Determination of inter-ionic and intra-ionic interactions in a monofluorinated imidazolium ionic liquid by a combination of X-ray crystallography and NOE NMR spectroscopy

Article abstract of DOI:10.1002/mrc.4608

We report the very first application of a Transient 1D ¹H{¹⁹F} NOE NMR experiment in neat ionic liquids. In comparison with classical 2D HOESY NMR spectroscopy, a substantial reduction in measurement time is gained with comparable quality and information content of the spectra. In combination with classical X-ray crystallography, we have applied this technique for the determination of inter-ionic distances (i.e. probabilities of presence) utilizing an ionic liquid containing a monofluorinated imidazolium cation. Copyright

Full text of DOI:10.1002/mrc.4608

Determination of Inter- and Intra-ionic Interactions
in a Monofluorinated Imidazolium Ionic Liquid by a
Combination of X-Ray Crystallography and NOE
NMR Spectroscopy
Yves Lingscheid, Mathias Paul, Andreas Bröhl, Jörg-Martin Neudörfl,* and Ralf Giernoth*
Universität zu Köln, Department für Chemie, Köln, Germany
Abstract
We report the very first application of a Transient 1D 1H{19F} NOE NMR experiment in neat
ionic liquids. In comparison to classical 2D HOESY NMR spectroscopy,[1,2] a substantial
reduction in measurement time is gained with comparable quality and information content of
the spectra. In combination with classical X-ray crystallography, we have applied this
technique for the determination of inter-ionic distances (i.e. probabilities of presence)
utilizing an ionic liquid containing a monofluorinated imidazilium cation.
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1002/mrc.4608
This article is protected by copyright. All rights reserved.

Introduction
Many ionic liquids (ILs) are well-known to exhibit a certain defined supramolecular
structure. Historically, there was quite early literature precedence that ILs may even show
some kind of middle-ranged order like what is known from liquid crystals.[3-5] The authors
were able to derive their picture by combining 1H NOESY NMR data with crystal structures
of their ILs. Based on these fundamental publications, other authors, predominantly Mele,
have exploited NOE NMR methods to learn about the solution structure of various ionic
liquids.[6-8] In contrast, molecular dynamics simulations show that the liquid-phase structure
of common ionic liquids resembles clustered regions of high and low polarity, rather than a
high degree of long-range order.[9]
NOESY NMR data are of limited value for the determination of absolute distances, for the
quality of their information content substantially suffers from spin diffusion. For that reason
we demonstrated the application of HOESY NMR experiments of neat ionic liquids in an
earlier publication[1] with the goal to determine the highest preferred probability of presence
(i.e. the average inter-ionic distance) of the anions around the cations. Much to our surprise,
the results of first experiments gave a fairly inconsistent picture. Some of the “distances”
reported there seemed reasonable while others did not. One possible reason for these findings
was delivered by Weingärtner and Steinhauser shortly after in a series of publications
utilizing our own published NMR data.[10-12] Essentially, by reinvestigating the fundamental
mathematics of the NOE theory, they stated that for our ionic liquids the so-called “extreme
narrowing limit” condition was not met. In consequence, NOE signals measured that way
would stem from direct next neighbors for only 50 % of the cases. The rest of the signal
intensity would stem from “the bulk”.
The application of HOESY NMR spectroscopy in neat ionic liquids can be limited due to
extremely long measuring times. For a typical reproducible experiment measuring times of
20–30 h per NOE build-up curve was needed in our set-up.[1] Therefore, we were looking for
a faster but still reliable alternative. In 1999, Gerig published a NMR pulse sequence for
1H{19F} NOE experiments which was intended to be applied to enzymes.[13] This sequence
applies gradient pulses for removal of unwanted magnetization. In contrast to conventional
HOESY NMR, the proton magnetization is rotated to the transverse plane during mixing
time, suppressing classical NOE artefacts. This way, 1H{19F} NOE enhancement can be
measured in a one-dimensional experiment like in a classical NOE Difference experiment
without the necessity of a difference spectrum. A modified version of this experiment has
been used to study solvent-solute interactions.[14]
To be able to test the new experimental setup and to cross-check the results with the
Weingärtner/Steinhauser theory, we synthesized a monofluorinated imidazolium ionic liquid
in which the fluorine nucleus can be used as an internal standard for referencing the NOE
signal intensities with a known H-F distance. With this fluorinated IL, H-F distances within
the cation should be accessible.
This article is protected by copyright. All rights reserved.

