Jian Song,Claudia Li,Shao Zhang,Xiuxia Meng,Bo Meng,Jaka Sunarso
1 Department of Chemical Engineering,Shandong University of Technology,Zibo 255049,China
2 State Key Laboratory of Separation Membranes and Membrane Processes,Department of Chemical Engineering,Tiangong University,Tianjin 300387,China
3 Research Centre for Sustainable Technologies,Faculty of Engineering,Computing and Science,Swinburne University of Technology,Jalan Simpang Tiga,93350 Kuching,Sarawak,Malaysia
Keywords:Hollow fiber Membrane reactor Oxidative coupling of methane (OCM)Perovskite
ABSTRACT The catalytic oxidative coupling of methane(OCM)to C2 hydrocarbons(C2H6 and C2H4)represents one of the most effective ways to convert natural gas to more useful products,which can be performed effectively using La0.6Sr0.4Co0.2Fe0.8O3-δ (LSCF) perovskite hollow fiber membrane microreactor.In this work,the effects of adding a thin BaCe0.8Gd0.2O3-δ (BCG) catalyst film onto the inner LSCF fiber surface as the OCM catalyst and a porous Ba0.5Sr0.5Co0.8Fe0.2O3-δ(BSCF)perovskite layer onto the outer LSCF surface to improve the oxygen permeation were evaluated.Between 700 °C and 1000 °C,methane conversion increased in the order of uncoated,BCG and BSCF-coated,and BCG-coated LSCF hollow fiber while C2-selectivity and C2-yield increased in the order of BCG and BSCF-coated,uncoated,and BCG-coated LSCF hollow fiber.Oxygen permeation flux at the same temperature range,on the other hand,was enhanced in the order of uncoated,BCG-coated,and BCG and BSCF-coated LSCF hollow fiber.This finding demonstrates the complex interplay between oxygen permeation and OCM performance.The BCG and BSCFcoated hollow fiber was also subjected to thermal cycles between 850°C and 900°C for up to 175 hours during which the fiber showed minor degradation in oxygen permeation fluxes and relatively stable OCM performance.
Oxidative coupling of methane(OCM)serves as an economically promising route to directly obtain value-added C2hydrocarbons(ethane and ethylene)from natural gas feedstock.The large potential of OCM to become a key technology for the petrochemical and chemical industries has attracted considerable research interest over the past few decades [1].Presently,the large-scale industrial commercialization of the OCM process is hindered by the low C2yield that is typical for this conversion-selectivity traded-off reaction,whereby an increase in the O2feed will lead to high CH4conversion but with poor product selectivity.The yield of the desired C2products is diminished by the non-selective reaction of methyl radicals with the catalyst surface and gaseous O2,ultimately leading to higher yield of undesirable combustion products (COx,i.e.,CO and CO2) [2].It has been estimated that a single-pass conversion of 35%–37%and selectivity of 85%–88%,which is equivalent to a C2yield of more than 30%,is required to attain commercial competitiveness for OCM[3].Considerable efforts have been made to improve the product yields of OCM catalysts up to a commercially feasible level.The general consensus is that basic metal oxides promoted by alkali and alkaline-earth metals show an acceptable performance as OCM catalysts [1].The use of oxide catalyst in OCM follows a unique heterogeneous-homogeneous reaction mechanism,i.e.,methyl radicals are generated on the solid surface and coupled to form gaseous C2hydrocarbons.The carboncontaining species are then oxidized into CO2mostly in the gas phase,and possibly also on the catalyst surface.Consequently,the per-pass C2yield on all the reported catalysts in conventional packed-bed reactors have never exceeded 25% [4].
