MEETINGS FOR THE WEEK.

Monday, November 1.

Royal Institution, 5. General Meeting.

Wednesday, November 3

Society of Public Analysts, 8. "Gravimetric Estimation of Bismuth as Phosphate and its applications in Ore Analysis" by W. R. Schoeller, Ph. D., and E. F. Waterhouse.

"Time Factor in Saponification" by Percival J. Fryer, F.I.C. "Position of Analytical Chemistry in France" by Victor Cofman, B.Sc.

"Apparatus for collecting samples of water at great depths" by W. T. Burgess, F I.Č.

Thursday, November 4.

Royal Society. "Vibrations of an Elastic Plate in Contact with Water" by Prof. H. Lainb.

"Transmission of Electric Waves around the Earth's Surface" by Prof. H. M. Macdonald.

"A Re-examination of the Light scattered by Gases in respect of polarization." II. "Experiments on Helium and Argon" by Lord Rayleigh.

"Dilatation and Compressibility of Liquid Carbonic Acid" by Prof. C. F. Jenkin.

"Radiation in Explosions of Hydrogen and Air" by W. T. David "Photochemical Investigations of the Photographic Plate" by R. E. Slade, D.Sc., and G. I. Higson.

"Relationship between Pressure and Temperature at the same Level in the free Atmosphere" by E. H. Chapman, D.Sc. "Note on Vacuum Grating Spectroscopy" by Prof. J. C. McLennan, F.R.S.

Chemical Society, 8. "The Preparation of 4-, 5- and 6-methylcoumaran-2-ones, and some derivatives of O-, m- and ptolyloxyacetic acids" by L. Higginbotham and H. Stephen. "A New Method for the Preparation of 2: : 4-dihydroxy- and 2:44 -trihydroxy-benzophenone, and some Observations Relating to the Hoesch Reaction" by H. Stephen.

Triphenylarsine and Diphenylarseníous Salts" by W. J. Pope and E. É. Turner.

"The Preparation and Physical Properties of Carbonyl Chloride" by R. H. Atkinson, C. T. Heycock and W. J. Pope. "Interaction of Ethylene and Selenium Monochloride " by H. W. Bauser, C. S. Gibson and W. J. Pope.

"A Study of the Reactions of Sugars and Polyatomic Alcohols

LECTURESHIP IN PHARMACY.

APPLICATIONS are invited for the above

lectureship. Salary £400 per annum. The appointment is a whole-time one in the Chemistry Department of the College. The lecturer will be chiefly required to give instruction in Pharmacy and Materia Medica to students preparing for the Professional Examinations of the Pharmaceutical Society of Great Britain, but qualifications in Botany and Chemistry are desirable. Applications, stating age and particulars regarding training qualifications and experience, accompanied by copies of testmonials to be sent to the undersigned not later than Monday, November 1st. T. J. REES, B.A., Director of Education, Education Offices, Swansea.

11th October, 1920.

Applications" by G. Van B. Gilmour.

"The Sulphonation of Glyoxalines" by F. L. Pyman and L. A. Ravald.

"O- and _p- tolueneazoglyoxalines" by F. L. Pyman and L. A. Ravald.

'Investigation of Sodium Oleate Solutions in the Three Physical States of Curd, Gel and Sol" by M. E. Laing and J. W. Mc Bain.

"The Constitution of Polysaccharides." Part I. "The Relationship of Inulin to Fructose" by J. C. Irvine and E. S. Steele. "The Preparation of Ethyl Iodide" by B. E. Hunt. "Action of Sulphur Trioxide on Aromatic Ethers" by R. C. Menzies.

Association are desirous of appointing a Senior Research Physicist and an Assistant Research Physicist to conduct work of a physico-chemical and physical nature in connection with the Linen Industry. According to qualification for the position the remuneration offered will be about £700 and £400 respectively. Application Forms may be had on application to the SECRETARY, Research Institute, Lambeg, Belfast.

PUPIL ASSISTANT.-A London Public

analyst with a large practice has a vacancy in his Laboratory. No premium. Apply "ANALYST," 24, Aldgate, London, E1.

"Acetylacetones of Selinium and Tellurium" by G. T. Morgan and H. D. K. Drew.

