FIGURE 11.5 Variation of laser frequency and signal with current for a typical lead salt diode laser (adapted from Werle et al, 1992).

the frequency output of the laser by modulating the current and thus the temperature of the diode (Reid et al., 1978). Absorbances down to ~f(T5 to 3 - l(Th can be measured using multipass cells, corresponding to ppb to sub-ppb concentrations for many pollutants of atmospheric interest; the limits for open-reflection systems are not as good (10~3-10~4) due to interference from atmospheric turbulence (Schiff et al., f994a, 1994b).

Figure 11.6 shows the major elements of a typical TDLS apparatus used for aircraft measurements (Hastie et al, 1983). Two diode lasers can be mounted on the dewar cold finger used for temperature/wavelength tuning; one is chosen for use by moving it into the appropriate position. A series of flat and off-axis parabolic mirrors are used to direct the laser beam into a White cell through which the air is pumped and back out to the sample detector. The He-Ne laser is used for alignment. A reference cell containing a high concentration of the species of interest can be inserted into the light path for calibration.

Figure 11.7 shows a typical 2/ spectrum for the 1597-cm"1 line of N02 obtained using this apparatus, compared to a calibration obtained using f .4 ppb N02. Fitting the ambient air spectrum to the reference gives an ambient air concentration of 72 ppt (Schiff et al., 1990).

TDLS is particularly useful for species such as H202 that are present at small concentrations and while very important in atmospheric chemistry, are difficult to measure. Figure 11.8 compares ambient H202 concentrations measured using TDLS and a wet scrubbing with enzyme fluorescence technique (Kleindienst et al., 1988a; Schiff et al., f994a, 1994b). The two are generally in agreement to within about 30%.

Table If.2 gives reported detection limits for some gases that have also been measured in the atmosphere by FTIR. As expected, the sensitivity of TDLS is significantly better than that of FTfR. For most species of atmospheric interest, detection limits are ~0.1 ppb for measurement times of f min in a 200-m White cell (G. Mackay, personal communication, 1998).

(4) Nondispersive infrared spectroscopy (NDIR) Figure 11.9 is a schematic diagram of the major components of an NDIR device (Skoog et al., 1998). As the name implies, it measures infrared-absorbing gases without dispersing the radiation or using FT techniques to derive wavelength-dependent signals. This method is also referred to as gas filter correlation. Infrared radiation is directed into two cells, one of which (the reference cell) is filled with a non-infrared-absorbing gas and the second of which (the sample cell) holds the sample (in a flow mode). The IR beams passing through the two cells then individually strike the compartments of the sensor cell, which are filled with the gas of interest and are separated by a thin, flexible metal diaphragm. When IR reaches this sensor cell, it is absorbed, causing heating and hence changes in pressure.

If the concentration of the absorbing gas is zero in the sample cell, the radiation striking both compartments is the same, and hence the heating is the same and there is no movement of the diaphragm separating

FIGURE lt.6 Schematic diagram of a TDLS apparatus (adapted from Hastie et al., 1983).


1.4 ppb

1.4 ppb

Was this article helpful?

0 0

Post a comment