Photochemistry of imidazolium cations. Water addition to methylimidazolium ring induced by UV radiation in aqueous solution

Article abstract of DOI:10.1016/j.jphotochem.2015.07.018

The UV-C induced photoaddition of water to N-alkyl-N′-methylimidazolium cations was studied. The main photoreaction products exhibit chemical additions of a proton and a hydroxyl group to either positions 4 or 5 of the imidazolium ring. For unsymmetrical imidazolium cations, two positional isomers were obtained as products. In these cases, the most abundant isomer is the one in which the hydroxyl group adds at the side of the ring having the longer alkyl substituent. Experiments performed in D₂O solutions reveal that the additions of proton and hydroxyl group never take place at the same carbon atom, in a reaction that produces equal amounts of diastereoisomers. Moreover, the formation of diastereoisomers at equal proportions suggests that the reaction proceeds in an unconcerted fashion.

Full text of DOI:10.1016/j.jphotochem.2015.07.018

Journal of Photochemistry and Photobiology A: Chemistry 314 (2016) 155–163
Contents lists available at ScienceDirect
Journal of Photochemistry and Photobiology A:
Chemistry
journal homepage: www.elsevier.com/locate/jphotochem
Photochemistry of imidazolium cations. Water addition to
methylimidazolium ring induced by UV radiation in aqueous solution
Gabriela P. Sarmientoa, Andrés Zelcerb,c, Mariela S. Espinosaa, Paola A. Babaya,
Martin Mirendaa,b,d,
*
a
Gerencia Química, Centro Atómico Constituyentes, Comisión Nacional de Energía Atómica, Av. Gral. Paz 1499, B1650KNA San Martín, Buenos Aires,
Argentina
b
CONICET, Av. Rivadavia 1917, C1033AAJ Buenos Aires, Argentina
ECyT-UNSAM, 25 de Mayo y Francia, C.P.: 1650 San Martín, Buenos Aires, Argentina
Departamento de Química Inorgánica, Analítica y Química Física, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Ciudad
c
d
Universitaria, Pab. II, C1428EHA, Buenos Aires, Argentina
A R T I C L E I N F O
A B S T R A C T
The UV-C induced photoaddition of water to N-alkyl-N0-methylimidazolium cations was studied. The
main photoreaction products exhibit chemical additions of a proton and a hydroxyl group to either
positions 4 or 5 of the imidazolium ring. For unsymmetrical imidazolium cations, two positional isomers
were obtained as products. In these cases, the most abundant isomer is the one in which the hydroxyl
group adds at the side of the ring having the longer alkyl substituent. Experiments performed in D2O
solutions reveal that the additions of proton and hydroxyl group never take place at the same carbon
atom, in a reaction that produces equal amounts of diastereoisomers. Moreover, the formation of
diastereoisomers at equal proportions suggests that the reaction proceeds in an unconcerted fashion.
ã 2015 Elsevier B.V. All rights reserved.
Article history:
Received 22 April 2015
Received in revised form 13 July 2015
Accepted 30 July 2015
Available online 6 August 2015
Keywords:
Imidazolium
Ionic liquid
Photoaddition
UV-C reaction
Water addition
1. Introduction
The degradation methods proposed in the literature include; (a)
chemical degradation by ozonation and Fenton and Fenton like
Methylimidazolium cations are one of the most common
constituents of ionic liquids (ILs), a newfangled kind of organic
salts that melt near or below 100 ꢀC [1–4]. The distinctive
physicochemical properties of these compounds—low flammabil-
ity, tunable chemical solubility, low vapor pressure and high
thermal stability, among others—make them interesting alter-
natives to common volatile organic solvents in many areas such as
catalysis [5,6], organic synthesis [7,8], nanotechnology [9,10], and
nuclear technology [11,12].
ILs were assiduously proposed as solvents for applications in
green technologies since their negligible vapor pressure avoids air
pollution due to evaporation and volatilization processes. Howev-
er, most of them have a significant solubility in water, the most
likely medium through which they would be eventually released
into the environment [13]. In recent years, some research groups
have succeeded in analyzing different mechanisms of degradation
of methylimidazolium cations in water, since the latter are not
normally considered as readily biodegradable compounds [14].
reactions, (b) electrochemical degradation and, (c) photochemical
degradation by UV, UV/TiO2, UV/H2O2 and ultrasound/H2O2/
CH3COOH [15]. In many cases, not only is the efficiency of
degradation process the important issue. The chemical nature of
the degradation products is also relevant in order to evaluate if
these compounds are environmentally friendlier than their
precursors. Siedlecka et al. [16] and more recently Muñoz et al.
[17] showed that the main degradation products of methylimida-
zolium cations in aqueous solutions under Fenton oxidation are
imidazolones. On the other hand, Czerwicka et al. [18] showed that
the photochemical treatment of methylimidazolium cations in
solution with UV/H2O2 yields main degradation products with two
hydroxyl groups at the imidazolium ring. While different
degradation processes may lead to a variety of products, all of
them have in common the oxidation, at least to some degree, of the
imidazolium ring. The incorporation of oxygen atoms in the form
of hydroxyl, aldehyde or carboxylic groups is interesting, since
these functional groups represent possible sites for attack by
oxygenases [19]. It is important to note that, although ten years ago
Stepnowski and Zaleska [20] reported a pioneering work on the
degradation of imidazolium ILs under UV light, they failed to
inform the chemical products obtained in that process.
* Corresponding author.
