Wende Tian, Haoran Zhang, Zhe Cui,*, Xiude Hu

1 College of Chemical Engineering, Qingdao University of Science & Technology, Qingdao 266042, China

2 State Key Laboratory of High-efficiency Utilization of Coal and Green Chemical Engineering, Ningxia University, Yinchuan 750000, China

Keywords:ReaxFF MD simulation CCLG-MMA system simulation Sensitivity analysis Plant wide control

ABSTRACT Nowadays, the efficient and cleaner utilization of coal have attracted wide attention due to the rich coal and rare oil/gas resources structure in China.Coal chemical looping gasification (CCLG) is a promising coal utilization technology to achieve energy conservation and emission reduction targets for highly pure synthesis gas.As a downstream product of synthesis gas,methyl methacrylate(MMA),is widely used as raw material for synthesizing polymethyl methacrylate and resin products with excellent properties.So this paper proposes a novel system integrating MMA production and CCLG (CCLG-MMA) processes aiming at ‘‘energy saving and low emission”, in which the synthesis gas produced by CCLG and purified by dry methane reforming (DMR) reaction and Rectisol process reacts with ethylene for synthesizing MMA.Firstly, the reaction mechanism of CCLG is investigated by using Reactive force field (ReaxFF)MD simulation based on atomic models of char and oxygen carrier (Fe2O3) for obtaining optimum reaction temperature of fuel reactor(FR).Secondly,the steady-state simulation of CCLG-MMA system is carried out to verify the feasibility of MMA production.The amount of CO2 emitted by CCLG process and DMR reaction is 0.0028(kg CO2)-1·(kg MMA)-1.The total energy consumption of the CCLG-MMA system is 45521 kJ·(kg MMA)-1,among which the consumption of MMA production part is 25293 kJ·(kg MMA)-1.The results show that the CCLG-MMA system meets CO2 emission standard and has lower energy consumption compared to conventional MMA production process.Finally, one control scheme is designed to verify the stability of CCLG-MMA system.The CCLG-MMA integration strategy aims to obtain highly pure MMA from multi-scale simulation perspectives, so this is an optimal design regarding all factors influencing cleaner MMA production.

1.Introduction

Fossil fuels such as coal, petroleum, and natural gas are essential in the current energy consumption structure in China.However, more than 340 billion tons of CO2are generated from fossil fuel and released into the atmosphere each year, of which about 20 billion tons are absorbed in ocean, 7 billion tons are absorbed in terrestrial ecosystem, and only no more than 10 billion tons are utilized by humans [1].How to reduce the emission of CO2from fossil fuel utilization process has become a serious challenge now.Therefore, the development of efficient cleaner utilization technology becomes an increasing focus of industrial practitioners as well as academic researchers.Coal chemical looping gasification(CCLG) consisting of fuel reactor (FR) and air reactor (AR) is a promising coal utilization process using oxygen carrier to separate CO2and improve the conversion efficiency of coal.CCLG has lower CO2emission compared with traditional coal gasification process in producing highly pure synthesis gas.In this process, char, as a pyrolysis production of coal,reacts in FR with oxygen carrier recycled from AR, as shown in Fig.1.

Fig.1.Simplified schematic of CCLG process.

Oxygen carrier occupies an important role in CCLG process as it has a decisive effect on the gasification rate of coal[2].As it carries oxygen as well as heat in this process,its oxygen-carrying capacity and reactivity determine the reaction time and recycle rate of the whole CCLG process,which further affect the production efficiency of synthesis gas [3].Due to the characteristic of continuous oxidation and reduction in high temperature, the oxygen carrier should have low production cost, easy availability, anti-sintering, wear resistance, and good fluidization properties [4].In addition, high combustion efficiency is also an important basis for the selection of oxygen carrier.Currently, the commonly used oxygen carriers are mainly oxides of Ni, Cu, Fe, Co, and Mn.In particular, ironbased oxygen carriers are widely used because of their good reactivity, easy availability, and low environmental pollution [5].The feasibility of using iron oxide as an oxygen carrier in chemical looping combustion process was investigated to prove the stable conversion rate of Fe2O3[6].The reaction conditions between coal and oxygen carrier (Fe2O3) using steam as gasification agent was explored, showing that the carbon conversion efficiency increases from 55.74% to 81% with the increase of O/C ratio [7].Therefore,Fe2O3is selected as oxygen carrier in this paper.Doubtlessly, it necessitates to study and discuss the reaction mechanism between Fe2O3and coal by a microscope approach such as molecular dynamics (MD) simulation to further optimize CCLG process.

