Recently, there have been grown interests in improving the performance of a direct absorption solar collector (DASC). The present work reports a systematic optimization to further improve the performance of a DASC, especially for a system using plasmonic nanofluids. Plasmonic nanofluids has adaptable absorption characteristics such that its absorption coefficient can be controlled to be nearly constant with respect to the wavelength by mixing different shapes (or sizes) of metallic nanoparticles. At the same time, the magnitude of the absorption coefficient can also be changed by varying the particle concentration. Therefore, the absorption coefficient of plasmonic nanofluids should be taken into account in the optimization processes; however, it has not drawn much attention in existing studies so far. In the present work, the collector geometry (i.e., depth and length of the channel), flow characteristics (i.e., flow rate), and fluid property (i.e., absorption coefficient) are taken as design variables in the optimization process. Effects of each variable on the temperature gain, thermal efficiency, and thermal loss parameter are theoretically analyzed. In order to reduce the cost of a DASC and avoid particle agglomeration when using plasmonic nanofluids, we also explore the configuration with the lowest possible absorption coefficient but with the reasonably high temperature gain as well as efficiency. The results of this study will facilitate the development of highly efficient solar thermal collectors using plasmonic nanofluids. (C) 2017 Elsevier Ltd, All rights reserved.