Pathway and Kinetic Analysis on the Iso-Propyl Radical + O 2

Size: px

Start display at page:

Download "Pathway and Kinetic Analysis on the Iso-Propyl Radical + O 2"

Percival Ramsey
6 years ago
Views:

1 URL-J PREPRINT Pathway and Kinetic Analysis on the Iso-Propyl Radical + O 2 Reaction System J.W. Bozzelli W.J. Pitz This paper was prepared for submittal to the 1997 Spring Meeting of the Western States Institute Livermore, A April 14-15, 1997 April 1997 Lawrence Livermore National Laboratory This is a preprint of a paper intended for publication in a journal or proceedings. Since changes may be made before publication, this preprint is made available with the understanding that it will not be cited or reproduced without the permission of the author.

2 DISLAIMER This document was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor the University of alifornia nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or the University of alifornia. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or the University of alifornia, and shall not be used for advertising or product endorsement purposes.

3 Pathway and Kinetic Analysis on the Iso-Propyl Radical + O System Joseph W. Bozzelli hemistry and hemical Engineering Department New Jersey Institute of Technology Newark, NJ William J. Pitz Lawrence Livermore National Laboratories Livermore, A April 7, 1997 This paper was prepared for submittal to the 1997 Spring Meeting of the Western States Institute Sandia National Laboratories Livermore, A April 1997

4 Pathway and Kinetic Analysis on the Iso-Propyl Radical Reaction + System O 2 ABSTRAT Joseph W. Bozzelli hemistry and hemical Engineering Department New Jersey Institute of Technology, Newark, NJ William J. Pitz Lawrence Livermore National Laboratories Livermore, A We analyze the iso-propyl + O2 reaction system using thermochemical Transition State Theory (TST), molecular thermodynamic properties, analysis (quantum RRK) for k(e) and modified strong collision analy yclic transition states for both hydrogen transfer and concerted e propylene from isopropylperoxy are calculated using semi-empirical theory in addition to transition states for HO2 elimination and ep from hydroperoxy-isopropyl. omputed rate constants are compared t constant measurements of Gulati and Walker (1988) for isopropyl + O HO2. INTRODUTION The reaction of alkyl radicals with molecular oxygen plays a key the oxidation of alkanes. In the negative temperature coefficient alkane oxidation, this reaction leads to the production of alkenes, and chain branching. At higher temperatures, it leads primarily to production of a conjugate alkene and the HO2 radical. The reaction radicals with O2 is important in atmospheric chemistry and in many combustion problems including automotive and gas turbine engines. In the case of 2H5 + O2, it was generally accepted that the rea proceeds to 2H4 through the path: 2H5 + O2 <=TST1=> OO.# <= TST2=>.OOH# <=TST3=> 2H4 + HO2 + M V OO.

5 3 Recent work by Ignatyev et al. [1] using density functional theo suggested a fourth transition state (TST4), where direct production + HO2 can result via concerted elimination of HO2 from OO#. This postulated reaction path does not proceed through the hydroperoxy-e radical (.OOH) intermediate. The existence of the cyclic transiti first hypothesized by McAdam and Walker [2]. They reported that Mechanism (I) could not explain the observed pressure dependence of 2H5 + O2 = 2H4 + HO2 at 298 K and the observed decrease in rate constant with increasing temperature from K. They suggested the existence of a long cyclic transition state and a concerted elimination of ethylene wou these experimental observations. In subsequent work, Gutlati and W postulated the existence of an analogous, long-lived, cyclic transi the i-3h7 + O2 system as well. However, Wagner [4], Bozzelli and D Kaiser [6], Koert et al. [7] have shown that Mechanism I can explai observed temperature and pressure dependencies. The iso-propyl + O2 system is addressed in the present work. In study, we analyze the propyl + O2 reaction system using thermochemi kinetics, Transition State Theory (TST), molecular thermodynamic properties, quantum Kassel analysis (quantum RRK) for k(e) and modi strong collision analysis for fall off. yclic transition states f transfer and the HO2 concerted elimination from isopropylperoxy (2 are calculated using semi-empirical (MOPA PM3) calculations [8] in to transition states for HO2 elimination and epoxide formation fro hydroperoxy-isopropyl (2.OOH). omputed rate constants for isopro O2 are compared to the values of Gulati and Walker who measured the constants at 50 torr and over a temperature range of 653 to 773 K. APPROAH Thermodynamic properties listed in Table 1 for the relevant rad stable parents were obtained by group additivity using THERM [9] wit

