2i6 E.\amti!ifli!n niiit Impnvtmeiit! cf the Aujl^'fs nf Steel. 



of pruffiateof pot-afh, afforded prufiiatcof iron as beautiful as at firft. The fecoiid pre- 

 cipitate, treated as before, and ftrongly heated under n niuffle, became at fiift black, but af- 

 terwards acquired a brown-red colour, which began to deliroy the opinion I had conceived 

 refpefting the exillence of maiiganefe. 



Neverthelefs, to clear up this doubt, I applied the fulphurcous acid to the calcined pre- 

 cipitate, in the hope of diffolving the mangancfc, if it (houid exift, without toucliing the 

 iron. But I was convinced that a fmall portion cf iron only had been taken up ; for the 

 fulphureous acid atTorded a blue precipitate by the pruDTiate of pot-a(h, and a deep green by 

 caufkic.pot-afli. This proves at the fame time that the iron does not contain manganefe, 

 and tliat this method is infullicient to denionRrate its prcfence ; more efpccially to feparate 

 it from the iron. For, on the one hand, the muriatic acid decompofes, or at leaft diflblves, 

 the pruffute of iron ; and, on the otlier, the fulphureous acid likcwife diffolves the oxide 

 of iron. 



This procefs being founded on the alTertion of chemifts, that the prufliate of iron is in- 

 foluble and not decompofed in acids; and that the prufliate of inanganefe, on the contrary, 

 is cafilydilTolved by the fame agents ; — if the laft affertion be true, the prefent experiments 

 prove that the firll is falfe. It is neverthelefs probable, that the weaker acids, fuch as the 

 acetous and fulphureous, greatly diluted, may not have fo evident an aftion on the prufliate 

 of iron ; and that they may be preferable to feparate it from the prufliate of manganefe. But 

 experiment has not yet decided in this refpefl, though it is certain that it is by no means a 

 matter of indifference what acid is employed in this operation. 



The Method of Bergman to fepnrc.te Manganefe from Iron. 



THE bad fuccefsof this procefs engaged me to follow another method propofed by Berg- 

 man, in his Differtation on the White Ores of Iron, and other parts of his works, to fepa- 

 rate iron from manganefe. 



I took one part of the oxide of iron which had been precipitated from the folution of 

 the fame fteel by the cauftic alkali, and afterwafds walhed, dried, and calcined in the cu- 

 pelling furnace. Upon this I repeatedly boiled the nitric acid to drynefs, and calcining it 

 in each operation. Laftly, I treated it with the fame acid in which a fmall quantity of 

 fugar had been diffolved. During this operation, a great quantity of nitrous gas was dif- 

 engaged, and the fluid affumed a brown colour. After evaporation to drynefs, and wafli- 

 ing of the refidue, part was diffolved in water. The folution afforded a beautiful blue 

 precipitate by the pruffiate of pot-afli, green by cauftic pot-afli, and white by the car- 

 bonate of pot-afli. 



The fulphuric acid, poured on the ferruginous folution, did not difengage nitrous va- 

 pours, but a white fume, the fmell of which confidcrably refembled that of the acetous acid. 



It is therefore proved, that the method propofed by IJergman, and followed and repeated 

 by chemifts in general, is very faulty, bccaufe it favours the folution of iron, and the acid 

 does not, when applied to a mixture of this metallic gas and the oxide of manganefe, a£b 

 cxclufivcly on the latter, unlefs by accident fuch a quantity of fugar fliould be employed 

 as might be capable of forming fo much acid only as could diffolve the oxide of manga- 

 nefe alone. But how-is it poffible to afcertain this quantity, when the proportion of the 

 two metals is unknown ? 



a A titw 



