7564 appears quite apt in terms of electron movements, but in this case a slightly smaller A factor would seem more
I1
I
1
1.5
appropriate. If the decomposition is akin to those of the alkyl halideslg reaction would involve polarization of the molecule in the manner of 11. Kinetic investigations with other acyl halides are required to obtain information on the effects of substitutions at significant molecular sites.
1.6
10p.
+,
Figure 2. Arrhenius plot for temperatures 3.54-426"; without additives; 0, with cyclohexene; 0,with isobutylene; A, in packed vessel.
(9) A. Maccoll, Aduan. Phys. Org. Chem., 3, 91 (1965).
+ *
+
Kinetic Study of the Reaction CHI, HI CHJ, I,. A Summary of Tliermochemical Properties of Halomethanes and Halomethyl FLadicals'" Shozo Furuyama,lb David M. Golden, and Sidney W. Benson Contribution from the Department of Thermochemistry and Chemical Kinetics, Stanford Research Institute, Menlo Park, California 94025. Received M a y 2, 1969
+
+
Abstract: The rate of the reaction CHI3 H I CH& I2 has been followed spectrophotometrically from 158.1 to 206.1'. The rate constant for the reaction I CHI3Ffc CHI2 I2fits the equation,log kl (M-l sec-l) = (11.75 f 0.17) - (9.63 f 0.36)/8. This value, combined with the assumption that & = 0 f 1 kcal/mole, leads to AH12' 98 (CHI2.) = 79.9 f 2.3, DHoZss(CHI2-I) = 45.7 f 1.2, and DH02ss(CH12-H) = 102.7 f 2.5 kcal/mole, respectively. The heats of formation of (unmixed) halomethanes and halomethyl radicals are summarized and discussed in terms of Bernstein's interaction scheme and Pauling's bond-energy equations.
In
+
+
previous papers, we have reported on studies of the Experimental Section equilibrium 2CH31 CHI CH212,2a the kinetic!; kIaterials. Mallinckrodt reagent grade, resublimed, iodine was 12,and the iodine-catalyzed of CH212 HI Ffc CHJ used. Matheson anhydrous hydrogen iodide was used after purification by distillation under vacuum. Eastman iodoform was CH212,2b in which the heats of process 2CHJ Ffc CHI purified by recrystallization in methanol and resublimation under formation, AHfo(CH212)and AHfo(CH21),and bond vacuum. The purity of CHI3 was determined to be greater than dissociation energies, DHo(CH21-I)and DHo(CH21-H), 99.5 by nmr measurements. were determined. Apparatus. The slightly modified Cary 15 spectrophotometer, The kinetics and thermochemistry of the reaction adapted for use with a quartz reaction vessel situated in an AI CH31 HI $ CHI I2have already been r e p ~ r t e d . ~ ) ~block oven, has been described in detail previou~ly.~ Procedure. The mechanism of the reaction R I + HI $ RH In this paper, the kinetics of CHI3 HI F?: CH212 IZ has been well explained by the scheme proposed by Benson I2 is discussed. This terminates our studies of iodoand O"ea15 methanes (c14 is not adaptable to study with thest: Iz I_21 KI,(equilibrium) methods). A summary of the thermochemical properties of halomethanes and halomethyl radicals is pre1 sented, as well. RI+IJ-R. + I ~ (132)
+
+
+
+ +
+
+
+
+ +
2
(1) (a) This work was supported in part by Grant AP0053-04, Public Health Service, Division of Air Pollution; (b) Postdoctoral Research Associate. (2) (a) S. Furuyama, D. M. Golden, and S. W. Benson, J . Phys. Chem., 72, 4713 (1968); (b) Intern. J. Chem. Kinetics, 1, 283 (1969). (3) M . C. Flowers and S. W. Benson,.J. Chem. Phys., 38, 882 (1963). (4) D. M. Golden, R. Walsh, and S. W. Benson, J . Am. Chem. SOC., 87, 4053 (1965).
3
R. +HI,RH+I
(3,4)
4
where R is an alkyl or aryl radical. The rate of deiodination of R I (5) S. W. Benson and E. O'Neal. J. Chem. Phys., 34,514 (1961).
