Agathe Laes

Copper is a critical trace element in marine environments, acting as an essential limiting micronutrient in open ocean waters while posing toxicity risks at elevated concentrations in industrialized coastal zones. Accurate assessment of these contrasting roles necessitates rapid, selective, and precise analytical methods capable of in situ deployment. In this study, we adapted a colorimetric reverse Flow Injection Analysis (rFIA) method with Liquid Core Waveguide (LCW) detection for use on an automated chemical analyzer. By optimizing the protocol established by (Wang et al., 2022) to address seawater-specific challenges—including temperature variations, salinity effects, and interferences from metals and sulfides—we achieved a subnanomolar detection limit (LOD = 0.03 nmol.L⁻¹ using a 1 m LCW). Method accuracy was validated through the analysis of certified reference materials NASS-7 and CASS-6. A critical operational constraint identified was the requirement to maintain a H₂S/Cu concentration ratio below 280 to prevent negative interference with the copper signal in sulfide-rich environments. The optimized method was successfully applied to diverse samples collected along a land-sea continuum—from the Aulne River and estuary, through Brest Bay to the Iroise Sea (France)—as well as to discrete depth profiles from the Black Sea and the Sea of Marmara (Türkiye). These results validate a robust, reproducible, and highly sensitive approach for quantifying copper across riverine, estuarine, and marine systems with widely varying physicochemical conditions.

Copper is a critical trace element in marine environments, acting as an essential limiting micronutrient in open ocean waters while posing toxicity risks at elevated concentrations in industrialized coastal zones. Accurate assessment of these contrasting roles necessitates rapid, selective, and precise analytical methods capable of in situ deployment. In this study, we adapted a colorimetric reverse Flow Injection Analysis (rFIA) method with Liquid Core Waveguide (LCW) detection for use on an automated chemical analyzer. By optimizing the protocol established by (Wang et al., 2022) to address seawater-specific challenges—including temperature variations, salinity effects, and interferences from metals and sulfides—we achieved a subnanomolar detection limit (LOD = 0.03 nmol.L⁻¹ using a 1 m LCW). Method accuracy was validated through the analysis of certified reference materials NASS-7 and CASS-6. A critical operational constraint identified was the requirement to maintain a H₂S/Cu concentration ratio below 280 to prevent negative interference with the copper signal in sulfide-rich environments. The optimized method was successfully applied to diverse samples collected along a land-sea continuum—from the Aulne River and estuary, through Brest Bay to the Iroise Sea (France)—as well as to discrete depth profiles from the Black Sea and the Sea of Marmara (Türkiye). These results validate a robust, reproducible, and highly sensitive approach for quantifying copper across riverine, estuarine, and marine systems with widely varying physicochemical conditions.

Copper is a trace element that has been the subject of numerous studies, as it plays an important role as a limiting micronutrient in certain areas, particularly in the open ocean. It can also a toxic substance when present in high concentrations such as industrialized coastal areas. It is now essential to determine its concentration using a rapid, selective, accurate, and precise method that can potentially be implemented in-situ. In this study, we adapted copper quantification for an automated chemical analyzer using a colorimetric method based on reverse Flow Injection Analysis (rFIA) with Liquid Core Waveguide (LCW) detection. By optimizing the method described by (Wang et al., 2022) for the seawater matrix (temperature, salinity effect, interference from metals and sulfides), we achieved high-performance detection limit in the subnanomolar range (LOD = 0.03 nmol.l-1, LCW 1 m), verified by the analysis of certified reference materials NASS-7 and CASS-6. However, care must be taken when analyzing seawater in the presence of sulfide: The H₂S/Cu concentration ratio must not exceed 280 to avoid any negative interference with the copper response signal. This method was applied to a diverse set of samples, initially collected along the land-sea continuum—from the Aulne River, its estuary, Brest Bay, and the Iroise Sea (France)—and then samples from the discrete depths from the Black Sea and the Sea of Marmara (Türkiye). Overall, this study enabled the validation of a rapid, reproducible, and highly accurate method for measuring copper in river, estuarine, and marine waters across contrasted study areas.

