Search Documents

By R. E. Brewer

Pilot scale flotation tests were conducted with a l?ft-diam column at the Barrick Resources (USA) Mercur mine, Tooele County, UT, to scavenge unleached (0.01 to 0.018 oz/st) gold values from the tailings of a carbon-in-leach (CIL) circuit. The optimal recovery was 35 pct of the unamenable-to-leaching gold, at a grade of 0.353 ounces of gold per short ton, using a mixture of dithiophosphate, xanthate, and diesel fuel as a collector. This was achieved by floating approximately 1.6 wt pet of the CIL tailings. The gold values are believed to be contained in arsenic-bearing minerals, attrited carbon, and carbonaceous material in the ore. Redox potentials were measured with a gold electrode during the conditioning stage prior to flotation. The correlation between flotation recovery and redox potential values could be used to control reagent additions. Treatment of a previously leached and weathered tailings material, averaging about 0.055 ounces of gold per short ton, was also investigated. The tailings were first water leached at approximately 60? C, and the water-leached tailings, containing approximately 0.03 ounces of gold per short ton, were then fed to the column flotation unit. The effects of pH' and sodium sulfide addition on gold flotation recovery and redox potentials were examined. Maximum gold flotation recovery, 36 pct, was achieved at pH 8,5 with a sodium sulfide addition of 0.18 Ib/st. Overall gold recovery by leaching and flotation was in the range of 70 to 80 pet.

Jan 1, 1989

By R. F. Hammen

Heavy metal contamination of water is a problem of immense national and international proportions. However, effective cleanup is complicated both by the toxicity and the low concentrations of metals in wastewater streams. Using high performance chelation chromatography, the solid phase extraction and purification of extremely dilute metals from water was highly effective. The process is based on selective displacement of metals from a chromatographic column containing a chelating resin. The resin is characterized by extremely efficient adsorption kinetics which allow rapid and complete removal of the metals.

Jan 1, 1993

By Yong J. Kim, J. M. Park

"Solid/melt interface shape during continuous casting process has a strong influence on the resulting microstructure of the ingot. The interface shape is primarily governed by the heat flow condition between the molten metal and mold (or container). The roll of the microstructural transition resulting from various solid/melt interface shapes is discussed using the Sn-22.6wt%Pb alloy. Three different types of mold such as cold mold, resistantly heated mold, and electromagnetically heated mold were used to generate those interfaces having different curvatures. Experimental measurements of complete temperature profile along the axis and surface of the ingot during downward continuous casting are presented to determine the solid/melt interface geometry. The results provide an explanation of the ingot property with emphasis on microstructure.IntroductionA large scale of continuous casting process of alloys often has a rather deep V channel type of solid/melt interface resulting from a limited heat flow condition. As a result, the deep V type solid/melt interface shape entraps defects such as pores and inclusions and also creates canter segregation. In order to reduce the defects concentration, electromagnetic stirring(EMS) mechanism has been applied to the melt in the channel[l]. Speith and Bungeroth introduced first experimental EMS device to the continuous caster in 1952.[2] Today, with wide ranges of benefit, a large variety of such technique is available. Electromagnetic field on the melt can be divided into three aspects. First, eddy current is induced on the metal surface by alternating magnetic field with high frequency, which results in heating the metal. Second, electromagnetic body force which is generated by the magnetic field and the eddy current reduces the melt pressure on mold or support the melt not to contact with the mold[3,4,5]. This force is used for electromagnetic casting process(EMC). Finally, it causes stirring in the melt. Many research efforts have been done on the EMC process to improve surface quality of slab[6] and control meniscus shape where three phases such as melt, solidified metal, and mold powder coexist[7,8]. However, the eddy current heating effect on the surface of the ingot was less considered. It is worth noting that induced eddy current on the surface of slab melts the solidified shell and thus changes the heat flow condition between the mold and melt. Mold heating effect was first proposed by A. Ohno[9] and its processing conditions were identified in details[ 10, 11]. The so-called Ohno process is based on the fact that by keeping the mold temperature slightly above the melting temperature of the liquid, no grain can be nucleated on the mold wall[12]."

