene produces three products, each with the same molecular weight and each in the same amount. Complete analysis of the three isomers, however, has not been carried In general, these results for the hydrogen abstraction from alkenes are not in agreement with those for the hydrogen abstraction from alkane~,~s~* *l A.V. Pankratov, A. N. Zercheninov, V. I. Chesnokov, and N. N. Zhdanova, Rum. J. Phys. Chem., 1969,43,212. 42 K. Mitteilungen,2.phys. Chem., 1963, 39, 262. S. W. Benson and H. E. O'Neal, 'Kinetic Data on Gas Phase Unimolecular Reactions', NSRDS-NBS 21, Washington D.C., 1970. 44 D. A. Edwards, J. A. Kerr, A. C. Lloyd, and A. F. Trotman-Dickenson,J. Chem. SOC.(A),1966, 621. 46 C. Dodwell, unpublished work. 46 D. G. E. Probert, Ph.D. Thesis, University of Wales, 1966. 47 C. Dodwell, personal communication. Table 5 Summarya of results for attack of alkyl radicals on dijluoroamino-radicals [Reaction (-13)] RH AHi"(RH)/ AHf"(Re)/ S"(RH)I S"(R9I E-131 log A-lJ kJ mol-l kJ mol-l J mol-1 K-l J mol-1 K-l kJ mo1-1 ml mol-1 s-l Propane (p) -104.0 87.9 269.9 286.3 18.0 & 8.4 11.8 4 0.4 (s) -104.0 73.6 269.9 278.8 17.6 k 8.4 9.7 k 0.6 n-Butane (s) -126.2 52.7 310.1 318.4 25.3 2 8.4 11.7 & 0.3 Isobutane (t) -134.5 28.0 294.5 312.1 14.4 k 8.4 10.0 2 0.3 Cyclopentane -77.4 102.1 292.9 301.2 13.7 k 8.4 10.8 k 0.4 a Theerrors are based upon errors in data used.Thermochemical values taken from refs. 40 and 43. Gas-phase Kinetics of the Difluoroamino-radical Table 6 Rates of hydrogen abstraction from alkenes by difluoroamino-radicals Alkene rangell[(Temperature kJ mol-lE/ ml mol-1 s-l log Al Ref.But-1 -ene 334 424 64.0 9.32 45 3-Me thy1 bu t- 1 -ene 3 52-424 60.2 9.20 45 Penta-l,4-diene 352424 46.0 7.7 45 Cyclopentene 334-391 56.1 9.1 22 Cyclohexene 373-405 61.5 10.22 46 the activation energies being some 40 kJ mob1 and the A factors about a factor of 10 lower than those for alkanes. These discrepancies are surprising as the results for the alkenes are based upon the rate of formation of the difluoroamine product, the rate of disappearance of both the difluoroamino-radical and the olefin owing to the addition reaction being taken into account. C. Aldehydes.-Aldehydic hydrogen abstraction by difluoroamino-radicals to form the corresponding NN-difluoroamide and difluoroamine (HNF2) was reported in the early sixties,4* but the reaction was not studied quantitatively until the late sixtie~.~~-~l Aldehydic hydrogen-abstraction reactions by alkyl radicals have been in~estigated,~~-~~ but the corresponding acyl radical decom- Table 7 Aldehydic hydrogen abstraction by difluoroamino-radicals El 81 1% A181 Tempera t urel Aldehyde kJ mo1-I ml mol-l s-l K Ref: Acetaldehyde 69.6 10.37 35348 50 Propionaldehyde 70.7 10.75 373-448 49 n-Butyraldehyde 68.1 10.57 353-423 49 Isobutyraldehyde 65.8 10.31 353-423 49 n-Valeraldehyde 66.7 10.78 353423 50 Isovaleraldehyde 66.7 10.78 353-423 50 R.C. Petry and J. P. Freeman, J. Amer. Chem. SOC.,1961, 83, 3912. 4sP. Cadman, C. Dodwell, A. F. Trotman-Dickenson, and A. J. White, J. Chem.SOC.(A), 1970,2371. P. Cadman, A. F. Trotman-Dickenson, and A. J. White, J. Chem. SOC.(A), 1970,3189. b1 A. J. White, Ph.D. Thesis, University of Wales, 1970. sa R. K. Brinton and D. H. Volman, J. Chem. Phys., 1952,20,1053. 63 G. 0. Pritchard, H. 0. Pritchard, and A. F. Trotman-Dickenson, J. Chem. Phys., 1953, 21, 748. 64 P.Ausloos and E. W. R. Steacie, Canad. J. Chem., 1955,33,31. sb R. E. Dodd, Canad. J. Chem., 1955,33, 699. E.~R. N. Birrell and A. F. Trotman-Dickenson, J. Chern. SOC.,1960,2059. 67 R. E. Dodd and J. W. Smith, J. Chem. SOC.,1957, 1465. ti8 D. H. Volman and R. K. Brinton, J. Chem. Phys., 1954,22, 929. 59 J. A. Kerr and A. F. Trotman-Dickenson, J. Chem. SOC.,1960, 1611. WJ J. A. Kerr and A. F. Trotman-Dickenson, Trans. Faraduy SOC.,1959,55, 572.61 J. A. Kerr and A. F. Trotman-Dickenson, Trans. Faraduy SOC.,1959, 55, 921. a* J. A. Kerr and A. F. Trotman-Dickenson, J. Chem. SOC.,1960, 1602. E. L. Metcalfe and A. F. Trotman-Dickenson, J. Chem. Soc., 1960, 5072. 