A STUDY ON THE BEHAVIOUR OF GALLIUM IN THE IRONMAKING PROCESS A STUDY ON THE BEHAVIOUR OF GALLIUM IN THE IRONMAKING PROCESS

A STUDY ON THE BEHAVIOUR OF GALLIUM IN THE IRONMAKING PROCESS

  • 期刊名字:金属学报
  • 文件大小:845kb
  • 论文作者:L.Savov,S.Garonin,O.Ivanov,Y.S
  • 作者单位:Institute of Iron and Steel Technology,Moscow State Institute of Steel and Alloys (MISIS)
  • 更新时间:2020-11-10
  • 下载次数:
论文简介

ACTA METALIJURGICA SINICA (ENGLII LETTERS)Vol. 16 No. 2 pp 78-83 April 2003A STUDY ON THE BEHAVIOUR OF GALLIUM IN THEIRONMA KING PROCESSL. Sao), S. Garonin), 0. Ioanov), Y.S. Yusfin2) ard D. Jankse)1) Istitte of Iron and Steel Technology, Freiberg Universty of Mining and Technology, D-09596 Freiberg,Germany2) Moscow State Institute of Steel and Alloys (MISIS), 4 Leninsky Prospekt, 117936 Moscow, RussiaManuscript received 9 May 2002The demand for gallum is increasing but production i8 limited since gallium is ex-tracted only as a by-product of bauxite processing. On the other hand coal, ironmakingcoke and iron ore gangue contain tmaces of galliun. However lttle is known about thebehaviour of gallium in ironmaking. The aim of the study ig to clarify the distributionof gllsum between hot metal, slag and top gas by means of laboratory erperimenls. Itwas found that Ga2O3 is not stable in blast furnace slags and that gllium is relarnedin hot metal. Vacum distillation erperiments with hot metal showed that galliumis not transferred to the gas phase. Data on the input and output of gallium at twoindustrial blast fumaces, as well as chemical analyses of the gallium content of severalcokes are presented, too.KEY WORDSgalium, thermodynamics, irnmaking, blast fumnace, cokeT151. IntroductionGallium is a valuable rare metal which is mainly being used in the production of GaAs.The demand for gallium has been increasing due to the growth of the wireless telecom-munications industry and the development of gallium-based devices for the optoelectronicmarket. In 2000 the world production of refined gallium was estinated to be about 110metric tons, and the price for crude gallium reached $600 per kilogram[4l.Most gallium is produced as a by-product of treating bauxite for aluminium production,and the remainder is obtained from zinc-processing residues. The gallium content of bauxiteis in the range of 0.003 to 0.008mass%l2. Because of this low content it is not economicto mine bauxite for the extraction of gallium alone. The amount of gallium available frombauxite is thus limited by the amount of aluminium oxide that is produced.Hence, it is necessary to search for alternative sources of gallium in order to satisfythe growing demand for this metal. It was reportedl3] that coal ash contains 0.01 to0.1mass% Ga and that during coal gasification gallimn is concentrated into the fty ash.Also it is known that sormne minerals which are present in the iron ore gangue as well asin coke ash can feature relatively high gallium contents(国, e.g. nepheline KNa3[AISiO4]4max.0.064mass% Ga), germanite Cug(Ge,Fe)(S,As)4 (max. 1.8mass% Ga), sphaleriteZnS (max. 0.018mnass% Ga) and wurtzite β ZnS (max. 0.040mass% Ga). Since ironmakingblast furnaces turn over largc amounts of iron-bearing materials and coke it is of interestwhether gallium can be extracted as a by-product of the ironmaking process. However,little is known about the behaviour of gallium in the blast furnace. Yusfin!4i and co-workers中国煤化工MYHCNMHG79determined the gllium input and output at two Russian blast furnace shops (Table 1).The industrial data show that almnost all of the gallium introduced into the blast furnaceis dissolved in hot metal, and only traces of galliuim are found in slag and dust. It shouldbe noted. too, that the glliumn content of blast furnace sludge is relatively