6 4 updated H//O groups and bond dissociation groups [10]. Ab initio calculations (MP4/631G**) and isodesmic reaction analysis were perf for selected peroxides [11]. Additional calculations show hydropero radicals 2.OOH have -H bond energy similar to that of a normal p alkyl radical [12], in contrast to values reported by Ignatyev et al 8 kcal/mole higher. The thermochemical data allow accurate calcula reverse reaction rate constants by microscopic reversibility. Entropies and heat capacities for relevant transition states ar using semi-empirical molecular orbital calculations (MOPA-PM3). L vibration from the reaction coordinate, hindered internal rotors, optical isomer and spin degeneracy are incorporated into the calcul entropies and heat capacities. The transition state (TS) geometrie PM3 method are identified by the existence of only one imaginary fr in the normal mode coordinate analysis. The overall transition state for HO2 molecular elimination fro propyl peroxy radical is like the products ethylene (SP2 like carbon The -2H4- moiety is nearly planar, near SP2 like structure, with t distance closer to that of a carbon = carbon double bond. The HO2 group is nearly perpendicular to the plane of the -2H4- and the - O--H distances, to the hydrogen leaving the methyl carbon, are near 1.29 and 1.31 A. The O=O bond is nearly that of a O=O double bond, arbon - carbons and oxygen - oxygen bonds both shorten; O-, O-H an bonds become longer. The transition state for H shift from isopropyl peroxy radical i non-planar, the carbon - hydrogen and O--O bonds become longer, the and --O bonds remain near constant in length, and 2H4 moiety is S The HO2 elimination TST's extended HO2 moiety, further away fro carbons, leads to lower vibrations and looser transition state re shift case. Energies are calculated in the PM3 case to be similar kcal/mole), but these data are not used.

7 5 High pressure limit pre-exponential factors (Arrhenius A factor for bimolecular addition and combination reactions are obtained fro literature and from trends in homologous series of reactions. A-f for unimolecular reactions are calculated using transition state th along with PM3-determined entropies of reactants and transition sta A = (eh p T/k b )exp( S ) where h p is Plank s constant, k b is the Boltzmann constant, and S to S 298,TS - S 298,reactant. A-factors are listed in Table 2 as a temperature. Activation energies are derived from the endothermic reaction ( U rxn ) and from analogy to similar reactions with known energetics. Activation energies were estimated in a consistent an manner with reference to the literature, experiments or theoretica calculation in each case. The activation energy for the forward ra internal H-atom shifts include analysis of Evans-Polanyi relationsh abstractions plus evaluation of ring-strain energy. The resultant pressure rate constants for each reaction pathway are given in Tabl Rate constants and product channels for the multi-channel react iso-propyl radicals with and Othe 2 reaction of important adducts were estimated with multifrequency quantum Kassel Theory (QRRK) [14,15] k(e) coupled with modified strong collision analysis of Gilbert et a fall-off and steady-state assumption for the adduct population at e A number of modifications have been made since the initial descrip QRRK and falloff were published. These are described in Reference RESULTS AND DISUSSION A potential energy diagram for the iso-propyl + O2 system is gi Figure 1. The four transition states calculated from the semi-emp molecular orbital calculations are sketched at the top of the figur

