THE DISCOVERY OF T H E ALKALI METALS BY HUMPHRY DAVY: THE BEARING OF T H E DISCOVERY UPON INDUSTRY. BY F. MOLLWO PERKIN, PH.D. ( A Paper read before the Faratlay Society, Tuesday, Dccenrber 17, 1907, Dr. H. BORNS in the Chair.) Humphry Davy was born on December 17, 1778, at Penzance. The Davys originally came from Norfolk, but settled down some generations earlier in Cornwall. Edmund Davy, the grandfather of Humphry, was a builder in the west of Cornwall, and Robert, his father, had a small property at Varfell, but he was thriftless, and when he died his affairs were found to be in a very embarrassed state ; so that upon Humphry, the eldest son, who at the time of his father’s death was sixteen years of age, fell the responsibility of looking after the family. Before this time Davy appears not to have been very studious, but of rather a restless disposition and by no means fond of restraint, but upon the death of his father he at once settled down to hard work and study, and a few weeks afterwards was apprenticed to an apothecary and surgeon who practised in Penzance.Not only did he whole-heartedly throw himself into the study of his profession, but he also mapped out for himself an arduous course of general study. In 1797 he commenced the study of natural philosophyand turned his attention to chemistry, as it helped him in the work of his profession. From the reading of chemistry to the practice with such crude apparatus as tea- cups and tobacco pipes was but a step, but beyond the fact of spoiling his sister’s dress with acids and frightening his friends by the noises of explosions which proceeded from his experinienting room, little is known of the outcome of his work at this time, which, in any case, was probably not much more than seeking in the dark with but little light to guide him.It was meeting with Gregory Watt, the son of the engineer James Watt, who had been educated at the University of Glasgow, that finally decided Davy to devote himself to science. Mr. Gilbert, who afterwards occupied the Presidential Chair of the Royal Society in succession to Davy, helped him also by allowing him the use of his library and physical apparatus In 1798 Dr. Beddoes, who had been trained as a medical man, and had many strange and unpractical ideas, founded a kind of sanatorium in Bristol, which he called the Pneumatic Institution, and asked Davy if he would take charge of the laboratories.The main object of the Institution was to study gases and their effects upon different ailments. Davy, with his energetic disposition, threw himself heartily into the work, and in 1799 we find him studying the action of nitrous oxide upon the human system. He first of all experimented upon himself and then tried its action upon others. About this time Maria Edgeworth, who was visiting her sister, Mrs. Beddoes, wrote :- ‘‘A young man. a Mr. Davy, at Dr. Beddoes’, who has applied himself 205206 DISCOVERY OF ALKALI METALS BY H. DAVY: much to chemistry, has made some discoveries of importance, and enthusi- astically expects wonders will be performed by the use of certain gases, which inebriate in the most delightful manner, having the oblivious effects of Lethe and at the same time giving rapturous sensations of the Nectar of the Gods.Pleasure even to madness is the consequence of this draught.” She adds, however, that some found nothing but a sick stomach and ;I giddy head. Although there are many points of great interest in Davy’s early days and in the gradual development of his power as he went on from step to step until he matured into one of the most brilliant experimenters, if not the most brilliant, which the nineteenth. century was to see, it is not my intention to describe the life of Davy, but to dwell upon his researches in connection with electrochemistry and to indicate the enormous progress in industry which SIR HUMPHRY DAVY.has been gradually evolved from his small, but record-breaking, com- mencements. In October of 1800 he wrote a letter to Mr. Gilbert, addressed from the Pneumatic Institution, in which he refers to experiments upon galvanism. This letter is of particular importance, because it fixes the date at which he commenced to work at electrical phenomena, and which, as it afterwards turned out, was to be the monument of his most enduring fame. He says in this letter :- “Galvanism I have found, by numerous experiments, to be a process purely chemical, and to depend wholly on the oxidation of metallic surfaces, having different degrees of electric conducting power. “Zinc is incapable of decomposing pure water ; and if the zinc plates be kept moist with pure water, the galvanic pile does not act ; but zinc is capable of oxidating itself when placed in contact with water, holding in solutionREARING OF THE DISCOVERY UPON INDUSTRY 207 either oxygen, atmospheric air, or nitrous or niuriatic acid ; and under such circumstances the galvanic phenomena are produced, and their intensity is in proportion to the rapidity with which the zinc is oxidated.