126
M. Lotfi et al. / Journal of Molecular Liquids 243 (2017) 124–131
x
energy of fusion ΔGfus is calculated by the QSPR approach and approxi-
under vacuum using a rotary evaporator, followed by further vacuum
drying at 70 °C for 24 h. H NMR: δ (ppm) 0.22 (t, 8H), 0.87 (m, 16H),
1
mated from the following COSMOtherm descriptors:
1
.56 (m, 12H), 7.86 (s, 1H).
ΔG ¼ c1μðiH2OÞ þ c2Ni þ c3V þ c4
x
ring
ð3Þ
fus
i
2.5.3. Choline acetate [Ch][OAc]
where c
of fusion. μ
1
to c
4
2
represent the QSPR parameters for the solute-free energy
[Ch][OAc] was synthesized as demonstrated by Fukaya [29]. In brief,
(
H
O)
ring
i
is the chemical potential of solute i in water, N
i
is the
choline hydroxide aqueous solution was slowly added to a slight excess
of the acetic acid aqueous solution in an ice bath. The mixture was
stirred at room temperature (RT) for 24 h. Water was then removed
by evaporation under reduced pressure, and the resulting product was
washed several times with diethyl ether to remove unreacted acid.
The resulting product was dried under vacuum at 70 °C for 24 h to
produce choline acetate. The chemical structure of the IL was confirmed
number of ring atoms in compound i, and V
ume. Thus, the chemical potential of a compound in pure water is im-
portant for calculation of the solubility in COSMO-RS.
i
is the solute molecular vol-
2
.3. Computational details, the ACV molecule, and the set of ILs
1
The molecular structure of ACV was drawn as a two-dimensional
by H NMR: δ (ppm) 1.95 (s, 3H), 3.20 (s, 9H), 3.51 (m, 2H), 4.02 (m,
structure and then converted to a three-dimensional geometry (Fig. 1)
by Chemdraw Ultra 7.0. It is necessary to determine the lowest energy
conformation of ACV, so the selection of cis–trans and conformational
isomerism of the ring structures must be carefully considered because
the rings can form internal H bonds. The different molecular conforma-
tions were then minimized. A COSMO polarization calculation of the
charge densities σ of the molecular surfaces of ACV was performed
using the TmoleX (version 4.1.1) program package at the density func-
tional theory with empirical dispersion corrections (DFT + disp) level
using the B3LYP (DFT-D3 BJ-damping) functional with TZVP basis set
2H). Differential scanning calorimetric (DSC) analysis revealed that
[Ch][OAc], which is solid at RT, has a relatively low melting point of
T = 51 °C.
m
2.6. Characterization of the ILs
1
The H NMR spectra of the ILs were recorded with a Bruker Ascend
2
500 NMR spectrometer in dimethylsulfoxide (DMSO)/D O. The Fourier
transform infrared (FTIR) spectra of the samples were recorded with
an FTIR 400 spectrometer (Agilent Technologies) in the range 400–
−
1
−1
(
a triple-f valence polarized basis set). Most of the anion and cation
4000 cm with a resolution of 1 cm using the KBr disk technique.
The water contents of the synthesized ILs were determined by coulomet-
ric Karl Fischer titration (Mettler Toledo DL 39) with Hydranal Coulomat
AG reagent (Riedel-de Haen). A viscometer (SVM-3000, Anton-Paar)
COSMO files used in this work were taken from the database of BP-
TZVP-IL (COSMOlogic GmbH & Co. KG, Leverkusen, Germany). The an-
ions and cations that were not available in the COSMO database were
generated by a similar procedure to that used to generate the ACV mol-
ecule. All of the calculations for predicting the solubility of ACV in the ILs
were performed using COSMOthermX (version C30-1601, COSMOlogic
GmbH & Co. KG). For all of the file inputs, only the conformers with
the lowest energies were used.
−
4
3
was used to determine the density (accuracy ± 5 × 10 g/cm ) and
viscosity (accuracy ± 0.35%) of the IPs over the temperature range
20–90 °C.
