Improving a variation of the DSC technique for measuring the boiling points of pure compounds at low pressures

https://doi.org/10.1016/j.jct.2016.04.023Get rights and content

Highlights

  • Improvement of a variation of the DSC technique for boiling points at low pressures.

  • Use of a ballpoint pen ball over the pinhole of the DSC crucible.

  • Effects of configuration variables of the DSC technique accounted by factorial design.

  • An optimized region was obtained and tested for selected compounds.

Abstract

This study aims to improve a variation of the differential scanning calorimetry (DSC) technique for measuring boiling points of pure compounds at low pressures. Using a well-known n-paraffin (n-hexadecane), experimental boiling points at a pressure of 3.47 kPa with u(P) = 0.07 kPa were obtained by using a variation of the DSC technique, which consists of placing samples inside hermetically sealed aluminum crucibles, with a pinhole (diameter of 0.8 mm) made on the lid and a tungsten carbide ball with a diameter of 1.0 mm over it. Experiments were configured at nine different combinations of heating rates (K·min−1) and sample sizes (mg) following a full factorial design (22 trials plus a star configuration and three central points). Individual and combined effects of these two independent variables on the difference between experimental and estimated boiling points (NIST Thermo Data Engine v. 5.0 – Aspen Plus v. 8.4) were investigated. The results obtained in this work reveal that although both factors affect individually the accuracy of this variation of the DSC technique, the effect of heating rate is the most important. An optimized region of combinations of heating rate and sample size for determining boiling points of pure compounds at low pressures was obtained using the response-surface methodology (RSM). Within this optimized region, a selected condition, combining a heating rate of 24.52 K·min−1 and a sample size of (4.6 ± 0.5) mg, was tested for six different compounds (92.094–302.37 g mol−1) comprising four fatty compounds (tributyrin, monocaprylin, octanoic acid and 1-octadecanol), glycerol and n-octadecane, besides n-hexadecane. This condition was also successfully applied for obtaining boiling points of n-hexadecane at pressures up to 18.66 kPa with u(P) = 0.18 kPa. The optimized region obtained in this work, in terms of heating rates and sample sizes, is specific for the crucible configuration tested.

Introduction

Vapor pressure is one of the most important thermophysical properties for characterization of substances and is essential for modeling and simulation of separation processes, such as distillation and stripping. Among other methods, DSC technique has been studied for determining vapor pressures of a variety of compounds [1], [2], [3], [4], [5], [6], [7], as an alternative method with remarkable advantages for costly, low-volatile and/or heat-sensitive compounds in terms of sample size (mg) and/or operation time [8], respectively. Recently, Siitsman and Oja [7] summarized seventeen studies published in the literature since 1972 that applied the DSC technique for measuring vapor–liquid equilibria data and/or vapor pressure data of different classes of compounds. Two other studies, not listed by Siitsman and Oja [7], involving analysis of fatty compounds, were lately released [1], [8]. As emphasized by Siitsman and Oja [7], the standardized method has rarely been fully applied, and these previous works, besides providing valuable experimental data, were also concerned with the development of the technique itself, namely, the best combination of operational conditions that delivered reliable results for a chosen application. In fact, a number of factors related to the accuracy of the DSC technique for measuring vapor pressures of pure compounds are identified in the literature [3], [5], [6], [7], [9], [10], [11], i.e., sample size, heating rate, sample purity, pressure range, crucible configuration and pinhole size. The standard test method for determining vapor pressure of pure compounds by thermal analysis at 0.2–2000 kPa of the American Society for Testing and Materials (ASTM E1782-14) [12] suggests a sample size of (1–5) mg, a heating rate of 5 K·min−1 and a pinhole size up to 0.350 mm. Brozena [6] successfully measured vapor pressures of 1-octanol using pinhole sizes up to 0.350 mm while keeping the heating rate at 5 K·min−1 at 0.197 kPa up to atmospheric pressure. Despite these recommendations, Matricarde Falleiro et al. [3] and Damaceno et al. [1] determined boiling points of fatty compounds and tocopherols by applying a combination of a higher heating rate (25 K·min−1) and the use of a small tungsten carbide sphere (1 mm of diameter) acting as a relief valve placed over a pinhole larger than the recommended value (0.8 mm of diameter). The first use of a ballpoint pen ball over the pinhole in a DSC crucible was described by Farritor and Tao [13]. Heating rates higher than the recommended value of 5 K·min−1 were also tested by Contreras et al. [10]. The authors investigated the variation of boiling points of isopropyl palmitate (298.5 g·mol−1) given by the DSC technique at pressures below 5 kPa configured with three different heating rates (5, 10 and 15 K·min−1) while fixing sample size (5.0 ± 0.5) mg and pinhole size (0.4 mm of diameter). Siitsman et al. [5] found out that sample size has a great influence on the accuracy of the DSC technique, especially around 15 kPa and over 250 kPa, while working with tobacco related materials with fixed pinhole size (0.5 mm) and heating rate (5 K·min−1).

