Info

At typical linear flow speeds of 1000 cm s_l, 1 cm along the tube corresponds to ~ 1-ms reaction time. Thus a flow tube of length 1 m can be used to study reactions at reaction times up to 100 ms.

Total pressures in most FFDS have typically been in the range 0.5-10 Torr where rapid diffusion across the flow tube ensures a relatively flat concentration profile of the reactants, so that Eq. (W) is valid. Maintaining the discharge used to generate atoms and free radicals is also difficult above a few Torr total pressure. The lower end of the pressure range is determined by the need to maintain viscous flow and to avoid significant axial concentration gradients. The latter may arise because of the lower concentrations of reactants at the downstream end of the flow tube; these can cause the true flow speed of the reactants to be greater than the calculated linear flow speed, due to their axial diffusion. Techniques for estimating errors due to such factors and for correcting measured rate constants for them are discussed in detail by Mulcahy (1973), Brown (1978), and Lambert et al. (1985).

However, flow tube systems for use at much higher pressures, up to several hundred Torr, have also been designed and applied to reactions of atmospheric interest (e.g., see Keyser, 1984; Abbatt et al., 1990, 1992; Seeley et al., 1993; and Donahue et al., 1996a). At these higher pressures, the velocity and radical axial and radial concentration profiles are experimentally determined and the full continuity equation describing the concentration profiles is solved.

A major factor in many FFDS studies is diffusion of the reactive species accompanied by their loss at the walls of the flow tube. Unfortunately, OH radicals are particularly sensitive to removal by wall reactions. While the mechanism and products of these wall reactions are unknown, it has been established that the rate of loss at the walls can be minimized by using various flow tube wall coatings or treatments. These include substances such as teflon or halocarbon waxes, which simply cover the entire surface so the incoming reactive species are only exposed to relatively unreactive carbon-halogen bonds, or treatment with boric or phosphoric acids. While such treatments have been shown to lower the rates of removal at the walls, why they do so is not clear.

Fortunately, the kinetics of the wall loss, measured from the decay of the reactive species in the absence of added reactant, are generally observed to be first order, so that corrections for these processes can be readily incorporated into the kinetic analyses. When these wall losses are significant, the integrated form of the rate expression (T) for reaction (17) of A + B becomes

where /cw is the observed first-order loss of A at the walls of the flow tube in the absence of B. The rate constant kxl can then be extracted from the slopes of plots of the pseudo-first-order rates of decay, r = (*17[B]„ + kj, against [B]„.

An example is shown in Figs. 5.5 and 5.6 for the reaction of OH with nitrosyl chloride, C1NO. Figure 5.5 shows the decay of OH resonance fluorescence emission intensity (proportional to the OH concentration) as a function of reaction time in a fast-flow discharge system at ~ 1 Torr total pressure as the concentration of C1NO is increased from 0 to 14.1 X 1013 molecules cm~3. As expected from Eq. (Y), the absolute value of the slope of the lines increases as [C1NO]0 increases. Figure 5.6 shows the plot of the absolute values of these slopes against [ClNO](). The slope of this plot gives the rate constant for the reaction of OH with C1NO,

under these conditions, which in this case gives kw = 5.6 X 10~13 cm3 molecule-1 s-1. The nonzero decay of OH when the C1NO concentration is zero is due to loss of OH at the walls of the flow tube.

These wall reactions can be a problem in FFDS studies. To avoid unrecognized interferences in the data associated with these heterogeneous reactions, as well as other secondary reactions, it is generally recommended that flow tube studies of a particular reaction be carried out using as many different wall coatings as possible. In addition, the use of different carrier gases

Coated Wall Flow Tube

Reaction time (ms)

FIGURE 5.5 Typical plot of OH resonance fluorescence intensity as a function of reaction time in the presence of increasing concentrations of C1NO (in units of fO11 molecules cm-3) at 373 K (adapted from Finlayson-Pitts et al., 1986).

Reaction time (ms)

FIGURE 5.5 Typical plot of OH resonance fluorescence intensity as a function of reaction time in the presence of increasing concentrations of C1NO (in units of fO11 molecules cm-3) at 373 K (adapted from Finlayson-Pitts et al., 1986).

and flow tubes of different diameters is recommended.

Flow tubes have also been used in combination with such techniques as mass spectrometry and FTIR for product studies. For example, high-pressure flow tubes with a White cell and FTIR at the downstream end

[CINO] (1013 molecules cm"3)

FIGURE 5.6 Typical plot of observed first-order rate constants for the decay of OH as a function of the initial CINO concentration at 373 K (adapted from Finlayson-Pitts et al., 1986).

[CINO] (1013 molecules cm"3)

FIGURE 5.6 Typical plot of observed first-order rate constants for the decay of OH as a function of the initial CINO concentration at 373 K (adapted from Finlayson-Pitts et al., 1986).

have been used with modulation of the reactants to obtain mechanistic information. In this approach, the radical source is modulated, and changes in the spectra with the source on and off are used to identify and quantify products (Donahue et al., 1996b).

