Designing and experimental verification of a photoacoustic flow sensor using computational fluid dynamics

TL;DR

This paper presents a photoacoustic gas sensor with CFD-designed flow cell, demonstrating high sensitivity and real-time detection of hydrocarbons using an FPGA-controlled system and mid-infrared laser spectroscopy.

Contribution

The study introduces a CFD-optimized flow cell design for a photoacoustic sensor, enabling fast, sensitive, and real-time gas detection with FPGA control and mid-IR laser excitation.

Findings

Achieved 0.4 ppb sensitivity for hexane at 1.7 L/min flow rate.

Demonstrated a detection limit of 0.25 ppbV for decane.

Validated flow noise immunity and high sensitivity through experiments.

Abstract

A photoacoustic (PA) sensor for fast and real-time gas sensing is demonstrated. The PA sensor is a standalone system controlled by a Field-Programmable Gate Array (FPGA). The PA cell has been designed for flow noise immunity using computational fluid dynamics (CFD) analysis. The aim of the CFD analysis was to investigate and minimize the influence of the gas distribution and the flow noise on the PA signal. PA measurements were conducted at different flow rates by exciting molecular C-H stretch vibrational bands of hexane (C$_6$H$_{14}$) and decane (C$_{10}$H$_{22}$) molecules in clean air at 2950 cm$^{-1}$ (3.38 $\mu$m) with a custom made mid-infrared interband cascade laser (ICL). We observe a (1$\sigma$, standard deviation) sensitivity of 0.4 $\pm0.1$ ppb (nmol/mol) for hexane in clean air at flow rates up to 1.7 L/min, corresponding to a normalized noise equivalent absorption (NNEA) coefficient of 2.5$\times 10^{-9}$ W cm$^{-1}$ Hz$^{-1/2}$, demonstrating high sensitivity and fast real-time gas analysis. An Allan deviation analysis for decane shows that the detection limit at optimum integration time is 0.25 ppbV (nmol/mol).