Experimental
Synthesis of the ionic liquid
4-Nitro-1H-imidazol:[15] A 250 mL round-bottomed flask was charged with 60 mL of
sulfuric acid (96 %). Under stirring and external cooling via a water bath, 20 g 1H-imidazole
was added in four portions of 5 g each. 60 mL of nitric acid (70 %) was added slowly while
the mixture heated up to 70 °C. (NB: After addidtion of 20 mL of the nitric acid the mixture
turned black and evolved nitrous gases. The mixture was cooled via an external water bath
until gas evolution was finished. Slow addition of the rest of the nitric acid resulted in a clear
solution again. The whole process lasted ca. 1 h.) The mixture was then heated to 100 °C and
stirred for 5 h until gas evolution had finished. The mixture was cooled to room temperature;
100 g ice was added in one portion and stirred. The solution was neutralized with
concentrated ammonia solution, which induced precipitation of colorless needles. The solids
were filtered off and washed with 100 mL of cold water. The product was dried in vacuo over
calcium chloride. Yield: 15.144 g (46 %) of a colorless crystalline solid. m.p.: decomposition
> 250 °C. FT IR (ATR):  [cm–1] = 3142 (m), 3007 (m), 2881 (m), 2785 (m), 2361 (m),
1557 (m), 1491 (s), 1431 (s), 1375 (vs), 1364 (vs), 1337 (s), 1244 (s), 1142 (m), 1080 (m),
987 (s), 931 (s), 860 (s), 758 (s), 652 (s), 608 (s). 1H NMR (300ꢀMHz, DMSO-d6):
δꢀ[ppm] = 12.80 (br. s, 1H, H-N), 8.32 (d, 3JHHꢀ=ꢀ1.2ꢀHz, 1H, H-5), 8.32 (d, 3JHHꢀ=ꢀ0.5ꢀHz, 1H,
H-2). 13C{1H} NMR (75ꢀMHz, DMSO-d6): δꢀ[ppm] = 147.5 (s, C-4), 136.9 (s, C-2), 119.1 (s,
C-5).
4-Fluoro-1H-imidazole:[16] In an quartz irradiation tube with cooling finger connected to a
cryostat 2.26 g (20 mmol, 1.0 ꢀeq) 4-nitro-1H-imidazole were dissolved in 100 mL of
tetrafluoroboric acid (50 %) and cooled to –10 °C. The solution was stirred while nitrogen
was bubbled through. Zinc dust (4.25ꢀg, 65 mmol, 3.25 eq) was added in small portions (ca.
25 mg each). Small samples were taken from the reaction mixture, diluted with water and
analyzed via UV spectroscopy. As soon as the nitro band at 297 nm disappeared from the
spectrum the reaction was complete. In the following, 1.52ꢀg (22ꢀmmol, 1.1ꢀeq) of sodium
nitrite were dissolved in 5 mL of water. This solution was added dropwise to the yellow
reaction mixture, turning it dark orange. It was diluted with 110 mL of tetrafluoroboric acid
(50 %) and the mixture was warmed to 0 °C. The reaction vessel was put into a photo reactor
operating at 254 nm. Again, samples were taken to follow the disappearance of the
diazonium-band at 270 nm (after approx. 3 h of irradiation). The reaction mixture was
transferred to a glass beaker and cooled with ice. The pH-value was carefully raised to pH=6
by addition of solid sodium hydroxide, then further to pH=7 by adding concentrated sodium
hydroxide solution. Without removal of the zinc hydroxide, the mixture was extracted four
times with 100 mL of ethyl acetate. The combined organic layers where dried over
magnesium sulfate and the solvent was evaporated. The residue was sonicated with
chloroform to give a suspension, which was filtered through a short silica gel column. The
This article is protected by copyright. All rights reserved.