3. Flash Photolysis Systems

As the name implies, this technique relies on flash photolysis to generate the reactive species A. In one of the most common configurations, resonance or induced fluorescence is used to monitor the decay of A—hence the name flash photolysis-resonance fluorescence (FP-RF). Since lasers are now frequently used as the photolysis source, the term laser flash photolysis-resonance fluorescence (LFP-RF) is also used.

Figure 5.7 is a schematic diagram of a typical FP-RF apparatus used to study chlorine atom reactions (Nicovich and Wine, 1996). For example, the fourth harmonic at 266 nm from an Nd:YAG laser can be used to generate chlorine atoms from the dissociation of phosgene, COCl2. After a preset time following the photolytic flash, the time decay of the reactive species is monitored using the fluorescence excited by a resonance lamp. Since B is present in concentrations in great excess compared to A, care must be taken to avoid impurities that may react with A or photolyze to produce reactive species that do. A restriction on the nature of B is that it must not photolyze significantly itself; reactions of such species as NOz and 03, which dissociate to produce highly reactive oxygen atoms, are often difficult to study with this technique. In addition, care must be taken to avoid the buildup of reaction products or of photolysis products in the photolysis cell, since some of these can photolyze and produce interfering secondary reactions. This is usually accomplished by using a slow flow of gas through the cell.

The limitations on the total pressure in the FP-RF cell are far less severe than those for FFDS. The lower end of the pressure range that can be used is determined by the need to minimize diffusion of the reactants out of the viewing zone. The upper end is determined primarily by the need to minimize both the absorption of the flash lamp radiation by the carrier gas and the quenching of the excited species being monitored by RF. In practice, pressures of ~ 5 Torr up to several atmospheres are used. The kinetic analysis is again typically pseudo-first-order with the "stable" reactant molecule B in great excess over the reactive species as outlined earlier. Table 5.5 gives some typical sources of reactive species used in FP-RF systems.

An example of the use of FP-RF to study the kinetics of an atmospherically relevant reaction is found in Fig. 5.8 (Stickel et al., 1992). Chlorine atoms were

FIGURE 5.7 Laser photolysis resonance fluorescence apparatus for studying the kinetics of gas-phase reactions of H, O, CI, and Br atoms with atmospheric trace gases. A/D, amplifier/discriminator; DDG, digital delay generator; FM, flow meter; IF, interference filter; MCS, multichannel scaler; PD, photodiode array detector; PG, pulse generator; PMT, photomultiplier. (Graciously provided by J. M. Nicovich and P. H. Wine, Georgia Institute of Technology.).

FIGURE 5.7 Laser photolysis resonance fluorescence apparatus for studying the kinetics of gas-phase reactions of H, O, CI, and Br atoms with atmospheric trace gases. A/D, amplifier/discriminator; DDG, digital delay generator; FM, flow meter; IF, interference filter; MCS, multichannel scaler; PD, photodiode array detector; PG, pulse generator; PMT, photomultiplier. (Graciously provided by J. M. Nicovich and P. H. Wine, Georgia Institute of Technology.).

formed by laser photolysis of COCl2 at 266 nm and detected using resonance fluorescence in the 135- to 140-nm region. As expected, the decay of CI in the presence of a great excess of CH3SCH3 (DMS) is exponential (Fig. 5.8a), and slopes of such decays are linear with the concentration of DMS (Fig. 5.8b). From the slope of the line in Fig. 5.8b, the rate constant at this temperature and pressure was determined to be (k = 2.71 ± 0.09) X 10"10 cm3 molecule"1 s"1.

4- Pulse Radiolysis

Pulse radiolysis has been used in a number of kinetic studies, for example by the group at RIS0 National Laboratory (Denmark) using the Febetron field emission accelerator facilities (e.g., see Nielsen and Sehested, 1993; Pagsberg et al., 1995; and Wallington et al., 1998). A short (30 ns) pulse of high-energy 2 MeV) electrons impacts a reaction cell containing an

TABLE 5.5 Some Typical Sources of Reactive Species in FP - RF Systems

Reactive species

Source

oh

Reactions of 0(' d), e.g., 03, n20 e.g., o, + hv -> o('d) + 02 o('d) + h2 -» oh + h or o('d) + h20 -> 20h hno, h2o2

ci

COCl2

oc3p)

o2

h

Alkanes, e.g., c^h^

ro

rono

atom or free radical source and reactant of interest. For example, if SFft is used as the bath gas, F atoms are generated, which can then undergo secondary reactions to form other reactant free radicals. For example, alkyl radicals can be generated from the F + RH reaction, and if carried out in the presence of 02, R02 is formed (e.g., see Nielsen and Sehested, 1993; and Wallington et al., 1998). Hydrogen atoms can be gener-

50 100 Time (us)

Was this article helpful?

0 0

Post a comment