Nuclear Spin Relaxation and Molecular Interactions of a Novel

Mar 7, 2013 - Department of Energy, National Energy Technology Laboratory-Regional University Alliance. ‡ Department of ... Nuclear spin relaxation,...
0 downloads 7 Views 1MB Size
Subscriber access provided by UIC Library

Article

Nuclear Spin Relaxation and Molecular Interactions of a Novel Triazolium-Based Ionic Liquid Jesse J Allen, Yanika Schneider, Brian W. Kail, David Richard Luebke, Hunaid B Nulwala, and Krishnan Damodaran J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/jp401188g • Publication Date (Web): 07 Mar 2013 Downloaded from http://pubs.acs.org on March 18, 2013

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry B is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Nuclear Spin Relaxation and Molecular Interactions of a Novel Triazolium-Based Ionic Liquid Jesse J. Allen, †,‡ Yanika Schneider, § Brian W. Kail,║ David R. Luebke, ┴ Hunaid Nulwala, ┴,# and Krishnan Damodaran,†, ‡* † Department of Energy, National Energy Technology Laboratory-Regional University Alliance ‡ Department of Chemistry, University of Pittsburgh, Pittsburgh, PA, 15260 § Lawrence Berkely National Laboratory, Berkeley, CA, 94720 ║ URS Corporation, P.O. Box 618, South Park, PA 15129 ┴ Department of Energy, National Energy Technology Laboratory, United States Department of Energy, P.O. Box 10940, Pittsburgh, PA 15129 # Department of Chemistry, Carnegie Mellon University, Pittsburgh, PA 15213

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 24

Abstract: Nuclear spin relaxation, small angle X-ray scattering (SAXS), and electrospray ionization mass spectrometry (ESI-MS) techniques are used to determine supramolecular arrangement of 3-methyl-1-octyl-4-phenyl-1H-triazol-1,2,3-ium bis(trifluoromethanesulfonyl)imide [OMPhTz][Tf2N], an example of a triazolium-based ionic liquid. The results obtained showed first order thermodynamic dependence for nuclear spin relaxation of the anion. First order relaxation dependence is interpreted as through-bond dipolar relaxation. Greater than first order dependence was found in the aliphatic protons, aromatic carbons (including nearest neighbors), and carbons at the end of the aliphatic tail. Greater than first order thermodynamic dependence of spin relaxation rates is interpreted as relaxation resulting from at least one mechanism additional to through-bond dipolar relaxation. In rigid portions of the cation, an additional spin relaxation mechanism is attributed to anisotropic effects, while greater than first order thermodynamic dependence of the octyl side chain’s spin relaxation rates is attributed to cation-cation interactions. Little interaction between the anion and the cation was observed by spin relaxation studies or by ESI-MS. No extended supramolecular structure was observed in this study, which was further supported by MS and SAXS. Nuclear Overhauser Enhancement (NOE) factors are used in conjunction with spin-lattice relaxation time (T1) measurements to calculate rotational correlation times for C-H bonds (The time it takes for the vector represented by the bond between the two atoms to rotate by one radian). The rotational correlation times are used to represent segmental reorientation dynamics of the cation. A combination of techniques are used to determine segmental interactions and dynamics of this example of a triazolium-based ionic liquid. Keywords: NMR Spectroscopy, Relaxation, Ionic Liquid, Kinetics, Molecular Dynamics

ACS Paragon Plus Environment

2

Page 3 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Introduction Ionic liquids (ILs) have shown potential in various technologically important applications, including: CO2 capture materials1-9, selective membranes10, batteries11,12, and sustainable reaction media13,14. Development of a fundamental understanding of ionic liquid interactions is very important for improving the properties of ILs. Formation of IL supramolecular structures can significantly affect IL properties.14 Ionic liquids have shown aggregation behavior in liquid state which affects both ion mobility and electrical conductivity15-18. Hence, it is very important to develop a better understanding of the causes of supramolecular arrangements. 1,3-disubstituted imidazolium-based ionic liquids have been an intense focus of research due to their ease of synthesis and high level of diversity.11 Imidazolium-based ionic liquids have also shown increasing levels of aggregation with increasing aliphatic group chain length (>3 carbons).15,19-22 Triazolium-based ILs; however, have several advantages over the more commonly studied ILs, such as imidazolium-based ILs. Triazolium-based ILs are available through a relatively simple modular synthetic route. The modular nature of the synthesis facilitates incorporation of various side groups, including a substituent on the fourth position of the ring. Previous studies of triazolium-based ILs have shown functional group dependence of CO2 solubility, viscosity, and thermal stability. Phenyl substituents in particular were found to have a significant effect on both thermal stability and viscosity, with viscosity increasing by up to three-fold when comparing triazolium-based ILs substituted with phenyl groups to analogous ILs with propyl groups.23,24 In this study, the intermolecular interactions of a triazolium-based ionic liquid, 3-methyl-1-octyl-4phenyl-1H-triazol-1,2,3-ium bis(trifluoromethanesulfonyl)imide ([OMPhTz][Tf2N]) (Scheme 1),

