Oral Presentation International Solvent Extraction Conference 2025

Characterization of Fluid Dynamics and Dense Packed Zones in Horizontal Gravity Separators for Design Optimization (122631)

Lukas Thiel 1 , Jörn Villwock 2 , Matthias Kraume 2 , Andreas Jupke 1
  1. Fluid Process Engineering (AVT.FVT), RWTH Aachen University, Aachen, Germany
  2. Chemical and Process Engineering, TU Berlin, Berlin, Germany

Introduction

In the chemical industry, horizontal gravity separators are fundamental unit operations for liquid-liquid phase separation. Despite the development of academic approaches for the design of separators, based on population balance equations (PBEs), the integration of these approaches into design practice has been limited, mainly due to their complexity and time consuming nature. As a result, simplified and standardized design methods, such as the Henschke model [1], are commonly used in industry [1 - 3]. However, these standardized methods are limited to liquid-liquid separators with a dispersion wedge and are therefore only applicable to systems where coalescence is not affected by surfactants [1, 3]. For systems containing surfactants or salts, referred to as coalescence hindered systems, liquid-liquid separators are typically operated with a dispersion band to utilize the maximum capacity of the unit. In such cases, the operation is governed by the fluid dynamics of the two-phase flow, as well as the free sedimentation, and coalescence in the dense packed zone (DPZ) [1]. Nevertheless, the mathematical description of the fluid dynamics and coalescence processes in liquid-liquid dispersions of coalescence hindered systems remains a significant challenge. To enable mathematical descriptions, experimental setups are required that are suitable for the parameterization and validation of partial models for coalescence and fluid dynamics.

Experimental Setups

To extend the standard design methodology for the operation of horizontal liquid-liquid gravity separators with a dispersion band, two experimental setups have been developed to analyze the fluid dynamics and the DPZ. The fluid dynamics within the separator is investigated using Particle Image Velocimetry (PIV) to measure flow velocities over the height and length of the separator to characterize the influence of the dispersion band (Figure 1). A second experimental setup was established to analyze the parameters influencing the DPZ (Figure 2). This setup allows the tracking of time-dependent DPZ formation and coalescence curves for DPZs up to 1 m in height, as observed in liquid-liquid systems with hindered coalescence. In addition, a shutdown method is used to quantify the local holdup in different sections of the DPZ.

Results

The DPZ experiments have revealed three distinct operational regions: the formation region, the steady-state region, and the coalescence phase. The evaluation of the time-dependent DPZ formation allows the parameterization of standardized coalescence models for the steady-state region. It also offers the possibility to validate a PBE approach for the formation region or the coalescence phase. By examining the influence of the hold-up in the free sedimentation and the Sauter mean diameter on the DPZ formation, it was observed that higher throughput and smaller Sauter mean diameter lead to increased steady-state heights of the DPZ. These results are consistent with the existing literature [5]. However, the question remains whether the model description applies to coalescence-hindered systems. Since it has been reported to influence the local hold-up [4], the implemented shutdown approach also considers the local drop size distribution in the DPZ. The hold-up in the DPZ is particularly important for the fluid dynamic approximation of the DPZ because the local viscosity in the DPZ correlates with the hold-up, which is important for describing the fluid dynamics of the DPZ in a separator.

In parallel, PIV was used to study the fluid dynamics outside the DPZ. The velocity profiles measured along the height and length of the separator provided detailed insights into the behavior of the continuous phases. Typically, single-phase CFD simulations that neglect the DPZ have been used to describe separator fluid dynamics. The experimental PIV data suggest that the dispersion band and the DPZ may have an impact on the flow patterns within the separator. Therefore, it is recommended that these effects be considered when improving the design methodology.

By integrating experimental results on DPZ behavior and fluid dynamics, we aim to establish a more robust and comprehensive design method for horizontal liquid-liquid gravity separators. The expected outcome is a design approach that optimizes separator capacity and efficiency by considering the complex interplay between coalescence dynamics and fluid flow. In the talk we will focus on the experimental results from both experimental setups. Future work will focus on refining and validating the models over a broader range of operating conditions and material systems.

67efbecc7642d-Separator-PIV-Setup.png

Figure 1. Separator-PIV-Setup

67efc958b8bd7-DPZ_Setup_neu_klein.png

Figure 2. DPZ-Setup

  1. Henschke, M. (1995): Dimensionierung liegender Flüssig-Flüssig-Abscheider anhand diskontinuierlicher Absetzversuche. RWTH Aachen, PhD thesis
  2. Kamp, J., Villwock, J., Kraume, M. 2017. Reviews in Chemical Engineering, vol. 33, no. 1, pp. 1-47. doi: 10.1515/revce-2015-0071
  3. Steinhoff, J., Bart, H.J. 2018. In: Davis, B., et al. Extraction. The Minerals, Metals & Material Series. Springer, Cham. Doi: 10.1007/978-3-319-95022-8_166
  4. Sibirtsev, S., Thiel, L., Kirsanov, A., Jupke, A. Droplet contact numbers and contact probabilities in liquid–liquid dense–packed zones, AIChE Journal. doi:10.1002/aic.18723.621
  5. Steinhoff, J., Gebauer, F., Flösch, D., Bart, H.J., Optimized Description of Dispersion Layers on a Vertical Separator. Chemie Ingenieur Technik 90 (7), pp. 979–987. doi: 10.1002/cite.201700092
  • Abstract category selection: