3754
C. DAVIDBASSAND GEORGEC. PIMENTEL
reported in the literature for k d k s and ka/ks. In Table I1 we have summarized recent data on disproportionation over recombination for normal alkyl radicals. The data reported for n-butyl radicals should be taken as maximum values since the authors assumed that all butene originated from disproportionation of n-butyl radicals. Other radicals were, however, present in their system. Butene may, therefore, be formed not only by TABLE I1 RATIOSOF DISPROPORTIONATION OVER RECOMBINATION FOR ALKYL RADICALS D/R
CHI
-36
+ +
CHI ~-CIH, CHI n-CdHo CHI t- n-CaH11
+
GH, C J I S
Author
0.04 .06 .05 I W .1
.12 .14 .15
[CONTRIBUTION FROM
1“B
Ausloos and Steacie” Helle8 Ausloos and Murad’ This investigation Ausloos and Steacie” James and Steacied Brinton and Steacie’
n-C,H,
+ n-CsHt
VOl. 83 .15
Ausloos and Murad’ .10 Blacet and Calvertf .17 Masona n-C,Ho n-C,Hp 0.57-1.09 Kerr and Tmtman-DickensonA This investigation a Ref. 7. * C. A. Heller, J . Chem. Phys., 28, 1255 (1958). OP. Ausloos and E. Murad, J . A m . Chcm. SOL, 80, 5929 (1958). d D. C.L. James and E. W. R. Steacie, Prpc. Roy. SOL.(London), A244, 289 (1958). * R . K. Brinton and E. W. R. Steacie, Can. J . Chem., 33, 1840 (1955). P. E. Blacet and J. G. Calvert, J . A w . Chcm. Soc.. 73,661 (1951). 0 Ref.4. * Ref.2a.
+
+
the reaction 2n-CpHe3 C ~ H I O C4Hs but also by R n G H 9 -t CaHs f RH. Excluding the data on n-butyl radicals Table I1 indicates that there is a general agreement on the relative importance of disproportionation and recombination reactions of normal alkyl radicals. Acknowledgment.-The author wishes to express his sincere thanks to Mr. J. A. Guercione for carrying out the experiments described in this paper.
+
DEPARTMENT OF CHEMISTRY, UNIVERSITY OF CALIFORNIA, BRRKRLEY, CALIFORNIA]
Hydrogen Abstraction from Hydrocarbons by Methyl Radicals from the Photolysis of Methyl Iodide in Solid Nitrogen BY C. DAVIDBASSAND GEORGEC. PIMENTEL RRCEIVED MARCH7, 1961 Methyl iodide has been photolyzed a t 20°K. in solid matrix materials, NI, Kr and Xe, containing hydrocarbons [GHs or (CH,),CH] or deuterated hydrocarbons [CD,, CHsCDs, or (CH&CD]. Hydrogen abstraction was studied by infrared detection of CHI and C H J I . I n the solid the abstraction products can be attributed to methyl radicals with an “effective temperature” in the range 1000-3000°K. Furthermore, the products obtained from photolysis of methyl iodide with ethane present as well as those from photolysis of ethyl iodide in nitrogen, indicate that about 85% of the reactions probably occur within the cage at the site of photon absorption. These studies provide information concerning the dissipation of the energy of a “hot” radical constrained within a reactive cage.
The “cage effect” hypothesis explains the characteristically low quantum yield of primary dissociation in a condensed medium.’ The solvent molecules surrounding a site of photon absorption form a “cage”; in collisions with this environment, primary fragments dissipate excess energy before separating far enough to escape recombination. Though the qualitative features of the cage effect have been discussed by a number of workers,2 much less is known about the quantitative aspects of this process. Evidence concerning the cage inhibition of photolytic decompositions in matrix isolation studies has been reviewed by Pimeritel.* It is clear that some substances, for example methyl iodide, resist photolytic decomposition even though the excitation energy which must be dissipated (per mole) may be more than one hundred times the (1) I . Franck and E. Rablnowitch, Trans. Faraday SOC.,30, 120 (1034). (2) See. for example, F. W. Lampe and R . Xi. Noyes, J . A m . Chenr Soc.. 7 6 , 2140 (1054); M. Szwarc, J . Polymer Sci., 16, 367 (1955); R. Luebbe and J. Willard. J . A m . Chcm. SOC.,81, 761 (1058). and references cited therein. (3) Chapter IV, “Radical Formation and Trapping in the Solid Phase,” by C. C. Pimentel, “Forination and Trapping of Free Radicals,” ed. A. Baa8 and H. Broida. Academic Presa, Inc.. New York, N. Y.. 1060.
