Innovative calibration strategies for quality assurance and quality control of reactive trace gas analyzers

Reliable sub-ppb monitoring of reactive trace gas concentrations is essential for industrial and air quality monitoring purposes. Calibrating reactive trace gas analyzers is challenging because of the lack of certified primary standards.

Here we present a three-fold approach for calibration (validation) of Picarro’s Cavity Ring-Down analyzers (CRDS) for reactive trace gas monitoring: (i) calibration of a golden analyzer, (ii) validation of linearity using a surrogate gas approach, and (iii) accurate determination of the zero value. We use formaldehyde (H₂CO) to highlight best practices for QA/QC of reactive trace gas measurements and we demonstrate that this strategy can be applied to other reactive trace gases, such as NH₃, HCl, H₂O₂, and HF.

• The golden analyzer approach is based on a carefully calibrated inhouse reference instrument that is used as a transfer standard to cross-calibrate production units. After initial scaling based on the spectroscopy of Saha et al. [1], we present new data of primary formaldehyde standards (Apel-Riemer) that are used to adjust the scale of the golden formaldehyde analyzer.
• The surrogate gas validation approach is based on the principle that the accurarcy and linearity of the analyzer can be validated using a surrogate gas standard that is non-reactive, commercially available, and has a spectral adsorption line adjacent to the primary gas. In the case of formaldehyde, methane (CH₄) meets these criteria, and using methane standards therefore remove the need for regularly measuring formaldehyde standards.
• Accurate and regular determination of the zero value of a trace gas analyzer is key to achieve the highest data quality. We discuss the use of different scrubbing agents (DrieRite, 4,2-DNPH cartridges, activated charcoal) in combination with an automated valve switching procedure to track the zero drift of the G2307 formaldehyde analyzer (typically <0.33ppb in 72hrs).

Reference

[1] Saha et al., Molecular Physics, 2007