DISODIUM HYDROGEN PHOSPHATE DODECAHTDRATE. 1589 CLXXV.-BisocZium Hyclr.oye7L Dodecahydrlc te. By DALZIEL LLEWELLYN HAMIWICK HECTOR and HENRY Boom. Phosphate KENNETH GOADBY, THE system disodium hydrogen phosphatewater has been investi-gated by Shionii ( i l l e m . Coll. Sci. E n g . Kyoto 1908 i 406) who showed that three hydrates (Na,HPO '1 2 H,O N a,H P 0 ,7 H,O , Na,HP0,,2H20) are capable of existence in contact with saturated aqueous solutions at different temperatures. I n the neighbour-hood of looo anhydrous salt is the stable solid phase. Shiomi records the following invariant points (breaks in a solubility curve) : Solid Phases. Temperature. Na2HP0,,12H,0-Na,HP04,7H,0. ........................ 36-48' Na,HP04 7H20-Na,HP0,,2H,0. ........................48.0 Na,HPO, 2H,O-Na,HP04 ................................. 95.2 The eutectic temperature (ice Na,HP04,12H20-solution) is given as -0.45O by Rudorf A 7 1 n . I'hys. Chern. 1864 [ii] 122, 337) and as -0.9O by Guthrie ("Recueil de Const'. Phys."). Certain peculiarities of the dodecahydrate having led to the suspicion that Shionii's analysis of the system was incomplete the various hydrates were examined by means of heating and cooling curves. Arsenic-f ree dodecahydrate was recrystallised and pre-parations of the other hydrates were made from it. The dihydrate is conveniently prepared by boiling the finely powdered dodeca-hydrate with ethyl alcohol. The heptahydrate was prepared by fusing together the appropriate mixture of dodecahydrate and dihydrate and cooling.The finely powdered hydrates were suspended in xylene and stirred with a thermometer in jacketed tubes immersed in a glycerol bath. Well-defined arrests were obtained on the heating and cooling curves at' the following temperatures : Na,HP04,2H20-Na2HP0 95.20 94.92 95.00 94.90 94.92 (corrected) ; mean 94.97O. Na,HP04,7H20-Na,HP0,,2H,0 48.5 48.0 48.0 47.8 48.0, 48.2; mean 48.09O. Na2HP0,,12H,0-Na2HP0,,7H20 35.0 35.1 35.4 35.0 35.0, 35.05 35.0 35.0 35.6 35.05; mean 35.0°. At the same time sharp breaks in the heating and 'cooling curves for the dodecahydrate were observed a t 29.6O (29.6 29.6 1690 HAMMICK QOADBY AND BOOTH: 29.5 29.55 29'65O). From these results it appeared that Shiomi's points a t 95.2O and 36-45O were probably too high. The transition temperature a t 35*0° found by the authors is in agreement with Tilden's (T.1884 45 409). It seemed moreover likely that tho break in the heating and cooling curves observed at 29.6O indicated a change of phase a t that temperature. In order to test this conclusion crystals separating from a solu-tion od sodium phosphate at 33O were collected and examined. They closely resemble in appearance the ordinary large monoclinic crystals obtained a t the ordinary temperatures; on keeping how-ever in a closed tube a t the ordinary temperatures (15-20°) they become opaque and friable. Analysis of the clear crystals gave the following result 2.3910 grams gave 0.7505 gram of magnesium pyrophosphate whence the number of molecules of water of crystallisation for one molecule of disodium hydrogen phosphate is 11.95.The solid phase in equilibrium with solutioas between 29'6O and 35.0° is therefore a dodecahydrate. Crystals of ordinary dodeca-hydrate kept in a closed tube in a thermostat above 29'6O (at 33') slowly lose their transparency and rigidity whereas crystals formed between 29'6O and 35-0° undergo no change. The conclusion is therefore drawn that disodium hydrogen phosphate dodecahydrate exists in two forms a and /3 the a-form being stable between 29.6O and 35*0° and the ordinary or p-form, below 29.6O. Shiomi's solubility data (Zoc. c i t . ) give no indication of any change in the solid phase in equilibrium between Oo and 36*5O. Solubilities were therefore redetermined from the eutectic tempera-ture to above the