Dec. 5, 1952
REACTIONS OF METHYL RADICALS WITH ACETONE [CONTRIBUTION FROM
THE
6005
CHEMISTRY DEPARTMENT, UNIVERSITYOF MANCHESTER 1
The Reactions of Methyl Radicals with Acetone BY M. T. JAQUISS, J. S. ROBERTS AND M. SZWARC* RECEIVED MAY21, 1952 The kinetics of the radical reaction (2) have been investigated, methyl radicals being generated by the thermal decomposition of di-t-butyl peroxide. The activation energy of (2) has been estimated a t 9.5 f 1.5 kcal./mole. This value and the value for kz/kt'lZ (kr being the rate constant for the bimolecular recombination of methyl radicals) are in fair agreement with those reported recently by Trotman-Dickenson and Steacie, who used a photochemical method for the generation of methyl radicals.
An extensive study of reactions of the type R H + .CH, --f R *+ CHI (1) has recently been carried out by Trotman-Dickenson and Steacie.' These authors have determined the activation energies and pre-exponential factors corresponding to various substrates, RH, by comparing the rate of reaction (1) with the rate of reaction ( 2 ) CHiCOCH3
+ .CHI
----f
CHBCOCH,.
+ CHI
(2)
Reaction 2 was investigated by many workers. Most of the studies being carried out by photolytic technique, and in view of the great importance of this reaction, we consider it useful to reinvestigate it by using a different technique and a different source of methyl radicals. The thermal decomposition of di-t-butyl peroxide seems to be a convenient generator of methyl radicals, and thus we investigated this decomposition in the presence of acetone. We have applied the technique described previously by Murawski, Roberts and Szwarc,2 and we refer the reader to this paper for all the details of experimental procedure. The Kinetics of the Decomposition of Di-t-butyl Peroxide.-The system di-t-butyl peroxide/acetone seems t o be relatively simple. The decomposition of di-t-butyl peroxide takes place according to the well established scheme, proved by the work of Rust, Vaughan and their colleague^,^ namely (CH,)aCOOC( CH3)3 --f 2( CHs)3CO*
(4)
+ CHsCOCH,
(5)
(CH,),CO.
.--f
CHJ.
Methyl radicals, generated in reaction (5), either dimerize 2CH3. +CiHB
(3)
or react with acetone4 producing methane according to reaction (2) CHoCOCH3 + .CH3 +CHSCOCH?. + CH, (2) It has been demonstrated by Raley, Rust and Vaughan3that the CHaCOCHz.radicals produced in reaction ( 2 ) recombine with methyl radicals to form methyl ethyl ketone CHICOCHZ.
+ CHI. + CHaCOCHxCH3
(6)
According to the above scheme, the rate of decomposition of di-t-butyl peroxide is measured by the rate of formation of GHP, CHI.
+
-d(peroxide)/dt
*
= d(CH4
+ C&s)/dt
College of Forestry, N. Y.State University, Syracuse 10, N. Y. (1) A. F. Trotman-Dickenson and E. W. R. Steacie, J. Chem. P h y s . , 18, 1097 (1950); 19, 163, 169, 329 (1951). (2) J. Murawski. J. S. Roberta and M. Szwarc, ;b;d., 19, 698 (1951). (3) J. H. Raley, F. F. Rust and W. E. Vaughan, THISJOWOY&, 70, 88 (1948). (4) Ie our experimonts acetone was alwayi in excera.
On this basis we calculated from our present results the uniniolecular rate constants of the thermal decomposition of di-t-butyl peroxide which are listed in Table I. TABLE I
Bxpt.
20 26 27 23 25 24 21 22 18 19
Temp., OC,
7 6 3 4 30 8 10 28 9 29 l6 17 34
126.5 127 127 127.5 127.5 129 131 131 146 147 148 148 148 150 150 151.5 152 152 152 153 154 162 164.5 165
13 14 15 31 33 12 32 11
165 165.5 167 167 168.5 169 168 175
5
Di+ butyl peroxide IO3 mole/l.
