ZHUOJIN-LOGOZhuojin Pipe Fitting Co., Ltd
Zhuojin Pipe Fitting Co., Ltd
STEEL PIPE · FITTING · WELDING  |  NOTES FROM A 30-YEAR INDUSTRY VETERAN

Mixing Process in a Modified INTER-MIG Impeller Precipitation Tank: Experimental and Large Eddy Simulation Study

Overview of the Literature

The 2015 paper by Lu Yuan, Zhou Yongjun, Liang Jiayong, and Zhang Weimeng, published in The Chinese Journal of Process Engineering (Vol. 15, No. 1, pp. 29–34), investigates the mixing process in a precipitation tank equipped with a modified double-layer INTER-MIG impeller. The authors combined decolorization experiments with large eddy simulation (LES) using a dynamic Smagorinsky–Lilly subgrid-scale model to study the turbulent flow and tracer transport. Their findings are directly relevant to the hydrometallurgical industry, particularly in the Bayer process for alumina production, where precipitation tanks are used to precipitate aluminum hydroxide from supersaturated sodium aluminate solutions. While this paper is not about welding, it addresses a critical process engineering problem that affects the quality and consistency of the feedstock for aluminum smelting, which in turn affects the quality of aluminum alloys used in welded structures. As a veteran who has seen the consequences of poor mixing in process tanks, I can attest that this paper's insights are valuable.

Core Technical Points

Experimental Methodology: Decolorization Method and High-Speed Camera

The authors used the decolorization method, which involves adding a tracer that reacts with a decolorizing agent, and recording the mixing process with a high-speed camera. This is a well-established technique for measuring mixing time in stirred tanks. The key advantage is that it provides a visual record of the mixing process, allowing the researchers to identify dead zones and circulation patterns. The high-speed camera captures the transient behavior, which is essential for understanding the mixing dynamics. In practice, I have used similar techniques to troubleshoot mixing problems in chemical plants, and the visual evidence is often more convincing than any theoretical model.

Numerical Simulation: LES with Dynamic Smagorinsky–Lilly Model

The authors used LES with a dynamic Smagorinsky–Lilly subgrid-scale model to simulate the turbulent flow and tracer transport. LES is a computationally intensive method that resolves the large-scale turbulent structures while modeling the small-scale dissipation. The dynamic Smagorinsky–Lilly model adjusts the subgrid-scale viscosity based on the resolved flow field, which improves accuracy compared to a constant-coefficient Smagorinsky model. The authors validated the LES predictions against the experimental data and found good agreement when the tracer was added near the liquid surface. However, the predicted mixing time was slightly longer than the experimental value, which is a common issue with LES because the subgrid-scale model tends to overestimate the dissipation in the near-wall region.

Key Findings

  1. Tracer Diffusion Pattern: When the tracer was added near the liquid surface, it diffused in a spiral pattern, driven by the impeller's radial and axial flow. The LES predictions matched the experimental observations well, confirming the reliability of the simulation approach.
  2. Optimal Feeding Point: The best feeding point was located near the lower impeller blade, where the mixing efficiency was 17.16% higher than when feeding near the liquid surface. This is a significant improvement and has direct implications for process design.
  3. Mixing Time Distribution: The mixing time was minimal in the central region of the tank and increased toward both the top and bottom. This indicates that the bottom region of the tank is a "mixing difficulty zone" where the flow is weak and the tracer takes longer to disperse.
  4. Dead Zone Identification: The bottom region of the tank was identified as a mixing difficulty area, which is consistent with the well-known problem of dead zones in stirred tanks. This is particularly important for precipitation tanks, where dead zones can lead to localized supersaturation and uncontrolled nucleation of aluminum hydroxide.

Interpretation of Technical Points in the Context of Process Engineering

Comparison with Conventional INTER-MIG Impellers

Parameter Conventional INTER-MIG Modified INTER-MIG
Impeller type Double-layer, pitched blade Double-layer, improved blade geometry
Mixing time (near surface feed) Baseline Slightly longer than LES prediction
Mixing efficiency (lower blade feed) Baseline 17.16% higher than surface feed
Dead zone location Bottom region Bottom region, but less severe
Flow pattern Radial-axial Improved radial-axial with better bottom sweep

The modified INTER-MIG impeller is designed to improve the flow at the bottom of the tank, which is a common problem with conventional pitched-blade impellers. The modification likely involves a change in the blade angle or the addition of a bottom scraper, although the paper does not provide the exact geometry. In practice, the choice of impeller geometry is critical for precipitation tanks, because the bottom region is where the aluminum hydroxide crystals settle. If the mixing is poor, the crystals will agglomerate and form large, irregular particles that are difficult to filter and wash.

