1H and 17O NMR study of H-bond dynamics in picolinic acid N-oxide solutions in acetonitrile-h3 and acetonitrile?d3: Novel aspects of old casus

Article abstract of DOI:10.1016/j.chemphys.2018.06.020

Despite apparent similarity, two isotopomers of acetonitrile (ACN-h₃ and ACN-d₃) show unexpected differences in their structural and dynamic properties. These differences cannot be fully understood within the usual isotope effect and therefore called as unresolved casus. Sensing the solvent properties, the H-bond dynamics of picolinic acid N-oxide (PANO) in ACN was studied using ¹H and ¹⁷O NMR. Several overlapping factors, like the presence of micro-traces of water, liquid-liquid equilibrium, effect of ionic ingredient, etc., which may cause the observed anomalies in the dependences of PANO chemical shift and signal width on temperature, were analyzed. The deuterium isotope effect on the phase separation in organic solute/water mixtures was deduced to be the crucial factor inducing those surprising differences between ACN-h₃ and ACN-d₃. Thus the casus could be resolved within the second-order phase transition model, in which the vibrational contributions to the intermolecular interaction energies would be properly taken into account.

Full text of DOI:10.1016/j.chemphys.2018.06.020

Accepted Manuscript
1H and 17O NMR study of H-bond dynamics in picolinic acid N-oxide solutions
in acetonitrile-h 3 and acetonitrile-d 3: novel aspects of old casus
Arūnas Maršalka, Laurynas Dagys, Virginija Jakubkienė, Sigitas Tumkevičius,
Vytautas Balevicius
PII:
S0301-0104(18)30356-2
DOI:
https://doi.org/10.1016/j.chemphys.2018.06.020
CHEMPH 10061
Reference:
Chemical Physics
To appear in:
Received Date:
Revised Date:
Accepted Date:
15 April 2018
25 June 2018
26 June 2018
Please cite this article as: A. Maršalka, L. Dagys, V. Jakubkienė, S. Tumkevičius, V. Balevicius, 1H and 17O NMR
study of H-bond dynamics in picolinic acid N-oxide solutions in acetonitrile-h 3 and acetonitrile-d 3: novel aspects
of old casus, Chemical Physics (2018), doi: https://doi.org/10.1016/j.chemphys.2018.06.020
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

17
1H and O NMR study of H-bond dynamics in picolinic acid N-oxide solutions
in acetonitrile-h3 and acetonitrile-d3: novel aspects of old casus
Arūnas Maršalka1, Laurynas Dagys1, Virginija Jakubkienė2, Sigitas Tumkevičius2, Vytautas
Balevicius1*
1Institute of Chemical Physics, Vilnius University, Sauletekio av. 3, LT-10257 Vilnius, Lithuania
2Institute of Chemistry, Vilnius University, Naugarduko str. 24, LT-03225 Vilnius, Lithuania
Abstract
Despite apparent similarity, two isotopomers of acetonitrile (ACN-h3 and ACN-d3) show
unexpected differences in their structural and dynamic properties. These differences cannot be fully
understood within the usual isotope effect and therefore called as unresolved casus. Sensing the
solvent properties, the H-bond dynamics of picolinic acid N-oxide (PANO) in ACN was studied
1
17
using H and O NMR. Several overlapping factors, like the presence of micro-traces of water,
liquid-liquid equilibrium, effect of ionic ingredient, etc., which may cause the observed anomalies
in the dependences of PANO chemical shift and signal width on temperature, were analyzed. The
deuterium isotope effect on the phase separation in organic solute/water mixtures was deduced to be
the crucial factor inducing those surprising differences between ACN-h3 and ACN-d3. Thus the
casus could be resolved within the second-order phase transition model, in which the vibrational
contributions to the intermolecular interaction energies would be properly taken into account.
Keywords: NMR spectra; hydrogen bond; isotopic effects; liquid-liquid phase separation; phase
transitions
*Corresponding author. E-mail address: vytautas.balevicius@ff.vu.lt (V. Balevicius).
1