“The galvanic pile only acts for a few minutes, when introduced into hydrogen, nitrogen, . . . that is, only as long as the water between its plates holds some oxygen in solution : immerse it for a few seconds in water con- taining air and it acts again.” This shows that Davy had already a very acute insight into the working of the primary battery, and in this connection it must be born in mind that it was only in May, 1800, that the discovery of the Voltaic pile was made known by Volta in a letter to Sir Joseph Ranks, the then President of the Royal Society. On February 16, 1801, Davy was clected to the post of Assistant Lecturer in Chemistry at the Royal Institution, Director of the Laboratory and Assistant Editor of Journals of the Institution.It is of particular interest to notice that the first lecture lie gavc was upon “Galvanic Phenomena.” Davy at this time was rather ungainly in appearance and not particularly taking in manner. So much so, in fact, that Count Runiford, the head of the Royal Institution, was by no ineatis prepossessed when he first met him. But after hearing him lecture he exclaimed, Let him command any arrangements which the Institution can afford.” The chemist appointed to the Royal Institution in those days had not 311 altogether enviable position, because the Board of Managers were wont to meet and draw up suggestions as to the class of work which should he lectured upon and upon which research work should be carried out.::: Davy was, for example, asked to lecture upon the “ Principlcs and Art of Tanning,” then upon “ Agriculture,” and the work entailed upon these subjects must have prevented him from giving so much time as he probably otherwise would have done to the subject of electricity, although i n 1806, i n the Bakerian Lecture before the Royal Society, the subject of his paper was upon the decompositions and chemical changes produced in substances of known composition by electricity.He had also shown that it was due to impurities in the water used by Nicholson and Carlisle when they succeeded in deconiposiiig it electrolytically, that caused the appearance of acidity at the positive electrode and alkalinity at the negative electrode. Beside this he had advanced his theory of electro-affinity, so that it is obvious he had had the subject of electricity and its applications to chemistry continually before him. In October, 1807, he commenced the study of the action of Voltaic electricity upon the alkalis, a subject which he had evidently considered before, because we find in one of his note-books under thc date August 6, 1800 :- (‘ Would not potash dissolved in spirits of wine become a conductor ? ” To-day we know that potash dissolved in absolute alcohol is an electrolyte, although an extremely poor one.On November 19, 1807, Davy delivered the Bakerian Lecture before the Royal Society and astonished the scientific world by describing the metals potassium and sodium.+ The lecture was entitled, “ On some New Phenomena of Chemical Changes produced by Electricity, particularly the Decomposi- tion of the Fixed Alkalies, and the Exhibition of the New Substances which * The Royal Institution was originally founded on philanthropic lines to benefit t The first intimation of his discovery is found in his laboratory note-book, the poor. preserved at the Royal Institution under the date October 19, 1807.208 DISCOVERY OF ALKALI METALS BY H. DAVY: constitute their Bases ; and on the General Nature of Alkaline Bodies ( P l d .Traits., 1808, vol. 98, pp. 1-44). Davy commenced by referring to the lecture delivered by hini the previous year where he described a number of deconipositions and chemical changes produced in substances of known composition by electricity, when he also stated with prophetic foresight that “the new methods of investiga- tion promised to lead to a inore intimate knowledge than had hitherto been obtained concerning the true elements of bodies.” He then describes the “ methods used for the decomposition of the fixed alkalies.” The first attempts were made upon saturated aqueous solutions of the metallic hydroxides, using the highest electrical power he could command. This I‘ was produced by a combination of Voltaic batteries belonging to the Royal Institution, containing 24 plates of copper and zinc of 12 in.square, 100 plates of 6 in., and 150 of 4 in. square, charged with a solution of alum and nitrous acid (nitric acid) ; but in these cases, although there was a high intensity of action, the water of the solution was affected and hydrogene and oxygene disengaged with the production of much heat and violent effervescence. He therefore had a cell of 274 pairs, which was probably coupled up in series, so that the voltage obtained would be considerable. I find that if a zinc and copper plate are placed, the one in a saturated solutioii of alum and the other in 30 per cent. nitric acid, that an E.M.F. of 1.2 volts is obtained. If Davy had used such a battery composed of 274 cells he would hare obtained an E.M.F.of 328 volts. Davy, however, probably used very much weaker nitric acid and also employed a mixture of alum and nitric acid, the zinc and copper plates not being separated by a porous cell, as porous cells do not appear to have been used in these early days. In fact, Faraday in his first electromagnetic experiments used a battery consisting of concentric plates of zinc and copper separated from contact with each other by means of cloth. The first reference to porous earthenware cells which I can find is in 1836, when they were employed by Becquerel (Pogg. Aizia. Plzys. Chem. Now by placing zinc and copper plates in a saturated solution of alum and 5 per cent. of nitric acid an E.M.F. of 0.8 volt is shown. Therefore with such a battery the potential at Davy’s disposal odd have been 219.2 volts.The current shown by two such cells when shorted through