Qinwen Liu,Yan Shi,Wenqi Zhong, *,Aibing Yu

1 Key Laboratory of Energy Thermal Conversion and Control of the Ministry of Education,School of Energy and Environment,Southeast University,Nanjing 210096,China

2 Center for Simulation and Modelling of Particulate Systems,Southeast University-Monash University Joint Research Institute,Suzhou 215000,China

3 ARC Research Hub for Computational Particle Technology,Department of Chemical Engineering,Monash University,Clayton,VIC 3800,Australia

Keywords:Oxy-fuel combustion Co-firing of coal and biomass Oxy-fuel fluidized bed CFD simulation

ABSTRACT The co-firing of coal and biomass in oxy-fuel fluidized beds is one of the most promising technologies for capturing CO2.This technology has attracted wide attention from academia and industry in recent years as a negative emission method to capture CO2produced by carbon contained in biomass.In the past decades,many studies have been carried out regarding experiments and numerical simulations under oxy-fuel combustion conditions.This paper firstly briefly discusses the techno-economic viability of the biomass and coal co-firing with oxycombustion and then presents a review of recent advancements involving experimental research and computational fluid dynamics(CFD)simulations in this field.Experimental studies on mechanism research,such as thermogravimetric analysis and tube furnace experiments,and fluidized bed experiments based on oxy-fuel fluidized beds with different sizes as well as the main findings,are summarized as a part of this review.It has been recognized that CFD is a useful approach for understanding the behaviors of the co-firing of coal and biomass in oxyfuel fluidized beds.We summarize a recent survey of published CFD research on oxy-fuel fluidized bed combustion,which categorized into Eulerian and Lagrangian methods.Finally,we discuss the challenges and interests for future research.

1.Introduction

To address climate change,reducing carbon emissions is an urgent challenge,and limiting carbon dioxide(CO2)emissions is a key method.Carbon capture and storage(CCS)technology,as a potential method of limiting atmospheric emissions of CO2from power plant,has received increasing attention over the past decade[1].Oxy-fuel combustion is one of the most competitive CCS technologies.It transforms the traditional air(O2/N2)combustion mode into oxygen-rich(O2/CO2,O2concentration＞21%)combustion,thus greatly increasing the concentration of CO2in flue gas by usually more than 90%.Compared with air combustion,the combustion efficiency can be improved and pollution emissions can be reduced[2-4].An illustration of an oxy-fuel combustion system with flue-gas recirculation is shown in Fig.1[5].It demonstrates that the high-purity oxygen obtained from the air separation unit(ASU)through the air compression and separation process mixes with the recirculated flue gas.The mixed gas,as the replacement of air,is the oxidant for fuel combustion[3,4,6-8].After combustion,the flue gas is purified in a dust elimination unit,followed by further purification and compression in a carbon dioxide purification and compression unit(CPU);pure liquid CO2is retained for storage,transportation,and utilization.This process results in zero carbon emissions,as the CO2generated by combustion is completely separated and collected without any emission to the atmosphere.Due to the feasibility for boiler retrofitting,oxy-fuel combustion has been widely considered as one of the most promising technologies for industrial application since it was initially introduced by Horne and Steinburg in 1981.(See Figs.2 and 3.)

Currently,two approaches are applied for oxy-fuel combustion,namely pulverized coal(PC)boiler combustion and fluidized bed boiler combustion.Compared to a PC boiler,fluidized bed oxy-fuel combustion gives a better performance due to its wide fuel adaptability,for example coal,biomass and solid waste,high combustion intensity,smooth regulation characteristics,and low NOxemission.It has thus attracted worldwide attention in the field of multiphase combustion science and technology.The oxy-fuel combustion conducted on a circulating fluidized bed(CFB)boiler can reduce the energy consumption of recirculation flue gas fan because of the higher oxygen concentration in the furnace[11,12].Moreover,it is thought that,to reach the same combustion performance,an excess of 1%oxygen is possible for oxy-fuel combustion CFB,while that for PC boiler is 3%.Thus,energy consumption of the ASU can be reduced by around 0.5%due to less excess oxygen for CFB[12].Some studies have indicated that because the fluidized bed contains a large quantity of inert bed material,the process of switching from air combustion to the oxy-fuel combustion will be smooth,safe,and rapid[13,14].In view of these advantages,the fluidized bed is the most suitable reactor for oxy-fuel combustion.

Fig.1.System diagram of the oxy-coal combustion technology[5].

