]\gt ; [ 237 ] VII .
Colours in Metal Glasses , in Metallic Filmw , and in Metallic Solutions .
\mdash ; II .
J. C. MAXWELL GARNETT .
by Professor LARMOR , Sec. R.S. Received May 15 , \mdash ; Read June 8 , 1905 .
CONTENTS .
Pages 1 .
Introduction . . . . .
237-239 2 .
Expressions for Ghe optical constants of media containing metal in amorphous or granular forms . . . . . . . . . .
239-242 3 .
Formulae applicable only when the volume proportion of metal is small . . . . . .
242-243 4 .
Calculated numerical value of the optical constants of metal glasses , \amp ; c. . . . . .
243-248 5 .
Diffusions of gold .
The nature and form of the suspended particles . . . . . .
248-255 6 .
Diffusions of silver .
The nature and form of the suspended particles . . . . .
255-259 7 .
Blue reflection from the stained face of silver glass . . . . . . .
269-263 8 .
Diffusions of copper .
The nature and form of the suspended particles . . . . .
263-265 9 .
Colouring effects of the radiation from radium , cathode rays , \amp ; c. . . . . .
10 .
Numerical values of the optical constants of media containing large proportions of certain metals . . . . . .
267-276 11 .
Colour changes caused by heating metal films 277-282 12 .
The exceptional case of beaten metal leaf .
282-283 13 .
CAREY " " allotropic\ldquo ; silver . . . . .
283-285 14 .
HERMANN VOGEL'S silver . . . . .
286-286 15 .
Allotropic forms of metal . . . . .
286-288 1 .
Introduction .
THIS paper is an extension of a previous memoir on the " " Colours in Metal Glasses and in Metallic Films\ldquo ; * ; it is concerned with the application of mathematical analysis , akin to that already there developed , to the explanation and coordination of the colours which certain metals are , under a great variety of circumstances , capable of causing .
* Phil. Trans , vol. 203 , pp. 385-420 .
( 393 .
) 11.10.05 From observations on gold and copper ruby glasses , it has been shown* that the first stage in the formation of a crystal of those metals is the small sphere ; and observations on the growth of sulphur crystals in , VOGELSANG arrived at the conclusion that the small sphere is always the first stage in the formation of a crystal .
He remarked , however , that it is by no means necessary that each of the small spheres , formed as crystallisation commences , should give rise to a separate clystal : the small spheres tend to coagulate , forming first rows and.then groups of other and more complicated shapes , until the crystal is ultimately formed .
To the intermediate bodies he gives the name of crystallites .
That the spherical form of the nascent crystal is governed by surface tension , was suggested in the former paper .
If this suggestion is corl'ect , we should expect that when the conditions are not the same in all directions , the spherical form of the nascent crystal will be replaced by an ellipsoidal form .
In particular , when a very thin film of amorphous metal is heated until the molecules are sufficiently free to allow crystallisation to commence , the nascent crystals may be expected to be spheroids of the planetary type , having their axes normal to the film .
Mr. G. T. BEILBYS has observed such spheroids in thin films of gold and silver .
Now it will appear below that metals a , re not only dichroic , exhibiting one colour by reflected light and , in thin films , another by transmitted light ; but that one and the same metal may , as its physical condition is altered , show a great variety of colours by reflected light , and a corresponding other series of tints by transmitted light .
The ultimate cause of all these colours is to be found in the structure of the molecule itself Juxtaposition , however , causes one molecule to affect the vibrations of another .
Thus consider a substance composed of molecules of a given metal separated from each by the aether or by any other non-absorbing medium : the " " effective period\ldquo ; of the molecule of such a substance is dependent on the geometrical arrange- meant and density of distribution of the molecules in question .
The optical properties of the substance will therefore depend on its microstructure .
The object of this paper is to obtain information concerning the ultramicroscopic structure of various metal glasses , colloidal solutions , and metallic films , by calculating optical properties corresponding to certain assumed microstructures , and by comparing the calculated properties with those observed . .
( pp. 388-392 ) .
When writing the former paper here cited I was unaware of VOGEoeANG'S work .
H. VOGELSANG , " " Sir les Cristallites ' Archives erlandaises , V. ( 1870 ) , p. 156 ; VI .
, p. VII .
1872 , pp. 38-385 .
, p. 392 .
The further suggestion there made that in the colourless gold glass there are tho molecules of gold present is , as will appear below , p. 261 , erroneous .
It is almost certain that in AQ colourless glass a gold salt is in solution , so that the heating has first to leduce the gold and then to allow the isolated molecules to run together into spheres .
S 'Hurter Memorial cture , ' Glasgow , 1903 , p. 46 .
As , for example , glass in a metal glass , or water in a colloidal solution .
ON COLOURS IN METAL GLASSES , ETC .
microstructures to be assumed are suggested by the preceding remarks on crystaUisation .
Calculations will be made for three types of microstructure , namely , ( 1 ) amorphous\mdash ; that in which the metal molecules are distributed at random ; ( 2 ) granular\mdash ; that in which the metal molecules are arranged in spherical groups ; ( 3 ) spicular*\mdash ; that in which these small spheres are replaced by oblate spheroids .
It will subsequently appear that when the surrounding non-absorbing medium is of refractive index unity , an amorphous and a granular microstructure produce the same colours .
In order to calculate the optical constants\mdash ; the refractive index and the coefficient of absorption\mdash ; which correspond to any given microstructure , it is necessary to know the values of the constants for some standard amorphous state of the metal .
Now BEILBY has shown that the process of polishing a metal surface causes the surface layer to " " flow\ldquo ; as a liquid , and thus the polished surface is that of the metal in the amorphous state .
It follows that the optical constants which we are to use as data for our calculations should , so far as possible , be those which have been determined by means of reflection from the polished surface of the metal iu its normal state according to DRUDE'S method , rather than those obtained by means of the light transmitted through thin prisms of the metal , after the method adopted by KUNDT .
2 .
Expressions for Optical Constcmts of containing Metad in Amorphous or Forms .
The optical properties of a homogeneous isol , ropic medium are determined wheu the values of the refractive index and the absorption coefficient , which correspond to light of every frequency , are known .
We proceed to obtain the values of and for a substance composed of molecules of metal embedded in an isotropic non-absorbing medium , the microstructure being amorphous .
Consider then a medium consisting of one substance , in solution in another , so that the molecules of each substance are distributed at random .
Let the number of molecules present per unit volume in the standard amorphous forms of A and be respectively and , and let number of molecules present per unit volume of the composite medium be and respectively .
We shall assume and to be constant throughout the medium ; or , more precisely , we assume that a very small compared with a wave-length of light , can be found such that , for all values of greater than , the number of molecules of A contained by a spherical surface situated wholly within the medium having a radius and its centre being * The calculations for a spicular microstructure are resorved for subsequent publication , see note p. 241 .
, Lord RAVLEIGH Royal Institution Polish , March , also holds the view that the process of polishing is a one .
'Ann .
Phys XXXIX .
( 1889 ) .
situated at any point , is independent of the position of that point ; thus depends only on : and similarly for C. Suppose that when electromagnetic waves traverse this medium , the moments of the average molecule of A and in the vicinity of the point are and .
Then and fc are both proportional to , the electric force exciting the average molecule* ; thus The polarisation of the compound medium is given by .
Writing now for in ths general equation we obtain so that .
But MAXWELL'S equations for the composite medium are curl curl where is the velocity of light in vacuo .
These may be written curl curl E , when we put .
( 1 ) .
If now we write ( 2 ) , then and are the refractive index and absorption coefficient of the composite medium .
But the same analysis will show that if and , where and are the optical constants of A and of , then and *Cf .
' Phil. Trans , vol. 203 , pp. 392 , 393 .
, equation ( 9 ) , p. 393 .
ON COLOURS IN METAL GLASSES , ETC .
these expressions in equation ( 1 ) we obtain If , now , we suppose that is a transparent isotropic substance of refractive index and that A is a metal , we have , by omitting the suffix A and putting ; or , ( 3 ) .
When is very small this equation becomes .
( 4 ) .
These equations give the optical constants of the metalliferous medium in the amorphous state .
When the microstructure is granular , these equations ( 3 ) and ( 4 ) are , as has been already shown , *replaced by ( 5 ) , so that when is small , ( 6 ) .
Comparison of equations ( 3 ) and ( 5 ) shows that the optical properties of a metalliferous medium , a given volume proportion of the metal , vary according as the metal is in small spheres or in a state of molecular subdivision , except when .
Thus when metal is in solution in water or glass the colour of the compound medium will change as crystallisation commences .
When , however , , the equations ( 3 ) and ( 5 ) both reduce to ( 7 ) .
It follows that the optical properties of a metal in a state in which its specific * Lo .
, equations ( 11 ) and ( 12 ) , p. 394 .
The mathematical treatment of the optical properties of media containing minute metallic ellipsoids , instead of the spheres which give the granular microstructure , is under consideration , but , with the exception of the case wherein the volume proportion , , of metal is small , it is not yet complete .
[ Note adckd 1st August , investigation of the general case ( any value of ) has now been completed .
The results for the case when is small which , when this memoir was communicated to the Royal Society , were given in S12 , have therefore been reserved for subsequent publication in a more complete form .
] VOL. COV.\mdash ; A. 2 I gravity has any known value , are unaltered by a change in the amorphous to granular .
* Or , again , Professor R. W. WOOD'S clouds of sodium vapour , for which nearJy , do not change colour as condensation commenoes .
3 .
Formuloe Applicable only when the Volume of Maal The volume proportion of metal present in all the coloured glasses and colloidal solutions which we shall discuss below is small We proceed to obtain , from equations ( 4 ) and ( 6 ) above , expressions for the optical constants of media , such as glass or water , holding in suspension metal in the amorphous and granular states .
Let denote the optical constants of the compound medium when the metal is in the amorphous state\mdash ; in true solution .
Then , replacing by in equation ( 4 ) , we have , say .
Equating real and imaginary parts , we find that ( 8 ) .
From ( 8 ) we have so that , neglecting , we obtain The corresponding values of , the optical constants when the metal is in the granular state , have already been obtained .
They are reproduced here for convenience of reference : thus ( 10 ) , where ( 11 ) .
