98 



Van Name and Huff — Estimation of 



free from impurities which reduce iodine), and the excess 

 of iodine is at once titrated in the usual way with thio- 

 sulphate, adding starch just before the end point is 

 reached. If the iodine solution is more dilute than N/10 

 about a gram of potassium iodide should be added before 

 or during the titration to increase the sharpness of the 

 end point. The amount of acid taken for the acidifica- 

 tion should be slightly more than twice the quantity, as 

 determined by a rough preliminary test, which will just 

 suffice to give an acid reaction toward litmus to a solu- 

 tion of disodium phosphate containing the same weight 

 of that salt as has actually been added. 



A number of test analyses of standard phosphorous 

 acid solutions by this method are recorded in Table V and 

 VI. For Table VI the solution used was the same as 





Table V. 







lodometi 



■ic Estimation of Phosphorous Acid. 





Iodine solution 1022 N/50. 



Thiosulphate solution 1-010 N/50. 







NaJ3 2 3 







H 3 PO a 



Iodine 



solution 



H 3 P0 3 



Found 



taken 



solution 



required 



found 



% 



cm 3 grm. 



cm 3 



em 3 



grm. 





15 001077 



4500 



32-53 



001078 



10009 



20 001436 



4500 



28-22 



001435 



99-93 



25 001796 



4502 



23-85 



001799 



100-17 



30 002155 



4500 



19-52 



002156 



10005 



35 002514 



4500 



15-21 



002513 



99-96 



40 002873 



45-00 



10-80 



0-02879 



100-21 







45-00 



45-52 











4500 



45-58 







Beagents used in each determination. 25 cm 3 of an 11% solution of 

 NaJHPCX.lSlLO. 5 cm 3 of 42% H 3 P0 4 solution. 10 em 3 of 10% KI 

 solution. 



in Table II, and a portion of the same solution, carefully 

 diluted to five times its volume with the aid of a certified 

 burette and a certified flask, was employed in the experi- 

 ments of Table V. The results of Table V, obtained with 

 N/50 solutions of iodine and of thiosulphate, show a very 

 satisfactory accuracy. Those of Table VI, for which 

 N/10 solutions were used, though remarkably concord- 

 ant, all show a small positive error. Whether this is 

 due in some way to the fact that in this series of experi- 

 ments the solution was acidified with acetic instead of 



