Metallic impurities, deposition

L. Mouche, P. Tardif, and J. Derrien, Mechanisms of metallic impurity deposition on silicon substrates dipped in cleaning solution, J. Electrochem. Soc. 142, 2395, 1995. 

Crude lead contains traces of a number of metals. The desilvering of lead is considered later under silver (Chapter 14). Other metallic impurities are removed by remelting under controlled conditions when arsenic and antimony form a scum of lead(II) arsenate and antimonate on the surface while copper forms an infusible alloy which also takes up any sulphur, and also appears on the surface. The removal of bismuth, a valuable by-product, from lead is accomplished by making the crude lead the anode in an electrolytic bath consisting of a solution of lead in fluorosilicic acid. Gelatin is added so that a smooth coherent deposit of lead is obtained on the pure lead cathode when the current is passed. The impurities here (i.e. all other metals) form a sludge in the electrolytic bath and are not deposited on the cathode. 

The most important undesired metallic impurities are nickel and vanadium, present in porphyrinic structures that originate from plants and are predominantly found in the heavy residues. In addition, iron may be present due to corrosion in storage tanks. These metals deposit on catalysts and give rise to enhanced carbon deposition (nickel in particular). Vanadium has a deleterious effect on the lattice structure of zeolites used in fluid catalytic cracking. A host of other elements may also be present. Hydrodemetallization is strictly speaking not a catalytic process, because the metallic elements remain in the form of sulfides on the catalyst. Decomposition of the porphyrinic structures is a relatively rapid reaction and as a result it occurs mainly in the front end of the catalyst bed, and at the outside of the catalyst particles. 

The metallic impurities present in an impure metal can be broadly divided into two groups those nobler (less electronegative) and those less noble or baser (more electronegative) as compared to the metal to be purified. Purification with respect to these two classes of impurities occurs due to the chemical and the electrochemical reactions that take place at the anode and at the cathode. At the anode, the impurities which are baser than the metal to be purified would go into solution by chemical displacement and by electrochemical reactions whereas the nobler impurities would remain behind as sludges. At the cathode, the baser impurities would not get electrolytically deposited because of the unfavorable electrode potential and the concentration of these impurities would build up in the electrolyte. If, however, the baser impurities enter the cell via the electrolyte or from the construction materials of the cell, there would be no accumulation or build up because these would readily co-deposit at the cathode and contaminate the metal. It is for this reason that it is extremely important to select the electrolyte and the construction materials of the cell carefully. In actual practice, some of the baser impurities do get transferred to the cathode due to chemical reactions. As an example, let the case of the electrorefining of vanadium in a molten electrolyte composed of sodium chloride-potassium chloride-vanadium dichloride be considered. Aluminum and iron are typically considered as baser and nobler impurities in the metal. When the impure metal is brought into contact with the molten electrolyte, the following reaction occurs... 

Efficient refining of the more volatile actinide metals (Pu, Am, Cm, Bk, and Cf) is achieved by selective vaporization for those (Pu, Am, Cm) available in macro quantities. The metal is sublimed at the lowest possible temperature to avoid co-evaporation of the less volatile impurities and then deposited at the highest possible temperature to allow vaporization of the more volatile impurities. Deposition occurs below the melting point of the metal to avoid potential corrosion of the condenser by the liquid metal. Very good decontamination factors can be obtained for most metallic impurities. However, Ag, Ca, Be, Sn, Dy, and Ho are not separated from Am metal nor are Co, Fe, Cr, Ni, Si, Ge, Gd, Pr, Nd, Sc, Tb, and Lu from Cm and Pu metals. 

If an actinide metal is available in sufficient quantity to form a rod or an electrode, very efficient methods of purification are applicable electrorefining, zone melting, and electrotransport. Thorium, uranium, neptunium, and plutonium metals have been refined by electrolysis in molten salts (84). An electrode of impure metal is dissolved anodically in a molten salt bath (e.g., in LiCl/KCl eutectic) the metal is deposited electrochemically on the cathode as a solid or a liquid (19, 24). To date, the purest Np and Pu metals have been produced by this technique. 

All subsequent preparations of Cf metal have used the method of choice, that is, reduction of californium oxide by La metal and deposition of the vaporized Cf metal (Section II,B) on a Ta collector 10, 30, 32, 45, 91, 97, 120). The apparatus used in this work is pictured schematically in Fig. 16. Complete analysis of Cf metal for cationic and anionic impurities has not been obtained due to the small (milligram) scale of the metal preparations to date. Since Cf is the element of highest atomic number available for measurement of its bulk properties in the metallic state, accurate measurement of its physical properties is important for predicting those of the still heavier actinides. Therefore, further studies of the metallic state of californium are necessary. 

