Experimental and Theoretical Investigation of the Kinetics of the

ARTICLE pubs.acs.org/JPCA

Experimental and Theoretical Investigation of the Kinetics of the Reaction of Atomic Chlorine with 1,4-Dioxane Binod R. Giri† and John M. Roscoe* Department of Chemistry, Acadia University, Wolfville, NS B4P 2R6, Canada

Nuria Gonzalez-García Theory Department, Fritz-Haber-Institut der Max-Planck-Gesellschaft, Faradayweg 4-6,14195 Berlin, Germany

Matthias Olzmann* Institut f€ur Physikalische Chemie, Karlsruher Institut f€ur Technologie (KIT), Kaiserstr. 12, 76131 Karlsruhe, Germany

John M. H. Lo and Robert A. Marriott Alberta Sulphur Research Ltd., #6, 3535 Research Road NW, Calgary, Alberta T2L 2K8, Canada

bS Supporting Information ABSTRACT: The rate coefficients for the reaction of 1,4-dioxane with atomic chlorine were measured from T = 292360 K using the relative rate method. The reference reactant was isobutane and the experiments were made in argon with atomic chlorine produced by photolysis of small concentrations of Cl2. The rate coefficients were put on an absolute basis by using the published temperature dependence of the absolute rate coefficients for the reference reaction. The rate coefficients for the reaction of Cl with 1,4-dioxane were found to be independent of total pressure from p = 290 to 782 Torr. The experimentally measured rate coefficients showed a weak temperature dependence, given by kexp(T) = þ3.1 ) 1010 exp((470 ( 110)/(T/K)) cm3 molecule1 s1. (8.4 2.3 The experimental results are rationalized in terms of statistical rate theory on the basis of molecular data obtained from quantum-chemical calculations. Molecular geometries and frequencies were obtained from MP2/aug-cc-pVDZ calculations, while single-point energies of the stationary points were computed at CCSD(T) level of theory. The calculations indicate that the reaction proceeds by an overall exothermic additionelimination mechanism via two intermediates, where the rate-determining step is the initial barrier-less association reaction between the chlorine atom and the chair conformer of 1,4-dioxane. This is in contrast to the Br plus 1,4-dioxane reaction studied earlier, where the rate-determining step is a chair-to-boat conformational change of the brominedioxane adduct, which is necessary for this reaction to proceed. The remarkable difference in the kinetic behavior of the reactions of 1,4-dioxane with these two halogen atoms can be consistently explained by this change in the reaction mechanism.

’ INTRODUCTION 1,4-Dioxane is a cyclic ether that finds wide use as a solvent in both industrial and laboratory work. It is toxic and is slow to biodegrade. Volatilization appears to be an important fate process.1 The release of ethers to the atmosphere can result in both their conversion to other volatile organic compounds (VOCs) by oxidation and the removal of reactive species such as OH, Cl, and Br. The atmospheric oxidation of cyclic ethers is expected to be initiated primarily through hydrogen abstraction by OH2 and unexpected reactivity trends in these reactions have been observed and reviewed.3,4 The reactions of halogen atoms with cyclic ethers are also expected to proceed by hydrogen r 2011 American Chemical Society

abstraction, producing the same organic free radicals as those resulting from attack by OH. This can initiate atmospheric oxidation of these cyclic ethers through the reaction of these organic free radicals with atmospheric O2 to produce a range of oxygenated organic compounds whose reactivity in the atmosphere can be quite different from that of the parent ether. The halogen-initiated atmospheric oxidation of VOCs can become significant in regions such as the marine boundary layer and the Received: February 23, 2011 Revised: April 19, 2011 Published: April 29, 2011 5105

dx.doi.org/10.1021/jp201803g | J. Phys. Chem. A 2011, 115, 5105–5111

The Journal of Physical Chemistry A Arctic where concentrations of Cl and Br are sometimes observed to be particularly large relative to OH.2,5,6 In recent publications,7,8 we presented experimental results for the reactions of tetrahydrofuran, tetrahydropyran, and 1,4dioxane with atomic bromine in which the reactivity of 1,4dioxane was quite unusual compared to that of tetrahydrofuran and tetrahydropyran. This unusual behavior was explained in terms of a complex reaction path in which a chair-to-boat conformation change of a dioxanebromine adduct appeared to be the bottleneck for the whole process. More recently, we have reported the results for the reactions of Cl with dimethyl ether, tetrahydrofuran, and tetrahydropyran.9 In the current paper, we report the results for the reaction of 1,4-dioxane with Cl, which were found to be qualitatively different from those for the reactions of the other ethers. We present here both the experimental results and a computational examination of the reaction of Cl with 1,4-dioxane in an effort to account for this unexpected behavior.

’ EXPERIMENTAL SECTION A detailed description of the experimental methods used in this work has been presented previously711 and is summarized briefly here. The reactions took place in a 70 L Pyrex reaction vessel enclosed in an insulated housing and surrounded by four fluorescent lamps equally spaced around the circumference of the reactor. The lamps provided a very broad spectrum centered at 460 nm with full width at half-maximum ≈ 170 nm. Four strong mercury lines were superimposed on the continuous emission. The temperature was controlled by circulating heated or cooled air through the space between the housing and the reaction vessel and was measured with three ironconstantan thermocouples placed inside the reaction vessel along its long axis. Pressures were measured by 10 Torr Baratron and 1000 Torr piezoelectric pressure gauges. Reaction mixtures were prepared from measured partial pressures of dilute mixtures of known composition of Cl2, isobutane and 1,4-dioxane in argon, stored in separate Pyrex bulbs. The final reaction mixtures were brought to the desired total pressure by adding measured amounts of argon. The gas handling system was constructed of Pyrex, was mercury free, and used exclusively Teflon stopcocks. Typical reaction mixtures contained Cl2 concentrations of the order of 4 1016 molecules cm3 and concentrations of organic reactants that were approximately an order of magnitude smaller than this. The output of the lamps was sufficient for photolysis of Cl2 providing concentrations of atomic chlorine of the order of 107 atoms cm3. This estimate was obtained from the slopes of pseudofirst order plots of loss of isobutane and the known value of the absolute rate coefficient for its reaction with Cl as in our previous work.7,8,11 The concentrations of the organic reactants remaining after a measured period of irradiation were determined by gas chromatography using a flame ionization detector. The analytes were separated on either an 8 ft 1/8 in. column packed with 80/100 mesh Porapak Q or a 8 ft 1/8 in. column packed with 1% SP1000 on 60/80 mesh Carbopak B. These columns gave identical analytical results but the Porapak Q column was preferred because it gave less bleed and consequently provided better signal-to-noise and improved sensitivity. The carrier gas was helium at a flow rate of typically 50 mL/min. The columns were operated isothermally at 120 C for the SP 1000 column or 150 for the Porapak Q column. Gas samples of known pressure were injected directly from a sample loop connected to the inlet

ARTICLE

of the gas chromatograph. The samples removed from the reaction vessel were sufficiently small that several successive irradiations could be made on a reaction mixture without significantly altering the total pressure. The kinetic analysis assumes that loss of the organic reactants occurs only by the following reactions and that there are no processes that regenerate the organic reactants Cl þ C4 H8 O2 f HCl þ C4 H7 O2 •

ð1Þ

Cl þ i-C4 H10 f HCl þ i-C4 H9 •

ð2Þ

The ratio of the rate coefficients for reactions 1 and 2 is given by the relation ( ) ( ) ½C4 H8 O2 t0 ½i-C4 H10 t0 k1 ¼ ln ð3Þ ln ½C4 H8 O2 t k2 ½i-C4 H10 t A plot of ln{[C4H8O2]t0/[C4H8O2]t} against ln{[i-C4H10]t0/ [i-C4H10]t} is predicted to be linear with a slope of k1/k2 and a zero intercept. The areas of the peaks in the chromatograms are proportional to the analyte concentration so the concentrations in eq 3 can be replaced by the peak areas in the chromatograms after normalization to a constant pressure in the sample loop. The absolute rate coefficients were calculated by multiplication of the values of k1/k2 by the known value of the absolute rate coefficient for the reference reaction, k2, at the temperature of the experiment. As shown in Figure 1, plots of the data in the form of eq 3 were linear, suggesting that the assumptions of the kinetic analysis are valid and that interference by secondary reactions has been avoided. The intercepts of these plots were zero within the standard deviation of the regression line. As in our previous work,711 the organic reactants were tested for interference due to photolysis or due to dark reaction with Cl2. In all cases, no loss of the organic reactant could be detected in the dark in the presence of Cl2 or with the lamps turned on in the absence of Cl2 over a period of time that was longer than the time required for a series of kinetic experiments. We conclude from this that our kinetic measurements were unaffected by dark reaction with Cl2 or by photolysis of the parent organic reactant. As discussed in our earlier work,9 the measurements were carried out in the absence of O2 to avoid the possibility of interference by attack of OH on the organic reactant and to reduce the potential challenge of obtaining adequate chromatographic resolution of a more complex mixture of reaction products. 1,4-Dioxane (Aldrich, 99.8%, anhydrous, without stabilizer) was thoroughly degassed followed by distillation of the required amount into its storage bulb. Isobutane (Aldrich, 99.995%), chlorine (Matheson, 99.5%), Ar (Praxair, UHP, 99.9995%), and the gases used for gas chromatography, He (Praxair, UHP, 99.9995%), H2 (Praxair, 99.95%), and medical air (Praxair) were used without further purification.

