WET GBIND1NG. 



TABLE I. The relative cementing value of various rock samples, etc. Continued. 



MARKED EFFECT OF WATER UPON THE BINDING POWER OF 

 CERTAIN SUBSTANCES. 



In some of the earlier investigations conducted in this laboratory a 

 number of experiments had been made on powdered glass, which was 

 found to have little or no binding power, even if prepared in the finest 

 state of subdivision by dry grinding and elutriation. When, how- 

 ever, following up the results on rock powders, plate glass was ground 

 wet in the ball mill, the surprising fact was established that a very 

 high binding power had been developed, so high, indeed, that bri- 

 quettes made from this glass powder became so hard that they could 

 not be broken in the hands or indented by the finger nail. It is a 

 matter of common knowledge that glass, even in the form of glass 

 vessels, is acted upon by water. In a powdered condition this action 

 would, of course, go farther, but under the influence of wet grinding, 

 in which the products of the reactions were continually being worn 

 off of the surfaces of the particles, it was presumable that a sufficient 

 amount of sodic silicate had been formed to cement the coarser parti- 

 cles together. This was manifestly a point which could be very simply 

 investigated. Phenolphthalein showed an immediate slight alkaline 

 reaction on moistening glass which had been ground dry, but a deep 

 and strong reaction with glass that had been ground wet. Two hun- 

 dred grams of glass that had been ground wet were thrown on a large 

 double filter and extracted with several liters of hot water. The fil- 

 trates were evaporated in platinum and weighed. More than 10 grams 

 (5 per cent) of a hard, glue-like residue were obtained which was 

 without doubt mainly sodic silicate. When this residue was obtained 

 there could be no longer any question as to the reason for the binding 

 power of wet-ground glass powder, the glue-like nature of sodic sili- 

 cate solutions being too well known to require discussion. The inter- 



