30 S. L. Penfield — Methods for Determination of Water 



Art. VI. — On son,, Methods for the Determination of 

 Water ; by S. L. Penfield. 



The accurate determination of water is sometimes one of 

 the difficult operations of analytical chemistry. This is espe- 

 cially the case when water is not completely expelled from a 

 mineral at a temperature which can be obtained in a hard 

 glass tube with an ordinary gas combustion furnace, and es- 

 pecially when either the loss of some volatile constituent or an 

 increase in weight by oxidation renders it impossible to make 

 the determination by loss on ignition. The tubulated plati- 

 num crucible described by Gooch* furnished a solution of the 

 problem, but the apparatus is expensive, and the author has 

 found it necessary to surround the crucible in which the fusion 

 is made with ail outer one containing some fusible salt like 

 sodium carbonate, in order to prevent the passage of gases 

 through the red hot platinum, which would render the results 

 of the determination too high. This makes the apparatus 

 more complicated, and the expansion and contraction of the 

 fused salt between the crucibles strains the apparatus to such 

 an extent that repairs are frequently necessary. The use of a 

 platinum tube, wrapped with repeated layers of asbestus paper 

 soaked in borax to prevent the passage of gases, has been sug- 

 gested by Chatard,f but the apparatus is both expensive and 

 complicated. 



The author's experience in testing minerals for water has 

 suggested the possibility of making an accurate determination 

 by heating a weighed quantitj^ of mineral in a closed glass 

 tube, weighing the tube plus the water, then drying and weigh- 

 ing again. It has been found, as will be shown in the course 

 of this article, that the method is very accurate and makes 

 it possible to obtain ' direct water determinations without the 

 use of absorption tubes, or any system of drying apparatus. 

 The method is not altogether new, as it has been used by Prof. 

 G. J. Brushy in the analysis of Sussexite, H(MnMg)B0 3 . In 

 this case after the water had been expelled and the tube 

 weighed, the water was dried out in a vacuum over chloride of 

 calcium, when the tube was again weighed and the percentage 

 calculated from the loss. 



From the author's experiments the following conditions may 

 be recommended as favorable. The shape of the closed tubes 

 depends upon the quantity of mineral that is to be experi- 



*Am. Chem. Jour., ii, p. 247, 1880. 

 f Amer. Chem. Jour., xiii, p. 110, 1891. 

 % This Journal, II, xlvi, p. 240, 1868. 



