G L A

G L A

( 490 )

drinking veffeis. The arfenic is conftantly vo-latilizing from the arfenical glafs when pre-paring, that is, till it is intimately united withthe reft of the glafs, after which however itcannot again be feparated by heat or any com-mon means.

As arfenic is entirely volatilized when in con-tact with any carbonaceous matter another ufehas been made of it, which is to difperfe anyfuch matter which may remain in the glafs,owing to a defect in the calcination of thealkali, or any other caufe. c When this hap-pens, fmall lumps of white arfenic are fometimesthruft down to the bottom of the glafs-pots andftirred in with the contents, and the arfenicmeeting with the unburnt carbon diffufedthrough the glafs unites with it, is fpeedilyvolatilized with it, and the glafs is left freedboth from the adhering carbon, and alfo frommolt of the arfenic that was added. Themotion excited through the melting glafs by thevolatilization of the arfenic is alfo thought ufefulin haftening the compleat vitrification of theingredients.

Nitre is ufed in glafs-making only in fmallquantities, and is an acceffary ingredient tofulfil particular purpofes. This fait at a heateven much below that of the glafs-pots is readilydecompofed, giving out much oxygen, nitrousgas and azot, and leaving behind its pure pot-afh. It is of fervice in deftroying any carbo-naceous matter in the ingredients of the glafs(with which it fhould be mixed before fufion)and its ufe in fixing arfenic, and in keeping upthe tinging power of manganefe has been al-ready mentioned. The fame circumftance ofkeeping metallic oxyds up to their highefl ftateof oxygenation, alfo renders this fait often ufeful,and even neceffary in the preparation of fomeothers of the coloured glalfes.

Ofthe proportion of Silex to the Saline fluxes.

The exa£t proportion and number of ingre-uients that enter into every fpecies of glafs can-not eafily be obtained with any certainty, thoughmany apparently good mixtures are given bydifferent authors which will be afterwards men-tioned. But fome obfervations may be madeon the folvent power of mere alkali over filiceousearth, which is the bafis of the rules for thecompofiticn of glafs, .as, ftridly fpeaking, no-thing elfe is neceffary for the formation of glafsthan a folution of filex in alkali by a meltingheat, and all the other additions are more pro-perly ufeful than neceffary ingredients.

* Lej'fei, Art, de la Vcrreric, 4 I

When filex is melted with twice its weight,or more, of dry carbonated alkali, either potafhpr foda, the refult is a very foft deliquefcentvitreous mafs, always more or lefs opake,ftrongly alkaline to the talte, and which onexpofure to moift air, or more fpeedily if putinto water, totally diffolves into a clear liquorwhich is a folution of filex in alkali. Thisfilicited alkaline folution is decompofed by allthe acids, which feparate the filex in the form ofa white powder.

When the alkali employed (meaning all alongin this place the dry carbonated alkali) onlyequals the filex in weight, or at leaft does notmuch exceed it, the glafs is now tranfparent, butis ftill foluble in water as before.

It is not till the alkali is diminilhed to aboutone half of the weight of the filex that the glafsbecomes perfectly hard and infoluble in anycorrofive liquors, (the fluoric acid excepted) andin fhort, acquires the character of a perfectglafs.

This proportion therefore, that is, two partsof fand to one of alkali is ufually the datum onwhich the dofes of the fpecies of alkalies actu-ally ufed are regulated. d Thus if commonwood alhes (of which the alkaline part is reck-oned at no more than io per cent.) are em-ployed, ioo lbs. of thefe would require nomore than about 20 lb. of fand. If the beltSpaniffi barilla containing from 45 to 50 percent, of carbonat of foda, be ufed, an equalweight of fand may be added; but if purifiedpearlalh be taken, it will melt down perfectlytwice its own weight of fand.

But glaffes compofed merely of pure alkaliand fand require a very ftrong fire for theirfufion, and are hard, harfh, and difficult towork. They are therefore never ufed alone,for even when common affies and fand are theonly ingredients, the affies contain an abun-dance of earth and other fubftances. As onehalf the weight of the fand is reckoned anabundant allowance of alkali, it follows ofcourfe that when litharge, arfenic, borax, orany other fluxes are employed, the quantityof alkali will be proportionally diminilhed.

It mud not be fuppofed that the glaflesthemfelves contain nearly the original propor-tion of filex and alkali that was put into theglafs-pots. Of thefe two materials the alkaliis abundantly volatile in the intenfe five ne-ceffary for glafs-making, but the filex is abfo-lutely fixed, and hence in proportion to theftrength and continuance of the fire the relative■cyclop. Arts & Metiers. Article Verre.