Extraction and Preparation of Vanadium

For industrial purposes vanadium is not required in the elemental state. More than 90 per cent, of the world's production of vanadium is used in the manufacture of special steels, for which purpose an iron-vanadium alloy, known as ferrovanadium, containing from 30 to 40 per cent, of vanadium, is marketed. The method of manufacture of this alloy from vanadium-bearing ores varies considerably with the composition of the ore and the value of the by-products. The process is conveniently divided into two stages:

1. The preparation of a complex mixture of vanadates or of crude vanadium pentoxide or of crude iron vanadate.
2. The conversion of these products into the iron-vanadium alloy.

Dry Process

The ore is roasted in a reverberatory furnace of about eighty tons capacity. The time occupied in passing the ore through the furnace is about two days, this time being necessary in order to burn off the asphaltic material which, together with any free sulphur also present, renders the addition of fuel unnecessary except towards the end of the heating. The roasted product contains from 40 to 50 per cent, of vanadium pentoxide and not more than 0.5 per cent, of sulphur, the rest being made up of silica, alumina, lime, magnesia, iron and nickel in varying proportions. This material is mixed with suitable fluxes and subjected to a matte smelting in a second reverberatory furnace. A matte is formed of all the foreign metals present in the ore, and a supernatant slag is produced, which contains all the vanadium combined with the gangue material (silica, lime, alumina, magnesia) as mixed vanadates.

Wet Processes

These vary considerably in detail according to the nature and amount of constituents other than vanadium in the ore. An outline of the operations involved in the case of patronite is as follows: The ore is roasted with common salt or sodium carbonate and then extracted either (a) with water to give an alkaline solution of sodium vanadate and soluble vanadates of other metals, any lead, zinc, copper, etc., being left in the residue; or (h) with sulphuric acid to produce a solution of vanadyl sulphate. Acid extraction is usually employed when the vanadium content of the material is low. The alkaline extract from (a) is treated with excess of sodium carbonate in order to precipitate calcium and aluminium, after removal of which, addition of ferrous sulphate throws down a precipitate of iron vanadate of uncertain composition. The acid solution of vanadyl sulphate from (b) is either evaporated to a cake and the residue calcined to vanadium pentoxide, or the solution is treated directly with oxidising agents, e.g. hypochlorites, whereupon precipitation of vanadium pentoxide takes place:

2VOSO₄ + Cl₂ + 3H₂O = V₂O₅ + 2H₂SO₄ + 2HCl.

For the treatment of carnotite several methods are available. The method recommended by the United States Bureau of Mines is as follows: The ore is leached with concentrated nitric acid at 100° C., neutralised with caustic soda, and barium chloride and sulphuric acid added to the solution to precipitate the radium as barium-radium sulphate. The precipitate settles in three or four days, after which time the clear liquid is decanted into tanks and is treated with excess of boiling sodium carbonate solution in order to precipitate any iron, aluminium and chromium present. The solution now contains sodium uranyl carbonate and sodium vanadate. It is nearly neutralised with nitric acid, and caustic soda is added in sufficient quantity to precipitate the uranium as sodium uranate. After filtering, the remaining solution is neutralised with nitric acid and ferrous sulphate added, whereupon iron vanadate is thrown down. By this method it is claimed that 90 per cent, of the radium, all the uranium, and 50 per cent, of the vanadium in the carnotite are recovered.

Electrolytic methods for the separation of vanadates of the metals have also been suggested, but do not appear to have come into general use.

The preparation of the iron-vanadium alloy from the crude vanadates obtained in any of the foregoing processes is carried out almost entirely at Bridgville, Pennsylvania. The process consists in reduction of the material with carbon in the electric furnace. Three graphite rods of 12-inch diameter are suspended in an intimate mixture of vanadium compound, iron ore or scale, fluxing agent (lime or fluorspar), and coke, contained in a cast-iron furnace lined successively with bricks and carbon blocks. The material is moved by a worm-conveyor into the high-temperature zone, and thence is immediately removed in order to prevent reoxidation of the vanadium. Two tap-holes are provided, one for alloy and one for slag, and continuous feed is employed. A good sample of the alloy produced in this manner gave the following analysis:

	%
Fe	48.65
V	49.20
Si	0.72
C	0.55
Al	0.31
Mn	0.32
S	0.15
Ni	0.10

The usual vanadium content of commercial ferrovanadium is, however, between 30 and 40 per cent.

Carbon has a great tendency to combine with vanadium to form carbides, the presence of which in the alloy renders it unsuitable for use in steel manufacture. The successful employment of carbon as the reducing agent is in fact quite recent. Formerly silicon, an iron- silicon alloy, or aluminium was used in place of carbon, but it was difficult to obtain a product which was free from silicon or aluminium, and considerable loss of vanadium took place in the slags.

Modifications of the Goldschmidt thermite process may also be employed for the preparation of the iron-vanadium alloy. The crushed vanadates or vanadium pentoxide are mixed with the necessary amount of iron scalings or turnings and fluxes, and introduced into a gas-fired open-hearth furnace or into an iron crucible provided with a refractory lining and previously heated to redness. The reactions taking place are:

(i) 3V₂O₅ + 10Al = 6V + 5Al₂O₃.
(ii) Fe₂O₃ + 2Al = 2Fe + Al₂O₃.

