Air pollution caused by exhaust particulate matter (PM) from vehicular traffic is a major health issue. Increasingly strict regulations of vehicle emission have been introduced and efforts have been put on both the suppression of particulate formation inside the engine cylinders and the development of after-treatment technologies such as filters. With modern direct injected engines that produce a large number of really small sub-micron particles, the focus has increased even further and now also include a number count. The problem of calculating particle trajectories in flow ducts like vehicle exhaust systems is challenging but important to further improve the technology. The interaction between particles and oscillating flows may lead to the formation of particle groups (regions where the particle concentration is increased) yielding a possibility of realizing particle agglomeration. The oscillating flow may simply be hydrodynamic or as assumed here: the flow oscillations are created by sound propagation rather than aerodynamic approaches. An analysis is presented which gives the relationship between the speed of sound, the mean flow velocity and the amplitude of the acoustic particle velocity for particle agglomeration to be feasible. It is shown that it can be achieved if the speed of sound is reduced to be of the same order as the mean flow velocity. It is therefore suggested to use the so-called acoustic metamaterials, which can help control, direct and manipulate sound waves. At this stage a phenomenological 1D model is used for the analysis. It may seem too simplistic for direct practical engineering applications but allow the authors to build a clear picture of the effect of the sound wave and flow oscillations on particle motion and pave the way for further analysis on particle agglomeration.