摘要:

J . Chem. SOC., Faraday Trans. 1, 1984, 80, 3503-3511 Studies of Reactions of Atoms in a Discharge-flow Stirred Reactor Part 5.-0(3P) + Trimethylamine BY DONALD L. BAULCH, IAN M. CAMPBELL* AND ROBERT HAINSWORTH School of Chemistry, The University, Leeds LS2 9JT Received 16th May, 1984 A discharge-flow system incorporating a stirred-flow reactor has been used to obtain stoichiometric coefficients for the rapid reaction between O(,P) and trimethylamine (TMA). The amounts of 0 atoms consumed and NO molecules produced when TMA was added to the stirred-flow reactor at 420 K were determined directly by measurements of chemiluminescent intensities. The yield of H atoms was estimated by a kinetic method when an excess of NO was present at relatively high pressures. On addition of CO to induce competition for OH between the reactions CO+OH+CO,+H (4) and O+OH + O,+H (5) the resultant decrease in 0-atom consumption indicated that 5.2 & 1.4 OH radicals were generated per TMA molecule reacted.Reaction (5) is the only likely source of 0, from the reaction. The stoichiometric equation deduced is (14.6&0.9)0+(CH3),N --+ NO+3(CO+C02)+(7+ 1)H+(5.2f 1.4)0, with the likelihood that a small yield of H, is generated also. The stoichiometric coefficient of 0 atoms was found to decrease by 1.2 0.3 in 0 + N, + CO systems when the point of addition of TMA was switched from the stirred-flow reactor to a section of the entry tubing, the times for reaction prior to observation changing from ca. 1 s (at 420 K) to ca. 15 ms (at 292 K) correspondingly.In view of the limited extent of possible reaction of, formaldehyde in 15 ms this excludes formaldehyde and hence its precursor CH, radicals from being major intermediates in the reaction. The major reactions composing the mechanism have rate constants > lo9 dm3 mol-' SKI at 292 K. In previous studies in our stirred-flow rea~torl-~ we have examined kinetics and mechanisms of reactions of O(3P) atoms in which the initial step has been relatively slow. Here we apply the technique to the study of a very fast reaction, O(3P) with trimethylamine (TMA), to obtain the overall reaction stoichiometry and hence information on elements of the kinetics and mechanism. Atkinson and Pitts5 obtained a rate constant of ca. 1.3 x 1O1O dm3 mol-1 s-l at room temperature for the initial step, and the small negative activation energy found over the range 298-440 K suggested formation of an 0(3P)-TMA complex.Anticipating our found overall stoichiometry of ca. 15 0 atoms consumed per TMA reacted and applying the above rate-constant value to our typical initial conditions of [O],/[TMA1o = 30 with [O], = lo-' mol dm-3, 99% consumption of TMA occurs on a time-scale < 10 ms, assuming that subsequent steps under the discharge-flow conditions are also rapid. 35033 504 DISCHARGE-FLOW STUDY OF o(3~) + TRIMETHYLAMINE < -81 - TMA/N2 Fig. 1. Diagram of the reaction system with numbers representing dimensions in mm. J1, 52 and 53 are jets for reagent addition, L1 and L2 are viewing points for the photomultiplier. Slagle et a1.,6 from a crossed-jet study of O(,P) + TMA, considered that an adduct was the initial product and found evidence that the adduct, regarded as an amine- N-oxide initially formed with 250-293 kJ mol-l of internal energy, decomposed predominantly via OH ejection but also via some CH, ejection.Kleinermanns and Luntz7 examined the reaction in a crossed-molecular-beam system and also proposed that addition is the rate-determining step in the thermal reaction. All of these previous studies have been concerned with the initial stages of the reaction only, and no investigation under discharge-flow conditions has been reported. In the present work we established directly the numbers of oxygen atoms consumed and NO molecules produced per TMA molecule consumed in the discharge-flow reactor from measurements of chemiluminescent intensities.Also, by the