One of the proposed strategies to achieve high C2yield is by using membrane reactors instead of conventional packed-bed reactors [3].In membrane reactors,the oxygen is supplied to the reactor in a controlled manner,allowing the local methane-tooxygen ratio in the reaction zone to be kept at a minimum,and thus a much higher C2-selectivity relative to the conventional reactors can be attained.Higher CH4conversion can also be achieved by operating the reactor at a relatively higher overall oxygen-tomethane ratio than that of a conventional co-feed reactor.Moreover,the controlled delivery of O2can mitigate thermal hotspots for the highly exothermic reaction of OCM.Dense membrane reactors,in particular,demonstrate 100% selectivity towards oxygen permeation and can prevent loss of CH4due to back-permeation when used for OCM [5,6].Furthermore,air can be used as the O2source without the need of downstream N2separation from the product stream and the process can be operated in a safe mode,leading to reduced operation costs [7].More importantly,the dense mixed ionic-electronic conducting (MIEC) membranes supply O2in a form that circumvents the gas phase and forms active oxygen species,hence the undesired gas-phase reactions of oxygen with CH4or its intermediates can be greatly reduced[8].However,due to the low oxygen concentration in the reaction side,the reaction rate is slow and a long contact time is required for high CH4conversion[1].Table 1 lists several dense ceramic membrane compositions that were reported in experimental studies on OCM.The results indicate large variations between the different studies with the per-pass C2product yields ranging between 3%–20% in most studies,which is far less than the desired threshold for commercial consideration.The barrier to obtain a higher C2product yield in these membrane reactors can be attributed to the inherent problems pertaining to poor membrane surface catalytic properties and low oxygen fluxes in the OCM reaction temperature range,i.e.700–1000 °C,and unfavorable reactor configuration [9].
The membrane catalytic activity (in terms of both yield and selectivity) is of crucial importance for the OCM reaction in dense ceramic membrane reactors [10].Zeng and Lin [9] carried out the OCM reaction in a Bi1.5Y0.3Sm0.2O3-δtubular membrane reactor and reported the highest C2yield of 35%,which was attributed to the high catalytic activity and selectivity of the δ-phase Bi2O3-based membrane materials for OCM.However,it is difficult for a membrane composition to simultaneously possess both high catalytic activity and high oxygen permeability.Although the reaction can be improved by using a highly efficient OCM catalyst packed in the reactor,a more effective strategy is to coat the catalyst on the membrane surface with high oxygen permeability.The presence of the oxide ions on the catalyst surface will help to promote the generation of C2products.Ideally,gaseous O2should not be present to prevent any undesirable reactions with the reactant and product to maximize the selectivity for desired products [11].Furthermore,the exposure of catalytically modified membrane surfaces can also be maximized in this case for higher CH4conversion.
During the membrane reactor operation for OCM,the rate of O2permeation needs to conform to the rate of O2consumption to attain high C2product yield[7,12].When there is insufficient oxygen supply,comparably poor conversion are observed,while higher oxygen fluxes lead to increased CH4conversion,especially in the presence of an efficient catalyst.In fact,a high C2yield can only be achieved when the oxygen permeation flux,the methane flow rate,and the intrinsic reaction rates match well [10].Among the commonly studied membrane configurations,hollow fiber membranes fabricated via the phase inversion/sintering technique are suited to obtain high oxygen permeation fluxes given their asymmetric structure (i.e.,a thin dense and separating layer integrated with a porous substrate of the same material)[13–17].Furthermore,the large membrane surface area to reactor volume ratios of these hollow fiber membranes favor the C2yield improvement in OCM [18,19].
Overall,La0.6Sr0.4Co0.2Fe0.8O3-δ(LSCF) perovskite hollow fiber membranes appear to consistently produce high C2yield and selectivity(Table 1).To further enhance their OCM performance,a thin BaCe0.8Gd0.2O3-δ(BCG) perovskite film can be coated onto the inner surface as the OCM catalyst for LSCF hollow fiber [20].Likewise,to improve the oxygen permeation flux performance,a porous Ba0.5Sr0.5Co0.8Fe0.2O3-δ(BSCF) perovskite layer can be coated onto the outer surface of the LSCF hollow fiber [21,22].Three different LSCF hollow fibers were thus synthesized here,i.e.,LSCF hollow fiber,BCG-coated LSCF hollow fiber,and BCG and BSCF-coated hollow fiber.This work aims to compare the oxygen fluxes and OCM performances of these 3 hollow fibers to evaluate the relationship between the oxygen flux and OCM performance.
