A SLIGHT MODIFICATION IN VICTOR MEYER'S VAPOUR DENSITY APPARATUS. By T. S. PATTERSON.

IN view of Mr. A. Price's letter in your issue of November 19 (p. 249), I should like to say that for the past thirteen or fourteen years, I have used copper in place of glass for the outer tube of Victor Meyer's Vapour density Apparatus, with most satisfactory results. The apparatus I use consists of a copper tube, A, closed, by brazing, at the lower end. It has a short side tube, B, brazed on at the upper end, as shown in the accompanying sketch. Into the upright portion of B, a small condenser is fitted.

B

The ordinary inner tube of the Victor Meyer apparatus is then cut across at H and the tube E is fitted, as shown, to a large rubber stopper. The two parts of the apparatus are then joined together again at H by a piece of fairly thick india-rubber tubing. For the release of the little vessel containing the substance, the arrangement described by me in the CHEMICAL NEWS (1908, xcvii., 73), is used. The side tube, G, is connected by means of a convenient length of narrow rubber tubing to the delivery tube, P. A small quantity of mercury is put into the bulb, D, to break the fall of the little glass tube. The apparatus is heated by boiling any suitable liquid in A. I generally use amyl alcohol. The apparatus thus fitted gives very accurate results.

Organic Chemistry Department, University of Glasgow. Nov. 22, 1920.

PROCEEDINGS OF SOCIETIES.

ROYAL SOCIETY.

Ordinary Meeting, December 9, 1920.

PROF. C. S. SHERRINGTON, President, in the Chair.

THE following papers were read :—

"Double Refraction and Crystalline Structure of Silica Glass." By LORD RAYLEIGH, F.R.S.

1. Although glasses in general have no double refraction, except that due to bad annealing, yet silica glass is found to have a doubly refracting structure, which cannot be so accounted for, and must rather be regarded as crystalline.

2. The double refraction is very weak, of the order of 1/60 that of crystalline quartz.

3. In a mass of silica which has been melted, but not drawn or blown, the structure consists of doubly refracting grains with dimensions of about

mm., oriented at random. These grains may correspond to the original fragments of quartz which were melted to form the glass; but this is uncertain. At all events, the grains are very persistent, and individually survive a re-melting of the material.

4. If the grained material is drawn out while soft, the grains are elongated into crystalline fibres or ribbons. These fibres always give "straight extinction" in the polariscope, and their length is along one axis of the ellipsoid of optical elasticity, but apparently not always along the same axis. The fact that the crystalline grains can be made to flow in this way assimilates them to Lehmann's liquid crystals.

5. If the material is bent or twisted, the fibres follow its course unbroken, and are always extinguished in the polariscope if the nicols are set along the tangent and normal to their direction at any point.

6. Fused silica sometimes contains isolated small inclusions of quartz, with angular outlines, which have escaped vitrification. These are conspicuous in the polariscope by the strain effects they produce in the surrounding glass.

7. Discs of "optical silica," which are built up by a special process, show a most curious strain structure, on which the crystalline grains are superposed.

A heat treatment is described which resulted in the complete disappearance of all this structure and its subsequent reappearance.

"Effect of Asymmetry on Wave-length Determinations." By Prof. J. W. NICHOLSON, F.R.S., and Prof. T. R. MERTON, M.A., D.Sc., F.R.S.

(1) The apparent displacement of an unsymmetrical spectrum line caused by the finite resolving power of the spectroscope can be calculated on certain simple assumptions.

(2) The displacement is independent of the actual widths of the lines.

(3) It is considered that the general practice of measuring spectrum lines to a degree of accuracy far transcending the resolving power is not justified

"Effect of Concentration on the Spectra of Luminous Gases." By Prof. T. R. MERTON, M.A., D.Sc., F.R.S.

Certain spectroscopic phenomena appear to be associated with the concentration of the radiating atoms in the source. An increase in concentration may result in a broadening of the lines, a change in the structure of the lines, and changes in the relative intensities. Sources containing lithium exhibit these three phenomena, and the broadening is familiar in sodium flames. A study has been made of the behaviour of sources containing sodium and lithium. The results seem to exclude a temporary association of atoms as the cause of the changes, for the addition of large quantities of sodium to a source containing a trace of lithium produces no change in the lithium spectrum. Mixtures of hydrogen and helium have also been investigated.

