Numberless other sources of soda have been proposed from time to time. Hunt and Gossage fluxed salt cake with small coal, lixiviated the sulphide of sodium cake obtained with water, and passed into the solution, heated with steam, a current of carbonic anhydride. Carbonate of soda is produced which may be crystallized out from the cooled liquors, and purified by dissolving and recrystallizing. The sulphide of sodium cake is, however, a material difficult to make, aud uncertain in character. A considerable amount of soda is lost by volatilization, and the wear and tear of the plant is very great. The burning of the sulphuretted hydrogen evolved, and utilization of the sulphurous acid thereby produced for the manufacture of sulphuric acid, has been only a failure. It has been proposed in several patents to obtain sulphide of sodium in the manner described, and to manufacture caustic soda from it by the oxides of iron, manganese, zinc, or copper. Here, again, the sulphide of sodium difficulties have been insuperable. The decompo sition of salt by silica and steam, by electricity, by superheated steam, and various other agencies, has formed a series of interesting but practically unsuccessful experiments upon a more or less large scale. The most likely process, excepting the ammonia method described above, consists in the extraction of soda from felspar, cryolite, and other minerals. Of the many patents bearing upon this ides, that of Julius Thomsen (1850), has been the most successfully worked. Cryolite is crushed to powder and mixed with slaked lime, or powdered chalk in large excess. These materials are then calcined at a gentle beat, for about two hours, fusion being especially avoided, and the resulting white powder is lixiviated by water. A solution of soda and sodium aluminate is obtained standing about 28° B. This is transferred to cylinders fitted with agitators, and a stream of pure carbonic anhydride passed in, which converts the soda compounds into carbonate, freeing alumina. The following equation represents, roughly, the reaction :— = The liquors are run off, the alumina is settled out, and the soda solution boiled down and calcined.
If caustic soda be required, about 15 equivalents of lime are added to 2 equivalents of cryolite, producing caustic soda, aluminate of lime, and fluoride of calcium. The decomposition is usually performed in a large upright cylinder of cast iron, heated by steam to boiling point. The mixture is thoroughly agitated and boiled for about three hours, and then settled, evaporated, fished, and calcined. Carbonate of soda may be obtained by this latter, wet, method also, by treating the liquors from the decomposers with carbonic anhydride, settling and calcining. About 20,000 tons of cryolite and allied minerals are annually consumed by this process.
It only remains to glance at some of the modifications and improvements of the soda manu facture which are engaging the attention of manufacturers. The discovery of enormous salt deposits in the Cleveland district has for many years past excited among the Tyne manufacturers the hope of procuring their supplies of salt nearer home than at present. Considerable sums of money have been spent by Messrs. Bell Brothers aud Bolckow, Vaughan, and Company, in turning this discovery to practical account ; but hitherto the nature of the deposits—the necessity for making an artificial brine, and the great depth at which the salt lies—has vetoed all attempts. Doubtless, before many years are past, unless the use of salt is superseded, more perfect mechanical contrivances than have yet been applied, will bring the Cleveland deposits into active competition with the brines of Cheshire and Worcestershire.
The salt question leads naturally to the consideration of a process that has been on its trial for many years with ouly a limited success—the ammonia soda process. So long ago as 1838, Harrison, Grey, Dyar, and John Hemming patented the use of ammonia in the production of soda, and since that time many specifications have been filed, having as their object the bringing of the discovery to the test of practice upon a largo scale. The patents of Delaunay, Young, Rolland, Goasage, SehlOssing, Deacon, and finally Solvay may be mentioned. All hut the last named have failed to
command even a moderate degree of success, owing to the delicate nature of the manufacture and tho costly character of the materials employed. The simplicity of the process has, however, always tempted fresh investigators into the field, and to Solvay belongs the honour of finally establish ing the manufacture upon something like a large scale. Solvay's first patents were taken out in 1861 and 1863. Previously to that time works had been established in Cheshire, at Widues, Leeds, and Newton, in this country, by Messrs. Deacon, Gossage, Bowker, Muspratt, &c. ; also near Nancy and at Putcaux upon the Continent. The Turck process was adopted at Nancy, and at Putoaux, Schliissing carried out his ideas assisted by Rolland, a scientific engineer, Solvay's works were first erected about the year 1866, at Couillet, in Belgium. About 1872, a second establish ment was set up at Varangdville-Dombasle, and both works have been recently enlarged, the total output reaching about 25,000 tons per annum— 20,000 tons at Varangeville-Dombasle, and 5000 tons at Couillet. In England the only works in active operation are those of Briinner, Mond, and Co., at Northwich and Saudbacb. At these two works, the turn-out of carbonate is about 10,000 tons per annum.
The process is founded upon the well-known laboratory reaction, that when carbou dioxide is passed into a solution of common salt and ammonia, bicarbonate of soda and chloride of ammonia aro formed : NH, + + NaC1 + 11,0 = NH,CI + Or, in shorter form, bicarbonate of ammonia and chloride of sodium in solution produce bicarbonate of soda and chloride of ammonium. The former is decomposed by heat to yield neutral carbonate, the latter is, or may be, distilled with limo or magnesia, the ammonia recovered, and used over again. As a matter of fact, however, this part of the process has practically broken down, and, in the English works at least, has been abandoned. At the best a serious loss of ammonia is incurred, which tells fearfully against the costs. A thoroughly good article is undoubtedly produced, es the following analysis shows : Carbonate of soda .. Iron and Alumina .. •• Moisture 0.15 Lime Siliea .• 0.10 Magnesia 0.06 Chloride of sodium .. 0.13In this respect, the ammonia soda certainly has the advantage over the ordinary carbonate, which eontains up to 7 or 8 per cent. of sodium sulphate, of chloride of sodium, and 1 per cent. of caustic soda. It is, moreover, claimed for the process that tho noxious exhalations of the Le Blanc manufacture are avoided, that there is little or no waste product to encumber and render offensive the land in the neighbourhood of the works, that the purity of the article and its freedom from caustic soda render unnecessary the manufacture of refined alkali and soda crystals, and that the absence of iron and sodium sulphate peculiarly fit it for fine glass-making. There is a great deal of truth in all this, hut a good deal may be said on the other side. For many purposes, e.g. in the soap manufaeture—the presence of sulphate of soda is a positive advantage and must be actually added if the ammonia soda is used. Where an artiele free from the ordinary impurities of soda ash is required, it can be produced in the form of refined alkali or soda crystals at a cast that is, still, lower than that of ammonia soda. Moreover, the attendant evils of the Le Blanc process, are capable of regeneration. Much has lately been done in this way, and much more will yet be accom plished. The present plant from its very roughness of structure is easily put right when any little mishap occurs, whereas the constant breaking down of the more delicate parts of the ammonia soda process seriously enhances the costs of the product. Upon this question of costs it may be stated that a ton of the purest alkali of the Le Blanc process—refined alkali—costs 61. 10s., against 81. 5s. of the ammonia process, so that even allowing for the difference in streligth-53 per. cent. against 58 per cent.—there is still a considerable margin in favour of the former. The further consideration that the ammonia soda is a purer, as well as a stronger article, only lessens, does not remove, this disparity.