OLASSES , ETC .
243 traverses a thickness of a metalliferous light is reduced in the proportion* or ' as the metal is in true solution or in spherical aggregates .
Suppose now that two kinds of monochromatic light , of wave-lenghs and , are , traversing a distance in an absorbing medium , reduced in intensity by and respectively .
Then the absorbing medium reduces the proportion of the intenities of the two kinds of light in the ratio , which is a function of .
Thus the tint of a coloured medium , viewed by transmitted light , depends on its thickness .
We shaU , however , speak of two absorbing media as possessing the same colour when , whatever be the values of and , the ratio is the same for either medium ; for , if suitable thicknesses of such media be chosen , the light transmitted by them wiJJ be of precisely the same tint .
Since , therefore , it appears from equations ( 8 ) and ( 9 ) above that the ratio is independent of , it follows that a molecularly subdivided metal produces the same coloration ( by transmitted light ) in all non-dispersive transparent isotropic " " solvents irrespective of their refractive indices .
Thus , neglecting the small dispersion , borax bead and a glass bead , each containing a metal in solution , will be of the same colour ; but so soon as crystallisation of the metal begins , so that part of the metal is in small spheres , the beads will cease to be of the same colour , since the ratio is not independent of 4 .
Numerical Values of Optical Constants of Metal Glasses , Consider any transparent isotropic non-dispersive medium of refractive index containing either molecules or smaU spheres of a metal , the optical constants of which , for light of wave-length , are and , the particles of metal being so tributed that there are many of them to a wave-length of light .
The " " absorptions and , of the compound medium can be easily determined by means of equations ( 9 ) and ( 10 ) , when the values of , are known for light of wave-length .
These values can be calculated by means of equations ( 11 ) *Cf .
'Phil .
Trans , 1904 , p. 395 .
Thus , for example , a thin sheet of gold ruby glass will appear pink , a considerable amount of blue light being transmitted , whereas a thick sheet of the same glass will appear deep red , almost like a copper ruby .
Again , by increasing the depth of a silver stain on glass , we get all gradations in colour canary through amber to red .
This , then , must be the oolour of the vapour of the metal provided that the molecules are monatomic , or , at least , do not dissociate when the metal is vapolnised .
We shall term it the ' vapour colour and ( 8 ) , when the quantities and have been determined for the light in queition by direct experiment on the metal in the standard amorphous state .
R. S. MINOB* has made such experimental determinations , for various kinds of monochromatic light , from the polished surfaces of silver and copper .
His values of and for silver and copper , together with the numerical values of , and of certain other functions as calculated for various values of , are shown in Tables II .
and IIL The values of and , for and only , have also been determined by DRUDr from the polished surface of gold ; but values of and for other values of have been obtained by HAGEN and from gold prisms deposited on glass .
Since however the state of the metal in the prisms is not known , these latter values cannot be depended upon for our purpose ; S but as a rough estimate of the values of \amp ; c. , may be formed by their means , the numerical values of these quantities have been calculated ; the results are shown , together with all the observed values of and , in Table L The wave-length of the light will throughout be measured in thousandths of a milimetre .
All the calculated numbers given in Tables I. , II .
, and III .
have been carefully checked with a " " Brunsviga\ldquo ; machine .
I believe that in no case does an error amounting to 1 per cent. survive in these Tables , which must accordingly supersede those given in the former communication .
* R. S. MINOR , 'Ann .
der Phys vol. X. , 1903 .
'P .
DRUDE , 'Physik .
Zeitschr January , 1900 .
RUBENS , 'Wied .
Ann vol. , 1889 .
S It will appear in the sequel that on this account the optical properties of silver and of colloidal solutions of silver are much more accurately represented by our calculations than is the case with gold ruby glasses and colloidal solutions of goltL Lo .
ait .
, p. 396 .
In fig. 2 the graph .
of for glass , shown by a broken line.in fig. 1 , is again represented ; but here on such a scale as to have the same ordinate at as that of the graphs of the observed absorptions of Au ( A ) and Au ( B ) .
The calculated and observed curves resemble one another in having a minimum in the red and a maximum in the green , although the calculated maximum occurs at about , while the observed maximum falls at 533 .
Also both calculated and observed absorptions fall from green to blue , while the dotted curve in fig. 1 shows that the absorptions produced by molecularly divided gold will increase ffom green to blue , having a maximum at about .
These results then , so far as they go , are in accordance with the suggestion , put forward in the former memoir , that the colouring agent of gold ruby glass consists primarily of diffused spheres of gold , although some discrete molecules may also be present .
The following is the evidence which has accumulated to show that a gold ruby glass contains minute spheres 01 gold , many to a wave-length of light , and that it is to these small gold spheres that the pink colour of the glass is primarily ( 1 ) There are particles , presumably of gold , visible in all specimens of gold ruby glass in which the colour has been deveIoped .
* ( 2 ) Whenever these palticles are of diameter less than centim. they are spherical in shape .
SIEDENTOPF and ZSIGMONDY state1 : " " It is only in the case of glasses that the particles are so dense that they cannot be fully separated under the microscope In other words , whnever there are many small spheres to a wave-length of light , the glass is ruby .
( 4 ) We have just seen that , within the limits of experimental error , this ruby colour is that which would be produced by small spheres ( but not by molecules ) of gold , many to a wave-length , embedded in the glass .
( 5 ) The polarisation of the cone of light emittod by the particles in the path of a beam of white light traversing any of the three ruby glasses examined by SIEDENTOPF and ZSXGmONDYS is that which would be possessed by the cone of light emitted by small spheres of metal imbedded in the glass .
Further , the colour of the cone of light in the case of these three glasses was .
while it has been shown in the former that the intensity of light of wave-length emitted by small spheres of gold embedded in the glass is proportional to , and , according to Table I. above , this expression for gold spheres in a glass of refractive index has a maximum in the neighbourhood of , i.e. , in the We conclude then that the colouring agent of gold ruby is metallic gold , the major portion of which is in the form of small spheres .
The irregular blue and purple which often appear , instead of the ruby at which the glass manufacturer aims , can be explained as indicated in the appendix to the former and ZSIGMONDY , ' Ann. der Phys January , 1903 . .
former paper , ' Phil. Trans , 1904 , p. 301 . .
cit. , p. 27 .
S See their table reproduced at p. $97 of the former paper and discussion following it on pp. 398-401 Phil. Trans , p. 400 .
Not aurous oxide , as stated in the text-books on glass making .
ON COLOURS IN METAL GLASSES , ETC .
seen that if gold glass when first made in the furnace be cooled , the glass remains colourless .
In order to obtain ruby glass , the molten glass must be left in the annealing oven and maintained at a high temperature for three days .
If the glass is too violently heated or is kept too long at a high it becomes turbid , reflects brown light , and develops first an amethystine and then a blue tint by transmitted light .
But it now appears that the gold cannot , as previously suggested , be in solution in the colourless glass when first heated ; for if metallic gold were in true solution in the glass it would have the vapour colour indicated by the dotted curve in fig. 1 .
The gold must therefore be gradually reduced during the annealing process .
So long as the glass remains hot enough to admit of molecular movement , the molecules of gold go together to form spheres , and these small spheres tend coagulate into crystaUites .
If the glass coo before the coagulation of the small spheres , a gold ruby glass is obtained .
If , however , some of the small spheres have coagulated into crystallites , the density of which exceeds of that of normal gold , S these crystallites will reflect light which is poedominantly yellow or red The glass will thus reflect brownish light ; and since the more refrangible rays are less reflect , ed than those of longer wave-length , the red end of the absorption curve will , owing to the crystallites , be raised relative to the blue .
The glass will thus appear bluer than when no coagulation has occurred .
Further , as these crystallites may be of dimensions comparable with a wave-length of light , they destroy the optical continuity of the medium and produce turbidity .
Now the blue colour of a gold glass is always associated with turbidity and a brown appearance by reflected light , so that the formation of crystallites of gold in the glass accounts for the irregular blue and purple colours which gold glass sometimes exhibits .
Diffusions of gold particles in water\mdash ; the ' ' colloidal solutions\ldquo ; of gold\mdash ; have been prepared by ZsIGMONDY , and STOEKL and VANINOn , who , p. 392 .
, p. 392 .
In the case of copper ruby glasses the process continues until actual crystals of copper are formed , but I have not seen gold crystals in a glass , although it is probable that they are occasionally formed .
S KIRCnNER and ZSIGMONDY ( ' Ann. der Phys XIIL , 1904 , p. estimate that a clump of gold particles in a blue gold-gelatine preparation contains at least 50 per cent. of gold .
See below , p. 254 .
See fig. 12 , below .
The blue and violet colours of the glasses and in SImDENTOPF and ZSIGMONDY'S table Phil. Trans , as well as the red , yellow and brown colours of the cone of light emitted by them , are thus explained .
STOKES ( Royal Institution Lecture .
1864 , 'Collected Works , ' vol. IV .
, p. , without entering into the question why gold glass ordinarily transmits pink light , says that .
it being the property of gold to transmit bluish light , the metallic gold in suspension causes the blue appearance .
FARADAY , Bakerian Lecture for 1857 , printed in 'Phil .
Trans. ' for , and reprinted in his ' Researches in Chemistry and Physioe .
' Referenoes will be made to the pages of the reprint .
ZSIGMONDY , ' LIEB .
Ann $01 ( 1898 ) , p. 29 , and ' Zeitschr .
Electrochom vol. IV .
, p. STOEKL and VANINO , 'Zoitschr . .
Phys. Chem XXX .
, p. 98 : precipitated the gold from its chloride by means of various reducing agents ; and by BREDIG* and later by EHRENHAFT , who used a gold terminal for an electric arc which was caused to spark under water .
All these preparations exhibited a gradual change in colour ffom red through purple to blue ; this change was greatly accelerated by the introduction of a trace of salt into the water .
gives the absorption curves of a number of ' ' solutions " " of gold .
STOEKL and VANINOS measured the absorptions of a red suspension containing a known volume proportion of gold .
Lastly , EHRENHAFr has made careful measurements of the absorptions of " " colloidal\ldquo ; gold .
The curves plotted ffom his measurements of the red " " solutions\ldquo ; resemble the continuous curve shown in fig. 1 .
