外文翻译(英文)陶瓷颗粒细度对氧化锆-莫来石复合材料性能的影响

EFFECT OF CERAMIC POWDER FINENESSON MULLITE-ZIRCONIUM CERAMIC PROPERTIESG .P.Sedmale,1A.V .Khmelev,1and I.É.Shperberga 1Translated from Novye Ogneupory ,No.1,pp.41–46,January 2011.Original article submitted October 29,2010.Results are provided for a study of the development of high-temperature mullite-zirconium ceramic with use of activated ceramic powders prepared by grinding for different times with addition of illite clay,and from pure oxide powders.It is shown that increased activity and amorphicity of ground particles considerably pro-motes formation of mullite phase at 1200°C,and also transition of the monoclinic modification of ZrO 2to tetragonal,particularly with an increase in firing temperature.It is proposed that as a result of rapid “freezing”the structure retains the high-temperature modification of ZrO 2,having a tendency with slow ceramic cooling to transform into the monoclinic modification.Keywords:mullite-zirconium ceramic,illite clay,grinding,particle size.INTRODUCTIONMullite-zirconia (mullite-corundum)ceramic is one of the materials used extensively in high-temperature produc-tion processes.A distinguishing feature of mullite-zirconia ceramic is the retention of high strength,including at ele-vated temperature and with temperature falls.The set of these properties predetermines further application of the ce-ramic and use of it in high-temperature production processes.It has been established [1]that mullite-zirconia ceramic may be prepared from mixed starting compositions,includ-ing g-Al 2O 3,silica-gel,ZrO 2mon,Y 2O 3with addition of 7.75–8.75wt.%illite clay,promoting sintering and forma-tion of a mullite phase at lower temperatures [2].On the other hand,the importance is indicated in [3,4]of the degree of grinding of the starting powders.It has been established grinding ceramic powders leads to destruction of the particle crystal lattice,and as a consequence to amorphization.Here higher sintering indices are achieved for ceramic material and correspondingly density,and ultimate strength in bend-ing and compression.However,a distinguishing feature of rapid grinding [3]is formation of coarse agglomerates con-sisting of particles strongly sintered to each other.It has been proposed that agglomeration is due to heat liberated during grinding.Therefore,as indicated in [5],the duration of grind-ing should severely limited and determined by experimentfor each specific case.According to the authors of the pres-ent articles,grinding duration for starting powders is 5–6h.It is also noted [6]that grinding of starting powder pro-motes crystallization of mullite and tegtragonal ZrO 2in ce-ramic material.After a short period of grinding (4–6h)rapid mullite formation during firing is observed at 1180–1280°C;further mullite formation occurs over the ex-tent of the next 350°C.It has been established [7]that with an average content of particles with a size of 0.5m m in the starting powder the increase in ultimate strength in compression and even elas-ticity modulus was about 30%compared with a specimen containing particles with a size of more than 5m m.It has been noted that the size of mullite crystal particle that form,which are 50–70nm,and sometimes 80–95nm,is of con-siderable importance.The aim of this work includes determining the effect of ceramic powder fineness,including with addition of illite clay,on formation and development of high-temperature crystalline phases (mullite and ZrO 2),and the mechanical properties of mullite-zirconia ceramic.STUDY METHODSThe starting powder was prepared from a mixture con-sisting of synthetic materials,i.e.,g -Al 2O 3prepared from calcined Al(OH)3at 550°C,amorphous SiO 2,ZrO 2mon,and Y 2O 3.A mineral raw material was used in one part of the Refractories and Industrial Ceramics Vol.52,No.1,May,2011351083-4877/11/05201-0035©2011Springer Science+Business Media,Inc.1Riga Technical University,Institute of Silicate Materials,Riga,Latvia.starting powder as an additional component,i.e.illite clay containing about 65%illite fraction.The starting powder compositions are provided in Table 1.The chemical and mineral compositions,and also the av-erage grain size of illite clay are provided below:Chemical composition,%:SiO 2...............................50.5Al 2O 3...............................20.8Fe 2O 3...............................7.5TiO 2................................1.2CaO ................................1.9MgO................................3.6K 2O ................................6.0Na 2O ...............................0.1D m cal (1000°C)..........................8.4Mineral composition,wt.p:illite K 0.5(H 3O)0.5Al 2(OH)2[(Al,Si)4O 10]·n H 2O .......65–70quartz SiO 2..........................18–20calcite CaCO 3..........................5–6goethite a -FeOOH .......................7–8kaolinite Al 2(OH)4[Si 2O 5]...................5–7Content,%,particles with sizes,mm:63–20..............................20.520–6.3..............................21.56.3–2.0.............................28.5<2.0...............................29.5The starting powder mixes were ground and homoge-nized in a laboratory planetary mill for 4,10and 24h,using corundum balls as the grinding bodies.The size of ground powder particles was determined by means of an SEM mi-croscope (model JSM-T200,Japan),and the average particle distribution was evaluated by means of a photon correlation spectrometer using a strongly diluted suspension of 10–2N KCl (40mg KCl +100ml H 2O);the surfactant used was Fairy washing substance.In order to determine the crystal-line particle size an x-ray diffractometer (model Rigaku,Ja-pan,with Cu Ka-radiation with a scanning interval 2è=10–60°and a speed of 4°/min)was used.The size of crystal particles D ,nm,in the original ground powders was determined by the Sherrer equation [8]:D =k l /B cos q ,where k is Bolzmann constant (k =0.87–1.0);l is x-ray beam wavelength,nm (l =0.15418nm).The value of B ,rad,was calculated from the difference in x-ray beam reflection angles,B =è2–è1.A schematic image of the diffraction maximum with a note of the values required for calculation is shown in Fig.1.Specimens for phase composition,structure and proper-ties of the prepared ceramic materials were manufactured in the form of disks 30mm in diameter and 3mm thick,cylin-ders with a diameter of 35mm and a height of 45mm,and rods 55mm long and 3mm thick from powders by axial compaction (pressure 120MPa).Firing was carried out in an air atmosphere in the range from 1200to 1500°C after each 100°C (in a laboratory muffle furnace Nabertherm 3000,Germany,with a heating rate of 6°C/min with soaking at the maximum temperature for 30min).Specimen density after firing was evaluated according to apparent density from the ratio of specimen weight to its volume,and also from deter-mination of open porosity and overall shrinkage according to EN 63-2:2001.The composition of phases,as for the size of crystals,and also microstructure of fired specimens were determined from x-ray phase analysis (XPA)data and scanning a scan-ning electron microscope (JSM-T200,Japan)data after each firing temperature cycle.Thermal shock resistance was de-termined according to EN820-3:2004,using ceramic rod specimens fired at 1400°C.Specimens were studied after each 200°C in the range 500–1000°C.Thermal shock resis-tance after firing cycle was evaluated from the change in elasticity modulus and ultimate strength in bending.Elasticity modulus was determined in a Buzz-o-sonic in-strument (firm Buzz-Mac International LLC,USA)accord-ing tot eh principle of measuring shock wave propagation with a ceramic specimens placed perpendicular to two paral-lel installed metal bases,covered with a thin polymer layer.Shock waves within a specimen were created by means of small polymer hammer,in one end of which there was a steel ball 4mm in diameter,recorded by a special microphone and analyzed by means of a Fourier arrangement.Elasticity modulus E ,GPa,was calculated by the equa-tionE =0.9465r f 2L 4T 1/t 2,where r is specimen density,g/cm 3;f is frequency,Hz;L is specimen length,mm;T 1is correlation value which depends on test specimen dimensions;t is specimen thickness,mm.The ultimate strength in compression of ceramic speci-mens was determined according to EN 658-2:2003using a TONI Technik Controller TG0995instrument,and ultimate strength in bending was determined by a three-point method according to EN 843-1:2006using a Zwick/Role model 1486instrument.36G.P.Sedmale,A.V .Khmelev,and I.É.ShperbergaTABLE 1.Starting Powder Compositions for Preparing Mullite-Zirconia Ceramic,wt.%Powder compositiong -Al 2O 3SiO 2·n H 2OZrO 2Y 2O 3Illite clay1062.3028.00 5.20 4.50–10i 57.3025.85 4.70 4.158.00Fig.1.Schematic image of diffraction maximum with marked value for calcu-lating crystal particle size by the Sherrer equation.RESULTS AND DISCUSSIONMicrophotographs are shown in Fig.2of ceramic pow-der compositions 10and 10i,ground for 4and 24h.After 4h of grinding powder composition 10is represented by densely compacted amorphous particles of circular shape with sizes of about 3to 7–10m m,being both individual formations and in the form of agglomerates.At the same time the struc-ture of ceramic powder 10i with additions of illite clay,ground for 24h,is in the from of amorphous agglomerates of finer particles.According to XPA data,the average size of ZrO 2mon,Y 2O 3and quartz SiO 2crystals (introduced with illite clay)in powders after 4,10and 24h of grinding with an without ad-dition of clay varies within limits 55–100nm (Fig.3).It may seen that additions of clay affects the dispersion of crys-tals in the original powder.There is especially a reduction in crystal size of baddeleyite after 24h of grinding powder and Y 2O 3(less markedly);a sharp reduction in the size of quartz and ZrO 2crystals in composition 10i,reaching 55–59nm (see Fig.3).On the other hand,from the results of determin-ing the size of particles an agglomerates in ground powder using a correlation spectrometer it follows that the size of the maximum amount of particles is seen within the limits 200–520nm (Fig.4).Formation of mullite and corundum phases,and also tetragonal and monoclinic ZrO 2(according to XPA data)in specimens sintered at different temperatures,is shown in Fig.5.It may be seen that mullite formation commences at 1200°C and it develops actively up to 1600°C,although the first mullite phase nuclei possibly form at 1100°C due to the activity and amorphicity of particles in the original powder.Crystalline phases of SiO 2(cristobalite and quartz)are pres-ent up to 1250°C,and at higher temperatures (1330and 1400°C)a reaction sets in promoting further mullite and also corundum formation.It should be noted that over the whole temperature range apart from a diffraction maximum from ZrO 2mon,there is formation of a clearly expressed maximum from ZrO 2tetr,Effect of Ceramic Powder Fineness on Mullite-Zirconium Ceramic Properties37Fig.2.Microstructure of starting powder composites 10and 10i re-spectively after grinding for 4(a )and 20h (b).Fig.3.Change in crystal dimensions in relation to starting powder grinding duration for composites 10and 10i:¯and ¡)baddeleyite (correspondingly compositions 10and 10i);*)corundum (composi-tion 10i);r and ´)Y 2O 3(correspondingly compositions 10and 10i);+)quartz (composition10i).Fig.4.Most possible distribution of particles P and their aggregates in starting powder compositions 10and 10i after grinding for 4and 10h.Fig.5.X-ray patterns of crystalline phase formation in a 10i speci-men in relation to sintering temperature:M is mullite 3Al 2O 3·2SiO 2;C is corundum a -Al 2O 3;Q is quartz;Cr is cristobalite;Z m is ZrO 2mon;Z t is ZrO 2tetr.which points to partial stabilization of zirconium dioxide due to introduction of Y 2O 3into the ZrO 2structure.The microstructure of a specimen of this ceramic,shown in Fig.6a ,is represented by densely packed crystalline for-mations,mainly mullite of prismatic and pseudo-prismatic habit;for a specimen without a clay addition (Fig.6b )the shape of mullite is more clearly observed.As may be seen from data for shrinkage of specimens of composition 10i (Fig.7),the effect of grinding duration t ,i.e.,increase in fineness of the starting powder,is quite marked.Specimen shrinkage increases from 15to 25%with an increase in t for the starting powder for each temperature.Such a marked difference in development of shrinkage with an increase in t may be explained by assuming that with presence of a liquid phase there is “contraction”(drawing to-gether)of particles followed by sintering by a solid-phase mechanism.An increase in shrinkage with an increase in sintering temperature to 1500°C causes a reduction in liquid phase viscosity and consequently acceleration of ion diffu-sion,leading to specimen contraction during cooling.A simi-lar observation is also given by the authors in [4].By analyzing the change in apparent density r app and ul-timate strength in compression s co (Fig.8)it may be noted that a significant role in increasing specimen r app (up to 3.0and ~2.6g/cm 3correspondingly for compositions 10i and 10)is played by addition of illite clay,which is particularly typical for specimens ground for 10and 24h.This is ex-plained by more active diffusion and reaction of particles in the presence fo a liquid phase,whose formation is pro-moted by addition of illite clay.A positive role is also pro-posed for addition of illite clay on the transformation ZrO 2tetr ®ZrO 2mon on cooling specimens.At the sametime,ceramic specimens without addition of illite clay have lower values of r app since in this case sintering is determined mainly by the activity and amorphicity of particles in the starting powder.A more intense increase in r app and s co of specimens of powders after the first 4h of grinding should also be noted.A further increase in grinding duration (t >10h)is less effec-tive,which is probably due to agglomeration of particles.Addition of illite clay at a high level promotes an increase in the value of s co ,reaching a maximum value of 165MPa.An increase in s co to 98MPa is also demonstrated by specimens without addition of illite clay from powders after 24h of grinding,although the lower values of this index are due to presence of internal pores within specimens,which is indi-cated by the power values of r app .The elasticity modulus E of specimens after thermal shock D T ,particularly with an increase in the temperature range during thermal shock,and also specimens prepared from finer powders,has a tendency to increase (Fig.9).Such a clear difference in the value of E for specimens is appar-ently connected mainly with phase transformations of ZrO 2[6,11](ZrO 2mon >ZrO 2tetr),and in all probability with im-provement and growth of prismatic crystalline mullite for-mations.For specimens with addition of clay the value of E increased by 2–3GPa or more uniformly,particularly with a sharper temperature drop (800/20and 1000/20).For speci-38G.P.Sedmale,A.V .Khmelev,and I.É.ShperbergaFig.6.Microstructure of specimens of compositions 10i (a )and 10(b ),fired at1300°C.Fig.8.Change in apparent density r app (––)and strength in com-pression s co (——).Fig.7.Change in shrinkage Y for specimens of composition 10i:t is temperature;t is starting powder grinding time.mens after the first thermal shock cycle(500/20)a similar tendency is retained.A similar tendency is observed for the ultimate strength in bending s ben of specimens after thermal shock.As may be seen from Fig.10a,the overall tendency of a change in s ben involves a marked increase after thermal shock.For exam-ple,whereas for an original specimens(after24h grinding of starting powder)the value of s ben is25MPa,with an in-crease in D T s ben reaches42.5MPa.It may be proposed that compliance of specimen to a sharp change in temperature followed by rapid b rapid“freezing”of the structure pro-motes a marked or total transfer ZrO2mon>ZrO2tetr,and to a certain extent agrees with expressions provided in[10].Changes of s ben of specimens with addition of clay (Fig.10b)are more uniform.The previous tendency is ob-served towards a more marked increase in s ben of specimens of powders after24h of grinding,and also with an increase in D T,which is also explained by ZrO2polymorphic transfor-mation.CONCLUSIONResults of development of high-temperature mullite-zir-conia ceramic using activated powders,ground for different times without or with addition of illite clay,showed that the particle size of the starting powder is determined by grinding e of different methods for evaluating powder particle size,and also particle agglomeration in a powder, gives different results.It is clear that the size of crystal parti-cles is~50–100nm,whereas for amorphous particles and aggregates it is200–500nm.An increase in activity and amorphicity of ground particles markedly promotes forma-tion of mullite phase starting from1200°C,and also a transi-tion of the monoclinic modification of ZrO2into tetragonal, particularly with a high firing temperature.It has been established that specimens have relatively high shrinkage(15–25%),which increases considerably with an increase in temperature and starting powder grinding duration.The apparent density and ultimate strength in compres-sion of specimens are governed both by the duration of start-ing powder grinding and also presence of illite clay within it. After24h of starting powder grinding these indices for spec-imens without addition of illite clay reach2.55g/cm3and 80MPa;for specimens with additions they increase to 2.95g/cm3and15MPa respectively.Elasticity modulus and ultimate strength in bending for specimens with an increased temperature difference during thermal shock,both for specimens of powder with prolonged grinding(10and24h)have a tendency towards a marked in-crease,particularly for specimens without illite clay.It is assumed that the tendency of ceramic specimens to-wards a sharp temperature drop leads to rapid“freezing”of the structure,promoting retention of the high-temperature tetragonal modification of ZrO2.REFERENCES1.G.P.Sedmali,I.É.Sperberger,A.V.Khmelev,et al.,“Forma-tion of ceramic in the system Al2O3–SiO2–ZrO2in the presence of mineralizers,”Ogneupory Tekhn.Keram.,No.5,18–23 (2008).2.G.Sedmale,I.Sperberga,U.Sedmalis,et al.,“Formation ofhigh-temperature crystalline phases in ceramics from illite clay and dolomite,”J.Eur.Ceram.Soc.,26,No.15,3351–3355 (2006).Effect of Ceramic Powder Fineness on Mullite-Zirconium Ceramic Properties39 Fig.9.Change in elasticity modulus E for specimens of composi-tion10in relation to difference in temperature interval D T with ther-mal shock.Specimen firing temperature1400°C.Fig.10.Change in ultimate strength in bending s ben of specimenswithout addition(a)and with addition of illite clay(b)made frompowders ground for4,10and24h in relation to temperature D T.Specimen firing temperature1400°C.3.N.Behmanesh and S.H.Manesh,“Role of mechanical activa-tion of precursors in solid-state processing of nano-structured mullite phase,”J.of Alloys and Compounds,450,421–425 (2008).4.Y.Lin and Yi.Chen,“Fabrication of mullite composites by cy-clic infiltration and reaction sintering,”Materials Science and Engineering A,298,179–186(2001).5.L.B.Kong,T.S.Zhang,J.Marr,et al.,“Anisotropic graingrowth of mullite in high-energy ball milling powders doped with transition metal oxides,”,”J.Eur.Ceram.Soc.,23, 2247–2256(2003).6.E.Medvedovski,“Alumina-mullite ceramics for structuring ap-plications,”Ceramics International,32,369–375(2006).7.C.Aksel,“The effect of mullite on mechanical properties andthermal shock behavior of alumina-mullite refractory materi-als,”Ceramics International,29,183–188(2003).8.S.Junaid,S.Quazi,and R.Andrian,“Use of wide-angle x-raydiffraction to measure shape and size of dispersed colloidal par-ticles,”J.of Colloid and Interface Science,338,No.1, 105–110(2009).9.W.Yoon,P.Sarin,and W.M.Kriven,“Growth of textured mul-lite fibers using a quadrupole lamp furnace,”,”J.Eur.Ceram.Soc.,28,455–463(2008).10.N.M.Rendtorft,L.B.Garrido,and E.F.Aglietti,“Thermalshock behavior of dense mullite-zirconia composites obtained by two processing routes,”Ceramics International,34, 2017–2024(2008).40G.P.Sedmale,A.V.Khmelev,and I.É.Shperberga。