Effect of the interactions between volatiles

合集下载

跨音速涡轮定常与非定常叶尖泄漏流机理分析

跨音速涡轮定常与非定常叶尖泄漏流机理分析

第63卷,2()21年第2期 Vol.63, 2021,No.2Chinese Journal of Turbomachinery跨音速涡轮定常与非定常叶尖泄漏流机理分析**基金项目:国家自然科学基金资助项目(No.51836008)谭 炜" 张子卿1'2屈 骁12张英杰12卢新根1张燕峰1(1.中国科学院工程热物理研究所轻型动力实验室;2冲国科学院大学)摘要:针对格拉兹大学单级跨音涡轮流场进行数值模拟并分析其叶尖泄漏涡结构与非定常影响机理。

结果表 明:泄漏涡于破碎前为类Rankine 涡结构,破碎后卷吸入间隙流体,同时受涡核卷吸而围绕其运动的部分间隙流与 主流掺混,形成高剪切流区域,形成类Sullivan 涡结构的亚临界状态。

非定常尾迹含有的流向涡量与叶尖泄漏涡- 刮削涡涡对相互作用下,引起叶尖泄漏涡周向位置与尺度发生波动;跨音涡轮中非定常的动、静叶排位势场作用导致激波与波后静压场分别对泄漏涡的位置与尺度产生影响。

在动、静叶栅距比为1.5时,尾迹与位势场的非定常 作用同样重要,导致泄漏涡损失加大。

未来设计高压涡轮需要考虑非定常的相互作用以控制叶尖泄漏损失。

关键词:叶尖泄漏涡;涡破碎;Sullivan 涡;尾迹;尾缘激波;静压场中图分类号:TK14;TK05 文章编号:1006-8155-(2021)02-0001-12文献标志码:ADOI : 10.16492/j.fjjs.2021.02.0001Mechanism Analysis of Steady and Unsteady Tip-LeakageFlow of Transonic TurbineWei Tan 1,2 Zi-qing Zhang 1,2 Xiao Qu 1,2 Ying-jie Zhang 1,2 Xin-gen Lu 1 Yan-feng Zhang 1(1. Laboratory of Light-duty Gas-turbine, Institute of Engineering Thermophysics,2. University of Chinese Academy of Science)Abstract : The flow field of single-stage transonic turbine of University of Graz is numerical simulated, the structure of tipleakage vortex and the influence of unsteadiness on tip leakage vortex is carried on. The research conclusion shows that the leakage vortex is Rankine-like vortex before break - down, it inhales the leakage air and changes into subcritical state Sulli -van-like vortex after break-down. Unsteady wakes influence the fluctuation size and the circumferential position of the vortex core by the interaction between tip leakage flow and streamwise vortex. Shock-wave and static pressure field has effect on theposition and size of tip leakage which caused by interaction of unsteady rotor blade and static blade in transonic turbine. Forthis transonic turbine stage, the unsteady effect of wakes and the potential field are of equal importance under the condition of the pitch ratio of 1.5 which increases mixing loss of leakage vortex. Future high-pressure turbine designs need to consider un ­steady interactions to control tip leakage losses.Keywords : Tip-Leakage Vortex;Vortex Break-down; Sullivan Vortex; Wake; Trailing Edge Shock-wave; Static PressureField0引言叶顶泄漏流是航空发动机涡轮动叶流场中的重要 组成部分,其造成的损失可达到转子总损失的45%,级总损失的30%[|]。

