AUTHOR=Geißler Felix , Achterberg Eric P. , Beaton Alexander D. , Hopwood Mark J. , Clarke Jennifer S. , Mutzberg André , Mowlem Matt C. , Connelly Douglas P. TITLE=Evaluation of a Ferrozine Based Autonomous in Situ Lab-on-Chip Analyzer for Dissolved Iron Species in Coastal Waters JOURNAL=Frontiers in Marine Science VOLUME=4 YEAR=2017 URL=https://www.frontiersin.org/journals/marine-science/articles/10.3389/fmars.2017.00322 DOI=10.3389/fmars.2017.00322 ISSN=2296-7745 ABSTRACT=

The trace metal iron (Fe) is an essential micronutrient for phytoplankton growth and limits, or co-limits primary production across much of the world's surface ocean. Iron is a redox sensitive element, with Fe(II) and Fe(III) co-existing in natural waters. Whilst Fe(II) is the most soluble form, it is also transient with rapid oxidation rates in oxic seawater. Measurements of Fe(II) are therefore preferably undertaken in situ. For this purpose an autonomous wet chemical analyzer based on lab-on-chip technology was developed for the in situ determination of the concentration of dissolved (<0.45 μm) Fe species (Fe(II) and labile Fe) suitable for deployments in a wide range of aquatic environments. The spectrophotometric approach utilizes a buffered ferrozine solution and a ferrozine/ascorbic acid mixture for Fe(II) and labile Fe(III) analyses, respectively. Diffusive mixing, color development and spectrophotometric detection take place in three separate flow cells with different lengths such that the analyzer can measure a broad concentration range from low nM to several μM of Fe, depending on the desired application. A detection limit of 1.9 nM Fe was found. The microfluidic analyzer was tested in situ for nine days in shallow waters in the Kiel Fjord (Germany) along with other sensors as a part of the SenseOCEAN EU-project. The analyzer's performance under natural conditions was assessed with discrete samples collected and processed according to GEOTRACES protocol [acidified to pH < 2 and analyzed via inductively coupled plasma mass spectrometry (ICP-MS)]. The mechanical performance of the analyzer over the nine day period was good (consistent high precision of Fe(II) and Fe(III) standards with a standard deviation of 2.7% (n = 214) and 1.9% (n = 217), respectively, and successful completion of every programmed data point). However, total dissolved Fe was consistently low compared to ICP-MS data. Recoveries between 16 and 75% were observed, indicating that the analyzer does not measure a significant fraction of natural dissolved Fe species in coastal seawater. It is suggested that an acidification step would be necessary in order to ensure that the analyzer derived total dissolved Fe concentration is reproducible and consistent with discrete values.