We present an innovative electrochemical sensing method for the detection of heavy metal cations that departs from conventional pulsed electroanalytical techniques. The keystone of this approach is the nanostructuring of trace amounts of preconcentrated metal on a functionalized electrode, thereby exploiting the heterogeneous catalytic properties of the resulting metal nanodeposits. Rather than introducing a new selective sensing interface, this work focuses on a general detection strategy. In this approach, a diazonium modified graphite electrode is used as a robust platform to demonstrate that sparsely pre accumulated metal cations can be reduced into nanodeposits, which behave as a nanoelectrode array. The macroscopic electrochemical response of this array towards the hydrogen evolution reaction depends on the size of the nanoelectrodes: higher metal concentrations in the incubation solution lead to more extensive metal deposition and larger nanostructures, which shift the hydrogen evolution reaction catalytic peak potential positively in cyclic voltammetry. This shift continues until the metal becomes detectable by standard pulsed analytical methods. As an illustration, the proposed approach enables the detection of copper and iron with improved sensitivity compared to the corresponding reference pulsed techniques. Moreover, this gain in sensitivity does not rely on a specific surface chemistry. Therefore, the more affine a given accumulation layer is for its target metal, the lower the detection limits that can be achieved with this method. This study points toward a promising route for next generation electrochemical sensors for trace metal detection, offering sensitivity beyond that of traditional faradaic detection schemes.

In this study, we adapted copper quantification to the CHEMINI instrumentation [1] using the colorimetric method based on reverse flow analysis (rFIA) with liquid core wavelength (GOF) detection, as recently described [2]. This method relies on the reduction of ferricyanide by L-glutathione in acidic media, with copper acting as the reaction catalyst. Various parameters (flow rates, reaction, injection loop and GOF length) were optimized using chemometric experimental designs. By adapting the method to seawater (salinity effect, metal and organic matter interferences), we achieved high-performance analytical limits in the subnanomolar range (detection limit = 0.03 nmol.L-1, GOF 1 m). The first application of the method was conducted along the land-sea continuum from the Aulne estuary towards the Iroise Sea, measuring both soluble and dissolved copper concentrations (FeLINE project, [3]). The average copper concentrations obtained were comparable (5-20 nmol.L-1) to those reported earlier in this system [4]. Concentrations decreased almost conservatively from the Aulne freshwaters (22.0 nM at S=0) towards the Bay of Brest (8.1 nM at S=34). and slightly increased toward the Iroise Sea (11.2 nM at S=35.2). The soluble copper fraction represented between 72% and 100% of the dissolved copper fraction, indicating a low contribution of high molecular weight colloidal Cu species (0.02-0.45 µm) in the Aulne estuary. Figure : Evolution of dissolved copper DCu (

The international BioGeoChemical-Argo program recommends the measurement of 6 essential ocean variables (EOVs), one of which being the inorganic carbon. Estimates of pH variation in the open ocean using traditional sampling methods show a decrease of 0.002 units per year [1]. Understanding and predicting pH variations using precise, certified in situ detection over oceans around the world over the long term is currently essential. In the frame of the PIANO (Plan d’Investissement Argo Nouvelles Observations, Ifremer) project we designed and developed a new submersible pHT colorimetric sensor deployable on profiling floats sensible enough to be suitable for the acidification monitoring. This development is aimed on the use of the referenced colorimetric method ([2]–[4]), applied to microfluidic channels with in situ calibration using on-board standards. The first phase of the project was to determine the miniaturized actuators able of working under 350 bars and 4°C. Then, a 3D printed manifold was fabricated using stereolithography. It was optimized to integrate a micromixing area, an optical detection and a temperature sensor (pt100). Once the final architecture of the manifold was fixed, the system's repeatability and accuracy were then tested. Electronic and mechanic design were established considering minimum energy consumption and space. In parallel, tests were set up with the metrology laboratory to study the ageing of on-board pHT standards over time (2-years test). An overall protocol was devised, focusing on the procedure for rinsing the bags to minimize bacterial contamination and purchasing stable standards. The first results of this future submersible pHT sensor against robust benchtop spectroscopic measurements of referenced waters will be discussed.Figure 1: 3D printing of the optical cellFigure 2: 3D printing mixing unit (based on tesla design) [1] F. J. Millero, “The marine inorganic carbon cycle.,” Chem. Rev., vol. 107, no. Table 1, pp. 308–341, 2007.[2] X. Liu, M. C. Patsavas, and R. H. Byrne, “Purification and characterization of meta-cresol purple for spectrophotometric seawater ph measurements,” Environ. Sci. Technol., vol. 45, no. 11, pp. 4862–4868, 2011.[3] J. D. Müller et al., “Metrology for pH measurements in brackish waters-part 1: Extending electrochemical pHT measurements of TRIS buffers to salinities 5-20,” Front. Mar. Sci., vol. 5, no. JUL, pp. 1–12, 2018.[4] J. D. Müller and G. Rehder, “Metrology of pH measurements in brackish waters-part 2: Experimental characterization of purified meta-cresol purple for spectrophotometric pHT measurements,” Front. Mar. Sci., vol. 5, no. JUL, pp. 1–9, 2018.

Understanding and predicting pH distributions through accurate and certified in situ detection is currently crucial. As part of the European BRIDGE Black Sea project, surface pHT transects were carried out in July 2024 across the Marmara Sea, the Bosphorus Strait and the southern Black Sea. This study compares high-resolution temporal pHT measurements from 3 sensors deployed in parallel: SAMI-pHTM (Sunburst Sensors, LLC, colorimetric sensor [1]), SeaFET™ (Sea-Bird Scientific, ion selective field effect transistor (ISFET) [2]) and pHNX (Ifremer Homemade colorimetric sensor). The sensors continuously monitored pHT distributions of surface seawater passing through a SeaBird Scientific SBE45 thermosalinograph. Data collected by SAMI-pHTM (n=319), SeaFET™(n=8424) and pHNX (n=488) were corrected from salinity and temperature (data provided by SBE45 thermosalinograph). Additionally, the SeaFETTM sensor underwent evaluation and correction against robust benchtop measurements, with a one-point correction from the pHNX sensor, ensuring full traceability to international system unit [3]. The three instruments recorded the highest pHT values (8.2225 for pHNX) at the exit of the Bosphorus Strait, extending in a north-easterly direction into the Black Sea. The lowest surface values were obtained in the northern part of the transect (8.1360 for pHNX, 41°33'04.7"N 29°39'01.1"E) and in the Marmara Sea (8.0623 for pHNX, 40°48'49.3"N 28°53'58.6"E). These findings are in line with a previous study carried out in the same area in June 2023 (data not published yet), and the general trend towards the entry of a pH-enriched flow into a semi-enclosed sea has been observed in other parts of the world [3].Figure 1: pHT sensors and sample inletFigure 2: Distribution of pHT from pHNX[1] A. M. Nightingale, A. D. Beaton, and M. C. Mowlem, “Trends in microfluidic systems for in situ chemical analysis of natural waters,” Sensors Actuators B Chem., vol. 221, pp. 1398–1405, 2015.[2] T. R. Martz, J. G. Connery, and K. S. Johnson, “Testing the Honeywell Durafet® for seawater pH applications,” Limnol. Oceanogr. Methods, vol. 8, no. MAY, pp. 172–184, 2010.[3] X. Liu, Z. A. Wang, R. H. Byrne, E. A. Kaltenbacher, and R. E. Bernstein, “Spectrophotometric measurements of pH in-situ: Laboratory and field evaluations of instrumental performance,” Environ. Sci. Technol., vol. 40, no. 16, pp. 5036–5044, 2006.

