稀有金属(英文版) 2019,38(12),1160-1168

Numerical simulation of preparation of ultrafine cerium oxides using jet-flow pyrolysis

Chao Lv Ting-An Zhang Zhi-He Dou Qiu-Yue Zhao

Key Laboratory of Ecological Metallurgy of Multi-metal Intergrown Ores of Ministry of Education,Special Metallurgy and Process Engineering Institute,Northeastern University

School of Control Engineering,Northeastern University at Qinhuangdao

作者简介：*Ting-An Zhang e-mail:zta2000@163.net;

收稿日期：13 June 2018

基金：financially supported by the National Natural Science Foundation of China (No.51904069);the Natural Science Foundation of Hebei Province of China (No.E2019501085);the Colleges and Universities in Hebei Province Science and Technology Research Youth Fund (No.QN2019312);the Fundamental Research Funds for the Central Universities (No.N172303012);the National Science and Technology Support Program (No.2012BAE01B02);

Numerical simulation of preparation of ultrafine cerium oxides using jet-flow pyrolysis

Chao Lv Ting-An Zhang Zhi-He Dou Qiu-Yue Zhao

Key Laboratory of Ecological Metallurgy of Multi-metal Intergrown Ores of Ministry of Education,Special Metallurgy and Process Engineering Institute,Northeastern University

School of Control Engineering,Northeastern University at Qinhuangdao

Abstract：

Ultrafine rare-earth oxides(REOs) are widely applied in all fields of daily life,but the conventional preparation methods are limited by a long procedure,low efficiency and severe environmental pollution.Our team has independently developed a jet pyrolysis reactor for the preparation of ultrafine cerium oxides,and this process has theoretical significance and practical application values.In this study,gas-solid pyrolysis reactions inside the jet-flow pyrolysis reactor were numerically simulated.We performed a coupling computation of the combustion,phase transformation and gas-solid reaction on Fluent and userdefined functions.We characterized the flows of different phases as well as the compositions and distributive laws of the reactants/products in the reactor.The gas-phase inlet velocity and dynamic pressure/additional pressure were related by a quadratic function.The velocity at the throat inlet changed the most,and the outlet velocity was very stable.The CeO₂ concentrations were obviously stratified.This study enriches theories of jet-flow pyrolysis and theoretically underlies the optimization and popularization of self-developed pyrolysis reactors.

Keyword：

Jet-flow; Pyrolysis; Reactor; Cerium oxides; Numerical simulation;

Received： 13 June 2018

1 Introduction

Ultrafine rare-earth oxides (REOs) are micro-or nanoscale powder particles that display the quantum effect,interface effect,small-size effect and macroscopic quantum tunnel effect [ 1] .Ultrafine REOs are widely used in daily life because of their demonstrated structural,photoelectrical and chemical changes.However,REO preparation methods have led to differences in the shapes,particle sizes and distribution of REOs and affect their performance [ 2, 3, 4] .Currently,the main industrial production methods are the oxalic acid precipitation method and carbonate precipitation method [ 5] ;however,these methods are limited by their low efficiency,large material consumption and severe environmental risks.Zhang et al.applied jet-flow pyrolysis technology to REO production and prepared REOs from the direct roasting of rare-earth chlorides,which is a green production method that does not generate waste gas or wastewater [ 6, 7] .Jet pyrolysis reactors are outstanding,with a fast reaction speed,short reaction time,high product purity,uniform particle sizes,high-quality and complete gas-phase absorption.

Pyrolysis is a complicated gas-solid reaction process inside a reactor,and this gas-solid hydromechanical and chelnical reaction has been widely studied through experiments [ 8, 9, 10, 11] and simulations [ 12, 13, 14, 15, 16, 17, 18, 19, 20] .Li et al. [ 21] numerically simulated and analyzed the three-dimensional turbulence cold-state flow fields inside a spurt precalciner using standard k-epsilon (k-ε) two-equation models [ 22, 23] and semi-implicit method for pressure-linked equations (SIMPLE) and focused on the effects of the vortex-carrying tertiary air,column sizes and middle neck on the gas flow fields inside the precalciner.Pan and Kong [ 24] stated that the airflow of peripheral particles should be considered when studying intra-particle transport at the particle scales of fast biomass pyrolysis and numerically simulated the evolution of biomass particles under fast pyrolysis conditions.The comparison and model validation between numerical simulated data and experimental data for single biomass particles under pyrolysis conditions suggested that the simulated temperatures and conversion rates were consistent with the experimental data.The lattice Boltzmann method [ 26] is able to reveal the detailed transformation of biomass particles under different pyrolysis conditions and can be used to improve reactorscale simulation engineering models and optimize the reactor design and operation conditions.Motasemi and Gerber [ 27] built complex coupling models involving the key factors of pyrolysis kinetics,phase transformation,fast variation of mixing characteristics and gas-phase transport;the group described heat transfer and mass transfer [ 28, 29, 30] during biomass particle pyrolysis and experimentally validated the results in the self-built pyrolysis reactor.Many basic works have been conducted,and many studies have raised relevant theories that have been validated by experiments and simulations.However,there are few experiments or simulations on cerium chloride pyrolysis inside macroscopic main reactors,and reports on the complex coupling simulations of combustion,phase transformation or pyrolysis reactions are rare.

