Elsevier

Fuel

Volume 165, 1 February 2016, Pages 41-49
Fuel

Predicting ash deposition behaviour for co-combustion of palm kernel with coal based on CFD modelling of particle impaction and sticking

https://doi.org/10.1016/j.fuel.2015.10.056Get rights and content

Abstract

A CFD model that simulates particle impaction and sticking has been developed for predicting the ash deposition characteristics for the co-combustion of South African Coal (SAC) and palm kernel expeller (PKE) in an entrained flow reactor. The numerical related errors, caused by interception and the improper resolving of the flow-field within the boundary layer near the deposition surface, are investigated. In order to minimize the numerical related errors without excessive meshing, a new revised particle impaction model has been developed and accomplished using an impaction correction factor. The particle sticking is predicted based on the molten fraction results that have been obtained from the chemical equilibrium calculations using the chemical fractionation data in order to consider the short residence time of fly ash particles. The simulation results show that a reasonable coarse mesh, coupled with the revised particle impaction model, is suitable to accurately resolve the particle impaction without using a prohibitive meshing size. The ash deposition behaviour is determined by the particle impaction and sticking properties. Good agreements are obtained between the predicted results and the experimental data for the ash deposition behaviour.

Introduction

Co-combustion of biomass with coal has been used as a near term measure to reduce CO2 emission from coal fired power plants [1], [2]. Currently, co-firing 10–20% (thermal) biomass with coal has been widely used in power stations in the UK and Europe and a higher co-firing rate is also used. Further, some power stations, such as the Drax in the UK, are being converted to firing 100% biomass. With the recent announcement of the new EU targets of reducing gas emissions, fuel flexibility is likely to be one of the key factors influencing the operation of the power stations in the future, and the uses of various biomass and waste for power in the EU are expected to substantially increase. Currently most large scale power stations are using relatively clean biomass such as wood pellets, and to some extent straw, olive stones and palm kernel expeller (PKE). An addition of up to about 10–20% biomass has only moderate effects on the ash deposition in the furnace. However, with an increased co-firing rate and the use of a wide range of biomass sources, ash related problems are ranking high on the list of significant operational constraints in co-firing power plants [3], [4]. Ash deposition reduces the efficiency of the heat transfer through the water walls and heat exchangers and causes corrosion of boiler tubes, which may lead to reduced generating capacity and unscheduled outages [5]. Therefore, an improved understanding of the ash deposition in firing various types of biomass is imperative for an efficient boiler operation and optimization in the future [6].

The optimum biomass co-firing rate in coal-fired boilers has still been mainly determined by experiments up to now [4]. Computational Fluid Dynamics (CFD) has been widely used for solid fuel combustion simulations and various sub-models have been developed for predicting ash depositions in lab-scale test facilities as well as for full scale boilers [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13]. Considerable progress has been made in the last decades in developing ash deposition models for CFD simulations [3], [4], [5], [14], [15], [16], [17], and more detailed and accurate sub-models for combustion, fuel/ash particle transport, and sticking and deposition rate predictions have been developed [3], [16]. Typically, Lagrangian methods are employed to compute the trajectories of ash particles, coupled with an Eulerian method for flow and gaseous phase reaction, where the inertial impaction of particles are often considered as the only or main mechanism for ash deposition formation. Therefore, accurate prediction of the particle impaction is a critical factor that affects the modelling of the ash deposition. The impaction efficiency of the particles is usually assumed to be unity which represents the worst scenario in terms of ash deposition rate [3], [16]. In practice the impaction efficiency can be much lower than this depending on the size, shape and density of the particles and the nature of the depositing surface. Weber et al. [17] investigated the requirements for accurate predictions of the impaction efficiency of fly ashes in a 2D geometry using the RANS-based CFD methods. It was concluded that only when the flow-field in the neighbourhood of the deposition surfaces is accurately resolved can accurate predictions of the particle impaction be obtainable by using the RANS-based CFD methods, especially for small particles since their trajectories are strongly affected by the boundary layer development. Haugen et al. [18], [19] applied direct numerical simulation (DNS) to investigate the particle impaction behaviour on cylinders and superheater tube bundles in a crossflow in order to accurately resolve the boundary layers around the cylinders. It should be noted that in these models [17], [18], [19], extremely fine grid for RANS or DNS is needed. However, this requirement is difficult to be satisfied in the simulation of an industrial boiler in order to predict ash deposition behaviours [17]. Often predicting major operational parameters such as the boiler temperatures and/or combustion properties may be achieved by using a reasonably coarse computer mesh. However, this reasonably coarse mesh can still lead to a significant error in ash particle impaction efficiency calculation [17].

In addition to the particle impaction, the stickiness of the ash particles plays a critical role in the formation of ash deposit and related slagging and fouling. The stickiness of an ash particle can be determined based on such as the viscosity, kinetic energy and degree of molten of fly ash particles. In terms of viscosity based sticking models, a reference viscosity is used to determine the stickiness. The value of the reference viscosity ranges within 8–108 Pa s which makes the sticking model strongly sensitive to the reference viscosity and may contribute to an inaccurate stickiness prediction [4], [20]. In addition, the kinetic energy thresholding sticking model, based on the Johnson–Kendall–Roberts (JKR) theory [14], [21], [22], has been proposed which takes into account the kinetic energy of the particles and the surface energy of both the particle and the impacted surface. However, a fitting process was necessary to develop the effective Young’s modulus versus the particle temperature and the particle diameter by matching the experimental data with the simulation results [14]. Further, the molten fraction-based sticking model has been developed using slag calculations based on the chemical equilibrium of the ash composition and it was found that deposition models based on the molten fraction of ash particles calculated from chemical equilibrium are promising [20].