Journal of the American Chemical Society / 91:27 / December 31, I969
7565 (i2l%/kapp1 0
1
1
1.6
0
1
3
2
/
/
3
2
/
4
5
200
500
400
300 TIME
+
-
with HI is expressed as
Equation 5 is transformed into a pseudo-first-order kinetic equation in a case where the pressures of HI and IZare so high in comparison with that of RI that they can be regarded as almost constant during the reaction.
(E)
+
+
Here is [(I& (IP),YZ, (HI) is [(HI)o (HI)m]/2, and (130, etc., are the initial, or final, pressures of 12 and HI. (RI)can be equated with ( O D - CYRH(RI)O - CYI~(IZ) - aadHI))/(aRr- RH) which, to a good approximation, equals (OD - O D J / ( ~ R IOLRE), where OD and OD, are optical densities a t time t and at equilibrium, respectively (RI = CHII, RH = CHA, here). The a's are the extinction coefficients at fixed wavelengths of the components indicated by the suffixes. By substituting this into eq 6 , the usual first-order integrated equation is obtained.
(7) kappwas obtained by following OD - OD, a t 350 ml.6 k 4 k 3 and &IKI~'/Z _were_then determined from the intercept and the slope of a plot of (Iz)/(HI) against (g)'/2/kaPpby the equation
Cu. 0.2-0.4 Torr of iodoform' (OD0 OD, was 4 . 2 - 0 . 4 OD unit) was deiodinated to methylene iodide by excess hydrogen iodide (-4-120 Torr) in the presence of excess iodide (-6-20 Torr) at 158-206". It seems possible that the consecutive reaction, CHZTZ4- HI CHaI 12,might take place simultaneously during this experiment.
+
(6) aiZand LYHIare negligibly small at 350 mp (ref 2). (7) This is the highest vapor pressure of iodoform at -100-200".
7
8
10
9
l
l
9
10
- torr% .seconds
However, kl (for CHzIP I -+ C H J . $. 12) has been determined as 1011.%-16.11'%2bwhich is about 240 times smaller than kl (for Iz). This was confirmed by the fact that CHIP I -.+ CHIz. increase of the pressure of iodine after the reaction was always comparable with the initial pressure of CHIa, and OD, - ODI,, - ODHI, was also the same as the expected O D C H ~ I ~ .
+
+
Results k,, was obtained from a plot of log (OD - OD,) us. time. Typical plots are shown in Figure 1. (L)/ (HI)was then plotted against (~)'/'/kapp.at 159 186, and 205.5" in Figure 2. Good straight lines were obtained at each temperature where k3/kz and klKIz'IZ were determined directly from their intercepts and slopes. Data are summarized in Table I. All values of k3/kz fall in the range -0.10-0.12. Log (k3/kZ) is (-0.9 =t0.9) - (0.2 f 1.9)/0, which agrees well with the relation log ( k 3 / k 2 )= (-0.5 f 1) - (1 f l)/e (0 = 2.303RT kcal/mole) established for the analogous reactions in other system^.^^^-^^' k3/k2= 0.10 at 158" coincides very closely with the e3timated value from log (k3/k2) = -0.5 - (1/0). Thi!; is probably the most reliable value because of the suitable reaction velocity ,for measurements (see Figure l), and the excellent straight line of (Iz)/(HI) us. (h)'/'/kapp(see Figure 2) at this temperature. k3/kZ at 1.86-206' is corrected to obey log (ke/kz) = -0.5 - ( l / @ and substituted into eq 6 or 8, together with the known K I ~ ' l/ 1~to obtain kl. Log kl,which is plotted againcst 1/Tin Figure 3, is
logkl(M-' sec-*) = (11.75
f
0.17) - (9.63
f
0.36)p
where the errors are standard deviations. kl, which is directly obtained from the plots in Figure 2, is given by log kl(M-l sec-l)
-
6
/
+
600
- seconds
Figure 1. Plot of log (OD OD,) us. time for CHIo HI Ft CH212 HI. VaIues (in Torr) for three temperatures are given for (CHI&, (HI)o, and (I&, respectively, as follows: (205") 0.30, 4.42,6.30; (186")0.28,4.04,11.83; (158")0.37,116.2,9.02.
+
8
7 /
us. (&'/Z/karlp.