Biomass is one of the most abundant renewable energy resources,and widely distributed geographically.The oxy-fuel combustion with coal and biomass co-firing is able to capture CO2produced by carbon contained in biomass,thus called the clean combustion technology[14-16].And it has attracted wide attention from academia and industry in recent years.Utilizing oxy-fuel combustion with coal and biomass co-firing as a“negative emission method”to capture CO2was firstly proposed by Tan et al.[14,15].In addition to the significant carbon reduction effect,the co-firing of coal and biomass in CFB boilers can offset the disadvantages of biomass caused by low calorific value and high moisture content and make it possible to reduce the investment cost and save coal consumption.Furthermore,the ignition and burnout performance can be effectively improved by co-firing with coal due to the high reactivity(high volatile content,low ignition temperature)of biomass[17-19].So this technology promotes the efficient and clean utilization of low calorific value coal such as lignite(45% of global coal reserves)and low calorific value solid waste such as coal gangue[20,21].

Over the past decades,much research has been conducted regarding fluidized bed oxy-fuel combustion by means of experiments and numerical simulations[22-27],and the performance of coal and biomass co-firing with oxy-fuel combustion has become one of the most attractive topics in worldwide[28-32].In this review paper,the techno-economic viability of the biomass and coal co-firing with oxycombustion in power or heating generations is firstly discussed in Section 2 and then the recent advances in experimental research,including fundamental analysis and pilot scale facilities related to oxyfuel combustion with coal and biomass co-firing in a CFB boiler are summarized in Section 3.The development of numerical simulations in this field is introduced,with the application of Eulerian and Eulerian-Lagrangian methods reviewed in Section 4.In Section 5,we discuss challenges and future research.Our conclusions are described in Section 6.

2.Techno-economic Efficiency and Carbon Emission Reduction Effect

The co-firing of coal and biomass under oxy-fuel conditions is a promising technology that can be implemented immediately in nearly all existing coal-fired plants for heating or power generations with the aim to reduce the greenhouse gas emission,as well as to realize the industrial-scale utilization of biomass in a relatively short period of time.However,the techno-economic and environmental assessment are necessary.The techno-economic evaluation of co-firing under oxyfuel conditions in coal-fired plants often consider a wide range of factors generally referring to the following aspects:(a)the additional capital costs in retrofitting the existing coal-fired plants or building the new ones for co-firing with biomass and for the oxy-combustion;(b)the possible energy efficiency penalties due to the oxy-combustion and the addition of biomass.

Fig.2.Life cycle CO2emissions(a)and cost input(b)of the pulverized coal/biomass co-firing power plant with and without post-combustion CO2capture[9].

Fig.3.Experimental system of oxy-fuel combustion under fluidization condition[10].

Comparing to the PC boiler,retrofitting the existing CFB boiler to oxygen firing capability requires relatively minimal modification,mainly involving the additions of a cryogenic ASU,a gas processing system(GPS)that to process the flue gas stream leaving the boiler island to provide CO2,the associated new controls and instrumentation for these systems,and the auxiliary piping and draft systems such as the new oxygen supply piping,new flue gas recirculation system,new CO2product ductwork to new gas processing system and so on[33-35].The major new equipment ASU and GPS have significant land area requirements for the location of new equipment.For example,Alstom Power Inc.reported that for an existing 90 MWe boiler island that covers 3600 m2,the ASU requires about 3600 m2and the GPS about 6500 m2;only for oxy-combustion,the plant retrofit was estimated to cost about 1545 USD·kW-1,of which 85%were for the ASU and GPS and the modifications to the existing boiler cost 72 USD·kW-1[33].Nevertheless,the capital cost for plant retrofit with oxy-fuel is believed to prominently lower than that for postcombustion carbon capture retrofit or newly built integrated gasification combined cycle plants[31,35,36].

On the other hand,as CFB boilers have higher flexibility for the fuel type,the direct co-firing that is straightforward and less expensive is often adopted[37,38].So with only minor modifications,the CFB boiler can easily adapted for efficient co-combustion of coal and biomass and costs less benefiting from it being in no need of additional dedicated mills or burners for biomass[37,39].It was believed that co-firing of biomass and coal with the biomass fraction under 3%(energy based)did not need significant investment costs[37,38].If the biomass fraction was up to 20%,taking Polish energy system for example,the total cost of retrofit(in 2007)was estimated in range of 1.1-3.1[37,39].