high (Table 1).Table 1 Measured input and output of gallium in g/t material and g/t hot metal(HM), respectively, at blast furnaces in Russial4IronmakingInputOutputplantCokeSinterPelletsHot MetalSlag DustSludgeSeverstal5-10 70- 100 5- 7(*)60- 110 Max.2 Traces 150 - 250(ing/t HM) 2-5 60 - 1202- 360- 110 Traces Traces 1.5 - 3.0Tulachermet 30- 40 20 - 40 5- 10(**) 30- 80 Max.3 Traces Max.200(in g/t HM) 15-25 25- 50 .2- 530 - 80 Traces TracesMax.2Note: (*) 一Kostamuscba pellets; (**)一: Lebedinski pellets.One can expect that Ga and Al in the blast furnace behave in a similar way since theseelements are vertical neighbours in the IIA group of the Periodic Table. However theindustrial data in Table 1 show the opposite because it is well known that most aluminiumintroduced into the blast furnace is found as Al2O3 in the slag. Therefore the aim ofthe present study is to clarify the distribution of gallium between the products of theironmaking process: hot metal, slag and top gas, by means of laboratory experimentscarried out under defined conditions. Chemical analysis of various types of coke werecarried out, too, in order to check whether the gallium content varies strongly dependingon the coke.2. Physico-chemical Behaviour of Gallium at Elevated TemperaturesSome physical and thermal properties[5,6] of gallium are as follows: atormic number 31问,atomic weight 69.7235, melting point (m.p.) 29.76C51, boiling point (b.p.) 2204C5, den-sity: sol. at 29.69C 5.904g.cm - -3[5] and liq. at 29.89C 6.095g.cm -3[5, viscosity 1.94mPa.sl6]at m.p, 1.266mPasl5 at 200°C and 0.578mPa.sl5 at 11009C, enthalpy of fusion (at m.p.)80.2J.g -3[5, heat capacity (at 25C) 25.86J mol 1.K - 1例), thermal conductivity (at 27C)0.406W.cmn-1.K-16. Gallium is one of the few metals which can be liquid near roomtemperature and features the longest liquid range of any element. The vapour pressures ofgallium and several other metals as a function of temperature can be seen in Ref.[6]. .In its corpounds, the valency of gallium is usually 3. The monovalent state is unstable,although the monovalent gallium compounds Ga2O and GaCl can be isolated2. Liquidgallium oxidises rapidly and forms a protective layer of oxide. The oxygen compounds ofgallium resemnble those of aluninium in that there are bigh and low temperature formsof Ga2O3 and two hydroxides, Ga(OH)3 and GaO.0H. The melting point of Ga2O3 is1800°C171. It is evident from Ref.[7] that Ga2O3 is less stable than Al2O3 but more stablethan the oxides of other group IIA elements. Data in Ref.[7] also shows that Ga2O3 isreduced by carbon at temperatures higher than 1275K.With elements of group V, gallium forms binary compounds, of which GaAs and GaPare of great industrial importance. They can be prepared by direct combination of the中国煤化工MYHCNMHG80.elements at high temperaturel2. The melting point of GaAs is 12389C/7].Gallium readily alloys with most metals. The Fe-Ga phase diagraml8] shows that at atemperature of 16009C iron and galliun are completely niscible. The solubility of Ga iny-Fe is limited to 3.5mass% at 1140°C but there is a large solubility range for c-Fe. Atroom temperature, the solubility of Ga in a-Fe is estimated at approximately 14mass%l8.3. Consumption and Application of GalliumGalliumn has limited commercial applications in its metallic form. Its principal use isin the manufacture of sermiconducting compounds, mainly gallium arsenide (GaAs), butalso GaP and GaN9. GaAs is u8ed in optoelectronic devices which include light-emittingdiodes (LED), laser diodes, photodetectors and solar cells. Gallium arsenide-based laserdiodes are used in such items as CD and DVD players10]. GaAs is also 1used in integratedcircuits for the wireless communication market as well as in defence