KINETICS AND MECHANISMS OF THE PYROLYSIS OF DIMETHYL ETHER

111. THE REACTION ACCELERATED BY HYDROGEN SULPHIDE

D. J. MCKENNEY AND K. J. LAIDLER

Department of Chemistry, University of Ottawa, Ottawa, Canada Received November 23. 1962 ABSTRACT The thermal decomposition of dimethyl ether in the presence of hydrogen sulphide was studied in the temperature range of 480 to 530' C, and over the pressure range of 5 to 500 mm of Hg. For a given pressure of dimethyl ether the rate of decomposition increased with the addition of increasing quantities of hydrogen sulphide, reaching a plateau after approximately 30% H S had been added. The relative rate of decomposition then remained constant until more than 50% HIS had been added. Further increase in the hydrogen sulphide concentration produced a linear increase in the relative rate. I n the region between 30 and 50Yo added HzS, the decomposition of ditnethyl ether was three-halves order with respect to ether pressure, and zero order with respect to hydrogen sulphide pressure. I n this region the three-halves-order cc1I2mole - ' I 2 sec-I. In the region rate constant can be expressed by k = 1.06X1014e-53,20Q'RT beyond 50% added HIS, the reaction was first order with respect to dimethyl ether, and onehalf order with respect to hydrogen sulphide. I n this region the three-halves-order rate constant cc1I2in01e-~1~ set-l. The experimental facts are shown is given by k = 4.98X 1014e-52,5001RT t o be consistent with a mechanism involving hydrosulphide radicals as the principal chain carriers. These H S radicals are produced mainly from the reaction of methyl radicals with hydrogen sulphide. The work leads to a value of 85.6 kcal for the dissociation energy of HIS into H SH.

Can. J. Chem. Downloaded from www.nrcresearchpress.com by 112.201.27.17 on 01/24/12 For personal use only.

INTRODUCTION

A number of workers (1-7) have suggested that thiyl radicals (RS) are more reactive than alkyl radicals. Thus thiols generally accelerate free-radical reactions involving saturated aldehydes; this has been shown to be the case in the liquid phase ( I ) , in photodecomposition reactions in the gas phase (2, 3) and in thermal decompositions in the gas phase (4-7). Birrell and co-workers (2) observed that various thiols (H2S, CH3SI-I, C2H6SH, etc.) catalyzed the photodecomposition of acetaldehyde, and in most cases the thiols increased the rate of the decomposition. The rates increased to a plateau with the addition of thiol and remained constant over a range of concentrations. When the thiol concentration was about three-quarters of the acetaldehyde concentration the relative rate (ratio of catalyzed to uncatalyzed rate) increased further. Hydrogen sulphide was found to be the most effective catalyst, and the catalytic effect decreased with increasing complexity of the thiol. The mechanism proposed for catalysis by methanethiol involved the chain transfer reaction
CH3

+ CH3SH

CH3S

+ CH, + CH3 + CO

followed by
CH3S

+ CH3CHO
2CH3S

CH3SH

and termination by
-+

CH3SSCHa.

Such a simple ~llechanism could not, however, explain the effects observed for other thiols. More recently Imai and Toyama have postulated mechanisms to explain the acceleration by hydrogen sulphide and methanethiol of the decomposition of acetaldehyde (4, 5, 7). The effect of hydrogen sulphide on the decomposition of dimethyl ether was also studied (6). These workers suggest that the catalytic effect of hydrogen sulphide is
Canadian Journal of Chemistry. Volume 41 (1963)

2009

2010

CANADIAN J O U R N I L O F

CHEMISTRY. V O L . 41,

1963

Can. J. Chem. Downloaded from www.nrcresearchpress.com by 112.201.27.17 on 01/24/12 For personal use only.