In term of energy efficiency of the coal-fired plants,it has been widely accepted up to today that the oxy-combustion usually performs an energy penalty of 8%-12%,with the ASU energy consumption accounts to almost 45%of total ancillary power consumption[31,35].In detail,Cormos[40]presented a techno-economic analysis for oxycombustion power plant of 350 MW net power with a carbon capture rate higher than 90%at the generation parameters of 582°C/29 MPa achieved by both fossil fuels(coal and lignite)and renewable sawdust,and found that compared with coal-firing systems without oxycombustion,the oxy-combustion carbon capture results in an energy penalty of 9-12 net efficiency percentage points,37%-50%increase in total capital investment,7%-15%increase in the O&M costs and the electricity cost would increase 54%-95%.In spite of this,the oxy-combustion is still one of the most economic technologies to capture CO2from industrial combustion processes[36,41].However,when it comes to the biomass co-firing,things become more complicated as the energy efficiency is closely related to many factors such as the biomass and coal fuel characteristics,co-firing method,co-firing ratio and so on.Specifically,the biomass fuel characteristics affect the energy efficiency of cofiring systems mainly in two aspects which are the effects of the higher moisture contents and lower heating value of biomass on boiler efficiency and the auxiliary power consumption for pumps,fans and mills[35,37,38].Among these auxiliary power consumption devices,milling characteristic of various fuels usually is the dominate factor that determines net plant efficiency[38].In such condition,the torrefied biomass is supposed to one of the most favorable options due to the effective improvements for biomass in heating value and grindability during torrefaction process[37,38].When torrrefied biomass is co-fired,boiler efficiency was observed to be similar with that of the reference coal,while for other biomass co-firing cases boiler efficiency decrease was often shown,although when the co-firing ratio of biomass is under 10%,the negative effect on the boiler efficiency is usually not very obvious[37,42-44]especially considering the more flexible operational range of CFB boilers in biomass fraction.

Some studies have been implemented in detail to reveal the effects of different fuel types and biomass fractions.For example,Fogarasi et al.[31]assessed a conceptual design for co-firing sawdust and coal using the oxy-combustion technology and found that increasing the sawdust feedstock content would cause power plant performance to decrease.Mun et al.[38]investigated the boiler efficiency and power plant efficiency of an existing 500 MWe coal-fired power plant through process simulation for co-firing various biomasses along with two coal blends.Their study introduced the Hardgrove grindability index tests of the mixture of biomass(10 wt%)and coal(90 wt%)into the system model to accurately obtain the milling power consumption.The results showed that the addition of biomass would reduce the net power plant efficiency due to the lower calorific value of biomass and the increasing milling power consumption.But for the low rank coals,co-firing with the biomass especially the torrrefied biomass possibly increase the boiler efficiency and net power plant efficiency[38,42-44].Therefore,the internal waste heat utilization drying process in the plants was proposed to dry the biomass when co-fired with low rank coal such as lignite[40].These negative effects of biomass co-firing and oxycombustion technology in plant economic viability can be effectively compensated by increasing the plant scaling[31,38].For the larger power plants with the capacity of such as 300 MW,blending 40%-50%of biomass is still believed attractive as a compromise between economy and risk[38].

Parallel to the economic cost,the detailed greenhouse gas emissions of the coal and biomass co-firing power/heat generation with oxycombustion is also highly concerned by many researchers and policymakers.The life cycle approach has been widely used to evaluate the energy use,CO2emissions and cost requirements for the entire process of power/heat generation that starts from raw material acquisition through processing distribution,use until the final disposal by analyzing both internal and external factors[9,37,45].However,up to now there are very limited life cycle analysis on the coal and biomass co-firing power/heat generation of CFB with oxy-combustion.But some referential attempts have been conducted respectively on the biomass,coal or biomass-coal co-firing with or without CCS[32,46-50].For example,Agbor[32]developed the data-intensive computational model covering processes of the fuel harvesting,transportation and power generation to evaluate the technical potential and costs,as well as the environmental benefits of different co-firing technologies in a 500 MW subcritical PC plant in Alberta,Canada.They found the higher delivery costs of biomass feedstock than coal,as well as the additional capital costs to modify a plant to co-fire biomass actively contribute to the typical higher cost of co-firing generating electricity and thus the case of fully paidoff coal-fired power plant co-fired with forest residues which has the favorable plant modification costs and biomass obtaining and delivering costs shows the best economic and environmental benefits.The study of Yi et al.[9]also believed that the high costs from biomass supply chain process is one of the biggest handicap for biomass power generation and related measures are required to push forward development and application of coal/biomass power generation with CCS.

3.Experimental Advances

Currently,studies on oxy-fuel combustion with coal and biomass cofiring in CFB boilers primarily focus on fundamental experiments and experiments conducted using CFB boilers.

3.1.Fundamental experiments:brief introduction

In-depth understanding the co-combustion characteristic of coal and biomass is of fundamental importance to develop the efficient oxy-co-firing of biomass and coal.A series of fundamental experiments were conducted on bench scale reactors such as thermogravimetric analysis(TGA)[17,51-54],tube furnace[55-58],and fixed bed[59,60],combined with other analysis methods.