E-mail address: mirenda@cnea.gov.ar (M. Mirenda).
http://dx.doi.org/10.1016/j.jphotochem.2015.07.018
1010-6030/ã 2015 Elsevier B.V. All rights reserved.

156
G.P. Sarmiento et al. / Journal of Photochemistry and Photobiology A: Chemistry 314 (2016) 155–163
The subject of the interaction of methylimidazolium com-
The mixture was then kept at room temperature and maintained
under stirring for 6 days. The excess of iodomethane was removed
by distillation and captured in an ammonium hydroxide solution.
After distillation, a viscous yellow liquid of MMIMI was obtained. In
order to replace iodide by chloride anions, MMIMI was passed
through an ion exchange resin (Aldrich, Dorwex 1 ꢁ4 chloride
form). The complete exchange of iodide anion was confirmed by
the absence of the UV signal at 226 nm on eluted aliquots.
The excess of water was removed by evaporation under reduced
pounds with UV radiation itself also constitutes a relevant topic in
photochemistry. These compounds are used as solvents in photo-
catalysis [21], photoisomerizations [22], photoinduced charge-
transfer processes [23], and in dye sensitized solar cells [24,25],
since they are optically transparent in a sizeable fraction of the
UV–vis spectrum [26,27]. Although the photophysical character-
istics of the methylimidazolium ionic liquids (ImILs) have been
analyzed in several publications [28–30], none of them have
focused on the photochemical effects produced by the irradiation
of these compounds in the UV-C region, where they absorb.
The present work is focuses on the photochemical reactions
that occur when methylimidazolium cations are exposed to UV
radiation in aqueous medium. The experiments were performed
on three different N-alkyl-N0-methylimidazolium chlorides, vary-
ing the lengths of the alkyl sustituent chains of the reactants in a
systematic way. The separation and identification of the main
reaction products were performed using a variety of techniques
including chromatography, mass spectrometry and single and
multidimensional NMR. The results show that all methylimida-
zolium cations have similar photochemical behaviors, whereas
isotopic substitution experiments provide additional insights
about mechanistic aspects of the photoreaction.
pressure, yielding
(500 MHz, D2O,
3.88 (s, 6H-C(6-7)). 13C NMR: (500 MHz, D2O,
a
colorless solid of MMIMCl. 1H NMR:
in ppm) 8.64 (s, 1H-C(2)), 7.41 (s, 2H-C(4-5)),
in ppm) 136.49 (C(2)),
d
d
123.35 (C(4-5)), 35.57 (C(6-7)). The Scheme 1 shows the chemical
structures of three ionic liquids used for the experiments.
Deionized water (18 M
V
ꢁ cm), used for solution preparation,
was obtained from a Milli-Q system. Water solutions of ILs were
prepared by weighing, one hour before irradiation experiments.
Studies were carried out under normal atmospheric air conditions,
since control experiments performed under CO2 and N2 atmos-
pheres did not show different results, compared with the former
ones.
2.2. UV irradiation
2. Materials and methods
A deuterium lamp (Oriel 6316 STD D2-30W) placed inside an
air-cooled housing was used for the broadband irradiation of the
samples. In order to obtain the photon irradiance between 190 and
235 nm, where BMIMCl absorbs, first it was determined the
2.1. Sample preparation
1-Butyl-3-methylimidazolium chloride (BMIMCl) was obtained
by chemical synthesis in our laboratories following standard
procedures [8]. Freshly distilled 1-chlorobutane (Merck-Reagent
for synthesis) was added dropwise to distilled 1-methylimidazole
(Merck-Reagent for synthesis) (1.3:1 mole proportion) under
vigorous stirring. The mixture was kept at 50 ꢀC for 5 days under
continuous stirring, in a N2 atmosphere, and subsequently purified
by recrystallization from acetonitrile-ethyl acetate mixtures. 1H
irradiance between 320 and 400 nm, E = 100 m
Watts/cm2, using a
SpectrolineTM DM-365XA radiometer from Spectronic Corpora-
tion. With this information and, taking into account the spectral
irradiance of the lamp provided by the manufacturers, it was
estimated a photon irradiance amount basis [31] En,p = 17 mEisteins
mꢂ2 sꢂ1. The irradiation spot was large enough (ꢃ1 cm2) to avoid
small irradiation volumes. No temperature changes were regis-
tered in the sample during the irradiation. In order to ensure that
the photolysis was produced by UV light, control experiments were
performed using a narrow band excitation, using a commercial
PTI-Spectrofluorometer provided with a Xenon lamp and mono-
chromators. The excitation wavelength was fixed at 211 nm with a
narrow excitation slit of ꢃ1 nm. The results obtained in these
experiments were the same as those recorded using broadband
illumination. However, the latter alternative was chosen since it
required shorter irradiation times.
NMR: (500 MHz, D2O,
d in ppm) 8.70 (s. n. r., 1H-C(2)), 7.47 (s. n. r.,
1H-C(5)), 7.42 (s. n. r., 1H-C(4)), 4.19 (t, 2H-C(7)), 3.88 (s, 3H-C(6)),1.84
(m, 2H-C(8)), 1.31 (m, 2H-C(9)), 0.91 (t, 2H-C(10)). 13C NMR:
(500 MHz, D2O,
d in ppm) 135.79 (C(2)), 123.42 (C(4)), 122.17
(C(5)), 49.24 (C(7)), 35.57 (C(6)), 31.22 (C(8)), 18.70 (C(9)), 12.57 (C(10)).
