Formation of Alkali Cyanides

The synthesis of potassium cyanide by heating a mixture of potassium carbonate and carbon to a high temperature in an atmosphere of nitrogen was first discovered by Scheele in 1784. Bunsen and Playfair in 1848 investigated the reaction between atmospheric nitrogen, potassium and carbon compounds, following the discovery of Dawes in 1835 that potassium cyanide was a product of the blast furnace. Incidentally it may be mentioned that it has recently been suggested that about 1 per cent, of the nitrogen of the air blast could be recovered as cyanide without interfering with the production of iron in the blast furnace.

Many attempts have been made to manufacture potassium cyanide from the carbonate, carbon, and nitrogen, but the greatest drawback is the deterioration of the furnace by the materials used at the high temperature required. It has been found possible, however, to prepare technically pure potassium cyanide (and sodium cyanide) indirectly by fusing crude calcium cyanamide, CaNCN, with the corresponding chloride or carbonate:

CaNCN + 2KCl + C = CaCl₂ + 2KCN;
CaNCN + Na₂CO₃ + C = CaCO₃ + 2NaCN.

Interest in the original cyanide synthesis was revived by the exhaustive investigation of J. E. Bucher into the production of cyanides as a method of nitrogen fixation, and various patents were taken out during the years 1914 to 1917. Essentially the process consists in the reaction between sodium carbonate, carbon, and nitrogen at a high temperature in the presence of finely divided iron which acts catalytically:

Na₂CO₃ + 4C + N₂ = 2NaCN + 3CO -138,500 calories.

As regards raw materials, sodium carbonate in the commercial form of soda ash is used; coke, either metallurgical, petroleum, or pitch, may be employed with technical success, the latter form being practically ashless, which obviates the periodic removal of material; with regard to the nitrogen, it seems that at present there is insufficient data to determine whether pure nitrogen or producer gas has any special advantages.

The finely divided mixture of soda ash, coke, and iron is made up into briquettes, which are fed into externally heated vertical retorts heated to 900°-950° C., and nitrogen passed into the retorts under slight pressure. The reaction is of a complex nature, but probably the iron assists in the reduction of sodium carbonate to metallic sodium. After four hours heating the final product yielded cyanide to the extent of a 50 per cent, conversion of sodium carbonate. Bucher proposed lixiviation of the eyanised material with water to extract the cyanide as such, but the chief disadvantage is the solution of unchanged sodium carbonate. If sodium cyanide is required, the process worked for some time by the Nitrogen Products Company would seem to be advantageous, whereby liquid ammonia was used to leach out the cyanide, upon which it exerted a selective solvent action.

The original idea of Bucher was to fix nitrogen as ammonia or ammonia derivatives, chiefly by hydrolysis of the cyanide. Thus, at relatively low temperatures the production of ammonia and sodium formate occurs,

NaCN + 2H₂O = HCOONa + NH₃ +64,000 calories.,

and any cyanate which has been formed by oxidation is converted into ammonia, carbon dioxide, and sodium carbonate:

2NaNCO + 3H₂O = Na₂CO₃ + 2NH₃ + CO₂ +15,340 calories.

At higher temperatures (400° to 500° C.) the hydrolysis of the cyanide occurs, with the formation of sodium formate and ammonia as before, but the formate is broken down at that temperature into the carbonate, hydrogen, and carbon monoxide:

2HCOONa = Na₂CO₃ + CO + H₂.

The sodium carbonate produced could be returned to the cyanising retorts.

Bucher suggested that further developments of this process might consist of heating the cyanide in air, whereby oxygen would be removed from the latter to form cyanate, and the nitrogen could be used for the cyanising furnace. If, then, one-half of the cyanate were heated with water, production of ammonia and sodium bicarbonate would occur:

NaNCO + 2H₂O = NaHCO₃ + NH₃.

The other half of the cyanate would react with this ammonia and carbon dioxide (produced by burning the carbon monoxide from the cyanising furnace) in the presence of water to form ammonium cyanate:

NaNCO + NH₃ + H₂OH + CO₂ = NaHCO₃ + NH₄NCO.

Intramolecular change converts ammonium cyanate into urea: NH₄NCO = CO(NH₂)₂.

The possibility of a cheap large scale production of urea is attractive, as modern research indicates that this substance is ideal for fertilising purposes.

An interesting modification of the Bucher process is the subject of patents by C. B. Jacobs of the E. I. du Pont de Nemours Powder Company. Instead of briquetting, the addition of 5 per cent, sodium fluoride or chloride is made, which acts as a flux. Further, a reactive form of carbon is used which is obtained by incinerating black liquor at 200° to 300° C. This latter substance is produced from the digestion of wood and other cellulosic materials with caustic soda.

The composition of a typical charge is:

Sodium carbonate	40 per cent.
Carbon	40 per cent.
Ferric oxide	15 per cent.
Sodium fluoride	5 per cent.

It is claimed that at 925° to 950° C. a 95 to 98 per cent, conversion of sodium carbonate is obtained in two hours.

After all volatile matter has been expelled by heating, nitrogen in the form of producer gas is passed in under a pressure of 20 pounds above atmospheric, and this is continued until no more carbon monoxide is evolved. This gas is utilised by mixing with producer gas and burning under the retorts.

Extraction of the mass with water yields a 96 to 98 per cent, pure cyanide after crystallisation. The above description of the modified Bucher process is given, because it would appear that the original process was not economical enough to fulfil expectations. According to Guernsey and Sherman, the mechanism of the reaction is the reduction of sodium carbonate to metallic sodium, which then forms sodium carbide, and this latter compound reacts with nitrogen to form the cyanide, assisted by the iron as catalyst.

