COP

COP

r 350 )

The great principle on which all the modesof purifying bell-metal depend, is the muchmore ready oxidability of tin by the unitedadtion of heat and oxygen, than copper. Henceeven when bell-metal is limply kept melted inan open veflel, a degree of feparation of the twometals begins inftantly, the tin oxidating muchrafter and fooner than the copper, and of courfethe proportion of copper being therefore con-ftantly increafing in the fluid metal below.

Another equally important circumftance alfodepending on the much ftronger affinity of tinfor oxygen is, that when oxyd of copper ismixed with tin at its loweft ftate of oxygena-tion, the tin ftill retains its fuperior affinity foroxygen, and deprives the oxyd of copper of thisprinciple, and the products are, tin highly oxy-genated and in a pafty femi-fluid mafs, andcopper in the reguline ftate, partly collected atthe bottom of the veflel and partly in fmall glo-bules entangled in the oxyd of tin.

The direct experiments of Fourcroy on thispoint are valuable.

A bell-metal was firft made with 80 parts ofcopper and 20 of tin, both pure.

Of this alloy ioo parts were put on a muffleand expofed for a certain time to a red heat,with accefs of air. The whole was convertedto a grey oxyd and weighed 104. parts. It wasthen heated ftrongly tor half an hour, andyielded a button of pure copper weighing 54parts. Coul'equently 2 6 parts of copper re-mained unreduced and mixed with the fcoriteof the oxyd of tin.

A hundred parts were calcined as above tillthey increafed to 117. This being ftronglyheated for half an hour gave a brown mafs, fromwhich no copper feparated, except a few inter-fperfed globules. This therefore was too muchoxygenated, that is, the tin had combined withfo much oxygen in the firft operation that itcould not afterwards feparate any from theoxyd of copper.

A hundred parts were treated in the fame waytill they increafed to 112. This melted into abrown mafs containing more globules of cop-per than the laft, but ftill the greater part re-mained unreduced.

Hence it appears that even an increafe of 12on 100 is too much, and by other experiments,the above chemift fixes the proper point ofoxidation at which oxyd of copper is reducibleby oxyd of tin to be from 5 to 7 on 100 ofcommon bell-metal. However, when too highlyoxygenated metal is mixed with the requifite

proportion of frefh metal, the former ferves asa reducing flux to the latter, and the whole maythen be made to yield a large portion of puri-fied copper. This very ingenious plan has beenfound to anfwer extremely well in experimentsmade in the large way, as will prefcntly bementioned.

Several fubftances have been tried that mightaflift in the firft oxygenation of the bell-metaland fhorten the procels. Of thefe, nitre andoxyd of manganefe anfwer the mod completely.For experiment in the fmall way, ioo parts ofbell-metal were powdered and mixed with about14 of nitre, and heated, at firft flowly, after-wards ftrongly, and gave from 57 to 63 parts ofcopper, very malleable, but not quite pure.With a larger quantity of nitre the copper ispurer, but more is loft in the fcorise.

A hundred parts of bell-metal mixed with 25parts of oxyd of manganefe, covered with brokenglafs, and ftrongly heated for an hour, gave36 parts of very good copper. It is of fomeimportance to add glafs or fome other falineflux, as this brings the whole into thoroughfufion, and allows the copper to fubfide throughthe vitrified oxyd of manganefe. Where thisis not done (and in the large way it would addto the expenfe) much of the copper is entangledin the vitrified manganefe, and a fubfequentoperation is required to feparate it.

In the precedes of two other chemifts, faitand fand have each been found ufeful additionsin the refining of bell-metal.

We may add the detail of two trials made inthe large way by z Pelletier and Darcet by orderof the French government, the one in which aportion was firft oxydated, and this ufed as areducing flux for the remainder, without anyforeign addition whatfoever: and the other inwhich the refining was affifted by oxyd of man-ganefe. It may be premifed that the manipu-lation of large quantities is not the fame as withfmall, particularly iri the cafe of reduction togrofs powder, which may be eafily done in aniron mortar on a few ounces or pounds, but notwith any economy on large weights. Largemaffes therefore are firft heated red-hot, whenthey may be eafily broken up and fpread aboutby an iron bar, and this is the way conftantlypra£tifed.

The firft procefs was that of refining withoutaddition. For this purpofe 400 pounds ofmetal were put on a furnace, and when red-hotbroken up with an iron bar, and fpread aboutwith conftant ftirring till the whole was con-

* Memoires de Pelletier, tom. a,