26 Table 8 Summarya of results for attack of acyl radicals on dijluoroamine [reaction (-lS)] AH!" AH!" R (RCHO)/ Ref. (RCO)/ Ref. S"(RCHO)/ Ref. S"(ReO)/ Ref. AH,,/ E-,$ log &d kJ mol-l kJ mol-l J mol-l K-l J mol-l K-l kJ mol-l kJ mol-l mlm01-i r1 Me -165.9 64 -24.2 64 263.8 64 265.4 40 41.0f8.4 28.4f8.4 10.lfl.O Et -190.6 65 -41.8 estimated 306.8 66 304.7 estimated 48.1 f12.5 22.6k 12.5 10.7fl.0 Prn -204.4 65 -62.3 estimated 345.3 66 344.0 estimated 41.4f 12.5 26.8i- 12.5 10.4f 1.0 Pri -216.1 estimated -71.1 estimated 349.9 estimated 341.9 estimated 41.0f 12.5 24.7 f12.5 10.6f 1.0 Bun -227.8 estimated -86.5 estimated 391.2 estimated 383.7 estimated 40.5 f12.5 32.6f 12.5 11.3 f1.0 Bui -236.6 estimated -94.9 estimated 389.2 estimated 381.6 estimated 41.0f12.5 25.9f 12.5 11.Ofl.O a The errors are estimates based upon the errors of the data used.Estimated values based upon the additivity rules in reference 40. AHf"(*NF,) taken as 35.5 kJ mol-l (ref. 13) and AHfo(HNF,) as -65.2 kJ mol-l (ref. 41).S"(0NF.J taken as 249.5 J mol-l K-' (ref. 40) and So(HNFz)as 252.9 J mol-l K-' (ref. 42). O4 J. A. Devore and H. E. O'Neal, J. Phys. Chem., 1969,73,2644. 65 E. Buckley and J. D. Cox, Trans. Faraday SOC.,1967, 63, 895. I. A. Vasil'ev and A. A.Vvedenskii, Russ. J. Phys. Chem., 1966, 40,453. Gas-phase Kinetics of the Dijluoroamino-radica I position reaction was not studied because of the complexity of the systems. Aldehydic hydrogen abstraction by difluoroamino-radicals, however, afforded a simple method of producing the acyl radical and, because the difluoroamino- radical acts as a ‘radical-trap’, the acyl radical decomposition reactions were studied quantitati~ely.~~-~~ The suggested mechanism is given in equations (1 8)--(23) : RCHO + *NF2 RCO + HNF2 (18)---f Re0 + M -+ RCO* + M (19) RCO* + M +Re0 + M (20) RCO*4 R*+ CO (21) R* + *NF, +RNF, (22) Re0 + *NFz+RCONF, (23) The results for the aldehydic hydrogen-abstraction reaction by the difluoroamino- radical are given in Table 7.The A factors for the attack of difluoroamino- radicals on aldehydes are less than those for the corresponding attack by alkyl radical^.^^-^^ Low A factors for hydrogen abstraction by difluoroamino-radicals appear to be the norm rather than the exception (for a more detailed discussion of this see Section 5). Table 7 shows that the Arrhenius parameters are very similar for one particular radical attacking a series of aldehydes. This is indica- tive, though not conclusive, of constant RCO-H bond energies in this series of aldehydes. This is also indicated by the fairly consistent values of AHl8 in Table 8. As AH, = D(F,N-H) -D(RC0-H), and D(F2N-H) is a constant, AH, 8 will be constant if D(RC0-H) is independent of the R group.The average value of AHl8 is 42.2 kJ mol-l, hence the mean aldehydic bond strength in this series of aldehydes is 361 kJ mol-1 [assuming50 that D(F,N-H) is 319 kJ mol-l]. AHl8 is also related to the activation energies of the forward and back reactions of reaction (18) by AHlg = E18-E-ls (An = 0). Hence can be estimated (see Table 8). The A factors of the forward and back reactions of reaction (18) are also related by log (AlE/A-lg) = AS18/2.3R, hence log A-18 can be estimated (see Table 8). Very few abstractions are known with which the values calculated for Eland log can be compared. The energy of abstrac- tion of hydrogen from hydrogen iodides7 by acetyl radicals has been found to be 6.3 kJ mol-l. From thermochemical considerations, the reactions of acyl radicals with difluoroamine might be expected to have slightly higher activation energies as the bond in difluoroamine is about 20 kJ mol-1 stronger than in hydrogen iodide.The calculated values of are in the region 20-30 kJ mol-1 and therefore seem plausible. The abstraction of hydrogen from hydrogen bromides8 by acetyl radicals does not fit the calculated results, but as this reaction was only inferred and not measured directly, the results might be considered suspect. 67 H. E. O’Neal and S. W. Benson, J. Chem. Phys., 1962, 37, 540. 6* M,J, Ridge and E, W, R. Steacie, Canad.J. Chem., 1955, 33, 383. 28 White D. Ketones.-Table 9 summarizes the results of the hydrogen abstraction by difluoroamino-radicals from ketone^.