Reference should be made here to the fact that at present the only large scale production of sodium cyanide is the Castner-Roessler process, the sole rights of which are owned by the Deutsche Gold und Silber Scheidenanstalt. In this process metallic sodium and ammonia are caused to react at 300° to 400° C., with the formation of sodamide:

1. 2Na + 2NH₃ = 2NaNH₂ + H₂.
   
   The molten sodamide is then brought into contact with red-hot charcoal, the final product being sodium cyanide, while sodium cyanamide is formed intermediately:
2. 2NaNH₂ + C = Na₂NCN + H₂;
3. Na₂NCN + C = 2NaCN.

It has already been mentioned that the sodium cyanide may be used as a potential source of ammonia by hydrolysis with steam. It must not be overlooked, however, that sodium cyanide is being used largely in the metallurgical industry - especially in gold extraction, and also that it has a limited use as an insecticide.

In the description of the cyanamide process (infra) reference is made to the simultaneous formation of calcium and barium cyanides and cyanamides by heating the carbonates and carbon in nitrogen. Many attempts have been made to manufacture barium cyanide as a source of ammonia, but at the present time it would seem that this process has not developed into a commercial success. There are two lines along which investigations have been made: the first a single-stage process which resembles the Bucher process for sodium cyanide, and the second a two-stage process closely allied to the manufacture of cyanamide. The first process consists in briquetting a mixture of barium carbonate and carbon (pitch or tar) and passing nitrogen over the mixture heated to 1000°-1400° C.:

BaCO₃ + 3C + N₂ = Ba(CN)₂ + 2CO.

In the second process, the first stage is the production of barium carbide by heating barium oxide and carbon in the electric furnace:

BaO + 3C = BaC₂ + 2CO.

The second stage is the formation of barium cyanide by treating the carbide with nitrogen at about 1000° C. It is probable that barium cyanamide is formed as an intermediate product:

BaC₂ + N₂ = BaNCN + C;
BaNCN + C = Ba(CN)₂.

The barium cyanide from either process may be treated with steam, when hydrolysis occurs, with the formation of ammonia and barium hydroxide:

Ba(CN)₂ + 4H₂O = Ba(OH)₂ + 2NH₃ + 2CO.

Theoretically, the barium hydroxide can be utilised again for the production of cyanide, but considerable practical difficulties are encountered. The high initial cost of the barium carbonate (and hydroxide) compared with that of calcium carbonate is a great disadvantage in making a comparison of the cyanide and cyanamide processes. One of the great drawbacks of the barium cyanide process is the corrosive action of the molten cyanide upon the retorts owing to the high temperature employed. Recently, Askenasy has suggested that this may be obviated to a large extent by using a carbon of low ash content, such as petroleum coke or soot. This should be preheated to a temperature above that employed for its production, and should also be treated with oxygen or chlorine.

Another method for obtaining an intimate mixture of barium oxide and carbon is by decomposing methane with hot barium oxide, the particles of which become coated with carbon particles. The hydrogen from the methane is utilised for providing heat for the furnace, and also for the reaction of the mixture of barium oxide and carbon with atmospheric nitrogen as cyanide.

It would seem desirable to indicate processes which have been proposed from time to time to fix atmospheric nitrogen in the form of hydrogen cyanide. So far, however, there is no large scale process of any importance which makes practical use of the synthesis of hydrocyanic acid.

Berthelot discovered that acetylene and nitrogen combined together with the aid of electric sparks to form hydrogen cyanide, and that the reaction was reversible:

C₂H₂ + N₂ ⇔ 2HCN.

Later, Dewar showed that a mixture of hydrogen and nitrogen when passed through a carbon tube heated externally resulted in the production of hydrocyanic acid.

Hoyermann repeated Berthelot's synthesis by passing a mixture of 2 volumes of nitrogen and 1 volume of acetylene through a carbon arc, and obtained a 60 to 70 per cent, conversion of acetylene into hydrocyanic acid.

Gruskiewicz found that hydrogen cyanide was produced when sparks were passed through a mixture of carbon monoxide, nitrogen, and hydrogen. Lipinsky considered that the carbon monoxide was first reduced to methane, on account of his experiments with a mixture of methane (from natural gas), nitrogen, and hydrogen:

2CH₄ + N₂ = 2HCN + 3H₂.

Lipinsky's method was tried on the technical scale at Neuhausen in Switzerland in 1914, when a quantitative conversion of methane into hydrocyanic acid was claimed. The mixture consisted of 20 per cent, methane, 70 per cent, nitrogen, and 10 per cent, hydrogen, which was passed through an arc between platinum electrodes from a 2200-volt alternating current supply. The hydrogen and excess of nitrogen were employed merely as a diluent, and it is stated that 30 grams of hydrocyanic acid per K.W. hour were obtained.

A somewhat different method of causing gaseous carbon and nitrogen compounds to combine was used by Beindl, who employed a catalyst material consisting of a metal, its oxide and nitride.

The Chemischerfabrik Griesheim Elektron propose the passing of a nitrogen-hydrogen mixture over coke in an electric furnace; while a more recent process by the Norsk Hydroelektrisk Kvaelstofaktieselskab makes use of a blown-arc furnace of the Birkeland-Eyde type, through which a mixture of nitrogen, hydrogen, and a hydrocarbon is passed.

The subsequent treatment of the hydrocyanic acid is either for the production of cyanides by absorption with alkalies, oxidation by burning in air to nitric oxide for conversion into nitric acid, or hydrolysis with steam at a high temperature with the formation of ammonia.