PART II |

FLUIDITY AND OTHER PHYSICAL AND !

CHEMICAL PROPERTIES .

CHAPTER I

I

VISCOSITY AND FLUIDITY

It has been tacitly assumed by the great majority of workers
that when two liquids are mixed, the viscosity of the mixture
is normally a linear function of the composition. This appeared
as early as 1876 in the work of Wijkander. In a great many
mixtures, including practically all of those in which water is a
component, the viscosity is certainly very far from being a linear
function of the composition, there being often a maximum in the |

viscosity curves. However water mixtures should not be con-
sidered as "normal," but since it is difficult to decide what shall
be considered normal mixtures, the question whether the
viscosities are additive or not is admittedly difficult of solution. f

Dunstan (1905) classifies as normal those mixtures whose vis- l\

cosity-weight concentration curves do not show a maximum or ^

a minimum. This classification is not satisfactory not only i;

because it lacks a theoretical justification but also because many ^

of the so-defined normal mixtures give curves which depart |

considerably from the linear, so that the suspicion is aroused that /

the occurrence of a maximum or minimum may depend upon f

accidental circumstances such as the nearness to equality of the |

viscosity of the components. The accidental character of such jj

a classification is very striking in mixtures which fall into the 3

normal class at one temperature but at a slightly different tern- |

perature must be classified as abnormal. ]

Such light as can be gained from a study of the viscosities J

of mixtures, seems to lead to the conclusion that viscosities are f

i
not additive, as has been assumed. Thus Dunstan (1904) ]

remarks, "The law of mixtures is never accurately obeyed and

6 81