The recently discovered deep-sea Capelinhos hydrothermal edifice, ~ 1.5 km of the main Lucky Strike (LS) vent field (northern Mid-Atlantic Ridge), contrasts with the other LS edifices in having poorly-altered end-member hydrothermal fluids with low pH and chlorine, and high metal concentrations. Capelinhos unique chemistry and location offer the opportunity to test the effects of local abiotic filters on faunal community structure while avoiding the often-correlated influence of dispersal limitation and depth. In this paper, we characterize for the first time the distribution patterns of the Capelinhos faunal communities, and analyze the benthic invertebrates (> 250 µm) inhabiting diffusive-flow areas and their trophic structures (δ¹³C, δ¹⁵N and δ³⁴S). We hypothesized that faunal communities would differ from those of the nearest LS vent edifices, showing an impoverished species subset due to the potential toxicity of the chemical environment. Conversely, our results show that: (1) community distribution resembles that of other LS edifices, with assemblages visually dominated by shrimps (close to high-temperature focused-fluid areas) and mussels (at low-temperature diffuse flow areas); (2) most species from diffuse flow areas are well-known LS inhabitants, including the bed-forming and chemosymbiotic mussel Bathymodiolus azoricus and (3) communities are as diverse as those of the most diverse LS edifices. On the contrary, stable isotopes suggest different trophodynamics at Capelinhos. The high δ¹⁵N and, especially, δ¹³C and δ³⁴S values suggest an important role of methane oxidation (i.e., methanotrophy), rather than the sulfide oxidation (i.e., thiotrophy) that predominates at most LS edifices. Our results indicate that Capelinhos shows unique environmental conditions, trophic structure and trophodynamics, yet similar fauna, compared to other LS edifices, which suggest a great environmental and trophic plasticity of the vent faunal communities at the LS.

The aim of FeLINE project (Fer Ligands In the aulNe Estuary) was to determine the distribution of iron and associated ligands concentrations along the land sea continuum of the Iroise Sea (Bay of Brest, France). Iron porphyrin like ligands (Fe-Py) such as heme and hemoproteins are relevant complexes in iron biogeochemical cycling as they can persist in seawater and on marine particulates. This work reveals for the first time the distribution of Fe-Py concentrations (dissolved plus reactive particulate) along a temperate macrotidal estuary. Unfiltered samples were collected in October 2019 across a transect of the Aulne river and estuary / Rade of Brest / Iroise Sea during low tidal coefficient (39). Fe-Py concentrations were determined using flow injection analysis with chemiluminescence detection adapted from Vong et al. (2007). Various interferences (organic, metallic, pH and salinity) were tested. The detection limit attained was 11 pmol.l-1 and the time of analysis 1min30s per sample. The Fe-Py concentrations varied from 0.007 ±0.002 nmol.l-1 for S=33.98 and 1.177 ±0.007 nmol.l-1 for S = 0.92. The Fe-Py concentrations clearly showed a non-conservative behavior due to various processes other than simple mixing of natural and seawater. The highest values revealing a Fe-Py enrichment were observed in the Estuarine Turbidity Maximum (ETM) for which concentrations varied between 1.177 ±0.007, S = 5.2 and 0.738 ±0.004 nmol.l-1 S = 8.59. This positive anomaly of Fe-Py concentrations (40%) also corresponded to the lowest pH values (pH =7.27-7.32). The distal part of the transect displayed a negative anomaly for salinities comprised between 15 and 25 (loss of 37%). The four last points geographically corresponding to the Bay of Brest (S>35) exhibited low and stable Fe-Py concentrations of 0.007±0.002 and 0.024 ± 0.003 nmol.l-1. The supply and removal fluxes were respectively estimated at 2.4±0.2g/d and 8.1 ± 0.8g/d, revealing an average Fe-Py removal of 39.8% that is probably due to particle flocculation.