a low-resistance ammeter is only 0.8 to 0.9 ampere. With 10 per cent. nitric acid, which acts violently on the plates, a current of 3-2 amperes was obtained, the E.M.F. being o.8j volt per cell. The batteries used by Davy, which are still preserved at the Royal Institution, were in the %oC form of a trough of baked wood, as shown i n Fig. I. The plates of zinc and copper were connected face to face and then slipped into slots and FIG. I . - fastened leak - tight with pitch or some similar sub- stance, the electrolyte being placed in each compartment. Thus i n the electrolyte zinc faced copper and vice versd. Davy recognised that the presence of water interfered with his experi- ments, and he therefore resolved to try the action of the electric current upon potash in “igneous fusion.” The potash was placed ir.a platinum 1836, 37. 429).BEARING OF THE DISCOVERY UPON INDUSTRY 209 spoon and was melted by heating it with a spirit lamp, a stream of oxygen gas being applied to the flame. The spoon was in connection with the positive pole of a battery of roo 6-in. plates “highly charged.” The connection with the negative side was made by a platinum wire. The potash appeared a conductor in a high degree, and some brilliant phenomena were produced. So long as connection was preserved a most intense light was exhibited at the negative wire, aiid a column of flame which seemed to be owing to the development of combustible matter arose from the point of contact.“When the order was changed, so that the platina spoon was made negative, a vivid and constant light appeared at the opposite point : there was no effect of inflammation round it ; but xriform globules, which inflamed in the atmosphere, rose through the potash.’’ But as the potash was dry this inflammable matter apparently arose from its decomposition. Davy also noticed that the platinum was considerably acted upon. Although a number of experiments were carried out under these conditions it was not found possible to collect any of the products of decom- position. He noticed that potash dried by ignition w3s a non-conductor, but was rendered conducting by the presence of a very small trace of moisture, and that it then readily fused and was decomposed by strong electricai powers.Davy now altered the experiment as follows : “ A small piece of pure potash, which had been exposed for a few seconds to the atmosphere, so as to give conducting power to the surface, was placed upon an insulated disc of platina, connected with the negative side of the battery, power 250 of 6 and 4 in a state of intense activity; and a platina wire, communicating with the positive side, was brought in contact with the upper surface of the a1 kali.” This experiment is easily shown if a piece of dry potassium hydroxide is placed upon a nickel plate connected with one pole of a lamp resistance and a piece of nickel wire connected with the other pole is placed in contact with the side of the hydroxide furthest away from the plate.No action then ensues. But if one now breathes upon the potassium hydroxide so as t o moisten it externally, action almost immediately commences and the hydroxide fuses. Electrolysis then takes place, and the metal as it is produced rises to the surface of the fused alkali often inflaming explosively. I n carrying out this experiment it is therefore wiser to wear glasses. Davy noticed under these circumstances that the “ potash began to fuse at both its points of electrisation. There was a violent effervescence on the upper surface ; at the lower or negative surface there was no liberation of elastic fluid, but sinall globules having a high metallic lustre, and being precisely similar in visible characters to quicksilver appeared, some of which burnt with explosion and bright flame as soon as they were formed, and others remained and were merely tarnished, and finally covered by a white film which formed on their surfaces.” From a number of experiments Davy managed to collect enough of the, what he termed, “ inflammable principle ” of potash, in order to experi- ment with it.He then tried caustic soda and found that similar phenomena were produced. But in this case the action was less intense, and it was necessary to work with thinner pieces of the substance. The potassillin produced by Davy remained fluid at ordinary tempera- tures, which shows that it must have been contaminated with sodium, because an alloy of sodium and potassium is liquid at ordinary temperatures.So great was Davy’s joy when he saw the globules of potassium ascending through the fused potash that we are told he danced round the210 DISCOVERY OF ALKALI METALS BY H. DAVY: room, and it was some minutes before he was calm enough to continue his experiments. With Davy’s theoretical considerations we will not here concern ourselves. He studied the properties of the new substances, and his experi- mental work with the minute quantities of the substance at his disposal is little short of the marvellous. He noticed that in all “ sensible properties ” these substances resembled the metals. The action upon oxygen, hydro- chloric acid gas, water, and mercury was noticed. It was fused with sulphur, heated with various oils, also with metallic oxides, and its power in reducing these was noticed.Less than a year later-June 30, 1808-Davy presented another paper before the Royal Society upon the ‘( Electro-chemical Researches on the Decomposition of the Earths ; with Observations