Canadian Journa of Chemistry Published by THENATIONAL RESEARCH COUNCIL OF CANADA - -- FEBRUARY 1, 1970 Can. J. Chem. Downloaded from www.nrcresearchpress.com by 3.236.55.199 on 06/11/20 For personal use only. VOLUME 48 NUMBER 3 Heats of formation of some alkylthio radicals D. H. FINE'AND J. B. WESTMORE Chemistry Department, University of Manitoba, Winnipeg, Manitoba Received August 29, 1969 From a comparison of the dissociation energies of carbon-oxygen and carbon-sulfur bonds in alcohols, thioalcohols, ethers, and thioethers, a self-consistent set of values for the heats of formation of the methylthio-, ethylthio-, n-butylthio-, iso-butylthio-, sec-butylthio-, t-butylthio-, phenylthio-, and benzylthio-radicals is derived. Canadian Journal of Chemistry, 48, 395 (1970) Introduction An understanding of sulfur chemistry would be enhanced by the availability of reliable thermochemical data. In this paper we derive values for the heats of formation of the alkylthio radicals by comparing the few reliable (1) carbon-sulfur bond dissociation energies with the carbon-oxygen values obtained from the relatively well-studied alkoxy radicals (2, 3). The principal path towards the heats of formation of free radicals is by the union of thermodynamic and kinetic data. This method combines heats of formation determined calorimetrically with activation energies, derived from kinetic measurements, which are identified with dissociation energies of particular bonds. For many ethers and thioethers, represented by RXR', the dissociation energy of the R-XR' bond is too large to be determined directly by the kinetic method. It may nevertheless be evaluated indirectly from the relationship provided that the heats of formation of the radicals R - and .XR' and of the parent molecule RXR' are known. Heats of combustion of comlPresent address: Chemical Engineering Department, Massachusetts Institute of Technology, Cambridge, Mass. 02139. pounds containing carbon, hydrogen, oxygen, or sulfur can be measured to high precision by combustion calorimetry. Evaluation of the heat of formation of the gaseous compounds also requires knowledge of the heat of sublimation or vaporization of RXR' at 25 "C. Source of Data Available experimental values for the standard heats of formation of the alkyl-, aryl-, oxy-, and thio-radicals (3, 4) are listed in Table 1. The heats of formation of the parent oxygenand sulfur-compounds in the gas phase at 25 "C are given in Table 2. Experimental values are available for only some of the ethers and thioethers, and it is therefore necessary to make use of estimated values. Pilcher, Pell, and Coleman (17) calculated the heats of formation of the ethers by using the same Allen-type (25) equation that Skinner (26) had derived for the alcohols. The correlation was not good for iso-propyl-t-butyl ether (error 2.5 kcal/mole) and for di-t-butyl ether (error 10.6 kcal mole). We ascribe this discrepancy to steric interference when alkyl groups are attached to both carbon atoms of the C-0-C grouping, i.e. the hydrogen atom of an a methyl group interferes sterically with the hydrogen atom of the opposite a' carbon atom. The interference is small and we take account of it C C I by adding 0.91 kcal/mole for every C-C-0-C I 396 CANADIAN JOURNAL OF CHEMISTRY. TABLE 1 Standard heats of formation of alkyl-, aryl-, oxy-, and thio-radicals at 25 "C Can. J. Chem. Downloaded from www.nrcresearchpress.com by 3.236.55.199 on 06/11/20 For personal use only. Radical Heat of formation (kcal/mole) Reference VOL. 48, 1970 difficult