Density measurements on binary mixtures (nitrogen + carbon dioxide and argon + carbon dioxide) at temperatures from (298.15 to 423.15) K with pressures from (11 to 31) MPa using a single-sinker densimeter

https://doi.org/10.1016/j.jct.2015.07.014Get rights and content

Highlights

  • A new single-sinker densimeter was built for temperatures from (273.15 to 423.15) K and pressures up to 35 MPa.

  • The expanded measurement uncertainties (k = 2) are 35 mK for temperature, 3.39 kPa for pressure, and 0.033% for density.

  • Measurements of (N2 + CO2) and (Ar + CO2) at compositions (0.01 and 0.05) mole fraction CO2 were not seen in literatures.

  • These binary systems were measured at temperatures from (298.15 to 423.14) K with pressures from (11 to 31) MPa.

  • Relative deviations of the new experimental data to the GERG-2008 EOS were mostly within 0.5%.

Abstract

A single-sinker densimeter was built to specifically investigate the (p, ρ, T, x) behavior of fluid mixtures relevant for carbon capture and storage (CCS). Due to the use of a magnetic-suspension coupling, the densimeter enables measurements over the temperature range from (273.15 to 423.15) K with pressures up to 35 MPa. A comprehensive analysis of the experimental uncertainties was undertaken. The expanded uncertainties (k = 2) are 35 mK for temperature, 3.39 kPa for pressure, and 0.033% for density determination. The apparatus was used for measurements on the binary systems (nitrogen + carbon dioxide) and (argon + carbon dioxide). The compositions for both systems were (0.05 and 0.01) mole fraction carbon dioxide. Density measurements were carried out at temperatures from (298.15 to 423.15) K with pressures from (11 to 31) MPa. The relative combined expanded uncertainty (k = 2) in density was 0.15% for the (nitrogen + carbon dioxide) mixtures and 0.12% for the (argon + carbon dioxide) mixtures. A major contribution to this uncertainty emerged from the uncertainty in the gas mixture composition. The new experimental data were compared to the GERG-2008 equation of state (EOS) for natural-gas mixtures as implemented in the NIST REFPROP database and to the EOS-CG, another new Helmholtz energy model for CCS mixtures as implemented in the TREND software package of Ruhr-University Bochum. Relative deviations were mostly within 0.5%. The agreement of the new density values with the only available literature data closest to the composition range under study was better than 0.1%.

Introduction

The single-sinker densimeter (SSD) is one of the state-of-the-art instruments to accurately measure the density of fluid substances over large temperature and pressure ranges. The method is based on the Archimedes buoyancy principle and can be used as an absolute technique without the need for calibration fluids. The commonly used reference equations of state (EOS), e.g., for nitrogen [1], argon [2], carbon dioxide [3], ethane [4], ethylene [5], propane [6], and sulfur hexafluoride [7] were all modeled with the help of data measured with SSDs. Moreover, with the increasing demand for knowledge of volumetric properties of fluid mixtures, more measurements [8], [9], [10], [11] were carried out utilizing different SSDs. However, in this context only accurate density data of binary mixtures serve to further improve the performance of multi-parameter EOS for mixtures, especially those based on the GERG EOS (GERG-2004 of Kunz et al. [12] and GERG-2008 of Kunz and Wagner [13]). All reference EOS play important roles in various engineering applications, such as refrigeration engineering [14], oil and gas engineering [15], and in the field of carbon capture and storage (CCS) [16].

The SSD was developed by Brachthäuser et al. [17], [18] in the early 1990s. It simplifies the complex design of the two-sinker densimeter [19] when only measurements of medium and high densities are the goal. An overview of this general type of instrument incorporating a magnetic suspension coupling was published by Wagner and Kleinrahm [20] as well as in chapter 2 of [21] by McLinden. Different versions of the SSD were built by research groups all over the globe [22], [23], [24], [25], [26], [27]. A selection of different SSDs, including measuring range, system uncertainty, and measured fluids, is listed in table 1. Each SSD has its unique features, either of extreme low uncertainty, i.e., a few parts per million in Kuramoto’s group [22], or of a large pressure range, i.e., up to 200 MPa in Hall’s group [9], or for the use at cryogenic temperatures as at Ruhr-University Bochum [26], [27]. Recently a new SSD (type: FluiDENS, Rubotherm, Germany) was installed in our laboratory at Tsinghua University. The apparatus was built to specifically investigate the (p, ρ, T, x) behavior of fluid mixtures relevant for CCS. To enable the investigation of supercritical states, as well as of corrosive substances, the densimeter was designed for measurements over the temperature range from (273.15 to 423.15) K with pressures up to 35 MPa.