MD methods have a wide application in a variety of fields such as chemistry, physics, and materials science, which has been greatly improved with the continuous development of force field theory [8].Reactive force field (ReaxFF), as a typical type of force field in MD, is capable of handling processes involving millions of atoms in spatial scale and 100 nanoseconds in time scale, filling the gap between quantum chemical and empirical force field based computational chemical methods [9].Reactions can be simulated to take place in ReaxFF when the temperature and chemical environment are given properly.ReaxFF MD simulation has shown good performance for rapid reaction systems such as pyrolysis,combustion, metal oxidation, and explosion [10].ReaxFF MD simulation is also widely used in research of metals and metal oxides.The chain-like nucleation and growth of oxides on the aluminum nanoparticle surface were studied, providing a novel view in the oxidation mechanism of aluminum from molecular scale [11].ReaxFF force field data in C/H/O/N was retrained to incorporate aluminum interactions,and the pyrolysis process of RDX on different aluminum oxide layers was studied subsequently[12].The formation and growth of iron oxide nanoparticles were simulated using a fast-computational protocol based on atomistic reactive molecular dynamics in the synthetic and proliferation process of the nanoparticles with a homogeneous medium [13].The kinetics of surface oxide growth on single-crystal of copper was investigated, with results well coinciding with experimental results[14].At the same time, ReaxFF MD simulation is used to calculate the complicated system that includes more than 10 thousand atoms of coal.A Fugu sub-bituminous coal model containing 23898 atoms was developed based on the results of proximate element analysis,13C NMR, and solvent extraction experiments[15].The Liulin coal model containing 28351 atoms were constructed to analyze the product distributions in pyrolysis process of coal and the effects of temperature on pyrolysis[16].In a word,the ReaxFF MD simulation provides a novel approach to probe complicated systems with high accuracy, so this is a proper method to apply in the research of CCLG.

From macro-perspective, the simulation of chemical looping process has been carried out for optimization purposes recently.A biomass chemical looping model was simulated,and its thermodynamic analysis was carried out based on principle of Gibbs free energy minimization [17].A poly-generation system coupled coal pyrolysis and CCLG was proposed to realize coproduction of synthesis gas fuel and heat/electric on the base of thermodynamic results [18].Dynamic simulation as an important simulation means has also been reported in recently years for process stability test.For example,a 25 kW pilot plant was simulated to analyze the influence of chemical reaction on the flow of gas[19].A 1 MW fluidized bed chemical looping combustion process was designed and optimized based on a dynamic mathematical model to predict distribution of temperature,velocity of solid and gas,and composition of gas [20].

Synthesis gas from CCLG process is an important raw material for producing industrial organic chemicals [21].Methyl methacrylate(MMA)occupies a vital role in the downstream productions of synthesis gas.MMA can be synthesized through several methods,such as acetylcholine (ACH) method, isobutylene oxidation method, and ethylene method.ACH method was developed in 1934 by ICI Company in UK and industrialized in 1937.It is the most widely used route because of the high yield of MMA [22].But ACH method has inevitable limitations including the highly toxic raw material and environmentally inefficient process.Isobutylene oxidation method has advantage of environmental friendly and safe process.But the high cost of raw material, catalyst, and utility in the recovery of methanol process hinders its wide application in industry[23].Comparatively,ethylene method using synthesis gas and ethylene as raw materials is a promising route as it has a simple and environmental friendly process[24].In addition,a lot of H2O and MMA ultimately generated make it an ideal postprocess for CCLG in the high atom economy view.