ACS Paragon Plus Environment

3

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 24

has been evaluated utilizing nuclear magnetic resonance (NMR), mass spectrometry (MS), and small angle X-ray scattering (SAXS). Scheme 1. Structure and numbering of the ionic liquid 3-methyl-1-octyl-4-phenyl-1H-triazol1,2,3-ium bis(trifluoromethanesulfonyl)imide

Kunze et al. used spin-spin relaxation measurements of deuterated alkyl chains to show the high sensitivity of NMR to intermolecular interactions.21 Nuclear spin relaxation experiments have shown that 1,3-dialkylimidazolium ionic liquids form ordered supramolecular aggregates both in solution and as pure liquids.21,22,25 Evidence of ionic liquid supramolecular structure is a relatively new occurrence, and NMR promises to be a powerful tool for investigating aggregation in neat ionic liquid samples.15,19-22,25,26 Isotropic carbon atoms directly bound to protons undergo spin relaxation almost exclusively through the dipolar relaxation mechanism. Spin relaxation is also most effective when the rotation of the vector created by a bond is similar to the Larmor frequency of the nucleus.27 The rate of rotation of the vector created by a bond is referred to as the rotational correlation time, or correlation time. This correlation time is related to the relaxation rate, and can be calculated using a combination of spin-lattice relaxation time (T1) and Nuclear Overhauser Enhancement (NOE) factor measurements.28-32

ACS Paragon Plus Environment

4

Page 5 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Studies utilizing mass spectrometry (MS) have been used in literature to show both aggregation and ion-pairing of ILs due to coulombic effects.15,33 Small angle X-ray scattering (SAXS) has shown the formation of supramolecular structures in ILs containing aliphatic groups of approximately four carbons or greater.20,25,33,34 While other techniques such as SAXS and electrospray ionization mass spectrometry (ESIMS) can be used to determine molecular-level properties of materials, the properties of molecular segments are more difficult to determine. Regio-specific information on intermolecular interactions can be obtained by NMR, as nuclear spin relaxation is sensitive to both intra- and intermolecular relaxation mechanisms. By measuring the relaxation rates of each nucleus in the system, a relaxation “map” can be made of the molecule. Regiospecific information is valuable in narrowing down the region of the molecule that is related to behaviors such as Van der Waals interactions or hydrogen bonding. We report here an investigation of interactions within a triazolium-based ionic liquid; starting with intermolecular and intramolecular interactions narrowing in on segmental motion of the molecule; and ultimately viewing supramolecular properties.

Experimental Section [OMPhTz][Tf2N] was synthesized according to published procedures.23 Small-angle X-ray scattering (SAXS) was performed at beamline 7.3.3 of the Advanced Light Source (ALS) using X-rays of wavelength λ = 1.240 A° focused on a 50 by 300 µm spot. Samples were prepared by placing ~ 20 mg of ionic liquid in a sealed sample holder equipped with X-ray transparent Kapton windows. Two-dimensional scattering patterns were collected on a Pilatus 1M detector with an active area of 188 by 188 mm. The scattering patterns were

ACS Paragon Plus Environment

5

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 24

radially averaged, and the scattering intensity was corrected with the position chamber intensity using Nika version 1.51. Electrospray ionization mass spectrometry (ESI-MS) was performed in acetonitrile at 0.4 mL/min scanning from 50-1500 m/z and a direct infusion method with 4.5 KeV fragmentor voltage. The desolvation line was set to 140 °C, and the heat block was set to 200 °C. NMR measurements were performed using a 600 MHz Ultrashield Bruker NMR with an Avance III console. The employed probe was a direct observe BBFO plus (broadband including fluorine) probe. The sample of [OMPhTz][Tf2N] was placed in a Young tube under high vacuum (10-15 µm Hg) for one hour before being introduced to an argon atmosphere. This was repeated three times before the sample was placed under high vacuum for seven days prior to being sealed under an argon atmosphere. Relaxation measurements were carried out with relaxation delays at least five times T1, and the instrument was carefully tuned, shimmed, and the 90 degree pulse calibrated before each measurement. T1 relaxation measurements were done with an inversion recovery pulse sequence. 13

