The properties of nanocomposite materials depend on the dispersion of the nanoparticles/nanofibers within the matrix. The addition of surfactants and varied processing techniques are used to increase the dispersion of the nanoparticles in the final composite. A method for the quantitative prediction of the interactions between nanoparticles in solution would aid in the design of processing schedules. In this study, molecular dynamics simulations are used to compute for the potential of mean force as a function of the distance and orientation between a pair of single-walled carbon nanotubes (CNTs) in water. An adaptive biasing force method is used to speed up the calculations. Simulation results show that CNT orientation and the addition of surfactant can significantly affect CNT interactions and inturn dispersion.