ABSTRACT
Theoretical and experimental analyses are the major approaches for scientific studies in most research areas. One objective of a theoretical analysis is to find the solutions of certain governing equations, such as the Navier–Stokes and Poisson–Boltzmann equations. For simple fluid flows (e.g., Poiseuille’s flow), the analytical solution of the Navier–Stokes equation can be obtained. However, in many cases, flows are rather complex and it is difficult to theoretically solve the governing equations. This leads to the development of various numerical methods, which employ different strategies in interpreting and solving the governing equations. Numerical methods or simulations usually involve intensive calculations and can only be performed using computers. Describing fluid flows through numerically solving the governing equations is an important approach and has been developed as a branch of fluid mechanics, which is called computational fluid dynamics (CFD). As most of CFD methods numerically solve the Navier–Stokes equation, they can accurately describe macroscale flows and have been widely used for various flow analyses. For nanoscale flows, as discussed in Chapter 2, the Navier–Stokes equation may become invalid and the traditional CFD method is not regarded as a quantitatively reliable approach for catching nanoscale flow physics. The most popular numerical method for nanofluidics is molecular dynamics (MD) simulation, which considers multi-body interactions of fluid molecules and solid surface atoms. Some other numerical methods, such as Monte Carlo simulation, can also be used for nanoscale flow analyses, especially for gas flows. In this chapter, only the method of MD simulation is introduced.