A jet-flow pyrolysis reactor can be used to prepare REOs,such as cerium oxide and lanthanum oxide.In this study,CeCl₃ pyrolysis in a self-developed jet-flow pyrolysis reactor was conducted.The coupling operation of combustion,phase transformation and gas-solid reaction was realized using Fluent with a user-defined function (UDF).Specifically,a fluid flow simulation was first conducted.After the operations stabilized,combustion reactions were simulated.After the operations stabilized,the phase transformation was uploaded.Finally,after the operations stabilized,UDF was uploaded to simulate gas-solid chemical reactions.In particular,UDF was encoded according to chemic al reaction rate equations and the retract model.

2 Numerical simulation methods

2.1 Mathematical model

The dimensions of the self-built Venturi pipe jet-flow reactor are shown in Fig.1:The total length was 0.85 m,the left-side gas-phase inlets consisted of a fuel inlet (inlet l) and an O₂ inlet (inlet2),and the pipe diameters wereΦ₁=0.01 m andΦ₂=0.02 m.The sectional diameter of the material inlet (inlet3) was d=Φ₁=0.01 m;the sectional diameters of the straight pipeline were d₁=d₂=0.05 m;the lengths of the straight pipeline in front of and behind the throat were L₃=0.15 m and L₄=0.30 m,respectively;the throat had a diameter of d_e=d₁/2=0.025 m and length of L_e=4^*d_e=0.1 m;the length of the diameter-variable pipeline was L₁=L₂=0.14 m(tan8°<[0.5^*(d₁-d_e)/L₁]<tan20°),L₅=0.02 m;and the sectional diameter of the outlet was 0.05 m.The Venturi jet-flow reactor was pided into meshes on the Integrated Computer Engineering and Manufacturing code for Computational Fluid Dynamics (ICEMCFD).Specifically,the straight pipeline and diameter-variable pipeline were both pided into hexahedron meshes,while the throat and material drainage pipes were mesh-encrypted and pided into tetrahedrons.The total number of meshes exceeded 230 k (Fig.2).

2.2 Computational domain and boundary conditions

The numerical simulation was based on a three-dimensional (3D) non-steady algorithm,Euler multi-phase flow model and volume-finite discrete differential equations.The intra-reactor turbulence flows were described using standard k-εtwo-equation models and the SIMPLE algorithm based on pressure-velocity coupling.The control equations were differentially treated using the 2-order upwind scheme,the wall surfaces were set as adiabatic,and all the items converged to 1×10^-4 except for energy (to1×10^-6).This study focuses on CeC1₃ pyrolysis inside the reactor,and the main reactions involved include:

Fig.1 Jet-flow pyrolysis reactor:a real image,b dimension,and c section of gas-phase inlet

Fig.2 Mesh generation

The kinetic equation of pyrolysis is:

whereαis the reaction rate,R is the molar gas constant,8.314 J·mol^-1·K^-1,T (K) is the absolute temperature and t (s) is the reaction time.

According to the data for the pre-exponential factor and apparent activation energy determined from the kinetic equations of CeCl₃ reaction from experiments,we used UDFs to simulate CeCl₃ pyrolysis reactions.The velocities of CH₄inlet are 1.44-14.40 m·s^-1,the velocities of O₂ inlet are10.575-105.75 m·s^-1,and the velocity of material inlet is0.03 m·s^-1.The tested materials included CH₄,O₂,CO₂,H₂O,CeCl₃ and CeO₂.The physical parameters for all substances were identified by searching inorganic thermodynamics manuals,and the CeO₂ particle size was set at 1μm.