In this paper,a novel CCLG-MMA system aiming at‘‘energy saving and low emission”is proposed.The main steps of this proposal are shown in Fig.2.It is divided into three parts including MD simulation, steady-state simulation, and dynamics simulation.MD simulation is aimed to provide specific thermodynamic parameters for steady-state simulation of the CCLG-MMA system.Based on the results of steady-state simulation, a control scheme is proposed using dynamic simulation to ensure the high purity of MMA production under disturbances.The innovation of this paper lies in the exploration of MD mechanism analysis to guide process simulation.The main structure of this paper is as follows:the details of char and Fe2O3molecular models, the ReaxFF MD simulation, and the mechanism analysis are described in Section 2.The CCLG-MMA system including CCLG, Rectisol, and MMA is simulated in Section 3.A control scheme of CCLG-MMA system is established to verify the purity of MMA in face of disturbances in Section 4.Several important conclusions are summarized in the last section.

2.MD Research on CCLG Mechanism

MD simulation is an effective method for investigating reaction mechanism and intermolecular interaction in various chemical environments [25].The research of CCLG reaction mechanism includes four steps: the Fe2O3, H2O, and char molecular model building, energy and structure optimization of Fe2O3-H2O-Char system, ReaxFF MD simulation, and post processing of ReaxFF MD results.The procedures of MD research are shown in Fig.3.

Fig.2.The framework of the design method of the CCLG-MMA system.

2.1.Molecular configuration

Fe2O3is an ionic crystal belonging to B-3C space group with lattice parameters of lengths as a = b = 5.035 ?, c = 13.72 ? (1 ?=0.1 nm), angles in α = β = 90°,γ = 120°, which is a cuboid lattice with rhombic cross section [26].Above lattice parameters are used to construct an empty lattice, to which the coordinate information of Fe and O are imported.The structure of Fe2O3hexagonal crystal is shown in Figs.4 and 5.

Fig.3.The procedures of MD research.

Fig.4.The top view of Fe2O3 lattice.Iron atoms are colored purple, oxygen atoms are red.

Fig.5.The side view of Fe2O3 lattice.

Char is the pyrolysis product of coal and the intermediate product of CCLG [27].To better describe the reaction between C element and lattice oxygen in Fe2O3, char is chosen as a representative for coal to facilitate model construction and chemical property estimation[28].Three molecular scripts about carbon,oxygen,and hydrogen atoms are chosen as shown in Fig.6.

2.2.Molecular and system optimization

The 3D model of char scripts are also constructed on the basis of chemical formula.The carbon and oxygen skeletons are firstly built by sketch modeling, and the hydrogen atoms are automatically added afterward.Due to the instability of initial structure, the energy and geometry of char models need further be optimized.The algorithm is smart method, which is an integration of the steepest descent, ABNR, and quasi-Newton methods.The specific parameters are as follows: the convergence tolerance of energy is 0.00002 kcal·mol-1(1 cal=4.18 J), force is 0.001 kcal·mol-1·?-1(1 ?=0.1 nm), displacement is 0.00001 ?, the maximum number of iterations is 5000, the summation method is used for electrostatic interactions, and van der Waals interactions is atom based.The changes of energy in geometry optimization procedure are shown in Figs.7-9, where the potential energy of C19H30O is converged at 15.3 kcal·mol-1after 118 steps, the potential energy of C20H30O is converged at 69.37 kcal·mol-1after 143 steps, and the potential energy of C50H68O2is converged at 135 kcal·mol-1after 82 steps.The energies of char models before and after energy and geometry optimization are compared in Table 1.