indirect kinetic technique developed in our previous w ~ r k l - ~ we obtain an estimate of the number of hydrogen atoms produced per TMA molecule reacted. Further, by comparison of the 0 and TMA stoichiometries in N, carrier and in N, carrier containing a large proportion of CO (which competes with 0 atoms for any OH produced) we obtained an estimate of the number of OH radicals produced as intermediates from a TMA molecule, the general method being as discussed before., EXPERIMENTAL The discharge-flow stirred reactor system has been described in detail before:' the reactor is shown in fig. 1. All internal surfaces were coated with a thin layer of Teflon to inhibit heterogeneous recombination of atom^.^ Tubular aluminium light guides were fitted at the viewing points L1 and L2, allowing these to be viewed by an RCA 1P28 photomultiplier through the appropriate one of two optical filters described later.The total pressure, in the range 0.2-0.5 kPa, was measured through 53 using a silicone oil manometer backed by a running vacuum. Most of the experiments were conducted with the hot box surrounding the reactor maintained at a temperature of 420 K, so that kinetic parameters from our previous are directly applicable. The general procedures for gas handling, purification, flowrate control and measurement were as described before.l-* Small flowrates of TMA in N, diluent were added usually as a split flow through the two sidearms to the sphere. TMA vapour was taken from a B.D.H.Chemicals Ltd lecture bottle (> 99% stated purity) and was used without further purification. A bulb (2 dm3) was filled to a measured pressure of ca. 1.3 kPa with TMA vapour and N, (B.O.C. White Spot) was addedD. L. BAULCH, I. M. CAMPBELL AND R. HAINSWORTH 3505 to a measured pressure of ca. 100 kPa: at least 24 h was allowed for mixing prior to use. When required, carbon monoxide (B.O.C. Technical Grade purified as described before,) was added through the sidearm in the entry tube to replace ca. 60% of the N, carrier. In some experiments the TMA+N, mixture was added with this CO flow. Upstream a microwave discharge unit (Microtron 2000) partially dissociated ( < 1 % ) N, flowing through the quartz tubing within the cavity. Typical N, flowrates were ca.80 pmol s-l, reduced pro rata when CO was added. The resultant N(4S) atoms were titrated with NO at J1, completely converting them into O(3P) atoms through the very fast reaction N(4S) +NO -+ N, + O(3P). The flowrate of NO required to achieve the endpoint (no glows present) established the oxygen-atom concentration, [O],, at J1 and hence also at the entry to the reactor (except in those experiments in which CO and TMA were added through the entry tubing sidearm): no significant decay of oxygen atoms on the time-scale of 15-30 ms between J l and the entry to the sphere was expected. With addition of a small excess of NO at J 1, the greenish air-afterglow was produced, associated with the combination reaction O+NO+M+NO,+M with NO regenerated by the rapid subsequent reaction O+NO, +NO+O,.The galvanometer reading (IG1) produced with the photomultiplier viewing L1 through a Wratten 61 filter (transmission range 478-602 nm, maximum at ca. 525 nm) is proportional to the product of entry concentrations [O],[NO], in the experiments without CO addition. The increase in IG1 when a measured small flowrate of NO was added through 52 allowed evaluation of [NO],. The galvanometer readings measured similarly with the photomultiplier viewing L2 were corrected by the methods described before' to obtain a value of IGz, which was directly comparable with IG1. IG2 is proportional to [O] [NO], these concentrations being those in the exit tubing and hence throughout the sphere when it was operating in its diffusion-stirred mode,' as was the case in this work.The increase in I,, when a measured small flowrate of NO was added through 53 allowed evaluation of [NO]. In turn this allowed evaluation of [O] through the calibrated proportionality factor for IG1, also applying to IG2. The changes in concentrations due to reactions when TMA is added through the sphere sidearms are defined as A[NO] = [NO]-[NO], and -A[O] = [O],-[O]. The residence time, At, in the reactor was calculated by dividing the volume of the sphere (0.54 dm3) by the total volume flowrate of the gases: it was of the order of 1 s in this work. Hydrogen atoms are an expected product of the overall reaction O(3P)+TMA. As will be explained in the discussion section, the reaction effectively occurs externally to the main reactor volume and the sphere can be regarded as containing the mixture of H and 0 atoms resultant from the reaction.Thus when substantial excesses of NO are added with total pressures of ca. 0.5 kPa, various cycles initiated by three-body combination reactions of 0 and H atoms (as detailed in the discussion section) increase - A[O]/[O] At with increasing [NO] to an extent reflecting [H]/[O]. Evaluation of this ratio by the methods described before' and in the discussion section then allow A[H] = [HI to be obtained. When ca. 60% of the N, carrier was replaced with CO, the bluish chemiluminescence associated with the combination reaction O+CO+M +CO,+M was observed downstream of the sidarm. This emission was viewed by the photomultiplier through an Oriel Optics G-774-3550 coloured glass filter (transmission range 290-410 nm, maximum at ca.360 nm). At L1 the corresponding galvanometer reading IB1 is proportional to [O],[CO], while at L2 IB2 (obtained with correction of the galvanometer reading as before2 to be comparable with 1,') is proportional to [O] [CO]. The small amounts of NO generated by the O+TMA reaction produced very small contributions to I , values, corrected for as described b e f ~ r e . ~ [CO] is very large and therefore effectively unchanged throughout.3506 DISCHARGE-FLOW STUDY OF o(3q + TRIMETHYLAMINE Table 1. Selection of the data obtained for the ratios of amounts of NO produced (A[NO]) to added concentrations ([TMA],) of trimethylamine for a range of values of VMA], and [O],/[TMA], in an O+N, system with [N,] = 6.5 x mol dm-3 (0.23 kPa) at T=420K [TMA],/ mol dm-3 [O],/[TMA], A[NO]/[TMA], 5.12 6.13 7.32 8.95 9.67 10.1 12.2 47.5 37.2 32.2 25.3 24.4 23.3 19.3 mean (all results) 1.04 1.13 0.96 1.01 0.92 0.94 1.08 = 0.97f0.07 A kinetic test was devised to establish if any comparatively slow reactions are involved in the 0 + TMA system.Oxygen-atom consumptions were compared when the same small flowrate of TMA (insufficient to consume > 70% of [O],) was switched from addition through the reactor sidearms to addition with the CO through the entry-tubing sidearm, with otherwise steady running. In the first mode (IB1 - IB2) measured the oxygen-atom consumption. The total pressure in these experiments is low, inhibiting three-body combination rates (for which small corrections were applied).Here the time-scale allowed for reaction is ca. 1 s. In the second mode, IB1 was measured before and after the TMA addition through the entry-tubing sidearm. In this case the difference in the two readings measures the oxygen-atom consumption to L1 on the time scale of ca. 15 ms. Essentially complete consumption of the TMA is expected (see Introduction), but any intermediate species which reacts with O(3P) with a rate constant of the order of < lo8 dm3 mol-1 s-l will not be significantly consumed. However, in the first mode such an intermediate will be essentially completely consumed provided that this rate constant exceeds ca. 3 x lo7 dm3 mol-l s-l. The generation of such an intermediate will be reflected in significant differences of oxygen-atom consumption in the two modes.RESULTS Table 1 shows a selection of values of the ratio of the amount of NO produced to the amount of TMA consumed (expressed as A[NO]/[TMA],, where [TMA], is