’ CALCULATIONS The experimental results are rationalized in terms of statistical rate theory with the molecular and transition-state parameters obtained from quantum chemical calculations. The methods used are briefly outlined below. Structures and Energies. All quantum chemical calculations were performed in complete analogy to our earlier work on the Br plus 1,4-dioxane reaction.8 We calculated geometries and frequencies of the stationary points, using second-order 5106

dx.doi.org/10.1021/jp201803g |J. Phys. Chem. A 2011, 115, 5105–5111

The Journal of Physical Chemistry A

ARTICLE

Figure 1. Plots of concentration ratios according to eq 3. All experiments were performed in argon buffer gas. Conditions: (b) [Cl2] = 4.17 1016 molecules cm3, [i-C4H10] = 2.44 1015 molecules cm3, [C4H8O2] = 5.72 1015 molecules cm3, total pressure = 782 Torr, temperature = 346.5 K; (O) [Cl2] = 6.56 1016 molecules cm3, [i-C4H10] = 2.71 1015 molecules cm3, [C4H8O2] = 6.64 1015 molecules cm3, total pressure = 792 Torr, temperature = 301.3 K; (9) [Cl2] = 3.76 1016 molecules cm3, [i-C4H10] = 2.95 1015 molecules cm3, [C4H8O2] = 4.12 1015 molecules cm3, total pressure = 329 Torr, temperature = 335.8 K; (0) [Cl2] = 5.26 1016 molecules cm3, [i-C4H10] = 2.45 1015 molecules cm3, [C4H8O2] = 6.36 1015 molecules cm3, total pressure = 777 Torr, temperature = 321.1 K.

MøllerPlesset perturbation theory (MP2)12,13 with Dunning’s augmented correlation-consistent polarized valence double-zeta basis set (aug-cc-pVDZ).14,15 Energies were obtained from single-point coupled cluster calculations with single and double excitations including noniterative triples corrections (CCSD(T)).1618 In these calculations, Dunning’s correlation consistent polarized valence doublezeta (cc-pVDZ) and triple-zeta (cc-pVTZ) basis sets were used,14,15 and the complete basis set (CBS) limits were obtained by employing the extrapolation scheme developed by Helgaker et al.19,20 We calculated the zero-point corrections from the frequencies obtained at the MP2/aug-cc-pVDZ level of theory, using a scaling factor of 0.9615.21 All computations were performed with the Gaussian03 program package22 employing the spin-unrestricted formalism.13 Rate Coefficients. The rate coefficient for the barrier-less association reaction c-C4H8O2 þ Cl f c-C4H8O2Cl, which is the rate determining step of reaction 1 (see below), was calculated from the simplified Statistical Adiabatic Channel Model (SACM) in its canonical version.2325 Specific rate coefficients for the decomposition reactions of c-C4H8O2Cl were obtained from RRKM theory2528 or from the microcanonical version25,29 of SACM depending on whether or not the reaction considered has a localized transition state. It turns out that these calculations, while important to characterize the general behavior of the reaction system, do not quantitatively influence our results concerning the rate coefficients. Therefore, we refrain from giving extensive details of our implementation here but refer to a recent publication by two of the authors.30 All molecular and transition-state data used are compiled in Table IS of the Supporting Information.

’ RESULTS AND DISCUSSION The experiments for the reaction of Cl with 1,4-dioxane were carried out over a temperature range of 292360 K and at total

pressures from 290 to 782 Torr. The rate coefficients were measured relative to the reaction of Cl with isobutane. The temperature dependence of the rate coefficient for the reference reaction was taken from recent work from our laboratory9 and is 10 exp((99 ( 44)/ given by kref(T) = k2(T) = (1.02þ0.16 0.14) 10 (T/K)) cm3 molecule1 s1. This function was used to calculate the absolute values of the rate coefficients k1, presented in Table 1 and Figure 2, from the measured relative values. The reaction revealed very little T-dependence of the rate coefficients and exhibited no discernible pressure dependence over the range of our experimental conditions. Least-squares analysis of the experimental data in Figure 2 leads to a small activation energy of 3.9 ( 0.9 kJ/mol. The small temperature dependence of the rate coefficients for this reaction is given by eq 4, in which the uncertainties represent one standard deviation 10 expðð470 ( 110Þ k1 ðTÞ ¼ ð8:4þ3:1 2:3 Þ 10 exp

