The valorization of massive strandings of holopelagic Sargassum spp. is strongly limited by high levels of inorganic arsenic (Asi) that are potentially above the limit of current regulations. Monitoring Asi in algal biomass is currently achieved using standard chromatographic separation followed by spectroscopic detection. Here, we propose an alternative simpler procedure based on the extraction of Asi from the freeze-dried algal powder in deionized water and the electroanalytical detection of the diluted extract at a gold-microwire electrode. The protocol was optimized both in terms of extraction (powder/water ratio, extraction time, temperature) and electrolyte used for the voltammetric detection. Two electrolytes were tested: one composed of citric acid, sulfamic acid and KCl (pH 2.0) and another composed of an acetate buffer (pH 4.7) and NaCl. We demonstrate here that Asi determination is possible with the first electrolyte but it is necessary to deal with a relative unstable signal. Measurement of Asi was best achieved with the second electrolyte (acetate buffer and NaCl) with the following optimized electrochemical conditions: deposition potential of -1.2 V, deposition time of 30 seconds and linear scan voltammetry. Voltammetric results were then compared to a reference method (HPLC-ICP-MS) using different morphotypes of holopelagic Sargassum spp. (S. natans VIII, S. natans I and S. fluitans III), using commercial extracts of brown seaweeds and using a Hijiki certified reference material. Very good agreement was obtained between our novel method and HPLC-ICP-MS. Both methods show that inorganic arsenic is almost entirely present as As(V) in Sargassum spp. extracts.

Metal partitioning between the dissolved and particulate phases is still poorly constrained within the early mixing of hydrothermal fluids and deep seawater. In this study, in situ filtration has been used to collect early buoyant plume fluids. This has provided the unique opportunity to reassess precisely metal partitioning along the mixing gradient by limiting chemical exchange processes between the dissolved (<0.45 μm) and particulate (>0.45 μm) phases during sampling. We report on the partitioning of three major metals (Fe, Cu, Zn) in the early buoyant plume of six black and clear smokers from the Lucky Strike hydrothermal field (37°N, MAR; EMSO-Azores deep sea observatory). We show that chemical changes are limited in the warmest part of the plume [50–150°C, dMn > 40 μM; dilution factor (DF) of ~1–10 by NADW] as metal partitioning displays a chemical signature similar to the end-member one. However, as the dilution ratio between the hydrothermal fluid and North-Atlantic Deep-water (NADW) increases (4–50°C, dMn < 40 μM; DF of 10–100 by NADW), metal partitioning is affected by different precipitation and oxidation processes. Molar ratios normalized to Fe in the particles highlight the onset of Fe oxides formation, the precipitation of barite and the decreasing contribution of sulfide minerals (mainly Cu–Fe sulfides and sphalerite/wurtzite) along with fluid dilution.

Deep-sea hydrothermal venting is now recognized as a major source of iron (Fe), an essential trace element that controls marine productivity. However, the reactions occurring during dispersal from buoyant plumes to neutrally buoyant hydrothermal plumes are still poorly constrained. Here we report for the first time on the dissolved-particulate partition of Fe after in situ filtration at the early stage of mixing at different hydrothermal discharges, i.e., Lucky Strike (37 degrees N), TAG (26 degrees N), and Snakepit (23 degrees N) on the Mid-Atlantic Ridge. We found that hydrothermal iron is almost completely preserved (>90%) in the dissolved fraction, arguing for low iron-bearing sulfide precipitation of iron in basalt-hosted systems with low Fe:H2S ratios. This result can only be explained by a kinetically limited formation of pyrite. The small part of Fe being precipitated as sulfides in the mixing gradient (<10%) is restricted to the inclusion of Fe in minerals of high Cu and Zn content. We also show that secondary venting is a source of Fe-depleted hydrothermal solutions. These results provide new constrains on Fe fluxes from hydrothermal venting.

In this study, we focused on the behavior of molybdenum in a macrotidal estuarine system (Aulne-Bay of Brest, France). A systematic and pronounced loss of dissolved molybdenum was observed in the inner part of the system which was ∼10–30 fold higher than the inputs from the fluvial waters. Further to the examination of Mo concentrations in the water column and in the sediment pore waters over a 6-month period at a coastal station, we found that the systematic deficit in Mo concentrations in the coastal area was related to the loss of metal in the inner part of the system. We also estimated that benthic inputs in the downstream part of the system could be sufficient to counterbalance the upstream losses. After its incorporation in particles within the high turbidity area of the estuary, we postulate that Mo is redistributed downward following the deposition of the particles in areas of sedimentation and the redissolution of Mo in pore waters.