1-Etyl-3-methylimidazolium chloride (EMIMCl) was purchased
from Aldrich (Purity 98%) and was used as received. 1H NMR:
(500 MHz, D2O,
J = 1.8 Hz), 7.41 (d, 1H-C(4), J = 1.8 Hz), 4.23 (q, 2H-C(7)), 3.88 (s, 3H-
(6)), 1.49 (t, 3H-C(8)). 13C NMR: (500 MHz, D2O,
in ppm) 135.48
d in ppm) 8.70 (s, 1H-C(2)), 7.48 (d, 1H-C(5),
C
d
2.3. HPLC and NMR measurements
(C(2)), 123.39 (C(4)), 121.81 (C(5)), 44.73 (C(7)), 35.55 (C(6)), 14.45
(C(8)).
Three different equipments were used for HPLC runs for the
following purposes: equipment 1 was used for the separation and
identification of the main components of the photoreaction
mixtures; equipment 2 was used with the only purpose to
determine the high resolution mass spectra of reactants and
products and, equipment 3 was used for the isolation and
1,3-Dimethylimidazolium chloride (MMIMCl) was obtained by
chemical synthesis in our laboratories. 1.7 mL of iodomethane
(Merck-Reagent for synthesis) was added dropwise to 2 mL of
distilled 1-methylimidazole (Merck-Reagent for synthesis)
(1.08:1 mole proportion) under vigorous stirring at 0 ꢀC for 3 h.
Scheme 1. Chemical structures of ionic liquids used for the experiments: (a) BMIMCl, (b) EMIMCl and (c) MMIMCl.

G.P. Sarmiento et al. / Journal of Photochemistry and Photobiology A: Chemistry 314 (2016) 155–163
157
subsequent chemical characterization of the main photoreaction
products. Pure water and diluted acidic solutions were tested as
mobile phases, in order to find the best experimental run
conditions. The diluted acidic solutions were finally preferred
because they yield the chromatograms with the narrowest peaks.
Equipment 1 was a Shimadzu 2020 LC-PDA-ESI–MS instrument.
ESI was in positive mode under the following conditions: scan
range 15 < m/z < 1000; ESI voltage 2.5 kV; desolvation line (DL)
temperature 100 ꢀC; heated block temperature 200 ꢀC; DL voltage
+80 V; Q-array DC voltage +10 V; Q-array RF voltage 60 V;
nebulizing gas flow rate 1.5 L/min; drying gas flow rate 18 L/min.
Chromatographic and PDA conditions were: Gemini–NX
To separate and characterize the main reaction products,
ꢃ1 ꢁ10ꢂ4 M of BMIMCl, EMIMCl and MMIMCl water solutions
were irradiated during 1 h and injected into equipments 1 and 2.
For the NMR measurements, 1 ꢁ10ꢂ2 M ImILs solutions in
either D2O or H2O were irradiated during 22 h and injected into
equipment 3. Different elution volumes per run were collected for
the narrower peaks corresponding to reactant and products. The
collected fractions were frozen, freeze-dried and re-dissolved in
0.6 mL of D2O for NMR measurements.
NMR spectra were recorded on a Bruker Avance II 500, using
D2O and water/D2O as solvents. Measurements include 1H NMR,
13C NMR, COSY, HSQC and NOESY experiments.
C-18 column (Phenomenex), 3
ature 25 ꢀC; PDA range 200–350 nm; PDA flow-cell temperature
25 ꢀC; injection volume 20
L. A flow rate of 1 mL/min was
m
m, 150 ꢁ 4.6 mm; column temper-
3. Results and discussion
m
established with a flow reduction down to 0.2 mL/min at the ESI
entrance. A washing step with methanol was included between
runs.
Equipment 2 was a Bruker micrOTOF-Q instrument. ESI was
operated in positive mode under the following conditions: scan
range 50 < m/z < 950; capillary voltage 4.5 kV; end plate offset
ꢂ500 V; collision cell RF 100.0 VPP, nebulizing gas pressure 3.5 bar;
drying heater temperature 200 ꢀC; drying gas flow rate 7 L/min.
Chromatographic and PDA conditions were: Luna C-18 column
3.1. Separation of photoproducts from the reaction media
BMIMCl, EMIMCl and MMIMCl aqueous solutions were exposed
to UV radiation. The irradiated solutions were injected into a
HPLC-PDA-ESI–MS (see Section 2.3 for details). Fig. 1 shows the
chromatograms of the three irradiated solutions, obtained by
monitoring the UV absorption at two different wavelengths: (i) at
211 nm, where the imidazolium cation has an absorption maxi-
mum and, (ii) at 230 nm, where the imidazolium cation does not
appreciably absorb. The figure also includes the mass spectra at
some selected retention times.
(Phenomenex),
3
m
m, 100 ꢁ 2 mm; injection volume 10
mL;
column temperature 25 ꢀC; PDA range 200–350 nm; PDA flow-
cell temperature 25 ꢀC. Elution with 0.1% aqueous formic acid was
employed at a flow rate of 0.2 mL/min; a linear solvent program for
methanol to clean the column was as follows: 0 min 5%, 6 min 30%,
12 min 55%, 18 min 80%, 20 min 5%.
BMIM+ chromatogram (Fig. 1a) shows three well-resolved
peaks at 3.96, 4.53 and 5.32 min. The peak at 5.32 min is associated
with a mass spectrum containing a single signal at m/z = 139+. This
peak is assigned to unreacted BMIM+, as it was confirmed by
independent runs performed with non-irradiated solutions of pure
ILs. In the same chromatogram two peaks are observed at shorter
retention times, 3.96 and 4.53 min. The mass spectra associated
with both peaks show single signals at m/z = 157+. The differences
observed in: (i) the retention times, (ii) the m/z ratios and (iii) the
absorption characteristics of these compounds compared with the
reactants, clearly indicate the presence and the effective separation
of, at least, two photoreaction products.