3 Result validation

The errors of the outlet products HCl and CeO₂ between physical experiments and numerical simulation were less than 5%(Table 1),which was acceptable and confirmed that the model selection,boundary condition setting and kinetic models in the numerical simulation were basically correct.The errors may be caused by an insufficient heat supply or CeO₂ intratubular sedimentation during combu stion.

  下载原图

Table 1 Model verification

To further analyze the intra-reactor fluid flow and pyrolysis reactions,we selected three intra-reactor monitoring lines,including the horizontal axis and lower and upper parallel lines 0.01 m from the axis,which were marked as y₁=-0.01 m,y₂=0 m and y₃=0.01 m,respectively.Ten monitoring planes were selected,marked as z₁=0.020 m,z₂=0.095 m,z₃=0.170 m,z₄=0.310 m,z₅=0.365 m,z₆=0.410 m,z₇=0.480 m,z₈=0.550 m,z₉=0.700 m and z₁₀=0.850 m,and widely distributed at key positions of the reactor (Fig.3).

All pyrolysis reactions inside the reactor were simulated with 90 k,230 k and 440 k grids and the same computation model.The effect of the grid size on the simulation results was investigated.The velocity field and temperature field change trends in the same computation model were investigated when the grid density was changed (Fig.4a,b).Under different grid pisions,the velocity changes for the monitoring lines in the reactor were not largely different (<5%).The temperature pision changes were not largely different(<5%) for the monitoring lines between grid numbers 230 k and 440 k,which largely differed from the grid pision at 90 k.Based on the above results,we chose the grid pision 230 k for use in the simulations,which reduced the amount of computation and led to stable results independent of the grid count.

4 Simulation of flow characteristics inside reactor

4.1 Analysis of pressure field

During CeCl₃ pyrolysis,the reaction materials were jointly affected by the negative pressure inside the reactor and applied pressure outside the reactor,which together ensured sufficient utilization of fuel-generated heat.In addition to increasing the fuel velocity (v) at the gas-phase inlet,the dynamic pressure (P) was rising at the throat of the jet-flow reactor (Fig.5),and the two parameters obeyed a quadratic function:P=2v²-11.29v+14.5.The gasphase inlet velocity and required additional pressure also fit a quadratic function P=1.62v²-1.55v-0.97.The gasphase inlet velocity and materials to be absorbed were linearly related,and the material supplement and gas-phase inlet velocity that ensured sufficient heat utilization were also linearly correlated (Fig.6).

Fig.3 Selection of monitoring line and monitoring surface (m)

Fig.4 Profile of monitoring line y₂ at different grid numbers for profile:a velocity profile and b temperature profile

Generally,the intra-reactor pressure changed as follows:A gas phase rapidly entered the left-side gas-phase inlet,producing an instantaneous negative pressure.The gasphase velocity declined,and the pressure rose rapidly in the pergent pipeline and forepart straight pipeline.The gasphase velocity rose,and the pressure decreased in the convergent pipeline and at the throat.The fluid velocity was maximized,and the pressure was negative in the second half of the throat.The fluid velocity then declined,and the pressure rose in the pergent pipeline and the outlet straight pipeline.The flow pressures stabilized at the outlet (Fig.7).

Fig.5 Relationships of gas inlet velocity with adsorption dynamic pressure and additional pressure

Fig.6 Relationships of gas inlet velocity with adsorbed material and supplementary material

4.2 Analysis of velocity field

During the simulation of CeCl₃ pyrolysis,the gas-phase mixture was set as phase 1,including CH₄,O₂,CO₂,H₂O(g) and HCl.The fluid mixed phase was set as phase2,including H₂O(1),CeCl₃ and CeO₂.The velocity distributions of the two phases inside the jet-flow reactor were analyzed using velocity nephograms and contours (Fig.8)and by monitoring online speed changes (Fig.9).

For phase l,the velocity at the left-side inlet rapidly increased after the entrance of the gas phase,but did not decrease until reaching the straight pipeline,and the velocity increased again after reaching the convergent pipeline and throat.At the throat,due to the joint effect of the two phases,the intra-reactor velocity was maximized,but rapidly declined after reaching the pergent pipeline and stabilized in the straight pipeline.The velocity distributions of phase l were slightly different among the three monitoring lines.The velocity of the monitoring line in the before-throat pipeline was very symmetrical,and the phase l velocity on line y₂ was slightly higher than those of the other lines because the gas-phase inlet was at the central position and the gas-phase fluid velocity on the central axis was slightly higher.Beyond the interface between the throat and drainage mouth,the velocity change trends for the three monitoring lines were as follows:The velocity maximized for line y₁ and minimized for line y₃,and a velocity blind spot appeared in the second half of the throat due to the impact and collision of the two-phase fluids.