essentially due to reaction of methyl radicals with hydrogen sulphide, with the production of hydrosulphide (SH) radicals which in turn abstract hydrogen atoms from the substrate. These reactions are suggested to be faster than the reaction between methyl radicals and the substrate. For the thermal decomposition of dimethyl ether in the presence of hydrogen sulphide the rate increased .with increasing concentration of hydrogen sulphide and reached a limit where further addition of hydrogen sulphide had no effect. T h e maximum amount of hydrogen sulphide which was added was approximately 67%, and the reactions were carried out over the temperature range of 360 to 440' C. T h e mechanism suggested for the reaction in the presence of sufficient hydrogen sulphide to ensure "liiniting" acceleration was as follows:"

+ CH30 + H2S CHZOH + HS CH3 + H2S CH, + HS CH30CH3+ HS --+ H2S + CH3OCH2 CH30CH, + CH20 + CH3 HS + HS ,products.
CH30CH, -+ CH3
--t

Ill
[241 PI] 1201

CHZO

-t

[31
la31

The rate equation to which this mechanism leads is

and a three-halves-order dependence on ether concentration was obtained experimentally. set-l. ccli2 The results led to a rate constant of k = 1.82 X 1015e-512800iRT In the light of recent work (Part 11) in this laboratory on the action of inhibitors on the thermal decomposition of dimethyl ether, it was thought t h a t a mechanism similar t o inhibition mechanisms could be applied to the reaction accelerated by hj~drogensulphide. I t also seemed necessary t o extend the experimental investigation t o higher temperatures and higher pressures of H2S. The fact that there was an increase in relative rate a t very high pressures of H2S (>50% added HZ) indicated that the mechanism postulated by Imai and Toyama (6) is not sufficient to describe the reaction completely.
EXPERIMENTAL The apparatus used was essentially the same as described previously (8; see also Part I of this series). The dimethyl ether and hydrogen sulphide were 99.5% pure, and were used after three trap-to-trap distillations. T h e reactants were mixed thoroughly prior to a run, the procedure described earlier (Part 111 being employed. The reaction vessel was a cylindrical quartz tube of about 250-in1 volume. 4 Thermo Electronic controller, activated by means of a thermocouple in the copper furnace block, maintained the tenlperature constant to about A0.2" C. T h e temperature was measured by means of a thermocouple in the reaction vessel and a Thermo Electric Mini-Mite potentiometer. Before each run the reaction vessel was pumped down to less than 10 p. This small residual pressure had no detectable effect on the rates. A Beclcman photopen recorder in conjunction with a quartz spiral gauge automatically recorded the pressure change as a function of time. The details of this arrangement have been described by Laidler and Wojciechowslri (8). Rates were calculated from the slopes of the pressure-time curves a t the inflection point, and the results were found t o be fully reproducible over the entire range of pressure and temperature. The induction period for the reaction in the presence of H2S was generally quite short, so t h a t t h e inflection point occurred in all cases a t less than lOy0 decomposition. The rates can therefore be considered to be initial rates to a very close approximation.

*The numbering i s writfen so as to be consistent with the mechanism postulated below, as well as with the numbering in Parts I and 11 of this series.

McKENNEY AND LAIDLER: DIMETHYL ETHER

PYROLYSIS. 111

2011

RESULTS

Can. J. Chem. Downloaded from www.nrcresearchpress.com by 112.201.27.17 on 01/24/12 For personal use only.

Preliminary experiments were carried out with hydrogen sulphide alone in the reaction vessel a t GOO0 C. I t mas found t h a t the total pressure decreased by approximately 1% to a limit. This slight pressure change is insignificant compared to the pressure change for the decomposition of dimethyl ether in the presence of H2S, and was therefore ignored. In order to verify the experimental results of previous workers (6) the rate of decotnposition of dimethyl ether was measured for various added amounts of hq-drogen sulphide. A plot of relative rate (i.e., the rate in the presence of H2S divided by the rate in the absence of H2S) is shown in Fig. 1. The curves drawn have been calculated from Fig. 2 ,

FIG. I. Relative rate (vluu) against pressure of hydrogen sulphide for three different pressures of ether: @ 100 mm, and 0 20 mm of ether. The solid lines have been calculated from Fig. 2. FIG.2. R e l a h e rate (v/vo) against the percentage of hydrogen sulphide for three different temperatures: A 530" C, 0 520" C, and X 480' C.

