Billions of people use biomass burning cookstoves in their homes and suffer serious health repercussions. Additionally, global warming is exacerbated by cookstove emissions containing greenhouse gases and particulate matter. Improved cookstoves (ICSs) mitigate the problem, but accurate and affordable emission gas measurements, particularly of Carbon Dioxide (CO2) and Carbon Monoxide (CO), are required in order to confidently declare ICSs cleaner burning than traditional cookstoves. The aim of this research is to assess the suitability of photoacoustic (PA) CO2 detection technology for cookstove emissions monitoring. The designs of several longitudinally resonant, photoacoustic, LED, CO2 sensors of varying levels of functionality are presented. Three aluminum cell designs allowed the detection of a photoacoustic signal: a 4cm long cylinder with a ~1cm diameter (Design 3), a 3.9cm long cylindrical resonator with ~1in diameter and quarter-acoustic-wavelength buffer volumes (Designs 4a,b), and a 3.7cm long cylinder with ~1in diameter (Design 5). All three cell designs operate in the longitudinal resonant mode via the irradiation of gases inside the PA cell with a 4.3um wavelength LED, driven at an on-off frequency in the kHz range by a square wave from an Arduino. A rudimentary lock-in amplifier (LIA) based on the AD630 was considered, but the SR830 LIA was actually used to extract the desired MEMS microphone signal from noise. Designs 3-4b produced PA signals dominated by wall-absorption, but the final design (Design 5) yielded a resonant PA signal proportional to CO2 concentration. It was discovered that photoacoustic gas detection is challenging to design and set up without extensive experience and equipment. Practical lessons learned are shared. Primary limitations with the presented designs are identified as the extremely low power of the 4.3um LEDs, wall absorption due to insufficient collimation of LED radiation, dependence on temperature, and reliance on an expensive, high performance, lock-in amplifier. Further testing and development of designs like Design 5 (short cylinder with large diameter-to-length ratio) is necessary to evaluate their potential for in-field, real-time CO2 concentration measurement. Though LED PA CO2 sensing was demonstrated to be possible, it is concluded that NDIR CO2 sensors are currently better suited for cookstove use. In addition to photoacoustic detection, a method of detecting CO2 concentration by measuring resonant frequency of the gas cell (The Acoustic Method) is presented.



College and Department

Ira A. Fulton College of Engineering and Technology



Date Submitted


Document Type





longitudinal, resonant, LED, photoacoustic, gas detection, CO2, resonant frequency, mid IR, LED, combustion emissions monitoring, biomass cookstoves



Included in

Engineering Commons