Jan 1, 1998

By Jinjing Dua, Jun Zhua, Kuiren Liub, Bin Wang, Xiaolei Wua

V-Ti-Fe as hydrogen storage alloy was prepared by reducing metal oxides with a metallothermic reduction method. The deoxidation process of the V-Ti-Fe alloy was performed by electro-deoxidation and vacuum induction melting method, respectively. The results showed that the content of oxygen impurity can be controlled at about 0.2wt% after electrolyzing for 240min. Both the maximum values of hydrogen absorption capacity and the effective hydrogen desorption capacity are 2.65wt% and 1.01wt%, respectively. The deoxidization effect of vacuum induction melting process is more remarkable, with Ce as the reducing agent. It is found that Ce addition is beneficial for reducing the componential segregation, promoting the hydrogen absorption activation of V-Ti-Fe alloy. When m(Ce)/m(alloy) is up to 5%, the oxygen content has been reduced to 0.05wt% and the maximum hydrogen absorption capacity and the effective hydrogen desorption capacity are up to 3.36wt% and 1.85wt%, respectively.

Jan 1, 2015

By Simon Lekakh, Von Richards, C. Hank Rawlins, Kent Peaslee

"Batch aqueous leaching and carbonation tests were conducted using industrial steelmaking slags to determine the effect of carbonic anhydrase enzyme as a catalyst. Calcium leaching is a strong function of particle surface area, and the extent can be expressed as a function of time and particle size. Carbonic anhydrase did not affect the calcium-leaching rate, however, it did catalyze calcium carbonate formation to achieve a neutralization time near the theoretical rate. Additionally, carbonic anhydrase modified the precipitate morphology due to accelerated particle nucleation. Time controlled tests in which the pH dropped to ~6 decreased the amount of carbonate produced, and this effect was exaggerated by carbonic anhydrase, while pH controlled tests (>8.5) exhibited the highest rate of carbonation. Because the leaching rate was ~50% faster than the carbonation rate, a further increase in the amount of carbonation may be realized by using carbonic anhydrase however pH must be >10.3.IntroductionSteelmaking slag contains high fractions of alkaline earth oxide based phases (i.e., CaO and MgO) that exothermically form carbonates. Thus, it is being considered as a means of permanent carbon dioxide sequestration. At current U.S. steel production rates, approximately 13-17 million tons/year of slag are generated from basic oxygen furnace (BOF), electric arc furnace (EAF), and ladle metallurgy furnace (LMF) processes [1]. BOF and EAF slag is used as high quality mineral aggregate or cement clinker, while LMF slag, produced at ~15% the rate of BOF and EAF slags, has limited use as blast furnace flux or acid mine neutralization with the majority going to landfill [2]. Based on slag’s chemical nature and immediate availability as a co-product from steel production, the steel industry initiated a project to investigate hydrous carbonate formation in steelmaking slag as a method of sequestering carbon dioxide emitted in steelmaking offgas [3]. The goal of this project is to determine the reactor parameters suitable for the design of an industrial system for treating steelmaking offgas with raw or minimally processed slag."

Jan 1, 2009

By G. S. Hill

A large amount of the stereological error associated with liberation studies can he eliminated by assessing areas instead of linear intercepts. The presentation of a simple geometrical model as a stereological 'worst case' allows calculation of the worst possible grade/recovery curve for a given assemblage of areas. The true results are expected to lie in the window between this worst case and the hest possible case, represented by uncorrected data. The width of this window typically corresponds to an error margin of about 5% grade at a given recovery. An indication of where the data lies within the range of possible grade/recovery results may he obtained from a qualitative analysis of the locked section complexity. Assemblages consisting of texturally complex locked sections (ie. containing many mineral nuclei) are likely to exhibit minimal stereological bias and therefore will tend to lie close to the uncorrected side of the grade/recovery window.

Jan 11, 2011

By G. A. Irons, D. Colbert

Several electric arc furnace (EAF) shops have increased the zinc and lead grades of EAF dust by reinjecting it into their furnaces; however, the kinetics of this operation are poorly understood. To understand the kinetics, EAF dust was injected into 70 kg heats of iron or steel in an induction furnace. Supplemental additions of carbon and oxygen to overcome the heat load on the system were also investigated. In all cases, the dust was chemically transformed, and the zinc content increased from 16% Zn to 60 - 80% Zn. About 50% of the zinc in the EAF dust dissolved in the iron or steel. The zinc oxide in the dust is thought to be reduced by carbon and the zinc vapour subsequently dissolved in the iron. A kinetic model for the dissolution process is proposed, and the implications for EAF dust recycling are discussed.