A 300 mm,

to be considered later. For a given pressure of ether the relative rate increases to a plateau with a n increasing concentration of hydrogen sulphide. In this region the order with respect to hydrogen sulphide is obviously zero. As still greater quantities of H2S are added the relative rate increases linearly. I t is apparent from the curves shown in Fig. 1 that the length of the plateau depends on the pressure of ether; the lower the ether pressure, the shorter the plateau. This suggests t h a t the eflects depend only on the proportion of H2S relative to ether, rather than on the absolute amount of H2S. This result differs from that found in the nitric oxide inhibited decomposition of dimethyl ether (Part 11). A plot of relative rate against the percentage of hydrogen sulphide (Fig. 2) shows this dependence more clearly. Points were determined a t three different temperatures-480, 520, and 530" C. Within the experimental error, the plateau occurs a t approxinlately the same relative rate in all cases, although there appears to be a slight increase withdecreasing temperature, i.e. the plateau seems to be higher for the low temperatures. Kinetic measurelnents were carried out with about 40% added H a , and with about 75 to 99% added H S . Hydrogen sulphide in the amount of 40% was sufficient to ensure "limiting" acceleration (independent of H2S), while over 75y0H2S ensured t h a t the reaction was in the region dependent on the percentage of hydrogen sulphide. A typical pressure-time curve is shown in Fig. 3. Curve A is an actual tracing of the record obtained for the decomposition of 90 mm of dimethyl ether in the absence of H2S. Curve B is a similar tracing showing the decomposition of 85 mm of dimethyl ether

2012

CANADIAN JOURXAL O F

CHEMISTRY. V O L .
10.6

41, 1963

+0.4

r0.2

0.0

-0.2

Can. J. Chem. Downloaded from www.nrcresearchpress.com by 112.201.27.17 on 01/24/12 For personal use only.
LT

-0.4

:. m
J

-0.6

-0.5

-1.0

-1.2

-1.4 -1.5

-18

200

400

600

600

1000

1200

1.2

1 . 4

1 . 6

1.8

2.0

2.2

2.4

2.6

2.8

TIME ( s e c l

Log

P ,

lmml

FIG. 3. Typical pressure-time curves. Curve A is a tracing of the record obtained for the decomposition of 90 mm of ether in the absence of hydrogen sulphide. Curve B is a similar tracing showing the decomposition of 85 mm of ether in the presence of 475 mm of H2S. The temperature in both cases was 510" C. FIG. 4. Double logarithmic plot of inflection rate against ether pressure for various temperatures. The lines all have slopes of 1.5. These curves were determined for the decomposition in the presence of about 40% HBS(i.e. the plateau region).

in the presence of 475 mnl of H2S. Both runs were carried out a t 510 C. The accelerating effect of the added H2S is readily seen. Figure-4 shows plots of the logarithm of the rate against the logarithm of the ether pressure, for the reaction in the plateau region where about 40% H2S was added. All of the lines have slopes of 1.5. The corresponding Arrhe~liuscurve is shown in Fig. 5.

-1.4 1.24 1.25 1.26 1.27 1.28 1.29 1.30 1 . 3 1 1.32 1.33 2.1 2.2 2.3 2.4 Log 2.5

P ,
2

103

2.6

2.7

28

2.9

(rnrnl

in the presence of about 40y0 HBS(i.e. the plateau region). FIG. 5. Arrhenius plot for the deco~nposition FIG. 6. Double logarithmic plot of inflection rate against pressure of hydrogen sulphide a t 520' C. The ether pressure was coilstant a t about 20 mm. The slope of the line is 0.5.