Again , EHRENHAFT statesl that the absorption band of a gold suspension which possessed a beautiful red colour began at and attained a maximum at , while the solution was almost transparent in the ultra-violet .
Now the maximum of the calculated absorption curve for spheres of gold in water occurs at Again , the dotted curve in fig. 1 , which will represent the absorption produced by a true solution of gold , does not sufficiently agree with the measured absorptions to admit of the gold being in true solution in the water .
These results suggest that the coloration is due to diffused spheres of gold , although some discrete gold molecules may also be present .
BREDIG , ' Zeitschr . .
Phys. Chem XXXIL , p. 127 .
F. ExRENHArr , 'Ann .
der Phys XI .
( 1903 ) , p. 489 .
ZSIGMONDY , 'LIEB .
Ann vol. 301 ( 1898 ) , pp. 46-48 .
S , p. 108 .
For a discussion of their results see below ( footnote , p. 253 ) .
Ixe .
, pp. 505 , 506 .
Cf .
, table given , , p. 607 .
Thus the differences in wave-length between the observed maximum absorption of gold ruby glass and of the calculated maximum for gold spheres in ghss , between the observed maximum for colloidal gold and the calculated maximum for gold spheres in water , are respectively 017 and 013 , and these differences are of the same size .
EHRENHAFr also supposed that the gold was present in the form of small spheres ; but he proceeded to define the average size of these spheres ( and also of those of Ag , Pt , \amp ; c. , in the ' colloidal\ldquo ; solutions of these metals ) by means of J. J. THONSON'S equation connecting the radius of a conducting sphere with the wave-length corresponding to the free periods of its vibration , this wave-length being assumed to be that of the absorption maximum .
KIRCHNER and ZSIGXONDY ( Ann. der Phys 1904 , p. 576 ) , however , point out that there is no between size of particles and the absorption of light produced by them , and this we have seen to be the case , provided there are many particles to a wave-length ; also the very small size spherical , their average diameters wouYd of the particles of gold , the gold content of which ZSIGMONDY measured would require the absorption maximum to be in the ultra-violet .
KIRCHNER and ZSIGMONDY add that it would only be possible to get a large enough linear dimension.to .
give a free period if the particles were not dimensional , and they conclude therefore that the gold particles must be in the form of leaves or of rods ; but they do not reconcile such a form with the polarisation and green colour of the cone of light emitted by the smaller .
Since , however , we find that the small-sphere hypothesis accounts for the rved phenomena , we must agree with EHRENHAFr .
that the particles are spherical , although we cannot admit thbt the average diameter of spheres is correlated to the wave-length of the light most absorbed .
ON COLOURS IN METAL GLASSES , ETC .
the same time explained .
The following quotations must suffice to describe the phenomena in question .
FARADAY observes that " " A gradual change goes on amongst the particles diffused through these fluids , especially in the cases where the gold is apparently abundant .
It appears to of an aggregation .
Fluids at first clear , or almost clear , to ordinary observation , buome being left to stand for a few days , a deposit \ldquo ; When common salt , or any other substance which dissociates in water , was added to the fluid " " . .
The salt diffused gradually through the whole , first turning the gold it came in contact with blue , and then causing its precipitation .
" " Such results would seem to show that this blue gold is aggregated gold , i.e. , gold in larger particles than before.\ldquo ; Again " " The supernatant fluid in specimens that had stood long and deposited was always ruby .
there was every reason to believe that the gold was there in separate particles , and that such specimens afforded cases of extreme division.\ldquo ; S Observations made by subsequent physicists agree with those of FARADAY .
Thus ZSIGNONDY writes " " In every case the bright red colour [ of suspensions of gold in water ] changed to blue on the addition of salt ; and decoloration of the upper part of the liquid showed that precipitation has then begun Again , STOEKL and VANINO , who examined a large number of suspensions of gold in water prepared by many different methods , state that " " When the particles [ of gold ] are very small . .
fluid appears red-yellow , ruby-rod .
When , however , the particles increase in size , the red and yellow rays are quite cut off and the transmitted light consists only of blue and violet rays , the fluid appearing blue-violet.\ldquo ; lI 'Researches in Physics and Chemistry , ' p. We may suppose that by friction against the water the gold spheres obtain that negative change which ZSIGMONDY ( ' LIEB .
Ann vol. 301 , p. 36 ) found that they possess .
The mutual repulsion of these like charges prevents the spheres from coagulating and thus keeps the gold in suspension in the water .
But when an electrolyte is introduced into the fluid , the positive ions discharge the gold spheres , so that coagulation and precipitation result .
FARADAY , M. cit. , p. 420 .
S FARADAY , , p. 418 .
ZSIGMONDY , ' LIEB .
Ann ( 1898 ) , p. 34 .
' Zeitschr . .
Phys. Chem XXX .
( 1899 ) , p. 108 .
As already stated , STOEKL and VANINO measured the absorption of light , for six different values of by a suspension contaiming a known volume proportion of gold .
Using their value of to determine the scale of the continuous curve in fig. 1 , and comparing the values of so obtained with Finally , KIRCHNER and that in a gold .
A given ( generally large ) number of particles which diffract green light brought together by the addition of an electrolyte into a single particle which light with much greater intensity than its components .
With this , uniting of partioles in the colour of the fluid from red to blue.\ldquo ; We have already shown that , theoretically , the coagulation of the small spheoes of gold should produce a colour change in the fluid , from red through purple.to blue ; and the above quotations have indicated that coagulation accompanies the change of colour .
But that the coagulation takes place in the manner assumed for the purposes of the theory has been shown by KIRCHNER and ZSIGMONDY , who prepared suspensions of gold in gelatine , some of which preparations were red when wet , and changed to blue on being dried , at the same time developing a gold-bronze reflection .
Now these dry blue membranes contained a number of clumps , each comp of hundreds of ultra-microscopic resonatorsS ( small spheres ) ; and these clumps were comparable in size with a wave-length of light , being directly visible when:examined with a numerical aperture of : they would therefore be capable of reflecting light .
Further , the change of colour to blue was most marked in those preparations in which the individual clumps were most dense and it appears from below that the selective absorption of red and yellow light by a gold crystallite is greater the greater its density .
The theoretical explanation of the change to blue requires the rays of lower refrangibility to be stopped by reflections from crystallites , and this requirement is thus satisfied .
the absorption curve obtained from STOEKL VANINO'S observations , we find that the observed curve lies below the calculated curve , except for red light .
But STOEKL and VANINO record that the obs6ved fluid had a yellowish reflection , so that large particles ( crystallites ) must have been present in it and the presence of these crystallites requires the volume of gold , which per unit volume of the liquid is in the form of small spheres , to be less than the total volume )roportion .
This diminishes the absorptions throughout the spectrum .
But the volume proportion of crystallites produces absorption which is much greater for the red and yellow than for the green and blue rays .
The superposition of the absorptions produced by and by would thus produce an absorption curve in accordance with that observed .
* The aggregate may be supposed to be comparable in size with a wave-length of light ; the intensity of the light reflected from it would thus be proportional to the square of its diameter , while the intoIl of the light diffracted by the small spheres is proportional only to the sixih power of their diameters .
KIRCHNER and ZSIGMONDY , loc. cit. , p. 592 . .
cit. , p. 589 .
S KIRCHNER and ZSIGMONDY , .
cit. , p. 576 . .
cit. , p. 677 .
A similar explanation possibly applies to the fact that when light , transmitted membrane containing gold in suspension , is polarised in the direction red , but when the incident light is polarised in a perpendicular clumps being comparable with a wave-length in the .
ON COLOUBS IN METAL , ETC .
In conclusion , we remark that mos " " colloidal solutions\ldquo ; of gold , even those which are of a ruby , contain crystallites in addition to the spheres to which the colour is primarily due .
Thus FARADAY could detect the green " " cone of light which indicates the absence of large aggregations , only in those liquids which had been cleared by prolonged precipitation and frequent decantation ; and STOEKL and VANINO found that all the gold suspensions which they examined showed a yellowish reflection .
A small ber of the large aggregations may , however , cause the cone oflight to appear yellow or red without appreciably altering the colour of the transmitted light .
For , whereas the intensity of the ( green ) light emitted by a small sphere is proportional to the sixth power of the diameter , the intensity of the ( brown ) light reflected from a gold crystallite is proportional to the square of its linear dimensions .
Gold solutions prepared chemically appear , however , to be freer from aggregated gold than are those prepared by BREDIG'S method .
* 6 .
Diffusions of Silver .
The Nature Form of the Suspended Partictes .
We proceed to consider the absorption of light produced .
by diffused particles of silver .
The values of for , and given in Table II .
are plotted in fig. 3 , the positions of the maximum of each curve being determined as in the case of gold .
Since ( cf. above , S4 ) values of and for silver were all determined from the polished surface of the metal , these curves should represent the absorption produced by diffused spheres of silver in glass , in water , and in vacuo , with only a small error .
The dotted curve in fig. 1 , which represents the absorptions of diffused molecules of silver vacuo ( and , on different scales , in other non-absorbing and non-dispersive media ) , shows that the silver molecule has a free period corresponding to , about .
The existence of this free period is possibly responsible for the sensitiveness of silver salts to ultra-violet light .
In fig. 4 the graphs of for glasses of refractive indices and are shown on such a as to have the same ordinate at the line as the graph of for the measured glass Ag ( B ) , of which the measured refractive index at the line was .
The measured curve resembles those calculated , following them very closely from to , and having a maximum for a value of intermediate between those values of which correspond to the maxima of the two calculated curves .
This close approximation of the observed absorptions to those ZSIGMONDY , ' Zeitschr . .
Electrochem p. 547 .
BREDIG'S remark , that his gold solutions were blue red , points to the same conclusion .
These curves show that in each case the absorption is less for red than for yellow .
This is contrary to statements mad6 in the previous paper ( 100 .
cit. , pp. 399 and ; the errors therein made were due to miscalculation for silver Table .
The cause of the depression of the observed maximum below those calculated is doubtless to be found in the fact , which Lord RAYLEIGH has called attention in a recent lecture at the loyal stitution , that ON COLOURS IN METAL GLASSES , ETC .
deulated suggests that the colouring agent of the yellow silver glass consists primarily of diffused spheres of silver .
