An Escherichia coli cadAp::luc-based whole-cell sensor was constructed to measure cadmium (Cd) bioavailability and assess the immobilizing efficiency of phosphate and silicate on Cd. In previous induction experiments, a linear response (R(2) = 0.97, P < 0.01) from 0.1 to 5 μmol L(-1) of Cd was detected by this sensor after a 2 h incubation. The sensor was then used to estimate Cd bioavailability in soils spiked with different amounts of dipotassium phosphate (DKP, K₂HPO₄) or sodium silicate (SS, Na₂SiO₃·9H₂O). The total Cd in soil-water extracts (TSWE) was determined with ICP-MS, and the bioavailable Cd in soil-water extracts (BSWE) and bioavailable Cd in soil-water suspensions (BSWS) were measured by the E. coli cadAp::luc-based whole-cell sensor. Final results showed that spiked SS (Si : Cd = 2 : 1, mol mol(-1)) reduced the different forms of Cd (TSWE, BSWE and BSWS) from 56.47 mg kg(-1), 42.11 mg kg(-1), and 206.72 mg kg(-1) to 16.63 mg kg(-1), 15.90 mg kg(-1), and 67.57 mg kg(-1), respectively. In other words, SS had 25.68%, 19.5%, and 9.54% better immobilizing efficiency, respectively, compared with DKP. All the results supported SS was more efficient than DKP at immobilizing Cd in soil, and higher soil pH and higher solubility of the immobilizing agents may have been the major factor affecting immobilizing efficiency. In addition, the total and bioavailable Cd in soil-water extracts was only 16.13-35.41% of the sensor contact assay-determined Cd (BSWS), which indicated that the whole-cell sensor-based contact assay was more practical in assessing the risk of Cd in soil after immobilization since it would not overrate the immobilizing capacity of the agents.