We investigate the existence of multiple turbulent states in highly turbulent TaylorCouette flow in the range of Ta = 10 11 to 9 × 10 12 by measuring the global torques and the local velocities while probing the phase space spanned by the rotation rates of the inner and outer cylinders. The multiple states are found to be very robust and are expected to persist beyond Ta = 10 13 . The rotation ratio is the parameter that most strongly controls the transitions between the flow states; the transitional values only weakly depend on the Taylor number. However, complex paths in the phase space are necessary to unlock the full region of multiple states. By mapping the flow structures for various rotation ratios in a Taylor-Couette setup with an equal radius ratio but a larger aspect ratio than before, multiple states are again observed. Here they are characterized by even richer roll structure phenomena, including an antisymmetrical roll state.

We investigate the effect of fluctuations in thermal boundary layer on heat transfer in turbulent Rayleigh-Bénard convection for Prandtl number greater than one in the regime where the thermal dissipation rate is dominated by boundary layer contribution and in the presence of a large-scale circulating flow.

A new vertical water tunnel with global temperature control and the possibility for bubble and local heat & mass injection has been designed and constructed. The new facility offers the possibility to accurately study heat and mass transfer in turbulent multiphase flow (gas volume fraction up to 8%) with a Reynolds-number range from 1.5 × 10 4 to 3 × 10 5 in the case of water at room temperature. The tunnel is made of highgrade stainless steel permitting the use of salt solutions in excess of 15% mass fraction. The tunnel has a volume of 300 L. The tunnel has three interchangeable measurement sections of 1 m height but with different cross sections (0.3 × 0.04 m 2 , 0.3 × 0.06 m 2 , 0.3 × 0.08 m 2 ). The glass vertical measurement sections allow for optical access to the flow, enabling techniques such as laser Doppler anemometry, particle image velocimetry, particle tracking velocimetry, and laser-induced fluorescent imaging. Local sensors can be introduced from the top and can be traversed using a built-in traverse system, allowing for e.g. local temperature, hot-wire, or local phase measurements. Combined with simultaneous velocity measurements, the local heat flux in single phase and two phase turbulent flows can thus be studied quantitatvely and precisely. arXiv:1902.05871v1 [physics.flu-dyn]

In this work we study the heat transport in inhomogeneous bubbly flow. The experiments were performed in a rectangular bubble column heated from one side wall and cooled from the other, with millimetric bubbles introduced through one half of the injection section (close to the hot wall or close to the cold wall). We characterise the global heat transport while varying two parameters: the gas volume fraction α = 0.4% − 5.1%, and the Rayleigh number Ra H = 4 × 10 9 − 2.2 × 10 10 . As captured by imaging and characterised using Laser Doppler Anemometry (LDA), different flow regimes occur with increasing gas flow rates. In the generated inhomogeneous bubbly flow there are three main contributions to the mixing: (i ) transport by the buoyancy driven recirculation, (ii ) bubble induced turbulence (BIT) and (iii ) shear-induced turbulence (SIT). The strength of these contributions and their interplay depends on the gas volume fraction which is reflected in the measured heat transport enhancement. We compare our results with the findings for heat transport in homogeneous bubbly flow from Gvozdić et al. (2018) [1]. We find that for the lower gas volume fractions (α < 4%), inhomogeneous bubbly injection results in better heat transport due to induced large-scale circulation. In contrast, for α > 4%, when the contribution of SIT becomes stronger, but so does the competition between all three contributions, the homogeneous injection is more efficient.

The continuous injection case with a mean flow can be derived by integrating the point source solution since Eq. 1.3 is linear for passive scalar [2,17].