After the geometric optimization, the optimal structure with a minimum potential energy is found by annealing algorithm.By periodically increasing and decreasing the temperature, the annealing operation can avoid the local energy minimum in finding the global minimum energy structure.In this task,the connectivity of atoms is constant.The specific parameters of anneal task are as follows: the temperature is increased first from 300 K to 1000 K and then decreased to 300 K, there are 2000 heating ramps with 200,000 steps in each cycle, and the cycle is repeated 5 times in total 1,000,000 steps.Take char model C50H68O2as an example,the temperature and energy in anneal task are shown in Figs.10 and 11.Five structures with different potential energy of C50H68O2are obtained by anneal task.The energy results are listed in Table 2.The anneal process for C19H30O and C20H30O are the same as C50H68O2, results listed in Tables 3 and 4 respectively.

Table 2 The C50H68O2 frame energy of anneal results

Table 3 The C19H30O frame energy of anneal results

Table 4 The C20H30O frame energy of anneal results

Fig.6.Chemical formula of char scripts C19H30O (a), C20H30O (b), and C50H68O2 (c).

Fig.7.The potential energies of C19H30O in energy optimization process.

Fig.8.The potential energies of C20H30O in energy optimization process.

Fig.9.The potential energies of C50H68O2 in energy optimization process.

Table 1 The energies of char models before and after energy and geometry optimization

The minimum potential energy structures of char models are thus obtained from anneal results, as shown in Fig.12, where carbon atoms are colored as grey,oxygen atoms are red,and hydrogen atoms are white.

Fig.10.The temperature of C50H68O2 in anneal task.

Fig.11.The pressure of C50H68O2 in anneal task.

Steam is an important gasification agent, which provides oxygen and hydrogen elements to react with carbon to produce CO and H2[29].Therefore, H2O molecular is also constructed in this work.The molecular of char, non-periodic Fe2O3, and H2O are put into a box of 45 ? × 45 ? × 45 ? size with a mole ratio of 3:1:25 and initial density of 0.297 g·cm-3.The box structure is shown in Fig.13.

After the molecular construction,the CLG box is relaxed in temperature and pressure terms.A volume and temperature constant(NVT)simulation with 100 picosecond(ps)and 300 K is performed with a time step of 1.0 fs and thermostat method of nose.The temperature fluctuates violently at the beginning and then tends to be converged, as shown in Fig.14.A pressure and temperature constant (NPT) simulation with 100 ps and 0.1 MPa is also performed with a time step of 1.0 fs and barostat of Berendsen.The density is increased at first and then gradually converged to 0.315 g·cm-3,as shown in Fig.15.The change of length when shrinking the box volume is shown in Fig.16, the length of the box is decreased at first and then gradually converged to 0.315 Angstrom.

Fig.15.The changes of system density during 100 ps NPT dynamics.

2.3.CLG mechanism investigation via ReaxFF MD

The ReaxFF MD simulation of CLG process is performed by using NVT ensemble.In this case, the temperature is increased from300 K to 1700 K in 100 ps, which is controlled by the Berendsen ensemble with a time step of 0.25 fs.The distribution of production species is then analyzed,as shown in Fig.17.It can be seen that the fragments of small molecule hydrocarbon are maintained at a low level.The piece number of H2O is decreased remarkably at first and converged gradually when the temperature is 1200 K.Conversely,the piece number of CO and H2are increased at first and finally converged when reaction temperatures are 900 K and 1200 K respectively.This may be due to the decomposition of H2O molecular and the breaking of C-H bond in char molecular.

Fig.12.The minimum potential energy structure of C19H30O (a), C20H30O (b), and C50H68O2 (c).

Fig.13.The 45 ? × 45 ? × 45 ? box of char-Fe2O3-H2O system.Iron atoms are colored purple, oxygen atoms are red, carbon atoms are grey, and hydrogen atoms are white.

Fig.14.The changes of system temperature during 100 ps NVT dynamics.

Fig.16.The length of the box in 100 ps NPT dynamics.