Table 1 Dense MIEC ceramic membrane reactors used in oxidative coupling of methane (OCM)
In this work,La0.6Sr0.4Co0.2Fe0.8O3-δ(LSCF)perovskite was used as the membrane material with BaCe0.8Gd0.2O3-δ(BCG)as the OCM catalyst,and Ba0.6Sr0.4Co0.2Fe0.8O3-δ(BSCF) as the oxygen permeation promoter.All composite oxides were prepared via a microwave-assisted sol–gel combustion method as described elsewhere [33].Metal ions were sourced from nitrates such as La(NO3)3·6H2O,Sr(NO3)2,Ba(NO3)2,Co(NO3)3·6H2O,Fe(NO3)3·9H2O,Ce(NO3)3·6H2O,and Gd(NO3)3·6H2O.Glycine (NH2CH2COOH,99.5% purity) and polyethylene glycol (PEG-2000,99.5% purity)were used as the chelating agent and the dispersant,respectively.The composite oxide precursors were calcined at 800°C in static air for 4 h for LSCF and at 1000 °C for 2 h for BSCF and BCG in a boxtype furnace to remove the residual carbon and to attain the desired perovskite structure.
The LSCF hollow fiber membranes were prepared using the phase inversion/sintering technique with the detailed preparation procedures as described elsewhere [13].The calcined LSCF powders were ball milled for 48 h and then sieved through a 200-mesh sifter to remove agglomerates.The spinning suspension used in this study consisted of 64.33%(mass)LSCF powder,7.51%(mass)polyethersulfone (PESf),and 28.16% (mass) 1-methyl-2-pyrrolidinone (NMP).A spinneret with the orifice diameter/inner diameter of 3.0 mm/1.5 mm was used to spin the hollow fiber precursors.De-ionized (DI) water and tap water were used as the internal and the external coagulants,respectively.After drying and straightening,the hollow fiber precursors were sintered at 1300 °C for 4 h to form dense hollow fiber membranes.The gastightness of the hollow fibers were measured using an N2gas permeation test prior to subsequent surface coating and oxygen permeation tests [14].
The calcined BCG perovskite powders were ball milled in ethanol using agate balls in an agate bottle for 24 h.PVP-K30(30 mg·g-1perovskite) and polyvinyl butyral (40 mg·g-1perovskite) were mixed in as additives to form the BCG slurry.The homogenous BCG slurry was injected into the hollow fiber lumen to form the OCM catalytic layer and was purged using N2sweep gas to remove the excess solution.The coated fibers were then fired at 600°C for 1 h to remove the organics,and later sintered at 1100 °C in stagnant air for 3 h with heating and cooling rates of 2–3 °C·min-1to integrate the coating layer onto the membrane surface.Similarly,a BSCF slurry was prepared and coated on the outer surface of the hollow fibers by brushing and subsequent sintering.To obtain a thicker perovskite layer,the coating/sintering steps were repeated for 2–3 times.
