2282 THORPE AND SIMMONDS: LEAD SILICATES IXCCXXX1X.-Lead Silicates in Relation to PotteryManufacture. Part 11.By SIR EDWARD ‘FHORPE AND CHABLES SIMMONDS.SONE years ago, in connexion with the question of lead poisoning inthe pottery industry, we made a study of various complex lead silicatesemployed in the production of ceramic glazes (“Lead Silicates inRelation to Pottery Manufacture,” Trans., 1901, 79, 792). I nparticular, we showed that the quantity of lead which could bedissolved from a silicate by dilute acids depended primarily on the typeof silicate. Monosilicates are easily attackable. Polysilicates anddisilicates are but slightly so, and this is the case whether theycontain little or much lead. We further showed that in the caseof the polysilicates and disilicates, as ordinarily prepared, there isfrequently a small quantity of an easily soluble lead compound mixedwith the bulk; this could be extracted with dilute acid, leaving theresidue practically unattackable under the conditions of the experi-ment.A fairly complete summary of present knowledge of lend silicatesand glazes in respect of the foregoing properties has recently beenpublished by Beck, Ldwe, and Stegmuller,* who have studied the action* “Zur Kenntniss der bleihaltigen Glasuren und deren Bleiabgabe an saureFliissigkeiten ” (Arbeit.K. Gesundhcitsamte, 1910, 33, No. 2)RELATION TO POTTERY MANUFACTURE. PART 11. 2283of acids both on the powdered silicate and on the fused glaze as itexists on the finished pottery ware.With certain modifications, theyhave repeated our experiments; but whereas our studies, made onsilicates in actual commercial use, were of necessity chiefly concernedwith the complex silicates, containing not only lead but other bases,Beck, Lome, and Stegmiilier have adopted what in some respects is thebetter plan of studying first the simple lead silicates.The results they obtain lead, however, to the same conclusions as ourown. Thus the proportion of lead oxide dissolved from three simplesilicates was found to be as follows :PbO dissolved by 1 per cent. HNO,.A/- >PbO present Percentage ofSilicate grams Amount total PbOPhO,SiO, ............ 1’529 1.525 99-7Pb0,2Si02.. .......... 1’232 0-106 8.6Pb0,3Si02.. .......... 1 *038 0.019 1 -9That is, practically all the lead is dissolved from the monosilicate,but a relatively small quantity only from the disilicate and trisilicate.On a further treatment of the di- and tri-silicate residues, very littleadditional lead was extracted (0.5 and 0.6 per cant.respectively).This confirms what has already been stated (Zoc. cit., p. 802).An explanation of the fact that a single treatment with acid extractssubstantially the whole of the “soluble ” lead from the di- and the tri-silicates is suggested by the German authors. A t the temperature offusion a certain amount of dissociation may occur ; thus the disilicatePbSi,05 may partly dissociate into PbO + 2SiO,, or into PbSiO, + SiO,,and a portion of these components may remain dissociated when thesilicate cools.Since both lead oxide and monosilicate are readily solublein dilute acids,they would, if the silicate were sufficiently finely powdered,be all extracted on a single treatment with acid, whilst even from arelatively coarse powder the bulk of this “soluble” lead would beremoved.In the paper quoted we give a table (pp. 796-797) illustrating thefact that whether a lead silicate yields much soluble lead or notdepends mainly on the value of the ratio : number of acidic molecules/number of basic molecules. Beck, Lowe, and Stegmuller remark t h a twhilst this appears capable of rendering good service for rapid sorting-out purposes, it gives no special insight into the effect of a particularconstituent in individual cases.They note that the effects of boricoxide and alumina are contrary to what would be expected from thetable, and suggest that the possibility of the formation of complexborosilicates and aluminosilicates may have been overlooked. Also theyremark that the possibility of alumina acting as an acidic oxide mustnot be excluded (Zoc. cit., p. 226).The authors in question, however, have quoted the table from a,7 ~ 2284 THORPE AND SIMMONDS: LEAD SILICATES INParliamentary Paper in which it was reproduced (‘‘ Lead Compoundsin Pottery,” 1901, Cd. 679, p. 26-27). On reference to the originalpaper already referred to ( h e . cit., p. 799), it will be seen that thepoints they mention had not been lost sight of. We note there that“ subsidiary factors may exist in the possible different states of com-bination in which aIumina and boric oxide may occur in the silicate ” ;and we also say ‘‘ it is conceivable that in some cases the alumina mayact as anacid constituent.In such cases, the amount of lead dissolvedwould presumably be less than indicated by the value of the ratio.”Further on we briefly discussed the influence of boric oxide.These views, put forward tentatively, were arrived at from acomparison of the results yielded by various complex silicates of verydiveroe composition; but it was recognised that the best way ofstudying the matter was to ascertain the effect of each oxide singlyrather than to deduce it from such comparisons.A number of silicates were therefore prepared by fusion of theingredients in the proportions required for certain polysili2ates anddisilicates of definite formula. These silicates, in a finely-powderedcondition, were then shaken continuously for an hour with 1000times their weight of dilute hydrochloric acid (0.25 per cent.), and thequantity of lead dissolved was determined by a colorimetric methoddepending on the comparison of the depth of tint with that given byknown quantities of freshly produced lead sulphide.The results may be thus stated :(I.) IfiJuence of Alumina.Polysilicates&SiO,........... 