The Pyrolysis and Photolysis of Acetone-d6 in the

Mixtures ofacetone-dg and methane were pyrolyzed in the range 475-525° and photolyzed in the temperature range. 350-428°. The rate of reaction of CD...
0 downloads 0 Views 329KB Size
J . K. MCNESBYAND A. S.GORDON

4196

Vol. 76

u.s.S A V A L ORDXANCE TESTS T A r I O S ]

[CONTRIBUTIONFROM THE RESEARCH DEPARTMENT,

The Pyrolysis and Photolysis of Acetone-& in the Presence of Methane BY J. K. MC~\'ESBY ASD

X.S.GORDON

RECEIVED APRIL 17, 1954 Mixtures of acetone-ds and methane n-ere pyrolyzed in the range 475-525" and photolyzed in the temperature range 350-428'. The rate of reaction of CD3 with CH, to form CDIH relative to the rate of formation of CD, from CD3 and acetone-& was studied as a function of temperature. The activation energy for the reaction of CDBwith methane t o form CDIH is 2.74 kcal./mole greater than for CD3 with acetone-ds to form CD4. From this value the activation energy for the abstraction of hydrogen from methane by CD3 is shown to be 14.0 kcal.!mole. I t has been demonstrated that the CDaH/ CD, ratio is somewhat surface dependent; a sixfold increase in surface/volume ratio increases the CD3H/CDa ratio by about 10%.

Introduction The kinetics of reactions of methyl radicals with a number of alkanes have been studied' with the notable exception of methane. I t is obviously impossible to measure the rate of formation of CHI from the reaction of CH3 with CH4, but a study of the reaction of CD3with CH4 to give CD3H appeared to be feasible. Such a study has been attempted by Trotman-Dickenson and S t e a ~ i e who ~ , ~ employed a photochemical technique without success. In the present work it was noted that the reaction of methyl-& radicals with methane would be much slower than with the parent acetone-ds, and the study was made with large CHd/acetone-dG ratios. This technique increased the rate of reaction of CD3 with CHI relative to CD3 with the acetone-& impurity so that most of the CD3H formed would arise from the former reaction and the blank correction would be relatively small. 1.22

Experimental Preparation of Materials. A . Deuterioacetone.-'i'hc preparation of this compound has been described in a prcvlous paper.4 I t was stored in a flask equipped with a mercury cut-off valve. The composition was 95 parts acetoned6 and 5 parts acetone-&, with an unknown but small arnouiit of acetone-&. B. Methane.-The methane was obtained from the Southern California Gas Company. Mass Spectrometer analysis showed the material t o be 99.6yc methane and 0.4yc ethane. 2. Procedures.-Mixtures of methane and acetone-ds were made up and the compositions examined mass spectrometrically. A master mixture of known composition was kept in a %liter flask equipped with a mercury cut-off valve. Two such mixtures were used in the course of this work. A series of pyrolyses and photolyses were run using an initial methane:deuterioacetone ratio of 5.76. A second mixture whose methane: deuterioacetone ratio was 5.34 was used for the pyrolyses a t 475 and 523', and also for the photolyses a t 326 and 432'. In these photolyses the surface/ volume ratio was increased by a factor of about six by inserting the appropriate amount of fused silica rod into the reaction vessel. An aluminum block furnace was used to heat the cylindrical reaction vessel which was of fused silica ivith plane parallel windows. The reaction vessel was 4 cm. in diameter and 4 cm. high. The maximum gradient The across the vessel was 3.5' a t 350' and 5' a t 428'. light source was a Hanovia SC-2537 mercury vapor lamp in the form of a flat spiral. The pyrolyses were carried out, using a technique described previously. In the photolyses the system was evacuated, brought t o temperature, and the lamp turned on. IVith the lamp shielded, a sample of the master mixture was admitted t o the system until the total pressure was about 90 mm. The shield was removed and the photolysis begun. After a time the lamp was turned off and the reaction mixture allowed to expand into an evacuated Pyrex vessel equipped with a break-off seal and an Apiezon wax cut-off. The wax was melted, trapping the sample in the sampling vessel. 3 . Analysis.-A gross analysis of the sample was performed in order t o ascertain the relative amounts of methane and acetone-& remaining a t the end of the reaction. A sample volatile from liquid nitrogen was admitted into the mass spectrometer and the ratio of the '9/2~ mass peaks (CD3H/CDa)was recorded. This procedure was followed for all the experiments reported in this paper.