Fig.17.The number of species in 300 K to 1700 K ReaxFF MD simulation.

After above MD simulation,a conclusion can be drawn that the ideal reaction temperature is 1200 K, which is consistent with the experiment result [30].There is no obvious changes in number of pieces when the reaction temperature is higher than 1200 K.

3.Steady-state Simulation of CCLG-MMA System

From above discussions, the optimized parameters in CLG process with high purity of synthesis gas are obtained by ReaxFF MD simulation, providing a good guidance for CCLG-MMA system simulation.

The CCLG-MMA system includes three parts: crude synthesis gas production, synthesis gas purification, and MMA production.The steady-state simulation by Aspen Plus software is an appropriate approach to verify the feasibility of this process.The first and also the most basic step for steady-state simulation of CCLGMMA system is to define unconventional components such as char and ash in Aspen Physical Property Database[31].Peng Robinson-Boston Mathias (PR-BM) is chosen as property method, which has been widely used in gas-processing, refinery, and petrochemical situations [32].The simulation flow diagram of crude synthesis gas production is shown in Fig.18, in which char stream is the raw material as well as the main product of coal pyrolysis.In this process,char is decomposed into simple substances by Decompose module with premise of element balance to simplify the CLG reaction modeling.The reaction between coal and Fe2O3is complicated as it includes a variety of products and reactions,so the Reduction reactor is simulated using RGibbs module in Aspen plus, which uses Gibbs free energy minimization phase to correctly calculate reactions.The optimum reaction temperature of FR is given by ReaxFF MD simulation as 1200 K.After pyrolysis reaction,the produced synthesis gas composed of CO,H2,H2S,N2,unreacted water,and FeO leaves FR for further purification.

To reduce CO2emission, synthesis gas is separated from solid FeO and then reacted with coke oven gas (COG) in DMR reactor.As a substitute for natural gas, COG is a hydrogen-rich byproduct of coal coking process, which is mainly composed of 30% CH4and 60% H2[33].Dry methane reforming reactions are crucial for synthesis gas production, which are performed in DMR Reactor module with Eqs.(1) and (2) [34].For the highly stable and special properties requirement for methane reforming with CO2at high temperature,Ni/SBA-15 is selected as catalyst in DMR reactor with a voidage of 0.68.Besides,research has revealed that the impurities such as H2S in synthesis gas has no effect on DMR reaction[35],so impurity effect is ignored in this reaction.After FR,FeO is oxidized to Fe2O3by air (78% N2and 22% O2) in AR reactor.The simulation results of important streams in synthesis gas production part are listed in Table 5, and the specific data of reactors are listed in Table 6.

Table 5 Simulation results of important streams in crude synthesis gas production part

Table 6 Specific data of reactors in crude synthesis gas production part

Fig.18.The simulation flow diagram of crude synthesis gas production part.

The simulation flow diagram of synthesis gas purification is shown in Fig.19, in which the low temperature methanol is used to absorb H2S, CO2, and small molecular hydrocarbon by Rectisol process[36].The liquid methanol is refined by desorption and distillation processes, part of which is recycled as absorbent, and the remaining is used to produce MMA.After absorption,the synthesis gas that contains only N2,CO,and H2is delivered to next part.The simulation results of inlet and outlet streams in synthesis gas purify section are listed in Table 7.

Table 7 Simulation results of inlet and outlet streams in synthesis gas purification part

Fig.19.The simulation flow diagram of synthesis gas purification part.

The simulation flow diagram of MMA production is shown in Fig.20, including three reactions and their post treatment operations.The ethylene hydroformylation reaction is given in Eq.(3),the propionaldehyde condensation reaction reactions are shown in Eqs.(4) and (5), and the methyl acrolein methylation reaction reactions are shown in Eqs.(6)and(7).The specific data of reactors in MMA production part are listed in Table 8.

