3572
J. Phys. Chem. 1983, 87, 3572-3578
c h l o r ~ p r o p a n e .The ~ ~ ratio of isomers was the same for unsensitized focused decomposition at a different COz laser wavelength and the focused SiF,-sensitized decomposition. The ratios of the reaction rates for formation of the different isomers were calculated as a function of energy (photolytic mechanism) and temperature (thermal mechanism). The focused SiF,-sensitized ratios (and the focused unsensitized ratios) were consistent with 1,2-dichloropropane reacting with an energy content of 117 kcal/mol or at a temperature of 2000 K. From arguments about the calculated reaction time, the photolytic mechanism was preferred. Similarly, the ratios of isomers with unfocused SiF,-sensitized decompositon was attributed to a thermal reaction at 1100 K. For the focused SiF,-sensitized decomposition, a mechanism of vibrational excitation of 1,2-dichloropropane into the quasicontinuum by SiF, followed by C 0 2laser pumping was suggested. Based on the present work, an alternative mechanism would be electronic energy transfer from an excited electronic state of SiF,. It is surprising, however, that with focused radiation the isomer ratios do not change with pressure up to 32 torr at a SiF4/1,2-dichloropropaneratio of 2.
The sensitization of acceptor molecules by IR multiphoton pumped SiF4in an excited electronic state should not be unique to SiF,. Inverse electronic relaxation has been observed in a number of IR multiphoton pumped molecules. The possibility of other molecules acting as electronic state sensitizers will depend on the energy and lifetime of the electronic state. The photophysical properties of BC13bear a strong resemblence to SiF4.31 BC13 exhibits a prompt broad-band luminescence from 440 to 660 nm. As with SiF,, the emission does not correspond to any known transition of the parent molecule or its fragments and could well arise from a Franck-Condon shifted electronic state. Collisionally induced emission from an excited electronic state of BC1 is observed in BCl, comparable to the SiF emission in SiF,.
(30) W. Tsang, J. A. Walker, and W. Braun, J. Phys. Chem., 86,719 (1982).
(31) V. N. Bourimov, V. S. Letokhov, and E. A. Ryabov, J . Photochem., 5 , 49 (1976).
Kinetics of CI(,P,) and CH,CICHCI,
Acknowledgment. I thank Sherman McCutcheon, 111, for his excellent technical assistance and Dr. Wayne Sharfin for his help in setting up the initial experiments. I also thank one of the referees for suggesting SF, as a sensitizer. Registry NO.CF2,2154-59-8;SiF4,7783-61-1;CFZHCl, 75-45-6.
Reactions with the Chloroethanes CH3CH,CI, CH,CHCI,, CH2CICH2CI,
P. H. Wine* and D. H. Semmes Molecular Sciences Group, Engineering Experiment Station, Georgia Instnute of Technology, Atlanta, Georgla 30332 (Received: December 10, 1982)
-
The kinetics of the reactions Cl('PJ) + RHCl RC1+ HCl were investigated over the temperature range 257426 K for RHCl = CH3CH2C1( k 2 ) ,CH3CHC12(k3),CHzCICHzCl(k,), and CHzCICHClz(k5). c1(2PJ)was produced by 355-nm pulsed laser photolysis of Clzand monitored by time-resolved resonance fluorescence spectroscopy. The data are adequately described by the following Arrhenius expressions (units are cm3molecule-' s-l, errors are 2u and refer to precision only): k z = (2.34 f 0.42) X lo-" exp[-(310 f 56)/T], k3 = (8.19 f 1.84) x exp[-(554 f 71)/T], k4 = (2.21 f 0.51) X lo-" exp[-(793 f 73)/T], and k5 = (4.88 f 1.41) X exp[-(786 f 88)/77. Under some experimental conditions evidence for Cl('PJ) regeneration via a secondary reaction was observed. At 258 f 1 K, deviations of Cl('PJ) temporal profiles from first-order behavior were attributable to the reactions RC1+ clz RClz + c1(2PJ)(k,). By modeling the observed Cl('PJ) temporal profiles, we found the rate constants kj to lie in the range (5-12) X lo-', cm3molecule-' s-l for all RC1 investigated. The reactivity trends observed in reactions of Cl('PJ) with C2H,C16,, x = 3-6, are discussed.