solvent was evaporated and the crystalline residue was purified further by sublimation at
40 °C and 9.0ꢀ·ꢀ10–3 mbar. Yield: 414.0 mg (24 %) of colorless crystals. m.p.: 101–103 °C. FT
IR (ATR): [cm–1] = 3132 (m), 3038 (m), 2650 (w), 1579 (s), 1489 (m), 1474 (m), 1325 (s),
1287 (s), 1229 (m), 1180 (w), 1072 (w), 1001 (s), 880 (brm), 816 (s), 739 (s), 621 (s). 1H
NMR (300ꢀMHz, chloroform-d1): δꢀ[ppm] = 10.51 (br. s, 1H, H-N), 7.28 (s, 1H, H-2), 8.32
(dd, 3JHFꢀ=ꢀ8.0ꢀHz, 3JHHꢀ=ꢀ1.5ꢀHz, 1H, H-5). 13C{1H} NMR (75ꢀMHz, chloroform-d1):
δꢀ[ppm] = 157.9 (d, 1JCFꢀ=ꢀ233.7ꢀHz, C-4), 129.0 (d, 3JCFꢀ=ꢀ16.1ꢀHz, C-2), 94.6 (d,
2JCFꢀ=ꢀ38.5ꢀHz, C-5). 15N NMR (41ꢀMHz, MeCN-d3): δꢀ[ppm] = 154 (s, 1N, N-1), 235 (s, 1N,
N-3). 19F NMR (282ꢀMHz, DCM-d2): δꢀ[ppm] = –138.75 (d, 3JHFꢀ=ꢀ8.0ꢀHz, F-4).
1-Butyl-4-fluoro-1H-imidazole:[17] An inert Schlenk tube was loaded with 140.0 ꢀmg
(1.63 mmol, 1.0 eq) of 4-fluoro-1H-imidazole, dissolved in 4 mL N,N-dimethyl formamide
(anhydrous, 99.8 %) and externally cooled with ice. 78.2 mg (1.96 mmol, 60 %, 1.2 eq) of
sodium hydride in mineral oil were added and the mixture was stirred vigorously for 45 min.
0.2 mL (1.96ꢀ mmol, 1.1 eq) n-butyl iodide was added and stirred for one more hour. The ice
bath was removed and the mixture was stirred for 20 h. The solution was diluted with 20 mL
of diethyl ether and washed with saturated sodium chloride solution. The aqueous layer was
extracted two more times with 10 mL of ether before drying the combined organic layers
over sodium sulfate. The solvent was evaporated under reduced pressure and
N,N-dimethylformamide was distilled off using Kugelrohr distillation (monitored by TLC:
EtOAc/Et2O 1/1 on silica gel. Rf-values: 0.14–0.25 = N,N-dimethylformamide; 0.38 = 4-
fluoro-1H-imidazole; 0.59 =1-butyl-4-fluoro-1H-imidazol; 0.85 = aliphatic ring-opening
products). The crude product was purified further via column chromatography under the
same conditions as the TLC. Yield: 218 mg (94 %) of a pale yellow oil. Elemental analysis:
Calculated: C: 59.14 %, H: 7.80 %, N: 19.70 %. Found: C: 59.27 %, H: 7.52 %, N: 19.30 %.
= 142.092 (M+). FT IR (ATR): [cm–1] = 3157 (vw), 3113 (vw), 2961 (m),
HR EI MS:
2936 (m), 2876 (m), 1566 (vs), 1504 (s), 1460 (brm), 1416 (m), 1354 (m), 1325 (s), 1229
(m), 997 (vs), 806 (brm), 716 (brm), 644 (m), 621 (m). 1H NMR (400ꢀMHz, DCM-d2):
δꢀ[ppm] = 6.99 (s, 1H, H-2), 6.42 (dd, 3JHFꢀ=ꢀ8.0ꢀHz, 4JHHꢀ=ꢀ1.6ꢀHz, 1H, H-5), 3.83 (t,
3JHHꢀ=ꢀ7.1ꢀHz, 2H, H-1‘), 1.72 (m, 2H, H-2‘), 1,.0 (m, 2H, H-3‘), 0.92 (t, 3JHHꢀ=ꢀ7.4ꢀHz, 3H,
H-4‘).3C{1H} NMR (101ꢀMHz, DCM-d2): δꢀ[ppm] = 157.6 (d, 1JCFꢀ=ꢀ231.3ꢀHz, C-4), 130.5 (d,
3JCFꢀ=ꢀ16.8ꢀHz, C-2), 97.4 (d, 2JCFꢀ=ꢀ39.2ꢀHz, C-5), 48.1 (s, C-1‘), 33.1 (s, C-2‘), 20.0 (s, C-3‘),
13.7 (s, C-4‘).19F NMR (376ꢀMHz, DCM-d2): δꢀ[ppm] = –136.36 (d, 3JHFꢀ=ꢀ8.0ꢀHz, 1F, F-4).
1-Butyl-4-fluoro-3-methyl-1H-imidazol-3-ium tetrafluoroborate: A Schlenk tube with
stirring bar and inert atmosphere was loaded with 128.0ꢀmg (0.823ꢀmmol, 95%, 0.92 eq) of
trimethyloxonium tetrafluoroborate and suspended in 7.5 mL of dry dichloromethane.
127 mg (0,893 mmol, 1,00 eq) of 1-butyl-4-fluoro-1H-imidazole was dissolved in 7.5 mL of
dry dichloromethane and added to the suspension. After stirring for 19 h the solvent was
evaporated under reduced pressure. The crude product was purified by column
chromatography. Thereby the desired product was first washed with dichloromethane. Then–
–gradually––the eluent was changed to a mixture of dichloromethane and methanol (95/5) to
This article is protected by copyright. All rights reserved.