Since discrete silver molecules would produce an tion maximum at , no more than a comparatively small amount of silver can be present nolecularly subdivided condition .
The conclusion that silver glass owes its colour to diffused spheres of silver will be verified in the following section .
The absorption spectra of some colloidal solutions of silver , prepared by BREDIG'S method , *have been measured by EHRENHAFT .
The continuous .
curve shown in fig. 3 , representing the calculated absorptions of a diffusion of silver spheres in water , is of the same form as that which , according to EHRENHAFP'S measurements , represents the absorption of visible light by a colloidal solution of silver .
Using ultra-violet light , he further found that a brown colloidal solution of silver , examined before coagulation seriously affected its colour , showed an absorption band which began at and attained a maximum at , while the fluid was again quite transparent at .
Except for the fact that the maximum ordinate of the calculated curve for silver spheres in water is at instead of at , the above observations admirably describe the continuous curve shown in fig. 3 .
Since the dotted curve given in that figure shows a maximum at , and the absorption band does not begin until , about , the colour of the colloidal\ldquo ; solution is not that which would be exhibited by a suspension of discrete silver molecules , i.e. , by a true solution .
We conclude , therefore , that the silver in a " " colloidal\ldquo ; solution is present in the form of small spheres ; discrete molecules may , however , also be present , and , as indicated above in the case of gold , -prepared by BREDIG'S method , probably ako crystalites , the number and size of which will increase with the age of the solution .
That the silver in a colloidal solution is in the form of small spheres is further shown by an experiment of BARUS and SCHNEIDER , measured the refractive index of such a fluid .
Their results are given in the following table , S in which represents the measured refractive index:\mdash ; the spectrum formed by the light which has traversed the glass is not quite pure , so that that image of the slit which should be illuminated only by light of wave-length , say , , is also , owing to reflections from dust particles , \amp ; c. , illuminated by light of other wave-lengths which has experienced a less absorption .
* BREDIG , ' Zeitschr .
Electrochemie , ' 4 , pp. 614 , 647 .
EHRENHAFr , loc. cit. , p. 606 .
BARUS and SCHNmIDER on " " The Nature of Colloidal Solutions 'Zeitschr . .
Phys. Chem p. 278 .
S Tabelle 6 , .
cit. , p. 296 .
VOL. CCV.\mdash ; A. 2 ON COLOURS IN METAL , ETC .
requires that practically the whole of the silver must.have been in silspension in the form of small spheres .
* Once more , CAREY .
prepared suspensions of silver in water by precipitating the silver from the nitrate by means of a mixture of ferrous sulphate and sodic citrate .
He describes how , after careful washing , the silver frequently " " dissolved forming a liquid which varied from red to yellow and was generally blood red ; he adds : " " On one occasion the substance was obtained in a crystalline form .
Some crude red solution had been set aside in a corked vial .
Some weeks after the solution had become decoloured with crystalline deposit on the bottom .
The bottle was carefully broken ; the deposit , examined by a lens , consisted of short black needles and tbin prisms If , then , the diffused particles of silver when aggregated and precipitated had become crystalline , they must before have been in the form of nascent crystals , and for gold and for all the substances examined by VOGELSANG , S such nascent crystals were spherical .
7 .
Blue Reflection from the Stained Face of Silver Glass .
When clear glass is flashed with silver glass , or when a clear glass is so stained on one face with silver that the volume proportion of silver does not gradualy diminish to zero as we proceed inwards from the stained faoe , but that the stained region ends abruptly , a blue reflection from the interface can be observed if the glass is held with the stained face away from the eye .
No blue reflection can be seen from the air-glass interface when the stain is held towards the eye .
STOKES observed this blue reflection , and stated that the interface presented the appearance of being coated with a fine blue powder We.proceed to examine whether the presence of small spheres of silver , which has been shown to account for the colour of the light transmitted by silver glass , will also account for this blue reflection .
Consider , then , polarised light travelling in a medium of refractive index and directly incident on surface , , of an absorbing medium whose optical constants are and , where .
Then we may take as the electric and magnetic vectors for * BARUS and SCHNEIDER ( kc .
cit. , p. 297 ) make the following comment on their experiment : " " KUNDT has found for normal metallic silver a refractive index of about .
It would , therefore , be expected that the presence of the silver would diminish the refractive index of the water .
It is by no means denied that it might be possible to explain the normal refractive indices of the above table in accordance with MAXWELL'S Theory of Light The investigation in the text attempts to give such an explanation .
CAREY LEA , ' Amer .
Journal of Science , ' 1889 , and ' Phil. Mag 1891 .
Cf .
above , p. 243 , especially second footnot S .
above , S1 .
STOKES , ' Collected Works , ' III .
, p. 316 .
Incident ligh :\mdash ; Reflected light:\mdash ; Bexp Light inside absorbing medium:\mdash ; Making X and continuous at , we have .
Hence Taking the square of the modulus , we have , for the value of the ratio of the intensity of the reflected light to that of the incident light , ( 12 ) .
If , now , the absorbing medium consist of minute spheres of metal embedded in a transparent medium of refractive index , we have equations ( 10 ) , namely , , ( 10 ) .
Substituting these values of and of in ( 12 ) we obtain ( 13 ) , in which : powers of higher than the second have been neglected .
lppose first .that , so that we consider the reflection at the front face of the stained glass .
Omitting powers of except the lowest which occur , we then have from ( 13 ) ( 14 ) .
It appears from equation ( 14 ) that light is reflected from the stained glass almost as if the stain did not exist , the effect of the stain being slightly increase the reflection of those colours ( in the blue ) for which,.accord.ing to Tab , le II is greatest .
ON COLOURS IN METAL GLASSES , ETC .
Now , however , suppose that , so that the light is reflected at the interface between colourless glass and the same glass containing small spheres of metal .
Neglecting , equation ( 13 ) then reduces to ( 15 ) .
Since this expression for contains no large constant term , the light from the interface will in this case be highly coloured in the case of those metals for which varies greatly for different values of If , however , the absorbing medium contain molecularly divided metal , equations ( 10 ) are replaced by . . .
( 11 ) .
Replacing and in equations ( 14 ) and ( 15 ) by and respectively , we obtain , as the intensities of the light reflect from the front face of the stain and from the interface respectively , ( 16 ) , , ( 17 ) .
As before , it appears that when the stain is held towards the eye the reflection is almost as if the stain were not there ; while when the stained face is away from the eye , the reflection is coloured .
Sir WILLTAM ABNEY has kindly measured for me the intensities of light reflected from the interface between the unstained and stained regions of one of STOKES ' specimens of silver glass .
The values of are given in the following table : TABLE VI.\mdash ; Blue Reflection from Silver Glass .
Measured Value of at D Lines ON COLOURS IN METAL GLASSES , ETC .
It.appears that while the graphs of and Ro widely differ , the positions of the respective maxima falling near and respectively , the graph of closely resembles that of the maxima of and of occurring at almost the same value of .
We conclude that the presence of small spheres of silver throughout the stained region of the glass will account for the blue reflection ; and we thus confirm : : .
the view , to which absorption phenomena led us , that silver glass consists of a suspension of small spheres of silver in a colourless glass .
Before leaving the consideration of the blue reflection from silver glass , it may be noticed that the light is not reflected as from a plane interface between glass and silver glass .
Thus when the source of light is an electric arc , the blue colour is clearly discernible by an observer whose eye is not in the straight line determined by the ordinary law of reflection .
This effect is due to the irregularity of the interface , the silver not having penetrated the glass to a uniform depth .
An alternative explanation , however , suggests itself , \mdash ; the blue colour might be due to independent radiation from discrete spheres ( or molecules ) of silver so far apart as not to form an opticaly homogeneous medium .
The intensity of the emitted light would then be proportional to case spheres , in the case of molecules ) .
Further , the blue colour would be equally visible if the light illuminating the discrete spheres ( or molecules ) entered the silver glass from the air side or the clear glass side ; and this is not the case .
It is of interest to notice that while each individual sphere in glass radiates out light of an intensity proportional to , a surface separating a glass , containing many of the spheres to a wave-length of light , from a region of the same glass in which no spheres are present , reflects light with an intensity proportional to This is due to the fact that the number of spheres ( on the reflecting surface ) , the phase of the forced vibrations of which lies at any instant between given limits , is proportional to ; so that the intensity of the reflected light is proportional to times the intensity of the light emitted by a singJe sphere .
8 .
Diffusions Copper .
The Nature of the Suspended Particles .
We proceed to discuss the colours produced by diffused particles of copper in order to discover the cause of colour of copper ruby glass .
The values of the expression for , and given in Table III .
, are plotted in fig. 6 , the maxima being determined as in the case of fig. 1 ( cf. S5 above ) .
As in the case of silver , these curves should fairly accurately represent the absorptions produced by copper spheres in glass , in glass , and by copper spheres or molecules * The fact that increases from yellow to red , while diminishes in the same range , would be accounted for if the black paper with which Sir WILLIAM ABNEY backed the stained face of the glass reflected 2 per cent. oi the light incident on .
Further experiments are to be made on this .
ON COLOURS IN METAL GLASSES , ETC .
( oopper vapour ) in vcecuo respectively .
The absorption band in the yellow green shown by the top two curves in fig. 6 was observed by STOKES in the spectrum of a copper ruby glass .
In fig. 7 the graphs of for glass and of are reproduced ffom fig. 6 on such a scale as to have the same ordinate at the line as that possessed by the continuous curve which has been fitted to the plots of the measured absorption of the glass Cu ( X ) .
The curves in fig. 7 all have a minimum in the red or infra red and a maximum in the yellow-green ; but while that which represents the absorptions of spheres in glass has a secondary maximum near , the dotted curve shows that the absorption of copper molecules in glass continues to increase till .
Also the maximum in the yellow green for the " " sphere\ldquo ; curve occurs for approximately the same value of as corresponds to the maximum observed absorption ; while the value of at the maximum of the dotted ( ' ' molecule curve is about less , and the latter maximum is much less marked than are the former two .
