arXiv subjects

Roberto Verzicco

Publications and source records attributed to Roberto Verzicco.

At least 127 records · Page 7Linked to original sources

arXiv2015

Logarithmic mean temperature profiles and their connection to plume emissions in turbulent Rayleigh-Bénard convection

Two-dimensional simulations of Rayleigh-Bénard convection at $Ra = 5\times10^{10}$ show that vertical logarithmic mean temperature profiles can be observed in regions of the boundary layer where thermal plumes are emitted. The profile is logarithmic only in these regions and not in the rest of the boundary layer where it is sheared by the large scale wind and impacted by plumes. In addition, the logarithmic behavior is not visible in the horizontal average. The findings reveal that the temperature profiles are strongly connected to thermal plume emission and support a perception that parts of the boundary layer can be turbulent, while others are not. The transition to the ultimate regime, in which the boundary layers are considered to be fully turbulent, can therefore be understood as a gradual increases in fraction of the plume-emitting ('turbulent') regions of the boundary layer.

physics.flu-dyn↗

arXiv2015

Deformation statistics of sub-Kolmogorov-scale ellipsoidal neutrally buoyant drops in isotropic turbulence

Small droplets in turbulent flows can undergo highly variable deformations and orientational dynamics. For neutrally buoyant droplets smaller than the Kolmogorov scale, the dominant effects from the surrounding turbulent flow arise through Lagrangian time histories of the velocity gradient tensor. Here we study the evolution of representative droplets using a model that includes rotation and stretching effects from the surrounding fluid, and restoration effects from surface tension including a constant droplet volume constraint, while assuming that the droplets maintain an ellipsoidal shape. The model is combined with Lagrangian time histories of the velocity gradient tensor extracted from DNS of turbulence to obtain simulated droplet evolutions. These are used to characterize the size, shape and orientation statistics of small droplets in turbulence. A critical capillary number, $Ca_c$ is identified associated with unbounded growth of one or two of the droplet's semi-axes. Exploiting analogies with dynamics of polymers in turbulence, the $Ca_c$ number can be predicted based on the large deviation theory for the largest Finite Time Lyapunov exponent. Also, for sub-critical $Ca$ the theory enables predictions of the slope of the power-law tails of droplet size distributions in turbulence. For cases when the viscosities of droplet and outer fluid differ in a way that enables vorticity to decorrelate the shape from the straining directions, the large deviation formalism based on the stretching properties of the velocity gradient tensor loses validity and its predictions fail. Even considering the limitations of the assumed ellipsoidal droplet shape, the results highlight the complex coupling between droplet deformation, orientation and the local fluid velocity gradient tensor to be expected when small viscous drops interact with turbulent flows.

physics.flu-dyn↗

arXiv2015

A multiple--resolution strategy for Direct Numerical Simulation of scalar turbulence

In this paper a numerical procedure to simulate low diffusivity scalar turbulence is presented. The method consists of using a grid for the advected scalar with a higher spatial resolutions than that of the momentum. The latter usually requires a less refined mesh and integrating both fields on a single grid tailored to the most demanding variable, produces an unnecessary computational overhead. A multiple resolution approach is used also in the time integration in order to maintain the stability of the scalars on the finer grid. The method is the more advantageous the less diffusive the scalar is with respect to momentum, therefore it is particularly well suited for large Prandtl or Schmidt number flows. However, even in the case of equal diffusivities the present procedure gives CPU time and memory occupation savings. The reason is that the absence of the pressure term in the scalar equation leads to much steeper gradients in the scalar field as compared to the velocity field.

physics.flu-dyn↗

arXiv2015

The effect of velocity boundary conditions on the heat transfer and flow topology in two-dimensional Rayleigh-Bénard convection

The effect of various velocity boundary condition is studied in two-dimensional Rayleigh-Bénard convection. Combinations of no-slip, stress-free and periodic boundary conditions are used on both the sidewalls and the horizontal plates. For the studied Rayleigh numbers Ra between $10^8$ and $10^{11}$ the heat transport is lower for $Γ= 0.33$ than for $Γ= 1$ in case of no-slip sidewalls. This is surprisingly opposite for stress-free sidewalls, where the heat transport increases for lower aspect-ratio. In wider cells the aspect-ratio dependence is observed to disappear for $\text{Ra} \ge 10^{10}$. Two distinct flow types with very different dynamics can be seen, mostly dependent on the plate velocity boundary condition, namely roll-like flow and horizontal zonal flow, which have a substantial effect on the dynamics and heat transport in the system. The predominantly horizontal zonal flow suppresses heat flux and is observed for stress-free and asymmetric plates. Low aspect-ratio periodic sidewall simulations with a no-slip boundary condition on the plates also exhibit zonal flow. In all the other cases, the flow is roll-like. In two-dimensional Rayleigh-Bénard convection, the velocity boundary conditions thus have large implications on both roll-like and zonal flow that have to be taken into consideration before the boundary conditions are imposed.