=ðT=KÞÞ cm3 molecule1 s1

ð4Þ

The rate coefficients have a standard deviation of 10% from the line representing this temperature dependence and all the experimental values of the rate coefficients are within two standard deviations of the regression line. Equation 4 leads to a rate coefficient of (1.74 ( 0.26) 1010 cm3 molecule1 s1 at 298 K, which agrees well with k1(298 K) = (2.0 ( 0.3) 1010 cm3 molecule1 s1 reported by Platz et al.31 using dimethyl ether and ethylene as the reference compounds. Our value is also in excellent agreement with that from a recent report of Li and Pirasteh32 who measured the rate coefficients for reaction 1 in their flow reactor at T = 240 to 340 K and P = 1.0 to 1.1 Torr using the relative rate method. They found that the rate coefficients for reaction 1 had such a small temperature 5107



ARTICLE

Table 1. Rate Coefficients for the Reaction of Atomic Chlorine with 1,4-Dioxanea absolute temperature

a

[i-C4H10]0 1015

[C4H8O2]0 1015

[Cl2]0 1015

(molec. cm3)

(molec. cm3)

(molec. cm3)

k1 1010 (k1)/(k2)

(cm3 molec.1 s1)

291.5

3.10

6.04

31.1

1.17 ( 0.13

1.67 ( 0.19

301.3

2.71

6.64

65.6

1.06 ( 0.05

1.50 ( 0.07

302.6

3.26

6.13

47.8

1.47 ( 0.06

2.08 ( 0.08

303.9

2.93

6.59

32.0

1.35 ( 0.12

1.91 ( 0.17

311.5

2.74

5.27

30.9

1.26 ( 0.15

1.77 ( 0.21

311.9

2.85

5.23

27.8

1.27 ( 0.18

1.77 ( 0.25

316.7 320.0

2.61 2.81

5.91 6.62

28.3 43.4

1.24 ( 0.19 1.51 ( 0.14

1.73 ( 0.26 2.10 ( 0.19

320.6

3.06

6.80

42.9

1.40 ( 0.10

1.94 ( 0.14

321.1

2.45

6.36

52.6

1.45 ( 0.12

2.01 ( 0.17

334.2

2.57

4.44

26.5

1.26 ( 0.24

1.73 ( 0.33

335.8

2.95

4.12

37.6

1.37 ( 0.08

1.88 ( 0.10

345.8

2.31

6.36

43.6

1.82 ( 0.09

2.47 ( 0.12

346.5

2.44

5.72

41.7

1.88 ( 0.13

2.55 ( 0.18

355.3 355.6

2.53 2.18

5.43 5.75

42.3 47.4

1.58 ( 0.03 1.69 ( 0.01

2.13 ( 0.04 2.28 ( 0.01

355.8

2.26

5.19

44.0

1.47 ( 0.04

1.98 ( 0.05

356.6

2.76

6.67

45.7

1.62 ( 0.09

2.18 ( 0.12

358.9

3.20

3.29

37.1

1.77 ( 0.14

2.38 ( 0.19

359.8

2.64

4.10

43.5

1.78 ( 0.14

2.39 ( 0.19

360.4

2.82

4.23

46.8

1.59 ( 0.08

2.14 ( 0.10

The uncertainties are one standard deviation and do not include the uncertainty in the value of the reference rate coefficient.

Figure 2. Temperature dependence of the rate coefficient k1: (b) experimental points; (O) experimental results of reference 32; (0) reference 31; ---calculated values, eq 5.

dependence that this reaction is essentially independent of temperature and reported an average value of (1.93 ( 0.2) 1010 cm3 molecule1 s1 for k1. While the results of the two previous measurements of the rate coefficient for the reaction of Cl with 1,4-dioxane all lie within the error bars of our measurements, it is worth also examining the accuracy inherent in our calculation of the absolute rate coefficients for this reaction. The principal error contributor to the

relative rate coefficients is the uncertainty in determining the concentration ratios of 1,4-dioxane and the reference compound, isobutane, chromatographically required in the relative rate plots such as those in Figure 1. Obviously, the error is large for the smallest extent of reaction. However, we estimate the maximum uncertainty in the slopes of the relative rate plots to be approximately (20% and for most cases it is about (10%. It is expected that this will represent a random source of error and is adequately 5108



