59th Annual Report on Research 2014

Figure 1: Synthesis and structure of IL monomer and poly(IL) used in this work.

Investigation of the swelling behavior of the poly(IL) focused on variety of organic solvents with a wide range of solubility parameters, with water included as a control.  As expected, solvents that were fully immiscible with the IL monomer (e.g. hexanes, water) were incapable of swelling the poly(IL).   Yet, we also observed that some solvents which were fully miscible with the IL monomer such as dichloromethane (CH₂Cl₂), tetrahydrofuran (THF), and ethanol (EtOH) had virtually no ability to swell the poly(IL).  The largest swelling capacity initially observed was in dimethyl sulfoxide (DMSO) where the swelling capacity (Q) was determined to be 202.4.  Figure 2 provides an example of the difference in size between a “dry” poly IL sample and the fully swollen sample in DMSO.

Figure 2:   Digital photographs displaying the 10 mm diameter dry poly(IL) sample (left) and the material fully swollen, ~89 mm diameter poly(IL) in DMSO (right). 

Analysis of solvent solubility parameters (Figure 3a) appears to reveal that there is “sweet spot” with regard poly(IL) swelling behavior.  Based on the large degree of swelling (Q = 202.4) in dimethyl sulfoxide (DMSO) (d = 26.7 MPa^1/2), it can be inferred that DMSO is the most well-matched solvent in terms of solubility parameter to the poly(IL).  Interestingly, this value is nearly identical to that calculated by Bara, et al. for a very similar (non-polymerizable) IL, 1-benzyl-3-methylimidazolium bistriflimide (d = 27.2 MPa^1/2).   Yet, not all solvents with solubility parameters near 27 MPa^1/2 resulted in a high swollen poly(IL) (i.e., Q > 40).  For example, ethanol (d = 26.5) had a value of Q = 0.11, indicative of nearly zero swelling.  Contrastingly, exposure of the poly(IL) to methyl ethyl ketone (MEK) (d = 19.1) still resulted in a greater level of swelling than would be expected (Q = 41.5).  These distinctly different behaviors indicate that for a poly(IL) system, the solubility parameter value alone cannot be taken as a predictor of swelling and a more relationship between other solvent properties and swelling must exist.

Based on observations reported in other works, we then examined correlations between poly(IL) swelling and solvent dielectric constant (Figure 3b).  Here, the organic solvent with the greatest Q value, DMSO, also has the largest dielectric constant (e = 47.2).  Figure 3b illustrates a distinct tendency of the poly(IL) to swell more in organic solvents with increasing dielectric constant, although the near zero swelling in EtOH (e = 24.3) compared to that in MEK (e = 18.5) is anomalous.  As also can be seen in Figure 3b, there is a poor correlation between Q and e when the solvent with the largest dielectric constant, water (e = 80.4) is considered.  Experimental results, as visualized in Figures 2a,b, thus suggest that solvents with high dielectric constants that also have solubility parameters near that of the poly(IL) will yield greater swelling.

Still, we endeavored to find a more intuitive correlation by which to rationalize the swelling behavior of the poly(IL).  Q values for the solvents were then plotted against solvent dipole moment (m) (Figure 3c), resulting in a much more obvious trend between swelling and an intrinsic solvent property.  Solvents with dipole moments < 2 D are observed to have no ability to swell the poly(IL), while those with dipole moments > 2.5 D result in highly swollen materials.  Figure 3c also illustrates a nearly exponential increase in swelling as dipole moment increases, with the large degree of swelling in DMSO again correlated with the largest dipole moment.   This result led us to repeat our experiments with an additional solvent with a larger dipole moment.  We selected propylene carbonate (PC) (m = 4.94 D), which also has a larger dielectric constant (e = 64.9) than DMSO.  The result of this experiment was consistent with the trend, with the largest swelling observed for any solvent (Q = 389).
                These results indicate that poly(IL) materials are capable of expanding more than two orders of magnitude their original size likely due to charge screening, as evident from dielectric constant and dipole moment.  This reveals that ionic interactions can be disrupted and the material will expand.  However, in the case of CO₂, which has no dipole moment, such swelling behavior would not be expected even at high pressures.  This behavior may also indicate that in order to increase CO₂ affinities for ILs and poly(ILs) the cation-anion interactions will need to be disrupted by a secondary force. 

Figure 3a: Relationship between poly(IL) swelling and solvent solubility parameter (d).

Figure 3b: Relationship between poly(IL) swelling and solvent dielectric constant (e).

Figure 3c: Relationship between poly(IL) swelling and solvent dipole moment (m).

                The ACS-PRF DNI program has benefitted me greatly in my career thus far.  I have been able to explore some very important aspects of ionic liquids from both experimental and computational/modeling approaches.  My group has been able to publish several key papers that will serve as a foundation for future work for years to come.  The award has also enabled me to attend two international conferences related to ionic liquids, COIL-5 (Portugal, 2013) and IL-SEPT2 (Canada, 2014).  At these meetings, I was able to interact with many well-known researchers from around the world and foster collaborations.  The DNI award has helped support a Ph.D. student, W. Jeffrey Horne, who is on track to graduate summer 2015.   Furthermore, this award has enabled me to work with a number of very talented undergraduates, several of whom are named as co-authors on publications that have resulted from this funding.