physics.flu-dyn↗

arXiv2015

A pencil distributed finite difference code for strongly turbulent wall-bounded flows

We present a numerical scheme geared for high performance computation of wall-bounded turbulent flows. The number of all-to-all communications is decreased to only six instances by using a two-dimensional (pencil) domain decomposition and utilizing the favourable scaling of the CFL time-step constraint as compared to the diffusive time-step constraint. As the CFL condition is more restrictive at high driving, implicit time integration of the viscous terms in the wall-parallel directions is no longer required. This avoids the communication of non-local information to a process for the computation of implicit derivatives in these directions. We explain in detail the numerical scheme used for the integration of the equations, and the underlying parallelization. The code is shown to have very good strong and weak scaling to at least 64K cores.

physics.flu-dyn↗

arXiv2014

Sidewall effects in Rayleigh-Bénard convection

We investigate the influence of the temperature boundary conditions at the sidewall on the heat transport in Rayleigh-Bénard (RB) convection using direct numerical simulations. For relatively low Rayleigh numbers Ra the heat transport is higher when the sidewall is isothermal, kept at a temperature $T_c+Δ/2$ (where $Δ$ is the temperature difference between the horizontal plates and $T_c$ the temperature of the cold plate), than when the sidewall is adiabatic. The reason is that in the former case part of the heat current avoids the thermal resistance of the fluid layer by escaping through the sidewall that acts as a short-circuit. For higher Ra the bulk becomes more isothermal and this reduces the heat current through the sidewall. Therefore the heat flux in a cell with an isothermal sidewall converges to the value obtained with an adiabatic sidewall for high enough Ra ($\simeq 10^{10}$). However, when the sidewall temperature deviates from $T_c+Δ/2$ the heat transport at the bottom and top plates is different from the value obtained using an adiabatic sidewall. In this case the difference does not decrease with increasing Ra thus indicating that the ambient temperature of the experimental apparatus can influence the heat transfer. A similar behavior is observed when only a very small sidewall region close to the horizontal plates is kept isothermal, while the rest of the sidewall is adiabatic. The reason is that in the region closest to the horizontal plates the temperature difference between the fluid and the sidewall is highest. This suggests that one should be careful with the placement of thermal shields outside the fluid sample to minimize spurious heat currents.

physics.flu-dyn↗

arXiv2014

Heat transport in boiling turbulent Rayleigh-Bénard convection

Boiling is an extremely effective way to promote heat transfer from a hot surface to a liquid due to several mechanisms many of which are not understood in quantitative detail. An important component of the overall process is that the buoyancy of the bubbles compounds with that of the liquid to give rise to a much enhanced natural convection. In this paper we focus specifically on this enhancement and present a numerical study of the resulting two-phase Rayleigh-Bénard convection process. We make no attempt to model other aspects of the boiling process such as bubble nucleation and detachment. We consider a cylindrical cell with a diameter equal to its height. The cell base and top are held at temperatures above and below the boiling point of the liquid, respectively. By keeping the temperature difference constant and changing the liquid pressure we study the effect of the liquid superheat in a Rayleigh number range that, in the absence of boiling, would be between $2\times10^6$ and $5\times10^9$. We find a considerable enhancement of the heat transfer and study its dependence on the bubble number, the degree of superheat of the hot cell bottom and the Rayleigh number. The increased buoyancy provided by the bubbles leads to more energetic hot plumes detaching from the hot cell bottom and, as a consequence, the strength of the circulation in the cell is significantly increased. Our results are in general agreement with recent experimental results of Zhong et al., Phys. Rev. Lett. {\bf 102}, 124501 (2009) for boiling Rayleigh-Bénard convection.

