Atmospheric Environment 41 (2007) 1604–1615 Comparing field performances of denuder techniques in the high Arctic Antonietta Ianniello a,Ã , Harry J. Beine a , Matthew S. Landis b , Robert K. Stevens c , Giulio Esposito a , Antonio Amoroso a , Ivo Allegrini a a C.N.R.-Istituto sull’Inquinamento Atmosferico (IIA), Via Salaria Km 29.3, I-00016 Monterotondo Scalo (Roma), Italy b U.S. EPA Office of Research and Development, 109 T.W. Alexander Drive, Research Triangle Park, NC 27709, USA c Florida Department of Environmental Protection on assignment to U.S. EPA, Research Triangle Park, NC 27709, USA Received 18 May 2006; received in revised form 19 October 2006; accepted 24 October 2006 Abstract A field evaluation between two annular denuder configurations was conducted during the spring of 2003 in the marine Arctic at Ny-A ˚ lesund, Svalbard. The IIA annular denuder system (ADS) employed a series of five single-channel annular denuders, a cyclone and a filter pack to discriminate between gas and aerosol species, while the EPA-Versatile Air Pollution Sampler (VAPS) configuration used a single multi-channel annular denuder to protect the integrity of PM 2.5 sample filters by collecting acidic gases. We compared the concentrations of gaseous nitric acid (HNO 3 ), nitrous acid (HONO), sulfur dioxide (SO 2 ) and hydrochloric acid (HCl) measured by the two systems. Results for HNO 3 and SO 2 suggested losses of gas phase species within the EPA-VAPS inlet surfaces due to low temperatures, high relative humidities, and coarse particle sea-salt deposition to the VAPS inlet during sampling. The difference in HNO 3 concentrations (55%) between the two data sets might also be due to the reaction between HNO 3 and NaCl on inlet surfaces within the EPA- VAPS system. Furthermore, we detected the release of HCl from marine aerosol particles in the EPA-VAPS inlet during sampling contributing to higher observed concentrations. Based on this work we present recommendations on the application of denuder sampling techniques for low-concentration gaseous species in Arctic and remote marine locations to minimize sampling biases. We suggest an annular denuder technique without a large surface area inlet device in order to minimize retention and/or production of gaseous atmospheric pollutants during sampling. r 2006 Elsevier Ltd. All rights reserved. Keywords: Arctic; Atmospheric chemistry; Annular denuder; Multi-channel denuder 1. Introduction Annular diffusion denuders have been widely used to separate gas phase species from particles (diameter40.01 mm) without mutual interference in a variety of environments (Possanzini et al., 1983, 1999; Febo et al., 1989; Perrino et al., 2001; Beine et al., 2001, 2003; Ianniello et al., 2002). Gases have diffusion coefficients 3–6 orders of magnitude higher than those of submicron atmospheric parti- cles. When a laminar air stream passes through a suitably coated annular denuder, reactive gases (e.g., nitric acid and sulfur dioxide) diffuse to the denuder walls, and are retained. Particles, on the ARTICLE IN PRESS www.elsevier.com/locate/atmosenv 1352-2310/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.atmosenv.2006.10.040 Ã Corresponding author. Tel.: +39 06906 72262; fax: +39 06906 72660. E-mail address: ianniello@iia.cnr.it (A. Ianniello).