FLECK: THE RELATION OF URANOUS SALTS TO THORIUM. 247XXVIL- The Relation of Uranous Salts to Thorium.By ALEXANDER FLECK.IN the beginning of 1913 the law governing the evolution of theradio-elements through the periodic table was formulated, and itwas then shown that when an element gave off an a-ray, the numberof the group of the periodic system t o which the resulting productbelonged was diminished by two units, and that when an elementgave off a &ray the number of the group was increased by oneunit. It was further shown that when any number of radioactiveelements occupied one place of the periodic table these elementswere non-separable from one another (Russell, Chem. News, 1913,107, 49; Fajans, Physikal. Zeitsch., 1913, 14, 131; Ber., 1913,47, 422; Soddy, Chem.News, 1913, 107, 97; Juhrb. Badioaktiv.Elektronik., 1913, 10, 188). Thus mesothorium-1 gives off two8-particles and then one a-particle to become thorium-X. Thesetwo substances cannot be separated from one another once theyhave been mixea fogether in solution, although they have a differentatomic weight. Such non-separable elements have been termed“ isotopic elements ” or “ isotopes ” (Soddy, Nature, 1913, 92, 399).So far as radioactive changes show, the successive places betweenthallium and uranium correspond with unit difference of chargein the atomic constitution, as has been proposed generally for thewhole table h y Van der Broek (iVature, 1913, 92, 372), andrecently verified for the elements calcium t o zinc by the experimentsof Moseley (Phil.Mug., 1913, [vi], 26, 1024). The point hereexperimentally tested was whether the change of atomic chargein ordinary electrochemical change of valency in one element wouldproduce changes of chemical character indistinguishable from thoseobserved in radioactive changes. The oxidation of quadrivalenturanium (uranous compounds) t o sexavalent uranium (uranylcompounds) may be regarded as a process in which the uraniumatom loses two electrons, and it is analogous t o the process in whichuranium-X, isotopic with thorium, loses tpio 8-particles, and formsuranium-2, isotopic with uranium. The point examined was there-fore whether uranous salts, so long as their valency is not allowedto change, would prove to be non-separable from thorium-salts.If so, it might reasonably be concluded that the electrons concernedin the two cases came from t.he same part of the atom, the outershell rather than the central nucleus on Rutherford’s theory.When the literature concerning thorium and uranous compoundswas examined, it was seen that, apart from their differences ofcolour, both possessed chemical properties very much alike.Ther248 FLECK: THE RELATION OF URANOUS SAT,TS TO THORIUM.did not appear to be any reaction in which it was definitely provedthat thorium could be completely separated from uranous com-pounds by one operation. It was therefore decided t o try some ofthese reactions which precipitated thorium and uranous com-pounds in a fractional manner, and then to examine the relativeproportions of uranium and thorium in the various precipitates.Ifthorium and uranous salts are chemically non-separable, then theratio of the quantity of thorium to the quantity of uranium willbe constant throughout one experiment. This ratio will, of course,not be affected by estimating the uranium in the uranyl condition.Uranous compounds are easily oxidised to uranyl compounds ifallowed to come into contact with the air. It was thereforeessential that the various precipitates should be collected, and thefiltrate again treated with a further quantity of the reagentwithout permitting the entrance of air into the mixture of thoriumand uranium in the lower state of oxidation.The principle employed in operating an apparatus for this purposewas to use differences of gas pressure t o move the liquid, eitherclear or holding a precipitate in suspension, from a vessel into thefilter pump and back again to the vessel.I n the case of the firstapparatus that was constructed, the air was exhausted by meansof a water-pump, and any gas required t o give a difference ofpressure to force the liquid from the reaction vessel or back againinto it was obtained from a Kipp’s carbon dioxide apparatus. Theliquid was thus during the progress of the experiment a t a pressureof only a few centimetres of mercury. This first apparatus wasopen to the objection that through any small leak air would travelinwards and come into contact) with the mixture of thorium anduranous salts.Two reactions were used with this apparatus,namely, the fractional precipitation of a mixture of thorium anduranous salts, (1) by gradually decomposing by boiling the excessof ammonium carbonate holding these salts in solution, and (2) bythe addition of successive small quantities of oxalic acid to anacid solution of the mixed salts. I n both of these experiments theprecipitates were treated with nitric acid and the uranous saltsthus oxidised to the uranyl condition before the quantities ofthorium and uranium were estimated by the usual well-knowngravimetrical methods. I n the first case it was found that uranouscarbonate is more insoluble than thorium carbonate, that is, thefirst fractions that were precipitated contained more uranium thanthe latter fractions.In the oxalic acid experiment the contrarywas found to be the case, and it was seen that uranous oxalate wasmore soluble than thorium oxalate.I n both of these experiments it was noticed that after no furtheFLECK: THE RELATION OF URAKOUS SALTS TO THORIUM. 