physics.flu-dyn↗

arXiv2014

Exploring the phase diagram of fully turbulent Taylor-Couette flow

Direct numerical simulations of Taylor-Couette flow (TC). Shear Reynolds numbers of up to $3\cdot10^5$, corresponding to Taylor numbers of $Ta=4.6\cdot10^{10}$, were reached. Effective scaling laws for the torque are presented. The transition to the ultimate regime, in which asymptotic scaling laws (with logarithmic corrections) for the torque are expected to hold up to arbitrarily high driving, is analysed for different radius ratios $η$, different aspect ratios $Γ$ and different rotation ratios $Ro$. It is shown that the transition is approximately independent of $Ro$ and $Γ$, but depends significantly on $η$. We furthermore calculate the local angular velocity profiles and visualize different flow regimes that depend both on the shearing of the flow, and the Coriolis force originating from the outer cylinder rotation. Two main regimes are distinguished, based on the magnitude of the Coriolis force, namely the co-rotating and weakly counter-rotating regime dominated by Rayleigh-unstable regions, and the strongly counter-rotating regime where a mixture of stable and unstable regions exist. Furthermore, an analogy between $η$ and outer-cylinder rotation is revealed, namely that smaller gaps behave like a wider gap with co-rotating cylinders, and that wider gaps behave like smaller gaps with weakly counter-rotating cylinders. Finally, the effect of $Γ$ on the effective torque versus $Ta$ scaling is analysed and it is shown that different branches of the torque-versus-$Ta$ relationships associated to different aspect ratios are found to cross within $15%$ of the $Re$ associated to the transition to the ultimate regime. The paper culminates in phase diagram in the inner vs outer $Re$ number parameter space and in the $Ta$ vs $Ro$ parameter space, which can be seen as the extension of the Andereck \emph{et al.} phase diagram towards the ultimate regime.

physics.flu-dyn↗

arXiv2013

Mixing in thermally stratified nonlinear spin-up with sources and sinks

Stratified spin-up experiments in enclosed cylinders have reported the presence of small pockets of well-mixed fluids but quantitative measurements of the mixedness of the fluid has been lacking. Previous numerical simulations have not addressed these measurements. Here we present numerical simulations that address how the combined effect of spin-up and thermal boundary conditions enhances or hinders mixing of a fluid in a cylinder. Measurements of efficiency of mixing are based on the variance of temperature and explained in terms of the potential energy available. The numerical simulations of the Navier--Stokes equations for the problem with different sets of thermal boundary conditions at the horizontal walls helped shed some light on the physical mechanisms of mixing, for which a clear explanation was lacking.

physics.flu-dyn↗

arXiv2013

Turbulence decay towards the linearly-stable regime of Taylor-Couette flow

Taylor-Couette (TC) flow is used to probe the hydrodynamical stability of astrophysical accretion disks. Experimental data on the subcritical stability of TC are in conflict about the existence of turbulence (cf. Ji et al. Nature, 444, 343-346 (2006) and Paoletti et al., A$\&$A, 547, A64 (2012)), with discrepancies attributed to end-plate effects. In this paper we numerically simulate TC flow with axially periodic boundary conditions to explore the existence of sub-critical transitions to turbulence when no end-plates are present. We start the simulations with a fully turbulent state in the unstable regime and enter the linearly stable regime by suddenly starting a (stabilizing) outer cylinder rotation. The shear Reynolds number of the turbulent initial state is up to $Re_s \sim10^5$ and the radius ratio is $η=0.714$. The stabilization causes the system to behave as a damped oscillator and correspondingly the turbulence decays. The evolution of the torque and turbulent kinetic energy is analysed and the periodicity and damping of the oscillations are quantified and explained as a function of shear Reynolds number. Though the initially turbulent flow state decays, surprisingly, the system is found to absorb energy during this decay.

physics.flu-dyn↗

arXiv2013

Optimal Taylor-Couette flow: Radius ratio dependence

Taylor-Couette flow with independently rotating inner (i) and outer (o) cylinders is explored numerically and experimentally to determine the effects of the radius ratio η on the system response. Numerical simulations reach Reynolds numbers of up to Re_i=9.5 x 10^3 and Re_o=5x10^3, corresponding to Taylor numbers of up to Ta=10^8 for four different radius ratios η=r_i/r_o between 0.5 and 0.909. The experiments, performed in the Twente Turbulent Taylor-Couette (T^3C) setup, reach Reynolds numbers of up to Re_i=2x10^6$ and Re_o=1.5x10^6, corresponding to Ta=5x10^{12} for η=0.714-0.909. Effective scaling laws for the torque J^ω(Ta) are found, which for sufficiently large driving Ta are independent of the radius ratio η. As previously reported for η=0.714, optimum transport at a non-zero Rossby number Ro=r_i|ω_i-ω_o|/[2(r_o-r_i)ω_o] is found in both experiments and numerics. Ro_opt is found to depend on the radius ratio and the driving of the system. At a driving in the range between {Ta\sim3\cdot10^8} and {Ta\sim10^{10}}, Ro_opt saturates to an asymptotic η-dependent value. Theoretical predictions for the asymptotic value of Ro_{opt} are compared to the experimental results, and found to differ notably. Furthermore, the local angular velocity profiles from experiments and numerics are compared, and a link between a flat bulk profile and optimum transport for all radius ratios is reported.

