According to the results of chemical analysis, compositions SiaOcCd and SiaBbOcCd of samples SiOC, SiBOC1, SiBOC2, and SiBOC3 are located within the three-phase region SiO2–SiC–C and four-phase region SiO2–SiC–C–B2O3 of the phase diagram of the Si–B–O–C system. Thus, the aim of the calorimetric measurements in this work was to obtain the enthalpy of formation of amorphous Si–(B–)O–C PDCs with respect to a mixture of the most stable crystalline components: SiO2, SiC, and C (graphite) for sample SiOC; and SiO2, SiC, C, and B2O3 for samples SiBOC1–3. This enthalpy, ΔHf,comp of Si–(B–)O–C, is associated with the following reaction:
When a pellet is dissolved in the solvent during the calorimetric measurement, the oxidative dissolution reaction, represented by Eq. (2), has a corresponding enthalpy, which we call ΔHds of Si–(B–)O–C:
In this equation, the final state of the sample is SiO2 (cristobalite) and dissolved B2O3 (for boron-containing ceramics) and CO2 gas is evolved as a result of oxidation. The rationale and evidence for the final state of SiO2 and completeness of oxidation of carbon are explained elsewhere. The measured values of ΔHds of samples SiOC–SiBOC3 are given in Table 4. These values combined with other known thermodynamic quantities were used to calculate the oxidation enthalpy at 25°C, ΔHox, the formation enthalpy from the corresponding elements, ΔHf,elem, and the enthalpy of formation from the components, ΔHf,comp, through the thermodynamic cycles given in Table 5. The enthalpy values obtained for the Si–(B–)O–C glasses are given in three right-hand columns of Table 4. In Fig. 4, the values of ΔHf,comp are plotted versus boron content. For both pyrolysis temperatures, the enthalpy of formation as a function of the boron content changes approximately linearly with a positive slope, indicating that the glasses become enthalpically less stable with increasing boron content. For samples pyrolyzed at 1000°C, while the value of ΔHf,comp for sample SiOC is exothermic [−9.9 ± 2.0 kJ (g·at.)−1], the corresponding values for the B-containing samples are endothermic [4.9 ± 2.2, 11.6 ± 2.5, and 23.8 ± 2.4 kJ (g·at.)−1], which means that the B-doped glasses pyrolyzed at 1000°C are energetically less stable than the competitive crystalline components. For identical compositions, the glasses pyrolyzed at 1200°C have enthalpies of formation more exothermic than those pyrolyzed at 1000°C, indicating that the higher the pyrolysis temperature, the more stable the Si–(B–)O–C glasses. Using the enthalpy values given in Table 4, the exothermic enthalpy changes from 1000°C to 1200°C for the samples containing 0.0, 3.1, and 5.7 at.% were obtained to be −11.6 ± 3.1, −17.7 ± 3.2, and −14.1 ± 3.6 kJ (g·at.)−1, respectively. As seen, the enthalpy changes are almost the same within errors and do not vary systematically with the boron content. Calorimetry was not performed on sample SiBOC3 pyrolyzed at 1200°C because the sample was partially crystalline.
Figure 4. Enthalpies of formation of samples SiOC, SiBOC1, SiBOC2, and SiBOC3 with respect to crystalline SiO2 (cristobalite), SiC, graphite, and crystalline B2O3, ΔHf, comp, plotted as a function of the boron content.
Download figure to PowerPoint
As expected, the values of ΔHf,elem are all significantly exothermic (see Table 4) because the enthalpies of formation of components SiO2, SiC, and B2O3 from elements are very exothermic, as seen in Table 5. The change in ΔHf,elem as a function of the boron content follows the same trend as explained above in the case of ΔHf,comp.