994 SEYLER AND LLOYD: LXXXV.-Studies of the Carbonates. Part 111. Lithium Calciuni and iMacgnesiurn Carbonates. By CLARENCE ARTHUR SEYLER and PERCY VIVIAN LLOYD. I; it h iu m G’nr 71 o ti n t e . AIR was drawn through an N/10-solution of lithium hydrogen carbonate a t 25O. The percentage of carbonic acid in the air was determined as in the previous paper (this vol. p. 138)’ by draw-ing air through an N/lO-solution of sodium carbonate. c = 100.5 x 10-3 gram-equivalents per litre [H,CO,] = 0.0105 x 1OOz = hydrogen carbonate 40.65 per cent,. 100( 1 - x) =carbonate 59.35 per cent. This result is identical within the limits of experimental error with that for sodium carbonate a t the same concentration and 10-3. gives k2 k= 53%. k a2 Air was drawn through water containing lithium carbonate in 1OOz =hydrogen carbonate 23.4 per cent.lOO(1- x) = carbonate 76.6 per cent. This result agrees with McCoy’s value for sodium carbonate at Lithium carbonate when shaken with water alone (c = 338 x It is evident, therefore that the ionisation of lithium carbonate and hydrogen carbonate can bO taken as equal to that of sodium carbonate and hydrogen carbonate. A series of experiments was made the results of which although only preliminary are worth recording. Lithium carbonate was shaken with water containing increasing amounts of carbonic acid in bottles without air space until equilibrium was presumed to be reached. The proportion of carbonate and hydrogen carbonate to total base was ascertained by double titration without special pre-cautions except against loss of carbonic acid.The solid phase, even when the solution contained chiefly hydrogen carbonate, appeared to be lithium carbonate as determined by dissolving a portion in pure water and titrating as above. The concentration of the free carbonic acid was not determined but can be approxim-excess. c = 363 x 10-3 equivalents. }t 5 = 4272. the same concentration. behaved analytically like a pure carbonate. ately calculated by the equation [H2C03]=-s k .-j CHC%I2 and is k IC% STUDIES O F THE CARBONATES. PART 111. 995 small in any case. The object was to ascertain whether the solu-bility product [Li]2 x [CO,] was approximately constant. Such experiments would afford a criterion as t o whether the intermediate ion LiCO really exists since i f it does the lithium ion concentra-tion will be [HCO,] + [LiCO,1 + Z[CO,] whilst if it does not it will be [HCO,] + Z[CO,].If b is the molecular concentration of the total hydrogen carbonate and c that of the carbonate as determined by analysis, we should have in the first case [Li]=a(b + c ) + p c and in the second [Li’] = ah + 2 P c ; j3 is taken to be the same as for sodium carbonate a t the same total lithium concentration and a to be the same as for sodium acetate at the same molecular concentration, that is a t the concentration b - t c . The non-ioiiised [Li,CO,] will be (I -a)b and the [LiCO,]==c(a-P). Equivalent concentrat ion. b+ 2c. b+c. b. 338 x 10-3 109 x 10-3 0 363 224 85 364 215 66 386 248 110 406 282 158 706 635 570 818 773 728 a.0-75 0.72 0.72 0.71 0.70 0.595 0.56 CHCO3I. C. o x 10-3 169 x 10-3 61.2 139 47.5 149 78.1 138 110.6 124 339.0 65 407.7 45 8. 0.295 0.285 0.285 0-275 0.267 0.187 0.165 W,l. [Li.] [Li’]. [Li] x z[CO,]. [Li’] x 2[C0,]. 