"The Formation of 2:3:6-Trinitrotoluene in the Nitration of Toluene" by R. B. Drew.

"The Formation and Reactions of imino compounds." Part XX. "The Condensation of Aldehydes with Cyanoacetamide" by J. N. E. Day and J. F. Thorpe.

"The Formation and Stability of Spiro-Compounds." Part III. "Spiro-Compounds from Cyclopentane "by O. Becker and J. F. Thorpe.

"Condensation of Dimethyldihydroresorcin with Aromatic Aldehydes" by H. Chattopadhyaya and P. C. Ghosh. "The Influence of Lead on the Catalytic Activity of Platinum' by E. B. Maxted.

If in good condition, Sixpence per copy will be

paid for any of the undermentioned numbers of the CHEMICAL NEWS which may be forwarded to this office:

3048, May 24th, 1918.e 3051, July 5th, 1918. 3053, August 2nd, 1918. 3054, August 16th, 1918.

3056, September 13th, 1918. 3059, October 25th, 1918

3062, December 6th, 1918.

3063, December 20th, 1918.

3064, January 3rd, 1919.

3066, January 17th, 1919.

3068, January 31st, 1919.

3069, February 7th, 1919.

3070, February 14th, 1919.

3075, March 21st, 1919.

CHEMICAL NEWS OFFICE, 97, SHOE LANE, LONDON, E.C.4

qualified. Special knowledge of either biological or colloidal chemistry essential. Apply, giving full particulars, to MESSRS. CADBURY BROS., LTD., Bournville, Birmingham.

KIESELGUHR (DIATOMITE).

All

grades for every purpose always in stock at principal ports. A. EGGERMONT, Victoria Chambers, Middlesbrough-on-Tees.

PATENTS, TRADE MARKS. Handbook

and advice free-B. T. KING, British and U. S. Regd. Patent Attorney, 146a, Queen Victoria St., London. 35 years reference

FOR Sale.-Filter-press, by Manlove Alliott;

practically new; centre feed, with pressure pump; offers invited.-CANNINGS, Salop Street, Highgate Birmingham.

COVERS FOR BINDING.

Cloth, Gilt-lettered, Covers for binding the Half-yearly Volumes of the

CHEMICAL NEWS

may now be obtained. Price 2/- each (post free 2/3). Volumes bound in Cloth Cases, Lettered, and Numbered at 35. per volume.

CHEMICAL NEWS OFFICE,

97, SHOE LANE, LONDON, E.C.4.

THE CHEMICAL NEWS.

VOL. CXXI., No. 3160.

EDITORIAL.

THE Joint Meeting of the Faraday Society and the Physical Society that was held at the Hall of the Institution of Mechanical Engineers on Monday, October 25, adds one more to the important series of joint meetings that have recently been arranged by the Faraday Society under the Presidency of Sir Robert Hadfield. The subject, "The Physics and Chemistry of Colloids," is one of the greatest importance at the present time; the bearing upon manufactures, and indeed upon every sphere of industry of those bodies whose dimensions lie between that of the molecule and the smallest macroscopic particle, is only beginning to be realised.

The Science of Colloids, although dating from the time of Graham, is still in its infancy, but judging from the volume of the papers that were read at the meeting and from the enthusiasm shown by the speakers, that "infancy" is a very vigorous one.

We understand that the complete "Discussion" will be published before long, but the "Introduction," by Professor Dr. The Svedberg, of the University of Upsala, is of such interest that we are reproducing it at once and hope to publish others of the papers as space may permit.

THE PHYSICS AND CHEMISTRY OF COLLOIDS AND THEIR BEARING ON INDUSTRIAL QUESTIONS.

A SHORT SURVEY OF THE PHYSICS AND CHEMISTRY OF COLLOIDS.

By DR. THE SVEDBERG, Professor of Physical Chemistry, University of Upsala.

