Highly precise and real-time measurements of 13CO2/12CO2 isotopic ratio in breath using a 2 μm diode laser

Abstract

Real-time breath gas analysis with high accuracy, precision and time resolution, as a promising, non-invasive, fast and reliable tool, is important in medical diagnostics. Especially stable isotopologues of carbon dioxide is applied to multiple research areas including the diagnosis of gastrointestinal diseases. Helicobacter pylori (H. pylori) is one of the most frequent bacterial infectious diseases in human beings and is now recognized as one of the key risk factors for chronic gastritis, peptic ulcers, stomach cancer and lymphoma. In contrast to traditional invasive tests, the most reliable non-invasive method in the diagnosis of the H. pylori infection is considered to be 13C-urea breath test which is implemented by measuring the 13CO2/12CO2 isotope ratio in human breath. Tunable diode laser absorption spectroscopy (TDLAS) has the advantages of fast response, low drift, good gas selectivity and high detection sensitivity, and it is very convenient to develop a high precision, real-time and online measurement system. A precision laser spectrometer for the measurement of CO2 isotope abundance in human breath (with CO2 concentration of 4%-5%) or high concentration gas is designed and evaluated based on TDLAS technology. The spectrometer contains a novel compact dense-pattern multipass cell with a small volume of 280 cm3 and an effective optical path length of 26. 4 m. The cell is in conjunction with a fiber-coupled distributed feedback diode laser operating at 2.008 μm. Wavelength modulation spectroscopy approach is used. The mass flow, pressure and temperature of the cell are actively controlled, and able to keep long-term stability. The influence of laser power fluctuation is eliminated by fitting the baseline with cubic polynomial to normalize the raw spectrum. Moving window regression is used to remove the influence of frequency drift on measuring isotope abundance. The system measurement precision is improved by wavelet denosing and Kalman filtering. The experimental results demonstrate that moving window regression method not only extends the stability time of the system but also improves the measurement precision of isotope abundance well, the wavelet denoising improves the signal-to-noise ratio by 2 times that by the method of multi-spectral average, the stability time of the system is 100 s given by Allan variance, and the measurement precision of CO2 isotope ratio is 0. 067‰ after Kalman filtering. The use of small multi-pass cell and the default of denoising devices make the system more portable and improve the real-time and online measurement performance of the system. In addition to the measurement of 13CO2/12CO2 isotope ratio in human breath, by replacing different lasers, the spectrometer can also be used to measure trace gas concentration and the stable isotope abundance of many gas molecules in atmosphere. Therefore, the spectrometer will have broad applications in the areas of medical diagnosis, carbon cycle study and environmental monitoring.