1. Introduction
Acetonitrile (CH3CN, ACN) is a widely used solvent or co-solvent, in which many hydrophobic,
hydrophilic and ionic materials can be dissolved [1]. ACN-water mixtures exhibit anomalous
thermodynamic properties and complex structures with the regimes of microheterogeneity
dependent on concentration and temperature ([1] and Refs cited therein). In separation science
(liquid - liquid and solid - phase extraction) ACN plays a key role as an extraction medium [2].
Therefore, the molecular systems containing ACN as one of the components have been widely
studied with particular emphasis on the structure, intermolecular interactions, vibrational and
orientational dynamics.
Two isotopomers of acetonitrile (ACN-h3 and ACN-d3) were studied in a series of
works using IR, Raman and depolarized Rayleigh scattering in order to elucidate their dynamic
behavior and determine their local structural organization [3-6]. The unusual behavior was revealed
analyzing the IR and Raman spectra of C–H/C–D symmetric stretching and CH3/CD3 symmetric
bending vibrations of the pure ACN, equimolar isotopic mixtures and equimolarly diluted in CCl4.
The general conclusion emerging from those works was that despite their apparent similarity, ACN-
h3 and ACN-d3 show unusual and surprising differences in their structural and dynamic properties,
which are not fully understood within the usual isotope effect.
Another group of works deal with electronic processes, like charge transfer reactions
[7], resonance Raman scattering and fluorescence of dye molecules [8-10], which occur in the
solutions of ACN-h3 and ACN-d3. No effects from solvent isotope substitution on the reaction rate
and reaction yield have been found in the case of the intramolecular charge transfer in 4-(1-
azetidinyl) benzonitrile [7]. However, the Raman cross sections of betaine-30 were found in ACN-
d3 to be lower than those in ACN-h3, and different solvent reorganization energies were determined
[8,9]. Probably the most interesting results have been obtained in Ref. 10 studying the solvation
dynamics of dye molecules in methanol and ACN. The effects of solvent deuterium substitution on
2

the fluorescence spectra and the time domain data were found to be significant in both solvents. The
ultrafast components (200–260 fs) were found to be not sensitive to the deuterium substitution.
Beside the components in ultrafast (fs and ps) solvation time scale, a large amplitude component in
the scale of nanoseconds was observed in both solvents. The amplitudes of this component are
substantially different in isotopomers, showing the opposite trend in ACN and methanol [10].
In summary, the different experimental trends shown by the normal and deuterated
solvents can be not always explained by a regular dynamic isotopic effect, i.e. their kinetic/transport
coefficients to be proportional to the square root of the ratio between the isotopomer masses [3]. In
Ref. 6 this was called as an unresolved casus. In the current work we present a complementary
NMR observation of non-trivial effect of isotopomers of acetonitrile, as reaction media, on the
intramolecular hydrogen bond (H-bond) dynamics in picolinic acid N-oxide. The obtained results
can be understood accepting that the isotopic substitution of solvent molecules drastically influents
the micro-structuring and the local fluctuation processes culminating with liquid-liquid phase
transition (phase separation) therein.
2. Experimental
Commercial picolinic acid N-oxide (PANO), 97 %, from Sigma-Aldrich was dried
keeping for 2 days under vacuum at  380 K. Acetonitrile (ACN-h3), anhydrous, 99.8 % (I) and
deuterated acetonitrile (ACN-d3),  99.8 atom % D (II), both from Sigma-Aldrich and ACN-h3, LC-
MS,  99.9 % (III) from Fluka have been used as solvents preparing the samples on two different
levels: a) the preparation under ambient conditions – i. e. without using of vacuum techniques,
molecular sieves and without keeping a sample closed in a dry environment during all stages of
preparation (weighting of components, tube filling, etc.); b) the high-level preparation - all stages
3