^^^^^^^ The Arrhenius parameters for this Table 9 Hydrogen abstraction by difluoroamino-radicals from ketones Ketone range/K Temperature kJ mol-1 E/ ml mol-1 s-I 1% Al Ref.Me2C0 451-553 81.3 10.7 5 Et ,CO 453-555 72.5 10.2 69 Pr'COMe 453-555 79.2 10.8 70 attack were obtained by following the consumption of ketone with time. Both the A factors and the activation energies are lower than those obtained for the attack of difluoroamino-radicals on alkanes4 using this same method, and they are considerably lower than those obtained by measuring the rate of formation of the alkyl-difluoroamine.28 This indicates that the method of measuring the rate of disappearance of the ketone must be suspect, probably owing to im-purities released by the attack of difluoroamine and/or the difluoroamino- radical on the Pyrex glass at the higher temperatures employed.E. Formafe~.-Thynne~~-~~investigated the decomposition of alkyl formates by methyl radical photosensitization. He concluded that the formyl hydrogen atom was attacked exclusively and that the decomposition of the aikoxy-carbonyl radical so produced was a good 'thermal' source of alkyl radicals. Grotewold and KerrY7* however, claimed that abstraction from the alkoxy-group of n-propyl formate occurs to a significant extent. Arthur and Gray7B have employed the use of isotopic labelling to determine the position and extent of hydrogen abstraction from the formyl and methoxy sites in methyl formate by methyl and trifluoro- methyl radicals.They found that attack was principally at the formyl group, but at 455 K a significant proportion of abstraction occurred from the methoxy- group. Similar conclusions were reached by Donovan et aZ.76 using methyl and [2H,]methyl radicals and methyl formate and methyl [2H]formate. The reaction of n-propyl formate and the difluoroamino-radical has been investigated," the proposed mechanism being given by equations (24-27) : *NFa+ HCO,Prn -+ HNF2 + *C02Prn (24) J. Grzechowiak, Roczniki Chem., 1966, 40,895. 70 J. Grzechowiak, Chem. Stosowana (A), 1967, 11, 215. 71 J. C. J. Thynne, Trans. Faraday SOC.,1962, 58, 676. J. C. J. Thynne, Trans. Faraday SOC.,1962, 58, 1394. 73 J. C. J. Thynne, Trans. Faraday SOC.,1962, 58, 1533.74 J. Grotewold and J. A. Ken,J. Chem. SOC.,1963,4342. 76 N. L. Arthur and P. Gray, Trans. Faraday SOC.,1969,65,424. T. R. Donovan, W. Dorko, and A. G. Harrison, Canad. J. Chem., 1971,48, 828. 77 P. Cadman, A. J. White, and A. F. Trotman-Dickenson, J.C.S. Faraday I, 1972, 68, 506. Gas-phase Kinetics of the Difluoroamino-radical Prn-+ *NF2+ PrnNF2 (26) *NF2+ *C02Prn-+ NF2C02Prn (27) The rate equation was found to be given by log kza(in ml mol-l s-l) = 8.48 & 0.88 -(77 800 k 7200)/2.3RTwhere R = 8.314 J mol-l K-l. The A factor for this formyl hydrogen abstraction reaction by difluoroamino- radicals is about lo2 lower than a ‘normal’ A factor for hydrogen-abstraction reactions by difluoroamino-radicals. Abstraction of the formyl hydrogen by methyl radicals has also been found to have a lower A factor than found for other methyl radical reactions.No hydrogen abstraction from the n-propoxy-group was observed even at the higher temperatures -the product was looked for but was not found. Because of the unusually low A factor, the results were tested fairly rigorously for consistency with the proposed mechanism.77 Although the extent of methyl radical attack on the alkoxy-group has not completely been resolved, it seems certain that the formyl hydrogen is much more reactive than the alkoxy-group. It is expected that difluoroamino-radicals, which are much less reactive than methyl, would be more discriminating and hence would be even less likely to attack the n-propoxy-group.This is confirmed by the absence of any products from this reaction. Comparison of the activation energies for the attack of meth~l,~l-~~ iodine,78 and difluoroamino-radicals on alkyl formates shows that difluoroamino-radicals are slightly more reactive than iodine atoms but much less reactive than methyl. This is the same order of reactivity found for the attack on alkanes4s2* and aldehyde^.^^^^^ 5 Combination Reactions The cross-combination reaction of difluoroamino-radicals with ethyl and isopropyl radicals was investigated’ by photolysing the corresponding dialkyl ketone in the presence of very small concentrations of tetrafluorohydrazine, and hence the difluoroamino-radical. The results can be discussed in terms of the mechanism in equations (28)-(32): RCOR + hv -R* + RCO+ RCO+ -R* + CO (28) (29) R + R*-tR-R (30) R*+ *NF24RNFa (31) Rt‘O + *NF2--+ RCONF2 (32) where RCO+ is vibrationally excited.The mechanism involved in the photolysis of both diethyl and di-isopropyl 78 R. K. Solly and S. W. Benson, Internat. J. Chem. Kinetics, 1969,1,427. 