Figure 3. Energy profile obtained at the CCSD(T)/CBS level of theory for the C4H8O2 þ Cl reaction system (including scaled zero-point energies, see text); TS denotes transition states; the prefixes c- and b- denote the chair and boat conformer, respectively.

represented by the results of the least-squares analysis of the data. The other significant source of uncertainty in the absolute rate coefficients is the value of the absolute rate coefficient used for the reference reaction of Cl with isobutane. We used the rate coefficient for this reaction that we reported recently.9 Those measurements were made relative to the reaction of Cl with propane for which a temperature-independent value of (1.4 ( 0.06) 1010 cm3 molecule1 s1 has been recommended.33 This, combined with the mean uncertainty of (4% reported for the measured relative rate coefficients, leads to an accuracy of about (8% in the rate coefficient for the reaction of Cl with isobutane. When this is combined with the standard deviation of (10% in the rate coefficients for the reaction of Cl with 1,4-dioxane measured relative to the reaction of Cl with isobutane, the absolute rate coefficients calculated from these relative values will have an accuracy of approximately (18%. If we compare our results with those in the literature at a common temperature of 298 K, our calculated absolute rate coefficients agree with those in the literature to better than 18% as illustrated in Figure 2. We next rationalize our results in terms of statistical rate theory to account for the small magnitude of the T-dependence for reaction 1 and to explain the remarkable differences in terms of reactivity between the chlorine and bromine atom reactions with 1,4-dioxane. Figure 3 depicts the energy diagram of the overall reaction 1 obtained at the CCSD(T)/CBS level of theory. Selected geometrical parameters of the stationary points are collected in Table IIS of the Supporting Information, and the structures are displayed in Figures IS and IIS of the Supporting Information. It is important to note that 1,4-dioxane can occur in two stable isomers, namely a chair and a boat conformer. As outlined in ref 8, the equilibrium fraction of the boat conformer under our experimental conditions is below 104. Consequently, the reactivity of 1,4-dioxane is completely governed by the chair conformer, and the reaction pathway via TS(c f b) (see Figure 3) needs not to be considered further. The overall exothermicity of reaction 1 at T = 0 K was calculated to be 23.85 kJ/mol. In complete analogy to the Br þ C4H8O2 reaction, the initial step is the formation of an association complex, c-C4H8O2Cl, which is primarily stabilized by the interaction between the chlorine atom and one of the oxygen atoms in the ether moiety. This association complex is lower in energy by 22.70 kJ/mol relative to the reactants and can lead to the products via two different reaction pathways.

ARTICLE

Like the Br þ C4H8O2 reaction, the hydrogen abstraction is an intramolecular process and can happen after a ring inversion from c-C4H8O2Cl to b-C4H8O2Cl. A transition-state (TS(A) in Figure 3) linking these two conformers was found to be 46.76 kJ/mol higher in energy than the c-C4H8O2Cl complex; the isomerization product, b-C4H8O2Cl, lies 2.65 kJ/ mol above the bimolecular reactants. The intramolecular Habstraction in the boat conformer has a threshold energy of 9.79 kJ/mol and proceeds via the transition state TS(B) (see Figure 3), which is product-like because both the forming (HCl) and breaking (CH) bonds are already very close to their values in b-C4H7O2HCl. The latter is a predissociative complex, which is stabilized by long-range interactions between the abstracted hydrogen and the oxygen in the 1,4-dioxane radical part. Eventually, it decomposes into the separate fragments of the dioxanyl radical (C4H7O2•) and HCl. Unlike the Br þ C4H8O2 reaction,8 a H-abstraction pathway for reaction 1 via TS(A) involving the ring inversion appears to be an energetically unfavorable route in our T-range because a low-lying intramolecular abstraction pathway that retains the chair conformation of the dioxane ring exists. This is a notable difference between the reactions of Cl and Br atoms with 1,4-dioxane. Starting from the c-C4H8O2Cl association complex, the intramolecular hydrogen abstraction proceeds via TS(1b) with a barrier of 8.50 kJ/mol to produce first a predissociative complex, c-C4H7O2HCl, before it can finally decompose into C4H7O2• and HCl. As seen in Figure 3, all stationary points along this reaction pathway lie well below the energy of the reactants. This gives rise to a chemical activation mechanism, that governs the kinetics of reaction 1. The parallel reaction pathway via ring inversion is unimportant in the temperature range of our study. In view of the above results, our further kinetic analysis is based on the following mechanism for reaction 1 c-C4 H8 O2 þ Cl f c-C4 H8 O2 Cl

