188 EXPERIMENT STATION. [Jan. 



The sodium oxide was calculated from the alkalinity, deter- 

 mined by direct titration with methyl orange as indicator, a 

 process sufficiently accurate for the purpose of checking quality. 



The several reactions taking place in the manufacture of slow 

 process green may he summarized in a single equation : — 



4 CuS045 H2O + 8 NaAs02 + 2 C2H4O2 



= Cu(C2H302)2 • 3 Cu(As02)2 + AS2O3 + 4 Na2S04 + 6 H2O. 



Sodium arsenite reacts upon the l)lue yitriol with the produc- 

 tion of a bulky, yellowish-green precipitate of copper arsenite 

 (Scheele's green), whicli iii turn is acted upon by the acetic 

 acid with the formation of a greatly reduced yolunie of Paris 

 green. Experience has shown, as indicated by the aboye for- 

 mula, that about Yi more arsenic is required for the production 

 of the green than actually enters combination, as was the case 

 with the instantaneous process. Acetic acid in excess- of the 2 

 molecules stated (by nearly 66 per cent.) is needed for the 

 reaction. It is evident from what has been said that the man- 

 ufacture of slow process green requires considerable equipment, 

 expensiye reagents and expert control which, together with the 

 imayoidable waste of chemicals, insures a costly product. 



The two general processes for making Paris green and their 

 several reactions were carefully studied in the station laboratory 

 to ascertain the character of the product that might reasonably 

 be expected. As a result of numerous experiments a combina- 

 tion process, using copper acetate and sodium arsenite, together 

 with sufficient acetic acid to offset the alkalinity of the arsenite, 

 was found the most acceptable. 



4 Cu(C2H302)2H20 + G NaAsOs + 6 C2H4O2 



= Cu(C2H302)2 • 3 Cu(As02)2 + fi NaC2H302. 



The reaction was easy to control, could be carried out at any 

 temperature from that of the laboratory to boiling and gave a 

 product of variable ])hysical structure and of fine color. Solu- 

 tions of different concentrations were tried, of which % molec- 

 ular (M/f)) for the acetate and V2 molecular (lM/2) for the 

 arsenite proved satisfactory. This process appeared to require 

 less arsenic in excess than the ordinary method, although the 