-
Introduction Reactions involving abstraction of a hydrogen atom by ground-state atomic chlorine, Cl('PJ), have been studied extensively by kineticists for many years. Early work, which employed end product analysis techniques to determine relative rate coefficients for reactions of Cl('PJ) with hydrogen, alkanes, and chlorinated alkanes, was motivated by the desire to test the ability of various theories to predict the rate coefficients for a series of related reactions.'V2 In recent years, the controversy concerning
the extent of chlorine-catalyzed destruction of stratospheric ozone3 has led to renewed interest in reactions which convert c1(2PJ) into the relatively stable reservoir species HC1; this has motivated the application of modern "direct" kinetic techniques in numerous investigations of c1(2PJ) reactions with hydrogen-containing atmospheric constituents such as Hz, hydrocarbons, H02, and H2COe4 A number of Cl('PJ) + RH reactions have also been investigated as initiation reactions in model systems for (3) F. S. Rowland and M. J. Molina, Reu. Geophys. Space Phys., 13,
(1) G. C. Fettis and J. H. Knox, Prog. React. Kinet., 2 , l (1964), and references therein. (2) H. S. Johnston and P. Goldfinger, J.Chem. Phys., 37, 700 (1962). 0022-365418312087-3572$01.50/0
1 (1975).
(4) "Chemical Kinetic and Photochemical Data for Use in Stratospheric Modelling", Evaluation No. 5, J P L Publication 82-57, Jet Propulsion Laboratory, Pasadena, CA, 1982, and references therein.
0 1983 American Chemical Society
The Journal of Physical Chemistry, Vol. 87,No. 18, 1983 3573
Cl(2PJ)Reactions with the Chloroethanes
studying chain-reaction kinetic^.^ Because c h l o r d a n e concentrations in the stratosphere are relatively low, their reactions with Cl(?J) have received relatively little attention. Two studies of Cl(,P J) reactions with chloromethanes have been reported,6i7while the only kinetic investigation of a Cl(,PJ) + chloroethane reaction reported to date is a recent study in our laboratog of the reaction Cl(,Pj) + CH3CC13 CH2CC13 HC1 (1)
+
-+
We found reaction 1 to be surprisingly slow-at 298 K C1(2PJ)reacts with ethane a t least 1300 times faster than with CH3CC13. To better understand this unexpected result we decided to extend our measurements to include the following reactions:
-
c1(2PJ)+ CH3CH2Cl
products
(2)
c1(2PJ) + CH3CHC12 products
(3)
+ CH2ClCH2Cl- CHzClCHCl + HC1 Cl(,PJ) + CH2C1CHC1, products
Cl(,Pj)
(4)
(5)
The results of our investigations of reactions 2-5 are reported in this paper.
Experimental Section The experimental apparatus is described elsewherea8 c1(2PJ) was produced by 355-nm pulsed laser photolysis of C12 and detected by time-resolved resonance fluorescence spectroscopy. A CW electrodeless discharge lamp (gas mixture: 0.1% Cl, in He) was used as the fluorescence excitation light source. The lamp output was filtered with a calcium fluoride window and an N20gas filter to prevent impurity emission of H, 0, and N resonance radiation from entering the reactor. Signals were obtained by using photon counting techniques in conjunction with multichannel scaling. All experiments were carried out under “slow-flow” conditions. The linear flow rate through the Pyrex reactor was 3 cm s-l and the laser repetition rate was 1Hz. Under these conditions, a fresh reaction mixture was available for each laser shot. Dilute mixtures of the chloroethane reactant (RHCl) in argon were prepared in 12-L bulbs. RHCl, Cl,, Ar, and, in some cases, O2 were premixed before entering the reactor; the concentration of each component in the reaction mixture was determined from measurement of the appropriate mass flow rates and the total pressure. The fraction of RHCl in the RHCl/Ar mixtures was checked frequently by UV photometry at 185.0 nm (mercury resonance line). The required absorption cross sections were measured during the course of the investigation. They are as follows (in units of cm2): CH3CHC12,1.31 X CH2C1CH3CH2C1,1.34 X CH2Cl, 5.92 X 10-19;and CH2C1CHC1,, 1.58 X The CH3CH2Cl cross section agrees well with a previously published value.g The gases used in this study had the following stated purities: Ar > 99.995%, O2 > 99.9970, C1, > 99.970, He > 99.99970, and CH3CH2C1> 99.7%. CH3CH2C1was degassed repeatedly at 77 K before use while all other gases were used as supplied. The CH2ClCH2Clsample was Fisher reagent grade (no stabilizers added) and had an ( 5 ) D. J. Nesbitt and S. R. Leone, J. Chem. Phys., 72,1722(1980);75, 4949 (1981);J . Phys. Chem., 86,4962(1982). (6)R.G.Manning and M. J. Kurylo, J.Phys. Chem., 81,291 (1977). (7)M. A. A. Clyne and R. F. Walker, J. Chem. Soc., Faraday Trans. 1 , 69,1547 (1973). ( 8 ) P. H. Wine, D. H. Semmes, and A. R. Ravishankara, Chem. Phys. Lett., 90,128 (1982). (9) C . Hubrich and F. Stuhl, J.Photochem., 12,93(1980).
