Experimental study of turbulent mixing in a T-shaped mixer
Huixin Li, Mohammad Mehdi Zamani Asl, Bastian Bäuerlein, Kerstin Avila, Duo Xu, Marc Avila · 2025
Read Experimental study of turbulent mixing in a T-shaped mixer by Huixin Li, Mohammad Mehdi Zamani Asl, Bastian Bäuerlein, Kerstin Avila, Duo Xu, Marc Avila (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-10264.txt.
Introduction
Inside a T-shaped mixer, two liquid streams collide head-on. They must blend as they travel down one channel. Huixin Li and five co-authors study the turbulent version of that flow in this 2025 fluids paper. Their new setup has a channel four centimeters wide. It allows optical measurements at small scales.
The team uses particle image velocimetry and laser-induced fluorescence. First, they copy flow patterns known from tiny mixers at gentle speeds. Then they turn to turbulence. They measure the flow's kinetic energy and how fast it dissipates. They also track the spread of the mixed substance and its spectrum. The spectrum shows early signs of Batchelor scaling. Last, they tie the velocity and concentration fields together. They check all of it against standard models and their own simulations.
Catalog record
| CREATOR | Huixin Li, Mohammad Mehdi Zamani Asl, Bastian Bäuerlein, Kerstin Avila, Duo Xu, Marc Avila |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 60 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-10264
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Experimental study of turbulent mixing in a T-shaped mixer Huixin Li, Mohammad Mehdi Zamani Asl, Bastian Bäuerlein, Kerstin Avila, Duo Xu, Marc Avila Abstract One of the most widespread canonical devices for fluid mixing is the T-shaped mixer, in which two opposing miscible liquid streams meet at a junction and then mix along a main channel. Laminar steady and time-periodic flows in T-shaped mixers have been thoroughly studied, but turbulent flows have received much less scrutiny despite their prevalence in applications. We here introduce a novel experimental setup with a hydraulic diameter of four centimetres that enables the optical study of turbulent mixing at small scales. Using this setup, we perform two-dimensional particle image velocimetry and planar laser-induced fluorescence measurements. First, we successfully replicate characteristic flow regimes observed in micro-scale T-shaped mixers at low Reynolds numbers. We then focus on the turbulent regime and characterize the turbulent kinetic energy and dissipation along the mixing channel. Further, we measure the scalar concentration variance and its corresponding probability density function and spectra. The latter exhibits an incipient Batchelor scaling. We estimate the mechanical-to-scalar timescale ratio and examine the link between the turbulent velocity and scalar fields. The measurement data are compared with model predictions and correlations used in engineering practice, and with our own direct numerical simulations performed with a spectral-element code. I Introduction Fluid mixing is an important multi-scale process in nature and in engineering applications. The T-shaped mixer, hereafter referred as T-mixer, is a canonical system widely used for fundamental research in fluid mechanics and as a testbed in chemical engineering. In its simplest and most studied form the inlets have square cross-section, and the main (outlet) channel is of rectangular shape. The macroscopic mixing properties in the laminar and transitional regimes (Reynolds number Re=HU_{0}/\nu\lesssim 500 , where U_{0} is the mean flow speed in the inlet channels, H their hydraulic diameter and \nu the fluid’s kinematic viscosity) are reasonably well understood. Specifically, the mixing quality has been unambiguously linked to the changes in the large-scale structures emerging in a sequence of flow instabilities, as explained in what follows (see e.g., the review Camarri et al. 2020 ). At low Reynolds numbers, Re\lesssim 120 , the two fluid streams are laminar and flow parallel to each other along the main channel. Here the mixing is purely diffusive, because the vortices in the flow are reflection symmetric and thus do not contribute to mixing the two streams Hoffmann et al. 2006a ; Bothe et al. 2006 . When the Reynolds number is increased, Kelvin–Helmholtz-like vortices emerge at the T-junction, and form a steady engulfment regime, where the vortices in the flow are asymmetric Hoffmann et al. 2006b ; Bothe et al. 2006 ; Fani et al. 2013 ; Mariotti et al. 2018 . In this regime, the contact surface of the two streams becomes strongly convoluted and leads to enhanced