Therefore, this paper aims to develop an improved ash deposition CFD model through (i) a new revised particle impaction model to minimize the numerical related errors with an affordable number of computational mesh, and (ii) an appropriate particle sticking model based on the ash chemistry and the particle momentum for the PKE where there is relatively a scarce amount of data available. The model developed has been tested using the experimental data from Imperial College’s entrained flow reactor [23], [24], where PKE with the high level of phosphorus has been considered.

Section snippets

Source of experimental data

Fig. 1 shows a schematic geometry of the entrained flow reactor (EFR) located at Imperial College London. It consists of four electrically heated furnaces with a diameter of 0.1 m and a length of about 5 m. The burner consists of a primary inlet through which the pulverized coal and the primary air are fed, and a secondary inlet for the heated air. The uncooled ceramic probe is placed at the sample port 2 to collect the ash deposits, which has a furnace temperature of approximately 1250 °C. More

Combustion models

In this paper, the combustion of coal and biomass is modelled in a combined Eulerian–Lagrangian frame of reference where the volatile combustion is modelled in the Eulerian frame of reference and the fuel/ash particles are tracked in a Lagrangian frame of reference. As stated in our previous papers, the single kinetic rate model was employed for the devolatilizations of the coal and biomass, where the rate of devolatilization depends on both the temperature and the volatile content of the

Model set up

In this paper, co-firing PKE with South Africa coal with four different co-firing rates have been investigated, namely the SAC are blended with 0, 20, 40 and 60 wt.% of the PKE. The same EFR operational conditions as indicated in Fig. 1 were employed for all four cases investigated. The coal flow rate of 0.014 g s−1, the primary air flow rate of 0.067 kg s−1 at 70 °C, and the secondary air flow of 1.167 kg s−1 at 300 °C [8] have been used for all the cases. Only the biomass additions were different to

Conclusions

An ash deposition model based on modelling the particle impaction and sticking has been developed for the modelling of ash deposition for co-combustion of coal and palm kernel. A revised particle impaction modelling approach is proposed in order to minimize the numerical related errors and avoid using an excessive mesh size. A particle impaction correction factor that takes account of both the effect of particle interception and errors in the particle impaction prediction when a coarse

Acknowledgments

The authors wish to thank Prof. J. Williamson (Imperial College, London) for the helpful discussions on the deposition behaviours in the co-firing experiments, Prof. R. Weber and Dr. A.M. Beckmann (Clausthal University of Technology) for their very helpful advice on modelling particle impaction. Also X. Yang would like to acknowledge the China Scholarship Council and the University of Sheffield and University Of Leeds, for funding his research studies. Support from the RCUK is also acknowledged.

References (62)

  • N.E.L. Haugen et al.

    MSWI super heater tube bundle: particle impaction efficiency and size distribution

    Fuel Process Technol

    (2013)
  • C. Wieland et al.

    Evaluation, comparison and validation of deposition criteria for numerical simulation of slagging

    Appl Energy

    (2012)
  • F. Wigley et al.

    Ash deposition at higher levels of coal replacement by biomass

    Fuel Process Technol

    (2007)
  • H. Zhou et al.

    Research on the slagging characteristics of easy to slagging coal in a pilot scale furnace

    Fuel

    (2013)
  • X. Wang et al.

    The ash deposition mechanism in boilers burning Zhundong coal with high contents of sodium and calcium: a study from ash evaporating to condensing

    Appl Therm Eng

    (2015)
  • Y. Niu et al.

    Investigations on biomass slagging in utility boiler: criterion numbers and slagging growth mechanisms

    Fuel Process Technol

    (2014)
  • Y. Niu et al.

    Further study on biomass ash characteristics at elevated ashing temperatures: the evolution of K, Cl, S and the ash fusion characteristics

    Bioresour Technol

    (2013)
  • G. Wang et al.

    Investigation on ash deposit formation during the co-firing of coal with agricultural residues in a large-scale laboratory furnace

    Fuel

    (2014)
  • G. Lu et al.

    Impact of co-firing coal and biomass on flame characteristics and stability

    Fuel

    (2008)
  • J.P. Smart et al.

    Radiation and convective heat transfer, and burnout in oxy-coal combustion

    Fuel

    (2010)
  • L. Ma et al.

    Modelling methods for co-fired pulverised fuel furnaces

    Fuel

    (2009)
  • B.F. Magnussen et al.

    On mathematical modeling of turbulent combustion with special emphasis on soot formation and combustion

    Proc Combust Inst

    (1977)
  • L. Ma et al.

    Modelling the combustion of pulverized biomass in an industrial combustion test furnace

    Fuel

    (2007)
  • L. Chen et al.

    Development of a three-dimensional computational slag flow model for coal combustion and gasification

    Fuel

    (2013)
  • B.A. Kader

    Temperature and concentration profiles in fully turbulent boundary layers

    Int J Heat Mass Transfer

    (1981)
  • D. Nutalapati et al.

    Assessing slagging and fouling during biomass combustion: a thermodynamic approach allowing for alkali/ash reactions

    Fuel Process Technol

    (2007)
  • C.W. Bale et al.

    FactSage thermochemical software and databases—recent developments

    Calphad

    (2009)
  • L. Fryda et al.

    Study on ash deposition under oxyfuel combustion of coal/biomass blends

    Fuel

    (2010)
  • A.A. Tortosa Masià et al.

    Characterising ash of biomass and waste

    Fuel Process Technol

    (2007)
  • M. Aho et al.

    Importance of coal ash composition in protecting the boiler against chlorine deposition during combustion of chlorine-rich biomass

    Fuel

    (2005)
  • O. Yakaboylu et al.

    Supercritical water gasification of manure: a thermodynamic equilibrium modeling approach

    Biomass Bioenergy

    (2013)
  • Cited by (0)

    View full text