6 CIRCULAR 7 65, U. S. DEPARTMENT OF AGRICULTURE 



The two wires of the 110-volt alternating-current line were attached 

 to the two poles of the primary coil of a 25.000-volt transformer. An 

 electric switch, a rheostat, and a voltmeter were placed in this line 

 One of the poles of the secondary coil was connected to the brass rod 

 leading to the three copper wires of the ozonizer tubes: the other pole 

 was connected to the copper coil immersed in the water of the battery 

 jar. The voltage during operation of the ozonizer was varied from 

 70 to 90 in the primary coil as desired. 



During operation the electric discharge occurred in the annular 

 spaces between the mercury cylinders and the outer cylinders of the 

 Berthelot tubes ; in the dark it appeared as a continuous glow in the 

 tubes. No opportunity was afforded for sparking between electrodes 

 such as often occurs in ozone generators which employ metallic elec- 

 trodes separated by insufficient dielectrics. As a consequence, the pos- 

 sibility of the formation of oxides of nitrogen, which also have 

 bactericidal properties (26) , was obviated; and no trace of them could 

 be found in the storage room. 



Method of Analysis 



Sampling of atmosphere. — A Pyrex tube 3 was extended into the 

 center of the storage room through a small aperture in the wall. The 

 atmosphere from the room was drawn through a gas-washing bottle by 

 means of suction, the volume of air being accurately measured with a 

 precision wet -test gas meter placed between the gas-absorption bottle 

 and the suction pump. Connections to the gas-absorption bottle were 

 made by means of Pyrex standard taper joints. 



Ozone determination. — Thorp's modification (25) of! the starch- 

 iodide method was used for the ozone analysis. 



Solutions needed. — The solutions used consisted of (1) a 2 X solution 

 of potassium iodide; (2) a buffer solution consisting of 5 grams of 

 aluminum chloride hexahydrate and 1 gram of ammonium chloride 

 made up to 1 liter with distilled water ; (3) a 0.01 N solution of sodium 

 thiosulfate; and (4) a starch solution as an indicator. 



Procedure. — To 100 milliliters of the 2 N potassium iodide solu- 

 tion 5 milliliters of the buffer solution of aluminum chloride hexa- 

 hydrate and ammonium chloride was added, and 100 milliliters of 

 the solution was placed in the gas-washing bottle. The ozonized air 

 to be analyzed was drawn through the solution until a yellow color 

 was obtained. The volume of air was accurately read from the gas 

 meter, and the free iodine that had been liberated from the potas- 

 sium iodide by the ozone was titrated with a 0.01 X solution of sodium 

 thiosulfate ; starch solution was used as the indicator. An equal volume 

 of air free from ozone drawn through a second gas-washing bottle of 



3 Glass is inert to the action of ozone : because of this property it was used 

 wherever contact with ozonized atmospheres was necessary. Rubber lubing 

 rapidly disintegrates in the presence of ozone and cannot be used in any operation 

 when it is in contact with that gas. Because the question arose whether it would 

 be permissible to sample an ozonized atmosphere through a copper tube, a series 

 of determinations was made to compare the analytical results of duplicate sam- 

 ples of air drawn through copper and Pyrex tubing. No differences were obtained. 

 Consequently, it was concluded that copper tubing of the comparatively short 

 length required for sampling ozonized atmospheres from storage rooms would be 

 satisfactory. 



