摘要:

Thc Faraday Society is not respoizsible f o ~ opirlions expressed bcforc it by Authors or Speakers. O F FOUNDED 1903. To PROMOTE THE STUDY OF ELECTROCHEMISTRY, ELECTROMETALLURQY, CHEMICAL PHYSICS, METALLOGRAPHY, AND KINDRED 8UBdECTS. VOL. IV. APRIL, 1909. PART 3. T H E INFLUENCE O F CHEAP ELECTRICITY ON ELECTROLYTIC AND ELECTROTHERMAL INDUSTRIES. BY E. A. ASHCROFT, A.M.I.C.E., M.I.E.E., &c.::: ( A Paper read before the Faraday Society, Monday, December 21, 1908, DR. N. T. M. WILSMORE in !he Chair.) DISC USSZON. Mr. W. R. Cooper (conzmuiticated) : There is no doubt, of course, that cheap power is the essence of success in certain electrochemical industries, namely, wherever the cost of power is a considerable portion of the whole. There is always a difficulty when discussing the cost of water-power, for the simple reason that the cost of such power must depend upon the price that it will fetch, and this price is not likely to have any fixed value.I n dealing with coal, we know that the price of coal cannot vary beyond certain limits, but if a contract is made for water-power for, say, seven years, it does not follow that the price of the water-power subsequently will not be raised, owing to the greater demand for power in that particular neighbourhood. I do not suppose there is any doubt that the price of water-power will tend to go up considerably; but, on the other hand, the cost of power from various kinds of fuel is likely to go down, though possibly not to a very serious extent. It follows, therefore, that water-power should be purchased outright by the companies running electrolytic or electrothermal industries, and not merely leased.Mr. Bertram Blount thought the author had been a little hard on sources of power other than water. In many electrochemical industries questions of freight, labour, &c., gave the advantage to more costly sources. The notion frequently held that nothing but the cheapest power, say, 30s. per kw. year, was of any use to electrochemical methods was erroneous. I t was a strange but undoubted fact that however costly electric heating appeared to be, its economy in application frequently counterbalanced its * This Paper was published in the volume of Transacfions (vol. iv.? part 2, p. 134) issued in October, 1908.150 T H E INFLUENCE OF CHEAP ELECTRICITY ON high prime cost.It was for this reason that he believed there was a great future for processes such as electric zinc-smelting, although up to the present so little had been achieved commercially in this direction. The 13s. per kw. year allowed by the author for capital charges in the case of the best water-powers (p. 135) seemed to him high, but, of course, the shortsighted policy of impeding industry pursued by the Norwegian Govern- ment had to be borne in mind. The Paper was based on calculations of cost of power to bc made and sold as such, but a different point of view might be adopted, according to which the power was produced by the consumer. This was certainly a more favourable state of affairs. Mr. W. Murray Morrison agreed with the author that the subject had not received sufficient attention, and he was glad that it had been brought before the Society for discussion.He took it, though it was not stated in the Paper, that in all cases the author’s figures were based upon a load factor of IOO per cent., and he thought Mr. Blount was right in his criticism of the author’s treatment of steam and gas engines. The cost of €8 6s. 6d. per kw. year given for electric energy generated by the former was high, and he thought that under favourable conditions electricity could be produced from steam for probably as low as ;Gs per kw. year. In the case of gas engines, also, he knew of cases where very good results had been obtained on a large scale, especially where valuable by-products were pro- duced. He thought, however, that the author’s charges for labour were considerably understated in both these cases.It might very possibly be the case that the capital cost, and hence interest charges, in the case of steam and gas plants were less than those for a water-power plant of an equivalent size, but the depreciation charges would be much greater, as also labour and stores, and the fuel costs would be an extra. He did not altogether agree with Mr. Ashcroft in his classifications of water-powers. His Class I. was a very good example of a fall, but one not often met with, and certainly not on a large scale conveniently situated. His Class II., however, was a bad specimen, and there were many intermediate in value between the two classes. The author had unduly penalised Class 11.as far as labour and upkeep were concerned. There would be greater depre- ciation in the case of cheap powers, where the capital would be mostly sunk in plant, than in dearer powers, where the moiiey would be chiefly spent in earthworks, &c., and in royalties. Again, the ;Go 2s. per kw. year capital charge in the case of Class 11. meant that about ;Gqo per kw. must be spent on plant, &c. This was an exorbitant figure. It was these considerations that made him think that the differences between Classes I. and 11. stated in the Paper were excessive. The English plant quoted by the author (p. 138), on which _G60 per h.p. had been spent in capital expenditure, must have been a very small one, as the price was one which no engineer or capitalist would care to consider.Referring now to Table II., he noted that the cost of construction amounted to A7 8s., contingencies to ISS., and purchase consideration (z.e., existing mortgages and payment in shares) to 610 14s. per kw. installed. This gave a grand total of ;G19 per kw.-by no means a low figure. Even these figures, however, were based largely on estimates, and without great experience at the back of estimates of this kind they usually fell short of what was required. He certainly thought the author had allowed too low a figure for the plant, and had not left sufficient margin for contingencies. He himself had been pretty well’all over Norway, and he had many ‘‘ raw ” properties (that is LO say, land and undeveloped water-rights) offeredELECTROLYTIC & ELECTROTHERMAL INDUSTRIES 151 him at prices equivalent to about 10s.per kw. If Class I. powers were brought down to this figure, they would be more attractive, but he feared not many such were to be found. There were one or two other considerations worth drawing attention to. The position of the source of capital and of the management was of some importance in connection with labour questions. Foreign capital usually carried with it its own labour, and unless this was done on a very large scale labour troubles were pretty sure to arise. This was a contingency worth bearing in mind. Added to the safeguarding of capital, management, and labour, numerous other considerations called for careful investigation before fixing on the site of a power supply, such as source of raw materials, market for finished product, freights, possible tariffs, or other restrictions.While agreeing with the author that well-chosen and well-designed water-powers opened the way for the successful operation of many electrometallurgical and electrochemical industries, in a manner which no other source of energy was likely to do, he did not go so far as the author in saying, as he did, that such alternative sources of energy had in all cases ‘’ no chance at all,” or that water-powers strictly of Class 1. were the only ones to be looked to for the attaining of the desired and necessary economy. He quite agreed with the author in thinking that 110 two water-power schemes were alike, and each should be considered on its merits. Generali- sations in this connection were therefore to be accepted with great caution, for conclusions reached for a particular water-power scheme, as well as for a particular industry and a particular country, could not be applied elsewhere without taking every possible difference of conditions into consideration. Mr.Charles Weiss said he had taken especial interest in the available water-powers in Bavaria and the Bavarian Highlands. Apart from private owners of water-rights, the swift-flowing rivers and several lakes belonged to the State, and the latter had introduced legislation retaining State control of their water-rights, and only leasing them to manufacturers on conditions beneficial to the State. But it was proposed to use the electrical energy generated first for the shorter State Railway lines in South Bavaria.Some people in England might call this socialistic legislation, but in Germany the State took much greater powers for the benefit of the people. He agreed with previous speakers who had dwelt on the existence of cheaper steam-produced electrical energy than was allowed by the author ; and he knew of at least one case in this country where electricity in bulk for a roo per cent. load factor could be had at under ;G6 per kw. year froin steam turbo-generators. For certain electrochemical industries this price was quite acceptable. Mr. Leon Gaster : I fully appreciate the importance of employing waterfalls, &c., for the generation of electric power for electrochemical purposes, and so preserving our diminishing stores of coal ; as water-power, unlike coal consumption, involves no actual *using up of valuable and unre- placeable material, this course is obviously desirable. However, the feasibility of the system is often influenced by other factors, such as transport facilities, supply of material, and labour, besides the actual economy of power conversion.The nature of the water supply, &c., as the lecturer pointed out, may render the original cost of installation very heavy, and this may be accentuated by ill-advised and even prohibitive legislation. In Norway, where the conditions are specially favourable for the develop- ment of water-power, the legislation ought to be much modified in order to stimulate the investment in the country of foreign capital. Inconvenient legislation is also sometimes resorted to on aesthetic grounds.152 THE INFLUENCE O F CHEAP ELECTRICITY ON For instance, I was informed that at Niagara an agitation against destroying the natural beauties of the falls (originated, it is believed, by parties interested in coal in the United States) resulted in legislation restricting the use of the water to but a fraction of that originally contemplated. Naturally, therefore, the first cost of the installation now appears needlessly high for the present With the increased utilisation of water-power manufacturers have greatly improved the design of turbines, which are now suited to both high and low falls, and answer all requirements.The matter is also engaging the attention of many Governments whose representatives are undertaking a survey of the available sources of power, with the view of framing legislation for their ultimate extensive use.Gas engines and steam turbines can now be con- structed with higher efficiencies than those mentioned by the author. I should, therefore, like to know what was the size of the plant which the author had in view in compiling the comparative costs shown in Table I. on page 135. With cheap water-power as described by the author it may be possible to economically establish many lucrative electrochemical manu- factures in Europe which are now developed in the United States of America. Dr. H. Borns remarked that the author had not drawn attention to some of the disadvantages of water-powers, such as the troubles due to occasional dearth of water and to ice. Mr.E. A. Ashcroft : Ice difficulties do not exist in Norway, where water can usually be drawn off many feet below the surface of the supply, although they do exist in Sweden and Finland. output. Was there no need of a steam-power reserve? Dr. H. Borns : This must give Norway a great advantage. Professor R. H. S m i t h said the ice question referred to by the last speaker was an important one, for although W. Norway appeared to be particularly favoured in this respect, most localities were not so. He had in mind the power plant at Valtellina in the Alps for supplying the railway with current, where the exclusion of ice from the channels cost considerable money and trouble in the clesign of the works. The enormous variations in supply during the year and the consequent ex- pense of provision therefor was another point which told against water-powers. He assumed that the figures in Table I.were calculated on a load-factor basis of IOO or full load for 8,760 hours per year. Of course under those conditions it was easy to get good results. On that basis of 8,760 hours full load per year his figure of &6 9s. 6d. for steam engines worked out to 0.18 of a penny per 1i.p. hour. On the same basis his (Prof. Smith’s) calculations in the chapter on “ Capital and Working Costs ” of his book upon (( Commercial Economy of Steam and other Heat Eiigines” would give practically the same result for condensing engines ; but the figures given in that chapter were worked out for 50 x 54=2,700 hours full-load working per year.The cost varied greatly with the class of steam engine, gas engine, or oil engine, and on the size of the installatibn, as also, of course, with the price of fuel. These variations were fully illustrated on pages 129 and 137 of the above book. He did not see how it was possible even with water-power to reach the low figure of ;G2 per kw.-year, and certainly not in Great Britain. In the figures given on page 140 of the Paper the author allowed 10 per cent. upkeep for buildings and only 5 per cent. for machinery. This relationship was surely extravagant. Did he rightly understand these allowances to be for upkeep only, seeing that depreciation was not specially mentioned ? In his opinion the importance of depreciation was not fully recognised in England, and the result was that electric power was frequently being sold liere for under cost price.This was being done in the case of theELECTROLYTIC & ELECTROTHERMAL INDUSTRIES 153 older electric lighting stations which were built for lighting only and could not compete with the modern stations. In order to fill up partially the valleys in their lighting load curves, they sold current for power without debiting this power supply with any reasonable proportion, sometimes with no share at all, of capital charges. Chemical engineers would be wise to reckon on the impossibility of electric power being produced in this country for less than Id. per unit delivered to the consumers. Dr. E. Feilmann said that the success which had lately attended the production of gaseous fuel from peat had very favourably influenced the cost of power from gas engines ; large installations for gasifying peat by the well-known Mond process were now either at work or in construction in England, Germany, and Italy, that in the latter country being designed for a production of 3,000 i.h.p.The ammonium sulphate produced as a by-product was stated to bring down the cost of the gas to almost zero, and in some cases it was said that it even paid to burn it to waste. Under such conditions the cost of power from a gas engine must approach very nearly to that from the author’s Class 11. waterfall. He did not believe that the electric smelting of iron could be carried out on a commercial scale for many years to come. According to Hard& (Elccf. and Met.I d . 7. 16) we had an expenditure per ton of iron in electric smelting of $4’80 in fuel for reduction, $3’90 for electric power, and $ I for electrodes ; this was on the assumption that power costs $20 per kw. year, but even at half that price, or E 2 10s. per kw. year, the cost was $7.75 per ton for electric smelting, as against $6.84 per ton for smelting in the blast furnace, according to Hardkn’s figures, which omitted the cost of labour, which was estimated to be about the same in the two cases. Mr. E. A. Ashcroft, in reply, said lie was glad the Paper had given rise to so interesting a discussion. As regarded the criticisms that had been made, he thought these, in many cases, had answered one another. The special object of the Paper seemed not to have been made perfectly clear.He was not advocating water-powers as against steam or gas per se; that was a much larger question, and had been pretty thoroughly discussed and written about in other places. He was merely trying to call attention by this Paper to the very special kinds of water-power that still existed in ;L few favoured localities, but were somewhat rare, and which, if developed, would allow electrochemical industries to grow up that otherwise could not exist. It seemed to him a matter o e r e a t import- ance to call general attention to the existence of economical conditions in the Norwegian Falls which will in the future render practicable electrolytic and electrothermal manufactures which are, and must always remain, economically impossible in any other place.He absolutely differed from Professor Smith in what he said regarding the future minimum cost of electrical energy (5d. per unit). Professor S m i t h : I beg your pardon, I meant 0.5d. per unit. Mr. Ashcroft : Of course that makes a great difference, but Professor Smith’s figure is still too high. He knew cases where electricity was sold in bulk, on contract, for ad. per unit (equivalent to about ;G9 per h.p. year), or even half that price, and he knew for certain that ad. per unit price could be made to pay the vendors handsomely. But he had in mind something nearer 30s. per h.p. year, for at A9 iron smelting, for example, offered little or no inducement, low-grade sulphide ore treatment became too expensive, and nitrogen fixation was out of the question.But in Norway iron ores can be, and will be, smelted electrothermally at a sooner or later date in virtue of the cheap power and cheap freights-there were no insuperable difficulties in the154 T H E INFLUENCE OF CHEAP ELECTRICITY way of establishing such an industry from the engineer’s point of view. ‘The same applied to the manufacture of nitrates, which by two different methods were already an established success in Norway, whilst the same industry was not yet established anywhere else in the world. He understood that a com- pany had been formed in one of the Eastern States of America to work the Birkeland-Eyde lime nitrogen process ; he doubted whether this would pay there, but it could and did pay in Norway, with power at one-third the price which it is possible to obtain in America, and with cheap fjord-borne carriage. These figures had been worked out some time ago and no doubt might be somewhat open to criticism to-day, but as far as the author could get up-to-date informa- tion, they nevertheless represented a fair average.The sizes of the plants in Table I. (replying to Mr. Gaster) were 5,000 kw. He knew that a large Company in Newcastle was supplying electric energy generated by steam plant at prices said to be ;Gs per kw. year to electrochemical consumers. He could only say he failed to see how this was profitable to the Company, if fair allowances were made on all usual accounts, such as depreciation and also having in view the present prices and known consumption of coal, but even at that price steam-power could not compete in the special fields he was dealing with.His figures for depreciation in the capital charges for Falls of Class 11. had been criticised by Mr. Morrison. But he thought Mr. Morrison could not have clearly grasped his figures, which were ample for the purpose, and more drastic than are usually allowed in calculating steam plants ; for if only 3 per cent. were allowed for earthworks-which was sufficient-there was still 5 per cent. for machinery, leaving also a large difference (of 8s. to 10s. per h.p.) for other contingencies, such as debenture intere$, royalties, &c. He would have liked the efficiencies of electric furnaces to have been dis- cussed. The futiire of electrometallurgy, especially of electrothermal pro- cesses, much depended on this point.For instance, what actual thermal efficiency can be obtained from an iron-smelting electric furnace, or again from electric zinc retorts? It seemed to him probable that the latter might be easily more than three times as great as the efficiency of the old retort furnaces, so that the line of economical limitations in this case may be already crossed. The realisation of the figures in Tabl,: 11. in practice had been doubted by some of the sBeakers, but many reliable firms were willing to contract at the figures given, and he knew of several falls nearly as cheap to develop as the one referred to-for example, a fall of 40,000 h.p., from which power could be very profitably sold at 30s. per k w . year. His figures, however, represented a probable average for Norwegian power rates when sonic of the few exces- sively cheap powers at present available were snatched up out of the market. Replying to Dr. Borns, to have reserves of generating plant to take up the load during temporary stoppages-that is to say, duplicating the generating plant-was out of the range of practical politics in electrolytic and electro- thermal work. Certainly in connection with the cheap powers in Norway it would seldom prove economical except in a very limitcd sense-such, for instance, as installing five units of plant and operating four, thus keeping one in reserve. If a crust of ice formed on the top of the lakes, the water was drawn away from underneath. At Niagara the charge for removing ice from the intakes was a factor to be reckoned with. He was sorry not to be able to give Dr. Wilsmore any information regard- ing the Falls of the Upper Nile and the Zambesi. The figures for steam plant in his Table I. had been criticised. The ice in Norway gave little or no trouble.