THE science of colloids is a science of the microstructure of matter. In it is reflected the tendency of modern natural science to deal more and more intensely with problem of structure in its full extent. In the great science of the structure of matter, the science of colloids forms the domain that lies above molecular dimensions and beneath macroscopic dimensions. In this domain we have a great number of those systems which are the basis of our material culture and the basis of life as a whole. All living beings are built up of colloids; almost all our food, our articles of clothing, our building materials, are colloids. Or, to mention some special systems, protoplasm, proteins, glue, starch, all kinds of fibres, wood, brick, mortar, cement, certain kinds of glass, rubber, celluloid, &c. The importance of colloid science for many industrial questions is, therefore, beyond all doubt.

The science of colloids is a rather young one. The field of study which it concerns has for a long time been disregarded. In order to be able to treat with success all the questions presented to us by industry, there is still much to be carried out in the department of pure colloid science. In what follows, I will try to give a short review of

the scientific results so far obtained and of the problems which, in my opinion, need especial attention.

The various branches of colloid science are connected to such an extent that it is very difficult to treat the different questions separately. We will try to fix our attention on two principal problems: (1) the formation of colloids, and (2) the changes of structure in colloid systems. Connected with both these is the problem of the properties of colloids and the changes in these properties during the processes mentioned under (2).

The formation of a colloid may be effected in two ways, different in principle, viz., viz., by condensation and by dispersion, according as one tries to obtain a microstructural system, a colloid, from a molecular structural or a macrostructural system.

In the case of a condensation process, the degree of "graininess" or the degree of dispersion will become higher as the degree of supersaturation increases; which must, of course, always precede condensation. This is the case when fogs are formed by adiabatic expansion of gases. e.g., cloud-formation in the atmosphere, when metal colloids are prepared by condensation of metal gas from the electric arc, when a slightly soluble substance, e.g., barium sulphate, is precipitated by means of a reaction between ions. The condensation always proceeds from certain heterogenities in the medium; condensation centres or nuclei. As such nuclei, we may have particles of the substance which is to be formed by the condensation, e.g., precipitation of gold on small gold particles when preparing gold hydrosols or gold ruby glass, or gas ions, e.g., fog formation in gases at low degrees of supersaturation, or complex molecules, e.g. fog-formation in gases at high degrees of supersaturation. The manner in which these nuclei are introduced into the system under condensation is of great importance for the degree of dispersion of the colloid resultant. If the nuclei are not introduced into the system all at once, but gradually in the course of the condensation process, the particles will be very unequal in size.

The biologically important colloids, e.g., the proteins, are evidently all formed by condensation, but no details of this process are known. The tendency towards condensation manifests itself in the fact that even the protein molecules are, from a purely chemical point of view, condensation products.

We

In case of a dispersion process, there is always work to be performed against the surface tension or the cohesion force. Accordingly, such a process is, in contradistinction to a condensation process, a forced and not a spontaneous one. know very little as to the relation between degree of dispersion and experimental conditions. When emulsifying fats and hydrocarbons, the surface tension may be lowered by the addition of small quantities of alkalies or soaps. Grinding, in general, does not lead to a very high degree of sub-division; but it is possible to increase the latter by adding an indifferent solid diluent which can easily be dissolved and removed, leaving the disperse phase suspended in the solvent used. Thus, colloid sulphur has been prepared by grinding sulphur with urea and putting the substance in water. A combination of grinding and

chemical effects on the material may also be used. It seems that, in many cases, a prevention of the aggregation of the particles by means of suitable additions ought to render possible the preparation of high disperse systems by pure grinding. Bubbles and foam might be regarded as a kind of colloids formed by dispersion. On spontaneously breaking up, they form new disperse systems in which the phase that was previously the continuous one becomes the non-continuous one, and vice versa. The system: "soap foam-air," with the soap solution as the continuous and the air as the non-continuous phase, is transferred, on account of surface reduction, i.e. condensation, into the system: "soap solution drops-air," with the air as the continuous phase. Mercury foam in water (produced by means of pressing water through mercury) breaks up into a mercury hydrosol partly very fine-grained.

A newly-formed colloid may, immediately after its formation, undergo changes of structure of a more or less profound nature. On the other hand, it is nearly always possible to prevent the occurrence of such changes, and therefore we have a right to distinguish and investigate the primary structure as the direct result of the colloid-forming process. Colloids with primary structure may conveniently be called primary colloids.

