Frequency modulation and wavelength modulation spectroscopies: comparison of experimental methods using a lead-salt diode laser

TLDR

Modulation and detection frequencies span over five orders of magnitude, with WMS below 100 kHz limited by laser excess (1/f) noise. The study compares WMS and one‑tone and two‑tone FMS by measuring the minimum detectable absorbances with a mid‑IR lead‑salt diode laser. The authors compare WMS and one‑tone and two‑tone FMS by measuring minimum detectable absorbances with a mid‑IR lead‑salt diode laser, test high‑frequency WMS on tropospheric N₂O, and discuss sensitivity limits from interference fringes and étalon reduction schemes. High‑frequency WMS at 10 MHz achieves absorbances in the 10⁻⁷ range over a 1‑Hz bandwidth, limited by detector thermal noise, enabling sub‑ppb detection with multiple‑pass cells, whereas FMS methods at ≥150 MHz perform poorly due to inefficient modulation coupling, yet the results agree well with theory.

Abstract

Wavelength modulation spectroscopy (WMS) and one-tone and two-tone frequency modulation spectroscopy (FMS) are compared by measuring the minimum detectable absorbances achieved using a mid-IR lead-salt diode laser. The range of modulation and detection frequencies spans over 5 orders of magnitude. The best results, absorbances in the low-to-mid 10(-7) range in a 1-Hz bandwidth, are obtained by using high-frequency WMS (10-MHz detection frequency) and are limited by detector thermal noise. This sensitivity can provide species detection limits well below 1 part per billion for molecules with moderate line strengths if multiple-pass cells are used. High-frequency WMS is also tested by measuring the absorbance due to tropospheric N(2)O at 1243.795 cm(-1). WMS at frequencies < 100 kHz is limited by laser excess (1/f) noise. Both of the FMS methods, which require modulating the laser at frequencies >/= 150 MHz, give relatively poor results due to inefficient coupling of the modulation waveform to the laser current. The re ults obtained agree well with theory. We also discuss the sensitivity limitations due to interference fringes from unintentional étalons and the effectiveness of étalon reduction schemes.