run in a glove-box filled with extremely dry nitrogen (humidity  0.1 ppm) and in some cases
keeping ACN over the molecular sieves (3 Å) for several days before.
The 4-nitro derivative of PANO (4-nitro-PANO) was synthesized in the Department
of Organic Chemistry of Vilnius University by nitration of PANO using a slightly modified
procedure than that reported in [11]. A mixture of 2-picolinic acid N-oxide (1 g, 6.36 mmol), conc.
H2SO4 (3 mL), and conc. HNO3 (2.2 mL, d = 1.5 g/cm3) was heated at 408 K under stirring for 3 hr.
After cooling to room temperature, the reaction mixture was poured to a crushed ice. The formed
precipitate was filtered off, washed with water till pH = 7 and recrystallized from ethanol to give
0.4 g (31%) of 4-nitropicolinic acid N-oxide, m.p. 422 – 423 K, Ref. 11: m.p. 421 K. The solutions
were prepared by weighting ( 0.1 mg) the components.
17
1H and O NMR experiments were carried out on a BRUKER AVANCE III 400
1
17
WB NMR spectrometer operating at 400 and 54 MHz for H and O, respectively, using BBO
probe-head. The variable-temperature measurements in the range of 230 K - 350 K were
controlled with the accuracy of  0.5 K. The signal of TSPSA (3-(trimethylsilyl)propane sulfonic
1
acid sodium salt) in ACN-d3 solution in capillary insert was used as the reference for H NMR
spectra and for lock. The D2O in the capillary insert was used for reference in 17O NMR spectra.
In order to get better signal-to-noise ratio the natural abundance 17O spectra were recorded using 10
mm tubes. The /2 pulses of 22.5 s at the transmitter power of 70 W were applied using the
acquisition time of 0.02 s and the recycle delay of 0.4 s. The number of scans was 400000. When
1
13
necessary, H and C chemical shifts of PANO were confirmed by the analysis of 2D homo-
1
and hetero-nuclear NMR experiments (1H–1H COSY, 1H–13C HSQC and H–13C HMBC). NMR
spectra were processed using Topsin 3.2 software. Additionally, the signal shapes were processed
using Microcal Origin 9 package.
4

3. Results and discussion
The NMR experiments have been performed in wide temperature range approaching
the melting (M.P.) to boiling (B.P.) points of ACN (227 – 232 K and 354 K, respectively [2,12]).
This was because the warning given in [6] that no anomalies could be evidenced in the isotopic
dilution experiments if experiments are carried out at the fixed temperature.
PANO was chosen for the study because of several reasons. The H-bond complexes
between carboxylic acids and pyridine N-oxide or its derivatives are promising benchmark systems
for short O–H…O hydrogen bonding [13]. These systems exhibit in most cases asymmetric and flat
single well proton potentials, which result in large-amplitude proton dynamics that often culminates
with occurrences of proton transfer (PT). PANO is maybe the most prominent member of this
family [13]. Quantum mechanical simulation of the proton dynamics by Car-Parrinello method
shows that the H-bond proton in crystalline PANO is located about 99% of time in the energy
minimum near the carboxylic oxygen (O1) and the jumps to the N–O(3) acceptor are rare (for atom
numbering see Fig. 1) [14]. However, the H-bond in PANO in liquid solutions looks to be more soft
and flexible. Even the solvent effect is clearly pronounced on the proton potential functions at
increasing polarity of the solvents [15]. Thus, a large amplitude motion of the H-bond proton in
PANO, including PT process, could be expected in ACN solutions. And indeed, the 17O NMR
spectrum of PANO dissolved in ACN-h3 confirms that the fast (in the sense of NMR time scale, i.e.
 10–5 -10–6 s) intramolecular H-bond proton dynamics coupled with the interchange of O1 and O2
oxygen atoms occurs in this system (Fig. 1). The probable scheme for such interchange is shown in
Fig. 1. It is the intramolecular process and could be realized if PT from O1 to O3 atoms occurs. The
intermolecular exchange could run too, however, if PANO molecules formed the cyclic dimers or
even higher H-bond aggregates. In order to confirm or to deny the possibility of the intermolecular
1
route, PANO was mixed with one of its derivatives - 4-nitro-PANO, and H NMR spectra were
registered.
5