on the Metals obtained from the Alkaline Earths and on the Amalgam produced from Ammonia” (PhiE. Tram., 1808, 98. p. 333). Although he never obtained the metals calcium, strontium, and barium in the pure state, he did obtain them in the form of their amalgams. He likewise obtained the metal magnesium as an amalgam. His method of procedure was to “slightly moisten the earths (that is, the oxides of the alkaline metals), mixed with one-third of the red oxide of mercury, the mixture was placed on a platina spoon, a cavity was made in the upper part of it to receive a globule of mercury .. . the whole was covered by a film of naphtha, and the plate was made positive, the mercury negative.” He distilled the amalgams which he obtainccl in this manner in plate-glass tubes, the tubes being bent in the middle and the extremities enlarged and rendered globular so as to act as a retort. The amalgam was introduced into the tube ; it was tlieii filled with naphtha, which was expelled by boiling through a small orifice in the receiver end. When all the naphtha except vapour had been driven out the tube was sealed up. Davy found no difficulty in separating the bulk of the mercury, but he was never sure, even after prolonged heating, whether he had actually driven it all off. He, however, describes the metals as being of a silver colour.In further experiments he showed that ammonium behaved like the metals and prepared the amalgam. History, it is usually considered, does justice to the work of men. It places upon an eminence those who have in their own day and generation been scoffed at, whilst others who stood high in popular esteem are thrown down. But ;in reference to scientific pioneers history may not always be able to judge, because those who to-day have to examine into the claims of those who lived in bygone generations cannot enter properly into the diffi- culties which beset their work. There was practically no theory ; what there was had often been erected upoii an erroneous basis, and therefore hindered rat her than helped investigators. Apparatus and material were defective. The wonder, therefore, is not that they did not do more, but that, seeking as they often did in an Egyptian darkness, they did so much.There was one thing, however, which was to their advantage : they knew nothing of the rush and headlong hurry of the present day, and that must have counted for much. It is often said, what marvels a Faraday or a Davy would have produced if they had had the apparatus at their disposal which we have to-clay. They might not have done ; they were pioneers and had pioneering instinct ; they knew how to sweep aside the entangling under- growth which blocked their path. But if all had been cleared away, it is His best results were obtained with barium.BEARING OF THE DISCOVERY UPON INDUSTRY 211 possible they would not have made their mark as they did.Each generation has those who meet its needs. D a y , thc acute thinker, the rapid and impatient worker, was the man for his time. He had also, what is not for every one who strives after scientific truth, recognition in his day. He loved popular applause and he loved fame-these he had in full measure. He was acknowledged in his lifetime, we acknowledge him to-day. And now let us look at the enormous harvest which we to-day reap from the seed sown by Davy, seed which took a long time to germinate, but when once the roots of the plant had taken hold of the soil, the face of an industry, nay of many industries, was changed. Cheiiiical Metkods. The first alkali metal to be produced upon nn industrial scale was potassium in 1827, by Wohler, who wanted to use it for the production of aluminium.Wohler hcated an intimate mixture of potassium carbonate with carbon- KZC03 + 2C: = 2 K + 3CO. About 18j4 Saiiite Claire Deville manufactured sodium in a similar manner. The process was first carried out at Salindres, in France, the method being to heat anhydrous sodium carbonate and chalk with charcoal. The manufacture was carried on in this manner for over thirty years. At the end of this time some 5,000 to 6,000 kilos. of metallic sodium were aniiually manufactured at Salindres. In 1886 the process of Castner was introduced in England, the method being to reduce sodium hydroxide with an iron carbide, prepared by heating a mixture of iron oxide and tar to red heat. This mixture was then heated to 1,000~ with the sodium hydroxide. The iron, however, according to Roscoe, took no part in the reaction, which may therefore be formulated as- 3NaOH + C = Na,C03 + Na + 3H.Netto, about the same time, introduced n somewhat similar method. I n this process melted sodium hydroxide was allowed to flow down over red hot coke. This caused decomposition to take place at the moment of contact between the molten hydroxide and the coke. The appa- ratus (Fig. 2) consisted of a cast-iron retort b filled with coke and wood charcoal, and heated to redness. The molten sodium hydroxide was run in from the container e through the funnel d, the sodium vapour being condensed in the receiver g. In order to produce 100 kg. of sodium, 1,000 kilos. of sodium hydroxide and 150 kilos. of carbon was required as reducer. The chief object in manufacturing sodium at this time was to employ it in the manufacture of aluminium, because the cheapening in the cost of sodium meant a corresponding drop in the price of aluminium.FIG 2. Electrolytic Methods. In 1890 Castner introduced