2.0 2.5 1.4 2.4 1.6 2.5 1.7 1.2 0.54 .49 .77 .88 1.5 0.60 .68 .77 .48 .31 .53 .27 .86 .49 .51 .39 .29 .50 .52 .46 .33 .33 .37 .15
Acetone 1OJ niule/l.
11 6.6 13 12 14 13 14 7.2 15 3.4 4.2 1.5 7.7 9.7 9.4 14 15 21 14 21 5.6 18 18
8.6 16 17 9.4 16 7.8 7.7 9.5 3.9
Reaction time, min.
120 120 240 120 240 180 200 200 100 100 75 30 35 75 75 50 60 90 64 110 30 18 18 23 18 16 16 17 18 16 16 18
Decomp. of perk X IO'. oxide, yo sec.-L
5.2 4.9 9.9 5.2 10.3 9.4 12.7 12.3 30 29 26 13.0 15.3 35 33 26 33 49 36 61 22 32 35 44 38 36 41 41 43 45 44 74
0.74 -70 .72 .71 .75 .91 1.13 1.10 6.0 5.7 6.7 7.7 7 9 9.5 9.0
10.3 11.5 12.3 11.8 14.4 13.9 36 41 42 45 47 55 52 5.5 62 60 125
These unimolecular rate constants are independent of the initial concentration of peroxide, of the initial concentration of acetone and of the time of heating. Moreover, the present results compare well with those reported previously by Roberts and Szwarc, 2.6 who investigated this decomposition in the presence of toluene. The agreement between these two investigations is excellent within the higher temperature range. At lower temperatures, the rate constants obtained in the experiments in which acetone was present are slightly lower than those obtained in the experiments carried out in the presence of toluene. (6)
M.gsrarc pad J. 9. Roberte, J . Chem. Phyr.. ld. 66L (1950).
1
The plot of tlir logarithm of the uniinolecular iate constant against 11T is given in Fig. 1. The "best" straight line corresponds to an activation energy of 37 kcal./mole. The activation energy reported by Roberts and Szwarc2s5(the result of an ili\estigation carried out ill a static system in the ~resenceof toluene) is 34 k 2 kcal. mole, the results of Murawski, Roberts and Szwarc' (obtained by a flow technique) lead to an activation energy of :j(i ri: 1 kcal./mole, and the value reported by Raley, Rust and Vaughan3 is 39 kcal. mole. In spite of tliese sinall diffcrences in the observed activation energies, we infer that in all these decompositions the rate of the same initial process generating methyl radicals has been measured, since the absolute rate constants reported i n all these studies q r e e \'cry well.'"
TABLE I1 Expl
L'tj I
t-
-.>
>$ ',
23 24 21
-.. I).,
1s 19 5 6 :3 1
28 2q 16 17 34 13 14 15 31 33 12 *'I 2
-4.n.
c
M
11
CH4, mmoles
1.33 I).KI 1.Y7 t .4t; 2.34 2.85 "88 I .53 2 . ?A 0.91 1.21 1.52 2.25 2.35 2.48 2.18 2.60 1.09 2.02 2.32 1.55 1.55 2.24 1.79 2.27 1.40 1.28 I .52 0,;s
7.0 .I 7 0 4 r, . 1
I37 127' 3 127 3
l
no x
s'
Temp, QC
126 5 127
20 i
LIolar ratio, acetone/ peroxide
,
8.(i 3.4
129 121 131 146 147 148 149 150
x 0 3 X ,
28 7.0 5.5 5.0 16 14 18 31 27 6 .;, 36 35 22 56 31 18
130 151 .?I 1.52 132 151 162 164.5 165 165 16,5 5 167 16i 168 5 169
34
32 23
25 25
lfi!)
17,:
CzHe,
mmoles
0.97 1.71
1.11 1. :io t ,3:: 2.27 1,xo 1.84 1.19
2.23 3.2 3.5 2.33 2.65 1.9s 1.29 1.63 3 .1 1.43 1.66 2.24 0.92 1.77 2.94 1.92 2.13 3.02 2,O(i 1 . 7,3
T h e plot of log (k2 'ksjl'j)against 1 1 2' is given in Fig. 2 . The "best" straight line corresponds to
I 22
-., ., .I
24
25
lo; 1'
I
Fig 1 -( CH:),COOC( CH3)3 '-f 2( CH3)tCO.