the added concentration of TMA) when the TMA was added to the reactor through the sidearms of the sphere at relatively low total pressures. These and other data make it clear that there is a 1 : 1 stoichiometry between TMA consumed and NO produced. Fig. 2 shows a selection of values of the oxygen-atom stoichiometry parameter defined as n = -A[O]/[TMA], (corrected for the minor effects of cycles of reactions initiated by three-body combination processes) as a function of [O],/[TMA], under the same conditions as for table 1 (upper data). These and other similar data for TMA addition to 0 + N, mixtures in the reactor indicate that 14.6 f 0.9 oxygen atoms are consumed per TMA molecule reacted, not significantly dependent on the [O],/[TMA], ratio in a range extending over more than a factor of two.This is interpreted as indicating that the mechanism of the overall reaction does not vary significantly, so that it is likely that TMA and intermediate species react predominantly with O(3P) atoms under our conditions. Also shown in fig. 2 are data representing the measured values of n in theD. L. BAULCH, I. M. CAMPBELL AND R. HAINSWORTH 3 507 '* t I I I 20 30 40 50 Fig. 2. Plots of the oxygen-atom stoichometry number n = -A[O]/[TMA], against the initial concentration ratio of oxygen atoms to trimethylamine. ., TMA addition via sphere sidearms to O+N, system at 420 K: observation at L2.0, TMA addition via sphere sidearms to O+N,+CO systems at 420 K (CO is ca. 60% of carrier gas): observation at L2. A, TMA addition via entry-tubing sidearm to 0 + N, + CO systems at 292 K (CO is ca. 60% of carrier gas): observation at L1. Total pressure 0.23 kPa in all cases. [olo/[TMAlo experiments where TMA was added through the sidearms of the sphere and, subsequently, through the sidearm in the entry tubing, to O+N,+CO systems in which CO composed ca. 60% of the total gas and [CO]/[O] ratios at the observation points were in the range 900-2000. This switching procedure produced pairs of points. It is evident that n is depressed in the presence of CO, the mean value of n = 9.8 f 0.9 being obtained for TMA addition through the sphere sidearms; a change of An = - 1.2 f 0.3 results when TMA is switched to addition through the entry-tubing sidearm.It is notable that with these pairs of points there is no case in which An is not found to be negative. A set of experiments in which TMA was added via the sphere sidearms to 0 +NO + N, systems at relatively high pressures of ca. 0.45 kPa was conducted. Here an excess of NO was added at J1 to produce a substantial ratio of FJO] to [O,] (> 5) in the reactor, in order to minimize the effect of atom-removal cycles involving 0,. The results of these experiments are incorporated into the discussion section, the data obtained being shown in table 2 to follow. DISCUSSION It is assumed that the overall reaction is represented by the stoichiometric equation involving the likely products: nO+(CH,),N -+ NO+3(CO+CO,)+xH+yO,.3508 DISCHARGE-FLOW STUDY OF o(3q + TRIMETHYLAMI~ Water is discounted as a significant product: Slagle et aL6 did not detect H,O formation in the initial steps and its formation in later steps seems unlikely on mechanistic grounds.The equation incorporates our finding of unit stoichiometry between TMA and NO. Our measured - A[O]/[TMA], = n = 14.6 f 0.9. When TMA is added through the sphere sidearms to an excess of oxygen atoms, the large rate constant of the initial step ensures that the primary reaction zones will be in the throats of the sidearms. Only the ultimate reaction products will achieve uniform concentrations throughout the sphere. The analysis of the behaviour of the system is then based on the reasoned assumption that the 0(3P)+TMA reaction occurs effectively externally to the sphere.At relatively high pressures and with [NO] substantial, the significant reactions in the main volume are initiated by O+NO+N, +NO,+N, (1) and H+NO+N, +HNO+N, (2) Reaction (1) is followed by the stoichiometrically equivalent fast reactions O+NO,+NO+O, or H+NO,-+OH+NO O+OH+O,+H. Reaction (2) is followed by the fast reactions O+HNO-+OH+NO O+OH +O,+H. There is also some contribution to -A[O] from the cycle of reactions H+O,+N, +HO,+N, H +HO, + 20H O+HO,+OH+O, O+OH+O,+H. (3) Kinetic data for this have been discussed bef0re.