70 P. Cadman, Y. Inel, A. F. Trotman-Dickenson, and A. J. White, J. Chem. SOC.(A), 1971, 1353. White ketones is well knowneo and occurs via reactions (28)--(30). Reactions (31) and (32) are invoked to explain the products found in the presence of difluoroamino- radicals. On photolysing diethyl ketone in the presence of small concentrations of the difluoroamino-radical above 373 K, the only products found were n-butane and NN-difluoroethylamine. No NN-difluoropropionamide was formed.From the proposed mechanism, and GH 5' I = (Rc~H~~)*/k3 0 * (34) Substituting for ethyl in (33) and integrating between the limits [.NF,]i at time = 0 and [.NF& at time = t gives As [N2F41i= [N2F41f+ HCZHJWI (36) and the initial and final concentrations of [.NF2] are related to the concentrations of [NzF4], k31 can be calculated from equation (35). Below 360 K appreciable amounts of NN-difluoropropionamide were formed by reaction (32) occurring as well as reaction (29). The above method could not then be used to calculate k31. In these runs the conversion of tetrafluoro- hydrazine was kept to less than 10% and equation (37) was used to calculate k 31, where [.NF,] Bv is the average concentration of difluoroamino-radicals : [.NF2lav (37)ktu = (Rcpp,) ~~O+/(RC~HJ' WAIi = [NAIf + HC2HSNFzl + &EC2H5CONFzl (38) In all cases good agreement existed between [N2F4]i calculated by equations (36) or (38) and that measured on the gas burette, although the calculated values were used as they were thought to be more accurate.The interference of the disproportionation reaction (39) in regard to the use of (36) or (38) is Photolysis of diethyl ketone in the presence of varying ratios of difluoroamino-radicals and tetrafluorohydrazine indicates that reaction (40)was not important. CIH5. + *NFz--t C2H4 + HNFS (39) C2H5* + N2Fp +C2H5NFa + *NFs (40) J. C. Calvert and J. N. Pitts, jun., 'Photochemistry', Wiley, New York, 1966, pp.396,402. P. Cadman, Y.Inel, and A. F. Trotman-Dickenson, J. Chem. SOC.(A), 1971, 2859. 31 2 Gas-phase Kinetics of the Di’uoroamino-radical The rate of attack of difluoroamino-radicals on diethyl ketone [equation (41) J is smallss in this temperature range and so can be neglected. *NF2+ CzH ,COC2H HNF2 + czH4COC2H (41)--f Using Hiatt and Benson’s values2 of ml mol-l s-l for k30, log kS1 (in ml mol-l s-l) = (8.2 _+ 0.5) -(1100 f 3400)/2.3RTwhere R = 8.314 J mol-l K-l. The results obtained for the di-isopropyl ketone photolysis in the presence of small concentrations of difluoroamino-radicals showed that no NN-difluoro- butyramide was detected even at room temperature, so presumably any iso- butyryl radicals formed as in equation (28) decompose by equation (29).This is supported by the mass balance of difluoroamino-radicals obtained using equa- tion (36) above. k31 was calculated using equations (35) and (36) together with Hiatt and Benson’s values3 of ml mol-l s-l for k30. The disproportionation of isopropyl and difluoroamino-radicals [equation (42)] has been founds1 to be much smaller than the combination [equation (31)] and can be disregarded by comparison with equation (34). i-C3H,* + *NF2-HNFa + C3H6 (42) Least-mean-square analysis of the results gave log k31 (in ml mol-l s-l) = (9.2 f 0.3) -(5300 -1700)/2.3RTwhereR = 8.314 J mol-l K-l. The results show that the rate of combination of both ethyl and isopropyl radicals with diiluoroamino-radicals is much smaller than that for ethyl-ethyl and isopropyl-isopropyl recombinations.They are, however, in the same region as that for t-butyl-t-butyl radical re~ombination.