17
Fig. 1. Natural abundance O NMR spectrum of PANO dissolved in ACN-h3 confirming a fast
intramolecular H-bond proton dynamics with sequent interchange of O1 and O2 oxygen atoms (0.3
M, T = 293 K).
Because of the limited solubility of PANO and its derivatives in ACN only  0.3 M
1
concentration was achieved. Two separated signals of H-bond protons are seen in the H NMR
spectrum of PANO mixed with 4-nitro-PANO and dissolved in ACN-h3 (Fig. 2). This means that at
such relatively high dilution the intermolecular H-bond exchange does not occur at all or it is very
slow. Hence, it can be concluded that the intermolecular contribution can be neglected, and only the
intramolecular H-bond dynamics as well as its coupling with the solvent has to be considered.
6

1
Fig. 2. H NMR spectrum of PANO + 4-nitro-PANO (R = NO2) dissolved in ACN-h3 as the
evidence of slow intermolecular H-bond proton exchange (0.3 M, T = 293 K).
The proton transfer from the donor (O1) to the acceptor (O3) has to be considered in
this case. The driving force for PT is the electric field created by the electric dipoles of solvent
molecules at the H-bond site [16]. Therefore, a crucial role of dielectric media in PT processes is
stressed almost in all studies on this topic. The strength of the proton driving field depends on the
ordering of individual dipoles of solvent molecules. The questions concerning the extent of this
dipole ordering, like - is it either an overall or structuring effect that covers only a part of solvent
molecules in the closest surrounding of H-bond complex, and the rate of solvent reorganization are
still open. Not only dipoles of solvent molecules are important in PT processes. The concept of
solvatomers [17-19] can be also considered explaining the changes in the proton position in the
intramolecular H-bond and relating it to the solvation pattern around the oxygen atoms. According
to this concept, a solution is disorganized, with one of the sites instantaneously solvated better than
the other [18], leading to the presence of equilibrating solvatomers (isomers, stereoisomers, or
7

tautomers that differ in solvation) [19]. Thus, any changes in the dynamics of solvent molecules
could then modify the rate and depth of PT in the solute molecules.
1
It is quite logical to expect that the H chemical shift of the bridge proton can be used
as the most effective diagnostic tool sensing the solvent properties via the H-bond dynamics. This is
because the PT path and the motion of solvent molecules are coupled. Thus the rates of solvent
molecular reorganization are very important for PT reactions. Therefore, the temperature
dependences of 1H NMR parameters (chemical shift and width) of –COOH proton in PANO in both
isotopomers (ACN-h3 and ACN-d3) were studied very precisely covering the whole liquid phase
from ACN melting to boiling points. The results obtained using the samples prepared under ambient
conditions (see Experimental) are presented in Fig. 3.
No isotopic differences are seen on the dependences of chemical shift and signal width
in the temperature range up to  292 K (Fig. 3). The dependencies of chemical shift are linear,
characterized by the same slope (0.0062  0.0001 ppm/deg) in both solvents. The picture is
drastically changing at temperatures above 292 K. The changes in the ACN-h3 solution look like a
second-order liquid-liquid phase transition occurred. The linear dependence of chemical shift
transforms to the more complex one, whereas no changes in temperature regime is seen in ACN-d3
(Fig. 3a). The differences between ACN-h3 and ACN-d3 become more obvious looking at the
dependences of signal width on temperature (Fig. 3b). These dependences remind those formerly
observed in IR and Raman experiments on pure ACN-h3 and ACN-d3 together with isotopic and
chemical equimolar mixtures, where this observation was called as an unresolved casus [6]. There
was also claimed that a qualitative interpretation is possible, if it is accepted that the isotopic
substitution can substantially modify the local structural organization of these molecular liquids.
8

Fig. 3. Temperature dependences of 1H chemical shift (a) and signal width (b) of –COOH proton of
PANO in solutions prepared under ambient conditions in ACN-h3 and ACN-d3, respectively.
The results obtained in the present work using NMR spectroscopy reveal some new
aspects of this problem in more details. Several overlapping factors that can cause the observed
anomalies in ACN-h3 and ACN-d3 solutions can to be noted:
i) The presence of traces of water in the samples prepared under ambient and high-level conditions.
9