Equipment 3 was used to separate the photoproducts to be used
in the subsequent 1H NMR and HSQC measurements. It consisted of
a Gemini–NX C-18 column (Phenomenex), 3
m
m, 150 ꢁ 4.6 mm, a
binary pump (Thermo Spectra series P200, USA), a UV–vis detector
(Thermo UV 100, USA) set at 220 nm, and a manual valve
(Rheodyne 2895, USA) for fraction collection. Elution was carried
out isocratically with 0.0025% acetic acid/water mixture at a flow
rate of 1 mL/min. A washing step with methanol was included
between runs.
Fig.1. HPLC chromatograms of irradiated solutions of: (a) BMIMCl, (b) EMIMCl and, (c) MMIMCl, obtained in a C18 column. The detection was performed at 211 nm, where the
imidazolium cation has an absorption maximum (dashed line) and at 230 nm where the imidazolium cation does not appreciably absorb (full line). Mass spectra were
obtained at the selected retention times in the chromatogram.

158
G.P. Sarmiento et al. / Journal of Photochemistry and Photobiology A: Chemistry 314 (2016) 155–163
EMIM+ chromatogram (Fig. 1b) shows similar features, al-
addition of a water molecule to the reactant. Considering that the
relationship between the isotopic abundances of Cl37 and Cl35 is
24.23/75.77 ꢄ 1/3, the absence of mass signals at [(m/z) + 2] in the
mass spectra (positive mode) of the irradiated samples provides
clear evidence of the absence of chlorine atoms in the photo-
products of the reaction. On the other hand, the high resolution
mass spectra of the products reveals molecular formulas of the type
C5H11N2O-(CH2)n+ with n = 0, 1 or 3 for MMIM+, EMIM+ and BMIM+
respectively, confirming the addition of a water molecule to each
methylimidazolium reactant. The incorporation of an oxygen
functional group leads to products with higher polarities than
reactants and justifies the lower retention times observed in the
chromatograms.
Table 1 summarizes the main characteristics of reactants and
products: retention times, m/z ratios, molecular formulas and UV
spectra. It is remarkable that all products show similar UV spectra
with absorption maxima around 230 nm. In particular, the two
products of BMIM+ photolysis have the same absorption spectra.
These photoproducts absorption spectra, in turn, present different
absorption bands than reagents, which show absorption maxima
around 210 nm. The addition of water has a significant effect on the
chromophoric unit of the original reactant.
though with less resolved peaks. At t = 2.91 min a peak with an
associated mass of m/z = 111+ is found. This peak corresponds to
unreacted EMIM+. The shorter retention time of EMIM+ in
C18 column compared to BMIM+ is consistent with the shorter
ethyl chain that confers a major polarity to the reactant. In the
same figure, a peak at 2.78 min with a mass spectrum showing a
single signal at m/z = 129+ is also observed. This signal confirms the
presence of, at least, one reaction product.
Finally, MMIM+ chromatogram is presented in Fig. 1c. It can be
observed a peak at 2.80 min associated to a mass spectrum with a
single signal at m/z = 97+ that corresponds to MMIM+. This reagent
exhibits the shortest retention time of the three methylimidazo-
lium cations, since the presence of a methyl group turns it into the
reagent with highest polarity. As in the two previous cases, the
presence of a peak with a shorter retention time than the reactive,
at 2.78 min, is also observed. This feature is associated to a mass
spectrum with a single signal at m/z = 115+.
In the three cases, the retention times of products are always less
thanthose of the correspondingreagents. Furthermore, the m/z ratio
of the products is related to that of the reactants according to the
following simple expression: m/zprod = m/zreact + 18, suggesting the
Table 1
HPLC retention times, m/z ratios, molecular formulas and UV spectra of reactants and products.
Reactant
Product
Rt/min
m/z molecular formula
Absorption spectrum
Rt/min
m/z molecular formula
Absorption spectrum
3.96
157.134
C8H17N2O+
157.134
5.32
139.123
C8H15N2
+
4.53
2.78
2.78
129.102
C6H13N2O+
129.102
2.91
111.092
C6H11N2
+
2.80
97.076
C5H9N2
2.78
115.087
C5H11N2O+
+

G.P. Sarmiento et al. / Journal of Photochemistry and Photobiology A: Chemistry 314 (2016) 155–163
159
Table 2
Chemical structures and 1H NMR and 13C NMR signals of the photoreaction products.
