Airborne Measurements of Formaldehyde Employing A Tunable Diode Laser Absorption Spectrometer

The Dual Channel Airborne Laser Spectrometer (DCALS), as configured for simultaneous measurements of formaldehyde (CH₂O) and hydrogen peroxide (H₂O₂), is schematically shown in Figure 1. During TRACE-P, only the CH₂ O channel was operational, and the discussion that follows refers only to measurements of CH₂O. Infrared (IR) radiation at a wavelength of 3.5mm, which is emitted from a liquid nitrogen-cooled lead-salt diode laser, is collected and collimated by a 90-degree off-axis parabolic mirror (OAP). The radiation is then directed onto a second OAP mirror, which focuses the beam through a pinhole. The radiation then expands, hits an off-axis-elliptical (OAE) mirror and is ultimately re-focused into the center of a multipass astigmatic Herriott cell (Aerodyne Incorporated). The IR beam, which trace out a Lissajous pattern in the cell, achieves a total optical pathlength of 100 m in a 3-liter sampling volume. After 182 reflections, the IR beam become re-entrant and exits the cell through the entrance aperture. The exit beam is then directed onto two different liquid-nitrogen cooled detectors (sample detector and reference detector). As shown, the beam in the reference arm passes through a reference cell containing a few Torr of pure CH₂O. This cell serves as wavelength references by which to lock the laser output to the center of the absorption feature.

Data are acquired by repetitively sweeping the diode laser wavelength across an isolated CH ₂ O absorption feature (2831.6417 cm ^-1 ) at a frequency of 50-Hz and simultaneously applying a synchronous 50-kHz quasi-square modulation waveform to the laser tuning current. This line is free from all spectral interferences with the exception of a small positive interference (3.9%) from methanol. Both the sample and wavelength reference channel is demodulated at twice the modulation frequency employing commercial lock-in amplifiers. The amplifier outputs are then digitized and co-averaged by computer. The line center of the wavelength reference, which has very high S/N, is determined on every scan and used to appropriately shift the spectrum in memory to align the peak center before co-averaging. Additional details regarding data acquisition and on-line processing can be found in Fried et al. [1998a].

 Sample air is continuously drawn through the Herriott cell at flow rates around 9 standard liters minute^-1 (slm) using a heated 1/2-inch OD PFA Teflon line, which protrudes outside the aircraft boundary layer. A winglet structure external to the aircraft allows us to heat the inlet line to » 35 °C to within a few cm of the inlet entrance as well as add zero air to nearly the entire inlet. Zero air is generated by employing a second inlet using a diaphragm pump and a heated Pd/Al₂O₃ scrubber. This scrubbing system very effectively removes CH₂O without significantly affecting the ambient water vapor concentration. During ambient sampling, this zero air is dumped using a 3-way Teflon valve. During background acquisition, the zero air flow is re-routed and reintroduced back into the inlet line a few inches from the tip. Typically the zero air flow is » 2 to 3-slm higher than the total inlet line flow. As discussed by Fried et al. [1998a,b], background spectra are acquired approximately twice every minute, and this approach very effectively captures and removes optical noise as well as the effects of sample line outgassing. In fact laboratory measurements under a variety of relative humidities, temperatures, and sampling pressures, indicate accurate CH₂O retrieval using this approach [Wert et al., 2001].

  For calibration purposes, CH ₂ O standards are generated using a temperature and pressure-controlled permeation system. The CH₂O permeation system contains two permeation devices with different emission rates for calibration at two different concentrations. The CH₂O emissions rates are periodically determined by comparing the permeation output with that generated from a laboratory Henry’s Law calibration system. As discussed by Fried et al. [1997a], and Gilpin et al. [1997], the calculated output of this laboratory system has been verified by direct absorption spectroscopy, by two different cartridge methods, and by additional permeation devices calibrated by gravimetry. Based upon all the collective calibrations, we estimate a total calibration uncertainty of ± 6% at the 1s level for both permeation devices. The calibration system output is periodically added to the main inlet line a few inches downstream of the zero air addition port using a separate 1/8-inch addition port. Thus we calibrate and zero nearly the entire inlet line upstream of the sampling cell.

 During airborne operation, TDLAS measurements are acquired and stored in 5-second increments, and 12 such measurements are obtained before a 10-second background acquisition is acquired. Background spectra are acquired before and after each 1-minute period using an appropriate delay of 7-seconds (approximately 5 inlet/cell e-folding times) after each switch. The backgrounds surrounding each ambient block are averaged and subtracted point by point from each of the 12 ambient spectra.

 Each background-subtracted ambient spectrum thus acquired is fit in real time to a background-subtracted calibration spectrum acquired for each gas previously employing a multiple linear regression approach. Typically, the ambient-flush-background acquisition sequence is repeated for 60 minutes before a new calibration spectrum is acquired.  Each ambient measurement is corrected for drifts in laser power between calibration acquisitions, by measuring the detector dc voltages with and without the laser beams blocked using the lifting stage shown in Figure 1 [Fried et al., 1998a].

 Airborne replicate measurements of CH₂O employing 1-minute averages typically yield 1s standard deviations between 25 and 40 pptv.  The preliminary archived data are 1-minute averages. Longer averages can be applied for improved precision. Alternatively, averages as short as 5-seconds can be obtained for highly structured elevated plume measurements. The final data set will present these additional time-bases for selected interesting time periods. In addition, the final data will be corrected for the weak methanol interference using the onboard methanol measurements acquired by other groups.

References

 Fried, A., B. Henry, B. Wert, S. Sewell, and J.R. Drummond, Laboratory, ground-based, and airborne tunable diode laser systems: performance characteristics and applications in atmospheric studies, Appl. Phys. B, 67, 317 ^_ 330, 1998a.

 Fried, A., B.P. Wert, B. Henry, and J.R. Drummond, Airborne tunable diode laser measurements of trace atmospheric gases, SPIE Proc., 3285, 154 ^_ 162, 1998b.

 Fried, A., S. Sewell, B. Henry, B.P. Wert, T. Gilpin, and J.R. Drummond, Tunable diode laser absorption spectrometer for ground-based measurements of formaldehyde, J. Geophys. Res., 102, 6253 ^_ 6266, 1997a.

 Gilpin, T., E. Apel, A. Fried, B. Wert, J. Calvert, Z. Genfa, P. Dasgupta, J.W. Harder, B. Heikes, B. Hopkins, H. Westberg, T. Kleindienst, Y.-N. Lee, X. Zhou, W. Lonneman, and S. Sewell, Intercomparison of six ambient[CH₂O] measurement techniques, J. Geophys. Res., 102,  21,161 ^_ 21,188, 1997.