Operation parameters and design optimization based on CFD simulations on a novel spray dispersion desulfurization tower

Highlights

• Semi-empirical correlations are coupled in CFD to simulate a novel spray-dispersion tower.

• Desulfurization rate ratios of spray and bubble section under different operation conditions are compared.

• The suitable tubes immersion depth and liquid-to-gas ratio are 0.14 m–0.16 m and 3.03 L/Nm³–3.6 L/Nm³

• Twin entrances spray section and circular type bubble section are recommended to improve performance

Abstract

In this work, simulations were performed on a novel two-stage SO₂ absorption tower, namely, spray dispersion tower which consists of a spray section as the first stage and a bubble section as the second stage. The desulfurization performance of the spray dispersion tower was investigated by Computational Fluid Dynamics (CFD) method. After calculation, the predicted results were in good agreement with the measured results to validate the reliability of the simulation model.

The operational parameters optimization indicated that the appropriate tubes' immersion depth and liquid-to-gas ratios were 0.14 m–0.16 m and 3.03 L/Nm³–3.6 L/Nm³ to attain higher energy and economic savings, respectively. In the optimal bubble section, less time was required for the multiphase flow and chemical reaction values to reach the quasi-static state than that of the original one.

Overall, the SO₂ removal efficiency was 3% higher in the optimal spray section than the original spray section, and the desulfurization efficiency was 8% higher in the optimal bubble section than the original bubble section.

Introduction

The purification of sulfur dioxide and other relevant pollutants has garnered attention for many years. Environmental protection and human health are under threat by the discharge of sulfur dioxide and other pollutants. Acid rain, city haze or smog, atmospheric ozone depletion, and even the greenhouse effect are directly or indirectly caused by these air pollutants [1,2]. Normally, coal-fired power generation and sintering processes in the iron and steel industry produce the majority of the SO₂ emissions [3]. Nowadays, many companies and organizations are striving to achieve the ‘near-zero’ emissions required by the new stringent laws and regulations to decrease the SO₂ emissions [4].

Wet flue gas desulfurization (WFGD) is the most widespread technology for flue gas purification. There are two main types of WFGD systems, namely, spray scrubber and jet bubble reactor [5,6]. To meet the ‘ultra-low’ exhaust emission requirement, one common method is to increase the internal structures [7]. This is because the slurry residence times are relatively short, leading to insufficient gas-liquid contact in some spray towers. Another method involves increasing the amount of slurry or using a combination of two separate towers. However, these methods can lead to a series of problems. Adding more spray layers for the spray tower could raise the building or maintenance costs. Raising slurry layer height could decrease the equipment lifetime, due to the larger proportion of soot blockage, soil erosion, etc. [8]. For example, some distributor tubes or orifice plates are placed inside the slurry tank with slurry insertion depth of 400 mm or above, causing heavy pressure drop of flue gas in bubble tower [9]. As these operation parameters and structural layouts are not ideal, the industrial applications of these towers are limited to some extent [10].

For a long time, the design of the two kinds of WFGD towers has been based on experience, semi-empirical correlations, or trial and error. There are some limitations in the WFGD tower design and optimization processes because these designs are based on experience instead of simulation and pilot-test.

Recently, Ren et al. proposed a novel WFGD tower that combines these two technologies in one spray dispersion tower [11] (the spray dispersion tower is a two-stage desulfurization device that consists of a spray section and a bubble section). The spray dispersion system is divided into three parts: the primary section is the middle slurry spray section, which uses the contact between the gas and downward alkaline liquid flow from the nozzles. The jet bubble section is located below the spray section, which uses perforated plates or sparger pipes to generate a large number of bubbles to interact with liquid slurry. The demister device and gasoloid suppressor are placed at the upper layer. Ren and coworkers reported the design of the spray dispersion tower but gave no information on the parameters and conditions for optimized SO₂ purification [11].

Therefore, there is a crucial need to better understand which factors affect the combined mechanisms involved in the spray section and the bubble section together.

In this work, a systematic approach using Computational Fluid Dynamics (CFD) simulations together with pilot-test results is proposed for the operation and design optimizations of a spray dispersion tower that includes both a spray and a bubble section. After comparison, the ideal working conditions and reasonable design of spray and bubble sections are presented for the first time.