8 6 the barrier for the H-atom shift channel of 2OO. is estimated to to the barrier for HO2 molecular elimination. This is in contrast of Ignatyev et al. [1] who calculated that the barrier for H-atom sh kcal/mole higher than the barrier for HO2 elimination. Ignatyev et calculate that the H shift TST is ca 8 kcal/mole higher than the HO elimination pathway for the ethyl + O2 system, using density functi calculations, B3LYP/6-31g* level. onsideration of the H shift bar similar to that of HO2 molecular elimination is however justified. Ignatyev et al. calculated value for the stable 2.OOH alkyl hydrop radical leads to a primary --H bond energy of 109 kcal/mole; signif higher than a normal primary, 101.6, or a --H bond on a carbon wit hydroxyl group (96 kcal/mole). It is reasonable that the error in functional calculation of 2.OOH carries over to the transition st shift to the peroxy group. Here the hybridization (SP3 on 's) and atomic charges remain similar in both structures. A second support argument is the calculated activation energy (Ea) for hydrogen shif peroxy, Morokuma and Lin - modified G2 calculation [17], is 2 to 3 lower than Ea for the corresponding HO2 elimination. The lower va Ea for the H shift are also in agreement with the Ea calculated by methyl peroxy system, after allowance for ring strain in a 4 versus ring is considered. The A-factor (Table 2) for the elimination channel is about a f higher than for H-shift due to a looser transition state for eli compared to H-atom shift. This difference in A-factor results in a constant for elimination versus H shift as discussed below. The A- unimolecular dissociation of 2.OOH differ by a factor of 15, also Table 2. The transition state to propene oxide (cy2o) + OH is mu than the TST to propene + HO2. The calculated (chemical activation) rate constants of iso-propy function of temperature and at one atmosphere are given in Fig. 2. dominant channel for iso-propyl + O2 is to isopropylperoxy radical (

9 7 (stabilization). The rate constant for this channel drops off at h temperature due to dissociation of 2OO. to reactants. The QRRK calculation for 2OO. (stabilized) shows that it principally dissoc reactants, 2. + O2. The reaction 2. + O2 = 2OO. is in equilibrium above 600 K and a small amount of the reactants ( percent) goes to products. The second channel shown in Fig. 2 is t to 2. + O2. This is the fraction of the 2. + O2 that goes to 2 comes back to reactants rather than to products or stabilization (2 HO2 elimination (3H6-E + HO2) and H-shift (3H6-S + HO2) are the dominant channels to products. The labeling of 3H6-E and 3H6-S i distinguish between propene made by the elimination channel versus shift channel. The rate constant for the elimination channel is a of 3 higher than the H-shift channel, as seen in Fig. 2. The diffe primarily due to the A-factor being higher for the elimination cha its looser transition state. The next channel shown is the one hydroperoxy-isopropyl radical (2.OOH). The QRRK dissociation resu (not shown) indicate that this radical principally proceeds to prop One can add the 2.OOH and 3H6-S + HO2 channels to get the total contribution of the H-shift channels to propene production, when r 2.OOH with O2 is not important. The remaining product channels h small contributions, with the highest of these being the channel t oxide (cy2o) and OH whose rate constant increases at high temperat Results from QRRK calculations are compared to rate constant measurements of Gulati and Walker [3]. They examined the reaction isobutyraldehyde, oxygen and nitrogen in a closed vessel at torr. The isopropyl formyl radical formed from isobutyraldehyde ra decomposes to iso-propyl + O. The yields of propene and propane f isobutyraldehyde oxidation were used to derive the rate constant fo

10 8 propyl + O2 = 3H6 + HO2. Gulati and Walker extrapolated their prod yields back to zero extent of reaction, so that reverse reaction of HO2 radicals should not play a role in interpretation of the data. potential concern about the effect of reaction of molecular oxygen 2.OOH in interpreting the experimental results. If this reaction important under their experimental conditions, it would lead to lo propene yields at higher oxygen concentrations. They examined a ra oxygen concentrations from 6-20% O2 and did not report a decrease i yield with increasing O2 level. Results from the QRRK calculations are compared to the rate con reported by Gulati and Walker. Three different cases were consider 1) Inclusion of both HO2 elimination and H-shift channels. 2) HO2 elimination only 3) H-shift channel only Activation energies for HO2 elimination and/or H-shift were adj give the iso-propyl + O2 rate constant reported by Gulati and Walke 1, the activation energies for HO2 elimination and H shift are both The rate constants obtained are listed in Table 3 (Reaction 2 and 5 constants are our current best estimate for these reactions. The a obtained with Gulati and Walker for this case is shown in Fig. 3. calculated curve exhibits a slightly positive activation energy com slightly negative activation energy shown by the experimental resul QRRK calculated curve includes contributions from both the HO2 elimination channel and the H-shift channel. In these comparisons constant to the 2.OOH channel was added into the total rate const propene + HO2. This is because the QRRK dissociation calculations 2.OOH show that the main fate of this radical is to propene + HO2 In case 2, the activation energy for the H-shift channel (React raised 8 kcal/mole (turned H-shift channel off) above the HO2 elim channel (Reaction 2). The 8 kcal/mole was chosen to agree with the difference in the H-shift and HO2 elimination barrier heights found