1H RMN
13C from HSQC
Atom
Molecule
Atom
Signal type
d
d
H–C(2)
H–C(5)
H–C(4)
H–C(4)
H3–C(6)
H2–C(7)
H2–C(8)
H2–C(9)
H3–C(10)
s
8.20
5.74
4.09
3.69
3.20
3.52
1.66
1.33
0.91
C(2)
C(5)
C(4)
C(6)
C(7)
C(8)
C(9)
C(10)
156.53
83.11
57.53
34.32
44.14
29.34
18.98
12.75
dd (J = 4 and 9)
dd (J = 9 and 14)
dd (J = 4 and 14)
s
t (J = 7)
q (J = 7)
sex (J = 7)
t (J = 7)
H–C(2)
H–C(4)
H–C(5)
H–C(5)
H3–C(6)
H2–C(7)
H2–C(8)
H2–C(9)
H3–C(10)
s
8.20
5.65
4.11
3.72
3.18
3.53
1.64
1.32
0.91
C(2)
C(4)
C(5)
C(6)
C(7)
C(8)
C(9)
C(10)
156.45
84.13
55.52
31.59
47.73
28.53
19.01
12.75
dd (J = 4 and 9)
dd (J = 9 and 14)
dd (J = 4 and 14)
s
t (J = 7)
q (J = 7)
sex (J = 7)
t (J = 7)
H–C(2)
H–C(5)
H–C(4)
H–C(4)
H3–C(6)
H2–C(7)
H3–C(8)
s
8.10
5.66
3.99
3.59
3.11
3.46
1.20
C(2)
C(5)
C(4)
C(6)
C(7)
C(8)
156.16
83.04
57.60
34.25
43.09
12.84
dd (J = 4 and 9)
dd (J = 9 and 14)
dd (J = 4 and 14)
s
q (J = 7)
t (J = 7)
H–C(2)
H–C(4)
H–C(5)
H–C(5)
H3–C(6)
H2–C(7)
H3–C(8)
s
8.09
5.55
4.02
3.62
3.08
3.38
1.18
C(2)
C(5)
C(4)
C(6)
C(7)
C(8)
156.16
55.39
84.20
31.98
45.08
12.12
dd (J = 4 and 9)
dd (J = 9 and 14)
dd (J = 4 and 14)
s
q (J = 7)
t (J = 7)
H–C(2)
H–C(5)
H–C(5)
H–C(4)
H3–C(6)
H3- C(7)
s
8.03
3.59
3.99
5.55
3.08
3.10
C(2)
C(4)
C(5)
C(6)
C(7)
157.07
84.41
57.52
30.48
34.25
dd (J = 4 and 14)
dd (J = 9 and 14)
dd (J = 4 and 9)
s
s
3.2. Chemical structures of photoproducts
1H NMR spectra of the products show clear differences when
compared to reactants. In particular, it is noticeable the simulta-
neous disappearance of the signals corresponding to the hydrogens
located at positions 4 and 5 of the imidazolium ring and the growth
of three new hydrogen signals with gem, syn and anti coupling
between 3.6 and 5.8 ppm. These features imply that the hydroxyl
and proton additions take place at either positions 4 (next to the
methyl group) or 5 (next to the alkyl chain) on the imidazolium
ring, leading to two positional isomers. Table 2 summarizes the
NMR signals and the proposed chemical structures of the products.
Once the spectra of the reaction products were identified, the
next step in the analysis involved the consideration of coupling
signals in NOESY spectra, in order to assign NMR spectra to each
positional isomer. As an example, the particular case of EMIM+ will
be analyzed, although similar considerations apply for the case of
BMIM+. Fig. 2 shows a selected region of NOESY spectrum of EMIM+
The products of the three photoreactions were separated from
each reaction medium by HPLC. Once isolated, they were frozen,
freeze-dried and analyzed by single and multidimensional NMR
(see Section 2.3 for details).
In the case of BMIM+, one product was isolated from each
chromatogram peak. The chemical structures of these products
were obtained by combining information provided by mass and
NMR spectra, including isotopic substitution experiments
(vide infra). Unfortunately, the peaks in EMIM+ and MMIM+
chromatograms could not be properly resolved; in particular, the
product peak in the case of EMIM+ stands an unresolved peak.
Nevertheless, in the latter case, the NMR spectra can be understood
as a combination of just two products structurally similar to those
obtained in the BMIM+ photolysis.

160
G.P. Sarmiento et al. / Journal of Photochemistry and Photobiology A: Chemistry 314 (2016) 155–163
Fig. 2. Selected region of NOESY spectrum of EMIMCl irradiation products.
products. It can be seen that the signal of the hydrogen located at
the hydroxylated carbon (5.55 ppm) couples to those of hydrogens
at methyl position (3.08 ppm) only for the isomer that show
hydroxyl addition to C4 position. Thereby, it was possible to assign
each spectrum with their respective isomer.
shown). Consequently, the products formed by addition of water to
the imidazolium ring constitute about 50% of the total products
and, evidently, represent the principal reaction products. The
existence of other reaction products with very high retention times
in HPLC experiments has also been observed. Additional irradia-
tion experiments were performed on aqueous solutions of HPLC
isolated A and B isomers, in order to evaluate the hypothesis that
these unidentified products could derive from further photo-
degradation of the hydroxylated products. The results of these
experiments showed that no photochemical degradation was
observed for these compounds, at least at the irradiation wave-
lengths at which they absorb. These results confirm that
unidentified compounds are originated from the direct irradiation
of imidazolium cations in aqueous solution. The separation and
chemical characterization of these photoproducts, due to their
lower concentration and complex structure, deserve a separate
treatment that is beyond the scope of this paper.
To facilitate the following discussion, it will be useful to denote
the isomers with the hydroxyl group at the position 5 as of the “A”
type, whereas those with the hydroxyl group at the position 4 as of
the “B” type. A/B ratios are far from unit for the Bmim+ and Emim+
reactions, whereas the symmetry of the MMIM+ cation precludes
the existence of two positional isomers. This magnitude can be
calculated in each case by integrating 1H NMR signals of the
reaction mixtures corresponding to the hydrogens at position 5 in
A isomers and at position 4 in B isomers. The calculations yield
A/B = 1.50 and A/B = 1.12 ratios for R = Butyl and R = Ethyl products
respectively. In all cases, A isomers are more abundant, suggesting
that the addition is not driven by steric effects. As such, it may be
possible that the stability of an eventual intermediate dictates the
position of the substitution, but further experiment should be
performed in order to confirm this hypothesis.