We have at our disposal several methods for the closer study of the structure. Almost every property of a colloid depends on the structure, and therefore, conversely, from the study of the properties of colloids we may draw conclusions as to their structure.

The most important and most obvious means is the microscope and the ultramicroscope. With their aid, it is often possible to settle whether the colloid under investigation is of a grainy structure, e.g. a colloid gold solution in water, a gold hydrosol, or of a foamy structure, e.g. high disperse soap foam, or of a fibrous structure, e.g. soap solutions of a certain concertrations. The number and approximate size of the discontinuities, e.g. the particles, may also be determined in this way. One may, for instance, count in the ultramicroscope the number of particles observed in a certain volume of gold hydrosol, and by means of analysis determine the content of gold present in the sol. From these figures we get the mass and approximate size if, for instance, we make the assumption that they are spherical. On the other hand, the ultramicroscope gives little or no information about the form or structure of the particles.

A means of deciding whether the particles are symmetrical is to be found in the study of the behaviour of the colloids in magnetic or electric fields. Non-symmetrical particles are oriented by such fields and thereby impart to the colloid a certain, though in general very slight, degree of double refraction, which may easily be measured with great accuracy. In this way, we have been able to settle that the particles in common sulphur hydrosols, prepared by oxidation of hydrogen sulphide, are spherical; but sulphur hydrosols, prepared by grinding sulphur with urea, are dissymetrical. The particles in gold hydrosols prepared by reduction are dissymetrical to a high degree.

Two other optical properties, viz., the light absorption and its accompanying phenomena, the scattering and polarisation of light-the Tyndall

phenomenon-may also in certain cases be used for structure studies. Theory, as well as practice, proves that these phenomena are, for instance, dependent to a great extent on the degree of dispersion of the sol. The form and structure of the particles also influence the said properties, but in a manner hitherto unknown. The emission of light from illuminated particles especially-the Tyndall light cone-varies to a great extent with the size of the particles or the degree of dispersion. Colloid solutions that contain very small particles, e.g. Faraday's gold hydrosols, give only a slight emission; the light one is scarcely visible. Those with large par ticles, e.g. gold sols made from Faraday's gold sols by allowing the particles to grow in a reduction mixture, emit very much light; the Tyndall cone is very prominent. The optical properties of the particles also play a great part. metal particles emit much light, particles of silicic acid or gelatin only a little.

Thus

The resistance exerted on the particles by the surrounding medium when they move under the influence of a force is a means of investigation that is now often used for the determination of the size of the particles. In some cases, e.g. when a sol is filtered through a membrane of collodium or gelatin (ultrafiltration), the connection between the resistance and the size of the particles is not known in detail, but we are justified in assuming that the resistance rises with the size of particles. The small particles are more rapidly pressed through the filter than the large ones. Certain kinds of filters transmit molecularly dissolved substances, but not colloids; an important method of separation, especially in biochemistry. If a colloid is separated from a great quantity of dispersion medium by such a membrane, the molecularly dissolved substances-the crystalloidsdiffuse through the membrane, leaving a colloid of a purer state-Graham's dialysis. If the particles are spherical and move through a liquid, the resistance is Cnry, where is the viscosity, r the hadius, and v the velocity. If the force that causes the movement is known, e.g., gravity. the radius may be calculated.

Thus, by measuring the velocity of sedimentation, the radius can be found. We have

wheres, is the specific gravity of the particle, s, that of the liquid, and g the gravity constant.

Even when no exterior forces are acting, the particles in a colloid solution are in movement because of the impacts from the surrounding molecules. This is the so-called Brownian movement, which has attracted so much attention of late. According to the kinetic theory of the Brownian movement, which has been fully confirmed experimentally, each particle, whatever its size and nature, has the same translatory energy as a 3RT molecule, i.e. where R is the gas constant, 2N

T the absolute temperature, and N the Avogadro constant. Because of the resistance of the surrounding medium, the mean value of the quadrate of the distance traversed in the time t by the particle is 2Dt, where D, the diffusion constant

RT 1

Ν' (πηγ particle is In certain

of the particles, has the value Thus, if the displacement of the measured the radius must be found. cases it is more convenient to measure D directly and then calculater by means of this experimental value.

