Quantitative HPLC Analysis and Thermodynamics of Sulfur Melts

Abstract

poldahafen 2, by quoting the depository number CSD 5 1 119, the names of the authors, and the journal citation. 161 Magnetic measurements: The molar susceptibilities of 1 and 2 were measured on powdered samples in the temperature region 93-293 K Because of its unique thermal behavior, liquid sulfur has been the subject of numerous physicochemical and theoretical investigations. As a rule, it is assumed that, below the polymerization temperature (159"C), the melt essentially consists of S, molecules. However, we have already previously shown by means of vibrational spectra"] that rings other than S, (mainly S,, S6, and S, with n > 8) are present at all temperatures in liquid sulfur, and that these-in comparison with S8-are considerably more reactive species and strongly influence the properties of the melt[21.The separation of sulfur homocycles by means of highpressure liquid chromatography (HPLC)[31 now enables a much more accurate quantitative analysis. Using this technique, we have now found that, after establishment of the equilibrium at constant temperature (23 samples, 13 temperatures) and then quenched in liquid nitrogen, accurately weighed at 20 "C (4.1-7.2g), and immediately extracted quantitatively with ca. 200 mL CS2[41. After filtration and dilution to a prescribed volume, the solution was analyzed quantitatively by HPLC. The insoluble residue (polymeric sulfur, S,) was washed with CS,, dried, and weighed.In the chroma tog ram^'^^, the areas A of 18 S, ring-molecules (n = 6-23) could be measured by electronic integration. For the calibration, at least ten solutions of different concentration c of each of the pure compounds were measured in the case of S6, S,, s8, Sl0, S12, and SZ0. The calibration curve was linear (c = aA) with a correlation coefficient better than 0.9999 over the range 5-500 bg S/mL CS2. Since the slope a is proportional to the number of S atoms in the molecule (except in the case of S7), the calibrations could be calculated for the remaining molecules.Representative analytical values at 13 different temperatures are listed in Table 1. As expected[61, the concentrations of the S, rings ( n f 8 ) pass through a maximum at 159°C: besides 3% S,, only 83.4% S, is present. The missing "rest" of the 100% assay is ascribed to soluble s,, molecules (n > 23) and analytical errors (maximally 3"/0). Molecules with n < 6 were not detectable.After S8 and S,, the most frequently occurring molecules in the melt are S6, S9, and Si2. The relevant reaction enthalpies AH(1) of Equation (1) can be calculated from the temperature-dependence of the concentrations of the molecules listed in Table 1. From the linear regressions in the temperature range 116-159 "C the values shown in Table 2 are obtained.Assuming an equally strong intermolecular interaction (per atom) for all molecules in the melt (zero enthalpy of mixing), the mean bond energies (m.B.E.) in the sulfur rings shown in Figure 1 can be calculated from the AH( I)

Geochemical implications of subaqueous molten sulfur at Yugama crater lake, Kusatsu-Shirane volcano, Japan.

“…3 never shifts from the homocyclic sulfur line toward the sulfane line, suggesting that this peak includes an unstable sulfur species other than sulfanes. Calculated stabilities of cyclic sul fur suggest that cyclic S12 is more stable than cy clic S10 (Steudel et al, 1985). In harmony with this calculation, chromatograms of quenched pure molten sulfur show a higher S12 peak than S10.…”

Comment on “Probing the sulfur polymerization transition in situ with Raman spectroscopy” [J. Chem. Phys. 118, 8460 (2003)]

“…50 atoms ͑which are still soluble in carbondisulfide͒ amount to 5% at the triple point of sulfur ͑115°C͒ and are responsible for the well-known melting point depression of 5 K. 1,9 At 200°C the concentration of the soluble non-S 8 rings is 11% and at 300°C it amounts to 9%. 1,7 Expectedly, the Raman spectra of some of these species are very similar to those of S 8 or of polymeric sulfur, in particular in the region of the SS stretching vibrations. 10 In fact, a mixture of soluble sulfur rings larger than S 8 , prepared from liquid sulfur and termed as S x in the literature, shows just one strong and broad Raman line at 460 cm Ϫ1 and a weaker line at 416 cm Ϫ1 in the region 400-500 cm Ϫ1 .…”

“…The authors claimed that the in situ Raman method applied yields more accurate results on the polymer content of liquid sulfur at various temperatures than quenching-dissolution methods. Previously, many authors had determined the polymer content of sulfur melts by quenching the melt either on metal sheets at room temperature, 3,4 in liquid air 5 or in liquid nitrogen 6,7 followed by extraction of the soluble low-molecular weight constituents with carbondisulfide at 20°C and weighing the insoluble residue after drying in a vacuum. By the latter techniques considerably lower polymer contents were obtained ͑e.g., 0.2% at 120°C, 29% at 200°C, and 45%-55% at 300°C 7 ͒, especially if highly purified sulfur was used, if highly efficient quenching techniques were applied ͑very thin stream of liquid sulfur poured into liquid nitrogen͒, and if the quenched powderlike melt was extracted immediately without storage at ambient temperature, as descibed in Refs.…”

Response to “Comment on ‘Probing in situ the sulfur polymerization transition with Raman spectroscopy’ ” [J. Chem. Phys. 121, 6573 (2004)]

“…The much slower quenching rate is less efficient than that employed by KK and this fact inevitably led to the lower polymer content found in Ref. 10. ͑c͒ Investigation of sulfur's structure with neutron diffraction experiments 12 before and immediately after quenching from 240°C has shown that the diffraction pattern is different between the two states, i.e., the melt and the quenched product.…”

“…The lower the quenched mass of sulfur the faster is the quenching rate. In the most commonly accepted work using the QD method, 7͑a͒ Koh and Klement ͑KK͒ quenched a quantity of liquid sulfur of about 0.1 g. On the other hand, Steudel et al 10 used much bigger quantities, i.e., 4.1-7.2 g thus corresponding to very slow cooling rates. The much slower quenching rate is less efficient than that employed by KK and this fact inevitably led to the lower polymer content found in Ref.…”