\O
The Rate of the Reaction (Z).-Since all the evidence indicates that methane is produced by reaction ( 2 ) and that ethane is formed in the bimolecular recombination ( 3 ) we may calculate the ratio k? kyl '2 from the equation k k,'/? = (r,tte of formation of CHa)/(rate of formatioil of CLH6)'/z[acetone] = ('irnotlnt of CH, formed)/{(amount of CJIE formed) X t'/? x [acetone])
denoting the time of reaction. This derivation iieglects the correction for the slight time depenctriice of the stationary concentration of methyl radicals,6 this correction being quite small. The values of kJka'/? are listed in Table 11. For a given temperature, kz k?'/f remains constant even if the conditions of experiments are changed in such a manner that the ratio of CH4/CzHe which are produced in the decomposition varies by a factor of 6.
\
_i_-______,
/
(3a) A paper by Lossing and Tickner, J . Chcm. P h y s . , 40, 907 :19.52), appeared after the present communication was sent for pdbiication. These authors extended the temperature range of the decornposition of di-1-butyl peroxide up to 350'. and they found that their rewlts, Rlnrawski, Roberts and Szwarc results, and Raley, Rust and Vaughan results give a set of mutually eonsistent date. (6\ V ,Szwnrc. { b i d . . 19,296 (1951>,
\
-0.7
2.2
2.3
I:ig. 7 . - CHI.
+
3.4 2.5 1031~. CN.COCH1 .-+CHI f .CIIrCOCIIL
TABLE I11
7', o c .
Steacie and TrotmanDickensonu k,/ka'/z, rc.'/f set, -1/2 moles--'/2
Jaqiii.;3, Roberts atid
Szwarc k2/k3'i/?, c c , '/? ser.
-'I2 m a l e s - ' / ?
121 0.27 0.20 131 .37 ,3n 150 .63 .50 167 .99 .75 a The values listed in the second column of this table were obtained from the graph reported b y Trotman-Tlickensotl rind Steacie fnr aretone pressure of 100 m m .
KIKETICS O F HOTHYDROGEN ATOMS
Dee. 5, 1952
E? - 1,/2E3= 9.5 kcal./mole and we consider this value to be correct within 1.5 kcal./mole. We note the very good agreement between this value and that reported by Trotman-Dickenson and Steaciel (9.7 + 0.1 kcal./mole). The values of kl/knl/Z obtained by us are lower by about !G%]than those recorded by Trotninn-T~ickeiisori niicl Steacie (see Table 111).
[CONTRIBUTION FROM
IN P H O T O L Y S I S OF
HYDROGEN HALIDES
6007
The agreement in the estimated activation energy and the observed rates for the processes producing methane in the photolytic system investigated by 'Trotman-Dickenson and Steacie and in the pyrolytic system investigated by us is particiilarly gratifying. I t assures us that we do indeed measure the sctivation energy for the radical reaction ( 2 ) . MANCHESTER 13, ENGLAND
DEPARTMENT OF CHEMISTRY, VX:IVERSITP.OF TOTRE DAME]
THE
The Kinetics of Hot Hydrogen Atoms in the Photolysis of the Hydrogen Halide~l-~ BY
HAROLD A.
SCHWARZ,4
RUSSELLR. WILLIAMS, JR.,
AND
WILLIAMH. HAMILL
RECEIVED APRIL29, 1952 The photolyses of hydrogen iodide, deuterium iodide and hydrogen bromide were studied in the presence of added inert gases. The inhibiting effect of the free halogens was found to be a function of the concentration of inert gas, the effect being independent of its nature a t large pressure of inert gas. The results are interpreted in terms of hot atom mechanisms. The moderating effect of the inert gas is accounted for semi-quantitatively and the difference of activation energies between the thermal reactions, El - E2: [H H I + H2 f I, kl; H I? H I I, k?] is found to be 4.5 rt 0.8 kcal. mole-'.
