THE GIBBS FREE-ENERGY OF FORMATION AND HEAT-CAPACITY OF BETA-RH2O3 AND MGRH2O4, THE MGO-RH-O PHASE-DIAGRAM, AND CONSTRAINTS ON THE STABILITY OF MG2RH4+O4

The oxygen potentials of the reactions 4/3Rh + O-2 = 2/3 beta-Rh2O3 an d 2/3MgO + 4/3Rh + O-2 = 2/3MgRh(2)(3+)O(4) were measured using electr ochemical cells of the type Pt, metal + oxide \CSZ\YDT (air), Pt. Rela tive to a reference pressure of 1 bar we find (mu O2(beta-Rh2O3))(+/-6 3 J.mol(-1)) = -278500 + 283.8T - 11.69T ln T (860 K < T < 1355 K) and (mu O2(MgRh2O4))(+/-110 J.mol(-1)) = -297314 +/- 369.405T - 23.338T I n T (940 K < T < 1495 K). The constant pressure heat capacities of bet a-Rh2O3 and MgRh2O4 were measured with a differential scanning calorim eter operated in step heating mode between 360 K and 1065 K. Best fits to the data (in J.mol(-1).K-1 with an uncertainty of +/-2 J.mol(-1).K -1) give C-p(beta-Rh2O3) = 123.6 + 0.0141T - 208.8T(-0.5) - 2312000T(- 2) and C-p(MgRh2O4) = 174.0 + 0.014T - 4297000T(-2) A third law analys is showed satisfactory internal consistency of the Gibbs free energy o f formation and heat capacity data of beta-Rh2O3, but with a much lowe r value for S-298.15,S-beta-Rh2O3 (71.5 +/- 1.5 J.mol(-1).K-1 compared with 106.27 J.mol(-1).K-1; Barin, 1989). This is attributed to the ne w C-p(beta-Rh2O3) data that are significantly different from the origi nal measurements of Wohler and Jochum (1933) and the adjusted values o f Barin (1989). Spinels prepared in the MgO-Rh-O system are solid solu tions between MgRh23+O4 and Mg2Rh4+O4 and the interpretation of the da ta for mu(O2(MgRh2O4)) requires an understanding of phase relationship s in the MgO-Rh-O system. From an isothermal projection of oxygen pote ntials onto the Mg-Rh binary at 1373 K, the mol fraction MgRh23+O4 in spinel (X-MgRh2O4) in equilibrium with MgO and Rh at 1373 K was estima ted to be about 0.92. (i.e., X-Mg2RhO4 approximate to 0.08). This prov ides a calibration point for determining the temperature dependent of a(MgRh2O4)(spinel) in MgRh23+O4-Mg2Rh4+O4 solid solutions. A third-law analysis showed that, once corrected for a(MgRh2O4)(spinel), our data for mu(O2(MgRh2O4)) and C-p(MgRh2O4) are fully consistent. The calcul ated value for S-298.15,S-MgRh2O4 is 105.75 +/- 2 J.mol(-1).K-1. This is in reasonable agreement with the assumption of additive oxide entro pies (S-298.15,S-MgRh2O4 approximate to 98.4 J.mol(-1).K-1), using our new value of S-298.15,S-beta-Rh2O3. We, therefore, conclude that our data for beta-Rh2O3 and MgRh23+O4 are internally consistent. From the third-law analysis it is also possible to determine the activity of Mg 2Rh4+O4, in MgRh23+O4-Mg2Rh4+O4 solid solutions (a(Mg2RhO4)(spinel)) a s a function of temperature. The data for a(Mg2RhO4)(spinel) may be co mbined with the emf measurements for the spinel + MgO + Rh assemblage to evaluate the Gibbs free energy of formation of Mg2Rh4+O4: Delta(f)G (Mg2RhO4,T)degrees (+/-6000 J.mol(-1)) = -100076.3 + 100.0T (1008 < T < 1495 K) We conclude that Mg2Rh4+O4 is an important component in rhod ate spinels at high temperatures, thus extending the stability field o f spinel in the MgO-Rh-O system. Copyright (C) 1997 Elsevier Science L td.