11 9 Ignatyev et al. [1]. The activation energy for HO2 elimination was to be 29.0 kcal/mole where agreement with Gulati and Walker s rate constants was obtained. The calculated curve is nearly identical t shown in Fig. 3. The resulting rate constants are given in Table 4 In case 3, the activation energy for the HO2 elimination channe (Reaction 2) was raised 8 kcal/mole above the H-shift channel (Reac order to investigate the case where the H-shift channel dominates. activation energy for the H-shift channel was found to be 26.9 kcal addition, the activation energy of 2.OOH = 3H6 + HO2 (Reaction 7) lowered from 20.7 to 18.2 kcal/mole which was the lower bound of ou estimate for this activation energy. If the H-shift channel domin barrier from 2.OOH to 3H6 + HO2 affects relative rates of further with O2 versus dissociation. The calculated curve is nearly identi one shown in Fig. 3. The rate constant measurements of Gulati and Walker can be expl by HO2 elimination or H-shift channels of iso-propyl + O2. The measurements do not distinguish between the two channels. Our cur best estimate is to include both channels with the rate constants g Table 3. Experimental data on epoxide product levels will provide data to distinguish between these two reaction pathways. ONLUSIONS The properties of four transition states important to the react propyl with O2 have been calculated and used to derive A-factors fo relevant reaction pathways. Activation energies for these pathways been estimated. It was found that the transition state for elimina has a looser transition state than that for the internal transfer o This leads to a higher A-factor for HO2 elimination than for H-tran estimated activation energies for HO2 elimination and H-transfer ar to be both 29.8 kcal/mole based on agreement with the data of Gulat

12 10 Walker. However, the experimental data did not provide any further information to distinguish whether the main reaction pathway is HO2 elimination or H-atom shift. Predictions of the rate constants fo channels over a wide temperature range at 1 atm are given. AKNOWLEDGMENTS The work at LLNL was carried out under the auspices of the U. S. Department of Energy by the Lawrence Livermore National Laboratory contract No. W-7405-ENG-48. The work at NJIT is carried out under from the US EPA Airborne Organic Research enter. REFERENES 1. Ignatyev, I. S., Xie, Y., Allen, W. D., and Schaefer III, H. F. of the 2H5 + O2 Reaction J. Phys. hem, In press, (1997). 2. McAdam, K. G., and Walker, R. W. J. hem. Soc., Faraday Trans 83: (1987). 3. Gutlati, S. K., and Walker, R. W. J. hem. Soc., Faraday Tran 407 (1988). 4. Wagner, A. F., Slagle, I. R., Sarzynski, D., and Gutman, D. J. 94: (1990). 5. Bozzelli, J.W., and Dean, A.M. J. Phys. hem. 94: ( Kaiser, E. W. J. Phys. hem. 99: (1995). 7. Koert, D., Pitz, W. J., Bozzelli, J. W., and ernansky, N. P. T Symposium (International) on ombustion, The ombustion Instit Pittsburgh, PA, 1996, pp MOPA: A General Molecular Orbital Package (QPE 445): APE Bull 43, MOPA6.0: Frank J. Seiler Research Lab., U.S. Air Forc Academy, O, 1990.

13 11 9. Ritter, E. R. and Bozzelli, J. W. Int. J. hem. Kinet. 23:767( 10. Lay, T., Bozzelli, J. W., Dean, A. M., and Ritter, E. R. J. Phy 99:14514 (1995). 11. Lay, T., Krasnoperov, L., Venanzi,., and Bozzelli, J. J. Phys 100: (1996). 12. Lay, T. and Bozzelli, J. W., personal communication. 13. Benson, S. W., Thermochemical Kinetics, John Wiley & Sons, N York, Dean, A. M., J. Phys. hem. 89:4600 (1985). 15. Dean, A. M., Bozzelli, J. W., Ritter, J. W. ombust. Sci. Tech (1991). 16. Gilbert, R. G., Luther, K., and Troe, J., Ber. Bunsenges. Phys 87:164 (1983). 17. Mebel, A. M., Diau, E. W. G., Lin, M.., and Morokuma, K. Ab i and RRKM calculations for multichannel rate constants of the 2 O2 reaction, J. Am. hem. Soc., In press, 1996.