ð1aÞ

c-C4 H8 O2 Cl f c-C4 H7 O2 HCl

ð1bÞ

c-C4 H7 O2 HCl f c-C4 H7 O2 • þ HCl

ð1cÞ

In the exothermic reaction 1a, the c-C4H8O2Cl intermediate is formed with an average internal energy of ÆEæ ∼ (22.7 þ 12.8) kJ/mol = 35.5 kJ/mol. Here the first term is the energy difference at 0 K (see Figure 3), and the second term represents the average thermal energy of the reactants at T = 300 K, which was estimated from the sum of states of c-C4H8O2Cl.26 For the specific rate coefficients of the back dissociation (1a) and forward reaction 1b, we obtained k1a(ÆEæ) = 2.8 1011 s1 and k1b(ÆEæ) = 3.1 1012 s1 from the SACM (R/β = 0.845, see below) and RRKM theory, respectively. From these values, it follows that back dissociation as well as collisional stabilization of the chemically activated c-C4H8O2Cl adduct at atmospheric pressure (collision frequency in the order of 1010 s1) cannot compete with the intramolecular hydrogen abstraction, reaction 1b. As the subsequent reaction 1c has an even higher excess energy and, moreover, a loose transition state, the decomposition of the cC4H7O2HCl adduct into the products, C4H7O2 þ HCl, is even faster and kinetically irrelevant under the conditions of this work. Accordingly, the rate determining step of the overall reaction 1 is the barrierless association step, reaction 1a, and its rate coefficient is identical to the high-pressure limiting value. Consequently, we calculated this high-pressure limit from the 5109



ARTICLE

canonical version of the simplified SACM.23 The anistropy parameter, R/β,24,34 was fitted so as to match the experimental rate coefficients near 320 K, which is the median temperature of our experiments. For R/β = 0.845, we obtained k1a(320 K) = 2.0 1010 cm3 molecule1 s1. Our calculation resulted in a small T-dependence for k1, which can be best represented by the following Arrhenius expression in the range from T = 290360 K 10 kcalc expð36:5=ðT=KÞÞ cm3 molecule1 s1 1 ðTÞ ¼ k1a ðTÞ ¼ 2:27 10

ð5Þ Figure 2 displays a comparison between the calculated and experimental rate coefficients for reaction 1. With values around 2 1010 cm3 molecule1 s1, the rate coefficient is close to the collision limit. Our optimum value of R/β = 0.845 seems somewhat higher than the empirically favored standard range of 0.3 e R/β e 0.6.24 However, we note here that the rate coefficients calculated with R/β = 0.5 are just a factor of 2 lower at all temperatures, which does not alter the general picture of this reaction. Our SACM calculations may appear to have somewhat underestimated our experimental results. However, considering the uncertainties associated with both of these methods, the weak temperature dependence of the rate coefficient is well reproduced by the theory which is demonstrated in Figure 2. As far as the quality of the predicted values of the rate coefficients is concerned, the intermolecular potential of the dissociating fragments comes into play. In our calculations we used a Morse function, which if combined with the exponential interpolation scheme of the simplified SACM is known to predict in most cases a weak positive temperature dependence.24 For only a very slightly more accurate description, a much more accurate and extended potential energy surface would be required because the finer details of the temperature dependence are governed by the complicated interplay between the radial and angular parts of the potential. Such calculations, however, are far beyond the scope of the present work.34

’ CONCLUSIONS The reaction of Cl with 1,4-dioxane was investigated in a smog chamber over a temperature range from 292 to 360 K and pressures from 290 to 782 Torr using the relative rate method. Rate coefficients for this reaction showed no discernible pressure dependence but exhibited a weak positive temperature dependence with values near the collision limit of k1 ∼ 2.0 1010 cm3 molecule1 s1 that can be described by an activation energy of approximately 4 kJ/mol. We estimate that our absolute rate coefficients have an accuracy of approximately (18% which also includes the uncertainty of the reference rate coefficient. Our value at T = 298 K agrees well with the earlier reports from two other groups31,32 within the quoted uncertainty. There is now accumulating evidence that the reactions of halogen atoms with ethers proceed via the initial formation of an addition complex,8,35 and similar examples exist for other organic compounds containing oxygen or sulfur atoms.3638 However, they exhibit a wide variation in their reactivity when the atomic reactant is Br rather than Cl under similar conditions. Reactions of Br with ethers are found to be several orders of magnitude slower than those of Cl with ethers. In the case of 1,4-dioxane, the rate coefficients for reaction with Cl are 5 orders of magnitude larger than those for the reaction with Br. Our theoretical analysis