CCS is considered to be one of the most important strategies in reducing global carbon emissions [44]. The fluids involved in the pipeline transportation in CCS are mainly carbon dioxide, argon, nitrogen, oxygen, water, sulfur oxide, nitric oxide, and other trace substances [16]. The exact compositions and quantities of the impurities vary with different carbon dioxide sources and different capture and purification processes [16], [44], [45]. The density of carbon dioxide mixtures is a fundamental parameter in the process design for CCS. However, to the best of our knowledge, the density data situation is poor and the available data are mostly of lacking accuracy. In the present project we focused on the binary mixtures (nitrogen + carbon dioxide) and (argon + carbon dioxide). The available experimental data concerning these binaries are summarized in table 2. No data are available for compositions lower than 0.10 mol fraction carbon dioxide for the binary mixtures under study. To fill this data gap, comprehensive density measurements at temperatures from (298.15 to 423.15) K with pressures from (11 to 31) MPa were carried out on the two binary systems both with compositions (0.05 and 0.01) mole fraction carbon dioxide utilizing the new SSD. Although mixtures with a composition of only (0.01 to 0.05) mole fraction carbon dioxide seem to differ just slightly from the pure major component, experimental densities will reveal helpful information in terms of performance testing (and maybe for the improvement) of multi-parameter EOS, as will be discussed in Section 4.3.

Section snippets

Apparatus description

The density measurement system is schematically illustrated in figure 1. It was divided into four sections: gas source, gas pre-treatment, density measurement, and exhaust-gas treatment. The gas source section provided the gas sample under study as well as compressed air to operate the booster pump. In the gas pre-treatment section, the gas was heated by the preheater to a supercritical state. This was required to avoid condensation of gas during compression. Afterwards, the gas was compressed

Uncertainty analysis

This section provides a detailed uncertainty analysis for the new SSD. The combined uncertainty of each measurement parameter (temperature, pressure, density, force transmission error (FTE) of the magnetic suspension coupling, and composition) is calculated according to the error propagation principle suggested by the Guide to the Expression of Uncertainty in Measurement (GUM) [76]. For our considerations we assume that the contributors to uncertainty are independent from each other.

Results from commissioning the new single-sinker densimeter

Within the scope of commissioning the new SSD, measurements on three pure substances were carried out: nitrogen, at T = (298.15, 348.15, and 398.15) K with pressures up to 33 MPa; argon, at T = (298.15, 348.15, and 398.15) K with pressures up to 33 MPa; carbon dioxide, at T = 323.15 K with pressures up to 29 MPa. The relative deviation σr, is defined byσr=100·ρexp-ρEOSρEOS,where ρexp is the experimental density, and ρEOS is the respective value calculated with the state-of-the-art EOS (nitrogen [1], argon

Conclusions

A single-sinker magnetic suspension densimeter was built to specifically investigate the (p, ρ, T, x) behavior of fluid mixtures relevant for CCS. To enable the investigation of supercritical states, also of corrosive substances, the densimeter was designed for measurements over the temperature range from (273.15 to 423.15) K with pressures up to 35 MPa. The expanded uncertainties (k = 2) were 35 mK for temperature, 3.39 kPa for pressure, and 0.033% for density. Within the scope of commissioning the new

Acknowledgements

The research leading to these results has received funding from the European Community’s Seventh Framework Programme (FP7-ENERGY-20121-1-2STAGE) under Grant agreement No. 308809 (The IMPACTS Project). The authors acknowledge the project partners and the following funding partners for their contributions: Statoil Petroleum AS, Lundin Norway AS, Gas Natural Fenosa, MAN Diesel & Turbo SE and Vattenfall AB.