McKENNEY AND LAIDLER: DIMETHYL ETHER PYROLYSIS. 111

2013

The three-halves-order rate constant may be expressed as k = 1.06X1014e-53,2001RT cc1I2 mole-lJ2 sec-l. This is to be compared with k = 1.82X1016e-61~8001RT cc1I2m01e-~/~ sec-I obtained by Imai and Toyama (6). The kinetic measurements made with 75 to 99% H2S are shown in Figs. 6, 7, and 8. Figure 6 is a double logarithmic plot of rate against pressure of hydrogen sulphide, for a

Can. J. Chem. Downloaded from www.nrcresearchpress.com by 112.201.27.17 on 01/24/12 For personal use only.

FIG.7. Double logarithmic plot of inflection rate against ether pressure for various temperatures. The lines all have slopes of 1.5. These curves were determined for the decomposition in the presence of 7 5 9 9 %

H2S. FIG.8. Arrhenius plot for the decomposition in the presence of 75-99% HZS.

constant (-20 mm) pressure of dimethyl ether. T h e Iine drawn has a slope of 0.5, showing that the order with respect to H2S pressure is one-half in the region of very high percentages of HZS. The logarithm of the rate minus one-half the logarithm of the hydrogen sulphide pressure is plotted against the logarithm of the ether pressure in Fig. 7. A plot of this kind enabled the dependence on ether pressure t o be determined, without the tedious task of holding the H2S pressure constant. All of the lines drawn in Fig. 7 have slopes of unity. The corresponding Arrhenius plot is shown in Fig. 8. The calculated activation energy for the reaction accelerated by very high percentages of hydrogen sulphide is 52.5 kcal mole-'. The rate constant may be expressed by k = 4.98X1014 e-52 . 5 0 0 / R T cc1/2 mole-1/2 sec-l.
DISCUSSION

The fact that the rate of decomposition of dimethyl ether in the presence of hydrogen sulphide relative to the rate in the absence of hydrogen sulphide depends on the proportion of H2Sto ether (see Figs. 1 and 2) suggests that competitive reactions are involved in the process. At these temperatures it is well known that dimethyl ether decomposes to form methyl and methoxy radicals. Hydrosulphide radicals are almost certainly formed from the decomposition of hydrogen sulphide. This is suggested by the fact that there is a slight decrease in pressure when H2S alone is present in the reaction vessel a t 600' C. The thermal decomposition of hydrogen sulphide has in fact been studied in the temperature range employed in this research (9). The evidence suggests that the decomposition

2014

CANADIAN JOURNAL O F

CHEMISTRY. V O L . 41,

1963

proceeds mainly by heterogeneous processes a t temperatures below 600" C. The radical HS would probably be formed in large quantities, and can also be produced from reactions such as
CH3

+ H2S

--t

CHa

+ HS.

Consequently, during the decomposition of dimethyl ether in the presence of hydrogen sulphide, there should be a plentiful supply of hydrosulphide radicals. These could initiate the ether decomposition by abstraction of a hydrogen atom,
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 112.201.27.17 on 01/24/12 For personal use only.
CHaOCH3

+ HS

--t

H2S

+ CHaOCHz.

The overall rate expression for the thermal decomposition of pure dimethyl ether is of the form (Part I and ref. 10)

T h e shape of the curves shown in Figs. 1 and 2 suggests the following relationships:

This reduces t o

Thus on a plot of relative rate (v/vo) against the percentage of hydrogen sulphide a plateau would be observed when the proportion of H2S to CHBOCH? is small and k"jk is the dominant term. However, as the ratio of H2S to CHBOCHB becomes larger the relative rate should then increase with increasing percentage of hydrogen sulphide. The rate expression for the reaction in the presence of H2S should therefore involve two terms-one predicting three-halves-order behavior with respect t o dimethyl ether and no dependence on hydrogen sulphide, the second predicting dependence with respect t o both hydrogen sulphide and dimethyl ether. Figures 6, 7, and 8 show that the order is one-half in H2S and first in CHBOCHZ, i.e. n = 1/2 and m = 1, for the reaction with high percentages of H2S. The mechanism to which these considerations lead can be written as:

hlcKESKEY AKD LAIDLER: DIMETHYL ETHER PYROLYSIS. I11

2015

where 14 represents a third body. Application of the conventional steady-state approximation yields for the radical concentrations:

Can. J. Chem. Downloaded from www.nrcresearchpress.com by 112.201.27.17 on 01/24/12 For personal use only.

~ Z (2k19) O [CH~OCHZ] = - --k~ k23a

lt2

[CH3

0 ~ ~ 3 1

[H~s]~"

+ k3 (2k -A [CH k23a) 'I2

'

The rate is given by

The first and fourth terms in this expression are small in comparisonTto the second and third terms, and hence may be neglected. The rate-can therefore-be written as

A crude estimate of the percentage of 1-12s a t which the change in order takes place can be made using the values for 81, and k19 from Table I. The ratio of the first to the second term is

This expression becoines greater than unity when the ratio [H2S]/[CH30CH3]becomes greater than 9.1. Thus the first term in the rate expression predominates when the pressure of hydrogen sulphide exceeds 90%. The transition actually takes place with about 50-7070 H2S. The agreement is therefore quite satisfactory in view of the uncertainties in the rate constants. Consequently, to describe the rate for thc plateau region the follou~ing expression can be written :

This expression is identical with that postulated by Imai and Toyama (6). The reason

2016

CANADIAN JOURNAL O F

CHEMISTRY. V O L . 41,

1963

TABLE I Kinetic parameters Frequency factor (sec-I, cc mole-I sec-I, or cc2 set-I) Activation energy (kcal mole-')

Reaction

k a t 800" K (sec-', cc mole-' sec-l, or cc2 mole@ sec-')

Reference See Part I Estimated (see footnote) Assumed See Part I Imai and Toyama (11) Estimated (see footnote) See Part I See Part I

Can. J. Chem. Downloaded from www.nrcresearchpress.com by 112.201.27.17 on 01/24/12 For personal use only.

NOTE:T h e values for reactions [I91 and [23a] were calculated in the same manner as in the Appendix of Part I. Four effective SH, by analogy with hydrogen peroxide, and three effective degrees of freedom were degrees of freedom were assumed for HS H2S + H HS 4- M. taken for M

for this is that reaction [19] is unimportant until large relative quantities of hydrogen sulphide are used. Thus with the neglect of reactions [I91 and [5], the mechanism is essentially the same as that postulated by Imai and Toyama (6). The reaction in this region is therefore predicted t o be three-halves order with respect to ether, in agreement with experiment. The predicted activation energy is given by E = E20+3(Ela-EZ3&). Substitution of the appropriate values given in Table I yields E = 52.8 kcal mole-l, in good agreement with the experimental value of 53.2 kcal mole-l. The predicted frequency factor is A = A20(2A1a/A23a)1/2 = 1.0 X lo1*cc1I2 sec-l. This is also in good agreem e n t with the experimental value of 1.06X 1014cc1I2mole-'I2 sec-'. In the region of high percentages of H2S the rate is given by

Experimentally the reaction is one-half order with respect to hydrogen sulphide and first order with respect to dimethyl ether in this region, in agreement with this rate expression. The predicted activation energy is E = E ~ o + $ ( E ~ ~ - E = z 56.3 ~ ~ ) kcal mole-l, in fair agreement with the experimental value of 52.5 kcal mole-'. The predicted frequency factor is A = Azo(2A19/A23a)112 = l.0X1014 cc1I2 sec-', whereas the experimental value is A = 4.98 X 1014CC'/~ mole-112 sec-l.