Water molecules are always present in solutions in certain amounts in the samples prepared under
ambient conditions. Note, this level of sample preparation is often met in many works done using
‘fast’ methods of spectroscopy (FTIR, Raman). According to a wide spread opinion, the presence of
water in micro-amounts can be neglected as the vibrational bands of water can be easily recognized,
separated from others and analysed. Often this is true. Therefore FTIR spectroscopy was applied in
the present work in order to check and to compare the water content in ACN-h3 and ACN-d3. As it
follows from the FTIR spectra (Fig. 4), the amount of water was practically the same as in ACN-h3
as well as in ACN-d3.
Fig. 4. FTIR ATR spectra of ACN-h3 (III) and ACN-d3 (II) under ambient conditions. The spectra
were recorded on BRUKER-ALPHA spectrometer.
The situation differs when NMR methods are used. Because NMR is ‘slow’
spectroscopy and the averaging of signals of nuclei involved in fast dynamic processes occurs, the
NMR users always sought to minimize in their works the amount of water. It is hard to recognize,
however, some micro traces of water have survival in the samples prepared under ambient
10

conditions and studied in the present work. The corresponding signals were detected in the 1H NMR
1
spectra at very large zooming of intensities (Fig. 5). It is worthy to note that H NMR spectroscopy
looks to be the unique method that reveals very fine differences in water structuring/aggregation in
ACN-h3 and ACN-d3 solutions. These differences can be identified analyzing the signal shapes in
the range of 2 - 6 ppm (Fig. 5) and some peaks can be attributed to water in various states of
aggregation.
Fig. 5. The traces of water observed at 2 - 6 ppm in the 1H NMR spectra of PANO in ACN-h3 (III)
and ACN-d3 (II) solutions. The evolution of signal shapes in both solvents is shown separately in
two inserts at increasing temperature (240, 273, 323 and 345 K).
11

The effect of high-level sample preparation is shown in Fig. 6. In order additionally to
reduce the presence of water that may come to solutions together with PANO, despite it was dried
under vacuum (see Experimental), its concentration was reduced to 0.03 M. The convergence of
two regimes of H-bond dynamics in PANO in both solvents (ACN-h3 and ACN-d3) can be stated by
reducing of the amount of water (Fig.6).
Fig. 6. The effect of high-level sample preparation on the temperature dependences of signal width
of –COOH proton of PANO. The samples prepared under ambient conditions: 1, 2 – solutions of
0.3 M of PANO in ACN-d3 (II) and ACN-h3 (III), respectively; the samples prepared in the glove-
box: 3 – 0.03 M of PANO in ACN-d3 (II), 4 – in ACN-h3 (III) kept over molecular sieves, 5 - in
anhydrous ACN-h3 (I) and no molecular sieves. More details are given in Experimental.
ii) Liquid-liquid equilibrium in ACN/water solutions.
It is well-known [20-22] that the aqueous ACN solution exhibits the second-order phase transition
characterized by the phase diagram with the upper critical solution temperature (UCST) at  272 K
and at the concentration 0.36 - 0.38 mole fraction of ACN (Fig. 7a). The microscopic and
12

mesoscopic mixing scheme changes drastically, dependent on concentration and temperature,
especially in the neighborhood of the critical point. The small-angle X-ray scattering experiments
confirm that indeed the fluctuations of concentration became very large approaching to UCST [20]
(see the dashed area in Fig. 7a.).
Fig. 7. The possible scenario of the evolution of phase diagram (T vs XACN (concentration in mole
fraction)) of aqueous ACN-h3 and ACN-d3 solutions upon ionic and isotopic effects. The data taken
from [21,22] were used illustrating the changes of phase diagrams a  b and a  c. The critical
regions of large fluctuations are shown by the dashed areas. More comments in text.
13

