Flow Characteristics of Eccentric Double-Layer Improved INTER-MIG Impeller Studied by PIV

Literature Overview and Research Context

This study, published in Petrochemical Equipment in 2017 by Wang Guangwu and colleagues from Nanjing Tech University and the National Chemical Equipment Supervision and Inspection Center, investigates the hydrodynamic behavior inside a cylindrical stirred tank with an inner diameter of 420 mm, employing an eccentrically mounted double-layer improved INTER-MIG impeller. The research adopts Particle Image Velocimetry (PIV) to capture full-field velocity data under turbulent flow conditions, while a torque sensor is used to characterize the power consumption behavior. The work is significant because the INTER-MIG (Impeller with Tangential and Radial Flow combined with Mixed Impeller Group) design is widely applied in chemical reactors, mixing vessels, and process equipment where both radial and axial flow components are required simultaneously. Understanding the flow field evolution as a function of inter-impeller spacing and Reynolds number is critical for optimizing mixing efficiency, heat transfer, and mass transfer in industrial-scale equipment.

Core Technical Findings

Effect of Inter-Impeller Spacing on Axial Velocity Distribution

The study systematically varied the relative inter-impeller spacing ratio C2/H across the range of 0.148 to 0.400 and observed distinct flow regime transitions. When C2/H ranges from 0.148 to 0.400, the total axial velocity profile progressively approaches a normal (Gaussian) distribution as the spacing increases. This finding indicates that at smaller spacings, the two impellers interact strongly, creating a non-uniform velocity field dominated by localized recirculation zones. As the spacing increases toward C2/H = 0.300, the connecting flow between the two impellers becomes most pronounced, suggesting that this spacing ratio represents an optimal coupling point where the discharge flow from the upper impeller and the suction flow from the lower impeller overlap constructively. Beyond C2/H = 0.400, a compartmentalized flow phenomenon emerges, meaning the tank is effectively divided into separate flow regions with limited intercommunication. This compartmentalization is undesirable in most mixing applications because it creates stagnant zones that reduce overall mixing efficiency.

Reynolds Number Effects on Flow and Power Characteristics

The Reynolds number was varied across the range of Re = (3.75 to 9) × 10^4, which places the operation firmly in the turbulent regime. The power number NP decreases gradually as Re increases within this range, which is consistent with the general trend observed for most impeller types in turbulent flow. However, a notable finding is that when Re reaches or exceeds 9 × 10^4, the power number stabilizes, indicating that the system has entered a fully turbulent regime where viscous effects become negligible and the power consumption becomes essentially independent of Reynolds number. This plateau behavior is critical for scale-up calculations, as it allows engineers to use constant power number values for design purposes when operating above this threshold.

Process and Standards Analysis

The eccentric mounting of the impeller is a deliberate design choice to break the axisymmetric flow pattern that typically results from centrally mounted impellers. In centrally mounted configurations, the flow field is rotationally symmetric, which can lead to poor radial mixing and dead zones near the tank wall. Eccentric placement introduces an asymmetry that promotes radial circulation and enhances bulk mixing. This principle is analogous to the use of baffles in stirred tanks, although the mechanism is fundamentally different—eccentricity modifies the impeller-to-wall clearance asymmetrically, creating differential pressure zones that drive radial flow.

Integration with Engineering Practice

From a practical engineering standpoint, this research has direct implications for the design of chemical reactors, mixers used in petrochemical processing, and blending equipment in pharmaceutical and food industries. The finding that C2/H = 0.300 provides the strongest inter-impeller flow coupling suggests that this spacing should be adopted as a design benchmark for double-layer INTER-MIG configurations. Engineers should avoid spacing ratios below 0.200, where the impellers may interfere with each other's flow fields and create excessive turbulence without contributing to overall mixing. Similarly, spacing ratios above 0.400 should be avoided because the compartmentalized flow regime reduces mixing efficiency and may lead to material accumulation in stagnant zones.

The power number stabilization at Re ≥ 9 × 10^4 is particularly valuable for scale-up. In industrial applications, the actual Reynolds number in a production-scale tank is often significantly higher than in laboratory studies. The fact that NP plateaus means that the power consumption can be predicted reliably using the laboratory-derived power number, simplifying motor sizing and energy consumption calculations. For a typical petrochemical mixing vessel operating with a liquid viscosity in the range of 1 to 5 mPa·s and a tank diameter of 2 to 5 meters, the Reynolds number would easily exceed 10^5, placing the operation well within the stabilized NP regime.

Key Questions and Reflections

A critical question that arises from this study is whether the optimal spacing ratio of C2/H = 0.300 is universally applicable or whether it depends on other geometric parameters such as impeller-to-tank diameter ratio (D/T), impeller submergence depth, and tank aspect ratio (H/T). The study does not address these variables, which limits the generality of the findings. In my own engineering experience, the optimal impeller spacing is often sensitive to the D/T ratio, and a spacing that works well for a D/T = 0.5 configuration may not be optimal for a D/T = 0.3 configuration. Future research should investigate the interaction between spacing ratio and other geometric parameters to develop a more comprehensive design guideline.

Another point of reflection is the measurement technique itself. PIV provides excellent spatial resolution and is non-intrusive, which is a significant advantage over traditional hot-wire or laser Doppler anemometry techniques. However, PIV is limited to the measurement plane and does not directly provide three-dimensional flow information. For a complete understanding of the flow field in a stirred tank, especially one with an eccentrically mounted impeller, three-dimensional PIV or tomographic PIV would be more appropriate. The absence of three-dimensional data in this study means that some flow features, particularly those occurring out of the measurement plane, may not have been captured.

Study Insights and Implications

The most important takeaway from this research is the identification of a clear flow regime transition at C2/H = 0.400, beyond which the tank experiences compartmentalized flow. This is a critical design boundary that engineers must respect when configuring double-impeller systems. The practical implication is straightforward: for double-layer INTER-MIG impeller systems, the inter-impeller spacing should be maintained below 0.400 H, with a preferred value around 0.300 H, to ensure effective inter-impeller flow coupling and avoid compartmentalization. This guideline, combined with the power number stabilization criterion at Re ≥ 9 × 10^4, provides a solid foundation for the design and scale-up of eccentrically mounted double-impeller mixing systems in chemical and petrochemical applications.

Reference Value and Outlook

This paper contributes meaningfully to the limited body of literature on double-impeller stirred tank hydrodynamics, particularly for the INTER-MIG impeller type which has received relatively less attention compared to conventional Rushton turbines or Rushton-Kirk impellers. The combination of PIV measurement with torque-based power characterization provides a relatively complete picture of the flow and power behavior. For engineers involved in the design of chemical reactors, mixing vessels, and process equipment, this study offers practical design parameters and clear operating boundaries. The next logical step in this research direction would be to investigate the mixing performance metrics—such as mixing time, blend uniformity index, and mass transfer coefficient—directly, rather than inferring performance from flow field data alone. Such a study would bridge the gap between hydrodynamic characterization and process performance, providing a more actionable design tool for industrial applications.