Table 8 Specific data of reactors in MMA production part

In the MMA process, the synthesis gas is mixed and then reacted with ethylene in R01 (Hydroformylation reactor) under 373.15 K and 3 MPa[37].As the boiling points of ethylene and propionaldehyde are very close, the common distillation operation is unable to meet the separation requirements.Extraction distillation is therefore used to separate ethylene from propionaldehyde,where formaldehyde is used as extractant since it is also the rawmaterial of propionaldehyde condensation reaction.After distillation, the unreacted ethylene is recycled to R01.Propionaldehyde and formaldehyde in bottom stream of T0301 (Hydroformylation product separating column) are delivered to R02 (Condensation reactor).The reaction temperature and pressure of Condensation reactor are 323.15 K and 4 MPa [38].Purified methyl acrolein is reacted with methanol and oxygen in R03 (Methylation reactor)under 323.15 K and 0.1 MPa.The Extractant module is used to separate MMA and methanol azeotropic system with water as extractant.At last, the 99.9%(mass) MMA product is obtained from extract phase of distillation process.

The simulation results of main streams in MMA process are listed in Table 9.It can be seen that the mass fraction of CO2in SYNGAS-1, SYNGAS-2, and Cleaner Gas streams are 17.47%,0.25%, and 0.02% respectively, which meet the requirement of CO2reduction standard [39].The mass fraction of MMA stream is 99.9% and the molecular fraction of H2and CO in Cleaner Gas stream are 45.89% and 53.27%, which manifest the high quality of MMA product and synthesis gas in CCLG-MMA system.The amount of CO2emission after CCLG process, DMR reaction, and Rectisol process is 0.0028 kg CO2·(kg MMA)-1.The energy consumption of utility in each part of CCLG-MMA system are listed in Table 10.The total consumption is 45521 kJ·(kg MMA)-1, and the consumption of MMA production part is 25293 kJ·(kg MMA)-1,showing that the energy savings of CCLG-MMA is 27.4%compared to conventional process with 34839 kJ·(kg MMA)-1.

Table 9 Simulation results of important streams in CCLG-MMA system

Table 10 Energy consumptions of each part in CCLG-MMA system

Fig.20.The simulation flow diagram of MMA production part.

Based on above steady-state simulations,the influence of tower operating parameters and the mass flow of extractant on the purity of MMA are further analyzed to obtain the optimal value of these key parameters.The effect of extractant mass flow on MMA production quality is shown in Fig.21, where the mass fraction of MMA increases significantly with the increase of extractant mass flow.The MMA stream achieves the required purity of 99.9%when the mass flow of extractant is 5700 kg·h-1.The effect of distillation reflux ratio on MMA product quality is shown in Fig.22, showing that the mass fraction of MMA is qualified when the mole reflux ratio is 2.2.The effect of total distillation stage on MMA product quality is shown in Fig.23, showing that the mass fraction of MMA is qualified when the total stage number is 20.The effect of distillation feed stage on MMA product quality is shown in Fig.24, showing that the mass fraction of MMA reaches its zenith when feed stage is 7.Therefore, the optimum parameters are obtained by above sensitivity analysis as follows: the mass flow of extractant is 5700 kg·h-1, the mole reflux ratio of T0305 is 2.2,the total stage number of T0305 is 20, and the feed stage is 7.

Fig.21.The effect of extractant mass flow on MMA quality.

Finally,the mass balance from the steady state simulation with above optimum parameters is calculated and listed in Table 11.It can be seen that the total mass flow of inlet streams in three parts are equaled with the total mass flow of outlet streams.

4.Plant Wide Control Design Using Dynamic Simulation

Fig.22.The effect of distillation column reflux ratio on MMA quality.

Fig.23.The effect of distillation column total stage on MMA quality.

Fig.24.The effect of distillation column feed stage on MMA quality.