~~ The small activation energies are probably not significant considering the experimental error. From the collision diameters of ethylyS5 isopropyl,s6 and diiluoroamino- radicalss7 the collisional efficiency for combination of these alkyl radicals with difluoroamino-radicals can be calculated to be 104-10-6, compared with the value of about obtained for small alkyl-radical combinations.s2-84 This slower rate of combination involving difluoroamino-radicals was predicted by Simonsao from MO calculations. He concluded that the unpaired electron, being in a 2bl.rr-orbital perpendicular to the molecular plane and held near the nitrogen by the inductive effect of the fluorine atoms, hampers the reactions of difluoro-amino-radical.Suitable orientation and close contact must both occur before any orbital overlap is possible. This factor also explains the low pre-exponential fact or obtained in abstract ion reactions of difluor oamino-radical . Table 10 shows the results for difluoroamino-radical Combination reactions. The acyl radical-difluoroamino-radical combination must have a value close to the collision frequency in order to give reasonable A factors for the acyl radical R. Hiatt and S. W. Benson, J. Amer. Chem. SOC.,1972, 94, 6886. as R. Hiatt and S. W. Benson, Internut. J. Chem. Kinetics, 1972, 4, 151.R. Hiatt and S. W. Benson, Internat. J. Chem. Kinetics, 1973, 5, 385. 86 H. S. Johnston, ‘Gas Phase Reaction Rate Theory’, Ronald Press, New York,1966, p. 153. J. S. Rowlinson, Quart. Rev.,1954, 8, 168. 87 L. M. Brown and B. deB. Darwent, J. Chem. Phys., 1965,42,2158. White Table 10 Difluoroamino-radical-radical combination reactions Radical Temperature1 Rate constant1 Ref. K ml mol-1 s-I *NFo 400 3 x 1Ol0 87 *Et 297-448 1.6 x lo8 79b *Pri 29748 log k = 79b 9.2 -530012.3RT Re0a Prnoco 353-8 398-463 1014 1013 -1014 49, 50 77 ButO* 373423 3.16 x 1O1O 88 a R group in RdO = Et, Prn, Pri, Bun, or Bui.* Recalculated using refs. 82 and 83. decomposition and formation reaction^^^^^^ and to correlate with previous result^.^^^^^ It would therefore be surprising if this rate was much less than 1014mlmo1-1 s-l.These arguments also apply to the combination of n-propoxy- carbonyl radicals with difluoroamino-radicals.77If this rate was much lower than lo1*,then the A factor for the n-propoxycarbonyl radical decomposition would be low, whereas evidence7* favours a normal value for this type of radical decom- position. The rate of combination of t-butoxyl and difluoroamino-radicals has been assumed8* to be 1010-6ml mol-l s-l. In order to rationalize these different values and the results of Simons’ calculations,20 it is possible that combination rates are slow except where there is a carbonyl multiple bond adjacent to the odd electron.The carbonyl x-orbital may be able to overlap much more easily with the 2b, n;-orbital of the ditluoro- amino-radical than can the o-orbital of the free electron. The rates of combination reactions of nitric oxide, nitrogen dioxide, and arnino-radical~*~~~~~~~-~~with themselves and with alkyl and alkoxyl radicals have also been found to be in the range 109-1012 ml mob1 s-l, i.e. similar to the difluoroamino-radical. BensonQ3 has discussed these low rates. The results presented in this review support his suggestion that the combination rates of species containing the unpaired electron on a nitrogen atom are often anomalous. It is unfortunate that no acyl radical-nitric oxide or -nitrogen dioxide cross- Combination reactions have been reported with which to compare the acyl radical-difluoroamino-radical combination and its suggested high value.The entropies of NN-difluoroethylamine and NN-difluoroisopropylamine can a* P. Cadman, A. F. Trotman-Dickenson, and A. J. White, J. Chem. SOC.(A), 1971, 2296. J. A. Kerr and A. C. Lloyd, Trans. Faraday SOC.,1967, 63,2480. eo H. E. O’Neal and S. W. Benson, J. Chem. Phys., 1964,40, 302. 91L. Phillips and R. Shaw, ‘10th International Symposium on Combustion’, Pittsburg, Pennsylvania, The Combustion Institute, 1964, p. 453. 92 I. M. Napier and R. G. W. Norrish, Proc. Roy. SOC.,1967, A299,313. e3 W. C. Sleppy and J. G. Calvert, J. Amer. Chem. SOC.,1959, 81, 769. e4 M. I. Christie and J. S. Frost, Trans. Faraday SOC.,1965, 61, 468. e6 D.L. Cox,R. A. Livermore, and L. Phillips, J. Chem. SOC.(B), 1966, 245. O8 T. Carrington and N. Davidson, J. Phys. Chem., 1953,57,418. Gas-phase Kinetics of the Difluoroarnino-radical be calculated from that of NN-difluoromethylaminee7by bond additivity principles to be 321.3 and 358.6 J mol-1 K-l. Using the known entropiesPo of ethyl, isopropyl, and difluoroamino-radicals, AhS31O can be calculated to be -178.7 and -170.3 J mol-1 K-l at 298 K, respectively (standard state 1 mol ml-l). From the pre-exponential factors given above for equation (31), log A-31 can then be calculated as 13.9 (R = Et) and 14.4 (R = Pri). These values are slightly lower than usually found for the decomposition reactions of compounds into radicals. 