a b s t r a c tThe electrochemical determination of Hg(II) at trace level using gold nanoparticles-modified glassy carbon (AuNPs-GC) electrodes is described. Starting from HAuCl 4 in NaNO 3 , gold nanoparticles (AuNPs) were deposited onto Glassy Carbon (GC) electrodes using Cyclic Voltammetry (CV). Different deposits were obtained by varying the global charge consumed during the whole electroreduction step, depending on the number of cyclic potential scans (N). AuNPs were characterized as a function of the charge using both CV in H 2 SO 4 and Field Emission Gun Scanning Electron Microscopy (FEG-SEM). The AuNPs-GC electrodes were then applied to determine low Hg(II) concentrations using Square Wave Anodic Stripping Voltammetry (SWASV). The AuNPs-GC electrodes provided significantly improved performances in Hg(II) determination compared to unmodified GC and bare Au electrodes. It was shown that the physico-chemical properties of the deposits are correlated to the performances of the AuNPs-GC electrode with respect to Hg(II) assay. The best results were obtained for four electrodeposition cyclic scans, where small-sized particles (36 ± 13 nm) with high density (73 particles lm À2 ) were obtained. Under these conditions, a linearity range from 0.64 to 4.00 nM and a limit of detection of 0.42 nM were obtained.

Gold nanoparticles (AuNPs) were deposited on Glassy Carbon (GC) substrate by using three electrochemical techniques: Cyclic Voltammetry (CV), Chronoamperometry (CA) and Potentiostatic Double-Pulse (PDP). For each electrodeposition method, the resulting AuNPs-modified electrodes were characterized by CV in H 2 SO 4 and Field Emission Gun Scanning Electron Microscopy (FEG-SEM). CA was found to be the best electrodeposition mode for controlling the morphology and the density of AuNPs. The modified electrodes were used for low Hg(II) concentration detection using Square Wave Anodic Stripping Voltammetry (SWASV). AuNPs obtained by CA afforded the best amperometric response while involving the lowest amount of charge during the electrodeposition step (Q Au(III)). This analytical response is correlated to both the smallest particle size (ca. 17 nm in diameter) and the highest particle density (332 particles m −2), thus displaying high electrode effective surface area. In these optimal conditions, using a Hg(II) preconcentration time of 300 s, the nanosensor array exhibited a linearity range from 0.80 to 9.9 nM with a sensitivity of 1.16 A nM −1. A detection limit of 0.40 nM (s/n = 3) was reached.

Gold nanoparticles (AuNPs) were deposited on a glassy carbon (GC) substrate by constant potential electrolysis and characterized by cyclic voltammetry in H2SO4 and field emission gun scanning electron microscopy (FEG-SEM). The modified AuNPs-GC electrode was used for low Hg(II) concentration detection using a Square Wave Anodic Stripping Voltammetry (SWASV) procedure which included a chloride desorption step. The comparison of the obtained results with our previous work in which no desorption step was used showed that this latter step significantly improved the analytical performances, providing a three time higher sensitivity and a limit of detection of 80pM for 300s preconcentration, as well as a lower average standard deviation. The influence of chloride concentration on the AuNPs-GC electrode response to Hg(II) trace amounts was also studied and its optimal value confirmed to be in the 10(-2)M range. Finally, the AuNPs-GC electrode was used for the determination of Hg(II) in a natural groundwater sample from south of France. By using a preconcentration time of 3000s, a Hg(II) concentration of 19±3pM was found, which compared well with the result obtained by cold vapor atomic fluorescence spectroscopy (22±2pM).