49.85 x 10-3 176.5 x 99.7 x lo-” 1-553 x 0.495 x 39.60 200.9 140.4 1.598 0.780 42-50 197.3 132.5 1-654 0.740 37.9 214.0 153.9 1.736 0-89 33.2 230.6 177.0 1.765 1.04 12.1 389.9 363.2 1-84 1-69 7.42 441.2 422.5 1.46 1.32 On the assumption of an intermediate ion the value of [Li] x [CO,] is fairly constant whereas on the other assumption there is no constancy until the effect of the intermediate ion becomes negligible. I t will be found that although [LiI2 x [CO,] is approximately constant the non-ionised [Li,CO,] diminishes so that [Li12 f c r ) 2 1 = kakb increases.Since [LiI [“3] =kb is [Li VV ] Li CO,] roughly constant* (see this vol. p. 143) it follows that the p r e duct [Li] x [LiCO,] must also be approximately constant. It seems therefore that i t is the non-ionised molecule that does not follow the law of mass action o r of which the concentration is not a true measure of the active mass. These experiments indicate a line of research which is worth pursuing by experiments in which all the conditions art3 accurately controlled. * kb increases with the concentration but much less rapidly than k,. Q Q VOL. CXI 996 SEYLER AMD LLOYD: Calcium Carbonate. Stieglitz (Carnegie Institution Pub. 1909 No. 107) and McCoy and Smit,h ( J .Anter. C'kem. Soc. 1911 33 468) have calculated the solubilit,y product [Ca] x [CO,] = li of calcium carbonate from the value of 2$[Ca] x [CO,] obtained by Schloesing (or by McCoy 7% and Smith). Stieglitz used the value k3=6*2 x 10-11 and McCoy and Smith 5.5 x 10-11. It has been shown (Part II. Zoc. cit.) that these values are prob-ably too high. = 7124 and 5 = 19-2 x 10-5. Consequenbly the value of [Ca] x [CO,]= k, obtained by Stieglitz and by McCoy and Smith is too high. Taking the result of McCoy and Smith's experiments we have 2 4 = 102.5 x 10-6 whence [Ca] x [CO,] = X = 71.9 x 10-lo a t 2 5 O . We found the values 1;,=4*27 x ki3 k, k k, McCoy and Smith calculated i t t o be 93 x 10-10 and Stieglitz 126 x 10-10. This high result explains why the calculations of the solubility of calcium carbonate in pure water have hitherto been materially larger than the value found by experiment.gram-molecules per litre at 2 5 O whilst Kendall (Phil. Mag. 1912, [vi] 23 958) found only 14-33 x 10-5. We recalculate it on a basis of [Ca] x [C03]=71*9 x 10-10 as follows The pure calcium carbonate is largely hydrolysed a t this dilution thus : where the concentration [HCO,] = [OH]. McCoy and Smith for instance calculated it to be 16.6 x CaCO + 2H,O = Ca(O13)2 + Ca(HC0,)2, Therefore [Gal = [CO,] + R CHCO 1 + - [OH] - [CO,] + [HCO,]. 2 2 Also If 1 0 0 ~ is the percentage of the calcium carbonate hydrolysed, [HCO,] = z[Ca], [CO,] = (1 - x)[ca], But [Ca] x [C03]=[Ca]2(1-x)=71.9 x 10-lO STUDIES OF THE CARBONATES.PART 111. 997 We have two equations for [Ca] and x which are satisfied by [Ca]= 14.6 x 10-5 and x=O.666; that is to say the solubility of calcium carbonate will be 14.6 x 10-5 gram-molecules per litre of which two-thirds will be hydrolysed. The alteration of to 1-67 x 10-4 will make very little change. [Ca] will become 14.24 x 10-5 and x=0*645. The solubility of calcium carbonate using normal air can also be calculatced : [HC0q]2 LCOsI * L'%CO,I Therefore, k3 = 7124 [Ca] x [CO,] =71.9 x 10-10 [HUO,] = 2 ([Cat] - EC03I ) * ([Ca] - [CO,] i 2 - '7124 - - -[ H,COs]. [CO,I 4 F o r air containing 0.037 per cent. of carblon dioxide [H,CO,] a t 2 5 O will be 1-221 x 10-5. This will give [a]=54.7 x 10-5 but Ken-dall fouad considerably less.I n an experimegt conducted a t about 15O it was found t h a t after twenty-eight days the solubility of powdered limestone suspended in water through which air from outside the laboratory was drawn was 33.7 x 10-5. After sixty-nine days it rose t o 54.2 x 10-5 and after two hundred and forty days to 57.5 x 101-5 so that equilibrium is only slo cvly relach ed. Assuming 0.033 per cent. of carbon dioxide in the air [H,CO,] a t 15O = 1.475 x 10-5. This will give [a] = 58.2 x 10-5 against 57.5 x 10-5 fo'und. F o r air containing 0.0333 per cent. of carbon dioxide Schloesing found (at 18.) 