Since the chloride anions are not present as part of the
photoproducts, an interesting question arises concerning whether
they might intervene at any stage in the reaction. The absorption
spectrum of the chloride anion in aqueous solution shows a
The analysis of 1H NMR spectra of the reaction mixtures before
and after irradiation shows that the total number of hydrogen
atoms in A and B isomers accounts approximately for half the
number of imidazolium hydrogens that have reacted—which are
roughly estimated in ꢃ25 % of the total reactant—(data non
maximum in 175 nm(e(
= 160 Mꢂ1 cmꢂ1) extending up to
l
=175nm)
190 nm [32]. In the experiments performed on diluted solutions
with a narrow bandwidth excitation of 211 ꢅ1 nm, the fraction of
light absorbed by this species is negligible compared to the one of
Scheme 2. Different isomers feasible to be obtained by irradiation of unsymetrical methylimidazolium cations in D2O.

G.P. Sarmiento et al. / Journal of Photochemistry and Photobiology A: Chemistry 314 (2016) 155–163
161
Fig. 3. 1H NMR signals of the hydrogens located at positions 4 and 5 of the BMIM+ photoreaction products in: (i) aqueous solution, (ii) D2O solution.
the imidazolium cations. In order to experimentally confirm that
chloride anions do not play any relevant role in the reaction,
additional irradiation experiments were performed by adding NaCl
to diluted solutions of BMIMCl. No differences in the formation of
the photoproducts were observed for different chloride concen-
trations.
The same line of reasoning used for the chloride case can be
applied to the hydroxyl anion. The molar absorption coefficient of
the hydroxyl anion at 211 nm is ꢃ300 Mꢂ1 cmꢂ1 [33]. In neutral
solutions, the lower concentration of hydroxyl ions makes the
fraction of light absorbed by this specie at 211 nm also negligible.
Experiments performed by adding NaOH to BMIMCl solutions do
not show changes in the formation of photoproducts.
3.3. Stereochemistry of the photoreaction
In order to study the stereochemistry of the photoreaction,
isotopic substitution experiments were performed by irradiating
methylimidazolium cations in D2O. The reaction of unsymetrical
methylimidazolim cations in D2O yields eight different isomers,
some of them indistinguishable by conventional 1H NMR
spectroscopy. Scheme 2 shows the chemical structures of these
isomers. The two A isomers exhibiting syn addition are indistin-
guishable by 1H NMR because they are enantiomers, and will be
referred hereinafter as A0. On the other hand, the two A isomers
showing anti addition are also enantiomers and will be referred as
AA0 and A00 isomers are diastereoisomers. A similar description
100
150
a
b
100
50
0
50
0
5.70
5.72
5.74
ppm
5.76
5.78
5.62
5.64
5.66
5.68
ppm
Fig. 4. 1H NMR signals of hydrogens at position: (a) C5 for A-type isomer and, (b) C4 for B-type isomer, obtained by irradiation of a D2O solution of BMIMCl at 211 nm.
The dotted lines represent the Lorentzian functions used for the convolution and the dashed line indicates the sum of all contributions.

162
G.P. Sarmiento et al. / Journal of Photochemistry and Photobiology A: Chemistry 314 (2016) 155–163
Table 3
the presence of syn and anti additions in equal proportions,
suggesting that the reaction proceeds in an unconcerted fashion.
Spectral analysis of the mixtures obtained by irradiation of
EMIMCl and MMIMCl in D2O shows similar results (data not
shown), indicating the formation of diastereoisomers in the same
proportion for all cases.
Peak position and Lorentzian functions areas used for the reproduction of the 1H
NMR spectra of the hydrogens at the positions: (a) C5 for A-type isomer and, (b)
C4 for B-type isomer, obtained by irradiation of D2O solution of BMIM+ at 211 nm.
Type-A isomers
Lorentzian
Type-B isomers
Lorentzian
Area
Center
Area
Center
1
2
3
4
0.51
0.54
0.53
0.52
5.732
5.737
5.745
5.750
1
2
3
4
0.34
0.31
0.34
0.32
5.640
5.645
5.652
5.657
4. Conclusions
In this work, photochemical reactions of methylimidazolium
cations induced by UV-C radiation in aqueous solution have been
investigated. The main result can be summarized by the following
reaction Scheme 3:
The principal products of the photochemical reaction are the
result of the chemical addition of a proton and a hydroxyl group to
applies for B isomers. The two B isomers with addition syn will be
referred to as B0, while the two B isomers with addition anti will be
Scheme 3. Photochemical reaction of methylimidazole cations in aqueous solution under UV-C radiation.
referred to as B00. The following analysis, performed for stereo-
chemitry characterization of the photoreactions, was carried out
on NMR spectra of reaction mixtures instead of NMR of individual
compounds since they provide extra information about A0/A00 and
B0/B00 ratios, as will be shown in the next paragraphs.
the imidazolium ring, which comprise ꢃ50% of the total reaction
products. The addition of the hydroxyl group takes place either at
positions 4 or 5 of the imidazolium ring, leading the formation of
two positional isomers. For MMIM+ only one substitution product
is obtained, due to the symmetry of the molecular ion. The ratios
between isomer populations depend on the lengths of the alkyl
chains bonded to the nitrogen atoms. The most abundant isomers
correspond to those that show the addition of the hydroxyl group
at the ring side having the longer alkyl substituent.