Figure 2. Plot of 100
seconds
I;
l
(i2)x/kappl
0
5
4
=
(12.07
f 0.18)
- (10.28
f 0.37)p
The differences in the Arrheniu:; parameters of both kl's are within the 95 confidence limits. The bond dissociation en xgy, DHo(CH12-I), is (8) D. B. Hartley and S. W. Benson, J . Chem. Phys., 39, 132 (1963). (9) A. S. Rodgers, D. M. Golden, ard S. W. Benson, J. A m , Chem. SOC.,89, 4578 (1967). (10) H. E. O'NeaI and S. W. Benson, J . Chem. Phys., 37, 540 (1962). (11) "JANAF Interim Thermochemical Tables," D. R. Stull, Ed., Dow Chemical Co., Midland, Mich., 1963.
Furuyama, Golden, Bemon 1 Halomethanes and Halomethyl Radicals
7566 Table I. Kinetic Data for CHI3
+ HI + CHaIz+
I2 in
the Presence of Excess HI and Iz
Temp,
(HVo
(cHI3)o
"C
102kapp,a sec-'
(1210
lO-lk1, M-l sec-l
k1K12'/2,
k3/k2
M-'/lsec-l
(k3/kz)estc
10-7(kl)corr,d M-l sec-l
206.1 205.8 205.7 205.2 201.4Q 186.0 186.0 186.0 158.3 158.1 158.1
0.33 27.55 14.32 3.66 2.43 0.20 7.77 11.63 1.12 2.10 0.22 36.55 5.04 5.30 0.10 9.70 2.46 0.111 2.35 0.30 4.42 6.30 0.877 2.16 0.35 30.30 7.54 3.42 ... ... *.. 0.110 2.00 0.30 19.90 11.41 0.902 1.61 0.28 9.18 11.69 0.419 0.12 2.51 1.38 0.106 1.52 0.30 49.2 7.48 1.77 1.47 0.27 44.2 12.46 0.233 0.772 0.10 0.398 0.743 0.10 0.695 0.37 118.2 9.02 0.387 0.39 19.92 19.83 0.0970 0.745 k,,, = klZG2'/a(12)'/2/(l kZ(I>/ka(E)). Q The consecutive reaction CHZL HI $ CH3I 12 was followed after the reaction with addiCorrected value, using (k3/kZ)eat. tion of more HI. See ref 2b, Table 111. c Estimated values from the equation log (k3/kZ) = -0.5 - ( l / O
+
+
Table 11. Gas-Phase Thermochemical Data
related to heats of formation by
+
AHi ', So, kcal/mole gibbs/mole -Cp 298 O 298 O 298 O
DHTO(CHI2-I) E AHf"T(CHI2.) AHfOT(1) AHfOT(CHJ.3) AH1.zo(T) DfI~"(1-1) (9)
+
The usual a ~ s u m p t i o nthat ~ ~ Ez ~ ~ = 0 i 1 kcal/mole leads to AH1.2(4550K)= El- Ez
-
+
1 kcal/mole Using values of the appropriate thermochemical data listed in Table I1
El = 9.6
f
CHI2
79.8. 14.924 25.537 59.8a
12c I C
CHIs
-
20
I
I
I
2.1
2.3
2.2
!eT
24
+
Figure 3. Arrhenius plot for the reaction CHI3 f 1. CHI, Iz: 0, ----,kl with experimental k3/kz; 0 , -, kl with estimated k3/k2. -+
+
~
+
where ACpo = (ACpo2g~ ACpo455)/2 = -0.2 i= 2 cal/(mole deg). We cannot determine the precise value of AHfo(CH12.); however, it is possible to estimate it using the extrapolated valuezaof AHf0(CHI3) = 59.8 f 2 kcal/mole. AHf02g8(CHIz*)= DHOzas(CHIz-I) f AHf029a(CH13) AHf0zs8(I) = 79.9 2.3 kcal/mole
-
+
Thus, DHoz9s(CH12-H) = AHf'zgs(CH1z) AHf0z9s(H) - AHf029a(CHz12) = 102.7 f 2.5 kcal/mole.