6 Disproportionation Reactions The disproportionation reactions of alkyl radicals to give alkenes and alkanes are well known,98-100 although the question whether this reaction takes place via the same transition state as occurs in combination or via a different one has not yet been settled.It has been suggested that loose bending frequencies occur- ring in the combination transition state are responsible for the occurrence of disproportionation reaction^.^^^^^^ Bensonlo2 has suggested an ionic transition state for disproportionation, different from that for combination. Disproportionation reactions between radicals other than two alkyls have not been as well studied, although they have been reported to occur between alkyl radicals and nitric oxide,lo3 alkyl and amino-radical~,~~~~~~~ aIkoxyl radicals and nitric oxide,lo6-ll1 and also alkyl and difluoroamino-radicals.*l This 1at ter reaction was studied by photolysing di-isopropyl ketone and methyl t-butyl ketone in the presence of tetrafluorohydrazine.The photolysis of di-isopropyl ketone is well known,80 and the products formed in the presence of tetrafluorohydrazine can be explained in terms of reactions (43)-(46) : (i-C3H7)2C0+ hu -2 i-C3H7*+ CO (43) i-C3H7*+ *NF2+i-C3H7NF2 (45) -C3H6 + HNFa (46) 97 L. P. Pierce, R. G. Hayes, and J. F. Beecher, J. Chem. Phys., 1967, 46,4352. 98 J. A. Kerr and A. F. Trotman-Dickenson, Progr. Reaction Kinetics, 1961,1, 107. ss A. F. Trotman-Dickenson and G. S. Milne, ‘Tables of Bimolecular Reactions’, NSRDS- NBS9, Washington D.C., 1967.looE. Ratajczak and A. F. Trotman-Dickenson, ‘Supplementary Tables of Bimolecular Gas Reactions’, UWIST, Cardiff, 1970. lol J. N. Bradley, J. Chem. Phys., 1961, 35, 748. loaS. W. Benson, Adv. Photochem., 1964, 2, 1. lo3J. Heicklen and N. Cohen, Adv. Photochem., 1968, 5, 284. Io4 W. E. Groth, U. Schurath, and R. N. Schlindler, J. Phys. Chem., 1968,72, 3914. Io5 U. Schurath, P. Tiedemann, and R. N. Schlindler, J. Phys. Chem., 1969,73,456. In8E. A. Arden, L. Phillips, and R. Shaw, J. Chem. Soc., 1964, 5126. lo7 R. A. Livermore and L. Phillips, J. Chem. SOC.(B), 1966, 640. Io8 G. R. McMillan, J. Amer. Chem. SOC.,1961, 83, 3018. log R. F. Walker and L. Phillips, J. Chem. SOC.(A), 1968, 2103. IlnR.L. East and L. Phillips, J. Chem. SOC.(A),1970, 331. G. R. McMillan, J. Amer. Chem. Soc., 1962, 84,2514. White Now difluoroamino-radicals add to propene to give 1,2-bis(difluoroamino)pro-panea1 [reaction (47)] and this reaction consumes between 0 and 15% of the propene formed in reaction (46), depending on the temperature. The rate of addition has been shown to be limited by the addition of one difluoroamino- radical to the double bond of the alkene and this rate is given by equation (48), where kq7is a composite rate coefficient calculated from the rate coefficients of the individual steps of the mechanism and depending on the total pressure of the system. d[C aH 6(NF2)21 dt As propene is formed in reaction (46) and consumed in reaction (47), d(.NF2, -Pri) is corrected for the loss of propene via reaction (47) in equation (49) : Values of k4, were calculated from the rate coefficients of the individual steps in the addition of difluoroamino-radicals to propene which have been published previously.21 The consumption of propene in terms of percentage loss calculated from relation (49) was found to be significant only at the higher temperatures and never > 15 %.Variation of the concentrations of di-isopropyl ketone and difluoroamino- radicals was found to have no effect on the values of d(-NF2,*Pri). The reaction of isopropyl radicals with tetrafluorohydrazine itself can be neglected.79 The reaction of ethyl radicals with tetrafluorohydrazine [equation (SO)] was shown to be unimportant.CaH6. + NzF4 -t C2HbNF2 + *NF2 (50) d(*NF2,*Pri) was found to be given by equation (51): d(*NF,, *Pri) = (0.0535 f 0.005) exp[+ (595 k 300)/RT] (51) Methyl t-butyl ketone-tetrafluorohydrazinemixtures were photolysed between 291 and 485 K. The products from the reactions of methyl radicals or R-C=O (where R = Me or But) with difluoroamino-radicals