to prepare and Smutny and Bondi (18) have shown that the strain due to overcrowding is about 7.6 kcal/mole, which agrees well with the 7.0 kcal/mole deviation between experimental and calculated values which we obtain here (see Table 2). For the thioethers, McCullough and Good's (23) scheme gives an average deviation between the experimental and calculated values of + 0.28 kcal/mole, which is again less than the average experimental uncertainty interval of 10.35 kcal/mole. Since the carbon-sulfur bond is longer than the carbon-oxygen bond (1.82 A compared with 1.42 A), steric interference is negligible in the thioethers and no correction is applied. Paucity of experimental data makes it difficult to extend the correlation to aryl ethers or aryl thioethers. Discussion The ROR' and RSR' bond dissociation energies, calculated from the data in Tables 1 and 2 are listed in Table 3. Comparisons (1) of bond dissociation energies have been made between corresponding oxygen- and sulfur-compounds and lead to heats of formation for H S - , C,H,S., CH,Se, and C,H,S. radicals believed to be interaction. The equations (17, 23) we use are reliable to _+ 3 kcal/mole. From Table 3 certain trends are immediately obvious. The alkoxy-alkyl bond dissociation energies are almost independent of the nature of the alkoxy or alkyl groups: the average dissociation energies for the methoxy-, ethoxy-, npropoxy-, iso-propoxy-, n-butoxy-, sec-butoxy-, where a is the number of C atoms, b the number iso-butoxy-, and t-butoxy-alkyl bonds are 80, of C-C-C interactions, c and c' the number of 82, 78, 81, 80, 81, 80, and 81 kcal/mole, respecC-C-0 and C-C-S interactions, respectively, tively. Only for compounds containing the n-propoxy group are the bond dissociation c energies seen to be out of line. The experimental I d the number of C-C-C interactions, e and e' values, based on pyrolyses of n-propyl nitrite (2), nitrate (2), and peroxide (28) are - 15.1, - 12.0, e c and - 10.7 kcal/mole, respectively. Kerr (4) 1 I the number of 6-C-0 and C-C-S inter- decided that, in keeping with the heat of formaactions, respectively, and f the number of tion of the methoxy- and ethoxy-radicals, the value based on the peroxide data was likely to be C c the more reliable. We support this argument as I I C-C-0-C interactions. The ethers, apart there is no reason why D(n-C,H,O-R) should from di-t-butyl ether, give an average deviation be 2 kcal/mole less than that for other alkoxybetween the experimental and calculated values alkyl bonds. of _f 0.21 kcal/mole, which is considerably less Apart from the methylthio- and ethylthiothan the average experimental uncertainty inter- containing alkylthio ethers, neither the heat of val of i 0 . 3 5 kcal/mole. Di-t-butyl ether is formation of the alkylthio radical nor the alkyl- 397 FINE AND WESTMORE: AHP VALUES OF SOME ALKYLTHIO RADICALS TABLE 2 Heats of formation (kcal/mole) of the parent oxygen- and sulfur-compounds, in the gas phase at 25 "C X = sulfur X = oxygen -AH: Obsd. - ca1cd.t Reference -AHfO Obsd. - calcd.? Reference Can. J. Chem. Downloaded from www.nrcresearchpress.com by 3.236.55.199 on 06/11/20 For personal use only. Compound *Calculated values. ?Difference between observed and calculated values of A H P . thio-alkyl bond dissociation energies are known. However, the data for the methylthio- and ethylthio-ethers closely parallel the data for the ethers. Thus, whereas the dissociation energy of the alkoxy-alkyl bond is 80 $ 2 kcal/mole, that + of the alkylthio-alkyl bond is 11 3 kcal/mole