overall purity of -99%. Special analyzed samples of unstabilized CH3CHC1, and CHC1,CH2C1 were obtained from Dow Chemical. The CH3CHC12sample had an overall purity of 99.76 mol %; major impurities were CH3CH2Cl(0.15 mol %), CHCl=CHC1(0.07 mol %), and CH3CC13(0.015 mol %). The CH,ClCHCl, sample had an overall purity of 99.20 mol %; major impurities were CH2C1CC13(0.48 mol 5%) and CH2ClCH2Cl(0.32 mol %). All liquid samples were purified by vacuum distillation with only the middle fraction used in experiments.
Results All experiments were carried out under pseudo-firstorder conditions with RHCl in large excess over Cl(,PJ). Argon was used as the buffer gas at a pressure of 100 torr, thus ensuring that a thermal distribution of Cl(,PJ) spinorbit states was maintained throughout the course of the reaction.’O Typical atomic and molecular chlorine concentrations were [C1(2PJ)]o= 2 X 10” atoms cm-3 and [Cl,] = 6 X l O I 3 molecules ~ m -although ~, both concentrations were varied over a wide range as checks on the system chemistry. To study the kinetics of reactions 2-5, it is desirable to establish experimental conditions where the Cl(,PJ) temporal profile is governed entirely by the following processes: c12 Cl(,PJ)
-
+ RHCl-!.
(6)
2Cl(,PJ) RC1
+ HCl
i = 2-5
(i)
Cl(,PJ) loss by diffusion from detector field of view and reaction with background impurities (7) Then, since [RHCI] >> [c1(?J)], simple first-order kinetics are obeyed: In ~[c1(2P~)]o/[c1(2P~)]~) = (ki[RHCl] + k,)t = k’t
(8)
The bimolecular rate constant, k,,is determined from the slope of a k’vs. [RHCl] plot. Observation of Cl(,PJ) temporal profiles which are exponential (i.e., obey eq 81, a linear dependence of k’on [RHCl], and invariance of k’ to variations in laser photon fluence and [Cl,] strongly suggests that reactions i, 6, and 7 are the only processes which affect the c1(2PJ)time history and, therefore, validates the measurement of k,. The presence of reactive impurities in the RHCl samples would, of course, not be elucidated by the above set of observations (as long as [impurity] >> [c1(2PJ)]). Over the entire range of experimental conditions investigated, factor-of-5 variations in laser photon fluence had no effect on the observed c1(2PJ) temporal profiles. This observation proves that biradical processes, such as reaction of Cl(,PJ) with a product of reactions i or with a photolytically produced radical, were unimportant. At T = 298 K, all observed Cl(,PJ) temporal profiles were exponential with k’ independent of both laser power and [Cl,]. However, at both the lowest and highest temperatures investigated, nonexponential decays were observed. In all cases, the initial c1(2PJ) decay rate was faster than the decay rate at later times. Since the aforementioned variation in laser power demonstrated that the initial fast decay rate was not due to a radical-radical reaction, it must be that regeneration of Cl(?J) at long times via secondary reactions was the cause of the observed nonexponential temporal profiles. At T < 298 K, nonexponential behavior could be eliminated either by reducing [Cl,] or by adding O2to the reaction mixture to scavenge the intermediate (10)A. R.Ravishankara and P. H. Wine, J.Chem. Phys., 72,25(1980).
3574
The Journal of Physical Chemistry, Vol. 87, No. 18, 1983
Semmes
Figure 2. k' vs. [CH,CHCI,] data obtained under conditions where secondary prcductbn of Cy+,) was negligible. SOIUlines are obtained from unwwted llnear least-squares analyses; the sbpes of these lines give the bimolecular rate constants k,(T).
t (ms) Figure 1. Typical data obtained under conditions where signal-to-nolse ratio per laser shot was high, Le., b w [RHCI] and [O,] = 0. Reaction: Cl('P,) CH,CH,CI. Experimental conditions: T = 258 K; [Cl,] = 8.3 X loi3molecules cm-,; laser energy fluence = 20 mJ om-'; [CH,C~ 0,(b) 1.61,(c) 4.28,(d) H,CI] in units of loi3 molecules ~ m =- (a) 9.22; number of laser shots averaged = (a) 8,(b) 64,(c)128,(d) 256. Pseudo-first-order CI('P,) decay rates obtained from the data are (a) 27, (b) 187,(c) 410, and (d) 784 s-'.