The broadened lines of both these elements from vacuum tubes excited by condensed discharges are accounted for completely by the electrical resolution of the lines by the electric fields of neighbouring charged particles. Although the broadening of lines of the diffuse series of helium is less than that of the Balmer lines of hydrogen of equal term number, the helium line λ=4471 A. is found to be very broad, whilst the hydrogen line His sharp when the proportion of hydrogen present is very small. With an increased proportion of hydrogen the Balmer lines begin to broaden. It would appear that the electric resolution due to the proximity of a charged particle to the radiating atom depends on the nature of the charged particle, and that the mutual influence of charged atoms of the same kind on one another is vastly greater than that of different kinds of charged atoms.

Phenomena have been observed which suggest that when condensed discharges pass through vacuum tubes containing mixtures of hydrogen and helium, a partial separation of the gases is effected, the hydrogen being selectively driven out of the capillary into the bulbs of the vacuum tube.

"Measurement of Low Magnetic Susceptibility by an Instrument of a New Type." By Prof. E. WILSON, M. Inst. C. E., M. Inst. E.E.

The paper deals with the design, construction, and working of an instrument for the measurement of susceptibility (of low order) over a wide range of magnetic force, and thus avoids the difficulty met with in the Curie balance, the defections of which follow the square law, and, in fact, limit

the measurement of susceptibility of a given specimen to a very narrow range of magnetic force. The force due to torsion in a suspending fibre is replaced by an electro-magnetic system in which the mechanical force is due to two components-one proportional to the magnetic force impressed upon the specimen and the other variable if the susceptibility varies. The expression for the susceptibility is that of the reciprocal of a resistance multiplied by a constant, and thus the instrument lends itself to a great accuracy in the detection of variations in susceptibility.

"The Internal Energy of Inflammable Mixtures of Coal-gas and Air after Explosion." By Prof. W. T. DAVID, M.A., D.Sc.

In the first part of this paper an empirical law of cooling of exploded mixtures of coal-gas and air contained in a closed vessel has been formulated. This is based upon measurements of the heat loss by conduction and by radiation made during the explosion and subsequent cooling of the inflammable mixtures.

In the second part the heat loss measurements have been applied to the estimation of the internal energy of the gaseous mixtures at the moment of various stages maximum temperature and at during cooling. Curves connecting the internal energy and gas temperature have been prepared for mixtures of various strengths. An examination of these curves indicates that the internal energy during the early stages of cooling is a function of the time after explosion, as well as of the temperature of the gas. The interpretation of this is that combustion is not complete at the moment of maximum temperature and continues during the early stages of cooling. It is estimated that only about 90 per cent of the heat of combustion is transformed into thermal energy by the time the maximum pressure is reached, and that combustion is proceeding for at least o'25 seconds after this moment.

The factors limiting the pressures developed are discussed and a quantitative estimate made of their relative importance in mixtures of various strengths.

"Multenions and Differential Invariants." Prof. A. MCAULAY.

By

The paper is a summary of the properties of a linear associative algebra suitable for electromagnetic relations, differential invariants, and relativity. There are n fundamental units, otherwise it is the same algebra as that considered in a paper by W. J. Johnston and read to the Royal Society on November 20, 1919.

THE FARADAY SOCIETY AND THE INSTITUTE OF METALS. (SHEFFIELD SECTION).

AT a joint meeting of the Faraday Society and the Sheffield Section of the Institute of Physics held on Friday, November 19th, 1920, in the Mappin Hall of the Department of Applied Science of the University of Sheffield, a series of Papers on "Electro-deposition and Electroplating" were presented and discussed.

The afternoon session, which was presided over by Prof. C. H. DESCH, D.Sc., was devoted to problems relating to Electro-deposition generally.

"Electroplating for the Prevention of Corrosion." By Dr. LESLIE AITCHISON (communicated by W. R. BARCLAY).

The paper dealt more especially with the protection of iron and steel and their alloys. The conditions for proper protective coatings, namely, permanence and resistance and non-permeability to corrosive agents, were defined and the value of the various protective coatings discussed with relation thereto. In particular, the relative advantages of hot coating and cold electrodeposition were considered and the general advantage of zinc deposition over galvanising was emphasised.

The properties required in a metallic protective coating were classified as follows:-

I.

2

2.

3.

4.

5.

6.

7.

8.

9.

10.

Uniformity.
Permanence.
Good appearance.

Freedom from porosity.
Freedom from pinholes.

Good adhesion to base metal. '
Reasonable ductility.

Freedom from scaling or flaking.
Penetration into all parts of a complicated
surface.