To form the membrane reactor for OCM experiments,the 260 mm-length catalyst-modified LSCF hollow fiber membrane was assembled in a quartz tube with the dimensions of 10/8 mm in external/internal diameter and 30 cm in length.A flexible silicone rubber tube was used on one end of the fiber to offset the thermal expansion mismatch between the hollow fiber membrane and the module shell at high temperatures,as shown in Fig.1.The sealing was achieved by using an organic sealant.A K-type thermocouple was positioned close to the center of the hollow fiber to measure the temperature during operation.The membrane reactor was placed in a custom-made tubular furnace with a heating length of 5 cm.During operation,the CH4-Ar mixture(31 m·min-1,with volumetric ratio of 11:20) and air (100 mL·min-1) was introduced in co-current configuration into the lumen and the shell side of the reactor,respectively.The gas flow rates were controlled using mass flow controllers (D07-7B,Beijing Sevenstar Electronics Co.,Ltd.,China),which were calibrated using a soap bubble flow meter that was also used to measure the effluent gas flow rates.All gas flows were quoted at standard temperature and pressure(STP).Compositions of the products were detected using a gas chromatograph (Agilent 6890 N) equipped with 5A carbon molecular sieve column and thermal conductivity detector (TCD) and flame ionization detector (FID).Three analyses were conducted at each specified experimental condition and the average value was calculated.The methane conversion (XCH4),species selectivity(SC2),and yield (YC2) were calculated using Eqs.(1)–(3).
where Finand Foutare the flow rates of methane feed (methaneargon mixture) and product stream,respectively;yfand xC2are the methane concentration in feed and the C2hydrocarbon concentration (C2H4+C2H6) in the product stream,respectively.
Fig.1.Experimental setup for OCM reaction.
The oxygen permeability of the LSCF hollow fiber membrane was also evaluated using the same setup.High purity argon was used as the carrier gas in the fiber lumen with a flow rate of 30 ml·min-1while air was passed in the shell side at a 100 ml·min-1flow rate.The oxygen permeation fluxwas calculated based on the change in oxygen concentration in the air stream (Eq.(4)).
where FAirandis the air feed and outlet flow rate,ml·min-1;andis the feed and outlet molar fraction of oxygen in the air stream,respectively;Amis the effective membrane area calculated by Eq.(5) in which L is the effective length for oxygen permeation of the hollow fiber membrane.The heating length of the furnace(5 cm)was taken as L in this work.D and d are the outer and inner diameter of the fiber sample,respectively,in cm,as obtained from the SEM micrographs.
The crystal structure of the powders was identified by powder X-ray diffraction (XRD,Bruker D8 Advance,Germany) using Cu-Kα radiation.Continuous scan mode was used to collect 2θ data between 20°and 80°with a 0.02°sampling pitch and a 2(°)·min-1scan rate.The X-ray tube voltage and current were set at 40 kV and 40 mA,respectively.The structure and morphology of the hollow fiber membrane were observed using a scanning electron microscope (SEM,Hitachi S-4800,Japan) on samples coated with gold sputter.
The powder XRD patterns of the calcined LSCF,BSCF,and BCG powders,as well as the crushed powder from the sintered LSCF hollow fiber membrane are displayed in Fig.2.The perovskite phase of the LSCF powders calcined at 800 °C for 4 h (Fig.2(a)) is characterized by the perovskite peaks of cubic La0.6Sr0.4Co0.9Fe0.1-O3-δ(PDF#01-089-5720) with Pm3m space group.In comparison,the XRD analysis reveals no further changes in the LSCF crystalline structure following sintering of the hollow fiber membranes at 1300 °C for 4 h (Fig.2(b)),indicating proper preservation of the perovskite structure during the spinning and sintering process.However,the intensity of the corresponding characteristic peaks of the perovskite phase in the sintered LSCF hollow fibers are slightly higher than those in the calcined LSCF powders,which suggests that the crystal size in the hollow fibers has become larger due to the high-temperature sintering.The 1000 °C-calcined BSCF powder also exhibited pure perovskite structure(Fig.2(c)).Due to the larger ionic radius of Ba2+(0.161 nm) compared to La3+(0.136 nm),the lattice expansion of the BSCF perovskite cells translates to the slight shift in the characteristic peaks to lower angle values [34].Additionally,the powder XRD pattern of calcined BCG powders in Fig.2(d)displays no secondary phases and reveals an orthorhombic crystal structure with a Pbnm space group(PDF#01-070-1429),similar to the other Gd-doped BaCeO3composites [35,36].