applications. Smallamounts of gallium are 11sed in specialtyalloys as well as for rcsearch purposes,0 I~for example in the gallium solar neutrinodetector4l.Since the first commercial application ofLEDs for digital watches and hand-held cal-culators in the mid- 1960s, the demand fogallium has been growing. The developmentof the world primary gallium production12]is ilustrated in Fig.1.The rising demand for galliun and the1990 1992 1994 1996 1998 2000act that primary gallium production deYearpends on the amount of processed bauxiteand zinc residues implies that alternativerig.! Estimatedworld primary galiumsources of gallium may become attractive.4. Experimental ProcedureTwo types of experiments were carried out. The first consists in achieving equilib-rium with respect to gallium between hot metal and synthetic slag which were held in agraphite crucible at a temperature of 1673K under a flow of forming gas F10 (90vol.%N2+ 10vol.%H2). The second consists in treating hot metal alloyed with Ga at a reducedpressure of the gas phase in a vacuum induction furnace.The metal/slag experiments were carried out in a furnace featuring a horizontal graphitereaction tube with 40mm diameter and 160mm in length, which is heated by HF induction.Two types of synthetic slags with compositions similar to those of blast furnace slags wereprepared by premelting (Table 2). After solidifcation in air, the slags were ground to finepowders. Approximately 2g of slag were put together with 2g of pig iron in a graphitecrucible with an inner diameter of 16mm and a height of 27m. The composition of thepig iron is also given in Table 2. Gallium was added either as pure metal (99.99%) tothe iron phasc or as Ga2O3 to the slag phasc in order to achieve equilibrium from bothsides. The crucible with the charge was inserted into the hot zone of the furnace and was中国煤化工MYHCNMHG. 81held there for 30nin at the experimental temperature of 1673K. Preliminary experimentsshowed tbat this time is sufficient to reach equilibrium. Longer holding times would havecaused excessive loss of Na2O due to its volatilisation. The temperature was neasuredusing a PtRh18 thermocouple placed in the gap between the crucible and the reactionchamber. At the end of experiment the crucible was withdrawn from the furnace andcooled in air. The Ga content of metal and slag was analysed by inductively coupledplasma mass spectrometry (ICP-MS) using an ARL 35 000 ICP spectromneter.Table 2 Chemical composition of premelted slags and pig iron (in mass%)CaOSiO2MgOAl2O3Na2OBSlag A36.440.911.46.74.600.9 1.0Slag B32.746.910.96.4.140.70.8Pig Iron4.5%C0.7%Si0.2%Mn 0.03%SNote: B1 = CaO 7 SiO2; B3 = (CaO+MgO) 7 (SiO2 + Al2O3).The second type of experiment consisted in vcuum treatment of hot metal alloyedwith Ga. The experimental set-up is described in Ref.[13]. The procedure is as follows:20kg pig iron the composition of which is given in Table 2 was charged in a graphitecrucible featuring an inner diamneter of 15cm and smelted under an argon atmosphere.After stabilising the temperature at 1673K, the melt was alloyed with gallium. Afterhomogenisation, the vacuum pumps were switched on and chamber pressure was reducedto 30Pa. The pressure was measured using a Pirani galuge and was kept constant at 30Paby introducing argon into the gas-tight reaction chamber at a controlled rate. Samples weretaken from the liquid metal at time intervals of 10min. The samples were analysed with theaforementioned ICP-MS spectrometer. The melt temperature was monitored continouslyduring the treatment using an optical pyrometer and occasional measurements were madeby immersing a Pt-Pt/18%Rh thermocouple into the melt, too. It is assumed that thetemperature was controlled within +15K.5. Results and Discussion0.030The experimental results on the distribu-tion of Ga between metal and slag are illus-trated in Fig.2. This figure shows that most0.020gallium is found in the liquid iron, and amaximum of 0.027mass% remains in the slag.The data shown in Fig.2 correspond to dis-告0.010.tribution ratios LGa in the range of 0.003 to▲O SlagA;▲ ■Ga added to metal0.03, where LGa= mas8%(Ga)/ mass%[Ga].