ISSN:0014-7672    

DOI:10.1039/TF9090400149

出版商:RSC    

年代:1909

数据来源: RSC

摘要:

NEW APPLICATIONS OF ELECTROMETALLURGICAL ALLOYS. BY AD. JOUVE, Ingknieur-Conseil, Ancien preparateur de Chimie 1’Ecole Polytechnique, Directeur de la Revue d’Electrochimie et d’Elecfrome‘tallurgie. ( A Paper rend before the Faraduy Society, Tuesday, June 23, 1908, PROFESSOR A. K. HUNTINGTON, VICE-PRESIDEST, in the Chair.) Among the alloys prepared by means of the electric furnace the most important are those containing iron, and particularly the ferro-silicons. Substances prepared in this way have a percentage of silicon varying from 20 to 75 per cent,, the most usual being 20 to 25, 50 to 55, and 70 to 75 per cent. Their uses are well known to all metallurgists, while the analytical methods for determining them are very numerous, accurate, and wholly without complication, being all based on the sclutions of the substance in alkalies in the presence of an oxidising agent such as peroxide of sodium, sodium oxide, or free oxygen. I t is a fact well known to analysts that no methods of analysis are based on the effect of acid, except in the case of hydrofluoric acid, for the reason that ferro-silicons with percentages above 2 0 per cent.are insoluble in acids. It is this peculiar property, due to silicon alone, which has been employed in an altogether unusual way ; i.e., instead of placing ferro-silicons in acids to dissolve them, acids during their preparation may be contained in f erro-silicon apparatus. Thus, by using high percentage ferro-silicons, apparatus practically un- affected by acids are obtained. Ferro-silicons are not, however, the only substances which possess this property.For, as this unalterability is due to the silicon, any alloy of a metal with this metalloid will behave in the same way to a greater or less degree according to the nature of the metal. I t is evident, for instance, that calcium silicide will be unaffected by acid ; the same phenomenon, in fact, occurs, for it is evidently much less affected than metallic calcium would be. In answer to the question whether the problem of constructing apparatus for the preservation of acids is solved by making use of metallic silicides, it may be said that (I) these substances are very fragile ; (2) they contract con- siderably on cooling, often breaking in the mould when scarcely solid and still at a bright red heat ; (3) their melting-point is low compared with the temperatures occurring in ordinary foundry practice--e.g., the second melting in a coke or furnace gas-heated cupola ; and (4) certain of these compounds are powerful reducing agents, so that an effect similar to that produced by aluminium may occur.In consequence, fusion in the presence of the air necessary for combustion of the coke or furnace gas is impossible without 1.55156 NEW APPLICATIONS OF considerably lowering the percentage of silicon (especially in rich ferro- silicons-i.e., about 35 per cent.) to such a point that sometimes it will become as low as 15 per cent., and thus give alloys which are attacked by moderately strong acids. Manufacture of Afparatus.-It is not proposed to enter here into complete details of the foundry methods employed, which belong rather to technical metallurgy, and it will simply be shown that by quite special and perfect arrangements, as well as by the correction of the mechanical properties with alloys of different natures, it has been possible to obtain, after six years' research, apparatus of all kinds and descriptions.The first part of the process consists in obtaining melted metal suitable for pouring into moulds. The simplest and most used of these methods consists in pouring alloys, melted in that apparatus, directly from the electric furnace into a mould lined with some refractory material. The mixture is then made. In a second method, the alloys obtained from the electric furnace are remelted in an oil, or in certain cases in a crucible, furnace and then cast into ingots.Once the liquid metal is obtained, metals for corrective purposes are added. The number and nature of these is practically infinite for any par- ticular case. They are added directly to the metal in the casting ladle, usually with aluminium, which raises the temperature of the bath and increases the fluidity. When the ladle is ready it is used mechanically, by a travelling crane, and carried quickly, to prevent cooling, to the mould. The pouring is effected by ordinary means without any special precautions. The casting is removed as soon as it solidifies, is turned out into the warm casting sand, and thus annealed at a bright red heat. It is then ready for use. Resistaizce to Acids.-The resistance of the metallic alloys of silicon, which have been called '' Mktillures," to acids is quite interesting, as the following examples will show.Nitric acid, even as a vapour, such as is obtained at the exit of a bisulphate retort or when mixed with nitrous acid, does not affect them a t all. A striking example of this is given by a pipe which has been submitted for nearly five years to the daily passage of 660 lbs. of nitric acid vapour at temperatures varying from 150" to 200" without its loss in weight exceeding a few decigramnies in a total weight of a score of kilogrammcs. This loss occurred quite at the beginning of the period, and was probably due to a few impurities remaining on the inner surface of the pipe after fusion. Sulphuric acid has still less effect, and acids of strength from 20 to 25, known as petites eaux, may be concentrated in vessels of this kind by direct heating and continuous evaporation.Hydrochloric acid is analogous in its action, and use has been made of pipes of the above material for carrying and condensing its gases. Acetic acid and the mixture produced by treating calcium acetate with sulphuric acid are without effect on " Mktillures," even in the presence of air. Resistance to other Reagents.-An important English firm has obtained the following figures relating to the resistance of " Mktillures" to the action of cyanides, sulpho-cyanides, and prussic acid. The results are very remarkable, for they show both that the unattackability increases with the percentage of silicon, while for a suitable percentage it is practically negligible.ELECTROMETALLURGICAL ALLOYS I57 NO.I 2 3 4 5 6 I 2 3 4 5 6 '' M~TILLURES "-20 PER CENT. NITRIC ACID. Weight in grms. 6.4410 5'1669 8.1524 I4.5013 I -7060 1'4941 6.4018 5.1622 8.1487 14'5025 1'6344 1.4907 No. Change in Weight. After 24 hours. - 0'0392 - 0.0047 - 0.0037 + 0'0012 - 0.0716 - 0.0034 - 0'0207 - 0.0006 - 0*0032 + 0.0045 - 0*0007 After 48 hours. - 0'0323 Reaction with Sulpho- Cyanide. MhTILLURES-13 PER CENT. SULPHO-CYANIDE OF ANILINE. Violent. Slight. Moderate. Moderate. Very violent. Very slight. Violent. Very slight. Moderate. Slight. Violent. None. Weight in grms. 6.381 I 5.1616 8.1455 14'5070 I '602 I 1 '4900 Change in Weight. After 24 hours. - 0.0387 - om065 - 0'04.55 + 0.0055 - 0.0186 - 0.0160 After 48 hours.- 0.0675 - 0-0219 - 0.0433 - 0.0063 - 0.02 16 - 0'001 I Reaction with Potassium Ferri-cyanide. Very violent. Moderate. Violent. Slight. Violent. Very slight. Very violent. Moderate. Violent. Slight. Violent. Very slight. Under the same conditions, iron or any other metal in commercial use would be rapidly attacked, thus preventing its employment. The apparatus constructed up to the present time is extremely varied both in form and dimensions, ranging from evaporating dishes, troughs, right-angled bends, elbows, worms, ventilators, &c., and in the other from those weighing about 20 lbs. and having a diameter of I ft. to those weighing over I ton and measuring 7 ft. across, as well as pipes with a diameter of 8 ft. and weighing about 4 of a ton. To sum up, it may be said that the production of a material capable of resisting acids is an accomplished industrial fact, and the principal desiderata are realised.It remains, however, to remedy the unmalleability of these substances. This question is actually being studied, and a solution seems probable in the near future. As electrometallurgists we may congratulate ourselves that without the electric furnace, which alone permits sufliciently high temperatures for the158 ELECTROMETALLURGICAL ALLOYS reduction of silicon or silicates by carbon to be obtained, the high perccntage alloys of metals with silicon could not have been manufactured. Further, without the electric furnace the fusion of high-percentage silicides would not have been possible. For when the percentage exceeds 50 the temperature of a coke or gas furnace is not sufficient to permit these alloys to be cast, not because they will not run, but because their specific heat is too low for the metal to be poured. In fact, it will not leave the crucible, but forms a pasty mass. This latest conquest of the electric furnace has come at a most suitable time. For platinum, the most stable industrial metal, is becoming more and more rare, while its price is increasing by leaps and bounds, making its employment a matter of greater and greater difficulty.

ISSN:0014-7672    

DOI:10.1039/TF9090400155

出版商:RSC    

年代:1909

数据来源: RSC

摘要:

