I. Introduction
Hydrothermal vents located underwater near volcanic rifts are a place where unique ecosystems have developed, relying on the energy provided by the hot water and the organic carbon produced by chemosynthetic microbes [1]. Moreover these underwater hot springs play an important role in ocean geochemistry since many metal ions are coming into the sea through them [2]. In the long run, polymetallic nodules (including manganese, iron, copper, etc.) deposit on the seafloor near the vents, which may become a new source for metallic ore mining [3]. Manganese (Mn) is an element that is usually found at the nanomolar level in the ocean [4] but can attain several micromolars in hydrothermal fluids [5]. It is thus a good candidate for a marker for the detection of hydrothermal vents. The ideal method for detection of Mn2 + should show both a high sensitivity and a good linearity over a wide range of concentrations, and should also be able to be miniaturized sufficiently to fit in compact autonomous underwater vehicles (AUVs) or remotely operated vehicles (ROVs) for in situ real-time monitoring. The in situ measurement further avoids the contamination problems during handling and storage of bottled samples, and allows the detailed mapping of Mn distribution in the ocean. Only a few detection systems actually performing in situ monitoring of Mn concentration in seawater are reported and they are either based on chemiluminescence [6] or on spectrophotometry [7]–[9]. Such techniques could benefit from microfluidics [10] since it allows the development of miniaturized devices that have a faster response, use less quantity of reagents, occupy less volume, and consume less power than their macroworld counterparts, and which have already proved efficient in ocean environment [11].