Fig.7 Pressure distribution nephogram and contour map of the reactor

Fig.8 Velocity nephogram and contour map in reactor at phase:a phase l and b phase2

Fig.9 Velocity distribution variation at monitoring line in reactor

Phase2 was from the drainage mouth to the reactor,and its velocity zeroed in the pipeline in front of the interface between the drainage mouth and throat.The joint velocity due to the joint action of the two-phase fluids surpassed the initial velocities and was concentrated in the middle and upper parts of the reactor and then decreased rapidly in the pergent section and stabilized at the outlet.At the lower part of the interface between the pergent pipeline and the terminal straight pipeline was a section where the twophase fluid velocities were very low,which easily led to product accumulation and reduced the reaction efficiency.

4.3 Simulation of combustion effect

The numerical simulation of CeCl₃ pyrolysis involved combustion chemical reactions,evaporation,phase transformation and gas-solid pyrolysis reactions.In particular,the chemical reactions of fuel combustion should offer heat for subsequent evaporation,phase transformation and pyrolysis reaction and thus were very significant.When only combustion and two-phase coupled mass/heat transfer were considered,the fuel types required for experiments and simulation were determined by considering the combustion reactions of various gas fuels.Commonly used gas fuels include methane,propane and hydrogen.The atmospheric combustion of various fuels was simulated,and the average temperature change trends at the monitoring planes were observed (Fig.10).Among the various fuels,the highest temperature of 2400 K was found in C₃H₈ and the lowest average temperature of 620 K was generated by H_2.Since the temperatures required for CeCl₃ pyrolysis were from 750 to 1050 K,the heat generated from C₃H₈ or CH₄was used for water-phase evaporation and CeCl₃ pyrolysis.However,the price of C₃H₈ is higher than that of CH₄,fluctuates with the oil market and is not as stable as that of natural gas.Figure 11 shows temperature changes at different monitoring lines in the jet-flow pyrolysis reactor during the combustion reactions for various fuels.Clearly,the heat release from C₃H₈ leads to the greatest increase in the intra-reactor temperature,followed by heat released by CH₄ and hydrogen.The heat release from CH₄ combustion meets the demands for evaporation,phase transformation and intra-reactor out-of-phase chemical reactions.

Fig.10 Change trends of average temperature for various fuels on each monitoring surface

4.4 Analysis of concentration field

During the numerical simulation,the combustion reaction,phase transformation and pyrolysis reaction were computed successively,and the next reaction was not started until the previous reaction stabilized and until the intra-reactor temperature field changes and material concentration distributions all stabilized.Finally,the material concentration change trends inside the jet-flow pyrolysis reactor were analyzed.

The CH₄ concentration was 1 at the gas-phase inlet and decreased after the fast combustion in the reactor;at the center of the forepart straight pipeline,the reactions completely proceeded and the concentration was 0 (Fig.12).To ensure the generation of sufficient heat from combustion and the progression of evaporation phase transformation and CeCl₃ pyrolysis,we set the initial O₂concentration to 0.3 in the simulations (oxygen-rich status).Along with combustion reactions,the concentrations in the center that were under much contact with CH₄ decreased rapidly,and the concentration in the center of the forepart straight pipeline stabilized at 0.02;the O₂ concentration in the second half section decreased for two reasons:The new phase2 fluid diluted the O₂ concentration,and oxygen was further consumed as a fuel during pyrolysis.

The initial CO₂ concentration was 0,but CO₂ was generated during the combustion reactions and was diluted by the new phase2.In terms of the general change trends,the CH₄,O₂ and CO₂ concentrations were symmetrically distributed in the forepart pipeline.The HCl concentration was initially very high due to the formation of CeCl₃pyrolysis,but then decreased because HCl was diffused and diluted to the outlet due to the flowing carriage of intra-reactor fluids.CeCl₃ went flowed through the drainage pipe to the reactor.To ensure the feeding amount at a unit of time,we set the initial CeCl₃ concentration at the material inlet as 0.052,but it decreased rapidly at the throat due to water evaporation and dilution by mixed fluid phases,as well as CeCl₃ pyrolysis,and was zeroed after the reaction at the interface between the throat and the distension pipeline was completed.The CeO₂ concentration was initially 0,but with the CeCl₃ pyrolysis reactions,since CeO₂ was far denser than the gas-phase fluid,CeO₂under gravity action was deposited at the throat and middle and lower parts of the pergent pipeline.The concentration was obviously stratified,forming a significant concentration gradient.