Effect of Surfactants on the Interfacial

Effect of Surfactants on the Interfacial

Effect of Surfactants on the Interfacial Tension and Emulsion Formation between Water and Carbon Dioxide Sandro R.P.da Rocha,Kristi L.Harrison,and Keith P.Johnston*Department of Chemical Engineering,University of Texas,Austin,Texas78712Received July8,1998.In Final Form:October7,1998 The lowering of the interfacial tension(γ)between water and carbon dioxide by various classes of surfactants is reported and used to interpret complementary measurements of the capacity,stability,and average drop size of water-in-CO2emulsions.γis lowered from∼20to∼2mN/m for the best poly(propylene oxide)-b-poly(ethylene oxide)-b-poly(propylene oxide)(PPO-b-PEO-b-PPO)and PEO-b-PPO-b-PEO Pluronic triblock copolymers,1.4mN/m for a poly(butylene oxide)-b-PEO copolymer,0.8mN/m for a perfluoropolyether (PFPE)ammonium carboxylate and0.2mN/m for PDMS24-g-EO22.The hydrophilic-CO2-philic balance (HCB)of the triblock Pluronic and PDMS-g-PEO-PPO surfactants is characterized by the CO2-to-water distribution coefficient and“V-shaped”plots of logγvs wt%EO.A minimum inγis observed for the optimum HCB.As the CO2-philicity of the surfactant tail is increased,the molecular weight of the hydrophilic segment increases for an optimum HCB.The stronger interactions on both sides of the interface lead to a lowerγ.Consequently,more water was emulsified for the PDMS-based copolymers than either the PPO-or PBO-based copolymers.IntroductionSupercritical fluid(SCF)carbon dioxide(T c)31°C,P c )73.8bar)is an environmentally benign alternative to organic solvents for waste minimization.It is nontoxic, nonflammable,and inexpensive.However,because of its very low dielectric constant, ,and polarizability per volume,R/v,CO2is a poor solvent for most nonvolatile lipophilic and hydrophilic solutes.1It may be considered a third type of condensed phase,different from lipophilic and hydrophilic phases.Consequently,it is possible to disperse either lipophilic or hydrophilic phases into CO2, in the form of microemulsions,emulsions,and latexes, given an appropriate surfactant.Because of the low values of and R/v for CO2,the most CO2-philic types of functional groups have low cohesive energy densities,e.g.,fluoro-carbons,fluoroethers,and siloxanes.2-6The solvent strength of carbon dioxide may be understood by the fact that the solubility of a polymer in carbon dioxide is highly correlated with the surface tension of the pure polymer melt.7For example,poly(fluoroacrylates)with low surface tensions of10-15mN/m are highly soluble,whereas poly-(dimethylsiloxanes)with surface tensions of20mN/m are moderately soluble,and hydrocarbon polymers with higher surface tensions show very low solubility.For nonpolar or slightly polar polymers,the surface tension is a measure of the van der Waals forces and is related to the cohesive energy density.Because R/v is so small for CO2,polymers with low cohesive densities and surface tensions are the most soluble.The first generation of research involving surfactants in SCFs addressed reverse micelles and water-in-SCF microemulsions,for fluids such as ethane and propane8,9 as reviewed recently.10,11Microemulsions are thermody-namically stable and optically transparent,with typical droplet diameters of about2-10nm.The mechanistic insight gained from these studies of phase equilibria, interfacial curvature,and droplet interactions in a su-percritical fluid is directly applicable to carbon dioxide. Attempts to form water-in-CO2(w/c)microemulsions have been elusive.6,12,13For PFPE COO-NH4+w/c microemul-sions,FTIR,UV-visible absorbance,fluorescence,and electron paramagnetic resonance(EPR)experiments have demonstrated the existence of an aqueous domain in CO2 with a polarity approaching that of bulk water,14as has also been shown by small-angle neutron scattering (SANS).15Organic-in-CO2microemulsions have also been formed for600molecular weight poly(ethylene glycol) (PEG600)and for polystyrene oligomers.16,17In many previous studies,surfactant activity in CO2has been characterized in terms of water uptake into a CO2 microemulsion.Since the results were negative most of the time,it has been difficult to determine how to design surfactants to the water-CO2interface.A more direct property,such as the interfacial tension,is needed to understand the activity of surfactants at various interfaces containing carbon dioxide.In SCF systems,only a few studies have measured the interfacial tension(γ)even for simple binary systems(1)O’Shea,K.;Kirmse,K.;Fox,M.A.;Johnston,K.P.J.Phys.Chem. 1991,95,7863.(2)McHugh,M.A.;Krukonis,V.J.Supercritical Fluid Extraction: Priciples and Practice,2nd ed.;Butterworth:Stonham,MA,1994.(3)Hoefling,T.A.;Newman,D.A.;Enick,R.M.;Beckman,E.J.J. Supercrit.Fluids1993,6,165-171.(4)Newman,D.A.;Hoefling,T.A.;Beitle,R.R.;Beckman,E.J.; Enick,R.M.J.Supercrit.Fluids1993,6,205-210.(5)DeSimone,J.M.;Guan,Z.;Elsbernd,C.S.Science1992,257, 945.(6)Harrison,K.;Goveas,J.;Johnston,K.P.;O’Rear,ngmuir 1994,10,3536.(7)O’Neill,M.L.;Cao,Q.;Fang,M.;Johnston,K.P.;Wilkinson,S. P.;Smith,C.D.;Kerschner,J.;Jureller,S.Ind.Chem.Eng.Res.1998, 37,3067-3079.(8)Fulton,J.L.;Smith,R.D.J.Phys.Chem.1988,92,2903-2907.(9)Johnston,K.P.;McFann,G.;Lemert,R.M.Am.Chem.Soc.Symp. Ser.1989,406,140-164.(10)Bartscherer,K.A.;Minier,M.;Renon,H.Fluid Phase Equilib. 1995,107,93-150.(11)McFann,G.J.;Johnston,K.P.In Microemulsions:Fundamental and Applied Aspects;Kumar,P.,Ed.;Dekker:New York,1998;Vol.in press.(12)Iezzi,A.;Enick,R.;Brady,J.Am.Chem.Soc.Symp.Ser.1989, No.406,122-139.(13)Consani,K.A.;Smith,R.D.J.Supercrit.Fluids1990,3,51-65.(14)Johnston,K.P.;Harrison,K.L.;Clarke,M.J.;Howdle,S.M.; Heitz,M.P.;Bright,F.V.;Carlier,C.;Randolph,T.W.Science1996, 271,624-626.(15)Zielinski,R.G.;Kline,S.R.;Kaler,E.W.;Rosov,ngmuir 1997,13,3934-3937.419Langmuir1999,15,419-428including carbon dioxide and a liquid phase.18-20None of these studies included a surfactant.Surfactants have been studied for the generation of CO2foams in water21typically for water-soluble surfactants.The effects of various surfactants on theγbetween supercritical CO2and PEG (600MW)were reported recently.16At276bar,the addition of1%PFPE COO-NH4+reducesγfrom3.2to2.1mN/m, and the interfacial area of the surfactant is437Å2/ molecule.Interfacial tension measurements have also been made between poly(2-ethylhexyl acrylate)(PEHA)and CO222and styrene oligomers and CO2.23As is well-known for water-in-oil(w/o)emulsions and microemulsions,the phase behavior,γ,and curvature are interrelated,as shown in Figure1.24A minimum inγis observed at the phase inversion point where the system is balanced with respect to the partitioning of the surfactant between the phases.25,26Upon change of any of the formulation variables away from this point,for example,the temperature or the hydrophilicity/hydro-phobicity ratio(in our case the hydrophilic/CO2-philic ratio),the surfactant will migrate toward one of the phases. This phase usually becomes the external phase,according to the Bancroft rule.27Unlike the case for conventional solvents,a small change in pressure or temperature can have a large influence on the density and thus on the solvent strength of a supercritical fluid.By“tuning”the interactions between the surfactant tail and the solvent,it becomes possible to manipulate the phase behavior,and therefore the activity of the surfactant at the interface and curvature,and also the extension of the surfactant tails.As an example of pressure tuning,a water-in-propane microemulsion is inverted to a propane-in-water microemulsion by varying the pressure by50bar in the C12EO6/brine/propane system, at constant temperature.28This system undergoes a phase inversion density,by analogy with the phase inversion temperature,for conventional systems.If the density is changed so that the surfactant prefers either phase over the other,the surfactant is less interfacially active and γincreases.16,22,23The objective of this study is to achieve a fundamental understanding of the lowering of the water-CO2inter-facial tension by different classes of surfactants and to use this knowledge to explain the formation and stability of water-in-CO2(w/c)emulsions.The surfactants include PFPE COO-NH4+,Pluronic R(PPO-b-PEO-b-PPO)and Pluronic L(PEO-b-PPO-b-PEO)triblock copolymers,poly-(butylene oxide-b-ethylene oxide)(PBO-b-PEO),and poly-(dimethylsiloxane)(PDMS)copolymers with PEO-PPO grafts(PDMS-g-PEO-PPO).Fromγmeasurements ver-sus concentration,the adsorption is investigated for PFPE COO-NH4+and used to determine the critical micro-emulsion concentration.For the PPO-and PDMS-based surfactants,the concept of a hydrophilic-CO2-philic bal-ance(HCB)is introduced by relatingγand the distribution coefficient of the surfactant to the EO fraction(see Figure 1).To understand howγand the HCB influence colloid stability,we chose to study w/c emulsions in contrast to previous studies of microemulsions,since so few of these surfactants form microemulsions.Emulsions are ther-modynamically unstable,but may be kinetically stable, with droplets from100nm to several micrometers in diameter.The presence of the surfactant at the interface lowers theγand thus the Laplace pressure,reducing the energy necessary to deform the interface.29The emulsions may be stabilized against flocculation due to van der Waals forces by steric stabilization,as has been analyzed theoretically,30-33and/or Marangoni stresses,due to gradients in interfacial tension at the interface.To characterize emulsion capacity,stability,and the average droplet size of the emulsions,an in-situ turbidity technique has been applied in addition to visual observations.The ability to design surfactants for the interface between CO2 and an aqueous phase based upon knowledge of the relationship between colloid formation and stability,phase behavior,andγis of interest for a wide variety of heterogeneous reactions and separation processes in CO2. Examples include dry cleaning,extraction with micro-(16)Harrison,K.L.;Johnston,K.P.;Sanchez,ngmuir1996, 12,2637-2644.(17)McClain,J.B.;Betts,D.E.;Canelas,D.A.;Samulski,E.T.; DeSimone,J.M.;Londono,J.D.;Cochran,H.D.;Wignall,G.D.;Chillura-Martino,D.;Triolo,R.Science1996,274,2049.(18)Heurer,G.Ph.D.Thesis,The University of Texas at Austin, 1957.(19)Chun,B.-S.;Wilkinson,G.T.Ind.Eng.Chem.Res.1995,34, 4371-4377.(20)Schiemann,H.;Wiedner,E.;Peter,S.J.Supercrit.Fluids1993, 6,181-189.(21)Lee,H.O.;Heller,J.P.;Hoefer,A.M.W.SPE Reservoir Eng. 1991,11,421-428.(22)O’Neill,M.;Yates,M.Z.;Harrison,K.L.;Johnston,P.K.;Canelas,D.A.;Betts,D.E.;DeSimone,J.M.;Wilkinson,S.P.Macromolecules1997,30,5050-5059.(23)Harrison,K.L.;da Rocha,S.R.P.;Yates,M.Z.;Johnston,K. P.;Canelas,D.;DeSimone,ngmuir1998,14,6855-6863.(24)Aveyard,R.;Binks,B.P.;Clark,S.;Fletcher,P.D.I.J.Chem. Technol.Biotechnol.1990,48,161-171.(25)Bourrel,M.;Schechter,R.S.Microemulsions and Related Systems:Formulation,Solvency and Physical Properties;Marcel(27)Ruckentein,ngmuir1996,12,6351-6353.(28)McFann,G.J.;Johnston,ngmuir1993,9,2942.(29)Walstra,P.Chem.Eng.Sci.1993,48,333-349.(30)Peck,D.G.;Johnston,K.P.Macromolecules1993,26,1537.(31)Meredith,J.C.;Johnston,K.P.Macromolecules1998,31,5507-5555.(32)Meredith,J.C.;Sanchez,I.C.;Johnston,K.P.;Pablo,J.J.d.Figure1.Schematic representation of phase behavior andinterfacial tension for mixtures of water,CO2,and nonionicsurfactants as a function of formulation variables.420Langmuir,Vol.15,No.2,1999da Rocha et al.emulsions and emulsions,phase transfer reactions,34,35and emulsion polymerization.36Experimental SectionMaterials.All of the surfactants were used as received,unless indicated.The CF 3O(CF 2CF(CF 3)O)∼3CF 2COO -NH 4+(PFPE COO -NH 4+),a gift from A.Chittofrati,37was stored in a desiccator.The single tail Krytox-sulfate,R -COOCH 2CH 2OSO 3--Na +,where R )CF 3(CF 2CF(CF 3)O)n CF 2CF 2-,and the triple tail Krytox-sorbitol surfactants were synthesized by E.Singley and Dr.E.J.Beckman at the University of Pittsburgh.38Pluronic L,PEO-b -PPO-b -PEO (PEO -PPO -PEO),and Pluronic R,PPO-b -PEO-b -PPO (PPO -PEO -PPO),surfactants were a gift from BASF.The block copolymer PEO-b -PBO (EO 15-BO 12,SAM185)(where the subscripts indicate the number of repeat units of each moiety)was provided by Pittsburgh Paint and Glass.The surfactant (CH 3)3SiO[Si(CH 3)2O]20[Si(CH 3)(R)]2OSi(CH 3)3,with graft R )(CH 2)3O(C 2H 4O)∼11H,(PDMS 24-g -EO 22),M w ∼2600,was a gift synthesized by Unilever.7SILWET L-7500(M w )3000),(CH 3)3SiO(Si(CH 3)2O)x (Si(CH 3)(R))y OSi(CH 3)3,with R )(CH 2)3O-(C 3H 6O)n Bu (PDMS 11-g -PO 39),with n ,x ,and y not specified,and SILWET L-7622(M w )10000),with a similar backbone,but R )(CH 2)3O(C 2H 4O)m Me (PDMS 105-g -EO 68),were provided by OSi Specialties,Inc.ABIL B 8851(M w ∼6000),(CH 3)3SiO(Si-(CH 3)2O)22(Si(CH 3)(R)O)4Si(CH 3)3,with R )(CH 2)3O(C 2H 4O)∼17-(C 3H 6O)∼4H (PDMS 28-g -EO 67-PO 17),and ABIL B 88184(M w ∼13000),(CH 3)3SiO(Si(CH 3)2O)73(Si(CH 3)(R)O)4Si(CH 3)3,with R ∼(CH 2)3O(C 2H 4O)∼32(C 3H 6O)∼7H (PDMS 79-g -EO 126-PO 28)were obtained from Goldschmidt AG.PDMS homopolymer with a M w of 13000was synthesized by J.M.DeSimone at U.N.Carolina.Poly(ethylene glycol)with a molecular weight of 600was obtained from Polysciences,Inc.Poly(butylene glycol)monoether,composed of an ethylene oxide backbone with an ethyl side group (PBO,800g/mol)was supplied by Air Products.Poly(propylene glycol)(1025g/mol)was obtained from Polysciences,Inc.,and used as received.Deionized water (NANOpureII;Barnstead)and instrument grade carbon dioxide (99.99%)were used for all experiments.Phase Behavior.Phase boundaries were determined in the variable-volume view cell as described in further detail else-where.7For a given weight of surfactant and CO 2,the pressure of the system was increased until a single phase was observed in the view cell.The pressure was then decreased slowly until the solution became slightly turbid.The pressure was then increased again,and the process was repeated.The pressure where the system became turbid was classified as the cloud point pressure.The pressure and temperature were measured to (0.2bar and (0.1°C,respectively.Interfacial Tension Measurements.The tandem variable-volume pendant drop tensiometer described previously 16was used to measure the interfacial tension between CO and water (γ).The apparatus consisted of two variable volume view cells (the drop phase cell and the measurement cell (continuous phase cell)),an optical rail for proper alignment,a light source,a video camera,and a computer.The drop phase cell contained water saturated with an excess amount of pure CO 2,and the continuous phase cell contained CO 2and surfactant (if present).In this configuration,the surfactant only has to diffuse short distances in the small volume of the droplet phase.Pendant drops were formed on the end of a stainless steel or PEEK capillary tube with an inside diameter ranging from 0.01to 0.03in.Once a suitable drop was formed,the six-port switching valve connecting the two cells was closed and timing of the drop age was started.Several images were recorded as a function of drop age.Images of the drop were obtained in a tagged imagefile format (TIFF)and the edge of the drop was extracted from data at various global threshold values using a C ++program.From the shape of the interface,the γmay be obtained from the Laplace equationwhere ∆P is the pressure differential across the interface,R 0is the radius of curvature at the apex of the drop,and z is the vertical distance from the apex.A set of three first-order differential equations was used to express Laplace’s equation,and a computer program 39,40was used to solve for γ.The density difference between the two phases was calculated by using an equation of state for pure CO 241and steam tables for pure water.The aqueous phase density was assumed to change less than 0.0025g/cm 3for the concentrations of surfactant studied.Emulsion Formation,Stability,and Average Droplet Size Estimation.Figure 2shows a schematic representation of the experimental apparatus,similar to a previous version,for turbidimetric measurement and visual observation of emulsion formation and stability.22The system consists of a 28-mL variable-volume view cell,an optical cell (0.1cm path length)which was mounted in a spectrophotometer (Cary 3E UV -vis),a high-pressure reciprocating pump (minipump with a flow rate of 8-80mL/min),and a manual pressure generator (High-Pressure Equip.,model 87-6-5).A six-port switching valve (Valco Instru-ments Co.,Inc.)with an external sampling loop was used to add water to the system.The pressure was monitored to (0.2bar with a strain gauge pressure transducer (Sensotec),and the temperature was controlled to within (0.1°C.Surfactant was initially loaded into the view cell,and the desired amount of CO 2was added with the pressure generator.The pressure was increased,and the system equilibrated at the desired T ,for ∼2h,by using a magnetic stir bar.The cloud point of the surfactant was obtained as described above.The solution was then recirculated,and deionized water was injected into the system via the 150-µL sample loop in the switching valve.The solution was sheared through a 130µm i.d.×50mm long stainless steel capillary tube upstream of the optical cell.Emulsion formation and stability were characterized based upon turbidity measurements versus time (t )at a constant wavelength (λ)650nm)and also visual observation.The turbidity is a measure of the reduction in transmitted intensity,τ)(1/l )ln(I 0/I ),where l is the path length and I 0and I are the incident and transmitted intensities,respectively.After the injection of each increment of water,the emulsion was stirred and recirculated for ∼20min (approximate time required for the absorbance to reach a maximum value).Immediately after recirculation and stirring were stopped,τmeasurements started.The stability was assessed from τas a function of t ,while the(34)Jacobson,G.B.;Lee,C.T.;daRocha,S.R.P.;Johnston,.Chem.,in press.(35)Jacobson,G.B.;Lee,C.T.;Johnston,.Chem.,in press.(36)Adamsky,F.A.;Beckman,E.J.Macromolecules 1994,27,312-314.(37)Chittofrati,A.;Lenti,D.;Sanguineti,A.;Visca,M.;Gambi,C.M.C.;Senatra,D.;Zhou,Z.Prog.Colloid Polym.Sci.1989,79,218-(39)Jennings,J.W.;Pallas,ngmuir 1988,4,959-967.Figure 2.Apparatus for emulsion formation and turbidimetry measurement.∆P )2γ/R 0+(∆F )gz(1)Surfactant Effect on Interfacial Tension Langmuir,Vol.15,No.2,1999421effective average droplet size was determined fromτversusλ.For a monodisperse system of nonabsorbing spheres in theabsence of multiple scatteringτis given byτ)3K*φ/2D,42where φis the dispersed phase volume fraction,D is the droplet diameter, and K*is the scattering coefficient.According to Mie theory,Κ*is a complex function of R(R∼D/λ,whereλis the wavelengthof the incident light)and m the ratio of the refractive indices ofthe dispersed and continuous phases.The refractive indices wereapproximated by those of the pure components,water(1.333)and CO2.43By evaluation of turbidities at two wavelengths,theaverage droplet size can be determined by an iteration proce-dure.44Results and DiscussionInterfacial Tension of the CO2-Water Binary System.The interfacial tension between pure CO2and water is shown in Figure3for two temperatures as a function of pressure,along with the data of Heurer18and Chun and Wilkinson.19Our interfacial tensions were measured1h after drop formation.Theγvalues obtained by Chun and Wilkinson19were measured with the capillary rise technique.Whereas local equilibrium was achieved within the capillary tube,the entire system was not at equilibrium.Heurer used the pendant drop technique; however,the values reported were obtained from the drop profile within10s of drop formation.Therefore,the lower values ofγin the present study suggest a closer approach to true equilibrium.A simple physical picture may be used to explain the behavior for most of the pressure range studied.16At pressures below70bar,γdecreases with increasing pressure.The cohesive energy density or free energy density of CO2is well below that of water at all pressures. The density and free energy density of CO2change over a wide range with pressure,whereas the values for essentially incompressible water are constant.As the density of the CO2phase increases,its free energy density becomes closer to that of water,andγdecreases.At low pressures where the density and free energy density change a great deal with pressure,the decrease inγis pronounced.At high pressures,where CO2is more “liquidlike”,it is much less compressible and the decrease inγwith pressure is small.For the CO2-PEG600interface,γwas predicted quantitatively with a gradientmodel and the lattice fluid equation of state.16The latticefluid model is less applicable for water due to thecomplexities resulting from hydrogen bonding and car-bonic acid formation.A cusp in the curve ofγversus pressure is observed attemperatures and pressures near the critical point of CO2.The region of the cusp inγshifts to slightly higherpressures as the temperature is increased above the criticaltemperature of CO2.For supercritical temperatures,themagnitude of the cusp increases as the temperature isdecreased toward the critical temperature.At25,1935,and38°C,the cusp in the interfacial tension is verynoticeable,while it becomes small at45°C and is notvisible at71°C.18The following argument explains how the cusp is relatedto the large compressibility of CO2.An upward pointingcusp has been observed for the surface excess of ethyleneon graphitized carbon black.45The excess adsorption canbe defined in terms of the density of the bulk phase andthe density of the interfacial region46where F(z)is the molar density of the fluid at a distancez from the surface.At pressures below the critical pressureregion,F(z)can be much larger than F,due to attractionof solvent to the surface,leading to a largeΓex.At higherpressures,the bulk fluid is much denser,so that thedifference between F(z)and F is much smaller resultingin a smallerΓex.As temperature increases above thecritical temperature of the solvent,the tendency of thesurface to raise F(z)to“liquidlike”densities diminishesandΓex decreases.Similar arguments apply to theadsorption of CO2at the water-CO2interface.TheenhancedΓex is manifested as the downward cusp inγ.Inboth examples,the cusps become broader and shift tohigher pressures at higher temperatures.Similar behavioris observed for peaks in plots of the isothermal compress-ibility of pure CO2versus pressure at constant temper-ature.To put the above results in perspective,new interfacialtension data are shown for the PEG600-CO2interface tocomplement earlier data16only at45°C(Figure4).Thevalues ofγfor the water-CO2interface are considerablylarger than those for the PEG600-CO2,PS(M n)1850),23CO2-PEHA(M n)32k)interfaces.22This result is dueprimarily to the much larger surface tension of water,∼72mN/m,versus that of PEG,∼35mN/m,and PEHA, 30mN/m.However,it is interesting thatγbetween CO2and water at high pressures,20mN/m,is below that forwater-hydrocarbon interfaces.For heptane and octane,the hydrocarbon-waterγis about50mN/m.This lower γis consistent with the higher miscibility between CO2 and water47versus hydrocarbons and water.The stronger interactions between CO2and water versus hydrocarbons and water are due to the small size of CO2which causes a smaller penalty in hydrophobic hydration,CO2’s quad-rupole moment,and,finally,Lewis and Bronsted acid-base interactions.Over the entire pressure range for PEG600-CO2at25and45°C,the interfacial tension decreased monotonicallywith increasing pressure,unlike the case for CO2-water(42)Yang,K.C.;Hogg,R.Anal.Chem.1979,51,758-763.(43)Burns,R.C.;Graham,C.;Weller,A.R.M.Mol.Phys.1986,59,(45)Findenegg,G.H.In Fundamentals of Adsorption;Myers,A.L., Belfort,G.,Eds.;Engineering Foundation:New York,1983;p207.Figure3.Interfacial tension at the CO2-water interface asa function of pressure at various temperatures.Γex≡∫(F(z)-F bulk)d z(2) 422Langmuir,Vol.15,No.2,1999da Rocha et al.at 35°C.The lack of a dip near the critical pressure may be due to the much lower compressibility at 25and 45°C versus 35°C.This contrast in behavior may also be due to a difference in the density gradient and thickness in the interfacial region for the two systems,for example,greater miscibility for the CO 2-PEG600system.Interfacial Tension:PFPE Ammonium Carboxy-late.The addition of small amounts of PFPE COO -NH 4+decreases γsubstantially as shown at 45°C and 276bar in Figure 5.As the concentration is raised above 0.03%surfactant,a discontinuity is observed,and the magnitude of the slope becomes much smaller.Because it has been shown that w/c microemulsions are formed in this system,14the discontinuity can be attributed to a critical microemulsion concentration (c µc)for the PFPE COO --NH 4+surfactant,as has been done for oil -water inter-faces.24At concentrations above the c µc,the less negative slope is caused by the addition of surfactant primarily to adsorption at the pendant drop interface,the change in γis reduced.The adsorption obtained from the Gibbs’adsorption equationfor the PFPE COO -NH 4+surfactant was 1.77×10-10mol/cm 2,which corresponds to a surface coverage of ∼100Å2/molecule.Such a high surface coverage is sufficient for the formation of microemulsions.A comparable value of ∼140Å2/molecule was measured by Eastoe et al.48at 500bar and 25°C for the hybrid hydrocarbon -fluorocarbon C 7F 15CH(OSO 3-Na +)C 7H 15surfactant in CO 2.This value was determined by assuming that all the surfactant is adsorbed at the interface of spherical droplets of 25Å2radius,as measured by SANS,with a polydispersity of ∼0.2.The substantial reduction in γand relatively high surfactant adsorption explain why it was possible to form a w/c microemulsion with PFPE COO -NH 4+.The same surfactant had an absorption of 400Å2/molecule at the CO 2-PEG interface.16Phase behavior studies indicated that PEG-in-CO 2microemulsions are also formed with this surfactant,but the nature of the core has not been characterized.16Interfacial Tension:Fluoroether Sulfate and Sorbitol Surfactant.The phase behavior of fluoroether sulfates and fluoroether sorbitols was measured by Singley et al.38for various molecular weights of single-,twin-,and triple-tailed surfactants.The surfactants were soluble in CO 2at 33°C and moderate pressure (<300bar).The sorbitol surfactants were found to be more soluble in CO 2than the sulfate ones,as expected due to the low solubilities of ions in CO 2,because of its low dielectric constant.The results showed that branching depresses the cloud point curve of a surfactant until the solubility becomes domi-nated by the overall molecular weight.These surfactants were used to form CO 2-in-water and middle-phase emul-sions with excess CO 2and water.38The interfacial tension was measured at the water -CO 2interface for the single-tailed M w 2500sulfate and the triple-tailed (7500g/mol total)sorbitol surfactants.Our measured cloud point for the 1.4%(w/w)CO 2sorbitol surfactant was 215.6bar at 45°C.For 0.56%sulfate surfactant,it was 139.8bar at 45°C.The sulfate surfactant did not lower the interfacial tension significantly over the pressure range of 180-283bar 45°C at a concentration of 0.56%.The interfacial tension was difficult to determine accurately,because bubbles and possibly surfactant precipitate appeared on the surface of the pendant drop within 15min of drop formation.The interfacial tension was estimated to be ∼15mN/m by using manual edge detection of the pendant drop.For the sorbitol surfactant,the interfacial tension decreased to ∼5.5mN/m at 276bar and 45°C with a concentration of 1.4%.Relative to other surfactants reported in this study,these surfactants were less successful in lowering the interfacial tension.Interfacial Tension:PPO -PEO -PPO,PEO -PPO -PEO,and PBO -PEO Surfactants.Block co-polymers containing CO 2-philic and hydrophilic (CO 2-phobic)functional groups may be designed to be active at the CO 2-water interface.In this section,the CO 2-philic blocks are poly(propylene oxide)and poly(butylene oxide),while the CO 2-phobic block is poly(ethylene oxide).TheFigure 4.Interfacial tension for the PEG600-CO 2interface at varioustemperatures.Figure 5.Interfacial tension for the water -CO 2-PFPE COO -NH 4+system at 45°C and 276bar.The dotted line is used to determine the surfactant adsorption via the Gibbs adsorption equation.A discontinuity is present at the critical micromemulsion concentration.Γ2)-1RT (d γd ln c 2)T ,P(3)Surfactant Effect on Interfacial Tension Langmuir,Vol.15,No.2,1999423。

Effects of in-vivo and in-vitro environments on the

Effects of in-vivo and in-vitro environments on the

Effects of in-vivo and in-vitro environments on the metabolism of the cumulus±oocyte complex and itsin¯uence on oocyte developmental capacityM.L.Sutton,R.B.Gilchrist and J.G.Thompson1Reproductive Medicine Unit,Department of Obstetrics and Gynaecology,University of Adelaide,The Queen Elizabeth Hospital, Woodville Road,Woodville,SA,5011,Australia1To whom correspondence should be addressed at:E-mail:jeremy.thompson@.auThere has been an improvement in the blastocyst rates achieved following in-vitro embryo production that can largely be attributed to improved embryo culture conditions based on an increased knowledge of the in-vivo environment,as well as the metabolic needs of the embryo.Despite this,in-vitro oocyte maturation(IVM)conditions have remained largely unchanged.Within the antral follicle,numerous events affect oocyte maturation and the acquisition of developmental competency,including:interactions between somatic cells of the follicle(in particular cumulus cells)and the oocyte;the composition of follicular¯uid;and the temperature and vascularity of the follicular environment.Many of these factors change with follicle size and oocyte growth.In contrast,culture conditions for IVM are based on somatic cells that often do not re¯ect the follicular environment,and/or have complex compositions or additives such as macromolecule supplements that are unde®ned in nature.Metabolites included in media such as glucose,pyruvate,oxygen and amino acids have been shown to have differential in¯uences on oocyte maturation and competency.Manipulation of these factors and application of gained knowledge of the in-vivo environment may result in improved in-vitro oocyte maturation and overall in-vitro embryo production. Key words:culture conditions/follicular¯uid/in-vitro maturation/metabolism/oocyte maturationIntroductionOocyte maturation is the culmination of a prolonged period of oocyte growth and development within the growing follicle,and the short interval of meiotic maturation at ovulation.It is over the long phase of weeks to months that the oocyte,in a highly co-ordinated manner,gradually acquires the cellular machinery required to support early embryonic development.This capacity of the oocyte to sustain early development,called oocyte developmental competence,is intrinsically linked to the process of folliculogenesis and to the health of the developing follicle. The follicular environment also maintains oocytes in an arrested state of meiosis,at the diplotene stage of prophase I[also called the germinal vesicle(GV)stage].The last phase of oocyte maturation,meiotic maturation of the immature GV oocyte, germinal vesicle breakdown(GVBD)and progression to meta-phase II(MII),is induced in vivo by the pre-ovulatory gonadotrophic surge.Alternatively,arti®cial release of the oocyte from the inhibitory environment of the follicle leads to spontaneous meiotic maturation in vitro(Pincus and Enzmann, 1935).Oocyte in-vitro maturation(IVM)is a viable phenomenon as oocytes matured,fertilized and cultured in vitro can generate embryos with full developmental potential after embryo transfer.Meiotic maturation following liberation of the oocyte from the follicle was®rst described during the1930s(Pincus and Enzmann,1935),but it was not until the mid-1960s that the potential for IVM as a step in the process of embryo production was recognized(Edwards,1965).However,the ability of the oocyte to undergo meiotic maturation is a poor marker of oocyte developmental capacity(Moor and Trounson,1977).In most species examined,oocytes matured in vitro are compromised in their developmental capacity compared with oocytes matured in vivo(Bousquet et al.,1999;Farin et al.,2001;Yang et al.,2001; Combelles et al.,2002;Dieleman et al.,2002;Holm et al.,2002). Furthermore,the proportion of pregnancies achieved following IVM of human oocytes from unstimulated patients is minute (Trounson et al.,1994;Cha et al.,2000).With further research, IVM has the potential to become a viable alternative to ovarian stimulation,especially for the treatment of patients with fertility disorders who are at an increased risk of developing ovarian hyperstimulation syndrome when treated with exogenous hor-mones,for example polycystic ovarian syndrome.Our understanding of what constitutes a developmentally competent oocyte recovered from antral follicles remains poor, although it is clear that the quality of the follicular environmentHuman Reproduction Update,Vol.9,No.1pp.35±48,2003DOI:10.1093/humupd/dmg009ÓEuropean Society of Human Reproduction and Embryology35 by guest on July 14, 2011 Downloaded fromfrom which the oocyte originates is a major determining factor.Despite this,little is known about how the nutrient requirements of the cumulus±oocyte complex(COC)impact on subsequent embryo development.For example,the most commonly used oocyte maturation media used today are formulations designed many years ago for culture of non-ovarian somatic cells.There are no studies that directly correlate the metabolic needs of the COC with developmental outcomes. However,the pioneering work of Downs and colleagues has clearly shown that availability of energy substrates can regulate meiotic resumption in oocytes from antral follicles,with small alterations in substrate concentrations either suppressing or inducing meiosis(Downs and Mastropolo,1994;Downs and Hudson,2000).In contrast,the effect of cell±cell signalling between the oocyte and granulosa cells during the earliest stages of folliculogenesis on metabolism of the oocyte is unknown and is likely to remain technically dif®cult to study.In this review,we will examine the composition of the antral follicular environment and how this relates to developmental outcome,and also the metabolism of the oocyte and the surrounding cellular vestment and relate these to developmental outcome and the current development of IVM media.Oocyte±follicular cell interactionsOocyte±follicular cell communication pathwaysThe follicular environment`programmes'oocyte developmental competence.Clearly,oocyte growth and development are absolutely dependent on the nurturing capacity of the follicle,in particular of the granulosa munication between the germ cell and somatic cell compartments of the follicle occurs via paracrine and gap-junctional signalling(Figure1).Indeed,both forms of communication are essential for normal oogenesis and folliculogenesis(Dong et al.,1996;Simon et al.,1997). Traditionally,research has focused on just one direction of this communication axisÐthat is,on granulosa cell support of the developing oocyteÐbut recent studies have demonstrated the importance of a bi-directional communication axis(Albertini et al.,2001).It is now becoming clear that oocyte paracrine signals are pivotal regulators of granulosa cell and ovarian function (Eppig,2001).Two key oocyte molecules identi®ed so far are growth differentiation factor9(GDF-9)and GDF-9B[also called bone morphogenic protein15(BMP-15)].These oocyte growth factors are critical for progression of the very earliest stages of folliculogenesis(Dong et al.,1996;Galloway et al.,2000),and then in late follicular development these oocyte-secreted factors play an important role in the differentiation of different granulosa cell lineages(Eppig et al.,1997;Li et al.,2000)and in the regulation of key granulosa cell functions(Elvin et al.,1999; Joyce et al.,2000;Otsuka et al.,2001).The highly specialized cumulus cells have distinctive trans-zonal cytoplasmic processes(TZP),which penetrate through the zona pellucida and abut the oolemma.Gap junctions at the ends of these TZP(and between cumulus cells)allow the transfer of low molecular-weight molecules between oocyte and cumulus cell, and also between cumulus cells(Eppig,1991).Gap-junctional communication in the follicle is essential for development and fertility.Both folliculogenesis and oogenesis fail in mice homozygous null for either connexin-37(the protein building block of oocyte±cumulus cell gap junctions;Simon et al.,1997), or connexin-43(the protein associated with gap junctions between granulosa cells;Ackert et al.,2001).Glucose metabolites,amino acids and nucleotides are all able to pass between oocyte and cumulus cells.In addition,gap junctions participate in oocyte meiotic regulation by allowing the passage of small regulatory molecules such as cAMP and purines(Dekel and Beers,1980; Salustri and Siracusa,1983;Eppig and Downs,1984;Racowsky, 1985;Racowsky and Satterlie,1985).Such intimate metabolic contact between oocyte and cumulus cells is thought to play a key role in disseminating local and endocrine signals to the oocyte via the cumulus cells.Hence,an understanding of the nutritional, metabolic or hormonal factors conferring oocyte developmental competence,by necessity,must entail an examination of the COC as a whole(as opposed to isolated oocytes).However,the majority of studies investigating energy substrates for maturing oocytes involve the addition of substrates to intact COCs and determining either developmental outcome or the metabolism of the denuded oocyte.Clearly,the metabolic pro®le of denuded oocytes(DOs)differs signi®cantly from that of COCs(Colonna and Mangia,1983;Zuelke and Brackett,1993;Khurana and Niemann,2000a).Importance of cumulus cells to oocyte IVMApart from the importance of granulosa cells and cumulus cells to the oocyte throughout follicle growth,the cumulus cells also play a critical role during spontaneous meiotic maturation in vitro.At around the time of meiotic resumption,cumulus cell±TZP begin to withdraw from the oocyte and there is almost complete loss of gap-junctional communication by the time oocytes reach metaphase I(MI).Considerable extracellular production of hyaluronic acid by cumulus cells causes dispersion of cumulus cells or cumulus expansion(Eppig,1981;Salustri et al.,1989; Chen et al.,1990).However,during this phase cumulus cells presumably continue to communicate with the oocyte,as removal of the cumulus cells prior to IVF results in compromised fertilization and embryo development compared with removing them post-IVF,regardless of co-culture with cumulus cells (Zhang et al.,1995;Fatehi et al.,2002).One of the most commonly used selection criteria for IVM is the morphology of the COC,in particular the cumulus vestment.Factors such as increased cell layers and degree of compaction are related to improved developmental out-come compared with oocytes surrounded by compromised vestments and DOs(Shioya et al.,1988;Madison et al.,1992; Lonergan et al.,1994;Goud et al.,1998),as well as there being a positive relationship between increased cumulus cell number in co-culture and developmental competence(Hashimoto et al., 1998).Follicular¯uid compositionThe follicular antrum is formed early in folliculogenesis. Follicular¯uid(FF)bathes the COC and contains a variety of proteins,cytokine/growth factors and other peptide hormones, steroids,energy metabolites and other unde®ned factors. Granulosa cells are separated by20nm-diameter channels, potentially allowing molecules up to M r500000in size to enterM.L.Sutton,R.B.Gilchrist and J.G.Thompson36 by guest on July 14, 2011 Downloaded fromthe antrum (Gosden et al .,1988).The porous nature of the follicular epithelium results in FF composition being comparable with that of `®ltered'venous plasma (Table I).Protein contentMean protein concentration is signi®cantly lower in bovine FF compared with blood serum,regardless of follicle size (Desjardins et al .,1966),and this is largely accounted for by the partial exclusion of most proteins with MW >250000(i.e.a 1-lipoprotein,a 2-macroglobulin and IgM)(Andersen et al .,1976).There is a positive relationship between increasing follicle size and the concentration of proteins with high molecular weight,indicative of increased permeability of serum proteins with follicular growth.In general,the concentration of globulins in human FF are not signi®cantly different to that in plasma,while albumin is 35%higher in FF compared with plasma (Velazquez et al .,1977).The total concentrations of amino acids in FF are also higher than in blood plasma,with the exception of cysteine (0.19mmol/l in plasma versus 0.062mmol/l in FF)(Velazquez et al .,1977),possibly due to its oxidation to cystine or use by the COC.The concentration of cysteine in a commonly used medium for IVM (Tissue culture medium 199;TCM199)is 0.6m mol/l,which is 10-fold lower than physiological levels.ElectrolytesThe concentrations of electrolytes such as chloride,calcium and magnesium in FF from large follicles (mostly pre-ovulatory)are highly comparable with serum and plasma levels (Gosden et al .,1988).Potassium levels may be elevated (1.5-to 3-fold)in FF in some species (possibly indicating active transport systems)(Schuetz and Anisowicz,1974;Gosden et al .,1988).Energy substratesThe concentration of energy metabolites in human FF has been studied with samples obtained from pre-ovulatory follicles of hyperstimulated patients undergoing assisted reproduction treat-ments.One group (Leese and Lenton,1990)reported that follicular lactate levels were 3-to 4-fold higher than serum levels (6.12versus 1.5±2mmol/l)and exist in a 2:1ratio with glucose.This contradicts later studies showing that glucose and lactate levels in human FF were 3.39and 3.17mmol/l respectively (Gull et al .,1999).Differences may have arisen from the methods used for analysis of the FF and serum and the storage of samples.Glucose-6-phosphate dehydrogenase activity and lactate dehydrogenase-1(LDH-1)synthesis increase signi®-cantly with oocyte growth,plateauing in medium-sizedfolliclesFigure 1.Oocyte±cumulus cell communication.Both paracrine (bold arrow)and gap-junctional (dashed arrow)communication between the oocyte and cumulus cells are required for normal oocyte and follicle development.Both communications pathways are bi-directional.Factors transmitted via these pathways include follicular ¯uid meiosis-activating sterol (FF-MAS),cAMP,purines and pyrimidines,metabolites,amino acids,growth differentiation factor-9(GDF-9)and GDF-9B or bone morphogenic protein (BMP-15),®broblast growth factor (FGF)and activin.Cumulus±oocyte complex metabolism37by guest on July 14, 2011 Downloaded from(Mangia et al .,1976).A positive correlation between glucose utilization and lactate production exists,and it is postulated that as the follicle grows then energy requirements increase with decreasing O 2availability (due to thickening of the avascular epithelium),leading to an increase in glycolysis and increased lactate production (Boland et al .,1993;Gull et al .,1999).This is supported by a 2-fold decrease in FF O 2tension (59.8mmHg in FF versus 102mmHg in maternal blood)and higher CO 2tension (46.9mmHg in FF versus 38.3mmHg in blood),resulting in a lower pH of FF compared with blood (7.33and 7.41respectively)(Fischer et al .,1992).All of these events are associated with increasing follicular growth leading to ovulation.Follicular vascularity and dissolved O 2content in FF are positively related to oocyte developmental outcome in humans.Measurements of follicular vascularity prior to oocyte collection demonstrated that oocytes derived from follicles with >50%blood ¯ow on their circumference had signi®cantly higher rates of clinical pregnancies following IVF and embryo transfer,com-pared to oocytes with poor vascularity (Chui et al .,1997;Coulam et al .,1999).Furthermore,only embryos resulting from oocytes collected from follicles with a high degree of vascularity (blood ¯ow identi®ed on 76±100%of the follicular circumference)resulted in successful pregnancies following embryo transfer.Poor vascularity and low dissolved O 2content are associated withdevelopmental defects such as aneuploidy,abnormal spindle organization and cytoplasmic structure (Van Blerkom et al .,1997).Oocytes from follicles with higher dissolved O 2in FF are more competent than oocytes from lower oxygenated follicles (as measured by development to 6-to 8-cell stage)(Van Blerkom et al .,1997;Huey et al .,1999).These studies suggest that hypoxic conditions have adverse effects on subsequent oocyte quality.LipidsIn general,fatty acid concentration of follicular ¯uid decreases with follicle size (Yao et al .,1980).In particular,linoleic acid is negatively correlated to follicle size,and its addition to culture medium inhibits GVBD in bovine oocytes,possibly by indirectly stimulating cAMP levels by affecting adenylate cyclase activity (Homa and Brown,1992).In general,there appears to be little information on the role of lipids during oocyte growth and maturation.There is,however,an important exception to this and that is with regard to a group of sterols,the meiosis-activating sterols (MAS),that are intermediates in the cholesterol biosyn-thetic pathway.Follicular ¯uid MAS (FF-MAS)and testicular MAS (T-MAS,®rst puri®ed from testicular tissue)are present in the FF of pre-ovulatory follicles in micromolar concentrations (Byskov et al .,2002).Their potential roles in the regulation of oocyte maturation are discussed later.Table I.The composition of sheep,pig,human and cow follicular ¯uid (FF)from pooled,small or large folliclesSheep Pig Human Cow PooledPooledPCOSPost LHPooledUnstimulated (pre LH)Stimulated (post LH)SmallLarge Small Large Na +(mmol/l)149b128h 139b 133.5b 132b 177.7i 109.2i 102.7i 88.1i K +(mmol/l) 4.7b 15.9h 8.05b 4.9b 9.2b 10.2i 7.4i 11.4i 5.6i Cl ±(mmol/l)107b97.3b 124.5b 149.5b Ca 2+(mmol/l) 2.29b2.34h 2.3b 0.94b3.1b 1.9i 2.1i 2.2i 1.8i Mg 2+(mmol/l)0.89b0.75b 0.76b 0.90i 0.89i 1.3i 0.73i Protein (g/100ml) 6.84h7.28c 7.08a 6.94f247j 33j Albumin (mg/ml)48.2c 43.4i36i54.1i47.4iTotal aa (m g/ml)236cGlucose (mmol/l) 3.44d 3.39e Lactate (mmol/l) 6.27d 3.17e pO 2(mmHg)59.8g 100.5k pCO 2(mmHg)46.9g 34.8k pH7.33g 7.35kNH 4+(m mol/l)134jSuperscripts indicate references.a Desjardins et al .,1966;b Gosden et al .,1988;c Velazquez et al .,1977;d Leese and Lenton,1990;e Gull et al .,1999;fAndersen et al .,1976;g Fischer et al .,1992;h Schuetz and Anisowicz,1974;i Wise,1987;j Hammon et al .,2000;k Huey et al .,1999.PCOS =polycystic ovary syndrome.M.L.Sutton,R.B.Gilchrist and J.G.Thompson38by guest on July 14, 2011 Downloaded fromTemperature and pHTemperature gradients exist within the ovarian environment,with pre-ovulatory follicles approximately1.5±2°C cooler than the ovarian stroma in pigs(Hunter et al.,1997,2000),humans (Grinsted et al.,1985)and cows(Grùndahl et al.,1996).How such temperature gradients are established and maintained is dif®cult to explain,and may yet re¯ect inadequate technologies to make such measurements.However,no differences in tempera-ture were observed between the stromal tissue and small antral follicles(Grùndahl et al.,1996;Hunter et al.,1997).It has been argued(Hunter et al.,1997)that the variations in temperature are established due to the follicle becoming largely avascular compared to the surrounding tissue,as well as an increase in endothermic activity associated with ovulatory processes. Decreased temperatures may decrease the viscosity of porcine FF,which would facilitate entry of the oocyte into the Fallopian tubes.However,the application of temperature gradients to IVM did not alter the developmental rates of bovine oocytes(Shi et al.,1998),indicating that although the temperature used for IVM is based on visceral temperature(and is higher than that within the ovary;Grùndahl et al.,1996;Hunter et al.,1997,2000),this seems to be adequate for IVM.The adverse effects of short-term heat shock during IVM are seen when temperatures are increased by approximately4°C and for>30min culture periods(Ju et al., 1999).IVM mediaCommercially available cell culture mediaThe maturation of oocytes in vitro is typically undertaken in commercially available complex medium,originally intended for the culture of non-ovarian somatic cells.Several commercially supplied media are commonly used for the base of IVM systems, such as TCM199,Waymouth MB752/1,Ham's F-12,Minimum Essential Medium(MEM),and Dulbecco's modi®cation of Eagle's medium(DMEM).The composition of the most commonly used IVM media are given in Table II.Table II.The composition of commercially supplied media commonly used for in-vitro oocyte maturationCompound(mmol/l)MediumTCM199Waymouth MB752/1Ham's F-12MEM DMEM HECM CaCl2 1.8020.820.23 1.36 1.36 1.9 MgSO40.788 3.960.580.790.79KCl 5.367 2.013 5.37 5.373 NaCl116.359102.67130.05116.36109.51113.8 NaHCO326.661426.1944.0425 Na2HPO4 1.017 2.5 1.18 1.17 1.04DL-alanine0.5610.1L-arginine0.3320.3610.60.4DL-aspartic acid0.4510.450.10.01 Asparagine0.01 L-cysteine 6.98Q10±40.510.220.01 L-cystine0.0830.060.10.2DL-glutamic acid0.908 1.020.10.01 L-glutamine0.684 2.41240.2 Glycine0.6660.670.10.40.01 L-histidine0.1040.780.170.20.20.01 Hydroxy-L-proline0.0763DL-isoleucine0.3050.190.030.40.8DL-leucine0.9150.380.10.40.8L-lysine0.479 1.640.250.510.01 DL-methionine0.2010.340.030.10.2DL-phenylalanine0.3030.30.030.20.4L-proline0.3480.430.30.01 DL-serine0.4760.10.40.01 Taurine0.5 DL-threonine0.5040.630.10.40.8DL-tryptophan0.09790.200.010.050.08L-tyrosine0.2560.260.230.46DL-valine0.4270.560.10.40.8Glucose 5.5527.7510 5.5524.97DL-lactate 4.5 Pyruvate1Glutathione 1.62Q10±40.16Hypoxanthine0.00220.180.04TCM=tissue culture medium;MEM=Minimum Essential Medium;DMEM=Dulbecco's modi®cation of Eagle's medium;mBM-3=Basic salt medium3; HECM=hamster embryo culture medium.Cumulus±oocyte complex metabolism39 by guest on July 14, 2011 Downloaded fromA range of different IVM base media is commonly used since oocytes from different species vary in their response to different media.Bovine oocytes matured in TCM199,SFRE(serum-free medium based on TCM199)and MEM have superior blastocyst development rates(12±19%)compared with oocytes matured in Waymouth MB752/1,Ham's F-12(3%and1%respectively; Rose and Bavister,1992)or MeÂneÂzo's B2(Hasler,2000).This is contrary to murine oocytes,where the highest cleavage rates were observed with IVM systems that used Waymouth MB752/1and MEM+non-essential amino acids(NEA),Ham's F-12and a MEM(van de Sandt et al.,1990).For porcine IVM,the composition of Waymouth MB752/1more favourably supports male pronucleus formation than TCM199or TLP-PVA(Tyrode's with lactate,pyruvate and polyvinyl alcohol)media(Yoshida et al.,1992).This may be related to high cysteine and cystine levels in Waymouth MB752/1,leading to increased cytoplasmic integrity through elevated axoplasmic glutathione(GSH)levels (Yoshida et al.,1993).Given the apparent need to test the different IVM base media in different species,the choice of base medium for human IVM is particularly dif®cult.Clearly,it is not possible to conduct an experiment large enough using human oocytes to test thoroughly the different IVM media.IVM of human oocytes is typically conducted using either TCM199(Trounson et al.,1994;Cha and Chian,1998;Mikkelsen et al.,1999)or Ham's F10(Cha et al., 1991).Waymouth MB752/1has been used for IVM of marmoset monkey oocytes(Gilchrist et al.,1995,1997),while modi®ed Connaught Medical Research Laboratories medium(CMRL-1066)is the most commonly used rhesus oocyte IVM medium (Schramm and Bavister,1994,1996;Schramm et al.,1994). The use of simple inorganic salt-based media is useful in determining which of the multitude of factors in complex media are important for successful oocyte maturation.In serum-free systems,mBM-3supplemented with glucose and a mixture of11 amino acids(in particular glutamine)(Rose-Hellekant et al., 1998),or supplemented with NEA alone,or NEA+essential amino acids(EA)(Avery et al.,1998)during IVM,led to improved blastocyst development compared with that achieved with TCM199.Embryo development has also been achieved from human oocytes matured in simple balanced salt solutions,such as human tubal¯uid(HTF;Jaroudi et al.,1997;Hwu et al.,1998) and human oocyte maturation medium(HOM;Trounson et al., 1998,2001).As IVM media trials are exceptionally dif®cult using human oocytes,such experiments are more feasible using non-human primate oocytes.With appropriate amino acid additives,a simple protein-free medium such as hamster embryo culture medium-10(HECM-10)is equally effective as the complex medium,CMRL-1066during IVM,at supporting development of rhesus oocytes through to the blastocyst stage(Zheng et al., 2001b).The formulation of IVM media speci®cally based on the composition of FFs has not been attempted.Substantial improve-ments in embryo culture media have been made over the past decade by basing media formulations on the major cation and anion concentrations and metabolic substrates of reproductive tract¯uids,for example sheep oviduct¯uid(SOF;Tervit et al., 1972),HTF(Quinn et al.,1985)and G1/G2;human tubal and uterine¯uids(Gardner et al.,1996),MTF;mouse tubal¯uid (Gardner and Leese,1990)and PL3(based on bovine blood and sheep oviductal¯uid;Park and Lin,1993).IVM ef®ciency may be improved with the design of an IVM medium along similar principles.Macromolecule supplementationThere is a long-running debate as to whether protein and macromolecule supplements should be added to IVM media and subsequent IVF and in-vitro embryo culture(IVC)media. Numerous protein supplements are used(Fukui and Ono,1989; Wiemer et al.,1991)such as fetal calf serum(FCS),estrous cow serum,estrous gilt serum,anestrous cow serum,steer serum, newborn calf serum,bovine serum albumin(BSA)and for human IVM,autologous patient serum and human serum albumin.FCS and BSA are the most commonly used protein supplements in IVM,with bovine oocytes matured in the presence of FCS having higher frequencies of oocyte nuclear maturation,cleavage and blastocyst formation compared to supplementation with or without other macromolecules(Fukui and Ono,1989;Wiemer et al.,1991;OcanÄa-Quero et al.,1999;Hasler,2000).Fetal serum contains numerous factors thought to be bene®cial to oocyte maturation and embryo development such as growth factors, lipids,albumin,hormones,steroids,cholesterol,peptides and many other unde®ned factors.The highly unde®ned nature of protein supplements makes them undesirable for many research aspects,due to the risk of batch variation and contaminating compounds of unde®ned nature.Although high-grade BSA has some degree of variability,it is less variable than serum itself. BSA has also been shown to contain steroids,especially estradiol, at levels high enough to allow for adequate cytoplasmic and nuclear maturation that supplementation with estradiol alone is unnecessary(Mingoti et al.,2002).Polyvinyl alcohol(PVA)and polyvinyl pyrolidone(PVP)are commonly used non-biological alternatives to protein supple-ments to aid in the handling of oocytes and embryos.Although oocytes matured in media supplemented with PVA or PVP have lower rates of polyspermic fertilization,development to the blastocyst stage is compromised compared with that of oocytes matured in the presence of proteins(Eckert and Niemann,1995; Fukui et al.,2000).Despite this,supplementation of PVA-based IVM media with hormones(LH,FSH and estradiol),growth factors(epidermal growth factor)and other bene®cial factors(b-mercaptoethanol,hypotaurine)can increase blastocyst develop-ment to rates comparable with oocytes matured in the presence of proteins(Avery et al.,1998;Abeydeera et al.,2000;Mizushima and Fukui,2001).This indicates that inorganic macromolecules together with de®ned protein additives can potentially replace serum/BSA supplements in IVM medium.FF as a mediumWhen FF is used as a substitute for serum in IVM media,embryo development is not in¯uenced by the size of the follicle from which the¯uid originated,nor are there any differences between bovine oocytes matured in the presence of FF or serum(Lonergan et al.,1994;Carolan et al.,1995;Kim et al.,1996).Although the size of the follicle from which the FF is sourced has little in¯uence on embryo development,¯uid obtained from non-atretic follicles supported oocyte developmental competence to a greater extent than FF from atretic follicles(CognieÂa et al.,1995).In contrast,FF from non-atretic dominant follicles when added toM.L.Sutton,R.B.Gilchrist and J.G.Thompson40 by guest on July 14, 2011 Downloaded from。

柚子酒发酵条件优化

柚子酒发酵条件优化

柚子酒发酵条件优化李敏杰;熊亚【摘要】Using pomelo as raw material, on the basis of single-factor experiments, the fermentation technology was optimized by Box-Benhnken design using alcohol degree as response value. The results showed that the optimum fermentation conditions were as follows: water to pomelo juice ratio 1∶1.3 (V/V), SO2 addition 60.9 mg/L, yeast inoculum 0.09%, fermentation temperature 28 ℃ and time 14 d. Under the optimal conditions, the alcohol content of the pomelo wine was 13.4%vol.%以柚子为原料,在单因素验的基础上,以酒精度为响应值,通过响应面分析法优化柚子酒的发酵条件.结果表明,柚子酒发酵的最佳条件为水∶柚子汁=1∶1.3(V/V),SO2添加量60.9 mg/L,酵母菌接种量0.09%,发酵温度28 ℃,发酵时间14 d,在此优化条件下,柚子酒酒精度为13.4%vol.【期刊名称】《中国酿造》【年(卷),期】2019(038)004【总页数】5页(P198-202)【关键词】柚子酒;响应面分析法;发酵条件优化【作者】李敏杰;熊亚【作者单位】攀枝花学院生物与化学工程学院,四川攀枝花 617000;攀枝花市干热河谷特色生物资源工程技术中心,四川攀枝花 617000;攀枝花学院生物与化学工程学院,四川攀枝花 617000;攀枝花市干热河谷特色生物资源工程技术中心,四川攀枝花 617000【正文语种】中文【中图分类】TS261.2柚子(Citrus maxima)又名文旦、香栾、朱栾、内紫等[1]。

双流体模型中曳力及恢复系数对气固流动的影响

双流体模型中曳力及恢复系数对气固流动的影响

2006 年 4 月Journal of Chemical Engineering of Chinese Universities Apr. 2006文章编号:10013-9015(2006)02-0164-05双流体模型中曳力及恢复系数对气固流动的影响王嘉骏, 顾雪萍, 杨富军, 冯连芳(浙江大学化学工程与生物工程学系化学工程联合国家重点实验室聚合反应工程分室, 浙江杭州 310027)摘要:应用双流体模型CFD模拟的方法,从恢复系数和曳力两方面,研究了气固密相流化床中颗粒之间和气固相之间的相互作用对床内非均匀流动结构形成与变化的影响。

计算结果表明颗粒间非弹性碰撞和气固间曳力的增大均使气固两相流动的非均匀性增大。

通过比较二者对非均匀流动结构的影响,发现气固间曳力是形成非均匀流动结构的决定因素。

从碰撞耗散、颗粒动能和颗粒势能的角度分析了二者的作用机理,发现恢复系数和曳力对流动结构的作用主要区别在于对颗粒团聚和床层膨胀的影响程度不同。

关键词:气固流化床;双流体模型;非均匀流动结构;恢复系数;曳力中图分类号:TQ021.1;TQ051.3 文献标识码:AEffects of Restitution Coefficient and Drag Force on Gas-solid Flow Behavior:A Two-fluid Model SimulationWANG Jia-jun, GU Xue-ping, YANG Fu-jun, FENG Lian-fang(State Key Laboratory of Chemical Engineering (Polymer Reaction Engineering Division), Department of Chemical and Biochemical Engineering, Zhejiang University, Hangzhou 310027, China)Abstract: In the dense gas-fluidized beds, the heterogeneous flow structure of the gas-particle flow affects the gas-solid contact and transport process in the bed seriously. In order to explore the influence of the interactions between particles and between gas and particles on the formation of heterogeneous flow structure, a computational study was carried out by using a two-fluid model based on the particle kinetic theory. The results show that the stronger the particle-particle collisional dissipation and gas-solid drag force, the more heterogeneous the flow structure. It also shows that though the collisional dissipation dramatically intensifies the formation of heterogeneous flow structure, the gas-solid drag force is the fundamental factor which triggers the pattern formation. Based on the energy budget analysis of collisional dissipation, particle kinetic energy and particle potential energy, it shows that the main different between the effect of restitution coefficient and drag force on the heterogeneous flow structure is that the former effects the particle agglomeration chiefly and the later effects the bed height mainly.Key words: gas-fluidized bed; two-fluid model; heterogeneous flow structure; restitution coefficient;drag force1 前言气固流态化技术在很多工业过程中得到广泛应用[1~2]。

2002_neuroreport_beyond syntax P600

2002_neuroreport_beyond syntax P600

Beyond syntax:Language-related positivities re£ect the revision of hierarchiesIna Bornkessel,CA Matthias Schlesewsky1and Angela D.FriedericiM ax-Planck-Institute ofCognitive Neuroscience,P.O.Box500355,04303Leipzig;1Department ofLinguistics,University ofPotsdam,GermanyCA Corresponding AuthorReceived19December2001;accepted3January2002INTRODUCTIO NSince Kutas and Hillyard’s seminal work on the application of the event-related brain potentials(ERPs)methodology to language processing[1],numerous studies concerned with how humans understand language have made use of this method.In this way,a number of language-related ERP components have been identified,among them the P600or syntactic positive shift,a parietally distributed positivity between B500and900ms after the onset of a critical word. This component has been associated with the processing of syntactic information in general[2],with processes of syntactic reanalysis and repair[3–5],and with syntactic integration cost[6].An early positive component(P345) interpreted as reflecting the diagnosis preceding reanalysis has also been reported[7,8].Thus,language-related positiv-ities have hitherto been generally associated with a broad range of syntax-related operations.The present study examines the hypothesis that language-related positivities may be elicited by operations(e.g. reanalyses)applying to hierarchical information in general, rather than to syntactic information per se.To this end,we draw upon a type of information that is not syntactic but nevertheless hierarchically ordered,namely thematic infor-mation.Thematic information essentially determines who is doing what to whom in a given sentence.In the sentence John kissed Mary,for example,John is the initiator of the kissing event(the Agent)and Mary the undergoer of this event(the Patient).Thus,thematic information provides a general conceptual specification of the relations between the arguments of a sentence to one another and to the verb. Most importantly for present purposes,thematic informa-tion is hierarchically structured by way of thematic dependencies.Broadly speaking,this type of dependency arises since‘if one participant of a predicate is causally affected,the predicate necessarily selects a causer as another participant’[9,p.52].In the following,we will only be concerned with this hierarchical nature of the thematic relations between the arguments(i.e.it will be irrelevant whether an argument represents a willfully causing Agent or a,perhaps acciden-tal,Causer,for example).In this way,we may refer to the thematically higher argument as a Proto-Agent and to the thematically lower argument as a Proto-Patient[9,10]. Previous psycholinguistic studies have shown that lan-guages with morphological case marking allow an immedi-ate mapping of arguments to thematic(proto-)roles,which is made without reference to the verb[11].This is possible because a particular case marking is only compatible with a certain subset of thematically relevant features(e.g.control), thus restricting the potential range of proto-roles that this argument may be mapped onto.The immediate association of arguments to proto-roles implies that,in verb-final clauses such as the German(1),a hierarchical thematic ordering is established between the arguments on the basis of their case marking even before the verb is encountered. Thus,the nominative argument der Professor(‘The profes-sor’)will be assigned the Proto-Agent role and the dative argument dem Ga¨rtner(‘the gardener’)the Proto-Patient role.(1)y dass der Professor dem Ga¨rtner dankt/gefa¨llty that[the professor]-a NOMINATIVE[the gardener]-DATIVE thanks/pleasesy that the Professor thanks/is pleasing to the gardener This preferential thematic ordering is confirmed by a clause-final active(Agent-Patient)verb such as danken(‘to thank’),COGNITIVE NEUROSCIENCE AND NEUROPHYSIOLOGY N EURO R EPOR Tbut disconfirmed by an object-experiencer verb such as gefallen(‘to be pleasing to’),since this type of verb assigns the higher-ranked experiencer role to the dative object [9,12].Thus,when a verb-final sentence is completed by an object-experiencer verb,the hierarchical thematic ordering between subject and object must be revised.If the hypoth-esis that language-related positivities are a reflex of hierarchically ordered information rather than syntactic information is correct,object-experiencer verbs in sentences such as(1)should give rise to a positivity in comparison to their active counterparts.In order to ensure that any effects observed for object-experiencer verbs are indeed attributable to a thematic reanalysis,we must contrast constructions such as(1)with constructions in which no thematic preferences arise before the verb is processed.Since such preferences are crucially linked to the morphological case marking of an argument, sentences analogous to(1)but without unambiguous case marking should be suitable controls in this regard(cf.2).(2)y dass Maria Sa¨ngerinnen dankt/gefa¨llty that Maria-AMB singers-AMB thanks/pleasesy that Maria thanks/is pleasing to singersIn(2),both arguments are three-way ambiguous between nominative,accusative,and dative,i.e.maximally under-specified with regard to thematic features.In view of this unrestrictedness,no hierarchical thematic ordering can be established before the verb is reached and therefore no effect of verb type should be observable.MA TERIALS ANDMETH OD STwenty students of the University of Leipzig participated in the experiment(11female;age20–27years;mean22.6years) after giving informed consent.All were right handed,native speakers of German with normal or corrected-to-normal vision.The participants were paid DM13/h.The experimental sentences comprised a matrix clause of the form Maria ho¨rte(‘Maria heard’)and an embedded clause of the form shown in(1/2).Each sentence was completed by an adjunct clause in order to avoid placing the crucial verb clause-finally.The experimental design in-volved a manipulation of the factors verb type(VERB:active vs object-experiencer)and ambiguity(AMB:unambiguous vs ambiguous case marking).Participants were presented with80sentences for each of the four critical conditions in a randomised manner.Each condition included an equal number of subject-and object-initial sentences in order to control for possible influences of word order.After each experimental sentence,participants were required to judge whether a subsequently presented declarative sentence correctly expressed the content of the preceding sentence or not.This comprehension task required the answer‘yes’equally as often as the answer‘no’,and the incorrect sentences involved an inverted assignment of grammatical functions(subject and object)to the arguments. Sentences were presented visually in the centre of a computer screen in a phrase-by-phrase manner(i.e.NPs were presented as a whole).Single words were presented for450ms and phrases for500ms with an inter-stimulus interval of100ms.Participants were asked to avoid move-ments and to only blink their eyes between their response to the comprehension task and the presentation of the next sentence.The experimental session began with a short training session followed by8experimental blocks compris-ing40sentences each,between which the participants took short breaks.The entire experiment(including electrode preparation)lasted B2.5h.The EEG was recorded by means of58Ag/AgCl electrodes fixed at the scalp by means of an elastic cap (Electro Cap International).The ground electrode was positioned above the sternum.Recordings were referenced to the left mastoid,but rereferenced to linked mastoids offline.The electro-oculogram(EOG)was monitored by means of electrodes placed at the outer canthus of each eye for the horizontal EOG and above and below the partici-pant’s right eye for the vertical EOG.Electrode impedances were kept o5kOhm.All EEG and EOG channels were amplified using a Neuroscan synamps amplifier(DC to50Hz)and recorded continuously with a digitisation rate of250Hz.The plots of grand average ERPs were smoothed off-line with a10Hz low pass filter,but all statistical analyses were computed on unfiltered data.Average ERPs were calculated per condition per partici-pant from the onset of the critical stimulus item(i.e.the verb)to1000ms post-onset,before grand averages were computed over all participants.Averaging took place relative to a baseline interval fromÀ200to0ms before the onset of the verb.Trials for which the comprehension task was not performed correctly were excluded from the averaging procedure,as were trials containing ocular, amplifier saturation or other artefacts(the EOG rejection criterion was40m V).For the statistical analysis of the ERP data,repeated measures ANOVAS were calculated for mean amplitude values per time window per condition.The statistical analysis was carried out in a hierarchical manner,i.e.only significant interactions(p o0.05)were resolved.Addition-ally,no main effects of or interactions between topographi-cal factors will be reported.Topographical factors were chosen as follows.For the midline electrodes,the factor Electrode included the eight electrodes AFZ,FZ,FCZ,CZ,CPZ,PZ,POZ and OZ as levels;for the lateral electrodes,the factors Hemisphere(left vs right)and Region(anterior,posterior)were fully crossed, thus giving rise to the following four lateral regions of interest(ROIs):left-anterior(AF7,AF3,F7,F5,F3,FT7,FC5, FC3),left-posterior(TP7,CP5,CP3,P7,P5,P3,PO7,PO3), right-anterior(AF8,AF4,F8,F6,F4,FT8,FC6,FC4),right-posterior(TP8,CP6,CP4,P8,P6,P4,PO8,PO4).RESUL TSFigure1and Fig.2show grand average ERPs at the position of the verb for unambiguous and ambiguous sentences, respectively.As is apparent from Fig.1,for unambiguous structures,object-experiencer verbs give rise to a posterior positivity between200and600ms in comparison to active verbs.By contrast,no such difference is apparent for the ambiguous structures(Fig.2).For the time-window200–600ms,the statistical analysis for the lateral electrodes revealed a main effect ofN EURO R EPOR T I.BORNKESSEL,M.SCHLESEWSKYAND A.D.FRIEDERICIAMB(F(1,19)=21.02,p o0.001),which was due to more negative waveforms for ambiguous in comparison to unambiguous structures.Furthermore,there were signifi-cant interactions for HemisphereÂAMB(F(1,19)=5.11,p o0.04)and RegionÂAMBÂVERB(F(1,19)=7.15, p o0.02).Planned comparisons for each of the four ROIs revealed significant main effects of AMB for the right-anterior(F(1,19)=11.64,p o0.01),right-posterior (F(1,19)=34.84,p=0.0001)and left-posterior quadrants (F(1,19)=8.54,p o0.01),as well as a marginal main effect of AMB in the left-anterior quadrant(F(1,19)=3.94, p o0.07).The interaction AMBÂVERB was significant in the left-posterior(F(1,19)=9.78,p o0.01)and marginal in the right-posterior quadrant(F(1,19)=3.71,p o0.07).Resol-ving the left-posterior interaction AMBÂVERB by AMB revealed that only unambiguous structures gave rise to a VERB effect(F(1,19)=19.05,p o0.001),with object-experi-encer verbs more positive than active verbs.There was no such effect for the ambiguous structures(F o1).With regard to the midline electrodes,there was again a significant main effect of AMB(F(1,19)=23.81,p=0.0001), with ambiguous structures more negative than unambig-uous structures,as well as an interaction AMBÂVERB (F(1,19)=6.76,p o0.02).Planned comparisons for unambig-uous and ambiguous sentences revealed an effect of VERB for unambiguous(F(1,19)=7.71,p o0.02,object-experien-cers more positive),but not for ambiguous structures (F o1).DISCUSSIO NOur data show that in unambiguously case marked German verb-final clauses,object-experiencer verbs give rise to an early parietal positivity in contrast to active verbs.There was no such difference in analogous structures with ambiguous case marking.Crucially,the early positivity was observable indepen-dently of syntactic manipulations.Furthermore,this effect cannot be ascribed to a simple lexical difference between active and object-experiencer verbs,since no main effect of verb class was observable for ambiguous structures.Rather, these results support the assumption that the non-confirma-tion of a preferential thematic ordering between arguments leads to a thematic reanalysis.The hierarchical reordering of non-syntactic information thus required is reflected in a positive deflection in the ERP.When no thematic hierarchis-ing can take place before the verb is processed,as is the case with ambiguous structures which provide no specification of thematic features,the processing of object-experiencer verbs does not differ from that of active verbs.The finding that revisions of non-syntactic,but never-theless hierarchical information lead to a positivity in the ERP indicates that a characterisation of language-related positivities in terms of syntax-related operations is too restrictive.It rather appears more accurate to describe early and late positivities as reflecting operations related to the manipulation of hierarchical structure in general,of which syntactic structure is,of course,a special case.Why,then,should the latency of language-related positivities vary?One possibility is that the latency differences reflect which processing stage the operations (e.g.reanalysis)giving rise to the positivity are associated with.Thus,Friederici’s neurocognitive model of sentence processing[5,13]assumes three processing phases,the first involving word category processing and the second being constituted by parallel,but independent functionaland SYNT AX-INDEPENDENT POSITIVITIES N EURO R EPOR Tinterpretative processes resulting from the application of the lexical information associated with the word currently being processed.In the third stage,reanalysis/repair processes (reflected by the P600)set in,when the syntactic and conceptual representations built up in the second phase cannot be mapped onto one another.Thus,the P600is a late positivity because it can only set in after phase2of processing is complete.In the present experiment,however, reanalysis could take place solely within the interpretative processing route of phase2,thus giving rise to an early positivity.CO NCL USIO NLanguage-related positivities reflect processing operations applying to hierarchically structured information,of which syntactic information is but a special case.The latency of these components appears to be influenced by the types of information upon which the respective processes operate.REFERENCES1.Kutas M and Hillyard SA.Science207,203–205(1980).2.Hagoort P,Brown C and Groothusen ng Cogn Processes8,43–83(1993).3.Osterhout L,Holcomb PJ and Swinney DA.J Exp Psychol Learn MemCogn20,786–803(1994).4.Osterhout L and Holcomb PJ.J Mem Lang31,785–804(1992).5.Friederici AD.Brain Lang50,259–281(1995).6.Kaan E,Harris A,Gibson E et ng Cogn Processes15,159–201(2000).7.Mecklinger A,Schriefers H,Steinhauer K et al.Mem Cogn23,477–494(1995).8.Friederici AD.Diagnosis and reanalysis:two processing aspects thebrain may differentiate.In:Fodor JD and Ferreira F,eds.Reanalysis in Sentence Processing.Dordrecht:Kluwer;1998,pp.177–200.9.Primus B.Cases and thematic roles.Tu¨bingen:Niemeyer;1999.10.Dowty nguage67,547–619(1991).11.Frisch S and Schlesewsky M.Neuroreport12,3391–3394(2001).12.Wunderlich D.Linguistic Inquiry28,27–68(1997).13.Friederici AD.The Neurobiology of Language Comprehension.In:Friederici AD,nguage Comprehension:A Biological Perspective.Berlin:Springer;1999,pp.265–292.Acknowledgements:The research reported here was supported by the grants FOR375/1-4awarded to M.S.and A.F.and FI848/1-1awarded to A.F.from the Deutsche Forschungsgemeinschaft(DFG).We would like to thank Stefan Frisch and Gisbert Fanselow forhelpful comments and discussions and Ulrike Barth for the data acquisition.N EURO R EPOR T I.BORNKESSEL,M.SCHLESEWSKYAND A.D.FRIEDERICI。

基于CFD-DPM的旋风分离器结构设计优化

基于CFD-DPM的旋风分离器结构设计优化

中国粉体技术CHINA POWDER SCIENCE AND TECHNOLOGY第27卷第2期2021年3月Vol. 27 No. 2Mar. 2021文章编号:1008-5548 (2021 )02-0063-11 doi :10.13732/j.issn.l008-5548.2021.02.009基于CFD-DPM 的旋风分离器结构设计优化彭丽,柳冠青,董方,石战胜(华电电力科学研究院有限公司多相流分离技术研究及应用中心,浙江杭州310030)摘要:采用计算流体力学离散颗粒模型(CFD-DPM),结合响应曲面法,通过系列正交实验,对旋风分离器结构进 行优化设计;考察旋风分离器的7个结构参数以及参数间的交互作用对其性能的影响。

结果表明:对压降和分离效率影响最显著的结构参数为排气管直径,然后分别是入口高度、入口宽度、旋风分离器长度、排气管插入深度;入口尺寸与排气管直径对压降的影响存在很强的交互作用;旋风分离器长度与排气管插入深度、入口宽度与排气管直 径、入口宽度与旋风分离器长度及排气管直径与旋风分离器长度对分离效率的影响存在较强的交互作用,其余因素影响不显著;通过对各结构参数的响应面进行优化,获得该旋风分离器在最小压降和最大分离效率时对应的几何结构 参数。

关键词:旋风分离器;响应曲面法;计算流体力学;两相流;模型优化中图分类号:TH31 文献标志码:AStructure optimization and design of cyclone separatorbased on CFD-DPMPENG Li , LIU Guanqing , DONG Fang , SHI Zhansheng(Research and Application Center of Multiphase Flow Separation Technology , Huadian Electric Power Research Institute Co., Ltd., Hangzhou 310030, China)Abstract : A series of orthographic experiments were designed to optimize and design the cyclone separator geometry by adoptingthe CFD-DPM (computational fluid dynamics-discrete particle model ) and the response surface method. The effect of sevencyclone geometrical parameters and their interactions on the performance were investigated. The results show that the most signifi ­cant geometrical parameter is the vortex finder diameter. Other factors of the inlet width , inlet height , total cyclone height , and vortex finder length have significant effects on the cyclone performance. In addition , there are strong interactions between theeffect of the inlet dimensions and the vortex finder diameter on the pressure drop. There are strong interactions between the effectof the vortex finder length and total cyclone height , inlet width and vortex finder diameter , inlet width and total cyclone height ,vortex finder diameter and total cyclone height on the separation efficiency. Finally, a new set of geometrical ratios are obtained toachieve minimum pressure drop maximum separation efficiency by optimization of the response surface of each index.Keywords : cyclone separator ; response surface method ; computational fluid dynamics ; two-phase flow ; model optimization旋风分离器是一种极其重要的颗粒分离设备,在火力发电、石油、化工、水泥、钢铁、冶金等工业领域应用广泛。

有调节的中介效应

有调节的中介效应
Farh, J., Lee, C., & Frah, C.I.C. (2010) Task conflict and team creativity: A question of how much and when. Journal of Applied Psychology, 95(6), 1173-1180.
3.71 1.38
Y 1 4 2 1 2 4 2
2.29 1.25
15
Post hoc analysis: An example
X 5 4 5 1 1 6 1
M SD 3.29 2.21
Z 5 3 5 2 2 4 5
3.71 1.38
Y 1 4 2 1 2 4 2
2.29 1.25
16
Post hoc analysis: Two questions
Mediation and Moderation
Kenneth Law (罗胜强)
Department of Management The Chinese University of Hong Kong
香港中文大学管理系
1
Interaction and Mediation
Moderator Interaction Mediator Moderated
Main effects
Gender (x1) Career stage (x2) Work-family conflict (x3) x1 x2 x1 x3 x2 x3 x1 x2 x3 Model R2
Significant interaction
R2 ***
11
Graphing interaction effects
  1. 1、下载文档前请自行甄别文档内容的完整性,平台不提供额外的编辑、内容补充、找答案等附加服务。
  2. 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
  3. 3、如文档侵犯您的权益,请联系客服反馈,我们会尽快为您处理(人工客服工作时间:9:00-18:30)。

