Chemotaxis is an important mechanism by which microswimmers achieve directed migration in chemical concentration gradients and is a central issue in the study of the environmental responses of artificial active colloids. Here, a hybrid mesoscale simulation method combining multiparticle collision dynamics and molecular dynamics is used to systematically investigate the chemotactic behavior of self-diffusiophoretic Janus particles in a chemical concentration gradient. The solvent consists of reactant
F, product
P, and inert particles
N, while chemical reactions at the boundary walls continuously maintain the external concentration gradient. A Janus particle converts reactant
F into product
P through chemical reactions on its catalytic hemisphere, generating a nonuniform chemical field around the particle and enabling autonomous motion through self-diffusiophoresis. By tuning the types of Lennard-Jones interactions between the different fluid species and the Janus particle, self-propulsion can be achieved either along or opposite to the particle’s orientation vector. The simulations show that Janus particles in nonuniform chemical environments can exhibit either positive chemotaxis, migrating toward regions of higher concentration, or negative chemotaxis, migrating toward regions of lower concentration. Further analysis reveals that the macroscopic direction of chemotaxis is not determined solely by orientational polarization induced by the concentration gradient. Instead, it results from the competition and cooperation among several microscopic mechanisms, including orientational effects, finite particle size effects, and orientation-dependent motility. Together, these mechanisms regulate the net migration direction and chemotactic strength of Janus particles. This study elucidates the microscopic physical picture underlying the chemotactic response of self-diffusiophoretic Janus particles in nonuniform chemical environments and provides a theoretical basis for understanding the directed transport of artificial active colloids and designing micro/nanomachines with tunable environmental responses.