Jan 1, 2000

By Y. K. Rao

Relatively pure crystalline lead sulfate (PbSO4) can be produced by chemical vapor transport from an impure source such as the lead-acid battery paste or the roast-ore-leach-residue from a hy¬drometallurgical process-route. In a closed-reactor, transport of matter is induced by a tempera¬ture-gradient and the steady-state transfer rate can be calculated using the data on the standard free energies of formation for the species present and the pertinent binary diffusion coefficients. The present work focusses on chemical vapor transport in a tubular reactor (1.4-cm inner diameter x 10-cm long) with the source- and deposition-ends held at 850°C (T,) and 750°C (Td), respectively; PbC12(g) is the 'solvent' or 'transport agent' and NiSO4(s) placed in the source-zone -serves to maintain a sufficiently high S0,-potential such that the stability of the lead sulfate species is ensured. For the purposes of analysis, the gas-phase in the reactor is assumed to consist of eight species: H2O, PbC12, SO3, SO2. O2 HO, C12, and PbC14. The reaction between C12(g) and PbSO4(s) in the high-temperature source-zone produces PbCl2, SO2, and 02; four more reaction-equilibria in the same zone yield other species. In the low-temperature deposition-zone, reaction equilibrium involving PbC12, H2O, and SO, leads to the nucleation and growth of PbSO4 crystals; three other reaction-equilibria occur in this zone. By employing the virtual equilibrium state for the end-zones and coupling it to the Stefan-Maxwell diffusion equations for each of the gas species, the respective partial pressures for these, in the source- and deposition-zones, are computed at steady-state. From these data, the molar fluxes of PbCl2. PbCl4, and so forth, are determined. Allowing for the transport-enhancing effects of thermal convection, the net transport rate of lead (or PbSO4) is duly calculated for specified initial conditions. The predicted rates show very good agreement with measured rate data reported in the literature [ V. Plies, T. Kohlmann and R. Gruehn: Z. anorg. allg. Chem, 568, 62 (1989)

Jan 1, 1995

By Md. Maniruzzaman

The quenching performance of a liquid quenchant can be controlled by controlling the agitation in the quench tank. Without agitation, a vapor layer or bubbles form at the hot part surface and the natural convention of the quenchant limit the heat extraction rate at the quenched part surface. Resistance to the heat extraction rate can be reduced by breaking the vapor layer or by inducing a turbulent convection at the part surface through agitation (Le. forced circulation of the quenchant). To determine the effects of agitation on the quench performance of a mineral oil based quenchant, the CHTE quench probe system was used with AISI4140 steel probe tip. Quenching performance of the oil is evaluated by computing the effective heat transfer coefficients of the steel probe as a function of surface temperature using Lumped Parameter Analysis. In addition, computational fluid dynamics (CFO) simulation of the quench tank with an impeller type agitator has been done to understand and predict the fluid flow pattern in the tank. CFO simulation also provides insights for determination of optimal location of the impeller and the quench probe tip. Keywords: CHTE Quench Probe, agitation, CFD simulation, liquid quenchant, 4140 steel

Jan 1, 2003

By Donald Voit

"The reaction between zirconia/zircon, coke, and chlorine involves a slow reaction between the coke and the zirconia/zircon and a fast reaction with chlorine. Because of the fast chlorine reaction, the reaction rate for zirconia/zircon is nearly independent of chlorine concentration for most conditions. The reaction between zirconia/zircon and coke is particle size controlled and related to the degree of physical contact between the zirconia/zircon and the coke. Since particle-particle contact is critical, compaction of finely ground coke and zirconia/zircon produces the highest reaction rates. Uncompacted, coarse coke and zirconia/zircon has the slowest reaction behavior with static bed tests.SummaryZirconia reacts with chlorine in the presence of carbon to form zirconium tetrachloride, carbon monoxide, and carbon dioxide. A similar reaction occurs with zirconium silicate with the additional formation of silicon tetrachloride. The reaction rate depends primarily on temperature and the intimacy of contact between the zirconia and the coke. The reaction is multi-stepped with a slow reaction between coke and zirconia to form an intermediate that reacts rapidly with chlorine, causing the order of the reaction for chlorine to be near zero. Chlorination rates in the range of 1 µmole/hr/cm2 are seen for fluid bed reactors. Plug flow, and static bed reactors have reaction rates of the order of 1 mmole/hr/cm2, where the area is based on the zircon surface area in plug flow and fluid bed reactors and the boat cross sectional area in a static bed. Chlorination rates measured in zirconia/coke briquettes are higher with rates as high as 400 mmole/hr/cm2 reported. High chlorine diffusion resistance, not accounted for in the analysis of experimental results, has led to a lot of confusion in the literature interpretation of chlorination data.In general coke and zirconia/zircon particle size, coke and zirconia/zircon intimacy, coke to zirconia/zircon ratio, chlorine concentration, and temperature are important variables for chlorination. Decreasing the coke and zirconia/zircon particle size generally increase the reaction rate. The coke to zirconia/zircon ratio affects the reaction in a more complex way. Increasing the chlorine concentration generally increases the reaction rate, but the effect appears to be more related to the concentration dependency of diffusion than to its role in the kinetics. The reaction rate increases with temperature but the exact behavior is complicated by temperature effects on diffusion."