Fig.2.Powder XRD patterns of the (a) calcined LSCF powder,(b) sintered LSCF membrane,(c) calcined BSCF powder,and (d) calcined BCG powders.
The morphology of the asymmetric LSCF hollow fiber membrane coated with BCG inner layer and BSCF outer layer is displayed in the SEM micrograph in Fig.3.The outer and inner diameter of the hollow fiber membrane is 1.92 mm and 1.18 mm,respectively (Fig.3(a)).Despite the presence of numerous pores on the inner and outer membrane walls,the dense middle layer of the membrane ensures the gas tightness of the membrane,as confirmed by the nitrogen permeation test.The dense layer is sandwiched between a~40 μm-thick BCG inner layer and a~45 μm-thick BSCF outer layer (Fig.3(b)).Prior to applying the BCG and BSCF coatings,the inner surface of the dense LSCF membrane was found to have some micropores (Fig.3(c))while the outer membrane surface was dense and smooth (Fig.3(d)).Fig.3(e) and (f) display the fiber membrane inner and outer surface after loading with BCG and BSCF particles,respectively.Several discrete holes on the surface of the BCG and BSCF layers were produced through the combustion of organic additives during calcination,which can increase the surface area for oxygen permeation effectively.
The oxygen permeability of the uncoated LSCF hollow fiber membrane was first measured using air as the feed gas(100 mL·min-1) in the shell side of the membrane and Ar as the sweep gas (30 mL·min-1) in the fiber lumen.The oxygen permeation flux as well as the oxygen concentration in the permeate side are shown in Fig.4.As expected,the oxygen permeation rate increases with increasing temperature since both the surface exchange reactions and the ionic bulk diffusion are temperatureactivated.This leads to an increase in the oxygen concentration in the permeate stream,as shown in Fig.4,implying that the operating temperature plays a more significant role than the driving force in oxygen permeation through the LSCF perovskite membrane.The permeation equation through the LSCF hollow fiber membrane was obtained by fitting the experimental data using Eq.(6).
Fig.3.SEM micrographs of the LSCF hollow fiber membranes.Cross section of (a)the uncoated hollow fiber and(b) the coated hollow fiber;(c) Inner surface and (d)outer surface of the uncoated hollow fiber;(e) BCG coating on the inner surface;(f) BSCF coating on the outer surface.
Fig.4.Oxygen permeation through the uncoated LSCF hollow fiber membrane under air feed and Ar sweep between 700 °C and 1000 °C.
where the activation energy for oxygen permeation,Eawas regressed from the Arrhenius equation as 88.012 kJ·mol-1,which is close to the reported literature values for LSCF [37,38].
Following the oxygen permeation measurements,the OCM performance of the uncoated LSCF hollow fiber membrane was tested in comparison to the LSCF hollow fiber coated with BCG inner layer and the LSCF hollow fiber coated with BCG inner layer and BSCF outer layer.CH4-Ar gas mixture was fed at 31 mL·min-1flow rate with volumetric ratio of 11:20 in the hollow fiber membrane lumen while air passed at 100 mL·min-1flow rate in the membrane shell side.The OCM performances of the uncoated,BCGcoated,and BCG and BSCF-coated LSCF hollow fibers are displayed in Fig.5.Fig.5(a) depicts the methane conversion,C2-selectivity,and C2-yield as a function of temperature for the three hollow fiber membrane reactors.The methane conversion and C2-yield of the uncoated LSCF membrane increase with an increase in the operating temperature,whereas the C2-selectivity exhibits a maximum value of 30.77% at 850 °C,similar to a previous report on LSCF membrane [26].With the exception of C2-selectivity and C2-yield trends above 900 °C,methane conversion,C2-selectivity,and C2-yield of the BCG-coated hollow fiber are improved relative to the uncoated hollow fiber due to the presence of the BCG catalysts on the inner surface,which introduces higher methane catalytic activity relative to the uncoated membrane.Coating with BCG and BSCF at the inner and outer surface of the hollow fiber,respectively,on the other hand,serves to lower the methane conversion,C2-selectivity,and C2-yield relative to BCG-coated hollow fiber.While the methane conversion for BCG and BSCF-coated hollow fiber lies between those of uncoated and BCG-coated hollow fibers,the C2-selectivity and C2-yield generally increase in the order of BCG and BSCF-coated,uncoated,and BCG-coated hollow fiber.