Minimum tendency to promote corrosion of the base metal if the coating is pierced. Special emphasis was laid on the need for ductility and condition 10 was discussed and some apparent anomalies set forth. Finally the preparaption of the surface was dealt with.

t3ཕྱགག

2:

level

for

"Some Applications of Electro-Deposition in Aeronautical Engineering." By W. A. THAIN,

A.M.I.C.E.

Three cases of the electro-deposition of copper were considered, viz. (i.) as a protection against carburisation in case-hardening practice; (ii.) as a means of increasing heat conductivity; (iii.) as a means of building up a definite constructional detail.

As an example of (i.) the treatment of a camshaft was considered. In case (ii.) copper-plating of the steel cylinder of an aeronautical engine was examined. The deposit is of the order of 0'003inch thick, being greatest at the top of the flues and least at the foot. Finally an account was given of the details in the building up of the copper water jacket of a Beardmore aero-engine cylinder. The final thickness of the jacket wall is 17 mm.; it is tested under water pressure of 25-30lbs. per sq. inch, and as hundreds of such cylinders had stood the test of active service, the method had been thoroughly proved and must be considered a sound practical proposition. "The Electro-deposition of Cobalt.' By BYRON CARR. From a bath containing 4lbs. of cobalt sulphate crystals, 5 ozs. boric acid, and 24ozs. of sodium chloride per gallon and used at 34° C., excellent hard adherent deposits of cobalt were obtained, provided that these were not too thick, with 150 and 72 amperes per square foot respectively. Periods of immersion not exceeding 2 and 4 minutes respectively were suitable. The deposit is exceedingly resistant to atmospheric corrosion and is superior to nickel in the rapidity of deposition and hardness. Articles of domestic use so

treated were, however, subject to the deleterious action of fruit juice.

"The Use of Colloids in the Electro-deposition of Metals." By W. E. HUGHES, B.A.

Colloids and substances producing the same effects were often usefully employed in electroplating solutions. In one case at least-that of lead-no coherent deposit could be obtained unless some colloid, for instance, glue or peptone, was used. The effect usually produced by the colloid addition was a diminution of the size of grain of which the deposit consists, the diminution being, in some cases, so extreme that, to the naked eye, the deposit appears plastic and non-crystalline. The mechanism of the change from macrocrystalline to micro-crystalline was unknown. The differing views of Marc, Kohlschütter, and Freundlich and Fischer were referred to, and it was suggested that this difference of opinion was in itself sufficient reason for drawing attention to the operation of colloids in plating solutions. which was of interest both practically and theoretically.

"The Commercial Electrolysis of Zinc Sulphate Solutions." By S. FIELD, A.R.C.Sc.

The electrolysis of zinc sulphate solutions is carried on commercially in cold galvanising with a zinc anode, and in the electrolytic recovery of zinc, using an insoluble lead anode. In the latter case the solution contains a percentage of acid prior to electrolysis, and this increases as the zinc is deposited, usually on aluminium cathodes. A neutral sulphate solution containing 10 per cent of zinc may be produced by leaching ores or other zinc-bearing materials with sulphuric acid or acid zinc sulphate liquors from the electrolytic cells, and this composition may conveniently be altered to 8 per cent zinc and 3 per cent acid before electrolysis. Commercial electrolysis aims at the maximum extraction of zinc from such a liquor with a minimum of energy. The greater the extraction of zinc, the smaller the volume of liquor which circulates through the whole of the extraction plant, and the smaller the proportion of zinc which demands repeated purification. limit to this extraction is set by the cost of increased energy necessary to take out zinc from dilute liquors, due mainly to a considerable falling off in the current efficiency. Current efficiency is dependent upon a number of factors including current density, amount of zinc present, temperature, and last, but by no means least, the presence of impurities.

A

Several impurities normally occurring in these crude zinc sulphate liquors exert a marked deleterious effect on the deposition of the metal from these acid solutions. The effect of, among others, nickel and cobalt has been carefully studied. The presence of a few parts of either of these metals per million parts of liquor disastrously lowers current efficiency and increases energy consumption. Traces of cobalt produce an extraordinarily honey-combed deposit, while even smaller amounts of nickel produce large and clean-cut holes in the deposited metal. The types of corrosion exhibit a marked contrast.

Traces of these impurities thus seriously militate against successful deposition. Neither impurity is deposited with the metal but remains in the solution and, concentrating during the cycle of operations, soon places the operation outside

the sphere of economy. Colloids have an ameliorating effect, but of a temporary character only. The elimination of these impurities is an essential to successful deposition.

It is suggested that as the beneficial effects of traces of colloids are known and applied there is justification for keeping a close watch on all impurities in electro-depositing solutions even though the normal conditions of deposition may widely differ.

