We decided to increase fuel injection velocity at 20 m/s from previous results at 5m/s : we take a restart file at 50000 iterations i.e t=22 ms.
We note that the flow has lost a part of his symmetry on the following picture. However, the cross sections show quite symmetrical profiles, and also falls in fuel mass fraction near the wall behind the injector at x=0.1. We can make the same commentary as the previous case about recirculation areas.
Moreover, we note that there are fuel rich areas which are coming undone from the fuel jet. In fact, these areas traduce the existence of vortex. Those vortices are traduced by an increasing of U velocity on the corresponding shaded contours. Those contours also show recirculation areas behind fuel jet, where velocity is negative.
On the bottom left corner ( in front of the fuel jet), we can see a fuel rich area which are coming undone from the fuel jet with negative velocity ( it moves to the entry direction). This phenomenon is certainly due to numerical diffusion. If we don't see the see thing on the top, it's because the grid is not symmetrical.
The mean fuel mass fraction have been ploted after the last solution (at 44 ms i.e. 100000 iterations) during 30000
iterations. We can conclude the same remarks than the instantaneous fuel mass fraction contours : the flow is not
symmetrical, the outlet flow is not mixed.  Particulary, the instantaneous cross section at x=0.21 (outlet) is the same as the mean outlet cross section of fuel mass fraction.

As the previous case,  fuel flow rate is not enough important because fuel and air are not mixed in the exit.