A novel,simple,rapid,efficient and environment-friendly method for the determination of trace copper in cereal samples was developed by using dispersive liquid-liquid microextraction based on solidification of floating organic drop(DLLME-SFO) followed by flame atomic absorption spectrometry.In the DLLME-SFO,copper was complexed with 8-hydroxy quinoline and extracted into a small volume of 1-dodecanol,which is of low density,low toxicity and proper melting point near room temperature. The experimental parameters affecting the extraction efficiency were investigated and optimized.Under the optimum conditions, the calibration graph exhibited linearity over the range of 0.5—500 ng/mL with the correlation coefficient(r) of 0.9996.The enrichment factor was 122 and the limit of detection was 0.1 ng/mL.The method was applied to the determination of copper in the complex matrix samples such as rice and millet with the recoveries for the spiked samples at 5.0 and 10.0 u,g/g falling in the range of 92.0-98.0%and the relative standard deviation of 3.9-5.7%. A novel, simple, rapid, efficient and environment-friendly method for the determination of trace copper in cereal samples was developed by using dispersive liquid-liquid microextraction based on solidification of floating organic drop (DLLME-SFO) followed by flame atomic absorption spectrometry. In the DLLME-SFO, copper was complexed with 8-hydroxy quinoline and extracted into a small volume of 1-dodecanol, which is of low density, low toxicity and proper melting point near room temperature. The experimental parameters affecting the extraction efficiency were investigated and optimized. Un optimum conditions, the calibration graph showed linearity over the range of 0.5-500 ng / mL with the correlation coefficient (r) of 0.9996. The enrichment factor was 122 and the limit of detection was 0.1 ng / mL. method was applied to the determination of copper in the complex matrix samples such as rice and millet with the recoveries for the spiked samples at 5.0 and 10.0 u, g / g falling in the rang e of 92.0-98.0% and the relative standard deviation of 3.9-5.7%.