1Institut für Ionenphysik und Angewandte Physik, Leopold-Franzens-Universität Innsbruck (IAP-LFUI), Innsbruck, Austria
2Atmospheric Chemistry Division, National Center for Atmospheric Research (NCAR), Boulder, USA
3Institut für Chemie und Dynamik der Geosphäre, ICG-II: Troposphäre, Forschungszentrum Jülich, Germany
4Institut für Meteorologie u. Klimaforschung, Atmosph. Umweltforschung (IMK-IFU), Garmisch-Partenkirchen, Germany
5Leibniz-Institut für Troposphärenforschung (ift), Leipzig, Germany
6Institut für Umweltphysik, Universität Bremen (iup-UB), Bremen, Germany
*now at: Cohausz~&~Florack, patent attorneys, Düsseldorf, Germany
**now at: Fachbereich C, Atmosphärenphysik, Bergische Universität Wuppertal, Wuppertal, Germany
Abstract. The atmosphere simulation chamber SAPHIR at the Research Centre Jülich was used to test the suitability of state-of-the-art analytical instruments for the measurement of gas-phase formaldehyde (HCHO) in air. Five analyzers based on four different sensing principles were deployed: a differential optical absorption spectrometer (DOAS), cartridges for 2,4-dinitro\-phenyl\-hydrazine (DNPH) derivatization followed by off-line high pressure liquid chromatography (HPLC) analysis, two different types of commercially available wet chemical sensors based on Hantzsch fluorimetry, and a proton-transfer-reaction mass spectrometer (PTR-MS). A new optimized mode of operation was used for the PTR-MS instrument which significantly enhanced its performance for online HCHO detection at low absolute humidities.
The instruments were challenged with typical ambient levels of HCHO ranging from zero to several ppb. Synthetic air of high purity and particulate-filtered ambient air were used as sample matrices in the atmosphere simulation chamber onto which HCHO was spiked under varying levels of humidity and ozone. Measurements were compared to mixing ratios calculated from the chamber volume and the known amount of HCHO injected into the chamber; measurements were also compared between the different instruments. The formal and blind intercomparison exercise was conducted under the control of an independent referee. A number of analytical problems associated with the experimental set-up and with individual instruments were identified, the overall agreement between the methods was fair.