54.88 x 10-3. The solution of the carbonate in equilibrium with air was found t o have a slightly alkaline reaction to phenolphthalein. (For a full discussion of the subject see Johnston and Williamson J .Anzer. Chem. Soc. 1916 38 075.) His result appears to be low. Magnesium Cnrb oilate. Bodlander has calculated the ratio from Engel's experiments on the solubility of crystallised magnesium carbonate MgCO,,SH,O in water containing carbonic acid under pressures of from 4 to 6 atmospheres. A t these tensions the carbonate is a stable solid phase. The difficulty presents itself t h a t the carbonate unlike calcium barium or strontium carbon 998 SEYLER AND LLOYD: ates is not sparingly soluble. Engel states t h a t it is soluble in pure water t o the extent of 11.5 x 10-3 gram-molecules per litre a t 12.5O, and Bodlander assunies t h a t 56 per cent. is dissociated giving a concentration of 5.06 x 10-3 for the non-ionised part which Bod-lander dedncts in calculating the concentration of the magnesium and hydrogen carbonate.However this is erroneous. Crystallised magnesium carbonate has no definite solubility in pure water. It is decomposed into basic compounds and the solution contains magnesium hydrogen carbonate with a certain amount of ionised and non-ionised carbonate. The magnesium in solution depends on the ratio of water to solid employed and the equilibrium takes some time to reach completion. The following results bearing on this question were obtained. The total carbonic acid was determined by Dittmar's ' vacuum method ' (" Quantitative Chemical Analysis," p. 227) used by him for sea-water the ' fixed ' carbonic acid by titration with methyl-orange as indicator : [COZl Time Water C 3 .CWHCO:~)~. [MgOI' (days) - P a x * 2 2 cI1IgCO.p in solid C%CO& 440 unknown 5.98 x 10-3 1.58 x 10-3 2.20 x 10-3 0.775 440 370 8.48 3.48 2.50 0.772 - 15.96 10.21 2.87 - 0.0103 x lo-' 335 unknown 31.0 22.15 4.57 0.79 47 17 47.5 34.1 6.70 0.84 338 unknown 48.9 33.9 7.50 -47 8.5 60.0 46.6 6.70 0.79 47 2.0 76.62 63.5 6.56 0.79 The subject deserves careful investigation but i t is evident t h a t the carhonate below a certain tension of carbonic acid is decom-posed until the solid phase is not far removed from hydromag-nesite 3MgC03,B!Ig(OH),,3H,0 ( ratio ~ ~ ~ ~ ~ ~ 1 - 0 * 7 5 ' ) - which is, stable over a fairly wide range of concentration of carbonic acid. The magnesium remains in solution largely as hydrogen carbonate, with a proportion of carbonate both ionise'd and non-ionised.The main rextion is approximately 5MgC03 + 2H20 = 3MgC03,Mg(OH) + BTg(HCO,),. Consequently the greater thel amount of magnesium carbonate in relation to t h s water the greater is the concentration of the solu-tion. The dissolved carbonate with increasing concentration (and when sufficient time is allowed) becomes nearly constant a t a value about 7.5 x 10-3 gram-molecules per litre. This is probably almost entirely non-ionised. The following experiments show t h a t mag-nwium c a r h n a t e is ,only slightly ionised even a t high dilutions. Air wm drawn through water containing magnesium carbonat STUDIES O F THE CARBONATES. PART 111. 999 in suspension a t 2 5 O . The percentage of carbonic acid in the air was determined by drawing it through an N/lO-solution of sodium carbonate and determining the proportion of hydrogen carbonate and carbonate.Total magnesium in solation = 15.96 x 10-3 gram-equivalents per litre. Hydrogen carbonate = 63.97 per cent. ; carbonate = 36.03 per cent. Concentration of free carbonic acid = 0*0103 x 10-3 : A similar experinleiit was made with a solution of about the same concentration b u t no solid was present. Total magnesium in solution z 18.65 x 10-3 gram-equivalents. Hydrogen carbonate = 60.48 per cent. ; carbonate = 39.52 per cent. : The mean value of It is 3566 for 2 concentration of 17.3 x 10-3 