The UCST in ACN/water solution looks to be rather far away from  292 K, i.e. the
1
point where the onset of anomalies was observed in H NMR spectra in the present work (Fig. 3).
However, it is also known that even very small amount of ionic solute can drastically influent the
phase equilibrium shifting the UCST to much higher temperatures [22,23].
iii) Effect of ionic solute.
Binary mixtures of water with added organic components like ACN or tetrahydrofuran (THF) are
often used in modern bio- and nano-technological applications. Usually ACN and THF solutions in
H2O or D2O are accepted to be homogeneous at room temperature for any molar ratio of the
components. However, dissolved ions, even a miserly amount, can change their phase diagrams
very significantly [22,23]. Namely, the phase separation temperature (or UCST) in ACN/water
mixtures can be shifted upon dissolving of ions (K+, Cl–, KPS, etc.) from 272 K to 280 K or even to
higher values, particularly, if water is present in small amount [22]. The ionic effect on the phase
diagram (liquid - liquid coexistence curve) is shown in Fig. 7b. Furthermore, as it was noted in the
studies of very strong H-bond between pyridine N-oxide (PyO) and HCl in ACN solution, the H-
bond systems with PT can act as an ionic solute [24]. It was observed that very low solubility of the
PyO.HCl co-crystal in ACN (ca. 0.1 M at the room temperature, i.e. even lower than PANO in the
present case) can be improved by adding of some amount of H2O or D2O. However, a rather
inhomogeneous composition of ACN/H2O/PyO/HCl was obtained in this case. The presence of
micro inhomogeneities in the sample is not always noticeable visually. However, even a small
perturbation of temperature renders the sample turbid, and consequently the phase separation is
induced. The ionic structures PyO.H+...Cl– that appear thanks to PT process are acting as ionic
ingredients. It would be logically to assume, that PANO molecules could also play a role of ionic
ingredients while the virtual ionic pairs appear during the fast intramolecular PT (Fig. 1). The time
14

scale of the oscillation of ionic forms in PT process in the range of  10–5 - 10–6 s, as it that follows
from the 17O NMR spectrum (Fig. 1), should be noted particularly. This seems to be a very
interesting parallel with the intramolecular charge transfer and fluorescence results obtained
studying the solvation dynamics of dye molecules in ACN solutions [10]. There was deduced that
the nanosecond-component in both ACN isotopomers were substantially different, however, no
effect in ultrafast (fs and ps) solvation time scale.
iv) Deuterium isotope effects on phase separation in organic solute/water mixtures.
According to the FTIR data (Fig. 4), the amount of traces of water was practically the same in both
solvents (ACN-h3 and ACN-d3). Therefore, the above factors (i - iii) solely cannot explain the
differences in the observed isotopic effects (Fig. 3). However note, the deuteration of organic solute
in aqueous solutions leads sometimes to rather unexpected phenomena. A very revealing example
of such behavior is met in the case of THF/water solutions. The solutions of THF-h8 in D2O and
H2O possess the so-called 'closed-loop' phase diagram possessing the lower and the upper critical
solution temperatures (LSCT and UCST; 336.9 K and 416.7 K in D2O, respectively) [23]. It is very
surprising, but the closed-loop phase diagram disappears in THF-d8 aqueous solutions. THF-d8 was
found to be perfectly miscible with water, either H2O or D2O, for any molar ratios. The phase
separation in these mixtures can be induced by dissolving of salts. Depending on the concentration
of ions, the liquid - liquid phase separation can appear even at the room temperature. The play with
solute deuteration can be continued - the miscibility gape appears again in the THF-h4d4/water
solutions, however, it is narrower than in the case of THF-h8 [23].
Unfortunately, the complete phase diagram for the ACN-d3/water solutions, to the best
of our knowledge, was not determined or published. Only the increase of UCST by + 0.6 K upon
deuteration of ACN and about + 6 K upon replacement of H2O by D2O was reported [21]. This
situation is symbolically pictured in Fig. 7c. The total effect of all of these contributions (i-iv) is
15