Research has focused on the influence of oxy-fuel combustion condition and the co-combustion of biomass and coal on the thermophysical properties,for example ignition temperature,temperature profile,NOxemission and ash composition.The results showed that oxy-fuel condition would cause a slight delay of the combustion when the oxygen concentration is the same as air[52].However,as the oxygen concentration increases,the ignition temperature will become lower than that in air condition[61],which is beneficial to the combustion.

The supplement of biomass can improve the ignition and thermal reactivity,promote the combustion and conversion of coal,decrease the heat flux,and shorten the total reaction time[10,17,51,53,56,61].It is also an effective solution for reducing environmental impacts such as CO2[10,54,59],and NOxand SO2emissions[57,58,60,62-64].Although some of the phenomena described above are influenced by the fuel properties[52],it is apparent that the coal and biomass co-firing in oxy-fuel combustion with CFB has many advantages that require further exploration.

3.2.Oxy-fuel fluidized bed experiments

Due to the superiorities of coal and biomass co-firing,some bench scale and pilot scale experiments were implemented in recent years(Table 1).

In Quebec,Canada,Canmet ENERGY has established a 0.8 MW oxyfuel combustion pilot scale device with a CFB boiler.In the facility,a series of experimental studies were conducted with different kinds of fuels including lignite,asphalt,and biomass especially for the co-firing of coal and wood particles with a weight ratio of 20%-50%(wood particles)[14,15].The purpose of the experiments was to evaluate the combustion characteristics of oxy-fuel combustion with coal and biomass co-firing,including the flue gas composition,organic volatile compounds emissions,and trace metal emissions.The experimental results showed that it is possible to operate reliably in oxy-fuel atmosphere and produce flue gas which contains high concentrations of CO2(-90%),and the addition of sawdust will not significantly affect the combustion.These results have provided information for the capture of CO2by cofiring of fossil fuels such as coal and petroleum coke and biomass under oxy-fuel combustion conditions,which is a reasonable way to achieve negative CO2emissions.

In Spain,CIUDEN[16,27,76,77]has constructed a multifunctional oxy-fuel combustion system with the largest thermal capacity in the world.The maximum input thermal power of the CFB boiler is 30 MW with a square furnace,and the cross-sectional area is 2.9×1.7 m2and the total height is 20 m.The system is suited for the combustion of different kinds of fuels,for example coal,biomass,and petroleum coke in traditional air or oxy-fuel combustion modes.They concluded that the characteristics of wide fuel adaptability in traditional CFB boilers are also applicable to the oxy-fuel combustion process.Compared with traditional combustion technology,the oxy-fuel combustion with CFB can reduce CO2emissions by 91%,while the coal and biomass co-firing oxyfuel combustion system with fluidized bed can reduce the CO2emissions by 120%,achieving negative CO2emissions.The system provides direct CO2capture,however,significant coking and corrosion problems were found during operations.

The research group of Liu et al.at Nottingham University established a 20 kW oxy-fuel combustion system firing biomass based on a CFB boiler(Fig.4)[65].The effects of combustion atmosphere(air or oxyfuel)and oxygen concentration in oxy-fuel combustion on the gas emissions and temperature distribution were systematically studied.Biomass,including two types of non-wood fuels(awn grass and straw particles)and a wood fuel(wood chips particles produced in China)were used in the experiment,while the oxygen ratio for combustionremained unchanged.Their results showed that the combustion temperature decreased significantly when the oxidants consisted of 21%O2and 79%CO2instead of air,and eventually led to the quench of combustion flame,and this was because the specific heat of CO2is larger than that of N2.To retain the temperature distribution similar to that under air combustion modes,the oxygen concentration in oxy-fuel combustion mode must be increased to 30%.

Table 1 List of the oxy-fuel fluidized bed experiments

At the Middle East University of Technology in Turkey,Varol et al.[66]studied the co-firing process of high sulfur lignite and biomass blend(mass ratio 50%)in an oxy-fuel combustion pilot plant based on CFB combustion(Fig.5).The purpose was to study the effects of biomass fraction on the emissions of NOx,SO2and CO.In the study,the effects of increasing biomass fraction under air combustion were investigated,and then comparative experiments were conducted under oxy-fuel combustion conditions.The results showed that the effect of biomass share on the pollutant emissions was very insensitive,similar to other results obtained from coal combustion,and may be far lower than any emission criteria currently in place.There appears to be no direct challenge to the co-firing oxy-fuel combustion in terms of gas emissions.However,the lack of research to date suggests that more extensive CFB oxy-fuel combustion data of biomass and coal co-firing are needed.Notably,K-doping was also detected in this experiment,while no obvious K-phase formed on the deposited probe.