With a vertical-shaft furnace a much higher temperature, 2500° to 2800° C., and a much larger output can be obtained than with a crucible. A furnace 9 feet 3 inches high and 4 feet 6 inches wide will produce 125,000 lb. of alloy in one " run."

There is no demand for pure vanadium, and the isolation of the metal is therefore not an industrial process. Even on the small scale the operation is attended with considerable difficulty, owing to the very high temperature necessary for the reduction of vanadium compounds and the tendency for re- oxidation to take place. The following methods have given products of variable purity:

Modifications of the Goldschmidt Process

Vanadium pentoxide, V₂O₅, is mixed with twice its weight of an alloy of the rare earths obtained in the manufacture of thorium nitrate, and consisting roughly of 45 per cent, cerium, 20 per cent, lanthanum, 15 per cent. " didymium," and about 20 per cent, of other rare metals. The reaction is carried out in a magnesia-lined crucible and is started with a firing mixture of barium peroxide, potassium chlorate, and aluminium powder. Considerable evolution of heat takes place. It is claimed that vanadium of 99.7 per cent, purity can be obtained by this method. Samples of vanadium, which in some cases were 100 per cent, pure, have recently been obtained by reducing the pentoxide with a mixture of finely milled calcium and calcium chloride in a bomb heated electrically for an hour at 900° to 950° C. The presence of hydrogen or carbon should be avoided, and the operation is best conducted in vacuo.

Vanadium pentoxide is not easily reduced by means of aluminium, which also tends to alloy with the product. Even with the addition of carbon, calcium fluoride, or calcium carbide to the reaction mixture, complete reduction does not ensue. Meyer and Backa obtained vanadium of only 93.5 per cent, purity using vanadium pentoxide and aluminium as in the Goldschmidt process. Vogel and Tammann claim to have prepared vanadium of more than 99.07 per cent, purity by the same method, but did not ascertain the conditions necessary for success. Ruff and Martin prepared 95 per cent, pure vanadium by using vanadium trioxide, V₂O₃, and aluminium.

The use of calcium in place of the rare-earth alloy as the reducing agent gives a product containing from 91 to 93 per cent, of vanadium, while a mixture of calcium and aluminium produces 94.5 per cent, pure vanadium. Lithium has been used as reducing agent, vanadium of 99 per cent, purity being claimed.

Reduction of Chlorides

Roscoe reduced vanadium dichloride, VCl₂, at a bright red heat with hydrogen, every precaution being taken to prevent the entry of moisture and oxygen into the apparatus. The product was 95.8 per cent, pure metal, the impurity being mainly hydrogen. This method is of interest in that by its means metallic vanadium was first obtained; the process is, however, very slow. Reduction of the chlorides of vanadium by means of sodium gives a product of doubtful purity. Billy claims to have prepared pure vanadium by passing the vapour of vanadium tetrachloride, VCl₄, over sodium hydride, prepared in Situ, at 400° C.

Electrolytic Reduction at High Temperatures

The deposition of metallic vanadium by electrolysis of a solution of a vanadium salt at ordinary pressures has not hitherto proved successful. The reason is that vanadium compounds of low valency frequently decompose water with evolution of hydrogen and undergo oxidation with increase of valency, so that the formation of the free metal does not ensue. The electrolytic isolation of other strongly electropositive metals is attended with the same difficulty. Electrolysis of anhydrous fused vanadium salts or reduction of vanadium oxides in the electric furnace can, however, be successfully employed. Thus, the metal has been obtained by electrolysing vanadium trioxide or pentoxide dissolved in a bath of molten vanadium tetrafluoride and calcium fluoride. The anode is made of carbon and the cathode of lead. A lead-vanadium alloy is obtained from which the lead is subsequently volatilised. This process is similar to the Heroult method for the extraction of aluminium. Gin electrolysed molten calcium fluoride using a steel cathode and an anode composed of a mixture of carbon and vanadium trioxide, V₂O₃. Vanadium trifluoride is formed on the anode, passes into the molten electrolyte and is then decomposed, the vanadium being deposited on the cathode. Beckman electrolysed a vanadium oxide in a bath of molten lime.

Reduction of Oxides of Vanadium in the Electric Furnace

The reduction of vanadium pentoxide or trioxide by means of carbon yields a product which contains some of the carbon as carbide. Friederich and Sittig were unable to obtain a sample containing more than 82 per cent, of vanadium when they reduced a mixture of vanadous oxide, V₂O₃, and carbon in an atmosphere of hydrogen. Using sugar-charcoal and carrying out the reduction in an atmosphere of hydrogen in an electric furnace, Moissan was able to reduce the carbon content to 4.4 per cent. Ruff and Martin obtained 98.11 per cent, vanadium by heating to 1950° C. a mixture of vanadium carbide and vanadium trioxide pressed into a rod in a zirconia crucible. The reduction of vanadium trioxide has also been effected by passing an electric current through rods of the material in a good vacuum, and by the action of hydrogen at a pressure of 5 atmospheres and a temperature of 2500° C.

The isolation of vanadium can be effected on a very small scale, suitable as a lecture experiment, by passing an electric current through a platinum wire filament immersed in vanadium oxytrichloride, VOCl₃, either in vacuo or in an atmosphere of hydrogen. The metal is obtained as a smooth, silver-grey deposit.