transition temperature of the a-dodecahydrate into hep t a hydrate.The temperature of the eutectic was found by stirring solutions of sodium phosphate in freshly dihlled water in jacketed tubes cooled in a brine-bath a t -3.00 to - 4 ~ 0 ~ . Steady temperatures* were obtained a t the following points which remained unchanged for at least half a minute on remosving the tube from the brine-bath -0*42O -0*47O - 0 ' 4 7 O '-0-/7O. In the first two experiments the solid phase separating first was ice. In order to determine the composition of the liquid phase a t the eutectic a solution that had been brought to the eutectic point was stirred in a freezing mixture of ice and sodium phosphate for about half an hour solid matter allowed to settle and about 5 grams of the supernatant liquid were removed with a pipette and analysed.The thermometer W~EI graduated in 1/60" and was standardised before 1lae DISODIUM HYDROGEN PHOSPHATE DODECAHYDRATE. 1591 The result is shown in the table below. The other solubilities there given were determined as follows. Saturated solutions were prepared by stirring the appropriate solid phase with distilled water in an electrically heated and con-trolled thermostat temperature being colnstant to within k0.02O. In order to make certain that' the true solid phase in equilibrium was present during the whole period of stirring an approximately saturated solution was prepared a t a higher temperature than that of the thermostat' and introduced into the solubility apparatus, together with a few crystals of the expected solid phase.Stirring was continued for three to four hours,* a portion of the saturated solution being then siphoned through a glass-wool filter into a 40' 30 20 10 0 %- ' 3 SOLUBII -,-I-L 0 5 10 small weighed bottle. The whole operation was carried o u t in the thermostat. ~~ The composition of the saturated solution was determined by conversion of the dissolved phosphate into magnesium pyre phosphate. The temperature of the thermostat was recorded with standard thermometers graduated in 1 / 20° the mercury thread being totally immersed. Seventeen determinations were com-pleted; two results were discarded owing to suspicion of leakage of water from the thermostat into! the solubility apparatus. The remainder are given below as grams of anhydrous disodium hydrogen phosphate in 100 grams of solution.* It was found that no difference greater than 0.05 per cent. was pro-duced in the value of a solubility by continuing the stirring for eight houru 1592 HUNT THE PREPARATION OF ETHYL IODIDE. Temperature. Solubility. - 0.47' (eut,ectic). 1-46 -t 6.00 * 2.73 19-95 7-26 22.77 8-93 24.15 9.53 25-78 10.90 27.80 14.16 28.65 15.57 29.05 16.04 Temperature. + 29.50" 30.10 3O.YC1 32.50 33.70 34.70 36.50 40.02 Solubility. 17-18 19-45 20.08 22-57 54.63 29.75 31-18 35.56 * Temperature at which a solution containing 2.81 grams in 100 grams of water begins to crystalliae. These results are1 plotted in the figure. The solubility curve shows distinct breaks a t 29-5-29*6O corresponding with the invariant point a-Na,HP0,,12H,0-~-Na.L~IP0,,12H,0-solutioii, and at 35.0° corresponding with a-Na2HP0,,12H,O-Na,HP04,7H,0-solution. Summary. (1) Disodium hydrogen phosphate dodecahydrate exists in two forms a and j3. The transition temperature between the a- and &hydrates is 2 9 . 6 O . The a-hydrate passes into heptahydrate at 35'0° (not 36*45O as given by Shiomi Zoc. cit.). (2) The euteclic point /3-dodecahydrate-ice is found to be at - 0*47O agreeing closely with Riidorf's value - 0 ~ 4 5 ~ (Zoc. cit.). (3) The solubilities of the two dodecahydrates have been deter-mined ; from the solubility curves the transition temperatures are found to agree with those deduced from heating and cooling curves. CHEMICAL LABORATORY, THE COLLEGE, WINCHESTER. [Received November 15th 1920.