~9 The experimentally measured A[O] is then expected to be predicted by the equation - A[O] = n[TMAl, + 2“,1 ANN01 (kl[O1 + k2[HI) + k3WI [021>.With rearrangement and making the close approximation that [O,] = -A[O]/2 this yields incorporating the mean value of n = 14.6. The mean rate-constant values used at the operating temperature of 420 K were (in units of 1Olo dm3 mol-1 s-l) k, = 1.8 (measured in the course of other work in this reactor), k, = 1.2 [ref. (l)] and k, = 1.5 [ref. (3)]. In these experiments suitable conditions were [O],/[TMA], > 50, -A[O]/[O] > 5 and [N0]/[0,] 2 5. Table 2 shows a selection of the measured parameters and resultant [H]/[TMA], values derived. The indicated value for the coefficient for H in the stoichiometric equation is x = 7_+ 1, the error limit being approximately one standard deviation.D. L. BAULCH, I. M.CAMPBELL AND R. HAINSWORTH 3509 Table 2. Selection of experimental parameters used in the determination of the hydrogen- atom production stoichiometry (T = 420 K, all concentrations in mol dm-3) 1.20 1 . 1 1 96.6 1.19 1.11 82.4 79.1 82.1 77.0 76.1 1.17 0.97 77.1 78.6 81.0 1.16 0.97 82.1 81.4 8.88 8.71 8.63 7.86 7.8 1 7.8 1 7.44 8.37 8.98 9.55 10.3 8.02 2.4 7.4 9.39 3.1 9.50 3.2 6.95 2.5 7.62 2.9 7.92 3.1 5.75 2.6 6.21 2.3 7.23 2.7 8.0 7.8 6.0 6.6 6.9 6.8 5.9 7.1 8.15 2.9 7.9 8.15 2.4 6.4 mean (all data) 7 f 1 In the experiments (at relatively low total pressures) in which TMA was added through the sphere sidearms to O+N,+CO systems, the primary reaction zones in the throats had [CO]/[O] ratios which probably exceeded those in the main volume to some extent. In any case, the OH produced in the reaction of 0 + TMA is consumed predominantly by with only minor competition from CO+OH -+ CO,+H (4) O+OH +O,+H.( 5 ) A value of k, = 1 .O x lo8 dm3 mol-1 s-l has been assessed8 as applicable for 300-500 K (at low total pressures). Lewis and Watsong measured k, in the range 221-499 K and their results, producing a mean value of k, = 1.6 x 1O1O dm3 mol-l s-l at 420 K, are in good agreement with other recent values. For [CO]/[O] = 900-2000 (1 8 O M O O O ) , the above k, and k, imply that 85-93% (92-96%) of OH produced is consumed uia reaction (4), the bracketed values representing the likely maximum average ratios in the primary reaction zones. The measured value of n = 9.8f0.9 in these systems compared to n = 14.6 f 0.9 without CO present is consistent with the generation of 5.2 f 1.4 OH radicals per TMA molecule reacted.Since reaction ( 5 ) appears to be the only source of 0, in the O+TMA mechanism, the coefficient for 0, in the stoichiometric equation is indicated to be y = 5.2+ 1.4. The experiments in which TMA injection was switched from the sphere to the entry tubing test for the involvement of CH, and HCHO as intermediates in the reaction. Under our conditions (excess CO present) the subsequent reactions will be O+CH,+HCHO+H (6) O+HCHO+OH+HCO ( 7 4 +H+HCO, (7 b) O+HCO-+OH+CO (8 a) +CO,+H (8 b)3510 DISCHARGE-FLOW STUDY OF o(3~) + TRIMETHYLAMINE O+HCO, +OH+CO, (9) CO+OH -+CO,+H. (4) Reaction (6) will be effectively instantaneous since k, = 6 x 1Olo dm3 mol-1 s-l at room temperature.1° Chang and Barker’s measurementsll over the range 296-437 K yield k, = lo8 dm3 mol-1 s-l at room temperature, much smaller than the other rate constants involved here.The subsequent reaction (8) will also be effectively instantaneous in the light of a rneasurementl2 of k, = 1.3 x loll dm3 mol-l s-l. Since [HI and [O] may be of the same order of magnitude