Numerical Simulation and Demisting Efficiency Analysis of a Fold-line Three-channel Demister

Literature Overview

This paper by Hong Wenpeng, Deng Guangqiang, and Lei Jianqi from Northeast Electric Power University presents a computational fluid dynamics (CFD) study of a fold-line type three-channel demister, which is an internal component used in flue gas desulfurization (FGD) systems to remove entrained gypsum droplets before the gas exits the absorber tower. The research was funded under a key technology program addressing the prevention of "gypsum rain" — a persistent operational problem in wet FGD systems where liquid droplets escape the demister and settle downstream, causing corrosion and equipment degradation. The study employs the RNG k-ε turbulence model for the gas phase and the Euler-Lagrange DPM (Discrete Phase Model) for the liquid droplet trajectory prediction. The work appeared in Thermal Power Engineering, Volume 31, Issue 1, 2016, pages 54–58.

Core Technical Approach

The simulation methodology is well-structured for this type of multiphase flow problem. The gas phase is modeled using the Reynolds-averaged Navier-Stokes equations with the RNG k-ε turbulence model, which is an appropriate choice for internal flow with moderate turbulence intensity typical of demister operating conditions. The RNG variant offers improved accuracy for swirling and recirculating flows compared to the standard k-ε model, which is significant because demisters inherently create flow separation zones at the plate edges.

The liquid phase is treated as discrete droplets using the DPM approach, where individual droplet trajectories are tracked through the continuous gas phase. This Euler-Lagrange methodology is standard practice for particle-laden flow simulations and allows the researcher to calculate droplet impaction efficiency by tracking whether droplets of various sizes collide with the demister plates.

Key Findings and Technical Parameters

The study identifies three primary factors affecting demisting efficiency: plate geometry, gas velocity, and droplet diameter. The quantitative relationships are summarized below.

A critical finding is the velocity-dependent performance advantage of the three-channel configuration. When gas velocity is below 2 m/s or above 5 m/s, the performance difference between two-channel and three-channel demisters is marginal. However, in the intermediate range of 2–5 m/s — which is the typical design operating window for FGD demisters — the fold-line three-channel configuration demonstrates a clear advantage. This is because the fold-line geometry creates additional flow turning points, increasing the probability of droplet impaction at moderate velocities where inertial forces alone are insufficient for complete collection.

Connection to Engineering Practice

From the perspective of piping and equipment design, demister performance directly affects the sizing and material selection of downstream ductwork and stack components. Gypsum rain causes accelerated corrosion of carbon steel ducts and can lead to premature failure of flue gas exit systems. The typical design gas velocity for FGD demisters ranges from 1.5 to 3.5 m/s, placing most operating conditions squarely within the range where the three-channel fold-line design offers meaningful performance gains.

The finding that efficiency decreases with increasing plate spacing has direct implications for demister module design. In practice, plate spacing is a compromise between collection efficiency and pressure drop — tighter spacing improves efficiency but increases the risk of plugging from gypsum buildup and raises the fan power requirement. The study's results support the engineering practice of using 75–100 mm plate spacing for the final demister stage while potentially using wider spacing for upstream pre-demister stages.

Key Questions and Reflections

Several questions arise from this study that merit further investigation. First, the simulation does not address the long-term operational effects of gypsum accumulation on the fold-line geometry, which in practice significantly reduces the effective plate spacing over time. Second, the DPM approach assumes monodisperse droplet populations, whereas actual FGD droplet size distributions are broad and multimodal. Third, the study does not evaluate the erosion resistance of the demister plates at higher gas velocities, which is a critical concern for materials selection in high-temperature or abrasive service.

The fold-line geometry itself presents a manufacturing challenge. The precise angle and radius of the fold determine the flow turning characteristics, and any deviation during fabrication can alter the local flow field significantly. This has implications for the welding and forming processes used to manufacture demister plates, particularly for stainless steel or alloy plate constructions where weld distortion can shift the fold angle.

Study Insights and Implications

This paper provides valuable quantitative guidance for demister selection in FGD systems. The identification of the 2–5 m/s velocity range as the optimal operating window for three-channel fold-line demisters is particularly useful for engineers specifying equipment for new installations or retrofit projects. The CFD methodology described is transferable to other internal flow components in piping systems, including cyclone separators, mist eliminators, and flow distributors in heat exchanger bundles. The approach of combining RANS turbulence modeling with DPM particle tracking represents a mature and validated methodology for multiphase flow analysis in process equipment design.