physics.flu-dyn↗

arXiv2013

Boundary layer dynamics at the transition between the classical and the ultimate regime of Taylor-Couette flow

Direct numerical simulations of turbulent Taylor-Couette flow are performed up to inner cylinder Reynolds numbers of {Re_i=10^5} for a radius ratio of {η=r_i/r_o=0.714} between the inner and outer cylinder. With increasing {Re_i}, the flow undergoes transitions between three different regimes: (i) a flow dominated by large coherent structures, (ii) an intermediate transitional regime, and (iii) a flow with developed turbulence. In the first regime the large--scale rolls completely drive the meridional flow while in the second one the coherent structures recover only on average. The presence of a mean flow allows for the coexistence of laminar and turbulent boundary layer dynamics. In the third regime the mean flow effects fade away and the flow becomes dominated by plumes. The effect of the local driving on the azimuthal and angular velocity profiles is quantified, in particular we show when and where those profiles develop.

physics.flu-dyn↗

arXiv2012

Optimal Taylor-Couette flow: direct numerical simulations

We numerically simulate turbulent Taylor-Couette flow for independently rotating inner and outer cylinders, focusing on the analogy with turbulent Rayleigh-Bénard flow. Reynolds numbers of $Re_i=8\cdot10^3$ and $Re_o=\pm4\cdot10^3$ of the inner and outer cylinders, respectively, are reached, corresponding to Taylor numbers Ta up to $10^8$. Effective scaling laws for the torque and other system responses are found. Recent experiments with the Twente turbulent Taylor-Couette ($T^3C$) setup and with a similar facility in Maryland at very high Reynolds numbers have revealed an optimum transport at a certain non-zero rotation rate ratio $a = -ω_o / ω_i$ of about $a_{opt}=0.33-0.35$. For large enough $Ta$ in the numerically accessible range we also find such an optimum transport at non-zero counter-rotation. The position of this maximum is found to shift with the driving, reaching a maximum of $a_{opt}=0.15$ for $Ta=2.5\cdot10^7$. An explanation for this shift is elucidated, consistent with the experimental result that $a_{opt}$ becomes approximately independent of the driving strength for large enough Reynolds numbers. We furthermore numerically calculate the angular velocity profiles and visualize the different flow structures for the various regimes. By writing the equations in a frame co-rotating with the outer cylinder a link is found between the local angular velocity profiles and the global transport quantities.

physics.flu-dyn↗

arXiv2012

Logarithmic temperature profiles in turbulent Rayleigh-Bénard convection

We report results for the temperature profiles of turbulent Rayleigh-Bénard convection (RBC) in the interior of a cylindrical sample of aspect ratio $Γ\equiv D/L = 0.50$ ($D$ and $L$ are the diameter and height respectively). Results from experiment over the Rayleigh number range $4\times 10^{12} \alt Ra \alt 10^{15}$ for a Prandtl number $\Pra \simeq 0.8$ and from direct numerical simulation (DNS) at $Ra = 2 \times 10^{12}$ for $\Pra = 0.7$ are presented. We find that the temperature varies as $A*ln(z/L) + B$ where $z$ is the distance from the bottom or top plate. This is the case in the classical as well as in the ultimate state of RBC. From DNS we find that $A$ in the classical state decreases in the radial direction as the distance from the side wall increases and becomes small near the sample center.

physics.flu-dyn↗

arXiv2012

Spatial distribution of heat flux and fluctuations in turbulent Rayleigh-Bénard convection