+
free radicals RC1. Hence, it appears that, at low temperatures, nonexponential behavior was due to the reactions RC1 + c12 RCl2 Cl(,Pj) 0') At high temperatures the observed temporal profiles were insensitive to variations in [Cl,]. However, nonexponential decays were not observed if 0, was added to the reaction mixture. When sufficient 0, was present (usually -0.5 torr), exponential decays were observed and k i ( T ) was found to be independent of laser power, [Cl,], and further increases in [O,]. At temperatures in the 400 K range, and in the absence of 02,drastic changes in kinetic behavior were observed for small (- 10 K) temperature increases, suggesting that the secondary reaction(s) of importance have large activation energies. Two possible secondary reactions which would result in Cl(,PJ) temporal profiles which are independent of variations in [Cl,] and laser photon fluence are +
+ Cl(,Pj) ClRRH + C1(2Pj)
RC1+ R RC1 + RHCl
(k)
(1) Attempts to quantitatively model the secondary chemistry are described later in the paper. For determining ki(T ) , the important conclusion is that at all temperatures and for all RHCl it was possible to establish experimental conditions where Cl(,PJ) temporal profiles were exponential with a characteristic lifetime which was independent of [Cl,], [O,], and laser photon fluence but linearly dependent on [RHCl]. Nearly 300 experiments (experiment determination of one pseudo-first-order rate constant) were carried out. The attainable signal-to-noise ratio (S/N) was limited by absorption of resonance lamp radiation by RHCl and 0,; as a result S/N decreased dramatically as [RHCl] and [O,] increased. Typical high S / N data are shown in Figure 1 while typical data obtained under conditions where S/N was low is included in Figure 5. Typical plots of k' vs. [RHCl] are shown in Figure 2. Linear dependences of k' on [RHCl] were observed in all cases where decays were exponential. The experimental results are summarized in Table I. (Results of some preliminary experiments, which +
qualitatively examined the dependence of Cl(?PJ)temporal profiles on [Cl,] and laser photon fluence, are not included in Table I). Those experiments where secondary chemistry complications were encountered are identified with asterisks; none of the data so identified were used to determine ki(T). Errors quoted for individual ki determinations are 2u and refer only to the precision of linear least-squares fits of the k'vs. [RHCl] data. Where two or more kivalues were averaged to obtain a rate constant, the overall precision is conservatively chosen to bracket all individual ki's and their 20 uncertainties. In those cases where only a single ki(T) was determined, the precision is set at *lo%, which is a little larger than the typical uncertainty observed when several determinations were averaged. The absolute accuracy of the results is limited by precision and uncertainties in the determination of the reactant concentration. We estimate the absolute accuracy of each reported k i ( T ) to be f15%. The data for reactions 2-5 are adequately described in Arrhenius form (i.e., a linear In k vs. T 1dependence). Unweighted linear leasbsquares analyses give the following Arrhenius expressions (units are cm3 molecule-' s-l):
k2 = (2.34 f 0.42)
X
k3 = (8.19 f 1.84) X
lo-'' exp[-(310 f 5 6 ) / q
lo-',
exp[-(544 f 71)/7")
k, = (2.21 f 0.51) X lo-'' exp[-(793
f 73)/7")
k, = (4.88 f 1.41) X lo-', exp[-(786
f 88)/T]
The errors in the above expressions are 2a and represent precision only (uA = AaM). Experimental rate constants and best-fit Arrhenius lines are plotted in Figure 3.
Secondary Chemistry On the basis of the observations discussed above, we have attempted to model secondary chemistry using the following reaction mechanism: C1(2Pj)+ RHCl RCl
+ c12
k, 4
-
HC1 + RCl
RC12 + c1(2Pj)
RCl -% R
i- Cl(,PJ)
RC1 + RHCl-% HRRCl
-
+ C1(,PJ)
(9
ci) (k) (2)
k7
c1(2PJ) loss by processes other than reaction i RC1-
km
(7)
loss by processes which do not result in Cl(,PJ) production (m)
The Journal of Physical Chemistty, Vol. 87,No. 18, 1983 3575
Cl(2PJ)Reactions with the Chloroethanes
1
J
!\
4,-
*.\
t
2O
2.5
3.0
3.5
4.0
iOOO/ T Figure 3. Arrhenius plots for reactions 2-5. SolM lines are obtained from unwelghted llnear least-squares analyses.