his method for maiiufacturing sodium by the electrolysis of sodium hydroxide.212 DISCOVERY OF ALKALI METALS BY H. DAVY: Other attempts had been made to electrolyse salts of the alkali metals- Thus, in 1848, Linnemann (Yourr~fiir Prnktisclze Chenz., 1848, 415), suggested electrolysing fused potassium cyanide. He found if the temperature was held sufficiently low so that the surface had a solid skin upon it, that the metal collected below the surface, and was not therefore oxidised by the atmos- phere.In 1851 Charles Watt in this country patented an apparatus (Eng. Pat. 13,755) by means of which he hoped to produce sodium on a manufacturing scale by the electrolysis of fused sodium chloride. The apparatus consisted of a vessel A , which was madc of earthenware, or of iron lined with earthenware, to protect it from the action of the heat. F is the distilling head for the metal. C C are the electrodes, which, it will be noticed, are partially separated by means of the division E . The electrode beneath the still head consisted of carbon, the other one of gold. One recognises at once that a gold electrode would be useless if used for the electrolysis of an alkali halogen salt. The temperature of the apparatus was to be kept sufficiently high to cause the metal to volatilise as it was produced ; this, of course, would also be an objection, The motive power was to consist of 10.FIG. 3. Daniel1 cells. One point of particular note about this cell is the partiaP partition. Within ihe last few years cells with partial partitions have been patented for quite a number of diverse operations, in order to get over the difficulties always at tendant with the use of diaphragms. I t will be remembered that Davy was not able to isolate potassium so long as he employed external heat in order to fuse the alkali hydroxide, but that when he used the current first to fuse and then to electrolyse the fused salt he obtained the metal. The reason, as is now well known, was due to the fact that at high temperatures the metal dissolves again in the fused electrolyte, whereas it can be separated if the temperature is only allowed to rise a little higher than the fusing point of the hydroxide.In the Castner apparatus, although in the first place gas or other firing is used to melt the sodium hydroxide, very little heat is required after the passage of the current, because the heat generated is sufficient to keep the material fused, and one of the particular points of the patent is the temperature at which the electrolysis is to be conducted. The apparatus consists of an iron holder A for the fused hydroxide, which is narrowed to an opening at the bottom and connected with the opening is aBEARING OF THE DISCOVERY UPON INDUSTRY 213 cast-iron continuation B. Through this the iron cathode C passes.The anodes D D are of iron or ferro-nickel, which withstands the action of the evolved gases ; this anode is concentric with the cathode. E is a collecting vessel for the metallic sodium, which is closed at the bottom by means of an iron gauze net, shown in the dotted line. This net harigs down at the sides and forms a cylinder which encloses the cathode, thus preventing diffusion, and the gases from the anode coining into con- tact with the sodium produced at the cathode. The sodium as it is produced being molten and speci- fically lighter than the electrolyte, rises to the surface and passes through the gauze into the con- tainer E. From this it is removed from time to time by means of perforated ladles, which allow the molten alkali to drain out, but not the sodium.The gases pro- duced at the anode pass out at the opening P; this opening is also FIG. 4. ' useful for inserting-a thermometer in order to test the temperature of the electrolyte. The electrolyte is kept from leaking out at the cathode by the elongation B, which is in the first place filled with molten sodium hydroxide ; it, however, solidifies owing to this portion of the apparatus being kept cool. In order that the process may be satisfactorily carried on, it is necessary to regulate the distance apart of the electrodes and also their size according to the current to be employed. If the surfaces of the electrodes are too great then a portion of the sodium is absorbed by the fused alkali-this, of course, means loss of electrical energy.On the other hand, if the electrodes are too small, then the temperature of the bath becomes too high and again loss takes place. Pure sodium hydroxide contains only the ions Na. and OH'; hydroxyl is therefore given up at the anode. This causes the formation of water and the evolution of oxygen zOH= H,O + 0. At the high temperature of the electrolyte only a portion of the water will dissolve, but this will also take part in the electrolysis and yield H and 0, which, of course, is explosive. If no secondary reactions take place then only oxygen should be given up at the anode and sodium at the cathode ; but owing to the formation of water a certain quantity of hydrogen is invariably given up at the cathode, As a matter of fact this is not altogether a disadvantage, because it helps to cause the sodium to rise up into the catch box through the gauze.As, how- ever, the hydrogen also mixes with the oxygen and therefore tends to form an explosive mixture, explosions do frequently occur, but as a rule they are harmless. Probably for this reason it has not been found advisable to build electrolysers which will take