lower, at least for the methylthio- and ethylthioethers. As we may reasonably expect the relationship between the dissociation energies of the oxygen- and sulfur-compounds to persist for Can. J. Chem. Downloaded from www.nrcresearchpress.com by 3.236.55.199 on 06/11/20 For personal use only. a w oubrbbb$b$bQ&hufi c a m c c p c z r g r g r grp CANADIAN JOURNAL OF CHEMISTRY. VOL. 48, 1970 1 $1 399 FINE AND WESTMORE: AHro VALUES O F SOME ALKYLTHIO RADICALS Can. J. Chem. Downloaded from www.nrcresearchpress.com by 3.236.55.199 on 06/11/20 For personal use only. other alkyl ethers and alkylthioethers we are able to derive "best fit" values for the bond dissociation energies D(RIS-R) (see the values in parentheses in Table 3). From these "best fit" bond dissociation energies we derive the heats of formation of the remaining alkylthio radicals (see Table 4). TABLE 4 Standard heats of for~nationof the alkylthio radicals derived in this work Heat of formation (kcal/mole) Radical CH3S. C2H5S. n-C3H7S. iso-C,H,S. n-C4H9S. sec-C4H9S. iso-C4HyS. t-C4HyS. CsH5CH2S. CsH5S. This work* Literature? 2913 25i3 18k3 17i3 13i 3 12+3 11i3 9i3 55+5 50i3 30 (1, 11, 12) 25 (1,13) 20 (13) 18 (13) 15 (13) formation of the phenylthio radical. The weakest bond was broken, since D(CH3-SC6H,) = 60 < D(CH,S-C6H5) = 87 kcal/mole For methyl benzyl sulfide, the principal pyrolysis products, methanethiol and bibenzyl, are formed at the rame rate. The decomposition is most readily explained by the reactions C6H5CH2SCH3+ C6H5CHZ. C6H5CH3(carrier gas) + .SCH, + CsH5CHZ. + .SCH3 + H$CH3 2C6H5CHz. + CzH5CHzCHZC6H5 The sulfur atom remains attached to the methyl group to give the methylthio radical, because - 11 (13) - 50 (4) *Derived in this work from "best fit" bond dissociation energies. tReference numbers in parentheses. A knowledge of the relevant bond dissociation energies in the asymmetrical ethers and thioethers enables one to predict which bond is the weakest and hence where the molecule will split on pyrolysis. For example, the pyrolysis of methyl phenyl ether (29) in quartz tubes at 500-600 "C yields phenol as the major product. The decomposition is readily explained if the first step is This is to be expected, as the methyl-oxygen bond is 34 kcal/mole weaker than the phenyloxygen bond, viz. D(CH,-0C6H5) = 67 and D(CH,O-C6H ,) = 101 kcal/mole. The thermal decompositions of methyl phenyl sulfide (30) and methyl benzyl sulfide (1 I) have been investigated in a toluene flow system at 470-700 "C. For methyl phenyl sulfide the main products are methane, benzenethiol, and bibenzyl (from the toluene). Back and Sehon (30) ascribed the primary mode of reaction to the methyl-sulfur bond rupture The activation energy of 60 kcal/mole which they measured was ascribed to this process and they were hence able to derive a value for the heat of In methanethiol and ethanethiol the carbonsulfur bond is weaker than the sulfur-hydrogen bond D(CH,-SH) = 73 kcal/mole D(CH,S-H) = 87 kcal/mole D(C2H,-SH) = 70 kcal/mole D(C,H,S-H) = 88 kcal/mole and Sehon and Darwent (31) have shown that the first step in the pyrolysis of methanethiol is For ethanethiol the decomposition is more complex, being predominantly a molecular rearrangement at low temperature and a free-radical process at higher temperatures On pyrolysis, benzenethiol (31) and 2-methyl-2propanethiol(32) also split predominantly at the carbon-sulfur bond. The results listed in Table 4 depend upon a comparison of bond dissociation energies in many analogous pairs of compounds. Since a significant body of independently obtained data is involved, this would suggest that most values are within the error limits given of f3 kcal/mole. The reliability Can. J. Chem. Downloaded from www.nrcresearchpress.com by 3.236.55.199 on 06/11/20 For personal use only. 400 CANADIAN JOURNAL OF CHEMISTRY. VOL. 48, 1970 7. C. LEGGETT and J. C. J. THYNNE.Trans. Faraday of the data depend upon the reliability of the gasSOC.63, 2504 (1967). phase heats of formation of (i) the parent com8. S. W. BENSON. J. Amer. Chem. Soc. 87. 