The rate equations for the above reaction scheme can be solved analytically to yield the Cl(,PJ) temporal profile:
where
K = kj[Cl2]
+ k k + k[RHCl] + k,
X1 = 0.5((a2- 40)'/, - CY)
+ CY) = ki[RHCl] + K + k7 0 = kkkl + kjkl[Cl,] + kik7[RHC1] + klk7 = - O . ~ ( ( C Y-~40)'/,
X2
(Y
According to eq 9, if any (or all) of the secondary reactions j , k, or 1 are important, a double exponential decay is observed with one component, i.e., X1 or A,, faster than the
"primary" decay rate ki[RHC1] + k7. Kinetic data for reactions k have been critically reviewed by Benson and 0'Neal.l' On the basis of their recommendations, rate constants for reactions k are calculated to be s-l at 258 K and about 100 s-' at 400 K. Reactions 1 are approximately thermoneutral and are expected to have substantial activation energies; hence, it seems safe to assume that these reactions are also negligibly slow at 258 K. It was not surprising, therefore, that the nonexponential decays observed at relatively high [Cl,] became well-behaved (i.e., exponential) at low [Cl,]. Except for a few trial calculations, which are discussed in more detail below, the model calculations for the 258 f 1 K data assumed kk = kl = 0. Directly measured values for k7,and ki values obtained at low [Cl,], were used in the calculations. A trial-and-error procedure was employed to obtain the best pair of rate constants kj and k, to fit each temporal profile. A typical fit is shown in Figure 4, while a summary of all results is given in Table 11. The data are fitted very well by eq 9, giving k j values of (5-12) x cm3 molecule-' s-' for all RC1 reactants. We estimate the accuracy of the kj's determined from these model calculations to be f50% -provided, of course, that the assumed mechanism includes all reactions of impor~
~~
~
(11) S. W.Benson and H.E. ONeal, Natl. Stand. Ref. Data Ser. (US., Natl. Bur. Stand.), No.21, 1970.
1
40
80
t (ms) Figure 4. Typical Cl(2PJ)temporal profile at low T , high [Cl,]. Reaction: CI('PJ) CH,CICH,CI. Experimental conditions: T = 257 K, [CI,l = 1.09 X 1014 molecules cm-', laser energy fluence = 20 mJ cm- , 50 laser shots averaged. The solid line is obtained from a two-parameter fit to eq 9 using k,[CH,CICH,CI] = 160 s-l and k, = 19 s-l as determined in independent experiments and assuming kk = k, = 0. The best-fit line Is obtained for the choice of parameters k, = 38 s-l and k, = 7 X lo-'' om3 molecule-' s-'.
+
tance in producing and destroying Cl(,PJ). If we relax the restriction that kk = kl = 0 at 258 K, then it is possible to reproduce the exponential decays observed at low [Cl,] by assuming values for ki greater than or equal to those obtained under the assumption that kk = kl = 0 in conjunction with very fast rates for (kk kl[RHC1])and k,. However, when such a choice of rate coefficients is employed, it is impossible to reproduce the nonexponential decays observed at higher [Cl,] because the slower components in the observed decays cannot accommodate the required fast k,. Hence, the observation of nonexponential decays at high [Cl,] and exponential decays at low [Cl,] leads to a fit for k j and k, which is unique within the confines of the proposed mechanism. Even though reactions j certainly become faster with increasing temperature, no evidence for nonexponential c1(pJ)temporal profiles was observed at T = 298 K. Due to an unidentified background reaction, C1(,PJ) removal in the absence of RHC1, i.e., reaction 7, was found to be much