C T1 relaxation measurements were done with 1H decoupling throughout the relaxation time, to

prevent any cross relaxation effects.28,31,32 T2 measurements were made using a CPMG pulse sequence. Measured

13

C NOE enhancement factors were calculated as in Equation 1, where ηmeasured is

the NOE enhancement factor, NH is the number of protons directly bound to the carbon atom being measured, IPG is the intensity of the peak resulting from full NOE enhancement by polarization transfer from protons in a power gated experiment, and IIG is the intensity of the peak resulting from an inverse gated experiment, experiencing no NOE enhancement by proton

ACS Paragon Plus Environment

6

Page 7 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

polarization transfer.35 Pulse sequences used are included as text files in the supplemental information.

Calculations of rotational correlation times were performed by the method developed by Carper et al.28,29,31,32 and are included in the supplemental information. Iterative calculations were performed using the GoalseekTM function of Excel®.

Results and Discussion Spin Relaxation Relaxation measurements of [OMPhTz][Tf2N] were used to characterize segmental motions and molecular interactions. Spin-lattice relaxation times (T1) and spin-spin relaxation times (T2) were measured using NMR spectroscopy. These characteristics are known to be sensitive to several factors such as electronic effects, mobility, and intermolecular interactions. Magnetic field fluctuations are the major cause of nuclear spin relaxation. Localized association of nuclear spins can cause increased relaxation rates; in this way relaxation measurements and NOE factors can be used as evidence of localized interactions.22,25,36,37 Rotational correlation times are calculated from relaxation measurements and used to represent motions of molecule segments. Segmental motion can be very important for interactions with small molecules and for understanding interactions related to the physical properties of ionic liquids. Spin-Lattice Relaxation The spin-lattice relaxation rate (R1) is defined as the inverse of the spin-lattice relaxation time (T1). R1 represents the rate at which alpha and beta spin states establish equilibrium. R1 is

ACS Paragon Plus Environment

7

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 24

particularly sensitive to reorientation in a magnetic field and dipolar relaxation. Isotropic carbons with hydrogen covalently bound, due to the very high magnetic moment of protons, are relaxed primarily by the dipolar relaxation mechanism in the absence of paramagnetic relaxants. R1 is strongly influenced by dipole moments, bond distances, and molecular motion.28,30 Plotting the natural log of R1 against the inverse of temperature (from the Arrhenius Equation, Equation 2, where m and b are constants, and R is a rate constant) gives indication of the temperaturedependency of relaxation kinetics.38 A strong linear correlation in the Arrhenius plot is evidence of a single rate-limited thermally activated process.

The relationship between temperature and spin-lattice relaxation rate (R1) is shown in relaxation measurements of 1H nuclei in Figures 1 and 2. The nonlinear thermally dependent correlation shown by the aliphatic protons 6, 7, and 14 (Figure 1) indicates that at least two thermally dependent processes are affecting relaxation rates. It is suggested that the most likely secondary cause of relaxation is through intermolecular dipolar interactions (Van Der Waals), as this would be expected with the length of the aliphatic tail.22

ACS Paragon Plus Environment

8

Page 9 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 1. 1H R1 spin-lattice relaxation rates show non-linear behavior of the aliphatic protons 6, 7, and 14 in the Arrhenius plot. This is taken to represent the Van Der Waals interactions of aliphatic protons on different molecules. Lines are provided as a guide to the eye.