attributes this large variation in kinetic behavior to the change in the reaction mechanism. The reactions of 1,4-dioxane with Cl and Br differ mainly in two ways. First, the reaction of Br occurs with an overall endothermicity of 23 kJ/mol and shows a stronger T-dependence,8 whereas the corresponding reaction of Cl is overall an exothermic process (ΔH0 = 23.85 kJ/mol) with a very weak T-dependence. Second, the critical intramolecular hydrogen abstraction in the reaction of Cl with 1,4-dioxane can occur within the chair conformation as indicated in Figure 3, whereas this step is energetically not accessible in the reaction of Br with 1,4-dioxane.8 Instead, a chair-to-boat isomerization has to precede the intramolecular abstraction step and this chair-toboat isomerization turns out to be the rate-determining step in the reaction of Br with 1,4-dioxane. This accounts for the temperature dependence of the reaction.8 In the reaction of Cl with 1,4-dioxane, all transition states are energetically located below the energy of the reactants. Therefore, the rate determining step in this case is the association reaction of the Cl atom and 1,4-dioxane to yield the chair conformer of the Cldioxane complex. This results in a much higher overall rate coefficient with a much weaker temperature dependence compared to that of the reaction of 1,4-dioxane with Br. The reason for these differences can be attributed to the much stronger HCl bond as compared to the HBr bond. This turns the overall reaction from being endothermic in the case of Br8 to exothermic in the case of Cl. It is clear that the strength of the halogenhydrogen interaction determines the energy profile along the reaction pathway, resulting in an energy decrease of the stationary points relative to the reactants if the atomic reactant is Cl rather than Br. In the specific case of 1,4-dioxane, a new lowlying hydrogen-abstraction pathway within the chair conformation appears when the atomic reactant is Cl. This leads to a change in the rate-determining step and explains the different kinetic behavior observed in our experiments. It is further informative to summarize the overall picture for the reaction pathways of the reactions of Cl with monoethers and alcohols, in which formation of adducts plays a crucial role in determining the kinetics of these reactions. The first elementary step for the reaction of the Cl atom with methanol involves the formation of a loose molecular complex which, being thermally unstable, eventually dissociates to yield CH2OH• and HCl.36 In the reaction of Cl with tetrahydropyran,39 the lowest energy channel involves attack by the Cl atom at the axial hydrogen R- to the ether oxygen. However, we note that these authors did not examine the role of adduct formation resulting from the coordination of the Cl atom to the ether oxygen. Formation of such an adduct as the first elementary step, followed by intramolecular R-hydrogen abstraction from the axial position as indicated by these authors, would be consistent with other work on reactions of Cl and Br in the literature8,35,36 as well as with our analysis of the reaction of Cl with 1,4-dioxane. In the current work, all the hydrogens in 1,4-dioxane are R- to an ether oxygen, so the only energetic distinction is between the abstraction pathways for the axial and equatorial hydrogens. Our results are in agreement with those of reference 39 in concluding that the lowest energy path involves abstraction of an R-H atom from the axial position.

’ ASSOCIATED CONTENT

bS

Supporting Information. Figures and tables with geometric parameters and harmonic wave numbers of all relevant

5110


The Journal of Physical Chemistry A species. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Telephone (902) 585-1353. Fax (902)585-1114. E-mail: (J.M.R.) [email protected]; (M.O.) [email protected]. Present Addresses †

Alberta Sulfur Research Ltd., #6, 3535 Research Road NW, Calgary, Alberta T2L 2K8, Canada.