The evening session was presided over by Mr. E. A. SMITH. The Papers presented all dealt with various aspects of the electroplating of silver.

"Electro-Silver Plating and its Technical Development." By W. R. BARCLAY, O.B.E., A.M.I.E.E.

This paper dealt with the history of technical investigation, and research into the electrodeposition of silver, so far as the more practical aspect of electroplating is concerned

The work of early authorities, such as Smee, Gore, Langbein, and others, was briefly outlined, but it was pointed out that the main advances belonged to the last twenty-five years, and followed on the work of the physical chemists of the latter part of the nineteenth century, who laid the foundations of modern electro-chemistry.

The investigations of such research workers as Brunner, Foerster, Bancroft, Frary, and Porter, and others, on the influence of current density, metal and free cyanide concentration, and other cognate factors, on the character of silver deposits was discussed, and an outline given of the author's own experience in workshop practice.

Emphasis was laid on the necessity for careful co-ordination of the factors of metal and free cyanide content to that of current density. It was shown that though considerable latitude is allowable in practice, the best results and highest efficiency lie within fairly well-defined limits. With a current density of 2.75 to 3 amperes per square foot (for example) a solution containing 3 troy oz. per gallon with not less than twice a molecular equivalent of free cyanide, gave excellent results as to both quality of deposit, efficiency, and time occupied.

Among other matters it was pointed out that silver solutions in existence over sixty years still continued to yield excellent results, and that generally old solutions yielded better deposits at higher current densities than those newly made up from pure materials. Recent investigations had shown that this was due, in a great measure, to the presence of substances other than the simple double cyanide of silver and potassium, especially potassium carbonate, which salt is usually to be found in a higher proportion than any other, and increases the conductivity to a considerable degree.

While the technical efficiency of a modern silver-plating plant was fairly high, further investigation was necessary as to the exact influence of substances such as potassium chloride, sulphate, cyanate, and formate, all of which had been shown to be present in commercial plating baths, and necessarily made the chemical and electrochemical reactions of these baths somewhat complex. A question of further importance remaining to be solved was the influence of organic matter on the physical structure of deposits.

Silver on they

"The Chemical Composition of Old Plating Solutions, with Observations Working Properties." By G. B. BROOK, F.I.C., and L. W. HOLMES.

The paper dealt with a large number of soletions varying in age from one to fifty years, furnishing historical, chemical, and physical data with regard to each, and correlating the compoșition with the working properties in actual works practice. The graph showing the intimate relation of the carbonate content to the electrical conductivity was the outstanding feature of the paper, and constituted in itself a valuable discovery, which promised to have a far-reaching effect on the composition of the plating solution of the future The comparison of solutions, used respectively by "trade" platers and private manufacturers' plating departments, was of considerable interest.

"A New Maximum Current Density in Commer cial Silver-piating." By FRANK MASON, A. M. I. E. E.

During extensive researches on the physical structures of electro-silver deposits, which are at present being carried out at the University of Sheffield, the author observed that in certain instances exceedingly fine deposits of silver were obtained at an extraordinarily rapid rate.

None of the mechanical devices usually associated with rapid deposition are as yet available for production on a commercial scale of thick deposits of silver with a crystalline structure easy to "finish."

Microscopic examination of such silver structures show that these ideal conditions for a high finish are to a large extent governed by the size and position of the crystals of the deposit. Under normal conditions of commercial electroplating the higher the current density used, the larger is the crystal obtained, until the crystalline structure is such that "finishing" is impossible, and the deposit is said to be "burnt." The limiting current density permissible is 3 to 4 amperes per square foot.

During the investigation already referred to the electrolytes from which abnormal silver deposits were obtained always contained potassium carbonate. The metal from these was precipitated in a fine reguline condition, until the percentage of the latter compound reached the neighbourhood of 10 per cent, and the current conditions were normal. Any further increase produced large crystalline structures, very difficult to burnish or polish, and a cathode current density of 6 amperes per square foot brought down a deposit from which the desired final "finish" could not satisfactorily be produced. The solutions containing a higher percentage of potassium cyanide in conjunction with potassium carbonate, somewhat counteracted this undesired effect. It was further found that with a free cyanide content in excess of that usually advised in any up-to-date text-book, and with potassium carbonate in considerable quantities, not only could the current density be increased enormously, but that the deposit was of a finer texture altogether. Sufficient evidence was obtained that the potassium carbonate does really assist in the conductance of the current in an electrolyte of the double cyanide of silver and potassium.

The author has had the process installed on a commercial scale with most satisfactory results,