Fig.5.(a)OCM performance,(b)oxygen permeation flux,and concentration of(c)CO2 and(d)other products in the permeate stream of the uncoated,BCG-coated,and BCG and BSCF-coated LSCF hollow fiber membrane reactors between 700 and 1000 °C.
As the temperature rises,the oxygen flux through the LSCF hollow fiber increases,leading to a relative excess of gaseous oxygen in the fiber lumen side(Fig.5(b)).Further inspection of the oxygen flux profiles in Fig.5(b) reveals that at the same temperature,the oxygen permeation rate through the catalyst coated membrane reactor is higher than that through the uncoated membrane.The presence of inner BCG coating and outer BSCF coating clearly leads to the higher oxygen permeation fluxes relative to the uncoated hollow fiber due to the improved surface reaction rates on the inner and outer surfaces.With rising temperature,the enhanced reaction with methane will consume more oxygen and further decrease the oxygen partial pressure on the methane side.Overall,the increasing oxygen permeation rates in the coated and uncoated membrane reactors is due to the combined effects of temperature on defect diffusion,surface reaction rate,and oxygen permeation driving force [39].However,the presence of the excess gaseous oxygen further converts the produced ethane and ethylene into COxat high temperature.Consequently,the C2-selectivity of the coated membrane decreases greatly due to the excessive oxygen permeation[7].The highest O2fluxes for BCG and BSCF-coated hollow fiber relative to BCG-coated and uncoated hollow fibers translates to its lowest C2-selectivity and yield (Fig.5(a) and Fig.5(b)).These observations highlight the presence of a fine trade-off between the oxygen permeation and OCM performance.
Fig.5(c) and (d) display the compositions of the product streams for the uncoated,BCG-coated,and BCG and BSCF-coated LSCF hollow fibers between 700°C and 1000°C.The water concentration is not presented in this work due to the unavailability of GC for water measurement.Interestingly,O2is observed to coexist with H2,CO,and C2-hydrocarbons,implying that the reaction rate on the membrane surface is not fast enough to completely consume the oxygen permeated through the membrane.The concentrations of CO,CO2,H2,and C2H6generally increase with an increase in temperature.On the other hand,the O2concentration in the product stream for uncoated membrane case demonstrates a two-stage increasing trend as a function of temperature,i.e.,from 0.44%at 700°C to 0.92%at 850°C,and then from 0.67%at 900°C to 1.04% at 1000 °C,respectively,implying that higher temperature range (900–1000 °C) promotes the reaction rate more noticeably compared to between 850–900 °C.The dip in O2concentration between 850 °C and 900 °C,which is also present in the BCGcoated LSCF hollow fiber does not occur in the BCG and BSCFcoated LSCF hollow fiber since the O2permeation rate is promoted by BSCF catalyst on the outer membrane surface.The concentration of ethylene peaks at between 900 °C and 950 °C,indicating that the oxidation of ethylene is more prominent than the ethylene production rate from ethane above this temperature range.These results support the gaseous radical reaction mechanism [9,10],whereby methane is first adsorbed and reacted with the lattice oxygen () to form methyl radicals,which are then coupled in gas phase into ethane or further react with gaseous oxygen to form carbon oxides.Ethylene is formed by further dehydrogenation of ethane on the membrane surface.In contrast,the concentration of CO2in the coated LSCF hollow fiber membrane reactors is generally higher relative to the uncoated membrane,which is attributed to the promotion of oxygen permeation.Accordingly,the concentration of O2in the exhaust gas for the coated membrane is promoted.This means that between 700 °C and 1000 °C,the increase in temperature promotes the oxygen permeation rate more significantly compared to the reaction rate.At the same time,C2H4attains a maximum concentration at 850–950 °C,which is consistent with the results of the uncoated hollow fiber membrane reactor.