In dealing with wet desulfurization towers, there are two kinds of two-phase flow interaction simulation methods: the Eulerian-Lagrangian approach and the Eulerian-Eulerian approach [5,12]. Herein, the proposed approach combines the E-L approach to study the spray section, and the E-E approach to investigate the bubble section. This strategy can help to understand the physicochemical mechanisms of the original spray dispersion tower and its optimal structures.

The processes of gas-liquid multiphase flow, mass transfer, and SO₂ absorption reactions of the original and optimal structures are obtained and analyzed by these two approaches. The times to reach chemical reaction equilibriums in two different bubble sections are also compared to evaluate their operational performance. Moreover, the SO₂ removal efficiencies of two spray sections and two bubble sections are compared. Finally, this numerical model is applied to predict and optimize a pilot-scale spray dispersion tower, and the simulated results are validated by the pilot-test.

This work lays a solid foundation for further applications of the spray dispersion tower in industrial settings.

It is generally recognized that the design and operation of WFGD are based on simple experimental studies or semi-empirical correlations. However, even advanced measuring techniques cannot fully reveal the complex multiphase flow structure and pollutant purification processes inside the WFGD. This is due to the limited spatial and temporal resolutions, which cause a lack of understanding of its working mechanism. In recent years, CFD is considered as a valuable tool to deal with complex hydrodynamics, and a great number of efforts have been made to model the gas-liquid interaction and SO₂ absorption. CFD can provide a huge amount of process data, which could be verified by experiments [12,13].

There are some differences between the E-L approach and E-E approach [14]. As the Lagrange approach can track the real behavior of every liquid droplet when the liquid phase volume fraction is diluted to less than 10%, it is adopted in many liquid spray systems [15]. Meanwhile, the Eulerian–Eulerian two-phase flow approach is used in various kinds of jet bubble towers. This is because the general liquid volume fraction can be larger than 0.1, which contradicts the assumption of the Eulerian-Lagrangian model [16].

The difficulty in simulating the multiphase flow and mass transfer in spray and bubble towers is due to the fact that many parameters could influence the calculation results [[17], [18], [19]]. In the spray simulations, the discrete phase is mainly influenced by the turbulence. The movement of liquid droplets is random and a reasonable tracking method is necessary [20].

Moreover, as the SO₂ absorption is an exothermic reaction, the temperature change caused by vapor evaporation could be important. In the jet bubble simulations, the two-fluid flow characteristics and SO₂ removal performance are affected by a large number of hydrodynamic parameters, such as gas holdup, gas and liquid velocities, gas temperature, bubble size density/bubble diameter distribution and other related parameters [21,22].

Furthermore, the bubble size distribution could be affected by many interfacial forces and turbulence, leading to the internal energy shift of bubbles. The entire morphological evolution of the bubbles in the bubble reactors mainly includes the formation, aggregation, fragmentation, and death of bubbles. Therefore, a proper bubble size distribution model that incorporates the closure model of gas-liquid interaction forces is of significant importance. For the calculation of SO₂ removal efficiency based on acid-base neutralization reaction, a suitable absorption theory with a specific mass transfer rate plays a critical role [16,23,24]. Although there are many relevant transfer theories and diffusion theories, the application of these theories to the pilot-scale is still difficult. Also, in these mass transfer theories, the dimensional numbers, such as Re and Sc, should be taken into account.

Generally, the operation parameters of many working conditions such as liquid/gas ratio (L/G) and tube immersion depth have a great influence on desulfurization efficiency. Other operation conditions such as flow patterns, pressure drop, temperature, and the velocity of gas-phase could be strongly influenced by the internal geometry structures [18].

As the physical and chemical changes inside these towers are very complex, few reports that combine the macroscopic method with the microscopic method to modify the working processes of spray and bubble towers.

In this work, a series of combined numerical models, based on Navier-Stokes theory, was first applied to simulate a novel spray dispersion tower and validated by measurement to ensure model applicability. These models include the k-ε turbulent model, discrete phase model, droplet evaporation model and two-film SO₂ absorption model for the spray section, and the interfacial forces such as drag and lift force, RNG k-ε turbulent model, PBM model and two-membrane SO₂ absorption model for the jet bubble section. Then, the desulfurization performances of spray dispersion tower under various operating parameters and different setups were evaluated.