The efficient removal of arsenite [As(III)] from groundwater remains a great challenge. Nanoscale oxides of Fe(III), Zr(IV), and Al(III) can selectively remove arsenic from groundwater through inner-sphere complexation. However, owing to polysilicate coatings formation on nanoparticles surface, the ubiquitous silicate exerts remarkably adverse effects on As(III) removal. Herein, we propose a new strategy to enhance silicate resistance of nanoscale oxides by embedding them inside the redox polymer host. As a proof-ofconcept, the nanocomposite HFO@PS-Cl was employed to remove As(III) from silicate-containing water. The polymer host (PS-Cl) contains active chlorine to oxidize As(III) into arsenate [As(V)], and the embedded Fe(III) oxides enabling specific adsorption toward arsenic. Silicate exerts negligible effects on As(III) removal by HFO@PS-Cl in pH 3-7, but increasing the residual arsenic concentration from 49 mu g/L to 166 mu g/L for the solutions treated by HFO@PS-N, i.e., the nanoscale Fe(III) oxides embedded inside the polymer host without active chlorine. During the six cyclic decontamination-regeneration assays, HFO@PS-Cl steadily reduces As(III) below 10 mu g/L. As for HFO@PS-N, however, the residual arsenic increases to 57 mu g/L in the sixth run. In column mode, HFO@PS-Cl column generates >3200-bed volume (BV) clean water ([As]<10 mu g/L) from the simulated As(III)-contaminated groundwater. In contrast, the values for As(V)-contaminated water and HFO@PS-N column are only similar to 650 BV and similar to 608 BV, respectively. The stoichiometric assays, XPS, and in-situ ATR-FTIR analysis demonstrate that silicate polymerization is intensively suppressed by the protons produced during As(III) oxidation, thus rendering HFO@PS-Cl with excellent silicate resistant properties. (C) 2020 Elsevier Ltd. All rights reserved.