Finally , the last readings obtained for in the violet indicate that the continuous ( observed ) curve rapidly approaches a maximum near We conclude that copper ruby glass is coloured by metallic copper , and that the greater part of the copper is present in .
the form of small spheres , although some probably remains in the form of discrete molecules .
The manufacture of copper ruby glass closely resembles that of gold ruby .
S Like gold ruby , the copper ruby glass becomes turbid if kept too long at a high temperature .
This turbidity is also probably due to the formation of crystalites by the coagulation of small spheres , since , when the conditions necessary for the development of turbidity are long maintained , actual crystals , apparently of copper , are formed in the glass 9 .
Colouring Effects of the Radiation from Radium , Cathode It has long been well known that cathode rays produce a blue-violet coloration in soda glass .
Soda glass tubes , after containing the emanation from radium , show the * The graph of for can , by increasing all the ordinates in the proper constant proportion , be changed into the graph of for any value of .
Cf .
S3 above .
STOKES Math. Phys. Papers , supposed that the colouring agent was suboxide of copper .
The blue colour exhibited by overheated specimens of the glass is probably caused by the coagulation of the small spheres into crystallites and crystals which reflect out the red light .
Measurements will have to be made with ultra-violet light in order to determine how much copper remains in the molecularly subdivided condition .
S Cf .
above , S5 , p. 251 .
Cf .
'Phil .
Trans ( 1904 ) , p. 392 .
Some of the crystalline glazes made by Mr. BURTON at PILKINGTON'S the works exhibit the same effect .
I have seen a pot with a copper glaze in parts of which the copper was apparently reduced , for in passing from the colourless glaze ( where the copper was not reduced ) into legions where the reduction had been effected , a deep red ( copper ruby ) was first reached ; that colour increased in intensity until , in central portions of the region , crystals , htly of copper , could be seen .
VOL same colour , and crystals of each salt acquire under cathode rays a beautiful violet tint .
* Experiment has also shown that exposure to the emanation from radium gives to gold glass a ruby colour , to silver glass a yellow colour , and to potash glass a brown colour .
Now we have seen that a molecularly subdivided metal possesses the same colour by transmitted light whatever be the nature of the surrounding transparent medium , supposed non-dispersive and isotropic .
This colour may be called the vapour-colour of the metal .
It has further appeared that although the transmitted colour of a metal subdivided into small spheres , many to a wave-length of light , does depend on the refractive index of the medium in which the small spheres are " " embedded yet this colour approaches to the vapour-colour as approximates to unity .
As is shown by the dotted curve in fig. 1 , the vapour-colour of gold must be red .
The colour of glass containing molecularly distributed gold is thus red , although when the gold is collected into spheres the glass is pink .
Similarly , reference to the relative values of in Table II .
shows that the vapour-colour of silver is yellow .
Glass coloured by small spheres of silver is also yellow .
Again , Professor R. W. WOOD showed to the British AssociationS in Cambridge that the vapour-colour of sodium is violet , this colour being due to the absorption at the lines .
This violet colour is also produced at the cathode in the electrolysis of sodium chloride the molecules of sodium formed at the cathode being distributed throughout the water in its neighbourhood and giving rise to the vapour-colour .
Analogy with the cases of gold and of silver indicates that small spheres of sodium would produce in glass a colour not greatly different from the vapour-colour produced by the molecularly subdivided metal .
Thus the colours developed in gold , silver , or soda glass by the radiation from the emanation from radium are approximately the same as the colours which would be given to the glass by the presence of the reduced metal , either molecularly divided or in small spheres ( nascent crystals ) .
It is therefore.very probable that the metal in the glass is reduced by the action of the radiation .
This view finds considerable support in the discovery of VILLARD , that cathode rays exert a reducing action , as well as from the fact , already cited , that ELSTER and GEITEL found the salts of the alkali metals , which had been coloured by posure to cathode rays , to exhibit photo-electric effects as if they contained traces of the free metal .
GOLDSTEIN , ' WIED .
Ann liv .
, p. 371 , 1898 .
i\amp ; ante , 1 , .
243 .
Or yellow ; see the second footnote , p. 243 .
S August , 1904 .
Cf .
J. J. THOnSON , ' Conduction of Electricity through Gases , ' pp. 495 , 496 .
BUNSEN found that common salt , after heating to about C. , exhibited a violet colour , due apparently to .
the reduced metal , although BUNSEN suggested a subchloride , 'Journal de Phys Series , VIII .
, p. 140 , 1899 .
' Phil. Trans , 1904 , p. 400 , ON COLOURS IN METAL GLASSES , ETC .
Sir WILLIAM RAMSAY , when I first called his attention to the explanation of the ooloration of glass by radium which is afforded by supposing the radiation to reduce metal in the glass , suggested that the reduction might be effected by the discharge of free ions of the metal .
Since that time the further evidence that has accumulated seems to favour the truth of this theory .
Thus , as all the colour-changes from pink to blue exhibited by gold glass can be imitated with suspensions of gold in water , the glass appears to behave as a liquid , although a very viscous one ; and it seems , therefore , reasonable to suppose that the salt of a metal which will dissociate in water will dissociate also in glass .
As an alternative hypothesis , we might suppose the compound molecules broken up by the rays .
But , were this the case , the -rays would be far more efficient than the in producing the colour .
And this is not true ; for the coloration produced in the splinters of gold and silver glass , as well as in soda and potash glasses , are not , apparently , stronger on the sides of the glass , but seem to be of uniform strength throughout .
From this it appears that the -rays are alone capable of producing the colour .
This is in accord with the former hypothesis .
For the ions of the metal in the glass would be positively charged , and their discharge by the negatively-charged particles ( or cathode rays ) would change them into molecules\mdash ; just as the sodium ions in the electrolysis of common salt are discharged at the cathode , and thus are transformed into molecules of sodium , impa.rting a violet colour to the water and capable of forming caustic soda .
It appears , therefore , possible that all glasses contain free ions of metal , and that it is by the discharge of these ions , and consequent reduction of the metal , that cathode and Becquerel rays are able to produce coloration in them .
10 .
Values of the Optical Constants of Media containing Large Volume Proportions of Metals .
The preceding sections of this paper have treated only of the optical properties of those media for which the volume proportion , , of metal is very small .
The consideration of media in which may have any value up to unity will now , however , be resumed , in order to discover what may be the physical explanation of those colours and changes of colour which FARADAY , *BEILBy , and others have found to be exhibited by thin metallic films .
fn S 11 of the former communication the question whether films built up of small spheres of silver or of gold would , for any given volume proportion of metal , transmit red or yellow light more easily , was discussed , and the conclusions reached were compared with the results of Mr. BEILBY'S experimentst on the effect of heat on thin ffims of metal .
The present section extends the scope of that enquiry .
* Bakerian Lecture for 1857 , 'Phil .
Trans , 1867 .
Reprinted in FARADAY'S 'Researches in Chemistry and Physics , ' pp. 391 et seq. ( Reference be made to the pages of the reprint .
) 'Roy .
Soc. Proc vol. 72 , 1903 , p. 226. .
'Phil .
Trams , 1904 , p. 416 .
It has been shown that the optical properties of a metal , so diffuaed in vacuo that has some definite value , are the same whether the microstructure be amorphous ( molecularly sub-divided ) or consist of small spheres , these optical properties bning in either case deducible from equation 7 ) , If then , in accordance with the notation adopted in the former communication ( pp. 403 et seq the accents hitherto used to denote the optical constants , and , when differs ffom unity , be now omitted , and the values of those constants corresponding to any particular value of be denoted by a suffix , the values of and are given by equation ( 17 ) , p. 404 , namely , ( 18 ) , where , as in equation , p. , .
( 19 ) .
By these equations the values of and , determining optical properties of amorphous or " " small-sphere\ldquo ; metallic media of any density , may be calculated for light of wave-len th , in terms of the values of and of for the same mono- chromatic light .
The values of and of for gold and for silver have already been calculated for all values of in the case of red light and in that of yellow light .
The results are given in Table .
of the former communication .
But in order to obtain a true conception of the colours of such media , corresponding calculations must be effected for other colours also .
Now , in the case of silver , the numerical values of all those functions of and which have hitherto been calculated for the case of \mdash ; those functions , in fact , which relate to molecules or smau spheres of silver in \mdash ; vary continuously from red to blue .
If , therefore , the values of and corresponding to and be now calculated for all values of , the values of and for other colours may be obtained approximately by interpolation between and The values of and corresponding to and have therefore been calculated for certain values of , by means of equations ( 1S ) and ( 19 ) .
The values of and used for the calculations were carefully determined by R. S. MINOR .
S The results are tabulated below ( Table VIIL ) .
In the case of gold it is not so easy to apply this method of interpolation .
The values of and corresponding both to blue and to green * Phil. Trans , 1904 , equation ( 16 ) , ' Phil. Trana , ' , 1904 .
'Phil .
Trans , 1904 , p. 406. .
S ( mde ante , Table have been calculated ; but the values of and used for the purpose are those given by HAGEN and RUBENS , are not by any means so accurate , as those which MINOR has determined for silver .
The results are included in Table VIL We return to the consideration of the transmission and reflection of light by a metallic film .
We confine our present attention to films the microstructure of which is either amorphous or consists of small spheres of metal ; the films in question are thus optically isotropic .
Suppose that , as explained in the preceding section , the optical constants of the film are and when its specific gravity is times that of the metal of which it is composed .
When light of wave-length in vacuo is directly incident on such a film , of thickness , let and denote the ratios of the intensities of the reflected and of the transmitted light to that of the incident beain .
Adopting the analysis already given in the previous paper , p. 409 , we suppose the film to be bounded by and , and that Incident wave is ; Reflected wave is ; Wave in film , i.e. , between and , is Transmitted wave is , Cexp ; -Cexp where .
We shall that , so that we shall be correct within 2 per oent .
when we neglect in comparison with .
The boundary conditions at namely the continuity of the components of and which are parallel to the interface , then give ; Eliminating we obtain , by taking the squares of moduli , .
( 20 ) . .
cit. .
above , S4 .
Cf .
PhiL Trans , 1904 , p. 409 .
ON COLOURS IN METAL GLASSES , ETC .
In equation(26 ) , on p. we have already proved that ( 21 ) If we write ( 22 ) , equation ( 21 ) may be written ( 23 ) .