Lupiáñez et al.[72-74]also studied emissions of SO2,NOx,and HCl,along with the deposition rates and ash mineralogy based on a fluidized bed of 30 kW.The results showed while the emissions of SO2were affected by the chlorine content supplied with the biomass,NOxemissions were much more dependent on operating conditions in a way similar to conventional combustion.Furthermore,oxy-fuel combustion increased the chlorine detected in fly ash compared to the air-fired tests.

At the Yonsei University of Korea,Sung et al.[67]constructed a 30 kW pilot-scale CFB boiler combined with flue gas recycling to study the oxy-fuel combustion of waste sludge,and biomass and CO2enrichment characteristics(Fig.6).In this study,the fuel mixing ratio(wood pellets)was 0-70%and the oxygen concentration ratio was 21%-30%.By increasing the mixing ratio of biomass and oxygen concentration ratio,the combustion accelerated,and the ignition time was shortened.Furthermore,the performance of heat recovery and CO2enrichment was highest when the mixed biomass ratio was 30%and the oxygen concentration ratio was 23%.Under the condition of 60%flue gas recirculation,the oxy-fuel co-firing of sludge and biomass was optimized in high concentrations of CO2(more than 90%)with less pollutant emissions,of which CO was 0.91%and NO was 14×10-6.

Similar research was conducted by Kumar et al.[70]in a 20 kW bubbling fluidized bed(BFB).They observed the temperature profile,flue gas emission,combustion efficiency,energy consumption,exergy efficiency,and energy destruction of the combustor when changing the blending ratio of the co-firing coal to biomass.They found that the cofiring of coal and biomass was successful inside the combustor,and observed the maximum temperature in the splash zone.By using the blend of 75%coal/25%biomass,a maximum conceivable combustion efficiency of 97.09%was accomplished in oxy-fuel mode.

In the past two years,institutions such as the University of Zaragoza in Spain[72-74],the Institute of Science and Technology in Bura,India[70]and the TUBITAK Malmara Research Center in Turkey[71]have successively proven the feasibility of oxy-fuel combustion of biomass and coal co-firing in a circulating/bubbling fluidized bed at the laboratory scale.They have also preliminarily studied the effects of oxygen concentration,coal and biomass properties and biomass substitution on combustion characteristics,conversion efficiency,carbon capture efficiency,pollutant emissions,dust deposition,corrosion characteristics,and other effects.

Some research institutes in China have also carried out relevant studies and obtained some preliminary conclusions.In recent years,the Institute of Engineering Thermophysics of the Chinese Academy of Sciences has realized oxy-fuel combustion with the co-firing of biomass and coal based on 50 kW and 0.1 MW CFB test benches respectively[18,68,69,78,79].The effects of operating parameters such as oxygen concentration,oxidants distribution mode,biomass-mixing ratio,and flue gas recirculation modes on temperature distribution,fly ash characteristics,flue gas composition,and pollutant emission properties of the system have been systematically studied.For example,the NOxemissions and characteristics of coal and biomass(buckwheat,cotton,and corn stalks)combustion under different gas distributions(50%O2/50%CO2)and flue gas recirculation(50%O2/50%recycling flue gas(RFG))were studied on the 0.1 MW CFB oxy-fuel combustion test platform(Fig.7).Compared with coal combustion,the temperature of coal and biomass co-combustion condition decreased in each zone of the furnace,particularly in the dense phase zone,while other operating parameters such as air volume ad secondary air ratio remained unchanged.

Fig.4.The oxy-fuel combustion system based on CFB firing biomass[65].

The Chongqing University[19]and the Southeast University[75]in China studied the effects of oxygen concentration,system temperature,and biomass mixing ratio on pollutant emissions in an oxy-fuel combustion system with BFB and CFB,and preliminarily discussed the formation mechanism of NO and other pollutants.

Additionally,the higher alkali and chlorine contents of biomass may bring severe operational problems regarding to heating transfer degradation,slagging,fouling,or corrosion when combusted in furnaces[37,53,80],as the alkali oxides or alkali salts will promote the formation of agglomeration by reacting with SiO2and Al2O3in ash to form the eutectic with the lower melting temperature.The oxy-fuel atmosphere usually with the higher SO3partial pressure due to the flue gas recycling in plants and co-firing with the coal that has the relatively higher sulfur content will obviously influence the gaseous environments and ash deposition.However,the related studies for oxy-co-firing of biomass and coal are very limited with the fundamental mechanisms on the ash deposition and slagging being still unclear and many problems have not achieved the consistent conclusions.Limited by the length of the review,the comprehensive discussion is not shown here.Some interesting studies and discussion can be found in the reference[55,73,80-82].

4.Numerical Advances

Numerical simulation is an important method for studying the dense gas-solid flow and characteristics of the oxy-fuel combustion process in a fluidized bed,which has great significance for the design,operation and optimization of the fluidized bed reactor[83].