This work reports the determination of ultra-trace of Sb(III) in seawater by using a stripping chronopotentiometric (SCP) method with a mercury film electrode. A sensitivity and detection limit of 360 ms L microg(-1) and 8 ng L(-1) (70 pM), respectively, were accomplished for a 15-min electrolysis time. Compared to the only two chronopotentiometric methods reported for Sb(III) determination in seawater, our method is more sensitive and does not need to use a medium exchange procedure before the stripping step. Moreover, the use of a double electrolysis potential (-450 mV and -250 mV) allows the analysis of Sb(III) independently from the Cu level in the sample. The method was successfully used to study the behaviour of dissolved Sb(III) in the Penzé estuary, NW France.

The study reported here was carried out in the macrotidal estuary of Penzé (Brittany, Western Channel, France). Ten field stations along the freshwater–seawater mixing zone were sampled each month in order to examine the speciation of dissolved and particulate cadmium. Different biogeochemical parameters (suspended particulate matter, Chlorophyll-a, pH and dissolved organic carbon) were also measured. Levels of total dissolved and total particulate metal ranged from 0.06 to 0.46nM and from 0.4 to 2.4μgg−1, respectively. Our data showed a non-conservative behaviour for cadmium as a result of desorption from particles. The rise in total dissolved cadmium in the salinity range 0–20 was about 0.2nM. Incidentally, a decrease of total particulate concentrations of about 0.5μgg−1, which corresponds approximately to 0.2nM, was observed. The study of the various cadmium species led us to identify desorption of cadmium occurred from the iron/manganese oxide-associated fraction. This desorption process leads to the formation of labile complexes, i.e. chlorocomplexes in spring and summer. However, in winter and autumn, a large part of the dissolved fraction was constituted by organic complexes. These complexes can be formed from suspended particles desorption or could be a released from sediment. Analysis of flux indicated that most of the metal inputs to the estuary (usually more than 80%) corresponded to organic cadmium. Spring and summer outputs were mainly constituted of chlorocomplexes whereas cadmium organic complexes were predominant in winter and autumn output fluxes.

This one-year survey conducted in the macrotidal estuary of Penzé (Brittany, Western part of the Channel, France) was aimed at examining the variations of the various dissolved and particulate copper species. Ten field stations along the whole freshwater–seawater mixing zone were sampled each month. Different biogeochemical parameters (SPM, chl-a, pH and DOC) were also measured. The levels in total dissolved and total particulate copper ranged from 1.8 to 9.5nM and from 5 to 98μgg−1, respectively; such amounts are indicative of a pollution-free system. Extractable C18 copper (non-polar hydrophobic organic copper species), in winter and spring, accounted for 30–40% of the total dissolved copper. In summer, this contribution rapidly rose to 60% in the salinity range 20–30; over the same period of time, total particulate copper decreased. The change in dissolved copper speciation and the lowering of particulate copper concentrations were attributed to the release of strong organic ligands by phytoplankton. Our field data evidenced a highly variable behaviour for the various copper species over the seasonal cycle, and then led us to identify the following mechanisms: (i) metal desorption from organic river-flown particles (winter and spring), (ii) metal desorption from resuspended sediment in the upstream section (summer), (iii) competition between particles, non-extractable C18 organic ligands and phytoplankton-released extractable C18 organic ligands to complex copper in the downstream section (summer), and (iv) removal of non-extractable C18 organic copper by adsorption (autumn). Dissolved copper species fluxes were assessed: most of metal inputs to the estuary (60–74%) corresponded to non-extractable C18 organic copper. Winter and spring metal output fluxes were mainly constituted of non-extractable C18 organic complexes; on the other hand, extractable C18 organic complexes were predominant in summer and autumn output fluxes.

Vertical distribution of cadmium (27–133 pM) at 13 stations in the Alboran Sea (western Mediterranean) showed profiles that were different from those typically known from oceanic waters. A subsurface cadmium-rich layer, related to the presence of Winter Intermediate Water (WIW), was a prominent feature. The WIW is formed in the northwestern coastal sector of the western Mediterranean Sea, and the high cadmium concentrations in this water mass are therefore likely to be of coastal origin. The surface concentrations of cadmium at all stations in the Alboran Sea were higher and increased, but not progressively, from west to east, with a substantially higher concentrations around the Almeria–Oran front than on either side. As shown by the cadmium–salinity diagram and the rates of diffusive transfer of cadmium from the subsurface into the surface waters, flux from WIW is an important source of cadmium for surface waters in the Alboran Sea, with a marked increase in the rates near the front.