Results from experiments performed in D2O solutions reveal
that the addition of hydrogen and hydroxyl group: (i) takes place at
different carbon atoms on the imidazolium ring and, (ii) produce
equal amounts of syn and anti additions.
The formation of diastereoisomers at equal proportions could
be a strong indication that the reaction does not proceed via a
mechanism that involves a single step. In principle, a concerted
attack on the double bond should promote a preferential syn or anti
addition. As such, further experiments should be performed in the
future to establish the mechanism of the photochemical reaction.
Fig. 3 shows 1H NMR signals related with the two products of
the BMIM+ photoreaction in: (i) aqueous solution and, (ii) D2O
solution. In each case, the selected signals correspond to hydrogens
at positions 4 and 5: three for A isomers and three for B isomers.
The hydrogen signals of the products obtained in water show three
different coupling constants: Jgem = 14, Jcis = 9 and Jtrans = 4. On the
other hand, the signals of the products obtained in D2O show the
disappearance of the large 1H-1H gem coupling (Jgem = 14). The
absence of this coupling confirms that the addition of deuterium
atom always occurs on the carbon that is not hydroxylated. In other
words, the results indicate that the addition of D and OD never
takes place at the same carbon.
Although small shifts and widening are evident—most likely as
a result from hydrogen and deuterium coupling—none of the above
mentioned signals fully disappear. It is note that if the addition had
proceeded completely either in a syn or anti manner, some of the
signals of the hydrogens attached to un-hydroxylated carbon in the
aqueous mixture (C4 for A isomer or C5 for B isomer) should have
disappeared. The fact that this is not observed indicates that the
addition does not occur exclusively with a syn or anti stereochem-
istry, but proceeds in the two ways simultaneously.
The signal at 5.74 in Fig. 3ii can be interpreted as a sum
of the two signals corresponding to syn and anti additions. The
different dihedral angles in each isomer yield different coupling
constants—Jcis = 9 and Jtrans = 4—[34], leading to two distinguish-
able doublets. As such, it is possible to quantitatively measure the
relative abundance of syn and anti additions by evaluating the
contribution of each doublet signal to the total signal intensity. The
same line of reasoning can be applied for the signal at 5.65 ppm.
Fig. 4 shows the decomposition of these signals as a sum of several
Lorentzian functions. The corresponding maxima and areas of each
curve are listed in Table 3. It is noteworthy that the areas of the four
functions are similar, indicating that both diasteroisomers
contribute to the total intensity in equal form. This result reveals
Acknowledgments
Funding was obtained from ANPCyT (PICT 2008-1490,
2012-0852) A. Z. and M. M. are staff members of CONICET.
Discussions with Andrea Bordoni, Maria Laura Japas, Javier
Rodriguez, Karin Gutkowski, Pablo Hoijemberg and Daniel Laria
are greatly acknowledged.
References
[1] R.D. Rogers, G.A. Voth, Ionic liquids, Acc. Chem. Res. 40 (2007) 1077–1078.
[2] E.W. Castner Jr, J.F. Wishart, Spotlight on ionic liquids, J. Chem. Phys.132 (2010)
120901–120909.
[3] C.P. Fredlake, J.M. Crosthwaite, D.G. Hert, S.N.V.K. Aki, J.F. Brennecke,
Thermophysical properties of imidazolium-based ionic liquids, J. Chem. Eng.
Data 49 (2004) 954–964.
[4] H. Weingärtner, Understanding ionic liquids at the molecular level: facts,
problems, and controversies, Angew. Chem. Int. Ed. 47 (2008) 654–670.
[5] J. Dupont, G.S. Fonseca, A.P. Umpierre, P.F.P. Fichtner, S.R. Teixeira, Transition-
metal nanoparticles in imidazolium ionic liquids: recyclable catalysts for
biphasic hydrogenation reactions, J. Am. Chem. Soc. 124 (2002) 4228–4229.
[6] T. Welton, Ionic liquids in catalysis, Coord. Chem. Rev. 248 (2004) 2459–2477.

G.P. Sarmiento et al. / Journal of Photochemistry and Photobiology A: Chemistry 314 (2016) 155–163
163
[7] T. Welton, Room-temperature ionic liquids. Solvent for synthesis and catalysis,
Chem. Rev. 99 (1999) 2071–2084.
[8] J.P. Hallett, T. Welton, Room-temperature ionic liquids. Solvent for synthesis
[21] M. Álvaro, E. Carbonell, B. Ferrer, H. Garcia, J.R. Herance, Ionic liquids as a novel
medium for photochemical reactions. Ru(bpy) 32+/Viologen in imidazolium
ionic liquid as a photocatalytic system mimicking the oxido-reductase
enzyme, Photochem. Photobiol. 82 (2006) 185–190.
and catalysis. 2, Chem. Rev. 111 (2011) 3508–3576.
[9] P. Dash, RWJ, Scott, 1-Methylimidazole stabilization of gold nanoparticles in
imidazolium ionic liquids, Chem. Commun. 43 (2009) 812–814.
[10] M. Antonietti, D. Kuang, B. Smarsly, Y. Zhou, Ionic liquids for the convenient
synthesis of functional nanoparticles and other inorganic nanostructures,
Angew. Chem. Int. Ed. 43 (2004) 4988–4992.