wash the ionic liquid off the column. The solvents were removed under reduced pressure. (To
remove dissolved silica gel, the IL was dissolved in DCM and passed through a glass frit
(Por. 4).) To get rid of any solvent residues and water, the ionic liquid was dried in high
vacuum. Yield: 218 mg (94 %) of a colorless solid or a slightly yellow supercooled liquid.
m.p.: 45 °C. Phase transitions (DSC): –10 °C; –70 °C. HR ESI MS:
= 157.11349
([K]+ꢀ=ꢀC8H14N2F); 401.23061 ([K2A]+ꢀ=ꢀC16H28N4BF6); 87.00371 ([A]– = BF4); 331.12033
([KA2]– = C8H14N2B2F9). FT IR (ATR): [cm–1] = 3152 (vw), 3107 (vw), 2963 (vw), 2926
(vw), 2359 (m), 1647 (vs), 1576 (m), 1472 (brm), 1152 (m), 1024 (brs), 839 (brm), 777
(brm), 646 (m), 619 (m). 1H NMR (400ꢀMHz, DCM-d2): δꢀ[ppm] = 8.54 (t, Jꢀ=ꢀ2.0ꢀHz, 1H,
H-2), 7.08 (dd, 3JHFꢀ=ꢀ5.9ꢀHz, 3JHHꢀ=ꢀ2.0ꢀHz, 1H, H-5), 4.15 (t, 3JHHꢀ=ꢀ7.4ꢀHz, 2H, H-1‘), 3.84 (s,
3H, H-1‘‘), 1.85 (m, 2H, H-2‘), 1.37 (m, 2H, H-3‘), 0.95 (t, 3JHHꢀ=ꢀ7.4ꢀHz, 3H, H-4‘).13C{1H}
NMR (101ꢀMHz, DCM-d2): δꢀ[ppm] = 147.6 (d, 1JCFꢀ=ꢀ267.8ꢀHz, C-4), 131.4 (s, C-2), 102.4
(d, 2JCFꢀ=ꢀ21.5ꢀHz, C-5), 51.5 (s, C-1‘), 32.5 (s, C-1’’), 31.8 (s, C-2‘), 19.6 (s, C-3‘), 13.4 (s,
C-4‘).15N NMR (51ꢀMHz, DCM-d2): δꢀ[ppm] = 176 (s, 1N, N-1), 157 (s, 1N, N-3). 19F NMR
2(82ꢀMHz, DCM-d2): δꢀ[ppm] = –145.35 (d, 3JHFꢀ=ꢀ5.9ꢀHz, 1F, F-4), –151.53 (s, 0,8F, 10BF4), –
151.53 (s, 3,2F, 11BF4). 11B NMR (96ꢀMHz, DCM-d2): δꢀ[ppm] = 10.35 (m, BF4). Water
content (Karl-Fischer titration): 36.6 ppm H2O. Ion chromatography: The sample is halogen-
free. X-ray chrystallographic data can be found in the Supporting Information.
NOE NMR experiments: 2D HOESY NMR experiments have been run and mathematically
fitted as described earlier.[1] The fits were used to extract the cross relaxation rates for each
individual proton resonance following the given mathematical equation (vide infra). They
were normalized with a proportionality factor p using the following equation:
with NI = number of detected nuclei, NS = number of excited nuclei.
Transient NOE experiments were set up and run as described by Gerig.[14] The resulting NOE
build-up curves were fitted exactly the same way as for the 2D HOESY NMR experiments.
X-ray crystallography: Crystals suitable for single crystal X-ray diffraction were grown
from melt. The orthorhombic crystals were measured on a Bruker D8 diffraktometer
equipped with a copper microfocus X-ray source and a PHOTON 100 detector. The structure
was solved with SHELXT[18] and refined with SHELXL2013.[19,20] CCDC 1541424 contains
the supplementary crystallographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/structures.
This article is protected by copyright. All rights reserved.