Discussion DH0298(CHIrI) = 45.7 f 1.2 kcal/mole obtained here agrees within the error with a value of 50 f 4
14.90* 8,901 4.968 19.44d
16.31* 8.948 4.968 21.73d
-
-
+-
kcal/mole obtained from a study of radical reactions in diffusion flames of alkali metals with organic halides. l 2 Log A1 of the step RI 1 3 R . 4- IZ for CHI3 obtained here is 11.75, which is of comparable size to that for CHJ (11.40)3 and for CH21z(11.45).2b The A factor may be expressed in terms of collision theory as
+
+
DH0zg8(CHIz-I) = Aff1.2(455 OK) DHozss(I-I) 0.16ACp0 = 45.7 i 1.2 kcal/mole
13.38* 8.814 4.968 17.11d
gibbs/mole---400" 500"
S o and Cpoof CHIz are determined a Estimated in this work. as follows. When CHL is produced from CH212by abstraction of one hydrogen atom, three modes of vibration, that is, C-H stretch0 H-C-H bending ( V -1350 cm-l), and H-C-I ing (v ~ 3 0 0 cm-I), 0 are lost: E. K. Plyler and W. S. Benedict, bending (v ~ 1 1 0 cm-I), J. Res. Nut. Bur. Srd., 47, 202 (1951). S'(CHI2) and C," are then given by S0(CH212) ZS"(vib) - 4.58(Z log 0 - log 2) and Cqo(CH212) 2Cpo(vib), respectively. u = symmetry. Log 2 arises from the degeneracy of the electronic state. S"(vib) and C, "(vib) are obtained from S. W. Benson, "Thermochemical Kinetics,'' John Wiley and Sons,Inc., New York, N. Y .,1968, p 208. c Reference 11. d SO(CHI3) is approximated by 2S0(CHzI2) So(CHJ) 4.582 log (symmetry change). C,"(CHL) 2Cp0CPoand Soof CHZLand CH3I are obtained (CH21z) C,"(CHJ). from E, Gelles and K. S. Pitzer, J. Am. Chem. SOC.,75, 5259 (1953).
-
6.7
75.2gb 62.281 43.184 85.37d
O,
A
=
pZe'l2
wherep is a steric factor, and 2 is the collision frequency. (The ell2 arises as a result of the Til2dependeye in Z.) By taking the mean collision diameter, u,5.0 A13*14 for a CHI3-I pair, log Ze'/' = 11.42 at 500"K, so that p Z 1. For the case of CH31, log Ze'/' = 10.96, when CT is taken as 2.7 k. This is slightly smaller than the experimental value of Flowers and Benson ; however, p 1 within experimental error. For the case 9f CH212, log Ze'I2is 11.33, when CT is taken as 4.4 A. This agrees well with the experimental value, log kl = 11.45, which means p Gi 1. In the cases of RI = CzH61: CsH61,$and CH3COI,1athe steric factors are also almost unity, A value of p near unity implies little need for orientation in the collision pair and corre(12) W. J. Miller and H.B. Palmer, J . Chem. Phys., 40, 701 (1964). (13) Covalent radii of C and I = 0.77 and 1.33 A.14 (14) C. A. Coulson, "Valence," Clarendon Press, Oxford, 1953, p 180.
Journal of the American Chemical Society J 91:27 / December 31, 1969
7567
sponds to a loose transition state in the language of transition-state theory.15 Reactions of R I I a R. I2 probably pass through the same type of transition state for all R's which are not resonance stabilized. I e C13 This might suggest that the reaction CII I2 would proceed through a loose transition state, too. Therefore, log A l for the reaction would probably be 11.8 bFcause log Ze'/zcalculated at 500°K by taking u = 5.0Ais 11.40. Considering that the activation energies, El's, for CH31,3CH212,2b and CHIa are 20.5, 15.5, and 9.6 kcal/ mole, respectively, it may be plausible to predict by a simple extrapolation that El for C14 I CIS I2 would be 4 f 1 kcal/mole. Combining this with A1 estimated above, log kl for CIe would be (1 1.8 f 0.2) - (4 lye. An activation energy of 4 kcal/mole implies DHTo(C13-I) DH02ss(C13-I) = 40 kcal/mole. This, together with an extrapolation of the double difference scheme applied in ref 2a, to yield AH02ss(C14,,) 95 kcal/mole, S 110 kcal/mole. The yields a value of AHf0298(C13,g) latter value would mean that DH0298(C13-H) i Z 102 kcal/mole.