249precipitates were obtained the solution that remained possessed thecharacteristic green coloration of uranyl compounds. The reactionswere carried out in a partial vacuum a t a temperature approaching100°, and there were present large amounts of compounds con-taining oxygen. Under these conditions these compounds mightvery readily be reduced, oxidising some of the uranous salts in theprocess. Thif, oxidation process will be continuous during theprogress of the experiment, and consequently, even if thorium anduranous salts are chemically identical, the ratios of the quantitiesof the two elements will vary in different precipitates.Neitherof these two experiments can therefore be regarded as conclusivethat thorium and uranous ions have different chemical properties.KIjIn order t o overcome the objections mentioned above t o thoseexperiments, it was decided t o precipitate fractionally a mixtureof thorium and uranous salts in the cold, and so decrease thereadiness with which oxidation would take place. For this purposea cold solution of potassium fluoride was used as the precipitant.The experiment was also conducted in an apparatus free from airand containing carbon dioxide a t a pressure greater than atmo-spheric.In this way, if there was any leak in the apparatus, gaswould force its passage outwards, allowing no air t o come in. Theapparatus shown in the figure was constructed. Excess of pressureis obtained from a carbon dioxide cylinder, and a safety valve isformed by dipping the long limb of a T-tube below four feet ofwater. By means of a three-way tap a pressure can be applie250 FLECK: THE RELATION O F URANOUS SALTS TO THORIUM.to the liquid in the tap funnel to force it into the reaction vessel A ,or to the liquid in the reaction vessel t o force i t into the filterpump. By exerting a pressure on the liquid in A , and openingthe taps 31 and N , the liquid is forced into the Buchner funnel,and passes into B in a clear condition.The taps ill and N areclosed, and the three-way taps a t K turned so that pressure isexerted on the liquid in B. The tap €2 is opened in the properdirection, and the liquid in B flows back into the vessel A , whereit can be subjected t o a further treatment with the reagent. Awooden collar is used t o clamp the ground glass to the top of thefilter filnnel. The mixture of uranium and thorium salts is placedin A and reduced there by means of pure zinc and sulphuric acid,a considerable evolution of hydrogen being maintained for one toone and a-half hours. The reduced liquid is then forced into B,and excess of sulphuric acid added to the vessel A . When all thezinc is dissolved, the tube entering the filter funnel is momentarilydisconnected, and the zinc sulphate solution blown out.This tubeis then replaced, and the reduced solutions forced back into thevessel A by applying a presssure on t o the top of the liquid in B.Small quantities of the precipitant are forced into this vessel fromthe tap funnel, the liquid being filtered between the addition ofeach quantity of the reagent. After the liquid is filtered it isforced back into the vessel A , and all the taps closed before thecover of the filter funnel is lifted to take out and to replace thefilter paper. I n this way, by rxsing a cold solution of potassiumfluoride, a number of successive fractions-usually five-of amixture of thorium and uranous fluorides was obtained. Thoriumfluoride is a very insoluble substance, and it was found that con-centrated nitric acid had no effect, and that sometimes even a largeexcess of aqua regia failed t o dissolve completely the uranium whenan attempt 1778s made to convert the uranous fluoride into asexavalent condition.The method that was subsequently adoptedto obtain both the thorium and uranous fluorides in solution wasto treat the precipitate with concentrated ammonium carbonatesolution and a few C.C. of 10 per cent. solution of hydrogenperoxide. On warming the zolution the whole of the mixed pre-cipitates was dissolved. The quantities of thorium and uraniumcould then be estimated in the usual way. The result of oneexperiment is as followsPOPE : FLUORONE DERIVATIVES. PART Ir. 251Potassium Fluoride Experiment.Number of Weight of Weight of Ratio ofprecipitate.uranium oxide. thorium oxide. UsO$ThO,.1 0’0201 0.1230 0.16342 0.1242 o m 5 7 0.1813 0,2841 1.0392 0 2744 0‘4586 0.2744 1-675 1.1501 0’1417 8-12A t the conclusion of this experiment the solution which remainedin the flask was quite colourless, showing that no oxidation hadgone on during the progress of the experiment. None of theobjections that were raised t o the former experiments could there-fore be applied to this one.It is quite evident that although the properties of the uranousand thorium ions are similar, yet there is a distinct difference, andthat they can be separated chemically from one another. Thismeans, then, that, there is an essential difference between the lossof two electrons by means of electrochemical change and the lossof two electrons expelled as two successive &rays. The theoreticalconsiderations involved have been discussed by Soddy in a letterto Nafure (1913, 92, 399) on “Intra-atomic Charge,” and allthat requires to be said here is that so far as these experimentsare concerned one must definitely regard the electrons expelled inP-ray changes as coming from the central nucleus of the atom, onRutherford’s theory, and not from its external ring. The electronsof the external ring can either give up or receive other electronsfrom the exterior as in ordinary electrochemical change of valency.There are electrons in the nucleus, but in chemical changes thereis no transference of them t o the outer ring.I desire to thank Mr. Soddy for suggesting this research, andfor his advice in connexion with itPHYSICAL CHEMISTRY DEPARTMENT,GLASGOW UNIVERSITY