Results and Discussion
Synthesis of the monofluorinated ionic liquid
The synthetic pathway to a monofluorinated imidazole derivative is depicted in Scheme 1.
On first sight, the synthesis looks pretty much straightforward. Nonetheless, it was already
known that a simple Baltz-Schiemann reaction does not work for aromatic heterocycles of
this type, since one of the intermediates proved to be fairly unstable.[21] We were able to
circumvent these problems by utilizing a photochemical version of the Baltz-Schiemann
fluorination that operates at much lower temperature (according to Kirk and Cohen).[16] The
monofluorinated ionic liquid (m.p.: 40 °C) could then be synthesized by double alkylation of
the imidazole derivative (Scheme 2). The high selectivity of the first alkylation step is a result
of the strong electronic influence of the fluoride substituent on the aromatic ring.
Crystal structure of the monoflurinated ionic liquid
The crystal structure gives a visualization of the respective surroundings of the cation and the
anion (Figure 1). From the X-ray data, we determined all relevant H–F distances, intra- as
well as inter-ionic. To this end, we focussed on next-neighbors only (first coordination
sphere). The classical distance limit between two nuclei showing a detectable NOE is roughly
5 Å.[22] Therefore, we chose 4.5 Å as a cut-off criterion. Complete lists with all
crystallographic data can be found in the Supporting Information.
NOE data and distance measurement
Figure 2 exemplarily shows a typical NOE build-up curve and the respective mathematical fit
(as described earlier in detail).[1] The data are taken from a Transient NOE experiment of the
neat monofluorinated IL.
The NOE build-up curve visualizes the time dependence of the NOE between two nuclei
(here, H and F). Mathematically derived from the fundamental Solomon equations, it can be
described as follows:
– –
–
–
(1)
This article is protected by copyright. All rights reserved.