Cadmium also may be recovered from zinc ores and separated from other metals present as impurities by fractional distillation. Alternatively, the cadmium dust obtained from the roasting of zinc ore is mixed with sulfuric acid. Zinc dust is added in small quantities to precipitate out copper and other impurities. The metal impurities are removed by filtration. An excess amount of zinc dust is added to the solution. A spongy cadmium-rich precipitate is formed which may he oxidized and dissolved in dilute sulfuric acid. Cadmium sulfate solution is then electrolyzed using aluminum cathodes and lead anodes. The metal is deposited at the cathode, stripped out regularly, washed and melted in an iron retort in the presence of caustic soda, and drawn into desired shapes. More than half of the world s production of cadmium is obtained by elecrolytic processes. 

Metallic impurities complicated zinc electrowinning [361-368] from acidic sulfate solutions and can adversely affect the cathodic current efficiency, cell potential, power consumption, deposit quality, and the overall polarization behavior of the cathode. 

When the pulverulent gold has entirely deposited, the liquid must be decanted or filtered off with the greatest precaution. Care must bo taken that not the smallest particle of the gold powder is allowed to pass away with tho liquid, A little hydrochloric acid, which must ho quite free from any admixture of nitric acid, is then to be poured upon the precipitate. This will remove any iron or other metallic impurities without dissolving the gold. The latter is then to be washed, at least six times, with successive portions of distilled waterj and lastly, it is transferred to a email porcelain Or platinum crucible, in which it is heated over a spirit lamp, till the last portions of water are expelled. It ought, in fact, to bo raised to a red heat, or even to be fused with a small quantity of borax and nitrate of soda, as formerly recommended, to expel the last traces of ohlorlde of silver. 

Other contributions to the polarization are activation polarizations (nact) caused by inhibition of the passage of ions through the phase boundary which may arise in the discharge mechanism. Films on the electrode may also contribute (e.g. oxide, metal already deposited, impurity) by offering a resistance to current flow differing from the bath resistance (ohmic polarization, Tlohm)- Hence the observed overvoltage is given by... 

In the case of oxides or sulphides, poisoning has been observed to be very weak [99,168]. One reason is that these materials, especially oxides, are usually obtained as porous layers [169, 170]. But another, more exciting explanation is that metal deposition on non-metallic surfaces is more difficult than on metals, since the bond formed between the surface and deposited metal atoms is weaker [168], Therefore, no underpotential deposition takes place, which pushes the potential range of impurity deposition to more cathodic potentials. If deposition takes place, it occurs at a high overpotential, and the resulting discharged particles form clusters rather than monolayers, thus leaving most of the active surface uncovered. 

Stability tests have shown [359] that these modified surfaces can be stable for more than 3000 hours. It is interesting that no deactivation has been observed in the presence of metallic impurities in solution. An activation of the surface has even been observed, which is to be related to the decrease in Tafel slope brought about by the deposition of small metal particles [167]. This is rather a proof that the original surface is not especially active for H2 evolution (cf. Section 4.6). 

Another source of deactivation is the usual presence of metallic impurities, in particular Fe, in technical solutions [25, 151, 446]. Fe is deposited on NiSx and can produce deactivation, but the effect is reduced on account of the larger surface area and the semimetallic nature of the surface. The addition of MoS2 is reported to result in an improvement in this direction as well. Ni impurities have been found not to deactivate NiSx electrodes. 

Oxide electrodes have been observed to be almost immune from poisoning effects due to traces of metallic impurities in solution [99]. This is undoubtedly due primarily to the extended surface area. It can be anticipated that the calcination temperature must have a sizable effect. But in addition, a different mechanism of electrodeposition must be operative. Chemisorption on wet oxides is usually weak because metal cations are covered by OH groups [479]. As a consequence, underpotential deposition of metals is not observed on Ru02, although metal electrodeposition does takes place. However, electrodeposited metals give rise to clusters or islands and not to a monomolecular layer like on Pt. Therefore, the oxide active surface remains largely uncovered even if metallic impurities are deposited [168]. Thus, the weak tendency of oxides to adsorb ions, and its dependence on the pH of the solution is linked to their favorable behavior observed as cathodes in the presence of metallic impurities. 

Clays are natural compounds of silica and alumina, containing major amounts of the oxides of sodium, potassium, magnesium, calcium, and other alkali and alkaline earth metals. Iron and other transition metals are often found in natural clays, substituted for the aluminum cations. Oxides of virtually every metal are found as impurity deposits in clay minerals. 

As mentioned earlier, the starting materials are of high purity. Because we work in a closed system and because we have an electrodeless discharge there should be no sources of additional impurities. Neutron activation analysis revealed that all the transition metal impurities that strongly affect the transmission properties of the optical fibers are lower than 1 ppm. From fiber transmission measurements we know that, besides traces of OH, some impurities must be lower than 1 ppb because only the intrinsic attenuation of the material is found. The chlorine content is rather large at 0.1%, even at the deposition temperature of 1000 °C. Fortunately the chlorine does not affect the optical properties in the interesting region of 0.6 pm - 1.5 pm. 

---