faster at T 1 298 K than at lower temperatures (k7 -20 s-' at 258 K and -75 s-l at 298 K). Equation 9 predicts that, for k j = 2 X cm3 molecule-' s-' and k7 = 75 s-', deviations from exponential decays would be unobservable under the experimental conditions employed. Hence, the more rapid background Cl('PJ) removal rate at higher temperatures made it unfeasible to determine activation energies for reactions j . It is worth noting that, in the presence of sufficiently large levels of Cl,, evidence for reactions j should have been observable even with the faster value for k7. However, an experimental problem, interference due to vacuum-UV C12fluorescence excited by the chlorine resonance lamp, prevented experiments from being carried out under the required conditions. Because nonexponential Cl(,PJ) temporal profiles were not observed at 298 K, the possibility that the observed Cl(?J) kinetics were influenced not only by reactions i and 7 but also by reactions k, 1, and m cannot be completely ruled out. However, the fact that addition of -2 X 10l6 O2 cm-3 to the reaction mixture had no effect on the observed kinetics argues strongly against this possibility. If we assume that addition of O2 to RC1 proceeds with a cm3 reasonable bimolecular rate coefficient of 5 X molecule-' s-l at 100 torr of Ar, then addition of 2 X 1OI6 O2cm-3 to the reaction mixture would increase k , by lo4
+
3570
Wine and Semmes
The Journal of Physical Chemistty, Vol. 87, No. 18, 1983
TABLE I: Kinetic Data for the Reactions of Cl(*P.,) with RHCla 10-13 x 10-16 x laser [Cl, I, [O,], fluence, no. of T. K cm-3 cm-3 mJ expts range of k ' . s-' 258 27 5 299 352 355 416 419
258 298
349 412
257 278 298
303 3 26 3 50 353 365 377 389 401 404 413 426 258 275 298
8.3 7.2 6.0 6.4 5.9 4.9 5.4 4.7 5.2 5.2
0
0 0 0 1.3 0 1.3 0 0.88 1.8
20 20 25 5 20 25 20 20 20 20
6 5 5 4 5 5 5 6 4 4
CH,CH.CI = i7-784 40-766 96-1050 87-826 98-1 1 5 0 80-707 78-1180 76-842 86-1070 79-995
4.0 11.3 6.0 6.5 6.5 6.5 6.6 5.8 5.8 4.9 4.9
0 0 0 0 0 1.8 2.1 0 2.2 0 2.0
20 20 20 20 4 20 20 20 20 20 20
6 3 5 4 4 6 6 4 5 4 5
CH,CHCl, 23-683 20-480 94-84 1 69-712 55-666 104-629 78-652 66-658 102-624 85-636 79-716
3.1 10.9 5.8 3.1 10.8 10.8 6.5 5.1 6.0 6.5 5.9 5.4 5.4 5.2 5.0 5.1 2.4 9.6 9.6 5.4 4.8 4.8 4.3 4.2
0 0 0 0 0 0 0 0 2.3 1.9 0 0 2.4 0 0 0 0 0 0 1.4 3.2 6.0 0 0
20 20 20 20 20 4 20 20 20 20 20 20 20 20 20 20 20 20 4 20 20 20 20 20
5 3 5 6 4 4 4 5 5 5 5 6 5 5 5 5 3 3 3 5 5 2 5 3
CH, ClCH,Cl 22-574 19-452 26-607 103-63 8 107-602 104-524 84-592 69-635 78-539 122-568 75-704 84-829 108-628 67-644 74-675 76-643 77-361 62-412 75-378 125-500 85-643 82-458 b 82-245
3.7 7.6 13.9 2.9 6.6 2.5
0 0 0 0 0 0
20 20 20 20 20 20
5 2 2 3 5 3
CH,ClCHCI, 32-347 23-162 18-162 54-368 38-464 72-430
1oi3(kit 20), cm3 ( 1 0 i 3 k , ( ~ )cm3 ), molecule-l s - ' molecule-' s-l 70.3 f 3.9 76.1 t 2.9 79.2 t 1 . 5 79.6 * 4.9 82.4 f 3.3 83.8 t 8.7* 102 t 3 82.1 i. 4.3* 110 * 1 107 c 11 9.74 t 0.55 10.6 * 0.3* 13.8 i. 0.6 12.0 i. 0.6 1 3 . 5 t 0.8 11.0 * 0.5 12.6 f 0.6 16.2 i. 0.7 16.0 t 1 . 0 18.6 t 0.5* 22.0 t 0 . 6 9.70 t 0.44 9.69 i. 1.05* 12.9 t 0.9 14.9 t 1.1 1 5 . 5 t 0.4 13.9 t 0.6 14.0 t 0.2 14.3 t 0.2 15.0 t 0.6 1 7 . 3 t 1.1 20.6 f 0.9 23.1 t 0.4 23.2 f 0.7 24.3 ?r 0.8 22.6 t 0.6* 20.1 f 1.4* 15.2 t 0.8* 18.8 t 0.3* 16.4 f 1.3* 28.7 t 1.3 31.6 t 1.4 30.6 b 5.3 f 0.2* 2.38 t 2.75* 2.81 * 2.68 t 2.82 * 3.24 ?
70.3 76.1 80.4
* i.
7.3 7.6
f
5.7
102 t 1 0 1 0 9 f 13 9.74
12.6
f
*
0.97
2.1
16.1 t 1.1 22.0
i.
