Computational fluid dynamic simulation of the supersonic CO2 flow during champagne cork popping

Abstract

Behind the iconic "pop!" accompanying the uncorking of a champagne bottle hides a gas flow of surprising complexity. Its modeling is made delicate by its supersonic nature, its interaction with the cork stopper, the eminently unsteady character of the flow escaping from the bottle, and the continuous change of the geometry of the computational flow domain due to the displacement of the cork. Computational Fluid Dynamics (CFD) simulations revealed the formation, evolution and dissipation of shock wave patterns during the first millisecond following champagne cork popping. A first crown-shaped shock wave pattern develops radially, then followed by the formation of a detached shock wave, or bow shock, induced by the presence of the cork in the axial path of the supersonic gas flow. Moreover, the good agreement between the position of the bow shock previously observed through high-speed imaging and that determined through CFD simulations argues in favor of the numerical method used to describe the ejection of the gas mixture expelled from the bottleneck immediately after the cork popping process.

Simulating the opening of a champagne bottle

“…In the case (A), the stopper undergoes an exit speed of around m s, a value closely resembling the experimental data found in Liger-Belair et al. (2019) and used in Benidar et al. (2022).…”

“…A negative acceleration only occurs in the first case, thereby explaining the local minimum of the green graph in figure 6(a). In the case (A), the stopper undergoes an exit speed of around 18 m s −1 , a value closely resembling the experimental data found in Liger-Belair et al ( 2019) and used in Benidar et al (2022).…”

“…As the data put forward by Liger-Belair et al. (2019) and Benidar et al. (2022) shall validate our simulation results, we only consider the two combinations of the pressure and the temperature measured by Liger-Belair et al.…”

“…We remark that the neglect of heat conduction and dissipation and the associated boundary layers renders the temperatures at the solid surfaces as just resulting from convection. Therefore, the (questionable) insulation conditions employed by Benidar et al. (2022) are irrelevant in our inviscid approach.…”

“…Special care is taken to facilitate the complete gas–stopper interaction in a fully implicit manner and the stage of stopper–bottleneck interaction initiating the release of the stopper. Therefore, our approach contrasts with the recent computational study by Benidar, Georges, Kulkarni, Cordier, and Liger-Belair (2022), complementing the aforementioned motivating experimental work: these authors employ the widely used commercial flow simulation software package Ansys Fluent to solve the fully Reynolds-averaged Navier–Stokes equations using turbulence closures but disregard that initial stage entirely. The present investigation, however, aims at resolving the coupled gas and stopper dynamics of the whole process in the greatest possible detail, where our focus is on the formation of the Mach disc.…”