presented in Fig. 7d. The region of critical mixing/demixing regimes (the dashed area in Fig. 7d)
can be pushed then to the conditions actual for the present work ( 292 K, at the micro-amount of
water, i.e. XACN  1). Until the precise phase diagrams for ACN-d3/ACN-h3/water/ions are not
determined for a wide range of concentration, the drawing in Fig. 7d has to be treated as the
hypothesis or as a directing-post for the future experiments. However, a such scenario is at a certain
level supported by the NMR experiments carried out in this work (Fig. 5) as well as by the things
known for the related systems (refs 21-24).
Concluding remarks
The obtained results allow to conclude that the latter factor, viz. - the deuterium
isotope effects on the phase separation in organic solute/water mixtures should be considered as the
most crucial. It causes the observed differences in the properties of ACN-h3 and ACN-d3, as
solvents, called in the literature as anomalies or casus. The liquid-liquid phase separation in its
physical matter is the second-order phase transition. Therefore, rigorously the casus with ACN-h3
and ACN-d3 can be explained constructing the proper second-order phase transition
model/treatment, in which the vibrational contributions of different isotopes to the intermolecular
interaction energies would be properly taken into account.
As ACN is widely used as a solvent or co-solvent in many kinetic and spectroscopic
experiments, this is very revealing example illustrating how the traces of water, even in the micro-
amount, can modify the properties of solvents and thus can distort the experimental outcomes.
1H NMR spectroscopy is a very proper tool that can be effectively used sensing does
the micro-structuring or phase separation occur in the chosen solvent at all. Very fine details of
micro-heterogeneity, clustering or confinement of water molecules can be recognized applying this
16

method (Fig. 5). Then, looking ahead, this technique can be applied monitoring the liquid-liquid
phase separation at the conditions where other experimental methods traditionally used determining
the phase coexistence curves (light scattering, viscosity, refractive index, etc.) appear to be not
capable.
Acknowledgement
The authors acknowledge Center of Spectroscopic Characterization of Materials and
Electronic/Molecular
Processes
(Scientific
infrastructure
"SPECTROVERSUM"
www.spectroversum.ff.vu.lt) at Lithuanian National Center for Physical Sciences and Technology
for use of spectroscopic equipment. We thank Dr. Kristijonas Genevičius for the assistance during
sample preparation in glove-box and vacuum technique.
References
[1] J.R. Reimers, L.E. Hall, The Solvation of Acetonitrile, J. Am. Chem. Soc. 121 (1999) 3730–
3744.
[2] P.K. Zarzycki, M.B. Zarzycka, M.M. Ślączka, V.L. Clifton, Acetonitrile, the polarity
chameleon, Anal. Bioanal. Chem. 397 (2010) 905–908.
[3] P. Sassi, A. Morresi, G. Paliani, R.S. Cataliotti, Reorientational relaxation in a simple polar
liquid: CD3CN. A comparison with light acetonitrile, J. Phys.: Condens. Matter 12, (2000) 3615–
3623.
[4] E. Marri, A. Morresi, G. Paliani, R.S. Cataliotti, M.G. Giorgini, Isotopic and chemical dilution
effects on the vibrational relaxation rate of some totally symmetric motions of liquid acetonitrile,
Chem. Phys. 243 (1999) 323–332.
17

[5] A. Morresi, P. Sassi, M. Paolantoni, S. Santini, R.S. Cataliotti, Solvent effect on the vibrational
dephasing of the 2(CN) and 4(CC) stretching modes in liquid acetonitrile and acetonitrile-d3,
Chem. Phys. 254 (2000) 337–347.
[6] A. Morresi, M. Ombelli, P. Sassi, S. Santini, Light and deuterated acetonitrile: an unresolved
casus? J. Raman Spectr. 33 (2002) 71–79.
[7] T. Pradhan, P. Ghoshal, R. Biswas, Excited state intramolecular charge transfer reaction in 4-(1-
azetidinyl) benzonitrile: Solvent isotope effects, J. Chem. Sci. 121 (2009) 95–101.
[8] Y. Zong, J.L. Mc Hale, Raman study of solvent dynamics in electron transfer. I. Betaine-30 in
CH3CN and CD3CN, J. Chem. Phys. 106 (1997) 4963–4972.
[9] Y. Zong, J.L. Mc Hale, Erratum: ‘‘Resonance Raman study of solvent dynamics in electron
transfer. I. Betaine-30 in CH3CN and CD3CN’’, J. Chem. Phys. 120 (2004) 11333.
[10] S.H. Lee, J.H. Lee, T. Joo, Deuterium isotope effect on the solvation dynamics of a dye
molecule in methanol and acetonitrile, J. Chem. Phys. 110 (1999) 10969–10976.
[11] E. Profft, W. Steinke, N-Oxyde substituierter Picolinsäuren, J. prakt. Chem. 13 (1961) 58–75.
[12] C. Reichardt, T. Welton, Solvents and Solvent Effects in Organic Chemistry (Wiley-VCH
Verlag GmbH & Co. KGaA, Weinheim, 2011) p. 550.
[13] J. Stare, M. Hartl, L. Daemen, J. Eckert, The Very Short Hydrogen Bond in the Pyridine N-
Oxide – Trichloroacetic Acid Complex: An Inelastic Neutron Scattering and Computational Study,
Acta Chim. Slov. 58 (2011) 521–527.
[14] J. Stare, J. Panek, J. Eckert, J. Grdadolnik, J. Mavri, D. Hadži, Proton Dynamics in the Strong
Chelate Hydrogen Bond of Crystalline Picolinic Acid N-Oxide. A New Computational Approach
and Infrared, Raman and INS Study, J. Phys. Chem. A 112 (2008) 1576–1586.
[15] J. Stare, J. Mavri, G. Ambrožič, D. Hadži, Strong intramolecular hydrogen bonds. Part I.
Vibrational frequencies of the OH group in some picolinic acid N-oxides predicted from DFT
18

