We use molecular dynamics simulations to investigate the thermal conductivity of argon-based nanofluid with copper nanoparticles through the Green-Kubo formalism. To describe the interaction between argon-argon atoms, we used the well-known Lennard-Jones (L-J) potential, while the copper-copper interactions are modeled using the embedded atom method (EAM) potential that takes the metallic bonding into account. The thermal conductivity of the pure argon liquid obtained in the present simulation agreed with available experimental results. In the case of nanofluid, our simulation predicted thermal conductivity values larger than those found by the existing analytical models, but in a good accordance with experimental results. This implies that our simulation is more adequate, to describe the thermal conductivity of nanofluids than the previous analytical models. The efficiency of nanofluids is improved and the thermal conductivity enhancement is appeared when the particle size and temperature increase. (C) 2018 The Society of Powder Technology Japan. Published by Elsevier B.V. and The Society of Powder Technology Japan. All rights reserved.