Analytical Methods For Gases And Particles In The Atmosphere

282 nm

308 nm x^n

FIGURE 11.43 Schematic diagram of OH energy levels used in LIF measurements.

(e.g., see Copeland et al., 1985; Copeland and Crosley, 1986; Crosley, 1989).

The major problem with LIF measurements in the past has been what might be called the "atmospheric uncertainty principle;" i.e., in the act of carrying out the measurement, the system is perturbed and artifact formation of OH can occur (e.g., see Smith and Crosley, 1990; and Hard et al., 1992b). This is primarily due to the photolysis of 03 to generate O('D), which in the presence of water vapor forms OH:

This artifact formation of OH is less severe when the excitation is at 308 nm, rather than 282 nm, since both the absorption cross sections and quantum yields for ozone photolysis decrease rapidly with wavelength in this region (see Chapter 4.B). Another advantage is that the absorption cross section for the (0,0) transition is about a factor of four larger than for the (0,1) transition, increasing the amount of excited OH. As a result, most LIF systems now use 308-nm excitation (e.g., Chan et al., 1990; Stevens et al., 1994; Hard et al., 1995; Holland et al., 1995). The disadvantage is a larger background signal at the fluorescence wavelength due to scattered laser light.

A second approach to minimizing the artifactual formation of OH in these measurements has been to sample the air through a nozzle into a low-pressure region operated at ~4 Torr. This was pioneered by O'Brien, Hard, and co-workers (Hard et al., 1984; Chan et al., 1990; Hard et al., 1995) and is known as the FAGE technique (/luorescence assay with gas expansion). The advantage is that the rate of generation of OH from the O('D) + H20 reaction is smaller, providing less in situ generation of OH in the laser beam. While the OH concentration in air is reduced proportionately with the pressure, collisional quenching of the electronically excited OH is as well; the result is that the OH LIF signal does not change substantially on reducing the pressure.

The "zero" signal in such instruments is usually established by adding an organic such as isobutane (e.g., Hard et al., 1992b) or ChF6 (Stevens et al., 1994; Dubey et al., 1996) that reacts rapidly with the OH. The difference in signal when the compound is added compared to when it is not is then a measure of the OH present. Another approach is to tune the laser on and off resonance with the OH absorption, permitting measurement of the background signal, which can be subtracted (e.g., Hofzumahaus et al., 1996).

Figure If.44 is a schematic diagram of one Li F instrument (Stevens et al., f 994; Brune et al., 1998). An air-cooled copper-vapor laser pumps a dye laser whose output at 616 nm is doubled to generate the 308-nm exciting radiation. An OH reference cell in which OH is generated from the thermal dissociation of water

Air inlet

OH detection chamber

OH reference cell

White cell

Doubling crystal Dye laser ^

Air inlet

OH detection chamber

OH reference cell

White cell

Doubling crystal Dye laser ^

Collecting optics White cell.

Mirror-

Laser focussing lens

Mirror-

Laser focussing lens i Detected „ , ' ^ airflow Bypass filter

Optical guide

Laser beam from dye laser

Collecting optics White cell.

> MicroChannel plate detector

Laser beam from dye laser

Optical guide

Photodiode laser power monitor

FIGURE 11.44 (a) Overall schematic diagram of an LIF instrument used for OH and H02 and (b) sample chamber in this instrument. (Adapted from Stevens et al., 1994.)

provides the reference for tuning the dye laser into resonance with the OH absorption. The beam is directed into a multipass cell with White cell optics as shown in Fig. If.44b.

One disadvantage of LIF compared to absorption measurements is the need for field calibration. It is a nontrivial issue to generate known concentrations of OH under ambient conditions for this purpose. A variety of approaches are used. These include photolysis of water vapor at 185 nm where the HzO absorption cross section as well as that of 02 are needed (e.g., Holland et al., 1998). However, there has been considerable uncertainty associated with these absorption cross sections (e.g., see Lazendorf et al., 1997; and Hofzumahaus et al., 1997, 1998). Stevens et al. (1994) used an internal calibration by titration of known concentrations of N02 with an excess of H atoms which generates OH via H + N02 -> OH + NO combined with external calibration using water vapor photolysis to account for transmission of OH through the sampling system. Sampling from a sample chamber in which a VOC-NO, mixture is irradiated and the rate of decay of the organics used to obtain the OH concentration has also been used (Hard et al., 1984; Chan et al., 1990).

As discussed later, LIF has also been used to measure HOz by conversion to OH by reaction with NO.

Mass spectrometry. Reaction of OH to form an ion, IIS04. which can be measured by mass spectrometry was first demonstrated by Eisele and Tanner (1991). Figure 11.45 is a schematic diagram of this approach (Tanner et al., 1997). Air is sampled through an inlet system described in detail by Eisele et al. (1997) and mixed with isotopically labeled 34S02, forming H24S04 via reactions discussed in Chapter 8.C.2:

Air flow

10 cm diameter

34S02 ± C3H8 injector -

Exhaust air

10 cm diameter

34S02 ± C3H8 injector -

HN03/C3H8 injector'

Exhaust air

Ion optics

Quadrupole mass filter

FIGURE 11.45 Schematic diagram of mass spectrometer used for OH measurements using derivatization approach (adapted from Tanner et al„ 1997).

HN03/C3H8 injector'

Collisional dissociation — chamber

Ion optics

Quadrupole mass filter

FIGURE 11.45 Schematic diagram of mass spectrometer used for OH measurements using derivatization approach (adapted from Tanner et al„ 1997).

Sufficient 34S02 is added to convert more than 99% of the OH in air to the acid. The use of isotopically labeled S02 forms labeled H2S04 which is not present in measurable quantities in air. Thus, labeled H2S04 is equal to the initial OH and allows H232S04 present in air to be measured simultaneously. Periodically during the measurements, propane is added simultaneously with the 34S02 at concentrations that will remove most of the OH, providing a background signal.

As discussed shortly, H02 and R02 react in the presence of NO to regenerate OH, which will lead to an overestimate of the OH concentration. To minimize this, propane is added downstream of the 34SOz injec tor to remove any of this regenerated OH. However, as discussed by Tanner et al. (1997), at high NOx concentrations, some regeneration does occur and the measurements must be corrected for that.

At this downstream port, HN03 is also added at concentrations such that the N0^(HN03) ion adduct is the major nitrate ion (see discussion of mass spectrometry in Section A.2). Since H2S04 is a stronger acid, it proton transfers to the cluster:

Figure 11.46a shows a typical mass spectrum. In addition to the N0^(HN03) ionizing agent, smaller amounts of NO^ and N0^"(HN03)2 are present. The HN03 adducts of both the naturally occurring 32 S and the added 34 S isotopes of HS04 are seen as well as the corresponding HS04 ions. These ions then enter a

11. ANALYTICAL METHODS FOR GASES ANL) PARTICLES IN THE ATMOSPHERE

Was this article helpful?

0 0

Post a comment