calculated potentials and located in the infrared spectra, J. Mol. Struct. (Theochem), 500 (2000)
429–440.
[16] N.S. Golubev, I.G. Shenderovich, S.N. Smirnov, G.S. Denisov, H.H. Limbach, Nuclear Scalar
Spin-Spin Coupling Reveals Novel Properties of Low-Barier Hydrogen Bonds in a Polar
Environment, Chem. Eur. J. 5 (1999) 492–497.
[17] P. Dopieralski, C.L. Perrin, Z. Latajka, On the Intramolecular Hydrogen Bond in Solution:
Car–Parrinello and Path Integral Molecular Dynamics Perspective, J. Chem. Theory Comput. 7
(2011) 3505–3513.
[18] C.L. Perrin, B.K. Ohta, Symmetry of NHN hydrogen bonds in solution, J. Mol. Sruct. 644
(2003) 1–12.
[19] C.L. Perrin, J.S. Lau, Hydrogen-Bond Symmetry in Zwitterionic Phthalate Anions: Symmetry
Breaking by Solvation, J.Am. Chem. Soc. 128 (2006) 11820–11824.
[20] K. Nishikawa, Y. Kasahara, T. Ichioka, Inhomogeneity of Mixing in Acetonitrile Aqueous
Solution Studied by Small-Angle X-ray Scattering, J. Phys. Chem. B 106 (2002) 693–700.
[21] M. Hürth, D. Woermann, Deuterium Isotope Effects in Binary Critical Mixtures Study of the
Systems Methanol/Isooctane and Acetonitrile/Water, Ber. Bunsenges. Phys. Chem. 91 (1987) 614–
616.
[22] D. Pfenning, D. Woermann, Modification of the transport properties of a membrane by
selective solvation: Hyperfiltration of acetonitrile/water mixtures, J. Membrane Sc. 32 (1987) 105–
116.
[23] V. Balevicius , N. Weiden, A. Weiss, Deuteron NMR relaxation, phase diagrams and isotope
effects in liquid mixtures of tetrahydrofuran/D2O/ salt, Ber. Bunsenges. Phys. Chem. 98 (1994)
785–792.
19

[24] V. Balevicius, K. Aidas, I. Svoboda, H. Fuess, Hydrogen bonding in pyridine N-oxide/acid
systems: proton transfer and fine details revealed by FTIR, NMR and X-ray diffraction, J. Phys.
Chem. A 116 (2012) 8753–8761.
20

Graphic abstract
Highlights
Non-trivial effect of isotopomers of acetonitrile, as reaction media, on H-bond dynamics was
deduced
Traces of water, even in the micro-amount, substantially modify the structuring of acetonitrile
Isotopic substitution CH3 CD3 influents the micro-structuring and local fluctuations in solvent
Structuring can distort the experimental outcomes of kinetics that occur in micro- to nanoseconds
scales
21