I am concerned looking for an exact Riemann solver for compressible 1D multi-component Euler equations, supplied with the ideal gases equation of state and under the assumptions:

  • Mechanical equilibrium
  • Thermal equilibrium

For example for the form used by Gouasmi et al. 2020 [1]. In a question [2] of this forum, it is very well explained. For me, it seems that he/she is using the Riemann solver for single-component model in the multi-component model. If doing this, I do not see how to obtain the partial densities/mass fractions in the intermediate state. Borisov & Rikov [3] provide the exact Riemann solver for a different equation of state, and Declercq et al. [4] provide it for two-phase model. However, despite used in the literature, I don't find an exact Riemann solver for this model.

If anyone has an idea, I would be very grateful,

Have a nice day,

References: [1] https://www.sciencedirect.com/science/article/pii/S0045782520300955 [2] Exact Riemann solver for perfect gas mixture: problem with Newton's method convergence [3] https://www.researchgate.net/publication/325098653_An_exact_Riemann_solver_in_the_algorithms_for_multicomponent_gas_dynamics [4] https://hal.science/hal-01580049/document


1 Answer 1


I finally found the solution. In the reference [3] the procedure is very detailed, it was just I did not know that mass fractions (Y) do not change across shocks or rarefactions. They just change across the contact discontinuity, between states 3 and 4 in the figure

Spatio-temporal diagram of the possible wave patterns in the Riemann problems for multi-component Euler model, extracted from [3].

The only complication I found is that if the objective intermediate pressure (the one arising from the solver) is zero, a relaxation coefficient must be applied to the Newton-Raphson loop.

Please notice two possible mistakes in the reference

  1. the velociy of the left rarefaction should be $D^{l,*}=U-c^{l,*}$ in case 1 page 14
  2. in $P=p_l$ or $P=p_r$, the formulas for calculating the shock velocity blow up (cases 2 and 4 page 14), so it is better to put the case $P=p_l$ and $P=p_r$ in cases 1 and 3 respectively.

For any question do not hesitate to contact me.

  • $\begingroup$ My apologizes for the mistake, in the third paragraph of this answer I was meaning NEARLY ZERO intermediate pressure (understood as a very low value in comparison to the left and right states, like a recieding flow problem) $\endgroup$
    – ReroDani
    Commented Jul 9 at 12:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.