+
+
+
+
+
+
-
Summary of Thermochemical Data for Halomethanes and Halomethyl Radicals Table I11 summarizes the known thermochemical properties of the (unmixed) halomethanes and halomethyl radicals. Bernstein l6 has developed an interaction scheme for predicting the properties of fluoro-, chloro-, and bromoethanes. This scheme may be expanded to include iodomethanes and the various halomethyl radicals. According to Bernstein's scheme, AHf"(CH,-,X,) may be expressed as
+
AHfo(CH4-nX,) = nCX (4 - n)CH
+ 4 4 - n)AXH
(10) and CH are the effective bond contributions where of C-X and C-H and correspond to AHf"(CXe)/4 and AHro(CH4)/4,respectively. AXH is the effective interaction between X and -H, and corresponds to -A[A(AHto)]/2.l7 CH, CX, and AXH may be defined in terms of 3 AHfo(CXe)/4 = CX - CX - XX (11) 2
+
AHfo(CH4)/4 =
CH
= CH
+ -23 H H
-2A[A(AHio)]/2 = 2AXH = 2XH - XX
(12)
- HH
(13) where CX and CH are implicit bond contributions, and XX, HH, and XH are implicit interactions between X and X, and so on. Implicit bond contributions and interactions are obtained as follows. CA (A = X or H) is defined in terms of CA = (1/4)AHro(CA4*) = (1/4)AH1"[C(g)l
+
AHf"[A(g)]
- (1/4)AHao(CA4*)
(14)
( 1 5 ) S. W. Benson, "Foundations of Chemical Kinetics," McGrawHill Book Co., Inc., New York, N. Y., 1960, pp 271-281. (16) H. J. Bernstein. J . Phvs. Chem.. 69. 1550 (19651. (17j For example, A[A(Akf ")] = [AHf"(CH?) -'AHi'(CHaX)I -
[AHt"(CHaX)
- AHro(CHzXz)l.
Furuyama, Golden, Benson 1 Halomethane;;and Halomethyl Radicals
7568 Table IV. Bond or Interaction Contributions t o AHf“(at 25“ in Gas)= Atom AHtaA (A)b
H F C1 Br I C
52.1 18.9 28.9 26.7 25.5 170.9
D(AA). 104.2 37.8 57.8 46.0 36.1 83.1
XAC
2.1 4.0 3.0 2.8 2.5 2.5
D(C-A)c
CA
98.8 117.2d 78.5 65.9 57.4 83.1
-4.5 -55.7 -5.7 9.5 23.8
All in kcal/mole, except for XA in column 4.
b
ex’
AXH
AX”
11.4 1 . l C -37.5 -0.8 6.2 -1.6 24.0 -2.4 36.4
Reference 11.
c
1.9’ 0.4 -2.6 -2.0
{see text).
where CA4*is a hypothetical methane (or halomethane) gas which has no nonbonded interactions and AH,’ i!r the heat of atomization. AH,’(CA4*)/4 is identical with the “bond energy” of C-A, D(C-A). (14) i:; transformed into
+
CA = (l/4)AHfo[C(g) AHf”[A(g)] - D(C-A) (15) D(C-A) may be calculated from Pauling’s empirical equations l8* l9
+ 3 0 ( ~ ,- XA)’ (16)
or D(C-A)
=
1 $D(C-C)
+ D(A-A)] + 23(xC - xA)’ (17)
where D(A-A) is the bond dissociation energy of A.2 gas, and x is the electronegativity value of the elements indicated by the suffix. D(C-C) is the bond dissociation energy of a typical C-C bond between sp3 carbon atoms. Pauling uses 83.1 kcal/mole. 18119 Substituting D(C-A) obtained by Pauling18into (15), CA is calculated for A = H, F, C1, Br, and I. XX: and HH are, then, obtained by substituting CA’s into eq 11 and 12. Finally, XH is calculated from relation 13. The data are summarized in Table IV. The result that I1 = 8.7 and BrBr = 3.3 kcal/mole, while HH, FF, and ClCl 0 kcal/mole, does not seem unreasonable. The heat of formation of the halomethyl radical may be expressed in a similar way as (10)
-
AH,’(*CH,,X,)
+
nCX’ (3 - n)CH’
=
+ n(3 - n)AXH’
(18)
where the definitions of CX’, etc., are the same as those .)/3, for halomethanes and correspond to AHfo(CX3 AHfo(CH3.)/3, and -A[A(AHf “)]/2, respectively. CX’, etc., are then given by
CX’ AHfo(.CH3)/3 = CH’ AHfo(-CX3)/3=
AIA(AHfo)] = 2AXH’
=
+ XX’ = CH’ + HH’ = CX’
2XH‘
- XX’
AA -0.3 -0.1 0.3 3.3 8.7
AH
CA’
AA’
AH’
0.9 -0.8 -0.1 1.8
11.7 -39.9 8.8 20.2 26.5
-0.3 2.4 -2.6 3.8 9.9
2.9 -1.1 -0.9 2.8
References 18 and 19. d Calculated value in the present paper (see text). AHf’(CH3) is omitted to calculate AFH’ = -A[A(AHtD(R.))]/2
AHr”(CH4)isomitted to calculate AFH = -A[A(AHf“(RX))]/2 (see text).