Figure 2. 1H R1 spin-lattice relaxation rates show linear behavior of protons 5, 13, 16, 17, and 18 in the Arrhenius plot. This is taken to represent that through bond dipolar relaxation is dominant. The legend refers to the numbering system delineated in Scheme 1. The lines provided are linear regressions. Non-linear behavior for

13

C spin-lattice relaxation of aromatic species in Figure 3 shows the

influence of chemical shift anisotropy (CSA) upon T1 relaxation. CSA is known to occur when molecules experience hindered rotation or tumbling and have orientation dependent electronic environments, and as such is a known contributor to relaxation of aromatic groups and those atoms close in proximity to aromatic groups.29 Higher temperatures cause decreased amounts of relaxation from CSA due to increased molecular tumbling. The decreased contribution of CSA to relaxation is visible in Figure 3 as a maximum in the graph. The linear correlation for the aliphatic carbons 12 and 13 in figure 4 implies first order thermodynamics of spin-lattice relaxation. The non-linear relaxation of carbons 6 and 7 placed in

ACS Paragon Plus Environment

9

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 24

Figure 4 provides a clear contrast that, while these carbons are still sp3 carbons, the direct link to an aromatic carbon causes the CSA effect to be strong enough to compete with dipolar relaxation.

Figure 3.

13

C R1 cation spin-lattice relaxation measurements show non-linear behavior for

aromatic carbons 4, 5, 15, 16, 17, and 18 in the Arrhenius plot. Chemical shift anisotropy is suggested to be the secondary mechanism causing greater than first order thermodynamic dependence of nuclear spin relaxation. The legend refers to the numbering system delineated in Scheme 1. Lines are provided as guides to the eye.

ACS Paragon Plus Environment

10

Page 11 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 4.

13

C R1 cation spin-lattice relaxation measurements show non-linear behavior for

carbons attached to aromatics (6 and 14) and linear behavior for aliphatic carbons (12 and 13) in the Arrhenius plot. The legend refers to the numbering system delineated in Scheme 1. Lines for carbons 6 and 14 are provided as guides to the eye. Lines for carbons 12 and 13 are linear regressions.

Spin-Spin Relaxation The spin-spin relaxation rate (R2) is defined as the inverse of the spin-spin relaxation time (T2). R2 represents the rate of phase coherence loss in spin states. Spin-spin interactions causing relaxation can be both intermolecular and intramolecular. In the case of [OMPhTz][Tf2N], the molecule is considered too rigid for the functional groups to undergo intramolecular interactions. Consequently, R2 can be very sensitive to intermolecular interactions.21 An Arrhenius plot (Figure 5) clearly shows first order relaxation kinetics with respect to temperature for the anion, Tf2N, for both 13C and 19F spin relaxation. This first order rate kinetics implies that the CF3 groups undergo relaxation mainly via through-bond dipolar relaxation, and that the CF3 groups are not involved in spin relaxation of cation nuclei. Close association of the

ACS Paragon Plus Environment

11

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 24

cation and anion would result in CF3 groups being close enough for spin interactions with the cation. This suggests that interactions between the cation and anion are weak.

Figure 5. R1 and R2 relaxation rates of 13C and 19F nuclei show linear behavior for the anion in the Arrhenius plot. The first order thermodynamic dependence of nuclear spin relaxation in the anion is attributed to exclusively through-bond C-F dipolar relaxation. Lines provided are linear regressions.

Figure 6. 1H T2 measurements show linear behavior of the aliphatic protons 5 and 14-18 in the Arrhenius plot. The first order thermodynamic dependence of spin relaxation in the aromatic

ACS Paragon Plus Environment

12

Page 13 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

regions of the molecule imply that spin-spin interactions are minimal, and so intermolecular distances are larger for this region. The legend refers to the numbering system delineated in Scheme 1. Lines provided are linear regressions. The relationship between temperature and cation spin relaxation rate is shown by 1H spin-spin relaxation (R2) kinetics in Figure 6. A strong linear correlation is evidence of a single ratelimited thermally activated process, which is taken to be spin-relaxation caused by through-bond dipolar interactions. The non-linear properties shown by the aliphatic protons 6, 7, and 13 in Figure 7 are presented as evidence of regio-specific intermolecular interactions. Specifically, these interactions are interpreted as aliphatic tail groups of neighboring molecules in close enough proximity to allow spin-spin relaxation.

Figure 7. 1H T2 measurements show non-linear behavior of the aliphatic protons 6, 7, and 13 in the Arrhenius plot. This is interpreted as being caused by spin-spin interactions with the aliphatic tails of neighboring molecules. The legend refers to the numbering system delineated in Scheme 1. Lines are provided as a guide to the eye.

ACS Paragon Plus Environment

13

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 24

Figure 8 shows first order 13C spin-spin relaxation of the aromatic groups and carbons directly bound to the aromatic groups. This clean relationship again is taken to represent a lack of interactions between this group and neighboring molecules. The terminal carbons of the octyl chain show non-linear relationships (Figure 9), which are taken as evidence of Van Der Waals interactions between the aliphatic tails of neighboring cations.