Jan 1, 2011

By R. Hurman Eric

The electrode in a slag-resistance electric furnace is the central factor in the design and operation of the furnace by its role of conveying the smelting power, i.e. heat needed for the operation. The control parameters are the size and positioning in the slag bath which are defined by slag resistivity, furnace resistance, cell constant and power input sequence all of which are closely interrelated. The flow of heat generated by the electrode current determines the energy distributions and thus the furnace dimensions. Heat flow models concerning vertical, horizontal and combined two-dimensional flows are discussed. All these considerations are based on the experimentally determined slag properties such as its resistivity.

Jan 1, 1995

By Srinath Viswanathan, Adrian Sabau

"In general, the occurrence of microporosity during metal casting is due to the combined effects of solidification shrinkage and gas precipitation. The governing equations for fluid flow and hydrogen evolution indicate that porosity formation and fluid flow are strongly coupled. However, in most studies on microporosity, it is considered that the porosity formation does not influence the fluid flow in the mushy zone. In this study, a computational methodology is presented for the numerical simulation of interdendritic fluid flow and microporosity evolution. The solution algorithm presented includes a fully coupled, implicit treatment of microporosity and local pressure in the mushy zone. The effects of microporosity evolution due to the local pressure drop in the mushy zone, and pore expansion in casting regions where liquid feeding alone cannot compensate for the solidification shrinkage are also considered. It is shown that neglecting the effect of porosity formation on the pressure in the mushy zone yields higher pressure drops. By its growth, microporosity partially compensates for the solidification shrinkage, reducing the feeding demand. A reduced feeding demand requires less fluid flow to compensate for solidification shrinkage, and results in smaller pressure drops in the mushy zone. Therefore, in order to accurately describe casting defects, comprehensive models of fluid flow, heat transfer, and solidification must include the effect of microporosity as well.IntroductionIn the terminology commonly used by foundrymen, porosity is usually considered to be either ""hydrogen"" or ""shrinkage"" porosity. Hydrogen porosity is the term given to porosity that is generally rounded, isolated, and well distributed. Porosity that is interconnected or clustered, and of an irregular shape corresponding to the shape of the interdendritic region, is usually termed shrinkage. In general, the occurrence of microporosity in aluminum alloys is due to the combined effects of solidification shrinkage and gas precipitation [1]. Continuum models [1, 2- 7] have been used to predict the level of porosity in castings. A number of other studies have also attempted to understand the phenomena of porosity formation and pore growth [8-10] and pore morphology [11].The governing equations for fluid flow and hydrogen evolution indicate that porosity formation and fluid flow are strongly coupled. However, in most studies [4, 9, 10], it is considered that the porosity formation does not influence the fluid flow in the mushy zone. Kubo and Pehlke [3] presented a methodology for the prediction of microporosity distribution in shaped castings by considering the effects of both the hydrogen precipitation during solidification and the pressure drop in the mushy zone. In their approach, the local pressure and porosity were computed in an uncoupled manner based on a ""flux of interdendritic liquid"" criterion, i.e., either pressure or porosity was first computed using the other variable at the previous time step. Their approach has been used with little change in numerous studies [6, 12, 13]. Recently, Kuznetsov and Vafai [14] presented a fully coupled, one-dimensional model for the computation of pressure and porosity. They showed that neglecting the effect of porosity formation on the pressure in the mushy zone yields higher pressure drops and an over-prediction of final porosity. They also showed that the influence of porosity formation on the pressure is larger at lower pressures in the mushy zone."

