(502d) Using Carbon Dioxide as an Antisolvent Separation Aid with Ionic Liquids: Thermodynamics and Solvent Strength

Ionic Liquids (ILs) have potential for applications in many different areas, including as reaction and separation solvents.  One of the many advantages of ILs is that they are tunable, the anions and cations can be chosen for specific applications.  However, a limitation for their use is being able to separate compounds from ILs.  Many products, impurities, and, especially, spent catalyst, are difficult to separate from ILs.  Distillation is not an option for low volatility or thermally labile compounds.  Supercritical CO₂ extraction has shown potential, but is limited to compounds that dissolve in supercritical CO₂.  Another method to separate compounds utilizes a gas as an antisolvent?essentially a dissolved gas lowers the solvation capability of the liquid and creates supersaturation, which facilitates separation.  In fact, this method is commonly used to separate and recrystallize many compounds from organic solvents.  This method can also be used for ILs, for both liquid[1] and solid[2] solutes, although sometimes the presence of an organic cosolvent is necessary.  When CO₂ is dissolved in an organic, the volume expands greatly.  This lowers the solvent strength, and thus changes the saturation concentration of a dissolved solute, inducing a separation.  However, ILs do not expand greatly under CO₂ pressure.  Therefore, a way to determine the changes in ILs and IL/organic mixtures upon addition of CO₂ would be useful in designing gas antisolvent separation systems for ILs.  One possible method is to determine the solvent strength using solvatochromic parameters.

The work encompassed here includes a systematic investigation of the antisolvent ability of CO₂ for ILs and IL/organic cosolvent mixtures and coordinated solvent strength studies.  The ILs and organics were chosen to investigate the specific effects of polarity and hydrogen bond donating ability of the cation, anion, and cosolvent.  The main IL in the study in the IUPAC standard, [hmim][Tf₂N].  The organic cosolvents acetonitrile, 2-butanone, and 2,2,2-trifluoroethanol encompass different degrees of polarity and hydrogen bonding ability.  The solid solute is a copper probe that not only acts as a model catalyst, but also as a spectroscopic probe.  Thus, it is possible to observe the solvent strength during the nucleation process.

The antisolvent ability of CO₂ for liquid/liquid mixtures is defined in terms of the lower critical end point (LCEP), which is the temperature and pressure at which a phase split occurs.  The LCEP for [hmim][Tf₂N] with the different organics depends on the strength of interactions between the IL and organic (Table 1).  For instance, more CO₂ pressure is required to induce a phase split for [hmim][Tf₂N] with a more polar molecule.  For 10 mole%  IL mixtures, the 2,2,2-trifluoroethanol requires the most CO₂ pressure and the 2-butanone requires the least.

The antisolvent ability of CO₂ for a dissolved solid solute in a liquid is defined as the temperature and pressure at which the first crystal of the solid solute appears?the nucleation pressure.  The nucleation pressures for the pure organic acetonitrile follow published trends?as the concentration of the copper probe increases, the nucleation pressure decreases.  For [hmim][Tf₂N]/acetonitrile mixtures with dissolved copper probe, the lower mole fractions of IL exhibited phase separation, while mixtures at 10 mol% IL did not.  The organic cosolvent is important to facilitate the CO₂ induced nucleation.

Various mixtures were tested with the spectroscopic probes in addition to the copper probe.  These include IL/organic binary mixtures, IL/CO₂ and organic/CO₂ binary mixtures, and IL/organic/CO₂ binary mixtures.  The solvent strength is defined in terms of the Kamlet-Taft parameters: p* for dipolarity/polarizability, a for hydrogen bond donating ability, and b for hydrogen bond accepting ability.  The IL/organic mixtures showed that a large amount of organic, typically at least 50 mol%, is need for a significant effect on solvent strength.  The solvent strength (p*) of the pure organics significantly decreases with increasing CO₂ pressure, while the solvent strength of the pure IL doesn't change significantly (Figure 1).  The mixtures, however, do show a more significant drop.  The organic with the most significant drop in solvent strength with CO₂ pressure is 2-butanone, which can explain why the 10% [hmim][Tf₂N]/2-butanone mixture has the lowest LCEP of the organics measured. Thus, in this work we will show the connection between CO₂-induced phase separation in IL/organic/CO₂ systems and the solvent strength of the mixtures, as measured by spectroscopic probes.

Table 1. The LCEP, XCO2 at LCEP, and K-point for various IL/organic mixtures at 40 °C.

Figure 1. The Kamlet-Taft p* parameter for IL/CO₂, organic/CO₂, and IL/organic/CO₂ mixtures at 40 °C.

1.   S. N. V. K. Aki, A. M. Scurto, and J. F. Brennecke. Ternary Phase Behavior of Ionic Liquid (IL)-Organic-CO₂ Systems. Industrial & Engineering Chemistry Research. 2006;ASAP Article.

2.   E. M. Saurer, S. N. V. K. Aki, and J. F. Brennecke. Removal of Ammonium Bromide, Ammonium Chloride, and Zinc Acetate from Ionic Liquid/Organic Mixtures Using Carbon Dioxide. Green Chemistry. 2006;2:141-143.