equivalents. Taking the ionisation of the magnesium hydrogen carbonate as equal t o that of magnesium nitrate say 0.84 we cal-culate 6 as follows: 3566 7124 p = - a2 = 0.353.Evidently even a t an equivalent concentration of 17.3 x 10-3 magnesium carbonate is only slightly ionised. From the above experiments we calculate the relation : It would be possible by careful experiments to determine 6 f o r higher concentrations but i t is evident t h a t a t such concentrations as obtained in Engel's experiments the magnesium carbonate will be practically non-ionised. Some experiments of Treadwell and Reuter (Zeitsch. nnorg. Chem. 1898 17 ZOO) are unfortunately not available for calculating fl since the solutions are not in true equilibrium with the gaseous phase but they show t h a t beyond concentrations of magnesium of 50 x 10-3 equivalenh the amount of carbonate is practically constant.The maximum value is 9.0 x 10-3 gram-molecules per litre. We will take this as the con-centration of the non-ionised carbonate in Engel's experiments. These wsre carried out a t 12.5O. The concentration of the free carboniz acid has to be calculated from the pressure which is not strictly correct for carbonic acid 1000 STUDIES OF THE CARBONATES. PART 111. Pressure. Atmo-spheres. 0.5 1.0 1.5 2.0 2.5 3.0 4.0 6.0 CMg-9). a. 245 x 0.686 316 0.680 374 0.675 407 0.672 435 0.670 456 0.669 509 0.666 603 0.662 %k,. 168.1 x 336.2 x 10-3 21.55 x 88.17 219.8 429.6 41.80 94.80 252-4 504.8 60.9 105.6 273.5 547.0 79.8 102.5 29 1-5 583.0 99.75 99.3 305-0 610.0 116.4 97.5 339.0 678.0 151.6 102.8 227.2 117.2 399.2 798.4 Wgl.CHCO,I. [H,CO,I. k, Mean ............... 100.57 The concentration of the CO ion is calculated from the relation : [CO 3 = ______-- [HC0,I2 a [H,CO,] x 7120’ From these figures we may calculate the value of the (‘solubility These estimates of the (‘ solubility product ” must be subject to product,” [Mg] x [CO,] namely 141.2 x k correction for the value of -2 which is strictly only known f o r 2 5 O . 4 co IT clzc s io 71 s . (1) Lithium carbonate is ionised tcj the same extent as sodium carbonate. The ionisation takes place in two stages Li,C03 = Li + LiCO and LiCO,=Li + CO,. Consequently the concentration of the lithium ion is more than double t h a t of the CO ion and can be calculated. If this is done it is found t h a t the “solubility product,” [LiI2 x [CO,] is practically constant over a wide range, thus confirming the assumption made.(2) The solubility product of calcium carbonate [Ca] x [CO,] is about 71.9 x 19-10 and has hitherto been put a t too high a value. It is shown that- this agrees with a 2olubility of calcium carbonate in pure water of [Ca]= 14.6 x 10-5 and t h a t the salt is hydrolysed to the extent of 66 per cent. (3) Crystallised magnesium carbonate has no definite solubility in pure water. It decomposes into basic carbonates and magnesium hydrogen carbonate whilst a certain amount of carbonate is also dissolved. Over a wide range the reaction approximates t o 5MgC0 + 2H,O = 3MgCO,,Mg(OH) + hlg(HCO,),. The larger the amount of carbonate in relation to the water the larger is the amount of dissolved hydrogen carbonat’e but the carbonate tends t o reach a limit. (4) It is shown t h a t a t an equivalent concentration of 17.3 x 10-3 magnesium carbonate is only ionised (as regards the CO ion) to th INTERNATIONAL ATOMIC WEIGHTS. 1001 extent of about 35 per cent. and probably a t high concentrations it is only very slightly ionised. The " solubility product " of magnesium carbonate a t concentra-tions of free carbonic acid a t which the carbonate' is a stable solid phase has been calculated from Engel's experiments and is found to have the1 value 141 x 10-6. [Received August 25th 1917.