To properly function as a high temperature OCM reactor,strong stability is required to sustain alternate heating and cooling cycles.Fig.6(a) shows the consistent flux reproducibility during heating and cooling between 850°C and 900°C for up to 175 hours,corresponding to 5 thermal cycles.The oxygen permeation flux gradually decreases from 1.5 mL·cm-2min-1to 1.24 mL·cm-2min-1at 850 °C during the first 72 h.When the temperature is increased to 900 °C,the oxygen permeation flux is boosted up to 2.9 mL·cm-2·min-1and gradually decreases to 2.1 mL·cm-2·min-1after 12 h.When the temperature is adjusted back to 850°C again,the oxygen flux is unable to return to the initial level in the initial stage and continues to decay slowly for 12 hours.After 2 cycles,i.e.,by the 110-h mark,stable and reproducible permeation fluxes between 850°C and 900°C are attained.This rules out the possibility of the porous layer deteriorating at high temperatures,indicating the stability of the membrane reactor.
Fig.6(b) depicts the stability of methane catalysis performance of the coated LSCF hollow fiber membrane reactor under thermal cycling.At the initial stage of 850 °C,the methane conversion first increases and then decreases with time,while the C2-selectivity displays the opposite trend,i.e.,the C2-selectivity decreases first and then increases.Consequently,the C2-yield remains in a relatively stable state.This indicates that the stability of the methane conversion and C2-selectivity of the coated hollow fiber membrane reactor in the initial stage of the reaction still needs to be improved.When the operating temperature is raised to 900 °C,the methane conversion is significantly increased to a higher level while the corresponding C2-selectivity is reduced.Compared with 850 °C case,the C2-yield is slightly improved,and for the next 12 hours,it remains almost stable.The same trend is observed throughout the 5 thermal cycles,i.e.,at 900 °C,the methane conversion gradually increases,while the C2-selectivity gradually decreases;and when the temperature drops to 850 °C,the methane conversion gradually decreases,while the C2-selectivity gradually increases.The C2-yield remains relatively stable even with the increase in the number of cycles.This indicates that C2can be generated in consistent manner from the fabricated hollow fiber membrane reactor.
Fig.6.(a) Oxygen permeation stability and (b) OCM catalytic performance over 5 thermal cycles within 175 h in the BSCF and BCG coated LSCF hollow fiber membrane reactor.
La0.6Sr0.4Co0.2Fe0.8O3-δ(LSCF) membrane reactor generally possesses good C2-selectivity and oxygen permeation rate,but the conversion of methane is low,which makes pure LSCF unsuited for application in oxidative coupling of methane(OCM).Three different fibers,i.e.,uncoated,BCG-coated,and BCG and BSCF-coated LSCF hollow fibers were made.At temperature range of 700–1000 °C,BCG-coated LSCF hollow fiber exhibited the highest methane conversion,C2-selectivity,and C2-yield among the three fibers while BCG and BSCF-coated LSCF hollow fiber exhibited the highest O2fluxes but the lowest C2-selectivity and C2-yield.Enhancing the OCM reaction by adding OCM catalyst layer (BCG)onto the inner surface of LSCF hollow fiber presented more significant effect towards improving C2-selectvity and C2-yield compared to improving the O2flux by adding porous BSCF layer onto the outer surface,highlighting the complex role that O2flux plays in OCM reaction.The BCG and BSCF-coated LSCF hollow fiber membrane was also able to demonstrate consistent flux reproducibility and OCM performance during heating and cooling between 850 and 900°C for up to 175 hours,corresponding to 5 thermal cycles.
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 the research funding provided by the National Natural Science Foundation of China(21805206,22179073).
Chinese Journal of Chemical Engineering2022年1期