In order to better understand the fate of the toxic element Ag(I), sorption of Ag(I) was studied from batch experiments, at different pHs (2-8) and at 298 K. A pure quartz sand (99.999% SiO2) and "natural" quartz sand (99% SiO2, and traces of Fe, Al, Mn (hydr)oxides, of clays and of pyrite) were used as sorbents. The Ag(I) sorption behavior depends strongly on pH with isotherm shapes characteristic of Langmuir-type relationship for initial Ag concentration [Ag(I)], range between 5.0 x 10(-7) and 1.0 x 10(-3) M. Even if the Ag (I) sorption capacity on pure quartz sand is very low compared to the natural quartz sands, its affinity is rather high. From speciation calculations, several sites were proposed: at pH(i) 4,6 and 8. the first surface site is assumed to be due to iron (hydr)oxides while the second surface site is attributed to silanols. At pH(i) 2, sorption of Ag(I) was assumed to be on two surface sites of iron (hydr)oxides and a third surface site on silanol groups. Even if the sand is mainly composed of silica, the trace minerals play an important role in sorption capacity compared to silica. The conditional surface complexation constants of Ag(I) depend on pH. On the other hand, it is shown that the Ag speciation depends strongly on the history of "natural" quartz sand due to initial applied treatment, little rinsing or longer washing. In the presence of low amount of pyrite, strong complexes between Ag(I) and sulfur compounds such as thiosulfates due to oxidative dissolution of pyrite are formed what decreases Ag sorption capability. SEM-EDS analyses highlighted the surface complexation-precipitation of Ag2S and Ag(0) colloids which confirmed the important role of pyrite on Ag(I) speciation. (C) 2015 Elsevier Inc. All rights reserved.