Figure 8. 13C T2 measurements show linear behavior for carbons 4-6 and 14-18 in the Arrhenius plot. All carbons on or adjacent to aromatic rings show first order thermodynamic dependence of spin-spin relaxation. First order dependence of relaxation is taken as relaxation being dominated by through-bond dipolar relaxation. The legend refers to the numbering system delineated in Scheme 1. Lines provided are linear regressions.

ACS Paragon Plus Environment

14

Page 15 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 9. 13C T2 measurements show non-linear behavior for carbons 12 and 13 in the Arrhenius plot. This greater than first order dependence of relaxation is attributed to the spin-spin association of the aliphatic tails of neighboring molecules. The legend refers to the numbering system delineated in Scheme 1. Lines provided are linear regressions.

Segmental Dynamics Relaxation of spin ½ nuclei is highly dependent upon motion. For carbon atoms directly bound to hydrogen atoms, careful design of relaxation experiments allows for measurement of T1 relaxation times which almost exclusively represent relaxation through the dipolar relaxation mechanism.28,29,31,32 The dipolar coupling constant can then be calculated using Equation 3:

where Dij is the dipolar coupling constant, µ0 is the permeability of a vacuum, γC is the gyromagnetic ratio of carbon, γC is the gyromagnetic ratio of hydrogen, ħ is the reduced Planck’s constant, and rij is the length of the bond between carbon i and proton j (assumed to be 1.09 Å). Dipolar relaxation can be related to both the rotation rate of the vector between the spin systems and the NOE factor by Equation 4:

ACS Paragon Plus Environment

15

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 24

Where τc represents the rotational correlation time (The time it takes for the vector defined by the C-H bond to rotate by one radian), NH is the number of directly bound protons, and ωC and ωH are the Larmor frequencies of carbon and hydrogen, respectively. τc, found iteratively from inserting Equation 3 into Equation 4, can be corrected using the measured and theoretical maximum NOE factor (ηmeasured and ηmax, respectively) by Equations 5 and 6.29

A table containing the values of constants can be found in the supplemental information with an example of the rotational correlation time calculation. τc has been calculated for all C-H bonds which have unambiguously assigned

13

C

resonances.28,29,31,32,39 These atoms are labeled as 12, 13, 14, 6, 5, and 18 in Scheme 1. Calculated τc values are shown in Table 1 for varying temperatures. The values in the table agree with expectations: methyl groups have very short τc values due to the rotation of the H3C-C bond in the case of 13 and the H3C-N bond in the case of 14. The 6 and 12 positions show significantly slower motions, with the 6 position being further slowed by its proximity to the heteroaromatic ring. The 18 and 5 positions are the least flexible with motions being slowed by their presence in aromatic ring structures, requiring rotation of the entire ring to rotate the spin systems. In particular, carbon position 5 shows further slowing by its ring being bound to two pendant groups. Carbon position 18 shows a slight increase in τc due to being in the para phenyl position along the axis of rotation for the phenyl ring. Table 1. τc values (ps) for all measureable C-H bonds

ACS Paragon Plus Environment

16

Page 17 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

T (K)

Assigned carbon positions 12

13

14

6

5

18

302

63

17

17

113

263

198

313

62

15

15

96

274

234

324

46

11

16

61

234

228

329

45

10

20

60

252

225

336

45

10

21

58

260

228

347

38

9

26

71

406

334

358

38

9

26

69

408

365

369

32

8

27

91

488

536

Assigned carbon positions are labeled by carbon # in header. τc (rotational correlation time) values (ps) in [OMPhTz][Tf2N] are calculated from T1 and NOE factor measurements.

Aggregation ESI-MS ESI-MS methods have been used previously for the detection of cation clusters in solution. As mass spectrometry detects charge to mass ratios, the ratio of the cation ([OMPhTz]+) peak intensity relative to the peak representing two cations and one anion ([OMPhTz]2[Tf2N]+) can be used to indicate the relative level of anion-cation clustering.25

ACS Paragon Plus Environment

17

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 24

Figure 10. ESI-MS of [OMPhTz][Tf2N] with expansion of the 550-1050 m/z region shows the lack of aggregate formation. A peak representing 2MTf2N+ (m/z=842) can be seen in the expansion at