more than 1,000 to 1,200 amperes. Owing to the fact that for every two equivalents of sodium hydroxide decomposed, one equivalcnt of water is produced, it is not possible to obtain more than a 50 per cent. yield of sodium-as a matter of fact 40 per cent. is as high as is usually obtained. The E.M.F. at the terminals of the apparatus is from The process is worked in this country by the Castner Kellner Company at 4 to 5 volts.214 DISCOVEKY OF ALKALI METALS BY H.DAVY: Weston Point, in Cheshire, and at Wallsend-on-'l'yne ; in Germany by Farbwerke vorm. Meister, Lucius and Brunning at Hochst, near Frankfurt, by die Elektrochemischen Werke Natruni, in Rheinfelden ; in France by the Cie. d'Electro-Chimie at Gavet ; also in America by the Electro- Chemical Company at Niagara Falls. Another process for the production of sodium by the electrolysis of sodium hydroxide is that of Rathenau and Suter, which has been in operation since 1895 at the Elektrochemischen Werken at Bitterfeld, aiid also in a modified form in the Aluminiumaktiengesellschaft a t Neuhausen. We have already noticed that one of the chief difficulties in obtaining sodium by the electrolysis of sodium hydroxide is the tendency of the sodium as it is produced to dissolve in the molten electrolyte. Castner gets over the difficulty by surround- i n g the cathode with gauze, which prevents the metal from diffusing throughout the electrolyte, and directs its path upward to the col- lecting recepticle, also by paying attention to the temperature of the bath.Rathenau and Suter get over the difficulty by allowing the electrodes to just dip in the molten electrolyte, Fig. 5. The surface tension causes the electrolyte to rise up slightly, SO that the electrodes are not actually below the surface of the molten hydroxide. On passing the current metallic sodium is produced at the ends of the electrodes, and here it partially solidifies and can be removed by means of ladels. In order to prevent over-heating at the electrode the inventers state that the current should not be of greater intensity than 10 amperes per square centimeter.The process of Becker, which is worked in France (Usines de Rioup&oux), is cf interest. It differs from the Castner method in that a mixture of sodium car- bonate and hydroxide is electrolysed. The apparatus employed is depicted in Fig. 6. The method of making the cathode tight so that the molten electrolyte may not leak out is much the same as that in the Castner apparatus. The cathode, however, is cone-shaped, so that the metal as it is produced may readily ascend to the collecting hood d ; or the cathode may be made of a number of slightly coned rods. By this means a greater surface is obtained and the metal ascends hetter.The hood is of iron or ferro-nickel, and is connected to the FIG. 6. negative source of current through a resistance. Thus, when thc metal comes into contact with the sides of the hood, it is in electrical connection with the negative pole, and, consequently,.. T _ _ - - FIG. j. 1BEARING OF T H E DISCOVERY UPON INDUSTRY 215 is not further acted upon by the molten electrolyte. The hood is air-cooled, or, if necessary, cooled by water dropping upon it so as to prevent the metal from actually distilling over, through the tube f, should the temperature rise too high. The metal, being specifically lighter than the electrolyte, rises up in the hood arid flows out through the side tube f l . The temperature of the electrolyte is much higher than in the Castner process, where it is only molten sodium hydroxide which is being electrolysed, whereas the average temperature of the Castner electrolyte during electrolysis is about 325'.In this case the temperature of the electrolyte between the anode and the cathode is about 550~. Every cell is worked with 1,250 amperes. Each apparatus produces about 40 kilos. of metal every 24 hours, and one man can work three such apparatus, consequently over IOO kilos. can be produced in one day by one furnace attendant. Many attempts have been made to produce metallic sodium on a com- mercial scale by the electrolysis of the cheaper sodium chloride in place of the more expensive hydroxide. In this case, however, owing to the very much higher temperature of fusion of the electrolyte, there is tendency for the sodium to be vaporised.The temperature at which sodium chloride melts is about yo', and metallic sodium distils at a little over v', conse- quently it is necessary to have the temperature of the bath very carefully regulated. Another drawback is the production of chlorine which at the high temperature at which it is produced is intensely active. But, of course, once the difficulty of dealing with the chlorine is surmounted, instead of being a disadvantage, it is an asset, being a useful by-product. The apparatus of Grabau-which, I believe, is not at present worked-was originally patented in 1889. I t is illustrated in Fig. 7. The decomposition vessel A, made of porcelain or other fire-resisting mate- rial, is externally heated by furnace gases which pass round it by means of the flues G, the decomposition vessel itself being placed in an air bath L L.The anodes of carbon C C are separated from the cathode by means of a porcelain partition which reaches down in the electro- lyte as far as the bottom of the electrodes, and is hollow so that heated air or gases may be passed through it. This prevents the sodium produced at the cathode and the chlorine