Equations ( 20 ) and ( 21 ) are thus correct within 2 per cent. for directly incident visible light , and for in the case of gold if , or in the case of silver if , where millim. For convenience of reference the corresponding results for obliquely incident light are given below .
Let be the angle of incidence .
When the incident light is polarised in the plane of incidence , the ratios of the intensities of the reflected and of the transmitted beams to that of the incident light are given by .
( 24 ) , , say ( 25 ) , where and are defined by the equation .
( 26 ) .
When , however the incident light is polarised perpendicular to the plane of incidence , the corresponding ratios are given by .
( 27 ) , ( 28 ) , where . . .
( 29 ) .
Putting , we obtain It appears from equation ( 23 ) that the colour of the light transmitted by a metallic ffim , although principally dependent on the values of for different values of , is also affected by the corresponding values of .
The thicker the film , however , the less important is in determining the colour .
'Phil .
Trans , 1904 .
ON COLOURS IN METAL GLASSES , ETC .
11 .
Cotour Changes by Heating Metal Films .
In the Bakerian Lecture*for 1857 , FARADAY described a number of experiments concerning the colours which gold and other metals were , in various conditions , capable of exhibiting .
Mr. G. T. BEILBY'S investigations on the colour and structure of films of metal are described in his paper on " " The Surface Structure of Solids.\ldquo ; The average thickness of the gold leaf which FARADAY used in his earlier experiments was about 90 Reference to p. 271 above will show that , with a probable error of 2 per cent. , the optical properties of such a leaf will be subject to the analysis given in S10 above .
Thus , for example , if we assume that , in a gold leaf as it leaves the beater , the gold is in an amorphous state , its colour by directly transmitted light is that for which , as given by equation ( 23 ) , namely .
( 23 ) , is a maximum .
If , further , the metal has its normal specific gravity , so that the values of and of iri this equation are those given for in Tables VIII .
and IX , or figs. 8 and 10 .
Now , when , the value of is much smaller for blue and green than for yellow and red , and is slightly smaller for blue than for green ; while the value of is greater for green and blue than for yellow and red , and is considerably greater for green than for blue .
Thus both the factors of in equation ( 23 ) are greater for blue and green than for yellow and red .
The former colours therefore predominate in the transmitted light .
Further , in very thick ffims , for which is of supreme importance , blue should , in the transmitted beam , predominate over green ; while in thinner films , on account of the greater value of for green than for blue , green light should be more intense .
FARADAY found that all his gold leaf appeared olive-green by transmitted light .
Again , Table IX .
, or fig. 12 , shows that the reflecting.power , , is , when much greater for red and yellow than for green and blue ; and this result is again in accordance with the observed colour of gold leaf by reflected light .
FARADAY , S however , states that gold leaf still appeared green by transmitted light when thickness was reduced to only or 5 .
Now equation ( 24 ) of the former communication namely , * Reprinted from the 'Phil .
Trans. ' in his ' Researches in Chemistry and Physics , ' p. 391 .
ncoe be made to the pages of this reprint .
Ghsgow , 1903 .
, p. 394 .
, p. $96 .
'Phil .
Trana , 1904 , p. 408 .
shows that , for a film of such thinness , the intensity of the transmitted light is greatest for red light .
* This red colour has been seen both by and by parts of their green films .
FARADAY says the red colour was extremely faint but appeared to have an objective reality , while BEILBY describes the effect as that of " " an irregular film of pink jelly It appears that extremely thin films of gold are , by surface tension , drawn up into green patches , leaving larger areas covered by an almost transparent , but faintly red , film .
The effect on the unaided eye is that of a transparent green .
The silver leaf used by BEILBY was over 300 thick .
It therefore comes well inside the range for which the analysis of S10 applies .
Now Table VIII .
, or fig. 9 , shows that , for amorphous silver of norm.al specific gravity , is least for the more refrangible rays .
Again , Table IX .
, or fig. 11 , shows that , for is greatest for the same rays .
It follows therefore , from equation ( 23 ) above , that , on both these accounts , the light transmitted by silver leaf should be blue ; and , in fact , silver leaf transmits .
a deep blue The approximately equal values of the reflecting power , , shown in Table IX .
, or fig. 13 , for , correspond to the almost colourless reflection from polished normal silver .
Consider now the colour changes which , according to fige .
8 to 13 , deduced ffom the calculations of S10 , should accompany a diminution the density of gold and silver films from its normal value to zero This diminution of density -may be conceived either as an increase of the distance between adjacent molecules or as due to the aggregation of groups of neighbouring molecules into small spheres .
For geometrical considerations show that so soon as two sphere form adjacent to one another in an otherwise amorphous mass of metal , the density of the mass must begin to diminish .
And it has been shown that the calculations in question are applicable whether the metal is in small spheres or in an amorphous state , and thus when it is partly in the one condition and partly in the other .
Taking first the case of gold , it appears from figs. 8 and 10 , in conjunction.with equation ( 23 ) , that , as begins to diminish from unity , the absorptions of red and yellow light increase rapidly , owing both to the increase of and to the decrease in .
Meanwhile , owing to the decrease of theratio of ( green ) to ( blue ) and to the increase of green ) ( blue ) , the relative intensity of green to .
blue in the transmitted beam diminishes .
Thus the first effect is to make the transmitted light.bluer , and this effect continues until about75 .
As continues to diminish below this value , the absorption of red rapidly decreases until , at , in a very thick film , S the absorption of red has become as small as that of blue .
The film is *Cf .
Table IV .
, p. 406 of former paper , and Table VIIL abovt , p. 400 .
, p. 40 .
S When is the dominating factor in thinner films for which is very important .
ON COLOURS IN METAL GLASSES , ETC .
then purple .
As still diminishes , the relative absorption of red continues to beOome small , so that the film becomes pink .
Finally , at , the absorption of is less than that of blue .
' and the colour has changed from pink red .
* It is rther seen , ffom fig. 12 , that the reflecting power has , as began to diminish , become more yelow .
At , when the colour of is still purple or pink , the ur of has become green ; and thenceforward remains green as diminishes ffom6 to zero .
Similar consideration of figs. 9 and 11 shows that in the case of silver , as beglins to diminish from unity , the colour by transmitted light becomes at first bluer , then changes to purple in the of , and thence , through pink , to red or " " amber\ldquo ; as ffirther diminishes to .
In fact , it appears from the four colours for which calculations have been made , that there is , for any value of wel-defined absorption band at some position in the ( visible or invisible ) spectrum , and that , as diminishes , position of this absorption band moves from the infra-red through the visible spectrum towards the shorter wave-lengths , being at for about83 , at for about80 , at for about69 , and at for about55 .
Fig. 13 shows that the colour of reflected light becomes distinctly.blue at about , and remains blue down to With a view to determining what may be the explanation of the colours and .
changes of colour exhibited by gold and silver films , we have now to compare the latter colours with those which we have found bove to be consequent upon a mere otropic change in density .
BEILBY has prepared gold ffims by using paints in which " " the metal had been brought into solution in an essential oil Having smoothly coated a plate of glass or mica with the paint , he heated it to a temperature of about , thereby driving off the oil and other volatile constituents .
A film of pure gold with full metallic reflection , and transmitting green light , is left adhering the glass .
When these films are kept at a high temperature for some time , they change colour .
By transmitted light , the original olive-green colour becomes at first bluer , then changes to purple , in which , as the annealing process is still continued , the red predominates more and more over the blue , until finally the purple has given place to pink .
The reflecting power of the has , meanwhile , diminished .
But the colour of the light reflected from the blue films remains yellow , while the pink films reflect a green colour .
I have before me a gold film prepared in this way and subjected to lengthy annealing .
By.transmitted light it appears striated with pink and blue bands .
By reflected light the blue striae become golden , : but the pink striae green .
Under the microscope the film appears continuous , and is quite thick .
These changes , both with transmitted and with reflected light , are just *Or ydlaw , if Ro is the .
footnote on p. 243 above .
, .
40 .
those which have been shown above to be consequent upon a continuous diminution in the density of a gold film , which throughout remains either amorphous or " " granular\ldquo ; ( i.e. , possessing a microstructure of small spheres ) .
The view that the fikn is initially amorphous or granular , and that heating diminishes its density , is supported , as has already been pointed out , the fact that the curves in figs. 8 and 10 show that the absorption of light increases from unity , while BEILBy'S films exhibited just such an increase of absorptive power when first heated .
This view is also in accordance with the loosening of structure which is suggested by the great decrease in electric conductivity which accompanies heating .
But direct evidence of the correctness of the view that heating produoes deCreaSe in denSity not wanting , f0r BEILBY haS eStimated the thiCkneSS of a film which had been annealed to the purple stage .
He found , by weighing the gold from a given area , that , had the density of the gold been then normal , the ffim would have been thick , whereas , under the microscope , the thickness seemed to be much greater than this .
The density of the gold in the purple film thus appeared to be less than in the normal green films .
We conclude , therefore , that ( a ) the flms , as first prepared , are amorphous .
or granular in structure ; and ( b ) heating diminishes the density of the , while pressure is abte to increase the density a.qain .
Further , BEILBy foundt that , when the heating of a film was continued after it had reached the purple stage , " " the ffim assumes a ffosted appearance by reflected light and becomes paler by transmitted light The frosted surface appeared , under the microscope , to consist of granules at least in This phenomenon suggests that in the earlier stages of annealing , smaller granules were formed , which , as annealing proceeded , ran together to form larger granules : and the formation of .
such minute granules , while , according to our analysis , it does not affect the optical continuity of the film , will explain the diminution in density which occurs on heating .
It is , therefore , most probable that ( c ) the diminution in density produoed by heating is effected by the passage of moetal fom the amnorphous to the granular phase and the growth of the granules at the expense of the smaller ; and the increase in density produced by pressure be mpanied by the passage of metal from the to the amorphous The optical properties of the films of gold which FARADAY produced by reducing that metal from its solution by means of phosphorus , tend to show that these films are composed of amorphous or granular gold of density less than the normal .
The appeared to consist of pure gold ; when first prepared .
the film6 appeared of a grey * ' Phil. Trans , 1904 , p. 416 .
, p. 4 .
the of pressure on FARADAY'S " " phosphorous\ldquo ; films after S [ Note addd August , 1906.\mdash ; Subsequent analysis ha4 hOwever , of the granules would cause the colours of the standard metal to FARADAY , , p. 408 .
ON COLOURS IN GLASSES , ETC .
lour , which was ffequently resolvable into a mixture of green and amethystine striae .
These colours would be shown by an amorphous or granular film for which the density was in parts as low as .
Moreover , such a structure agrees with the fact that " " the Alms did not sensibly conduct electricity\ldquo ; and that " " the films cannot be regarded as continuous.\ldquo ; * FARADAY further states that , though they are certainly porous to gas and to water-vapour , the films have evident optical continuity .
Heating diminished the conducting power and changed the colour to amethyst or ruby , just as with BEILBY'S fihns , pressure , which we should expect to increase the density of the film , changed the transmitted colour to green and increased the reflecting power ; and these are precisely the changes which would , according to calculation , accompany an increase in to the neighbourhood of unity in the case of an amorphous or granular film .
Closely allied to these phosphorous films are the deposits of gold on glass which FARADAY obtained\ldquo ; by deflagrating a gold wire by explosions of a Leyden battery ' ' There is no reason to doubt that these deposits consisted of metallic gold in a state of extreme division This method of preparing these deposits is similar to BREDIG'SS method of obtaining suspensions of gold in water ; it is , therefore , to be expected that the deposits consist of smaU spheres of gold together with some large crystallites .
The ffims were so discontinuous as to be unable to conduct electricity ; but they were such as to present an optical continuity .
FARADAY sums up their colour changes as " " Fine gold particles , loosely deposited , can in one state transmit light of a blue-grey colour about .
or can by heat be made to transmit light of a colour , or can by pressure from either of the former states be made to transmit light of a green colour , these changes being due to modifications of gold as gold and independent of the presence of the bodies on which for the time the gold is supported It appears , therefore , that the conclusions , ( b ) , , arrived at on p. 280 for BEILBY'S films , are also applicable to FARADAY'S " " phosphorous\ldquo ; films and to FARADAY'S " " deflagration\ldquo ; ffims .
One more experiment of on coloured gold deposits remains to be noticed .
When a drop of solution of chloride of gold is evaporated in a watch-glass until the gold is reduced , a portion of the gold is generally found to have been carried by the vapour on to the neighbouring part of the glass .
This part has the ruby tint ; and we have seen that a ruby tint is characteristic of the light transmitted by , p. 407 .
, p. 439 .
, p. 401 .
S Cf .
above , p. , and footnote , p. 266 .
, p. 402 .
, p. 439 .
Probably nearly ; but see fourth footnote on p. 280 , above .
, p. 428 .
VOL OOV.\mdash ; A. 2 amorphous or granular gold , the density of which is in the neighbourhood of 6 of that of normal gold .
The similarity of this method of prepariug a metallic film with to that by which Professor R. W. WOOD prepared the sodium and potassium films , deeoribed in S 12 of the former communication , deserving of notice , and , \amp ; om a different standpoint , tends to confirm the view there expressed as to the physical nature and structure of Professor WOOD'S films .
The conclusions , ( b ) , ( c ) arrived at above as to the effect of heat and pressure on metallic films do not apply only to gold , as the following observations on silver ffims show .
FARADAY obtained silver films by reducing silver \amp ; om a solution of the nitrate .
The thinner parts of these films transmitted light of a " " warm brown or sepia tint .
Pressure brought out the full metallic lustre and converts the colour ffom brown to blue \ldquo ; The behaviour 01 these ffims corresponds to that of the gold films obtained with phosphorus .
Again , R. W. WOOD prepared films by chemicaly depositing silver on glass .
These films , as originally prepared , show the same reddish-brown colour by transmitted light , and have a good blue-green reflection .
It has been shown above that both these colours are characteristic of amorphous or granular silver , for which is appreciably less than .
These films showed no electrical conductivity ; S so that , as in the case of BEILBY'S gold films the evidence of a loose structure afforded by the colours exhibited is confirmed by .
the evidence from conductivity .
12 .
The Exceptional Case of Beaten Metat Leaf .
There is one class of metallic film which , when heated , does not exhibit the lour changes that , according to our calculations , correspond to a gradual diminution in the density of the ffim .
To such films the conclusions , of p. 280 do not directly apply .
Instead of being obtained from finely divided metal by chemical deposition , deflagration , \amp ; c. , the ffims in question are prepared by beating sheets of the solid metal into thin leaves .
observed that heat caused gold leaf to lose its olive-green colour and silver leaf to lose its deep blue colour , the films at the same time .
becoming more .
cit. , p. 412 .
, p. 409 .
'Phil .
Mag August , 1903 .
The silver was prepared by the method of CAREY LBA ( ' Amer .
Journ. of Sc 1889 ) .
A further memoir on WOOD'S silver films is now in course of preparation .
S Cf .
BARUS and SCHNEIDER , 'Zeitschr .
Phys. Chem VIII .
, p. , 1891 , who attempted to the conductivity of a silver film prepared by CAREY LBA'S and found that , so drop of the silver suspension dried , so that the charged particles of cowld no longer move about , the conductivity of the drop vanished .
See above , p. 280 .
, p. $96 et seq. ON COLOURS IN METAL GLASSES , ETC .
and tonding to shrink during the process .
* Thus a silver leaf which before heating was opaque , or only able to transmit deep blue light , and that very ffibly , was so altered by heating that the light of a candle could be seen through forty thicknesses .
But in every case the original colour of the leaf , whether of gold or of silver , returns when the leaf is subjected to pressure .
The differences between the effect of heat on chemically prepared films and on beaten leaf correspond to differences between the laminated strueture of the leaf and " " the closer and more horn-like texture of the films deposited by chemical agents.\ldquo ; The optical properties of a Iaminated metal leaf may be estimated and compared with the corresponding properties of an amorphous or granular film of the same metal , if the optioal constants of a plate built up of a number of flat spheroidsS with their polar axes normal to the plate can be calculated .
The general problem of the transmission of electromagnetic waves by a medium composed of a number of minute similar and similarly situated elipsoids , distributed at random many to a wave-length , has now been solved , and it is hoped that the discussion of the optical properties of gold and silver leaf , of the change in those properties which is produced by heat , and of the relations of metal films ( spheroidal , granular , and amorphous ) to polarised light , may form the subject of a future memoir .
With exceptions , namely , the properties peculiar to beaten leaf and the relations of metal films to polarised light , all the experimental relations of gold ( and other metals ) to light , which FARADAY described in his Bakerian Lecture have now been discussed , and we are led to the conclusion that the phenomena exhibited\mdash ; whether by ohemically or electrically deposited films , or by particles of gold diffused in glass , jelly , or water\mdash ; are due to different groupings of the metal molecules and to variations in the mean distance between adjacent molecules , and in no case are they due to allotropic modifications of the molecules themselves .
13 .
CAREY LEA'S " " Allotropic\ldquo ; Silver .
In the former communication it was suggested that CAREY LEA'S " " aUotl.opic\ldquo ; silver was in reality only finely divided silver , the division being sufficiently fine to admit of the ffims being optically continuous .
He two principal arguments , p. 396 .
FARADAY , , p. 399 .
BEILBY , , p. 43 .
The difference in structure is shown by the faot that while mercury will diflUse slowly and uniformly in the compact film , in the leaf thin streams of mercury may be seen shooting rapidly in all directions .
1 BEILBY has shown that a layer of exceedingly flat spheroids is generally found on the surface of a metal .
'Phil .
Trans , 1904 , p. 419 .
It is not sary to suppose the microstructure of the finely divided silver to be granular , as was done in the former paper .
It may be in part granular and in part amorphous .
' Amer .
Journal of ' 1889 , and ' British Journal of Photography , ' March , 1901 .
Also ' Phil. Mag vols .
31 , 32 ( 1891 ) .
for the allotropy of silver in the form in which he prepared the metal .
to examine these arguments .
In the first place , then , all CAREY LEA'S silver ffims were prepared from silver suspensions .
He claims that these suspensions were " " true solutions and that the abihty of the silver to remain in solution in water was evidence that the molecules of the silver in question diflered from those of normal silver , or , in other words , the silver was in an allotropic form .
We are now , however , familiar with the fact that particles of normal silver , as of many other metals , are able , in consequence of mutual electrostatic repulsions , keep themselves in suspension in quite pure water .
Again , we have seen that , when a silver solution is prepared by BREDIG'S method , its refractive index is that which is possessed by a suspension of small spheres , but not of molecules , of silver in w.ater , and in the same case there is a strong absorption band at exactly that point of the spectrum at which small spheres , but not molecules , of silver in water would produce a maximum ; that in this case the greater part of the silver is certainly present in the form of small spheres .
Further , if , when prepared by deflagration , silver in suspension in water takes the small sphere form , it is pnmd probable that it does the same when obtained by CAREY LBA'S method , S and this probability is increased by the fact that CAREY LEA'S silver suspensions exhibited the same red , , and yellow colours which are shown by BREDIG'S suspensions of different densities .
We conclude that CAREY LEA'S " " solutions of allotropic silver\ldquo ; consisted of small spheres of normal silver in suspension We should therefore expect that the films obtained by CAREY LEA would be similar in constitution and behaviour to BEILBY'S " " gold paint\ldquo ; ffimsl and to FARADAY'S phosphorous ffims .
This leads us to CAREY LEA'S second argument for the allotropy of his silver ; he states:\mdash ; ' ' The brittleness of the substances and [ blue and gold coloured respectively , by reflected light ] , the facility with which they can be reduced to the finest powder makes a striking point of difference between allotropic and normal silver .
It is probable that normal silver , precipitated in fine powder and set aside moist to dry gradually may cohere into brittle lumps , but there would be mere aggregations of discontinuous material .
With allotropic silver the case is very different , the particles dry in optical contact with each other , the surfaces are brilliant ; and the material evidently continuous .
That this should be brittle indicates a totally different state of molecular constitution from that of normal silvo *See footnote p. 263 above .
See above , p. 268 .
See above , p. 267 .
S Above p. 269 .