A REDETERMINATION OF T H E ELECTROLYTIC POTENTIALS OF SILVER AND THALLIUM. BY F. J. BRISLEE, D.Sc., Muspratt Laboratory of Physical and Electrochemistry, University of Liverpool. ( A Paficr read before the Favaday Socicty on Tuesday, Dccenzbcv 15, r y 8 , Dr. T. M. LOWRY irz the Chair.) The potential difference between a metallic electrode and a solution con- taining ions of the same metal is given by the equation- where C is the ionic concentration of the solution surrounding the electrode, R the gas constant, T the absolute temperature, F the charge per gram equivalent, and n the valency. The electrolytic potential, E.P., is defined as the potential of a metal against a normal ionic solution of the same metal. The E.P. is measured by employing the calomel or hydrogen normal electrode, and determining the E.M.F.between the electrodes of the combination, metal -soh tion of known ionic concentration j! normal electrode. The measured E.M.F. includes the liquid potential at the junction of the two liquids, which can be calculated, in certain cases, by the Planck formula (Wied. A m . 40, 561, 18go), or as Cumniing (Zeitsclzr. Elektrochem., 13, 17, 1907, and Trans. Far. SOC., 2, 213, 1906) has shown, it can be eliminated in some instances by employing a saturated solution of ammonium nitrate as connecting liquid. The ionic concentration can be calculated from the degree of dissociation given by conductivity or freezing-point measurements, and the value for the E.P. calculated. For room temperature 17' C. the equation becomes- The E.P.at constant temperature is a constant for each metal, and indepen- dent of the nature of the salt employed, but depending upon the solvent. Jahn (Zeitschr. Physik. Clzem., 33, 545, 1900) has shown by very exact potential measurements that the degree of dissociation of strongly dissociated electro- lytes is not correctly given by conductivity determinations, but Cumniing (Zoc. cit.) has shown that, in the case of silver nitrate, the degree of dis- sociation, calculated in the usual way from the conductivity, is a very exact measure of the dissociation between the concentrations 4 and normal. I n the following research the electrolytic potentials of silver and thallium have been redetermined over a range of concentrations, and in different salt solutions, the ionic concentration being calculated from the conductivity, the most recent conductivity measurements being used for the purpose.EXPERIMENTAL PART. The E.M.F. measurements were made by the compensation method, using This instrument is designed for exact E.M.F. a Clark Fisher Potentiometer. I59160 A REDETERMINATION OF THE ELECTROLYTIC measurements, and has a range of 3’5 volts and rcading to O*OOOI volt. An Ayrton-Mather reflecting galvanometer of the d’Arsonva1 type was used as a detector, and the Weston cadmium cell as a standard. The cadmium cell was compared with a standard cadmium cell, standardised at the National Physical Laboratory. All measurements were made at room temperature, viz., 17’ C., and are given to the nearest millivolt. SILVER. The silver was prepared from the chloride by fusion with sodium car- bonate.The metal was then fused and cast into sticks 3-4 mm. in diameter. The sticks were cut into lengths about 20 mm. long and fused on to copper leads. The electrodes were cemented into glass tubes with marine glue, which was quite unacted upon by the liquids employed, and made water-tight joints. The electrodes were always freshly silvered before use by electrolysis of a silver potassium cyanide solution, then thoroughly washed and allowed to stand, all connected together, in a solution of the same concentration as that used in the subsequent measurements, for at least a couple of hours. Electrodes were also silvered by electrolysis of silver nitrate, and electrodes made of silver-foil and gauze were employed.Electrodes prepared as indicated showed no E.M.F. between each other when dipping into the same solution. The electrode vessels were the half elements described in Ostwald- Luther (Hand utzd Hiilfsbuch, p. 377), and were connected through a vessel containing a saturated solution of ammonium nitrate to eliminate the liquid potential (Cumming, Zoc. cit.). In order to prevent the liquid surrounding the electrode from becoming impure owing to the diffusion of the saturated ammonium nitrate, a tap was interposed and measurements made through the closed tap. The calomel decinormal elcctrode Hg 1 Hg,CI, - KCI / / was em- ployed as comparison electrode. The E.M.F. of this electrode being better defined than that of the normal calomel electrode, the mercury is 0.613 volt more positive than the solution (Richards, Zeifschv.PhysiR. Clzent., 24, 37, 1897). The calomel electrode was also protected from contamination by the connecting liquid by means of a tap. N I0 Preliminary measurements of the cell- showed that the potential could be reproduced to within two millivolts, using a number of different solutions and electrodes and repeatedly resilver- ing the electrodes. The current flowed in the cell from the mercury to the silver. The combination gave a constant E.M.F. after standing for about an hour and remained constant for a considerable time, in most cases overnight. The following values were obtained for the E.M.F. of the combination- N Y Hg I HgZCL ~ K c l II NHdNO, AgNO, 1 Ag, two electrodes of silver being measured in each case ; I .2 . 3. 4. 5. 0.4002 0~4002 0.4007 0.4018 0.4016 volts. A rotating electrode was next measured ; for, as Kistiakowsky (Zeifsclzr. Elektrochem., 14, 113, 1908) has shown, different values were obtained with certain metals as electrodes when rotating electrodes were employed. The solutions were made up with recently boiled out water and care takenPOTENTIALS OF SILVER AND THALLIUM 161 to exclude air from the electrode vessel and in some cases hydrogen was passed through the solution during measurement. The results obtained were- I. 2 . 3. 4. 0.40 I 2 0.40 I 2 0-4010 0.4010 practically the same value as that obtained with the stationary electrodes. Stationary electrodes were used in the subsequent measurements, at least three different cells being measured at each dilution, the three deter- minations agreeing to within less than I millivolt.The following table shows the results obtained for silver on silver nitrate the concentrations varying from 0.1 to 0*002 gram equivalent per litre. TABLE I . Concen- trated AgN03 0.5 0'1 0'2 0'0 I 0'0002 0.005 u o k J . - 0.663 0.81 j 0.900 0'924 0.950 0.968 E.M.F. of Combina- tion in volts. 0'438 0.400 0'344 0'345 0.328 0'305 + 1.0g1 + 0'977 + 0.958 + 0.94' + 0.918 + 1'013 + 0'774 + 0.736 + 0'700 + 0.681 + 0.664 + 0.641 I The degree of dissociation is calculated 1 0058 log C. I E.P. (d. I + 0'027 -+ 0'003 + 0'101 + 0.118 + 0.135 -k 0'157 + 1.078 + 1.076 + 1.078 + 1.076 + 1.076 + 1.075 E.P. (h). + 0.801 + 0'799 + 0.801 + 0'799 + 0'799 + 0'798 from the measurements of Xeinwehr (Ber.kgl. fiveuss. conductivity of AgNO, by Kohlrausch and Akad. dev Wiss., 26, 581-587 ; Zed. Elektrochem., 8, 628, 1402). The'mean value for the E.P. calculated from the above results is 1.076 + O-OOOG volts. a value in gooa agreement witn rne recenL uecei-miriacioii or u. iv. Lewis (Lezi. Plays. Clzem., 55, 473, 1906). A series of measurements was next made, employing silver acetate as electrolyte. The silver acetate was prepared by precipitating silver nitrate solutioii by a fairly concentrated solution of sodium acetate. The precipitate was three times recrystallised from hot water, and the resulting product, perfectly white and crystalline, was dried in a desiccator in the dark. The degree of dissociation was calculated from conductivity measurements, the limiting value pm =89-4 at r 8 O being taken from Kohlrausch and Holborn (Leitverm 6gen der E Zeklrolyle).The conductivity was measured at 18" C. by the usual Kohlrausch- Ostwald method and gave the following figures- PO I*m -. Dilution. Molecular Conductivity at 180 C. 0.060 0.040 0.025 0'020 0'010 0.005 0.004 0.0025 0'0020 0'0010 63-68 74-00 80. I 3 84.62 86-00 86-20 71.48 75'96 83'93 87'44 0.712 0'799 0.827 0'849 0.896 0.938 0'944 0.962 0,904 0'978 * 0.1 N.E. indicates the decinormal calomel electrodes. t H.N.E. indicates the hydrogen normal electrodes.162 A REDETERMINATION OF THE ELECTROLYTIC The electrodes were freshly silvered by electrolysis of silver potassium The E.M.F. of the cyanide and of silver nitrate before each measurement. combination- N N Ag I ; Ag C2H302 II N H4N03 (I lo KC1Hg2CL I Hg, gradually increased until a constant value was reached, in about two hours, which remained constant for a considerable time, and in many cases gave the same value after standing for three days.The following table shows the values obtained for the E.P. of silver in silver acetate solutions. TABLE 11. Concen- trated A gC2H30, 0.060 0.025 0'020 0'010 0'002 0.004 P O Pm -. 0.7 I 2 0.827 0.849 0.896 0'944 0.964 E.M.F. of Combin- ation. 0.382 0.362 0.358 0.340 0'323 0.311 f eZeL-tvode - E SOliZ. 0.1 N.E. = 0613 volt. e electrode - e sol9t. H.N. E. = 0. + 0.718 + 0.698 + 0'694 + 0.676 + 0.647 + 0.659 0058 log C. E.PC. + 1-74 + 1.073 + 1'072 + 1.076 + 1.081 + 1.073 E.Ph. + 0'797 + 0'796 + 0'795 + 0'799 + 0'804 + 0'796 The mean E.P.from all these results is + 1.075 volt t 0-0013 volt. Rejecting the last number, viz., + 1-081, the mean is + 1-074 & 0-0007. The general mean of all the measurements of the E.P. of silver, in both nitrate and acetate solutions, is + 1'0757 -+ 0*0006 volt. This value agrees with that determined by Lewis (Zoc. cit.), and is the most probable value for the E.P. of silver in a solution that is normal with respect to silver ions. Abegg (Haizdbzrclz der Auorg. Clzcm., Band ii., abt. i., 857) refers to an unpublished research by A. Jacques and gives + 0.796 volt for the electrolytic potential of silver from measurements made in silver acetate solutions. T HA LLI u M. The thallium employed in 'the following experiments was the thallium puriss supplied by the Litophone Fabrikation, Triebes, and on testing was found to contain only the merest trace of impurity.The thallium was cast into sticks 3-4 mm. in diameter, which were fused on to copper leads and then cemented into glass protecting tubes with marine glue. The electrodes were carefully polished with fine emery-paper and then washed in water and allowed to stand, connected together in a solution of the same thallium salt which was to be used in the subsequent measurements. The thallium nitrate, chloride, and hydroxide employed were prepared from the metal. The nitrate was repeatedly recrystallised and the purity checked by analysis. The chloride was prepared from the nitrate by precipitation, and after thoroughly washing was carefully dried. The hydroxide was prepared by bubbling CO, free air through distilled water containing a quantity of metallic thallium in small fragments.In this way a & normal solution of thallous hydroxide could be prepared in a few hours. The strength of the thallous hydroxide was deter- mined by ti tration with standard acid, employing phenolphthalein as indicator. The first measurements of the E.M.F. of the combination- N N T1 I TINO,; jJ NH4N03 11 GKClHg2C1, I HgPOTENTIALS OF SILVER AND THALLIUM 163 Concen- trated TINO,. -- 0.10 0.02 0.005 0.004 0.002 0'01 gave inconstant values, owing to the solutions becoming alkaline, by the solution of the thallium, forming thallous hydroxide, but by employing boiled-out water, and passing hydrogen through the solution during measure- ment, consistent values were obtained and the solutions remained neutral.The degree of dissociation of both thallium nitrate and thallium chloride was calcuiated from the conductivity measurements of Kohlrausch and von Steinwehr (Ber. d . kgl. Preuss. Akad. der Wiss., 26, 581-587 ; Zeitschr. Elekirochem., 8, 628, 1902), the limiting values being obtained from these results, viz., pco for TlNO, = 127.75 and pco for TlCl= 131'47 at 18' C. The degree of dissociation of thallous hydroxide was calculated from Ostwald's determinations (Lehrbuch der Allg. Chenz., 771). The current flowed from thallium to mercury in the cell. The following tables show the results obtained- 0.78 0.84 0.95 0.96 0.97 0'93 TABLE 111. Thallium in Thallous Nitrate. 0.058 log C. EPC.E.M.F. of Combina- tion. -0.118 - 0.133 - 0.140 -0157 0'724 0'753 0'775 0.788 0'793 0.813 - 0.042 - 0.042 - 0.043 -0.043 ' T1- solrr. 0.1 N.E.= 0-613 volt. - 0'111 - 0.140 - 0.162 - 0.175 - 0.180 - 0'200 ' TI - ' soln. H.N.E. = 0. - 0.388 - 0.417 - 0'439 - 0.452 - 0'457 - 0'477 3.058 log C. 0.065 0.1 18 0.134 0.140 0'157 0'100 E.Pc. - 0.046 - 0.040 - 0.044 - 0*041 - 0.040 - 0.043 E.P/z. ~ - 0.323 - 0 3 17 - 0.321 - 0.318 - 0.317 - 0.320 The mean value of the E.P. from these results is - 0'042 & 0-001 volt for 0'1 normal electrode = + 0.613 volt or - 0'319 volt for H. normal electrode =o. The next series of determinations were made in thallous chloride solutions, and in order to ensure that the values were constant and could be teproduced, several electrodes were used in each experiment and the solution renewed from time to time.The following values were obtained- TABLE IV. Thallium in Thallous Chloride. Concen- trated TIC1. 0'0 I 0.005 0.004 0'002 P O Pee, ' - E.M.F. of Com- bination. 0'773 0.788 0'796 0.8 I 3 -- T I - s o h . 0.1 N.E. = o.613 volt. -~ - 0.160 - 0.183 - 0'175 - 0'200 ~~ TI - soln. H.N.E. = 0. - 0'43 7 - 0.452 - 0'460 - 0'477 _____ EPh. - 0.3 19 - 0'3 19 - 0.320 - 0'320 The mean value from these results is- E.P.(,) = - 0'042 & 0-0004 volt. E.P.(h)= - 0'319 volt. An attempt to determine the E.P. of thallium was then made, using thallium hydroxide as electrolyte and using the calomel electrode saturated164 A REDETERMINATION OF T H E ELECTROLYTIC solution of NH,NO, as connecting liquid, but the values obtained for the E.P. decreased with increasing dilution of the thallous hydroxide solution. This was thought to be due to the NH,NO, not eliminating the liquid potential at the junction, or, at any rate, not completely eliminating it, so an indirect method was employed, using a thallium electrode in thallous nitrate as comparison electrode and calculating the liquid potential difference by Planck's formula. The combination- N N N T1 I 2oo TlNO, 11 2oo TINO, (1 TlOH I T1 gave an E.M.F. of 0.014 volt, the current flowing from the TlNO, to the TlOH in the cell. The liquid potential l'lN0,I)TlOH calculated from Planck's formula amounts to 0.017 volt, whence from the value for T1 I 200TlN0, given above the E.P. can be calculated- N a value fairly close to that obtained above and within the limit of experimental error of this indirect method. The most probable value for the E.P. of thallium from the above measure- ments is - 0.0425 & 0*0005 volt. This agrees with that calculated from the measurements of Neumann (Zeifschr. PlzysiR. Chem., 14, 193, 1894) and Wilsmore and Ostwald (Ibid., 35, 291, 338, 1900). In conclusion, the author expresses his thanks to Professor Donnan and Dr. N. T. M. Wilsmore for their criticism and advice during the conduct of the above measurements.

ISSN:0014-7672    

DOI:10.1039/TF9090400159

出版商:RSC    

年代:1909

数据来源: RSC

摘要:

164 A REDETERMINATION OF T H E ELECTROLYTIC DISC USSZON. Professor R. Abegg (communicated) : May I add the following short remarks to the account of the very valuable and exact redeterminations of Dr. Brislee ?- I. Dr. A. C. Cumming has himself found, and published in the Faraday Society Transactions, that N H,NO, solutions are not applicable for cutting out liquid potentials between basic solutions, in agreement with what Dr. Brislee notices on p. 163. 2. As a member of the " potential committee " of the Deutsche Bunsen- Gesellschaft, I might use this opportunity of expressing the hope that their British colleagues will accept the symbols proposed in 1903 by the Bunsen- Gesellschaft (see Zeitschriff fiir Elektrochenzie, 9, 685), and accepted by the 5th International Congress of Applied Chemistry in Berlin the same year.According to this is the potential counted from the N-hydrogen electrode, E~ from the N-calomel electrode as zero-points, the latter not to be mistaken for the potential numbers formerly used by Osfwald and his pupils calculating from the N-calomel electrode as being + 0.56 volts. Compare also 2. f. Elektrochemie, part 50 (1908), p. 818. The notation of Dr. Brislee does not fully agree with this convention. Dr. G. Senter asked whether temperature differences might not account for the variations found by the author in his results. The Chairman said he was not convinced as to the usefulness of mathematical methods of calculating probable errors when applied to actualPOTENTIALS OF SILVER AND THALLIUM 165 experiments. His experience had been that concordant series of observations were easy to obtain, showing a '( probable error " of absurdly small magni- tude, but that a subsequent attempt to repeat the observations often gave a fresh concordant series of figures differing widely from that which had first been obtained. The mathematical method had the fatal disadvantage that it entirely neglected the sources of error which persisted throughout a series of experiments and were often far more serious than the mere casual variations in the readings.Dr. F. J. Brislee (comwurzicafed reply) : In reply to Dr. Senter, it is quite certain that temperature differences do not account for the variations found by the author. The temperature of the solutions was observed at the time of measurement, and it never deviated from 17" C. by more than o-zo C. The author is in entire agreement with the Chairman in regard to the diffi- culty of getting two series of concordant results, and believes that the repetition of the observations should be the test of reliability of the measurements.