Fig.11 Temperature distribution nephograms for various fuels on four monitoring surfaces

Fig.12 Concentration distribution nephogram of each material in the reactor

During pyrolysis reactions,we focused on the concentrations of the product CeO₂,which involved the heat supply from methane combustion.To study the two reactions inside the reactor,it was necessary to quantify the changing trends in the CO₂ concentration in the first reaction and CeO₂ in the second reaction.The CO₂ concentration was initially 0 and was maximized under the combustion reactions,but at the interface between the throat and drainage pipe,it decreased for lines yl and y₂due to the entrance of phase2 and was stabilized at the outlet (Fig.13).The CO₂ concentration at line y₃ was zeroed and did not recover or stabilize until it approached the outlet.The CeO₂ concentration was 0 in the pipeline in front of the interface between the throat and the drainage pipe,but beyond the interface,it increased and then decreased on the monitoring line.The CeO₂ concentration was maximized for line y₃,which was located at the middle and lower parts of the reactor.

Fig.13 Resultant concentration variation at monitoring line in reactor

4.5 Analysis of temperature field

The temperature field changes inside the reactor are shown in Fig.14.Clearly,the highest temperature was 1600 K and was mainly distributed in the forepart straight pipeline due to the heat released from combustion reactions.The temperature at the throat stabilized was near 1400 K,and the temperature required for CeCl₃ pyrolysis was from 750to 1050 K,indicating that the reaction could occur.Owing to the entrance of phase2 at the drainage mouth,the temperature rapidly dropped because of three reasons:First,the phase2 initially at 300 K transferred heat with phase l;second,the water in the mixed fluid phases in phase2 evaporated and was transformed into vapor,absorbing heat;and third,the CeCl₃ pyrolysis in phase2was an endothermic reaction,and its progression depended on heat absorption.

Fig.14 Temperature distribution nephogram in reactor

5 Conclusion

The relationships of the gas-phase inlet velocity with the dynamic pressure and additional pressure obeyed quadratic functions P=2v²-11.29v+14.5 and P=1.62v²-1.55v-0.97,respectively.The velocity of phase l was maximized at the throat and rapidly decreased at the pergent pipeline.The velocity of phase2 was maximized at the middle and upper parts of the throat,rapidly declined in the pergent segment and stabilized at the outlet.The initial CeC₂ concentration was 0,and along with pyrolysis reactions,CeO₂ under the action of gravity was deposited at the throat and at the middle and lower parts of the pergent pipeline,and the CeO₂ concentrations were obviously stratified.

参考文献

[1] Chen WZ,Yin HR,Ma ZH,Guo HW.Research progress of SmCo_5 nanoparticles and nanocomposites.Chin J Rare Met.2018;42(1):88.

[2] Liu JL,Zhang P,Zhang XK,Xie QQ,Pan DJ,Zhang J,Zhang M.Synthesis and microwave absorbing properties of La-doped Sr-hexaferrite nanopowders via sol-gel auto-combustion method.Rare Met.2017;36(9):1.

[3] Yang Q,Hu H,Wang SS.Preparation and desulfurization activity of nano-CeO_2/γ-Al_2O_3,catalysts.Rare Met 2018;37(7):554.

[4] Peyghambari SM,Yousefpour M.Electrodeposition of nanostructured Ti/(Ru+Ti+Ce)O_2 coatings.Rare Met.2018;37(1):13.

[5] Xiang L.Study on the Preparation of Light Rare Earth Oxide from Rare Earth Chloride Roasted Statically.Shenyang:Northeastern University;2010.15.

[6] Lv C,Zhao QY,Zhang ZM,Dou ZH,Zhang TA,Zhao HL.Numerical simulation of fluid dynamics in rare earth chloride solution in jet-flow pyrolysis reactor.Trans Nonferrous Met Soc China.2015;25(3):997.

[7] Lv C,Zhang ZM,Zhao QY,Dou ZH,Zhang TA,Zhao HL.Numerical simulation:preparation of La_2O_3 in a jet pyrolysis reactor.Rare Met 2015;34(8):600.