In this context, this work presents a systematic study of the influence of two selected factors, sample size (mg) and heating rate (K·min−1), in the accuracy of a variation of the DSC technique for measuring boiling points of pure compounds at low pressures. n-Hexadecane, an n-paraffin of sixteen carbons with a molar mass of 226.45 g·mol−1 was selected for this investigation. Experiments were configured following a full factorial design (22 trials plus a star configuration and three central points). Individual and combined effects of these two independent variables on the difference between experimental and estimated boiling points for the pure compound (NIST Thermo Data Engine v. 5.0 – Aspen Plus v. 8.4) were investigated while fixing other two measurement conditions: (1) samples were placed inside hermetically sealed aluminum crucibles, with a fixed pinhole diameter (0.8 mm) on the lid and a tungsten carbide ball with a diameter of 1.0 mm over it, following our previous works [1], [3], [8], and (2) pressure set at 3.47 kPa with u(P) = 0.07 kPa. Surfaces and contour plots were sketched using the quadratic models for the statistically significant variables according to response surface methodology. The ranges of the independent variables investigated were (3.5–5.0) mg for sample size and (5–30) K·min−1 for heating rates. For further analysis, six different compounds (92.094–302.37 g·mol−1) comprising four fatty compounds (tributyrin, monocaprylin, octanoic acid and 1-octadecanol), glycerol and n-octadecane were selected for testing the reliability of the optimized conditions. Complementarily, the selected condition within the optimized region was tested for n-hexadecane at four different pressures: 10.37 kPa with u(P) = 0.19 kPa, 15.17 kPa with u(P) = 0.19 kPa, 18.66 kPa with u(P) = 0.18 kPa, and 22.39 kPa with u(P) = 0.23 kPa.

Section snippets

Material

Table 1 lists the reagents used in this work together with CAS Registry numbers, IUPAC names, purities in mass fraction, molar mass (MW), normal boiling points, and suppliers. All chemicals were used without any further purification steps.

Hermetic aluminum crucibles (pans + lids) for DSC (max 10 μL) were purchased from TA Instruments. A pinhole of a diameter of 0.8 mm was made on each lid using a system consisting of a fixation assembly, mandrel and drills. From the disassemble of a ballpoint pen,

Results and discussion

The experimental results shown in Table 2 reveal that the accuracy of the DSC technique (DIF/K) is a function of the two variables studied. It can be observed, in a comparison of trials 5 and 6 (X1 varying from −α to +α), and trials 7 and 8 (X2 varying from −α to +α), that the difference between the experimental and estimated boiling points for n-hexadecane at 3.47 kPa with u(P) = 0.07 kPa decreased with an increase in heating rate or an increase in sample size, respectively. Combined effects were

Conclusions

In this study, different configurations of a variation of the DSC technique in terms of heating rates and sample sizes were tested for determining boiling points of pure compounds at low pressures. An optimized region for the combination of these two variables was achieved at 3.47 kPa with u(P) = 0.07 kPa, considering a sealed aluminum crucible, with a pinhole (diameter of 0.8 mm) made on the lid and a tungsten carbide ball with a diameter of 1.0 mm over it, and proved to be valid for other selected

Acknowledgements

R. Ceriani acknowledges FAPESP (2013/12735-5) for the financial support and CNPq (304303/2013-5) for her individual grant. K.L. Troni and D.S. Damaceno acknowledge respectively, CAPES and CNPq for their individual grants.

References (30)

  • R.M. Matricarde Falleiro et al.

    J. Chem. Thermodyn.

    (2010)
  • R.M. Matricarde Falleiro et al.

    Thermochim. Acta

    (2012)
  • L.Y. Akisawa Silva et al.

    Thermochim. Acta

    (2011)
  • C. Siitsman et al.

    Thermochim. Acta

    (2014)
  • A. Brozena

    Thermochim. Acta

    (2013)
  • C. Siitsman et al.

    Thermochim. Acta

    (2015)
  • L.P. Cunico et al.

    J. Chem. Thermodyn.

    (2015)
  • R.J. Seyler

    Thermochim. Acta

    (1976)
  • M.D. Contreras et al.

    Thermochim. Acta

    (1993)
  • R.R. Farritor et al.

    Thermochim. Acta

    (1970)
  • J. Marrero et al.

    Fluid Phase Equilib.

    (2001)
  • D. Ambrose et al.

    J. Chem. Thermodyn.

    (1987)
  • D.L. Morgan et al.

    Fluid Phase Equilib.

    (1994)
  • F. Zuiderweg

    J. Chem. Eng. Sci.

    (1952)
  • D. Ambrose et al.

    J. Chem. Thermodyn.

    (1974)
  • Cited by (13)

    • Vapour pressure and VLE data of fatty compounds

      2021, Journal of Chemical Thermodynamics
      Citation Excerpt :

      The DSC thermal analysis consists in monitoring the difference between the heat flow to a crucible containing the substance of interest and a reference crucible, while both crucibles are subjected to temperature control [4]. The use of the DSC technique has been studied and improved in our research group for measuring vapour pressure data and VLE of fatty mixtures, especially in the case of saturated fatty compounds [5,6,7,8,9,10]. The technique developed in this work was used for measuring data of unsaturated fatty compounds, for which experimental data are very scarce or not available in the literature.

    • Experimental data and prediction of normal boiling points of partial acylglycerols

      2018, Fuel
      Citation Excerpt :

      Monophasic events were also observed for monobutyrin (Fig. S6), dicaprylin (Fig. S7), dinonanoin (Fig. S8) and dicaprin (Fig. S9). On the other hand, monopalmitin (Fig. S10) and monoestearin (Fig. S11) showed a perturbation in the baseline of the DTG, which is indication of the migration of the acyl group of the monoacylglycerol molecule [4,5,14,20,25,26]. Figs. S12–S15 of monocaprin (two lots) and monolaurin (two lots) showed in the DTG curves the presence of more than one peak (a deep perturbation in the baseline followed by a peak).

    View all citing articles on Scopus
    View full text