Cf .
also the fact that the silver in a silver-stained glass is in the form of small Cf .
ako evidence given on p. 259 above .
See above , p. 279 .
above , p. 281 .
This expectation is verified by a further examination of WOOD'S films .
See note above p. 282 .
'Brit .
Jour .
Phot March 1901 , p. 21 .
ON COLOURS IN METAL GLASSES , ETC .
An these operties are shared by FARADAY'S " " phosphorous\ldquo ; gold , that our expectation is , so far , ffilfilled .
We are , in fact , perfectly familiar with ' ' mere aggregations of discontinuous material\ldquo ; which are optically continuous\mdash ; for example , gold ruby glass .
Many of the observations which CAREY LEA has recorded on the colours of his silver ffims are in accordance with the expectation that these films , like BEILBY'S gold films and FARADAY'S " " phosphorous\ldquo ; gold , should behave according to the laws , ( b ) , and ( c ) stated above .
But wo difficulties arise in the way of this accordance , for , in the first place LEA'S recorded observations do not sufficiently distinguish between transmitted and reflected light .
For example he records that his freely precipitated silver dissolves to a blood-red colour , and proceeds " " When the substance is brushed over paper and dried rapidly it exhibits a beautiful succession of colours .
At the moment of applying it it appears blood red ; when half dry it has a splendid blue colour and lustrous talhc reflection ; S when quite dry this metallic effect disappears and the colour is matt blue .
Lastly , in the case of the films discussed in S11 above , the colour depended on the fact that the density of the film was less than that of the metal composing the film when in its normal state ; but pressure increased the density to its normal value , at the same time bringing out the normal colour , both by reflected and by transmitted light , of the metal .
And CAREY LEA'S silvers " " show a lower specific gravity than that of normal silver ; \ldquo ; and pressure " " instantly converted gold-coloured alotropic silver into normal silver .
We conclude ffom the above evidence that this silver was not " " allotropic but consisted of normal silver in a finely divided state .
14 .
HERMANN VOGEL'S Silver .
Before leaving the consideration of these discontinuous forms of silver , reference must be made to a paper by HERMANN VOGEL , in which the author describes how * The least touch of the finger removed the film of gold . .
These films , though they are certainly porous to gas .
have evident optical continuity\ldquo ; FARADAY , .
Cf .
also the facts that films analogous to CAREY LEA'S did not conduct BARUS SCHNXIDER , , and that the phosphorous films did not sensibly conduct ioity FARADAY , 'Brit .
Journ. Phot March , 1901 , p. 19 .
This is the colour by transmitted light when is fairly small .
Cf .
figs. 9 and 11 .
S This is the reflected colour for values of from zero to nearly8 .
Cf .
fig. 13 .
Professor W. WOOD repeated this experiment , using glass instead of paper to support the silver film .
The metallic effect , then , does not disappear , but remains after the film has become quite dry .
Cf .
above , p. 282 .
'Brit .
Journ. Phot March , 1901 , p. 21 .
PhiL Mag vol. 31 , p. 244 , 1891 .
'Pogg .
Ann CXVII .
, p. 316 , 1861 .
he prepared silver of less specific gravity than that of normal silver , by depoeiting that metal on the platinum electrode of a platinum-zinc battery .
He also prepared silver in suspension in water by chemical means , observing the characteristic amber colour and noticing that precipitation could be accelerated by the addition of salt to the water .
VOGEL concludes ( loc. cit. , p. 837 ) that there are three forms of silver , ( 1 ) regular dendritic silver [ crystalline ] ; ( 2 ) granular powdery silver [ small spheres ] ; ( 3 ) mirror silver [ amorphous ] .
He found that the second type " " tended to the formation of a coloured powder but could be changed into the third type by pressure .
He adds that the silver precipitated by photography is of the second type , and this is the view suggested in the preceding memoir ( p. 417 ) , because of the redbrown transmitted colour and the green colour of the reflection from fogged photographic films , which , according to the analysis given above , S 10 , are the colours exhibited by ffims of amorphous or granular silver , less than standard density .
15 .
Allotropic Forms of Metal .
In the course of the preceding investigations we have been led to recognise that variation of the relative position of the moleoules of a metal will cause the metal to change colour , whether it be examined by reflected or by transmitted light .
It has been shown , for example , that mere variation in density causes gold in one state to transmit green light , in another blue , in another purple , and , in another again , ruby .
Further , this discovery has led us to the conclusion that , in order to account for the properties of CAREY LEA'S anomalous silvers , it is not necessary to assume the existence of an " " allotropic\ldquo ; molecule of silver .
The question thus arises : Are there any other cases in which an allotropic molecule has been unnecessarily postulated collected particulars of a large number of supposed cases of allotropicS states of metals .
We proceed to the examination of these particulars in order to determine whether the effects , for the explanation of which the allotropic molecule was postulated , are not merely those which , according to the analysis of S10 above , would be due to a decrease in the density of the metal in a granular or amorphous state .
In the first place , then , the discovery that metals in different statee , corresponding to different methods of preparation , possessed different densities and had widely different properties , although chemical analysis could detect no change in the *Cf .
figs. 9 , 11 and 13 , and also p. 282 above , where the same colours , exhibited by one of W. WOOD'S silver fllms , are discussed .
Above , p. 286 .
Metallurgy , ' pp. 87 et seq. S ROBBRTS-AUSTEN defines " " allotropy\ldquo ; as follows ( loc. it .
, p. 89 ) : The ocourrenoe of elements in . .
allotropic states means that . .
th atoms are differently arranged in the moIeoulea\ldquo ; ON COLOURS IN METAL GLASSES , ETC .
does not require those different states to have been allotropic .
Again , it is unnecessary to suppose that BOLLEY'S lead , prepared by electrolysis , and similar in composition to sheet-lead , is allotropic because it oxidises rapidly in air while sheetlead does not : for the electrolysis gives the essential fine division , and the consequent large amount of surface exposed to the air greatly accelerates oxidation .
Lastly , SCHUTZENBERGER supposed that the copper deposited on the platinum electrode of a copper-platinum cell was allotropic because it was ry fragile , its density was only about9 of that of normal copper , it oxidised rapidly in air , and it could be converted into normal copper by prolonged contact with dilute sulphuric acid .
Here , too , the supposition of allotropy is not required to account for the facts .
For the low density , the fragility and the rapid dation are all accounted for by the loose structure which we should expect in such a deposit of copper , while CAREY LEA found that his silvers , which , if our conclusion at p. 285 is correct , were only finely divided silver , could be transformed to normal silver by contact with sulphuric acid .
Similar remarks apply to SCHUTZENBERGER'S silver .
Consider now MATTHIESSEN'S important generalisation , S that metals may sustain change in their molecular condition by union with each other in a fused state .
ROBERTS-AUSTEN points that the evidence that metals ever assume allotropic states , when they enter into union with each other , is difficult to obtain .
When obtained , the evidence is generally composed of the facts that the specific gravity of the normal metal is greater.than that of the metal in the state alleged to be allotropic ; that the chemical activity is less in amount , although the same in kin , for the former than for the latter state ; and that the appearance of the metal is different in the two states .
Reference is also sometimes made to a difference in physical properties which is accounted for by lack of continuity , and consequently of electric conductivity , in the supposed allotropic state .
Occasional reference is ako made to a readiness to form hydrates which the metal in the latter state exhibits .
Setting this last property aside , as not yet established , the remaining evidence is not conclusive , for all the facts in question are also characteristic of continuous granular ( or amorphous ) pieces of metal .
Increase of chemical activity , for example , is a consequence of the enormous effective surface in a ium built up of independent granules .
Further , when one metal is united with another in a fused state , a chemical compound is not , in general , formed , but the molecules of the two metals freely mix .
Thus one metal is in solution in the other .
So long , therefore , as the temperature remains sufficiently high to permit the molecules to move about freely , the molecules of each metal tend to segregate , and to group themselves into separate crystals as the *JOULn and LYON PLAYFAIR , ' Memoirs of the Ohem .
Soc vol. , p. 57 ( 1846 ) .
RoBBRlS-AUSTBN , , p. 90 .
'Bull .
Soc. Chim XXX .
, p. 3 BOBBRIS AUSTnN , , p. 87 .
, p. 91 .
Cf .
PXTBRSmN on " " Allotropic Forms of Metals\ldquo ; Zeitschr .
Phys. temperature is slowly lowered .
It is , however , probable that , as in the case of gold and copper ruby glasses , the molecules of each metal first group themselves into small spheres .
If the temperature were rapidly lowered at this stage , this granular structure would be fixed in the alloy .
If , then , one metal\mdash ; that , suppose , of which the larger volume is present\mdash ; were suddenly annihilated , the other metal would remain in a granular form , possessing a colour*quite difIerent from that exhibited by the normal form of that metal .
Now when an aUoy of potassium and gold containing about 10 per cent. of the precious metal is thrown on to water , the potassium is , in effect , annihilated , and the gold is released as a black or dark brown powder .
It will be seen from fig. 12 that granular gold , with a density slightly over6 of that of normal gold , would reflect light of a brown colour , while the reflecting power would not exceed granular structure is thus in accordance with the dull appearance and with the colour of the powder .
Similarly when a silver-gold alloy containing two parts of silver to one of gold is treated with nitric acid the silver is removed , the gold remaimng in the form of a dull brown powder , which can be converted into bright metallic gold by slight pressure or by heating to redness .
It appears , therefore , that this brown powder is probably granular gold , the component particles being small compared with a wave-length of light ; so that , once more , the does not require us to suppose this form of gold to be allotropic .
Finally , it seems unnecessary to assert that iron released ffom its amalgam by distilling away the mercury is in an alotropic form because it takes fire on exposure to the air .
For this burning of the iron would be the consequence of the large surface exposed to the air by an extremely finely divided form of the metal .
We conclude , therefore , that in none of the cases of supposed allotropy , which we have examined in this section , has the existence of an allotropic form of metal been established .
*See S10 above .
Cf .
ROBBRTS-AUSTEN , , p. 91 .
The potassium does not catch fire , but combines with the water to form KHO ( which immediately passes into solution and is thus removed ) and which catches We must except that of the alleged formation of aunc hydrate , but I have been unable to obtain any confirmation of the existence of such a compound .