For controlling the quality of MMA production, the dynamic simulation is performed on the basis of steady-state simulation of MMA production in this paper.Dynamic simulation is widely used in chemical industry, petrochemical process, and hydropurification process for operation optimization and safety analysis[40,41].Dynamic simulation is coupled with various subjects such as control theory, process system engineering, thermodynamic model, chemical engineering, and dynamic data processing [42],so it can provide comprehensive information about state transition of chemical process.In dynamic simulation, the MMA process is simplified reasonably to reduce the unnecessary computational load.Additionally, the parameters including dimensions and pressure drop of devices are provided to into simulation system, as listed in Table 12.Afterwards,a control scheme is designed to keep the production purity of MMA and the stability of CCLG-MMA system.Table 13 lists all controllers and their information.

Table 11 Mass balance calculation results of CCLG-MMA system

Table 12 The dimensions and pressure drops of all devices

Table 13 Controller parameters

The dynamic model of MMA producing section is established through Aspen Dynamics software, which diagram is shown in Fig.25.The stability of system under the control scheme is verified by adding reboiler load and reactor temperature disturbances.

Fig.25.The dynamics simulation diagram of MMA production part.

The reboiler duty of T02 is selected as disturbance to verify the stability of MMA production at first.The changes of temperature and pressure of T02 top stream after increasing reboiler duty by 10% at 2 h are shown in Fig.26.The pressure fluctuates at first and then approaches a stable value of 0.109 MPa,and the temperature has a similar trend with a stable value of 360 K.Secondly,the temperature of methyl acrolein methylation reactor has a great influence on the MMA producing process, thus the reaction temperature is taken as another example to test the dynamic behavior of reactor.The change of the mass fraction of MMA in product after reducing reactor temperature by 10% at 2 h is shown in Fig.27.It can be seen that the mass fraction of MMA in product is decreased at first and then increased to stable value of 0.99896.The results in above two cases thus prove the stability and feasibility of the control scheme depicted in Fig.25.

Fig.26.Effect of increase reboiler duty at 2 h on the pressure (a) and temperature (b) of T0305 top.

Fig.27.Effect of increase reactor temperature at 2 h on the mass fraction of MMA.

5.Conclusions

In this paper,a novel CCLG-MMA system is designed,which couples CCLG and DMR process to reduce the emission of CO2.Several methods including MD,steady-state simulation,and dynamics simulation are performed in the investigation of CCLG-MMA system.The reaction mechanism of CCLG is analyzed by ReaxFF MD simulation, with results showing that the optimum temperature of CLG process is 1200 K.The steady-state simulation of CCLG-MMA system is carried out for crude synthesis gas production, synthesis gas purification, and MMA production.The results show that the CO2molecular fraction and molecular flow of synthesis gas after CLG and DMR processes are 0.25%and 0.084 kmol·h-1respectively,and the mass fraction of MMA stream is 99.9%.The amount of CO2emission after CCLG process,DMR reaction,and Rectisol process is 0.0028 kg CO2·(kg MMA)-1,which meets the aim of low CO2emission.The total energy consumption is 45521 kJ·(kg MMA)-1and the energy savings of CCLG-MMA is 27.4% compared to conventional process.The optimum parameters are also obtained by sensitivity analysis as follows:the mass flow of extractant is 5700 kg·h-1,the mole reflux ratio of T0305 is 2.2,the total stage number of T0305 is 20,and the feed stage is 7.A control scheme is finally designed and its stability is verified by adding reactor temperature and reboiler load disturbances.Dynamic simulation results show that the mass fraction of MMA is stable in the face of 10%disturbance.

The contribution of this CCLG-MMA system consists in the exploration of MD mechanism analysis based process simulation.The conclusion observed in this research suggests the potential application of cleaner production to more real chemical processes.This work will take into consideration the heat integration issue in the near future.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors gratefully acknowledge that this work is supported by the National Natural Science Foundation of China (21576143)and Foundation of State Key Laboratory of High-efficiency Utilization of Coal and Green Chemical Engineering (2020-KF-13).