V2G ^^' THOMSON S PURIFICATION OF GOLD AND SILVER. 



ftrongly ignited coiied Up fplrally, and put Into a crucible, the bottom of 



an'^hour*'^'^'^ ° which was covered with hlftJL-k o>iide of manganefe. I then 



All the metal added more oxide till the filver was covered, and ail the 



wasoxided. t'pace between the coils completely filled. ; A cover was then 



luted to the crucible, and a fmall hole left for the efcape of 



axigcn gafs. When this had been ex-pofed for a quarter of 



an hour to a heat fudiciejit to melt filver, I found the furface 



of the manganefe brown from the lofs of oxigen ; but, where 



the filver had been, the whole vVas one uniform black powder, 



without the leaft appearance of metallic luftre, fo that I had 



no doubt, that even the filver was become an oxide. 



The wV^c con- I then put the whole contents of the firft crucible into a 



pat'^rnto a^ai-^er ^^^^"^ ^^ ^ larger fize, into the bottom of which I put a 



crucible with gu,antity of pounded green glafs, about three times the bulk 



thrice its bulk of ^f ^j^g contents of the firft. crucible, and, luted on a cover as 

 green glais. , ^ ' ' ■ ■ , 



before, to prevent the accefs of any inflammable f(d:>fiance. 



Strong heat- The crucible was then expofed to a heat fufiiciently ftrong 



ladieducfdfhe *^ ^^^^ ^^^ S^^^^ very fluid. Upon cooling an^ breaking the 

 filver pure, and crucible, I found the filver at the bottom perfectly pure, as 

 alone in a button, j^^ oxide alone could part from its oxigen without the accefs 

 This procefs of fome inflammable fubllance. I find this procefs anfwers 

 aniwer§ e(iua y equally vvell for purifying gold, and to me it feems to polTefs 

 - • fome advantages over all the former methods. The materials 



ufed are cheap, and a large quantity can be refined as foon, 

 and as eafily as a fuiall quantity, by merely altering the capa- 

 city of the crucible you ufe. 

 The metal muft I tried the fame operation on gold and filver in round 



be in thin or iTjaifes, but found it went on very flowlv, and what I fcarcely 

 tnallmafles. ^ , . , ^ ^ ,. , r r -i • . 



expected, in the nrit part ot the procels ot oxidating the 



• metals, the remaining metal continued uniformly impure or 



nearly fo, until the whole was oxidated. 



The proporf^ons, J regret that 1 have been forced to make this matter public, 



siven^^becaufe ^ ^^^^"^^ ^ could do it in • a manner fatisfa6lory to mylelf, 1 



the author has wilhed to have given the exa6l proportions of allpy, manga- 



^nk'aJe'hi'/"'* nefe, and glafs to be generally ufed, and to have afcertained 



procefs. if there is any truth in the old opinion, that faltpetre melted 



with gold dcftroys a part of it. I fuppofe that idea may have 



arizen from the oxigen given out by the nitre in a high heat, 



oxigenating the copper contained in the impure gold, which 



has been the fubjed of the experiment. 



5 Since 