The Termination Reaction Darwent and Roberts (9) concluded, from a study the photolysis of H2S, that about 13% of the HS radicals disappeared by the reaction 2HS -+ Sz H z while about 87% reacted by the process 2HS --+ H2S S followed by 2 s M -+Sz M. In view of this evidence, the equilibrium process

has been included in the mechanism. The small quantity of S atoms produced may react with H2S a s shown or with CH3OCH3:

I t is suggested that reaction [-221 is faster than [24], and hence [24] can be ignored.

McKENNEY AND LAIDLER: DIMETHYL ETHER

PYROLYSIS. 111

2017

The concentration of HS radicals is large in this system and their combination will give rise to H2S2. This reaction will require a third body under these experimental conditions. The Initiation Reaction An important conclusion arises from this study with regard to the order of the initiation reaction: the dissociation of the ether molecule into the radicals CH3 and CH30. Consideration of the steady-state equations shows that the only apparent way t o obtain the correct pressure dependence in both H2Sand CH3OCH3 is by having the initiation in its low-pressure second-order region. This observation is important with reference to the mechanism of the uninhibited reaction, as discussed in Part I. The Dissociation Energy of Hydrogen Sulphide The present work provides a method of estimating the dissociation energy (D(H-SH)) of hydrogen sulphide. The difference between the activation energies for the reaction in the plateau region and the region of high H2S percentages is given from the overall rate expression as

Can. J. Chem. Downloaded from www.nrcresearchpress.com by 112.201.27.17 on 01/24/12 For personal use only.

The experimental difference is 0.7 kcal mole-'. Since El, is known (Part I) calculated as, a t high pressures, E19= El,-2AE
=

E l 9

can be

85.6 kcal mole-'.

Previous work has led to conflicting values for this dissociation energy; values varying from 81 to 95 kcal mole-' have been reported (12). Johns and Ramsay (13) recently determined the dissociation energy of the hydrosulphide radical and obtained D(H-S) = 81.4~t2.9 kcal mole-'. This can be used t o calculate a value for D(H-SH), since the heats of formation of S, H, and H2S are known (14). For the process SH -+ S H,

D(S-H)

AHlO(S)+AHP(H) --AHrO(SH)

whence AHfo(SH) For H2S -+ H

=

23.9 kcal mole-'.

+ SH
=

80.9 kcal mole-l.

Cottrell (12) favors a more recent value of 57 kcal mole-' for AHfO(S), and Ansdell and Page (15) support this value on the basis of their determination of the electron affinity of the sulphur atom. This value yields D(H-SH) = 84.65 kcal mole, in good agreement with our value. On the other hand, Mackle and McClean (16) prefer a value of 35.2 kcal mole-' for AHfO(HS), which was determined from a calorimetric study of benzyl mercaptan. This yields D(H-SH) = 92 kcal mole-'. Palmer and Lossing (17) recently determined AHfO(HS) < 33.7 by mass spectrometric techniques, and this leads to an estimate

2018

CANADIAN JOURNAL O F

CHEMISTRY. VOL. 41,

1963

of D(H-SH) ,< 90.6 kcal mole-l. The value is therefore still in doubt, but our value of 85.6 kcal mole-' calculated from the difference in activation energies supports the lower values. Admittedly the experimental errors involved in this estimation are large. The value of 85.6 kcal mole-l was taken for the calculations considered above.

The Efect of Temperature on the Relative Rate The main features of the thermal decomposition of dimethyl ether can be represented by the following scheme (see Part I ) :
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 112.201.27.17 on 01/24/12 For personal use only.

+ CH30CH3 -+ CH3 + CH3OCH3

M
CH3OCHZ

+
+

M + CH3O + CHaOCH, M + CH, + CH,

H

-+
-t

+ CH3O + &/I CH4 + CH3OCHZ CHZO + CH3 CH20 + H + M Hz + CHZOCHZ CzH6 + M.