at the anode froin diffusing and recom- FIG. 7. bining during the electrolysis. This pole partition is connected air-tight round the iron head E, which has a side tube connecting it with the receiver M. The alkali metal being very fluid at the high temperature, almost at its distilling point, rises up in the tube E and flows out through the tube a.The chlorine gas passes out by d. Fresh supplies of sodium chloride::: * Grabau suggests a mixture of NaC1, SrC1, and KCl as having a Iower melting- point. With such a mixture, however, pure sodium could not be produced. VOL. 111-T8216 DISCOVERY OF ALKALI METALS BY H. DAVY: can be added from time to time through e. As the temperature at which the metal flows into the receiver is very high it is advisable to lead an indifferent gas through it by c. J. D. Darling, in America, in the works of the Brothers Harrison, a t Phila- delphia, electrolyses sodium nitrate. The difficulty in this process is that the sodium when liberated reacts with the fused nitrate, and in the first place reduces it to nitrite, further electrolysis results in pure sodium being obtained.At the anode NO, is give!i up, and this then splits up into NO, and 0. Tlie nitrogen peroxide is absorbed i n water, and from this nitric acid is produced. I t was not found possible to work successfully with a common electrolyte for anode and cathode owing to the reduction referred to above. Darling therefore divided his cell up by meaiis of a diaphragm, the anode com- partment contained fused sodium nitrate, aiid the cathode compartment fused sodium hydroxide. It is, therefore, only necessary to feed sodium nitrate into the anode cell, and here decomposition takes place, NO', anions being given up and the Na cations passing to the cathode cell where they in turn are yielded up. The difficulty which now faced Darling was to obtain a diaphragm which could be employed in a fused electrolyte.The form of cell adopted is n . c B FIG. 8. illustrated in-Fig. 8. It consists of an iiiiier cell of perforated sheet iron ~vhicli is placed in an outer cell also of perforated iron. Tlie spaces between the two cells may he filled in with sand, magnesia, or brokeii glass, but the substance actually used is a mixture of Portland cement and magnesium oxide. The product is made by heating magnesite until it becomes hard and dense. It is then broken up so that it will pass through a ao-mesh sieve, but not through one of 30 meshes. This is mixed with Portland cement, moistened with water, and filled in between the two portions of the cell. The cell is made the cathode compartment, and is numbered 17 i n the diagram, Fig.9. It is placed within a closed iron vessel which coil- tains the fused sodium nitrate, fused caustic soda being in the cathode cell. The cathode compart- ment is surrounded by the iron anode, and the gases produced in the anode chamber'pass out through the tube marked with two arrows. It is found that the life of the metal part of the diaphragm is very much prolonged if about 5 per cent. of the current is allowed to pass through it. This, of course, meaiis a certain waste of energy, but it well repays itself by preventing the destruction of the cell. As shown in the diagram, the current which passes through the diaphragm is regulated by the resistance 3 1 . In practice twelve such cells are worked together, each cell taking 400 amperes at 15 volts. Every apparatus is made of an outer iron cell 55 cm.in diameter. The bottom of the cell is covered to a depth of 15 cm. with insulating material, such as Portland cement. The cathode cell rests upon this, and has a height of 75 cm., aiid an outer diameter of 40 cm., the inner diameter being 20 cm., and the depth 65 cm. The iron outer cell is connected with the + source of the current, the cathode, which consists of a 10 cm. iron tube, with the negative pole, At the coinnieiiceineiit of the electrolysis the cell is externally heated. Potassium is only innnufactured upon a very small scale, either elec-BEARING OF THE DISCOVERY UPON INDUSTRY 217 trolytically or by purely chemical processes. When it is manufactured by electrolytic means it is made by the electrolysis of potassium hydroxide, in similar apparatus to that used for the production of sodium.The practical difficulties, however, are greater than those met with in the manufacture of sodium, owing in part, at any rate, to the more ready oxidisability of the metal. The advance in these manufactures within the last decade is enormous, and the astonishing fact is that the process originally employed by Davy when he discovered the metals one hundred years ago is that adopted to-day. The process of Becker, in which a mixture of sodium hydroxide and carbonate is electrolysed is really only a modification of the Castner process. Darling's process by the electrolysis of sodium nitrate is the only one worked to-day in which an electrolyte other than sodium hydroxide is employed. The present-day production of metallic sodium is about 5,000 tons annually, and the price is 8d. per lb.; fifteen years ago it was nearly as many shillings. The preparation of the amalgams of the metals calcium, strontium, and barium has already been mentioned, p. 210. Davy prepared these amalgams by the electrolysis of the oxides. The only successful methods proposed for preparing these metals to-day are by means of the electrolysis of their fused salts. In 1856 Bunsen and Matthiessen succeeded in preparing