The investigations reported here focused on the amounts of various dissolved cadmium and copper species in the whole salinity gradient of the system composed of the Loire estuary and the Biscay Bay continental shelf (France). Electrochemical labile cadmium, extractable C-18 organic copper and total dissolved species of these two metals were analysed. Within the estuary, levels of total dissolved cadmium and copper ranged from 0.08 to 0.29 nM and 8.0 to 21.2 nM, respectively. These values are higher than those currently reported in metal input-free estuarine waters, but remain lower than those found in heavily industrialised environments. At the sampling stations located on the continental shelf, concentrations of total dissolved cadmium varied within 0.11 and 0.26 nM, whereas those of copper were in the range 1.7-8.0 nM. Both these series of data are higher than those reported in the literature for areas in the close vicinity of the continental shelf and in the open Atlantic Ocean.Our experimental values highlighted a non-conservative behaviour on estuarine mixing for both metals, indicative of additional metal inputs. For cadmium, the rise in concentration was particularly strong: 360%. Further to the study of labile cadmium and non-polar hydrophobic organic copper fractions, the observed elevations in concentrations were related to particle desorption phenomena or to inputs from sediments. Discharges in dissolved cadmium and copper species to the continental shelf by the Loire river were assessed as 14 and 343 kg day(-1), respectively. Among them 22% of cadmium flux corresponded to labile complexes, 36% of copper constituted of non-polar hydrophobic complexes. Concerning both metals a large area of the continental shelf was affected by waters from the Loire river. No change in the partition between hydrophobic and non-hydrophobic organic copper complexes was observed in this area, contrary to the cadmium one for which changes in chemical speciation were noticeable along the salinity gradient. (C) 2003 Elsevier Ltd. All rights reserved.

A stripping chronopotentiometric method, using a rotating gold disk electrode for mercury measurements in sea water is described. Compared with a same method using a stationary gold film electrode, this method has a eight times higher sensitivity and a detection limit of 5ngl−1 after 10min deposition time. Moreover, the time needed for gold plating is eliminated. Compared with other electrochemical methods capable of measuring mercury at low concentrations, the present method is more simplified with no degassing step and no need to use a medium-exchange procedure before the stripping step. These characteristics render the method easily practicable on board oceanographic vessels for ‘in situ’ measurements.

A chronopotentiometric stripping method [constant current stripping analysis (CCSA)] using a rotating gold disk electrode for measurements of copper in sea-water is described. Compared with a CCSA method using a stationary gold film electrode, the new method has a sensitivity 2.5 times higher and a detection limit of 11 ng l–1 after a 5 min deposition time, and eliminates from the analysis the time needed for gold plating. The precision and accuracy of this method agree well with those obtained by other techniques. Vertical profiles of copper obtained by two different chronopotentiometric stripping methods, CCSA and potentiometric stripping analysis (PSA) in west Brittany coastal waters show that the measurement of copper by CCSA coupled with a gold disk electrode appears to be less affected by organic matter than those made by PSA with a mercury electrode.

A potentiometric stripping method for a simultaneous measurement of copper, lead and cadmium in oceanic waters using a TraceLab PSU22 unit has been developed. Compared to the earlier TraceLab PSU20 unit, the PSU22 has a sampling rate of the potential during the stripping step that is three times lower. This results in the decrease of the detection limit by about four times. The detection limits obtained with a pre-electrolysis time of 15 min were 44 ng l−1 (0.7 nM), 3 ng l−1 (14 pM) and 1 ng l−1 (9 pM) for Cu, Pb and Cd, respectively. The precision, accuracy and reproducibility of the method agree well with those obtained by other techniques for trace metals analysis in seawater. Vertical profiles of Cd in the Mediterranean waters obtained by AA and PSA were concordant.

The vertical distribution of cadmium, copper and lead was studied in the 0–500 m layer of the Eastern Alboran Sea. The concentrations measured compare favourably with recent data for the neighbouring area (53 pM < Cd < 124 pM; 1.4 nM< Cu < 2.8 nM, and 90 pM < Pb < 310 pM). The analysis of six profiles showed the presence of a sub-surface water mass, the northwestern Mediterranean Water (NWMW, 75-30 m), with high concentration of trace metals. Metal-salinity relationships, typified by cadmium, showed that the Atlantic surface waters (ASW) with low metal concentrations before entering the Mediterranean Sea become enriched with metals from some as yet not clearly understood source near the Straits of Gibraltar. The salinity-Cd relationship, taken together with flux rates of salt and Cd, however, shows that the Modified Atlantic water (MAW) gains its high metal concentrations during its transit in the Alboran Sea principally from a mixing with the underlying metal-rich NWMW.

Seasonal variation of copper, nickel and lead were studied in the Douarnenez Bay and the Iroise Sea during April 1983-February 1984 period. Our results compare favourably with those found in neighbouring coastal areas (1·5 nmol 1-1 <Cu <8·0 nmol l-1; 1·4 nmol l-1 <Ni <6·8 nmol l-1; and 0·07 nmol l-1 <Pb < 2·70 nmol l-1).