Here, R is the total relaxation rate,  is the mixing time, and  represents the cross-relaxation
rate. In our earlier publication,[1] this equation proved to be insufficient for fitting the NOE
build-up (which we need for determination of HF), so we applied a more precise version
which takes into account all individual relaxation rates for all proton and fluorine nuclei:
–
(2)
(3)
–
(The original publication contains a mistake, writing “sin” instead of “sinh”.)
R in this equation denotes the individual relaxation rate for each nucleus; M0 represents the
intensity of the respective diagonal signal in the 2D experiment for  = 0. Equations 2 and 3
have been used to fit the NOE build-up as depicted in Figure 2 and to determine all relevant
cross relaxation rates.
In this classical form of the NOE theory, the intensity of an NOE[23] signal is proportional to
the distance between the two nuclei to the power of –6:[22]
–
(4)
It does not matter whether inter- or intramolecular NOEs are considered. Therefore, the
following ratio can be drafted:
–
–
(5)
Since one given H-F distance is known from the crystal structure of our ionic liquid (see
Supporting Information), and since the NOE data can be measured, the interionic distance
can be obtained via:
–
–
(6)
The measured NOE is the sum of the interaction between the two nuclei and the individual
relaxation rates. Therefore, the result of this equation differs substantially for different mixing
times. To circumvent this effect, the complete NOE kinetics must be taken into account. By
fitting the measured NOE build-up curve for each cross signal to the mathematical equation
for the NOE (as conclusively described earlier[1]), we get the cross relaxation rate  for every
observed nucleus. With this time-independent observable, the formula transforms into:
–
–
(7)
or:
This article is protected by copyright. All rights reserved.

(8)
This equation is qualified and used for the determination of preferred probabilities of
presence (“distances”).
From the crystal structure of our monofluorinated IL, we determined two intra-ionic distances
as potential internal references: the distance F––H-5 and F––H-1’’ (for nomenclature see
picture in Table 1). Since we found two slightly different ion pairs within the elemental cell,
the basis for the internal reference was the average of the two (H-5) or two closest (H-1’’)
distances. With this approach, we ended up with an intramolecular F––H-5 distance of
2.7925 Å and a F––H-1’’ distance of 2.812 Å.
Results of the NOE experiments
By incorporation of the measured cross relaxation rates into Equation 8, each intramolecular
H-F distance that has been determined from the crystal structure can be used to produce a set
of normalized “distances” between anion and cation for our monofluorinated IL. The results
of this process based on the data from the two different NOE NMR experiments are listed in
Table 1. In addition, we have determined all relevant H-F distances also from our crystal
structure, to be able to compare the values. This approach has successfully been followed
earlier already by other groups.[24]
The data from Table 1 are visualized with the help of bar graphs in Figure 3.
For the 2D HOESY NMR data, the resulting numbers perfectly match with a slight
systematic overestimation in the case the interionic F––H1’’ distance is used for
normalization. This result fits to the picture that the methyl group has the potential of free
rotation, which incorporates some uncertainty into the “fixed” F-H distance.
The data taken from the Transient NOE NMR experiment look differently. Here, the numbers
are substantially higher throughout, and the difference based on the two normalization
standards is also higher. Obviously, the substantial gain in measurement time comes for the
price of lower absolute accuracy.
The shortest inter-ionic F-H distance is always between the anion and H-2, with the next
closest being H-1’’. For the HOESY data, H-2’ is next, followed by H-1’, while this order is
reversed in the Transient data.
Interestingly, the HOESY-based data match very well with the data taken from the actual
crystal structure. Only H-4’ seems to deviate substantially, which might be explained by the
fact that this terminal group of the alkyl chain comes close to the next-neighbor anion from
the next “elemental cell” already, which must result in a seemingly shorter distance.
This article is protected by copyright. All rights reserved.

