Case Study CS-TC09 Rev. A

Page 3

© 2024 • Sales & Technical Support: (406) 587-4910 • email: sales@teamWavelength.com • web: www.teamWavelength.com

absorption wavelengths of all three sample gases. The

signals from the two microphones from both channels were

differentially amplified by a preamplifier and passed onto a

lock-in amplifier. These devices are seen in

Figure 1.

Test conditions, such as laser modulation frequency, laser

modulation depth, phase angle, and demodulation method

are all optimized to meet the best measurement conditions

of the three different gases.1 the resonance frequency was

determined to be at 710 Hz, and the photoacoustic signal at

different modulation depths is measured at this resonance

frequency for methane to find sufficient signal intensities.

To avoid spectral line broadening and wavelength shift, a

modulation depth of 10 mA is selected.

To determine the detection current of the three gases, the

absorption lines are observed over the full dynamic range

of the ICL.1 The laser's injection current is scanned from 50

to 100 mA to create a full profile of the gases' absorption

spectra with the DHR signals demodulated at the first (1f)

and second (2f) harmonic frequencies. These DHR signals

can be seen in

Figure 4. The 2f demodulation method is

undesirable for multi-component detection as the response

signals of ethylene and methane are almost in reverse

phases and will cancel each other out. These are the

blue and violet highlighted regions in

Figure 4b. A similar

situation arises with the 1f signal in the blue and violet

highlighted regions in

Figure 4a, but a detection window

is presented for ethylene at a higher injection current in

the range of 73.35 to 79.83 mA. Because of this detection

window, the demodulation method with the first harmonic

signal is chosen for signal acquisition.

The lock-in amplifier and the phase angle play an important

role in extracting weak signals from noisy backgrounds to

measure them accurately. The reference signal and the

detected signal will have some phase difference due to

the relaxation of exited gas molecules, energy loss from

friction of the photoacoustic cell wall, and the hysteresis

of the electronic circuit. To maximize the signal of all three

gases, a detection phase angle is set at 12.1º throughout

the scanning process.

RESULTS

Due to the cross-interference of the gas absorption signals,

each gas absorption signal will correspond to different laser

injection currents, and, therefore, emission wavelengths:

CH

4 - 62.35 mA, C2H4 - 73.88 mA, and C2H6 - 88.23 mA.

Detecting the gases at different wavelengths will help

increase signal intensity without signals canceling each

other.

Methane, ethylene, and ethane sample gases with varying

concentrations (2-95 ppm) were prepared for absorption

detection in the photoacoustic cell. At each selected laser

injection current, each concentration was measured to

generate a spectra curve. At these current levels, each gas

signal has a good linear relationship, but each has unique

responsivity levels derived from the linear fit. From the first

harmonic signals of each of the three gases (

Figure 5),

an Allan bias analysis can give the detection limits for

the appropriate integration time. This detection limit can

be compared to detection limits obtained in a separate

experiment.

Figure 4. (a) 1f and (b) 2f signals of the three analytes in the full dynamic range of the ICL current (50 to 100 mA).1