296 Viscosity, Density, and Temperature of Salt Solutions 



From these tables it appears that the greatest difference 

 between experimental and calculated values occurs at about 

 30° C, but the differences (a little over 2 per cent, being the 

 greatest) are not great when one considers the experimental 

 difficulties involved in the measurement of viscosity. 



Several other figures for different salts were taken from 

 the Smithsonian Tables and the log. graphs plotted, and in 

 no case was there any great deviation from the straight line. 

 Unfortunately, however, in the majority of cases only three 

 points are obtainable from the given figures, so that the 

 comparison with these other solutions is not so rigid as is 

 desirable. Again, some of the solutions actually diminish in 

 viscosity as salt is added until a certain minimum viscosity 

 is attained, after which the viscosity increases regularly with 

 an increase of density as in the majority of solutions. One 

 of the most noteworthy examples of this behaviour is found 

 in ammonium nitrate, the density-viscosity curve of which 

 is shown in fig. 9 (PL IV.). It might be thought that such 

 a case would form a departure from the above law. It is 

 found, however, that on giving the formula the form 



± (,-,)»=A(l-£) 



where fi is the minimum viscosity, and where the negative 

 sign is used when x is greater than p, x being the density at 

 the minimum viscosity point, then the law does hold very 

 approximately. Five points are given in the case of 

 ammonium nitrate, and on plotting the log. graph for these, 

 they were found to lie very sensibly on a straight line as 

 shown in fig. 10. The same thing is shown in fig. 10 for 

 manganese chloride for which four points are given in the 

 tables, the log. graph of these also being practically a straight 

 line. Manganese chloride, however, does not have a minimum 

 viscosity point but behaves like the majority of salts. 



The reason for the difference in the value of " n" for 

 sodium chloride as found in the two experiments * lies, of 

 course, in the different conditions under which these were 

 carried out, the solutions in the first set of experiments being 

 unfiltered. 



If this law holds generally for all salt solutions, as seems 

 probable, the best method of obtaining the formula for each 

 salt would be to make three accurate determinations (at the 

 same temperature) of the viscosity and density at points as 

 far apart as possible and deduce the value of " log A " and 



* Pp. 292 & 203. 