[22] E.W. Castner Jr, C.J. Margulis, M. Maroncelli, J.F. Wishart, Ionic liquids:
structure and photochemical reactions, Annu. Rev. Phys. Chem. 62 (2011)
85–105.
[23] S. Marquis, B. Ferrer, M. Alvaro, H. Garcia, H.D. Roth, Photoinduced electron
transfer in ionic liquids: use of 2,4,6-triphenylthiapyrylium as a
photosensitizer probe, J. Phys. Chem. B 110 (2006) 14956–14960.
[24] M. Grätzel, Dye-sensitized solar cells, J. Photochem. Photobiol. C: Photochem.
Rev. 4 (2003) 145–153.
[25] N. Yamanaka, R. Kawano, W. Kubo, N. Masaki, T. Kitamura, Y. Wada, M.
Watanabe, S. Yanagida, Dye-sensitized TiO2 solar cells using imidazolium-
type ionic liquid crystal systems as effective electrolytes, J. Phys. Chem. B 111
(2007) 4763–4769.
[11] X. Sun, H. Luo, S. Dai, Ionic liquids-based extraction: a promising strategy for
the advanced nuclear fuel cycle, Chem. Rev. 112 (2012) 2100–2128.
[12] M. Mirenda, D. Rodrigues, P. Arenillas, K. Gutkowski, X. Mougeot, Ionic liquids
ꢀ
as solvents for liquid scintillation technology. Cerenkov counting with
1-butyl-3-methylimidazolium chloride, Radiat. Phys. Chem. 98 (2014) 98–102.
[13] T.P.T. Pham, C.-W. Cho, Y.-S. Yun, Environmental fate and toxicity of ionic
liquids: a review, Water Res. 44 (2010) 352–372.
ꢁ
ꢁ
[14] M.C. Bubalo, K. Radoševic, I.R. Redovnikovic, J. Halambek, V.G. Src9ek, A brief
overview of the potential environmental hazards of ionic liquids, Ecotox.
Environ. Safe 99 (2014) 1–12.
[26] A.-V. Mudring, Optical spectroscopy and ionic liquids, Top. Curr. Chem. 290
(2009) 285–310.
[27] P.K. Mandal, A. Paul, A. Samanta, Room temperature ionic liquids as media for
photophysical studies, J. Chin. Chem. Soc. 53 (2006) 247–252.
[28] A. Paul, P.K. Mandal, A. Samanta, On the optical properties of the imidazolium
ionic liquids, J. Phys. Chem. B 109 (2005) 9148–9153.
[29] P.K. Mandal, A. Paul, A. Samanta, Excitation wavelength dependent
fluorescence behavior of the room temperature ionic liquids and dissolved
dipolar solutes, J. Photochem. Photobiol. A Chem 182 (2006) 113–120.
[30] A. Paul, A. Samanta, Optical absorption and fluorescence studies on
imidazolium ionic liquids comprising the bis(trifluoromethanesulphonyl)
imide anion, J. Chem. Sci. 118 (2006) 335–340.
[15] E. M. Siedlecka, M., Czerwicka, J., Neumann, P., Stepnowski, J. F. Fernández, J.
Thöming, (2011) Ionic Liquids: Methods of Degradation and Recovery, Ionic
Liquids: Theory, Properties, New Approaches, Prof. Alexander Kokorin (Ed.),
ISBN: 978-953-307-349-1, InTech, DOI: 10.5772/15463. Available from: http://
www.intechopen.com/books/ionic-liquids-theory-properties-
newapproaches/ionic-liquids-methods-of-degradation-and-recovery.
[16] E.M. Siedlecka, M. Golebiowski, J. Kumirska, P. Stepnowski, Identification of
1-butyl-3-methylimidazolium chloride degradation products formed in Fe
(III)/H2O2 oxidation system, Chem. Anal. (Warsaw) 53 (2008) 943–951.
[17] M. Muñoz, C.M. Domínguez, Z.M. de Pedro, A. Quintanilla, J.A. Casas, J.J.
Rodriguez, Ionic liquids breakdown by Fenton oxidation, Catal. Today 240
(2015) 16–21.
[18] M. Czerwicka, W., Mrozik, A. Müller, M., Golebiowski, J., Kumirska, E., Malinski,
P. Stepnowski, Study on degradation products of imidazolium ionic liquids in
H2O2/UV oxidation system, Proceedings of the 10th international conference
on environmental science and technology, (2007) A-281–A-286.
[19] N. Gathergood, M.T. Garcia, P.J. Scammells, Biodegradable ionic liquids: part I.
Concept, preliminary targets and evaluation, Green Chem. 6 (2004) 166–175.
[20] P. Stepnowski, A. Zaleska, Comparison of different advanced oxidation
processes for the degradation of room temperature ionic liquids, J. Photochem.
Photobiol. A Chem. 170 (2005) 45–50.
[31] S.E. Braslavsky, Glossary of terms used in photochemistry, Pure Appl. Chem. 79
(2007) 293–465.
[32] M.F. Fox, B.E. Barker, E. Hayon, Far-ultraviolet solution spectroscopy of chloride
ion, J. Chem. Soc., Faraday Trans. 174 (1978) 1776–1785.
[33] S.O. Nielsen, B.D. Michael, E.J. Hart, Ultraviolet absorption spectra of eaq-, H,
OH, D and OD from pulse radiolysis of aqueous solutions, J. Phys. Chem 80
(1976) 2482–2488.
[34] M. Karplus, Vecinal proton coupling in nuclear magnetic resonance, J. Am.
Chem. Soc. 85 (1963) 2870–2871.