do not interfere with the analysis or reaction scheme and were not separated in the chromatography. The concentration of difluoroamino-radicals was kept much larger than the con- centration of the alkyl radicals to eliminate the reactions between alkyl radicals. No trace of products arising from the reactions between methyl and t-butyl radicals was found. The products, isobutene and t-butyldifluoroamine, can be explained in terms of the same scheme as before -equations (52)--(54): Gas-phase Kinetics of the Difluoroamino-radical t'C4HsCOCHs + hv +t-C,Hs* + CMSCO' (52) t-C,H,* + 'NFB -+ t-C4HSNFz (53) C4Hs + HNFo (54) The rate of addition of difluoroamino-radicals to the product isobutene is faster than the rate of addition to propene.The correction for the loss of isobutene via reaction (47) was therefore larger than in the case of propene but was made using the same method. No Arrhenius parameters were reported for the addition of difluoroamino- radicals to isobutene22 but the reaction was studied up to 373 K. The large values obtained for kp7necessitated larger oleb to difluoroamino-radical ratios to be able to measure kp7and this resulted in telomerization.The olefin to difluoro- amino-radical ratio is much lower and so telomerization is less likely. The results obtained for the addition of difluoroamino-radicals to isobutene were extra- polated. d(-NFz, *But) was found to be given by equation (55): A(.~B,*But)= (1.98 k 0.20) exp[-(9140 & 330)/RT] (55) The disproportionation-combination ratio between ethyl and difluoroamino- radicals could not be measured with the experimental arrangement used for isopropyl and t-butyl radicals because the disproportionation product, ethylene, has the same molecular weight as the carrier gas, nitrogen. From the results it is likely that this ratio is probably also small. The disproportionation-combinationratio of isopropyl radicals with difluoro- amino-radicals is less than that of the corresponding ratio of t-butyl radicals, which is as expected if the relationship which has been found for alkyl radicalsBs is also true for these pairs of radicals.This relationshiplos showed that d per hydrogen atom available for transfer increased from isopropyl to t-butyl and is also consistent with a lower value of d(-NF,, *Et). The experimental results showed that d(.NF,, *hi)is nearly independent of temperature, i.e. the difference in activation energies -E,J is nearly zero within the experimental error. This independence of temperature of d is the same as has generally been found previously for the disproportionation-combination of alkyl-alkyl radicals.has been measured70 and found to be 5.28 kJ mol-l, so E,, is probably in the region 4-6 kJ mol-l. The activation energy difference -Ess for t-butyl radicals is 9.1 kJ mol-l. This value is obviously very dependent upon the correction for isobutene con- sumption by difluoroamino-radicals, which is much larger than for propene. Examination of a graph of log d(.NF2, -But) vs. 1/T showed that the higher- temperature points lie near the line drawn through those obtained at lower temperatures, where the correction for olefin loss is small. This indicates the general validity of the correction used. The absolute rate of disproportionation of t-butyl radicals and the activation energy for this reaction could be found because the corresponding rate of combination has not been measured.As the White activation energy difference between disproportionation and combination is not the same in the case of the t-butyl as in the isopropyl radical, it would be interest- ing to find whether it is disproportionation or Combination which causes this dissimilarity and has a different activation energy. From the rate of combination of isopropyl and difluoroamino-radicals [equation (56)] the absolute rate coefficient for disproportionation of isopropyl and difluoroamino-radicals can be calculated [equation (57)] : log k,, (in ml mol-l s-l) = 9.2 -5300/2.3 RT (56) log kd8(in ml mol-1 s-l) = 9.15 -5900/2.3 RT (57) The combination rate of both ethyl and isopropyl with difluoroamino-radicals has been found to be much slower than the rate of combination of alkyl-alkyl These results show that the disproportionation rate is slightly slower than