9.70
2.2
t
0.97
12.9 t 1.3
14.6 t 17.3 t 20.6 i. 23.1 t 23.2 t 24.3 t
1.3 1.7 2.1 2.3 2.3 2.4
30.3 t 2.9
0.11
2.38
0.04 0.21 0.17
2.75 t 0.28
i. 0.24
6.1 0 4 61-476 20 3.44 t 0.06 4 6.1 0 4 63-460 3.31 t 0.14 6.2 20 1.7 3.26 t 0.34 5 88-484 3.31 f 0.39 347 5.4 20 0 4 84-534 4.81 ?r 0.28* 5.4 20 1.7 5 108-609 5.38 t 0.22 5.38 t 0.54 411 2.3 0 20 3 75-375 4.09 f 0.14* 4.9 0 20 4.20 t 0.07* 3 93-399 5.1 2.1 20 3 99-505 7.06 t 0.73 3.5 5.1 4 94-412 20 7.05 t 0.81 7.04 f 0.80 a Data affected by secondary chemistry and, therefore, not used to determine h i are indicated by asterisks. Decays were such that k' values could not be estimated, i.e., not exponential for even a single l i e time. s-l. The fact that the Cl('PJ) kinetics were unaffected by suggest k , 30 s-l at that temperature. addition of 0, means, therefore, either that reactions k , At temperatures above 400 K, very complicated kinetic
-
1, and m are of negligible importance or that, in the absence of 02,k, >> lo4 s-l. The latter possibility seems virtually impossible in light of the fact that the data at 258 K
behavior was observed. Cl(,PJ) temporal profiles were nonexponential but independent of laser photon fluence and [Cl,]. However, nonexponential decays were elimi-
The Journal of Physical Chemistry, Vol. 87, No. 18, 1983 3577
CI('PJ) Reactions wRh the Chloroethanes
TABLE 11: Kinetic Data for the Reactions RC1 + C1,
ki --f
RCI,
+ CI('PJ)
hi [ RHCl ] ,
a t 258
t
1 Ka
1O-'4[cl2],
best-fit km, s - '
k,, s ' '
best-fit k ., 10-14 cm1 molecule-' s"
possible RC1
S-'
CH,CHCI CH,CH,Cl
162
CH,CH,Cl 27
0.828
30
5
CH3CC1, CHC~,CH,
134 478 141
CH,CHCl, 20 20 23
1.13 1.13 0.398
25 40 40
9 13 13
CH,CICHC~
121 424 118
CH,ClCH,Cl 19 19 23
1.09 1.08 0.318
38 38 10
7 7 7
CH,ClCCl, CHCI,CHC~
131 142
CH,ClCHCl, 18 23
1.39 0.764
27 25
7 9
Uncertainties in rate constants are estimated to be *50%. TABLE 111: Arrhenius Parameters and k ( 2 9 8 K ) for Reactions of Cl('PJ) with Ethane and Mono-, Di-, and Trichloroethanesa
E/R k(298 K ) re E CH,CH, 90 57 4 CH,CH,Cl 310 8.3 this work CH,CHCI, 5 54 1.3 this work CH,ClCH,Cl 793 1.5 this work CH,ClCHCl, 7 86 0.35 this work 8 CH,CCl, >1200 400 K. Time-resolved detection of additional reactive intermediates will be required to unravel the complex chemistry in this temperature regime.
Discussion Arrhenius parameters and 298 K rate coefficients for reactions 1-5 and the additional reaction Cl('PJ)
-
+ CH3CH3
CH2CH3+ HCl
(10)
are tabulated in Table 111. The parameters for reaction 10 are those recommended by the NASA panel for chemical kinetic and photochemical data evaluation! Data from other laboratories with which we can compare our results are extremely sparse. There have been no other absolute measurements of any of the rate coefficients kl-k5. A competitive kinetics measurement of the ratio k2/klo has been reported by Pritchard et a1.12 By measuring the relative consumption of CH3CH3and CH3CH2Clin the presence of Cl(?J) these workers obtained the result k2/klo = 0.38 exp[-(250 f 50)/T]. Using our results for kp(T) and the klo(7') taken from ref 4, we obtain k2/klo = (0.304 f 0.042) exp[-(220 f 1 4 0 ) / q . The agreement with the competitive kinetics results is excellent. Our results demonstrate that substitution of C1 for H in ethane and chloroethanes dramatically reduces the "abstractability" of remaining H atoms. Both A factors and activation energies are affected by C1 substitution, but the increase in activation energy with increasing C1 substitution is the dominant factor controlling the observed decrease in reactivity. The activation energy for the re(12) H.0.Pritchard, J. B. Pyke, and A. F. Trotman-Dickenson,J.Am. Chem. Soc., 77, 2629 (1955).