Owing to the fact that the size of the particles is rather great in comparison with that of the molecules, colloids diffuse very slowly compared with crystalloids. As a matter of fact, Graham the founder of colloid chemistry, regarded this property as the fundamental difference between colloids and crystalloids. We know now that between colloids and crystalloids-so very different in their extremes-there exist all degrees of transition forms, and therefore all degrees of diffusibility.

The size of particles may also be determined by measuring the sedimentation equilibrium, i.e. the distribution of the number of particles per cc. under the joint influence of gravity and the Brownian movement. In this equilibrium, the concentration of the colloid diminishes exponentially with increasing height, as is the case with the atmosphere surrounding the earth. We have

Consequently, the osmotic pressure is a measure of the degree of dispersion. The osmotic pressure of a colloid is often determined by means of a common osmometer, provided with a membrane permeable to crystalloids, but impermeable to the colloid particles. Now, however, owing to ion adsorption, the particles are, in most cases, surrounded by an electric double-layer of ions, and the colloid thus acts as an electrolyte, one ion of which is able to penetrate the membrane, but not the other. This causes complications. called membrane equilibrium is formed, and the osmotic pressure found is not a real measure of the structure of the colloid.

A so

Owing to the Brownian movement, the number of particles in every small volume of a sol undergoes spontaneous and incessant fluctuations. Hence the value of every property of the colloid in the small volume fluctuates.

This phenomenon, predicted by the kinetic theory for colloidal as well as for molecular solutions has been the subject of extensive investigations. The results are of importance Decause they show that Boyle's law holds good very exactly for dilute colloind solutions, and because of the light they have thrown on the applicability of the probability calculus to a natural phenomenon. From these studies we have also obtained a deeper comprehension of the conception of entropy. For they have shown, in a direct and experimental way, that the law of the incessant growth of entropy only holds for macroscopic systems.

It often occurs that the particles of a colloid are to small to be measured directly, e.g. by means of the ultramicroscope or by determining the velocity of sedimentation. In some of these cases one can overcome the difficulty by depositing gold on the particles, thus increasing their size. This method has already been applied to sols of almost all the metals and to sols of some sulphides. If the quantity of gold in a particle is known, it is easy to calculate the radius in the usual manner.

In most colloid solutions and precipitates there are particles of various sizes, and the investigator should, of course, be able to determine not only the mean size, but also the real structure of the sol, e.g. the law governing the distribution of the various sizes of particles. Among the phenomena reviewed above only the velocity of sedimentation, the sedimentation equilibrium, and the Brownian movement have been used in studying the distribution of the size of the particles. On the basis of the former phenomenon just mentioned a method has been worked out which has already provided us with much valuable information as to the formation of the various particles in colloids and the agglomeration processes taking place in them.

(To be continued.)

THE CONSTITUTION AND STRUCTURE
OF THE CHEMICAL ELEMENTS
THE FIFTH PRINCIPLE (continued).
By HAWKSWORTH COLLINS.

SINCE it has been proved that it is not a matter of chance that in 26 cases out of 44, an element has been found to exist in its ultimate matrix in association with simpler elements, the sum of whose atomic weights is equal to the atomic weight of that element, and whose valences explain the valency of that element; and since it has been shown that in many cases the particular integer obtained as the atomic weight of an element is not only especially suitable for explaining the properties of the element in comparison with the two adjacent integers, but also that frequently it is especially suitable in comparison with several adjacent integers; and since it has been shown that chemists admit the possibility of large errors in the atomic weights; and since also it has been proved that there are large errors; it is evident that some may be expected in the integers obtained from the experimental atomic weights, and that these errors may be corrected by observing the special suitability of adjacent integers.

Thallium (204) = Cu,Se.

The mineral crookesite is Cu,Se, Tl,Se, Ag,Se. Thallium is widely diffused in Fe, Cu pyrites.

Berzelianite Cu2Se with Ag, Tl, Fe.