’ ACKNOWLEDGMENT Funding for the experimental work was provided by the Natural Sciences and Engineering Research Council of Canada through a Discovery Grant to J.M.R. N.G.-G. and M.O. are grateful to the Alexander von Humboldt Foundation for a Research Fellowship to N.G.-G. ’ REFERENCES (1) National Library of Medicine HSDB Database, http//toxnet. nlm.nih.gov/cgi-bin/sis/htmgen?HSDB (accessed April 8, 2011). (2) Finlayson-Pitts, B. J.; Pitts, J. N. Chemistry of the Upper and Lower Atmosphere Theory, Experiments, and Applications; Academic Press: New York, 2000. (3) Moriarti, J.; Sidebottom, H.; Wenger, J.; Mellouki, A.; Le Bras, G. J. Phys. Chem. A 2003, 107, 1499. (4) Mellouki, A.; LeBras, G.; Sidebottom, H. Chem. Rev. 2003, 103, 5077. (5) Bierbach, A.; Barnes, I.; Becker, K. H. Atmos. Environ. 1999, 33, 2981. (6) Ramacher, B.; Orlando, J. J.; Tyndall, G. S. Int. J. Chem. Kinet. 2001, 33, 198. (7) Giri, B. R.; Roscoe, J. M. J. Phys. Chem. A 2009, 113, 8001. (8) Giri, B. R.; Roscoe, J. M.; Gonzalez-García, N.; Olzmann, M. J. Phys. Chem. A 2010, 114, 291. (9) Giri, B. R.; Roscoe, J. M. J. Phys. Chem. A. 2010, 114, 8369. (10) Anthony, L. M.; Roscoe, J. M. J. Phys. Chem. A 2004, 108, 7535. (11) Wheeler, M.; Mills, R.; Roscoe, J. M. J. Phys. Chem. A 2008, 112, 858. (12) Møller, C.; Plesset, M. S. Phys. Rev. 1934, 46, 618. (13) Hehre, W. J.; Radom, L.; Schleyer, P. v. R.; Pople, J. A. Ab initio Molecular Orbital Theory; Wiley: New York, 1986. (14) Dunning, T. H., Jr. J. Chem. Phys. 1989, 90, 1007. (15) Kendall, R. A.; Dunning, T. H., Jr.; Harrison, R. J. J. Chem. Phys. 1992, 96, 6796. ízek, J. J. Chem. Phys. 1966, 45, 4256. (16) C (17) Purvis, J. D., III; Bartlett, R. J. J. Chem. Phys. 1982, 76, 1910. (18) Pople, J. A.; Head-Gordon, M.; Raghavachari, K. J. Chem. Phys. 1987, 87, 5968. (19) Helgaker, T.; Klopper, W.; Koch, H.; Noga, J. J. Chem. Phys. 1997, 106, 9636. (20) Halkier, A.; Helgaker, T.; Jorgensen, P.; Klopper, W.; Koch, H.; Olsen, J.; Wilson, A. K. Chem. Phys. Lett. 1998, 286, 243. (21) Merrick, J. P.; Moran, D.; Radom, L. J. Phys. Chem. A 2007, 111, 11683. (22) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, Jr., J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;

ARTICLE

Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; and Pople, J. A. Gaussian 03, revision C.02; Gaussian, Inc.: Wallingford, CT, 2004. (23) Troe, J. J. Chem. Phys. 1981, 75, 226. (24) Cobos, C. J.; Troe, J. J. Chem. Phys. 1985, 83, 1010. (25) Gilbert, R. G.; Smith, S. C. Theory of Unimolecular and Recombination Reactions; Blackwell: Oxford, 1990. (26) Forst, W. Theory of Unimolecular Reactions; Academic Press: New York, 1973. (27) Marcus, R. A.; Rice, O. K. J. Phys. Colloid Chem. 1951, 55, 894. (28) Marcus, R. A. J. Chem. Phys. 1952, 20, 359. (29) Troe, J. J. Chem. Phys. 1983, 79, 6017. (30) Gonzalez-García, N.; Olzmann, M. Phys. Chem. Chem. Phys. 2010, 12, 12290. (31) Platz, J; Sehested, J.; Møgelberg, T.; Nielsen, O. J.; Wallington, T. J. J. Chem. Soc., Faraday Trans. 1997, 93, 2855. (32) Li, Z.; Pirasteh, A. Int. J. Chem. Kinet. 2006, 38, 386. (33) Atkinson, R.; Baulch, D.; Cox, R.; Crowley, J.; Hampson, R.; Hynes, R.; Jenkin, M.; Rossi, M.; Troe, J. Atmos. Chem. Phys. 2006, 6, 3625. (34) Troe, J. Z. Phys. Chem. Neue Folge 1989, 161, 209. (35) Lo, J. H.; Marriott, R. A.; Giri, B. R.; Roscoe, J. M.; Klobukowski, M. Can. J. Chem. 2010, 88, 1136. (36) Jodkowski, J. T.; Rayez, M.-T.; Rayez, J.-C.; Berces, T.; D obe, S. J. Phys. Chem. A 1998, 102, 9230. (37) Garzon, A.; Notario, A.; Albaladejo, J.; Pe~ na-Ruiz, T.; FernandezGomez, M. Chem. Phys. Lett. 2007, 438, 184. (38) Kleissas, K. M.; Nicovich, J. M.; Wine, P. H. J. Photochem. Photobiol., A 2007, 187, 1. (39) Ballesteros, B.; Ceavero-Vega, A. A.; Garzon, A.; Jimenez, E.; Albaladejo, J. J. Photochem. Photobiol., A 2009, 208, 186.

5111


Experimental and Theoretical Investigation of the Kinetics of the

Recommend Documents