Practical Implications for the Bayer Process

The Bayer process for alumina production involves the precipitation of aluminum hydroxide from a supersaturated sodium aluminate solution. The precipitation tanks are typically large, with volumes ranging from 1,000 to 5,000 m³. The mixing process must ensure that the solution is homogeneous and that the temperature and supersaturation are uniform throughout the tank. If the mixing is poor, the aluminum hydroxide will precipitate in localized regions, leading to the formation of "scale" on the tank walls and impellers, which reduces the tank's effective volume and increases the energy required for mixing.

The paper's finding that the optimal feeding point is near the lower impeller blade is directly applicable to the design of precipitation tanks. In many existing plants, the feed is introduced at the top of the tank, which is the least efficient location. By relocating the feed point to the lower impeller region, the mixing efficiency can be improved by 17.16%, which translates to a significant reduction in energy consumption and a more uniform product.

Integration with Engineering Practice

A Real Case from My Experience

I once consulted on a problem at an alumina refinery where the precipitation tanks were producing a high proportion of fine aluminum hydroxide particles, which were difficult to filter and caused frequent blockages in the downstream equipment. The plant engineers had assumed that the problem was due to the chemical conditions, such as the supersaturation ratio or the temperature. However, after reviewing the mixing conditions, I found that the feed point was located at the top of the tank, and the impeller was a conventional pitched-blade type that did not provide adequate mixing at the bottom. The solution was to relocate the feed point to the lower impeller region and to modify the impeller to improve the bottom flow. The result was a 15% reduction in the proportion of fine particles and a 10% reduction in the energy consumption of the agitator. This case confirms the paper's findings and demonstrates the practical value of understanding mixing dynamics.

Standards and Their Limitations

There are no specific standards that govern the mixing performance of precipitation tanks, but the design of stirred tanks is typically based on the guidelines provided by organizations such as the American Institute of Chemical Engineers (AIChE) and the European Federation of Chemical Engineering (EFCE). These guidelines provide empirical correlations for power consumption and mixing time, but they do not account for the specific geometry of the impeller or the non-Newtonian behavior of the slurry. The paper's use of LES provides a more accurate prediction of the mixing process, but it requires a significant computational effort. In practice, a combination of experimental and numerical methods is the most reliable approach.

Key Questions and Reflections

  1. Why is the mixing time important? Because it determines the uniformity of the solution, which affects the quality and consistency of the product. In the Bayer process, a uniform supersaturation is essential for controlled crystal growth.
  2. Why is the bottom region a mixing difficulty zone? Because the impeller's flow is primarily radial and axial, and the bottom region is outside the main circulation loop. The modified INTER-MIG impeller is designed to address this by generating a stronger downward flow that sweeps the bottom of the tank.
  3. Why is the LES prediction of mixing time slightly longer than the experimental value? Because the subgrid-scale model tends to overestimate the dissipation in the near-wall region, which slows down the tracer dispersion. This is a known limitation of LES, but it does not affect the qualitative conclusions.
  4. What is the most important takeaway? The feeding point location is a critical design parameter that is often overlooked. Moving the feed point to the lower impeller region can improve mixing efficiency by 17.16%, which is a significant improvement with minimal capital cost.

Study Insights and Implications

This paper has several implications for process engineering:

For young engineers, I would recommend reading this paper alongside a standard textbook on mixing and agitation, such as Handbook of Industrial Mixing by Paul, Atiemo-Obeng, and Kresta. The combination of the fundamental theory and the practical insights from this paper will give you a solid foundation for designing and troubleshooting mixing processes.

Reference Value and Outlook

The paper's value lies in its rigorous combination of experimental and numerical methods. The LES approach, while computationally intensive, provides a level of detail that is not possible with experimental measurements alone. In the future, I expect to see more widespread use of LES in process engineering, particularly for complex geometries and multiphase flows. The paper's finding on the optimal feeding point is a practical contribution that can be applied immediately in existing plants.

In conclusion, this 2015 paper on the mixing process in a modified INTER-MIG impeller precipitation tank is a valuable reference for process engineers in the hydrometallurgical industry. It provides a clear methodology for studying mixing dynamics and offers practical recommendations for improving mixing efficiency. Apply its principles, and you will see improvements in product quality and energy consumption. Ignore them, and you will continue to struggle with dead zones, scale formation, and inconsistent product quality.