There is evidently only one possible combination of elements, viz., Cu,Se, which can be said to be especially associated with T1; and it happens that its molecular weight is exactly equal to the experimental atomic weight of T1, when O=16. The probability that the weight of the only associated combination of elements would accidentally happen to be 204 is something like 1: 100. T will probably be found later to be a tetrad although at present it seems to be recognised only either as a monad or a triad, for Cu2Se must form an artiad to be in accordance with the theory.

Bismuth (209) CoAs, AuC=ZrSb. Native bismuth is abundant and intimately mixed with CoAs, especially in Saxony.

Bismutite is found with auriferous quartz in the Transvaal. Bismuthinite, a sulphide of bismuth, occurs with gold in Rowan Co., N.C. Bismite, an oxide of Bi, occurs with native gold at Berezov in Siberia. Bismuth-gold, Au,Bi is found in quartz at Maldon, Victoria.

Bismuth is also found with Sb in kobellite, hauchecornite, and chiviatite; and as Zr is especially found with Au it must also be found with Bi.

These mineralogical facts suggest the three combinations, each of which has a weight of 209, and which are connected with one another by the interrelationship of the elements in the following

S As Bi= Al Cu

C

:-

where AlCu-Zr, AIS=Co, SAs Ag, CuC=As, AgC = Sb.

Whilst the number 209 is exceedingly suitable, 207 is extremely unsuitable for explaining scientific facts with regard to Bi. Each of the pairs HPb, OsO, NaW, TeBr, SrSb, YSr, RuPd, make up a total weight of 207, but not one of them can be said to be connected with Bi. Also 207 equals 9×23, but the non-metallic nature of Bi could not be explained by this combination.

Since it has been proved that the actual atomic weight of an element is distinguished by its ability to co-ordinate mineralogical and chemical facts, it is clear that the evidence in favour of the number 209 is very great indeed. The experimental percentage error involved is not as great as that which is suspected by the International Committee in the cases of Zr and B.

The six elements Cu, As; Ag, Sb; Au, Bi, all of which are concerned in the above matter, are connected in the Periodic Table in the following

with calamine, an oxide of Zn and Si. Seamon (Dana) says that the calamine has been gradually crystallised out of the zinciferous clays, these having been first formed.

Zinc is found in the ashes of the yellow pansy growing in Rhenish Prussia in soil which contains Zn. The power of vegetable life has probably atomised the KCN in the plant and thus produced in the soil a deposit of Zn, the accumulation of ages.

Zn and KCN are both employed for dissolving silver, and their properties are similar in several other ways.

The integer 65 is nearer than 66 to the experimental atomic weight.

Cadmium (111) = Na,Zn.

Cd is found in smithsonite (ZnCO1). Greenockite (CdS) is found as a coating on sphalerite (ZnS).

Zn and Cd are the only elements, besides N, which according to this theory are exceptions to the Odd and Even Rule, the reason being that they contain N.

Caesium (131) = Na,K.

This element is found with Rb (Na2K), K and Na in rhodizite.

There is no combination of associated elements which is at all suitable for explaining the properties of Cs when 133 is taken as its atomic weight, but the above formula for the atomic weight 131 is exactly suitable in all respects, for it explains why Cs is similar to Rb and K, why it is generally a monad and sometimes a pentad as in CsI,, also why it cannot act as a non-metal, and it is in beautiful conformity with the mineralogical facts and also with the large generalisation with regard to Na.

In addition to all this the number 131 is confirmed as the atomic weight of Cs by the following independent investigation, which is only a small part of a very extensive and exact discovery of which a considerable portion has been sent to the Nobel Institute of Sweden. In the following Table, the first column gives the formula of the atom or molecule, the second column gives the relative volumes, the third the difference between the relative volumes of K and Na, the fourth the calculated S.G., the fifth the observed S.G., and the sixth gives the names of those scientists who obtained the exact, or, who were the nearest to the calculated S.G. The relative volumes are in every case calculated from specific gravities obtained at 15° C. or reduced to 15° C.

It is evident from the Table that atoms of K and Na in combination occupy exactly half the volume that they occupy respectively when uncombined. Therefore the volume of an atom of K in combination in the above eight molecules is 44 58/2=2229; and the volume of an atom of Na in combination is 237/2=1185. The S.G. of Cs is given as :