■口SlagB; 0口Ga added to slagIt seems that slag basicity has no effect on0.000.02.04.0the distribution of Ga but due to the largeGa in metal, mass%dlata scatter this should be further investi-gated. It is evident that Ga2O3 is not stable Fig.2 Distribution of gllium between hot metalunder the current experimental conditions.and slag (T= 1673K, holding time 30min,Gallium oxide is easily reduced by carbongas atmosphere: 90vol.%N2 + 10vol.%H2,gas rate 801/h).fromn the crucible and/or by carbon dissolved中国煤化工MYHCNMHG82in liquid iron, whereby glliuim dislves in liquid iron, e.g. according to the fllowingreaction(Ga2O3) + 3Cig) = 2[Ga] + 3C0Thermodynamic datal7l for the reaction given by the following equationGa2O3 + 3Cig) = 2Ga() + 3COindicate that the reduction of pure Ga2O3 with plre solid carbon procceds at tempera-tures higher than 1275K. However, if the reaction product, gallium, dissolves in iron, itsactivity becornes lower than unity, which facilitates reduction at even lower temperatures.It can be seen from Fe-Ga phase diagram[8l that Fe and Ga are completely miscible atsteelmaking tenperatures and that even at room temperature the solubility of Ga in Feis considerable. It is clear that in the presence of iron and carbon Ga2O3 is not stableat elevated temperatures. Hence the output of gallium from the blast furnace takcs placewith the hot metal since Ga2Oz in the slag is not stable under the reducing conditions ofthe blast furnace ironmaking process.On the other hand, gallium features a rclatively high vapour pressure at steelmakingtemperaturee, e.g. at 1873 K the vapour pressure of pure liquid gallium P&a is 1355Pal6)This value is comparable to the pressure of manganese at the same temperature, PMn= 5365Pa, while the vapour pressure of iron is only 5Pa at 1873K(6l. That is why thepossibility to extract solute gallium from the hot metal by means of vacuum distillationwas studied. It was found that alfter 20min of treatment at a pressure of 30Pa the galliumcontent of hot metal remained esentially unchanged at 0.016mass%, while the manganesecontent was reduced from 0.16 to 0.12mass%.It is known that a prerequisite for the evaporation of a solute element i from liquidiron is a relatively high partial vapour pressure Pi = pq .Yi.Xi (T = const.), where p,7i; and X; denote vapour pressure of pure liquid i(Pa), Raoultian activity cofficient of iat infinite dilution in liquid iron and mole fraction of i in the melt, respectively. Sincegallium was not removed from the iron melt by vacuurm treatment, while Mn evaporated,it follows that pGa《PMn. Under the current experimental conditions XGa=0.0011, XMn= 0.0014, Pra = 181Pal6, pin = 893Pal6] and 7Mn = 0.4314. The relationship PGa《PMnholds only if 7Ga <27 (T = 1673K).If it is assumed that R . T . lmyGa = const., it follows that at a temperature of 1873K,7YGa <19. Taking into account that carbon increases the activity of the solute elementsaluminiumn and boron at infinite dilution in liquid ironl15, it can be expe.ted that theactivity coeficient of gallium in carbon-saturated iron melts is higher than that in pureliquid iron (YEe- Cat.