The high level of data matching between HOESY and crystal structure derived data
seemingly contradicts the mathematical forcast by Steinhauser and Weingärtner. In a series of
publications from 2014 onwards[10-12] they claim that for viscous ionic liquids the extreme
narrowing limit condition is not met. This results in additional relaxation processes that stem
from slow tumbling of the ions in the liquid phase via residual dipolar coupling.
Consequently, only a fraction of the NOE signal stems from direct next neighbors
(approximately 50 %). The rest of the signal comes from “the bulk”. Therefore, equations 6–
8 of our approach (vide supra) should not be suitable for the measurement of distances in neat
ILs. Apparently, at least in this special case, it is possible to measure inter-ionic “distances”
in terms of highest probabilities of presence for neat ionic liquids this way. The assumption
that our IL is below the extreme narrowing limit seems reasonable. Therefore, the only
sensible explanation of our data is that the liquid-phase structure of this IL is reasonably close
to the structure of its crystal. Thus, at least in major areas, there seems to be a substantial
amount of short- as well as long-range order.
Conclusion
We have synthesized a new monofluorinated imidazolium tetrafluoroborate ionic liquid for
NMR based inter-ionic distance measurement in the liquid phase. From the crystal structure
of the IL we extracted all accessible F-H distances in the first coordination sphere.
Conventional 2D HOESY NMR experiments generated reliable data for the determination of
distances (i.e. highest probabilities of presence) in the liquid phase, but the experiments prove
to be highly time-consuming. A much faster Transient NOE experiment was applied on neat
ILs for the first time. NOE data produced this way gave the same results on a qualitative level
but for the cost of lower accuracy. The overall picture of anion-cation distances drawn by
these methods matches our current picture, which is based on X-ray crystallography, neutron
diffraction and MD simulations, surprizingly well.
Acknowledgements
Financial support was granted by the German Science Foundation (DFG, SPP 1191 “Ionic
Liquids”), by the Fonds of the German Chemical Industry (FCI, fellowship for A. B.), and by
the project "Sustainable Chemical Synthesis (SusChemSys)" which is co-financed by the
European Regional Development Fund (ERDF) and the state of North Rhine-Westphalia,
Germany, under the Operational Programme "Regional Competitiveness and Employment"
2007 – 2013.
This article is protected by copyright. All rights reserved.

References
[1]
[2]
[3]
Y. Lingscheid, S. Arenz, R. Giernoth, ChemPhysChem 2012, 13, 261–266.
A. Mele, Chimica Oggi-Chemistry Today 2010, 28, 48–55.
J. Dupont, P. Suarez, R. F. de Souza, R. A. Burrow, J. P. Kintzinger, Chem. Eur.
J. 2000, 6, 2377–2381.
[4]
C. S. Consorti, P. Suarez, R. F. de Souza, R. A. Burrow, D. H. Farrar, A. J.
Lough, W. Loh, L. da Silva, J. Dupont, J. Phys. Chem. B 2005, 109, 4341–4349.
R. Hayes, G. G. Warr, R. Atkin, Chem. Rev. 2015, 115, 6357–6426.
M. Moreno, F. Castiglione, A. Mele, C. Pasqui, G. Raos, J. Phys. Chem. B 2008,
112, 7826–7836.
[5]
[6]
[7]
A. Mele, G. Romanò, M. Giannone, E. Ragg, G. Fronza, G. Raos, V. Marcon,
Angew. Chem. Int. Ed. 2006, 45, 1123–1126.
[8]
A. Mele, C. D. Tran, S. H. De Paoli Lacerda, Angew. Chem. Int. Ed. 2003, 42,
4364–4366.
[9]
J. N. A. Canongia Lopes, A. A. H. Pádua, J. Phys. Chem. B 2006, 110, 3330–
3335.
[10]
[11]
S. Gabl, O. Steinhauser, H. Weingärtner, Angew. Chem. Int. Ed. 2013, 52, 9242–
9246.
S. Gabl, C. Schröder, D. Braun, H. Weingärtner, O. Steinhauser, J. Chem. Phys.
2014, 140, 184503.
[12]
[13]
[14]
[15]
D. Braun, O. Steinhauser, Phys. Chem. Chem. Phys. 2015, 17, 8509–8517.
J. T. Gerig, Magn. Reson. Chem. 1999, 37, 647–652.
J. T. Gerig, J. Org. Chem. 2003, 68, 5244–5248.
G. Wuellner, F. W. Herkenrath, A. Juelich, Methods for the Production of 2-
Halo-4-Nitroimidazole and Intermediates Thereof, 2013, US-Pat.
WO2010021409A1.
[16]
[17]
[18]
[19]
K. L. Kirk, L. A. Cohen, J. Org. Chem. 1973, 38, 3647–3648.
H. Zhao, F. W. J. Foss, R. Breslow, J. Am. Chem. Soc. 2008, 130, 12590–12591.
G. M. Sheldrick, Acta Crystallogr. A Found. Adv. 2015, 71, 3–8.
C. B. Huebschle, G. M. Sheldrick, B. Dittrich, J. Appl. Crystallogr. 2011, 44,
1281–1284.
[20]
[21]
[22]
G. M. Sheldrick, Acta Crystallogr. C Struct. Chem. 2015, 71, 3–8.
K. L. Kirk, L. A. Cohen, J. Am. Chem. Soc. 1971, 93, 3060–3061.
D. Neuhaus, M. Williamson, The Nuclear Overhauser Effect in Structural and
Conformational Analysis., Wiley, New York, 2000.
A. W. Overhauser, Phys. Rev. 1953, 92, 411–415.
[23]
[24]
F. Castiglione, G. B. Appetecchi, S. Passerini, W. Panzeri, S. Indelicato, A.
Mele, J. Mol. Liquids 2015, 1–8.
This article is protected by copyright. All rights reserved.

