816 SODA.— Reactions in Black-ash Furnace.

could only be formed by a small excess of causticlime existing alongside of it.

Kolb tested the question in a different manner,viz.—on a large scale. He composed two mixtures,one corresponding to Dumas’ equation, but withthe excess of coal necessary in practical furnacework, thus:—

I. 2Na 2 S0 4 + 3CaC0 3 + 120 = 2Na 2 C0 3 + CaO,2CaS

+ 10CO + 3C.

The other corresponding to a formula proposed byDubrunfaut, and supposing only CaS, viz.:—

II. 3Na 2 S0 4 + 3CaCOo + 120 = 3Na 2 C0 3 + 3CaS+ 1200 .

If, however, the reaction does not consist in theformation of CaS but of CaO,2CaS, according toDumas, Kolb’s second mixture could only give thefollowing results, as Landolt has shown :—

II. a. 3Na 2 S0 4 + 3CaC0 3 + 120 = 2Na 2 C0 3 + Na 2 S

+ Ca020aS + 1200 + C0 2 .

That is to say, only two-thirds of the sodium sul-phate employed are converted into carbonate, andone-third into sulphide. The two mixtures werefurnaced in a hall-furnace under exactly equal con-ditions, and lixiviated in the same manner. Theanalysis of the liquors showed that the followingproducts had been formed from 100 parts of sodiumsulphate:—

Mixture.

I.

2Na2S04 + 3CaC03.

ii.

3Na>iSo4 -{- 3CaCo3.

Na 2 C0 3 .

52-601 = 67-0

59-511 =65-3

Na 2 0,.

8-45) Na 2 C0 3 .

3-39 J Na 2 C0 3 .

Na. 2 S,.

0-15

2*16

Na 2 S0 4 ,.

0-93

6-71

This shows that the quantity of sodium carbonateformed was equal in both cases; there was in thesecond case only 2T6 Na 2 S formed against 18-3required by equation ILa., and the decompositionmust, therefore, have taken place according to equa-tion II. The mixture I. yields a surplus of freelime, which explains the larger percentage of Na 2 0in the liquors.

Kolb’s experiments, which, it will be observed,agree with Gossage’s statements, have never beenrefuted, and since that time (1866) it has beengenerally assumed that black-ash does not contain acalcium oxysulphide ; but that it must be consideredas essentially a mixture of sodium carbonate, calciumsulphide, and free lime.

Dumas’ view of the process which takes place inthe furnace itself has also been opposed. Dumas’very simple equations quoted above were soon re-cognized as untenable, the process being decidedlymore complicated; but Unger, Brown, Gossage,and Dubrunfaut, who each suggested a differentformula, agreed with Dumas in assuming that thecoal is converted entirely into CO. Scheurer-Kest-ner found (1864) that Na 2 S and CaC0 3 ignitedtogether entirely decompose each other; he alsofound (like Unger) that on heating sodium sulphate

and coal in a retort no CO, but merely C0 2 , isevolved, whichever of the two ingredients may bein excess. On the other hand, when chalk and coalwere ignited together CO was largely formed (88per cent. CO against 12 per cent. C0 2 ); but thedecomposition requires a much higher temperaturethan the reduction of sodium sulphate. On thisScheurer - Kestner bases the following statementof the process in the ball-furnace, in which the threeequations a b c are supposed to occur consecutively,and d to express the sum total of all:—•

a. 5Na 2 S0 4 + IOC = 5 Na 2 S + 10CO 2 .

b. 5Na 2 S + 5CaC0 3 = 5Na 2 C0 3 + SCaS.

c. 2CaC0 3 + 2C = 2CaO + 4CO.

cl. 5Na 2 S0 4 + 7CaC0 3 + 12C = 5Na 2 C0 3 + 5CaS+ 2CaO + 10CO 2 + 4CO.

This theory assumes that all the C0 2 required forthe formation of sodium carbonate is furnished bythe chalk.

Kolb based his view, different from the above,on the following experiments (1866) :—Two retorts,one containing a mixture of sodium sulphate andcoal, the other chalk and coal, were exposed in thesame furnace to a gradually increasing temperature.At a low red heat there was given off, almost exactlyat the same time, C0 2 from the first and CO fromthe second retort. Pure chalk, contained in a thirdretort, gave off C0 2 when bright red heat had beenreached. When a mixture of Na 2 S0 4 + CaC0 3 +3C was heated, a mixture of C0 2 and CO wasevolved, showing that the coal acted contemporane-ously on both bodies. If therefore the chalk loses itsC0 2 at the same time whilst the sulphate is being re-duced to sulphide, the reaction, lsra 2 S + CaC0 3 =Na 2 C0 3 + CaS, becomes impossible, and it must beindifferent whether calcium carbonate, quick lime,or hydrated lime is used for the soda mixture.Kolb tested this on a large scale, and found itentirely borne out by the facts. He concluded thatthe C0 2 of the Na 2 C0 3 must be furnished either bythe reduction of Na 2 S0 4 with C, or by the gases ofthe fire. He tried to decide between these two, byigniting a normal soda mixture in covered crucibles,and he found that this did not yield a good blackash; he therefore concluded that the C0 2 of thefire gas is absolutely necessary for the formation ofgood black ash, and thus explained the importanceof thorough “ paddling ” and “ raking ” in thefurnace. His theory is embodied in the followingequations:—

a. Na 2 S0 4 + 2C = Na 2 S + 2C0 2 .

b. CaC0 3 + C = CaO + 2C0.

c. Na 2 S '+ CaO + C0 2 = Na 2 C0 3 + CaS.

In c the C0 2 is supposed to be furnished by thefire gas.

This theory was opposed by Scheurer-Kestner (1867), who, like Kynaston and even Leblanchimself, found that altogether normal black ash canbe obtained by heating the mixture in closed cruci-bles. He also found that this reaction took placebefore a mixture of chalk and coal or pure chalkwere decomposed under exactly similar conditions.The C0 2 must therefore be furnished by the chalk,