[8] Alhaddad M,Rendek E,Corriou JP,Mauviel G.Biomass fast pyrolysis:experimental analysis and modeling approach.Energy Fuels.2010;24(9):4689.

[9] Blasi CD,Branca C,Masotta F,Biase ED.Experimental analysis of reaction heat effects during beech wood pyrolysis.Energy Fuels.2013;27(5):2665.

[10] Dupont C,Commandre JM,Gauthier P,Boissonnet G,Salvador S,Schweich D.Biomass pyrolysis experiments in an analytical entrained flow reactor between 1073 K and 1273 K.Fuel.2008;87(7):1155.

[11] Suriapparao DV,Vinu R.Effects of biomass particle size on slow pyrolysis kinetics and fast pyrolysis product distribution.Rare Met 2018;39(3):465.

[12] Babu B,Chaurasia A.Modeling for pyrolysis of solid particle:kinetics and heat transfer effects.Energy Convers Manag.2003;44(14):2251.

[13] Bharadwaj A,Baxter LL,Robinson AL.Effects of intraparticle heat and mass transfer on biomass devolatilization:experimental results and model predictions.Energy Fuels.2004;18(4):1021.

[14] Blondeau J,Jeanmart H.Biomass pyrolysis at high temperatures:prediction of gaseous species yields from an anisotropic particle.Biomass Bioenerg.2012;41(41):107.

[15] Grnli MG,Melaaen MC.Mathematical model for wood pyrolysis comparison of experimental measurements with model predictions.Energy Fuels.2000;14(4):791.

[16] Kung HC.A mathematical model of wood pyrolysis.Combust Flame.1972;18(2):18.

[17] Krutof A,Hawboldt KA.Upgrading of biomass sourced pyrolysis oil review:focus on co-pyrolysis and vapour upgrading during pyrolysis.Biomass Conv Bioref.2018;8(3):775.

[18] Park WC,Atreya A,Baum HR.Experimental and theoretical investigation of heat and mass transfer processes during wood pyrolysis.Combust Flame.2010;157(3):481.

[19] Larfeldt J,Leckner B,Melaaen MC.Modelling and measurements of the pyrolysis of large wood particles.Fuel.2000;79(13):1637.

[20] Sadhukhan AK,Gupta P,Saha RK.Modelling and experimental studies on pyrolysis of biomass particles.J Anal Appl Pyrol.2008;81(2):183.

[21] Li XG,Ma BG,Wu B,Jian SW,Luo ZT.Numerical simulation and optimization of cold airflow field in sprayed calciners.J Harbin Inst Technol.2009;41(4):226.

[22] Zhao HL,Liu Y,Zhang TA,Gu SQ.Computational fluid dynamics(CFD)simulations on multiphase flow in mechanically agitated seed precipitation tank.JOM.2014;66(7):1218.

[23] Zhao HL,Zhang ZM,Zhang TA,Liu Y,Gu SQ,Zhang C.Experimental and CFD studies of solid-liquid slurry tank stirredwith an improved Intermig impeller.Trans Nonferrous Met Soc China.2014;24(8):2650.

[24] Pan YY,Kong SC.Simulation of biomass particle evolution under pyrolysis conditions using lattice Boltzmann method.Combust Flame.2017;178(1):21.

[251 He X,Chen S,Doolen GD.A novel thermal model for the lattice Boltzmann method in incompressible limit.J Comput Phys.1998;146(1):282.

[26] Haseli Y,Oijen JV,Goey LD.A detailed one-dimensional model of combustion of a woody biomass particle.Bioresour Tcchnol.2011;102(20):9772.

[27] Motasemi F,Gerber AG.Multicomponent conjugate heat and mass transfer in biomass materials during microwave pyrolysis for biofuel production.Fuel.2018;211(1):649.

[28] Chen H,Ma J,Liu HT.Least square spectral collocation method for nonlinear heat transfer in moving porous plate with convective and radiative boundary conditions.Int J Therm Sci.2018;132(10):335.

[29] Ma J,Sun YS,Li BW.Spectral collocation method for transient thermal analysis of coupled conductive,convective and radiative heat transfer in the moving plate with temperature dependent properties and heat generation.Int J Heat Mass Iransf.2017;114(3):469.

[30] Ma J,Sun YS,Li BW.Simulation of combined conductive,convective and radiative heat transfer in moving irregular porous fins by spectral element method.Int J Therm Sci.2017;118(8):475.