Fluidized bed oxy-fuel combustion is a complex system involving multiphase flow,mass transfer,and heat transfer,as well as complicated homogeneous reactions and heterogeneous reactions.It is difficult to determine its complex flow behaviors,combustion characteristics,and heat and mass transfer mechanisms thoroughly by experimental research.For example,the characteristics of particle motion,gas concentration distribution,temperature field,and pressure distribution in the fluidized bed are barely observable with the existing experimental measurement methods.Numerical simulations thereby offer a method to overcome these difficulties.There are primarily two typical multiphase CFD modeling techniques for studying the gas-solid reaction system,namely the Eulerian method and the Eulerian-Lagrangian method[83].

Fig.5.Pilot scale plant for oxy-fuel combustion of high sulfur lignite and biomass in CFB[66].

Fig.7.Effect of flow staging:(a)temperature profiles in the furnace and emission factors of nitrogenous gases(b)in O2/CO2with the addition of corn straw[68].

However,up to now,there is nearly no research simulated the cocombustion of coal and biomass in oxy-fuel fluidized beds.Several numerical simulations of fluidized bed oxy-fuel combustion is shown in the following:

4.1.Eulerian method

A two-dimensional CFB oxy-fuel combustion model was established by Zhou et al.[84,85]using an Eulerian-Eulerian based simulation,and studied the gas-solid flow,combustion temperature,and composition distribution.At the North China Electric Power University,Yu[86]has also conducted similar work using Fluent software.The combustion process in the furnace was simulated,and the combustion of coal particles and temperature distribution in the furnace under different air volume ratios and oxygen concentrations was calculated.

Wu et al.[87]established a three-dimensional model for oxy-fuel combustion based on Eulerian-Eulerian model to simulate the flow characteristics,combustion,and pollutant emissions under hot flue gas cycle conditions.This study was similar that of Zhou et al.that only simulated the riser,without considering the circulation of particles and flue gas.Amoo et al.[88]also established a three-dimensional oxyfuel combustion numerical model based on Eulerian-Eulerian model with the consideration of the particles'cycle.The CFD simulation was conducted to study the flow,heat transfer,and emission characteristics of coal particles of varied sizes under oxy-fuel combustion conditions.

Further simulation studies were conducted to describe the characteristics of CFB combustion based on oxy-fuel combustion,among which Ren Lulu[89]established a comprehensive mathematical model of oxy-fuel combustion with a fluidized bed,based on a onedimensional and half-cell model including a series of sub-models.Through the establishment of gas,solid mass balance and energy balance equations in each sub-section,the fluidized bed in different atmospheres was simulated to study the temperature distribution,gas distribution and heat transfer coefficient distribution in the furnace during combustion.Zhou Hui[90,91]and Mao Yuru[92-94]also conducted similar work,focusing on the simulation of NOxgeneration,char combustion and hydrodynamic in the furnace under oxy-fuel combustion in a CFB boiler.Based in the Turow power plant in Poland,Krzywanski et al.[95,96]established a one-dimensional model of a 670 t·h-1CFB boiler.The combustion characteristics of the dense and dilute phase zone in the oxy-fuel CFB combustion process were studied,considering air classification,the desulfurization process,and the generation and reduction of NOx.

At the Southeast University,Fong Bo[97]established a twodimensional calculation model of coal combustion and coal biomass co-firing in fluidized beds under O2/CO2atmosphere considering the characteristics of fluidized bed and oxy-fuel combustion.The parameters suitable for O2/CO2atmosphere were selected in the combustion and chemical reaction simulation,and the coke combustion model was revised,adding the calculation model of the gasification reaction of coke with CO2and H2O.

The numerical simulation of oxy-fuel combustion in fluidized beds based on the Eulerian-Eulerian model cannot generate satisfactory results due to shortcomings in the approach.For example,the Eulerian-Eulerian approach cannot accurately describe the characteristics of real particles such as particle size distribution.In addition,recycling of particles in the CFB boiler suggest that simulating the riser only will not lead to accurate results.The Euler-Lagrange model provides an important method for overcoming these problems.

4.2.Eulerian-Lagrangian method

Compared to the Eulerian method,the Eulerian-Lagrangian method can provide detailed information regarding the solid phase,such as trajectories of particles,particle-particle,gas-particle,and particle-wall interactions.However,the simulations of oxy-fuel combustion in fluidized beds based on Euler-Lagrange approach have been rarely reported in the literature to date.