+
Introduction The photodissociaiion of the hydrogen halides with light of X 2537 A. yields hydrogen atoms with considerable kinetic energy, since the bond strength is much less than the quantum energy and this energy difference will be divided in the inverse ratio of the atomic masses. In the case of hydrogen iodide, spectral investigation^^,^ indicate the production of the 2Fa/zstate of the iodine atom and, therefore, hydrogen atoms of cu. 42 kcal. mole-' excess energy should be produced. The inhibitory effect of iodine in this photolysis, represented by the ratio kz/kl = 3.8 (independent of temperature) in pure hydrogen iodide7 increases to a value of 7.0
+
H H I +HP H+I?-HI+I
+I
ki k2
a t pressures of cyclohexane greater than 300 mm.* This effect was attributed to the moderation of the hot hydrogen atoms by cyclohexane and appropriate reactions were added to the classical mechanism H +HI-HP + I ka H+I:!---tHI+I kq H M+ H h4 k:,
+
+
where H is a hot hydrogen atom and M the moderating substance. In a similar fashion, the photolysis of hydrogen bromide should yield hydrogen atoms of ca. 26 kcal. mole-1 excess energy. In this case, the inhibition constant k7/k~ has not been measured H H
+ HBr
--f
Hz
+ Br
+ Br2 +HBr + Br
kg
ki
(1) From the doctoral dissertation of H. A. Schwarz. (2) Presented at the 121st Meeting of the Americap Chemical Society, Buffalo, N. y . , March, 1952. (3) Work supported in part under A.E.C. contract A t ( l l - 1 ) - 3 8 . (4) A.E.C. Predoctoral Fellow, 1949-1951. (5) C. F. Goodeve and A. W. C. Taylor, Proc. Roy.SOC.,London, 164, 181 (1936). (6) J. Romand, Comgt. rend., 227, 117 (1948). (7) R. R. Williams, Jr., and R. A. Ogg, Jr., J . Chem. Phys., 15, 691 (1947). (8) R.A. Ogg, Jr., and R.R. Williams, ihiif., 18. ,586 (1945).
+
+
directly, but is expected to have a value of 8.6, independent of temperature, from experiments on the photobroiiiination of h y d r ~ g e n . ~ It is the purpose of the present work to investigate more extensively the effect of inert diluents on these photolyses in the light of the hot atom mechanism previously proposed. Experimental The extent of photolysis was measured through titration of the halogen produced. The high degree of precisioii required in the analyses did not permit initial additions of halogen t o the reaction mixture, which consisted of the hydrogen halide with various amounts of inert gas. Four photolyses were conducted simultaneously in four quartz cells held in rigid positions with respect t o each other and rotated past a light source. In each group one cell, containing only the hydrogen halide, was used as an actinometer for the other three, which contained added moderator. The relative extents of photolysis in the various cells could be but since the quantity repeated with a reproducibility of 170, of interest, the inhibition constant, is obtained by a difference calculation (see equation 2), the reproducibility of this quantity is approximately 10%. Apparatus.-Each quartz cell (35 X 75 mm.) was provided with a fused, quartz window (Fig. 1). The other end was constricted to 8-10 mm. and led t o a graded seal. A side-arm was provided for freezing the hydrogen iodide while adding the inert gas. The cells were secured permanently in copper tubes soldered to a brass sleeve which could be fastened to a rotating steel shaft. The cells were rotated in a furnace a t 140 r.p.m. No attempt was made to stabilize the speed of rotation since only the average value of the product (intensity) X (time) for each cell is significant. The cells passed the light source about 17,000 times during an experiment and even transient variations due to rotation or illumination were effectively averaged out. The furnace consisted of an aluminum box about 150 mm. on each edge with two nichrome heating elements in opposite sides. A 35-mm. square window in the bottom of the furnace was covered with a quartz plate. Temperature was measured with a 360" t h e r m o m p - in an aluminum sheath and was regulated within f l by manually adjustinga Variac The Hanovia Sc2537 lamp employed in this work is stated by the manufacturer to give about 85% of its light as the 2537 A. resonance line. Materials.-All reagents, including those used in preparations, were C.P. or analytical grade. Hydrogen iodide
.
(9) M, Bodenstein and G. 2, Jung, 2. pkrsik. Chem., 121, 127 (1926).