We numerically investigate the radial dependence of the velocity and temperature fluctuations and of the time-averaged heat flux $\bar{j}(r)$ in a cylindrical Rayleigh-Bénard cell with aspect ratio Γ=1 for Rayleigh numbers Ra between $2 \times 10^6$ and $2\times 10^{9}$ at a fixed Prandtl number Pr = 5.2. The numerical results reveal that the heat flux close to the sidewall is larger than in the center and that, just as the global heat transport, it has an effective power law dependence on the Rayleigh number, $\bar{j}(r)\propto Ra^{γ_j(r)}$. The scaling exponent $γ_j(r)$ decreases monotonically from 0.43 near the axis ($r \approx 0$) to 0.29 close to the side walls ($r \approx D/2$). The effective exponents near the axis and the side wall agree well with the measurements of Shang et al. (Phys.\ Rev.\ Lett.\ \textbf{100}, 244503, 2008) and the predictions of Grossmann and Lohse (Phys.\ Fluids \textbf{16}, 1070, 2004). Extrapolating our results to large Rayleigh number would imply a crossover at $Ra\approx 10^{15}$, where the heat flux near the axis would begin to dominate. In addition, we find that the local heat flux is more than twice as high at the location where warm or cold plumes go up or down, than in the plume depleted regions.

physics.flu-dyn↗

arXiv2012

Thermal boundary layer profiles in turbulent Rayleigh-Bénard convection in a cylindrical sample

We numerically investigate the structures of the near-plate temperature profiles close to the bottom and top plates of turbulent Rayleigh-Bénard flow in a cylindrical sample at Rayleigh numbers Ra=10^8 to Ra=2\times10^{12} and Prandtl numbers Pr=6.4 and Pr=0.7 with the dynamical frame method [Q. Zhou and K.-Q. Xia, Phys. Rev. Lett. 104, 104301 (2010)] thus extending previous results for quasi-2-dimensional systems to 3D systems for the first time. The dynamical frame method shows that the measured temperature profiles in the spatially and temporally local frame are much closer to the temperature profile of a laminar, zero-pressure gradient boundary layer according to Pohlhausen than in the fixed reference frame. The deviation between the measured profiles in the dynamical reference frame and the laminar profiles increases with decreasing Pr, where the thermal BL is more exposed to the bulk fluctuations due to the thinner kinetic BL, and increasing Ra, where more plumes are passing the measurement location.

physics.flu-dyn↗

arXiv2011

Prandtl and Rayleigh number dependence of heat transport in high Rayleigh number thermal convection

Results from direct numerical simulation for three-dimensional Rayleigh-Bénard convection in samples of aspect ratio $Γ=0.23$ and $Γ=0.5$ up to Rayleigh number $Ra=2\times10^{12}$ are presented. The broad range of Prandtl numbers $0.5<Pr<10$ is considered. In contrast to some experiments, we do not see any increase in $Nu/Ra^{1/3}$, neither due to $Pr$ number effects, nor due to a constant heat flux boundary condition at the bottom plate instead of constant temperature boundary conditions. Even at these very high $Ra$, both the thermal and kinetic boundary layer thicknesses obey Prandtl-Blasius scaling.

physics.flu-dyn↗

arXiv2011

Axially-homogeneous Rayleigh-Benard convection in a cylindrical cell

Previous numerical studies have shown that the "ultimate regime of thermal convection" can be attained in a Rayleigh-Benard cell when the kinetic and thermal boundary layers are eliminated by replacing the walls with periodic boundary conditions (homogeneous Rayleigh-Benard convection). Then, the heat transfer scales like Nu ~ Ra^{1/2} and turbulence intensity as Re ~ Ra^{1/2}, where the Rayleigh number Ra indicates the strength of the driving force. However, experiments never operate in unbounded domains and it is important to understand how confinement might alter the approach to this ultimate regime. Here we consider homogeneous Rayleigh-Benard convection in a laterally confined geometry - a small aspect-ratio vertical cylindrical cell - and show evidence of the ultimate regime as Ra is increased: In spite of the confinement and the resulting kinetic boundary layers, we still find Nu ~ Re ~ Ra^{1/2}. The system supports exact solutions composed of modes of exponentially growing vertical velocity and temperature fields, with Ra as the critical parameter determining the properties of these modes. Counterintuitively, in the low Ra regime, or for very narrow cylinders, the numerical simulations are susceptible to these solutions which can dominate the dynamics and lead to very high and unsteady heat transfer. As Ra is increased, interaction between modes stabilizes the system, evidenced by the increasing homogeneity and reduced fluctuations in the r.m.s. velocity and temperature fields. We also test that physical results become independent of the periodicity length of the cylinder, a purely numerical parameter, as the aspect ratio is increased.

physics.flu-dyn↗