3578
The Journal of Physical Chemistry, Vol. 87, No. 18, 1983
action of Cl('PJ) with CH3CC13is found to be much larger than that for reaction with CH2C1CHC1, even though both RHCl reactants contain the same number of H atoms and C1 atoms. A similar result is obtained for OH reactions with CH3CC13and CH2C1CHC12.13 Although C-H bond dissociation energies (BDE) do not seem to be available for any of the chloroethanes except CHC12CC13 and CHZC1CCl3,l4 it is tempting to speculate that substitution of C1 on one carbon atom (atom a) substantially increases the strength of the C-H bonds on the adjacent carbon atom (atom b). Support for such speculation comes from the fact that CH3CF3,the fluorosubstituted analogue of CH3CC13,is known to have extremely strong C-H bonds (BDE = 106.7 kcal mol-' (ref 14)). On the other hand, thermodynamic data are a ~ a i l a b l e which '~ demonstrate that substitution of C1 for H in methane and chloromethanes decreases the dissociation energy of remaining C-H bonds. Kinetic data for H abstraction from CH,C14, by Cl(?PJ) and OH(X211)are consistent with the idea that the rate of abstraction depends on the BDE; i.e., the fastest reactions are with CHC1, which has the weakest C-H bond and the slowest reactions are with CHI which has the strongest C-H bonds. If the reactivity trends observed in the chloromethane series are applicable to the chloroethanes as well, then substitution of C1 or H on carbon a not only strengthens C-H bonds on carbon b but also weakens remaining C-H bonds on carbon a. Some end product analysis data are available which support the contention that C1 substitution at carbon a increases the probability that H abstraction will be from carbon a; for example, photolysis of C12in the presence of CHDzCD2C1 gives exclusively CHD2CDC12as a photoproduct.16 The lower activation energy and lower A factor that we obtain for reaction of C1('PJ) with CH3CHC12compared to reaction with CHzCICHzClis consistent with the hypothesis that H abstraction occurs primarily from the more substituted carbon atom; i.e., the lower activation energy suggests a weaker C-H bond at the reactive site while the lower A factor is probably due, at least in part, to the presence of a single weakest C-H bond in CH3CHC12 compared to four equivalent C-H bonds in CH2C1CH2C1. Rate constants for reactions of C2H,Clk, radicals with Clz have been obtained previously from studies of the (13)K. Jeong and F. Kaufman, Geophys. Res. Lett., 6 , 757 (1979). (14)D. F.McMillen and D. M. Golden, Annu. Rev. Phys. Chem., 33, 493 (1982). (15)S.'Furuyama, D. M. Golden, and S. W . Benson, J. Am. Chem. SOC.,91,7564 (1969). (16)D. C.McKean and B. W. Laurie, J.Mol. Struct., 27,317(1975).
Wine and Semmes
photochlorination of ethylene and chloroethylenes. Most of this work was carried out by Dainton and co-workers and by Goldfinger and co-workers during the 1960s. Pertinent references and recommended rate constants are given by Kerr." Addition of c1(2PJ) to unsymmetric chloroethylenes is believed to occur at the least substituted carbon atom.l* Hence, many of the radicals produced in the photochlorination of ethylenes are different from those which are probably produced by H abstraction from ethanes. Nevertheless, a comparison of results is instructive. Based on the photochlorination results, A factors for all C1, reactions are within a factor of 2 of 6 X C2H,Cl, cm3molecule-' s-l while activation energies increase from 0 to -5 kcal mot1 with increasing C1 substitution. Repulsion of the incoming Clz by chlorine atoms on the carbon atom bearing the unpaired electron is thought to be the primary factor affecting reactivity while delocalization of the unpaired electron due to chlorine atoms on the carbon adjacent to that bearing the unpaired electron is thought to play a significant but less important role.ls The magnitude of the rate constants that we obtain at 258 K is consistent with an A factor of 6 X cm3molecule-' s-' and an activation energy of -1 kcal mol-'. The reactivity trend predicted by the photochlorination studies is not evident in our results, but the large uncertainties associated with both sets of measurements, indirect identification of the identity of the CZH,C1,, reactants, and the fact that the photochlorination studies were done at T > 298 K while our results were obtained only at T = 258 f 1K make quantitative comparisons difficult. Real time measurements featuring direct detection of C2H,C1,, are needed to definitively elucidate the reactivity trends in C2H,Cl,-, + Clz reactions.
+
Acknowledgment. We thank Dr. A. R. Ravishankara for many helpful discussions during the course of this work. We also thank D. Gerard and S. Collier of Dow Chemical for supplying us with analyzed samples of unstabilized CH3CHClZand CHZC1CHClz.This work was supported in part by the National Science Foundation through grant No. ATM-80-19040 and in part by the National Aeronautics and Space Administration through subcontract No. 954814 from the Jet Propulsion Laboratory. Registry NO. C1,22537-15-1; CH&H&1,75-0(t3; CH3CHC12, 75-34-3; CH2ClCH2C1, 107-06-2; CH2ClCHCl2, 79-00-5. (17)J. A. Kerr, Ed., 'Handbook of Bimolecular and Termolecular Gas Reactions", Vol. I, CRC Press, Boca Raton, FL, 1981,p 319. (18)P.B. Ayscough, F. S. Dainton, and B. E. Fleischfresser, Tram. Faraday SOC.,62,1838 (1966).