-Gay > N(Fe -Ga) since B, Al and Ga pertain to the 8ame group IIA inthe Periodic Table. It is also known that aluminium and boron feature very low Raoultianactivity coffcients at infinite dilution in liquid iron, i.e. YAI = 0.029 and rB = 0.022[15].Hence it is supposed that at low gallium contents Fe and Ga form solutions which do not ;feature strong positive deviation from Raoult's law. It is more likely that Fe and Ga formidcal solutions or solutions featuring negative deviation from Raoult's law, ie: YGa≤1 (T= 1873K). This may explain the fact that only traces of gallium are found in the fue dustof blast furnaccs (Table 1). Obvimusly, gallium which is introduced into the blast furnace中国煤化工MYHCNMHG33as a trace element in complex minerals of the iron ore gangue and coke ash dissolves readilyin iron and is not released to the gas phase due to its low activity in iron solutions.Another aim of this study was to determine the gallium content of various types of coke.The analysis was carried out in the chemical laboratory of the Institute of Iron and SteelTechnology (Freiberg). The ash content was determined according to DIN 51719. Thegallium contents (mass%) of the ash in various types of coke are 0.030 in north Goonjella(Australia), 0.029 in Burton (Australia), 0.028 in Lolberg (Germany) and 0.028 in Elkview(Canada), respectively. It was found that various cokes feature approximately the sameGa content. At a coke rate of 400kg/t hot metal (HM) and a content of 0.03mass% Ga incoke ash the input of gallium with coke in the blast furnaces is 12g/t HM asuning anaverage ash content of 10%).6. ConclusionsCoke ash contains approximately 0.03mass% gallium. Ga2O3 is not stable in blastfurnace slags. This oxide is easily reduced whereby gallium is dissolved in iron. At lowgallium contents Fe and Ga probably form ideal solutions or solutions featuring a slightnegative deviation from Raoult's law. In the blast furnace gallium is retained in hot mctaland not transferred to the gas phase. It is concluded that the extraction of galliun fromn themain products of the ironmaking process is not attractive. It is recornmended to investigatethe possibility of gallium extraction from blast furnace sludge.Acknowleigerments- The financial support of the EU-sponsored TACIS Project“Joint Education in NaturalResources Management" is gratfully acknowledged.REFERENCES1 U.S. Geological Survey, Mineral Commodity Summaries 2001, Galium.2 Ulmann's Encyclopedia of Industrial Chemistry, 6-th Ed., 1998 Electronic Release.3 A.N. Selikman and B.G. Korschunov, Metalurgija redkih Metallov (Metllurgja, Moskwa, 1991) p.283.4 Y.S. Yusin, PI. Chernousov, A.K. Zaitsev, A.L. Petelin and JA. Karpov, Izv. VUZ Chern. Met.41(11) (1998) 8.5 D.R. Lide (Ed.), CRC Handbook of Chemistry and Pbysics, 74-th Ed. (CRC Press, Boca Raton FL.1994).6 T. lida and R.I.L. Guthrie, The Physical Properties of Liquid Metals (Clarendon Press, Oxford, 1988)p.223.’ Outokumpu HSC Chemistry for Windows. Chemical Reaction and Equilibrium Software with ExtensiveThermochemical Database, Version 4.0 (999).8 T.B. Massalski (Ed.), Binary Alloy Phase Diagrams, 2-nd Edition (ASM International, 1990) p.1703.9 D.A. Kramer, Galltum and Galium Compounds, ASM Handbook, Volume 2, Properties and Selection:Nonferrous Alloys and Special-Purpose Materials, Section: Specific Metals and Alloys, ASM Inter-national 1990, in ASM Handbook: Metals Properties and Performance Collection CD-ROM (ASMInternational and The Dialog Corporation, 1998)10 D.A. Cramer, Gallium (U.S. Geological Survey, Minerals Yearbook 2000).11 Laboratori Nazionali del Gran Sasso (Italy), Anual Report 2000.12 Private communication by D. A. Crarmer (U.S. Geological Survey) on 22.01 .2002.13 L. Savov and D. Janke, ISIJ Intermational 40(2) (2000) 95.14 D. Drester, Can. Metall. Quarterly 28(2) (1989) 109.15 G. Sigworth and J. Elliott, Metal Science 8 (1974) 298.中国煤化工MYHCNMHG

论文截图
版权:如无特殊注明,文章转载自网络,侵权请联系cnmhg168#163.com删除!文件均为网友上传,仅供研究和学习使用,务必24小时内删除。