Elsevier

Fluid Phase Equilibria

Volume 404, 25 October 2015, Pages 124-130
Fluid Phase Equilibria

Physical data for a process to separate krypton from air by selective absorption in an ionic liquid

https://doi.org/10.1016/j.fluid.2015.06.037Get rights and content

Abstract

Ionic liquids provide a possible absorption process to extract krypton from air. The feed for such a process is an oxygen stream from a liquid-air plant. An effective ionic liquid is [P(14)666][TMPP]; in that solvent, the solubilities of some pertinent common gases are appreciably larger than those in conventional ionic liquids, and the selectivity Kr/O2 is close to 3. A nonvolatile ionic liquid is preferred over a hydrocarbon solvent because of safety and simpler solvent recovery. Because, the viscosity of [P(14)666][TMPP] is very high, 20 wt.% [BHMIM][AC] is added to reduce the viscosity by one order of magnitude without significantly reducing solvent capacity and selectivity. This work provides extensive fundamental data (solubility, density and viscosity) required for process design.

Introduction

Double (or triple)-pane windows provide an effective way for reducing heat loss in the winter or heat gain in the summer in buildings; such windows commonly use argon for the narrow space between the glass panes [1]. Krypton is a better choice because, like argon, it is safe and chemically inert but, compared to argon, it has a lower thermal conductivity. In addition to multi-pane windows, krypton is used in lighting and electronic devices [2], [3].

However, krypton is more expensive than argon. At present, krypton from air is produced by cryogenic distillation of liquefied oxygen [4]. Because the concentration of krypton in air is very small, and because krypton is less volatile than oxygen, this distillation requires much energy [4]. The present work explores a possible alternate method to produce krypton: an absorption process at ambient temperature. The feed to the absorption column is an oxygen stream from a liquid-air plant. The solvent should have a high capacity for krypton and oxygen, and a high Kr/O2 selectivity.

A possible solvent is a heavy hydrocarbon, but hydrocarbons are volatile and not safe (explosion with oxygen). Ionic liquids are attractive because of their negligible vapor pressure, nonflammability and chemical stability [5], [6]. Because the polarizability of Kr (16.8 a03) [7] is larger than that of O2 (10.6 a03) [8], selectivity Kr/O2 is likely to be good. Experimental data from our laboratory show that the solubilities of small hydrocarbons in ionic liquid trihexyl tetradecylphosphonium bis(2,4,4-trimethylpentyl)phosphinate ([P(14)666][TMPP]) [9], are appreciably larger than those in other ionic liquids. Therefore, [P(14)666][TMPP] may be a suitable solvent for an absorption process producing krypton.

However, the viscosity of [P(14)666][TMPP] is too high for a separation process (1004 cP at 298 K [10]). We use 1-butyl-3-H-imidazolium acetate ([BHMIM][AC]) as a diluent to reduce the viscosity of [P(14)666][TMPP]. We choose [BHMIM][AC] because it has low viscosity (7 mPa s at 298 K) [11] and exhibits good solubilities for krypton and oxygen [12]. A simpler alternative is water. However, water is volatile and exhibits very low solubilities for krypton and oxygen.

In this work, we report densities, viscosities and solubilities of krypton, oxygen, nitrogen, xenon and argon in [P(14)666][TMPP] from 313 to 353 K up to 6 MPa. We studied the effect of diluents on solubilities and selectivity for krypton and oxygen in [P(14)666][TMPP]. Because solubiliites of gases in [P(14)666][TMPP] are appreciably larger than those in [BHMIM][AC], we prefer not to use [BHMIM][AC] alone.

Section snippets

Experimental

Krypton and oxygen were purchased from Praxair with purities ≥99.9%. Ionic liquids [P(14)666][TMPP] and [BHMIM][AC] were obtained from Ionic Liquid Technologies (Io-Li-Tec); purities are higher than 95 and 98%, respectively. To remove volatile impurities, the ionic liquids were dried for 24 h in a vacuum oven at 373 K prior to use. Water contents of ionic liquids were measured using the Karl-Fischer method (Aquastar C2000 Titrator); they are less than 0.6 wt.%. The contents of chloride in

Results

Table S1 and Fig. S2 show the Px data for krypton, oxygen, argon, xenon and nitrogen in [P(14)666][TMPP] from 313 to 353 K up to 6 MPa. The uncertainty in mole fraction is less than ±5%. Fig. 1 shows the Px data for krypton, oxygen, argon, xenon and nitrogen in [P(14)666][TMPP] at 313 K. From these data, we obtain Henry’s constants defined byH=limx0fxwhere Henry’s constant H is in MPa, f is fugacity and x is mole fraction of solute.

Table 1 gives Henry’s constants for five gases in

Preliminary process flow diagram

Fig. 12, Fig. 13 show two possible process flow diagrams for the separation of Kr from a side stream of a liquid-air distillation column containing 2–5 × 103 ppm Kr (private communication from Praxair, Inc.).

Fig. 12 shows a conventional absorption/stripping process. The feed gas is introduced to a packed-bed column absorber. The solvent enters at the top of the column while the feed is introduced at the bottom. Following counter-current contact, the solvent becomes saturated with gas. The ‘rich’

Conclusions

This work reports extensive experimental physico-chemical data required to design a potential new separation process for obtaining krypton from the oxygen stream of an air-liquefaction plant. Experimental results suggest that diluted trihexyl tetradecylphosphonium bis(2,4,4-trimethylpentyl)phosphinate [P(14)666][TMPP] is a suitable solvent for an absorption process. It has high capacity for dissolved gases as desired to permit a small solvent flow. Also, the diluted ionic liquid exhibits good

Acknowledgments

This work was supported by the Assistant Secretary for Energy Efficiency and Renewable Energy, Office of Building Technology, Building Technologies Program of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. The authors are grateful to Robert Hart, Dragan Curcija, Alexis Bell, Scott Lynn and co-workers for valuable advice, and to Maogang He for assistance in preparing the manuscript.

References (31)

  • P.L. Kebabian et al.

    Determination of argon-filled insulated glass window seal failure by spectroscopic detection of oxygen

    Meas. Sci. Technol.

    (2003)
  • R. Betzendahl

    The rare gases market recovers with global economies

    CryoGas Int.

    (2011)
  • F.G. Kerry

    Industrial Gas Handbook: Gas Separation and Purification

    (2010)
  • F. Jutz et al.

    Ionic liquids and dense carbon dioxide: a beneficial biphasic system for catalysis

    Chem. Rev.

    (2011)
  • H. Tokuda et al.

    Physicochemical properties and structures of room-temperature ionic liquids. 3. Variation of cationic structures

    J. Phys. Chem. B

    (2006)
  • Cited by (0)

    View full text