Using the ANSYS FLUENT platform and a self-defining function UDF,Adamczyk et al.[98,99]established an Euler-Lagrange model of air combustion and oxy-fuel combustion in a CFB boiler.The rectangular section of the furnace was 22×10 m2and the height was 42 m(Fig.8).The real particle size distribution(PSD)was considered in the model,and the combustion process was based on a single particle size,which can be used to detect the corrosion of the boiler wall.In this Euler-Lagrangian hybrid method,four-directions coupling was considered to represent the relationship between continuous and dispersed phases in mass,momentum and energy transfer,and particle flow dynamics theory(KTGF)was applied to consider the interaction between particles.The simulation results obtained were in good agreement with the actual test results.The established model can be used for mixed numerical calculations.This Euler-Lagrange model can be used as an accurate approach to calculate the inputs of semi-empirical models under the steady combustion process in a CFB boiler.

Fig.8.Geometric size of the boiler and simulation results of oxy-fuel combustion in CFB by the Euler-Lagrange method[99].

To describe the flow patterns of the oxy-fuel combustion in fluidized beds,Upadhyay et al.[100]adopted the multiphase particle-in-cell(MP-PIC)approach to simulate the gas-solid flow characteristics with air and O2/CO2as the fluidized gas.Their model successfully captured the gas-solid flow behaviors,including the typical annular nuclear flow structure.However,theirs is a cold model,which does not consider the chemical reactions occurring.At present,research focused on simulating the co-firing of coal and biomass in oxy-fuel fluidized bed is scarce,with some described below.

At the University of Leeds in the United Kingdom,Black et al.[101]used CFD commercial software to simulate the oxy-fuel combustion behaviors of biomass and coal co-firing based on a 500 MW in-service boiler.The simulation results and internal empirical models,which were in agreement with the experimental data under conventional air-coal combustion conditions,were applied to oxygen-enriched combustion.The study revealed the possible effects of changing fuel and combustion temperature on the heat transfer characteristics of inservice boilers.

Alvarez et al.[102,103]conducted a CFD numerical simulation experiment on the combustion of olive residue(0,10 wt%,20 wt%)and two different kinds of coal in different oxy-fuel condition and atmosphere.The results showed that the co-combustion of biomass and coal performed a beneficial synergistic effect,which significantly improved fuel burnout and reduced NOxemissions.

At the University of Technology in Sydney,Australia,Bhuiyan and Naser et al.[104]used the CFD method to simulate the air combustion and oxy-fuel combustion of powdered Russian coal and high volatile biomass in a 0.5 MW combustion device at different ratios.Furthermore,Bhuiyan et al.[105]conducted a CFD numerical simulation of co-combustion of 550 MW tangentially-fired pulverized coal boilers(Fig.9),calculated in detail the phenomena of component transfer and blending,and studied the ignition,combustion and pollutant generations in the furnace.However,all these studies were focused on pulverized coal boilers.

These studies showed that the overall modeling strategy of mixed fuels oxy-fuel combustion is the same as that of air combustion.Most of the effects of combustion atmosphere can be modeled,but it is still necessary to improve the combustion chemical reaction mechanism model to make it suitable for oxy-fuel combustion[106].

5.Discussion and Future Research Interests

5.1.Main findings

Fig.9.Numerical simulation of oxy-fuel combustion firing coal and biomass[105].

The techno-economic viability analysis shows the coal and biomass oxy-firing is a promising technology that can be implemented in nearly all existing or newly built coal-fired plants for heating/power generations with the aim to reduce the greenhouse gas emission,as well as to realize the industrial-scale utilization of biomass in a relatively short period of time.However,apart from the additional retrofitting costs,the usual obvious energy efficiency penalties in boiler efficiency or plant output efficiency due to the addition of biomass and oxycombustion process urgently require the in-depth understanding and effective developments and optimizations on the oxy-co-firing of biomass and coal.Studies of the co-firing of coal and biomass in oxy-fuel fluidized beds have made some progress over the last decades.Both experimental research and numerical simulations have been adopted to investigate the characteristics of the co-firing of coal and biomass in a fluidized bed.

These studies illustrate the benefits of oxy-fuel combustion with coal and biomass co-firing based on a CFB boiler.For example,it was demonstrated in some research that oxy-fuel combustion with higher oxygen concentrations could reduce the ignition temperature and NOxemission,except for the aggregation of CO2.The supplement of biomass can improve the ignition,thermal reactivity,and burnout of the fuel as well as decrease the heat flux.In addition,the emissions of NOxand SO2are also reduced and the system is able to achieve negative emission of CO2.However,these studies focus on the combustion dynamics under static conditions through thermogravimetric analysis and tube furnace experiments,which are quite different from fluidized combustion.