2

1,25

1.35

1.13 lOOO/

1.,'j,i

1.65

T.

Fig. 1.-Reactiori of CU, with methane and acctoiic-&. Dashed line is effect of 6 X increase iri surfacelvolume ratio. ( I ) A . P. Trotman-Dickenson, J . R Birchard a n d E. 1". R Steacie, J . Chsni.Phys., 10, 163 (1951). ( 2 ) A . F. T r o t m a n - D i c k e n s o n a n d E. U'. R . Stearje, ihad , 19, 835 i1 9.7 1 ) $ 3 ) A 1:. ' I ' r o I ~ i ~ u tIi) i r b - t . s ( 4 < i r 1 iiii,l lc. \V il. Str:tr.ir, i / i i * l , 1 9 . !)I!) (Iu.;I)

Results and Discussion The results are given in Tables I and 11, and the -1rrhenius plot is shown in Fig. 1. The experiments without added surface are designated by open circles for photolyses and filled circles for pyrolyses. Experiments with added surface are denoted by half filled circles. Considerable difficulty was encountered in obtaining reproducible results in the photolyses but not in the pyrolyses. In the latter cases methane was formed almost to the exclusion of ethane, while in the photolyses a t low temperat i r e s , ethane is the nixjor Iiydrocitrbon product. ( 4 ) J , R \Ic,Nrsl,\,. 'l' 7 0 . X?R ( I r J T I I .

\ V l),ivi,

,cii,l

4 S . ( ; ~ > r < l r ~' tl ',i i i s I ~ ~ I ~ , I ~ Y A I

REACTION OF

CD??

Run no.

T, 'C.

1 2 3

4 0

6 7 8 9 10 11 12 13 14 15

16 17 18 19 20 21 22 23 24

348 347 350 351 349 348 348 428 428 428 428 429 429 428 475 475 475 475 475 499 496 498 498 498 523 523 523 524 523

4197

PYROLYSIS AND PHOTOLYSIS OF ACETONE-&

Aug. 20, 1954

CD5

i, min.

0.688 ,753 ,862 ,074 ,071 ,069

4 8 16 4 8

16 0

.800 .830 .864 ,062 .061 ,062

3 6 10 3 6 10 0 10 20 30 30 0 5 15 15 30 0 2 4 6 5

TABLE I CD3 RADICALS W I T H METHANE kd4 M1

ki(A-da

+ A-dz

(A-ds

+ A-da)

2

ka/ki

+ log k d k i

1000/1'

0.9877

1.fins

5 76

0.614 ,682 .792

0 0 0 5.76

,560 ,738 ,769 ,802

0.0972

0

,711

0 0 5.76

.123-1-

1.091s

1,427

0 5.34

,1416

1.1511

1.337

.1476

1.1691

1.297

.1586

1,2003

1.256

,851

,837 .834 ,085 .756 ,940 ,930 ,085

0 0

,085 .850

5.76

.847

5.34

,945 ,928 ,938 .091

0

0 TABLE I1

INFLUENCE OF 6 X INCREASE IN SURFACE ON REACTIOK OF CD3 WITH METHANE Run no.

25 26 27 28 29 30 31

T , 'C.

t , min.