mixing, although the velocity and concentration fields are time-independent. When the Reynolds number is larger than about 220 , the engulfment regime becomes time-periodic Fani et al. 2014 ; Mariotti et al. 2018 and chaotic Thomas and Ameel 2010 . As the Reynolds number further increases, the engulfment regime is superseded by a symmetric regime, where the flow is time-periodic and regains the left-right reflection symmetry. Consequently, the mixing efficiency drops after this transition Thomas et al. 2010 ; Schikarski et al. 2017 . Subsequently the flow becomes chaotic and transitions to turbulence. For Re\gtrsim 650 , the flow in the outlet channel is already turbulent ( Schikarski et al. 2019a ) . We note that only a few numerical studies have explored the transition to turbulence and the turbulent regime in the outlet channel ( Minakov et al. 2013 ; Schikarski et al. 2019a ) . A few experiments and simulations focusing on particle precipitation in T-mixers have covered wide Reynolds number ranges, Re\in[250,15000] and have demonstrated the key influence of flow regimes and mixing on the particle size distributions Gradl et al. 2006 ; Schwarzer et al. 2006 ; Schikarski et al. 2019b ; Schikarski et al. 2022 . In addition to the Reynolds number, the mixing dynamics is also governed by the Schmidt number Sc=\nu/D , where D is the molecular diffusivity of the scalar. In liquid mixing, Sc=\mathcal{O}(10^{3}) , which is technically challenging in both experiments and numerical simulations due to the very small scale dynamics associated with large Sc . Specifically, when the flow is turbulent, the scalar mixing occurs down to the Batchelor length scale ( Batchelor 1959 ) , \eta_{b}=(\nu D^{2}/\epsilon)^{1/4} , where \epsilon is the viscous dissipation of turbulent kinetic energy. The Batchelor scale is a factor \sqrt{Sc} smaller than the smallest length scale in the velocity field, the Kolmogorov scale \eta_{k}=(\nu^{3}/\epsilon)^{1/4} ( Pope 2000 ) . Thus, in direct numerical simulations of liquid mixing, resolving the Batchelor scale demands enormous computing power Gotoh and Yeung 2012 ; Buaria et al. 2021 . Specifically, for Sc=1000 , about a factor of Sc^{3/2}\approx 31,623 more grid points are necessary to resolve the Batchelor scale in the scalar equation than for the Kolmogorv scale in the Navier–Stokes equations. In the experiments, measuring the fine fluctuations of the scalar field is also a technical challenge due to the limited measurement dynamic range in scale and limited signal-to-noise ratio for the weak scalar signal at small scales. For large Re and isotropic conditions, the spectrum of the velocity field features the classical Kolmogorov \kappa^{-5/3} in the inertial-convection range, followed by a sharper viscous-convection range down to the wave number ( \kappa ) corresponding to the Kolmogorov scale. In the spectrum of passive scalar fluctuations, Batchelor 1959 and Kraichnan 1968 theoretically predicted that there exists a \kappa^{-1} scaling for the length scales between the Kolmogorov and the Batchelor scales. Past evidence for this scaling law for the power spectrum is scarce and ambiguous. Gibson and Schwarz 1963 measured the passive scalar in grid turbulence at a point, used Taylor’s hypothesis to infer the spatial dynamics and fitted a \kappa^{-1} to the resulting spectrum, which however did not have a clear scaling range. Grant et al. 1968 in an ocean and tidal channel reported that the temperature fluctuations are consistent with a \kappa^{-1} scaling. In a turbulent jet flow, Talbot et al. 2009 used hot-wire anemometry and Rayleigh light scattering to simultaneously measure the velocity and concentration field, in which the scalar spectra exhibits \kappa^{-1} scaling law (again leveraging Taylor’s frozen hypothesis). Similarly, Iwano et al. 2021 used an optical fiber laser-induced fluorescence probe with a 2.8 \mu m spatial resolution to investigate the power spectrum in a high Schmidt number turbulent jet. Their measured spectrum approximately follows the \kappa^{-1} scaling but has a small bump. Similar results were also reported in other measurements with turbulent jets ( Miller and Dimotakis 1991a ; Miller and Dimotakis 1991b ; Miller and Dimotakis 1996 ) . Recently, Clay 2017 provided the evidence that the \kappa^{-1} scaling region does exist and becomes better defined with increasing Sc through direct numerical simulation. Despite this great progress, it appears necessary to revisit this scaling through experiments and without Taylor’s frozen hypothesis. Measuring the mixing processes down to length scales as close to the Batchelor