CH,

[la]
[2I

[31
141

El
[71

T h e overall rate expression derived by means of the usual steady-state approximation is given by

I t was shown above that the rate expression for the reaction in the plateau region (-30 to 50y0 added HpS) can be written as

The relative rate is therefore

which reduces to

The predicted activation energy for the relative rate in this region is therefore

E~o-Ez+$(E~-Ez~~).

Substitution of the appropriate values from Table I yields E = -0.45 kcal mole-l. This is in good agreement with the experimental value determined by subtracting the activation energy for the reaction in the absence of H2S from the activation energ)- for the reaction in the presence of 40y0 HpS, i.e. 53.2-54.9 = -1.7 kcal mole-l. This explains why the relative rate seems to increase with decreasing temperature. The fact that -1.7 is so small is indicated from Fig. 2, which shows that the curves for the three temperatures fall very close to one another. Similarly for the region using high percentages of hydrogen sulphide the relative rate is given by

McKENXEY AND LAIDLER: DIMETHI'L ETHER PYROLYSIS. I11

The activation energy for the relative rate is predicted to be

E,o-E2+~(E,,+E7-E1,-E23,) 3.85 kcal mole-l.

This is in fair agreement with the experimental value of -2.4 kcal mole-l, considering the uncertainties in the individual activation energies.
ACKNOWLEDGMEXTS

Can. J. Chem. Downloaded from www.nrcresearchpress.com by 112.201.27.17 on 01/24/12 For personal use only.

T h e authors are indebted to Dr. B. W. Wojciechowski for valuable suggestions. The work was supported by grants from the National Research Council and the Petroleum Research Fund.
REFERENCES
1. 2. 3. 4. 5.
6.

K. E. J. BARRETT and W. A. WATERS. Discussions Faraday Soc. 14, 221 (1953). R. h-. BIKKELL, R. F. SMITH, A. F. TROTMAN-DICKENSON, and H. ~VILKIE.J. Chem. Soc. 2807 (1957).
J. A. KERRand A. F. TROTMA~;-DICKEKSOX. ' J . Chem. Soc. 3322 (1957). N. IMAIand 0. TOYAMA. Bull. Chem. Soc. Japan, 33, 1120 (1960). N. IMAI, Y. YOSHIDA, and 0. TOYAAIA.Bull. Chem. Soc. Japan, 35, 752, 759 (1962). N. IMAIand 0. TOYAMA. Bull. C h e n ~ Soc. . Japan, 34, 328 (1961). N. IMAIand 0. TOYAMA. Bull. Chem. Soc. Japan, 33, 1408 (1960). K. J. LAIDLER and B. W. WOJCIECHOWSKI. Proc. Roy. Soc. (London), Ser. A, 259, 257 (1960). and R. ROBERTS. Proc. Roy. Sac. (London), Ser. A, 216, 344 (1953). B. DE B. DARWEKT S. 1%'. BENSON and D. V. S. JAIX. J. Chem. Phys. 31, 1008 (1959). N.IVAI and 0. TOYAMA. Bull. Chem. Soc. Japan, 33, 652 (1960). T. L. COTTRELL. The strengths of chemical bonds. Butter\vorths Scientific Publications, London. 1958. p. 188. and D . -q. RAMSAY. Can. J . Phys. 391 210(1961). J . W. C. JOHXS SELECTED VALUESOF CHEMICAL THER~IODYNA~IIC PROPERTIES. Natl. Bur. Std. (U.S.), Circr. 500. 1952. D. A.ANSDELL and F. M. PAGE. Trans. Faraday Soc. 58, 1084 (1962). H. MACKLE and R. T . B. MCCLEAN. Trans. Faraday Soc. 58, 895 (1962). F . PALnfER-and F. P. LOSSING. J. Am. Chem. Soc. 84, 4661 (1962).

7. 8. 9. 10. 11. 12. 13. 14: 15. 16. 17.