combination.Similar slow rates of both disproportionation and combination have also been found for alkyl and alkoxyl radicals with nitric oxide. This seems to indicate a connection somehow between the transition state for disproportionation and combination, as slow combination rates have even slower disproportionation rates. This may, of course, just be coincidence or it is possible that these reactions occur via an energized molecule undergoing molecular elimination [equations (5f9-W) 1: CHS CH3* I 1 H8G-C. + *NF9-+ HSG-G-NF, I 1 There was no indication of any pressure dependence of d occurring although this was not specifically looked for.A comparison of the combination-disproportionationresults obtained for nitrogen-containing radicals is shown in Table 11. The values obtained for the difluoroamino-radicals are generally less than those obtained for nitric oxide and amino-radicals. The difference in entropies of the disproportionation and combination Gas-phase Kinetics of the Difluoroamino-radical products (zSou-zSocomb) has been found to be related to A(alky1, alkyl) radicals.lol The results for A(*NF,, R) obtained here fall near the same line as that for A(alky1, alkyl) (within the scatter obtained for alkyl radical dispropor- tionations). Table 11 A Valuesfor nitrogen-containing radicals Radicals A Ref.-NF, *Pri 0.064 81 *NF2 *But 0.123 81 0.31 105 0.21 104 NO Me00 0.5 106 NO EtO- 0.3 106 0.45 107 NO PrnO* 0.4-0.5 110 NO PriO- 0.15 108 0.19 108 NO Bu*O* 0.26 109 NO ButO* 0 111 7 Di-t-butyl Peroxide Pyrolysis The kinetics and pressure dependence of the decomposition of t-butoxyl radicals have been studied in the gas phase between 373 and 423 K by pyrolysing di-t- butyl peroxide (DTBP) in the presence of difluoroamino-radicals.88The results for the pyrolysis of DTBP were consistent with those recommended by Shaw and Pritchard,l12 and the rate of decomposition of the butoxyl radical was similar to the value suggested by O’Neal and Benson,@ showing that the difluoroamino- radical may be used as an effective radical trap.8 Miscellaneous Thermochemistry Inells calculated that So(total) (CH,NF,) = 280.2 J mol-1 K-l and So(P.A.C.)* (CH,NF2) = 280.7 J mol-l K-l Hence So(CH,NFJ = 280.5 J mol-l K-l Now using Benson’s additivity rules,40 * P.A.C. = partial atomic contributions. 11* D. H. Shaw and H. 0. Pritchard, Canad.J. Chern., 1968,46,2721. White S"(CHSNF2) = So(C -(N)(H)3) + S"(N -(C)(F),} S"(N -(C)(F)2} = S"(CHSNF2) -So{C-fN)(H),} = 280.5 -127.2 = 153.3 J mol-1 K-l Using this value of S"{N-(C)(F)2}, Table 12 can be drawn up; Table 13 may also be constructed. Table 12Estimated entropy values for NF2 compounds Sol Sol Compound J mol-l K-l J mol-l K-l (P.A.C.)b (P.B.C.) CH3NFz 280.5a 280.7 275.3 C*H SNF, 321.3a 320.1 310.0 n-C,H,NF, 360.7" --iso-C,H ,NF2 358.6a 350.2 336.0 n-C,H gNFa 400.0a --iso-C,H ,NF, 397.9a --t-C,H gNF2 392.0a 380.7 370.7 a Values estimated using Benson's additivity Errors are probably not more than f6.3 J mol-' K-1.These values based on partial atomic contributions (P.A.C.) and partial bond contributions (P.B.C.) are taken from Inel." Table 13 AHf" Values and D(C--N) values" Compoundc AHi'l "-N/kJ mol-l kJ mol-l CH3NF2 -74.9 253 C2H SNFZ -102.5 247 n-C 3H ,NF, -123.0 247 iso-C3H,NF -138.9 249 n-C4HgNFz -143.9 247 iso-C,H gNF2 -149.4 243 t-C,H gNF2 -172.4 239 a Errors probably not more than f11.8 kJ mol-'. D(C-N) values based on N -(C)(F), = 32.6 kJ mol-l. Applied a gauche correction of 3.3 kJ mol-' owing to (CH,), and NF2 on adjacent carbon atoms.C For (CH,),CHCH,(NF,)CH,NF,, AHr" (calc) = -227.6 f 15.5 kJ mol-l (gauche correction omitted) and AHr" (measured) = -207.4 kJ mol-l.lla 9 Conclusion The difluoroamino-radical is probably one of the easiest nitrogen-containing radicals to study. Through the gas-phase kinetic study with organic compounds, much knowledge has been acquired about not only the reactions of this radical but, because of its acting as a convenient radical trap, also those of organic radical decompositions. Furthermore, the study of the difluoroamino-radical- organic radicals cross-combination reactions has led to a better overall under- standing of the nature of such cross-combination reactions. W. D.Good, D. R. Douslin, and J. P. McCullough, J. Phys. Chem., 1963,67, 1312. 39