14 12 Table 1: Thermochemical Properties of Important Species SPEIES Hf(298 S p ) kcal/ mole cal/ mol-k cal/ mol-k OH HO H * TIPRXOOH TIPROXOH TIPRYQE TIPROOXH OOH OOH OO OO OOH OO OOH OOH OO OO OO OOH OOH Y2O YO *O Notes: Dot denotes radical site on atom to left. arbon atoms are fully saturated with H-atoms. * = double bond T = transition state IPR = isopropyl E = elimination Y = cyclic X = represents the elimination of the species after x

15 13 Table 2: A-factors for Various Key Reactions alculated from Transition State Theory Temperature A-factors (K) 2OO. => 2OO. => 2.OOH => 2.OOH => 3H6 + HO2 2.OOH 3H6 + HO2 cy2o + OH E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E E+10 Table 3: High-Pressure Rate onstants Used for QRRK alculation of Isopropyl + O2 => Products (Used as input for chemical activation and unimolecular dissoci calculations) k=at n exp(-ea/rt) (mole-cm-kcal-sec units) Reaction A n Ea O2 => 2OO. 9.11E OO. => 2. + O2 7.78E OO. => 3H6 + HO2 5.47E OO. => O. + O 1.09E OO. => 2*O + OH 1.22E OO. => 2.OOH 5.06E OOH => 2OO. 2.56E OOH => 3H6 +HO2 1.01E OOH => cy2o+oh 1.31E OOH => *OOH + H3 3.89E Notes: Dot denotes radical site on atom to left. arbon atoms are saturated with H-atoms. * = double bond cy = cyclic structure containing the atoms that follow

16 14 Table 4: High-Pressure Rate onstants Used for ases 1 through 3 k=at n exp(-ea/rt) (mole-cm-kcal-sec units) ase Reaction A n Ea 1 2 2OO. => 3H6 + HO2 5.47E OO. => 2.OOH 5.06E OOH => 2OO. 2.56E OOH => 3H6 +HO2 1.01E OO. => 3H6 + HO2 5.47E OO. => 2.OOH 5.06E OOH => 2OO. 2.56E OOH => 3H6 +HO2 1.01E OO. => 3H6 + HO2 5.47E OO. => 2.OOH 5.06E OOH => 2OO. 2.56E OOH => 3H6 +HO2 1.01E Notes: Dot denotes radical site on atom to left. arbon atoms are saturated with H-atoms. * = double bond cy = cyclic structure containing the atoms that follow

17 O2 (21.0) O O H O O H O (18.0) (16.2) (16.2) (19.0) OH O O H 10 (8.5) Enthalpy (kcal/mole) O (-13.8) O O OH (-0.9) = + HO2 O + OH (-12.4) Fig. 1 Potential energy diagram of isopropyl + O2 => products.

18 16 13 k [cm 3 /mole-sec] OO. 2.+O2 10 3H6-E+HO2 3H6-S+HO2 2.OOH Y2O+OH 9 2*O+OH *OOH+H3 O.+O /T [K] Fig. 2 Results of QRRK calculation of isopropyl + O2 => products fo pressure of 1 atm. 3H6-E denotes propene via the molecular elimi channel and 3H6-S denotes propene formed via the H-atom shift cha

19 k (cm 3 /mole-sec) Gulati and Walker (1988) QRRK calculation /T(K) Fig. 3 Rate constant of isopropyl + O2 => propene + HO2 calculated analysis and measured by Gulati and Walker [3]. The calculated rat constants include the contributions from internal H-atom transfer elimination. The product 2.OOH was assumed to react to propene + which is its main fate.

20 Technical Information Department Lawrence Livermore National Laboratory University of alifornia Livermore, alifornia 94551

Pathway and Kinetic Analysis on the Propyl Radical + O2 Reaction System

URL-J-126375 Rev. 1 PREPRINT Pathway and Kinetic Analysis on the Propyl Radical + 2 Reaction System J.W. Bozzelli W.J. Pitz This paper was prepared for submittal to the Fourth International onference on