Effect of the interactions between volatiles and char on sulfur transformation during brown coal upgrade by pyrolysisYulong Zhang a ,Meijun Wang a ,Zheng Qin a ,Yunlong Yang a ,Chunhui Fu a ,Li Feng b ,Liping Chang a ,⇑a Key Laboratory of Coal Science and Technology,Taiyuan University of Technology,Ministry of Education and Shanxi Province,Taiyuan 030024,People’s Republic of China bSchool of Chemical Engineering and Technology,China University of Mining and Technology,Xuzhou,Jiangsu 221116,People’s Republic of Chinah i g h l i g h t s"A modified two-stage quartz reactor has been designed."Sulfur distribution based on interactions between volatiles and char."H 2S formation combined with H radicals and organic sulfur in raw coal."COS formation combined with C–O radicals and organic sulfur in raw coal."Comprehensive mechanism of sulfur transformation in different forms.a r t i c l e i n f o Article history:Received 29April 2012Received in revised form 31August 2012Accepted 24September 2012Available online 6October 2012Keywords:CoalPyrolysisVolatiles–char interactions Sulfur transformationa b s t r a c tA modified two-stage quartz reactor was designed to further investigate the mechanism of sulfur trans-formation based on the interactions between volatiles and char during brown coal pyrolysis at different temperatures.With the two-stage reactor,the active volatiles generated from the coal fast pyrolysis can be in direct contact with the nascent char produced from the coal temperature-programmed pyrolysis.The results show that the interactions between volatiles and char have a pronounced effect on the sulfur transformation in the upgrading process of brown coal by pyrolysis,with the effect being a function of temperature.The interactions between volatiles and char promote more sulfur into the gas phase at lower temperatures of below 600°C,while remain more sulfur in char at higher temperatures of above 600°C.The evolutions of H 2,CH 4,CO and CO 2at different reaction stages play important roles for the for-mation of sulfur species in char,which are the main evidences to reveal the mechanism of sulfur trans-formation.The XPS data of chars obtained from the different types of experiments indicate that the distribution of sulfur forms on the char surface is closely related to sulfur transformation.A comprehen-sive mechanism of sulfur transformation due to the interactions between volatiles and char during brown coal pyrolysis was also explored.Ó2012Elsevier Ltd.All rights reserved.1.IntroductionLow grade coals with a high water and/or high sulfur contents have gradually become important energy resources because of their abundant minable reserves and the substantial consumption of high-quality coals.However,the sulfur and water contained in these coals largely inhibit its effective and extensive utilization.The research to improve inferior coal quality has been widely car-ried out,which mainly includes the drying,washing and pyrolysis techniques.The appropriate thermal decomposition of coal pre-sents a potentially low-cost alternative for yielding liquid fuels from condensed volatile matter and upgrading char as value-added carbon products.The pyrolysis process can make sulfur in the raw coal transfer to coal-based gas,tar,and char,forming different forms of sulfur and regulating the designed removal of sulfur according to the required destination.However,the previous research mainly considered coal pyrolysis as a whole and few resolved the whole pyrolysis into its elementary reactions.In the process of coal pyrolysis,the interactions between volatiles and char are the main factors influencing gaseous products,except for the interactions of volatiles itself and the change of the coal matrix itself [1].In addition to sulfur containing gases (H 2S,COS,CS 2),the vola-tiles contain a large amount of active groups,such as H radicals.Generally,20–50%of sulfur in raw coal is retained in the solid phase when coal is pyrolyzed,which leads to an increase of sulfur content in char [2,3]and limits the range of application for char0016-2361/$-see front matter Ó2012Elsevier Ltd.All rights reserved./10.1016/j.fuel.2012.09.061Corresponding author.Tel./fax:+863516010482.E-mail address:lpchang@ (L.Chang).after pyrolysis of brown coal.With increasing temperature,the sul-fur is gradually transferred from the bulk coal to the char surface during coal pyrolysis[4,5]and then the gaseous H radicals could easily react with the surface S to form H2S[4].During the fast pyro-lysis,the free radicals generated as a‘‘self-gasification’’process of char by the volatiles[6],would help to produce H2S and COS.At thesame time,on the surface of nascent char,there are considerable functional groups and active sites.Consequently,it is obvious that there are strong interactions between volatiles and char based on the highly reactive nature of volatiles and the vulnerable structure of nascent char.In fact,the competitive reactions of sulfur transformation exist in coal pyrolysis.For example,the active groups in volatiles can react with the nascent char and promote the sulfur in nascent char into gaseous phase;the sulfur-containing gases can react with the nascent char and are retained in the char.The effects of the interactions between volatiles and char on the volatilization of alkali(alkaline)metal and the reactivity of nascent char during coal pyrolysis have been studied[7,8],but their effects on sulfur transformation has not been presented.The effect of H2S on organic sulfur behavior and the kinetics and dynamic behavior of sulfur transformation have been investigated during heat treat-ments[9–11].However,sulfurs transformation during pyrolysis is also not very clearly understood.In order to achieve the effective and clean staged conversion of brown coal with high-sulfur content,it is necessary to obtain direct experimental evidence about the effects of interactions between volatiles and nascent char on sulfur transformation at different pyrolysis temperatures.A novel two-stage quartz reactor was designed and a series of 2.2.PyrolysisA modified two-stage quartz reactor was designed based on the research in relation to the effects of interactions between volatiles and char on volatilization of alkali(alkaline)and reactivity of nas-cent char in pyrolysis experiments on brown coal samples[7,8]. The schematic diagram of the two-stage reactor is shown in Fig.1.The modified two-stage reactor has some features:two frits were installed in the quartz tube for charging coal and/or the char sample,in the form of twofixed beds(i.d.24mm);the upper and lowerfixed bed are used to produce volatiles and prepare char, respectively.The distance between the twofixed beds is5cm. The most important feature of the two-stage reactor is that it al-lows the study of the interactions between volatiles and char and their effects on the sulfur transformation during coal pyrolysis. Two groups of symmetrical T-pipes,glass ellipsoids and stop valves constitute the coal reservoirs and were installed above the reac-tors.Before the coal is fed on the upper bed,the two stop valves are closed.The coal sample is dropped on the upperfixed bed by the carrier gas and its gravity in the experimental process.The coal sample was pre-loaded on the lowerfixed bed through the middle quartz glass tube before experiment was done.Ultra high purity argon was used as the pyrolysis carrier gas.A total gasflow rate Table2Ash composition analyses of raw coal sample used in the experiments(wt.%).SiO2Al2O3Fe2O3TiO2CaO MgO SO3K2O Na2O P2O530.314.69.7 1.218.2 4.918.60.60.90.6Fig.1.A schematic diagram of the two-stage quartz reactor.916Y.Zhang et al./Fuel103(2013)915–922actions between volatiles and char during coal pyrolysis is shown in Fig.2.In thefirst single experiment(denoted as B1),the reactor was first heated up to the desiredfinal temperatures of300,400,500, 600,700,800,and900°C,respectively.About0.5g of coal sample was then fed into the upperfixed bed and was held for a pre-set time of about30min.In this case,the amount of volatiles produced was greater than that in the experiment using a slow heating rate. This type of experiment was termed the‘‘producing volatiles’’pro-cess.As soon as the pre-set holding time was reached,the reactor was lifted out of the furnace to be cooled down naturally with a continuous Ar gasflow.The cooled chars are termed‘B1-char’and the H2S and COS released in this stage are denoted as ‘B1-H2S’and‘B1-COS’,respectively.In the second single experiment(denoted as B2),about0.5g of coal sample was pre-loaded into the lower bed.The coal was then heated up at a heating rate of10K minÀ1,from room temperature to the requiredfinal temperature,and held for the pre-set time of about30min.This type of experiment was termed the‘‘preparing char’’process.As soon as the pre-set holding time was reached,the reactor was lifted out of the furnace to be cooled down naturally with a continuous Ar gasflow.The cooled chars are termed ‘B2-char’.The H2S and COS released from chars generated in the holding stage at differentfinal temperature are denoted as ‘B2-H2S’and‘B2-COS’,respectively.It needs to be highlighted that the yields of H2S and COS for B2type of experiment do not include that from temperature-rise period,which is just the process of pre-paring char sample used in the experiments.In the comprehensive experiment(denoted as C),the two single experiments were integrated.When the requiredfinal temperature of B2experiment was reached,B1experiment was commenced. The comprehensive experiment was notfinished until B1experi-ment was accomplished.The volatiles generated in the upperfixed bed as a result of rapid heating can pass through the lowerfixed bed and react with the nascent char.As soon as the pre-set holding time was reached,the reactor was lifted out of the furnace to be cooled down naturally with a continuous Ar gasflow.In this case, the cooled chars of the upper and lowerfixed bed are termed ‘C1-char’and‘C2-Char’,respectively.The H2S and COS are denoted as‘C-H2S’or‘C-COS’,respectively.A large amount of work has been done to ensure the accuracy of the experimental results.Repeated experiments were also conducted and the results showed that the relative error is less than3%for all experimental results.2.3.Chromatographic analysis of sulfur-containing gasesIt needs to be highlighted that the product gas stream was col-lected by once per minute only at the holding stage using a gasbag. The concentrations of H2S and COS in the outlet gaseous products were directly measured by a GC-950gas chromatography equipped with a Flame Photometric Detector.The total amounts of these gases in pyrolysis were obtained by accumulating each da-tum in the whole period of the holding stage.The yields(Y i)of B1-H2S(B1-COS)and B2-H2S(B2-COS)were calculated by following equation:Y i¼XP=Qð1ÞP¼ðC1;iþC2;iÞÂVÂtÂ273:15ÂM S2Â22:4Âð273:15þTÞÂ1000=Q¼m oÂð1ÀM adÞÂð1ÀA dÞÂS dafwhere C1,i and C2,i are the release concentration of H2S or COS at the adjacent sampling time period(%);V is theflow rate of the carrier gas(ml minÀ1);t is the sampling time interval(min);M S is the mo-lar mass of the sulfur atom(g molÀ1);22.4is the volume of mole gas (L/mol);T is the room temperature(K);M ad is the moisture content in coal at air-dried basis(wt.%);A d is the ash content in coal at air-dried basis(wt.%);S daf is the sulfur content in coal or char at the dry and ash-free basis(wt.%).In the B1type of experiment,m o is the weight of coal sample on the upperfixed-bed in each experiment ($0.5g)and S daf is the sulfur content in raw coal at the dry and ash-free basis.In the B2type of experiment,m o is the weight of char on the lowerfixed-bed at the beginning of the holding stage in each experiment(<0.5g)and S daf is the sulfur content of the correspond-ingchar.Y.Zhang et al./Fuel103(2013)915–922917The yields (Y i )of C-H 2S (C-COS)and B1-H 2S +B2-H 2S (B1-COS +B2-COS)were calculated by following equation:Y i ¼XP =Q 0ð2ÞQ 0¼m o Âð1ÀM ad ÞÂð1ÀA d ÞÂS daf þm 0o ð1ÀM ad ÞÂð1ÀA d ÞÂS 0dafIn the C type of experiment,m o is the weight of the coal sample on the upper fixed-bed in each experiment and S daf is the sulfur content in raw coal on the dry and ash-free basis;m 0o is the weight of char on the lower fixed-bed at the begining of the holding stage in each experiment and S 0daf is the sulfur content of the correspond-ing char.In the calculation of B1-H 2S +B2-H 2S (B1-COS +B2-COS),m 0and S daf are the same as that in the B1type of experiment;m 0o and S 0daf are the same as that in the B2type of experiment.2.4.Analysis of bulk sulfur in the coal and charThe total sulfur content in the coal and char were analyzed according to GB/T 214-1996using a chemical method.Pyrite and sulfate contents were determined according to GB/T 215-2003,and organic sulfur content was obtained by the difference.The pro-portion of sulfur forms in total sulfur of coal sample are also shown in Table 1.The bulk sulfurs mean the majority of sulfur in a large mass compared to the sulfur on sample surface.The bulk sulfur content is approximately considered as the total sulfur content of the sample from chemical analysis method,an average sulfur con-tent in coal or char.2.5.Py-MS experimentThe aforementioned three types of pyrolysis experiments were repeated under the same operational conditions.The on-line mea-surements of H 2,CH 4,CO,CO 2,H 2S,COS and SO 2gaseous product were carried out by MS in order to discuss the effect of ambient atmosphere on sulfur transformation during coal pyrolysis.The MS system (Hiden QIC-20)automatically samples and analyzes the gases evolved from the reactor.2.6.XPS test of sulfur forms on the surface of coal and charXPS permits analysis of the sample surface (20atomic layers).Usually the S 2p signal can be resolved into peaks corresponding to pyrite,sulfides,thiophenes,sulphoxides,sulphones and sulpho-nates/sulfates on the surface of sample.XPS analyses with a V.G.Scientific ESCALAB250using focused (diameter of the irradiated area was approximately 500l m)monochromated Al K a (hv =1486.6eV,150W)were performed.The residual pressure in-side the analysis chamber was 6Â10À10mbar.The relative con-centration of sulfur forms was based in the peak values of 162.5±0.5,163.3±0.4,164.1±0.2,166.0±0.5,168.0±0.5and 170.0±0.5eV binding energy for pyrite,sulfide,thiophene,sulfox-ides,sulfones and sulfates,respectively [12,13].The samples were placed in a sealed bag,cooled in liquid nitrogen and then placed in the refrigerator for characterizing really by XPS method.3.Results and discussion3.1.Effect of interactions between volatiles and char on H 2S formation Fig.3presents the yields of H 2S from B1,B2,and C types of experiments during coal pyrolysis at different final temperature.The yields of H 2S from the three types of experiments present the similar trend—increasing to a maximum and then decreasing with the increase of pyrolysis temperature.The peak temperatureis not entirely consistent for the three types of experiments,but it can be seen that the release of H 2S in the C experiment is domi-nated by the B1experiment.The yields of H 2S present the peak va-lue at 600°C in the B1experiment,which is consistent with the instantaneous evolution profile by MS (shown in Fig.4).The yields of C-H 2S are higher than the sum of two single exper-iments (B1-H 2S +B2-H 2S)at temperature below 550°C,suggesting that the volatiles promote the sulfur in nascent char to H 2S in the lower temperature.Fig.4presents the evolution profiles of H 2S,CH 4and H 2in the process of the B1experiment.Generally,the coal pyrolysis presents in the thermal cracking stage.The breaking of the bridged linkage lead to the formation of the H-containing rad-icals,such as ÅCH 2ÀR ÀCH 2Å;R ÀCH 2Å;CH 3Å;ÅCH 2Å;H Åand so on at 300–500°C,and further react to form CH 4[14–16],which has been proved by the results from Fig.4B.H 2S is produced by a combina-tion of hydrogen radicals and sulfur-containing species formed by decomposition of sulfur containing compounds in the coal [17].When the volatiles pass through the char on the lower fixed-bed,these H-containing radicals can react with the sulfur functional groups or sulfur radicals (HS Å,S Å)in char to form H 2S and are there-by released as gaseous pared with Figs.3and 4B,it can be found that the yields of H 2S in the C experiment show the same tendency with the evolution profiles of CH 4.The result shows that the active H radicals related to the formation of CH 4can pro-mote the formation of H 2S.The yields of C-H 2S are lower than that of B1-H 2S +B2-H 2S at the temperature above 550°C.This indicates that the interactions between volatiles and nascent char retain sulfur species in char and decrease the release of H 2S at higher temperatures when the H 2S produced in the upper fixed bed passes through the lower fixed bed.The H 2S was captured by organic matrix in char to form organic sulfur with complex molecular structures,which can be attributed to the increase of thiophenes and sulfides in the char [9,18].When the temperature is risen to above 550°C,the dehy-drogenation of aromatic structures and condensation of the aro-matic nuclei into coal char are the main sources of the molecular hydrogen [19]and H radicals may penetrate into the char matrix and favor the condensation of aromatic rings [20].H 2S produced from the upper fixed-bed can react with the nascent char on the lower fixed-bed and form sulfur species in the char matrix.The hydrogen free radicals,in related to the formation of H 2in volatiles,would favor the condensation of the aromatic ring,and accelerate the nascent sulfur species ‘‘aging’’,which is a very stable form and hard to decompose.These should explain the decrease of H 2S yields from the C experiment at a high temperature of above 550°C caused by the interactions between volatiles and char.The distinction of the H 2S yields between C-H 2S and B1-H 2S +B2-H 2S becomes larger with the increase of temperature in the final tem-918Y.Zhang et al./Fuel 103(2013)915–922perature range of600–900°C in Fig.3,which is related to the hydrogen free radicals in relation to the formation of H2during coal pyrolysis(shown in Fig.4C).The amount of active H radicals formed from indigenous hydrogen in coal during pyrolysis have been getting larger with the increase of temperature,which can be sufficient to accelerate the nascent sulfur species‘‘aging’’. Therefore,the distinction of H2S yields between C-H2S and B1-H2S+B2-H2S becomes larger at the tail section of Fig.3.Formation of H2S in coal pyrolysis can be considered as a heter-ogeneous reaction,which is controlled by chemical reaction and/or gases diffusion.In the B1experiment,with increasing temperature from300to600°C,the yields and instantaneous content of H2S in-crease rapidly,which show that the sulfur release is dominated by chemical reaction control in this region.While when the tempera-ture is above600°C,the decrease of H2S yield and instantaneous content with the increase of temperature,which shows that the sulfur transformation may be dominated by gas diffusion control. Fig.5presents the effects of carrier gasflow rate on the yields of H2S in the B1experiment.In the temperature range of300–600°C,the change caused by gasflow rate is small,which indirectly proves that the formation of H2S is a reaction controlled process in this region.When the temperature is higher than600°C, the yields of H2S in the condition of high gasflow rate are obvi-ously bigger than that under the low gasflow rate.The appreciable variation in H2S yields from different gasflow rate suggests that the sulfur release is dominated by diffusion control at600–900°C.The effects of the interactions between volatiles and char on sulfur transformation in different carrier gasflow rate have also been studied and the results are shown in Fig.5.For the gasflow rate of600ml minÀ1,the volatiles in nitrogen gas stream have a superficial velocity of0.02m sÀ1.The effect of the gas diffusion resistance can be neglected when the gas velocity exceeded 0.15m sÀ1at900°C or less[9].This suggests the interaction be-tween H2S and nascent char is dominated by gas diffusion control. The apparent residence time of tar and gaseous products in the lowerfixed bed is0.12s.H2S appears to have been remarkably adsorbed by the char through the formation of thermally stable organic forms of sulfur species at the initial stage of the interaction at800°C[9,10].Therefore,the appreciable difference in H2S yields was observed between C-H2S and B1-H2S+B2-H2S.Fig.5also shows that the difference of H2S evolution profiles under different carrier gas velocities in the C experiment is larger than that in the B1experiment over the whole temperature range.This means that the effects of carrier gasflow rate on the C experiment are larger than that on the B1experiment.By increasingflow rate,H2S can be quickly brought out from the reaction zone,which also mini-mizes the occurrence of secondary reactions.3.2.Effect of interactions between volatiles and char on COS formationFig.6presents the changes of yields of COS from different types of coal pyrolysis as a function of temperature.The formation of COS presents a similar trend with that of H2S,but the amounts of COS yields are one order of magnitude lower than that of H2S.ThisY.Zhang et al./Fuel103(2013)915–922919suggests that H2S is the main component of sulfur containing gas during coal fast pyrolysis.The yields of C-COS are higher than that of B1-COS+B2-COS at the temperature of below500°C and are lower at the temperature of above500°C,which show that the interactions promote the sulfur in nascent char or other sulfur spe-cies in volatiles to transform into COS at the lower temperature and retain the gaseous COS into nascent char at the higher temper-ature.This phenomenon is similar to the trend of H2S release but with a more obvious variation.This means that the effects of the interactions between volatiles and char on the formation of COS are bigger than H2S,which suggests that COS can be formed from other route besides the secondary reaction of H2S.COS during coal pyrolysis is mainly produced from pyrite, organic sulfur or secondary reactions of sulfur-containing groups [21].From the data in Table1,it can be seen that there is no pyrite in the coal sample,so the COS is mainly produced from organic sul-fur and secondary reactions of sulfur species.The formation of COS is related to the formation of CO2and CO by recombining C–S and C–O[21,22].CO2and CO in volatiles can react with the sulfur species producing COS,such as H2S+CO2=COS+H2O and H2S+CO=COS+H2.When the volatiles from fast pyrolysis of coal in the upperfixed-bed pass through the nascent char on the lower fixed-bed,the equilibrium relationship between hydrogenolysis/ hydrolysis reactions of COS and its reverse reactions at different temperatures are the main factors influencing the COS release, which is very sensitive to the interactions of volatiles and char.Fig.7presents the instantaneous release profiles of COS,CO and CO2from the B1experiment at different pyrolysis temperature with time.The evolution profiles of COS and CO2show a similar trend during fast coal pyrolysis,which are different from that of CO.The release amounts of COS and CO2present peak values at 600°C,but that of CO increases gradually with increasing pyrolysis temperature.This indicates that the C–O groups related to the for-mation of CO2are much easier to bond sulfur radicals than that for CO.At300–500°C,the volatiles containing much C–O groups related to the formation of CO2pass through the nascent char, and promote the sulfur species in nascent char and/or H2S in gases to transform into COS.Although the volatiles contain abundant C–O groups related to the formation of CO at600–900°C,the C–O groups show weak activity to promote the evolution of COS. On the contrary,COS can react with the active char and becomea part of the char-S.3.3.Effect of interactions between volatiles and char on sulfur distribution in charFig.8shows the sulfur content in chars at different temperature from the three types of experiments.Corresponding to the change trends of S-containing gases at different pyrolysis temperatures, the amounts of char-S from three types of experiments present the trend of decreasing to a minimum and then increase with the increase of pyrolysis temperature.The sulfur content in C1-char920Y.Zhang et al./Fuel103(2013)915–922and B1-char at different temperature is nearly identical,while there is a significant difference between B2-char and C2-char due to the interactions of volatiles and char in the C experiment.The sulfur on the surface of the chars obtained from B2-char and C-char and raw coal were studied by XPS and the analysis re-sults are presented in Fig.9.The sulfur on char surface mainly pres-ent in the forms of sulfide,thiophene,and pared with the total sulfur content of2.9wt.%in raw coal by chemical analysis method,the sulfur content of0.55at.%on the surface of raw coal is much lower,which suggests that sulfur is rich in the bulk and poor on the surface of coal.When coal is heated in the Ar gas,the sulfur content on the char surface is greater than the raw coal and in-creases gradually with increasing pyrolysis temperature,which can be mainly attributed to the increase in the sulfide and thio-phene on the char surface.The result suggests that the sulfur could transfer from the coal matrix to the char surface during pyrolysis and there exists a dynamic equilibrium of the sulfur content in bulk and on the surface of pared with the B2char,the total sulfur content in the C2char is lower at below600°C and higher at above600°C,which presents the corresponding relation with the release of sulfur-containing gases.The content of sulfide in C2char is lower than that in B2char over the whole experimental temperature range,which suggests that the interactions of volatiles and char decrease the content of sulfide in char over the whole experimental temperature range. The effects of interactions of volatiles and char on the transforma-tion of sulfide are mainly shown in two aspects.Firstly,the resi-dence time of volatiles is increased,which increase the chance for the interaction between sulfide in char of lowerfixed-bed and volatiles containing extensive H radicals from upper fixed-bed.Secondly,the alkali metallic species(e.g.potassium and sodium)contained in the volatiles increase the decomposition of sulfide in char.The alkaline earth metallic species,volatilized from the upperfixed bed during the coal fast pyrolysis[7,8],would be trapped by the char on lowerfixed bed.The trapped alkaline earth metallic species can transform the thermally stable organic sulfur(mainly sulfide)to more labile species[23],and enhance the decrease of the sulfide content in char.The change of thiophene content in C2char with temperature compared with B2char is lower at below500°C and higher at above500°C,which is consis-tent with the trend of H2S release.This result indicates that the ef-fect mechanisms of the interactions between volatiles and char on the transformation of sulfide and thiophene are different.At high temperature,the volatiles containing extensive quantities of H rad-icals related to formation of H2age labile sulfur compounds to more complex and stable species such as thiophene.Sulfone content is low and the effect of interactions between volatiles and char on the transformation of sulfone is insignificant at higher temperatures of above500°C.The content of sulfate on the char surface shows no change between C and B2experiments at400°C,while sulfide,thiophene and sulfone appear to decrease significantly,which show that the decrease of sulfur content on char surface owes to the decomposition of the organic sulfur.The decrease of the sulfate content in C2char at500and600°C impliesY.Zhang et al./Fuel103(2013)915–922921。

相关文档
最新文档