972 11965). 9. PETERGRAY. Personal communication. (1968). ' pounds, (ii) the alkyl- and alkoxy-radicals, and 10. B. A. THRUSH. Progr. React. Kinet. 3, 64 (1965). (iii) the mercapto-, methylthio-, and ethylthio- 11. E. H. BRAYE, A. H. SEHON,and B. DE B. DARWENT. radicals. Although the experimental heat of J. Amer. Chem. Soc. 77, 5282 (1955). and F. P. LOSSING. J. Amer. Chem. formation data for the parent compounds have 12. T. F. PALMER SOC.84, 4661 (1962). been supplemented by calculated values, we have 13. H. MACKLE.Tetrahedron, 19, 1159 (1963). shown the uncertainties to be less than k0.5 14. J. H. S. GREEN. Ouart. Rev. 15. 125 (1961). kcal/mole. The uncertainties in the heats of 15. H. A. GUNDRY,A. J. HEAD,and G. B'. LEWIS. Trans. Faraday Soc. 58, 1309 (1962). formation of the alkyl and alkoxy radicals are 16. D. R. STULL. Ind. Eng. Chem. 39, 517 (1947). generally about 1 to 2 kcal/mole. The largest 17. G. PILCHER,A. S. PELL,and D. J. COLEMAN.Trans. Soc. 60, 499 (1964). uncertainties ($.3 kcal/mole) are associated with 18. Faraday E. J. SMUTNY and A. BOKDI. 3. Phys. Chem. 65, the mercapto-, methylthio-, and ethylthio-radi546 (1961). cals (I). The heats of formation of the alkylthio 19. G. PILCHER,H. A. SKINNER,A. S. PELL,and A. E. Trans. Faraday Soc. 59, 316 (1963). radicals presented here are self-consistent. If 20. POPE. M. COLOMINA, A. S. PELL,H. A. SKINNER, and D. J. subsequent data suggest a revision of the values COLEMAN.Trans. Faraday Soc. 61, 2641 (1965). D. C. GINNINGS, P. E. MCCOSKEY, for the mercapto-, methylthio-, or ethylthio- 21. G. T. FURUKAWA, and R. A. NELSON. J. Res. Natl. Bur. Stand. 46, radicals, it should only be necessary to adjust the 195 (1951). and P. A. G. O'HARE. Tetrahedron, 19, values for the other sulfur-containing radicals by 22. H. MACKLE 961 (1963). -, a similar amount. 23. J-%. MCCULLOUGH and W. D. GOOD. J. Phys. Chem. 65, 1430 (1961). 24. M. J. S. DEWARand H. N. SCHMEISING.TetraWe thank the National Research Council of Canada hedron, 5, 166 (1959); 11, 96 (1960). and the Faculty of Graduate Studies, University of 25. T. L. ALLEN. J. Chem. Phvs. 31. 1039 (1959). Manitoba for financial assistance. 26. H. A. SKINNER.J. Chem. SOC. 4396 (19621, ' 27. S. W. BENSON.J. Chem. Educ. 42, 502 (1965). 28. E. J. HARRIS. Proc. Roy. Soc. London, Ser. A, 173, 1 . D. H. FIXEand J. B. WESTMORE.Chem. Commun. 126 (1939). 273 (1969). 29. Yu. K. SHAPOSHNIKOV and L. V. KOSYUKOVA. 2. P. GRAYand A. WILLIAMS. Chem. Rev. 59, 239 Khim. Pererab. Drev. Ref. Inform. No. 3, 6-9 (1965). (1959). Chem. Abstr. 66, 37557a. 3. P. GRAY,R. SHAW,and J. C. J. THYNNE. Progr. 30. M. H. BACKand A. H. SEHON. Can. J. Chem. 38, React. Kinet. 4, 63 (1967). 1076 11960). 4. J. A. KERR. Chem. Rev. 66.465 (1966). 31. A. H: SEHON and B. DE B. DARWENT.J. Amer. 5. A. S. RODGERS. D . M. GOLDEN. and s.'w. BENSON. Chem. Soc. 76, 4806 (1956). J. ~ m & .hem: Soc. 89, 4578 (i967). 32. C. J. THOMPSON, R. A. MEYER,and J. S. BELL. 6. Natl. Bur. Stand. Circ. 500. Washington, D.C. (1952). J. Amer. Chem. Soc. 74, 32 \ - ~