46 



NATURE 



[March io, 192 i 



Colloids and Colloidal Electrolytes. 

 By Prof. J. W. McBain. 



COLLOIDS comprise all matter that is made 

 up of particles smaller than a wave-length 

 of light, but larger than a single molecule of an 

 ordinary crystalloidal substance such as sugar, 

 salt, or water. It would appear that in some 

 cases the chemical molecules are linked together 

 into particles of colloidal dimensions, and then 

 from these particles are built up the familiar 

 structures such as rubber, fibres of cotton, wood, 

 or earthenware. It is a moot question as to 

 whether, in the case of certain highly complex 

 organic substances, the single molecules them- 

 selves may not be large enough to exhibit the 

 distinctive properties of colloidal particles. 



Scientific study has been devoted almost ex- 

 clusively to mixtures in which colloidal particles 

 are dispersed throughout a second continuous 

 medium ; such as in many precious stones, 

 ink, the body fluids, or a bar of soap 

 where the continuous medium is water. 

 Furthermore, the investigations of physical 

 chemists have been directed almost entirely to 

 the study of very dilute colloidal solutions (sols) 

 such as dilute suspensions of gold or arsenic tri- 

 sulphide in water, whilst biologists have devoted 

 a great deal of attention to gelatin and protein, 

 colloids of a very different type. For this reason 

 the innumerable observations that have been made 

 on colloids have not been well linked up either 

 with each other or with our general scientific 

 knowledge. There are, however, two outstanding 

 instances in which some of the familiar and un- 

 ambiguous methods of classical physical chem- 

 istry have been extended to the study of highly 

 characteristic colloids— namely, soaps, chiefly 

 studied in this country, and proteins, chiefly eluci- 

 dated by W. B. Hardy and by the professor of 

 biophysical chemistry in Vienna University, Wolf- 

 gang Pauli.^ It now appears that soaps, proteins, 

 and gelatin salts are closely similar types of sub- 

 stances, whilst soaps are by far the most accessible 

 to quantitative measurements. 



A very important characteristic of most colloidal 

 solutions which have received careful study is the 

 fact that the colloidal particles possess electrical 

 charges. For instance, silver particles of dia- 

 meter of about 500 millionths of a millimetre, sus- 

 pended in water, move under the influence of an 

 electric field. This must be ascribed to electrical 

 charges on the particles, and calculation shows 

 that on each such particle there are anything up 

 to 100 million negative charges or electrons. This 

 electrical charge seems enormous until we reflect 

 that it is relatively ever so much less than the 

 number of atoms of silver, and that in an ordinary 

 ion there is one electrical charge for each atom. 



The stability of the dilute suspensions of such 



1 A comprehensive summarv of Pauli's raasterly researches on this particu- 

 larly compile ted material is to be found in his " Kolloid-chemie der 

 EiweisskSrper." Pp. 109. (Dresden and Leipzig : Th. Steiiikoph, 1920.) 



NO. 2680, VOL. 107] 



insoluble substances greatly depends upon these 

 electrical charges. 



As will be shown, these "irreversible " or "sus- 

 pensoid " particles, which have been so largely 

 studied, occupy an intermediate position between 

 electrically neutral colloidal particles, such as 

 rubber in solution in benzene, and the much more 

 highly charged colloidal particles known as the 

 ionic micelle that occur in such aqueous solutions 

 as those of soap. In the ionic micelle or particle 

 the number of electrical charges is commensurate 

 with the number of molecules or ions which have 

 aggregated together. 



Another prominent characteristic which physical 

 chemists have met in attempting to study sus- 

 pensoid colloids is their extreme variability and 

 sensitiveness to all sorts of disturbing influences. 

 It has become almost an axiom that only variable 

 and non-reproducible results can be expected, and 

 that they depend on the individual specimen exam- 

 ined. It is all the more fortunate, then, that in 

 the case of soap solutions it is possible to obtain 

 quantitative reproducible results depending only 

 upon the composition and the state of the system. 

 This has enabled us to investigate through these 

 comparatively simple substances of known mole- 

 cular formulae and structure some of the charac- 

 teristic properties exhibited by solutions of so 

 many of those extraordinarily complex chemical 

 substances, mostly of unknown formulae, which 

 are involved in all life processes, and are 

 frequently of very great industrial importance. 

 Salmon's suggestion is that these colloids should 

 be called "equilibrium colloids," a classification 

 that would in practice more or less correspond to 

 the present modified use of Hardy's term "re- 

 versible colloids," now used chiefly with reference 

 to the properties of dried residues. The expres- 

 sion " equilibrium colloids " has the advantage of 

 possessing a rather deeper significance. 



In the study of soap solutions in the Bristol 

 University laboratory, it was first established that 

 they exhibited excellent electrical conductivity 

 even in the most concentrated viscous solutions. 

 The change in conductivity with concentration 

 exhibited remarkable anomalies such as had 

 hitherto been met with only in certain non- 

 aqueous solutions. The curve passes through 

 both a maximum and a minimum in moderately 

 strong solution. At this time it had been gene- 

 rally considered that colloids as such could not 

 exhibit conductivity, and if observed it was 

 ascribed to impurities and admixtures. 



Although there were no admixtures in the case 

 of these specially pure soap solutions, no data 

 at all existed with regard to the amount of alkali 

 set free in the solution through hydrolysis of the 

 soap by the solvent water. Direct measurements 

 succeeded in showing through two independent 

 methods, electromotive force and rate of cata- 