in some of our systems, it is also necessary to consider the reaction H+HCO-+H,+CO (10) since k,, = 2k,., Reaction (10) can account for the consumption of a significant amount of any HCO formed and leads to the formation of H,. On the other hand, attack of HCHO by H atoms is unlikely to be of significance since the corresponding rate constant is only ca. 25% of k, at ambient temperatures.13 If CH, and/or HCHO are formed in the 0 + TMA reaction, this should be evident in that n will be decreased when the TMA injection is switched from the sphere sidearms (observation at L2) to the entry-tubing sidearm (observation at Ll).With a typical [O] = lo-, mol dm-, in the reaction zone, k, = 10, dm3 mot1 s-l leads to a half-life of HCHO of 70 ms, much longer than the 15 ms time-scale involved in the entry tubing, precluding reaction (7) from significance in the latter type of experiment. Accordingly the change in n on switching, An, should be substantial if either HCHO or its precursor CH, is a major intermediate. Our measured upper limit of An = - 1.5 would not be consistent with the formation of more than one CH, radical per TMA molecule reacted, taking into account the partitioning ratios of k7a: k7b: :0.7: 0.3 [ref.(1 I)] and k,, = keb [ref. (12)]. Within the error limits An = - 1.2k0.3 could be accommodated by the proposal of Slagle et alSs that some degree of CH, ejection results from the initial step of the 0 + TMA reaction, with no further ejection of CH, by subsequent reactions under our discharge-flow conditions. The most likely set of events is that subsequent to the initial step, the remaining CH, groups undergo substantial attack and conversion while remaining attached to the N atom, and HCHO is not released as a result of these processes. Further our experiments reveal that the major part of the 0 + TMA reaction under discharge-flow conditions involves steps with rate constants > lo9 dm3 mob1 s-l at 292 K. The overall stoichiometric equation for relatively long reaction times indicated by our measurements (in the absence of CO) is (14.6 f 0.9) 0 + (CH,),N -+ NO + 3(CO + CO,) + (7 f l)H + (5.2 1 .4)O,.The upper limit of the coefficient for H atoms appears to suggest that some H, is produced also, the obvious route being by way of formation of HCO and its subsequent reaction with H atoms. The relative extents of formation of CO and CO, are unknown. Within these uncertainties, the above equation can produce a mass balance provided that at least 0.5 H, is formed per TMA molecule reacted. Perhaps fortuitously our mean values themselves appoach a mass balance, within the uncertainties imposed by the lack of information on yields of CO, CO, and H,. R.H. expresses his gratitude to the S.E.R.C. for the award of a studentship.D. L. BAULCH, I. M. CAMPBELL AND R. HAINSWORTH 351 1 Part 1. I. M. Campbell and B. J. Handy, J. Chem. SOC., Faraday Trans. I , 1975, 71, 2097. Part 2. I. M. Campbell and B. J. Handy, J. Chem. SOC., Faraday Trans. I , 1978,74, 316. Part 3. I. M. Campbell, J. S. Rogerson and B. J. Handy, J. Chem. SOC., Faraday Trans. I , 1978, 74, 2672. Part 4. D. L. Baulch, I. M. Campbell and R. Hainsworth, J , Chem. Soc., Faraday Trans. I , 1984,840, 2525. R. Atkinson and J. N. Pitts, J. Chem. Phys., 1978,68, 911. I. R. Slagle, J. F. Dudich and D. Gutman, J. Phys. Chem., 1979,83, 3065. ' K. Kleinermanns and A. C. Luntz, J. Chem. Phys., 1982, 77, 3537. I. M. Campbell and B. J. Handy, Chem. Phys. Lett., 1977, 47,475. R. S. Lewis and R. T. Watson, J. Phys. Chem., 1980,84, 3495. lo N. Washida and K. D. Bayes, Int. J. Chem. Kinet., 1976, 8, 777. l1 J. S. Chang and J. R. Barker, J. Phys. Chem., 1979, 83, 3059. l 2 N. Washida, R. I. Martinez and K. D. Bayes, Z. Naturforsch., Ted A , 1974, 29, 251. l 3 A. A. Westenberg and N. deHaas, J. Phys. Chem., 1972, 76, 2213. (PAPER 4/505)

ISSN:0300-9599    

DOI:10.1039/F19848003503

出版商:RSC    

年代:1984

数据来源: RSC