ISSN:0014-7672    

DOI:10.1039/TF9090400164

出版商:RSC    

年代:1909

数据来源: RSC

摘要:

THE HEATS OF COMBUSTION OF ALUMINIUM, CALCIUM, AND MAGNESIUM, BY F. E. WESTON, B.Sc., AND €4. R. ELLIS, B.Sc. ( A Paper read befare the Faraday Society o n Tuesday, December 15, 1908, Dr. T. M. LOWRY in tlzc Cltnir.) The values given by various experimenters for the heats of formation of A1,0,, CaO, and MgO differ very considerably, as is shown in the following table :- I. Mg + 0 = MgO- Heat Evolution. 148,000 calories . . . 143,300 9 , -.- 143,400 ? Y ... 11. Ca + 0 = CaO- 151,goo calories ... 131,500 >, ... 145,000 ,, 131,648 ,, ... 130,900 ,, ... -4 u t hority . Dr. Strauss, Minet, Productiorc of Al, p. 209. Beketoff. Electrochemical a i d Illetallurgical Industry, 6-8. A. Guntz arid H. Rassett, jun., C. R., 1905, 140, 863,864. Thornsen (see L. and B.). Moissan. Dr.Strauss, bliitet, Productioit of Al, p. 209, Thornsen, Guntz used pure Ca and formed pure compounds, from which he deter- mined the heats of solution ; this, of course, introduces complications, leading to various corrections. Forcrand, using pure CaO from Ca(OH), and deter- mining heats of solution, obtained the same value as Guntz (C. R., 1908, 146, 2 17-220). Thornsen used CaO prepared from calcium nitrate, and Forcrand (C. R., 146, 217-220) poiiits out that the low value obtained was probably due to impurities ; consequently all values based on Thomsen’s results will be low (see also J . C. S., Abst. II., 155, 1908). Moissan’s result was obtained by direct determination of the heat of combustion, and therefore is less liable to error than those results obtained by indirect means.111. 2 A1 + 3 0 = Al2O3-- 386,988 calories 380,200 ,, ... Thomsen (see L. and B.). 392,610 ,, ... Richards, Electrochemicar! aiid Metallurgical ... Dr. Strauss, Minet, Production of Al, p, 2 9 . Iiadustry, 6-8. Calculating these results for I gram-atom of 0, they become 128,996, 126,733, and 130,870 respectively. On account of the great divergence in the above numbers, experiments were carried out with a view of ascertaining as far as possible which element has the greatest heat of combustion by examining the reducibility of the oxides of aluminium, calcium, and magnesium by the metals Al, Ca, and Mg respectively. I 66T H E HEATS OF COMBUSTION 167 From the foregoing tables it will be seen that the average value of the heat of combustion of Mg is 145,000 calories, that of Ca is 137,000 calories, and that of A1 is 129,000 calories ; hence from the theory of greatest heat development Mg should reduce CaO and A1,0,, Ca should reduce A1,0, and not MgO, and A1 should not reduce CaO or MgO.Now Goldschmidt (Minet, Production of’ AZ, p. 213) has already shown that A1 can reduce CaO, and Dr. F. M. Perkin has succeeded in reducing A1,0, by Ca ; but it was thought advisable to repeat these experiments in order that a complete comparison could be made between the various reductions when carried out under the same conditions. EXPERIM EMTAL. Series I. The Action oj A1 Powder upon Al,O,, MgO, and CaO respectively. The A1 powder was the same as that used by the authors in their previous work (see Trans. Far.Soc., vol. iv., 1, 1908). The A1,0, was prepared by ignition of pure ammonia alum. The MgO was made by ignition of precipi- tated magnesium carbonate, and was heated for some hours in a muffle at 1,000~ C. and was in an extremely fine state of division. The CaO was obtained by ignition of marble, finely powdered, passed through a 40,000- meshed sieve, and the resulting powder again heated at 1,000~ C. for one hour. This experiment was first carried out by Duboin (C. R., cxxxii., No. 13). The mixture readily reacts with a fuse, forming a blackish- grey product which dissolves in HC1 with the evolution of H ; it contains, probably, a suboxide of aluminium, but very little nitride. The mixture, contained in a Hessian crucible, would not react in the cold, even when a fairly large fuse was used ; when heated over a bunsen burner for two hours (the bottom of the crucible was at a dull red heat), a fuse again brought about no reaction.The crucible was then placed in the muffle at its hottest part ; after two minutes, the crucible then being bright red, action started at the surface and then very rapidly spread throughout the whole mass, giving a somewhat violent reaction. The reaction consisted first of the oxidation of the surface A1 by the air, and the heat produced by this reaction, together with the heat supplied by the furnace, brought about the reduction of the CaO by the Al; this reaction appears to us to be endo- thermic, since it was only possible to bring about the reaction under the conditions stated. The product of the reaction consisted of a hard, black, fused mass, and contained free Ca, calcium aluminate, free A1 (and possibly some Al,O), A1,0,, and CaO ; it also contained a small amount of nitride and carbide.Estimation of Free Ca and A1.-The finely powdered substance was treated with cold water and the gas evolved collected and analysed. From volume of H evolved the percentage of free Ca was calculated ; a small quantity of C,H, and CH, was found present. After treatment with water, HC1 was added and the process repeated. = 8.3 ,, ,, free A1 = 6.1 ,, ,, nitrogen = 0.375 I. A1,0, + 4Al. 2. 2 A1 + 3 CaO. Per cent. free Ca (as nitride) Hence it is evident that A1 powder can reduce CaO at high temperatures, and it is probable that the reaction 2 A1 + 3 CaO z+ 3 Ca + A1,0, is a rever- sible one and that a stage of equilibrium is reached under the above con- ditions.168 THE HEATS OF COMBUSTION OF ALUMINIUM, 3.2 A1 + 3 MgO. The reduction could not be brought about even at the highest temperature attainable (e.g., about 1,100~ C.) ; the only change obtained was a superficial oxidation of the Al; the contents of the crucible were otherwise unacted upon. Hence it appears that the heat of formation of MgO is greater than that of CaO. Series I I . The calcium was first converted into fine shavings in a lathe, and these were then crushed as small as possible in an iron mortar, and that portion which passed through a 120-meshed sieve was used. The mixture, in a Hessian crucible, would not react in the cold even when a moderately large fuse was used, but on heating with a bunsen burner until the bottom of the crucible was at a dull red heat, a violent reaction took place and the larger portion of the contents of the crucible was ejected.From the way in which the reaction took place it is fair to assume that if calcium powder in a sufficiently fine state of division were employed the reduction would take place readily in the cold, and if air were excluded and sufficiently large quantities of mixture employed, fused A1 could be obtained (see also Dr. Perkin, T. F . S., iii., 3, 184). In this reaction the A1 produced probably volatilised, and consequently oxidised under the very high temperature obtained, and this may account for the small percentage of A1 found and also for Dr. Perkin’s failure to obtain globules of metallic Al.The authors intend repeating this experiment under such conditions as to avoid loss, &c. The presence of free A1 was shown by treating the finely powdered product with water until no more gas was evolved, even on heating, and then adding HCl ; a rapid effervescence of H took place, and A1 was found in the solution. The product contained about 8 per cent. of free Al. The mixture, placed in a Hessian crucible, was found to react slowly when the surface was heated with a bunsen burner ; the surface Ca ignited, and the incandescence slowly spread throughout the mass from the lop downwards. On cooling the contents of the crucible were found to be white at the top to a depth of about + inch, and then of a deep yellow. I t was thought at first the action was due to reaction of the Ca with the atmosphere forming a small amount of lime and a large amount of calcium nitride.That this was not so was evident from ( I ) the analysis of the resulting mixture ; (2) that the calcium when heated in a crucible by itself under the same conditions only burnt on the surface and remained unchanged below ; and (3) a mixture of CaO and Ca when treated in the same manner or even when strongly heated by a Bunsen flame only reacted at the surface. I. 3Ca + A1,0,. 2 . Ca + MgO. Analysis of the product gave- N as nitride = 5-79 per cent. Free Ca = 4.16 ,, ,, 7 ) hlg = 0.58 ,, ), I t was not possible to decide whether the nitride was calcium nitride or magnesium nitride or a mixture of both. But from the large percentage of nitride (about 30 per cent.if calcium nitride) it appears that the atmospheric N plays a great part in the reaction observed. Series I I I . I. Mg + CaO. That Mg will readily reduce CaO in the cold is, of course well known, this reaction serving €or the preparation of argon from the air. A mixture made in the proportion Mg + CaO can easily be fired by a lightedCALCIUM, AND MAGNESIUM 169 match, the reaction proceeding steadily throughout the mass. product is of a bright yellow colour, and consists chiefly of calcium nitride. The resulting Analysis of product gave- Ca,N, = 33-65 per cent. Free Ca = 1-29 ,, ,, ,, Mg = 0'47 ,Y 9 , In this reaction, in order to obtain a yield of Ca, air would have to be excluded. 2. 3 Mg + A1,0,. This mixture, in a Hessian crucible, easily reacted with a fuse ; the reaction was somewhat violent, the mass swelling up considerably.The resulting product was quite black. Analysis of product gave- N as nitride = 7'43 per cent. Free A1 = 0'71 ,, ,, 9 , Mg = 3'87 9 , , I In this reaction, as in the case of Ca and A1,0,, it is probable that most of the A1 liberated is converted into nitride and some re-oxidised to A1,0,. Although in these experiments, which were only carried out on a small scale (20 to 50 grams), complications arise partly from the interaction of the various hot metals with the air and with the oxides, it is quite evident that both Mg and Ca reduce A1,0, in the cold and also that Mg easily reduces CaO with the formation of the free metal. It is thus evident that the heat of combustion of Mg is greater than that of Ca, since MgO is not reduced by Al, whilst CaO is at a very high temperature ; however, the heat of combustion of Mg is not much greater than that of Ca, since it is possible to cause Ca to interact with MgO ; also the heats of com- bustion of Mg and Ca are much higher than that of Al. The partial reductions of CaO by A1 and of MgO by Ca-probably endothermic reactions-are analogous to the reduction of B,O, by K and Na respectively. Gay-Lussac prepared B by heating B,O, with I( to a red heat in an iron tube. Now, the heat of combustion of B to B,O, is given as 317,200 (Troost and Hautefeuille-Watts, Dictionary), 314,821 by Roscoe and Schorlemner ; whilst the heat of combustion of K to K,O is given by Beketoff as 97,100 (L. and B.), 86,800 by Rengarde ( J . C. S., Abs. II., 156, 1908)) and 84,800 (ibid.), Hence B,O, + 6 K = 3 K,O f B, becomes thermally (using average values) - [(316,000) - 3 (89,500)I 2.e. = - 47,500 ; i.e., heat must be supplied ; a similar reaction occurs with Na. CHEMICAL LABORATORY, THE POLYTECHNIC, REGENT STREET, W.

ISSN:0014-7672    

DOI:10.1039/TF9090400166

出版商:RSC    

年代:1909

数据来源: RSC

摘要:

CALCIUM, AND MAGNESIUM 169 DISCUSSION. Dr. H. Borns : Though Mr. Weston has stated that he is repeating his experiments in vacuo, I cannot but point out that experiments such as shown -interesting certainly in themselves-cannot decide anything about the uncertain heats of formation of the oxides of Al, Ca, Mg. Mr. Weston is170 THE HEATS OF COMBUSTION OF ALUMINIUM, working in a i r ; he powders his metals, the mixtures possibly stand for hours, and I doubt that his metal is free of oxygen when he starts his experiment. Electrolytic calcium shavings turned on a lathe would hardly be free of oxide. As Mr. Weston starts his combustion with a piece of magnesium wire, the ammonia might possibly be due to magnesium nitride in all these cases. The observation of the influence of CaCO, is interesting with regard to the formation of calcium cyanamide. It is well known that carbide and nitrogen alone will hardly yield any cyanamide ; we want the presence of some other substance to render the process remunerative, and several investigators have mentioned CaCO, among the recommendable additions, of which CaCl, is most generally employed.As that reaction has not sufficiently been explained so far, this observation of Mr. Weston’s deserves attention. If Mr. Weston performs his experiments in air, however, the presence of some carbonate would generally have to be assumed in the Ca and CaO tests. Mr. H. K. Picard said he had assisted Mr. Vautin in 1893 and 1894 in his early experiments, which formed the foundation of what was now known as the “ thermit ” process, and he had occasion to carry out reductions of a large number of oxides and sulphides by means of aluminium.As far as he had gone he had reached the same conclusions as the author. They suc- ceeded, for example, in producing calcium by the reduction of lime with aluminium, but they failed to reduce magnesia. They had not tried the reverse reduction of alumina by calciur-, and he was much interested to learn that this was possible at the temperature of a muffle furnace. An interesting reaction to study would be that between alumina and aluminium sulphide. He saw no reason why such a reaction should not take place, although he had not succeeded in bringing it about himself-he always was troubled by the formation of what he believed to be a suboxide of aluminium.He hoped the authors would try it, because it might prove to be a possible method of manufacturing aluminium. Mr. H. R. Ellis : In the reductions when magnesium was employed it is evident from the following considerations that the nitride obtained was not magnesium nitride. In the first place, when magnesium burns in air very little nitride is produced, and when magnesium reduces the oxide of a metal which does not form a nitride only a very small amount of nitride is produced. Now, magnesium nitride readily burns to magnesium oxide if it is heated, consequently when magnesium burns in air, since a very great heat evolution takes place, any nitride produced would be destroyed, and when magnesium powder reduces an oxide which has only a small heat of formation such would likewise be the case, e.g.- The heat evolution when Mg reduces CaO is not very great- CaO + Mg = MgO + Ca + 3,000 cals ; 145.000 148,000 (Moissan’s value) consequently, we might expect that the nitride is present as magnesium nitride, but the following considerations render this unlikely. When the amount of Mg is in excess the amount of nitride increases butCALCIUM, AND MAGNESIUM 171 is never more than the Ca can account for ; and again, SrO and BaO with a much greater heat evolution on reduction give a larger yield of nitride. Again, when B,O, and SiO, are reduced by Mg the yield of nitride is very small, and in both these cases the heat evolution is not very great- + [BzO, + 3Mg = 3MgO + B,] + 42,600. Every mol. of MgO formed liberates 42,600 cals. 8 [SO, + 2 Mg = 2MgO + Si] + 20,000.Every mol. of MgO formed liberates 20,000 cals. Dr. N. T. M. Wilsmore agreed with previous speakers that the reactions were too complicated to allow of deductions of or from the heats of formation of the simple oxides. Assuming with the authors that lower oxides were formed, which, unless the air' took some notable part in the reactions, seemed to be the only explanation possible of the phenomena described, the heats of formation of these lower oxides would be determining factors in the changes observed. For as all the heats of formation involved were probably high, they might be taken to be roughly proportional to the free energy changes. He looked forward with interest to the promised experiments in absence of air.Answering Dr. Borns, he thought that metallic calcium was nearly as stable as metallic magnesium. Mr. F. E. Weston : In reply to Dr. Borns I should like to state that the magnesium ribbon used in the experiments shown at the meeting was simply a convenient means of starting the reaction, i.e., an easy means of heating up the mixture ; in the experiments described in the Paper the action was started without magnesium, and yet nitrides were formed. That lime, left exposed to air, readily takes up CO, is well known, and hence the authors took precautions that the lime used was free from CO,. It is hardly likely that the calcium contained oxide, since the turnings used were extremely bright ; calcium is much more stable than is usually supposed ; pieces of calcium turned up in a lathe possess a high metallic lustre which remains unaltered for days when kept in an airtight dry bottle. The authors are quite aware of the weakness of the deductions drawn from their experiments, due to the probable interaction of the air ; but since all the experiments were carried out under the same conditions, the same error would affect all to the same extent, and hence the results are compar- able. The only error which is not common to a11 the experiments is the one pointed out by Dr. Wilsmore, namely, the heat changes due to the formation of low oxides; this error undoubtedly affects some of those reactions in which A1 or A1,0, takes part. The authors, however, hope to state in a subsequent Paper what part the atmosphere plays in the reactions described, since experiments are now in progress in which the reactions are being studied in vucuo. 3 16,000 444,000 2 15,490 296,000 VOL. IV-T9