C Z C D4

kaM ki(A-ds A-dr)

326 326 326 326 43 1 432 433 434 432

3.0 5.0 4.0

0.654 .661 ,066

0.586 ,593

0 1.1 3.0 1.0 3.0 0

,796 ,808 .063 .062

+

.575 .733 * 745

.726

(A-da

.M/

+ A-da)

kdki

2

+ log k d k r

lOOO/T

0 0.534

0.1077

1.0322

1.669

0 0 0.584

0.1360

1,1335

1.418

+

+ +

(1b) I n order to obtain enough methane for analysis, the CD3 CD2HCOCD3 +CD4 CDHCOCD3 low temperature (150-250°) photolyses had to be CD3 + CHD2COCD3+CDsH CDzCOCD3 (2a) carried out to large percentage conversions of ace- C D ~ H+ C D ~ C O C D+ ~ C D ~ H+ CD~COCD, (2b) tone-ds. This caused large variations in the CD&/ CD2H + CD2HCOCD3 CD3H + CDHCOCD3( 2 c ) CD4 ratio when runs were made to different percentaEe conversions. The Dhotochemical studv CDI CH4 +CDJH f CH, (3) was lone in the range 350 td 428' in order to in'- CD2H CDJK!OCD1 + CD2HL CDzCOCDJ ( 4 ) CDLH + CHd +CD2H2 CH? crease the proportion of methane in the products. (5) In I and 4J " CD3H kPa(CD3)(A-dS) + k P ~ ( C D ~ H ) ( A d ~ k2,(CDzH)(A-d0) ) ka(CD3)M (I) G, 10, 11, 12, 16, 19, 20, 24, k~a(CDa)(A-d,) klb(CDa)(A-do) CD4 27, 30 and 31 were made with acetone-ds alone. The CDIH arises from the CDIH - kax(CD3)(A-ds) (kZb(A-d6) k d A - d , ) ) c D ~ H (kia(A-ds) $- kib(A-ds))CDa reactions 2 and 3 , while CD4comes from reaction 1. CD4 The following reactions are important k3M (11) +

+

+

+

+ +

+

+

hv

CD3

+

+ +

CDaCOCD? --+ 2CD1 CO CO CDiCOCD? +2CDj CDBCOCD3 +CDI CDtCOCDs

+

(MA-&)

(in) (Ib)

(In)

+

+

+ kidA-d,))

where A-ds and A-d6represent the concentrations of acetone-d6 and ncetone-d6, respectively. Early in the reaction CDzH/CDa = LY = constant.

EDWIXJ. HART

4198

Since a is initially independent of the amount of added CH4, the first fraction on the right-hand side of equation I1 can be obtained to a close approximation as the CD3H/CD4 ratio in the absence of added CH4. If the reasonable assumption kla = k l b is made, equation I1 becomes CDsH/CDc = (CDaH/CDd)a f (k3/ki)(Jf/[(A-&)

+

(A-&)l) (111)

where (CD3H/CD4)ois the ratio in the absence of added methane. Thus k3/kl can be obtained from the initial ratio of mass 19/niass 20, (CD,H/CD4)o and a knowledge of the composition of the initial mixture. I n the temperature range 350428' the CD3H/CD4 ratio was dependent upon the time of photolysis for methane : acetone-& mixtures. Since this ratio was found to be almost linear with time, a short extrapolation to zero time was applied to the ratio after correction for (CD3H/CD4)0. This gave (k3/k1)(M/[(A-ds) (A&)]). I n contrast to the photolysis, the M/[(A-d6) (Ads)] ratio did not change appreciably in the course of pyrolysis, SO that it was not necessary to extrapolate to zero time. The constancy of this ratio was reflected in the constancy with time of the CD3H/CD4 ratio in the pyrolyses. In the case of the photolyses, the smaller percentage of methane in the products made i t necessary to carry out the reactions to greater percentage conversions of deuterioacetone in order to obtain enough products for analysis. Hence extrapolation of CD3H/CD4 t o zero time was necessary in the photolyses. The line drawn through the points in Fig. 1 was drawn by the method of least squares. Since

+

+

k 3 / k l = (A8/Al)e(EI-Ea)'RT

(IV)

where A , and A, are the Arrhenius pre-exponential factors for reactions 3 and 1, respectively, a plot of log k3/kl ZIS. 1/T should give a straight line of slope ( E , - E3)/4.576 and intercept log A3/AI. Using this treatment and applying the method of least squares, E3 - El = 2.74 kcal./mole and A3/A1 = [ C O N T R I B U T I O N FROM T H E C H E M I S T R Y

