A Remote Sensor System Based on TDLAS Technique for Ammonia Leakage Monitoring

Abstract

The development of an efficient, portable, real-time, and high-precision ammonia (NH₃) remote sensor system is of great significance for environmental protection and citizens' health. We developed a NH₃ remote sensor system based on tunable diode laser absorption spectroscopy (TDLAS) technique to measure the NH₃ leakage. In order to eliminate the interference of water vapor on NH₃ detection, the wavelength-locked wavelength modulation spectroscopy technique was adopted to stabilize the output wavelength of the laser at 6612.7 cm^-1, which significantly increased the sampling frequency of the sensor system. To solve the problem in that the light intensity received by the detector keeps changing, the 2f/1f signal processing technique was adopted. The practical application results proved that the 2f/1f signal processing technique had a satisfactory suppression effect on the signal fluctuation caused by distance changing. Using Allan deviation analysis, we determined the stability and limit of detection (LoD). The system could reach a LoD of 16.6 ppm·m at an average time of 2.8 s, and a LoD of 0.5 ppm·m at an optimum averaging time of 778.4 s. Finally, the measurement result of simulated ammonia leakage verified that the ammonia remote sensor system could meet the need for ammonia leakage detection in the industrial production process.

Keywords: 2f/1f signal processing technique; remote sensor; tunable diode laser absorption spectroscopy; wavelength modulation spectroscopy; wavelength-locked.

Figure 1

(a) Absorption spectra of NH₃, H₂O, CO₂ from 6550 cm⁻¹ to 6650 cm⁻¹. (b) Absorption spectra of NH₃ (1000 ppm·m, blue line), H₂O (1%, red line), and CO₂ (0.03%, yellow line) at an optical path of 60 m based on HITRAN database and the plot of drive current versus wavenumber of the distributed feedback (DFB) laser (black line), the value of modulation current at the central wavenumber 6612.7 cm⁻¹ is 69.48 mA.

Figure 2

Output wavenumber of the DFB laser versus drive current at 20 °C, 22 °C, 24 °C, 26 °C, 28 °C, 30 °C, and 32 °C. The red line is the linear fitting line of the selected temperature (32 °C). The slope of the fitting line is 0.06156, which is the reciprocal of the slope of the black line shown in Figure 1b.

Figure 3

Schematic of the ammonia remote sensor system. This system consists of three parts: an optical subsystem, an electrical subsystem, and an upper computer.

Figure 4

Modulation depth optimization performed at 2000 ppm∙m NH₃, the optimal modulation depth is 0.4199 cm⁻¹ with a modulation amplitude of drive current at 6 mA.

Figure 5

(a) Calibration of system by using a gas sampling bag with integral concentration from 0ppm·m to 2000 ppm·m, each concentration level was measured and recorded for about 3 min. (b) Measured data dots and linear fitting curve of the NH₃ integral concentration versus 2f/1f signal amplitude.

Figure 6

The change trend of the first harmonic, second harmonic, and the 2f/1f signal of the absorbed signal versus the remote sensing distance.

Figure 7

(a) The stability test of the remote sensor system for 1 h in the corridor of the building. (b) Allan deviation analysis of the sensor based on the stability test data shown in (a).

Figure 8

(a) The outdoor experiment where NH₃ gas cylinders were set at a distance of 25 m to simulate industrial NH₃ leakage; the concentration order from right to left is 2000 ppm, 10,000 ppm, 50,000 ppm; and the distance from target is 30 m. (b) The integral concentration of NH₃ at the three leakage ports obtained by the NH₃ remote sensor system based on the test conditions shown in (a).

Figure 9

(a) The horizontal distribution of NH₃ concentration 0.2 m above the leakage source. (b) The vertical distribution of NH₃ concentration at the leakage source.