ISSN:0014-7672    

DOI:10.1039/TF9090400169

出版商:RSC    

年代:1909

数据来源: RSC

摘要:

T H E FORMATION OF GRAPHITE BY T H E INTERACTION OF MAGNESIUM POWDER AND CARBONATES. BY H. RUSSELL ELLIS, BSc., F.C.S. ( A Paper read before the Faraday Society on Tuesday, December 15, 1908, Dr. T. M. LOWRY in the Chair.) Winkler, in an investigation of the action of magnesium on the carbonates of the alkaline earths, showed that the carbonates were readily reduced on heating, and amorphous carbon was produced in the reaction.:k The pro- perties of this carbon were not investigated by Winkler or by Maquenne, who also studied the reaction between magnesium and the alkaline earth carbonates. During an investigation of the action of magnesium powder on various carbonates quantities of carbon were produced, and the object of the present Paper is to show that this carbon is a mixture of amorphous carbon and graphite; in some cases it consists almost entirely of graphite.The graphite is non-intumescent. A mixture of a carbonate and magnesium powder does not require heating before action commences. If such a mixture be ignited in one portion by a match or by means of burning magnesium, a reaction similar to Goldschmidt’s occurs, and proceeds throughout the whole mass more or less vigorously ; in many cases CO is evolved, which causes much of the contents of the crucible to be ejected ; if the metal is of comparatively low boiling-point, this distils off and burns and so renders the reaction more violent. Since this reaction was accompanied by very high temperature, and since at high temperatures graphite is the stable form of carbon, it was thought that the carbon would be present in this form.In all cases this was found to be so, although amorphous carbon was also present, the graphite being formed as such or formed from amorphous carbon at the extremely high temperature of the reaction. Detection of’ Graphite i n the Reaction Products. The contents of the crucible, after a reduction had taken place, were boiled for some time with hydrochloric acid, washed, filtered, and dried, and then treated with fuming nitric acid prepared from boiled sulphuric acid, dry potassium nitrate, and dry potassium chlorate. The mixture was kept cold at first and then placed on a water bath and the temperature slowly brought to IOOOC. and the solution evaporated to dryness. If the mixture be made at a high temperature, explosion occurs, and at times the whole mass inflames ; by mixing in the cold and warming as described above, in no case was any explosion produced.Small quantities only were used ; in nearly every case a deep green coloration was produced at the first treat- ment, and a yellow powder was left after dissolving the humic acids and potassium chloride in the dry residue ; if any black powder was left, this was dried and again treated with the oxidising mixture. The complete conversion * Bey. xxiii. 2,642, 1890 ; see also Moissan, Electric Furnace, p. 38. 172THE FORMATION OF GRAPHITE I73 of the carbon into humic and graphitic acid seldom required more than one treatment. The graphitic acid obtained readily exploded on heating, forming pyrographitic acid.Estimation of the Amounts of Graphife and Amorphous Carbon in the Mixture. In Berthelot's memoir on carbon," it was shown that fuming nitric acid attacked amorphous carbon, producing humic acids which were soluble in water and did not attack graphite. This observation has served as the basis of a method of estimating graphite and amorphous carbon in a mixture of the two. The total ash is estimated in the mixture by ignition in the muffle furnace ; the amorphous carbon quickly burns away, whilst the graphite remains for some time and only slowly burns. Another portion (0*1-0*4 gr.) of the mixture is heated on a sand bath in a conical flask with fuming nitric acid for several days until no more brown fumes are evolved, fresh additions of nitric acid being added from time to time.After the oxidation is complete the solution is poured out into an evaporating basin and evaporated to dryness on the water bath to remove all the nitric acid. The residue, consisting of humic acid, graphite, and insoluble mineral matter, is extracted with water and filtered through a tared filter-paper, washed, dried, and weighed. It is then finally burnt in the muffle, the loss of which gives the amount of graphite present. It is essential that the nitric acid be removed, as the mixture is difficult to filter and the filter-paper is attacked. The method was tested by using a mixture of pure graphite+ and amorphous carbon, and it was found to give fairly good results, the graphite being 1-2 per cent. too high, probably owing to the great difficulty of removing traces of moisture. This method is very tedious and requires much time, so it was not con- sidered worth while to estimate any of the mixtures obtained.Graphite was obtained from the carbonates of the alkaline earths--ammonium carbonate, cadmium carbonate, and magnesium carbonate-and by the action of mag- nesium on carbon dioxide. Reduction of Carbonates of Barium, Calcium, and Stronfium, Mixtures were made in the following proportions :- (a) RCO, + Mg, (b) RCO, + 2 Mg, (4 RCO, + 3 Mg, (4 RCO, + 4 Mg, ( e ) RCO, + 3 Mg + 2 C, (f) RCO,+3Mg+6C, and the reaction carried out in open crucibles and in closed iron tins at high temperatures. In all cases the reaction was violent, especially when carbon was initially present, and in all cases graphite was found in the products of the reaction ; in the case of reactions ( c ) and ( d ) very little amorphous carbon was produced.The reaction between barium and strontium carbonates and magnesium powder when carried out in open crucibles always produced the carbide, nitride, and large amounts of cyanide; when the reaction took place in * Ann. de Chim. ef de Phys., ser. 4, xix. 392, 1870. t Trans. Far. SOL, vol. iv., pt. I, p. 70.I74 THE FORMATION OF GRAPHITE closed tins, the carbides only were produced. With calcium carbonate no carbide, nitride, cyanamide, or cyanide was produced when open crucibles were used ; when the reaction took place at high temperatures in closed tins, calcium carbide only was produced. Since results differing from those of Winkler and Maquenne were obtained, a subsequent Paper on this subject will be given.It is well known that most carbides dissolve carbon and deposit them in the form of graphite; that the graphite was not obtained in this way is evident from the fact that the mass had not a fused appearance, but was a very soft powder. Action of Magnesium on Ammonium Carbonate, Magnesium Carbonate, and Cadmium Carbonate. The action was vigorous and produced graphite, but no nitride, cyanide, or carbide. Action of Magnesium on Carbon Dioxide. A rapid stream of dry carbon dioxide was passed over 25 gms. of dry magnesium powder, and this was heated strongly at one point. After a time the magnesium caught fire and burnt rapidly, producing a black and white residue. This powder was treated with hydrochloric acid and the carbon filtered and dried ; about 2-5 gms. were obtained. Only a very slight amount of carbide was formed, no gas being evolved when the powder above was dropped in water ; the smell of acetylene, which is very easily observed from minute quantities of the gas, was, however, noticed. The black powder contained graphite, but only in small quantity. Percentage of ash . . . . . . . . . . . . . . . = 6.64 ,, graphite . . . . . . . . . . . . = 12.09 ,, amorphous carbon . . . . . . . . . = 81.27 The reaction in this case is comparatively slow, and the mixture was spread out over a long tube and consequently no very high temperature was attained ; on the other hand, when the reduction takes place in crucibles, the heat is concentrated, and a much higher temperature results and more of the carbon is changed into graphite. It seems probable that amorphous carbon is first produced by the action of the magnesium on the carbonate, and that more or less of this is converted into graphite owing to the enormous heat evolved. CHEMICAL LABORATORY, THE POLYTECHNIC, REGENT STREET, W.

ISSN:0014-7672    

DOI:10.1039/TF9090400172

出版商:RSC    

年代:1909

数据来源: RSC

摘要:

COLLOIDAL BARIUM SULPHATE. By ERNEST FEILMANN, B.Sc., Ph.D., F.I.C. ( A Paper read before the Favnda-y Socidy, Tuesday, DrcevMber 15, 1908, DK. T. M. LOWRY in the Chair.) PRE LI MI NARY CO RIM u NI CATI OX. Descriptions of the preparation of colloidal barium sulphate are compara- tively rare ; G. Buchner (Clzem. Zfg., 17, 1893,878) describes the precipitation of a gelatinous form of this substance on mixing concentrated aqueous solu- tions of barium acetate and aluminium sulphate, but on dilution or on standing this reverts to the ordinary form. Neuberg and Neumann (Bioclzem. Zed., I, 1906, 166) obtain it by the addition of aqueous sulphuric acid to a methyl alcoholic solution of barium hydroxide, as a jelly which forms a porcelain- like mass on ignition. v. Weinarm (Zeif. f.Claem. at. Ind. d. Kolloide, 1@3,2,89) describes the preparation of numerous salts of the alkali earth and alkali metals, including barium sulphate, in the form of jellies. Some years ago the author of the present communication experimented on the protective action of alkali-casein solutions in the production of colloidal solutions of organic and inorganic substances. For various reasons these were not published at the time ; the most various substances, including many of the heavy metals and their more insoluble salts, were thus obtained in solutions of great strength and somewhat surprising stability, and it was thought that a description of such a solution and its method of preparation might not be without interest, especially in view of the ready accessibility in large quantities, and cheapness, of the materials employed.Twenty-three grms. of commercial casein were dissolved at below 40' C. in water containing 0.09 grms. of sodium hydroxide and diluted to 230 C.C. To this solution was added a solution of 14.2 grms. (0.1 mol.) of anhydrous sodium sulphate in water to zoo C.C. To this mixed solution a solution of 245 grms. (0.1 mol.) of crystallised barium chloride in water to IOO C.C. was then slowly added at 2oOC. with vigorous stirring. A thick, creamy, and almost homogeneous liquid was so obtained, though there was a certain amount of sedimentation on standing. I t was precipitated by adding 12 C.C. of glacial acetic acid, which caused the formation of a white, curd-like pre- cipitate, allowed to stand, and the precipitate washed by repeated decantation till free from acid reaction.It was then stirred with 400 C.C. of water and 2 C.C. of 15 per cent. sodium hydroxide solution at 4oOC. for three hours, filtered through an open paper filter and evaporated to dryness on the water- bath. Before evaporation the liquid had much the appearance of ordinary milk by reflected light ; by transmitted light it was transparent in thin layers, showing the peculiar reddish-yellow tinge so characteristic of many colloidal solutions. The solid obtained by evaporation was translucent, and of a horny appearance and texture. It was quite soluble in water to a similar solution to the one just described, but on standing it appears to undergo some slight change which renders it necessary to employ a very little dilute alkali in order to obtain a perfect colloidal solution.On ignition 56.7 per cent. of barium 175176 COLLOIDAL BARIUM SULPHATE sulphate was left. After keeping for six months 2 grms. of the substance dissolved in 70 C.C. of water and 0.5 C.C. of normal sodium hydroxide solution, leaving only a negligible amount of residue. Under a magnification of 200 diameters particles in rapid movement can just be made out under the microscope, and the appearance of the liquid under magnification is much the same as that of the colloidal solutions of tungsten prepared by Kuzel's method. The solutions are very stable ; only a slight amount of sedimenta- tion takes place after standing for many hours, and this is very possibly due to the action of the air, which appears in some way to affect the stability of the solution.The protective action of organic and other colloids in cases such as the one just considered is one that has not yet been adequately explained, though the phenomenon has been known for a very long time. Some light is thrown on this question by the experiments of Michaelis and Pincussohn (Biologische Zeil., 2, 25 I), who experimented with colloidal suspensions of indophenol and mastix. These were more stable than suspensions of indophenol alone, and microscopic and optical examination of the mixture led to the conclusion that each particle of indophenol was surrounded by a layer of mastix. There is reason to suppose that the colloidal barium sulphate and other preparations of the present author are in a very similar condition, and that each particle of inorganic material is surrounded by a layer of casein ; they behave towards acids and alkalies in a very similar manner to casein itself, being precipitated by the former from solution and redissolved by the latter, and this process can apparently be carried on alternately almost ad Zibilzun. These facts suggest that the surface in actual contact with the liquid media consists of casein only, and that it is this surface which conditions the stability of the solutions obtained. This view is also in accordance with the fact that a certain minimum quantity of casein is in each case necessary to hold the inorganic substance in solution.

ISSN:0014-7672    

DOI:10.1039/TF9090400175

出版商:RSC    

年代:1909

数据来源: RSC

摘要:

COLLOIDAL BARIUM SULPHATE DISC USSI 0 N . Dr. V. H. Veley (cornmunicafed) : The preliminary communication of Dr. Feilmann on colloidal barium sulphate presents various interesting points, and it is hoped that the author will continue his investigations upon this subject, using other colloidal menstrua, such as gelatine, agar-agar jelly, various albumens, &c. the appearance of ordinary inilk by reflected light,” it would seem that they were of the nature of emulsions (cf. Ramsden, Proc. Roy. Soc., 1903, 72, 156, and Pickering, Juum. Chenz. SOC. Tram., 1907, ZOOI), and their general behaviour is very similar to that of those described by the last writer. It would be a further point of interest to examine more fully the size of the barium sulphate particles under a microscope of high power, preferably using a hot stage according to usual bacteriological methods.Some time ago the writer had occasion to examine the size of barium sulphate particles precipitated from barium chloride and a sulphate without addition of free acid (which makes the precipitate more crystalline and coarser grained) ; the powder was subsequently ignited and sifted through fine muslin (Journ. Chenz. Soc. Trans., 1903, 749). The average diameter of the particles was found to be I-2p (=O-OOI to 0.002 mm.), and their edges were perfectly sharp and well defined. It appears probable that particles of barium sulphate in the colloidal condition would be larger and without definite edges. From the description given of the liquids as havingCOLLOIDAL BARIUM SULPHATE I77 By arranging matters so that the organic menstruum could be slowly melted off, the appearance of the barium sulphate under the microscope could be ascertained, and the temperature at which the salt passes from a colloidal to a crystalline condition could be determined within a few degrees of temperature. The conditions of precipitation by acids, and resolution by alkalies, could similarly be studied, and thus more information obtained as to the relative size of the particles according to the nature of acid or alkali used and their respective concentrations.Another line of investigation might be to determine the size and form of barium sulphate particles by a repetition of the experiments of Lodge (Brit. Assoc. Report, 1884, 309), varying the conditions of tube-diameter, potential gradient, and current, as it appears possible that every variety from well- defined crystals to amorphous masses of barium sulphate could thus be obtained.Professor W. W. Haldane Gee (conznrunicated): Since the work of Graham in 1861, colloidal chemistry has, chiefly by the efforts of German in- vestigators, been much extended in its scope. It now promises to become a very important branch of physical chemistry, and to be fertile in technical appli- cations. Dr. Feilmann’s communication is to be welcomed more especially as a contribution on the protective action of organic colloids in producing stable colloidal solutions. This action, which has been much studied by Zsigmondy and others, has not, as the author states, been completely explained.The particles seem to be coated by the protector. This is in accordance with the fact shown by Quincke, that if we have three immiscible liquids A, B, and C, then B will tend to form a layer separating the other two if- T A B + TBC < TAC, where T A B is the surface tension between A and B, whilst the other quantities are corresponding values for the other boundaries. It has been suggested that the coating may form a permanent electric ‘‘ double layer,” with the effect of preventing the neutralisation of the nucleus by ions of an opposite charge. I t would be interesting to extend the method described to the salts of calcium. Using albumen and like substances, much might be learnt about the method in which dentine and other hard substances are produced in an iin a1 tissues .Dr. G. Senter said that although the title of the paper was “Colloidal Barium Sulphate,” he understood from the author’s statement that he regarded the system as made up of particles of ordinary barium sulphate surrounded by a skin of casein, and he himself quite agreed with the author that the results were best accounted for in this way. This view might perhaps be tested by grinding the coagulated precipitate with finely-divided sand, with the object of breaking up the particles, and then examining as regards solubility. An ultra-microscopic examination of the particles should also prove of interest. The proportion of casein was rather large; was it not possible to obtain complete solution with relatively less casein ? Dr. H. Borns : The remarks I intended to make concern the point already alluded to by Dr.Senter. In the experiments of Neuberg and Rewald on colloidal salts of the alkaline earths, there is no colloidal envelope for the particles of an organic substance. They dissolve barium oxide in aqueous methyl-alcohol, pass CO, through the solution, and obtain a gelatinous solu- tion which, as long as some methyl alcohol is left in it, will completely redissolve in water. With sulphuric acid this solution gives colloidal bariumCOLLOIDAL BARIUM SULPHATE sulphate, but I do not remember whether that sulphate will completely redissolve likewise. As regards a possible application of those colloids, I might mention the street photometer of A. Wright, shown recently at the exhibition of scientific apparatus arranged by the Physical Society, A copper sphere is coated inside with barium sulphate to provide a light-reflecting surface ; that coating might more easily and uniformly be produced with Dr.Feilmann’s colloidal sulphate than with ordinary sulphates, I should think. Dr. C. H. Desch suggested that as alkalies peptonised the coagulated colloid, this was rather against the existence of the BaSO, itself in a colloidal condition, since colloidal BaSO,, coagulated by acid, would not be taken up again by alkali. Dr. Feiimann’s suggestion that small particles of BaSO, are merely enclosed in colloidal envelopes of casein was probably right. He called attention to the work of von Weimarn, who has shown (Zeiischr. fiir Kolloide and Journ. Russ. Phys.Chern. SOC.) that the so-called “ amorphous ” precipitates of HaSO,, &c., are probably minutely crystalline, and has photographed the stages of their growth. Dr. Feilmann ( i i t reply) was much obliged to Dr. Veley for his very valuable suggestions, but was afraid that to carry them out at all adequately would imply a very considerable amount of work. Pickering’s results were, in the author’s opinion, in no way analogous to his; Pickering’s emulsions were very much coarser in texture and he obtained a system of oil globules surrounded by a layer of powdered solid, so that there was a liquid nucleus surrounded by a solid envelope, whereas the properties of the liquids described by the author all pointed to the presence of solid nuclei surrounded by a liquid, or semi-liquid, envelope.Also, of course, Pickering’s emulsions were destroyed completely by the application of heat, whilst the author’s liquids could be evaporated to dryness, and the solid so produced could be re-dissolved to a liquid of the same character as before. In other characteristics, also, such as viscosity, the preparations of Pickering and of the author showed very different characteristics. In reply to Dr. Senter, he had referred to the system as a whole as a ‘‘ colloidal solution” in deference to the generally accepted nomenclature ; from the work of v. M’eimarn it was doubtful whether the term “ colloidal,’’ as applied to a minute particle, had any very definite meaning. He had hoped to get some insight into the nature of the protective action of colloids--this appeared to him to be the really interesting question. He would certainly try grinding up with sand, but was not very sanguine as to the result. The quantity of casein used was the minimum with which he had succeeded in preparing stable preparations of barium sulphate. It took a very exceptionally large amount, probably owing to its very crystalline character. Other inorganic solids, such as silver, could be con- verted into pseudo-soluble preparations containing very much smaller proportions of casein, and the author hoped to publish further results on the subject shortly. As a result of his work, he had come to the general conclusion that the greater the tendency of a substance to crystallise, the more of the protective colloid was needed in order to produce a stable hydrosol. He was much obliged to Dr. Borns for his suggested application of the material ; he hoped to extend his investigations to other protective colloids, as suggested by Professor Haldane Gee and Dr. Veley.

ISSN:0014-7672    

DOI:10.1039/TF9090400176

出版商:RSC    

年代:1909

数据来源: RSC

摘要:

APPLICATION O F ELECTROLYTIC CHLORINE TO SEWAGE PURIFICATION AND DEODORISATION BY THE “OXYCHLORIDES ” PROCESS. By SAMUEL RIDEAL, D.Sc. (Lond,), F.I.C., F.C.S. ( A Paper read before the Faraday Society, Tuesday, February 9, 1909, Dr. V. H. VELEY, F.R.S., in the Chair.) SECTION I.-THE APPARATUS. The first electrolyser of the Oxychlorides Company was a semicircular wooden trough lined with graphite on a metallic base which formed the anode. In this trough revolved the cathode, a wooden cylinder covered with lead, attached to a copper ring with a rubbing contact-block. The two poles were connected with the terminals of a low-potential dynamo. The later form of electrolyser is of larger size, and the cross-section is completely circular except as to about 4 in. at the top, which is left open for the free escape of gases evolved in the electrolysis.Both machines occupy the same floor space, and have respectively an electric surface of 30 and of 57-5 square feet. The graphite blocks are of the highest grade Acheson make, treated at the Oxychloride Works by a process which adds materially to their life as electrodes, so that a small machine has been in constant use at a distillery for four years without repairs, and is still carrying its full current. This form of electrolyser gave satisfactory results in some directions, but in applying the process to bleaching of cotton, linen, and paper-making material, it was found that the salt consumption was too high, and that loss occurred in circulating the solutions from electrolysers to bleaching vessels and vice versLi.To meet these conditions it was decided to produce a more concentrated solution by using a diaphragm, and after some months of experiment the Company have succeeded in obtaining a diaphragm which has a commercially long life, gives a good current efficiency, and does not set up too much resistance. This diaphragm machine shows results in the following figures supplied by the Oxychloride Company. With a current of 1,500 amperes at 4’6 to 5 volts, according to density of salt solutions, the current efficiency in available chlorine at end of continuous daily runs of eight and ten hours was 68 per cent. It is, of course, essential that the salt solutions be purified. In the earlier runs of this apparatus it was found that chlorate was formed in the anode chamber equivalent to 9‘5 to 12.5 per cent.of the current efficiency in av. C1, but further experience showed that this formation of chlorate could be minimised, and also that the salt consumption could be still further reduced. It appears that ultimately the chlorate in the anode chamber will not exceed the equivalent of 3 per cent of the current efficiency in av. C1, with, of course, a corresponding increase in percentage of available chlorine, and without raising the voltage. I may mention that in tests of the The daily salt consumption was 90 lbs. I79180 SEWAGE PURIFICATION AND DEODORISATION liquid as used at Guildford I have found over 8 per cent. of av. C1 and have not found chlorate. The main advantages resulting from this diaphragm machine are- Elimination of the circulation of electrolyte.Reduced quantity of electrolyte handled daily, thereby reducing cost of A less salt consumption. Increased production per horse-power hour. Reduced cost per kilogramme of av. C1, labour and power. The degree of concentration in av. C1 is purely a matter of manipulation, and up to 15 per cent. av. C1 calls for no special care. Whilst the diaphragm machine was essentially developed for bleaching and allied trades, it is equally applicable to sewage treatment, and a merit of the new solution is that it is very stable, and therefore particularly suitable for sending out in carboys to small sewage works where the quantity of effluent is small, and would not call for the installation of an electrolyser. The chlorides in the treated liquid are only very slightly increased by the use of-this new form of the solution. SECTION 2.-APPLICATION OF CHLORISE TO SEWAGE.My first series of experiments was made in 1904, and demonstrated the efficiency of chlorine so derived in acting as an oxidiser, deodoriser, and germicide on sewages and effluents. The publication of my results in the JournnE of fhe R . Sun. Insfitule (vol. xxvi., No. 7, 1905) and elsewhere* was succeeded by the investigations on similar lines of Phelps and Carpenter on the sterilisation of sewage filter effluents (U.S. Technology Quarterly, xix, 47, pp. 382-403, Dec., 1906), and of Kellerman, Pratt, and Kimberley for the Ohio State Board (U.S. Dept. of Agriculture, Bull., 115, Oct., 1907). In these cases a solution of chloride of lime was used as being more locally available, but Phelps remarks that ‘‘ the use of gaseous chlorine manufactured at the dis- posal works by electrolytic methods would, in the case of larger works, considerably reduce the cost.The process as thus outlined is entirely feasible, and in the case of large works it is cheaper than sand filtration.” Kellermann, Pratt, and Kimberley found similar results when working with calcium hypochlorite on sewages in America, and came to the conclusion that with a hypochlorite complete disinfection was effected in partially purified sewage at a lower cost than with copper sulphate. In their experi- ments they have determined the ratio of chlorine immediately absorbed and the oxygen consumed figure obtained by five minutes’ boiling with acid per- manganate, and find that there is no relation between these two figures.They confirm my 1904 work in showing that the chlorine consumption of sewage is influenced by the concentration of the chlorine, since when 10 parts per IOO,OOO of chlorine is added more than twice the amount of chlorine is absorbed as when 5 parts per 100,ooo only are added; this points to the advantage, when using chlorine for sterilising, of adding it in small quantities gradually, in order that its sterilising power may not be interfered with or absorbed by the organic matter present. The American writers have compared the ratio of the amount of hypo- chlorite used by the organic matter in the co2d in five minutes with the amount of permanganate with five minutes’ boiling, whereas in my experiments I have compared the amount of chlorine required for producing sterility, irrespective * See also Sanitary Record, Oct.6, 1904 ; Rideal, Sewage and Bacterial Purification of Sewage, 1906, p. 186.General View of " Oxychlorides " Electrolyser.BY THE ‘( OXYCHLORIDES” PROCESS 181 of the organic matter present, with the amount destroyed by the permanganate in the cold. It is evident that the organic matter in different sewages in various stages of purification must be amenable in a varying extent to oxidising agents, but as it is in all cases a time reaction the essential point to observe in investigations of this character is not so much the quantity of chlorine absorbed by the organic and other reducing matters, as whether there is sufficient free or available chlorine present in the liquid for a length of time sufficient to cause necessary sterilisation.It is further obvious that, in ensuring sufficient available chlorine for sterilising work, some of the oxidising agent at the same time must be destroyed by the organic matter, This will be greater in strong sewages than in weak sewages, and will be greater in liquids in which the organic matter is in a condition to be amenable to chlorine oxidation. Sulphuretted hydrogen present in a sewage must influence very considerably the amount of chlorine required to be added, as it reacts immediately on free chlorine or hypochlorites. Nitrites, ammonia, and urea all form compounds with the av. C1 which have more or less sterilising effect.I found that the oxygen consumed from permanganate in five minutes in the cold as measuring the sulphuretted hydrogen and other easily oxidisable substances was a guide to the amount of chlorine likely to be quickly absorbed by the liquid, and in the earlier experiments found that when a maximum amount equal to 1’7 times this figure was used, the availability of the chlorine was sufficiently prolonged to ensure the desired sterilising effect. We have, however, frequently since obtained excellent results when the amount of chlorine was only equal to the amount of oxygen consumed from permanganate in five minutes in the cold. In November, 1904, I was consulted in regard to a town sewage where complaints had been made of aerial nuisance, and oxychloride treatment was proposed as a remedy.The Council had already had experience with hypo- chlorites, and were then using permanganate, so that they were fully aware of the advantages of employing an oxidising agent for deodorising their sewage, They visited Guildford to examine the oxychloride process, and approved of it. The Local Government Board, however, suggested further experim ts. and expressed a doubt as to whether subsequent bacterial purification m ht possibly be interfered with. These points I had previously elucidated, as shown by the following instances from my earlier experiments :- Experiment 59.-Rapidity with which oxychloride effects deodorisation. One part per ~oo,ooo of av. C1 was added at the inlet of the branch pipe through which the septic effluent was passing at the rate of 1,000 gallons per hour.This pipe was 40 ft. long and 2& in. bore. At the commencement a strong foul odour existed at the receiving chamber. The liquid was deodorised in passing through the pipe, and in less than an hour the whole tank had lost its odour. Experiment 33.-A septic effluent giving a five minutes’ oxygen figure of 3-41 parts per IOO,OOO was treated with z parts of av. C1. This quantity reduced the five minutes’ oxygen figure to 2-37 parts, was sufficient to deodorise the liquid, but not to destroy the organisms, therefore would not interfere with subsequent bacterial purification. Other experiments made by Mr. Fieldhouse under my direction at Guild- ford in 1905 proved that oxychloride when used in small quantities sufficient to destroy foul odours, as from a septic effluent, at the same time acts as a powerful oxidiser, as seen in the five minutes’ oxygen figure, and consequently must reduce the work required of the filters.The Local Government Board, nevertheless, as is their usual practice with182 SEWAGE PURIFICATION AND DEODORISATION . . . . . . Untreated septic liquor Treated septic liquor . . . . . . No. I filter . . . . . . . . . . . . No. 2 filter . . . . . . . . . . . . new work, were not satisfied without further prolonged experiments, and referred the matter to the Royal Commission on Sewage, whose experts recommended the construction of two exactly similar filters, one to receive oxychloride-treated septic liquor and the other untreated liquor.These were 9 ft. diameter, 6 ft. deep, and filled with broken coke of I to 3 in. size. Second Series of Oxychlol-ide ExlSerimertfs. - The untreated liquor was conveyed from the main carrier of the Guildford Corporation septic tank to a small tank used as a constant head supply, thence to a second small tank furnished with a float to register the quantity flowing over two V notches, which led to two separate compartments containing syphons discharging periodically by sprinklers on to the beds. Oxychloride was stored in a vat connected with a constant level arrangement by which the solution was supplied in the amount required. The quantity of septic liquor passing on to the sprinkler filters was 50 gallons per cubic yard per 24 hours. The maturing of the beds commenced on August 8th, and continued until November 22, 1906, when the Royal Commission’s chemist, Mr.A. C. Carter, began to make investigations commencing with 25 per cent. of the five minutes’ oxygen figure. It is necessary to know the five minutes’ oxygen figure as in the Guildford experiments to get comparative results, by other methods factors must be introduced. Guildford Method.--ro C.C. of sewage is placed in a bottle of about 250 C.C. capacity and 90 C.C. of tap water added, to which also is added 10 C.C. of 10 per cent. sulphuric acid and 25 C.C. permanganate of strength O’OOOI grms. 0 per c.c., using thio = 0~00008 grms. 0 per c.c., adding a crystal of potassium iodide after five minutes, doing a blank with 100 C.C. of tap water at the same time. This method gives a lower oxygen figure than by other methods, except direct titration by permanganate.Having obtained the oxygen consumed in five minutes in parts per IOO,OOO, 25 per cent. of this figure was added of av. C1; e.g., if five minutes’ oxygen figure = 2.0 parts per IOO,OOO, 0.5 parts of av. C1 = 25 per cent. of that figure. Therefore, using oxychloride at 0.2 per cent. av. C1,it requires 2.5 gallons of oxychloride to 1,000 gallons septic effluent or 5 gallons of a 0.1 per cent. solution. 0.1 per cent. solution was used during the earlier experiments as being more suitable for small flows, because of the gauging and the mixing. No. I bed received untreated sewage, No. 2 bed received treated sewage, and 25 per cent. treatment was carried out from November 22nd to December 6, 1906.Chemical analyses showed that oxidation by bacteria and other agencies had not in the least been interfered with, there being a slight improvement in the effluent from the treated sewage as compared with the untreated. The five minutes’ oxygen consumed figure and nitric nitrogen were as follows-- 1 - 2-07 , - 1 - I . 0.51 j 2-20 I I o.50 j 2-02 I I j Average of Last Week prior , to Treatment. ~ Five Mins.’ 1 Nitric Oxygen. Nitrogen. Average of Last Week of 25 per cent. Treatment. Five hlins.’ 1 Nitric Oxygen. Nitrogen. I 0.58 1 2.0BY THE “ OXYCHLORIDES” PROCESS 183 Therefore, prior to treatment the nitrates were higher in No. I filter, but at end of treatment No. 2 filter, which received the treated effluent, was slightly higher in nitrates.The av. C1 was sufficient to itearly eliminate aerial nuisance around the filter, and, in my opinion, would be sufficient for practical purposes, although sulphuretted hydrogen was not entirely destroyed when tested with lead papers fixed on the arms of the sprinkler. Four-fifths of the H,S was removed (see Report of Royal Commissiora 011 Sewage, 1908, On December 6th we raised the amount of av. C1 to 35 per cent. of the five minutes’ oxygen figure, and with this amount it was agreed (I) that complete deodorisation was attained on No. 2 bed (all traces of H,S being absent to lead papers placed on arms of sprinkler), and ( 2 ) that further oxidation had not been interfered with. On December 10th it was decided to further increase the av. Cl. to 50 per cent.of the five minutes’ oxygen figure,it being suggested that if aman carelessly put on an extra amount of oxychloride it might interfere with further purifi- cation. This augmentation was continued from December 11th to January ~ 3 t h , and unexpectedly proved of decided advantage to the filter, there being an increase in oxidation, also a reduction of grey gelatinous growths ; whilst the B. coli, which in No. I effluent were at least 50,000 per c.c., were less than 10,000 per C.C. in No. 2 . The following figures show averages of the first and the last week of this treatment, and prove that even if a larger quantity of oxychloride is added than that required to deodorise, oxidation is not interfered with- pp. 19s-201). Untreated septic liquor ... Treated septic liquor ...No. I filter . . . . . . No. 2 filter . . . . . . First Week. Five hlins.’ Oxygen. 2’39 2-29 0.78 0.80 Nitric Nitrogen. - - 1’10 I ‘07 Last Week. Five Mins.’ Oxygen. 2.70 2’55 0’5 5 0’52 Nitric Nitrogen. .___ - - 3’0 3’53 The next point was to ascertain how much av. C1 would really be required to retard nitrification, and to this end the dose was increased to IOO per cent., four times the quantity that would be used in most cases for purposes of deodorising. This amount is sufficient to reduce coli from hundreds of thousands to less than I per C.C. when given necessary contact. This experimental plant was not designed for this purpose, so that av. C1 was passing on to the filters, dissolving up the growths and cleansing the media, which at first tended to produce an indifferent effluent ; yet under these conditions the filter speedily recovered, and gradually surpassed No.I (receiving untreated liquor), as the following figures prove (January 14 to May 13, 1907)-184 SEWAGE PURIFICATION AND DEODORISATION Untreated septic liquor ... ... Treated septic liquor ... ... Effluent from No. I filter ... Effluent from No. 2 filter ... ______~_ - Early Part of Treatment. -___ Fire Mins.' Oxygen. 2.56 2.07 0'94 1.08 ~ _ _ _ _ Nitric Nitrogen. ______ - - Last Week in April. Fire hiins.' Oxygen. 3'85 3.01 1'22 1.17 Nitric Nitrogen. - - 3'39 4.10 -____ These results show how much No. 2 was behind and how quickly it has surpassed No. I. The 100 per cent. addition broke up the growths that were on the surface of No.2 bed, whilst No. I had a considerable growth which would either have to pass away as solids or else block up gradually the spaces betwixt the media. The pipes conveying the treated liquor were clean and free from growths, whilst those to No. I had to be repeatedly cleaned out to allow the syphons to work properly. Bacteriological tests showed coli to be reduced from 200,ooo to nil per C.C. in the liquor, and that when No. I efluent contained 20,000 per C.C. No. 2 had only 2,000; and these came from the growths passing away as solids in suspension. Thus the IOO per cent. treat- ment surpassed expectations in that the filter improved even with such a large amount of av. C1 as was passing on to it. Consequently a still further increase to 200 per cent. of the five minutes' oxygen figure was tried from May 14th to May 3oth, with the following average results for the last three tests- __-.____ Untreated septic liquor Treated septic liquor No.I effluent ... ... No. 2 effluent ... ... Five, Mins. Oxygen. 3'40 2-36 0.80 0.70 Four Hours' Oxygen. 10.16 994 3 '49 2.96 Nitrogen. As Nitrite. - - 0.09 0.07 As Nitrate. - - 4'85 4'95 By the analyses, even with this excessive amount of av. C1, the filter was only temporarily interfered with and finally was every bit as good as No. I , receiving untreated liquor-slightly better if anything. B. coli were absent from 2 C.C. in the treated liquor, and spores of enteritidis were less than 10 per C.C. More coli and spores were passing away with the effluent, €3. coli 2,000 per c.c., and enteritidis spores 20 per C.C.in No. 2, whilst from No. I the coli present in the effluent = 10,000 and enteritidis spores zoo per C.C. Here again the B. coli in effluent from No. 2 filter were apparently traceable to growths dissolved from the media and passing away as suspended solids. Seeing that zoo per cent. did not stop the process of oxidation, it was decided to again increase the amount of av. C1, and from May 31st to June 10th the amount added was 500 per cent. This now produced an appreciable difference in the amount of oxidised nitrogen in No. z effluent.BY T H E L i OXYCHLORIDES" PROCESS 185 The average of the analysis of the last three days' samples was- No. 2 effluent ... 1 Five Mins.' Oxygen. 0.87 r Shaken. I ' Five Mins.' . . . . . . I ~ Oxygen.Untreated 1 4.88 No. I . . . . . . . . . I 0.89 . . . . . . Paper-filtered 0.46 Paper-filtered . . . . . . I 1-40 . . . . . . ' 1'42 '*. i No. z I Untreated liquor 1 3-04 Nitrogen. As Nitrite. As Nitrate. ~ _ _ _ _____ i - - I - Four Hours' Oxygen. I 1-84 4-32 1 - 2.08 0.30 4'33 8-64 6.88 11 il 0.25 - - 1 Treated liquor ... I 1-18 Settled. Four Hours' Oxygen. 1 Nitric 1 Nitrogen. Shaken. 8*8j 7'78 3'57 3'94 Settled. 1 __._ ____ 2'45 3'64 I 5'37 Opacity. Settled. - - over 12" over I 2 ' I Bacteriological tests- Untreated. Treated. No. I Effluent. No. z Effluent. B. coli . . . . . . 500,000 per Absent from 5,000 per C.C. Absent from Enteritidis spores 500 per C.C. I per C.C. 500 per C.C. 20 per C.C. C.C. 5 C.C. I C.C. Av. C1 was absent from the effluent. During the last week of treatment No.2 filter showed distinct signs of recovery, nitrates rising from 2-90 to 3.75 parts per IOO,OOO, and it was decided to increase the av. C1 to 1,000 per cent. of the five minutes' oxygen figure from June Irth, and as the septic tank effluent was very foul at times, the amount of av. C1 was often 50 parts per IOO,OOO. This treatment was continued up to June 16th, and from June 27th to July 12th. On July 2nd av. C1 began to pass away with the effluent, and after this No. 2 filter went back rapidly, as seen by the figures below of special samples collected on July 11, 1907. Bacteriological tests of the liquor passing on to No. 2 filter showed B. coli to be absent from 5 C.C. probably more, and enteritidis spores absent from 0'5 c.c., whilst in the effluent both coli and spores were absent from Recovery of No.2 Filfer after 1,000 per ceizf. Treatrnenf.-To see how long No. 2 would take to become normal again, the filters were rested eleven days, when again untreated liquor was passed on to both. After three weeks No. 2 2 C.C.186 SEWAGE PURIFICATION AND DEODORISATION surpassed No. I in oxidised nitrogen, as on August 12th it contained 3.82 parts, as against 3-51 parts in No. 1. This difference was maintained up to the time of dismantling the filters (for a fresh series of experiments) on August 29, 1907, when it was found that the filtering medium in No. 2 was considerably cleaner than that in No. I. These experiments from August, 1906, to August, 1907, were carried out by Mr. Fieldhouse under the superintendence of the Royal Commission on Sewage, who published an extract of the results in their Report of 1908, p.200, and summarised their conclusions to the following effect- I. That “ 35 per cent. to 50 per cent. treatment ” with the hypochlorite solution is sufficient to do away with the offensive sulphuretted hydrogen smell of the Guildford septic tank liquor, leaving in its place a comparatively inoffensive odour of spent bleach and fresh sewage. 2. That the addition of this quantity of “oxychloride ” in no way interferes with the efficiency of the filter. Indeed, it rather adds to it, by helping to keep down an excess of grey growths on the top of the filter. 3. That a very much larger quantity of oxychloride may be added, without any danger to the filter.In other words, there is a large margin of safety, so far as regards the accidental addition of too much oxychloride. Somewhere between “ 200 per cent. and 500 per cent. treatment” appears to be the limit of the quantity of oxychloride which can be added with safety. A number of bacteriological examinations were made in connection with the experiments, and we ascertained that a dose of oxychloride, more than sufficient to remove sulphuretted hydrogen smell and kill B. coli in the liquid to be treated, may be used without prejudice to the purifying ability of a mature percolating filter. The subsequent paragraph (281, p. 201) of the Report is somewhat surprising : “ Our own experiments have shown that the offensive character of Dorking septic tank liquor can be destroyed by the addition to it of three grains of lime per gallon,” as it is general experience that lime alone tends to rather increase than diminish nuisance.Beyond the above conclusions, in my opinion the following are equally obvious- (a) That sewage from hospitals or infirmaries may be freed from dangerous organisms by the use of oxychloride before passing on to sprinkler or contact beds, or land, or before entering into ordinary sewage systems, without in any way interfering with the usual methods of purification. (b) That when beds are clogged up with growths internally or on the surface, these can be dissolved and washed through speedily, and that such beds quickly recover normal conditions. It is evident if growths are kept under control by at least occasional doses of oxychloride, that sludging up will be considerably delayed, and that the excessive quantities of solids which pass away-particularly when heavy growths are breaking up-will be avoided.BY THE 6‘OXYCHLORIDES” PROCESS Five Mins.’ Oxygen._ _ _ _ _ ~ 2.40 2-16 187 Four Hours’ Oxygen. _~___ 7’74 7.66 THIRD SERIES OF OXYCHLORIDE EXPERIMENTS, 1907 AND 1908. Nitrous Nitrogen. The next point was to ascertain whether a 25 to 35 per cent treatment might be safely applied duriizg the maturing of a filter, and how far it would retard or prevent growths and assist in the work of oxidation. It was also thought probable that oxychloride might enable fine media to be employed for sprinkler beds in cases where at present excessive growths prevent their adoption.On the other hand, we wished to determine whether oxychloride used in such quantities for deodorising on a continued and pro- longed run might disturb the balance of organic life and effect some alteration in the results. For these objects the construction of the filters was slightly modified. The media were removed, and each filter divided into two compartments, having separate exits. Also galvanised iron channels were fixed 2’ 3’’ and 4’ from the surface in each compartment for the purpose of collecting special samples. The filters had one half filled with coarse medium as before (new coke from I” to 3”), the other half with new coke which had passed through a I” sieve and remained on a g’’ square mesh, so that each sprinkler had now one-half rough and .one-half fine medium 6‘ in depth.The plant in other respects was the same as previously. At the commencement, on September 12, 1907, the rate of flow was 50 gallons per cubic yard on each filter, No. 2 receiving the treated liquor as before, the amount of av. C1 for the first two weeks being 25 per cent. of the five minutes’ oxygen figure. Afterwards the dose was increased to 35 per cent. of the five minutes’ oxygen figure, and continued at this throughout the whole test. The first run was from September 12 to October 2, I907 (11 days), when the Corpcxation shut down the septic tank (which had been in use 12 months) for cleaning. On the last day of this short run the average results were as follows (see also Curve IA)- Nitric Nitrogen. Untreated liquor ...Treated liquor . . . No. I rough ... ... No. I fine ... ... No. 2 rough ... ... No. 2 fine ... ... 0’00 0’20 0.90 0’10 0’00 0-5 I 0’34 2-26 I Opacity. The above figures indicate a rapid rate in the maturing of filter No. 2, receiving the treated liquor ; no growths had formed on either filter, and the material was clean. The experiments were not recommenced until the tank had been filled 2 weeks to allow the liquor to become somewhat septicised, the beds resting for 15 weeks. On January 17, 1908, after the sprinklers had run for a short time, special samples were collected, and oxidised nitrogen was found to be absent from the effluents, so that the beds had to be matured again. The following were the progressive results of the dates mentioned. The beds were first worked intermittently for 6 or 12 hours a day, or on alternate days but on February 22nd 24 hours’ continuous runnings were comnienced.188 SEWAGE PURIFICATION AND DEODORISATIONBY THE OXYCHLORIDES ” PROCESS190 SEWAGE PURIFICATION AND DEODORISATION NO.I fine" ... Jan. 22nd Untreated liquid . . . Treated liquid ... No. I rough ... No. I fine ... ... No. 2 rough ... No. 2 fine ... ... Untreated liquid ... Treated liquid ... No. I rough ... No. I fine ... ... No. 2 rough ... No. 2 fine ... ... Untreated liquid ... Treated liquid ... No. I rough ... No. I fine ... ... No. 2 rough ... No. 2 fine ... ... Untreated liquid ... Treated liquid . . . No. I rough ... Feb. 12th. F eb . 2 7 t It. Marclc 12th. 0.46 Oxygen Consumed. ive blins. - 1.60 I -60 0'57 0'44 0'73 0'47 1-68 1-60 0'44 0-40 0.5 I 0.38 1-68 I -64 0'44 0.4 I 0.48 0'3.5 2-04 I -88 0'59 April 6th.Untreated liquid ... Treated liquid ... No. I rough ... No. I fine ... ... No. 2 rough ... No. 2 fine ... ... I -92 I '92 1 '44 0.3 5 0.40 0.38 Four Hours 8-20 8.56 2-86 1-36 4.08 2-48 8.96 8*80 2-40 2.08 2-56 2.08 8-00 8-24 2-24 1-92 2-48 1-44 8.64 8-64 2-80 2-24 3 '04 2-08 7'44 7-30 I '90 I '40 1-72 1.17 Nitrogen. -. - trace ,, 7 , ,, - - 0.005 0.005 0.015 0'010 - - 0'0 I 0'0 I 0'0 I 0'02 - I 0'010 0'020 0.005 0.030 - - 0'02 0'02 0'02 0'02 Nitric. - - 0.19 0.27 0'34 0'22 - - 1-13 0.6 I 2.9 I 2'02 I - 1.81 0'49 2.3 I 1-21 - - 1-20 2'11 1-17 3-11 - - 1-89 3.00 2.69 3'49 (For further particulars of analysis see Curve IIA.) I t will be seen that bed No. z is again maturing at a quicker rate, but the difference is not so marked as with the liquor from the septic tank before cleaning out.On February 19th growths were forming on No. I bed, but not on No. 2 ; on the 23rd and after the growths increased rapidly on No. I , and there was only a slight formation on No. 2 bed. On March 5th the syphon of No. I began to choke, and the sprinkler refused to face the wind, owing to the slow discharge. The pipes and syphons were washed out with strong oxychloride, and the working went on properly. It was not until about eight months afterwards (November 18th) that in No. 2 cleansing became necessary on account of slowly accumulating growths ; they were cleaned out with strong oxychloride without disconnectiiig the pipes. Samples of the media were examined on December 7, 1908, for sludge.No. I, rough and fine, contained respectively 24.7 and 63.9 per cent. more deposit than No. 2 ; the latter was also visibly cleaner. The liquids from the washing of the media were tested by incubation ; of the rough, No. I developed H,S, No. 2 a trace ; of the fine, No. I gave heavy H,S, No. 2 none.BY THE ' I OXYCHLORIDES" PROCESS U O O h N *U)W ""0 ?P d- d- !n N 9 E or, * X , d-m h? *PI *rr N N U U U U 9 * su 00 M ? 0 "0 I 0 M qo 0 2 _____. . . . . . . . . . . . . . . . _ _ ~ _ ~ . . . . . . . . . . . . . . . ~~~ ~- . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .192 SEWAGE PURIFICATION AND DEODORISATION A number of tests on effluents were made from the various depths ; the 2’ 0’’ rough in all cases would not keep sweet after 2 4 hours.The fine 2’ 0‘’ on several occasions remained sweet and cleared themselves on standing. (See Table IA, also Curve IIIA.) The fine samples show a better result than the rough both in five minutes’ oxygen and in nitration. At first sight it would appear that the 4‘ 0’’ depth gives very nearly the same average purification as the 6’ o”, therefore that the depth of the filter could be reduced, but this is not the case during the winter months nor at times of special impurity; and in general the greater depth is required for safety, particularly with rough medium. Conclusions from the third series of experiments- I. That a bed can be successfully matured when using oxychloride in such quantities as are required for preventing aerial nuisance.2. That a bed may be thus matured at a quicker rate, as is shown by Curve IA. 3. That this treatment renders possible the use of fine-grade filters, which are usually not available on account of clogging by growths. 4. That it also renders easy the clearing of pipes, sprinklers, and syphons blocked by growths, without disconnecting the system. 5. That any place adopting the treatment for aerial nuisance has the additional advantage of ( I ) the retarding and destruction of growths ; (2) assistance in the work of oxidation ; (3) the means at hand of a very active deodorising and disinfectant agent, which may be made use of for general purposes. The fact that oxychloride is a clear liquid free from suspended solids favours the use of automatic appliances for its addition in regulated quantities, and for this purpose Mr.Fieldhouse has designed a patent automatic supply and recording valve used at Guildford, which has been found simple and reli- able for adding any determined amount in proportion to the volume treated. SECTION 3.-ADDITIOXAL NOTES ON SPECIAL POINTS. Behaviour of Treated and Zl-nti-caied Efluent as mgarn‘s Dissolved Oxygen.- A septic effluent showing a five minutes’ oxygen figure of 2.16 parts per IOO,OOO was treated with oxychloride containing 0.1 per cent. of av. C1 in the propor- tions of 35, 50, and IOO per cent. of the five minutes’ oxygen figure. The treated and untreated samples were all shakenup with air in the same way at the beginning, so as to saturate them with oxygen ; they were then allowed to stand in closed vessels at the ordinary temperature.The dissolved oxygen determinations have the following figures- _ _ _ ~ Untreated ... ... Treated, 35 per cent .... Treated, 50 per cent... . Treated, IOO per cent. Cubic Centimetres of Dissolved Oxygen per Litre. In Original Liquids at Beginning. 3’43 3’43 3’53 3.62 After Shaking for two hiinutes with Air. 6.22 6 - 2 2 6-27 6.50 After Incubating for Six Hours. 0’37 3.06 3‘34 3’99BY THE '' OXYCHLORIDES" PROCESS194 SEWAGE PURIFICATION AND DEODOKISATION Five Miris.' Oxygen. Liquid Under Experiment. I. Septic liquor . . . . . . . . . 3-6 + 75 % of fully aerated water - 11. Septic liquor . . . . . . . . . 2-24 + 75 % of fully aerated water - These results prove that the treatment prevents the rapid de-aeration which septic effluents naturally undergo, and that the treated effluents maintain a satisfactory aeration. The following are other examples- Av.C1. Added. ____ 3.6 .- 2-24 - Dissolved Imme- diate. Oxygen C.C. per Litre after 24 Hours. 6.13 7-15 1 7-10 5'44 This proves that the effluent would not cause any appreciable de-aeration if discharged into even a small body of water. The treated liquor takes up oxygen more rapidly than the untreated. Thus a sample which had received oxychloride equal to only 35 per cent. of the five minutes' oxygen figure showid when collected from the sprinkler arms 2-23 C.C. of dissolved oxygen per litre, while the untreated had only 1.30 C.C.Besides being better in other respects, the effluent from a sprinkler bed that has received treated septic liquor also contains more dissolved oxygen than that from one receiving untreated liquor. The following are examples taken at a very unfavourable period, and also at a more favourable season- Untreated . . . . . . Treated 50 per cent. Treated IOO per cent. ~~ I Cubic Centimetres of Dissolved Oxygen per Litre. 3-71 ~ 5-71 6'22 - - - That a treated liquor will more quickly become fully aerated when agitated with air is clearly shown in the following instance on May Ist, where the relative difference is greater- C.C. per litre of dissolved oxygen Ditto after shaking with air for I+ minutes . . . . . . . . . Percentage of full aeration originally .. . . . . . . . acquired . . . . . . . . . In Untreated. 0'74 3-16 about 50 per cent. In Treated. 2-23 4'55 about 75 per cent.BY THE “OXYCHLORIDES” PROCESS I95 The fact that a septic liquor treated with oxychloride for sterilisation when added to a well-aerated river water does not deprive it of its dissolved oxygen, goes to prove that the treatment might be adopted without the expen- sive addition of the usual bacterial installation : as the presence of traces of av. C1 does not form a precipitate of solids, therefore it is particularly applicable in tidal and other large flowing rivers, or to shellfish or water- cress beds. Efect of the Presence of Available Chlorine upon the Dissoived Oxygen Test.- The dissolved oxygens were determined by Rideal and Stewart’s modification of the Winkler process.This is not appreciably interfered with by the presence of available chlorine in the amounts occurring in the oxychloride treatment, as the following example proves. A septic liquor was treated with 0.1 per cent. strength oxychloride, and after two hours’ contact the available chlorine was 1-41 parts per IOO,OOO, which is over the ordinary amount. Two bottles were filled, one having its available chlorine neutralised by arsenious acid ; the other was untouched. The dissolved oxygen estimated at once was in the former 3-34 C.C. per litre, in the latter 3-53 C.C. If the av. C1 had yielded its equivalent of oxygen it would have added 0.32 part or over 2 C.C. Disappearance of the av. Chlorine when mixed with Septic Liquors.-The sample of septic liquor showed a five minutes’ oxygen figure of 5-28 parts per 100,ooo.It was treated in each case with the same proportion of av. C1, namely, 6 parts per IOO,OOO, or a little over 100 per cent. of the five minutes’ oxygen. The solutions of oxychloride used were, however, of different strengths and characters, as below- Strength of Oxychloride Solution used in per cents. of av. C1. New oxychloride, 0.169 per cent. ... ... New oxychloride, 0.084 per cent. . . . Old oxychloride, 0.174 per cent. ... Old oxychloride, 0.082 per cent. ... A sample of the . . . ... ... g:: 0.10 per cent. ... ... long time J Parts of av. C1 added per 100,000. Time Required for the Disappearance of the av. C1 Reaction to KI and Starch. 24 hours 3 hours 15 miiiutes I hour 20 minutes I hour 50 minutes The words ‘‘ old oxychloride ” refer to a preparation made in the original manner ; “ new oxychloride ” means the liquid obtained by the new process, in which certain additions improve the keeping properties ; it will be seen that its available chlorine also has a more lasting action in contact with the septic liquor.Another important point proved is that in all cases the liquid should be added in a diluted rather than in a concentrated state ; with a more dilute oxychloride when the same total amount of av. C1 is added, this av. Cl lasts longer, due to better mixing, and to its not being taken up so rapidly by the organic matter. This is shown by tables A, B, C. It will be noticed in Table A 4-29 parts of av. C1 have been required, in Table B 2.2 parts, and in Table C 1-13 parts.The bacterial results show that the presence of traces of av. C1 has been in each case sufficient to removeTABLE A.-Comparative rates of absorption of av. C1 with the old and with the new forin of oxychloride solution. The test was carried out by adding to one Eifre of the septic effluenl (with a five minutes' oxygen consumed of 4'56 parts per IOO,OOO) measured quantities of the two solutions of 9.195 per cent. strength av. C1 at intervals as below. E; ZTABLE B.-Absorption of av. C1 when added progressively, and destruction of R. coli by the slight traces left. Tests as in Table A with a septic efilueiit giving a five minutes’ oxygen figure of 3.04 parts per IOO,OOO, but the oxycbloride was added to ticlo Zilvcs ol eflfluent.198 SEWAGE PURIFICATION AND DEOEORISATIONBY THE ii OXYCHLORIDES” PROCESS I99 coli with one hour to two hours’ contact.The percentages of av. C1 employed in relation to the five minutes’ oxygen figure have been ’in A, with old form, 101.7, with new form, 94 ; in B, with new form, 72.3 ; in C, with ditto, 50.44. In my former experiments with the “ old oxychloride ” I found that the best proportion to use on the average could be calculated by multiplying the five minutes’ oxygen figure by 1.7, or as it is now phrased “ 170 per cent.” This ratio has now to be modified in the use of the new oxychloride for effluents, as the following experiment shows. A tertiary effluent having a five minutes’ oxygen figure of 0.366 was treated with an equal quantity, namely, 0.366 parts per IOO,OOO, of C1, or IOO per cent.; available chlorine was found to be present after eighteen Jzouq B. coli was absent from 10 C.C. after two hours’ contact, and from 20 C.C. after three hours’ contact. Hence an av. C1 equal to the five minutes’ oxygen, or IOO per cent. treatment, in this case has been more than suficieut. Therefore it is not generally necessary to add more than this amount. AERIAL NUISANCE. (See Royal Commission on Semage Report, 1908, pp. 195-201.) Septic tank liquor possesses a stronger odour than raw sewage, owing to the breaking up of the organic matters in the tank and the formation of gases, and this liquor frequently gives rise to aerial nuisance when discharged by sprinklers on to percolation beds, The smell may be caused by a number of bodies, the more offensive of which are capable of deodorisation by oxychloride. The most prominent, especially when gas liquor enters the sewage, is sulphuretted hy drogen. The following table gives the amount of oxychloride required to be added to obviate the nuisance. Strong septic liquor yielding a five minutes’ oxygen figure of 5.2 parts per IOO,OOO, and showing a considerable evolution of sulphurettcd hydrogen by darkening a lead paper suspended over it, was tested with the lead paper after the addition of various proportions of oxychloride solution containing 0.2 per cent. of available chlorine. Results of two trials are given below- Av;dlable C1 added in per cents. of the live minutes’ Oxygen. Per cent. I 0 20 25 35 40 50 75 90 I00 Sulphuretted Hydrogen evolved after Treatment. -_ - I. strong 1 di;kdig;t 1 trace nil nil nil nil nil - 11. - - distinct nil nil nil nil - nil Immediate Test by K1 and Starch for available C1 left. I. nil nil nil nil nil trace I very distinct distinct quantity re- maining after 1 4 hours - I nil nil nil trace trace remaining after 5 minutes - i very distinct This shows that 25 per cent will be ordinarily sufficient to remove the odour, and that with such strong effluents 50 to 75 per cent. leaves a residue of av.200 SEWAGE PURIFICATION AND DEODORISATION C1 for further purification. In each case 35 per cent. has removed all the sulphuretted hydrogen. The oxycliloride also prevents or retards the further development of sulphuretted hydrogen. Thus I found in comparing the effluents from sprinkler beds receiving treated and untreated septic liquors, that although both might at first be free from sulphuretted hydrogen, the latter developed it on incubation or keeping, whilc the treated usually did not, owing to the effect of the oxycliloride on the organisms.

ISSN:0014-7672    

DOI:10.1039/TF9090400179

出版商:RSC    

年代:1909

数据来源: RSC