(511c) Solid-Liquid Flow and Mixing in Fluidic Oscillator

Fluidic oscillators (FOs) originated in the early 60s as a replacement for the then-infant stage transistors and logic circuits¹. Since its inception, FOs have found a variety of like flow meters, actuators, sensors, liquid-liquid extractors, emulsions, and wastewater treatment. FOs have reported having excellent mixing characteristics owing to which they have been reported to be used for process and fine chemical engineering applications^{2, 3}. Recently Pandit and Ranade⁴ and Yang Yu et al.⁵ have demonstrated the use of FO in continuous antisolvent crystallization of Paracetamol. Experimental and numerical characterizations of flow in FO have mainly been based on single-phase in past studies. The crystallization process generates solid particles resulting in solid-liquid flow in the FO. Therefore evaluating solid-liquid flow in FO becomes essential for establishing FO as a good crystallizer. Madane et al.³ have computationally investigated an improved version of FO for single-phase flow, mixing, residence time distribution, and heat transfer characteristics. However, extending this computational modelling for solid-liquid flow and quantifying solid-liquid flow, solids residence time distribution and mixing in the presence of solids will be certainly valuable in further developing and optimizing FO geometry for crystallization. Solid-liquid flow in FOs has not been reported extensively in the open literature. In this work, we have both experimentally and computationally studied and quantified the solid-liquid flow dynamics in the FO. The current study focuses on (i) Experimental validation of the modelled solid-liquid flow and mixing in the FO through high-speed photography, (ii) Quantification of solid residence time and mixing performance in the presence of solids, and (iii) Studying the effect of the particles size on solid residence time and flow dynamics.

Experiments and Modelling

The solid-liquid flow in the FO was modelled with the help of the Eulerian-Lagrangian approach using Discrete particle modelling (DPM) in Ansys Fluent 2021 R2. In this numerical study, Paracetamol (ρ = 1263 kg/m³) was selected as the discrete solid phase to mimic an actual model crystallization system. Single-sized spherical particles of diameters (D_p) of 50, 100, 150, and 200 μm were modelled. Two-way coupling of the continuous and discrete phases was employed. The turbulence in the FO was modelled by using the SST k−ω model with water as a continuous phase (ρ = 1000 kg/m³, μ = 0.001 kg/m-s). High quality and a well-structured grid was used for the numerical solution to ensure the maximum possible accuracy of the solution (Figure 1). The simulations were performed in an unsteady mode owing to the unsteady periodic pattern of the oscillating jet. The particles were traced every time step that was set at s. This work will investigate the effects of particle size on axial dispersion, residence times, and velocities of particles. The effect of particles on the jet oscillation frequency will also be determined to quantify the effect of particles on the overall performance of the FO.

To investigate the solid-liquid flow in the FO, coloured particles of known sizes (distribution) will be injected along with the continuous flow of water via a T-junction. FO used for this experiment will be made of transparent material to allow visibility of the injected particles. The particles will be traced with a high-speed camera. The obtained video will be post-processed using the image analysis technique. By image analysis, the stagnation zones for the particles will be identified to determine solid-dead zones. The concentration of the particles will also be measured at the outlet with the help of LDA with LIF capabilities to determine the solid residence times distribution in the FO.

Key Results and Conclusions

Initially, a two-way coupled DPM simulation was performed with 200 μm particles in the FO with an inlet velocity of 2.70m/s (3.5 LPM | τ ~ 1s). The particles were released from the inlet surface, normal to the surface with the same velocity as that of water. Figure 2 shows the velocity contour along with the particle trace and mixing for the different time steps. It was observed that the particles, once they enter the oscillation chamber, they follow the path of the oscillating jet. Appropriate high-speed measurements will be performed to validate the simulated solid-liquid flow dynamics and mixing in FO. The effect of the particles sizes (distribution) on the flow dynamics (mixing) will also be investigated. The modelling and experimental technique given in this work will serve as a base for establishing FO as a crystallizer that will be able to handle solids. The approach shown in this work will also be helpful in designing and optimizing new FOs for a variety of process engineering applications.

Acknowledgement

The authors greatly acknowledge SSPC (https://sspc.ie/), SFI Research Centre for Pharmaceuticals, University of Limerick, Grant number 12/RC/2275_p2) for supporting the work. The authors also acknowledge the support provided by Ansys for CFD licenses under the academic partnership program. The authors also wish to acknowledge the Irish Centre for High-End Computing (ICHEC | https://www.ichec.ie/) for the provision of computational facilities and support

References:

1. Woszidlo, R.; Krüger, O., Editorial for the Special Issue on “Fluidic Oscillators—Devices and Applications”. MDPI: 2022; Vol. 7, p 91.
2. Khalde, C. M.; Pandit, A. V.; Sangwai, J. S.; Ranade, V. V., Flow, mixing, and heat transfer in fluidic oscillators. The Canadian Journal of Chemical Engineering 2019, 97 (2), 542-559.
3. Madane, K.; Khalde, C.; Pandit, A.; Ranade, V., Flow Physics of Planar Bistable Fluidic Oscillator with Backflow Limbs. AIChE Journal, e17621.
4. Pandit, A. V.; Ranade, V. V., Fluidic Oscillator as a Continuous Crystallizer: Feasibility Evaluation. Industrial & Engineering Chemistry Research 2020, 59 (9), 3996-4006.
5. Yu, Y.; Pandit, A. V.; Robertson, P.; Ranade, V. V., Antisolvent Crystallization using a Fluidic Oscillator: Modeling and Validation. Industrial & Engineering Chemistry Research 2021, 60 (34), 12752-12766.