Vol. 70

0.48. It has been shown4J that El = 11.3 kcal./ mole, and therefore E3 = 14.0 kcal./mole. These activation energies are based upon an energy of activation of 9.6 kcal./mole for the abstraction of H from acetone by CD3 and an activation energy of zero for the methyl-methyl combination. Since CH3 and CD, have previously been shown5 to take a hydrogen atom from normal acetone with equal facility, it is reasonable that CD, and CH3 abstract a hydrogen atom from methane with equal facility. Figure 1 demonstrates that i t is entirely practical to study reactions of methyl radicals using both photochemical and pyrolytic techniques concurrently. In addition to giving results in harmony with photolysis, the pyrolytic technique can be used, as it has been here, to establish the presence or absence of hot radicals. I n the present case it is evident that hot radicals cannot be present in the photolysis since they are absent in the pyrolysis and the two techniques give consistent results. Table I1 and Fig. 1 show clearly that the reaction of CD3 radicals with CH4 is partly heterogeneous relative to the reaction of CD3 with CD3COCD3. X sixfold increase in the surface area caused a ten per cent. increase in the ratio k3/kl. If the reaction of CD3 with acetone-& is homogeneous, it may be concluded that the rate of the homogeneous reaction between CD3 and methane is at least 98% of the observed rate. Therefore, if the CD3acetoneds reaction is homogeneous, the data obtained without added surface represent the homogeneous reaction of CD3 with methane within the limits of experimental error. If homogeneity of acetone decomposition were established beyond any doubt, the relative heterogeneity of methyl radical reactions could be studied conveniently and quantitatively using the technique outlined here. Acknowledgment.-The authors wish to thank Dr. S. Iiuven Smith and X r . William Golyer for the mass spectrometric analyses. (.5) J. K SIciXesby and Alvin S. Gordon, THISJ O U R N A L , 7 6 , 1416 (1924).

INYOKERN, CHINALAKE,CALIF.

DIVISION,ARGONNEN A T I O N A L

LABORATORY]

y-Ray-Induced Oxidation of Aqueous Formic Acid-Oxygen Solutions.

Effect of pH

BY EDWIN J. HART RECEIVED

MARCH 29, 1954

Radiation yields of carbon dioxide, hydrogen and hydrogen peroxide formed and oxygen reacted have been measured a t 0.01 'iA formic acid concentration over the pH range from 0.32 t o 11.58. In the pH range above 3, molecular hydrogen and hydrogen peroxide are formed in equimolar amounts. Below this range of pH, an excess of hydrogen peroxide is formed. The results are interpreted in terms of number of water molecules dissociated in three primary reactions: (1) HxO = H OH, ( 2 ) H%O= '/2Hz02 f 1/2HZ,and (3) H20 = H 1/2H202. Each of these three reactions is pH dependent. However, the total number of water molecules decomposedjlO0 e.v. IS substantially constant in the PH range 0.32 to 11.58 and equals 3.99 f 0.14.

+

+

The formic acid-oxygen system has been employed to measure the radical pair and molecular product yields of ionizing radiations in aqueous solutions.' Specifically, the yield of oxygen consumption, G(-oz), provides a measure of free rltdicals :Lvailable, whereas the yieltf of hytlrogell extrap(,-

lated to infinite oxygen concentration, G(H*),is a measure of reaction 2.

(1) (a) li. J. I I a r t , J. P h y s . Chem., 66, 594 (1952); ( b ) THISJ O I I K N A I . , 76, Srgt. (1984); (c) R a d . Rescurch, 1, 53 (1984).

i~ I , ; % l ~ r ~ lr ~ r proyr t H S I . - I . I ? I Y ( 2 ) A. 0 Alleu, H r < ~ , k h n v rNsliirnd (1963); R a d . Kescurch, 1, 85 (lY54).

HZ0 = H I I , ~= */.tr2

+ OH -+

~