D(C-A) = [D(C-C)D(A-A)]”z
CA -4.0 -55.6 -6.9 4.5 10.8
(19) (20)
’
where the definitions of CX’, etc., are the same as those for halomethanes. Similarly, CA’ can be expressed as CA‘ = (1/3)AHf0(.CA3*)= D(C-A)
this value also fits well the heats of formation of some fluorinated ethanes. The difference in the values given by (16) and (17) may reflect, in part, some of the difficulties of fitting fluorine-containing compounds into a simple scheme. (19) L. Pauling, “The Nature of the Chemical Bond,” 3rd ed, Cornel1 University Press, Ithaca, N. Y.,1960, pp 79-95.
(22)
23(xc. - xA)’ (23) where the definitions are analogous to those in (14)(16). D(C-C) is then the hypothetical bond dissociation energy of a bond between carbons in the case where a radical center would exist at each carbon, but the dissociation products would be neither stabilized nor destabilized. This is clearly equal to D(C-C) and the same reasoning requires D(C-A) = D(C-A). Here we may make use of the following two approximations: (a) HH’ = HH, (b) CH - CH’ = CX - CX’. (a) seems reasonable as there is cause to suspect that the small H-H interaction will be even smaller in the radical. (b) seems reasonable, based on the following considerations. From (17) and (23) CH
- CH’
= (1/2)[D(C-C)
- D(C-C)]
23[xC2 - 2xCxH -
cx - CX’
=
+
XCe2
+ 2xC.xH]
- D(C-C)] + 23[xc2 - 2xCxH - XC.’ + 2xc.xH]
(1/2)[D(C-C)
A = (CH - CH’) - (CX - CX’) = 46(xx
- xH)(xC - xc.)
= 0 kcal/mole
As xx # xH, (b) .requires XC. = xc (=2.5). This, in turn, leads to D(C-C) = 83.2 f 0.5 kcal/mole by (22) and (23) (where A = H, taking into account assumption a), which is the same as D(C-C) = 83.1 kcal/mole, which is expected. (a) and (b) lead to CH’
=
- HH
- 11.7 kcal/mole
(24)
and CX’ = CX - CH
+ CH’ = CX
- HH’ (21)
(18) We have used B(C-A) from Tables 3-4, p 85 of ref 19, except for D(C-F),which is listed in this table as 105.4 kcal/mole. Calculations using eq 16 and 17 yield values of 111.7 and 122.6 kcal/mole, re. spectively. We have used the average value of 117.2 kcal/mole, since
+ AHfo[A(g)l- W - A ) = [D(C-C) + D(A-A)]/2 +
(1/3)AHfo[c(g)l
+ 15.7 kcal/mole
(25)
From the known data, CX’, XX’, and XH’ are calculated by (24), (19), and (21). The results are summarized in Table IV. It is interesting to note that any XX’ or XH’ is quite close to the corresponding XX or XH (except for, perhaps, FF‘ and ClCl’). Comparing the structure of CH4-,Xn with that of .CH3-,X,, the distance between X and X (or X and H) in the radical is only a maximum
Journal of the American Chemical Society / 91:27 / Decembcr 31, 1969
7569 of 7 %longer than that in the molecule. It is reasonable to expect, therefore, that XX (or XH) is close to XX' (or XH'). The coincidence between XX and XX' (or XH and XH') obtained here seems to be quite satisfactory and shows that the approximations a and b are not unreasonable. It should be noted that the second differences in the fluorocarbon series (both molecule and radical) are not nearly as constant as the other halocarbons. Discussion of the interaction parameters for these compounds is therefore somewhat tenuous. In the previous papers2it has been observed that AH0(2CH31
CH4
+ CHzI2) =
4.7 kcal/mole
DHO(CH3-H)
- DH"(CHJ-H)
AHO(CH4
+ .CHJ
(26)
=
CH3I
+ .CHB) =
0.3 kcal/mole (27) and DHO(CH3-I)
- DH"(CH2I-I)
AHO(CH31
+ .CHJ
=
CH&
+ aCH3) = 5 kcal/mole (28)
Since (26) is derived from (27) and (28), a problem is why AHOM follows a simple bond additivity rule,
while AHo2,does not. Generally DH"( CH,-,I,-H)
- DHo(CH3-,p11,+ 1-H)
N
1 kcal/mole (29)
while DHo(CHcnIn-l-I) - DHo(CHkt-lIn-I) -N 6 kcal/mole (30) This can be explained as follow3. Both (29) and (30) are expressed in terms of CX, XX, etc., through the relations in (10x13) and (18)-(21). They are finally approximated by (29) = IH and (30) = I1 - IH, since I1 sz TI' and IH ss IH'. Both IH = 2, and I1 - I H = 7 kcal/mole agree with (29) == 1 and (30) = 6 kcal/ mole, respectively. The data for 11, II', IH, and IH' come from the same sources as eq 26-30, so this agreement is not surprising. However, formulation of (29) and (30), in terms of the interaction parameters of the Bernstein scheme, makes the differences in BDE's physically understandable. For difference in bond stren,gths, such as (30), the number of 1-1 interactions change so that the difference is far from zero. For differenccx like (29), where only the transfer of a hydrogen atom occurs, it is the number of I-H interactions (a quantity much smaller than 1-1) which changes and the difference is small.
The Thermal Decomposition of 6-Methyl-3,4-djhydro-2H-pyran' Charlotte S. Caton2 Contribution from the Department of Chemistry, University of Rochester, Rochester, New York 14627. Receiued October 7 , 1968 Abstract: The gas-phase thermal decomposition of 6-methyl-3,4-dihydro-2H-pyran (6-MDHP) into ethylene and methyl vinyl ketone has been investigated over the temperature range 330-370' and at initial pressures of 5-25 mm. The reaction is a first-order homogeneous process and is not affected by the addition of nitric oxide or propylene. An activation energy of 51.2 f 0.5 kcal/mole has been found and the first-order rate constant is k = 2.82 f 0.09 x 1014 exp(-51,200/RT) sec-I. 6-MDHP has been found by gas chromatography to be among the
products formed in the decomposition of methyl cyclobutyl ketone.
T
he thermal decomposition of several derivatives of cyclopropane and cyclobutane (i.e., those derivatives with 4-CH8 and 4-R
II
CHz
II
0
where R is H and CHI) have been found to yield corresponding five- and six-membered ring compounds. Isopropenylcyclobutane gave nearly equal amounts of ethylene-isoprene and 1-methylcyclohexene. Roquitte' (1) Th.is work was supported by a grant from the National Science Foundation. (2) Abstracted by C. S. Caton from her M.S. thesis written under the supervision of the late W. D. Walters, University of Rochester, 1967. Address correspondence to author at 255 Dolly Varden Blvd., #38, Scarborough 722, Ontario, Canada. (3) R. J. Ellis and H. M. Frey, Tram. Furuduy Soc., 59, 2076 (1963). (4) B. C. Roquitte and W. D. Walters, J. Am. Chem. SOC.,84,4049 (1962).
established that a small amount of 3,4-dihydro-2Hpyran was formed during the decomposition of cyclobutanecarboxaldehyde. In contrast, no ring-enlargement reaction was reported for the decomposition of methyl cyclobutyl ketone.6 Our present reinvestigation of the decomposition product; of this compound by means of gas chromatography indicates the presence of a very small amount of a ring-enlargement product, 6methyl-3,4-dihydro-2H-pyran(6-MDHP). Subsequently, the decomposition of 6-MDHP has been studied in some detail. Experimental Section Materials. Two methods were used to prepare 6-methyl-3,4dihydro-2H-pyran (6MDHP). Sample I was obtained by de(5) L. G. Daignault and
Caton
w.D. Waltm, ibfd., 80, 541 (1958).
Decomposition of 6-A!fefhyl-3,#-dihyclro-2Zi-pyran