Jan 1, 2001

By Michel Boisvert

Noranda Metallurgy is presently completing construction of the new Noranda Converter (NCV) at its Home Smelter in Rouyn-Noranda. This capital investment of CDN$56M is the continuation of the Emissions and SO2 Reduction Program that was initiated in the late 1980s with the acid plant for Reactor off-gas treatment. The Patented Noranda Converter was developed and pilot tested at the Home Smelter in the early 1990s. The project will replace the Peirce-Smith Converters and enable sulphur fixation to be 'increased from 70% to over 90% by 2002 with an accompanying reduction of dust emissions to the atmosphere. The construction site is an extension of the present Noranda Reactor area. A number of existing facilities including the acid plant and process air supply were used to minimize the capital cost of the project. This paper provides a brief history of the Noranda converting process, the Home Smelter Emissions and SO2 Reduction Program, ,project implementation and construction, a description of the different facilities, and preparation of human resources for start-up.

Jan 1, 1998

By Luciano E. Ramírez, Jesús L. Valenzuela, José R. Parga

"An innovative process for removing copper cyanide complexes including silver and zinc from cyanidation circuits has been developed. The technology was based on the inducing the nucleated precipitation of copper and silver in a tube serpentine reactor, using sodium sulfide as the precipitate and sulfuric acid as pH control. The results showed that pH had a great effect on copper cyanide removal efficiency and the optimum pH was about 3 to 3.5. At this pH value copper cyanide removal efficiency could be achieved above 97 and 99 % ,when influent copper concentration ions were 650 and 900 ppm respectively. In this process (sulphidizaton-acidification-recycle-thickening), the cyanide associated with the copper cyanide complexes is released as HCN gas under weakly acidic conditions, allowing it to be recycled back to the cyanidation process as free cyanide. This system was successfully commercial applied at Minera William in Mexico.IntroductionThe recovery of gold and silver from minerals and concentrates with cyanide is an important hydrometallurgical process, which has been studied for more than 124 years. Patented by MacArthur in 1888 [1], the process relies on the fact that gold dissolves in aerated cyanide solutions to produce the gold cyanide complex. The stoichiometry by wich this reaction occurs is known as the Elsner equation as shown here. Silver is accomplished in the same fashion."

Jan 1, 2012

By Wenjun Duan, Kun Wang, Qingbo Yu, Qin Qin

"A novel oxygen production technology noted chemical looping air separation (CLAS) was introduced and properties of Cu-based oxygen carriers used were investigated. The oxygen carriers were prepared by mechanical mixing. Al2O3, Fe2O3, ZrO2, TiO2 and SiO2 were used as binders during the preparation. XRD patterns show that spine! oxides are formed using Al2O3 and Fe2O3 as binders. Phases of CuO and binders keep stable using ZrO2, TiO2 and SiO2 as binders. BET values of oxygen carriers increase in the order of ZrO2>TiO2>SiO2>Fe2O3>Al2O3. The effects of temperatures, binders, particle sizes and flow rates on reduction/oxidation reaction rates were studied in TG experiments. The reduction reaction rate increases in the order of ZrO2>TiO2>SiO2. The effect of binders on the reactivity of oxidation is not obvious. The reduction and oxidation reaction rates both have the tendencies of increasing with the temperature increasing and with gas flow decreasing. However, the effects of gas flow and particle size are little.IntroductionChemical looping air separation (CLAS) is a novel oxygen production technology. The oxygen carriers used in CLAS release oxygen in reduction reactor and regenerate in the oxidation reactor. Selecting suitable oxygen carriers is the key for CLAS. Oxides of CuO, Mn2O3, Co3O4, Pb3O4, CrO2, SrO2, PdO2 and CeO2 have the property of releasing oxygen at suitable temperature and oxygen partial pressure [1]. Among these oxides, CuO, Mn2O3, Co3O4 are the most common compositions used to prepared oxygen carriers for chemical looping combustion (CLC) or chemical looping with oxygen uncoupling (CLOU) [2-5]. Moreover, thermodynamic calculations and preliminary experiments were carried out on oxides of Cu (CuO/Cu2O), Mn (Mn2O3/Mn3O4) and Co (Co3OJCoO) by Moghtaderi to ascertain their feasibility for CLAS process [6]. CuO has the highest oxygen transport capacity (0.1 g 02/g CuO compared to 0.03 g 02/g Mn2O3 and 0.06 g 02/g Co3O4). However, agglomeration is the biggest issue for copper oxides used at high temperature because of its low melting point (l 450°C for CuO, 1235°C for Cu2O and 1075°C for the mixture of CuO and Cu2O) [7]. In order to improve the capability of anti-sintering and stability ofCuO, binders (ZrO2, MgAl2O4, TiO2, SiO2, Fe2O3, Al2O3, cement) are added during the preparation of Cu-based oxygen carriers [8]. In this paper, the Cu-based oxygen carriers were prepared by mechanical mixing method. TiO2, Al2O3, Fe2O3, ZrO2 and SiO2 were selected as binder. XRD, BET and SEM were used to analyze the phase, surface area and surface morphology of oxygen carrier prepared. The reactivity of oxygen reduction and oxidation of oxygen carriers prepared were investigated in STA409PC thermal analyzer."

