Section III, 1919 [97] Trans. R.S.C. 



Note on a Modified form of Freezing Point Apparatus 



By O. Maass 



Presented by Professor R. F. Ruttan 

 (Read May Meeting, 1919.) 



The molecular weight of a dissolved substance is most commonly 

 determined by the freezing-point method. Familiarity with this 

 method may be assumed so that it is possible to proceed immediately 

 with the subject of this note. 



M the molecular weight of a dissolved substance is given by 

 M = k "'/Wet., where k is the cryoscopic constant, t the lowering of the 

 freezing point when w gms. of solute are dissolved in W gms. of solvent. 

 In the ordinary experimental determination the temperature of the 

 freezing-bath must be as close as possible to the freezing-point of the 

 solution, otherwise the observed reading of the freezing point will be 

 too low. A similar error is brought about by supercooling, for if t' 

 represents the supercooling, s the specific heat of the solution and L 

 the latent heat of fusion, the W in the equation given above is dim- 

 inished by st'/L. The combined errors may easily result in a 3 per 

 cent error in the molecular weight. 



These errors persist to a certain extent. It is generally necessary 

 to have the freezing bath at least Z" lower than the freezing point and 

 supercooling of several degrees is unavoidable in the case of certain 

 solutions even when much time is spent in trying to get the solution to 

 "freeze out" by vigorous stirring, scraping the sides, etc. The only 

 certain method to prevent excessive supercooling is that of introducing 

 a crystal of the solvent. If this is done in the ordinary way the method 

 is cumbersome and the results are uncertain, the crystal having a 

 habit of sticking to the sides of the test tube, or of melting before it 

 reaches the solution. Besides this, there are certain unavoidable 

 errors connected with bodily introducing a crystal; for one thing — • 

 the concentration of the solution is altered and subsequent redetermina- 

 tions rendered inaccurate. 



By means of the apparatus shown in the accompanying diagram 

 a crystal can be introduced without any of the objections mentioned 

 in the last paragraph. The inner test tube, fitted out in the ordinary 

 way with a Beckmann T and a stirrer R, has a small piece of platinum 

 wire E sealed into the glass at a short distance below the surface of 

 the solution. The platinum wire ends in the form of a tack. The outer 