molar heat of fusion of the solid environment (as it is in solid argon or nitrogen). We have sought a more detailed understanding of the fate of the excitation energy through studies of the abstraction of hydrogen and deuterium from hydrocarbons by methyl radicals produced through photolysis of methyl iodide in solid nitrogen. Experimental The gases were mixed in a three-liter flask in a vacuum system. The bulb was painted black to prevent gas phase photolysis reactions and it contained Teflon chips for mixing. Approximately 10’ micromoles of the hydrocarbon was admitted to the bulb and condensed at 77’K. Then 10’ micromoles of methyl iodide mas measured into a smaller calibrated flask and transferred to the larger flask a t 77°K. Finally, 25 X 100 micromoles of the matris gas was expanded into the sample flask. The concentrations, expressed in mole ratios, are approximately: M / R H = 25 and M/CH,I = 250 (M =. matrix, R H = hydrocarbon). The flask was shaken to agltate the Teflon chips and thus t o ensure mixing. The vapor pressure of CD, a t 77’K. is about 20 mm. Therefore. when R H was CD,, the order of introducing hydrocarbon and methyl iodide was reversed. The low temperature cell was a duplicate of a cell designed by Van Thiel.‘ The gaseous mixture was admitted t o the cell a t reduced pressure by passing it through a metal needle (4) M. Van Thiel, Ph.D. Thesis. University of California, Berkeley, 1058.
Sept. 20, 1961
PHOTOLYSIS OF METHYLIODIDE IN SOLID NITROGEN
3755
valve (Hoke) greased with Apiezon-N lubricant. When the Table I in terms of the peak value of log Io/I matrix was N1, the spray-on rate was 40 t o 60 mm./hour per micromole. When a range is given, deviation from the three liter bulb (about 160 micrornoles/inin.). from Beer's law was observed. The lowest value The matrix materials Kr and Xe required slower deposition (about 0 mni./hour) to reduce light scattering. Approxi- corresponds to the limiting slope and the higher mately 5 x 10' micrornoles of the gas mixture was deposited value gives the slope when 30 micromoles of sample in each experiment. was deposited. All quantitative interpretations Infrared spectra were recorded using a Beckmati IR-7 discussed here are based on the actual curves of or a Perkin-Elmer Model 21 spectrophotometer, as indicated in the tables. Frequency measurements were made with growth. In any experiment in which both CHaD the IR-7 instrument arid the reproducibility was 1 2 ern.-'. and CHI were produced, the absorption a t 1304 Inteusity reproducibility was better than 10%. Slit widths cm.-l caused by CHaD was estimated from the were small compared to band widths: effective band widths intensity of the band a t 1155 an.-'. This small were 1.3 cm.-1 a t 960 cm-1, 1.7 cm.-I a t 11m1250 and 1.5 cm.-1 at 1300 cm.-l. (For experiments using the correction was subtracted and the remainder was Perkin-Elmer 21, the effective band widths were somewhat ascribed to CHI. larger but still smaller than the observed widths of the bands.) TABLE I The samples were irradiated with Light from a General APPARENT ABSORPTION CORFIWIENTS OF VARIOUS SUBElectric H A 4 medium pressure mercury arc with the outer STANCES SUSPENDED IN SOLIDNITROGEN^ Pyrex jacket removed. The beam was focused on the sample with a fused quartz lens. Output of the lamp was T =I 2OoK., DO =I peak value of log & / I per micromole. a t least 1018 quanta per hour. Substance .(em. -1) D, Chemical Preparations.-CHaD.-This compound, syuCHJ 1243 0.0098-0.0061 thesized by Grignard reaction from CHJ, contained 95% CIHJ 1215 .011 CHID and 6% CH..' There was a negligible amount of CHID,. GH, 960 .038 ( 0 . 0 1 ~ 5 ~ ) ( CHa)rCD.-This compound, synthesized by Grignard CIL 1304 .0137 reaction,' was washed through concentrated HiSol with CHaD 1155 ,0055 f 0.0003 nitrogen gas (which removed butenes) aud frozen a t 77'K. 1304 .0027 f 0.0003 The sample contaiued about 70% (CHa)rCD and 30% (CHs)sCH.' 'The curves of growth are presented in the Ph. D. GHs (Phillips), CH, (Phillips), (CH&CH, (Matheson), thesis of C. D. Bass, University of California, Berkeley, Cd& (Phillips), CD, (Bio-Rad Laboratones), C&CD, 1961. b Peak absorption after warming sample to about (Merck).