References (89)

  • M.E. Mondéjar et al.

    J. Chem. Thermodyn.

    (2012)
  • R. Hernández-Gómez et al.

    J. Chem. Thermodyn.

    (2014)
  • E. De Visser et al.

    Int. J. Greenhouse Gas Control

    (2008)
  • R. Kleinrahm et al.

    J. Chem. Thermodyn.

    (1986)
  • M.E. Mondéjar et al.

    Measurement

    (2011)
  • P. Patil et al.

    J. Chem. Thermodyn.

    (2007)
  • C.R. Chamorro et al.

    J. Chem. Thermodyn.

    (2006)
  • M.E. Mondéjar et al.

    J. Chem. Thermodyn.

    (2011)
  • P. Claus et al.

    J. Chem. Thermodyn.

    (2003)
  • D. Sommer et al.

    J. Chem. Thermodyn.

    (2011)
  • G. Schilling et al.

    J. Chem. Thermodyn.

    (2008)
  • J. Klimeck et al.

    J. Chem. Thermodyn.

    (1998)
  • J. Klimeck et al.

    J. Chem. Thermodyn.

    (2001)
  • G. Pipitone et al.

    Int. J. Greenhouse Gas Control

    (2009)
  • M. Mantovani et al.

    J. Supercrit. Fluids

    (2012)
  • H.B. Brugge et al.

    Physica A

    (1989)
  • P. McElroy et al.

    J. Chem. Thermodyn.

    (1989)
  • G.J. Esper et al.

    Fluid Phase Equilib.

    (1989)
  • J.F. Ely et al.

    J. Chem. Thermodyn.

    (1989)
  • S. Jiang et al.

    Fluid Phase Equilib.

    (1990)
  • J.C. Seitz et al.

    Geochim. Cosmochim. Acta

    (1994)
  • J.C. Seitz et al.

    J. Chem. Thermodyn.

    (1996)
  • J.C. Seitz et al.

    J. Chem. Thermodyn.

    (1996)
  • M. Richter et al.

    J. Chem. Thermodyn.

    (2014)
  • R. Span et al.

    J. Phys. Chem. Ref. Data

    (2000)
  • C. Tegeler et al.

    J. Phys. Chem. Ref. Data

    (1999)
  • R. Span et al.

    J. Phys. Chem. Ref. Data

    (1996)
  • D. Buecker et al.

    J. Phys. Chem. Ref. Data

    (2006)
  • J. Smukala et al.

    J. Phys. Chem. Ref. Data

    (2000)
  • E.W. Lemmon et al.

    J. Chem. Eng. Data

    (2009)
  • C. Guder et al.

    J. Phys. Chem. Ref. Data

    (2009)
  • M. Atilhan et al.

    J. Chem. Eng. Data

    (2011)
  • M. Atilhan et al.

    J. Chem. Eng. Data

    (2011)
  • O. Kunz, R. Klimeck, W. Wagner, M. Jaeschke, The GERG-2004 wide-range equation of state for natural gases and other...
  • O. Kunz et al.

    J. Chem. Eng. Data

    (2012)
  • E.W. Lemmon et al.

    J. Phys. Chem. Ref. Data

    (2004)
  • M. Atilhan et al.

    J. Chem. Eng. Data

    (2010)
  • W. Wagner et al.

    Int. J. Thermophys.

    (1995)
  • K. Brachthäuser, R. Kleinrahm, H.W. Lösch, W. Wagner, Entwicklung eines neuen Dichtemeßverfahrens und Aufbau einer...
  • W. Wagner et al.

    Metrologia

    (2004)
  • E. Wilhelm et al.

    Volume Properties: Liquids, Solutions and Vapours

    (2014)
  • N. Kuramoto et al.

    Metrologia

    (2004)
  • M. Atilhan, High-accuracy, P–ρ–T Measurements up to 200MPa Between 200K to 500K using a Single Sinker Magnetic...
  • S. Ejaz, High-Accuracy P–ρ–T Measurements of Pure Gas and Natural Gas like Mixtures using a Compact Magnetic Suspension...
  • Cited by (0)

    View full text