Although the research conducted on the fluidized bed is scant,there is a preliminary consensus that co-combustion of biomass and coal in oxy-fuel combustion can achieve negative CO2emissions.Compared with the traditional combustion technology,the oxy-fuel combustion with coal and biomass co-firing in CFB boiler can even reduce the CO2emissions by 120%.So far these studies mainly focused on the feasible discovery of oxy-fuel combustion with coal and biomass mixtures in a fluidized bed.Biomass addition to coal appears to have an increasing synergistic effect on combustion,for example by accelerating the combustion and shortening the ignition time as the oxygen enrichment and biomass portion in the mix increases.NOxis much more dependent on operating conditions in a way similar to conventional combustion,while SO2emissions are affected by the chlorine content supplied with the biomass.

For the numerical simulation,there are some preliminary studies on the oxy-fuel combustion process in fluidized beds,which mainly focus on theoretical or empirical modeling,and CFD simulations under the Eulerian-Eulerian framework,while the Eulerian-Lagrangian simulation of oxy-fuel combustion in fluidized beds has rarely been reported to date.However,simulation research on the oxy-fuel combustion of coal and biomass mixtures with a fluidized bed has yet to be reported.

5.2.Future research interests

Two major research gaps are identified in both academia and industry.The first is how biomass affects the oxy-fuel combustion characteristics of coal,and the second is how to establish a high-efficiency and low-pollution emissions coal and biomass co-combustion system to capture CO2.Therefore,for both experimental research and numerical simulations,the implementation of systematic and in-depth research in order to enrich the understanding of oxy-fuel combustion characteristics of coal and biomass co-firing system with fluidized beds is urgently needed.

In further research,well-designed experiments,measurements and analyses are necessary to implement the development of this technology.These key research areas are 1)determine how to implement experiments in fluidized oxy-fuel combustion conditions,meaning,how to develop static experiments into dynamic experiments that are closer to the real situation,2)determine how to characterize and analyze the characteristics of oxy-fuel combustion and 3)determine the oxy-fuel combustion characteristics of coal and biomass co-combustion system in a fluidized bed,in terms of how the biomass ratio,combustion atmosphere,oxygen concentration,combustion temperature and other parameters affect combustion characteristics.

Furthermore,many parameters cannot be measured due to the limitations of experimental conditions.By relying solely on experiments,the reaction process in a fluidized bed is difficult to observe,and many pertinent questions cannot be answered.As an effective approach,numerical simulations can compensate for the shortcomings of experiments.By establishing a three-dimensional mathematical and physical model which is suitable for the co-firing of coal and biomass in an oxy-fuel fluidized bed,the mechanism of oxy-fuel combustion with a CFB boiler will be further understood,for example the distribution and evolutionary characteristics of gas-solid flow,different products concentration and temperature distribution under different time and locations.It will be helpful for optimizing and retrofitting the boiler for a better performance.

On the other hand,to abate the remarkable energy efficiency penalties due to the higher compression consumptions and pressure drop loss from oxy-fuel process,supplying and milling consumptions of fuels and the possible combustion performance decline resulted from the addition of biomass,elevating the combustion pressure of boiler is a possible means to effectively increase the commercial competitiveness of the oxy-co-firing of biomass and coal.In fact,pressurized oxycombustion has also become the main direction of the development of oxy-fuel combustion technology.For the complicated oxy-co-firing of biomass and coal under higher pressure,the fundamental researches on the techno-economic and environmental assessments,cocombustion characteristics and mechanisms,pollutant emissions,combustion dynamic controls and optimizations and so on,are very interesting and urgently required.

6.Conclusions

This paper presents a brief review of oxy-fuel combustion with CFB boiler co-firing coal and biomass,with a key focus on experiments and CFD simulations.The main findings are summarized as follows:

(1)Co-firing of coal and biomass under oxy-fuel condition is found to be a promising and competitive way of realizing negative CO2emissions,as well as improving combustion performance.Fluidized bed oxy-fuel combustion is a feasible and important technology.

(2)The combustion behavior of different kinds of coal and biomass remains unclear and the influence of different operating parameters,such as fuel properties,oxy concentration,and bed temperature on the hydrodynamic and heat transfer have yet to be clarified.Furthermore,the enrichment of CO2in the flue gas is an important indicator in CCS for the capture and sequestration and needs further exploration.

(3)Although some CFD simulations have been conducted in the past few years,the sub-models adopted were developed in air combustion conditions,which reduces the accuracy and reliability.It is therefore urgent to update the sub-models through comprehensive experiments and theoretical research.

(4)System simulations and optimizations are also needed in conjunction with the life cycle analysis of the entire electricity generation process and this remains lacking.

Acknowledgments

This work was supported by the Key Program of the National Natural Science Foundation of China(51736002)and the Natural Science Foundation of Jiangsu Province(BK20180386).ABY is also grateful to the Australian Research Council,for the partial financial support.