-These samples were purchased, and each was 40'K. and recooling to 2OoK. stated t o be 99 mole %pure. C&I (Eastman Kodak, reagent).-Methyl iodide was TABLB I1 outgassed a t 77'K. three times and then once again before PIIOTOI.YSIS OF CHJ IN VARIOUSM A T R I CCONTAININQ ~ each experiment. NS(General Dynamics Gorp.).-Nitrogen gas was passed GHi The over copper filings a t 500' t o remove traces of gas was passed through a trap at 77'K. and collected iu il CHII = 210; (&Hi 21; T = 20'K. large storage flask. Xe (Air Reduction Sales Co.).-Stated impurities (by mass spectrometric analysis) in mole % were: HI, 0.005; Nz, 0.005; Kr, 0.005; 0 1 , O . O O l . Kr (Air Reduction Sales Co.).-Stated irnnurities (by .Nn 0 0.163 17.0 miss ippkctrometric analysis) in 'mole per cent: were: H*, 10 .117 0.068 11.8 0.79 less than 0.001; Ns, 0.005; Xe, 0.010. 30 ,091 -105 21 9.1 .95 0 . 6 9 Results 80 .076 ,126 32 7.6 .95 .89 120 .063 .151 33 6 . 3 1.00 .78 Absorption Coefficients of Products.-QualitaR3 0 .149 15.3 tive and quantitative analyses were based on refer10 .116 ,030 11.8 0.60 ence spectra of known samples suspended in the 8.8 .64 .69 .088 .059 10 30 appropriate matrix under conditions duplicating ,084 18 7.7 .77 .77 60 .077 the photolysis experiments. The sample was -78 1.00 120 .lo4 30 5.7 ,057 deposited in several measured fractions to provide a curve of growth. The spectrum was recorded Xe 0 .117 , 11.8 .. for each fraction and the peak intensity was plotted 8.4 .47 0.77 10 .084 .Or8 5 vs. micromoles deposited.' Unfortunately, only .64 * .045 6.8 .068 30 a fraction of the sample actually is condensed in .77 1 . 0 0 60 .053 .071 20 5.3 the optical path (probably more than two thirds .86 1.10 120 .040 4.0 .094 31 of the sample). Though the absolute extinction a Integrated intensities.' ACH:I equals the decrease in coefficients are not known, the ratios of true micromoles of C H d from time =I 0. Not measured.
a.
-
,.. ..
..
..
...
..
..
..
.
..
..
...
..
..
..
*
extinction coefficients can be measured provided only that all of the compounds studied have the same accommodation coefficients. The ratios are sufficient for the purposes of a material balance. The apparent extinction coefficients are given in (6) We are grateful to Dr. B. H. Mahan and Dr. A. Newton for mass spectral analyses of the CHID nod (CHahCD samples. (6) We are grateful to Dr. A. C. McRowe for his preparation of (CHdCD. (7) The area, rather than the peuk absorption, was measured for the band of ethylene at 960 cm.-1 because the band w a s found to broaden from a half-width near 6 c m - 1 to about 20 cm. -1 if dlfiusioo was permitted. This change In band shape resulted in no change In the band area, within experimental uncertainty.
CZHJ in N2.-Ethyl iodide, methane, ethane and nitrogen were mixed in the mole ratios 1 :1:20:480 and condensed a t 20'K. The sample, containing 27.6 micromoles of ethyl iodide (log 10/1= 0.302 a t 1215 cm.-'), was photolyzed for 30 minutes and 79% of the GHJ was decomposed. Ethylene was produced (16.0 micronloles) as revealed by the absorption a t 960 ern.-* (see footnote 7). The ratio of micromoles GH4 produced per micromole C2Hddecomposed was 0.76. CHJ C a s in N1, Kr, and Xe.-The results of experiments in which CHaI was photolyzed in the
+
C. DAVIDBASSAND GEORGEC. PIMENTEL
Vol. 83
TABLE I11 PHOTOLYSIS OF CHII IN NITROGEN CONTAININQ HYDROCAREONS T = 206K..M/CHd = 220 Expt no.
1
Cum position
CD: M/CD, = 25 CzHs CD4 M/CzHo = 45 M/CD4 = 35 CHaCDs M/CHsCDI 21
Photolysis Time (inin.)
CHII
1248 Em.-!
log
c
690
rei
1166 cm.-l
+
CHI CHID
c
...
CHII
...
...
.005
I
I
+
CH4 CHsD ACHII
21.6 ... ... 8.7 > 12 0.77 -005 7.6 > 13 .80 ... ... ... 18.1 14.0 1.04 ,009 .78 1.17 .016 10.9 .84 9.6 1.10 .021 .91 19.6 ... ... ... 16.0 1.01 ,010 .99 12.3 ,016 .89 1.32 9.7 .021 .85 1.27 ... ... ... 19.6 14.5 .OlO 1.21 .77 11.4 .018 1.17 .84 1.15 .023 .87 9.6 26.1 ... ... ... 20.7 ... ... -71 16.7 .86 ... ... 13.2 .96 ... 1.10 10.4 ... ... 24.6 ... ... ... 18.6 ... 0.72 ... 0.81 14.0 ... ... 12.2 1.01 ... ... 9.3 1.08 ... ... 23.6 ... ... 16.6 < ,005 0.43 > 4 12.6 < .005 .60 > .a .91 8.7 .014 4.6 ACHJ equals the decrease in micromoles of CHJ
<