28’8 35.1PbO ............ 71.2 60.9A1,0, ......... - 4.0100.0 100.0(1) (2)- -- -PbO dissolved 25 5Disilicates&(3) (4)35.1 41 *964’9 54‘53%100-0 100*0-- -- -5 5 per cent.No.1 is the simple lead polysilicate 2Pb0,3Si02( = 10Pb0,15Si02),and No. 2 corresponds with the complex polysilicate7Pb0,A120,,1 5Si0, ;thus the substitution in No. 2 of one molecule of A1203 for threemolecules of PbO has had a very marked effect in decreasing thesolubility of the lead.No. 3 is the simple disilicate Pb0,2Si02( = 10Pb0,20Si02), andNo. 4 is the complex disilicate 7PbO,A1,0,,2OSi0,. Here thesubstitution of Al,03 for 3Pb0 has not affected the proportion ofsoluble lead, or at least, not sufficiently to be evident under thRELATION TO POTTERY MANUFACTURE. PART 11. 2285conditions of the experiment. A similar conclusion appears tohave been arrived a t recently by H. Eisenlohr (Sprechsnal, 1910,43, 389).(11.) Iwfluence of Sodium Oxide.Polysilicate Disilicate( 5 ) (6)5iOZ .....................31 *O 37.9Na,O ..................... 3 -1 3’3PbO ..................... 65.9 58 -8- -100.0 100*0 - -YbO dissolved . , . . , , .. , 44 13 per cent.No. 5 is the polysilicate 2Na20,12Pb0,21Si0, ; it is to be comparedwith No. 1 (= 14Pb0,21Si02).No. 6 is the disilicate Na20,5Pb0,12SiO2 ; its comparison sample isNo. 3 (=6Pb0,L2Si02)’; thus in both cases the substitution ofone molecule of Na,O for one of PbO has materially increased theproportion of soluble lead.(111.) Influence of AZunaina and Xodiunz Oxide Yogether.Polysilicate(7)SiO., .................... 83’8N%O ..................... 1.5PbO ................... 62 *3h1,0,. ...................2 -4-100’084-PbO dissolved . , . , , .Disilicate40‘456 ‘22 -11.3100’0(8)--5 per cent,No. 7 is the polyailicate Na,O,Al2O,,12Pb0,24Si0,. It is to becompared with No. 1 and No. 2, also with No. 5. Relatively toNo. 1, a considerable lowering of the proportion of soluble leadhas been effected, and this, in accordance with (I) and (11), is tobe attributed to the alumina. As compared with No. 2, a largerquantit,y of soluble lead is shown, in agreement with the factthat No. 7 contains a smaller percentage of alumina than No. 2.The proportion ofsoluble lead agrees with that in the simple disilicate No. 3.No. 8 is the disilicate Na,0,A120,,1 2Pb0,32Si02.(IV.) Influence of Bm-on Trioxide.No borosilicates of defioite molecular formulae were made, butthe following results were obtained with modifications of the simpledisilicate 2286 LEAD SILICATES IN RELATION TO POTTERY MANUFACTURE.Modifications30Did icatc - (9) (3150) (1)B,O, -SiO, ........................35 '1PbO 64.9 65 60 ....................... 5 5100.0 100 100........................- - -- - -PbO dissolved ............ 5 53 46 per cent.Thus the substitution oE 5 per cent. of boron trioxide for the samequantity of either silica or lead oxide in the simple disilicate enormouslyincreases the quantity of soluble lead. This effect, however, is notproduced with the complex disilicates, or a t least, not with those of aparticular type. We had already noted (Zoc. cit., p. '799) t h a t from4 to 6 per cent.of boron trioxide was present in certain borosilicateswhich yielded mere traces of lead to the action of solvents. Butin these cases the proportion of silica was higher, and that of the leadoxide much lower, than in the two experiments above described, andnotable quantities of lime and alumina were also present. Which ofthese factors determines the behaviour of the boron trioxide is atpresent obscure. It was precisely the influence of the variousconstituents that we had hoped to ascertain if the work had beencompleted.The results obtained point clearly to the following conclusions :(1) Alumina has a marked effect in promoting the stability of thepolysilicates towards acids. Possibly this is due to its function in thesilicate being that of an acidic oxide.I n the disilicates this effect isless apparent, since these are already tolerably stable.(2) Sodium oxide appears to increase the solubility of the lead,(3) Boron trioxide in some cases renders the silicates much moreeasily attackable, but in others has no such effect.Some experiments described by Beck, Lowe, and Stegmuller lead, inthe main, t o similar conclusions (Zoc. cit., p. 214). With the simplelead silicates they mixed 2+ or 5 per cent. of one or other of thefollowing compounds : Al,O,, B203, CaO, Na,B,07. The mixtures werefused, and the proportion o€ soluble lead was determined in the resultingcomplex silicates. Since by this procedure the acidity of the originalsilicates is modified, the results are not closely comparable withthose that we adduce. Thus the addition of 5 per cent. of lime to thedisilicate raised the percentage of soluble lead from 5.3 to 20.0 ; butthis experiment is not analogous to those we show in (11) with sodiumoxide, because the new compound is notably more basic than theoriginal, and would therefore in any case be expected to yield moresoluble lead. Nevertheless, the general results obtained point in thespecially in the polysilicatesABSORPTION SPECTRA OF SOME SULPHUR COMPOUNDS. 2281same direction as our own. Thus the effect of an addition of5 per cent. of alumina could be traced in the decrease, by about 2 percent., of the soluble lead in the disilicate (no experiments on the poly-silicate are shown), and whilst 5 per cent. of borax raised thesolubility figure of the disilicate from 5.3 to 32.0, the effect onthe trisilicate was very slight, namely, an increase of only 1 per cent.Results of the same order were obtained with 5 per cent. of borontrioxide