Hostname: page-component-7b7567c8c-fvjsw Total loading time: 0 Render date: 2026-11-16T02:59:37.904Z Has data issue: false hasContentIssue false

Turbulent transfer and entrainment in a low-density jet

Published online by Cambridge University Press:  08 August 2023

P. Salizzoni*
Affiliation:
Laboratoire de Mécanique des Fluides et d'Acoustique, University of Lyon, CNRS UMR 5509 Ecole Centrale de Lyon, INSA Lyon, Université Claude Bernard, 36, avenue Guy de Collongue, 69134 Ecully, FranceDepartment of Environmental, Land, and Infrastructure Engineering, Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129 Turin, Italy
S. Vaux
Affiliation:
Institut de Radioprotection et de Sûreté Nucléaire (IRSN), PSN-RES, SA2I, LIE, Cadarache, 13115 St Paul-Lez-Durance, France
M. Creyssels
Affiliation:
Laboratoire de Mécanique des Fluides et d'Acoustique, University of Lyon, CNRS UMR 5509 Ecole Centrale de Lyon, INSA Lyon, Université Claude Bernard, 36, avenue Guy de Collongue, 69134 Ecully, France
M. Amielh
Affiliation:
Aix Marseille University, CNRS, Centrale Marseille, IRPHE UMR 7342, 49, rue Frédéric Joliot-Curie, BP 146, 13384 Marseille Cedex 13, France
L. Pietri
Affiliation:
Aix Marseille University, CNRS, Centrale Marseille, IRPHE UMR 7342, 49, rue Frédéric Joliot-Curie, BP 146, 13384 Marseille Cedex 13, France
F. Anselmet
Affiliation:
Aix Marseille University, CNRS, Centrale Marseille, IRPHE UMR 7342, 49, rue Frédéric Joliot-Curie, BP 146, 13384 Marseille Cedex 13, France
*
†Email address for correspondence: pietro.salizzoni@ec-lyon.fr

Abstract

We investigate the dynamics of a low-density round jet, with a focus on the mechanisms governing the turbulent momentum and mass transfers as well as on the entrainment of ambient fluid. To that purpose, we combine a theoretical analysis, laboratory experiments and numerical simulations. The theoretical analysis relies on a general formulation of the entrainment decomposition for the case of large density differences, revealing the role of the processes contributing to the entrainment: turbulent kinetic energy production and variation in the shape of the mean velocity radial profiles. The spatial evolution of these terms has been evaluated by means of challenging experiments, providing a unique data set of combined velocity and density statistics of a low-density jet and an air jet. The same flows are investigated by means of large-eddy simulation (LES). Other than for providing complementary information on flow statistics, LES is here used to investigate the role of varying conditions imposed at the source, notably concerning the shape of the inlet velocity profile and the presence of a bottom wall surrounding the source. Experimental and numerical results provide clear insight on how a reduced density within the jet enhances the turbulent kinetic energy production (compared to an iso-density jet) and modifies the shape of the mean velocity profiles. Despite its clear influence on the flow statistics, the reduced density has overall little influence on the entrainment rate, which also shows little sensitivity to varying source conditions.

Information

Type
JFM Papers
Copyright
© The Author(s), 2023. Published by Cambridge University Press
Figure 0

Figure 1.Schematic details of the boundary conditions at the source used for the numerical simulations (a) J1a–c (i.e. no bottom wall and pipe flow shaped inlet mean velocity profile) and (b) J3 (bottom wall and top-hat inlet mean velocity profile). Also shown for each case are the development of the jet and mean velocity profile for two distances from the source. (c) Instantaneous snapshot of the air-helium mixture mass fraction provided by the LES results (white, $C=0$; green, $C=1$). (d) Image of the jet obtained by planar laser-induced fluorescence ($22 \times 22$ mm$^2$ field of view).

Figure 1

Table 1.Experimental conditions. The main results concern the light jet (figures 3–5 and 6–11) and they are compared to air jet results in figure 11.

Figure 2

Figure 2.Sketch of the experimental set-up (Moutte 2018). The main elements are as follows: (0) camera for PIV, (1) $200$ mm lens, (2) pulsed laser, (3) camera for PLIF, (4) $532$ nm optical filter, (5) light intensifier, (6) dichroic mirror, (7) optical table, (8) containment box, and (9) jet maker.

Figure 3

Table 2.Boundary conditions, subgrid models and density ratio at the source adopted in the numerical simulations.

Figure 4

Figure 3.Decay of (a) the mean streamwise velocity and (b) the mean helium concentration along the jet axis. (c) Increase of the mean streamwise velocity half-width ($r_w$) and the mean concentration half-width ($r_C$).  Axial evolution of the (d) longitudinal and (e) radial velocity fluctuations, and (f) of the concentration fluctuations. Black circles, experiments; black dotted lines, reference simulation (J1a); red dashed lines, simulation without bottom wall and top-hat profile for the inlet velocity (J2); blue dash–dotted lines, simulation with bottom wall and using top-hat profile for the inlet velocity (J3).

Figure 5

Figure 4.(a–c) Radial evolution of the mean streamwise velocity, (d–f) the mean concentration and (g–i) the mean radial velocity. The distance to the source of the jet is fixed to (a,d,g) $z/r_s = 2$, (b,e,h) $z/r_s = 16$ and (c,f,i) $z/r_s = 36$. Same symbols and lines as in figure 3.

Figure 6

Figure 5.Radial evolution of the (a–c) longitudinal and (d–f) radial velocity fluctuations, (g–i) concentration fluctuations, ( j–l) Reynolds stress, (m–o) radial velocity-concentration correlation and (p–r) turbulent Schmidt number. The distance to the source of the jet is fixed to (a,d,g,j,m,p) $z/r_s = 2$, (b,e,h,k,n,q) $z/r_s = 16$ and (c,f,i,l,o,r) $z/r_s = 36$. Same symbols and lines as in figure 3.

Figure 7

Figure 6.Influence of the subgrid model used on the numerical results. Black dotted lines, J1a (Vreman); brown plusses, J1b (WALE); purple stars, J1c (dynamic Smagorinsky).

Figure 8

Figure 7.Effect of the Reynolds or Favre averaging on the radial evolution of the (a–c) mean streamwise velocity and (d–f) Reynolds stress. The distance to the source of the jet is fixed to (a,d) $z/r_s = 2$, (b,e) $z/r_s = 16$ and (c,f) $z/r_s = 36$. Symbols, experiments; lines, reference simulation (J1a); black circles and dotted lines, Reynolds averaging; red dots and lines, Favre averaging.

Figure 9

Figure 8.Evolution (a) of the momentum, volume and mass fluxes, (b) of the mean characteristic velocity and radius and (c) of the density ratio, as a function of the distance to the source of the jet. Black circles, experiments; black dotted lines, reference simulation (J1a); red dashed lines, J2; blue dash–dotted lines, J3.

Figure 10

Figure 9.(a) Axial evolution of the entrainment coefficient $\alpha$ for simulations (black dotted lines, J1a; red dashed lines, J2; blue dash–dotted lines, J3) and experiment (symbols). In panels (b) and (c), $\bar {\alpha }$ defined by (5.4) and $\bar {\alpha }_{ap}$ defined by (5.5) are also plotted against $z/r_s$ for simulation J1a and experiment.

Figure 11

Figure 10.Variations of the mean profile coefficients (a) $\beta _f$, (b) $\gamma _m$ and $\gamma _f$, (c) $-\delta _m$, and the entrainment coefficients (d) $\alpha _{prod}^m$ and (e) $\alpha$ as a function of the distance to the source of the light jet. Circles, experiments; lines, simulation J1a. Also shown (in grey, only for simulations): (a) $-\beta _p$; (b) $\gamma _g$ and $-\gamma _p$; (c) $-\delta _g$, $\delta _f$ and $-\delta _p$; (d) $\alpha _{prod}$ and (e) $\alpha _{E}$ (see (5.6)).

Figure 12

Figure 11.Longitudinal profile of the mean profile coefficients (a) $\gamma _m$, (b) $-\delta _m$, (c) $\alpha ^{m}_{shape}$ and (d) $\alpha _{prod}^m$. Circles, experiments; black dotted lines, reference simulation J1a; red dashed lines, J2 (without bottom wall and using top-hat profile for the inlet velocity); blue dash–dotted lines, J3 (with bottom wall and using top-hat profile for the inlet velocity).

Figure 13

Figure 12.Longitudinal profile of the (a) non-dimensional mass flux, (b) mean characteristic radius, (c) entrainment coefficient estimated by (2.12), (d) mean profile coefficients $\gamma _m$ and (e) $-\delta _m$, and (f) ${\alpha ^{m}_{prod}}$ (see (5.6)), for the low-density jet (J1a, black dotted line and circles) and the iso-density jet (J0, green dashed line and crosses). Also shown is the DNS of van Reeuwijk et al. (2016) for an iso-density jet (green solid line).