July 15, 1919 Disperse Colloids in Bituminous Road Materials 171 



In comparing these values it will be noted that the average depth 

 of all cells is but slightly in excess of o.io mm., while the maximum 

 variation in depth and in the thickness of slide is approximately the 

 same (0.007 rnii^-)' indicating a cell floor closely approaching a true 

 plane. 



Dial measurements also were undertaken to determine the depth of 

 cell, including cover glass after mounting in the asphaltic oil solution 

 employed in counting (see below) and in air to form an estimate of the 

 relative thickness of the liquid film between cover glass and slide beyond 

 the cell area. In every case lesser values were obtained for cells mounted 

 in this solution than in air, indicating a more perfect contact through 

 the release of atmospheric pressure and the adhesive character of the 

 bitumen. 



PREPARATION OF SOLUTIONS 



Before describing the method of counting colloidal particles employed 

 in this investigation, it will be found desirable to outline briefly the 

 general character of bituminous solutions containing colloidal matter 

 and the manner in which these solutions have been prepared for micro- 

 scopic analysis. 



When viewed under the ultra-microscope the colloidal portion of the 

 solution will appear as a mass of very finely divided and more or less 

 widely dispersed particles undergoing a constant and, under certain 

 conditions, perpetual movement (Brownian movement). This move- 

 ment has been ascribed to the molecular energy of the suspending liquid 

 and may be regarded as a function of the size of the particles and their 

 degree of dispersion which, in turn, is limited by the viscosity of the 

 solution.^ In order, therefore, to count successfully these suspended 

 particles it was found necessary either to retard their movement by 

 suitable concentration of solution or to increase it by dilution to such a 

 degree that they settled out within the cell inclosure.^ In the latter 

 case, however, it frequently happened that the particles were in part 

 resorbed on exposure to light, thus destroying the accuracy of the count. 

 To assure concordant results, therefore, the particles were always counted 

 in a somewhat viscous solution of colloid-free asphaltic oil to which a 

 definite amount of paraffin had been added. This solution was pre- 

 pared by fluxing 2.5 gm. Mexican oil asphalt (penetration 148) with 

 0.5 gm. crystalline paraffin and diluting to 100 cc. with benzol contain- 

 ing 10 per cent alcohol. This was then evaporated to constant weight 

 on the water bath, brought to original consistency with benzol and 

 passed through an alundum tube or filter tube clogged with macerated 

 filter paper until approximately all suspended matter had been removed. 



• OsTWALD, Wolfgang, die welt der vernachlassigten dimensionen. p. 34-35. Dresden and 

 Leipzig, 1915. 



2 The solvent used in these investigations was c. p. benzol, since carbon bisulphid was found to contain 

 an appreciable quantity of colloidal sulphur. 