small quantities of calcium by electrolysing the fused chlorides ; but Moissan, in 1898, was the first to prepare any quantity of calcium by means of the electro- lysis of fused calcium iodide.;k Owing to the high price of calcium iodide, this method cannot be employed on a commercial scale.On the other hand, it is a good laboratory method, because the calcium so obtained possesses a high Comfit. rend%, 1898, 126, 1753. VOJ;. III--T8*218 DISCOVERY OF ALKALI METALS BY H. DAVY’: degree of purity. Borchers and Stockem ::: obtained the metal as a spongy mass by electrolysiiig fused calcium chloride at a temperature below the melting-point of the metal. The spongy product so obtained contained from 50-60 per cent. of metal. By compressing this the excess of calcium chloride was forced out and a metal of c)o per cent. calcium obtained. The object of electrolysing the calcium chloride at a temperature below the fusing-point of the inelal was to prevent reaction ensuing between the calcium chloride and the metal. At a high temperature, a sub-chloride being produced- Ca + CaC1, = Ca,Cl,.Suter and Redlich have succeeded in getting over this difficulty by employing an electrode after the samc principle as that adopted by IZathenau and Suter for the manufacture of sodium, which has already been referred to, €7. 214. This kind of electrode might perhaps be describcd as as a surface-tension electrode, in Gcrniany it is called ‘‘ Berulirungselektrode.” Calcium chloride is thoroughly dehydrated, and then fused in a suitable electrolysing cell, modes of carbon being employed. The cathode is of iron, and is arranged to just touch the surface of the fused salt. Shortly after the current has been passed a globule of metallic calcium forms at the end of the cathodc. The cathodc is now slowly raised, so that the connection between the electrolyte and the electrode is between the metallic calcium.As iiior~: calcium is deposited out so the electrode is raised, until finally a long rod of the metal is produced at the ciid of the iron cathode. The bath being hot, and ;L heavy current being employed, the electrode remains This prevents the thin protective coating of calcium chloride which clings the calcium is unacted upon by the atmosphere during FIG 10. the course of the operation. Fig. 10 represents the elec- trode as it appears after the electrolysis has been in progress for some time. A is the iron cathode, I3 the metallic calcium, and C the electrolyte. At the comnienceinerit of the operation it is advisable not to employ too high a current density, or the temperature may rise too high and cause the calcium to redissolve as it is formed.According to P. Wohlcr,+ the commencing current should be 40 amperes per square centimeter, which should be gradually raised to 250 aniperes per squarc centinieter. This process is carried out at the Elektrochemischen Werke, at Bitter- feld. In France Poulenc Frkres also manufacture metallic ca1cium.f In this case a cathode of molten aluminium is employed, and an alloy con- taining 97 per cent. of calcium is produced. J. H. Goodwins has also described a laboratory furnace, from which very good results can be obtained. The furnace itself is made of Acheson graphite, and this constitutes the anode. The cathode is of iron, and can be raised or lowered by means of a screw arrangement. ‘l’he electrolyte consists of calcium chloride. The furnace is depicted in Fig. 11. E is a copper coil for circulation of water, through which is a thick piece of asbestos C, holding by the blocks D, the copper coil E well up in the Acheson graphite anode F, and insulated from it by the asbestos G. For __ hot and the calcium also keeps hot. -_ .. - - ___ -. - - - to the metallic calcium from delequescing, consequently * German Patent 144,667. German Patent 144,777. t Zeifschrift f u r Elektrocltemic, 1()05, 612. 5 Elecfrocltcrtricnl nrzd Mcinllurgicul lrtdustry I l l . , 80.BEARING OF THE DISCOVERY UPON INDUSTRY 219 the rest the apparatus explains itself. A metal of over 98 per cent. is obtained in this manner. The metals barium and strontium are also made from the electrolysis of their fused chlorides, but so far therc have been considerable difficulties in preparing them in the pure condition, and they have not been manufactured on a commercial scale. Davy also endeavoured to prepare mag- nesium by electrolysis, and apparently did prepare an amalgam of this metal with mercury. He was unsuccessiul in his attempts to prepare pure magnesium, but to-day both of these metals are manufactured by electrolytic methods. Calcium carbide was also iirst pre- pared by Davy, and now, as every one knows, this substmce is tnanufacturcd upon a very large scale by electrothermic methods. The work of Davy upon pieces of metal the size of pin hcads has resulted in the manu- facture of one of these metals by thousands of tons, and has branched out and borne fruits, which have resulted in the eiiiploynient of thousands of persons in various new indus- tries ; furthermore, science has been enriched in a manlier that even Davy in his wildest dreams could not possibly have foreseen. FIG. 11. NoTE.-The biographical portion of this paper was taken from Thorpe’s “ Life of Davy.” The other portions in connection with Davy’s work from two original papers. The portrait of Davy is reproduced from Ostwald’s Elecirochemie. Figs. 8 and 9 have been taken from Die EIektrometallurgie der Alkalimetalle, by H. Becker.