Jan 1, 2014

By L. Sh. Tsemekhman

A new technology for the treatment of copper nickel-containing concentrate after high-grade matte flotation to blister copper with the production of liquid silicate slags in a vertical autogenous unit with upper oxygen blow has been considered. The copper and nickel behavior in the process of sulphide copper concentrate interaction with silicate melt containing oxide copper has been studied. The general mechanisms of such interaction are found.

Jan 1, 2003

By Dai Yongnian

The methods of treating zinc crust and its development in the last hundred years are compared and reviewed. There are better effects from normal pressure metallurgy turned into vacuum metallurgy. The traditional methods dezinc only. But we must consider that delead on further step. From lab tests to industrial scale experiment are researched. The first step-dezinc, second-delead, crude ,metal of, zinc, lead and silver are produced, The flowsheet of treating zinc crust are simplified, material consumption reduced, work condition improved, environment protect mended. This method is pointed out that its application can be expanded.

Jan 1, 1992

By Guangshi Li, Xionggang Lu, Xueliang Xie, Xingli Zou, Yuyang Bian, Hongwei Cheng, Kai Zheng, Changyuan Lu

The corrosion resistance and corroded microstructure of MgO-C refractory in molten slag are investigated by immersion test in muffle furnace at 1600 °C . The microstructure and chemical compositions of the specimens after erosion experiment are characterized by scanning electron microscope (SEM) and energy dispersive X-ray (EDX). The results confirm that the oxidation of graphite is the most important degradation mode of MgO-C refractory. The structure of the refractory material is damaged due to the decrease of graphite, as a result, the periclase substrate reacts with the molten slag directly. The corrosion process is controlled by the dissolution of the refractory materials into slag, low melting point oxides are therefore generated which lead to the damage of MgO-C refractory finally.

Jan 1, 2015

By S. Yuan

Oxygen ion conducting yttria-stabilized zirconia (YSZ) tube was used to remove oxygen from molten Cu at 1373K. Outer surface of the tube was in contact with the melt and the inner surface containing a thin layer of porous Ni-ZrO2 cermet electrode was in contact with a reducing gas. The migration of oxygen species (O2-) from the Cu melt through the YSZ tube is accompanied by an equal and opposite flux of electrons maintaining electroneutrality. Without applying an external EMF, the fluxes are maximized by short circuiting the melt with the cermet electrode. For efficient oxygen removal, the YSZ tube must be thin, contact resistances should be negligible, and the electrodes must be porous and cover the entire electrolyte surface for charge transfer. This refining process is environmentally sound, can use reducing gases such as H2, CH4 and CO without running the risk of increasing H or C content of the melt.

Jan 1, 1993

By KNona C. Liddell

Goethite is an important naturally occurring sorbent for divalent heavy metals and has often been used as a model compound in studies of adsorption on oxide mineral surfaces. Surface complexation modeling, surface spectroscopy, and electrochemical techniques have all been applied to goethite adsorption. Modeling studies have shown that usually several different sets of fitted parameters are equally successful in representing surface charging and adsorption data. Surface spectroscopic data are proving valuable in determining the nature of the surface sites, and refined models now incorporate structural information on interatomic distances and coordination numbers. Multisite surface complexation models can give very good fits to surface charging and adsorption data, but further improvements are needed to better represent surface reactions involving exchange of protons and divalent cations. Recent experimental and modeling studies on adsorption of divalent transition metals on goethite are described and the present status of the field is assessed.

Jan 1, 1999