2.7. Solubility study of ACV in the ILs
To comprehensively evaluate the solubility of ACV in the ILs, a set of
cations (imidazolium, ammonium, pyridinium, and phosphonium) was
selected. Alkyl groups with various chain lengths were added to the cat-
ion parent structures at various positions to investigate their influence
on the solubility of ACV. Thirty-two anions were considered, including
both hydrophilic and hydrophobic anions.
The solubility of ACV in the ILs was determined as shown by
Moniruzzaman [17]. The measurements were performed at RT, except
for the measurements with [Ch][OAc] IL which were performed at 60 °C
because it is solid at RT. The solubility of ACV in the selected ILs was de-
termined as follows. ACV was added in excess to the IL and the mixture
was stirred for 24 h at a constant temperature. The undissolved ACV
was removed by a 0.45 μm Millipore filter. The amount of ACV remaining
in the clear filtrate was measured using an ultraviolet (UV) spectropho-
tometer at 252 nm after appropriate dilution with methanol.
2
.4. Materials
ACV was purchased from Biomarketing Services (M) Sdn Bhd (KL,
Malaysia) and used as received. The [EMIM][OAc] IL (≥95%) was obtain-
ed from Ionic Liquids Technologies GmbH (Heilbronn, Germany) and
used as received. All of the other reagents, including the starting mate-
rials for IL synthesis, were of analytical grade. Milli-Q water was used in
2.8. In vitro cytotoxicity study
The cytotoxicities of the selected ILs were evaluated using the MCF-
10 normal breast epithelial cell line as demonstrated by Ismail and
group [30]. The cells were cultured at the Roswell Park Memorial Insti-
tute (RPMI) 1640 medium supplemented with 1% penicillin/streptomy-
−
1
all of the experiments (conductivity b0.1 S cm ).
2
2
.5. Synthesis of the ILs
2
cin and 10% horse serum at 37 °C (5% CO ). After trypsinization, most of
.5.1. Diethylammonium acetate [DEA][OAc] and triethylammonium ace-
the cells were detached from the plate and then centrifuged (1000 rpm
for 10 s). The plate was re-suspended with phosphate buffered solution
(PBS) containing about 5–10% DMSO. The cell count was performed in a
hemocytometer using a microscope, where the cell concentration was
tate [TEA][OAc]
The [DEA][OAc] and [TEA][OAc] ILs were synthesized as described by
Attri et al. [28]. The purities of the ILs were verified by elementary anal-
1
1
6
ysis and H NMR. For [DEA][OAc], H NMR: δ (ppm) 1.3 (t, 6H), 1.97 (s,
3
estimated at a density of about 10 cells/mL by dilution with RPMI me-
1
H), 2.95 (m, 3H), 9.20 (s, 2H). For [TEA][OAc], H NMR: δ (ppm) 0.77 (t,
dium. 100 μL of the freshly prepared medium was added to each well
followed by 100 μL of test IL by serial dilution. 100 μL of the cell suspen-
sion was then added to each well and the plate was transferred to an in-
cubator. After 48 h of growth, the cell viability in each well was
measured by 1-(4,5-dimethylthiazol-2-yl)-3,5-diphenylformazan
(MTT) assay as follows: 20 μL of 2.5 mg/mL MTT was added to each
well in PBS followed by incubation at 37 °C for 3 h. The resulting liquid
(170 μL) was then aspirated, and the formazan purple crystals were dis-
solved in DMSO (100 μL). The absorbance was measured by a UV spec-
trophotometer at 570 nm using an enzyme-linked immunosorbent
9
H), 1.46 (s, 3H), 2.58 (m, 6H), 11.0 (s, 1H).
2
.5.2. Tributylmethylphosphonium methylsulfate [TBMP][MeSO
4
]
This IL was synthesized by ion exchange reaction between
tributylmethylphsophonium bromide and sodium methylsulfate. Sodi-
um methylsulfate was added to an aqueous solution of
tributylmethylphsophonium bromide. The mixture then was stirred
continuously for 12 h at 27 °C. Sodium bromide as a solid precipitate
was then filtered using filter paper. The obtained IL was dried for 6 h