Table 1: Preferred probabilities of presence (= “distances”, given in Å) between all protons and the
F-atoms of the anion for [C4C1F4im]BF4, based on Equation 8. The numbers in the last row are
average distances of the two anions in the elemental cell taken directly from the crystal structure.
Distance Anion F––X: H-2 H-5 H-1' H-1'' H-2' H-3' H-4'
2D HOESY
normalized via...
F––H-5intra = 2.79
F––H-1’’intra = 2.81
Transient NOE
2.45 2.60 3.04 2.51 2.61 3.07 3.32
2.46 2.62 3.06 2.53 2.63 3.09 3.35
normalized via...
F––H-5intra = 2.79
F––H-1’’intra = 2.81
Crystal Structure
2.95 3.18 3.45 3.26 3.64 3.79 3.87
3.07 3.31 3.60 3.39 3.79 3.95 4.03
2.30 2.39 2.81 2.67 2.62 3.18 2.85
This article is protected by copyright. All rights reserved.

Figure 1: Parts of the crystal structure of the monofluorinated ionic liquid. Surrounding of the anion
(left) and the cation (right) within 4.5 Å H–F distance in solid state. For the disordered terminal
methyl group only one of the two possible orientations is shown.
This article is protected by copyright. All rights reserved.

Figure 2: Heterocuclear H,F NOE build-up curve and mathematical fit[1] of the H-2 proton of the
described monofluorinated ionic liquid with fluorine of the anion. The shown example has been
produced by application of the Transient NOE experiment.[14]
This article is protected by copyright. All rights reserved.

4.50
4.00
3.50
3.00
2.50
2.00
1.50
1.00
0.50
0.00
H-2
H-5
H-1'
H-1''
H-2'
H-3'
H-4'
HOESY via H-5
HOESY via H-1''
Transient via H-5
Transient via H-1''
Crystal Structure
Figure 3: Bar graph derived from the data in Table 1. For each proton in the cation, the derived
“distance” to the anion is plotted, based on the two different NOE NMR experiments (with two
different modes of internal normalization) plus data taken from the crystal structure.
This article is protected by copyright. All rights reserved.

Graphical Table of Contents
Determination of Inter- and Intra-ionic Interactions in a Monofluorinated Imidazolium
Ionic Liquid by a Combination of X-Ray Crystallography and NOE NMR Spectroscopy
Yves Lingscheid, Mathias Paul, Andreas Bröhl, Jörg-Martin Neudörfl,* and Ralf Giernoth*
A Transient 1D 1H{19F} NOE NMR experiment has been applied to a monofluorinated ionic
liquid to determine inter-ionic distances. The results have been compared and validated with
classical 2D HOESY experiments and the x-ray crystal structure.
This article is protected by copyright. All rights reserved.