Enthalpy of formation of pargasite by high-temperature solution calorimetry and heat capacity of pargasite and fluoropargasite by differential scanning calorimetry

Abstract

Synthetic pargasite with near end-member cation stoichiometry (X F = F/((OH) + F) = 0) and natural, gem-quality fluoropargasite from Sri Lanka (X F = 0.605) have been studied by high-temperature lead-borate drop-solution calorimetry under dynamic measuring conditions in flowing Ar. IR-spectroscopy shows that both samples are highly Mg-Al disordered over M(2) and M(3). For synthetic pargasite, five mutually consistent reaction cycles were constructed in which the critical volatile component H 2 O was treated in fundamentally different ways, namely as the simple “oxide” ( i.e. , as the enthalpy difference [H TCal - H 298.15 ] only), or as the (OH)-component in both low and high water-content phases. The molar enthalpies of drop solution of the participating solid phases pargasite, brucite, corundum, jadeite, diopside, enstatite, calcite, magnesite, quartz, tremolite and talc were measured under identical conditions in the same calorimeter. Our refined value for the enthalpy of formation from the elements at 298.15 K for Mg-Al disordered, end-member pargasite with X F = 0 lies between - 12651.4 ± 12.7 and - 12659.4 ± 14.0 kJ/mol, depending on the internally consistent thermodynamic data set used for the reaction components. This value is somewhat less negative than the thermodynamically derived - 12678 ± 19 kJ/mol suggested by Westrich & Holloway (1981), but differs considerably from the value of - 12719.6 ± 5.3 kJ/mol currently used in the data set of Holland & Powell (1998), or the value of - 12623.396 kJ/mol quoted by Helgeson et al. (1978). The enthalpy of solution for natural fluoropargasite is almost identical to that of synthetic pargasite when cationic deviations from end-member composition are accounted for. This result corroborates the enthalpy of formation obtained for synthetic pargasite and shows that not only H 2 O but also fluorine is expelled from the system without interacting with the lead borate melt. Heat capacity data from both samples have been obtained by differential scanning calorimetry in step-scanning mode for the temperature range 50 - 50 °C (pargasite) and 50 - 420 °C (fluoropargasite). The following best-fit equations are valid for compositions corrected to ideal end-member cation stoichiometry (T in Kelvin, C p in J mol −1 K -1 ): synthetic pargasite , NaCa 2 Mg 4 Al 3 Si 6 O 22 (OH) 2 : C p = 1511.14731 - 0.045825 T - 2.00898 10 6 T −2 - 13389.1321 T −0.5 , natural fluoropargasite , NaCa 2 Mg 4 Al 3 Si 6 O 22 (OH) 0.79 F 1.21 : C p = 1585.2892 - 0.10582 T - 1.30629 10 6 T −2 - 14647.5800 T 0.5 . The polynomial for corrected natural fluoropargasite was then used to model the heat capacities of the OH and F end-members of the pargasite-fluoropargasite series employing literature data for brucite, Mg(OH) 2 , and sellaite, MgF 2 , and assuming ideal (OH, F) mixing. The pargasite estimate is within 0.8% of the above polynomial for measured synthetic pargasite. The modelled polynomials differ by about 0.3% (pargasite) and 1% (fluoropargasite) from the respective end-member entries in the internally consistent data set of Holland & Powell (1998). The enthalpy difference [H 985 K - H 298.15 K ] of 625.59 kJ/mol obtained from the modelled polynomial for end-member fluoropargasite compares well (0.6%) with a value of 621.55 ± 0.85 kJ/mol obtained experimentally by Westrich & Navrotsky (1981) for synthetic end-member fluoropargasite by transposed temperature drop calorimetry.